From 76d08bb3f09054edc45326ce5c698a3f6c45f5d0 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Mon, 5 Jun 2006 13:54:14 -0700
Subject: [IA64] Add "model name" to /proc/cpuinfo

Linux ia64 port tried to decode the processor family number
to something human-readable, but Intel brandnames don't change
synchronously with updates to the family number.  Adopt a more
i386-like approach and just print the family number in decimal.
Add a new field "model name" that uses PAL_BRAND_INFO to find
the official name for the cpu, or on older systems, falls back
to using the well-known codenames (Merced, McKinley, Madison).

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/pal.h       | 10 ++++++++++
 include/asm-ia64/processor.h |  1 +
 2 files changed, 11 insertions(+)

(limited to 'include')

diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h
index 37e52a2836b0..312109d7be97 100644
--- a/include/asm-ia64/pal.h
+++ b/include/asm-ia64/pal.h
@@ -78,6 +78,7 @@
 #define PAL_VM_TR_READ		261	/* read contents of translation register */
 #define PAL_GET_PSTATE		262	/* get the current P-state */
 #define PAL_SET_PSTATE		263	/* set the P-state */
+#define PAL_BRAND_INFO		274	/* Processor branding information */
 
 #ifndef __ASSEMBLY__
 
@@ -1133,6 +1134,15 @@ ia64_pal_set_pstate (u64 pstate_index)
 	return iprv.status;
 }
 
+/* Processor branding information*/
+static inline s64
+ia64_pal_get_brand_info (char *brand_info)
+{
+	struct ia64_pal_retval iprv;
+	PAL_CALL_STK(iprv, PAL_BRAND_INFO, 0, (u64)brand_info, 0);
+	return iprv.status;
+}
+
 /* Cause the processor to enter LIGHT HALT state, where prefetching and execution are
  * suspended, but cache and TLB coherency is maintained.
  */
diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h
index b3bd58e80690..03100c4272c6 100644
--- a/include/asm-ia64/processor.h
+++ b/include/asm-ia64/processor.h
@@ -164,6 +164,7 @@ struct cpuinfo_ia64 {
 	__u8 family;
 	__u8 archrev;
 	char vendor[16];
+	char *model_name;
 
 #ifdef CONFIG_NUMA
 	struct ia64_node_data *node_data;
-- 
cgit v1.2.3


From 2ab561a116e16cdee3ae0e13d51910634c15aee9 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <David.Mosberger@acm.org>
Date: Wed, 21 Jun 2006 11:19:22 -0700
Subject: [IA64] esi-support

Add support for making ESI calls [1].  ESI stands for "Extensible SAL
specification" and is basically a way for invoking firmware
subroutines which are identified by a GUID.  I don't know whether ESI
is used by vendors other than HP (if you do, please let me know) but
as firmware "backdoors" go, this seems one of the cleaner methods, so
it seems reasonable to support it, even though I'm not aware of any
publicly documented ESI calls.  I'd have liked to make the ESI module
completely stand-alone, but unfortunately that is not easily (or not
at all) possible because in order to make ESI calls in physical mode,
a small stub similar to the EFI stub is needed in the kernel proper.
I did try to create a stub that would work in user-level, but it
quickly got ugly beyond recognition (e.g., the stub had to make
assumptions about how the module-loader generated call-stubs work) and
I didn't even get it to work (that's probably fixable, but I didn't
bother because I concluded it was too ugly anyhow).  While it's not
terribly elegant to have kernel code which isn't actively used in the
kernel proper, I think it might be worth making an exception here for
two reasons: the code is trivially small (all that's really needed is
esi_stub.S) and by including it in the normal kernel distro, it might
encourage other OEMs to also use ESI, which I think would be far
better than each inventing their own firmware "backdoor".

The code was originally written by Alex.  I just massaged and packaged
it a bit (and perhaps messed up some things along the way...).

Changes since first version of patch that was posted to mailing list:
* Export ia64_esi_call and ia64_esi_call_phys() as GPL symbols.
* Disallow building esi.c as a module for now.  Building as a module
  would currently lead to an unresolved reference to "sal_lock" on SMP kernels
  because that symbol doesn't get exported.
* Export esi_call_phys() only if ESI is enabled.
* Remove internal stuff from esi.h and add a "proc_type" argument to
  ia64_esi_call() such that serialization-requirements can be expressed (ESI
  follows SAL here, where procedure calls may have to be serialized, are
  MP-safe, or MP-safe andr reentrant).

[1] h21007.www2.hp.com/dspp/tech/tech_TechDocumentDetailPage_IDX/1,1701,919,00.html

Signed-off-by: David Mosberger <David.Mosberger@acm.org>
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/Kconfig             |   8 ++
 arch/ia64/kernel/Makefile     |   5 ++
 arch/ia64/kernel/esi.c        | 205 ++++++++++++++++++++++++++++++++++++++++++
 arch/ia64/kernel/esi_stub.S   |  96 ++++++++++++++++++++
 arch/ia64/kernel/ia64_ksyms.c |   4 +
 include/asm-ia64/esi.h        |  30 +++++++
 6 files changed, 348 insertions(+)
 create mode 100644 arch/ia64/kernel/esi.c
 create mode 100644 arch/ia64/kernel/esi_stub.S
 create mode 100644 include/asm-ia64/esi.h

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 0f3076a820c3..bc106519482c 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -413,6 +413,14 @@ config IA64_PALINFO
 config SGI_SN
 	def_bool y if (IA64_SGI_SN2 || IA64_GENERIC)
 
+config IA64_ESI
+	bool "ESI (Extensible SAL Interface) support"
+	help
+	  If you say Y here, support is built into the kernel to
+	  make ESI calls.  ESI calls are used to support vendor-specific
+	  firmware extensions, such as the ability to inject memory-errors
+	  for test-purposes.  If you're unsure, say N.
+
 source "drivers/sn/Kconfig"
 
 source "drivers/firmware/Kconfig"
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index 09a0dbc17fb6..66ac8e3dca9c 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -31,6 +31,11 @@ obj-$(CONFIG_KPROBES)		+= kprobes.o jprobes.o
 obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR)	+= uncached.o
 mca_recovery-y			+= mca_drv.o mca_drv_asm.o
 
+obj-$(CONFIG_IA64_ESI)		+= esi.o
+ifneq ($(CONFIG_IA64_ESI),)
+obj-y				+= esi_stub.o	# must be in kernel proper
+endif
+
 # The gate DSO image is built using a special linker script.
 targets += gate.so gate-syms.o
 
diff --git a/arch/ia64/kernel/esi.c b/arch/ia64/kernel/esi.c
new file mode 100644
index 000000000000..ebf4e988e78c
--- /dev/null
+++ b/arch/ia64/kernel/esi.c
@@ -0,0 +1,205 @@
+/*
+ * Extensible SAL Interface (ESI) support routines.
+ *
+ * Copyright (C) 2006 Hewlett-Packard Co
+ * 	Alex Williamson <alex.williamson@hp.com>
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/string.h>
+
+#include <asm/esi.h>
+#include <asm/sal.h>
+
+MODULE_AUTHOR("Alex Williamson <alex.williamson@hp.com>");
+MODULE_DESCRIPTION("Extensible SAL Interface (ESI) support");
+MODULE_LICENSE("GPL");
+
+#define MODULE_NAME	"esi"
+
+#define ESI_TABLE_GUID					\
+    EFI_GUID(0x43EA58DC, 0xCF28, 0x4b06, 0xB3,		\
+	     0x91, 0xB7, 0x50, 0x59, 0x34, 0x2B, 0xD4)
+
+enum esi_systab_entry_type {
+	ESI_DESC_ENTRY_POINT = 0
+};
+
+/*
+ * Entry type:	Size:
+ *	0	48
+ */
+#define ESI_DESC_SIZE(type)	"\060"[(unsigned) (type)]
+
+typedef struct ia64_esi_desc_entry_point {
+	u8 type;
+	u8 reserved1[15];
+	u64 esi_proc;
+	u64 gp;
+	efi_guid_t guid;
+} ia64_esi_desc_entry_point_t;
+
+struct pdesc {
+	void *addr;
+	void *gp;
+};
+
+static struct ia64_sal_systab *esi_systab;
+
+static int __init esi_init (void)
+{
+	efi_config_table_t *config_tables;
+	struct ia64_sal_systab *systab;
+	unsigned long esi = 0;
+	char *p;
+	int i;
+
+	config_tables = __va(efi.systab->tables);
+
+	for (i = 0; i < (int) efi.systab->nr_tables; ++i) {
+		if (efi_guidcmp(config_tables[i].guid, ESI_TABLE_GUID) == 0) {
+			esi = config_tables[i].table;
+			break;
+		}
+	}
+
+	if (!esi)
+		return -ENODEV;;
+
+	systab = __va(esi);
+
+	if (strncmp(systab->signature, "ESIT", 4) != 0) {
+		printk(KERN_ERR "bad signature in ESI system table!");
+		return -ENODEV;
+	}
+
+	p = (char *) (systab + 1);
+	for (i = 0; i < systab->entry_count; i++) {
+		/*
+		 * The first byte of each entry type contains the type
+		 * descriptor.
+		 */
+		switch (*p) {
+		      case ESI_DESC_ENTRY_POINT:
+			break;
+		      default:
+			printk(KERN_WARNING "Unkown table type %d found in "
+			       "ESI table, ignoring rest of table\n", *p);
+			return -ENODEV;
+		}
+
+		p += ESI_DESC_SIZE(*p);
+	}
+
+	esi_systab = systab;
+	return 0;
+}
+
+
+int ia64_esi_call (efi_guid_t guid, struct ia64_sal_retval *isrvp,
+		   enum esi_proc_type proc_type, u64 func,
+		   u64 arg1, u64 arg2, u64 arg3, u64 arg4, u64 arg5, u64 arg6,
+		   u64 arg7)
+{
+	struct ia64_fpreg fr[6];
+	unsigned long flags = 0;
+	int i;
+	char *p;
+
+	if (!esi_systab)
+		return -1;
+
+	p = (char *) (esi_systab + 1);
+	for (i = 0; i < esi_systab->entry_count; i++) {
+		if (*p == ESI_DESC_ENTRY_POINT) {
+			ia64_esi_desc_entry_point_t *esi = (void *)p;
+			if (!efi_guidcmp(guid, esi->guid)) {
+				ia64_sal_handler esi_proc;
+				struct pdesc pdesc;
+
+				pdesc.addr = __va(esi->esi_proc);
+				pdesc.gp = __va(esi->gp);
+
+				esi_proc = (ia64_sal_handler) &pdesc;
+
+				ia64_save_scratch_fpregs(fr);
+				if (proc_type == ESI_PROC_SERIALIZED)
+					spin_lock_irqsave(&sal_lock, flags);
+				else if (proc_type == ESI_PROC_MP_SAFE)
+					local_irq_save(flags);
+				else
+					preempt_disable();
+				*isrvp = (*esi_proc)(func, arg1, arg2, arg3,
+						     arg4, arg5, arg6, arg7);
+				if (proc_type == ESI_PROC_SERIALIZED)
+					spin_unlock_irqrestore(&sal_lock,
+							       flags);
+				else if (proc_type == ESI_PROC_MP_SAFE)
+					local_irq_restore(flags);
+				else
+					preempt_enable();
+				ia64_load_scratch_fpregs(fr);
+				return 0;
+			}
+		}
+		p += ESI_DESC_SIZE(*p);
+	}
+	return -1;
+}
+EXPORT_SYMBOL_GPL(ia64_esi_call);
+
+int ia64_esi_call_phys (efi_guid_t guid, struct ia64_sal_retval *isrvp,
+			u64 func, u64 arg1, u64 arg2, u64 arg3, u64 arg4,
+			u64 arg5, u64 arg6, u64 arg7)
+{
+	struct ia64_fpreg fr[6];
+	unsigned long flags;
+	u64 esi_params[8];
+	char *p;
+	int i;
+
+	if (!esi_systab)
+		return -1;
+
+	p = (char *) (esi_systab + 1);
+	for (i = 0; i < esi_systab->entry_count; i++) {
+		if (*p == ESI_DESC_ENTRY_POINT) {
+			ia64_esi_desc_entry_point_t *esi = (void *)p;
+			if (!efi_guidcmp(guid, esi->guid)) {
+				ia64_sal_handler esi_proc;
+				struct pdesc pdesc;
+
+				pdesc.addr = (void *)esi->esi_proc;
+				pdesc.gp = (void *)esi->gp;
+
+				esi_proc = (ia64_sal_handler) &pdesc;
+
+				esi_params[0] = func;
+				esi_params[1] = arg1;
+				esi_params[2] = arg2;
+				esi_params[3] = arg3;
+				esi_params[4] = arg4;
+				esi_params[5] = arg5;
+				esi_params[6] = arg6;
+				esi_params[7] = arg7;
+				ia64_save_scratch_fpregs(fr);
+				spin_lock_irqsave(&sal_lock, flags);
+				*isrvp = esi_call_phys(esi_proc, esi_params);
+				spin_unlock_irqrestore(&sal_lock, flags);
+				ia64_load_scratch_fpregs(fr);
+				return 0;
+			}
+		}
+		p += ESI_DESC_SIZE(*p);
+	}
+	return -1;
+}
+EXPORT_SYMBOL_GPL(ia64_esi_call_phys);
+
+static void __exit esi_exit (void)
+{
+}
+
+module_init(esi_init);
+module_exit(esi_exit);	/* makes module removable... */
diff --git a/arch/ia64/kernel/esi_stub.S b/arch/ia64/kernel/esi_stub.S
new file mode 100644
index 000000000000..6b3d6c1f99b6
--- /dev/null
+++ b/arch/ia64/kernel/esi_stub.S
@@ -0,0 +1,96 @@
+/*
+ * ESI call stub.
+ *
+ * Copyright (C) 2005 Hewlett-Packard Co
+ *	Alex Williamson <alex.williamson@hp.com>
+ *
+ * Based on EFI call stub by David Mosberger.  The stub is virtually
+ * identical to the one for EFI phys-mode calls, except that ESI
+ * calls may have up to 8 arguments, so they get passed to this routine
+ * through memory.
+ *
+ * This stub allows us to make ESI calls in physical mode with interrupts
+ * turned off.  ESI calls may not support calling from virtual mode.
+ *
+ * Google for "Extensible SAL specification" for a document describing the
+ * ESI standard.
+ */
+
+/*
+ * PSR settings as per SAL spec (Chapter 8 in the "IA-64 System
+ * Abstraction Layer Specification", revision 2.6e).  Note that
+ * psr.dfl and psr.dfh MUST be cleared, despite what this manual says.
+ * Otherwise, SAL dies whenever it's trying to do an IA-32 BIOS call
+ * (the br.ia instruction fails unless psr.dfl and psr.dfh are
+ * cleared).  Fortunately, SAL promises not to touch the floating
+ * point regs, so at least we don't have to save f2-f127.
+ */
+#define PSR_BITS_TO_CLEAR						\
+	(IA64_PSR_I | IA64_PSR_IT | IA64_PSR_DT | IA64_PSR_RT |		\
+	 IA64_PSR_DD | IA64_PSR_SS | IA64_PSR_RI | IA64_PSR_ED |	\
+	 IA64_PSR_DFL | IA64_PSR_DFH)
+
+#define PSR_BITS_TO_SET							\
+	(IA64_PSR_BN)
+
+#include <asm/processor.h>
+#include <asm/asmmacro.h>
+
+/*
+ * Inputs:
+ *	in0 = address of function descriptor of ESI routine to call
+ *	in1 = address of array of ESI parameters
+ *
+ * Outputs:
+ *	r8 = result returned by called function
+ */
+GLOBAL_ENTRY(esi_call_phys)
+	.prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(2)
+	alloc loc1=ar.pfs,2,7,8,0
+	ld8 r2=[in0],8			// load ESI function's entry point
+	mov loc0=rp
+	.body
+	;;
+	ld8 out0=[in1],8		// ESI params loaded from array
+	;;				// passing all as inputs doesn't work
+	ld8 out1=[in1],8
+	;;
+	ld8 out2=[in1],8
+	;;
+	ld8 out3=[in1],8
+	;;
+	ld8 out4=[in1],8
+	;;
+	ld8 out5=[in1],8
+	;;
+	ld8 out6=[in1],8
+	;;
+	ld8 out7=[in1]
+	mov loc2=gp			// save global pointer
+	mov loc4=ar.rsc			// save RSE configuration
+	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
+	;;
+	ld8 gp=[in0]			// load ESI function's global pointer
+	movl r16=PSR_BITS_TO_CLEAR
+	mov loc3=psr			// save processor status word
+	movl r17=PSR_BITS_TO_SET
+	;;
+	or loc3=loc3,r17
+	mov b6=r2
+	;;
+	andcm r16=loc3,r16	// get psr with IT, DT, and RT bits cleared
+	br.call.sptk.many rp=ia64_switch_mode_phys
+.ret0:	mov loc5=r19			// old ar.bsp
+	mov loc6=r20			// old sp
+	br.call.sptk.many rp=b6		// call the ESI function
+.ret1:	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
+	mov r16=loc3			// save virtual mode psr
+	mov r19=loc5			// save virtual mode bspstore
+	mov r20=loc6			// save virtual mode sp
+	br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode
+.ret2:	mov ar.rsc=loc4			// restore RSE configuration
+	mov ar.pfs=loc1
+	mov rp=loc0
+	mov gp=loc2
+	br.ret.sptk.many rp
+END(esi_call_phys)
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index bbcfd08378a6..9583f310d486 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -106,5 +106,9 @@ EXPORT_SYMBOL(ia64_spinlock_contention);
 # endif
 #endif
 
+#if defined(CONFIG_IA64_ESI) || defined(CONFIG_IA64_ESI_MODULE)
+extern void esi_call_phys (void);
+EXPORT_SYMBOL_GPL(esi_call_phys);
+#endif
 extern char ia64_ivt[];
 EXPORT_SYMBOL(ia64_ivt);
diff --git a/include/asm-ia64/esi.h b/include/asm-ia64/esi.h
new file mode 100644
index 000000000000..84aac0e0b583
--- /dev/null
+++ b/include/asm-ia64/esi.h
@@ -0,0 +1,30 @@
+/*
+ * ESI service calls.
+ *
+ * Copyright (c) Copyright 2005-2006 Hewlett-Packard Development Company, L.P.
+ * 	Alex Williamson <alex.williamson@hp.com>
+ */
+#ifndef esi_h
+#define esi_h
+
+#include <linux/efi.h>
+
+#define ESI_QUERY			0x00000001
+#define ESI_OPEN_HANDLE			0x02000000
+#define ESI_CLOSE_HANDLE		0x02000001
+
+enum esi_proc_type {
+	ESI_PROC_SERIALIZED,	/* calls need to be serialized */
+	ESI_PROC_MP_SAFE,	/* MP-safe, but not reentrant */
+	ESI_PROC_REENTRANT	/* MP-safe and reentrant */
+};
+
+extern int ia64_esi_init (void);
+extern struct ia64_sal_retval esi_call_phys (void *, u64 *);
+extern int ia64_esi_call(efi_guid_t, struct ia64_sal_retval *,
+			 enum esi_proc_type,
+			 u64, u64, u64, u64, u64, u64, u64, u64);
+extern int ia64_esi_call_phys(efi_guid_t, struct ia64_sal_retval *, u64, u64,
+                              u64, u64, u64, u64, u64, u64);
+
+#endif /* esi_h */
-- 
cgit v1.2.3


From c37bb26654bb8981ea237076e333eb37d4aa2dc6 Mon Sep 17 00:00:00 2001
From: Dennis Munsie <dmunsie@cecropia.com>
Date: Tue, 20 Jun 2006 14:55:55 -0400
Subject: intelfb: add preliminary i2c support

[01/07] i2c: add intelfb bit algorithm id

Adds the intelfb bit algorithm id to i2c-id.h.

Signed-off-by: Dennis Munsie <dmunsie@cecropia.com>
---
 include/linux/i2c-id.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index c8b81f419fd8..cce6074dd754 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -189,6 +189,7 @@
 #define I2C_HW_B_RADEON		0x01001e /* radeon framebuffer driver */
 #define I2C_HW_B_EM28XX		0x01001f /* em28xx video capture cards */
 #define I2C_HW_B_CX2341X	0x010020 /* Conexant CX2341X MPEG encoder cards */
+#define I2C_HW_B_INTELFB	0x010021 /* intel framebuffer driver */
 
 /* --- PCF 8584 based algorithms					*/
 #define I2C_HW_P_LP		0x020000 /* Parallel port interface */
-- 
cgit v1.2.3


From 26c5032eaa64090b2a01973b0c6ea9e7f6a80fa7 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 4 Jul 2006 14:16:28 +1000
Subject: [POWERPC] Add briq support to CHRP

The support for Briq machines has been floating around as patches for
ages. This cleans it up and adds it once for all.

Some of this is based on initial code provided by Karsten Jeppesen
<karsten@jeppesens.com> and mostly rewritten from scratch by me.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/prom_init.c     | 10 +++++++--
 arch/powerpc/platforms/chrp/pci.c   | 42 +++++++++++++++++++++++++++++++++----
 arch/powerpc/platforms/chrp/setup.c | 27 +++++++++++++++++++++++-
 include/asm-powerpc/processor.h     |  1 +
 4 files changed, 73 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index 462bced40c12..90972ef6c471 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -2033,16 +2033,22 @@ static void __init fixup_device_tree_maple(void)
 #endif
 
 #ifdef CONFIG_PPC_CHRP
-/* Pegasos lacks the "ranges" property in the isa node */
+/* Pegasos and BriQ lacks the "ranges" property in the isa node */
 static void __init fixup_device_tree_chrp(void)
 {
 	phandle isa;
 	u32 isa_ranges[6];
+	u32 rloc = 0x01006000; /* IO space; PCI device = 12 */
 	char *name;
 	int rc;
 
 	name = "/pci@80000000/isa@c";
 	isa = call_prom("finddevice", 1, 1, ADDR(name));
+	if (!PHANDLE_VALID(isa)) {
+		name = "/pci@ff500000/isa@6";
+		isa = call_prom("finddevice", 1, 1, ADDR(name));
+		rloc = 0x01003000; /* IO space; PCI device = 6 */
+	}
 	if (!PHANDLE_VALID(isa))
 		return;
 
@@ -2054,7 +2060,7 @@ static void __init fixup_device_tree_chrp(void)
 
 	isa_ranges[0] = 0x1;
 	isa_ranges[1] = 0x0;
-	isa_ranges[2] = 0x01006000;
+	isa_ranges[2] = rloc;
 	isa_ranges[3] = 0x0;
 	isa_ranges[4] = 0x0;
 	isa_ranges[5] = 0x00010000;
diff --git a/arch/powerpc/platforms/chrp/pci.c b/arch/powerpc/platforms/chrp/pci.c
index 6802cdc3168a..6d7ac649b45e 100644
--- a/arch/powerpc/platforms/chrp/pci.c
+++ b/arch/powerpc/platforms/chrp/pci.c
@@ -257,7 +257,7 @@ chrp_find_bridges(void)
 		else
 			printk(KERN_INFO "PCI buses %d..%d",
 			       bus_range[0], bus_range[1]);
-		printk(" controlled by %s", dev->type);
+		printk(" controlled by %s", dev->full_name);
 		if (!is_longtrail)
 			printk(" at %llx", (unsigned long long)r.start);
 		printk("\n");
@@ -289,6 +289,19 @@ chrp_find_bridges(void)
 			setup_indirect_pci(hose, 0xfec00cf8, 0xfee00cfc);
 		} else if (is_pegasos == 2) {
 			setup_peg2(hose, dev);
+		} else if (!strncmp(model, "IBM,CPC710", 10)) {
+			setup_indirect_pci(hose,
+					   r.start + 0x000f8000,
+					   r.start + 0x000f8010);
+			if (index == 0) {
+				dma = get_property(dev, "system-dma-base",&len);
+				if (dma && len >= sizeof(*dma)) {
+					dma = (unsigned int *)
+						(((unsigned long)dma) +
+						len - sizeof(*dma));
+						pci_dram_offset = *dma;
+				}
+			}
 		} else {
 			printk("No methods for %s (model %s), using RTAS\n",
 			       dev->full_name, model);
@@ -306,8 +319,29 @@ chrp_find_bridges(void)
 			printk("pci_dram_offset = %lx\n", pci_dram_offset);
 		}
 	}
+}
+
+/* SL82C105 IDE Control/Status Register */
+#define SL82C105_IDECSR                0x40
+
+/* Fixup for Winbond ATA quirk, required for briq */
+void chrp_pci_fixup_winbond_ata(struct pci_dev *sl82c105)
+{
+	u8 progif;
 
-	/* Do not fixup interrupts from OF tree on pegasos */
-	if (is_pegasos)
-		ppc_md.pcibios_fixup = NULL;
+	/* If non-briq machines need that fixup too, please speak up */
+	if (!machine_is(chrp) || _chrp_type != _CHRP_briq)
+		return;
+
+	if ((sl82c105->class & 5) != 5) {
+		printk("W83C553: Switching SL82C105 IDE to PCI native mode\n");
+		/* Enable SL82C105 PCI native IDE mode */
+		pci_read_config_byte(sl82c105, PCI_CLASS_PROG, &progif);
+		pci_write_config_byte(sl82c105, PCI_CLASS_PROG, progif | 0x05);
+		sl82c105->class |= 0x05;
+		/* Disable SL82C105 second port */
+		pci_write_config_word(sl82c105, SL82C105_IDECSR, 0x0003);
+	}
 }
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_WINBOND, PCI_DEVICE_ID_WINBOND_82C105,
+		chrp_pci_fixup_winbond_ata);
diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c
index 9c08ff322290..be39742db809 100644
--- a/arch/powerpc/platforms/chrp/setup.c
+++ b/arch/powerpc/platforms/chrp/setup.c
@@ -74,6 +74,9 @@ extern irqreturn_t xmon_irq(int, void *, struct pt_regs *);
 
 extern unsigned long loops_per_jiffy;
 
+/* To be replaced by RTAS when available */
+static unsigned int *briq_SPOR;
+
 #ifdef CONFIG_SMP
 extern struct smp_ops_t chrp_smp_ops;
 #endif
@@ -92,6 +95,15 @@ static const char *gg2_cachemodes[4] = {
 	"Disabled", "Write-Through", "Copy-Back", "Transparent Mode"
 };
 
+static const char *chrp_names[] = {
+	"Unknown",
+	"","","",
+	"Motorola",
+	"IBM or Longtrail",
+	"Genesi Pegasos",
+	"Total Impact Briq"
+};
+
 void chrp_show_cpuinfo(struct seq_file *m)
 {
 	int i, sdramen;
@@ -229,6 +241,14 @@ static void __init pegasos_set_l2cr(void)
 	}
 }
 
+static void briq_restart(char *cmd)
+{
+	local_irq_disable();
+	if (briq_SPOR)
+		out_be32(briq_SPOR, 0);
+	for(;;);
+}
+
 void __init chrp_setup_arch(void)
 {
 	struct device_node *root = find_path_device ("/");
@@ -245,11 +265,16 @@ void __init chrp_setup_arch(void)
 		_chrp_type = _CHRP_IBM;
 	} else if (machine && strncmp(machine, "MOT", 3) == 0) {
 		_chrp_type = _CHRP_Motorola;
+	} else if (machine && strncmp(machine, "TotalImpact,BRIQ-1", 18) == 0) {
+		_chrp_type = _CHRP_briq;
+		/* Map the SPOR register on briq and change the restart hook */
+		briq_SPOR = (unsigned int *)ioremap(0xff0000e8, 4);
+		ppc_md.restart = briq_restart;
 	} else {
 		/* Let's assume it is an IBM chrp if all else fails */
 		_chrp_type = _CHRP_IBM;
 	}
-	printk("chrp type = %x\n", _chrp_type);
+	printk("chrp type = %x [%s]\n", _chrp_type, chrp_names[_chrp_type]);
 
 	rtas_initialize();
 	if (rtas_token("display-character") >= 0)
diff --git a/include/asm-powerpc/processor.h b/include/asm-powerpc/processor.h
index 22e54a2a6604..6cb6fb19e57f 100644
--- a/include/asm-powerpc/processor.h
+++ b/include/asm-powerpc/processor.h
@@ -32,6 +32,7 @@
 #define _CHRP_Motorola	0x04	/* motorola chrp, the cobra */
 #define _CHRP_IBM	0x05	/* IBM chrp, the longtrail and longtrail 2 */
 #define _CHRP_Pegasos	0x06	/* Genesi/bplan's Pegasos and Pegasos2 */
+#define _CHRP_briq	0x07	/* TotalImpact's briQ */
 
 #if defined(__KERNEL__) && defined(CONFIG_PPC32)
 
-- 
cgit v1.2.3


From b5a1a9abe1a54ba40a9612001920f98bbdd0c56f Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@ozlabs.org>
Date: Tue, 4 Jul 2006 16:46:44 +1000
Subject: [POWERPC] Use const qualifiers for prom parsing utilites

The of_bus callbacks map and get_flags can be constified, as they don't
alter the range or addr arguments. of_dump_addr and of_read_addr can
also be constified.

Built for 32- and 64-bit powerpc

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/prom_parse.c | 22 ++++++++++++----------
 include/asm-powerpc/prom.h       |  2 +-
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/prom_parse.c b/arch/powerpc/kernel/prom_parse.c
index 21009b1f7869..e9960170667b 100644
--- a/arch/powerpc/kernel/prom_parse.c
+++ b/arch/powerpc/kernel/prom_parse.c
@@ -27,7 +27,7 @@
 
 /* Debug utility */
 #ifdef DEBUG
-static void of_dump_addr(const char *s, u32 *addr, int na)
+static void of_dump_addr(const char *s, const u32 *addr, int na)
 {
 	printk("%s", s);
 	while(na--)
@@ -35,7 +35,7 @@ static void of_dump_addr(const char *s, u32 *addr, int na)
 	printk("\n");
 }
 #else
-static void of_dump_addr(const char *s, u32 *addr, int na) { }
+static void of_dump_addr(const char *s, const u32 *addr, int na) { }
 #endif
 
 
@@ -46,9 +46,10 @@ struct of_bus {
 	int		(*match)(struct device_node *parent);
 	void		(*count_cells)(struct device_node *child,
 				       int *addrc, int *sizec);
-	u64		(*map)(u32 *addr, u32 *range, int na, int ns, int pna);
+	u64		(*map)(u32 *addr, const u32 *range,
+				int na, int ns, int pna);
 	int		(*translate)(u32 *addr, u64 offset, int na);
-	unsigned int	(*get_flags)(u32 *addr);
+	unsigned int	(*get_flags)(const u32 *addr);
 };
 
 
@@ -65,7 +66,8 @@ static void of_bus_default_count_cells(struct device_node *dev,
 		*sizec = prom_n_size_cells(dev);
 }
 
-static u64 of_bus_default_map(u32 *addr, u32 *range, int na, int ns, int pna)
+static u64 of_bus_default_map(u32 *addr, const u32 *range,
+		int na, int ns, int pna)
 {
 	u64 cp, s, da;
 
@@ -93,7 +95,7 @@ static int of_bus_default_translate(u32 *addr, u64 offset, int na)
 	return 0;
 }
 
-static unsigned int of_bus_default_get_flags(u32 *addr)
+static unsigned int of_bus_default_get_flags(const u32 *addr)
 {
 	return IORESOURCE_MEM;
 }
@@ -118,7 +120,7 @@ static void of_bus_pci_count_cells(struct device_node *np,
 		*sizec = 2;
 }
 
-static u64 of_bus_pci_map(u32 *addr, u32 *range, int na, int ns, int pna)
+static u64 of_bus_pci_map(u32 *addr, const u32 *range, int na, int ns, int pna)
 {
 	u64 cp, s, da;
 
@@ -143,7 +145,7 @@ static int of_bus_pci_translate(u32 *addr, u64 offset, int na)
 	return of_bus_default_translate(addr + 1, offset, na - 1);
 }
 
-static unsigned int of_bus_pci_get_flags(u32 *addr)
+static unsigned int of_bus_pci_get_flags(const u32 *addr)
 {
 	unsigned int flags = 0;
 	u32 w = addr[0];
@@ -178,7 +180,7 @@ static void of_bus_isa_count_cells(struct device_node *child,
 		*sizec = 1;
 }
 
-static u64 of_bus_isa_map(u32 *addr, u32 *range, int na, int ns, int pna)
+static u64 of_bus_isa_map(u32 *addr, const u32 *range, int na, int ns, int pna)
 {
 	u64 cp, s, da;
 
@@ -203,7 +205,7 @@ static int of_bus_isa_translate(u32 *addr, u64 offset, int na)
 	return of_bus_default_translate(addr + 1, offset, na - 1);
 }
 
-static unsigned int of_bus_isa_get_flags(u32 *addr)
+static unsigned int of_bus_isa_get_flags(const u32 *addr)
 {
 	unsigned int flags = 0;
 	u32 w = addr[0];
diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h
index b095a285c84b..56f6ea0c76de 100644
--- a/include/asm-powerpc/prom.h
+++ b/include/asm-powerpc/prom.h
@@ -198,7 +198,7 @@ extern int release_OF_resource(struct device_node* node, int index);
 
 
 /* Helper to read a big number */
-static inline u64 of_read_number(u32 *cell, int size)
+static inline u64 of_read_number(const u32 *cell, int size)
 {
 	u64 r = 0;
 	while (size--)
-- 
cgit v1.2.3


From dac411e7aa92d23dadbcb8721845ab88577294c7 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 13 Jul 2006 17:52:04 +1000
Subject: [POWERPC] iseries: Move e2a()/strne2a() into their only caller

The ASCII -> EBCDIC functions, e2a() and strne2a() are now only used in
dt.c, so move them in there.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/lib/Makefile           |   1 -
 arch/powerpc/lib/e2a.c              | 116 ------------------------------------
 arch/powerpc/platforms/iseries/dt.c |  98 +++++++++++++++++++++++++++++-
 include/asm-powerpc/system.h        |   5 --
 4 files changed, 97 insertions(+), 123 deletions(-)
 delete mode 100644 arch/powerpc/lib/e2a.c

(limited to 'include')

diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
index ff7096458249..336dd191f768 100644
--- a/arch/powerpc/lib/Makefile
+++ b/arch/powerpc/lib/Makefile
@@ -14,7 +14,6 @@ endif
 obj-$(CONFIG_PPC64)	+= checksum_64.o copypage_64.o copyuser_64.o \
 			   memcpy_64.o usercopy_64.o mem_64.o string.o \
 			   strcase.o
-obj-$(CONFIG_PPC_ISERIES) += e2a.o
 obj-$(CONFIG_XMON)	+= sstep.o
 
 ifeq ($(CONFIG_PPC64),y)
diff --git a/arch/powerpc/lib/e2a.c b/arch/powerpc/lib/e2a.c
deleted file mode 100644
index 4b72ed8fd50e..000000000000
--- a/arch/powerpc/lib/e2a.c
+++ /dev/null
@@ -1,116 +0,0 @@
-/*
- *  EBCDIC to ASCII conversion
- *
- * This function moved here from arch/powerpc/platforms/iseries/viopath.c 
- *
- * (C) Copyright 2000-2004 IBM Corporation
- *
- * This program is free software;  you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation; either version 2 of the
- * License, or (at your option) anyu later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software Foundation,
- * Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- *
- */
-
-#include <linux/module.h>
-
-unsigned char e2a(unsigned char x)
-{
-	switch (x) {
-	case 0xF0:
-		return '0';
-	case 0xF1:
-		return '1';
-	case 0xF2:
-		return '2';
-	case 0xF3:
-		return '3';
-	case 0xF4:
-		return '4';
-	case 0xF5:
-		return '5';
-	case 0xF6:
-		return '6';
-	case 0xF7:
-		return '7';
-	case 0xF8:
-		return '8';
-	case 0xF9:
-		return '9';
-	case 0xC1:
-		return 'A';
-	case 0xC2:
-		return 'B';
-	case 0xC3:
-		return 'C';
-	case 0xC4:
-		return 'D';
-	case 0xC5:
-		return 'E';
-	case 0xC6:
-		return 'F';
-	case 0xC7:
-		return 'G';
-	case 0xC8:
-		return 'H';
-	case 0xC9:
-		return 'I';
-	case 0xD1:
-		return 'J';
-	case 0xD2:
-		return 'K';
-	case 0xD3:
-		return 'L';
-	case 0xD4:
-		return 'M';
-	case 0xD5:
-		return 'N';
-	case 0xD6:
-		return 'O';
-	case 0xD7:
-		return 'P';
-	case 0xD8:
-		return 'Q';
-	case 0xD9:
-		return 'R';
-	case 0xE2:
-		return 'S';
-	case 0xE3:
-		return 'T';
-	case 0xE4:
-		return 'U';
-	case 0xE5:
-		return 'V';
-	case 0xE6:
-		return 'W';
-	case 0xE7:
-		return 'X';
-	case 0xE8:
-		return 'Y';
-	case 0xE9:
-		return 'Z';
-	}
-	return ' ';
-}
-EXPORT_SYMBOL(e2a);
-
-unsigned char* strne2a(unsigned char *dest, const unsigned char *src, size_t n)
-{
-	int i;
-
-	n = strnlen(src, n);
-
-	for (i = 0; i < n; i++)
-		dest[i] = e2a(src[i]);
-
-	return dest;
-}
diff --git a/arch/powerpc/platforms/iseries/dt.c b/arch/powerpc/platforms/iseries/dt.c
index d194140c1ebf..39c676ab9d6d 100644
--- a/arch/powerpc/platforms/iseries/dt.c
+++ b/arch/powerpc/platforms/iseries/dt.c
@@ -1,5 +1,6 @@
 /*
- *    Copyright (c) 2005-2006 Michael Ellerman, IBM Corporation
+ *    Copyright (C) 2005-2006 Michael Ellerman, IBM Corporation
+ *    Copyright (C) 2000-2004, IBM Corporation
  *
  *    Description:
  *      This file contains all the routines to build a flattened device
@@ -76,6 +77,101 @@ static char __initdata device_type_pci[] = "pci";
 static char __initdata device_type_vdevice[] = "vdevice";
 static char __initdata device_type_vscsi[] = "vscsi";
 
+
+/* EBCDIC to ASCII conversion routines */
+
+unsigned char e2a(unsigned char x)
+{
+	switch (x) {
+	case 0xF0:
+		return '0';
+	case 0xF1:
+		return '1';
+	case 0xF2:
+		return '2';
+	case 0xF3:
+		return '3';
+	case 0xF4:
+		return '4';
+	case 0xF5:
+		return '5';
+	case 0xF6:
+		return '6';
+	case 0xF7:
+		return '7';
+	case 0xF8:
+		return '8';
+	case 0xF9:
+		return '9';
+	case 0xC1:
+		return 'A';
+	case 0xC2:
+		return 'B';
+	case 0xC3:
+		return 'C';
+	case 0xC4:
+		return 'D';
+	case 0xC5:
+		return 'E';
+	case 0xC6:
+		return 'F';
+	case 0xC7:
+		return 'G';
+	case 0xC8:
+		return 'H';
+	case 0xC9:
+		return 'I';
+	case 0xD1:
+		return 'J';
+	case 0xD2:
+		return 'K';
+	case 0xD3:
+		return 'L';
+	case 0xD4:
+		return 'M';
+	case 0xD5:
+		return 'N';
+	case 0xD6:
+		return 'O';
+	case 0xD7:
+		return 'P';
+	case 0xD8:
+		return 'Q';
+	case 0xD9:
+		return 'R';
+	case 0xE2:
+		return 'S';
+	case 0xE3:
+		return 'T';
+	case 0xE4:
+		return 'U';
+	case 0xE5:
+		return 'V';
+	case 0xE6:
+		return 'W';
+	case 0xE7:
+		return 'X';
+	case 0xE8:
+		return 'Y';
+	case 0xE9:
+		return 'Z';
+	}
+	return ' ';
+}
+EXPORT_SYMBOL(e2a);
+
+unsigned char* strne2a(unsigned char *dest, const unsigned char *src, size_t n)
+{
+	int i;
+
+	n = strnlen(src, n);
+
+	for (i = 0; i < n; i++)
+		dest[i] = e2a(src[i]);
+
+	return dest;
+}
+
 static struct iseries_flat_dt * __init dt_init(void)
 {
 	struct iseries_flat_dt *dt;
diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h
index d075725bf444..5deb7bc7bb1f 100644
--- a/include/asm-powerpc/system.h
+++ b/include/asm-powerpc/system.h
@@ -169,11 +169,6 @@ extern u32 booke_wdt_enabled;
 extern u32 booke_wdt_period;
 #endif /* CONFIG_BOOKE_WDT */
 
-/* EBCDIC -> ASCII conversion for [0-9A-Z] on iSeries */
-extern unsigned char e2a(unsigned char);
-extern unsigned char* strne2a(unsigned char *dest,
-		const unsigned char *src, size_t n);
-
 struct device_node;
 extern void note_scsi_host(struct device_node *, void *);
 
-- 
cgit v1.2.3


From c59acae85409fdf5d7574e90009c8410daf38938 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 13 Jul 2006 17:52:09 +1000
Subject: [POWERPC] iseries: Make ItExtVpdPanel private to iSeries

No one outside platforms/iseries needs ItExtVpdPanel anymore, so move
it in there. It used to be needed by lparcfg, and so was exported, but
isn't needed anymore, so unexport it.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/kernel/lparcfg.c                     |  1 -
 arch/powerpc/platforms/iseries/dt.c               |  2 +-
 arch/powerpc/platforms/iseries/it_exp_vpd_panel.h | 51 +++++++++++++++++++++++
 arch/powerpc/platforms/iseries/lpardata.c         |  3 +-
 include/asm-powerpc/iseries/it_exp_vpd_panel.h    | 51 -----------------------
 5 files changed, 53 insertions(+), 55 deletions(-)
 create mode 100644 arch/powerpc/platforms/iseries/it_exp_vpd_panel.h
 delete mode 100644 include/asm-powerpc/iseries/it_exp_vpd_panel.h

(limited to 'include')

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 23f34daa044a..2d94b372d49b 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -32,7 +32,6 @@
 #include <asm/rtas.h>
 #include <asm/system.h>
 #include <asm/time.h>
-#include <asm/iseries/it_exp_vpd_panel.h>
 #include <asm/prom.h>
 #include <asm/vdso_datapage.h>
 
diff --git a/arch/powerpc/platforms/iseries/dt.c b/arch/powerpc/platforms/iseries/dt.c
index c5f59a8b9ef3..aa582262aefd 100644
--- a/arch/powerpc/platforms/iseries/dt.c
+++ b/arch/powerpc/platforms/iseries/dt.c
@@ -34,13 +34,13 @@
 #include <asm/iseries/hv_types.h>
 #include <asm/iseries/hv_lp_config.h>
 #include <asm/iseries/hv_call_xm.h>
-#include <asm/iseries/it_exp_vpd_panel.h>
 #include <asm/udbg.h>
 
 #include "processor_vpd.h"
 #include "call_hpt.h"
 #include "call_pci.h"
 #include "pci.h"
+#include "it_exp_vpd_panel.h"
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
diff --git a/arch/powerpc/platforms/iseries/it_exp_vpd_panel.h b/arch/powerpc/platforms/iseries/it_exp_vpd_panel.h
new file mode 100644
index 000000000000..6de9097b7f57
--- /dev/null
+++ b/arch/powerpc/platforms/iseries/it_exp_vpd_panel.h
@@ -0,0 +1,51 @@
+/*
+ * Copyright (C) 2002  Dave Boutcher IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+#ifndef _PLATFORMS_ISERIES_IT_EXT_VPD_PANEL_H
+#define _PLATFORMS_ISERIES_IT_EXT_VPD_PANEL_H
+
+/*
+ *	This struct maps the panel information
+ *
+ * Warning:
+ *	This data must match the architecture for the panel information
+ */
+
+#include <asm/types.h>
+
+struct ItExtVpdPanel {
+	/* Definition of the Extended Vpd On Panel Data Area */
+	char	systemSerial[8];
+	char	mfgID[4];
+	char	reserved1[24];
+	char	machineType[4];
+	char	systemID[6];
+	char	somUniqueCnt[4];
+	char	serialNumberCount;
+	char	reserved2[7];
+	u16	bbu3;
+	u16	bbu2;
+	u16	bbu1;
+	char	xLocationLabel[8];
+	u8	xRsvd1[6];
+	u16	xFrameId;
+	u8	xRsvd2[48];
+};
+
+extern struct ItExtVpdPanel	xItExtVpdPanel;
+
+#endif /* _PLATFORMS_ISERIES_IT_EXT_VPD_PANEL_H */
diff --git a/arch/powerpc/platforms/iseries/lpardata.c b/arch/powerpc/platforms/iseries/lpardata.c
index a7769445d6c7..13e9bd132254 100644
--- a/arch/powerpc/platforms/iseries/lpardata.c
+++ b/arch/powerpc/platforms/iseries/lpardata.c
@@ -18,7 +18,6 @@
 #include <asm/iseries/it_lp_reg_save.h>
 #include <asm/paca.h>
 #include <asm/iseries/lpar_map.h>
-#include <asm/iseries/it_exp_vpd_panel.h>
 #include <asm/iseries/it_lp_queue.h>
 
 #include "naca.h"
@@ -27,6 +26,7 @@
 #include "ipl_parms.h"
 #include "processor_vpd.h"
 #include "release_data.h"
+#include "it_exp_vpd_panel.h"
 
 /* The HvReleaseData is the root of the information shared between
  * the hypervisor and Linux.
@@ -134,7 +134,6 @@ struct ItIplParmsReal xItIplParmsReal __attribute__((__section__(".data")));
 
 /* May be filled in by the hypervisor so cannot end up in the BSS */
 struct ItExtVpdPanel xItExtVpdPanel __attribute__((__section__(".data")));
-EXPORT_SYMBOL(xItExtVpdPanel);
 
 #define maxPhysicalProcessors 32
 
diff --git a/include/asm-powerpc/iseries/it_exp_vpd_panel.h b/include/asm-powerpc/iseries/it_exp_vpd_panel.h
deleted file mode 100644
index 304a609ae21a..000000000000
--- a/include/asm-powerpc/iseries/it_exp_vpd_panel.h
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * Copyright (C) 2002  Dave Boutcher IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
- */
-#ifndef _ASM_POWERPC_ISERIES_IT_EXT_VPD_PANEL_H
-#define _ASM_POWERPC_ISERIES_IT_EXT_VPD_PANEL_H
-
-/*
- *	This struct maps the panel information
- *
- * Warning:
- *	This data must match the architecture for the panel information
- */
-
-#include <asm/types.h>
-
-struct ItExtVpdPanel {
-	/* Definition of the Extended Vpd On Panel Data Area */
-	char	systemSerial[8];
-	char	mfgID[4];
-	char	reserved1[24];
-	char	machineType[4];
-	char	systemID[6];
-	char	somUniqueCnt[4];
-	char	serialNumberCount;
-	char	reserved2[7];
-	u16	bbu3;
-	u16	bbu2;
-	u16	bbu1;
-	char	xLocationLabel[8];
-	u8	xRsvd1[6];
-	u16	xFrameId;
-	u8	xRsvd2[48];
-};
-
-extern struct ItExtVpdPanel	xItExtVpdPanel;
-
-#endif /* _ASM_POWERPC_ISERIES_IT_EXT_VPD_PANEL_H */
-- 
cgit v1.2.3


From a2ced11b6af59854cc2a2791dccd8b6c0da2f733 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 13 Jul 2006 17:52:12 +1000
Subject: [POWERPC] iseries: Make HvLpConfig_get(Primary)LpIndex functions

HvLpConfig_get(Primary)LpIndex are currently static inlines that return
fields from the itLpNaca, if we make them real functions we can make the
itLpNaca private to iSeries.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/platforms/iseries/hvlpconfig.c | 13 +++++++++++++
 arch/powerpc/platforms/iseries/setup.c      |  1 +
 include/asm-powerpc/iseries/hv_lp_config.h  | 13 ++-----------
 3 files changed, 16 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/iseries/hvlpconfig.c b/arch/powerpc/platforms/iseries/hvlpconfig.c
index 663a1affb4bb..cfcedaae6ea2 100644
--- a/arch/powerpc/platforms/iseries/hvlpconfig.c
+++ b/arch/powerpc/platforms/iseries/hvlpconfig.c
@@ -18,9 +18,22 @@
 
 #include <linux/module.h>
 #include <asm/iseries/hv_lp_config.h>
+#include <asm/iseries/it_lp_naca.h>
 
 HvLpIndex HvLpConfig_getLpIndex_outline(void)
 {
 	return HvLpConfig_getLpIndex();
 }
 EXPORT_SYMBOL(HvLpConfig_getLpIndex_outline);
+
+HvLpIndex HvLpConfig_getLpIndex(void)
+{
+	return itLpNaca.xLpIndex;
+}
+EXPORT_SYMBOL(HvLpConfig_getLpIndex);
+
+HvLpIndex HvLpConfig_getPrimaryLpIndex(void)
+{
+	return itLpNaca.xPrimaryLpIndex;
+}
+EXPORT_SYMBOL_GPL(HvLpConfig_getPrimaryLpIndex);
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index c9605d773a77..c34299b18d8b 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -48,6 +48,7 @@
 #include <asm/iseries/hv_call_event.h>
 #include <asm/iseries/hv_call_xm.h>
 #include <asm/iseries/it_lp_queue.h>
+#include <asm/iseries/it_lp_naca.h>
 #include <asm/iseries/mf.h>
 #include <asm/iseries/hv_lp_event.h>
 #include <asm/iseries/lpar_map.h>
diff --git a/include/asm-powerpc/iseries/hv_lp_config.h b/include/asm-powerpc/iseries/hv_lp_config.h
index df8b20739719..a006fd1e4a2c 100644
--- a/include/asm-powerpc/iseries/hv_lp_config.h
+++ b/include/asm-powerpc/iseries/hv_lp_config.h
@@ -25,7 +25,6 @@
 
 #include <asm/iseries/hv_call_sc.h>
 #include <asm/iseries/hv_types.h>
-#include <asm/iseries/it_lp_naca.h>
 
 enum {
 	HvCallCfg_Cur	= 0,
@@ -44,16 +43,8 @@ enum {
 #define HvCallCfgGetHostingLpIndex			HvCallCfg + 32
 
 extern HvLpIndex HvLpConfig_getLpIndex_outline(void);
-
-static inline HvLpIndex	HvLpConfig_getLpIndex(void)
-{
-	return itLpNaca.xLpIndex;
-}
-
-static inline HvLpIndex	HvLpConfig_getPrimaryLpIndex(void)
-{
-	return itLpNaca.xPrimaryLpIndex;
-}
+extern HvLpIndex HvLpConfig_getLpIndex(void);
+extern HvLpIndex HvLpConfig_getPrimaryLpIndex(void);
 
 static inline u64 HvLpConfig_getMsChunks(void)
 {
-- 
cgit v1.2.3


From 06a36db1d712242a00cb30aaebdd088b4be28082 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 13 Jul 2006 17:52:17 +1000
Subject: [POWERPC] iseries: Move ItLpNaca into platforms/iseries

Move ItLpNaca into platforms/iseries now that it's not used elsewhere.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/kernel/setup_64.c              |  1 -
 arch/powerpc/platforms/iseries/hvlpconfig.c |  2 +-
 arch/powerpc/platforms/iseries/it_lp_naca.h | 80 +++++++++++++++++++++++++++++
 arch/powerpc/platforms/iseries/lpardata.c   |  3 +-
 arch/powerpc/platforms/iseries/lpevents.c   |  2 +-
 arch/powerpc/platforms/iseries/setup.c      |  2 +-
 include/asm-powerpc/iseries/it_lp_naca.h    | 80 -----------------------------
 7 files changed, 84 insertions(+), 86 deletions(-)
 create mode 100644 arch/powerpc/platforms/iseries/it_lp_naca.h
 delete mode 100644 include/asm-powerpc/iseries/it_lp_naca.h

(limited to 'include')

diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index fd1785e4c9bb..e2447aef3a8f 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -56,7 +56,6 @@
 #include <asm/page.h>
 #include <asm/mmu.h>
 #include <asm/lmb.h>
-#include <asm/iseries/it_lp_naca.h>
 #include <asm/firmware.h>
 #include <asm/xmon.h>
 #include <asm/udbg.h>
diff --git a/arch/powerpc/platforms/iseries/hvlpconfig.c b/arch/powerpc/platforms/iseries/hvlpconfig.c
index cfcedaae6ea2..f0475f0b1853 100644
--- a/arch/powerpc/platforms/iseries/hvlpconfig.c
+++ b/arch/powerpc/platforms/iseries/hvlpconfig.c
@@ -18,7 +18,7 @@
 
 #include <linux/module.h>
 #include <asm/iseries/hv_lp_config.h>
-#include <asm/iseries/it_lp_naca.h>
+#include "it_lp_naca.h"
 
 HvLpIndex HvLpConfig_getLpIndex_outline(void)
 {
diff --git a/arch/powerpc/platforms/iseries/it_lp_naca.h b/arch/powerpc/platforms/iseries/it_lp_naca.h
new file mode 100644
index 000000000000..9bbf58986819
--- /dev/null
+++ b/arch/powerpc/platforms/iseries/it_lp_naca.h
@@ -0,0 +1,80 @@
+/*
+ * Copyright (C) 2001  Mike Corrigan IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+#ifndef _PLATFORMS_ISERIES_IT_LP_NACA_H
+#define _PLATFORMS_ISERIES_IT_LP_NACA_H
+
+#include <linux/types.h>
+
+/*
+ *	This control block contains the data that is shared between the
+ *	hypervisor (PLIC) and the OS.
+ */
+
+struct ItLpNaca {
+// CACHE_LINE_1 0x0000 - 0x007F Contains read-only data
+	u32	xDesc;			// Eye catcher			x00-x03
+	u16	xSize;			// Size of this class		x04-x05
+	u16	xIntHdlrOffset;		// Offset to IntHdlr array	x06-x07
+	u8	xMaxIntHdlrEntries;	// Number of entries in array	x08-x08
+	u8	xPrimaryLpIndex;	// LP Index of Primary		x09-x09
+	u8	xServiceLpIndex;	// LP Ind of Service Focal Pointx0A-x0A
+	u8	xLpIndex;		// LP Index			x0B-x0B
+	u16	xMaxLpQueues;		// Number of allocated queues	x0C-x0D
+	u16	xLpQueueOffset;		// Offset to start of LP queues	x0E-x0F
+	u8	xPirEnvironMode;	// Piranha or hardware		x10-x10
+	u8	xPirConsoleMode;	// Piranha console indicator	x11-x11
+	u8	xPirDasdMode;		// Piranha dasd indicator	x12-x12
+	u8	xRsvd1_0[5];		// Reserved for Piranha related	x13-x17
+	u8	flags;			// flags, see below		x18-x1F
+	u8	xSpVpdFormat;		// VPD areas are in CSP format	...
+	u8	xIntProcRatio;		// Ratio of int procs to procs	...
+	u8	xRsvd1_2[5];		// Reserved			...
+	u16	xRsvd1_3;		// Reserved			x20-x21
+	u16	xPlicVrmIndex;		// VRM index of PLIC		x22-x23
+	u16	xMinSupportedSlicVrmInd;// Min supported OS VRM index	x24-x25
+	u16	xMinCompatableSlicVrmInd;// Min compatible OS VRM index x26-x27
+	u64	xLoadAreaAddr;		// ER address of load area	x28-x2F
+	u32	xLoadAreaChunks;	// Chunks for the load area	x30-x33
+	u32	xPaseSysCallCRMask;	// Mask used to test CR before  x34-x37
+					// doing an ASR switch on PASE
+					// system call.
+	u64	xSlicSegmentTablePtr;	// Pointer to Slic seg table.   x38-x3f
+	u8	xRsvd1_4[64];		//				x40-x7F
+
+// CACHE_LINE_2 0x0080 - 0x00FF Contains local read-write data
+	u8	xRsvd2_0[128];		// Reserved			x00-x7F
+
+// CACHE_LINE_3-6 0x0100 - 0x02FF Contains LP Queue indicators
+// NB: Padding required to keep xInterrruptHdlr at x300 which is required
+// for v4r4 PLIC.
+	u8	xOldLpQueue[128];	// LP Queue needed for v4r4	100-17F
+	u8	xRsvd3_0[384];		// Reserved			180-2FF
+
+// CACHE_LINE_7-8 0x0300 - 0x03FF Contains the address of the OS interrupt
+//  handlers
+	u64	xInterruptHdlr[32];	// Interrupt handlers		300-x3FF
+};
+
+extern struct ItLpNaca		itLpNaca;
+
+#define ITLPNACA_LPAR		0x80	/* Is LPAR installed on the system */
+#define ITLPNACA_PARTITIONED	0x40	/* Is the system partitioned */
+#define ITLPNACA_HWSYNCEDTBS	0x20	/* Hardware synced TBs */
+#define ITLPNACA_HMTINT		0x10	/* Utilize MHT for interrupts */
+
+#endif /* _PLATFORMS_ISERIES_IT_LP_NACA_H */
diff --git a/arch/powerpc/platforms/iseries/lpardata.c b/arch/powerpc/platforms/iseries/lpardata.c
index 13e9bd132254..8162049bb04d 100644
--- a/arch/powerpc/platforms/iseries/lpardata.c
+++ b/arch/powerpc/platforms/iseries/lpardata.c
@@ -13,7 +13,6 @@
 #include <asm/processor.h>
 #include <asm/ptrace.h>
 #include <asm/abs_addr.h>
-#include <asm/iseries/it_lp_naca.h>
 #include <asm/lppaca.h>
 #include <asm/iseries/it_lp_reg_save.h>
 #include <asm/paca.h>
@@ -27,6 +26,7 @@
 #include "processor_vpd.h"
 #include "release_data.h"
 #include "it_exp_vpd_panel.h"
+#include "it_lp_naca.h"
 
 /* The HvReleaseData is the root of the information shared between
  * the hypervisor and Linux.
@@ -127,7 +127,6 @@ struct ItLpNaca itLpNaca = {
 		(u64)instruction_access_slb_iSeries /* 0x480 I-SLB */
 	}
 };
-EXPORT_SYMBOL(itLpNaca);
 
 /* May be filled in by the hypervisor so cannot end up in the BSS */
 struct ItIplParmsReal xItIplParmsReal __attribute__((__section__(".data")));
diff --git a/arch/powerpc/platforms/iseries/lpevents.c b/arch/powerpc/platforms/iseries/lpevents.c
index 2a9f81ea27d6..98c1c2440aad 100644
--- a/arch/powerpc/platforms/iseries/lpevents.c
+++ b/arch/powerpc/platforms/iseries/lpevents.c
@@ -20,7 +20,7 @@
 #include <asm/iseries/it_lp_queue.h>
 #include <asm/iseries/hv_lp_event.h>
 #include <asm/iseries/hv_call_event.h>
-#include <asm/iseries/it_lp_naca.h>
+#include "it_lp_naca.h"
 
 /*
  * The LpQueue is used to pass event data from the hypervisor to
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index c34299b18d8b..7f1953066ff8 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -48,7 +48,6 @@
 #include <asm/iseries/hv_call_event.h>
 #include <asm/iseries/hv_call_xm.h>
 #include <asm/iseries/it_lp_queue.h>
-#include <asm/iseries/it_lp_naca.h>
 #include <asm/iseries/mf.h>
 #include <asm/iseries/hv_lp_event.h>
 #include <asm/iseries/lpar_map.h>
@@ -60,6 +59,7 @@
 #include "irq.h"
 #include "vpd_areas.h"
 #include "processor_vpd.h"
+#include "it_lp_naca.h"
 #include "main_store.h"
 #include "call_sm.h"
 #include "call_hpt.h"
diff --git a/include/asm-powerpc/iseries/it_lp_naca.h b/include/asm-powerpc/iseries/it_lp_naca.h
deleted file mode 100644
index 4fdcf052927f..000000000000
--- a/include/asm-powerpc/iseries/it_lp_naca.h
+++ /dev/null
@@ -1,80 +0,0 @@
-/*
- * Copyright (C) 2001  Mike Corrigan IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
- */
-#ifndef _ASM_POWERPC_ISERIES_IT_LP_NACA_H
-#define _ASM_POWERPC_ISERIES_IT_LP_NACA_H
-
-#include <linux/types.h>
-
-/*
- *	This control block contains the data that is shared between the
- *	hypervisor (PLIC) and the OS.
- */
-
-struct ItLpNaca {
-// CACHE_LINE_1 0x0000 - 0x007F Contains read-only data
-	u32	xDesc;			// Eye catcher			x00-x03
-	u16	xSize;			// Size of this class		x04-x05
-	u16	xIntHdlrOffset;		// Offset to IntHdlr array	x06-x07
-	u8	xMaxIntHdlrEntries;	// Number of entries in array	x08-x08
-	u8	xPrimaryLpIndex;	// LP Index of Primary		x09-x09
-	u8	xServiceLpIndex;	// LP Ind of Service Focal Pointx0A-x0A
-	u8	xLpIndex;		// LP Index			x0B-x0B
-	u16	xMaxLpQueues;		// Number of allocated queues	x0C-x0D
-	u16	xLpQueueOffset;		// Offset to start of LP queues	x0E-x0F
-	u8	xPirEnvironMode;	// Piranha or hardware		x10-x10
-	u8	xPirConsoleMode;	// Piranha console indicator	x11-x11
-	u8	xPirDasdMode;		// Piranha dasd indicator	x12-x12
-	u8	xRsvd1_0[5];		// Reserved for Piranha related	x13-x17
-	u8	flags;			// flags, see below		x18-x1F
-	u8	xSpVpdFormat;		// VPD areas are in CSP format	...
-	u8	xIntProcRatio;		// Ratio of int procs to procs	...
-	u8	xRsvd1_2[5];		// Reserved			...
-	u16	xRsvd1_3;		// Reserved			x20-x21
-	u16	xPlicVrmIndex;		// VRM index of PLIC		x22-x23
-	u16	xMinSupportedSlicVrmInd;// Min supported OS VRM index	x24-x25
-	u16	xMinCompatableSlicVrmInd;// Min compatible OS VRM index x26-x27
-	u64	xLoadAreaAddr;		// ER address of load area	x28-x2F
-	u32	xLoadAreaChunks;	// Chunks for the load area	x30-x33
-	u32	xPaseSysCallCRMask;	// Mask used to test CR before  x34-x37
-					// doing an ASR switch on PASE
-					// system call.
-	u64	xSlicSegmentTablePtr;	// Pointer to Slic seg table.   x38-x3f
-	u8	xRsvd1_4[64];		//				x40-x7F
-
-// CACHE_LINE_2 0x0080 - 0x00FF Contains local read-write data
-	u8	xRsvd2_0[128];		// Reserved			x00-x7F
-
-// CACHE_LINE_3-6 0x0100 - 0x02FF Contains LP Queue indicators
-// NB: Padding required to keep xInterrruptHdlr at x300 which is required
-// for v4r4 PLIC.
-	u8	xOldLpQueue[128];	// LP Queue needed for v4r4	100-17F
-	u8	xRsvd3_0[384];		// Reserved			180-2FF
-
-// CACHE_LINE_7-8 0x0300 - 0x03FF Contains the address of the OS interrupt
-//  handlers
-	u64	xInterruptHdlr[32];	// Interrupt handlers		300-x3FF
-};
-
-extern struct ItLpNaca		itLpNaca;
-
-#define ITLPNACA_LPAR		0x80	/* Is LPAR installed on the system */
-#define ITLPNACA_PARTITIONED	0x40	/* Is the system partitioned */
-#define ITLPNACA_HWSYNCEDTBS	0x20	/* Hardware synced TBs */
-#define ITLPNACA_HMTINT		0x10	/* Utilize MHT for interrupts */
-
-#endif /* _ASM_POWERPC_ISERIES_IT_LP_NACA_H */
-- 
cgit v1.2.3


From 8bff05b052db7a4cfaaf0eee7f8145600548e9c9 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 13 Jul 2006 18:51:22 +1000
Subject: [POWERPC] iseries: A new iSeries console

This driver uses the hvc_console.c infrastructure that is used by the
pSeries virtual and RTAS consoles.  This will allow us to make viocons.c
obsolete and is another step along the way to a combined kernel (as
viocons could not coexist with CONFIG_VT).

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/platforms/iseries/Kconfig |   8 +-
 arch/powerpc/platforms/iseries/dt.c    |   3 +-
 drivers/char/Kconfig                   |   7 +
 drivers/char/Makefile                  |   1 +
 drivers/char/hvc_iseries.c             | 593 +++++++++++++++++++++++++++++++++
 drivers/char/viocons.c                 |  31 +-
 include/asm-powerpc/iseries/vio.h      |  28 ++
 7 files changed, 638 insertions(+), 33 deletions(-)
 create mode 100644 drivers/char/hvc_iseries.c

(limited to 'include')

diff --git a/arch/powerpc/platforms/iseries/Kconfig b/arch/powerpc/platforms/iseries/Kconfig
index 3d957a30c8c2..887b68804e6d 100644
--- a/arch/powerpc/platforms/iseries/Kconfig
+++ b/arch/powerpc/platforms/iseries/Kconfig
@@ -3,13 +3,17 @@ menu "iSeries device drivers"
 	depends on PPC_ISERIES
 
 config VIOCONS
-	tristate "iSeries Virtual Console Support"
+	tristate "iSeries Virtual Console Support (Obsolete)"
+	help
+	  This is the old virtual console driver for legacy iSeries.
+	  You should use the iSeries Hypervisor Virtual Console
+	  support instead.
 
 config VIODASD
 	tristate "iSeries Virtual I/O disk support"
 	help
 	  If you are running on an iSeries system and you want to use
- 	  virtual disks created and managed by OS/400, say Y.
+	  virtual disks created and managed by OS/400, say Y.
 
 config VIOCD
 	tristate "iSeries Virtual I/O CD support"
diff --git a/arch/powerpc/platforms/iseries/dt.c b/arch/powerpc/platforms/iseries/dt.c
index d194140c1ebf..894b534669d0 100644
--- a/arch/powerpc/platforms/iseries/dt.c
+++ b/arch/powerpc/platforms/iseries/dt.c
@@ -298,7 +298,8 @@ static void __init dt_vdevices(struct iseries_flat_dt *dt)
 	dt_prop_u32(dt, "#address-cells", 1);
 	dt_prop_u32(dt, "#size-cells", 0);
 
-	dt_do_vdevice(dt, "vty", reg, -1, device_type_serial, NULL, 1);
+	dt_do_vdevice(dt, "vty", reg, -1, device_type_serial,
+			"IBM,iSeries-vty", 1);
 	reg++;
 
 	dt_do_vdevice(dt, "v-scsi", reg, -1, device_type_vscsi,
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 11de59ff4229..a7ef542afbc2 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -610,6 +610,13 @@ config HVC_CONSOLE
 	  console. This driver allows each pSeries partition to have a console
 	  which is accessed via the HMC.
 
+config HVC_ISERIES
+	bool "iSeries Hypervisor Virtual Console support"
+	depends on PPC_ISERIES && !VIOCONS
+	select HVC_DRIVER
+	help
+	  iSeries machines support a hypervisor virtual console.
+
 config HVC_RTAS
 	bool "IBM RTAS Console support"
 	depends on PPC_RTAS
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index 7a7ee5721279..8c6dfc621520 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_AMIGA_BUILTIN_SERIAL) += amiserial.o
 obj-$(CONFIG_SX)		+= sx.o generic_serial.o
 obj-$(CONFIG_RIO)		+= rio/ generic_serial.o
 obj-$(CONFIG_HVC_CONSOLE)	+= hvc_vio.o hvsi.o
+obj-$(CONFIG_HVC_ISERIES)	+= hvc_iseries.o
 obj-$(CONFIG_HVC_RTAS)		+= hvc_rtas.o
 obj-$(CONFIG_HVC_DRIVER)	+= hvc_console.o
 obj-$(CONFIG_RAW_DRIVER)	+= raw.o
diff --git a/drivers/char/hvc_iseries.c b/drivers/char/hvc_iseries.c
new file mode 100644
index 000000000000..256afc8e5838
--- /dev/null
+++ b/drivers/char/hvc_iseries.c
@@ -0,0 +1,593 @@
+/*
+ * iSeries vio driver interface to hvc_console.c
+ *
+ * This code is based heavily on hvc_vio.c and viocons.c
+ *
+ * Copyright (C) 2006 Stephen Rothwell, IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+#include <stdarg.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/console.h>
+
+#include <asm/hvconsole.h>
+#include <asm/vio.h>
+#include <asm/prom.h>
+#include <asm/iseries/vio.h>
+#include <asm/iseries/hv_call.h>
+#include <asm/iseries/hv_lp_config.h>
+#include <asm/iseries/hv_lp_event.h>
+
+#include "hvc_console.h"
+
+#define VTTY_PORTS 10
+
+static DEFINE_SPINLOCK(consolelock);
+static DEFINE_SPINLOCK(consoleloglock);
+
+static const char hvc_driver_name[] = "hvc_console";
+
+#define IN_BUF_SIZE	200
+
+/*
+ * Our port information.
+ */
+static struct port_info {
+	HvLpIndex lp;
+	u64 seq;	/* sequence number of last HV send */
+	u64 ack;	/* last ack from HV */
+	struct hvc_struct *hp;
+	int in_start;
+	int in_end;
+	unsigned char in_buf[IN_BUF_SIZE];
+} port_info[VTTY_PORTS] = {
+	[ 0 ... VTTY_PORTS - 1 ] = {
+		.lp = HvLpIndexInvalid
+	}
+};
+
+#define viochar_is_console(pi)	((pi) == &port_info[0])
+
+static struct vio_device_id hvc_driver_table[] __devinitdata = {
+	{"serial", "IBM,iSeries-vty"},
+	{ "", "" }
+};
+MODULE_DEVICE_TABLE(vio, hvc_driver_table);
+
+static void hvlog(char *fmt, ...)
+{
+	int i;
+	unsigned long flags;
+	va_list args;
+	static char buf[256];
+
+	spin_lock_irqsave(&consoleloglock, flags);
+	va_start(args, fmt);
+	i = vscnprintf(buf, sizeof(buf) - 1, fmt, args);
+	va_end(args);
+	buf[i++] = '\r';
+	HvCall_writeLogBuffer(buf, i);
+	spin_unlock_irqrestore(&consoleloglock, flags);
+}
+
+/*
+ * Initialize the common fields in a charLpEvent
+ */
+static void init_data_event(struct viocharlpevent *viochar, HvLpIndex lp)
+{
+	struct HvLpEvent *hev = &viochar->event;
+
+	memset(viochar, 0, sizeof(struct viocharlpevent));
+
+	hev->flags = HV_LP_EVENT_VALID | HV_LP_EVENT_DEFERRED_ACK |
+		HV_LP_EVENT_INT;
+	hev->xType = HvLpEvent_Type_VirtualIo;
+	hev->xSubtype = viomajorsubtype_chario | viochardata;
+	hev->xSourceLp = HvLpConfig_getLpIndex();
+	hev->xTargetLp = lp;
+	hev->xSizeMinus1 = sizeof(struct viocharlpevent);
+	hev->xSourceInstanceId = viopath_sourceinst(lp);
+	hev->xTargetInstanceId = viopath_targetinst(lp);
+}
+
+static int get_chars(uint32_t vtermno, char *buf, int count)
+{
+	struct port_info *pi;
+	int n = 0;
+	unsigned long flags;
+
+	if (vtermno >= VTTY_PORTS)
+		return -EINVAL;
+	if (count == 0)
+		return 0;
+
+	pi = &port_info[vtermno];
+	spin_lock_irqsave(&consolelock, flags);
+
+	if (pi->in_end == 0)
+		goto done;
+
+	n = pi->in_end - pi->in_start;
+	if (n > count)
+		n = count;
+	memcpy(buf, &pi->in_buf[pi->in_start], n);
+	pi->in_start += n;
+	if (pi->in_start == pi->in_end) {
+		pi->in_start = 0;
+		pi->in_end = 0;
+	}
+done:
+	spin_unlock_irqrestore(&consolelock, flags);
+	return n;
+}
+
+static int put_chars(uint32_t vtermno, const char *buf, int count)
+{
+	struct viocharlpevent *viochar;
+	struct port_info *pi;
+	HvLpEvent_Rc hvrc;
+	unsigned long flags;
+	int sent = 0;
+
+	if (vtermno >= VTTY_PORTS)
+		return -EINVAL;
+
+	pi = &port_info[vtermno];
+
+	spin_lock_irqsave(&consolelock, flags);
+
+	if (viochar_is_console(pi) && !viopath_isactive(pi->lp)) {
+		spin_lock_irqsave(&consoleloglock, flags);
+		HvCall_writeLogBuffer(buf, count);
+		spin_unlock_irqrestore(&consoleloglock, flags);
+		sent = count;
+		goto done;
+	}
+
+	viochar = vio_get_event_buffer(viomajorsubtype_chario);
+	if (viochar == NULL) {
+		hvlog("\n\rviocons: Can't get viochar buffer.");
+		goto done;
+	}
+
+	while ((count > 0) && ((pi->seq - pi->ack) < VIOCHAR_WINDOW)) {
+		int len;
+
+		len = (count > VIOCHAR_MAX_DATA) ? VIOCHAR_MAX_DATA : count;
+
+		if (viochar_is_console(pi)) {
+			spin_lock_irqsave(&consoleloglock, flags);
+			HvCall_writeLogBuffer(buf, len);
+			spin_unlock_irqrestore(&consoleloglock, flags);
+		}
+
+		init_data_event(viochar, pi->lp);
+
+		viochar->len = len;
+		viochar->event.xCorrelationToken = pi->seq++;
+		viochar->event.xSizeMinus1 =
+			offsetof(struct viocharlpevent, data) + len;
+
+		memcpy(viochar->data, buf, len);
+
+		hvrc = HvCallEvent_signalLpEvent(&viochar->event);
+		if (hvrc)
+			hvlog("\n\rerror sending event! return code %d\n\r",
+				(int)hvrc);
+		sent += len;
+		count -= len;
+		buf += len;
+	}
+
+	vio_free_event_buffer(viomajorsubtype_chario, viochar);
+done:
+	spin_unlock_irqrestore(&consolelock, flags);
+	return sent;
+}
+
+static struct hv_ops hvc_get_put_ops = {
+	.get_chars = get_chars,
+	.put_chars = put_chars,
+};
+
+static int __devinit hvc_vio_probe(struct vio_dev *vdev,
+			const struct vio_device_id *id)
+{
+	struct hvc_struct *hp;
+	struct port_info *pi;
+
+	/* probed with invalid parameters. */
+	if (!vdev || !id)
+		return -EPERM;
+
+	if (vdev->unit_address >= VTTY_PORTS)
+		return -ENODEV;
+
+	pi = &port_info[vdev->unit_address];
+
+	hp = hvc_alloc(vdev->unit_address, vdev->irq, &hvc_get_put_ops);
+	if (IS_ERR(hp))
+		return PTR_ERR(hp);
+	pi->hp = hp;
+	dev_set_drvdata(&vdev->dev, pi);
+
+	return 0;
+}
+
+static int __devexit hvc_vio_remove(struct vio_dev *vdev)
+{
+	struct port_info *pi = dev_get_drvdata(&vdev->dev);
+	struct hvc_struct *hp = pi->hp;
+
+	return hvc_remove(hp);
+}
+
+static struct vio_driver hvc_vio_driver = {
+	.id_table	= hvc_driver_table,
+	.probe		= hvc_vio_probe,
+	.remove		= hvc_vio_remove,
+	.driver		= {
+		.name	= hvc_driver_name,
+		.owner	= THIS_MODULE,
+	}
+};
+
+static void hvc_open_event(struct HvLpEvent *event)
+{
+	unsigned long flags;
+	struct viocharlpevent *cevent = (struct viocharlpevent *)event;
+	u8 port = cevent->virtual_device;
+	struct port_info *pi;
+	int reject = 0;
+
+	if (hvlpevent_is_ack(event)) {
+		if (port >= VTTY_PORTS)
+			return;
+
+		spin_lock_irqsave(&consolelock, flags);
+
+		pi = &port_info[port];
+		if (event->xRc == HvLpEvent_Rc_Good) {
+			pi->seq = pi->ack = 0;
+			/*
+			 * This line allows connections from the primary
+			 * partition but once one is connected from the
+			 * primary partition nothing short of a reboot
+			 * of linux will allow access from the hosting
+			 * partition again without a required iSeries fix.
+			 */
+			pi->lp = event->xTargetLp;
+		}
+
+		spin_unlock_irqrestore(&consolelock, flags);
+		if (event->xRc != HvLpEvent_Rc_Good)
+			printk(KERN_WARNING
+			       "hvc: handle_open_event: event->xRc == (%d).\n",
+			       event->xRc);
+
+		if (event->xCorrelationToken != 0) {
+			atomic_t *aptr= (atomic_t *)event->xCorrelationToken;
+			atomic_set(aptr, 1);
+		} else
+			printk(KERN_WARNING
+			       "hvc: weird...got open ack without atomic\n");
+		return;
+	}
+
+	/* This had better require an ack, otherwise complain */
+	if (!hvlpevent_need_ack(event)) {
+		printk(KERN_WARNING "hvc: viocharopen without ack bit!\n");
+		return;
+	}
+
+	spin_lock_irqsave(&consolelock, flags);
+
+	/* Make sure this is a good virtual tty */
+	if (port >= VTTY_PORTS) {
+		event->xRc = HvLpEvent_Rc_SubtypeError;
+		cevent->subtype_result_code = viorc_openRejected;
+		/*
+		 * Flag state here since we can't printk while holding
+		 * the consolelock spinlock.
+		 */
+		reject = 1;
+	} else {
+		pi = &port_info[port];
+		if ((pi->lp != HvLpIndexInvalid) &&
+				(pi->lp != event->xSourceLp)) {
+			/*
+			 * If this is tty is already connected to a different
+			 * partition, fail.
+			 */
+			event->xRc = HvLpEvent_Rc_SubtypeError;
+			cevent->subtype_result_code = viorc_openRejected;
+			reject = 2;
+		} else {
+			pi->lp = event->xSourceLp;
+			event->xRc = HvLpEvent_Rc_Good;
+			cevent->subtype_result_code = viorc_good;
+			pi->seq = pi->ack = 0;
+		}
+	}
+
+	spin_unlock_irqrestore(&consolelock, flags);
+
+	if (reject == 1)
+		printk(KERN_WARNING "hvc: open rejected: bad virtual tty.\n");
+	else if (reject == 2)
+		printk(KERN_WARNING "hvc: open rejected: console in exclusive "
+				"use by another partition.\n");
+
+	/* Return the acknowledgement */
+	HvCallEvent_ackLpEvent(event);
+}
+
+/*
+ * Handle a close charLpEvent.  This should ONLY be an Interrupt because the
+ * virtual console should never actually issue a close event to the hypervisor
+ * because the virtual console never goes away.  A close event coming from the
+ * hypervisor simply means that there are no client consoles connected to the
+ * virtual console.
+ */
+static void hvc_close_event(struct HvLpEvent *event)
+{
+	unsigned long flags;
+	struct viocharlpevent *cevent = (struct viocharlpevent *)event;
+	u8 port = cevent->virtual_device;
+
+	if (!hvlpevent_is_int(event)) {
+		printk(KERN_WARNING
+			"hvc: got unexpected close acknowlegement\n");
+		return;
+	}
+
+	if (port >= VTTY_PORTS) {
+		printk(KERN_WARNING
+			"hvc: close message from invalid virtual device.\n");
+		return;
+	}
+
+	/* For closes, just mark the console partition invalid */
+	spin_lock_irqsave(&consolelock, flags);
+
+	if (port_info[port].lp == event->xSourceLp)
+		port_info[port].lp = HvLpIndexInvalid;
+
+	spin_unlock_irqrestore(&consolelock, flags);
+}
+
+static void hvc_data_event(struct HvLpEvent *event)
+{
+	unsigned long flags;
+	struct viocharlpevent *cevent = (struct viocharlpevent *)event;
+	struct port_info *pi;
+	int n;
+	u8 port = cevent->virtual_device;
+
+	if (port >= VTTY_PORTS) {
+		printk(KERN_WARNING "hvc: data on invalid virtual device %d\n",
+				port);
+		return;
+	}
+	if (cevent->len == 0)
+		return;
+
+	/*
+	 * Change 05/01/2003 - Ryan Arnold: If a partition other than
+	 * the current exclusive partition tries to send us data
+	 * events then just drop them on the floor because we don't
+	 * want his stinking data.  He isn't authorized to receive
+	 * data because he wasn't the first one to get the console,
+	 * therefore he shouldn't be allowed to send data either.
+	 * This will work without an iSeries fix.
+	 */
+	pi = &port_info[port];
+	if (pi->lp != event->xSourceLp)
+		return;
+
+	spin_lock_irqsave(&consolelock, flags);
+
+	n = IN_BUF_SIZE - pi->in_end;
+	if (n > cevent->len)
+		n = cevent->len;
+	if (n > 0) {
+		memcpy(&pi->in_buf[pi->in_end], cevent->data, n);
+		pi->in_end += n;
+	}
+	spin_unlock_irqrestore(&consolelock, flags);
+	if (n == 0)
+		printk(KERN_WARNING "hvc: input buffer overflow\n");
+}
+
+static void hvc_ack_event(struct HvLpEvent *event)
+{
+	struct viocharlpevent *cevent = (struct viocharlpevent *)event;
+	unsigned long flags;
+	u8 port = cevent->virtual_device;
+
+	if (port >= VTTY_PORTS) {
+		printk(KERN_WARNING "hvc: data on invalid virtual device\n");
+		return;
+	}
+
+	spin_lock_irqsave(&consolelock, flags);
+	port_info[port].ack = event->xCorrelationToken;
+	spin_unlock_irqrestore(&consolelock, flags);
+}
+
+static void hvc_config_event(struct HvLpEvent *event)
+{
+	struct viocharlpevent *cevent = (struct viocharlpevent *)event;
+
+	if (cevent->data[0] == 0x01)
+		printk(KERN_INFO "hvc: window resized to %d: %d: %d: %d\n",
+		       cevent->data[1], cevent->data[2],
+		       cevent->data[3], cevent->data[4]);
+	else
+		printk(KERN_WARNING "hvc: unknown config event\n");
+}
+
+static void hvc_handle_event(struct HvLpEvent *event)
+{
+	int charminor;
+
+	if (event == NULL)
+		return;
+
+	charminor = event->xSubtype & VIOMINOR_SUBTYPE_MASK;
+	switch (charminor) {
+	case viocharopen:
+		hvc_open_event(event);
+		break;
+	case viocharclose:
+		hvc_close_event(event);
+		break;
+	case viochardata:
+		hvc_data_event(event);
+		break;
+	case viocharack:
+		hvc_ack_event(event);
+		break;
+	case viocharconfig:
+		hvc_config_event(event);
+		break;
+	default:
+		if (hvlpevent_is_int(event) && hvlpevent_need_ack(event)) {
+			event->xRc = HvLpEvent_Rc_InvalidSubtype;
+			HvCallEvent_ackLpEvent(event);
+		}
+	}
+}
+
+static int send_open(HvLpIndex remoteLp, void *sem)
+{
+	return HvCallEvent_signalLpEventFast(remoteLp,
+			HvLpEvent_Type_VirtualIo,
+			viomajorsubtype_chario | viocharopen,
+			HvLpEvent_AckInd_DoAck, HvLpEvent_AckType_ImmediateAck,
+			viopath_sourceinst(remoteLp),
+			viopath_targetinst(remoteLp),
+			(u64)(unsigned long)sem, VIOVERSION << 16,
+			0, 0, 0, 0);
+}
+
+static int hvc_vio_init(void)
+{
+	atomic_t wait_flag;
+	int rc;
+
+	/* +2 for fudge */
+	rc = viopath_open(HvLpConfig_getPrimaryLpIndex(),
+			viomajorsubtype_chario, VIOCHAR_WINDOW + 2);
+	if (rc)
+		printk(KERN_WARNING "hvc: error opening to primary %d\n", rc);
+
+	if (viopath_hostLp == HvLpIndexInvalid)
+		vio_set_hostlp();
+
+	/*
+	 * And if the primary is not the same as the hosting LP, open to the
+	 * hosting lp
+	 */
+	if ((viopath_hostLp != HvLpIndexInvalid) &&
+	    (viopath_hostLp != HvLpConfig_getPrimaryLpIndex())) {
+		printk(KERN_INFO "hvc: open path to hosting (%d)\n",
+				viopath_hostLp);
+		rc = viopath_open(viopath_hostLp, viomajorsubtype_chario,
+				VIOCHAR_WINDOW + 2);	/* +2 for fudge */
+		if (rc)
+			printk(KERN_WARNING
+				"error opening to partition %d: %d\n",
+				viopath_hostLp, rc);
+	}
+
+	if (vio_setHandler(viomajorsubtype_chario, hvc_handle_event) < 0)
+		printk(KERN_WARNING
+			"hvc: error seting handler for console events!\n");
+
+	/*
+	 * First, try to open the console to the hosting lp.
+	 * Wait on a semaphore for the response.
+	 */
+	atomic_set(&wait_flag, 0);
+	if ((viopath_isactive(viopath_hostLp)) &&
+	    (send_open(viopath_hostLp, &wait_flag) == 0)) {
+		printk(KERN_INFO "hvc: hosting partition %d\n", viopath_hostLp);
+		while (atomic_read(&wait_flag) == 0)
+			mb();
+		atomic_set(&wait_flag, 0);
+	}
+
+	/*
+	 * If we don't have an active console, try the primary
+	 */
+	if ((!viopath_isactive(port_info[0].lp)) &&
+	    (viopath_isactive(HvLpConfig_getPrimaryLpIndex())) &&
+	    (send_open(HvLpConfig_getPrimaryLpIndex(), &wait_flag) == 0)) {
+		printk(KERN_INFO "hvc: opening console to primary partition\n");
+		while (atomic_read(&wait_flag) == 0)
+			mb();
+	}
+
+	/* Register as a vio device to receive callbacks */
+	rc = vio_register_driver(&hvc_vio_driver);
+
+	return rc;
+}
+module_init(hvc_vio_init); /* after drivers/char/hvc_console.c */
+
+static void hvc_vio_exit(void)
+{
+	vio_unregister_driver(&hvc_vio_driver);
+}
+module_exit(hvc_vio_exit);
+
+/* the device tree order defines our numbering */
+static int hvc_find_vtys(void)
+{
+	struct device_node *vty;
+	int num_found = 0;
+
+	for (vty = of_find_node_by_name(NULL, "vty"); vty != NULL;
+			vty = of_find_node_by_name(vty, "vty")) {
+		uint32_t *vtermno;
+
+		/* We have statically defined space for only a certain number
+		 * of console adapters.
+		 */
+		if ((num_found >= MAX_NR_HVC_CONSOLES) ||
+				(num_found >= VTTY_PORTS))
+			break;
+
+		vtermno = (uint32_t *)get_property(vty, "reg", NULL);
+		if (!vtermno)
+			continue;
+
+		if (!device_is_compatible(vty, "IBM,iSeries-vty"))
+			continue;
+
+		if (num_found == 0)
+			add_preferred_console("hvc", 0, NULL);
+		hvc_instantiate(*vtermno, num_found, &hvc_get_put_ops);
+		++num_found;
+	}
+
+	return num_found;
+}
+console_initcall(hvc_find_vtys);
diff --git a/drivers/char/viocons.c b/drivers/char/viocons.c
index 766f7864c6c6..f3efeaf2826e 100644
--- a/drivers/char/viocons.c
+++ b/drivers/char/viocons.c
@@ -43,7 +43,6 @@
 #include <linux/sysrq.h>
 
 #include <asm/iseries/vio.h>
-
 #include <asm/iseries/hv_lp_event.h>
 #include <asm/iseries/hv_call_event.h>
 #include <asm/iseries/hv_lp_config.h>
@@ -67,35 +66,6 @@ static int vio_sysrq_pressed;
 extern int sysrq_enabled;
 #endif
 
-/*
- * The structure of the events that flow between us and OS/400.  You can't
- * mess with this unless the OS/400 side changes too
- */
-struct viocharlpevent {
-	struct HvLpEvent event;
-	u32 reserved;
-	u16 version;
-	u16 subtype_result_code;
-	u8 virtual_device;
-	u8 len;
-	u8 data[VIOCHAR_MAX_DATA];
-};
-
-#define VIOCHAR_WINDOW		10
-#define VIOCHAR_HIGHWATERMARK	3
-
-enum viocharsubtype {
-	viocharopen = 0x0001,
-	viocharclose = 0x0002,
-	viochardata = 0x0003,
-	viocharack = 0x0004,
-	viocharconfig = 0x0005
-};
-
-enum viochar_rc {
-	viochar_rc_ebusy = 1
-};
-
 #define VIOCHAR_NUM_BUF		16
 
 /*
@@ -1183,6 +1153,7 @@ static int __init viocons_init(void)
 		port_info[i].magic = VIOTTY_MAGIC;
 	}
 	HvCall_setLogBufferFormatAndCodepage(HvCall_LogBuffer_ASCII, 437);
+	add_preferred_console("viocons", 0, NULL);
 	register_console(&viocons_early);
 	return 0;
 }
diff --git a/include/asm-powerpc/iseries/vio.h b/include/asm-powerpc/iseries/vio.h
index 72a97d37aac3..7a95d296abd1 100644
--- a/include/asm-powerpc/iseries/vio.h
+++ b/include/asm-powerpc/iseries/vio.h
@@ -122,6 +122,34 @@ enum viorc {
 	viorc_openRejected = 0x0301
 };
 
+/*
+ * The structure of the events that flow between us and OS/400 for chario
+ * events.  You can't mess with this unless the OS/400 side changes too.
+ */
+struct viocharlpevent {
+	struct HvLpEvent event;
+	u32 reserved;
+	u16 version;
+	u16 subtype_result_code;
+	u8 virtual_device;
+	u8 len;
+	u8 data[VIOCHAR_MAX_DATA];
+};
+
+#define VIOCHAR_WINDOW		10
+
+enum viocharsubtype {
+	viocharopen = 0x0001,
+	viocharclose = 0x0002,
+	viochardata = 0x0003,
+	viocharack = 0x0004,
+	viocharconfig = 0x0005
+};
+
+enum viochar_rc {
+	viochar_rc_ebusy = 1
+};
+
 struct device;
 
 extern struct device *iSeries_vio_dev;
-- 
cgit v1.2.3


From 54f5cd8afa1c9c9f8b152a946b0a7e0ecdef1631 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 13 Jul 2006 18:56:56 +1000
Subject: [POWERPC] iseries: Remove unnecessary include of
 iseries/hv_lp_event.h

Also remove unnecessary reference to struct HvLpEvent.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/kernel/asm-offsets.c         | 1 -
 include/asm-powerpc/iseries/it_lp_queue.h | 2 --
 2 files changed, 3 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index 7ee84968087b..ac0631958b20 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -40,7 +40,6 @@
 #ifdef CONFIG_PPC64
 #include <asm/paca.h>
 #include <asm/lppaca.h>
-#include <asm/iseries/hv_lp_event.h>
 #include <asm/cache.h>
 #include <asm/compat.h>
 #endif
diff --git a/include/asm-powerpc/iseries/it_lp_queue.h b/include/asm-powerpc/iseries/it_lp_queue.h
index 284c5a7db3ac..3f6814769295 100644
--- a/include/asm-powerpc/iseries/it_lp_queue.h
+++ b/include/asm-powerpc/iseries/it_lp_queue.h
@@ -27,8 +27,6 @@
 #include <asm/types.h>
 #include <asm/ptrace.h>
 
-struct HvLpEvent;
-
 #define IT_LP_MAX_QUEUES	8
 
 #define IT_LP_NOT_USED		0	/* Queue will not be used by PLIC */
-- 
cgit v1.2.3


From ca652c9396fa052815518e2b2ce2ebee6d9fb861 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Fri, 14 Jul 2006 14:25:33 +1000
Subject: [POWERPC] iseries: Move iommu_table_cb into platforms/iseries

Although we pass the address of an iommu_table_cb to HvCallXm_getTceTableParms,
we don't actually need the structure definition anywhere except in the
iseries iommu code, so move the struct in there.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/platforms/iseries/iommu.c   | 17 +++++++++++++++++
 include/asm-powerpc/iseries/hv_call_xm.h | 17 -----------------
 2 files changed, 17 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c
index e3bd2015f2c9..2c3dbcd4613c 100644
--- a/arch/powerpc/platforms/iseries/iommu.c
+++ b/arch/powerpc/platforms/iseries/iommu.c
@@ -87,6 +87,23 @@ static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages)
 	}
 }
 
+/*
+ * Structure passed to HvCallXm_getTceTableParms
+ */
+struct iommu_table_cb {
+	unsigned long	itc_busno;	/* Bus number for this tce table */
+	unsigned long	itc_start;	/* Will be NULL for secondary */
+	unsigned long	itc_totalsize;	/* Size (in pages) of whole table */
+	unsigned long	itc_offset;	/* Index into real tce table of the
+					   start of our section */
+	unsigned long	itc_size;	/* Size (in pages) of our section */
+	unsigned long	itc_index;	/* Index of this tce table */
+	unsigned short	itc_maxtables;	/* Max num of tables for partition */
+	unsigned char	itc_virtbus;	/* Flag to indicate virtual bus */
+	unsigned char	itc_slotno;	/* IOA Tce Slot Index */
+	unsigned char	itc_rsvd[4];
+};
+
 /*
  * Call Hv with the architected data structure to get TCE table info.
  * info. Put the returned data into the Linux representation of the
diff --git a/include/asm-powerpc/iseries/hv_call_xm.h b/include/asm-powerpc/iseries/hv_call_xm.h
index ca9202cb01ed..392ac3f54df0 100644
--- a/include/asm-powerpc/iseries/hv_call_xm.h
+++ b/include/asm-powerpc/iseries/hv_call_xm.h
@@ -16,23 +16,6 @@
 #define HvCallXmSetTce			HvCallXm + 11
 #define HvCallXmSetTces			HvCallXm + 13
 
-/*
- * Structure passed to HvCallXm_getTceTableParms
- */
-struct iommu_table_cb {
-	unsigned long	itc_busno;	/* Bus number for this tce table */
-	unsigned long	itc_start;	/* Will be NULL for secondary */
-	unsigned long	itc_totalsize;	/* Size (in pages) of whole table */
-	unsigned long	itc_offset;	/* Index into real tce table of the
-					   start of our section */
-	unsigned long	itc_size;	/* Size (in pages) of our section */
-	unsigned long	itc_index;	/* Index of this tce table */
-	unsigned short	itc_maxtables;	/* Max num of tables for partition */
-	unsigned char	itc_virtbus;	/* Flag to indicate virtual bus */
-	unsigned char	itc_slotno;	/* IOA Tce Slot Index */
-	unsigned char	itc_rsvd[4];
-};
-
 static inline void HvCallXm_getTceTableParms(u64 cb)
 {
 	HvCall1(HvCallXmGetTceTableParms, cb);
-- 
cgit v1.2.3


From 804af2cf6e7af31d2e664b54e657dddd9b531dbd Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Wed, 26 Jul 2006 21:39:49 +0100
Subject: [AGPGART] remove private page protection map

AGP keeps its own copy of the protection_map, upcoming DRM changes will
also require access to this map from modules.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/char/agp/frontend.c | 27 ++-------------------------
 include/linux/mm.h          |  1 +
 mm/mmap.c                   |  7 +++++++
 3 files changed, 10 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/drivers/char/agp/frontend.c b/drivers/char/agp/frontend.c
index d9c5a9142ad1..0f2ed2aa2d81 100644
--- a/drivers/char/agp/frontend.c
+++ b/drivers/char/agp/frontend.c
@@ -151,35 +151,12 @@ static void agp_add_seg_to_client(struct agp_client *client,
 	client->segments = seg;
 }
 
-/* Originally taken from linux/mm/mmap.c from the array
- * protection_map.
- * The original really should be exported to modules, or
- * some routine which does the conversion for you
- */
-
-static const pgprot_t my_protect_map[16] =
-{
-	__P000, __P001, __P010, __P011, __P100, __P101, __P110, __P111,
-	__S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111
-};
-
 static pgprot_t agp_convert_mmap_flags(int prot)
 {
-#define _trans(x,bit1,bit2) \
-((bit1==bit2)?(x&bit1):(x&bit1)?bit2:0)
-
 	unsigned long prot_bits;
-	pgprot_t temp;
-
-	prot_bits = _trans(prot, PROT_READ, VM_READ) |
-	    _trans(prot, PROT_WRITE, VM_WRITE) |
-	    _trans(prot, PROT_EXEC, VM_EXEC);
-
-	prot_bits |= VM_SHARED;
 
-	temp = my_protect_map[prot_bits & 0x0000000f];
-
-	return temp;
+	prot_bits = calc_vm_prot_bits(prot) | VM_SHARED;
+	return vm_get_page_prot(prot_bits);
 }
 
 static int agp_create_segment(struct agp_client *client, struct agp_region *region)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 990957e0929f..4fba4560699b 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1012,6 +1012,7 @@ static inline unsigned long vma_pages(struct vm_area_struct *vma)
 	return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
 }
 
+pgprot_t vm_get_page_prot(unsigned long vm_flags);
 struct vm_area_struct *find_extend_vma(struct mm_struct *, unsigned long addr);
 struct page *vmalloc_to_page(void *addr);
 unsigned long vmalloc_to_pfn(void *addr);
diff --git a/mm/mmap.c b/mm/mmap.c
index c1868ecdbc5f..c7ed061f4507 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -60,6 +60,13 @@ pgprot_t protection_map[16] = {
 	__S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111
 };
 
+pgprot_t vm_get_page_prot(unsigned long vm_flags)
+{
+	return protection_map[vm_flags &
+				(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
+}
+EXPORT_SYMBOL(vm_get_page_prot);
+
 int sysctl_overcommit_memory = OVERCOMMIT_GUESS;  /* heuristic overcommit */
 int sysctl_overcommit_ratio = 50;	/* default is 50% */
 int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT;
-- 
cgit v1.2.3


From d8e2be90d301a0381e9b2528fe2835cf2992bca3 Mon Sep 17 00:00:00 2001
From: Daniel Drake <dsd@gentoo.org>
Date: Tue, 18 Jul 2006 21:30:34 +0100
Subject: [PATCH] ieee80211: small ERP handling additions

This adds a flag to the ieee80211_network structure which indicates whether
the stored erp_value is valid (a check against 0 is not enough, since an ERP
of 0 is valid and very meaningful).

I also added the ERP IE bit-definitions to ieee80211.h.

This is needed by some upcoming softmac patches.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211.h      | 7 +++++++
 net/ieee80211/ieee80211_rx.c | 1 +
 2 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index ecc42864b001..aa007f49bf29 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -240,6 +240,11 @@ struct ieee80211_snap_hdr {
 #define WLAN_CAPABILITY_SHORT_SLOT_TIME (1<<10)
 #define WLAN_CAPABILITY_DSSS_OFDM (1<<13)
 
+/* 802.11g ERP information element */
+#define WLAN_ERP_NON_ERP_PRESENT (1<<0)
+#define WLAN_ERP_USE_PROTECTION (1<<1)
+#define WLAN_ERP_BARKER_PREAMBLE (1<<2)
+
 /* Status codes */
 enum ieee80211_statuscode {
 	WLAN_STATUS_SUCCESS = 0,
@@ -747,6 +752,8 @@ struct ieee80211_txb {
 #define NETWORK_HAS_IBSS_DFS            (1<<8)
 #define NETWORK_HAS_TPC_REPORT          (1<<9)
 
+#define NETWORK_HAS_ERP_VALUE           (1<<10)
+
 #define QOS_QUEUE_NUM                   4
 #define QOS_OUI_LEN                     3
 #define QOS_OUI_TYPE                    2
diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c
index 72d4d4e04d42..f73fc164b9ef 100644
--- a/net/ieee80211/ieee80211_rx.c
+++ b/net/ieee80211/ieee80211_rx.c
@@ -1166,6 +1166,7 @@ static int ieee80211_parse_info_param(struct ieee80211_info_element
 
 		case MFIE_TYPE_ERP_INFO:
 			network->erp_value = info_element->data[0];
+			network->flags |= NETWORK_HAS_ERP_VALUE;
 			IEEE80211_DEBUG_MGMT("MFIE_TYPE_ERP_SET: %d\n",
 					     network->erp_value);
 			break;
-- 
cgit v1.2.3


From 5acd0c4153be25269d7cb9a4b09fd6db571c5cc1 Mon Sep 17 00:00:00 2001
From: Daniel Drake <dsd@gentoo.org>
Date: Tue, 18 Jul 2006 21:33:27 +0100
Subject: [PATCH] softmac: ERP handling and driver-level notifications

This patch implements ERP handling in softmac so that the drivers can support
protection and preambles properly.

I added a new struct, ieee80211softmac_bss_info, which is used for
BSS-dependent variables like these.

A new hook has been added (bssinfo_change), which allows the drivers to be
notified when anything in bssinfo changes.

I modified the txrates_change API to match the bssinfo_change API. The
existing one is a little messy and the usefulness of providing the old rates
is questionable (and can be implemented at driver level if really necessary).
No drivers are using this API (yet), so this should be safe.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211softmac.h                  | 52 +++++++++++++++--
 net/ieee80211/softmac/ieee80211softmac_assoc.c  | 21 +++++--
 net/ieee80211/softmac/ieee80211softmac_io.c     | 14 +++++
 net/ieee80211/softmac/ieee80211softmac_module.c | 78 ++++++++++++++++++-------
 net/ieee80211/softmac/ieee80211softmac_priv.h   |  8 ++-
 5 files changed, 142 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h
index 00ad810eb883..c27d03433c21 100644
--- a/include/net/ieee80211softmac.h
+++ b/include/net/ieee80211softmac.h
@@ -86,9 +86,6 @@ struct ieee80211softmac_assoc_info {
 	
 	/* BSSID we're trying to associate to */
 	char bssid[ETH_ALEN];
-
-	/* Rates supported by the network */
-	struct ieee80211softmac_ratesinfo supported_rates;
 	
 	/* some flags.
 	 * static_essid is valid if the essid is constant,
@@ -103,6 +100,7 @@ struct ieee80211softmac_assoc_info {
 	 * bssfixed is used for SIOCSIWAP.
 	 */
 	u8 static_essid:1,
+	   short_preamble_available:1,
 	   associating:1,
 	   assoc_wait:1,
 	   bssvalid:1,
@@ -115,6 +113,19 @@ struct ieee80211softmac_assoc_info {
 	struct work_struct timeout;
 };
 
+struct ieee80211softmac_bss_info {
+	/* Rates supported by the network */
+	struct ieee80211softmac_ratesinfo supported_rates;
+
+	/* This indicates whether frames can currently be transmitted with
+	 * short preamble (only use this variable during TX at CCK rates) */
+	u8 short_preamble:1;
+
+	/* This indicates whether protection (e.g. self-CTS) should be used
+	 * when transmitting with OFDM modulation */
+	u8 use_protection:1;
+};
+
 enum {
 	IEEE80211SOFTMAC_AUTH_OPEN_REQUEST	= 1,
 	IEEE80211SOFTMAC_AUTH_OPEN_RESPONSE	= 2,
@@ -157,6 +168,10 @@ struct ieee80211softmac_txrates {
 #define IEEE80211SOFTMAC_TXRATECHG_MCAST		(1 << 2) /* mcast_rate */
 #define IEEE80211SOFTMAC_TXRATECHG_MGT_MCAST		(1 << 3) /* mgt_mcast_rate */
 
+#define IEEE80211SOFTMAC_BSSINFOCHG_RATES		(1 << 0) /* supported_rates */
+#define IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE	(1 << 1) /* short_preamble */
+#define IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION		(1 << 2) /* use_protection */
+
 struct ieee80211softmac_device {
 	/* 802.11 structure for data stuff */
 	struct ieee80211_device *ieee;
@@ -200,10 +215,16 @@ struct ieee80211softmac_device {
 	 * The driver just needs to read them.
 	 */
 	struct ieee80211softmac_txrates txrates;
-	/* If the driver needs to do stuff on TX rate changes, assign this callback. */
+
+	/* If the driver needs to do stuff on TX rate changes, assign this
+	 * callback. See IEEE80211SOFTMAC_TXRATECHG for change flags. */
 	void (*txrates_change)(struct net_device *dev,
-			       u32 changes, /* see IEEE80211SOFTMAC_TXRATECHG flags */
-			       const struct ieee80211softmac_txrates *rates_before_change);
+			       u32 changes);
+
+	/* If the driver needs to do stuff when BSS properties change, assign
+	 * this callback. see IEEE80211SOFTMAC_BSSINFOCHG for change flags. */
+	void (*bssinfo_change)(struct net_device *dev,
+			       u32 changes);
 
 	/* private stuff follows */
 	/* this lock protects this structure */
@@ -216,6 +237,7 @@ struct ieee80211softmac_device {
 	
 	struct ieee80211softmac_scaninfo *scaninfo;
 	struct ieee80211softmac_assoc_info associnfo;
+	struct ieee80211softmac_bss_info bssinfo;
 
 	struct list_head auth_queue;
 	struct list_head events;
@@ -279,6 +301,24 @@ static inline u8 ieee80211softmac_suggest_txrate(struct ieee80211softmac_device
 		return txrates->mcast_rate;
 }
 
+/* Helper function which advises you when it is safe to transmit with short
+ * preamble.
+ * You should only call this function when transmitting at CCK rates. */
+static inline int ieee80211softmac_short_preamble_ok(struct ieee80211softmac_device *mac,
+						    int is_multicast,
+						    int is_mgt)
+{
+	return (is_multicast && is_mgt) ? 0 : mac->bssinfo.short_preamble;
+}
+
+/* Helper function which advises you whether protection (e.g. self-CTS) is
+ * needed. 1 = protection needed, 0 = no protection needed
+ * Only use this function when transmitting with OFDM modulation. */
+static inline int ieee80211softmac_protection_needed(struct ieee80211softmac_device *mac)
+{
+	return mac->bssinfo.use_protection;
+}
+
 /* Start the SoftMAC. Call this after you initialized the device
  * and it is ready to run.
  */
diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c
index 44215ce64d4e..589f6d2c548a 100644
--- a/net/ieee80211/softmac/ieee80211softmac_assoc.c
+++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c
@@ -96,7 +96,7 @@ ieee80211softmac_disassoc(struct ieee80211softmac_device *mac)
 	mac->associated = 0;
 	mac->associnfo.bssvalid = 0;
 	mac->associnfo.associating = 0;
-	ieee80211softmac_init_txrates(mac);
+	ieee80211softmac_init_bss(mac);
 	ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_DISASSOCIATED, NULL);
 	spin_unlock_irqrestore(&mac->lock, flags);
 }
@@ -334,11 +334,19 @@ ieee80211softmac_associated(struct ieee80211softmac_device *mac,
 	struct ieee80211_assoc_response * resp,
 	struct ieee80211softmac_network *net)
 {
+	u16 cap = le16_to_cpu(resp->capability);
+	u8 erp_value = net->erp_value;
+
 	mac->associnfo.associating = 0;
-	mac->associnfo.supported_rates = net->supported_rates;
+	mac->bssinfo.supported_rates = net->supported_rates;
 	ieee80211softmac_recalc_txrates(mac);
 
 	mac->associated = 1;
+
+	mac->associnfo.short_preamble_available =
+		(cap & WLAN_CAPABILITY_SHORT_PREAMBLE) != 0;
+	ieee80211softmac_process_erp(mac, erp_value);
+
 	if (mac->set_bssid_filter)
 		mac->set_bssid_filter(mac->dev, net->bssid);
 	memcpy(mac->ieee->bssid, net->bssid, ETH_ALEN);
@@ -351,9 +359,9 @@ ieee80211softmac_associated(struct ieee80211softmac_device *mac,
 int
 ieee80211softmac_handle_assoc_response(struct net_device * dev,
 				       struct ieee80211_assoc_response * resp,
-				       struct ieee80211_network * _ieee80211_network_do_not_use)
+				       struct ieee80211_network * _ieee80211_network)
 {
-	/* NOTE: the network parameter has to be ignored by
+	/* NOTE: the network parameter has to be mostly ignored by
 	 *       this code because it is the ieee80211's pointer
 	 *       to the struct, not ours (we made a copy)
 	 */
@@ -385,6 +393,11 @@ ieee80211softmac_handle_assoc_response(struct net_device * dev,
 	/* now that we know it was for us, we can cancel the timeout */
 	cancel_delayed_work(&mac->associnfo.timeout);
 
+	/* if the association response included an ERP IE, update our saved
+	 * copy */
+	if (_ieee80211_network->flags & NETWORK_HAS_ERP_VALUE)
+		network->erp_value = _ieee80211_network->erp_value;
+
 	switch (status) {
 		case 0:
 			dprintk(KERN_INFO PFX "associated!\n");
diff --git a/net/ieee80211/softmac/ieee80211softmac_io.c b/net/ieee80211/softmac/ieee80211softmac_io.c
index 6ae5a1dc7956..82bfddbf33a2 100644
--- a/net/ieee80211/softmac/ieee80211softmac_io.c
+++ b/net/ieee80211/softmac/ieee80211softmac_io.c
@@ -467,3 +467,17 @@ ieee80211softmac_send_mgt_frame(struct ieee80211softmac_device *mac,
 	kfree(pkt);
 	return 0;
 }
+
+/* Beacon handling */
+int ieee80211softmac_handle_beacon(struct net_device *dev,
+	struct ieee80211_beacon *beacon,
+	struct ieee80211_network *network)
+{
+	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
+
+	if (mac->associated && memcmp(network->bssid, mac->associnfo.bssid, ETH_ALEN) == 0)
+		ieee80211softmac_process_erp(mac, network->erp_value);
+
+	return 0;
+}
+
diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c
index 4b2e57d12418..c275646b2269 100644
--- a/net/ieee80211/softmac/ieee80211softmac_module.c
+++ b/net/ieee80211/softmac/ieee80211softmac_module.c
@@ -44,6 +44,7 @@ struct net_device *alloc_ieee80211softmac(int sizeof_priv)
 	softmac->ieee->handle_assoc_response = ieee80211softmac_handle_assoc_response;
 	softmac->ieee->handle_reassoc_request = ieee80211softmac_handle_reassoc_req;
 	softmac->ieee->handle_disassoc = ieee80211softmac_handle_disassoc;
+ 	softmac->ieee->handle_beacon = ieee80211softmac_handle_beacon;
 	softmac->scaninfo = NULL;
 
 	softmac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
@@ -209,35 +210,59 @@ static u8 highest_supported_rate(struct ieee80211softmac_device *mac,
 	return user_rate;
 }
 
+void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac,
+	u8 erp_value)
+{
+ 	int use_protection;
+	int short_preamble;
+ 	u32 changes = 0;
+
+	/* Barker preamble mode */
+	short_preamble = ((erp_value & WLAN_ERP_BARKER_PREAMBLE) == 0
+			  && mac->associnfo.short_preamble_available) ? 1 : 0;
+
+	/* Protection needed? */
+	use_protection = (erp_value & WLAN_ERP_USE_PROTECTION) != 0;
+
+	if (mac->bssinfo.short_preamble != short_preamble) {
+		changes |= IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE;
+		mac->bssinfo.short_preamble = short_preamble;
+	}
+
+	if (mac->bssinfo.use_protection != use_protection) {
+		changes |= IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION;
+		mac->bssinfo.use_protection = use_protection;
+	}
+
+	if (mac->bssinfo_change && changes)
+		mac->bssinfo_change(mac->dev, changes);
+}
+
 void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac)
 {
 	struct ieee80211softmac_txrates *txrates = &mac->txrates;
-	struct ieee80211softmac_txrates oldrates;
 	u32 change = 0;
 
-	if (mac->txrates_change)
-		oldrates = mac->txrates;
-
 	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
-	txrates->default_rate = highest_supported_rate(mac, &mac->associnfo.supported_rates, 0);
+	txrates->default_rate = highest_supported_rate(mac, &mac->bssinfo.supported_rates, 0);
 
 	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK;
 	txrates->default_fallback = lower_rate(mac, txrates->default_rate);
 
 	change |= IEEE80211SOFTMAC_TXRATECHG_MCAST;
-	txrates->mcast_rate = highest_supported_rate(mac, &mac->associnfo.supported_rates, 1);
+	txrates->mcast_rate = highest_supported_rate(mac, &mac->bssinfo.supported_rates, 1);
 
 	if (mac->txrates_change)
-		mac->txrates_change(mac->dev, change, &oldrates);
+		mac->txrates_change(mac->dev, change);
 
 }
 
-void ieee80211softmac_init_txrates(struct ieee80211softmac_device *mac)
+void ieee80211softmac_init_bss(struct ieee80211softmac_device *mac)
 {
 	struct ieee80211_device *ieee = mac->ieee;
 	u32 change = 0;
 	struct ieee80211softmac_txrates *txrates = &mac->txrates;
-	struct ieee80211softmac_txrates oldrates;
+	struct ieee80211softmac_bss_info *bssinfo = &mac->bssinfo;
 
 	/* TODO: We need some kind of state machine to lower the default rates
 	 *       if we loose too many packets.
@@ -245,8 +270,6 @@ void ieee80211softmac_init_txrates(struct ieee80211softmac_device *mac)
 	/* Change the default txrate to the highest possible value.
 	 * The txrate machine will lower it, if it is too high.
 	 */
-	if (mac->txrates_change)
-		oldrates = mac->txrates;
 	/* FIXME: We don't correctly handle backing down to lower
 	   rates, so 801.11g devices start off at 11M for now. People
 	   can manually change it if they really need to, but 11M is
@@ -272,7 +295,23 @@ void ieee80211softmac_init_txrates(struct ieee80211softmac_device *mac)
 	change |= IEEE80211SOFTMAC_TXRATECHG_MGT_MCAST;
 
 	if (mac->txrates_change)
-		mac->txrates_change(mac->dev, change, &oldrates);
+		mac->txrates_change(mac->dev, change);
+
+	change = 0;
+
+	bssinfo->supported_rates.count = 0;
+	memset(bssinfo->supported_rates.rates, 0,
+		sizeof(bssinfo->supported_rates.rates));
+	change |= IEEE80211SOFTMAC_BSSINFOCHG_RATES;
+
+	bssinfo->short_preamble = 0;
+	change |= IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE;
+
+	bssinfo->use_protection = 0;
+	change |= IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION;
+
+	if (mac->bssinfo_change)
+		mac->bssinfo_change(mac->dev, change);
 
 	mac->running = 1;
 }
@@ -282,7 +321,7 @@ void ieee80211softmac_start(struct net_device *dev)
 	struct ieee80211softmac_device *mac = ieee80211_priv(dev);
 
 	ieee80211softmac_start_check_rates(mac);
-	ieee80211softmac_init_txrates(mac);
+	ieee80211softmac_init_bss(mac);
 }
 EXPORT_SYMBOL_GPL(ieee80211softmac_start);
 
@@ -335,7 +374,6 @@ u8 ieee80211softmac_lower_rate_delta(struct ieee80211softmac_device *mac, u8 rat
 static void ieee80211softmac_add_txrates_badness(struct ieee80211softmac_device *mac,
 						 int amount)
 {
-	struct ieee80211softmac_txrates oldrates;
 	u8 default_rate = mac->txrates.default_rate;
 	u8 default_fallback = mac->txrates.default_fallback;
 	u32 changes = 0;
@@ -348,8 +386,6 @@ printk("badness %d\n", mac->txrate_badness);
 	mac->txrate_badness += amount;
 	if (mac->txrate_badness <= -1000) {
 		/* Very small badness. Try a faster bitrate. */
-		if (mac->txrates_change)
-			memcpy(&oldrates, &mac->txrates, sizeof(oldrates));
 		default_rate = raise_rate(mac, default_rate);
 		changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
 		default_fallback = get_fallback_rate(mac, default_rate);
@@ -358,8 +394,6 @@ printk("badness %d\n", mac->txrate_badness);
 printk("Bitrate raised to %u\n", default_rate);
 	} else if (mac->txrate_badness >= 10000) {
 		/* Very high badness. Try a slower bitrate. */
-		if (mac->txrates_change)
-			memcpy(&oldrates, &mac->txrates, sizeof(oldrates));
 		default_rate = lower_rate(mac, default_rate);
 		changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
 		default_fallback = get_fallback_rate(mac, default_rate);
@@ -372,7 +406,7 @@ printk("Bitrate lowered to %u\n", default_rate);
 	mac->txrates.default_fallback = default_fallback;
 
 	if (changes && mac->txrates_change)
-		mac->txrates_change(mac->dev, changes, &oldrates);
+		mac->txrates_change(mac->dev, changes);
 }
 
 void ieee80211softmac_fragment_lost(struct net_device *dev,
@@ -416,7 +450,11 @@ ieee80211softmac_create_network(struct ieee80211softmac_device *mac,
 	memcpy(&softnet->supported_rates.rates[softnet->supported_rates.count], net->rates_ex, net->rates_ex_len);
 	softnet->supported_rates.count += net->rates_ex_len;
 	sort(softnet->supported_rates.rates, softnet->supported_rates.count, sizeof(softnet->supported_rates.rates[0]), rate_cmp, NULL);
-	
+
+	/* we save the ERP value because it is needed at association time, and
+	 * many AP's do not include an ERP IE in the association response. */
+	softnet->erp_value = net->erp_value;
+
 	softnet->capabilities = net->capability;
 	return softnet;
 }
diff --git a/net/ieee80211/softmac/ieee80211softmac_priv.h b/net/ieee80211/softmac/ieee80211softmac_priv.h
index fa1f8e3acfc0..0642e090b8a7 100644
--- a/net/ieee80211/softmac/ieee80211softmac_priv.h
+++ b/net/ieee80211/softmac/ieee80211softmac_priv.h
@@ -116,9 +116,11 @@ ieee80211softmac_get_network_by_essid(struct ieee80211softmac_device *mac,
 	struct ieee80211softmac_essid *essid);
 
 /* Rates related */
+void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac,
+	u8 erp_value);
 int ieee80211softmac_ratesinfo_rate_supported(struct ieee80211softmac_ratesinfo *ri, u8 rate);
 u8 ieee80211softmac_lower_rate_delta(struct ieee80211softmac_device *mac, u8 rate, int delta);
-void ieee80211softmac_init_txrates(struct ieee80211softmac_device *mac);
+void ieee80211softmac_init_bss(struct ieee80211softmac_device *mac);
 void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac);
 static inline u8 lower_rate(struct ieee80211softmac_device *mac, u8 rate) {
 	return ieee80211softmac_lower_rate_delta(mac, rate, 1);
@@ -133,6 +135,9 @@ static inline u8 get_fallback_rate(struct ieee80211softmac_device *mac, u8 rate)
 /*** prototypes from _io.c */
 int ieee80211softmac_send_mgt_frame(struct ieee80211softmac_device *mac,
 	void* ptrarg, u32 type, u32 arg);
+int ieee80211softmac_handle_beacon(struct net_device *dev,
+	struct ieee80211_beacon *beacon,
+	struct ieee80211_network *network);
 
 /*** prototypes from _auth.c */
 /* do these have to go into the public header? */
@@ -189,6 +194,7 @@ struct ieee80211softmac_network {
 	    authenticated:1,
 	    auth_desynced_once:1;
 
+	u8 erp_value;				/* Saved ERP value */
 	u16 capabilities;			/* Capabilities bitfield */
 	u8 challenge_len;			/* Auth Challenge length */
 	char *challenge;			/* Challenge Text */
-- 
cgit v1.2.3


From d7712ac254a4ae2e9c927e29e37b8c7ac334e6ad Mon Sep 17 00:00:00 2001
From: Daniel Drake <dsd@gentoo.org>
Date: Tue, 18 Jul 2006 21:34:56 +0100
Subject: [PATCH] softmac: export highest_supported_rate function

zd1211 needs this functionality, no point duplicating it.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211softmac.h                  |  8 ++++++++
 net/ieee80211/softmac/ieee80211softmac_module.c | 16 +++++-----------
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h
index c27d03433c21..425b3a57ac74 100644
--- a/include/net/ieee80211softmac.h
+++ b/include/net/ieee80211softmac.h
@@ -279,6 +279,14 @@ extern void ieee80211softmac_fragment_lost(struct net_device *dev,
  * Note that the rates need to be sorted. */
 extern void ieee80211softmac_set_rates(struct net_device *dev, u8 count, u8 *rates);
 
+/* Finds the highest rate which is:
+ *  1. Present in ri (optionally a basic rate)
+ *  2. Supported by the device
+ *  3. Less than or equal to the user-defined rate
+ */
+extern u8 ieee80211softmac_highest_supported_rate(struct ieee80211softmac_device *mac,
+	struct ieee80211softmac_ratesinfo *ri, int basic_only);
+
 /* Helper function which advises you the rate at which a frame should be
  * transmitted at. */
 static inline u8 ieee80211softmac_suggest_txrate(struct ieee80211softmac_device *mac,
diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c
index c275646b2269..addea1cf73ae 100644
--- a/net/ieee80211/softmac/ieee80211softmac_module.c
+++ b/net/ieee80211/softmac/ieee80211softmac_module.c
@@ -179,21 +179,14 @@ int ieee80211softmac_ratesinfo_rate_supported(struct ieee80211softmac_ratesinfo
 	return 0;
 }
 
-/* Finds the highest rate which is:
- *  1. Present in ri (optionally a basic rate)
- *  2. Supported by the device
- *  3. Less than or equal to the user-defined rate
- */
-static u8 highest_supported_rate(struct ieee80211softmac_device *mac,
+u8 ieee80211softmac_highest_supported_rate(struct ieee80211softmac_device *mac,
 	struct ieee80211softmac_ratesinfo *ri, int basic_only)
 {
 	u8 user_rate = mac->txrates.user_rate;
 	int i;
 
-	if (ri->count == 0) {
-		dprintk(KERN_ERR PFX "empty ratesinfo?\n");
+	if (ri->count == 0)
 		return IEEE80211_CCK_RATE_1MB;
-	}
 
 	for (i = ri->count - 1; i >= 0; i--) {
 		u8 rate = ri->rates[i];
@@ -209,6 +202,7 @@ static u8 highest_supported_rate(struct ieee80211softmac_device *mac,
 	/* If we haven't found a suitable rate by now, just trust the user */
 	return user_rate;
 }
+EXPORT_SYMBOL_GPL(ieee80211softmac_highest_supported_rate);
 
 void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac,
 	u8 erp_value)
@@ -244,13 +238,13 @@ void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac)
 	u32 change = 0;
 
 	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT;
-	txrates->default_rate = highest_supported_rate(mac, &mac->bssinfo.supported_rates, 0);
+	txrates->default_rate = ieee80211softmac_highest_supported_rate(mac, &mac->bssinfo.supported_rates, 0);
 
 	change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK;
 	txrates->default_fallback = lower_rate(mac, txrates->default_rate);
 
 	change |= IEEE80211SOFTMAC_TXRATECHG_MCAST;
-	txrates->mcast_rate = highest_supported_rate(mac, &mac->bssinfo.supported_rates, 1);
+	txrates->mcast_rate = ieee80211softmac_highest_supported_rate(mac, &mac->bssinfo.supported_rates, 1);
 
 	if (mac->txrates_change)
 		mac->txrates_change(mac->dev, change);
-- 
cgit v1.2.3


From f2060f039e8a8bc83b10e6d0f8fb440425560569 Mon Sep 17 00:00:00 2001
From: Daniel Drake <dsd@gentoo.org>
Date: Tue, 18 Jul 2006 21:38:05 +0100
Subject: [PATCH] ieee80211: Make ieee80211_rx_any usable

ieee80211_rx_any is new to 2.6.18-rc1, even though it appears this function
was never completed:

http://lists.sipsolutions.net/pipermail/softmac-dev/2006-February/000103.html

This patch changes ieee80211_rx_any to always claim the skb, which avoids
further driver complexity and the possibility of leaking management frames.
It also exports the function so that people can actually use it.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211.h      |  2 ++
 net/ieee80211/ieee80211_rx.c | 38 ++++++++++++++++++++++++++++----------
 2 files changed, 30 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index aa007f49bf29..b174ebb277a9 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -1259,6 +1259,8 @@ extern int ieee80211_tx_frame(struct ieee80211_device *ieee,
 			      int total_len, int encrypt_mpdu);
 
 /* ieee80211_rx.c */
+extern void ieee80211_rx_any(struct ieee80211_device *ieee,
+		     struct sk_buff *skb, struct ieee80211_rx_stats *stats);
 extern int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb,
 			struct ieee80211_rx_stats *rx_stats);
 /* make sure to set stats->len */
diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c
index f73fc164b9ef..d60358d702d7 100644
--- a/net/ieee80211/ieee80211_rx.c
+++ b/net/ieee80211/ieee80211_rx.c
@@ -779,33 +779,44 @@ int ieee80211_rx(struct ieee80211_device *ieee, struct sk_buff *skb,
 	return 0;
 }
 
-/* Filter out unrelated packets, call ieee80211_rx[_mgt] */
-int ieee80211_rx_any(struct ieee80211_device *ieee,
+/* Filter out unrelated packets, call ieee80211_rx[_mgt]
+ * This function takes over the skb, it should not be used again after calling
+ * this function. */
+void ieee80211_rx_any(struct ieee80211_device *ieee,
 		     struct sk_buff *skb, struct ieee80211_rx_stats *stats)
 {
 	struct ieee80211_hdr_4addr *hdr;
 	int is_packet_for_us;
 	u16 fc;
 
-	if (ieee->iw_mode == IW_MODE_MONITOR)
-		return ieee80211_rx(ieee, skb, stats) ? 0 : -EINVAL;
+	if (ieee->iw_mode == IW_MODE_MONITOR) {
+		if (!ieee80211_rx(ieee, skb, stats))
+			dev_kfree_skb_irq(skb);
+		return;
+	}
+
+	if (skb->len < sizeof(struct ieee80211_hdr))
+		goto drop_free;
 
 	hdr = (struct ieee80211_hdr_4addr *)skb->data;
 	fc = le16_to_cpu(hdr->frame_ctl);
 
 	if ((fc & IEEE80211_FCTL_VERS) != 0)
-		return -EINVAL;
+		goto drop_free;
 		
 	switch (fc & IEEE80211_FCTL_FTYPE) {
 	case IEEE80211_FTYPE_MGMT:
+		if (skb->len < sizeof(struct ieee80211_hdr_3addr))
+			goto drop_free;
 		ieee80211_rx_mgt(ieee, hdr, stats);
-		return 0;
+		dev_kfree_skb_irq(skb);
+		return;
 	case IEEE80211_FTYPE_DATA:
 		break;
 	case IEEE80211_FTYPE_CTL:
-		return 0;
+		return;
 	default:
-		return -EINVAL;
+		return;
 	}
 
 	is_packet_for_us = 0;
@@ -849,8 +860,14 @@ int ieee80211_rx_any(struct ieee80211_device *ieee,
 	}
 
 	if (is_packet_for_us)
-		return (ieee80211_rx(ieee, skb, stats) ? 0 : -EINVAL);
-	return 0;
+		if (!ieee80211_rx(ieee, skb, stats))
+			dev_kfree_skb_irq(skb);
+	return;
+
+drop_free:
+	dev_kfree_skb_irq(skb);
+	ieee->stats.rx_dropped++;
+	return;
 }
 
 #define MGMT_FRAME_FIXED_PART_LENGTH		0x24
@@ -1730,5 +1747,6 @@ void ieee80211_rx_mgt(struct ieee80211_device *ieee,
 	}
 }
 
+EXPORT_SYMBOL_GPL(ieee80211_rx_any);
 EXPORT_SYMBOL(ieee80211_rx_mgt);
 EXPORT_SYMBOL(ieee80211_rx);
-- 
cgit v1.2.3


From 1c57e86d75cf162bdadb3a5fe0cd3f65aa1a9ca3 Mon Sep 17 00:00:00 2001
From: Erich Chen <erich@areca.com.tw>
Date: Wed, 12 Jul 2006 08:59:32 -0700
Subject: [SCSI] arcmsr: initial driver, version 1.20.00.13

arcmsr is a driver for the Areca Raid controller, a host based RAID
subsystem that speaks SCSI at the firmware level.

This patch is quite a clean up over the initial submission with
contributions from:

Randy Dunlap <rdunlap@xenotime.net>
Christoph Hellwig <hch@lst.de>
Matthew Wilcox <matthew@wil.cx>
Adrian Bunk <bunk@stusta.de>

Signed-off-by: Erich Chen <erich@areca.com.tw>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 Documentation/scsi/ChangeLog.arcmsr |   56 ++
 Documentation/scsi/arcmsr_spec.txt  |  574 ++++++++++++++
 drivers/scsi/Kconfig                |   14 +
 drivers/scsi/Makefile               |    1 +
 drivers/scsi/arcmsr/Makefile        |    6 +
 drivers/scsi/arcmsr/arcmsr.h        |  472 +++++++++++
 drivers/scsi/arcmsr/arcmsr_attr.c   |  392 +++++++++
 drivers/scsi/arcmsr/arcmsr_hba.c    | 1496 +++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h             |   17 +
 9 files changed, 3028 insertions(+)
 create mode 100644 Documentation/scsi/ChangeLog.arcmsr
 create mode 100644 Documentation/scsi/arcmsr_spec.txt
 create mode 100644 drivers/scsi/arcmsr/Makefile
 create mode 100644 drivers/scsi/arcmsr/arcmsr.h
 create mode 100644 drivers/scsi/arcmsr/arcmsr_attr.c
 create mode 100644 drivers/scsi/arcmsr/arcmsr_hba.c

(limited to 'include')

diff --git a/Documentation/scsi/ChangeLog.arcmsr b/Documentation/scsi/ChangeLog.arcmsr
new file mode 100644
index 000000000000..162c47fdf45f
--- /dev/null
+++ b/Documentation/scsi/ChangeLog.arcmsr
@@ -0,0 +1,56 @@
+**************************************************************************
+** History
+**
+**   REV#         DATE             NAME         DESCRIPTION
+** 1.00.00.00    3/31/2004       Erich Chen     First release
+** 1.10.00.04    7/28/2004       Erich Chen     modify for ioctl
+** 1.10.00.06    8/28/2004       Erich Chen     modify for 2.6.x
+** 1.10.00.08    9/28/2004       Erich Chen     modify for x86_64
+** 1.10.00.10   10/10/2004       Erich Chen     bug fix for SMP & ioctl
+** 1.20.00.00   11/29/2004       Erich Chen     bug fix with arcmsr_bus_reset when PHY error
+** 1.20.00.02   12/09/2004       Erich Chen     bug fix with over 2T bytes RAID Volume
+** 1.20.00.04    1/09/2005       Erich Chen     fits for Debian linux kernel version 2.2.xx
+** 1.20.00.05    2/20/2005       Erich Chen     cleanly as look like a Linux driver at 2.6.x
+**                                              thanks for peoples kindness comment
+**						Kornel Wieliczek
+**						Christoph Hellwig
+**						Adrian Bunk
+**						Andrew Morton
+**						Christoph Hellwig
+**						James Bottomley
+**						Arjan van de Ven
+** 1.20.00.06    3/12/2005       Erich Chen     fix with arcmsr_pci_unmap_dma "unsigned long" cast,
+**						modify PCCB POOL allocated by "dma_alloc_coherent"
+**						(Kornel Wieliczek's comment)
+** 1.20.00.07    3/23/2005       Erich Chen     bug fix with arcmsr_scsi_host_template_init
+**						occur segmentation fault,
+**						if RAID adapter does not on PCI slot
+**						and modprobe/rmmod this driver twice.
+**						bug fix enormous stack usage (Adrian Bunk's comment)
+** 1.20.00.08    6/23/2005       Erich Chen     bug fix with abort command,
+**						in case of heavy loading when sata cable
+**						working on low quality connection
+** 1.20.00.09    9/12/2005       Erich Chen     bug fix with abort command handling, firmware version check
+**						and firmware update notify for hardware bug fix
+** 1.20.00.10    9/23/2005       Erich Chen     enhance sysfs function for change driver's max tag Q number.
+**						add DMA_64BIT_MASK for backward compatible with all 2.6.x
+**						add some useful message for abort command
+**						add ioctl code 'ARCMSR_IOCTL_FLUSH_ADAPTER_CACHE'
+**						customer can send this command for sync raid volume data
+** 1.20.00.11    9/29/2005       Erich Chen     by comment of Arjan van de Ven fix incorrect msleep redefine
+**						cast off sizeof(dma_addr_t) condition for 64bit pci_set_dma_mask
+** 1.20.00.12    9/30/2005       Erich Chen     bug fix with 64bit platform's ccbs using if over 4G system memory
+**						change 64bit pci_set_consistent_dma_mask into 32bit
+**						increcct adapter count if adapter initialize fail.
+**						miss edit at arcmsr_build_ccb....
+**						psge += sizeof(struct _SG64ENTRY *) =>
+**						psge += sizeof(struct _SG64ENTRY)
+**						64 bits sg entry would be incorrectly calculated
+**						thanks Kornel Wieliczek give me kindly notify
+**						and detail description
+** 1.20.00.13   11/15/2005       Erich Chen     scheduling pending ccb with FIFO
+**						change the architecture of arcmsr command queue list
+**						for linux standard list
+**						enable usage of pci message signal interrupt
+**						follow Randy.Danlup kindness suggestion cleanup this code
+**************************************************************************
\ No newline at end of file
diff --git a/Documentation/scsi/arcmsr_spec.txt b/Documentation/scsi/arcmsr_spec.txt
new file mode 100644
index 000000000000..5e0042340fd3
--- /dev/null
+++ b/Documentation/scsi/arcmsr_spec.txt
@@ -0,0 +1,574 @@
+*******************************************************************************
+**                            ARECA FIRMWARE SPEC
+*******************************************************************************
+**	Usage of IOP331 adapter
+**	(All In/Out is in IOP331's view)
+**	1. Message 0 --> InitThread message and retrun code
+**	2. Doorbell is used for RS-232 emulation
+**		inDoorBell :    bit0 -- data in ready
+**			(DRIVER DATA WRITE OK)
+**				bit1 -- data out has been read
+**			(DRIVER DATA READ OK)
+**		outDooeBell:    bit0 -- data out ready
+**			(IOP331 DATA WRITE OK)
+**				bit1 -- data in has been read
+**			(IOP331 DATA READ OK)
+**	3. Index Memory Usage
+**	offset 0xf00 : for RS232 out (request buffer)
+**	offset 0xe00 : for RS232 in  (scratch buffer)
+**	offset 0xa00 : for inbound message code message_rwbuffer
+**			(driver send to IOP331)
+**	offset 0xa00 : for outbound message code message_rwbuffer
+**			(IOP331 send to driver)
+**	4. RS-232 emulation
+**		Currently 128 byte buffer is used
+**			1st uint32_t : Data length (1--124)
+**			Byte 4--127  : Max 124 bytes of data
+**	5. PostQ
+**	All SCSI Command must be sent through postQ:
+**	(inbound queue port)	Request frame must be 32 bytes aligned
+**	#bit27--bit31 => flag for post ccb
+**	#bit0--bit26  => real address (bit27--bit31) of post arcmsr_cdb
+**		bit31 :
+**			0 : 256 bytes frame
+**			1 : 512 bytes frame
+**		bit30 :
+**			0 : normal request
+**			1 : BIOS request
+**		bit29 : reserved
+**		bit28 : reserved
+**		bit27 : reserved
+**  ---------------------------------------------------------------------------
+**	(outbount queue port)	Request reply
+**	#bit27--bit31
+**		=> flag for reply
+**	#bit0--bit26
+**		=> real address (bit27--bit31) of reply arcmsr_cdb
+**			bit31 : must be 0 (for this type of reply)
+**			bit30 : reserved for BIOS handshake
+**			bit29 : reserved
+**			bit28 :
+**			0 : no error, ignore AdapStatus/DevStatus/SenseData
+**			1 : Error, error code in AdapStatus/DevStatus/SenseData
+**			bit27 : reserved
+**	6. BIOS request
+**		All BIOS request is the same with request from PostQ
+**		Except :
+**			Request frame is sent from configuration space
+**		offset: 0x78 : Request Frame (bit30 == 1)
+**		offset: 0x18 : writeonly to generate
+**					IRQ to IOP331
+**		Completion of request:
+**			(bit30 == 0, bit28==err flag)
+**	7. Definition of SGL entry (structure)
+**	8. Message1 Out - Diag Status Code (????)
+**	9. Message0 message code :
+**		0x00 : NOP
+**		0x01 : Get Config
+**		->offset 0xa00 :for outbound message code message_rwbuffer
+**		(IOP331 send to driver)
+**		Signature             0x87974060(4)
+**		Request len           0x00000200(4)
+**		numbers of queue      0x00000100(4)
+**		SDRAM Size            0x00000100(4)-->256 MB
+**		IDE Channels          0x00000008(4)
+**		vendor                40 bytes char
+**		model                  8 bytes char
+**		FirmVer               16 bytes char
+**		Device Map            16 bytes char
+**		FirmwareVersion DWORD <== Added for checking of
+**						new firmware capability
+**		0x02 : Set Config
+**		->offset 0xa00 :for inbound message code message_rwbuffer
+**		(driver send to IOP331)
+**		Signature             0x87974063(4)
+**		UPPER32 of Request Frame  (4)-->Driver Only
+**		0x03 : Reset (Abort all queued Command)
+**		0x04 : Stop Background Activity
+**		0x05 : Flush Cache
+**		0x06 : Start Background Activity
+**			(re-start if background is halted)
+**		0x07 : Check If Host Command Pending
+**			(Novell May Need This Function)
+**		0x08 : Set controller time
+**		->offset 0xa00 : for inbound message code message_rwbuffer
+**		(driver to IOP331)
+**		byte 0 : 0xaa <-- signature
+**		byte 1 : 0x55 <-- signature
+**		byte 2 : year (04)
+**		byte 3 : month (1..12)
+**		byte 4 : date (1..31)
+**		byte 5 : hour (0..23)
+**		byte 6 : minute (0..59)
+**		byte 7 : second (0..59)
+*******************************************************************************
+*******************************************************************************
+**      	RS-232 Interface for Areca Raid Controller
+**      The low level command interface is exclusive with VT100 terminal
+**  --------------------------------------------------------------------
+**      1. Sequence of command execution
+**  --------------------------------------------------------------------
+**    	(A) Header : 3 bytes sequence (0x5E, 0x01, 0x61)
+**    	(B) Command block : variable length of data including length,
+**    		command code, data and checksum byte
+**    	(C) Return data : variable length of data
+**  --------------------------------------------------------------------
+**    2. Command block
+**  --------------------------------------------------------------------
+**    	(A) 1st byte : command block length (low byte)
+**    	(B) 2nd byte : command block length (high byte)
+**                note ..command block length shouldn't > 2040 bytes,
+**    		length excludes these two bytes
+**    	(C) 3rd byte : command code
+**    	(D) 4th and following bytes : variable length data bytes
+**    		depends on command code
+**    	(E) last byte : checksum byte (sum of 1st byte until last data byte)
+**  --------------------------------------------------------------------
+**    3. Command code and associated data
+**  --------------------------------------------------------------------
+**    	The following are command code defined in raid controller Command
+**    	code 0x10--0x1? are used for system level management,
+**    	no password checking is needed and should be implemented in separate
+**    	well controlled utility and not for end user access.
+**    	Command code 0x20--0x?? always check the password,
+**    	password must be entered to enable these command.
+**    	enum
+**    	{
+**    		GUI_SET_SERIAL=0x10,
+**    		GUI_SET_VENDOR,
+**    		GUI_SET_MODEL,
+**    		GUI_IDENTIFY,
+**    		GUI_CHECK_PASSWORD,
+**    		GUI_LOGOUT,
+**    		GUI_HTTP,
+**    		GUI_SET_ETHERNET_ADDR,
+**    		GUI_SET_LOGO,
+**    		GUI_POLL_EVENT,
+**    		GUI_GET_EVENT,
+**    		GUI_GET_HW_MONITOR,
+**    		//    GUI_QUICK_CREATE=0x20, (function removed)
+**    		GUI_GET_INFO_R=0x20,
+**    		GUI_GET_INFO_V,
+**    		GUI_GET_INFO_P,
+**    		GUI_GET_INFO_S,
+**    		GUI_CLEAR_EVENT,
+**    		GUI_MUTE_BEEPER=0x30,
+**    		GUI_BEEPER_SETTING,
+**    		GUI_SET_PASSWORD,
+**    		GUI_HOST_INTERFACE_MODE,
+**    		GUI_REBUILD_PRIORITY,
+**    		GUI_MAX_ATA_MODE,
+**    		GUI_RESET_CONTROLLER,
+**    		GUI_COM_PORT_SETTING,
+**    		GUI_NO_OPERATION,
+**    		GUI_DHCP_IP,
+**    		GUI_CREATE_PASS_THROUGH=0x40,
+**    		GUI_MODIFY_PASS_THROUGH,
+**    		GUI_DELETE_PASS_THROUGH,
+**    		GUI_IDENTIFY_DEVICE,
+**    		GUI_CREATE_RAIDSET=0x50,
+**    		GUI_DELETE_RAIDSET,
+**    		GUI_EXPAND_RAIDSET,
+**    		GUI_ACTIVATE_RAIDSET,
+**    		GUI_CREATE_HOT_SPARE,
+**    		GUI_DELETE_HOT_SPARE,
+**    		GUI_CREATE_VOLUME=0x60,
+**    		GUI_MODIFY_VOLUME,
+**    		GUI_DELETE_VOLUME,
+**    		GUI_START_CHECK_VOLUME,
+**    		GUI_STOP_CHECK_VOLUME
+**    	};
+**    Command description :
+**    	GUI_SET_SERIAL : Set the controller serial#
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x10
+**    		byte 3          : password length (should be 0x0f)
+**    		byte 4-0x13     : should be "ArEcATecHnoLogY"
+**    		byte 0x14--0x23 : Serial number string (must be 16 bytes)
+**      GUI_SET_VENDOR : Set vendor string for the controller
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x11
+**    		byte 3          : password length (should be 0x08)
+**    		byte 4-0x13     : should be "ArEcAvAr"
+**    		byte 0x14--0x3B : vendor string (must be 40 bytes)
+**      GUI_SET_MODEL : Set the model name of the controller
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x12
+**    		byte 3          : password length (should be 0x08)
+**    		byte 4-0x13     : should be "ArEcAvAr"
+**    		byte 0x14--0x1B : model string (must be 8 bytes)
+**      GUI_IDENTIFY : Identify device
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x13
+**    		                  return "Areca RAID Subsystem "
+**      GUI_CHECK_PASSWORD : Verify password
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x14
+**    		byte 3          : password length
+**    		byte 4-0x??     : user password to be checked
+**      GUI_LOGOUT : Logout GUI (force password checking on next command)
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x15
+**      GUI_HTTP : HTTP interface (reserved for Http proxy service)(0x16)
+**
+**      GUI_SET_ETHERNET_ADDR : Set the ethernet MAC address
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x17
+**    		byte 3          : password length (should be 0x08)
+**    		byte 4-0x13     : should be "ArEcAvAr"
+**    		byte 0x14--0x19 : Ethernet MAC address (must be 6 bytes)
+**      GUI_SET_LOGO : Set logo in HTTP
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x18
+**    		byte 3          : Page# (0/1/2/3) (0xff --> clear OEM logo)
+**    		byte 4/5/6/7    : 0x55/0xaa/0xa5/0x5a
+**    		byte 8          : TITLE.JPG data (each page must be 2000 bytes)
+**    		                  note page0 1st 2 byte must be
+**    					actual length of the JPG file
+**      GUI_POLL_EVENT : Poll If Event Log Changed
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x19
+**      GUI_GET_EVENT : Read Event
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x1a
+**    		byte 3          : Event Page (0:1st page/1/2/3:last page)
+**      GUI_GET_HW_MONITOR : Get HW monitor data
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x1b
+**    		byte 3 			: # of FANs(example 2)
+**    		byte 4 			: # of Voltage sensor(example 3)
+**    		byte 5 			: # of temperature sensor(example 2)
+**    		byte 6 			: # of power
+**    		byte 7/8        : Fan#0 (RPM)
+**    		byte 9/10       : Fan#1
+**    		byte 11/12 		: Voltage#0 original value in *1000
+**    		byte 13/14 		: Voltage#0 value
+**    		byte 15/16 		: Voltage#1 org
+**    		byte 17/18 		: Voltage#1
+**    		byte 19/20 		: Voltage#2 org
+**    		byte 21/22 		: Voltage#2
+**    		byte 23 		: Temp#0
+**    		byte 24 		: Temp#1
+**    		byte 25 		: Power indicator (bit0 : power#0,
+**    						 bit1 : power#1)
+**    		byte 26 		: UPS indicator
+**      GUI_QUICK_CREATE : Quick create raid/volume set
+**    	    byte 0,1        : length
+**    	    byte 2          : command code 0x20
+**    	    byte 3/4/5/6    : raw capacity
+**    	    byte 7 			: raid level
+**    	    byte 8 			: stripe size
+**    	    byte 9 			: spare
+**    	    byte 10/11/12/13: device mask (the devices to create raid/volume)
+**    		This function is removed, application like
+**    		to implement quick create function
+**    	need to use GUI_CREATE_RAIDSET and GUI_CREATE_VOLUMESET function.
+**      GUI_GET_INFO_R : Get Raid Set Information
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x20
+**    		byte 3          : raidset#
+**    	typedef struct sGUI_RAIDSET
+**    	{
+**    		BYTE grsRaidSetName[16];
+**    		DWORD grsCapacity;
+**    		DWORD grsCapacityX;
+**    		DWORD grsFailMask;
+**    		BYTE grsDevArray[32];
+**    		BYTE grsMemberDevices;
+**    		BYTE grsNewMemberDevices;
+**    		BYTE grsRaidState;
+**    		BYTE grsVolumes;
+**    		BYTE grsVolumeList[16];
+**    		BYTE grsRes1;
+**    		BYTE grsRes2;
+**    		BYTE grsRes3;
+**    		BYTE grsFreeSegments;
+**    		DWORD grsRawStripes[8];
+**    		DWORD grsRes4;
+**    		DWORD grsRes5; //     Total to 128 bytes
+**    		DWORD grsRes6; //     Total to 128 bytes
+**    	} sGUI_RAIDSET, *pGUI_RAIDSET;
+**      GUI_GET_INFO_V : Get Volume Set Information
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x21
+**    		byte 3          : volumeset#
+**    	typedef struct sGUI_VOLUMESET
+**    	{
+**    		BYTE gvsVolumeName[16]; //     16
+**    		DWORD gvsCapacity;
+**    		DWORD gvsCapacityX;
+**    		DWORD gvsFailMask;
+**    		DWORD gvsStripeSize;
+**    		DWORD gvsNewFailMask;
+**    		DWORD gvsNewStripeSize;
+**    		DWORD gvsVolumeStatus;
+**    		DWORD gvsProgress; //     32
+**    		sSCSI_ATTR gvsScsi;
+**    		BYTE gvsMemberDisks;
+**    		BYTE gvsRaidLevel; //     8
+**    		BYTE gvsNewMemberDisks;
+**    		BYTE gvsNewRaidLevel;
+**    		BYTE gvsRaidSetNumber;
+**    		BYTE gvsRes0; //     4
+**    		BYTE gvsRes1[4]; //     64 bytes
+**    	} sGUI_VOLUMESET, *pGUI_VOLUMESET;
+**      GUI_GET_INFO_P : Get Physical Drive Information
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x22
+**    		byte 3          : drive # (from 0 to max-channels - 1)
+**    	typedef struct sGUI_PHY_DRV
+**    	{
+**    		BYTE gpdModelName[40];
+**    		BYTE gpdSerialNumber[20];
+**    		BYTE gpdFirmRev[8];
+**    		DWORD gpdCapacity;
+**    		DWORD gpdCapacityX; //     Reserved for expansion
+**    		BYTE gpdDeviceState;
+**    		BYTE gpdPioMode;
+**    		BYTE gpdCurrentUdmaMode;
+**    		BYTE gpdUdmaMode;
+**    		BYTE gpdDriveSelect;
+**    		BYTE gpdRaidNumber; //     0xff if not belongs to a raid set
+**    		sSCSI_ATTR gpdScsi;
+**    		BYTE gpdReserved[40]; //     Total to 128 bytes
+**    	} sGUI_PHY_DRV, *pGUI_PHY_DRV;
+**    	GUI_GET_INFO_S : Get System Information
+**      	byte 0,1        : length
+**      	byte 2          : command code 0x23
+**    	typedef struct sCOM_ATTR
+**    	{
+**    		BYTE comBaudRate;
+**    		BYTE comDataBits;
+**    		BYTE comStopBits;
+**    		BYTE comParity;
+**    		BYTE comFlowControl;
+**    	} sCOM_ATTR, *pCOM_ATTR;
+**    	typedef struct sSYSTEM_INFO
+**    	{
+**    		BYTE gsiVendorName[40];
+**    		BYTE gsiSerialNumber[16];
+**    		BYTE gsiFirmVersion[16];
+**    		BYTE gsiBootVersion[16];
+**    		BYTE gsiMbVersion[16];
+**    		BYTE gsiModelName[8];
+**    		BYTE gsiLocalIp[4];
+**    		BYTE gsiCurrentIp[4];
+**    		DWORD gsiTimeTick;
+**    		DWORD gsiCpuSpeed;
+**    		DWORD gsiICache;
+**    		DWORD gsiDCache;
+**    		DWORD gsiScache;
+**    		DWORD gsiMemorySize;
+**    		DWORD gsiMemorySpeed;
+**    		DWORD gsiEvents;
+**    		BYTE gsiMacAddress[6];
+**    		BYTE gsiDhcp;
+**    		BYTE gsiBeeper;
+**    		BYTE gsiChannelUsage;
+**    		BYTE gsiMaxAtaMode;
+**    		BYTE gsiSdramEcc; //     1:if ECC enabled
+**    		BYTE gsiRebuildPriority;
+**    		sCOM_ATTR gsiComA; //     5 bytes
+**    		sCOM_ATTR gsiComB; //     5 bytes
+**    		BYTE gsiIdeChannels;
+**    		BYTE gsiScsiHostChannels;
+**    		BYTE gsiIdeHostChannels;
+**    		BYTE gsiMaxVolumeSet;
+**    		BYTE gsiMaxRaidSet;
+**    		BYTE gsiEtherPort; //     1:if ether net port supported
+**    		BYTE gsiRaid6Engine; //     1:Raid6 engine supported
+**    		BYTE gsiRes[75];
+**    	} sSYSTEM_INFO, *pSYSTEM_INFO;
+**    	GUI_CLEAR_EVENT : Clear System Event
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x24
+**      GUI_MUTE_BEEPER : Mute current beeper
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x30
+**      GUI_BEEPER_SETTING : Disable beeper
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x31
+**    		byte 3          : 0->disable, 1->enable
+**      GUI_SET_PASSWORD : Change password
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x32
+**    		byte 3 			: pass word length ( must <= 15 )
+**    		byte 4 			: password (must be alpha-numerical)
+**    	GUI_HOST_INTERFACE_MODE : Set host interface mode
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x33
+**    		byte 3 			: 0->Independent, 1->cluster
+**      GUI_REBUILD_PRIORITY : Set rebuild priority
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x34
+**    		byte 3 			: 0/1/2/3 (low->high)
+**      GUI_MAX_ATA_MODE : Set maximum ATA mode to be used
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x35
+**    		byte 3 			: 0/1/2/3 (133/100/66/33)
+**      GUI_RESET_CONTROLLER : Reset Controller
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x36
+**                            *Response with VT100 screen (discard it)
+**      GUI_COM_PORT_SETTING : COM port setting
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x37
+**    		byte 3 			: 0->COMA (term port),
+**    					  1->COMB (debug port)
+**    		byte 4 			: 0/1/2/3/4/5/6/7
+**    			(1200/2400/4800/9600/19200/38400/57600/115200)
+**    		byte 5 			: data bit
+**    					(0:7 bit, 1:8 bit : must be 8 bit)
+**    		byte 6 			: stop bit (0:1, 1:2 stop bits)
+**    		byte 7 			: parity (0:none, 1:off, 2:even)
+**    		byte 8 			: flow control
+**    			(0:none, 1:xon/xoff, 2:hardware => must use none)
+**      GUI_NO_OPERATION : No operation
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x38
+**      GUI_DHCP_IP : Set DHCP option and local IP address
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x39
+**    		byte 3          : 0:dhcp disabled, 1:dhcp enabled
+**    		byte 4/5/6/7    : IP address
+**      GUI_CREATE_PASS_THROUGH : Create pass through disk
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x40
+**    		byte 3 			: device #
+**    		byte 4 			: scsi channel (0/1)
+**    		byte 5 			: scsi id (0-->15)
+**    		byte 6 			: scsi lun (0-->7)
+**    		byte 7 			: tagged queue (1 : enabled)
+**    		byte 8 			: cache mode (1 : enabled)
+**    		byte 9 			: max speed (0/1/2/3/4,
+**    			async/20/40/80/160 for scsi)
+**    			(0/1/2/3/4, 33/66/100/133/150 for ide  )
+**      GUI_MODIFY_PASS_THROUGH : Modify pass through disk
+**    		byte 0,1        : length
+**    		byte 2 			: command code 0x41
+**    		byte 3 			: device #
+**    		byte 4 			: scsi channel (0/1)
+**    		byte 5 			: scsi id (0-->15)
+**    		byte 6 			: scsi lun (0-->7)
+**    		byte 7 			: tagged queue (1 : enabled)
+**    		byte 8 			: cache mode (1 : enabled)
+**    		byte 9 			: max speed (0/1/2/3/4,
+**    					async/20/40/80/160 for scsi)
+**    			(0/1/2/3/4, 33/66/100/133/150 for ide  )
+**      GUI_DELETE_PASS_THROUGH : Delete pass through disk
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x42
+**    		byte 3          : device# to be deleted
+**      GUI_IDENTIFY_DEVICE : Identify Device
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x43
+**    		byte 3          : Flash Method
+**    				(0:flash selected, 1:flash not selected)
+**    		byte 4/5/6/7    : IDE device mask to be flashed
+**                           note .... no response data available
+**    	GUI_CREATE_RAIDSET : Create Raid Set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x50
+**    		byte 3/4/5/6    : device mask
+**    		byte 7-22       : raidset name (if byte 7 == 0:use default)
+**      GUI_DELETE_RAIDSET : Delete Raid Set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x51
+**    		byte 3          : raidset#
+**    	GUI_EXPAND_RAIDSET : Expand Raid Set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x52
+**    		byte 3          : raidset#
+**    		byte 4/5/6/7    : device mask for expansion
+**    		byte 8/9/10     : (8:0 no change, 1 change, 0xff:terminate,
+**    				9:new raid level,
+**    				10:new stripe size
+**    				0/1/2/3/4/5->4/8/16/32/64/128K )
+**    		byte 11/12/13   : repeat for each volume in the raidset
+**      GUI_ACTIVATE_RAIDSET : Activate incomplete raid set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x53
+**    		byte 3          : raidset#
+**      GUI_CREATE_HOT_SPARE : Create hot spare disk
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x54
+**    		byte 3/4/5/6    : device mask for hot spare creation
+**    	GUI_DELETE_HOT_SPARE : Delete hot spare disk
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x55
+**    		byte 3/4/5/6    : device mask for hot spare deletion
+**    	GUI_CREATE_VOLUME : Create volume set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x60
+**    		byte 3          : raidset#
+**    		byte 4-19       : volume set name
+**    				(if byte4 == 0, use default)
+**    		byte 20-27      : volume capacity (blocks)
+**    		byte 28 		: raid level
+**    		byte 29 		: stripe size
+**    				(0/1/2/3/4/5->4/8/16/32/64/128K)
+**    		byte 30 		: channel
+**    		byte 31 		: ID
+**    		byte 32 		: LUN
+**    		byte 33 		: 1 enable tag
+**    		byte 34 		: 1 enable cache
+**    		byte 35 		: speed
+**    		(0/1/2/3/4->async/20/40/80/160 for scsi)
+**    		(0/1/2/3/4->33/66/100/133/150 for IDE  )
+**    		byte 36 		: 1 to select quick init
+**
+**    	GUI_MODIFY_VOLUME : Modify volume Set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x61
+**    		byte 3          : volumeset#
+**    		byte 4-19       : new volume set name
+**    		(if byte4 == 0, not change)
+**    		byte 20-27      : new volume capacity (reserved)
+**    		byte 28 		: new raid level
+**    		byte 29 		: new stripe size
+**    		(0/1/2/3/4/5->4/8/16/32/64/128K)
+**    		byte 30 		: new channel
+**    		byte 31 		: new ID
+**    		byte 32 		: new LUN
+**    		byte 33 		: 1 enable tag
+**    		byte 34 		: 1 enable cache
+**    		byte 35 		: speed
+**    		(0/1/2/3/4->async/20/40/80/160 for scsi)
+**    		(0/1/2/3/4->33/66/100/133/150 for IDE  )
+**    	GUI_DELETE_VOLUME : Delete volume set
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x62
+**    		byte 3          : volumeset#
+**    	GUI_START_CHECK_VOLUME : Start volume consistency check
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x63
+**    		byte 3          : volumeset#
+**    	GUI_STOP_CHECK_VOLUME : Stop volume consistency check
+**    		byte 0,1        : length
+**    		byte 2          : command code 0x64
+** ---------------------------------------------------------------------
+**    4. Returned data
+** ---------------------------------------------------------------------
+**    	(A) Header          : 3 bytes sequence (0x5E, 0x01, 0x61)
+**    	(B) Length          : 2 bytes
+**    			(low byte 1st, excludes length and checksum byte)
+**    	(C) status or data  :
+**           <1> If length == 1 ==> 1 byte status code
+**    		#define GUI_OK                    0x41
+**    		#define GUI_RAIDSET_NOT_NORMAL    0x42
+**    		#define GUI_VOLUMESET_NOT_NORMAL  0x43
+**    		#define GUI_NO_RAIDSET            0x44
+**    		#define GUI_NO_VOLUMESET          0x45
+**    		#define GUI_NO_PHYSICAL_DRIVE     0x46
+**    		#define GUI_PARAMETER_ERROR       0x47
+**    		#define GUI_UNSUPPORTED_COMMAND   0x48
+**    		#define GUI_DISK_CONFIG_CHANGED   0x49
+**    		#define GUI_INVALID_PASSWORD      0x4a
+**    		#define GUI_NO_DISK_SPACE         0x4b
+**    		#define GUI_CHECKSUM_ERROR        0x4c
+**    		#define GUI_PASSWORD_REQUIRED     0x4d
+**           <2> If length > 1 ==>
+**    		data block returned from controller
+**    		and the contents depends on the command code
+**    	(E) Checksum        : checksum of length and status or data byte
+**************************************************************************
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index 96a81cd17617..d61662c1a0ee 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -469,6 +469,20 @@ config SCSI_IN2000
 	  To compile this driver as a module, choose M here: the
 	  module will be called in2000.
 
+config SCSI_ARCMSR
+	tristate "ARECA ARC11X0[PCI-X]/ARC12X0[PCI-EXPRESS] SATA-RAID support"
+	depends on PCI && SCSI
+	help
+	  This driver supports all of ARECA's SATA RAID controller cards.
+	  This is an ARECA-maintained driver by Erich Chen.
+	  If you have any problems, please mail to: < erich@areca.com.tw >
+	  Areca supports Linux RAID config tools.
+
+	  < http://www.areca.com.tw >
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called arcmsr (modprobe arcmsr).
+
 source "drivers/scsi/megaraid/Kconfig.megaraid"
 
 config SCSI_SATA
diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index ebd0cf00bf3e..b2de9bfdfdcd 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -59,6 +59,7 @@ obj-$(CONFIG_SCSI_PSI240I)	+= psi240i.o
 obj-$(CONFIG_SCSI_BUSLOGIC)	+= BusLogic.o
 obj-$(CONFIG_SCSI_DPT_I2O)	+= dpt_i2o.o
 obj-$(CONFIG_SCSI_U14_34F)	+= u14-34f.o
+obj-$(CONFIG_SCSI_ARCMSR)	+= arcmsr/
 obj-$(CONFIG_SCSI_ULTRASTOR)	+= ultrastor.o
 obj-$(CONFIG_SCSI_AHA152X)	+= aha152x.o
 obj-$(CONFIG_SCSI_AHA1542)	+= aha1542.o
diff --git a/drivers/scsi/arcmsr/Makefile b/drivers/scsi/arcmsr/Makefile
new file mode 100644
index 000000000000..721aced39168
--- /dev/null
+++ b/drivers/scsi/arcmsr/Makefile
@@ -0,0 +1,6 @@
+# File: drivers/arcmsr/Makefile
+# Makefile for the ARECA PCI-X PCI-EXPRESS SATA RAID controllers SCSI driver.
+
+arcmsr-objs := arcmsr_attr.o arcmsr_hba.o
+
+obj-$(CONFIG_SCSI_ARCMSR) := arcmsr.o
diff --git a/drivers/scsi/arcmsr/arcmsr.h b/drivers/scsi/arcmsr/arcmsr.h
new file mode 100644
index 000000000000..aff96db9ccf6
--- /dev/null
+++ b/drivers/scsi/arcmsr/arcmsr.h
@@ -0,0 +1,472 @@
+/*
+*******************************************************************************
+**        O.S   : Linux
+**   FILE NAME  : arcmsr.h
+**        BY    : Erich Chen
+**   Description: SCSI RAID Device Driver for
+**                ARECA RAID Host adapter
+*******************************************************************************
+** Copyright (C) 2002 - 2005, Areca Technology Corporation All rights reserved.
+**
+**     Web site: www.areca.com.tw
+**       E-mail: erich@areca.com.tw
+**
+** This program is free software; you can redistribute it and/or modify
+** it under the terms of the GNU General Public License version 2 as
+** published by the Free Software Foundation.
+** This program is distributed in the hope that it will be useful,
+** but WITHOUT ANY WARRANTY; without even the implied warranty of
+** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+** GNU General Public License for more details.
+*******************************************************************************
+** Redistribution and use in source and binary forms, with or without
+** modification, are permitted provided that the following conditions
+** are met:
+** 1. Redistributions of source code must retain the above copyright
+**    notice, this list of conditions and the following disclaimer.
+** 2. Redistributions in binary form must reproduce the above copyright
+**    notice, this list of conditions and the following disclaimer in the
+**    documentation and/or other materials provided with the distribution.
+** 3. The name of the author may not be used to endorse or promote products
+**    derived from this software without specific prior written permission.
+**
+** THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
+** IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
+** OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+** IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+** INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES(INCLUDING, BUT
+** NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+** DATA, OR PROFITS; OR BUSINESS INTERRUPTION)HOWEVER CAUSED AND ON ANY
+** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+**(INCLUDING NEGLIGENCE OR OTHERWISE)ARISING IN ANY WAY OUT OF THE USE OF
+** THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+*******************************************************************************
+*/
+#include <linux/interrupt.h>
+
+struct class_device_attribute;
+
+#define ARCMSR_MAX_OUTSTANDING_CMD 						256
+#define ARCMSR_MAX_FREECCB_NUM							288
+#define ARCMSR_DRIVER_VERSION				"Driver Version 1.20.00.13"
+#define ARCMSR_SCSI_INITIATOR_ID						255
+#define ARCMSR_MAX_XFER_SECTORS							512
+#define ARCMSR_MAX_TARGETID							 17
+#define ARCMSR_MAX_TARGETLUN							  8
+#define ARCMSR_MAX_CMD_PERLUN				 ARCMSR_MAX_OUTSTANDING_CMD
+#define ARCMSR_MAX_QBUFFER						       4096
+#define ARCMSR_MAX_SG_ENTRIES							 38
+
+/*
+*******************************************************************************
+**        split 64bits dma addressing
+*******************************************************************************
+*/
+#define dma_addr_hi32(addr)               (uint32_t) ((addr>>16)>>16)
+#define dma_addr_lo32(addr)               (uint32_t) (addr & 0xffffffff)
+/*
+*******************************************************************************
+**        MESSAGE CONTROL CODE
+*******************************************************************************
+*/
+struct CMD_MESSAGE
+{
+      uint32_t HeaderLength;
+      uint8_t  Signature[8];
+      uint32_t Timeout;
+      uint32_t ControlCode;
+      uint32_t ReturnCode;
+      uint32_t Length;
+};
+/*
+*******************************************************************************
+**        IOP Message Transfer Data for user space
+*******************************************************************************
+*/
+struct CMD_MESSAGE_FIELD
+{
+    struct CMD_MESSAGE			cmdmessage;
+    uint8_t				messagedatabuffer[1032];
+};
+/* IOP message transfer */
+#define ARCMSR_MESSAGE_FAIL             0x0001
+/* DeviceType */
+#define ARECA_SATA_RAID				0x90000000
+/* FunctionCode */
+#define FUNCTION_READ_RQBUFFER			0x0801
+#define FUNCTION_WRITE_WQBUFFER			0x0802
+#define FUNCTION_CLEAR_RQBUFFER			0x0803
+#define FUNCTION_CLEAR_WQBUFFER			0x0804
+#define FUNCTION_CLEAR_ALLQBUFFER		0x0805
+#define FUNCTION_RETURN_CODE_3F			0x0806
+#define FUNCTION_SAY_HELLO			0x0807
+#define FUNCTION_SAY_GOODBYE			0x0808
+#define FUNCTION_FLUSH_ADAPTER_CACHE		0x0809
+/* ARECA IO CONTROL CODE*/
+#define ARCMSR_MESSAGE_READ_RQBUFFER       \
+	ARECA_SATA_RAID | FUNCTION_READ_RQBUFFER
+#define ARCMSR_MESSAGE_WRITE_WQBUFFER      \
+	ARECA_SATA_RAID | FUNCTION_WRITE_WQBUFFER
+#define ARCMSR_MESSAGE_CLEAR_RQBUFFER      \
+	ARECA_SATA_RAID | FUNCTION_CLEAR_RQBUFFER
+#define ARCMSR_MESSAGE_CLEAR_WQBUFFER      \
+	ARECA_SATA_RAID | FUNCTION_CLEAR_WQBUFFER
+#define ARCMSR_MESSAGE_CLEAR_ALLQBUFFER    \
+	ARECA_SATA_RAID | FUNCTION_CLEAR_ALLQBUFFER
+#define ARCMSR_MESSAGE_RETURN_CODE_3F      \
+	ARECA_SATA_RAID | FUNCTION_RETURN_CODE_3F
+#define ARCMSR_MESSAGE_SAY_HELLO           \
+	ARECA_SATA_RAID | FUNCTION_SAY_HELLO
+#define ARCMSR_MESSAGE_SAY_GOODBYE         \
+	ARECA_SATA_RAID | FUNCTION_SAY_GOODBYE
+#define ARCMSR_MESSAGE_FLUSH_ADAPTER_CACHE \
+	ARECA_SATA_RAID | FUNCTION_FLUSH_ADAPTER_CACHE
+/* ARECA IOCTL ReturnCode */
+#define ARCMSR_MESSAGE_RETURNCODE_OK              0x00000001
+#define ARCMSR_MESSAGE_RETURNCODE_ERROR           0x00000006
+#define ARCMSR_MESSAGE_RETURNCODE_3F              0x0000003F
+/*
+*************************************************************
+**   structure for holding DMA address data
+*************************************************************
+*/
+#define IS_SG64_ADDR                0x01000000 /* bit24 */
+struct  SG32ENTRY
+{
+	uint32_t					length;
+	uint32_t					address;
+};
+struct  SG64ENTRY
+{
+ 	uint32_t					length;
+ 	uint32_t					address;
+ 	uint32_t					addresshigh;
+};
+struct SGENTRY_UNION
+{
+	union
+	{
+		struct SG32ENTRY            sg32entry;
+		struct SG64ENTRY            sg64entry;
+	}u;
+};
+/*
+********************************************************************
+**      Q Buffer of IOP Message Transfer
+********************************************************************
+*/
+struct QBUFFER
+{
+	uint32_t      data_len;
+	uint8_t       data[124];
+};
+/*
+*******************************************************************************
+**      FIRMWARE INFO
+*******************************************************************************
+*/
+struct FIRMWARE_INFO
+{
+	uint32_t      signature;                /*0, 00-03*/
+	uint32_t      request_len;              /*1, 04-07*/
+	uint32_t      numbers_queue;            /*2, 08-11*/
+	uint32_t      sdram_size;               /*3, 12-15*/
+	uint32_t      ide_channels;             /*4, 16-19*/
+	char          vendor[40];               /*5, 20-59*/
+	char          model[8];                 /*15, 60-67*/
+	char          firmware_ver[16];         /*17, 68-83*/
+	char          device_map[16];           /*21, 84-99*/
+};
+/* signature of set and get firmware config */
+#define ARCMSR_SIGNATURE_GET_CONFIG                   0x87974060
+#define ARCMSR_SIGNATURE_SET_CONFIG                   0x87974063
+/* message code of inbound message register */
+#define ARCMSR_INBOUND_MESG0_NOP                      0x00000000
+#define ARCMSR_INBOUND_MESG0_GET_CONFIG               0x00000001
+#define ARCMSR_INBOUND_MESG0_SET_CONFIG               0x00000002
+#define ARCMSR_INBOUND_MESG0_ABORT_CMD                0x00000003
+#define ARCMSR_INBOUND_MESG0_STOP_BGRB                0x00000004
+#define ARCMSR_INBOUND_MESG0_FLUSH_CACHE              0x00000005
+#define ARCMSR_INBOUND_MESG0_START_BGRB               0x00000006
+#define ARCMSR_INBOUND_MESG0_CHK331PENDING            0x00000007
+#define ARCMSR_INBOUND_MESG0_SYNC_TIMER               0x00000008
+/* doorbell interrupt generator */
+#define ARCMSR_INBOUND_DRIVER_DATA_WRITE_OK           0x00000001
+#define ARCMSR_INBOUND_DRIVER_DATA_READ_OK            0x00000002
+#define ARCMSR_OUTBOUND_IOP331_DATA_WRITE_OK          0x00000001
+#define ARCMSR_OUTBOUND_IOP331_DATA_READ_OK           0x00000002
+/* ccb areca cdb flag */
+#define ARCMSR_CCBPOST_FLAG_SGL_BSIZE                 0x80000000
+#define ARCMSR_CCBPOST_FLAG_IAM_BIOS                  0x40000000
+#define ARCMSR_CCBREPLY_FLAG_IAM_BIOS                 0x40000000
+#define ARCMSR_CCBREPLY_FLAG_ERROR                    0x10000000
+/* outbound firmware ok */
+#define ARCMSR_OUTBOUND_MESG1_FIRMWARE_OK             0x80000000
+/*
+*******************************************************************************
+**    ARECA SCSI COMMAND DESCRIPTOR BLOCK size 0x1F8 (504)
+*******************************************************************************
+*/
+struct ARCMSR_CDB
+{
+	uint8_t							Bus;
+	uint8_t							TargetID;
+	uint8_t							LUN;
+	uint8_t							Function;
+
+	uint8_t							CdbLength;
+	uint8_t							sgcount;
+	uint8_t							Flags;
+#define ARCMSR_CDB_FLAG_SGL_BSIZE          0x01
+#define ARCMSR_CDB_FLAG_BIOS               0x02
+#define ARCMSR_CDB_FLAG_WRITE              0x04
+#define ARCMSR_CDB_FLAG_SIMPLEQ            0x00
+#define ARCMSR_CDB_FLAG_HEADQ              0x08
+#define ARCMSR_CDB_FLAG_ORDEREDQ           0x10
+	uint8_t							Reserved1;
+
+	uint32_t						Context;
+	uint32_t						DataLength;
+
+	uint8_t							Cdb[16];
+
+	uint8_t							DeviceStatus;
+#define ARCMSR_DEV_CHECK_CONDITION          0x02
+#define ARCMSR_DEV_SELECT_TIMEOUT			0xF0
+#define ARCMSR_DEV_ABORTED				0xF1
+#define ARCMSR_DEV_INIT_FAIL				0xF2
+	uint8_t							SenseData[15];
+
+	union
+	{
+		struct SG32ENTRY                sg32entry[ARCMSR_MAX_SG_ENTRIES];
+		struct SG64ENTRY                sg64entry[ARCMSR_MAX_SG_ENTRIES];
+	} u;
+};
+/*
+*******************************************************************************
+**     Messaging Unit (MU) of the Intel R 80331 I/O processor (80331)
+*******************************************************************************
+*/
+struct MessageUnit
+{
+	uint32_t	resrved0[4];			/*0000 000F*/
+	uint32_t	inbound_msgaddr0;		/*0010 0013*/
+	uint32_t	inbound_msgaddr1;		/*0014 0017*/
+	uint32_t	outbound_msgaddr0;		/*0018 001B*/
+	uint32_t	outbound_msgaddr1;		/*001C 001F*/
+	uint32_t	inbound_doorbell;		/*0020 0023*/
+	uint32_t	inbound_intstatus;		/*0024 0027*/
+	uint32_t	inbound_intmask;		/*0028 002B*/
+	uint32_t	outbound_doorbell;		/*002C 002F*/
+	uint32_t	outbound_intstatus;		/*0030 0033*/
+	uint32_t	outbound_intmask;		/*0034 0037*/
+	uint32_t	reserved1[2];			/*0038 003F*/
+	uint32_t	inbound_queueport;		/*0040 0043*/
+	uint32_t	outbound_queueport;     	/*0044 0047*/
+	uint32_t	reserved2[2];			/*0048 004F*/
+	uint32_t	reserved3[492];			/*0050 07FF 492*/
+	uint32_t	reserved4[128];			/*0800 09FF 128*/
+	uint32_t	message_rwbuffer[256];		/*0a00 0DFF 256*/
+	uint32_t	message_wbuffer[32];		/*0E00 0E7F  32*/
+	uint32_t	reserved5[32];			/*0E80 0EFF  32*/
+	uint32_t	message_rbuffer[32];		/*0F00 0F7F  32*/
+	uint32_t	reserved6[32];			/*0F80 0FFF  32*/
+};
+/*
+*******************************************************************************
+**                 Adapter Control Block
+*******************************************************************************
+*/
+struct AdapterControlBlock
+{
+	struct pci_dev *		pdev;
+	struct Scsi_Host *		host;
+	unsigned long			vir2phy_offset;
+	/* Offset is used in making arc cdb physical to virtual calculations */
+	uint32_t			outbound_int_enable;
+
+	struct MessageUnit __iomem *		pmu;
+	/* message unit ATU inbound base address0 */
+
+	uint32_t			acb_flags;
+#define ACB_F_SCSISTOPADAPTER         0x0001
+#define ACB_F_MSG_STOP_BGRB           0x0002
+	/* stop RAID background rebuild */
+#define ACB_F_MSG_START_BGRB          0x0004
+	/* stop RAID background rebuild */
+#define ACB_F_IOPDATA_OVERFLOW        0x0008
+	/* iop message data rqbuffer overflow */
+#define ACB_F_MESSAGE_WQBUFFER_CLEARED  0x0010
+	/* message clear wqbuffer */
+#define ACB_F_MESSAGE_RQBUFFER_CLEARED  0x0020
+	/* message clear rqbuffer */
+#define ACB_F_MESSAGE_WQBUFFER_READED   0x0040
+#define ACB_F_BUS_RESET               0x0080
+#define ACB_F_IOP_INITED              0x0100
+	/* iop init */
+
+	struct CommandControlBlock *			pccb_pool[ARCMSR_MAX_FREECCB_NUM];
+	/* used for memory free */
+	struct list_head		ccb_free_list;
+	/* head of free ccb list */
+	atomic_t			ccboutstandingcount;
+
+	void *				dma_coherent;
+	/* dma_coherent used for memory free */
+	dma_addr_t			dma_coherent_handle;
+	/* dma_coherent_handle used for memory free */
+
+	uint8_t				rqbuffer[ARCMSR_MAX_QBUFFER];
+	/* data collection buffer for read from 80331 */
+	int32_t				rqbuf_firstindex;
+	/* first of read buffer  */
+	int32_t				rqbuf_lastindex;
+	/* last of read buffer   */
+	uint8_t				wqbuffer[ARCMSR_MAX_QBUFFER];
+	/* data collection buffer for write to 80331  */
+	int32_t				wqbuf_firstindex;
+	/* first of write buffer */
+	int32_t				wqbuf_lastindex;
+	/* last of write buffer  */
+	uint8_t				devstate[ARCMSR_MAX_TARGETID][ARCMSR_MAX_TARGETLUN];
+	/* id0 ..... id15, lun0...lun7 */
+#define ARECA_RAID_GONE               0x55
+#define ARECA_RAID_GOOD               0xaa
+	uint32_t			num_resets;
+	uint32_t			num_aborts;
+	uint32_t			firm_request_len;
+	uint32_t			firm_numbers_queue;
+	uint32_t			firm_sdram_size;
+	uint32_t			firm_hd_channels;
+	char				firm_model[12];
+	char				firm_version[20];
+};/* HW_DEVICE_EXTENSION */
+/*
+*******************************************************************************
+**                   Command Control Block
+**             this CCB length must be 32 bytes boundary
+*******************************************************************************
+*/
+struct CommandControlBlock
+{
+	struct ARCMSR_CDB		arcmsr_cdb;
+	/*
+	** 0-503 (size of CDB=504):
+	** arcmsr messenger scsi command descriptor size 504 bytes
+	*/
+	uint32_t			cdb_shifted_phyaddr;
+	/* 504-507 */
+	uint32_t			reserved1;
+	/* 508-511 */
+#if BITS_PER_LONG == 64
+	/*  ======================512+64 bytes========================  */
+	struct list_head		list;
+	/* 512-527 16 bytes next/prev ptrs for ccb lists */
+	struct scsi_cmnd *		pcmd;
+	/* 528-535 8 bytes pointer of linux scsi command */
+	struct AdapterControlBlock *	acb;
+	/* 536-543 8 bytes pointer of acb */
+
+	uint16_t			ccb_flags;
+	/* 544-545 */
+	#define		CCB_FLAG_READ			0x0000
+	#define		CCB_FLAG_WRITE			0x0001
+	#define		CCB_FLAG_ERROR			0x0002
+	#define		CCB_FLAG_FLUSHCACHE		0x0004
+	#define		CCB_FLAG_MASTER_ABORTED		0x0008
+	uint16_t			startdone;
+	/* 546-547 */
+	#define		ARCMSR_CCB_DONE			0x0000
+	#define		ARCMSR_CCB_START		0x55AA
+	#define		ARCMSR_CCB_ABORTED		0xAA55
+	#define		ARCMSR_CCB_ILLEGAL		0xFFFF
+	uint32_t			reserved2[7];
+	/* 548-551 552-555 556-559 560-563 564-567 568-571 572-575 */
+#else
+	/*  ======================512+32 bytes========================  */
+	struct list_head		list;
+	/* 512-519 8 bytes next/prev ptrs for ccb lists */
+	struct scsi_cmnd *		pcmd;
+	/* 520-523 4 bytes pointer of linux scsi command */
+	struct AdapterControlBlock *	acb;
+	/* 524-527 4 bytes pointer of acb */
+
+	uint16_t			ccb_flags;
+	/* 528-529 */
+	#define		CCB_FLAG_READ			0x0000
+	#define		CCB_FLAG_WRITE			0x0001
+	#define		CCB_FLAG_ERROR			0x0002
+	#define		CCB_FLAG_FLUSHCACHE		0x0004
+	#define		CCB_FLAG_MASTER_ABORTED		0x0008
+	uint16_t			startdone;
+	/* 530-531 */
+	#define		ARCMSR_CCB_DONE			0x0000
+	#define		ARCMSR_CCB_START		0x55AA
+	#define		ARCMSR_CCB_ABORTED		0xAA55
+	#define		ARCMSR_CCB_ILLEGAL		0xFFFF
+	uint32_t			reserved2[3];
+	/* 532-535 536-539 540-543 */
+#endif
+	/*  ==========================================================  */
+};
+/*
+*******************************************************************************
+**    ARECA SCSI sense data
+*******************************************************************************
+*/
+struct SENSE_DATA
+{
+	uint8_t				ErrorCode:7;
+#define SCSI_SENSE_CURRENT_ERRORS	0x70
+#define SCSI_SENSE_DEFERRED_ERRORS	0x71
+	uint8_t				Valid:1;
+	uint8_t				SegmentNumber;
+	uint8_t				SenseKey:4;
+	uint8_t				Reserved:1;
+	uint8_t				IncorrectLength:1;
+	uint8_t				EndOfMedia:1;
+	uint8_t				FileMark:1;
+	uint8_t				Information[4];
+	uint8_t				AdditionalSenseLength;
+	uint8_t				CommandSpecificInformation[4];
+	uint8_t				AdditionalSenseCode;
+	uint8_t				AdditionalSenseCodeQualifier;
+	uint8_t				FieldReplaceableUnitCode;
+	uint8_t				SenseKeySpecific[3];
+};
+/*
+*******************************************************************************
+**  Outbound Interrupt Status Register - OISR
+*******************************************************************************
+*/
+#define     ARCMSR_MU_OUTBOUND_INTERRUPT_STATUS_REG                 0x30
+#define     ARCMSR_MU_OUTBOUND_PCI_INT                              0x10
+#define     ARCMSR_MU_OUTBOUND_POSTQUEUE_INT                        0x08
+#define     ARCMSR_MU_OUTBOUND_DOORBELL_INT                         0x04
+#define     ARCMSR_MU_OUTBOUND_MESSAGE1_INT                         0x02
+#define     ARCMSR_MU_OUTBOUND_MESSAGE0_INT                         0x01
+#define     ARCMSR_MU_OUTBOUND_HANDLE_INT                 \
+                    (ARCMSR_MU_OUTBOUND_MESSAGE0_INT      \
+                     |ARCMSR_MU_OUTBOUND_MESSAGE1_INT     \
+                     |ARCMSR_MU_OUTBOUND_DOORBELL_INT     \
+                     |ARCMSR_MU_OUTBOUND_POSTQUEUE_INT    \
+                     |ARCMSR_MU_OUTBOUND_PCI_INT)
+/*
+*******************************************************************************
+**  Outbound Interrupt Mask Register - OIMR
+*******************************************************************************
+*/
+#define     ARCMSR_MU_OUTBOUND_INTERRUPT_MASK_REG                   0x34
+#define     ARCMSR_MU_OUTBOUND_PCI_INTMASKENABLE                    0x10
+#define     ARCMSR_MU_OUTBOUND_POSTQUEUE_INTMASKENABLE              0x08
+#define     ARCMSR_MU_OUTBOUND_DOORBELL_INTMASKENABLE               0x04
+#define     ARCMSR_MU_OUTBOUND_MESSAGE1_INTMASKENABLE               0x02
+#define     ARCMSR_MU_OUTBOUND_MESSAGE0_INTMASKENABLE               0x01
+#define     ARCMSR_MU_OUTBOUND_ALL_INTMASKENABLE                    0x1F
+
+extern void arcmsr_post_Qbuffer(struct AdapterControlBlock *acb);
+extern struct class_device_attribute *arcmsr_host_attrs[];
+extern int arcmsr_alloc_sysfs_attr(struct AdapterControlBlock *acb);
+void arcmsr_free_sysfs_attr(struct AdapterControlBlock *acb);
+
diff --git a/drivers/scsi/arcmsr/arcmsr_attr.c b/drivers/scsi/arcmsr/arcmsr_attr.c
new file mode 100644
index 000000000000..0459f4194d7c
--- /dev/null
+++ b/drivers/scsi/arcmsr/arcmsr_attr.c
@@ -0,0 +1,392 @@
+/*
+*******************************************************************************
+**        O.S   : Linux
+**   FILE NAME  : arcmsr_attr.c
+**        BY    : Erich Chen
+**   Description: attributes exported to sysfs and device host
+*******************************************************************************
+** Copyright (C) 2002 - 2005, Areca Technology Corporation All rights reserved
+**
+**     Web site: www.areca.com.tw
+**       E-mail: erich@areca.com.tw
+**
+** This program is free software; you can redistribute it and/or modify
+** it under the terms of the GNU General Public License version 2 as
+** published by the Free Software Foundation.
+** This program is distributed in the hope that it will be useful,
+** but WITHOUT ANY WARRANTY; without even the implied warranty of
+** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+** GNU General Public License for more details.
+*******************************************************************************
+** Redistribution and use in source and binary forms, with or without
+** modification, are permitted provided that the following conditions
+** are met:
+** 1. Redistributions of source code must retain the above copyright
+**    notice, this list of conditions and the following disclaimer.
+** 2. Redistributions in binary form must reproduce the above copyright
+**    notice, this list of conditions and the following disclaimer in the
+**    documentation and/or other materials provided with the distribution.
+** 3. The name of the author may not be used to endorse or promote products
+**    derived from this software without specific prior written permission.
+**
+** THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
+** IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
+** OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+** IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+** INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES(INCLUDING,BUT
+** NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+** DATA, OR PROFITS; OR BUSINESS INTERRUPTION)HOWEVER CAUSED AND ON ANY
+** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+** (INCLUDING NEGLIGENCE OR OTHERWISE)ARISING IN ANY WAY OUT OF THE USE OF
+** THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+*******************************************************************************
+** For history of changes, see Documentation/scsi/ChangeLog.arcmsr
+**     Firmware Specification, see Documentation/scsi/arcmsr_spec.txt
+*******************************************************************************
+*/
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_transport.h>
+#include "arcmsr.h"
+
+struct class_device_attribute *arcmsr_host_attrs[];
+
+static ssize_t
+arcmsr_sysfs_iop_message_read(struct kobject *kobj, char *buf, loff_t off,
+    size_t count)
+{
+	struct class_device *cdev = container_of(kobj,struct class_device,kobj);
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	struct MessageUnit __iomem *reg = acb->pmu;
+	uint8_t *pQbuffer,*ptmpQbuffer;
+	int32_t allxfer_len = 0;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	/* do message unit read. */
+	ptmpQbuffer = (uint8_t *)buf;
+	while ((acb->rqbuf_firstindex != acb->rqbuf_lastindex)
+		&& (allxfer_len < 1031)) {
+		pQbuffer = &acb->rqbuffer[acb->rqbuf_firstindex];
+		memcpy(ptmpQbuffer, pQbuffer, 1);
+		acb->rqbuf_firstindex++;
+		acb->rqbuf_firstindex %= ARCMSR_MAX_QBUFFER;
+		ptmpQbuffer++;
+		allxfer_len++;
+	}
+	if (acb->acb_flags & ACB_F_IOPDATA_OVERFLOW) {
+		struct QBUFFER __iomem * prbuffer = (struct QBUFFER __iomem *)
+					&reg->message_rbuffer;
+		uint8_t __iomem * iop_data = (uint8_t __iomem *)prbuffer->data;
+		int32_t iop_len;
+
+		acb->acb_flags &= ~ACB_F_IOPDATA_OVERFLOW;
+		iop_len = readl(&prbuffer->data_len);
+		while (iop_len > 0) {
+			acb->rqbuffer[acb->rqbuf_lastindex] = readb(iop_data);
+			acb->rqbuf_lastindex++;
+			acb->rqbuf_lastindex %= ARCMSR_MAX_QBUFFER;
+			iop_data++;
+			iop_len--;
+		}
+		writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK,
+				&reg->inbound_doorbell);
+	}
+	return (allxfer_len);
+}
+
+static ssize_t
+arcmsr_sysfs_iop_message_write(struct kobject *kobj, char *buf, loff_t off,
+    size_t count)
+{
+	struct class_device *cdev = container_of(kobj,struct class_device,kobj);
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	int32_t my_empty_len, user_len, wqbuf_firstindex, wqbuf_lastindex;
+	uint8_t *pQbuffer, *ptmpuserbuffer;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (count > 1032)
+		return -EINVAL;
+	/* do message unit write. */
+	ptmpuserbuffer = (uint8_t *)buf;
+	user_len = (int32_t)count;
+	wqbuf_lastindex = acb->wqbuf_lastindex;
+	wqbuf_firstindex = acb->wqbuf_firstindex;
+	if (wqbuf_lastindex != wqbuf_firstindex) {
+		arcmsr_post_Qbuffer(acb);
+		return 0;	/*need retry*/
+	} else {
+		my_empty_len = (wqbuf_firstindex-wqbuf_lastindex - 1)
+				&(ARCMSR_MAX_QBUFFER - 1);
+		if (my_empty_len >= user_len) {
+			while (user_len > 0) {
+				pQbuffer =
+				&acb->wqbuffer[acb->wqbuf_lastindex];
+				memcpy(pQbuffer, ptmpuserbuffer, 1);
+				acb->wqbuf_lastindex++;
+				acb->wqbuf_lastindex %= ARCMSR_MAX_QBUFFER;
+				ptmpuserbuffer++;
+				user_len--;
+			}
+			if (acb->acb_flags & ACB_F_MESSAGE_WQBUFFER_CLEARED) {
+				acb->acb_flags &=
+					~ACB_F_MESSAGE_WQBUFFER_CLEARED;
+				arcmsr_post_Qbuffer(acb);
+			}
+			return count;
+		} else {
+			return 0;	/*need retry*/
+		}
+	}
+}
+
+static ssize_t
+arcmsr_sysfs_iop_message_clear(struct kobject *kobj, char *buf, loff_t off,
+    size_t count)
+{
+	struct class_device *cdev = container_of(kobj,struct class_device,kobj);
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	struct MessageUnit __iomem *reg = acb->pmu;
+	uint8_t *pQbuffer;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	if (acb->acb_flags & ACB_F_IOPDATA_OVERFLOW) {
+		acb->acb_flags &= ~ACB_F_IOPDATA_OVERFLOW;
+		writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK
+				, &reg->inbound_doorbell);
+	}
+	acb->acb_flags |=
+		(ACB_F_MESSAGE_WQBUFFER_CLEARED
+		| ACB_F_MESSAGE_RQBUFFER_CLEARED
+		| ACB_F_MESSAGE_WQBUFFER_READED);
+	acb->rqbuf_firstindex = 0;
+	acb->rqbuf_lastindex = 0;
+	acb->wqbuf_firstindex = 0;
+	acb->wqbuf_lastindex = 0;
+	pQbuffer = acb->rqbuffer;
+	memset(pQbuffer, 0, sizeof (struct QBUFFER));
+	pQbuffer = acb->wqbuffer;
+	memset(pQbuffer, 0, sizeof (struct QBUFFER));
+	return 1;
+}
+
+static struct bin_attribute arcmsr_sysfs_message_read_attr = {
+	.attr = {
+		.name = "mu_read",
+		.mode = S_IRUSR ,
+		.owner = THIS_MODULE,
+	},
+	.size = 1032,
+	.read = arcmsr_sysfs_iop_message_read,
+};
+
+static struct bin_attribute arcmsr_sysfs_message_write_attr = {
+	.attr = {
+		.name = "mu_write",
+		.mode = S_IWUSR,
+		.owner = THIS_MODULE,
+	},
+	.size = 1032,
+	.write = arcmsr_sysfs_iop_message_write,
+};
+
+static struct bin_attribute arcmsr_sysfs_message_clear_attr = {
+	.attr = {
+		.name = "mu_clear",
+		.mode = S_IWUSR,
+		.owner = THIS_MODULE,
+	},
+	.size = 1,
+	.write = arcmsr_sysfs_iop_message_clear,
+};
+
+int arcmsr_alloc_sysfs_attr(struct AdapterControlBlock *acb)
+{
+	struct Scsi_Host *host = acb->host;
+	int error;
+
+	error = sysfs_create_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_read_attr);
+	if (error) {
+		printk(KERN_ERR "arcmsr: alloc sysfs mu_read failed\n");
+		goto error_bin_file_message_read;
+	}
+	error = sysfs_create_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_write_attr);
+	if (error) {
+		printk(KERN_ERR "arcmsr: alloc sysfs mu_write failed\n");
+		goto error_bin_file_message_write;
+	}
+	error = sysfs_create_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_clear_attr);
+	if (error) {
+		printk(KERN_ERR "arcmsr: alloc sysfs mu_clear failed\n");
+		goto error_bin_file_message_clear;
+	}
+	return 0;
+error_bin_file_message_clear:
+	error = sysfs_remove_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_write_attr);
+	if (error)
+		printk(KERN_ERR "arcmsr: sysfs_remove_bin_file mu_write failed\n");
+error_bin_file_message_write:
+	error = sysfs_remove_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_read_attr);
+	if (error)
+		printk(KERN_ERR "arcmsr: sysfs_remove_bin_file mu_read failed\n");
+error_bin_file_message_read:
+	return error;
+}
+
+void
+arcmsr_free_sysfs_attr(struct AdapterControlBlock *acb) {
+	struct Scsi_Host *host = acb->host;
+	int error;
+
+	error = sysfs_remove_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_clear_attr);
+	if (error)
+		printk(KERN_ERR "arcmsr: free sysfs mu_clear failed\n");
+	error = sysfs_remove_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_write_attr);
+	if (error)
+		printk(KERN_ERR "arcmsr: free sysfs mu_write failed\n");
+	error = sysfs_remove_bin_file(&host->shost_classdev.kobj,
+				&arcmsr_sysfs_message_read_attr);
+	if (error)
+		printk(KERN_ERR "arcmsr: free sysfss mu_read failed\n");
+}
+
+
+static ssize_t
+arcmsr_attr_host_driver_version(struct class_device *cdev, char *buf) {
+	return snprintf(buf, PAGE_SIZE,
+			"ARCMSR: %s\n",
+			ARCMSR_DRIVER_VERSION);
+}
+
+static ssize_t
+arcmsr_attr_host_driver_posted_cmd(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	return snprintf(buf, PAGE_SIZE,
+			"Current commands posted:     %4d\n",
+			atomic_read(&acb->ccboutstandingcount));
+}
+
+static ssize_t
+arcmsr_attr_host_driver_reset(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	return snprintf(buf, PAGE_SIZE,
+			"SCSI Host Resets:            %4d\n",
+			acb->num_resets);
+}
+
+static ssize_t
+arcmsr_attr_host_driver_abort(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	return snprintf(buf, PAGE_SIZE,
+			"SCSI Aborts/Timeouts:        %4d\n",
+			acb->num_aborts);
+}
+
+static ssize_t
+arcmsr_attr_host_fw_model(struct class_device *cdev, char *buf) {
+    struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+	return snprintf(buf, PAGE_SIZE,
+			"Adapter Model: %s\n",
+			acb->firm_model);
+}
+
+static ssize_t
+arcmsr_attr_host_fw_version(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+
+	return snprintf(buf, PAGE_SIZE,
+			"Firmware Version:  %s\n",
+			acb->firm_version);
+}
+
+static ssize_t
+arcmsr_attr_host_fw_request_len(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+
+	return snprintf(buf, PAGE_SIZE,
+			"Reguest Lenth: %4d\n",
+			acb->firm_request_len);
+}
+
+static ssize_t
+arcmsr_attr_host_fw_numbers_queue(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+
+	return snprintf(buf, PAGE_SIZE,
+			"Numbers of Queue: %4d\n",
+			acb->firm_numbers_queue);
+}
+
+static ssize_t
+arcmsr_attr_host_fw_sdram_size(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+
+	return snprintf(buf, PAGE_SIZE,
+			"SDRAM Size: %4d\n",
+			acb->firm_sdram_size);
+}
+
+static ssize_t
+arcmsr_attr_host_fw_hd_channels(struct class_device *cdev, char *buf) {
+	struct Scsi_Host *host = class_to_shost(cdev);
+	struct AdapterControlBlock *acb = (struct AdapterControlBlock *) host->hostdata;
+
+	return snprintf(buf, PAGE_SIZE,
+			"Hard Disk Channels: %4d\n",
+			acb->firm_hd_channels);
+}
+
+static CLASS_DEVICE_ATTR(host_driver_version, S_IRUGO, arcmsr_attr_host_driver_version, NULL);
+static CLASS_DEVICE_ATTR(host_driver_posted_cmd, S_IRUGO, arcmsr_attr_host_driver_posted_cmd, NULL);
+static CLASS_DEVICE_ATTR(host_driver_reset, S_IRUGO, arcmsr_attr_host_driver_reset, NULL);
+static CLASS_DEVICE_ATTR(host_driver_abort, S_IRUGO, arcmsr_attr_host_driver_abort, NULL);
+static CLASS_DEVICE_ATTR(host_fw_model, S_IRUGO, arcmsr_attr_host_fw_model, NULL);
+static CLASS_DEVICE_ATTR(host_fw_version, S_IRUGO, arcmsr_attr_host_fw_version, NULL);
+static CLASS_DEVICE_ATTR(host_fw_request_len, S_IRUGO, arcmsr_attr_host_fw_request_len, NULL);
+static CLASS_DEVICE_ATTR(host_fw_numbers_queue, S_IRUGO, arcmsr_attr_host_fw_numbers_queue, NULL);
+static CLASS_DEVICE_ATTR(host_fw_sdram_size, S_IRUGO, arcmsr_attr_host_fw_sdram_size, NULL);
+static CLASS_DEVICE_ATTR(host_fw_hd_channels, S_IRUGO, arcmsr_attr_host_fw_hd_channels, NULL);
+
+struct class_device_attribute *arcmsr_host_attrs[] = {
+	&class_device_attr_host_driver_version,
+	&class_device_attr_host_driver_posted_cmd,
+	&class_device_attr_host_driver_reset,
+	&class_device_attr_host_driver_abort,
+	&class_device_attr_host_fw_model,
+	&class_device_attr_host_fw_version,
+	&class_device_attr_host_fw_request_len,
+	&class_device_attr_host_fw_numbers_queue,
+	&class_device_attr_host_fw_sdram_size,
+	&class_device_attr_host_fw_hd_channels,
+	NULL,
+};
diff --git a/drivers/scsi/arcmsr/arcmsr_hba.c b/drivers/scsi/arcmsr/arcmsr_hba.c
new file mode 100644
index 000000000000..475f978ff8f0
--- /dev/null
+++ b/drivers/scsi/arcmsr/arcmsr_hba.c
@@ -0,0 +1,1496 @@
+/*
+*******************************************************************************
+**        O.S   : Linux
+**   FILE NAME  : arcmsr_hba.c
+**        BY    : Erich Chen
+**   Description: SCSI RAID Device Driver for
+**                ARECA RAID Host adapter
+*******************************************************************************
+** Copyright (C) 2002 - 2005, Areca Technology Corporation All rights reserved
+**
+**     Web site: www.areca.com.tw
+**       E-mail: erich@areca.com.tw
+**
+** This program is free software; you can redistribute it and/or modify
+** it under the terms of the GNU General Public License version 2 as
+** published by the Free Software Foundation.
+** This program is distributed in the hope that it will be useful,
+** but WITHOUT ANY WARRANTY; without even the implied warranty of
+** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+** GNU General Public License for more details.
+*******************************************************************************
+** Redistribution and use in source and binary forms, with or without
+** modification, are permitted provided that the following conditions
+** are met:
+** 1. Redistributions of source code must retain the above copyright
+**    notice, this list of conditions and the following disclaimer.
+** 2. Redistributions in binary form must reproduce the above copyright
+**    notice, this list of conditions and the following disclaimer in the
+**    documentation and/or other materials provided with the distribution.
+** 3. The name of the author may not be used to endorse or promote products
+**    derived from this software without specific prior written permission.
+**
+** THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
+** IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
+** OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+** IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+** INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES(INCLUDING,BUT
+** NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+** DATA, OR PROFITS; OR BUSINESS INTERRUPTION)HOWEVER CAUSED AND ON ANY
+** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+** (INCLUDING NEGLIGENCE OR OTHERWISE)ARISING IN ANY WAY OUT OF THE USE OF
+** THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+*******************************************************************************
+** For history of changes, see Documentation/scsi/ChangeLog.arcmsr
+**     Firmware Specification, see Documentation/scsi/arcmsr_spec.txt
+*******************************************************************************
+*/
+#include <linux/module.h>
+#include <linux/reboot.h>
+#include <linux/spinlock.h>
+#include <linux/pci_ids.h>
+#include <linux/interrupt.h>
+#include <linux/moduleparam.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/timer.h>
+#include <linux/pci.h>
+#include <asm/dma.h>
+#include <asm/io.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_tcq.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_transport.h>
+#include <scsi/scsicam.h>
+#include "arcmsr.h"
+
+MODULE_AUTHOR("Erich Chen <erich@areca.com.tw>");
+MODULE_DESCRIPTION("ARECA (ARC11xx/12xx) SATA RAID HOST Adapter");
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_VERSION(ARCMSR_DRIVER_VERSION);
+
+static int arcmsr_iop_message_xfer(struct AdapterControlBlock *acb, struct scsi_cmnd *cmd);
+static int arcmsr_abort(struct scsi_cmnd *);
+static int arcmsr_bus_reset(struct scsi_cmnd *);
+static int arcmsr_bios_param(struct scsi_device *sdev,
+				struct block_device *bdev, sector_t capacity, int *info);
+static int arcmsr_queue_command(struct scsi_cmnd * cmd,
+				void (*done) (struct scsi_cmnd *));
+static int arcmsr_probe(struct pci_dev *pdev,
+				const struct pci_device_id *id);
+static void arcmsr_remove(struct pci_dev *pdev);
+static void arcmsr_shutdown(struct pci_dev *pdev);
+static void arcmsr_iop_init(struct AdapterControlBlock *acb);
+static void arcmsr_free_ccb_pool(struct AdapterControlBlock *acb);
+static void arcmsr_stop_adapter_bgrb(struct AdapterControlBlock *acb);
+static void arcmsr_flush_adapter_cache(struct AdapterControlBlock *acb);
+static uint8_t arcmsr_wait_msgint_ready(struct AdapterControlBlock *acb);
+static const char *arcmsr_info(struct Scsi_Host *);
+static irqreturn_t arcmsr_interrupt(struct AdapterControlBlock *acb);
+
+static int arcmsr_adjust_disk_queue_depth(struct scsi_device *sdev, int queue_depth)
+{
+	if (queue_depth > ARCMSR_MAX_CMD_PERLUN)
+		queue_depth = ARCMSR_MAX_CMD_PERLUN;
+	scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, queue_depth);
+	return queue_depth;
+}
+
+static struct scsi_host_template arcmsr_scsi_host_template = {
+	.module			= THIS_MODULE,
+	.name			= "ARCMSR ARECA SATA RAID HOST Adapter" ARCMSR_DRIVER_VERSION,
+	.info			= arcmsr_info,
+	.queuecommand		= arcmsr_queue_command,
+	.eh_abort_handler	= arcmsr_abort,
+	.eh_bus_reset_handler	= arcmsr_bus_reset,
+	.bios_param		= arcmsr_bios_param,
+	.change_queue_depth	= arcmsr_adjust_disk_queue_depth,
+	.can_queue		= ARCMSR_MAX_OUTSTANDING_CMD,
+	.this_id		= ARCMSR_SCSI_INITIATOR_ID,
+	.sg_tablesize		= ARCMSR_MAX_SG_ENTRIES,
+	.max_sectors    	= ARCMSR_MAX_XFER_SECTORS,
+	.cmd_per_lun		= ARCMSR_MAX_CMD_PERLUN,
+	.use_clustering		= ENABLE_CLUSTERING,
+	.shost_attrs		= arcmsr_host_attrs,
+};
+
+static struct pci_device_id arcmsr_device_id_table[] = {
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1110)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1120)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1130)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1160)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1170)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1210)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1220)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1230)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1260)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1270)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1280)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1380)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1381)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1680)},
+	{PCI_DEVICE(PCI_VENDOR_ID_ARECA, PCI_DEVICE_ID_ARECA_1681)},
+	{0, 0}, /* Terminating entry */
+};
+MODULE_DEVICE_TABLE(pci, arcmsr_device_id_table);
+static struct pci_driver arcmsr_pci_driver = {
+	.name			= "arcmsr",
+	.id_table		= arcmsr_device_id_table,
+	.probe			= arcmsr_probe,
+	.remove			= arcmsr_remove,
+	.shutdown		= arcmsr_shutdown
+};
+
+static irqreturn_t arcmsr_do_interrupt(int irq, void *dev_id,
+	struct pt_regs *regs)
+{
+	irqreturn_t handle_state;
+	struct AdapterControlBlock *acb;
+	unsigned long flags;
+
+	acb = (struct AdapterControlBlock *)dev_id;
+
+	spin_lock_irqsave(acb->host->host_lock, flags);
+	handle_state = arcmsr_interrupt(acb);
+	spin_unlock_irqrestore(acb->host->host_lock, flags);
+	return handle_state;
+}
+
+static int arcmsr_bios_param(struct scsi_device *sdev,
+		struct block_device *bdev, sector_t capacity, int *geom)
+{
+	int ret, heads, sectors, cylinders, total_capacity;
+	unsigned char *buffer;/* return copy of block device's partition table */
+
+	buffer = scsi_bios_ptable(bdev);
+	if (buffer) {
+		ret = scsi_partsize(buffer, capacity, &geom[2], &geom[0], &geom[1]);
+		kfree(buffer);
+		if (ret != -1)
+			return ret;
+	}
+	total_capacity = capacity;
+	heads = 64;
+	sectors = 32;
+	cylinders = total_capacity / (heads * sectors);
+	if (cylinders > 1024) {
+		heads = 255;
+		sectors = 63;
+		cylinders = total_capacity / (heads * sectors);
+	}
+	geom[0] = heads;
+	geom[1] = sectors;
+	geom[2] = cylinders;
+	return 0;
+}
+
+static int arcmsr_alloc_ccb_pool(struct AdapterControlBlock *acb)
+{
+	struct pci_dev *pdev = acb->pdev;
+	struct MessageUnit __iomem *reg = acb->pmu;
+	u32 ccb_phyaddr_hi32;
+	void *dma_coherent;
+	dma_addr_t dma_coherent_handle, dma_addr;
+	struct CommandControlBlock *ccb_tmp;
+	int i, j;
+
+	dma_coherent = dma_alloc_coherent(&pdev->dev,
+			ARCMSR_MAX_FREECCB_NUM *
+			sizeof (struct CommandControlBlock) + 0x20,
+			&dma_coherent_handle, GFP_KERNEL);
+	if (!dma_coherent)
+		return -ENOMEM;
+
+	acb->dma_coherent = dma_coherent;
+	acb->dma_coherent_handle = dma_coherent_handle;
+
+	if (((unsigned long)dma_coherent & 0x1F)) {
+		dma_coherent = dma_coherent +
+			(0x20 - ((unsigned long)dma_coherent & 0x1F));
+		dma_coherent_handle = dma_coherent_handle +
+			(0x20 - ((unsigned long)dma_coherent_handle & 0x1F));
+	}
+
+	dma_addr = dma_coherent_handle;
+	ccb_tmp = (struct CommandControlBlock *)dma_coherent;
+	for (i = 0; i < ARCMSR_MAX_FREECCB_NUM; i++) {
+		ccb_tmp->cdb_shifted_phyaddr = dma_addr >> 5;
+		ccb_tmp->acb = acb;
+		acb->pccb_pool[i] = ccb_tmp;
+		list_add_tail(&ccb_tmp->list, &acb->ccb_free_list);
+		dma_addr = dma_addr + sizeof (struct CommandControlBlock);
+		ccb_tmp++;
+	}
+
+	acb->vir2phy_offset = (unsigned long)ccb_tmp -
+			      (unsigned long)dma_addr;
+	for (i = 0; i < ARCMSR_MAX_TARGETID; i++)
+		for (j = 0; j < ARCMSR_MAX_TARGETLUN; j++)
+			acb->devstate[i][j] = ARECA_RAID_GOOD;
+
+	/*
+	** here we need to tell iop 331 our ccb_tmp.HighPart
+	** if ccb_tmp.HighPart is not zero
+	*/
+	ccb_phyaddr_hi32 = (uint32_t) ((dma_coherent_handle >> 16) >> 16);
+	if (ccb_phyaddr_hi32 != 0) {
+		writel(ARCMSR_SIGNATURE_SET_CONFIG, &reg->message_rwbuffer[0]);
+		writel(ccb_phyaddr_hi32, &reg->message_rwbuffer[1]);
+		writel(ARCMSR_INBOUND_MESG0_SET_CONFIG, &reg->inbound_msgaddr0);
+		if (arcmsr_wait_msgint_ready(acb))
+			printk(KERN_NOTICE "arcmsr%d: "
+			       "'set ccb high part physical address' timeout\n",
+				acb->host->host_no);
+	}
+
+	writel(readl(&reg->outbound_intmask) |
+			ARCMSR_MU_OUTBOUND_ALL_INTMASKENABLE,
+	       &reg->outbound_intmask);
+	return 0;
+}
+
+static int arcmsr_probe(struct pci_dev *pdev,
+	const struct pci_device_id *id)
+{
+	struct Scsi_Host *host;
+	struct AdapterControlBlock *acb;
+	uint8_t bus, dev_fun;
+	int error;
+
+	error = pci_enable_device(pdev);
+	if (error)
+		goto out;
+	pci_set_master(pdev);
+
+	host = scsi_host_alloc(&arcmsr_scsi_host_template,
+			sizeof(struct AdapterControlBlock));
+	if (!host) {
+		error = -ENOMEM;
+		goto out_disable_device;
+	}
+	acb = (struct AdapterControlBlock *)host->hostdata;
+	memset(acb, 0, sizeof (struct AdapterControlBlock));
+
+	error = pci_set_dma_mask(pdev, DMA_64BIT_MASK);
+	if (error) {
+		error = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
+		if (error) {
+			printk(KERN_WARNING
+			       "scsi%d: No suitable DMA mask available\n",
+			       host->host_no);
+			goto out_host_put;
+		}
+	}
+	bus = pdev->bus->number;
+	dev_fun = pdev->devfn;
+	acb->host = host;
+	acb->pdev = pdev;
+	host->max_sectors = ARCMSR_MAX_XFER_SECTORS;
+	host->max_lun = ARCMSR_MAX_TARGETLUN;
+	host->max_id = ARCMSR_MAX_TARGETID;/*16:8*/
+	host->max_cmd_len = 16;    /*this is issue of 64bit LBA, over 2T byte*/
+	host->sg_tablesize = ARCMSR_MAX_SG_ENTRIES;
+	host->can_queue = ARCMSR_MAX_FREECCB_NUM; /* max simultaneous cmds */
+	host->cmd_per_lun = ARCMSR_MAX_CMD_PERLUN;
+	host->this_id = ARCMSR_SCSI_INITIATOR_ID;
+	host->unique_id = (bus << 8) | dev_fun;
+	host->irq = pdev->irq;
+	error = pci_request_regions(pdev, "arcmsr");
+	if (error)
+		goto out_host_put;
+
+	acb->pmu = ioremap(pci_resource_start(pdev, 0),
+			   pci_resource_len(pdev, 0));
+	if (!acb->pmu) {
+		printk(KERN_NOTICE "arcmsr%d: memory"
+			" mapping region fail \n", acb->host->host_no);
+		goto out_release_regions;
+	}
+	acb->acb_flags |= (ACB_F_MESSAGE_WQBUFFER_CLEARED |
+			   ACB_F_MESSAGE_RQBUFFER_CLEARED |
+			   ACB_F_MESSAGE_WQBUFFER_READED);
+	acb->acb_flags &= ~ACB_F_SCSISTOPADAPTER;
+	INIT_LIST_HEAD(&acb->ccb_free_list);
+
+	error = arcmsr_alloc_ccb_pool(acb);
+	if (error)
+		goto out_iounmap;
+
+	error = request_irq(pdev->irq, arcmsr_do_interrupt,
+			SA_INTERRUPT | SA_SHIRQ, "arcmsr", acb);
+	if (error)
+		goto out_free_ccb_pool;
+
+	arcmsr_iop_init(acb);
+	pci_set_drvdata(pdev, host);
+
+	error = scsi_add_host(host, &pdev->dev);
+	if (error)
+		goto out_free_irq;
+
+	error = arcmsr_alloc_sysfs_attr(acb);
+	if (error)
+		goto out_free_sysfs;
+
+	scsi_scan_host(host);
+	return 0;
+ out_free_sysfs:
+ out_free_irq:
+	free_irq(pdev->irq, acb);
+ out_free_ccb_pool:
+	arcmsr_free_ccb_pool(acb);
+ out_iounmap:
+	iounmap(acb->pmu);
+ out_release_regions:
+	pci_release_regions(pdev);
+ out_host_put:
+	scsi_host_put(host);
+ out_disable_device:
+	pci_disable_device(pdev);
+ out:
+	return error;
+}
+
+static void arcmsr_abort_allcmd(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+
+	writel(ARCMSR_INBOUND_MESG0_ABORT_CMD, &reg->inbound_msgaddr0);
+	if (arcmsr_wait_msgint_ready(acb))
+		printk(KERN_NOTICE
+			"arcmsr%d: wait 'abort all outstanding command' timeout \n"
+			, acb->host->host_no);
+}
+
+static void arcmsr_pci_unmap_dma(struct CommandControlBlock *ccb)
+{
+	struct AdapterControlBlock *acb = ccb->acb;
+	struct scsi_cmnd *pcmd = ccb->pcmd;
+
+	if (pcmd->use_sg != 0) {
+		struct scatterlist *sl;
+
+		sl = (struct scatterlist *)pcmd->request_buffer;
+		pci_unmap_sg(acb->pdev, sl, pcmd->use_sg, pcmd->sc_data_direction);
+	}
+	else if (pcmd->request_bufflen != 0)
+		pci_unmap_single(acb->pdev,
+			pcmd->SCp.dma_handle,
+			pcmd->request_bufflen, pcmd->sc_data_direction);
+}
+
+static void arcmsr_ccb_complete(struct CommandControlBlock *ccb, int stand_flag)
+{
+	struct AdapterControlBlock *acb = ccb->acb;
+	struct scsi_cmnd *pcmd = ccb->pcmd;
+
+	arcmsr_pci_unmap_dma(ccb);
+	if (stand_flag == 1)
+		atomic_dec(&acb->ccboutstandingcount);
+	ccb->startdone = ARCMSR_CCB_DONE;
+	ccb->ccb_flags = 0;
+	list_add_tail(&ccb->list, &acb->ccb_free_list);
+	pcmd->scsi_done(pcmd);
+}
+
+static void arcmsr_remove(struct pci_dev *pdev)
+{
+	struct Scsi_Host *host = pci_get_drvdata(pdev);
+	struct AdapterControlBlock *acb =
+		(struct AdapterControlBlock *) host->hostdata;
+	struct MessageUnit __iomem *reg = acb->pmu;
+	int poll_count = 0;
+
+	arcmsr_free_sysfs_attr(acb);
+	scsi_remove_host(host);
+	arcmsr_stop_adapter_bgrb(acb);
+	arcmsr_flush_adapter_cache(acb);
+	writel(readl(&reg->outbound_intmask) |
+		ARCMSR_MU_OUTBOUND_ALL_INTMASKENABLE,
+		&reg->outbound_intmask);
+	acb->acb_flags |= ACB_F_SCSISTOPADAPTER;
+	acb->acb_flags &= ~ACB_F_IOP_INITED;
+
+	for (poll_count = 0; poll_count < 256; poll_count++) {
+		if (!atomic_read(&acb->ccboutstandingcount))
+			break;
+		arcmsr_interrupt(acb);
+		msleep(25);
+	}
+
+	if (atomic_read(&acb->ccboutstandingcount)) {
+		int i;
+
+		arcmsr_abort_allcmd(acb);
+		for (i = 0; i < ARCMSR_MAX_OUTSTANDING_CMD; i++)
+			readl(&reg->outbound_queueport);
+		for (i = 0; i < ARCMSR_MAX_FREECCB_NUM; i++) {
+			struct CommandControlBlock *ccb = acb->pccb_pool[i];
+			if (ccb->startdone == ARCMSR_CCB_START) {
+				ccb->startdone = ARCMSR_CCB_ABORTED;
+				ccb->pcmd->result = DID_ABORT << 16;
+				arcmsr_ccb_complete(ccb, 1);
+			}
+		}
+	}
+
+	free_irq(pdev->irq, acb);
+	iounmap(acb->pmu);
+	arcmsr_free_ccb_pool(acb);
+	pci_release_regions(pdev);
+
+	scsi_host_put(host);
+
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+}
+
+static void arcmsr_shutdown(struct pci_dev *pdev)
+{
+	struct Scsi_Host *host = pci_get_drvdata(pdev);
+	struct AdapterControlBlock *acb =
+		(struct AdapterControlBlock *)host->hostdata;
+
+	arcmsr_stop_adapter_bgrb(acb);
+	arcmsr_flush_adapter_cache(acb);
+}
+
+static int arcmsr_module_init(void)
+{
+	int error = 0;
+
+	error = pci_register_driver(&arcmsr_pci_driver);
+	return error;
+}
+
+static void arcmsr_module_exit(void)
+{
+	pci_unregister_driver(&arcmsr_pci_driver);
+}
+module_init(arcmsr_module_init);
+module_exit(arcmsr_module_exit);
+
+static u32 arcmsr_disable_outbound_ints(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	u32 orig_mask = readl(&reg->outbound_intmask);
+
+	writel(orig_mask | ARCMSR_MU_OUTBOUND_ALL_INTMASKENABLE,
+			&reg->outbound_intmask);
+	return orig_mask;
+}
+
+static void arcmsr_enable_outbound_ints(struct AdapterControlBlock *acb,
+		u32 orig_mask)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	u32 mask;
+
+	mask = orig_mask & ~(ARCMSR_MU_OUTBOUND_POSTQUEUE_INTMASKENABLE |
+			     ARCMSR_MU_OUTBOUND_DOORBELL_INTMASKENABLE);
+	writel(mask, &reg->outbound_intmask);
+}
+
+static void arcmsr_flush_adapter_cache(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg=acb->pmu;
+
+	writel(ARCMSR_INBOUND_MESG0_FLUSH_CACHE, &reg->inbound_msgaddr0);
+	if (arcmsr_wait_msgint_ready(acb))
+		printk(KERN_NOTICE
+			"arcmsr%d: wait 'flush adapter cache' timeout \n"
+			, acb->host->host_no);
+}
+
+static void arcmsr_report_sense_info(struct CommandControlBlock *ccb)
+{
+	struct scsi_cmnd *pcmd = ccb->pcmd;
+	struct SENSE_DATA *sensebuffer = (struct SENSE_DATA *)pcmd->sense_buffer;
+
+	pcmd->result = DID_OK << 16;
+	if (sensebuffer) {
+		int sense_data_length =
+			sizeof (struct SENSE_DATA) < sizeof (pcmd->sense_buffer)
+			? sizeof (struct SENSE_DATA) : sizeof (pcmd->sense_buffer);
+		memset(sensebuffer, 0, sizeof (pcmd->sense_buffer));
+		memcpy(sensebuffer, ccb->arcmsr_cdb.SenseData, sense_data_length);
+		sensebuffer->ErrorCode = SCSI_SENSE_CURRENT_ERRORS;
+		sensebuffer->Valid = 1;
+	}
+}
+
+static uint8_t arcmsr_wait_msgint_ready(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	uint32_t Index;
+	uint8_t Retries = 0x00;
+
+	do {
+		for (Index = 0; Index < 100; Index++) {
+			if (readl(&reg->outbound_intstatus)
+				& ARCMSR_MU_OUTBOUND_MESSAGE0_INT) {
+				writel(ARCMSR_MU_OUTBOUND_MESSAGE0_INT
+					, &reg->outbound_intstatus);
+				return 0x00;
+			}
+			msleep_interruptible(10);
+		}/*max 1 seconds*/
+	} while (Retries++ < 20);/*max 20 sec*/
+	return 0xff;
+}
+
+static void arcmsr_build_ccb(struct AdapterControlBlock *acb,
+	struct CommandControlBlock *ccb, struct scsi_cmnd *pcmd)
+{
+	struct ARCMSR_CDB *arcmsr_cdb = (struct ARCMSR_CDB *)&ccb->arcmsr_cdb;
+	int8_t *psge = (int8_t *)&arcmsr_cdb->u;
+	uint32_t address_lo, address_hi;
+	int arccdbsize = 0x30;
+
+	ccb->pcmd = pcmd;
+	memset(arcmsr_cdb, 0, sizeof (struct ARCMSR_CDB));
+	arcmsr_cdb->Bus = 0;
+	arcmsr_cdb->TargetID = pcmd->device->id;
+	arcmsr_cdb->LUN = pcmd->device->lun;
+	arcmsr_cdb->Function = 1;
+	arcmsr_cdb->CdbLength = (uint8_t)pcmd->cmd_len;
+	arcmsr_cdb->Context = (unsigned long)arcmsr_cdb;
+	memcpy(arcmsr_cdb->Cdb, pcmd->cmnd, pcmd->cmd_len);
+	if (pcmd->use_sg) {
+		int length, sgcount, i, cdb_sgcount = 0;
+		struct scatterlist *sl;
+
+		/* Get Scatter Gather List from scsiport. */
+		sl = (struct scatterlist *) pcmd->request_buffer;
+		sgcount = pci_map_sg(acb->pdev, sl, pcmd->use_sg,
+				pcmd->sc_data_direction);
+		/* map stor port SG list to our iop SG List. */
+		for (i = 0; i < sgcount; i++) {
+			/* Get the physical address of the current data pointer */
+			length = cpu_to_le32(sg_dma_len(sl));
+			address_lo = cpu_to_le32(dma_addr_lo32(sg_dma_address(sl)));
+			address_hi = cpu_to_le32(dma_addr_hi32(sg_dma_address(sl)));
+			if (address_hi == 0) {
+				struct SG32ENTRY *pdma_sg = (struct SG32ENTRY *)psge;
+
+				pdma_sg->address = address_lo;
+				pdma_sg->length = length;
+				psge += sizeof (struct SG32ENTRY);
+				arccdbsize += sizeof (struct SG32ENTRY);
+			} else {
+				struct SG64ENTRY *pdma_sg = (struct SG64ENTRY *)psge;
+
+				pdma_sg->addresshigh = address_hi;
+				pdma_sg->address = address_lo;
+				pdma_sg->length = length|IS_SG64_ADDR;
+				psge += sizeof (struct SG64ENTRY);
+				arccdbsize += sizeof (struct SG64ENTRY);
+			}
+			sl++;
+			cdb_sgcount++;
+		}
+		arcmsr_cdb->sgcount = (uint8_t)cdb_sgcount;
+		arcmsr_cdb->DataLength = pcmd->request_bufflen;
+		if ( arccdbsize > 256)
+			arcmsr_cdb->Flags |= ARCMSR_CDB_FLAG_SGL_BSIZE;
+	} else if (pcmd->request_bufflen) {
+		dma_addr_t dma_addr;
+		dma_addr = pci_map_single(acb->pdev, pcmd->request_buffer,
+				pcmd->request_bufflen, pcmd->sc_data_direction);
+		pcmd->SCp.dma_handle = dma_addr;
+		address_lo = cpu_to_le32(dma_addr_lo32(dma_addr));
+		address_hi = cpu_to_le32(dma_addr_hi32(dma_addr));
+		if (address_hi == 0) {
+			struct  SG32ENTRY *pdma_sg = (struct SG32ENTRY *)psge;
+			pdma_sg->address = address_lo;
+			pdma_sg->length = pcmd->request_bufflen;
+		} else {
+			struct SG64ENTRY *pdma_sg = (struct SG64ENTRY *)psge;
+			pdma_sg->addresshigh = address_hi;
+			pdma_sg->address = address_lo;
+			pdma_sg->length = pcmd->request_bufflen|IS_SG64_ADDR;
+		}
+		arcmsr_cdb->sgcount = 1;
+		arcmsr_cdb->DataLength = pcmd->request_bufflen;
+	}
+	if (pcmd->sc_data_direction == DMA_TO_DEVICE ) {
+		arcmsr_cdb->Flags |= ARCMSR_CDB_FLAG_WRITE;
+		ccb->ccb_flags |= CCB_FLAG_WRITE;
+	}
+}
+
+static void arcmsr_post_ccb(struct AdapterControlBlock *acb, struct CommandControlBlock *ccb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	uint32_t cdb_shifted_phyaddr = ccb->cdb_shifted_phyaddr;
+	struct ARCMSR_CDB *arcmsr_cdb = (struct ARCMSR_CDB *)&ccb->arcmsr_cdb;
+
+	atomic_inc(&acb->ccboutstandingcount);
+	ccb->startdone = ARCMSR_CCB_START;
+	if (arcmsr_cdb->Flags & ARCMSR_CDB_FLAG_SGL_BSIZE)
+		writel(cdb_shifted_phyaddr | ARCMSR_CCBPOST_FLAG_SGL_BSIZE,
+			&reg->inbound_queueport);
+	else
+		writel(cdb_shifted_phyaddr, &reg->inbound_queueport);
+}
+
+void arcmsr_post_Qbuffer(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	struct QBUFFER __iomem *pwbuffer = (struct QBUFFER __iomem *) &reg->message_wbuffer;
+	uint8_t __iomem *iop_data = (uint8_t __iomem *) pwbuffer->data;
+	int32_t allxfer_len = 0;
+
+	if (acb->acb_flags & ACB_F_MESSAGE_WQBUFFER_READED) {
+		acb->acb_flags &= (~ACB_F_MESSAGE_WQBUFFER_READED);
+		while ((acb->wqbuf_firstindex != acb->wqbuf_lastindex)
+			&& (allxfer_len < 124)) {
+			writeb(acb->wqbuffer[acb->wqbuf_firstindex], iop_data);
+			acb->wqbuf_firstindex++;
+			acb->wqbuf_firstindex %= ARCMSR_MAX_QBUFFER;
+			iop_data++;
+			allxfer_len++;
+		}
+		writel(allxfer_len, &pwbuffer->data_len);
+		writel(ARCMSR_INBOUND_DRIVER_DATA_WRITE_OK
+			, &reg->inbound_doorbell);
+	}
+}
+
+static void arcmsr_stop_adapter_bgrb(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+
+	acb->acb_flags &= ~ACB_F_MSG_START_BGRB;
+	writel(ARCMSR_INBOUND_MESG0_STOP_BGRB, &reg->inbound_msgaddr0);
+	if (arcmsr_wait_msgint_ready(acb))
+		printk(KERN_NOTICE
+			"arcmsr%d: wait 'stop adapter background rebulid' timeout \n"
+			, acb->host->host_no);
+}
+
+static void arcmsr_free_ccb_pool(struct AdapterControlBlock *acb)
+{
+	dma_free_coherent(&acb->pdev->dev,
+		ARCMSR_MAX_FREECCB_NUM * sizeof (struct CommandControlBlock) + 0x20,
+		acb->dma_coherent,
+		acb->dma_coherent_handle);
+}
+
+static irqreturn_t arcmsr_interrupt(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	struct CommandControlBlock *ccb;
+	uint32_t flag_ccb, outbound_intstatus, outbound_doorbell;
+
+	outbound_intstatus = readl(&reg->outbound_intstatus)
+		& acb->outbound_int_enable;
+	writel(outbound_intstatus, &reg->outbound_intstatus);
+	if (outbound_intstatus & ARCMSR_MU_OUTBOUND_DOORBELL_INT) {
+		outbound_doorbell = readl(&reg->outbound_doorbell);
+		writel(outbound_doorbell, &reg->outbound_doorbell);
+		if (outbound_doorbell & ARCMSR_OUTBOUND_IOP331_DATA_WRITE_OK) {
+			struct QBUFFER __iomem * prbuffer =
+				(struct QBUFFER __iomem *) &reg->message_rbuffer;
+			uint8_t __iomem * iop_data = (uint8_t __iomem *)prbuffer->data;
+			int32_t my_empty_len, iop_len, rqbuf_firstindex, rqbuf_lastindex;
+
+			rqbuf_lastindex = acb->rqbuf_lastindex;
+			rqbuf_firstindex = acb->rqbuf_firstindex;
+			iop_len = readl(&prbuffer->data_len);
+			my_empty_len = (rqbuf_firstindex - rqbuf_lastindex - 1)
+					&(ARCMSR_MAX_QBUFFER - 1);
+			if (my_empty_len >= iop_len) {
+				while (iop_len > 0) {
+					acb->rqbuffer[acb->rqbuf_lastindex] = readb(iop_data);
+					acb->rqbuf_lastindex++;
+					acb->rqbuf_lastindex %= ARCMSR_MAX_QBUFFER;
+					iop_data++;
+					iop_len--;
+				}
+				writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK,
+					&reg->inbound_doorbell);
+			} else
+				acb->acb_flags |= ACB_F_IOPDATA_OVERFLOW;
+		}
+		if (outbound_doorbell & ARCMSR_OUTBOUND_IOP331_DATA_READ_OK) {
+			acb->acb_flags |= ACB_F_MESSAGE_WQBUFFER_READED;
+			if (acb->wqbuf_firstindex != acb->wqbuf_lastindex) {
+				struct QBUFFER __iomem * pwbuffer =
+						(struct QBUFFER __iomem *) &reg->message_wbuffer;
+				uint8_t __iomem * iop_data = (uint8_t __iomem *) pwbuffer->data;
+				int32_t allxfer_len = 0;
+
+				acb->acb_flags &= (~ACB_F_MESSAGE_WQBUFFER_READED);
+				while ((acb->wqbuf_firstindex != acb->wqbuf_lastindex)
+					&& (allxfer_len < 124)) {
+					writeb(acb->wqbuffer[acb->wqbuf_firstindex], iop_data);
+					acb->wqbuf_firstindex++;
+					acb->wqbuf_firstindex %= ARCMSR_MAX_QBUFFER;
+					iop_data++;
+					allxfer_len++;
+				}
+				writel(allxfer_len, &pwbuffer->data_len);
+				writel(ARCMSR_INBOUND_DRIVER_DATA_WRITE_OK,
+					&reg->inbound_doorbell);
+			}
+			if (acb->wqbuf_firstindex == acb->wqbuf_lastindex)
+				acb->acb_flags |= ACB_F_MESSAGE_WQBUFFER_CLEARED;
+		}
+	}
+	if (outbound_intstatus & ARCMSR_MU_OUTBOUND_POSTQUEUE_INT) {
+		int id, lun;
+		/*
+		****************************************************************
+		**               areca cdb command done
+		****************************************************************
+		*/
+		while (1) {
+			if ((flag_ccb = readl(&reg->outbound_queueport)) == 0xFFFFFFFF)
+				break;/*chip FIFO no ccb for completion already*/
+			/* check if command done with no error*/
+			ccb = (struct CommandControlBlock *)(acb->vir2phy_offset +
+				(flag_ccb << 5));
+			if ((ccb->acb != acb) || (ccb->startdone != ARCMSR_CCB_START)) {
+				if (ccb->startdone == ARCMSR_CCB_ABORTED) {
+					struct scsi_cmnd *abortcmd=ccb->pcmd;
+					if (abortcmd) {
+					abortcmd->result |= DID_ABORT >> 16;
+					arcmsr_ccb_complete(ccb, 1);
+					printk(KERN_NOTICE
+						"arcmsr%d: ccb='0x%p' isr got aborted command \n"
+						, acb->host->host_no, ccb);
+					}
+					continue;
+				}
+				printk(KERN_NOTICE
+					"arcmsr%d: isr get an illegal ccb command done acb='0x%p'"
+					"ccb='0x%p' ccbacb='0x%p' startdone = 0x%x"
+					" ccboutstandingcount=%d \n"
+					, acb->host->host_no
+					, acb
+					, ccb
+					, ccb->acb
+					, ccb->startdone
+					, atomic_read(&acb->ccboutstandingcount));
+				continue;
+			}
+			id = ccb->pcmd->device->id;
+			lun = ccb->pcmd->device->lun;
+			if (!(flag_ccb & ARCMSR_CCBREPLY_FLAG_ERROR)) {
+				if (acb->devstate[id][lun] == ARECA_RAID_GONE)
+					acb->devstate[id][lun] = ARECA_RAID_GOOD;
+				ccb->pcmd->result = DID_OK << 16;
+				arcmsr_ccb_complete(ccb, 1);
+			} else {
+				switch(ccb->arcmsr_cdb.DeviceStatus) {
+				case ARCMSR_DEV_SELECT_TIMEOUT: {
+						acb->devstate[id][lun] = ARECA_RAID_GONE;
+						ccb->pcmd->result = DID_TIME_OUT << 16;
+						arcmsr_ccb_complete(ccb, 1);
+					}
+					break;
+				case ARCMSR_DEV_ABORTED:
+				case ARCMSR_DEV_INIT_FAIL: {
+						acb->devstate[id][lun] = ARECA_RAID_GONE;
+						ccb->pcmd->result = DID_BAD_TARGET << 16;
+						arcmsr_ccb_complete(ccb, 1);
+					}
+					break;
+				case ARCMSR_DEV_CHECK_CONDITION: {
+						acb->devstate[id][lun] = ARECA_RAID_GOOD;
+						arcmsr_report_sense_info(ccb);
+						arcmsr_ccb_complete(ccb, 1);
+					}
+					break;
+				default:
+					printk(KERN_NOTICE
+						"arcmsr%d: scsi id=%d lun=%d"
+						" isr get command error done,"
+						"but got unknown DeviceStatus = 0x%x \n"
+						, acb->host->host_no
+						, id
+						, lun
+						, ccb->arcmsr_cdb.DeviceStatus);
+						acb->devstate[id][lun] = ARECA_RAID_GONE;
+						ccb->pcmd->result = DID_NO_CONNECT << 16;
+						arcmsr_ccb_complete(ccb, 1);
+					break;
+				}
+			}
+		}/*drain reply FIFO*/
+	}
+	if (!(outbound_intstatus & ARCMSR_MU_OUTBOUND_HANDLE_INT))
+		return IRQ_NONE;
+	return IRQ_HANDLED;
+}
+
+static void arcmsr_iop_parking(struct AdapterControlBlock *acb)
+{
+	if (acb) {
+		/* stop adapter background rebuild */
+		if (acb->acb_flags & ACB_F_MSG_START_BGRB) {
+			acb->acb_flags &= ~ACB_F_MSG_START_BGRB;
+			arcmsr_stop_adapter_bgrb(acb);
+			arcmsr_flush_adapter_cache(acb);
+		}
+	}
+}
+
+static int arcmsr_iop_message_xfer(struct AdapterControlBlock *acb, struct scsi_cmnd *cmd)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	struct CMD_MESSAGE_FIELD *pcmdmessagefld;
+	int retvalue = 0, transfer_len = 0;
+	char *buffer;
+	uint32_t controlcode = (uint32_t ) cmd->cmnd[5] << 24 |
+						(uint32_t ) cmd->cmnd[6] << 16 |
+						(uint32_t ) cmd->cmnd[7] << 8  |
+						(uint32_t ) cmd->cmnd[8];
+					/* 4 bytes: Areca io control code */
+	if (cmd->use_sg) {
+		struct scatterlist *sg = (struct scatterlist *)cmd->request_buffer;
+
+		buffer = kmap_atomic(sg->page, KM_IRQ0) + sg->offset;
+		if (cmd->use_sg > 1) {
+			retvalue = ARCMSR_MESSAGE_FAIL;
+			goto message_out;
+		}
+		transfer_len += sg->length;
+	} else {
+		buffer = cmd->request_buffer;
+		transfer_len = cmd->request_bufflen;
+	}
+	if (transfer_len > sizeof(struct CMD_MESSAGE_FIELD)) {
+		retvalue = ARCMSR_MESSAGE_FAIL;
+		goto message_out;
+	}
+	pcmdmessagefld = (struct CMD_MESSAGE_FIELD *) buffer;
+	switch(controlcode) {
+	case ARCMSR_MESSAGE_READ_RQBUFFER: {
+			unsigned long *ver_addr;
+			dma_addr_t buf_handle;
+			uint8_t *pQbuffer, *ptmpQbuffer;
+			int32_t allxfer_len = 0;
+
+			ver_addr = pci_alloc_consistent(acb->pdev, 1032, &buf_handle);
+			if (!ver_addr) {
+				retvalue = ARCMSR_MESSAGE_FAIL;
+				goto message_out;
+			}
+			ptmpQbuffer = (uint8_t *) ver_addr;
+			while ((acb->rqbuf_firstindex != acb->rqbuf_lastindex)
+				&& (allxfer_len < 1031)) {
+				pQbuffer = &acb->rqbuffer[acb->rqbuf_firstindex];
+				memcpy(ptmpQbuffer, pQbuffer, 1);
+				acb->rqbuf_firstindex++;
+				acb->rqbuf_firstindex %= ARCMSR_MAX_QBUFFER;
+				ptmpQbuffer++;
+				allxfer_len++;
+			}
+			if (acb->acb_flags & ACB_F_IOPDATA_OVERFLOW) {
+				struct QBUFFER __iomem * prbuffer = (struct QBUFFER __iomem *)
+							&reg->message_rbuffer;
+				uint8_t __iomem * iop_data = (uint8_t __iomem *)prbuffer->data;
+				int32_t iop_len;
+
+				acb->acb_flags &= ~ACB_F_IOPDATA_OVERFLOW;
+				iop_len = readl(&prbuffer->data_len);
+				while (iop_len > 0) {
+					acb->rqbuffer[acb->rqbuf_lastindex] = readb(iop_data);
+					acb->rqbuf_lastindex++;
+					acb->rqbuf_lastindex %= ARCMSR_MAX_QBUFFER;
+					iop_data++;
+					iop_len--;
+				}
+				writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK,
+						&reg->inbound_doorbell);
+			}
+			memcpy(pcmdmessagefld->messagedatabuffer,
+				(uint8_t *)ver_addr, allxfer_len);
+			pcmdmessagefld->cmdmessage.Length = allxfer_len;
+			pcmdmessagefld->cmdmessage.ReturnCode = ARCMSR_MESSAGE_RETURNCODE_OK;
+			pci_free_consistent(acb->pdev, 1032, ver_addr, buf_handle);
+		}
+		break;
+	case ARCMSR_MESSAGE_WRITE_WQBUFFER: {
+			unsigned long *ver_addr;
+			dma_addr_t buf_handle;
+			int32_t my_empty_len, user_len, wqbuf_firstindex, wqbuf_lastindex;
+			uint8_t *pQbuffer, *ptmpuserbuffer;
+
+			ver_addr = pci_alloc_consistent(acb->pdev, 1032, &buf_handle);
+			if (!ver_addr) {
+				retvalue = ARCMSR_MESSAGE_FAIL;
+				goto message_out;
+			}
+			ptmpuserbuffer = (uint8_t *)ver_addr;
+			user_len = pcmdmessagefld->cmdmessage.Length;
+			memcpy(ptmpuserbuffer, pcmdmessagefld->messagedatabuffer, user_len);
+			wqbuf_lastindex = acb->wqbuf_lastindex;
+			wqbuf_firstindex = acb->wqbuf_firstindex;
+			if (wqbuf_lastindex != wqbuf_firstindex) {
+				struct SENSE_DATA *sensebuffer =
+					(struct SENSE_DATA *)cmd->sense_buffer;
+				arcmsr_post_Qbuffer(acb);
+				/* has error report sensedata */
+				sensebuffer->ErrorCode = 0x70;
+				sensebuffer->SenseKey = ILLEGAL_REQUEST;
+				sensebuffer->AdditionalSenseLength = 0x0A;
+				sensebuffer->AdditionalSenseCode = 0x20;
+				sensebuffer->Valid = 1;
+				retvalue = ARCMSR_MESSAGE_FAIL;
+			} else {
+				my_empty_len = (wqbuf_firstindex-wqbuf_lastindex - 1)
+						&(ARCMSR_MAX_QBUFFER - 1);
+				if (my_empty_len >= user_len) {
+					while (user_len > 0) {
+						pQbuffer =
+						&acb->wqbuffer[acb->wqbuf_lastindex];
+						memcpy(pQbuffer, ptmpuserbuffer, 1);
+						acb->wqbuf_lastindex++;
+						acb->wqbuf_lastindex %= ARCMSR_MAX_QBUFFER;
+						ptmpuserbuffer++;
+						user_len--;
+					}
+					if (acb->acb_flags & ACB_F_MESSAGE_WQBUFFER_CLEARED) {
+						acb->acb_flags &=
+							~ACB_F_MESSAGE_WQBUFFER_CLEARED;
+						arcmsr_post_Qbuffer(acb);
+					}
+				} else {
+					/* has error report sensedata */
+					struct SENSE_DATA *sensebuffer =
+						(struct SENSE_DATA *)cmd->sense_buffer;
+					sensebuffer->ErrorCode = 0x70;
+					sensebuffer->SenseKey = ILLEGAL_REQUEST;
+					sensebuffer->AdditionalSenseLength = 0x0A;
+					sensebuffer->AdditionalSenseCode = 0x20;
+					sensebuffer->Valid = 1;
+					retvalue = ARCMSR_MESSAGE_FAIL;
+				}
+			}
+			pci_free_consistent(acb->pdev, 1032, ver_addr, buf_handle);
+		}
+		break;
+	case ARCMSR_MESSAGE_CLEAR_RQBUFFER: {
+			uint8_t *pQbuffer = acb->rqbuffer;
+
+			if (acb->acb_flags & ACB_F_IOPDATA_OVERFLOW) {
+				acb->acb_flags &= ~ACB_F_IOPDATA_OVERFLOW;
+				writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK,
+					&reg->inbound_doorbell);
+			}
+			acb->acb_flags |= ACB_F_MESSAGE_RQBUFFER_CLEARED;
+			acb->rqbuf_firstindex = 0;
+			acb->rqbuf_lastindex = 0;
+			memset(pQbuffer, 0, ARCMSR_MAX_QBUFFER);
+			pcmdmessagefld->cmdmessage.ReturnCode =
+				ARCMSR_MESSAGE_RETURNCODE_OK;
+		}
+		break;
+	case ARCMSR_MESSAGE_CLEAR_WQBUFFER: {
+			uint8_t *pQbuffer = acb->wqbuffer;
+
+			if (acb->acb_flags & ACB_F_IOPDATA_OVERFLOW) {
+				acb->acb_flags &= ~ACB_F_IOPDATA_OVERFLOW;
+				writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK
+						, &reg->inbound_doorbell);
+			}
+			acb->acb_flags |=
+				(ACB_F_MESSAGE_WQBUFFER_CLEARED |
+					ACB_F_MESSAGE_WQBUFFER_READED);
+			acb->wqbuf_firstindex = 0;
+			acb->wqbuf_lastindex = 0;
+			memset(pQbuffer, 0, ARCMSR_MAX_QBUFFER);
+			pcmdmessagefld->cmdmessage.ReturnCode =
+				ARCMSR_MESSAGE_RETURNCODE_OK;
+		}
+		break;
+	case ARCMSR_MESSAGE_CLEAR_ALLQBUFFER: {
+			uint8_t *pQbuffer;
+
+			if (acb->acb_flags & ACB_F_IOPDATA_OVERFLOW) {
+				acb->acb_flags &= ~ACB_F_IOPDATA_OVERFLOW;
+				writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK
+						, &reg->inbound_doorbell);
+			}
+			acb->acb_flags |=
+				(ACB_F_MESSAGE_WQBUFFER_CLEARED
+				| ACB_F_MESSAGE_RQBUFFER_CLEARED
+				| ACB_F_MESSAGE_WQBUFFER_READED);
+			acb->rqbuf_firstindex = 0;
+			acb->rqbuf_lastindex = 0;
+			acb->wqbuf_firstindex = 0;
+			acb->wqbuf_lastindex = 0;
+			pQbuffer = acb->rqbuffer;
+			memset(pQbuffer, 0, sizeof (struct QBUFFER));
+			pQbuffer = acb->wqbuffer;
+			memset(pQbuffer, 0, sizeof (struct QBUFFER));
+			pcmdmessagefld->cmdmessage.ReturnCode = ARCMSR_MESSAGE_RETURNCODE_OK;
+		}
+		break;
+	case ARCMSR_MESSAGE_RETURN_CODE_3F: {
+			pcmdmessagefld->cmdmessage.ReturnCode = ARCMSR_MESSAGE_RETURNCODE_3F;
+		}
+		break;
+	case ARCMSR_MESSAGE_SAY_HELLO: {
+			int8_t * hello_string = "Hello! I am ARCMSR";
+
+			memcpy(pcmdmessagefld->messagedatabuffer, hello_string
+				, (int16_t)strlen(hello_string));
+			pcmdmessagefld->cmdmessage.ReturnCode = ARCMSR_MESSAGE_RETURNCODE_OK;
+		}
+		break;
+	case ARCMSR_MESSAGE_SAY_GOODBYE:
+		arcmsr_iop_parking(acb);
+		break;
+	case ARCMSR_MESSAGE_FLUSH_ADAPTER_CACHE:
+		arcmsr_flush_adapter_cache(acb);
+		break;
+	default:
+		retvalue = ARCMSR_MESSAGE_FAIL;
+	}
+ message_out:
+	if (cmd->use_sg) {
+		struct scatterlist *sg;
+
+		sg = (struct scatterlist *) cmd->request_buffer;
+		kunmap_atomic(buffer - sg->offset, KM_IRQ0);
+	}
+	return retvalue;
+}
+
+static struct CommandControlBlock *arcmsr_get_freeccb(struct AdapterControlBlock *acb)
+{
+	struct list_head *head = &acb->ccb_free_list;
+	struct CommandControlBlock *ccb = NULL;
+
+	if (!list_empty(head)) {
+		ccb = list_entry(head->next, struct CommandControlBlock, list);
+		list_del(head->next);
+	}
+	return ccb;
+}
+
+static void arcmsr_handle_virtual_command(struct AdapterControlBlock *acb,
+		struct scsi_cmnd *cmd)
+{
+	switch (cmd->cmnd[0]) {
+	case INQUIRY: {
+		unsigned char inqdata[36];
+		char *buffer;
+
+		if (cmd->device->lun) {
+			cmd->result = (DID_TIME_OUT << 16);
+			cmd->scsi_done(cmd);
+			return;
+		}
+		inqdata[0] = TYPE_PROCESSOR;
+		/* Periph Qualifier & Periph Dev Type */
+		inqdata[1] = 0;
+		/* rem media bit & Dev Type Modifier */
+		inqdata[2] = 0;
+		/* ISO,ECMA,& ANSI versions */
+		inqdata[4] = 31;
+		/* length of additional data */
+		strncpy(&inqdata[8], "Areca   ", 8);
+		/* Vendor Identification */
+		strncpy(&inqdata[16], "RAID controller ", 16);
+		/* Product Identification */
+		strncpy(&inqdata[32], "R001", 4); /* Product Revision */
+		if (cmd->use_sg) {
+			struct scatterlist *sg;
+
+			sg = (struct scatterlist *) cmd->request_buffer;
+			buffer = kmap_atomic(sg->page, KM_IRQ0) + sg->offset;
+		} else {
+			buffer = cmd->request_buffer;
+		}
+		memcpy(buffer, inqdata, sizeof(inqdata));
+		if (cmd->use_sg) {
+			struct scatterlist *sg;
+
+			sg = (struct scatterlist *) cmd->request_buffer;
+			kunmap_atomic(buffer - sg->offset, KM_IRQ0);
+		}
+		cmd->scsi_done(cmd);
+	}
+	break;
+	case WRITE_BUFFER:
+	case READ_BUFFER: {
+		if (arcmsr_iop_message_xfer(acb, cmd))
+			cmd->result = (DID_ERROR << 16);
+		cmd->scsi_done(cmd);
+	}
+	break;
+	default:
+		cmd->scsi_done(cmd);
+	}
+}
+
+static int arcmsr_queue_command(struct scsi_cmnd *cmd,
+	void (* done)(struct scsi_cmnd *))
+{
+	struct Scsi_Host *host = cmd->device->host;
+	struct AdapterControlBlock *acb =
+		(struct AdapterControlBlock *) host->hostdata;
+	struct CommandControlBlock *ccb;
+	int target = cmd->device->id;
+	int lun = cmd->device->lun;
+
+	cmd->scsi_done = done;
+	cmd->host_scribble = NULL;
+	cmd->result = 0;
+	if (acb->acb_flags & ACB_F_BUS_RESET) {
+		printk(KERN_NOTICE "arcmsr%d: bus reset"
+			" and return busy \n"
+			, acb->host->host_no);
+		return SCSI_MLQUEUE_HOST_BUSY;
+	}
+	if(target == 16) {
+		/* virtual device for iop message transfer */
+		arcmsr_handle_virtual_command(acb, cmd);
+		return 0;
+	}
+	if (acb->devstate[target][lun] == ARECA_RAID_GONE) {
+		uint8_t block_cmd;
+
+		block_cmd = cmd->cmnd[0] & 0x0f;
+		if (block_cmd == 0x08 || block_cmd == 0x0a) {
+			printk(KERN_NOTICE
+				"arcmsr%d: block 'read/write'"
+				"command with gone raid volume"
+				" Cmd=%2x, TargetId=%d, Lun=%d \n"
+				, acb->host->host_no
+				, cmd->cmnd[0]
+				, target, lun);
+			cmd->result = (DID_NO_CONNECT << 16);
+			cmd->scsi_done(cmd);
+			return 0;
+		}
+	}
+	if (atomic_read(&acb->ccboutstandingcount) >=
+			ARCMSR_MAX_OUTSTANDING_CMD)
+		return SCSI_MLQUEUE_HOST_BUSY;
+
+	ccb = arcmsr_get_freeccb(acb);
+	if (!ccb)
+		return SCSI_MLQUEUE_HOST_BUSY;
+	arcmsr_build_ccb(acb, ccb, cmd);
+	arcmsr_post_ccb(acb, ccb);
+	return 0;
+}
+
+static void arcmsr_get_firmware_spec(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	char *acb_firm_model = acb->firm_model;
+	char *acb_firm_version = acb->firm_version;
+	char __iomem *iop_firm_model = (char __iomem *) &reg->message_rwbuffer[15];
+	char __iomem *iop_firm_version = (char __iomem *) &reg->message_rwbuffer[17];
+	int count;
+
+	writel(ARCMSR_INBOUND_MESG0_GET_CONFIG, &reg->inbound_msgaddr0);
+	if (arcmsr_wait_msgint_ready(acb))
+		printk(KERN_NOTICE
+			"arcmsr%d: wait "
+			"'get adapter firmware miscellaneous data' timeout \n"
+			, acb->host->host_no);
+	count = 8;
+	while (count) {
+		*acb_firm_model = readb(iop_firm_model);
+		acb_firm_model++;
+		iop_firm_model++;
+		count--;
+	}
+	count = 16;
+	while (count) {
+		*acb_firm_version = readb(iop_firm_version);
+		acb_firm_version++;
+		iop_firm_version++;
+		count--;
+	}
+	printk(KERN_INFO
+		"ARECA RAID ADAPTER%d: FIRMWARE VERSION %s \n"
+		, acb->host->host_no
+		, acb->firm_version);
+	acb->firm_request_len = readl(&reg->message_rwbuffer[1]);
+	acb->firm_numbers_queue = readl(&reg->message_rwbuffer[2]);
+	acb->firm_sdram_size = readl(&reg->message_rwbuffer[3]);
+	acb->firm_hd_channels = readl(&reg->message_rwbuffer[4]);
+}
+
+static void arcmsr_polling_ccbdone(struct AdapterControlBlock *acb,
+	struct CommandControlBlock *poll_ccb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	struct CommandControlBlock *ccb;
+	uint32_t flag_ccb, outbound_intstatus, poll_ccb_done = 0, poll_count = 0;
+	int id, lun;
+
+ polling_ccb_retry:
+	poll_count++;
+	outbound_intstatus = readl(&reg->outbound_intstatus)
+					& acb->outbound_int_enable;
+	writel(outbound_intstatus, &reg->outbound_intstatus);/*clear interrupt*/
+	while (1) {
+		if ((flag_ccb = readl(&reg->outbound_queueport)) == 0xFFFFFFFF) {
+			if (poll_ccb_done)
+				break;
+			else {
+				msleep(25);
+				if (poll_count > 100)
+					break;
+				goto polling_ccb_retry;
+			}
+		}
+		ccb = (struct CommandControlBlock *)
+			(acb->vir2phy_offset + (flag_ccb << 5));
+		if ((ccb->acb != acb) ||
+			(ccb->startdone != ARCMSR_CCB_START)) {
+			if ((ccb->startdone == ARCMSR_CCB_ABORTED) ||
+				(ccb == poll_ccb)) {
+				printk(KERN_NOTICE
+					"arcmsr%d: scsi id=%d lun=%d ccb='0x%p'"
+					" poll command abort successfully \n"
+					, acb->host->host_no
+					, ccb->pcmd->device->id
+					, ccb->pcmd->device->lun
+					, ccb);
+				ccb->pcmd->result = DID_ABORT << 16;
+				arcmsr_ccb_complete(ccb, 1);
+				poll_ccb_done = 1;
+				continue;
+			}
+			printk(KERN_NOTICE
+				"arcmsr%d: polling get an illegal ccb"
+				" command done ccb='0x%p'"
+				"ccboutstandingcount=%d \n"
+				, acb->host->host_no
+				, ccb
+				, atomic_read(&acb->ccboutstandingcount));
+			continue;
+		}
+		id = ccb->pcmd->device->id;
+		lun = ccb->pcmd->device->lun;
+		if (!(flag_ccb & ARCMSR_CCBREPLY_FLAG_ERROR)) {
+			if (acb->devstate[id][lun] == ARECA_RAID_GONE)
+				acb->devstate[id][lun] = ARECA_RAID_GOOD;
+			ccb->pcmd->result = DID_OK << 16;
+			arcmsr_ccb_complete(ccb, 1);
+		} else {
+			switch(ccb->arcmsr_cdb.DeviceStatus) {
+			case ARCMSR_DEV_SELECT_TIMEOUT: {
+					acb->devstate[id][lun] = ARECA_RAID_GONE;
+					ccb->pcmd->result = DID_TIME_OUT << 16;
+					arcmsr_ccb_complete(ccb, 1);
+				}
+				break;
+			case ARCMSR_DEV_ABORTED:
+			case ARCMSR_DEV_INIT_FAIL: {
+					acb->devstate[id][lun] = ARECA_RAID_GONE;
+					ccb->pcmd->result = DID_BAD_TARGET << 16;
+					arcmsr_ccb_complete(ccb, 1);
+				}
+				break;
+			case ARCMSR_DEV_CHECK_CONDITION: {
+					acb->devstate[id][lun] = ARECA_RAID_GOOD;
+					arcmsr_report_sense_info(ccb);
+					arcmsr_ccb_complete(ccb, 1);
+				}
+				break;
+			default:
+				printk(KERN_NOTICE
+					"arcmsr%d: scsi id=%d lun=%d"
+					" polling and getting command error done"
+					"but got unknown DeviceStatus = 0x%x \n"
+					, acb->host->host_no
+					, id
+					, lun
+					, ccb->arcmsr_cdb.DeviceStatus);
+				acb->devstate[id][lun] = ARECA_RAID_GONE;
+				ccb->pcmd->result = DID_BAD_TARGET << 16;
+				arcmsr_ccb_complete(ccb, 1);
+				break;
+			}
+		}
+	}
+}
+
+static void arcmsr_iop_init(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	uint32_t intmask_org, mask, outbound_doorbell, firmware_state = 0;
+
+	do {
+		firmware_state = readl(&reg->outbound_msgaddr1);
+	} while (!(firmware_state & ARCMSR_OUTBOUND_MESG1_FIRMWARE_OK));
+	intmask_org = readl(&reg->outbound_intmask)
+			| ARCMSR_MU_OUTBOUND_MESSAGE0_INTMASKENABLE;
+	arcmsr_get_firmware_spec(acb);
+
+	acb->acb_flags |= ACB_F_MSG_START_BGRB;
+	writel(ARCMSR_INBOUND_MESG0_START_BGRB, &reg->inbound_msgaddr0);
+	if (arcmsr_wait_msgint_ready(acb)) {
+		printk(KERN_NOTICE "arcmsr%d: "
+			"wait 'start adapter background rebulid' timeout\n",
+			acb->host->host_no);
+	}
+
+	outbound_doorbell = readl(&reg->outbound_doorbell);
+	writel(outbound_doorbell, &reg->outbound_doorbell);
+	writel(ARCMSR_INBOUND_DRIVER_DATA_READ_OK, &reg->inbound_doorbell);
+	mask = ~(ARCMSR_MU_OUTBOUND_POSTQUEUE_INTMASKENABLE
+			| ARCMSR_MU_OUTBOUND_DOORBELL_INTMASKENABLE);
+	writel(intmask_org & mask, &reg->outbound_intmask);
+	acb->outbound_int_enable = ~(intmask_org & mask) & 0x000000ff;
+	acb->acb_flags |= ACB_F_IOP_INITED;
+}
+
+static void arcmsr_iop_reset(struct AdapterControlBlock *acb)
+{
+	struct MessageUnit __iomem *reg = acb->pmu;
+	struct CommandControlBlock *ccb;
+	uint32_t intmask_org;
+	int i = 0;
+
+	if (atomic_read(&acb->ccboutstandingcount) != 0) {
+		/* talk to iop 331 outstanding command aborted */
+		arcmsr_abort_allcmd(acb);
+		/* wait for 3 sec for all command aborted*/
+		msleep_interruptible(3000);
+		/* disable all outbound interrupt */
+		intmask_org = arcmsr_disable_outbound_ints(acb);
+		/* clear all outbound posted Q */
+		for (i = 0; i < ARCMSR_MAX_OUTSTANDING_CMD; i++)
+			readl(&reg->outbound_queueport);
+		for (i = 0; i < ARCMSR_MAX_FREECCB_NUM; i++) {
+			ccb = acb->pccb_pool[i];
+			if ((ccb->startdone == ARCMSR_CCB_START) ||
+				(ccb->startdone == ARCMSR_CCB_ABORTED)) {
+				ccb->startdone = ARCMSR_CCB_ABORTED;
+				ccb->pcmd->result = DID_ABORT << 16;
+				arcmsr_ccb_complete(ccb, 1);
+			}
+		}
+		/* enable all outbound interrupt */
+		arcmsr_enable_outbound_ints(acb, intmask_org);
+	}
+	atomic_set(&acb->ccboutstandingcount, 0);
+}
+
+static int arcmsr_bus_reset(struct scsi_cmnd *cmd)
+{
+	struct AdapterControlBlock *acb =
+		(struct AdapterControlBlock *)cmd->device->host->hostdata;
+	int i;
+
+	acb->num_resets++;
+	acb->acb_flags |= ACB_F_BUS_RESET;
+	for (i = 0; i < 400; i++) {
+		if (!atomic_read(&acb->ccboutstandingcount))
+			break;
+		arcmsr_interrupt(acb);
+		msleep(25);
+	}
+	arcmsr_iop_reset(acb);
+	acb->acb_flags &= ~ACB_F_BUS_RESET;
+	return SUCCESS;
+}
+
+static void arcmsr_abort_one_cmd(struct AdapterControlBlock *acb,
+		struct CommandControlBlock *ccb)
+{
+	u32 intmask;
+
+	ccb->startdone = ARCMSR_CCB_ABORTED;
+
+	/*
+	** Wait for 3 sec for all command done.
+	*/
+	msleep_interruptible(3000);
+
+	intmask = arcmsr_disable_outbound_ints(acb);
+	arcmsr_polling_ccbdone(acb, ccb);
+	arcmsr_enable_outbound_ints(acb, intmask);
+}
+
+static int arcmsr_abort(struct scsi_cmnd *cmd)
+{
+	struct AdapterControlBlock *acb =
+		(struct AdapterControlBlock *)cmd->device->host->hostdata;
+	int i = 0;
+
+	printk(KERN_NOTICE
+		"arcmsr%d: abort device command of scsi id=%d lun=%d \n",
+		acb->host->host_no, cmd->device->id, cmd->device->lun);
+	acb->num_aborts++;
+
+	/*
+	************************************************
+	** the all interrupt service routine is locked
+	** we need to handle it as soon as possible and exit
+	************************************************
+	*/
+	if (!atomic_read(&acb->ccboutstandingcount))
+		return SUCCESS;
+
+	for (i = 0; i < ARCMSR_MAX_FREECCB_NUM; i++) {
+		struct CommandControlBlock *ccb = acb->pccb_pool[i];
+		if (ccb->startdone == ARCMSR_CCB_START && ccb->pcmd == cmd) {
+			arcmsr_abort_one_cmd(acb, ccb);
+			break;
+		}
+	}
+
+	return SUCCESS;
+}
+
+static const char *arcmsr_info(struct Scsi_Host *host)
+{
+	struct AdapterControlBlock *acb =
+		(struct AdapterControlBlock *) host->hostdata;
+	static char buf[256];
+	char *type;
+	int raid6 = 1;
+
+	switch (acb->pdev->device) {
+	case PCI_DEVICE_ID_ARECA_1110:
+	case PCI_DEVICE_ID_ARECA_1210:
+		raid6 = 0;
+		/*FALLTHRU*/
+	case PCI_DEVICE_ID_ARECA_1120:
+	case PCI_DEVICE_ID_ARECA_1130:
+	case PCI_DEVICE_ID_ARECA_1160:
+	case PCI_DEVICE_ID_ARECA_1170:
+	case PCI_DEVICE_ID_ARECA_1220:
+	case PCI_DEVICE_ID_ARECA_1230:
+	case PCI_DEVICE_ID_ARECA_1260:
+	case PCI_DEVICE_ID_ARECA_1270:
+	case PCI_DEVICE_ID_ARECA_1280:
+		type = "SATA";
+		break;
+	case PCI_DEVICE_ID_ARECA_1380:
+	case PCI_DEVICE_ID_ARECA_1381:
+	case PCI_DEVICE_ID_ARECA_1680:
+	case PCI_DEVICE_ID_ARECA_1681:
+		type = "SAS";
+		break;
+	default:
+		type = "X-TYPE";
+		break;
+	}
+	sprintf(buf, "Areca %s Host Adapter RAID Controller%s\n        %s",
+			type, raid6 ? "( RAID6 capable)" : "",
+			ARCMSR_DRIVER_VERSION);
+	return buf;
+}
+
+
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index c09396d2c77b..df7b62676d87 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2004,6 +2004,23 @@
 #define PCI_DEVICE_ID_ALTIMA_AC9100	0x03ea
 #define PCI_DEVICE_ID_ALTIMA_AC1003	0x03eb
 
+#define PCI_VENDOR_ID_ARECA		0x17d3
+#define PCI_DEVICE_ID_ARECA_1110	0x1110
+#define PCI_DEVICE_ID_ARECA_1120	0x1120
+#define PCI_DEVICE_ID_ARECA_1130	0x1130
+#define PCI_DEVICE_ID_ARECA_1160	0x1160
+#define PCI_DEVICE_ID_ARECA_1170	0x1170
+#define PCI_DEVICE_ID_ARECA_1210	0x1210
+#define PCI_DEVICE_ID_ARECA_1220	0x1220
+#define PCI_DEVICE_ID_ARECA_1230	0x1230
+#define PCI_DEVICE_ID_ARECA_1260	0x1260
+#define PCI_DEVICE_ID_ARECA_1270	0x1270
+#define PCI_DEVICE_ID_ARECA_1280	0x1280
+#define PCI_DEVICE_ID_ARECA_1380	0x1380
+#define PCI_DEVICE_ID_ARECA_1381	0x1381
+#define PCI_DEVICE_ID_ARECA_1680	0x1680
+#define PCI_DEVICE_ID_ARECA_1681	0x1681
+
 #define PCI_VENDOR_ID_S2IO		0x17d5
 #define	PCI_DEVICE_ID_S2IO_WIN		0x5731
 #define	PCI_DEVICE_ID_S2IO_UNI		0x5831
-- 
cgit v1.2.3


From 3c5100c1c40cc5e27b4da4a736994c76d93392a0 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Wed, 26 Jul 2006 16:58:33 +0900
Subject: [PATCH] libata: cosmetic changes to PM functions

Unify pm_message_t argument to the new-style @mesg.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/scsi/libata-core.c | 12 ++++++------
 drivers/scsi/libata-scsi.c |  8 ++++----
 include/linux/libata.h     |  6 +++---
 3 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 386e5f21e191..66feebdb01c3 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -5767,11 +5767,11 @@ int pci_test_config_bits(struct pci_dev *pdev, const struct pci_bits *bits)
 	return (tmp == bits->val) ? 1 : 0;
 }
 
-void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t state)
+void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
 	pci_save_state(pdev);
 
-	if (state.event == PM_EVENT_SUSPEND) {
+	if (mesg.event == PM_EVENT_SUSPEND) {
 		pci_disable_device(pdev);
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
@@ -5785,24 +5785,24 @@ void ata_pci_device_do_resume(struct pci_dev *pdev)
 	pci_set_master(pdev);
 }
 
-int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t state)
+int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
 	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
 	int rc = 0;
 
-	rc = ata_host_set_suspend(host_set, state);
+	rc = ata_host_set_suspend(host_set, mesg);
 	if (rc)
 		return rc;
 
 	if (host_set->next) {
-		rc = ata_host_set_suspend(host_set->next, state);
+		rc = ata_host_set_suspend(host_set->next, mesg);
 		if (rc) {
 			ata_host_set_resume(host_set);
 			return rc;
 		}
 	}
 
-	ata_pci_device_do_suspend(pdev, state);
+	ata_pci_device_do_suspend(pdev, mesg);
 
 	return 0;
 }
diff --git a/drivers/scsi/libata-scsi.c b/drivers/scsi/libata-scsi.c
index 7ced41ecde86..1638e57028cb 100644
--- a/drivers/scsi/libata-scsi.c
+++ b/drivers/scsi/libata-scsi.c
@@ -400,7 +400,7 @@ void ata_dump_status(unsigned id, struct ata_taskfile *tf)
 /**
  *	ata_scsi_device_suspend - suspend ATA device associated with sdev
  *	@sdev: the SCSI device to suspend
- *	@state: target power management state
+ *	@mesg: target power management message
  *
  *	Request suspend EH action on the ATA device associated with
  *	@sdev and wait for the operation to complete.
@@ -411,7 +411,7 @@ void ata_dump_status(unsigned id, struct ata_taskfile *tf)
  *	RETURNS:
  *	0 on success, -errno otherwise.
  */
-int ata_scsi_device_suspend(struct scsi_device *sdev, pm_message_t state)
+int ata_scsi_device_suspend(struct scsi_device *sdev, pm_message_t mesg)
 {
 	struct ata_port *ap = ata_shost_to_port(sdev->host);
 	struct ata_device *dev = ata_scsi_find_dev(ap, sdev);
@@ -438,7 +438,7 @@ int ata_scsi_device_suspend(struct scsi_device *sdev, pm_message_t state)
 
 	/* request suspend */
 	action = ATA_EH_SUSPEND;
-	if (state.event != PM_EVENT_SUSPEND)
+	if (mesg.event != PM_EVENT_SUSPEND)
 		action |= ATA_EH_PM_FREEZE;
 	ap->eh_info.dev_action[dev->devno] |= action;
 	ap->eh_info.flags |= ATA_EHI_QUIET;
@@ -463,7 +463,7 @@ int ata_scsi_device_suspend(struct scsi_device *sdev, pm_message_t state)
 	spin_unlock_irqrestore(ap->lock, flags);
  out:
 	if (rc == 0)
-		sdev->sdev_gendev.power.power_state = state;
+		sdev->sdev_gendev.power.power_state = mesg;
 	return rc;
 }
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 66c3100c2b94..b9416708bba2 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -676,9 +676,9 @@ extern void ata_std_ports(struct ata_ioports *ioaddr);
 extern int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info,
 			     unsigned int n_ports);
 extern void ata_pci_remove_one (struct pci_dev *pdev);
-extern void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t state);
+extern void ata_pci_device_do_suspend(struct pci_dev *pdev, pm_message_t mesg);
 extern void ata_pci_device_do_resume(struct pci_dev *pdev);
-extern int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t state);
+extern int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
 extern int ata_pci_device_resume(struct pci_dev *pdev);
 extern int ata_pci_clear_simplex(struct pci_dev *pdev);
 #endif /* CONFIG_PCI */
@@ -697,7 +697,7 @@ extern int sata_scr_write_flush(struct ata_port *ap, int reg, u32 val);
 extern int ata_port_online(struct ata_port *ap);
 extern int ata_port_offline(struct ata_port *ap);
 extern int ata_scsi_device_resume(struct scsi_device *);
-extern int ata_scsi_device_suspend(struct scsi_device *, pm_message_t state);
+extern int ata_scsi_device_suspend(struct scsi_device *, pm_message_t mesg);
 extern int ata_host_set_suspend(struct ata_host_set *host_set,
 				pm_message_t mesg);
 extern void ata_host_set_resume(struct ata_host_set *host_set);
-- 
cgit v1.2.3


From a7f67bdf2c9f24509b8e81e0f35573b611987c80 Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@ozlabs.org>
Date: Wed, 12 Jul 2006 15:35:54 +1000
Subject: [POWERPC] Constify & voidify get_property()

Now that get_property() returns a void *, there's no need to cast its
return value. Also, treat the return value as const, so we can
constify get_property later.

powerpc core changes.

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/btext.c            | 24 ++++++------
 arch/powerpc/kernel/ibmebus.c          |  6 +--
 arch/powerpc/kernel/legacy_serial.c    | 35 ++++++++---------
 arch/powerpc/kernel/lparcfg.c          | 11 +++---
 arch/powerpc/kernel/machine_kexec_64.c | 10 ++---
 arch/powerpc/kernel/pci_32.c           | 39 ++++++++++---------
 arch/powerpc/kernel/pci_64.c           | 28 +++++++-------
 arch/powerpc/kernel/pci_dn.c           | 13 ++++---
 arch/powerpc/kernel/prom.c             | 23 ++++++------
 arch/powerpc/kernel/prom_parse.c       | 68 +++++++++++++++++-----------------
 arch/powerpc/kernel/rtas-proc.c        | 25 +++++++------
 arch/powerpc/kernel/rtas.c             | 28 +++++++-------
 arch/powerpc/kernel/rtas_pci.c         | 22 +++++------
 arch/powerpc/kernel/setup-common.c     | 19 ++++------
 arch/powerpc/kernel/setup_64.c         | 14 +++----
 arch/powerpc/kernel/sysfs.c            |  5 +--
 arch/powerpc/kernel/time.c             |  4 +-
 arch/powerpc/kernel/vio.c              | 16 ++++----
 arch/powerpc/mm/numa.c                 | 31 +++++++---------
 arch/powerpc/sysdev/fsl_soc.c          | 30 +++++++--------
 arch/powerpc/sysdev/mmio_nvram.c       |  4 +-
 include/asm-powerpc/ibmebus.h          |  2 +-
 include/asm-powerpc/prom.h             | 16 ++++----
 include/asm-powerpc/vio.h              |  4 +-
 24 files changed, 239 insertions(+), 238 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/btext.c b/arch/powerpc/kernel/btext.c
index f4e5e14ee2b6..995fcef156fd 100644
--- a/arch/powerpc/kernel/btext.c
+++ b/arch/powerpc/kernel/btext.c
@@ -158,35 +158,35 @@ int btext_initialize(struct device_node *np)
 {
 	unsigned int width, height, depth, pitch;
 	unsigned long address = 0;
-	u32 *prop;
+	const u32 *prop;
 
-	prop = (u32 *)get_property(np, "linux,bootx-width", NULL);
+	prop = get_property(np, "linux,bootx-width", NULL);
 	if (prop == NULL)
-		prop = (u32 *)get_property(np, "width", NULL);
+		prop = get_property(np, "width", NULL);
 	if (prop == NULL)
 		return -EINVAL;
 	width = *prop;
-	prop = (u32 *)get_property(np, "linux,bootx-height", NULL);
+	prop = get_property(np, "linux,bootx-height", NULL);
 	if (prop == NULL)
-		prop = (u32 *)get_property(np, "height", NULL);
+		prop = get_property(np, "height", NULL);
 	if (prop == NULL)
 		return -EINVAL;
 	height = *prop;
-	prop = (u32 *)get_property(np, "linux,bootx-depth", NULL);
+	prop = get_property(np, "linux,bootx-depth", NULL);
 	if (prop == NULL)
-		prop = (u32 *)get_property(np, "depth", NULL);
+		prop = get_property(np, "depth", NULL);
 	if (prop == NULL)
 		return -EINVAL;
 	depth = *prop;
 	pitch = width * ((depth + 7) / 8);
-	prop = (u32 *)get_property(np, "linux,bootx-linebytes", NULL);
+	prop = get_property(np, "linux,bootx-linebytes", NULL);
 	if (prop == NULL)
-		prop = (u32 *)get_property(np, "linebytes", NULL);
+		prop = get_property(np, "linebytes", NULL);
 	if (prop)
 		pitch = *prop;
 	if (pitch == 1)
 		pitch = 0x1000;
-	prop = (u32 *)get_property(np, "address", NULL);
+	prop = get_property(np, "address", NULL);
 	if (prop)
 		address = *prop;
 
@@ -214,11 +214,11 @@ int btext_initialize(struct device_node *np)
 
 int __init btext_find_display(int allow_nonstdout)
 {
-	char *name;
+	const char *name;
 	struct device_node *np = NULL; 
 	int rc = -ENODEV;
 
-	name = (char *)get_property(of_chosen, "linux,stdout-path", NULL);
+	name = get_property(of_chosen, "linux,stdout-path", NULL);
 	if (name != NULL) {
 		np = of_find_node_by_path(name);
 		if (np != NULL) {
diff --git a/arch/powerpc/kernel/ibmebus.c b/arch/powerpc/kernel/ibmebus.c
index 97ddc02a3d42..d9a0b087fa7f 100644
--- a/arch/powerpc/kernel/ibmebus.c
+++ b/arch/powerpc/kernel/ibmebus.c
@@ -167,7 +167,7 @@ static DEVICE_ATTR(name, S_IRUSR | S_IRGRP | S_IROTH, ibmebusdev_show_name,
 		   NULL);
 
 static struct ibmebus_dev* __devinit ibmebus_register_device_common(
-	struct ibmebus_dev *dev, char *name)
+	struct ibmebus_dev *dev, const char *name)
 {
 	int err = 0;
 
@@ -194,10 +194,10 @@ static struct ibmebus_dev* __devinit ibmebus_register_device_node(
 	struct device_node *dn)
 {
 	struct ibmebus_dev *dev;
-	char *loc_code;
+	const char *loc_code;
 	int length;
 
-	loc_code = (char *)get_property(dn, "ibm,loc-code", NULL);
+	loc_code = get_property(dn, "ibm,loc-code", NULL);
 	if (!loc_code) {
                 printk(KERN_WARNING "%s: node %s missing 'ibm,loc-code'\n",
 		       __FUNCTION__, dn->name ? dn->name : "<unknown>");
diff --git a/arch/powerpc/kernel/legacy_serial.c b/arch/powerpc/kernel/legacy_serial.c
index 359ab89748e0..ee1e0b8c7f1f 100644
--- a/arch/powerpc/kernel/legacy_serial.c
+++ b/arch/powerpc/kernel/legacy_serial.c
@@ -39,16 +39,17 @@ static int __init add_legacy_port(struct device_node *np, int want_index,
 				  phys_addr_t taddr, unsigned long irq,
 				  upf_t flags, int irq_check_parent)
 {
-	u32 *clk, *spd, clock = BASE_BAUD * 16;
+	const u32 *clk, *spd;
+	u32 clock = BASE_BAUD * 16;
 	int index;
 
 	/* get clock freq. if present */
-	clk = (u32 *)get_property(np, "clock-frequency", NULL);
+	clk = get_property(np, "clock-frequency", NULL);
 	if (clk && *clk)
 		clock = *clk;
 
 	/* get default speed if present */
-	spd = (u32 *)get_property(np, "current-speed", NULL);
+	spd = get_property(np, "current-speed", NULL);
 
 	/* If we have a location index, then try to use it */
 	if (want_index >= 0 && want_index < MAX_LEGACY_SERIAL_PORTS)
@@ -113,7 +114,7 @@ static int __init add_legacy_soc_port(struct device_node *np,
 				      struct device_node *soc_dev)
 {
 	u64 addr;
-	u32 *addrp;
+	const u32 *addrp;
 	upf_t flags = UPF_BOOT_AUTOCONF | UPF_SKIP_TEST | UPF_SHARE_IRQ;
 
 	/* We only support ports that have a clock frequency properly
@@ -140,15 +141,15 @@ static int __init add_legacy_soc_port(struct device_node *np,
 static int __init add_legacy_isa_port(struct device_node *np,
 				      struct device_node *isa_brg)
 {
-	u32 *reg;
-	char *typep;
+	const u32 *reg;
+	const char *typep;
 	int index = -1;
 	u64 taddr;
 
 	DBG(" -> add_legacy_isa_port(%s)\n", np->full_name);
 
 	/* Get the ISA port number */
-	reg = (u32 *)get_property(np, "reg", NULL);
+	reg = get_property(np, "reg", NULL);
 	if (reg == NULL)
 		return -1;
 
@@ -159,7 +160,7 @@ static int __init add_legacy_isa_port(struct device_node *np,
 	/* Now look for an "ibm,aix-loc" property that gives us ordering
 	 * if any...
 	 */
-	typep = (char *)get_property(np, "ibm,aix-loc", NULL);
+	typep = get_property(np, "ibm,aix-loc", NULL);
 
 	/* If we have a location index, then use it */
 	if (typep && *typep == 'S')
@@ -184,7 +185,7 @@ static int __init add_legacy_pci_port(struct device_node *np,
 				      struct device_node *pci_dev)
 {
 	u64 addr, base;
-	u32 *addrp;
+	const u32 *addrp;
 	unsigned int flags;
 	int iotype, index = -1, lindex = 0;
 
@@ -223,7 +224,7 @@ static int __init add_legacy_pci_port(struct device_node *np,
 	 * we get to their "reg" property
 	 */
 	if (np != pci_dev) {
-		u32 *reg = (u32 *)get_property(np, "reg", NULL);
+		const u32 *reg = get_property(np, "reg", NULL);
 		if (reg && (*reg < 4))
 			index = lindex = *reg;
 	}
@@ -281,13 +282,13 @@ static void __init setup_legacy_serial_console(int console)
 void __init find_legacy_serial_ports(void)
 {
 	struct device_node *np, *stdout = NULL;
-	char *path;
+	const char *path;
 	int index;
 
 	DBG(" -> find_legacy_serial_port()\n");
 
 	/* Now find out if one of these is out firmware console */
-	path = (char *)get_property(of_chosen, "linux,stdout-path", NULL);
+	path = get_property(of_chosen, "linux,stdout-path", NULL);
 	if (path != NULL) {
 		stdout = of_find_node_by_path(path);
 		if (stdout)
@@ -487,8 +488,8 @@ static int __init check_legacy_serial_console(void)
 {
 	struct device_node *prom_stdout = NULL;
 	int speed = 0, offset = 0;
-	char *name;
-	u32 *spd;
+	const char *name;
+	const u32 *spd;
 
 	DBG(" -> check_legacy_serial_console()\n");
 
@@ -509,7 +510,7 @@ static int __init check_legacy_serial_console(void)
 	}
 	/* We are getting a weird phandle from OF ... */
 	/* ... So use the full path instead */
-	name = (char *)get_property(of_chosen, "linux,stdout-path", NULL);
+	name = get_property(of_chosen, "linux,stdout-path", NULL);
 	if (name == NULL) {
 		DBG(" no linux,stdout-path !\n");
 		return -ENODEV;
@@ -521,12 +522,12 @@ static int __init check_legacy_serial_console(void)
 	}
 	DBG("stdout is %s\n", prom_stdout->full_name);
 
-	name = (char *)get_property(prom_stdout, "name", NULL);
+	name = get_property(prom_stdout, "name", NULL);
 	if (!name) {
 		DBG(" stdout package has no name !\n");
 		goto not_found;
 	}
-	spd = (u32 *)get_property(prom_stdout, "current-speed", NULL);
+	spd = get_property(prom_stdout, "current-speed", NULL);
 	if (spd)
 		speed = *spd;
 
diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 2d94b372d49b..3ce3a2d56fa8 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -309,12 +309,11 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 	int partition_potential_processors;
 	int partition_active_processors;
 	struct device_node *rtas_node;
-	int *lrdrp = NULL;
+	const int *lrdrp = NULL;
 
 	rtas_node = find_path_device("/rtas");
 	if (rtas_node)
-		lrdrp = (int *)get_property(rtas_node, "ibm,lrdr-capacity",
-		                            NULL);
+		lrdrp = get_property(rtas_node, "ibm,lrdr-capacity", NULL);
 
 	if (lrdrp == NULL) {
 		partition_potential_processors = vdso_data->processorCount;
@@ -519,7 +518,8 @@ static int lparcfg_data(struct seq_file *m, void *v)
 	const char *model = "";
 	const char *system_id = "";
 	const char *tmp;
-	unsigned int *lp_index_ptr, lp_index = 0;
+	const unsigned int *lp_index_ptr;
+	unsigned int lp_index = 0;
 
 	seq_printf(m, "%s %s \n", MODULE_NAME, MODULE_VERS);
 
@@ -539,8 +539,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
 			if (firmware_has_feature(FW_FEATURE_ISERIES))
 				system_id += 4;
 		}
-		lp_index_ptr = (unsigned int *)
-			get_property(rootdn, "ibm,partition-no", NULL);
+		lp_index_ptr = get_property(rootdn, "ibm,partition-no", NULL);
 		if (lp_index_ptr)
 			lp_index = *lp_index_ptr;
 	}
diff --git a/arch/powerpc/kernel/machine_kexec_64.c b/arch/powerpc/kernel/machine_kexec_64.c
index b438d45a068c..4efdaa9d3f43 100644
--- a/arch/powerpc/kernel/machine_kexec_64.c
+++ b/arch/powerpc/kernel/machine_kexec_64.c
@@ -33,8 +33,8 @@ int default_machine_kexec_prepare(struct kimage *image)
 	unsigned long begin, end;	/* limits of segment */
 	unsigned long low, high;	/* limits of blocked memory range */
 	struct device_node *node;
-	unsigned long *basep;
-	unsigned int *sizep;
+	const unsigned long *basep;
+	const unsigned int *sizep;
 
 	if (!ppc_md.hpte_clear_all)
 		return -ENOENT;
@@ -74,10 +74,8 @@ int default_machine_kexec_prepare(struct kimage *image)
 	/* We also should not overwrite the tce tables */
 	for (node = of_find_node_by_type(NULL, "pci"); node != NULL;
 			node = of_find_node_by_type(node, "pci")) {
-		basep = (unsigned long *)get_property(node, "linux,tce-base",
-							NULL);
-		sizep = (unsigned int *)get_property(node, "linux,tce-size",
-							NULL);
+		basep = get_property(node, "linux,tce-base", NULL);
+		sizep = get_property(node, "linux,tce-size", NULL);
 		if (basep == NULL || sizep == NULL)
 			continue;
 
diff --git a/arch/powerpc/kernel/pci_32.c b/arch/powerpc/kernel/pci_32.c
index 898dae8ab6d9..3f6bd36e9e14 100644
--- a/arch/powerpc/kernel/pci_32.c
+++ b/arch/powerpc/kernel/pci_32.c
@@ -633,12 +633,12 @@ pcibios_alloc_controller(void)
 static void
 make_one_node_map(struct device_node* node, u8 pci_bus)
 {
-	int *bus_range;
+	const int *bus_range;
 	int len;
 
 	if (pci_bus >= pci_bus_count)
 		return;
-	bus_range = (int *) get_property(node, "bus-range", &len);
+	bus_range = get_property(node, "bus-range", &len);
 	if (bus_range == NULL || len < 2 * sizeof(int)) {
 		printk(KERN_WARNING "Can't get bus-range for %s, "
 		       "assuming it starts at 0\n", node->full_name);
@@ -648,13 +648,13 @@ make_one_node_map(struct device_node* node, u8 pci_bus)
 
 	for (node=node->child; node != 0;node = node->sibling) {
 		struct pci_dev* dev;
-		unsigned int *class_code, *reg;
+		const unsigned int *class_code, *reg;
 	
-		class_code = (unsigned int *) get_property(node, "class-code", NULL);
+		class_code = get_property(node, "class-code", NULL);
 		if (!class_code || ((*class_code >> 8) != PCI_CLASS_BRIDGE_PCI &&
 			(*class_code >> 8) != PCI_CLASS_BRIDGE_CARDBUS))
 			continue;
-		reg = (unsigned int *)get_property(node, "reg", NULL);
+		reg = get_property(node, "reg", NULL);
 		if (!reg)
 			continue;
 		dev = pci_find_slot(pci_bus, ((reg[0] >> 8) & 0xff));
@@ -669,7 +669,7 @@ pcibios_make_OF_bus_map(void)
 {
 	int i;
 	struct pci_controller* hose;
-	u8* of_prop_map;
+	struct property *map_prop;
 
 	pci_to_OF_bus_map = (u8*)kmalloc(pci_bus_count, GFP_KERNEL);
 	if (!pci_to_OF_bus_map) {
@@ -691,9 +691,12 @@ pcibios_make_OF_bus_map(void)
 			continue;
 		make_one_node_map(node, hose->first_busno);
 	}
-	of_prop_map = get_property(find_path_device("/"), "pci-OF-bus-map", NULL);
-	if (of_prop_map)
-		memcpy(of_prop_map, pci_to_OF_bus_map, pci_bus_count);
+	map_prop = of_find_property(find_path_device("/"),
+			"pci-OF-bus-map", NULL);
+	if (map_prop) {
+		BUG_ON(pci_bus_count > map_prop->length);
+		memcpy(map_prop->value, pci_to_OF_bus_map, pci_bus_count);
+	}
 #ifdef DEBUG
 	printk("PCI->OF bus map:\n");
 	for (i=0; i<pci_bus_count; i++) {
@@ -712,7 +715,7 @@ scan_OF_pci_childs(struct device_node* node, pci_OF_scan_iterator filter, void*
 	struct device_node* sub_node;
 
 	for (; node != 0;node = node->sibling) {
-		unsigned int *class_code;
+		const unsigned int *class_code;
 	
 		if (filter(node, data))
 			return node;
@@ -722,7 +725,7 @@ scan_OF_pci_childs(struct device_node* node, pci_OF_scan_iterator filter, void*
 		 * a fake root for all functions of a multi-function device,
 		 * we go down them as well.
 		 */
-		class_code = (unsigned int *) get_property(node, "class-code", NULL);
+		class_code = get_property(node, "class-code", NULL);
 		if ((!class_code || ((*class_code >> 8) != PCI_CLASS_BRIDGE_PCI &&
 			(*class_code >> 8) != PCI_CLASS_BRIDGE_CARDBUS)) &&
 			strcmp(node->name, "multifunc-device"))
@@ -737,10 +740,10 @@ scan_OF_pci_childs(struct device_node* node, pci_OF_scan_iterator filter, void*
 static int
 scan_OF_pci_childs_iterator(struct device_node* node, void* data)
 {
-	unsigned int *reg;
+	const unsigned int *reg;
 	u8* fdata = (u8*)data;
 	
-	reg = (unsigned int *) get_property(node, "reg", NULL);
+	reg = get_property(node, "reg", NULL);
 	if (reg && ((reg[0] >> 8) & 0xff) == fdata[1]
 		&& ((reg[0] >> 16) & 0xff) == fdata[0])
 		return 1;
@@ -841,7 +844,7 @@ find_OF_pci_device_filter(struct device_node* node, void* data)
 int
 pci_device_from_OF_node(struct device_node* node, u8* bus, u8* devfn)
 {
-	unsigned int *reg;
+	const unsigned int *reg;
 	struct pci_controller* hose;
 	struct pci_dev* dev = NULL;
 	
@@ -854,7 +857,7 @@ pci_device_from_OF_node(struct device_node* node, u8* bus, u8* devfn)
 	if (!scan_OF_pci_childs(((struct device_node*)hose->arch_data)->child,
 			find_OF_pci_device_filter, (void *)node))
 		return -ENODEV;
-	reg = (unsigned int *) get_property(node, "reg", NULL);
+	reg = get_property(node, "reg", NULL);
 	if (!reg)
 		return -ENODEV;
 	*bus = (reg[0] >> 16) & 0xff;
@@ -885,8 +888,8 @@ pci_process_bridge_OF_ranges(struct pci_controller *hose,
 			   struct device_node *dev, int primary)
 {
 	static unsigned int static_lc_ranges[256] __initdata;
-	unsigned int *dt_ranges, *lc_ranges, *ranges, *prev;
-	unsigned int size;
+	const unsigned int *dt_ranges;
+	unsigned int *lc_ranges, *ranges, *prev, size;
 	int rlen = 0, orig_rlen;
 	int memno = 0;
 	struct resource *res;
@@ -897,7 +900,7 @@ pci_process_bridge_OF_ranges(struct pci_controller *hose,
 	 * that can have more than 3 ranges, fortunately using contiguous
 	 * addresses -- BenH
 	 */
-	dt_ranges = (unsigned int *) get_property(dev, "ranges", &rlen);
+	dt_ranges = get_property(dev, "ranges", &rlen);
 	if (!dt_ranges)
 		return;
 	/* Sanity check, though hopefully that never happens */
diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c
index 1d85fcba51e4..e795a7e2a38e 100644
--- a/arch/powerpc/kernel/pci_64.c
+++ b/arch/powerpc/kernel/pci_64.c
@@ -246,10 +246,10 @@ static void __init pcibios_claim_of_setup(void)
 #ifdef CONFIG_PPC_MULTIPLATFORM
 static u32 get_int_prop(struct device_node *np, const char *name, u32 def)
 {
-	u32 *prop;
+	const u32 *prop;
 	int len;
 
-	prop = (u32 *) get_property(np, name, &len);
+	prop = get_property(np, name, &len);
 	if (prop && len >= 4)
 		return *prop;
 	return def;
@@ -278,10 +278,11 @@ static void pci_parse_of_addrs(struct device_node *node, struct pci_dev *dev)
 	u64 base, size;
 	unsigned int flags;
 	struct resource *res;
-	u32 *addrs, i;
+	const u32 *addrs;
+	u32 i;
 	int proplen;
 
-	addrs = (u32 *) get_property(node, "assigned-addresses", &proplen);
+	addrs = get_property(node, "assigned-addresses", &proplen);
 	if (!addrs)
 		return;
 	DBG("    parse addresses (%d bytes) @ %p\n", proplen, addrs);
@@ -381,7 +382,7 @@ void __devinit of_scan_bus(struct device_node *node,
 				  struct pci_bus *bus)
 {
 	struct device_node *child = NULL;
-	u32 *reg;
+	const u32 *reg;
 	int reglen, devfn;
 	struct pci_dev *dev;
 
@@ -389,7 +390,7 @@ void __devinit of_scan_bus(struct device_node *node,
 
 	while ((child = of_get_next_child(node, child)) != NULL) {
 		DBG("  * %s\n", child->full_name);
-		reg = (u32 *) get_property(child, "reg", &reglen);
+		reg = get_property(child, "reg", &reglen);
 		if (reg == NULL || reglen < 20)
 			continue;
 		devfn = (reg[0] >> 8) & 0xff;
@@ -413,7 +414,7 @@ void __devinit of_scan_pci_bridge(struct device_node *node,
 			 	struct pci_dev *dev)
 {
 	struct pci_bus *bus;
-	u32 *busrange, *ranges;
+	const u32 *busrange, *ranges;
 	int len, i, mode;
 	struct resource *res;
 	unsigned int flags;
@@ -422,13 +423,13 @@ void __devinit of_scan_pci_bridge(struct device_node *node,
 	DBG("of_scan_pci_bridge(%s)\n", node->full_name);
 
 	/* parse bus-range property */
-	busrange = (u32 *) get_property(node, "bus-range", &len);
+	busrange = get_property(node, "bus-range", &len);
 	if (busrange == NULL || len != 8) {
 		printk(KERN_DEBUG "Can't get bus-range for PCI-PCI bridge %s\n",
 		       node->full_name);
 		return;
 	}
-	ranges = (u32 *) get_property(node, "ranges", &len);
+	ranges = get_property(node, "ranges", &len);
 	if (ranges == NULL) {
 		printk(KERN_DEBUG "Can't get ranges for PCI-PCI bridge %s\n",
 		       node->full_name);
@@ -892,13 +893,13 @@ static void __devinit pci_process_ISA_OF_ranges(struct device_node *isa_node,
 		unsigned int size;
 	};
 
-	struct isa_range *range;
+	const struct isa_range *range;
 	unsigned long pci_addr;
 	unsigned int isa_addr;
 	unsigned int size;
 	int rlen = 0;
 
-	range = (struct isa_range *) get_property(isa_node, "ranges", &rlen);
+	range = get_property(isa_node, "ranges", &rlen);
 	if (range == NULL || (rlen < sizeof(struct isa_range))) {
 		printk(KERN_ERR "no ISA ranges or unexpected isa range size,"
 		       "mapping 64k\n");
@@ -939,7 +940,8 @@ static void __devinit pci_process_ISA_OF_ranges(struct device_node *isa_node,
 void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose,
 					    struct device_node *dev, int prim)
 {
-	unsigned int *ranges, pci_space;
+	const unsigned int *ranges;
+	unsigned int pci_space;
 	unsigned long size;
 	int rlen = 0;
 	int memno = 0;
@@ -957,7 +959,7 @@ void __devinit pci_process_bridge_OF_ranges(struct pci_controller *hose,
 	 *			(size depending on dev->n_addr_cells)
 	 *   cells 4+5 or 5+6:	the size of the range
 	 */
-	ranges = (unsigned int *) get_property(dev, "ranges", &rlen);
+	ranges = get_property(dev, "ranges", &rlen);
 	if (ranges == NULL)
 		return;
 	hose->io_base_phys = 0;
diff --git a/arch/powerpc/kernel/pci_dn.c b/arch/powerpc/kernel/pci_dn.c
index 1c18953514c3..68df018dae0e 100644
--- a/arch/powerpc/kernel/pci_dn.c
+++ b/arch/powerpc/kernel/pci_dn.c
@@ -40,8 +40,8 @@
 static void * __devinit update_dn_pci_info(struct device_node *dn, void *data)
 {
 	struct pci_controller *phb = data;
-	int *type = (int *)get_property(dn, "ibm,pci-config-space-type", NULL);
-	u32 *regs;
+	const int *type = get_property(dn, "ibm,pci-config-space-type", NULL);
+	const u32 *regs;
 	struct pci_dn *pdn;
 
 	if (mem_init_done)
@@ -54,14 +54,14 @@ static void * __devinit update_dn_pci_info(struct device_node *dn, void *data)
 	dn->data = pdn;
 	pdn->node = dn;
 	pdn->phb = phb;
-	regs = (u32 *)get_property(dn, "reg", NULL);
+	regs = get_property(dn, "reg", NULL);
 	if (regs) {
 		/* First register entry is addr (00BBSS00)  */
 		pdn->busno = (regs[0] >> 16) & 0xff;
 		pdn->devfn = (regs[0] >> 8) & 0xff;
 	}
 	if (firmware_has_feature(FW_FEATURE_ISERIES)) {
-		u32 *busp = (u32 *)get_property(dn, "linux,subbus", NULL);
+		const u32 *busp = get_property(dn, "linux,subbus", NULL);
 		if (busp)
 			pdn->bussubno = *busp;
 	}
@@ -96,10 +96,11 @@ void *traverse_pci_devices(struct device_node *start, traverse_func pre,
 
 	/* We started with a phb, iterate all childs */
 	for (dn = start->child; dn; dn = nextdn) {
-		u32 *classp, class;
+		const u32 *classp;
+		u32 class;
 
 		nextdn = NULL;
-		classp = (u32 *)get_property(dn, "class-code", NULL);
+		classp = get_property(dn, "class-code", NULL);
 		class = classp ? *classp : 0;
 
 		if (pre && ((ret = pre(dn, data)) != NULL))
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index a1787ffb6319..2a3d84a39cb5 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -942,11 +942,11 @@ void __init early_init_devtree(void *params)
 int
 prom_n_addr_cells(struct device_node* np)
 {
-	int* ip;
+	const int *ip;
 	do {
 		if (np->parent)
 			np = np->parent;
-		ip = (int *) get_property(np, "#address-cells", NULL);
+		ip = get_property(np, "#address-cells", NULL);
 		if (ip != NULL)
 			return *ip;
 	} while (np->parent);
@@ -958,11 +958,11 @@ EXPORT_SYMBOL(prom_n_addr_cells);
 int
 prom_n_size_cells(struct device_node* np)
 {
-	int* ip;
+	const int* ip;
 	do {
 		if (np->parent)
 			np = np->parent;
-		ip = (int *) get_property(np, "#size-cells", NULL);
+		ip = get_property(np, "#size-cells", NULL);
 		if (ip != NULL)
 			return *ip;
 	} while (np->parent);
@@ -1034,7 +1034,7 @@ int device_is_compatible(struct device_node *device, const char *compat)
 	const char* cp;
 	int cplen, l;
 
-	cp = (char *) get_property(device, "compatible", &cplen);
+	cp = get_property(device, "compatible", &cplen);
 	if (cp == NULL)
 		return 0;
 	while (cplen > 0) {
@@ -1449,7 +1449,7 @@ static int of_finish_dynamic_node(struct device_node *node)
 {
 	struct device_node *parent = of_get_parent(node);
 	int err = 0;
-	phandle *ibm_phandle;
+	const phandle *ibm_phandle;
 
 	node->name = get_property(node, "name", NULL);
 	node->type = get_property(node, "device_type", NULL);
@@ -1466,8 +1466,7 @@ static int of_finish_dynamic_node(struct device_node *node)
 		return -ENODEV;
 
 	/* fix up new node's linux_phandle field */
-	if ((ibm_phandle = (unsigned int *)get_property(node,
-							"ibm,phandle", NULL)))
+	if ((ibm_phandle = get_property(node, "ibm,phandle", NULL)))
 		node->linux_phandle = *ibm_phandle;
 
 out:
@@ -1658,16 +1657,16 @@ struct device_node *of_get_cpu_node(int cpu, unsigned int *thread)
 	hardid = get_hard_smp_processor_id(cpu);
 
 	for_each_node_by_type(np, "cpu") {
-		u32 *intserv;
+		const u32 *intserv;
 		unsigned int plen, t;
 
 		/* Check for ibm,ppc-interrupt-server#s. If it doesn't exist
 		 * fallback to "reg" property and assume no threads
 		 */
-		intserv = (u32 *)get_property(np, "ibm,ppc-interrupt-server#s",
-					      &plen);
+		intserv = get_property(np, "ibm,ppc-interrupt-server#s",
+				&plen);
 		if (intserv == NULL) {
-			u32 *reg = (u32 *)get_property(np, "reg", NULL);
+			const u32 *reg = get_property(np, "reg", NULL);
 			if (reg == NULL)
 				continue;
 			if (*reg == hardid) {
diff --git a/arch/powerpc/kernel/prom_parse.c b/arch/powerpc/kernel/prom_parse.c
index e9960170667b..cdcd5d665468 100644
--- a/arch/powerpc/kernel/prom_parse.c
+++ b/arch/powerpc/kernel/prom_parse.c
@@ -270,7 +270,7 @@ static int of_translate_one(struct device_node *parent, struct of_bus *bus,
 			    struct of_bus *pbus, u32 *addr,
 			    int na, int ns, int pna)
 {
-	u32 *ranges;
+	const u32 *ranges;
 	unsigned int rlen;
 	int rone;
 	u64 offset = OF_BAD_ADDR;
@@ -287,7 +287,7 @@ static int of_translate_one(struct device_node *parent, struct of_bus *bus,
 	 * to translate addresses that aren't supposed to be translated in
 	 * the first place. --BenH.
 	 */
-	ranges = (u32 *)get_property(parent, "ranges", &rlen);
+	ranges = get_property(parent, "ranges", &rlen);
 	if (ranges == NULL || rlen == 0) {
 		offset = of_read_number(addr, na);
 		memset(addr, 0, pna * 4);
@@ -330,7 +330,7 @@ static int of_translate_one(struct device_node *parent, struct of_bus *bus,
  * that can be mapped to a cpu physical address). This is not really specified
  * that way, but this is traditionally the way IBM at least do things
  */
-u64 of_translate_address(struct device_node *dev, u32 *in_addr)
+u64 of_translate_address(struct device_node *dev, const u32 *in_addr)
 {
 	struct device_node *parent = NULL;
 	struct of_bus *bus, *pbus;
@@ -407,10 +407,10 @@ u64 of_translate_address(struct device_node *dev, u32 *in_addr)
 }
 EXPORT_SYMBOL(of_translate_address);
 
-u32 *of_get_address(struct device_node *dev, int index, u64 *size,
+const u32 *of_get_address(struct device_node *dev, int index, u64 *size,
 		    unsigned int *flags)
 {
-	u32 *prop;
+	const u32 *prop;
 	unsigned int psize;
 	struct device_node *parent;
 	struct of_bus *bus;
@@ -427,7 +427,7 @@ u32 *of_get_address(struct device_node *dev, int index, u64 *size,
 		return NULL;
 
 	/* Get "reg" or "assigned-addresses" property */
-	prop = (u32 *)get_property(dev, bus->addresses, &psize);
+	prop = get_property(dev, bus->addresses, &psize);
 	if (prop == NULL)
 		return NULL;
 	psize /= 4;
@@ -445,10 +445,10 @@ u32 *of_get_address(struct device_node *dev, int index, u64 *size,
 }
 EXPORT_SYMBOL(of_get_address);
 
-u32 *of_get_pci_address(struct device_node *dev, int bar_no, u64 *size,
+const u32 *of_get_pci_address(struct device_node *dev, int bar_no, u64 *size,
 			unsigned int *flags)
 {
-	u32 *prop;
+	const u32 *prop;
 	unsigned int psize;
 	struct device_node *parent;
 	struct of_bus *bus;
@@ -469,7 +469,7 @@ u32 *of_get_pci_address(struct device_node *dev, int bar_no, u64 *size,
 		return NULL;
 
 	/* Get "reg" or "assigned-addresses" property */
-	prop = (u32 *)get_property(dev, bus->addresses, &psize);
+	prop = get_property(dev, bus->addresses, &psize);
 	if (prop == NULL)
 		return NULL;
 	psize /= 4;
@@ -487,7 +487,7 @@ u32 *of_get_pci_address(struct device_node *dev, int bar_no, u64 *size,
 }
 EXPORT_SYMBOL(of_get_pci_address);
 
-static int __of_address_to_resource(struct device_node *dev, u32 *addrp,
+static int __of_address_to_resource(struct device_node *dev, const u32 *addrp,
 				    u64 size, unsigned int flags,
 				    struct resource *r)
 {
@@ -518,7 +518,7 @@ static int __of_address_to_resource(struct device_node *dev, u32 *addrp,
 int of_address_to_resource(struct device_node *dev, int index,
 			   struct resource *r)
 {
-	u32		*addrp;
+	const u32	*addrp;
 	u64		size;
 	unsigned int	flags;
 
@@ -532,7 +532,7 @@ EXPORT_SYMBOL_GPL(of_address_to_resource);
 int of_pci_address_to_resource(struct device_node *dev, int bar,
 			       struct resource *r)
 {
-	u32		*addrp;
+	const u32	*addrp;
 	u64		size;
 	unsigned int	flags;
 
@@ -543,13 +543,14 @@ int of_pci_address_to_resource(struct device_node *dev, int bar,
 }
 EXPORT_SYMBOL_GPL(of_pci_address_to_resource);
 
-void of_parse_dma_window(struct device_node *dn, unsigned char *dma_window_prop,
+void of_parse_dma_window(struct device_node *dn, const void *dma_window_prop,
 		unsigned long *busno, unsigned long *phys, unsigned long *size)
 {
-	u32 *dma_window, cells;
-	unsigned char *prop;
+	const u32 *dma_window;
+	u32 cells;
+	const unsigned char *prop;
 
-	dma_window = (u32 *)dma_window_prop;
+	dma_window = dma_window_prop;
 
 	/* busno is always one cell */
 	*busno = *(dma_window++);
@@ -578,13 +579,13 @@ static struct device_node *of_irq_dflt_pic;
 static struct device_node *of_irq_find_parent(struct device_node *child)
 {
 	struct device_node *p;
-	phandle *parp;
+	const phandle *parp;
 
 	if (!of_node_get(child))
 		return NULL;
 
 	do {
-		parp = (phandle *)get_property(child, "interrupt-parent", NULL);
+		parp = get_property(child, "interrupt-parent", NULL);
 		if (parp == NULL)
 			p = of_get_parent(child);
 		else {
@@ -646,11 +647,11 @@ void of_irq_map_init(unsigned int flags)
 
 }
 
-int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
-		   struct of_irq *out_irq)
+int of_irq_map_raw(struct device_node *parent, const u32 *intspec,
+		const u32 *addr, struct of_irq *out_irq)
 {
 	struct device_node *ipar, *tnode, *old = NULL, *newpar = NULL;
-	u32 *tmp, *imap, *imask;
+	const u32 *tmp, *imap, *imask;
 	u32 intsize = 1, addrsize, newintsize = 0, newaddrsize = 0;
 	int imaplen, match, i;
 
@@ -661,7 +662,7 @@ int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
 	 * is none, we are nice and just walk up the tree
 	 */
 	do {
-		tmp = (u32 *)get_property(ipar, "#interrupt-cells", NULL);
+		tmp = get_property(ipar, "#interrupt-cells", NULL);
 		if (tmp != NULL) {
 			intsize = *tmp;
 			break;
@@ -682,7 +683,7 @@ int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
 	 */
 	old = of_node_get(ipar);
 	do {
-		tmp = (u32 *)get_property(old, "#address-cells", NULL);
+		tmp = get_property(old, "#address-cells", NULL);
 		tnode = of_get_parent(old);
 		of_node_put(old);
 		old = tnode;
@@ -709,7 +710,7 @@ int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
 		}
 
 		/* Now look for an interrupt-map */
-		imap = (u32 *)get_property(ipar, "interrupt-map", &imaplen);
+		imap = get_property(ipar, "interrupt-map", &imaplen);
 		/* No interrupt map, check for an interrupt parent */
 		if (imap == NULL) {
 			DBG(" -> no map, getting parent\n");
@@ -719,7 +720,7 @@ int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
 		imaplen /= sizeof(u32);
 
 		/* Look for a mask */
-		imask = (u32 *)get_property(ipar, "interrupt-map-mask", NULL);
+		imask = get_property(ipar, "interrupt-map-mask", NULL);
 
 		/* If we were passed no "reg" property and we attempt to parse
 		 * an interrupt-map, then #address-cells must be 0.
@@ -766,14 +767,14 @@ int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
 			/* Get #interrupt-cells and #address-cells of new
 			 * parent
 			 */
-			tmp = (u32 *)get_property(newpar, "#interrupt-cells",
+			tmp = get_property(newpar, "#interrupt-cells",
 						  NULL);
 			if (tmp == NULL) {
 				DBG(" -> parent lacks #interrupt-cells !\n");
 				goto fail;
 			}
 			newintsize = *tmp;
-			tmp = (u32 *)get_property(newpar, "#address-cells",
+			tmp = get_property(newpar, "#address-cells",
 						  NULL);
 			newaddrsize = (tmp == NULL) ? 0 : *tmp;
 
@@ -819,14 +820,14 @@ EXPORT_SYMBOL_GPL(of_irq_map_raw);
 static int of_irq_map_oldworld(struct device_node *device, int index,
 			       struct of_irq *out_irq)
 {
-	u32 *ints;
+	const u32 *ints;
 	int intlen;
 
 	/*
 	 * Old machines just have a list of interrupt numbers
 	 * and no interrupt-controller nodes.
 	 */
-	ints = (u32 *) get_property(device, "AAPL,interrupts", &intlen);
+	ints = get_property(device, "AAPL,interrupts", &intlen);
 	if (ints == NULL)
 		return -EINVAL;
 	intlen /= sizeof(u32);
@@ -851,7 +852,8 @@ static int of_irq_map_oldworld(struct device_node *device, int index,
 int of_irq_map_one(struct device_node *device, int index, struct of_irq *out_irq)
 {
 	struct device_node *p;
-	u32 *intspec, *tmp, intsize, intlen, *addr;
+	const u32 *intspec, *tmp, *addr;
+	u32 intsize, intlen;
 	int res;
 
 	DBG("of_irq_map_one: dev=%s, index=%d\n", device->full_name, index);
@@ -861,13 +863,13 @@ int of_irq_map_one(struct device_node *device, int index, struct of_irq *out_irq
 		return of_irq_map_oldworld(device, index, out_irq);
 
 	/* Get the interrupts property */
-	intspec = (u32 *)get_property(device, "interrupts", &intlen);
+	intspec = get_property(device, "interrupts", &intlen);
 	if (intspec == NULL)
 		return -EINVAL;
 	intlen /= sizeof(u32);
 
 	/* Get the reg property (if any) */
-	addr = (u32 *)get_property(device, "reg", NULL);
+	addr = get_property(device, "reg", NULL);
 
 	/* Look for the interrupt parent. */
 	p = of_irq_find_parent(device);
@@ -875,7 +877,7 @@ int of_irq_map_one(struct device_node *device, int index, struct of_irq *out_irq
 		return -EINVAL;
 
 	/* Get size of interrupt specifier */
-	tmp = (u32 *)get_property(p, "#interrupt-cells", NULL);
+	tmp = get_property(p, "#interrupt-cells", NULL);
 	if (tmp == NULL) {
 		of_node_put(p);
 		return -EINVAL;
diff --git a/arch/powerpc/kernel/rtas-proc.c b/arch/powerpc/kernel/rtas-proc.c
index 9c9ad1fa9cce..2fe82abf1c52 100644
--- a/arch/powerpc/kernel/rtas-proc.c
+++ b/arch/powerpc/kernel/rtas-proc.c
@@ -246,12 +246,12 @@ struct file_operations ppc_rtas_rmo_buf_ops = {
 
 static int ppc_rtas_find_all_sensors(void);
 static void ppc_rtas_process_sensor(struct seq_file *m,
-	struct individual_sensor *s, int state, int error, char *loc);
+	struct individual_sensor *s, int state, int error, const char *loc);
 static char *ppc_rtas_process_error(int error);
 static void get_location_code(struct seq_file *m,
-	struct individual_sensor *s, char *loc);
-static void check_location_string(struct seq_file *m, char *c);
-static void check_location(struct seq_file *m, char *c);
+	struct individual_sensor *s, const char *loc);
+static void check_location_string(struct seq_file *m, const char *c);
+static void check_location(struct seq_file *m, const char *c);
 
 static int __init proc_rtas_init(void)
 {
@@ -446,11 +446,11 @@ static int ppc_rtas_sensors_show(struct seq_file *m, void *v)
 	for (i=0; i<sensors.quant; i++) {
 		struct individual_sensor *p = &sensors.sensor[i];
 		char rstr[64];
-		char *loc;
+		const char *loc;
 		int llen, offs;
 
 		sprintf (rstr, SENSOR_PREFIX"%04d", p->token);
-		loc = (char *) get_property(rtas_node, rstr, &llen);
+		loc = get_property(rtas_node, rstr, &llen);
 
 		/* A sensor may have multiple instances */
 		for (j = 0, offs = 0; j <= p->quant; j++) {
@@ -474,10 +474,10 @@ static int ppc_rtas_sensors_show(struct seq_file *m, void *v)
 
 static int ppc_rtas_find_all_sensors(void)
 {
-	unsigned int *utmp;
+	const unsigned int *utmp;
 	int len, i;
 
-	utmp = (unsigned int *) get_property(rtas_node, "rtas-sensors", &len);
+	utmp = get_property(rtas_node, "rtas-sensors", &len);
 	if (utmp == NULL) {
 		printk (KERN_ERR "error: could not get rtas-sensors\n");
 		return 1;
@@ -530,7 +530,7 @@ static char *ppc_rtas_process_error(int error)
  */
 
 static void ppc_rtas_process_sensor(struct seq_file *m,
-	struct individual_sensor *s, int state, int error, char *loc)
+	struct individual_sensor *s, int state, int error, const char *loc)
 {
 	/* Defined return vales */
 	const char * key_switch[]        = { "Off\t", "Normal\t", "Secure\t", 
@@ -682,7 +682,7 @@ static void ppc_rtas_process_sensor(struct seq_file *m,
 
 /* ****************************************************************** */
 
-static void check_location(struct seq_file *m, char *c)
+static void check_location(struct seq_file *m, const char *c)
 {
 	switch (c[0]) {
 		case LOC_PLANAR:
@@ -719,7 +719,7 @@ static void check_location(struct seq_file *m, char *c)
  * ${LETTER}${NUMBER}[[-/]${LETTER}${NUMBER} [ ... ] ]
  * the '.' may be an abbrevation
  */
-static void check_location_string(struct seq_file *m, char *c)
+static void check_location_string(struct seq_file *m, const char *c)
 {
 	while (*c) {
 		if (isalpha(*c) || *c == '.')
@@ -733,7 +733,8 @@ static void check_location_string(struct seq_file *m, char *c)
 
 /* ****************************************************************** */
 
-static void get_location_code(struct seq_file *m, struct individual_sensor *s, char *loc)
+static void get_location_code(struct seq_file *m, struct individual_sensor *s,
+		const char *loc)
 {
 	if (!loc || !*loc) {
 		seq_printf(m, "---");/* does not have a location */
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index 4a4cb5598402..10e10be324c9 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -177,10 +177,12 @@ void __init udbg_init_rtas_console(void)
 void rtas_progress(char *s, unsigned short hex)
 {
 	struct device_node *root;
-	int width, *p;
+	int width;
+	const int *p;
 	char *os;
 	static int display_character, set_indicator;
-	static int display_width, display_lines, *row_width, form_feed;
+	static int display_width, display_lines, form_feed;
+	const static int *row_width;
 	static DEFINE_SPINLOCK(progress_lock);
 	static int current_line;
 	static int pending_newline = 0;  /* did last write end with unprinted newline? */
@@ -191,16 +193,16 @@ void rtas_progress(char *s, unsigned short hex)
 	if (display_width == 0) {
 		display_width = 0x10;
 		if ((root = find_path_device("/rtas"))) {
-			if ((p = (unsigned int *)get_property(root,
+			if ((p = get_property(root,
 					"ibm,display-line-length", NULL)))
 				display_width = *p;
-			if ((p = (unsigned int *)get_property(root,
+			if ((p = get_property(root,
 					"ibm,form-feed", NULL)))
 				form_feed = *p;
-			if ((p = (unsigned int *)get_property(root,
+			if ((p = get_property(root,
 					"ibm,display-number-of-lines", NULL)))
 				display_lines = *p;
-			row_width = (unsigned int *)get_property(root,
+			row_width = get_property(root,
 					"ibm,display-truncation-length", NULL);
 		}
 		display_character = rtas_token("display-character");
@@ -293,10 +295,10 @@ EXPORT_SYMBOL(rtas_progress);		/* needed by rtas_flash module */
 
 int rtas_token(const char *service)
 {
-	int *tokp;
+	const int *tokp;
 	if (rtas.dev == NULL)
 		return RTAS_UNKNOWN_SERVICE;
-	tokp = (int *) get_property(rtas.dev, service, NULL);
+	tokp = get_property(rtas.dev, service, NULL);
 	return tokp ? *tokp : RTAS_UNKNOWN_SERVICE;
 }
 EXPORT_SYMBOL(rtas_token);
@@ -824,15 +826,15 @@ void __init rtas_initialize(void)
 	 */
 	rtas.dev = of_find_node_by_name(NULL, "rtas");
 	if (rtas.dev) {
-		u32 *basep, *entryp;
-		u32 *sizep;
+		const u32 *basep, *entryp, *sizep;
 
-		basep = (u32 *)get_property(rtas.dev, "linux,rtas-base", NULL);
-		sizep = (u32 *)get_property(rtas.dev, "rtas-size", NULL);
+		basep = get_property(rtas.dev, "linux,rtas-base", NULL);
+		sizep = get_property(rtas.dev, "rtas-size", NULL);
 		if (basep != NULL && sizep != NULL) {
 			rtas.base = *basep;
 			rtas.size = *sizep;
-			entryp = (u32 *)get_property(rtas.dev, "linux,rtas-entry", NULL);
+			entryp = get_property(rtas.dev,
+					"linux,rtas-entry", NULL);
 			if (entryp == NULL) /* Ugh */
 				rtas.entry = rtas.base;
 			else
diff --git a/arch/powerpc/kernel/rtas_pci.c b/arch/powerpc/kernel/rtas_pci.c
index cda022657324..5a798ac6aecf 100644
--- a/arch/powerpc/kernel/rtas_pci.c
+++ b/arch/powerpc/kernel/rtas_pci.c
@@ -57,7 +57,7 @@ static inline int config_access_valid(struct pci_dn *dn, int where)
 
 static int of_device_available(struct device_node * dn)
 {
-        char * status;
+        const char *status;
 
         status = get_property(dn, "status", NULL);
 
@@ -178,7 +178,7 @@ struct pci_ops rtas_pci_ops = {
 
 int is_python(struct device_node *dev)
 {
-	char *model = (char *)get_property(dev, "model", NULL);
+	const char *model = get_property(dev, "model", NULL);
 
 	if (model && strstr(model, "Python"))
 		return 1;
@@ -234,7 +234,7 @@ void __init init_pci_config_tokens (void)
 unsigned long __devinit get_phb_buid (struct device_node *phb)
 {
 	int addr_cells;
-	unsigned int *buid_vals;
+	const unsigned int *buid_vals;
 	unsigned int len;
 	unsigned long buid;
 
@@ -247,7 +247,7 @@ unsigned long __devinit get_phb_buid (struct device_node *phb)
 	if (phb->parent->parent)
 		return 0;
 
-	buid_vals = (unsigned int *) get_property(phb, "reg", &len);
+	buid_vals = get_property(phb, "reg", &len);
 	if (buid_vals == NULL)
 		return 0;
 
@@ -264,10 +264,10 @@ unsigned long __devinit get_phb_buid (struct device_node *phb)
 static int phb_set_bus_ranges(struct device_node *dev,
 			      struct pci_controller *phb)
 {
-	int *bus_range;
+	const int *bus_range;
 	unsigned int len;
 
-	bus_range = (int *) get_property(dev, "bus-range", &len);
+	bus_range = get_property(dev, "bus-range", &len);
 	if (bus_range == NULL || len < 2 * sizeof(int)) {
 		return 1;
  	}
@@ -325,15 +325,15 @@ unsigned long __init find_and_init_phbs(void)
 	 * in chosen.
 	 */
 	if (of_chosen) {
-		int *prop;
+		const int *prop;
 
-		prop = (int *)get_property(of_chosen, "linux,pci-probe-only",
-					   NULL);
+		prop = get_property(of_chosen,
+				"linux,pci-probe-only", NULL);
 		if (prop)
 			pci_probe_only = *prop;
 
-		prop = (int *)get_property(of_chosen,
-					   "linux,pci-assign-all-buses", NULL);
+		prop = get_property(of_chosen,
+				"linux,pci-assign-all-buses", NULL);
 		if (prop)
 			pci_assign_all_buses = *prop;
 	}
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index c6d7b98af7d5..aaf1727b3570 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -304,16 +304,15 @@ struct seq_operations cpuinfo_op = {
 void __init check_for_initrd(void)
 {
 #ifdef CONFIG_BLK_DEV_INITRD
-	unsigned long *prop;
+	const unsigned long *prop;
 
 	DBG(" -> check_for_initrd()\n");
 
 	if (of_chosen) {
-		prop = (unsigned long *)get_property(of_chosen,
-				"linux,initrd-start", NULL);
+		prop = get_property(of_chosen, "linux,initrd-start", NULL);
 		if (prop != NULL) {
 			initrd_start = (unsigned long)__va(*prop);
-			prop = (unsigned long *)get_property(of_chosen,
+			prop = get_property(of_chosen,
 					"linux,initrd-end", NULL);
 			if (prop != NULL) {
 				initrd_end = (unsigned long)__va(*prop);
@@ -366,15 +365,14 @@ void __init smp_setup_cpu_maps(void)
 	int cpu = 0;
 
 	while ((dn = of_find_node_by_type(dn, "cpu")) && cpu < NR_CPUS) {
-		int *intserv;
+		const int *intserv;
 		int j, len = sizeof(u32), nthreads = 1;
 
-		intserv = (int *)get_property(dn, "ibm,ppc-interrupt-server#s",
-					      &len);
+		intserv = get_property(dn, "ibm,ppc-interrupt-server#s", &len);
 		if (intserv)
 			nthreads = len / sizeof(int);
 		else {
-			intserv = (int *) get_property(dn, "reg", NULL);
+			intserv = get_property(dn, "reg", NULL);
 			if (!intserv)
 				intserv = &cpu;	/* assume logical == phys */
 		}
@@ -395,13 +393,12 @@ void __init smp_setup_cpu_maps(void)
 	if (machine_is(pseries) && firmware_has_feature(FW_FEATURE_LPAR) &&
 	    (dn = of_find_node_by_path("/rtas"))) {
 		int num_addr_cell, num_size_cell, maxcpus;
-		unsigned int *ireg;
+		const unsigned int *ireg;
 
 		num_addr_cell = prom_n_addr_cells(dn);
 		num_size_cell = prom_n_size_cells(dn);
 
-		ireg = (unsigned int *)
-			get_property(dn, "ibm,lrdr-capacity", NULL);
+		ireg = get_property(dn, "ibm,lrdr-capacity", NULL);
 
 		if (!ireg)
 			goto out;
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index e2447aef3a8f..77efe19ccd2c 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -106,7 +106,7 @@ static int smt_enabled_cmdline;
 static void check_smt_enabled(void)
 {
 	struct device_node *dn;
-	char *smt_option;
+	const char *smt_option;
 
 	/* Allow the command line to overrule the OF option */
 	if (smt_enabled_cmdline)
@@ -115,7 +115,7 @@ static void check_smt_enabled(void)
 	dn = of_find_node_by_path("/options");
 
 	if (dn) {
-		smt_option = (char *)get_property(dn, "ibm,smt-enabled", NULL);
+		smt_option = get_property(dn, "ibm,smt-enabled", NULL);
 
                 if (smt_option) {
 			if (!strcmp(smt_option, "on"))
@@ -292,7 +292,7 @@ static void __init initialize_cache_info(void)
 		 */
 
 		if ( num_cpus == 1 ) {
-			u32 *sizep, *lsizep;
+			const u32 *sizep, *lsizep;
 			u32 size, lsize;
 			const char *dc, *ic;
 
@@ -307,10 +307,10 @@ static void __init initialize_cache_info(void)
 
 			size = 0;
 			lsize = cur_cpu_spec->dcache_bsize;
-			sizep = (u32 *)get_property(np, "d-cache-size", NULL);
+			sizep = get_property(np, "d-cache-size", NULL);
 			if (sizep != NULL)
 				size = *sizep;
-			lsizep = (u32 *) get_property(np, dc, NULL);
+			lsizep = get_property(np, dc, NULL);
 			if (lsizep != NULL)
 				lsize = *lsizep;
 			if (sizep == 0 || lsizep == 0)
@@ -324,10 +324,10 @@ static void __init initialize_cache_info(void)
 
 			size = 0;
 			lsize = cur_cpu_spec->icache_bsize;
-			sizep = (u32 *)get_property(np, "i-cache-size", NULL);
+			sizep = get_property(np, "i-cache-size", NULL);
 			if (sizep != NULL)
 				size = *sizep;
-			lsizep = (u32 *)get_property(np, ic, NULL);
+			lsizep = get_property(np, ic, NULL);
 			if (lsizep != NULL)
 				lsize = *lsizep;
 			if (sizep == 0 || lsizep == 0)
diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c
index 010435095550..1d724aef438a 100644
--- a/arch/powerpc/kernel/sysfs.c
+++ b/arch/powerpc/kernel/sysfs.c
@@ -60,7 +60,7 @@ static int smt_snooze_cmdline;
 static int __init smt_setup(void)
 {
 	struct device_node *options;
-	unsigned int *val;
+	const unsigned int *val;
 	unsigned int cpu;
 
 	if (!cpu_has_feature(CPU_FTR_SMT))
@@ -70,8 +70,7 @@ static int __init smt_setup(void)
 	if (!options)
 		return -ENODEV;
 
-	val = (unsigned int *)get_property(options, "ibm,smt-snooze-delay",
-					   NULL);
+	val = get_property(options, "ibm,smt-snooze-delay", NULL);
 	if (!smt_snooze_cmdline && val) {
 		for_each_possible_cpu(cpu)
 			per_cpu(smt_snooze_delay, cpu) = *val;
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 774c0a3c5019..8d4ccf061a4d 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -859,14 +859,14 @@ EXPORT_SYMBOL(do_settimeofday);
 static int __init get_freq(char *name, int cells, unsigned long *val)
 {
 	struct device_node *cpu;
-	unsigned int *fp;
+	const unsigned int *fp;
 	int found = 0;
 
 	/* The cpu node should have timebase and clock frequency properties */
 	cpu = of_find_node_by_type(NULL, "cpu");
 
 	if (cpu) {
-		fp = (unsigned int *)get_property(cpu, name, NULL);
+		fp = get_property(cpu, name, NULL);
 		if (fp) {
 			found = 1;
 			*val = 0;
diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c
index fad8580f9081..cb87e71eec66 100644
--- a/arch/powerpc/kernel/vio.c
+++ b/arch/powerpc/kernel/vio.c
@@ -77,7 +77,7 @@ static struct iommu_table *vio_build_iommu_table(struct vio_dev *dev)
 	} else
 #endif
 	{
-		unsigned char *dma_window;
+		const unsigned char *dma_window;
 		struct iommu_table *tbl;
 		unsigned long offset, size;
 
@@ -217,7 +217,7 @@ static void __devinit vio_dev_release(struct device *dev)
 struct vio_dev * __devinit vio_register_device_node(struct device_node *of_node)
 {
 	struct vio_dev *viodev;
-	unsigned int *unit_address;
+	const unsigned int *unit_address;
 
 	/* we need the 'device_type' property, in order to match with drivers */
 	if (of_node->type == NULL) {
@@ -227,7 +227,7 @@ struct vio_dev * __devinit vio_register_device_node(struct device_node *of_node)
 		return NULL;
 	}
 
-	unit_address = (unsigned int *)get_property(of_node, "reg", NULL);
+	unit_address = get_property(of_node, "reg", NULL);
 	if (unit_address == NULL) {
 		printk(KERN_WARNING "%s: node %s missing 'reg'\n",
 				__FUNCTION__,
@@ -249,7 +249,7 @@ struct vio_dev * __devinit vio_register_device_node(struct device_node *of_node)
 	viodev->type = of_node->type;
 	viodev->unit_address = *unit_address;
 	if (firmware_has_feature(FW_FEATURE_ISERIES)) {
-		unit_address = (unsigned int *)get_property(of_node,
+		unit_address = get_property(of_node,
 				"linux,unit_address", NULL);
 		if (unit_address != NULL)
 			viodev->unit_address = *unit_address;
@@ -423,7 +423,7 @@ static int vio_hotplug(struct device *dev, char **envp, int num_envp,
 {
 	const struct vio_dev *vio_dev = to_vio_dev(dev);
 	struct device_node *dn = dev->platform_data;
-	char *cp;
+	const char *cp;
 	int length;
 
 	if (!num_envp)
@@ -431,7 +431,7 @@ static int vio_hotplug(struct device *dev, char **envp, int num_envp,
 
 	if (!dn)
 		return -ENODEV;
-	cp = (char *)get_property(dn, "compatible", &length);
+	cp = get_property(dn, "compatible", &length);
 	if (!cp)
 		return -ENODEV;
 
@@ -493,11 +493,11 @@ static struct vio_dev *vio_find_name(const char *kobj_name)
  */
 struct vio_dev *vio_find_node(struct device_node *vnode)
 {
-	uint32_t *unit_address;
+	const uint32_t *unit_address;
 	char kobj_name[BUS_ID_SIZE];
 
 	/* construct the kobject name from the device node */
-	unit_address = (uint32_t *)get_property(vnode, "reg", NULL);
+	unit_address = get_property(vnode, "reg", NULL);
 	if (!unit_address)
 		return NULL;
 	snprintf(kobj_name, BUS_ID_SIZE, "%x", *unit_address);
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index fbe23933f731..6c0f1c7d83e5 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -159,12 +159,12 @@ static struct device_node * __cpuinit find_cpu_node(unsigned int cpu)
 {
 	unsigned int hw_cpuid = get_hard_smp_processor_id(cpu);
 	struct device_node *cpu_node = NULL;
-	unsigned int *interrupt_server, *reg;
+	const unsigned int *interrupt_server, *reg;
 	int len;
 
 	while ((cpu_node = of_find_node_by_type(cpu_node, "cpu")) != NULL) {
 		/* Try interrupt server first */
-		interrupt_server = (unsigned int *)get_property(cpu_node,
+		interrupt_server = get_property(cpu_node,
 					"ibm,ppc-interrupt-server#s", &len);
 
 		len = len / sizeof(u32);
@@ -175,8 +175,7 @@ static struct device_node * __cpuinit find_cpu_node(unsigned int cpu)
 					return cpu_node;
 			}
 		} else {
-			reg = (unsigned int *)get_property(cpu_node,
-							   "reg", &len);
+			reg = get_property(cpu_node, "reg", &len);
 			if (reg && (len > 0) && (reg[0] == hw_cpuid))
 				return cpu_node;
 		}
@@ -186,9 +185,9 @@ static struct device_node * __cpuinit find_cpu_node(unsigned int cpu)
 }
 
 /* must hold reference to node during call */
-static int *of_get_associativity(struct device_node *dev)
+static const int *of_get_associativity(struct device_node *dev)
 {
-	return (unsigned int *)get_property(dev, "ibm,associativity", NULL);
+	return get_property(dev, "ibm,associativity", NULL);
 }
 
 /* Returns nid in the range [0..MAX_NUMNODES-1], or -1 if no useful numa
@@ -197,7 +196,7 @@ static int *of_get_associativity(struct device_node *dev)
 static int of_node_to_nid_single(struct device_node *device)
 {
 	int nid = -1;
-	unsigned int *tmp;
+	const unsigned int *tmp;
 
 	if (min_common_depth == -1)
 		goto out;
@@ -255,7 +254,7 @@ EXPORT_SYMBOL_GPL(of_node_to_nid);
 static int __init find_min_common_depth(void)
 {
 	int depth;
-	unsigned int *ref_points;
+	const unsigned int *ref_points;
 	struct device_node *rtas_root;
 	unsigned int len;
 
@@ -270,7 +269,7 @@ static int __init find_min_common_depth(void)
 	 * configuration (should be all 0's) and the second is for a normal
 	 * NUMA configuration.
 	 */
-	ref_points = (unsigned int *)get_property(rtas_root,
+	ref_points = get_property(rtas_root,
 			"ibm,associativity-reference-points", &len);
 
 	if ((len >= 1) && ref_points) {
@@ -297,7 +296,7 @@ static void __init get_n_mem_cells(int *n_addr_cells, int *n_size_cells)
 	of_node_put(memory);
 }
 
-static unsigned long __devinit read_n_cells(int n, unsigned int **buf)
+static unsigned long __devinit read_n_cells(int n, const unsigned int **buf)
 {
 	unsigned long result = 0;
 
@@ -435,15 +434,13 @@ static int __init parse_numa_properties(void)
 		unsigned long size;
 		int nid;
 		int ranges;
-		unsigned int *memcell_buf;
+		const unsigned int *memcell_buf;
 		unsigned int len;
 
-		memcell_buf = (unsigned int *)get_property(memory,
+		memcell_buf = get_property(memory,
 			"linux,usable-memory", &len);
 		if (!memcell_buf || len <= 0)
-			memcell_buf =
-				(unsigned int *)get_property(memory, "reg",
-					&len);
+			memcell_buf = get_property(memory, "reg", &len);
 		if (!memcell_buf || len <= 0)
 			continue;
 
@@ -787,10 +784,10 @@ int hot_add_scn_to_nid(unsigned long scn_addr)
 	while ((memory = of_find_node_by_type(memory, "memory")) != NULL) {
 		unsigned long start, size;
 		int ranges;
-		unsigned int *memcell_buf;
+		const unsigned int *memcell_buf;
 		unsigned int len;
 
-		memcell_buf = (unsigned int *)get_property(memory, "reg", &len);
+		memcell_buf = get_property(memory, "reg", &len);
 		if (!memcell_buf || len <= 0)
 			continue;
 
diff --git a/arch/powerpc/sysdev/fsl_soc.c b/arch/powerpc/sysdev/fsl_soc.c
index e983972132d8..07c47e8309ed 100644
--- a/arch/powerpc/sysdev/fsl_soc.c
+++ b/arch/powerpc/sysdev/fsl_soc.c
@@ -41,7 +41,7 @@ phys_addr_t get_immrbase(void)
 	soc = of_find_node_by_type(NULL, "soc");
 	if (soc) {
 		unsigned int size;
-		void *prop = get_property(soc, "reg", &size);
+		const void *prop = get_property(soc, "reg", &size);
 		immrbase = of_translate_address(soc, prop);
 		of_node_put(soc);
 	};
@@ -86,8 +86,8 @@ static int __init gfar_mdio_of_init(void)
 
 		while ((child = of_get_next_child(np, child)) != NULL) {
 			if (child->n_intrs) {
-				u32 *id =
-				    (u32 *) get_property(child, "reg", NULL);
+				const u32 *id =
+					get_property(child, "reg", NULL);
 				mdio_data.irq[*id] = child->intrs[0].line;
 			}
 		}
@@ -127,10 +127,10 @@ static int __init gfar_of_init(void)
 		struct resource r[4];
 		struct device_node *phy, *mdio;
 		struct gianfar_platform_data gfar_data;
-		unsigned int *id;
-		char *model;
-		void *mac_addr;
-		phandle *ph;
+		const unsigned int *id;
+		const char *model;
+		const void *mac_addr;
+		const phandle *ph;
 
 		memset(r, 0, sizeof(r));
 		memset(&gfar_data, 0, sizeof(gfar_data));
@@ -188,7 +188,7 @@ static int __init gfar_of_init(void)
 			    FSL_GIANFAR_DEV_HAS_VLAN |
 			    FSL_GIANFAR_DEV_HAS_EXTENDED_HASH;
 
-		ph = (phandle *) get_property(np, "phy-handle", NULL);
+		ph = get_property(np, "phy-handle", NULL);
 		phy = of_find_node_by_phandle(*ph);
 
 		if (phy == NULL) {
@@ -198,7 +198,7 @@ static int __init gfar_of_init(void)
 
 		mdio = of_get_parent(phy);
 
-		id = (u32 *) get_property(phy, "reg", NULL);
+		id = get_property(phy, "reg", NULL);
 		ret = of_address_to_resource(mdio, 0, &res);
 		if (ret) {
 			of_node_put(phy);
@@ -242,7 +242,7 @@ static int __init fsl_i2c_of_init(void)
 	     i++) {
 		struct resource r[2];
 		struct fsl_i2c_platform_data i2c_data;
-		unsigned char *flags = NULL;
+		const unsigned char *flags = NULL;
 
 		memset(&r, 0, sizeof(r));
 		memset(&i2c_data, 0, sizeof(i2c_data));
@@ -294,7 +294,7 @@ static int __init mpc83xx_wdt_init(void)
 	struct resource r;
 	struct device_node *soc, *np;
 	struct platform_device *dev;
-	unsigned int *freq;
+	const unsigned int *freq;
 	int ret;
 
 	np = of_find_compatible_node(NULL, "watchdog", "mpc83xx_wdt");
@@ -311,7 +311,7 @@ static int __init mpc83xx_wdt_init(void)
 		goto nosoc;
 	}
 
-	freq = (unsigned int *)get_property(soc, "bus-frequency", NULL);
+	freq = get_property(soc, "bus-frequency", NULL);
 	if (!freq) {
 		ret = -ENODEV;
 		goto err;
@@ -351,7 +351,7 @@ nodev:
 arch_initcall(mpc83xx_wdt_init);
 #endif
 
-static enum fsl_usb2_phy_modes determine_usb_phy(char * phy_type)
+static enum fsl_usb2_phy_modes determine_usb_phy(const char *phy_type)
 {
 	if (!phy_type)
 		return FSL_USB2_PHY_NONE;
@@ -379,7 +379,7 @@ static int __init fsl_usb_of_init(void)
 	     i++) {
 		struct resource r[2];
 		struct fsl_usb2_platform_data usb_data;
-		unsigned char *prop = NULL;
+		const unsigned char *prop = NULL;
 
 		memset(&r, 0, sizeof(r));
 		memset(&usb_data, 0, sizeof(usb_data));
@@ -428,7 +428,7 @@ static int __init fsl_usb_of_init(void)
 	     i++) {
 		struct resource r[2];
 		struct fsl_usb2_platform_data usb_data;
-		unsigned char *prop = NULL;
+		const unsigned char *prop = NULL;
 
 		memset(&r, 0, sizeof(r));
 		memset(&usb_data, 0, sizeof(usb_data));
diff --git a/arch/powerpc/sysdev/mmio_nvram.c b/arch/powerpc/sysdev/mmio_nvram.c
index 615350d46b52..ff23f5a4d4b9 100644
--- a/arch/powerpc/sysdev/mmio_nvram.c
+++ b/arch/powerpc/sysdev/mmio_nvram.c
@@ -80,7 +80,7 @@ static ssize_t mmio_nvram_get_size(void)
 int __init mmio_nvram_init(void)
 {
 	struct device_node *nvram_node;
-	unsigned long *buffer;
+	const unsigned long *buffer;
 	int proplen;
 	unsigned long nvram_addr;
 	int ret;
@@ -91,7 +91,7 @@ int __init mmio_nvram_init(void)
 		goto out;
 
 	ret = -EIO;
-	buffer = (unsigned long *)get_property(nvram_node, "reg", &proplen);
+	buffer = get_property(nvram_node, "reg", &proplen);
 	if (proplen != 2*sizeof(unsigned long))
 		goto out;
 
diff --git a/include/asm-powerpc/ibmebus.h b/include/asm-powerpc/ibmebus.h
index 7a42723d107c..7ab195a27888 100644
--- a/include/asm-powerpc/ibmebus.h
+++ b/include/asm-powerpc/ibmebus.h
@@ -48,7 +48,7 @@ extern struct dma_mapping_ops ibmebus_dma_ops;
 extern struct bus_type ibmebus_bus_type;
 
 struct ibmebus_dev {	
-	char *name;
+	const char *name;
 	struct of_device ofdev;
 };
 
diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h
index 56f6ea0c76de..abdf1be66e97 100644
--- a/include/asm-powerpc/prom.h
+++ b/include/asm-powerpc/prom.h
@@ -72,8 +72,8 @@ struct property {
 };
 
 struct device_node {
-	char	*name;
-	char	*type;
+	const char *name;
+	const char *type;
 	phandle	node;
 	phandle linux_phandle;
 	char	*full_name;
@@ -209,15 +209,15 @@ static inline u64 of_read_number(const u32 *cell, int size)
 /* Translate an OF address block into a CPU physical address
  */
 #define OF_BAD_ADDR	((u64)-1)
-extern u64 of_translate_address(struct device_node *np, u32 *addr);
+extern u64 of_translate_address(struct device_node *np, const u32 *addr);
 
 /* Extract an address from a device, returns the region size and
  * the address space flags too. The PCI version uses a BAR number
  * instead of an absolute index
  */
-extern u32 *of_get_address(struct device_node *dev, int index,
+extern const u32 *of_get_address(struct device_node *dev, int index,
 			   u64 *size, unsigned int *flags);
-extern u32 *of_get_pci_address(struct device_node *dev, int bar_no,
+extern const u32 *of_get_pci_address(struct device_node *dev, int bar_no,
 			       u64 *size, unsigned int *flags);
 
 /* Get an address as a resource. Note that if your address is
@@ -234,7 +234,7 @@ extern int of_pci_address_to_resource(struct device_node *dev, int bar,
 /* Parse the ibm,dma-window property of an OF node into the busno, phys and
  * size parameters.
  */
-void of_parse_dma_window(struct device_node *dn, unsigned char *dma_window_prop,
+void of_parse_dma_window(struct device_node *dn, const void *dma_window_prop,
 		unsigned long *busno, unsigned long *phys, unsigned long *size);
 
 extern void kdump_move_device_tree(void);
@@ -288,8 +288,8 @@ extern void of_irq_map_init(unsigned int flags);
  *
  */
 
-extern int of_irq_map_raw(struct device_node *parent, u32 *intspec, u32 *addr,
-			  struct of_irq *out_irq);
+extern int of_irq_map_raw(struct device_node *parent, const u32 *intspec,
+			  const u32 *addr, struct of_irq *out_irq);
 
 
 /***
diff --git a/include/asm-powerpc/vio.h b/include/asm-powerpc/vio.h
index dc9bd101ca14..4b51d42e1419 100644
--- a/include/asm-powerpc/vio.h
+++ b/include/asm-powerpc/vio.h
@@ -46,8 +46,8 @@ struct iommu_table;
  */
 struct vio_dev {
 	struct iommu_table *iommu_table;     /* vio_map_* uses this */
-	char *name;
-	char *type;
+	const char *name;
+	const char *type;
 	uint32_t unit_address;
 	unsigned int irq;
 	struct device dev;
-- 
cgit v1.2.3


From c61c27d58af61e5b78257019b173732c29ce0c64 Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@ozlabs.org>
Date: Wed, 12 Jul 2006 15:39:54 +1000
Subject: [POWERPC] cell: Constify & voidify get_property()

Now that get_property() returns a void *, there's no need to cast its
return value. Also, treat the return value as const, so we can
constify get_property later.

cell platform changes.

Built for cell_defconfig

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/cell/cbe_regs.c   |  8 +++-----
 arch/powerpc/platforms/cell/interrupt.c  |  5 ++---
 arch/powerpc/platforms/cell/iommu.c      | 22 ++++++++++++----------
 arch/powerpc/platforms/cell/spider-pic.c | 10 +++++-----
 arch/powerpc/platforms/cell/spu_base.c   | 18 +++++++++---------
 include/asm-powerpc/spu.h                |  2 +-
 6 files changed, 32 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/cell/cbe_regs.c b/arch/powerpc/platforms/cell/cbe_regs.c
index ce696c1cca75..3f3859d12e00 100644
--- a/arch/powerpc/platforms/cell/cbe_regs.c
+++ b/arch/powerpc/platforms/cell/cbe_regs.c
@@ -97,7 +97,7 @@ void __init cbe_regs_init(void)
 		struct cbe_regs_map *map = &cbe_regs_maps[cbe_regs_map_count++];
 
 		/* That hack must die die die ! */
-		struct address_prop {
+		const struct address_prop {
 			unsigned long address;
 			unsigned int len;
 		} __attribute__((packed)) *prop;
@@ -114,13 +114,11 @@ void __init cbe_regs_init(void)
 			if (cbe_thread_map[i].cpu_node == cpu)
 				cbe_thread_map[i].regs = map;
 
-		prop = (struct address_prop *)get_property(cpu, "pervasive",
-							   NULL);
+		prop = get_property(cpu, "pervasive", NULL);
 		if (prop != NULL)
 			map->pmd_regs = ioremap(prop->address, prop->len);
 
-		prop = (struct address_prop *)get_property(cpu, "iic",
-							   NULL);
+		prop = get_property(cpu, "iic", NULL);
 		if (prop != NULL)
 			map->iic_regs = ioremap(prop->address, prop->len);
 	}
diff --git a/arch/powerpc/platforms/cell/interrupt.c b/arch/powerpc/platforms/cell/interrupt.c
index 9d5da7896892..b26b496f6548 100644
--- a/arch/powerpc/platforms/cell/interrupt.c
+++ b/arch/powerpc/platforms/cell/interrupt.c
@@ -250,15 +250,14 @@ static int __init setup_iic(void)
 	struct resource r0, r1;
 	struct irq_host *host;
 	int found = 0;
- 	u32 *np;
+ 	const u32 *np;
 
 	for (dn = NULL;
 	     (dn = of_find_node_by_name(dn,"interrupt-controller")) != NULL;) {
 		if (!device_is_compatible(dn,
 				     "IBM,CBEA-Internal-Interrupt-Controller"))
 			continue;
- 		np = (u32 *)get_property(dn, "ibm,interrupt-server-ranges",
-					 NULL);
+		np = get_property(dn, "ibm,interrupt-server-ranges", NULL);
  		if (np == NULL) {
 			printk(KERN_WARNING "IIC: CPU association not found\n");
 			of_node_put(dn);
diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index a35004e14c69..d2b20eba5b87 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -308,15 +308,16 @@ static void cell_do_map_iommu(struct cell_iommu *iommu,
 
 static void iommu_devnode_setup(struct device_node *d)
 {
-	unsigned int *ioid;
-	unsigned long *dma_window, map_start, map_size, token;
+	const unsigned int *ioid;
+	unsigned long map_start, map_size, token;
+	const unsigned long *dma_window;
 	struct cell_iommu *iommu;
 
-	ioid = (unsigned int *)get_property(d, "ioid", NULL);
+	ioid = get_property(d, "ioid", NULL);
 	if (!ioid)
 		pr_debug("No ioid entry found !\n");
 
-	dma_window = (unsigned long *)get_property(d, "ibm,dma-window", NULL);
+	dma_window = get_property(d, "ibm,dma-window", NULL);
 	if (!dma_window)
 		pr_debug("No ibm,dma-window entry found !\n");
 
@@ -371,8 +372,9 @@ static int cell_map_iommu_hardcoded(int num_nodes)
 
 static int cell_map_iommu(void)
 {
-	unsigned int num_nodes = 0, *node_id;
-	unsigned long *base, *mmio_base;
+	unsigned int num_nodes = 0;
+	const unsigned int *node_id;
+	const unsigned long *base, *mmio_base;
 	struct device_node *dn;
 	struct cell_iommu *iommu = NULL;
 
@@ -381,7 +383,7 @@ static int cell_map_iommu(void)
 	for(dn = of_find_node_by_type(NULL, "cpu");
 	    dn;
 	    dn = of_find_node_by_type(dn, "cpu")) {
-		node_id = (unsigned int *)get_property(dn, "node-id", NULL);
+		node_id = get_property(dn, "node-id", NULL);
 
 		if (num_nodes < *node_id)
 			num_nodes = *node_id;
@@ -396,9 +398,9 @@ static int cell_map_iommu(void)
 	    dn;
 	    dn = of_find_node_by_type(dn, "cpu")) {
 
-		node_id = (unsigned int *)get_property(dn, "node-id", NULL);
-		base = (unsigned long *)get_property(dn, "ioc-cache", NULL);
-		mmio_base = (unsigned long *)get_property(dn, "ioc-translation", NULL);
+		node_id = get_property(dn, "node-id", NULL);
+		base = get_property(dn, "ioc-cache", NULL);
+		mmio_base = get_property(dn, "ioc-translation", NULL);
 
 		if (!base || !mmio_base || !node_id)
 			return cell_map_iommu_hardcoded(num_nodes);
diff --git a/arch/powerpc/platforms/cell/spider-pic.c b/arch/powerpc/platforms/cell/spider-pic.c
index ae7ef88f1a37..ab4c252a4d9b 100644
--- a/arch/powerpc/platforms/cell/spider-pic.c
+++ b/arch/powerpc/platforms/cell/spider-pic.c
@@ -230,7 +230,7 @@ static void spider_irq_cascade(unsigned int irq, struct irq_desc *desc,
 static unsigned int __init spider_find_cascade_and_node(struct spider_pic *pic)
 {
 	unsigned int virq;
-	u32 *imap, *tmp;
+	const u32 *imap, *tmp;
 	int imaplen, intsize, unit;
 	struct device_node *iic;
 	struct irq_host *iic_host;
@@ -248,25 +248,25 @@ static unsigned int __init spider_find_cascade_and_node(struct spider_pic *pic)
 #endif
 
 	/* Now do the horrible hacks */
-	tmp = (u32 *)get_property(pic->of_node, "#interrupt-cells", NULL);
+	tmp = get_property(pic->of_node, "#interrupt-cells", NULL);
 	if (tmp == NULL)
 		return NO_IRQ;
 	intsize = *tmp;
-	imap = (u32 *)get_property(pic->of_node, "interrupt-map", &imaplen);
+	imap = get_property(pic->of_node, "interrupt-map", &imaplen);
 	if (imap == NULL || imaplen < (intsize + 1))
 		return NO_IRQ;
 	iic = of_find_node_by_phandle(imap[intsize]);
 	if (iic == NULL)
 		return NO_IRQ;
 	imap += intsize + 1;
-	tmp = (u32 *)get_property(iic, "#interrupt-cells", NULL);
+	tmp = get_property(iic, "#interrupt-cells", NULL);
 	if (tmp == NULL)
 		return NO_IRQ;
 	intsize = *tmp;
 	/* Assume unit is last entry of interrupt specifier */
 	unit = imap[intsize - 1];
 	/* Ok, we have a unit, now let's try to get the node */
-	tmp = (u32 *)get_property(iic, "ibm,interrupt-server-ranges", NULL);
+	tmp = get_property(iic, "ibm,interrupt-server-ranges", NULL);
 	if (tmp == NULL) {
 		of_node_put(iic);
 		return NO_IRQ;
diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c
index 5d2313a6c82b..86d55675e1d2 100644
--- a/arch/powerpc/platforms/cell/spu_base.c
+++ b/arch/powerpc/platforms/cell/spu_base.c
@@ -488,10 +488,10 @@ int spu_irq_class_1_bottom(struct spu *spu)
 
 static int __init find_spu_node_id(struct device_node *spe)
 {
-	unsigned int *id;
+	const unsigned int *id;
 	struct device_node *cpu;
 	cpu = spe->parent->parent;
-	id = (unsigned int *)get_property(cpu, "node-id", NULL);
+	id = get_property(cpu, "node-id", NULL);
 	return id ? *id : 0;
 }
 
@@ -500,7 +500,7 @@ static int __init cell_spuprop_present(struct spu *spu, struct device_node *spe,
 {
 	static DEFINE_MUTEX(add_spumem_mutex);
 
-	struct address_prop {
+	const struct address_prop {
 		unsigned long address;
 		unsigned int len;
 	} __attribute__((packed)) *p;
@@ -511,7 +511,7 @@ static int __init cell_spuprop_present(struct spu *spu, struct device_node *spe,
 	struct zone *zone;
 	int ret;
 
-	p = (void*)get_property(spe, prop, &proplen);
+	p = get_property(spe, prop, &proplen);
 	WARN_ON(proplen != sizeof (*p));
 
 	start_pfn = p->address >> PAGE_SHIFT;
@@ -531,12 +531,12 @@ static int __init cell_spuprop_present(struct spu *spu, struct device_node *spe,
 static void __iomem * __init map_spe_prop(struct spu *spu,
 		struct device_node *n, const char *name)
 {
-	struct address_prop {
+	const struct address_prop {
 		unsigned long address;
 		unsigned int len;
 	} __attribute__((packed)) *prop;
 
-	void *p;
+	const void *p;
 	int proplen;
 	void* ret = NULL;
 	int err = 0;
@@ -570,14 +570,14 @@ static int __init spu_map_interrupts(struct spu *spu, struct device_node *np)
 {
 	struct irq_host *host;
 	unsigned int isrc;
-	u32 *tmp;
+	const u32 *tmp;
 
 	host = iic_get_irq_host(spu->node);
 	if (host == NULL)
 		return -ENODEV;
 
 	/* Get the interrupt source from the device-tree */
-	tmp = (u32 *)get_property(np, "isrc", NULL);
+	tmp = get_property(np, "isrc", NULL);
 	if (!tmp)
 		return -ENODEV;
 	spu->isrc = isrc = tmp[0];
@@ -593,7 +593,7 @@ static int __init spu_map_interrupts(struct spu *spu, struct device_node *np)
 
 static int __init spu_map_device(struct spu *spu, struct device_node *node)
 {
-	char *prop;
+	const char *prop;
 	int ret;
 
 	ret = -ENODEV;
diff --git a/include/asm-powerpc/spu.h b/include/asm-powerpc/spu.h
index c02d105d8294..b42b53c40f5d 100644
--- a/include/asm-powerpc/spu.h
+++ b/include/asm-powerpc/spu.h
@@ -106,7 +106,7 @@ struct spu_context;
 struct spu_runqueue;
 
 struct spu {
-	char *name;
+	const char *name;
 	unsigned long local_store_phys;
 	u8 *local_store;
 	unsigned long problem_phys;
-- 
cgit v1.2.3


From 018a3d1db7cdb6127656c1622ee1d2302e16436d Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@ozlabs.org>
Date: Wed, 12 Jul 2006 15:40:29 +1000
Subject: [POWERPC] powermac: Constify & voidify get_property()

Now that get_property() returns a void *, there's no need to cast its
return value. Also, treat the return value as const, so we can
constify get_property later.

powermac platform & macintosh driver changes.

Built for pmac32_defconfig, g5_defconfig

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/powermac/backlight.c  |  3 ++-
 arch/powerpc/platforms/powermac/cpufreq_32.c | 23 ++++++++---------
 arch/powerpc/platforms/powermac/cpufreq_64.c | 27 ++++++++++----------
 arch/powerpc/platforms/powermac/feature.c    | 30 +++++++++++-----------
 arch/powerpc/platforms/powermac/low_i2c.c    | 24 +++++++++---------
 arch/powerpc/platforms/powermac/pci.c        | 37 +++++++++++++++-------------
 arch/powerpc/platforms/powermac/pfunc_base.c |  2 +-
 arch/powerpc/platforms/powermac/pfunc_core.c |  5 ++--
 arch/powerpc/platforms/powermac/setup.c      | 18 ++++++--------
 arch/powerpc/platforms/powermac/smp.c        |  7 +++---
 arch/powerpc/platforms/powermac/udbg_scc.c   | 10 ++++----
 drivers/i2c/busses/i2c-powermac.c            |  3 ++-
 drivers/ide/ppc/pmac.c                       |  6 ++---
 drivers/macintosh/macio_asic.c               | 10 +++++---
 drivers/macintosh/macio_sysfs.c              |  8 +++---
 drivers/macintosh/smu.c                      | 19 +++++++-------
 drivers/macintosh/therm_adt746x.c            |  8 +++---
 drivers/macintosh/therm_pm72.c               | 14 +++++------
 drivers/macintosh/therm_windtunnel.c         |  4 +--
 drivers/macintosh/via-cuda.c                 |  4 +--
 drivers/macintosh/via-pmu-led.c              |  2 +-
 drivers/macintosh/via-pmu.c                  | 10 ++++----
 drivers/macintosh/windfarm_pm81.c            |  4 +--
 drivers/macintosh/windfarm_pm91.c            |  2 +-
 drivers/macintosh/windfarm_smu_controls.c    | 13 +++++-----
 drivers/macintosh/windfarm_smu_sat.c         |  8 +++---
 drivers/macintosh/windfarm_smu_sensors.c     | 12 ++++-----
 drivers/serial/pmac_zilog.c                  |  9 ++++---
 include/asm-powerpc/smu.h                    |  2 +-
 29 files changed, 166 insertions(+), 158 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/powermac/backlight.c b/arch/powerpc/platforms/powermac/backlight.c
index 69f65e215a5c..205b4a392862 100644
--- a/arch/powerpc/platforms/powermac/backlight.c
+++ b/arch/powerpc/platforms/powermac/backlight.c
@@ -38,7 +38,8 @@ int pmac_has_backlight_type(const char *type)
 	struct device_node* bk_node = find_devices("backlight");
 
 	if (bk_node) {
-		char *prop = get_property(bk_node, "backlight-control", NULL);
+		const char *prop = get_property(bk_node,
+				"backlight-control", NULL);
 		if (prop && strncmp(prop, type, strlen(type)) == 0)
 			return 1;
 	}
diff --git a/arch/powerpc/platforms/powermac/cpufreq_32.c b/arch/powerpc/platforms/powermac/cpufreq_32.c
index 62926248bdb8..c2b6b4134f68 100644
--- a/arch/powerpc/platforms/powermac/cpufreq_32.c
+++ b/arch/powerpc/platforms/powermac/cpufreq_32.c
@@ -421,7 +421,7 @@ static int pmac_cpufreq_cpu_init(struct cpufreq_policy *policy)
 
 static u32 read_gpio(struct device_node *np)
 {
-	u32 *reg = (u32 *)get_property(np, "reg", NULL);
+	const u32 *reg = get_property(np, "reg", NULL);
 	u32 offset;
 
 	if (reg == NULL)
@@ -497,7 +497,7 @@ static int pmac_cpufreq_init_MacRISC3(struct device_node *cpunode)
 								"frequency-gpio");
 	struct device_node *slew_done_gpio_np = of_find_node_by_name(NULL,
 								     "slewing-done");
-	u32 *value;
+	const u32 *value;
 
 	/*
 	 * Check to see if it's GPIO driven or PMU only
@@ -519,15 +519,15 @@ static int pmac_cpufreq_init_MacRISC3(struct device_node *cpunode)
 	 */
 	if (frequency_gpio && slew_done_gpio) {
 		int lenp, rc;
-		u32 *freqs, *ratio;
+		const u32 *freqs, *ratio;
 
-		freqs = (u32 *)get_property(cpunode, "bus-frequencies", &lenp);
+		freqs = get_property(cpunode, "bus-frequencies", &lenp);
 		lenp /= sizeof(u32);
 		if (freqs == NULL || lenp != 2) {
 			printk(KERN_ERR "cpufreq: bus-frequencies incorrect or missing\n");
 			return 1;
 		}
-		ratio = (u32 *)get_property(cpunode, "processor-to-bus-ratio*2", NULL);
+		ratio = get_property(cpunode, "processor-to-bus-ratio*2", NULL);
 		if (ratio == NULL) {
 			printk(KERN_ERR "cpufreq: processor-to-bus-ratio*2 missing\n");
 			return 1;
@@ -562,7 +562,7 @@ static int pmac_cpufreq_init_MacRISC3(struct device_node *cpunode)
 	/* If we use the PMU, look for the min & max frequencies in the
 	 * device-tree
 	 */
-	value = (u32 *)get_property(cpunode, "min-clock-frequency", NULL);
+	value = get_property(cpunode, "min-clock-frequency", NULL);
 	if (!value)
 		return 1;
 	low_freq = (*value) / 1000;
@@ -571,7 +571,7 @@ static int pmac_cpufreq_init_MacRISC3(struct device_node *cpunode)
 	if (low_freq < 100000)
 		low_freq *= 10;
 
-	value = (u32 *)get_property(cpunode, "max-clock-frequency", NULL);
+	value = get_property(cpunode, "max-clock-frequency", NULL);
 	if (!value)
 		return 1;
 	hi_freq = (*value) / 1000;
@@ -611,13 +611,14 @@ static int pmac_cpufreq_init_7447A(struct device_node *cpunode)
 static int pmac_cpufreq_init_750FX(struct device_node *cpunode)
 {
 	struct device_node *volt_gpio_np;
-	u32 pvr, *value;
+	u32 pvr;
+	const u32 *value;
 
 	if (get_property(cpunode, "dynamic-power-step", NULL) == NULL)
 		return 1;
 
 	hi_freq = cur_freq;
-	value = (u32 *)get_property(cpunode, "reduced-clock-frequency", NULL);
+	value = get_property(cpunode, "reduced-clock-frequency", NULL);
 	if (!value)
 		return 1;
 	low_freq = (*value) / 1000;
@@ -650,7 +651,7 @@ static int pmac_cpufreq_init_750FX(struct device_node *cpunode)
 static int __init pmac_cpufreq_setup(void)
 {
 	struct device_node	*cpunode;
-	u32			*value;
+	const u32		*value;
 
 	if (strstr(cmd_line, "nocpufreq"))
 		return 0;
@@ -661,7 +662,7 @@ static int __init pmac_cpufreq_setup(void)
 		goto out;
 
 	/* Get current cpu clock freq */
-	value = (u32 *)get_property(cpunode, "clock-frequency", NULL);
+	value = get_property(cpunode, "clock-frequency", NULL);
 	if (!value)
 		goto out;
 	cur_freq = (*value) / 1000;
diff --git a/arch/powerpc/platforms/powermac/cpufreq_64.c b/arch/powerpc/platforms/powermac/cpufreq_64.c
index a6a84ac5433e..c364c89adb4e 100644
--- a/arch/powerpc/platforms/powermac/cpufreq_64.c
+++ b/arch/powerpc/platforms/powermac/cpufreq_64.c
@@ -89,7 +89,7 @@ static DEFINE_MUTEX(g5_switch_mutex);
 
 #ifdef CONFIG_PPC_SMU
 
-static u32 *g5_pmode_data;
+static const u32 *g5_pmode_data;
 static int g5_pmode_max;
 
 static struct smu_sdbp_fvt *g5_fvt_table;	/* table of op. points */
@@ -391,7 +391,8 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 	unsigned int psize, ssize;
 	unsigned long max_freq;
 	char *freq_method, *volt_method;
-	u32 *valp, pvr_hi;
+	const u32 *valp;
+	u32 pvr_hi;
 	int use_volts_vdnap = 0;
 	int use_volts_smu = 0;
 	int rc = -ENODEV;
@@ -409,8 +410,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 	/* Get first CPU node */
 	for (cpunode = NULL;
 	     (cpunode = of_get_next_child(cpus, cpunode)) != NULL;) {
-		u32 *reg =
-			(u32 *)get_property(cpunode, "reg", NULL);
+		const u32 *reg = get_property(cpunode, "reg", NULL);
 		if (reg == NULL || (*reg) != 0)
 			continue;
 		if (!strcmp(cpunode->type, "cpu"))
@@ -422,7 +422,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 	}
 
 	/* Check 970FX for now */
-	valp = (u32 *)get_property(cpunode, "cpu-version", NULL);
+	valp = get_property(cpunode, "cpu-version", NULL);
 	if (!valp) {
 		DBG("No cpu-version property !\n");
 		goto bail_noprops;
@@ -434,7 +434,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 	}
 
 	/* Look for the powertune data in the device-tree */
-	g5_pmode_data = (u32 *)get_property(cpunode, "power-mode-data",&psize);
+	g5_pmode_data = get_property(cpunode, "power-mode-data",&psize);
 	if (!g5_pmode_data) {
 		DBG("No power-mode-data !\n");
 		goto bail_noprops;
@@ -442,7 +442,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 	g5_pmode_max = psize / sizeof(u32) - 1;
 
 	if (use_volts_smu) {
-		struct smu_sdbp_header *shdr;
+		const struct smu_sdbp_header *shdr;
 
 		/* Look for the FVT table */
 		shdr = smu_get_sdb_partition(SMU_SDB_FVT_ID, NULL);
@@ -493,7 +493,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 	 * half freq in this version. So far, I haven't yet seen a machine
 	 * supporting anything else.
 	 */
-	valp = (u32 *)get_property(cpunode, "clock-frequency", NULL);
+	valp = get_property(cpunode, "clock-frequency", NULL);
 	if (!valp)
 		return -ENODEV;
 	max_freq = (*valp)/1000;
@@ -541,8 +541,8 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpus)
 static int __init g5_pm72_cpufreq_init(struct device_node *cpus)
 {
 	struct device_node *cpuid = NULL, *hwclock = NULL, *cpunode = NULL;
-	u8 *eeprom = NULL;
-	u32 *valp;
+	const u8 *eeprom = NULL;
+	const u32 *valp;
 	u64 max_freq, min_freq, ih, il;
 	int has_volt = 1, rc = 0;
 
@@ -563,7 +563,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus)
 	/* Lookup the cpuid eeprom node */
         cpuid = of_find_node_by_path("/u3@0,f8000000/i2c@f8001000/cpuid@a0");
 	if (cpuid != NULL)
-		eeprom = (u8 *)get_property(cpuid, "cpuid", NULL);
+		eeprom = get_property(cpuid, "cpuid", NULL);
 	if (eeprom == NULL) {
 		printk(KERN_ERR "cpufreq: Can't find cpuid EEPROM !\n");
 		rc = -ENODEV;
@@ -573,7 +573,8 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus)
 	/* Lookup the i2c hwclock */
 	for (hwclock = NULL;
 	     (hwclock = of_find_node_by_name(hwclock, "i2c-hwclock")) != NULL;){
-		char *loc = get_property(hwclock, "hwctrl-location", NULL);
+		const char *loc = get_property(hwclock,
+				"hwctrl-location", NULL);
 		if (loc == NULL)
 			continue;
 		if (strcmp(loc, "CPU CLOCK"))
@@ -637,7 +638,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpus)
 	 */
 
 	/* Get max frequency from device-tree */
-	valp = (u32 *)get_property(cpunode, "clock-frequency", NULL);
+	valp = get_property(cpunode, "clock-frequency", NULL);
 	if (!valp) {
 		printk(KERN_ERR "cpufreq: Can't find CPU frequency !\n");
 		rc = -ENODEV;
diff --git a/arch/powerpc/platforms/powermac/feature.c b/arch/powerpc/platforms/powermac/feature.c
index f8313bf9a9f7..13fcaf5b1796 100644
--- a/arch/powerpc/platforms/powermac/feature.c
+++ b/arch/powerpc/platforms/powermac/feature.c
@@ -1058,8 +1058,8 @@ core99_reset_cpu(struct device_node *node, long param, long value)
 	if (np == NULL)
 		return -ENODEV;
 	for (np = np->child; np != NULL; np = np->sibling) {
-		u32 *num = (u32 *)get_property(np, "reg", NULL);
-		u32 *rst = (u32 *)get_property(np, "soft-reset", NULL);
+		u32 *num = get_property(np, "reg", NULL);
+		u32 *rst = get_property(np, "soft-reset", NULL);
 		if (num == NULL || rst == NULL)
 			continue;
 		if (param == *num) {
@@ -1087,7 +1087,7 @@ core99_usb_enable(struct device_node *node, long param, long value)
 {
 	struct macio_chip *macio;
 	unsigned long flags;
-	char *prop;
+	const char *prop;
 	int number;
 	u32 reg;
 
@@ -1096,7 +1096,7 @@ core99_usb_enable(struct device_node *node, long param, long value)
 	    macio->type != macio_intrepid)
 		return -ENODEV;
 
-	prop = (char *)get_property(node, "AAPL,clock-id", NULL);
+	prop = get_property(node, "AAPL,clock-id", NULL);
 	if (!prop)
 		return -ENODEV;
 	if (strncmp(prop, "usb0u048", 8) == 0)
@@ -1507,8 +1507,8 @@ static long g5_reset_cpu(struct device_node *node, long param, long value)
 	if (np == NULL)
 		return -ENODEV;
 	for (np = np->child; np != NULL; np = np->sibling) {
-		u32 *num = (u32 *)get_property(np, "reg", NULL);
-		u32 *rst = (u32 *)get_property(np, "soft-reset", NULL);
+		const u32 *num = get_property(np, "reg", NULL);
+		const u32 *rst = get_property(np, "soft-reset", NULL);
 		if (num == NULL || rst == NULL)
 			continue;
 		if (param == *num) {
@@ -2408,7 +2408,7 @@ static int __init probe_motherboard(void)
 	 */
 	dt = find_devices("device-tree");
 	if (dt != NULL)
-		model = (const char *) get_property(dt, "model", NULL);
+		model = get_property(dt, "model", NULL);
 	for(i=0; model && i<(sizeof(pmac_mb_defs)/sizeof(struct pmac_mb_def)); i++) {
 	    if (strcmp(model, pmac_mb_defs[i].model_string) == 0) {
 		pmac_mb = pmac_mb_defs[i];
@@ -2536,7 +2536,7 @@ found:
  */
 static void __init probe_uninorth(void)
 {
-	u32 *addrp;
+	const u32 *addrp;
 	phys_addr_t address;
 	unsigned long actrl;
 
@@ -2555,7 +2555,7 @@ static void __init probe_uninorth(void)
 	if (uninorth_node == NULL)
 		return;
 
-	addrp = (u32 *)get_property(uninorth_node, "reg", NULL);
+	addrp = get_property(uninorth_node, "reg", NULL);
 	if (addrp == NULL)
 		return;
 	address = of_translate_address(uninorth_node, addrp);
@@ -2596,7 +2596,7 @@ static void __init probe_one_macio(const char *name, const char *compat, int typ
 	struct device_node*	node;
 	int			i;
 	volatile u32 __iomem	*base;
-	u32			*addrp, *revp;
+	const u32		*addrp, *revp;
 	phys_addr_t		addr;
 	u64			size;
 
@@ -2639,7 +2639,7 @@ static void __init probe_one_macio(const char *name, const char *compat, int typ
 		return;
 	}
 	if (type == macio_keylargo || type == macio_keylargo2) {
-		u32 *did = (u32 *)get_property(node, "device-id", NULL);
+		const u32 *did = get_property(node, "device-id", NULL);
 		if (*did == 0x00000025)
 			type = macio_pangea;
 		if (*did == 0x0000003e)
@@ -2652,7 +2652,7 @@ static void __init probe_one_macio(const char *name, const char *compat, int typ
 	macio_chips[i].base	= base;
 	macio_chips[i].flags	= MACIO_FLAG_SCCB_ON | MACIO_FLAG_SCCB_ON;
 	macio_chips[i].name	= macio_names[type];
-	revp = (u32 *)get_property(node, "revision-id", NULL);
+	revp = get_property(node, "revision-id", NULL);
 	if (revp)
 		macio_chips[i].rev = *revp;
 	printk(KERN_INFO "Found a %s mac-io controller, rev: %d, mapped at 0x%p\n",
@@ -2695,15 +2695,15 @@ static void __init
 initial_serial_shutdown(struct device_node *np)
 {
 	int len;
-	struct slot_names_prop {
+	const struct slot_names_prop {
 		int	count;
 		char	name[1];
 	} *slots;
-	char *conn;
+	const char *conn;
 	int port_type = PMAC_SCC_ASYNC;
 	int modem = 0;
 
-	slots = (struct slot_names_prop *)get_property(np, "slot-names", &len);
+	slots = get_property(np, "slot-names", &len);
 	conn = get_property(np, "AAPL,connector", &len);
 	if (conn && (strcmp(conn, "infrared") == 0))
 		port_type = PMAC_SCC_IRDA;
diff --git a/arch/powerpc/platforms/powermac/low_i2c.c b/arch/powerpc/platforms/powermac/low_i2c.c
index 8677f50c2586..c2c7cf75dd5f 100644
--- a/arch/powerpc/platforms/powermac/low_i2c.c
+++ b/arch/powerpc/platforms/powermac/low_i2c.c
@@ -477,7 +477,8 @@ static int kw_i2c_xfer(struct pmac_i2c_bus *bus, u8 addrdir, int subsize,
 static struct pmac_i2c_host_kw *__init kw_i2c_host_init(struct device_node *np)
 {
 	struct pmac_i2c_host_kw *host;
-	u32			*psteps, *prate, *addrp, steps;
+	const u32		*psteps, *prate, *addrp;
+	u32			steps;
 
 	host = kzalloc(sizeof(struct pmac_i2c_host_kw), GFP_KERNEL);
 	if (host == NULL) {
@@ -490,7 +491,7 @@ static struct pmac_i2c_host_kw *__init kw_i2c_host_init(struct device_node *np)
 	 * on all i2c keywest nodes so far ... we would have to fallback
 	 * to macio parsing if that wasn't the case
 	 */
-	addrp = (u32 *)get_property(np, "AAPL,address", NULL);
+	addrp = get_property(np, "AAPL,address", NULL);
 	if (addrp == NULL) {
 		printk(KERN_ERR "low_i2c: Can't find address for %s\n",
 		       np->full_name);
@@ -504,13 +505,13 @@ static struct pmac_i2c_host_kw *__init kw_i2c_host_init(struct device_node *np)
 	host->timeout_timer.function = kw_i2c_timeout;
 	host->timeout_timer.data = (unsigned long)host;
 
-	psteps = (u32 *)get_property(np, "AAPL,address-step", NULL);
+	psteps = get_property(np, "AAPL,address-step", NULL);
 	steps = psteps ? (*psteps) : 0x10;
 	for (host->bsteps = 0; (steps & 0x01) == 0; host->bsteps++)
 		steps >>= 1;
 	/* Select interface rate */
 	host->speed = KW_I2C_MODE_25KHZ;
-	prate = (u32 *)get_property(np, "AAPL,i2c-rate", NULL);
+	prate = get_property(np, "AAPL,i2c-rate", NULL);
 	if (prate) switch(*prate) {
 	case 100:
 		host->speed = KW_I2C_MODE_100KHZ;
@@ -618,8 +619,8 @@ static void __init kw_i2c_probe(void)
 		} else {
 			for (child = NULL;
 			     (child = of_get_next_child(np, child)) != NULL;) {
-				u32 *reg =
-					(u32 *)get_property(child, "reg", NULL);
+				const u32 *reg = get_property(child,
+						"reg", NULL);
 				if (reg == NULL)
 					continue;
 				kw_i2c_add(host, np, child, *reg);
@@ -881,7 +882,7 @@ static void __init smu_i2c_probe(void)
 {
 	struct device_node *controller, *busnode;
 	struct pmac_i2c_bus *bus;
-	u32 *reg;
+	const u32 *reg;
 	int sz;
 
 	if (!smu_present())
@@ -904,7 +905,7 @@ static void __init smu_i2c_probe(void)
 		if (strcmp(busnode->type, "i2c") &&
 		    strcmp(busnode->type, "i2c-bus"))
 			continue;
-		reg = (u32 *)get_property(busnode, "reg", NULL);
+		reg = get_property(busnode, "reg", NULL);
 		if (reg == NULL)
 			continue;
 
@@ -948,9 +949,8 @@ struct pmac_i2c_bus *pmac_i2c_find_bus(struct device_node *node)
 		list_for_each_entry(bus, &pmac_i2c_busses, link) {
 			if (p == bus->busnode) {
 				if (prev && bus->flags & pmac_i2c_multibus) {
-					u32 *reg;
-					reg = (u32 *)get_property(prev, "reg",
-								  NULL);
+					const u32 *reg;
+					reg = get_property(prev, "reg", NULL);
 					if (!reg)
 						continue;
 					if (((*reg) >> 8) != bus->channel)
@@ -971,7 +971,7 @@ EXPORT_SYMBOL_GPL(pmac_i2c_find_bus);
 
 u8 pmac_i2c_get_dev_addr(struct device_node *device)
 {
-	u32 *reg = (u32 *)get_property(device, "reg", NULL);
+	const u32 *reg = get_property(device, "reg", NULL);
 
 	if (reg == NULL)
 		return 0;
diff --git a/arch/powerpc/platforms/powermac/pci.c b/arch/powerpc/platforms/powermac/pci.c
index 556b349797e8..787ffd999bc2 100644
--- a/arch/powerpc/platforms/powermac/pci.c
+++ b/arch/powerpc/platforms/powermac/pci.c
@@ -69,16 +69,16 @@ struct device_node *k2_skiplist[2];
 static int __init fixup_one_level_bus_range(struct device_node *node, int higher)
 {
 	for (; node != 0;node = node->sibling) {
-		int * bus_range;
-		unsigned int *class_code;
+		const int * bus_range;
+		const unsigned int *class_code;
 		int len;
 
 		/* For PCI<->PCI bridges or CardBus bridges, we go down */
-		class_code = (unsigned int *) get_property(node, "class-code", NULL);
+		class_code = get_property(node, "class-code", NULL);
 		if (!class_code || ((*class_code >> 8) != PCI_CLASS_BRIDGE_PCI &&
 			(*class_code >> 8) != PCI_CLASS_BRIDGE_CARDBUS))
 			continue;
-		bus_range = (int *) get_property(node, "bus-range", &len);
+		bus_range = get_property(node, "bus-range", &len);
 		if (bus_range != NULL && len > 2 * sizeof(int)) {
 			if (bus_range[1] > higher)
 				higher = bus_range[1];
@@ -96,13 +96,15 @@ static int __init fixup_one_level_bus_range(struct device_node *node, int higher
  */
 static void __init fixup_bus_range(struct device_node *bridge)
 {
-	int * bus_range;
-	int len;
+	int *bus_range, len;
+	struct property *prop;
 
 	/* Lookup the "bus-range" property for the hose */
-	bus_range = (int *) get_property(bridge, "bus-range", &len);
-	if (bus_range == NULL || len < 2 * sizeof(int))
+	prop = of_find_property(bridge, "bus-range", &len);
+	if (prop == NULL || prop->length < 2 * sizeof(int))
 		return;
+
+	bus_range = (int *)prop->value;
 	bus_range[1] = fixup_one_level_bus_range(bridge->child, bus_range[1]);
 }
 
@@ -240,7 +242,7 @@ static struct pci_ops macrisc_pci_ops =
 static int chaos_validate_dev(struct pci_bus *bus, int devfn, int offset)
 {
 	struct device_node *np;
-	u32 *vendor, *device;
+	const u32 *vendor, *device;
 
 	if (offset >= 0x100)
 		return  PCIBIOS_BAD_REGISTER_NUMBER;
@@ -248,8 +250,8 @@ static int chaos_validate_dev(struct pci_bus *bus, int devfn, int offset)
 	if (np == NULL)
 		return PCIBIOS_DEVICE_NOT_FOUND;
 
-	vendor = (u32 *)get_property(np, "vendor-id", NULL);
-	device = (u32 *)get_property(np, "device-id", NULL);
+	vendor = get_property(np, "vendor-id", NULL);
+	device = get_property(np, "device-id", NULL);
 	if (vendor == NULL || device == NULL)
 		return PCIBIOS_DEVICE_NOT_FOUND;
 
@@ -689,20 +691,21 @@ static void __init fixup_nec_usb2(void)
 
 	for (nec = NULL; (nec = of_find_node_by_name(nec, "usb")) != NULL;) {
 		struct pci_controller *hose;
-		u32 data, *prop;
+		u32 data;
+		const u32 *prop;
 		u8 bus, devfn;
 
-		prop = (u32 *)get_property(nec, "vendor-id", NULL);
+		prop = get_property(nec, "vendor-id", NULL);
 		if (prop == NULL)
 			continue;
 		if (0x1033 != *prop)
 			continue;
-		prop = (u32 *)get_property(nec, "device-id", NULL);
+		prop = get_property(nec, "device-id", NULL);
 		if (prop == NULL)
 			continue;
 		if (0x0035 != *prop)
 			continue;
-		prop = (u32 *)get_property(nec, "reg", NULL);
+		prop = get_property(nec, "reg", NULL);
 		if (prop == NULL)
 			continue;
 		devfn = (prop[0] >> 8) & 0xff;
@@ -901,7 +904,7 @@ static int __init add_bridge(struct device_node *dev)
 	struct pci_controller *hose;
 	struct resource rsrc;
 	char *disp_name;
-	int *bus_range;
+	const int *bus_range;
 	int primary = 1, has_address = 0;
 
 	DBG("Adding PCI host bridge %s\n", dev->full_name);
@@ -910,7 +913,7 @@ static int __init add_bridge(struct device_node *dev)
 	has_address = (of_address_to_resource(dev, 0, &rsrc) == 0);
 
 	/* Get bus range if any */
-	bus_range = (int *) get_property(dev, "bus-range", &len);
+	bus_range = get_property(dev, "bus-range", &len);
 	if (bus_range == NULL || len < 2 * sizeof(int)) {
 		printk(KERN_WARNING "Can't get bus-range for %s, assume"
 		       " bus 0\n", dev->full_name);
diff --git a/arch/powerpc/platforms/powermac/pfunc_base.c b/arch/powerpc/platforms/powermac/pfunc_base.c
index 6d66359ec8c8..829dacec96e5 100644
--- a/arch/powerpc/platforms/powermac/pfunc_base.c
+++ b/arch/powerpc/platforms/powermac/pfunc_base.c
@@ -114,7 +114,7 @@ static void macio_gpio_init_one(struct macio_chip *macio)
 	 * we just create them all
 	 */
 	for (gp = NULL; (gp = of_get_next_child(gparent, gp)) != NULL;) {
-		u32 *reg = (u32 *)get_property(gp, "reg", NULL);
+		const u32 *reg = get_property(gp, "reg", NULL);
 		unsigned long offset;
 		if (reg == NULL)
 			continue;
diff --git a/arch/powerpc/platforms/powermac/pfunc_core.c b/arch/powerpc/platforms/powermac/pfunc_core.c
index b117adbf9571..7651f278615a 100644
--- a/arch/powerpc/platforms/powermac/pfunc_core.c
+++ b/arch/powerpc/platforms/powermac/pfunc_core.c
@@ -813,14 +813,15 @@ struct pmf_function *__pmf_find_function(struct device_node *target,
 	struct pmf_device *dev;
 	struct pmf_function *func, *result = NULL;
 	char fname[64];
-	u32 *prop, ph;
+	const u32 *prop;
+	u32 ph;
 
 	/*
 	 * Look for a "platform-*" function reference. If we can't find
 	 * one, then we fallback to a direct call attempt
 	 */
 	snprintf(fname, 63, "platform-%s", name);
-	prop = (u32 *)get_property(target, fname, NULL);
+	prop = get_property(target, fname, NULL);
 	if (prop == NULL)
 		goto find_it;
 	ph = *prop;
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 31a9da769fa2..824a618396ab 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -116,7 +116,7 @@ extern struct smp_ops_t core99_smp_ops;
 static void pmac_show_cpuinfo(struct seq_file *m)
 {
 	struct device_node *np;
-	char *pp;
+	const char *pp;
 	int plen;
 	int mbmodel;
 	unsigned int mbflags;
@@ -134,12 +134,12 @@ static void pmac_show_cpuinfo(struct seq_file *m)
 	seq_printf(m, "machine\t\t: ");
 	np = of_find_node_by_path("/");
 	if (np != NULL) {
-		pp = (char *) get_property(np, "model", NULL);
+		pp = get_property(np, "model", NULL);
 		if (pp != NULL)
 			seq_printf(m, "%s\n", pp);
 		else
 			seq_printf(m, "PowerMac\n");
-		pp = (char *) get_property(np, "compatible", &plen);
+		pp = get_property(np, "compatible", &plen);
 		if (pp != NULL) {
 			seq_printf(m, "motherboard\t:");
 			while (plen > 0) {
@@ -163,10 +163,8 @@ static void pmac_show_cpuinfo(struct seq_file *m)
 	if (np == NULL)
 		np = of_find_node_by_type(NULL, "cache");
 	if (np != NULL) {
-		unsigned int *ic = (unsigned int *)
-			get_property(np, "i-cache-size", NULL);
-		unsigned int *dc = (unsigned int *)
-			get_property(np, "d-cache-size", NULL);
+		const unsigned int *ic = get_property(np, "i-cache-size", NULL);
+		const unsigned int *dc = get_property(np, "d-cache-size", NULL);
 		seq_printf(m, "L2 cache\t:");
 		has_l2cache = 1;
 		if (get_property(np, "cache-unified", NULL) != 0 && dc) {
@@ -254,7 +252,7 @@ static void __init l2cr_init(void)
 		if (np == 0)
 			np = find_type_devices("cpu");
 		if (np != 0) {
-			unsigned int *l2cr = (unsigned int *)
+			const unsigned int *l2cr =
 				get_property(np, "l2cr-value", NULL);
 			if (l2cr != 0) {
 				ppc_override_l2cr = 1;
@@ -277,7 +275,7 @@ static void __init l2cr_init(void)
 static void __init pmac_setup_arch(void)
 {
 	struct device_node *cpu, *ic;
-	int *fp;
+	const int *fp;
 	unsigned long pvr;
 
 	pvr = PVR_VER(mfspr(SPRN_PVR));
@@ -287,7 +285,7 @@ static void __init pmac_setup_arch(void)
 	loops_per_jiffy = 50000000 / HZ;
 	cpu = of_find_node_by_type(NULL, "cpu");
 	if (cpu != NULL) {
-		fp = (int *) get_property(cpu, "clock-frequency", NULL);
+		fp = get_property(cpu, "clock-frequency", NULL);
 		if (fp != NULL) {
 			if (pvr >= 0x30 && pvr < 0x80)
 				/* PPC970 etc. */
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index 827b7121ffb8..653eeb64d1e2 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -548,7 +548,7 @@ static void __init smp_core99_setup_i2c_hwsync(int ncpus)
 	struct device_node *cc = NULL;	
 	struct device_node *p;
 	const char *name = NULL;
-	u32 *reg;
+	const u32 *reg;
 	int ok;
 
 	/* Look for the clock chip */
@@ -562,7 +562,7 @@ static void __init smp_core99_setup_i2c_hwsync(int ncpus)
 		pmac_tb_clock_chip_host = pmac_i2c_find_bus(cc);
 		if (pmac_tb_clock_chip_host == NULL)
 			continue;
-		reg = (u32 *)get_property(cc, "reg", NULL);
+		reg = get_property(cc, "reg", NULL);
 		if (reg == NULL)
 			continue;
 		switch (*reg) {
@@ -707,8 +707,7 @@ static void __init smp_core99_setup(int ncpus)
 		core99_tb_gpio = KL_GPIO_TB_ENABLE;	/* default value */
 		cpu = of_find_node_by_type(NULL, "cpu");
 		if (cpu != NULL) {
-			tbprop = (u32 *)get_property(cpu, "timebase-enable",
-						     NULL);
+			tbprop = get_property(cpu, "timebase-enable", NULL);
 			if (tbprop)
 				core99_tb_gpio = *tbprop;
 			of_node_put(cpu);
diff --git a/arch/powerpc/platforms/powermac/udbg_scc.c b/arch/powerpc/platforms/powermac/udbg_scc.c
index 37e5b1eff911..ce1a235855f7 100644
--- a/arch/powerpc/platforms/powermac/udbg_scc.c
+++ b/arch/powerpc/platforms/powermac/udbg_scc.c
@@ -68,11 +68,11 @@ static unsigned char scc_inittab[] = {
 
 void udbg_scc_init(int force_scc)
 {
-	u32 *reg;
+	const u32 *reg;
 	unsigned long addr;
 	struct device_node *stdout = NULL, *escc = NULL, *macio = NULL;
 	struct device_node *ch, *ch_def = NULL, *ch_a = NULL;
-	char *path;
+	const char *path;
 	int i, x;
 
 	escc = of_find_node_by_name(NULL, "escc");
@@ -81,7 +81,7 @@ void udbg_scc_init(int force_scc)
 	macio = of_get_parent(escc);
 	if (macio == NULL)
 		goto bail;
-	path = (char *)get_property(of_chosen, "linux,stdout-path", NULL);
+	path = get_property(of_chosen, "linux,stdout-path", NULL);
 	if (path != NULL)
 		stdout = of_find_node_by_path(path);
 	for (ch = NULL; (ch = of_get_next_child(escc, ch)) != NULL;) {
@@ -96,13 +96,13 @@ void udbg_scc_init(int force_scc)
 	ch = ch_def ? ch_def : ch_a;
 
 	/* Get address within mac-io ASIC */
-	reg = (u32 *)get_property(escc, "reg", NULL);
+	reg = get_property(escc, "reg", NULL);
 	if (reg == NULL)
 		goto bail;
 	addr = reg[0];
 
 	/* Get address of mac-io PCI itself */
-	reg = (u32 *)get_property(macio, "assigned-addresses", NULL);
+	reg = get_property(macio, "assigned-addresses", NULL);
 	if (reg == NULL)
 		goto bail;
 	addr += reg[2];
diff --git a/drivers/i2c/busses/i2c-powermac.c b/drivers/i2c/busses/i2c-powermac.c
index 2a0b3be7cdd0..e8a6de5a1517 100644
--- a/drivers/i2c/busses/i2c-powermac.c
+++ b/drivers/i2c/busses/i2c-powermac.c
@@ -209,7 +209,8 @@ static int i2c_powermac_probe(struct device *dev)
 	struct pmac_i2c_bus *bus = dev->platform_data;
 	struct device_node *parent = NULL;
 	struct i2c_adapter *adapter;
-	char name[32], *basename;
+	char name[32];
+	const char *basename;
 	int rc;
 
 	if (bus == NULL)
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index ebf961f1718d..fa46856e8068 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1154,7 +1154,7 @@ static int
 pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 {
 	struct device_node *np = pmif->node;
-	int *bidp;
+	const int *bidp;
 
 	pmif->cable_80 = 0;
 	pmif->broken_dma = pmif->broken_dma_warn = 0;
@@ -1176,14 +1176,14 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 		pmif->broken_dma = 1;
 	}
 
-	bidp = (int *)get_property(np, "AAPL,bus-id", NULL);
+	bidp = get_property(np, "AAPL,bus-id", NULL);
 	pmif->aapl_bus_id =  bidp ? *bidp : 0;
 
 	/* Get cable type from device-tree */
 	if (pmif->kind == controller_kl_ata4 || pmif->kind == controller_un_ata6
 	    || pmif->kind == controller_k2_ata6
 	    || pmif->kind == controller_sh_ata6) {
-		char* cable = get_property(np, "cable-type", NULL);
+		const char* cable = get_property(np, "cable-type", NULL);
 		if (cable && !strncmp(cable, "80-", 3))
 			pmif->cable_80 = 1;
 	}
diff --git a/drivers/macintosh/macio_asic.c b/drivers/macintosh/macio_asic.c
index 80c0c665b5f6..7817cf286d0c 100644
--- a/drivers/macintosh/macio_asic.c
+++ b/drivers/macintosh/macio_asic.c
@@ -139,7 +139,9 @@ static int macio_uevent(struct device *dev, char **envp, int num_envp,
 {
 	struct macio_dev * macio_dev;
 	struct of_device * of;
-	char *scratch, *compat, *compat2;
+	char *scratch;
+	const char *compat, *compat2;
+
 	int i = 0;
 	int length, cplen, cplen2, seen = 0;
 
@@ -173,7 +175,7 @@ static int macio_uevent(struct device *dev, char **envp, int num_envp,
          * it's not really legal to split it out with commas. We split it
          * up using a number of environment variables instead. */
 
-	compat = (char *) get_property(of->node, "compatible", &cplen);
+	compat = get_property(of->node, "compatible", &cplen);
 	compat2 = compat;
 	cplen2= cplen;
 	while (compat && cplen > 0) {
@@ -454,7 +456,7 @@ static struct macio_dev * macio_add_one_device(struct macio_chip *chip,
 					       struct resource *parent_res)
 {
 	struct macio_dev *dev;
-	u32 *reg;
+	const u32 *reg;
 	
 	if (np == NULL)
 		return NULL;
@@ -489,7 +491,7 @@ static struct macio_dev * macio_add_one_device(struct macio_chip *chip,
 #endif
 			MAX_NODE_NAME_SIZE, np->name);
 	} else {
-		reg = (u32 *)get_property(np, "reg", NULL);
+		reg = get_property(np, "reg", NULL);
 		sprintf(dev->ofdev.dev.bus_id, "%1d.%08x:%.*s",
 			chip->lbus.index,
 			reg ? *reg : 0, MAX_NODE_NAME_SIZE, np->name);
diff --git a/drivers/macintosh/macio_sysfs.c b/drivers/macintosh/macio_sysfs.c
index cae24a13526a..8566bdfdd4b8 100644
--- a/drivers/macintosh/macio_sysfs.c
+++ b/drivers/macintosh/macio_sysfs.c
@@ -16,12 +16,12 @@ static ssize_t
 compatible_show (struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct of_device *of;
-	char *compat;
+	const char *compat;
 	int cplen;
 	int length = 0;
 
 	of = &to_macio_device (dev)->ofdev;
-	compat = (char *) get_property(of->node, "compatible", &cplen);
+	compat = get_property(of->node, "compatible", &cplen);
 	if (!compat) {
 		*buf = '\0';
 		return 0;
@@ -42,12 +42,12 @@ static ssize_t modalias_show (struct device *dev, struct device_attribute *attr,
 			      char *buf)
 {
 	struct of_device *of;
-	char *compat;
+	const char *compat;
 	int cplen;
 	int length;
 
 	of = &to_macio_device (dev)->ofdev;
-	compat = (char *) get_property (of->node, "compatible", &cplen);
+	compat = get_property(of->node, "compatible", &cplen);
 	if (!compat) compat = "", cplen = 1;
 	length = sprintf (buf, "of:N%sT%s", of->node->name, of->node->type);
 	buf += length;
diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c
index f139a74696fe..6f358600536e 100644
--- a/drivers/macintosh/smu.c
+++ b/drivers/macintosh/smu.c
@@ -447,7 +447,7 @@ EXPORT_SYMBOL(smu_present);
 int __init smu_init (void)
 {
 	struct device_node *np;
-	u32 *data;
+	const u32 *data;
 
         np = of_find_node_by_type(NULL, "smu");
         if (np == NULL)
@@ -483,7 +483,7 @@ int __init smu_init (void)
 		printk(KERN_ERR "SMU: Can't find doorbell GPIO !\n");
 		goto fail;
 	}
-	data = (u32 *)get_property(np, "reg", NULL);
+	data = get_property(np, "reg", NULL);
 	if (data == NULL) {
 		of_node_put(np);
 		printk(KERN_ERR "SMU: Can't find doorbell GPIO address !\n");
@@ -506,7 +506,7 @@ int __init smu_init (void)
 		np = of_find_node_by_name(NULL, "smu-interrupt");
 		if (np == NULL)
 			break;
-		data = (u32 *)get_property(np, "reg", NULL);
+		data = get_property(np, "reg", NULL);
 		if (data == NULL) {
 			of_node_put(np);
 			break;
@@ -959,11 +959,11 @@ static struct smu_sdbp_header *smu_create_sdb_partition(int id)
 /* Note: Only allowed to return error code in pointers (using ERR_PTR)
  * when interruptible is 1
  */
-struct smu_sdbp_header *__smu_get_sdb_partition(int id, unsigned int *size,
-						int interruptible)
+const struct smu_sdbp_header *__smu_get_sdb_partition(int id,
+		unsigned int *size, int interruptible)
 {
 	char pname[32];
-	struct smu_sdbp_header *part;
+	const struct smu_sdbp_header *part;
 
 	if (!smu)
 		return NULL;
@@ -980,8 +980,7 @@ struct smu_sdbp_header *__smu_get_sdb_partition(int id, unsigned int *size,
 	} else
 		mutex_lock(&smu_part_access);
 
-	part = (struct smu_sdbp_header *)get_property(smu->of_node,
-						      pname, size);
+	part = get_property(smu->of_node, pname, size);
 	if (part == NULL) {
 		DPRINTK("trying to extract from SMU ...\n");
 		part = smu_create_sdb_partition(id);
@@ -992,7 +991,7 @@ struct smu_sdbp_header *__smu_get_sdb_partition(int id, unsigned int *size,
 	return part;
 }
 
-struct smu_sdbp_header *smu_get_sdb_partition(int id, unsigned int *size)
+const struct smu_sdbp_header *smu_get_sdb_partition(int id, unsigned int *size)
 {
 	return __smu_get_sdb_partition(id, size, 0);
 }
@@ -1071,7 +1070,7 @@ static ssize_t smu_write(struct file *file, const char __user *buf,
 		pp->mode = smu_file_events;
 		return 0;
 	} else if (hdr.cmdtype == SMU_CMDTYPE_GET_PARTITION) {
-		struct smu_sdbp_header *part;
+		const struct smu_sdbp_header *part;
 		part = __smu_get_sdb_partition(hdr.cmd, NULL, 1);
 		if (part == NULL)
 			return -EINVAL;
diff --git a/drivers/macintosh/therm_adt746x.c b/drivers/macintosh/therm_adt746x.c
index 7f86478bdd36..a0f30d0853ea 100644
--- a/drivers/macintosh/therm_adt746x.c
+++ b/drivers/macintosh/therm_adt746x.c
@@ -47,7 +47,7 @@ static u8 FAN_SPD_SET[2] = {0x30, 0x31};
 
 static u8 default_limits_local[3] = {70, 50, 70};    /* local, sensor1, sensor2 */
 static u8 default_limits_chip[3] = {80, 65, 80};    /* local, sensor1, sensor2 */
-static char *sensor_location[3] = {NULL, NULL, NULL};
+static const char *sensor_location[3] = {NULL, NULL, NULL};
 
 static int limit_adjust = 0;
 static int fan_speed = -1;
@@ -553,7 +553,7 @@ static int __init
 thermostat_init(void)
 {
 	struct device_node* np;
-	u32 *prop;
+	const u32 *prop;
 	int i = 0, offset = 0;
 	
 	np = of_find_node_by_name(NULL, "fan");
@@ -566,13 +566,13 @@ thermostat_init(void)
 	else
 		return -ENODEV;
 
-	prop = (u32 *)get_property(np, "hwsensor-params-version", NULL);
+	prop = get_property(np, "hwsensor-params-version", NULL);
 	printk(KERN_INFO "adt746x: version %d (%ssupported)\n", *prop,
 			 (*prop == 1)?"":"un");
 	if (*prop != 1)
 		return -ENODEV;
 
-	prop = (u32 *)get_property(np, "reg", NULL);
+	prop = get_property(np, "reg", NULL);
 	if (!prop)
 		return -ENODEV;
 
diff --git a/drivers/macintosh/therm_pm72.c b/drivers/macintosh/therm_pm72.c
index 20bf67244e2c..d00c0c37a12e 100644
--- a/drivers/macintosh/therm_pm72.c
+++ b/drivers/macintosh/therm_pm72.c
@@ -660,7 +660,7 @@ static int read_eeprom(int cpu, struct mpu_data *out)
 {
 	struct device_node *np;
 	char nodename[64];
-	u8 *data;
+	const u8 *data;
 	int len;
 
 	/* prom.c routine for finding a node by path is a bit brain dead
@@ -673,7 +673,7 @@ static int read_eeprom(int cpu, struct mpu_data *out)
 		printk(KERN_ERR "therm_pm72: Failed to retrieve cpuid node from device-tree\n");
 		return -ENODEV;
 	}
-	data = (u8 *)get_property(np, "cpuid", &len);
+	data = get_property(np, "cpuid", &len);
 	if (data == NULL) {
 		printk(KERN_ERR "therm_pm72: Failed to retrieve cpuid property from device-tree\n");
 		of_node_put(np);
@@ -1336,7 +1336,7 @@ static int init_backside_state(struct backside_pid_state *state)
 	 */
 	u3 = of_find_node_by_path("/u3@0,f8000000");
 	if (u3 != NULL) {
-		u32 *vers = (u32 *)get_property(u3, "device-rev", NULL);
+		const u32 *vers = get_property(u3, "device-rev", NULL);
 		if (vers)
 			if (((*vers) & 0x3f) < 0x34)
 				u3h = 0;
@@ -2111,8 +2111,8 @@ static void fcu_lookup_fans(struct device_node *fcu_node)
 
 	while ((np = of_get_next_child(fcu_node, np)) != NULL) {
 		int type = -1;
-		char *loc;
-		u32 *reg;
+		const char *loc;
+		const u32 *reg;
 
 		DBG(" control: %s, type: %s\n", np->name, np->type);
 
@@ -2128,8 +2128,8 @@ static void fcu_lookup_fans(struct device_node *fcu_node)
 			continue;
 
 		/* Lookup for a matching location */
-		loc = (char *)get_property(np, "location", NULL);
-		reg = (u32 *)get_property(np, "reg", NULL);
+		loc = get_property(np, "location", NULL);
+		reg = get_property(np, "reg", NULL);
 		if (loc == NULL || reg == NULL)
 			continue;
 		DBG(" matching location: %s, reg: 0x%08x\n", loc, *reg);
diff --git a/drivers/macintosh/therm_windtunnel.c b/drivers/macintosh/therm_windtunnel.c
index c7d1c290cb0c..738faab1b22c 100644
--- a/drivers/macintosh/therm_windtunnel.c
+++ b/drivers/macintosh/therm_windtunnel.c
@@ -484,14 +484,14 @@ struct apple_thermal_info {
 static int __init
 g4fan_init( void )
 {
-	struct apple_thermal_info *info;
+	const struct apple_thermal_info *info;
 	struct device_node *np;
 
 	init_MUTEX( &x.lock );
 
 	if( !(np=of_find_node_by_name(NULL, "power-mgt")) )
 		return -ENODEV;
-	info = (struct apple_thermal_info*)get_property(np, "thermal-info", NULL);
+	info = get_property(np, "thermal-info", NULL);
 	of_node_put(np);
 
 	if( !info || !machine_is_compatible("PowerMac3,6") )
diff --git a/drivers/macintosh/via-cuda.c b/drivers/macintosh/via-cuda.c
index 69d5452fd22f..7512d1c15207 100644
--- a/drivers/macintosh/via-cuda.c
+++ b/drivers/macintosh/via-cuda.c
@@ -123,7 +123,7 @@ int __init find_via_cuda(void)
 {
     struct adb_request req;
     phys_addr_t taddr;
-    u32 *reg;
+    const u32 *reg;
     int err;
 
     if (vias != 0)
@@ -132,7 +132,7 @@ int __init find_via_cuda(void)
     if (vias == 0)
 	return 0;
 
-    reg = (u32 *)get_property(vias, "reg", NULL);
+    reg = get_property(vias, "reg", NULL);
     if (reg == NULL) {
 	    printk(KERN_ERR "via-cuda: No \"reg\" property !\n");
 	    goto fail;
diff --git a/drivers/macintosh/via-pmu-led.c b/drivers/macintosh/via-pmu-led.c
index 5189d5454b1f..179af10105d9 100644
--- a/drivers/macintosh/via-pmu-led.c
+++ b/drivers/macintosh/via-pmu-led.c
@@ -120,7 +120,7 @@ static int __init via_pmu_led_init(void)
 	dt = of_find_node_by_path("/");
 	if (dt == NULL)
 		return -ENODEV;
-	model = (const char *)get_property(dt, "model", NULL);
+	model = get_property(dt, "model", NULL);
 	if (model == NULL)
 		return -ENODEV;
 	if (strncmp(model, "PowerBook", strlen("PowerBook")) != 0 &&
diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c
index 06ca80bfd6b9..80e88b4a1d07 100644
--- a/drivers/macintosh/via-pmu.c
+++ b/drivers/macintosh/via-pmu.c
@@ -287,7 +287,7 @@ static char *pbook_type[] = {
 int __init find_via_pmu(void)
 {
 	u64 taddr;
-	u32 *reg;
+	const u32 *reg;
 
 	if (via != 0)
 		return 1;
@@ -295,7 +295,7 @@ int __init find_via_pmu(void)
 	if (vias == NULL)
 		return 0;
 
-	reg = (u32 *)get_property(vias, "reg", NULL);
+	reg = get_property(vias, "reg", NULL);
 	if (reg == NULL) {
 		printk(KERN_ERR "via-pmu: No \"reg\" property !\n");
 		goto fail;
@@ -337,7 +337,7 @@ int __init find_via_pmu(void)
 		
 		gpiop = of_find_node_by_name(NULL, "gpio");
 		if (gpiop) {
-			reg = (u32 *)get_property(gpiop, "reg", NULL);
+			reg = get_property(gpiop, "reg", NULL);
 			if (reg)
 				gaddr = of_translate_address(gpiop, reg);
 			if (gaddr != OF_BAD_ADDR)
@@ -486,9 +486,9 @@ static int __init via_pmu_dev_init(void)
 		pmu_batteries[1].flags |= PMU_BATT_TYPE_SMART;
 	} else {
 		struct device_node* prim = find_devices("power-mgt");
-		u32 *prim_info = NULL;
+		const u32 *prim_info = NULL;
 		if (prim)
-			prim_info = (u32 *)get_property(prim, "prim-info", NULL);
+			prim_info = get_property(prim, "prim-info", NULL);
 		if (prim_info) {
 			/* Other stuffs here yet unknown */
 			pmu_battery_count = (prim_info[6] >> 16) & 0xff;
diff --git a/drivers/macintosh/windfarm_pm81.c b/drivers/macintosh/windfarm_pm81.c
index f1df6efcbe68..2ff546e4c92f 100644
--- a/drivers/macintosh/windfarm_pm81.c
+++ b/drivers/macintosh/windfarm_pm81.c
@@ -396,7 +396,7 @@ static void wf_smu_sys_fans_tick(struct wf_smu_sys_fans_state *st)
 static void wf_smu_create_cpu_fans(void)
 {
 	struct wf_cpu_pid_param pid_param;
-	struct smu_sdbp_header *hdr;
+	const struct smu_sdbp_header *hdr;
 	struct smu_sdbp_cpupiddata *piddata;
 	struct smu_sdbp_fvt *fvt;
 	s32 tmax, tdelta, maxpow, powadj;
@@ -702,7 +702,7 @@ static struct notifier_block wf_smu_events = {
 
 static int wf_init_pm(void)
 {
-	struct smu_sdbp_header *hdr;
+	const struct smu_sdbp_header *hdr;
 
 	hdr = smu_get_sdb_partition(SMU_SDB_SENSORTREE_ID, NULL);
 	if (hdr != 0) {
diff --git a/drivers/macintosh/windfarm_pm91.c b/drivers/macintosh/windfarm_pm91.c
index 0d6372e96d32..59e9ffe37c39 100644
--- a/drivers/macintosh/windfarm_pm91.c
+++ b/drivers/macintosh/windfarm_pm91.c
@@ -144,7 +144,7 @@ static struct wf_smu_slots_fans_state *wf_smu_slots_fans;
 static void wf_smu_create_cpu_fans(void)
 {
 	struct wf_cpu_pid_param pid_param;
-	struct smu_sdbp_header *hdr;
+	const struct smu_sdbp_header *hdr;
 	struct smu_sdbp_cpupiddata *piddata;
 	struct smu_sdbp_fvt *fvt;
 	s32 tmax, tdelta, maxpow, powadj;
diff --git a/drivers/macintosh/windfarm_smu_controls.c b/drivers/macintosh/windfarm_smu_controls.c
index a9e88edc0c72..bff1f372f188 100644
--- a/drivers/macintosh/windfarm_smu_controls.c
+++ b/drivers/macintosh/windfarm_smu_controls.c
@@ -159,14 +159,15 @@ static struct smu_fan_control *smu_fan_create(struct device_node *node,
 					      int pwm_fan)
 {
 	struct smu_fan_control *fct;
-	s32 *v; u32 *reg;
-	char *l;
+	const s32 *v;
+	const u32 *reg;
+	const char *l;
 
 	fct = kmalloc(sizeof(struct smu_fan_control), GFP_KERNEL);
 	if (fct == NULL)
 		return NULL;
 	fct->ctrl.ops = &smu_fan_ops;
-	l = (char *)get_property(node, "location", NULL);
+	l = get_property(node, "location", NULL);
 	if (l == NULL)
 		goto fail;
 
@@ -223,17 +224,17 @@ static struct smu_fan_control *smu_fan_create(struct device_node *node,
 		goto fail;
 
 	/* Get min & max values*/
-	v = (s32 *)get_property(node, "min-value", NULL);
+	v = get_property(node, "min-value", NULL);
 	if (v == NULL)
 		goto fail;
 	fct->min = *v;
-	v = (s32 *)get_property(node, "max-value", NULL);
+	v = get_property(node, "max-value", NULL);
 	if (v == NULL)
 		goto fail;
 	fct->max = *v;
 
 	/* Get "reg" value */
-	reg = (u32 *)get_property(node, "reg", NULL);
+	reg = get_property(node, "reg", NULL);
 	if (reg == NULL)
 		goto fail;
 	fct->reg = *reg;
diff --git a/drivers/macintosh/windfarm_smu_sat.c b/drivers/macintosh/windfarm_smu_sat.c
index e295a07a1ebc..aceb61d9fbc8 100644
--- a/drivers/macintosh/windfarm_smu_sat.c
+++ b/drivers/macintosh/windfarm_smu_sat.c
@@ -233,15 +233,15 @@ static void wf_sat_create(struct i2c_adapter *adapter, struct device_node *dev)
 {
 	struct wf_sat *sat;
 	struct wf_sat_sensor *sens;
-	u32 *reg;
-	char *loc, *type;
+	const u32 *reg;
+	const char *loc, *type;
 	u8 addr, chip, core;
 	struct device_node *child;
 	int shift, cpu, index;
 	char *name;
 	int vsens[2], isens[2];
 
-	reg = (u32 *) get_property(dev, "reg", NULL);
+	reg = get_property(dev, "reg", NULL);
 	if (reg == NULL)
 		return;
 	addr = *reg;
@@ -268,7 +268,7 @@ static void wf_sat_create(struct i2c_adapter *adapter, struct device_node *dev)
 	isens[0] = isens[1] = -1;
 	child = NULL;
 	while ((child = of_get_next_child(dev, child)) != NULL) {
-		reg = (u32 *) get_property(child, "reg", NULL);
+		reg = get_property(child, "reg", NULL);
 		type = get_property(child, "device_type", NULL);
 		loc = get_property(child, "location", NULL);
 		if (reg == NULL || loc == NULL)
diff --git a/drivers/macintosh/windfarm_smu_sensors.c b/drivers/macintosh/windfarm_smu_sensors.c
index bed25dcf8a1e..defe9922ebd1 100644
--- a/drivers/macintosh/windfarm_smu_sensors.c
+++ b/drivers/macintosh/windfarm_smu_sensors.c
@@ -198,14 +198,14 @@ static struct wf_sensor_ops smu_slotspow_ops = {
 static struct smu_ad_sensor *smu_ads_create(struct device_node *node)
 {
 	struct smu_ad_sensor *ads;
-	char *c, *l;
-	u32 *v;
+	const char *c, *l;
+	const u32 *v;
 
 	ads = kmalloc(sizeof(struct smu_ad_sensor), GFP_KERNEL);
 	if (ads == NULL)
 		return NULL;
-	c = (char *)get_property(node, "device_type", NULL);
-	l = (char *)get_property(node, "location", NULL);
+	c = get_property(node, "device_type", NULL);
+	l = get_property(node, "location", NULL);
 	if (c == NULL || l == NULL)
 		goto fail;
 
@@ -255,7 +255,7 @@ static struct smu_ad_sensor *smu_ads_create(struct device_node *node)
 	} else
 		goto fail;
 
-	v = (u32 *)get_property(node, "reg", NULL);
+	v = get_property(node, "reg", NULL);
 	if (v == NULL)
 		goto fail;
 	ads->reg = *v;
@@ -382,7 +382,7 @@ smu_cpu_power_create(struct wf_sensor *volts, struct wf_sensor *amps)
 
 static void smu_fetch_param_partitions(void)
 {
-	struct smu_sdbp_header *hdr;
+	const struct smu_sdbp_header *hdr;
 
 	/* Get CPU voltage/current/power calibration data */
 	hdr = smu_get_sdb_partition(SMU_SDB_CPUVCP_ID, NULL);
diff --git a/drivers/serial/pmac_zilog.c b/drivers/serial/pmac_zilog.c
index bfd2a22759eb..a3b99caf80e6 100644
--- a/drivers/serial/pmac_zilog.c
+++ b/drivers/serial/pmac_zilog.c
@@ -1400,8 +1400,8 @@ static struct uart_ops pmz_pops = {
 static int __init pmz_init_port(struct uart_pmac_port *uap)
 {
 	struct device_node *np = uap->node;
-	char *conn;
-	struct slot_names_prop {
+	const char *conn;
+	const struct slot_names_prop {
 		int	count;
 		char	name[1];
 	} *slots;
@@ -1458,7 +1458,7 @@ no_dma:
 		uap->flags |= PMACZILOG_FLAG_IS_IRDA;
 	uap->port_type = PMAC_SCC_ASYNC;
 	/* 1999 Powerbook G3 has slot-names property instead */
-	slots = (struct slot_names_prop *)get_property(np, "slot-names", &len);
+	slots = get_property(np, "slot-names", &len);
 	if (slots && slots->count > 0) {
 		if (strcmp(slots->name, "IrDA") == 0)
 			uap->flags |= PMACZILOG_FLAG_IS_IRDA;
@@ -1470,7 +1470,8 @@ no_dma:
 	if (ZS_IS_INTMODEM(uap)) {
 		struct device_node* i2c_modem = find_devices("i2c-modem");
 		if (i2c_modem) {
-			char* mid = get_property(i2c_modem, "modem-id", NULL);
+			const char* mid =
+				get_property(i2c_modem, "modem-id", NULL);
 			if (mid) switch(*mid) {
 			case 0x04 :
 			case 0x05 :
diff --git a/include/asm-powerpc/smu.h b/include/asm-powerpc/smu.h
index 51e65fc46a03..e49f644ca63a 100644
--- a/include/asm-powerpc/smu.h
+++ b/include/asm-powerpc/smu.h
@@ -517,7 +517,7 @@ struct smu_sdbp_cpupiddata {
  * This returns the pointer to an SMU "sdb" partition data or NULL
  * if not found. The data format is described below
  */
-extern struct smu_sdbp_header *smu_get_sdb_partition(int id,
+extern const struct smu_sdbp_header *smu_get_sdb_partition(int id,
 					unsigned int *size);
 
 /* Get "sdb" partition data from an SMU satellite */
-- 
cgit v1.2.3


From 931b261f442e779b0656d9b04c7ffe4939ef8c0a Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@ozlabs.org>
Date: Wed, 12 Jul 2006 15:42:06 +1000
Subject: [POWERPC] Make get_property() return a const void *

Previous changes have treated the return values of get_property as
const, so now we can make the actual change to get_property(). There
shouldn't be a need to cast the return values anymore.

We will now get compiler warnings when property values are assigned to
a non-const variable.

If properties need to be updated, there's still the of_find_property
function.

Built for cell_defconfig, chrp32_defconfig, g5_defconfig,
iseries_defconfig, maple_defconfig, pmac32_defconfig, ppc64_defconfig
and pseries_defconfig.

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/prom.c | 2 +-
 include/asm-powerpc/prom.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 2a3d84a39cb5..bf2005b2feb6 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -1527,7 +1527,7 @@ struct property *of_find_property(struct device_node *np, const char *name,
  * Find a property with a given name for a given node
  * and return the value.
  */
-void *get_property(struct device_node *np, const char *name, int *lenp)
+const void *get_property(struct device_node *np, const char *name, int *lenp)
 {
 	struct property *pp = of_find_property(np,name,lenp);
 	return pp ? pp->value : NULL;
diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h
index abdf1be66e97..31bfea4686a6 100644
--- a/include/asm-powerpc/prom.h
+++ b/include/asm-powerpc/prom.h
@@ -160,7 +160,7 @@ extern void unflatten_device_tree(void);
 extern void early_init_devtree(void *);
 extern int device_is_compatible(struct device_node *device, const char *);
 extern int machine_is_compatible(const char *compat);
-extern void *get_property(struct device_node *node, const char *name,
+extern const void *get_property(struct device_node *node, const char *name,
 		int *lenp);
 extern void print_properties(struct device_node *node);
 extern int prom_n_addr_cells(struct device_node* np);
-- 
cgit v1.2.3


From cb18bd40030c879cd93fef02fd579f74dbab473d Mon Sep 17 00:00:00 2001
From: Mike Kravetz <kravetz@us.ibm.com>
Date: Thu, 20 Jul 2006 23:39:51 -0700
Subject: [POWERPC] Instrument Hypervisor Calls: merge headers

Move all the Hypervisor call definitions to to a single header file.

Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 drivers/net/ibmveth.h        | 10 ----------
 include/asm-powerpc/hvcall.h |  8 ++++++++
 2 files changed, 8 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/net/ibmveth.h b/drivers/net/ibmveth.h
index 8385bf836507..149191cef2f0 100644
--- a/drivers/net/ibmveth.h
+++ b/drivers/net/ibmveth.h
@@ -41,16 +41,6 @@
 #define IbmVethMcastRemoveFilter     0x2UL
 #define IbmVethMcastClearFilterTable 0x3UL
 
-/* hcall numbers */
-#define H_VIO_SIGNAL             0x104
-#define H_REGISTER_LOGICAL_LAN   0x114
-#define H_FREE_LOGICAL_LAN       0x118
-#define H_ADD_LOGICAL_LAN_BUFFER 0x11C
-#define H_SEND_LOGICAL_LAN       0x120
-#define H_MULTICAST_CTRL         0x130
-#define H_CHANGE_LOGICAL_LAN_MAC 0x14C
-#define H_FREE_LOGICAL_LAN_BUFFER 0x1D4
-
 /* hcall macros */
 #define h_register_logical_lan(ua, buflst, rxq, fltlst, mac) \
   plpar_hcall_norets(H_REGISTER_LOGICAL_LAN, ua, buflst, rxq, fltlst, mac)
diff --git a/include/asm-powerpc/hvcall.h b/include/asm-powerpc/hvcall.h
index 0d3c4e85711a..f07ae50cbc2c 100644
--- a/include/asm-powerpc/hvcall.h
+++ b/include/asm-powerpc/hvcall.h
@@ -164,9 +164,15 @@
 #define H_VIO_SIGNAL		0x104
 #define H_SEND_CRQ		0x108
 #define H_COPY_RDMA		0x110
+#define H_REGISTER_LOGICAL_LAN	0x114
+#define H_FREE_LOGICAL_LAN	0x118
+#define H_ADD_LOGICAL_LAN_BUFFER 0x11C
+#define H_SEND_LOGICAL_LAN	0x120
+#define H_MULTICAST_CTRL	0x130
 #define H_SET_XDABR		0x134
 #define H_STUFF_TCE		0x138
 #define H_PUT_TCE_INDIRECT	0x13C
+#define H_CHANGE_LOGICAL_LAN_MAC 0x14C
 #define H_VTERM_PARTNER_INFO	0x150
 #define H_REGISTER_VTERM	0x154
 #define H_FREE_VTERM		0x158
@@ -196,11 +202,13 @@
 #define H_GET_HCA_INFO          0x1B8
 #define H_GET_PERF_COUNT        0x1BC
 #define H_MANAGE_TRACE          0x1C0
+#define H_FREE_LOGICAL_LAN_BUFFER 0x1D4
 #define H_QUERY_INT_STATE       0x1E4
 #define H_POLL_PENDING		0x1D8
 #define H_JOIN			0x298
 #define H_VASI_STATE            0x2A4
 #define H_ENABLE_CRQ		0x2B0
+#define MAX_HCALL_OPCODES	(H_ENABLE_CRQ >> 2)
 
 #ifndef __ASSEMBLY__
 
-- 
cgit v1.2.3


From b9377ffc3a03cde558d76349a262a1adbb6d3112 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Wed, 19 Jul 2006 08:01:28 +1000
Subject: [POWERPC] clean up pseries hcall interfaces

Our pseries hcall interfaces are out of control:

	plpar_hcall_norets
	plpar_hcall
	plpar_hcall_8arg_2ret
	plpar_hcall_4out
	plpar_hcall_7arg_7ret
	plpar_hcall_9arg_9ret

Create 3 interfaces to cover all cases:

	plpar_hcall_norets:	7 arguments no returns
	plpar_hcall:		6 arguments 4 returns
	plpar_hcall9:		9 arguments 9 returns

There are only 2 cases in the kernel that need plpar_hcall9, hopefully
we can keep it that way.

Pass in a buffer to stash return parameters so we avoid the &dummy1,
&dummy2 madness.

Signed-off-by: Anton Blanchard <anton@samba.org>
--
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/lparcfg.c                   |  18 ++-
 arch/powerpc/kernel/rtas.c                      |  11 +-
 arch/powerpc/platforms/pseries/hvCall.S         | 207 ++++--------------------
 arch/powerpc/platforms/pseries/hvconsole.c      |   5 +-
 arch/powerpc/platforms/pseries/lpar.c           |  12 +-
 arch/powerpc/platforms/pseries/plpar_wrappers.h |  97 +++++++++--
 arch/powerpc/platforms/pseries/xics.c           |  22 +--
 drivers/net/ibmveth.c                           |   3 +-
 drivers/net/ibmveth.h                           |  17 +-
 include/asm-powerpc/hvcall.h                    | 105 +++---------
 10 files changed, 184 insertions(+), 313 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 3ce3a2d56fa8..41c05dcd68f4 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -182,8 +182,14 @@ static unsigned int h_get_ppp(unsigned long *entitled,
 			      unsigned long *resource)
 {
 	unsigned long rc;
-	rc = plpar_hcall_4out(H_GET_PPP, 0, 0, 0, 0, entitled, unallocated,
-			      aggregation, resource);
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_GET_PPP, retbuf);
+
+	*entitled = retbuf[0];
+	*unallocated = retbuf[1];
+	*aggregation = retbuf[2];
+	*resource = retbuf[3];
 
 	log_plpar_hcall_return(rc, "H_GET_PPP");
 
@@ -193,8 +199,12 @@ static unsigned int h_get_ppp(unsigned long *entitled,
 static void h_pic(unsigned long *pool_idle_time, unsigned long *num_procs)
 {
 	unsigned long rc;
-	unsigned long dummy;
-	rc = plpar_hcall(H_PIC, 0, 0, 0, 0, pool_idle_time, num_procs, &dummy);
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_PIC, retbuf);
+
+	*pool_idle_time = retbuf[0];
+	*num_procs = retbuf[1];
 
 	if (rc != H_AUTHORITY)
 		log_plpar_hcall_return(rc, "H_PIC");
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index 10e10be324c9..14353b8789dd 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -668,15 +668,14 @@ static int rtas_ibm_suspend_me(struct rtas_args *args)
 	int i;
 	long state;
 	long rc;
-	unsigned long dummy;
-
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
 	struct rtas_suspend_me_data data;
 
 	/* Make sure the state is valid */
-	rc = plpar_hcall(H_VASI_STATE,
-			 ((u64)args->args[0] << 32) | args->args[1],
-			 0, 0, 0,
-			 &state, &dummy, &dummy);
+	rc = plpar_hcall(H_VASI_STATE, retbuf,
+			 ((u64)args->args[0] << 32) | args->args[1]);
+
+	state = retbuf[0];
 
 	if (rc) {
 		printk(KERN_ERR "rtas_ibm_suspend_me: vasi_state returned %ld\n",rc);
diff --git a/arch/powerpc/platforms/pseries/hvCall.S b/arch/powerpc/platforms/pseries/hvCall.S
index c9ff547f9d25..9a99b056bd27 100644
--- a/arch/powerpc/platforms/pseries/hvCall.S
+++ b/arch/powerpc/platforms/pseries/hvCall.S
@@ -1,7 +1,6 @@
 /*
  * This file contains the generic code to perform a call to the
  * pSeries LPAR hypervisor.
- * NOTE: this file will go away when we move to inline this work.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
@@ -16,42 +15,6 @@
 
 	.text
 
-/* long plpar_hcall(unsigned long opcode,		R3
-			unsigned long arg1,		R4
-			unsigned long arg2,		R5
-			unsigned long arg3,		R6
-			unsigned long arg4,		R7
-			unsigned long *out1,		R8
-			unsigned long *out2,		R9
-			unsigned long *out3);		R10
- */
-_GLOBAL(plpar_hcall)
-	HMT_MEDIUM
-
-	mfcr	r0
-
-	std	r8,STK_PARM(r8)(r1)	/* Save out ptrs */
-	std	r9,STK_PARM(r9)(r1)
-	std	r10,STK_PARM(r10)(r1)
-
-	stw	r0,8(r1)
-
-	HVSC				/* invoke the hypervisor */
-
-	lwz	r0,8(r1)
-
-	ld	r8,STK_PARM(r8)(r1)	/* Fetch r4-r6 ret args */
-	ld	r9,STK_PARM(r9)(r1)
-	ld	r10,STK_PARM(r10)(r1)
-	std	r4,0(r8)
-	std	r5,0(r9)
-	std	r6,0(r10)
-
-	mtcrf	0xff,r0
-	blr				/* return r3 = status */
-
-
-/* Simple interface with no output values (other than status) */
 _GLOBAL(plpar_hcall_norets)
 	HMT_MEDIUM
 
@@ -64,164 +27,64 @@ _GLOBAL(plpar_hcall_norets)
 	mtcrf	0xff,r0
 	blr				/* return r3 = status */
 
-
-/* long plpar_hcall_8arg_2ret(unsigned long opcode,	R3
-			unsigned long arg1,		R4
-			unsigned long arg2,		R5
-			unsigned long arg3,		R6
-			unsigned long arg4,		R7
-			unsigned long arg5,		R8
-			unsigned long arg6,		R9
-			unsigned long arg7,		R10
-			unsigned long arg8,		112(R1)
-			unsigned long *out1);		120(R1)
- */
-_GLOBAL(plpar_hcall_8arg_2ret)
+_GLOBAL(plpar_hcall)
 	HMT_MEDIUM
 
 	mfcr	r0
-	ld	r11,STK_PARM(r11)(r1)	/* put arg8 in R11 */
 	stw	r0,8(r1)
 
-	HVSC				/* invoke the hypervisor */
-
-	lwz	r0,8(r1)
-	ld	r10,STK_PARM(r12)(r1)	/* Fetch r4 ret arg */
-	std	r4,0(r10)
-	mtcrf	0xff,r0
-	blr				/* return r3 = status */
-
-
-/* long plpar_hcall_4out(unsigned long opcode,		R3
-		 	unsigned long arg1,		R4
-		 	unsigned long arg2,		R5
-		 	unsigned long arg3,		R6
-		 	unsigned long arg4,		R7
-		 	unsigned long *out1,		R8
-		 	unsigned long *out2,		R9
-		 	unsigned long *out3,		R10
-		 	unsigned long *out4);		112(R1)
- */
-_GLOBAL(plpar_hcall_4out)
-	HMT_MEDIUM
-
-	mfcr	r0
-	stw	r0,8(r1)
+	std     r4,STK_PARM(r4)(r1)     /* Save ret buffer */
 
-	std	r8,STK_PARM(r8)(r1)	/* Save out ptrs */
-	std	r9,STK_PARM(r9)(r1)
-	std	r10,STK_PARM(r10)(r1)
+	mr	r4,r5
+	mr	r5,r6
+	mr	r6,r7
+	mr	r7,r8
+	mr	r8,r9
+	mr	r9,r10
 
 	HVSC				/* invoke the hypervisor */
 
-	lwz	r0,8(r1)
-
-	ld	r8,STK_PARM(r8)(r1)	/* Fetch r4-r7 ret args */
-	ld	r9,STK_PARM(r9)(r1)
-	ld	r10,STK_PARM(r10)(r1)
-	ld	r11,STK_PARM(r11)(r1)
-	std	r4,0(r8)
-	std	r5,0(r9)
-	std	r6,0(r10)
-	std	r7,0(r11)
-
-	mtcrf	0xff,r0
-	blr				/* return r3 = status */
-
-/* plpar_hcall_7arg_7ret(unsigned long opcode,		R3
-			 unsigned long arg1,		R4
-			 unsigned long arg2,		R5
-			 unsigned long arg3,		R6
-			 unsigned long arg4,		R7
-			 unsigned long arg5,		R8
-			 unsigned long arg6,		R9
-			 unsigned long arg7,		R10
-			 unsigned long *out1,		112(R1)
-			 unsigned long *out2,		110(R1)
-			 unsigned long *out3,		108(R1)
-			 unsigned long *out4,		106(R1)
-			 unsigned long *out5,		104(R1)
-			 unsigned long *out6,		102(R1)
-			 unsigned long *out7);		100(R1)
-*/
-_GLOBAL(plpar_hcall_7arg_7ret)
-	HMT_MEDIUM
-
-	mfcr	r0
-	stw	r0,8(r1)
-
-	HVSC				/* invoke the hypervisor */
+	ld	r12,STK_PARM(r4)(r1)
+	std	r4,  0(r12)
+	std	r5,  8(r12)
+	std	r6, 16(r12)
+	std	r7, 24(r12)
 
 	lwz	r0,8(r1)
-
-	ld	r11,STK_PARM(r11)(r1)	/* Fetch r4 ret arg */
-	std	r4,0(r11)
-	ld	r11,STK_PARM(r12)(r1)	/* Fetch r5 ret arg */
-	std	r5,0(r11)
-	ld	r11,STK_PARM(r13)(r1)	/* Fetch r6 ret arg */
-	std	r6,0(r11)
-	ld	r11,STK_PARM(r14)(r1)	/* Fetch r7 ret arg */
-	std	r7,0(r11)
-	ld	r11,STK_PARM(r15)(r1)	/* Fetch r8 ret arg */
-	std	r8,0(r11)
-	ld	r11,STK_PARM(r16)(r1)	/* Fetch r9 ret arg */
-	std	r9,0(r11)
-	ld	r11,STK_PARM(r17)(r1)	/* Fetch r10 ret arg */
-	std	r10,0(r11)
-
 	mtcrf	0xff,r0
 
 	blr				/* return r3 = status */
 
-/* plpar_hcall_9arg_9ret(unsigned long opcode,		R3
-			 unsigned long arg1,		R4
-			 unsigned long arg2,		R5
-			 unsigned long arg3,		R6
-			 unsigned long arg4,		R7
-			 unsigned long arg5,		R8
-			 unsigned long arg6,		R9
-			 unsigned long arg7,		R10
-			 unsigned long arg8,		112(R1)
-			 unsigned long arg9,		110(R1)
-			 unsigned long *out1,		108(R1)
-			 unsigned long *out2,		106(R1)
-			 unsigned long *out3,		104(R1)
-			 unsigned long *out4,		102(R1)
-			 unsigned long *out5,		100(R1)
-			 unsigned long *out6,		 98(R1)
-			 unsigned long *out7);		 96(R1)
-			 unsigned long *out8,		 94(R1)
-		         unsigned long *out9,            92(R1)
-*/
-_GLOBAL(plpar_hcall_9arg_9ret)
+_GLOBAL(plpar_hcall9)
 	HMT_MEDIUM
 
 	mfcr	r0
 	stw	r0,8(r1)
 
-	ld	r11,STK_PARM(r11)(r1)	 /* put arg8 in R11 */
-	ld	r12,STK_PARM(r12)(r1)    /* put arg9 in R12 */
+	std     r4,STK_PARM(r4)(r1)     /* Save ret buffer */
+
+	mr	r4,r5
+	mr	r5,r6
+	mr	r6,r7
+	mr	r7,r8
+	mr	r8,r9
+	mr	r9,r10
+	ld	r10,STK_PARM(r11)(r1)	 /* put arg7 in R10 */
+	ld	r11,STK_PARM(r12)(r1)	 /* put arg8 in R11 */
+	ld	r12,STK_PARM(r13)(r1)    /* put arg9 in R12 */
 
 	HVSC				/* invoke the hypervisor */
 
-	ld	r0,STK_PARM(r13)(r1)	/* Fetch r4 ret arg */
-	stdx	r4,r0,r0
-	ld	r0,STK_PARM(r14)(r1)	/* Fetch r5 ret arg */
-	stdx	r5,r0,r0
-	ld	r0,STK_PARM(r15)(r1)	/* Fetch r6 ret arg */
-	stdx	r6,r0,r0
-	ld	r0,STK_PARM(r16)(r1)	/* Fetch r7 ret arg */
-	stdx	r7,r0,r0
-	ld	r0,STK_PARM(r17)(r1)	/* Fetch r8 ret arg */
-	stdx	r8,r0,r0
-	ld	r0,STK_PARM(r18)(r1)	/* Fetch r9 ret arg */
-	stdx	r9,r0,r0
-	ld	r0,STK_PARM(r19)(r1)	/* Fetch r10 ret arg */
-	stdx	r10,r0,r0
-	ld	r0,STK_PARM(r20)(r1)	/* Fetch r11 ret arg */
-	stdx	r11,r0,r0
-	ld	r0,STK_PARM(r21)(r1)	/* Fetch r12 ret arg */
-	stdx	r12,r0,r0
+	ld	r12,STK_PARM(r4)(r1)
+	std	r4,  0(r12)
+	std	r5,  8(r12)
+	std	r6, 16(r12)
+	std	r7, 24(r12)
+	std	r8, 32(r12)
+	std	r9, 40(r12)
+	std	r10,48(r12)
+	std	r11,56(r12)
+	std	r12,64(r12)
 
 	lwz	r0,8(r1)
 	mtcrf	0xff,r0
diff --git a/arch/powerpc/platforms/pseries/hvconsole.c b/arch/powerpc/platforms/pseries/hvconsole.c
index a72a987f1d4d..3f6a89b09816 100644
--- a/arch/powerpc/platforms/pseries/hvconsole.c
+++ b/arch/powerpc/platforms/pseries/hvconsole.c
@@ -27,6 +27,7 @@
 #include <linux/module.h>
 #include <asm/hvcall.h>
 #include <asm/hvconsole.h>
+#include "plpar_wrappers.h"
 
 /**
  * hvc_get_chars - retrieve characters from firmware for denoted vterm adatper
@@ -40,9 +41,9 @@ int hvc_get_chars(uint32_t vtermno, char *buf, int count)
 {
 	unsigned long got;
 
-	if (plpar_hcall(H_GET_TERM_CHAR, vtermno, 0, 0, 0, &got,
-		(unsigned long *)buf, (unsigned long *)buf+1) == H_SUCCESS)
+	if (plpar_get_term_char(vtermno, &got, buf) == H_SUCCESS)
 		return got;
+
 	return 0;
 }
 
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 4cb7ff227f72..6cbf14266d5e 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -48,13 +48,11 @@
 #define DBG_LOW(fmt...) do { } while(0)
 #endif
 
-/* in pSeries_hvCall.S */
+/* in hvCall.S */
 EXPORT_SYMBOL(plpar_hcall);
-EXPORT_SYMBOL(plpar_hcall_4out);
+EXPORT_SYMBOL(plpar_hcall9);
 EXPORT_SYMBOL(plpar_hcall_norets);
-EXPORT_SYMBOL(plpar_hcall_8arg_2ret);
-EXPORT_SYMBOL(plpar_hcall_7arg_7ret);
-EXPORT_SYMBOL(plpar_hcall_9arg_9ret);
+
 extern void pSeries_find_serial_port(void);
 
 
@@ -277,7 +275,6 @@ long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	unsigned long flags;
 	unsigned long slot;
 	unsigned long hpte_v, hpte_r;
-	unsigned long dummy0, dummy1;
 
 	if (!(vflags & HPTE_V_BOLTED))
 		DBG_LOW("hpte_insert(group=%lx, va=%016lx, pa=%016lx, "
@@ -302,8 +299,7 @@ long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	if (rflags & (_PAGE_GUARDED|_PAGE_NO_CACHE))
 		hpte_r &= ~_PAGE_COHERENT;
 
-	lpar_rc = plpar_hcall(H_ENTER, flags, hpte_group, hpte_v,
-			      hpte_r, &slot, &dummy0, &dummy1);
+	lpar_rc = plpar_pte_enter(flags, hpte_group, hpte_v, hpte_r, &slot);
 	if (unlikely(lpar_rc == H_PTEG_FULL)) {
 		if (!(vflags & HPTE_V_BOLTED))
 			DBG_LOW(" full\n");
diff --git a/arch/powerpc/platforms/pseries/plpar_wrappers.h b/arch/powerpc/platforms/pseries/plpar_wrappers.h
index 3bd1b3e06003..ebd15de7597e 100644
--- a/arch/powerpc/platforms/pseries/plpar_wrappers.h
+++ b/arch/powerpc/platforms/pseries/plpar_wrappers.h
@@ -5,20 +5,17 @@
 
 static inline long poll_pending(void)
 {
-	unsigned long dummy;
-	return plpar_hcall(H_POLL_PENDING, 0, 0, 0, 0, &dummy, &dummy, &dummy);
+	return plpar_hcall_norets(H_POLL_PENDING);
 }
 
 static inline long prod_processor(void)
 {
-	plpar_hcall_norets(H_PROD);
-	return 0;
+	return plpar_hcall_norets(H_PROD);
 }
 
 static inline long cede_processor(void)
 {
-	plpar_hcall_norets(H_CEDE);
-	return 0;
+	return plpar_hcall_norets(H_CEDE);
 }
 
 static inline long vpa_call(unsigned long flags, unsigned long cpu,
@@ -42,21 +39,47 @@ static inline long register_vpa(unsigned long cpu, unsigned long vpa)
 
 extern void vpa_init(int cpu);
 
+static inline long plpar_pte_enter(unsigned long flags,
+		unsigned long hpte_group, unsigned long hpte_v,
+		unsigned long hpte_r, unsigned long *slot)
+{
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_ENTER, retbuf, flags, hpte_group, hpte_v, hpte_r);
+
+	*slot = retbuf[0];
+
+	return rc;
+}
+
 static inline long plpar_pte_remove(unsigned long flags, unsigned long ptex,
 		unsigned long avpn, unsigned long *old_pteh_ret,
 		unsigned long *old_ptel_ret)
 {
-	unsigned long dummy;
-	return plpar_hcall(H_REMOVE, flags, ptex, avpn, 0, old_pteh_ret,
-			old_ptel_ret, &dummy);
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_REMOVE, retbuf, flags, ptex, avpn);
+
+	*old_pteh_ret = retbuf[0];
+	*old_ptel_ret = retbuf[1];
+
+	return rc;
 }
 
 static inline long plpar_pte_read(unsigned long flags, unsigned long ptex,
 		unsigned long *old_pteh_ret, unsigned long *old_ptel_ret)
 {
-	unsigned long dummy;
-	return plpar_hcall(H_READ, flags, ptex, 0, 0, old_pteh_ret,
-			old_ptel_ret, &dummy);
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_READ, retbuf, flags, ptex);
+
+	*old_pteh_ret = retbuf[0];
+	*old_ptel_ret = retbuf[1];
+
+	return rc;
 }
 
 static inline long plpar_pte_protect(unsigned long flags, unsigned long ptex,
@@ -68,9 +91,14 @@ static inline long plpar_pte_protect(unsigned long flags, unsigned long ptex,
 static inline long plpar_tce_get(unsigned long liobn, unsigned long ioba,
 		unsigned long *tce_ret)
 {
-	unsigned long dummy;
-	return plpar_hcall(H_GET_TCE, liobn, ioba, 0, 0, tce_ret, &dummy,
-			&dummy);
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_GET_TCE, retbuf, liobn, ioba);
+
+	*tce_ret = retbuf[0];
+
+	return rc;
 }
 
 static inline long plpar_tce_put(unsigned long liobn, unsigned long ioba,
@@ -94,9 +122,17 @@ static inline long plpar_tce_stuff(unsigned long liobn, unsigned long ioba,
 static inline long plpar_get_term_char(unsigned long termno,
 		unsigned long *len_ret, char *buf_ret)
 {
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
 	unsigned long *lbuf = (unsigned long *)buf_ret;	/* TODO: alignment? */
-	return plpar_hcall(H_GET_TERM_CHAR, termno, 0, 0, 0, len_ret,
-			lbuf + 0, lbuf + 1);
+
+	rc = plpar_hcall(H_GET_TERM_CHAR, retbuf, termno);
+
+	*len_ret = retbuf[0];
+	lbuf[0] = retbuf[1];
+	lbuf[1] = retbuf[2];
+
+	return rc;
 }
 
 static inline long plpar_put_term_char(unsigned long termno, unsigned long len,
@@ -107,4 +143,31 @@ static inline long plpar_put_term_char(unsigned long termno, unsigned long len,
 			lbuf[1]);
 }
 
+static inline long plpar_eoi(unsigned long xirr)
+{
+	return plpar_hcall_norets(H_EOI, xirr);
+}
+
+static inline long plpar_cppr(unsigned long cppr)
+{
+	return plpar_hcall_norets(H_CPPR, cppr);
+}
+
+static inline long plpar_ipi(unsigned long servernum, unsigned long mfrr)
+{
+	return plpar_hcall_norets(H_IPI, servernum, mfrr);
+}
+
+static inline long plpar_xirr(unsigned long *xirr_ret)
+{
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
+
+	rc = plpar_hcall(H_XIRR, retbuf);
+
+	*xirr_ret = retbuf[0];
+
+	return rc;
+}
+
 #endif /* _PSERIES_PLPAR_WRAPPERS_H */
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index 1eab4688be17..c88ec63129f3 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -34,6 +34,7 @@
 #include <asm/i8259.h>
 
 #include "xics.h"
+#include "plpar_wrappers.h"
 
 #define XICS_IPI		2
 #define XICS_IRQ_SPURIOUS	0
@@ -110,27 +111,6 @@ static inline void direct_qirr_info(int n_cpu, u8 value)
 /* LPAR low level accessors */
 
 
-static inline long plpar_eoi(unsigned long xirr)
-{
-	return plpar_hcall_norets(H_EOI, xirr);
-}
-
-static inline long plpar_cppr(unsigned long cppr)
-{
-	return plpar_hcall_norets(H_CPPR, cppr);
-}
-
-static inline long plpar_ipi(unsigned long servernum, unsigned long mfrr)
-{
-	return plpar_hcall_norets(H_IPI, servernum, mfrr);
-}
-
-static inline long plpar_xirr(unsigned long *xirr_ret)
-{
-	unsigned long dummy;
-	return plpar_hcall(H_XIRR, 0, 0, 0, 0, xirr_ret, &dummy, &dummy);
-}
-
 static inline unsigned int lpar_xirr_info_get(int n_cpu)
 {
 	unsigned long lpar_rc;
diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c
index 0464e78f733a..e56eac88b809 100644
--- a/drivers/net/ibmveth.c
+++ b/drivers/net/ibmveth.c
@@ -702,7 +702,8 @@ static int ibmveth_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 					     desc[3].desc,
 					     desc[4].desc,
 					     desc[5].desc,
-					     correlator);
+					     correlator,
+					     &correlator);
 	} while ((lpar_rc == H_BUSY) && (retry_count--));
 
 	if(lpar_rc != H_SUCCESS && lpar_rc != H_DROPPED) {
diff --git a/drivers/net/ibmveth.h b/drivers/net/ibmveth.h
index 149191cef2f0..f5b25bff1540 100644
--- a/drivers/net/ibmveth.h
+++ b/drivers/net/ibmveth.h
@@ -51,8 +51,21 @@
 #define h_add_logical_lan_buffer(ua, buf) \
   plpar_hcall_norets(H_ADD_LOGICAL_LAN_BUFFER, ua, buf)
 
-#define h_send_logical_lan(ua, buf1, buf2, buf3, buf4, buf5, buf6, correlator) \
-  plpar_hcall_8arg_2ret(H_SEND_LOGICAL_LAN, ua, buf1, buf2, buf3, buf4, buf5, buf6, correlator, &correlator)
+static inline long h_send_logical_lan(unsigned long unit_address,
+		unsigned long desc1, unsigned long desc2, unsigned long desc3,
+		unsigned long desc4, unsigned long desc5, unsigned long desc6,
+		unsigned long corellator_in, unsigned long *corellator_out)
+{
+	long rc;
+	unsigned long retbuf[PLPAR_HCALL9_BUFSIZE];
+
+	rc = plpar_hcall9(H_SEND_LOGICAL_LAN, retbuf, unit_address, desc1,
+			desc2, desc3, desc4, desc5, desc6, corellator_in);
+
+	*corellator_out = retbuf[0];
+
+	return rc;
+}
 
 #define h_multicast_ctrl(ua, cmd, mac) \
   plpar_hcall_norets(H_MULTICAST_CTRL, ua, cmd, mac)
diff --git a/include/asm-powerpc/hvcall.h b/include/asm-powerpc/hvcall.h
index f07ae50cbc2c..63ce1ac8c1f4 100644
--- a/include/asm-powerpc/hvcall.h
+++ b/include/asm-powerpc/hvcall.h
@@ -212,94 +212,39 @@
 
 #ifndef __ASSEMBLY__
 
-/* plpar_hcall() -- Generic call interface using above opcodes
+/**
+ * plpar_hcall_norets: - Make a pseries hypervisor call with no return arguments
+ * @opcode: The hypervisor call to make.
  *
- * The actual call interface is a hypervisor call instruction with
- * the opcode in R3 and input args in R4-R7.
- * Status is returned in R3 with variable output values in R4-R11.
- * Only H_PTE_READ with H_READ_4 uses R6-R11 so we ignore it for now
- * and return only two out args which MUST ALWAYS BE PROVIDED.
- */
-long plpar_hcall(unsigned long opcode,
-		 unsigned long arg1,
-		 unsigned long arg2,
-		 unsigned long arg3,
-		 unsigned long arg4,
-		 unsigned long *out1,
-		 unsigned long *out2,
-		 unsigned long *out3);
-
-/* Same as plpar_hcall but for those opcodes that return no values
- * other than status.  Slightly more efficient.
+ * This call supports up to 7 arguments and only returns the status of
+ * the hcall. Use this version where possible, its slightly faster than
+ * the other plpar_hcalls.
  */
 long plpar_hcall_norets(unsigned long opcode, ...);
 
-/*
- * Special hcall interface for ibmveth support.
- * Takes 8 input parms. Returns a rc and stores the
- * R4 return value in *out1.
- */
-long plpar_hcall_8arg_2ret(unsigned long opcode,
-			   unsigned long arg1,
-			   unsigned long arg2,
-			   unsigned long arg3,
-			   unsigned long arg4,
-			   unsigned long arg5,
-			   unsigned long arg6,
-			   unsigned long arg7,
-			   unsigned long arg8,
-			   unsigned long *out1);
-
-/* plpar_hcall_4out()
+/**
+ * plpar_hcall: - Make a pseries hypervisor call
+ * @opcode: The hypervisor call to make.
+ * @retbuf: Buffer to store up to 4 return arguments in.
  *
- * same as plpar_hcall except with 4 output arguments.
+ * This call supports up to 6 arguments and 4 return arguments. Use
+ * PLPAR_HCALL_BUFSIZE to size the return argument buffer.
  *
+ * Used for all but the craziest of phyp interfaces (see plpar_hcall9)
  */
-long plpar_hcall_4out(unsigned long opcode,
-		      unsigned long arg1,
-		      unsigned long arg2,
-		      unsigned long arg3,
-		      unsigned long arg4,
-		      unsigned long *out1,
-		      unsigned long *out2,
-		      unsigned long *out3,
-		      unsigned long *out4);
+#define PLPAR_HCALL_BUFSIZE 4
+long plpar_hcall(unsigned long opcode, unsigned long *retbuf, ...);
 
-long plpar_hcall_7arg_7ret(unsigned long opcode,
-			   unsigned long arg1,
-			   unsigned long arg2,
-			   unsigned long arg3,
-			   unsigned long arg4,
-			   unsigned long arg5,
-			   unsigned long arg6,
-			   unsigned long arg7,
-			   unsigned long *out1,
-			   unsigned long *out2,
-			   unsigned long *out3,
-			   unsigned long *out4,
-			   unsigned long *out5,
-			   unsigned long *out6,
-			   unsigned long *out7);
-
-long plpar_hcall_9arg_9ret(unsigned long opcode,
-			   unsigned long arg1,
-			   unsigned long arg2,
-			   unsigned long arg3,
-			   unsigned long arg4,
-			   unsigned long arg5,
-			   unsigned long arg6,
-			   unsigned long arg7,
-			   unsigned long arg8,
-			   unsigned long arg9,
-			   unsigned long *out1,
-			   unsigned long *out2,
-			   unsigned long *out3,
-			   unsigned long *out4,
-			   unsigned long *out5,
-			   unsigned long *out6,
-			   unsigned long *out7,
-			   unsigned long *out8,
-			   unsigned long *out9);
+/**
+ * plpar_hcall9: - Make a pseries hypervisor call with up to 9 return arguments
+ * @opcode: The hypervisor call to make.
+ * @retbuf: Buffer to store up to 9 return arguments in.
+ *
+ * This call supports up to 9 arguments and 9 return arguments. Use
+ * PLPAR_HCALL9_BUFSIZE to size the return argument buffer.
+ */
+#define PLPAR_HCALL9_BUFSIZE 9
+long plpar_hcall9(unsigned long opcode, unsigned long *retbuf, ...);
 
 #endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From 9e5c50fa8686ede7c37b939a0b950df50346eb3d Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 4 Aug 2006 17:18:30 +0200
Subject: [SCSI] remove SCSI_STATE_ #defines

These aren't used anymore since the field in scsi_cmnd where it was
stored has been removed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 include/scsi/scsi_cmnd.h | 14 --------------
 1 file changed, 14 deletions(-)

(limited to 'include')

diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index 58e6444eebee..be117f812deb 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -118,20 +118,6 @@ struct scsi_cmnd {
 	unsigned long pid;	/* Process ID, starts at 0. Unique per host. */
 };
 
-/*
- * These are the values that scsi_cmd->state can take.
- */
-#define SCSI_STATE_TIMEOUT         0x1000
-#define SCSI_STATE_FINISHED        0x1001
-#define SCSI_STATE_FAILED          0x1002
-#define SCSI_STATE_QUEUED          0x1003
-#define SCSI_STATE_UNUSED          0x1006
-#define SCSI_STATE_DISCONNECTING   0x1008
-#define SCSI_STATE_INITIALIZING    0x1009
-#define SCSI_STATE_BHQUEUE         0x100a
-#define SCSI_STATE_MLQUEUE         0x100b
-
-
 extern struct scsi_cmnd *scsi_get_command(struct scsi_device *, gfp_t);
 extern void scsi_put_command(struct scsi_cmnd *);
 extern void scsi_io_completion(struct scsi_cmnd *, unsigned int);
-- 
cgit v1.2.3


From 4ff36718ede26ee2da73f2dae94d71e2b06845fc Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Tue, 4 Jul 2006 12:15:20 -0600
Subject: [SCSI] Improve inquiry printing

 - Replace scsi_device_types array API with scsi_device_type function API.
   Gets rid of a lot of common code, as well as being easier to use.
 - Add the new device types in SPC4 r05a, and rename some of the older ones.
 - Reformat the printing of inquiry data; now fits on one line and
   includes PQ.

I think I've addressed all the feedback from the previous versions.  My
current test box prints:

scsi 2:0:1:0: Direct access     HP 18.2G ATLAS10K3_18_SCA HP05 PQ: 0 ANSI: 2

Signed-off-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/block/cciss_scsi.c | 14 ++++------
 drivers/scsi/fcal.c        |  3 +--
 drivers/scsi/g_NCR5380.c   |  3 +--
 drivers/scsi/megaraid.c    |  4 +--
 drivers/scsi/scsi.c        | 36 ++++++++++++++++++-------
 drivers/scsi/scsi_proc.c   |  4 +--
 drivers/scsi/scsi_scan.c   | 66 ++++++----------------------------------------
 include/scsi/scsi.h        | 10 +++----
 8 files changed, 46 insertions(+), 94 deletions(-)

(limited to 'include')

diff --git a/drivers/block/cciss_scsi.c b/drivers/block/cciss_scsi.c
index afdff32f6724..05f79d7393f7 100644
--- a/drivers/block/cciss_scsi.c
+++ b/drivers/block/cciss_scsi.c
@@ -251,10 +251,6 @@ scsi_cmd_stack_free(int ctlr)
 	stk->pool = NULL;
 }
 
-/* scsi_device_types comes from scsi.h */
-#define DEVICETYPE(n) (n<0 || n>MAX_SCSI_DEVICE_CODE) ? \
-	"Unknown" : scsi_device_types[n]
-
 #if 0
 static int xmargin=8;
 static int amargin=60;
@@ -389,7 +385,7 @@ cciss_scsi_add_entry(int ctlr, int hostno,
 	   time anyway (the scsi layer's inquiries will show that info) */
 	if (hostno != -1)
 		printk("cciss%d: %s device c%db%dt%dl%d added.\n", 
-			ctlr, DEVICETYPE(sd->devtype), hostno, 
+			ctlr, scsi_device_type(sd->devtype), hostno,
 			sd->bus, sd->target, sd->lun);
 	return 0;
 }
@@ -407,7 +403,7 @@ cciss_scsi_remove_entry(int ctlr, int hostno, int entry)
 		ccissscsi[ctlr].dev[i] = ccissscsi[ctlr].dev[i+1];
 	ccissscsi[ctlr].ndevices--;
 	printk("cciss%d: %s device c%db%dt%dl%d removed.\n",
-		ctlr, DEVICETYPE(sd.devtype), hostno, 
+		ctlr, scsi_device_type(sd.devtype), hostno,
 			sd.bus, sd.target, sd.lun);
 }
 
@@ -458,7 +454,7 @@ adjust_cciss_scsi_table(int ctlr, int hostno,
 		if (found == 0) { /* device no longer present. */ 
 			changes++;
 			/* printk("cciss%d: %s device c%db%dt%dl%d removed.\n",
-				ctlr, DEVICETYPE(csd->devtype), hostno, 
+				ctlr, scsi_device_type(csd->devtype), hostno,
 					csd->bus, csd->target, csd->lun); */
 			cciss_scsi_remove_entry(ctlr, hostno, i);
 			/* note, i not incremented */
@@ -468,7 +464,7 @@ adjust_cciss_scsi_table(int ctlr, int hostno,
 			printk("cciss%d: device c%db%dt%dl%d type changed "
 				"(device type now %s).\n",
 				ctlr, hostno, csd->bus, csd->target, csd->lun,
-					DEVICETYPE(csd->devtype));
+					scsi_device_type(csd->devtype));
 			csd->devtype = sd[j].devtype;
 			i++;	/* so just move along. */
 		} else 		/* device is same as it ever was, */
@@ -1098,7 +1094,7 @@ cciss_update_non_disk_devices(int cntl_num, int hostno)
 			if (ncurrent >= CCISS_MAX_SCSI_DEVS_PER_HBA) {
 				printk(KERN_INFO "cciss%d: %s ignored, "
 					"too many devices.\n", cntl_num,
-					DEVICETYPE(devtype));
+					scsi_device_type(devtype));
 				break;
 			}
 			memcpy(&currentsd[ncurrent].scsi3addr[0], 
diff --git a/drivers/scsi/fcal.c b/drivers/scsi/fcal.c
index 7f891023aa15..c4e16c0775de 100644
--- a/drivers/scsi/fcal.c
+++ b/drivers/scsi/fcal.c
@@ -248,8 +248,7 @@ int fcal_proc_info (struct Scsi_Host *host, char *buffer, char **start, off_t of
 				if (scd->id == target) {
 					SPRINTF ("  [AL-PA: %02x, Id: %02d, Port WWN: %08x%08x, Node WWN: %08x%08x]  ",
 						alpa, target, u1[0], u1[1], u2[0], u2[1]);
-					SPRINTF ("%s ", (scd->type < MAX_SCSI_DEVICE_CODE) ?
-						scsi_device_types[(short) scd->type] : "Unknown device");
+					SPRINTF ("%s ", scsi_device_type(scd->type));
 
 					for (j = 0; (j < 8) && (scd->vendor[j] >= 0x20); j++)
 						SPRINTF ("%c", scd->vendor[j]);
diff --git a/drivers/scsi/g_NCR5380.c b/drivers/scsi/g_NCR5380.c
index 67f1100f3103..cdd893bb4e28 100644
--- a/drivers/scsi/g_NCR5380.c
+++ b/drivers/scsi/g_NCR5380.c
@@ -811,7 +811,6 @@ static int generic_NCR5380_proc_info(struct Scsi_Host *scsi_ptr, char *buffer, c
 	struct NCR5380_hostdata *hostdata;
 #ifdef NCR5380_STATS
 	struct scsi_device *dev;
-	extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE];
 #endif
 
 	NCR5380_setup(scsi_ptr);
@@ -851,7 +850,7 @@ static int generic_NCR5380_proc_info(struct Scsi_Host *scsi_ptr, char *buffer, c
 		long tr = hostdata->time_read[dev->id] / HZ;
 		long tw = hostdata->time_write[dev->id] / HZ;
 
-		PRINTP("  T:%d %s " ANDP dev->id ANDP(dev->type < MAX_SCSI_DEVICE_CODE) ? scsi_device_types[(int) dev->type] : "Unknown");
+		PRINTP("  T:%d %s " ANDP dev->id ANDP scsi_device_type(dev->type));
 		for (i = 0; i < 8; i++)
 			if (dev->vendor[i] >= 0x20)
 				*(buffer + (len++)) = dev->vendor[i];
diff --git a/drivers/scsi/megaraid.c b/drivers/scsi/megaraid.c
index 76edbb639d37..ccb0055ac73a 100644
--- a/drivers/scsi/megaraid.c
+++ b/drivers/scsi/megaraid.c
@@ -2822,9 +2822,7 @@ mega_print_inquiry(char *page, char *scsi_inq)
 
 	i = scsi_inq[0] & 0x1f;
 
-	len += sprintf(page+len, "  Type:   %s ",
-		i < MAX_SCSI_DEVICE_CODE ? scsi_device_types[i] :
-		   "Unknown          ");
+	len += sprintf(page+len, "  Type:   %s ", scsi_device_type(i));
 
 	len += sprintf(page+len,
 	"                 ANSI SCSI revision: %02x", scsi_inq[2] & 0x07);
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index b332caddd5b3..94df671d776a 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -96,24 +96,40 @@ unsigned int scsi_logging_level;
 EXPORT_SYMBOL(scsi_logging_level);
 #endif
 
-const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE] = {
-	"Direct-Access    ",
-	"Sequential-Access",
+static const char *const scsi_device_types[] = {
+	"Direct access    ",
+	"Sequential access",
 	"Printer          ",
 	"Processor        ",
 	"WORM             ",
-	"CD-ROM           ",
+	"CD/DVD           ",
 	"Scanner          ",
-	"Optical Device   ",
-	"Medium Changer   ",
+	"Optical memory   ",
+	"Media changer    ",
 	"Communications   ",
-	"Unknown          ",
-	"Unknown          ",
+	"ASC IT8          ",
+	"ASC IT8          ",
 	"RAID             ",
 	"Enclosure        ",
-	"Direct-Access-RBC",
+	"Direct access RBC",
+	"Optical card     ",
+	"Bridge controller",
+	"Object storage   ",
+	"Automation/Drive ",
 };
-EXPORT_SYMBOL(scsi_device_types);
+
+const char * scsi_device_type(unsigned type)
+{
+	if (type == 0x1e)
+		return "Well-known LUN   ";
+	if (type == 0x1f)
+		return "No Device        ";
+	if (type > ARRAY_SIZE(scsi_device_types))
+		return "Unknown          ";
+	return scsi_device_types[type];
+}
+
+EXPORT_SYMBOL(scsi_device_type);
 
 struct scsi_host_cmd_pool {
 	kmem_cache_t	*slab;
diff --git a/drivers/scsi/scsi_proc.c b/drivers/scsi/scsi_proc.c
index 55200e4fdf11..524a5f7a5193 100644
--- a/drivers/scsi/scsi_proc.c
+++ b/drivers/scsi/scsi_proc.c
@@ -178,9 +178,7 @@ static int proc_print_scsidevice(struct device *dev, void *data)
 
 	seq_printf(s, "\n");
 
-	seq_printf(s, "  Type:   %s ",
-		     sdev->type < MAX_SCSI_DEVICE_CODE ?
-	       scsi_device_types[(int) sdev->type] : "Unknown          ");
+	seq_printf(s, "  Type:   %s ", scsi_device_type(sdev->type));
 	seq_printf(s, "               ANSI"
 		     " SCSI revision: %02x", (sdev->scsi_level - 1) ?
 		     sdev->scsi_level - 1 : 1);
diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 1bd92b9b46d9..180399406510 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -133,59 +133,6 @@ static void scsi_unlock_floptical(struct scsi_device *sdev,
 			 SCSI_TIMEOUT, 3);
 }
 
-/**
- * print_inquiry - printk the inquiry information
- * @inq_result:	printk this SCSI INQUIRY
- *
- * Description:
- *     printk the vendor, model, and other information found in the
- *     INQUIRY data in @inq_result.
- *
- * Notes:
- *     Remove this, and replace with a hotplug event that logs any
- *     relevant information.
- **/
-static void print_inquiry(unsigned char *inq_result)
-{
-	int i;
-
-	printk(KERN_NOTICE "  Vendor: ");
-	for (i = 8; i < 16; i++)
-		if (inq_result[i] >= 0x20 && i < inq_result[4] + 5)
-			printk("%c", inq_result[i]);
-		else
-			printk(" ");
-
-	printk("  Model: ");
-	for (i = 16; i < 32; i++)
-		if (inq_result[i] >= 0x20 && i < inq_result[4] + 5)
-			printk("%c", inq_result[i]);
-		else
-			printk(" ");
-
-	printk("  Rev: ");
-	for (i = 32; i < 36; i++)
-		if (inq_result[i] >= 0x20 && i < inq_result[4] + 5)
-			printk("%c", inq_result[i]);
-		else
-			printk(" ");
-
-	printk("\n");
-
-	i = inq_result[0] & 0x1f;
-
-	printk(KERN_NOTICE "  Type:   %s ",
-	       i <
-	       MAX_SCSI_DEVICE_CODE ? scsi_device_types[i] :
-	       "Unknown          ");
-	printk("                 ANSI SCSI revision: %02x",
-	       inq_result[2] & 0x07);
-	if ((inq_result[2] & 0x07) == 1 && (inq_result[3] & 0x0f) == 1)
-		printk(" CCS\n");
-	else
-		printk("\n");
-}
-
 /**
  * scsi_alloc_sdev - allocate and setup a scsi_Device
  *
@@ -653,9 +600,8 @@ static int scsi_add_lun(struct scsi_device *sdev, char *inq_result, int *bflags)
 	if (*bflags & BLIST_ISROM) {
 		/*
 		 * It would be better to modify sdev->type, and set
-		 * sdev->removable, but then the print_inquiry() output
-		 * would not show TYPE_ROM; if print_inquiry() is removed
-		 * the issue goes away.
+		 * sdev->removable; this can now be done since
+		 * print_inquiry has gone away.
 		 */
 		inq_result[0] = TYPE_ROM;
 		inq_result[1] |= 0x80;	/* removable */
@@ -684,8 +630,6 @@ static int scsi_add_lun(struct scsi_device *sdev, char *inq_result, int *bflags)
 		printk(KERN_INFO "scsi: unknown device type %d\n", sdev->type);
 	}
 
-	print_inquiry(inq_result);
-
 	/*
 	 * For a peripheral qualifier (PQ) value of 1 (001b), the SCSI
 	 * spec says: The device server is capable of supporting the
@@ -715,6 +659,12 @@ static int scsi_add_lun(struct scsi_device *sdev, char *inq_result, int *bflags)
 	if (inq_result[7] & 0x10)
 		sdev->sdtr = 1;
 
+	sdev_printk(KERN_NOTICE "scsi", sdev, "%s %.8s %.16s %.4s PQ: %d "
+			"ANSI: %d%s\n", scsi_device_type(sdev->type),
+			sdev->vendor, sdev->model, sdev->rev,
+			sdev->inq_periph_qual, inq_result[2] & 0x07,
+			(inq_result[3] & 0x0f) == 1 ? " CCS" : "");
+
 	/*
 	 * End sysfs code.
 	 */
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index c60b8ff2f5e4..1bc675201413 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -24,13 +24,6 @@
 extern const unsigned char scsi_command_size[8];
 #define COMMAND_SIZE(opcode) scsi_command_size[((opcode) >> 5) & 7]
 
-/*
- *	SCSI device types
- */
-
-#define MAX_SCSI_DEVICE_CODE 15
-extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE];
-
 /*
  * Special value for scanning to specify scanning or rescanning of all
  * possible channels, (target) ids, or luns on a given shost.
@@ -225,6 +218,9 @@ static inline int scsi_status_is_good(int status)
 #define TYPE_RBC	    0x0e
 #define TYPE_NO_LUN         0x7f
 
+/* Returns a human-readable name for the device */
+extern const char * scsi_device_type(unsigned type);
+
 /*
  * standard mode-select header prepended to all mode-select commands
  */
-- 
cgit v1.2.3


From 40681b95a4ef798bc38c92e0d9b8c06bbdd34409 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Wed, 2 Aug 2006 11:13:50 +1000
Subject: [POWERPC] Make doc comments extractable

We don't have much in the way of doc comments, but some of those we do have
don't work because they start with "/***" or "/*", not "/**" which is what
kernel-doc requires.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/crash_dump.c |  2 +-
 arch/powerpc/sysdev/i8259.c      |  2 +-
 include/asm-powerpc/irq.h        | 24 ++++++++++++------------
 include/asm-powerpc/prom.h       |  8 ++++----
 4 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/crash_dump.c b/arch/powerpc/kernel/crash_dump.c
index 371973be8d71..2f6f5a7bc69e 100644
--- a/arch/powerpc/kernel/crash_dump.c
+++ b/arch/powerpc/kernel/crash_dump.c
@@ -80,7 +80,7 @@ static int __init parse_savemaxmem(char *p)
 }
 __setup("savemaxmem=", parse_savemaxmem);
 
-/*
+/**
  * copy_oldmem_page - copy one page from "oldmem"
  * @pfn: page frame number to be copied
  * @buf: target memory address for the copy; this can be in kernel address
diff --git a/arch/powerpc/sysdev/i8259.c b/arch/powerpc/sysdev/i8259.c
index 9855820b9548..26a6a3becd66 100644
--- a/arch/powerpc/sysdev/i8259.c
+++ b/arch/powerpc/sysdev/i8259.c
@@ -224,7 +224,7 @@ static struct irq_host_ops i8259_host_ops = {
 	.xlate = i8259_host_xlate,
 };
 
-/****
+/**
  * i8259_init - Initialize the legacy controller
  * @node: device node of the legacy PIC (can be NULL, but then, it will match
  *        all interrupts, so beware)
diff --git a/include/asm-powerpc/irq.h b/include/asm-powerpc/irq.h
index d903a62959da..4da41efb1319 100644
--- a/include/asm-powerpc/irq.h
+++ b/include/asm-powerpc/irq.h
@@ -137,7 +137,7 @@ struct irq_map_entry {
 extern struct irq_map_entry irq_map[NR_IRQS];
 
 
-/***
+/**
  * irq_alloc_host - Allocate a new irq_host data structure
  * @node: device-tree node of the interrupt controller
  * @revmap_type: type of reverse mapping to use
@@ -159,14 +159,14 @@ extern struct irq_host *irq_alloc_host(unsigned int revmap_type,
 				       irq_hw_number_t inval_irq);
 
 
-/***
+/**
  * irq_find_host - Locates a host for a given device node
  * @node: device-tree node of the interrupt controller
  */
 extern struct irq_host *irq_find_host(struct device_node *node);
 
 
-/***
+/**
  * irq_set_default_host - Set a "default" host
  * @host: default host pointer
  *
@@ -178,7 +178,7 @@ extern struct irq_host *irq_find_host(struct device_node *node);
 extern void irq_set_default_host(struct irq_host *host);
 
 
-/***
+/**
  * irq_set_virq_count - Set the maximum number of virt irqs
  * @count: number of linux virtual irqs, capped with NR_IRQS
  *
@@ -188,7 +188,7 @@ extern void irq_set_default_host(struct irq_host *host);
 extern void irq_set_virq_count(unsigned int count);
 
 
-/***
+/**
  * irq_create_mapping - Map a hardware interrupt into linux virq space
  * @host: host owning this hardware interrupt or NULL for default host
  * @hwirq: hardware irq number in that host space
@@ -202,13 +202,13 @@ extern unsigned int irq_create_mapping(struct irq_host *host,
 				       irq_hw_number_t hwirq);
 
 
-/***
+/**
  * irq_dispose_mapping - Unmap an interrupt
  * @virq: linux virq number of the interrupt to unmap
  */
 extern void irq_dispose_mapping(unsigned int virq);
 
-/***
+/**
  * irq_find_mapping - Find a linux virq from an hw irq number.
  * @host: host owning this hardware interrupt
  * @hwirq: hardware irq number in that host space
@@ -221,7 +221,7 @@ extern unsigned int irq_find_mapping(struct irq_host *host,
 				     irq_hw_number_t hwirq);
 
 
-/***
+/**
  * irq_radix_revmap - Find a linux virq from a hw irq number.
  * @host: host owning this hardware interrupt
  * @hwirq: hardware irq number in that host space
@@ -232,7 +232,7 @@ extern unsigned int irq_find_mapping(struct irq_host *host,
 extern unsigned int irq_radix_revmap(struct irq_host *host,
 				     irq_hw_number_t hwirq);
 
-/***
+/**
  * irq_linear_revmap - Find a linux virq from a hw irq number.
  * @host: host owning this hardware interrupt
  * @hwirq: hardware irq number in that host space
@@ -247,7 +247,7 @@ extern unsigned int irq_linear_revmap(struct irq_host *host,
 
 
 
-/***
+/**
  * irq_alloc_virt - Allocate virtual irq numbers
  * @host: host owning these new virtual irqs
  * @count: number of consecutive numbers to allocate
@@ -261,7 +261,7 @@ extern unsigned int irq_alloc_virt(struct irq_host *host,
 				   unsigned int count,
 				   unsigned int hint);
 
-/***
+/**
  * irq_free_virt - Free virtual irq numbers
  * @virq: virtual irq number of the first interrupt to free
  * @count: number of interrupts to free
@@ -300,7 +300,7 @@ extern unsigned int irq_of_parse_and_map(struct device_node *dev, int index);
 
 /* -- End OF helpers -- */
 
-/***
+/**
  * irq_early_init - Init irq remapping subsystem
  */
 extern void irq_early_init(void);
diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h
index 31bfea4686a6..7a457bd462a2 100644
--- a/include/asm-powerpc/prom.h
+++ b/include/asm-powerpc/prom.h
@@ -259,7 +259,7 @@ struct of_irq {
 	u32 specifier[OF_MAX_IRQ_SPEC];	/* Specifier copy */
 };
 
-/***
+/**
  * of_irq_map_init - Initialize the irq remapper
  * @flags:	flags defining workarounds to enable
  *
@@ -272,7 +272,7 @@ struct of_irq {
 
 extern void of_irq_map_init(unsigned int flags);
 
-/***
+/**
  * of_irq_map_raw - Low level interrupt tree parsing
  * @parent:	the device interrupt parent
  * @intspec:	interrupt specifier ("interrupts" property of the device)
@@ -292,7 +292,7 @@ extern int of_irq_map_raw(struct device_node *parent, const u32 *intspec,
 			  const u32 *addr, struct of_irq *out_irq);
 
 
-/***
+/**
  * of_irq_map_one - Resolve an interrupt for a device
  * @device:	the device whose interrupt is to be resolved
  * @index:     	index of the interrupt to resolve
@@ -305,7 +305,7 @@ extern int of_irq_map_raw(struct device_node *parent, const u32 *intspec,
 extern int of_irq_map_one(struct device_node *device, int index,
 			  struct of_irq *out_irq);
 
-/***
+/**
  * of_irq_map_pci - Resolve the interrupt for a PCI device
  * @pdev:	the device whose interrupt is to be resolved
  * @out_irq:	structure of_irq filled by this function
-- 
cgit v1.2.3


From 2f6093c84730b4bad65bcd0f2f904a5769b1dfc5 Mon Sep 17 00:00:00 2001
From: Michael Neuling <mikey@neuling.org>
Date: Mon, 7 Aug 2006 16:19:19 +1000
Subject: [POWERPC] Implement SLB shadow buffer

This adds a shadow buffer for the SLBs and regsiters it with PHYP.
Only the bolted SLB entries (top 3) are shadowed.

The SLB shadow buffer tells the hypervisor what the kernel needs to
have in the SLB for the kernel to be able to function.  The hypervisor
can use this information to speed up partition context switches.

Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/asm-offsets.c               |  2 ++
 arch/powerpc/kernel/entry_64.S                  | 13 +++++++++
 arch/powerpc/kernel/paca.c                      | 15 +++++++++-
 arch/powerpc/mm/slb.c                           | 37 ++++++++++++++++++++++---
 arch/powerpc/platforms/pseries/lpar.c           | 24 +++++++++++++---
 arch/powerpc/platforms/pseries/plpar_wrappers.h | 10 +++++++
 arch/powerpc/platforms/pseries/setup.c          | 12 ++++++--
 include/asm-powerpc/lppaca.h                    | 19 +++++++++++++
 include/asm-powerpc/paca.h                      |  3 ++
 9 files changed, 124 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index ac0631958b20..2ef7ea860379 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -135,11 +135,13 @@ int main(void)
 	DEFINE(PACA_STARTPURR, offsetof(struct paca_struct, startpurr));
 	DEFINE(PACA_USER_TIME, offsetof(struct paca_struct, user_time));
 	DEFINE(PACA_SYSTEM_TIME, offsetof(struct paca_struct, system_time));
+	DEFINE(PACA_SLBSHADOWPTR, offsetof(struct paca_struct, slb_shadow_ptr));
 
 	DEFINE(LPPACASRR0, offsetof(struct lppaca, saved_srr0));
 	DEFINE(LPPACASRR1, offsetof(struct lppaca, saved_srr1));
 	DEFINE(LPPACAANYINT, offsetof(struct lppaca, int_dword.any_int));
 	DEFINE(LPPACADECRINT, offsetof(struct lppaca, int_dword.fields.decr_int));
+	DEFINE(SLBSHADOW_SAVEAREA, offsetof(struct slb_shadow, save_area));
 #endif /* CONFIG_PPC64 */
 
 	/* RTAS */
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index 54d9f5cdaab4..5baea498ea64 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -323,6 +323,11 @@ _GLOBAL(ret_from_fork)
  * The code which creates the new task context is in 'copy_thread'
  * in arch/powerpc/kernel/process.c 
  */
+#define SHADOW_SLB_BOLTED_STACK_ESID \
+		(SLBSHADOW_SAVEAREA + 0x10*(SLB_NUM_BOLTED-1))
+#define SHADOW_SLB_BOLTED_STACK_VSID \
+		(SLBSHADOW_SAVEAREA + 0x10*(SLB_NUM_BOLTED-1) + 8)
+
 	.align	7
 _GLOBAL(_switch)
 	mflr	r0
@@ -375,6 +380,14 @@ BEGIN_FTR_SECTION
 	ld	r7,KSP_VSID(r4)	/* Get new stack's VSID */
 	oris	r0,r6,(SLB_ESID_V)@h
 	ori	r0,r0,(SLB_NUM_BOLTED-1)@l
+
+	/* Update the last bolted SLB */
+	ld	r9,PACA_SLBSHADOWPTR(r13)
+ 	li	r12,0
+  	std	r12,SHADOW_SLB_BOLTED_STACK_ESID(r9) /* Clear ESID */
+	std	r7,SHADOW_SLB_BOLTED_STACK_VSID(r9)  /* Save VSID */
+ 	std	r0,SHADOW_SLB_BOLTED_STACK_ESID(r9)  /* Save ESID */
+
 	slbie	r6
 	slbie	r6		/* Workaround POWER5 < DD2.1 issue */
 	slbmte	r7,r0
diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c
index c68741fed14b..55f1a25085cd 100644
--- a/arch/powerpc/kernel/paca.c
+++ b/arch/powerpc/kernel/paca.c
@@ -17,6 +17,7 @@
 #include <asm/lppaca.h>
 #include <asm/iseries/it_lp_reg_save.h>
 #include <asm/paca.h>
+#include <asm/mmu.h>
 
 
 /* This symbol is provided by the linker - let it fill in the paca
@@ -45,6 +46,17 @@ struct lppaca lppaca[] = {
 	},
 };
 
+/*
+ * 3 persistent SLBs are registered here.  The buffer will be zero
+ * initially, hence will all be invaild until we actually write them.
+ */
+struct slb_shadow slb_shadow[] __cacheline_aligned = {
+	[0 ... (NR_CPUS-1)] = {
+		.persistent = SLB_NUM_BOLTED,
+		.buffer_length = sizeof(struct slb_shadow),
+	},
+};
+
 /* The Paca is an array with one entry per processor.  Each contains an
  * lppaca, which contains the information shared between the
  * hypervisor and Linux.
@@ -59,7 +71,8 @@ struct lppaca lppaca[] = {
 	.lock_token = 0x8000,						    \
 	.paca_index = (number),		/* Paca Index */		    \
 	.kernel_toc = (unsigned long)(&__toc_start) + 0x8000UL,		    \
-	.hw_cpu_id = 0xffff,
+	.hw_cpu_id = 0xffff,						    \
+	.slb_shadow_ptr = &slb_shadow[number],
 
 #ifdef CONFIG_PPC_ISERIES
 #define PACA_INIT_ISERIES(number)					    \
diff --git a/arch/powerpc/mm/slb.c b/arch/powerpc/mm/slb.c
index de0c8842415c..d3733912adb4 100644
--- a/arch/powerpc/mm/slb.c
+++ b/arch/powerpc/mm/slb.c
@@ -22,6 +22,8 @@
 #include <asm/paca.h>
 #include <asm/cputable.h>
 #include <asm/cacheflush.h>
+#include <asm/smp.h>
+#include <linux/compiler.h>
 
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
@@ -50,9 +52,32 @@ static inline unsigned long mk_vsid_data(unsigned long ea, unsigned long flags)
 	return (get_kernel_vsid(ea) << SLB_VSID_SHIFT) | flags;
 }
 
-static inline void create_slbe(unsigned long ea, unsigned long flags,
-			       unsigned long entry)
+static inline void slb_shadow_update(unsigned long esid, unsigned long vsid,
+				     unsigned long entry)
 {
+	/*
+	 * Clear the ESID first so the entry is not valid while we are
+	 * updating it.
+	 */
+	get_slb_shadow()->save_area[entry].esid = 0;
+	barrier();
+	get_slb_shadow()->save_area[entry].vsid = vsid;
+	barrier();
+	get_slb_shadow()->save_area[entry].esid = esid;
+
+}
+
+static inline void create_shadowed_slbe(unsigned long ea, unsigned long flags,
+					unsigned long entry)
+{
+	/*
+	 * Updating the shadow buffer before writing the SLB ensures
+	 * we don't get a stale entry here if we get preempted by PHYP
+	 * between these two statements.
+	 */
+	slb_shadow_update(mk_esid_data(ea, entry), mk_vsid_data(ea, flags),
+			  entry);
+
 	asm volatile("slbmte  %0,%1" :
 		     : "r" (mk_vsid_data(ea, flags)),
 		       "r" (mk_esid_data(ea, entry))
@@ -77,6 +102,10 @@ void slb_flush_and_rebolt(void)
 	if ((ksp_esid_data & ESID_MASK) == PAGE_OFFSET)
 		ksp_esid_data &= ~SLB_ESID_V;
 
+	/* Only third entry (stack) may change here so only resave that */
+	slb_shadow_update(ksp_esid_data,
+			  mk_vsid_data(ksp_esid_data, lflags), 2);
+
 	/* We need to do this all in asm, so we're sure we don't touch
 	 * the stack between the slbia and rebolting it. */
 	asm volatile("isync\n"
@@ -209,9 +238,9 @@ void slb_initialize(void)
 	asm volatile("isync":::"memory");
 	asm volatile("slbmte  %0,%0"::"r" (0) : "memory");
 	asm volatile("isync; slbia; isync":::"memory");
-	create_slbe(PAGE_OFFSET, lflags, 0);
+	create_shadowed_slbe(PAGE_OFFSET, lflags, 0);
 
-	create_slbe(VMALLOC_START, vflags, 1);
+	create_shadowed_slbe(VMALLOC_START, vflags, 1);
 
 	/* We don't bolt the stack for the time being - we're in boot,
 	 * so the stack is in the bolted segment.  By the time it goes
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 6cbf14266d5e..1820a0b0a8c6 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -252,18 +252,34 @@ out:
 void vpa_init(int cpu)
 {
 	int hwcpu = get_hard_smp_processor_id(cpu);
-	unsigned long vpa = __pa(&lppaca[cpu]);
+	unsigned long addr;
 	long ret;
 
 	if (cpu_has_feature(CPU_FTR_ALTIVEC))
 		lppaca[cpu].vmxregs_in_use = 1;
 
-	ret = register_vpa(hwcpu, vpa);
+	addr = __pa(&lppaca[cpu]);
+	ret = register_vpa(hwcpu, addr);
 
-	if (ret)
+	if (ret) {
 		printk(KERN_ERR "WARNING: vpa_init: VPA registration for "
 				"cpu %d (hw %d) of area %lx returns %ld\n",
-				cpu, hwcpu, vpa, ret);
+				cpu, hwcpu, addr, ret);
+		return;
+	}
+	/*
+	 * PAPR says this feature is SLB-Buffer but firmware never
+	 * reports that.  All SPLPAR support SLB shadow buffer.
+	 */
+	addr = __pa(&slb_shadow[cpu]);
+	if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
+		ret = register_slb_shadow(hwcpu, addr);
+		if (ret)
+			printk(KERN_ERR
+			       "WARNING: vpa_init: SLB shadow buffer "
+			       "registration for cpu %d (hw %d) of area %lx "
+			       "returns %ld\n", cpu, hwcpu, addr, ret);
+	}
 }
 
 long pSeries_lpar_hpte_insert(unsigned long hpte_group,
diff --git a/arch/powerpc/platforms/pseries/plpar_wrappers.h b/arch/powerpc/platforms/pseries/plpar_wrappers.h
index ebd15de7597e..3eb7b294d92f 100644
--- a/arch/powerpc/platforms/pseries/plpar_wrappers.h
+++ b/arch/powerpc/platforms/pseries/plpar_wrappers.h
@@ -37,6 +37,16 @@ static inline long register_vpa(unsigned long cpu, unsigned long vpa)
 	return vpa_call(0x1, cpu, vpa);
 }
 
+static inline long unregister_slb_shadow(unsigned long cpu, unsigned long vpa)
+{
+	return vpa_call(0x7, cpu, vpa);
+}
+
+static inline long register_slb_shadow(unsigned long cpu, unsigned long vpa)
+{
+	return vpa_call(0x3, cpu, vpa);
+}
+
 extern void vpa_init(int cpu);
 
 static inline long plpar_pte_enter(unsigned long flags,
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index de214d86ff44..6ebeecfd6bcb 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -234,9 +234,17 @@ static void pseries_kexec_cpu_down_xics(int crash_shutdown, int secondary)
 {
 	/* Don't risk a hypervisor call if we're crashing */
 	if (firmware_has_feature(FW_FEATURE_SPLPAR) && !crash_shutdown) {
-		unsigned long vpa = __pa(get_lppaca());
+		unsigned long addr;
 
-		if (unregister_vpa(hard_smp_processor_id(), vpa)) {
+		addr = __pa(get_slb_shadow());
+		if (unregister_slb_shadow(hard_smp_processor_id(), addr))
+			printk("SLB shadow buffer deregistration of "
+			       "cpu %u (hw_cpu_id %d) failed\n",
+			       smp_processor_id(),
+			       hard_smp_processor_id());
+
+		addr = __pa(get_lppaca());
+		if (unregister_vpa(hard_smp_processor_id(), addr)) {
 			printk("VPA deregistration of cpu %u (hw_cpu_id %d) "
 					"failed\n", smp_processor_id(),
 					hard_smp_processor_id());
diff --git a/include/asm-powerpc/lppaca.h b/include/asm-powerpc/lppaca.h
index 4dc514aabfe7..942bb450baff 100644
--- a/include/asm-powerpc/lppaca.h
+++ b/include/asm-powerpc/lppaca.h
@@ -27,7 +27,9 @@
 //
 //
 //----------------------------------------------------------------------------
+#include <linux/cache.h>
 #include <asm/types.h>
+#include <asm/mmu.h>
 
 /* The Hypervisor barfs if the lppaca crosses a page boundary.  A 1k
  * alignment is sufficient to prevent this */
@@ -133,5 +135,22 @@ struct lppaca {
 
 extern struct lppaca lppaca[];
 
+/*
+ * SLB shadow buffer structure as defined in the PAPR.  The save_area
+ * contains adjacent ESID and VSID pairs for each shadowed SLB.  The
+ * ESID is stored in the lower 64bits, then the VSID.
+ */
+struct slb_shadow {
+	u32	persistent;		// Number of persistent SLBs	x00-x03
+	u32	buffer_length;		// Total shadow buffer length	x04-x07
+	u64	reserved;		// Alignment			x08-x0f
+	struct	{
+		u64     esid;
+		u64	vsid;
+	} save_area[SLB_NUM_BOLTED];	//				x10-x40
+} ____cacheline_aligned;
+
+extern struct slb_shadow slb_shadow[];
+
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_LPPACA_H */
diff --git a/include/asm-powerpc/paca.h b/include/asm-powerpc/paca.h
index 2d4585f06209..7ffa2512524e 100644
--- a/include/asm-powerpc/paca.h
+++ b/include/asm-powerpc/paca.h
@@ -23,6 +23,7 @@
 register struct paca_struct *local_paca asm("r13");
 #define get_paca()	local_paca
 #define get_lppaca()	(get_paca()->lppaca_ptr)
+#define get_slb_shadow()	(get_paca()->slb_shadow_ptr)
 
 struct task_struct;
 
@@ -98,6 +99,8 @@ struct paca_struct {
 	u64 user_time;			/* accumulated usermode TB ticks */
 	u64 system_time;		/* accumulated system TB ticks */
 	u64 startpurr;			/* PURR/TB value snapshot */
+
+	struct slb_shadow *slb_shadow_ptr;
 };
 
 extern struct paca_struct paca[];
-- 
cgit v1.2.3


From 5cf13911b1e72707b6f0eb39b2d819ec6e343d76 Mon Sep 17 00:00:00 2001
From: Michael Neuling <mikey@neuling.org>
Date: Mon, 7 Aug 2006 17:34:50 +1000
Subject: [POWERPC] Update lppaca offset comments

Update offset comments.  No functional change.

Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/lppaca.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-powerpc/lppaca.h b/include/asm-powerpc/lppaca.h
index 942bb450baff..821ea0c512b4 100644
--- a/include/asm-powerpc/lppaca.h
+++ b/include/asm-powerpc/lppaca.h
@@ -116,7 +116,7 @@ struct lppaca {
 
 
 //=============================================================================
-// CACHE_LINE_3 0x0100 - 0x007F: This line is shared with other processors
+// CACHE_LINE_3 0x0100 - 0x017F: This line is shared with other processors
 //=============================================================================
 	// This is the yield_count.  An "odd" value (low bit on) means that
 	// the processor is yielded (either because of an OS yield or a PLIC
@@ -128,7 +128,7 @@ struct lppaca {
 	u8	reserved6[124];		// Reserved                     x04-x7F
 
 //=============================================================================
-// CACHE_LINE_4-5 0x0100 - 0x01FF Contains PMC interrupt data
+// CACHE_LINE_4-5 0x0180 - 0x027F Contains PMC interrupt data
 //=============================================================================
 	u8	pmc_save_area[256];	// PMC interrupt Area           x00-xFF
 } __attribute__((__aligned__(0x400)));
-- 
cgit v1.2.3


From b03732f006bd1ecee32587ec8235c41af5ad905f Mon Sep 17 00:00:00 2001
From: Brian King <brking@us.ibm.com>
Date: Mon, 7 Aug 2006 14:27:10 -0500
Subject: [PATCH] libata: Add ata_host_set_init

Add ata_host_set_init in preparation for SAS attached SATA.

Signed-off-by: Brian King <brking@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/scsi/libata-core.c | 28 ++++++++++++++++++++++++----
 include/linux/libata.h     |  2 ++
 2 files changed, 26 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 351544d3653a..886440b128a5 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -5350,6 +5350,28 @@ err_out:
 	return NULL;
 }
 
+/**
+ *	ata_sas_host_init - Initialize a host_set struct
+ *	@host_set:	host_set to initialize
+ *	@dev:		device host_set is attached to
+ *	@flags:	host_set flags
+ *	@ops:		port_ops
+ *
+ *	LOCKING:
+ *	PCI/etc. bus probe sem.
+ *
+ */
+
+void ata_host_set_init(struct ata_host_set *host_set,
+		       struct device *dev, unsigned long flags,
+		       const struct ata_port_operations *ops)
+{
+	spin_lock_init(&host_set->lock);
+	host_set->dev = dev;
+	host_set->flags = flags;
+	host_set->ops = ops;
+}
+
 /**
  *	ata_device_add - Register hardware device with ATA and SCSI layers
  *	@ent: Probe information describing hardware device to be registered
@@ -5381,15 +5403,12 @@ int ata_device_add(const struct ata_probe_ent *ent)
 			   (ent->n_ports * sizeof(void *)), GFP_KERNEL);
 	if (!host_set)
 		return 0;
-	spin_lock_init(&host_set->lock);
 
-	host_set->dev = dev;
+	ata_host_set_init(host_set, dev, ent->host_set_flags, ent->port_ops);
 	host_set->n_ports = ent->n_ports;
 	host_set->irq = ent->irq;
 	host_set->mmio_base = ent->mmio_base;
 	host_set->private_data = ent->private_data;
-	host_set->ops = ent->port_ops;
-	host_set->flags = ent->host_set_flags;
 
 	/* register each port bound to this device */
 	for (i = 0; i < ent->n_ports; i++) {
@@ -5908,6 +5927,7 @@ EXPORT_SYMBOL_GPL(sata_deb_timing_hotplug);
 EXPORT_SYMBOL_GPL(sata_deb_timing_long);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
 EXPORT_SYMBOL_GPL(ata_std_ports);
+EXPORT_SYMBOL_GPL(ata_host_set_init);
 EXPORT_SYMBOL_GPL(ata_device_add);
 EXPORT_SYMBOL_GPL(ata_port_detach);
 EXPORT_SYMBOL_GPL(ata_host_set_remove);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index b9416708bba2..be15ef5d8d85 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -684,6 +684,8 @@ extern int ata_pci_clear_simplex(struct pci_dev *pdev);
 #endif /* CONFIG_PCI */
 extern int ata_device_add(const struct ata_probe_ent *ent);
 extern void ata_port_detach(struct ata_port *ap);
+extern void ata_host_set_init(struct ata_host_set *, struct device *,
+			      unsigned long, const struct ata_port_operations *);
 extern void ata_host_set_remove(struct ata_host_set *host_set);
 extern int ata_scsi_detect(struct scsi_host_template *sht);
 extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg);
-- 
cgit v1.2.3


From 80289167fd3ebaeb7b2641e69cbec44b61165fe7 Mon Sep 17 00:00:00 2001
From: Brian King <brking@us.ibm.com>
Date: Mon, 7 Aug 2006 14:27:31 -0500
Subject: [PATCH] libata: Add support for SATA attachment to SAS adapters

The following patch enhances libata to allow SAS device drivers
to utilize libata to talk to SATA devices. It introduces some
new APIs which allow libata to be used without allocating a
virtual scsi host.

New APIs:

ata_sas_port_alloc - Allocate an ata_port
ata_sas_port_init - Initialize an ata_port (probe device, etc)
ata_sas_port_destroy - Free an ata_port allocated by ata_sas_port_alloc
ata_sas_slave_configure - configure scsi device
ata_sas_queuecmd - queue a scsi command, similar to ata_scsi_queuecomand

These new APIs can be used either directly by a SAS LLDD or could be used
by the SAS transport class.

Possible usage for a SAS LLDD would be:

scsi_scan_host
	target_alloc
		ata_sas_port_alloc
	slave_alloc
		ata_sas_port_init
	slave_configure
		ata_sas_slave_configure

Commands received by the LLDD for SATA devices would call ata_sas_queuecmd.

Device teardown would occur with:

slave_destroy
	port_disable
target_destroy
	ata_sas_port_destroy

Signed-off-by: Brian King <brking@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/scsi/libata-core.c |   2 +-
 drivers/scsi/libata-scsi.c | 149 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/scsi/libata.h      |   1 +
 include/linux/libata.h     |   9 +++
 4 files changed, 160 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index dc35cccd2898..83d93fc0751b 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -1528,7 +1528,7 @@ err_out_nosup:
  *	Zero on success, negative errno otherwise.
  */
 
-static int ata_bus_probe(struct ata_port *ap)
+int ata_bus_probe(struct ata_port *ap)
 {
 	unsigned int classes[ATA_MAX_DEVICES];
 	int tries[ATA_MAX_DEVICES];
diff --git a/drivers/scsi/libata-scsi.c b/drivers/scsi/libata-scsi.c
index 1638e57028cb..37ec02661433 100644
--- a/drivers/scsi/libata-scsi.c
+++ b/drivers/scsi/libata-scsi.c
@@ -3158,3 +3158,152 @@ void ata_scsi_dev_rescan(void *data)
 			scsi_rescan_device(&(dev->sdev->sdev_gendev));
 	}
 }
+
+/**
+ *	ata_sas_port_alloc - Allocate port for a SAS attached SATA device
+ *	@pdev: PCI device that the scsi device is attached to
+ *	@port_info: Information from low-level host driver
+ *	@host: SCSI host that the scsi device is attached to
+ *
+ *	LOCKING:
+ *	PCI/etc. bus probe sem.
+ *
+ *	RETURNS:
+ *	ata_port pointer on success / NULL on failure.
+ */
+
+struct ata_port *ata_sas_port_alloc(struct ata_host_set *host_set,
+				    struct ata_port_info *port_info,
+				    struct Scsi_Host *host)
+{
+	struct ata_port *ap = kzalloc(sizeof(*ap), GFP_KERNEL);
+	struct ata_probe_ent *ent;
+
+	if (!ap)
+		return NULL;
+
+	ent = ata_probe_ent_alloc(host_set->dev, port_info);
+	if (!ent) {
+		kfree(ap);
+		return NULL;
+	}
+
+	ata_port_init(ap, host_set, ent, 0);
+	ap->lock = host->host_lock;
+	kfree(ent);
+	return ap;
+}
+EXPORT_SYMBOL_GPL(ata_sas_port_alloc);
+
+/**
+ *	ata_sas_port_start - Set port up for dma.
+ *	@ap: Port to initialize
+ *
+ *	Called just after data structures for each port are
+ *	initialized.  Allocates DMA pad.
+ *
+ *	May be used as the port_start() entry in ata_port_operations.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+int ata_sas_port_start(struct ata_port *ap)
+{
+	return ata_pad_alloc(ap, ap->dev);
+}
+EXPORT_SYMBOL_GPL(ata_sas_port_start);
+
+/**
+ *	ata_port_stop - Undo ata_sas_port_start()
+ *	@ap: Port to shut down
+ *
+ *	Frees the DMA pad.
+ *
+ *	May be used as the port_stop() entry in ata_port_operations.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ */
+
+void ata_sas_port_stop(struct ata_port *ap)
+{
+	ata_pad_free(ap, ap->dev);
+}
+EXPORT_SYMBOL_GPL(ata_sas_port_stop);
+
+/**
+ *	ata_sas_port_init - Initialize a SATA device
+ *	@ap: SATA port to initialize
+ *
+ *	LOCKING:
+ *	PCI/etc. bus probe sem.
+ *
+ *	RETURNS:
+ *	Zero on success, non-zero on error.
+ */
+
+int ata_sas_port_init(struct ata_port *ap)
+{
+	int rc = ap->ops->port_start(ap);
+
+	if (!rc)
+		rc = ata_bus_probe(ap);
+
+	return rc;
+}
+EXPORT_SYMBOL_GPL(ata_sas_port_init);
+
+/**
+ *	ata_sas_port_destroy - Destroy a SATA port allocated by ata_sas_port_alloc
+ *	@ap: SATA port to destroy
+ *
+ */
+
+void ata_sas_port_destroy(struct ata_port *ap)
+{
+	ap->ops->port_stop(ap);
+	kfree(ap);
+}
+EXPORT_SYMBOL_GPL(ata_sas_port_destroy);
+
+/**
+ *	ata_sas_slave_configure - Default slave_config routine for libata devices
+ *	@sdev: SCSI device to configure
+ *	@ap: ATA port to which SCSI device is attached
+ *
+ *	RETURNS:
+ *	Zero.
+ */
+
+int ata_sas_slave_configure(struct scsi_device *sdev, struct ata_port *ap)
+{
+	ata_scsi_sdev_config(sdev);
+	ata_scsi_dev_config(sdev, ap->device);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ata_sas_slave_configure);
+
+/**
+ *	ata_sas_queuecmd - Issue SCSI cdb to libata-managed device
+ *	@cmd: SCSI command to be sent
+ *	@done: Completion function, called when command is complete
+ *	@ap:	ATA port to which the command is being sent
+ *
+ *	RETURNS:
+ *	Zero.
+ */
+
+int ata_sas_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *),
+		     struct ata_port *ap)
+{
+	ata_scsi_dump_cdb(ap, cmd);
+
+	if (likely(ata_scsi_dev_enabled(ap->device)))
+		__ata_scsi_queuecmd(cmd, done, ap->device);
+	else {
+		cmd->result = (DID_BAD_TARGET << 16);
+		done(cmd);
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ata_sas_queuecmd);
diff --git a/drivers/scsi/libata.h b/drivers/scsi/libata.h
index 0b7a37c2785d..d4a4f82360ec 100644
--- a/drivers/scsi/libata.h
+++ b/drivers/scsi/libata.h
@@ -111,6 +111,7 @@ extern void ata_scsi_rbuf_fill(struct ata_scsi_args *args,
                                            u8 *rbuf, unsigned int buflen));
 extern void ata_schedule_scsi_eh(struct Scsi_Host *shost);
 extern void ata_scsi_dev_rescan(void *data);
+extern int ata_bus_probe(struct ata_port *ap);
 
 /* libata-eh.c */
 extern enum scsi_eh_timer_return ata_scsi_timed_out(struct scsi_cmnd *cmd);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index be15ef5d8d85..cf5eb1da3e32 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -691,6 +691,15 @@ extern int ata_scsi_detect(struct scsi_host_template *sht);
 extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg);
 extern int ata_scsi_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *));
 extern int ata_scsi_release(struct Scsi_Host *host);
+extern void ata_sas_port_destroy(struct ata_port *);
+extern struct ata_port *ata_sas_port_alloc(struct ata_host_set *,
+					   struct ata_port_info *, struct Scsi_Host *);
+extern int ata_sas_port_init(struct ata_port *);
+extern int ata_sas_port_start(struct ata_port *ap);
+extern void ata_sas_port_stop(struct ata_port *ap);
+extern int ata_sas_slave_configure(struct scsi_device *, struct ata_port *);
+extern int ata_sas_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *),
+			    struct ata_port *ap);
 extern unsigned int ata_host_intr(struct ata_port *ap, struct ata_queued_cmd *qc);
 extern int sata_scr_valid(struct ata_port *ap);
 extern int sata_scr_read(struct ata_port *ap, int reg, u32 *val);
-- 
cgit v1.2.3


From 2ec7df0457b710d9201f211dbccdbecf0ad38b7e Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Thu, 10 Aug 2006 16:59:10 +0900
Subject: [PATCH] libata: rework legacy handling to remove much of the cruft

Kill host_set->next
Fix simplex support
Allow per platform setting of IDE legacy bases

Some of this can be tidied further later on, in particular all the
legacy port gunge belongs as a PCI quirk/PCI header decode to understand
the special legacy IDE rules in the PCI spec.

Longer term Jeff also wants to move the request_irq/free_irq out of core
which will make this even cleaner.

tj: folded in three followup patches - ata_piix-fix, broken-arch-fix
and fix-new-legacy-handling, and separated per-dev xfermask into
separate patch preceding this one.  Folded in fixes are...

* ata_piix-fix: fix build failure due to host_set->next removal
* broken-arch-fix: add missing include/asm-*/libata-portmap.h
* fix-new-legacy-handling:
	* In ata_pci_init_legacy_port(), probe_num was incorrectly
          incremented during initialization of the secondary port and
          probe_ent->n_ports was incorrectly fixed to 1.

	* Both legacy ports ended up having the same hard_port_no.

	* When printing port information, both legacy ports printed
	  the first irq.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/scsi/ata_piix.c              |   2 -
 drivers/scsi/libata-bmdma.c          | 135 +++++++++++++++++------------------
 drivers/scsi/libata-core.c           |  59 +++++++++------
 include/asm-alpha/libata-portmap.h   |   1 +
 include/asm-generic/libata-portmap.h |  12 ++++
 include/asm-i386/libata-portmap.h    |   1 +
 include/asm-ia64/libata-portmap.h    |   1 +
 include/asm-powerpc/libata-portmap.h |   1 +
 include/asm-sparc/libata-portmap.h   |   1 +
 include/asm-sparc64/libata-portmap.h |   1 +
 include/asm-x86_64/libata-portmap.h  |   1 +
 include/linux/libata.h               |   5 +-
 12 files changed, 123 insertions(+), 97 deletions(-)
 create mode 100644 include/asm-alpha/libata-portmap.h
 create mode 100644 include/asm-generic/libata-portmap.h
 create mode 100644 include/asm-i386/libata-portmap.h
 create mode 100644 include/asm-ia64/libata-portmap.h
 create mode 100644 include/asm-powerpc/libata-portmap.h
 create mode 100644 include/asm-sparc/libata-portmap.h
 create mode 100644 include/asm-sparc64/libata-portmap.h
 create mode 100644 include/asm-x86_64/libata-portmap.h

(limited to 'include')

diff --git a/drivers/scsi/ata_piix.c b/drivers/scsi/ata_piix.c
index 5e8afc876980..40ecab5793e3 100644
--- a/drivers/scsi/ata_piix.c
+++ b/drivers/scsi/ata_piix.c
@@ -932,8 +932,6 @@ static int piix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 
 static void piix_host_stop(struct ata_host_set *host_set)
 {
-	if (host_set->next == NULL)
-		kfree(host_set->private_data);
 	ata_host_stop(host_set);
 }
 
diff --git a/drivers/scsi/libata-bmdma.c b/drivers/scsi/libata-bmdma.c
index 3268cf5375ea..e694f6075c3b 100644
--- a/drivers/scsi/libata-bmdma.c
+++ b/drivers/scsi/libata-bmdma.c
@@ -854,7 +854,7 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int
 		if (bmdma) {
 			bmdma += 8;
 			if(inb(bmdma + 2) & 0x80)
-			probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
 			probe_ent->port[p].bmdma_addr = bmdma;
 		}
 		ata_std_ports(&probe_ent->port[p]);
@@ -867,44 +867,55 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int
 
 
 static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
-				struct ata_port_info *port, int port_num)
+				struct ata_port_info **port, int port_mask)
 {
 	struct ata_probe_ent *probe_ent;
-	unsigned long bmdma;
+	unsigned long bmdma = pci_resource_start(pdev, 4);
 
-	probe_ent = ata_probe_ent_alloc(pci_dev_to_dev(pdev), port);
+	int port_num = 0;
+
+	probe_ent = ata_probe_ent_alloc(pci_dev_to_dev(pdev), port[0]);
 	if (!probe_ent)
 		return NULL;
 
 	probe_ent->legacy_mode = 1;
-	probe_ent->n_ports = 1;
-	probe_ent->hard_port_no = port_num;
-	probe_ent->private_data = port->private_data;
-
-	switch(port_num)
-	{
-		case 0:
-			probe_ent->irq = 14;
-			probe_ent->port[0].cmd_addr = 0x1f0;
-			probe_ent->port[0].altstatus_addr =
-			probe_ent->port[0].ctl_addr = 0x3f6;
-			break;
-		case 1:
+	probe_ent->hard_port_no = 0;
+	probe_ent->private_data = port[0]->private_data;
+
+	if (port_mask & ATA_PORT_PRIMARY) {
+		probe_ent->irq = 14;
+		probe_ent->port[port_num].cmd_addr = ATA_PRIMARY_CMD;
+		probe_ent->port[port_num].altstatus_addr =
+		probe_ent->port[port_num].ctl_addr = ATA_PRIMARY_CTL;
+		if (bmdma) {
+			probe_ent->port[0].bmdma_addr = bmdma;
+			if (inb(bmdma + 2) & 0x80)
+				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+		}
+		ata_std_ports(&probe_ent->port[port_num]);
+		port_num ++;
+	}
+	if (port_mask & ATA_PORT_SECONDARY) {
+		if (port_num == 1)
+			probe_ent->irq2 = 15;
+		else {
+			/* Secondary only. IRQ 15 only and "first" port is port 1 */
 			probe_ent->irq = 15;
-			probe_ent->port[0].cmd_addr = 0x170;
-			probe_ent->port[0].altstatus_addr =
-			probe_ent->port[0].ctl_addr = 0x376;
-			break;
+			probe_ent->hard_port_no = 1;
+		}
+		probe_ent->port[port_num].cmd_addr = ATA_SECONDARY_CMD;
+		probe_ent->port[port_num].altstatus_addr =
+		probe_ent->port[port_num].ctl_addr = ATA_SECONDARY_CTL;
+		if (bmdma) {
+			probe_ent->port[port_num].bmdma_addr = bmdma + 8;
+			if (inb(bmdma + 10) & 0x80)
+				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+		}
+		ata_std_ports(&probe_ent->port[port_num]);
+		port_num ++;
 	}
 
-	bmdma = pci_resource_start(pdev, 4);
-	if (bmdma != 0) {
-		bmdma += 8 * port_num;
-		probe_ent->port[0].bmdma_addr = bmdma;
-		if (inb(bmdma + 2) & 0x80)
-			probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
-	}
-	ata_std_ports(&probe_ent->port[0]);
+	probe_ent->n_ports = port_num;
 
 	return probe_ent;
 }
@@ -924,6 +935,10 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
  *	regions, sets the dma mask, enables bus master mode, and calls
  *	ata_device_add()
  *
+ *	ASSUMPTION:
+ *	Nobody makes a single channel controller that appears solely as
+ *	the secondary legacy port on PCI.
+ *
  *	LOCKING:
  *	Inherited from PCI layer (may sleep).
  *
@@ -934,7 +949,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info,
 		      unsigned int n_ports)
 {
-	struct ata_probe_ent *probe_ent = NULL, *probe_ent2 = NULL;
+	struct ata_probe_ent *probe_ent = NULL;
 	struct ata_port_info *port[2];
 	u8 tmp8, mask;
 	unsigned int legacy_mode = 0;
@@ -983,35 +998,34 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info,
 		goto err_out;
 	}
 
-	/* FIXME: Should use platform specific mappers for legacy port ranges */
 	if (legacy_mode) {
-		if (!request_region(0x1f0, 8, "libata")) {
+		if (!request_region(ATA_PRIMARY_CMD, 8, "libata")) {
 			struct resource *conflict, res;
-			res.start = 0x1f0;
-			res.end = 0x1f0 + 8 - 1;
+			res.start = ATA_PRIMARY_CMD;
+			res.end = ATA_PRIMARY_CMD + 8 - 1;
 			conflict = ____request_resource(&ioport_resource, &res);
 			if (!strcmp(conflict->name, "libata"))
-				legacy_mode |= (1 << 0);
+				legacy_mode |= ATA_PORT_PRIMARY;
 			else {
 				disable_dev_on_err = 0;
-				printk(KERN_WARNING "ata: 0x1f0 IDE port busy\n");
+				printk(KERN_WARNING "ata: 0x%0X IDE port busy\n", ATA_PRIMARY_CMD);
 			}
 		} else
-			legacy_mode |= (1 << 0);
+			legacy_mode |= ATA_PORT_PRIMARY;
 
-		if (!request_region(0x170, 8, "libata")) {
+		if (!request_region(ATA_SECONDARY_CMD, 8, "libata")) {
 			struct resource *conflict, res;
-			res.start = 0x170;
-			res.end = 0x170 + 8 - 1;
+			res.start = ATA_SECONDARY_CMD;
+			res.end = ATA_SECONDARY_CMD + 8 - 1;
 			conflict = ____request_resource(&ioport_resource, &res);
 			if (!strcmp(conflict->name, "libata"))
-				legacy_mode |= (1 << 1);
+				legacy_mode |= ATA_PORT_SECONDARY;
 			else {
 				disable_dev_on_err = 0;
-				printk(KERN_WARNING "ata: 0x170 IDE port busy\n");
+				printk(KERN_WARNING "ata: 0x%X IDE port busy\n", ATA_SECONDARY_CMD);
 			}
 		} else
-			legacy_mode |= (1 << 1);
+			legacy_mode |= ATA_PORT_SECONDARY;
 	}
 
 	/* we have legacy mode, but all ports are unavailable */
@@ -1029,17 +1043,14 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info,
 		goto err_out_regions;
 
 	if (legacy_mode) {
-		if (legacy_mode & (1 << 0))
-			probe_ent = ata_pci_init_legacy_port(pdev, port[0], 0);
-		if (legacy_mode & (1 << 1))
-			probe_ent2 = ata_pci_init_legacy_port(pdev, port[1], 1);
+		probe_ent = ata_pci_init_legacy_port(pdev, port, legacy_mode);
 	} else {
 		if (n_ports == 2)
 			probe_ent = ata_pci_init_native_mode(pdev, port, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
 		else
 			probe_ent = ata_pci_init_native_mode(pdev, port, ATA_PORT_PRIMARY);
 	}
-	if (!probe_ent && !probe_ent2) {
+	if (!probe_ent) {
 		rc = -ENOMEM;
 		goto err_out_regions;
 	}
@@ -1047,35 +1058,17 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info,
 	pci_set_master(pdev);
 
 	/* FIXME: check ata_device_add return */
-	if (legacy_mode) {
-		struct device *dev = &pdev->dev;
-		struct ata_host_set *host_set = NULL;
-
-		if (legacy_mode & (1 << 0)) {
-			ata_device_add(probe_ent);
-			host_set = dev_get_drvdata(dev);
-		}
-
-		if (legacy_mode & (1 << 1)) {
-			ata_device_add(probe_ent2);
-			if (host_set) {
-				host_set->next = dev_get_drvdata(dev);
-				dev_set_drvdata(dev, host_set);
-			}
-		}
-	} else
-		ata_device_add(probe_ent);
+	ata_device_add(probe_ent);
 
 	kfree(probe_ent);
-	kfree(probe_ent2);
 
 	return 0;
 
 err_out_regions:
-	if (legacy_mode & (1 << 0))
-		release_region(0x1f0, 8);
-	if (legacy_mode & (1 << 1))
-		release_region(0x170, 8);
+	if (legacy_mode & ATA_PORT_PRIMARY)
+		release_region(ATA_PRIMARY_CMD, 8);
+	if (legacy_mode & ATA_PORT_SECONDARY)
+		release_region(ATA_SECONDARY_CMD, 8);
 	pci_release_regions(pdev);
 err_out:
 	if (disable_dev_on_err)
diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 4e6c2e8ac0f6..3634279d896d 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -5223,8 +5223,9 @@ void ata_port_init(struct ata_port *ap, struct ata_host_set *host_set,
 	ap->host_set = host_set;
 	ap->dev = ent->dev;
 	ap->port_no = port_no;
-	ap->hard_port_no =
-		ent->legacy_mode ? ent->hard_port_no : port_no;
+	ap->hard_port_no = port_no;
+	if (ent->legacy_mode)
+		ap->hard_port_no += ent->hard_port_no;
 	ap->pio_mask = ent->pio_mask;
 	ap->mwdma_mask = ent->mwdma_mask;
 	ap->udma_mask = ent->udma_mask;
@@ -5400,6 +5401,7 @@ int ata_device_add(const struct ata_probe_ent *ent)
 	ata_host_set_init(host_set, dev, ent->host_set_flags, ent->port_ops);
 	host_set->n_ports = ent->n_ports;
 	host_set->irq = ent->irq;
+	host_set->irq2 = ent->irq2;
 	host_set->mmio_base = ent->mmio_base;
 	host_set->private_data = ent->private_data;
 
@@ -5407,11 +5409,16 @@ int ata_device_add(const struct ata_probe_ent *ent)
 	for (i = 0; i < host_set->n_ports; i++) {
 		struct ata_port *ap;
 		unsigned long xfer_mode_mask;
+		int irq_line = ent->irq;
 
 		ap = ata_port_add(ent, host_set, i);
 		if (!ap)
 			goto err_out;
 
+		/* Report the secondary IRQ for second channel legacy */
+		if (i == 1 && ent->irq2)
+			irq_line = ent->irq2;
+
 		host_set->ports[i] = ap;
 		xfer_mode_mask =(ap->udma_mask << ATA_SHIFT_UDMA) |
 				(ap->mwdma_mask << ATA_SHIFT_MWDMA) |
@@ -5419,20 +5426,20 @@ int ata_device_add(const struct ata_probe_ent *ent)
 
 		/* print per-port info to dmesg */
 		ata_port_printk(ap, KERN_INFO, "%cATA max %s cmd 0x%lX "
-				"ctl 0x%lX bmdma 0x%lX irq %lu\n",
+				"ctl 0x%lX bmdma 0x%lX irq %d\n",
 				ap->flags & ATA_FLAG_SATA ? 'S' : 'P',
 				ata_mode_string(xfer_mode_mask),
 				ap->ioaddr.cmd_addr,
 				ap->ioaddr.ctl_addr,
 				ap->ioaddr.bmdma_addr,
-				ent->irq);
+				irq_line);
 
 		ata_chk_status(ap);
 		host_set->ops->irq_clear(ap);
 		ata_eh_freeze_port(ap);	/* freeze port before requesting IRQ */
 	}
 
-	/* obtain irq, that is shared between channels */
+	/* obtain irq, that may be shared between channels */
 	rc = request_irq(ent->irq, ent->port_ops->irq_handler, ent->irq_flags,
 			 DRV_NAME, host_set);
 	if (rc) {
@@ -5441,6 +5448,21 @@ int ata_device_add(const struct ata_probe_ent *ent)
 		goto err_out;
 	}
 
+	/* do we have a second IRQ for the other channel, eg legacy mode */
+	if (ent->irq2) {
+		/* We will get weird core code crashes later if this is true
+		   so trap it now */
+		BUG_ON(ent->irq == ent->irq2);
+
+		rc = request_irq(ent->irq2, ent->port_ops->irq_handler, ent->irq_flags,
+			 DRV_NAME, host_set);
+		if (rc) {
+			dev_printk(KERN_ERR, dev, "irq %lu request failed: %d\n",
+				   ent->irq2, rc);
+			goto err_out_free_irq;
+		}
+	}
+
 	/* perform each probe synchronously */
 	DPRINTK("probe begin\n");
 	for (i = 0; i < host_set->n_ports; i++) {
@@ -5514,6 +5536,8 @@ int ata_device_add(const struct ata_probe_ent *ent)
 	VPRINTK("EXIT, returning %u\n", ent->n_ports);
 	return ent->n_ports; /* success */
 
+err_out_free_irq:
+	free_irq(ent->irq, host_set);
 err_out:
 	for (i = 0; i < host_set->n_ports; i++) {
 		struct ata_port *ap = host_set->ports[i];
@@ -5605,6 +5629,8 @@ void ata_host_set_remove(struct ata_host_set *host_set)
 		ata_port_detach(host_set->ports[i]);
 
 	free_irq(host_set->irq, host_set);
+	if (host_set->irq2)
+		free_irq(host_set->irq2, host_set);
 
 	for (i = 0; i < host_set->n_ports; i++) {
 		struct ata_port *ap = host_set->ports[i];
@@ -5614,10 +5640,11 @@ void ata_host_set_remove(struct ata_host_set *host_set)
 		if ((ap->flags & ATA_FLAG_NO_LEGACY) == 0) {
 			struct ata_ioports *ioaddr = &ap->ioaddr;
 
-			if (ioaddr->cmd_addr == 0x1f0)
-				release_region(0x1f0, 8);
-			else if (ioaddr->cmd_addr == 0x170)
-				release_region(0x170, 8);
+			/* FIXME: Add -ac IDE pci mods to remove these special cases */
+			if (ioaddr->cmd_addr == ATA_PRIMARY_CMD)
+				release_region(ATA_PRIMARY_CMD, 8);
+			else if (ioaddr->cmd_addr == ATA_SECONDARY_CMD)
+				release_region(ATA_SECONDARY_CMD, 8);
 		}
 
 		scsi_host_put(ap->host);
@@ -5735,11 +5762,8 @@ void ata_pci_remove_one (struct pci_dev *pdev)
 {
 	struct device *dev = pci_dev_to_dev(pdev);
 	struct ata_host_set *host_set = dev_get_drvdata(dev);
-	struct ata_host_set *host_set2 = host_set->next;
 
 	ata_host_set_remove(host_set);
-	if (host_set2)
-		ata_host_set_remove(host_set2);
 
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
@@ -5807,14 +5831,6 @@ int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 	if (rc)
 		return rc;
 
-	if (host_set->next) {
-		rc = ata_host_set_suspend(host_set->next, mesg);
-		if (rc) {
-			ata_host_set_resume(host_set);
-			return rc;
-		}
-	}
-
 	ata_pci_device_do_suspend(pdev, mesg);
 
 	return 0;
@@ -5826,9 +5842,6 @@ int ata_pci_device_resume(struct pci_dev *pdev)
 
 	ata_pci_device_do_resume(pdev);
 	ata_host_set_resume(host_set);
-	if (host_set->next)
-		ata_host_set_resume(host_set->next);
-
 	return 0;
 }
 #endif /* CONFIG_PCI */
diff --git a/include/asm-alpha/libata-portmap.h b/include/asm-alpha/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-alpha/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-generic/libata-portmap.h b/include/asm-generic/libata-portmap.h
new file mode 100644
index 000000000000..9202fd02d5be
--- /dev/null
+++ b/include/asm-generic/libata-portmap.h
@@ -0,0 +1,12 @@
+#ifndef __ASM_GENERIC_LIBATA_PORTMAP_H
+#define __ASM_GENERIC_LIBATA_PORTMAP_H
+
+#define ATA_PRIMARY_CMD		0x1F0
+#define ATA_PRIMARY_CTL		0x3F6
+#define ATA_PRIMARY_IRQ		14
+
+#define ATA_SECONDARY_CMD	0x170
+#define ATA_SECONDARY_CTL	0x376
+#define ATA_SECONDARY_IRQ	15
+
+#endif
diff --git a/include/asm-i386/libata-portmap.h b/include/asm-i386/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-i386/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-ia64/libata-portmap.h b/include/asm-ia64/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-ia64/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-powerpc/libata-portmap.h b/include/asm-powerpc/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-powerpc/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-sparc/libata-portmap.h b/include/asm-sparc/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-sparc/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-sparc64/libata-portmap.h b/include/asm-sparc64/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-sparc64/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-x86_64/libata-portmap.h b/include/asm-x86_64/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-x86_64/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
diff --git a/include/linux/libata.h b/include/linux/libata.h
index cf5eb1da3e32..4504776570e4 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -36,6 +36,8 @@
 #include <linux/workqueue.h>
 #include <scsi/scsi_host.h>
 
+#include <asm/libata-portmap.h>
+
 /*
  * compile-time options: to be removed as soon as all the drivers are
  * converted to the new debugging mechanism
@@ -356,6 +358,7 @@ struct ata_probe_ent {
 	unsigned int		udma_mask;
 	unsigned int		legacy_mode;
 	unsigned long		irq;
+	unsigned long		irq2;
 	unsigned int		irq_flags;
 	unsigned long		host_flags;
 	unsigned long		host_set_flags;
@@ -367,6 +370,7 @@ struct ata_host_set {
 	spinlock_t		lock;
 	struct device 		*dev;
 	unsigned long		irq;
+	unsigned long		irq2;
 	void __iomem		*mmio_base;
 	unsigned int		n_ports;
 	void			*private_data;
@@ -374,7 +378,6 @@ struct ata_host_set {
 	unsigned long		flags;
 	int			simplex_claimed;	/* Keep seperate in case we
 							   ever need to do this locked */
-	struct ata_host_set	*next;		/* for legacy mode */
 	struct ata_port		*ports[0];
 };
 
-- 
cgit v1.2.3


From dd5b06c490de72440ec39f814de99a714a45a1a9 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 10 Aug 2006 16:59:12 +0900
Subject: [PATCH] libata: implement dummy port

Implement dummy port which can be requested by setting appropriate bit
in probe_ent->dummy_port_mask.  The dummy port is used as placeholder
for stolen legacy port.  This allows libata to guarantee that
index_of(ap) == ap->port_no == actual_device_port_no, and thus to
remove error-prone ap->hard_port_no.

As it's used only when one port of a legacy controller is reserved by
some other entity (e.g. IDE), the focus is on keeping the added *code*
complexity at minimum, so dummy port allocates all libata core
resources and acts as a normal port.  It just has all dummy port_ops.

This patch only implements dummy port.  The following patch will make
libata use it for stolen legacy ports.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/scsi/libata-core.c | 61 ++++++++++++++++++++++++++++++++++++++--------
 include/linux/libata.h     |  8 ++++++
 2 files changed, 59 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 3634279d896d..f2e7e2f13db9 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -5311,7 +5311,6 @@ static struct ata_port * ata_port_add(const struct ata_probe_ent *ent,
 {
 	struct Scsi_Host *shost;
 	struct ata_port *ap;
-	int rc;
 
 	DPRINTK("ENTER\n");
 
@@ -5333,15 +5332,7 @@ static struct ata_port * ata_port_add(const struct ata_probe_ent *ent,
 	ata_port_init(ap, host_set, ent, port_no);
 	ata_port_init_shost(ap, shost);
 
-	rc = ap->ops->port_start(ap);
-	if (rc)
-		goto err_out;
-
 	return ap;
-
-err_out:
-	scsi_host_put(shost);
-	return NULL;
 }
 
 /**
@@ -5415,11 +5406,27 @@ int ata_device_add(const struct ata_probe_ent *ent)
 		if (!ap)
 			goto err_out;
 
+		host_set->ports[i] = ap;
+
+		/* dummy? */
+		if (ent->dummy_port_mask & (1 << i)) {
+			ata_port_printk(ap, KERN_INFO, "DUMMY\n");
+			ap->ops = &ata_dummy_port_ops;
+			continue;
+		}
+
+		/* start port */
+		rc = ap->ops->port_start(ap);
+		if (rc) {
+			host_set->ports[i] = NULL;
+			scsi_host_put(ap->host);
+			goto err_out;
+		}
+
 		/* Report the secondary IRQ for second channel legacy */
 		if (i == 1 && ent->irq2)
 			irq_line = ent->irq2;
 
-		host_set->ports[i] = ap;
 		xfer_mode_mask =(ap->udma_mask << ATA_SHIFT_UDMA) |
 				(ap->mwdma_mask << ATA_SHIFT_MWDMA) |
 				(ap->pio_mask << ATA_SHIFT_PIO);
@@ -5940,6 +5947,39 @@ u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
 	return tmp;
 }
 
+/*
+ * Dummy port_ops
+ */
+static void ata_dummy_noret(struct ata_port *ap)	{ }
+static int ata_dummy_ret0(struct ata_port *ap)		{ return 0; }
+static void ata_dummy_qc_noret(struct ata_queued_cmd *qc) { }
+
+static u8 ata_dummy_check_status(struct ata_port *ap)
+{
+	return ATA_DRDY;
+}
+
+static unsigned int ata_dummy_qc_issue(struct ata_queued_cmd *qc)
+{
+	return AC_ERR_SYSTEM;
+}
+
+const struct ata_port_operations ata_dummy_port_ops = {
+	.port_disable		= ata_port_disable,
+	.check_status		= ata_dummy_check_status,
+	.check_altstatus	= ata_dummy_check_status,
+	.dev_select		= ata_noop_dev_select,
+	.qc_prep		= ata_noop_qc_prep,
+	.qc_issue		= ata_dummy_qc_issue,
+	.freeze			= ata_dummy_noret,
+	.thaw			= ata_dummy_noret,
+	.error_handler		= ata_dummy_noret,
+	.post_internal_cmd	= ata_dummy_qc_noret,
+	.irq_clear		= ata_dummy_noret,
+	.port_start		= ata_dummy_ret0,
+	.port_stop		= ata_dummy_noret,
+};
+
 /*
  * libata is essentially a library of internal helper functions for
  * low-level ATA host controller drivers.  As such, the API/ABI is
@@ -5950,6 +5990,7 @@ u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
 EXPORT_SYMBOL_GPL(sata_deb_timing_normal);
 EXPORT_SYMBOL_GPL(sata_deb_timing_hotplug);
 EXPORT_SYMBOL_GPL(sata_deb_timing_long);
+EXPORT_SYMBOL_GPL(ata_dummy_port_ops);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
 EXPORT_SYMBOL_GPL(ata_std_ports);
 EXPORT_SYMBOL_GPL(ata_host_set_init);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 4504776570e4..30bfe8f1666e 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -353,6 +353,7 @@ struct ata_probe_ent {
 	struct ata_ioports	port[ATA_MAX_PORTS];
 	unsigned int		n_ports;
 	unsigned int		hard_port_no;
+	unsigned int		dummy_port_mask;
 	unsigned int		pio_mask;
 	unsigned int		mwdma_mask;
 	unsigned int		udma_mask;
@@ -652,6 +653,8 @@ extern const unsigned long sata_deb_timing_normal[];
 extern const unsigned long sata_deb_timing_hotplug[];
 extern const unsigned long sata_deb_timing_long[];
 
+extern const struct ata_port_operations ata_dummy_port_ops;
+
 static inline const unsigned long *
 sata_ehc_deb_timing(struct ata_eh_context *ehc)
 {
@@ -661,6 +664,11 @@ sata_ehc_deb_timing(struct ata_eh_context *ehc)
 		return sata_deb_timing_normal;
 }
 
+static inline int ata_port_is_dummy(struct ata_port *ap)
+{
+	return ap->ops == &ata_dummy_port_ops;
+}
+
 extern void ata_port_probe(struct ata_port *);
 extern void __sata_phy_reset(struct ata_port *ap);
 extern void sata_phy_reset(struct ata_port *ap);
-- 
cgit v1.2.3


From 4852ba24f647199be797545226c6d325db231937 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Thu, 10 Aug 2006 16:59:18 +0900
Subject: [PATCH] libata: kill unused hard_port_no and legacy_mode

Kill unused probe_ent/ap->hard_port_no and probe_ent->legacy_mode.

Signed-off-by: Tejun Heo <htejun@gmail.com>
---
 drivers/scsi/libata-core.c | 1 -
 include/linux/libata.h     | 3 ---
 2 files changed, 4 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 2ef86b4e7f6d..3f963f206d4a 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -5223,7 +5223,6 @@ void ata_port_init(struct ata_port *ap, struct ata_host_set *host_set,
 	ap->host_set = host_set;
 	ap->dev = ent->dev;
 	ap->port_no = port_no;
-	ap->hard_port_no = port_no;
 	ap->pio_mask = ent->pio_mask;
 	ap->mwdma_mask = ent->mwdma_mask;
 	ap->udma_mask = ent->udma_mask;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 30bfe8f1666e..ed749f778697 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -352,12 +352,10 @@ struct ata_probe_ent {
 	struct scsi_host_template *sht;
 	struct ata_ioports	port[ATA_MAX_PORTS];
 	unsigned int		n_ports;
-	unsigned int		hard_port_no;
 	unsigned int		dummy_port_mask;
 	unsigned int		pio_mask;
 	unsigned int		mwdma_mask;
 	unsigned int		udma_mask;
-	unsigned int		legacy_mode;
 	unsigned long		irq;
 	unsigned long		irq2;
 	unsigned int		irq_flags;
@@ -509,7 +507,6 @@ struct ata_port {
 	unsigned int		pflags; /* ATA_PFLAG_xxx */
 	unsigned int		id;	/* unique id req'd by scsi midlyr */
 	unsigned int		port_no; /* unique port #; from zero */
-	unsigned int		hard_port_no;	/* hardware port #; from zero */
 
 	struct ata_prd		*prd;	 /* our SG list */
 	dma_addr_t		prd_dma; /* and its DMA mapping */
-- 
cgit v1.2.3


From 8b881b0410de0f72a43e814393abf3a4cb29ebb4 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Sun, 11 Jun 2006 09:59:27 -0400
Subject: [ATA] Increase lba48 max-sectors from 200 to 256.

Also, moved ATA_MAX_SECTORS and ATA_MAX_SECTORS_LBA48 from
linux/libata.h to linux/ata.h, now that they truly reflect the standard
(well... mostly; note TODO comment).

This changes the performance profile (and potential bug profile)
for a bunch of drivers, so be wary.
---
 include/linux/ata.h    | 2 ++
 include/linux/libata.h | 2 --
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ata.h b/include/linux/ata.h
index 3671af869696..8d708a3d505b 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -40,6 +40,8 @@ enum {
 	ATA_MAX_DEVICES		= 2,	/* per bus/port */
 	ATA_MAX_PRD		= 256,	/* we could make these 256/256 */
 	ATA_SECT_SIZE		= 512,
+	ATA_MAX_SECTORS		= 256,
+	ATA_MAX_SECTORS_LBA48	= 65535,/* TODO: 65536? */
 
 	ATA_ID_WORDS		= 256,
 	ATA_ID_SERNO_OFS	= 10,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index ed749f778697..060da736b3a8 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -114,8 +114,6 @@ enum {
 	/* tag ATA_MAX_QUEUE - 1 is reserved for internal commands */
 	ATA_MAX_QUEUE		= 32,
 	ATA_TAG_INTERNAL	= ATA_MAX_QUEUE - 1,
-	ATA_MAX_SECTORS		= 200,	/* FIXME */
-	ATA_MAX_SECTORS_LBA48	= 65535,
 	ATA_MAX_BUS		= 2,
 	ATA_DEF_BUSY_WAIT	= 10000,
 	ATA_SHORT_PAUSE		= (HZ >> 6) + 1,
-- 
cgit v1.2.3


From b352e57dc3bb5033996adaa67c2f69b795eddd39 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Thu, 10 Aug 2006 18:52:12 +0100
Subject: [PATCH] libata: Add CompactFlash support

The CFA world has some additional rules and drive modes we need to support for
newer expansion cards and on embedded boxes

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 62 +++++++++++++++++++++++++++++++++++++++++------
 include/linux/ata.h       | 20 ++++++++++++++-
 include/linux/libata.h    |  4 +--
 3 files changed, 75 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 0ac0b519cf2d..9092416a6301 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -386,9 +386,13 @@ static const char *ata_mode_string(unsigned int xfer_mask)
 		"PIO2",
 		"PIO3",
 		"PIO4",
+		"PIO5",
+		"PIO6",
 		"MWDMA0",
 		"MWDMA1",
 		"MWDMA2",
+		"MWDMA3",
+		"MWDMA4",
 		"UDMA/16",
 		"UDMA/25",
 		"UDMA/33",
@@ -875,6 +879,23 @@ static unsigned int ata_id_xfermask(const u16 *id)
 
 	mwdma_mask = id[ATA_ID_MWDMA_MODES] & 0x07;
 
+	if (ata_id_is_cfa(id)) {
+		/*
+		 *	Process compact flash extended modes
+		 */
+		int pio = id[163] & 0x7;
+		int dma = (id[163] >> 3) & 7;
+
+		if (pio)
+			pio_mask |= (1 << 5);
+		if (pio > 1)
+			pio_mask |= (1 << 6);
+		if (dma)
+			mwdma_mask |= (1 << 3);
+		if (dma > 1)
+			mwdma_mask |= (1 << 4);
+	}
+
 	udma_mask = 0;
 	if (id[ATA_ID_FIELD_VALID] & (1 << 2))
 		udma_mask = id[ATA_ID_UDMA_MODES] & 0xff;
@@ -1356,6 +1377,7 @@ int ata_dev_configure(struct ata_device *dev, int print_info)
 	struct ata_port *ap = dev->ap;
 	const u16 *id = dev->id;
 	unsigned int xfer_mask;
+	char revbuf[7];		/* XYZ-99\0 */
 	int rc;
 
 	if (!ata_dev_enabled(dev) && ata_msg_info(ap)) {
@@ -1399,6 +1421,15 @@ int ata_dev_configure(struct ata_device *dev, int print_info)
 
 	/* ATA-specific feature tests */
 	if (dev->class == ATA_DEV_ATA) {
+		if (ata_id_is_cfa(id)) {
+			if (id[162] & 1) /* CPRM may make this media unusable */
+				ata_dev_printk(dev, KERN_WARNING, "ata%u: device %u  supports DRM functions and may not be fully accessable.\n",
+					ap->id, dev->devno);
+			snprintf(revbuf, 7, "CFA");
+		}
+		else
+			snprintf(revbuf, 7, "ATA-%d",  ata_id_major_version(id));
+
 		dev->n_sectors = ata_id_n_sectors(id);
 
 		if (ata_id_has_lba(id)) {
@@ -1417,9 +1448,9 @@ int ata_dev_configure(struct ata_device *dev, int print_info)
 
 			/* print device info to dmesg */
 			if (ata_msg_drv(ap) && print_info)
-				ata_dev_printk(dev, KERN_INFO, "ATA-%d, "
+				ata_dev_printk(dev, KERN_INFO, "%s, "
 					"max %s, %Lu sectors: %s %s\n",
-					ata_id_major_version(id),
+					revbuf,
 					ata_mode_string(xfer_mask),
 					(unsigned long long)dev->n_sectors,
 					lba_desc, ncq_desc);
@@ -1440,9 +1471,9 @@ int ata_dev_configure(struct ata_device *dev, int print_info)
 
 			/* print device info to dmesg */
 			if (ata_msg_drv(ap) && print_info)
-				ata_dev_printk(dev, KERN_INFO, "ATA-%d, "
+				ata_dev_printk(dev, KERN_INFO, "%s, "
 					"max %s, %Lu sectors: CHS %u/%u/%u\n",
-					ata_id_major_version(id),
+					revbuf,
 					ata_mode_string(xfer_mask),
 					(unsigned long long)dev->n_sectors,
 					dev->cylinders, dev->heads,
@@ -1900,10 +1931,11 @@ int sata_set_spd(struct ata_port *ap)
  * drivers/ide/ide-timing.h and was originally written by Vojtech Pavlik
  */
 /*
- * PIO 0-5, MWDMA 0-2 and UDMA 0-6 timings (in nanoseconds).
+ * PIO 0-4, MWDMA 0-2 and UDMA 0-6 timings (in nanoseconds).
  * These were taken from ATA/ATAPI-6 standard, rev 0a, except
- * for PIO 5, which is a nonstandard extension and UDMA6, which
- * is currently supported only by Maxtor drives.
+ * for UDMA6, which is currently supported only by Maxtor drives.
+ *
+ * For PIO 5/6 MWDMA 3/4 see the CFA specification 3.0.
  */
 
 static const struct ata_timing ata_timing[] = {
@@ -1913,6 +1945,8 @@ static const struct ata_timing ata_timing[] = {
 	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
 	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
 
+	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20,  80,   0 },
+	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 100,   0 },
 	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
 	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
 	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
@@ -1927,7 +1961,8 @@ static const struct ata_timing ata_timing[] = {
 	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
 	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
 
-/*	{ XFER_PIO_5,     20,  50,  30, 100,  50,  30, 100,   0 }, */
+	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20,  80,   0 },
+	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 100,   0 },
 	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
 	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
 
@@ -3062,6 +3097,17 @@ static void ata_dev_xfermask(struct ata_device *dev)
 				       dev->mwdma_mask, dev->udma_mask);
 	xfer_mask &= ata_id_xfermask(dev->id);
 
+	/*
+	 *	CFA Advanced TrueIDE timings are not allowed on a shared
+	 *	cable
+	 */
+	if (ata_dev_pair(dev)) {
+		/* No PIO5 or PIO6 */
+		xfer_mask &= ~(0x03 << (ATA_SHIFT_PIO + 5));
+		/* No MWDMA3 or MWDMA 4 */
+		xfer_mask &= ~(0x03 << (ATA_SHIFT_MWDMA + 3));
+	}
+
 	if (ata_dma_blacklisted(dev)) {
 		xfer_mask &= ~(ATA_MASK_MWDMA | ATA_MASK_UDMA);
 		ata_dev_printk(dev, KERN_WARNING,
diff --git a/include/linux/ata.h b/include/linux/ata.h
index 8d708a3d505b..991b858acc30 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -170,12 +170,16 @@ enum {
 	XFER_UDMA_2		= 0x42,
 	XFER_UDMA_1		= 0x41,
 	XFER_UDMA_0		= 0x40,
+	XFER_MW_DMA_4		= 0x24,	/* CFA only */
+	XFER_MW_DMA_3		= 0x23,	/* CFA only */
 	XFER_MW_DMA_2		= 0x22,
 	XFER_MW_DMA_1		= 0x21,
 	XFER_MW_DMA_0		= 0x20,
 	XFER_SW_DMA_2		= 0x12,
 	XFER_SW_DMA_1		= 0x11,
 	XFER_SW_DMA_0		= 0x10,
+	XFER_PIO_6		= 0x0E,	/* CFA only */
+	XFER_PIO_5		= 0x0D,	/* CFA only */
 	XFER_PIO_4		= 0x0C,
 	XFER_PIO_3		= 0x0B,
 	XFER_PIO_2		= 0x0A,
@@ -274,7 +278,6 @@ struct ata_taskfile {
 };
 
 #define ata_id_is_ata(id)	(((id)[0] & (1 << 15)) == 0)
-#define ata_id_is_cfa(id)	((id)[0] == 0x848A)
 #define ata_id_is_sata(id)	((id)[93] == 0)
 #define ata_id_rahead_enabled(id) ((id)[85] & (1 << 6))
 #define ata_id_wcache_enabled(id) ((id)[85] & (1 << 5))
@@ -306,6 +309,9 @@ static inline unsigned int ata_id_major_version(const u16 *id)
 {
 	unsigned int mver;
 
+	if (id[ATA_ID_MAJOR_VER] == 0xFFFF)
+		return 0;
+
 	for (mver = 14; mver >= 1; mver--)
 		if (id[ATA_ID_MAJOR_VER] & (1 << mver))
 			break;
@@ -324,6 +330,18 @@ static inline int ata_id_current_chs_valid(const u16 *id)
 		id[56];    /* sectors in current translation */
 }
 
+static inline int ata_id_is_cfa(const u16 *id)
+{
+	u16 v = id[0];
+	if (v == 0x848A)	/* Standard CF */
+		return 1;
+	/* Could be CF hiding as standard ATA */
+	if (ata_id_major_version(id) >= 3 &&  id[82] != 0xFFFF &&
+			(id[82] & ( 1 << 2)))
+		return 1;
+	return 0;
+}
+
 static inline int atapi_cdb_len(const u16 *dev_id)
 {
 	u16 tmp = dev_id[0] & 0x3;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 060da736b3a8..806682603ac5 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -225,8 +225,8 @@ enum {
 	/* encoding various smaller bitmaps into a single
 	 * unsigned int bitmap
 	 */
-	ATA_BITS_PIO		= 5,
-	ATA_BITS_MWDMA		= 3,
+	ATA_BITS_PIO		= 7,
+	ATA_BITS_MWDMA		= 5,
 	ATA_BITS_UDMA		= 8,
 
 	ATA_SHIFT_PIO		= 0,
-- 
cgit v1.2.3


From 016131b8fffa1085b4ad165ab228116fdc278ebe Mon Sep 17 00:00:00 2001
From: James Smart <James.Smart@Emulex.Com>
Date: Mon, 14 Aug 2006 08:20:25 -0400
Subject: [SCSI] fc transport: convert fc_host symbolic_name attribute to a
 dynamic attribute

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_fc.c | 4 ++--
 include/scsi/scsi_transport_fc.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index b03aa85108e5..c1c5cdffca38 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -815,7 +815,6 @@ fc_private_host_rd_attr_cast(node_name, "0x%llx\n", 20, unsigned long long);
 fc_private_host_rd_attr_cast(port_name, "0x%llx\n", 20, unsigned long long);
 fc_private_host_rd_attr_cast(permanent_port_name, "0x%llx\n", 20,
 			     unsigned long long);
-fc_private_host_rd_attr(symbolic_name, "%s\n", (FC_SYMBOLIC_NAME_SIZE +1));
 fc_private_host_rd_attr(maxframe_size, "%u bytes\n", 20);
 fc_private_host_rd_attr(serial_number, "%s\n", (FC_SERIAL_NUMBER_SIZE +1));
 
@@ -858,6 +857,7 @@ fc_host_rd_attr(port_id, "0x%06x\n", 20);
 fc_host_rd_enum_attr(port_type, FC_PORTTYPE_MAX_NAMELEN);
 fc_host_rd_enum_attr(port_state, FC_PORTSTATE_MAX_NAMELEN);
 fc_host_rd_attr_cast(fabric_name, "0x%llx\n", 20, unsigned long long);
+fc_host_rd_attr(symbolic_name, "%s\n", FC_SYMBOLIC_NAME_SIZE + 1);
 
 
 /* Private Host Attributes */
@@ -1223,7 +1223,6 @@ fc_attach_transport(struct fc_function_template *ft)
 	SETUP_HOST_ATTRIBUTE_RD(permanent_port_name);
 	SETUP_HOST_ATTRIBUTE_RD(supported_classes);
 	SETUP_HOST_ATTRIBUTE_RD(supported_fc4s);
-	SETUP_HOST_ATTRIBUTE_RD(symbolic_name);
 	SETUP_HOST_ATTRIBUTE_RD(supported_speeds);
 	SETUP_HOST_ATTRIBUTE_RD(maxframe_size);
 	SETUP_HOST_ATTRIBUTE_RD(serial_number);
@@ -1234,6 +1233,7 @@ fc_attach_transport(struct fc_function_template *ft)
 	SETUP_HOST_ATTRIBUTE_RD(active_fc4s);
 	SETUP_HOST_ATTRIBUTE_RD(speed);
 	SETUP_HOST_ATTRIBUTE_RD(fabric_name);
+	SETUP_HOST_ATTRIBUTE_RD(symbolic_name);
 
 	/* Transport-managed attributes */
 	SETUP_PRIVATE_HOST_ATTRIBUTE_RW(tgtid_bind_type);
diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h
index 6d28b0317657..b7f62b85f0b3 100644
--- a/include/scsi/scsi_transport_fc.h
+++ b/include/scsi/scsi_transport_fc.h
@@ -409,6 +409,7 @@ struct fc_function_template {
 	void	(*get_host_active_fc4s)(struct Scsi_Host *);
 	void	(*get_host_speed)(struct Scsi_Host *);
 	void	(*get_host_fabric_name)(struct Scsi_Host *);
+	void	(*get_host_symbolic_name)(struct Scsi_Host *);
 
 	struct fc_host_statistics * (*get_fc_host_stats)(struct Scsi_Host *);
 	void	(*reset_fc_host_stats)(struct Scsi_Host *);
@@ -445,7 +446,6 @@ struct fc_function_template {
 	unsigned long	show_host_permanent_port_name:1;
 	unsigned long	show_host_supported_classes:1;
 	unsigned long	show_host_supported_fc4s:1;
-	unsigned long	show_host_symbolic_name:1;
 	unsigned long	show_host_supported_speeds:1;
 	unsigned long	show_host_maxframe_size:1;
 	unsigned long	show_host_serial_number:1;
@@ -456,6 +456,7 @@ struct fc_function_template {
 	unsigned long	show_host_active_fc4s:1;
 	unsigned long	show_host_speed:1;
 	unsigned long	show_host_fabric_name:1;
+	unsigned long	show_host_symbolic_name:1;
 };
 
 
-- 
cgit v1.2.3


From b8d08210126a7b769b857720a59721a453a57a1e Mon Sep 17 00:00:00 2001
From: James Smart <James.Smart@Emulex.Com>
Date: Thu, 17 Aug 2006 08:00:43 -0400
Subject: [SCSI] fc transport: add fc_host system_hostname attribute and
 u64_to_wwn()

This patch updates the fc transport for the following:

- Addition of a new attribute "system_hostname" which can be
  used to set the fully qualified hostname that the fc_host
  is attached to. The fc_host can then register this string
  as the FDMI-based host name attribute.
  Note: for NPIV, a fc_host could be associated with a system which
    is not the local system.

- Add the inline function u64_to_wwn(), which is the inverse of the
  existing wwn_to_u64() function.

- Slight reorg, just to keep dynamic attributes with each other, etc

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_fc.c | 30 ++++++++++++++++++++++++++++--
 include/scsi/scsi_transport_fc.h | 38 +++++++++++++++++++++++++++-----------
 2 files changed, 55 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index c1c5cdffca38..79d31ca2b741 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -301,8 +301,6 @@ static int fc_host_setup(struct transport_container *tc, struct device *dev,
 	fc_host->supported_classes = FC_COS_UNSPECIFIED;
 	memset(fc_host->supported_fc4s, 0,
 		sizeof(fc_host->supported_fc4s));
-	memset(fc_host->symbolic_name, 0,
-		sizeof(fc_host->symbolic_name));
 	fc_host->supported_speeds = FC_PORTSPEED_UNKNOWN;
 	fc_host->maxframe_size = -1;
 	memset(fc_host->serial_number, 0,
@@ -315,6 +313,8 @@ static int fc_host_setup(struct transport_container *tc, struct device *dev,
 		sizeof(fc_host->active_fc4s));
 	fc_host->speed = FC_PORTSPEED_UNKNOWN;
 	fc_host->fabric_name = -1;
+	memset(fc_host->symbolic_name, 0, sizeof(fc_host->symbolic_name));
+	memset(fc_host->system_hostname, 0, sizeof(fc_host->system_hostname));
 
 	fc_host->tgtid_bind_type = FC_TGTID_BIND_BY_WWPN;
 
@@ -688,6 +688,25 @@ store_fc_host_##field(struct class_device *cdev, const char *buf,	\
 	return count;							\
 }
 
+#define fc_host_store_str_function(field, slen)				\
+static ssize_t								\
+store_fc_host_##field(struct class_device *cdev, const char *buf,	\
+			   size_t count)				\
+{									\
+	struct Scsi_Host *shost = transport_class_to_shost(cdev);	\
+	struct fc_internal *i = to_fc_internal(shost->transportt);	\
+	unsigned int cnt=count;						\
+									\
+	/* count may include a LF at end of string */			\
+	if (buf[cnt-1] == '\n')						\
+		cnt--;							\
+	if (cnt > ((slen) - 1))						\
+		return -EINVAL;						\
+	memcpy(fc_host_##field(shost), buf, cnt);			\
+	i->f->set_host_##field(shost);					\
+	return count;							\
+}
+
 #define fc_host_rd_attr(field, format_string, sz)			\
 	fc_host_show_function(field, format_string, sz, )		\
 static FC_CLASS_DEVICE_ATTR(host, field, S_IRUGO,			\
@@ -859,6 +878,12 @@ fc_host_rd_enum_attr(port_state, FC_PORTSTATE_MAX_NAMELEN);
 fc_host_rd_attr_cast(fabric_name, "0x%llx\n", 20, unsigned long long);
 fc_host_rd_attr(symbolic_name, "%s\n", FC_SYMBOLIC_NAME_SIZE + 1);
 
+fc_private_host_show_function(system_hostname, "%s\n",
+		FC_SYMBOLIC_NAME_SIZE + 1, )
+fc_host_store_str_function(system_hostname, FC_SYMBOLIC_NAME_SIZE)
+static FC_CLASS_DEVICE_ATTR(host, system_hostname, S_IRUGO | S_IWUSR,
+		show_fc_host_system_hostname, store_fc_host_system_hostname);
+
 
 /* Private Host Attributes */
 
@@ -1234,6 +1259,7 @@ fc_attach_transport(struct fc_function_template *ft)
 	SETUP_HOST_ATTRIBUTE_RD(speed);
 	SETUP_HOST_ATTRIBUTE_RD(fabric_name);
 	SETUP_HOST_ATTRIBUTE_RD(symbolic_name);
+	SETUP_HOST_ATTRIBUTE_RW(system_hostname);
 
 	/* Transport-managed attributes */
 	SETUP_PRIVATE_HOST_ATTRIBUTE_RW(tgtid_bind_type);
diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h
index b7f62b85f0b3..c74be5dabfeb 100644
--- a/include/scsi/scsi_transport_fc.h
+++ b/include/scsi/scsi_transport_fc.h
@@ -312,7 +312,6 @@ struct fc_host_attrs {
 	u64 permanent_port_name;
 	u32 supported_classes;
 	u8  supported_fc4s[FC_FC4_LIST_SIZE];
-	char symbolic_name[FC_SYMBOLIC_NAME_SIZE];
 	u32 supported_speeds;
 	u32 maxframe_size;
 	char serial_number[FC_SERIAL_NUMBER_SIZE];
@@ -324,6 +323,8 @@ struct fc_host_attrs {
 	u8  active_fc4s[FC_FC4_LIST_SIZE];
 	u32 speed;
 	u64 fabric_name;
+	char symbolic_name[FC_SYMBOLIC_NAME_SIZE];
+	char system_hostname[FC_SYMBOLIC_NAME_SIZE];
 
 	/* Private (Transport-managed) Attributes */
 	enum fc_tgtid_binding_type  tgtid_bind_type;
@@ -354,8 +355,6 @@ struct fc_host_attrs {
 	(((struct fc_host_attrs *)(x)->shost_data)->supported_classes)
 #define fc_host_supported_fc4s(x)	\
 	(((struct fc_host_attrs *)(x)->shost_data)->supported_fc4s)
-#define fc_host_symbolic_name(x)	\
-	(((struct fc_host_attrs *)(x)->shost_data)->symbolic_name)
 #define fc_host_supported_speeds(x)	\
 	(((struct fc_host_attrs *)(x)->shost_data)->supported_speeds)
 #define fc_host_maxframe_size(x)	\
@@ -374,6 +373,10 @@ struct fc_host_attrs {
 	(((struct fc_host_attrs *)(x)->shost_data)->speed)
 #define fc_host_fabric_name(x)	\
 	(((struct fc_host_attrs *)(x)->shost_data)->fabric_name)
+#define fc_host_symbolic_name(x)	\
+	(((struct fc_host_attrs *)(x)->shost_data)->symbolic_name)
+#define fc_host_system_hostname(x)	\
+	(((struct fc_host_attrs *)(x)->shost_data)->system_hostname)
 #define fc_host_tgtid_bind_type(x) \
 	(((struct fc_host_attrs *)(x)->shost_data)->tgtid_bind_type)
 #define fc_host_rports(x) \
@@ -410,6 +413,7 @@ struct fc_function_template {
 	void	(*get_host_speed)(struct Scsi_Host *);
 	void	(*get_host_fabric_name)(struct Scsi_Host *);
 	void	(*get_host_symbolic_name)(struct Scsi_Host *);
+	void	(*set_host_system_hostname)(struct Scsi_Host *);
 
 	struct fc_host_statistics * (*get_fc_host_stats)(struct Scsi_Host *);
 	void	(*reset_fc_host_stats)(struct Scsi_Host *);
@@ -457,6 +461,7 @@ struct fc_function_template {
 	unsigned long	show_host_speed:1;
 	unsigned long	show_host_fabric_name:1;
 	unsigned long	show_host_symbolic_name:1;
+	unsigned long	show_host_system_hostname:1;
 };
 
 
@@ -492,6 +497,25 @@ fc_remote_port_chkready(struct fc_rport *rport)
 	return result;
 }
 
+static inline u64 wwn_to_u64(u8 *wwn)
+{
+	return (u64)wwn[0] << 56 | (u64)wwn[1] << 48 |
+	    (u64)wwn[2] << 40 | (u64)wwn[3] << 32 |
+	    (u64)wwn[4] << 24 | (u64)wwn[5] << 16 |
+	    (u64)wwn[6] <<  8 | (u64)wwn[7];
+}
+
+static inline void u64_to_wwn(u64 inm, u8 *wwn)
+{
+	wwn[0] = (inm >> 56) & 0xff;
+	wwn[1] = (inm >> 48) & 0xff;
+	wwn[2] = (inm >> 40) & 0xff;
+	wwn[3] = (inm >> 32) & 0xff;
+	wwn[4] = (inm >> 24) & 0xff;
+	wwn[5] = (inm >> 16) & 0xff;
+	wwn[6] = (inm >> 8) & 0xff;
+	wwn[7] = inm & 0xff;
+}
 
 struct scsi_transport_template *fc_attach_transport(
 			struct fc_function_template *);
@@ -503,12 +527,4 @@ void fc_remote_port_delete(struct fc_rport  *rport);
 void fc_remote_port_rolechg(struct fc_rport  *rport, u32 roles);
 int scsi_is_fc_rport(const struct device *);
 
-static inline u64 wwn_to_u64(u8 *wwn)
-{
-	return (u64)wwn[0] << 56 | (u64)wwn[1] << 48 |
-	    (u64)wwn[2] << 40 | (u64)wwn[3] << 32 |
-	    (u64)wwn[4] << 24 | (u64)wwn[5] << 16 |
-	    (u64)wwn[6] <<  8 | (u64)wwn[7];
-}
-
 #endif /* SCSI_TRANSPORT_FC_H */
-- 
cgit v1.2.3


From cca3974e48607c3775dc73b544a5700b2e37c21a Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Thu, 24 Aug 2006 03:19:22 -0400
Subject: libata: Grand renaming.

The biggest change is that ata_host_set is renamed to ata_host.

* ata_host_set			=> ata_host
* ata_probe_ent->host_flags	=> ata_probe_ent->port_flags
* ata_probe_ent->host_set_flags	=> ata_probe_ent->_host_flags
* ata_host_stats		=> ata_port_stats
* ata_port->host		=> ata_port->scsi_host
* ata_port->host_set		=> ata_port->host
* ata_port_info->host_flags	=> ata_port_info->flags
* ata_(.*)host_set(.*)\(\)	=> ata_\1host\2()

The leading underscore in ata_probe_ent->_host_flags is to avoid
reusing ->host_flags for different purpose.  Currently, the only user
of the field is libata-bmdma.c and probe_ent itself is scheduled to be
removed.

ata_port->host is reused for different purpose but this field is used
inside libata core proper and of different type.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ahci.c         |  98 +++++++--------
 drivers/ata/ata_piix.c     |  56 ++++-----
 drivers/ata/libata-bmdma.c |  34 +++---
 drivers/ata/libata-core.c  | 292 ++++++++++++++++++++++-----------------------
 drivers/ata/libata-eh.c    |  16 +--
 drivers/ata/libata-scsi.c  |  78 ++++++------
 drivers/ata/libata.h       |   2 +-
 drivers/ata/pdc_adma.c     |  46 +++----
 drivers/ata/sata_mv.c      |  98 ++++++++-------
 drivers/ata/sata_nv.c      |  58 ++++-----
 drivers/ata/sata_promise.c |  54 ++++-----
 drivers/ata/sata_qstor.c   |  44 +++----
 drivers/ata/sata_sil.c     |  47 ++++----
 drivers/ata/sata_sil24.c   |  59 +++++----
 drivers/ata/sata_sis.c     |  18 +--
 drivers/ata/sata_svw.c     |  10 +-
 drivers/ata/sata_sx4.c     |  64 +++++-----
 drivers/ata/sata_uli.c     |   8 +-
 drivers/ata/sata_via.c     |   4 +-
 drivers/ata/sata_vsc.c     |  16 +--
 include/linux/libata.h     |  39 +++---
 21 files changed, 568 insertions(+), 573 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 813031c01fba..3f1106fdaed1 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -277,7 +277,7 @@ static const struct ata_port_info ahci_port_info[] = {
 	/* board_ahci */
 	{
 		.sht		= &ahci_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA |
 				  ATA_FLAG_SKIP_D2H_BSY,
 		.pio_mask	= 0x1f, /* pio0-4 */
@@ -287,7 +287,7 @@ static const struct ata_port_info ahci_port_info[] = {
 	/* board_ahci_vt8251 */
 	{
 		.sht		= &ahci_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA |
 				  ATA_FLAG_SKIP_D2H_BSY |
 				  AHCI_FLAG_RESET_NEEDS_CLO | AHCI_FLAG_NO_NCQ,
@@ -709,7 +709,7 @@ static void ahci_fill_cmd_slot(struct ahci_port_priv *pp, unsigned int tag,
 static int ahci_clo(struct ata_port *ap)
 {
 	void __iomem *port_mmio = (void __iomem *) ap->ioaddr.cmd_addr;
-	struct ahci_host_priv *hpriv = ap->host_set->private_data;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
 	u32 tmp;
 
 	if (!(hpriv->cap & HOST_CAP_CLO))
@@ -741,7 +741,7 @@ static int ahci_prereset(struct ata_port *ap)
 static int ahci_softreset(struct ata_port *ap, unsigned int *class)
 {
 	struct ahci_port_priv *pp = ap->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	const u32 cmd_fis_len = 5; /* five dwords */
 	const char *reason = NULL;
@@ -850,7 +850,7 @@ static int ahci_hardreset(struct ata_port *ap, unsigned int *class)
 	struct ahci_port_priv *pp = ap->private_data;
 	u8 *d2h_fis = pp->rx_fis + RX_FIS_D2H_REG;
 	struct ata_taskfile tf;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	int rc;
 
@@ -1039,7 +1039,7 @@ static void ahci_error_intr(struct ata_port *ap, u32 irq_stat)
 
 static void ahci_host_intr(struct ata_port *ap)
 {
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	struct ata_eh_info *ehi = &ap->eh_info;
 	u32 status, qc_active;
@@ -1091,7 +1091,7 @@ static void ahci_irq_clear(struct ata_port *ap)
 
 static irqreturn_t ahci_interrupt(int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	struct ahci_host_priv *hpriv;
 	unsigned int i, handled = 0;
 	void __iomem *mmio;
@@ -1099,8 +1099,8 @@ static irqreturn_t ahci_interrupt(int irq, void *dev_instance, struct pt_regs *r
 
 	VPRINTK("ENTER\n");
 
-	hpriv = host_set->private_data;
-	mmio = host_set->mmio_base;
+	hpriv = host->private_data;
+	mmio = host->mmio_base;
 
 	/* sigh.  0xffffffff is a valid return from h/w */
 	irq_stat = readl(mmio + HOST_IRQ_STAT);
@@ -1108,22 +1108,22 @@ static irqreturn_t ahci_interrupt(int irq, void *dev_instance, struct pt_regs *r
 	if (!irq_stat)
 		return IRQ_NONE;
 
-        spin_lock(&host_set->lock);
+        spin_lock(&host->lock);
 
-        for (i = 0; i < host_set->n_ports; i++) {
+        for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap;
 
 		if (!(irq_stat & (1 << i)))
 			continue;
 
-		ap = host_set->ports[i];
+		ap = host->ports[i];
 		if (ap) {
 			ahci_host_intr(ap);
 			VPRINTK("port %u\n", i);
 		} else {
 			VPRINTK("port %u (no irq)\n", i);
 			if (ata_ratelimit())
-				dev_printk(KERN_WARNING, host_set->dev,
+				dev_printk(KERN_WARNING, host->dev,
 					"interrupt on disabled port %u\n", i);
 		}
 
@@ -1135,7 +1135,7 @@ static irqreturn_t ahci_interrupt(int irq, void *dev_instance, struct pt_regs *r
 		handled = 1;
 	}
 
-	spin_unlock(&host_set->lock);
+	spin_unlock(&host->lock);
 
 	VPRINTK("EXIT\n");
 
@@ -1157,7 +1157,7 @@ static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc)
 
 static void ahci_freeze(struct ata_port *ap)
 {
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 
 	/* turn IRQ off */
@@ -1166,7 +1166,7 @@ static void ahci_freeze(struct ata_port *ap)
 
 static void ahci_thaw(struct ata_port *ap)
 {
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	u32 tmp;
 
@@ -1181,7 +1181,7 @@ static void ahci_thaw(struct ata_port *ap)
 
 static void ahci_error_handler(struct ata_port *ap)
 {
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 
 	if (!(ap->pflags & ATA_PFLAG_FROZEN)) {
@@ -1198,7 +1198,7 @@ static void ahci_error_handler(struct ata_port *ap)
 static void ahci_post_internal_cmd(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 
 	if (qc->flags & ATA_QCFLAG_FAILED)
@@ -1213,9 +1213,9 @@ static void ahci_post_internal_cmd(struct ata_queued_cmd *qc)
 
 static int ahci_port_suspend(struct ata_port *ap, pm_message_t mesg)
 {
-	struct ahci_host_priv *hpriv = ap->host_set->private_data;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
 	struct ahci_port_priv *pp = ap->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	const char *emsg = NULL;
 	int rc;
@@ -1233,8 +1233,8 @@ static int ahci_port_suspend(struct ata_port *ap, pm_message_t mesg)
 static int ahci_port_resume(struct ata_port *ap)
 {
 	struct ahci_port_priv *pp = ap->private_data;
-	struct ahci_host_priv *hpriv = ap->host_set->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 
 	ahci_init_port(port_mmio, hpriv->cap, pp->cmd_slot_dma, pp->rx_fis_dma);
@@ -1244,8 +1244,8 @@ static int ahci_port_resume(struct ata_port *ap)
 
 static int ahci_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
-	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
-	void __iomem *mmio = host_set->mmio_base;
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
+	void __iomem *mmio = host->mmio_base;
 	u32 ctl;
 
 	if (mesg.event == PM_EVENT_SUSPEND) {
@@ -1264,9 +1264,9 @@ static int ahci_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 
 static int ahci_pci_device_resume(struct pci_dev *pdev)
 {
-	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
-	struct ahci_host_priv *hpriv = host_set->private_data;
-	void __iomem *mmio = host_set->mmio_base;
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
+	struct ahci_host_priv *hpriv = host->private_data;
+	void __iomem *mmio = host->mmio_base;
 	int rc;
 
 	ata_pci_device_do_resume(pdev);
@@ -1276,20 +1276,20 @@ static int ahci_pci_device_resume(struct pci_dev *pdev)
 		if (rc)
 			return rc;
 
-		ahci_init_controller(mmio, pdev, host_set->n_ports, hpriv->cap);
+		ahci_init_controller(mmio, pdev, host->n_ports, hpriv->cap);
 	}
 
-	ata_host_set_resume(host_set);
+	ata_host_resume(host);
 
 	return 0;
 }
 
 static int ahci_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
-	struct ahci_host_priv *hpriv = ap->host_set->private_data;
+	struct device *dev = ap->host->dev;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
 	struct ahci_port_priv *pp;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	void *mem;
 	dma_addr_t mem_dma;
@@ -1350,10 +1350,10 @@ static int ahci_port_start(struct ata_port *ap)
 
 static void ahci_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
-	struct ahci_host_priv *hpriv = ap->host_set->private_data;
+	struct device *dev = ap->host->dev;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
 	struct ahci_port_priv *pp = ap->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	void __iomem *mmio = ap->host->mmio_base;
 	void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no);
 	const char *emsg = NULL;
 	int rc;
@@ -1581,7 +1581,7 @@ static int ahci_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	memset(hpriv, 0, sizeof(*hpriv));
 
 	probe_ent->sht		= ahci_port_info[board_idx].sht;
-	probe_ent->host_flags	= ahci_port_info[board_idx].host_flags;
+	probe_ent->port_flags	= ahci_port_info[board_idx].flags;
 	probe_ent->pio_mask	= ahci_port_info[board_idx].pio_mask;
 	probe_ent->udma_mask	= ahci_port_info[board_idx].udma_mask;
 	probe_ent->port_ops	= ahci_port_info[board_idx].port_ops;
@@ -1599,9 +1599,9 @@ static int ahci_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc)
 		goto err_out_hpriv;
 
-	if (!(probe_ent->host_flags & AHCI_FLAG_NO_NCQ) &&
+	if (!(probe_ent->port_flags & AHCI_FLAG_NO_NCQ) &&
 	    (hpriv->cap & HOST_CAP_NCQ))
-		probe_ent->host_flags |= ATA_FLAG_NCQ;
+		probe_ent->port_flags |= ATA_FLAG_NCQ;
 
 	ahci_print_info(probe_ent);
 
@@ -1632,27 +1632,27 @@ err_out:
 static void ahci_remove_one (struct pci_dev *pdev)
 {
 	struct device *dev = pci_dev_to_dev(pdev);
-	struct ata_host_set *host_set = dev_get_drvdata(dev);
-	struct ahci_host_priv *hpriv = host_set->private_data;
+	struct ata_host *host = dev_get_drvdata(dev);
+	struct ahci_host_priv *hpriv = host->private_data;
 	unsigned int i;
 	int have_msi;
 
-	for (i = 0; i < host_set->n_ports; i++)
-		ata_port_detach(host_set->ports[i]);
+	for (i = 0; i < host->n_ports; i++)
+		ata_port_detach(host->ports[i]);
 
 	have_msi = hpriv->flags & AHCI_FLAG_MSI;
-	free_irq(host_set->irq, host_set);
+	free_irq(host->irq, host);
 
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 
-		ata_scsi_release(ap->host);
-		scsi_host_put(ap->host);
+		ata_scsi_release(ap->scsi_host);
+		scsi_host_put(ap->scsi_host);
 	}
 
 	kfree(hpriv);
-	pci_iounmap(pdev, host_set->mmio_base);
-	kfree(host_set);
+	pci_iounmap(pdev, host->mmio_base);
+	kfree(host);
 
 	if (have_msi)
 		pci_disable_msi(pdev);
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 0ca4c3b78dc5..22b2dba90b9a 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -151,7 +151,7 @@ struct piix_host_priv {
 
 static int piix_init_one (struct pci_dev *pdev,
 				    const struct pci_device_id *ent);
-static void piix_host_stop(struct ata_host_set *host_set);
+static void piix_host_stop(struct ata_host *host);
 static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev);
 static void piix_set_dmamode (struct ata_port *ap, struct ata_device *adev);
 static void piix_pata_error_handler(struct ata_port *ap);
@@ -362,7 +362,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* piix4_pata */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SLAVE_POSS,
+		.flags		= ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 #if 0
 		.mwdma_mask	= 0x06, /* mwdma1-2 */
@@ -376,7 +376,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* ich5_pata */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SLAVE_POSS | PIIX_FLAG_CHECKINTR,
+		.flags		= ATA_FLAG_SLAVE_POSS | PIIX_FLAG_CHECKINTR,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 #if 0
 		.mwdma_mask	= 0x06, /* mwdma1-2 */
@@ -390,7 +390,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* ich5_sata */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SATA | PIIX_FLAG_CHECKINTR |
+		.flags		= ATA_FLAG_SATA | PIIX_FLAG_CHECKINTR |
 				  PIIX_FLAG_IGNORE_PCS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -401,7 +401,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* i6300esb_sata */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SATA |
+		.flags		= ATA_FLAG_SATA |
 				  PIIX_FLAG_CHECKINTR | PIIX_FLAG_IGNORE_PCS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -412,7 +412,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* ich6_sata */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SATA |
+		.flags		= ATA_FLAG_SATA |
 				  PIIX_FLAG_CHECKINTR | PIIX_FLAG_SCR,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
@@ -423,7 +423,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* ich6_sata_ahci */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SATA |
+		.flags		= ATA_FLAG_SATA |
 				  PIIX_FLAG_CHECKINTR | PIIX_FLAG_SCR |
 				  PIIX_FLAG_AHCI,
 		.pio_mask	= 0x1f,	/* pio0-4 */
@@ -435,7 +435,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* ich6m_sata_ahci */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SATA |
+		.flags		= ATA_FLAG_SATA |
 				  PIIX_FLAG_CHECKINTR | PIIX_FLAG_SCR |
 				  PIIX_FLAG_AHCI,
 		.pio_mask	= 0x1f,	/* pio0-4 */
@@ -447,7 +447,7 @@ static struct ata_port_info piix_port_info[] = {
 	/* ich8_sata_ahci */
 	{
 		.sht		= &piix_sht,
-		.host_flags	= ATA_FLAG_SATA |
+		.flags		= ATA_FLAG_SATA |
 				  PIIX_FLAG_CHECKINTR | PIIX_FLAG_SCR |
 				  PIIX_FLAG_AHCI,
 		.pio_mask	= 0x1f,	/* pio0-4 */
@@ -485,7 +485,7 @@ MODULE_PARM_DESC(force_pcs, "force honoring or ignoring PCS to work around "
  */
 static void piix_pata_cbl_detect(struct ata_port *ap)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 tmp, mask;
 
 	/* no 80c support in host controller? */
@@ -517,7 +517,7 @@ cbl40:
  */
 static int piix_pata_prereset(struct ata_port *ap)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 
 	if (!pci_test_config_bits(pdev, &piix_enable_bits[ap->port_no])) {
 		ata_port_printk(ap, KERN_INFO, "port disabled. ignoring.\n");
@@ -551,8 +551,8 @@ static void piix_pata_error_handler(struct ata_port *ap)
  */
 static unsigned int piix_sata_present_mask(struct ata_port *ap)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
-	struct piix_host_priv *hpriv = ap->host_set->private_data;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct piix_host_priv *hpriv = ap->host->private_data;
 	const unsigned int *map = hpriv->map;
 	int base = 2 * ap->port_no;
 	unsigned int present_mask = 0;
@@ -631,7 +631,7 @@ static void piix_sata_error_handler(struct ata_port *ap)
 static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
 {
 	unsigned int pio	= adev->pio_mode - XFER_PIO_0;
-	struct pci_dev *dev	= to_pci_dev(ap->host_set->dev);
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
 	unsigned int is_slave	= (adev->devno != 0);
 	unsigned int master_port= ap->port_no ? 0x42 : 0x40;
 	unsigned int slave_port	= 0x44;
@@ -683,7 +683,7 @@ static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
 static void piix_set_dmamode (struct ata_port *ap, struct ata_device *adev)
 {
 	unsigned int udma	= adev->dma_mode; /* FIXME: MWDMA too */
-	struct pci_dev *dev	= to_pci_dev(ap->host_set->dev);
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
 	u8 maslave		= ap->port_no ? 0x42 : 0x40;
 	u8 speed		= udma;
 	unsigned int drive_dn	= (ap->port_no ? 2 : 0) + adev->devno;
@@ -835,13 +835,13 @@ static void __devinit piix_init_pcs(struct pci_dev *pdev,
 	if (force_pcs == 1) {
 		dev_printk(KERN_INFO, &pdev->dev,
 			   "force ignoring PCS (0x%x)\n", new_pcs);
-		pinfo[0].host_flags |= PIIX_FLAG_IGNORE_PCS;
-		pinfo[1].host_flags |= PIIX_FLAG_IGNORE_PCS;
+		pinfo[0].flags |= PIIX_FLAG_IGNORE_PCS;
+		pinfo[1].flags |= PIIX_FLAG_IGNORE_PCS;
 	} else if (force_pcs == 2) {
 		dev_printk(KERN_INFO, &pdev->dev,
 			   "force honoring PCS (0x%x)\n", new_pcs);
-		pinfo[0].host_flags &= ~PIIX_FLAG_IGNORE_PCS;
-		pinfo[1].host_flags &= ~PIIX_FLAG_IGNORE_PCS;
+		pinfo[0].flags &= ~PIIX_FLAG_IGNORE_PCS;
+		pinfo[1].flags &= ~PIIX_FLAG_IGNORE_PCS;
 	}
 }
 
@@ -881,7 +881,7 @@ static void __devinit piix_init_sata_map(struct pci_dev *pdev,
 		default:
 			printk(" P%d", map[i]);
 			if (i & 1)
-				pinfo[i / 2].host_flags |= ATA_FLAG_SLAVE_POSS;
+				pinfo[i / 2].flags |= ATA_FLAG_SLAVE_POSS;
 			break;
 		}
 	}
@@ -916,7 +916,7 @@ static int piix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct ata_port_info port_info[2];
 	struct ata_port_info *ppinfo[2] = { &port_info[0], &port_info[1] };
 	struct piix_host_priv *hpriv;
-	unsigned long host_flags;
+	unsigned long port_flags;
 
 	if (!printed_version++)
 		dev_printk(KERN_DEBUG, &pdev->dev,
@@ -935,9 +935,9 @@ static int piix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	port_info[0].private_data = hpriv;
 	port_info[1].private_data = hpriv;
 
-	host_flags = port_info[0].host_flags;
+	port_flags = port_info[0].flags;
 
-	if (host_flags & PIIX_FLAG_AHCI) {
+	if (port_flags & PIIX_FLAG_AHCI) {
 		u8 tmp;
 		pci_read_config_byte(pdev, PIIX_SCC, &tmp);
 		if (tmp == PIIX_AHCI_DEVICE) {
@@ -948,7 +948,7 @@ static int piix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	/* Initialize SATA map */
-	if (host_flags & ATA_FLAG_SATA) {
+	if (port_flags & ATA_FLAG_SATA) {
 		piix_init_sata_map(pdev, port_info,
 				   piix_map_db_table[ent->driver_data]);
 		piix_init_pcs(pdev, port_info,
@@ -961,7 +961,7 @@ static int piix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * MSI is disabled (and it is disabled, as we don't use
 	 * message-signalled interrupts currently).
 	 */
-	if (host_flags & PIIX_FLAG_CHECKINTR)
+	if (port_flags & PIIX_FLAG_CHECKINTR)
 		pci_intx(pdev, 1);
 
 	if (piix_check_450nx_errata(pdev)) {
@@ -976,11 +976,11 @@ static int piix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	return ata_pci_init_one(pdev, ppinfo, 2);
 }
 
-static void piix_host_stop(struct ata_host_set *host_set)
+static void piix_host_stop(struct ata_host *host)
 {
-	struct piix_host_priv *hpriv = host_set->private_data;
+	struct piix_host_priv *hpriv = host->private_data;
 
-	ata_host_stop(host_set);
+	ata_host_stop(host);
 
 	kfree(hpriv);
 }
diff --git a/drivers/ata/libata-bmdma.c b/drivers/ata/libata-bmdma.c
index 158f62dbf21b..760502859821 100644
--- a/drivers/ata/libata-bmdma.c
+++ b/drivers/ata/libata-bmdma.c
@@ -193,7 +193,7 @@ void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf)
  *	synchronization with interrupt handler / other threads.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_exec_command_pio(struct ata_port *ap, const struct ata_taskfile *tf)
@@ -216,7 +216,7 @@ static void ata_exec_command_pio(struct ata_port *ap, const struct ata_taskfile
  *	FIXME: missing write posting for 400nS delay enforcement
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_exec_command_mmio(struct ata_port *ap, const struct ata_taskfile *tf)
@@ -237,7 +237,7 @@ static void ata_exec_command_mmio(struct ata_port *ap, const struct ata_taskfile
  *	synchronization with interrupt handler / other threads.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf)
 {
@@ -422,7 +422,7 @@ u8 ata_altstatus(struct ata_port *ap)
  *	@qc: Info associated with this ATA transaction.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_bmdma_setup_mmio (struct ata_queued_cmd *qc)
@@ -452,7 +452,7 @@ static void ata_bmdma_setup_mmio (struct ata_queued_cmd *qc)
  *	@qc: Info associated with this ATA transaction.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_bmdma_start_mmio (struct ata_queued_cmd *qc)
@@ -483,7 +483,7 @@ static void ata_bmdma_start_mmio (struct ata_queued_cmd *qc)
  *	@qc: Info associated with this ATA transaction.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_bmdma_setup_pio (struct ata_queued_cmd *qc)
@@ -511,7 +511,7 @@ static void ata_bmdma_setup_pio (struct ata_queued_cmd *qc)
  *	@qc: Info associated with this ATA transaction.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_bmdma_start_pio (struct ata_queued_cmd *qc)
@@ -535,7 +535,7 @@ static void ata_bmdma_start_pio (struct ata_queued_cmd *qc)
  *	May be used as the bmdma_start() entry in ata_port_operations.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_bmdma_start(struct ata_queued_cmd *qc)
 {
@@ -557,7 +557,7 @@ void ata_bmdma_start(struct ata_queued_cmd *qc)
  *	May be used as the bmdma_setup() entry in ata_port_operations.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_bmdma_setup(struct ata_queued_cmd *qc)
 {
@@ -577,7 +577,7 @@ void ata_bmdma_setup(struct ata_queued_cmd *qc)
  *	May be used as the irq_clear() entry in ata_port_operations.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_bmdma_irq_clear(struct ata_port *ap)
@@ -605,7 +605,7 @@ void ata_bmdma_irq_clear(struct ata_port *ap)
  *	May be used as the bmdma_status() entry in ata_port_operations.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 u8 ata_bmdma_status(struct ata_port *ap)
@@ -629,7 +629,7 @@ u8 ata_bmdma_status(struct ata_port *ap)
  *	May be used as the bmdma_stop() entry in ata_port_operations.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_bmdma_stop(struct ata_queued_cmd *qc)
@@ -838,7 +838,7 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int
 		bmdma = pci_resource_start(pdev, 4);
 		if (bmdma) {
 			if (inb(bmdma + 2) & 0x80)
-				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+				probe_ent->_host_flags |= ATA_HOST_SIMPLEX;
 			probe_ent->port[p].bmdma_addr = bmdma;
 		}
 		ata_std_ports(&probe_ent->port[p]);
@@ -854,7 +854,7 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int
 		if (bmdma) {
 			bmdma += 8;
 			if(inb(bmdma + 2) & 0x80)
-				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+				probe_ent->_host_flags |= ATA_HOST_SIMPLEX;
 			probe_ent->port[p].bmdma_addr = bmdma;
 		}
 		ata_std_ports(&probe_ent->port[p]);
@@ -887,7 +887,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 		if (bmdma) {
 			probe_ent->port[0].bmdma_addr = bmdma;
 			if (inb(bmdma + 2) & 0x80)
-				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+				probe_ent->_host_flags |= ATA_HOST_SIMPLEX;
 		}
 		ata_std_ports(&probe_ent->port[0]);
 	} else
@@ -904,7 +904,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 		if (bmdma) {
 			probe_ent->port[1].bmdma_addr = bmdma + 8;
 			if (inb(bmdma + 10) & 0x80)
-				probe_ent->host_set_flags |= ATA_HOST_SIMPLEX;
+				probe_ent->_host_flags |= ATA_HOST_SIMPLEX;
 		}
 		ata_std_ports(&probe_ent->port[1]);
 	} else
@@ -957,7 +957,7 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info,
 	else
 		port[1] = port[0];
 
-	if ((port[0]->host_flags & ATA_FLAG_NO_LEGACY) == 0
+	if ((port[0]->flags & ATA_FLAG_NO_LEGACY) == 0
 	    && (pdev->class >> 8) == PCI_CLASS_STORAGE_IDE) {
 		/* TODO: What if one channel is in native mode ... */
 		pci_read_config_byte(pdev, PCI_CLASS_PROG, &tmp8);
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 9092416a6301..1c9315401f7a 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1335,7 +1335,7 @@ static void ata_dev_config_ncq(struct ata_device *dev,
 	}
 
 	if (ap->flags & ATA_FLAG_NCQ) {
-		hdepth = min(ap->host->can_queue, ATA_MAX_QUEUE - 1);
+		hdepth = min(ap->scsi_host->can_queue, ATA_MAX_QUEUE - 1);
 		dev->flags |= ATA_DFLAG_NCQ;
 	}
 
@@ -1349,12 +1349,13 @@ static void ata_set_port_max_cmd_len(struct ata_port *ap)
 {
 	int i;
 
-	if (ap->host) {
-		ap->host->max_cmd_len = 0;
+	if (ap->scsi_host) {
+		unsigned int len = 0;
+
 		for (i = 0; i < ATA_MAX_DEVICES; i++)
-			ap->host->max_cmd_len = max_t(unsigned int,
-						      ap->host->max_cmd_len,
-						      ap->device[i].cdb_len);
+			len = max(len, ap->device[i].cdb_len);
+
+		ap->scsi_host->max_cmd_len = len;
 	}
 }
 
@@ -1662,7 +1663,7 @@ int ata_bus_probe(struct ata_port *ap)
  *	Modify @ap data structure such that the system
  *	thinks that the entire port is enabled.
  *
- *	LOCKING: host_set lock, or some other form of
+ *	LOCKING: host lock, or some other form of
  *	serialization.
  */
 
@@ -1800,7 +1801,7 @@ struct ata_device *ata_dev_pair(struct ata_device *adev)
  *	never attempt to probe or communicate with devices
  *	on this port.
  *
- *	LOCKING: host_set lock, or some other form of
+ *	LOCKING: host lock, or some other form of
  *	serialization.
  */
 
@@ -2258,8 +2259,8 @@ int ata_set_mode(struct ata_port *ap, struct ata_device **r_failed_dev)
 	/* Record simplex status. If we selected DMA then the other
 	 * host channels are not permitted to do so.
 	 */
-	if (used_dma && (ap->host_set->flags & ATA_HOST_SIMPLEX))
-		ap->host_set->simplex_claimed = 1;
+	if (used_dma && (ap->host->flags & ATA_HOST_SIMPLEX))
+		ap->host->simplex_claimed = 1;
 
 	/* step5: chip specific finalisation */
 	if (ap->ops->post_set_mode)
@@ -2281,7 +2282,7 @@ int ata_set_mode(struct ata_port *ap, struct ata_device **r_failed_dev)
  *	other threads.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static inline void ata_tf_to_host(struct ata_port *ap,
@@ -2445,7 +2446,7 @@ static unsigned int ata_bus_softreset(struct ata_port *ap,
  *
  *	LOCKING:
  *	PCI/etc. bus probe sem.
- *	Obtains host_set lock.
+ *	Obtains host lock.
  *
  *	SIDE EFFECTS:
  *	Sets ATA_FLAG_DISABLED if bus reset fails.
@@ -3080,7 +3081,7 @@ static int ata_dma_blacklisted(const struct ata_device *dev)
 static void ata_dev_xfermask(struct ata_device *dev)
 {
 	struct ata_port *ap = dev->ap;
-	struct ata_host_set *hs = ap->host_set;
+	struct ata_host *host = ap->host;
 	unsigned long xfer_mask;
 
 	/* controller modes available */
@@ -3114,7 +3115,7 @@ static void ata_dev_xfermask(struct ata_device *dev)
 			       "device is on DMA blacklist, disabling DMA\n");
 	}
 
-	if ((hs->flags & ATA_HOST_SIMPLEX) && hs->simplex_claimed) {
+	if ((host->flags & ATA_HOST_SIMPLEX) && host->simplex_claimed) {
 		xfer_mask &= ~(ATA_MASK_MWDMA | ATA_MASK_UDMA);
 		ata_dev_printk(dev, KERN_WARNING, "simplex DMA is claimed by "
 			       "other device, disabling DMA\n");
@@ -3207,7 +3208,7 @@ static unsigned int ata_dev_init_params(struct ata_device *dev,
  *	Unmap all mapped DMA memory associated with this command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void ata_sg_clean(struct ata_queued_cmd *qc)
@@ -3267,7 +3268,7 @@ static void ata_sg_clean(struct ata_queued_cmd *qc)
  *	associated with the current disk command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  */
 static void ata_fill_sg(struct ata_queued_cmd *qc)
@@ -3319,7 +3320,7 @@ static void ata_fill_sg(struct ata_queued_cmd *qc)
  *	supplied PACKET command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS: 0 when ATAPI DMA can be used
  *               nonzero otherwise
@@ -3341,7 +3342,7 @@ int ata_check_atapi_dma(struct ata_queued_cmd *qc)
  *	Prepare ATA taskfile for submission.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_qc_prep(struct ata_queued_cmd *qc)
 {
@@ -3363,7 +3364,7 @@ void ata_noop_qc_prep(struct ata_queued_cmd *qc) { }
  *	to point to a single memory buffer, @buf of byte length @buflen.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen)
@@ -3394,7 +3395,7 @@ void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen)
  *	elements.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
@@ -3413,7 +3414,7 @@ void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg,
  *	DMA-map the memory buffer associated with queued_cmd @qc.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, negative on error.
@@ -3482,7 +3483,7 @@ skip_map:
  *	DMA-map the scatter-gather table associated with queued_cmd @qc.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, negative on error.
@@ -3991,7 +3992,7 @@ static inline int ata_hsm_ok_in_wq(struct ata_port *ap, struct ata_queued_cmd *q
  *	Finish @qc which is running on standard HSM.
  *
  *	LOCKING:
- *	If @in_wq is zero, spin_lock_irqsave(host_set lock).
+ *	If @in_wq is zero, spin_lock_irqsave(host lock).
  *	Otherwise, none on entry and grabs host lock.
  */
 static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
@@ -4003,8 +4004,8 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 		if (in_wq) {
 			spin_lock_irqsave(ap->lock, flags);
 
-			/* EH might have kicked in while host_set lock
-			 * is released.
+			/* EH might have kicked in while host lock is
+			 * released.
 			 */
 			qc = ata_qc_from_tag(ap, qc->tag);
 			if (qc) {
@@ -4369,7 +4370,7 @@ struct ata_queued_cmd *ata_qc_new_init(struct ata_device *dev)
  *	in case something prevents using it.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_qc_free(struct ata_queued_cmd *qc)
 {
@@ -4422,7 +4423,7 @@ void __ata_qc_complete(struct ata_queued_cmd *qc)
  *	command has completed, with either an ok or not-ok status.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_qc_complete(struct ata_queued_cmd *qc)
 {
@@ -4485,7 +4486,7 @@ void ata_qc_complete(struct ata_queued_cmd *qc)
  *	and commands are completed accordingly.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Number of completed commands on success, -errno otherwise.
@@ -4556,7 +4557,7 @@ static inline int ata_should_dma_map(struct ata_queued_cmd *qc)
  *	writing the taskfile to hardware, starting the command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_qc_issue(struct ata_queued_cmd *qc)
 {
@@ -4617,7 +4618,7 @@ err:
  *	May be used as the qc_issue() entry in ata_port_operations.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, AC_ERR_* mask on failure
@@ -4746,7 +4747,7 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc)
  *	handled via polling with interrupts disabled (nIEN bit).
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	One if interrupt was handled, zero if not (shared irq).
@@ -4833,14 +4834,14 @@ idle_irq:
 /**
  *	ata_interrupt - Default ATA host interrupt handler
  *	@irq: irq line (unused)
- *	@dev_instance: pointer to our ata_host_set information structure
+ *	@dev_instance: pointer to our ata_host information structure
  *	@regs: unused
  *
  *	Default interrupt handler for PCI IDE devices.  Calls
  *	ata_host_intr() for each port that is not disabled.
  *
  *	LOCKING:
- *	Obtains host_set lock during operation.
+ *	Obtains host lock during operation.
  *
  *	RETURNS:
  *	IRQ_NONE or IRQ_HANDLED.
@@ -4848,18 +4849,18 @@ idle_irq:
 
 irqreturn_t ata_interrupt (int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	unsigned int i;
 	unsigned int handled = 0;
 	unsigned long flags;
 
 	/* TODO: make _irqsave conditional on x86 PCI IDE legacy mode */
-	spin_lock_irqsave(&host_set->lock, flags);
+	spin_lock_irqsave(&host->lock, flags);
 
-	for (i = 0; i < host_set->n_ports; i++) {
+	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap;
 
-		ap = host_set->ports[i];
+		ap = host->ports[i];
 		if (ap &&
 		    !(ap->flags & ATA_FLAG_DISABLED)) {
 			struct ata_queued_cmd *qc;
@@ -4871,7 +4872,7 @@ irqreturn_t ata_interrupt (int irq, void *dev_instance, struct pt_regs *regs)
 		}
 	}
 
-	spin_unlock_irqrestore(&host_set->lock, flags);
+	spin_unlock_irqrestore(&host->lock, flags);
 
 	return IRQ_RETVAL(handled);
 }
@@ -5036,15 +5037,15 @@ int ata_flush_cache(struct ata_device *dev)
 	return 0;
 }
 
-static int ata_host_set_request_pm(struct ata_host_set *host_set,
-				   pm_message_t mesg, unsigned int action,
-				   unsigned int ehi_flags, int wait)
+static int ata_host_request_pm(struct ata_host *host, pm_message_t mesg,
+			       unsigned int action, unsigned int ehi_flags,
+			       int wait)
 {
 	unsigned long flags;
 	int i, rc;
 
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 
 		/* Previous resume operation might still be in
 		 * progress.  Wait for PM_PENDING to clear.
@@ -5084,11 +5085,11 @@ static int ata_host_set_request_pm(struct ata_host_set *host_set,
 }
 
 /**
- *	ata_host_set_suspend - suspend host_set
- *	@host_set: host_set to suspend
+ *	ata_host_suspend - suspend host
+ *	@host: host to suspend
  *	@mesg: PM message
  *
- *	Suspend @host_set.  Actual operation is performed by EH.  This
+ *	Suspend @host.  Actual operation is performed by EH.  This
  *	function requests EH to perform PM operations and waits for EH
  *	to finish.
  *
@@ -5098,11 +5099,11 @@ static int ata_host_set_request_pm(struct ata_host_set *host_set,
  *	RETURNS:
  *	0 on success, -errno on failure.
  */
-int ata_host_set_suspend(struct ata_host_set *host_set, pm_message_t mesg)
+int ata_host_suspend(struct ata_host *host, pm_message_t mesg)
 {
 	int i, j, rc;
 
-	rc = ata_host_set_request_pm(host_set, mesg, 0, ATA_EHI_QUIET, 1);
+	rc = ata_host_request_pm(host, mesg, 0, ATA_EHI_QUIET, 1);
 	if (rc)
 		goto fail;
 
@@ -5110,8 +5111,8 @@ int ata_host_set_suspend(struct ata_host_set *host_set, pm_message_t mesg)
 	 * This happens if hotplug occurs between completion of device
 	 * suspension and here.
 	 */
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 
 		for (j = 0; j < ATA_MAX_DEVICES; j++) {
 			struct ata_device *dev = &ap->device[j];
@@ -5126,30 +5127,30 @@ int ata_host_set_suspend(struct ata_host_set *host_set, pm_message_t mesg)
 		}
 	}
 
-	host_set->dev->power.power_state = mesg;
+	host->dev->power.power_state = mesg;
 	return 0;
 
  fail:
-	ata_host_set_resume(host_set);
+	ata_host_resume(host);
 	return rc;
 }
 
 /**
- *	ata_host_set_resume - resume host_set
- *	@host_set: host_set to resume
+ *	ata_host_resume - resume host
+ *	@host: host to resume
  *
- *	Resume @host_set.  Actual operation is performed by EH.  This
+ *	Resume @host.  Actual operation is performed by EH.  This
  *	function requests EH to perform PM operations and returns.
  *	Note that all resume operations are performed parallely.
  *
  *	LOCKING:
  *	Kernel thread context (may sleep).
  */
-void ata_host_set_resume(struct ata_host_set *host_set)
+void ata_host_resume(struct ata_host *host)
 {
-	ata_host_set_request_pm(host_set, PMSG_ON, ATA_EH_SOFTRESET,
-				ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 0);
-	host_set->dev->power.power_state = PMSG_ON;
+	ata_host_request_pm(host, PMSG_ON, ATA_EH_SOFTRESET,
+			    ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 0);
+	host->dev->power.power_state = PMSG_ON;
 }
 
 /**
@@ -5206,10 +5207,10 @@ void ata_port_stop (struct ata_port *ap)
 	ata_pad_free(ap, dev);
 }
 
-void ata_host_stop (struct ata_host_set *host_set)
+void ata_host_stop (struct ata_host *host)
 {
-	if (host_set->mmio_base)
-		iounmap(host_set->mmio_base);
+	if (host->mmio_base)
+		iounmap(host->mmio_base);
 }
 
 /**
@@ -5231,7 +5232,7 @@ void ata_dev_init(struct ata_device *dev)
 
 	/* High bits of dev->flags are used to record warm plug
 	 * requests which occur asynchronously.  Synchronize using
-	 * host_set lock.
+	 * host lock.
 	 */
 	spin_lock_irqsave(ap->lock, flags);
 	dev->flags &= ~ATA_DFLAG_INIT_MASK;
@@ -5247,7 +5248,7 @@ void ata_dev_init(struct ata_device *dev)
 /**
  *	ata_port_init - Initialize an ata_port structure
  *	@ap: Structure to initialize
- *	@host_set: Collection of hosts to which @ap belongs
+ *	@host: Collection of hosts to which @ap belongs
  *	@ent: Probe information provided by low-level driver
  *	@port_no: Port number associated with this ata_port
  *
@@ -5256,22 +5257,22 @@ void ata_dev_init(struct ata_device *dev)
  *	LOCKING:
  *	Inherited from caller.
  */
-void ata_port_init(struct ata_port *ap, struct ata_host_set *host_set,
+void ata_port_init(struct ata_port *ap, struct ata_host *host,
 		   const struct ata_probe_ent *ent, unsigned int port_no)
 {
 	unsigned int i;
 
-	ap->lock = &host_set->lock;
+	ap->lock = &host->lock;
 	ap->flags = ATA_FLAG_DISABLED;
 	ap->id = ata_unique_id++;
 	ap->ctl = ATA_DEVCTL_OBS;
-	ap->host_set = host_set;
+	ap->host = host;
 	ap->dev = ent->dev;
 	ap->port_no = port_no;
 	ap->pio_mask = ent->pio_mask;
 	ap->mwdma_mask = ent->mwdma_mask;
 	ap->udma_mask = ent->udma_mask;
-	ap->flags |= ent->host_flags;
+	ap->flags |= ent->port_flags;
 	ap->ops = ent->port_ops;
 	ap->hw_sata_spd_limit = UINT_MAX;
 	ap->active_tag = ATA_TAG_POISON;
@@ -5324,7 +5325,7 @@ void ata_port_init(struct ata_port *ap, struct ata_host_set *host_set,
  */
 static void ata_port_init_shost(struct ata_port *ap, struct Scsi_Host *shost)
 {
-	ap->host = shost;
+	ap->scsi_host = shost;
 
 	shost->unique_id = ap->id;
 	shost->max_id = 16;
@@ -5336,7 +5337,7 @@ static void ata_port_init_shost(struct ata_port *ap, struct Scsi_Host *shost)
 /**
  *	ata_port_add - Attach low-level ATA driver to system
  *	@ent: Information provided by low-level driver
- *	@host_set: Collections of ports to which we add
+ *	@host: Collections of ports to which we add
  *	@port_no: Port number associated with this host
  *
  *	Attach low-level ATA driver to system.
@@ -5348,7 +5349,7 @@ static void ata_port_init_shost(struct ata_port *ap, struct Scsi_Host *shost)
  *	New ata_port on success, for NULL on error.
  */
 static struct ata_port * ata_port_add(const struct ata_probe_ent *ent,
-				      struct ata_host_set *host_set,
+				      struct ata_host *host,
 				      unsigned int port_no)
 {
 	struct Scsi_Host *shost;
@@ -5357,7 +5358,7 @@ static struct ata_port * ata_port_add(const struct ata_probe_ent *ent,
 	DPRINTK("ENTER\n");
 
 	if (!ent->port_ops->error_handler &&
-	    !(ent->host_flags & (ATA_FLAG_SATA_RESET | ATA_FLAG_SRST))) {
+	    !(ent->port_flags & (ATA_FLAG_SATA_RESET | ATA_FLAG_SRST))) {
 		printk(KERN_ERR "ata%u: no reset mechanism available\n",
 		       port_no);
 		return NULL;
@@ -5371,32 +5372,31 @@ static struct ata_port * ata_port_add(const struct ata_probe_ent *ent,
 
 	ap = ata_shost_to_port(shost);
 
-	ata_port_init(ap, host_set, ent, port_no);
+	ata_port_init(ap, host, ent, port_no);
 	ata_port_init_shost(ap, shost);
 
 	return ap;
 }
 
 /**
- *	ata_sas_host_init - Initialize a host_set struct
- *	@host_set:	host_set to initialize
- *	@dev:		device host_set is attached to
- *	@flags:	host_set flags
- *	@ops:		port_ops
+ *	ata_sas_host_init - Initialize a host struct
+ *	@host:	host to initialize
+ *	@dev:	device host is attached to
+ *	@flags:	host flags
+ *	@ops:	port_ops
  *
  *	LOCKING:
  *	PCI/etc. bus probe sem.
  *
  */
 
-void ata_host_set_init(struct ata_host_set *host_set,
-		       struct device *dev, unsigned long flags,
-		       const struct ata_port_operations *ops)
+void ata_host_init(struct ata_host *host, struct device *dev,
+		   unsigned long flags, const struct ata_port_operations *ops)
 {
-	spin_lock_init(&host_set->lock);
-	host_set->dev = dev;
-	host_set->flags = flags;
-	host_set->ops = ops;
+	spin_lock_init(&host->lock);
+	host->dev = dev;
+	host->flags = flags;
+	host->ops = ops;
 }
 
 /**
@@ -5421,34 +5421,34 @@ int ata_device_add(const struct ata_probe_ent *ent)
 {
 	unsigned int i;
 	struct device *dev = ent->dev;
-	struct ata_host_set *host_set;
+	struct ata_host *host;
 	int rc;
 
 	DPRINTK("ENTER\n");
 	/* alloc a container for our list of ATA ports (buses) */
-	host_set = kzalloc(sizeof(struct ata_host_set) +
-			   (ent->n_ports * sizeof(void *)), GFP_KERNEL);
-	if (!host_set)
+	host = kzalloc(sizeof(struct ata_host) +
+		       (ent->n_ports * sizeof(void *)), GFP_KERNEL);
+	if (!host)
 		return 0;
 
-	ata_host_set_init(host_set, dev, ent->host_set_flags, ent->port_ops);
-	host_set->n_ports = ent->n_ports;
-	host_set->irq = ent->irq;
-	host_set->irq2 = ent->irq2;
-	host_set->mmio_base = ent->mmio_base;
-	host_set->private_data = ent->private_data;
+	ata_host_init(host, dev, ent->_host_flags, ent->port_ops);
+	host->n_ports = ent->n_ports;
+	host->irq = ent->irq;
+	host->irq2 = ent->irq2;
+	host->mmio_base = ent->mmio_base;
+	host->private_data = ent->private_data;
 
 	/* register each port bound to this device */
-	for (i = 0; i < host_set->n_ports; i++) {
+	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap;
 		unsigned long xfer_mode_mask;
 		int irq_line = ent->irq;
 
-		ap = ata_port_add(ent, host_set, i);
+		ap = ata_port_add(ent, host, i);
 		if (!ap)
 			goto err_out;
 
-		host_set->ports[i] = ap;
+		host->ports[i] = ap;
 
 		/* dummy? */
 		if (ent->dummy_port_mask & (1 << i)) {
@@ -5460,8 +5460,8 @@ int ata_device_add(const struct ata_probe_ent *ent)
 		/* start port */
 		rc = ap->ops->port_start(ap);
 		if (rc) {
-			host_set->ports[i] = NULL;
-			scsi_host_put(ap->host);
+			host->ports[i] = NULL;
+			scsi_host_put(ap->scsi_host);
 			goto err_out;
 		}
 
@@ -5484,13 +5484,13 @@ int ata_device_add(const struct ata_probe_ent *ent)
 				irq_line);
 
 		ata_chk_status(ap);
-		host_set->ops->irq_clear(ap);
+		host->ops->irq_clear(ap);
 		ata_eh_freeze_port(ap);	/* freeze port before requesting IRQ */
 	}
 
 	/* obtain irq, that may be shared between channels */
 	rc = request_irq(ent->irq, ent->port_ops->irq_handler, ent->irq_flags,
-			 DRV_NAME, host_set);
+			 DRV_NAME, host);
 	if (rc) {
 		dev_printk(KERN_ERR, dev, "irq %lu request failed: %d\n",
 			   ent->irq, rc);
@@ -5504,7 +5504,7 @@ int ata_device_add(const struct ata_probe_ent *ent)
 		BUG_ON(ent->irq == ent->irq2);
 
 		rc = request_irq(ent->irq2, ent->port_ops->irq_handler, ent->irq_flags,
-			 DRV_NAME, host_set);
+			 DRV_NAME, host);
 		if (rc) {
 			dev_printk(KERN_ERR, dev, "irq %lu request failed: %d\n",
 				   ent->irq2, rc);
@@ -5514,8 +5514,8 @@ int ata_device_add(const struct ata_probe_ent *ent)
 
 	/* perform each probe synchronously */
 	DPRINTK("probe begin\n");
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 		u32 scontrol;
 		int rc;
 
@@ -5526,7 +5526,7 @@ int ata_device_add(const struct ata_probe_ent *ent)
 		}
 		ap->sata_spd_limit = ap->hw_sata_spd_limit;
 
-		rc = scsi_add_host(ap->host, dev);
+		rc = scsi_add_host(ap->scsi_host, dev);
 		if (rc) {
 			ata_port_printk(ap, KERN_ERR, "scsi_add_host failed\n");
 			/* FIXME: do something useful here */
@@ -5574,29 +5574,29 @@ int ata_device_add(const struct ata_probe_ent *ent)
 
 	/* probes are done, now scan each port's disk(s) */
 	DPRINTK("host probe begin\n");
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 
 		ata_scsi_scan_host(ap);
 	}
 
-	dev_set_drvdata(dev, host_set);
+	dev_set_drvdata(dev, host);
 
 	VPRINTK("EXIT, returning %u\n", ent->n_ports);
 	return ent->n_ports; /* success */
 
 err_out_free_irq:
-	free_irq(ent->irq, host_set);
+	free_irq(ent->irq, host);
 err_out:
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 		if (ap) {
 			ap->ops->port_stop(ap);
-			scsi_host_put(ap->host);
+			scsi_host_put(ap->scsi_host);
 		}
 	}
 
-	kfree(host_set);
+	kfree(host);
 	VPRINTK("EXIT, returning 0\n");
 	return 0;
 }
@@ -5656,12 +5656,12 @@ void ata_port_detach(struct ata_port *ap)
 
  skip_eh:
 	/* remove the associated SCSI host */
-	scsi_remove_host(ap->host);
+	scsi_remove_host(ap->scsi_host);
 }
 
 /**
- *	ata_host_set_remove - PCI layer callback for device removal
- *	@host_set: ATA host set that was removed
+ *	ata_host_remove - PCI layer callback for device removal
+ *	@host: ATA host set that was removed
  *
  *	Unregister all objects associated with this host set. Free those
  *	objects.
@@ -5670,21 +5670,21 @@ void ata_port_detach(struct ata_port *ap)
  *	Inherited from calling layer (may sleep).
  */
 
-void ata_host_set_remove(struct ata_host_set *host_set)
+void ata_host_remove(struct ata_host *host)
 {
 	unsigned int i;
 
-	for (i = 0; i < host_set->n_ports; i++)
-		ata_port_detach(host_set->ports[i]);
+	for (i = 0; i < host->n_ports; i++)
+		ata_port_detach(host->ports[i]);
 
-	free_irq(host_set->irq, host_set);
-	if (host_set->irq2)
-		free_irq(host_set->irq2, host_set);
+	free_irq(host->irq, host);
+	if (host->irq2)
+		free_irq(host->irq2, host);
 
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 
-		ata_scsi_release(ap->host);
+		ata_scsi_release(ap->scsi_host);
 
 		if ((ap->flags & ATA_FLAG_NO_LEGACY) == 0) {
 			struct ata_ioports *ioaddr = &ap->ioaddr;
@@ -5696,13 +5696,13 @@ void ata_host_set_remove(struct ata_host_set *host_set)
 				release_region(ATA_SECONDARY_CMD, 8);
 		}
 
-		scsi_host_put(ap->host);
+		scsi_host_put(ap->scsi_host);
 	}
 
-	if (host_set->ops->host_stop)
-		host_set->ops->host_stop(host_set);
+	if (host->ops->host_stop)
+		host->ops->host_stop(host);
 
-	kfree(host_set);
+	kfree(host);
 }
 
 /**
@@ -5719,9 +5719,9 @@ void ata_host_set_remove(struct ata_host_set *host_set)
  *	One.
  */
 
-int ata_scsi_release(struct Scsi_Host *host)
+int ata_scsi_release(struct Scsi_Host *shost)
 {
-	struct ata_port *ap = ata_shost_to_port(host);
+	struct ata_port *ap = ata_shost_to_port(shost);
 
 	DPRINTK("ENTER\n");
 
@@ -5748,7 +5748,7 @@ ata_probe_ent_alloc(struct device *dev, const struct ata_port_info *port)
 	probe_ent->dev = dev;
 
 	probe_ent->sht = port->sht;
-	probe_ent->host_flags = port->host_flags;
+	probe_ent->port_flags = port->flags;
 	probe_ent->pio_mask = port->pio_mask;
 	probe_ent->mwdma_mask = port->mwdma_mask;
 	probe_ent->udma_mask = port->udma_mask;
@@ -5786,11 +5786,11 @@ void ata_std_ports(struct ata_ioports *ioaddr)
 
 #ifdef CONFIG_PCI
 
-void ata_pci_host_stop (struct ata_host_set *host_set)
+void ata_pci_host_stop (struct ata_host *host)
 {
-	struct pci_dev *pdev = to_pci_dev(host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(host->dev);
 
-	pci_iounmap(pdev, host_set->mmio_base);
+	pci_iounmap(pdev, host->mmio_base);
 }
 
 /**
@@ -5810,9 +5810,9 @@ void ata_pci_host_stop (struct ata_host_set *host_set)
 void ata_pci_remove_one (struct pci_dev *pdev)
 {
 	struct device *dev = pci_dev_to_dev(pdev);
-	struct ata_host_set *host_set = dev_get_drvdata(dev);
+	struct ata_host *host = dev_get_drvdata(dev);
 
-	ata_host_set_remove(host_set);
+	ata_host_remove(host);
 
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
@@ -5873,10 +5873,10 @@ void ata_pci_device_do_resume(struct pci_dev *pdev)
 
 int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
-	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
 	int rc = 0;
 
-	rc = ata_host_set_suspend(host_set, mesg);
+	rc = ata_host_suspend(host, mesg);
 	if (rc)
 		return rc;
 
@@ -5887,10 +5887,10 @@ int ata_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg)
 
 int ata_pci_device_resume(struct pci_dev *pdev)
 {
-	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
 
 	ata_pci_device_do_resume(pdev);
-	ata_host_set_resume(host_set);
+	ata_host_resume(host);
 	return 0;
 }
 #endif /* CONFIG_PCI */
@@ -6035,10 +6035,10 @@ EXPORT_SYMBOL_GPL(sata_deb_timing_long);
 EXPORT_SYMBOL_GPL(ata_dummy_port_ops);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
 EXPORT_SYMBOL_GPL(ata_std_ports);
-EXPORT_SYMBOL_GPL(ata_host_set_init);
+EXPORT_SYMBOL_GPL(ata_host_init);
 EXPORT_SYMBOL_GPL(ata_device_add);
 EXPORT_SYMBOL_GPL(ata_port_detach);
-EXPORT_SYMBOL_GPL(ata_host_set_remove);
+EXPORT_SYMBOL_GPL(ata_host_remove);
 EXPORT_SYMBOL_GPL(ata_sg_init);
 EXPORT_SYMBOL_GPL(ata_sg_init_one);
 EXPORT_SYMBOL_GPL(ata_hsm_move);
@@ -6105,8 +6105,8 @@ EXPORT_SYMBOL_GPL(sata_scr_write);
 EXPORT_SYMBOL_GPL(sata_scr_write_flush);
 EXPORT_SYMBOL_GPL(ata_port_online);
 EXPORT_SYMBOL_GPL(ata_port_offline);
-EXPORT_SYMBOL_GPL(ata_host_set_suspend);
-EXPORT_SYMBOL_GPL(ata_host_set_resume);
+EXPORT_SYMBOL_GPL(ata_host_suspend);
+EXPORT_SYMBOL_GPL(ata_host_resume);
 EXPORT_SYMBOL_GPL(ata_id_string);
 EXPORT_SYMBOL_GPL(ata_id_c_string);
 EXPORT_SYMBOL_GPL(ata_scsi_simulate);
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 2c476eee463f..b1b510493c2d 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -200,7 +200,7 @@ void ata_scsi_error(struct Scsi_Host *host)
 	/* synchronize with port task */
 	ata_port_flush_task(ap);
 
-	/* synchronize with host_set lock and sort out timeouts */
+	/* synchronize with host lock and sort out timeouts */
 
 	/* For new EH, all qcs are finished in one of three ways -
 	 * normal completion, error completion, and SCSI timeout.
@@ -377,7 +377,7 @@ void ata_port_wait_eh(struct ata_port *ap)
 	spin_unlock_irqrestore(ap->lock, flags);
 
 	/* make sure SCSI EH is complete */
-	if (scsi_host_in_recovery(ap->host)) {
+	if (scsi_host_in_recovery(ap->scsi_host)) {
 		msleep(10);
 		goto retry;
 	}
@@ -486,7 +486,7 @@ void ata_eng_timeout(struct ata_port *ap)
  *	other commands are drained.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_qc_schedule_eh(struct ata_queued_cmd *qc)
 {
@@ -513,14 +513,14 @@ void ata_qc_schedule_eh(struct ata_queued_cmd *qc)
  *	all commands are drained.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_port_schedule_eh(struct ata_port *ap)
 {
 	WARN_ON(!ap->ops->error_handler);
 
 	ap->pflags |= ATA_PFLAG_EH_PENDING;
-	scsi_schedule_eh(ap->host);
+	scsi_schedule_eh(ap->scsi_host);
 
 	DPRINTK("port EH scheduled\n");
 }
@@ -532,7 +532,7 @@ void ata_port_schedule_eh(struct ata_port *ap)
  *	Abort all active qc's of @ap and schedule EH.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Number of aborted qc's.
@@ -575,7 +575,7 @@ int ata_port_abort(struct ata_port *ap)
  *	is frozen.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 static void __ata_port_freeze(struct ata_port *ap)
 {
@@ -596,7 +596,7 @@ static void __ata_port_freeze(struct ata_port *ap)
  *	Abort and freeze @ap.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Number of aborted commands.
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index d168e3413661..3986ec8741b4 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -321,7 +321,7 @@ int ata_scsi_ioctl(struct scsi_device *scsidev, int cmd, void __user *arg)
  *	current command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Command allocated, or %NULL if none available.
@@ -537,7 +537,7 @@ int ata_scsi_device_resume(struct scsi_device *sdev)
  *	format sense blocks.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_to_sense_error(unsigned id, u8 drv_stat, u8 drv_err, u8 *sk, u8 *asc,
 			u8 *ascq, int verbose)
@@ -649,7 +649,7 @@ void ata_to_sense_error(unsigned id, u8 drv_stat, u8 drv_err, u8 *sk, u8 *asc,
  *	block. Clear sense key, ASC & ASCQ if there is no error.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 void ata_gen_ata_desc_sense(struct ata_queued_cmd *qc)
 {
@@ -918,7 +918,7 @@ int ata_scsi_change_queue_depth(struct scsi_device *sdev, int queue_depth)
  *	[See SAT revision 5 at www.t10.org]
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, non-zero on error.
@@ -986,7 +986,7 @@ invalid_fld:
  *	FLUSH CACHE EXT.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, non-zero on error.
@@ -1109,7 +1109,7 @@ static void scsi_16_lba_len(const u8 *scsicmd, u64 *plba, u32 *plen)
  *	Converts SCSI VERIFY command to an ATA READ VERIFY command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, non-zero on error.
@@ -1233,7 +1233,7 @@ nothing_to_do:
  *	%WRITE_16 are currently supported.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, non-zero on error.
@@ -1467,7 +1467,7 @@ static void ata_scsi_qc_complete(struct ata_queued_cmd *qc)
  *	issued to @dev.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	1 if deferring is needed, 0 otherwise.
@@ -1510,7 +1510,7 @@ static int ata_scmd_need_defer(struct ata_device *dev, int is_io)
  *	termination.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	0 on success, SCSI_ML_QUEUE_DEVICE_BUSY if the command
@@ -1589,7 +1589,7 @@ defer:
  *	Maps buffer contained within SCSI command @cmd.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Length of response buffer.
@@ -1623,7 +1623,7 @@ static unsigned int ata_scsi_rbuf_get(struct scsi_cmnd *cmd, u8 **buf_out)
  *	Unmaps response buffer contained within @cmd.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static inline void ata_scsi_rbuf_put(struct scsi_cmnd *cmd, u8 *buf)
@@ -1649,7 +1649,7 @@ static inline void ata_scsi_rbuf_put(struct scsi_cmnd *cmd, u8 *buf)
  *	and sense buffer are assumed to be set).
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_scsi_rbuf_fill(struct ata_scsi_args *args,
@@ -1680,7 +1680,7 @@ void ata_scsi_rbuf_fill(struct ata_scsi_args *args,
  *	with non-VPD INQUIRY command output.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_inq_std(struct ata_scsi_args *args, u8 *rbuf,
@@ -1736,7 +1736,7 @@ unsigned int ata_scsiop_inq_std(struct ata_scsi_args *args, u8 *rbuf,
  *	Returns list of inquiry VPD pages available.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_inq_00(struct ata_scsi_args *args, u8 *rbuf,
@@ -1764,7 +1764,7 @@ unsigned int ata_scsiop_inq_00(struct ata_scsi_args *args, u8 *rbuf,
  *	Returns ATA device serial number.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_inq_80(struct ata_scsi_args *args, u8 *rbuf,
@@ -1797,7 +1797,7 @@ unsigned int ata_scsiop_inq_80(struct ata_scsi_args *args, u8 *rbuf,
  *	   name ("ATA     "), model and serial numbers.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_inq_83(struct ata_scsi_args *args, u8 *rbuf,
@@ -1849,7 +1849,7 @@ unsigned int ata_scsiop_inq_83(struct ata_scsi_args *args, u8 *rbuf,
  *	that the caller should successfully complete this SCSI command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_noop(struct ata_scsi_args *args, u8 *rbuf,
@@ -1990,7 +1990,7 @@ static int ata_dev_supports_fua(u16 *id)
  *	descriptor for other device types.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_mode_sense(struct ata_scsi_args *args, u8 *rbuf,
@@ -2129,7 +2129,7 @@ saving_not_supp:
  *	Simulate READ CAPACITY commands.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf,
@@ -2204,7 +2204,7 @@ unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf,
  *	Simulate REPORT LUNS command.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 unsigned int ata_scsiop_report_luns(struct ata_scsi_args *args, u8 *rbuf,
@@ -2256,7 +2256,7 @@ void ata_scsi_set_sense(struct scsi_cmnd *cmd, u8 sk, u8 asc, u8 ascq)
  *	and the specified additional sense codes.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_scsi_badcmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *), u8 asc, u8 ascq)
@@ -2421,7 +2421,7 @@ static void atapi_qc_complete(struct ata_queued_cmd *qc)
  *	@scsicmd: SCSI CDB associated with this PACKET command
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Zero on success, non-zero on failure.
@@ -2500,7 +2500,7 @@ static struct ata_device * __ata_scsi_find_dev(struct ata_port *ap,
  *	Determine if commands should be sent to the specified device.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	0 if commands are not allowed / 1 if commands are allowed
@@ -2534,7 +2534,7 @@ static int ata_scsi_dev_enabled(struct ata_device *dev)
  *	SCSI command to be sent.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	Associated ATA device, or %NULL if not found.
@@ -2808,7 +2808,7 @@ static inline int __ata_scsi_queuecmd(struct scsi_cmnd *cmd,
  *	ATA and ATAPI devices appearing as SCSI devices.
  *
  *	LOCKING:
- *	Releases scsi-layer-held lock, and obtains host_set lock.
+ *	Releases scsi-layer-held lock, and obtains host lock.
  *
  *	RETURNS:
  *	Return value from __ata_scsi_queuecmd() if @cmd can be queued,
@@ -2852,7 +2852,7 @@ int ata_scsi_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *))
  *	that can be handled internally.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 void ata_scsi_simulate(struct ata_device *dev, struct scsi_cmnd *cmd,
@@ -2944,7 +2944,7 @@ void ata_scsi_scan_host(struct ata_port *ap)
 		if (!ata_dev_enabled(dev) || dev->sdev)
 			continue;
 
-		sdev = __scsi_add_device(ap->host, 0, i, 0, NULL);
+		sdev = __scsi_add_device(ap->scsi_host, 0, i, 0, NULL);
 		if (!IS_ERR(sdev)) {
 			dev->sdev = sdev;
 			scsi_device_put(sdev);
@@ -2958,11 +2958,11 @@ void ata_scsi_scan_host(struct ata_port *ap)
  *
  *	This function is called from ata_eh_hotplug() and responsible
  *	for taking the SCSI device attached to @dev offline.  This
- *	function is called with host_set lock which protects dev->sdev
+ *	function is called with host lock which protects dev->sdev
  *	against clearing.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  *
  *	RETURNS:
  *	1 if attached SCSI device exists, 0 otherwise.
@@ -2998,16 +2998,16 @@ static void ata_scsi_remove_dev(struct ata_device *dev)
 	 * be removed if there is __scsi_device_get() interface which
 	 * increments reference counts regardless of device state.
 	 */
-	mutex_lock(&ap->host->scan_mutex);
+	mutex_lock(&ap->scsi_host->scan_mutex);
 	spin_lock_irqsave(ap->lock, flags);
 
-	/* clearing dev->sdev is protected by host_set lock */
+	/* clearing dev->sdev is protected by host lock */
 	sdev = dev->sdev;
 	dev->sdev = NULL;
 
 	if (sdev) {
 		/* If user initiated unplug races with us, sdev can go
-		 * away underneath us after the host_set lock and
+		 * away underneath us after the host lock and
 		 * scan_mutex are released.  Hold onto it.
 		 */
 		if (scsi_device_get(sdev) == 0) {
@@ -3024,7 +3024,7 @@ static void ata_scsi_remove_dev(struct ata_device *dev)
 	}
 
 	spin_unlock_irqrestore(ap->lock, flags);
-	mutex_unlock(&ap->host->scan_mutex);
+	mutex_unlock(&ap->scsi_host->scan_mutex);
 
 	if (sdev) {
 		ata_dev_printk(dev, KERN_INFO, "detaching (SCSI %s)\n",
@@ -3176,7 +3176,7 @@ void ata_scsi_dev_rescan(void *data)
  *	ata_sas_port_alloc - Allocate port for a SAS attached SATA device
  *	@pdev: PCI device that the scsi device is attached to
  *	@port_info: Information from low-level host driver
- *	@host: SCSI host that the scsi device is attached to
+ *	@shost: SCSI host that the scsi device is attached to
  *
  *	LOCKING:
  *	PCI/etc. bus probe sem.
@@ -3185,9 +3185,9 @@ void ata_scsi_dev_rescan(void *data)
  *	ata_port pointer on success / NULL on failure.
  */
 
-struct ata_port *ata_sas_port_alloc(struct ata_host_set *host_set,
+struct ata_port *ata_sas_port_alloc(struct ata_host *host,
 				    struct ata_port_info *port_info,
-				    struct Scsi_Host *host)
+				    struct Scsi_Host *shost)
 {
 	struct ata_port *ap = kzalloc(sizeof(*ap), GFP_KERNEL);
 	struct ata_probe_ent *ent;
@@ -3195,14 +3195,14 @@ struct ata_port *ata_sas_port_alloc(struct ata_host_set *host_set,
 	if (!ap)
 		return NULL;
 
-	ent = ata_probe_ent_alloc(host_set->dev, port_info);
+	ent = ata_probe_ent_alloc(host->dev, port_info);
 	if (!ent) {
 		kfree(ap);
 		return NULL;
 	}
 
-	ata_port_init(ap, host_set, ent, 0);
-	ap->lock = host->host_lock;
+	ata_port_init(ap, host, ent, 0);
+	ap->lock = shost->host_lock;
 	kfree(ent);
 	return ap;
 }
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index d4a4f82360ec..a5ecb71390a9 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -69,7 +69,7 @@ extern int ata_flush_cache(struct ata_device *dev);
 extern void ata_dev_init(struct ata_device *dev);
 extern int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg);
 extern int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg);
-extern void ata_port_init(struct ata_port *ap, struct ata_host_set *host_set,
+extern void ata_port_init(struct ata_port *ap, struct ata_host *host,
 			  const struct ata_probe_ent *ent, unsigned int port_no);
 extern struct ata_probe_ent *ata_probe_ent_alloc(struct device *dev,
 						 const struct ata_port_info *port);
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index 61d2aa697b4d..912211ada816 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -127,7 +127,7 @@ static int adma_ata_init_one (struct pci_dev *pdev,
 static irqreturn_t adma_intr (int irq, void *dev_instance,
 				struct pt_regs *regs);
 static int adma_port_start(struct ata_port *ap);
-static void adma_host_stop(struct ata_host_set *host_set);
+static void adma_host_stop(struct ata_host *host);
 static void adma_port_stop(struct ata_port *ap);
 static void adma_phy_reset(struct ata_port *ap);
 static void adma_qc_prep(struct ata_queued_cmd *qc);
@@ -182,7 +182,7 @@ static struct ata_port_info adma_port_info[] = {
 	/* board_1841_idx */
 	{
 		.sht		= &adma_ata_sht,
-		.host_flags	= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST |
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST |
 				  ATA_FLAG_NO_LEGACY | ATA_FLAG_MMIO |
 				  ATA_FLAG_PIO_POLLING,
 		.pio_mask	= 0x10, /* pio4 */
@@ -237,7 +237,7 @@ static void adma_reset_engine(void __iomem *chan)
 static void adma_reinit_engine(struct ata_port *ap)
 {
 	struct adma_port_priv *pp = ap->private_data;
-	void __iomem *mmio_base = ap->host_set->mmio_base;
+	void __iomem *mmio_base = ap->host->mmio_base;
 	void __iomem *chan = ADMA_REGS(mmio_base, ap->port_no);
 
 	/* mask/clear ATA interrupts */
@@ -265,7 +265,7 @@ static void adma_reinit_engine(struct ata_port *ap)
 
 static inline void adma_enter_reg_mode(struct ata_port *ap)
 {
-	void __iomem *chan = ADMA_REGS(ap->host_set->mmio_base, ap->port_no);
+	void __iomem *chan = ADMA_REGS(ap->host->mmio_base, ap->port_no);
 
 	writew(aPIOMD4, chan + ADMA_CONTROL);
 	readb(chan + ADMA_STATUS);	/* flush */
@@ -412,7 +412,7 @@ static void adma_qc_prep(struct ata_queued_cmd *qc)
 static inline void adma_packet_start(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	void __iomem *chan = ADMA_REGS(ap->host_set->mmio_base, ap->port_no);
+	void __iomem *chan = ADMA_REGS(ap->host->mmio_base, ap->port_no);
 
 	VPRINTK("ENTER, ap %p\n", ap);
 
@@ -442,13 +442,13 @@ static unsigned int adma_qc_issue(struct ata_queued_cmd *qc)
 	return ata_qc_issue_prot(qc);
 }
 
-static inline unsigned int adma_intr_pkt(struct ata_host_set *host_set)
+static inline unsigned int adma_intr_pkt(struct ata_host *host)
 {
 	unsigned int handled = 0, port_no;
-	u8 __iomem *mmio_base = host_set->mmio_base;
+	u8 __iomem *mmio_base = host->mmio_base;
 
-	for (port_no = 0; port_no < host_set->n_ports; ++port_no) {
-		struct ata_port *ap = host_set->ports[port_no];
+	for (port_no = 0; port_no < host->n_ports; ++port_no) {
+		struct ata_port *ap = host->ports[port_no];
 		struct adma_port_priv *pp;
 		struct ata_queued_cmd *qc;
 		void __iomem *chan = ADMA_REGS(mmio_base, port_no);
@@ -476,13 +476,13 @@ static inline unsigned int adma_intr_pkt(struct ata_host_set *host_set)
 	return handled;
 }
 
-static inline unsigned int adma_intr_mmio(struct ata_host_set *host_set)
+static inline unsigned int adma_intr_mmio(struct ata_host *host)
 {
 	unsigned int handled = 0, port_no;
 
-	for (port_no = 0; port_no < host_set->n_ports; ++port_no) {
+	for (port_no = 0; port_no < host->n_ports; ++port_no) {
 		struct ata_port *ap;
-		ap = host_set->ports[port_no];
+		ap = host->ports[port_no];
 		if (ap && (!(ap->flags & ATA_FLAG_DISABLED))) {
 			struct ata_queued_cmd *qc;
 			struct adma_port_priv *pp = ap->private_data;
@@ -511,14 +511,14 @@ static inline unsigned int adma_intr_mmio(struct ata_host_set *host_set)
 
 static irqreturn_t adma_intr(int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	unsigned int handled = 0;
 
 	VPRINTK("ENTER\n");
 
-	spin_lock(&host_set->lock);
-	handled  = adma_intr_pkt(host_set) | adma_intr_mmio(host_set);
-	spin_unlock(&host_set->lock);
+	spin_lock(&host->lock);
+	handled  = adma_intr_pkt(host) | adma_intr_mmio(host);
+	spin_unlock(&host->lock);
 
 	VPRINTK("EXIT\n");
 
@@ -544,7 +544,7 @@ static void adma_ata_setup_port(struct ata_ioports *port, unsigned long base)
 
 static int adma_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct adma_port_priv *pp;
 	int rc;
 
@@ -582,10 +582,10 @@ err_out:
 
 static void adma_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct adma_port_priv *pp = ap->private_data;
 
-	adma_reset_engine(ADMA_REGS(ap->host_set->mmio_base, ap->port_no));
+	adma_reset_engine(ADMA_REGS(ap->host->mmio_base, ap->port_no));
 	if (pp != NULL) {
 		ap->private_data = NULL;
 		if (pp->pkt != NULL)
@@ -596,14 +596,14 @@ static void adma_port_stop(struct ata_port *ap)
 	ata_port_stop(ap);
 }
 
-static void adma_host_stop(struct ata_host_set *host_set)
+static void adma_host_stop(struct ata_host *host)
 {
 	unsigned int port_no;
 
 	for (port_no = 0; port_no < ADMA_PORTS; ++port_no)
-		adma_reset_engine(ADMA_REGS(host_set->mmio_base, port_no));
+		adma_reset_engine(ADMA_REGS(host->mmio_base, port_no));
 
-	ata_pci_host_stop(host_set);
+	ata_pci_host_stop(host);
 }
 
 static void adma_host_init(unsigned int chip_id,
@@ -684,7 +684,7 @@ static int adma_ata_init_one(struct pci_dev *pdev,
 	INIT_LIST_HEAD(&probe_ent->node);
 
 	probe_ent->sht		= adma_port_info[board_idx].sht;
-	probe_ent->host_flags	= adma_port_info[board_idx].host_flags;
+	probe_ent->port_flags	= adma_port_info[board_idx].flags;
 	probe_ent->pio_mask	= adma_port_info[board_idx].pio_mask;
 	probe_ent->mwdma_mask	= adma_port_info[board_idx].mwdma_mask;
 	probe_ent->udma_mask	= adma_port_info[board_idx].udma_mask;
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index a2915a56accd..34f1939b44c9 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -342,7 +342,7 @@ static u32 mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in);
 static void mv5_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val);
 static void mv_phy_reset(struct ata_port *ap);
 static void __mv_phy_reset(struct ata_port *ap, int can_sleep);
-static void mv_host_stop(struct ata_host_set *host_set);
+static void mv_host_stop(struct ata_host *host);
 static int mv_port_start(struct ata_port *ap);
 static void mv_port_stop(struct ata_port *ap);
 static void mv_qc_prep(struct ata_queued_cmd *qc);
@@ -480,35 +480,35 @@ static const struct ata_port_operations mv_iie_ops = {
 static const struct ata_port_info mv_port_info[] = {
 	{  /* chip_504x */
 		.sht		= &mv_sht,
-		.host_flags	= MV_COMMON_FLAGS,
+		.flags		= MV_COMMON_FLAGS,
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
 		.port_ops	= &mv5_ops,
 	},
 	{  /* chip_508x */
 		.sht		= &mv_sht,
-		.host_flags	= (MV_COMMON_FLAGS | MV_FLAG_DUAL_HC),
+		.flags		= (MV_COMMON_FLAGS | MV_FLAG_DUAL_HC),
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
 		.port_ops	= &mv5_ops,
 	},
 	{  /* chip_5080 */
 		.sht		= &mv_sht,
-		.host_flags	= (MV_COMMON_FLAGS | MV_FLAG_DUAL_HC),
+		.flags		= (MV_COMMON_FLAGS | MV_FLAG_DUAL_HC),
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
 		.port_ops	= &mv5_ops,
 	},
 	{  /* chip_604x */
 		.sht		= &mv_sht,
-		.host_flags	= (MV_COMMON_FLAGS | MV_6XXX_FLAGS),
+		.flags		= (MV_COMMON_FLAGS | MV_6XXX_FLAGS),
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
 		.port_ops	= &mv6_ops,
 	},
 	{  /* chip_608x */
 		.sht		= &mv_sht,
-		.host_flags	= (MV_COMMON_FLAGS | MV_6XXX_FLAGS |
+		.flags		= (MV_COMMON_FLAGS | MV_6XXX_FLAGS |
 				   MV_FLAG_DUAL_HC),
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
@@ -516,14 +516,14 @@ static const struct ata_port_info mv_port_info[] = {
 	},
 	{  /* chip_6042 */
 		.sht		= &mv_sht,
-		.host_flags	= (MV_COMMON_FLAGS | MV_6XXX_FLAGS),
+		.flags		= (MV_COMMON_FLAGS | MV_6XXX_FLAGS),
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
 		.port_ops	= &mv_iie_ops,
 	},
 	{  /* chip_7042 */
 		.sht		= &mv_sht,
-		.host_flags	= (MV_COMMON_FLAGS | MV_6XXX_FLAGS |
+		.flags		= (MV_COMMON_FLAGS | MV_6XXX_FLAGS |
 				   MV_FLAG_DUAL_HC),
 		.pio_mask	= 0x1f,	/* pio0-4 */
 		.udma_mask	= 0x7f,	/* udma0-6 */
@@ -618,12 +618,12 @@ static inline void __iomem *mv_port_base(void __iomem *base, unsigned int port)
 
 static inline void __iomem *mv_ap_base(struct ata_port *ap)
 {
-	return mv_port_base(ap->host_set->mmio_base, ap->port_no);
+	return mv_port_base(ap->host->mmio_base, ap->port_no);
 }
 
-static inline int mv_get_hc_count(unsigned long host_flags)
+static inline int mv_get_hc_count(unsigned long port_flags)
 {
-	return ((host_flags & MV_FLAG_DUAL_HC) ? 2 : 1);
+	return ((port_flags & MV_FLAG_DUAL_HC) ? 2 : 1);
 }
 
 static void mv_irq_clear(struct ata_port *ap)
@@ -809,7 +809,7 @@ static void mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val)
 
 /**
  *      mv_host_stop - Host specific cleanup/stop routine.
- *      @host_set: host data structure
+ *      @host: host data structure
  *
  *      Disable ints, cleanup host memory, call general purpose
  *      host_stop.
@@ -817,10 +817,10 @@ static void mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val)
  *      LOCKING:
  *      Inherited from caller.
  */
-static void mv_host_stop(struct ata_host_set *host_set)
+static void mv_host_stop(struct ata_host *host)
 {
-	struct mv_host_priv *hpriv = host_set->private_data;
-	struct pci_dev *pdev = to_pci_dev(host_set->dev);
+	struct mv_host_priv *hpriv = host->private_data;
+	struct pci_dev *pdev = to_pci_dev(host->dev);
 
 	if (hpriv->hp_flags & MV_HP_FLAG_MSI) {
 		pci_disable_msi(pdev);
@@ -828,7 +828,7 @@ static void mv_host_stop(struct ata_host_set *host_set)
 		pci_intx(pdev, 0);
 	}
 	kfree(hpriv);
-	ata_host_stop(host_set);
+	ata_host_stop(host);
 }
 
 static inline void mv_priv_free(struct mv_port_priv *pp, struct device *dev)
@@ -875,8 +875,8 @@ static void mv_edma_cfg(struct mv_host_priv *hpriv, void __iomem *port_mmio)
  */
 static int mv_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
-	struct mv_host_priv *hpriv = ap->host_set->private_data;
+	struct device *dev = ap->host->dev;
+	struct mv_host_priv *hpriv = ap->host->private_data;
 	struct mv_port_priv *pp;
 	void __iomem *port_mmio = mv_ap_base(ap);
 	void *mem;
@@ -965,17 +965,17 @@ err_out:
  *      Stop DMA, cleanup port memory.
  *
  *      LOCKING:
- *      This routine uses the host_set lock to protect the DMA stop.
+ *      This routine uses the host lock to protect the DMA stop.
  */
 static void mv_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct mv_port_priv *pp = ap->private_data;
 	unsigned long flags;
 
-	spin_lock_irqsave(&ap->host_set->lock, flags);
+	spin_lock_irqsave(&ap->host->lock, flags);
 	mv_stop_dma(ap);
-	spin_unlock_irqrestore(&ap->host_set->lock, flags);
+	spin_unlock_irqrestore(&ap->host->lock, flags);
 
 	ap->private_data = NULL;
 	ata_pad_free(ap, dev);
@@ -1330,7 +1330,7 @@ static void mv_err_intr(struct ata_port *ap, int reset_allowed)
 
 /**
  *      mv_host_intr - Handle all interrupts on the given host controller
- *      @host_set: host specific structure
+ *      @host: host specific structure
  *      @relevant: port error bits relevant to this host controller
  *      @hc: which host controller we're to look at
  *
@@ -1344,10 +1344,9 @@ static void mv_err_intr(struct ata_port *ap, int reset_allowed)
  *      LOCKING:
  *      Inherited from caller.
  */
-static void mv_host_intr(struct ata_host_set *host_set, u32 relevant,
-			 unsigned int hc)
+static void mv_host_intr(struct ata_host *host, u32 relevant, unsigned int hc)
 {
-	void __iomem *mmio = host_set->mmio_base;
+	void __iomem *mmio = host->mmio_base;
 	void __iomem *hc_mmio = mv_hc_base(mmio, hc);
 	struct ata_queued_cmd *qc;
 	u32 hc_irq_cause;
@@ -1371,7 +1370,7 @@ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant,
 
 	for (port = port0; port < port0 + MV_PORTS_PER_HC; port++) {
 		u8 ata_status = 0;
-		struct ata_port *ap = host_set->ports[port];
+		struct ata_port *ap = host->ports[port];
 		struct mv_port_priv *pp = ap->private_data;
 
 		hard_port = mv_hardport_from_port(port); /* range 0..3 */
@@ -1444,15 +1443,15 @@ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant,
  *      reported here.
  *
  *      LOCKING:
- *      This routine holds the host_set lock while processing pending
+ *      This routine holds the host lock while processing pending
  *      interrupts.
  */
 static irqreturn_t mv_interrupt(int irq, void *dev_instance,
 				struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	unsigned int hc, handled = 0, n_hcs;
-	void __iomem *mmio = host_set->mmio_base;
+	void __iomem *mmio = host->mmio_base;
 	struct mv_host_priv *hpriv;
 	u32 irq_stat;
 
@@ -1465,18 +1464,18 @@ static irqreturn_t mv_interrupt(int irq, void *dev_instance,
 		return IRQ_NONE;
 	}
 
-	n_hcs = mv_get_hc_count(host_set->ports[0]->flags);
-	spin_lock(&host_set->lock);
+	n_hcs = mv_get_hc_count(host->ports[0]->flags);
+	spin_lock(&host->lock);
 
 	for (hc = 0; hc < n_hcs; hc++) {
 		u32 relevant = irq_stat & (HC0_IRQ_PEND << (hc * HC_SHIFT));
 		if (relevant) {
-			mv_host_intr(host_set, relevant, hc);
+			mv_host_intr(host, relevant, hc);
 			handled++;
 		}
 	}
 
-	hpriv = host_set->private_data;
+	hpriv = host->private_data;
 	if (IS_60XX(hpriv)) {
 		/* deal with the interrupt coalescing bits */
 		if (irq_stat & (TRAN_LO_DONE | TRAN_HI_DONE | PORTS_0_7_COAL_DONE)) {
@@ -1491,12 +1490,12 @@ static irqreturn_t mv_interrupt(int irq, void *dev_instance,
 		       readl(mmio + PCI_IRQ_CAUSE_OFS));
 
 		DPRINTK("All regs @ PCI error\n");
-		mv_dump_all_regs(mmio, -1, to_pci_dev(host_set->dev));
+		mv_dump_all_regs(mmio, -1, to_pci_dev(host->dev));
 
 		writelfl(0, mmio + PCI_IRQ_CAUSE_OFS);
 		handled++;
 	}
-	spin_unlock(&host_set->lock);
+	spin_unlock(&host->lock);
 
 	return IRQ_RETVAL(handled);
 }
@@ -1528,7 +1527,7 @@ static unsigned int mv5_scr_offset(unsigned int sc_reg_in)
 
 static u32 mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in)
 {
-	void __iomem *mmio = mv5_phy_base(ap->host_set->mmio_base, ap->port_no);
+	void __iomem *mmio = mv5_phy_base(ap->host->mmio_base, ap->port_no);
 	unsigned int ofs = mv5_scr_offset(sc_reg_in);
 
 	if (ofs != 0xffffffffU)
@@ -1539,7 +1538,7 @@ static u32 mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in)
 
 static void mv5_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val)
 {
-	void __iomem *mmio = mv5_phy_base(ap->host_set->mmio_base, ap->port_no);
+	void __iomem *mmio = mv5_phy_base(ap->host->mmio_base, ap->port_no);
 	unsigned int ofs = mv5_scr_offset(sc_reg_in);
 
 	if (ofs != 0xffffffffU)
@@ -1904,8 +1903,8 @@ static void mv_channel_reset(struct mv_host_priv *hpriv, void __iomem *mmio,
 
 static void mv_stop_and_reset(struct ata_port *ap)
 {
-	struct mv_host_priv *hpriv = ap->host_set->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
+	struct mv_host_priv *hpriv = ap->host->private_data;
+	void __iomem *mmio = ap->host->mmio_base;
 
 	mv_stop_dma(ap);
 
@@ -1936,7 +1935,7 @@ static inline void __msleep(unsigned int msec, int can_sleep)
 static void __mv_phy_reset(struct ata_port *ap, int can_sleep)
 {
 	struct mv_port_priv *pp	= ap->private_data;
-	struct mv_host_priv *hpriv = ap->host_set->private_data;
+	struct mv_host_priv *hpriv = ap->host->private_data;
 	void __iomem *port_mmio = mv_ap_base(ap);
 	struct ata_taskfile tf;
 	struct ata_device *dev = &ap->device[0];
@@ -2035,7 +2034,7 @@ static void mv_phy_reset(struct ata_port *ap)
  *      chip/bus, fail the command, and move on.
  *
  *      LOCKING:
- *      This routine holds the host_set lock while failing the command.
+ *      This routine holds the host lock while failing the command.
  */
 static void mv_eng_timeout(struct ata_port *ap)
 {
@@ -2044,18 +2043,17 @@ static void mv_eng_timeout(struct ata_port *ap)
 
 	ata_port_printk(ap, KERN_ERR, "Entering mv_eng_timeout\n");
 	DPRINTK("All regs @ start of eng_timeout\n");
-	mv_dump_all_regs(ap->host_set->mmio_base, ap->port_no,
-			 to_pci_dev(ap->host_set->dev));
+	mv_dump_all_regs(ap->host->mmio_base, ap->port_no,
+			 to_pci_dev(ap->host->dev));
 
 	qc = ata_qc_from_tag(ap, ap->active_tag);
         printk(KERN_ERR "mmio_base %p ap %p qc %p scsi_cmnd %p &cmnd %p\n",
-	       ap->host_set->mmio_base, ap, qc, qc->scsicmd,
-	       &qc->scsicmd->cmnd);
+	       ap->host->mmio_base, ap, qc, qc->scsicmd, &qc->scsicmd->cmnd);
 
-	spin_lock_irqsave(&ap->host_set->lock, flags);
+	spin_lock_irqsave(&ap->host->lock, flags);
 	mv_err_intr(ap, 0);
 	mv_stop_and_reset(ap);
-	spin_unlock_irqrestore(&ap->host_set->lock, flags);
+	spin_unlock_irqrestore(&ap->host->lock, flags);
 
 	WARN_ON(!(qc->flags & ATA_QCFLAG_ACTIVE));
 	if (qc->flags & ATA_QCFLAG_ACTIVE) {
@@ -2236,7 +2234,7 @@ static int mv_init_host(struct pci_dev *pdev, struct ata_probe_ent *probe_ent,
 	if (rc)
 		goto done;
 
-	n_hc = mv_get_hc_count(probe_ent->host_flags);
+	n_hc = mv_get_hc_count(probe_ent->port_flags);
 	probe_ent->n_ports = MV_PORTS_PER_HC * n_hc;
 
 	for (port = 0; port < probe_ent->n_ports; port++)
@@ -2389,7 +2387,7 @@ static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	memset(hpriv, 0, sizeof(*hpriv));
 
 	probe_ent->sht = mv_port_info[board_idx].sht;
-	probe_ent->host_flags = mv_port_info[board_idx].host_flags;
+	probe_ent->port_flags = mv_port_info[board_idx].flags;
 	probe_ent->pio_mask = mv_port_info[board_idx].pio_mask;
 	probe_ent->udma_mask = mv_port_info[board_idx].udma_mask;
 	probe_ent->port_ops = mv_port_info[board_idx].port_ops;
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index be46df75ab5a..27c22feebf30 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -81,7 +81,7 @@ enum {
 };
 
 static int nv_init_one (struct pci_dev *pdev, const struct pci_device_id *ent);
-static void nv_ck804_host_stop(struct ata_host_set *host_set);
+static void nv_ck804_host_stop(struct ata_host *host);
 static irqreturn_t nv_generic_interrupt(int irq, void *dev_instance,
 					struct pt_regs *regs);
 static irqreturn_t nv_nf2_interrupt(int irq, void *dev_instance,
@@ -257,7 +257,7 @@ static struct ata_port_info nv_port_info[] = {
 	/* generic */
 	{
 		.sht		= &nv_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -266,7 +266,7 @@ static struct ata_port_info nv_port_info[] = {
 	/* nforce2/3 */
 	{
 		.sht		= &nv_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -275,7 +275,7 @@ static struct ata_port_info nv_port_info[] = {
 	/* ck804 */
 	{
 		.sht		= &nv_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 		.pio_mask	= NV_PIO_MASK,
 		.mwdma_mask	= NV_MWDMA_MASK,
 		.udma_mask	= NV_UDMA_MASK,
@@ -292,17 +292,17 @@ MODULE_VERSION(DRV_VERSION);
 static irqreturn_t nv_generic_interrupt(int irq, void *dev_instance,
 					struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	unsigned int i;
 	unsigned int handled = 0;
 	unsigned long flags;
 
-	spin_lock_irqsave(&host_set->lock, flags);
+	spin_lock_irqsave(&host->lock, flags);
 
-	for (i = 0; i < host_set->n_ports; i++) {
+	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap;
 
-		ap = host_set->ports[i];
+		ap = host->ports[i];
 		if (ap &&
 		    !(ap->flags & ATA_FLAG_DISABLED)) {
 			struct ata_queued_cmd *qc;
@@ -318,7 +318,7 @@ static irqreturn_t nv_generic_interrupt(int irq, void *dev_instance,
 
 	}
 
-	spin_unlock_irqrestore(&host_set->lock, flags);
+	spin_unlock_irqrestore(&host->lock, flags);
 
 	return IRQ_RETVAL(handled);
 }
@@ -354,12 +354,12 @@ static int nv_host_intr(struct ata_port *ap, u8 irq_stat)
 	return 1;
 }
 
-static irqreturn_t nv_do_interrupt(struct ata_host_set *host_set, u8 irq_stat)
+static irqreturn_t nv_do_interrupt(struct ata_host *host, u8 irq_stat)
 {
 	int i, handled = 0;
 
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 
 		if (ap && !(ap->flags & ATA_FLAG_DISABLED))
 			handled += nv_host_intr(ap, irq_stat);
@@ -373,14 +373,14 @@ static irqreturn_t nv_do_interrupt(struct ata_host_set *host_set, u8 irq_stat)
 static irqreturn_t nv_nf2_interrupt(int irq, void *dev_instance,
 				    struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	u8 irq_stat;
 	irqreturn_t ret;
 
-	spin_lock(&host_set->lock);
-	irq_stat = inb(host_set->ports[0]->ioaddr.scr_addr + NV_INT_STATUS);
-	ret = nv_do_interrupt(host_set, irq_stat);
-	spin_unlock(&host_set->lock);
+	spin_lock(&host->lock);
+	irq_stat = inb(host->ports[0]->ioaddr.scr_addr + NV_INT_STATUS);
+	ret = nv_do_interrupt(host, irq_stat);
+	spin_unlock(&host->lock);
 
 	return ret;
 }
@@ -388,14 +388,14 @@ static irqreturn_t nv_nf2_interrupt(int irq, void *dev_instance,
 static irqreturn_t nv_ck804_interrupt(int irq, void *dev_instance,
 				      struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	u8 irq_stat;
 	irqreturn_t ret;
 
-	spin_lock(&host_set->lock);
-	irq_stat = readb(host_set->mmio_base + NV_INT_STATUS_CK804);
-	ret = nv_do_interrupt(host_set, irq_stat);
-	spin_unlock(&host_set->lock);
+	spin_lock(&host->lock);
+	irq_stat = readb(host->mmio_base + NV_INT_STATUS_CK804);
+	ret = nv_do_interrupt(host, irq_stat);
+	spin_unlock(&host->lock);
 
 	return ret;
 }
@@ -418,7 +418,7 @@ static void nv_scr_write (struct ata_port *ap, unsigned int sc_reg, u32 val)
 
 static void nv_nf2_freeze(struct ata_port *ap)
 {
-	unsigned long scr_addr = ap->host_set->ports[0]->ioaddr.scr_addr;
+	unsigned long scr_addr = ap->host->ports[0]->ioaddr.scr_addr;
 	int shift = ap->port_no * NV_INT_PORT_SHIFT;
 	u8 mask;
 
@@ -429,7 +429,7 @@ static void nv_nf2_freeze(struct ata_port *ap)
 
 static void nv_nf2_thaw(struct ata_port *ap)
 {
-	unsigned long scr_addr = ap->host_set->ports[0]->ioaddr.scr_addr;
+	unsigned long scr_addr = ap->host->ports[0]->ioaddr.scr_addr;
 	int shift = ap->port_no * NV_INT_PORT_SHIFT;
 	u8 mask;
 
@@ -442,7 +442,7 @@ static void nv_nf2_thaw(struct ata_port *ap)
 
 static void nv_ck804_freeze(struct ata_port *ap)
 {
-	void __iomem *mmio_base = ap->host_set->mmio_base;
+	void __iomem *mmio_base = ap->host->mmio_base;
 	int shift = ap->port_no * NV_INT_PORT_SHIFT;
 	u8 mask;
 
@@ -453,7 +453,7 @@ static void nv_ck804_freeze(struct ata_port *ap)
 
 static void nv_ck804_thaw(struct ata_port *ap)
 {
-	void __iomem *mmio_base = ap->host_set->mmio_base;
+	void __iomem *mmio_base = ap->host->mmio_base;
 	int shift = ap->port_no * NV_INT_PORT_SHIFT;
 	u8 mask;
 
@@ -568,9 +568,9 @@ err_out:
 	return rc;
 }
 
-static void nv_ck804_host_stop(struct ata_host_set *host_set)
+static void nv_ck804_host_stop(struct ata_host *host)
 {
-	struct pci_dev *pdev = to_pci_dev(host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(host->dev);
 	u8 regval;
 
 	/* disable SATA space for CK804 */
@@ -578,7 +578,7 @@ static void nv_ck804_host_stop(struct ata_host_set *host_set)
 	regval &= ~NV_MCP_SATA_CFG_20_SATA_SPACE_EN;
 	pci_write_config_byte(pdev, NV_MCP_SATA_CFG_20, regval);
 
-	ata_pci_host_stop(host_set);
+	ata_pci_host_stop(host);
 }
 
 static int __init nv_init(void)
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index a5b3a7db7a9f..d627812ea73d 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -104,7 +104,7 @@ static void pdc_tf_load_mmio(struct ata_port *ap, const struct ata_taskfile *tf)
 static void pdc_exec_command_mmio(struct ata_port *ap, const struct ata_taskfile *tf);
 static void pdc_irq_clear(struct ata_port *ap);
 static unsigned int pdc_qc_issue_prot(struct ata_queued_cmd *qc);
-static void pdc_host_stop(struct ata_host_set *host_set);
+static void pdc_host_stop(struct ata_host *host);
 
 
 static struct scsi_host_template pdc_ata_sht = {
@@ -175,7 +175,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_2037x */
 	{
 		.sht		= &pdc_ata_sht,
-		.host_flags	= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
+		.flags		= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
 		.udma_mask	= 0x7f, /* udma0-6 ; FIXME */
@@ -185,7 +185,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_20319 */
 	{
 		.sht		= &pdc_ata_sht,
-		.host_flags	= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
+		.flags		= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
 		.udma_mask	= 0x7f, /* udma0-6 ; FIXME */
@@ -195,7 +195,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_20619 */
 	{
 		.sht		= &pdc_ata_sht,
-		.host_flags	= PDC_COMMON_FLAGS | ATA_FLAG_SLAVE_POSS,
+		.flags		= PDC_COMMON_FLAGS | ATA_FLAG_SLAVE_POSS,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
 		.udma_mask	= 0x7f, /* udma0-6 ; FIXME */
@@ -205,7 +205,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_20771 */
 	{
 		.sht		= &pdc_ata_sht,
-		.host_flags	= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
+		.flags		= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
 		.udma_mask	= 0x7f, /* udma0-6 ; FIXME */
@@ -215,7 +215,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_2057x */
 	{
 		.sht		= &pdc_ata_sht,
-		.host_flags	= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
+		.flags		= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
 		.udma_mask	= 0x7f, /* udma0-6 ; FIXME */
@@ -225,7 +225,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_40518 */
 	{
 		.sht		= &pdc_ata_sht,
-		.host_flags	= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
+		.flags		= PDC_COMMON_FLAGS | ATA_FLAG_SATA,
 		.pio_mask	= 0x1f, /* pio0-4 */
 		.mwdma_mask	= 0x07, /* mwdma0-2 */
 		.udma_mask	= 0x7f, /* udma0-6 ; FIXME */
@@ -292,7 +292,7 @@ static struct pci_driver pdc_ata_pci_driver = {
 
 static int pdc_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct pdc_port_priv *pp;
 	int rc;
 
@@ -326,7 +326,7 @@ err_out:
 
 static void pdc_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct pdc_port_priv *pp = ap->private_data;
 
 	ap->private_data = NULL;
@@ -336,11 +336,11 @@ static void pdc_port_stop(struct ata_port *ap)
 }
 
 
-static void pdc_host_stop(struct ata_host_set *host_set)
+static void pdc_host_stop(struct ata_host *host)
 {
-	struct pdc_host_priv *hp = host_set->private_data;
+	struct pdc_host_priv *hp = host->private_data;
 
-	ata_pci_host_stop(host_set);
+	ata_pci_host_stop(host);
 
 	kfree(hp);
 }
@@ -443,14 +443,14 @@ static void pdc_qc_prep(struct ata_queued_cmd *qc)
 
 static void pdc_eng_timeout(struct ata_port *ap)
 {
-	struct ata_host_set *host_set = ap->host_set;
+	struct ata_host *host = ap->host;
 	u8 drv_stat;
 	struct ata_queued_cmd *qc;
 	unsigned long flags;
 
 	DPRINTK("ENTER\n");
 
-	spin_lock_irqsave(&host_set->lock, flags);
+	spin_lock_irqsave(&host->lock, flags);
 
 	qc = ata_qc_from_tag(ap, ap->active_tag);
 
@@ -473,7 +473,7 @@ static void pdc_eng_timeout(struct ata_port *ap)
 		break;
 	}
 
-	spin_unlock_irqrestore(&host_set->lock, flags);
+	spin_unlock_irqrestore(&host->lock, flags);
 	ata_eh_qc_complete(qc);
 	DPRINTK("EXIT\n");
 }
@@ -509,15 +509,15 @@ static inline unsigned int pdc_host_intr( struct ata_port *ap,
 
 static void pdc_irq_clear(struct ata_port *ap)
 {
-	struct ata_host_set *host_set = ap->host_set;
-	void __iomem *mmio = host_set->mmio_base;
+	struct ata_host *host = ap->host;
+	void __iomem *mmio = host->mmio_base;
 
 	readl(mmio + PDC_INT_SEQMASK);
 }
 
 static irqreturn_t pdc_interrupt (int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	struct ata_port *ap;
 	u32 mask = 0;
 	unsigned int i, tmp;
@@ -526,12 +526,12 @@ static irqreturn_t pdc_interrupt (int irq, void *dev_instance, struct pt_regs *r
 
 	VPRINTK("ENTER\n");
 
-	if (!host_set || !host_set->mmio_base) {
+	if (!host || !host->mmio_base) {
 		VPRINTK("QUICK EXIT\n");
 		return IRQ_NONE;
 	}
 
-	mmio_base = host_set->mmio_base;
+	mmio_base = host->mmio_base;
 
 	/* reading should also clear interrupts */
 	mask = readl(mmio_base + PDC_INT_SEQMASK);
@@ -541,7 +541,7 @@ static irqreturn_t pdc_interrupt (int irq, void *dev_instance, struct pt_regs *r
 		return IRQ_NONE;
 	}
 
-	spin_lock(&host_set->lock);
+	spin_lock(&host->lock);
 
 	mask &= 0xffff;		/* only 16 tags possible */
 	if (!mask) {
@@ -551,9 +551,9 @@ static irqreturn_t pdc_interrupt (int irq, void *dev_instance, struct pt_regs *r
 
 	writel(mask, mmio_base + PDC_INT_SEQMASK);
 
-	for (i = 0; i < host_set->n_ports; i++) {
+	for (i = 0; i < host->n_ports; i++) {
 		VPRINTK("port %u\n", i);
-		ap = host_set->ports[i];
+		ap = host->ports[i];
 		tmp = mask & (1 << (i + 1));
 		if (tmp && ap &&
 		    !(ap->flags & ATA_FLAG_DISABLED)) {
@@ -568,7 +568,7 @@ static irqreturn_t pdc_interrupt (int irq, void *dev_instance, struct pt_regs *r
 	VPRINTK("EXIT\n");
 
 done_irq:
-	spin_unlock(&host_set->lock);
+	spin_unlock(&host->lock);
 	return IRQ_RETVAL(handled);
 }
 
@@ -581,8 +581,8 @@ static inline void pdc_packet_start(struct ata_queued_cmd *qc)
 
 	VPRINTK("ENTER, ap %p\n", ap);
 
-	writel(0x00000001, ap->host_set->mmio_base + (seq * 4));
-	readl(ap->host_set->mmio_base + (seq * 4));	/* flush */
+	writel(0x00000001, ap->host->mmio_base + (seq * 4));
+	readl(ap->host->mmio_base + (seq * 4));	/* flush */
 
 	pp->pkt[2] = seq;
 	wmb();			/* flush PRD, pkt writes */
@@ -743,7 +743,7 @@ static int pdc_ata_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	probe_ent->private_data = hp;
 
 	probe_ent->sht		= pdc_port_info[board_idx].sht;
-	probe_ent->host_flags	= pdc_port_info[board_idx].host_flags;
+	probe_ent->port_flags	= pdc_port_info[board_idx].flags;
 	probe_ent->pio_mask	= pdc_port_info[board_idx].pio_mask;
 	probe_ent->mwdma_mask	= pdc_port_info[board_idx].mwdma_mask;
 	probe_ent->udma_mask	= pdc_port_info[board_idx].udma_mask;
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 71bd6712b377..fa29dfe2a7b5 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -116,7 +116,7 @@ static void qs_scr_write (struct ata_port *ap, unsigned int sc_reg, u32 val);
 static int qs_ata_init_one (struct pci_dev *pdev, const struct pci_device_id *ent);
 static irqreturn_t qs_intr (int irq, void *dev_instance, struct pt_regs *regs);
 static int qs_port_start(struct ata_port *ap);
-static void qs_host_stop(struct ata_host_set *host_set);
+static void qs_host_stop(struct ata_host *host);
 static void qs_port_stop(struct ata_port *ap);
 static void qs_phy_reset(struct ata_port *ap);
 static void qs_qc_prep(struct ata_queued_cmd *qc);
@@ -174,7 +174,7 @@ static const struct ata_port_info qs_port_info[] = {
 	/* board_2068_idx */
 	{
 		.sht		= &qs_ata_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_SATA_RESET |
 				  //FIXME ATA_FLAG_SRST |
 				  ATA_FLAG_MMIO | ATA_FLAG_PIO_POLLING,
@@ -220,7 +220,7 @@ static void qs_irq_clear(struct ata_port *ap)
 
 static inline void qs_enter_reg_mode(struct ata_port *ap)
 {
-	u8 __iomem *chan = ap->host_set->mmio_base + (ap->port_no * 0x4000);
+	u8 __iomem *chan = ap->host->mmio_base + (ap->port_no * 0x4000);
 
 	writeb(QS_CTR0_REG, chan + QS_CCT_CTR0);
 	readb(chan + QS_CCT_CTR0);        /* flush */
@@ -228,7 +228,7 @@ static inline void qs_enter_reg_mode(struct ata_port *ap)
 
 static inline void qs_reset_channel_logic(struct ata_port *ap)
 {
-	u8 __iomem *chan = ap->host_set->mmio_base + (ap->port_no * 0x4000);
+	u8 __iomem *chan = ap->host->mmio_base + (ap->port_no * 0x4000);
 
 	writeb(QS_CTR1_RCHN, chan + QS_CCT_CTR1);
 	readb(chan + QS_CCT_CTR0);        /* flush */
@@ -342,7 +342,7 @@ static void qs_qc_prep(struct ata_queued_cmd *qc)
 static inline void qs_packet_start(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	u8 __iomem *chan = ap->host_set->mmio_base + (ap->port_no * 0x4000);
+	u8 __iomem *chan = ap->host->mmio_base + (ap->port_no * 0x4000);
 
 	VPRINTK("ENTER, ap %p\n", ap);
 
@@ -375,11 +375,11 @@ static unsigned int qs_qc_issue(struct ata_queued_cmd *qc)
 	return ata_qc_issue_prot(qc);
 }
 
-static inline unsigned int qs_intr_pkt(struct ata_host_set *host_set)
+static inline unsigned int qs_intr_pkt(struct ata_host *host)
 {
 	unsigned int handled = 0;
 	u8 sFFE;
-	u8 __iomem *mmio_base = host_set->mmio_base;
+	u8 __iomem *mmio_base = host->mmio_base;
 
 	do {
 		u32 sff0 = readl(mmio_base + QS_HST_SFF);
@@ -391,7 +391,7 @@ static inline unsigned int qs_intr_pkt(struct ata_host_set *host_set)
 			u8 sDST = sff0 >> 16;	/* dev status */
 			u8 sHST = sff1 & 0x3f;	/* host status */
 			unsigned int port_no = (sff1 >> 8) & 0x03;
-			struct ata_port *ap = host_set->ports[port_no];
+			struct ata_port *ap = host->ports[port_no];
 
 			DPRINTK("SFF=%08x%08x: sCHAN=%u sHST=%d sDST=%02x\n",
 					sff1, sff0, port_no, sHST, sDST);
@@ -421,13 +421,13 @@ static inline unsigned int qs_intr_pkt(struct ata_host_set *host_set)
 	return handled;
 }
 
-static inline unsigned int qs_intr_mmio(struct ata_host_set *host_set)
+static inline unsigned int qs_intr_mmio(struct ata_host *host)
 {
 	unsigned int handled = 0, port_no;
 
-	for (port_no = 0; port_no < host_set->n_ports; ++port_no) {
+	for (port_no = 0; port_no < host->n_ports; ++port_no) {
 		struct ata_port *ap;
-		ap = host_set->ports[port_no];
+		ap = host->ports[port_no];
 		if (ap &&
 		    !(ap->flags & ATA_FLAG_DISABLED)) {
 			struct ata_queued_cmd *qc;
@@ -457,14 +457,14 @@ static inline unsigned int qs_intr_mmio(struct ata_host_set *host_set)
 
 static irqreturn_t qs_intr(int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	unsigned int handled = 0;
 
 	VPRINTK("ENTER\n");
 
-	spin_lock(&host_set->lock);
-	handled  = qs_intr_pkt(host_set) | qs_intr_mmio(host_set);
-	spin_unlock(&host_set->lock);
+	spin_lock(&host->lock);
+	handled  = qs_intr_pkt(host) | qs_intr_mmio(host);
+	spin_unlock(&host->lock);
 
 	VPRINTK("EXIT\n");
 
@@ -491,9 +491,9 @@ static void qs_ata_setup_port(struct ata_ioports *port, unsigned long base)
 
 static int qs_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct qs_port_priv *pp;
-	void __iomem *mmio_base = ap->host_set->mmio_base;
+	void __iomem *mmio_base = ap->host->mmio_base;
 	void __iomem *chan = mmio_base + (ap->port_no * 0x4000);
 	u64 addr;
 	int rc;
@@ -530,7 +530,7 @@ err_out:
 
 static void qs_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct qs_port_priv *pp = ap->private_data;
 
 	if (pp != NULL) {
@@ -543,10 +543,10 @@ static void qs_port_stop(struct ata_port *ap)
 	ata_port_stop(ap);
 }
 
-static void qs_host_stop(struct ata_host_set *host_set)
+static void qs_host_stop(struct ata_host *host)
 {
-	void __iomem *mmio_base = host_set->mmio_base;
-	struct pci_dev *pdev = to_pci_dev(host_set->dev);
+	void __iomem *mmio_base = host->mmio_base;
+	struct pci_dev *pdev = to_pci_dev(host->dev);
 
 	writeb(0, mmio_base + QS_HCT_CTRL); /* disable host interrupts */
 	writeb(QS_CNFG3_GSRST, mmio_base + QS_HCF_CNFG3); /* global reset */
@@ -673,7 +673,7 @@ static int qs_ata_init_one(struct pci_dev *pdev,
 	INIT_LIST_HEAD(&probe_ent->node);
 
 	probe_ent->sht		= qs_port_info[board_idx].sht;
-	probe_ent->host_flags	= qs_port_info[board_idx].host_flags;
+	probe_ent->port_flags	= qs_port_info[board_idx].flags;
 	probe_ent->pio_mask	= qs_port_info[board_idx].pio_mask;
 	probe_ent->mwdma_mask	= qs_port_info[board_idx].mwdma_mask;
 	probe_ent->udma_mask	= qs_port_info[board_idx].udma_mask;
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index f17b3ae4dd37..c63dbabc0cd9 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -56,7 +56,7 @@ enum {
 	SIL_FLAG_RERR_ON_DMA_ACT = (1 << 29),
 	SIL_FLAG_MOD15WRITE	= (1 << 30),
 
-	SIL_DFL_HOST_FLAGS	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+	SIL_DFL_PORT_FLAGS	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_MMIO | ATA_FLAG_HRST_TO_RESUME,
 
 	/*
@@ -218,7 +218,7 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3112 */
 	{
 		.sht		= &sil_sht,
-		.host_flags	= SIL_DFL_HOST_FLAGS | SIL_FLAG_MOD15WRITE,
+		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_MOD15WRITE,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -227,7 +227,7 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3112_no_sata_irq */
 	{
 		.sht		= &sil_sht,
-		.host_flags	= SIL_DFL_HOST_FLAGS | SIL_FLAG_MOD15WRITE |
+		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_MOD15WRITE |
 				  SIL_FLAG_NO_SATA_IRQ,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
@@ -237,7 +237,7 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3512 */
 	{
 		.sht		= &sil_sht,
-		.host_flags	= SIL_DFL_HOST_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
+		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -246,7 +246,7 @@ static const struct ata_port_info sil_port_info[] = {
 	/* sil_3114 */
 	{
 		.sht		= &sil_sht,
-		.host_flags	= SIL_DFL_HOST_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
+		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -295,10 +295,9 @@ static unsigned char sil_get_device_cache_line(struct pci_dev *pdev)
 
 static void sil_post_set_mode (struct ata_port *ap)
 {
-	struct ata_host_set *host_set = ap->host_set;
+	struct ata_host *host = ap->host;
 	struct ata_device *dev;
-	void __iomem *addr =
-		host_set->mmio_base + sil_port[ap->port_no].xfer_mode;
+	void __iomem *addr = host->mmio_base + sil_port[ap->port_no].xfer_mode;
 	u32 tmp, dev_mode[2];
 	unsigned int i;
 
@@ -440,15 +439,15 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 static irqreturn_t sil_interrupt(int irq, void *dev_instance,
 				 struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
-	void __iomem *mmio_base = host_set->mmio_base;
+	struct ata_host *host = dev_instance;
+	void __iomem *mmio_base = host->mmio_base;
 	int handled = 0;
 	int i;
 
-	spin_lock(&host_set->lock);
+	spin_lock(&host->lock);
 
-	for (i = 0; i < host_set->n_ports; i++) {
-		struct ata_port *ap = host_set->ports[i];
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap = host->ports[i];
 		u32 bmdma2 = readl(mmio_base + sil_port[ap->port_no].bmdma2);
 
 		if (unlikely(!ap || ap->flags & ATA_FLAG_DISABLED))
@@ -466,14 +465,14 @@ static irqreturn_t sil_interrupt(int irq, void *dev_instance,
 		handled = 1;
 	}
 
-	spin_unlock(&host_set->lock);
+	spin_unlock(&host->lock);
 
 	return IRQ_RETVAL(handled);
 }
 
 static void sil_freeze(struct ata_port *ap)
 {
-	void __iomem *mmio_base = ap->host_set->mmio_base;
+	void __iomem *mmio_base = ap->host->mmio_base;
 	u32 tmp;
 
 	/* global IRQ mask doesn't block SATA IRQ, turn off explicitly */
@@ -488,7 +487,7 @@ static void sil_freeze(struct ata_port *ap)
 
 static void sil_thaw(struct ata_port *ap)
 {
-	void __iomem *mmio_base = ap->host_set->mmio_base;
+	void __iomem *mmio_base = ap->host->mmio_base;
 	u32 tmp;
 
 	/* clear IRQ */
@@ -567,7 +566,7 @@ static void sil_dev_config(struct ata_port *ap, struct ata_device *dev)
 }
 
 static void sil_init_controller(struct pci_dev *pdev,
-				int n_ports, unsigned long host_flags,
+				int n_ports, unsigned long port_flags,
 				void __iomem *mmio_base)
 {
 	u8 cls;
@@ -587,7 +586,7 @@ static void sil_init_controller(struct pci_dev *pdev,
 			   "cache line size not set.  Driver may not function\n");
 
 	/* Apply R_ERR on DMA activate FIS errata workaround */
-	if (host_flags & SIL_FLAG_RERR_ON_DMA_ACT) {
+	if (port_flags & SIL_FLAG_RERR_ON_DMA_ACT) {
 		int cnt;
 
 		for (i = 0, cnt = 0; i < n_ports; i++) {
@@ -658,7 +657,7 @@ static int sil_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	probe_ent->udma_mask = sil_port_info[ent->driver_data].udma_mask;
        	probe_ent->irq = pdev->irq;
        	probe_ent->irq_flags = IRQF_SHARED;
-	probe_ent->host_flags = sil_port_info[ent->driver_data].host_flags;
+	probe_ent->port_flags = sil_port_info[ent->driver_data].flags;
 
 	mmio_base = pci_iomap(pdev, 5, 0);
 	if (mmio_base == NULL) {
@@ -679,7 +678,7 @@ static int sil_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		ata_std_ports(&probe_ent->port[i]);
 	}
 
-	sil_init_controller(pdev, probe_ent->n_ports, probe_ent->host_flags,
+	sil_init_controller(pdev, probe_ent->n_ports, probe_ent->port_flags,
 			    mmio_base);
 
 	pci_set_master(pdev);
@@ -703,12 +702,12 @@ err_out:
 #ifdef CONFIG_PM
 static int sil_pci_device_resume(struct pci_dev *pdev)
 {
-	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
 
 	ata_pci_device_do_resume(pdev);
-	sil_init_controller(pdev, host_set->n_ports, host_set->ports[0]->flags,
-			    host_set->mmio_base);
-	ata_host_set_resume(host_set);
+	sil_init_controller(pdev, host->n_ports, host->ports[0]->flags,
+			    host->mmio_base);
+	ata_host_resume(host);
 
 	return 0;
 }
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 2d7cf3264587..39cb07baebae 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -316,7 +316,7 @@ struct sil24_port_priv {
 	struct ata_taskfile tf;			/* Cached taskfile registers */
 };
 
-/* ap->host_set->private_data */
+/* ap->host->private_data */
 struct sil24_host_priv {
 	void __iomem *host_base;	/* global controller control (128 bytes @BAR0) */
 	void __iomem *port_base;	/* port registers (4 * 8192 bytes @BAR2) */
@@ -337,7 +337,7 @@ static void sil24_error_handler(struct ata_port *ap);
 static void sil24_post_internal_cmd(struct ata_queued_cmd *qc);
 static int sil24_port_start(struct ata_port *ap);
 static void sil24_port_stop(struct ata_port *ap);
-static void sil24_host_stop(struct ata_host_set *host_set);
+static void sil24_host_stop(struct ata_host *host);
 static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 #ifdef CONFIG_PM
 static int sil24_pci_device_resume(struct pci_dev *pdev);
@@ -415,7 +415,7 @@ static const struct ata_port_operations sil24_ops = {
 };
 
 /*
- * Use bits 30-31 of host_flags to encode available port numbers.
+ * Use bits 30-31 of port_flags to encode available port numbers.
  * Current maxium is 4.
  */
 #define SIL24_NPORTS2FLAG(nports)	((((unsigned)(nports) - 1) & 0x3) << 30)
@@ -425,7 +425,7 @@ static struct ata_port_info sil24_port_info[] = {
 	/* sil_3124 */
 	{
 		.sht		= &sil24_sht,
-		.host_flags	= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(4) |
+		.flags		= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(4) |
 				  SIL24_FLAG_PCIX_IRQ_WOC,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
@@ -435,7 +435,7 @@ static struct ata_port_info sil24_port_info[] = {
 	/* sil_3132 */
 	{
 		.sht		= &sil24_sht,
-		.host_flags	= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(2),
+		.flags		= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(2),
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -444,7 +444,7 @@ static struct ata_port_info sil24_port_info[] = {
 	/* sil_3131/sil_3531 */
 	{
 		.sht		= &sil24_sht,
-		.host_flags	= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(1),
+		.flags		= SIL24_COMMON_FLAGS | SIL24_NPORTS2FLAG(1),
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -871,8 +871,8 @@ static inline void sil24_host_intr(struct ata_port *ap)
 
 static irqreturn_t sil24_interrupt(int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
-	struct sil24_host_priv *hpriv = host_set->private_data;
+	struct ata_host *host = dev_instance;
+	struct sil24_host_priv *hpriv = host->private_data;
 	unsigned handled = 0;
 	u32 status;
 	int i;
@@ -888,20 +888,20 @@ static irqreturn_t sil24_interrupt(int irq, void *dev_instance, struct pt_regs *
 	if (!(status & IRQ_STAT_4PORTS))
 		goto out;
 
-	spin_lock(&host_set->lock);
+	spin_lock(&host->lock);
 
-	for (i = 0; i < host_set->n_ports; i++)
+	for (i = 0; i < host->n_ports; i++)
 		if (status & (1 << i)) {
-			struct ata_port *ap = host_set->ports[i];
+			struct ata_port *ap = host->ports[i];
 			if (ap && !(ap->flags & ATA_FLAG_DISABLED)) {
-				sil24_host_intr(host_set->ports[i]);
+				sil24_host_intr(host->ports[i]);
 				handled++;
 			} else
 				printk(KERN_ERR DRV_NAME
 				       ": interrupt from disabled port %d\n", i);
 		}
 
-	spin_unlock(&host_set->lock);
+	spin_unlock(&host->lock);
  out:
 	return IRQ_RETVAL(handled);
 }
@@ -941,7 +941,7 @@ static inline void sil24_cblk_free(struct sil24_port_priv *pp, struct device *de
 
 static int sil24_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct sil24_port_priv *pp;
 	union sil24_cmd_block *cb;
 	size_t cb_size = sizeof(*cb) * SIL24_MAX_CMDS;
@@ -980,7 +980,7 @@ err_out:
 
 static void sil24_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct sil24_port_priv *pp = ap->private_data;
 
 	sil24_cblk_free(pp, dev);
@@ -988,10 +988,10 @@ static void sil24_port_stop(struct ata_port *ap)
 	kfree(pp);
 }
 
-static void sil24_host_stop(struct ata_host_set *host_set)
+static void sil24_host_stop(struct ata_host *host)
 {
-	struct sil24_host_priv *hpriv = host_set->private_data;
-	struct pci_dev *pdev = to_pci_dev(host_set->dev);
+	struct sil24_host_priv *hpriv = host->private_data;
+	struct pci_dev *pdev = to_pci_dev(host->dev);
 
 	pci_iounmap(pdev, hpriv->host_base);
 	pci_iounmap(pdev, hpriv->port_base);
@@ -999,7 +999,7 @@ static void sil24_host_stop(struct ata_host_set *host_set)
 }
 
 static void sil24_init_controller(struct pci_dev *pdev, int n_ports,
-				  unsigned long host_flags,
+				  unsigned long port_flags,
 				  void __iomem *host_base,
 				  void __iomem *port_base)
 {
@@ -1032,7 +1032,7 @@ static void sil24_init_controller(struct pci_dev *pdev, int n_ports,
 		}
 
 		/* Configure IRQ WoC */
-		if (host_flags & SIL24_FLAG_PCIX_IRQ_WOC)
+		if (port_flags & SIL24_FLAG_PCIX_IRQ_WOC)
 			writel(PORT_CS_IRQ_WOC, port + PORT_CTRL_STAT);
 		else
 			writel(PORT_CS_IRQ_WOC, port + PORT_CTRL_CLR);
@@ -1101,12 +1101,12 @@ static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	INIT_LIST_HEAD(&probe_ent->node);
 
 	probe_ent->sht		= pinfo->sht;
-	probe_ent->host_flags	= pinfo->host_flags;
+	probe_ent->port_flags	= pinfo->flags;
 	probe_ent->pio_mask	= pinfo->pio_mask;
 	probe_ent->mwdma_mask	= pinfo->mwdma_mask;
 	probe_ent->udma_mask	= pinfo->udma_mask;
 	probe_ent->port_ops	= pinfo->port_ops;
-	probe_ent->n_ports	= SIL24_FLAG2NPORTS(pinfo->host_flags);
+	probe_ent->n_ports	= SIL24_FLAG2NPORTS(pinfo->flags);
 
 	probe_ent->irq = pdev->irq;
 	probe_ent->irq_flags = IRQF_SHARED;
@@ -1144,14 +1144,14 @@ static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	/* Apply workaround for completion IRQ loss on PCI-X errata */
-	if (probe_ent->host_flags & SIL24_FLAG_PCIX_IRQ_WOC) {
+	if (probe_ent->port_flags & SIL24_FLAG_PCIX_IRQ_WOC) {
 		tmp = readl(host_base + HOST_CTRL);
 		if (tmp & (HOST_CTRL_TRDY | HOST_CTRL_STOP | HOST_CTRL_DEVSEL))
 			dev_printk(KERN_INFO, &pdev->dev,
 				   "Applying completion IRQ loss on PCI-X "
 				   "errata fix\n");
 		else
-			probe_ent->host_flags &= ~SIL24_FLAG_PCIX_IRQ_WOC;
+			probe_ent->port_flags &= ~SIL24_FLAG_PCIX_IRQ_WOC;
 	}
 
 	for (i = 0; i < probe_ent->n_ports; i++) {
@@ -1164,7 +1164,7 @@ static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		ata_std_ports(&probe_ent->port[i]);
 	}
 
-	sil24_init_controller(pdev, probe_ent->n_ports, probe_ent->host_flags,
+	sil24_init_controller(pdev, probe_ent->n_ports, probe_ent->port_flags,
 			      host_base, port_base);
 
 	pci_set_master(pdev);
@@ -1191,19 +1191,18 @@ static int sil24_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 #ifdef CONFIG_PM
 static int sil24_pci_device_resume(struct pci_dev *pdev)
 {
-	struct ata_host_set *host_set = dev_get_drvdata(&pdev->dev);
-	struct sil24_host_priv *hpriv = host_set->private_data;
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
+	struct sil24_host_priv *hpriv = host->private_data;
 
 	ata_pci_device_do_resume(pdev);
 
 	if (pdev->dev.power.power_state.event == PM_EVENT_SUSPEND)
 		writel(HOST_CTRL_GLOBAL_RST, hpriv->host_base + HOST_CTRL);
 
-	sil24_init_controller(pdev, host_set->n_ports,
-			      host_set->ports[0]->flags,
+	sil24_init_controller(pdev, host->n_ports, host->ports[0]->flags,
 			      hpriv->host_base, hpriv->port_base);
 
-	ata_host_set_resume(host_set);
+	ata_host_resume(host);
 
 	return 0;
 }
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index ac24f66897f6..9b17375d8056 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -128,7 +128,7 @@ static const struct ata_port_operations sis_ops = {
 
 static struct ata_port_info sis_port_info = {
 	.sht		= &sis_sht,
-	.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
+	.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 	.pio_mask	= 0x1f,
 	.mwdma_mask	= 0x7,
 	.udma_mask	= 0x7f,
@@ -158,7 +158,7 @@ static unsigned int get_scr_cfg_addr(unsigned int port_no, unsigned int sc_reg,
 
 static u32 sis_scr_cfg_read (struct ata_port *ap, unsigned int sc_reg)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	unsigned int cfg_addr = get_scr_cfg_addr(ap->port_no, sc_reg, pdev->device);
 	u32 val, val2 = 0;
 	u8 pmr;
@@ -178,7 +178,7 @@ static u32 sis_scr_cfg_read (struct ata_port *ap, unsigned int sc_reg)
 
 static void sis_scr_cfg_write (struct ata_port *ap, unsigned int scr, u32 val)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	unsigned int cfg_addr = get_scr_cfg_addr(ap->port_no, scr, pdev->device);
 	u8 pmr;
 
@@ -195,7 +195,7 @@ static void sis_scr_cfg_write (struct ata_port *ap, unsigned int scr, u32 val)
 
 static u32 sis_scr_read (struct ata_port *ap, unsigned int sc_reg)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u32 val, val2 = 0;
 	u8 pmr;
 
@@ -217,7 +217,7 @@ static u32 sis_scr_read (struct ata_port *ap, unsigned int sc_reg)
 
 static void sis_scr_write (struct ata_port *ap, unsigned int sc_reg, u32 val)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 pmr;
 
 	if (sc_reg > SCR_CONTROL)
@@ -275,17 +275,17 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* check and see if the SCRs are in IO space or PCI cfg space */
 	pci_read_config_dword(pdev, SIS_GENCTL, &genctl);
 	if ((genctl & GENCTL_IOMAPPED_SCR) == 0)
-		probe_ent->host_flags |= SIS_FLAG_CFGSCR;
+		probe_ent->port_flags |= SIS_FLAG_CFGSCR;
 
 	/* if hardware thinks SCRs are in IO space, but there are
 	 * no IO resources assigned, change to PCI cfg space.
 	 */
-	if ((!(probe_ent->host_flags & SIS_FLAG_CFGSCR)) &&
+	if ((!(probe_ent->port_flags & SIS_FLAG_CFGSCR)) &&
 	    ((pci_resource_start(pdev, SIS_SCR_PCI_BAR) == 0) ||
 	     (pci_resource_len(pdev, SIS_SCR_PCI_BAR) < 128))) {
 		genctl &= ~GENCTL_IOMAPPED_SCR;
 		pci_write_config_dword(pdev, SIS_GENCTL, genctl);
-		probe_ent->host_flags |= SIS_FLAG_CFGSCR;
+		probe_ent->port_flags |= SIS_FLAG_CFGSCR;
 	}
 
 	pci_read_config_byte(pdev, SIS_PMR, &pmr);
@@ -306,7 +306,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		port2_start = 0x20;
 	}
 
-	if (!(probe_ent->host_flags & SIS_FLAG_CFGSCR)) {
+	if (!(probe_ent->port_flags & SIS_FLAG_CFGSCR)) {
 		probe_ent->port[0].scr_addr =
 			pci_resource_start(pdev, SIS_SCR_PCI_BAR);
 		probe_ent->port[1].scr_addr =
diff --git a/drivers/ata/sata_svw.c b/drivers/ata/sata_svw.c
index baf259a966d0..2a7e3495cf16 100644
--- a/drivers/ata/sata_svw.c
+++ b/drivers/ata/sata_svw.c
@@ -169,7 +169,7 @@ static void k2_sata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
  *	@qc: Info associated with this ATA transaction.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void k2_bmdma_setup_mmio (struct ata_queued_cmd *qc)
@@ -199,7 +199,7 @@ static void k2_bmdma_setup_mmio (struct ata_queued_cmd *qc)
  *	@qc: Info associated with this ATA transaction.
  *
  *	LOCKING:
- *	spin_lock_irqsave(host_set lock)
+ *	spin_lock_irqsave(host lock)
  */
 
 static void k2_bmdma_start_mmio (struct ata_queued_cmd *qc)
@@ -261,12 +261,12 @@ static int k2_sata_proc_info(struct Scsi_Host *shost, char *page, char **start,
 		return 0;
 
 	/* Find the OF node for the PCI device proper */
-	np = pci_device_to_OF_node(to_pci_dev(ap->host_set->dev));
+	np = pci_device_to_OF_node(to_pci_dev(ap->host->dev));
 	if (np == NULL)
 		return 0;
 
 	/* Match it to a port node */
-	index = (ap == ap->host_set->ports[0]) ? 0 : 1;
+	index = (ap == ap->host->ports[0]) ? 0 : 1;
 	for (np = np->child; np != NULL; np = np->sibling) {
 		u32 *reg = (u32 *)get_property(np, "reg", NULL);
 		if (!reg)
@@ -423,7 +423,7 @@ static int k2_sata_init_one (struct pci_dev *pdev, const struct pci_device_id *e
 	writel(0x0, mmio_base + K2_SATA_SIM_OFFSET);
 
 	probe_ent->sht = &k2_sata_sht;
-	probe_ent->host_flags = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+	probe_ent->port_flags = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				ATA_FLAG_MMIO;
 	probe_ent->port_ops = &k2_sata_ops;
 	probe_ent->n_ports = 4;
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index 0da83cba5c12..091867e10ea3 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -160,7 +160,7 @@ static void pdc_port_stop(struct ata_port *ap);
 static void pdc20621_qc_prep(struct ata_queued_cmd *qc);
 static void pdc_tf_load_mmio(struct ata_port *ap, const struct ata_taskfile *tf);
 static void pdc_exec_command_mmio(struct ata_port *ap, const struct ata_taskfile *tf);
-static void pdc20621_host_stop(struct ata_host_set *host_set);
+static void pdc20621_host_stop(struct ata_host *host);
 static unsigned int pdc20621_dimm_init(struct ata_probe_ent *pe);
 static int pdc20621_detect_dimm(struct ata_probe_ent *pe);
 static unsigned int pdc20621_i2c_read(struct ata_probe_ent *pe,
@@ -218,7 +218,7 @@ static const struct ata_port_info pdc_port_info[] = {
 	/* board_20621 */
 	{
 		.sht		= &pdc_sata_sht,
-		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+		.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				  ATA_FLAG_SRST | ATA_FLAG_MMIO |
 				  ATA_FLAG_NO_ATAPI | ATA_FLAG_PIO_POLLING,
 		.pio_mask	= 0x1f, /* pio0-4 */
@@ -244,21 +244,21 @@ static struct pci_driver pdc_sata_pci_driver = {
 };
 
 
-static void pdc20621_host_stop(struct ata_host_set *host_set)
+static void pdc20621_host_stop(struct ata_host *host)
 {
-	struct pci_dev *pdev = to_pci_dev(host_set->dev);
-	struct pdc_host_priv *hpriv = host_set->private_data;
+	struct pci_dev *pdev = to_pci_dev(host->dev);
+	struct pdc_host_priv *hpriv = host->private_data;
 	void __iomem *dimm_mmio = hpriv->dimm_mmio;
 
 	pci_iounmap(pdev, dimm_mmio);
 	kfree(hpriv);
 
-	pci_iounmap(pdev, host_set->mmio_base);
+	pci_iounmap(pdev, host->mmio_base);
 }
 
 static int pdc_port_start(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct pdc_port_priv *pp;
 	int rc;
 
@@ -293,7 +293,7 @@ err_out:
 
 static void pdc_port_stop(struct ata_port *ap)
 {
-	struct device *dev = ap->host_set->dev;
+	struct device *dev = ap->host->dev;
 	struct pdc_port_priv *pp = ap->private_data;
 
 	ap->private_data = NULL;
@@ -453,8 +453,8 @@ static void pdc20621_dma_prep(struct ata_queued_cmd *qc)
 	struct scatterlist *sg;
 	struct ata_port *ap = qc->ap;
 	struct pdc_port_priv *pp = ap->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
-	struct pdc_host_priv *hpriv = ap->host_set->private_data;
+	void __iomem *mmio = ap->host->mmio_base;
+	struct pdc_host_priv *hpriv = ap->host->private_data;
 	void __iomem *dimm_mmio = hpriv->dimm_mmio;
 	unsigned int portno = ap->port_no;
 	unsigned int i, idx, total_len = 0, sgt_len;
@@ -514,8 +514,8 @@ static void pdc20621_nodata_prep(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	struct pdc_port_priv *pp = ap->private_data;
-	void __iomem *mmio = ap->host_set->mmio_base;
-	struct pdc_host_priv *hpriv = ap->host_set->private_data;
+	void __iomem *mmio = ap->host->mmio_base;
+	struct pdc_host_priv *hpriv = ap->host->private_data;
 	void __iomem *dimm_mmio = hpriv->dimm_mmio;
 	unsigned int portno = ap->port_no;
 	unsigned int i;
@@ -565,8 +565,8 @@ static void __pdc20621_push_hdma(struct ata_queued_cmd *qc,
 				 u32 pkt_ofs)
 {
 	struct ata_port *ap = qc->ap;
-	struct ata_host_set *host_set = ap->host_set;
-	void __iomem *mmio = host_set->mmio_base;
+	struct ata_host *host = ap->host;
+	void __iomem *mmio = host->mmio_base;
 
 	/* hard-code chip #0 */
 	mmio += PDC_CHIP0_OFS;
@@ -583,7 +583,7 @@ static void pdc20621_push_hdma(struct ata_queued_cmd *qc,
 				u32 pkt_ofs)
 {
 	struct ata_port *ap = qc->ap;
-	struct pdc_host_priv *pp = ap->host_set->private_data;
+	struct pdc_host_priv *pp = ap->host->private_data;
 	unsigned int idx = pp->hdma_prod & PDC_HDMA_Q_MASK;
 
 	if (!pp->doing_hdma) {
@@ -601,7 +601,7 @@ static void pdc20621_push_hdma(struct ata_queued_cmd *qc,
 static void pdc20621_pop_hdma(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	struct pdc_host_priv *pp = ap->host_set->private_data;
+	struct pdc_host_priv *pp = ap->host->private_data;
 	unsigned int idx = pp->hdma_cons & PDC_HDMA_Q_MASK;
 
 	/* if nothing on queue, we're done */
@@ -620,7 +620,7 @@ static void pdc20621_dump_hdma(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
 	unsigned int port_no = ap->port_no;
-	struct pdc_host_priv *hpriv = ap->host_set->private_data;
+	struct pdc_host_priv *hpriv = ap->host->private_data;
 	void *dimm_mmio = hpriv->dimm_mmio;
 
 	dimm_mmio += (port_no * PDC_DIMM_WINDOW_STEP);
@@ -638,9 +638,9 @@ static inline void pdc20621_dump_hdma(struct ata_queued_cmd *qc) { }
 static void pdc20621_packet_start(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	struct ata_host_set *host_set = ap->host_set;
+	struct ata_host *host = ap->host;
 	unsigned int port_no = ap->port_no;
-	void __iomem *mmio = host_set->mmio_base;
+	void __iomem *mmio = host->mmio_base;
 	unsigned int rw = (qc->tf.flags & ATA_TFLAG_WRITE);
 	u8 seq = (u8) (port_no + 1);
 	unsigned int port_ofs;
@@ -781,8 +781,8 @@ static inline unsigned int pdc20621_host_intr( struct ata_port *ap,
 
 static void pdc20621_irq_clear(struct ata_port *ap)
 {
-	struct ata_host_set *host_set = ap->host_set;
-	void __iomem *mmio = host_set->mmio_base;
+	struct ata_host *host = ap->host;
+	void __iomem *mmio = host->mmio_base;
 
 	mmio += PDC_CHIP0_OFS;
 
@@ -791,7 +791,7 @@ static void pdc20621_irq_clear(struct ata_port *ap)
 
 static irqreturn_t pdc20621_interrupt (int irq, void *dev_instance, struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	struct ata_port *ap;
 	u32 mask = 0;
 	unsigned int i, tmp, port_no;
@@ -800,12 +800,12 @@ static irqreturn_t pdc20621_interrupt (int irq, void *dev_instance, struct pt_re
 
 	VPRINTK("ENTER\n");
 
-	if (!host_set || !host_set->mmio_base) {
+	if (!host || !host->mmio_base) {
 		VPRINTK("QUICK EXIT\n");
 		return IRQ_NONE;
 	}
 
-	mmio_base = host_set->mmio_base;
+	mmio_base = host->mmio_base;
 
 	/* reading should also clear interrupts */
 	mmio_base += PDC_CHIP0_OFS;
@@ -822,16 +822,16 @@ static irqreturn_t pdc20621_interrupt (int irq, void *dev_instance, struct pt_re
 		return IRQ_NONE;
 	}
 
-        spin_lock(&host_set->lock);
+        spin_lock(&host->lock);
 
         for (i = 1; i < 9; i++) {
 		port_no = i - 1;
 		if (port_no > 3)
 			port_no -= 4;
-		if (port_no >= host_set->n_ports)
+		if (port_no >= host->n_ports)
 			ap = NULL;
 		else
-			ap = host_set->ports[port_no];
+			ap = host->ports[port_no];
 		tmp = mask & (1 << i);
 		VPRINTK("seq %u, port_no %u, ap %p, tmp %x\n", i, port_no, ap, tmp);
 		if (tmp && ap &&
@@ -845,7 +845,7 @@ static irqreturn_t pdc20621_interrupt (int irq, void *dev_instance, struct pt_re
 		}
 	}
 
-        spin_unlock(&host_set->lock);
+        spin_unlock(&host->lock);
 
 	VPRINTK("mask == 0x%x\n", mask);
 
@@ -857,13 +857,13 @@ static irqreturn_t pdc20621_interrupt (int irq, void *dev_instance, struct pt_re
 static void pdc_eng_timeout(struct ata_port *ap)
 {
 	u8 drv_stat;
-	struct ata_host_set *host_set = ap->host_set;
+	struct ata_host *host = ap->host;
 	struct ata_queued_cmd *qc;
 	unsigned long flags;
 
 	DPRINTK("ENTER\n");
 
-	spin_lock_irqsave(&host_set->lock, flags);
+	spin_lock_irqsave(&host->lock, flags);
 
 	qc = ata_qc_from_tag(ap, ap->active_tag);
 
@@ -885,7 +885,7 @@ static void pdc_eng_timeout(struct ata_port *ap)
 		break;
 	}
 
-	spin_unlock_irqrestore(&host_set->lock, flags);
+	spin_unlock_irqrestore(&host->lock, flags);
 	ata_eh_qc_complete(qc);
 	DPRINTK("EXIT\n");
 }
@@ -1429,7 +1429,7 @@ static int pdc_sata_init_one (struct pci_dev *pdev, const struct pci_device_id *
 	hpriv->dimm_mmio = dimm_mmio;
 
 	probe_ent->sht		= pdc_port_info[board_idx].sht;
-	probe_ent->host_flags	= pdc_port_info[board_idx].host_flags;
+	probe_ent->port_flags	= pdc_port_info[board_idx].flags;
 	probe_ent->pio_mask	= pdc_port_info[board_idx].pio_mask;
 	probe_ent->mwdma_mask	= pdc_port_info[board_idx].mwdma_mask;
 	probe_ent->udma_mask	= pdc_port_info[board_idx].udma_mask;
diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c
index 654aae2b25c5..8fc6e800011a 100644
--- a/drivers/ata/sata_uli.c
+++ b/drivers/ata/sata_uli.c
@@ -128,7 +128,7 @@ static const struct ata_port_operations uli_ops = {
 
 static struct ata_port_info uli_port_info = {
 	.sht            = &uli_sht,
-	.host_flags     = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
+	.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 	.pio_mask       = 0x1f,		/* pio0-4 */
 	.udma_mask      = 0x7f,		/* udma0-6 */
 	.port_ops       = &uli_ops,
@@ -143,13 +143,13 @@ MODULE_VERSION(DRV_VERSION);
 
 static unsigned int get_scr_cfg_addr(struct ata_port *ap, unsigned int sc_reg)
 {
-	struct uli_priv *hpriv = ap->host_set->private_data;
+	struct uli_priv *hpriv = ap->host->private_data;
 	return hpriv->scr_cfg_addr[ap->port_no] + (4 * sc_reg);
 }
 
 static u32 uli_scr_cfg_read (struct ata_port *ap, unsigned int sc_reg)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	unsigned int cfg_addr = get_scr_cfg_addr(ap, sc_reg);
 	u32 val;
 
@@ -159,7 +159,7 @@ static u32 uli_scr_cfg_read (struct ata_port *ap, unsigned int sc_reg)
 
 static void uli_scr_cfg_write (struct ata_port *ap, unsigned int scr, u32 val)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host_set->dev);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	unsigned int cfg_addr = get_scr_cfg_addr(ap, scr);
 
 	pci_write_config_dword(pdev, cfg_addr, val);
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index a0699a1728d4..7f087aef99de 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -176,7 +176,7 @@ static const struct ata_port_operations vt6421_sata_ops = {
 
 static struct ata_port_info vt6420_port_info = {
 	.sht		= &svia_sht,
-	.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
+	.flags		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY,
 	.pio_mask	= 0x1f,
 	.mwdma_mask	= 0x07,
 	.udma_mask	= 0x7f,
@@ -346,7 +346,7 @@ static struct ata_probe_ent *vt6421_init_probe_ent(struct pci_dev *pdev)
 	INIT_LIST_HEAD(&probe_ent->node);
 
 	probe_ent->sht		= &svia_sht;
-	probe_ent->host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY;
+	probe_ent->port_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY;
 	probe_ent->port_ops	= &vt6421_sata_ops;
 	probe_ent->n_ports	= N_PORTS;
 	probe_ent->irq		= pdev->irq;
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index 4c69a705a483..d0d92f33de54 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -123,7 +123,7 @@ static void vsc_intr_mask_update(struct ata_port *ap, u8 ctl)
 	void __iomem *mask_addr;
 	u8 mask;
 
-	mask_addr = ap->host_set->mmio_base +
+	mask_addr = ap->host->mmio_base +
 		VSC_SATA_INT_MASK_OFFSET + ap->port_no;
 	mask = readb(mask_addr);
 	if (ctl & ATA_NIEN)
@@ -206,20 +206,20 @@ static void vsc_sata_tf_read(struct ata_port *ap, struct ata_taskfile *tf)
 static irqreturn_t vsc_sata_interrupt (int irq, void *dev_instance,
 				       struct pt_regs *regs)
 {
-	struct ata_host_set *host_set = dev_instance;
+	struct ata_host *host = dev_instance;
 	unsigned int i;
 	unsigned int handled = 0;
 	u32 int_status;
 
-	spin_lock(&host_set->lock);
+	spin_lock(&host->lock);
 
-	int_status = readl(host_set->mmio_base + VSC_SATA_INT_STAT_OFFSET);
+	int_status = readl(host->mmio_base + VSC_SATA_INT_STAT_OFFSET);
 
-	for (i = 0; i < host_set->n_ports; i++) {
+	for (i = 0; i < host->n_ports; i++) {
 		if (int_status & ((u32) 0xFF << (8 * i))) {
 			struct ata_port *ap;
 
-			ap = host_set->ports[i];
+			ap = host->ports[i];
 
 			if (is_vsc_sata_int_err(i, int_status)) {
 				u32 err_status;
@@ -259,7 +259,7 @@ static irqreturn_t vsc_sata_interrupt (int irq, void *dev_instance,
 		}
 	}
 
-	spin_unlock(&host_set->lock);
+	spin_unlock(&host->lock);
 
 	return IRQ_RETVAL(handled);
 }
@@ -395,7 +395,7 @@ static int __devinit vsc_sata_init_one (struct pci_dev *pdev, const struct pci_d
 	pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE, 0x80);
 
 	probe_ent->sht = &vsc_sata_sht;
-	probe_ent->host_flags = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+	probe_ent->port_flags = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 				ATA_FLAG_MMIO;
 	probe_ent->port_ops = &vsc_sata_ops;
 	probe_ent->n_ports = 4;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 806682603ac5..563885cb0995 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -197,7 +197,7 @@ enum {
 	ATA_QCFLAG_EH_SCHEDULED = (1 << 18), /* EH scheduled (obsolete) */
 
 	/* host set flags */
-	ATA_HOST_SIMPLEX	= (1 << 0),	/* Host is simplex, one DMA channel per host_set only */
+	ATA_HOST_SIMPLEX	= (1 << 0),	/* Host is simplex, one DMA channel per host only */
 	
 	/* various lengths of time */
 	ATA_TMOUT_BOOT		= 30 * HZ,	/* heuristic */
@@ -357,13 +357,13 @@ struct ata_probe_ent {
 	unsigned long		irq;
 	unsigned long		irq2;
 	unsigned int		irq_flags;
-	unsigned long		host_flags;
-	unsigned long		host_set_flags;
+	unsigned long		port_flags;
+	unsigned long		_host_flags;
 	void __iomem		*mmio_base;
 	void			*private_data;
 };
 
-struct ata_host_set {
+struct ata_host {
 	spinlock_t		lock;
 	struct device 		*dev;
 	unsigned long		irq;
@@ -420,7 +420,7 @@ struct ata_queued_cmd {
 	void			*private_data;
 };
 
-struct ata_host_stats {
+struct ata_port_stats {
 	unsigned long		unhandled_irq;
 	unsigned long		idle_irq;
 	unsigned long		rw_reqbuf;
@@ -498,7 +498,7 @@ struct ata_eh_context {
 };
 
 struct ata_port {
-	struct Scsi_Host	*host;	/* our co-allocated scsi host */
+	struct Scsi_Host	*scsi_host; /* our co-allocated scsi host */
 	const struct ata_port_operations *ops;
 	spinlock_t		*lock;
 	unsigned long		flags;	/* ATA_FLAG_xxx */
@@ -523,7 +523,7 @@ struct ata_port {
 	unsigned int		hw_sata_spd_limit;
 	unsigned int		sata_spd_limit;	/* SATA PHY speed limit */
 
-	/* record runtime error info, protected by host_set lock */
+	/* record runtime error info, protected by host lock */
 	struct ata_eh_info	eh_info;
 	/* EH context owned by EH */
 	struct ata_eh_context	eh_context;
@@ -537,8 +537,8 @@ struct ata_port {
 	unsigned int		active_tag;
 	u32			sactive;
 
-	struct ata_host_stats	stats;
-	struct ata_host_set	*host_set;
+	struct ata_port_stats	stats;
+	struct ata_host		*host;
 	struct device 		*dev;
 
 	struct work_struct	port_task;
@@ -614,7 +614,7 @@ struct ata_port_operations {
 	int (*port_start) (struct ata_port *ap);
 	void (*port_stop) (struct ata_port *ap);
 
-	void (*host_stop) (struct ata_host_set *host_set);
+	void (*host_stop) (struct ata_host *host);
 
 	void (*bmdma_stop) (struct ata_queued_cmd *qc);
 	u8   (*bmdma_status) (struct ata_port *ap);
@@ -622,7 +622,7 @@ struct ata_port_operations {
 
 struct ata_port_info {
 	struct scsi_host_template	*sht;
-	unsigned long		host_flags;
+	unsigned long		flags;
 	unsigned long		pio_mask;
 	unsigned long		mwdma_mask;
 	unsigned long		udma_mask;
@@ -690,15 +690,15 @@ extern int ata_pci_clear_simplex(struct pci_dev *pdev);
 #endif /* CONFIG_PCI */
 extern int ata_device_add(const struct ata_probe_ent *ent);
 extern void ata_port_detach(struct ata_port *ap);
-extern void ata_host_set_init(struct ata_host_set *, struct device *,
-			      unsigned long, const struct ata_port_operations *);
-extern void ata_host_set_remove(struct ata_host_set *host_set);
+extern void ata_host_init(struct ata_host *, struct device *,
+			  unsigned long, const struct ata_port_operations *);
+extern void ata_host_remove(struct ata_host *host);
 extern int ata_scsi_detect(struct scsi_host_template *sht);
 extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg);
 extern int ata_scsi_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *));
 extern int ata_scsi_release(struct Scsi_Host *host);
 extern void ata_sas_port_destroy(struct ata_port *);
-extern struct ata_port *ata_sas_port_alloc(struct ata_host_set *,
+extern struct ata_port *ata_sas_port_alloc(struct ata_host *,
 					   struct ata_port_info *, struct Scsi_Host *);
 extern int ata_sas_port_init(struct ata_port *);
 extern int ata_sas_port_start(struct ata_port *ap);
@@ -715,9 +715,8 @@ extern int ata_port_online(struct ata_port *ap);
 extern int ata_port_offline(struct ata_port *ap);
 extern int ata_scsi_device_resume(struct scsi_device *);
 extern int ata_scsi_device_suspend(struct scsi_device *, pm_message_t mesg);
-extern int ata_host_set_suspend(struct ata_host_set *host_set,
-				pm_message_t mesg);
-extern void ata_host_set_resume(struct ata_host_set *host_set);
+extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg);
+extern void ata_host_resume(struct ata_host *host);
 extern int ata_ratelimit(void);
 extern unsigned int ata_busy_sleep(struct ata_port *ap,
 				   unsigned long timeout_pat,
@@ -742,7 +741,7 @@ extern u8 ata_altstatus(struct ata_port *ap);
 extern void ata_exec_command(struct ata_port *ap, const struct ata_taskfile *tf);
 extern int ata_port_start (struct ata_port *ap);
 extern void ata_port_stop (struct ata_port *ap);
-extern void ata_host_stop (struct ata_host_set *host_set);
+extern void ata_host_stop (struct ata_host *host);
 extern irqreturn_t ata_interrupt (int irq, void *dev_instance, struct pt_regs *regs);
 extern void ata_mmio_data_xfer(struct ata_device *adev, unsigned char *buf,
 			       unsigned int buflen, int write_data);
@@ -828,7 +827,7 @@ struct pci_bits {
 	unsigned long		val;
 };
 
-extern void ata_pci_host_stop (struct ata_host_set *host_set);
+extern void ata_pci_host_stop (struct ata_host *host);
 extern struct ata_probe_ent *
 ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int portmask);
 extern int pci_test_config_bits(struct pci_dev *pdev, const struct pci_bits *bits);
-- 
cgit v1.2.3


From f39b7a55a84e34e3074b168e30dc73b66e85261d Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Fri, 11 Aug 2006 00:07:08 -0500
Subject: [POWERPC] Cleanup CPU inits

Cleanup CPU inits a bit more, Geoff Levand already did some earlier.

* Move CPU state save to cpu_setup, since cpu_setup is only ever done
  on cpu 0 on 64-bit and save is never done more than once.
* Rename __restore_cpu_setup to __restore_cpu_ppc970 and add
  function pointers to the cputable to use instead. Powermac always
  has 970 so no need to check there.
* Rename __970_cpu_preinit to __cpu_preinit_ppc970 and check PVR before
  calling it instead of in it, it's too early to use cputable.
* Rename pSeries_secondary_smp_init to generic_secondary_smp_init since
  everyone but powermac and iSeries use it.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/asm-offsets.c      |  1 +
 arch/powerpc/kernel/cpu_setup_ppc970.S | 99 +++++++++-------------------------
 arch/powerpc/kernel/cputable.c         |  6 +++
 arch/powerpc/kernel/head_64.S          | 52 +++++++++++-------
 arch/powerpc/platforms/cell/smp.c      |  4 +-
 arch/powerpc/platforms/pseries/smp.c   |  4 +-
 include/asm-powerpc/cputable.h         |  3 ++
 7 files changed, 71 insertions(+), 98 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index a2f95e467a75..c53acd2a6dfc 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -246,6 +246,7 @@ int main(void)
 	DEFINE(CPU_SPEC_PVR_VALUE, offsetof(struct cpu_spec, pvr_value));
 	DEFINE(CPU_SPEC_FEATURES, offsetof(struct cpu_spec, cpu_features));
 	DEFINE(CPU_SPEC_SETUP, offsetof(struct cpu_spec, cpu_setup));
+	DEFINE(CPU_SPEC_RESTORE, offsetof(struct cpu_spec, cpu_restore));
 
 #ifndef CONFIG_PPC64
 	DEFINE(pbe_address, offsetof(struct pbe, address));
diff --git a/arch/powerpc/kernel/cpu_setup_ppc970.S b/arch/powerpc/kernel/cpu_setup_ppc970.S
index f69af2c5d7b3..f619932794e8 100644
--- a/arch/powerpc/kernel/cpu_setup_ppc970.S
+++ b/arch/powerpc/kernel/cpu_setup_ppc970.S
@@ -16,27 +16,12 @@
 #include <asm/asm-offsets.h>
 #include <asm/cache.h>
 
-_GLOBAL(__970_cpu_preinit)
-	/*
-	 * Do nothing if not running in HV mode
-	 */
+_GLOBAL(__cpu_preinit_ppc970)
+	/* Do nothing if not running in HV mode */
 	mfmsr	r0
 	rldicl.	r0,r0,4,63
 	beqlr
 
-	/*
-	 * Deal only with PPC970 and PPC970FX.
-	 */
-	mfspr	r0,SPRN_PVR
-	srwi	r0,r0,16
-	cmpwi	r0,0x39
-	beq	1f
-	cmpwi	r0,0x3c
-	beq	1f
-	cmpwi	r0,0x44
-	bnelr
-1:
-
 	/* Make sure HID4:rm_ci is off before MMU is turned off, that large
 	 * pages are enabled with HID4:61 and clear HID5:DCBZ_size and
 	 * HID5:DCBZ32_ill
@@ -72,21 +57,6 @@ _GLOBAL(__970_cpu_preinit)
 	isync
 	blr
 
-_GLOBAL(__setup_cpu_ppc970)
-	mfspr	r0,SPRN_HID0
-	li	r11,5			/* clear DOZE and SLEEP */
-	rldimi	r0,r11,52,8		/* set NAP and DPM */
-	mtspr	SPRN_HID0,r0
-	mfspr	r0,SPRN_HID0
-	mfspr	r0,SPRN_HID0
-	mfspr	r0,SPRN_HID0
-	mfspr	r0,SPRN_HID0
-	mfspr	r0,SPRN_HID0
-	mfspr	r0,SPRN_HID0
-	sync
-	isync
-	blr
-
 /* Definitions for the table use to save CPU states */
 #define CS_HID0		0
 #define CS_HID1		8
@@ -101,33 +71,28 @@ cpu_state_storage:
 	.balign	L1_CACHE_BYTES,0
 	.text
 
-/* Called in normal context to backup CPU 0 state. This
- * does not include cache settings. This function is also
- * called for machine sleep. This does not include the MMU
- * setup, BATs, etc... but rather the "special" registers
- * like HID0, HID1, HID4, etc...
- */
-_GLOBAL(__save_cpu_setup)
-	/* Some CR fields are volatile, we back it up all */
-	mfcr	r7
-
-	/* Get storage ptr */
-	LOAD_REG_IMMEDIATE(r5,cpu_state_storage)
 
-	/* We only deal with 970 for now */
-	mfspr	r0,SPRN_PVR
-	srwi	r0,r0,16
-	cmpwi	r0,0x39
-	beq	1f
-	cmpwi	r0,0x3c
-	beq	1f
-	cmpwi	r0,0x44
-	bne	2f
-
-1:	/* skip if not running in HV mode */
+_GLOBAL(__setup_cpu_ppc970)
+	/* Do nothing if not running in HV mode */
 	mfmsr	r0
 	rldicl.	r0,r0,4,63
-	beq	2f
+	beqlr
+
+	mfspr	r0,SPRN_HID0
+	li	r11,5			/* clear DOZE and SLEEP */
+	rldimi	r0,r11,52,8		/* set NAP and DPM */
+	mtspr	SPRN_HID0,r0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	sync
+	isync
+
+	/* Save away cpu state */
+	LOAD_REG_IMMEDIATE(r5,cpu_state_storage)
 
 	/* Save HID0,1,4 and 5 */
 	mfspr	r3,SPRN_HID0
@@ -139,35 +104,19 @@ _GLOBAL(__save_cpu_setup)
 	mfspr	r3,SPRN_HID5
 	std	r3,CS_HID5(r5)
 
-2:
-	mtcr	r7
 	blr
 
 /* Called with no MMU context (typically MSR:IR/DR off) to
  * restore CPU state as backed up by the previous
  * function. This does not include cache setting
  */
-_GLOBAL(__restore_cpu_setup)
-	/* Get storage ptr (FIXME when using anton reloc as we
-	 * are running with translation disabled here
-	 */
-	LOAD_REG_IMMEDIATE(r5,cpu_state_storage)
-
-	/* We only deal with 970 for now */
-	mfspr	r0,SPRN_PVR
-	srwi	r0,r0,16
-	cmpwi	r0,0x39
-	beq	1f
-	cmpwi	r0,0x3c
-	beq	1f
-	cmpwi	r0,0x44
-	bnelr
-
-1:	/* skip if not running in HV mode */
+_GLOBAL(__restore_cpu_ppc970)
+	/* Do nothing if not running in HV mode */
 	mfmsr	r0
 	rldicl.	r0,r0,4,63
 	beqlr
 
+	LOAD_REG_IMMEDIATE(r5,cpu_state_storage)
 	/* Before accessing memory, we make sure rm_ci is clear */
 	li	r0,0
 	mfspr	r3,SPRN_HID4
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 272e43622fd6..306da4cd37a0 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -39,7 +39,10 @@ extern void __setup_cpu_7400(unsigned long offset, struct cpu_spec* spec);
 extern void __setup_cpu_7410(unsigned long offset, struct cpu_spec* spec);
 extern void __setup_cpu_745x(unsigned long offset, struct cpu_spec* spec);
 #endif /* CONFIG_PPC32 */
+#ifdef CONFIG_PPC64
 extern void __setup_cpu_ppc970(unsigned long offset, struct cpu_spec* spec);
+extern void __restore_cpu_ppc970(void);
+#endif /* CONFIG_PPC64 */
 
 /* This table only contains "desktop" CPUs, it need to be filled with embedded
  * ones as well...
@@ -184,6 +187,7 @@ struct cpu_spec	cpu_specs[] = {
 		.dcache_bsize		= 128,
 		.num_pmcs		= 8,
 		.cpu_setup		= __setup_cpu_ppc970,
+		.cpu_restore		= __restore_cpu_ppc970,
 		.oprofile_cpu_type	= "ppc64/970",
 		.oprofile_type		= PPC_OPROFILE_POWER4,
 		.platform		= "ppc970",
@@ -199,6 +203,7 @@ struct cpu_spec	cpu_specs[] = {
 		.dcache_bsize		= 128,
 		.num_pmcs		= 8,
 		.cpu_setup		= __setup_cpu_ppc970,
+		.cpu_restore		= __restore_cpu_ppc970,
 		.oprofile_cpu_type	= "ppc64/970",
 		.oprofile_type		= PPC_OPROFILE_POWER4,
 		.platform		= "ppc970",
@@ -214,6 +219,7 @@ struct cpu_spec	cpu_specs[] = {
 		.dcache_bsize		= 128,
 		.num_pmcs		= 8,
 		.cpu_setup		= __setup_cpu_ppc970,
+		.cpu_restore		= __restore_cpu_ppc970,
 		.oprofile_cpu_type	= "ppc64/970",
 		.oprofile_type		= PPC_OPROFILE_POWER4,
 		.platform		= "ppc970",
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 6ff3cf506088..e9963d9f335a 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -132,7 +132,7 @@ _GLOBAL(__secondary_hold)
 	bne	100b
 
 #if defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
-	LOAD_REG_IMMEDIATE(r4, .pSeries_secondary_smp_init)
+	LOAD_REG_IMMEDIATE(r4, .generic_secondary_smp_init)
 	mtctr	r4
 	mr	r3,r24
 	bctr
@@ -1484,19 +1484,17 @@ fwnmi_data_area:
         . = 0x8000
 
 /*
- * On pSeries, secondary processors spin in the following code.
+ * On pSeries and most other platforms, secondary processors spin
+ * in the following code.
  * At entry, r3 = this processor's number (physical cpu id)
  */
-_GLOBAL(pSeries_secondary_smp_init)
+_GLOBAL(generic_secondary_smp_init)
 	mr	r24,r3
 	
 	/* turn on 64-bit mode */
 	bl	.enable_64b_mode
 	isync
 
-	/* Copy some CPU settings from CPU 0 */
-	bl	.__restore_cpu_setup
-
 	/* Set up a paca value for this processor. Since we have the
 	 * physical cpu id in r24, we need to search the pacas to find
 	 * which logical id maps to our physical one.
@@ -1522,15 +1520,28 @@ _GLOBAL(pSeries_secondary_smp_init)
 					/* start.			 */
 	sync
 
-	/* Create a temp kernel stack for use before relocation is on.	*/
+#ifndef CONFIG_SMP
+	b	3b			/* Never go on non-SMP		 */
+#else
+	cmpwi	0,r23,0
+	beq	3b			/* Loop until told to go	 */
+
+	/* See if we need to call a cpu state restore handler */
+	LOAD_REG_IMMEDIATE(r23, cur_cpu_spec)
+	ld	r23,0(r23)
+	ld	r23,CPU_SPEC_RESTORE(r23)
+	cmpdi	0,r23,0
+	beq	4f
+	ld	r23,0(r23)
+	mtctr	r23
+	bctrl
+
+4:	/* Create a temp kernel stack for use before relocation is on.	*/
 	ld	r1,PACAEMERGSP(r13)
 	subi	r1,r1,STACK_FRAME_OVERHEAD
 
-	cmpwi	0,r23,0
-#ifdef CONFIG_SMP
-	bne	.__secondary_start
+	b	.__secondary_start
 #endif
-	b 	3b			/* Loop until told to go	 */
 
 #ifdef CONFIG_PPC_ISERIES
 _STATIC(__start_initialization_iSeries)
@@ -1611,7 +1622,16 @@ _GLOBAL(__start_initialization_multiplatform)
 	bl	.enable_64b_mode
 
 	/* Setup some critical 970 SPRs before switching MMU off */
-	bl	.__970_cpu_preinit
+	mfspr	r0,SPRN_PVR
+	srwi	r0,r0,16
+	cmpwi	r0,0x39		/* 970 */
+	beq	1f
+	cmpwi	r0,0x3c		/* 970FX */
+	beq	1f
+	cmpwi	r0,0x44		/* 970MP */
+	bne	2f
+1:	bl	.__cpu_preinit_ppc970
+2:
 
 	/* Switch off MMU if not already */
 	LOAD_REG_IMMEDIATE(r4, .__after_prom_start - KERNELBASE)
@@ -1782,7 +1802,7 @@ _GLOBAL(pmac_secondary_start)
 	isync
 
 	/* Copy some CPU settings from CPU 0 */
-	bl	.__restore_cpu_setup
+	bl	.__restore_cpu_ppc970
 
 	/* pSeries do that early though I don't think we really need it */
 	mfmsr	r3
@@ -1932,12 +1952,6 @@ _STATIC(start_here_multiplatform)
 	mr	r5,r26
 	bl	.identify_cpu
 
-	/* Save some low level config HIDs of CPU0 to be copied to
-	 * other CPUs later on, or used for suspend/resume
-	 */
-	bl	.__save_cpu_setup
-	sync
-
 	/* Do very early kernel initializations, including initial hash table,
 	 * stab and slb setup before we turn on relocation.	*/
 
diff --git a/arch/powerpc/platforms/cell/smp.c b/arch/powerpc/platforms/cell/smp.c
index 46aef0640742..1c0acbad7425 100644
--- a/arch/powerpc/platforms/cell/smp.c
+++ b/arch/powerpc/platforms/cell/smp.c
@@ -57,7 +57,7 @@
  */
 static cpumask_t of_spin_map;
 
-extern void pSeries_secondary_smp_init(unsigned long);
+extern void generic_secondary_smp_init(unsigned long);
 
 /**
  * smp_startup_cpu() - start the given cpu
@@ -74,7 +74,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu)
 {
 	int status;
 	unsigned long start_here = __pa((u32)*((unsigned long *)
-					       pSeries_secondary_smp_init));
+					       generic_secondary_smp_init));
 	unsigned int pcpu;
 	int start_cpu;
 
diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
index f39dad8b99e0..c6624b8a0e77 100644
--- a/arch/powerpc/platforms/pseries/smp.c
+++ b/arch/powerpc/platforms/pseries/smp.c
@@ -62,7 +62,7 @@
  */
 static cpumask_t of_spin_map;
 
-extern void pSeries_secondary_smp_init(unsigned long);
+extern void generic_secondary_smp_init(unsigned long);
 
 #ifdef CONFIG_HOTPLUG_CPU
 
@@ -270,7 +270,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu)
 {
 	int status;
 	unsigned long start_here = __pa((u32)*((unsigned long *)
-					       pSeries_secondary_smp_init));
+					       generic_secondary_smp_init));
 	unsigned int pcpu;
 	int start_cpu;
 
diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 1ba3c9983614..748bc1805da9 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -36,6 +36,7 @@
 struct cpu_spec;
 
 typedef	void (*cpu_setup_t)(unsigned long offset, struct cpu_spec* spec);
+typedef	void (*cpu_restore_t)(void);
 
 enum powerpc_oprofile_type {
 	PPC_OPROFILE_INVALID = 0,
@@ -65,6 +66,8 @@ struct cpu_spec {
 	 * BHT, SPD, etc... from head.S before branching to identify_machine
 	 */
 	cpu_setup_t	cpu_setup;
+	/* Used to restore cpu setup on secondary processors and at resume */
+	cpu_restore_t	cpu_restore;
 
 	/* Used by oprofile userspace to select the right counters */
 	char		*oprofile_cpu_type;
-- 
cgit v1.2.3


From 869d7f381e8c32de85ddfa9621125fb10a885f87 Mon Sep 17 00:00:00 2001
From: Jon Loeliger <jdl@freescale.com>
Date: Tue, 15 Aug 2006 16:19:02 -0500
Subject: [POWERPC] Allow MPC8641 HPCN to build with CONFIG_PCI disabled too.

Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/86xx/mpc86xx_hpcn.c | 9 +++++++--
 arch/powerpc/platforms/86xx/pci.c          | 3 ++-
 include/asm-powerpc/mpc86xx.h              | 2 --
 3 files changed, 9 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c
index 4a33d95e7ad7..496cc7c3a54c 100644
--- a/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c
+++ b/arch/powerpc/platforms/86xx/mpc86xx_hpcn.c
@@ -52,6 +52,7 @@ unsigned long pci_dram_offset = 0;
 #endif
 
 
+#ifdef CONFIG_PCI
 static void mpc86xx_8259_cascade(unsigned int irq, struct irq_desc *desc,
 				 struct pt_regs *regs)
 {
@@ -60,14 +61,18 @@ static void mpc86xx_8259_cascade(unsigned int irq, struct irq_desc *desc,
 		generic_handle_irq(cascade_irq, regs);
 	desc->chip->eoi(irq);
 }
+#endif	/* CONFIG_PCI */
 
 void __init
 mpc86xx_hpcn_init_irq(void)
 {
 	struct mpic *mpic1;
-	struct device_node *np, *cascade_node = NULL;
-	int cascade_irq;
+	struct device_node *np;
 	phys_addr_t openpic_paddr;
+#ifdef CONFIG_PCI
+	struct device_node *cascade_node = NULL;
+	int cascade_irq;
+#endif
 
 	np = of_find_node_by_type(NULL, "open-pic");
 	if (np == NULL)
diff --git a/arch/powerpc/platforms/86xx/pci.c b/arch/powerpc/platforms/86xx/pci.c
index d7050c1108ff..481e18ed5be9 100644
--- a/arch/powerpc/platforms/86xx/pci.c
+++ b/arch/powerpc/platforms/86xx/pci.c
@@ -188,7 +188,8 @@ int __init add_bridge(struct device_node *dev)
 
 	printk(KERN_INFO "Found MPC86xx PCIE host bridge at 0x%08lx. "
 	       "Firmware bus number: %d->%d\n",
-		rsrc.start, hose->first_busno, hose->last_busno);
+	       (unsigned long) rsrc.start,
+	       hose->first_busno, hose->last_busno);
 
 	DBG(" ->Hose at 0x%p, cfg_addr=0x%p,cfg_data=0x%p\n",
 		hose, hose->cfg_addr, hose->cfg_data);
diff --git a/include/asm-powerpc/mpc86xx.h b/include/asm-powerpc/mpc86xx.h
index f260382739fa..2d6ad859df7f 100644
--- a/include/asm-powerpc/mpc86xx.h
+++ b/include/asm-powerpc/mpc86xx.h
@@ -23,8 +23,6 @@
 #define _ISA_MEM_BASE   isa_mem_base
 #ifdef CONFIG_PCI
 #define PCI_DRAM_OFFSET pci_dram_offset
-#else
-#define PCI_DRAM_OFFSET 0
 #endif
 
 #define CPU0_BOOT_RELEASE 0x01000000
-- 
cgit v1.2.3


From 6f3d5d3cc4b1447578ae8484166bbc34a64150c5 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Wed, 16 Aug 2006 22:04:14 +1000
Subject: [POWERPC] Add a helper for calculating RTAS "config_addr" parameters

Several RTAS calls take a "config_addr" parameter, which is a particular
way of specifying a PCI busno, devfn and register number into a 32-bit word.
Currently these are open-coded, and I'll be adding another soon, replace
them with a helper that encapsulates the logic. Be more strict about masking
the busno too, just in case.

Booted on P5 LPAR.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/rtas_pci.c |  6 ++----
 include/asm-powerpc/rtas.h     | 16 ++++++++++++++++
 2 files changed, 18 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/rtas_pci.c b/arch/powerpc/kernel/rtas_pci.c
index 5a798ac6aecf..b4a0de79c060 100644
--- a/arch/powerpc/kernel/rtas_pci.c
+++ b/arch/powerpc/kernel/rtas_pci.c
@@ -81,8 +81,7 @@ int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val)
 	if (!config_access_valid(pdn, where))
 		return PCIBIOS_BAD_REGISTER_NUMBER;
 
-	addr = ((where & 0xf00) << 20) | (pdn->busno << 16) |
-		(pdn->devfn << 8) | (where & 0xff);
+	addr = rtas_config_addr(pdn->busno, pdn->devfn, where);
 	buid = pdn->phb->buid;
 	if (buid) {
 		ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
@@ -134,8 +133,7 @@ int rtas_write_config(struct pci_dn *pdn, int where, int size, u32 val)
 	if (!config_access_valid(pdn, where))
 		return PCIBIOS_BAD_REGISTER_NUMBER;
 
-	addr = ((where & 0xf00) << 20) | (pdn->busno << 16) |
-		(pdn->devfn << 8) | (where & 0xff);
+	addr = rtas_config_addr(pdn->busno, pdn->devfn, where);
 	buid = pdn->phb->buid;
 	if (buid) {
 		ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr,
diff --git a/include/asm-powerpc/rtas.h b/include/asm-powerpc/rtas.h
index 82a27e9a041f..d34f9e1f242c 100644
--- a/include/asm-powerpc/rtas.h
+++ b/include/asm-powerpc/rtas.h
@@ -230,5 +230,21 @@ extern unsigned long rtas_rmo_buf;
 
 #define GLOBAL_INTERRUPT_QUEUE 9005
 
+/**
+ * rtas_config_addr - Format a busno, devfn and reg for RTAS.
+ * @busno: The bus number.
+ * @devfn: The device and function number as encoded by PCI_DEVFN().
+ * @reg: The register number.
+ *
+ * This function encodes the given busno, devfn and register number as
+ * required for RTAS calls that take a "config_addr" parameter.
+ * See PAPR requirement 7.3.4-1 for more info.
+ */
+static inline u32 rtas_config_addr(int busno, int devfn, int reg)
+{
+	return ((reg & 0xf00) << 20) | ((busno & 0xff) << 16) |
+			(devfn << 8) | (reg & 0xff);
+}
+
 #endif /* __KERNEL__ */
 #endif /* _POWERPC_RTAS_H */
-- 
cgit v1.2.3


From f4ad7b5807385ad1fed0347d966e51a797cd1013 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Fri, 25 Aug 2006 13:48:18 -0500
Subject: [SCSI] scsi_transport_sas: remove local_attached flag

This flag denotes local attachment of the phy.  There are two problems
with it:

1) It's actually redundant ... you can get the same information simply
by seeing whether a host is the phys parent
2) we condition a lot of phy parameters on it on the false assumption
that we can only control local phys.  I'm wiring up phy resets in the
aic94xx now, and it will be able to reset non-local phys as well.

I fixed 2) by moving the local check into the reset and stats function
of the mptsas, since that seems to be the only HBA that can't
(currently) control non-local phys.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptsas.c   | 11 ++++++++---
 drivers/scsi/scsi_transport_sas.c | 10 ++--------
 include/scsi/scsi_transport_sas.h |  5 ++---
 3 files changed, 12 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c
index dfdd1e445768..b752a479f6db 100644
--- a/drivers/message/fusion/mptsas.c
+++ b/drivers/message/fusion/mptsas.c
@@ -852,6 +852,10 @@ static int mptsas_get_linkerrors(struct sas_phy *phy)
 	dma_addr_t dma_handle;
 	int error;
 
+	/* FIXME: only have link errors on local phys */
+	if (!scsi_is_sas_phy_local(phy))
+		return -EINVAL;
+
 	hdr.PageVersion = MPI_SASPHY1_PAGEVERSION;
 	hdr.ExtPageLength = 0;
 	hdr.PageNumber = 1 /* page number 1*/;
@@ -924,6 +928,10 @@ static int mptsas_phy_reset(struct sas_phy *phy, int hard_reset)
 	unsigned long timeleft;
 	int error = -ERESTARTSYS;
 
+	/* FIXME: fusion doesn't allow non-local phy reset */
+	if (!scsi_is_sas_phy_local(phy))
+		return -EINVAL;
+
 	/* not implemented for expanders */
 	if (phy->identify.target_port_protocols & SAS_PROTOCOL_SMP)
 		return -ENXIO;
@@ -1570,9 +1578,6 @@ static int mptsas_probe_one_phy(struct device *dev,
 
 	if (!phy_info->phy) {
 
-		if (local)
-			phy->local_attached = 1;
-
 		error = sas_phy_add(phy);
 		if (error) {
 			sas_phy_free(phy);
diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
index 5a625c3fddae..d518c1207fb4 100644
--- a/drivers/scsi/scsi_transport_sas.c
+++ b/drivers/scsi/scsi_transport_sas.c
@@ -266,9 +266,6 @@ show_sas_phy_##field(struct class_device *cdev, char *buf)		\
 	struct sas_internal *i = to_sas_internal(shost->transportt);	\
 	int error;							\
 									\
-	if (!phy->local_attached)					\
-		return -EINVAL;						\
-									\
 	error = i->f->get_linkerrors ? i->f->get_linkerrors(phy) : 0;	\
 	if (error)							\
 		return error;						\
@@ -299,9 +296,6 @@ static ssize_t do_sas_phy_reset(struct class_device *cdev,
 	struct sas_internal *i = to_sas_internal(shost->transportt);
 	int error;
 
-	if (!phy->local_attached)
-		return -EINVAL;
-
 	error = i->f->phy_reset(phy, hard_reset);
 	if (error)
 		return error;
@@ -849,7 +843,7 @@ show_sas_rphy_enclosure_identifier(struct class_device *cdev, char *buf)
 	 * Only devices behind an expander are supported, because the
 	 * enclosure identifier is a SMP feature.
 	 */
-	if (phy->local_attached)
+	if (scsi_is_sas_phy_local(phy))
 		return -EINVAL;
 
 	error = i->f->get_enclosure_identifier(rphy, &identifier);
@@ -870,7 +864,7 @@ show_sas_rphy_bay_identifier(struct class_device *cdev, char *buf)
 	struct sas_internal *i = to_sas_internal(shost->transportt);
 	int val;
 
-	if (phy->local_attached)
+	if (scsi_is_sas_phy_local(phy))
 		return -EINVAL;
 
 	val = i->f->get_bay_identifier(rphy);
diff --git a/include/scsi/scsi_transport_sas.h b/include/scsi/scsi_transport_sas.h
index 6cc2314098cf..eeb2200de855 100644
--- a/include/scsi/scsi_transport_sas.h
+++ b/include/scsi/scsi_transport_sas.h
@@ -57,9 +57,6 @@ struct sas_phy {
 	enum sas_linkrate	maximum_linkrate_hw;
 	enum sas_linkrate	maximum_linkrate;
 
-	/* internal state */
-	unsigned int		local_attached : 1;
-
 	/* link error statistics */
 	u32			invalid_dword_count;
 	u32			running_disparity_error_count;
@@ -196,4 +193,6 @@ scsi_is_sas_expander_device(struct device *dev)
 		rphy->identify.device_type == SAS_EDGE_EXPANDER_DEVICE;
 }
 
+#define scsi_is_sas_phy_local(phy)	scsi_is_host_device((phy)->dev.parent)
+
 #endif /* SCSI_TRANSPORT_SAS_H */
-- 
cgit v1.2.3


From 2908d778ab3e244900c310974e1fc1c69066e450 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Tue, 29 Aug 2006 09:22:51 -0500
Subject: [SCSI] aic94xx: new driver

This is the end point of the separate aic94xx driver based on the
original driver and transport class from Luben Tuikov
<ltuikov@yahoo.com>

The log of the separate development is:

Alexis Bruemmer:
  o aic94xx: fix hotplug/unplug for expanderless systems
  o aic94xx: disable split completion timer/setting by default
  o aic94xx: wide port off expander support
  o aic94xx: remove various inline functions
  o aic94xx: use bitops
  o aic94xx: remove queue comment
  o aic94xx: remove sas_common.c
  o aic94xx: sas remove depot's
  o aic94xx: use available list_for_each_entry_safe_reverse()
  o aic94xx: sas header file merge

James Bottomley:
  o aic94xx: fix TF_TMF_NO_CTX processing
  o aic94xx: convert to request_firmware interface
  o aic94xx: fix hotplug/unplug
  o aic94xx: add link error counts to the expander phys
  o aic94xx: add transport class phy reset capability
  o aic94xx: remove local_attached flag
  o Remove README
  o Fixup Makefile variable for libsas rename
  o Rename sas->libsas
  o aic94xx: correct return code for sas_discover_event
  o aic94xx: use parent backlink port
  o aic94xx: remove channel abstraction
  o aic94xx: fix routing algorithms
  o aic94xx: add backlink port
  o aic94xx: fix cascaded expander properties
  o aic94xx: fix sleep under lock
  o aic94xx: fix panic on module removal in complex topology
  o aic94xx: make use of the new sas_port
  o rename sas_port to asd_sas_port
  o Fix for eh_strategy_handler move
  o aic94xx: move entirely over to correct transport class formulation
  o remove last vestages of sas_rphy_alloc()
  o update for eh_timed_out move
  o Preliminary expander support for aic94xx
  o sas: remove event thread
  o minor warning cleanups
  o remove last vestiges of id mapping arrays
  o Further updates
  o Convert aic94xx over entirely to the transport class end device and
  o update aic94xx/sas to use the new sas transport class end device
  o [PATCH] aic94xx: attaching to the sas transport class
  o Add missing completion removal from prior patch
  o [PATCH] aic94xx: attaching to the sas transport class
  o Build fixes from akpm

Jeff Garzik:
  o [scsi aic94xx] Remove ->owner from PCI info table

Luben Tuikov:
  o initial aic94xx driver

Mike Anderson:
  o aic94xx: fix panic on module insertion
  o aic94xx: stub out SATA_DEV case
  o aic94xx: compile warning cleanups
  o aic94xx: sas_alloc_task
  o aic94xx: ref count update
  o aic94xx nexus loss time value
  o [PATCH] aic94xx: driver assertion in non-x86 BIOS env

Randy Dunlap:
  o libsas: externs not needed

Robert Tarte:
  o aic94xx: sequence patch - fixes SATA support

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 Documentation/scsi/libsas.txt          |  484 +++++++
 drivers/scsi/Kconfig                   |    5 +-
 drivers/scsi/Makefile                  |    2 +
 drivers/scsi/aic94xx/Kconfig           |   41 +
 drivers/scsi/aic94xx/Makefile          |   39 +
 drivers/scsi/aic94xx/aic94xx.h         |  114 ++
 drivers/scsi/aic94xx/aic94xx_dev.c     |  353 +++++
 drivers/scsi/aic94xx/aic94xx_dump.c    |  959 +++++++++++++
 drivers/scsi/aic94xx/aic94xx_dump.h    |   52 +
 drivers/scsi/aic94xx/aic94xx_hwi.c     | 1376 ++++++++++++++++++
 drivers/scsi/aic94xx/aic94xx_hwi.h     |  397 ++++++
 drivers/scsi/aic94xx/aic94xx_init.c    |  860 ++++++++++++
 drivers/scsi/aic94xx/aic94xx_reg.c     |  332 +++++
 drivers/scsi/aic94xx/aic94xx_reg.h     |  302 ++++
 drivers/scsi/aic94xx/aic94xx_reg_def.h | 2398 ++++++++++++++++++++++++++++++++
 drivers/scsi/aic94xx/aic94xx_sas.h     |  785 +++++++++++
 drivers/scsi/aic94xx/aic94xx_scb.c     |  732 ++++++++++
 drivers/scsi/aic94xx/aic94xx_sds.c     | 1136 +++++++++++++++
 drivers/scsi/aic94xx/aic94xx_seq.c     | 1401 +++++++++++++++++++
 drivers/scsi/aic94xx/aic94xx_seq.h     |   70 +
 drivers/scsi/aic94xx/aic94xx_task.c    |  642 +++++++++
 drivers/scsi/aic94xx/aic94xx_tmf.c     |  636 +++++++++
 drivers/scsi/libsas/Kconfig            |   39 +
 drivers/scsi/libsas/Makefile           |   36 +
 drivers/scsi/libsas/sas_discover.c     |  749 ++++++++++
 drivers/scsi/libsas/sas_dump.c         |   76 +
 drivers/scsi/libsas/sas_dump.h         |   42 +
 drivers/scsi/libsas/sas_event.c        |   75 +
 drivers/scsi/libsas/sas_expander.c     | 1862 +++++++++++++++++++++++++
 drivers/scsi/libsas/sas_init.c         |  227 +++
 drivers/scsi/libsas/sas_internal.h     |  146 ++
 drivers/scsi/libsas/sas_phy.c          |  157 +++
 drivers/scsi/libsas/sas_port.c         |  279 ++++
 drivers/scsi/libsas/sas_scsi_host.c    |  786 +++++++++++
 include/scsi/libsas.h                  |  627 +++++++++
 include/scsi/sas.h                     |  644 +++++++++
 include/scsi/scsi.h                    |    6 +
 37 files changed, 18866 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/scsi/libsas.txt
 create mode 100644 drivers/scsi/aic94xx/Kconfig
 create mode 100644 drivers/scsi/aic94xx/Makefile
 create mode 100644 drivers/scsi/aic94xx/aic94xx.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_dev.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_dump.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_dump.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_hwi.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_hwi.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_init.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_reg.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_reg.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_reg_def.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_sas.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_scb.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_sds.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_seq.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_seq.h
 create mode 100644 drivers/scsi/aic94xx/aic94xx_task.c
 create mode 100644 drivers/scsi/aic94xx/aic94xx_tmf.c
 create mode 100644 drivers/scsi/libsas/Kconfig
 create mode 100644 drivers/scsi/libsas/Makefile
 create mode 100644 drivers/scsi/libsas/sas_discover.c
 create mode 100644 drivers/scsi/libsas/sas_dump.c
 create mode 100644 drivers/scsi/libsas/sas_dump.h
 create mode 100644 drivers/scsi/libsas/sas_event.c
 create mode 100644 drivers/scsi/libsas/sas_expander.c
 create mode 100644 drivers/scsi/libsas/sas_init.c
 create mode 100644 drivers/scsi/libsas/sas_internal.h
 create mode 100644 drivers/scsi/libsas/sas_phy.c
 create mode 100644 drivers/scsi/libsas/sas_port.c
 create mode 100644 drivers/scsi/libsas/sas_scsi_host.c
 create mode 100644 include/scsi/libsas.h
 create mode 100644 include/scsi/sas.h

(limited to 'include')

diff --git a/Documentation/scsi/libsas.txt b/Documentation/scsi/libsas.txt
new file mode 100644
index 000000000000..9e2078b2a615
--- /dev/null
+++ b/Documentation/scsi/libsas.txt
@@ -0,0 +1,484 @@
+SAS Layer
+---------
+
+The SAS Layer is a management infrastructure which manages
+SAS LLDDs.  It sits between SCSI Core and SAS LLDDs.  The
+layout is as follows: while SCSI Core is concerned with
+SAM/SPC issues, and a SAS LLDD+sequencer is concerned with
+phy/OOB/link management, the SAS layer is concerned with:
+
+      * SAS Phy/Port/HA event management (LLDD generates,
+        SAS Layer processes),
+      * SAS Port management (creation/destruction),
+      * SAS Domain discovery and revalidation,
+      * SAS Domain device management,
+      * SCSI Host registration/unregistration,
+      * Device registration with SCSI Core (SAS) or libata
+        (SATA), and
+      * Expander management and exporting expander control
+        to user space.
+
+A SAS LLDD is a PCI device driver.  It is concerned with
+phy/OOB management, and vendor specific tasks and generates
+events to the SAS layer.
+
+The SAS Layer does most SAS tasks as outlined in the SAS 1.1
+spec.
+
+The sas_ha_struct describes the SAS LLDD to the SAS layer.
+Most of it is used by the SAS Layer but a few fields need to
+be initialized by the LLDDs.
+
+After initializing your hardware, from the probe() function
+you call sas_register_ha(). It will register your LLDD with
+the SCSI subsystem, creating a SCSI host and it will
+register your SAS driver with the sysfs SAS tree it creates.
+It will then return.  Then you enable your phys to actually
+start OOB (at which point your driver will start calling the
+notify_* event callbacks).
+
+Structure descriptions:
+
+struct sas_phy --------------------
+Normally this is statically embedded to your driver's
+phy structure:
+	struct my_phy {
+	       blah;
+	       struct sas_phy sas_phy;
+	       bleh;
+	};
+And then all the phys are an array of my_phy in your HA
+struct (shown below).
+
+Then as you go along and initialize your phys you also
+initialize the sas_phy struct, along with your own
+phy structure.
+
+In general, the phys are managed by the LLDD and the ports
+are managed by the SAS layer.  So the phys are initialized
+and updated by the LLDD and the ports are initialized and
+updated by the SAS layer.
+
+There is a scheme where the LLDD can RW certain fields,
+and the SAS layer can only read such ones, and vice versa.
+The idea is to avoid unnecessary locking.
+
+enabled -- must be set (0/1)
+id -- must be set [0,MAX_PHYS)
+class, proto, type, role, oob_mode, linkrate -- must be set
+oob_mode --  you set this when OOB has finished and then notify
+the SAS Layer.
+
+sas_addr -- this normally points to an array holding the sas
+address of the phy, possibly somewhere in your my_phy
+struct.
+
+attached_sas_addr -- set this when you (LLDD) receive an
+IDENTIFY frame or a FIS frame, _before_ notifying the SAS
+layer.  The idea is that sometimes the LLDD may want to fake
+or provide a different SAS address on that phy/port and this
+allows it to do this.  At best you should copy the sas
+address from the IDENTIFY frame or maybe generate a SAS
+address for SATA directly attached devices.  The Discover
+process may later change this.
+
+frame_rcvd -- this is where you copy the IDENTIFY/FIS frame
+when you get it; you lock, copy, set frame_rcvd_size and
+unlock the lock, and then call the event.  It is a pointer
+since there's no way to know your hw frame size _exactly_,
+so you define the actual array in your phy struct and let
+this pointer point to it.  You copy the frame from your
+DMAable memory to that area holding the lock.
+
+sas_prim -- this is where primitives go when they're
+received.  See sas.h. Grab the lock, set the primitive,
+release the lock, notify.
+
+port -- this points to the sas_port if the phy belongs
+to a port -- the LLDD only reads this. It points to the
+sas_port this phy is part of.  Set by the SAS Layer.
+
+ha -- may be set; the SAS layer sets it anyway.
+
+lldd_phy -- you should set this to point to your phy so you
+can find your way around faster when the SAS layer calls one
+of your callbacks and passes you a phy.  If the sas_phy is
+embedded you can also use container_of -- whatever you
+prefer.
+
+
+struct sas_port --------------------
+The LLDD doesn't set any fields of this struct -- it only
+reads them.  They should be self explanatory.
+
+phy_mask is 32 bit, this should be enough for now, as I
+haven't heard of a HA having more than 8 phys.
+
+lldd_port -- I haven't found use for that -- maybe other
+LLDD who wish to have internal port representation can make
+use of this.
+
+
+struct sas_ha_struct --------------------
+It normally is statically declared in your own LLDD
+structure describing your adapter:
+struct my_sas_ha {
+       blah;
+       struct sas_ha_struct sas_ha;
+       struct my_phy phys[MAX_PHYS];
+       struct sas_port sas_ports[MAX_PHYS]; /* (1) */
+       bleh;
+};
+
+(1) If your LLDD doesn't have its own port representation.
+
+What needs to be initialized (sample function given below).
+
+pcidev
+sas_addr -- since the SAS layer doesn't want to mess with
+	 memory allocation, etc, this points to statically
+	 allocated array somewhere (say in your host adapter
+	 structure) and holds the SAS address of the host
+	 adapter as given by you or the manufacturer, etc.
+sas_port
+sas_phy -- an array of pointers to structures. (see
+	note above on sas_addr).
+	These must be set.  See more notes below.
+num_phys -- the number of phys present in the sas_phy array,
+	 and the number of ports present in the sas_port
+	 array.  There can be a maximum num_phys ports (one per
+	 port) so we drop the num_ports, and only use
+	 num_phys.
+
+The event interface:
+
+	/* LLDD calls these to notify the class of an event. */
+	void (*notify_ha_event)(struct sas_ha_struct *, enum ha_event);
+	void (*notify_port_event)(struct sas_phy *, enum port_event);
+	void (*notify_phy_event)(struct sas_phy *, enum phy_event);
+
+When sas_register_ha() returns, those are set and can be
+called by the LLDD to notify the SAS layer of such events
+the SAS layer.
+
+The port notification:
+
+	/* The class calls these to notify the LLDD of an event. */
+	void (*lldd_port_formed)(struct sas_phy *);
+	void (*lldd_port_deformed)(struct sas_phy *);
+
+If the LLDD wants notification when a port has been formed
+or deformed it sets those to a function satisfying the type.
+
+A SAS LLDD should also implement at least one of the Task
+Management Functions (TMFs) described in SAM:
+
+	/* Task Management Functions. Must be called from process context. */
+	int (*lldd_abort_task)(struct sas_task *);
+	int (*lldd_abort_task_set)(struct domain_device *, u8 *lun);
+	int (*lldd_clear_aca)(struct domain_device *, u8 *lun);
+	int (*lldd_clear_task_set)(struct domain_device *, u8 *lun);
+	int (*lldd_I_T_nexus_reset)(struct domain_device *);
+	int (*lldd_lu_reset)(struct domain_device *, u8 *lun);
+	int (*lldd_query_task)(struct sas_task *);
+
+For more information please read SAM from T10.org.
+
+Port and Adapter management:
+
+	/* Port and Adapter management */
+	int (*lldd_clear_nexus_port)(struct sas_port *);
+	int (*lldd_clear_nexus_ha)(struct sas_ha_struct *);
+
+A SAS LLDD should implement at least one of those.
+
+Phy management:
+
+	/* Phy management */
+	int (*lldd_control_phy)(struct sas_phy *, enum phy_func);
+
+lldd_ha -- set this to point to your HA struct. You can also
+use container_of if you embedded it as shown above.
+
+A sample initialization and registration function
+can look like this (called last thing from probe())
+*but* before you enable the phys to do OOB:
+
+static int register_sas_ha(struct my_sas_ha *my_ha)
+{
+	int i;
+	static struct sas_phy   *sas_phys[MAX_PHYS];
+	static struct sas_port  *sas_ports[MAX_PHYS];
+
+	my_ha->sas_ha.sas_addr = &my_ha->sas_addr[0];
+
+	for (i = 0; i < MAX_PHYS; i++) {
+		sas_phys[i] = &my_ha->phys[i].sas_phy;
+		sas_ports[i] = &my_ha->sas_ports[i];
+	}
+
+	my_ha->sas_ha.sas_phy  = sas_phys;
+	my_ha->sas_ha.sas_port = sas_ports;
+	my_ha->sas_ha.num_phys = MAX_PHYS;
+
+	my_ha->sas_ha.lldd_port_formed = my_port_formed;
+
+	my_ha->sas_ha.lldd_dev_found = my_dev_found;
+	my_ha->sas_ha.lldd_dev_gone = my_dev_gone;
+
+	my_ha->sas_ha.lldd_max_execute_num = lldd_max_execute_num; (1)
+
+	my_ha->sas_ha.lldd_queue_size = ha_can_queue;
+	my_ha->sas_ha.lldd_execute_task = my_execute_task;
+
+	my_ha->sas_ha.lldd_abort_task     = my_abort_task;
+	my_ha->sas_ha.lldd_abort_task_set = my_abort_task_set;
+	my_ha->sas_ha.lldd_clear_aca      = my_clear_aca;
+	my_ha->sas_ha.lldd_clear_task_set = my_clear_task_set;
+	my_ha->sas_ha.lldd_I_T_nexus_reset= NULL; (2)
+	my_ha->sas_ha.lldd_lu_reset       = my_lu_reset;
+	my_ha->sas_ha.lldd_query_task     = my_query_task;
+
+	my_ha->sas_ha.lldd_clear_nexus_port = my_clear_nexus_port;
+	my_ha->sas_ha.lldd_clear_nexus_ha = my_clear_nexus_ha;
+
+	my_ha->sas_ha.lldd_control_phy = my_control_phy;
+
+	return sas_register_ha(&my_ha->sas_ha);
+}
+
+(1) This is normally a LLDD parameter, something of the
+lines of a task collector.  What it tells the SAS Layer is
+whether the SAS layer should run in Direct Mode (default:
+value 0 or 1) or Task Collector Mode (value greater than 1).
+
+In Direct Mode, the SAS Layer calls Execute Task as soon as
+it has a command to send to the SDS, _and_ this is a single
+command, i.e. not linked.
+
+Some hardware (e.g. aic94xx) has the capability to DMA more
+than one task at a time (interrupt) from host memory.  Task
+Collector Mode is an optional feature for HAs which support
+this in their hardware.  (Again, it is completely optional
+even if your hardware supports it.)
+
+In Task Collector Mode, the SAS Layer would do _natural_
+coalescing of tasks and at the appropriate moment it would
+call your driver to DMA more than one task in a single HA
+interrupt. DMBS may want to use this by insmod/modprobe
+setting the lldd_max_execute_num to something greater than
+1.
+
+(2) SAS 1.1 does not define I_T Nexus Reset TMF.
+
+Events
+------
+
+Events are _the only way_ a SAS LLDD notifies the SAS layer
+of anything.  There is no other method or way a LLDD to tell
+the SAS layer of anything happening internally or in the SAS
+domain.
+
+Phy events:
+	PHYE_LOSS_OF_SIGNAL, (C)
+	PHYE_OOB_DONE,
+	PHYE_OOB_ERROR,      (C)
+	PHYE_SPINUP_HOLD.
+
+Port events, passed on a _phy_:
+	PORTE_BYTES_DMAED,      (M)
+	PORTE_BROADCAST_RCVD,   (E)
+	PORTE_LINK_RESET_ERR,   (C)
+	PORTE_TIMER_EVENT,      (C)
+	PORTE_HARD_RESET.
+
+Host Adapter event:
+	HAE_RESET
+
+A SAS LLDD should be able to generate
+	- at least one event from group C (choice),
+	- events marked M (mandatory) are mandatory (only one),
+	- events marked E (expander) if it wants the SAS layer
+	  to handle domain revalidation (only one such).
+	- Unmarked events are optional.
+
+Meaning:
+
+HAE_RESET -- when your HA got internal error and was reset.
+
+PORTE_BYTES_DMAED -- on receiving an IDENTIFY/FIS frame
+PORTE_BROADCAST_RCVD -- on receiving a primitive
+PORTE_LINK_RESET_ERR -- timer expired, loss of signal, loss
+of DWS, etc. (*)
+PORTE_TIMER_EVENT -- DWS reset timeout timer expired (*)
+PORTE_HARD_RESET -- Hard Reset primitive received.
+
+PHYE_LOSS_OF_SIGNAL -- the device is gone (*)
+PHYE_OOB_DONE -- OOB went fine and oob_mode is valid
+PHYE_OOB_ERROR -- Error while doing OOB, the device probably
+got disconnected. (*)
+PHYE_SPINUP_HOLD -- SATA is present, COMWAKE not sent.
+
+(*) should set/clear the appropriate fields in the phy,
+    or alternatively call the inlined sas_phy_disconnected()
+    which is just a helper, from their tasklet.
+
+The Execute Command SCSI RPC:
+
+	int (*lldd_execute_task)(struct sas_task *, int num,
+				 unsigned long gfp_flags);
+
+Used to queue a task to the SAS LLDD.  @task is the tasks to
+be executed.  @num should be the number of tasks being
+queued at this function call (they are linked listed via
+task::list), @gfp_mask should be the gfp_mask defining the
+context of the caller.
+
+This function should implement the Execute Command SCSI RPC,
+or if you're sending a SCSI Task as linked commands, you
+should also use this function.
+
+That is, when lldd_execute_task() is called, the command(s)
+go out on the transport *immediately*.  There is *no*
+queuing of any sort and at any level in a SAS LLDD.
+
+The use of task::list is two-fold, one for linked commands,
+the other discussed below.
+
+It is possible to queue up more than one task at a time, by
+initializing the list element of struct sas_task, and
+passing the number of tasks enlisted in this manner in num.
+
+Returns: -SAS_QUEUE_FULL, -ENOMEM, nothing was queued;
+	 0, the task(s) were queued.
+
+If you want to pass num > 1, then either
+A) you're the only caller of this function and keep track
+   of what you've queued to the LLDD, or
+B) you know what you're doing and have a strategy of
+   retrying.
+
+As opposed to queuing one task at a time (function call),
+batch queuing of tasks, by having num > 1, greatly
+simplifies LLDD code, sequencer code, and _hardware design_,
+and has some performance advantages in certain situations
+(DBMS).
+
+The LLDD advertises if it can take more than one command at
+a time at lldd_execute_task(), by setting the
+lldd_max_execute_num parameter (controlled by "collector"
+module parameter in aic94xx SAS LLDD).
+
+You should leave this to the default 1, unless you know what
+you're doing.
+
+This is a function of the LLDD, to which the SAS layer can
+cater to.
+
+int lldd_queue_size
+	The host adapter's queue size.  This is the maximum
+number of commands the lldd can have pending to domain
+devices on behalf of all upper layers submitting through
+lldd_execute_task().
+
+You really want to set this to something (much) larger than
+1.
+
+This _really_ has absolutely nothing to do with queuing.
+There is no queuing in SAS LLDDs.
+
+struct sas_task {
+	dev -- the device this task is destined to
+	list -- must be initialized (INIT_LIST_HEAD)
+	task_proto -- _one_ of enum sas_proto
+	scatter -- pointer to scatter gather list array
+	num_scatter -- number of elements in scatter
+	total_xfer_len -- total number of bytes expected to be transfered
+	data_dir -- PCI_DMA_...
+	task_done -- callback when the task has finished execution
+};
+
+When an external entity, entity other than the LLDD or the
+SAS Layer, wants to work with a struct domain_device, it
+_must_ call kobject_get() when getting a handle on the
+device and kobject_put() when it is done with the device.
+
+This does two things:
+     A) implements proper kfree() for the device;
+     B) increments/decrements the kref for all players:
+     domain_device
+	all domain_device's ... (if past an expander)
+	    port
+		host adapter
+		     pci device
+			 and up the ladder, etc.
+
+DISCOVERY
+---------
+
+The sysfs tree has the following purposes:
+    a) It shows you the physical layout of the SAS domain at
+       the current time, i.e. how the domain looks in the
+       physical world right now.
+    b) Shows some device parameters _at_discovery_time_.
+
+This is a link to the tree(1) program, very useful in
+viewing the SAS domain:
+ftp://mama.indstate.edu/linux/tree/
+I expect user space applications to actually create a
+graphical interface of this.
+
+That is, the sysfs domain tree doesn't show or keep state if
+you e.g., change the meaning of the READY LED MEANING
+setting, but it does show you the current connection status
+of the domain device.
+
+Keeping internal device state changes is responsibility of
+upper layers (Command set drivers) and user space.
+
+When a device or devices are unplugged from the domain, this
+is reflected in the sysfs tree immediately, and the device(s)
+removed from the system.
+
+The structure domain_device describes any device in the SAS
+domain.  It is completely managed by the SAS layer.  A task
+points to a domain device, this is how the SAS LLDD knows
+where to send the task(s) to.  A SAS LLDD only reads the
+contents of the domain_device structure, but it never creates
+or destroys one.
+
+Expander management from User Space
+-----------------------------------
+
+In each expander directory in sysfs, there is a file called
+"smp_portal".  It is a binary sysfs attribute file, which
+implements an SMP portal (Note: this is *NOT* an SMP port),
+to which user space applications can send SMP requests and
+receive SMP responses.
+
+Functionality is deceptively simple:
+
+1. Build the SMP frame you want to send. The format and layout
+   is described in the SAS spec.  Leave the CRC field equal 0.
+open(2)
+2. Open the expander's SMP portal sysfs file in RW mode.
+write(2)
+3. Write the frame you built in 1.
+read(2)
+4. Read the amount of data you expect to receive for the frame you built.
+   If you receive different amount of data you expected to receive,
+   then there was some kind of error.
+close(2)
+All this process is shown in detail in the function do_smp_func()
+and its callers, in the file "expander_conf.c".
+
+The kernel functionality is implemented in the file
+"sas_expander.c".
+
+The program "expander_conf.c" implements this. It takes one
+argument, the sysfs file name of the SMP portal to the
+expander, and gives expander information, including routing
+tables.
+
+The SMP portal gives you complete control of the expander,
+so please be careful.
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index d61662c1a0ee..7de5fdfdab67 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -209,7 +209,7 @@ config SCSI_LOGGING
 	  there should be no noticeable performance impact as long as you have
 	  logging turned off.
 
-menu "SCSI Transport Attributes"
+menu "SCSI Transports"
 	depends on SCSI
 
 config SCSI_SPI_ATTRS
@@ -242,6 +242,8 @@ config SCSI_SAS_ATTRS
 	  If you wish to export transport-specific information about
 	  each attached SAS device to sysfs, say Y.
 
+source "drivers/scsi/libsas/Kconfig"
+
 endmenu
 
 menu "SCSI low-level drivers"
@@ -431,6 +433,7 @@ config SCSI_AIC7XXX_OLD
 	  module will be called aic7xxx_old.
 
 source "drivers/scsi/aic7xxx/Kconfig.aic79xx"
+source "drivers/scsi/aic94xx/Kconfig"
 
 # All the I2O code and drivers do not seem to be 64bit safe.
 config SCSI_DPT_I2O
diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index b2de9bfdfdcd..83da70decdd1 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -32,6 +32,7 @@ obj-$(CONFIG_SCSI_SPI_ATTRS)	+= scsi_transport_spi.o
 obj-$(CONFIG_SCSI_FC_ATTRS) 	+= scsi_transport_fc.o
 obj-$(CONFIG_SCSI_ISCSI_ATTRS)	+= scsi_transport_iscsi.o
 obj-$(CONFIG_SCSI_SAS_ATTRS)	+= scsi_transport_sas.o
+obj-$(CONFIG_SCSI_SAS_LIBSAS)	+= libsas/
 
 obj-$(CONFIG_ISCSI_TCP) 	+= libiscsi.o	iscsi_tcp.o
 obj-$(CONFIG_INFINIBAND_ISER) 	+= libiscsi.o
@@ -68,6 +69,7 @@ obj-$(CONFIG_SCSI_AIC7XXX)	+= aic7xxx/
 obj-$(CONFIG_SCSI_AIC79XX)	+= aic7xxx/
 obj-$(CONFIG_SCSI_AACRAID)	+= aacraid/
 obj-$(CONFIG_SCSI_AIC7XXX_OLD)	+= aic7xxx_old.o
+obj-$(CONFIG_SCSI_AIC94XX)	+= aic94xx/
 obj-$(CONFIG_SCSI_IPS)		+= ips.o
 obj-$(CONFIG_SCSI_FD_MCS)	+= fd_mcs.o
 obj-$(CONFIG_SCSI_FUTURE_DOMAIN)+= fdomain.o
diff --git a/drivers/scsi/aic94xx/Kconfig b/drivers/scsi/aic94xx/Kconfig
new file mode 100644
index 000000000000..0ed391d8ee84
--- /dev/null
+++ b/drivers/scsi/aic94xx/Kconfig
@@ -0,0 +1,41 @@
+#
+# Kernel configuration file for aic94xx SAS/SATA driver.
+#
+# Copyright (c) 2005 Adaptec, Inc.  All rights reserved.
+# Copyright (c) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+#
+# This file is licensed under GPLv2.
+#
+# This file is part of the aic94xx driver.
+#
+# The aic94xx driver is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation; version 2 of the
+# License.
+#
+# The aic94xx driver is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with Aic94xx Driver; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#
+#
+
+config SCSI_AIC94XX
+	tristate "Adaptec AIC94xx SAS/SATA support"
+	depends on PCI
+	select SCSI_SAS_LIBSAS
+	help
+		This driver supports Adaptec's SAS/SATA 3Gb/s 64 bit PCI-X
+		AIC94xx chip based host adapters.
+
+config AIC94XX_DEBUG
+	bool "Compile in debug mode"
+	default y
+	depends on SCSI_AIC94XX
+	help
+		Compiles the aic94xx driver in debug mode.  In debug mode,
+		the driver prints some messages to the console.
diff --git a/drivers/scsi/aic94xx/Makefile b/drivers/scsi/aic94xx/Makefile
new file mode 100644
index 000000000000..e6b70123940c
--- /dev/null
+++ b/drivers/scsi/aic94xx/Makefile
@@ -0,0 +1,39 @@
+#
+# Makefile for Adaptec aic94xx SAS/SATA driver.
+#
+# Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+# Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+#
+# This file is licensed under GPLv2.
+#
+# This file is part of the the aic94xx driver.
+#
+# The aic94xx driver is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation; version 2 of the
+# License.
+#
+# The aic94xx driver is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with the aic94xx driver; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+
+ifeq ($(CONFIG_AIC94XX_DEBUG),y)
+	EXTRA_CFLAGS += -DASD_DEBUG -DASD_ENTER_EXIT
+endif
+
+obj-$(CONFIG_SCSI_AIC94XX) += aic94xx.o
+aic94xx-y += aic94xx_init.o \
+	     aic94xx_hwi.o  \
+	     aic94xx_reg.o  \
+	     aic94xx_sds.o  \
+	     aic94xx_seq.o  \
+	     aic94xx_dump.o \
+	     aic94xx_scb.o  \
+	     aic94xx_dev.o  \
+	     aic94xx_tmf.o  \
+	     aic94xx_task.o
diff --git a/drivers/scsi/aic94xx/aic94xx.h b/drivers/scsi/aic94xx/aic94xx.h
new file mode 100644
index 000000000000..cb7caf1c9ce1
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx.h
@@ -0,0 +1,114 @@
+/*
+ * Aic94xx SAS/SATA driver header file.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ * $Id: //depot/aic94xx/aic94xx.h#31 $
+ */
+
+#ifndef _AIC94XX_H_
+#define _AIC94XX_H_
+
+#include <linux/slab.h>
+#include <linux/ctype.h>
+#include <scsi/libsas.h>
+
+#define ASD_DRIVER_NAME		"aic94xx"
+#define ASD_DRIVER_DESCRIPTION	"Adaptec aic94xx SAS/SATA driver"
+
+#define asd_printk(fmt, ...)	printk(KERN_NOTICE ASD_DRIVER_NAME ": " fmt, ## __VA_ARGS__)
+
+#ifdef ASD_ENTER_EXIT
+#define ENTER  printk(KERN_NOTICE "%s: ENTER %s\n", ASD_DRIVER_NAME, \
+		__FUNCTION__)
+#define EXIT   printk(KERN_NOTICE "%s: --EXIT %s\n", ASD_DRIVER_NAME, \
+		__FUNCTION__)
+#else
+#define ENTER
+#define EXIT
+#endif
+
+#ifdef ASD_DEBUG
+#define ASD_DPRINTK asd_printk
+#else
+#define ASD_DPRINTK(fmt, ...)
+#endif
+
+/* 2*ITNL timeout + 1 second */
+#define AIC94XX_SCB_TIMEOUT  (5*HZ)
+
+extern kmem_cache_t *asd_dma_token_cache;
+extern kmem_cache_t *asd_ascb_cache;
+extern char sas_addr_str[2*SAS_ADDR_SIZE + 1];
+
+static inline void asd_stringify_sas_addr(char *p, const u8 *sas_addr)
+{
+	int i;
+	for (i = 0; i < SAS_ADDR_SIZE; i++, p += 2)
+		snprintf(p, 3, "%02X", sas_addr[i]);
+	*p = '\0';
+}
+
+static inline void asd_destringify_sas_addr(u8 *sas_addr, const char *p)
+{
+	int i;
+	for (i = 0; i < SAS_ADDR_SIZE; i++) {
+		u8 h, l;
+		if (!*p)
+			break;
+		h = isdigit(*p) ? *p-'0' : *p-'A'+10;
+		p++;
+		l = isdigit(*p) ? *p-'0' : *p-'A'+10;
+		p++;
+		sas_addr[i] = (h<<4) | l;
+	}
+}
+
+struct asd_ha_struct;
+struct asd_ascb;
+
+int  asd_read_ocm(struct asd_ha_struct *asd_ha);
+int  asd_read_flash(struct asd_ha_struct *asd_ha);
+
+int  asd_dev_found(struct domain_device *dev);
+void asd_dev_gone(struct domain_device *dev);
+
+void asd_invalidate_edb(struct asd_ascb *ascb, int edb_id);
+
+int  asd_execute_task(struct sas_task *, int num, unsigned long gfp_flags);
+
+/* ---------- TMFs ---------- */
+int  asd_abort_task(struct sas_task *);
+int  asd_abort_task_set(struct domain_device *, u8 *lun);
+int  asd_clear_aca(struct domain_device *, u8 *lun);
+int  asd_clear_task_set(struct domain_device *, u8 *lun);
+int  asd_lu_reset(struct domain_device *, u8 *lun);
+int  asd_query_task(struct sas_task *);
+
+/* ---------- Adapter and Port management ---------- */
+int  asd_clear_nexus_port(struct asd_sas_port *port);
+int  asd_clear_nexus_ha(struct sas_ha_struct *sas_ha);
+
+/* ---------- Phy Management ---------- */
+int  asd_control_phy(struct asd_sas_phy *phy, enum phy_func func);
+
+#endif
diff --git a/drivers/scsi/aic94xx/aic94xx_dev.c b/drivers/scsi/aic94xx/aic94xx_dev.c
new file mode 100644
index 000000000000..6f8901b748f7
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_dev.c
@@ -0,0 +1,353 @@
+/*
+ * Aic94xx SAS/SATA DDB management
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ * $Id: //depot/aic94xx/aic94xx_dev.c#21 $
+ */
+
+#include "aic94xx.h"
+#include "aic94xx_hwi.h"
+#include "aic94xx_reg.h"
+#include "aic94xx_sas.h"
+
+#define FIND_FREE_DDB(_ha) find_first_zero_bit((_ha)->hw_prof.ddb_bitmap, \
+					       (_ha)->hw_prof.max_ddbs)
+#define SET_DDB(_ddb, _ha) set_bit(_ddb, (_ha)->hw_prof.ddb_bitmap)
+#define CLEAR_DDB(_ddb, _ha) clear_bit(_ddb, (_ha)->hw_prof.ddb_bitmap)
+
+static inline int asd_get_ddb(struct asd_ha_struct *asd_ha)
+{
+	unsigned long flags;
+	int ddb, i;
+
+	spin_lock_irqsave(&asd_ha->hw_prof.ddb_lock, flags);
+	ddb = FIND_FREE_DDB(asd_ha);
+	if (ddb >= asd_ha->hw_prof.max_ddbs) {
+		ddb = -ENOMEM;
+		spin_unlock_irqrestore(&asd_ha->hw_prof.ddb_lock, flags);
+		goto out;
+	}
+	SET_DDB(ddb, asd_ha);
+	spin_unlock_irqrestore(&asd_ha->hw_prof.ddb_lock, flags);
+
+	for (i = 0; i < sizeof(struct asd_ddb_ssp_smp_target_port); i+= 4)
+		asd_ddbsite_write_dword(asd_ha, ddb, i, 0);
+out:
+	return ddb;
+}
+
+#define INIT_CONN_TAG   offsetof(struct asd_ddb_ssp_smp_target_port, init_conn_tag)
+#define DEST_SAS_ADDR   offsetof(struct asd_ddb_ssp_smp_target_port, dest_sas_addr)
+#define SEND_QUEUE_HEAD offsetof(struct asd_ddb_ssp_smp_target_port, send_queue_head)
+#define DDB_TYPE        offsetof(struct asd_ddb_ssp_smp_target_port, ddb_type)
+#define CONN_MASK       offsetof(struct asd_ddb_ssp_smp_target_port, conn_mask)
+#define DDB_TARG_FLAGS  offsetof(struct asd_ddb_ssp_smp_target_port, flags)
+#define DDB_TARG_FLAGS2 offsetof(struct asd_ddb_stp_sata_target_port, flags2)
+#define EXEC_QUEUE_TAIL offsetof(struct asd_ddb_ssp_smp_target_port, exec_queue_tail)
+#define SEND_QUEUE_TAIL offsetof(struct asd_ddb_ssp_smp_target_port, send_queue_tail)
+#define SISTER_DDB      offsetof(struct asd_ddb_ssp_smp_target_port, sister_ddb)
+#define MAX_CCONN       offsetof(struct asd_ddb_ssp_smp_target_port, max_concurrent_conn)
+#define NUM_CTX         offsetof(struct asd_ddb_ssp_smp_target_port, num_contexts)
+#define ATA_CMD_SCBPTR  offsetof(struct asd_ddb_stp_sata_target_port, ata_cmd_scbptr)
+#define SATA_TAG_ALLOC_MASK offsetof(struct asd_ddb_stp_sata_target_port, sata_tag_alloc_mask)
+#define NUM_SATA_TAGS   offsetof(struct asd_ddb_stp_sata_target_port, num_sata_tags)
+#define SATA_STATUS     offsetof(struct asd_ddb_stp_sata_target_port, sata_status)
+#define NCQ_DATA_SCB_PTR offsetof(struct asd_ddb_stp_sata_target_port, ncq_data_scb_ptr)
+#define ITNL_TIMEOUT    offsetof(struct asd_ddb_ssp_smp_target_port, itnl_timeout)
+
+static inline void asd_free_ddb(struct asd_ha_struct *asd_ha, int ddb)
+{
+	unsigned long flags;
+
+	if (!ddb || ddb >= 0xFFFF)
+		return;
+	asd_ddbsite_write_byte(asd_ha, ddb, DDB_TYPE, DDB_TYPE_UNUSED);
+	spin_lock_irqsave(&asd_ha->hw_prof.ddb_lock, flags);
+	CLEAR_DDB(ddb, asd_ha);
+	spin_unlock_irqrestore(&asd_ha->hw_prof.ddb_lock, flags);
+}
+
+static inline void asd_set_ddb_type(struct domain_device *dev)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	int ddb = (int) (unsigned long) dev->lldd_dev;
+
+	if (dev->dev_type == SATA_PM_PORT)
+		asd_ddbsite_write_byte(asd_ha,ddb, DDB_TYPE, DDB_TYPE_PM_PORT);
+	else if (dev->tproto)
+		asd_ddbsite_write_byte(asd_ha,ddb, DDB_TYPE, DDB_TYPE_TARGET);
+	else
+		asd_ddbsite_write_byte(asd_ha,ddb,DDB_TYPE,DDB_TYPE_INITIATOR);
+}
+
+static int asd_init_sata_tag_ddb(struct domain_device *dev)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	int ddb, i;
+
+	ddb = asd_get_ddb(asd_ha);
+	if (ddb < 0)
+		return ddb;
+
+	for (i = 0; i < sizeof(struct asd_ddb_sata_tag); i += 2)
+		asd_ddbsite_write_word(asd_ha, ddb, i, 0xFFFF);
+
+	asd_ddbsite_write_word(asd_ha, (int) (unsigned long) dev->lldd_dev,
+			       SISTER_DDB, ddb);
+	return 0;
+}
+
+static inline int asd_init_sata(struct domain_device *dev)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	int ddb = (int) (unsigned long) dev->lldd_dev;
+	u32 qdepth = 0;
+	int res = 0;
+
+	asd_ddbsite_write_word(asd_ha, ddb, ATA_CMD_SCBPTR, 0xFFFF);
+	if ((dev->dev_type == SATA_DEV || dev->dev_type == SATA_PM_PORT) &&
+	    dev->sata_dev.identify_device &&
+	    dev->sata_dev.identify_device[10] != 0) {
+		u16 w75 = le16_to_cpu(dev->sata_dev.identify_device[75]);
+		u16 w76 = le16_to_cpu(dev->sata_dev.identify_device[76]);
+
+		if (w76 & 0x100) /* NCQ? */
+			qdepth = (w75 & 0x1F) + 1;
+		asd_ddbsite_write_dword(asd_ha, ddb, SATA_TAG_ALLOC_MASK,
+					(1<<qdepth)-1);
+		asd_ddbsite_write_byte(asd_ha, ddb, NUM_SATA_TAGS, qdepth);
+	}
+	if (dev->dev_type == SATA_DEV || dev->dev_type == SATA_PM ||
+	    dev->dev_type == SATA_PM_PORT) {
+		struct dev_to_host_fis *fis = (struct dev_to_host_fis *)
+			dev->frame_rcvd;
+		asd_ddbsite_write_byte(asd_ha, ddb, SATA_STATUS, fis->status);
+	}
+	asd_ddbsite_write_word(asd_ha, ddb, NCQ_DATA_SCB_PTR, 0xFFFF);
+	if (qdepth > 0)
+		res = asd_init_sata_tag_ddb(dev);
+	return res;
+}
+
+static int asd_init_target_ddb(struct domain_device *dev)
+{
+	int ddb, i;
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	u8 flags = 0;
+
+	ddb = asd_get_ddb(asd_ha);
+	if (ddb < 0)
+		return ddb;
+
+	dev->lldd_dev = (void *) (unsigned long) ddb;
+
+	asd_ddbsite_write_byte(asd_ha, ddb, 0, DDB_TP_CONN_TYPE);
+	asd_ddbsite_write_byte(asd_ha, ddb, 1, 0);
+	asd_ddbsite_write_word(asd_ha, ddb, INIT_CONN_TAG, 0xFFFF);
+	for (i = 0; i < SAS_ADDR_SIZE; i++)
+		asd_ddbsite_write_byte(asd_ha, ddb, DEST_SAS_ADDR+i,
+				       dev->sas_addr[i]);
+	asd_ddbsite_write_word(asd_ha, ddb, SEND_QUEUE_HEAD, 0xFFFF);
+	asd_set_ddb_type(dev);
+	asd_ddbsite_write_byte(asd_ha, ddb, CONN_MASK, dev->port->phy_mask);
+	if (dev->port->oob_mode != SATA_OOB_MODE) {
+		flags |= OPEN_REQUIRED;
+		if ((dev->dev_type == SATA_DEV) ||
+		    (dev->tproto & SAS_PROTO_STP)) {
+			struct smp_resp *rps_resp = &dev->sata_dev.rps_resp;
+			if (rps_resp->frame_type == SMP_RESPONSE &&
+			    rps_resp->function == SMP_REPORT_PHY_SATA &&
+			    rps_resp->result == SMP_RESP_FUNC_ACC) {
+				if (rps_resp->rps.affil_valid)
+					flags |= STP_AFFIL_POL;
+				if (rps_resp->rps.affil_supp)
+					flags |= SUPPORTS_AFFIL;
+			}
+		} else {
+			flags |= CONCURRENT_CONN_SUPP;
+			if (!dev->parent &&
+			    (dev->dev_type == EDGE_DEV ||
+			     dev->dev_type == FANOUT_DEV))
+				asd_ddbsite_write_byte(asd_ha, ddb, MAX_CCONN,
+						       4);
+			else
+				asd_ddbsite_write_byte(asd_ha, ddb, MAX_CCONN,
+						       dev->pathways);
+			asd_ddbsite_write_byte(asd_ha, ddb, NUM_CTX, 1);
+		}
+	}
+	if (dev->dev_type == SATA_PM)
+		flags |= SATA_MULTIPORT;
+	asd_ddbsite_write_byte(asd_ha, ddb, DDB_TARG_FLAGS, flags);
+
+	flags = 0;
+	if (dev->tproto & SAS_PROTO_STP)
+		flags |= STP_CL_POL_NO_TX;
+	asd_ddbsite_write_byte(asd_ha, ddb, DDB_TARG_FLAGS2, flags);
+
+	asd_ddbsite_write_word(asd_ha, ddb, EXEC_QUEUE_TAIL, 0xFFFF);
+	asd_ddbsite_write_word(asd_ha, ddb, SEND_QUEUE_TAIL, 0xFFFF);
+	asd_ddbsite_write_word(asd_ha, ddb, SISTER_DDB, 0xFFFF);
+
+	if (dev->dev_type == SATA_DEV || (dev->tproto & SAS_PROTO_STP)) {
+		i = asd_init_sata(dev);
+		if (i < 0) {
+			asd_free_ddb(asd_ha, ddb);
+			return i;
+		}
+	}
+
+	if (dev->dev_type == SAS_END_DEV) {
+		struct sas_end_device *rdev = rphy_to_end_device(dev->rphy);
+		if (rdev->I_T_nexus_loss_timeout > 0)
+			asd_ddbsite_write_word(asd_ha, ddb, ITNL_TIMEOUT,
+					       min(rdev->I_T_nexus_loss_timeout,
+						   (u16)ITNL_TIMEOUT_CONST));
+		else
+			asd_ddbsite_write_word(asd_ha, ddb, ITNL_TIMEOUT,
+					       (u16)ITNL_TIMEOUT_CONST);
+	}
+	return 0;
+}
+
+static int asd_init_sata_pm_table_ddb(struct domain_device *dev)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	int ddb, i;
+
+	ddb = asd_get_ddb(asd_ha);
+	if (ddb < 0)
+		return ddb;
+
+	for (i = 0; i < 32; i += 2)
+		asd_ddbsite_write_word(asd_ha, ddb, i, 0xFFFF);
+
+	asd_ddbsite_write_word(asd_ha, (int) (unsigned long) dev->lldd_dev,
+			       SISTER_DDB, ddb);
+
+	return 0;
+}
+
+#define PM_PORT_FLAGS offsetof(struct asd_ddb_sata_pm_port, pm_port_flags)
+#define PARENT_DDB    offsetof(struct asd_ddb_sata_pm_port, parent_ddb)
+
+/**
+ * asd_init_sata_pm_port_ddb -- SATA Port Multiplier Port
+ * dev: pointer to domain device
+ *
+ * For SATA Port Multiplier Ports we need to allocate one SATA Port
+ * Multiplier Port DDB and depending on whether the target on it
+ * supports SATA II NCQ, one SATA Tag DDB.
+ */
+static int asd_init_sata_pm_port_ddb(struct domain_device *dev)
+{
+	int ddb, i, parent_ddb, pmtable_ddb;
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	u8  flags;
+
+	ddb = asd_get_ddb(asd_ha);
+	if (ddb < 0)
+		return ddb;
+
+	asd_set_ddb_type(dev);
+	flags = (dev->sata_dev.port_no << 4) | PM_PORT_SET;
+	asd_ddbsite_write_byte(asd_ha, ddb, PM_PORT_FLAGS, flags);
+	asd_ddbsite_write_word(asd_ha, ddb, SISTER_DDB, 0xFFFF);
+	asd_ddbsite_write_word(asd_ha, ddb, ATA_CMD_SCBPTR, 0xFFFF);
+	asd_init_sata(dev);
+
+	parent_ddb = (int) (unsigned long) dev->parent->lldd_dev;
+	asd_ddbsite_write_word(asd_ha, ddb, PARENT_DDB, parent_ddb);
+	pmtable_ddb = asd_ddbsite_read_word(asd_ha, parent_ddb, SISTER_DDB);
+	asd_ddbsite_write_word(asd_ha, pmtable_ddb, dev->sata_dev.port_no,ddb);
+
+	if (asd_ddbsite_read_byte(asd_ha, ddb, NUM_SATA_TAGS) > 0) {
+		i = asd_init_sata_tag_ddb(dev);
+		if (i < 0) {
+			asd_free_ddb(asd_ha, ddb);
+			return i;
+		}
+	}
+	return 0;
+}
+
+static int asd_init_initiator_ddb(struct domain_device *dev)
+{
+	return -ENODEV;
+}
+
+/**
+ * asd_init_sata_pm_ddb -- SATA Port Multiplier
+ * dev: pointer to domain device
+ *
+ * For STP and direct-attached SATA Port Multipliers we need
+ * one target port DDB entry and one SATA PM table DDB entry.
+ */
+static int asd_init_sata_pm_ddb(struct domain_device *dev)
+{
+	int res = 0;
+
+	res = asd_init_target_ddb(dev);
+	if (res)
+		goto out;
+	res = asd_init_sata_pm_table_ddb(dev);
+	if (res)
+		asd_free_ddb(dev->port->ha->lldd_ha,
+			     (int) (unsigned long) dev->lldd_dev);
+out:
+	return res;
+}
+
+int asd_dev_found(struct domain_device *dev)
+{
+	int res = 0;
+
+	switch (dev->dev_type) {
+	case SATA_PM:
+		res = asd_init_sata_pm_ddb(dev);
+		break;
+	case SATA_PM_PORT:
+		res = asd_init_sata_pm_port_ddb(dev);
+		break;
+	default:
+		if (dev->tproto)
+			res = asd_init_target_ddb(dev);
+		else
+			res = asd_init_initiator_ddb(dev);
+	}
+	return res;
+}
+
+void asd_dev_gone(struct domain_device *dev)
+{
+	int ddb, sister_ddb;
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+
+	ddb = (int) (unsigned long) dev->lldd_dev;
+	sister_ddb = asd_ddbsite_read_word(asd_ha, ddb, SISTER_DDB);
+
+	if (sister_ddb != 0xFFFF)
+		asd_free_ddb(asd_ha, sister_ddb);
+	asd_free_ddb(asd_ha, ddb);
+	dev->lldd_dev = NULL;
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_dump.c b/drivers/scsi/aic94xx/aic94xx_dump.c
new file mode 100644
index 000000000000..e6ade5996d95
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_dump.c
@@ -0,0 +1,959 @@
+/*
+ * Aic94xx SAS/SATA driver dump interface.
+ *
+ * Copyright (C) 2004 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2004 David Chaw <david_chaw@adaptec.com>
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ * 2005/07/14/LT  Complete overhaul of this file.  Update pages, register
+ * locations, names, etc.  Make use of macros.  Print more information.
+ * Print all cseq and lseq mip and mdp.
+ *
+ */
+
+#include "linux/pci.h"
+#include "aic94xx.h"
+#include "aic94xx_reg.h"
+#include "aic94xx_reg_def.h"
+#include "aic94xx_sas.h"
+
+#include "aic94xx_dump.h"
+
+#ifdef ASD_DEBUG
+
+#define MD(x)	    (1 << (x))
+#define MODE_COMMON (1 << 31)
+#define MODE_0_7    (0xFF)
+
+static const struct lseq_cio_regs {
+	char	*name;
+	u32	offs;
+	u8	width;
+	u32	mode;
+} LSEQmCIOREGS[] = {
+	{"LmMnSCBPTR",    0x20, 16, MD(0)|MD(1)|MD(2)|MD(3)|MD(4) },
+	{"LmMnDDBPTR",    0x22, 16, MD(0)|MD(1)|MD(2)|MD(3)|MD(4) },
+	{"LmREQMBX",      0x30, 32, MODE_COMMON },
+	{"LmRSPMBX",      0x34, 32, MODE_COMMON },
+	{"LmMnINT",       0x38, 32, MODE_0_7 },
+	{"LmMnINTEN",     0x3C, 32, MODE_0_7 },
+	{"LmXMTPRIMD",    0x40, 32, MODE_COMMON },
+	{"LmXMTPRIMCS",   0x44,  8, MODE_COMMON },
+	{"LmCONSTAT",     0x45,  8, MODE_COMMON },
+	{"LmMnDMAERRS",   0x46,  8, MD(0)|MD(1) },
+	{"LmMnSGDMAERRS", 0x47,  8, MD(0)|MD(1) },
+	{"LmMnEXPHDRP",   0x48,  8, MD(0) },
+	{"LmMnSASAALIGN", 0x48,  8, MD(1) },
+	{"LmMnMSKHDRP",   0x49,  8, MD(0) },
+	{"LmMnSTPALIGN",  0x49,  8, MD(1) },
+	{"LmMnRCVHDRP",   0x4A,  8, MD(0) },
+	{"LmMnXMTHDRP",   0x4A,  8, MD(1) },
+	{"LmALIGNMODE",   0x4B,  8, MD(1) },
+	{"LmMnEXPRCVCNT", 0x4C, 32, MD(0) },
+	{"LmMnXMTCNT",    0x4C, 32, MD(1) },
+	{"LmMnCURRTAG",   0x54, 16, MD(0) },
+	{"LmMnPREVTAG",   0x56, 16, MD(0) },
+	{"LmMnACKOFS",    0x58,  8, MD(1) },
+	{"LmMnXFRLVL",    0x59,  8, MD(0)|MD(1) },
+	{"LmMnSGDMACTL",  0x5A,  8, MD(0)|MD(1) },
+	{"LmMnSGDMASTAT", 0x5B,  8, MD(0)|MD(1) },
+	{"LmMnDDMACTL",   0x5C,  8, MD(0)|MD(1) },
+	{"LmMnDDMASTAT",  0x5D,  8, MD(0)|MD(1) },
+	{"LmMnDDMAMODE",  0x5E, 16, MD(0)|MD(1) },
+	{"LmMnPIPECTL",   0x61,  8, MD(0)|MD(1) },
+	{"LmMnACTSCB",    0x62, 16, MD(0)|MD(1) },
+	{"LmMnSGBHADR",   0x64,  8, MD(0)|MD(1) },
+	{"LmMnSGBADR",    0x65,  8, MD(0)|MD(1) },
+	{"LmMnSGDCNT",    0x66,  8, MD(0)|MD(1) },
+	{"LmMnSGDMADR",   0x68, 32, MD(0)|MD(1) },
+	{"LmMnSGDMADR",   0x6C, 32, MD(0)|MD(1) },
+	{"LmMnXFRCNT",    0x70, 32, MD(0)|MD(1) },
+	{"LmMnXMTCRC",    0x74, 32, MD(1) },
+	{"LmCURRTAG",     0x74, 16, MD(0) },
+	{"LmPREVTAG",     0x76, 16, MD(0) },
+	{"LmMnDPSEL",     0x7B,  8, MD(0)|MD(1) },
+	{"LmDPTHSTAT",    0x7C,  8, MODE_COMMON },
+	{"LmMnHOLDLVL",   0x7D,  8, MD(0) },
+	{"LmMnSATAFS",    0x7E,  8, MD(1) },
+	{"LmMnCMPLTSTAT", 0x7F,  8, MD(0)|MD(1) },
+	{"LmPRMSTAT0",    0x80, 32, MODE_COMMON },
+	{"LmPRMSTAT1",    0x84, 32, MODE_COMMON },
+	{"LmGPRMINT",     0x88,  8, MODE_COMMON },
+        {"LmMnCURRSCB",   0x8A, 16, MD(0) },
+	{"LmPRMICODE",    0x8C, 32, MODE_COMMON },
+	{"LmMnRCVCNT",    0x90, 16, MD(0) },
+	{"LmMnBUFSTAT",   0x92, 16, MD(0) },
+	{"LmMnXMTHDRSIZE",0x92,  8, MD(1) },
+	{"LmMnXMTSIZE",   0x93,  8, MD(1) },
+	{"LmMnTGTXFRCNT", 0x94, 32, MD(0) },
+	{"LmMnEXPROFS",   0x98, 32, MD(0) },
+	{"LmMnXMTROFS",   0x98, 32, MD(1) },
+	{"LmMnRCVROFS",   0x9C, 32, MD(0) },
+	{"LmCONCTL",      0xA0, 16, MODE_COMMON },
+	{"LmBITLTIMER",   0xA2, 16, MODE_COMMON },
+	{"LmWWNLOW",      0xA8, 32, MODE_COMMON },
+	{"LmWWNHIGH",     0xAC, 32, MODE_COMMON },
+	{"LmMnFRMERR",    0xB0, 32, MD(0) },
+	{"LmMnFRMERREN",  0xB4, 32, MD(0) },
+	{"LmAWTIMER",     0xB8, 16, MODE_COMMON },
+	{"LmAWTCTL",      0xBA,  8, MODE_COMMON },
+	{"LmMnHDRCMPS",   0xC0, 32, MD(0) },
+	{"LmMnXMTSTAT",   0xC4,  8, MD(1) },
+	{"LmHWTSTATEN",   0xC5,  8, MODE_COMMON },
+	{"LmMnRRDYRC",    0xC6,  8, MD(0) },
+        {"LmMnRRDYTC",    0xC6,  8, MD(1) },
+	{"LmHWTSTAT",     0xC7,  8, MODE_COMMON },
+	{"LmMnDATABUFADR",0xC8, 16, MD(0)|MD(1) },
+	{"LmDWSSTATUS",   0xCB,  8, MODE_COMMON },
+	{"LmMnACTSTAT",   0xCE, 16, MD(0)|MD(1) },
+	{"LmMnREQSCB",    0xD2, 16, MD(0)|MD(1) },
+	{"LmXXXPRIM",     0xD4, 32, MODE_COMMON },
+	{"LmRCVASTAT",    0xD9,  8, MODE_COMMON },
+	{"LmINTDIS1",     0xDA,  8, MODE_COMMON },
+	{"LmPSTORESEL",   0xDB,  8, MODE_COMMON },
+	{"LmPSTORE",      0xDC, 32, MODE_COMMON },
+	{"LmPRIMSTAT0EN", 0xE0, 32, MODE_COMMON },
+	{"LmPRIMSTAT1EN", 0xE4, 32, MODE_COMMON },
+	{"LmDONETCTL",    0xF2, 16, MODE_COMMON },
+	{NULL, 0, 0, 0 }
+};
+/*
+static struct lseq_cio_regs LSEQmOOBREGS[] = {
+   {"OOB_BFLTR"        ,0x100, 8, MD(5)},
+   {"OOB_INIT_MIN"     ,0x102,16, MD(5)},
+   {"OOB_INIT_MAX"     ,0x104,16, MD(5)},
+   {"OOB_INIT_NEG"     ,0x106,16, MD(5)},
+   {"OOB_SAS_MIN"      ,0x108,16, MD(5)},
+   {"OOB_SAS_MAX"      ,0x10A,16, MD(5)},
+   {"OOB_SAS_NEG"      ,0x10C,16, MD(5)},
+   {"OOB_WAKE_MIN"     ,0x10E,16, MD(5)},
+   {"OOB_WAKE_MAX"     ,0x110,16, MD(5)},
+   {"OOB_WAKE_NEG"     ,0x112,16, MD(5)},
+   {"OOB_IDLE_MAX"     ,0x114,16, MD(5)},
+   {"OOB_BURST_MAX"    ,0x116,16, MD(5)},
+   {"OOB_XMIT_BURST"   ,0x118, 8, MD(5)},
+   {"OOB_SEND_PAIRS"   ,0x119, 8, MD(5)},
+   {"OOB_INIT_IDLE"    ,0x11A, 8, MD(5)},
+   {"OOB_INIT_NEGO"    ,0x11C, 8, MD(5)},
+   {"OOB_SAS_IDLE"     ,0x11E, 8, MD(5)},
+   {"OOB_SAS_NEGO"     ,0x120, 8, MD(5)},
+   {"OOB_WAKE_IDLE"    ,0x122, 8, MD(5)},
+   {"OOB_WAKE_NEGO"    ,0x124, 8, MD(5)},
+   {"OOB_DATA_KBITS"   ,0x126, 8, MD(5)},
+   {"OOB_BURST_DATA"   ,0x128,32, MD(5)},
+   {"OOB_ALIGN_0_DATA" ,0x12C,32, MD(5)},
+   {"OOB_ALIGN_1_DATA" ,0x130,32, MD(5)},
+   {"OOB_SYNC_DATA"    ,0x134,32, MD(5)},
+   {"OOB_D10_2_DATA"   ,0x138,32, MD(5)},
+   {"OOB_PHY_RST_CNT"  ,0x13C,32, MD(5)},
+   {"OOB_SIG_GEN"      ,0x140, 8, MD(5)},
+   {"OOB_XMIT"         ,0x141, 8, MD(5)},
+   {"FUNCTION_MAKS"    ,0x142, 8, MD(5)},
+   {"OOB_MODE"         ,0x143, 8, MD(5)},
+   {"CURRENT_STATUS"   ,0x144, 8, MD(5)},
+   {"SPEED_MASK"       ,0x145, 8, MD(5)},
+   {"PRIM_COUNT"       ,0x146, 8, MD(5)},
+   {"OOB_SIGNALS"      ,0x148, 8, MD(5)},
+   {"OOB_DATA_DET"     ,0x149, 8, MD(5)},
+   {"OOB_TIME_OUT"     ,0x14C, 8, MD(5)},
+   {"OOB_TIMER_ENABLE" ,0x14D, 8, MD(5)},
+   {"OOB_STATUS"       ,0x14E, 8, MD(5)},
+   {"HOT_PLUG_DELAY"   ,0x150, 8, MD(5)},
+   {"RCD_DELAY"        ,0x151, 8, MD(5)},
+   {"COMSAS_TIMER"     ,0x152, 8, MD(5)},
+   {"SNTT_DELAY"       ,0x153, 8, MD(5)},
+   {"SPD_CHNG_DELAY"   ,0x154, 8, MD(5)},
+   {"SNLT_DELAY"       ,0x155, 8, MD(5)},
+   {"SNWT_DELAY"       ,0x156, 8, MD(5)},
+   {"ALIGN_DELAY"      ,0x157, 8, MD(5)},
+   {"INT_ENABLE_0"     ,0x158, 8, MD(5)},
+   {"INT_ENABLE_1"     ,0x159, 8, MD(5)},
+   {"INT_ENABLE_2"     ,0x15A, 8, MD(5)},
+   {"INT_ENABLE_3"     ,0x15B, 8, MD(5)},
+   {"OOB_TEST_REG"     ,0x15C, 8, MD(5)},
+   {"PHY_CONTROL_0"    ,0x160, 8, MD(5)},
+   {"PHY_CONTROL_1"    ,0x161, 8, MD(5)},
+   {"PHY_CONTROL_2"    ,0x162, 8, MD(5)},
+   {"PHY_CONTROL_3"    ,0x163, 8, MD(5)},
+   {"PHY_OOB_CAL_TX"   ,0x164, 8, MD(5)},
+   {"PHY_OOB_CAL_RX"   ,0x165, 8, MD(5)},
+   {"OOB_PHY_CAL_TX"   ,0x166, 8, MD(5)},
+   {"OOB_PHY_CAL_RX"   ,0x167, 8, MD(5)},
+   {"PHY_CONTROL_4"    ,0x168, 8, MD(5)},
+   {"PHY_TEST"         ,0x169, 8, MD(5)},
+   {"PHY_PWR_CTL"      ,0x16A, 8, MD(5)},
+   {"PHY_PWR_DELAY"    ,0x16B, 8, MD(5)},
+   {"OOB_SM_CON"       ,0x16C, 8, MD(5)},
+   {"ADDR_TRAP_1"      ,0x16D, 8, MD(5)},
+   {"ADDR_NEXT_1"      ,0x16E, 8, MD(5)},
+   {"NEXT_ST_1"        ,0x16F, 8, MD(5)},
+   {"OOB_SM_STATE"     ,0x170, 8, MD(5)},
+   {"ADDR_TRAP_2"      ,0x171, 8, MD(5)},
+   {"ADDR_NEXT_2"      ,0x172, 8, MD(5)},
+   {"NEXT_ST_2"        ,0x173, 8, MD(5)},
+   {NULL, 0, 0, 0 }
+};
+*/
+#define STR_8BIT   "   %30s[0x%04x]:0x%02x\n"
+#define STR_16BIT  "   %30s[0x%04x]:0x%04x\n"
+#define STR_32BIT  "   %30s[0x%04x]:0x%08x\n"
+#define STR_64BIT  "   %30s[0x%04x]:0x%llx\n"
+
+#define PRINT_REG_8bit(_ha, _n, _r) asd_printk(STR_8BIT, #_n, _n,      \
+					     asd_read_reg_byte(_ha, _r))
+#define PRINT_REG_16bit(_ha, _n, _r) asd_printk(STR_16BIT, #_n, _n,     \
+					      asd_read_reg_word(_ha, _r))
+#define PRINT_REG_32bit(_ha, _n, _r) asd_printk(STR_32BIT, #_n, _n,      \
+					      asd_read_reg_dword(_ha, _r))
+
+#define PRINT_CREG_8bit(_ha, _n) asd_printk(STR_8BIT, #_n, _n,      \
+					     asd_read_reg_byte(_ha, C##_n))
+#define PRINT_CREG_16bit(_ha, _n) asd_printk(STR_16BIT, #_n, _n,     \
+					      asd_read_reg_word(_ha, C##_n))
+#define PRINT_CREG_32bit(_ha, _n) asd_printk(STR_32BIT, #_n, _n,      \
+					      asd_read_reg_dword(_ha, C##_n))
+
+#define MSTR_8BIT   "   Mode:%02d %30s[0x%04x]:0x%02x\n"
+#define MSTR_16BIT  "   Mode:%02d %30s[0x%04x]:0x%04x\n"
+#define MSTR_32BIT  "   Mode:%02d %30s[0x%04x]:0x%08x\n"
+
+#define PRINT_MREG_8bit(_ha, _m, _n, _r) asd_printk(MSTR_8BIT, _m, #_n, _n,   \
+					     asd_read_reg_byte(_ha, _r))
+#define PRINT_MREG_16bit(_ha, _m, _n, _r) asd_printk(MSTR_16BIT, _m, #_n, _n, \
+					      asd_read_reg_word(_ha, _r))
+#define PRINT_MREG_32bit(_ha, _m, _n, _r) asd_printk(MSTR_32BIT, _m, #_n, _n, \
+					      asd_read_reg_dword(_ha, _r))
+
+/* can also be used for MD when the register is mode aware already */
+#define PRINT_MIS_byte(_ha, _n) asd_printk(STR_8BIT, #_n,CSEQ_##_n-CMAPPEDSCR,\
+                                           asd_read_reg_byte(_ha, CSEQ_##_n))
+#define PRINT_MIS_word(_ha, _n) asd_printk(STR_16BIT,#_n,CSEQ_##_n-CMAPPEDSCR,\
+                                           asd_read_reg_word(_ha, CSEQ_##_n))
+#define PRINT_MIS_dword(_ha, _n)                      \
+        asd_printk(STR_32BIT,#_n,CSEQ_##_n-CMAPPEDSCR,\
+                   asd_read_reg_dword(_ha, CSEQ_##_n))
+#define PRINT_MIS_qword(_ha, _n)                                       \
+        asd_printk(STR_64BIT, #_n,CSEQ_##_n-CMAPPEDSCR,                \
+                   (unsigned long long)(((u64)asd_read_reg_dword(_ha, CSEQ_##_n))     \
+                 | (((u64)asd_read_reg_dword(_ha, (CSEQ_##_n)+4))<<32)))
+
+#define CMDP_REG(_n, _m) (_m*(CSEQ_PAGE_SIZE*2)+CSEQ_##_n)
+#define PRINT_CMDP_word(_ha, _n) \
+asd_printk("%20s 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x\n", \
+	#_n, \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 0)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 1)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 2)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 3)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 4)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 5)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 6)), \
+	asd_read_reg_word(_ha, CMDP_REG(_n, 7)))
+
+#define PRINT_CMDP_byte(_ha, _n) \
+asd_printk("%20s 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x 0x%04x\n", \
+	#_n, \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 0)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 1)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 2)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 3)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 4)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 5)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 6)), \
+	asd_read_reg_byte(_ha, CMDP_REG(_n, 7)))
+
+static void asd_dump_cseq_state(struct asd_ha_struct *asd_ha)
+{
+	int mode;
+
+	asd_printk("CSEQ STATE\n");
+
+	asd_printk("ARP2 REGISTERS\n");
+
+	PRINT_CREG_32bit(asd_ha, ARP2CTL);
+	PRINT_CREG_32bit(asd_ha, ARP2INT);
+	PRINT_CREG_32bit(asd_ha, ARP2INTEN);
+	PRINT_CREG_8bit(asd_ha, MODEPTR);
+	PRINT_CREG_8bit(asd_ha, ALTMODE);
+	PRINT_CREG_8bit(asd_ha, FLAG);
+	PRINT_CREG_8bit(asd_ha, ARP2INTCTL);
+	PRINT_CREG_16bit(asd_ha, STACK);
+	PRINT_CREG_16bit(asd_ha, PRGMCNT);
+	PRINT_CREG_16bit(asd_ha, ACCUM);
+	PRINT_CREG_16bit(asd_ha, SINDEX);
+	PRINT_CREG_16bit(asd_ha, DINDEX);
+	PRINT_CREG_8bit(asd_ha, SINDIR);
+	PRINT_CREG_8bit(asd_ha, DINDIR);
+	PRINT_CREG_8bit(asd_ha, JUMLDIR);
+	PRINT_CREG_8bit(asd_ha, ARP2HALTCODE);
+	PRINT_CREG_16bit(asd_ha, CURRADDR);
+	PRINT_CREG_16bit(asd_ha, LASTADDR);
+	PRINT_CREG_16bit(asd_ha, NXTLADDR);
+
+	asd_printk("IOP REGISTERS\n");
+
+	PRINT_REG_32bit(asd_ha, BISTCTL1, CBISTCTL);
+	PRINT_CREG_32bit(asd_ha, MAPPEDSCR);
+
+	asd_printk("CIO REGISTERS\n");
+
+	for (mode = 0; mode < 9; mode++)
+		PRINT_MREG_16bit(asd_ha, mode, MnSCBPTR, CMnSCBPTR(mode));
+	PRINT_MREG_16bit(asd_ha, 15, MnSCBPTR, CMnSCBPTR(15));
+
+	for (mode = 0; mode < 9; mode++)
+		PRINT_MREG_16bit(asd_ha, mode, MnDDBPTR, CMnDDBPTR(mode));
+	PRINT_MREG_16bit(asd_ha, 15, MnDDBPTR, CMnDDBPTR(15));
+
+	for (mode = 0; mode < 8; mode++)
+		PRINT_MREG_32bit(asd_ha, mode, MnREQMBX, CMnREQMBX(mode));
+	for (mode = 0; mode < 8; mode++)
+		PRINT_MREG_32bit(asd_ha, mode, MnRSPMBX, CMnRSPMBX(mode));
+	for (mode = 0; mode < 8; mode++)
+		PRINT_MREG_32bit(asd_ha, mode, MnINT, CMnINT(mode));
+	for (mode = 0; mode < 8; mode++)
+		PRINT_MREG_32bit(asd_ha, mode, MnINTEN, CMnINTEN(mode));
+
+	PRINT_CREG_8bit(asd_ha, SCRATCHPAGE);
+	for (mode = 0; mode < 8; mode++)
+		PRINT_MREG_8bit(asd_ha, mode, MnSCRATCHPAGE,
+				CMnSCRATCHPAGE(mode));
+
+	PRINT_REG_32bit(asd_ha, CLINKCON, CLINKCON);
+	PRINT_REG_8bit(asd_ha, CCONMSK, CCONMSK);
+	PRINT_REG_8bit(asd_ha, CCONEXIST, CCONEXIST);
+	PRINT_REG_16bit(asd_ha, CCONMODE, CCONMODE);
+	PRINT_REG_32bit(asd_ha, CTIMERCALC, CTIMERCALC);
+	PRINT_REG_8bit(asd_ha, CINTDIS, CINTDIS);
+
+	asd_printk("SCRATCH MEMORY\n");
+
+	asd_printk("MIP 4 >>>>>\n");
+	PRINT_MIS_word(asd_ha, Q_EXE_HEAD);
+	PRINT_MIS_word(asd_ha, Q_EXE_TAIL);
+	PRINT_MIS_word(asd_ha, Q_DONE_HEAD);
+	PRINT_MIS_word(asd_ha, Q_DONE_TAIL);
+	PRINT_MIS_word(asd_ha, Q_SEND_HEAD);
+	PRINT_MIS_word(asd_ha, Q_SEND_TAIL);
+	PRINT_MIS_word(asd_ha, Q_DMA2CHIM_HEAD);
+	PRINT_MIS_word(asd_ha, Q_DMA2CHIM_TAIL);
+	PRINT_MIS_word(asd_ha, Q_COPY_HEAD);
+	PRINT_MIS_word(asd_ha, Q_COPY_TAIL);
+	PRINT_MIS_word(asd_ha, REG0);
+	PRINT_MIS_word(asd_ha, REG1);
+	PRINT_MIS_dword(asd_ha, REG2);
+	PRINT_MIS_byte(asd_ha, LINK_CTL_Q_MAP);
+	PRINT_MIS_byte(asd_ha, MAX_CSEQ_MODE);
+	PRINT_MIS_byte(asd_ha, FREE_LIST_HACK_COUNT);
+
+	asd_printk("MIP 5 >>>>\n");
+	PRINT_MIS_qword(asd_ha, EST_NEXUS_REQ_QUEUE);
+	PRINT_MIS_qword(asd_ha, EST_NEXUS_REQ_COUNT);
+	PRINT_MIS_word(asd_ha, Q_EST_NEXUS_HEAD);
+	PRINT_MIS_word(asd_ha, Q_EST_NEXUS_TAIL);
+	PRINT_MIS_word(asd_ha, NEED_EST_NEXUS_SCB);
+	PRINT_MIS_byte(asd_ha, EST_NEXUS_REQ_HEAD);
+	PRINT_MIS_byte(asd_ha, EST_NEXUS_REQ_TAIL);
+	PRINT_MIS_byte(asd_ha, EST_NEXUS_SCB_OFFSET);
+
+	asd_printk("MIP 6 >>>>\n");
+	PRINT_MIS_word(asd_ha, INT_ROUT_RET_ADDR0);
+	PRINT_MIS_word(asd_ha, INT_ROUT_RET_ADDR1);
+	PRINT_MIS_word(asd_ha, INT_ROUT_SCBPTR);
+	PRINT_MIS_byte(asd_ha, INT_ROUT_MODE);
+	PRINT_MIS_byte(asd_ha, ISR_SCRATCH_FLAGS);
+	PRINT_MIS_word(asd_ha, ISR_SAVE_SINDEX);
+	PRINT_MIS_word(asd_ha, ISR_SAVE_DINDEX);
+	PRINT_MIS_word(asd_ha, Q_MONIRTT_HEAD);
+	PRINT_MIS_word(asd_ha, Q_MONIRTT_TAIL);
+	PRINT_MIS_byte(asd_ha, FREE_SCB_MASK);
+	PRINT_MIS_word(asd_ha, BUILTIN_FREE_SCB_HEAD);
+	PRINT_MIS_word(asd_ha, BUILTIN_FREE_SCB_TAIL);
+	PRINT_MIS_word(asd_ha, EXTENDED_FREE_SCB_HEAD);
+	PRINT_MIS_word(asd_ha, EXTENDED_FREE_SCB_TAIL);
+
+	asd_printk("MIP 7 >>>>\n");
+	PRINT_MIS_qword(asd_ha, EMPTY_REQ_QUEUE);
+	PRINT_MIS_qword(asd_ha, EMPTY_REQ_COUNT);
+	PRINT_MIS_word(asd_ha, Q_EMPTY_HEAD);
+	PRINT_MIS_word(asd_ha, Q_EMPTY_TAIL);
+	PRINT_MIS_word(asd_ha, NEED_EMPTY_SCB);
+	PRINT_MIS_byte(asd_ha, EMPTY_REQ_HEAD);
+	PRINT_MIS_byte(asd_ha, EMPTY_REQ_TAIL);
+	PRINT_MIS_byte(asd_ha, EMPTY_SCB_OFFSET);
+	PRINT_MIS_word(asd_ha, PRIMITIVE_DATA);
+	PRINT_MIS_dword(asd_ha, TIMEOUT_CONST);
+
+	asd_printk("MDP 0 >>>>\n");
+	asd_printk("%-20s %6s %6s %6s %6s %6s %6s %6s %6s\n",
+		   "Mode: ", "0", "1", "2", "3", "4", "5", "6", "7");
+	PRINT_CMDP_word(asd_ha, LRM_SAVE_SINDEX);
+	PRINT_CMDP_word(asd_ha, LRM_SAVE_SCBPTR);
+	PRINT_CMDP_word(asd_ha, Q_LINK_HEAD);
+	PRINT_CMDP_word(asd_ha, Q_LINK_TAIL);
+	PRINT_CMDP_byte(asd_ha, LRM_SAVE_SCRPAGE);
+
+	asd_printk("MDP 0 Mode 8 >>>>\n");
+	PRINT_MIS_word(asd_ha, RET_ADDR);
+	PRINT_MIS_word(asd_ha, RET_SCBPTR);
+	PRINT_MIS_word(asd_ha, SAVE_SCBPTR);
+	PRINT_MIS_word(asd_ha, EMPTY_TRANS_CTX);
+	PRINT_MIS_word(asd_ha, RESP_LEN);
+	PRINT_MIS_word(asd_ha, TMF_SCBPTR);
+	PRINT_MIS_word(asd_ha, GLOBAL_PREV_SCB);
+	PRINT_MIS_word(asd_ha, GLOBAL_HEAD);
+	PRINT_MIS_word(asd_ha, CLEAR_LU_HEAD);
+	PRINT_MIS_byte(asd_ha, TMF_OPCODE);
+	PRINT_MIS_byte(asd_ha, SCRATCH_FLAGS);
+	PRINT_MIS_word(asd_ha, HSB_SITE);
+	PRINT_MIS_word(asd_ha, FIRST_INV_SCB_SITE);
+	PRINT_MIS_word(asd_ha, FIRST_INV_DDB_SITE);
+
+	asd_printk("MDP 1 Mode 8 >>>>\n");
+	PRINT_MIS_qword(asd_ha, LUN_TO_CLEAR);
+	PRINT_MIS_qword(asd_ha, LUN_TO_CHECK);
+
+	asd_printk("MDP 2 Mode 8 >>>>\n");
+	PRINT_MIS_qword(asd_ha, HQ_NEW_POINTER);
+	PRINT_MIS_qword(asd_ha, HQ_DONE_BASE);
+	PRINT_MIS_dword(asd_ha, HQ_DONE_POINTER);
+	PRINT_MIS_byte(asd_ha, HQ_DONE_PASS);
+}
+
+#define PRINT_LREG_8bit(_h, _lseq, _n) \
+        asd_printk(STR_8BIT, #_n, _n, asd_read_reg_byte(_h, Lm##_n(_lseq)))
+#define PRINT_LREG_16bit(_h, _lseq, _n) \
+        asd_printk(STR_16BIT, #_n, _n, asd_read_reg_word(_h, Lm##_n(_lseq)))
+#define PRINT_LREG_32bit(_h, _lseq, _n) \
+        asd_printk(STR_32BIT, #_n, _n, asd_read_reg_dword(_h, Lm##_n(_lseq)))
+
+#define PRINT_LMIP_byte(_h, _lseq, _n)                              \
+	asd_printk(STR_8BIT, #_n, LmSEQ_##_n(_lseq)-LmSCRATCH(_lseq), \
+		   asd_read_reg_byte(_h, LmSEQ_##_n(_lseq)))
+#define PRINT_LMIP_word(_h, _lseq, _n)                              \
+	asd_printk(STR_16BIT, #_n, LmSEQ_##_n(_lseq)-LmSCRATCH(_lseq), \
+		   asd_read_reg_word(_h, LmSEQ_##_n(_lseq)))
+#define PRINT_LMIP_dword(_h, _lseq, _n)                             \
+	asd_printk(STR_32BIT, #_n, LmSEQ_##_n(_lseq)-LmSCRATCH(_lseq), \
+		   asd_read_reg_dword(_h, LmSEQ_##_n(_lseq)))
+#define PRINT_LMIP_qword(_h, _lseq, _n)                                \
+	asd_printk(STR_64BIT, #_n, LmSEQ_##_n(_lseq)-LmSCRATCH(_lseq), \
+		 (unsigned long long)(((unsigned long long) \
+		 asd_read_reg_dword(_h, LmSEQ_##_n(_lseq))) \
+	          | (((unsigned long long) \
+		 asd_read_reg_dword(_h, LmSEQ_##_n(_lseq)+4))<<32)))
+
+static void asd_print_lseq_cio_reg(struct asd_ha_struct *asd_ha,
+				   u32 lseq_cio_addr, int i)
+{
+	switch (LSEQmCIOREGS[i].width) {
+	case 8:
+		asd_printk("%20s[0x%x]: 0x%02x\n", LSEQmCIOREGS[i].name,
+			   LSEQmCIOREGS[i].offs,
+			   asd_read_reg_byte(asd_ha, lseq_cio_addr +
+					     LSEQmCIOREGS[i].offs));
+
+		break;
+	case 16:
+		asd_printk("%20s[0x%x]: 0x%04x\n", LSEQmCIOREGS[i].name,
+			   LSEQmCIOREGS[i].offs,
+			   asd_read_reg_word(asd_ha, lseq_cio_addr +
+					     LSEQmCIOREGS[i].offs));
+
+		break;
+	case 32:
+		asd_printk("%20s[0x%x]: 0x%08x\n", LSEQmCIOREGS[i].name,
+			   LSEQmCIOREGS[i].offs,
+			   asd_read_reg_dword(asd_ha, lseq_cio_addr +
+					      LSEQmCIOREGS[i].offs));
+		break;
+	}
+}
+
+static void asd_dump_lseq_state(struct asd_ha_struct *asd_ha, int lseq)
+{
+	u32 moffs;
+	int mode;
+
+	asd_printk("LSEQ %d STATE\n", lseq);
+
+	asd_printk("LSEQ%d: ARP2 REGISTERS\n", lseq);
+	PRINT_LREG_32bit(asd_ha, lseq, ARP2CTL);
+	PRINT_LREG_32bit(asd_ha, lseq, ARP2INT);
+	PRINT_LREG_32bit(asd_ha, lseq, ARP2INTEN);
+	PRINT_LREG_8bit(asd_ha, lseq, MODEPTR);
+	PRINT_LREG_8bit(asd_ha, lseq, ALTMODE);
+	PRINT_LREG_8bit(asd_ha, lseq, FLAG);
+	PRINT_LREG_8bit(asd_ha, lseq, ARP2INTCTL);
+	PRINT_LREG_16bit(asd_ha, lseq, STACK);
+	PRINT_LREG_16bit(asd_ha, lseq, PRGMCNT);
+	PRINT_LREG_16bit(asd_ha, lseq, ACCUM);
+	PRINT_LREG_16bit(asd_ha, lseq, SINDEX);
+	PRINT_LREG_16bit(asd_ha, lseq, DINDEX);
+	PRINT_LREG_8bit(asd_ha, lseq, SINDIR);
+	PRINT_LREG_8bit(asd_ha, lseq, DINDIR);
+	PRINT_LREG_8bit(asd_ha, lseq, JUMLDIR);
+	PRINT_LREG_8bit(asd_ha, lseq, ARP2HALTCODE);
+	PRINT_LREG_16bit(asd_ha, lseq, CURRADDR);
+	PRINT_LREG_16bit(asd_ha, lseq, LASTADDR);
+	PRINT_LREG_16bit(asd_ha, lseq, NXTLADDR);
+
+	asd_printk("LSEQ%d: IOP REGISTERS\n", lseq);
+
+	PRINT_LREG_32bit(asd_ha, lseq, MODECTL);
+	PRINT_LREG_32bit(asd_ha, lseq, DBGMODE);
+	PRINT_LREG_32bit(asd_ha, lseq, CONTROL);
+	PRINT_REG_32bit(asd_ha, BISTCTL0, LmBISTCTL0(lseq));
+	PRINT_REG_32bit(asd_ha, BISTCTL1, LmBISTCTL1(lseq));
+
+	asd_printk("LSEQ%d: CIO REGISTERS\n", lseq);
+	asd_printk("Mode common:\n");
+
+	for (mode = 0; mode < 8; mode++) {
+		u32 lseq_cio_addr = LmSEQ_PHY_BASE(mode, lseq);
+		int i;
+
+		for (i = 0; LSEQmCIOREGS[i].name; i++)
+			if (LSEQmCIOREGS[i].mode == MODE_COMMON)
+				asd_print_lseq_cio_reg(asd_ha,lseq_cio_addr,i);
+	}
+
+	asd_printk("Mode unique:\n");
+	for (mode = 0; mode < 8; mode++) {
+		u32 lseq_cio_addr = LmSEQ_PHY_BASE(mode, lseq);
+		int i;
+
+		asd_printk("Mode %d\n", mode);
+		for  (i = 0; LSEQmCIOREGS[i].name; i++) {
+			if (!(LSEQmCIOREGS[i].mode & (1 << mode)))
+				continue;
+			asd_print_lseq_cio_reg(asd_ha, lseq_cio_addr, i);
+		}
+	}
+
+	asd_printk("SCRATCH MEMORY\n");
+
+	asd_printk("LSEQ%d MIP 0 >>>>\n", lseq);
+	PRINT_LMIP_word(asd_ha, lseq, Q_TGTXFR_HEAD);
+	PRINT_LMIP_word(asd_ha, lseq, Q_TGTXFR_TAIL);
+	PRINT_LMIP_byte(asd_ha, lseq, LINK_NUMBER);
+	PRINT_LMIP_byte(asd_ha, lseq, SCRATCH_FLAGS);
+	PRINT_LMIP_qword(asd_ha, lseq, CONNECTION_STATE);
+	PRINT_LMIP_word(asd_ha, lseq, CONCTL);
+	PRINT_LMIP_byte(asd_ha, lseq, CONSTAT);
+	PRINT_LMIP_byte(asd_ha, lseq, CONNECTION_MODES);
+	PRINT_LMIP_word(asd_ha, lseq, REG1_ISR);
+	PRINT_LMIP_word(asd_ha, lseq, REG2_ISR);
+	PRINT_LMIP_word(asd_ha, lseq, REG3_ISR);
+	PRINT_LMIP_qword(asd_ha, lseq,REG0_ISR);
+
+	asd_printk("LSEQ%d MIP 1 >>>>\n", lseq);
+	PRINT_LMIP_word(asd_ha, lseq, EST_NEXUS_SCBPTR0);
+	PRINT_LMIP_word(asd_ha, lseq, EST_NEXUS_SCBPTR1);
+	PRINT_LMIP_word(asd_ha, lseq, EST_NEXUS_SCBPTR2);
+	PRINT_LMIP_word(asd_ha, lseq, EST_NEXUS_SCBPTR3);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_SCB_OPCODE0);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_SCB_OPCODE1);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_SCB_OPCODE2);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_SCB_OPCODE3);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_SCB_HEAD);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_SCB_TAIL);
+	PRINT_LMIP_byte(asd_ha, lseq, EST_NEXUS_BUF_AVAIL);
+	PRINT_LMIP_dword(asd_ha, lseq, TIMEOUT_CONST);
+	PRINT_LMIP_word(asd_ha, lseq, ISR_SAVE_SINDEX);
+	PRINT_LMIP_word(asd_ha, lseq, ISR_SAVE_DINDEX);
+
+	asd_printk("LSEQ%d MIP 2 >>>>\n", lseq);
+	PRINT_LMIP_word(asd_ha, lseq, EMPTY_SCB_PTR0);
+	PRINT_LMIP_word(asd_ha, lseq, EMPTY_SCB_PTR1);
+	PRINT_LMIP_word(asd_ha, lseq, EMPTY_SCB_PTR2);
+	PRINT_LMIP_word(asd_ha, lseq, EMPTY_SCB_PTR3);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_SCB_OPCD0);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_SCB_OPCD1);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_SCB_OPCD2);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_SCB_OPCD3);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_SCB_HEAD);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_SCB_TAIL);
+	PRINT_LMIP_byte(asd_ha, lseq, EMPTY_BUFS_AVAIL);
+
+	asd_printk("LSEQ%d MIP 3 >>>>\n", lseq);
+	PRINT_LMIP_dword(asd_ha, lseq, DEV_PRES_TMR_TOUT_CONST);
+	PRINT_LMIP_dword(asd_ha, lseq, SATA_INTERLOCK_TIMEOUT);
+	PRINT_LMIP_dword(asd_ha, lseq, SRST_ASSERT_TIMEOUT);
+	PRINT_LMIP_dword(asd_ha, lseq, RCV_FIS_TIMEOUT);
+	PRINT_LMIP_dword(asd_ha, lseq, ONE_MILLISEC_TIMEOUT);
+	PRINT_LMIP_dword(asd_ha, lseq, TEN_MS_COMINIT_TIMEOUT);
+	PRINT_LMIP_dword(asd_ha, lseq, SMP_RCV_TIMEOUT);
+
+	for (mode = 0; mode < 3; mode++) {
+		asd_printk("LSEQ%d MDP 0 MODE %d >>>>\n", lseq, mode);
+		moffs = mode * LSEQ_MODE_SCRATCH_SIZE;
+
+		asd_printk(STR_16BIT, "RET_ADDR", 0,
+			   asd_read_reg_word(asd_ha, LmSEQ_RET_ADDR(lseq)
+					     + moffs));
+		asd_printk(STR_16BIT, "REG0_MODE", 2,
+			   asd_read_reg_word(asd_ha, LmSEQ_REG0_MODE(lseq)
+					     + moffs));
+		asd_printk(STR_16BIT, "MODE_FLAGS", 4,
+			   asd_read_reg_word(asd_ha, LmSEQ_MODE_FLAGS(lseq)
+					     + moffs));
+		asd_printk(STR_16BIT, "RET_ADDR2", 0x6,
+			   asd_read_reg_word(asd_ha, LmSEQ_RET_ADDR2(lseq)
+					     + moffs));
+		asd_printk(STR_16BIT, "RET_ADDR1", 0x8,
+			   asd_read_reg_word(asd_ha, LmSEQ_RET_ADDR1(lseq)
+					     + moffs));
+		asd_printk(STR_8BIT, "OPCODE_TO_CSEQ", 0xB,
+			   asd_read_reg_byte(asd_ha, LmSEQ_OPCODE_TO_CSEQ(lseq)
+					     + moffs));
+		asd_printk(STR_16BIT, "DATA_TO_CSEQ", 0xC,
+			   asd_read_reg_word(asd_ha, LmSEQ_DATA_TO_CSEQ(lseq)
+					     + moffs));
+	}
+
+	asd_printk("LSEQ%d MDP 0 MODE 5 >>>>\n", lseq);
+	moffs = LSEQ_MODE5_PAGE0_OFFSET;
+	asd_printk(STR_16BIT, "RET_ADDR", 0,
+		   asd_read_reg_word(asd_ha, LmSEQ_RET_ADDR(lseq) + moffs));
+	asd_printk(STR_16BIT, "REG0_MODE", 2,
+		   asd_read_reg_word(asd_ha, LmSEQ_REG0_MODE(lseq) + moffs));
+	asd_printk(STR_16BIT, "MODE_FLAGS", 4,
+		   asd_read_reg_word(asd_ha, LmSEQ_MODE_FLAGS(lseq) + moffs));
+	asd_printk(STR_16BIT, "RET_ADDR2", 0x6,
+		   asd_read_reg_word(asd_ha, LmSEQ_RET_ADDR2(lseq) + moffs));
+	asd_printk(STR_16BIT, "RET_ADDR1", 0x8,
+		   asd_read_reg_word(asd_ha, LmSEQ_RET_ADDR1(lseq) + moffs));
+	asd_printk(STR_8BIT, "OPCODE_TO_CSEQ", 0xB,
+	   asd_read_reg_byte(asd_ha, LmSEQ_OPCODE_TO_CSEQ(lseq) + moffs));
+	asd_printk(STR_16BIT, "DATA_TO_CSEQ", 0xC,
+	   asd_read_reg_word(asd_ha, LmSEQ_DATA_TO_CSEQ(lseq) + moffs));
+
+	asd_printk("LSEQ%d MDP 0 MODE 0 >>>>\n", lseq);
+	PRINT_LMIP_word(asd_ha, lseq, FIRST_INV_DDB_SITE);
+	PRINT_LMIP_word(asd_ha, lseq, EMPTY_TRANS_CTX);
+	PRINT_LMIP_word(asd_ha, lseq, RESP_LEN);
+	PRINT_LMIP_word(asd_ha, lseq, FIRST_INV_SCB_SITE);
+	PRINT_LMIP_dword(asd_ha, lseq, INTEN_SAVE);
+	PRINT_LMIP_byte(asd_ha, lseq, LINK_RST_FRM_LEN);
+	PRINT_LMIP_byte(asd_ha, lseq, LINK_RST_PROTOCOL);
+	PRINT_LMIP_byte(asd_ha, lseq, RESP_STATUS);
+	PRINT_LMIP_byte(asd_ha, lseq, LAST_LOADED_SGE);
+	PRINT_LMIP_byte(asd_ha, lseq, SAVE_SCBPTR);
+
+	asd_printk("LSEQ%d MDP 0 MODE 1 >>>>\n", lseq);
+	PRINT_LMIP_word(asd_ha, lseq, Q_XMIT_HEAD);
+	PRINT_LMIP_word(asd_ha, lseq, M1_EMPTY_TRANS_CTX);
+	PRINT_LMIP_word(asd_ha, lseq, INI_CONN_TAG);
+	PRINT_LMIP_byte(asd_ha, lseq, FAILED_OPEN_STATUS);
+	PRINT_LMIP_byte(asd_ha, lseq, XMIT_REQUEST_TYPE);
+	PRINT_LMIP_byte(asd_ha, lseq, M1_RESP_STATUS);
+	PRINT_LMIP_byte(asd_ha, lseq, M1_LAST_LOADED_SGE);
+	PRINT_LMIP_word(asd_ha, lseq, M1_SAVE_SCBPTR);
+
+	asd_printk("LSEQ%d MDP 0 MODE 2 >>>>\n", lseq);
+	PRINT_LMIP_word(asd_ha, lseq, PORT_COUNTER);
+	PRINT_LMIP_word(asd_ha, lseq, PM_TABLE_PTR);
+	PRINT_LMIP_word(asd_ha, lseq, SATA_INTERLOCK_TMR_SAVE);
+	PRINT_LMIP_word(asd_ha, lseq, IP_BITL);
+	PRINT_LMIP_word(asd_ha, lseq, COPY_SMP_CONN_TAG);
+	PRINT_LMIP_byte(asd_ha, lseq, P0M2_OFFS1AH);
+
+	asd_printk("LSEQ%d MDP 0 MODE 4/5 >>>>\n", lseq);
+	PRINT_LMIP_byte(asd_ha, lseq, SAVED_OOB_STATUS);
+	PRINT_LMIP_byte(asd_ha, lseq, SAVED_OOB_MODE);
+	PRINT_LMIP_word(asd_ha, lseq, Q_LINK_HEAD);
+	PRINT_LMIP_byte(asd_ha, lseq, LINK_RST_ERR);
+	PRINT_LMIP_byte(asd_ha, lseq, SAVED_OOB_SIGNALS);
+	PRINT_LMIP_byte(asd_ha, lseq, SAS_RESET_MODE);
+	PRINT_LMIP_byte(asd_ha, lseq, LINK_RESET_RETRY_COUNT);
+	PRINT_LMIP_byte(asd_ha, lseq, NUM_LINK_RESET_RETRIES);
+	PRINT_LMIP_word(asd_ha, lseq, OOB_INT_ENABLES);
+	PRINT_LMIP_word(asd_ha, lseq, NOTIFY_TIMER_TIMEOUT);
+	PRINT_LMIP_word(asd_ha, lseq, NOTIFY_TIMER_DOWN_COUNT);
+
+	asd_printk("LSEQ%d MDP 1 MODE 0 >>>>\n", lseq);
+	PRINT_LMIP_qword(asd_ha, lseq, SG_LIST_PTR_ADDR0);
+	PRINT_LMIP_qword(asd_ha, lseq, SG_LIST_PTR_ADDR1);
+
+	asd_printk("LSEQ%d MDP 1 MODE 1 >>>>\n", lseq);
+	PRINT_LMIP_qword(asd_ha, lseq, M1_SG_LIST_PTR_ADDR0);
+	PRINT_LMIP_qword(asd_ha, lseq, M1_SG_LIST_PTR_ADDR1);
+
+	asd_printk("LSEQ%d MDP 1 MODE 2 >>>>\n", lseq);
+	PRINT_LMIP_dword(asd_ha, lseq, INVALID_DWORD_COUNT);
+	PRINT_LMIP_dword(asd_ha, lseq, DISPARITY_ERROR_COUNT);
+	PRINT_LMIP_dword(asd_ha, lseq, LOSS_OF_SYNC_COUNT);
+
+	asd_printk("LSEQ%d MDP 1 MODE 4/5 >>>>\n", lseq);
+	PRINT_LMIP_dword(asd_ha, lseq, FRAME_TYPE_MASK);
+	PRINT_LMIP_dword(asd_ha, lseq, HASHED_SRC_ADDR_MASK_PRINT);
+	PRINT_LMIP_byte(asd_ha, lseq, NUM_FILL_BYTES_MASK);
+	PRINT_LMIP_word(asd_ha, lseq, TAG_MASK);
+	PRINT_LMIP_word(asd_ha, lseq, TARGET_PORT_XFER_TAG);
+	PRINT_LMIP_dword(asd_ha, lseq, DATA_OFFSET);
+
+	asd_printk("LSEQ%d MDP 2 MODE 0 >>>>\n", lseq);
+	PRINT_LMIP_dword(asd_ha, lseq, SMP_RCV_TIMER_TERM_TS);
+	PRINT_LMIP_byte(asd_ha, lseq, DEVICE_BITS);
+	PRINT_LMIP_word(asd_ha, lseq, SDB_DDB);
+	PRINT_LMIP_word(asd_ha, lseq, SDB_NUM_TAGS);
+	PRINT_LMIP_word(asd_ha, lseq, SDB_CURR_TAG);
+
+	asd_printk("LSEQ%d MDP 2 MODE 1 >>>>\n", lseq);
+	PRINT_LMIP_qword(asd_ha, lseq, TX_ID_ADDR_FRAME);
+	PRINT_LMIP_dword(asd_ha, lseq, OPEN_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, SRST_AS_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, LAST_LOADED_SG_EL);
+
+	asd_printk("LSEQ%d MDP 2 MODE 2 >>>>\n", lseq);
+	PRINT_LMIP_dword(asd_ha, lseq, CLOSE_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, BREAK_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, DWS_RESET_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, SATA_INTERLOCK_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, MCTL_TIMER_TERM_TS);
+
+	asd_printk("LSEQ%d MDP 2 MODE 4/5 >>>>\n", lseq);
+	PRINT_LMIP_dword(asd_ha, lseq, COMINIT_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, RCV_ID_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, RCV_FIS_TIMER_TERM_TS);
+	PRINT_LMIP_dword(asd_ha, lseq, DEV_PRES_TIMER_TERM_TS);
+}
+
+/**
+ * asd_dump_ddb_site -- dump a CSEQ DDB site
+ * @asd_ha: pointer to host adapter structure
+ * @site_no: site number of interest
+ */
+void asd_dump_target_ddb(struct asd_ha_struct *asd_ha, u16 site_no)
+{
+	if (site_no >= asd_ha->hw_prof.max_ddbs)
+		return;
+
+#define DDB_FIELDB(__name)                                        \
+	asd_ddbsite_read_byte(asd_ha, site_no,                    \
+			      offsetof(struct asd_ddb_ssp_smp_target_port, __name))
+#define DDB2_FIELDB(__name)                                       \
+	asd_ddbsite_read_byte(asd_ha, site_no,                    \
+			      offsetof(struct asd_ddb_stp_sata_target_port, __name))
+#define DDB_FIELDW(__name)                                        \
+	asd_ddbsite_read_word(asd_ha, site_no,                    \
+			      offsetof(struct asd_ddb_ssp_smp_target_port, __name))
+
+#define DDB_FIELDD(__name)                                         \
+	asd_ddbsite_read_dword(asd_ha, site_no,                    \
+			       offsetof(struct asd_ddb_ssp_smp_target_port, __name))
+
+	asd_printk("DDB: 0x%02x\n", site_no);
+	asd_printk("conn_type: 0x%02x\n", DDB_FIELDB(conn_type));
+	asd_printk("conn_rate: 0x%02x\n", DDB_FIELDB(conn_rate));
+	asd_printk("init_conn_tag: 0x%04x\n", be16_to_cpu(DDB_FIELDW(init_conn_tag)));
+	asd_printk("send_queue_head: 0x%04x\n", be16_to_cpu(DDB_FIELDW(send_queue_head)));
+	asd_printk("sq_suspended: 0x%02x\n", DDB_FIELDB(sq_suspended));
+	asd_printk("DDB Type: 0x%02x\n", DDB_FIELDB(ddb_type));
+	asd_printk("AWT Default: 0x%04x\n", DDB_FIELDW(awt_def));
+	asd_printk("compat_features: 0x%02x\n", DDB_FIELDB(compat_features));
+	asd_printk("Pathway Blocked Count: 0x%02x\n",
+		   DDB_FIELDB(pathway_blocked_count));
+	asd_printk("arb_wait_time: 0x%04x\n", DDB_FIELDW(arb_wait_time));
+	asd_printk("more_compat_features: 0x%08x\n",
+		   DDB_FIELDD(more_compat_features));
+	asd_printk("Conn Mask: 0x%02x\n", DDB_FIELDB(conn_mask));
+	asd_printk("flags: 0x%02x\n", DDB_FIELDB(flags));
+	asd_printk("flags2: 0x%02x\n", DDB2_FIELDB(flags2));
+	asd_printk("ExecQ Tail: 0x%04x\n",DDB_FIELDW(exec_queue_tail));
+	asd_printk("SendQ Tail: 0x%04x\n",DDB_FIELDW(send_queue_tail));
+	asd_printk("Active Task Count: 0x%04x\n",
+		   DDB_FIELDW(active_task_count));
+	asd_printk("ITNL Reason: 0x%02x\n", DDB_FIELDB(itnl_reason));
+	asd_printk("ITNL Timeout Const: 0x%04x\n", DDB_FIELDW(itnl_timeout));
+	asd_printk("ITNL timestamp: 0x%08x\n", DDB_FIELDD(itnl_timestamp));
+}
+
+void asd_dump_ddb_0(struct asd_ha_struct *asd_ha)
+{
+#define DDB0_FIELDB(__name)                                  \
+	asd_ddbsite_read_byte(asd_ha, 0,                     \
+			      offsetof(struct asd_ddb_seq_shared, __name))
+#define DDB0_FIELDW(__name)                                  \
+	asd_ddbsite_read_word(asd_ha, 0,                     \
+			      offsetof(struct asd_ddb_seq_shared, __name))
+
+#define DDB0_FIELDD(__name)                                  \
+	asd_ddbsite_read_dword(asd_ha,0 ,                    \
+			       offsetof(struct asd_ddb_seq_shared, __name))
+
+#define DDB0_FIELDA(__name, _o)                              \
+	asd_ddbsite_read_byte(asd_ha, 0,                     \
+			      offsetof(struct asd_ddb_seq_shared, __name)+_o)
+
+
+	asd_printk("DDB: 0\n");
+	asd_printk("q_free_ddb_head:%04x\n", DDB0_FIELDW(q_free_ddb_head));
+	asd_printk("q_free_ddb_tail:%04x\n", DDB0_FIELDW(q_free_ddb_tail));
+	asd_printk("q_free_ddb_cnt:%04x\n",  DDB0_FIELDW(q_free_ddb_cnt));
+	asd_printk("q_used_ddb_head:%04x\n", DDB0_FIELDW(q_used_ddb_head));
+	asd_printk("q_used_ddb_tail:%04x\n", DDB0_FIELDW(q_used_ddb_tail));
+	asd_printk("shared_mem_lock:%04x\n", DDB0_FIELDW(shared_mem_lock));
+	asd_printk("smp_conn_tag:%04x\n",    DDB0_FIELDW(smp_conn_tag));
+	asd_printk("est_nexus_buf_cnt:%04x\n", DDB0_FIELDW(est_nexus_buf_cnt));
+	asd_printk("est_nexus_buf_thresh:%04x\n",
+		   DDB0_FIELDW(est_nexus_buf_thresh));
+	asd_printk("conn_not_active:%02x\n", DDB0_FIELDB(conn_not_active));
+	asd_printk("phy_is_up:%02x\n",       DDB0_FIELDB(phy_is_up));
+	asd_printk("port_map_by_links:%02x %02x %02x %02x "
+		   "%02x %02x %02x %02x\n",
+		   DDB0_FIELDA(port_map_by_links, 0),
+		   DDB0_FIELDA(port_map_by_links, 1),
+		   DDB0_FIELDA(port_map_by_links, 2),
+		   DDB0_FIELDA(port_map_by_links, 3),
+		   DDB0_FIELDA(port_map_by_links, 4),
+		   DDB0_FIELDA(port_map_by_links, 5),
+		   DDB0_FIELDA(port_map_by_links, 6),
+		   DDB0_FIELDA(port_map_by_links, 7));
+}
+
+static void asd_dump_scb_site(struct asd_ha_struct *asd_ha, u16 site_no)
+{
+
+#define SCB_FIELDB(__name)                                                 \
+	asd_scbsite_read_byte(asd_ha, site_no, sizeof(struct scb_header)   \
+			      + offsetof(struct initiate_ssp_task, __name))
+#define SCB_FIELDW(__name)                                                 \
+	asd_scbsite_read_word(asd_ha, site_no, sizeof(struct scb_header)   \
+			      + offsetof(struct initiate_ssp_task, __name))
+#define SCB_FIELDD(__name)                                                 \
+	asd_scbsite_read_dword(asd_ha, site_no, sizeof(struct scb_header)  \
+			       + offsetof(struct initiate_ssp_task, __name))
+
+	asd_printk("Total Xfer Len: 0x%08x.\n", SCB_FIELDD(total_xfer_len));
+	asd_printk("Frame Type: 0x%02x.\n", SCB_FIELDB(ssp_frame.frame_type));
+	asd_printk("Tag: 0x%04x.\n", SCB_FIELDW(ssp_frame.tag));
+	asd_printk("Target Port Xfer Tag: 0x%04x.\n",
+		   SCB_FIELDW(ssp_frame.tptt));
+	asd_printk("Data Offset: 0x%08x.\n", SCB_FIELDW(ssp_frame.data_offs));
+	asd_printk("Retry Count: 0x%02x.\n", SCB_FIELDB(retry_count));
+}
+
+/**
+ * asd_dump_scb_sites -- dump currently used CSEQ SCB sites
+ * @asd_ha: pointer to host adapter struct
+ */
+void asd_dump_scb_sites(struct asd_ha_struct *asd_ha)
+{
+	u16	site_no;
+
+	for (site_no = 0; site_no < asd_ha->hw_prof.max_scbs; site_no++) {
+		u8 opcode;
+
+		if (!SCB_SITE_VALID(site_no))
+			continue;
+
+		/* We are only interested in SCB sites currently used.
+		 */
+		opcode = asd_scbsite_read_byte(asd_ha, site_no,
+					       offsetof(struct scb_header,
+							opcode));
+		if (opcode == 0xFF)
+			continue;
+
+		asd_printk("\nSCB: 0x%x\n", site_no);
+		asd_dump_scb_site(asd_ha, site_no);
+	}
+}
+
+/**
+ * ads_dump_seq_state -- dump CSEQ and LSEQ states
+ * @asd_ha: pointer to host adapter structure
+ * @lseq_mask: mask of LSEQs of interest
+ */
+void asd_dump_seq_state(struct asd_ha_struct *asd_ha, u8 lseq_mask)
+{
+	int lseq;
+
+	asd_dump_cseq_state(asd_ha);
+
+	if (lseq_mask != 0)
+		for_each_sequencer(lseq_mask, lseq_mask, lseq)
+			asd_dump_lseq_state(asd_ha, lseq);
+}
+
+void asd_dump_frame_rcvd(struct asd_phy *phy,
+			 struct done_list_struct *dl)
+{
+	unsigned long flags;
+	int i;
+
+	switch ((dl->status_block[1] & 0x70) >> 3) {
+	case SAS_PROTO_STP:
+		ASD_DPRINTK("STP proto device-to-host FIS:\n");
+		break;
+	default:
+	case SAS_PROTO_SSP:
+		ASD_DPRINTK("SAS proto IDENTIFY:\n");
+		break;
+	}
+	spin_lock_irqsave(&phy->sas_phy.frame_rcvd_lock, flags);
+	for (i = 0; i < phy->sas_phy.frame_rcvd_size; i+=4)
+		ASD_DPRINTK("%02x: %02x %02x %02x %02x\n",
+			    i,
+			    phy->frame_rcvd[i],
+			    phy->frame_rcvd[i+1],
+			    phy->frame_rcvd[i+2],
+			    phy->frame_rcvd[i+3]);
+	spin_unlock_irqrestore(&phy->sas_phy.frame_rcvd_lock, flags);
+}
+
+static inline void asd_dump_scb(struct asd_ascb *ascb, int ind)
+{
+	asd_printk("scb%d: vaddr: 0x%p, dma_handle: 0x%llx, next: 0x%llx, "
+		   "index:%d, opcode:0x%02x\n",
+		   ind, ascb->dma_scb.vaddr,
+		   (unsigned long long)ascb->dma_scb.dma_handle,
+		   (unsigned long long)
+		   le64_to_cpu(ascb->scb->header.next_scb),
+		   le16_to_cpu(ascb->scb->header.index),
+		   ascb->scb->header.opcode);
+}
+
+void asd_dump_scb_list(struct asd_ascb *ascb, int num)
+{
+	int i = 0;
+
+	asd_printk("dumping %d scbs:\n", num);
+
+	asd_dump_scb(ascb, i++);
+	--num;
+
+	if (num > 0 && !list_empty(&ascb->list)) {
+		struct list_head *el;
+
+		list_for_each(el, &ascb->list) {
+			struct asd_ascb *s = list_entry(el, struct asd_ascb,
+							list);
+			asd_dump_scb(s, i++);
+			if (--num <= 0)
+				break;
+		}
+	}
+}
+
+#endif /* ASD_DEBUG */
diff --git a/drivers/scsi/aic94xx/aic94xx_dump.h b/drivers/scsi/aic94xx/aic94xx_dump.h
new file mode 100644
index 000000000000..0c388e7da6bb
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_dump.h
@@ -0,0 +1,52 @@
+/*
+ * Aic94xx SAS/SATA driver dump header file.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#ifndef _AIC94XX_DUMP_H_
+#define _AIC94XX_DUMP_H_
+
+#ifdef ASD_DEBUG
+
+void asd_dump_ddb_0(struct asd_ha_struct *asd_ha);
+void asd_dump_target_ddb(struct asd_ha_struct *asd_ha, u16 site_no);
+void asd_dump_scb_sites(struct asd_ha_struct *asd_ha);
+void asd_dump_seq_state(struct asd_ha_struct *asd_ha, u8 lseq_mask);
+void asd_dump_frame_rcvd(struct asd_phy *phy,
+			 struct done_list_struct *dl);
+void asd_dump_scb_list(struct asd_ascb *ascb, int num);
+#else /* ASD_DEBUG */
+
+static inline void asd_dump_ddb_0(struct asd_ha_struct *asd_ha) { }
+static inline void asd_dump_target_ddb(struct asd_ha_struct *asd_ha,
+				     u16 site_no) { }
+static inline void asd_dump_scb_sites(struct asd_ha_struct *asd_ha) { }
+static inline void asd_dump_seq_state(struct asd_ha_struct *asd_ha,
+				      u8 lseq_mask) { }
+static inline void asd_dump_frame_rcvd(struct asd_phy *phy,
+				       struct done_list_struct *dl) { }
+static inline void asd_dump_scb_list(struct asd_ascb *ascb, int num) { }
+#endif /* ASD_DEBUG */
+
+#endif /* _AIC94XX_DUMP_H_ */
diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c
new file mode 100644
index 000000000000..075cea85b56b
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_hwi.c
@@ -0,0 +1,1376 @@
+/*
+ * Aic94xx SAS/SATA driver hardware interface.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+
+#include "aic94xx.h"
+#include "aic94xx_reg.h"
+#include "aic94xx_hwi.h"
+#include "aic94xx_seq.h"
+#include "aic94xx_dump.h"
+
+u32 MBAR0_SWB_SIZE;
+
+/* ---------- Initialization ---------- */
+
+static void asd_get_user_sas_addr(struct asd_ha_struct *asd_ha)
+{
+	extern char sas_addr_str[];
+	/* If the user has specified a WWN it overrides other settings
+	 */
+	if (sas_addr_str[0] != '\0')
+		asd_destringify_sas_addr(asd_ha->hw_prof.sas_addr,
+					 sas_addr_str);
+	else if (asd_ha->hw_prof.sas_addr[0] != 0)
+		asd_stringify_sas_addr(sas_addr_str, asd_ha->hw_prof.sas_addr);
+}
+
+static void asd_propagate_sas_addr(struct asd_ha_struct *asd_ha)
+{
+	int i;
+
+	for (i = 0; i < ASD_MAX_PHYS; i++) {
+		if (asd_ha->hw_prof.phy_desc[i].sas_addr[0] == 0)
+			continue;
+		/* Set a phy's address only if it has none.
+		 */
+		ASD_DPRINTK("setting phy%d addr to %llx\n", i,
+			    SAS_ADDR(asd_ha->hw_prof.sas_addr));
+		memcpy(asd_ha->hw_prof.phy_desc[i].sas_addr,
+		       asd_ha->hw_prof.sas_addr, SAS_ADDR_SIZE);
+	}
+}
+
+/* ---------- PHY initialization ---------- */
+
+static void asd_init_phy_identify(struct asd_phy *phy)
+{
+	phy->identify_frame = phy->id_frm_tok->vaddr;
+
+	memset(phy->identify_frame, 0, sizeof(*phy->identify_frame));
+
+	phy->identify_frame->dev_type = SAS_END_DEV;
+	if (phy->sas_phy.role & PHY_ROLE_INITIATOR)
+		phy->identify_frame->initiator_bits = phy->sas_phy.iproto;
+	if (phy->sas_phy.role & PHY_ROLE_TARGET)
+		phy->identify_frame->target_bits = phy->sas_phy.tproto;
+	memcpy(phy->identify_frame->sas_addr, phy->phy_desc->sas_addr,
+	       SAS_ADDR_SIZE);
+	phy->identify_frame->phy_id = phy->sas_phy.id;
+}
+
+static int asd_init_phy(struct asd_phy *phy)
+{
+	struct asd_ha_struct *asd_ha = phy->sas_phy.ha->lldd_ha;
+	struct asd_sas_phy *sas_phy = &phy->sas_phy;
+
+	sas_phy->enabled = 1;
+	sas_phy->class = SAS;
+	sas_phy->iproto = SAS_PROTO_ALL;
+	sas_phy->tproto = 0;
+	sas_phy->type = PHY_TYPE_PHYSICAL;
+	sas_phy->role = PHY_ROLE_INITIATOR;
+	sas_phy->oob_mode = OOB_NOT_CONNECTED;
+	sas_phy->linkrate = PHY_LINKRATE_NONE;
+
+	phy->id_frm_tok = asd_alloc_coherent(asd_ha,
+					     sizeof(*phy->identify_frame),
+					     GFP_KERNEL);
+	if (!phy->id_frm_tok) {
+		asd_printk("no mem for IDENTIFY for phy%d\n", sas_phy->id);
+		return -ENOMEM;
+	} else
+		asd_init_phy_identify(phy);
+
+	memset(phy->frame_rcvd, 0, sizeof(phy->frame_rcvd));
+
+	return 0;
+}
+
+static int asd_init_phys(struct asd_ha_struct *asd_ha)
+{
+	u8 i;
+	u8 phy_mask = asd_ha->hw_prof.enabled_phys;
+
+	for (i = 0; i < ASD_MAX_PHYS; i++) {
+		struct asd_phy *phy = &asd_ha->phys[i];
+
+		phy->phy_desc = &asd_ha->hw_prof.phy_desc[i];
+
+		phy->sas_phy.enabled = 0;
+		phy->sas_phy.id = i;
+		phy->sas_phy.sas_addr = &phy->phy_desc->sas_addr[0];
+		phy->sas_phy.frame_rcvd = &phy->frame_rcvd[0];
+		phy->sas_phy.ha = &asd_ha->sas_ha;
+		phy->sas_phy.lldd_phy = phy;
+	}
+
+	/* Now enable and initialize only the enabled phys. */
+	for_each_phy(phy_mask, phy_mask, i) {
+		int err = asd_init_phy(&asd_ha->phys[i]);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/* ---------- Sliding windows ---------- */
+
+static int asd_init_sw(struct asd_ha_struct *asd_ha)
+{
+	struct pci_dev *pcidev = asd_ha->pcidev;
+	int err;
+	u32 v;
+
+	/* Unlock MBARs */
+	err = pci_read_config_dword(pcidev, PCI_CONF_MBAR_KEY, &v);
+	if (err) {
+		asd_printk("couldn't access conf. space of %s\n",
+			   pci_name(pcidev));
+		goto Err;
+	}
+	if (v)
+		err = pci_write_config_dword(pcidev, PCI_CONF_MBAR_KEY, v);
+	if (err) {
+		asd_printk("couldn't write to MBAR_KEY of %s\n",
+			   pci_name(pcidev));
+		goto Err;
+	}
+
+	/* Set sliding windows A, B and C to point to proper internal
+	 * memory regions.
+	 */
+	pci_write_config_dword(pcidev, PCI_CONF_MBAR0_SWA, REG_BASE_ADDR);
+	pci_write_config_dword(pcidev, PCI_CONF_MBAR0_SWB,
+			       REG_BASE_ADDR_CSEQCIO);
+	pci_write_config_dword(pcidev, PCI_CONF_MBAR0_SWC, REG_BASE_ADDR_EXSI);
+	asd_ha->io_handle[0].swa_base = REG_BASE_ADDR;
+	asd_ha->io_handle[0].swb_base = REG_BASE_ADDR_CSEQCIO;
+	asd_ha->io_handle[0].swc_base = REG_BASE_ADDR_EXSI;
+	MBAR0_SWB_SIZE = asd_ha->io_handle[0].len - 0x80;
+	if (!asd_ha->iospace) {
+		/* MBAR1 will point to OCM (On Chip Memory) */
+		pci_write_config_dword(pcidev, PCI_CONF_MBAR1, OCM_BASE_ADDR);
+		asd_ha->io_handle[1].swa_base = OCM_BASE_ADDR;
+	}
+	spin_lock_init(&asd_ha->iolock);
+Err:
+	return err;
+}
+
+/* ---------- SCB initialization ---------- */
+
+/**
+ * asd_init_scbs - manually allocate the first SCB.
+ * @asd_ha: pointer to host adapter structure
+ *
+ * This allocates the very first SCB which would be sent to the
+ * sequencer for execution.  Its bus address is written to
+ * CSEQ_Q_NEW_POINTER, mode page 2, mode 8.  Since the bus address of
+ * the _next_ scb to be DMA-ed to the host adapter is read from the last
+ * SCB DMA-ed to the host adapter, we have to always stay one step
+ * ahead of the sequencer and keep one SCB already allocated.
+ */
+static int asd_init_scbs(struct asd_ha_struct *asd_ha)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int bitmap_bytes;
+
+	/* allocate the index array and bitmap */
+	asd_ha->seq.tc_index_bitmap_bits = asd_ha->hw_prof.max_scbs;
+	asd_ha->seq.tc_index_array = kzalloc(asd_ha->seq.tc_index_bitmap_bits*
+					     sizeof(void *), GFP_KERNEL);
+	if (!asd_ha->seq.tc_index_array)
+		return -ENOMEM;
+
+	bitmap_bytes = (asd_ha->seq.tc_index_bitmap_bits+7)/8;
+	bitmap_bytes = BITS_TO_LONGS(bitmap_bytes*8)*sizeof(unsigned long);
+	asd_ha->seq.tc_index_bitmap = kzalloc(bitmap_bytes, GFP_KERNEL);
+	if (!asd_ha->seq.tc_index_bitmap)
+		return -ENOMEM;
+
+	spin_lock_init(&seq->tc_index_lock);
+
+	seq->next_scb.size = sizeof(struct scb);
+	seq->next_scb.vaddr = dma_pool_alloc(asd_ha->scb_pool, GFP_KERNEL,
+					     &seq->next_scb.dma_handle);
+	if (!seq->next_scb.vaddr) {
+		kfree(asd_ha->seq.tc_index_bitmap);
+		kfree(asd_ha->seq.tc_index_array);
+		asd_ha->seq.tc_index_bitmap = NULL;
+		asd_ha->seq.tc_index_array = NULL;
+		return -ENOMEM;
+	}
+
+	seq->pending = 0;
+	spin_lock_init(&seq->pend_q_lock);
+	INIT_LIST_HEAD(&seq->pend_q);
+
+	return 0;
+}
+
+static inline void asd_get_max_scb_ddb(struct asd_ha_struct *asd_ha)
+{
+	asd_ha->hw_prof.max_scbs = asd_get_cmdctx_size(asd_ha)/ASD_SCB_SIZE;
+	asd_ha->hw_prof.max_ddbs = asd_get_devctx_size(asd_ha)/ASD_DDB_SIZE;
+	ASD_DPRINTK("max_scbs:%d, max_ddbs:%d\n",
+		    asd_ha->hw_prof.max_scbs,
+		    asd_ha->hw_prof.max_ddbs);
+}
+
+/* ---------- Done List initialization ---------- */
+
+static void asd_dl_tasklet_handler(unsigned long);
+
+static int asd_init_dl(struct asd_ha_struct *asd_ha)
+{
+	asd_ha->seq.actual_dl
+		= asd_alloc_coherent(asd_ha,
+			     ASD_DL_SIZE * sizeof(struct done_list_struct),
+				     GFP_KERNEL);
+	if (!asd_ha->seq.actual_dl)
+		return -ENOMEM;
+	asd_ha->seq.dl = asd_ha->seq.actual_dl->vaddr;
+	asd_ha->seq.dl_toggle = ASD_DEF_DL_TOGGLE;
+	asd_ha->seq.dl_next = 0;
+	tasklet_init(&asd_ha->seq.dl_tasklet, asd_dl_tasklet_handler,
+		     (unsigned long) asd_ha);
+
+	return 0;
+}
+
+/* ---------- EDB and ESCB init ---------- */
+
+static int asd_alloc_edbs(struct asd_ha_struct *asd_ha, unsigned int gfp_flags)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int i;
+
+	seq->edb_arr = kmalloc(seq->num_edbs*sizeof(*seq->edb_arr), gfp_flags);
+	if (!seq->edb_arr)
+		return -ENOMEM;
+
+	for (i = 0; i < seq->num_edbs; i++) {
+		seq->edb_arr[i] = asd_alloc_coherent(asd_ha, ASD_EDB_SIZE,
+						     gfp_flags);
+		if (!seq->edb_arr[i])
+			goto Err_unroll;
+		memset(seq->edb_arr[i]->vaddr, 0, ASD_EDB_SIZE);
+	}
+
+	ASD_DPRINTK("num_edbs:%d\n", seq->num_edbs);
+
+	return 0;
+
+Err_unroll:
+	for (i-- ; i >= 0; i--)
+		asd_free_coherent(asd_ha, seq->edb_arr[i]);
+	kfree(seq->edb_arr);
+	seq->edb_arr = NULL;
+
+	return -ENOMEM;
+}
+
+static int asd_alloc_escbs(struct asd_ha_struct *asd_ha,
+			   unsigned int gfp_flags)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	struct asd_ascb *escb;
+	int i, escbs;
+
+	seq->escb_arr = kmalloc(seq->num_escbs*sizeof(*seq->escb_arr),
+				gfp_flags);
+	if (!seq->escb_arr)
+		return -ENOMEM;
+
+	escbs = seq->num_escbs;
+	escb = asd_ascb_alloc_list(asd_ha, &escbs, gfp_flags);
+	if (!escb) {
+		asd_printk("couldn't allocate list of escbs\n");
+		goto Err;
+	}
+	seq->num_escbs -= escbs;  /* subtract what was not allocated */
+	ASD_DPRINTK("num_escbs:%d\n", seq->num_escbs);
+
+	for (i = 0; i < seq->num_escbs; i++, escb = list_entry(escb->list.next,
+							       struct asd_ascb,
+							       list)) {
+		seq->escb_arr[i] = escb;
+		escb->scb->header.opcode = EMPTY_SCB;
+	}
+
+	return 0;
+Err:
+	kfree(seq->escb_arr);
+	seq->escb_arr = NULL;
+	return -ENOMEM;
+
+}
+
+static void asd_assign_edbs2escbs(struct asd_ha_struct *asd_ha)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int i, k, z = 0;
+
+	for (i = 0; i < seq->num_escbs; i++) {
+		struct asd_ascb *ascb = seq->escb_arr[i];
+		struct empty_scb *escb = &ascb->scb->escb;
+
+		ascb->edb_index = z;
+
+		escb->num_valid = ASD_EDBS_PER_SCB;
+
+		for (k = 0; k < ASD_EDBS_PER_SCB; k++) {
+			struct sg_el *eb = &escb->eb[k];
+			struct asd_dma_tok *edb = seq->edb_arr[z++];
+
+			memset(eb, 0, sizeof(*eb));
+			eb->bus_addr = cpu_to_le64(((u64) edb->dma_handle));
+			eb->size = cpu_to_le32(((u32) edb->size));
+		}
+	}
+}
+
+/**
+ * asd_init_escbs -- allocate and initialize empty scbs
+ * @asd_ha: pointer to host adapter structure
+ *
+ * An empty SCB has sg_elements of ASD_EDBS_PER_SCB (7) buffers.
+ * They transport sense data, etc.
+ */
+static int asd_init_escbs(struct asd_ha_struct *asd_ha)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int err = 0;
+
+	/* Allocate two empty data buffers (edb) per sequencer. */
+	int edbs = 2*(1+asd_ha->hw_prof.num_phys);
+
+	seq->num_escbs = (edbs+ASD_EDBS_PER_SCB-1)/ASD_EDBS_PER_SCB;
+	seq->num_edbs = seq->num_escbs * ASD_EDBS_PER_SCB;
+
+	err = asd_alloc_edbs(asd_ha, GFP_KERNEL);
+	if (err) {
+		asd_printk("couldn't allocate edbs\n");
+		return err;
+	}
+
+	err = asd_alloc_escbs(asd_ha, GFP_KERNEL);
+	if (err) {
+		asd_printk("couldn't allocate escbs\n");
+		return err;
+	}
+
+	asd_assign_edbs2escbs(asd_ha);
+	/* In order to insure that normal SCBs do not overfill sequencer
+	 * memory and leave no space for escbs (halting condition),
+	 * we increment pending here by the number of escbs.  However,
+	 * escbs are never pending.
+	 */
+	seq->pending   = seq->num_escbs;
+	seq->can_queue = 1 + (asd_ha->hw_prof.max_scbs - seq->pending)/2;
+
+	return 0;
+}
+
+/* ---------- HW initialization ---------- */
+
+/**
+ * asd_chip_hardrst -- hard reset the chip
+ * @asd_ha: pointer to host adapter structure
+ *
+ * This takes 16 cycles and is synchronous to CFCLK, which runs
+ * at 200 MHz, so this should take at most 80 nanoseconds.
+ */
+int asd_chip_hardrst(struct asd_ha_struct *asd_ha)
+{
+	int i;
+	int count = 100;
+	u32 reg;
+
+	for (i = 0 ; i < 4 ; i++) {
+		asd_write_reg_dword(asd_ha, COMBIST, HARDRST);
+	}
+
+	do {
+		udelay(1);
+		reg = asd_read_reg_dword(asd_ha, CHIMINT);
+		if (reg & HARDRSTDET) {
+			asd_write_reg_dword(asd_ha, CHIMINT,
+					    HARDRSTDET|PORRSTDET);
+			return 0;
+		}
+	} while (--count > 0);
+
+	return -ENODEV;
+}
+
+/**
+ * asd_init_chip -- initialize the chip
+ * @asd_ha: pointer to host adapter structure
+ *
+ * Hard resets the chip, disables HA interrupts, downloads the sequnecer
+ * microcode and starts the sequencers.  The caller has to explicitly
+ * enable HA interrupts with asd_enable_ints(asd_ha).
+ */
+static int asd_init_chip(struct asd_ha_struct *asd_ha)
+{
+	int err;
+
+	err = asd_chip_hardrst(asd_ha);
+	if (err) {
+		asd_printk("couldn't hard reset %s\n",
+			    pci_name(asd_ha->pcidev));
+		goto out;
+	}
+
+	asd_disable_ints(asd_ha);
+
+	err = asd_init_seqs(asd_ha);
+	if (err) {
+		asd_printk("couldn't init seqs for %s\n",
+			   pci_name(asd_ha->pcidev));
+		goto out;
+	}
+
+	err = asd_start_seqs(asd_ha);
+	if (err) {
+		asd_printk("coudln't start seqs for %s\n",
+			   pci_name(asd_ha->pcidev));
+		goto out;
+	}
+out:
+	return err;
+}
+
+#define MAX_DEVS ((OCM_MAX_SIZE) / (ASD_DDB_SIZE))
+
+static int max_devs = 0;
+module_param_named(max_devs, max_devs, int, S_IRUGO);
+MODULE_PARM_DESC(max_devs, "\n"
+	"\tMaximum number of SAS devices to support (not LUs).\n"
+	"\tDefault: 2176, Maximum: 65663.\n");
+
+static int max_cmnds = 0;
+module_param_named(max_cmnds, max_cmnds, int, S_IRUGO);
+MODULE_PARM_DESC(max_cmnds, "\n"
+	"\tMaximum number of commands queuable.\n"
+	"\tDefault: 512, Maximum: 66047.\n");
+
+static void asd_extend_devctx_ocm(struct asd_ha_struct *asd_ha)
+{
+	unsigned long dma_addr = OCM_BASE_ADDR;
+	u32 d;
+
+	dma_addr -= asd_ha->hw_prof.max_ddbs * ASD_DDB_SIZE;
+	asd_write_reg_addr(asd_ha, DEVCTXBASE, (dma_addr_t) dma_addr);
+	d = asd_read_reg_dword(asd_ha, CTXDOMAIN);
+	d |= 4;
+	asd_write_reg_dword(asd_ha, CTXDOMAIN, d);
+	asd_ha->hw_prof.max_ddbs += MAX_DEVS;
+}
+
+static int asd_extend_devctx(struct asd_ha_struct *asd_ha)
+{
+	dma_addr_t dma_handle;
+	unsigned long dma_addr;
+	u32 d;
+	int size;
+
+	asd_extend_devctx_ocm(asd_ha);
+
+	asd_ha->hw_prof.ddb_ext = NULL;
+	if (max_devs <= asd_ha->hw_prof.max_ddbs || max_devs > 0xFFFF) {
+		max_devs = asd_ha->hw_prof.max_ddbs;
+		return 0;
+	}
+
+	size = (max_devs - asd_ha->hw_prof.max_ddbs + 1) * ASD_DDB_SIZE;
+
+	asd_ha->hw_prof.ddb_ext = asd_alloc_coherent(asd_ha, size, GFP_KERNEL);
+	if (!asd_ha->hw_prof.ddb_ext) {
+		asd_printk("couldn't allocate memory for %d devices\n",
+			   max_devs);
+		max_devs = asd_ha->hw_prof.max_ddbs;
+		return -ENOMEM;
+	}
+	dma_handle = asd_ha->hw_prof.ddb_ext->dma_handle;
+	dma_addr = ALIGN((unsigned long) dma_handle, ASD_DDB_SIZE);
+	dma_addr -= asd_ha->hw_prof.max_ddbs * ASD_DDB_SIZE;
+	dma_handle = (dma_addr_t) dma_addr;
+	asd_write_reg_addr(asd_ha, DEVCTXBASE, dma_handle);
+	d = asd_read_reg_dword(asd_ha, CTXDOMAIN);
+	d &= ~4;
+	asd_write_reg_dword(asd_ha, CTXDOMAIN, d);
+
+	asd_ha->hw_prof.max_ddbs = max_devs;
+
+	return 0;
+}
+
+static int asd_extend_cmdctx(struct asd_ha_struct *asd_ha)
+{
+	dma_addr_t dma_handle;
+	unsigned long dma_addr;
+	u32 d;
+	int size;
+
+	asd_ha->hw_prof.scb_ext = NULL;
+	if (max_cmnds <= asd_ha->hw_prof.max_scbs || max_cmnds > 0xFFFF) {
+		max_cmnds = asd_ha->hw_prof.max_scbs;
+		return 0;
+	}
+
+	size = (max_cmnds - asd_ha->hw_prof.max_scbs + 1) * ASD_SCB_SIZE;
+
+	asd_ha->hw_prof.scb_ext = asd_alloc_coherent(asd_ha, size, GFP_KERNEL);
+	if (!asd_ha->hw_prof.scb_ext) {
+		asd_printk("couldn't allocate memory for %d commands\n",
+			   max_cmnds);
+		max_cmnds = asd_ha->hw_prof.max_scbs;
+		return -ENOMEM;
+	}
+	dma_handle = asd_ha->hw_prof.scb_ext->dma_handle;
+	dma_addr = ALIGN((unsigned long) dma_handle, ASD_SCB_SIZE);
+	dma_addr -= asd_ha->hw_prof.max_scbs * ASD_SCB_SIZE;
+	dma_handle = (dma_addr_t) dma_addr;
+	asd_write_reg_addr(asd_ha, CMDCTXBASE, dma_handle);
+	d = asd_read_reg_dword(asd_ha, CTXDOMAIN);
+	d &= ~1;
+	asd_write_reg_dword(asd_ha, CTXDOMAIN, d);
+
+	asd_ha->hw_prof.max_scbs = max_cmnds;
+
+	return 0;
+}
+
+/**
+ * asd_init_ctxmem -- initialize context memory
+ * asd_ha: pointer to host adapter structure
+ *
+ * This function sets the maximum number of SCBs and
+ * DDBs which can be used by the sequencer.  This is normally
+ * 512 and 128 respectively.  If support for more SCBs or more DDBs
+ * is required then CMDCTXBASE, DEVCTXBASE and CTXDOMAIN are
+ * initialized here to extend context memory to point to host memory,
+ * thus allowing unlimited support for SCBs and DDBs -- only limited
+ * by host memory.
+ */
+static int asd_init_ctxmem(struct asd_ha_struct *asd_ha)
+{
+	int bitmap_bytes;
+
+	asd_get_max_scb_ddb(asd_ha);
+	asd_extend_devctx(asd_ha);
+	asd_extend_cmdctx(asd_ha);
+
+	/* The kernel wants bitmaps to be unsigned long sized. */
+	bitmap_bytes = (asd_ha->hw_prof.max_ddbs+7)/8;
+	bitmap_bytes = BITS_TO_LONGS(bitmap_bytes*8)*sizeof(unsigned long);
+	asd_ha->hw_prof.ddb_bitmap = kzalloc(bitmap_bytes, GFP_KERNEL);
+	if (!asd_ha->hw_prof.ddb_bitmap)
+		return -ENOMEM;
+	spin_lock_init(&asd_ha->hw_prof.ddb_lock);
+
+	return 0;
+}
+
+int asd_init_hw(struct asd_ha_struct *asd_ha)
+{
+	int err;
+	u32 v;
+
+	err = asd_init_sw(asd_ha);
+	if (err)
+		return err;
+
+	err = pci_read_config_dword(asd_ha->pcidev, PCIC_HSTPCIX_CNTRL, &v);
+	if (err) {
+		asd_printk("couldn't read PCIC_HSTPCIX_CNTRL of %s\n",
+			   pci_name(asd_ha->pcidev));
+		return err;
+	}
+	pci_write_config_dword(asd_ha->pcidev, PCIC_HSTPCIX_CNTRL,
+					v | SC_TMR_DIS);
+	if (err) {
+		asd_printk("couldn't disable split completion timer of %s\n",
+			   pci_name(asd_ha->pcidev));
+		return err;
+	}
+
+	err = asd_read_ocm(asd_ha);
+	if (err) {
+		asd_printk("couldn't read ocm(%d)\n", err);
+		/* While suspicios, it is not an error that we
+		 * couldn't read the OCM. */
+	}
+
+	err = asd_read_flash(asd_ha);
+	if (err) {
+		asd_printk("couldn't read flash(%d)\n", err);
+		/* While suspicios, it is not an error that we
+		 * couldn't read FLASH memory.
+		 */
+	}
+
+	asd_init_ctxmem(asd_ha);
+
+	asd_get_user_sas_addr(asd_ha);
+	if (!asd_ha->hw_prof.sas_addr[0]) {
+		asd_printk("No SAS Address provided for %s\n",
+			   pci_name(asd_ha->pcidev));
+		err = -ENODEV;
+		goto Out;
+	}
+
+	asd_propagate_sas_addr(asd_ha);
+
+	err = asd_init_phys(asd_ha);
+	if (err) {
+		asd_printk("couldn't initialize phys for %s\n",
+			    pci_name(asd_ha->pcidev));
+		goto Out;
+	}
+
+	err = asd_init_scbs(asd_ha);
+	if (err) {
+		asd_printk("couldn't initialize scbs for %s\n",
+			    pci_name(asd_ha->pcidev));
+		goto Out;
+	}
+
+	err = asd_init_dl(asd_ha);
+	if (err) {
+		asd_printk("couldn't initialize the done list:%d\n",
+			    err);
+		goto Out;
+	}
+
+	err = asd_init_escbs(asd_ha);
+	if (err) {
+		asd_printk("couldn't initialize escbs\n");
+		goto Out;
+	}
+
+	err = asd_init_chip(asd_ha);
+	if (err) {
+		asd_printk("couldn't init the chip\n");
+		goto Out;
+	}
+Out:
+	return err;
+}
+
+/* ---------- Chip reset ---------- */
+
+/**
+ * asd_chip_reset -- reset the host adapter, etc
+ * @asd_ha: pointer to host adapter structure of interest
+ *
+ * Called from the ISR.  Hard reset the chip.  Let everything
+ * timeout.  This should be no different than hot-unplugging the
+ * host adapter.  Once everything times out we'll init the chip with
+ * a call to asd_init_chip() and enable interrupts with asd_enable_ints().
+ * XXX finish.
+ */
+static void asd_chip_reset(struct asd_ha_struct *asd_ha)
+{
+	struct sas_ha_struct *sas_ha = &asd_ha->sas_ha;
+
+	ASD_DPRINTK("chip reset for %s\n", pci_name(asd_ha->pcidev));
+	asd_chip_hardrst(asd_ha);
+	sas_ha->notify_ha_event(sas_ha, HAE_RESET);
+}
+
+/* ---------- Done List Routines ---------- */
+
+static void asd_dl_tasklet_handler(unsigned long data)
+{
+	struct asd_ha_struct *asd_ha = (struct asd_ha_struct *) data;
+	struct asd_seq_data *seq = &asd_ha->seq;
+	unsigned long flags;
+
+	while (1) {
+		struct done_list_struct *dl = &seq->dl[seq->dl_next];
+		struct asd_ascb *ascb;
+
+		if ((dl->toggle & DL_TOGGLE_MASK) != seq->dl_toggle)
+			break;
+
+		/* find the aSCB */
+		spin_lock_irqsave(&seq->tc_index_lock, flags);
+		ascb = asd_tc_index_find(seq, (int)le16_to_cpu(dl->index));
+		spin_unlock_irqrestore(&seq->tc_index_lock, flags);
+		if (unlikely(!ascb)) {
+			ASD_DPRINTK("BUG:sequencer:dl:no ascb?!\n");
+			goto next_1;
+		} else if (ascb->scb->header.opcode == EMPTY_SCB) {
+			goto out;
+		} else if (!ascb->uldd_timer && !del_timer(&ascb->timer)) {
+			goto next_1;
+		}
+		spin_lock_irqsave(&seq->pend_q_lock, flags);
+		list_del_init(&ascb->list);
+		seq->pending--;
+		spin_unlock_irqrestore(&seq->pend_q_lock, flags);
+	out:
+		ascb->tasklet_complete(ascb, dl);
+
+	next_1:
+		seq->dl_next = (seq->dl_next + 1) & (ASD_DL_SIZE-1);
+		if (!seq->dl_next)
+			seq->dl_toggle ^= DL_TOGGLE_MASK;
+	}
+}
+
+/* ---------- Interrupt Service Routines ---------- */
+
+/**
+ * asd_process_donelist_isr -- schedule processing of done list entries
+ * @asd_ha: pointer to host adapter structure
+ */
+static inline void asd_process_donelist_isr(struct asd_ha_struct *asd_ha)
+{
+	tasklet_schedule(&asd_ha->seq.dl_tasklet);
+}
+
+/**
+ * asd_com_sas_isr -- process device communication interrupt (COMINT)
+ * @asd_ha: pointer to host adapter structure
+ */
+static inline void asd_com_sas_isr(struct asd_ha_struct *asd_ha)
+{
+	u32 comstat = asd_read_reg_dword(asd_ha, COMSTAT);
+
+	/* clear COMSTAT int */
+	asd_write_reg_dword(asd_ha, COMSTAT, 0xFFFFFFFF);
+
+	if (comstat & CSBUFPERR) {
+		asd_printk("%s: command/status buffer dma parity error\n",
+			   pci_name(asd_ha->pcidev));
+	} else if (comstat & CSERR) {
+		int i;
+		u32 dmaerr = asd_read_reg_dword(asd_ha, DMAERR);
+		dmaerr &= 0xFF;
+		asd_printk("%s: command/status dma error, DMAERR: 0x%02x, "
+			   "CSDMAADR: 0x%04x, CSDMAADR+4: 0x%04x\n",
+			   pci_name(asd_ha->pcidev),
+			   dmaerr,
+			   asd_read_reg_dword(asd_ha, CSDMAADR),
+			   asd_read_reg_dword(asd_ha, CSDMAADR+4));
+		asd_printk("CSBUFFER:\n");
+		for (i = 0; i < 8; i++) {
+			asd_printk("%08x %08x %08x %08x\n",
+				   asd_read_reg_dword(asd_ha, CSBUFFER),
+				   asd_read_reg_dword(asd_ha, CSBUFFER+4),
+				   asd_read_reg_dword(asd_ha, CSBUFFER+8),
+				   asd_read_reg_dword(asd_ha, CSBUFFER+12));
+		}
+		asd_dump_seq_state(asd_ha, 0);
+	} else if (comstat & OVLYERR) {
+		u32 dmaerr = asd_read_reg_dword(asd_ha, DMAERR);
+		dmaerr = (dmaerr >> 8) & 0xFF;
+		asd_printk("%s: overlay dma error:0x%x\n",
+			   pci_name(asd_ha->pcidev),
+			   dmaerr);
+	}
+	asd_chip_reset(asd_ha);
+}
+
+static inline void asd_arp2_err(struct asd_ha_struct *asd_ha, u32 dchstatus)
+{
+	static const char *halt_code[256] = {
+		"UNEXPECTED_INTERRUPT0",
+		"UNEXPECTED_INTERRUPT1",
+		"UNEXPECTED_INTERRUPT2",
+		"UNEXPECTED_INTERRUPT3",
+		"UNEXPECTED_INTERRUPT4",
+		"UNEXPECTED_INTERRUPT5",
+		"UNEXPECTED_INTERRUPT6",
+		"UNEXPECTED_INTERRUPT7",
+		"UNEXPECTED_INTERRUPT8",
+		"UNEXPECTED_INTERRUPT9",
+		"UNEXPECTED_INTERRUPT10",
+		[11 ... 19] = "unknown[11,19]",
+		"NO_FREE_SCB_AVAILABLE",
+		"INVALID_SCB_OPCODE",
+		"INVALID_MBX_OPCODE",
+		"INVALID_ATA_STATE",
+		"ATA_QUEUE_FULL",
+		"ATA_TAG_TABLE_FAULT",
+		"ATA_TAG_MASK_FAULT",
+		"BAD_LINK_QUEUE_STATE",
+		"DMA2CHIM_QUEUE_ERROR",
+		"EMPTY_SCB_LIST_FULL",
+		"unknown[30]",
+		"IN_USE_SCB_ON_FREE_LIST",
+		"BAD_OPEN_WAIT_STATE",
+		"INVALID_STP_AFFILIATION",
+		"unknown[34]",
+		"EXEC_QUEUE_ERROR",
+		"TOO_MANY_EMPTIES_NEEDED",
+		"EMPTY_REQ_QUEUE_ERROR",
+		"Q_MONIRTT_MGMT_ERROR",
+		"TARGET_MODE_FLOW_ERROR",
+		"DEVICE_QUEUE_NOT_FOUND",
+		"START_IRTT_TIMER_ERROR",
+		"ABORT_TASK_ILLEGAL_REQ",
+		[43 ... 255] = "unknown[43,255]"
+	};
+
+	if (dchstatus & CSEQINT) {
+		u32 arp2int = asd_read_reg_dword(asd_ha, CARP2INT);
+
+		if (arp2int & (ARP2WAITTO|ARP2ILLOPC|ARP2PERR|ARP2CIOPERR)) {
+			asd_printk("%s: CSEQ arp2int:0x%x\n",
+				   pci_name(asd_ha->pcidev),
+				   arp2int);
+		} else if (arp2int & ARP2HALTC)
+			asd_printk("%s: CSEQ halted: %s\n",
+				   pci_name(asd_ha->pcidev),
+				   halt_code[(arp2int>>16)&0xFF]);
+		else
+			asd_printk("%s: CARP2INT:0x%x\n",
+				   pci_name(asd_ha->pcidev),
+				   arp2int);
+	}
+	if (dchstatus & LSEQINT_MASK) {
+		int lseq;
+		u8  lseq_mask = dchstatus & LSEQINT_MASK;
+
+		for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+			u32 arp2int = asd_read_reg_dword(asd_ha,
+							 LmARP2INT(lseq));
+			if (arp2int & (ARP2WAITTO | ARP2ILLOPC | ARP2PERR
+				       | ARP2CIOPERR)) {
+				asd_printk("%s: LSEQ%d arp2int:0x%x\n",
+					   pci_name(asd_ha->pcidev),
+					   lseq, arp2int);
+				/* XXX we should only do lseq reset */
+			} else if (arp2int & ARP2HALTC)
+				asd_printk("%s: LSEQ%d halted: %s\n",
+					   pci_name(asd_ha->pcidev),
+					   lseq,halt_code[(arp2int>>16)&0xFF]);
+			else
+				asd_printk("%s: LSEQ%d ARP2INT:0x%x\n",
+					   pci_name(asd_ha->pcidev), lseq,
+					   arp2int);
+		}
+	}
+	asd_chip_reset(asd_ha);
+}
+
+/**
+ * asd_dch_sas_isr -- process device channel interrupt (DEVINT)
+ * @asd_ha: pointer to host adapter structure
+ */
+static inline void asd_dch_sas_isr(struct asd_ha_struct *asd_ha)
+{
+	u32 dchstatus = asd_read_reg_dword(asd_ha, DCHSTATUS);
+
+	if (dchstatus & CFIFTOERR) {
+		asd_printk("%s: CFIFTOERR\n", pci_name(asd_ha->pcidev));
+		asd_chip_reset(asd_ha);
+	} else
+		asd_arp2_err(asd_ha, dchstatus);
+}
+
+/**
+ * ads_rbi_exsi_isr -- process external system interface interrupt (INITERR)
+ * @asd_ha: pointer to host adapter structure
+ */
+static inline void asd_rbi_exsi_isr(struct asd_ha_struct *asd_ha)
+{
+	u32 stat0r = asd_read_reg_dword(asd_ha, ASISTAT0R);
+
+	if (!(stat0r & ASIERR)) {
+		asd_printk("hmm, EXSI interrupted but no error?\n");
+		return;
+	}
+
+	if (stat0r & ASIFMTERR) {
+		asd_printk("ASI SEEPROM format error for %s\n",
+			   pci_name(asd_ha->pcidev));
+	} else if (stat0r & ASISEECHKERR) {
+		u32 stat1r = asd_read_reg_dword(asd_ha, ASISTAT1R);
+		asd_printk("ASI SEEPROM checksum 0x%x error for %s\n",
+			   stat1r & CHECKSUM_MASK,
+			   pci_name(asd_ha->pcidev));
+	} else {
+		u32 statr = asd_read_reg_dword(asd_ha, ASIERRSTATR);
+
+		if (!(statr & CPI2ASIMSTERR_MASK)) {
+			ASD_DPRINTK("hmm, ASIERR?\n");
+			return;
+		} else {
+			u32 addr = asd_read_reg_dword(asd_ha, ASIERRADDR);
+			u32 data = asd_read_reg_dword(asd_ha, ASIERRDATAR);
+
+			asd_printk("%s: CPI2 xfer err: addr: 0x%x, wdata: 0x%x, "
+				   "count: 0x%x, byteen: 0x%x, targerr: 0x%x "
+				   "master id: 0x%x, master err: 0x%x\n",
+				   pci_name(asd_ha->pcidev),
+				   addr, data,
+				   (statr & CPI2ASIBYTECNT_MASK) >> 16,
+				   (statr & CPI2ASIBYTEEN_MASK) >> 12,
+				   (statr & CPI2ASITARGERR_MASK) >> 8,
+				   (statr & CPI2ASITARGMID_MASK) >> 4,
+				   (statr & CPI2ASIMSTERR_MASK));
+		}
+	}
+	asd_chip_reset(asd_ha);
+}
+
+/**
+ * asd_hst_pcix_isr -- process host interface interrupts
+ * @asd_ha: pointer to host adapter structure
+ *
+ * Asserted on PCIX errors: target abort, etc.
+ */
+static inline void asd_hst_pcix_isr(struct asd_ha_struct *asd_ha)
+{
+	u16 status;
+	u32 pcix_status;
+	u32 ecc_status;
+
+	pci_read_config_word(asd_ha->pcidev, PCI_STATUS, &status);
+	pci_read_config_dword(asd_ha->pcidev, PCIX_STATUS, &pcix_status);
+	pci_read_config_dword(asd_ha->pcidev, ECC_CTRL_STAT, &ecc_status);
+
+	if (status & PCI_STATUS_DETECTED_PARITY)
+		asd_printk("parity error for %s\n", pci_name(asd_ha->pcidev));
+	else if (status & PCI_STATUS_REC_MASTER_ABORT)
+		asd_printk("master abort for %s\n", pci_name(asd_ha->pcidev));
+	else if (status & PCI_STATUS_REC_TARGET_ABORT)
+		asd_printk("target abort for %s\n", pci_name(asd_ha->pcidev));
+	else if (status & PCI_STATUS_PARITY)
+		asd_printk("data parity for %s\n", pci_name(asd_ha->pcidev));
+	else if (pcix_status & RCV_SCE) {
+		asd_printk("received split completion error for %s\n",
+			   pci_name(asd_ha->pcidev));
+		pci_write_config_dword(asd_ha->pcidev,PCIX_STATUS,pcix_status);
+		/* XXX: Abort task? */
+		return;
+	} else if (pcix_status & UNEXP_SC) {
+		asd_printk("unexpected split completion for %s\n",
+			   pci_name(asd_ha->pcidev));
+		pci_write_config_dword(asd_ha->pcidev,PCIX_STATUS,pcix_status);
+		/* ignore */
+		return;
+	} else if (pcix_status & SC_DISCARD)
+		asd_printk("split completion discarded for %s\n",
+			   pci_name(asd_ha->pcidev));
+	else if (ecc_status & UNCOR_ECCERR)
+		asd_printk("uncorrectable ECC error for %s\n",
+			   pci_name(asd_ha->pcidev));
+	asd_chip_reset(asd_ha);
+}
+
+/**
+ * asd_hw_isr -- host adapter interrupt service routine
+ * @irq: ignored
+ * @dev_id: pointer to host adapter structure
+ * @regs: ignored
+ *
+ * The ISR processes done list entries and level 3 error handling.
+ */
+irqreturn_t asd_hw_isr(int irq, void *dev_id, struct pt_regs *regs)
+{
+	struct asd_ha_struct *asd_ha = dev_id;
+	u32 chimint = asd_read_reg_dword(asd_ha, CHIMINT);
+
+	if (!chimint)
+		return IRQ_NONE;
+
+	asd_write_reg_dword(asd_ha, CHIMINT, chimint);
+	(void) asd_read_reg_dword(asd_ha, CHIMINT);
+
+	if (chimint & DLAVAIL)
+		asd_process_donelist_isr(asd_ha);
+	if (chimint & COMINT)
+		asd_com_sas_isr(asd_ha);
+	if (chimint & DEVINT)
+		asd_dch_sas_isr(asd_ha);
+	if (chimint & INITERR)
+		asd_rbi_exsi_isr(asd_ha);
+	if (chimint & HOSTERR)
+		asd_hst_pcix_isr(asd_ha);
+
+	return IRQ_HANDLED;
+}
+
+/* ---------- SCB handling ---------- */
+
+static inline struct asd_ascb *asd_ascb_alloc(struct asd_ha_struct *asd_ha,
+					      unsigned int gfp_flags)
+{
+	extern kmem_cache_t *asd_ascb_cache;
+	struct asd_seq_data *seq = &asd_ha->seq;
+	struct asd_ascb *ascb;
+	unsigned long flags;
+
+	ascb = kmem_cache_alloc(asd_ascb_cache, gfp_flags);
+
+	if (ascb) {
+		memset(ascb, 0, sizeof(*ascb));
+		ascb->dma_scb.size = sizeof(struct scb);
+		ascb->dma_scb.vaddr = dma_pool_alloc(asd_ha->scb_pool,
+						     gfp_flags,
+						    &ascb->dma_scb.dma_handle);
+		if (!ascb->dma_scb.vaddr) {
+			kmem_cache_free(asd_ascb_cache, ascb);
+			return NULL;
+		}
+		memset(ascb->dma_scb.vaddr, 0, sizeof(struct scb));
+		asd_init_ascb(asd_ha, ascb);
+
+		spin_lock_irqsave(&seq->tc_index_lock, flags);
+		ascb->tc_index = asd_tc_index_get(seq, ascb);
+		spin_unlock_irqrestore(&seq->tc_index_lock, flags);
+		if (ascb->tc_index == -1)
+			goto undo;
+
+		ascb->scb->header.index = cpu_to_le16((u16)ascb->tc_index);
+	}
+
+	return ascb;
+undo:
+	dma_pool_free(asd_ha->scb_pool, ascb->dma_scb.vaddr,
+		      ascb->dma_scb.dma_handle);
+	kmem_cache_free(asd_ascb_cache, ascb);
+	ASD_DPRINTK("no index for ascb\n");
+	return NULL;
+}
+
+/**
+ * asd_ascb_alloc_list -- allocate a list of aSCBs
+ * @asd_ha: pointer to host adapter structure
+ * @num: pointer to integer number of aSCBs
+ * @gfp_flags: GFP_ flags.
+ *
+ * This is the only function which is used to allocate aSCBs.
+ * It can allocate one or many. If more than one, then they form
+ * a linked list in two ways: by their list field of the ascb struct
+ * and by the next_scb field of the scb_header.
+ *
+ * Returns NULL if no memory was available, else pointer to a list
+ * of ascbs.  When this function returns, @num would be the number
+ * of SCBs which were not able to be allocated, 0 if all requested
+ * were able to be allocated.
+ */
+struct asd_ascb *asd_ascb_alloc_list(struct asd_ha_struct
+				     *asd_ha, int *num,
+				     unsigned int gfp_flags)
+{
+	struct asd_ascb *first = NULL;
+
+	for ( ; *num > 0; --*num) {
+		struct asd_ascb *ascb = asd_ascb_alloc(asd_ha, gfp_flags);
+
+		if (!ascb)
+			break;
+		else if (!first)
+			first = ascb;
+		else {
+			struct asd_ascb *last = list_entry(first->list.prev,
+							   struct asd_ascb,
+							   list);
+			list_add_tail(&ascb->list, &first->list);
+			last->scb->header.next_scb =
+				cpu_to_le64(((u64)ascb->dma_scb.dma_handle));
+		}
+	}
+
+	return first;
+}
+
+/**
+ * asd_swap_head_scb -- swap the head scb
+ * @asd_ha: pointer to host adapter structure
+ * @ascb: pointer to the head of an ascb list
+ *
+ * The sequencer knows the DMA address of the next SCB to be DMAed to
+ * the host adapter, from initialization or from the last list DMAed.
+ * seq->next_scb keeps the address of this SCB.  The sequencer will
+ * DMA to the host adapter this list of SCBs.  But the head (first
+ * element) of this list is not known to the sequencer.  Here we swap
+ * the head of the list with the known SCB (memcpy()).
+ * Only one memcpy() is required per list so it is in our interest
+ * to keep the list of SCB as long as possible so that the ratio
+ * of number of memcpy calls to the number of SCB DMA-ed is as small
+ * as possible.
+ *
+ * LOCKING: called with the pending list lock held.
+ */
+static inline void asd_swap_head_scb(struct asd_ha_struct *asd_ha,
+				     struct asd_ascb *ascb)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	struct asd_ascb *last = list_entry(ascb->list.prev,
+					   struct asd_ascb,
+					   list);
+	struct asd_dma_tok t = ascb->dma_scb;
+
+	memcpy(seq->next_scb.vaddr, ascb->scb, sizeof(*ascb->scb));
+	ascb->dma_scb = seq->next_scb;
+	ascb->scb = ascb->dma_scb.vaddr;
+	seq->next_scb = t;
+	last->scb->header.next_scb =
+		cpu_to_le64(((u64)seq->next_scb.dma_handle));
+}
+
+/**
+ * asd_start_timers -- (add and) start timers of SCBs
+ * @list: pointer to struct list_head of the scbs
+ * @to: timeout in jiffies
+ *
+ * If an SCB in the @list has no timer function, assign the default
+ * one,  then start the timer of the SCB.  This function is
+ * intended to be called from asd_post_ascb_list(), just prior to
+ * posting the SCBs to the sequencer.
+ */
+static inline void asd_start_scb_timers(struct list_head *list)
+{
+	struct asd_ascb *ascb;
+	list_for_each_entry(ascb, list, list) {
+		if (!ascb->uldd_timer) {
+			ascb->timer.data = (unsigned long) ascb;
+			ascb->timer.function = asd_ascb_timedout;
+			ascb->timer.expires = jiffies + AIC94XX_SCB_TIMEOUT;
+			add_timer(&ascb->timer);
+		}
+	}
+}
+
+/**
+ * asd_post_ascb_list -- post a list of 1 or more aSCBs to the host adapter
+ * @asd_ha: pointer to a host adapter structure
+ * @ascb: pointer to the first aSCB in the list
+ * @num: number of aSCBs in the list (to be posted)
+ *
+ * See queueing comment in asd_post_escb_list().
+ *
+ * Additional note on queuing: In order to minimize the ratio of memcpy()
+ * to the number of ascbs sent, we try to batch-send as many ascbs as possible
+ * in one go.
+ * Two cases are possible:
+ *    A) can_queue >= num,
+ *    B) can_queue < num.
+ * Case A: we can send the whole batch at once.  Increment "pending"
+ * in the beginning of this function, when it is checked, in order to
+ * eliminate races when this function is called by multiple processes.
+ * Case B: should never happen if the managing layer considers
+ * lldd_queue_size.
+ */
+int asd_post_ascb_list(struct asd_ha_struct *asd_ha, struct asd_ascb *ascb,
+		       int num)
+{
+	unsigned long flags;
+	LIST_HEAD(list);
+	int can_queue;
+
+	spin_lock_irqsave(&asd_ha->seq.pend_q_lock, flags);
+	can_queue = asd_ha->hw_prof.max_scbs - asd_ha->seq.pending;
+	if (can_queue >= num)
+		asd_ha->seq.pending += num;
+	else
+		can_queue = 0;
+
+	if (!can_queue) {
+		spin_unlock_irqrestore(&asd_ha->seq.pend_q_lock, flags);
+		asd_printk("%s: scb queue full\n", pci_name(asd_ha->pcidev));
+		return -SAS_QUEUE_FULL;
+	}
+
+	asd_swap_head_scb(asd_ha, ascb);
+
+	__list_add(&list, ascb->list.prev, &ascb->list);
+
+	asd_start_scb_timers(&list);
+
+	asd_ha->seq.scbpro += num;
+	list_splice_init(&list, asd_ha->seq.pend_q.prev);
+	asd_write_reg_dword(asd_ha, SCBPRO, (u32)asd_ha->seq.scbpro);
+	spin_unlock_irqrestore(&asd_ha->seq.pend_q_lock, flags);
+
+	return 0;
+}
+
+/**
+ * asd_post_escb_list -- post a list of 1 or more empty scb
+ * @asd_ha: pointer to a host adapter structure
+ * @ascb: pointer to the first empty SCB in the list
+ * @num: number of aSCBs in the list (to be posted)
+ *
+ * This is essentially the same as asd_post_ascb_list, but we do not
+ * increment pending, add those to the pending list or get indexes.
+ * See asd_init_escbs() and asd_init_post_escbs().
+ *
+ * Since sending a list of ascbs is a superset of sending a single
+ * ascb, this function exists to generalize this.  More specifically,
+ * when sending a list of those, we want to do only a _single_
+ * memcpy() at swap head, as opposed to for each ascb sent (in the
+ * case of sending them one by one).  That is, we want to minimize the
+ * ratio of memcpy() operations to the number of ascbs sent.  The same
+ * logic applies to asd_post_ascb_list().
+ */
+int asd_post_escb_list(struct asd_ha_struct *asd_ha, struct asd_ascb *ascb,
+		       int num)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&asd_ha->seq.pend_q_lock, flags);
+	asd_swap_head_scb(asd_ha, ascb);
+	asd_ha->seq.scbpro += num;
+	asd_write_reg_dword(asd_ha, SCBPRO, (u32)asd_ha->seq.scbpro);
+	spin_unlock_irqrestore(&asd_ha->seq.pend_q_lock, flags);
+
+	return 0;
+}
+
+/* ---------- LED ---------- */
+
+/**
+ * asd_turn_led -- turn on/off an LED
+ * @asd_ha: pointer to host adapter structure
+ * @phy_id: the PHY id whose LED we want to manupulate
+ * @op: 1 to turn on, 0 to turn off
+ */
+void asd_turn_led(struct asd_ha_struct *asd_ha, int phy_id, int op)
+{
+	if (phy_id < ASD_MAX_PHYS) {
+		u32 v = asd_read_reg_dword(asd_ha, LmCONTROL(phy_id));
+		if (op)
+			v |= LEDPOL;
+		else
+			v &= ~LEDPOL;
+		asd_write_reg_dword(asd_ha, LmCONTROL(phy_id), v);
+	}
+}
+
+/**
+ * asd_control_led -- enable/disable an LED on the board
+ * @asd_ha: pointer to host adapter structure
+ * @phy_id: integer, the phy id
+ * @op: integer, 1 to enable, 0 to disable the LED
+ *
+ * First we output enable the LED, then we set the source
+ * to be an external module.
+ */
+void asd_control_led(struct asd_ha_struct *asd_ha, int phy_id, int op)
+{
+	if (phy_id < ASD_MAX_PHYS) {
+		u32 v;
+
+		v = asd_read_reg_dword(asd_ha, GPIOOER);
+		if (op)
+			v |= (1 << phy_id);
+		else
+			v &= ~(1 << phy_id);
+		asd_write_reg_dword(asd_ha, GPIOOER, v);
+
+		v = asd_read_reg_dword(asd_ha, GPIOCNFGR);
+		if (op)
+			v |= (1 << phy_id);
+		else
+			v &= ~(1 << phy_id);
+		asd_write_reg_dword(asd_ha, GPIOCNFGR, v);
+	}
+}
+
+/* ---------- PHY enable ---------- */
+
+static int asd_enable_phy(struct asd_ha_struct *asd_ha, int phy_id)
+{
+	struct asd_phy *phy = &asd_ha->phys[phy_id];
+
+	asd_write_reg_byte(asd_ha, LmSEQ_OOB_REG(phy_id, INT_ENABLE_2), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_OOB_REG(phy_id, HOT_PLUG_DELAY),
+			   HOTPLUG_DELAY_TIMEOUT);
+
+	/* Get defaults from manuf. sector */
+	/* XXX we need defaults for those in case MS is broken. */
+	asd_write_reg_byte(asd_ha, LmSEQ_OOB_REG(phy_id, PHY_CONTROL_0),
+			   phy->phy_desc->phy_control_0);
+	asd_write_reg_byte(asd_ha, LmSEQ_OOB_REG(phy_id, PHY_CONTROL_1),
+			   phy->phy_desc->phy_control_1);
+	asd_write_reg_byte(asd_ha, LmSEQ_OOB_REG(phy_id, PHY_CONTROL_2),
+			   phy->phy_desc->phy_control_2);
+	asd_write_reg_byte(asd_ha, LmSEQ_OOB_REG(phy_id, PHY_CONTROL_3),
+			   phy->phy_desc->phy_control_3);
+
+	asd_write_reg_dword(asd_ha, LmSEQ_TEN_MS_COMINIT_TIMEOUT(phy_id),
+			    ASD_COMINIT_TIMEOUT);
+
+	asd_write_reg_addr(asd_ha, LmSEQ_TX_ID_ADDR_FRAME(phy_id),
+			   phy->id_frm_tok->dma_handle);
+
+	asd_control_led(asd_ha, phy_id, 1);
+
+	return 0;
+}
+
+int asd_enable_phys(struct asd_ha_struct *asd_ha, const u8 phy_mask)
+{
+	u8  phy_m;
+	u8  i;
+	int num = 0, k;
+	struct asd_ascb *ascb;
+	struct asd_ascb *ascb_list;
+
+	if (!phy_mask) {
+		asd_printk("%s called with phy_mask of 0!?\n", __FUNCTION__);
+		return 0;
+	}
+
+	for_each_phy(phy_mask, phy_m, i) {
+		num++;
+		asd_enable_phy(asd_ha, i);
+	}
+
+	k = num;
+	ascb_list = asd_ascb_alloc_list(asd_ha, &k, GFP_KERNEL);
+	if (!ascb_list) {
+		asd_printk("no memory for control phy ascb list\n");
+		return -ENOMEM;
+	}
+	num -= k;
+
+	ascb = ascb_list;
+	for_each_phy(phy_mask, phy_m, i) {
+		asd_build_control_phy(ascb, i, ENABLE_PHY);
+		ascb = list_entry(ascb->list.next, struct asd_ascb, list);
+	}
+	ASD_DPRINTK("posting %d control phy scbs\n", num);
+	k = asd_post_ascb_list(asd_ha, ascb_list, num);
+	if (k)
+		asd_ascb_free_list(ascb_list);
+
+	return k;
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.h b/drivers/scsi/aic94xx/aic94xx_hwi.h
new file mode 100644
index 000000000000..c7d505388fed
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_hwi.h
@@ -0,0 +1,397 @@
+/*
+ * Aic94xx SAS/SATA driver hardware interface header file.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#ifndef _AIC94XX_HWI_H_
+#define _AIC94XX_HWI_H_
+
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/dma-mapping.h>
+
+#include <scsi/libsas.h>
+
+#include "aic94xx.h"
+#include "aic94xx_sas.h"
+
+/* Define ASD_MAX_PHYS to the maximum phys ever. Currently 8. */
+#define ASD_MAX_PHYS       8
+#define ASD_PCBA_SN_SIZE   12
+
+/* Those are to be further named properly, the "RAZORx" part, and
+ * subsequently included in include/linux/pci_ids.h.
+ */
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR10 0x410
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR12 0x412
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR1E 0x41E
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR30 0x430
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR32 0x432
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR3E 0x43E
+#define PCI_DEVICE_ID_ADAPTEC2_RAZOR3F 0x43F
+
+struct asd_ha_addrspace {
+	void __iomem  *addr;
+	unsigned long  start;       /* pci resource start */
+	unsigned long  len;         /* pci resource len */
+	unsigned long  flags;       /* pci resource flags */
+
+	/* addresses internal to the host adapter */
+	u32 swa_base; /* mmspace 1 (MBAR1) uses this only */
+	u32 swb_base;
+	u32 swc_base;
+};
+
+struct bios_struct {
+	int    present;
+	u8     maj;
+	u8     min;
+	u32    bld;
+};
+
+struct unit_element_struct {
+	u16    num;
+	u16    size;
+	void   *area;
+};
+
+struct flash_struct {
+	u32    bar;
+	int    present;
+	int    wide;
+	u8     manuf;
+	u8     dev_id;
+	u8     sec_prot;
+
+	u32    dir_offs;
+};
+
+struct asd_phy_desc {
+	/* From CTRL-A settings, then set to what is appropriate */
+	u8     sas_addr[SAS_ADDR_SIZE];
+	u8     max_sas_lrate;
+	u8     min_sas_lrate;
+	u8     max_sata_lrate;
+	u8     min_sata_lrate;
+	u8     flags;
+#define ASD_CRC_DIS  1
+#define ASD_SATA_SPINUP_HOLD 2
+
+	u8     phy_control_0; /* mode 5 reg 0x160 */
+	u8     phy_control_1; /* mode 5 reg 0x161 */
+	u8     phy_control_2; /* mode 5 reg 0x162 */
+	u8     phy_control_3; /* mode 5 reg 0x163 */
+};
+
+struct asd_dma_tok {
+	void *vaddr;
+	dma_addr_t dma_handle;
+	size_t size;
+};
+
+struct hw_profile {
+	struct bios_struct bios;
+	struct unit_element_struct ue;
+	struct flash_struct flash;
+
+	u8     sas_addr[SAS_ADDR_SIZE];
+	char   pcba_sn[ASD_PCBA_SN_SIZE+1];
+
+	u8     enabled_phys;	  /* mask of enabled phys */
+	struct asd_phy_desc phy_desc[ASD_MAX_PHYS];
+	u32    max_scbs;	  /* absolute sequencer scb queue size */
+	struct asd_dma_tok *scb_ext;
+	u32    max_ddbs;
+	struct asd_dma_tok *ddb_ext;
+
+	spinlock_t ddb_lock;
+	void  *ddb_bitmap;
+
+	int    num_phys;	  /* ENABLEABLE */
+	int    max_phys;	  /* REPORTED + ENABLEABLE */
+
+	unsigned addr_range;	  /* max # of addrs; max # of possible ports */
+	unsigned port_name_base;
+	unsigned dev_name_base;
+	unsigned sata_name_base;
+};
+
+struct asd_ascb {
+	struct list_head list;
+	struct asd_ha_struct *ha;
+
+	struct scb *scb;	  /* equals dma_scb->vaddr */
+	struct asd_dma_tok dma_scb;
+	struct asd_dma_tok *sg_arr;
+
+	void (*tasklet_complete)(struct asd_ascb *, struct done_list_struct *);
+	u8     uldd_timer:1;
+
+	/* internally generated command */
+	struct timer_list timer;
+	struct completion completion;
+	u8        tag_valid:1;
+	__be16    tag;		  /* error recovery only */
+
+	/* If this is an Empty SCB, index of first edb in seq->edb_arr. */
+	int    edb_index;
+
+	/* Used by the timer timeout function. */
+	int    tc_index;
+
+	void   *uldd_task;
+};
+
+#define ASD_DL_SIZE_BITS   0x8
+#define ASD_DL_SIZE        (1<<(2+ASD_DL_SIZE_BITS))
+#define ASD_DEF_DL_TOGGLE  0x01
+
+struct asd_seq_data {
+	spinlock_t pend_q_lock;
+	u16    scbpro;
+	int    pending;
+	struct list_head pend_q;
+	int    can_queue;	  /* per adapter */
+	struct asd_dma_tok next_scb; /* next scb to be delivered to CSEQ */
+
+	spinlock_t tc_index_lock;
+	void **tc_index_array;
+	void *tc_index_bitmap;
+	int   tc_index_bitmap_bits;
+
+	struct tasklet_struct dl_tasklet;
+	struct done_list_struct *dl; /* array of done list entries, equals */
+	struct asd_dma_tok *actual_dl; /* actual_dl->vaddr */
+	int    dl_toggle;
+	int    dl_next;
+
+	int    num_edbs;
+	struct asd_dma_tok **edb_arr;
+	int    num_escbs;
+	struct asd_ascb **escb_arr; /* array of pointers to escbs */
+};
+
+/* This is the Host Adapter structure.  It describes the hardware
+ * SAS adapter.
+ */
+struct asd_ha_struct {
+	struct pci_dev   *pcidev;
+	const char       *name;
+
+	struct sas_ha_struct sas_ha;
+
+	u8                revision_id;
+
+	int               iospace;
+	spinlock_t        iolock;
+	struct asd_ha_addrspace io_handle[2];
+
+	struct hw_profile hw_prof;
+
+	struct asd_phy    phys[ASD_MAX_PHYS];
+	struct asd_sas_port   ports[ASD_MAX_PHYS];
+
+	struct dma_pool  *scb_pool;
+
+	struct asd_seq_data  seq; /* sequencer related */
+};
+
+/* ---------- Common macros ---------- */
+
+#define ASD_BUSADDR_LO(__dma_handle) ((u32)(__dma_handle))
+#define ASD_BUSADDR_HI(__dma_handle) (((sizeof(dma_addr_t))==8)     \
+                                    ? ((u32)((__dma_handle) >> 32)) \
+                                    : ((u32)0))
+
+#define dev_to_asd_ha(__dev)  pci_get_drvdata(to_pci_dev(__dev))
+#define SCB_SITE_VALID(__site_no) (((__site_no) & 0xF0FF) != 0x00FF   \
+				 && ((__site_no) & 0xF0FF) > 0x001F)
+/* For each bit set in __lseq_mask, set __lseq to equal the bit
+ * position of the set bit and execute the statement following.
+ * __mc is the temporary mask, used as a mask "counter".
+ */
+#define for_each_sequencer(__lseq_mask, __mc, __lseq)                        \
+	for ((__mc)=(__lseq_mask),(__lseq)=0;(__mc)!=0;(__lseq++),(__mc)>>=1)\
+		if (((__mc) & 1))
+#define for_each_phy(__lseq_mask, __mc, __lseq)                              \
+	for ((__mc)=(__lseq_mask),(__lseq)=0;(__mc)!=0;(__lseq++),(__mc)>>=1)\
+		if (((__mc) & 1))
+
+#define PHY_ENABLED(_HA, _I) ((_HA)->hw_prof.enabled_phys & (1<<(_I)))
+
+/* ---------- DMA allocs ---------- */
+
+static inline struct asd_dma_tok *asd_dmatok_alloc(unsigned int flags)
+{
+	return kmem_cache_alloc(asd_dma_token_cache, flags);
+}
+
+static inline void asd_dmatok_free(struct asd_dma_tok *token)
+{
+	kmem_cache_free(asd_dma_token_cache, token);
+}
+
+static inline struct asd_dma_tok *asd_alloc_coherent(struct asd_ha_struct *
+						     asd_ha, size_t size,
+						     unsigned int flags)
+{
+	struct asd_dma_tok *token = asd_dmatok_alloc(flags);
+	if (token) {
+		token->size = size;
+		token->vaddr = dma_alloc_coherent(&asd_ha->pcidev->dev,
+						  token->size,
+						  &token->dma_handle,
+						  flags);
+		if (!token->vaddr) {
+			asd_dmatok_free(token);
+			token = NULL;
+		}
+	}
+	return token;
+}
+
+static inline void asd_free_coherent(struct asd_ha_struct *asd_ha,
+				     struct asd_dma_tok *token)
+{
+	if (token) {
+		dma_free_coherent(&asd_ha->pcidev->dev, token->size,
+				  token->vaddr, token->dma_handle);
+		asd_dmatok_free(token);
+	}
+}
+
+static inline void asd_init_ascb(struct asd_ha_struct *asd_ha,
+				 struct asd_ascb *ascb)
+{
+	INIT_LIST_HEAD(&ascb->list);
+	ascb->scb = ascb->dma_scb.vaddr;
+	ascb->ha = asd_ha;
+	ascb->timer.function = NULL;
+	init_timer(&ascb->timer);
+	ascb->tc_index = -1;
+	init_completion(&ascb->completion);
+}
+
+/* Must be called with the tc_index_lock held!
+ */
+static inline void asd_tc_index_release(struct asd_seq_data *seq, int index)
+{
+	seq->tc_index_array[index] = NULL;
+	clear_bit(index, seq->tc_index_bitmap);
+}
+
+/* Must be called with the tc_index_lock held!
+ */
+static inline int asd_tc_index_get(struct asd_seq_data *seq, void *ptr)
+{
+	int index;
+
+	index = find_first_zero_bit(seq->tc_index_bitmap,
+				    seq->tc_index_bitmap_bits);
+	if (index == seq->tc_index_bitmap_bits)
+		return -1;
+
+	seq->tc_index_array[index] = ptr;
+	set_bit(index, seq->tc_index_bitmap);
+
+	return index;
+}
+
+/* Must be called with the tc_index_lock held!
+ */
+static inline void *asd_tc_index_find(struct asd_seq_data *seq, int index)
+{
+	return seq->tc_index_array[index];
+}
+
+/**
+ * asd_ascb_free -- free a single aSCB after is has completed
+ * @ascb: pointer to the aSCB of interest
+ *
+ * This frees an aSCB after it has been executed/completed by
+ * the sequencer.
+ */
+static inline void asd_ascb_free(struct asd_ascb *ascb)
+{
+	if (ascb) {
+		struct asd_ha_struct *asd_ha = ascb->ha;
+		unsigned long flags;
+
+		BUG_ON(!list_empty(&ascb->list));
+		spin_lock_irqsave(&ascb->ha->seq.tc_index_lock, flags);
+		asd_tc_index_release(&ascb->ha->seq, ascb->tc_index);
+		spin_unlock_irqrestore(&ascb->ha->seq.tc_index_lock, flags);
+		dma_pool_free(asd_ha->scb_pool, ascb->dma_scb.vaddr,
+			      ascb->dma_scb.dma_handle);
+		kmem_cache_free(asd_ascb_cache, ascb);
+	}
+}
+
+/**
+ * asd_ascb_list_free -- free a list of ascbs
+ * @ascb_list: a list of ascbs
+ *
+ * This function will free a list of ascbs allocated by asd_ascb_alloc_list.
+ * It is used when say the scb queueing function returned QUEUE_FULL,
+ * and we do not need the ascbs any more.
+ */
+static inline void asd_ascb_free_list(struct asd_ascb *ascb_list)
+{
+	LIST_HEAD(list);
+	struct list_head *n, *pos;
+
+	__list_add(&list, ascb_list->list.prev, &ascb_list->list);
+	list_for_each_safe(pos, n, &list) {
+		list_del_init(pos);
+		asd_ascb_free(list_entry(pos, struct asd_ascb, list));
+	}
+}
+
+/* ---------- Function declarations ---------- */
+
+int  asd_init_hw(struct asd_ha_struct *asd_ha);
+irqreturn_t asd_hw_isr(int irq, void *dev_id, struct pt_regs *regs);
+
+
+struct asd_ascb *asd_ascb_alloc_list(struct asd_ha_struct
+				     *asd_ha, int *num,
+				     unsigned int gfp_mask);
+
+int  asd_post_ascb_list(struct asd_ha_struct *asd_ha, struct asd_ascb *ascb,
+			int num);
+int  asd_post_escb_list(struct asd_ha_struct *asd_ha, struct asd_ascb *ascb,
+			int num);
+
+int  asd_init_post_escbs(struct asd_ha_struct *asd_ha);
+void asd_build_control_phy(struct asd_ascb *ascb, int phy_id, u8 subfunc);
+void asd_control_led(struct asd_ha_struct *asd_ha, int phy_id, int op);
+void asd_turn_led(struct asd_ha_struct *asd_ha, int phy_id, int op);
+int  asd_enable_phys(struct asd_ha_struct *asd_ha, const u8 phy_mask);
+void asd_build_initiate_link_adm_task(struct asd_ascb *ascb, int phy_id,
+				      u8 subfunc);
+
+void asd_ascb_timedout(unsigned long data);
+int  asd_chip_hardrst(struct asd_ha_struct *asd_ha);
+
+#endif
diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c
new file mode 100644
index 000000000000..3ec2e46f80c6
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_init.c
@@ -0,0 +1,860 @@
+/*
+ * Aic94xx SAS/SATA driver initialization.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+
+#include <scsi/scsi_host.h>
+
+#include "aic94xx.h"
+#include "aic94xx_reg.h"
+#include "aic94xx_hwi.h"
+#include "aic94xx_seq.h"
+
+/* The format is "version.release.patchlevel" */
+#define ASD_DRIVER_VERSION "1.0.2"
+
+static int use_msi = 0;
+module_param_named(use_msi, use_msi, int, S_IRUGO);
+MODULE_PARM_DESC(use_msi, "\n"
+	"\tEnable(1) or disable(0) using PCI MSI.\n"
+	"\tDefault: 0");
+
+static int lldd_max_execute_num = 0;
+module_param_named(collector, lldd_max_execute_num, int, S_IRUGO);
+MODULE_PARM_DESC(collector, "\n"
+	"\tIf greater than one, tells the SAS Layer to run in Task Collector\n"
+	"\tMode.  If 1 or 0, tells the SAS Layer to run in Direct Mode.\n"
+	"\tThe aic94xx SAS LLDD supports both modes.\n"
+	"\tDefault: 0 (Direct Mode).\n");
+
+char sas_addr_str[2*SAS_ADDR_SIZE + 1] = "";
+
+static struct scsi_transport_template *aic94xx_transport_template;
+
+static struct scsi_host_template aic94xx_sht = {
+	.module			= THIS_MODULE,
+	/* .name is initialized */
+	.name			= "aic94xx",
+	.queuecommand		= sas_queuecommand,
+	.target_alloc		= sas_target_alloc,
+	.slave_configure	= sas_slave_configure,
+	.slave_destroy		= sas_slave_destroy,
+	.change_queue_depth	= sas_change_queue_depth,
+	.change_queue_type	= sas_change_queue_type,
+	.bios_param		= sas_bios_param,
+	.can_queue		= 1,
+	.cmd_per_lun		= 1,
+	.this_id		= -1,
+	.sg_tablesize		= SG_ALL,
+	.max_sectors		= SCSI_DEFAULT_MAX_SECTORS,
+	.use_clustering		= ENABLE_CLUSTERING,
+};
+
+static int __devinit asd_map_memio(struct asd_ha_struct *asd_ha)
+{
+	int err, i;
+	struct asd_ha_addrspace *io_handle;
+
+	asd_ha->iospace = 0;
+	for (i = 0; i < 3; i += 2) {
+		io_handle = &asd_ha->io_handle[i==0?0:1];
+		io_handle->start = pci_resource_start(asd_ha->pcidev, i);
+		io_handle->len   = pci_resource_len(asd_ha->pcidev, i);
+		io_handle->flags = pci_resource_flags(asd_ha->pcidev, i);
+		err = -ENODEV;
+		if (!io_handle->start || !io_handle->len) {
+			asd_printk("MBAR%d start or length for %s is 0.\n",
+				   i==0?0:1, pci_name(asd_ha->pcidev));
+			goto Err;
+		}
+		err = pci_request_region(asd_ha->pcidev, i, ASD_DRIVER_NAME);
+		if (err) {
+			asd_printk("couldn't reserve memory region for %s\n",
+				   pci_name(asd_ha->pcidev));
+			goto Err;
+		}
+		if (io_handle->flags & IORESOURCE_CACHEABLE)
+			io_handle->addr = ioremap(io_handle->start,
+						  io_handle->len);
+		else
+			io_handle->addr = ioremap_nocache(io_handle->start,
+							  io_handle->len);
+		if (!io_handle->addr) {
+			asd_printk("couldn't map MBAR%d of %s\n", i==0?0:1,
+				   pci_name(asd_ha->pcidev));
+			goto Err_unreq;
+		}
+	}
+
+	return 0;
+Err_unreq:
+	pci_release_region(asd_ha->pcidev, i);
+Err:
+	if (i > 0) {
+		io_handle = &asd_ha->io_handle[0];
+		iounmap(io_handle->addr);
+		pci_release_region(asd_ha->pcidev, 0);
+	}
+	return err;
+}
+
+static void __devexit asd_unmap_memio(struct asd_ha_struct *asd_ha)
+{
+	struct asd_ha_addrspace *io_handle;
+
+	io_handle = &asd_ha->io_handle[1];
+	iounmap(io_handle->addr);
+	pci_release_region(asd_ha->pcidev, 2);
+
+	io_handle = &asd_ha->io_handle[0];
+	iounmap(io_handle->addr);
+	pci_release_region(asd_ha->pcidev, 0);
+}
+
+static int __devinit asd_map_ioport(struct asd_ha_struct *asd_ha)
+{
+	int i = PCI_IOBAR_OFFSET, err;
+	struct asd_ha_addrspace *io_handle = &asd_ha->io_handle[0];
+
+	asd_ha->iospace = 1;
+	io_handle->start = pci_resource_start(asd_ha->pcidev, i);
+	io_handle->len   = pci_resource_len(asd_ha->pcidev, i);
+	io_handle->flags = pci_resource_flags(asd_ha->pcidev, i);
+	io_handle->addr  = (void __iomem *) io_handle->start;
+	if (!io_handle->start || !io_handle->len) {
+		asd_printk("couldn't get IO ports for %s\n",
+			   pci_name(asd_ha->pcidev));
+		return -ENODEV;
+	}
+	err = pci_request_region(asd_ha->pcidev, i, ASD_DRIVER_NAME);
+	if (err) {
+		asd_printk("couldn't reserve io space for %s\n",
+			   pci_name(asd_ha->pcidev));
+	}
+
+	return err;
+}
+
+static void __devexit asd_unmap_ioport(struct asd_ha_struct *asd_ha)
+{
+	pci_release_region(asd_ha->pcidev, PCI_IOBAR_OFFSET);
+}
+
+static int __devinit asd_map_ha(struct asd_ha_struct *asd_ha)
+{
+	int err;
+	u16 cmd_reg;
+
+	err = pci_read_config_word(asd_ha->pcidev, PCI_COMMAND, &cmd_reg);
+	if (err) {
+		asd_printk("couldn't read command register of %s\n",
+			   pci_name(asd_ha->pcidev));
+		goto Err;
+	}
+
+	err = -ENODEV;
+	if (cmd_reg & PCI_COMMAND_MEMORY) {
+		if ((err = asd_map_memio(asd_ha)))
+			goto Err;
+	} else if (cmd_reg & PCI_COMMAND_IO) {
+		if ((err = asd_map_ioport(asd_ha)))
+			goto Err;
+		asd_printk("%s ioport mapped -- upgrade your hardware\n",
+			   pci_name(asd_ha->pcidev));
+	} else {
+		asd_printk("no proper device access to %s\n",
+			   pci_name(asd_ha->pcidev));
+		goto Err;
+	}
+
+	return 0;
+Err:
+	return err;
+}
+
+static void __devexit asd_unmap_ha(struct asd_ha_struct *asd_ha)
+{
+	if (asd_ha->iospace)
+		asd_unmap_ioport(asd_ha);
+	else
+		asd_unmap_memio(asd_ha);
+}
+
+static const char *asd_dev_rev[30] = {
+	[0] = "A0",
+	[1] = "A1",
+	[8] = "B0",
+};
+
+static int __devinit asd_common_setup(struct asd_ha_struct *asd_ha)
+{
+	int err, i;
+
+	err = pci_read_config_byte(asd_ha->pcidev, PCI_REVISION_ID,
+				   &asd_ha->revision_id);
+	if (err) {
+		asd_printk("couldn't read REVISION ID register of %s\n",
+			   pci_name(asd_ha->pcidev));
+		goto Err;
+	}
+	err = -ENODEV;
+	if (asd_ha->revision_id < AIC9410_DEV_REV_B0) {
+		asd_printk("%s is revision %s (%X), which is not supported\n",
+			   pci_name(asd_ha->pcidev),
+			   asd_dev_rev[asd_ha->revision_id],
+			   asd_ha->revision_id);
+		goto Err;
+	}
+	/* Provide some sane default values. */
+	asd_ha->hw_prof.max_scbs = 512;
+	asd_ha->hw_prof.max_ddbs = 128;
+	asd_ha->hw_prof.num_phys = ASD_MAX_PHYS;
+	/* All phys are enabled, by default. */
+	asd_ha->hw_prof.enabled_phys = 0xFF;
+	for (i = 0; i < ASD_MAX_PHYS; i++) {
+		asd_ha->hw_prof.phy_desc[i].max_sas_lrate = PHY_LINKRATE_3;
+		asd_ha->hw_prof.phy_desc[i].min_sas_lrate = PHY_LINKRATE_1_5;
+		asd_ha->hw_prof.phy_desc[i].max_sata_lrate= PHY_LINKRATE_1_5;
+		asd_ha->hw_prof.phy_desc[i].min_sata_lrate= PHY_LINKRATE_1_5;
+	}
+
+	return 0;
+Err:
+	return err;
+}
+
+static int __devinit asd_aic9410_setup(struct asd_ha_struct *asd_ha)
+{
+	int err = asd_common_setup(asd_ha);
+
+	if (err)
+		return err;
+
+	asd_ha->hw_prof.addr_range = 8;
+	asd_ha->hw_prof.port_name_base = 0;
+	asd_ha->hw_prof.dev_name_base = 8;
+	asd_ha->hw_prof.sata_name_base = 16;
+
+	return 0;
+}
+
+static int __devinit asd_aic9405_setup(struct asd_ha_struct *asd_ha)
+{
+	int err = asd_common_setup(asd_ha);
+
+	if (err)
+		return err;
+
+	asd_ha->hw_prof.addr_range = 4;
+	asd_ha->hw_prof.port_name_base = 0;
+	asd_ha->hw_prof.dev_name_base = 4;
+	asd_ha->hw_prof.sata_name_base = 8;
+
+	return 0;
+}
+
+static ssize_t asd_show_dev_rev(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct asd_ha_struct *asd_ha = dev_to_asd_ha(dev);
+	return snprintf(buf, PAGE_SIZE, "%s\n",
+			asd_dev_rev[asd_ha->revision_id]);
+}
+static DEVICE_ATTR(revision, S_IRUGO, asd_show_dev_rev, NULL);
+
+static ssize_t asd_show_dev_bios_build(struct device *dev,
+				       struct device_attribute *attr,char *buf)
+{
+	struct asd_ha_struct *asd_ha = dev_to_asd_ha(dev);
+	return snprintf(buf, PAGE_SIZE, "%d\n", asd_ha->hw_prof.bios.bld);
+}
+static DEVICE_ATTR(bios_build, S_IRUGO, asd_show_dev_bios_build, NULL);
+
+static ssize_t asd_show_dev_pcba_sn(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	struct asd_ha_struct *asd_ha = dev_to_asd_ha(dev);
+	return snprintf(buf, PAGE_SIZE, "%s\n", asd_ha->hw_prof.pcba_sn);
+}
+static DEVICE_ATTR(pcba_sn, S_IRUGO, asd_show_dev_pcba_sn, NULL);
+
+static void asd_create_dev_attrs(struct asd_ha_struct *asd_ha)
+{
+	device_create_file(&asd_ha->pcidev->dev, &dev_attr_revision);
+	device_create_file(&asd_ha->pcidev->dev, &dev_attr_bios_build);
+	device_create_file(&asd_ha->pcidev->dev, &dev_attr_pcba_sn);
+}
+
+static void asd_remove_dev_attrs(struct asd_ha_struct *asd_ha)
+{
+	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_revision);
+	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_bios_build);
+	device_remove_file(&asd_ha->pcidev->dev, &dev_attr_pcba_sn);
+}
+
+/* The first entry, 0, is used for dynamic ids, the rest for devices
+ * we know about.
+ */
+static struct asd_pcidev_struct {
+	const char * name;
+	int (*setup)(struct asd_ha_struct *asd_ha);
+} asd_pcidev_data[] = {
+	/* Id 0 is used for dynamic ids. */
+	{ .name  = "Adaptec AIC-94xx SAS/SATA Host Adapter",
+	  .setup = asd_aic9410_setup
+	},
+	{ .name  = "Adaptec AIC-9410W SAS/SATA Host Adapter",
+	  .setup = asd_aic9410_setup
+	},
+	{ .name  = "Adaptec AIC-9405W SAS/SATA Host Adapter",
+	  .setup = asd_aic9405_setup
+	},
+};
+
+static inline int asd_create_ha_caches(struct asd_ha_struct *asd_ha)
+{
+	asd_ha->scb_pool = dma_pool_create(ASD_DRIVER_NAME "_scb_pool",
+					   &asd_ha->pcidev->dev,
+					   sizeof(struct scb),
+					   8, 0);
+	if (!asd_ha->scb_pool) {
+		asd_printk("couldn't create scb pool\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/**
+ * asd_free_edbs -- free empty data buffers
+ * asd_ha: pointer to host adapter structure
+ */
+static inline void asd_free_edbs(struct asd_ha_struct *asd_ha)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int i;
+
+	for (i = 0; i < seq->num_edbs; i++)
+		asd_free_coherent(asd_ha, seq->edb_arr[i]);
+	kfree(seq->edb_arr);
+	seq->edb_arr = NULL;
+}
+
+static inline void asd_free_escbs(struct asd_ha_struct *asd_ha)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int i;
+
+	for (i = 0; i < seq->num_escbs; i++) {
+		if (!list_empty(&seq->escb_arr[i]->list))
+			list_del_init(&seq->escb_arr[i]->list);
+
+		asd_ascb_free(seq->escb_arr[i]);
+	}
+	kfree(seq->escb_arr);
+	seq->escb_arr = NULL;
+}
+
+static inline void asd_destroy_ha_caches(struct asd_ha_struct *asd_ha)
+{
+	int i;
+
+	if (asd_ha->hw_prof.ddb_ext)
+		asd_free_coherent(asd_ha, asd_ha->hw_prof.ddb_ext);
+	if (asd_ha->hw_prof.scb_ext)
+		asd_free_coherent(asd_ha, asd_ha->hw_prof.scb_ext);
+
+	if (asd_ha->hw_prof.ddb_bitmap)
+		kfree(asd_ha->hw_prof.ddb_bitmap);
+	asd_ha->hw_prof.ddb_bitmap = NULL;
+
+	for (i = 0; i < ASD_MAX_PHYS; i++) {
+		struct asd_phy *phy = &asd_ha->phys[i];
+
+		asd_free_coherent(asd_ha, phy->id_frm_tok);
+	}
+	if (asd_ha->seq.escb_arr)
+		asd_free_escbs(asd_ha);
+	if (asd_ha->seq.edb_arr)
+		asd_free_edbs(asd_ha);
+	if (asd_ha->hw_prof.ue.area) {
+		kfree(asd_ha->hw_prof.ue.area);
+		asd_ha->hw_prof.ue.area = NULL;
+	}
+	if (asd_ha->seq.tc_index_array) {
+		kfree(asd_ha->seq.tc_index_array);
+		kfree(asd_ha->seq.tc_index_bitmap);
+		asd_ha->seq.tc_index_array = NULL;
+		asd_ha->seq.tc_index_bitmap = NULL;
+	}
+	if (asd_ha->seq.actual_dl) {
+			asd_free_coherent(asd_ha, asd_ha->seq.actual_dl);
+			asd_ha->seq.actual_dl = NULL;
+			asd_ha->seq.dl = NULL;
+	}
+	if (asd_ha->seq.next_scb.vaddr) {
+		dma_pool_free(asd_ha->scb_pool, asd_ha->seq.next_scb.vaddr,
+			      asd_ha->seq.next_scb.dma_handle);
+		asd_ha->seq.next_scb.vaddr = NULL;
+	}
+	dma_pool_destroy(asd_ha->scb_pool);
+	asd_ha->scb_pool = NULL;
+}
+
+kmem_cache_t *asd_dma_token_cache;
+kmem_cache_t *asd_ascb_cache;
+
+static int asd_create_global_caches(void)
+{
+	if (!asd_dma_token_cache) {
+		asd_dma_token_cache
+			= kmem_cache_create(ASD_DRIVER_NAME "_dma_token",
+					    sizeof(struct asd_dma_tok),
+					    0,
+					    SLAB_HWCACHE_ALIGN,
+					    NULL, NULL);
+		if (!asd_dma_token_cache) {
+			asd_printk("couldn't create dma token cache\n");
+			return -ENOMEM;
+		}
+	}
+
+	if (!asd_ascb_cache) {
+		asd_ascb_cache = kmem_cache_create(ASD_DRIVER_NAME "_ascb",
+						   sizeof(struct asd_ascb),
+						   0,
+						   SLAB_HWCACHE_ALIGN,
+						   NULL, NULL);
+		if (!asd_ascb_cache) {
+			asd_printk("couldn't create ascb cache\n");
+			goto Err;
+		}
+	}
+
+	return 0;
+Err:
+	kmem_cache_destroy(asd_dma_token_cache);
+	asd_dma_token_cache = NULL;
+	return -ENOMEM;
+}
+
+static void asd_destroy_global_caches(void)
+{
+	if (asd_dma_token_cache)
+		kmem_cache_destroy(asd_dma_token_cache);
+	asd_dma_token_cache = NULL;
+
+	if (asd_ascb_cache)
+		kmem_cache_destroy(asd_ascb_cache);
+	asd_ascb_cache = NULL;
+}
+
+static int asd_register_sas_ha(struct asd_ha_struct *asd_ha)
+{
+	int i;
+	struct asd_sas_phy   **sas_phys =
+		kmalloc(ASD_MAX_PHYS * sizeof(struct asd_sas_phy), GFP_KERNEL);
+	struct asd_sas_port  **sas_ports =
+		kmalloc(ASD_MAX_PHYS * sizeof(struct asd_sas_port), GFP_KERNEL);
+
+	if (!sas_phys || !sas_ports) {
+		kfree(sas_phys);
+		kfree(sas_ports);
+		return -ENOMEM;
+	}
+
+	asd_ha->sas_ha.sas_ha_name = (char *) asd_ha->name;
+	asd_ha->sas_ha.lldd_module = THIS_MODULE;
+	asd_ha->sas_ha.sas_addr = &asd_ha->hw_prof.sas_addr[0];
+
+	for (i = 0; i < ASD_MAX_PHYS; i++) {
+		sas_phys[i] = &asd_ha->phys[i].sas_phy;
+		sas_ports[i] = &asd_ha->ports[i];
+	}
+
+	asd_ha->sas_ha.sas_phy = sas_phys;
+	asd_ha->sas_ha.sas_port= sas_ports;
+	asd_ha->sas_ha.num_phys= ASD_MAX_PHYS;
+
+	asd_ha->sas_ha.lldd_queue_size = asd_ha->seq.can_queue;
+
+	return sas_register_ha(&asd_ha->sas_ha);
+}
+
+static int asd_unregister_sas_ha(struct asd_ha_struct *asd_ha)
+{
+	int err;
+
+	err = sas_unregister_ha(&asd_ha->sas_ha);
+
+	sas_remove_host(asd_ha->sas_ha.core.shost);
+	scsi_remove_host(asd_ha->sas_ha.core.shost);
+	scsi_host_put(asd_ha->sas_ha.core.shost);
+
+	kfree(asd_ha->sas_ha.sas_phy);
+	kfree(asd_ha->sas_ha.sas_port);
+
+	return err;
+}
+
+static int __devinit asd_pci_probe(struct pci_dev *dev,
+				   const struct pci_device_id *id)
+{
+	struct asd_pcidev_struct *asd_dev;
+	unsigned asd_id = (unsigned) id->driver_data;
+	struct asd_ha_struct *asd_ha;
+	struct Scsi_Host *shost;
+	int err;
+
+	if (asd_id >= ARRAY_SIZE(asd_pcidev_data)) {
+		asd_printk("wrong driver_data in PCI table\n");
+		return -ENODEV;
+	}
+
+	if ((err = pci_enable_device(dev))) {
+		asd_printk("couldn't enable device %s\n", pci_name(dev));
+		return err;
+	}
+
+	pci_set_master(dev);
+
+	err = -ENOMEM;
+
+	shost = scsi_host_alloc(&aic94xx_sht, sizeof(void *));
+	if (!shost)
+		goto Err;
+
+	asd_dev = &asd_pcidev_data[asd_id];
+
+	asd_ha = kzalloc(sizeof(*asd_ha), GFP_KERNEL);
+	if (!asd_ha) {
+		asd_printk("out of memory\n");
+		goto Err;
+	}
+	asd_ha->pcidev = dev;
+	asd_ha->sas_ha.pcidev = asd_ha->pcidev;
+	asd_ha->sas_ha.lldd_ha = asd_ha;
+
+	asd_ha->name = asd_dev->name;
+	asd_printk("found %s, device %s\n", asd_ha->name, pci_name(dev));
+
+	SHOST_TO_SAS_HA(shost) = &asd_ha->sas_ha;
+	asd_ha->sas_ha.core.shost = shost;
+	shost->transportt = aic94xx_transport_template;
+	shost->max_id = ~0;
+	shost->max_lun = ~0;
+	shost->max_cmd_len = 16;
+
+	err = scsi_add_host(shost, &dev->dev);
+	if (err) {
+		scsi_host_put(shost);
+		goto Err_free;
+	}
+
+
+
+	err = asd_dev->setup(asd_ha);
+	if (err)
+		goto Err_free;
+
+	err = -ENODEV;
+	if (!pci_set_dma_mask(dev, DMA_64BIT_MASK)
+	    && !pci_set_consistent_dma_mask(dev, DMA_64BIT_MASK))
+		;
+	else if (!pci_set_dma_mask(dev, DMA_32BIT_MASK)
+		 && !pci_set_consistent_dma_mask(dev, DMA_32BIT_MASK))
+		;
+	else {
+		asd_printk("no suitable DMA mask for %s\n", pci_name(dev));
+		goto Err_free;
+	}
+
+	pci_set_drvdata(dev, asd_ha);
+
+	err = asd_map_ha(asd_ha);
+	if (err)
+		goto Err_free;
+
+	err = asd_create_ha_caches(asd_ha);
+        if (err)
+		goto Err_unmap;
+
+	err = asd_init_hw(asd_ha);
+	if (err)
+		goto Err_free_cache;
+
+	asd_printk("device %s: SAS addr %llx, PCBA SN %s, %d phys, %d enabled "
+		   "phys, flash %s, BIOS %s%d\n",
+		   pci_name(dev), SAS_ADDR(asd_ha->hw_prof.sas_addr),
+		   asd_ha->hw_prof.pcba_sn, asd_ha->hw_prof.max_phys,
+		   asd_ha->hw_prof.num_phys,
+		   asd_ha->hw_prof.flash.present ? "present" : "not present",
+		   asd_ha->hw_prof.bios.present ? "build " : "not present",
+		   asd_ha->hw_prof.bios.bld);
+
+	if (use_msi)
+		pci_enable_msi(asd_ha->pcidev);
+
+	err = request_irq(asd_ha->pcidev->irq, asd_hw_isr, SA_SHIRQ,
+			  ASD_DRIVER_NAME, asd_ha);
+	if (err) {
+		asd_printk("couldn't get irq %d for %s\n",
+			   asd_ha->pcidev->irq, pci_name(asd_ha->pcidev));
+		goto Err_irq;
+	}
+	asd_enable_ints(asd_ha);
+
+	err = asd_init_post_escbs(asd_ha);
+	if (err) {
+		asd_printk("couldn't post escbs for %s\n",
+			   pci_name(asd_ha->pcidev));
+		goto Err_escbs;
+	}
+	ASD_DPRINTK("escbs posted\n");
+
+	asd_create_dev_attrs(asd_ha);
+
+	err = asd_register_sas_ha(asd_ha);
+	if (err)
+		goto Err_reg_sas;
+
+	err = asd_enable_phys(asd_ha, asd_ha->hw_prof.enabled_phys);
+	if (err) {
+		asd_printk("coudln't enable phys, err:%d\n", err);
+		goto Err_en_phys;
+	}
+	ASD_DPRINTK("enabled phys\n");
+	/* give the phy enabling interrupt event time to come in (1s
+	 * is empirically about all it takes) */
+	ssleep(1);
+	/* Wait for discovery to finish */
+	scsi_flush_work(asd_ha->sas_ha.core.shost);
+
+	return 0;
+Err_en_phys:
+	asd_unregister_sas_ha(asd_ha);
+Err_reg_sas:
+	asd_remove_dev_attrs(asd_ha);
+Err_escbs:
+	asd_disable_ints(asd_ha);
+	free_irq(dev->irq, asd_ha);
+Err_irq:
+	if (use_msi)
+		pci_disable_msi(dev);
+	asd_chip_hardrst(asd_ha);
+Err_free_cache:
+	asd_destroy_ha_caches(asd_ha);
+Err_unmap:
+	asd_unmap_ha(asd_ha);
+Err_free:
+	kfree(asd_ha);
+	scsi_remove_host(shost);
+Err:
+	pci_disable_device(dev);
+	return err;
+}
+
+static void asd_free_queues(struct asd_ha_struct *asd_ha)
+{
+	unsigned long flags;
+	LIST_HEAD(pending);
+	struct list_head *n, *pos;
+
+	spin_lock_irqsave(&asd_ha->seq.pend_q_lock, flags);
+	asd_ha->seq.pending = 0;
+	list_splice_init(&asd_ha->seq.pend_q, &pending);
+	spin_unlock_irqrestore(&asd_ha->seq.pend_q_lock, flags);
+
+	if (!list_empty(&pending))
+		ASD_DPRINTK("Uh-oh! Pending is not empty!\n");
+
+	list_for_each_safe(pos, n, &pending) {
+		struct asd_ascb *ascb = list_entry(pos, struct asd_ascb, list);
+		list_del_init(pos);
+		ASD_DPRINTK("freeing from pending\n");
+		asd_ascb_free(ascb);
+	}
+}
+
+static void asd_turn_off_leds(struct asd_ha_struct *asd_ha)
+{
+	u8 phy_mask = asd_ha->hw_prof.enabled_phys;
+	u8 i;
+
+	for_each_phy(phy_mask, phy_mask, i) {
+		asd_turn_led(asd_ha, i, 0);
+		asd_control_led(asd_ha, i, 0);
+	}
+}
+
+static void __devexit asd_pci_remove(struct pci_dev *dev)
+{
+	struct asd_ha_struct *asd_ha = pci_get_drvdata(dev);
+
+	if (!asd_ha)
+		return;
+
+	asd_unregister_sas_ha(asd_ha);
+
+	asd_disable_ints(asd_ha);
+
+	asd_remove_dev_attrs(asd_ha);
+
+	/* XXX more here as needed */
+
+	free_irq(dev->irq, asd_ha);
+	if (use_msi)
+		pci_disable_msi(asd_ha->pcidev);
+	asd_turn_off_leds(asd_ha);
+	asd_chip_hardrst(asd_ha);
+	asd_free_queues(asd_ha);
+	asd_destroy_ha_caches(asd_ha);
+	asd_unmap_ha(asd_ha);
+	kfree(asd_ha);
+	pci_disable_device(dev);
+	return;
+}
+
+static ssize_t asd_version_show(struct device_driver *driver, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%s\n", ASD_DRIVER_VERSION);
+}
+static DRIVER_ATTR(version, S_IRUGO, asd_version_show, NULL);
+
+static void asd_create_driver_attrs(struct device_driver *driver)
+{
+	driver_create_file(driver, &driver_attr_version);
+}
+
+static void asd_remove_driver_attrs(struct device_driver *driver)
+{
+	driver_remove_file(driver, &driver_attr_version);
+}
+
+static struct sas_domain_function_template aic94xx_transport_functions = {
+	.lldd_port_formed	= asd_update_port_links,
+
+	.lldd_dev_found		= asd_dev_found,
+	.lldd_dev_gone		= asd_dev_gone,
+
+	.lldd_execute_task	= asd_execute_task,
+
+	.lldd_abort_task	= asd_abort_task,
+	.lldd_abort_task_set	= asd_abort_task_set,
+	.lldd_clear_aca		= asd_clear_aca,
+	.lldd_clear_task_set	= asd_clear_task_set,
+	.lldd_I_T_nexus_reset	= NULL,
+	.lldd_lu_reset		= asd_lu_reset,
+	.lldd_query_task	= asd_query_task,
+
+	.lldd_clear_nexus_port	= asd_clear_nexus_port,
+	.lldd_clear_nexus_ha	= asd_clear_nexus_ha,
+
+	.lldd_control_phy	= asd_control_phy,
+};
+
+static const struct pci_device_id aic94xx_pci_table[] __devinitdata = {
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR10),
+	 0, 0, 1},
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR12),
+	 0, 0, 1},
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR1E),
+	 0, 0, 1},
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR30),
+	 0, 0, 2},
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR32),
+	 0, 0, 2},
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR3E),
+	 0, 0, 2},
+	{PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR3F),
+	 0, 0, 2},
+	{}
+};
+
+MODULE_DEVICE_TABLE(pci, aic94xx_pci_table);
+
+static struct pci_driver aic94xx_pci_driver = {
+	.name		= ASD_DRIVER_NAME,
+	.id_table	= aic94xx_pci_table,
+	.probe		= asd_pci_probe,
+	.remove		= __devexit_p(asd_pci_remove),
+};
+
+static int __init aic94xx_init(void)
+{
+	int err;
+
+
+	asd_printk("%s version %s loaded\n", ASD_DRIVER_DESCRIPTION,
+		   ASD_DRIVER_VERSION);
+
+	err = asd_create_global_caches();
+	if (err)
+		return err;
+
+	aic94xx_transport_template =
+		sas_domain_attach_transport(&aic94xx_transport_functions);
+	if (err)
+		goto out_destroy_caches;
+
+	err = pci_register_driver(&aic94xx_pci_driver);
+	if (err)
+		goto out_release_transport;
+
+	asd_create_driver_attrs(&aic94xx_pci_driver.driver);
+
+	return err;
+
+ out_release_transport:
+	sas_release_transport(aic94xx_transport_template);
+ out_destroy_caches:
+	asd_destroy_global_caches();
+
+	return err;
+}
+
+static void __exit aic94xx_exit(void)
+{
+	asd_remove_driver_attrs(&aic94xx_pci_driver.driver);
+	pci_unregister_driver(&aic94xx_pci_driver);
+	sas_release_transport(aic94xx_transport_template);
+	asd_destroy_global_caches();
+	asd_printk("%s version %s unloaded\n", ASD_DRIVER_DESCRIPTION,
+		   ASD_DRIVER_VERSION);
+}
+
+module_init(aic94xx_init);
+module_exit(aic94xx_exit);
+
+MODULE_AUTHOR("Luben Tuikov <luben_tuikov@adaptec.com>");
+MODULE_DESCRIPTION(ASD_DRIVER_DESCRIPTION);
+MODULE_LICENSE("GPL v2");
+MODULE_VERSION(ASD_DRIVER_VERSION);
diff --git a/drivers/scsi/aic94xx/aic94xx_reg.c b/drivers/scsi/aic94xx/aic94xx_reg.c
new file mode 100644
index 000000000000..f210dac3203d
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_reg.c
@@ -0,0 +1,332 @@
+/*
+ * Aic94xx SAS/SATA driver register access.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/pci.h>
+#include "aic94xx_reg.h"
+#include "aic94xx.h"
+
+/* Writing to device address space.
+ * Offset comes before value to remind that the operation of
+ * this function is *offs = val.
+ */
+static inline void asd_write_byte(struct asd_ha_struct *asd_ha,
+				  unsigned long offs, u8 val)
+{
+	if (unlikely(asd_ha->iospace))
+		outb(val,
+		     (unsigned long)asd_ha->io_handle[0].addr + (offs & 0xFF));
+	else
+		writeb(val, asd_ha->io_handle[0].addr + offs);
+	wmb();
+}
+
+static inline void asd_write_word(struct asd_ha_struct *asd_ha,
+				  unsigned long offs, u16 val)
+{
+	if (unlikely(asd_ha->iospace))
+		outw(val,
+		     (unsigned long)asd_ha->io_handle[0].addr + (offs & 0xFF));
+	else
+		writew(val, asd_ha->io_handle[0].addr + offs);
+	wmb();
+}
+
+static inline void asd_write_dword(struct asd_ha_struct *asd_ha,
+				   unsigned long offs, u32 val)
+{
+	if (unlikely(asd_ha->iospace))
+		outl(val,
+		     (unsigned long)asd_ha->io_handle[0].addr + (offs & 0xFF));
+	else
+		writel(val, asd_ha->io_handle[0].addr + offs);
+	wmb();
+}
+
+/* Reading from device address space.
+ */
+static inline u8 asd_read_byte(struct asd_ha_struct *asd_ha,
+			       unsigned long offs)
+{
+	u8 val;
+	if (unlikely(asd_ha->iospace))
+		val = inb((unsigned long) asd_ha->io_handle[0].addr
+			  + (offs & 0xFF));
+	else
+		val = readb(asd_ha->io_handle[0].addr + offs);
+	rmb();
+	return val;
+}
+
+static inline u16 asd_read_word(struct asd_ha_struct *asd_ha,
+				unsigned long offs)
+{
+	u16 val;
+	if (unlikely(asd_ha->iospace))
+		val = inw((unsigned long)asd_ha->io_handle[0].addr
+			  + (offs & 0xFF));
+	else
+		val = readw(asd_ha->io_handle[0].addr + offs);
+	rmb();
+	return val;
+}
+
+static inline u32 asd_read_dword(struct asd_ha_struct *asd_ha,
+				 unsigned long offs)
+{
+	u32 val;
+	if (unlikely(asd_ha->iospace))
+		val = inl((unsigned long) asd_ha->io_handle[0].addr
+			  + (offs & 0xFF));
+	else
+		val = readl(asd_ha->io_handle[0].addr + offs);
+	rmb();
+	return val;
+}
+
+static inline u32 asd_mem_offs_swa(void)
+{
+	return 0;
+}
+
+static inline u32 asd_mem_offs_swc(void)
+{
+	return asd_mem_offs_swa() + MBAR0_SWA_SIZE;
+}
+
+static inline u32 asd_mem_offs_swb(void)
+{
+	return asd_mem_offs_swc() + MBAR0_SWC_SIZE + 0x20;
+}
+
+/* We know that the register wanted is in the range
+ * of the sliding window.
+ */
+#define ASD_READ_SW(ww, type, ord)                                     \
+static inline type asd_read_##ww##_##ord (struct asd_ha_struct *asd_ha,\
+					  u32 reg)                     \
+{                                                                      \
+	struct asd_ha_addrspace *io_handle = &asd_ha->io_handle[0];    \
+	u32 map_offs=(reg - io_handle-> ww##_base )+asd_mem_offs_##ww ();\
+	return asd_read_##ord (asd_ha, (unsigned long) map_offs);      \
+}
+
+#define ASD_WRITE_SW(ww, type, ord)                                    \
+static inline void asd_write_##ww##_##ord (struct asd_ha_struct *asd_ha,\
+				  u32 reg, type val)                   \
+{                                                                      \
+	struct asd_ha_addrspace *io_handle = &asd_ha->io_handle[0];    \
+	u32 map_offs=(reg - io_handle-> ww##_base )+asd_mem_offs_##ww ();\
+	asd_write_##ord (asd_ha, (unsigned long) map_offs, val);       \
+}
+
+ASD_READ_SW(swa, u8,  byte);
+ASD_READ_SW(swa, u16, word);
+ASD_READ_SW(swa, u32, dword);
+
+ASD_READ_SW(swb, u8,  byte);
+ASD_READ_SW(swb, u16, word);
+ASD_READ_SW(swb, u32, dword);
+
+ASD_READ_SW(swc, u8,  byte);
+ASD_READ_SW(swc, u16, word);
+ASD_READ_SW(swc, u32, dword);
+
+ASD_WRITE_SW(swa, u8,  byte);
+ASD_WRITE_SW(swa, u16, word);
+ASD_WRITE_SW(swa, u32, dword);
+
+ASD_WRITE_SW(swb, u8,  byte);
+ASD_WRITE_SW(swb, u16, word);
+ASD_WRITE_SW(swb, u32, dword);
+
+ASD_WRITE_SW(swc, u8,  byte);
+ASD_WRITE_SW(swc, u16, word);
+ASD_WRITE_SW(swc, u32, dword);
+
+/*
+ * A word about sliding windows:
+ * MBAR0 is divided into sliding windows A, C and B, in that order.
+ * SWA starts at offset 0 of MBAR0, up to 0x57, with size 0x58 bytes.
+ * SWC starts at offset 0x58 of MBAR0, up to 0x60, with size 0x8 bytes.
+ * From 0x60 to 0x7F, we have a copy of PCI config space 0x60-0x7F.
+ * SWB starts at offset 0x80 of MBAR0 and extends to the end of MBAR0.
+ * See asd_init_sw() in aic94xx_hwi.c
+ *
+ * We map the most common registers we'd access of the internal 4GB
+ * host adapter memory space.  If a register/internal memory location
+ * is wanted which is not mapped, we slide SWB, by paging it,
+ * see asd_move_swb() in aic94xx_reg.c.
+ */
+
+/**
+ * asd_move_swb -- move sliding window B
+ * @asd_ha: pointer to host adapter structure
+ * @reg: register desired to be within range of the new window
+ */
+static inline void asd_move_swb(struct asd_ha_struct *asd_ha, u32 reg)
+{
+	u32 base = reg & ~(MBAR0_SWB_SIZE-1);
+	pci_write_config_dword(asd_ha->pcidev, PCI_CONF_MBAR0_SWB, base);
+	asd_ha->io_handle[0].swb_base = base;
+}
+
+static void __asd_write_reg_byte(struct asd_ha_struct *asd_ha, u32 reg, u8 val)
+{
+	struct asd_ha_addrspace *io_handle=&asd_ha->io_handle[0];
+	BUG_ON(reg >= 0xC0000000 || reg < ALL_BASE_ADDR);
+	if (io_handle->swa_base <= reg
+	    && reg < io_handle->swa_base + MBAR0_SWA_SIZE)
+		asd_write_swa_byte (asd_ha, reg,val);
+	else if (io_handle->swb_base <= reg
+		 && reg < io_handle->swb_base + MBAR0_SWB_SIZE)
+		asd_write_swb_byte (asd_ha, reg, val);
+	else if (io_handle->swc_base <= reg
+		 && reg < io_handle->swc_base + MBAR0_SWC_SIZE)
+		asd_write_swc_byte (asd_ha, reg, val);
+	else {
+		/* Ok, we have to move SWB */
+		asd_move_swb(asd_ha, reg);
+		asd_write_swb_byte (asd_ha, reg, val);
+	}
+}
+
+#define ASD_WRITE_REG(type, ord)                                  \
+void asd_write_reg_##ord (struct asd_ha_struct *asd_ha, u32 reg, type val)\
+{                                                                 \
+	struct asd_ha_addrspace *io_handle=&asd_ha->io_handle[0]; \
+	unsigned long flags;                                      \
+	BUG_ON(reg >= 0xC0000000 || reg < ALL_BASE_ADDR);         \
+	spin_lock_irqsave(&asd_ha->iolock, flags);                \
+	if (io_handle->swa_base <= reg                            \
+	    && reg < io_handle->swa_base + MBAR0_SWA_SIZE)        \
+		asd_write_swa_##ord (asd_ha, reg,val);            \
+	else if (io_handle->swb_base <= reg                       \
+		 && reg < io_handle->swb_base + MBAR0_SWB_SIZE)   \
+		asd_write_swb_##ord (asd_ha, reg, val);           \
+	else if (io_handle->swc_base <= reg                       \
+		 && reg < io_handle->swc_base + MBAR0_SWC_SIZE)   \
+		asd_write_swc_##ord (asd_ha, reg, val);           \
+	else {                                                    \
+		/* Ok, we have to move SWB */                     \
+		asd_move_swb(asd_ha, reg);                        \
+		asd_write_swb_##ord (asd_ha, reg, val);           \
+	}                                                         \
+	spin_unlock_irqrestore(&asd_ha->iolock, flags);           \
+}
+
+ASD_WRITE_REG(u8, byte);
+ASD_WRITE_REG(u16,word);
+ASD_WRITE_REG(u32,dword);
+
+static u8 __asd_read_reg_byte(struct asd_ha_struct *asd_ha, u32 reg)
+{
+	struct asd_ha_addrspace *io_handle=&asd_ha->io_handle[0];
+	u8 val;
+	BUG_ON(reg >= 0xC0000000 || reg < ALL_BASE_ADDR);
+	if (io_handle->swa_base <= reg
+	    && reg < io_handle->swa_base + MBAR0_SWA_SIZE)
+		val = asd_read_swa_byte (asd_ha, reg);
+	else if (io_handle->swb_base <= reg
+		 && reg < io_handle->swb_base + MBAR0_SWB_SIZE)
+		val = asd_read_swb_byte (asd_ha, reg);
+	else if (io_handle->swc_base <= reg
+		 && reg < io_handle->swc_base + MBAR0_SWC_SIZE)
+		val = asd_read_swc_byte (asd_ha, reg);
+	else {
+		/* Ok, we have to move SWB */
+		asd_move_swb(asd_ha, reg);
+		val = asd_read_swb_byte (asd_ha, reg);
+	}
+	return val;
+}
+
+#define ASD_READ_REG(type, ord)                                   \
+type asd_read_reg_##ord (struct asd_ha_struct *asd_ha, u32 reg)   \
+{                                                                 \
+	struct asd_ha_addrspace *io_handle=&asd_ha->io_handle[0]; \
+	type val;                                                 \
+	unsigned long flags;                                      \
+	BUG_ON(reg >= 0xC0000000 || reg < ALL_BASE_ADDR);         \
+	spin_lock_irqsave(&asd_ha->iolock, flags);                \
+	if (io_handle->swa_base <= reg                            \
+	    && reg < io_handle->swa_base + MBAR0_SWA_SIZE)        \
+		val = asd_read_swa_##ord (asd_ha, reg);           \
+	else if (io_handle->swb_base <= reg                       \
+		 && reg < io_handle->swb_base + MBAR0_SWB_SIZE)   \
+		val = asd_read_swb_##ord (asd_ha, reg);           \
+	else if (io_handle->swc_base <= reg                       \
+		 && reg < io_handle->swc_base + MBAR0_SWC_SIZE)   \
+		val = asd_read_swc_##ord (asd_ha, reg);           \
+	else {                                                    \
+		/* Ok, we have to move SWB */                     \
+		asd_move_swb(asd_ha, reg);                        \
+		val = asd_read_swb_##ord (asd_ha, reg);           \
+	}                                                         \
+	spin_unlock_irqrestore(&asd_ha->iolock, flags);           \
+	return val;                                               \
+}
+
+ASD_READ_REG(u8, byte);
+ASD_READ_REG(u16,word);
+ASD_READ_REG(u32,dword);
+
+/**
+ * asd_read_reg_string -- read a string of bytes from io space memory
+ * @asd_ha: pointer to host adapter structure
+ * @dst: pointer to a destination buffer where data will be written to
+ * @offs: start offset (register) to read from
+ * @count: number of bytes to read
+ */
+void asd_read_reg_string(struct asd_ha_struct *asd_ha, void *dst,
+			 u32 offs, int count)
+{
+	u8 *p = dst;
+	unsigned long flags;
+
+	spin_lock_irqsave(&asd_ha->iolock, flags);
+	for ( ; count > 0; count--, offs++, p++)
+		*p = __asd_read_reg_byte(asd_ha, offs);
+	spin_unlock_irqrestore(&asd_ha->iolock, flags);
+}
+
+/**
+ * asd_write_reg_string -- write a string of bytes to io space memory
+ * @asd_ha: pointer to host adapter structure
+ * @src: pointer to source buffer where data will be read from
+ * @offs: start offset (register) to write to
+ * @count: number of bytes to write
+ */
+void asd_write_reg_string(struct asd_ha_struct *asd_ha, void *src,
+			  u32 offs, int count)
+{
+	u8 *p = src;
+	unsigned long flags;
+
+	spin_lock_irqsave(&asd_ha->iolock, flags);
+	for ( ; count > 0; count--, offs++, p++)
+		__asd_write_reg_byte(asd_ha, offs, *p);
+	spin_unlock_irqrestore(&asd_ha->iolock, flags);
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_reg.h b/drivers/scsi/aic94xx/aic94xx_reg.h
new file mode 100644
index 000000000000..2279307fd27e
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_reg.h
@@ -0,0 +1,302 @@
+/*
+ * Aic94xx SAS/SATA driver hardware registers definitions.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#ifndef _AIC94XX_REG_H_
+#define _AIC94XX_REG_H_
+
+#include <asm/io.h>
+#include "aic94xx_hwi.h"
+
+/* Values */
+#define AIC9410_DEV_REV_B0            0x8
+
+/* MBAR0, SWA, SWB, SWC, internal memory space addresses */
+#define REG_BASE_ADDR                 0xB8000000
+#define REG_BASE_ADDR_CSEQCIO         0xB8002000
+#define REG_BASE_ADDR_EXSI            0xB8042800
+
+#define MBAR0_SWA_SIZE                0x58
+extern  u32    MBAR0_SWB_SIZE;
+#define MBAR0_SWC_SIZE                0x8
+
+/* MBAR1, points to On Chip Memory */
+#define OCM_BASE_ADDR                 0xA0000000
+#define OCM_MAX_SIZE                  0x20000
+
+/* Smallest address possible to reference */
+#define ALL_BASE_ADDR                 OCM_BASE_ADDR
+
+/* PCI configuration space registers */
+#define PCI_IOBAR_OFFSET              4
+
+#define PCI_CONF_MBAR1                0x6C
+#define PCI_CONF_MBAR0_SWA            0x70
+#define PCI_CONF_MBAR0_SWB            0x74
+#define PCI_CONF_MBAR0_SWC            0x78
+#define PCI_CONF_MBAR_KEY             0x7C
+#define PCI_CONF_FLSH_BAR             0xB8
+
+#include "aic94xx_reg_def.h"
+
+u8  asd_read_reg_byte(struct asd_ha_struct *asd_ha, u32 reg);
+u16 asd_read_reg_word(struct asd_ha_struct *asd_ha, u32 reg);
+u32 asd_read_reg_dword(struct asd_ha_struct *asd_ha, u32 reg);
+
+void asd_write_reg_byte(struct asd_ha_struct *asd_ha, u32 reg, u8 val);
+void asd_write_reg_word(struct asd_ha_struct *asd_ha, u32 reg, u16 val);
+void asd_write_reg_dword(struct asd_ha_struct *asd_ha, u32 reg, u32 val);
+
+void asd_read_reg_string(struct asd_ha_struct *asd_ha, void *dst,
+			 u32 offs, int count);
+void asd_write_reg_string(struct asd_ha_struct *asd_ha, void *src,
+			  u32 offs, int count);
+
+#define ASD_READ_OCM(type, ord, S)                                    \
+static inline type asd_read_ocm_##ord (struct asd_ha_struct *asd_ha,  \
+					 u32 offs)                    \
+{                                                                     \
+	struct asd_ha_addrspace *io_handle = &asd_ha->io_handle[1];   \
+	type val = read##S (io_handle->addr + (unsigned long) offs);  \
+	rmb();                                                        \
+	return val;                                                   \
+}
+
+ASD_READ_OCM(u8, byte, b);
+ASD_READ_OCM(u16,word, w);
+ASD_READ_OCM(u32,dword,l);
+
+#define ASD_WRITE_OCM(type, ord, S)                                    \
+static inline void asd_write_ocm_##ord (struct asd_ha_struct *asd_ha,  \
+					 u32 offs, type val)          \
+{                                                                     \
+	struct asd_ha_addrspace *io_handle = &asd_ha->io_handle[1];   \
+	write##S (val, io_handle->addr + (unsigned long) offs);       \
+	return;                                                       \
+}
+
+ASD_WRITE_OCM(u8, byte, b);
+ASD_WRITE_OCM(u16,word, w);
+ASD_WRITE_OCM(u32,dword,l);
+
+#define ASD_DDBSITE_READ(type, ord)                                        \
+static inline type asd_ddbsite_read_##ord (struct asd_ha_struct *asd_ha,   \
+					   u16 ddb_site_no,                \
+					   u16 offs)                       \
+{                                                                          \
+	asd_write_reg_word(asd_ha, ALTCIOADR, MnDDB_SITE + offs);          \
+	asd_write_reg_word(asd_ha, ADDBPTR, ddb_site_no);                  \
+	return asd_read_reg_##ord (asd_ha, CTXACCESS);                     \
+}
+
+ASD_DDBSITE_READ(u32, dword);
+ASD_DDBSITE_READ(u16, word);
+
+static inline u8 asd_ddbsite_read_byte(struct asd_ha_struct *asd_ha,
+				       u16 ddb_site_no,
+				       u16 offs)
+{
+	if (offs & 1)
+		return asd_ddbsite_read_word(asd_ha, ddb_site_no,
+					     offs & ~1) >> 8;
+	else
+		return asd_ddbsite_read_word(asd_ha, ddb_site_no,
+					     offs) & 0xFF;
+}
+
+
+#define ASD_DDBSITE_WRITE(type, ord)                                       \
+static inline void asd_ddbsite_write_##ord (struct asd_ha_struct *asd_ha,  \
+					u16 ddb_site_no,                   \
+					u16 offs, type val)                \
+{                                                                          \
+	asd_write_reg_word(asd_ha, ALTCIOADR, MnDDB_SITE + offs);          \
+	asd_write_reg_word(asd_ha, ADDBPTR, ddb_site_no);                  \
+	asd_write_reg_##ord (asd_ha, CTXACCESS, val);                      \
+}
+
+ASD_DDBSITE_WRITE(u32, dword);
+ASD_DDBSITE_WRITE(u16, word);
+
+static inline void asd_ddbsite_write_byte(struct asd_ha_struct *asd_ha,
+					  u16 ddb_site_no,
+					  u16 offs, u8 val)
+{
+	u16 base = offs & ~1;
+	u16 rval = asd_ddbsite_read_word(asd_ha, ddb_site_no, base);
+	if (offs & 1)
+		rval = (val << 8) | (rval & 0xFF);
+	else
+		rval = (rval & 0xFF00) | val;
+	asd_ddbsite_write_word(asd_ha, ddb_site_no, base, rval);
+}
+
+
+#define ASD_SCBSITE_READ(type, ord)                                        \
+static inline type asd_scbsite_read_##ord (struct asd_ha_struct *asd_ha,   \
+					   u16 scb_site_no,                \
+					   u16 offs)                       \
+{                                                                          \
+	asd_write_reg_word(asd_ha, ALTCIOADR, MnSCB_SITE + offs);          \
+	asd_write_reg_word(asd_ha, ASCBPTR, scb_site_no);                  \
+	return asd_read_reg_##ord (asd_ha, CTXACCESS);                     \
+}
+
+ASD_SCBSITE_READ(u32, dword);
+ASD_SCBSITE_READ(u16, word);
+
+static inline u8 asd_scbsite_read_byte(struct asd_ha_struct *asd_ha,
+				       u16 scb_site_no,
+				       u16 offs)
+{
+	if (offs & 1)
+		return asd_scbsite_read_word(asd_ha, scb_site_no,
+					     offs & ~1) >> 8;
+	else
+		return asd_scbsite_read_word(asd_ha, scb_site_no,
+					     offs) & 0xFF;
+}
+
+
+#define ASD_SCBSITE_WRITE(type, ord)                                       \
+static inline void asd_scbsite_write_##ord (struct asd_ha_struct *asd_ha,  \
+					u16 scb_site_no,                   \
+					u16 offs, type val)                \
+{                                                                          \
+	asd_write_reg_word(asd_ha, ALTCIOADR, MnSCB_SITE + offs);          \
+	asd_write_reg_word(asd_ha, ASCBPTR, scb_site_no);                  \
+	asd_write_reg_##ord (asd_ha, CTXACCESS, val);                      \
+}
+
+ASD_SCBSITE_WRITE(u32, dword);
+ASD_SCBSITE_WRITE(u16, word);
+
+static inline void asd_scbsite_write_byte(struct asd_ha_struct *asd_ha,
+					  u16 scb_site_no,
+					  u16 offs, u8 val)
+{
+	u16 base = offs & ~1;
+	u16 rval = asd_scbsite_read_word(asd_ha, scb_site_no, base);
+	if (offs & 1)
+		rval = (val << 8) | (rval & 0xFF);
+	else
+		rval = (rval & 0xFF00) | val;
+	asd_scbsite_write_word(asd_ha, scb_site_no, base, rval);
+}
+
+/**
+ * asd_ddbsite_update_word -- atomically update a word in a ddb site
+ * @asd_ha: pointer to host adapter structure
+ * @ddb_site_no: the DDB site number
+ * @offs: the offset into the DDB
+ * @oldval: old value found in that offset
+ * @newval: the new value to replace it
+ *
+ * This function is used when the sequencers are running and we need to
+ * update a DDB site atomically without expensive pausing and upausing
+ * of the sequencers and accessing the DDB site through the CIO bus.
+ *
+ * Return 0 on success; -EFAULT on parity error; -EAGAIN if the old value
+ * is different than the current value at that offset.
+ */
+static inline int asd_ddbsite_update_word(struct asd_ha_struct *asd_ha,
+					  u16 ddb_site_no, u16 offs,
+					  u16 oldval, u16 newval)
+{
+	u8  done;
+	u16 oval = asd_ddbsite_read_word(asd_ha, ddb_site_no, offs);
+	if (oval != oldval)
+		return -EAGAIN;
+	asd_write_reg_word(asd_ha, AOLDDATA, oldval);
+	asd_write_reg_word(asd_ha, ANEWDATA, newval);
+	do {
+		done = asd_read_reg_byte(asd_ha, ATOMICSTATCTL);
+	} while (!(done & ATOMICDONE));
+	if (done & ATOMICERR)
+		return -EFAULT;	  /* parity error */
+	else if (done & ATOMICWIN)
+		return 0;	  /* success */
+	else
+		return -EAGAIN;	  /* oldval different than current value */
+}
+
+static inline int asd_ddbsite_update_byte(struct asd_ha_struct *asd_ha,
+					  u16 ddb_site_no, u16 offs,
+					  u8 _oldval, u8 _newval)
+{
+	u16 base = offs & ~1;
+	u16 oval;
+	u16 nval = asd_ddbsite_read_word(asd_ha, ddb_site_no, base);
+	if (offs & 1) {
+		if ((nval >> 8) != _oldval)
+			return -EAGAIN;
+		nval = (_newval << 8) | (nval & 0xFF);
+		oval = (_oldval << 8) | (nval & 0xFF);
+	} else {
+		if ((nval & 0xFF) != _oldval)
+			return -EAGAIN;
+		nval = (nval & 0xFF00) | _newval;
+		oval = (nval & 0xFF00) | _oldval;
+	}
+	return asd_ddbsite_update_word(asd_ha, ddb_site_no, base, oval, nval);
+}
+
+static inline void asd_write_reg_addr(struct asd_ha_struct *asd_ha, u32 reg,
+				      dma_addr_t dma_handle)
+{
+	asd_write_reg_dword(asd_ha, reg,   ASD_BUSADDR_LO(dma_handle));
+	asd_write_reg_dword(asd_ha, reg+4, ASD_BUSADDR_HI(dma_handle));
+}
+
+static inline u32 asd_get_cmdctx_size(struct asd_ha_struct *asd_ha)
+{
+	/* DCHREVISION returns 0, possibly broken */
+	u32 ctxmemsize = asd_read_reg_dword(asd_ha, LmMnINT(0,0)) & CTXMEMSIZE;
+	return ctxmemsize ? 65536 : 32768;
+}
+
+static inline u32 asd_get_devctx_size(struct asd_ha_struct *asd_ha)
+{
+	u32 ctxmemsize = asd_read_reg_dword(asd_ha, LmMnINT(0,0)) & CTXMEMSIZE;
+	return ctxmemsize ? 8192 : 4096;
+}
+
+static inline void asd_disable_ints(struct asd_ha_struct *asd_ha)
+{
+	asd_write_reg_dword(asd_ha, CHIMINTEN, RST_CHIMINTEN);
+}
+
+static inline void asd_enable_ints(struct asd_ha_struct *asd_ha)
+{
+	/* Enable COM SAS interrupt on errors, COMSTAT */
+	asd_write_reg_dword(asd_ha, COMSTATEN,
+			    EN_CSBUFPERR | EN_CSERR | EN_OVLYERR);
+	/* Enable DCH SAS CFIFTOERR */
+	asd_write_reg_dword(asd_ha, DCHSTATUS, EN_CFIFTOERR);
+	/* Enable Host Device interrupts */
+	asd_write_reg_dword(asd_ha, CHIMINTEN, SET_CHIMINTEN);
+}
+
+#endif
diff --git a/drivers/scsi/aic94xx/aic94xx_reg_def.h b/drivers/scsi/aic94xx/aic94xx_reg_def.h
new file mode 100644
index 000000000000..b79f45f3ad47
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_reg_def.h
@@ -0,0 +1,2398 @@
+/*
+ * Aic94xx SAS/SATA driver hardware registers defintions.
+ *
+ * Copyright (C) 2004 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2004 David Chaw <david_chaw@adaptec.com>
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * Luben Tuikov: Some register value updates to make it work with the window
+ * agnostic register r/w functions.  Some register corrections, sizes,
+ * etc.
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ * $Id: //depot/aic94xx/aic94xx_reg_def.h#27 $
+ *
+ */
+
+#ifndef _ADP94XX_REG_DEF_H_
+#define _ADP94XX_REG_DEF_H_
+
+/*
+ * Common definitions.
+ */
+#define CSEQ_MODE_PAGE_SIZE	0x200		/* CSEQ mode page size */
+#define LmSEQ_MODE_PAGE_SIZE	0x200		/* LmSEQ mode page size */
+#define LmSEQ_HOST_REG_SIZE   	0x4000		/* LmSEQ Host Register size */
+
+/********************* COM_SAS registers definition *************************/
+
+/* The base is REG_BASE_ADDR, defined in aic94xx_reg.h.
+ */
+
+/*
+ * CHIM Registers, Address Range : (0x00-0xFF)
+ */
+#define COMBIST		(REG_BASE_ADDR + 0x00)
+
+/* bits 31:24 */
+#define		L7BLKRST		0x80000000
+#define		L6BLKRST		0x40000000
+#define		L5BLKRST		0x20000000
+#define		L4BLKRST		0x10000000
+#define		L3BLKRST		0x08000000
+#define		L2BLKRST		0x04000000
+#define		L1BLKRST		0x02000000
+#define		L0BLKRST		0x01000000
+#define		LmBLKRST		0xFF000000
+#define LmBLKRST_COMBIST(phyid)		(1 << (24 + phyid))
+
+#define		OCMBLKRST		0x00400000
+#define		CTXMEMBLKRST		0x00200000
+#define		CSEQBLKRST		0x00100000
+#define		EXSIBLKRST		0x00040000
+#define		DPIBLKRST		0x00020000
+#define		DFIFBLKRST		0x00010000
+#define		HARDRST			0x00000200
+#define		COMBLKRST		0x00000100
+#define		FRCDFPERR		0x00000080
+#define		FRCCIOPERR		0x00000020
+#define		FRCBISTERR		0x00000010
+#define		COMBISTEN		0x00000004
+#define		COMBISTDONE		0x00000002	/* ro */
+#define 	COMBISTFAIL		0x00000001	/* ro */
+
+#define COMSTAT		(REG_BASE_ADDR + 0x04)
+
+#define		REQMBXREAD		0x00000040
+#define 	RSPMBXAVAIL		0x00000020
+#define 	CSBUFPERR		0x00000008
+#define		OVLYERR			0x00000004
+#define 	CSERR			0x00000002
+#define		OVLYDMADONE		0x00000001
+
+#define		COMSTAT_MASK		(REQMBXREAD | RSPMBXAVAIL | \
+					 CSBUFPERR | OVLYERR | CSERR |\
+					 OVLYDMADONE)
+
+#define COMSTATEN	(REG_BASE_ADDR + 0x08)
+
+#define		EN_REQMBXREAD		0x00000040
+#define		EN_RSPMBXAVAIL		0x00000020
+#define		EN_CSBUFPERR		0x00000008
+#define		EN_OVLYERR		0x00000004
+#define		EN_CSERR		0x00000002
+#define		EN_OVLYDONE		0x00000001
+
+#define SCBPRO		(REG_BASE_ADDR + 0x0C)
+
+#define		SCBCONS_MASK		0xFFFF0000
+#define		SCBPRO_MASK		0x0000FFFF
+
+#define CHIMREQMBX	(REG_BASE_ADDR + 0x10)
+
+#define CHIMRSPMBX	(REG_BASE_ADDR + 0x14)
+
+#define CHIMINT		(REG_BASE_ADDR + 0x18)
+
+#define		EXT_INT0		0x00000800
+#define		EXT_INT1		0x00000400
+#define		PORRSTDET		0x00000200
+#define		HARDRSTDET		0x00000100
+#define		DLAVAILQ		0x00000080	/* ro */
+#define		HOSTERR			0x00000040
+#define		INITERR			0x00000020
+#define		DEVINT			0x00000010
+#define		COMINT			0x00000008
+#define		DEVTIMER2		0x00000004
+#define		DEVTIMER1		0x00000002
+#define		DLAVAIL			0x00000001
+
+#define		CHIMINT_MASK		(HOSTERR | INITERR | DEVINT | COMINT |\
+					 DEVTIMER2 | DEVTIMER1 | DLAVAIL)
+
+#define 	DEVEXCEPT_MASK		(HOSTERR | INITERR | DEVINT | COMINT)
+
+#define CHIMINTEN	(REG_BASE_ADDR + 0x1C)
+
+#define		RST_EN_EXT_INT1		0x01000000
+#define		RST_EN_EXT_INT0		0x00800000
+#define		RST_EN_HOSTERR		0x00400000
+#define		RST_EN_INITERR		0x00200000
+#define		RST_EN_DEVINT		0x00100000
+#define		RST_EN_COMINT		0x00080000
+#define		RST_EN_DEVTIMER2	0x00040000
+#define		RST_EN_DEVTIMER1	0x00020000
+#define		RST_EN_DLAVAIL		0x00010000
+#define		SET_EN_EXT_INT1		0x00000100
+#define		SET_EN_EXT_INT0		0x00000080
+#define		SET_EN_HOSTERR		0x00000040
+#define		SET_EN_INITERR		0x00000020
+#define		SET_EN_DEVINT		0x00000010
+#define		SET_EN_COMINT		0x00000008
+#define		SET_EN_DEVTIMER2	0x00000004
+#define		SET_EN_DEVTIMER1	0x00000002
+#define		SET_EN_DLAVAIL		0x00000001
+
+#define		RST_CHIMINTEN		(RST_EN_HOSTERR | RST_EN_INITERR | \
+					 RST_EN_DEVINT | RST_EN_COMINT | \
+					 RST_EN_DEVTIMER2 | RST_EN_DEVTIMER1 |\
+					 RST_EN_DLAVAIL)
+
+#define		SET_CHIMINTEN		(SET_EN_HOSTERR | SET_EN_INITERR |\
+					 SET_EN_DEVINT | SET_EN_COMINT |\
+					 SET_EN_DLAVAIL)
+
+#define OVLYDMACTL	(REG_BASE_ADDR + 0x20)
+
+#define		OVLYADR_MASK		0x07FF0000
+#define		OVLYLSEQ_MASK		0x0000FF00
+#define		OVLYCSEQ		0x00000080
+#define		OVLYHALTERR		0x00000040
+#define		PIOCMODE		0x00000020
+#define		RESETOVLYDMA		0x00000008	/* wo */
+#define		STARTOVLYDMA		0x00000004
+#define		STOPOVLYDMA		0x00000002	/* wo */
+#define		OVLYDMAACT		0x00000001	/* ro */
+
+#define OVLYDMACNT	(REG_BASE_ADDR + 0x24)
+
+#define		OVLYDOMAIN1		0x20000000	/* ro */
+#define		OVLYDOMAIN0		0x10000000
+#define		OVLYBUFADR_MASK		0x007F0000
+#define		OVLYDMACNT_MASK		0x00003FFF
+
+#define OVLYDMAADR	(REG_BASE_ADDR + 0x28)
+
+#define DMAERR		(REG_BASE_ADDR + 0x30)
+
+#define		OVLYERRSTAT_MASK	0x0000FF00	/* ro */
+#define		CSERRSTAT_MASK		0x000000FF	/* ro */
+
+#define SPIODATA	(REG_BASE_ADDR + 0x34)
+
+/* 0x38 - 0x3C are reserved  */
+
+#define T1CNTRLR	(REG_BASE_ADDR + 0x40)
+
+#define		T1DONE			0x00010000	/* ro */
+#define		TIMER64			0x00000400
+#define		T1ENABLE		0x00000200
+#define		T1RELOAD		0x00000100
+#define		T1PRESCALER_MASK	0x00000003
+
+#define	T1CMPR		(REG_BASE_ADDR + 0x44)
+
+#define T1CNTR		(REG_BASE_ADDR + 0x48)
+
+#define T2CNTRLR	(REG_BASE_ADDR + 0x4C)
+
+#define		T2DONE			0x00010000	/* ro */
+#define		T2ENABLE		0x00000200
+#define		T2RELOAD		0x00000100
+#define		T2PRESCALER_MASK	0x00000003
+
+#define	T2CMPR		(REG_BASE_ADDR + 0x50)
+
+#define T2CNTR		(REG_BASE_ADDR + 0x54)
+
+/* 0x58h - 0xFCh are reserved */
+
+/*
+ * DCH_SAS Registers, Address Range : (0x800-0xFFF)
+ */
+#define CMDCTXBASE	(REG_BASE_ADDR + 0x800)
+
+#define DEVCTXBASE	(REG_BASE_ADDR + 0x808)
+
+#define CTXDOMAIN	(REG_BASE_ADDR + 0x810)
+
+#define		DEVCTXDOMAIN1		0x00000008	/* ro */
+#define		DEVCTXDOMAIN0		0x00000004
+#define		CMDCTXDOMAIN1		0x00000002	/* ro */
+#define		CMDCTXDOMAIN0		0x00000001
+
+#define DCHCTL		(REG_BASE_ADDR + 0x814)
+
+#define		OCMBISTREPAIR		0x00080000
+#define		OCMBISTEN		0x00040000
+#define		OCMBISTDN		0x00020000	/* ro */
+#define		OCMBISTFAIL		0x00010000	/* ro */
+#define		DDBBISTEN		0x00004000
+#define		DDBBISTDN		0x00002000	/* ro */
+#define		DDBBISTFAIL		0x00001000	/* ro */
+#define		SCBBISTEN		0x00000400
+#define		SCBBISTDN		0x00000200	/* ro */
+#define		SCBBISTFAIL		0x00000100	/* ro */
+
+#define		MEMSEL_MASK		0x000000E0
+#define		MEMSEL_CCM_LSEQ		0x00000000
+#define		MEMSEL_CCM_IOP		0x00000020
+#define		MEMSEL_CCM_SASCTL	0x00000040
+#define		MEMSEL_DCM_LSEQ		0x00000060
+#define		MEMSEL_DCM_IOP		0x00000080
+#define		MEMSEL_OCM		0x000000A0
+
+#define		FRCERR			0x00000010
+#define		AUTORLS			0x00000001
+
+#define DCHREVISION	(REG_BASE_ADDR + 0x818)
+
+#define		DCHREVISION_MASK	0x000000FF
+
+#define DCHSTATUS	(REG_BASE_ADDR + 0x81C)
+
+#define		EN_CFIFTOERR		0x00020000
+#define		CFIFTOERR		0x00000200
+#define		CSEQINT			0x00000100	/* ro */
+#define		LSEQ7INT		0x00000080	/* ro */
+#define		LSEQ6INT		0x00000040	/* ro */
+#define		LSEQ5INT		0x00000020	/* ro */
+#define		LSEQ4INT		0x00000010	/* ro */
+#define		LSEQ3INT		0x00000008	/* ro */
+#define		LSEQ2INT		0x00000004	/* ro */
+#define		LSEQ1INT		0x00000002	/* ro */
+#define		LSEQ0INT		0x00000001	/* ro */
+
+#define		LSEQINT_MASK		(LSEQ7INT | LSEQ6INT | LSEQ5INT |\
+					 LSEQ4INT | LSEQ3INT | LSEQ2INT	|\
+					 LSEQ1INT | LSEQ0INT)
+
+#define DCHDFIFDEBUG	(REG_BASE_ADDR + 0x820)
+#define		ENFAIRMST		0x00FF0000
+#define		DISWRMST9		0x00000200
+#define		DISWRMST8		0x00000100
+#define		DISRDMST		0x000000FF
+
+#define ATOMICSTATCTL	(REG_BASE_ADDR + 0x824)
+/* 8 bit wide */
+#define		AUTOINC			0x80
+#define		ATOMICERR		0x04
+#define		ATOMICWIN		0x02
+#define		ATOMICDONE		0x01
+
+
+#define ALTCIOADR	(REG_BASE_ADDR + 0x828)
+/* 16 bit; bits 8:0 define CIO addr space of CSEQ */
+
+#define ASCBPTR		(REG_BASE_ADDR + 0x82C)
+/* 16 bit wide */
+
+#define ADDBPTR		(REG_BASE_ADDR + 0x82E)
+/* 16 bit wide */
+
+#define ANEWDATA	(REG_BASE_ADDR + 0x830)
+/* 16 bit */
+
+#define AOLDDATA	(REG_BASE_ADDR + 0x834)
+/* 16 bit */
+
+#define CTXACCESS	(REG_BASE_ADDR + 0x838)
+/* 32 bit */
+
+/* 0x83Ch - 0xFFCh are reserved */
+
+/*
+ * ARP2 External Processor Registers, Address Range : (0x00-0x1F)
+ */
+#define ARP2CTL		0x00
+
+#define		FRCSCRPERR		0x00040000
+#define		FRCARP2PERR		0x00020000
+#define		FRCARP2ILLOPC		0x00010000
+#define		ENWAITTO		0x00008000
+#define		PERRORDIS		0x00004000
+#define		FAILDIS			0x00002000
+#define		CIOPERRDIS		0x00001000
+#define		BREAKEN3		0x00000800
+#define		BREAKEN2		0x00000400
+#define		BREAKEN1		0x00000200
+#define		BREAKEN0		0x00000100
+#define		EPAUSE			0x00000008
+#define		PAUSED			0x00000004	/* ro */
+#define		STEP			0x00000002
+#define		ARP2RESET		0x00000001	/* wo */
+
+#define ARP2INT		0x04
+
+#define		HALTCODE_MASK		0x00FF0000	/* ro */
+#define		ARP2WAITTO		0x00000100
+#define		ARP2HALTC		0x00000080
+#define		ARP2ILLOPC		0x00000040
+#define		ARP2PERR		0x00000020
+#define		ARP2CIOPERR		0x00000010
+#define		ARP2BREAK3		0x00000008
+#define		ARP2BREAK2		0x00000004
+#define		ARP2BREAK1		0x00000002
+#define		ARP2BREAK0		0x00000001
+
+#define ARP2INTEN	0x08
+
+#define		EN_ARP2WAITTO		0x00000100
+#define		EN_ARP2HALTC		0x00000080
+#define		EN_ARP2ILLOPC		0x00000040
+#define		EN_ARP2PERR		0x00000020
+#define		EN_ARP2CIOPERR		0x00000010
+#define		EN_ARP2BREAK3		0x00000008
+#define		EN_ARP2BREAK2		0x00000004
+#define		EN_ARP2BREAK1		0x00000002
+#define		EN_ARP2BREAK0		0x00000001
+
+#define ARP2BREAKADR01	0x0C
+
+#define		BREAKADR1_MASK		0x0FFF0000
+#define		BREAKADR0_MASK		0x00000FFF
+
+#define	ARP2BREAKADR23	0x10
+
+#define		BREAKADR3_MASK		0x0FFF0000
+#define		BREAKADR2_MASK		0x00000FFF
+
+/* 0x14h - 0x1Ch are reserved */
+
+/*
+ * ARP2 Registers, Address Range : (0x00-0x1F)
+ * The definitions have the same address offset for CSEQ and LmSEQ
+ * CIO Bus Registers.
+ */
+#define MODEPTR		0x00
+
+#define		DSTMODE			0xF0
+#define		SRCMODE			0x0F
+
+#define ALTMODE		0x01
+
+#define		ALTDMODE		0xF0
+#define		ALTSMODE		0x0F
+
+#define ATOMICXCHG	0x02
+
+#define FLAG		0x04
+
+#define		INTCODE_MASK		0xF0
+#define		ALTMODEV2		0x04
+#define		CARRY_INT		0x02
+#define		CARRY			0x01
+
+#define ARP2INTCTL	0x05
+
+#define 	PAUSEDIS		0x80
+#define		RSTINTCTL		0x40
+#define		POPALTMODE		0x08
+#define		ALTMODEV		0x04
+#define		INTMASK			0x02
+#define		IRET			0x01
+
+#define STACK		0x06
+
+#define FUNCTION1	0x07
+
+#define PRGMCNT		0x08
+
+#define ACCUM		0x0A
+
+#define SINDEX		0x0C
+
+#define DINDEX		0x0E
+
+#define ALLONES		0x10
+
+#define ALLZEROS	0x11
+
+#define SINDIR		0x12
+
+#define DINDIR		0x13
+
+#define JUMLDIR		0x14
+
+#define ARP2HALTCODE	0x15
+
+#define CURRADDR	0x16
+
+#define LASTADDR	0x18
+
+#define NXTLADDR	0x1A
+
+#define DBGPORTPTR	0x1C
+
+#define DBGPORT		0x1D
+
+/*
+ * CIO Registers.
+ * The definitions have the same address offset for CSEQ and LmSEQ
+ * CIO Bus Registers.
+ */
+#define MnSCBPTR      	0x20
+
+#define MnDDBPTR      	0x22
+
+#define SCRATCHPAGE	0x24
+
+#define MnSCRATCHPAGE	0x25
+
+#define SCRATCHPAGESV	0x26
+
+#define MnSCRATCHPAGESV	0x27
+
+#define MnDMAERRS	0x46
+
+#define MnSGDMAERRS	0x47
+
+#define MnSGBUF		0x53
+
+#define MnSGDMASTAT	0x5b
+
+#define MnDDMACTL	0x5c	/* RAZOR.rspec.fm rev 1.5 is wrong */
+
+#define MnDDMASTAT	0x5d	/* RAZOR.rspec.fm rev 1.5 is wrong */
+
+#define MnDDMAMODE	0x5e	/* RAZOR.rspec.fm rev 1.5 is wrong */
+
+#define MnDMAENG	0x60
+
+#define MnPIPECTL	0x61
+
+#define MnSGBADR	0x65
+
+#define MnSCB_SITE	0x100
+
+#define MnDDB_SITE	0x180
+
+/*
+ * The common definitions below have the same address offset for both
+ * CSEQ and LmSEQ.
+ */
+#define BISTCTL0	0x4C
+
+#define BISTCTL1	0x50
+
+#define MAPPEDSCR	0x800
+
+/*
+ * CSEQ Host Register, Address Range : (0x000-0xFFC)
+ */
+#define CSEQ_HOST_REG_BASE_ADR		0xB8001000
+
+#define CARP2CTL			(CSEQ_HOST_REG_BASE_ADR	+ ARP2CTL)
+
+#define CARP2INT			(CSEQ_HOST_REG_BASE_ADR	+ ARP2INT)
+
+#define CARP2INTEN			(CSEQ_HOST_REG_BASE_ADR	+ ARP2INTEN)
+
+#define CARP2BREAKADR01			(CSEQ_HOST_REG_BASE_ADR+ARP2BREAKADR01)
+
+#define CARP2BREAKADR23			(CSEQ_HOST_REG_BASE_ADR+ARP2BREAKADR23)
+
+#define CBISTCTL			(CSEQ_HOST_REG_BASE_ADR	+ BISTCTL1)
+
+#define		CSEQRAMBISTEN		0x00000040
+#define		CSEQRAMBISTDN		0x00000020	/* ro */
+#define		CSEQRAMBISTFAIL		0x00000010	/* ro */
+#define		CSEQSCRBISTEN		0x00000004
+#define		CSEQSCRBISTDN		0x00000002	/* ro */
+#define		CSEQSCRBISTFAIL		0x00000001	/* ro */
+
+#define CMAPPEDSCR			(CSEQ_HOST_REG_BASE_ADR	+ MAPPEDSCR)
+
+/*
+ * CSEQ CIO Bus Registers, Address Range : (0x0000-0x1FFC)
+ * 16 modes, each mode is 512 bytes.
+ * Unless specified, the register should valid for all modes.
+ */
+#define CSEQ_CIO_REG_BASE_ADR		REG_BASE_ADDR_CSEQCIO
+
+#define CSEQm_CIO_REG(Mode, Reg) \
+		(CSEQ_CIO_REG_BASE_ADR  + \
+		((u32) (Mode) * CSEQ_MODE_PAGE_SIZE) + (u32) (Reg))
+
+#define CMODEPTR	(CSEQ_CIO_REG_BASE_ADR + MODEPTR)
+
+#define CALTMODE	(CSEQ_CIO_REG_BASE_ADR + ALTMODE)
+
+#define CATOMICXCHG	(CSEQ_CIO_REG_BASE_ADR + ATOMICXCHG)
+
+#define CFLAG		(CSEQ_CIO_REG_BASE_ADR + FLAG)
+
+#define CARP2INTCTL	(CSEQ_CIO_REG_BASE_ADR + ARP2INTCTL)
+
+#define CSTACK		(CSEQ_CIO_REG_BASE_ADR + STACK)
+
+#define CFUNCTION1	(CSEQ_CIO_REG_BASE_ADR + FUNCTION1)
+
+#define CPRGMCNT	(CSEQ_CIO_REG_BASE_ADR + PRGMCNT)
+
+#define CACCUM		(CSEQ_CIO_REG_BASE_ADR + ACCUM)
+
+#define CSINDEX		(CSEQ_CIO_REG_BASE_ADR + SINDEX)
+
+#define CDINDEX		(CSEQ_CIO_REG_BASE_ADR + DINDEX)
+
+#define CALLONES	(CSEQ_CIO_REG_BASE_ADR + ALLONES)
+
+#define CALLZEROS	(CSEQ_CIO_REG_BASE_ADR + ALLZEROS)
+
+#define CSINDIR		(CSEQ_CIO_REG_BASE_ADR + SINDIR)
+
+#define CDINDIR		(CSEQ_CIO_REG_BASE_ADR + DINDIR)
+
+#define CJUMLDIR	(CSEQ_CIO_REG_BASE_ADR + JUMLDIR)
+
+#define CARP2HALTCODE	(CSEQ_CIO_REG_BASE_ADR + ARP2HALTCODE)
+
+#define CCURRADDR	(CSEQ_CIO_REG_BASE_ADR + CURRADDR)
+
+#define CLASTADDR	(CSEQ_CIO_REG_BASE_ADR + LASTADDR)
+
+#define CNXTLADDR	(CSEQ_CIO_REG_BASE_ADR + NXTLADDR)
+
+#define CDBGPORTPTR	(CSEQ_CIO_REG_BASE_ADR + DBGPORTPTR)
+
+#define CDBGPORT	(CSEQ_CIO_REG_BASE_ADR + DBGPORT)
+
+#define CSCRATCHPAGE	(CSEQ_CIO_REG_BASE_ADR + SCRATCHPAGE)
+
+#define CMnSCBPTR(Mode)       CSEQm_CIO_REG(Mode, MnSCBPTR)
+
+#define CMnDDBPTR(Mode)       CSEQm_CIO_REG(Mode, MnDDBPTR)
+
+#define CMnSCRATCHPAGE(Mode)		CSEQm_CIO_REG(Mode, MnSCRATCHPAGE)
+
+#define CLINKCON	(CSEQ_CIO_REG_BASE_ADR + 0x28)
+
+#define	CCIOAACESS	(CSEQ_CIO_REG_BASE_ADR + 0x2C)
+
+/* mode 0-7 */
+#define MnREQMBX 0x30
+#define CMnREQMBX(Mode)			CSEQm_CIO_REG(Mode, 0x30)
+
+/* mode 8 */
+#define CSEQCON				CSEQm_CIO_REG(8, 0x30)
+
+/* mode 0-7 */
+#define MnRSPMBX 0x34
+#define CMnRSPMBX(Mode)			CSEQm_CIO_REG(Mode, 0x34)
+
+/* mode 8 */
+#define CSEQCOMCTL			CSEQm_CIO_REG(8, 0x34)
+
+/* mode 8 */
+#define CSEQCOMSTAT			CSEQm_CIO_REG(8, 0x35)
+
+/* mode 8 */
+#define CSEQCOMINTEN			CSEQm_CIO_REG(8, 0x36)
+
+/* mode 8 */
+#define CSEQCOMDMACTL			CSEQm_CIO_REG(8, 0x37)
+
+#define		CSHALTERR		0x10
+#define		RESETCSDMA		0x08		/* wo */
+#define		STARTCSDMA		0x04
+#define		STOPCSDMA		0x02		/* wo */
+#define		CSDMAACT		0x01		/* ro */
+
+/* mode 0-7 */
+#define MnINT 0x38
+#define CMnINT(Mode)			CSEQm_CIO_REG(Mode, 0x38)
+
+#define		CMnREQMBXE		0x02
+#define		CMnRSPMBXF		0x01
+#define		CMnINT_MASK		0x00000003
+
+/* mode 8 */
+#define CSEQREQMBX			CSEQm_CIO_REG(8, 0x38)
+
+/* mode 0-7 */
+#define MnINTEN 0x3C
+#define CMnINTEN(Mode)			CSEQm_CIO_REG(Mode, 0x3C)
+
+#define		EN_CMnRSPMBXF		0x01
+
+/* mode 8 */
+#define CSEQRSPMBX			CSEQm_CIO_REG(8, 0x3C)
+
+/* mode 8 */
+#define CSDMAADR			CSEQm_CIO_REG(8, 0x40)
+
+/* mode 8 */
+#define CSDMACNT			CSEQm_CIO_REG(8, 0x48)
+
+/* mode 8 */
+#define CSEQDLCTL			CSEQm_CIO_REG(8, 0x4D)
+
+#define		DONELISTEND		0x10
+#define 	DONELISTSIZE_MASK	0x0F
+#define		DONELISTSIZE_8ELEM	0x01
+#define		DONELISTSIZE_16ELEM	0x02
+#define		DONELISTSIZE_32ELEM	0x03
+#define		DONELISTSIZE_64ELEM	0x04
+#define		DONELISTSIZE_128ELEM	0x05
+#define		DONELISTSIZE_256ELEM	0x06
+#define		DONELISTSIZE_512ELEM	0x07
+#define		DONELISTSIZE_1024ELEM	0x08
+#define		DONELISTSIZE_2048ELEM	0x09
+#define		DONELISTSIZE_4096ELEM	0x0A
+#define		DONELISTSIZE_8192ELEM	0x0B
+#define		DONELISTSIZE_16384ELEM	0x0C
+
+/* mode 8 */
+#define CSEQDLOFFS			CSEQm_CIO_REG(8, 0x4E)
+
+/* mode 11 */
+#define CM11INTVEC0			CSEQm_CIO_REG(11, 0x50)
+
+/* mode 11 */
+#define CM11INTVEC1			CSEQm_CIO_REG(11, 0x52)
+
+/* mode 11 */
+#define CM11INTVEC2			CSEQm_CIO_REG(11, 0x54)
+
+#define	CCONMSK	  			(CSEQ_CIO_REG_BASE_ADR + 0x60)
+
+#define	CCONEXIST			(CSEQ_CIO_REG_BASE_ADR + 0x61)
+
+#define	CCONMODE			(CSEQ_CIO_REG_BASE_ADR + 0x62)
+
+#define CTIMERCALC			(CSEQ_CIO_REG_BASE_ADR + 0x64)
+
+#define CINTDIS				(CSEQ_CIO_REG_BASE_ADR + 0x68)
+
+/* mode 8, 32x32 bits, 128 bytes of mapped buffer */
+#define CSBUFFER			CSEQm_CIO_REG(8, 0x80)
+
+#define	CSCRATCH			(CSEQ_CIO_REG_BASE_ADR + 0x1C0)
+
+/* mode 0-8 */
+#define CMnSCRATCH(Mode)		CSEQm_CIO_REG(Mode, 0x1E0)
+
+/*
+ * CSEQ Mapped Instruction RAM Page, Address Range : (0x0000-0x1FFC)
+ */
+#define CSEQ_RAM_REG_BASE_ADR		0xB8004000
+
+/*
+ * The common definitions below have the same address offset for all the Link
+ * sequencers.
+ */
+#define MODECTL		0x40
+
+#define DBGMODE		0x44
+
+#define CONTROL		0x48
+#define LEDTIMER		0x00010000
+#define LEDTIMERS_10us		0x00000000
+#define LEDTIMERS_1ms		0x00000800
+#define LEDTIMERS_100ms		0x00001000
+#define LEDMODE_TXRX		0x00000000
+#define LEDMODE_CONNECTED	0x00000200
+#define LEDPOL			0x00000100
+
+#define LSEQRAM		0x1000
+
+/*
+ * LmSEQ Host Registers, Address Range : (0x0000-0x3FFC)
+ */
+#define LSEQ0_HOST_REG_BASE_ADR		0xB8020000
+#define LSEQ1_HOST_REG_BASE_ADR		0xB8024000
+#define LSEQ2_HOST_REG_BASE_ADR		0xB8028000
+#define LSEQ3_HOST_REG_BASE_ADR		0xB802C000
+#define LSEQ4_HOST_REG_BASE_ADR		0xB8030000
+#define LSEQ5_HOST_REG_BASE_ADR		0xB8034000
+#define LSEQ6_HOST_REG_BASE_ADR		0xB8038000
+#define LSEQ7_HOST_REG_BASE_ADR		0xB803C000
+
+#define LmARP2CTL(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					ARP2CTL)
+
+#define LmARP2INT(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					ARP2INT)
+
+#define LmARP2INTEN(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					ARP2INTEN)
+
+#define LmDBGMODE(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					DBGMODE)
+
+#define LmCONTROL(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					CONTROL)
+
+#define LmARP2BREAKADR01(LinkNum)	(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					ARP2BREAKADR01)
+
+#define LmARP2BREAKADR23(LinkNum)	(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					ARP2BREAKADR23)
+
+#define LmMODECTL(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					MODECTL)
+
+#define		LmAUTODISCI		0x08000000
+#define		LmDSBLBITLT		0x04000000
+#define		LmDSBLANTT		0x02000000
+#define		LmDSBLCRTT		0x01000000
+#define		LmDSBLCONT		0x00000100
+#define		LmPRIMODE		0x00000080
+#define		LmDSBLHOLD		0x00000040
+#define		LmDISACK		0x00000020
+#define		LmBLIND48		0x00000010
+#define		LmRCVMODE_MASK		0x0000000C
+#define		LmRCVMODE_PLD		0x00000000
+#define		LmRCVMODE_HPC		0x00000004
+
+#define LmDBGMODE(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					DBGMODE)
+
+#define		LmFRCPERR		0x80000000
+#define		LmMEMSEL_MASK		0x30000000
+#define		LmFRCRBPERR		0x00000000
+#define		LmFRCTBPERR		0x10000000
+#define		LmFRCSGBPERR		0x20000000
+#define		LmFRCARBPERR		0x30000000
+#define		LmRCVIDW		0x00080000
+#define		LmINVDWERR		0x00040000
+#define		LmRCVDISP		0x00004000
+#define		LmDISPERR		0x00002000
+#define		LmDSBLDSCR		0x00000800
+#define		LmDSBLSCR		0x00000400
+#define		LmFRCNAK		0x00000200
+#define		LmFRCROFS		0x00000100
+#define		LmFRCCRC		0x00000080
+#define		LmFRMTYPE_MASK		0x00000070
+#define		LmSG_DATA		0x00000000
+#define		LmSG_COMMAND		0x00000010
+#define		LmSG_TASK		0x00000020
+#define		LmSG_TGTXFER		0x00000030
+#define		LmSG_RESPONSE		0x00000040
+#define		LmSG_IDENADDR		0x00000050
+#define		LmSG_OPENADDR		0x00000060
+#define		LmDISCRCGEN		0x00000008
+#define		LmDISCRCCHK		0x00000004
+#define		LmSSXMTFRM		0x00000002
+#define		LmSSRCVFRM		0x00000001
+
+#define LmCONTROL(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					CONTROL)
+
+#define		LmSTEPXMTFRM		0x00000002
+#define		LmSTEPRCVFRM		0x00000001
+
+#define LmBISTCTL0(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	  \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) + \
+					BISTCTL0)
+
+#define		ARBBISTEN		0x40000000
+#define		ARBBISTDN		0x20000000	/* ro */
+#define		ARBBISTFAIL		0x10000000	/* ro */
+#define		TBBISTEN		0x00000400
+#define		TBBISTDN		0x00000200	/* ro */
+#define		TBBISTFAIL		0x00000100	/* ro */
+#define		RBBISTEN		0x00000040
+#define		RBBISTDN		0x00000020	/* ro */
+#define		RBBISTFAIL		0x00000010	/* ro */
+#define		SGBISTEN		0x00000004
+#define		SGBISTDN		0x00000002	/* ro */
+#define		SGBISTFAIL		0x00000001	/* ro */
+
+#define LmBISTCTL1(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	 \
+					((LinkNum)*LmSEQ_HOST_REG_SIZE) +\
+					BISTCTL1)
+
+#define		LmRAMPAGE1		0x00000200
+#define		LmRAMPAGE0		0x00000100
+#define		LmIMEMBISTEN		0x00000040
+#define		LmIMEMBISTDN		0x00000020	/* ro */
+#define		LmIMEMBISTFAIL		0x00000010	/* ro */
+#define		LmSCRBISTEN		0x00000004
+#define		LmSCRBISTDN		0x00000002	/* ro */
+#define		LmSCRBISTFAIL		0x00000001	/* ro */
+#define		LmRAMPAGE		(LmRAMPAGE1 + LmRAMPAGE0)
+#define		LmRAMPAGE_LSHIFT	0x8
+
+#define LmSCRATCH(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	   \
+					((LinkNum) * LmSEQ_HOST_REG_SIZE) +\
+					MAPPEDSCR)
+
+#define LmSEQRAM(LinkNum)		(LSEQ0_HOST_REG_BASE_ADR +	   \
+					((LinkNum) * LmSEQ_HOST_REG_SIZE) +\
+					LSEQRAM)
+
+/*
+ * LmSEQ CIO Bus Register, Address Range : (0x0000-0xFFC)
+ * 8 modes, each mode is 512 bytes.
+ * Unless specified, the register should valid for all modes.
+ */
+#define LmSEQ_CIOBUS_REG_BASE		0x2000
+
+#define  LmSEQ_PHY_BASE(Mode, LinkNum) \
+		(LSEQ0_HOST_REG_BASE_ADR + \
+		(LmSEQ_HOST_REG_SIZE * (u32) (LinkNum)) + \
+		LmSEQ_CIOBUS_REG_BASE + \
+		((u32) (Mode) * LmSEQ_MODE_PAGE_SIZE))
+
+#define  LmSEQ_PHY_REG(Mode, LinkNum, Reg) \
+                 (LmSEQ_PHY_BASE(Mode, LinkNum) + (u32) (Reg))
+
+#define LmMODEPTR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, MODEPTR)
+
+#define LmALTMODE(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ALTMODE)
+
+#define LmATOMICXCHG(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ATOMICXCHG)
+
+#define LmFLAG(LinkNum)			LmSEQ_PHY_REG(0, LinkNum, FLAG)
+
+#define LmARP2INTCTL(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ARP2INTCTL)
+
+#define LmSTACK(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, STACK)
+
+#define LmFUNCTION1(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, FUNCTION1)
+
+#define LmPRGMCNT(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, PRGMCNT)
+
+#define LmACCUM(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ACCUM)
+
+#define LmSINDEX(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, SINDEX)
+
+#define LmDINDEX(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, DINDEX)
+
+#define LmALLONES(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ALLONES)
+
+#define LmALLZEROS(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ALLZEROS)
+
+#define LmSINDIR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, SINDIR)
+
+#define LmDINDIR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, DINDIR)
+
+#define LmJUMLDIR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, JUMLDIR)
+
+#define LmARP2HALTCODE(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, ARP2HALTCODE)
+
+#define LmCURRADDR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, CURRADDR)
+
+#define LmLASTADDR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, LASTADDR)
+
+#define LmNXTLADDR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, NXTLADDR)
+
+#define LmDBGPORTPTR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, DBGPORTPTR)
+
+#define LmDBGPORT(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, DBGPORT)
+
+#define LmSCRATCHPAGE(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, SCRATCHPAGE)
+
+#define LmMnSCRATCHPAGE(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 	\
+						      MnSCRATCHPAGE)
+
+#define LmTIMERCALC(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x28)
+
+#define LmREQMBX(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x30)
+
+#define LmRSPMBX(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x34)
+
+#define LmMnINT(LinkNum, Mode)		LmSEQ_PHY_REG(Mode, LinkNum, 0x38)
+
+#define		CTXMEMSIZE		0x80000000	/* ro */
+#define		LmACKREQ		0x08000000
+#define		LmNAKREQ		0x04000000
+#define		LmMnXMTERR		0x02000000
+#define		LmM5OOBSVC		0x01000000
+#define		LmHWTINT		0x00800000
+#define		LmMnCTXDONE		0x00100000
+#define		LmM2REQMBXF		0x00080000
+#define		LmM2RSPMBXE		0x00040000
+#define		LmMnDMAERR		0x00020000
+#define		LmRCVPRIM		0x00010000
+#define		LmRCVERR		0x00008000
+#define		LmADDRRCV		0x00004000
+#define		LmMnHDRMISS		0x00002000
+#define		LmMnWAITSCB		0x00001000
+#define		LmMnRLSSCB		0x00000800
+#define		LmMnSAVECTX		0x00000400
+#define		LmMnFETCHSG		0x00000200
+#define		LmMnLOADCTX		0x00000100
+#define		LmMnCFGICL		0x00000080
+#define		LmMnCFGSATA		0x00000040
+#define		LmMnCFGEXPSATA		0x00000020
+#define		LmMnCFGCMPLT		0x00000010
+#define		LmMnCFGRBUF		0x00000008
+#define		LmMnSAVETTR		0x00000004
+#define		LmMnCFGRDAT		0x00000002
+#define		LmMnCFGHDR		0x00000001
+
+#define LmMnINTEN(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x3C)
+
+#define		EN_LmACKREQ		0x08000000
+#define		EN_LmNAKREQ		0x04000000
+#define		EN_LmMnXMTERR		0x02000000
+#define		EN_LmM5OOBSVC		0x01000000
+#define		EN_LmHWTINT		0x00800000
+#define		EN_LmMnCTXDONE		0x00100000
+#define		EN_LmM2REQMBXF		0x00080000
+#define		EN_LmM2RSPMBXE		0x00040000
+#define		EN_LmMnDMAERR		0x00020000
+#define		EN_LmRCVPRIM		0x00010000
+#define		EN_LmRCVERR		0x00008000
+#define		EN_LmADDRRCV		0x00004000
+#define		EN_LmMnHDRMISS		0x00002000
+#define		EN_LmMnWAITSCB		0x00001000
+#define		EN_LmMnRLSSCB		0x00000800
+#define		EN_LmMnSAVECTX		0x00000400
+#define		EN_LmMnFETCHSG		0x00000200
+#define		EN_LmMnLOADCTX		0x00000100
+#define		EN_LmMnCFGICL		0x00000080
+#define		EN_LmMnCFGSATA		0x00000040
+#define		EN_LmMnCFGEXPSATA	0x00000020
+#define		EN_LmMnCFGCMPLT		0x00000010
+#define		EN_LmMnCFGRBUF		0x00000008
+#define		EN_LmMnSAVETTR		0x00000004
+#define		EN_LmMnCFGRDAT		0x00000002
+#define		EN_LmMnCFGHDR		0x00000001
+
+#define		LmM0INTEN_MASK		(EN_LmMnCFGCMPLT | EN_LmMnCFGRBUF | \
+					 EN_LmMnSAVETTR | EN_LmMnCFGRDAT | \
+					 EN_LmMnCFGHDR | EN_LmRCVERR | \
+					 EN_LmADDRRCV | EN_LmMnHDRMISS | \
+					 EN_LmMnRLSSCB | EN_LmMnSAVECTX | \
+					 EN_LmMnFETCHSG | EN_LmMnLOADCTX | \
+					 EN_LmHWTINT | EN_LmMnCTXDONE | \
+					 EN_LmRCVPRIM | EN_LmMnCFGSATA | \
+					 EN_LmMnCFGEXPSATA | EN_LmMnDMAERR)
+
+#define		LmM1INTEN_MASK		(EN_LmMnCFGCMPLT | EN_LmADDRRCV | \
+					 EN_LmMnRLSSCB | EN_LmMnSAVECTX | \
+					 EN_LmMnFETCHSG | EN_LmMnLOADCTX | \
+					 EN_LmMnXMTERR | EN_LmHWTINT | \
+					 EN_LmMnCTXDONE | EN_LmRCVPRIM | \
+					 EN_LmRCVERR | EN_LmMnDMAERR)
+
+#define		LmM2INTEN_MASK		(EN_LmADDRRCV | EN_LmHWTINT | \
+					 EN_LmM2REQMBXF | EN_LmRCVPRIM | \
+					 EN_LmRCVERR)
+
+#define		LmM5INTEN_MASK		(EN_LmADDRRCV | EN_LmM5OOBSVC | \
+					 EN_LmHWTINT | EN_LmRCVPRIM | \
+					 EN_LmRCVERR)
+
+#define LmXMTPRIMD(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x40)
+
+#define LmXMTPRIMCS(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x44)
+
+#define LmCONSTAT(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x45)
+
+#define LmMnDMAERRS(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x46)
+
+#define LmMnSGDMAERRS(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x47)
+
+#define LmM0EXPHDRP(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x48)
+
+#define LmM1SASALIGN(LinkNum)		LmSEQ_PHY_REG(1, LinkNum, 0x48)
+#define SAS_ALIGN_DEFAULT		0xFF
+
+#define LmM0MSKHDRP(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x49)
+
+#define LmM1STPALIGN(LinkNum)		LmSEQ_PHY_REG(1, LinkNum, 0x49)
+#define STP_ALIGN_DEFAULT		0x1F
+
+#define LmM0RCVHDRP(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x4A)
+
+#define LmM1XMTHDRP(LinkNum)		LmSEQ_PHY_REG(1, LinkNum, 0x4A)
+
+#define LmM0ICLADR(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x4B)
+
+#define LmM1ALIGNMODE(LinkNum)		LmSEQ_PHY_REG(1, LinkNum, 0x4B)
+
+#define		LmDISALIGN		0x20
+#define		LmROTSTPALIGN		0x10
+#define		LmSTPALIGN		0x08
+#define		LmROTNOTIFY		0x04
+#define		LmDUALALIGN		0x02
+#define		LmROTALIGN		0x01
+
+#define LmM0EXPRCVNT(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x4C)
+
+#define LmM1XMTCNT(LinkNum)		LmSEQ_PHY_REG(1, LinkNum, 0x4C)
+
+#define LmMnBUFSTAT(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x4E)
+
+#define		LmMnBUFPERR		0x01
+
+/* mode 0-1 */
+#define LmMnXFRLVL(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x59)
+
+#define		LmMnXFRLVL_128		0x05
+#define		LmMnXFRLVL_256		0x04
+#define		LmMnXFRLVL_512		0x03
+#define		LmMnXFRLVL_1024		0x02
+#define		LmMnXFRLVL_1536		0x01
+#define		LmMnXFRLVL_2048		0x00
+
+ /* mode 0-1 */
+#define LmMnSGDMACTL(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x5A)
+
+#define 	LmMnRESETSG		0x04
+#define 	LmMnSTOPSG		0x02
+#define 	LmMnSTARTSG		0x01
+
+/* mode 0-1 */
+#define LmMnSGDMASTAT(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x5B)
+
+/* mode 0-1 */
+#define LmMnDDMACTL(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x5C)
+
+#define 	LmMnFLUSH		0x40		/* wo */
+#define 	LmMnRLSRTRY		0x20		/* wo */
+#define 	LmMnDISCARD		0x10		/* wo */
+#define 	LmMnRESETDAT		0x08		/* wo */
+#define 	LmMnSUSDAT		0x04		/* wo */
+#define 	LmMnSTOPDAT		0x02		/* wo */
+#define 	LmMnSTARTDAT		0x01		/* wo */
+
+/* mode 0-1 */
+#define LmMnDDMASTAT(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x5D)
+
+#define		LmMnDPEMPTY		0x80
+#define		LmMnFLUSHING		0x40
+#define		LmMnDDMAREQ		0x20
+#define		LmMnHDMAREQ		0x10
+#define		LmMnDATFREE		0x08
+#define		LmMnDATSUS		0x04
+#define		LmMnDATACT		0x02
+#define		LmMnDATEN		0x01
+
+/* mode 0-1 */
+#define LmMnDDMAMODE(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x5E)
+
+#define 	LmMnDMATYPE_NORMAL		0x0000
+#define 	LmMnDMATYPE_HOST_ONLY_TX	0x0001
+#define 	LmMnDMATYPE_DEVICE_ONLY_TX	0x0002
+#define 	LmMnDMATYPE_INVALID		0x0003
+#define 	LmMnDMATYPE_MASK	0x0003
+
+#define 	LmMnDMAWRAP		0x0004
+#define 	LmMnBITBUCKET		0x0008
+#define 	LmMnDISHDR		0x0010
+#define 	LmMnSTPCRC		0x0020
+#define 	LmXTEST			0x0040
+#define 	LmMnDISCRC		0x0080
+#define 	LmMnENINTLK		0x0100
+#define 	LmMnADDRFRM		0x0400
+#define 	LmMnENXMTCRC		0x0800
+
+/* mode 0-1 */
+#define LmMnXFRCNT(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x70)
+
+/* mode 0-1 */
+#define LmMnDPSEL(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x7B)
+#define 	LmMnDPSEL_MASK		0x07
+#define 	LmMnEOLPRE		0x40
+#define 	LmMnEOSPRE		0x80
+
+/* Registers used in conjunction with LmMnDPSEL and LmMnDPACC registers */
+/* Receive Mode n = 0 */
+#define LmMnHRADDR			0x00
+#define LmMnHBYTECNT			0x01
+#define LmMnHREWIND			0x02
+#define LmMnDWADDR			0x03
+#define LmMnDSPACECNT			0x04
+#define LmMnDFRMSIZE			0x05
+
+/* Registers used in conjunction with LmMnDPSEL and LmMnDPACC registers */
+/* Transmit Mode n = 1 */
+#define LmMnHWADDR			0x00
+#define LmMnHSPACECNT			0x01
+/* #define LmMnHREWIND			0x02 */
+#define LmMnDRADDR			0x03
+#define LmMnDBYTECNT			0x04
+/* #define LmMnDFRMSIZE			0x05 */
+
+/* mode 0-1 */
+#define LmMnDPACC(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x78)
+#define 	LmMnDPACC_MASK		0x00FFFFFF
+
+/* mode 0-1 */
+#define LmMnHOLDLVL(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x7D)
+
+#define LmPRMSTAT0(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x80)
+#define LmPRMSTAT0BYTE0			0x80
+#define LmPRMSTAT0BYTE1			0x81
+#define LmPRMSTAT0BYTE2			0x82
+#define LmPRMSTAT0BYTE3			0x83
+
+#define		LmFRAMERCVD		0x80000000
+#define		LmXFRRDYRCVD		0x40000000
+#define		LmUNKNOWNP		0x20000000
+#define		LmBREAK			0x10000000
+#define		LmDONE			0x08000000
+#define		LmOPENACPT		0x04000000
+#define		LmOPENRJCT		0x02000000
+#define		LmOPENRTRY		0x01000000
+#define		LmCLOSERV1		0x00800000
+#define		LmCLOSERV0		0x00400000
+#define		LmCLOSENORM		0x00200000
+#define		LmCLOSECLAF		0x00100000
+#define		LmNOTIFYRV2		0x00080000
+#define		LmNOTIFYRV1		0x00040000
+#define		LmNOTIFYRV0		0x00020000
+#define		LmNOTIFYSPIN		0x00010000
+#define		LmBROADRV4		0x00008000
+#define		LmBROADRV3		0x00004000
+#define		LmBROADRV2		0x00002000
+#define		LmBROADRV1		0x00001000
+#define		LmBROADSES		0x00000800
+#define		LmBROADRVCH1		0x00000400
+#define		LmBROADRVCH0		0x00000200
+#define		LmBROADCH		0x00000100
+#define		LmAIPRVWP		0x00000080
+#define		LmAIPWP			0x00000040
+#define		LmAIPWD			0x00000020
+#define		LmAIPWC			0x00000010
+#define		LmAIPRV2		0x00000008
+#define		LmAIPRV1		0x00000004
+#define		LmAIPRV0		0x00000002
+#define		LmAIPNRML		0x00000001
+
+#define		LmBROADCAST_MASK	(LmBROADCH | LmBROADRVCH0 | \
+					 LmBROADRVCH1)
+
+#define LmPRMSTAT1(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0x84)
+#define LmPRMSTAT1BYTE0			0x84
+#define LmPRMSTAT1BYTE1			0x85
+#define LmPRMSTAT1BYTE2			0x86
+#define LmPRMSTAT1BYTE3			0x87
+
+#define		LmFRMRCVDSTAT		0x80000000
+#define		LmBREAK_DET		0x04000000
+#define		LmCLOSE_DET		0x02000000
+#define		LmDONE_DET		0x01000000
+#define		LmXRDY			0x00040000
+#define 	LmSYNCSRST		0x00020000
+#define 	LmSYNC			0x00010000
+#define 	LmXHOLD			0x00008000
+#define 	LmRRDY			0x00004000
+#define 	LmHOLD			0x00002000
+#define 	LmROK			0x00001000
+#define 	LmRIP			0x00000800
+#define 	LmCRBLK			0x00000400
+#define 	LmACK			0x00000200
+#define 	LmNAK			0x00000100
+#define 	LmHARDRST		0x00000080
+#define 	LmERROR			0x00000040
+#define 	LmRERR			0x00000020
+#define 	LmPMREQP		0x00000010
+#define 	LmPMREQS		0x00000008
+#define 	LmPMACK			0x00000004
+#define 	LmPMNAK			0x00000002
+#define 	LmDMAT			0x00000001
+
+/* mode 1 */
+#define	LmMnSATAFS(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x7E)
+#define	LmMnXMTSIZE(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0x93)
+
+/* mode 0 */
+#define LmMnFRMERR(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0xB0)
+
+#define		LmACRCERR		0x00000800
+#define		LmPHYOVRN		0x00000400
+#define		LmOBOVRN		0x00000200
+#define 	LmMnZERODATA		0x00000100
+#define		LmSATAINTLK		0x00000080
+#define		LmMnCRCERR		0x00000020
+#define		LmRRDYOVRN		0x00000010
+#define		LmMISSSOAF		0x00000008
+#define		LmMISSSOF		0x00000004
+#define		LmMISSEOAF		0x00000002
+#define		LmMISSEOF		0x00000001
+
+#define LmFRMERREN(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0xB4)
+
+#define 	EN_LmACRCERR		0x00000800
+#define 	EN_LmPHYOVRN		0x00000400
+#define 	EN_LmOBOVRN		0x00000200
+#define 	EN_LmMnZERODATA		0x00000100
+#define 	EN_LmSATAINTLK		0x00000080
+#define 	EN_LmFRMBAD		0x00000040
+#define 	EN_LmMnCRCERR		0x00000020
+#define 	EN_LmRRDYOVRN		0x00000010
+#define 	EN_LmMISSSOAF		0x00000008
+#define 	EN_LmMISSSOF		0x00000004
+#define 	EN_LmMISSEOAF		0x00000002
+#define 	EN_LmMISSEOF		0x00000001
+
+#define 	LmFRMERREN_MASK  	(EN_LmSATAINTLK | EN_LmMnCRCERR | \
+					 EN_LmRRDYOVRN | EN_LmMISSSOF | \
+					 EN_LmMISSEOAF | EN_LmMISSEOF | \
+					 EN_LmACRCERR | LmPHYOVRN | \
+					 EN_LmOBOVRN | EN_LmMnZERODATA)
+
+#define LmHWTSTATEN(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0xC5)
+
+#define		EN_LmDONETO		0x80
+#define		EN_LmINVDISP		0x40
+#define		EN_LmINVDW		0x20
+#define		EN_LmDWSEVENT		0x08
+#define		EN_LmCRTTTO		0x04
+#define		EN_LmANTTTO		0x02
+#define		EN_LmBITLTTO		0x01
+
+#define		LmHWTSTATEN_MASK	(EN_LmINVDISP | EN_LmINVDW | \
+					 EN_LmDWSEVENT | EN_LmCRTTTO | \
+					 EN_LmANTTTO | EN_LmDONETO | \
+					 EN_LmBITLTTO)
+
+#define LmHWTSTAT(LinkNum) 		LmSEQ_PHY_REG(0, LinkNum, 0xC7)
+
+#define		LmDONETO		0x80
+#define		LmINVDISP		0x40
+#define		LmINVDW			0x20
+#define		LmDWSEVENT		0x08
+#define		LmCRTTTO		0x04
+#define		LmANTTTO		0x02
+#define		LmBITLTTO		0x01
+
+#define LmMnDATABUFADR(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0xC8)
+#define		LmDATABUFADR_MASK	0x0FFF
+
+#define LmMnDATABUF(LinkNum, Mode)	LmSEQ_PHY_REG(Mode, LinkNum, 0xCA)
+
+#define	LmPRIMSTAT0EN(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0xE0)
+
+#define 	EN_LmUNKNOWNP 		0x20000000
+#define 	EN_LmBREAK		0x10000000
+#define 	EN_LmDONE		0x08000000
+#define 	EN_LmOPENACPT		0x04000000
+#define 	EN_LmOPENRJCT		0x02000000
+#define 	EN_LmOPENRTRY		0x01000000
+#define 	EN_LmCLOSERV1		0x00800000
+#define 	EN_LmCLOSERV0		0x00400000
+#define 	EN_LmCLOSENORM		0x00200000
+#define 	EN_LmCLOSECLAF		0x00100000
+#define 	EN_LmNOTIFYRV2		0x00080000
+#define 	EN_LmNOTIFYRV1		0x00040000
+#define 	EN_LmNOTIFYRV0		0x00020000
+#define 	EN_LmNOTIFYSPIN		0x00010000
+#define 	EN_LmBROADRV4		0x00008000
+#define 	EN_LmBROADRV3		0x00004000
+#define 	EN_LmBROADRV2		0x00002000
+#define 	EN_LmBROADRV1		0x00001000
+#define 	EN_LmBROADRV0		0x00000800
+#define 	EN_LmBROADRVCH1		0x00000400
+#define 	EN_LmBROADRVCH0		0x00000200
+#define 	EN_LmBROADCH		0x00000100
+#define 	EN_LmAIPRVWP		0x00000080
+#define 	EN_LmAIPWP		0x00000040
+#define 	EN_LmAIPWD		0x00000020
+#define 	EN_LmAIPWC		0x00000010
+#define 	EN_LmAIPRV2		0x00000008
+#define 	EN_LmAIPRV1		0x00000004
+#define 	EN_LmAIPRV0		0x00000002
+#define 	EN_LmAIPNRML		0x00000001
+
+#define		LmPRIMSTAT0EN_MASK	(EN_LmBREAK | \
+					 EN_LmDONE | EN_LmOPENACPT | \
+					 EN_LmOPENRJCT | EN_LmOPENRTRY | \
+					 EN_LmCLOSERV1 | EN_LmCLOSERV0 | \
+					 EN_LmCLOSENORM | EN_LmCLOSECLAF | \
+					 EN_LmBROADRV4 | EN_LmBROADRV3 | \
+					 EN_LmBROADRV2 | EN_LmBROADRV1 | \
+					 EN_LmBROADRV0 | EN_LmBROADRVCH1 | \
+					 EN_LmBROADRVCH0 | EN_LmBROADCH | \
+					 EN_LmAIPRVWP | EN_LmAIPWP | \
+					 EN_LmAIPWD | EN_LmAIPWC | \
+					 EN_LmAIPRV2 | EN_LmAIPRV1 | \
+					 EN_LmAIPRV0 | EN_LmAIPNRML)
+
+#define LmPRIMSTAT1EN(LinkNum)		LmSEQ_PHY_REG(0, LinkNum, 0xE4)
+
+#define		EN_LmXRDY		0x00040000
+#define		EN_LmSYNCSRST		0x00020000
+#define		EN_LmSYNC		0x00010000
+#define 	EN_LmXHOLD		0x00008000
+#define 	EN_LmRRDY		0x00004000
+#define 	EN_LmHOLD		0x00002000
+#define 	EN_LmROK		0x00001000
+#define 	EN_LmRIP		0x00000800
+#define 	EN_LmCRBLK		0x00000400
+#define 	EN_LmACK		0x00000200
+#define 	EN_LmNAK		0x00000100
+#define 	EN_LmHARDRST		0x00000080
+#define 	EN_LmERROR		0x00000040
+#define 	EN_LmRERR		0x00000020
+#define 	EN_LmPMREQP		0x00000010
+#define 	EN_LmPMREQS		0x00000008
+#define 	EN_LmPMACK		0x00000004
+#define 	EN_LmPMNAK		0x00000002
+#define 	EN_LmDMAT		0x00000001
+
+#define LmPRIMSTAT1EN_MASK		(EN_LmHARDRST | \
+					 EN_LmSYNCSRST | \
+					 EN_LmPMREQP | EN_LmPMREQS | \
+					 EN_LmPMACK | EN_LmPMNAK)
+
+#define LmSMSTATE(LinkNum) 		LmSEQ_PHY_REG(0, LinkNum, 0xE8)
+
+#define LmSMSTATEBRK(LinkNum) 		LmSEQ_PHY_REG(0, LinkNum, 0xEC)
+
+#define LmSMDBGCTL(LinkNum) 		LmSEQ_PHY_REG(0, LinkNum, 0xF0)
+
+
+/*
+ * LmSEQ CIO Bus Mode 3 Register.
+ * Mode 3: Configuration and Setup, IOP Context SCB.
+ */
+#define LmM3SATATIMER(LinkNum) 		LmSEQ_PHY_REG(3, LinkNum, 0x48)
+
+#define LmM3INTVEC0(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x90)
+
+#define LmM3INTVEC1(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x92)
+
+#define LmM3INTVEC2(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x94)
+
+#define LmM3INTVEC3(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x96)
+
+#define LmM3INTVEC4(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x98)
+
+#define LmM3INTVEC5(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x9A)
+
+#define LmM3INTVEC6(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x9C)
+
+#define LmM3INTVEC7(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0x9E)
+
+#define LmM3INTVEC8(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0xA4)
+
+#define LmM3INTVEC9(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0xA6)
+
+#define LmM3INTVEC10(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0xB0)
+
+#define LmM3FRMGAP(LinkNum)		LmSEQ_PHY_REG(3, LinkNum, 0xB4)
+
+#define LmBITL_TIMER(LinkNum) 		LmSEQ_PHY_REG(0, LinkNum, 0xA2)
+
+#define LmWWN(LinkNum) 			LmSEQ_PHY_REG(0, LinkNum, 0xA8)
+
+
+/*
+ * LmSEQ CIO Bus Mode 5 Registers.
+ * Mode 5: Phy/OOB Control and Status.
+ */
+#define LmSEQ_OOB_REG(phy_id, reg)	LmSEQ_PHY_REG(5, (phy_id), (reg))
+
+#define OOB_BFLTR	0x100
+
+#define		BFLTR_THR_MASK		0xF0
+#define		BFLTR_TC_MASK		0x0F
+
+#define OOB_INIT_MIN	0x102
+
+#define OOB_INIT_MAX	0x104
+
+#define OOB_INIT_NEG	0x106
+
+#define	OOB_SAS_MIN	0x108
+
+#define OOB_SAS_MAX	0x10A
+
+#define OOB_SAS_NEG	0x10C
+
+#define OOB_WAKE_MIN	0x10E
+
+#define OOB_WAKE_MAX	0x110
+
+#define OOB_WAKE_NEG	0x112
+
+#define OOB_IDLE_MAX	0x114
+
+#define OOB_BURST_MAX	0x116
+
+#define OOB_DATA_KBITS	0x126
+
+#define OOB_ALIGN_0_DATA	0x12C
+
+#define OOB_ALIGN_1_DATA	0x130
+
+#define D10_2_DATA_k		0x00
+#define SYNC_DATA_k		0x02
+#define ALIGN_1_DATA_k		0x04
+#define ALIGN_0_DATA_k		0x08
+#define BURST_DATA_k		0x10
+
+#define OOB_PHY_RESET_COUNT	0x13C
+
+#define OOB_SIG_GEN	0x140
+
+#define		START_OOB		0x80
+#define		START_DWS		0x40
+#define		ALIGN_CNT3		0x30
+#define 	ALIGN_CNT2		0x20
+#define 	ALIGN_CNT1		0x10
+#define 	ALIGN_CNT4		0x00
+#define		STOP_DWS		0x08
+#define		SEND_COMSAS		0x04
+#define		SEND_COMINIT		0x02
+#define		SEND_COMWAKE		0x01
+
+#define OOB_XMIT	0x141
+
+#define		TX_ENABLE		0x80
+#define		XMIT_OOB_BURST		0x10
+#define		XMIT_D10_2		0x08
+#define		XMIT_SYNC		0x04
+#define		XMIT_ALIGN_1		0x02
+#define		XMIT_ALIGN_0		0x01
+
+#define FUNCTION_MASK	0x142
+
+#define		SAS_MODE_DIS		0x80
+#define		SATA_MODE_DIS		0x40
+#define		SPINUP_HOLD_DIS		0x20
+#define		HOT_PLUG_DIS		0x10
+#define		SATA_PS_DIS		0x08
+#define		FUNCTION_MASK_DEFAULT	(SPINUP_HOLD_DIS | SATA_PS_DIS)
+
+#define OOB_MODE	0x143
+
+#define		SAS_MODE		0x80
+#define		SATA_MODE		0x40
+#define		SLOW_CLK		0x20
+#define		FORCE_XMIT_15		0x08
+#define		PHY_SPEED_60		0x04
+#define		PHY_SPEED_30		0x02
+#define		PHY_SPEED_15		0x01
+
+#define	CURRENT_STATUS	0x144
+
+#define		CURRENT_OOB_DONE	0x80
+#define		CURRENT_LOSS_OF_SIGNAL	0x40
+#define		CURRENT_SPINUP_HOLD	0x20
+#define		CURRENT_HOT_PLUG_CNCT	0x10
+#define		CURRENT_GTO_TIMEOUT	0x08
+#define		CURRENT_OOB_TIMEOUT	0x04
+#define		CURRENT_DEVICE_PRESENT	0x02
+#define		CURRENT_OOB_ERROR	0x01
+
+#define 	CURRENT_OOB1_ERROR	(CURRENT_HOT_PLUG_CNCT | \
+					 CURRENT_GTO_TIMEOUT)
+
+#define 	CURRENT_OOB2_ERROR	(CURRENT_HOT_PLUG_CNCT | \
+					 CURRENT_OOB_ERROR)
+
+#define		DEVICE_ADDED_W_CNT	(CURRENT_OOB_DONE | \
+					 CURRENT_HOT_PLUG_CNCT | \
+					 CURRENT_DEVICE_PRESENT)
+
+#define		DEVICE_ADDED_WO_CNT	(CURRENT_OOB_DONE | \
+					 CURRENT_DEVICE_PRESENT)
+
+#define 	DEVICE_REMOVED		CURRENT_LOSS_OF_SIGNAL
+
+#define		CURRENT_PHY_MASK	(CURRENT_OOB_DONE | \
+					 CURRENT_LOSS_OF_SIGNAL | \
+					 CURRENT_SPINUP_HOLD | \
+					 CURRENT_HOT_PLUG_CNCT | \
+					 CURRENT_GTO_TIMEOUT | \
+					 CURRENT_DEVICE_PRESENT | \
+					 CURRENT_OOB_ERROR )
+
+#define		CURRENT_ERR_MASK	(CURRENT_LOSS_OF_SIGNAL | \
+					 CURRENT_GTO_TIMEOUT | \
+					 CURRENT_OOB_TIMEOUT | \
+					 CURRENT_OOB_ERROR )
+
+#define SPEED_MASK	0x145
+
+#define		SATA_SPEED_30_DIS	0x10
+#define		SATA_SPEED_15_DIS	0x08
+#define		SAS_SPEED_60_DIS	0x04
+#define		SAS_SPEED_30_DIS	0x02
+#define		SAS_SPEED_15_DIS	0x01
+#define		SAS_SPEED_MASK_DEFAULT	0x00
+
+#define OOB_TIMER_ENABLE	0x14D
+
+#define		HOT_PLUG_EN		0x80
+#define		RCD_EN			0x40
+#define 	COMTIMER_EN		0x20
+#define		SNTT_EN			0x10
+#define		SNLT_EN			0x04
+#define		SNWT_EN			0x02
+#define		ALIGN_EN		0x01
+
+#define OOB_STATUS		0x14E
+
+#define		OOB_DONE		0x80
+#define		LOSS_OF_SIGNAL		0x40		/* ro */
+#define		SPINUP_HOLD		0x20
+#define		HOT_PLUG_CNCT		0x10		/* ro */
+#define		GTO_TIMEOUT		0x08		/* ro */
+#define		OOB_TIMEOUT		0x04		/* ro */
+#define		DEVICE_PRESENT		0x02		/* ro */
+#define		OOB_ERROR		0x01		/* ro */
+
+#define		OOB_STATUS_ERROR_MASK	(LOSS_OF_SIGNAL | GTO_TIMEOUT | \
+					 OOB_TIMEOUT | OOB_ERROR)
+
+#define OOB_STATUS_CLEAR	0x14F
+
+#define		OOB_DONE_CLR		0x80
+#define		LOSS_OF_SIGNAL_CLR 	0x40
+#define		SPINUP_HOLD_CLR		0x20
+#define		HOT_PLUG_CNCT_CLR     	0x10
+#define		GTO_TIMEOUT_CLR		0x08
+#define		OOB_TIMEOUT_CLR		0x04
+#define		OOB_ERROR_CLR		0x01
+
+#define HOT_PLUG_DELAY		0x150
+/* In 5 ms units. 20 = 100 ms. */
+#define	HOTPLUG_DELAY_TIMEOUT		20
+
+
+#define INT_ENABLE_2		0x15A
+
+#define		OOB_DONE_EN		0x80
+#define		LOSS_OF_SIGNAL_EN	0x40
+#define		SPINUP_HOLD_EN		0x20
+#define		HOT_PLUG_CNCT_EN	0x10
+#define		GTO_TIMEOUT_EN		0x08
+#define		OOB_TIMEOUT_EN		0x04
+#define		DEVICE_PRESENT_EN	0x02
+#define		OOB_ERROR_EN		0x01
+
+#define PHY_CONTROL_0		0x160
+
+#define		PHY_LOWPWREN_TX		0x80
+#define		PHY_LOWPWREN_RX		0x40
+#define		SPARE_REG_160_B5	0x20
+#define		OFFSET_CANCEL_RX	0x10
+
+/* bits 3:2 */
+#define		PHY_RXCOMCENTER_60V	0x00
+#define		PHY_RXCOMCENTER_70V	0x04
+#define		PHY_RXCOMCENTER_80V	0x08
+#define		PHY_RXCOMCENTER_90V	0x0C
+#define 	PHY_RXCOMCENTER_MASK	0x0C
+
+#define		PHY_RESET		0x02
+#define		SAS_DEFAULT_SEL		0x01
+
+#define PHY_CONTROL_1		0x161
+
+/* bits 2:0 */
+#define		SATA_PHY_DETLEVEL_50mv	0x00
+#define		SATA_PHY_DETLEVEL_75mv	0x01
+#define		SATA_PHY_DETLEVEL_100mv	0x02
+#define		SATA_PHY_DETLEVEL_125mv	0x03
+#define		SATA_PHY_DETLEVEL_150mv	0x04
+#define		SATA_PHY_DETLEVEL_175mv	0x05
+#define		SATA_PHY_DETLEVEL_200mv	0x06
+#define		SATA_PHY_DETLEVEL_225mv	0x07
+#define		SATA_PHY_DETLEVEL_MASK	0x07
+
+/* bits 5:3 */
+#define		SAS_PHY_DETLEVEL_50mv	0x00
+#define		SAS_PHY_DETLEVEL_75mv	0x08
+#define		SAS_PHY_DETLEVEL_100mv	0x10
+#define		SAS_PHY_DETLEVEL_125mv	0x11
+#define		SAS_PHY_DETLEVEL_150mv	0x20
+#define		SAS_PHY_DETLEVEL_175mv	0x21
+#define		SAS_PHY_DETLEVEL_200mv	0x30
+#define		SAS_PHY_DETLEVEL_225mv	0x31
+#define		SAS_PHY_DETLEVEL_MASK	0x38
+
+#define PHY_CONTROL_2		0x162
+
+/* bits 7:5 */
+#define 	SATA_PHY_DRV_400mv	0x00
+#define 	SATA_PHY_DRV_450mv	0x20
+#define 	SATA_PHY_DRV_500mv	0x40
+#define 	SATA_PHY_DRV_550mv	0x60
+#define 	SATA_PHY_DRV_600mv	0x80
+#define 	SATA_PHY_DRV_650mv	0xA0
+#define 	SATA_PHY_DRV_725mv	0xC0
+#define 	SATA_PHY_DRV_800mv	0xE0
+#define		SATA_PHY_DRV_MASK	0xE0
+
+/* bits 4:3 */
+#define 	SATA_PREEMP_0		0x00
+#define 	SATA_PREEMP_1		0x08
+#define 	SATA_PREEMP_2		0x10
+#define 	SATA_PREEMP_3		0x18
+#define 	SATA_PREEMP_MASK	0x18
+
+#define 	SATA_CMSH1P5		0x04
+
+/* bits 1:0 */
+#define 	SATA_SLEW_0		0x00
+#define 	SATA_SLEW_1		0x01
+#define 	SATA_SLEW_2		0x02
+#define 	SATA_SLEW_3		0x03
+#define 	SATA_SLEW_MASK		0x03
+
+#define PHY_CONTROL_3		0x163
+
+/* bits 7:5 */
+#define 	SAS_PHY_DRV_400mv	0x00
+#define 	SAS_PHY_DRV_450mv	0x20
+#define 	SAS_PHY_DRV_500mv	0x40
+#define 	SAS_PHY_DRV_550mv	0x60
+#define 	SAS_PHY_DRV_600mv	0x80
+#define 	SAS_PHY_DRV_650mv	0xA0
+#define 	SAS_PHY_DRV_725mv	0xC0
+#define 	SAS_PHY_DRV_800mv	0xE0
+#define		SAS_PHY_DRV_MASK	0xE0
+
+/* bits 4:3 */
+#define 	SAS_PREEMP_0		0x00
+#define 	SAS_PREEMP_1		0x08
+#define 	SAS_PREEMP_2		0x10
+#define 	SAS_PREEMP_3		0x18
+#define 	SAS_PREEMP_MASK		0x18
+
+#define 	SAS_CMSH1P5		0x04
+
+/* bits 1:0 */
+#define 	SAS_SLEW_0		0x00
+#define 	SAS_SLEW_1		0x01
+#define 	SAS_SLEW_2		0x02
+#define 	SAS_SLEW_3		0x03
+#define 	SAS_SLEW_MASK		0x03
+
+#define PHY_CONTROL_4		0x168
+
+#define		PHY_DONE_CAL_TX		0x80
+#define		PHY_DONE_CAL_RX		0x40
+#define		RX_TERM_LOAD_DIS	0x20
+#define		TX_TERM_LOAD_DIS	0x10
+#define		AUTO_TERM_CAL_DIS	0x08
+#define		PHY_SIGDET_FLTR_EN	0x04
+#define		OSC_FREQ		0x02
+#define		PHY_START_CAL		0x01
+
+/*
+ * HST_PCIX2 Registers, Addresss Range: (0x00-0xFC)
+ */
+#define PCIX_REG_BASE_ADR		0xB8040000
+
+#define PCIC_VENDOR_ID	0x00
+
+#define PCIC_DEVICE_ID	0x02
+
+#define PCIC_COMMAND	0x04
+
+#define		INT_DIS			0x0400
+#define		FBB_EN			0x0200		/* ro */
+#define		SERR_EN			0x0100
+#define		STEP_EN			0x0080		/* ro */
+#define		PERR_EN			0x0040
+#define		VGA_EN			0x0020		/* ro */
+#define		MWI_EN			0x0010
+#define		SPC_EN			0x0008
+#define		MST_EN			0x0004
+#define		MEM_EN			0x0002
+#define		IO_EN			0x0001
+
+#define	PCIC_STATUS	0x06
+
+#define		PERR_DET		0x8000
+#define		SERR_GEN		0x4000
+#define		MABT_DET		0x2000
+#define		TABT_DET		0x1000
+#define		TABT_GEN		0x0800
+#define		DPERR_DET		0x0100
+#define		CAP_LIST		0x0010
+#define		INT_STAT		0x0008
+
+#define	PCIC_DEVREV_ID	0x08
+
+#define	PCIC_CLASS_CODE	0x09
+
+#define	PCIC_CACHELINE_SIZE	0x0C
+
+#define	PCIC_MBAR0	0x10
+
+#define 	PCIC_MBAR0_OFFSET	0
+
+#define	PCIC_MBAR1	0x18
+
+#define 	PCIC_MBAR1_OFFSET	2
+
+#define	PCIC_IOBAR	0x20
+
+#define 	PCIC_IOBAR_OFFSET	4
+
+#define	PCIC_SUBVENDOR_ID	0x2C
+
+#define PCIC_SUBSYTEM_ID	0x2E
+
+#define PCIX_STATUS		0x44
+#define 	RCV_SCE		0x20000000
+#define 	UNEXP_SC	0x00080000
+#define 	SC_DISCARD	0x00040000
+
+#define ECC_CTRL_STAT		0x48
+#define 	UNCOR_ECCERR	0x00000008
+
+#define PCIC_PM_CSR		0x5C
+
+#define		PWR_STATE_D0		0
+#define		PWR_STATE_D1		1	/* not supported */
+#define		PWR_STATE_D2		2 	/* not supported */
+#define		PWR_STATE_D3		3
+
+#define PCIC_BASE1	0x6C	/* internal use only */
+
+#define		BASE1_RSVD		0xFFFFFFF8
+
+#define PCIC_BASEA	0x70	/* internal use only */
+
+#define		BASEA_RSVD		0xFFFFFFC0
+#define 	BASEA_START		0
+
+#define PCIC_BASEB	0x74	/* internal use only */
+
+#define		BASEB_RSVD		0xFFFFFF80
+#define		BASEB_IOMAP_MASK	0x7F
+#define 	BASEB_START		0x80
+
+#define PCIC_BASEC	0x78	/* internal use only */
+
+#define		BASEC_RSVD		0xFFFFFFFC
+#define 	BASEC_MASK		0x03
+#define 	BASEC_START		0x58
+
+#define PCIC_MBAR_KEY	0x7C	/* internal use only */
+
+#define 	MBAR_KEY_MASK		0xFFFFFFFF
+
+#define PCIC_HSTPCIX_CNTRL	0xA0
+
+#define 	REWIND_DIS		0x0800
+#define		SC_TMR_DIS		0x04000000
+
+#define PCIC_MBAR0_MASK	0xA8
+#define		PCIC_MBAR0_SIZE_MASK 	0x1FFFE000
+#define		PCIC_MBAR0_SIZE_SHIFT 	13
+#define		PCIC_MBAR0_SIZE(val)	\
+		    (((val) & PCIC_MBAR0_SIZE_MASK) >> PCIC_MBAR0_SIZE_SHIFT)
+
+#define PCIC_FLASH_MBAR	0xB8
+
+#define PCIC_INTRPT_STAT 0xD4
+
+#define PCIC_TP_CTRL	0xFC
+
+/*
+ * EXSI Registers, Addresss Range: (0x00-0xFC)
+ */
+#define EXSI_REG_BASE_ADR		REG_BASE_ADDR_EXSI
+
+#define	EXSICNFGR	(EXSI_REG_BASE_ADR + 0x00)
+
+#define		OCMINITIALIZED		0x80000000
+#define		ASIEN			0x00400000
+#define		HCMODE			0x00200000
+#define		PCIDEF			0x00100000
+#define		COMSTOCK		0x00080000
+#define		SEEPROMEND		0x00040000
+#define		MSTTIMEN		0x00020000
+#define		XREGEX			0x00000200
+#define		NVRAMW			0x00000100
+#define		NVRAMEX			0x00000080
+#define		SRAMW			0x00000040
+#define		SRAMEX			0x00000020
+#define		FLASHW			0x00000010
+#define		FLASHEX			0x00000008
+#define		SEEPROMCFG		0x00000004
+#define		SEEPROMTYP		0x00000002
+#define		SEEPROMEX		0x00000001
+
+
+#define EXSICNTRLR	(EXSI_REG_BASE_ADR + 0x04)
+
+#define		MODINT_EN		0x00000001
+
+
+#define PMSTATR		(EXSI_REG_BASE_ADR + 0x10)
+
+#define		FLASHRST		0x00000002
+#define		FLASHRDY		0x00000001
+
+
+#define FLCNFGR		(EXSI_REG_BASE_ADR + 0x14)
+
+#define		FLWEH_MASK		0x30000000
+#define		FLWESU_MASK		0x0C000000
+#define		FLWEPW_MASK		0x03F00000
+#define		FLOEH_MASK		0x000C0000
+#define 	FLOESU_MASK		0x00030000
+#define 	FLOEPW_MASK		0x0000FC00
+#define 	FLCSH_MASK		0x00000300
+#define 	FLCSSU_MASK		0x000000C0
+#define 	FLCSPW_MASK		0x0000003F
+
+#define SRCNFGR		(EXSI_REG_BASE_ADR + 0x18)
+
+#define		SRWEH_MASK		0x30000000
+#define		SRWESU_MASK		0x0C000000
+#define		SRWEPW_MASK		0x03F00000
+
+#define		SROEH_MASK		0x000C0000
+#define 	SROESU_MASK		0x00030000
+#define 	SROEPW_MASK		0x0000FC00
+#define		SRCSH_MASK		0x00000300
+#define		SRCSSU_MASK		0x000000C0
+#define		SRCSPW_MASK		0x0000003F
+
+#define NVCNFGR		(EXSI_REG_BASE_ADR + 0x1C)
+
+#define 	NVWEH_MASK		0x30000000
+#define 	NVWESU_MASK		0x0C000000
+#define 	NVWEPW_MASK		0x03F00000
+#define 	NVOEH_MASK		0x000C0000
+#define 	NVOESU_MASK		0x00030000
+#define 	NVOEPW_MASK		0x0000FC00
+#define 	NVCSH_MASK		0x00000300
+#define 	NVCSSU_MASK		0x000000C0
+#define 	NVCSPW_MASK		0x0000003F
+
+#define XRCNFGR		(EXSI_REG_BASE_ADR + 0x20)
+
+#define 	XRWEH_MASK		0x30000000
+#define 	XRWESU_MASK		0x0C000000
+#define 	XRWEPW_MASK		0x03F00000
+#define 	XROEH_MASK		0x000C0000
+#define 	XROESU_MASK		0x00030000
+#define 	XROEPW_MASK		0x0000FC00
+#define 	XRCSH_MASK		0x00000300
+#define 	XRCSSU_MASK		0x000000C0
+#define		XRCSPW_MASK		0x0000003F
+
+#define XREGADDR	(EXSI_REG_BASE_ADR + 0x24)
+
+#define 	XRADDRINCEN		0x80000000
+#define 	XREGADD_MASK		0x007FFFFF
+
+
+#define XREGDATAR	(EXSI_REG_BASE_ADR + 0x28)
+
+#define		XREGDATA_MASK 		0x0000FFFF
+
+#define GPIOOER		(EXSI_REG_BASE_ADR + 0x40)
+
+#define GPIOODENR	(EXSI_REG_BASE_ADR + 0x44)
+
+#define GPIOINVR	(EXSI_REG_BASE_ADR + 0x48)
+
+#define GPIODATAOR	(EXSI_REG_BASE_ADR + 0x4C)
+
+#define GPIODATAIR	(EXSI_REG_BASE_ADR + 0x50)
+
+#define GPIOCNFGR	(EXSI_REG_BASE_ADR + 0x54)
+
+#define		GPIO_EXTSRC		0x00000001
+
+#define SCNTRLR		(EXSI_REG_BASE_ADR + 0xA0)
+
+#define 	SXFERDONE		0x00000100
+#define 	SXFERCNT_MASK		0x000000E0
+#define 	SCMDTYP_MASK		0x0000001C
+#define 	SXFERSTART		0x00000002
+#define 	SXFEREN			0x00000001
+
+#define	SRATER		(EXSI_REG_BASE_ADR + 0xA4)
+
+#define	SADDRR		(EXSI_REG_BASE_ADR + 0xA8)
+
+#define 	SADDR_MASK		0x0000FFFF
+
+#define SDATAOR		(EXSI_REG_BASE_ADR + 0xAC)
+
+#define	SDATAOR0	(EXSI_REG_BASE_ADR + 0xAC)
+#define SDATAOR1	(EXSI_REG_BASE_ADR + 0xAD)
+#define SDATAOR2	(EXSI_REG_BASE_ADR + 0xAE)
+#define SDATAOR3	(EXSI_REG_BASE_ADR + 0xAF)
+
+#define SDATAIR		(EXSI_REG_BASE_ADR + 0xB0)
+
+#define SDATAIR0	(EXSI_REG_BASE_ADR + 0xB0)
+#define SDATAIR1	(EXSI_REG_BASE_ADR + 0xB1)
+#define SDATAIR2	(EXSI_REG_BASE_ADR + 0xB2)
+#define SDATAIR3	(EXSI_REG_BASE_ADR + 0xB3)
+
+#define ASISTAT0R	(EXSI_REG_BASE_ADR + 0xD0)
+#define 	ASIFMTERR		0x00000400
+#define 	ASISEECHKERR		0x00000200
+#define 	ASIERR			0x00000100
+
+#define ASISTAT1R	(EXSI_REG_BASE_ADR + 0xD4)
+#define 	CHECKSUM_MASK		0x0000FFFF
+
+#define ASIERRADDR	(EXSI_REG_BASE_ADR + 0xD8)
+#define ASIERRDATAR	(EXSI_REG_BASE_ADR + 0xDC)
+#define ASIERRSTATR	(EXSI_REG_BASE_ADR + 0xE0)
+#define 	CPI2ASIBYTECNT_MASK	0x00070000
+#define 	CPI2ASIBYTEEN_MASK      0x0000F000
+#define 	CPI2ASITARGERR_MASK	0x00000F00
+#define 	CPI2ASITARGMID_MASK	0x000000F0
+#define 	CPI2ASIMSTERR_MASK	0x0000000F
+
+/*
+ * XSRAM, External SRAM (DWord and any BE pattern accessible)
+ */
+#define XSRAM_REG_BASE_ADDR             0xB8100000
+#define XSRAM_SIZE                        0x100000
+
+/*
+ * NVRAM Registers, Address Range: (0x00000 - 0x3FFFF).
+ */
+#define		NVRAM_REG_BASE_ADR	0xBF800000
+#define		NVRAM_MAX_BASE_ADR	0x003FFFFF
+
+/* OCM base address */
+#define		OCM_BASE_ADDR		0xA0000000
+#define		OCM_MAX_SIZE		0x20000
+
+/*
+ * Sequencers (Central and Link) Scratch RAM page definitions.
+ */
+
+/*
+ * The Central Management Sequencer (CSEQ) Scratch Memory is a 1024
+ * byte memory.  It is dword accessible and has byte parity
+ * protection. The CSEQ accesses it in 32 byte windows, either as mode
+ * dependent or mode independent memory. Each mode has 96 bytes,
+ * (three 32 byte pages 0-2, not contiguous), leaving 128 bytes of
+ * Mode Independent memory (four 32 byte pages 3-7). Note that mode
+ * dependent scratch memory, Mode 8, page 0-3 overlaps mode
+ * independent scratch memory, pages 0-3.
+ * - 896 bytes of mode dependent scratch, 96 bytes per Modes 0-7, and
+ * 128 bytes in mode 8,
+ * - 259 bytes of mode independent scratch, common to modes 0-15.
+ *
+ * Sequencer scratch RAM is 1024 bytes.  This scratch memory is
+ * divided into mode dependent and mode independent scratch with this
+ * memory further subdivided into pages of size 32 bytes. There are 5
+ * pages (160 bytes) of mode independent scratch and 3 pages of
+ * dependent scratch memory for modes 0-7 (768 bytes). Mode 8 pages
+ * 0-2 dependent scratch overlap with pages 0-2 of mode independent
+ * scratch memory.
+ *
+ * The host accesses this scratch in a different manner from the
+ * central sequencer. The sequencer has to use CSEQ registers CSCRPAGE
+ * and CMnSCRPAGE to access the scratch memory. A flat mapping of the
+ * scratch memory is avaliable for software convenience and to prevent
+ * corruption while the sequencer is running. This memory is mapped
+ * onto addresses 800h - BFFh, total of 400h bytes.
+ *
+ * These addresses are mapped as follows:
+ *
+ *        800h-83Fh   Mode Dependent Scratch Mode 0 Pages 0-1
+ *        840h-87Fh   Mode Dependent Scratch Mode 1 Pages 0-1
+ *        880h-8BFh   Mode Dependent Scratch Mode 2 Pages 0-1
+ *        8C0h-8FFh   Mode Dependent Scratch Mode 3 Pages 0-1
+ *        900h-93Fh   Mode Dependent Scratch Mode 4 Pages 0-1
+ *        940h-97Fh   Mode Dependent Scratch Mode 5 Pages 0-1
+ *        980h-9BFh   Mode Dependent Scratch Mode 6 Pages 0-1
+ *        9C0h-9FFh   Mode Dependent Scratch Mode 7 Pages 0-1
+ *        A00h-A5Fh   Mode Dependent Scratch Mode 8 Pages 0-2
+ *                    Mode Independent Scratch Pages 0-2
+ *        A60h-A7Fh   Mode Dependent Scratch Mode 8 Page 3
+ *                    Mode Independent Scratch Page 3
+ *        A80h-AFFh   Mode Independent Scratch Pages 4-7
+ *        B00h-B1Fh   Mode Dependent Scratch Mode 0 Page 2
+ *        B20h-B3Fh   Mode Dependent Scratch Mode 1 Page 2
+ *        B40h-B5Fh   Mode Dependent Scratch Mode 2 Page 2
+ *        B60h-B7Fh   Mode Dependent Scratch Mode 3 Page 2
+ *        B80h-B9Fh   Mode Dependent Scratch Mode 4 Page 2
+ *        BA0h-BBFh   Mode Dependent Scratch Mode 5 Page 2
+ *        BC0h-BDFh   Mode Dependent Scratch Mode 6 Page 2
+ *        BE0h-BFFh   Mode Dependent Scratch Mode 7 Page 2
+ */
+
+/* General macros */
+#define CSEQ_PAGE_SIZE			32  /* Scratch page size (in bytes) */
+
+/* All macros start with offsets from base + 0x800 (CMAPPEDSCR).
+ * Mode dependent scratch page 0, mode 0.
+ * For modes 1-7 you have to do arithmetic. */
+#define CSEQ_LRM_SAVE_SINDEX		(CMAPPEDSCR + 0x0000)
+#define CSEQ_LRM_SAVE_SCBPTR		(CMAPPEDSCR + 0x0002)
+#define CSEQ_Q_LINK_HEAD		(CMAPPEDSCR + 0x0004)
+#define CSEQ_Q_LINK_TAIL		(CMAPPEDSCR + 0x0006)
+#define CSEQ_LRM_SAVE_SCRPAGE		(CMAPPEDSCR + 0x0008)
+
+/* Mode dependent scratch page 0 mode 8 macros. */
+#define CSEQ_RET_ADDR			(CMAPPEDSCR + 0x0200)
+#define CSEQ_RET_SCBPTR			(CMAPPEDSCR + 0x0202)
+#define CSEQ_SAVE_SCBPTR		(CMAPPEDSCR + 0x0204)
+#define CSEQ_EMPTY_TRANS_CTX		(CMAPPEDSCR + 0x0206)
+#define CSEQ_RESP_LEN			(CMAPPEDSCR + 0x0208)
+#define CSEQ_TMF_SCBPTR			(CMAPPEDSCR + 0x020A)
+#define CSEQ_GLOBAL_PREV_SCB		(CMAPPEDSCR + 0x020C)
+#define CSEQ_GLOBAL_HEAD		(CMAPPEDSCR + 0x020E)
+#define CSEQ_CLEAR_LU_HEAD		(CMAPPEDSCR + 0x0210)
+#define CSEQ_TMF_OPCODE			(CMAPPEDSCR + 0x0212)
+#define CSEQ_SCRATCH_FLAGS		(CMAPPEDSCR + 0x0213)
+#define CSEQ_HSB_SITE                   (CMAPPEDSCR + 0x021A)
+#define CSEQ_FIRST_INV_SCB_SITE		(CMAPPEDSCR + 0x021C)
+#define CSEQ_FIRST_INV_DDB_SITE		(CMAPPEDSCR + 0x021E)
+
+/* Mode dependent scratch page 1 mode 8 macros. */
+#define CSEQ_LUN_TO_CLEAR		(CMAPPEDSCR + 0x0220)
+#define CSEQ_LUN_TO_CHECK		(CMAPPEDSCR + 0x0228)
+
+/* Mode dependent scratch page 2 mode 8 macros */
+#define CSEQ_HQ_NEW_POINTER		(CMAPPEDSCR + 0x0240)
+#define CSEQ_HQ_DONE_BASE		(CMAPPEDSCR + 0x0248)
+#define CSEQ_HQ_DONE_POINTER		(CMAPPEDSCR + 0x0250)
+#define CSEQ_HQ_DONE_PASS		(CMAPPEDSCR + 0x0254)
+
+/* Mode independent scratch page 4 macros. */
+#define CSEQ_Q_EXE_HEAD			(CMAPPEDSCR + 0x0280)
+#define CSEQ_Q_EXE_TAIL			(CMAPPEDSCR + 0x0282)
+#define CSEQ_Q_DONE_HEAD                (CMAPPEDSCR + 0x0284)
+#define CSEQ_Q_DONE_TAIL                (CMAPPEDSCR + 0x0286)
+#define CSEQ_Q_SEND_HEAD		(CMAPPEDSCR + 0x0288)
+#define CSEQ_Q_SEND_TAIL		(CMAPPEDSCR + 0x028A)
+#define CSEQ_Q_DMA2CHIM_HEAD		(CMAPPEDSCR + 0x028C)
+#define CSEQ_Q_DMA2CHIM_TAIL		(CMAPPEDSCR + 0x028E)
+#define CSEQ_Q_COPY_HEAD		(CMAPPEDSCR + 0x0290)
+#define CSEQ_Q_COPY_TAIL		(CMAPPEDSCR + 0x0292)
+#define CSEQ_REG0			(CMAPPEDSCR + 0x0294)
+#define CSEQ_REG1			(CMAPPEDSCR + 0x0296)
+#define CSEQ_REG2			(CMAPPEDSCR + 0x0298)
+#define CSEQ_LINK_CTL_Q_MAP		(CMAPPEDSCR + 0x029C)
+#define CSEQ_MAX_CSEQ_MODE		(CMAPPEDSCR + 0x029D)
+#define CSEQ_FREE_LIST_HACK_COUNT	(CMAPPEDSCR + 0x029E)
+
+/* Mode independent scratch page 5 macros. */
+#define CSEQ_EST_NEXUS_REQ_QUEUE	(CMAPPEDSCR + 0x02A0)
+#define CSEQ_EST_NEXUS_REQ_COUNT	(CMAPPEDSCR + 0x02A8)
+#define CSEQ_Q_EST_NEXUS_HEAD		(CMAPPEDSCR + 0x02B0)
+#define CSEQ_Q_EST_NEXUS_TAIL		(CMAPPEDSCR + 0x02B2)
+#define CSEQ_NEED_EST_NEXUS_SCB		(CMAPPEDSCR + 0x02B4)
+#define CSEQ_EST_NEXUS_REQ_HEAD		(CMAPPEDSCR + 0x02B6)
+#define CSEQ_EST_NEXUS_REQ_TAIL		(CMAPPEDSCR + 0x02B7)
+#define CSEQ_EST_NEXUS_SCB_OFFSET	(CMAPPEDSCR + 0x02B8)
+
+/* Mode independent scratch page 6 macros. */
+#define CSEQ_INT_ROUT_RET_ADDR0		(CMAPPEDSCR + 0x02C0)
+#define CSEQ_INT_ROUT_RET_ADDR1		(CMAPPEDSCR + 0x02C2)
+#define CSEQ_INT_ROUT_SCBPTR		(CMAPPEDSCR + 0x02C4)
+#define CSEQ_INT_ROUT_MODE		(CMAPPEDSCR + 0x02C6)
+#define CSEQ_ISR_SCRATCH_FLAGS		(CMAPPEDSCR + 0x02C7)
+#define CSEQ_ISR_SAVE_SINDEX		(CMAPPEDSCR + 0x02C8)
+#define CSEQ_ISR_SAVE_DINDEX		(CMAPPEDSCR + 0x02CA)
+#define CSEQ_Q_MONIRTT_HEAD		(CMAPPEDSCR + 0x02D0)
+#define CSEQ_Q_MONIRTT_TAIL		(CMAPPEDSCR + 0x02D2)
+#define CSEQ_FREE_SCB_MASK		(CMAPPEDSCR + 0x02D5)
+#define CSEQ_BUILTIN_FREE_SCB_HEAD	(CMAPPEDSCR + 0x02D6)
+#define CSEQ_BUILTIN_FREE_SCB_TAIL	(CMAPPEDSCR + 0x02D8)
+#define CSEQ_EXTENDED_FREE_SCB_HEAD	(CMAPPEDSCR + 0x02DA)
+#define CSEQ_EXTENDED_FREE_SCB_TAIL	(CMAPPEDSCR + 0x02DC)
+
+/* Mode independent scratch page 7 macros. */
+#define CSEQ_EMPTY_REQ_QUEUE		(CMAPPEDSCR + 0x02E0)
+#define CSEQ_EMPTY_REQ_COUNT		(CMAPPEDSCR + 0x02E8)
+#define CSEQ_Q_EMPTY_HEAD		(CMAPPEDSCR + 0x02F0)
+#define CSEQ_Q_EMPTY_TAIL		(CMAPPEDSCR + 0x02F2)
+#define CSEQ_NEED_EMPTY_SCB		(CMAPPEDSCR + 0x02F4)
+#define CSEQ_EMPTY_REQ_HEAD		(CMAPPEDSCR + 0x02F6)
+#define CSEQ_EMPTY_REQ_TAIL		(CMAPPEDSCR + 0x02F7)
+#define CSEQ_EMPTY_SCB_OFFSET		(CMAPPEDSCR + 0x02F8)
+#define CSEQ_PRIMITIVE_DATA		(CMAPPEDSCR + 0x02FA)
+#define CSEQ_TIMEOUT_CONST		(CMAPPEDSCR + 0x02FC)
+
+/***************************************************************************
+* Link m Sequencer scratch RAM is 512 bytes.
+* This scratch memory is divided into mode dependent and mode
+* independent scratch with this memory further subdivided into
+* pages of size 32 bytes. There are 4 pages (128 bytes) of
+* mode independent scratch and 4 pages of dependent scratch
+* memory for modes 0-2 (384 bytes).
+*
+* The host accesses this scratch in a different manner from the
+* link sequencer. The sequencer has to use LSEQ registers
+* LmSCRPAGE and LmMnSCRPAGE to access the scratch memory. A flat
+* mapping of the scratch memory is avaliable for software
+* convenience and to prevent corruption while the sequencer is
+* running. This memory is mapped onto addresses 800h - 9FFh.
+*
+* These addresses are mapped as follows:
+*
+*        800h-85Fh   Mode Dependent Scratch Mode 0 Pages 0-2
+*        860h-87Fh   Mode Dependent Scratch Mode 0 Page 3
+*                    Mode Dependent Scratch Mode 5 Page 0
+*        880h-8DFh   Mode Dependent Scratch Mode 1 Pages 0-2
+*        8E0h-8FFh   Mode Dependent Scratch Mode 1 Page 3
+*                    Mode Dependent Scratch Mode 5 Page 1
+*        900h-95Fh   Mode Dependent Scratch Mode 2 Pages 0-2
+*        960h-97Fh   Mode Dependent Scratch Mode 2 Page 3
+*                    Mode Dependent Scratch Mode 5 Page 2
+*        980h-9DFh   Mode Independent Scratch Pages 0-3
+*        9E0h-9FFh   Mode Independent Scratch Page 3
+*                    Mode Dependent Scratch Mode 5 Page 3
+*
+****************************************************************************/
+/* General macros */
+#define LSEQ_MODE_SCRATCH_SIZE		0x80 /* Size of scratch RAM per mode */
+#define LSEQ_PAGE_SIZE			0x20 /* Scratch page size (in bytes) */
+#define LSEQ_MODE5_PAGE0_OFFSET 	0x60
+
+/* Common mode dependent scratch page 0 macros for modes 0,1,2, and 5 */
+/* Indexed using LSEQ_MODE_SCRATCH_SIZE * mode, for modes 0,1,2. */
+#define LmSEQ_RET_ADDR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0000)
+#define LmSEQ_REG0_MODE(LinkNum)	(LmSCRATCH(LinkNum) + 0x0002)
+#define LmSEQ_MODE_FLAGS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0004)
+
+/* Mode flag macros (byte 0) */
+#define		SAS_SAVECTX_OCCURRED		0x80
+#define		SAS_OOBSVC_OCCURRED		0x40
+#define		SAS_OOB_DEVICE_PRESENT		0x20
+#define		SAS_CFGHDR_OCCURRED		0x10
+#define		SAS_RCV_INTS_ARE_DISABLED	0x08
+#define		SAS_OOB_HOT_PLUG_CNCT		0x04
+#define		SAS_AWAIT_OPEN_CONNECTION	0x02
+#define		SAS_CFGCMPLT_OCCURRED		0x01
+
+/* Mode flag macros (byte 1) */
+#define		SAS_RLSSCB_OCCURRED		0x80
+#define		SAS_FORCED_HEADER_MISS		0x40
+
+#define LmSEQ_RET_ADDR2(LinkNum)	(LmSCRATCH(LinkNum) + 0x0006)
+#define LmSEQ_RET_ADDR1(LinkNum)	(LmSCRATCH(LinkNum) + 0x0008)
+#define LmSEQ_OPCODE_TO_CSEQ(LinkNum)	(LmSCRATCH(LinkNum) + 0x000B)
+#define LmSEQ_DATA_TO_CSEQ(LinkNum)	(LmSCRATCH(LinkNum) + 0x000C)
+
+/* Mode dependent scratch page 0 macros for mode 0 (non-common) */
+/* Absolute offsets */
+#define LmSEQ_FIRST_INV_DDB_SITE(LinkNum)	(LmSCRATCH(LinkNum) + 0x000E)
+#define LmSEQ_EMPTY_TRANS_CTX(LinkNum)		(LmSCRATCH(LinkNum) + 0x0010)
+#define LmSEQ_RESP_LEN(LinkNum)			(LmSCRATCH(LinkNum) + 0x0012)
+#define LmSEQ_FIRST_INV_SCB_SITE(LinkNum)	(LmSCRATCH(LinkNum) + 0x0014)
+#define LmSEQ_INTEN_SAVE(LinkNum)		(LmSCRATCH(LinkNum) + 0x0016)
+#define LmSEQ_LINK_RST_FRM_LEN(LinkNum)		(LmSCRATCH(LinkNum) + 0x001A)
+#define LmSEQ_LINK_RST_PROTOCOL(LinkNum)	(LmSCRATCH(LinkNum) + 0x001B)
+#define LmSEQ_RESP_STATUS(LinkNum)		(LmSCRATCH(LinkNum) + 0x001C)
+#define LmSEQ_LAST_LOADED_SGE(LinkNum)		(LmSCRATCH(LinkNum) + 0x001D)
+#define LmSEQ_SAVE_SCBPTR(LinkNum)		(LmSCRATCH(LinkNum) + 0x001E)
+
+/* Mode dependent scratch page 0 macros for mode 1 (non-common) */
+/* Absolute offsets */
+#define LmSEQ_Q_XMIT_HEAD(LinkNum)		(LmSCRATCH(LinkNum) + 0x008E)
+#define LmSEQ_M1_EMPTY_TRANS_CTX(LinkNum)	(LmSCRATCH(LinkNum) + 0x0090)
+#define LmSEQ_INI_CONN_TAG(LinkNum)		(LmSCRATCH(LinkNum) + 0x0092)
+#define LmSEQ_FAILED_OPEN_STATUS(LinkNum)	(LmSCRATCH(LinkNum) + 0x009A)
+#define LmSEQ_XMIT_REQUEST_TYPE(LinkNum)	(LmSCRATCH(LinkNum) + 0x009B)
+#define LmSEQ_M1_RESP_STATUS(LinkNum)		(LmSCRATCH(LinkNum) + 0x009C)
+#define LmSEQ_M1_LAST_LOADED_SGE(LinkNum)	(LmSCRATCH(LinkNum) + 0x009D)
+#define LmSEQ_M1_SAVE_SCBPTR(LinkNum)		(LmSCRATCH(LinkNum) + 0x009E)
+
+/* Mode dependent scratch page 0 macros for mode 2 (non-common) */
+#define LmSEQ_PORT_COUNTER(LinkNum)		(LmSCRATCH(LinkNum) + 0x010E)
+#define LmSEQ_PM_TABLE_PTR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0110)
+#define LmSEQ_SATA_INTERLOCK_TMR_SAVE(LinkNum)	(LmSCRATCH(LinkNum) + 0x0112)
+#define LmSEQ_IP_BITL(LinkNum)			(LmSCRATCH(LinkNum) + 0x0114)
+#define LmSEQ_COPY_SMP_CONN_TAG(LinkNum)	(LmSCRATCH(LinkNum) + 0x0116)
+#define LmSEQ_P0M2_OFFS1AH(LinkNum)		(LmSCRATCH(LinkNum) + 0x011A)
+
+/* Mode dependent scratch page 0 macros for modes 4/5 (non-common) */
+/* Absolute offsets */
+#define LmSEQ_SAVED_OOB_STATUS(LinkNum)		(LmSCRATCH(LinkNum) + 0x006E)
+#define LmSEQ_SAVED_OOB_MODE(LinkNum)		(LmSCRATCH(LinkNum) + 0x006F)
+#define LmSEQ_Q_LINK_HEAD(LinkNum)		(LmSCRATCH(LinkNum) + 0x0070)
+#define LmSEQ_LINK_RST_ERR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0072)
+#define LmSEQ_SAVED_OOB_SIGNALS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0073)
+#define LmSEQ_SAS_RESET_MODE(LinkNum)		(LmSCRATCH(LinkNum) + 0x0074)
+#define LmSEQ_LINK_RESET_RETRY_COUNT(LinkNum)	(LmSCRATCH(LinkNum) + 0x0075)
+#define LmSEQ_NUM_LINK_RESET_RETRIES(LinkNum)	(LmSCRATCH(LinkNum) + 0x0076)
+#define LmSEQ_OOB_INT_ENABLES(LinkNum)		(LmSCRATCH(LinkNum) + 0x007A)
+#define LmSEQ_NOTIFY_TIMER_TIMEOUT(LinkNum)	(LmSCRATCH(LinkNum) + 0x007C)
+#define LmSEQ_NOTIFY_TIMER_DOWN_COUNT(LinkNum)	(LmSCRATCH(LinkNum) + 0x007E)
+
+/* Mode dependent scratch page 1, mode 0 and mode 1 */
+#define LmSEQ_SG_LIST_PTR_ADDR0(LinkNum)        (LmSCRATCH(LinkNum) + 0x0020)
+#define LmSEQ_SG_LIST_PTR_ADDR1(LinkNum)        (LmSCRATCH(LinkNum) + 0x0030)
+#define LmSEQ_M1_SG_LIST_PTR_ADDR0(LinkNum)     (LmSCRATCH(LinkNum) + 0x00A0)
+#define LmSEQ_M1_SG_LIST_PTR_ADDR1(LinkNum)     (LmSCRATCH(LinkNum) + 0x00B0)
+
+/* Mode dependent scratch page 1 macros for mode 2 */
+/* Absolute offsets */
+#define LmSEQ_INVALID_DWORD_COUNT(LinkNum)	(LmSCRATCH(LinkNum) + 0x0120)
+#define LmSEQ_DISPARITY_ERROR_COUNT(LinkNum) 	(LmSCRATCH(LinkNum) + 0x0124)
+#define LmSEQ_LOSS_OF_SYNC_COUNT(LinkNum)	(LmSCRATCH(LinkNum) + 0x0128)
+
+/* Mode dependent scratch page 1 macros for mode 4/5 */
+#define LmSEQ_FRAME_TYPE_MASK(LinkNum)	      (LmSCRATCH(LinkNum) + 0x00E0)
+#define LmSEQ_HASHED_DEST_ADDR_MASK(LinkNum)  (LmSCRATCH(LinkNum) + 0x00E1)
+#define LmSEQ_HASHED_SRC_ADDR_MASK_PRINT(LinkNum) (LmSCRATCH(LinkNum) + 0x00E4)
+#define LmSEQ_HASHED_SRC_ADDR_MASK(LinkNum)   (LmSCRATCH(LinkNum) + 0x00E5)
+#define LmSEQ_NUM_FILL_BYTES_MASK(LinkNum)    (LmSCRATCH(LinkNum) + 0x00EB)
+#define LmSEQ_TAG_MASK(LinkNum)		      (LmSCRATCH(LinkNum) + 0x00F0)
+#define LmSEQ_TARGET_PORT_XFER_TAG(LinkNum)   (LmSCRATCH(LinkNum) + 0x00F2)
+#define LmSEQ_DATA_OFFSET(LinkNum)	      (LmSCRATCH(LinkNum) + 0x00F4)
+
+/* Mode dependent scratch page 2 macros for mode 0 */
+/* Absolute offsets */
+#define LmSEQ_SMP_RCV_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0040)
+#define LmSEQ_DEVICE_BITS(LinkNum)		(LmSCRATCH(LinkNum) + 0x005B)
+#define LmSEQ_SDB_DDB(LinkNum)			(LmSCRATCH(LinkNum) + 0x005C)
+#define LmSEQ_SDB_NUM_TAGS(LinkNum)		(LmSCRATCH(LinkNum) + 0x005E)
+#define LmSEQ_SDB_CURR_TAG(LinkNum)		(LmSCRATCH(LinkNum) + 0x005F)
+
+/* Mode dependent scratch page 2 macros for mode 1 */
+/* Absolute offsets */
+/* byte 0 bits 1-0 are domain select. */
+#define LmSEQ_TX_ID_ADDR_FRAME(LinkNum)		(LmSCRATCH(LinkNum) + 0x00C0)
+#define LmSEQ_OPEN_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x00C8)
+#define LmSEQ_SRST_AS_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x00CC)
+#define LmSEQ_LAST_LOADED_SG_EL(LinkNum)	(LmSCRATCH(LinkNum) + 0x00D4)
+
+/* Mode dependent scratch page 2 macros for mode 2 */
+/* Absolute offsets */
+#define LmSEQ_STP_SHUTDOWN_TIMER_TERM_TS(LinkNum) (LmSCRATCH(LinkNum) + 0x0140)
+#define LmSEQ_CLOSE_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0144)
+#define LmSEQ_BREAK_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0148)
+#define LmSEQ_DWS_RESET_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x014C)
+#define LmSEQ_SATA_INTERLOCK_TIMER_TERM_TS(LinkNum) \
+						(LmSCRATCH(LinkNum) + 0x0150)
+#define LmSEQ_MCTL_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0154)
+
+/* Mode dependent scratch page 2 macros for mode 5 */
+#define LmSEQ_COMINIT_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0160)
+#define LmSEQ_RCV_ID_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0164)
+#define LmSEQ_RCV_FIS_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0168)
+#define LmSEQ_DEV_PRES_TIMER_TERM_TS(LinkNum)	(LmSCRATCH(LinkNum) + 0x016C)
+
+/* Mode dependent scratch page 3 macros for modes 0 and 1 */
+/* None defined */
+
+/* Mode dependent scratch page 3 macros for modes 2 and 5 */
+/* None defined */
+
+/* Mode Independent Scratch page 0 macros. */
+#define LmSEQ_Q_TGTXFR_HEAD(LinkNum)	(LmSCRATCH(LinkNum) + 0x0180)
+#define LmSEQ_Q_TGTXFR_TAIL(LinkNum)	(LmSCRATCH(LinkNum) + 0x0182)
+#define LmSEQ_LINK_NUMBER(LinkNum)	(LmSCRATCH(LinkNum) + 0x0186)
+#define LmSEQ_SCRATCH_FLAGS(LinkNum)	(LmSCRATCH(LinkNum) + 0x0187)
+/*
+ * Currently only bit 0, SAS_DWSAQD, is used.
+ */
+#define		SAS_DWSAQD			0x01  /*
+						       * DWSSTATUS: DWSAQD
+						       * bit las read in ISR.
+						       */
+#define  LmSEQ_CONNECTION_STATE(LinkNum) (LmSCRATCH(LinkNum) + 0x0188)
+/* Connection states (byte 0) */
+#define		SAS_WE_OPENED_CS		0x01
+#define		SAS_DEVICE_OPENED_CS		0x02
+#define		SAS_WE_SENT_DONE_CS		0x04
+#define		SAS_DEVICE_SENT_DONE_CS		0x08
+#define		SAS_WE_SENT_CLOSE_CS		0x10
+#define		SAS_DEVICE_SENT_CLOSE_CS	0x20
+#define		SAS_WE_SENT_BREAK_CS		0x40
+#define		SAS_DEVICE_SENT_BREAK_CS	0x80
+/* Connection states (byte 1) */
+#define		SAS_OPN_TIMEOUT_OR_OPN_RJCT_CS	0x01
+#define		SAS_AIP_RECEIVED_CS		0x02
+#define		SAS_CREDIT_TIMEOUT_OCCURRED_CS	0x04
+#define		SAS_ACKNAK_TIMEOUT_OCCURRED_CS	0x08
+#define		SAS_SMPRSP_TIMEOUT_OCCURRED_CS	0x10
+#define		SAS_DONE_TIMEOUT_OCCURRED_CS	0x20
+/* Connection states (byte 2) */
+#define		SAS_SMP_RESPONSE_RECEIVED_CS	0x01
+#define		SAS_INTLK_TIMEOUT_OCCURRED_CS	0x02
+#define		SAS_DEVICE_SENT_DMAT_CS		0x04
+#define		SAS_DEVICE_SENT_SYNCSRST_CS	0x08
+#define		SAS_CLEARING_AFFILIATION_CS	0x20
+#define		SAS_RXTASK_ACTIVE_CS		0x40
+#define		SAS_TXTASK_ACTIVE_CS		0x80
+/* Connection states (byte 3) */
+#define		SAS_PHY_LOSS_OF_SIGNAL_CS	0x01
+#define		SAS_DWS_TIMER_EXPIRED_CS	0x02
+#define		SAS_LINK_RESET_NOT_COMPLETE_CS	0x04
+#define		SAS_PHY_DISABLED_CS		0x08
+#define		SAS_LINK_CTL_TASK_ACTIVE_CS	0x10
+#define		SAS_PHY_EVENT_TASK_ACTIVE_CS	0x20
+#define		SAS_DEVICE_SENT_ID_FRAME_CS	0x40
+#define		SAS_DEVICE_SENT_REG_FIS_CS	0x40
+#define		SAS_DEVICE_SENT_HARD_RESET_CS	0x80
+#define  	SAS_PHY_IS_DOWN_FLAGS	(SAS_PHY_LOSS_OF_SIGNAL_CS|\
+					 SAS_DWS_TIMER_EXPIRED_CS |\
+					 SAS_LINK_RESET_NOT_COMPLETE_CS|\
+					 SAS_PHY_DISABLED_CS)
+
+#define		SAS_LINK_CTL_PHY_EVENT_FLAGS   (SAS_LINK_CTL_TASK_ACTIVE_CS |\
+						SAS_PHY_EVENT_TASK_ACTIVE_CS |\
+						SAS_DEVICE_SENT_ID_FRAME_CS  |\
+						SAS_DEVICE_SENT_HARD_RESET_CS)
+
+#define LmSEQ_CONCTL(LinkNum)		(LmSCRATCH(LinkNum) + 0x018C)
+#define LmSEQ_CONSTAT(LinkNum)		(LmSCRATCH(LinkNum) + 0x018E)
+#define LmSEQ_CONNECTION_MODES(LinkNum)	(LmSCRATCH(LinkNum) + 0x018F)
+#define LmSEQ_REG1_ISR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0192)
+#define LmSEQ_REG2_ISR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0194)
+#define LmSEQ_REG3_ISR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0196)
+#define LmSEQ_REG0_ISR(LinkNum)		(LmSCRATCH(LinkNum) + 0x0198)
+
+/* Mode independent scratch page 1 macros. */
+#define LmSEQ_EST_NEXUS_SCBPTR0(LinkNum)	(LmSCRATCH(LinkNum) + 0x01A0)
+#define LmSEQ_EST_NEXUS_SCBPTR1(LinkNum)	(LmSCRATCH(LinkNum) + 0x01A2)
+#define LmSEQ_EST_NEXUS_SCBPTR2(LinkNum)	(LmSCRATCH(LinkNum) + 0x01A4)
+#define LmSEQ_EST_NEXUS_SCBPTR3(LinkNum)	(LmSCRATCH(LinkNum) + 0x01A6)
+#define LmSEQ_EST_NEXUS_SCB_OPCODE0(LinkNum)	(LmSCRATCH(LinkNum) + 0x01A8)
+#define LmSEQ_EST_NEXUS_SCB_OPCODE1(LinkNum)	(LmSCRATCH(LinkNum) + 0x01A9)
+#define LmSEQ_EST_NEXUS_SCB_OPCODE2(LinkNum)	(LmSCRATCH(LinkNum) + 0x01AA)
+#define LmSEQ_EST_NEXUS_SCB_OPCODE3(LinkNum)	(LmSCRATCH(LinkNum) + 0x01AB)
+#define LmSEQ_EST_NEXUS_SCB_HEAD(LinkNum)	(LmSCRATCH(LinkNum) + 0x01AC)
+#define LmSEQ_EST_NEXUS_SCB_TAIL(LinkNum)	(LmSCRATCH(LinkNum) + 0x01AD)
+#define LmSEQ_EST_NEXUS_BUF_AVAIL(LinkNum)	(LmSCRATCH(LinkNum) + 0x01AE)
+#define LmSEQ_TIMEOUT_CONST(LinkNum)		(LmSCRATCH(LinkNum) + 0x01B8)
+#define LmSEQ_ISR_SAVE_SINDEX(LinkNum)	        (LmSCRATCH(LinkNum) + 0x01BC)
+#define LmSEQ_ISR_SAVE_DINDEX(LinkNum)	        (LmSCRATCH(LinkNum) + 0x01BE)
+
+/* Mode independent scratch page 2 macros. */
+#define LmSEQ_EMPTY_SCB_PTR0(LinkNum)	(LmSCRATCH(LinkNum) + 0x01C0)
+#define LmSEQ_EMPTY_SCB_PTR1(LinkNum)	(LmSCRATCH(LinkNum) + 0x01C2)
+#define LmSEQ_EMPTY_SCB_PTR2(LinkNum)	(LmSCRATCH(LinkNum) + 0x01C4)
+#define LmSEQ_EMPTY_SCB_PTR3(LinkNum)	(LmSCRATCH(LinkNum) + 0x01C6)
+#define LmSEQ_EMPTY_SCB_OPCD0(LinkNum)	(LmSCRATCH(LinkNum) + 0x01C8)
+#define LmSEQ_EMPTY_SCB_OPCD1(LinkNum)	(LmSCRATCH(LinkNum) + 0x01C9)
+#define LmSEQ_EMPTY_SCB_OPCD2(LinkNum)	(LmSCRATCH(LinkNum) + 0x01CA)
+#define LmSEQ_EMPTY_SCB_OPCD3(LinkNum)	(LmSCRATCH(LinkNum) + 0x01CB)
+#define LmSEQ_EMPTY_SCB_HEAD(LinkNum)	(LmSCRATCH(LinkNum) + 0x01CC)
+#define LmSEQ_EMPTY_SCB_TAIL(LinkNum)	(LmSCRATCH(LinkNum) + 0x01CD)
+#define LmSEQ_EMPTY_BUFS_AVAIL(LinkNum)	(LmSCRATCH(LinkNum) + 0x01CE)
+#define LmSEQ_ATA_SCR_REGS(LinkNum)	(LmSCRATCH(LinkNum) + 0x01D4)
+
+/* Mode independent scratch page 3 macros. */
+#define LmSEQ_DEV_PRES_TMR_TOUT_CONST(LinkNum)	(LmSCRATCH(LinkNum) + 0x01E0)
+#define LmSEQ_SATA_INTERLOCK_TIMEOUT(LinkNum)	(LmSCRATCH(LinkNum) + 0x01E4)
+#define LmSEQ_STP_SHUTDOWN_TIMEOUT(LinkNum)	(LmSCRATCH(LinkNum) + 0x01E8)
+#define LmSEQ_SRST_ASSERT_TIMEOUT(LinkNum)	(LmSCRATCH(LinkNum) + 0x01EC)
+#define LmSEQ_RCV_FIS_TIMEOUT(LinkNum)		(LmSCRATCH(LinkNum) + 0x01F0)
+#define LmSEQ_ONE_MILLISEC_TIMEOUT(LinkNum)	(LmSCRATCH(LinkNum) + 0x01F4)
+#define LmSEQ_TEN_MS_COMINIT_TIMEOUT(LinkNum)	(LmSCRATCH(LinkNum) + 0x01F8)
+#define LmSEQ_SMP_RCV_TIMEOUT(LinkNum)		(LmSCRATCH(LinkNum) + 0x01FC)
+
+#endif
diff --git a/drivers/scsi/aic94xx/aic94xx_sas.h b/drivers/scsi/aic94xx/aic94xx_sas.h
new file mode 100644
index 000000000000..64d231712345
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_sas.h
@@ -0,0 +1,785 @@
+/*
+ * Aic94xx SAS/SATA driver SAS definitions and hardware interface header file.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#ifndef _AIC94XX_SAS_H_
+#define _AIC94XX_SAS_H_
+
+#include <scsi/libsas.h>
+
+/* ---------- DDBs ---------- */
+/* DDBs are device descriptor blocks which describe a device in the
+ * domain that this sequencer can maintain low-level connections for
+ * us.  They are be 64 bytes.
+ */
+
+struct asd_ddb_ssp_smp_target_port {
+	u8     conn_type;	  /* byte 0 */
+#define DDB_TP_CONN_TYPE 0x81	  /* Initiator port and addr frame type 0x01 */
+
+	u8     conn_rate;
+	__be16 init_conn_tag;
+	u8     dest_sas_addr[8];  /* bytes 4-11 */
+
+	__le16 send_queue_head;
+	u8     sq_suspended;
+	u8     ddb_type;	  /* DDB_TYPE_TARGET */
+#define DDB_TYPE_UNUSED    0xFF
+#define DDB_TYPE_TARGET    0xFE
+#define DDB_TYPE_INITIATOR 0xFD
+#define DDB_TYPE_PM_PORT   0xFC
+
+	__le16 _r_a;
+	__be16 awt_def;
+
+	u8     compat_features;	  /* byte 20 */
+	u8     pathway_blocked_count;
+	__be16 arb_wait_time;
+	__be32 more_compat_features; /* byte 24 */
+
+	u8     conn_mask;
+	u8     flags;	  /* concurrent conn:2,2 and open:0(1) */
+#define CONCURRENT_CONN_SUPP 0x04
+#define OPEN_REQUIRED        0x01
+
+	u16    _r_b;
+	__le16 exec_queue_tail;
+	__le16 send_queue_tail;
+	__le16 sister_ddb;
+
+	__le16 _r_c;
+
+	u8     max_concurrent_conn;
+	u8     num_concurrent_conn;
+	u8     num_contexts;
+
+	u8     _r_d;
+
+	__le16 active_task_count;
+
+	u8     _r_e[9];
+
+	u8     itnl_reason;	  /* I_T nexus loss reason */
+
+	__le16 _r_f;
+
+	__le16 itnl_timeout;
+#define ITNL_TIMEOUT_CONST 0x7D0 /* 2 seconds */
+
+	__le32 itnl_timestamp;
+} __attribute__ ((packed));
+
+struct asd_ddb_stp_sata_target_port {
+	u8     conn_type;	  /* byte 0 */
+	u8     conn_rate;
+	__be16 init_conn_tag;
+	u8     dest_sas_addr[8];  /* bytes 4-11 */
+
+	__le16 send_queue_head;
+	u8     sq_suspended;
+	u8     ddb_type;	  /* DDB_TYPE_TARGET */
+
+	__le16 _r_a;
+
+	__be16 awt_def;
+	u8     compat_features;	  /* byte 20 */
+	u8     pathway_blocked_count;
+	__be16 arb_wait_time;
+	__be32 more_compat_features; /* byte 24 */
+
+	u8     conn_mask;
+	u8     flags;	  /* concurrent conn:2,2 and open:0(1) */
+#define SATA_MULTIPORT     0x80
+#define SUPPORTS_AFFIL     0x40
+#define STP_AFFIL_POL      0x20
+
+	u8     _r_b;
+	u8     flags2;		  /* STP close policy:0 */
+#define STP_CL_POL_NO_TX    0x00
+#define STP_CL_POL_BTW_CMDS 0x01
+
+	__le16 exec_queue_tail;
+	__le16 send_queue_tail;
+	__le16 sister_ddb;
+	__le16 ata_cmd_scbptr;
+	__le32 sata_tag_alloc_mask;
+	__le16 active_task_count;
+	__le16 _r_c;
+	__le32 sata_sactive;
+	u8     num_sata_tags;
+	u8     sata_status;
+	u8     sata_ending_status;
+	u8     itnl_reason;	  /* I_T nexus loss reason */
+	__le16 ncq_data_scb_ptr;
+	__le16 itnl_timeout;
+	__le32 itnl_timestamp;
+} __attribute__ ((packed));
+
+/* This struct asd_ddb_init_port, describes the device descriptor block
+ * of an initiator port (when the sequencer is operating in target mode).
+ * Bytes [0,11] and [20,27] are from the OPEN address frame.
+ * The sequencer allocates an initiator port DDB entry.
+ */
+struct asd_ddb_init_port {
+	u8     conn_type;	  /* byte 0 */
+	u8     conn_rate;
+	__be16 init_conn_tag;     /* BE */
+	u8     dest_sas_addr[8];
+	__le16 send_queue_head;   /* LE, byte 12 */
+	u8     sq_suspended;
+	u8     ddb_type;	  /* DDB_TYPE_INITIATOR */
+	__le16 _r_a;
+	__be16 awt_def;		  /* BE */
+	u8     compat_features;
+	u8     pathway_blocked_count;
+	__be16 arb_wait_time;	  /* BE */
+	__be32 more_compat_features; /* BE */
+	u8     conn_mask;
+	u8     flags;		  /* == 5 */
+	u16    _r_b;
+	__le16 exec_queue_tail;	  /* execution queue tail */
+	__le16 send_queue_tail;
+	__le16 sister_ddb;
+	__le16 init_resp_timeout; /* initiator response timeout */
+	__le32 _r_c;
+	__le16 active_tasks;	  /* active task count */
+	__le16 init_list;	  /* initiator list link pointer */
+	__le32 _r_d;
+	u8     max_conn_to[3]; /* from Conn-Disc mode page, in us, LE */
+	u8     itnl_reason;	  /* I_T nexus loss reason */
+	__le16 bus_inact_to; /* from Conn-Disc mode page, in 100 us, LE */
+	__le16 itnl_to;		  /* from the Protocol Specific Port Ctrl MP */
+	__le32 itnl_timestamp;
+} __attribute__ ((packed));
+
+/* This struct asd_ddb_sata_tag, describes a look-up table to be used
+ * by the sequencers.  SATA II, IDENTIFY DEVICE data, word 76, bit 8:
+ * NCQ support.  This table is used by the sequencers to find the
+ * corresponding SCB, given a SATA II tag value.
+ */
+struct asd_ddb_sata_tag {
+	__le16 scb_pointer[32];
+} __attribute__ ((packed));
+
+/* This struct asd_ddb_sata_pm_table, describes a port number to
+ * connection handle look-up table.  SATA targets attached to a port
+ * multiplier require a 4-bit port number value.  There is one DDB
+ * entry of this type for each SATA port multiplier (sister DDB).
+ * Given a SATA PM port number, this table gives us the SATA PM Port
+ * DDB of the SATA port multiplier port (i.e. the SATA target
+ * discovered on the port).
+ */
+struct asd_ddb_sata_pm_table {
+	__le16 ddb_pointer[16];
+	__le16 _r_a[16];
+} __attribute__ ((packed));
+
+/* This struct asd_ddb_sata_pm_port, describes the SATA port multiplier
+ * port format DDB.
+ */
+struct asd_ddb_sata_pm_port {
+	u8     _r_a[15];
+	u8     ddb_type;
+	u8     _r_b[13];
+	u8     pm_port_flags;
+#define PM_PORT_MASK  0xF0
+#define PM_PORT_SET   0x02
+	u8     _r_c[6];
+	__le16 sister_ddb;
+	__le16 ata_cmd_scbptr;
+	__le32 sata_tag_alloc_mask;
+	__le16 active_task_count;
+	__le16 parent_ddb;
+	__le32 sata_sactive;
+	u8     num_sata_tags;
+	u8     sata_status;
+	u8     sata_ending_status;
+	u8     _r_d[9];
+} __attribute__ ((packed));
+
+/* This struct asd_ddb_seq_shared, describes a DDB shared by the
+ * central and link sequencers.  port_map_by_links is indexed phy
+ * number [0,7]; each byte is a bit mask of all the phys that are in
+ * the same port as the indexed phy.
+ */
+struct asd_ddb_seq_shared {
+	__le16 q_free_ddb_head;
+	__le16 q_free_ddb_tail;
+	__le16 q_free_ddb_cnt;
+	__le16 q_used_ddb_head;
+	__le16 q_used_ddb_tail;
+	__le16 shared_mem_lock;
+	__le16 smp_conn_tag;
+	__le16 est_nexus_buf_cnt;
+	__le16 est_nexus_buf_thresh;
+	u32    _r_a;
+	u8     settable_max_contexts;
+	u8     _r_b[23];
+	u8     conn_not_active;
+	u8     phy_is_up;
+	u8     _r_c[8];
+	u8     port_map_by_links[8];
+} __attribute__ ((packed));
+
+/* ---------- SG Element ---------- */
+
+/* This struct sg_el, describes the hardware scatter gather buffer
+ * element.  All entries are little endian.  In an SCB, there are 2 of
+ * this, plus one more, called a link element of this indicating a
+ * sublist if needed.
+ *
+ * A link element has only the bus address set and the flags (DS) bit
+ * valid.  The bus address points to the start of the sublist.
+ *
+ * If a sublist is needed, then that sublist should also include the 2
+ * sg_el embedded in the SCB, in which case next_sg_offset is 32,
+ * since sizeof(sg_el) = 16; EOS should be 1 and EOL 0 in this case.
+ */
+struct sg_el {
+	__le64 bus_addr;
+	__le32 size;
+	__le16 _r;
+	u8     next_sg_offs;
+	u8     flags;
+#define ASD_SG_EL_DS_MASK   0x30
+#define ASD_SG_EL_DS_OCM    0x10
+#define ASD_SG_EL_DS_HM     0x00
+#define ASD_SG_EL_LIST_MASK 0xC0
+#define ASD_SG_EL_LIST_EOL  0x40
+#define ASD_SG_EL_LIST_EOS  0x80
+} __attribute__ ((packed));
+
+/* ---------- SCBs ---------- */
+
+/* An SCB (sequencer control block) is comprised of a common header
+ * and a task part, for a total of 128 bytes.  All fields are in LE
+ * order, unless otherwise noted.
+ */
+
+/* This struct scb_header, defines the SCB header format.
+ */
+struct scb_header {
+	__le64 next_scb;
+	__le16 index;		  /* transaction context */
+	u8     opcode;
+} __attribute__ ((packed));
+
+/* SCB opcodes: Execution queue
+ */
+#define INITIATE_SSP_TASK       0x00
+#define INITIATE_LONG_SSP_TASK  0x01
+#define INITIATE_BIDIR_SSP_TASK 0x02
+#define ABORT_TASK              0x03
+#define INITIATE_SSP_TMF        0x04
+#define SSP_TARG_GET_DATA       0x05
+#define SSP_TARG_GET_DATA_GOOD  0x06
+#define SSP_TARG_SEND_RESP      0x07
+#define QUERY_SSP_TASK          0x08
+#define INITIATE_ATA_TASK       0x09
+#define INITIATE_ATAPI_TASK     0x0a
+#define CONTROL_ATA_DEV         0x0b
+#define INITIATE_SMP_TASK       0x0c
+#define SMP_TARG_SEND_RESP      0x0f
+
+/* SCB opcodes: Send Queue
+ */
+#define SSP_TARG_SEND_DATA      0x40
+#define SSP_TARG_SEND_DATA_GOOD 0x41
+
+/* SCB opcodes: Link Queue
+ */
+#define CONTROL_PHY             0x80
+#define SEND_PRIMITIVE          0x81
+#define INITIATE_LINK_ADM_TASK  0x82
+
+/* SCB opcodes: other
+ */
+#define EMPTY_SCB               0xc0
+#define INITIATE_SEQ_ADM_TASK   0xc1
+#define EST_ICL_TARG_WINDOW     0xc2
+#define COPY_MEM                0xc3
+#define CLEAR_NEXUS             0xc4
+#define INITIATE_DDB_ADM_TASK   0xc6
+#define ESTABLISH_NEXUS_ESCB    0xd0
+
+#define LUN_SIZE                8
+
+/* See SAS spec, task IU
+ */
+struct ssp_task_iu {
+	u8     lun[LUN_SIZE];	  /* BE */
+	u16    _r_a;
+	u8     tmf;
+	u8     _r_b;
+	__be16 tag;		  /* BE */
+	u8     _r_c[14];
+} __attribute__ ((packed));
+
+/* See SAS spec, command IU
+ */
+struct ssp_command_iu {
+	u8     lun[LUN_SIZE];
+	u8     _r_a;
+	u8     efb_prio_attr;	  /* enable first burst, task prio & attr */
+#define EFB_MASK        0x80
+#define TASK_PRIO_MASK	0x78
+#define TASK_ATTR_MASK  0x07
+
+	u8    _r_b;
+	u8     add_cdb_len;	  /* in dwords, since bit 0,1 are reserved */
+	union {
+		u8     cdb[16];
+		struct {
+			__le64 long_cdb_addr;	  /* bus address, LE */
+			__le32 long_cdb_size;	  /* LE */
+			u8     _r_c[3];
+			u8     eol_ds;		  /* eol:6,6, ds:5,4 */
+		} long_cdb;	  /* sequencer extension */
+	};
+} __attribute__ ((packed));
+
+struct xfer_rdy_iu {
+	__be32 requested_offset;  /* BE */
+	__be32 write_data_len;	  /* BE */
+	__be32 _r_a;
+} __attribute__ ((packed));
+
+/* ---------- SCB tasks ---------- */
+
+/* This is both ssp_task and long_ssp_task
+ */
+struct initiate_ssp_task {
+	u8     proto_conn_rate;	  /* proto:6,4, conn_rate:3,0 */
+	__le32 total_xfer_len;
+	struct ssp_frame_hdr  ssp_frame;
+	struct ssp_command_iu ssp_cmd;
+	__le16 sister_scb;	  /* 0xFFFF */
+	__le16 conn_handle;	  /* index to DDB for the intended target */
+	u8     data_dir;	  /* :1,0 */
+#define DATA_DIR_NONE   0x00
+#define DATA_DIR_IN     0x01
+#define DATA_DIR_OUT    0x02
+#define DATA_DIR_BYRECIPIENT 0x03
+
+	u8     _r_a;
+	u8     retry_count;
+	u8     _r_b[5];
+	struct sg_el sg_element[3]; /* 2 real and 1 link */
+} __attribute__ ((packed));
+
+/* This defines both ata_task and atapi_task.
+ * ata: C bit of FIS should be 1,
+ * atapi: C bit of FIS should be 1, and command register should be 0xA0,
+ * to indicate a packet command.
+ */
+struct initiate_ata_task {
+	u8     proto_conn_rate;
+	__le32 total_xfer_len;
+	struct host_to_dev_fis fis;
+	__le32 data_offs;
+	u8     atapi_packet[16];
+	u8     _r_a[12];
+	__le16 sister_scb;
+	__le16 conn_handle;
+	u8     ata_flags;	  /* CSMI:6,6, DTM:4,4, QT:3,3, data dir:1,0 */
+#define CSMI_TASK           0x40
+#define DATA_XFER_MODE_DMA  0x10
+#define ATA_Q_TYPE_MASK     0x08
+#define	ATA_Q_TYPE_UNTAGGED 0x00
+#define ATA_Q_TYPE_NCQ      0x08
+
+	u8     _r_b;
+	u8     retry_count;
+	u8     _r_c;
+	u8     flags;
+#define STP_AFFIL_POLICY   0x20
+#define SET_AFFIL_POLICY   0x10
+#define RET_PARTIAL_SGLIST 0x02
+
+	u8     _r_d[3];
+	struct sg_el sg_element[3];
+} __attribute__ ((packed));
+
+struct initiate_smp_task {
+	u8     proto_conn_rate;
+	u8     _r_a[40];
+	struct sg_el smp_req;
+	__le16 sister_scb;
+	__le16 conn_handle;
+	u8     _r_c[8];
+	struct sg_el smp_resp;
+	u8     _r_d[32];
+} __attribute__ ((packed));
+
+struct control_phy {
+	u8     phy_id;
+	u8     sub_func;
+#define DISABLE_PHY            0x00
+#define ENABLE_PHY             0x01
+#define RELEASE_SPINUP_HOLD    0x02
+#define ENABLE_PHY_NO_SAS_OOB  0x03
+#define ENABLE_PHY_NO_SATA_OOB 0x04
+#define PHY_NO_OP              0x05
+#define EXECUTE_HARD_RESET     0x81
+
+	u8     func_mask;
+	u8     speed_mask;
+	u8     hot_plug_delay;
+	u8     port_type;
+	u8     flags;
+#define DEV_PRES_TIMER_OVERRIDE_ENABLE 0x01
+#define DISABLE_PHY_IF_OOB_FAILS       0x02
+
+	__le32 timeout_override;
+	u8     link_reset_retries;
+	u8     _r_a[47];
+	__le16 conn_handle;
+	u8     _r_b[56];
+} __attribute__ ((packed));
+
+struct control_ata_dev {
+	u8     proto_conn_rate;
+	__le32 _r_a;
+	struct host_to_dev_fis fis;
+	u8     _r_b[32];
+	__le16 sister_scb;
+	__le16 conn_handle;
+	u8     ata_flags;	  /* 0 */
+	u8     _r_c[55];
+} __attribute__ ((packed));
+
+struct empty_scb {
+	u8     num_valid;
+	__le32 _r_a;
+#define ASD_EDBS_PER_SCB 7
+/* header+data+CRC+DMA suffix data */
+#define ASD_EDB_SIZE (24+1024+4+16)
+	struct sg_el eb[ASD_EDBS_PER_SCB];
+#define ELEMENT_NOT_VALID  0xC0
+} __attribute__ ((packed));
+
+struct initiate_link_adm {
+	u8     phy_id;
+	u8     sub_func;
+#define GET_LINK_ERROR_COUNT      0x00
+#define RESET_LINK_ERROR_COUNT    0x01
+#define ENABLE_NOTIFY_SPINUP_INTS 0x02
+
+	u8     _r_a[57];
+	__le16 conn_handle;
+	u8     _r_b[56];
+} __attribute__ ((packed));
+
+struct copy_memory {
+	u8     _r_a;
+	__le16 xfer_len;
+	__le16 _r_b;
+	__le64 src_busaddr;
+	u8     src_ds;		  /* See definition of sg_el */
+	u8     _r_c[45];
+	__le16 conn_handle;
+	__le64 _r_d;
+	__le64 dest_busaddr;
+	u8     dest_ds;		  /* See definition of sg_el */
+	u8     _r_e[39];
+} __attribute__ ((packed));
+
+struct abort_task {
+	u8     proto_conn_rate;
+	__le32 _r_a;
+	struct ssp_frame_hdr ssp_frame;
+	struct ssp_task_iu ssp_task;
+	__le16 sister_scb;
+	__le16 conn_handle;
+	u8     flags;	  /* ovrd_itnl_timer:3,3, suspend_data_trans:2,2 */
+#define SUSPEND_DATA_TRANS 0x04
+
+	u8     _r_b;
+	u8     retry_count;
+	u8     _r_c[5];
+	__le16 index;  /* Transaction context of task to be queried */
+	__le16 itnl_to;
+	u8     _r_d[44];
+} __attribute__ ((packed));
+
+struct clear_nexus {
+	u8     nexus;
+#define NEXUS_ADAPTER  0x00
+#define NEXUS_PORT     0x01
+#define NEXUS_I_T      0x02
+#define NEXUS_I_T_L    0x03
+#define NEXUS_TAG      0x04
+#define NEXUS_TRANS_CX 0x05
+#define NEXUS_SATA_TAG 0x06
+#define NEXUS_T_L      0x07
+#define NEXUS_L        0x08
+#define NEXUS_T_TAG    0x09
+
+	__le32 _r_a;
+	u8     flags;
+#define SUSPEND_TX     0x80
+#define RESUME_TX      0x40
+#define SEND_Q         0x04
+#define EXEC_Q         0x02
+#define NOTINQ         0x01
+
+	u8     _r_b[3];
+	u8     conn_mask;
+	u8     _r_c[19];
+	struct ssp_task_iu ssp_task; /* LUN and TAG */
+	__le16 _r_d;
+	__le16 conn_handle;
+	__le64 _r_e;
+	__le16 index;  /* Transaction context of task to be cleared */
+	__le16 context;		  /* Clear nexus context */
+	u8     _r_f[44];
+} __attribute__ ((packed));
+
+struct initiate_ssp_tmf {
+	u8     proto_conn_rate;
+	__le32 _r_a;
+	struct ssp_frame_hdr ssp_frame;
+	struct ssp_task_iu ssp_task;
+	__le16 sister_scb;
+	__le16 conn_handle;
+	u8     flags;	  /* itnl override and suspend data tx */
+#define OVERRIDE_ITNL_TIMER  8
+
+	u8     _r_b;
+	u8     retry_count;
+	u8     _r_c[5];
+	__le16 index;  /* Transaction context of task to be queried */
+	__le16 itnl_to;
+	u8     _r_d[44];
+} __attribute__ ((packed));
+
+/* Transmits an arbitrary primitive on the link.
+ * Used for NOTIFY and BROADCAST.
+ */
+struct send_prim {
+	u8     phy_id;
+	u8     wait_transmit; 	  /* :0,0 */
+	u8     xmit_flags;
+#define XMTPSIZE_MASK      0xF0
+#define XMTPSIZE_SINGLE    0x10
+#define XMTPSIZE_REPEATED  0x20
+#define XMTPSIZE_CONT      0x20
+#define XMTPSIZE_TRIPLE    0x30
+#define XMTPSIZE_REDUNDANT 0x60
+#define XMTPSIZE_INF       0
+
+#define XMTCONTEN          0x04
+#define XMTPFRM            0x02	  /* Transmit at the next frame boundary */
+#define XMTPIMM            0x01	  /* Transmit immediately */
+
+	__le16 _r_a;
+	u8     prim[4];		  /* K, D0, D1, D2 */
+	u8     _r_b[50];
+	__le16 conn_handle;
+	u8     _r_c[56];
+} __attribute__ ((packed));
+
+/* This describes both SSP Target Get Data and SSP Target Get Data And
+ * Send Good Response SCBs.  Used when the sequencer is operating in
+ * target mode...
+ */
+struct ssp_targ_get_data {
+	u8     proto_conn_rate;
+	__le32 total_xfer_len;
+	struct ssp_frame_hdr ssp_frame;
+	struct xfer_rdy_iu  xfer_rdy;
+	u8     lun[LUN_SIZE];
+	__le64 _r_a;
+	__le16 sister_scb;
+	__le16 conn_handle;
+	u8     data_dir;	  /* 01b */
+	u8     _r_b;
+	u8     retry_count;
+	u8     _r_c[5];
+	struct sg_el sg_element[3];
+} __attribute__ ((packed));
+
+/* ---------- The actual SCB struct ---------- */
+
+struct scb {
+	struct scb_header header;
+	union {
+		struct initiate_ssp_task ssp_task;
+		struct initiate_ata_task ata_task;
+		struct initiate_smp_task smp_task;
+		struct control_phy       control_phy;
+		struct control_ata_dev   control_ata_dev;
+		struct empty_scb         escb;
+		struct initiate_link_adm link_adm;
+		struct copy_memory       cp_mem;
+		struct abort_task        abort_task;
+		struct clear_nexus       clear_nexus;
+		struct initiate_ssp_tmf  ssp_tmf;
+	};
+} __attribute__ ((packed));
+
+/* ---------- Done List ---------- */
+/* The done list entry opcode field is defined below.
+ * The mnemonic encoding and meaning is as follows:
+ * TC - Task Complete, status was received and acknowledged
+ * TF - Task Failed, indicates an error prior to receiving acknowledgment
+ *   for the command:
+ *   - no conn,
+ *   - NACK or R_ERR received in response to this command,
+ *   - credit blocked or not available, or in the case of SMP request,
+ *   - no SMP response was received.
+ *   In these four cases it is known that the target didn't receive the
+ *   command.
+ * TI - Task Interrupted, error after the command was acknowledged.  It is
+ *   known that the command was received by the target.
+ * TU - Task Unacked, command was transmitted but neither ACK (R_OK) nor NAK
+ *   (R_ERR) was received due to loss of signal, broken connection, loss of
+ *   dword sync or other reason.  The application client should send the
+ *   appropriate task query.
+ * TA - Task Aborted, see TF.
+ * _RESP - The completion includes an empty buffer containing status.
+ * TO - Timeout.
+ */
+#define TC_NO_ERROR             0x00
+#define TC_UNDERRUN             0x01
+#define TC_OVERRUN              0x02
+#define TF_OPEN_TO              0x03
+#define TF_OPEN_REJECT          0x04
+#define TI_BREAK                0x05
+#define TI_PROTO_ERR            0x06
+#define TC_SSP_RESP             0x07
+#define TI_PHY_DOWN             0x08
+#define TF_PHY_DOWN             0x09
+#define TC_LINK_ADM_RESP        0x0a
+#define TC_CSMI                 0x0b
+#define TC_ATA_RESP             0x0c
+#define TU_PHY_DOWN             0x0d
+#define TU_BREAK                0x0e
+#define TI_SATA_TO              0x0f
+#define TI_NAK                  0x10
+#define TC_CONTROL_PHY          0x11
+#define TF_BREAK                0x12
+#define TC_RESUME               0x13
+#define TI_ACK_NAK_TO           0x14
+#define TF_SMPRSP_TO            0x15
+#define TF_SMP_XMIT_RCV_ERR     0x16
+#define TC_PARTIAL_SG_LIST      0x17
+#define TU_ACK_NAK_TO           0x18
+#define TU_SATA_TO              0x19
+#define TF_NAK_RECV             0x1a
+#define TA_I_T_NEXUS_LOSS       0x1b
+#define TC_ATA_R_ERR_RECV       0x1c
+#define TF_TMF_NO_CTX           0x1d
+#define TA_ON_REQ               0x1e
+#define TF_TMF_NO_TAG           0x1f
+#define TF_TMF_TAG_FREE         0x20
+#define TF_TMF_TASK_DONE        0x21
+#define TF_TMF_NO_CONN_HANDLE   0x22
+#define TC_TASK_CLEARED         0x23
+#define TI_SYNCS_RECV           0x24
+#define TU_SYNCS_RECV           0x25
+#define TF_IRTT_TO              0x26
+#define TF_NO_SMP_CONN          0x27
+#define TF_IU_SHORT             0x28
+#define TF_DATA_OFFS_ERR        0x29
+#define TF_INV_CONN_HANDLE      0x2a
+#define TF_REQUESTED_N_PENDING  0x2b
+
+/* 0xc1 - 0xc7: empty buffer received,
+   0xd1 - 0xd7: establish nexus empty buffer received
+*/
+/* This is the ESCB mask */
+#define ESCB_RECVD              0xC0
+
+
+/* This struct done_list_struct defines the done list entry.
+ * All fields are LE.
+ */
+struct done_list_struct {
+	__le16 index;		  /* aka transaction context */
+	u8     opcode;
+	u8     status_block[4];
+	u8     toggle;		  /* bit 0 */
+#define DL_TOGGLE_MASK     0x01
+} __attribute__ ((packed));
+
+/* ---------- PHYS ---------- */
+
+struct asd_phy {
+	struct asd_sas_phy        sas_phy;
+	struct asd_phy_desc   *phy_desc; /* hw profile */
+
+	struct sas_identify_frame *identify_frame;
+	struct asd_dma_tok  *id_frm_tok;
+
+	u8         frame_rcvd[ASD_EDB_SIZE];
+};
+
+
+#define ASD_SCB_SIZE sizeof(struct scb)
+#define ASD_DDB_SIZE sizeof(struct asd_ddb_ssp_smp_target_port)
+
+/* Define this to 0 if you do not want NOTIFY (ENABLE SPINIP) sent.
+ * Default: 0x10 (it's a mask)
+ */
+#define ASD_NOTIFY_ENABLE_SPINUP  0x10
+
+/* If enabled, set this to the interval between transmission
+ * of NOTIFY (ENABLE SPINUP). In units of 200 us.
+ */
+#define ASD_NOTIFY_TIMEOUT        2500
+
+/* Initial delay after OOB, before we transmit NOTIFY (ENABLE SPINUP).
+ * If 0, transmit immediately. In milliseconds.
+ */
+#define ASD_NOTIFY_DOWN_COUNT     0
+
+/* Device present timer timeout constant, 10 ms. */
+#define ASD_DEV_PRESENT_TIMEOUT   0x2710
+
+#define ASD_SATA_INTERLOCK_TIMEOUT 0
+
+/* How long to wait before shutting down an STP connection, unless
+ * an STP target sent frame(s). 50 usec.
+ * IGNORED by the sequencer (i.e. value 0 always).
+ */
+#define ASD_STP_SHUTDOWN_TIMEOUT  0x0
+
+/* ATA soft reset timer timeout. 5 usec. */
+#define ASD_SRST_ASSERT_TIMEOUT   0x05
+
+/* 31 sec */
+#define ASD_RCV_FIS_TIMEOUT       0x01D905C0
+
+#define ASD_ONE_MILLISEC_TIMEOUT  0x03e8
+
+/* COMINIT timer */
+#define ASD_TEN_MILLISEC_TIMEOUT  0x2710
+#define ASD_COMINIT_TIMEOUT ASD_TEN_MILLISEC_TIMEOUT
+
+/* 1 sec */
+#define ASD_SMP_RCV_TIMEOUT       0x000F4240
+
+#endif
diff --git a/drivers/scsi/aic94xx/aic94xx_scb.c b/drivers/scsi/aic94xx/aic94xx_scb.c
new file mode 100644
index 000000000000..fc1b7438a913
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_scb.c
@@ -0,0 +1,732 @@
+/*
+ * Aic94xx SAS/SATA driver SCB management.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/pci.h>
+
+#include "aic94xx.h"
+#include "aic94xx_reg.h"
+#include "aic94xx_hwi.h"
+#include "aic94xx_seq.h"
+
+#include "aic94xx_dump.h"
+
+/* ---------- EMPTY SCB ---------- */
+
+#define DL_PHY_MASK      7
+#define BYTES_DMAED      0
+#define PRIMITIVE_RECVD  0x08
+#define PHY_EVENT        0x10
+#define LINK_RESET_ERROR 0x18
+#define TIMER_EVENT      0x20
+#define REQ_TASK_ABORT   0xF0
+#define REQ_DEVICE_RESET 0xF1
+#define SIGNAL_NCQ_ERROR 0xF2
+#define CLEAR_NCQ_ERROR  0xF3
+
+#define PHY_EVENTS_STATUS (CURRENT_LOSS_OF_SIGNAL | CURRENT_OOB_DONE   \
+			   | CURRENT_SPINUP_HOLD | CURRENT_GTO_TIMEOUT \
+			   | CURRENT_OOB_ERROR)
+
+static inline void get_lrate_mode(struct asd_phy *phy, u8 oob_mode)
+{
+	switch (oob_mode & 7) {
+	case PHY_SPEED_60:
+		/* FIXME: sas transport class doesn't have this */
+		phy->sas_phy.linkrate = PHY_LINKRATE_6;
+		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_6_0_GBPS;
+		break;
+	case PHY_SPEED_30:
+		phy->sas_phy.linkrate = PHY_LINKRATE_3;
+		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_3_0_GBPS;
+		break;
+	case PHY_SPEED_15:
+		phy->sas_phy.linkrate = PHY_LINKRATE_1_5;
+		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_1_5_GBPS;
+		break;
+	}
+	if (oob_mode & SAS_MODE)
+		phy->sas_phy.oob_mode = SAS_OOB_MODE;
+	else if (oob_mode & SATA_MODE)
+		phy->sas_phy.oob_mode = SATA_OOB_MODE;
+}
+
+static inline void asd_phy_event_tasklet(struct asd_ascb *ascb,
+					 struct done_list_struct *dl)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct sas_ha_struct *sas_ha = &asd_ha->sas_ha;
+	int phy_id = dl->status_block[0] & DL_PHY_MASK;
+	struct asd_phy *phy = &asd_ha->phys[phy_id];
+
+	u8 oob_status = dl->status_block[1] & PHY_EVENTS_STATUS;
+	u8 oob_mode   = dl->status_block[2];
+
+	switch (oob_status) {
+	case CURRENT_LOSS_OF_SIGNAL:
+		/* directly attached device was removed */
+		ASD_DPRINTK("phy%d: device unplugged\n", phy_id);
+		asd_turn_led(asd_ha, phy_id, 0);
+		sas_phy_disconnected(&phy->sas_phy);
+		sas_ha->notify_phy_event(&phy->sas_phy, PHYE_LOSS_OF_SIGNAL);
+		break;
+	case CURRENT_OOB_DONE:
+		/* hot plugged device */
+		asd_turn_led(asd_ha, phy_id, 1);
+		get_lrate_mode(phy, oob_mode);
+		ASD_DPRINTK("phy%d device plugged: lrate:0x%x, proto:0x%x\n",
+			    phy_id, phy->sas_phy.linkrate, phy->sas_phy.iproto);
+		sas_ha->notify_phy_event(&phy->sas_phy, PHYE_OOB_DONE);
+		break;
+	case CURRENT_SPINUP_HOLD:
+		/* hot plug SATA, no COMWAKE sent */
+		asd_turn_led(asd_ha, phy_id, 1);
+		sas_ha->notify_phy_event(&phy->sas_phy, PHYE_SPINUP_HOLD);
+		break;
+	case CURRENT_GTO_TIMEOUT:
+	case CURRENT_OOB_ERROR:
+		ASD_DPRINTK("phy%d error while OOB: oob status:0x%x\n", phy_id,
+			    dl->status_block[1]);
+		asd_turn_led(asd_ha, phy_id, 0);
+		sas_phy_disconnected(&phy->sas_phy);
+		sas_ha->notify_phy_event(&phy->sas_phy, PHYE_OOB_ERROR);
+		break;
+	}
+}
+
+/* If phys are enabled sparsely, this will do the right thing. */
+static inline unsigned ord_phy(struct asd_ha_struct *asd_ha,
+			       struct asd_phy *phy)
+{
+	u8 enabled_mask = asd_ha->hw_prof.enabled_phys;
+	int i, k = 0;
+
+	for_each_phy(enabled_mask, enabled_mask, i) {
+		if (&asd_ha->phys[i] == phy)
+			return k;
+		k++;
+	}
+	return 0;
+}
+
+/**
+ * asd_get_attached_sas_addr -- extract/generate attached SAS address
+ * phy: pointer to asd_phy
+ * sas_addr: pointer to buffer where the SAS address is to be written
+ *
+ * This function extracts the SAS address from an IDENTIFY frame
+ * received.  If OOB is SATA, then a SAS address is generated from the
+ * HA tables.
+ *
+ * LOCKING: the frame_rcvd_lock needs to be held since this parses the frame
+ * buffer.
+ */
+static inline void asd_get_attached_sas_addr(struct asd_phy *phy, u8 *sas_addr)
+{
+	if (phy->sas_phy.frame_rcvd[0] == 0x34
+	    && phy->sas_phy.oob_mode == SATA_OOB_MODE) {
+		struct asd_ha_struct *asd_ha = phy->sas_phy.ha->lldd_ha;
+		/* FIS device-to-host */
+		u64 addr = be64_to_cpu(*(__be64 *)phy->phy_desc->sas_addr);
+
+		addr += asd_ha->hw_prof.sata_name_base + ord_phy(asd_ha, phy);
+		*(__be64 *)sas_addr = cpu_to_be64(addr);
+	} else {
+		struct sas_identify_frame *idframe =
+			(void *) phy->sas_phy.frame_rcvd;
+		memcpy(sas_addr, idframe->sas_addr, SAS_ADDR_SIZE);
+	}
+}
+
+static inline void asd_bytes_dmaed_tasklet(struct asd_ascb *ascb,
+					   struct done_list_struct *dl,
+					   int edb_id, int phy_id)
+{
+	unsigned long flags;
+	int edb_el = edb_id + ascb->edb_index;
+	struct asd_dma_tok *edb = ascb->ha->seq.edb_arr[edb_el];
+	struct asd_phy *phy = &ascb->ha->phys[phy_id];
+	struct sas_ha_struct *sas_ha = phy->sas_phy.ha;
+	u16 size = ((dl->status_block[3] & 7) << 8) | dl->status_block[2];
+
+	size = min(size, (u16) sizeof(phy->frame_rcvd));
+
+	spin_lock_irqsave(&phy->sas_phy.frame_rcvd_lock, flags);
+	memcpy(phy->sas_phy.frame_rcvd, edb->vaddr, size);
+	phy->sas_phy.frame_rcvd_size = size;
+	asd_get_attached_sas_addr(phy, phy->sas_phy.attached_sas_addr);
+	spin_unlock_irqrestore(&phy->sas_phy.frame_rcvd_lock, flags);
+	asd_dump_frame_rcvd(phy, dl);
+	sas_ha->notify_port_event(&phy->sas_phy, PORTE_BYTES_DMAED);
+}
+
+static inline void asd_link_reset_err_tasklet(struct asd_ascb *ascb,
+					      struct done_list_struct *dl,
+					      int phy_id)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct sas_ha_struct *sas_ha = &asd_ha->sas_ha;
+	struct asd_sas_phy *sas_phy = sas_ha->sas_phy[phy_id];
+	u8 lr_error = dl->status_block[1];
+	u8 retries_left = dl->status_block[2];
+
+	switch (lr_error) {
+	case 0:
+		ASD_DPRINTK("phy%d: Receive ID timer expired\n", phy_id);
+		break;
+	case 1:
+		ASD_DPRINTK("phy%d: Loss of signal\n", phy_id);
+		break;
+	case 2:
+		ASD_DPRINTK("phy%d: Loss of dword sync\n", phy_id);
+		break;
+	case 3:
+		ASD_DPRINTK("phy%d: Receive FIS timeout\n", phy_id);
+		break;
+	default:
+		ASD_DPRINTK("phy%d: unknown link reset error code: 0x%x\n",
+			    phy_id, lr_error);
+		break;
+	}
+
+	asd_turn_led(asd_ha, phy_id, 0);
+	sas_phy_disconnected(sas_phy);
+	sas_ha->notify_port_event(sas_phy, PORTE_LINK_RESET_ERR);
+
+	if (retries_left == 0) {
+		int num = 1;
+		struct asd_ascb *cp = asd_ascb_alloc_list(ascb->ha, &num,
+							  GFP_ATOMIC);
+		if (!cp) {
+			asd_printk("%s: out of memory\n", __FUNCTION__);
+			goto out;
+		}
+		ASD_DPRINTK("phy%d: retries:0 performing link reset seq\n",
+			    phy_id);
+		asd_build_control_phy(cp, phy_id, ENABLE_PHY);
+		if (asd_post_ascb_list(ascb->ha, cp, 1) != 0)
+			asd_ascb_free(cp);
+	}
+out:
+	;
+}
+
+static inline void asd_primitive_rcvd_tasklet(struct asd_ascb *ascb,
+					      struct done_list_struct *dl,
+					      int phy_id)
+{
+	unsigned long flags;
+	struct sas_ha_struct *sas_ha = &ascb->ha->sas_ha;
+	struct asd_sas_phy *sas_phy = sas_ha->sas_phy[phy_id];
+	u8  reg  = dl->status_block[1];
+	u32 cont = dl->status_block[2] << ((reg & 3)*8);
+
+	reg &= ~3;
+	switch (reg) {
+	case LmPRMSTAT0BYTE0:
+		switch (cont) {
+		case LmBROADCH:
+		case LmBROADRVCH0:
+		case LmBROADRVCH1:
+		case LmBROADSES:
+			ASD_DPRINTK("phy%d: BROADCAST change received:%d\n",
+				    phy_id, cont);
+			spin_lock_irqsave(&sas_phy->sas_prim_lock, flags);
+			sas_phy->sas_prim = ffs(cont);
+			spin_unlock_irqrestore(&sas_phy->sas_prim_lock, flags);
+			sas_ha->notify_port_event(sas_phy,PORTE_BROADCAST_RCVD);
+			break;
+
+		case LmUNKNOWNP:
+			ASD_DPRINTK("phy%d: unknown BREAK\n", phy_id);
+			break;
+
+		default:
+			ASD_DPRINTK("phy%d: primitive reg:0x%x, cont:0x%04x\n",
+				    phy_id, reg, cont);
+			break;
+		}
+		break;
+	case LmPRMSTAT1BYTE0:
+		switch (cont) {
+		case LmHARDRST:
+			ASD_DPRINTK("phy%d: HARD_RESET primitive rcvd\n",
+				    phy_id);
+			/* The sequencer disables all phys on that port.
+			 * We have to re-enable the phys ourselves. */
+			sas_ha->notify_port_event(sas_phy, PORTE_HARD_RESET);
+			break;
+
+		default:
+			ASD_DPRINTK("phy%d: primitive reg:0x%x, cont:0x%04x\n",
+				    phy_id, reg, cont);
+			break;
+		}
+		break;
+	default:
+		ASD_DPRINTK("unknown primitive register:0x%x\n",
+			    dl->status_block[1]);
+		break;
+	}
+}
+
+/**
+ * asd_invalidate_edb -- invalidate an EDB and if necessary post the ESCB
+ * @ascb: pointer to Empty SCB
+ * @edb_id: index [0,6] to the empty data buffer which is to be invalidated
+ *
+ * After an EDB has been invalidated, if all EDBs in this ESCB have been
+ * invalidated, the ESCB is posted back to the sequencer.
+ * Context is tasklet/IRQ.
+ */
+void asd_invalidate_edb(struct asd_ascb *ascb, int edb_id)
+{
+	struct asd_seq_data *seq = &ascb->ha->seq;
+	struct empty_scb *escb = &ascb->scb->escb;
+	struct sg_el     *eb   = &escb->eb[edb_id];
+	struct asd_dma_tok *edb = seq->edb_arr[ascb->edb_index + edb_id];
+
+	memset(edb->vaddr, 0, ASD_EDB_SIZE);
+	eb->flags |= ELEMENT_NOT_VALID;
+	escb->num_valid--;
+
+	if (escb->num_valid == 0) {
+		int i;
+		/* ASD_DPRINTK("reposting escb: vaddr: 0x%p, "
+			    "dma_handle: 0x%08llx, next: 0x%08llx, "
+			    "index:%d, opcode:0x%02x\n",
+			    ascb->dma_scb.vaddr,
+			    (u64)ascb->dma_scb.dma_handle,
+			    le64_to_cpu(ascb->scb->header.next_scb),
+			    le16_to_cpu(ascb->scb->header.index),
+			    ascb->scb->header.opcode);
+		*/
+		escb->num_valid = ASD_EDBS_PER_SCB;
+		for (i = 0; i < ASD_EDBS_PER_SCB; i++)
+			escb->eb[i].flags = 0;
+		if (!list_empty(&ascb->list))
+			list_del_init(&ascb->list);
+		i = asd_post_escb_list(ascb->ha, ascb, 1);
+		if (i)
+			asd_printk("couldn't post escb, err:%d\n", i);
+	}
+}
+
+static void escb_tasklet_complete(struct asd_ascb *ascb,
+				  struct done_list_struct *dl)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct sas_ha_struct *sas_ha = &asd_ha->sas_ha;
+	int edb = (dl->opcode & DL_PHY_MASK) - 1; /* [0xc1,0xc7] -> [0,6] */
+	u8  sb_opcode = dl->status_block[0];
+	int phy_id = sb_opcode & DL_PHY_MASK;
+	struct asd_sas_phy *sas_phy = sas_ha->sas_phy[phy_id];
+
+	if (edb > 6 || edb < 0) {
+		ASD_DPRINTK("edb is 0x%x! dl->opcode is 0x%x\n",
+			    edb, dl->opcode);
+		ASD_DPRINTK("sb_opcode : 0x%x, phy_id: 0x%x\n",
+			    sb_opcode, phy_id);
+		ASD_DPRINTK("escb: vaddr: 0x%p, "
+			    "dma_handle: 0x%llx, next: 0x%llx, "
+			    "index:%d, opcode:0x%02x\n",
+			    ascb->dma_scb.vaddr,
+			    (unsigned long long)ascb->dma_scb.dma_handle,
+			    (unsigned long long)
+			    le64_to_cpu(ascb->scb->header.next_scb),
+			    le16_to_cpu(ascb->scb->header.index),
+			    ascb->scb->header.opcode);
+	}
+
+	sb_opcode &= ~DL_PHY_MASK;
+
+	switch (sb_opcode) {
+	case BYTES_DMAED:
+		ASD_DPRINTK("%s: phy%d: BYTES_DMAED\n", __FUNCTION__, phy_id);
+		asd_bytes_dmaed_tasklet(ascb, dl, edb, phy_id);
+		break;
+	case PRIMITIVE_RECVD:
+		ASD_DPRINTK("%s: phy%d: PRIMITIVE_RECVD\n", __FUNCTION__,
+			    phy_id);
+		asd_primitive_rcvd_tasklet(ascb, dl, phy_id);
+		break;
+	case PHY_EVENT:
+		ASD_DPRINTK("%s: phy%d: PHY_EVENT\n", __FUNCTION__, phy_id);
+		asd_phy_event_tasklet(ascb, dl);
+		break;
+	case LINK_RESET_ERROR:
+		ASD_DPRINTK("%s: phy%d: LINK_RESET_ERROR\n", __FUNCTION__,
+			    phy_id);
+		asd_link_reset_err_tasklet(ascb, dl, phy_id);
+		break;
+	case TIMER_EVENT:
+		ASD_DPRINTK("%s: phy%d: TIMER_EVENT, lost dw sync\n",
+			    __FUNCTION__, phy_id);
+		asd_turn_led(asd_ha, phy_id, 0);
+		/* the device is gone */
+		sas_phy_disconnected(sas_phy);
+		sas_ha->notify_port_event(sas_phy, PORTE_TIMER_EVENT);
+		break;
+	case REQ_TASK_ABORT:
+		ASD_DPRINTK("%s: phy%d: REQ_TASK_ABORT\n", __FUNCTION__,
+			    phy_id);
+		break;
+	case REQ_DEVICE_RESET:
+		ASD_DPRINTK("%s: phy%d: REQ_DEVICE_RESET\n", __FUNCTION__,
+			    phy_id);
+		break;
+	case SIGNAL_NCQ_ERROR:
+		ASD_DPRINTK("%s: phy%d: SIGNAL_NCQ_ERROR\n", __FUNCTION__,
+			    phy_id);
+		break;
+	case CLEAR_NCQ_ERROR:
+		ASD_DPRINTK("%s: phy%d: CLEAR_NCQ_ERROR\n", __FUNCTION__,
+			    phy_id);
+		break;
+	default:
+		ASD_DPRINTK("%s: phy%d: unknown event:0x%x\n", __FUNCTION__,
+			    phy_id, sb_opcode);
+		ASD_DPRINTK("edb is 0x%x! dl->opcode is 0x%x\n",
+			    edb, dl->opcode);
+		ASD_DPRINTK("sb_opcode : 0x%x, phy_id: 0x%x\n",
+			    sb_opcode, phy_id);
+		ASD_DPRINTK("escb: vaddr: 0x%p, "
+			    "dma_handle: 0x%llx, next: 0x%llx, "
+			    "index:%d, opcode:0x%02x\n",
+			    ascb->dma_scb.vaddr,
+			    (unsigned long long)ascb->dma_scb.dma_handle,
+			    (unsigned long long)
+			    le64_to_cpu(ascb->scb->header.next_scb),
+			    le16_to_cpu(ascb->scb->header.index),
+			    ascb->scb->header.opcode);
+
+		break;
+	}
+
+	asd_invalidate_edb(ascb, edb);
+}
+
+int asd_init_post_escbs(struct asd_ha_struct *asd_ha)
+{
+	struct asd_seq_data *seq = &asd_ha->seq;
+	int i;
+
+	for (i = 0; i < seq->num_escbs; i++)
+		seq->escb_arr[i]->tasklet_complete = escb_tasklet_complete;
+
+	ASD_DPRINTK("posting %d escbs\n", i);
+	return asd_post_escb_list(asd_ha, seq->escb_arr[0], seq->num_escbs);
+}
+
+/* ---------- CONTROL PHY ---------- */
+
+#define CONTROL_PHY_STATUS (CURRENT_DEVICE_PRESENT | CURRENT_OOB_DONE   \
+			    | CURRENT_SPINUP_HOLD | CURRENT_GTO_TIMEOUT \
+			    | CURRENT_OOB_ERROR)
+
+/**
+ * control_phy_tasklet_complete -- tasklet complete for CONTROL PHY ascb
+ * @ascb: pointer to an ascb
+ * @dl: pointer to the done list entry
+ *
+ * This function completes a CONTROL PHY scb and frees the ascb.
+ * A note on LEDs:
+ *  - an LED blinks if there is IO though it,
+ *  - if a device is connected to the LED, it is lit,
+ *  - if no device is connected to the LED, is is dimmed (off).
+ */
+static void control_phy_tasklet_complete(struct asd_ascb *ascb,
+					 struct done_list_struct *dl)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct scb *scb = ascb->scb;
+	struct control_phy *control_phy = &scb->control_phy;
+	u8 phy_id = control_phy->phy_id;
+	struct asd_phy *phy = &ascb->ha->phys[phy_id];
+
+	u8 status     = dl->status_block[0];
+	u8 oob_status = dl->status_block[1];
+	u8 oob_mode   = dl->status_block[2];
+	/* u8 oob_signals= dl->status_block[3]; */
+
+	if (status != 0) {
+		ASD_DPRINTK("%s: phy%d status block opcode:0x%x\n",
+			    __FUNCTION__, phy_id, status);
+		goto out;
+	}
+
+	switch (control_phy->sub_func) {
+	case DISABLE_PHY:
+		asd_ha->hw_prof.enabled_phys &= ~(1 << phy_id);
+		asd_turn_led(asd_ha, phy_id, 0);
+		asd_control_led(asd_ha, phy_id, 0);
+		ASD_DPRINTK("%s: disable phy%d\n", __FUNCTION__, phy_id);
+		break;
+
+	case ENABLE_PHY:
+		asd_control_led(asd_ha, phy_id, 1);
+		if (oob_status & CURRENT_OOB_DONE) {
+			asd_ha->hw_prof.enabled_phys |= (1 << phy_id);
+			get_lrate_mode(phy, oob_mode);
+			asd_turn_led(asd_ha, phy_id, 1);
+			ASD_DPRINTK("%s: phy%d, lrate:0x%x, proto:0x%x\n",
+				    __FUNCTION__, phy_id,phy->sas_phy.linkrate,
+				    phy->sas_phy.iproto);
+		} else if (oob_status & CURRENT_SPINUP_HOLD) {
+			asd_ha->hw_prof.enabled_phys |= (1 << phy_id);
+			asd_turn_led(asd_ha, phy_id, 1);
+			ASD_DPRINTK("%s: phy%d, spinup hold\n", __FUNCTION__,
+				    phy_id);
+		} else if (oob_status & CURRENT_ERR_MASK) {
+			asd_turn_led(asd_ha, phy_id, 0);
+			ASD_DPRINTK("%s: phy%d: error: oob status:0x%02x\n",
+				    __FUNCTION__, phy_id, oob_status);
+		} else if (oob_status & (CURRENT_HOT_PLUG_CNCT
+					 | CURRENT_DEVICE_PRESENT))  {
+			asd_ha->hw_prof.enabled_phys |= (1 << phy_id);
+			asd_turn_led(asd_ha, phy_id, 1);
+			ASD_DPRINTK("%s: phy%d: hot plug or device present\n",
+				    __FUNCTION__, phy_id);
+		} else {
+			asd_ha->hw_prof.enabled_phys |= (1 << phy_id);
+			asd_turn_led(asd_ha, phy_id, 0);
+			ASD_DPRINTK("%s: phy%d: no device present: "
+				    "oob_status:0x%x\n",
+				    __FUNCTION__, phy_id, oob_status);
+		}
+		break;
+	case RELEASE_SPINUP_HOLD:
+	case PHY_NO_OP:
+	case EXECUTE_HARD_RESET:
+		ASD_DPRINTK("%s: phy%d: sub_func:0x%x\n", __FUNCTION__,
+			    phy_id, control_phy->sub_func);
+		/* XXX finish */
+		break;
+	default:
+		ASD_DPRINTK("%s: phy%d: sub_func:0x%x?\n", __FUNCTION__,
+			    phy_id, control_phy->sub_func);
+		break;
+	}
+out:
+	asd_ascb_free(ascb);
+}
+
+static inline void set_speed_mask(u8 *speed_mask, struct asd_phy_desc *pd)
+{
+	/* disable all speeds, then enable defaults */
+	*speed_mask = SAS_SPEED_60_DIS | SAS_SPEED_30_DIS | SAS_SPEED_15_DIS
+		| SATA_SPEED_30_DIS | SATA_SPEED_15_DIS;
+
+	switch (pd->max_sas_lrate) {
+	case PHY_LINKRATE_6:
+		*speed_mask &= ~SAS_SPEED_60_DIS;
+	default:
+	case PHY_LINKRATE_3:
+		*speed_mask &= ~SAS_SPEED_30_DIS;
+	case PHY_LINKRATE_1_5:
+		*speed_mask &= ~SAS_SPEED_15_DIS;
+	}
+
+	switch (pd->min_sas_lrate) {
+	case PHY_LINKRATE_6:
+		*speed_mask |= SAS_SPEED_30_DIS;
+	case PHY_LINKRATE_3:
+		*speed_mask |= SAS_SPEED_15_DIS;
+	default:
+	case PHY_LINKRATE_1_5:
+		/* nothing to do */
+		;
+	}
+
+	switch (pd->max_sata_lrate) {
+	case PHY_LINKRATE_3:
+		*speed_mask &= ~SATA_SPEED_30_DIS;
+	default:
+	case PHY_LINKRATE_1_5:
+		*speed_mask &= ~SATA_SPEED_15_DIS;
+	}
+
+	switch (pd->min_sata_lrate) {
+	case PHY_LINKRATE_3:
+		*speed_mask |= SATA_SPEED_15_DIS;
+	default:
+	case PHY_LINKRATE_1_5:
+		/* nothing to do */
+		;
+	}
+}
+
+/**
+ * asd_build_control_phy -- build a CONTROL PHY SCB
+ * @ascb: pointer to an ascb
+ * @phy_id: phy id to control, integer
+ * @subfunc: subfunction, what to actually to do the phy
+ *
+ * This function builds a CONTROL PHY scb.  No allocation of any kind
+ * is performed. @ascb is allocated with the list function.
+ * The caller can override the ascb->tasklet_complete to point
+ * to its own callback function.  It must call asd_ascb_free()
+ * at its tasklet complete function.
+ * See the default implementation.
+ */
+void asd_build_control_phy(struct asd_ascb *ascb, int phy_id, u8 subfunc)
+{
+	struct asd_phy *phy = &ascb->ha->phys[phy_id];
+	struct scb *scb = ascb->scb;
+	struct control_phy *control_phy = &scb->control_phy;
+
+	scb->header.opcode = CONTROL_PHY;
+	control_phy->phy_id = (u8) phy_id;
+	control_phy->sub_func = subfunc;
+
+	switch (subfunc) {
+	case EXECUTE_HARD_RESET:  /* 0x81 */
+	case ENABLE_PHY:          /* 0x01 */
+		/* decide hot plug delay */
+		control_phy->hot_plug_delay = HOTPLUG_DELAY_TIMEOUT;
+
+		/* decide speed mask */
+		set_speed_mask(&control_phy->speed_mask, phy->phy_desc);
+
+		/* initiator port settings are in the hi nibble */
+		if (phy->sas_phy.role == PHY_ROLE_INITIATOR)
+			control_phy->port_type = SAS_PROTO_ALL << 4;
+		else if (phy->sas_phy.role == PHY_ROLE_TARGET)
+			control_phy->port_type = SAS_PROTO_ALL;
+		else
+			control_phy->port_type =
+				(SAS_PROTO_ALL << 4) | SAS_PROTO_ALL;
+
+		/* link reset retries, this should be nominal */
+		control_phy->link_reset_retries = 10;
+
+	case RELEASE_SPINUP_HOLD: /* 0x02 */
+		/* decide the func_mask */
+		control_phy->func_mask = FUNCTION_MASK_DEFAULT;
+		if (phy->phy_desc->flags & ASD_SATA_SPINUP_HOLD)
+			control_phy->func_mask &= ~SPINUP_HOLD_DIS;
+		else
+			control_phy->func_mask |= SPINUP_HOLD_DIS;
+	}
+
+	control_phy->conn_handle = cpu_to_le16(0xFFFF);
+
+	ascb->tasklet_complete = control_phy_tasklet_complete;
+}
+
+/* ---------- INITIATE LINK ADM TASK ---------- */
+
+static void link_adm_tasklet_complete(struct asd_ascb *ascb,
+				      struct done_list_struct *dl)
+{
+	u8 opcode = dl->opcode;
+	struct initiate_link_adm *link_adm = &ascb->scb->link_adm;
+	u8 phy_id = link_adm->phy_id;
+
+	if (opcode != TC_NO_ERROR) {
+		asd_printk("phy%d: link adm task 0x%x completed with error "
+			   "0x%x\n", phy_id, link_adm->sub_func, opcode);
+	}
+	ASD_DPRINTK("phy%d: link adm task 0x%x: 0x%x\n",
+		    phy_id, link_adm->sub_func, opcode);
+
+	asd_ascb_free(ascb);
+}
+
+void asd_build_initiate_link_adm_task(struct asd_ascb *ascb, int phy_id,
+				      u8 subfunc)
+{
+	struct scb *scb = ascb->scb;
+	struct initiate_link_adm *link_adm = &scb->link_adm;
+
+	scb->header.opcode = INITIATE_LINK_ADM_TASK;
+
+	link_adm->phy_id = phy_id;
+	link_adm->sub_func = subfunc;
+	link_adm->conn_handle = cpu_to_le16(0xFFFF);
+
+	ascb->tasklet_complete = link_adm_tasklet_complete;
+}
+
+/* ---------- SCB timer ---------- */
+
+/**
+ * asd_ascb_timedout -- called when a pending SCB's timer has expired
+ * @data: unsigned long, a pointer to the ascb in question
+ *
+ * This is the default timeout function which does the most necessary.
+ * Upper layers can implement their own timeout function, say to free
+ * resources they have with this SCB, and then call this one at the
+ * end of their timeout function.  To do this, one should initialize
+ * the ascb->timer.{function, data, expires} prior to calling the post
+ * funcion.  The timer is started by the post function.
+ */
+void asd_ascb_timedout(unsigned long data)
+{
+	struct asd_ascb *ascb = (void *) data;
+	struct asd_seq_data *seq = &ascb->ha->seq;
+	unsigned long flags;
+
+	ASD_DPRINTK("scb:0x%x timed out\n", ascb->scb->header.opcode);
+
+	spin_lock_irqsave(&seq->pend_q_lock, flags);
+	seq->pending--;
+	list_del_init(&ascb->list);
+	spin_unlock_irqrestore(&seq->pend_q_lock, flags);
+
+	asd_ascb_free(ascb);
+}
+
+/* ---------- CONTROL PHY ---------- */
+
+/* Given the spec value, return a driver value. */
+static const int phy_func_table[] = {
+	[PHY_FUNC_NOP]        = PHY_NO_OP,
+	[PHY_FUNC_LINK_RESET] = ENABLE_PHY,
+	[PHY_FUNC_HARD_RESET] = EXECUTE_HARD_RESET,
+	[PHY_FUNC_DISABLE]    = DISABLE_PHY,
+	[PHY_FUNC_RELEASE_SPINUP_HOLD] = RELEASE_SPINUP_HOLD,
+};
+
+int asd_control_phy(struct asd_sas_phy *phy, enum phy_func func)
+{
+	struct asd_ha_struct *asd_ha = phy->ha->lldd_ha;
+	struct asd_ascb *ascb;
+	int res = 1;
+
+	if (func == PHY_FUNC_CLEAR_ERROR_LOG)
+		return -ENOSYS;
+
+	ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL);
+	if (!ascb)
+		return -ENOMEM;
+
+	asd_build_control_phy(ascb, phy->id, phy_func_table[func]);
+	res = asd_post_ascb_list(asd_ha, ascb , 1);
+	if (res)
+		asd_ascb_free(ascb);
+
+	return res;
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_sds.c b/drivers/scsi/aic94xx/aic94xx_sds.c
new file mode 100644
index 000000000000..eec1e0db0e0f
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_sds.c
@@ -0,0 +1,1136 @@
+/*
+ * Aic94xx SAS/SATA driver access to shared data structures and memory
+ * maps.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/pci.h>
+#include <linux/delay.h>
+
+#include "aic94xx.h"
+#include "aic94xx_reg.h"
+
+/* ---------- OCM stuff ---------- */
+
+struct asd_ocm_dir_ent {
+	u8 type;
+	u8 offs[3];
+	u8 _r1;
+	u8 size[3];
+} __attribute__ ((packed));
+
+struct asd_ocm_dir {
+	char sig[2];
+	u8   _r1[2];
+	u8   major;          /* 0 */
+	u8   minor;          /* 0 */
+	u8   _r2;
+	u8   num_de;
+	struct asd_ocm_dir_ent entry[15];
+} __attribute__ ((packed));
+
+#define	OCM_DE_OCM_DIR			0x00
+#define	OCM_DE_WIN_DRVR			0x01
+#define	OCM_DE_BIOS_CHIM		0x02
+#define	OCM_DE_RAID_ENGN		0x03
+#define	OCM_DE_BIOS_INTL		0x04
+#define	OCM_DE_BIOS_CHIM_OSM		0x05
+#define	OCM_DE_BIOS_CHIM_DYNAMIC	0x06
+#define	OCM_DE_ADDC2C_RES0		0x07
+#define	OCM_DE_ADDC2C_RES1		0x08
+#define	OCM_DE_ADDC2C_RES2		0x09
+#define	OCM_DE_ADDC2C_RES3		0x0A
+
+#define OCM_INIT_DIR_ENTRIES	5
+/***************************************************************************
+*  OCM dircetory default
+***************************************************************************/
+static struct asd_ocm_dir OCMDirInit =
+{
+	.sig = {0x4D, 0x4F},	/* signature */
+	.num_de = OCM_INIT_DIR_ENTRIES,	/* no. of directory entries */
+};
+
+/***************************************************************************
+*  OCM dircetory Entries default
+***************************************************************************/
+static struct asd_ocm_dir_ent OCMDirEntriesInit[OCM_INIT_DIR_ENTRIES] =
+{
+	{
+		.type = (OCM_DE_ADDC2C_RES0),	/* Entry type  */
+		.offs = {128},			/* Offset */
+		.size = {0, 4},			/* size */
+	},
+	{
+		.type = (OCM_DE_ADDC2C_RES1),	/* Entry type  */
+		.offs = {128, 4},		/* Offset */
+		.size = {0, 4},			/* size */
+	},
+	{
+		.type = (OCM_DE_ADDC2C_RES2),	/* Entry type  */
+		.offs = {128, 8},		/* Offset */
+		.size = {0, 4},			/* size */
+	},
+	{
+		.type = (OCM_DE_ADDC2C_RES3),	/* Entry type  */
+		.offs = {128, 12},		/* Offset */
+		.size = {0, 4},			/* size */
+	},
+	{
+		.type = (OCM_DE_WIN_DRVR),	/* Entry type  */
+		.offs = {128, 16},		/* Offset */
+		.size = {128, 235, 1},		/* size */
+	},
+};
+
+struct asd_bios_chim_struct {
+	char sig[4];
+	u8   major;          /* 1 */
+	u8   minor;          /* 0 */
+	u8   bios_major;
+	u8   bios_minor;
+	__le32  bios_build;
+	u8   flags;
+	u8   pci_slot;
+	__le16  ue_num;
+	__le16  ue_size;
+	u8  _r[14];
+	/* The unit element array is right here.
+	 */
+} __attribute__ ((packed));
+
+/**
+ * asd_read_ocm_seg - read an on chip memory (OCM) segment
+ * @asd_ha: pointer to the host adapter structure
+ * @buffer: where to write the read data
+ * @offs: offset into OCM where to read from
+ * @size: how many bytes to read
+ *
+ * Return the number of bytes not read. Return 0 on success.
+ */
+static int asd_read_ocm_seg(struct asd_ha_struct *asd_ha, void *buffer,
+			    u32 offs, int size)
+{
+	u8 *p = buffer;
+	if (unlikely(asd_ha->iospace))
+		asd_read_reg_string(asd_ha, buffer, offs+OCM_BASE_ADDR, size);
+	else {
+		for ( ; size > 0; size--, offs++, p++)
+			*p = asd_read_ocm_byte(asd_ha, offs);
+	}
+	return size;
+}
+
+static int asd_read_ocm_dir(struct asd_ha_struct *asd_ha,
+			    struct asd_ocm_dir *dir, u32 offs)
+{
+	int err = asd_read_ocm_seg(asd_ha, dir, offs, sizeof(*dir));
+	if (err) {
+		ASD_DPRINTK("couldn't read ocm segment\n");
+		return err;
+	}
+
+	if (dir->sig[0] != 'M' || dir->sig[1] != 'O') {
+		ASD_DPRINTK("no valid dir signature(%c%c) at start of OCM\n",
+			    dir->sig[0], dir->sig[1]);
+		return -ENOENT;
+	}
+	if (dir->major != 0) {
+		asd_printk("unsupported major version of ocm dir:0x%x\n",
+			   dir->major);
+		return -ENOENT;
+	}
+	dir->num_de &= 0xf;
+	return 0;
+}
+
+/**
+ * asd_write_ocm_seg - write an on chip memory (OCM) segment
+ * @asd_ha: pointer to the host adapter structure
+ * @buffer: where to read the write data
+ * @offs: offset into OCM to write to
+ * @size: how many bytes to write
+ *
+ * Return the number of bytes not written. Return 0 on success.
+ */
+static void asd_write_ocm_seg(struct asd_ha_struct *asd_ha, void *buffer,
+			    u32 offs, int size)
+{
+	u8 *p = buffer;
+	if (unlikely(asd_ha->iospace))
+		asd_write_reg_string(asd_ha, buffer, offs+OCM_BASE_ADDR, size);
+	else {
+		for ( ; size > 0; size--, offs++, p++)
+			asd_write_ocm_byte(asd_ha, offs, *p);
+	}
+	return;
+}
+
+#define THREE_TO_NUM(X) ((X)[0] | ((X)[1] << 8) | ((X)[2] << 16))
+
+static int asd_find_dir_entry(struct asd_ocm_dir *dir, u8 type,
+			      u32 *offs, u32 *size)
+{
+	int i;
+	struct asd_ocm_dir_ent *ent;
+
+	for (i = 0; i < dir->num_de; i++) {
+		if (dir->entry[i].type == type)
+			break;
+	}
+	if (i >= dir->num_de)
+		return -ENOENT;
+	ent = &dir->entry[i];
+	*offs = (u32) THREE_TO_NUM(ent->offs);
+	*size = (u32) THREE_TO_NUM(ent->size);
+	return 0;
+}
+
+#define OCM_BIOS_CHIM_DE  2
+#define BC_BIOS_PRESENT   1
+
+static int asd_get_bios_chim(struct asd_ha_struct *asd_ha,
+			     struct asd_ocm_dir *dir)
+{
+	int err;
+	struct asd_bios_chim_struct *bc_struct;
+	u32 offs, size;
+
+	err = asd_find_dir_entry(dir, OCM_BIOS_CHIM_DE, &offs, &size);
+	if (err) {
+		ASD_DPRINTK("couldn't find BIOS_CHIM dir ent\n");
+		goto out;
+	}
+	err = -ENOMEM;
+	bc_struct = kmalloc(sizeof(*bc_struct), GFP_KERNEL);
+	if (!bc_struct) {
+		asd_printk("no memory for bios_chim struct\n");
+		goto out;
+	}
+	err = asd_read_ocm_seg(asd_ha, (void *)bc_struct, offs,
+			       sizeof(*bc_struct));
+	if (err) {
+		ASD_DPRINTK("couldn't read ocm segment\n");
+		goto out2;
+	}
+	if (strncmp(bc_struct->sig, "SOIB", 4)
+	    && strncmp(bc_struct->sig, "IPSA", 4)) {
+		ASD_DPRINTK("BIOS_CHIM entry has no valid sig(%c%c%c%c)\n",
+			    bc_struct->sig[0], bc_struct->sig[1],
+			    bc_struct->sig[2], bc_struct->sig[3]);
+		err = -ENOENT;
+		goto out2;
+	}
+	if (bc_struct->major != 1) {
+		asd_printk("BIOS_CHIM unsupported major version:0x%x\n",
+			   bc_struct->major);
+		err = -ENOENT;
+		goto out2;
+	}
+	if (bc_struct->flags & BC_BIOS_PRESENT) {
+		asd_ha->hw_prof.bios.present = 1;
+		asd_ha->hw_prof.bios.maj = bc_struct->bios_major;
+		asd_ha->hw_prof.bios.min = bc_struct->bios_minor;
+		asd_ha->hw_prof.bios.bld = le32_to_cpu(bc_struct->bios_build);
+		ASD_DPRINTK("BIOS present (%d,%d), %d\n",
+			    asd_ha->hw_prof.bios.maj,
+			    asd_ha->hw_prof.bios.min,
+			    asd_ha->hw_prof.bios.bld);
+	}
+	asd_ha->hw_prof.ue.num = le16_to_cpu(bc_struct->ue_num);
+	asd_ha->hw_prof.ue.size= le16_to_cpu(bc_struct->ue_size);
+	ASD_DPRINTK("ue num:%d, ue size:%d\n", asd_ha->hw_prof.ue.num,
+		    asd_ha->hw_prof.ue.size);
+	size = asd_ha->hw_prof.ue.num * asd_ha->hw_prof.ue.size;
+	if (size > 0) {
+		err = -ENOMEM;
+		asd_ha->hw_prof.ue.area = kmalloc(size, GFP_KERNEL);
+		if (!asd_ha->hw_prof.ue.area)
+			goto out2;
+		err = asd_read_ocm_seg(asd_ha, (void *)asd_ha->hw_prof.ue.area,
+				       offs + sizeof(*bc_struct), size);
+		if (err) {
+			kfree(asd_ha->hw_prof.ue.area);
+			asd_ha->hw_prof.ue.area = NULL;
+			asd_ha->hw_prof.ue.num  = 0;
+			asd_ha->hw_prof.ue.size = 0;
+			ASD_DPRINTK("couldn't read ue entries(%d)\n", err);
+		}
+	}
+out2:
+	kfree(bc_struct);
+out:
+	return err;
+}
+
+static void
+asd_hwi_initialize_ocm_dir (struct asd_ha_struct *asd_ha)
+{
+	int i;
+
+	/* Zero OCM */
+	for (i = 0; i < OCM_MAX_SIZE; i += 4)
+		asd_write_ocm_dword(asd_ha, i, 0);
+
+	/* Write Dir */
+	asd_write_ocm_seg(asd_ha, &OCMDirInit, 0,
+			  sizeof(struct asd_ocm_dir));
+
+	/* Write Dir Entries */
+	for (i = 0; i < OCM_INIT_DIR_ENTRIES; i++)
+		asd_write_ocm_seg(asd_ha, &OCMDirEntriesInit[i],
+				  sizeof(struct asd_ocm_dir) +
+				  (i * sizeof(struct asd_ocm_dir_ent))
+				  , sizeof(struct asd_ocm_dir_ent));
+
+}
+
+static int
+asd_hwi_check_ocm_access (struct asd_ha_struct *asd_ha)
+{
+	struct pci_dev *pcidev = asd_ha->pcidev;
+	u32 reg;
+	int err = 0;
+	u32 v;
+
+	/* check if OCM has been initialized by BIOS */
+	reg = asd_read_reg_dword(asd_ha, EXSICNFGR);
+
+	if (!(reg & OCMINITIALIZED)) {
+		err = pci_read_config_dword(pcidev, PCIC_INTRPT_STAT, &v);
+		if (err) {
+			asd_printk("couldn't access PCIC_INTRPT_STAT of %s\n",
+					pci_name(pcidev));
+			goto out;
+		}
+
+		printk(KERN_INFO "OCM is not initialized by BIOS,"
+		       "reinitialize it and ignore it, current IntrptStatus"
+		       "is 0x%x\n", v);
+
+		if (v)
+			err = pci_write_config_dword(pcidev,
+						     PCIC_INTRPT_STAT, v);
+		if (err) {
+			asd_printk("couldn't write PCIC_INTRPT_STAT of %s\n",
+					pci_name(pcidev));
+			goto out;
+		}
+
+		asd_hwi_initialize_ocm_dir(asd_ha);
+
+	}
+out:
+	return err;
+}
+
+/**
+ * asd_read_ocm - read on chip memory (OCM)
+ * @asd_ha: pointer to the host adapter structure
+ */
+int asd_read_ocm(struct asd_ha_struct *asd_ha)
+{
+	int err;
+	struct asd_ocm_dir *dir;
+
+	if (asd_hwi_check_ocm_access(asd_ha))
+		return -1;
+
+	dir = kmalloc(sizeof(*dir), GFP_KERNEL);
+	if (!dir) {
+		asd_printk("no memory for ocm dir\n");
+		return -ENOMEM;
+	}
+
+	err = asd_read_ocm_dir(asd_ha, dir, 0);
+	if (err)
+		goto out;
+
+	err = asd_get_bios_chim(asd_ha, dir);
+out:
+	kfree(dir);
+	return err;
+}
+
+/* ---------- FLASH stuff ---------- */
+
+#define FLASH_RESET			0xF0
+#define FLASH_MANUF_AMD                 1
+
+#define FLASH_SIZE                      0x200000
+#define FLASH_DIR_COOKIE                "*** ADAPTEC FLASH DIRECTORY *** "
+#define FLASH_NEXT_ENTRY_OFFS		0x2000
+#define FLASH_MAX_DIR_ENTRIES		32
+
+#define FLASH_DE_TYPE_MASK              0x3FFFFFFF
+#define FLASH_DE_MS                     0x120
+#define FLASH_DE_CTRL_A_USER            0xE0
+
+struct asd_flash_de {
+	__le32   type;
+	__le32   offs;
+	__le32   pad_size;
+	__le32   image_size;
+	__le32   chksum;
+	u8       _r[12];
+	u8       version[32];
+} __attribute__ ((packed));
+
+struct asd_flash_dir {
+	u8    cookie[32];
+	__le32   rev;		  /* 2 */
+	__le32   chksum;
+	__le32   chksum_antidote;
+	__le32   bld;
+	u8    bld_id[32];	  /* build id data */
+	u8    ver_data[32];	  /* date and time of build */
+	__le32   ae_mask;
+	__le32   v_mask;
+	__le32   oc_mask;
+	u8    _r[20];
+	struct asd_flash_de dir_entry[FLASH_MAX_DIR_ENTRIES];
+} __attribute__ ((packed));
+
+struct asd_manuf_sec {
+	char  sig[2];		  /* 'S', 'M' */
+	u16   offs_next;
+	u8    maj;           /* 0 */
+	u8    min;           /* 0 */
+	u16   chksum;
+	u16   size;
+	u8    _r[6];
+	u8    sas_addr[SAS_ADDR_SIZE];
+	u8    pcba_sn[ASD_PCBA_SN_SIZE];
+	/* Here start the other segments */
+	u8    linked_list[0];
+} __attribute__ ((packed));
+
+struct asd_manuf_phy_desc {
+	u8    state;         /* low 4 bits */
+#define MS_PHY_STATE_ENABLEABLE 0
+#define MS_PHY_STATE_REPORTED   1
+#define MS_PHY_STATE_HIDDEN     2
+	u8    phy_id;
+	u16   _r;
+	u8    phy_control_0; /* mode 5 reg 0x160 */
+	u8    phy_control_1; /* mode 5 reg 0x161 */
+	u8    phy_control_2; /* mode 5 reg 0x162 */
+	u8    phy_control_3; /* mode 5 reg 0x163 */
+} __attribute__ ((packed));
+
+struct asd_manuf_phy_param {
+	char  sig[2];		  /* 'P', 'M' */
+	u16   next;
+	u8    maj;           /* 0 */
+	u8    min;           /* 2 */
+	u8    num_phy_desc;  /* 8 */
+	u8    phy_desc_size; /* 8 */
+	u8    _r[3];
+	u8    usage_model_id;
+	u32   _r2;
+	struct asd_manuf_phy_desc phy_desc[ASD_MAX_PHYS];
+} __attribute__ ((packed));
+
+#if 0
+static const char *asd_sb_type[] = {
+	"unknown",
+	"SGPIO",
+	[2 ... 0x7F] = "unknown",
+	[0x80] = "ADPT_I2C",
+	[0x81 ... 0xFF] = "VENDOR_UNIQUExx"
+};
+#endif
+
+struct asd_ms_sb_desc {
+	u8    type;
+	u8    node_desc_index;
+	u8    conn_desc_index;
+	u8    _recvd[0];
+} __attribute__ ((packed));
+
+#if 0
+static const char *asd_conn_type[] = {
+	[0 ... 7] = "unknown",
+	"SFF8470",
+	"SFF8482",
+	"SFF8484",
+	[0x80] = "PCIX_DAUGHTER0",
+	[0x81] = "SAS_DAUGHTER0",
+	[0x82 ... 0xFF] = "VENDOR_UNIQUExx"
+};
+
+static const char *asd_conn_location[] = {
+	"unknown",
+	"internal",
+	"external",
+	"board_to_board",
+};
+#endif
+
+struct asd_ms_conn_desc {
+	u8    type;
+	u8    location;
+	u8    num_sideband_desc;
+	u8    size_sideband_desc;
+	u32   _resvd;
+	u8    name[16];
+	struct asd_ms_sb_desc sb_desc[0];
+} __attribute__ ((packed));
+
+struct asd_nd_phy_desc {
+	u8    vp_attch_type;
+	u8    attch_specific[0];
+} __attribute__ ((packed));
+
+#if 0
+static const char *asd_node_type[] = {
+	"IOP",
+	"IO_CONTROLLER",
+	"EXPANDER",
+	"PORT_MULTIPLIER",
+	"PORT_MULTIPLEXER",
+	"MULTI_DROP_I2C_BUS",
+};
+#endif
+
+struct asd_ms_node_desc {
+	u8    type;
+	u8    num_phy_desc;
+	u8    size_phy_desc;
+	u8    _resvd;
+	u8    name[16];
+	struct asd_nd_phy_desc phy_desc[0];
+} __attribute__ ((packed));
+
+struct asd_ms_conn_map {
+	char  sig[2];		  /* 'M', 'C' */
+	__le16 next;
+	u8    maj;		  /* 0 */
+	u8    min;		  /* 0 */
+	__le16 cm_size;		  /* size of this struct */
+	u8    num_conn;
+	u8    conn_size;
+	u8    num_nodes;
+	u8    usage_model_id;
+	u32   _resvd;
+	struct asd_ms_conn_desc conn_desc[0];
+	struct asd_ms_node_desc node_desc[0];
+} __attribute__ ((packed));
+
+struct asd_ctrla_phy_entry {
+	u8    sas_addr[SAS_ADDR_SIZE];
+	u8    sas_link_rates;  /* max in hi bits, min in low bits */
+	u8    flags;
+	u8    sata_link_rates;
+	u8    _r[5];
+} __attribute__ ((packed));
+
+struct asd_ctrla_phy_settings {
+	u8    id0;		  /* P'h'y */
+	u8    _r;
+	u16   next;
+	u8    num_phys;	      /* number of PHYs in the PCI function */
+	u8    _r2[3];
+	struct asd_ctrla_phy_entry phy_ent[ASD_MAX_PHYS];
+} __attribute__ ((packed));
+
+struct asd_ll_el {
+	u8   id0;
+	u8   id1;
+	__le16  next;
+	u8   something_here[0];
+} __attribute__ ((packed));
+
+static int asd_poll_flash(struct asd_ha_struct *asd_ha)
+{
+	int c;
+	u8 d;
+
+	for (c = 5000; c > 0; c--) {
+		d  = asd_read_reg_byte(asd_ha, asd_ha->hw_prof.flash.bar);
+		d ^= asd_read_reg_byte(asd_ha, asd_ha->hw_prof.flash.bar);
+		if (!d)
+			return 0;
+		udelay(5);
+	}
+	return -ENOENT;
+}
+
+static int asd_reset_flash(struct asd_ha_struct *asd_ha)
+{
+	int err;
+
+	err = asd_poll_flash(asd_ha);
+	if (err)
+		return err;
+	asd_write_reg_byte(asd_ha, asd_ha->hw_prof.flash.bar, FLASH_RESET);
+	err = asd_poll_flash(asd_ha);
+
+	return err;
+}
+
+static inline int asd_read_flash_seg(struct asd_ha_struct *asd_ha,
+				     void *buffer, u32 offs, int size)
+{
+	asd_read_reg_string(asd_ha, buffer, asd_ha->hw_prof.flash.bar+offs,
+			    size);
+	return 0;
+}
+
+/**
+ * asd_find_flash_dir - finds and reads the flash directory
+ * @asd_ha: pointer to the host adapter structure
+ * @flash_dir: pointer to flash directory structure
+ *
+ * If found, the flash directory segment will be copied to
+ * @flash_dir.  Return 1 if found, 0 if not.
+ */
+static int asd_find_flash_dir(struct asd_ha_struct *asd_ha,
+			      struct asd_flash_dir *flash_dir)
+{
+	u32 v;
+	for (v = 0; v < FLASH_SIZE; v += FLASH_NEXT_ENTRY_OFFS) {
+		asd_read_flash_seg(asd_ha, flash_dir, v,
+				   sizeof(FLASH_DIR_COOKIE)-1);
+		if (memcmp(flash_dir->cookie, FLASH_DIR_COOKIE,
+			   sizeof(FLASH_DIR_COOKIE)-1) == 0) {
+			asd_ha->hw_prof.flash.dir_offs = v;
+			asd_read_flash_seg(asd_ha, flash_dir, v,
+					   sizeof(*flash_dir));
+			return 1;
+		}
+	}
+	return 0;
+}
+
+static int asd_flash_getid(struct asd_ha_struct *asd_ha)
+{
+	int err = 0;
+	u32 reg, inc;
+
+	reg = asd_read_reg_dword(asd_ha, EXSICNFGR);
+
+	if (!(reg & FLASHEX)) {
+		ASD_DPRINTK("flash doesn't exist\n");
+		return -ENOENT;
+	}
+	if (pci_read_config_dword(asd_ha->pcidev, PCI_CONF_FLSH_BAR,
+				  &asd_ha->hw_prof.flash.bar)) {
+		asd_printk("couldn't read PCI_CONF_FLSH_BAR of %s\n",
+			   pci_name(asd_ha->pcidev));
+		return -ENOENT;
+	}
+	asd_ha->hw_prof.flash.present = 1;
+	asd_ha->hw_prof.flash.wide = reg & FLASHW ? 1 : 0;
+	err = asd_reset_flash(asd_ha);
+	if (err) {
+		ASD_DPRINTK("couldn't reset flash(%d)\n", err);
+		return err;
+	}
+	/* Get flash info. This would most likely be AMD Am29LV family flash.
+	 * First try the sequence for word mode.  It is the same as for
+	 * 008B (byte mode only), 160B (word mode) and 800D (word mode).
+	 */
+	reg = asd_ha->hw_prof.flash.bar;
+	inc = asd_ha->hw_prof.flash.wide ? 2 : 1;
+	asd_write_reg_byte(asd_ha, reg + 0x555, 0xAA);
+	asd_write_reg_byte(asd_ha, reg + 0x2AA, 0x55);
+	asd_write_reg_byte(asd_ha, reg + 0x555, 0x90);
+	asd_ha->hw_prof.flash.manuf = asd_read_reg_byte(asd_ha, reg);
+	asd_ha->hw_prof.flash.dev_id= asd_read_reg_byte(asd_ha,reg+inc);
+	asd_ha->hw_prof.flash.sec_prot = asd_read_reg_byte(asd_ha,reg+inc+inc);
+	/* Get out of autoselect mode. */
+	err = asd_reset_flash(asd_ha);
+
+	if (asd_ha->hw_prof.flash.manuf == FLASH_MANUF_AMD) {
+		ASD_DPRINTK("0Found FLASH(%d) manuf:%d, dev_id:0x%x, "
+			    "sec_prot:%d\n",
+			    asd_ha->hw_prof.flash.wide ? 16 : 8,
+			    asd_ha->hw_prof.flash.manuf,
+			    asd_ha->hw_prof.flash.dev_id,
+			    asd_ha->hw_prof.flash.sec_prot);
+		return 0;
+	}
+
+	/* Ok, try the sequence for byte mode of 160B and 800D.
+	 * We may actually never need this.
+	 */
+	asd_write_reg_byte(asd_ha, reg + 0xAAA, 0xAA);
+	asd_write_reg_byte(asd_ha, reg + 0x555, 0x55);
+	asd_write_reg_byte(asd_ha, reg + 0xAAA, 0x90);
+	asd_ha->hw_prof.flash.manuf = asd_read_reg_byte(asd_ha, reg);
+	asd_ha->hw_prof.flash.dev_id = asd_read_reg_byte(asd_ha, reg + 2);
+	asd_ha->hw_prof.flash.sec_prot = asd_read_reg_byte(asd_ha, reg + 4);
+	err = asd_reset_flash(asd_ha);
+
+	if (asd_ha->hw_prof.flash.manuf == FLASH_MANUF_AMD) {
+		ASD_DPRINTK("1Found FLASH(%d) manuf:%d, dev_id:0x%x, "
+			    "sec_prot:%d\n",
+			    asd_ha->hw_prof.flash.wide ? 16 : 8,
+			    asd_ha->hw_prof.flash.manuf,
+			    asd_ha->hw_prof.flash.dev_id,
+			    asd_ha->hw_prof.flash.sec_prot);
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
+static u16 asd_calc_flash_chksum(u16 *p, int size)
+{
+	u16 chksum = 0;
+
+	while (size-- > 0)
+		chksum += *p++;
+
+	return chksum;
+}
+
+
+static int asd_find_flash_de(struct asd_flash_dir *flash_dir, u32 entry_type,
+			     u32 *offs, u32 *size)
+{
+	int i;
+	struct asd_flash_de *de;
+
+	for (i = 0; i < FLASH_MAX_DIR_ENTRIES; i++) {
+		u32 type = le32_to_cpu(flash_dir->dir_entry[i].type);
+
+		type &= FLASH_DE_TYPE_MASK;
+		if (type == entry_type)
+			break;
+	}
+	if (i >= FLASH_MAX_DIR_ENTRIES)
+		return -ENOENT;
+	de = &flash_dir->dir_entry[i];
+	*offs = le32_to_cpu(de->offs);
+	*size = le32_to_cpu(de->pad_size);
+	return 0;
+}
+
+static int asd_validate_ms(struct asd_manuf_sec *ms)
+{
+	if (ms->sig[0] != 'S' || ms->sig[1] != 'M') {
+		ASD_DPRINTK("manuf sec: no valid sig(%c%c)\n",
+			    ms->sig[0], ms->sig[1]);
+		return -ENOENT;
+	}
+	if (ms->maj != 0) {
+		asd_printk("unsupported manuf. sector. major version:%x\n",
+			   ms->maj);
+		return -ENOENT;
+	}
+	ms->offs_next = le16_to_cpu((__force __le16) ms->offs_next);
+	ms->chksum = le16_to_cpu((__force __le16) ms->chksum);
+	ms->size = le16_to_cpu((__force __le16) ms->size);
+
+	if (asd_calc_flash_chksum((u16 *)ms, ms->size/2)) {
+		asd_printk("failed manuf sector checksum\n");
+	}
+
+	return 0;
+}
+
+static int asd_ms_get_sas_addr(struct asd_ha_struct *asd_ha,
+			       struct asd_manuf_sec *ms)
+{
+	memcpy(asd_ha->hw_prof.sas_addr, ms->sas_addr, SAS_ADDR_SIZE);
+	return 0;
+}
+
+static int asd_ms_get_pcba_sn(struct asd_ha_struct *asd_ha,
+			      struct asd_manuf_sec *ms)
+{
+	memcpy(asd_ha->hw_prof.pcba_sn, ms->pcba_sn, ASD_PCBA_SN_SIZE);
+	asd_ha->hw_prof.pcba_sn[ASD_PCBA_SN_SIZE] = '\0';
+	return 0;
+}
+
+/**
+ * asd_find_ll_by_id - find a linked list entry by its id
+ * @start: void pointer to the first element in the linked list
+ * @id0: the first byte of the id  (offs 0)
+ * @id1: the second byte of the id (offs 1)
+ *
+ * @start has to be the _base_ element start, since the
+ * linked list entries's offset is from this pointer.
+ * Some linked list entries use only the first id, in which case
+ * you can pass 0xFF for the second.
+ */
+static void *asd_find_ll_by_id(void * const start, const u8 id0, const u8 id1)
+{
+	struct asd_ll_el *el = start;
+
+	do {
+		switch (id1) {
+		default:
+			if (el->id1 == id1)
+		case 0xFF:
+				if (el->id0 == id0)
+					return el;
+		}
+		el = start + le16_to_cpu(el->next);
+	} while (el != start);
+
+	return NULL;
+}
+
+/**
+ * asd_ms_get_phy_params - get phy parameters from the manufacturing sector
+ * @asd_ha: pointer to the host adapter structure
+ * @manuf_sec: pointer to the manufacturing sector
+ *
+ * The manufacturing sector contans also the linked list of sub-segments,
+ * since when it was read, its size was taken from the flash directory,
+ * not from the structure size.
+ *
+ * HIDDEN phys do not count in the total count.  REPORTED phys cannot
+ * be enabled but are reported and counted towards the total.
+ * ENEBLEABLE phys are enabled by default and count towards the total.
+ * The absolute total phy number is ASD_MAX_PHYS.  hw_prof->num_phys
+ * merely specifies the number of phys the host adapter decided to
+ * report.  E.g., it is possible for phys 0, 1 and 2 to be HIDDEN,
+ * phys 3, 4 and 5 to be REPORTED and phys 6 and 7 to be ENEBLEABLE.
+ * In this case ASD_MAX_PHYS is 8, hw_prof->num_phys is 5, and only 2
+ * are actually enabled (enabled by default, max number of phys
+ * enableable in this case).
+ */
+static int asd_ms_get_phy_params(struct asd_ha_struct *asd_ha,
+				 struct asd_manuf_sec *manuf_sec)
+{
+	int i;
+	int en_phys = 0;
+	int rep_phys = 0;
+	struct asd_manuf_phy_param *phy_param;
+	struct asd_manuf_phy_param dflt_phy_param;
+
+	phy_param = asd_find_ll_by_id(manuf_sec, 'P', 'M');
+	if (!phy_param) {
+		ASD_DPRINTK("ms: no phy parameters found\n");
+		ASD_DPRINTK("ms: Creating default phy parameters\n");
+		dflt_phy_param.sig[0] = 'P';
+		dflt_phy_param.sig[1] = 'M';
+		dflt_phy_param.maj = 0;
+		dflt_phy_param.min = 2;
+		dflt_phy_param.num_phy_desc = 8;
+		dflt_phy_param.phy_desc_size = sizeof(struct asd_manuf_phy_desc);
+		for (i =0; i < ASD_MAX_PHYS; i++) {
+			dflt_phy_param.phy_desc[i].state = 0;
+			dflt_phy_param.phy_desc[i].phy_id = i;
+			dflt_phy_param.phy_desc[i].phy_control_0 = 0xf6;
+			dflt_phy_param.phy_desc[i].phy_control_1 = 0x10;
+			dflt_phy_param.phy_desc[i].phy_control_2 = 0x43;
+			dflt_phy_param.phy_desc[i].phy_control_3 = 0xeb;
+		}
+
+		phy_param = &dflt_phy_param;
+
+	}
+
+	if (phy_param->maj != 0) {
+		asd_printk("unsupported manuf. phy param major version:0x%x\n",
+			   phy_param->maj);
+		return -ENOENT;
+	}
+
+	ASD_DPRINTK("ms: num_phy_desc: %d\n", phy_param->num_phy_desc);
+	asd_ha->hw_prof.enabled_phys = 0;
+	for (i = 0; i < phy_param->num_phy_desc; i++) {
+		struct asd_manuf_phy_desc *pd = &phy_param->phy_desc[i];
+		switch (pd->state & 0xF) {
+		case MS_PHY_STATE_HIDDEN:
+			ASD_DPRINTK("ms: phy%d: HIDDEN\n", i);
+			continue;
+		case MS_PHY_STATE_REPORTED:
+			ASD_DPRINTK("ms: phy%d: REPORTED\n", i);
+			asd_ha->hw_prof.enabled_phys &= ~(1 << i);
+			rep_phys++;
+			continue;
+		case MS_PHY_STATE_ENABLEABLE:
+			ASD_DPRINTK("ms: phy%d: ENEBLEABLE\n", i);
+			asd_ha->hw_prof.enabled_phys |= (1 << i);
+			en_phys++;
+			break;
+		}
+		asd_ha->hw_prof.phy_desc[i].phy_control_0 = pd->phy_control_0;
+		asd_ha->hw_prof.phy_desc[i].phy_control_1 = pd->phy_control_1;
+		asd_ha->hw_prof.phy_desc[i].phy_control_2 = pd->phy_control_2;
+		asd_ha->hw_prof.phy_desc[i].phy_control_3 = pd->phy_control_3;
+	}
+	asd_ha->hw_prof.max_phys = rep_phys + en_phys;
+	asd_ha->hw_prof.num_phys = en_phys;
+	ASD_DPRINTK("ms: max_phys:0x%x, num_phys:0x%x\n",
+		    asd_ha->hw_prof.max_phys, asd_ha->hw_prof.num_phys);
+	ASD_DPRINTK("ms: enabled_phys:0x%x\n", asd_ha->hw_prof.enabled_phys);
+	return 0;
+}
+
+static int asd_ms_get_connector_map(struct asd_ha_struct *asd_ha,
+				    struct asd_manuf_sec *manuf_sec)
+{
+	struct asd_ms_conn_map *cm;
+
+	cm = asd_find_ll_by_id(manuf_sec, 'M', 'C');
+	if (!cm) {
+		ASD_DPRINTK("ms: no connector map found\n");
+		return 0;
+	}
+
+	if (cm->maj != 0) {
+		ASD_DPRINTK("ms: unsupported: connector map major version 0x%x"
+			    "\n", cm->maj);
+		return -ENOENT;
+	}
+
+	/* XXX */
+
+	return 0;
+}
+
+
+/**
+ * asd_process_ms - find and extract information from the manufacturing sector
+ * @asd_ha: pointer to the host adapter structure
+ * @flash_dir: pointer to the flash directory
+ */
+static int asd_process_ms(struct asd_ha_struct *asd_ha,
+			  struct asd_flash_dir *flash_dir)
+{
+	int err;
+	struct asd_manuf_sec *manuf_sec;
+	u32 offs, size;
+
+	err = asd_find_flash_de(flash_dir, FLASH_DE_MS, &offs, &size);
+	if (err) {
+		ASD_DPRINTK("Couldn't find the manuf. sector\n");
+		goto out;
+	}
+
+	if (size == 0)
+		goto out;
+
+	err = -ENOMEM;
+	manuf_sec = kmalloc(size, GFP_KERNEL);
+	if (!manuf_sec) {
+		ASD_DPRINTK("no mem for manuf sector\n");
+		goto out;
+	}
+
+	err = asd_read_flash_seg(asd_ha, (void *)manuf_sec, offs, size);
+	if (err) {
+		ASD_DPRINTK("couldn't read manuf sector at 0x%x, size 0x%x\n",
+			    offs, size);
+		goto out2;
+	}
+
+	err = asd_validate_ms(manuf_sec);
+	if (err) {
+		ASD_DPRINTK("couldn't validate manuf sector\n");
+		goto out2;
+	}
+
+	err = asd_ms_get_sas_addr(asd_ha, manuf_sec);
+	if (err) {
+		ASD_DPRINTK("couldn't read the SAS_ADDR\n");
+		goto out2;
+	}
+	ASD_DPRINTK("manuf sect SAS_ADDR %llx\n",
+		    SAS_ADDR(asd_ha->hw_prof.sas_addr));
+
+	err = asd_ms_get_pcba_sn(asd_ha, manuf_sec);
+	if (err) {
+		ASD_DPRINTK("couldn't read the PCBA SN\n");
+		goto out2;
+	}
+	ASD_DPRINTK("manuf sect PCBA SN %s\n", asd_ha->hw_prof.pcba_sn);
+
+	err = asd_ms_get_phy_params(asd_ha, manuf_sec);
+	if (err) {
+		ASD_DPRINTK("ms: couldn't get phy parameters\n");
+		goto out2;
+	}
+
+	err = asd_ms_get_connector_map(asd_ha, manuf_sec);
+	if (err) {
+		ASD_DPRINTK("ms: couldn't get connector map\n");
+		goto out2;
+	}
+
+out2:
+	kfree(manuf_sec);
+out:
+	return err;
+}
+
+static int asd_process_ctrla_phy_settings(struct asd_ha_struct *asd_ha,
+					  struct asd_ctrla_phy_settings *ps)
+{
+	int i;
+	for (i = 0; i < ps->num_phys; i++) {
+		struct asd_ctrla_phy_entry *pe = &ps->phy_ent[i];
+
+		if (!PHY_ENABLED(asd_ha, i))
+			continue;
+		if (*(u64 *)pe->sas_addr == 0) {
+			asd_ha->hw_prof.enabled_phys &= ~(1 << i);
+			continue;
+		}
+		/* This is the SAS address which should be sent in IDENTIFY. */
+		memcpy(asd_ha->hw_prof.phy_desc[i].sas_addr, pe->sas_addr,
+		       SAS_ADDR_SIZE);
+		asd_ha->hw_prof.phy_desc[i].max_sas_lrate =
+			(pe->sas_link_rates & 0xF0) >> 4;
+		asd_ha->hw_prof.phy_desc[i].min_sas_lrate =
+			(pe->sas_link_rates & 0x0F);
+		asd_ha->hw_prof.phy_desc[i].max_sata_lrate =
+			(pe->sata_link_rates & 0xF0) >> 4;
+		asd_ha->hw_prof.phy_desc[i].min_sata_lrate =
+			(pe->sata_link_rates & 0x0F);
+		asd_ha->hw_prof.phy_desc[i].flags = pe->flags;
+		ASD_DPRINTK("ctrla: phy%d: sas_addr: %llx, sas rate:0x%x-0x%x,"
+			    " sata rate:0x%x-0x%x, flags:0x%x\n",
+			    i,
+			    SAS_ADDR(asd_ha->hw_prof.phy_desc[i].sas_addr),
+			    asd_ha->hw_prof.phy_desc[i].max_sas_lrate,
+			    asd_ha->hw_prof.phy_desc[i].min_sas_lrate,
+			    asd_ha->hw_prof.phy_desc[i].max_sata_lrate,
+			    asd_ha->hw_prof.phy_desc[i].min_sata_lrate,
+			    asd_ha->hw_prof.phy_desc[i].flags);
+	}
+
+	return 0;
+}
+
+/**
+ * asd_process_ctrl_a_user - process CTRL-A user settings
+ * @asd_ha: pointer to the host adapter structure
+ * @flash_dir: pointer to the flash directory
+ */
+static int asd_process_ctrl_a_user(struct asd_ha_struct *asd_ha,
+				   struct asd_flash_dir *flash_dir)
+{
+	int err, i;
+	u32 offs, size;
+	struct asd_ll_el *el;
+	struct asd_ctrla_phy_settings *ps;
+	struct asd_ctrla_phy_settings dflt_ps;
+
+	err = asd_find_flash_de(flash_dir, FLASH_DE_CTRL_A_USER, &offs, &size);
+	if (err) {
+		ASD_DPRINTK("couldn't find CTRL-A user settings section\n");
+		ASD_DPRINTK("Creating default CTRL-A user settings section\n");
+
+		dflt_ps.id0 = 'h';
+		dflt_ps.num_phys = 8;
+		for (i =0; i < ASD_MAX_PHYS; i++) {
+			memcpy(dflt_ps.phy_ent[i].sas_addr,
+			       asd_ha->hw_prof.sas_addr, SAS_ADDR_SIZE);
+			dflt_ps.phy_ent[i].sas_link_rates = 0x98;
+			dflt_ps.phy_ent[i].flags = 0x0;
+			dflt_ps.phy_ent[i].sata_link_rates = 0x0;
+		}
+
+		size = sizeof(struct asd_ctrla_phy_settings);
+		ps = &dflt_ps;
+	}
+
+	if (size == 0)
+		goto out;
+
+	err = -ENOMEM;
+	el = kmalloc(size, GFP_KERNEL);
+	if (!el) {
+		ASD_DPRINTK("no mem for ctrla user settings section\n");
+		goto out;
+	}
+
+	err = asd_read_flash_seg(asd_ha, (void *)el, offs, size);
+	if (err) {
+		ASD_DPRINTK("couldn't read ctrla phy settings section\n");
+		goto out2;
+	}
+
+	err = -ENOENT;
+	ps = asd_find_ll_by_id(el, 'h', 0xFF);
+	if (!ps) {
+		ASD_DPRINTK("couldn't find ctrla phy settings struct\n");
+		goto out2;
+	}
+
+	err = asd_process_ctrla_phy_settings(asd_ha, ps);
+	if (err) {
+		ASD_DPRINTK("couldn't process ctrla phy settings\n");
+		goto out2;
+	}
+out2:
+	kfree(el);
+out:
+	return err;
+}
+
+/**
+ * asd_read_flash - read flash memory
+ * @asd_ha: pointer to the host adapter structure
+ */
+int asd_read_flash(struct asd_ha_struct *asd_ha)
+{
+	int err;
+	struct asd_flash_dir *flash_dir;
+
+	err = asd_flash_getid(asd_ha);
+	if (err)
+		return err;
+
+	flash_dir = kmalloc(sizeof(*flash_dir), GFP_KERNEL);
+	if (!flash_dir)
+		return -ENOMEM;
+
+	err = -ENOENT;
+	if (!asd_find_flash_dir(asd_ha, flash_dir)) {
+		ASD_DPRINTK("couldn't find flash directory\n");
+		goto out;
+	}
+
+	if (le32_to_cpu(flash_dir->rev) != 2) {
+		asd_printk("unsupported flash dir version:0x%x\n",
+			   le32_to_cpu(flash_dir->rev));
+		goto out;
+	}
+
+	err = asd_process_ms(asd_ha, flash_dir);
+	if (err) {
+		ASD_DPRINTK("couldn't process manuf sector settings\n");
+		goto out;
+	}
+
+	err = asd_process_ctrl_a_user(asd_ha, flash_dir);
+	if (err) {
+		ASD_DPRINTK("couldn't process CTRL-A user settings\n");
+		goto out;
+	}
+
+out:
+	kfree(flash_dir);
+	return err;
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_seq.c b/drivers/scsi/aic94xx/aic94xx_seq.c
new file mode 100644
index 000000000000..9050c6f3f6bd
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_seq.c
@@ -0,0 +1,1401 @@
+/*
+ * Aic94xx SAS/SATA driver sequencer interface.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * Parts of this code adapted from David Chaw's adp94xx_seq.c.
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/firmware.h>
+#include "aic94xx_reg.h"
+#include "aic94xx_hwi.h"
+
+#include "aic94xx_seq.h"
+#include "aic94xx_dump.h"
+
+/* It takes no more than 0.05 us for an instruction
+ * to complete. So waiting for 1 us should be more than
+ * plenty.
+ */
+#define PAUSE_DELAY 1
+#define PAUSE_TRIES 1000
+
+static const struct firmware *sequencer_fw;
+static const char *sequencer_version;
+static u16 cseq_vecs[CSEQ_NUM_VECS], lseq_vecs[LSEQ_NUM_VECS], mode2_task,
+	cseq_idle_loop, lseq_idle_loop;
+static u8 *cseq_code, *lseq_code;
+static u32 cseq_code_size, lseq_code_size;
+
+static u16 first_scb_site_no = 0xFFFF;
+static u16 last_scb_site_no;
+
+/* ---------- Pause/Unpause CSEQ/LSEQ ---------- */
+
+/**
+ * asd_pause_cseq - pause the central sequencer
+ * @asd_ha: pointer to host adapter structure
+ *
+ * Return 0 on success, negative on failure.
+ */
+int asd_pause_cseq(struct asd_ha_struct *asd_ha)
+{
+	int	count = PAUSE_TRIES;
+	u32	arp2ctl;
+
+	arp2ctl = asd_read_reg_dword(asd_ha, CARP2CTL);
+	if (arp2ctl & PAUSED)
+		return 0;
+
+	asd_write_reg_dword(asd_ha, CARP2CTL, arp2ctl | EPAUSE);
+	do {
+		arp2ctl = asd_read_reg_dword(asd_ha, CARP2CTL);
+		if (arp2ctl & PAUSED)
+			return 0;
+		udelay(PAUSE_DELAY);
+	} while (--count > 0);
+
+	ASD_DPRINTK("couldn't pause CSEQ\n");
+	return -1;
+}
+
+/**
+ * asd_unpause_cseq - unpause the central sequencer.
+ * @asd_ha: pointer to host adapter structure.
+ *
+ * Return 0 on success, negative on error.
+ */
+int asd_unpause_cseq(struct asd_ha_struct *asd_ha)
+{
+	u32	arp2ctl;
+	int	count = PAUSE_TRIES;
+
+	arp2ctl = asd_read_reg_dword(asd_ha, CARP2CTL);
+	if (!(arp2ctl & PAUSED))
+		return 0;
+
+	asd_write_reg_dword(asd_ha, CARP2CTL, arp2ctl & ~EPAUSE);
+	do {
+		arp2ctl = asd_read_reg_dword(asd_ha, CARP2CTL);
+		if (!(arp2ctl & PAUSED))
+			return 0;
+		udelay(PAUSE_DELAY);
+	} while (--count > 0);
+
+	ASD_DPRINTK("couldn't unpause the CSEQ\n");
+	return -1;
+}
+
+/**
+ * asd_seq_pause_lseq - pause a link sequencer
+ * @asd_ha: pointer to a host adapter structure
+ * @lseq: link sequencer of interest
+ *
+ * Return 0 on success, negative on error.
+ */
+static inline int asd_seq_pause_lseq(struct asd_ha_struct *asd_ha, int lseq)
+{
+	u32    arp2ctl;
+	int    count = PAUSE_TRIES;
+
+	arp2ctl = asd_read_reg_dword(asd_ha, LmARP2CTL(lseq));
+	if (arp2ctl & PAUSED)
+		return 0;
+
+	asd_write_reg_dword(asd_ha, LmARP2CTL(lseq), arp2ctl | EPAUSE);
+	do {
+		arp2ctl = asd_read_reg_dword(asd_ha, LmARP2CTL(lseq));
+		if (arp2ctl & PAUSED)
+			return 0;
+		udelay(PAUSE_DELAY);
+	} while (--count > 0);
+
+	ASD_DPRINTK("couldn't pause LSEQ %d\n", lseq);
+	return -1;
+}
+
+/**
+ * asd_pause_lseq - pause the link sequencer(s)
+ * @asd_ha: pointer to host adapter structure
+ * @lseq_mask: mask of link sequencers of interest
+ *
+ * Return 0 on success, negative on failure.
+ */
+int asd_pause_lseq(struct asd_ha_struct *asd_ha, u8 lseq_mask)
+{
+	int lseq;
+	int err = 0;
+
+	for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+		err = asd_seq_pause_lseq(asd_ha, lseq);
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
+/**
+ * asd_seq_unpause_lseq - unpause a link sequencer
+ * @asd_ha: pointer to host adapter structure
+ * @lseq: link sequencer of interest
+ *
+ * Return 0 on success, negative on error.
+ */
+static inline int asd_seq_unpause_lseq(struct asd_ha_struct *asd_ha, int lseq)
+{
+	u32 arp2ctl;
+	int count = PAUSE_TRIES;
+
+	arp2ctl = asd_read_reg_dword(asd_ha, LmARP2CTL(lseq));
+	if (!(arp2ctl & PAUSED))
+		return 0;
+
+	asd_write_reg_dword(asd_ha, LmARP2CTL(lseq), arp2ctl & ~EPAUSE);
+	do {
+		arp2ctl = asd_read_reg_dword(asd_ha, LmARP2CTL(lseq));
+		if (!(arp2ctl & PAUSED))
+			return 0;
+		udelay(PAUSE_DELAY);
+	} while (--count > 0);
+
+	ASD_DPRINTK("couldn't unpause LSEQ %d\n", lseq);
+	return 0;
+}
+
+
+/**
+ * asd_unpause_lseq - unpause the link sequencer(s)
+ * @asd_ha: pointer to host adapter structure
+ * @lseq_mask: mask of link sequencers of interest
+ *
+ * Return 0 on success, negative on failure.
+ */
+int asd_unpause_lseq(struct asd_ha_struct *asd_ha, u8 lseq_mask)
+{
+	int lseq;
+	int err = 0;
+
+	for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+		err = asd_seq_unpause_lseq(asd_ha, lseq);
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
+/* ---------- Downloading CSEQ/LSEQ microcode ---------- */
+
+static int asd_verify_cseq(struct asd_ha_struct *asd_ha, const u8 *_prog,
+			   u32 size)
+{
+	u32 addr = CSEQ_RAM_REG_BASE_ADR;
+	const u32 *prog = (u32 *) _prog;
+	u32 i;
+
+	for (i = 0; i < size; i += 4, prog++, addr += 4) {
+		u32 val = asd_read_reg_dword(asd_ha, addr);
+
+		if (le32_to_cpu(*prog) != val) {
+			asd_printk("%s: cseq verify failed at %u "
+				   "read:0x%x, wanted:0x%x\n",
+				   pci_name(asd_ha->pcidev),
+				   i, val, le32_to_cpu(*prog));
+			return -1;
+		}
+	}
+	ASD_DPRINTK("verified %d bytes, passed\n", size);
+	return 0;
+}
+
+/**
+ * asd_verify_lseq - verify the microcode of a link sequencer
+ * @asd_ha: pointer to host adapter structure
+ * @_prog: pointer to the microcode
+ * @size: size of the microcode in bytes
+ * @lseq: link sequencer of interest
+ *
+ * The link sequencer code is accessed in 4 KB pages, which are selected
+ * by setting LmRAMPAGE (bits 8 and 9) of the LmBISTCTL1 register.
+ * The 10 KB LSEQm instruction code is mapped, page at a time, at
+ * LmSEQRAM address.
+ */
+static int asd_verify_lseq(struct asd_ha_struct *asd_ha, const u8 *_prog,
+			   u32 size, int lseq)
+{
+#define LSEQ_CODEPAGE_SIZE 4096
+	int pages =  (size + LSEQ_CODEPAGE_SIZE - 1) / LSEQ_CODEPAGE_SIZE;
+	u32 page;
+	const u32 *prog = (u32 *) _prog;
+
+	for (page = 0; page < pages; page++) {
+		u32 i;
+
+		asd_write_reg_dword(asd_ha, LmBISTCTL1(lseq),
+				    page << LmRAMPAGE_LSHIFT);
+		for (i = 0; size > 0 && i < LSEQ_CODEPAGE_SIZE;
+		     i += 4, prog++, size-=4) {
+
+			u32 val = asd_read_reg_dword(asd_ha, LmSEQRAM(lseq)+i);
+
+			if (le32_to_cpu(*prog) != val) {
+				asd_printk("%s: LSEQ%d verify failed "
+					   "page:%d, offs:%d\n",
+					   pci_name(asd_ha->pcidev),
+					   lseq, page, i);
+				return -1;
+			}
+		}
+	}
+	ASD_DPRINTK("LSEQ%d verified %d bytes, passed\n", lseq,
+		    (int)((u8 *)prog-_prog));
+	return 0;
+}
+
+/**
+ * asd_verify_seq -- verify CSEQ/LSEQ microcode
+ * @asd_ha: pointer to host adapter structure
+ * @prog: pointer to microcode
+ * @size: size of the microcode
+ * @lseq_mask: if 0, verify CSEQ microcode, else mask of LSEQs of interest
+ *
+ * Return 0 if microcode is correct, negative on mismatch.
+ */
+static int asd_verify_seq(struct asd_ha_struct *asd_ha, const u8 *prog,
+			      u32 size, u8 lseq_mask)
+{
+	if (lseq_mask == 0)
+		return asd_verify_cseq(asd_ha, prog, size);
+	else {
+		int lseq, err;
+
+		for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+			err = asd_verify_lseq(asd_ha, prog, size, lseq);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+#define ASD_DMA_MODE_DOWNLOAD
+#ifdef ASD_DMA_MODE_DOWNLOAD
+/* This is the size of the CSEQ Mapped instruction page */
+#define MAX_DMA_OVLY_COUNT ((1U << 14)-1)
+static int asd_download_seq(struct asd_ha_struct *asd_ha,
+			    const u8 * const prog, u32 size, u8 lseq_mask)
+{
+	u32 comstaten;
+	u32 reg;
+	int page;
+	const int pages = (size + MAX_DMA_OVLY_COUNT - 1) / MAX_DMA_OVLY_COUNT;
+	struct asd_dma_tok *token;
+	int err = 0;
+
+	if (size % 4) {
+		asd_printk("sequencer program not multiple of 4\n");
+		return -1;
+	}
+
+	asd_pause_cseq(asd_ha);
+	asd_pause_lseq(asd_ha, 0xFF);
+
+	/* save, disable and clear interrupts */
+	comstaten = asd_read_reg_dword(asd_ha, COMSTATEN);
+	asd_write_reg_dword(asd_ha, COMSTATEN, 0);
+	asd_write_reg_dword(asd_ha, COMSTAT, COMSTAT_MASK);
+
+	asd_write_reg_dword(asd_ha, CHIMINTEN, RST_CHIMINTEN);
+	asd_write_reg_dword(asd_ha, CHIMINT, CHIMINT_MASK);
+
+	token = asd_alloc_coherent(asd_ha, MAX_DMA_OVLY_COUNT, GFP_KERNEL);
+	if (!token) {
+		asd_printk("out of memory for dma SEQ download\n");
+		err = -ENOMEM;
+		goto out;
+	}
+	ASD_DPRINTK("dma-ing %d bytes\n", size);
+
+	for (page = 0; page < pages; page++) {
+		int i;
+		u32 left = min(size-page*MAX_DMA_OVLY_COUNT,
+			       (u32)MAX_DMA_OVLY_COUNT);
+
+		memcpy(token->vaddr, prog + page*MAX_DMA_OVLY_COUNT, left);
+		asd_write_reg_addr(asd_ha, OVLYDMAADR, token->dma_handle);
+		asd_write_reg_dword(asd_ha, OVLYDMACNT, left);
+		reg = !page ? RESETOVLYDMA : 0;
+		reg |= (STARTOVLYDMA | OVLYHALTERR);
+		reg |= (lseq_mask ? (((u32)lseq_mask) << 8) : OVLYCSEQ);
+		/* Start DMA. */
+		asd_write_reg_dword(asd_ha, OVLYDMACTL, reg);
+
+		for (i = PAUSE_TRIES*100; i > 0; i--) {
+			u32 dmadone = asd_read_reg_dword(asd_ha, OVLYDMACTL);
+			if (!(dmadone & OVLYDMAACT))
+				break;
+			udelay(PAUSE_DELAY);
+		}
+	}
+
+	reg = asd_read_reg_dword(asd_ha, COMSTAT);
+	if (!(reg & OVLYDMADONE) || (reg & OVLYERR)
+	    || (asd_read_reg_dword(asd_ha, CHIMINT) & DEVEXCEPT_MASK)){
+		asd_printk("%s: error DMA-ing sequencer code\n",
+			   pci_name(asd_ha->pcidev));
+		err = -ENODEV;
+	}
+
+	asd_free_coherent(asd_ha, token);
+ out:
+	asd_write_reg_dword(asd_ha, COMSTATEN, comstaten);
+
+	return err ? : asd_verify_seq(asd_ha, prog, size, lseq_mask);
+}
+#else /* ASD_DMA_MODE_DOWNLOAD */
+static int asd_download_seq(struct asd_ha_struct *asd_ha, const u8 *_prog,
+			    u32 size, u8 lseq_mask)
+{
+	int i;
+	u32 reg = 0;
+	const u32 *prog = (u32 *) _prog;
+
+	if (size % 4) {
+		asd_printk("sequencer program not multiple of 4\n");
+		return -1;
+	}
+
+	asd_pause_cseq(asd_ha);
+	asd_pause_lseq(asd_ha, 0xFF);
+
+	reg |= (lseq_mask ? (((u32)lseq_mask) << 8) : OVLYCSEQ);
+	reg |= PIOCMODE;
+
+	asd_write_reg_dword(asd_ha, OVLYDMACNT, size);
+	asd_write_reg_dword(asd_ha, OVLYDMACTL, reg);
+
+	ASD_DPRINTK("downloading %s sequencer%s in PIO mode...\n",
+		    lseq_mask ? "LSEQ" : "CSEQ", lseq_mask ? "s" : "");
+
+	for (i = 0; i < size; i += 4, prog++)
+		asd_write_reg_dword(asd_ha, SPIODATA, *prog);
+
+	reg = (reg & ~PIOCMODE) | OVLYHALTERR;
+	asd_write_reg_dword(asd_ha, OVLYDMACTL, reg);
+
+	return asd_verify_seq(asd_ha, _prog, size, lseq_mask);
+}
+#endif /* ASD_DMA_MODE_DOWNLOAD */
+
+/**
+ * asd_seq_download_seqs - download the sequencer microcode
+ * @asd_ha: pointer to host adapter structure
+ *
+ * Download the central and link sequencer microcode.
+ */
+static int asd_seq_download_seqs(struct asd_ha_struct *asd_ha)
+{
+	int 	err;
+
+	if (!asd_ha->hw_prof.enabled_phys) {
+		asd_printk("%s: no enabled phys!\n", pci_name(asd_ha->pcidev));
+		return -ENODEV;
+	}
+
+	/* Download the CSEQ */
+	ASD_DPRINTK("downloading CSEQ...\n");
+	err = asd_download_seq(asd_ha, cseq_code, cseq_code_size, 0);
+	if (err) {
+		asd_printk("CSEQ download failed:%d\n", err);
+		return err;
+	}
+
+	/* Download the Link Sequencers code. All of the Link Sequencers
+	 * microcode can be downloaded at the same time.
+	 */
+	ASD_DPRINTK("downloading LSEQs...\n");
+	err = asd_download_seq(asd_ha, lseq_code, lseq_code_size,
+			       asd_ha->hw_prof.enabled_phys);
+	if (err) {
+		/* Try it one at a time */
+		u8 lseq;
+		u8 lseq_mask = asd_ha->hw_prof.enabled_phys;
+
+		for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+			err = asd_download_seq(asd_ha, lseq_code,
+					       lseq_code_size, 1<<lseq);
+			if (err)
+				break;
+		}
+	}
+	if (err)
+		asd_printk("LSEQs download failed:%d\n", err);
+
+	return err;
+}
+
+/* ---------- Initializing the chip, chip memory, etc. ---------- */
+
+/**
+ * asd_init_cseq_mip - initialize CSEQ mode independent pages 4-7
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_init_cseq_mip(struct asd_ha_struct *asd_ha)
+{
+	/* CSEQ Mode Independent, page 4 setup. */
+	asd_write_reg_word(asd_ha, CSEQ_Q_EXE_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_EXE_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_DONE_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_DONE_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_SEND_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_SEND_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_DMA2CHIM_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_DMA2CHIM_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_COPY_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_COPY_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_REG0, 0);
+	asd_write_reg_word(asd_ha, CSEQ_REG1, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_REG2, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_LINK_CTL_Q_MAP, 0);
+	{
+		u8 con = asd_read_reg_byte(asd_ha, CCONEXIST);
+		u8 val = hweight8(con);
+		asd_write_reg_byte(asd_ha, CSEQ_MAX_CSEQ_MODE, (val<<4)|val);
+	}
+	asd_write_reg_word(asd_ha, CSEQ_FREE_LIST_HACK_COUNT, 0);
+
+	/* CSEQ Mode independent, page 5 setup. */
+	asd_write_reg_dword(asd_ha, CSEQ_EST_NEXUS_REQ_QUEUE, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_EST_NEXUS_REQ_QUEUE+4, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_EST_NEXUS_REQ_COUNT, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_EST_NEXUS_REQ_COUNT+4, 0);
+	asd_write_reg_word(asd_ha, CSEQ_Q_EST_NEXUS_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_EST_NEXUS_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_NEED_EST_NEXUS_SCB, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_EST_NEXUS_REQ_HEAD, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_EST_NEXUS_REQ_TAIL, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_EST_NEXUS_SCB_OFFSET, 0);
+
+	/* CSEQ Mode independent, page 6 setup. */
+	asd_write_reg_word(asd_ha, CSEQ_INT_ROUT_RET_ADDR0, 0);
+	asd_write_reg_word(asd_ha, CSEQ_INT_ROUT_RET_ADDR1, 0);
+	asd_write_reg_word(asd_ha, CSEQ_INT_ROUT_SCBPTR, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_INT_ROUT_MODE, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_ISR_SCRATCH_FLAGS, 0);
+	asd_write_reg_word(asd_ha, CSEQ_ISR_SAVE_SINDEX, 0);
+	asd_write_reg_word(asd_ha, CSEQ_ISR_SAVE_DINDEX, 0);
+	asd_write_reg_word(asd_ha, CSEQ_Q_MONIRTT_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_MONIRTT_TAIL, 0xFFFF);
+	/* Calculate the free scb mask. */
+	{
+		u16 cmdctx = asd_get_cmdctx_size(asd_ha);
+		cmdctx = (~((cmdctx/128)-1)) >> 8;
+		asd_write_reg_byte(asd_ha, CSEQ_FREE_SCB_MASK, (u8)cmdctx);
+	}
+	asd_write_reg_word(asd_ha, CSEQ_BUILTIN_FREE_SCB_HEAD,
+			   first_scb_site_no);
+	asd_write_reg_word(asd_ha, CSEQ_BUILTIN_FREE_SCB_TAIL,
+			   last_scb_site_no);
+	asd_write_reg_word(asd_ha, CSEQ_EXTENDED_FREE_SCB_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_EXTENDED_FREE_SCB_TAIL, 0xFFFF);
+
+	/* CSEQ Mode independent, page 7 setup. */
+	asd_write_reg_dword(asd_ha, CSEQ_EMPTY_REQ_QUEUE, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_EMPTY_REQ_QUEUE+4, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_EMPTY_REQ_COUNT, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_EMPTY_REQ_COUNT+4, 0);
+	asd_write_reg_word(asd_ha, CSEQ_Q_EMPTY_HEAD, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_Q_EMPTY_TAIL, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_NEED_EMPTY_SCB, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_EMPTY_REQ_HEAD, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_EMPTY_REQ_TAIL, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_EMPTY_SCB_OFFSET, 0);
+	asd_write_reg_word(asd_ha, CSEQ_PRIMITIVE_DATA, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_TIMEOUT_CONST, 0);
+}
+
+/**
+ * asd_init_cseq_mdp - initialize CSEQ Mode dependent pages
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_init_cseq_mdp(struct asd_ha_struct *asd_ha)
+{
+	int	i;
+	int	moffs;
+
+	moffs = CSEQ_PAGE_SIZE * 2;
+
+	/* CSEQ Mode dependent, modes 0-7, page 0 setup. */
+	for (i = 0; i < 8; i++) {
+		asd_write_reg_word(asd_ha, i*moffs+CSEQ_LRM_SAVE_SINDEX, 0);
+		asd_write_reg_word(asd_ha, i*moffs+CSEQ_LRM_SAVE_SCBPTR, 0);
+		asd_write_reg_word(asd_ha, i*moffs+CSEQ_Q_LINK_HEAD, 0xFFFF);
+		asd_write_reg_word(asd_ha, i*moffs+CSEQ_Q_LINK_TAIL, 0xFFFF);
+		asd_write_reg_byte(asd_ha, i*moffs+CSEQ_LRM_SAVE_SCRPAGE, 0);
+	}
+
+	/* CSEQ Mode dependent, mode 0-7, page 1 and 2 shall be ignored. */
+
+	/* CSEQ Mode dependent, mode 8, page 0 setup. */
+	asd_write_reg_word(asd_ha, CSEQ_RET_ADDR, 0xFFFF);
+	asd_write_reg_word(asd_ha, CSEQ_RET_SCBPTR, 0);
+	asd_write_reg_word(asd_ha, CSEQ_SAVE_SCBPTR, 0);
+	asd_write_reg_word(asd_ha, CSEQ_EMPTY_TRANS_CTX, 0);
+	asd_write_reg_word(asd_ha, CSEQ_RESP_LEN, 0);
+	asd_write_reg_word(asd_ha, CSEQ_TMF_SCBPTR, 0);
+	asd_write_reg_word(asd_ha, CSEQ_GLOBAL_PREV_SCB, 0);
+	asd_write_reg_word(asd_ha, CSEQ_GLOBAL_HEAD, 0);
+	asd_write_reg_word(asd_ha, CSEQ_CLEAR_LU_HEAD, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_TMF_OPCODE, 0);
+	asd_write_reg_byte(asd_ha, CSEQ_SCRATCH_FLAGS, 0);
+	asd_write_reg_word(asd_ha, CSEQ_HSB_SITE, 0);
+	asd_write_reg_word(asd_ha, CSEQ_FIRST_INV_SCB_SITE,
+			   (u16)last_scb_site_no+1);
+	asd_write_reg_word(asd_ha, CSEQ_FIRST_INV_DDB_SITE,
+			   (u16)asd_ha->hw_prof.max_ddbs);
+
+	/* CSEQ Mode dependent, mode 8, page 1 setup. */
+	asd_write_reg_dword(asd_ha, CSEQ_LUN_TO_CLEAR, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_LUN_TO_CLEAR + 4, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_LUN_TO_CHECK, 0);
+	asd_write_reg_dword(asd_ha, CSEQ_LUN_TO_CHECK + 4, 0);
+
+	/* CSEQ Mode dependent, mode 8, page 2 setup. */
+	/* Tell the sequencer the bus address of the first SCB. */
+	asd_write_reg_addr(asd_ha, CSEQ_HQ_NEW_POINTER,
+			   asd_ha->seq.next_scb.dma_handle);
+	ASD_DPRINTK("First SCB dma_handle: 0x%llx\n",
+		    (unsigned long long)asd_ha->seq.next_scb.dma_handle);
+
+	/* Tell the sequencer the first Done List entry address. */
+	asd_write_reg_addr(asd_ha, CSEQ_HQ_DONE_BASE,
+			   asd_ha->seq.actual_dl->dma_handle);
+
+	/* Initialize the Q_DONE_POINTER with the least significant
+	 * 4 bytes of the first Done List address. */
+	asd_write_reg_dword(asd_ha, CSEQ_HQ_DONE_POINTER,
+			    ASD_BUSADDR_LO(asd_ha->seq.actual_dl->dma_handle));
+
+	asd_write_reg_byte(asd_ha, CSEQ_HQ_DONE_PASS, ASD_DEF_DL_TOGGLE);
+
+	/* CSEQ Mode dependent, mode 8, page 3 shall be ignored. */
+}
+
+/**
+ * asd_init_cseq_scratch -- setup and init CSEQ
+ * @asd_ha: pointer to host adapter structure
+ *
+ * Setup and initialize Central sequencers. Initialiaze the mode
+ * independent and dependent scratch page to the default settings.
+ */
+static void asd_init_cseq_scratch(struct asd_ha_struct *asd_ha)
+{
+	asd_init_cseq_mip(asd_ha);
+	asd_init_cseq_mdp(asd_ha);
+}
+
+/**
+ * asd_init_lseq_mip -- initialize LSEQ Mode independent pages 0-3
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_init_lseq_mip(struct asd_ha_struct *asd_ha, u8 lseq)
+{
+	int i;
+
+	/* LSEQ Mode independent page 0 setup. */
+	asd_write_reg_word(asd_ha, LmSEQ_Q_TGTXFR_HEAD(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_Q_TGTXFR_TAIL(lseq), 0xFFFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_LINK_NUMBER(lseq), lseq);
+	asd_write_reg_byte(asd_ha, LmSEQ_SCRATCH_FLAGS(lseq),
+			   ASD_NOTIFY_ENABLE_SPINUP);
+	asd_write_reg_dword(asd_ha, LmSEQ_CONNECTION_STATE(lseq),0x08000000);
+	asd_write_reg_word(asd_ha, LmSEQ_CONCTL(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_CONSTAT(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_CONNECTION_MODES(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_REG1_ISR(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_REG2_ISR(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_REG3_ISR(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_REG0_ISR(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_REG0_ISR(lseq)+4, 0);
+
+	/* LSEQ Mode independent page 1 setup. */
+	asd_write_reg_word(asd_ha, LmSEQ_EST_NEXUS_SCBPTR0(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_EST_NEXUS_SCBPTR1(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_EST_NEXUS_SCBPTR2(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_EST_NEXUS_SCBPTR3(lseq), 0xFFFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_SCB_OPCODE0(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_SCB_OPCODE1(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_SCB_OPCODE2(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_SCB_OPCODE3(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_SCB_HEAD(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_SCB_TAIL(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EST_NEXUS_BUF_AVAIL(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_TIMEOUT_CONST(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_ISR_SAVE_SINDEX(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_ISR_SAVE_DINDEX(lseq), 0);
+
+	/* LSEQ Mode Independent page 2 setup. */
+	asd_write_reg_word(asd_ha, LmSEQ_EMPTY_SCB_PTR0(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_EMPTY_SCB_PTR1(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_EMPTY_SCB_PTR2(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_EMPTY_SCB_PTR3(lseq), 0xFFFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_SCB_OPCD0(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_SCB_OPCD1(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_SCB_OPCD2(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_SCB_OPCD3(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_SCB_HEAD(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_SCB_TAIL(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_EMPTY_BUFS_AVAIL(lseq), 0);
+	for (i = 0; i < 12; i += 4)
+		asd_write_reg_dword(asd_ha, LmSEQ_ATA_SCR_REGS(lseq) + i, 0);
+
+	/* LSEQ Mode Independent page 3 setup. */
+
+	/* Device present timer timeout */
+	asd_write_reg_dword(asd_ha, LmSEQ_DEV_PRES_TMR_TOUT_CONST(lseq),
+			    ASD_DEV_PRESENT_TIMEOUT);
+
+	/* SATA interlock timer disabled */
+	asd_write_reg_dword(asd_ha, LmSEQ_SATA_INTERLOCK_TIMEOUT(lseq),
+			    ASD_SATA_INTERLOCK_TIMEOUT);
+
+	/* STP shutdown timer timeout constant, IGNORED by the sequencer,
+	 * always 0. */
+	asd_write_reg_dword(asd_ha, LmSEQ_STP_SHUTDOWN_TIMEOUT(lseq),
+			    ASD_STP_SHUTDOWN_TIMEOUT);
+
+	asd_write_reg_dword(asd_ha, LmSEQ_SRST_ASSERT_TIMEOUT(lseq),
+			    ASD_SRST_ASSERT_TIMEOUT);
+
+	asd_write_reg_dword(asd_ha, LmSEQ_RCV_FIS_TIMEOUT(lseq),
+			    ASD_RCV_FIS_TIMEOUT);
+
+	asd_write_reg_dword(asd_ha, LmSEQ_ONE_MILLISEC_TIMEOUT(lseq),
+			    ASD_ONE_MILLISEC_TIMEOUT);
+
+	/* COM_INIT timer */
+	asd_write_reg_dword(asd_ha, LmSEQ_TEN_MS_COMINIT_TIMEOUT(lseq),
+			    ASD_TEN_MILLISEC_TIMEOUT);
+
+	asd_write_reg_dword(asd_ha, LmSEQ_SMP_RCV_TIMEOUT(lseq),
+			    ASD_SMP_RCV_TIMEOUT);
+}
+
+/**
+ * asd_init_lseq_mdp -- initialize LSEQ mode dependent pages.
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_init_lseq_mdp(struct asd_ha_struct *asd_ha,  int lseq)
+{
+	int    i;
+	u32    moffs;
+	u16 ret_addr[] = {
+		0xFFFF,		  /* mode 0 */
+		0xFFFF,		  /* mode 1 */
+		mode2_task,	  /* mode 2 */
+		0,
+		0xFFFF,		  /* mode 4/5 */
+		0xFFFF,		  /* mode 4/5 */
+	};
+
+	/*
+	 * Mode 0,1,2 and 4/5 have common field on page 0 for the first
+	 * 14 bytes.
+	 */
+	for (i = 0; i < 3; i++) {
+		moffs = i * LSEQ_MODE_SCRATCH_SIZE;
+		asd_write_reg_word(asd_ha, LmSEQ_RET_ADDR(lseq)+moffs,
+				   ret_addr[i]);
+		asd_write_reg_word(asd_ha, LmSEQ_REG0_MODE(lseq)+moffs, 0);
+		asd_write_reg_word(asd_ha, LmSEQ_MODE_FLAGS(lseq)+moffs, 0);
+		asd_write_reg_word(asd_ha, LmSEQ_RET_ADDR2(lseq)+moffs,0xFFFF);
+		asd_write_reg_word(asd_ha, LmSEQ_RET_ADDR1(lseq)+moffs,0xFFFF);
+		asd_write_reg_byte(asd_ha, LmSEQ_OPCODE_TO_CSEQ(lseq)+moffs,0);
+		asd_write_reg_word(asd_ha, LmSEQ_DATA_TO_CSEQ(lseq)+moffs,0);
+	}
+	/*
+	 *  Mode 5 page 0 overlaps the same scratch page with Mode 0 page 3.
+	 */
+	asd_write_reg_word(asd_ha,
+			 LmSEQ_RET_ADDR(lseq)+LSEQ_MODE5_PAGE0_OFFSET,
+			   ret_addr[5]);
+	asd_write_reg_word(asd_ha,
+			 LmSEQ_REG0_MODE(lseq)+LSEQ_MODE5_PAGE0_OFFSET,0);
+	asd_write_reg_word(asd_ha,
+			 LmSEQ_MODE_FLAGS(lseq)+LSEQ_MODE5_PAGE0_OFFSET, 0);
+	asd_write_reg_word(asd_ha,
+			 LmSEQ_RET_ADDR2(lseq)+LSEQ_MODE5_PAGE0_OFFSET,0xFFFF);
+	asd_write_reg_word(asd_ha,
+			 LmSEQ_RET_ADDR1(lseq)+LSEQ_MODE5_PAGE0_OFFSET,0xFFFF);
+	asd_write_reg_byte(asd_ha,
+		         LmSEQ_OPCODE_TO_CSEQ(lseq)+LSEQ_MODE5_PAGE0_OFFSET,0);
+	asd_write_reg_word(asd_ha,
+		         LmSEQ_DATA_TO_CSEQ(lseq)+LSEQ_MODE5_PAGE0_OFFSET, 0);
+
+	/* LSEQ Mode dependent 0, page 0 setup. */
+	asd_write_reg_word(asd_ha, LmSEQ_FIRST_INV_DDB_SITE(lseq),
+			   (u16)asd_ha->hw_prof.max_ddbs);
+	asd_write_reg_word(asd_ha, LmSEQ_EMPTY_TRANS_CTX(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_RESP_LEN(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_FIRST_INV_SCB_SITE(lseq),
+			   (u16)last_scb_site_no+1);
+	asd_write_reg_word(asd_ha, LmSEQ_INTEN_SAVE(lseq),
+			    (u16) LmM0INTEN_MASK & 0xFFFF0000 >> 16);
+	asd_write_reg_word(asd_ha, LmSEQ_INTEN_SAVE(lseq) + 2,
+			    (u16) LmM0INTEN_MASK & 0xFFFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_LINK_RST_FRM_LEN(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_LINK_RST_PROTOCOL(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_RESP_STATUS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_LAST_LOADED_SGE(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_SAVE_SCBPTR(lseq), 0);
+
+	/* LSEQ mode dependent, mode 1, page 0 setup. */
+	asd_write_reg_word(asd_ha, LmSEQ_Q_XMIT_HEAD(lseq), 0xFFFF);
+	asd_write_reg_word(asd_ha, LmSEQ_M1_EMPTY_TRANS_CTX(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_INI_CONN_TAG(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_FAILED_OPEN_STATUS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_XMIT_REQUEST_TYPE(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_M1_RESP_STATUS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_M1_LAST_LOADED_SGE(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_M1_SAVE_SCBPTR(lseq), 0);
+
+	/* LSEQ Mode dependent mode 2, page 0 setup */
+	asd_write_reg_word(asd_ha, LmSEQ_PORT_COUNTER(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_PM_TABLE_PTR(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_SATA_INTERLOCK_TMR_SAVE(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_IP_BITL(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_COPY_SMP_CONN_TAG(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_P0M2_OFFS1AH(lseq), 0);
+
+	/* LSEQ Mode dependent, mode 4/5, page 0 setup. */
+	asd_write_reg_byte(asd_ha, LmSEQ_SAVED_OOB_STATUS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_SAVED_OOB_MODE(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_Q_LINK_HEAD(lseq), 0xFFFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_LINK_RST_ERR(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_SAVED_OOB_SIGNALS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_SAS_RESET_MODE(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_LINK_RESET_RETRY_COUNT(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_NUM_LINK_RESET_RETRIES(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_OOB_INT_ENABLES(lseq), 0);
+	/*
+	 * Set the desired interval between transmissions of the NOTIFY
+	 * (ENABLE SPINUP) primitive.  Must be initilized to val - 1.
+	 */
+	asd_write_reg_word(asd_ha, LmSEQ_NOTIFY_TIMER_TIMEOUT(lseq),
+			   ASD_NOTIFY_TIMEOUT - 1);
+	/* No delay for the first NOTIFY to be sent to the attached target. */
+	asd_write_reg_word(asd_ha, LmSEQ_NOTIFY_TIMER_DOWN_COUNT(lseq),
+			   ASD_NOTIFY_DOWN_COUNT);
+
+	/* LSEQ Mode dependent, mode 0 and 1, page 1 setup. */
+	for (i = 0; i < 2; i++)	{
+		int j;
+		/* Start from Page 1 of Mode 0 and 1. */
+		moffs = LSEQ_PAGE_SIZE + i*LSEQ_MODE_SCRATCH_SIZE;
+		/* All the fields of page 1 can be intialized to 0. */
+		for (j = 0; j < LSEQ_PAGE_SIZE; j += 4)
+			asd_write_reg_dword(asd_ha, LmSCRATCH(lseq)+moffs+j,0);
+	}
+
+	/* LSEQ Mode dependent, mode 2, page 1 setup. */
+	asd_write_reg_dword(asd_ha, LmSEQ_INVALID_DWORD_COUNT(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_DISPARITY_ERROR_COUNT(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_LOSS_OF_SYNC_COUNT(lseq), 0);
+
+	/* LSEQ Mode dependent, mode 4/5, page 1. */
+	for (i = 0; i < LSEQ_PAGE_SIZE; i+=4)
+		asd_write_reg_dword(asd_ha, LmSEQ_FRAME_TYPE_MASK(lseq)+i, 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_FRAME_TYPE_MASK(lseq), 0xFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_HASHED_DEST_ADDR_MASK(lseq), 0xFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_HASHED_DEST_ADDR_MASK(lseq)+1,0xFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_HASHED_DEST_ADDR_MASK(lseq)+2,0xFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_HASHED_SRC_ADDR_MASK(lseq), 0xFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_HASHED_SRC_ADDR_MASK(lseq)+1, 0xFF);
+	asd_write_reg_byte(asd_ha, LmSEQ_HASHED_SRC_ADDR_MASK(lseq)+2, 0xFF);
+	asd_write_reg_dword(asd_ha, LmSEQ_DATA_OFFSET(lseq), 0xFFFFFFFF);
+
+	/* LSEQ Mode dependent, mode 0, page 2 setup. */
+	asd_write_reg_dword(asd_ha, LmSEQ_SMP_RCV_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_DEVICE_BITS(lseq), 0);
+	asd_write_reg_word(asd_ha, LmSEQ_SDB_DDB(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_SDB_NUM_TAGS(lseq), 0);
+	asd_write_reg_byte(asd_ha, LmSEQ_SDB_CURR_TAG(lseq), 0);
+
+	/* LSEQ Mode Dependent 1, page 2 setup. */
+	asd_write_reg_dword(asd_ha, LmSEQ_TX_ID_ADDR_FRAME(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_TX_ID_ADDR_FRAME(lseq)+4, 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_OPEN_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_SRST_AS_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_LAST_LOADED_SG_EL(lseq), 0);
+
+	/* LSEQ Mode Dependent 2, page 2 setup. */
+	/* The LmSEQ_STP_SHUTDOWN_TIMER_TERM_TS is IGNORED by the sequencer,
+	 * i.e. always 0. */
+	asd_write_reg_dword(asd_ha, LmSEQ_STP_SHUTDOWN_TIMER_TERM_TS(lseq),0);
+	asd_write_reg_dword(asd_ha, LmSEQ_CLOSE_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_BREAK_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_DWS_RESET_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha,LmSEQ_SATA_INTERLOCK_TIMER_TERM_TS(lseq),0);
+	asd_write_reg_dword(asd_ha, LmSEQ_MCTL_TIMER_TERM_TS(lseq), 0);
+
+	/* LSEQ Mode Dependent 4/5, page 2 setup. */
+	asd_write_reg_dword(asd_ha, LmSEQ_COMINIT_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_RCV_ID_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_RCV_FIS_TIMER_TERM_TS(lseq), 0);
+	asd_write_reg_dword(asd_ha, LmSEQ_DEV_PRES_TIMER_TERM_TS(lseq),	0);
+}
+
+/**
+ * asd_init_lseq_scratch -- setup and init link sequencers
+ * @asd_ha: pointer to host adapter struct
+ */
+static void asd_init_lseq_scratch(struct asd_ha_struct *asd_ha)
+{
+	u8 lseq;
+	u8 lseq_mask;
+
+	lseq_mask = asd_ha->hw_prof.enabled_phys;
+	for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+		asd_init_lseq_mip(asd_ha, lseq);
+		asd_init_lseq_mdp(asd_ha, lseq);
+	}
+}
+
+/**
+ * asd_init_scb_sites -- initialize sequencer SCB sites (memory).
+ * @asd_ha: pointer to host adapter structure
+ *
+ * This should be done before initializing common CSEQ and LSEQ
+ * scratch since those areas depend on some computed values here,
+ * last_scb_site_no, etc.
+ */
+static void asd_init_scb_sites(struct asd_ha_struct *asd_ha)
+{
+	u16	site_no;
+	u16     max_scbs = 0;
+
+	for (site_no = asd_ha->hw_prof.max_scbs-1;
+	     site_no != (u16) -1;
+	     site_no--) {
+		u16	i;
+
+		/* Initialize all fields in the SCB site to 0. */
+		for (i = 0; i < ASD_SCB_SIZE; i += 4)
+			asd_scbsite_write_dword(asd_ha, site_no, i, 0);
+
+		/* Workaround needed by SEQ to fix a SATA issue is to exclude
+		 * certain SCB sites from the free list. */
+		if (!SCB_SITE_VALID(site_no))
+			continue;
+
+		if (last_scb_site_no == 0)
+			last_scb_site_no = site_no;
+
+		/* For every SCB site, we need to initialize the
+		 * following fields: Q_NEXT, SCB_OPCODE, SCB_FLAGS,
+		 * and SG Element Flag. */
+
+		/* Q_NEXT field of the last SCB is invalidated. */
+		asd_scbsite_write_word(asd_ha, site_no, 0, first_scb_site_no);
+
+		/* Initialize SCB Site Opcode field to invalid. */
+		asd_scbsite_write_byte(asd_ha, site_no,
+				       offsetof(struct scb_header, opcode),
+				       0xFF);
+
+		/* Initialize SCB Site Flags field to mean a response
+		 * frame has been received.  This means inadvertent
+		 * frames received to be dropped. */
+		asd_scbsite_write_byte(asd_ha, site_no, 0x49, 0x01);
+
+		first_scb_site_no = site_no;
+		max_scbs++;
+	}
+	asd_ha->hw_prof.max_scbs = max_scbs;
+	ASD_DPRINTK("max_scbs:%d\n", asd_ha->hw_prof.max_scbs);
+	ASD_DPRINTK("first_scb_site_no:0x%x\n", first_scb_site_no);
+	ASD_DPRINTK("last_scb_site_no:0x%x\n", last_scb_site_no);
+}
+
+/**
+ * asd_init_cseq_cio - initialize CSEQ CIO registers
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_init_cseq_cio(struct asd_ha_struct *asd_ha)
+{
+	int i;
+
+	asd_write_reg_byte(asd_ha, CSEQCOMINTEN, 0);
+	asd_write_reg_byte(asd_ha, CSEQDLCTL, ASD_DL_SIZE_BITS);
+	asd_write_reg_byte(asd_ha, CSEQDLOFFS, 0);
+	asd_write_reg_byte(asd_ha, CSEQDLOFFS+1, 0);
+	asd_ha->seq.scbpro = 0;
+	asd_write_reg_dword(asd_ha, SCBPRO, 0);
+	asd_write_reg_dword(asd_ha, CSEQCON, 0);
+
+	/* Intialize CSEQ Mode 11 Interrupt Vectors.
+	 * The addresses are 16 bit wide and in dword units.
+	 * The values of their macros are in byte units.
+	 * Thus we have to divide by 4. */
+	asd_write_reg_word(asd_ha, CM11INTVEC0, cseq_vecs[0]);
+	asd_write_reg_word(asd_ha, CM11INTVEC1, cseq_vecs[1]);
+	asd_write_reg_word(asd_ha, CM11INTVEC2, cseq_vecs[2]);
+
+	/* Enable ARP2HALTC (ARP2 Halted from Halt Code Write). */
+	asd_write_reg_byte(asd_ha, CARP2INTEN, EN_ARP2HALTC);
+
+	/* Initialize CSEQ Scratch Page to 0x04. */
+	asd_write_reg_byte(asd_ha, CSCRATCHPAGE, 0x04);
+
+	/* Initialize CSEQ Mode[0-8] Dependent registers. */
+	/* Initialize Scratch Page to 0. */
+	for (i = 0; i < 9; i++)
+		asd_write_reg_byte(asd_ha, CMnSCRATCHPAGE(i), 0);
+
+	/* Reset the ARP2 Program Count. */
+	asd_write_reg_word(asd_ha, CPRGMCNT, cseq_idle_loop);
+
+	for (i = 0; i < 8; i++) {
+		/* Intialize Mode n Link m Interrupt Enable. */
+		asd_write_reg_dword(asd_ha, CMnINTEN(i), EN_CMnRSPMBXF);
+		/* Initialize Mode n Request Mailbox. */
+		asd_write_reg_dword(asd_ha, CMnREQMBX(i), 0);
+	}
+}
+
+/**
+ * asd_init_lseq_cio -- initialize LmSEQ CIO registers
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_init_lseq_cio(struct asd_ha_struct *asd_ha, int lseq)
+{
+	u8  *sas_addr;
+	int  i;
+
+	/* Enable ARP2HALTC (ARP2 Halted from Halt Code Write). */
+	asd_write_reg_dword(asd_ha, LmARP2INTEN(lseq), EN_ARP2HALTC);
+
+	asd_write_reg_byte(asd_ha, LmSCRATCHPAGE(lseq), 0);
+
+	/* Initialize Mode 0,1, and 2 SCRATCHPAGE to 0. */
+	for (i = 0; i < 3; i++)
+		asd_write_reg_byte(asd_ha, LmMnSCRATCHPAGE(lseq, i), 0);
+
+	/* Initialize Mode 5 SCRATCHPAGE to 0. */
+	asd_write_reg_byte(asd_ha, LmMnSCRATCHPAGE(lseq, 5), 0);
+
+	asd_write_reg_dword(asd_ha, LmRSPMBX(lseq), 0);
+	/* Initialize Mode 0,1,2 and 5 Interrupt Enable and
+	 * Interrupt registers. */
+	asd_write_reg_dword(asd_ha, LmMnINTEN(lseq, 0), LmM0INTEN_MASK);
+	asd_write_reg_dword(asd_ha, LmMnINT(lseq, 0), 0xFFFFFFFF);
+	/* Mode 1 */
+	asd_write_reg_dword(asd_ha, LmMnINTEN(lseq, 1), LmM1INTEN_MASK);
+	asd_write_reg_dword(asd_ha, LmMnINT(lseq, 1), 0xFFFFFFFF);
+	/* Mode 2 */
+	asd_write_reg_dword(asd_ha, LmMnINTEN(lseq, 2), LmM2INTEN_MASK);
+	asd_write_reg_dword(asd_ha, LmMnINT(lseq, 2), 0xFFFFFFFF);
+	/* Mode 5 */
+	asd_write_reg_dword(asd_ha, LmMnINTEN(lseq, 5), LmM5INTEN_MASK);
+	asd_write_reg_dword(asd_ha, LmMnINT(lseq, 5), 0xFFFFFFFF);
+
+	/* Enable HW Timer status. */
+	asd_write_reg_byte(asd_ha, LmHWTSTATEN(lseq), LmHWTSTATEN_MASK);
+
+	/* Enable Primitive Status 0 and 1. */
+	asd_write_reg_dword(asd_ha, LmPRIMSTAT0EN(lseq), LmPRIMSTAT0EN_MASK);
+	asd_write_reg_dword(asd_ha, LmPRIMSTAT1EN(lseq), LmPRIMSTAT1EN_MASK);
+
+	/* Enable Frame Error. */
+	asd_write_reg_dword(asd_ha, LmFRMERREN(lseq), LmFRMERREN_MASK);
+	asd_write_reg_byte(asd_ha, LmMnHOLDLVL(lseq, 0), 0x50);
+
+	/* Initialize Mode 0 Transfer Level to 512. */
+	asd_write_reg_byte(asd_ha,  LmMnXFRLVL(lseq, 0), LmMnXFRLVL_512);
+	/* Initialize Mode 1 Transfer Level to 256. */
+	asd_write_reg_byte(asd_ha, LmMnXFRLVL(lseq, 1), LmMnXFRLVL_256);
+
+	/* Initialize Program Count. */
+	asd_write_reg_word(asd_ha, LmPRGMCNT(lseq), lseq_idle_loop);
+
+	/* Enable Blind SG Move. */
+	asd_write_reg_dword(asd_ha, LmMODECTL(lseq), LmBLIND48);
+	asd_write_reg_word(asd_ha, LmM3SATATIMER(lseq),
+			   ASD_SATA_INTERLOCK_TIMEOUT);
+
+	(void) asd_read_reg_dword(asd_ha, LmREQMBX(lseq));
+
+	/* Clear Primitive Status 0 and 1. */
+	asd_write_reg_dword(asd_ha, LmPRMSTAT0(lseq), 0xFFFFFFFF);
+	asd_write_reg_dword(asd_ha, LmPRMSTAT1(lseq), 0xFFFFFFFF);
+
+	/* Clear HW Timer status. */
+	asd_write_reg_byte(asd_ha, LmHWTSTAT(lseq), 0xFF);
+
+	/* Clear DMA Errors for Mode 0 and 1. */
+	asd_write_reg_byte(asd_ha, LmMnDMAERRS(lseq, 0), 0xFF);
+	asd_write_reg_byte(asd_ha, LmMnDMAERRS(lseq, 1), 0xFF);
+
+	/* Clear SG DMA Errors for Mode 0 and 1. */
+	asd_write_reg_byte(asd_ha, LmMnSGDMAERRS(lseq, 0), 0xFF);
+	asd_write_reg_byte(asd_ha, LmMnSGDMAERRS(lseq, 1), 0xFF);
+
+	/* Clear Mode 0 Buffer Parity Error. */
+	asd_write_reg_byte(asd_ha, LmMnBUFSTAT(lseq, 0), LmMnBUFPERR);
+
+	/* Clear Mode 0 Frame Error register. */
+	asd_write_reg_dword(asd_ha, LmMnFRMERR(lseq, 0), 0xFFFFFFFF);
+
+	/* Reset LSEQ external interrupt arbiter. */
+	asd_write_reg_byte(asd_ha, LmARP2INTCTL(lseq), RSTINTCTL);
+
+	/* Set the Phy SAS for the LmSEQ WWN. */
+	sas_addr = asd_ha->phys[lseq].phy_desc->sas_addr;
+	for (i = 0; i < SAS_ADDR_SIZE; i++)
+		asd_write_reg_byte(asd_ha, LmWWN(lseq) + i, sas_addr[i]);
+
+	/* Set the Transmit Size to 1024 bytes, 0 = 256 Dwords. */
+	asd_write_reg_byte(asd_ha, LmMnXMTSIZE(lseq, 1), 0);
+
+	/* Set the Bus Inactivity Time Limit Timer. */
+	asd_write_reg_word(asd_ha, LmBITL_TIMER(lseq), 9);
+
+	/* Enable SATA Port Multiplier. */
+	asd_write_reg_byte(asd_ha, LmMnSATAFS(lseq, 1), 0x80);
+
+	/* Initialize Interrupt Vector[0-10] address in Mode 3.
+	 * See the comment on CSEQ_INT_* */
+	asd_write_reg_word(asd_ha, LmM3INTVEC0(lseq), lseq_vecs[0]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC1(lseq), lseq_vecs[1]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC2(lseq), lseq_vecs[2]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC3(lseq), lseq_vecs[3]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC4(lseq), lseq_vecs[4]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC5(lseq), lseq_vecs[5]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC6(lseq), lseq_vecs[6]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC7(lseq), lseq_vecs[7]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC8(lseq), lseq_vecs[8]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC9(lseq), lseq_vecs[9]);
+	asd_write_reg_word(asd_ha, LmM3INTVEC10(lseq), lseq_vecs[10]);
+	/*
+	 * Program the Link LED control, applicable only for
+	 * Chip Rev. B or later.
+	 */
+	asd_write_reg_dword(asd_ha, LmCONTROL(lseq),
+			    (LEDTIMER | LEDMODE_TXRX | LEDTIMERS_100ms));
+
+	/* Set the Align Rate for SAS and STP mode. */
+	asd_write_reg_byte(asd_ha, LmM1SASALIGN(lseq), SAS_ALIGN_DEFAULT);
+	asd_write_reg_byte(asd_ha, LmM1STPALIGN(lseq), STP_ALIGN_DEFAULT);
+}
+
+
+/**
+ * asd_post_init_cseq -- clear CSEQ Mode n Int. status and Response mailbox
+ * @asd_ha: pointer to host adapter struct
+ */
+static void asd_post_init_cseq(struct asd_ha_struct *asd_ha)
+{
+	int i;
+
+	for (i = 0; i < 8; i++)
+		asd_write_reg_dword(asd_ha, CMnINT(i), 0xFFFFFFFF);
+	for (i = 0; i < 8; i++)
+		asd_read_reg_dword(asd_ha, CMnRSPMBX(i));
+	/* Reset the external interrupt arbiter. */
+	asd_write_reg_byte(asd_ha, CARP2INTCTL, RSTINTCTL);
+}
+
+/**
+ * asd_init_ddb_0 -- initialize DDB 0
+ * @asd_ha: pointer to host adapter structure
+ *
+ * Initialize DDB site 0 which is used internally by the sequencer.
+ */
+static void asd_init_ddb_0(struct asd_ha_struct *asd_ha)
+{
+	int	i;
+
+	/* Zero out the DDB explicitly */
+	for (i = 0; i < sizeof(struct asd_ddb_seq_shared); i+=4)
+		asd_ddbsite_write_dword(asd_ha, 0, i, 0);
+
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, q_free_ddb_head), 0);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, q_free_ddb_tail),
+			       asd_ha->hw_prof.max_ddbs-1);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, q_free_ddb_cnt), 0);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, q_used_ddb_head), 0xFFFF);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, q_used_ddb_tail), 0xFFFF);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, shared_mem_lock), 0);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, smp_conn_tag), 0);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, est_nexus_buf_cnt), 0);
+	asd_ddbsite_write_word(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, est_nexus_buf_thresh),
+			       asd_ha->hw_prof.num_phys * 2);
+	asd_ddbsite_write_byte(asd_ha, 0,
+		 offsetof(struct asd_ddb_seq_shared, settable_max_contexts),0);
+	asd_ddbsite_write_byte(asd_ha, 0,
+	       offsetof(struct asd_ddb_seq_shared, conn_not_active), 0xFF);
+	asd_ddbsite_write_byte(asd_ha, 0,
+	       offsetof(struct asd_ddb_seq_shared, phy_is_up), 0x00);
+	/* DDB 0 is reserved */
+	set_bit(0, asd_ha->hw_prof.ddb_bitmap);
+}
+
+/**
+ * asd_seq_setup_seqs -- setup and initialize central and link sequencers
+ * @asd_ha: pointer to host adapter structure
+ */
+static void asd_seq_setup_seqs(struct asd_ha_struct *asd_ha)
+{
+	int 		lseq;
+	u8		lseq_mask;
+
+	/* Initialize SCB sites. Done first to compute some values which
+	 * the rest of the init code depends on. */
+	asd_init_scb_sites(asd_ha);
+
+	/* Initialize CSEQ Scratch RAM registers. */
+	asd_init_cseq_scratch(asd_ha);
+
+	/* Initialize LmSEQ Scratch RAM registers. */
+	asd_init_lseq_scratch(asd_ha);
+
+	/* Initialize CSEQ CIO registers. */
+	asd_init_cseq_cio(asd_ha);
+
+	asd_init_ddb_0(asd_ha);
+
+	/* Initialize LmSEQ CIO registers. */
+	lseq_mask = asd_ha->hw_prof.enabled_phys;
+	for_each_sequencer(lseq_mask, lseq_mask, lseq)
+		asd_init_lseq_cio(asd_ha, lseq);
+	asd_post_init_cseq(asd_ha);
+}
+
+
+/**
+ * asd_seq_start_cseq -- start the central sequencer, CSEQ
+ * @asd_ha: pointer to host adapter structure
+ */
+static int asd_seq_start_cseq(struct asd_ha_struct *asd_ha)
+{
+	/* Reset the ARP2 instruction to location zero. */
+	asd_write_reg_word(asd_ha, CPRGMCNT, cseq_idle_loop);
+
+	/* Unpause the CSEQ  */
+	return asd_unpause_cseq(asd_ha);
+}
+
+/**
+ * asd_seq_start_lseq -- start a link sequencer
+ * @asd_ha: pointer to host adapter structure
+ * @lseq: the link sequencer of interest
+ */
+static int asd_seq_start_lseq(struct asd_ha_struct *asd_ha, int lseq)
+{
+	/* Reset the ARP2 instruction to location zero. */
+	asd_write_reg_word(asd_ha, LmPRGMCNT(lseq), lseq_idle_loop);
+
+	/* Unpause the LmSEQ  */
+	return asd_seq_unpause_lseq(asd_ha, lseq);
+}
+
+static int asd_request_firmware(struct asd_ha_struct *asd_ha)
+{
+	int err, i;
+	struct sequencer_file_header header, *hdr_ptr;
+	u32 csum = 0;
+	u16 *ptr_cseq_vecs, *ptr_lseq_vecs;
+
+	if (sequencer_fw)
+		/* already loaded */
+		return 0;
+
+	err = request_firmware(&sequencer_fw,
+			       SAS_RAZOR_SEQUENCER_FW_FILE,
+			       &asd_ha->pcidev->dev);
+	if (err)
+		return err;
+
+	hdr_ptr = (struct sequencer_file_header *)sequencer_fw->data;
+
+	header.csum = le32_to_cpu(hdr_ptr->csum);
+	header.major = le32_to_cpu(hdr_ptr->major);
+	header.minor = le32_to_cpu(hdr_ptr->minor);
+	sequencer_version = hdr_ptr->version;
+	header.cseq_table_offset = le32_to_cpu(hdr_ptr->cseq_table_offset);
+	header.cseq_table_size = le32_to_cpu(hdr_ptr->cseq_table_size);
+	header.lseq_table_offset = le32_to_cpu(hdr_ptr->lseq_table_offset);
+	header.lseq_table_size = le32_to_cpu(hdr_ptr->lseq_table_size);
+	header.cseq_code_offset = le32_to_cpu(hdr_ptr->cseq_code_offset);
+	header.cseq_code_size = le32_to_cpu(hdr_ptr->cseq_code_size);
+	header.lseq_code_offset = le32_to_cpu(hdr_ptr->lseq_code_offset);
+	header.lseq_code_size = le32_to_cpu(hdr_ptr->lseq_code_size);
+	header.mode2_task = le16_to_cpu(hdr_ptr->mode2_task);
+	header.cseq_idle_loop = le16_to_cpu(hdr_ptr->cseq_idle_loop);
+	header.lseq_idle_loop = le16_to_cpu(hdr_ptr->lseq_idle_loop);
+
+	for (i = sizeof(header.csum); i < sequencer_fw->size; i++)
+		csum += sequencer_fw->data[i];
+
+	if (csum != header.csum) {
+		asd_printk("Firmware file checksum mismatch\n");
+		return -EINVAL;
+	}
+
+	if (header.cseq_table_size != CSEQ_NUM_VECS ||
+	    header.lseq_table_size != LSEQ_NUM_VECS) {
+		asd_printk("Firmware file table size mismatch\n");
+		return -EINVAL;
+	}
+
+	ptr_cseq_vecs = (u16 *)&sequencer_fw->data[header.cseq_table_offset];
+	ptr_lseq_vecs = (u16 *)&sequencer_fw->data[header.lseq_table_offset];
+	mode2_task = header.mode2_task;
+	cseq_idle_loop = header.cseq_idle_loop;
+	lseq_idle_loop = header.lseq_idle_loop;
+
+	for (i = 0; i < CSEQ_NUM_VECS; i++)
+		cseq_vecs[i] = le16_to_cpu(ptr_cseq_vecs[i]);
+
+	for (i = 0; i < LSEQ_NUM_VECS; i++)
+		lseq_vecs[i] = le16_to_cpu(ptr_lseq_vecs[i]);
+
+	cseq_code = &sequencer_fw->data[header.cseq_code_offset];
+	cseq_code_size = header.cseq_code_size;
+	lseq_code = &sequencer_fw->data[header.lseq_code_offset];
+	lseq_code_size = header.lseq_code_size;
+
+	return 0;
+}
+
+int asd_init_seqs(struct asd_ha_struct *asd_ha)
+{
+	int err;
+
+	err = asd_request_firmware(asd_ha);
+
+	if (err) {
+		asd_printk("Failed to load sequencer firmware file %s, error %d\n",
+			   SAS_RAZOR_SEQUENCER_FW_FILE, err);
+		return err;
+	}
+
+	asd_printk("using sequencer %s\n", sequencer_version);
+	err = asd_seq_download_seqs(asd_ha);
+	if (err) {
+		asd_printk("couldn't download sequencers for %s\n",
+			   pci_name(asd_ha->pcidev));
+		return err;
+	}
+
+	asd_seq_setup_seqs(asd_ha);
+
+	return 0;
+}
+
+int asd_start_seqs(struct asd_ha_struct *asd_ha)
+{
+	int err;
+	u8  lseq_mask;
+	int lseq;
+
+	err = asd_seq_start_cseq(asd_ha);
+	if (err) {
+		asd_printk("couldn't start CSEQ for %s\n",
+			   pci_name(asd_ha->pcidev));
+		return err;
+	}
+
+	lseq_mask = asd_ha->hw_prof.enabled_phys;
+	for_each_sequencer(lseq_mask, lseq_mask, lseq) {
+		err = asd_seq_start_lseq(asd_ha, lseq);
+		if (err) {
+			asd_printk("coudln't start LSEQ %d for %s\n", lseq,
+				   pci_name(asd_ha->pcidev));
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * asd_update_port_links -- update port_map_by_links and phy_is_up
+ * @sas_phy: pointer to the phy which has been added to a port
+ *
+ * 1) When a link reset has completed and we got BYTES DMAED with a
+ * valid frame we call this function for that phy, to indicate that
+ * the phy is up, i.e. we update the phy_is_up in DDB 0.  The
+ * sequencer checks phy_is_up when pending SCBs are to be sent, and
+ * when an open address frame has been received.
+ *
+ * 2) When we know of ports, we call this function to update the map
+ * of phys participaing in that port, i.e. we update the
+ * port_map_by_links in DDB 0.  When a HARD_RESET primitive has been
+ * received, the sequencer disables all phys in that port.
+ * port_map_by_links is also used as the conn_mask byte in the
+ * initiator/target port DDB.
+ */
+void asd_update_port_links(struct asd_sas_phy *sas_phy)
+{
+	struct asd_ha_struct *asd_ha = sas_phy->ha->lldd_ha;
+	const u8 phy_mask = (u8) sas_phy->port->phy_mask;
+	u8  phy_is_up;
+	u8  mask;
+	int i, err;
+
+	for_each_phy(phy_mask, mask, i)
+		asd_ddbsite_write_byte(asd_ha, 0,
+				       offsetof(struct asd_ddb_seq_shared,
+						port_map_by_links)+i,phy_mask);
+
+	for (i = 0; i < 12; i++) {
+		phy_is_up = asd_ddbsite_read_byte(asd_ha, 0,
+			  offsetof(struct asd_ddb_seq_shared, phy_is_up));
+		err = asd_ddbsite_update_byte(asd_ha, 0,
+				offsetof(struct asd_ddb_seq_shared, phy_is_up),
+				phy_is_up,
+				phy_is_up | phy_mask);
+		if (!err)
+			break;
+		else if (err == -EFAULT) {
+			asd_printk("phy_is_up: parity error in DDB 0\n");
+			break;
+		}
+	}
+
+	if (err)
+		asd_printk("couldn't update DDB 0:error:%d\n", err);
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_seq.h b/drivers/scsi/aic94xx/aic94xx_seq.h
new file mode 100644
index 000000000000..42281c36153b
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_seq.h
@@ -0,0 +1,70 @@
+/*
+ * Aic94xx SAS/SATA driver sequencer interface header file.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#ifndef _AIC94XX_SEQ_H_
+#define _AIC94XX_SEQ_H_
+
+#define CSEQ_NUM_VECS	3
+#define LSEQ_NUM_VECS	11
+
+#define SAS_RAZOR_SEQUENCER_FW_FILE "aic94xx-seq.fw"
+
+/* Note:  All quantites in the sequencer file are little endian */
+struct sequencer_file_header {
+	/* Checksum of the entire contents of the sequencer excluding
+	 * these four bytes */
+	u32	csum;
+	/* numeric major version */
+	u32	major;
+	/* numeric minor version */
+	u32	minor;
+	/* version string printed by driver */
+	char	version[16];
+	u32	cseq_table_offset;
+	u32	cseq_table_size;
+	u32	lseq_table_offset;
+	u32	lseq_table_size;
+	u32	cseq_code_offset;
+	u32	cseq_code_size;
+	u32	lseq_code_offset;
+	u32	lseq_code_size;
+	u16	mode2_task;
+	u16	cseq_idle_loop;
+	u16	lseq_idle_loop;
+} __attribute__((packed));
+
+#ifdef __KERNEL__
+int asd_pause_cseq(struct asd_ha_struct *asd_ha);
+int asd_unpause_cseq(struct asd_ha_struct *asd_ha);
+int asd_pause_lseq(struct asd_ha_struct *asd_ha, u8 lseq_mask);
+int asd_unpause_lseq(struct asd_ha_struct *asd_ha, u8 lseq_mask);
+int asd_init_seqs(struct asd_ha_struct *asd_ha);
+int asd_start_seqs(struct asd_ha_struct *asd_ha);
+
+void asd_update_port_links(struct asd_sas_phy *phy);
+#endif
+
+#endif
diff --git a/drivers/scsi/aic94xx/aic94xx_task.c b/drivers/scsi/aic94xx/aic94xx_task.c
new file mode 100644
index 000000000000..285e70dae933
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_task.c
@@ -0,0 +1,642 @@
+/*
+ * Aic94xx SAS/SATA Tasks
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/spinlock.h>
+#include "aic94xx.h"
+#include "aic94xx_sas.h"
+#include "aic94xx_hwi.h"
+
+static void asd_unbuild_ata_ascb(struct asd_ascb *a);
+static void asd_unbuild_smp_ascb(struct asd_ascb *a);
+static void asd_unbuild_ssp_ascb(struct asd_ascb *a);
+
+static inline void asd_can_dequeue(struct asd_ha_struct *asd_ha, int num)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&asd_ha->seq.pend_q_lock, flags);
+	asd_ha->seq.can_queue += num;
+	spin_unlock_irqrestore(&asd_ha->seq.pend_q_lock, flags);
+}
+
+/* PCI_DMA_... to our direction translation.
+ */
+static const u8 data_dir_flags[] = {
+	[PCI_DMA_BIDIRECTIONAL] = DATA_DIR_BYRECIPIENT,	/* UNSPECIFIED */
+	[PCI_DMA_TODEVICE]      = DATA_DIR_OUT, /* OUTBOUND */
+	[PCI_DMA_FROMDEVICE]    = DATA_DIR_IN, /* INBOUND */
+	[PCI_DMA_NONE]          = DATA_DIR_NONE, /* NO TRANSFER */
+};
+
+static inline int asd_map_scatterlist(struct sas_task *task,
+				      struct sg_el *sg_arr,
+				      unsigned long gfp_flags)
+{
+	struct asd_ascb *ascb = task->lldd_task;
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct scatterlist *sc;
+	int num_sg, res;
+
+	if (task->data_dir == PCI_DMA_NONE)
+		return 0;
+
+	if (task->num_scatter == 0) {
+		void *p = task->scatter;
+		dma_addr_t dma = pci_map_single(asd_ha->pcidev, p,
+						task->total_xfer_len,
+						task->data_dir);
+		sg_arr[0].bus_addr = cpu_to_le64((u64)dma);
+		sg_arr[0].size = cpu_to_le32(task->total_xfer_len);
+		sg_arr[0].flags |= ASD_SG_EL_LIST_EOL;
+		return 0;
+	}
+
+	num_sg = pci_map_sg(asd_ha->pcidev, task->scatter, task->num_scatter,
+			    task->data_dir);
+	if (num_sg == 0)
+		return -ENOMEM;
+
+	if (num_sg > 3) {
+		int i;
+
+		ascb->sg_arr = asd_alloc_coherent(asd_ha,
+						  num_sg*sizeof(struct sg_el),
+						  gfp_flags);
+		if (!ascb->sg_arr) {
+			res = -ENOMEM;
+			goto err_unmap;
+		}
+		for (sc = task->scatter, i = 0; i < num_sg; i++, sc++) {
+			struct sg_el *sg =
+				&((struct sg_el *)ascb->sg_arr->vaddr)[i];
+			sg->bus_addr = cpu_to_le64((u64)sg_dma_address(sc));
+			sg->size = cpu_to_le32((u32)sg_dma_len(sc));
+			if (i == num_sg-1)
+				sg->flags |= ASD_SG_EL_LIST_EOL;
+		}
+
+		for (sc = task->scatter, i = 0; i < 2; i++, sc++) {
+			sg_arr[i].bus_addr =
+				cpu_to_le64((u64)sg_dma_address(sc));
+			sg_arr[i].size = cpu_to_le32((u32)sg_dma_len(sc));
+		}
+		sg_arr[1].next_sg_offs = 2 * sizeof(*sg_arr);
+		sg_arr[1].flags |= ASD_SG_EL_LIST_EOS;
+
+		memset(&sg_arr[2], 0, sizeof(*sg_arr));
+		sg_arr[2].bus_addr=cpu_to_le64((u64)ascb->sg_arr->dma_handle);
+	} else {
+		int i;
+		for (sc = task->scatter, i = 0; i < num_sg; i++, sc++) {
+			sg_arr[i].bus_addr =
+				cpu_to_le64((u64)sg_dma_address(sc));
+			sg_arr[i].size = cpu_to_le32((u32)sg_dma_len(sc));
+		}
+		sg_arr[i-1].flags |= ASD_SG_EL_LIST_EOL;
+	}
+
+	return 0;
+err_unmap:
+	pci_unmap_sg(asd_ha->pcidev, task->scatter, task->num_scatter,
+		     task->data_dir);
+	return res;
+}
+
+static inline void asd_unmap_scatterlist(struct asd_ascb *ascb)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct sas_task *task = ascb->uldd_task;
+
+	if (task->data_dir == PCI_DMA_NONE)
+		return;
+
+	if (task->num_scatter == 0) {
+		dma_addr_t dma = (dma_addr_t)
+		       le64_to_cpu(ascb->scb->ssp_task.sg_element[0].bus_addr);
+		pci_unmap_single(ascb->ha->pcidev, dma, task->total_xfer_len,
+				 task->data_dir);
+		return;
+	}
+
+	asd_free_coherent(asd_ha, ascb->sg_arr);
+	pci_unmap_sg(asd_ha->pcidev, task->scatter, task->num_scatter,
+		     task->data_dir);
+}
+
+/* ---------- Task complete tasklet ---------- */
+
+static void asd_get_response_tasklet(struct asd_ascb *ascb,
+				     struct done_list_struct *dl)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct sas_task *task = ascb->uldd_task;
+	struct task_status_struct *ts = &task->task_status;
+	unsigned long flags;
+	struct tc_resp_sb_struct {
+		__le16 index_escb;
+		u8     len_lsb;
+		u8     flags;
+	} __attribute__ ((packed)) *resp_sb = (void *) dl->status_block;
+
+/* 	int  size   = ((resp_sb->flags & 7) << 8) | resp_sb->len_lsb; */
+	int  edb_id = ((resp_sb->flags & 0x70) >> 4)-1;
+	struct asd_ascb *escb;
+	struct asd_dma_tok *edb;
+	void *r;
+
+	spin_lock_irqsave(&asd_ha->seq.tc_index_lock, flags);
+	escb = asd_tc_index_find(&asd_ha->seq,
+				 (int)le16_to_cpu(resp_sb->index_escb));
+	spin_unlock_irqrestore(&asd_ha->seq.tc_index_lock, flags);
+
+	if (!escb) {
+		ASD_DPRINTK("Uh-oh! No escb for this dl?!\n");
+		return;
+	}
+
+	ts->buf_valid_size = 0;
+	edb = asd_ha->seq.edb_arr[edb_id + escb->edb_index];
+	r = edb->vaddr;
+	if (task->task_proto == SAS_PROTO_SSP) {
+		struct ssp_response_iu *iu =
+			r + 16 + sizeof(struct ssp_frame_hdr);
+
+		ts->residual = le32_to_cpu(*(__le32 *)r);
+		ts->resp = SAS_TASK_COMPLETE;
+		if (iu->datapres == 0)
+			ts->stat = iu->status;
+		else if (iu->datapres == 1)
+			ts->stat = iu->resp_data[3];
+		else if (iu->datapres == 2) {
+			ts->stat = SAM_CHECK_COND;
+			ts->buf_valid_size = min((u32) SAS_STATUS_BUF_SIZE,
+					 be32_to_cpu(iu->sense_data_len));
+			memcpy(ts->buf, iu->sense_data, ts->buf_valid_size);
+			if (iu->status != SAM_CHECK_COND) {
+				ASD_DPRINTK("device %llx sent sense data, but "
+					    "stat(0x%x) is not CHECK_CONDITION"
+					    "\n",
+					    SAS_ADDR(task->dev->sas_addr),
+					    ts->stat);
+			}
+		}
+	}  else {
+		struct ata_task_resp *resp = (void *) &ts->buf[0];
+
+		ts->residual = le32_to_cpu(*(__le32 *)r);
+
+		if (SAS_STATUS_BUF_SIZE >= sizeof(*resp)) {
+			resp->frame_len = le16_to_cpu(*(__le16 *)(r+6));
+			memcpy(&resp->ending_fis[0], r+16, 24);
+			ts->buf_valid_size = sizeof(*resp);
+		}
+	}
+
+	asd_invalidate_edb(escb, edb_id);
+}
+
+static void asd_task_tasklet_complete(struct asd_ascb *ascb,
+				      struct done_list_struct *dl)
+{
+	struct sas_task *task = ascb->uldd_task;
+	struct task_status_struct *ts = &task->task_status;
+	unsigned long flags;
+	u8 opcode = dl->opcode;
+
+	asd_can_dequeue(ascb->ha, 1);
+
+Again:
+	switch (opcode) {
+	case TC_NO_ERROR:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAM_GOOD;
+		break;
+	case TC_UNDERRUN:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_DATA_UNDERRUN;
+		ts->residual = le32_to_cpu(*(__le32 *)dl->status_block);
+		break;
+	case TC_OVERRUN:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_DATA_OVERRUN;
+		ts->residual = 0;
+		break;
+	case TC_SSP_RESP:
+	case TC_ATA_RESP:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_PROTO_RESPONSE;
+		asd_get_response_tasklet(ascb, dl);
+		break;
+	case TF_OPEN_REJECT:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_OPEN_REJECT;
+		if (dl->status_block[1] & 2)
+			ts->open_rej_reason = 1 + dl->status_block[2];
+		else if (dl->status_block[1] & 1)
+			ts->open_rej_reason = (dl->status_block[2] >> 4)+10;
+		else
+			ts->open_rej_reason = SAS_OREJ_UNKNOWN;
+		break;
+	case TF_OPEN_TO:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_OPEN_TO;
+		break;
+	case TF_PHY_DOWN:
+	case TU_PHY_DOWN:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_PHY_DOWN;
+		break;
+	case TI_PHY_DOWN:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_PHY_DOWN;
+		break;
+	case TI_BREAK:
+	case TI_PROTO_ERR:
+	case TI_NAK:
+	case TI_ACK_NAK_TO:
+	case TF_SMP_XMIT_RCV_ERR:
+	case TC_ATA_R_ERR_RECV:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_INTERRUPTED;
+		break;
+	case TF_BREAK:
+	case TU_BREAK:
+	case TU_ACK_NAK_TO:
+	case TF_SMPRSP_TO:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_DEV_NO_RESPONSE;
+		break;
+	case TF_NAK_RECV:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_NAK_R_ERR;
+		break;
+	case TA_I_T_NEXUS_LOSS:
+		opcode = dl->status_block[0];
+		goto Again;
+		break;
+	case TF_INV_CONN_HANDLE:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_DEVICE_UNKNOWN;
+		break;
+	case TF_REQUESTED_N_PENDING:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_PENDING;
+		break;
+	case TC_TASK_CLEARED:
+	case TA_ON_REQ:
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_ABORTED_TASK;
+		break;
+
+	case TF_NO_SMP_CONN:
+	case TF_TMF_NO_CTX:
+	case TF_TMF_NO_TAG:
+	case TF_TMF_TAG_FREE:
+	case TF_TMF_TASK_DONE:
+	case TF_TMF_NO_CONN_HANDLE:
+	case TF_IRTT_TO:
+	case TF_IU_SHORT:
+	case TF_DATA_OFFS_ERR:
+		ts->resp = SAS_TASK_UNDELIVERED;
+		ts->stat = SAS_DEV_NO_RESPONSE;
+		break;
+
+	case TC_LINK_ADM_RESP:
+	case TC_CONTROL_PHY:
+	case TC_RESUME:
+	case TC_PARTIAL_SG_LIST:
+	default:
+		ASD_DPRINTK("%s: dl opcode: 0x%x?\n", __FUNCTION__, opcode);
+		break;
+	}
+
+	switch (task->task_proto) {
+	case SATA_PROTO:
+	case SAS_PROTO_STP:
+		asd_unbuild_ata_ascb(ascb);
+		break;
+	case SAS_PROTO_SMP:
+		asd_unbuild_smp_ascb(ascb);
+		break;
+	case SAS_PROTO_SSP:
+		asd_unbuild_ssp_ascb(ascb);
+	default:
+		break;
+	}
+
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	task->task_state_flags &= ~SAS_TASK_STATE_PENDING;
+	task->task_state_flags |= SAS_TASK_STATE_DONE;
+	if (unlikely((task->task_state_flags & SAS_TASK_STATE_ABORTED))) {
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+		ASD_DPRINTK("task 0x%p done with opcode 0x%x resp 0x%x "
+			    "stat 0x%x but aborted by upper layer!\n",
+			    task, opcode, ts->resp, ts->stat);
+		complete(&ascb->completion);
+	} else {
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+		task->lldd_task = NULL;
+		asd_ascb_free(ascb);
+		mb();
+		task->task_done(task);
+	}
+}
+
+/* ---------- ATA ---------- */
+
+static int asd_build_ata_ascb(struct asd_ascb *ascb, struct sas_task *task,
+			      unsigned long gfp_flags)
+{
+	struct domain_device *dev = task->dev;
+	struct scb *scb;
+	u8     flags;
+	int    res = 0;
+
+	scb = ascb->scb;
+
+	if (unlikely(task->ata_task.device_control_reg_update))
+		scb->header.opcode = CONTROL_ATA_DEV;
+	else if (dev->sata_dev.command_set == ATA_COMMAND_SET)
+		scb->header.opcode = INITIATE_ATA_TASK;
+	else
+		scb->header.opcode = INITIATE_ATAPI_TASK;
+
+	scb->ata_task.proto_conn_rate = (1 << 5); /* STP */
+	if (dev->port->oob_mode == SAS_OOB_MODE)
+		scb->ata_task.proto_conn_rate |= dev->linkrate;
+
+	scb->ata_task.total_xfer_len = cpu_to_le32(task->total_xfer_len);
+	scb->ata_task.fis = task->ata_task.fis;
+	scb->ata_task.fis.fis_type = 0x27;
+	if (likely(!task->ata_task.device_control_reg_update))
+		scb->ata_task.fis.flags |= 0x80; /* C=1: update ATA cmd reg */
+	scb->ata_task.fis.flags &= 0xF0; /* PM_PORT field shall be 0 */
+	if (dev->sata_dev.command_set == ATAPI_COMMAND_SET)
+		memcpy(scb->ata_task.atapi_packet, task->ata_task.atapi_packet,
+		       16);
+	scb->ata_task.sister_scb = cpu_to_le16(0xFFFF);
+	scb->ata_task.conn_handle = cpu_to_le16(
+		(u16)(unsigned long)dev->lldd_dev);
+
+	if (likely(!task->ata_task.device_control_reg_update)) {
+		flags = 0;
+		if (task->ata_task.dma_xfer)
+			flags |= DATA_XFER_MODE_DMA;
+		if (task->ata_task.use_ncq &&
+		    dev->sata_dev.command_set != ATAPI_COMMAND_SET)
+			flags |= ATA_Q_TYPE_NCQ;
+		flags |= data_dir_flags[task->data_dir];
+		scb->ata_task.ata_flags = flags;
+
+		scb->ata_task.retry_count = task->ata_task.retry_count;
+
+		flags = 0;
+		if (task->ata_task.set_affil_pol)
+			flags |= SET_AFFIL_POLICY;
+		if (task->ata_task.stp_affil_pol)
+			flags |= STP_AFFIL_POLICY;
+		scb->ata_task.flags = flags;
+	}
+	ascb->tasklet_complete = asd_task_tasklet_complete;
+
+	if (likely(!task->ata_task.device_control_reg_update))
+		res = asd_map_scatterlist(task, scb->ata_task.sg_element,
+					  gfp_flags);
+
+	return res;
+}
+
+static void asd_unbuild_ata_ascb(struct asd_ascb *a)
+{
+	asd_unmap_scatterlist(a);
+}
+
+/* ---------- SMP ---------- */
+
+static int asd_build_smp_ascb(struct asd_ascb *ascb, struct sas_task *task,
+			      unsigned long gfp_flags)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	struct domain_device *dev = task->dev;
+	struct scb *scb;
+
+	pci_map_sg(asd_ha->pcidev, &task->smp_task.smp_req, 1,
+		   PCI_DMA_FROMDEVICE);
+	pci_map_sg(asd_ha->pcidev, &task->smp_task.smp_resp, 1,
+		   PCI_DMA_FROMDEVICE);
+
+	scb = ascb->scb;
+
+	scb->header.opcode = INITIATE_SMP_TASK;
+
+	scb->smp_task.proto_conn_rate = dev->linkrate;
+
+	scb->smp_task.smp_req.bus_addr =
+		cpu_to_le64((u64)sg_dma_address(&task->smp_task.smp_req));
+	scb->smp_task.smp_req.size =
+		cpu_to_le32((u32)sg_dma_len(&task->smp_task.smp_req)-4);
+
+	scb->smp_task.smp_resp.bus_addr =
+		cpu_to_le64((u64)sg_dma_address(&task->smp_task.smp_resp));
+	scb->smp_task.smp_resp.size =
+		cpu_to_le32((u32)sg_dma_len(&task->smp_task.smp_resp)-4);
+
+	scb->smp_task.sister_scb = cpu_to_le16(0xFFFF);
+	scb->smp_task.conn_handle = cpu_to_le16((u16)
+						(unsigned long)dev->lldd_dev);
+
+	ascb->tasklet_complete = asd_task_tasklet_complete;
+
+	return 0;
+}
+
+static void asd_unbuild_smp_ascb(struct asd_ascb *a)
+{
+	struct sas_task *task = a->uldd_task;
+
+	BUG_ON(!task);
+	pci_unmap_sg(a->ha->pcidev, &task->smp_task.smp_req, 1,
+		     PCI_DMA_FROMDEVICE);
+	pci_unmap_sg(a->ha->pcidev, &task->smp_task.smp_resp, 1,
+		     PCI_DMA_FROMDEVICE);
+}
+
+/* ---------- SSP ---------- */
+
+static int asd_build_ssp_ascb(struct asd_ascb *ascb, struct sas_task *task,
+			      unsigned long gfp_flags)
+{
+	struct domain_device *dev = task->dev;
+	struct scb *scb;
+	int    res = 0;
+
+	scb = ascb->scb;
+
+	scb->header.opcode = INITIATE_SSP_TASK;
+
+	scb->ssp_task.proto_conn_rate  = (1 << 4); /* SSP */
+	scb->ssp_task.proto_conn_rate |= dev->linkrate;
+	scb->ssp_task.total_xfer_len = cpu_to_le32(task->total_xfer_len);
+	scb->ssp_task.ssp_frame.frame_type = SSP_DATA;
+	memcpy(scb->ssp_task.ssp_frame.hashed_dest_addr, dev->hashed_sas_addr,
+	       HASHED_SAS_ADDR_SIZE);
+	memcpy(scb->ssp_task.ssp_frame.hashed_src_addr,
+	       dev->port->ha->hashed_sas_addr, HASHED_SAS_ADDR_SIZE);
+	scb->ssp_task.ssp_frame.tptt = cpu_to_be16(0xFFFF);
+
+	memcpy(scb->ssp_task.ssp_cmd.lun, task->ssp_task.LUN, 8);
+	if (task->ssp_task.enable_first_burst)
+		scb->ssp_task.ssp_cmd.efb_prio_attr |= EFB_MASK;
+	scb->ssp_task.ssp_cmd.efb_prio_attr |= (task->ssp_task.task_prio << 3);
+	scb->ssp_task.ssp_cmd.efb_prio_attr |= (task->ssp_task.task_attr & 7);
+	memcpy(scb->ssp_task.ssp_cmd.cdb, task->ssp_task.cdb, 16);
+
+	scb->ssp_task.sister_scb = cpu_to_le16(0xFFFF);
+	scb->ssp_task.conn_handle = cpu_to_le16(
+		(u16)(unsigned long)dev->lldd_dev);
+	scb->ssp_task.data_dir = data_dir_flags[task->data_dir];
+	scb->ssp_task.retry_count = scb->ssp_task.retry_count;
+
+	ascb->tasklet_complete = asd_task_tasklet_complete;
+
+	res = asd_map_scatterlist(task, scb->ssp_task.sg_element, gfp_flags);
+
+	return res;
+}
+
+static void asd_unbuild_ssp_ascb(struct asd_ascb *a)
+{
+	asd_unmap_scatterlist(a);
+}
+
+/* ---------- Execute Task ---------- */
+
+static inline int asd_can_queue(struct asd_ha_struct *asd_ha, int num)
+{
+	int res = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&asd_ha->seq.pend_q_lock, flags);
+	if ((asd_ha->seq.can_queue - num) < 0)
+		res = -SAS_QUEUE_FULL;
+	else
+		asd_ha->seq.can_queue -= num;
+	spin_unlock_irqrestore(&asd_ha->seq.pend_q_lock, flags);
+
+	return res;
+}
+
+int asd_execute_task(struct sas_task *task, const int num,
+		     unsigned long gfp_flags)
+{
+	int res = 0;
+	LIST_HEAD(alist);
+	struct sas_task *t = task;
+	struct asd_ascb *ascb = NULL, *a;
+	struct asd_ha_struct *asd_ha = task->dev->port->ha->lldd_ha;
+
+	res = asd_can_queue(asd_ha, num);
+	if (res)
+		return res;
+
+	res = num;
+	ascb = asd_ascb_alloc_list(asd_ha, &res, gfp_flags);
+	if (res) {
+		res = -ENOMEM;
+		goto out_err;
+	}
+
+	__list_add(&alist, ascb->list.prev, &ascb->list);
+	list_for_each_entry(a, &alist, list) {
+		a->uldd_task = t;
+		t->lldd_task = a;
+		t = list_entry(t->list.next, struct sas_task, list);
+	}
+	list_for_each_entry(a, &alist, list) {
+		t = a->uldd_task;
+		a->uldd_timer = 1;
+		if (t->task_proto & SAS_PROTO_STP)
+			t->task_proto = SAS_PROTO_STP;
+		switch (t->task_proto) {
+		case SATA_PROTO:
+		case SAS_PROTO_STP:
+			res = asd_build_ata_ascb(a, t, gfp_flags);
+			break;
+		case SAS_PROTO_SMP:
+			res = asd_build_smp_ascb(a, t, gfp_flags);
+			break;
+		case SAS_PROTO_SSP:
+			res = asd_build_ssp_ascb(a, t, gfp_flags);
+			break;
+		default:
+			asd_printk("unknown sas_task proto: 0x%x\n",
+				   t->task_proto);
+			res = -ENOMEM;
+			break;
+		}
+		if (res)
+			goto out_err_unmap;
+	}
+	list_del_init(&alist);
+
+	res = asd_post_ascb_list(asd_ha, ascb, num);
+	if (unlikely(res)) {
+		a = NULL;
+		__list_add(&alist, ascb->list.prev, &ascb->list);
+		goto out_err_unmap;
+	}
+
+	return 0;
+out_err_unmap:
+	{
+		struct asd_ascb *b = a;
+		list_for_each_entry(a, &alist, list) {
+			if (a == b)
+				break;
+			t = a->uldd_task;
+			switch (t->task_proto) {
+			case SATA_PROTO:
+			case SAS_PROTO_STP:
+				asd_unbuild_ata_ascb(a);
+				break;
+			case SAS_PROTO_SMP:
+				asd_unbuild_smp_ascb(a);
+				break;
+			case SAS_PROTO_SSP:
+				asd_unbuild_ssp_ascb(a);
+			default:
+				break;
+			}
+			t->lldd_task = NULL;
+		}
+	}
+	list_del_init(&alist);
+out_err:
+	if (ascb)
+		asd_ascb_free_list(ascb);
+	asd_can_dequeue(asd_ha, num);
+	return res;
+}
diff --git a/drivers/scsi/aic94xx/aic94xx_tmf.c b/drivers/scsi/aic94xx/aic94xx_tmf.c
new file mode 100644
index 000000000000..61234384503b
--- /dev/null
+++ b/drivers/scsi/aic94xx/aic94xx_tmf.c
@@ -0,0 +1,636 @@
+/*
+ * Aic94xx Task Management Functions
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This file is part of the aic94xx driver.
+ *
+ * The aic94xx driver is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * The aic94xx driver is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the aic94xx driver; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/spinlock.h>
+#include "aic94xx.h"
+#include "aic94xx_sas.h"
+#include "aic94xx_hwi.h"
+
+/* ---------- Internal enqueue ---------- */
+
+static int asd_enqueue_internal(struct asd_ascb *ascb,
+		void (*tasklet_complete)(struct asd_ascb *,
+					 struct done_list_struct *),
+				void (*timed_out)(unsigned long))
+{
+	int res;
+
+	ascb->tasklet_complete = tasklet_complete;
+	ascb->uldd_timer = 1;
+
+	ascb->timer.data = (unsigned long) ascb;
+	ascb->timer.function = timed_out;
+	ascb->timer.expires = jiffies + AIC94XX_SCB_TIMEOUT;
+
+	add_timer(&ascb->timer);
+
+	res = asd_post_ascb_list(ascb->ha, ascb, 1);
+	if (unlikely(res))
+		del_timer(&ascb->timer);
+	return res;
+}
+
+static inline void asd_timedout_common(unsigned long data)
+{
+	struct asd_ascb *ascb = (void *) data;
+	struct asd_seq_data *seq = &ascb->ha->seq;
+        unsigned long flags;
+
+	spin_lock_irqsave(&seq->pend_q_lock, flags);
+        seq->pending--;
+        list_del_init(&ascb->list);
+        spin_unlock_irqrestore(&seq->pend_q_lock, flags);
+}
+
+/* ---------- CLEAR NEXUS ---------- */
+
+static void asd_clear_nexus_tasklet_complete(struct asd_ascb *ascb,
+					     struct done_list_struct *dl)
+{
+	ASD_DPRINTK("%s: here\n", __FUNCTION__);
+	if (!del_timer(&ascb->timer)) {
+		ASD_DPRINTK("%s: couldn't delete timer\n", __FUNCTION__);
+		return;
+	}
+	ASD_DPRINTK("%s: opcode: 0x%x\n", __FUNCTION__, dl->opcode);
+	ascb->uldd_task = (void *) (unsigned long) dl->opcode;
+	complete(&ascb->completion);
+}
+
+static void asd_clear_nexus_timedout(unsigned long data)
+{
+	struct asd_ascb *ascb = (void *) data;
+
+	ASD_DPRINTK("%s: here\n", __FUNCTION__);
+	asd_timedout_common(data);
+	ascb->uldd_task = (void *) TMF_RESP_FUNC_FAILED;
+	complete(&ascb->completion);
+}
+
+#define CLEAR_NEXUS_PRE         \
+	ASD_DPRINTK("%s: PRE\n", __FUNCTION__); \
+        res = 1;                \
+	ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL); \
+	if (!ascb)              \
+		return -ENOMEM; \
+                                \
+	scb = ascb->scb;        \
+	scb->header.opcode = CLEAR_NEXUS
+
+#define CLEAR_NEXUS_POST        \
+	ASD_DPRINTK("%s: POST\n", __FUNCTION__); \
+	res = asd_enqueue_internal(ascb, asd_clear_nexus_tasklet_complete, \
+				   asd_clear_nexus_timedout);              \
+	if (res)                \
+		goto out_err;   \
+	ASD_DPRINTK("%s: clear nexus posted, waiting...\n", __FUNCTION__); \
+	wait_for_completion(&ascb->completion); \
+	res = (int) (unsigned long) ascb->uldd_task; \
+	if (res == TC_NO_ERROR) \
+		res = TMF_RESP_FUNC_COMPLETE;   \
+out_err:                        \
+	asd_ascb_free(ascb);    \
+	return res
+
+int asd_clear_nexus_ha(struct sas_ha_struct *sas_ha)
+{
+	struct asd_ha_struct *asd_ha = sas_ha->lldd_ha;
+	struct asd_ascb *ascb;
+	struct scb *scb;
+	int res;
+
+	CLEAR_NEXUS_PRE;
+	scb->clear_nexus.nexus = NEXUS_ADAPTER;
+	CLEAR_NEXUS_POST;
+}
+
+int asd_clear_nexus_port(struct asd_sas_port *port)
+{
+	struct asd_ha_struct *asd_ha = port->ha->lldd_ha;
+	struct asd_ascb *ascb;
+	struct scb *scb;
+	int res;
+
+	CLEAR_NEXUS_PRE;
+	scb->clear_nexus.nexus = NEXUS_PORT;
+	scb->clear_nexus.conn_mask = port->phy_mask;
+	CLEAR_NEXUS_POST;
+}
+
+#if 0
+static int asd_clear_nexus_I_T(struct domain_device *dev)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	struct asd_ascb *ascb;
+	struct scb *scb;
+	int res;
+
+	CLEAR_NEXUS_PRE;
+	scb->clear_nexus.nexus = NEXUS_I_T;
+	scb->clear_nexus.flags = SEND_Q | EXEC_Q | NOTINQ;
+	if (dev->tproto)
+		scb->clear_nexus.flags |= SUSPEND_TX;
+	scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long)
+						   dev->lldd_dev);
+	CLEAR_NEXUS_POST;
+}
+#endif
+
+static int asd_clear_nexus_I_T_L(struct domain_device *dev, u8 *lun)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	struct asd_ascb *ascb;
+	struct scb *scb;
+	int res;
+
+	CLEAR_NEXUS_PRE;
+	scb->clear_nexus.nexus = NEXUS_I_T_L;
+	scb->clear_nexus.flags = SEND_Q | EXEC_Q | NOTINQ;
+	if (dev->tproto)
+		scb->clear_nexus.flags |= SUSPEND_TX;
+	memcpy(scb->clear_nexus.ssp_task.lun, lun, 8);
+	scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long)
+						   dev->lldd_dev);
+	CLEAR_NEXUS_POST;
+}
+
+static int asd_clear_nexus_tag(struct sas_task *task)
+{
+	struct asd_ha_struct *asd_ha = task->dev->port->ha->lldd_ha;
+	struct asd_ascb *tascb = task->lldd_task;
+	struct asd_ascb *ascb;
+	struct scb *scb;
+	int res;
+
+	CLEAR_NEXUS_PRE;
+	scb->clear_nexus.nexus = NEXUS_TAG;
+	memcpy(scb->clear_nexus.ssp_task.lun, task->ssp_task.LUN, 8);
+	scb->clear_nexus.ssp_task.tag = tascb->tag;
+	if (task->dev->tproto)
+		scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long)
+							  task->dev->lldd_dev);
+	CLEAR_NEXUS_POST;
+}
+
+static int asd_clear_nexus_index(struct sas_task *task)
+{
+	struct asd_ha_struct *asd_ha = task->dev->port->ha->lldd_ha;
+	struct asd_ascb *tascb = task->lldd_task;
+	struct asd_ascb *ascb;
+	struct scb *scb;
+	int res;
+
+	CLEAR_NEXUS_PRE;
+	scb->clear_nexus.nexus = NEXUS_TRANS_CX;
+	if (task->dev->tproto)
+		scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long)
+							  task->dev->lldd_dev);
+	scb->clear_nexus.index = cpu_to_le16(tascb->tc_index);
+	CLEAR_NEXUS_POST;
+}
+
+/* ---------- TMFs ---------- */
+
+static void asd_tmf_timedout(unsigned long data)
+{
+	struct asd_ascb *ascb = (void *) data;
+
+	ASD_DPRINTK("tmf timed out\n");
+	asd_timedout_common(data);
+	ascb->uldd_task = (void *) TMF_RESP_FUNC_FAILED;
+	complete(&ascb->completion);
+}
+
+static int asd_get_tmf_resp_tasklet(struct asd_ascb *ascb,
+				    struct done_list_struct *dl)
+{
+	struct asd_ha_struct *asd_ha = ascb->ha;
+	unsigned long flags;
+	struct tc_resp_sb_struct {
+		__le16 index_escb;
+		u8     len_lsb;
+		u8     flags;
+	} __attribute__ ((packed)) *resp_sb = (void *) dl->status_block;
+
+	int  edb_id = ((resp_sb->flags & 0x70) >> 4)-1;
+	struct asd_ascb *escb;
+	struct asd_dma_tok *edb;
+	struct ssp_frame_hdr *fh;
+	struct ssp_response_iu   *ru;
+	int res = TMF_RESP_FUNC_FAILED;
+
+	ASD_DPRINTK("tmf resp tasklet\n");
+
+	spin_lock_irqsave(&asd_ha->seq.tc_index_lock, flags);
+	escb = asd_tc_index_find(&asd_ha->seq,
+				 (int)le16_to_cpu(resp_sb->index_escb));
+	spin_unlock_irqrestore(&asd_ha->seq.tc_index_lock, flags);
+
+	if (!escb) {
+		ASD_DPRINTK("Uh-oh! No escb for this dl?!\n");
+		return res;
+	}
+
+	edb = asd_ha->seq.edb_arr[edb_id + escb->edb_index];
+	ascb->tag = *(__be16 *)(edb->vaddr+4);
+	fh = edb->vaddr + 16;
+	ru = edb->vaddr + 16 + sizeof(*fh);
+	res = ru->status;
+	if (ru->datapres == 1)	  /* Response data present */
+		res = ru->resp_data[3];
+#if 0
+	ascb->tag = fh->tag;
+#endif
+	ascb->tag_valid = 1;
+
+	asd_invalidate_edb(escb, edb_id);
+	return res;
+}
+
+static void asd_tmf_tasklet_complete(struct asd_ascb *ascb,
+				     struct done_list_struct *dl)
+{
+	if (!del_timer(&ascb->timer))
+		return;
+
+	ASD_DPRINTK("tmf tasklet complete\n");
+
+	if (dl->opcode == TC_SSP_RESP)
+		ascb->uldd_task = (void *) (unsigned long)
+			asd_get_tmf_resp_tasklet(ascb, dl);
+	else
+		ascb->uldd_task = (void *) 0xFF00 + (unsigned long) dl->opcode;
+
+	complete(&ascb->completion);
+}
+
+static inline int asd_clear_nexus(struct sas_task *task)
+{
+	int res = TMF_RESP_FUNC_FAILED;
+	struct asd_ascb *tascb = task->lldd_task;
+	unsigned long flags;
+
+	ASD_DPRINTK("task not done, clearing nexus\n");
+	if (tascb->tag_valid)
+		res = asd_clear_nexus_tag(task);
+	else
+		res = asd_clear_nexus_index(task);
+	wait_for_completion_timeout(&tascb->completion,
+				    AIC94XX_SCB_TIMEOUT);
+	ASD_DPRINTK("came back from clear nexus\n");
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	if (task->task_state_flags & SAS_TASK_STATE_DONE)
+		res = TMF_RESP_FUNC_COMPLETE;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	return res;
+}
+
+/**
+ * asd_abort_task -- ABORT TASK TMF
+ * @task: the task to be aborted
+ *
+ * Before calling ABORT TASK the task state flags should be ORed with
+ * SAS_TASK_STATE_ABORTED (unless SAS_TASK_STATE_DONE is set) under
+ * the task_state_lock IRQ spinlock, then ABORT TASK *must* be called.
+ *
+ * Implements the ABORT TASK TMF, I_T_L_Q nexus.
+ * Returns: SAS TMF responses (see sas_task.h),
+ *          -ENOMEM,
+ *          -SAS_QUEUE_FULL.
+ *
+ * When ABORT TASK returns, the caller of ABORT TASK checks first the
+ * task->task_state_flags, and then the return value of ABORT TASK.
+ *
+ * If the task has task state bit SAS_TASK_STATE_DONE set, then the
+ * task was completed successfully prior to it being aborted.  The
+ * caller of ABORT TASK has responsibility to call task->task_done()
+ * xor free the task, depending on their framework.  The return code
+ * is TMF_RESP_FUNC_FAILED in this case.
+ *
+ * Else the SAS_TASK_STATE_DONE bit is not set,
+ * 	If the return code is TMF_RESP_FUNC_COMPLETE, then
+ * 		the task was aborted successfully.  The caller of
+ * 		ABORT TASK has responsibility to call task->task_done()
+ *              to finish the task, xor free the task depending on their
+ *		framework.
+ *	else
+ * 		the ABORT TASK returned some kind of error. The task
+ *              was _not_ cancelled.  Nothing can be assumed.
+ *		The caller of ABORT TASK may wish to retry.
+ */
+int asd_abort_task(struct sas_task *task)
+{
+	struct asd_ascb *tascb = task->lldd_task;
+	struct asd_ha_struct *asd_ha = tascb->ha;
+	int res = 1;
+	unsigned long flags;
+	struct asd_ascb *ascb = NULL;
+	struct scb *scb;
+
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	if (task->task_state_flags & SAS_TASK_STATE_DONE) {
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+		res = TMF_RESP_FUNC_COMPLETE;
+		ASD_DPRINTK("%s: task 0x%p done\n", __FUNCTION__, task);
+		goto out_done;
+	}
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL);
+	if (!ascb)
+		return -ENOMEM;
+	scb = ascb->scb;
+
+	scb->header.opcode = ABORT_TASK;
+
+	switch (task->task_proto) {
+	case SATA_PROTO:
+	case SAS_PROTO_STP:
+		scb->abort_task.proto_conn_rate = (1 << 5); /* STP */
+		break;
+	case SAS_PROTO_SSP:
+		scb->abort_task.proto_conn_rate  = (1 << 4); /* SSP */
+		scb->abort_task.proto_conn_rate |= task->dev->linkrate;
+		break;
+	case SAS_PROTO_SMP:
+		break;
+	default:
+		break;
+	}
+
+	if (task->task_proto == SAS_PROTO_SSP) {
+		scb->abort_task.ssp_frame.frame_type = SSP_TASK;
+		memcpy(scb->abort_task.ssp_frame.hashed_dest_addr,
+		       task->dev->hashed_sas_addr, HASHED_SAS_ADDR_SIZE);
+		memcpy(scb->abort_task.ssp_frame.hashed_src_addr,
+		       task->dev->port->ha->hashed_sas_addr,
+		       HASHED_SAS_ADDR_SIZE);
+		scb->abort_task.ssp_frame.tptt = cpu_to_be16(0xFFFF);
+
+		memcpy(scb->abort_task.ssp_task.lun, task->ssp_task.LUN, 8);
+		scb->abort_task.ssp_task.tmf = TMF_ABORT_TASK;
+		scb->abort_task.ssp_task.tag = cpu_to_be16(0xFFFF);
+	}
+
+	scb->abort_task.sister_scb = cpu_to_le16(0xFFFF);
+	scb->abort_task.conn_handle = cpu_to_le16(
+		(u16)(unsigned long)task->dev->lldd_dev);
+	scb->abort_task.retry_count = 1;
+	scb->abort_task.index = cpu_to_le16((u16)tascb->tc_index);
+	scb->abort_task.itnl_to = cpu_to_le16(ITNL_TIMEOUT_CONST);
+
+	res = asd_enqueue_internal(ascb, asd_tmf_tasklet_complete,
+				   asd_tmf_timedout);
+	if (res)
+		goto out;
+	wait_for_completion(&ascb->completion);
+	ASD_DPRINTK("tmf came back\n");
+
+	res = (int) (unsigned long) ascb->uldd_task;
+	tascb->tag = ascb->tag;
+	tascb->tag_valid = ascb->tag_valid;
+
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	if (task->task_state_flags & SAS_TASK_STATE_DONE) {
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+		res = TMF_RESP_FUNC_COMPLETE;
+		ASD_DPRINTK("%s: task 0x%p done\n", __FUNCTION__, task);
+		goto out_done;
+	}
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	switch (res) {
+	/* The task to be aborted has been sent to the device.
+	 * We got a Response IU for the ABORT TASK TMF. */
+	case TC_NO_ERROR + 0xFF00:
+	case TMF_RESP_FUNC_COMPLETE:
+	case TMF_RESP_FUNC_FAILED:
+		res = asd_clear_nexus(task);
+		break;
+	case TMF_RESP_INVALID_FRAME:
+	case TMF_RESP_OVERLAPPED_TAG:
+	case TMF_RESP_FUNC_ESUPP:
+	case TMF_RESP_NO_LUN:
+		goto out_done; break;
+	}
+	/* In the following we assume that the managing layer
+	 * will _never_ make a mistake, when issuing ABORT TASK.
+	 */
+	switch (res) {
+	default:
+		res = asd_clear_nexus(task);
+		/* fallthrough */
+	case TC_NO_ERROR + 0xFF00:
+	case TMF_RESP_FUNC_COMPLETE:
+		break;
+	/* The task hasn't been sent to the device xor we never got
+	 * a (sane) Response IU for the ABORT TASK TMF.
+	 */
+	case TF_NAK_RECV + 0xFF00:
+		res = TMF_RESP_INVALID_FRAME;
+		break;
+	case TF_TMF_TASK_DONE + 0xFF00:	/* done but not reported yet */
+		res = TMF_RESP_FUNC_FAILED;
+		wait_for_completion_timeout(&tascb->completion,
+					    AIC94XX_SCB_TIMEOUT);
+		spin_lock_irqsave(&task->task_state_lock, flags);
+		if (task->task_state_flags & SAS_TASK_STATE_DONE)
+			res = TMF_RESP_FUNC_COMPLETE;
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+		goto out_done;
+	case TF_TMF_NO_TAG + 0xFF00:
+	case TF_TMF_TAG_FREE + 0xFF00: /* the tag is in the free list */
+	case TF_TMF_NO_CONN_HANDLE + 0xFF00: /* no such device */
+		res = TMF_RESP_FUNC_COMPLETE;
+		goto out_done;
+	case TF_TMF_NO_CTX + 0xFF00: /* not in seq, or proto != SSP */
+		res = TMF_RESP_FUNC_ESUPP;
+		goto out;
+	}
+out_done:
+	if (res == TMF_RESP_FUNC_COMPLETE) {
+		task->lldd_task = NULL;
+		mb();
+		asd_ascb_free(tascb);
+	}
+out:
+	asd_ascb_free(ascb);
+	ASD_DPRINTK("task 0x%p aborted, res: 0x%x\n", task, res);
+	return res;
+}
+
+/**
+ * asd_initiate_ssp_tmf -- send a TMF to an I_T_L or I_T_L_Q nexus
+ * @dev: pointer to struct domain_device of interest
+ * @lun: pointer to u8[8] which is the LUN
+ * @tmf: the TMF to be performed (see sas_task.h or the SAS spec)
+ * @index: the transaction context of the task to be queried if QT TMF
+ *
+ * This function is used to send ABORT TASK SET, CLEAR ACA,
+ * CLEAR TASK SET, LU RESET and QUERY TASK TMFs.
+ *
+ * No SCBs should be queued to the I_T_L nexus when this SCB is
+ * pending.
+ *
+ * Returns: TMF response code (see sas_task.h or the SAS spec)
+ */
+static int asd_initiate_ssp_tmf(struct domain_device *dev, u8 *lun,
+				int tmf, int index)
+{
+	struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha;
+	struct asd_ascb *ascb;
+	int res = 1;
+	struct scb *scb;
+
+	if (!(dev->tproto & SAS_PROTO_SSP))
+		return TMF_RESP_FUNC_ESUPP;
+
+	ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL);
+	if (!ascb)
+		return -ENOMEM;
+	scb = ascb->scb;
+
+	if (tmf == TMF_QUERY_TASK)
+		scb->header.opcode = QUERY_SSP_TASK;
+	else
+		scb->header.opcode = INITIATE_SSP_TMF;
+
+	scb->ssp_tmf.proto_conn_rate  = (1 << 4); /* SSP */
+	scb->ssp_tmf.proto_conn_rate |= dev->linkrate;
+	/* SSP frame header */
+	scb->ssp_tmf.ssp_frame.frame_type = SSP_TASK;
+	memcpy(scb->ssp_tmf.ssp_frame.hashed_dest_addr,
+	       dev->hashed_sas_addr, HASHED_SAS_ADDR_SIZE);
+	memcpy(scb->ssp_tmf.ssp_frame.hashed_src_addr,
+	       dev->port->ha->hashed_sas_addr, HASHED_SAS_ADDR_SIZE);
+	scb->ssp_tmf.ssp_frame.tptt = cpu_to_be16(0xFFFF);
+	/* SSP Task IU */
+	memcpy(scb->ssp_tmf.ssp_task.lun, lun, 8);
+	scb->ssp_tmf.ssp_task.tmf = tmf;
+
+	scb->ssp_tmf.sister_scb = cpu_to_le16(0xFFFF);
+	scb->ssp_tmf.conn_handle= cpu_to_le16((u16)(unsigned long)
+					      dev->lldd_dev);
+	scb->ssp_tmf.retry_count = 1;
+	scb->ssp_tmf.itnl_to = cpu_to_le16(ITNL_TIMEOUT_CONST);
+	if (tmf == TMF_QUERY_TASK)
+		scb->ssp_tmf.index = cpu_to_le16(index);
+
+	res = asd_enqueue_internal(ascb, asd_tmf_tasklet_complete,
+				   asd_tmf_timedout);
+	if (res)
+		goto out_err;
+	wait_for_completion(&ascb->completion);
+	res = (int) (unsigned long) ascb->uldd_task;
+
+	switch (res) {
+	case TC_NO_ERROR + 0xFF00:
+		res = TMF_RESP_FUNC_COMPLETE;
+		break;
+	case TF_NAK_RECV + 0xFF00:
+		res = TMF_RESP_INVALID_FRAME;
+		break;
+	case TF_TMF_TASK_DONE + 0xFF00:
+		res = TMF_RESP_FUNC_FAILED;
+		break;
+	case TF_TMF_NO_TAG + 0xFF00:
+	case TF_TMF_TAG_FREE + 0xFF00: /* the tag is in the free list */
+	case TF_TMF_NO_CONN_HANDLE + 0xFF00: /* no such device */
+		res = TMF_RESP_FUNC_COMPLETE;
+		break;
+	case TF_TMF_NO_CTX + 0xFF00: /* not in seq, or proto != SSP */
+		res = TMF_RESP_FUNC_ESUPP;
+		break;
+	default:
+		ASD_DPRINTK("%s: converting result 0x%x to TMF_RESP_FUNC_FAILED\n",
+			    __FUNCTION__, res);
+		res = TMF_RESP_FUNC_FAILED;
+		break;
+	}
+out_err:
+	asd_ascb_free(ascb);
+	return res;
+}
+
+int asd_abort_task_set(struct domain_device *dev, u8 *lun)
+{
+	int res = asd_initiate_ssp_tmf(dev, lun, TMF_ABORT_TASK_SET, 0);
+
+	if (res == TMF_RESP_FUNC_COMPLETE)
+		asd_clear_nexus_I_T_L(dev, lun);
+	return res;
+}
+
+int asd_clear_aca(struct domain_device *dev, u8 *lun)
+{
+	int res = asd_initiate_ssp_tmf(dev, lun, TMF_CLEAR_ACA, 0);
+
+	if (res == TMF_RESP_FUNC_COMPLETE)
+		asd_clear_nexus_I_T_L(dev, lun);
+	return res;
+}
+
+int asd_clear_task_set(struct domain_device *dev, u8 *lun)
+{
+	int res = asd_initiate_ssp_tmf(dev, lun, TMF_CLEAR_TASK_SET, 0);
+
+	if (res == TMF_RESP_FUNC_COMPLETE)
+		asd_clear_nexus_I_T_L(dev, lun);
+	return res;
+}
+
+int asd_lu_reset(struct domain_device *dev, u8 *lun)
+{
+	int res = asd_initiate_ssp_tmf(dev, lun, TMF_LU_RESET, 0);
+
+	if (res == TMF_RESP_FUNC_COMPLETE)
+		asd_clear_nexus_I_T_L(dev, lun);
+	return res;
+}
+
+/**
+ * asd_query_task -- send a QUERY TASK TMF to an I_T_L_Q nexus
+ * task: pointer to sas_task struct of interest
+ *
+ * Returns: TMF_RESP_FUNC_COMPLETE if the task is not in the task set,
+ * or TMF_RESP_FUNC_SUCC if the task is in the task set.
+ *
+ * Normally the management layer sets the task to aborted state,
+ * and then calls query task and then abort task.
+ */
+int asd_query_task(struct sas_task *task)
+{
+	struct asd_ascb *ascb = task->lldd_task;
+	int index;
+
+	if (ascb) {
+		index = ascb->tc_index;
+		return asd_initiate_ssp_tmf(task->dev, task->ssp_task.LUN,
+					    TMF_QUERY_TASK, index);
+	}
+	return TMF_RESP_FUNC_COMPLETE;
+}
diff --git a/drivers/scsi/libsas/Kconfig b/drivers/scsi/libsas/Kconfig
new file mode 100644
index 000000000000..aafdc92f8312
--- /dev/null
+++ b/drivers/scsi/libsas/Kconfig
@@ -0,0 +1,39 @@
+#
+# Kernel configuration file for the SAS Class
+#
+# Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+# Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+#
+# This file is licensed under GPLv2.
+#
+# This program is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation; version 2 of the
+# License.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+# USA
+#
+
+config SCSI_SAS_LIBSAS
+	tristate "SAS Domain Transport Attributes"
+	depends on SCSI
+	select SCSI_SAS_ATTRS
+	help
+	  This provides transport specific helpers for SAS drivers which
+	  use the domain device construct (like the aic94xxx).
+
+config SCSI_SAS_LIBSAS_DEBUG
+	bool "Compile the SAS Domain Transport Attributes in debug mode"
+	default y
+	depends on SCSI_SAS_LIBSAS
+	help
+		Compiles the SAS Layer in debug mode.  In debug mode, the
+		SAS Layer prints diagnostic and debug messages.
diff --git a/drivers/scsi/libsas/Makefile b/drivers/scsi/libsas/Makefile
new file mode 100644
index 000000000000..44d972a3b4bd
--- /dev/null
+++ b/drivers/scsi/libsas/Makefile
@@ -0,0 +1,36 @@
+#
+# Kernel Makefile for the libsas helpers
+#
+# Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+# Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+#
+# This file is licensed under GPLv2.
+#
+# This program is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation; version 2 of the
+# License.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+# USA
+
+ifeq ($(CONFIG_SCSI_SAS_LIBSAS_DEBUG),y)
+	EXTRA_CFLAGS += -DSAS_DEBUG
+endif
+
+obj-$(CONFIG_SCSI_SAS_LIBSAS) += libsas.o
+libsas-y +=  sas_init.o     \
+		sas_phy.o      \
+		sas_port.o     \
+		sas_event.o    \
+		sas_dump.o     \
+		sas_discover.o \
+		sas_expander.o \
+		sas_scsi_host.o
diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c
new file mode 100644
index 000000000000..d977bd492d8d
--- /dev/null
+++ b/drivers/scsi/libsas/sas_discover.c
@@ -0,0 +1,749 @@
+/*
+ * Serial Attached SCSI (SAS) Discover process
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/pci.h>
+#include <linux/scatterlist.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_eh.h>
+#include "sas_internal.h"
+
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+#include "../scsi_sas_internal.h"
+
+/* ---------- Basic task processing for discovery purposes ---------- */
+
+void sas_init_dev(struct domain_device *dev)
+{
+        INIT_LIST_HEAD(&dev->siblings);
+        INIT_LIST_HEAD(&dev->dev_list_node);
+        switch (dev->dev_type) {
+        case SAS_END_DEV:
+                break;
+        case EDGE_DEV:
+        case FANOUT_DEV:
+                INIT_LIST_HEAD(&dev->ex_dev.children);
+                break;
+        case SATA_DEV:
+        case SATA_PM:
+        case SATA_PM_PORT:
+                INIT_LIST_HEAD(&dev->sata_dev.children);
+                break;
+        default:
+                break;
+        }
+}
+
+static void sas_task_timedout(unsigned long _task)
+{
+	struct sas_task *task = (void *) _task;
+	unsigned long flags;
+
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	if (!(task->task_state_flags & SAS_TASK_STATE_DONE))
+		task->task_state_flags |= SAS_TASK_STATE_ABORTED;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	complete(&task->completion);
+}
+
+static void sas_disc_task_done(struct sas_task *task)
+{
+	if (!del_timer(&task->timer))
+		return;
+	complete(&task->completion);
+}
+
+#define SAS_DEV_TIMEOUT 10
+
+/**
+ * sas_execute_task -- Basic task processing for discovery
+ * @task: the task to be executed
+ * @buffer: pointer to buffer to do I/O
+ * @size: size of @buffer
+ * @pci_dma_dir: PCI_DMA_...
+ */
+static int sas_execute_task(struct sas_task *task, void *buffer, int size,
+			    int pci_dma_dir)
+{
+	int res = 0;
+	struct scatterlist *scatter = NULL;
+	struct task_status_struct *ts = &task->task_status;
+	int num_scatter = 0;
+	int retries = 0;
+	struct sas_internal *i =
+		to_sas_internal(task->dev->port->ha->core.shost->transportt);
+
+	if (pci_dma_dir != PCI_DMA_NONE) {
+		scatter = kzalloc(sizeof(*scatter), GFP_KERNEL);
+		if (!scatter)
+			goto out;
+
+		sg_init_one(scatter, buffer, size);
+		num_scatter = 1;
+	}
+
+	task->task_proto = task->dev->tproto;
+	task->scatter = scatter;
+	task->num_scatter = num_scatter;
+	task->total_xfer_len = size;
+	task->data_dir = pci_dma_dir;
+	task->task_done = sas_disc_task_done;
+
+	for (retries = 0; retries < 5; retries++) {
+		task->task_state_flags = SAS_TASK_STATE_PENDING;
+		init_completion(&task->completion);
+
+		task->timer.data = (unsigned long) task;
+		task->timer.function = sas_task_timedout;
+		task->timer.expires = jiffies + SAS_DEV_TIMEOUT*HZ;
+		add_timer(&task->timer);
+
+		res = i->dft->lldd_execute_task(task, 1, GFP_KERNEL);
+		if (res) {
+			del_timer(&task->timer);
+			SAS_DPRINTK("executing SAS discovery task failed:%d\n",
+				    res);
+			goto ex_err;
+		}
+		wait_for_completion(&task->completion);
+		res = -ETASK;
+		if (task->task_state_flags & SAS_TASK_STATE_ABORTED) {
+			int res2;
+			SAS_DPRINTK("task aborted, flags:0x%x\n",
+				    task->task_state_flags);
+			res2 = i->dft->lldd_abort_task(task);
+			SAS_DPRINTK("came back from abort task\n");
+			if (!(task->task_state_flags & SAS_TASK_STATE_DONE)) {
+				if (res2 == TMF_RESP_FUNC_COMPLETE)
+					continue; /* Retry the task */
+				else
+					goto ex_err;
+			}
+		}
+		if (task->task_status.stat == SAM_BUSY ||
+			   task->task_status.stat == SAM_TASK_SET_FULL ||
+			   task->task_status.stat == SAS_QUEUE_FULL) {
+			SAS_DPRINTK("task: q busy, sleeping...\n");
+			schedule_timeout_interruptible(HZ);
+		} else if (task->task_status.stat == SAM_CHECK_COND) {
+			struct scsi_sense_hdr shdr;
+
+			if (!scsi_normalize_sense(ts->buf, ts->buf_valid_size,
+						  &shdr)) {
+				SAS_DPRINTK("couldn't normalize sense\n");
+				continue;
+			}
+			if ((shdr.sense_key == 6 && shdr.asc == 0x29) ||
+			    (shdr.sense_key == 2 && shdr.asc == 4 &&
+			     shdr.ascq == 1)) {
+				SAS_DPRINTK("device %016llx LUN: %016llx "
+					    "powering up or not ready yet, "
+					    "sleeping...\n",
+					    SAS_ADDR(task->dev->sas_addr),
+					    SAS_ADDR(task->ssp_task.LUN));
+
+				schedule_timeout_interruptible(5*HZ);
+			} else if (shdr.sense_key == 1) {
+				res = 0;
+				break;
+			} else if (shdr.sense_key == 5) {
+				break;
+			} else {
+				SAS_DPRINTK("dev %016llx LUN: %016llx "
+					    "sense key:0x%x ASC:0x%x ASCQ:0x%x"
+					    "\n",
+					    SAS_ADDR(task->dev->sas_addr),
+					    SAS_ADDR(task->ssp_task.LUN),
+					    shdr.sense_key,
+					    shdr.asc, shdr.ascq);
+			}
+		} else if (task->task_status.resp != SAS_TASK_COMPLETE ||
+			   task->task_status.stat != SAM_GOOD) {
+			SAS_DPRINTK("task finished with resp:0x%x, "
+				    "stat:0x%x\n",
+				    task->task_status.resp,
+				    task->task_status.stat);
+			goto ex_err;
+		} else {
+			res = 0;
+			break;
+		}
+	}
+ex_err:
+	if (pci_dma_dir != PCI_DMA_NONE)
+		kfree(scatter);
+out:
+	return res;
+}
+
+/* ---------- Domain device discovery ---------- */
+
+/**
+ * sas_get_port_device -- Discover devices which caused port creation
+ * @port: pointer to struct sas_port of interest
+ *
+ * Devices directly attached to a HA port, have no parent.  This is
+ * how we know they are (domain) "root" devices.  All other devices
+ * do, and should have their "parent" pointer set appropriately as
+ * soon as a child device is discovered.
+ */
+static int sas_get_port_device(struct asd_sas_port *port)
+{
+	unsigned long flags;
+	struct asd_sas_phy *phy;
+	struct sas_rphy *rphy;
+	struct domain_device *dev;
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&port->phy_list_lock, flags);
+	if (list_empty(&port->phy_list)) {
+		spin_unlock_irqrestore(&port->phy_list_lock, flags);
+		kfree(dev);
+		return -ENODEV;
+	}
+	phy = container_of(port->phy_list.next, struct asd_sas_phy, port_phy_el);
+	spin_lock(&phy->frame_rcvd_lock);
+	memcpy(dev->frame_rcvd, phy->frame_rcvd, min(sizeof(dev->frame_rcvd),
+					     (size_t)phy->frame_rcvd_size));
+	spin_unlock(&phy->frame_rcvd_lock);
+	spin_unlock_irqrestore(&port->phy_list_lock, flags);
+
+	if (dev->frame_rcvd[0] == 0x34 && port->oob_mode == SATA_OOB_MODE) {
+		struct dev_to_host_fis *fis =
+			(struct dev_to_host_fis *) dev->frame_rcvd;
+		if (fis->interrupt_reason == 1 && fis->lbal == 1 &&
+		    fis->byte_count_low==0x69 && fis->byte_count_high == 0x96
+		    && (fis->device & ~0x10) == 0)
+			dev->dev_type = SATA_PM;
+		else
+			dev->dev_type = SATA_DEV;
+		dev->tproto = SATA_PROTO;
+	} else {
+		struct sas_identify_frame *id =
+			(struct sas_identify_frame *) dev->frame_rcvd;
+		dev->dev_type = id->dev_type;
+		dev->iproto = id->initiator_bits;
+		dev->tproto = id->target_bits;
+	}
+
+	sas_init_dev(dev);
+
+	switch (dev->dev_type) {
+	case SAS_END_DEV:
+		rphy = sas_end_device_alloc(port->port);
+		break;
+	case EDGE_DEV:
+		rphy = sas_expander_alloc(port->port,
+					  SAS_EDGE_EXPANDER_DEVICE);
+		break;
+	case FANOUT_DEV:
+		rphy = sas_expander_alloc(port->port,
+					  SAS_FANOUT_EXPANDER_DEVICE);
+		break;
+	case SATA_DEV:
+	default:
+		printk("ERROR: Unidentified device type %d\n", dev->dev_type);
+		rphy = NULL;
+		break;
+	}
+
+	if (!rphy) {
+		kfree(dev);
+		return -ENODEV;
+	}
+	rphy->identify.phy_identifier = phy->phy->identify.phy_identifier;
+	memcpy(dev->sas_addr, port->attached_sas_addr, SAS_ADDR_SIZE);
+	sas_fill_in_rphy(dev, rphy);
+	sas_hash_addr(dev->hashed_sas_addr, dev->sas_addr);
+	port->port_dev = dev;
+	dev->port = port;
+	dev->linkrate = port->linkrate;
+	dev->min_linkrate = port->linkrate;
+	dev->max_linkrate = port->linkrate;
+	dev->pathways = port->num_phys;
+	memset(port->disc.fanout_sas_addr, 0, SAS_ADDR_SIZE);
+	memset(port->disc.eeds_a, 0, SAS_ADDR_SIZE);
+	memset(port->disc.eeds_b, 0, SAS_ADDR_SIZE);
+	port->disc.max_level = 0;
+
+	dev->rphy = rphy;
+	spin_lock(&port->dev_list_lock);
+	list_add_tail(&dev->dev_list_node, &port->dev_list);
+	spin_unlock(&port->dev_list_lock);
+
+	return 0;
+}
+
+/* ---------- Discover and Revalidate ---------- */
+
+/* ---------- SATA ---------- */
+
+static void sas_get_ata_command_set(struct domain_device *dev)
+{
+	struct dev_to_host_fis *fis =
+		(struct dev_to_host_fis *) dev->frame_rcvd;
+
+	if ((fis->sector_count == 1 && /* ATA */
+	     fis->lbal         == 1 &&
+	     fis->lbam         == 0 &&
+	     fis->lbah         == 0 &&
+	     fis->device       == 0)
+	    ||
+	    (fis->sector_count == 0 && /* CE-ATA (mATA) */
+	     fis->lbal         == 0 &&
+	     fis->lbam         == 0xCE &&
+	     fis->lbah         == 0xAA &&
+	     (fis->device & ~0x10) == 0))
+
+		dev->sata_dev.command_set = ATA_COMMAND_SET;
+
+	else if ((fis->interrupt_reason == 1 &&	/* ATAPI */
+		  fis->lbal             == 1 &&
+		  fis->byte_count_low   == 0x14 &&
+		  fis->byte_count_high  == 0xEB &&
+		  (fis->device & ~0x10) == 0))
+
+		dev->sata_dev.command_set = ATAPI_COMMAND_SET;
+
+	else if ((fis->sector_count == 1 && /* SEMB */
+		  fis->lbal         == 1 &&
+		  fis->lbam         == 0x3C &&
+		  fis->lbah         == 0xC3 &&
+		  fis->device       == 0)
+		||
+		 (fis->interrupt_reason == 1 &&	/* SATA PM */
+		  fis->lbal             == 1 &&
+		  fis->byte_count_low   == 0x69 &&
+		  fis->byte_count_high  == 0x96 &&
+		  (fis->device & ~0x10) == 0))
+
+		/* Treat it as a superset? */
+		dev->sata_dev.command_set = ATAPI_COMMAND_SET;
+}
+
+/**
+ * sas_issue_ata_cmd -- Basic SATA command processing for discovery
+ * @dev: the device to send the command to
+ * @command: the command register
+ * @features: the features register
+ * @buffer: pointer to buffer to do I/O
+ * @size: size of @buffer
+ * @pci_dma_dir: PCI_DMA_...
+ */
+static int sas_issue_ata_cmd(struct domain_device *dev, u8 command,
+			     u8 features, void *buffer, int size,
+			     int pci_dma_dir)
+{
+	int res = 0;
+	struct sas_task *task;
+	struct dev_to_host_fis *d2h_fis = (struct dev_to_host_fis *)
+		&dev->frame_rcvd[0];
+
+	res = -ENOMEM;
+	task = sas_alloc_task(GFP_KERNEL);
+	if (!task)
+		goto out;
+
+	task->dev = dev;
+
+	task->ata_task.fis.command = command;
+	task->ata_task.fis.features = features;
+	task->ata_task.fis.device = d2h_fis->device;
+	task->ata_task.retry_count = 1;
+
+	res = sas_execute_task(task, buffer, size, pci_dma_dir);
+
+	sas_free_task(task);
+out:
+	return res;
+}
+
+static void sas_sata_propagate_sas_addr(struct domain_device *dev)
+{
+	unsigned long flags;
+	struct asd_sas_port *port = dev->port;
+	struct asd_sas_phy  *phy;
+
+	BUG_ON(dev->parent);
+
+	memcpy(port->attached_sas_addr, dev->sas_addr, SAS_ADDR_SIZE);
+	spin_lock_irqsave(&port->phy_list_lock, flags);
+	list_for_each_entry(phy, &port->phy_list, port_phy_el)
+		memcpy(phy->attached_sas_addr, dev->sas_addr, SAS_ADDR_SIZE);
+	spin_unlock_irqrestore(&port->phy_list_lock, flags);
+}
+
+#define ATA_IDENTIFY_DEV         0xEC
+#define ATA_IDENTIFY_PACKET_DEV  0xA1
+#define ATA_SET_FEATURES         0xEF
+#define ATA_FEATURE_PUP_STBY_SPIN_UP 0x07
+
+/**
+ * sas_discover_sata_dev -- discover a STP/SATA device (SATA_DEV)
+ * @dev: STP/SATA device of interest (ATA/ATAPI)
+ *
+ * The LLDD has already been notified of this device, so that we can
+ * send FISes to it.  Here we try to get IDENTIFY DEVICE or IDENTIFY
+ * PACKET DEVICE, if ATAPI device, so that the LLDD can fine-tune its
+ * performance for this device.
+ */
+static int sas_discover_sata_dev(struct domain_device *dev)
+{
+	int     res;
+	__le16  *identify_x;
+	u8      command;
+
+	identify_x = kzalloc(512, GFP_KERNEL);
+	if (!identify_x)
+		return -ENOMEM;
+
+	if (dev->sata_dev.command_set == ATA_COMMAND_SET) {
+		dev->sata_dev.identify_device = identify_x;
+		command = ATA_IDENTIFY_DEV;
+	} else {
+		dev->sata_dev.identify_packet_device = identify_x;
+		command = ATA_IDENTIFY_PACKET_DEV;
+	}
+
+	res = sas_issue_ata_cmd(dev, command, 0, identify_x, 512,
+				PCI_DMA_FROMDEVICE);
+	if (res)
+		goto out_err;
+
+	/* lives on the media? */
+	if (le16_to_cpu(identify_x[0]) & 4) {
+		/* incomplete response */
+		SAS_DPRINTK("sending SET FEATURE/PUP_STBY_SPIN_UP to "
+			    "dev %llx\n", SAS_ADDR(dev->sas_addr));
+		if (!le16_to_cpu(identify_x[83] & (1<<6)))
+			goto cont1;
+		res = sas_issue_ata_cmd(dev, ATA_SET_FEATURES,
+					ATA_FEATURE_PUP_STBY_SPIN_UP,
+					NULL, 0, PCI_DMA_NONE);
+		if (res)
+			goto cont1;
+
+		schedule_timeout_interruptible(5*HZ); /* More time? */
+		res = sas_issue_ata_cmd(dev, command, 0, identify_x, 512,
+					PCI_DMA_FROMDEVICE);
+		if (res)
+			goto out_err;
+	}
+cont1:
+	/* Get WWN */
+	if (dev->port->oob_mode != SATA_OOB_MODE) {
+		memcpy(dev->sas_addr, dev->sata_dev.rps_resp.rps.stp_sas_addr,
+		       SAS_ADDR_SIZE);
+	} else if (dev->sata_dev.command_set == ATA_COMMAND_SET &&
+		   (le16_to_cpu(dev->sata_dev.identify_device[108]) & 0xF000)
+		   == 0x5000) {
+		int i;
+
+		for (i = 0; i < 4; i++) {
+			dev->sas_addr[2*i] =
+	     (le16_to_cpu(dev->sata_dev.identify_device[108+i]) & 0xFF00) >> 8;
+			dev->sas_addr[2*i+1] =
+	      le16_to_cpu(dev->sata_dev.identify_device[108+i]) & 0x00FF;
+		}
+	}
+	sas_hash_addr(dev->hashed_sas_addr, dev->sas_addr);
+	if (!dev->parent)
+		sas_sata_propagate_sas_addr(dev);
+
+	/* XXX Hint: register this SATA device with SATL.
+	   When this returns, dev->sata_dev->lu is alive and
+	   present.
+	sas_satl_register_dev(dev);
+	*/
+	return 0;
+out_err:
+	dev->sata_dev.identify_packet_device = NULL;
+	dev->sata_dev.identify_device = NULL;
+	kfree(identify_x);
+	return res;
+}
+
+static int sas_discover_sata_pm(struct domain_device *dev)
+{
+	return -ENODEV;
+}
+
+int sas_notify_lldd_dev_found(struct domain_device *dev)
+{
+	int res = 0;
+	struct sas_ha_struct *sas_ha = dev->port->ha;
+	struct Scsi_Host *shost = sas_ha->core.shost;
+	struct sas_internal *i = to_sas_internal(shost->transportt);
+
+	if (i->dft->lldd_dev_found) {
+		res = i->dft->lldd_dev_found(dev);
+		if (res) {
+			printk("sas: driver on pcidev %s cannot handle "
+			       "device %llx, error:%d\n",
+			       pci_name(sas_ha->pcidev),
+			       SAS_ADDR(dev->sas_addr), res);
+		}
+	}
+	return res;
+}
+
+
+void sas_notify_lldd_dev_gone(struct domain_device *dev)
+{
+	struct sas_ha_struct *sas_ha = dev->port->ha;
+	struct Scsi_Host *shost = sas_ha->core.shost;
+	struct sas_internal *i = to_sas_internal(shost->transportt);
+
+	if (i->dft->lldd_dev_gone)
+		i->dft->lldd_dev_gone(dev);
+}
+
+/* ---------- Common/dispatchers ---------- */
+
+/**
+ * sas_discover_sata -- discover an STP/SATA domain device
+ * @dev: pointer to struct domain_device of interest
+ *
+ * First we notify the LLDD of this device, so we can send frames to
+ * it.  Then depending on the type of device we call the appropriate
+ * discover functions.  Once device discover is done, we notify the
+ * LLDD so that it can fine-tune its parameters for the device, by
+ * removing it and then adding it.  That is, the second time around,
+ * the driver would have certain fields, that it is looking at, set.
+ * Finally we initialize the kobj so that the device can be added to
+ * the system at registration time.  Devices directly attached to a HA
+ * port, have no parents.  All other devices do, and should have their
+ * "parent" pointer set appropriately before calling this function.
+ */
+int sas_discover_sata(struct domain_device *dev)
+{
+	int res;
+
+	sas_get_ata_command_set(dev);
+
+	res = sas_notify_lldd_dev_found(dev);
+	if (res)
+		return res;
+
+	switch (dev->dev_type) {
+	case SATA_DEV:
+		res = sas_discover_sata_dev(dev);
+		break;
+	case SATA_PM:
+		res = sas_discover_sata_pm(dev);
+		break;
+	default:
+		break;
+	}
+
+	sas_notify_lldd_dev_gone(dev);
+	if (!res) {
+		sas_notify_lldd_dev_found(dev);
+	}
+	return res;
+}
+
+/**
+ * sas_discover_end_dev -- discover an end device (SSP, etc)
+ * @end: pointer to domain device of interest
+ *
+ * See comment in sas_discover_sata().
+ */
+int sas_discover_end_dev(struct domain_device *dev)
+{
+	int res;
+
+	res = sas_notify_lldd_dev_found(dev);
+	if (res)
+		return res;
+
+	res = sas_rphy_add(dev->rphy);
+	if (res)
+		goto out_err;
+
+	/* do this to get the end device port attributes which will have
+	 * been scanned in sas_rphy_add */
+	sas_notify_lldd_dev_gone(dev);
+	sas_notify_lldd_dev_found(dev);
+
+	return 0;
+
+out_err:
+	sas_notify_lldd_dev_gone(dev);
+	return res;
+}
+
+/* ---------- Device registration and unregistration ---------- */
+
+static inline void sas_unregister_common_dev(struct domain_device *dev)
+{
+	sas_notify_lldd_dev_gone(dev);
+	if (!dev->parent)
+		dev->port->port_dev = NULL;
+	else
+		list_del_init(&dev->siblings);
+	list_del_init(&dev->dev_list_node);
+}
+
+void sas_unregister_dev(struct domain_device *dev)
+{
+	if (dev->rphy) {
+		sas_remove_children(&dev->rphy->dev);
+		sas_rphy_delete(dev->rphy);
+		dev->rphy = NULL;
+	}
+	if (dev->dev_type == EDGE_DEV || dev->dev_type == FANOUT_DEV) {
+		/* remove the phys and ports, everything else should be gone */
+		kfree(dev->ex_dev.ex_phy);
+		dev->ex_dev.ex_phy = NULL;
+	}
+	sas_unregister_common_dev(dev);
+}
+
+void sas_unregister_domain_devices(struct asd_sas_port *port)
+{
+	struct domain_device *dev, *n;
+
+	list_for_each_entry_safe_reverse(dev,n,&port->dev_list,dev_list_node)
+		sas_unregister_dev(dev);
+
+	port->port->rphy = NULL;
+
+}
+
+/* ---------- Discovery and Revalidation ---------- */
+
+/**
+ * sas_discover_domain -- discover the domain
+ * @port: port to the domain of interest
+ *
+ * NOTE: this process _must_ quit (return) as soon as any connection
+ * errors are encountered.  Connection recovery is done elsewhere.
+ * Discover process only interrogates devices in order to discover the
+ * domain.
+ */
+static void sas_discover_domain(void *data)
+{
+	int error = 0;
+	struct asd_sas_port *port = data;
+
+	sas_begin_event(DISCE_DISCOVER_DOMAIN, &port->disc.disc_event_lock,
+			&port->disc.pending);
+
+	if (port->port_dev)
+		return ;
+	else {
+		error = sas_get_port_device(port);
+		if (error)
+			return;
+	}
+
+	SAS_DPRINTK("DOING DISCOVERY on port %d, pid:%d\n", port->id,
+		    current->pid);
+
+	switch (port->port_dev->dev_type) {
+	case SAS_END_DEV:
+		error = sas_discover_end_dev(port->port_dev);
+		break;
+	case EDGE_DEV:
+	case FANOUT_DEV:
+		error = sas_discover_root_expander(port->port_dev);
+		break;
+	case SATA_DEV:
+	case SATA_PM:
+		error = sas_discover_sata(port->port_dev);
+		break;
+	default:
+		SAS_DPRINTK("unhandled device %d\n", port->port_dev->dev_type);
+		break;
+	}
+
+	if (error) {
+		kfree(port->port_dev); /* not kobject_register-ed yet */
+		port->port_dev = NULL;
+	}
+
+	SAS_DPRINTK("DONE DISCOVERY on port %d, pid:%d, result:%d\n", port->id,
+		    current->pid, error);
+}
+
+static void sas_revalidate_domain(void *data)
+{
+	int res = 0;
+	struct asd_sas_port *port = data;
+
+	sas_begin_event(DISCE_REVALIDATE_DOMAIN, &port->disc.disc_event_lock,
+			&port->disc.pending);
+
+	SAS_DPRINTK("REVALIDATING DOMAIN on port %d, pid:%d\n", port->id,
+		    current->pid);
+	if (port->port_dev)
+		res = sas_ex_revalidate_domain(port->port_dev);
+
+	SAS_DPRINTK("done REVALIDATING DOMAIN on port %d, pid:%d, res 0x%x\n",
+		    port->id, current->pid, res);
+}
+
+/* ---------- Events ---------- */
+
+int sas_discover_event(struct asd_sas_port *port, enum discover_event ev)
+{
+	struct sas_discovery *disc;
+
+	if (!port)
+		return 0;
+	disc = &port->disc;
+
+	BUG_ON(ev >= DISC_NUM_EVENTS);
+
+	sas_queue_event(ev, &disc->disc_event_lock, &disc->pending,
+			&disc->disc_work[ev], port->ha->core.shost);
+
+	return 0;
+}
+
+/**
+ * sas_init_disc -- initialize the discovery struct in the port
+ * @port: pointer to struct port
+ *
+ * Called when the ports are being initialized.
+ */
+void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *port)
+{
+	int i;
+
+	static void (*sas_event_fns[DISC_NUM_EVENTS])(void *) = {
+		[DISCE_DISCOVER_DOMAIN] = sas_discover_domain,
+		[DISCE_REVALIDATE_DOMAIN] = sas_revalidate_domain,
+	};
+
+	spin_lock_init(&disc->disc_event_lock);
+	disc->pending = 0;
+	for (i = 0; i < DISC_NUM_EVENTS; i++)
+		INIT_WORK(&disc->disc_work[i], sas_event_fns[i], port);
+}
diff --git a/drivers/scsi/libsas/sas_dump.c b/drivers/scsi/libsas/sas_dump.c
new file mode 100644
index 000000000000..f1246d2c9bef
--- /dev/null
+++ b/drivers/scsi/libsas/sas_dump.c
@@ -0,0 +1,76 @@
+/*
+ * Serial Attached SCSI (SAS) Dump/Debugging routines
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include "sas_dump.h"
+
+#ifdef SAS_DEBUG
+
+static const char *sas_hae_str[] = {
+	[0] = "HAE_RESET",
+};
+
+static const char *sas_porte_str[] = {
+	[0] = "PORTE_BYTES_DMAED",
+	[1] = "PORTE_BROADCAST_RCVD",
+	[2] = "PORTE_LINK_RESET_ERR",
+	[3] = "PORTE_TIMER_EVENT",
+	[4] = "PORTE_HARD_RESET",
+};
+
+static const char *sas_phye_str[] = {
+	[0] = "PHYE_LOSS_OF_SIGNAL",
+	[1] = "PHYE_OOB_DONE",
+	[2] = "PHYE_OOB_ERROR",
+	[3] = "PHYE_SPINUP_HOLD",
+};
+
+void sas_dprint_porte(int phyid, enum port_event pe)
+{
+	SAS_DPRINTK("phy%d: port event: %s\n", phyid, sas_porte_str[pe]);
+}
+void sas_dprint_phye(int phyid, enum phy_event pe)
+{
+	SAS_DPRINTK("phy%d: phy event: %s\n", phyid, sas_phye_str[pe]);
+}
+
+void sas_dprint_hae(struct sas_ha_struct *sas_ha, enum ha_event he)
+{
+	SAS_DPRINTK("ha %s: %s event\n", pci_name(sas_ha->pcidev),
+		    sas_hae_str[he]);
+}
+
+void sas_dump_port(struct asd_sas_port *port)
+{
+	SAS_DPRINTK("port%d: class:0x%x\n", port->id, port->class);
+	SAS_DPRINTK("port%d: sas_addr:%llx\n", port->id,
+		    SAS_ADDR(port->sas_addr));
+	SAS_DPRINTK("port%d: attached_sas_addr:%llx\n", port->id,
+		    SAS_ADDR(port->attached_sas_addr));
+	SAS_DPRINTK("port%d: iproto:0x%x\n", port->id, port->iproto);
+	SAS_DPRINTK("port%d: tproto:0x%x\n", port->id, port->tproto);
+	SAS_DPRINTK("port%d: oob_mode:0x%x\n", port->id, port->oob_mode);
+	SAS_DPRINTK("port%d: num_phys:%d\n", port->id, port->num_phys);
+}
+
+#endif /* SAS_DEBUG */
diff --git a/drivers/scsi/libsas/sas_dump.h b/drivers/scsi/libsas/sas_dump.h
new file mode 100644
index 000000000000..47b45d4f5258
--- /dev/null
+++ b/drivers/scsi/libsas/sas_dump.h
@@ -0,0 +1,42 @@
+/*
+ * Serial Attached SCSI (SAS) Dump/Debugging routines header file
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include "sas_internal.h"
+
+#ifdef SAS_DEBUG
+
+void sas_dprint_porte(int phyid, enum port_event pe);
+void sas_dprint_phye(int phyid, enum phy_event pe);
+void sas_dprint_hae(struct sas_ha_struct *sas_ha, enum ha_event he);
+void sas_dump_port(struct asd_sas_port *port);
+
+#else /* SAS_DEBUG */
+
+static inline void sas_dprint_porte(int phyid, enum port_event pe) { }
+static inline void sas_dprint_phye(int phyid, enum phy_event pe) { }
+static inline void sas_dprint_hae(struct sas_ha_struct *sas_ha,
+				  enum ha_event he) { }
+static inline void sas_dump_port(struct asd_sas_port *port) { }
+
+#endif /* SAS_DEBUG */
diff --git a/drivers/scsi/libsas/sas_event.c b/drivers/scsi/libsas/sas_event.c
new file mode 100644
index 000000000000..19110ed1c89c
--- /dev/null
+++ b/drivers/scsi/libsas/sas_event.c
@@ -0,0 +1,75 @@
+/*
+ * Serial Attached SCSI (SAS) Event processing
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <scsi/scsi_host.h>
+#include "sas_internal.h"
+#include "sas_dump.h"
+
+static void notify_ha_event(struct sas_ha_struct *sas_ha, enum ha_event event)
+{
+	BUG_ON(event >= HA_NUM_EVENTS);
+
+	sas_queue_event(event, &sas_ha->event_lock, &sas_ha->pending,
+			&sas_ha->ha_events[event], sas_ha->core.shost);
+}
+
+static void notify_port_event(struct asd_sas_phy *phy, enum port_event event)
+{
+	struct sas_ha_struct *ha = phy->ha;
+
+	BUG_ON(event >= PORT_NUM_EVENTS);
+
+	sas_queue_event(event, &ha->event_lock, &phy->port_events_pending,
+			&phy->port_events[event], ha->core.shost);
+}
+
+static void notify_phy_event(struct asd_sas_phy *phy, enum phy_event event)
+{
+	struct sas_ha_struct *ha = phy->ha;
+
+	BUG_ON(event >= PHY_NUM_EVENTS);
+
+	sas_queue_event(event, &ha->event_lock, &phy->phy_events_pending,
+			&phy->phy_events[event], ha->core.shost);
+}
+
+int sas_init_events(struct sas_ha_struct *sas_ha)
+{
+	static void (*sas_ha_event_fns[HA_NUM_EVENTS])(void *) = {
+		[HAE_RESET] = sas_hae_reset,
+	};
+
+	int i;
+
+	spin_lock_init(&sas_ha->event_lock);
+
+	for (i = 0; i < HA_NUM_EVENTS; i++)
+		INIT_WORK(&sas_ha->ha_events[i], sas_ha_event_fns[i], sas_ha);
+
+	sas_ha->notify_ha_event = notify_ha_event;
+	sas_ha->notify_port_event = notify_port_event;
+	sas_ha->notify_phy_event = notify_phy_event;
+
+	return 0;
+}
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
new file mode 100644
index 000000000000..b653a263f76a
--- /dev/null
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -0,0 +1,1862 @@
+/*
+ * Serial Attached SCSI (SAS) Expander discovery and configuration
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/pci.h>
+#include <linux/scatterlist.h>
+
+#include "sas_internal.h"
+
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+#include "../scsi_sas_internal.h"
+
+static int sas_discover_expander(struct domain_device *dev);
+static int sas_configure_routing(struct domain_device *dev, u8 *sas_addr);
+static int sas_configure_phy(struct domain_device *dev, int phy_id,
+			     u8 *sas_addr, int include);
+static int sas_disable_routing(struct domain_device *dev,  u8 *sas_addr);
+
+#if 0
+/* FIXME: smp needs to migrate into the sas class */
+static ssize_t smp_portal_read(struct kobject *, char *, loff_t, size_t);
+static ssize_t smp_portal_write(struct kobject *, char *, loff_t, size_t);
+#endif
+
+/* ---------- SMP task management ---------- */
+
+static void smp_task_timedout(unsigned long _task)
+{
+	struct sas_task *task = (void *) _task;
+	unsigned long flags;
+
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	if (!(task->task_state_flags & SAS_TASK_STATE_DONE))
+		task->task_state_flags |= SAS_TASK_STATE_ABORTED;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	complete(&task->completion);
+}
+
+static void smp_task_done(struct sas_task *task)
+{
+	if (!del_timer(&task->timer))
+		return;
+	complete(&task->completion);
+}
+
+/* Give it some long enough timeout. In seconds. */
+#define SMP_TIMEOUT 10
+
+static int smp_execute_task(struct domain_device *dev, void *req, int req_size,
+			    void *resp, int resp_size)
+{
+	int res;
+	struct sas_task *task = sas_alloc_task(GFP_KERNEL);
+	struct sas_internal *i =
+		to_sas_internal(dev->port->ha->core.shost->transportt);
+
+	if (!task)
+		return -ENOMEM;
+
+	task->dev = dev;
+	task->task_proto = dev->tproto;
+	sg_init_one(&task->smp_task.smp_req, req, req_size);
+	sg_init_one(&task->smp_task.smp_resp, resp, resp_size);
+
+	task->task_done = smp_task_done;
+
+	task->timer.data = (unsigned long) task;
+	task->timer.function = smp_task_timedout;
+	task->timer.expires = jiffies + SMP_TIMEOUT*HZ;
+	add_timer(&task->timer);
+
+	res = i->dft->lldd_execute_task(task, 1, GFP_KERNEL);
+
+	if (res) {
+		del_timer(&task->timer);
+		SAS_DPRINTK("executing SMP task failed:%d\n", res);
+		goto ex_err;
+	}
+
+	wait_for_completion(&task->completion);
+	res = -ETASK;
+	if ((task->task_state_flags & SAS_TASK_STATE_ABORTED)) {
+		SAS_DPRINTK("smp task timed out or aborted\n");
+		i->dft->lldd_abort_task(task);
+		if (!(task->task_state_flags & SAS_TASK_STATE_DONE)) {
+			SAS_DPRINTK("SMP task aborted and not done\n");
+			goto ex_err;
+		}
+	}
+	if (task->task_status.resp == SAS_TASK_COMPLETE &&
+	    task->task_status.stat == SAM_GOOD)
+		res = 0;
+	else
+		SAS_DPRINTK("%s: task to dev %016llx response: 0x%x "
+			    "status 0x%x\n", __FUNCTION__,
+			    SAS_ADDR(dev->sas_addr),
+			    task->task_status.resp,
+			    task->task_status.stat);
+ex_err:
+	sas_free_task(task);
+	return res;
+}
+
+/* ---------- Allocations ---------- */
+
+static inline void *alloc_smp_req(int size)
+{
+	u8 *p = kzalloc(size, GFP_KERNEL);
+	if (p)
+		p[0] = SMP_REQUEST;
+	return p;
+}
+
+static inline void *alloc_smp_resp(int size)
+{
+	return kzalloc(size, GFP_KERNEL);
+}
+
+/* ---------- Expander configuration ---------- */
+
+static void sas_set_ex_phy(struct domain_device *dev, int phy_id,
+			   void *disc_resp)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *phy = &ex->ex_phy[phy_id];
+	struct smp_resp *resp = disc_resp;
+	struct discover_resp *dr = &resp->disc;
+	struct sas_rphy *rphy = dev->rphy;
+	int rediscover = (phy->phy != NULL);
+
+	if (!rediscover) {
+		phy->phy = sas_phy_alloc(&rphy->dev, phy_id);
+
+		/* FIXME: error_handling */
+		BUG_ON(!phy->phy);
+	}
+
+	switch (resp->result) {
+	case SMP_RESP_PHY_VACANT:
+		phy->phy_state = PHY_VACANT;
+		return;
+	default:
+		phy->phy_state = PHY_NOT_PRESENT;
+		return;
+	case SMP_RESP_FUNC_ACC:
+		phy->phy_state = PHY_EMPTY; /* do not know yet */
+		break;
+	}
+
+	phy->phy_id = phy_id;
+	phy->attached_dev_type = dr->attached_dev_type;
+	phy->linkrate = dr->linkrate;
+	phy->attached_sata_host = dr->attached_sata_host;
+	phy->attached_sata_dev  = dr->attached_sata_dev;
+	phy->attached_sata_ps   = dr->attached_sata_ps;
+	phy->attached_iproto = dr->iproto << 1;
+	phy->attached_tproto = dr->tproto << 1;
+	memcpy(phy->attached_sas_addr, dr->attached_sas_addr, SAS_ADDR_SIZE);
+	phy->attached_phy_id = dr->attached_phy_id;
+	phy->phy_change_count = dr->change_count;
+	phy->routing_attr = dr->routing_attr;
+	phy->virtual = dr->virtual;
+	phy->last_da_index = -1;
+
+	phy->phy->identify.initiator_port_protocols = phy->attached_iproto;
+	phy->phy->identify.target_port_protocols = phy->attached_tproto;
+	phy->phy->identify.phy_identifier = phy_id;
+	phy->phy->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS;
+	phy->phy->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS;
+	phy->phy->minimum_linkrate = SAS_LINK_RATE_1_5_GBPS;
+	phy->phy->maximum_linkrate = SAS_LINK_RATE_3_0_GBPS;
+	switch (phy->linkrate) {
+	case PHY_LINKRATE_1_5:
+		phy->phy->negotiated_linkrate = SAS_LINK_RATE_1_5_GBPS;
+		break;
+	case PHY_LINKRATE_3:
+		phy->phy->negotiated_linkrate = SAS_LINK_RATE_3_0_GBPS;
+		break;
+	case PHY_LINKRATE_6:
+		phy->phy->negotiated_linkrate = SAS_LINK_RATE_6_0_GBPS;
+		break;
+	default:
+		phy->phy->negotiated_linkrate = SAS_LINK_RATE_UNKNOWN;
+		break;
+	}
+
+	if (!rediscover)
+		sas_phy_add(phy->phy);
+
+	SAS_DPRINTK("ex %016llx phy%02d:%c attached: %016llx\n",
+		    SAS_ADDR(dev->sas_addr), phy->phy_id,
+		    phy->routing_attr == TABLE_ROUTING ? 'T' :
+		    phy->routing_attr == DIRECT_ROUTING ? 'D' :
+		    phy->routing_attr == SUBTRACTIVE_ROUTING ? 'S' : '?',
+		    SAS_ADDR(phy->attached_sas_addr));
+
+	return;
+}
+
+#define DISCOVER_REQ_SIZE  16
+#define DISCOVER_RESP_SIZE 56
+
+static int sas_ex_phy_discover(struct domain_device *dev, int single)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int  res = 0;
+	u8   *disc_req;
+	u8   *disc_resp;
+
+	disc_req = alloc_smp_req(DISCOVER_REQ_SIZE);
+	if (!disc_req)
+		return -ENOMEM;
+
+	disc_resp = alloc_smp_req(DISCOVER_RESP_SIZE);
+	if (!disc_resp) {
+		kfree(disc_req);
+		return -ENOMEM;
+	}
+
+	disc_req[1] = SMP_DISCOVER;
+
+	if (0 <= single && single < ex->num_phys) {
+		disc_req[9] = single;
+		res = smp_execute_task(dev, disc_req, DISCOVER_REQ_SIZE,
+				       disc_resp, DISCOVER_RESP_SIZE);
+		if (res)
+			goto out_err;
+		sas_set_ex_phy(dev, single, disc_resp);
+	} else {
+		int i;
+
+		for (i = 0; i < ex->num_phys; i++) {
+			disc_req[9] = i;
+			res = smp_execute_task(dev, disc_req,
+					       DISCOVER_REQ_SIZE, disc_resp,
+					       DISCOVER_RESP_SIZE);
+			if (res)
+				goto out_err;
+			sas_set_ex_phy(dev, i, disc_resp);
+		}
+	}
+out_err:
+	kfree(disc_resp);
+	kfree(disc_req);
+	return res;
+}
+
+static int sas_expander_discover(struct domain_device *dev)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int res = -ENOMEM;
+
+	ex->ex_phy = kzalloc(sizeof(*ex->ex_phy)*ex->num_phys, GFP_KERNEL);
+	if (!ex->ex_phy)
+		return -ENOMEM;
+
+	res = sas_ex_phy_discover(dev, -1);
+	if (res)
+		goto out_err;
+
+	return 0;
+ out_err:
+	kfree(ex->ex_phy);
+	ex->ex_phy = NULL;
+	return res;
+}
+
+#define MAX_EXPANDER_PHYS 128
+
+static void ex_assign_report_general(struct domain_device *dev,
+					    struct smp_resp *resp)
+{
+	struct report_general_resp *rg = &resp->rg;
+
+	dev->ex_dev.ex_change_count = be16_to_cpu(rg->change_count);
+	dev->ex_dev.max_route_indexes = be16_to_cpu(rg->route_indexes);
+	dev->ex_dev.num_phys = min(rg->num_phys, (u8)MAX_EXPANDER_PHYS);
+	dev->ex_dev.conf_route_table = rg->conf_route_table;
+	dev->ex_dev.configuring = rg->configuring;
+	memcpy(dev->ex_dev.enclosure_logical_id, rg->enclosure_logical_id, 8);
+}
+
+#define RG_REQ_SIZE   8
+#define RG_RESP_SIZE 32
+
+static int sas_ex_general(struct domain_device *dev)
+{
+	u8 *rg_req;
+	struct smp_resp *rg_resp;
+	int res;
+	int i;
+
+	rg_req = alloc_smp_req(RG_REQ_SIZE);
+	if (!rg_req)
+		return -ENOMEM;
+
+	rg_resp = alloc_smp_resp(RG_RESP_SIZE);
+	if (!rg_resp) {
+		kfree(rg_req);
+		return -ENOMEM;
+	}
+
+	rg_req[1] = SMP_REPORT_GENERAL;
+
+	for (i = 0; i < 5; i++) {
+		res = smp_execute_task(dev, rg_req, RG_REQ_SIZE, rg_resp,
+				       RG_RESP_SIZE);
+
+		if (res) {
+			SAS_DPRINTK("RG to ex %016llx failed:0x%x\n",
+				    SAS_ADDR(dev->sas_addr), res);
+			goto out;
+		} else if (rg_resp->result != SMP_RESP_FUNC_ACC) {
+			SAS_DPRINTK("RG:ex %016llx returned SMP result:0x%x\n",
+				    SAS_ADDR(dev->sas_addr), rg_resp->result);
+			res = rg_resp->result;
+			goto out;
+		}
+
+		ex_assign_report_general(dev, rg_resp);
+
+		if (dev->ex_dev.configuring) {
+			SAS_DPRINTK("RG: ex %llx self-configuring...\n",
+				    SAS_ADDR(dev->sas_addr));
+			schedule_timeout_interruptible(5*HZ);
+		} else
+			break;
+	}
+out:
+	kfree(rg_req);
+	kfree(rg_resp);
+	return res;
+}
+
+static void ex_assign_manuf_info(struct domain_device *dev, void
+					*_mi_resp)
+{
+	u8 *mi_resp = _mi_resp;
+	struct sas_rphy *rphy = dev->rphy;
+	struct sas_expander_device *edev = rphy_to_expander_device(rphy);
+
+	memcpy(edev->vendor_id, mi_resp + 12, SAS_EXPANDER_VENDOR_ID_LEN);
+	memcpy(edev->product_id, mi_resp + 20, SAS_EXPANDER_PRODUCT_ID_LEN);
+	memcpy(edev->product_rev, mi_resp + 36,
+	       SAS_EXPANDER_PRODUCT_REV_LEN);
+
+	if (mi_resp[8] & 1) {
+		memcpy(edev->component_vendor_id, mi_resp + 40,
+		       SAS_EXPANDER_COMPONENT_VENDOR_ID_LEN);
+		edev->component_id = mi_resp[48] << 8 | mi_resp[49];
+		edev->component_revision_id = mi_resp[50];
+	}
+}
+
+#define MI_REQ_SIZE   8
+#define MI_RESP_SIZE 64
+
+static int sas_ex_manuf_info(struct domain_device *dev)
+{
+	u8 *mi_req;
+	u8 *mi_resp;
+	int res;
+
+	mi_req = alloc_smp_req(MI_REQ_SIZE);
+	if (!mi_req)
+		return -ENOMEM;
+
+	mi_resp = alloc_smp_resp(MI_RESP_SIZE);
+	if (!mi_resp) {
+		kfree(mi_req);
+		return -ENOMEM;
+	}
+
+	mi_req[1] = SMP_REPORT_MANUF_INFO;
+
+	res = smp_execute_task(dev, mi_req, MI_REQ_SIZE, mi_resp,MI_RESP_SIZE);
+	if (res) {
+		SAS_DPRINTK("MI: ex %016llx failed:0x%x\n",
+			    SAS_ADDR(dev->sas_addr), res);
+		goto out;
+	} else if (mi_resp[2] != SMP_RESP_FUNC_ACC) {
+		SAS_DPRINTK("MI ex %016llx returned SMP result:0x%x\n",
+			    SAS_ADDR(dev->sas_addr), mi_resp[2]);
+		goto out;
+	}
+
+	ex_assign_manuf_info(dev, mi_resp);
+out:
+	kfree(mi_req);
+	kfree(mi_resp);
+	return res;
+}
+
+#define PC_REQ_SIZE  44
+#define PC_RESP_SIZE 8
+
+int sas_smp_phy_control(struct domain_device *dev, int phy_id,
+			enum phy_func phy_func)
+{
+	u8 *pc_req;
+	u8 *pc_resp;
+	int res;
+
+	pc_req = alloc_smp_req(PC_REQ_SIZE);
+	if (!pc_req)
+		return -ENOMEM;
+
+	pc_resp = alloc_smp_resp(PC_RESP_SIZE);
+	if (!pc_resp) {
+		kfree(pc_req);
+		return -ENOMEM;
+	}
+
+	pc_req[1] = SMP_PHY_CONTROL;
+	pc_req[9] = phy_id;
+	pc_req[10]= phy_func;
+
+	res = smp_execute_task(dev, pc_req, PC_REQ_SIZE, pc_resp,PC_RESP_SIZE);
+
+	kfree(pc_resp);
+	kfree(pc_req);
+	return res;
+}
+
+static void sas_ex_disable_phy(struct domain_device *dev, int phy_id)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *phy = &ex->ex_phy[phy_id];
+
+	sas_smp_phy_control(dev, phy_id, PHY_FUNC_DISABLE);
+	phy->linkrate = PHY_DISABLED;
+}
+
+static void sas_ex_disable_port(struct domain_device *dev, u8 *sas_addr)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int i;
+
+	for (i = 0; i < ex->num_phys; i++) {
+		struct ex_phy *phy = &ex->ex_phy[i];
+
+		if (phy->phy_state == PHY_VACANT ||
+		    phy->phy_state == PHY_NOT_PRESENT)
+			continue;
+
+		if (SAS_ADDR(phy->attached_sas_addr) == SAS_ADDR(sas_addr))
+			sas_ex_disable_phy(dev, i);
+	}
+}
+
+static int sas_dev_present_in_domain(struct asd_sas_port *port,
+					    u8 *sas_addr)
+{
+	struct domain_device *dev;
+
+	if (SAS_ADDR(port->sas_addr) == SAS_ADDR(sas_addr))
+		return 1;
+	list_for_each_entry(dev, &port->dev_list, dev_list_node) {
+		if (SAS_ADDR(dev->sas_addr) == SAS_ADDR(sas_addr))
+			return 1;
+	}
+	return 0;
+}
+
+#define RPEL_REQ_SIZE	16
+#define RPEL_RESP_SIZE	32
+int sas_smp_get_phy_events(struct sas_phy *phy)
+{
+	int res;
+	struct sas_rphy *rphy = dev_to_rphy(phy->dev.parent);
+	struct domain_device *dev = sas_find_dev_by_rphy(rphy);
+	u8 *req = alloc_smp_req(RPEL_REQ_SIZE);
+	u8 *resp = kzalloc(RPEL_RESP_SIZE, GFP_KERNEL);
+
+	if (!resp)
+		return -ENOMEM;
+
+	req[1] = SMP_REPORT_PHY_ERR_LOG;
+	req[9] = phy->number;
+
+	res = smp_execute_task(dev, req, RPEL_REQ_SIZE,
+			            resp, RPEL_RESP_SIZE);
+
+	if (!res)
+		goto out;
+
+	phy->invalid_dword_count = scsi_to_u32(&resp[12]);
+	phy->running_disparity_error_count = scsi_to_u32(&resp[16]);
+	phy->loss_of_dword_sync_count = scsi_to_u32(&resp[20]);
+	phy->phy_reset_problem_count = scsi_to_u32(&resp[24]);
+
+ out:
+	kfree(resp);
+	return res;
+
+}
+
+#define RPS_REQ_SIZE  16
+#define RPS_RESP_SIZE 60
+
+static int sas_get_report_phy_sata(struct domain_device *dev,
+					  int phy_id,
+					  struct smp_resp *rps_resp)
+{
+	int res;
+	u8 *rps_req = alloc_smp_req(RPS_REQ_SIZE);
+
+	if (!rps_req)
+		return -ENOMEM;
+
+	rps_req[1] = SMP_REPORT_PHY_SATA;
+	rps_req[9] = phy_id;
+
+	res = smp_execute_task(dev, rps_req, RPS_REQ_SIZE,
+			            rps_resp, RPS_RESP_SIZE);
+
+	kfree(rps_req);
+	return 0;
+}
+
+static void sas_ex_get_linkrate(struct domain_device *parent,
+				       struct domain_device *child,
+				       struct ex_phy *parent_phy)
+{
+	struct expander_device *parent_ex = &parent->ex_dev;
+	struct sas_port *port;
+	int i;
+
+	child->pathways = 0;
+
+	port = parent_phy->port;
+
+	for (i = 0; i < parent_ex->num_phys; i++) {
+		struct ex_phy *phy = &parent_ex->ex_phy[i];
+
+		if (phy->phy_state == PHY_VACANT ||
+		    phy->phy_state == PHY_NOT_PRESENT)
+			continue;
+
+		if (SAS_ADDR(phy->attached_sas_addr) ==
+		    SAS_ADDR(child->sas_addr)) {
+
+			child->min_linkrate = min(parent->min_linkrate,
+						  phy->linkrate);
+			child->max_linkrate = max(parent->max_linkrate,
+						  phy->linkrate);
+			child->pathways++;
+			sas_port_add_phy(port, phy->phy);
+		}
+	}
+	child->linkrate = min(parent_phy->linkrate, child->max_linkrate);
+	child->pathways = min(child->pathways, parent->pathways);
+}
+
+static struct domain_device *sas_ex_discover_end_dev(
+	struct domain_device *parent, int phy_id)
+{
+	struct expander_device *parent_ex = &parent->ex_dev;
+	struct ex_phy *phy = &parent_ex->ex_phy[phy_id];
+	struct domain_device *child = NULL;
+	struct sas_rphy *rphy;
+	int res;
+
+	if (phy->attached_sata_host || phy->attached_sata_ps)
+		return NULL;
+
+	child = kzalloc(sizeof(*child), GFP_KERNEL);
+	if (!child)
+		return NULL;
+
+	child->parent = parent;
+	child->port   = parent->port;
+	child->iproto = phy->attached_iproto;
+	memcpy(child->sas_addr, phy->attached_sas_addr, SAS_ADDR_SIZE);
+	sas_hash_addr(child->hashed_sas_addr, child->sas_addr);
+	phy->port = sas_port_alloc(&parent->rphy->dev, phy_id);
+	BUG_ON(!phy->port);
+	/* FIXME: better error handling*/
+	BUG_ON(sas_port_add(phy->port) != 0);
+	sas_ex_get_linkrate(parent, child, phy);
+
+	if ((phy->attached_tproto & SAS_PROTO_STP) || phy->attached_sata_dev) {
+		child->dev_type = SATA_DEV;
+		if (phy->attached_tproto & SAS_PROTO_STP)
+			child->tproto = phy->attached_tproto;
+		if (phy->attached_sata_dev)
+			child->tproto |= SATA_DEV;
+		res = sas_get_report_phy_sata(parent, phy_id,
+					      &child->sata_dev.rps_resp);
+		if (res) {
+			SAS_DPRINTK("report phy sata to %016llx:0x%x returned "
+				    "0x%x\n", SAS_ADDR(parent->sas_addr),
+				    phy_id, res);
+			kfree(child);
+			return NULL;
+		}
+		memcpy(child->frame_rcvd, &child->sata_dev.rps_resp.rps.fis,
+		       sizeof(struct dev_to_host_fis));
+		sas_init_dev(child);
+		res = sas_discover_sata(child);
+		if (res) {
+			SAS_DPRINTK("sas_discover_sata() for device %16llx at "
+				    "%016llx:0x%x returned 0x%x\n",
+				    SAS_ADDR(child->sas_addr),
+				    SAS_ADDR(parent->sas_addr), phy_id, res);
+			kfree(child);
+			return NULL;
+		}
+	} else if (phy->attached_tproto & SAS_PROTO_SSP) {
+		child->dev_type = SAS_END_DEV;
+		rphy = sas_end_device_alloc(phy->port);
+		/* FIXME: error handling */
+		BUG_ON(!rphy);
+		child->tproto = phy->attached_tproto;
+		sas_init_dev(child);
+
+		child->rphy = rphy;
+		sas_fill_in_rphy(child, rphy);
+
+		spin_lock(&parent->port->dev_list_lock);
+		list_add_tail(&child->dev_list_node, &parent->port->dev_list);
+		spin_unlock(&parent->port->dev_list_lock);
+
+		res = sas_discover_end_dev(child);
+		if (res) {
+			SAS_DPRINTK("sas_discover_end_dev() for device %16llx "
+				    "at %016llx:0x%x returned 0x%x\n",
+				    SAS_ADDR(child->sas_addr),
+				    SAS_ADDR(parent->sas_addr), phy_id, res);
+			/* FIXME: this kfrees list elements without removing them */
+			//kfree(child);
+			return NULL;
+		}
+	} else {
+		SAS_DPRINTK("target proto 0x%x at %016llx:0x%x not handled\n",
+			    phy->attached_tproto, SAS_ADDR(parent->sas_addr),
+			    phy_id);
+	}
+
+	list_add_tail(&child->siblings, &parent_ex->children);
+	return child;
+}
+
+static struct domain_device *sas_ex_discover_expander(
+	struct domain_device *parent, int phy_id)
+{
+	struct sas_expander_device *parent_ex = rphy_to_expander_device(parent->rphy);
+	struct ex_phy *phy = &parent->ex_dev.ex_phy[phy_id];
+	struct domain_device *child = NULL;
+	struct sas_rphy *rphy;
+	struct sas_expander_device *edev;
+	struct asd_sas_port *port;
+	int res;
+
+	if (phy->routing_attr == DIRECT_ROUTING) {
+		SAS_DPRINTK("ex %016llx:0x%x:D <--> ex %016llx:0x%x is not "
+			    "allowed\n",
+			    SAS_ADDR(parent->sas_addr), phy_id,
+			    SAS_ADDR(phy->attached_sas_addr),
+			    phy->attached_phy_id);
+		return NULL;
+	}
+	child = kzalloc(sizeof(*child), GFP_KERNEL);
+	if (!child)
+		return NULL;
+
+	phy->port = sas_port_alloc(&parent->rphy->dev, phy_id);
+	/* FIXME: better error handling */
+	BUG_ON(sas_port_add(phy->port) != 0);
+
+
+	switch (phy->attached_dev_type) {
+	case EDGE_DEV:
+		rphy = sas_expander_alloc(phy->port,
+					  SAS_EDGE_EXPANDER_DEVICE);
+		break;
+	case FANOUT_DEV:
+		rphy = sas_expander_alloc(phy->port,
+					  SAS_FANOUT_EXPANDER_DEVICE);
+		break;
+	default:
+		rphy = NULL;	/* shut gcc up */
+		BUG();
+	}
+	port = parent->port;
+	child->rphy = rphy;
+	edev = rphy_to_expander_device(rphy);
+	child->dev_type = phy->attached_dev_type;
+	child->parent = parent;
+	child->port = port;
+	child->iproto = phy->attached_iproto;
+	child->tproto = phy->attached_tproto;
+	memcpy(child->sas_addr, phy->attached_sas_addr, SAS_ADDR_SIZE);
+	sas_hash_addr(child->hashed_sas_addr, child->sas_addr);
+	sas_ex_get_linkrate(parent, child, phy);
+	edev->level = parent_ex->level + 1;
+	parent->port->disc.max_level = max(parent->port->disc.max_level,
+					   edev->level);
+	sas_init_dev(child);
+	sas_fill_in_rphy(child, rphy);
+	sas_rphy_add(rphy);
+
+	spin_lock(&parent->port->dev_list_lock);
+	list_add_tail(&child->dev_list_node, &parent->port->dev_list);
+	spin_unlock(&parent->port->dev_list_lock);
+
+	res = sas_discover_expander(child);
+	if (res) {
+		kfree(child);
+		return NULL;
+	}
+	list_add_tail(&child->siblings, &parent->ex_dev.children);
+	return child;
+}
+
+static int sas_ex_discover_dev(struct domain_device *dev, int phy_id)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *ex_phy = &ex->ex_phy[phy_id];
+	struct domain_device *child = NULL;
+	int res = 0;
+
+	/* Phy state */
+	if (ex_phy->linkrate == PHY_SPINUP_HOLD) {
+		if (!sas_smp_phy_control(dev, phy_id, PHY_FUNC_LINK_RESET))
+			res = sas_ex_phy_discover(dev, phy_id);
+		if (res)
+			return res;
+	}
+
+	/* Parent and domain coherency */
+	if (!dev->parent && (SAS_ADDR(ex_phy->attached_sas_addr) ==
+			     SAS_ADDR(dev->port->sas_addr))) {
+		sas_add_parent_port(dev, phy_id);
+		return 0;
+	}
+	if (dev->parent && (SAS_ADDR(ex_phy->attached_sas_addr) ==
+			    SAS_ADDR(dev->parent->sas_addr))) {
+		sas_add_parent_port(dev, phy_id);
+		if (ex_phy->routing_attr == TABLE_ROUTING)
+			sas_configure_phy(dev, phy_id, dev->port->sas_addr, 1);
+		return 0;
+	}
+
+	if (sas_dev_present_in_domain(dev->port, ex_phy->attached_sas_addr))
+		sas_ex_disable_port(dev, ex_phy->attached_sas_addr);
+
+	if (ex_phy->attached_dev_type == NO_DEVICE) {
+		if (ex_phy->routing_attr == DIRECT_ROUTING) {
+			memset(ex_phy->attached_sas_addr, 0, SAS_ADDR_SIZE);
+			sas_configure_routing(dev, ex_phy->attached_sas_addr);
+		}
+		return 0;
+	} else if (ex_phy->linkrate == PHY_LINKRATE_UNKNOWN)
+		return 0;
+
+	if (ex_phy->attached_dev_type != SAS_END_DEV &&
+	    ex_phy->attached_dev_type != FANOUT_DEV &&
+	    ex_phy->attached_dev_type != EDGE_DEV) {
+		SAS_DPRINTK("unknown device type(0x%x) attached to ex %016llx "
+			    "phy 0x%x\n", ex_phy->attached_dev_type,
+			    SAS_ADDR(dev->sas_addr),
+			    phy_id);
+		return 0;
+	}
+
+	res = sas_configure_routing(dev, ex_phy->attached_sas_addr);
+	if (res) {
+		SAS_DPRINTK("configure routing for dev %016llx "
+			    "reported 0x%x. Forgotten\n",
+			    SAS_ADDR(ex_phy->attached_sas_addr), res);
+		sas_disable_routing(dev, ex_phy->attached_sas_addr);
+		return res;
+	}
+
+	switch (ex_phy->attached_dev_type) {
+	case SAS_END_DEV:
+		child = sas_ex_discover_end_dev(dev, phy_id);
+		break;
+	case FANOUT_DEV:
+		if (SAS_ADDR(dev->port->disc.fanout_sas_addr)) {
+			SAS_DPRINTK("second fanout expander %016llx phy 0x%x "
+				    "attached to ex %016llx phy 0x%x\n",
+				    SAS_ADDR(ex_phy->attached_sas_addr),
+				    ex_phy->attached_phy_id,
+				    SAS_ADDR(dev->sas_addr),
+				    phy_id);
+			sas_ex_disable_phy(dev, phy_id);
+			break;
+		} else
+			memcpy(dev->port->disc.fanout_sas_addr,
+			       ex_phy->attached_sas_addr, SAS_ADDR_SIZE);
+		/* fallthrough */
+	case EDGE_DEV:
+		child = sas_ex_discover_expander(dev, phy_id);
+		break;
+	default:
+		break;
+	}
+
+	if (child) {
+		int i;
+
+		for (i = 0; i < ex->num_phys; i++) {
+			if (ex->ex_phy[i].phy_state == PHY_VACANT ||
+			    ex->ex_phy[i].phy_state == PHY_NOT_PRESENT)
+				continue;
+
+			if (SAS_ADDR(ex->ex_phy[i].attached_sas_addr) ==
+			    SAS_ADDR(child->sas_addr))
+				ex->ex_phy[i].phy_state= PHY_DEVICE_DISCOVERED;
+		}
+	}
+
+	return res;
+}
+
+static int sas_find_sub_addr(struct domain_device *dev, u8 *sub_addr)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int i;
+
+	for (i = 0; i < ex->num_phys; i++) {
+		struct ex_phy *phy = &ex->ex_phy[i];
+
+		if (phy->phy_state == PHY_VACANT ||
+		    phy->phy_state == PHY_NOT_PRESENT)
+			continue;
+
+		if ((phy->attached_dev_type == EDGE_DEV ||
+		     phy->attached_dev_type == FANOUT_DEV) &&
+		    phy->routing_attr == SUBTRACTIVE_ROUTING) {
+
+			memcpy(sub_addr, phy->attached_sas_addr,SAS_ADDR_SIZE);
+
+			return 1;
+		}
+	}
+	return 0;
+}
+
+static int sas_check_level_subtractive_boundary(struct domain_device *dev)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct domain_device *child;
+	u8 sub_addr[8] = {0, };
+
+	list_for_each_entry(child, &ex->children, siblings) {
+		if (child->dev_type != EDGE_DEV &&
+		    child->dev_type != FANOUT_DEV)
+			continue;
+		if (sub_addr[0] == 0) {
+			sas_find_sub_addr(child, sub_addr);
+			continue;
+		} else {
+			u8 s2[8];
+
+			if (sas_find_sub_addr(child, s2) &&
+			    (SAS_ADDR(sub_addr) != SAS_ADDR(s2))) {
+
+				SAS_DPRINTK("ex %016llx->%016llx-?->%016llx "
+					    "diverges from subtractive "
+					    "boundary %016llx\n",
+					    SAS_ADDR(dev->sas_addr),
+					    SAS_ADDR(child->sas_addr),
+					    SAS_ADDR(s2),
+					    SAS_ADDR(sub_addr));
+
+				sas_ex_disable_port(child, s2);
+			}
+		}
+	}
+	return 0;
+}
+/**
+ * sas_ex_discover_devices -- discover devices attached to this expander
+ * dev: pointer to the expander domain device
+ * single: if you want to do a single phy, else set to -1;
+ *
+ * Configure this expander for use with its devices and register the
+ * devices of this expander.
+ */
+static int sas_ex_discover_devices(struct domain_device *dev, int single)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int i = 0, end = ex->num_phys;
+	int res = 0;
+
+	if (0 <= single && single < end) {
+		i = single;
+		end = i+1;
+	}
+
+	for ( ; i < end; i++) {
+		struct ex_phy *ex_phy = &ex->ex_phy[i];
+
+		if (ex_phy->phy_state == PHY_VACANT ||
+		    ex_phy->phy_state == PHY_NOT_PRESENT ||
+		    ex_phy->phy_state == PHY_DEVICE_DISCOVERED)
+			continue;
+
+		switch (ex_phy->linkrate) {
+		case PHY_DISABLED:
+		case PHY_RESET_PROBLEM:
+		case PHY_PORT_SELECTOR:
+			continue;
+		default:
+			res = sas_ex_discover_dev(dev, i);
+			if (res)
+				break;
+			continue;
+		}
+	}
+
+	if (!res)
+		sas_check_level_subtractive_boundary(dev);
+
+	return res;
+}
+
+static int sas_check_ex_subtractive_boundary(struct domain_device *dev)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int i;
+	u8  *sub_sas_addr = NULL;
+
+	if (dev->dev_type != EDGE_DEV)
+		return 0;
+
+	for (i = 0; i < ex->num_phys; i++) {
+		struct ex_phy *phy = &ex->ex_phy[i];
+
+		if (phy->phy_state == PHY_VACANT ||
+		    phy->phy_state == PHY_NOT_PRESENT)
+			continue;
+
+		if ((phy->attached_dev_type == FANOUT_DEV ||
+		     phy->attached_dev_type == EDGE_DEV) &&
+		    phy->routing_attr == SUBTRACTIVE_ROUTING) {
+
+			if (!sub_sas_addr)
+				sub_sas_addr = &phy->attached_sas_addr[0];
+			else if (SAS_ADDR(sub_sas_addr) !=
+				 SAS_ADDR(phy->attached_sas_addr)) {
+
+				SAS_DPRINTK("ex %016llx phy 0x%x "
+					    "diverges(%016llx) on subtractive "
+					    "boundary(%016llx). Disabled\n",
+					    SAS_ADDR(dev->sas_addr), i,
+					    SAS_ADDR(phy->attached_sas_addr),
+					    SAS_ADDR(sub_sas_addr));
+				sas_ex_disable_phy(dev, i);
+			}
+		}
+	}
+	return 0;
+}
+
+static void sas_print_parent_topology_bug(struct domain_device *child,
+						 struct ex_phy *parent_phy,
+						 struct ex_phy *child_phy)
+{
+	static const char ra_char[] = {
+		[DIRECT_ROUTING] = 'D',
+		[SUBTRACTIVE_ROUTING] = 'S',
+		[TABLE_ROUTING] = 'T',
+	};
+	static const char *ex_type[] = {
+		[EDGE_DEV] = "edge",
+		[FANOUT_DEV] = "fanout",
+	};
+	struct domain_device *parent = child->parent;
+
+	sas_printk("%s ex %016llx phy 0x%x <--> %s ex %016llx phy 0x%x "
+		   "has %c:%c routing link!\n",
+
+		   ex_type[parent->dev_type],
+		   SAS_ADDR(parent->sas_addr),
+		   parent_phy->phy_id,
+
+		   ex_type[child->dev_type],
+		   SAS_ADDR(child->sas_addr),
+		   child_phy->phy_id,
+
+		   ra_char[parent_phy->routing_attr],
+		   ra_char[child_phy->routing_attr]);
+}
+
+static int sas_check_eeds(struct domain_device *child,
+				 struct ex_phy *parent_phy,
+				 struct ex_phy *child_phy)
+{
+	int res = 0;
+	struct domain_device *parent = child->parent;
+
+	if (SAS_ADDR(parent->port->disc.fanout_sas_addr) != 0) {
+		res = -ENODEV;
+		SAS_DPRINTK("edge ex %016llx phy S:0x%x <--> edge ex %016llx "
+			    "phy S:0x%x, while there is a fanout ex %016llx\n",
+			    SAS_ADDR(parent->sas_addr),
+			    parent_phy->phy_id,
+			    SAS_ADDR(child->sas_addr),
+			    child_phy->phy_id,
+			    SAS_ADDR(parent->port->disc.fanout_sas_addr));
+	} else if (SAS_ADDR(parent->port->disc.eeds_a) == 0) {
+		memcpy(parent->port->disc.eeds_a, parent->sas_addr,
+		       SAS_ADDR_SIZE);
+		memcpy(parent->port->disc.eeds_b, child->sas_addr,
+		       SAS_ADDR_SIZE);
+	} else if (((SAS_ADDR(parent->port->disc.eeds_a) ==
+		    SAS_ADDR(parent->sas_addr)) ||
+		   (SAS_ADDR(parent->port->disc.eeds_a) ==
+		    SAS_ADDR(child->sas_addr)))
+		   &&
+		   ((SAS_ADDR(parent->port->disc.eeds_b) ==
+		     SAS_ADDR(parent->sas_addr)) ||
+		    (SAS_ADDR(parent->port->disc.eeds_b) ==
+		     SAS_ADDR(child->sas_addr))))
+		;
+	else {
+		res = -ENODEV;
+		SAS_DPRINTK("edge ex %016llx phy 0x%x <--> edge ex %016llx "
+			    "phy 0x%x link forms a third EEDS!\n",
+			    SAS_ADDR(parent->sas_addr),
+			    parent_phy->phy_id,
+			    SAS_ADDR(child->sas_addr),
+			    child_phy->phy_id);
+	}
+
+	return res;
+}
+
+/* Here we spill over 80 columns.  It is intentional.
+ */
+static int sas_check_parent_topology(struct domain_device *child)
+{
+	struct expander_device *child_ex = &child->ex_dev;
+	struct expander_device *parent_ex;
+	int i;
+	int res = 0;
+
+	if (!child->parent)
+		return 0;
+
+	if (child->parent->dev_type != EDGE_DEV &&
+	    child->parent->dev_type != FANOUT_DEV)
+		return 0;
+
+	parent_ex = &child->parent->ex_dev;
+
+	for (i = 0; i < parent_ex->num_phys; i++) {
+		struct ex_phy *parent_phy = &parent_ex->ex_phy[i];
+		struct ex_phy *child_phy;
+
+		if (parent_phy->phy_state == PHY_VACANT ||
+		    parent_phy->phy_state == PHY_NOT_PRESENT)
+			continue;
+
+		if (SAS_ADDR(parent_phy->attached_sas_addr) != SAS_ADDR(child->sas_addr))
+			continue;
+
+		child_phy = &child_ex->ex_phy[parent_phy->attached_phy_id];
+
+		switch (child->parent->dev_type) {
+		case EDGE_DEV:
+			if (child->dev_type == FANOUT_DEV) {
+				if (parent_phy->routing_attr != SUBTRACTIVE_ROUTING ||
+				    child_phy->routing_attr != TABLE_ROUTING) {
+					sas_print_parent_topology_bug(child, parent_phy, child_phy);
+					res = -ENODEV;
+				}
+			} else if (parent_phy->routing_attr == SUBTRACTIVE_ROUTING) {
+				if (child_phy->routing_attr == SUBTRACTIVE_ROUTING) {
+					res = sas_check_eeds(child, parent_phy, child_phy);
+				} else if (child_phy->routing_attr != TABLE_ROUTING) {
+					sas_print_parent_topology_bug(child, parent_phy, child_phy);
+					res = -ENODEV;
+				}
+			} else if (parent_phy->routing_attr == TABLE_ROUTING &&
+				   child_phy->routing_attr != SUBTRACTIVE_ROUTING) {
+				sas_print_parent_topology_bug(child, parent_phy, child_phy);
+				res = -ENODEV;
+			}
+			break;
+		case FANOUT_DEV:
+			if (parent_phy->routing_attr != TABLE_ROUTING ||
+			    child_phy->routing_attr != SUBTRACTIVE_ROUTING) {
+				sas_print_parent_topology_bug(child, parent_phy, child_phy);
+				res = -ENODEV;
+			}
+			break;
+		default:
+			break;
+		}
+	}
+
+	return res;
+}
+
+#define RRI_REQ_SIZE  16
+#define RRI_RESP_SIZE 44
+
+static int sas_configure_present(struct domain_device *dev, int phy_id,
+				 u8 *sas_addr, int *index, int *present)
+{
+	int i, res = 0;
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *phy = &ex->ex_phy[phy_id];
+	u8 *rri_req;
+	u8 *rri_resp;
+
+	*present = 0;
+	*index = 0;
+
+	rri_req = alloc_smp_req(RRI_REQ_SIZE);
+	if (!rri_req)
+		return -ENOMEM;
+
+	rri_resp = alloc_smp_resp(RRI_RESP_SIZE);
+	if (!rri_resp) {
+		kfree(rri_req);
+		return -ENOMEM;
+	}
+
+	rri_req[1] = SMP_REPORT_ROUTE_INFO;
+	rri_req[9] = phy_id;
+
+	for (i = 0; i < ex->max_route_indexes ; i++) {
+		*(__be16 *)(rri_req+6) = cpu_to_be16(i);
+		res = smp_execute_task(dev, rri_req, RRI_REQ_SIZE, rri_resp,
+				       RRI_RESP_SIZE);
+		if (res)
+			goto out;
+		res = rri_resp[2];
+		if (res == SMP_RESP_NO_INDEX) {
+			SAS_DPRINTK("overflow of indexes: dev %016llx "
+				    "phy 0x%x index 0x%x\n",
+				    SAS_ADDR(dev->sas_addr), phy_id, i);
+			goto out;
+		} else if (res != SMP_RESP_FUNC_ACC) {
+			SAS_DPRINTK("%s: dev %016llx phy 0x%x index 0x%x "
+				    "result 0x%x\n", __FUNCTION__,
+				    SAS_ADDR(dev->sas_addr), phy_id, i, res);
+			goto out;
+		}
+		if (SAS_ADDR(sas_addr) != 0) {
+			if (SAS_ADDR(rri_resp+16) == SAS_ADDR(sas_addr)) {
+				*index = i;
+				if ((rri_resp[12] & 0x80) == 0x80)
+					*present = 0;
+				else
+					*present = 1;
+				goto out;
+			} else if (SAS_ADDR(rri_resp+16) == 0) {
+				*index = i;
+				*present = 0;
+				goto out;
+			}
+		} else if (SAS_ADDR(rri_resp+16) == 0 &&
+			   phy->last_da_index < i) {
+			phy->last_da_index = i;
+			*index = i;
+			*present = 0;
+			goto out;
+		}
+	}
+	res = -1;
+out:
+	kfree(rri_req);
+	kfree(rri_resp);
+	return res;
+}
+
+#define CRI_REQ_SIZE  44
+#define CRI_RESP_SIZE  8
+
+static int sas_configure_set(struct domain_device *dev, int phy_id,
+			     u8 *sas_addr, int index, int include)
+{
+	int res;
+	u8 *cri_req;
+	u8 *cri_resp;
+
+	cri_req = alloc_smp_req(CRI_REQ_SIZE);
+	if (!cri_req)
+		return -ENOMEM;
+
+	cri_resp = alloc_smp_resp(CRI_RESP_SIZE);
+	if (!cri_resp) {
+		kfree(cri_req);
+		return -ENOMEM;
+	}
+
+	cri_req[1] = SMP_CONF_ROUTE_INFO;
+	*(__be16 *)(cri_req+6) = cpu_to_be16(index);
+	cri_req[9] = phy_id;
+	if (SAS_ADDR(sas_addr) == 0 || !include)
+		cri_req[12] |= 0x80;
+	memcpy(cri_req+16, sas_addr, SAS_ADDR_SIZE);
+
+	res = smp_execute_task(dev, cri_req, CRI_REQ_SIZE, cri_resp,
+			       CRI_RESP_SIZE);
+	if (res)
+		goto out;
+	res = cri_resp[2];
+	if (res == SMP_RESP_NO_INDEX) {
+		SAS_DPRINTK("overflow of indexes: dev %016llx phy 0x%x "
+			    "index 0x%x\n",
+			    SAS_ADDR(dev->sas_addr), phy_id, index);
+	}
+out:
+	kfree(cri_req);
+	kfree(cri_resp);
+	return res;
+}
+
+static int sas_configure_phy(struct domain_device *dev, int phy_id,
+				    u8 *sas_addr, int include)
+{
+	int index;
+	int present;
+	int res;
+
+	res = sas_configure_present(dev, phy_id, sas_addr, &index, &present);
+	if (res)
+		return res;
+	if (include ^ present)
+		return sas_configure_set(dev, phy_id, sas_addr, index,include);
+
+	return res;
+}
+
+/**
+ * sas_configure_parent -- configure routing table of parent
+ * parent: parent expander
+ * child: child expander
+ * sas_addr: SAS port identifier of device directly attached to child
+ */
+static int sas_configure_parent(struct domain_device *parent,
+				struct domain_device *child,
+				u8 *sas_addr, int include)
+{
+	struct expander_device *ex_parent = &parent->ex_dev;
+	int res = 0;
+	int i;
+
+	if (parent->parent) {
+		res = sas_configure_parent(parent->parent, parent, sas_addr,
+					   include);
+		if (res)
+			return res;
+	}
+
+	if (ex_parent->conf_route_table == 0) {
+		SAS_DPRINTK("ex %016llx has self-configuring routing table\n",
+			    SAS_ADDR(parent->sas_addr));
+		return 0;
+	}
+
+	for (i = 0; i < ex_parent->num_phys; i++) {
+		struct ex_phy *phy = &ex_parent->ex_phy[i];
+
+		if ((phy->routing_attr == TABLE_ROUTING) &&
+		    (SAS_ADDR(phy->attached_sas_addr) ==
+		     SAS_ADDR(child->sas_addr))) {
+			res = sas_configure_phy(parent, i, sas_addr, include);
+			if (res)
+				return res;
+		}
+	}
+
+	return res;
+}
+
+/**
+ * sas_configure_routing -- configure routing
+ * dev: expander device
+ * sas_addr: port identifier of device directly attached to the expander device
+ */
+static int sas_configure_routing(struct domain_device *dev, u8 *sas_addr)
+{
+	if (dev->parent)
+		return sas_configure_parent(dev->parent, dev, sas_addr, 1);
+	return 0;
+}
+
+static int sas_disable_routing(struct domain_device *dev,  u8 *sas_addr)
+{
+	if (dev->parent)
+		return sas_configure_parent(dev->parent, dev, sas_addr, 0);
+	return 0;
+}
+
+#if 0
+#define SMP_BIN_ATTR_NAME "smp_portal"
+
+static void sas_ex_smp_hook(struct domain_device *dev)
+{
+	struct expander_device *ex_dev = &dev->ex_dev;
+	struct bin_attribute *bin_attr = &ex_dev->smp_bin_attr;
+
+	memset(bin_attr, 0, sizeof(*bin_attr));
+
+	bin_attr->attr.name = SMP_BIN_ATTR_NAME;
+	bin_attr->attr.owner = THIS_MODULE;
+	bin_attr->attr.mode = 0600;
+
+	bin_attr->size = 0;
+	bin_attr->private = NULL;
+	bin_attr->read = smp_portal_read;
+	bin_attr->write= smp_portal_write;
+	bin_attr->mmap = NULL;
+
+	ex_dev->smp_portal_pid = -1;
+	init_MUTEX(&ex_dev->smp_sema);
+}
+#endif
+
+/**
+ * sas_discover_expander -- expander discovery
+ * @ex: pointer to expander domain device
+ *
+ * See comment in sas_discover_sata().
+ */
+static int sas_discover_expander(struct domain_device *dev)
+{
+	int res;
+
+	res = sas_notify_lldd_dev_found(dev);
+	if (res)
+		return res;
+
+	res = sas_ex_general(dev);
+	if (res)
+		goto out_err;
+	res = sas_ex_manuf_info(dev);
+	if (res)
+		goto out_err;
+
+	res = sas_expander_discover(dev);
+	if (res) {
+		SAS_DPRINTK("expander %016llx discovery failed(0x%x)\n",
+			    SAS_ADDR(dev->sas_addr), res);
+		goto out_err;
+	}
+
+	sas_check_ex_subtractive_boundary(dev);
+	res = sas_check_parent_topology(dev);
+	if (res)
+		goto out_err;
+	return 0;
+out_err:
+	sas_notify_lldd_dev_gone(dev);
+	return res;
+}
+
+static int sas_ex_level_discovery(struct asd_sas_port *port, const int level)
+{
+	int res = 0;
+	struct domain_device *dev;
+
+	list_for_each_entry(dev, &port->dev_list, dev_list_node) {
+		if (dev->dev_type == EDGE_DEV ||
+		    dev->dev_type == FANOUT_DEV) {
+			struct sas_expander_device *ex =
+				rphy_to_expander_device(dev->rphy);
+
+			if (level == ex->level)
+				res = sas_ex_discover_devices(dev, -1);
+			else if (level > 0)
+				res = sas_ex_discover_devices(port->port_dev, -1);
+
+		}
+	}
+
+	return res;
+}
+
+static int sas_ex_bfs_disc(struct asd_sas_port *port)
+{
+	int res;
+	int level;
+
+	do {
+		level = port->disc.max_level;
+		res = sas_ex_level_discovery(port, level);
+		mb();
+	} while (level < port->disc.max_level);
+
+	return res;
+}
+
+int sas_discover_root_expander(struct domain_device *dev)
+{
+	int res;
+	struct sas_expander_device *ex = rphy_to_expander_device(dev->rphy);
+
+	sas_rphy_add(dev->rphy);
+
+	ex->level = dev->port->disc.max_level; /* 0 */
+	res = sas_discover_expander(dev);
+	if (!res)
+		sas_ex_bfs_disc(dev->port);
+
+	return res;
+}
+
+/* ---------- Domain revalidation ---------- */
+
+static int sas_get_phy_discover(struct domain_device *dev,
+				int phy_id, struct smp_resp *disc_resp)
+{
+	int res;
+	u8 *disc_req;
+
+	disc_req = alloc_smp_req(DISCOVER_REQ_SIZE);
+	if (!disc_req)
+		return -ENOMEM;
+
+	disc_req[1] = SMP_DISCOVER;
+	disc_req[9] = phy_id;
+
+	res = smp_execute_task(dev, disc_req, DISCOVER_REQ_SIZE,
+			       disc_resp, DISCOVER_RESP_SIZE);
+	if (res)
+		goto out;
+	else if (disc_resp->result != SMP_RESP_FUNC_ACC) {
+		res = disc_resp->result;
+		goto out;
+	}
+out:
+	kfree(disc_req);
+	return res;
+}
+
+static int sas_get_phy_change_count(struct domain_device *dev,
+				    int phy_id, int *pcc)
+{
+	int res;
+	struct smp_resp *disc_resp;
+
+	disc_resp = alloc_smp_resp(DISCOVER_RESP_SIZE);
+	if (!disc_resp)
+		return -ENOMEM;
+
+	res = sas_get_phy_discover(dev, phy_id, disc_resp);
+	if (!res)
+		*pcc = disc_resp->disc.change_count;
+
+	kfree(disc_resp);
+	return res;
+}
+
+static int sas_get_phy_attached_sas_addr(struct domain_device *dev,
+					 int phy_id, u8 *attached_sas_addr)
+{
+	int res;
+	struct smp_resp *disc_resp;
+	struct discover_resp *dr;
+
+	disc_resp = alloc_smp_resp(DISCOVER_RESP_SIZE);
+	if (!disc_resp)
+		return -ENOMEM;
+	dr = &disc_resp->disc;
+
+	res = sas_get_phy_discover(dev, phy_id, disc_resp);
+	if (!res) {
+		memcpy(attached_sas_addr,disc_resp->disc.attached_sas_addr,8);
+		if (dr->attached_dev_type == 0)
+			memset(attached_sas_addr, 0, 8);
+	}
+	kfree(disc_resp);
+	return res;
+}
+
+static int sas_find_bcast_phy(struct domain_device *dev, int *phy_id,
+			      int from_phy)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int res = 0;
+	int i;
+
+	for (i = from_phy; i < ex->num_phys; i++) {
+		int phy_change_count = 0;
+
+		res = sas_get_phy_change_count(dev, i, &phy_change_count);
+		if (res)
+			goto out;
+		else if (phy_change_count != ex->ex_phy[i].phy_change_count) {
+			ex->ex_phy[i].phy_change_count = phy_change_count;
+			*phy_id = i;
+			return 0;
+		}
+	}
+out:
+	return res;
+}
+
+static int sas_get_ex_change_count(struct domain_device *dev, int *ecc)
+{
+	int res;
+	u8  *rg_req;
+	struct smp_resp  *rg_resp;
+
+	rg_req = alloc_smp_req(RG_REQ_SIZE);
+	if (!rg_req)
+		return -ENOMEM;
+
+	rg_resp = alloc_smp_resp(RG_RESP_SIZE);
+	if (!rg_resp) {
+		kfree(rg_req);
+		return -ENOMEM;
+	}
+
+	rg_req[1] = SMP_REPORT_GENERAL;
+
+	res = smp_execute_task(dev, rg_req, RG_REQ_SIZE, rg_resp,
+			       RG_RESP_SIZE);
+	if (res)
+		goto out;
+	if (rg_resp->result != SMP_RESP_FUNC_ACC) {
+		res = rg_resp->result;
+		goto out;
+	}
+
+	*ecc = be16_to_cpu(rg_resp->rg.change_count);
+out:
+	kfree(rg_resp);
+	kfree(rg_req);
+	return res;
+}
+
+static int sas_find_bcast_dev(struct domain_device *dev,
+			      struct domain_device **src_dev)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	int ex_change_count = -1;
+	int res;
+
+	res = sas_get_ex_change_count(dev, &ex_change_count);
+	if (res)
+		goto out;
+	if (ex_change_count != -1 &&
+	    ex_change_count != ex->ex_change_count) {
+		*src_dev = dev;
+		ex->ex_change_count = ex_change_count;
+	} else {
+		struct domain_device *ch;
+
+		list_for_each_entry(ch, &ex->children, siblings) {
+			if (ch->dev_type == EDGE_DEV ||
+			    ch->dev_type == FANOUT_DEV) {
+				res = sas_find_bcast_dev(ch, src_dev);
+				if (src_dev)
+					return res;
+			}
+		}
+	}
+out:
+	return res;
+}
+
+static void sas_unregister_ex_tree(struct domain_device *dev)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct domain_device *child, *n;
+
+	list_for_each_entry_safe(child, n, &ex->children, siblings) {
+		if (child->dev_type == EDGE_DEV ||
+		    child->dev_type == FANOUT_DEV)
+			sas_unregister_ex_tree(child);
+		else
+			sas_unregister_dev(child);
+	}
+	sas_unregister_dev(dev);
+}
+
+static void sas_unregister_devs_sas_addr(struct domain_device *parent,
+					 int phy_id)
+{
+	struct expander_device *ex_dev = &parent->ex_dev;
+	struct ex_phy *phy = &ex_dev->ex_phy[phy_id];
+	struct domain_device *child, *n;
+
+	list_for_each_entry_safe(child, n, &ex_dev->children, siblings) {
+		if (SAS_ADDR(child->sas_addr) ==
+		    SAS_ADDR(phy->attached_sas_addr)) {
+			if (child->dev_type == EDGE_DEV ||
+			    child->dev_type == FANOUT_DEV)
+				sas_unregister_ex_tree(child);
+			else
+				sas_unregister_dev(child);
+			break;
+		}
+	}
+	sas_disable_routing(parent, phy->attached_sas_addr);
+	memset(phy->attached_sas_addr, 0, SAS_ADDR_SIZE);
+	sas_port_delete_phy(phy->port, phy->phy);
+	if (phy->port->num_phys == 0)
+		sas_port_delete(phy->port);
+	phy->port = NULL;
+}
+
+static int sas_discover_bfs_by_root_level(struct domain_device *root,
+					  const int level)
+{
+	struct expander_device *ex_root = &root->ex_dev;
+	struct domain_device *child;
+	int res = 0;
+
+	list_for_each_entry(child, &ex_root->children, siblings) {
+		if (child->dev_type == EDGE_DEV ||
+		    child->dev_type == FANOUT_DEV) {
+			struct sas_expander_device *ex =
+				rphy_to_expander_device(child->rphy);
+
+			if (level > ex->level)
+				res = sas_discover_bfs_by_root_level(child,
+								     level);
+			else if (level == ex->level)
+				res = sas_ex_discover_devices(child, -1);
+		}
+	}
+	return res;
+}
+
+static int sas_discover_bfs_by_root(struct domain_device *dev)
+{
+	int res;
+	struct sas_expander_device *ex = rphy_to_expander_device(dev->rphy);
+	int level = ex->level+1;
+
+	res = sas_ex_discover_devices(dev, -1);
+	if (res)
+		goto out;
+	do {
+		res = sas_discover_bfs_by_root_level(dev, level);
+		mb();
+		level += 1;
+	} while (level <= dev->port->disc.max_level);
+out:
+	return res;
+}
+
+static int sas_discover_new(struct domain_device *dev, int phy_id)
+{
+	struct ex_phy *ex_phy = &dev->ex_dev.ex_phy[phy_id];
+	struct domain_device *child;
+	int res;
+
+	SAS_DPRINTK("ex %016llx phy%d new device attached\n",
+		    SAS_ADDR(dev->sas_addr), phy_id);
+	res = sas_ex_phy_discover(dev, phy_id);
+	if (res)
+		goto out;
+	res = sas_ex_discover_devices(dev, phy_id);
+	if (res)
+		goto out;
+	list_for_each_entry(child, &dev->ex_dev.children, siblings) {
+		if (SAS_ADDR(child->sas_addr) ==
+		    SAS_ADDR(ex_phy->attached_sas_addr)) {
+			if (child->dev_type == EDGE_DEV ||
+			    child->dev_type == FANOUT_DEV)
+				res = sas_discover_bfs_by_root(child);
+			break;
+		}
+	}
+out:
+	return res;
+}
+
+static int sas_rediscover_dev(struct domain_device *dev, int phy_id)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *phy = &ex->ex_phy[phy_id];
+	u8 attached_sas_addr[8];
+	int res;
+
+	res = sas_get_phy_attached_sas_addr(dev, phy_id, attached_sas_addr);
+	switch (res) {
+	case SMP_RESP_NO_PHY:
+		phy->phy_state = PHY_NOT_PRESENT;
+		sas_unregister_devs_sas_addr(dev, phy_id);
+		goto out; break;
+	case SMP_RESP_PHY_VACANT:
+		phy->phy_state = PHY_VACANT;
+		sas_unregister_devs_sas_addr(dev, phy_id);
+		goto out; break;
+	case SMP_RESP_FUNC_ACC:
+		break;
+	}
+
+	if (SAS_ADDR(attached_sas_addr) == 0) {
+		phy->phy_state = PHY_EMPTY;
+		sas_unregister_devs_sas_addr(dev, phy_id);
+	} else if (SAS_ADDR(attached_sas_addr) ==
+		   SAS_ADDR(phy->attached_sas_addr)) {
+		SAS_DPRINTK("ex %016llx phy 0x%x broadcast flutter\n",
+			    SAS_ADDR(dev->sas_addr), phy_id);
+	} else
+		res = sas_discover_new(dev, phy_id);
+out:
+	return res;
+}
+
+static int sas_rediscover(struct domain_device *dev, const int phy_id)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *changed_phy = &ex->ex_phy[phy_id];
+	int res = 0;
+	int i;
+
+	SAS_DPRINTK("ex %016llx phy%d originated BROADCAST(CHANGE)\n",
+		    SAS_ADDR(dev->sas_addr), phy_id);
+
+	if (SAS_ADDR(changed_phy->attached_sas_addr) != 0) {
+		for (i = 0; i < ex->num_phys; i++) {
+			struct ex_phy *phy = &ex->ex_phy[i];
+
+			if (i == phy_id)
+				continue;
+			if (SAS_ADDR(phy->attached_sas_addr) ==
+			    SAS_ADDR(changed_phy->attached_sas_addr)) {
+				SAS_DPRINTK("phy%d part of wide port with "
+					    "phy%d\n", phy_id, i);
+				goto out;
+			}
+		}
+		res = sas_rediscover_dev(dev, phy_id);
+	} else
+		res = sas_discover_new(dev, phy_id);
+out:
+	return res;
+}
+
+/**
+ * sas_revalidate_domain -- revalidate the domain
+ * @port: port to the domain of interest
+ *
+ * NOTE: this process _must_ quit (return) as soon as any connection
+ * errors are encountered.  Connection recovery is done elsewhere.
+ * Discover process only interrogates devices in order to discover the
+ * domain.
+ */
+int sas_ex_revalidate_domain(struct domain_device *port_dev)
+{
+	int res;
+	struct domain_device *dev = NULL;
+
+	res = sas_find_bcast_dev(port_dev, &dev);
+	if (res)
+		goto out;
+	if (dev) {
+		struct expander_device *ex = &dev->ex_dev;
+		int i = 0, phy_id;
+
+		do {
+			phy_id = -1;
+			res = sas_find_bcast_phy(dev, &phy_id, i);
+			if (phy_id == -1)
+				break;
+			res = sas_rediscover(dev, phy_id);
+			i = phy_id + 1;
+		} while (i < ex->num_phys);
+	}
+out:
+	return res;
+}
+
+#if 0
+/* ---------- SMP portal ---------- */
+
+static ssize_t smp_portal_write(struct kobject *kobj, char *buf, loff_t offs,
+				size_t size)
+{
+	struct domain_device *dev = to_dom_device(kobj);
+	struct expander_device *ex = &dev->ex_dev;
+
+	if (offs != 0)
+		return -EFBIG;
+	else if (size == 0)
+		return 0;
+
+	down_interruptible(&ex->smp_sema);
+	if (ex->smp_req)
+		kfree(ex->smp_req);
+	ex->smp_req = kzalloc(size, GFP_USER);
+	if (!ex->smp_req) {
+		up(&ex->smp_sema);
+		return -ENOMEM;
+	}
+	memcpy(ex->smp_req, buf, size);
+	ex->smp_req_size = size;
+	ex->smp_portal_pid = current->pid;
+	up(&ex->smp_sema);
+
+	return size;
+}
+
+static ssize_t smp_portal_read(struct kobject *kobj, char *buf, loff_t offs,
+			       size_t size)
+{
+	struct domain_device *dev = to_dom_device(kobj);
+	struct expander_device *ex = &dev->ex_dev;
+	u8 *smp_resp;
+	int res = -EINVAL;
+
+	/* XXX: sysfs gives us an offset of 0x10 or 0x8 while in fact
+	 *  it should be 0.
+	 */
+
+	down_interruptible(&ex->smp_sema);
+	if (!ex->smp_req || ex->smp_portal_pid != current->pid)
+		goto out;
+
+	res = 0;
+	if (size == 0)
+		goto out;
+
+	res = -ENOMEM;
+	smp_resp = alloc_smp_resp(size);
+	if (!smp_resp)
+		goto out;
+	res = smp_execute_task(dev, ex->smp_req, ex->smp_req_size,
+			       smp_resp, size);
+	if (!res) {
+		memcpy(buf, smp_resp, size);
+		res = size;
+	}
+
+	kfree(smp_resp);
+out:
+	kfree(ex->smp_req);
+	ex->smp_req = NULL;
+	ex->smp_req_size = 0;
+	ex->smp_portal_pid = -1;
+	up(&ex->smp_sema);
+	return res;
+}
+#endif
diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c
new file mode 100644
index 000000000000..b961664b8106
--- /dev/null
+++ b/drivers/scsi/libsas/sas_init.c
@@ -0,0 +1,227 @@
+/*
+ * Serial Attached SCSI (SAS) Transport Layer initialization
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ * USA
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/spinlock.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+
+#include "sas_internal.h"
+
+#include "../scsi_sas_internal.h"
+
+kmem_cache_t *sas_task_cache;
+
+/*------------ SAS addr hash -----------*/
+void sas_hash_addr(u8 *hashed, const u8 *sas_addr)
+{
+        const u32 poly = 0x00DB2777;
+        u32     r = 0;
+        int     i;
+
+        for (i = 0; i < 8; i++) {
+                int b;
+                for (b = 7; b >= 0; b--) {
+                        r <<= 1;
+                        if ((1 << b) & sas_addr[i]) {
+                                if (!(r & 0x01000000))
+                                        r ^= poly;
+                        } else if (r & 0x01000000)
+                                r ^= poly;
+                }
+        }
+
+        hashed[0] = (r >> 16) & 0xFF;
+        hashed[1] = (r >> 8) & 0xFF ;
+        hashed[2] = r & 0xFF;
+}
+
+
+/* ---------- HA events ---------- */
+
+void sas_hae_reset(void *data)
+{
+	struct sas_ha_struct *ha = data;
+
+	sas_begin_event(HAE_RESET, &ha->event_lock,
+			&ha->pending);
+}
+
+int sas_register_ha(struct sas_ha_struct *sas_ha)
+{
+	int error = 0;
+
+	spin_lock_init(&sas_ha->phy_port_lock);
+	sas_hash_addr(sas_ha->hashed_sas_addr, sas_ha->sas_addr);
+
+	if (sas_ha->lldd_queue_size == 0)
+		sas_ha->lldd_queue_size = 1;
+	else if (sas_ha->lldd_queue_size == -1)
+		sas_ha->lldd_queue_size = 128; /* Sanity */
+
+	error = sas_register_phys(sas_ha);
+	if (error) {
+		printk(KERN_NOTICE "couldn't register sas phys:%d\n", error);
+		return error;
+	}
+
+	error = sas_register_ports(sas_ha);
+	if (error) {
+		printk(KERN_NOTICE "couldn't register sas ports:%d\n", error);
+		goto Undo_phys;
+	}
+
+	error = sas_init_events(sas_ha);
+	if (error) {
+		printk(KERN_NOTICE "couldn't start event thread:%d\n", error);
+		goto Undo_ports;
+	}
+
+	if (sas_ha->lldd_max_execute_num > 1) {
+		error = sas_init_queue(sas_ha);
+		if (error) {
+			printk(KERN_NOTICE "couldn't start queue thread:%d, "
+			       "running in direct mode\n", error);
+			sas_ha->lldd_max_execute_num = 1;
+		}
+	}
+
+	return 0;
+
+Undo_ports:
+	sas_unregister_ports(sas_ha);
+Undo_phys:
+
+	return error;
+}
+
+int sas_unregister_ha(struct sas_ha_struct *sas_ha)
+{
+	if (sas_ha->lldd_max_execute_num > 1) {
+		sas_shutdown_queue(sas_ha);
+	}
+
+	sas_unregister_ports(sas_ha);
+
+	return 0;
+}
+
+static int sas_get_linkerrors(struct sas_phy *phy)
+{
+	if (scsi_is_sas_phy_local(phy))
+		/* FIXME: we have no local phy stats
+		 * gathering at this time */
+		return -EINVAL;
+
+	return sas_smp_get_phy_events(phy);
+}
+
+static int sas_phy_reset(struct sas_phy *phy, int hard_reset)
+{
+	int ret;
+	enum phy_func reset_type;
+
+	if (hard_reset)
+		reset_type = PHY_FUNC_HARD_RESET;
+	else
+		reset_type = PHY_FUNC_LINK_RESET;
+
+	if (scsi_is_sas_phy_local(phy)) {
+		struct Scsi_Host *shost = dev_to_shost(phy->dev.parent);
+		struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost);
+		struct asd_sas_phy *asd_phy = sas_ha->sas_phy[phy->number];
+		struct sas_internal *i =
+			to_sas_internal(sas_ha->core.shost->transportt);
+
+		ret = i->dft->lldd_control_phy(asd_phy, reset_type);
+	} else {
+		struct sas_rphy *rphy = dev_to_rphy(phy->dev.parent);
+		struct domain_device *ddev = sas_find_dev_by_rphy(rphy);
+		ret = sas_smp_phy_control(ddev, phy->number, reset_type);
+	}
+	return ret;
+}
+
+static struct sas_function_template sft = {
+	.phy_reset = sas_phy_reset,
+	.get_linkerrors = sas_get_linkerrors,
+};
+
+struct scsi_transport_template *
+sas_domain_attach_transport(struct sas_domain_function_template *dft)
+{
+	struct scsi_transport_template *stt = sas_attach_transport(&sft);
+	struct sas_internal *i;
+
+	if (!stt)
+		return stt;
+
+	i = to_sas_internal(stt);
+	i->dft = dft;
+	stt->create_work_queue = 1;
+	stt->eh_timed_out = sas_scsi_timed_out;
+	stt->eh_strategy_handler = sas_scsi_recover_host;
+
+	return stt;
+}
+EXPORT_SYMBOL_GPL(sas_domain_attach_transport);
+
+
+void sas_domain_release_transport(struct scsi_transport_template *stt)
+{
+	sas_release_transport(stt);
+}
+EXPORT_SYMBOL_GPL(sas_domain_release_transport);
+
+/* ---------- SAS Class register/unregister ---------- */
+
+static int __init sas_class_init(void)
+{
+	sas_task_cache = kmem_cache_create("sas_task", sizeof(struct sas_task),
+					   0, SLAB_HWCACHE_ALIGN, NULL, NULL);
+	if (!sas_task_cache)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void __exit sas_class_exit(void)
+{
+	kmem_cache_destroy(sas_task_cache);
+}
+
+MODULE_AUTHOR("Luben Tuikov <luben_tuikov@adaptec.com>");
+MODULE_DESCRIPTION("SAS Transport Layer");
+MODULE_LICENSE("GPL v2");
+
+module_init(sas_class_init);
+module_exit(sas_class_exit);
+
+EXPORT_SYMBOL_GPL(sas_register_ha);
+EXPORT_SYMBOL_GPL(sas_unregister_ha);
diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h
new file mode 100644
index 000000000000..89c397680846
--- /dev/null
+++ b/drivers/scsi/libsas/sas_internal.h
@@ -0,0 +1,146 @@
+/*
+ * Serial Attached SCSI (SAS) class internal header file
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ * USA
+ *
+ */
+
+#ifndef _SAS_INTERNAL_H_
+#define _SAS_INTERNAL_H_
+
+#include <scsi/scsi.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_transport_sas.h>
+#include <scsi/libsas.h>
+
+#define sas_printk(fmt, ...) printk(KERN_NOTICE "sas: " fmt, ## __VA_ARGS__)
+
+#ifdef SAS_DEBUG
+#define SAS_DPRINTK(fmt, ...) printk(KERN_NOTICE "sas: " fmt, ## __VA_ARGS__)
+#else
+#define SAS_DPRINTK(fmt, ...)
+#endif
+
+void sas_scsi_recover_host(struct Scsi_Host *shost);
+
+int sas_show_class(enum sas_class class, char *buf);
+int sas_show_proto(enum sas_proto proto, char *buf);
+int sas_show_linkrate(enum sas_phy_linkrate linkrate, char *buf);
+int sas_show_oob_mode(enum sas_oob_mode oob_mode, char *buf);
+
+int  sas_register_phys(struct sas_ha_struct *sas_ha);
+void sas_unregister_phys(struct sas_ha_struct *sas_ha);
+
+int  sas_register_ports(struct sas_ha_struct *sas_ha);
+void sas_unregister_ports(struct sas_ha_struct *sas_ha);
+
+enum scsi_eh_timer_return sas_scsi_timed_out(struct scsi_cmnd *);
+
+int  sas_init_queue(struct sas_ha_struct *sas_ha);
+int  sas_init_events(struct sas_ha_struct *sas_ha);
+void sas_shutdown_queue(struct sas_ha_struct *sas_ha);
+
+void sas_deform_port(struct asd_sas_phy *phy);
+
+void sas_porte_bytes_dmaed(void *);
+void sas_porte_broadcast_rcvd(void *);
+void sas_porte_link_reset_err(void *);
+void sas_porte_timer_event(void *);
+void sas_porte_hard_reset(void *);
+
+int sas_notify_lldd_dev_found(struct domain_device *);
+void sas_notify_lldd_dev_gone(struct domain_device *);
+
+int sas_smp_phy_control(struct domain_device *dev, int phy_id,
+			enum phy_func phy_func);
+int sas_smp_get_phy_events(struct sas_phy *phy);
+
+struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy);
+
+void sas_hae_reset(void *);
+
+static inline void sas_queue_event(int event, spinlock_t *lock,
+				   unsigned long *pending,
+				   struct work_struct *work,
+				   struct Scsi_Host *shost)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(lock, flags);
+	if (test_bit(event, pending)) {
+		spin_unlock_irqrestore(lock, flags);
+		return;
+	}
+	__set_bit(event, pending);
+	spin_unlock_irqrestore(lock, flags);
+	scsi_queue_work(shost, work);
+}
+
+static inline void sas_begin_event(int event, spinlock_t *lock,
+				   unsigned long *pending)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(lock, flags);
+	__clear_bit(event, pending);
+	spin_unlock_irqrestore(lock, flags);
+}
+
+static inline void sas_fill_in_rphy(struct domain_device *dev,
+				    struct sas_rphy *rphy)
+{
+	rphy->identify.sas_address = SAS_ADDR(dev->sas_addr);
+	rphy->identify.initiator_port_protocols = dev->iproto;
+	rphy->identify.target_port_protocols = dev->tproto;
+	switch (dev->dev_type) {
+	case SATA_DEV:
+		/* FIXME: need sata device type */
+	case SAS_END_DEV:
+		rphy->identify.device_type = SAS_END_DEVICE;
+		break;
+	case EDGE_DEV:
+		rphy->identify.device_type = SAS_EDGE_EXPANDER_DEVICE;
+		break;
+	case FANOUT_DEV:
+		rphy->identify.device_type = SAS_FANOUT_EXPANDER_DEVICE;
+		break;
+	default:
+		rphy->identify.device_type = SAS_PHY_UNUSED;
+		break;
+	}
+}
+
+static inline void sas_add_parent_port(struct domain_device *dev, int phy_id)
+{
+	struct expander_device *ex = &dev->ex_dev;
+	struct ex_phy *ex_phy = &ex->ex_phy[phy_id];
+
+	if (!ex->parent_port) {
+		ex->parent_port = sas_port_alloc(&dev->rphy->dev, phy_id);
+		/* FIXME: error handling */
+		BUG_ON(!ex->parent_port);
+		BUG_ON(sas_port_add(ex->parent_port));
+		sas_port_mark_backlink(ex->parent_port);
+	}
+	sas_port_add_phy(ex->parent_port, ex_phy->phy);
+}
+
+#endif /* _SAS_INTERNAL_H_ */
diff --git a/drivers/scsi/libsas/sas_phy.c b/drivers/scsi/libsas/sas_phy.c
new file mode 100644
index 000000000000..024ab00e70d2
--- /dev/null
+++ b/drivers/scsi/libsas/sas_phy.c
@@ -0,0 +1,157 @@
+/*
+ * Serial Attached SCSI (SAS) Phy class
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include "sas_internal.h"
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+#include "../scsi_sas_internal.h"
+
+/* ---------- Phy events ---------- */
+
+static void sas_phye_loss_of_signal(void *data)
+{
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PHYE_LOSS_OF_SIGNAL, &phy->ha->event_lock,
+			&phy->phy_events_pending);
+	phy->error = 0;
+	sas_deform_port(phy);
+}
+
+static void sas_phye_oob_done(void *data)
+{
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PHYE_OOB_DONE, &phy->ha->event_lock,
+			&phy->phy_events_pending);
+	phy->error = 0;
+}
+
+static void sas_phye_oob_error(void *data)
+{
+	struct asd_sas_phy *phy = data;
+	struct sas_ha_struct *sas_ha = phy->ha;
+	struct asd_sas_port *port = phy->port;
+	struct sas_internal *i =
+		to_sas_internal(sas_ha->core.shost->transportt);
+
+	sas_begin_event(PHYE_OOB_ERROR, &phy->ha->event_lock,
+			&phy->phy_events_pending);
+
+	sas_deform_port(phy);
+
+	if (!port && phy->enabled && i->dft->lldd_control_phy) {
+		phy->error++;
+		switch (phy->error) {
+		case 1:
+		case 2:
+			i->dft->lldd_control_phy(phy, PHY_FUNC_HARD_RESET);
+			break;
+		case 3:
+		default:
+			phy->error = 0;
+			phy->enabled = 0;
+			i->dft->lldd_control_phy(phy, PHY_FUNC_DISABLE);
+			break;
+		}
+	}
+}
+
+static void sas_phye_spinup_hold(void *data)
+{
+	struct asd_sas_phy *phy = data;
+	struct sas_ha_struct *sas_ha = phy->ha;
+	struct sas_internal *i =
+		to_sas_internal(sas_ha->core.shost->transportt);
+
+	sas_begin_event(PHYE_SPINUP_HOLD, &phy->ha->event_lock,
+			&phy->phy_events_pending);
+
+	phy->error = 0;
+	i->dft->lldd_control_phy(phy, PHY_FUNC_RELEASE_SPINUP_HOLD);
+}
+
+/* ---------- Phy class registration ---------- */
+
+int sas_register_phys(struct sas_ha_struct *sas_ha)
+{
+	int i;
+
+	static void (*sas_phy_event_fns[PHY_NUM_EVENTS])(void *) = {
+		[PHYE_LOSS_OF_SIGNAL] = sas_phye_loss_of_signal,
+		[PHYE_OOB_DONE] = sas_phye_oob_done,
+		[PHYE_OOB_ERROR] = sas_phye_oob_error,
+		[PHYE_SPINUP_HOLD] = sas_phye_spinup_hold,
+	};
+
+	static void (*sas_port_event_fns[PORT_NUM_EVENTS])(void *) = {
+		[PORTE_BYTES_DMAED] = sas_porte_bytes_dmaed,
+		[PORTE_BROADCAST_RCVD] = sas_porte_broadcast_rcvd,
+		[PORTE_LINK_RESET_ERR] = sas_porte_link_reset_err,
+		[PORTE_TIMER_EVENT] = sas_porte_timer_event,
+		[PORTE_HARD_RESET] = sas_porte_hard_reset,
+	};
+
+	/* Now register the phys. */
+	for (i = 0; i < sas_ha->num_phys; i++) {
+		int k;
+		struct asd_sas_phy *phy = sas_ha->sas_phy[i];
+
+		phy->error = 0;
+		INIT_LIST_HEAD(&phy->port_phy_el);
+		for (k = 0; k < PORT_NUM_EVENTS; k++)
+			INIT_WORK(&phy->port_events[k], sas_port_event_fns[k],
+				  phy);
+
+		for (k = 0; k < PHY_NUM_EVENTS; k++)
+			INIT_WORK(&phy->phy_events[k], sas_phy_event_fns[k],
+				  phy);
+		phy->port = NULL;
+		phy->ha = sas_ha;
+		spin_lock_init(&phy->frame_rcvd_lock);
+		spin_lock_init(&phy->sas_prim_lock);
+		phy->frame_rcvd_size = 0;
+
+		phy->phy = sas_phy_alloc(&sas_ha->core.shost->shost_gendev,
+					 i);
+		if (!phy->phy)
+			return -ENOMEM;
+
+		phy->phy->identify.initiator_port_protocols =
+			phy->iproto;
+		phy->phy->identify.target_port_protocols = phy->tproto;
+		phy->phy->identify.sas_address = SAS_ADDR(sas_ha->sas_addr);
+		phy->phy->identify.phy_identifier = i;
+		phy->phy->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS;
+		phy->phy->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS;
+		phy->phy->minimum_linkrate = SAS_LINK_RATE_1_5_GBPS;
+		phy->phy->maximum_linkrate = SAS_LINK_RATE_3_0_GBPS;
+		phy->phy->negotiated_linkrate = SAS_LINK_RATE_UNKNOWN;
+
+		sas_phy_add(phy->phy);
+	}
+
+	return 0;
+}
diff --git a/drivers/scsi/libsas/sas_port.c b/drivers/scsi/libsas/sas_port.c
new file mode 100644
index 000000000000..253cdcf306a2
--- /dev/null
+++ b/drivers/scsi/libsas/sas_port.c
@@ -0,0 +1,279 @@
+/*
+ * Serial Attached SCSI (SAS) Port class
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include "sas_internal.h"
+
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+#include "../scsi_sas_internal.h"
+
+/**
+ * sas_form_port -- add this phy to a port
+ * @phy: the phy of interest
+ *
+ * This function adds this phy to an existing port, thus creating a wide
+ * port, or it creates a port and adds the phy to the port.
+ */
+static void sas_form_port(struct asd_sas_phy *phy)
+{
+	int i;
+	struct sas_ha_struct *sas_ha = phy->ha;
+	struct asd_sas_port *port = phy->port;
+	struct sas_internal *si =
+		to_sas_internal(sas_ha->core.shost->transportt);
+
+	if (port) {
+		if (memcmp(port->attached_sas_addr, phy->attached_sas_addr,
+			   SAS_ADDR_SIZE) == 0)
+			sas_deform_port(phy);
+		else {
+			SAS_DPRINTK("%s: phy%d belongs to port%d already(%d)!\n",
+				    __FUNCTION__, phy->id, phy->port->id,
+				    phy->port->num_phys);
+			return;
+		}
+	}
+
+	/* find a port */
+	spin_lock(&sas_ha->phy_port_lock);
+	for (i = 0; i < sas_ha->num_phys; i++) {
+		port = sas_ha->sas_port[i];
+		spin_lock(&port->phy_list_lock);
+		if (*(u64 *) port->sas_addr &&
+		    memcmp(port->attached_sas_addr,
+			   phy->attached_sas_addr, SAS_ADDR_SIZE) == 0 &&
+		    port->num_phys > 0) {
+			/* wide port */
+			SAS_DPRINTK("phy%d matched wide port%d\n", phy->id,
+				    port->id);
+			break;
+		} else if (*(u64 *) port->sas_addr == 0 && port->num_phys==0) {
+			memcpy(port->sas_addr, phy->sas_addr, SAS_ADDR_SIZE);
+			break;
+		}
+		spin_unlock(&port->phy_list_lock);
+	}
+
+	if (i >= sas_ha->num_phys) {
+		printk(KERN_NOTICE "%s: couldn't find a free port, bug?\n",
+		       __FUNCTION__);
+		spin_unlock(&sas_ha->phy_port_lock);
+		return;
+	}
+
+	/* add the phy to the port */
+	list_add_tail(&phy->port_phy_el, &port->phy_list);
+	phy->port = port;
+	port->num_phys++;
+	port->phy_mask |= (1U << phy->id);
+
+	if (!port->phy)
+		port->phy = phy->phy;
+
+	SAS_DPRINTK("phy%d added to port%d, phy_mask:0x%x\n", phy->id,
+		    port->id, port->phy_mask);
+
+	if (*(u64 *)port->attached_sas_addr == 0) {
+		port->class = phy->class;
+		memcpy(port->attached_sas_addr, phy->attached_sas_addr,
+		       SAS_ADDR_SIZE);
+		port->iproto = phy->iproto;
+		port->tproto = phy->tproto;
+		port->oob_mode = phy->oob_mode;
+		port->linkrate = phy->linkrate;
+	} else
+		port->linkrate = max(port->linkrate, phy->linkrate);
+	spin_unlock(&port->phy_list_lock);
+	spin_unlock(&sas_ha->phy_port_lock);
+
+	if (!port->port) {
+		port->port = sas_port_alloc(phy->phy->dev.parent, port->id);
+		BUG_ON(!port->port);
+		sas_port_add(port->port);
+	}
+	sas_port_add_phy(port->port, phy->phy);
+
+	if (port->port_dev)
+		port->port_dev->pathways = port->num_phys;
+
+	/* Tell the LLDD about this port formation. */
+	if (si->dft->lldd_port_formed)
+		si->dft->lldd_port_formed(phy);
+
+	sas_discover_event(phy->port, DISCE_DISCOVER_DOMAIN);
+}
+
+/**
+ * sas_deform_port -- remove this phy from the port it belongs to
+ * @phy: the phy of interest
+ *
+ * This is called when the physical link to the other phy has been
+ * lost (on this phy), in Event thread context. We cannot delay here.
+ */
+void sas_deform_port(struct asd_sas_phy *phy)
+{
+	struct sas_ha_struct *sas_ha = phy->ha;
+	struct asd_sas_port *port = phy->port;
+	struct sas_internal *si =
+		to_sas_internal(sas_ha->core.shost->transportt);
+
+	if (!port)
+		return;		  /* done by a phy event */
+
+	if (port->port_dev)
+		port->port_dev->pathways--;
+
+	if (port->num_phys == 1) {
+		sas_unregister_domain_devices(port);
+		sas_port_delete(port->port);
+		port->port = NULL;
+	} else
+		sas_port_delete_phy(port->port, phy->phy);
+
+
+	if (si->dft->lldd_port_deformed)
+		si->dft->lldd_port_deformed(phy);
+
+	spin_lock(&sas_ha->phy_port_lock);
+	spin_lock(&port->phy_list_lock);
+
+	list_del_init(&phy->port_phy_el);
+	phy->port = NULL;
+	port->num_phys--;
+	port->phy_mask &= ~(1U << phy->id);
+
+	if (port->num_phys == 0) {
+		INIT_LIST_HEAD(&port->phy_list);
+		memset(port->sas_addr, 0, SAS_ADDR_SIZE);
+		memset(port->attached_sas_addr, 0, SAS_ADDR_SIZE);
+		port->class = 0;
+		port->iproto = 0;
+		port->tproto = 0;
+		port->oob_mode = 0;
+		port->phy_mask = 0;
+	}
+	spin_unlock(&port->phy_list_lock);
+	spin_unlock(&sas_ha->phy_port_lock);
+
+	return;
+}
+
+/* ---------- SAS port events ---------- */
+
+void sas_porte_bytes_dmaed(void *data)
+{
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PORTE_BYTES_DMAED, &phy->ha->event_lock,
+			&phy->port_events_pending);
+
+	sas_form_port(phy);
+}
+
+void sas_porte_broadcast_rcvd(void *data)
+{
+	unsigned long flags;
+	u32 prim;
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PORTE_BROADCAST_RCVD, &phy->ha->event_lock,
+			&phy->port_events_pending);
+
+	spin_lock_irqsave(&phy->sas_prim_lock, flags);
+	prim = phy->sas_prim;
+	spin_unlock_irqrestore(&phy->sas_prim_lock, flags);
+
+	SAS_DPRINTK("broadcast received: %d\n", prim);
+	sas_discover_event(phy->port, DISCE_REVALIDATE_DOMAIN);
+}
+
+void sas_porte_link_reset_err(void *data)
+{
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PORTE_LINK_RESET_ERR, &phy->ha->event_lock,
+			&phy->port_events_pending);
+
+	sas_deform_port(phy);
+}
+
+void sas_porte_timer_event(void *data)
+{
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PORTE_TIMER_EVENT, &phy->ha->event_lock,
+			&phy->port_events_pending);
+
+	sas_deform_port(phy);
+}
+
+void sas_porte_hard_reset(void *data)
+{
+	struct asd_sas_phy *phy = data;
+
+	sas_begin_event(PORTE_HARD_RESET, &phy->ha->event_lock,
+			&phy->port_events_pending);
+
+	sas_deform_port(phy);
+}
+
+/* ---------- SAS port registration ---------- */
+
+static void sas_init_port(struct asd_sas_port *port,
+			  struct sas_ha_struct *sas_ha, int i)
+{
+	port->id = i;
+	INIT_LIST_HEAD(&port->dev_list);
+	spin_lock_init(&port->phy_list_lock);
+	INIT_LIST_HEAD(&port->phy_list);
+	port->num_phys = 0;
+	port->phy_mask = 0;
+	port->ha = sas_ha;
+
+	spin_lock_init(&port->dev_list_lock);
+}
+
+int sas_register_ports(struct sas_ha_struct *sas_ha)
+{
+	int i;
+
+	/* initialize the ports and discovery */
+	for (i = 0; i < sas_ha->num_phys; i++) {
+		struct asd_sas_port *port = sas_ha->sas_port[i];
+
+		sas_init_port(port, sas_ha, i);
+		sas_init_disc(&port->disc, port);
+	}
+	return 0;
+}
+
+void sas_unregister_ports(struct sas_ha_struct *sas_ha)
+{
+	int i;
+
+	for (i = 0; i < sas_ha->num_phys; i++)
+		if (sas_ha->sas_phy[i]->port)
+			sas_deform_port(sas_ha->sas_phy[i]);
+
+}
diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
new file mode 100644
index 000000000000..43e0e4e36934
--- /dev/null
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -0,0 +1,786 @@
+/*
+ * Serial Attached SCSI (SAS) class SCSI Host glue.
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ * USA
+ *
+ */
+
+#include "sas_internal.h"
+
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_tcq.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_sas.h>
+#include "../scsi_sas_internal.h"
+
+#include <linux/err.h>
+#include <linux/blkdev.h>
+#include <linux/scatterlist.h>
+
+/* ---------- SCSI Host glue ---------- */
+
+#define TO_SAS_TASK(_scsi_cmd)  ((void *)(_scsi_cmd)->host_scribble)
+#define ASSIGN_SAS_TASK(_sc, _t) do { (_sc)->host_scribble = (void *) _t; } while (0)
+
+static void sas_scsi_task_done(struct sas_task *task)
+{
+	struct task_status_struct *ts = &task->task_status;
+	struct scsi_cmnd *sc = task->uldd_task;
+	unsigned ts_flags = task->task_state_flags;
+	int hs = 0, stat = 0;
+
+	if (unlikely(!sc)) {
+		SAS_DPRINTK("task_done called with non existing SCSI cmnd!\n");
+		list_del_init(&task->list);
+		sas_free_task(task);
+		return;
+	}
+
+	if (ts->resp == SAS_TASK_UNDELIVERED) {
+		/* transport error */
+		hs = DID_NO_CONNECT;
+	} else { /* ts->resp == SAS_TASK_COMPLETE */
+		/* task delivered, what happened afterwards? */
+		switch (ts->stat) {
+		case SAS_DEV_NO_RESPONSE:
+		case SAS_INTERRUPTED:
+		case SAS_PHY_DOWN:
+		case SAS_NAK_R_ERR:
+		case SAS_OPEN_TO:
+			hs = DID_NO_CONNECT;
+			break;
+		case SAS_DATA_UNDERRUN:
+			sc->resid = ts->residual;
+			if (sc->request_bufflen - sc->resid < sc->underflow)
+				hs = DID_ERROR;
+			break;
+		case SAS_DATA_OVERRUN:
+			hs = DID_ERROR;
+			break;
+		case SAS_QUEUE_FULL:
+			hs = DID_SOFT_ERROR; /* retry */
+			break;
+		case SAS_DEVICE_UNKNOWN:
+			hs = DID_BAD_TARGET;
+			break;
+		case SAS_SG_ERR:
+			hs = DID_PARITY;
+			break;
+		case SAS_OPEN_REJECT:
+			if (ts->open_rej_reason == SAS_OREJ_RSVD_RETRY)
+				hs = DID_SOFT_ERROR; /* retry */
+			else
+				hs = DID_ERROR;
+			break;
+		case SAS_PROTO_RESPONSE:
+			SAS_DPRINTK("LLDD:%s sent SAS_PROTO_RESP for an SSP "
+				    "task; please report this\n",
+				    task->dev->port->ha->sas_ha_name);
+			break;
+		case SAS_ABORTED_TASK:
+			hs = DID_ABORT;
+			break;
+		case SAM_CHECK_COND:
+			memcpy(sc->sense_buffer, ts->buf,
+			       max(SCSI_SENSE_BUFFERSIZE, ts->buf_valid_size));
+			stat = SAM_CHECK_COND;
+			break;
+		default:
+			stat = ts->stat;
+			break;
+		}
+	}
+	ASSIGN_SAS_TASK(sc, NULL);
+	sc->result = (hs << 16) | stat;
+	list_del_init(&task->list);
+	sas_free_task(task);
+	/* This is very ugly but this is how SCSI Core works. */
+	if (ts_flags & SAS_TASK_STATE_ABORTED)
+		scsi_finish_command(sc);
+	else
+		sc->scsi_done(sc);
+}
+
+static enum task_attribute sas_scsi_get_task_attr(struct scsi_cmnd *cmd)
+{
+	enum task_attribute ta = TASK_ATTR_SIMPLE;
+	if (cmd->request && blk_rq_tagged(cmd->request)) {
+		if (cmd->device->ordered_tags &&
+		    (cmd->request->flags & REQ_HARDBARRIER))
+			ta = TASK_ATTR_HOQ;
+	}
+	return ta;
+}
+
+static struct sas_task *sas_create_task(struct scsi_cmnd *cmd,
+					       struct domain_device *dev,
+					       unsigned long gfp_flags)
+{
+	struct sas_task *task = sas_alloc_task(gfp_flags);
+	struct scsi_lun lun;
+
+	if (!task)
+		return NULL;
+
+	*(u32 *)cmd->sense_buffer = 0;
+	task->uldd_task = cmd;
+	ASSIGN_SAS_TASK(cmd, task);
+
+	task->dev = dev;
+	task->task_proto = task->dev->tproto; /* BUG_ON(!SSP) */
+
+	task->ssp_task.retry_count = 1;
+	int_to_scsilun(cmd->device->lun, &lun);
+	memcpy(task->ssp_task.LUN, &lun.scsi_lun, 8);
+	task->ssp_task.task_attr = sas_scsi_get_task_attr(cmd);
+	memcpy(task->ssp_task.cdb, cmd->cmnd, 16);
+
+	task->scatter = cmd->request_buffer;
+	task->num_scatter = cmd->use_sg;
+	task->total_xfer_len = cmd->request_bufflen;
+	task->data_dir = cmd->sc_data_direction;
+
+	task->task_done = sas_scsi_task_done;
+
+	return task;
+}
+
+static int sas_queue_up(struct sas_task *task)
+{
+	struct sas_ha_struct *sas_ha = task->dev->port->ha;
+	struct scsi_core *core = &sas_ha->core;
+	unsigned long flags;
+	LIST_HEAD(list);
+
+	spin_lock_irqsave(&core->task_queue_lock, flags);
+	if (sas_ha->lldd_queue_size < core->task_queue_size + 1) {
+		spin_unlock_irqrestore(&core->task_queue_lock, flags);
+		return -SAS_QUEUE_FULL;
+	}
+	list_add_tail(&task->list, &core->task_queue);
+	core->task_queue_size += 1;
+	spin_unlock_irqrestore(&core->task_queue_lock, flags);
+	up(&core->queue_thread_sema);
+
+	return 0;
+}
+
+/**
+ * sas_queuecommand -- Enqueue a command for processing
+ * @parameters: See SCSI Core documentation
+ *
+ * Note: XXX: Remove the host unlock/lock pair when SCSI Core can
+ * call us without holding an IRQ spinlock...
+ */
+int sas_queuecommand(struct scsi_cmnd *cmd,
+		     void (*scsi_done)(struct scsi_cmnd *))
+{
+	int res = 0;
+	struct domain_device *dev = cmd_to_domain_dev(cmd);
+	struct Scsi_Host *host = cmd->device->host;
+	struct sas_internal *i = to_sas_internal(host->transportt);
+
+	spin_unlock_irq(host->host_lock);
+
+	{
+		struct sas_ha_struct *sas_ha = dev->port->ha;
+		struct sas_task *task;
+
+		res = -ENOMEM;
+		task = sas_create_task(cmd, dev, GFP_ATOMIC);
+		if (!task)
+			goto out;
+
+		cmd->scsi_done = scsi_done;
+		/* Queue up, Direct Mode or Task Collector Mode. */
+		if (sas_ha->lldd_max_execute_num < 2)
+			res = i->dft->lldd_execute_task(task, 1, GFP_ATOMIC);
+		else
+			res = sas_queue_up(task);
+
+		/* Examine */
+		if (res) {
+			SAS_DPRINTK("lldd_execute_task returned: %d\n", res);
+			ASSIGN_SAS_TASK(cmd, NULL);
+			sas_free_task(task);
+			if (res == -SAS_QUEUE_FULL) {
+				cmd->result = DID_SOFT_ERROR << 16; /* retry */
+				res = 0;
+				scsi_done(cmd);
+			}
+			goto out;
+		}
+	}
+out:
+	spin_lock_irq(host->host_lock);
+	return res;
+}
+
+static void sas_scsi_clear_queue_lu(struct list_head *error_q, struct scsi_cmnd *my_cmd)
+{
+	struct scsi_cmnd *cmd, *n;
+
+	list_for_each_entry_safe(cmd, n, error_q, eh_entry) {
+		if (cmd == my_cmd)
+			list_del_init(&cmd->eh_entry);
+	}
+}
+
+static void sas_scsi_clear_queue_I_T(struct list_head *error_q,
+				     struct domain_device *dev)
+{
+	struct scsi_cmnd *cmd, *n;
+
+	list_for_each_entry_safe(cmd, n, error_q, eh_entry) {
+		struct domain_device *x = cmd_to_domain_dev(cmd);
+
+		if (x == dev)
+			list_del_init(&cmd->eh_entry);
+	}
+}
+
+static void sas_scsi_clear_queue_port(struct list_head *error_q,
+				      struct asd_sas_port *port)
+{
+	struct scsi_cmnd *cmd, *n;
+
+	list_for_each_entry_safe(cmd, n, error_q, eh_entry) {
+		struct domain_device *dev = cmd_to_domain_dev(cmd);
+		struct asd_sas_port *x = dev->port;
+
+		if (x == port)
+			list_del_init(&cmd->eh_entry);
+	}
+}
+
+enum task_disposition {
+	TASK_IS_DONE,
+	TASK_IS_ABORTED,
+	TASK_IS_AT_LU,
+	TASK_IS_NOT_AT_LU,
+};
+
+static enum task_disposition sas_scsi_find_task(struct sas_task *task)
+{
+	struct sas_ha_struct *ha = task->dev->port->ha;
+	unsigned long flags;
+	int i, res;
+	struct sas_internal *si =
+		to_sas_internal(task->dev->port->ha->core.shost->transportt);
+
+	if (ha->lldd_max_execute_num > 1) {
+		struct scsi_core *core = &ha->core;
+		struct sas_task *t, *n;
+
+		spin_lock_irqsave(&core->task_queue_lock, flags);
+		list_for_each_entry_safe(t, n, &core->task_queue, list) {
+			if (task == t) {
+				list_del_init(&t->list);
+				spin_unlock_irqrestore(&core->task_queue_lock,
+						       flags);
+				SAS_DPRINTK("%s: task 0x%p aborted from "
+					    "task_queue\n",
+					    __FUNCTION__, task);
+				return TASK_IS_ABORTED;
+			}
+		}
+		spin_unlock_irqrestore(&core->task_queue_lock, flags);
+	}
+
+	for (i = 0; i < 5; i++) {
+		SAS_DPRINTK("%s: aborting task 0x%p\n", __FUNCTION__, task);
+		res = si->dft->lldd_abort_task(task);
+
+		spin_lock_irqsave(&task->task_state_lock, flags);
+		if (task->task_state_flags & SAS_TASK_STATE_DONE) {
+			spin_unlock_irqrestore(&task->task_state_lock, flags);
+			SAS_DPRINTK("%s: task 0x%p is done\n", __FUNCTION__,
+				    task);
+			return TASK_IS_DONE;
+		}
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+		if (res == TMF_RESP_FUNC_COMPLETE) {
+			SAS_DPRINTK("%s: task 0x%p is aborted\n",
+				    __FUNCTION__, task);
+			return TASK_IS_ABORTED;
+		} else if (si->dft->lldd_query_task) {
+			SAS_DPRINTK("%s: querying task 0x%p\n",
+				    __FUNCTION__, task);
+			res = si->dft->lldd_query_task(task);
+			if (res == TMF_RESP_FUNC_SUCC) {
+				SAS_DPRINTK("%s: task 0x%p at LU\n",
+					    __FUNCTION__, task);
+				return TASK_IS_AT_LU;
+			} else if (res == TMF_RESP_FUNC_COMPLETE) {
+				SAS_DPRINTK("%s: task 0x%p not at LU\n",
+					    __FUNCTION__, task);
+				return TASK_IS_NOT_AT_LU;
+			}
+		}
+	}
+	return res;
+}
+
+static int sas_recover_lu(struct domain_device *dev, struct scsi_cmnd *cmd)
+{
+	int res = TMF_RESP_FUNC_FAILED;
+	struct scsi_lun lun;
+	struct sas_internal *i =
+		to_sas_internal(dev->port->ha->core.shost->transportt);
+
+	int_to_scsilun(cmd->device->lun, &lun);
+
+	SAS_DPRINTK("eh: device %llx LUN %x has the task\n",
+		    SAS_ADDR(dev->sas_addr),
+		    cmd->device->lun);
+
+	if (i->dft->lldd_abort_task_set)
+		res = i->dft->lldd_abort_task_set(dev, lun.scsi_lun);
+
+	if (res == TMF_RESP_FUNC_FAILED) {
+		if (i->dft->lldd_clear_task_set)
+			res = i->dft->lldd_clear_task_set(dev, lun.scsi_lun);
+	}
+
+	if (res == TMF_RESP_FUNC_FAILED) {
+		if (i->dft->lldd_lu_reset)
+			res = i->dft->lldd_lu_reset(dev, lun.scsi_lun);
+	}
+
+	return res;
+}
+
+static int sas_recover_I_T(struct domain_device *dev)
+{
+	int res = TMF_RESP_FUNC_FAILED;
+	struct sas_internal *i =
+		to_sas_internal(dev->port->ha->core.shost->transportt);
+
+	SAS_DPRINTK("I_T nexus reset for dev %016llx\n",
+		    SAS_ADDR(dev->sas_addr));
+
+	if (i->dft->lldd_I_T_nexus_reset)
+		res = i->dft->lldd_I_T_nexus_reset(dev);
+
+	return res;
+}
+
+void sas_scsi_recover_host(struct Scsi_Host *shost)
+{
+	struct sas_ha_struct *ha = SHOST_TO_SAS_HA(shost);
+	unsigned long flags;
+	LIST_HEAD(error_q);
+	struct scsi_cmnd *cmd, *n;
+	enum task_disposition res = TASK_IS_DONE;
+	int tmf_resp;
+	struct sas_internal *i = to_sas_internal(shost->transportt);
+
+	spin_lock_irqsave(shost->host_lock, flags);
+	list_splice_init(&shost->eh_cmd_q, &error_q);
+	spin_unlock_irqrestore(shost->host_lock, flags);
+
+	SAS_DPRINTK("Enter %s\n", __FUNCTION__);
+
+	/* All tasks on this list were marked SAS_TASK_STATE_ABORTED
+	 * by sas_scsi_timed_out() callback.
+	 */
+Again:
+	SAS_DPRINTK("going over list...\n");
+	list_for_each_entry_safe(cmd, n, &error_q, eh_entry) {
+		struct sas_task *task = TO_SAS_TASK(cmd);
+
+		SAS_DPRINTK("trying to find task 0x%p\n", task);
+		list_del_init(&cmd->eh_entry);
+		res = sas_scsi_find_task(task);
+
+		cmd->eh_eflags = 0;
+		shost->host_failed--;
+
+		switch (res) {
+		case TASK_IS_DONE:
+			SAS_DPRINTK("%s: task 0x%p is done\n", __FUNCTION__,
+				    task);
+			task->task_done(task);
+			continue;
+		case TASK_IS_ABORTED:
+			SAS_DPRINTK("%s: task 0x%p is aborted\n",
+				    __FUNCTION__, task);
+			task->task_done(task);
+			continue;
+		case TASK_IS_AT_LU:
+			SAS_DPRINTK("task 0x%p is at LU: lu recover\n", task);
+			tmf_resp = sas_recover_lu(task->dev, cmd);
+			if (tmf_resp == TMF_RESP_FUNC_COMPLETE) {
+				SAS_DPRINTK("dev %016llx LU %x is "
+					    "recovered\n",
+					    SAS_ADDR(task->dev),
+					    cmd->device->lun);
+				task->task_done(task);
+				sas_scsi_clear_queue_lu(&error_q, cmd);
+				goto Again;
+			}
+			/* fallthrough */
+		case TASK_IS_NOT_AT_LU:
+			SAS_DPRINTK("task 0x%p is not at LU: I_T recover\n",
+				    task);
+			tmf_resp = sas_recover_I_T(task->dev);
+			if (tmf_resp == TMF_RESP_FUNC_COMPLETE) {
+				SAS_DPRINTK("I_T %016llx recovered\n",
+					    SAS_ADDR(task->dev->sas_addr));
+				task->task_done(task);
+				sas_scsi_clear_queue_I_T(&error_q, task->dev);
+				goto Again;
+			}
+			/* Hammer time :-) */
+			if (i->dft->lldd_clear_nexus_port) {
+				struct asd_sas_port *port = task->dev->port;
+				SAS_DPRINTK("clearing nexus for port:%d\n",
+					    port->id);
+				res = i->dft->lldd_clear_nexus_port(port);
+				if (res == TMF_RESP_FUNC_COMPLETE) {
+					SAS_DPRINTK("clear nexus port:%d "
+						    "succeeded\n", port->id);
+					task->task_done(task);
+					sas_scsi_clear_queue_port(&error_q,
+								  port);
+					goto Again;
+				}
+			}
+			if (i->dft->lldd_clear_nexus_ha) {
+				SAS_DPRINTK("clear nexus ha\n");
+				res = i->dft->lldd_clear_nexus_ha(ha);
+				if (res == TMF_RESP_FUNC_COMPLETE) {
+					SAS_DPRINTK("clear nexus ha "
+						    "succeeded\n");
+					task->task_done(task);
+					goto out;
+				}
+			}
+			/* If we are here -- this means that no amount
+			 * of effort could recover from errors.  Quite
+			 * possibly the HA just disappeared.
+			 */
+			SAS_DPRINTK("error from  device %llx, LUN %x "
+				    "couldn't be recovered in any way\n",
+				    SAS_ADDR(task->dev->sas_addr),
+				    cmd->device->lun);
+
+			task->task_done(task);
+			goto clear_q;
+		}
+	}
+out:
+	SAS_DPRINTK("--- Exit %s\n", __FUNCTION__);
+	return;
+clear_q:
+	SAS_DPRINTK("--- Exit %s -- clear_q\n", __FUNCTION__);
+	list_for_each_entry_safe(cmd, n, &error_q, eh_entry) {
+		struct sas_task *task = TO_SAS_TASK(cmd);
+		list_del_init(&cmd->eh_entry);
+		task->task_done(task);
+	}
+}
+
+enum scsi_eh_timer_return sas_scsi_timed_out(struct scsi_cmnd *cmd)
+{
+	struct sas_task *task = TO_SAS_TASK(cmd);
+	unsigned long flags;
+
+	if (!task) {
+		SAS_DPRINTK("command 0x%p, task 0x%p, timed out: EH_HANDLED\n",
+			    cmd, task);
+		return EH_HANDLED;
+	}
+
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	if (task->task_state_flags & SAS_TASK_STATE_DONE) {
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+		SAS_DPRINTK("command 0x%p, task 0x%p, timed out: EH_HANDLED\n",
+			    cmd, task);
+		return EH_HANDLED;
+	}
+	task->task_state_flags |= SAS_TASK_STATE_ABORTED;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	SAS_DPRINTK("command 0x%p, task 0x%p, timed out: EH_NOT_HANDLED\n",
+		    cmd, task);
+
+	return EH_NOT_HANDLED;
+}
+
+struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy)
+{
+	struct Scsi_Host *shost = dev_to_shost(rphy->dev.parent);
+	struct sas_ha_struct *ha = SHOST_TO_SAS_HA(shost);
+	struct domain_device *found_dev = NULL;
+	int i;
+
+	spin_lock(&ha->phy_port_lock);
+	for (i = 0; i < ha->num_phys; i++) {
+		struct asd_sas_port *port = ha->sas_port[i];
+		struct domain_device *dev;
+
+		spin_lock(&port->dev_list_lock);
+		list_for_each_entry(dev, &port->dev_list, dev_list_node) {
+			if (rphy == dev->rphy) {
+				found_dev = dev;
+				spin_unlock(&port->dev_list_lock);
+				goto found;
+			}
+		}
+		spin_unlock(&port->dev_list_lock);
+	}
+ found:
+	spin_unlock(&ha->phy_port_lock);
+
+	return found_dev;
+}
+
+static inline struct domain_device *sas_find_target(struct scsi_target *starget)
+{
+	struct sas_rphy *rphy = dev_to_rphy(starget->dev.parent);
+
+	return sas_find_dev_by_rphy(rphy);
+}
+
+int sas_target_alloc(struct scsi_target *starget)
+{
+	struct domain_device *found_dev = sas_find_target(starget);
+
+	if (!found_dev)
+		return -ENODEV;
+
+	starget->hostdata = found_dev;
+	return 0;
+}
+
+#define SAS_DEF_QD 32
+#define SAS_MAX_QD 64
+
+int sas_slave_configure(struct scsi_device *scsi_dev)
+{
+	struct domain_device *dev = sdev_to_domain_dev(scsi_dev);
+	struct sas_ha_struct *sas_ha;
+
+	BUG_ON(dev->rphy->identify.device_type != SAS_END_DEVICE);
+
+	sas_ha = dev->port->ha;
+
+	sas_read_port_mode_page(scsi_dev);
+
+	if (scsi_dev->tagged_supported) {
+		scsi_set_tag_type(scsi_dev, MSG_SIMPLE_TAG);
+		scsi_activate_tcq(scsi_dev, SAS_DEF_QD);
+	} else {
+		SAS_DPRINTK("device %llx, LUN %x doesn't support "
+			    "TCQ\n", SAS_ADDR(dev->sas_addr),
+			    scsi_dev->lun);
+		scsi_dev->tagged_supported = 0;
+		scsi_set_tag_type(scsi_dev, 0);
+		scsi_deactivate_tcq(scsi_dev, 1);
+	}
+
+	return 0;
+}
+
+void sas_slave_destroy(struct scsi_device *scsi_dev)
+{
+}
+
+int sas_change_queue_depth(struct scsi_device *scsi_dev, int new_depth)
+{
+	int res = min(new_depth, SAS_MAX_QD);
+
+	if (scsi_dev->tagged_supported)
+		scsi_adjust_queue_depth(scsi_dev, scsi_get_tag_type(scsi_dev),
+					res);
+	else {
+		struct domain_device *dev = sdev_to_domain_dev(scsi_dev);
+		sas_printk("device %llx LUN %x queue depth changed to 1\n",
+			   SAS_ADDR(dev->sas_addr),
+			   scsi_dev->lun);
+		scsi_adjust_queue_depth(scsi_dev, 0, 1);
+		res = 1;
+	}
+
+	return res;
+}
+
+int sas_change_queue_type(struct scsi_device *scsi_dev, int qt)
+{
+	if (!scsi_dev->tagged_supported)
+		return 0;
+
+	scsi_deactivate_tcq(scsi_dev, 1);
+
+	scsi_set_tag_type(scsi_dev, qt);
+	scsi_activate_tcq(scsi_dev, scsi_dev->queue_depth);
+
+	return qt;
+}
+
+int sas_bios_param(struct scsi_device *scsi_dev,
+			  struct block_device *bdev,
+			  sector_t capacity, int *hsc)
+{
+	hsc[0] = 255;
+	hsc[1] = 63;
+	sector_div(capacity, 255*63);
+	hsc[2] = capacity;
+
+	return 0;
+}
+
+/* ---------- Task Collector Thread implementation ---------- */
+
+static void sas_queue(struct sas_ha_struct *sas_ha)
+{
+	struct scsi_core *core = &sas_ha->core;
+	unsigned long flags;
+	LIST_HEAD(q);
+	int can_queue;
+	int res;
+	struct sas_internal *i = to_sas_internal(core->shost->transportt);
+
+	spin_lock_irqsave(&core->task_queue_lock, flags);
+	while (!core->queue_thread_kill &&
+	       !list_empty(&core->task_queue)) {
+
+		can_queue = sas_ha->lldd_queue_size - core->task_queue_size;
+		if (can_queue >= 0) {
+			can_queue = core->task_queue_size;
+			list_splice_init(&core->task_queue, &q);
+		} else {
+			struct list_head *a, *n;
+
+			can_queue = sas_ha->lldd_queue_size;
+			list_for_each_safe(a, n, &core->task_queue) {
+				list_move_tail(a, &q);
+				if (--can_queue == 0)
+					break;
+			}
+			can_queue = sas_ha->lldd_queue_size;
+		}
+		core->task_queue_size -= can_queue;
+		spin_unlock_irqrestore(&core->task_queue_lock, flags);
+		{
+			struct sas_task *task = list_entry(q.next,
+							   struct sas_task,
+							   list);
+			list_del_init(&q);
+			res = i->dft->lldd_execute_task(task, can_queue,
+							GFP_KERNEL);
+			if (unlikely(res))
+				__list_add(&q, task->list.prev, &task->list);
+		}
+		spin_lock_irqsave(&core->task_queue_lock, flags);
+		if (res) {
+			list_splice_init(&q, &core->task_queue); /*at head*/
+			core->task_queue_size += can_queue;
+		}
+	}
+	spin_unlock_irqrestore(&core->task_queue_lock, flags);
+}
+
+static DECLARE_COMPLETION(queue_th_comp);
+
+/**
+ * sas_queue_thread -- The Task Collector thread
+ * @_sas_ha: pointer to struct sas_ha
+ */
+static int sas_queue_thread(void *_sas_ha)
+{
+	struct sas_ha_struct *sas_ha = _sas_ha;
+	struct scsi_core *core = &sas_ha->core;
+
+	daemonize("sas_queue_%d", core->shost->host_no);
+	current->flags |= PF_NOFREEZE;
+
+	complete(&queue_th_comp);
+
+	while (1) {
+		down_interruptible(&core->queue_thread_sema);
+		sas_queue(sas_ha);
+		if (core->queue_thread_kill)
+			break;
+	}
+
+	complete(&queue_th_comp);
+
+	return 0;
+}
+
+int sas_init_queue(struct sas_ha_struct *sas_ha)
+{
+	int res;
+	struct scsi_core *core = &sas_ha->core;
+
+	spin_lock_init(&core->task_queue_lock);
+	core->task_queue_size = 0;
+	INIT_LIST_HEAD(&core->task_queue);
+	init_MUTEX_LOCKED(&core->queue_thread_sema);
+
+	res = kernel_thread(sas_queue_thread, sas_ha, 0);
+	if (res >= 0)
+		wait_for_completion(&queue_th_comp);
+
+	return res < 0 ? res : 0;
+}
+
+void sas_shutdown_queue(struct sas_ha_struct *sas_ha)
+{
+	unsigned long flags;
+	struct scsi_core *core = &sas_ha->core;
+	struct sas_task *task, *n;
+
+	init_completion(&queue_th_comp);
+	core->queue_thread_kill = 1;
+	up(&core->queue_thread_sema);
+	wait_for_completion(&queue_th_comp);
+
+	if (!list_empty(&core->task_queue))
+		SAS_DPRINTK("HA: %llx: scsi core task queue is NOT empty!?\n",
+			    SAS_ADDR(sas_ha->sas_addr));
+
+	spin_lock_irqsave(&core->task_queue_lock, flags);
+	list_for_each_entry_safe(task, n, &core->task_queue, list) {
+		struct scsi_cmnd *cmd = task->uldd_task;
+
+		list_del_init(&task->list);
+
+		ASSIGN_SAS_TASK(cmd, NULL);
+		sas_free_task(task);
+		cmd->result = DID_ABORT << 16;
+		cmd->scsi_done(cmd);
+	}
+	spin_unlock_irqrestore(&core->task_queue_lock, flags);
+}
+
+EXPORT_SYMBOL_GPL(sas_queuecommand);
+EXPORT_SYMBOL_GPL(sas_target_alloc);
+EXPORT_SYMBOL_GPL(sas_slave_configure);
+EXPORT_SYMBOL_GPL(sas_slave_destroy);
+EXPORT_SYMBOL_GPL(sas_change_queue_depth);
+EXPORT_SYMBOL_GPL(sas_change_queue_type);
+EXPORT_SYMBOL_GPL(sas_bios_param);
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
new file mode 100644
index 000000000000..72acdabe7f80
--- /dev/null
+++ b/include/scsi/libsas.h
@@ -0,0 +1,627 @@
+/*
+ * SAS host prototypes and structures header file
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ * USA
+ *
+ */
+
+#ifndef _LIBSAS_H_
+#define _LIBSAS_H_
+
+
+#include <linux/timer.h>
+#include <linux/pci.h>
+#include <scsi/sas.h>
+#include <linux/list.h>
+#include <asm/semaphore.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_transport_sas.h>
+
+struct block_device;
+
+enum sas_class {
+	SAS,
+	EXPANDER
+};
+
+enum sas_phy_role {
+	PHY_ROLE_NONE = 0,
+	PHY_ROLE_TARGET = 0x40,
+	PHY_ROLE_INITIATOR = 0x80,
+};
+
+enum sas_phy_type {
+        PHY_TYPE_PHYSICAL,
+        PHY_TYPE_VIRTUAL
+};
+
+/* The events are mnemonically described in sas_dump.c
+ * so when updating/adding events here, please also
+ * update the other file too.
+ */
+enum ha_event {
+	HAE_RESET             = 0U,
+	HA_NUM_EVENTS         = 1,
+};
+
+enum port_event {
+	PORTE_BYTES_DMAED     = 0U,
+	PORTE_BROADCAST_RCVD  = 1,
+	PORTE_LINK_RESET_ERR  = 2,
+	PORTE_TIMER_EVENT     = 3,
+	PORTE_HARD_RESET      = 4,
+	PORT_NUM_EVENTS       = 5,
+};
+
+enum phy_event {
+	PHYE_LOSS_OF_SIGNAL   = 0U,
+	PHYE_OOB_DONE         = 1,
+	PHYE_OOB_ERROR        = 2,
+	PHYE_SPINUP_HOLD      = 3, /* hot plug SATA, no COMWAKE sent */
+	PHY_NUM_EVENTS        = 4,
+};
+
+enum discover_event {
+	DISCE_DISCOVER_DOMAIN   = 0U,
+	DISCE_REVALIDATE_DOMAIN = 1,
+	DISCE_PORT_GONE         = 2,
+	DISC_NUM_EVENTS 	= 3,
+};
+
+/* ---------- Expander Devices ---------- */
+
+#define ETASK 0xFA
+
+#define to_dom_device(_obj) container_of(_obj, struct domain_device, dev_obj)
+#define to_dev_attr(_attr)  container_of(_attr, struct domain_dev_attribute,\
+                                         attr)
+
+enum routing_attribute {
+	DIRECT_ROUTING,
+	SUBTRACTIVE_ROUTING,
+	TABLE_ROUTING,
+};
+
+enum ex_phy_state {
+	PHY_EMPTY,
+	PHY_VACANT,
+	PHY_NOT_PRESENT,
+	PHY_DEVICE_DISCOVERED
+};
+
+struct ex_phy {
+	int    phy_id;
+
+	enum ex_phy_state phy_state;
+
+	enum sas_dev_type attached_dev_type;
+	enum sas_phy_linkrate linkrate;
+
+	u8   attached_sata_host:1;
+	u8   attached_sata_dev:1;
+	u8   attached_sata_ps:1;
+
+	enum sas_proto attached_tproto;
+	enum sas_proto attached_iproto;
+
+	u8   attached_sas_addr[SAS_ADDR_SIZE];
+	u8   attached_phy_id;
+
+	u8   phy_change_count;
+	enum routing_attribute routing_attr;
+	u8   virtual:1;
+
+	int  last_da_index;
+
+	struct sas_phy *phy;
+	struct sas_port *port;
+};
+
+struct expander_device {
+	struct list_head children;
+
+	u16    ex_change_count;
+	u16    max_route_indexes;
+	u8     num_phys;
+	u8     configuring:1;
+	u8     conf_route_table:1;
+	u8     enclosure_logical_id[8];
+
+	struct ex_phy *ex_phy;
+	struct sas_port *parent_port;
+};
+
+/* ---------- SATA device ---------- */
+enum ata_command_set {
+        ATA_COMMAND_SET   = 0,
+        ATAPI_COMMAND_SET = 1,
+};
+
+struct sata_device {
+        enum   ata_command_set command_set;
+        struct smp_resp        rps_resp; /* report_phy_sata_resp */
+        __le16 *identify_device;
+        __le16 *identify_packet_device;
+
+        u8     port_no;        /* port number, if this is a PM (Port) */
+        struct list_head children; /* PM Ports if this is a PM */
+};
+
+/* ---------- Domain device ---------- */
+struct domain_device {
+        enum sas_dev_type dev_type;
+
+        enum sas_phy_linkrate linkrate;
+        enum sas_phy_linkrate min_linkrate;
+        enum sas_phy_linkrate max_linkrate;
+
+        int  pathways;
+
+        struct domain_device *parent;
+        struct list_head siblings; /* devices on the same level */
+        struct asd_sas_port *port;        /* shortcut to root of the tree */
+
+        struct list_head dev_list_node;
+
+        enum sas_proto    iproto;
+        enum sas_proto    tproto;
+
+        struct sas_rphy *rphy;
+
+        u8  sas_addr[SAS_ADDR_SIZE];
+        u8  hashed_sas_addr[HASHED_SAS_ADDR_SIZE];
+
+        u8  frame_rcvd[32];
+
+        union {
+                struct expander_device ex_dev;
+                struct sata_device     sata_dev; /* STP & directly attached */
+        };
+
+        void *lldd_dev;
+};
+
+struct sas_discovery {
+	spinlock_t disc_event_lock;
+	struct work_struct disc_work[DISC_NUM_EVENTS];
+	unsigned long    pending;
+	u8     fanout_sas_addr[8];
+	u8     eeds_a[8];
+	u8     eeds_b[8];
+	int    max_level;
+};
+
+
+/* The port struct is Class:RW, driver:RO */
+struct asd_sas_port {
+/* private: */
+	struct completion port_gone_completion;
+
+	struct sas_discovery disc;
+	struct domain_device *port_dev;
+	spinlock_t dev_list_lock;
+	struct list_head dev_list;
+	enum   sas_phy_linkrate linkrate;
+
+	struct sas_phy *phy;
+	struct work_struct work;
+
+/* public: */
+	int id;
+
+	enum sas_class   class;
+	u8               sas_addr[SAS_ADDR_SIZE];
+	u8               attached_sas_addr[SAS_ADDR_SIZE];
+	enum sas_proto   iproto;
+	enum sas_proto   tproto;
+
+	enum sas_oob_mode oob_mode;
+
+	spinlock_t       phy_list_lock;
+	struct list_head phy_list;
+	int              num_phys;
+	u32              phy_mask;
+
+	struct sas_ha_struct *ha;
+
+	struct sas_port	*port;
+
+	void *lldd_port;	  /* not touched by the sas class code */
+};
+
+/* The phy pretty much is controlled by the LLDD.
+ * The class only reads those fields.
+ */
+struct asd_sas_phy {
+/* private: */
+	/* protected by ha->event_lock */
+	struct work_struct   port_events[PORT_NUM_EVENTS];
+	struct work_struct   phy_events[PHY_NUM_EVENTS];
+
+	unsigned long port_events_pending;
+	unsigned long phy_events_pending;
+
+	int error;
+
+	struct sas_phy *phy;
+
+/* public: */
+	/* The following are class:RO, driver:R/W */
+	int            enabled;	  /* must be set */
+
+	int            id;	  /* must be set */
+	enum sas_class class;
+	enum sas_proto iproto;
+	enum sas_proto tproto;
+
+	enum sas_phy_type  type;
+	enum sas_phy_role  role;
+	enum sas_oob_mode  oob_mode;
+	enum sas_phy_linkrate linkrate;
+
+	u8   *sas_addr;		  /* must be set */
+	u8   attached_sas_addr[SAS_ADDR_SIZE]; /* class:RO, driver: R/W */
+
+	spinlock_t     frame_rcvd_lock;
+	u8             *frame_rcvd; /* must be set */
+	int            frame_rcvd_size;
+
+	spinlock_t     sas_prim_lock;
+	u32            sas_prim;
+
+	struct list_head port_phy_el; /* driver:RO */
+	struct asd_sas_port      *port; /* Class:RW, driver: RO */
+
+	struct sas_ha_struct *ha; /* may be set; the class sets it anyway */
+
+	void *lldd_phy;		  /* not touched by the sas_class_code */
+};
+
+struct scsi_core {
+	struct Scsi_Host *shost;
+
+	spinlock_t        task_queue_lock;
+	struct list_head  task_queue;
+	int               task_queue_size;
+
+	struct semaphore  queue_thread_sema;
+	int               queue_thread_kill;
+};
+
+struct sas_ha_struct {
+/* private: */
+	spinlock_t       event_lock;
+	struct work_struct ha_events[HA_NUM_EVENTS];
+	unsigned long	 pending;
+
+	struct scsi_core core;
+
+/* public: */
+	char *sas_ha_name;
+	struct pci_dev *pcidev;	  /* should be set */
+	struct module *lldd_module; /* should be set */
+
+	u8 *sas_addr;		  /* must be set */
+	u8 hashed_sas_addr[HASHED_SAS_ADDR_SIZE];
+
+	spinlock_t      phy_port_lock;
+	struct asd_sas_phy  **sas_phy; /* array of valid pointers, must be set */
+	struct asd_sas_port **sas_port; /* array of valid pointers, must be set */
+	int             num_phys; /* must be set, gt 0, static */
+
+	/* The class calls this to send a task for execution. */
+	int lldd_max_execute_num;
+	int lldd_queue_size;
+
+	/* LLDD calls these to notify the class of an event. */
+	void (*notify_ha_event)(struct sas_ha_struct *, enum ha_event);
+	void (*notify_port_event)(struct asd_sas_phy *, enum port_event);
+	void (*notify_phy_event)(struct asd_sas_phy *, enum phy_event);
+
+	void *lldd_ha;		  /* not touched by sas class code */
+};
+
+#define SHOST_TO_SAS_HA(_shost) (*(struct sas_ha_struct **)(_shost)->hostdata)
+
+static inline struct domain_device *
+starget_to_domain_dev(struct scsi_target *starget) {
+	return starget->hostdata;
+}
+
+static inline struct domain_device *
+sdev_to_domain_dev(struct scsi_device *sdev) {
+	return starget_to_domain_dev(sdev->sdev_target);
+}
+
+static inline struct domain_device *
+cmd_to_domain_dev(struct scsi_cmnd *cmd)
+{
+	return sdev_to_domain_dev(cmd->device);
+}
+
+void sas_hash_addr(u8 *hashed, const u8 *sas_addr);
+
+/* Before calling a notify event, LLDD should use this function
+ * when the link is severed (possibly from its tasklet).
+ * The idea is that the Class only reads those, while the LLDD,
+ * can R/W these (thus avoiding a race).
+ */
+static inline void sas_phy_disconnected(struct asd_sas_phy *phy)
+{
+	phy->oob_mode = OOB_NOT_CONNECTED;
+	phy->linkrate = PHY_LINKRATE_NONE;
+}
+
+/* ---------- Tasks ---------- */
+/*
+      service_response |  SAS_TASK_COMPLETE  |  SAS_TASK_UNDELIVERED |
+  exec_status          |                     |                       |
+  ---------------------+---------------------+-----------------------+
+       SAM_...         |         X           |                       |
+       DEV_NO_RESPONSE |         X           |           X           |
+       INTERRUPTED     |         X           |                       |
+       QUEUE_FULL      |                     |           X           |
+       DEVICE_UNKNOWN  |                     |           X           |
+       SG_ERR          |                     |           X           |
+  ---------------------+---------------------+-----------------------+
+ */
+
+enum service_response {
+	SAS_TASK_COMPLETE,
+	SAS_TASK_UNDELIVERED = -1,
+};
+
+enum exec_status {
+	SAM_GOOD         = 0,
+	SAM_CHECK_COND   = 2,
+	SAM_COND_MET     = 4,
+	SAM_BUSY         = 8,
+	SAM_INTERMEDIATE = 0x10,
+	SAM_IM_COND_MET  = 0x12,
+	SAM_RESV_CONFLICT= 0x14,
+	SAM_TASK_SET_FULL= 0x28,
+	SAM_ACA_ACTIVE   = 0x30,
+	SAM_TASK_ABORTED = 0x40,
+
+	SAS_DEV_NO_RESPONSE = 0x80,
+	SAS_DATA_UNDERRUN,
+	SAS_DATA_OVERRUN,
+	SAS_INTERRUPTED,
+	SAS_QUEUE_FULL,
+	SAS_DEVICE_UNKNOWN,
+	SAS_SG_ERR,
+	SAS_OPEN_REJECT,
+	SAS_OPEN_TO,
+	SAS_PROTO_RESPONSE,
+	SAS_PHY_DOWN,
+	SAS_NAK_R_ERR,
+	SAS_PENDING,
+	SAS_ABORTED_TASK,
+};
+
+/* When a task finishes with a response, the LLDD examines the
+ * response:
+ * 	- For an ATA task task_status_struct::stat is set to
+ * SAS_PROTO_RESPONSE, and the task_status_struct::buf is set to the
+ * contents of struct ata_task_resp.
+ * 	- For SSP tasks, if no data is present or status/TMF response
+ * is valid, task_status_struct::stat is set.  If data is present
+ * (SENSE data), the LLDD copies up to SAS_STATUS_BUF_SIZE, sets
+ * task_status_struct::buf_valid_size, and task_status_struct::stat is
+ * set to SAM_CHECK_COND.
+ *
+ * "buf" has format SCSI Sense for SSP task, or struct ata_task_resp
+ * for ATA task.
+ *
+ * "frame_len" is the total frame length, which could be more or less
+ * than actually copied.
+ *
+ * Tasks ending with response, always set the residual field.
+ */
+struct ata_task_resp {
+	u16  frame_len;
+	u8   ending_fis[24];	  /* dev to host or data-in */
+	u32  sstatus;
+	u32  serror;
+	u32  scontrol;
+	u32  sactive;
+};
+
+#define SAS_STATUS_BUF_SIZE 96
+
+struct task_status_struct {
+	enum service_response resp;
+	enum exec_status      stat;
+	int  buf_valid_size;
+
+	u8   buf[SAS_STATUS_BUF_SIZE];
+
+	u32  residual;
+	enum sas_open_rej_reason open_rej_reason;
+};
+
+/* ATA and ATAPI task queuable to a SAS LLDD.
+ */
+struct sas_ata_task {
+	struct host_to_dev_fis fis;
+	u8     atapi_packet[16];  /* 0 if not ATAPI task */
+
+	u8     retry_count;	  /* hardware retry, should be > 0 */
+
+	u8     dma_xfer:1;	  /* PIO:0 or DMA:1 */
+	u8     use_ncq:1;
+	u8     set_affil_pol:1;
+	u8     stp_affil_pol:1;
+
+	u8     device_control_reg_update:1;
+};
+
+struct sas_smp_task {
+	struct scatterlist smp_req;
+	struct scatterlist smp_resp;
+};
+
+enum task_attribute {
+	TASK_ATTR_SIMPLE = 0,
+	TASK_ATTR_HOQ    = 1,
+	TASK_ATTR_ORDERED= 2,
+	TASK_ATTR_ACA    = 4,
+};
+
+struct sas_ssp_task {
+	u8     retry_count;	  /* hardware retry, should be > 0 */
+
+	u8     LUN[8];
+	u8     enable_first_burst:1;
+	enum   task_attribute task_attr;
+	u8     task_prio;
+	u8     cdb[16];
+};
+
+struct sas_task {
+	struct domain_device *dev;
+	struct list_head      list;
+
+	spinlock_t   task_state_lock;
+	unsigned     task_state_flags;
+
+	enum   sas_proto      task_proto;
+
+	/* Used by the discovery code. */
+	struct timer_list     timer;
+	struct completion     completion;
+
+	union {
+		struct sas_ata_task ata_task;
+		struct sas_smp_task smp_task;
+		struct sas_ssp_task ssp_task;
+	};
+
+	struct scatterlist *scatter;
+	int    num_scatter;
+	u32    total_xfer_len;
+	u8     data_dir:2;	  /* Use PCI_DMA_... */
+
+	struct task_status_struct task_status;
+	void   (*task_done)(struct sas_task *);
+
+	void   *lldd_task;	  /* for use by LLDDs */
+	void   *uldd_task;
+};
+
+
+
+#define SAS_TASK_STATE_PENDING  1
+#define SAS_TASK_STATE_DONE     2
+#define SAS_TASK_STATE_ABORTED  4
+
+static inline struct sas_task *sas_alloc_task(unsigned long flags)
+{
+	extern kmem_cache_t *sas_task_cache;
+	struct sas_task *task = kmem_cache_alloc(sas_task_cache, flags);
+
+	if (task) {
+		memset(task, 0, sizeof(*task));
+		INIT_LIST_HEAD(&task->list);
+		spin_lock_init(&task->task_state_lock);
+		task->task_state_flags = SAS_TASK_STATE_PENDING;
+		init_timer(&task->timer);
+		init_completion(&task->completion);
+	}
+
+	return task;
+}
+
+static inline void sas_free_task(struct sas_task *task)
+{
+	if (task) {
+		extern kmem_cache_t *sas_task_cache;
+		BUG_ON(!list_empty(&task->list));
+		kmem_cache_free(sas_task_cache, task);
+	}
+}
+
+struct sas_domain_function_template {
+	/* The class calls these to notify the LLDD of an event. */
+	void (*lldd_port_formed)(struct asd_sas_phy *);
+	void (*lldd_port_deformed)(struct asd_sas_phy *);
+
+	/* The class calls these when a device is found or gone. */
+	int  (*lldd_dev_found)(struct domain_device *);
+	void (*lldd_dev_gone)(struct domain_device *);
+
+	int (*lldd_execute_task)(struct sas_task *, int num,
+				 unsigned long gfp_flags);
+
+	/* Task Management Functions. Must be called from process context. */
+	int (*lldd_abort_task)(struct sas_task *);
+	int (*lldd_abort_task_set)(struct domain_device *, u8 *lun);
+	int (*lldd_clear_aca)(struct domain_device *, u8 *lun);
+	int (*lldd_clear_task_set)(struct domain_device *, u8 *lun);
+	int (*lldd_I_T_nexus_reset)(struct domain_device *);
+	int (*lldd_lu_reset)(struct domain_device *, u8 *lun);
+	int (*lldd_query_task)(struct sas_task *);
+
+	/* Port and Adapter management */
+	int (*lldd_clear_nexus_port)(struct asd_sas_port *);
+	int (*lldd_clear_nexus_ha)(struct sas_ha_struct *);
+
+	/* Phy management */
+	int (*lldd_control_phy)(struct asd_sas_phy *, enum phy_func);
+};
+
+extern int sas_register_ha(struct sas_ha_struct *);
+extern int sas_unregister_ha(struct sas_ha_struct *);
+
+extern int sas_queuecommand(struct scsi_cmnd *,
+		     void (*scsi_done)(struct scsi_cmnd *));
+extern int sas_target_alloc(struct scsi_target *);
+extern int sas_slave_alloc(struct scsi_device *);
+extern int sas_slave_configure(struct scsi_device *);
+extern void sas_slave_destroy(struct scsi_device *);
+extern int sas_change_queue_depth(struct scsi_device *, int new_depth);
+extern int sas_change_queue_type(struct scsi_device *, int qt);
+extern int sas_bios_param(struct scsi_device *,
+			  struct block_device *,
+			  sector_t capacity, int *hsc);
+extern struct scsi_transport_template *
+sas_domain_attach_transport(struct sas_domain_function_template *);
+extern void sas_domain_release_transport(struct scsi_transport_template *);
+
+int  sas_discover_root_expander(struct domain_device *);
+
+void sas_init_ex_attr(void);
+
+int  sas_ex_revalidate_domain(struct domain_device *);
+
+void sas_unregister_domain_devices(struct asd_sas_port *port);
+void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *);
+int  sas_discover_event(struct asd_sas_port *, enum discover_event ev);
+
+int  sas_discover_sata(struct domain_device *);
+int  sas_discover_end_dev(struct domain_device *);
+
+void sas_unregister_dev(struct domain_device *);
+
+void sas_init_dev(struct domain_device *);
+
+#endif /* _SASLIB_H_ */
diff --git a/include/scsi/sas.h b/include/scsi/sas.h
new file mode 100644
index 000000000000..752853a113dc
--- /dev/null
+++ b/include/scsi/sas.h
@@ -0,0 +1,644 @@
+/*
+ * SAS structures and definitions header file
+ *
+ * Copyright (C) 2005 Adaptec, Inc.  All rights reserved.
+ * Copyright (C) 2005 Luben Tuikov <luben_tuikov@adaptec.com>
+ *
+ * This file is licensed under GPLv2.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ * USA
+ *
+ */
+
+#ifndef _SAS_H_
+#define _SAS_H_
+
+#include <linux/types.h>
+#include <asm/byteorder.h>
+
+#define SAS_ADDR_SIZE        8
+#define HASHED_SAS_ADDR_SIZE 3
+#define SAS_ADDR(_sa) ((unsigned long long) be64_to_cpu(*(__be64 *)(_sa)))
+
+#define SMP_REQUEST             0x40
+#define SMP_RESPONSE            0x41
+
+#define SSP_DATA                0x01
+#define SSP_XFER_RDY            0x05
+#define SSP_COMMAND             0x06
+#define SSP_RESPONSE            0x07
+#define SSP_TASK                0x16
+
+#define SMP_REPORT_GENERAL       0x00
+#define SMP_REPORT_MANUF_INFO    0x01
+#define SMP_READ_GPIO_REG        0x02
+#define SMP_DISCOVER             0x10
+#define SMP_REPORT_PHY_ERR_LOG   0x11
+#define SMP_REPORT_PHY_SATA      0x12
+#define SMP_REPORT_ROUTE_INFO    0x13
+#define SMP_WRITE_GPIO_REG       0x82
+#define SMP_CONF_ROUTE_INFO      0x90
+#define SMP_PHY_CONTROL          0x91
+#define SMP_PHY_TEST_FUNCTION    0x92
+
+#define SMP_RESP_FUNC_ACC        0x00
+#define SMP_RESP_FUNC_UNK        0x01
+#define SMP_RESP_FUNC_FAILED     0x02
+#define SMP_RESP_INV_FRM_LEN     0x03
+#define SMP_RESP_NO_PHY          0x10
+#define SMP_RESP_NO_INDEX        0x11
+#define SMP_RESP_PHY_NO_SATA     0x12
+#define SMP_RESP_PHY_UNK_OP      0x13
+#define SMP_RESP_PHY_UNK_TESTF   0x14
+#define SMP_RESP_PHY_TEST_INPROG 0x15
+#define SMP_RESP_PHY_VACANT      0x16
+
+/* SAM TMFs */
+#define TMF_ABORT_TASK      0x01
+#define TMF_ABORT_TASK_SET  0x02
+#define TMF_CLEAR_TASK_SET  0x04
+#define TMF_LU_RESET        0x08
+#define TMF_CLEAR_ACA       0x40
+#define TMF_QUERY_TASK      0x80
+
+/* SAS TMF responses */
+#define TMF_RESP_FUNC_COMPLETE   0x00
+#define TMF_RESP_INVALID_FRAME   0x02
+#define TMF_RESP_FUNC_ESUPP      0x04
+#define TMF_RESP_FUNC_FAILED     0x05
+#define TMF_RESP_FUNC_SUCC       0x08
+#define TMF_RESP_NO_LUN          0x09
+#define TMF_RESP_OVERLAPPED_TAG  0x0A
+
+enum sas_oob_mode {
+	OOB_NOT_CONNECTED,
+	SATA_OOB_MODE,
+	SAS_OOB_MODE
+};
+
+/* See sas_discover.c if you plan on changing these.
+ */
+enum sas_dev_type {
+	NO_DEVICE   = 0,	  /* protocol */
+	SAS_END_DEV = 1,	  /* protocol */
+	EDGE_DEV    = 2,	  /* protocol */
+	FANOUT_DEV  = 3,	  /* protocol */
+	SAS_HA      = 4,
+	SATA_DEV    = 5,
+	SATA_PM     = 7,
+	SATA_PM_PORT= 8,
+};
+
+enum sas_phy_linkrate {
+	PHY_LINKRATE_NONE = 0,
+	PHY_LINKRATE_UNKNOWN = 0,
+	PHY_DISABLED,
+	PHY_RESET_PROBLEM,
+	PHY_SPINUP_HOLD,
+	PHY_PORT_SELECTOR,
+	PHY_LINKRATE_1_5 = 0x08,
+	PHY_LINKRATE_G1  = PHY_LINKRATE_1_5,
+	PHY_LINKRATE_3   = 0x09,
+	PHY_LINKRATE_G2  = PHY_LINKRATE_3,
+	PHY_LINKRATE_6   = 0x0A,
+};
+
+/* Partly from IDENTIFY address frame. */
+enum sas_proto {
+	SATA_PROTO    = 1,
+	SAS_PROTO_SMP = 2,	  /* protocol */
+	SAS_PROTO_STP = 4,	  /* protocol */
+	SAS_PROTO_SSP = 8,	  /* protocol */
+	SAS_PROTO_ALL = 0xE,
+};
+
+/* From the spec; local phys only */
+enum phy_func {
+	PHY_FUNC_NOP,
+	PHY_FUNC_LINK_RESET,		  /* Enables the phy */
+	PHY_FUNC_HARD_RESET,
+	PHY_FUNC_DISABLE,
+	PHY_FUNC_CLEAR_ERROR_LOG = 5,
+	PHY_FUNC_CLEAR_AFFIL,
+	PHY_FUNC_TX_SATA_PS_SIGNAL,
+	PHY_FUNC_RELEASE_SPINUP_HOLD = 0x10, /* LOCAL PORT ONLY! */
+};
+
+/* SAS LLDD would need to report only _very_few_ of those, like BROADCAST.
+ * Most of those are here for completeness.
+ */
+enum sas_prim {
+	SAS_PRIM_AIP_NORMAL = 1,
+	SAS_PRIM_AIP_R0     = 2,
+	SAS_PRIM_AIP_R1     = 3,
+	SAS_PRIM_AIP_R2     = 4,
+	SAS_PRIM_AIP_WC     = 5,
+	SAS_PRIM_AIP_WD     = 6,
+	SAS_PRIM_AIP_WP     = 7,
+	SAS_PRIM_AIP_RWP    = 8,
+
+	SAS_PRIM_BC_CH      = 9,
+	SAS_PRIM_BC_RCH0    = 10,
+	SAS_PRIM_BC_RCH1    = 11,
+	SAS_PRIM_BC_R0      = 12,
+	SAS_PRIM_BC_R1      = 13,
+	SAS_PRIM_BC_R2      = 14,
+	SAS_PRIM_BC_R3      = 15,
+	SAS_PRIM_BC_R4      = 16,
+
+	SAS_PRIM_NOTIFY_ENSP= 17,
+	SAS_PRIM_NOTIFY_R0  = 18,
+	SAS_PRIM_NOTIFY_R1  = 19,
+	SAS_PRIM_NOTIFY_R2  = 20,
+
+	SAS_PRIM_CLOSE_CLAF = 21,
+	SAS_PRIM_CLOSE_NORM = 22,
+	SAS_PRIM_CLOSE_R0   = 23,
+	SAS_PRIM_CLOSE_R1   = 24,
+
+	SAS_PRIM_OPEN_RTRY  = 25,
+	SAS_PRIM_OPEN_RJCT  = 26,
+	SAS_PRIM_OPEN_ACPT  = 27,
+
+	SAS_PRIM_DONE       = 28,
+	SAS_PRIM_BREAK      = 29,
+
+	SATA_PRIM_DMAT      = 33,
+	SATA_PRIM_PMNAK     = 34,
+	SATA_PRIM_PMACK     = 35,
+	SATA_PRIM_PMREQ_S   = 36,
+	SATA_PRIM_PMREQ_P   = 37,
+	SATA_SATA_R_ERR     = 38,
+};
+
+enum sas_open_rej_reason {
+	/* Abandon open */
+	SAS_OREJ_UNKNOWN   = 0,
+	SAS_OREJ_BAD_DEST  = 1,
+	SAS_OREJ_CONN_RATE = 2,
+	SAS_OREJ_EPROTO    = 3,
+	SAS_OREJ_RESV_AB0  = 4,
+	SAS_OREJ_RESV_AB1  = 5,
+	SAS_OREJ_RESV_AB2  = 6,
+	SAS_OREJ_RESV_AB3  = 7,
+	SAS_OREJ_WRONG_DEST= 8,
+	SAS_OREJ_STP_NORES = 9,
+
+	/* Retry open */
+	SAS_OREJ_NO_DEST   = 10,
+	SAS_OREJ_PATH_BLOCKED = 11,
+	SAS_OREJ_RSVD_CONT0 = 12,
+	SAS_OREJ_RSVD_CONT1 = 13,
+	SAS_OREJ_RSVD_INIT0 = 14,
+	SAS_OREJ_RSVD_INIT1 = 15,
+	SAS_OREJ_RSVD_STOP0 = 16,
+	SAS_OREJ_RSVD_STOP1 = 17,
+	SAS_OREJ_RSVD_RETRY = 18,
+};
+
+struct  dev_to_host_fis {
+	u8     fis_type;	  /* 0x34 */
+	u8     flags;
+	u8     status;
+	u8     error;
+
+	u8     lbal;
+	union { u8 lbam; u8 byte_count_low; };
+	union { u8 lbah; u8 byte_count_high; };
+	u8     device;
+
+	u8     lbal_exp;
+	u8     lbam_exp;
+	u8     lbah_exp;
+	u8     _r_a;
+
+	union { u8  sector_count; u8 interrupt_reason; };
+	u8     sector_count_exp;
+	u8     _r_b;
+	u8     _r_c;
+
+	u32    _r_d;
+} __attribute__ ((packed));
+
+struct host_to_dev_fis {
+	u8     fis_type;	  /* 0x27 */
+	u8     flags;
+	u8     command;
+	u8     features;
+
+	u8     lbal;
+	union { u8 lbam; u8 byte_count_low; };
+	union { u8 lbah; u8 byte_count_high; };
+	u8     device;
+
+	u8     lbal_exp;
+	u8     lbam_exp;
+	u8     lbah_exp;
+	u8     features_exp;
+
+	union { u8  sector_count; u8 interrupt_reason; };
+	u8     sector_count_exp;
+	u8     _r_a;
+	u8     control;
+
+	u32    _r_b;
+} __attribute__ ((packed));
+
+/* Prefer to have code clarity over header file clarity.
+ */
+#ifdef __LITTLE_ENDIAN_BITFIELD
+struct sas_identify_frame {
+	/* Byte 0 */
+	u8  frame_type:4;
+	u8  dev_type:3;
+	u8  _un0:1;
+
+	/* Byte 1 */
+	u8  _un1;
+
+	/* Byte 2 */
+	union {
+		struct {
+			u8  _un20:1;
+			u8  smp_iport:1;
+			u8  stp_iport:1;
+			u8  ssp_iport:1;
+			u8  _un247:4;
+		};
+		u8 initiator_bits;
+	};
+
+	/* Byte 3 */
+	union {
+		struct {
+			u8  _un30:1;
+			u8 smp_tport:1;
+			u8 stp_tport:1;
+			u8 ssp_tport:1;
+			u8 _un347:4;
+		};
+		u8 target_bits;
+	};
+
+	/* Byte 4 - 11 */
+	u8 _un4_11[8];
+
+	/* Byte 12 - 19 */
+	u8 sas_addr[SAS_ADDR_SIZE];
+
+	/* Byte 20 */
+	u8 phy_id;
+
+	u8 _un21_27[7];
+
+	__be32 crc;
+} __attribute__ ((packed));
+
+struct ssp_frame_hdr {
+	u8     frame_type;
+	u8     hashed_dest_addr[HASHED_SAS_ADDR_SIZE];
+	u8     _r_a;
+	u8     hashed_src_addr[HASHED_SAS_ADDR_SIZE];
+	__be16 _r_b;
+
+	u8     changing_data_ptr:1;
+	u8     retransmit:1;
+	u8     retry_data_frames:1;
+	u8     _r_c:5;
+
+	u8     num_fill_bytes:2;
+	u8     _r_d:6;
+
+	u32    _r_e;
+	__be16 tag;
+	__be16 tptt;
+	__be32 data_offs;
+} __attribute__ ((packed));
+
+struct ssp_response_iu {
+	u8     _r_a[10];
+
+	u8     datapres:2;
+	u8     _r_b:6;
+
+	u8     status;
+
+	u32    _r_c;
+
+	__be32 sense_data_len;
+	__be32 response_data_len;
+
+	u8     resp_data[0];
+	u8     sense_data[0];
+} __attribute__ ((packed));
+
+/* ---------- SMP ---------- */
+
+struct report_general_resp {
+	__be16  change_count;
+	__be16  route_indexes;
+	u8      _r_a;
+	u8      num_phys;
+
+	u8      conf_route_table:1;
+	u8      configuring:1;
+	u8      _r_b:6;
+
+	u8      _r_c;
+
+	u8      enclosure_logical_id[8];
+
+	u8      _r_d[12];
+} __attribute__ ((packed));
+
+struct discover_resp {
+	u8    _r_a[5];
+
+	u8    phy_id;
+	__be16 _r_b;
+
+	u8    _r_c:4;
+	u8    attached_dev_type:3;
+	u8    _r_d:1;
+
+	u8    linkrate:4;
+	u8    _r_e:4;
+
+	u8    attached_sata_host:1;
+	u8    iproto:3;
+	u8    _r_f:4;
+
+	u8    attached_sata_dev:1;
+	u8    tproto:3;
+	u8    _r_g:3;
+	u8    attached_sata_ps:1;
+
+	u8    sas_addr[8];
+	u8    attached_sas_addr[8];
+	u8    attached_phy_id;
+
+	u8    _r_h[7];
+
+	u8    hmin_linkrate:4;
+	u8    pmin_linkrate:4;
+	u8    hmax_linkrate:4;
+	u8    pmax_linkrate:4;
+
+	u8    change_count;
+
+	u8    pptv:4;
+	u8    _r_i:3;
+	u8    virtual:1;
+
+	u8    routing_attr:4;
+	u8    _r_j:4;
+
+	u8    conn_type;
+	u8    conn_el_index;
+	u8    conn_phy_link;
+
+	u8    _r_k[8];
+} __attribute__ ((packed));
+
+struct report_phy_sata_resp {
+	u8    _r_a[5];
+
+	u8    phy_id;
+	u8    _r_b;
+
+	u8    affil_valid:1;
+	u8    affil_supp:1;
+	u8    _r_c:6;
+
+	u32    _r_d;
+
+	u8    stp_sas_addr[8];
+
+	struct dev_to_host_fis fis;
+
+	u32   _r_e;
+
+	u8    affil_stp_ini_addr[8];
+
+	__be32 crc;
+} __attribute__ ((packed));
+
+struct smp_resp {
+	u8    frame_type;
+	u8    function;
+	u8    result;
+	u8    reserved;
+	union {
+		struct report_general_resp  rg;
+		struct discover_resp        disc;
+		struct report_phy_sata_resp rps;
+	};
+} __attribute__ ((packed));
+
+#elif defined(__BIG_ENDIAN_BITFIELD)
+struct sas_identify_frame {
+	/* Byte 0 */
+	u8  _un0:1;
+	u8  dev_type:3;
+	u8  frame_type:4;
+
+	/* Byte 1 */
+	u8  _un1;
+
+	/* Byte 2 */
+	union {
+		struct {
+			u8  _un247:4;
+			u8  ssp_iport:1;
+			u8  stp_iport:1;
+			u8  smp_iport:1;
+			u8  _un20:1;
+		};
+		u8 initiator_bits;
+	};
+
+	/* Byte 3 */
+	union {
+		struct {
+			u8 _un347:4;
+			u8 ssp_tport:1;
+			u8 stp_tport:1;
+			u8 smp_tport:1;
+			u8 _un30:1;
+		};
+		u8 target_bits;
+	};
+
+	/* Byte 4 - 11 */
+	u8 _un4_11[8];
+
+	/* Byte 12 - 19 */
+	u8 sas_addr[SAS_ADDR_SIZE];
+
+	/* Byte 20 */
+	u8 phy_id;
+
+	u8 _un21_27[7];
+
+	__be32 crc;
+} __attribute__ ((packed));
+
+struct ssp_frame_hdr {
+	u8     frame_type;
+	u8     hashed_dest_addr[HASHED_SAS_ADDR_SIZE];
+	u8     _r_a;
+	u8     hashed_src_addr[HASHED_SAS_ADDR_SIZE];
+	__be16 _r_b;
+
+	u8     _r_c:5;
+	u8     retry_data_frames:1;
+	u8     retransmit:1;
+	u8     changing_data_ptr:1;
+
+	u8     _r_d:6;
+	u8     num_fill_bytes:2;
+
+	u32    _r_e;
+	__be16 tag;
+	__be16 tptt;
+	__be32 data_offs;
+} __attribute__ ((packed));
+
+struct ssp_response_iu {
+	u8     _r_a[10];
+
+	u8     _r_b:6;
+	u8     datapres:2;
+
+	u8     status;
+
+	u32    _r_c;
+
+	__be32 sense_data_len;
+	__be32 response_data_len;
+
+	u8     resp_data[0];
+	u8     sense_data[0];
+} __attribute__ ((packed));
+
+/* ---------- SMP ---------- */
+
+struct report_general_resp {
+	__be16  change_count;
+	__be16  route_indexes;
+	u8      _r_a;
+	u8      num_phys;
+
+	u8      _r_b:6;
+	u8      configuring:1;
+	u8      conf_route_table:1;
+
+	u8      _r_c;
+
+	u8      enclosure_logical_id[8];
+
+	u8      _r_d[12];
+} __attribute__ ((packed));
+
+struct discover_resp {
+	u8    _r_a[5];
+
+	u8    phy_id;
+	__be16 _r_b;
+
+	u8    _r_d:1;
+	u8    attached_dev_type:3;
+	u8    _r_c:4;
+
+	u8    _r_e:4;
+	u8    linkrate:4;
+
+	u8    _r_f:4;
+	u8    iproto:3;
+	u8    attached_sata_host:1;
+
+	u8    attached_sata_ps:1;
+	u8    _r_g:3;
+	u8    tproto:3;
+	u8    attached_sata_dev:1;
+
+	u8    sas_addr[8];
+	u8    attached_sas_addr[8];
+	u8    attached_phy_id;
+
+	u8    _r_h[7];
+
+	u8    pmin_linkrate:4;
+	u8    hmin_linkrate:4;
+	u8    pmax_linkrate:4;
+	u8    hmax_linkrate:4;
+
+	u8    change_count;
+
+	u8    virtual:1;
+	u8    _r_i:3;
+	u8    pptv:4;
+
+	u8    _r_j:4;
+	u8    routing_attr:4;
+
+	u8    conn_type;
+	u8    conn_el_index;
+	u8    conn_phy_link;
+
+	u8    _r_k[8];
+} __attribute__ ((packed));
+
+struct report_phy_sata_resp {
+	u8    _r_a[5];
+
+	u8    phy_id;
+	u8    _r_b;
+
+	u8    _r_c:6;
+	u8    affil_supp:1;
+	u8    affil_valid:1;
+
+	u32   _r_d;
+
+	u8    stp_sas_addr[8];
+
+	struct dev_to_host_fis fis;
+
+	u32   _r_e;
+
+	u8    affil_stp_ini_addr[8];
+
+	__be32 crc;
+} __attribute__ ((packed));
+
+struct smp_resp {
+	u8    frame_type;
+	u8    function;
+	u8    result;
+	u8    reserved;
+	union {
+		struct report_general_resp  rg;
+		struct discover_resp        disc;
+		struct report_phy_sata_resp rps;
+	};
+} __attribute__ ((packed));
+
+#else
+#error "Bitfield order not defined!"
+#endif
+
+#endif /* _SAS_H_ */
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index 1bc675201413..84a6d5fe0920 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -429,4 +429,10 @@ struct scsi_lun {
 /* Used to obtain the PCI location of a device */
 #define SCSI_IOCTL_GET_PCI		0x5387
 
+/* Pull a u32 out of a SCSI message (using BE SCSI conventions) */
+static inline u32 scsi_to_u32(u8 *ptr)
+{
+	return (ptr[0]<<24) + (ptr[1]<<16) + (ptr[2]<<8) + ptr[3];
+}
+
 #endif /* _SCSI_SCSI_H */
-- 
cgit v1.2.3


From 669a5db411d85a14f86cd92bc16bf7ab5b8aa235 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Tue, 29 Aug 2006 18:12:40 -0400
Subject: [libata] Add a bunch of PATA drivers.

The vast majority of drivers and changes are from Alan Cox.  Albert Lee
contributed and maintains pata_pdc2027x.  Adrian Bunk, Andrew Morton,
and Tejun Heo contributed various minor fixes and updates.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/Kconfig             |  334 +++++++++++
 drivers/ata/Makefile            |   41 ++
 drivers/ata/ata_generic.c       |  252 ++++++++
 drivers/ata/ata_piix.c          |  436 ++++++++++----
 drivers/ata/pata_ali.c          |  679 +++++++++++++++++++++
 drivers/ata/pata_amd.c          |  707 ++++++++++++++++++++++
 drivers/ata/pata_artop.c        |  518 ++++++++++++++++
 drivers/ata/pata_atiixp.c       |  306 ++++++++++
 drivers/ata/pata_cmd64x.c       |  505 ++++++++++++++++
 drivers/ata/pata_cs5520.c       |  336 +++++++++++
 drivers/ata/pata_cs5530.c       |  387 ++++++++++++
 drivers/ata/pata_cs5535.c       |  291 +++++++++
 drivers/ata/pata_cypress.c      |  227 +++++++
 drivers/ata/pata_efar.c         |  342 +++++++++++
 drivers/ata/pata_hpt366.c       |  478 +++++++++++++++
 drivers/ata/pata_hpt37x.c       | 1257 +++++++++++++++++++++++++++++++++++++++
 drivers/ata/pata_hpt3x2n.c      |  597 +++++++++++++++++++
 drivers/ata/pata_hpt3x3.c       |  226 +++++++
 drivers/ata/pata_isapnp.c       |  156 +++++
 drivers/ata/pata_it8172.c       |  288 +++++++++
 drivers/ata/pata_it821x.c       |  847 ++++++++++++++++++++++++++
 drivers/ata/pata_jmicron.c      |  266 +++++++++
 drivers/ata/pata_legacy.c       |  949 +++++++++++++++++++++++++++++
 drivers/ata/pata_mpiix.c        |  313 ++++++++++
 drivers/ata/pata_netcell.c      |  175 ++++++
 drivers/ata/pata_ns87410.c      |  236 ++++++++
 drivers/ata/pata_oldpiix.c      |  339 +++++++++++
 drivers/ata/pata_opti.c         |  292 +++++++++
 drivers/ata/pata_optidma.c      |  547 +++++++++++++++++
 drivers/ata/pata_pcmcia.c       |  393 ++++++++++++
 drivers/ata/pata_pdc2027x.c     |  869 +++++++++++++++++++++++++++
 drivers/ata/pata_pdc202xx_old.c |  423 +++++++++++++
 drivers/ata/pata_qdi.c          |  403 +++++++++++++
 drivers/ata/pata_radisys.c      |  335 +++++++++++
 drivers/ata/pata_rz1000.c       |  205 +++++++
 drivers/ata/pata_sc1200.c       |  287 +++++++++
 drivers/ata/pata_serverworks.c  |  587 ++++++++++++++++++
 drivers/ata/pata_sil680.c       |  381 ++++++++++++
 drivers/ata/pata_sis.c          | 1030 ++++++++++++++++++++++++++++++++
 drivers/ata/pata_sl82c105.c     |  388 ++++++++++++
 drivers/ata/pata_triflex.c      |  285 +++++++++
 drivers/ata/pata_via.c          |  568 ++++++++++++++++++
 include/linux/libata.h          |    2 +-
 43 files changed, 18372 insertions(+), 111 deletions(-)
 create mode 100644 drivers/ata/ata_generic.c
 create mode 100644 drivers/ata/pata_ali.c
 create mode 100644 drivers/ata/pata_amd.c
 create mode 100644 drivers/ata/pata_artop.c
 create mode 100644 drivers/ata/pata_atiixp.c
 create mode 100644 drivers/ata/pata_cmd64x.c
 create mode 100644 drivers/ata/pata_cs5520.c
 create mode 100644 drivers/ata/pata_cs5530.c
 create mode 100644 drivers/ata/pata_cs5535.c
 create mode 100644 drivers/ata/pata_cypress.c
 create mode 100644 drivers/ata/pata_efar.c
 create mode 100644 drivers/ata/pata_hpt366.c
 create mode 100644 drivers/ata/pata_hpt37x.c
 create mode 100644 drivers/ata/pata_hpt3x2n.c
 create mode 100644 drivers/ata/pata_hpt3x3.c
 create mode 100644 drivers/ata/pata_isapnp.c
 create mode 100644 drivers/ata/pata_it8172.c
 create mode 100644 drivers/ata/pata_it821x.c
 create mode 100644 drivers/ata/pata_jmicron.c
 create mode 100644 drivers/ata/pata_legacy.c
 create mode 100644 drivers/ata/pata_mpiix.c
 create mode 100644 drivers/ata/pata_netcell.c
 create mode 100644 drivers/ata/pata_ns87410.c
 create mode 100644 drivers/ata/pata_oldpiix.c
 create mode 100644 drivers/ata/pata_opti.c
 create mode 100644 drivers/ata/pata_optidma.c
 create mode 100644 drivers/ata/pata_pcmcia.c
 create mode 100644 drivers/ata/pata_pdc2027x.c
 create mode 100644 drivers/ata/pata_pdc202xx_old.c
 create mode 100644 drivers/ata/pata_qdi.c
 create mode 100644 drivers/ata/pata_radisys.c
 create mode 100644 drivers/ata/pata_rz1000.c
 create mode 100644 drivers/ata/pata_sc1200.c
 create mode 100644 drivers/ata/pata_serverworks.c
 create mode 100644 drivers/ata/pata_sil680.c
 create mode 100644 drivers/ata/pata_sis.c
 create mode 100644 drivers/ata/pata_sl82c105.c
 create mode 100644 drivers/ata/pata_triflex.c
 create mode 100644 drivers/ata/pata_via.c

(limited to 'include')

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index 13027d56b7f6..cbda988692e7 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -145,6 +145,340 @@ config SATA_INTEL_COMBINED
 	depends on IDE=y && !BLK_DEV_IDE_SATA && (SATA_AHCI || ATA_PIIX)
 	default y
 
+config PATA_ALI
+	tristate "ALi PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the ALi ATA interfaces
+	  found on the many ALi chipsets.
+
+	  If unsure, say N.
+
+config PATA_AMD
+	tristate "AMD/NVidia PATA support (Experimental)"
+	depends on PCI
+	help
+	  This option enables support for the AMD and NVidia PATA
+	  interfaces found on the chipsets for Athlon/Athlon64.
+
+	  If unsure, say N.
+
+config PATA_ARTOP
+	tristate "ARTOP 6210/6260 PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for ARTOP PATA controllers.
+
+	  If unsure, say N.
+
+config PATA_ATIIXP
+	tristate "ATI PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the ATI ATA interfaces
+	  found on the many ATI chipsets.
+
+	  If unsure, say N.
+
+config PATA_CMD64X
+	tristate "CMD64x PATA support (Very Experimental)"
+	depends on PCI&& EXPERIMENTAL
+	help
+	  This option enables support for the CMD64x series chips
+	  except for the CMD640.
+
+	  If unsure, say N.
+
+config PATA_CS5520
+	tristate "CS5510/5520 PATA support"
+	depends on PCI
+	help
+	  This option enables support for the Cyrix 5510/5520
+	  companion chip used with the MediaGX/Geode processor family.
+
+	  If unsure, say N.
+
+config PATA_CS5530
+	tristate "CS5530 PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the Cyrix/NatSemi/AMD CS5530
+	  companion chip used with the MediaGX/Geode processor family.
+
+	  If unsure, say N.
+
+config PATA_CS5535
+	tristate "CS5535 PATA support (Experimental)"
+	depends on PCI && X86 && !X86_64 && EXPERIMENTAL
+	help
+	  This option enables support for the NatSemi/AMD CS5535
+	  companion chip used with the Geode processor family.
+
+	  If unsure, say N.
+
+config PATA_CYPRESS
+	tristate "Cypress CY82C693 PATA support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the Cypress/Contaq CY82C693
+	  chipset found in some Alpha systems
+
+	  If unsure, say N.
+
+config PATA_EFAR
+	tristate "EFAR SLC90E66 support"
+	depends on PCI
+	help
+	  This option enables support for the EFAR SLC90E66
+	  IDE controller found on some older machines.
+
+	  If unsure, say N.
+
+config ATA_GENERIC
+	tristate "Generic ATA support"
+	depends on PCI
+	help
+	  This option enables support for generic BIOS configured
+	  ATA controllers via the new ATA layer
+
+	  If unsure, say N.
+
+config PATA_HPT366
+	tristate "HPT 366/368 PATA support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the HPT 366 and 368
+	  PATA controllers via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_HPT37X
+	tristate "HPT 370/370A/371/372/374/302 PATA support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the majority of the later HPT
+	  PATA controllers via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_HPT3X2N
+	tristate "HPT 372N/302N PATA support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the N variant HPT PATA
+	  controllers via the new ATA layer
+
+	  If unsure, say N.
+
+config PATA_HPT3X3
+	tristate "HPT 343/363 PATA support (Experimental)"
+	depends on PCI
+	help
+	  This option enables support for the HPT 343/363
+	  PATA controllers via the new ATA layer
+
+	  If unsure, say N.
+
+config PATA_ISAPNP
+	tristate "ISA Plug and Play PATA support (Very Experimental)"
+	depends on EXPERIMENTAL && ISAPNP
+	help
+	  This option enables support for ISA plug & play ATA
+	  controllers such as those found on old soundcards.
+
+	  If unsure, say N.
+
+config PATA_IT8172
+	tristate "IT8172 PATA support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the ITE 8172 PATA controller
+	  via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_IT821X
+	tristate "IT821x PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the ITE 8211 and 8212
+	  PATA controllers via the new ATA layer, including RAID
+	  mode.
+
+	  If unsure, say N.
+
+config PATA_LEGACY
+	tristate "Legacy ISA PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for ISA/VLB bus legacy PATA
+	  ports and allows them to be accessed via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_TRIFLEX
+	tristate "Compaq Triflex PATA support"
+	depends on PCI
+	help
+	  Enable support for the Compaq 'Triflex' IDE controller as found
+	  on many Compaq Pentium-Pro systems, via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_MPIIX
+	tristate "Intel PATA MPIIX support"
+	depends on PCI
+	help
+	  This option enables support for MPIIX PATA support.
+
+	  If unsure, say N.
+
+config PATA_OLDPIIX
+	tristate "Intel PATA old PIIX support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for old(?) PIIX PATA support.
+
+	  If unsure, say N.
+
+config PATA_NETCELL
+	tristate "NETCELL Revolution RAID support"
+	depends on PCI
+	help
+	  This option enables support for the Netcell Revolution RAID
+	  PATA controller.
+
+	  If unsure, say N.
+
+config PATA_NS87410
+	tristate "Nat Semi NS87410 PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the National Semiconductor
+	  NS87410 PCI-IDE controller.
+
+	  If unsure, say N.
+
+config PATA_OPTI
+	tristate "OPTI621/6215 PATA support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables full PIO support for the early Opti ATA
+	  controllers found on some old motherboards.
+
+	  If unsure, say N.
+
+config PATA_OPTIDMA
+	tristate "OPTI FireStar PATA support (Veyr Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables DMA/PIO support for the later OPTi
+	  controllers found on some old motherboards and in some
+	  latops
+
+	  If unsure, say N.
+
+config PATA_PCMCIA
+	tristate "PCMCIA PATA support"
+	depends on PCMCIA
+	help
+	  This option enables support for PCMCIA ATA interfaces, including
+	  compact flash card adapters via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_PDC_OLD
+	tristate "Older Promise PATA controller support (Very Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the Promise 20246, 20262, 20263,
+	  20265 and 20267 adapters.
+
+	  If unsure, say N.
+
+config PATA_QDI
+	tristate "QDI VLB PATA support"
+	help
+	  Support for QDI 6500 and 6580 PATA controllers on VESA local bus.
+
+config PATA_RADISYS
+	tristate "RADISYS 82600 PATA support (Very experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the RADISYS 82600
+	  PATA controllers via the new ATA layer
+
+	  If unsure, say N.
+
+config PATA_RZ1000
+	tristate "PC Tech RZ1000 PATA support"
+	depends on PCI
+	help
+	  This option enables basic support for the PC Tech RZ1000/1
+	  PATA controllers via the new ATA layer
+
+	  If unsure, say N.
+
+config PATA_SC1200
+	tristate "SC1200 PATA support (Raving Lunatic)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the NatSemi/AMD SC1200 SoC
+	  companion chip used with the Geode processor family.
+
+	  If unsure, say N.
+
+config PATA_SERVERWORKS
+	tristate "SERVERWORKS OSB4/CSB5/CSB6/HT1000 PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for the Serverworks OSB4/CSB5/CSB6 and
+	  HT1000 PATA controllers, via the new ATA layer.
+
+	  If unsure, say N.
+
+config PATA_PDC2027X
+	tristate "Promise PATA 2027x support"
+	depends on PCI
+	help
+	  This option enables support for Promise PATA pdc20268 to pdc20277 host adapters.
+
+	  If unsure, say N.
+
+config PATA_SIL680
+	tristate "CMD / Silicon Image 680 PATA support"
+	depends on PCI
+	help
+	  This option enables support for CMD / Silicon Image 680 PATA.
+
+	  If unsure, say N.
+
+config PATA_SIS
+	tristate "SiS PATA support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  This option enables support for SiS PATA controllers
+
+	  If unsure, say N.
+
+config PATA_VIA
+	tristate "VIA PATA support"
+	depends on PCI
+	help
+	  This option enables support for the VIA PATA interfaces
+	  found on the many VIA chipsets.
+
+	  If unsure, say N.
+
+config PATA_WINBOND
+	tristate "Winbond SL82C105 PATA support"
+	depends on PCI
+	help
+	  This option enables support for SL82C105 PATA devices found in the
+	  Netwinder and some other systems
+
+	  If unsure, say N.
+
 endif
 endmenu
 
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index e260e3fe65c8..b183a04aeaf7 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -17,5 +17,46 @@ obj-$(CONFIG_SATA_ULI)		+= sata_uli.o
 obj-$(CONFIG_SATA_MV)		+= sata_mv.o
 obj-$(CONFIG_PDC_ADMA)		+= pdc_adma.o
 
+obj-$(CONFIG_PATA_ALI)		+= pata_ali.o
+obj-$(CONFIG_PATA_AMD)		+= pata_amd.o
+obj-$(CONFIG_PATA_ARTOP)	+= pata_artop.o
+obj-$(CONFIG_PATA_ATIIXP)	+= pata_atiixp.o
+obj-$(CONFIG_PATA_CMD64X)	+= pata_cmd64x.o
+obj-$(CONFIG_PATA_CS5520)	+= pata_cs5520.o
+obj-$(CONFIG_PATA_CS5530)	+= pata_cs5530.o
+obj-$(CONFIG_PATA_CS5535)	+= pata_cs5535.o
+obj-$(CONFIG_PATA_CYPRESS)	+= pata_cypress.o
+obj-$(CONFIG_PATA_EFAR)		+= pata_efar.o
+obj-$(CONFIG_PATA_HPT366)	+= pata_hpt366.o
+obj-$(CONFIG_PATA_HPT37X)	+= pata_hpt37x.o
+obj-$(CONFIG_PATA_HPT3X2N)	+= pata_hpt3x2n.o
+obj-$(CONFIG_PATA_HPT3X3)	+= pata_hpt3x3.o
+obj-$(CONFIG_PATA_ISAPNP)	+= pata_isapnp.o
+obj-$(CONFIG_PATA_IT8172)	+= pata_it8172.o
+obj-$(CONFIG_PATA_IT821X)	+= pata_it821x.o
+obj-$(CONFIG_PATA_NETCELL)	+= pata_netcell.o
+obj-$(CONFIG_PATA_NS87410)	+= pata_ns87410.o
+obj-$(CONFIG_PATA_OPTI)		+= pata_opti.o
+obj-$(CONFIG_PATA_OPTIDMA)	+= pata_optidma.o
+obj-$(CONFIG_PATA_MPIIX)	+= pata_mpiix.o
+obj-$(CONFIG_PATA_OLDPIIX)	+= pata_oldpiix.o
+obj-$(CONFIG_PATA_PCMCIA)	+= pata_pcmcia.o
+obj-$(CONFIG_PATA_PDC2027X)	+= pata_pdc2027x.o
+obj-$(CONFIG_PATA_PDC_OLD)	+= pata_pdc202xx_old.o
+obj-$(CONFIG_PATA_QDI)		+= pata_qdi.o
+obj-$(CONFIG_PATA_RADISYS)	+= pata_radisys.o
+obj-$(CONFIG_PATA_RZ1000)	+= pata_rz1000.o
+obj-$(CONFIG_PATA_SC1200)	+= pata_sc1200.o
+obj-$(CONFIG_PATA_SERVERWORKS)	+= pata_serverworks.o
+obj-$(CONFIG_PATA_SIL680)	+= pata_sil680.o
+obj-$(CONFIG_PATA_VIA)		+= pata_via.o
+obj-$(CONFIG_PATA_WINBOND)	+= pata_sl82c105.o
+obj-$(CONFIG_PATA_SIS)		+= pata_sis.o
+obj-$(CONFIG_PATA_TRIFLEX)	+= pata_triflex.o
+# Should be last but one libata driver
+obj-$(CONFIG_ATA_GENERIC)	+= ata_generic.o
+# Should be last libata driver
+obj-$(CONFIG_PATA_LEGACY)	+= pata_legacy.o
+
 libata-objs	:= libata-core.o libata-scsi.o libata-sff.o libata-eh.o
 
diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
new file mode 100644
index 000000000000..20b191dcf7ed
--- /dev/null
+++ b/drivers/ata/ata_generic.c
@@ -0,0 +1,252 @@
+/*
+ *  ata_generic.c - Generic PATA/SATA controller driver.
+ *  Copyright 2005 Red Hat Inc <alan@redhat.com>, all rights reserved.
+ *
+ *  Elements from ide/pci/generic.c 
+ *	    Copyright (C) 2001-2002	Andre Hedrick <andre@linux-ide.org>
+ *	    Portions (C) Copyright 2002  Red Hat Inc <alan@redhat.com>
+ *
+ *  May be copied or modified under the terms of the GNU General Public License
+ *  
+ *  Driver for PCI IDE interfaces implementing the standard bus mastering
+ *  interface functionality. This assumes the BIOS did the drive set up and
+ *  tuning for us. By default we do not grab all IDE class devices as they
+ *  may have other drivers or need fixups to avoid problems. Instead we keep
+ *  a default list of stuff without documentation/driver that appears to
+ *  work.  
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "ata_generic"
+#define DRV_VERSION "0.2.6"
+
+/*
+ *	A generic parallel ATA driver using libata
+ */
+
+/**
+ *	generic_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+ 
+static int generic_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+
+/**
+ *	generic_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *	@classes:
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+ 
+static void generic_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, generic_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	generic_set_mode	-	mode setting
+ *	@ap: interface to set up
+ *
+ *	Use a non standard set_mode function. We don't want to be tuned.
+ *	The BIOS configured everything. Our job is not to fiddle. We
+ *	read the dma enabled bits from the PCI configuration of the device
+ *	and respect them. 
+ */
+ 
+static void generic_set_mode(struct ata_port *ap)
+{
+	int dma_enabled = 0;
+	int i;
+
+	/* Bits 5 and 6 indicate if DMA is active on master/slave */
+	if (ap->ioaddr.bmdma_addr)
+		dma_enabled = inb(ap->ioaddr.bmdma_addr + ATA_DMA_CMD);
+	
+	for (i = 0; i < ATA_MAX_DEVICES; i++) {
+		struct ata_device *dev = &ap->device[i];
+		if (ata_dev_enabled(dev)) {
+			/* We don't really care */
+			dev->pio_mode = XFER_PIO_0;
+			dev->dma_mode = XFER_MW_DMA_0;
+			/* We do need the right mode information for DMA or PIO 
+			   and this comes from the current configuration flags */
+			if (dma_enabled & (1 << (5 + i))) {
+				dev->xfer_mode = XFER_MW_DMA_0;
+				dev->xfer_shift = ATA_SHIFT_MWDMA;
+				dev->flags &= ~ATA_DFLAG_PIO;
+			} else {
+				dev->xfer_mode = XFER_PIO_0;
+				dev->xfer_shift = ATA_SHIFT_PIO;
+				dev->flags |= ATA_DFLAG_PIO;
+			}
+		}
+	}
+}
+
+static struct scsi_host_template generic_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations generic_port_ops = {
+	.set_mode	= generic_set_mode,
+	
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+	
+	.data_xfer	= ata_pio_data_xfer,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= generic_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+	
+static int all_generic_ide;		/* Set to claim all devices */
+
+/**
+ *	ata_generic_init		-	attach generic IDE
+ *	@dev: PCI device found
+ *	@id: match entry
+ *
+ *	Called each time a matching IDE interface is found. We check if the
+ *	interface is one we wish to claim and if so we perform any chip 
+ *	specific hacks then let the ATA layer do the heavy lifting.
+ */
+ 
+static int ata_generic_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	u16 command;
+	static struct ata_port_info info = {
+		.sht = &generic_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &generic_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+		
+	/* Don't use the generic entry unless instructed to do so */
+	if (id->driver_data == 1 && all_generic_ide == 0)
+		return -ENODEV;
+
+	/* Devices that need care */
+	if (dev->vendor == PCI_VENDOR_ID_UMC &&
+	    dev->device == PCI_DEVICE_ID_UMC_UM8886A &&
+	    (!(PCI_FUNC(dev->devfn) & 1)))
+		return -ENODEV;
+
+	if (dev->vendor == PCI_VENDOR_ID_OPTI &&
+	    dev->device == PCI_DEVICE_ID_OPTI_82C558 &&
+	    (!(PCI_FUNC(dev->devfn) & 1)))
+		return -ENODEV;
+
+	/* Don't re-enable devices in generic mode or we will break some
+	   motherboards with disabled and unused IDE controllers */
+	pci_read_config_word(dev, PCI_COMMAND, &command);
+	if (!(command & PCI_COMMAND_IO))
+		return -ENODEV;
+		
+	if (dev->vendor == PCI_VENDOR_ID_AL)
+	    	ata_pci_clear_simplex(dev);
+
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id ata_generic[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_PCTECH, PCI_DEVICE_ID_PCTECH_SAMURAI_IDE), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_HOLTEK, PCI_DEVICE_ID_HOLTEK_6565), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_UMC,    PCI_DEVICE_ID_UMC_UM8673F), }, 
+	{ PCI_DEVICE(PCI_VENDOR_ID_UMC,    PCI_DEVICE_ID_UMC_UM8886A), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_UMC,    PCI_DEVICE_ID_UMC_UM8886BF), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_HINT,   PCI_DEVICE_ID_HINT_VXPROII_IDE), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA,    PCI_DEVICE_ID_VIA_82C561), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_OPTI,   PCI_DEVICE_ID_OPTI_82C558), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_1), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_2),  },
+	/* Must come last. If you add entries adjust this table appropriately */
+	{ PCI_ANY_ID,		PCI_ANY_ID,			   PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xFFFFFF00UL, 1},
+	{ 0, },
+};
+
+static struct pci_driver ata_generic_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= ata_generic,
+	.probe 		= ata_generic_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init ata_generic_init(void)
+{
+	return pci_module_init(&ata_generic_pci_driver);
+}
+
+
+static void __exit ata_generic_exit(void)
+{
+	pci_unregister_driver(&ata_generic_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for generic ATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, ata_generic);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(ata_generic_init);
+module_exit(ata_generic_exit);
+
+module_param(all_generic_ide, int, 0);
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 22b2dba90b9a..12b3a42fb356 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -93,7 +93,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME	"ata_piix"
-#define DRV_VERSION	"2.00"
+#define DRV_VERSION	"2.00ac6"
 
 enum {
 	PIIX_IOCFG		= 0x54, /* IDE I/O configuration register */
@@ -116,15 +116,18 @@ enum {
 	PIIX_80C_SEC		= (1 << 7) | (1 << 6),
 
 	/* controller IDs */
-	piix4_pata		= 0,
-	ich5_pata		= 1,
-	ich5_sata		= 2,
-	esb_sata		= 3,
-	ich6_sata		= 4,
-	ich6_sata_ahci		= 5,
-	ich6m_sata_ahci		= 6,
-	ich8_sata_ahci		= 7,
-
+	piix_pata_33		= 0,	/* PIIX3 or 4 at 33Mhz */
+	ich_pata_33		= 1,	/* ICH up to UDMA 33 only */
+	ich_pata_66		= 2,	/* ICH up to 66 Mhz */
+	ich_pata_100		= 3,	/* ICH up to UDMA 100 */
+	ich_pata_133		= 4,	/* ICH up to UDMA 133 */
+	ich5_sata		= 5,
+	esb_sata		= 6,
+	ich6_sata		= 7,
+	ich6_sata_ahci		= 8,
+	ich6m_sata_ahci		= 9,
+	ich8_sata_ahci		= 10,
+	
 	/* constants for mapping table */
 	P0			= 0,  /* port 0 */
 	P1			= 1,  /* port 1 */
@@ -152,19 +155,54 @@ struct piix_host_priv {
 static int piix_init_one (struct pci_dev *pdev,
 				    const struct pci_device_id *ent);
 static void piix_host_stop(struct ata_host *host);
-static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev);
-static void piix_set_dmamode (struct ata_port *ap, struct ata_device *adev);
 static void piix_pata_error_handler(struct ata_port *ap);
+static void ich_pata_error_handler(struct ata_port *ap);
 static void piix_sata_error_handler(struct ata_port *ap);
+static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev);
+static void piix_set_dmamode (struct ata_port *ap, struct ata_device *adev);
+static void ich_set_dmamode (struct ata_port *ap, struct ata_device *adev);
 
 static unsigned int in_module_init = 1;
 
 static const struct pci_device_id piix_pci_tbl[] = {
 #ifdef ATA_ENABLE_PATA
-	{ 0x8086, 0x7111, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix4_pata },
-	{ 0x8086, 0x24db, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
-	{ 0x8086, 0x25a2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
-	{ 0x8086, 0x27df, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
+	/* Intel PIIX4 for the 430TX/440BX/MX chipset: UDMA 33 */
+	/* Also PIIX4E (fn3 rev 2) and PIIX4M (fn3 rev 3) */
+	{ 0x8086, 0x7111, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 },
+	{ 0x8086, 0x24db, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	{ 0x8086, 0x25a2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* Intel PIIX4 */
+	{ 0x8086, 0x7199, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 },
+	/* Intel PIIX4 */
+	{ 0x8086, 0x7601, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 },
+	/* Intel PIIX */
+	{ 0x8086, 0x84CA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 },
+	/* Intel ICH (i810, i815, i840) UDMA 66*/
+	{ 0x8086, 0x2411, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_66 },
+	/* Intel ICH0 : UDMA 33*/
+	{ 0x8086, 0x2421, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_33 },
+	/* Intel ICH2M */
+	{ 0x8086, 0x244A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* Intel ICH2 (i810E2, i845, 850, 860) UDMA 100 */
+	{ 0x8086, 0x244B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/*  Intel ICH3M */
+	{ 0x8086, 0x248A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* Intel ICH3 (E7500/1) UDMA 100 */
+	{ 0x8086, 0x248B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* Intel ICH4 (i845GV, i845E, i852, i855) UDMA 100 */
+	{ 0x8086, 0x24CA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	{ 0x8086, 0x24CB, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* Intel ICH5 */
+	{ 0x8086, 0x24DB, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_133 },
+	/* C-ICH (i810E2) */
+	{ 0x8086, 0x245B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* ESB (855GME/875P + 6300ESB) UDMA 100  */	
+	{ 0x8086, 0x25A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* ICH6 (and 6) (i915) UDMA 100 */
+	{ 0x8086, 0x266F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
+	/* ICH7/7-R (i945, i975) UDMA 100*/
+	{ 0x8086, 0x27DF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_133 },
+	{ 0x8086, 0x269E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 },
 #endif
 
 	/* NOTE: The following PCI ids must be kept in sync with the
@@ -263,6 +301,39 @@ static const struct ata_port_operations piix_pata_ops = {
 	.host_stop		= piix_host_stop,
 };
 
+static const struct ata_port_operations ich_pata_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= piix_set_piomode,
+	.set_dmamode		= ich_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= ich_pata_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
 static const struct ata_port_operations piix_sata_ops = {
 	.port_disable		= ata_port_disable,
 
@@ -359,35 +430,56 @@ static const struct piix_map_db *piix_map_db_table[] = {
 };
 
 static struct ata_port_info piix_port_info[] = {
-	/* piix4_pata */
+	/* piix_pata_33: 0:  PIIX3 or 4 at 33MHz */
 	{
 		.sht		= &piix_sht,
-		.flags		= ATA_FLAG_SLAVE_POSS,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
 		.pio_mask	= 0x1f,	/* pio0-4 */
-#if 0
-		.mwdma_mask	= 0x06, /* mwdma1-2 */
-#else
-		.mwdma_mask	= 0x00, /* mwdma broken */
-#endif
+		.mwdma_mask	= 0x06, /* mwdma1-2 ?? CHECK 0 should be ok but slow */
 		.udma_mask	= ATA_UDMA_MASK_40C,
 		.port_ops	= &piix_pata_ops,
 	},
 
-	/* ich5_pata */
+	/* ich_pata_33: 1 	ICH0 - ICH at 33Mhz*/
+	{
+		.sht		= &piix_sht,
+		.flags		= ATA_FLAG_SRST | ATA_FLAG_SLAVE_POSS,
+		.pio_mask 	= 0x1f,	/* pio 0-4 */
+		.mwdma_mask	= 0x06, /* Check: maybe 0x07  */
+		.udma_mask	= ATA_UDMA2, /* UDMA33 */
+		.port_ops	= &ich_pata_ops,
+	},
+	/* ich_pata_66: 2 	ICH controllers up to 66MHz */
 	{
 		.sht		= &piix_sht,
-		.flags		= ATA_FLAG_SLAVE_POSS | PIIX_FLAG_CHECKINTR,
+		.flags		= ATA_FLAG_SRST | ATA_FLAG_SLAVE_POSS,
+		.pio_mask 	= 0x1f,	/* pio 0-4 */
+		.mwdma_mask	= 0x06, /* MWDMA0 is broken on chip */
+		.udma_mask	= ATA_UDMA4,
+		.port_ops	= &ich_pata_ops,
+	},
+	
+	/* ich_pata_100: 3 */
+	{
+		.sht		= &piix_sht,
+		.flags		= ATA_FLAG_SRST | ATA_FLAG_SLAVE_POSS | PIIX_FLAG_CHECKINTR,
 		.pio_mask	= 0x1f,	/* pio0-4 */
-#if 0
 		.mwdma_mask	= 0x06, /* mwdma1-2 */
-#else
-		.mwdma_mask	= 0x00, /* mwdma broken */
-#endif
-		.udma_mask	= 0x3f, /* udma0-5 */
-		.port_ops	= &piix_pata_ops,
+		.udma_mask	= ATA_UDMA5, /* udma0-5 */
+		.port_ops	= &ich_pata_ops,
 	},
 
-	/* ich5_sata */
+	/* ich_pata_133: 4 	ICH with full UDMA6 */
+	{
+		.sht		= &piix_sht,
+		.flags		= ATA_FLAG_SRST | ATA_FLAG_SLAVE_POSS | PIIX_FLAG_CHECKINTR,
+		.pio_mask 	= 0x1f,	/* pio 0-4 */
+		.mwdma_mask	= 0x06, /* Check: maybe 0x07  */
+		.udma_mask	= ATA_UDMA6, /* UDMA133 */
+		.port_ops	= &ich_pata_ops,
+	},
+
+	/* ich5_sata: 5 */
 	{
 		.sht		= &piix_sht,
 		.flags		= ATA_FLAG_SATA | PIIX_FLAG_CHECKINTR |
@@ -398,7 +490,7 @@ static struct ata_port_info piix_port_info[] = {
 		.port_ops	= &piix_sata_ops,
 	},
 
-	/* i6300esb_sata */
+	/* i6300esb_sata: 6 */
 	{
 		.sht		= &piix_sht,
 		.flags		= ATA_FLAG_SATA |
@@ -409,7 +501,7 @@ static struct ata_port_info piix_port_info[] = {
 		.port_ops	= &piix_sata_ops,
 	},
 
-	/* ich6_sata */
+	/* ich6_sata: 7 */
 	{
 		.sht		= &piix_sht,
 		.flags		= ATA_FLAG_SATA |
@@ -420,7 +512,7 @@ static struct ata_port_info piix_port_info[] = {
 		.port_ops	= &piix_sata_ops,
 	},
 
-	/* ich6_sata_ahci */
+	/* ich6_sata_ahci:8 */
 	{
 		.sht		= &piix_sht,
 		.flags		= ATA_FLAG_SATA |
@@ -432,7 +524,7 @@ static struct ata_port_info piix_port_info[] = {
 		.port_ops	= &piix_sata_ops,
 	},
 
-	/* ich6m_sata_ahci */
+	/* ich6m_sata_ahci: 9 */
 	{
 		.sht		= &piix_sht,
 		.flags		= ATA_FLAG_SATA |
@@ -455,6 +547,7 @@ static struct ata_port_info piix_port_info[] = {
 		.udma_mask	= 0x7f,	/* udma0-6 */
 		.port_ops	= &piix_sata_ops,
 	},
+
 };
 
 static struct pci_bits piix_enable_bits[] = {
@@ -483,7 +576,8 @@ MODULE_PARM_DESC(force_pcs, "force honoring or ignoring PCS to work around "
  *	LOCKING:
  *	None (inherited from caller).
  */
-static void piix_pata_cbl_detect(struct ata_port *ap)
+
+static void ich_pata_cbl_detect(struct ata_port *ap)
 {
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 tmp, mask;
@@ -503,14 +597,12 @@ static void piix_pata_cbl_detect(struct ata_port *ap)
 
 cbl40:
 	ap->cbl = ATA_CBL_PATA40;
-	ap->udma_mask &= ATA_UDMA_MASK_40C;
 }
 
 /**
  *	piix_pata_prereset - prereset for PATA host controller
  *	@ap: Target port
  *
- *	Prereset including cable detection.
  *
  *	LOCKING:
  *	None (inherited from caller).
@@ -524,9 +616,7 @@ static int piix_pata_prereset(struct ata_port *ap)
 		ap->eh_context.i.action &= ~ATA_EH_RESET_MASK;
 		return 0;
 	}
-
-	piix_pata_cbl_detect(ap);
-
+	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
 
@@ -536,6 +626,36 @@ static void piix_pata_error_handler(struct ata_port *ap)
 			   ata_std_postreset);
 }
 
+
+/**
+ *	ich_pata_prereset - prereset for PATA host controller
+ *	@ap: Target port
+ *
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+static int ich_pata_prereset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+
+	if (!pci_test_config_bits(pdev, &piix_enable_bits[ap->port_no])) {
+		ata_port_printk(ap, KERN_INFO, "port disabled. ignoring.\n");
+		ap->eh_context.i.action &= ~ATA_EH_RESET_MASK;
+		return 0;
+	}
+
+	ich_pata_cbl_detect(ap);
+
+	return ata_std_prereset(ap);
+}
+
+static void ich_pata_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, ich_pata_prereset, ata_std_softreset, NULL,
+			   ata_std_postreset);
+}
+
 /**
  *	piix_sata_present_mask - determine present mask for SATA host controller
  *	@ap: Target port
@@ -637,6 +757,13 @@ static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
 	unsigned int slave_port	= 0x44;
 	u16 master_data;
 	u8 slave_data;
+	u8 udma_enable;
+	int control = 0;
+	
+	/*
+	 *	See Intel Document 298600-004 for the timing programing rules
+	 *	for ICH controllers.
+	 */
 
 	static const	 /* ISP  RTC */
 	u8 timings[][2]	= { { 0, 0 },
@@ -645,20 +772,30 @@ static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
 			    { 2, 1 },
 			    { 2, 3 }, };
 
+	if (pio >= 2)
+		control |= 1;	/* TIME1 enable */
+	if (ata_pio_need_iordy(adev))
+		control |= 2;	/* IE enable */
+
+	/* Intel specifies that the PPE functionality is for disk only */	   
+	if (adev->class == ATA_DEV_ATA)
+		control |= 4;	/* PPE enable */
+
 	pci_read_config_word(dev, master_port, &master_data);
 	if (is_slave) {
+		/* Enable SITRE (seperate slave timing register) */
 		master_data |= 0x4000;
-		/* enable PPE, IE and TIME */
-		master_data |= 0x0070;
+		/* enable PPE1, IE1 and TIME1 as needed */
+		master_data |= (control << 4);
 		pci_read_config_byte(dev, slave_port, &slave_data);
 		slave_data &= (ap->port_no ? 0x0f : 0xf0);
-		slave_data |=
-			(timings[pio][0] << 2) |
-			(timings[pio][1] << (ap->port_no ? 4 : 0));
+		/* Load the timing nibble for this slave */
+		slave_data |= ((timings[pio][0] << 2) | timings[pio][1]) << (ap->port_no ? 4 : 0);
 	} else {
+		/* Master keeps the bits in a different format */
 		master_data &= 0xccf8;
-		/* enable PPE, IE and TIME */
-		master_data |= 0x0007;
+		/* Enable PPE, IE and TIME as appropriate */
+		master_data |= control;
 		master_data |=
 			(timings[pio][0] << 12) |
 			(timings[pio][1] << 8);
@@ -666,13 +803,23 @@ static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
 	pci_write_config_word(dev, master_port, master_data);
 	if (is_slave)
 		pci_write_config_byte(dev, slave_port, slave_data);
+
+	/* Ensure the UDMA bit is off - it will be turned back on if
+	   UDMA is selected */
+	   
+	if (ap->udma_mask) {
+		pci_read_config_byte(dev, 0x48, &udma_enable);
+		udma_enable &= ~(1 << (2 * ap->port_no + adev->devno));
+		pci_write_config_byte(dev, 0x48, udma_enable);
+	}
 }
 
 /**
- *	piix_set_dmamode - Initialize host controller PATA PIO timings
+ *	do_pata_set_dmamode - Initialize host controller PATA PIO timings
  *	@ap: Port whose timings we are configuring
- *	@adev: um
+ *	@adev: Drive in question
  *	@udma: udma mode, 0 - 6
+ *	@is_ich: set if the chip is an ICH device
  *
  *	Set UDMA mode for device, in host controller PCI config space.
  *
@@ -680,70 +827,140 @@ static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
  *	None (inherited from caller).
  */
 
-static void piix_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+static void do_pata_set_dmamode (struct ata_port *ap, struct ata_device *adev, int isich)
 {
-	unsigned int udma	= adev->dma_mode; /* FIXME: MWDMA too */
 	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
-	u8 maslave		= ap->port_no ? 0x42 : 0x40;
-	u8 speed		= udma;
-	unsigned int drive_dn	= (ap->port_no ? 2 : 0) + adev->devno;
-	int a_speed		= 3 << (drive_dn * 4);
-	int u_flag		= 1 << drive_dn;
-	int v_flag		= 0x01 << drive_dn;
-	int w_flag		= 0x10 << drive_dn;
-	int u_speed		= 0;
-	int			sitre;
-	u16			reg4042, reg4a;
-	u8			reg48, reg54, reg55;
-
-	pci_read_config_word(dev, maslave, &reg4042);
-	DPRINTK("reg4042 = 0x%04x\n", reg4042);
-	sitre = (reg4042 & 0x4000) ? 1 : 0;
-	pci_read_config_byte(dev, 0x48, &reg48);
-	pci_read_config_word(dev, 0x4a, &reg4a);
-	pci_read_config_byte(dev, 0x54, &reg54);
-	pci_read_config_byte(dev, 0x55, &reg55);
-
-	switch(speed) {
-		case XFER_UDMA_4:
-		case XFER_UDMA_2:	u_speed = 2 << (drive_dn * 4); break;
-		case XFER_UDMA_6:
-		case XFER_UDMA_5:
-		case XFER_UDMA_3:
-		case XFER_UDMA_1:	u_speed = 1 << (drive_dn * 4); break;
-		case XFER_UDMA_0:	u_speed = 0 << (drive_dn * 4); break;
-		case XFER_MW_DMA_2:
-		case XFER_MW_DMA_1:	break;
-		default:
-			BUG();
-			return;
-	}
+	u8 master_port		= ap->port_no ? 0x42 : 0x40;
+	u16 master_data;
+	u8 speed		= adev->dma_mode;
+	int devid		= adev->devno + 2 * ap->port_no;
+	u8 udma_enable;
+	
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 0 },
+			    { 2, 1 },
+			    { 2, 3 }, };
+
+	pci_read_config_word(dev, master_port, &master_data);
+	pci_read_config_byte(dev, 0x48, &udma_enable);
 
 	if (speed >= XFER_UDMA_0) {
-		if (!(reg48 & u_flag))
-			pci_write_config_byte(dev, 0x48, reg48 | u_flag);
-		if (speed == XFER_UDMA_5) {
-			pci_write_config_byte(dev, 0x55, (u8) reg55|w_flag);
-		} else {
-			pci_write_config_byte(dev, 0x55, (u8) reg55 & ~w_flag);
+		unsigned int udma = adev->dma_mode - XFER_UDMA_0;
+		u16 udma_timing;
+		u16 ideconf;
+		int u_clock, u_speed;
+		
+		/*
+	 	 * UDMA is handled by a combination of clock switching and
+		 * selection of dividers 
+		 *  
+		 * Handy rule: Odd modes are UDMATIMx 01, even are 02
+		 *	       except UDMA0 which is 00 
+		 */
+		u_speed = min(2 - (udma & 1), udma);
+		if (udma == 5)
+			u_clock = 0x1000;	/* 100Mhz */
+		else if (udma > 2)
+			u_clock = 1;		/* 66Mhz */
+		else
+			u_clock = 0;		/* 33Mhz */
+			
+		udma_enable |= (1 << devid);
+			
+		/* Load the CT/RP selection */
+		pci_read_config_word(dev, 0x4A, &udma_timing);
+		udma_timing &= ~(3 << (4 * devid));
+		udma_timing |= u_speed << (4 * devid);
+		pci_write_config_word(dev, 0x4A, udma_timing);
+
+		if (isich) {		
+			/* Select a 33/66/100Mhz clock */
+			pci_read_config_word(dev, 0x54, &ideconf);
+			ideconf &= ~(0x1001 << devid);
+			ideconf |= u_clock << devid;
+			/* For ICH or later we should set bit 10 for better
+			   performance (WR_PingPong_En) */
+			pci_write_config_word(dev, 0x54, ideconf);
 		}
-		if ((reg4a & a_speed) != u_speed)
-			pci_write_config_word(dev, 0x4a, (reg4a & ~a_speed) | u_speed);
-		if (speed > XFER_UDMA_2) {
-			if (!(reg54 & v_flag))
-				pci_write_config_byte(dev, 0x54, reg54 | v_flag);
-		} else
-			pci_write_config_byte(dev, 0x54, reg54 & ~v_flag);
 	} else {
-		if (reg48 & u_flag)
-			pci_write_config_byte(dev, 0x48, reg48 & ~u_flag);
-		if (reg4a & a_speed)
-			pci_write_config_word(dev, 0x4a, reg4a & ~a_speed);
-		if (reg54 & v_flag)
-			pci_write_config_byte(dev, 0x54, reg54 & ~v_flag);
-		if (reg55 & w_flag)
-			pci_write_config_byte(dev, 0x55, (u8) reg55 & ~w_flag);
+		/*
+		 * MWDMA is driven by the PIO timings. We must also enable
+		 * IORDY unconditionally along with TIME1. PPE has already
+		 * been set when the PIO timing was set.
+		 */
+		unsigned int mwdma	= adev->dma_mode - XFER_MW_DMA_0;
+		unsigned int control;
+		u8 slave_data;
+		const unsigned int needed_pio[3] = {
+			XFER_PIO_0, XFER_PIO_3, XFER_PIO_4
+		};
+		int pio = needed_pio[mwdma] - XFER_PIO_0;
+		
+		control = 3;	/* IORDY|TIME1 */
+		
+		/* If the drive MWDMA is faster than it can do PIO then
+		   we must force PIO into PIO0 */
+		   
+		if (adev->pio_mode < needed_pio[mwdma])
+			/* Enable DMA timing only */
+			control |= 8;	/* PIO cycles in PIO0 */
+
+		if (adev->devno) {	/* Slave */
+			master_data &= 0xFF4F;  /* Mask out IORDY|TIME1|DMAONLY */
+			master_data |= control << 4;
+			pci_read_config_byte(dev, 0x44, &slave_data);
+			slave_data &= (0x0F + 0xE1 * ap->port_no);
+			/* Load the matching timing */
+			slave_data |= ((timings[pio][0] << 2) | timings[pio][1]) << (ap->port_no ? 4 : 0);
+			pci_write_config_byte(dev, 0x44, slave_data);
+		} else { 	/* Master */
+			master_data &= 0xCCF4;	/* Mask out IORDY|TIME1|DMAONLY 
+						   and master timing bits */
+			master_data |= control;
+			master_data |=
+				(timings[pio][0] << 12) |
+				(timings[pio][1] << 8);
+		}
+		udma_enable &= ~(1 << devid);
+		pci_write_config_word(dev, master_port, master_data);
 	}
+	/* Don't scribble on 0x48 if the controller does not support UDMA */
+	if (ap->udma_mask)
+		pci_write_config_byte(dev, 0x48, udma_enable);
+}
+
+/**
+ *	piix_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: um
+ *
+ *	Set MW/UDMA mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void piix_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	do_pata_set_dmamode(ap, adev, 0);
+}
+
+/**
+ *	ich_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: um
+ *
+ *	Set MW/UDMA mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void ich_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	do_pata_set_dmamode(ap, adev, 1);
 }
 
 #define AHCI_PCI_BAR 5
@@ -872,7 +1089,7 @@ static void __devinit piix_init_sata_map(struct pci_dev *pdev,
 
 		case IDE:
 			WARN_ON((i & 1) || map[i + 1] != IDE);
-			pinfo[i / 2] = piix_port_info[ich5_pata];
+			pinfo[i / 2] = piix_port_info[ich_pata_100];
 			pinfo[i / 2].private_data = hpriv;
 			i++;
 			printk(" IDE IDE");
@@ -1007,4 +1224,3 @@ static void __exit piix_exit(void)
 
 module_init(piix_init);
 module_exit(piix_exit);
-
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
new file mode 100644
index 000000000000..0e0415ecfce7
--- /dev/null
+++ b/drivers/ata/pata_ali.c
@@ -0,0 +1,679 @@
+/*
+ * pata_ali.c 	- ALI 15x3 PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based in part upon
+ * linux/drivers/ide/pci/alim15x3.c		Version 0.17	2003/01/02
+ *
+ *  Copyright (C) 1998-2000 Michel Aubry, Maintainer
+ *  Copyright (C) 1998-2000 Andrzej Krzysztofowicz, Maintainer
+ *  Copyright (C) 1999-2000 CJ, cjtsai@ali.com.tw, Maintainer
+ *
+ *  Copyright (C) 1998-2000 Andre Hedrick (andre@linux-ide.org)
+ *  May be copied or modified under the terms of the GNU General Public License
+ *  Copyright (C) 2002 Alan Cox <alan@redhat.com>
+ *  ALi (now ULi M5228) support by Clear Zhang <Clear.Zhang@ali.com.tw>
+ *
+ *  Documentation
+ *	Chipset documentation available under NDA only
+ *
+ *  TODO/CHECK
+ *	Cannot have ATAPI on both master & slave for rev < c2 (???) but
+ *	otherwise should do atapi DMA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/dmi.h>
+
+#define DRV_NAME "pata_ali"
+#define DRV_VERSION "0.6.5"
+
+/*
+ *	Cable special cases
+ */
+
+static struct dmi_system_id cable_dmi_table[] = {
+	{
+		.ident = "HP Pavilion N5430",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_BOARD_NAME, "OmniBook N32N-736"),
+		},
+	},
+	{ }
+};
+
+static int ali_cable_override(struct pci_dev *pdev)
+{
+	/* Fujitsu P2000 */
+	if (pdev->subsystem_vendor == 0x10CF && pdev->subsystem_device == 0x10AF)
+	   	return 1;
+	/* Systems by DMI */
+	if (dmi_check_system(cable_dmi_table))
+		return 1;
+	return 0;
+}
+
+/**
+ *	ali_c2_cable_detect	-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection for C2 and later revisions
+ */
+
+static int ali_c2_cable_detect(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 ata66;
+
+	/* Certain laptops use short but suitable cables and don't
+	   implement the detect logic */
+
+	if (ali_cable_override(pdev))
+		return ATA_CBL_PATA80;
+
+	/* Host view cable detect 0x4A bit 0 primary bit 1 secondary
+	   Bit set for 40 pin */
+	pci_read_config_byte(pdev, 0x4A, &ata66);
+	if (ata66 & (1 << ap->port_no))
+		return ATA_CBL_PATA40;
+	else
+		return ATA_CBL_PATA80;
+}
+
+/**
+ *	ali_early_error_handler	-	reset for eary chip
+ *	@ap: ATA port
+ *
+ *	Handle the reset callback for the later chips with cable detect
+ */
+
+static int ali_c2_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ali_c2_cable_detect(ap);
+	return ata_std_prereset(ap);
+}
+
+static void ali_c2_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, ali_c2_pre_reset,
+			       ata_std_softreset, NULL,
+			       ata_std_postreset);
+}
+
+/**
+ *	ali_early_cable_detect	-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection for older chipsets. This turns out to be
+ *	rather easy to implement
+ */
+
+static int ali_early_cable_detect(struct ata_port *ap)
+{
+	return ATA_CBL_PATA40;
+}
+
+/**
+ *	ali_early_probe_init	-	reset for early chip
+ *	@ap: ATA port
+ *
+ *	Handle the reset callback for the early (pre cable detect) chips.
+ */
+
+static int ali_early_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ali_early_cable_detect(ap);
+	return ata_std_prereset(ap);
+}
+
+static void ali_early_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, ali_early_pre_reset,
+				     ata_std_softreset, NULL,
+				     ata_std_postreset);
+}
+
+/**
+ *	ali_20_filter		-	filter for earlier ALI DMA
+ *	@ap: ALi ATA port
+ *	@adev: attached device
+ *
+ *	Ensure that we do not do DMA on CD devices. We may be able to
+ *	fix that later on. Also ensure we do not do UDMA on WDC drives
+ */
+
+static unsigned long ali_20_filter(const struct ata_port *ap, struct ata_device *adev, unsigned long mask)
+{
+	char model_num[40];
+	/* No DMA on anything but a disk for now */
+	if (adev->class != ATA_DEV_ATA)
+		mask &= ~(ATA_MASK_MWDMA | ATA_MASK_UDMA);
+	ata_id_string(adev->id, model_num, ATA_ID_PROD_OFS, sizeof(model_num));
+	if (strstr(model_num, "WDC"))
+		return mask &= ~ATA_MASK_UDMA;
+	return ata_pci_default_filter(ap, adev, mask);
+}
+
+/**
+ *	ali_fifo_control	-	FIFO manager
+ *	@ap: ALi channel to control
+ *	@adev: device for FIFO control
+ *	@on: 0 for off 1 for on
+ *
+ *	Enable or disable the FIFO on a given device. Because of the way the
+ *	ALi FIFO works it provides a boost on ATA disk but can be confused by
+ *	ATAPI and we must therefore manage it.
+ */
+
+static void ali_fifo_control(struct ata_port *ap, struct ata_device *adev, int on)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int pio_fifo = 0x54 + ap->port_no;
+	u8 fifo;
+	int shift = 4 * adev->devno;
+
+	/* ATA - FIFO on set nibble to 0x05, ATAPI - FIFO off, set nibble to
+	   0x00. Not all the docs agree but the behaviour we now use is the
+	   one stated in the BIOS Programming Guide */
+	   
+	pci_read_config_byte(pdev, pio_fifo, &fifo);
+	fifo &= ~(0x0F << shift);
+	if (on)
+		fifo |= (on << shift);
+	pci_write_config_byte(pdev, pio_fifo, fifo);
+}
+
+/**
+ *	ali_program_modes	-	load mode registers
+ *	@ap: ALi channel to load
+ *	@adev: Device the timing is for
+ *	@cmd: Command timing
+ *	@data: Data timing
+ *	@ultra: UDMA timing or zero for off
+ *
+ *	Loads the timing registers for cmd/data and disable UDMA if
+ *	ultra is zero. If ultra is set then load and enable the UDMA
+ *	timing but do not touch the command/data timing.
+ */
+
+static void ali_program_modes(struct ata_port *ap, struct ata_device *adev, struct ata_timing *t, u8 ultra)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int cas = 0x58 + 4 * ap->port_no;	/* Command timing */
+	int cbt = 0x59 + 4 * ap->port_no;	/* Command timing */
+	int drwt = 0x5A + 4 * ap->port_no + adev->devno; /* R/W timing */
+	int udmat = 0x56 + ap->port_no;	/* UDMA timing */
+	int shift = 4 * adev->devno;
+	u8 udma;
+
+	if (t != NULL) {
+		t->setup = FIT(t->setup, 1, 8) & 7;
+		t->act8b = FIT(t->act8b, 1, 8) & 7;
+		t->rec8b = FIT(t->rec8b, 1, 16) & 15;
+		t->active = FIT(t->active, 1, 8) & 7;
+		t->recover = FIT(t->recover, 1, 16) & 15;
+
+		pci_write_config_byte(pdev, cas, t->setup);
+		pci_write_config_byte(pdev, cbt, (t->act8b << 4) | t->rec8b);
+		pci_write_config_byte(pdev, drwt, (t->active << 4) | t->recover);
+	}
+
+	/* Set up the UDMA enable */
+	pci_read_config_byte(pdev, udmat, &udma);
+	udma &= ~(0x0F << shift);
+	udma |= ultra << shift;
+	pci_write_config_byte(pdev, udmat, udma);
+}
+
+/**
+ *	ali_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the ALi registers for PIO mode. FIXME: add timings for
+ *	PIO5.
+ */
+
+static void ali_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct ata_device *pair = ata_dev_pair(adev);
+	struct ata_timing t;
+	unsigned long T =  1000000000 / 33333;	/* PCI clock based */
+
+	ata_timing_compute(adev, adev->pio_mode, &t, T, 1);
+	if (pair) {
+		struct ata_timing p;
+		ata_timing_compute(pair, pair->pio_mode, &p, T, 1);
+		ata_timing_merge(&p, &t, &t, ATA_TIMING_SETUP|ATA_TIMING_8BIT);
+		if (pair->dma_mode) {
+			ata_timing_compute(pair, pair->dma_mode, &p, T, 1);
+			ata_timing_merge(&p, &t, &t, ATA_TIMING_SETUP|ATA_TIMING_8BIT);
+		}
+	}
+
+	/* PIO FIFO is only permitted on ATA disk */
+	if (adev->class != ATA_DEV_ATA)
+		ali_fifo_control(ap, adev, 0x00);
+	ali_program_modes(ap, adev, &t, 0);
+	if (adev->class == ATA_DEV_ATA)
+		ali_fifo_control(ap, adev, 0x05);
+
+}
+
+/**
+ *	ali_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	FIXME: MWDMA timings
+ */
+
+static void ali_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static u8 udma_timing[7] = { 0xC, 0xB, 0xA, 0x9, 0x8, 0xF, 0xD };
+	struct ata_device *pair = ata_dev_pair(adev);
+	struct ata_timing t;
+	unsigned long T =  1000000000 / 33333;	/* PCI clock based */
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+
+
+	if (adev->class == ATA_DEV_ATA)
+		ali_fifo_control(ap, adev, 0x08);
+
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		ali_program_modes(ap, adev, NULL, udma_timing[adev->dma_mode - XFER_UDMA_0]);
+		if (adev->dma_mode >= XFER_UDMA_3) {
+			u8 reg4b;
+			pci_read_config_byte(pdev, 0x4B, &reg4b);
+			reg4b |= 1;
+			pci_write_config_byte(pdev, 0x4B, reg4b);
+		}
+	} else {
+		ata_timing_compute(adev, adev->dma_mode, &t, T, 1);
+		if (pair) {
+			struct ata_timing p;
+			ata_timing_compute(pair, pair->pio_mode, &p, T, 1);
+			ata_timing_merge(&p, &t, &t, ATA_TIMING_SETUP|ATA_TIMING_8BIT);
+			if (pair->dma_mode) {
+				ata_timing_compute(pair, pair->dma_mode, &p, T, 1);
+				ata_timing_merge(&p, &t, &t, ATA_TIMING_SETUP|ATA_TIMING_8BIT);
+			}
+		}
+		ali_program_modes(ap, adev, &t, 0);
+	}
+}
+
+/**
+ *	ali_lock_sectors	-	Keep older devices to 255 sector mode
+ *	@ap: ATA port
+ *	@adev: Device
+ *
+ *	Called during the bus probe for each device that is found. We use
+ *	this call to lock the sector count of the device to 255 or less on
+ *	older ALi controllers. If we didn't do this then large I/O's would
+ *	require LBA48 commands which the older ALi requires are issued by
+ *	slower PIO methods
+ */
+
+static void ali_lock_sectors(struct ata_port *ap, struct ata_device *adev)
+{
+	adev->max_sectors = 255;
+}
+
+static struct scsi_host_template ali_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	/* Keep LBA28 counts so large I/O's don't turn LBA48 and PIO
+	   with older controllers. Not locked so will grow on C5 or later */
+	.max_sectors		= 255,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+/*
+ *	Port operations for PIO only ALi
+ */
+
+static struct ata_port_operations ali_early_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= ali_set_piomode,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ali_early_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Port operations for DMA capable ALi without cable
+ *	detect
+ */
+static struct ata_port_operations ali_20_port_ops = {
+	.port_disable	= ata_port_disable,
+
+	.set_piomode	= ali_set_piomode,
+	.set_dmamode	= ali_set_dmamode,
+	.mode_filter	= ali_20_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+	.dev_config	= ali_lock_sectors,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ali_early_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Port operations for DMA capable ALi with cable detect
+ */
+static struct ata_port_operations ali_c2_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= ali_set_piomode,
+	.set_dmamode	= ali_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+	.dev_config	= ali_lock_sectors,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ali_c2_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Port operations for DMA capable ALi with cable detect and LBA48
+ */
+static struct ata_port_operations ali_c5_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= ali_set_piomode,
+	.set_dmamode	= ali_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ali_c2_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	ali_init_one		-	discovery callback
+ *	@pdev: PCI device ID
+ *	@id: PCI table info
+ *
+ *	An ALi IDE interface has been discovered. Figure out what revision
+ *	and perform configuration work before handing it to the ATA layer
+ */
+
+static int ali_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info_early = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.port_ops = &ali_early_port_ops
+	};
+	/* Revision 0x20 added DMA */
+	static struct ata_port_info info_20 = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST | ATA_FLAG_PIO_LBA48,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &ali_20_port_ops
+	};
+	/* Revision 0x20 with support logic added UDMA */
+	static struct ata_port_info info_20_udma = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST | ATA_FLAG_PIO_LBA48,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07, 
+		.udma_mask = 0x07,	/* UDMA33 */
+		.port_ops = &ali_20_port_ops
+	};
+	/* Revision 0xC2 adds UDMA66 */
+	static struct ata_port_info info_c2 = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST | ATA_FLAG_PIO_LBA48,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x1f,
+		.port_ops = &ali_c2_port_ops
+	};
+	/* Revision 0xC3 is UDMA100 */
+	static struct ata_port_info info_c3 = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST | ATA_FLAG_PIO_LBA48,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &ali_c2_port_ops
+	};
+	/* Revision 0xC4 is UDMA133 */
+	static struct ata_port_info info_c4 = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST | ATA_FLAG_PIO_LBA48,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &ali_c2_port_ops
+	};
+	/* Revision 0xC5 is UDMA133 with LBA48 DMA */
+	static struct ata_port_info info_c5 = {
+		.sht = &ali_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &ali_c5_port_ops
+	};
+
+	static struct ata_port_info *port_info[2];
+	u8 rev, tmp;
+	struct pci_dev *north, *isa_bridge;
+
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &rev);
+
+	/*
+	 * The chipset revision selects the driver operations and
+	 * mode data.
+	 */
+
+	if (rev < 0x20) {
+		port_info[0] = port_info[1] = &info_early;
+	} else if (rev < 0xC2) {
+		/* 1543-E/F, 1543C-C, 1543C-D, 1543C-E */
+		pci_read_config_byte(pdev, 0x4B, &tmp);
+		/* Clear CD-ROM DMA write bit */
+		tmp &= 0x7F;
+		pci_write_config_byte(pdev, 0x4B, tmp);
+        	port_info[0] = port_info[1] = &info_20;
+	} else if (rev == 0xC2) {
+        	port_info[0] = port_info[1] = &info_c2;
+	} else if (rev == 0xC3) {
+        	port_info[0] = port_info[1] = &info_c3;
+	} else if (rev == 0xC4) {
+        	port_info[0] = port_info[1] = &info_c4;
+	} else
+        	port_info[0] = port_info[1] = &info_c5;
+
+	if (rev >= 0xC2) {
+		/* Enable cable detection logic */
+		pci_read_config_byte(pdev, 0x4B, &tmp);
+		pci_write_config_byte(pdev, 0x4B, tmp | 0x08);
+	}
+
+	north = pci_get_slot(pdev->bus, PCI_DEVFN(0,0));
+	isa_bridge = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, NULL);
+
+	if (north && north->vendor == PCI_VENDOR_ID_AL) {
+		/* Configure the ALi bridge logic. For non ALi rely on BIOS.
+		   Set the south bridge enable bit */
+		pci_read_config_byte(isa_bridge, 0x79, &tmp);
+		if (rev == 0xC2)
+			pci_write_config_byte(isa_bridge, 0x79, tmp | 0x04);
+		else if (rev > 0xC2)
+			pci_write_config_byte(isa_bridge, 0x79, tmp | 0x02);
+	}
+
+	if (rev >= 0x20) {
+		if (rev < 0xC2) {
+			/* Are we paired with a UDMA capable chip */
+			pci_read_config_byte(isa_bridge, 0x5E, &tmp);
+			if ((tmp & 0x1E) == 0x12)
+		        	port_info[0] = port_info[1] = &info_20_udma;
+		}
+		/*
+		 * CD_ROM DMA on (0x53 bit 0). Enable this even if we want
+		 * to use PIO. 0x53 bit 1 (rev 20 only) - enable FIFO control
+		 * via 0x54/55.
+		 */
+		pci_read_config_byte(pdev, 0x53, &tmp);
+		if (rev <= 0x20)
+			tmp &= ~0x02;
+		if (rev == 0xc7)
+			tmp |= 0x03;
+		else
+			tmp |= 0x01;	/* CD_ROM enable for DMA */
+		pci_write_config_byte(pdev, 0x53, tmp);
+	}
+
+	pci_dev_put(isa_bridge);
+	pci_dev_put(north);
+
+	ata_pci_clear_simplex(pdev);
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static struct pci_device_id ali[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M5228), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M5229), },
+	{ 0, },
+};
+
+static struct pci_driver ali_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= ali,
+	.probe 		= ali_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init ali_init(void)
+{
+	return pci_register_driver(&ali_pci_driver);
+}
+
+
+static void __exit ali_exit(void)
+{
+	pci_unregister_driver(&ali_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for ALi PATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, ali);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(ali_init);
+module_exit(ali_exit);
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
new file mode 100644
index 000000000000..6ab568428a14
--- /dev/null
+++ b/drivers/ata/pata_amd.c
@@ -0,0 +1,707 @@
+/*
+ * pata_amd.c 	- AMD PATA for new ATA layer
+ *			  (C) 2005-2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ *  Based on pata-sil680. Errata information is taken from data sheets
+ *  and the amd74xx.c driver by Vojtech Pavlik. Nvidia SATA devices are
+ *  claimed by sata-nv.c.
+ *
+ *  TODO:
+ *	Variable system clock when/if it makes sense
+ *	Power management on ports
+ *
+ *
+ *  Documentation publically available.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_amd"
+#define DRV_VERSION "0.2.2"
+
+/**
+ *	timing_setup		-	shared timing computation and load
+ *	@ap: ATA port being set up
+ *	@adev: drive being configured
+ *	@offset: port offset
+ *	@speed: target speed
+ *	@clock: clock multiplier (number of times 33MHz for this part)
+ *
+ *	Perform the actual timing set up for Nvidia or AMD PATA devices.
+ *	The actual devices vary so they all call into this helper function
+ *	providing the clock multipler and offset (because AMD and Nvidia put
+ *	the ports at different locations).
+ */
+
+static void timing_setup(struct ata_port *ap, struct ata_device *adev, int offset, int speed, int clock)
+{
+	static const unsigned char amd_cyc2udma[] = {
+		6, 6, 5, 4, 0, 1, 1, 2, 2, 3, 3, 3, 3, 3, 3, 7
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct ata_device *peer = ata_dev_pair(adev);
+	int dn = ap->port_no * 2 + adev->devno;
+	struct ata_timing at, apeer;
+	int T, UT;
+	const int amd_clock = 33333;	/* KHz. */
+	u8 t;
+
+	T = 1000000000 / amd_clock;
+	UT = T / min_t(int, max_t(int, clock, 1), 2);
+
+	if (ata_timing_compute(adev, speed, &at, T, UT) < 0) {
+		dev_printk(KERN_ERR, &pdev->dev, "unknown mode %d.\n", speed);
+		return;
+	}
+
+	if (peer) {
+		/* This may be over conservative */
+		if (peer->dma_mode) {
+			ata_timing_compute(peer, peer->dma_mode, &apeer, T, UT);
+			ata_timing_merge(&apeer, &at, &at, ATA_TIMING_8BIT);
+		}
+		ata_timing_compute(peer, peer->pio_mode, &apeer, T, UT);
+		ata_timing_merge(&apeer, &at, &at, ATA_TIMING_8BIT);
+	}
+
+	if (speed == XFER_UDMA_5 && amd_clock <= 33333) at.udma = 1;
+	if (speed == XFER_UDMA_6 && amd_clock <= 33333) at.udma = 15;
+
+	/*
+	 *	Now do the setup work
+	 */
+
+	/* Configure the address set up timing */
+	pci_read_config_byte(pdev, offset + 0x0C, &t);
+	t = (t & ~(3 << ((3 - dn) << 1))) | ((FIT(at.setup, 1, 4) - 1) << ((3 - dn) << 1));
+	pci_write_config_byte(pdev, offset + 0x0C , t);
+
+	/* Configure the 8bit I/O timing */
+	pci_write_config_byte(pdev, offset + 0x0E + (1 - (dn >> 1)),
+		((FIT(at.act8b, 1, 16) - 1) << 4) | (FIT(at.rec8b, 1, 16) - 1));
+
+	/* Drive timing */
+	pci_write_config_byte(pdev, offset + 0x08 + (3 - dn),
+		((FIT(at.active, 1, 16) - 1) << 4) | (FIT(at.recover, 1, 16) - 1));
+
+	switch (clock) {
+		case 1:
+		t = at.udma ? (0xc0 | (FIT(at.udma, 2, 5) - 2)) : 0x03;
+		break;
+
+		case 2:
+		t = at.udma ? (0xc0 | amd_cyc2udma[FIT(at.udma, 2, 10)]) : 0x03;
+		break;
+
+		case 3:
+		t = at.udma ? (0xc0 | amd_cyc2udma[FIT(at.udma, 1, 10)]) : 0x03;
+		break;
+
+		case 4:
+		t = at.udma ? (0xc0 | amd_cyc2udma[FIT(at.udma, 1, 15)]) : 0x03;
+		break;
+
+		default:
+			return;
+	}
+
+	/* UDMA timing */
+	pci_write_config_byte(pdev, offset + 0x10 + (3 - dn), t);
+}
+
+/**
+ *	amd_probe_init		-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection. The BIOS stores this in PCI config
+ *	space for us.
+ */
+
+static int amd_pre_reset(struct ata_port *ap)
+{
+	static const u32 bitmask[2] = {0x03, 0xC0};
+	static const struct pci_bits amd_enable_bits[] = {
+		{ 0x40, 1, 0x02, 0x02 },
+		{ 0x40, 1, 0x01, 0x01 }
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 ata66;
+	
+	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+
+	pci_read_config_byte(pdev, 0x42, &ata66);
+	if (ata66 & bitmask[ap->port_no])
+		ap->cbl = ATA_CBL_PATA80;
+	else
+		ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+
+}
+
+static void amd_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, amd_pre_reset,
+				      ata_std_softreset, NULL,
+				      ata_std_postreset);
+}
+
+static int amd_early_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static struct pci_bits amd_enable_bits[] = {
+		{ 0x40, 1, 0x02, 0x02 },
+		{ 0x40, 1, 0x01, 0x01 }
+	};
+
+	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	/* No host side cable detection */
+	ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+
+}
+
+static void amd_early_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, amd_early_pre_reset,
+			       ata_std_softreset, NULL,
+			       ata_std_postreset);
+}
+
+/**
+ *	amd33_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the AMD registers for PIO mode.
+ */
+
+static void amd33_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->pio_mode, 1);
+}
+
+static void amd66_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->pio_mode, 2);
+}
+
+static void amd100_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->pio_mode, 3);
+}
+
+static void amd133_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->pio_mode, 4);
+}
+
+/**
+ *	amd33_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the MWDMA/UDMA modes for the AMD and Nvidia
+ *	chipset.
+ */
+
+static void amd33_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->dma_mode, 1);
+}
+
+static void amd66_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->dma_mode, 2);
+}
+
+static void amd100_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->dma_mode, 3);
+}
+
+static void amd133_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x40, adev->dma_mode, 4);
+}
+
+
+/**
+ *	nv_probe_init	-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection. The BIOS stores this in PCI config
+ *	space for us.
+ */
+
+static int nv_pre_reset(struct ata_port *ap) {
+	static const u8 bitmask[2] = {0x03, 0xC0};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 ata66;
+	u16 udma;
+
+	pci_read_config_byte(pdev, 0x52, &ata66);
+	if (ata66 & bitmask[ap->port_no])
+		ap->cbl = ATA_CBL_PATA80;
+	else
+		ap->cbl = ATA_CBL_PATA40;
+
+	/* We now have to double check because the Nvidia boxes BIOS
+	   doesn't always set the cable bits but does set mode bits */
+
+	pci_read_config_word(pdev, 0x62 - 2 * ap->port_no, &udma);
+	if ((udma & 0xC4) == 0xC4 || (udma & 0xC400) == 0xC400)
+		ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+static void nv_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, nv_pre_reset,
+			       ata_std_softreset, NULL,
+			       ata_std_postreset);
+}
+/**
+ *	nv100_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the AMD registers for PIO mode.
+ */
+
+static void nv100_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x50, adev->pio_mode, 3);
+}
+
+static void nv133_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x50, adev->pio_mode, 4);
+}
+
+/**
+ *	nv100_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the MWDMA/UDMA modes for the AMD and Nvidia
+ *	chipset.
+ */
+
+static void nv100_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x50, adev->dma_mode, 3);
+}
+
+static void nv133_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	timing_setup(ap, adev, 0x50, adev->dma_mode, 4);
+}
+
+static struct scsi_host_template amd_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations amd33_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= amd33_set_piomode,
+	.set_dmamode	= amd33_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= amd_early_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations amd66_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= amd66_set_piomode,
+	.set_dmamode	= amd66_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= amd_early_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations amd100_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= amd100_set_piomode,
+	.set_dmamode	= amd100_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= amd_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations amd133_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= amd133_set_piomode,
+	.set_dmamode	= amd133_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= amd_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations nv100_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= nv100_set_piomode,
+	.set_dmamode	= nv100_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= nv_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations nv133_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= nv133_set_piomode,
+	.set_dmamode	= nv133_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= nv_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int amd_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info[10] = {
+		{	/* 0: AMD 7401 */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,	/* No SWDMA */
+			.udma_mask = 0x07,	/* UDMA 33 */
+			.port_ops = &amd33_port_ops
+		},
+		{	/* 1: Early AMD7409 - no swdma */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x1f,	/* UDMA 66 */
+			.port_ops = &amd66_port_ops
+		},
+		{	/* 2: AMD 7409, no swdma errata */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x1f,	/* UDMA 66 */
+			.port_ops = &amd66_port_ops
+		},
+		{	/* 3: AMD 7411 */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x3f,	/* UDMA 100 */
+			.port_ops = &amd100_port_ops
+		},
+		{	/* 4: AMD 7441 */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x3f,	/* UDMA 100 */
+			.port_ops = &amd100_port_ops
+		},
+		{	/* 5: AMD 8111*/
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x7f,	/* UDMA 133, no swdma */
+			.port_ops = &amd133_port_ops
+		},
+		{	/* 6: AMD 8111 UDMA 100 (Serenade) */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x3f,	/* UDMA 100, no swdma */
+			.port_ops = &amd133_port_ops
+		},
+		{	/* 7: Nvidia Nforce */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x3f,	/* UDMA 100 */
+			.port_ops = &nv100_port_ops
+		},
+		{	/* 8: Nvidia Nforce2 and later */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x7f,	/* UDMA 133, no swdma */
+			.port_ops = &nv133_port_ops
+		},
+		{	/* 9: AMD CS5536 (Geode companion) */
+			.sht = &amd_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x3f,	/* UDMA 100 */
+			.port_ops = &amd100_port_ops
+		}
+	};
+	static struct ata_port_info *port_info[2];
+	static int printed_version;
+	int type = id->driver_data;
+	u8 rev;
+	u8 fifo;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev, "version " DRV_VERSION "\n");
+
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &rev);
+	pci_read_config_byte(pdev, 0x41, &fifo);
+
+	/* Check for AMD7409 without swdma errata and if found adjust type */
+	if (type == 1 && rev > 0x7)
+		type = 2;
+
+	/* Check for AMD7411 */
+	if (type == 3)
+		/* FIFO is broken */
+		pci_write_config_byte(pdev, 0x41, fifo & 0x0F);
+	else
+		pci_write_config_byte(pdev, 0x41, fifo | 0xF0);
+
+	/* Serenade ? */
+	if (type == 5 && pdev->subsystem_vendor == PCI_VENDOR_ID_AMD &&
+			 pdev->subsystem_device == PCI_DEVICE_ID_AMD_SERENADE)
+		type = 6;	/* UDMA 100 only */
+
+	if (type < 3)
+		ata_pci_clear_simplex(pdev);
+
+	/* And fire it up */
+
+	port_info[0] = port_info[1] = &info[type];
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id amd[] = {
+	{ PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_COBRA_7401,		PCI_ANY_ID, PCI_ANY_ID, 0, 0,  0 },
+	{ PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_VIPER_7409,		PCI_ANY_ID, PCI_ANY_ID, 0, 0,  1 },
+	{ PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_VIPER_7411,		PCI_ANY_ID, PCI_ANY_ID, 0, 0,  3 },
+	{ PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_OPUS_7441,		PCI_ANY_ID, PCI_ANY_ID, 0, 0,  4 },
+	{ PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_8111_IDE,		PCI_ANY_ID, PCI_ANY_ID, 0, 0,  5 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0,  7 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE2_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0,  8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE2S_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0,  8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE3_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0,  8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE3S_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8 },
+	{ PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE,	PCI_ANY_ID, PCI_ANY_ID, 0, 0, 8 },
+	{ PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_CS5536_IDE,		PCI_ANY_ID, PCI_ANY_ID, 0, 0, 9 },
+	{ 0, },
+};
+
+static struct pci_driver amd_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= amd,
+	.probe 		= amd_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init amd_init(void)
+{
+	return pci_register_driver(&amd_pci_driver);
+}
+
+static void __exit amd_exit(void)
+{
+	pci_unregister_driver(&amd_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for AMD PATA IDE");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, amd);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(amd_init);
+module_exit(amd_exit);
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
new file mode 100644
index 000000000000..d6ef3bf1bac7
--- /dev/null
+++ b/drivers/ata/pata_artop.c
@@ -0,0 +1,518 @@
+/*
+ *    pata_artop.c - ARTOP ATA controller driver
+ *
+ *	(C) 2006 Red Hat <alan@redhat.com>
+ *
+ *    Based in part on drivers/ide/pci/aec62xx.c
+ *	Copyright (C) 1999-2002	Andre Hedrick <andre@linux-ide.org>
+ *	865/865R fixes for Macintosh card version from a patch to the old
+ *		driver by Thibaut VARENE <varenet@parisc-linux.org>
+ *	When setting the PCI latency we must set 0x80 or higher for burst
+ *		performance Alessandro Zummo <alessandro.zummo@towertech.it>
+ *
+ *	TODO
+ *	850 serialization once the core supports it
+ *	Investigate no_dsc on 850R
+ *	Clock detect
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_artop"
+#define DRV_VERSION	"0.4.1"
+
+/*
+ *	The ARTOP has 33 Mhz and "over clocked" timing tables. Until we
+ *	get PCI bus speed functionality we leave this as 0. Its a variable
+ *	for when we get the functionality and also for folks wanting to
+ *	test stuff.
+ */
+
+static int clock = 0;
+
+static int artop6210_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	const struct pci_bits artop_enable_bits[] = {
+		{ 0x4AU, 1U, 0x02UL, 0x02UL },	/* port 0 */
+		{ 0x4AU, 1U, 0x04UL, 0x04UL },	/* port 1 */
+	};
+
+	if (!pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	artop6210_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6210_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, artop6210_pre_reset,
+				    ata_std_softreset, NULL,
+				    ata_std_postreset);
+}
+
+/**
+ *	artop6260_pre_reset	-	check for 40/80 pin
+ *	@ap: Port
+ *
+ *	The ARTOP hardware reports the cable detect bits in register 0x49.
+ *	Nothing complicated needed here.
+ */
+
+static int artop6260_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits artop_enable_bits[] = {
+		{ 0x4AU, 1U, 0x02UL, 0x02UL },	/* port 0 */
+		{ 0x4AU, 1U, 0x04UL, 0x04UL },	/* port 1 */
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 tmp;
+
+	/* Odd numbered device ids are the units with enable bits (the -R cards) */
+	if (pdev->device % 1 && !pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	pci_read_config_byte(pdev, 0x49, &tmp);
+	if (tmp & (1 >> ap->port_no))
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	artop6260_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6260_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, artop6260_pre_reset,
+				    ata_std_softreset, NULL,
+				    ata_std_postreset);
+}
+
+/**
+ *	artop6210_load_piomode - Load a set of PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device
+ *	@pio: PIO mode
+ *
+ *	Set PIO mode for device, in host controller PCI config space. This
+ *	is used both to set PIO timings in PIO mode and also to set the
+ *	matching PIO clocking for UDMA, as well as the MWDMA timings.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6210_load_piomode(struct ata_port *ap, struct ata_device *adev, unsigned int pio)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int dn = adev->devno + 2 * ap->port_no;
+	const u16 timing[2][5] = {
+		{ 0x0000, 0x000A, 0x0008, 0x0303, 0x0301 },
+		{ 0x0700, 0x070A, 0x0708, 0x0403, 0x0401 }
+
+	};
+	/* Load the PIO timing active/recovery bits */
+	pci_write_config_word(pdev, 0x40 + 2 * dn, timing[clock][pio]);
+}
+
+/**
+ *	artop6210_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring
+ *
+ *	Set PIO mode for device, in host controller PCI config space. For
+ *	ARTOP we must also clear the UDMA bits if we are not doing UDMA. In
+ *	the event UDMA is used the later call to set_dmamode will set the
+ *	bits as required.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6210_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int dn = adev->devno + 2 * ap->port_no;
+	u8 ultra;
+
+	artop6210_load_piomode(ap, adev, adev->pio_mode - XFER_PIO_0);
+
+	/* Clear the UDMA mode bits (set_dmamode will redo this if needed) */
+	pci_read_config_byte(pdev, 0x54, &ultra);
+	ultra &= ~(3 << (2 * dn));
+	pci_write_config_byte(pdev, 0x54, ultra);
+}
+
+/**
+ *	artop6260_load_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring
+ *	@pio: PIO mode
+ *
+ *	Set PIO mode for device, in host controller PCI config space. The
+ *	ARTOP6260 and relatives store the timing data differently.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6260_load_piomode (struct ata_port *ap, struct ata_device *adev, unsigned int pio)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int dn = adev->devno + 2 * ap->port_no;
+	const u8 timing[2][5] = {
+		{ 0x00, 0x0A, 0x08, 0x33, 0x31 },
+		{ 0x70, 0x7A, 0x78, 0x43, 0x41 }
+
+	};
+	/* Load the PIO timing active/recovery bits */
+	pci_write_config_byte(pdev, 0x40 + dn, timing[clock][pio]);
+}
+
+/**
+ *	artop6260_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring
+ *
+ *	Set PIO mode for device, in host controller PCI config space. For
+ *	ARTOP we must also clear the UDMA bits if we are not doing UDMA. In
+ *	the event UDMA is used the later call to set_dmamode will set the
+ *	bits as required.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6260_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	u8 ultra;
+
+	artop6260_load_piomode(ap, adev, adev->pio_mode - XFER_PIO_0);
+
+	/* Clear the UDMA mode bits (set_dmamode will redo this if needed) */
+	pci_read_config_byte(pdev, 0x44 + ap->port_no, &ultra);
+	ultra &= ~(7 << (4  * adev->devno));	/* One nibble per drive */
+	pci_write_config_byte(pdev, 0x44 + ap->port_no, ultra);
+}
+
+/**
+ *	artop6210_set_dmamode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: um
+ *
+ *	Set DMA mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6210_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int pio;
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int dn = adev->devno + 2 * ap->port_no;
+	u8 ultra;
+
+	if (adev->dma_mode == XFER_MW_DMA_0)
+		pio = 1;
+	else
+		pio = 4;
+
+	/* Load the PIO timing active/recovery bits */
+	artop6210_load_piomode(ap, adev, pio);
+
+	pci_read_config_byte(pdev, 0x54, &ultra);
+	ultra &= ~(3 << (2 * dn));
+
+	/* Add ultra DMA bits if in UDMA mode */
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		u8 mode = (adev->dma_mode - XFER_UDMA_0) + 1 - clock;
+		if (mode == 0)
+			mode = 1;
+		ultra |= (mode << (2 * dn));
+	}
+	pci_write_config_byte(pdev, 0x54, ultra);
+}
+
+/**
+ *	artop6260_set_dmamode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring
+ *
+ *	Set DMA mode for device, in host controller PCI config space. The
+ *	ARTOP6260 and relatives store the timing data differently.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void artop6260_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int pio	= adev->pio_mode - XFER_PIO_0;
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	u8 ultra;
+
+	if (adev->dma_mode == XFER_MW_DMA_0)
+		pio = 1;
+	else
+		pio = 4;
+
+	/* Load the PIO timing active/recovery bits */
+	artop6260_load_piomode(ap, adev, pio);
+
+	/* Add ultra DMA bits if in UDMA mode */
+	pci_read_config_byte(pdev, 0x44 + ap->port_no, &ultra);
+	ultra &= ~(7 << (4  * adev->devno));	/* One nibble per drive */
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		u8 mode = adev->dma_mode - XFER_UDMA_0 + 1 - clock;
+		if (mode == 0)
+			mode = 1;
+		ultra |= (mode << (4 * adev->devno));
+	}
+	pci_write_config_byte(pdev, 0x44 + ap->port_no, ultra);
+}
+
+static struct scsi_host_template artop_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations artop6210_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= artop6210_set_piomode,
+	.set_dmamode		= artop6210_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= artop6210_error_handler,
+	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.eng_timeout		= ata_eng_timeout,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static const struct ata_port_operations artop6260_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= artop6260_set_piomode,
+	.set_dmamode		= artop6260_set_dmamode,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= artop6260_error_handler,
+	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+
+/**
+ *	artop_init_one - Register ARTOP ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in artop_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int artop_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	static int printed_version;
+	static struct ata_port_info info_6210 = {
+		.sht		= &artop_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma0-2 */
+		.udma_mask 	= ATA_UDMA2,
+		.port_ops	= &artop6210_ops,
+	};
+	static struct ata_port_info info_626x = {
+		.sht		= &artop_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma0-2 */
+		.udma_mask 	= ATA_UDMA4,
+		.port_ops	= &artop6260_ops,
+	};
+	static struct ata_port_info info_626x_fast = {
+		.sht		= &artop_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma0-2 */
+		.udma_mask 	= ATA_UDMA5,
+		.port_ops	= &artop6260_ops,
+	};
+	struct ata_port_info *port_info[2];
+	struct ata_port_info *info;
+	int ports = 2;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev,
+			   "version " DRV_VERSION "\n");
+
+	if (id->driver_data == 0) {	/* 6210 variant */
+		info = &info_6210;
+		/* BIOS may have left us in UDMA, clear it before libata probe */
+		pci_write_config_byte(pdev, 0x54, 0);
+		/* For the moment (also lacks dsc) */
+		printk(KERN_WARNING "ARTOP 6210 requires serialize functionality not yet supported by libata.\n");
+		printk(KERN_WARNING "Secondary ATA ports will not be activated.\n");
+		ports = 1;
+	}
+	else if (id->driver_data == 1)	/* 6260 */
+		info = &info_626x;
+	else if (id->driver_data == 2)	{ /* 6260 or 6260 + fast */
+		unsigned long io = pci_resource_start(pdev, 4);
+		u8 reg;
+
+		info = &info_626x;
+		if (inb(io) & 0x10)
+			info = &info_626x_fast;
+		/* Mac systems come up with some registers not set as we
+		   will need them */
+
+		/* Clear reset & test bits */
+		pci_read_config_byte(pdev, 0x49, &reg);
+		pci_write_config_byte(pdev, 0x49, reg & ~ 0x30);
+
+		/* PCI latency must be > 0x80 for burst mode, tweak it
+		 * if required.
+		 */
+		pci_read_config_byte(pdev, PCI_LATENCY_TIMER, &reg);
+		if (reg <= 0x80)
+			pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x90);
+
+		/* Enable IRQ output and burst mode */
+		pci_read_config_byte(pdev, 0x4a, &reg);
+		pci_write_config_byte(pdev, 0x4a, (reg & ~0x01) | 0x80);
+
+	}
+	port_info[0] = port_info[1] = info;
+	return ata_pci_init_one(pdev, port_info, ports);
+}
+
+static const struct pci_device_id artop_pci_tbl[] = {
+	{ 0x1191, 0x0005, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{ 0x1191, 0x0006, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
+	{ 0x1191, 0x0007, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
+	{ 0x1191, 0x0008, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2},
+	{ 0x1191, 0x0009, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2},
+	{ }	/* terminate list */
+};
+
+static struct pci_driver artop_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= artop_pci_tbl,
+	.probe			= artop_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init artop_init(void)
+{
+	return pci_register_driver(&artop_pci_driver);
+}
+
+static void __exit artop_exit(void)
+{
+	pci_unregister_driver(&artop_pci_driver);
+}
+
+
+module_init(artop_init);
+module_exit(artop_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for ARTOP PATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, artop_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
new file mode 100644
index 000000000000..3f78a1e54a75
--- /dev/null
+++ b/drivers/ata/pata_atiixp.c
@@ -0,0 +1,306 @@
+/*
+ * pata_atiixp.c 	- ATI PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based on
+ *
+ *  linux/drivers/ide/pci/atiixp.c	Version 0.01-bart2	Feb. 26, 2004
+ *
+ *  Copyright (C) 2003 ATI Inc. <hyu@ati.com>
+ *  Copyright (C) 2004 Bartlomiej Zolnierkiewicz
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_atiixp"
+#define DRV_VERSION "0.4.2"
+
+enum {
+	ATIIXP_IDE_PIO_TIMING	= 0x40,
+	ATIIXP_IDE_MWDMA_TIMING	= 0x44,
+	ATIIXP_IDE_PIO_CONTROL	= 0x48,
+	ATIIXP_IDE_PIO_MODE	= 0x4a,
+	ATIIXP_IDE_UDMA_CONTROL	= 0x54,
+	ATIIXP_IDE_UDMA_MODE 	= 0x56
+};
+
+static int atiixp_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static struct pci_bits atiixp_enable_bits[] = {
+		{ 0x48, 1, 0x01, 0x00 },
+		{ 0x48, 1, 0x08, 0x00 }
+	};
+
+	if (!pci_test_config_bits(pdev, &atiixp_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+static void atiixp_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, atiixp_pre_reset, ata_std_softreset, NULL,   ata_std_postreset);
+}
+
+/**
+ *	atiixp_set_pio_timing	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called by both the pio and dma setup functions to set the controller
+ *	timings for PIO transfers. We must load both the mode number and
+ *	timing values into the controller.
+ */
+
+static void atiixp_set_pio_timing(struct ata_port *ap, struct ata_device *adev, int pio)
+{
+	static u8 pio_timings[5] = { 0x5D, 0x47, 0x34, 0x22, 0x20 };
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int dn = 2 * ap->port_no + adev->devno;
+
+	/* Check this is correct - the order is odd in both drivers */
+	int timing_shift = (16 * ap->port_no) + 8 * (adev->devno ^ 1);
+	u16 pio_mode_data, pio_timing_data;
+
+	pci_read_config_word(pdev, ATIIXP_IDE_PIO_MODE, &pio_mode_data);
+	pio_mode_data &= ~(0x7 << (4 * dn));
+	pio_mode_data |= pio << (4 * dn);
+	pci_write_config_word(pdev, ATIIXP_IDE_PIO_MODE, pio_mode_data);
+
+	pci_read_config_word(pdev, ATIIXP_IDE_PIO_TIMING, &pio_timing_data);
+	pio_mode_data &= ~(0xFF << timing_shift);
+	pio_mode_data |= (pio_timings[pio] << timing_shift);
+	pci_write_config_word(pdev, ATIIXP_IDE_PIO_TIMING, pio_timing_data);
+}
+
+/**
+ *	atiixp_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup. We use a shared helper for this
+ *	as the DMA setup must also adjust the PIO timing information.
+ */
+
+static void atiixp_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	atiixp_set_pio_timing(ap, adev, adev->pio_mode - XFER_PIO_0);
+}
+
+/**
+ *	atiixp_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the DMA mode setup. We use timing tables for most
+ *	modes but must tune an appropriate PIO mode to match.
+ */
+
+static void atiixp_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static u8 mwdma_timings[5] = { 0x77, 0x21, 0x20 };
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int dma = adev->dma_mode;
+	int dn = 2 * ap->port_no + adev->devno;
+	int wanted_pio;
+
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		u16 udma_mode_data;
+
+		dma -= XFER_UDMA_0;
+
+		pci_read_config_word(pdev, ATIIXP_IDE_UDMA_MODE, &udma_mode_data);
+		udma_mode_data &= ~(0x7 << (4 * dn));
+		udma_mode_data |= dma << (4 * dn);
+		pci_write_config_word(pdev, ATIIXP_IDE_UDMA_MODE, udma_mode_data);
+	} else {
+		u16 mwdma_timing_data;
+		/* Check this is correct - the order is odd in both drivers */
+		int timing_shift = (16 * ap->port_no) + 8 * (adev->devno ^ 1);
+
+		dma -= XFER_MW_DMA_0;
+
+		pci_read_config_word(pdev, ATIIXP_IDE_MWDMA_TIMING, &mwdma_timing_data);
+		mwdma_timing_data &= ~(0xFF << timing_shift);
+		mwdma_timing_data |= (mwdma_timings[dma] << timing_shift);
+		pci_write_config_word(pdev, ATIIXP_IDE_MWDMA_TIMING, mwdma_timing_data);
+	}
+	/*
+	 *	We must now look at the PIO mode situation. We may need to
+	 *	adjust the PIO mode to keep the timings acceptable
+	 */
+	 if (adev->dma_mode >= XFER_MW_DMA_2)
+	 	wanted_pio = 4;
+	else if (adev->dma_mode == XFER_MW_DMA_1)
+		wanted_pio = 3;
+	else if (adev->dma_mode == XFER_MW_DMA_0)
+		wanted_pio = 0;
+	else BUG();
+
+	if (adev->pio_mode != wanted_pio)
+		atiixp_set_pio_timing(ap, adev, wanted_pio);
+}
+
+/**
+ *	atiixp_bmdma_start	-	DMA start callback
+ *	@qc: Command in progress
+ *
+ *	When DMA begins we need to ensure that the UDMA control
+ *	register for the channel is correctly set.
+ */
+
+static void atiixp_bmdma_start(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int dn = (2 * ap->port_no) + adev->devno;
+	u16 tmp16;
+
+	pci_read_config_word(pdev, ATIIXP_IDE_UDMA_CONTROL, &tmp16);
+	if (adev->dma_mode >= XFER_UDMA_0)
+		tmp16 |= (1 << dn);
+	else
+		tmp16 &= ~(1 << dn);
+	pci_write_config_word(pdev, ATIIXP_IDE_UDMA_CONTROL, tmp16);
+	ata_bmdma_start(qc);
+}
+
+/**
+ *	atiixp_dma_stop	-	DMA stop callback
+ *	@qc: Command in progress
+ *
+ *	DMA has completed. Clear the UDMA flag as the next operations will
+ *	be PIO ones not UDMA data transfer.
+ */
+
+static void atiixp_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int dn = (2 * ap->port_no) + qc->dev->devno;
+	u16 tmp16;
+
+	pci_read_config_word(pdev, ATIIXP_IDE_UDMA_CONTROL, &tmp16);
+	tmp16 &= ~(1 << dn);
+	pci_write_config_word(pdev, ATIIXP_IDE_UDMA_CONTROL, tmp16);
+	ata_bmdma_stop(qc);
+}
+
+static struct scsi_host_template atiixp_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations atiixp_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= atiixp_set_piomode,
+	.set_dmamode	= atiixp_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= atiixp_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= atiixp_bmdma_start,
+	.bmdma_stop	= atiixp_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &atiixp_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x06,	/* No MWDMA0 support */
+		.udma_mask = 0x3F,
+		.port_ops = &atiixp_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id atiixp[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP200_IDE), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP300_IDE), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP400_IDE), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP600_IDE), },
+	{ 0, },
+};
+
+static struct pci_driver atiixp_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= atiixp,
+	.probe 		= atiixp_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init atiixp_init(void)
+{
+	return pci_register_driver(&atiixp_pci_driver);
+}
+
+
+static void __exit atiixp_exit(void)
+{
+	pci_unregister_driver(&atiixp_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for ATI IXP200/300/400");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, atiixp);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(atiixp_init);
+module_exit(atiixp_exit);
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
new file mode 100644
index 000000000000..03284611a727
--- /dev/null
+++ b/drivers/ata/pata_cmd64x.c
@@ -0,0 +1,505 @@
+/*
+ * pata_cmd64x.c 	- ATI PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based upon
+ * linux/drivers/ide/pci/cmd64x.c		Version 1.30	Sept 10, 2002
+ *
+ * cmd64x.c: Enable interrupts at initialization time on Ultra/PCI machines.
+ *           Note, this driver is not used at all on other systems because
+ *           there the "BIOS" has done all of the following already.
+ *           Due to massive hardware bugs, UltraDMA is only supported
+ *           on the 646U2 and not on the 646U.
+ *
+ * Copyright (C) 1998		Eddie C. Dost  (ecd@skynet.be)
+ * Copyright (C) 1998		David S. Miller (davem@redhat.com)
+ *
+ * Copyright (C) 1999-2002	Andre Hedrick <andre@linux-ide.org>
+ *
+ * TODO
+ *	Testing work
+ */
+ 
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_cmd64x"
+#define DRV_VERSION "0.2.1"
+
+/*
+ * CMD64x specific registers definition.
+ */
+ 
+enum {
+	CFR 		= 0x50,
+		CFR_INTR_CH0  = 0x02,
+	CNTRL 		= 0x51,
+		CNTRL_DIS_RA0 = 0x40,
+		CNTRL_DIS_RA1 = 0x80,
+		CNTRL_ENA_2ND = 0x08,
+	CMDTIM 		= 0x52,
+	ARTTIM0 	= 0x53,
+	DRWTIM0 	= 0x54,
+	ARTTIM1 	= 0x55,
+	DRWTIM1 	= 0x56,
+	ARTTIM23 	= 0x57,
+		ARTTIM23_DIS_RA2  = 0x04,
+		ARTTIM23_DIS_RA3  = 0x08,
+		ARTTIM23_INTR_CH1 = 0x10,
+	ARTTIM2 	= 0x57,
+	ARTTIM3 	= 0x57,
+	DRWTIM23	= 0x58,
+	DRWTIM2 	= 0x58,
+	BRST 		= 0x59,
+	DRWTIM3 	= 0x5b,
+	BMIDECR0	= 0x70,
+	MRDMODE		= 0x71,
+		MRDMODE_INTR_CH0 = 0x04,
+		MRDMODE_INTR_CH1 = 0x08,
+		MRDMODE_BLK_CH0  = 0x10,
+		MRDMODE_BLK_CH1	 = 0x20,
+	BMIDESR0	= 0x72,
+	UDIDETCR0	= 0x73,
+	DTPR0		= 0x74,
+	BMIDECR1	= 0x78,
+	BMIDECSR	= 0x79,
+	BMIDESR1	= 0x7A,
+	UDIDETCR1	= 0x7B,
+	DTPR1		= 0x7C
+};
+
+static int cmd64x_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+static int cmd648_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 r;
+
+	/* Check cable detect bits */
+	pci_read_config_byte(pdev, BMIDECSR, &r);
+	if (r & (1 << ap->port_no))
+		ap->cbl = ATA_CBL_PATA80;
+	else	
+		ap->cbl = ATA_CBL_PATA40;
+		
+	return ata_std_prereset(ap);
+}
+
+static void cmd64x_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, cmd64x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+static void cmd648_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, cmd648_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	cmd64x_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup.
+ */
+ 
+static void cmd64x_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct ata_timing t;
+	const unsigned long T = 1000000 / 33;
+	const u8 setup_data[] = { 0x40, 0x40, 0x40, 0x80, 0x00 };
+	
+	u8 reg;
+	
+	/* Port layout is not logical so use a table */
+	const u8 arttim_port[2][2] = { 
+		{ ARTTIM0, ARTTIM1 },
+		{ ARTTIM23, ARTTIM23 }
+	};
+	const u8 drwtim_port[2][2] = {
+		{ DRWTIM0, DRWTIM1 },
+		{ DRWTIM2, DRWTIM3 }
+	};
+	
+	int arttim = arttim_port[ap->port_no][adev->devno];
+	int drwtim = drwtim_port[ap->port_no][adev->devno];
+	
+	
+	if (ata_timing_compute(adev, adev->pio_mode, &t, T, 0) < 0) {
+		printk(KERN_ERR DRV_NAME ": mode computation failed.\n");
+		return;
+	}
+	if (ap->port_no) {
+		/* Slave has shared address setup */
+		struct ata_device *pair = ata_dev_pair(adev);
+		
+		if (pair) {
+			struct ata_timing tp;
+			ata_timing_compute(pair, pair->pio_mode, &tp, T, 0);
+			ata_timing_merge(&t, &tp, &t, ATA_TIMING_SETUP);
+		}
+	}
+	
+	printk(KERN_DEBUG DRV_NAME ": active %d recovery %d setup %d.\n",
+		t.active, t.recover, t.setup);
+	if (t.recover > 16) {
+		t.active += t.recover - 16;
+		t.recover = 16;
+	}
+	if (t.active > 16)
+		t.active = 16;
+		
+	/* Now convert the clocks into values we can actually stuff into
+	   the chip */
+	   
+	if (t.recover > 1)
+		t.recover--;
+	else
+		t.recover = 15;
+	
+	if (t.setup > 4)
+		t.setup = 0xC0;
+	else
+		t.setup = setup_data[t.setup];
+		
+	t.active &= 0x0F;	/* 0 = 16 */
+	
+	/* Load setup timing */
+	pci_read_config_byte(pdev, arttim, &reg);
+	reg &= 0x3F;
+	reg |= t.setup;
+	pci_write_config_byte(pdev, arttim, reg);
+	
+	/* Load active/recovery */
+	pci_write_config_byte(pdev, drwtim, (t.active << 4) | t.recover);	
+}
+
+/**
+ *	cmd64x_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the DMA mode setup.
+ */
+ 
+static void cmd64x_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u8 udma_data[] = {
+		0x31, 0x21, 0x11, 0x25, 0x15, 0x05
+	};
+	static const u8 mwdma_data[] = { 
+		0x30, 0x20, 0x10
+	};
+	
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 regU, regD;
+
+	int pciU = UDIDETCR0 + 8 * ap->port_no;
+	int pciD = BMIDESR0 + 8 * ap->port_no;
+	int shift = 2 * adev->devno;
+	
+	pci_read_config_byte(pdev, pciD, &regD);
+	pci_read_config_byte(pdev, pciU, &regU);
+
+	regD &= ~(0x20 << shift);
+	regU &= ~(0x35 << shift);
+	
+	if (adev->dma_mode >= XFER_UDMA_0)
+		regU |= udma_data[adev->dma_mode - XFER_UDMA_0] << shift;
+	else
+		regD |= mwdma_data[adev->dma_mode - XFER_MW_DMA_0] << shift;
+
+	regD |= 0x20 << adev->devno;
+
+	pci_write_config_byte(pdev, pciU, regU);
+	pci_write_config_byte(pdev, pciD, regD);
+}
+
+/**
+ *	cmd648_dma_stop	-	DMA stop callback
+ *	@qc: Command in progress
+ *
+ *	DMA has completed.
+ */
+
+static void cmd648_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 dma_intr;
+	int dma_reg = ap->port_no ? ARTTIM23_INTR_CH1 : CFR_INTR_CH0;
+	int dma_mask = ap->port_no ? ARTTIM2 : CFR;
+	
+	ata_bmdma_stop(qc);
+	
+	pci_read_config_byte(pdev, dma_reg, &dma_intr);
+	pci_write_config_byte(pdev, dma_reg, dma_intr | dma_mask);
+}
+		
+/**
+ *	cmd646r1_dma_stop	-	DMA stop callback
+ *	@qc: Command in progress
+ *
+ *	Stub for now while investigating the r1 quirk in the old driver.
+ */
+
+static void cmd646r1_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	ata_bmdma_stop(qc);
+}
+		
+static struct scsi_host_template cmd64x_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations cmd64x_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= cmd64x_set_piomode,
+	.set_dmamode	= cmd64x_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= cmd64x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static struct ata_port_operations cmd646r1_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= cmd64x_set_piomode,
+	.set_dmamode	= cmd64x_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= cmd64x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= cmd646r1_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static struct ata_port_operations cmd648_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= cmd64x_set_piomode,
+	.set_dmamode	= cmd64x_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= cmd648_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= cmd648_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+	
+static int cmd64x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	u32 class_rev;
+	
+	static struct ata_port_info cmd_info[6] = {
+		{	/* CMD 643 - no UDMA */
+			.sht = &cmd64x_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.port_ops = &cmd64x_port_ops
+		},
+		{	/* CMD 646 with broken UDMA */
+			.sht = &cmd64x_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.port_ops = &cmd64x_port_ops
+		},
+		{	/* CMD 646 with working UDMA */
+			.sht = &cmd64x_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = ATA_UDMA1,
+			.port_ops = &cmd64x_port_ops
+		},
+		{	/* CMD 646 rev 1  */
+			.sht = &cmd64x_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.port_ops = &cmd646r1_port_ops
+		},
+		{	/* CMD 648 */
+			.sht = &cmd64x_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = ATA_UDMA2,
+			.port_ops = &cmd648_port_ops
+		},
+		{	/* CMD 649 */
+			.sht = &cmd64x_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = ATA_UDMA3,
+			.port_ops = &cmd648_port_ops
+		}
+	};
+	static struct ata_port_info *port_info[2], *info;
+	u8 mrdmode;
+	
+	info = &cmd_info[id->driver_data];
+	
+	pci_read_config_dword(pdev, PCI_CLASS_REVISION, &class_rev);
+	class_rev &= 0xFF;
+	
+	if (id->driver_data == 0)	/* 643 */
+		ata_pci_clear_simplex(pdev);
+	
+	if (pdev->device == PCI_DEVICE_ID_CMD_646) {
+		/* Does UDMA work ? */
+		if (class_rev > 4)
+			info = &cmd_info[2];
+		/* Early rev with other problems ? */
+		else if (class_rev == 1)
+			info = &cmd_info[3];
+	}
+
+	pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 64);
+	pci_read_config_byte(pdev, MRDMODE, &mrdmode);
+	mrdmode &= ~ 0x30;	/* IRQ set up */
+	mrdmode |= 0x02;	/* Memory read line enable */
+	pci_write_config_byte(pdev, MRDMODE, mrdmode);
+	
+	/* Force PIO 0 here.. */
+	
+	/* PPC specific fixup copied from old driver */
+#ifdef CONFIG_PPC
+	pci_write_config_byte(pdev, UDIDETCR0, 0xF0);
+#endif
+	
+	port_info[0] = port_info[1] = info;
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static struct pci_device_id cmd64x[] = {
+	{ PCI_VENDOR_ID_CMD, PCI_DEVICE_ID_CMD_643, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{ PCI_VENDOR_ID_CMD, PCI_DEVICE_ID_CMD_646, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
+	{ PCI_VENDOR_ID_CMD, PCI_DEVICE_ID_CMD_648, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 4},
+	{ PCI_VENDOR_ID_CMD, PCI_DEVICE_ID_CMD_649, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 5},
+	{ 0, },
+};
+
+static struct pci_driver cmd64x_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= cmd64x,
+	.probe 		= cmd64x_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init cmd64x_init(void)
+{
+	return pci_register_driver(&cmd64x_pci_driver);
+}
+
+
+static void __exit cmd64x_exit(void)
+{
+	pci_unregister_driver(&cmd64x_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for CMD64x series PATA controllers");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, cmd64x);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(cmd64x_init);
+module_exit(cmd64x_exit);
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
new file mode 100644
index 000000000000..792ce4828510
--- /dev/null
+++ b/drivers/ata/pata_cs5520.c
@@ -0,0 +1,336 @@
+/*
+ *	IDE tuning and bus mastering support for the CS5510/CS5520
+ *	chipsets
+ *
+ *	The CS5510/CS5520 are slightly unusual devices. Unlike the
+ *	typical IDE controllers they do bus mastering with the drive in
+ *	PIO mode and smarter silicon.
+ *
+ *	The practical upshot of this is that we must always tune the
+ *	drive for the right PIO mode. We must also ignore all the blacklists
+ *	and the drive bus mastering DMA information. Also to confuse matters
+ *	further we can do DMA on PIO only drives.
+ *
+ *	DMA on the 5510 also requires we disable_hlt() during DMA on early
+ *	revisions.
+ *
+ *	*** This driver is strictly experimental ***
+ *
+ *	(c) Copyright Red Hat Inc 2002
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * Documentation:
+ *	Not publically available.
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"pata_cs5520"
+#define DRV_VERSION	"0.6.2"
+
+struct pio_clocks
+{
+	int address;
+	int assert;
+	int recovery;
+};
+
+static const struct pio_clocks cs5520_pio_clocks[]={
+	{3, 6, 11},
+	{2, 5, 6},
+	{1, 4, 3},
+	{1, 3, 2},
+	{1, 2, 1}
+};
+
+/**
+ *	cs5520_set_timings	-	program PIO timings
+ *	@ap: ATA port
+ *	@adev: ATA device
+ *
+ *	Program the PIO mode timings for the controller according to the pio
+ *	clocking table.
+ */
+
+static void cs5520_set_timings(struct ata_port *ap, struct ata_device *adev, int pio)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int slave = adev->devno;
+
+	pio -= XFER_PIO_0;
+
+	/* Channel command timing */
+	pci_write_config_byte(pdev, 0x62 + ap->port_no,
+				(cs5520_pio_clocks[pio].recovery << 4) |
+				(cs5520_pio_clocks[pio].assert));
+	/* FIXME: should these use address ? */
+	/* Read command timing */
+	pci_write_config_byte(pdev, 0x64 +  4*ap->port_no + slave,
+				(cs5520_pio_clocks[pio].recovery << 4) |
+				(cs5520_pio_clocks[pio].assert));
+	/* Write command timing */
+	pci_write_config_byte(pdev, 0x66 +  4*ap->port_no + slave,
+				(cs5520_pio_clocks[pio].recovery << 4) |
+				(cs5520_pio_clocks[pio].assert));
+}
+
+/**
+ *	cs5520_enable_dma	-	turn on DMA bits
+ *
+ *	Turn on the DMA bits for this disk. Needed because the BIOS probably
+ *	has not done the work for us. Belongs in the core SATA code.
+ */
+
+static void cs5520_enable_dma(struct ata_port *ap, struct ata_device *adev)
+{
+	/* Set the DMA enable/disable flag */
+	u8 reg = inb(ap->ioaddr.bmdma_addr + 0x02);
+	reg |= 1<<(adev->devno + 5);
+	outb(reg, ap->ioaddr.bmdma_addr + 0x02);
+}
+
+/**
+ *	cs5520_set_dmamode	-	program DMA timings
+ *	@ap: ATA port
+ *	@adev: ATA device
+ *
+ *	Program the DMA mode timings for the controller according to the pio
+ *	clocking table. Note that this device sets the DMA timings to PIO
+ *	mode values. This may seem bizarre but the 5520 architecture talks
+ *	PIO mode to the disk and DMA mode to the controller so the underlying
+ *	transfers are PIO timed.
+ */
+
+static void cs5520_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const int dma_xlate[3] = { XFER_PIO_0, XFER_PIO_3, XFER_PIO_4 };
+	cs5520_set_timings(ap, adev, dma_xlate[adev->dma_mode]);
+	cs5520_enable_dma(ap, adev);
+}
+
+/**
+ *	cs5520_set_piomode	-	program PIO timings
+ *	@ap: ATA port
+ *	@adev: ATA device
+ *
+ *	Program the PIO mode timings for the controller according to the pio
+ *	clocking table. We know pio_mode will equal dma_mode because of the
+ *	CS5520 architecture. At least once we turned DMA on and wrote a
+ *	mode setter.
+ */
+
+static void cs5520_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	cs5520_set_timings(ap, adev, adev->pio_mode);
+}
+
+
+static int cs5520_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+static void cs5520_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, cs5520_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+static struct scsi_host_template cs5520_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations cs5520_port_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= cs5520_set_piomode,
+	.set_dmamode		= cs5520_set_dmamode,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= cs5520_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	u8 pcicfg;
+	static struct ata_probe_ent probe[2];
+	int ports = 0;
+
+	/* IDE port enable bits */
+	pci_read_config_byte(dev, 0x60, &pcicfg);
+
+	/* Check if the ATA ports are enabled */
+	if ((pcicfg & 3) == 0)
+		return -ENODEV;
+
+	if ((pcicfg & 0x40) == 0) {
+		printk(KERN_WARNING DRV_NAME ": DMA mode disabled. Enabling.\n");
+		pci_write_config_byte(dev, 0x60, pcicfg | 0x40);
+	}
+
+	/* Perform set up for DMA */
+	if (pci_enable_device_bars(dev, 1<<2)) {
+		printk(KERN_ERR DRV_NAME ": unable to configure BAR2.\n");
+		return -ENODEV;
+	}
+	pci_set_master(dev);
+	if (pci_set_dma_mask(dev, DMA_32BIT_MASK)) {
+		printk(KERN_ERR DRV_NAME ": unable to configure DMA mask.\n");
+		return -ENODEV;
+	}
+	if (pci_set_consistent_dma_mask(dev, DMA_32BIT_MASK)) {
+		printk(KERN_ERR DRV_NAME ": unable to configure consistent DMA mask.\n");
+		return -ENODEV;
+	}
+
+	/* We have to do our own plumbing as the PCI setup for this
+	   chipset is non-standard so we can't punt to the libata code */
+
+	INIT_LIST_HEAD(&probe[0].node);
+	probe[0].dev = pci_dev_to_dev(dev);
+	probe[0].port_ops = &cs5520_port_ops;
+	probe[0].sht = &cs5520_sht;
+	probe[0].pio_mask = 0x1F;
+	probe[0].mwdma_mask = id->driver_data;
+	probe[0].irq = 14;
+	probe[0].irq_flags = 0;
+	probe[0].port_flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST;
+	probe[0].n_ports = 1;
+	probe[0].port[0].cmd_addr = 0x1F0;
+	probe[0].port[0].ctl_addr = 0x3F6;
+	probe[0].port[0].altstatus_addr = 0x3F6;
+	probe[0].port[0].bmdma_addr = pci_resource_start(dev, 2);
+
+	/* The secondary lurks at different addresses but is otherwise
+	   the same beastie */
+
+	probe[1] = probe[0];
+	INIT_LIST_HEAD(&probe[1].node);
+	probe[1].irq = 15;
+	probe[1].port[0].cmd_addr = 0x170;
+	probe[1].port[0].ctl_addr = 0x376;
+	probe[1].port[0].altstatus_addr = 0x376;
+	probe[1].port[0].bmdma_addr = pci_resource_start(dev, 2) + 8;
+
+	/* Let libata fill in the port details */
+	ata_std_ports(&probe[0].port[0]);
+	ata_std_ports(&probe[1].port[0]);
+
+	/* Now add the ports that are active */
+	if (pcicfg & 1)
+		ports += ata_device_add(&probe[0]);
+	if (pcicfg & 2)
+		ports += ata_device_add(&probe[1]);
+	if (ports)
+		return 0;
+	return -ENODEV;
+}
+
+/**
+ *	cs5520_remove_one	-	device unload
+ *	@pdev: PCI device being removed
+ *
+ *	Handle an unplug/unload event for a PCI device. Unload the
+ *	PCI driver but do not use the default handler as we manage
+ *	resources ourself and *MUST NOT* disable the device as it has
+ *	other functions.
+ */
+
+static void __devexit cs5520_remove_one(struct pci_dev *pdev)
+{
+	struct device *dev = pci_dev_to_dev(pdev);
+	struct ata_host *host = dev_get_drvdata(dev);
+
+	ata_host_remove(host);
+	dev_set_drvdata(dev, NULL);
+}
+
+/* For now keep DMA off. We can set it for all but A rev CS5510 once the
+   core ATA code can handle it */
+
+static struct pci_device_id pata_cs5520[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5510), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5520), },
+	{ 0, },
+};
+
+static struct pci_driver cs5520_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= pata_cs5520,
+	.probe 		= cs5520_init_one,
+	.remove		= cs5520_remove_one
+};
+
+
+static int __init cs5520_init(void)
+{
+	return pci_register_driver(&cs5520_pci_driver);
+}
+
+static void __exit cs5520_exit(void)
+{
+	pci_unregister_driver(&cs5520_pci_driver);
+}
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Cyrix CS5510/5520");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, pata_cs5520);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(cs5520_init);
+module_exit(cs5520_exit);
+
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
new file mode 100644
index 000000000000..f3d8a3bc1e78
--- /dev/null
+++ b/drivers/ata/pata_cs5530.c
@@ -0,0 +1,387 @@
+/*
+ * pata-cs5530.c 	- CS5530 PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based upon cs5530.c by Mark Lord.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Loosely based on the piix & svwks drivers.
+ *
+ * Documentation:
+ *	Available from AMD web site.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/dmi.h>
+
+#define DRV_NAME	"pata_cs5530"
+#define DRV_VERSION	"0.6"
+
+/**
+ *	cs5530_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Set our PIO requirements. This is fairly simple on the CS5530
+ *	chips.
+ */
+
+static void cs5530_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const unsigned int cs5530_pio_timings[2][5] = {
+		{0x00009172, 0x00012171, 0x00020080, 0x00032010, 0x00040010},
+		{0xd1329172, 0x71212171, 0x30200080, 0x20102010, 0x00100010}
+	};
+	unsigned long base = ( ap->ioaddr.bmdma_addr & ~0x0F) + 0x20 + 0x10 * ap->port_no;
+	u32 tuning;
+	int format;
+
+	/* Find out which table to use */
+	tuning = inl(base + 0x04);
+	format = (tuning & 0x80000000UL) ? 1 : 0;
+
+	/* Now load the right timing register */
+	if (adev->devno)
+		base += 0x08;
+
+	outl(cs5530_pio_timings[format][adev->pio_mode - XFER_PIO_0], base);
+}
+
+/**
+ *	cs5530_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	We cannot mix MWDMA and UDMA without reloading timings each switch
+ *	master to slave. We track the last DMA setup in order to minimise
+ *	reloads.
+ */
+
+static void cs5530_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned long base = ( ap->ioaddr.bmdma_addr & ~0x0F) + 0x20 + 0x10 * ap->port_no;
+	u32 tuning, timing = 0;
+	u8 reg;
+
+	/* Find out which table to use */
+	tuning = inl(base + 0x04);
+
+	switch(adev->dma_mode) {
+		case XFER_UDMA_0:
+			timing  = 0x00921250;break;
+		case XFER_UDMA_1:
+			timing  = 0x00911140;break;
+		case XFER_UDMA_2:
+			timing  = 0x00911030;break;
+		case XFER_MW_DMA_0:
+			timing  = 0x00077771;break;
+		case XFER_MW_DMA_1:
+			timing  = 0x00012121;break;
+		case XFER_MW_DMA_2:
+			timing  = 0x00002020;break;
+		default:
+			BUG();
+	}
+	/* Merge in the PIO format bit */
+	timing |= (tuning & 0x80000000UL);
+	if (adev->devno == 0) /* Master */
+		outl(timing, base + 0x04);
+	else {
+		if (timing & 0x00100000)
+			tuning |= 0x00100000;	/* UDMA for both */
+		else
+			tuning &= ~0x00100000;	/* MWDMA for both */
+		outl(tuning, base + 0x04);
+		outl(timing, base + 0x0C);
+	}
+
+	/* Set the DMA capable bit in the BMDMA area */
+	reg = inb(ap->ioaddr.bmdma_addr + ATA_DMA_STATUS);
+	reg |= (1 << (5 + adev->devno));
+	outb(reg, ap->ioaddr.bmdma_addr + ATA_DMA_STATUS);
+
+	/* Remember the last DMA setup we did */
+
+	ap->private_data = adev;
+}
+
+/**
+ *	cs5530_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings if
+ *	neccessary.  Specifically we have a problem that there is only
+ *	one MWDMA/UDMA bit.
+ */
+
+static unsigned int cs5530_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct ata_device *prev = ap->private_data;
+
+	/* See if the DMA settings could be wrong */
+	if (adev->dma_mode != 0 && adev != prev && prev != NULL) {
+		/* Maybe, but do the channels match MWDMA/UDMA ? */
+		if ((adev->dma_mode >= XFER_UDMA_0 && prev->dma_mode < XFER_UDMA_0) ||
+		    (adev->dma_mode < XFER_UDMA_0 && prev->dma_mode >= XFER_UDMA_0))
+		    	/* Switch the mode bits */
+		    	cs5530_set_dmamode(ap, adev);
+	}
+
+	return ata_qc_issue_prot(qc);
+}
+
+static int cs5530_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+static void cs5530_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, cs5530_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+
+static struct scsi_host_template cs5530_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations cs5530_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= cs5530_set_piomode,
+	.set_dmamode	= cs5530_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= cs5530_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= cs5530_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct dmi_system_id palmax_dmi_table[] = {
+	{
+		.ident = "Palmax PD1100",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Cyrix"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Caddis"),
+		},
+	},
+	{ }
+};
+
+static int cs5530_is_palmax(void)
+{
+	if (dmi_check_system(palmax_dmi_table)) {
+		printk(KERN_INFO "Palmax PD1100: Disabling DMA on docking port.\n");
+		return 1;
+	}
+	return 0;
+}
+
+/**
+ *	cs5530_init_one		-	Initialise a CS5530
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Install a driver for the newly found CS5530 companion chip. Most of
+ *	this is just housekeeping. We have to set the chip up correctly and
+ *	turn off various bits of emulation magic.
+ */
+
+static int cs5530_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	int compiler_warning_pointless_fix;
+	struct pci_dev *master_0 = NULL, *cs5530_0 = NULL;
+	static struct ata_port_info info = {
+		.sht = &cs5530_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x07,
+		.port_ops = &cs5530_port_ops
+	};
+	/* The docking connector doesn't do UDMA, and it seems not MWDMA */
+	static struct ata_port_info info_palmax_secondary = {
+		.sht = &cs5530_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.port_ops = &cs5530_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+
+	dev = NULL;
+	while ((dev = pci_get_device(PCI_VENDOR_ID_CYRIX, PCI_ANY_ID, dev)) != NULL) {
+		switch (dev->device) {
+			case PCI_DEVICE_ID_CYRIX_PCI_MASTER:
+				master_0 = pci_dev_get(dev);
+				break;
+			case PCI_DEVICE_ID_CYRIX_5530_LEGACY:
+				cs5530_0 = pci_dev_get(dev);
+				break;
+		}
+	}
+	if (!master_0) {
+		printk(KERN_ERR DRV_NAME ": unable to locate PCI MASTER function\n");
+		goto fail_put;
+	}
+	if (!cs5530_0) {
+		printk(KERN_ERR DRV_NAME ": unable to locate CS5530 LEGACY function\n");
+		goto fail_put;
+	}
+
+	pci_set_master(cs5530_0);
+	compiler_warning_pointless_fix = pci_set_mwi(cs5530_0);
+
+	/*
+	 * Set PCI CacheLineSize to 16-bytes:
+	 * --> Write 0x04 into 8-bit PCI CACHELINESIZE reg of function 0 of the cs5530
+	 *
+	 * Note: This value is constant because the 5530 is only a Geode companion
+	 */
+
+	pci_write_config_byte(cs5530_0, PCI_CACHE_LINE_SIZE, 0x04);
+
+	/*
+	 * Disable trapping of UDMA register accesses (Win98 hack):
+	 * --> Write 0x5006 into 16-bit reg at offset 0xd0 of function 0 of the cs5530
+	 */
+
+	pci_write_config_word(cs5530_0, 0xd0, 0x5006);
+
+	/*
+	 * Bit-1 at 0x40 enables MemoryWriteAndInvalidate on internal X-bus:
+	 * The other settings are what is necessary to get the register
+	 * into a sane state for IDE DMA operation.
+	 */
+
+	pci_write_config_byte(master_0, 0x40, 0x1e);
+
+	/*
+	 * Set max PCI burst size (16-bytes seems to work best):
+	 *	   16bytes: set bit-1 at 0x41 (reg value of 0x16)
+	 *	all others: clear bit-1 at 0x41, and do:
+	 *	  128bytes: OR 0x00 at 0x41
+	 *	  256bytes: OR 0x04 at 0x41
+	 *	  512bytes: OR 0x08 at 0x41
+	 *	 1024bytes: OR 0x0c at 0x41
+	 */
+
+	pci_write_config_byte(master_0, 0x41, 0x14);
+
+	/*
+	 * These settings are necessary to get the chip
+	 * into a sane state for IDE DMA operation.
+	 */
+
+	pci_write_config_byte(master_0, 0x42, 0x00);
+	pci_write_config_byte(master_0, 0x43, 0xc1);
+
+	pci_dev_put(master_0);
+	pci_dev_put(cs5530_0);
+
+	if (cs5530_is_palmax())
+		port_info[1] = &info_palmax_secondary;
+
+	/* Now kick off ATA set up */
+	return ata_pci_init_one(dev, port_info, 2);
+
+fail_put:
+	if (master_0)
+		pci_dev_put(master_0);
+	if (cs5530_0)
+		pci_dev_put(cs5530_0);
+	return -ENODEV;
+}
+
+static struct pci_device_id cs5530[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_IDE), },
+	{ 0, },
+};
+
+static struct pci_driver cs5530_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= cs5530,
+	.probe 		= cs5530_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init cs5530_init(void)
+{
+	return pci_register_driver(&cs5530_pci_driver);
+}
+
+
+static void __exit cs5530_exit(void)
+{
+	pci_unregister_driver(&cs5530_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the Cyrix/NS/AMD 5530");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, cs5530);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(cs5530_init);
+module_exit(cs5530_exit);
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
new file mode 100644
index 000000000000..69d6b4258724
--- /dev/null
+++ b/drivers/ata/pata_cs5535.c
@@ -0,0 +1,291 @@
+/*
+ * pata-cs5535.c 	- CS5535 PATA for new ATA layer
+ *			  (C) 2005-2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based upon cs5535.c from AMD <Jens.Altmann@amd.com> as cleaned up and
+ * made readable and Linux style by Wolfgang Zuleger <wolfgang.zuleger@gmx.de
+ * and Alexander Kiausch <alex.kiausch@t-online.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Loosely based on the piix & svwks drivers.
+ *
+ * Documentation:
+ *	Available from AMD web site.
+ * TODO
+ *	Review errata to see if serializing is neccessary
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <asm/msr.h>
+
+#define DRV_NAME	"cs5535"
+#define DRV_VERSION	"0.2.10"
+
+/*
+ *	The Geode (Aka Athlon GX now) uses an internal MSR based
+ *	bus system for control. Demented but there you go.
+ */
+
+#define MSR_ATAC_BASE    	0x51300000
+#define ATAC_GLD_MSR_CAP 	(MSR_ATAC_BASE+0)
+#define ATAC_GLD_MSR_CONFIG    (MSR_ATAC_BASE+0x01)
+#define ATAC_GLD_MSR_SMI       (MSR_ATAC_BASE+0x02)
+#define ATAC_GLD_MSR_ERROR     (MSR_ATAC_BASE+0x03)
+#define ATAC_GLD_MSR_PM        (MSR_ATAC_BASE+0x04)
+#define ATAC_GLD_MSR_DIAG      (MSR_ATAC_BASE+0x05)
+#define ATAC_IO_BAR            (MSR_ATAC_BASE+0x08)
+#define ATAC_RESET             (MSR_ATAC_BASE+0x10)
+#define ATAC_CH0D0_PIO         (MSR_ATAC_BASE+0x20)
+#define ATAC_CH0D0_DMA         (MSR_ATAC_BASE+0x21)
+#define ATAC_CH0D1_PIO         (MSR_ATAC_BASE+0x22)
+#define ATAC_CH0D1_DMA         (MSR_ATAC_BASE+0x23)
+#define ATAC_PCI_ABRTERR       (MSR_ATAC_BASE+0x24)
+
+#define ATAC_BM0_CMD_PRIM      0x00
+#define ATAC_BM0_STS_PRIM      0x02
+#define ATAC_BM0_PRD           0x04
+
+#define CS5535_CABLE_DETECT    0x48
+
+#define CS5535_BAD_PIO(timings) ( (timings&~0x80000000UL)==0x00009172 )
+
+/**
+ *	cs5535_pre_reset	-	detect cable type
+ *	@ap: Port to detect on
+ *
+ *	Perform cable detection for ATA66 capable cable. Return a libata
+ *	cable type.
+ */
+
+static int cs5535_pre_reset(struct ata_port *ap)
+{
+	u8 cable;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+
+	pci_read_config_byte(pdev, CS5535_CABLE_DETECT, &cable);
+	if (cable & 1)
+		ap->cbl = ATA_CBL_PATA80;
+	else
+		ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	cs5535_error_handler		-	reset/probe
+ *	@ap: Port to reset
+ *
+ *	Reset and configure a port
+ */
+
+static void cs5535_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, cs5535_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	cs5535_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Set our PIO requirements. The CS5535 is pretty clean about all this
+ */
+
+static void cs5535_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u16 pio_timings[5] = {
+		0xF7F4, 0x53F3, 0x13F1, 0x5131, 0x1131
+	};
+	static const u16 pio_cmd_timings[5] = {
+		0xF7F4, 0x53F3, 0x13F1, 0x5131, 0x1131
+	};
+	u32 reg, dummy;
+	struct ata_device *pair = ata_dev_pair(adev);
+
+	int mode = adev->pio_mode - XFER_PIO_0;
+	int cmdmode = mode;
+
+	/* Command timing has to be for the lowest of the pair of devices */
+	if (pair) {
+		int pairmode = pair->pio_mode - XFER_PIO_0;
+		cmdmode = min(mode, pairmode);
+		/* Write the other drive timing register if it changed */
+		if (cmdmode < pairmode)
+			wrmsr(ATAC_CH0D0_PIO + 2 * pair->devno,
+				pio_cmd_timings[cmdmode] << 16 | pio_timings[pairmode], 0);
+	}
+	/* Write the drive timing register */
+	wrmsr(ATAC_CH0D0_PIO + 2 * adev->devno,
+		pio_cmd_timings[cmdmode] << 16 | pio_timings[mode], 0);
+
+	/* Set the PIO "format 1" bit in the DMA timing register */
+	rdmsr(ATAC_CH0D0_DMA + 2 * adev->devno, reg, dummy);
+	wrmsr(ATAC_CH0D0_DMA + 2 * adev->devno, reg | 0x80000000UL, 0);
+}
+
+/**
+ *	cs5535_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ */
+
+static void cs5535_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u32 udma_timings[5] = {
+		0x7F7436A1, 0x7F733481, 0x7F723261, 0x7F713161, 0x7F703061
+	};
+	static const u32 mwdma_timings[3] = {
+		0x7F0FFFF3, 0x7F035352, 0x7F024241
+	};
+	u32 reg, dummy;
+	int mode = adev->dma_mode;
+
+	rdmsr(ATAC_CH0D0_DMA + 2 * adev->devno, reg, dummy);
+	reg &= 0x80000000UL;
+	if (mode >= XFER_UDMA_0)
+		reg |= udma_timings[mode - XFER_UDMA_0];
+	else
+		reg |= mwdma_timings[mode - XFER_MW_DMA_0];
+	wrmsr(ATAC_CH0D0_DMA + 2 * adev->devno, reg, 0);
+}
+
+static struct scsi_host_template cs5535_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations cs5535_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= cs5535_set_piomode,
+	.set_dmamode	= cs5535_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= cs5535_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	cs5535_init_one		-	Initialise a CS5530
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Install a driver for the newly found CS5530 companion chip. Most of
+ *	this is just housekeeping. We have to set the chip up correctly and
+ *	turn off various bits of emulation magic.
+ */
+
+static int cs5535_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &cs5535_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x1f,
+		.port_ops = &cs5535_port_ops
+	};
+	struct ata_port_info *ports[1] = { &info };
+
+	u32 timings, dummy;
+
+	/* Check the BIOS set the initial timing clock. If not set the
+	   timings for PIO0 */
+	rdmsr(ATAC_CH0D0_PIO, timings, dummy);
+	if (CS5535_BAD_PIO(timings))
+		wrmsr(ATAC_CH0D0_PIO, 0xF7F4F7F4UL, 0);
+	rdmsr(ATAC_CH0D1_PIO, timings, dummy);
+	if (CS5535_BAD_PIO(timings))
+		wrmsr(ATAC_CH0D1_PIO, 0xF7F4F7F4UL, 0);
+	return ata_pci_init_one(dev, ports, 1);
+}
+
+static struct pci_device_id cs5535[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_NS, 0x002D), },
+	{ 0, },
+};
+
+static struct pci_driver cs5535_pci_driver = {
+	.name		= DRV_NAME,
+	.id_table	= cs5535,
+	.probe 		= cs5535_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init cs5535_init(void)
+{
+	return pci_register_driver(&cs5535_pci_driver);
+}
+
+
+static void __exit cs5535_exit(void)
+{
+	pci_unregister_driver(&cs5535_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox, Jens Altmann, Wolfgan Zuleger, Alexander Kiausch");
+MODULE_DESCRIPTION("low-level driver for the NS/AMD 5530");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, cs5535);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(cs5535_init);
+module_exit(cs5535_exit);
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
new file mode 100644
index 000000000000..322043d8cf7f
--- /dev/null
+++ b/drivers/ata/pata_cypress.c
@@ -0,0 +1,227 @@
+/*
+ * pata_cypress.c 	- Cypress PATA for new ATA layer
+ *			  (C) 2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based heavily on
+ * linux/drivers/ide/pci/cy82c693.c		Version 0.40	Sep. 10, 2002
+ *
+ */
+ 
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_cypress"
+#define DRV_VERSION "0.1.2"
+
+/* here are the offset definitions for the registers */
+
+enum {
+	CY82_IDE_CMDREG		= 0x04,
+	CY82_IDE_ADDRSETUP	= 0x48,
+	CY82_IDE_MASTER_IOR	= 0x4C,
+	CY82_IDE_MASTER_IOW	= 0x4D,
+	CY82_IDE_SLAVE_IOR	= 0x4E,
+	CY82_IDE_SLAVE_IOW	= 0x4F,
+	CY82_IDE_MASTER_8BIT	= 0x50,
+	CY82_IDE_SLAVE_8BIT	= 0x51,
+
+	CY82_INDEX_PORT		= 0x22,
+	CY82_DATA_PORT		= 0x23,
+
+	CY82_INDEX_CTRLREG1	= 0x01,
+	CY82_INDEX_CHANNEL0	= 0x30,
+	CY82_INDEX_CHANNEL1	= 0x31,
+	CY82_INDEX_TIMEOUT	= 0x32
+};
+
+static int cy82c693_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+static void cy82c693_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, cy82c693_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	cy82c693_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup.
+ */
+ 
+static void cy82c693_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct ata_timing t;
+	const unsigned long T = 1000000 / 33;
+	short time_16, time_8;
+	u32 addr;
+	
+	if (ata_timing_compute(adev, adev->pio_mode, &t, T, 1) < 0) {
+		printk(KERN_ERR DRV_NAME ": mome computation failed.\n");
+		return;
+	}
+
+	time_16 = FIT(t.recover, 0, 15) | (FIT(t.active, 0, 15) << 4);
+	time_8 = FIT(t.act8b, 0, 15) | (FIT(t.rec8b, 0, 15) << 4);
+	
+	if (adev->devno == 0) {
+		pci_read_config_dword(pdev, CY82_IDE_ADDRSETUP, &addr);
+		
+		addr &= ~0x0F;	/* Mask bits */
+		addr |= FIT(t.setup, 0, 15);
+		
+		pci_write_config_dword(pdev, CY82_IDE_ADDRSETUP, addr);
+		pci_write_config_byte(pdev, CY82_IDE_MASTER_IOR, time_16);
+		pci_write_config_byte(pdev, CY82_IDE_MASTER_IOW, time_16);
+		pci_write_config_byte(pdev, CY82_IDE_MASTER_8BIT, time_8);
+	} else {
+		pci_read_config_dword(pdev, CY82_IDE_ADDRSETUP, &addr);
+		
+		addr &= ~0xF0;	/* Mask bits */
+		addr |= (FIT(t.setup, 0, 15) << 4);
+
+		pci_write_config_dword(pdev, CY82_IDE_ADDRSETUP, addr);
+		pci_write_config_byte(pdev, CY82_IDE_SLAVE_IOR, time_16);
+		pci_write_config_byte(pdev, CY82_IDE_SLAVE_IOW, time_16);
+		pci_write_config_byte(pdev, CY82_IDE_SLAVE_8BIT, time_8);
+	}
+}
+
+/**
+ *	cy82c693_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the DMA mode setup.
+ */
+ 
+static void cy82c693_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	int reg = CY82_INDEX_CHANNEL0 + ap->port_no;
+	
+	/* Be afraid, be very afraid. Magic registers  in low I/O space */
+	outb(reg, 0x22);
+	outb(adev->dma_mode - XFER_MW_DMA_0, 0x23);
+	
+	/* 0x50 gives the best behaviour on the Alpha's using this chip */
+	outb(CY82_INDEX_TIMEOUT, 0x22);
+	outb(0x50, 0x23);
+}
+
+static struct scsi_host_template cy82c693_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations cy82c693_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= cy82c693_set_piomode,
+	.set_dmamode	= cy82c693_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= cy82c693_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static int cy82c693_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &cy82c693_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &cy82c693_port_ops
+	};
+	static struct ata_port_info *port_info[1] = { &info };
+	
+	/* Devfn 1 is the ATA primary. The secondary is magic and on devfn2. For the
+	   moment we don't handle the secondary. FIXME */
+	   
+	if (PCI_FUNC(pdev->devfn) != 1)
+		return -ENODEV;
+	
+	return ata_pci_init_one(pdev, port_info, 1);
+}
+
+static struct pci_device_id cy82c693[] = {
+	{ PCI_VENDOR_ID_CONTAQ, PCI_DEVICE_ID_CONTAQ_82C693, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{ 0, },
+};
+
+static struct pci_driver cy82c693_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= cy82c693,
+	.probe 		= cy82c693_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init cy82c693_init(void)
+{
+	return pci_register_driver(&cy82c693_pci_driver);
+}
+
+
+static void __exit cy82c693_exit(void)
+{
+	pci_unregister_driver(&cy82c693_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the CY82C693 PATA controller");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, cy82c693);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(cy82c693_init);
+module_exit(cy82c693_exit);
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
new file mode 100644
index 000000000000..c30bc181304f
--- /dev/null
+++ b/drivers/ata/pata_efar.c
@@ -0,0 +1,342 @@
+/*
+ *    pata_efar.c - EFAR PIIX clone controller driver
+ *
+ *	(C) 2005 Red Hat <alan@redhat.com>
+ *
+ *    Some parts based on ata_piix.c by Jeff Garzik and others.
+ *
+ *    The EFAR is a PIIX4 clone with UDMA66 support. Unlike the later
+ *    Intel ICH controllers the EFAR widened the UDMA mode register bits
+ *    and doesn't require the funky clock selection.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_efar"
+#define DRV_VERSION	"0.4.1"
+
+/**
+ *	efar_pre_reset	-	check for 40/80 pin
+ *	@ap: Port
+ *
+ *	Perform cable detection for the EFAR ATA interface. This is
+ *	different to the PIIX arrangement
+ */
+
+static int efar_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits efar_enable_bits[] = {
+		{ 0x41U, 1U, 0x80UL, 0x80UL },	/* port 0 */
+		{ 0x43U, 1U, 0x80UL, 0x80UL },	/* port 1 */
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 tmp;
+
+	if (!pci_test_config_bits(pdev, &efar_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	pci_read_config_byte(pdev, 0x47, &tmp);
+	if (tmp & (2 >> ap->port_no))
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	efar_probe_reset - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void efar_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, efar_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	efar_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: um
+ *
+ *	Set PIO mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void efar_set_piomode (struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int pio	= adev->pio_mode - XFER_PIO_0;
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
+	unsigned int idetm_port= ap->port_no ? 0x42 : 0x40;
+	u16 idetm_data;
+	int control = 0;
+
+	/*
+	 *	See Intel Document 298600-004 for the timing programing rules
+	 *	for PIIX/ICH. The EFAR is a clone so very similar
+	 */
+
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 0 },
+			    { 2, 1 },
+			    { 2, 3 }, };
+
+	if (pio > 2)
+		control |= 1;	/* TIME1 enable */
+	if (ata_pio_need_iordy(adev))	/* PIO 3/4 require IORDY */
+		control |= 2;	/* IE enable */
+	/* Intel specifies that the PPE functionality is for disk only */
+	if (adev->class == ATA_DEV_ATA)
+		control |= 4;	/* PPE enable */
+
+	pci_read_config_word(dev, idetm_port, &idetm_data);
+
+	/* Enable PPE, IE and TIME as appropriate */
+
+	if (adev->devno == 0) {
+		idetm_data &= 0xCCF0;
+		idetm_data |= control;
+		idetm_data |= (timings[pio][0] << 12) |
+			(timings[pio][1] << 8);
+	} else {
+		int shift = 4 * ap->port_no;
+		u8 slave_data;
+
+		idetm_data &= 0xCC0F;
+		idetm_data |= (control << 4);
+
+		/* Slave timing in seperate register */
+		pci_read_config_byte(dev, 0x44, &slave_data);
+		slave_data &= 0x0F << shift;
+		slave_data |= ((timings[pio][0] << 2) | timings[pio][1]) << shift;
+		pci_write_config_byte(dev, 0x44, slave_data);
+	}
+
+	idetm_data |= 0x4000;	/* Ensure SITRE is enabled */
+	pci_write_config_word(dev, idetm_port, idetm_data);
+}
+
+/**
+ *	efar_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *
+ *	Set UDMA/MWDMA mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void efar_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
+	u8 master_port		= ap->port_no ? 0x42 : 0x40;
+	u16 master_data;
+	u8 speed		= adev->dma_mode;
+	int devid		= adev->devno + 2 * ap->port_no;
+	u8 udma_enable;
+
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 0 },
+			    { 2, 1 },
+			    { 2, 3 }, };
+
+	pci_read_config_word(dev, master_port, &master_data);
+	pci_read_config_byte(dev, 0x48, &udma_enable);
+
+	if (speed >= XFER_UDMA_0) {
+		unsigned int udma	= adev->dma_mode - XFER_UDMA_0;
+		u16 udma_timing;
+
+		udma_enable |= (1 << devid);
+
+		/* Load the UDMA mode number */
+		pci_read_config_word(dev, 0x4A, &udma_timing);
+		udma_timing &= ~(7 << (4 * devid));
+		udma_timing |= udma << (4 * devid);
+		pci_write_config_word(dev, 0x4A, udma_timing);
+	} else {
+		/*
+		 * MWDMA is driven by the PIO timings. We must also enable
+		 * IORDY unconditionally along with TIME1. PPE has already
+		 * been set when the PIO timing was set.
+		 */
+		unsigned int mwdma	= adev->dma_mode - XFER_MW_DMA_0;
+		unsigned int control;
+		u8 slave_data;
+		const unsigned int needed_pio[3] = {
+			XFER_PIO_0, XFER_PIO_3, XFER_PIO_4
+		};
+		int pio = needed_pio[mwdma] - XFER_PIO_0;
+
+		control = 3;	/* IORDY|TIME1 */
+
+		/* If the drive MWDMA is faster than it can do PIO then
+		   we must force PIO into PIO0 */
+
+		if (adev->pio_mode < needed_pio[mwdma])
+			/* Enable DMA timing only */
+			control |= 8;	/* PIO cycles in PIO0 */
+
+		if (adev->devno) {	/* Slave */
+			master_data &= 0xFF4F;  /* Mask out IORDY|TIME1|DMAONLY */
+			master_data |= control << 4;
+			pci_read_config_byte(dev, 0x44, &slave_data);
+			slave_data &= (0x0F + 0xE1 * ap->port_no);
+			/* Load the matching timing */
+			slave_data |= ((timings[pio][0] << 2) | timings[pio][1]) << (ap->port_no ? 4 : 0);
+			pci_write_config_byte(dev, 0x44, slave_data);
+		} else { 	/* Master */
+			master_data &= 0xCCF4;	/* Mask out IORDY|TIME1|DMAONLY
+						   and master timing bits */
+			master_data |= control;
+			master_data |=
+				(timings[pio][0] << 12) |
+				(timings[pio][1] << 8);
+		}
+		udma_enable &= ~(1 << devid);
+		pci_write_config_word(dev, master_port, master_data);
+	}
+	pci_write_config_byte(dev, 0x48, udma_enable);
+}
+
+static struct scsi_host_template efar_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations efar_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= efar_set_piomode,
+	.set_dmamode		= efar_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= efar_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+
+/**
+ *	efar_init_one - Register EFAR ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in efar_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int efar_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	static struct ata_port_info info = {
+		.sht		= &efar_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma1-2 */
+		.udma_mask 	= 0x0f, /* UDMA 66 */
+		.port_ops	= &efar_ops,
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev,
+			   "version " DRV_VERSION "\n");
+
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id efar_pci_tbl[] = {
+	{ 0x1055, 0x9130, PCI_ANY_ID, PCI_ANY_ID, },
+	{ }	/* terminate list */
+};
+
+static struct pci_driver efar_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= efar_pci_tbl,
+	.probe			= efar_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init efar_init(void)
+{
+	return pci_register_driver(&efar_pci_driver);
+}
+
+static void __exit efar_exit(void)
+{
+	pci_unregister_driver(&efar_pci_driver);
+}
+
+
+module_init(efar_init);
+module_exit(efar_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for EFAR PIIX clones");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, efar_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
new file mode 100644
index 000000000000..e54a0c51681d
--- /dev/null
+++ b/drivers/ata/pata_hpt366.c
@@ -0,0 +1,478 @@
+/*
+ * Libata driver for the highpoint 366 and 368 UDMA66 ATA controllers.
+ *
+ * This driver is heavily based upon:
+ *
+ * linux/drivers/ide/pci/hpt366.c		Version 0.36	April 25, 2003
+ *
+ * Copyright (C) 1999-2003		Andre Hedrick <andre@linux-ide.org>
+ * Portions Copyright (C) 2001	        Sun Microsystems, Inc.
+ * Portions Copyright (C) 2003		Red Hat Inc
+ *
+ *
+ * TODO
+ *	Maybe PLL mode
+ *	Look into engine reset on timeout errors. Should not be
+ *		required.
+ */
+
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"pata_hpt366"
+#define DRV_VERSION	"0.5"
+
+struct hpt_clock {
+	u8	xfer_speed;
+	u32	timing;
+};
+
+/* key for bus clock timings
+ * bit
+ * 0:3    data_high_time. inactive time of DIOW_/DIOR_ for PIO and MW
+ *        DMA. cycles = value + 1
+ * 4:8    data_low_time. active time of DIOW_/DIOR_ for PIO and MW
+ *        DMA. cycles = value + 1
+ * 9:12   cmd_high_time. inactive time of DIOW_/DIOR_ during task file
+ *        register access.
+ * 13:17  cmd_low_time. active time of DIOW_/DIOR_ during task file
+ *        register access.
+ * 18:21  udma_cycle_time. clock freq and clock cycles for UDMA xfer.
+ *        during task file register access.
+ * 22:24  pre_high_time. time to initialize 1st cycle for PIO and MW DMA
+ *        xfer.
+ * 25:27  cmd_pre_high_time. time to initialize 1st PIO cycle for task
+ *        register access.
+ * 28     UDMA enable
+ * 29     DMA enable
+ * 30     PIO_MST enable. if set, the chip is in bus master mode during
+ *        PIO.
+ * 31     FIFO enable.
+ */
+
+static const struct hpt_clock hpt366_40[] = {
+	{	XFER_UDMA_4,	0x900fd943	},
+	{	XFER_UDMA_3,	0x900ad943	},
+	{	XFER_UDMA_2,	0x900bd943	},
+	{	XFER_UDMA_1,	0x9008d943	},
+	{	XFER_UDMA_0,	0x9008d943	},
+
+	{	XFER_MW_DMA_2,	0xa008d943	},
+	{	XFER_MW_DMA_1,	0xa010d955	},
+	{	XFER_MW_DMA_0,	0xa010d9fc	},
+
+	{	XFER_PIO_4,	0xc008d963	},
+	{	XFER_PIO_3,	0xc010d974	},
+	{	XFER_PIO_2,	0xc010d997	},
+	{	XFER_PIO_1,	0xc010d9c7	},
+	{	XFER_PIO_0,	0xc018d9d9	},
+	{	0,		0x0120d9d9	}
+};
+
+static const struct hpt_clock hpt366_33[] = {
+	{	XFER_UDMA_4,	0x90c9a731	},
+	{	XFER_UDMA_3,	0x90cfa731	},
+	{	XFER_UDMA_2,	0x90caa731	},
+	{	XFER_UDMA_1,	0x90cba731	},
+	{	XFER_UDMA_0,	0x90c8a731	},
+
+	{	XFER_MW_DMA_2,	0xa0c8a731	},
+	{	XFER_MW_DMA_1,	0xa0c8a732	},	/* 0xa0c8a733 */
+	{	XFER_MW_DMA_0,	0xa0c8a797	},
+
+	{	XFER_PIO_4,	0xc0c8a731	},
+	{	XFER_PIO_3,	0xc0c8a742	},
+	{	XFER_PIO_2,	0xc0d0a753	},
+	{	XFER_PIO_1,	0xc0d0a7a3	},	/* 0xc0d0a793 */
+	{	XFER_PIO_0,	0xc0d0a7aa	},	/* 0xc0d0a7a7 */
+	{	0,		0x0120a7a7	}
+};
+
+static const struct hpt_clock hpt366_25[] = {
+	{	XFER_UDMA_4,	0x90c98521	},
+	{	XFER_UDMA_3,	0x90cf8521	},
+	{	XFER_UDMA_2,	0x90cf8521	},
+	{	XFER_UDMA_1,	0x90cb8521	},
+	{	XFER_UDMA_0,	0x90cb8521	},
+
+	{	XFER_MW_DMA_2,	0xa0ca8521	},
+	{	XFER_MW_DMA_1,	0xa0ca8532	},
+	{	XFER_MW_DMA_0,	0xa0ca8575	},
+
+	{	XFER_PIO_4,	0xc0ca8521	},
+	{	XFER_PIO_3,	0xc0ca8532	},
+	{	XFER_PIO_2,	0xc0ca8542	},
+	{	XFER_PIO_1,	0xc0d08572	},
+	{	XFER_PIO_0,	0xc0d08585	},
+	{	0,		0x01208585	}
+};
+
+static const char *bad_ata33[] = {
+	"Maxtor 92720U8", "Maxtor 92040U6", "Maxtor 91360U4", "Maxtor 91020U3", "Maxtor 90845U3", "Maxtor 90650U2",
+	"Maxtor 91360D8", "Maxtor 91190D7", "Maxtor 91020D6", "Maxtor 90845D5", "Maxtor 90680D4", "Maxtor 90510D3", "Maxtor 90340D2",
+	"Maxtor 91152D8", "Maxtor 91008D7", "Maxtor 90845D6", "Maxtor 90840D6", "Maxtor 90720D5", "Maxtor 90648D5", "Maxtor 90576D4",
+	"Maxtor 90510D4",
+	"Maxtor 90432D3", "Maxtor 90288D2", "Maxtor 90256D2",
+	"Maxtor 91000D8", "Maxtor 90910D8", "Maxtor 90875D7", "Maxtor 90840D7", "Maxtor 90750D6", "Maxtor 90625D5", "Maxtor 90500D4",
+	"Maxtor 91728D8", "Maxtor 91512D7", "Maxtor 91303D6", "Maxtor 91080D5", "Maxtor 90845D4", "Maxtor 90680D4", "Maxtor 90648D3", "Maxtor 90432D2",
+	NULL
+};
+
+static const char *bad_ata66_4[] = {
+	"IBM-DTLA-307075",
+	"IBM-DTLA-307060",
+	"IBM-DTLA-307045",
+	"IBM-DTLA-307030",
+	"IBM-DTLA-307020",
+	"IBM-DTLA-307015",
+	"IBM-DTLA-305040",
+	"IBM-DTLA-305030",
+	"IBM-DTLA-305020",
+	"IC35L010AVER07-0",
+	"IC35L020AVER07-0",
+	"IC35L030AVER07-0",
+	"IC35L040AVER07-0",
+	"IC35L060AVER07-0",
+	"WDC AC310200R",
+	NULL
+};
+
+static const char *bad_ata66_3[] = {
+	"WDC AC310200R",
+	NULL
+};
+
+static int hpt_dma_blacklisted(const struct ata_device *dev, char *modestr, const char *list[])
+{
+	unsigned char model_num[40];
+	char *s;
+	unsigned int len;
+	int i = 0;
+
+	ata_id_string(dev->id, model_num, ATA_ID_PROD_OFS, sizeof(model_num));
+	s = &model_num[0];
+	len = strnlen(s, sizeof(model_num));
+
+	/* ATAPI specifies that empty space is blank-filled; remove blanks */
+	while ((len > 0) && (s[len - 1] == ' ')) {
+		len--;
+		s[len] = 0;
+	}
+
+	while(list[i] != NULL) {
+		if (!strncmp(list[i], s, len)) {
+			printk(KERN_WARNING DRV_NAME ": %s is not supported for %s.\n", 
+				modestr, list[i]);
+			return 1;
+		}
+		i++;
+	}
+	return 0;
+}
+
+/**
+ *	hpt366_filter	-	mode selection filter
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Block UDMA on devices that cause trouble with this controller.
+ */
+ 
+static unsigned long hpt366_filter(const struct ata_port *ap, struct ata_device *adev, unsigned long mask)
+{
+	if (adev->class == ATA_DEV_ATA) {
+		if (hpt_dma_blacklisted(adev, "UDMA",  bad_ata33))
+			mask &= ~ATA_MASK_UDMA;
+		if (hpt_dma_blacklisted(adev, "UDMA3", bad_ata66_3))
+			mask &= ~(0x07 << ATA_SHIFT_UDMA);
+		if (hpt_dma_blacklisted(adev, "UDMA4", bad_ata66_4))
+			mask &= ~(0x0F << ATA_SHIFT_UDMA);
+	}
+	return ata_pci_default_filter(ap, adev, mask);
+}
+
+/**
+ *	hpt36x_find_mode	-	reset the hpt36x bus
+ *	@ap: ATA port
+ *	@speed: transfer mode
+ *
+ *	Return the 32bit register programming information for this channel
+ *	that matches the speed provided.
+ */
+ 
+static u32 hpt36x_find_mode(struct ata_port *ap, int speed)
+{
+	struct hpt_clock *clocks = ap->host->private_data;
+	
+	while(clocks->xfer_speed) {
+		if (clocks->xfer_speed == speed)
+			return clocks->timing;
+		clocks++;
+	}
+	BUG();
+	return 0xffffffffU;	/* silence compiler warning */
+}
+	
+static int hpt36x_pre_reset(struct ata_port *ap)
+{
+	u8 ata66;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	pci_read_config_byte(pdev, 0x5A, &ata66);
+	if (ata66 & (1 << ap->port_no))
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	hpt36x_error_handler	-	reset the hpt36x bus
+ *	@ap: ATA port to reset
+ *
+ *	Perform the reset handling for the 366/368
+ */
+ 
+static void hpt36x_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, hpt36x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	hpt366_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Perform PIO mode setup. 
+ */
+ 
+static void hpt366_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	if (fast & 0x80) {
+		fast &= ~0x80;
+		pci_write_config_byte(pdev, addr2, fast);
+	}
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt36x_find_mode(ap, adev->pio_mode);
+	mode &= ~0x8000000;	/* No FIFO in PIO */
+	mode &= ~0x30070000;	/* Leave config bits alone */
+	reg &= 0x30070000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt366_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	Set up the channel for MWDMA or UDMA modes. Much the same as with
+ *	PIO, load the mode number and then set MWDMA or UDMA flag.
+ */
+ 
+static void hpt366_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	if (fast & 0x80) {
+		fast &= ~0x80;
+		pci_write_config_byte(pdev, addr2, fast);
+	}
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt36x_find_mode(ap, adev->dma_mode);
+	mode |= 0x8000000;	/* FIFO in MWDMA or UDMA */
+	mode &= ~0xC0000000;	/* Leave config bits alone */
+	reg &= 0xC0000000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+static struct scsi_host_template hpt36x_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+/*
+ *	Configuration for HPT366/68
+ */
+ 
+static struct ata_port_operations hpt366_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt366_set_piomode,
+	.set_dmamode	= hpt366_set_dmamode,
+	.mode_filter	= hpt366_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt36x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/**
+ *	hpt36x_init_one		-	Initialise an HPT366/368
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Initialise an HPT36x device. There are some interesting complications
+ *	here. Firstly the chip may report 366 and be one of several variants.
+ *	Secondly all the timings depend on the clock for the chip which we must
+ *	detect and look up
+ *
+ *	This is the known chip mappings. It may be missing a couple of later
+ *	releases.
+ *
+ *	Chip version		PCI		Rev	Notes
+ *	HPT366			4 (HPT366)	0	UDMA66
+ *	HPT366			4 (HPT366)	1	UDMA66
+ *	HPT368			4 (HPT366)	2	UDMA66
+ *	HPT37x/30x		4 (HPT366)	3+	Other driver
+ *
+ */
+ 
+static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info_hpt366 = {
+		.sht = &hpt36x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x1f,
+		.port_ops = &hpt366_port_ops
+	};
+	struct ata_port_info *port_info[2] = {&info_hpt366, &info_hpt366};
+
+	u32 class_rev;
+	u32 reg1;
+	u8 drive_fast;
+
+	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class_rev);
+	class_rev &= 0xFF;
+	
+	/* May be a later chip in disguise. Check */
+	/* Newer chips are not in the HPT36x driver. Ignore them */
+	if (class_rev > 2)
+			return -ENODEV;
+
+	pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
+	pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
+	pci_write_config_byte(dev, PCI_MIN_GNT, 0x08);
+	pci_write_config_byte(dev, PCI_MAX_LAT, 0x08);
+
+	pci_read_config_byte(dev, 0x51, &drive_fast);
+	if (drive_fast & 0x80)
+		pci_write_config_byte(dev, 0x51, drive_fast & ~0x80);
+
+	pci_read_config_dword(dev, 0x40,  &reg1);
+	
+	/* PCI clocking determines the ATA timing values to use */
+	/* info_hpt366 is safe against re-entry so we can scribble on it */
+	switch(reg1 & 0x700) {
+		case 5:
+			info_hpt366.private_data = &hpt366_40;
+			break;
+		case 9:
+			info_hpt366.private_data = &hpt366_25;
+			break;
+		default:
+			info_hpt366.private_data = &hpt366_33;
+			break;
+	}
+	/* Now kick off ATA set up */
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id hpt36x[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT366), },
+	{ 0, },
+};
+
+static struct pci_driver hpt36x_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= hpt36x,
+	.probe 		= hpt36x_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init hpt36x_init(void)
+{
+	return pci_register_driver(&hpt36x_pci_driver);
+}
+
+
+static void __exit hpt36x_exit(void)
+{
+	pci_unregister_driver(&hpt36x_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the Highpoint HPT366/368");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, hpt36x);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(hpt36x_init);
+module_exit(hpt36x_exit);
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
new file mode 100644
index 000000000000..7c3da53f1e0c
--- /dev/null
+++ b/drivers/ata/pata_hpt37x.c
@@ -0,0 +1,1257 @@
+/*
+ * Libata driver for the highpoint 37x and 30x UDMA66 ATA controllers.
+ *
+ * This driver is heavily based upon:
+ *
+ * linux/drivers/ide/pci/hpt366.c		Version 0.36	April 25, 2003
+ *
+ * Copyright (C) 1999-2003		Andre Hedrick <andre@linux-ide.org>
+ * Portions Copyright (C) 2001	        Sun Microsystems, Inc.
+ * Portions Copyright (C) 2003		Red Hat Inc
+ *
+ * TODO
+ *	PLL mode
+ *	Look into engine reset on timeout errors. Should not be
+ *		required.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"pata_hpt37x"
+#define DRV_VERSION	"0.5"
+
+struct hpt_clock {
+	u8	xfer_speed;
+	u32	timing;
+};
+
+struct hpt_chip {
+	const char *name;
+	unsigned int base;
+	struct hpt_clock const *clocks[4];
+};
+
+/* key for bus clock timings
+ * bit
+ * 0:3    data_high_time. inactive time of DIOW_/DIOR_ for PIO and MW
+ *        DMA. cycles = value + 1
+ * 4:8    data_low_time. active time of DIOW_/DIOR_ for PIO and MW
+ *        DMA. cycles = value + 1
+ * 9:12   cmd_high_time. inactive time of DIOW_/DIOR_ during task file
+ *        register access.
+ * 13:17  cmd_low_time. active time of DIOW_/DIOR_ during task file
+ *        register access.
+ * 18:21  udma_cycle_time. clock freq and clock cycles for UDMA xfer.
+ *        during task file register access.
+ * 22:24  pre_high_time. time to initialize 1st cycle for PIO and MW DMA
+ *        xfer.
+ * 25:27  cmd_pre_high_time. time to initialize 1st PIO cycle for task
+ *        register access.
+ * 28     UDMA enable
+ * 29     DMA enable
+ * 30     PIO_MST enable. if set, the chip is in bus master mode during
+ *        PIO.
+ * 31     FIFO enable.
+ */
+
+/* from highpoint documentation. these are old values */
+static const struct hpt_clock hpt370_timings_33[] = {
+/*	{	XFER_UDMA_5,	0x1A85F442,	0x16454e31	}, */
+	{	XFER_UDMA_5,	0x16454e31	},
+	{	XFER_UDMA_4,	0x16454e31	},
+	{	XFER_UDMA_3,	0x166d4e31	},
+	{	XFER_UDMA_2,	0x16494e31	},
+	{	XFER_UDMA_1,	0x164d4e31	},
+	{	XFER_UDMA_0,	0x16514e31	},
+
+	{	XFER_MW_DMA_2,	0x26514e21	},
+	{	XFER_MW_DMA_1,	0x26514e33	},
+	{	XFER_MW_DMA_0,	0x26514e97	},
+
+	{	XFER_PIO_4,	0x06514e21	},
+	{	XFER_PIO_3,	0x06514e22	},
+	{	XFER_PIO_2,	0x06514e33	},
+	{	XFER_PIO_1,	0x06914e43	},
+	{	XFER_PIO_0,	0x06914e57	},
+	{	0,		0x06514e57	}
+};
+
+static const struct hpt_clock hpt370_timings_66[] = {
+	{	XFER_UDMA_5,	0x14846231	},
+	{	XFER_UDMA_4,	0x14886231	},
+	{	XFER_UDMA_3,	0x148c6231	},
+	{	XFER_UDMA_2,	0x148c6231	},
+	{	XFER_UDMA_1,	0x14906231	},
+	{	XFER_UDMA_0,	0x14986231	},
+
+	{	XFER_MW_DMA_2,	0x26514e21	},
+	{	XFER_MW_DMA_1,	0x26514e33	},
+	{	XFER_MW_DMA_0,	0x26514e97	},
+
+	{	XFER_PIO_4,	0x06514e21	},
+	{	XFER_PIO_3,	0x06514e22	},
+	{	XFER_PIO_2,	0x06514e33	},
+	{	XFER_PIO_1,	0x06914e43	},
+	{	XFER_PIO_0,	0x06914e57	},
+	{	0,		0x06514e57	}
+};
+
+/* these are the current (4 sep 2001) timings from highpoint */
+static const struct hpt_clock hpt370a_timings_33[] = {
+	{	XFER_UDMA_5,	0x12446231	},
+	{	XFER_UDMA_4,	0x12446231	},
+	{	XFER_UDMA_3,	0x126c6231	},
+	{	XFER_UDMA_2,	0x12486231	},
+	{	XFER_UDMA_1,	0x124c6233	},
+	{	XFER_UDMA_0,	0x12506297	},
+
+	{	XFER_MW_DMA_2,	0x22406c31	},
+	{	XFER_MW_DMA_1,	0x22406c33	},
+	{	XFER_MW_DMA_0,	0x22406c97	},
+
+	{	XFER_PIO_4,	0x06414e31	},
+	{	XFER_PIO_3,	0x06414e42	},
+	{	XFER_PIO_2,	0x06414e53	},
+	{	XFER_PIO_1,	0x06814e93	},
+	{	XFER_PIO_0,	0x06814ea7	},
+	{	0,		0x06814ea7	}
+};
+
+/* 2x 33MHz timings */
+static const struct hpt_clock hpt370a_timings_66[] = {
+	{	XFER_UDMA_5,	0x1488e673	},
+	{	XFER_UDMA_4,	0x1488e673	},
+	{	XFER_UDMA_3,	0x1498e673	},
+	{	XFER_UDMA_2,	0x1490e673	},
+	{	XFER_UDMA_1,	0x1498e677	},
+	{	XFER_UDMA_0,	0x14a0e73f	},
+
+	{	XFER_MW_DMA_2,	0x2480fa73	},
+	{	XFER_MW_DMA_1,	0x2480fa77	}, 
+	{	XFER_MW_DMA_0,	0x2480fb3f	},
+
+	{	XFER_PIO_4,	0x0c82be73	},
+	{	XFER_PIO_3,	0x0c82be95	},
+	{	XFER_PIO_2,	0x0c82beb7	},
+	{	XFER_PIO_1,	0x0d02bf37	},
+	{	XFER_PIO_0,	0x0d02bf5f	},
+	{	0,		0x0d02bf5f	}
+};
+
+static const struct hpt_clock hpt370a_timings_50[] = {
+	{	XFER_UDMA_5,	0x12848242	},
+	{	XFER_UDMA_4,	0x12ac8242	},
+	{	XFER_UDMA_3,	0x128c8242	},
+	{	XFER_UDMA_2,	0x120c8242	},
+	{	XFER_UDMA_1,	0x12148254	},
+	{	XFER_UDMA_0,	0x121882ea	},
+
+	{	XFER_MW_DMA_2,	0x22808242	},
+	{	XFER_MW_DMA_1,	0x22808254	},
+	{	XFER_MW_DMA_0,	0x228082ea	},
+
+	{	XFER_PIO_4,	0x0a81f442	},
+	{	XFER_PIO_3,	0x0a81f443	},
+	{	XFER_PIO_2,	0x0a81f454	},
+	{	XFER_PIO_1,	0x0ac1f465	},
+	{	XFER_PIO_0,	0x0ac1f48a	},
+	{	0,		0x0ac1f48a	}
+};
+
+static const struct hpt_clock hpt372_timings_33[] = {
+	{	XFER_UDMA_6,	0x1c81dc62	},
+	{	XFER_UDMA_5,	0x1c6ddc62	},
+	{	XFER_UDMA_4,	0x1c8ddc62	},
+	{	XFER_UDMA_3,	0x1c8edc62	},	/* checkme */
+	{	XFER_UDMA_2,	0x1c91dc62	},
+	{	XFER_UDMA_1,	0x1c9adc62	},	/* checkme */
+	{	XFER_UDMA_0,	0x1c82dc62	},	/* checkme */
+
+	{	XFER_MW_DMA_2,	0x2c829262	},
+	{	XFER_MW_DMA_1,	0x2c829266	},	/* checkme */
+	{	XFER_MW_DMA_0,	0x2c82922e	},	/* checkme */
+
+	{	XFER_PIO_4,	0x0c829c62	},
+	{	XFER_PIO_3,	0x0c829c84	},
+	{	XFER_PIO_2,	0x0c829ca6	},
+	{	XFER_PIO_1,	0x0d029d26	},
+	{	XFER_PIO_0,	0x0d029d5e	},
+	{	0,		0x0d029d5e	}
+};
+
+static const struct hpt_clock hpt372_timings_50[] = {
+	{	XFER_UDMA_5,	0x12848242	},
+	{	XFER_UDMA_4,	0x12ac8242	},
+	{	XFER_UDMA_3,	0x128c8242	},
+	{	XFER_UDMA_2,	0x120c8242	},
+	{	XFER_UDMA_1,	0x12148254	},
+	{	XFER_UDMA_0,	0x121882ea	},
+
+	{	XFER_MW_DMA_2,	0x22808242	},
+	{	XFER_MW_DMA_1,	0x22808254	},
+	{	XFER_MW_DMA_0,	0x228082ea	},
+
+	{	XFER_PIO_4,	0x0a81f442	},
+	{	XFER_PIO_3,	0x0a81f443	},
+	{	XFER_PIO_2,	0x0a81f454	},
+	{	XFER_PIO_1,	0x0ac1f465	},
+	{	XFER_PIO_0,	0x0ac1f48a	},
+	{	0,		0x0a81f443	}
+};
+
+static const struct hpt_clock hpt372_timings_66[] = {
+	{	XFER_UDMA_6,	0x1c869c62	},
+	{	XFER_UDMA_5,	0x1cae9c62	},
+	{	XFER_UDMA_4,	0x1c8a9c62	},
+	{	XFER_UDMA_3,	0x1c8e9c62	},
+	{	XFER_UDMA_2,	0x1c929c62	},
+	{	XFER_UDMA_1,	0x1c9a9c62	},
+	{	XFER_UDMA_0,	0x1c829c62	},
+
+	{	XFER_MW_DMA_2,	0x2c829c62	},
+	{	XFER_MW_DMA_1,	0x2c829c66	},
+	{	XFER_MW_DMA_0,	0x2c829d2e	},
+
+	{	XFER_PIO_4,	0x0c829c62	},
+	{	XFER_PIO_3,	0x0c829c84	},
+	{	XFER_PIO_2,	0x0c829ca6	},
+	{	XFER_PIO_1,	0x0d029d26	},
+	{	XFER_PIO_0,	0x0d029d5e	},
+	{	0,		0x0d029d26	}
+};
+
+static const struct hpt_clock hpt374_timings_33[] = {
+	{	XFER_UDMA_6,	0x12808242	},
+	{	XFER_UDMA_5,	0x12848242	},
+	{	XFER_UDMA_4,	0x12ac8242	},
+	{	XFER_UDMA_3,	0x128c8242	},
+	{	XFER_UDMA_2,	0x120c8242	},
+	{	XFER_UDMA_1,	0x12148254	},
+	{	XFER_UDMA_0,	0x121882ea	},
+
+	{	XFER_MW_DMA_2,	0x22808242	},
+	{	XFER_MW_DMA_1,	0x22808254	},
+	{	XFER_MW_DMA_0,	0x228082ea	},
+
+	{	XFER_PIO_4,	0x0a81f442	},
+	{	XFER_PIO_3,	0x0a81f443	},
+	{	XFER_PIO_2,	0x0a81f454	},
+	{	XFER_PIO_1,	0x0ac1f465	},
+	{	XFER_PIO_0,	0x0ac1f48a	},
+	{	0,		0x06814e93	}
+};
+
+static const struct hpt_chip hpt370 = {
+	"HPT370",
+	48,
+	{
+		hpt370_timings_33,
+		NULL,
+		NULL,
+		hpt370_timings_66
+	}
+};
+
+static const struct hpt_chip hpt370a = {
+	"HPT370A",
+	48,
+	{
+		hpt370a_timings_33,
+		NULL,
+		hpt370a_timings_50,
+		hpt370a_timings_66
+	}
+};
+
+static const struct hpt_chip hpt372 = {
+	"HPT372",
+	55,
+	{
+		hpt372_timings_33,
+		NULL,
+		hpt372_timings_50,
+		hpt372_timings_66
+	}
+};
+
+static const struct hpt_chip hpt302 = {
+	"HPT302",
+	66,
+	{
+		hpt372_timings_33,
+		NULL,
+		hpt372_timings_50,
+		hpt372_timings_66
+	}
+};
+
+static const struct hpt_chip hpt371 = {
+	"HPT371",
+	66,
+	{
+		hpt372_timings_33,
+		NULL,
+		hpt372_timings_50,
+		hpt372_timings_66
+	}
+};
+
+static const struct hpt_chip hpt372a = {
+	"HPT372A",
+	66,
+	{
+		hpt372_timings_33,
+		NULL,
+		hpt372_timings_50,
+		hpt372_timings_66
+	}
+};
+
+static const struct hpt_chip hpt374 = {
+	"HPT374",
+	48,
+	{
+		hpt374_timings_33,
+		NULL,
+		NULL,
+		NULL
+	}
+};
+
+/**
+ *	hpt37x_find_mode	-	reset the hpt37x bus
+ *	@ap: ATA port
+ *	@speed: transfer mode
+ *
+ *	Return the 32bit register programming information for this channel
+ *	that matches the speed provided.
+ */
+ 
+static u32 hpt37x_find_mode(struct ata_port *ap, int speed)
+{
+	struct hpt_clock *clocks = ap->host->private_data;
+	
+	while(clocks->xfer_speed) {
+		if (clocks->xfer_speed == speed)
+			return clocks->timing;
+		clocks++;
+	}
+	BUG();
+	return 0xffffffffU;	/* silence compiler warning */
+}
+
+static int hpt_dma_blacklisted(const struct ata_device *dev, char *modestr, const char *list[])
+{
+	unsigned char model_num[40];
+	char *s;
+	unsigned int len;
+	int i = 0;
+
+	ata_id_string(dev->id, model_num, ATA_ID_PROD_OFS,
+			  sizeof(model_num));
+	s = &model_num[0];
+	len = strnlen(s, sizeof(model_num));
+
+	/* ATAPI specifies that empty space is blank-filled; remove blanks */
+	while ((len > 0) && (s[len - 1] == ' ')) {
+		len--;
+		s[len] = 0;
+	}
+
+	while(list[i] != NULL) {
+		if (!strncmp(list[i], s, len)) {
+			printk(KERN_WARNING DRV_NAME ": %s is not supported for %s.\n", 
+				modestr, list[i]);
+			return 1;
+		}
+		i++;
+	}
+	return 0;
+}
+
+static const char *bad_ata33[] = {
+	"Maxtor 92720U8", "Maxtor 92040U6", "Maxtor 91360U4", "Maxtor 91020U3", "Maxtor 90845U3", "Maxtor 90650U2",
+	"Maxtor 91360D8", "Maxtor 91190D7", "Maxtor 91020D6", "Maxtor 90845D5", "Maxtor 90680D4", "Maxtor 90510D3", "Maxtor 90340D2",
+	"Maxtor 91152D8", "Maxtor 91008D7", "Maxtor 90845D6", "Maxtor 90840D6", "Maxtor 90720D5", "Maxtor 90648D5", "Maxtor 90576D4",
+	"Maxtor 90510D4",
+	"Maxtor 90432D3", "Maxtor 90288D2", "Maxtor 90256D2",
+	"Maxtor 91000D8", "Maxtor 90910D8", "Maxtor 90875D7", "Maxtor 90840D7", "Maxtor 90750D6", "Maxtor 90625D5", "Maxtor 90500D4",
+	"Maxtor 91728D8", "Maxtor 91512D7", "Maxtor 91303D6", "Maxtor 91080D5", "Maxtor 90845D4", "Maxtor 90680D4", "Maxtor 90648D3", "Maxtor 90432D2",
+	NULL
+};
+
+static const char *bad_ata100_5[] = {
+	"IBM-DTLA-307075",
+	"IBM-DTLA-307060",
+	"IBM-DTLA-307045",
+	"IBM-DTLA-307030",
+	"IBM-DTLA-307020",
+	"IBM-DTLA-307015",
+	"IBM-DTLA-305040",
+	"IBM-DTLA-305030",
+	"IBM-DTLA-305020",
+	"IC35L010AVER07-0",
+	"IC35L020AVER07-0",
+	"IC35L030AVER07-0",
+	"IC35L040AVER07-0",
+	"IC35L060AVER07-0",
+	"WDC AC310200R",
+	NULL
+};
+
+/**
+ *	hpt370_filter	-	mode selection filter
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Block UDMA on devices that cause trouble with this controller.
+ */
+ 
+static unsigned long hpt370_filter(const struct ata_port *ap, struct ata_device *adev, unsigned long mask)
+{
+	if (adev->class != ATA_DEV_ATA) {
+		if (hpt_dma_blacklisted(adev, "UDMA", bad_ata33))
+			mask &= ~ATA_MASK_UDMA;
+		if (hpt_dma_blacklisted(adev, "UDMA100", bad_ata100_5))
+			mask &= ~(0x1F << ATA_SHIFT_UDMA);
+	}
+	return ata_pci_default_filter(ap, adev, mask);
+}
+
+/**
+ *	hpt370a_filter	-	mode selection filter
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Block UDMA on devices that cause trouble with this controller.
+ */
+ 
+static unsigned long hpt370a_filter(const struct ata_port *ap, struct ata_device *adev, unsigned long mask)
+{
+	if (adev->class != ATA_DEV_ATA) {
+		if (hpt_dma_blacklisted(adev, "UDMA100", bad_ata100_5))
+			mask &= ~ (0x1F << ATA_SHIFT_UDMA);
+	}
+	return ata_pci_default_filter(ap, adev, mask);
+}
+	
+/**
+ *	hpt37x_pre_reset	-	reset the hpt37x bus
+ *	@ap: ATA port to reset
+ *
+ *	Perform the initial reset handling for the 370/372 and 374 func 0
+ */
+ 
+static int hpt37x_pre_reset(struct ata_port *ap)
+{
+	u8 scr2, ata66;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	pci_read_config_byte(pdev, 0x5B, &scr2);
+	pci_write_config_byte(pdev, 0x5B, scr2 & ~0x01);
+	/* Cable register now active */
+	pci_read_config_byte(pdev, 0x5A, &ata66);
+	/* Restore state */
+	pci_write_config_byte(pdev, 0x5B, scr2);
+	
+	if (ata66 & (1 << ap->port_no))
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+
+	/* Reset the state machine */
+	pci_write_config_byte(pdev, 0x50, 0x37); 
+	pci_write_config_byte(pdev, 0x54, 0x37); 
+	udelay(100);
+	
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	hpt37x_error_handler	-	reset the hpt374
+ *	@ap: ATA port to reset
+ *
+ *	Perform probe for HPT37x, except for HPT374 channel 2
+ */
+ 
+static void hpt37x_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, hpt37x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+static int hpt374_pre_reset(struct ata_port *ap)
+{
+	u16 mcr3, mcr6;
+	u8 ata66;
+	
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	/* Do the extra channel work */
+	pci_read_config_word(pdev, 0x52, &mcr3);
+	pci_read_config_word(pdev, 0x56, &mcr6);
+	/* Set bit 15 of 0x52 to enable TCBLID as input
+	   Set bit 15 of 0x56 to enable FCBLID as input
+	 */
+	pci_write_config_word(pdev, 0x52, mcr3 | 0x8000);
+	pci_write_config_word(pdev, 0x56, mcr6 | 0x8000);
+	pci_read_config_byte(pdev, 0x5A, &ata66);
+	/* Reset TCBLID/FCBLID to output */
+	pci_write_config_word(pdev, 0x52, mcr3);
+	pci_write_config_word(pdev, 0x56, mcr6);
+	
+	if (ata66 & (1 << ap->port_no))
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+
+	/* Reset the state machine */
+	pci_write_config_byte(pdev, 0x50, 0x37); 
+	pci_write_config_byte(pdev, 0x54, 0x37); 
+	udelay(100);
+	
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	hpt374_error_handler	-	reset the hpt374
+ *	@classes:
+ *
+ *	The 374 cable detect is a little different due to the extra
+ *	channels. The function 0 channels work like usual but function 1
+ *	is special
+ */
+ 
+static void hpt374_error_handler(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+		
+	if (!(PCI_FUNC(pdev->devfn) & 1))
+		hpt37x_error_handler(ap);
+	else
+		ata_bmdma_drive_eh(ap, hpt374_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	hpt370_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Perform PIO mode setup. 
+ */
+ 
+static void hpt370_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	fast &= ~0x02;
+	fast |= 0x01;
+	pci_write_config_byte(pdev, addr2, fast);
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt37x_find_mode(ap, adev->pio_mode);
+	mode &= ~0x8000000;	/* No FIFO in PIO */
+	mode &= ~0x30070000;	/* Leave config bits alone */
+	reg &= 0x30070000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt370_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	Set up the channel for MWDMA or UDMA modes. Much the same as with
+ *	PIO, load the mode number and then set MWDMA or UDMA flag.
+ */
+ 
+static void hpt370_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	fast &= ~0x02;
+	fast |= 0x01;
+	pci_write_config_byte(pdev, addr2, fast);
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt37x_find_mode(ap, adev->dma_mode);
+	mode |= 0x8000000;	/* FIFO in MWDMA or UDMA */
+	mode &= ~0xC0000000;	/* Leave config bits alone */
+	reg &= 0xC0000000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt370_bmdma_start		-	DMA engine begin
+ *	@qc: ATA command
+ *
+ *	The 370 and 370A want us to reset the DMA engine each time we
+ *	use it. The 372 and later are fine.
+ */
+ 
+static void hpt370_bmdma_start(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	pci_write_config_byte(pdev, 0x50 + 4 * ap->port_no, 0x37);
+	udelay(10);
+	ata_bmdma_start(qc);
+}
+
+/**
+ *	hpt370_bmdma_end		-	DMA engine stop
+ *	@qc: ATA command
+ *
+ *	Work around the HPT370 DMA engine.
+ */
+ 
+static void hpt370_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 dma_stat = inb(ap->ioaddr.bmdma_addr + 2);
+	u8 dma_cmd;
+	unsigned long bmdma = ap->ioaddr.bmdma_addr;
+	
+	if (dma_stat & 0x01) {
+		udelay(20);
+		dma_stat = inb(bmdma + 2);
+	}
+	if (dma_stat & 0x01) {
+		/* Clear the engine */
+		pci_write_config_byte(pdev, 0x50 + 4 * ap->port_no, 0x37);
+		udelay(10);
+		/* Stop DMA */
+		dma_cmd = inb(bmdma );
+		outb(dma_cmd & 0xFE, bmdma);
+		/* Clear Error */
+		dma_stat = inb(bmdma + 2);
+		outb(dma_stat | 0x06 , bmdma + 2);
+		/* Clear the engine */
+		pci_write_config_byte(pdev, 0x50 + 4 * ap->port_no, 0x37);
+		udelay(10);
+	}
+	ata_bmdma_stop(qc);
+}
+
+/**
+ *	hpt372_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Perform PIO mode setup. 
+ */
+ 
+static void hpt372_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	fast &= ~0x07;
+	pci_write_config_byte(pdev, addr2, fast);
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt37x_find_mode(ap, adev->pio_mode);
+	
+	printk("Find mode for %d reports %X\n", adev->pio_mode, mode);
+	mode &= ~0x80000000;	/* No FIFO in PIO */
+	mode &= ~0x30070000;	/* Leave config bits alone */
+	reg &= 0x30070000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt372_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	Set up the channel for MWDMA or UDMA modes. Much the same as with
+ *	PIO, load the mode number and then set MWDMA or UDMA flag.
+ */
+ 
+static void hpt372_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	fast &= ~0x07;
+	pci_write_config_byte(pdev, addr2, fast);
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt37x_find_mode(ap, adev->dma_mode);
+	printk("Find mode for DMA %d reports %X\n", adev->dma_mode, mode);
+	mode &= ~0xC0000000;	/* Leave config bits alone */
+	mode |= 0x80000000;	/* FIFO in MWDMA or UDMA */
+	reg &= 0xC0000000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt37x_bmdma_end		-	DMA engine stop
+ *	@qc: ATA command
+ *
+ *	Clean up after the HPT372 and later DMA engine
+ */
+ 
+static void hpt37x_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int mscreg = 0x50 + 2 * ap->port_no;
+	u8 bwsr_stat, msc_stat;
+	
+	pci_read_config_byte(pdev, 0x6A, &bwsr_stat);
+	pci_read_config_byte(pdev, mscreg, &msc_stat);
+	if (bwsr_stat & (1 << ap->port_no))
+		pci_write_config_byte(pdev, mscreg, msc_stat | 0x30);
+	ata_bmdma_stop(qc);
+}
+
+
+static struct scsi_host_template hpt37x_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+/*
+ *	Configuration for HPT370
+ */
+ 
+static struct ata_port_operations hpt370_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt370_set_piomode,
+	.set_dmamode	= hpt370_set_dmamode,
+	.mode_filter	= hpt370_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt37x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= hpt370_bmdma_start,
+	.bmdma_stop	= hpt370_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/*
+ *	Configuration for HPT370A. Close to 370 but less filters
+ */
+ 
+static struct ata_port_operations hpt370a_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt370_set_piomode,
+	.set_dmamode	= hpt370_set_dmamode,
+	.mode_filter	= hpt370a_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt37x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= hpt370_bmdma_start,
+	.bmdma_stop	= hpt370_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/*
+ *	Configuration for HPT372, HPT371, HPT302. Slightly different PIO
+ *	and DMA mode setting functionality.
+ */
+ 
+static struct ata_port_operations hpt372_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt372_set_piomode,
+	.set_dmamode	= hpt372_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt37x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= hpt37x_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/*
+ *	Configuration for HPT374. Mode setting works like 372 and friends
+ *	but we have a different cable detection procedure.
+ */
+ 
+static struct ata_port_operations hpt374_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt372_set_piomode,
+	.set_dmamode	= hpt372_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt374_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= hpt37x_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/**
+ *	htp37x_clock_slot	-	Turn timing to PC clock entry
+ *	@freq: Reported frequency timing
+ *	@base: Base timing
+ *
+ *	Turn the timing data intoa clock slot (0 for 33, 1 for 40, 2 for 50
+ *	and 3 for 66Mhz)
+ */
+ 
+static int hpt37x_clock_slot(unsigned int freq, unsigned int base)
+{
+	unsigned int f = (base * freq) / 192;	/* Mhz */
+	if (f < 40)
+		return 0;	/* 33Mhz slot */
+	if (f < 45)
+		return 1;	/* 40Mhz slot */
+	if (f < 55)
+		return 2;	/* 50Mhz slot */
+	return 3;		/* 60Mhz slot */
+}
+
+/**
+ *	hpt37x_calibrate_dpll		-	Calibrate the DPLL loop
+ *	@dev: PCI device 
+ *
+ *	Perform a calibration cycle on the HPT37x DPLL. Returns 1 if this
+ *	succeeds
+ */
+
+static int hpt37x_calibrate_dpll(struct pci_dev *dev)
+{
+	u8 reg5b;
+	u32 reg5c;
+	int tries;
+ 
+	for(tries = 0; tries < 0x5000; tries++) {
+		udelay(50);
+		pci_read_config_byte(dev, 0x5b, &reg5b);
+		if (reg5b & 0x80) {
+			/* See if it stays set */
+			for(tries = 0; tries < 0x1000; tries ++) {
+				pci_read_config_byte(dev, 0x5b, &reg5b);
+				/* Failed ? */
+				if ((reg5b & 0x80) == 0)
+					return 0;
+			}
+			/* Turn off tuning, we have the DPLL set */
+			pci_read_config_dword(dev, 0x5c, &reg5c);
+			pci_write_config_dword(dev, 0x5c, reg5c & ~ 0x100);
+			return 1;
+		}
+	}
+	/* Never went stable */
+	return 0;
+}
+/**
+ *	hpt37x_init_one		-	Initialise an HPT37X/302
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Initialise an HPT37x device. There are some interesting complications
+ *	here. Firstly the chip may report 366 and be one of several variants.
+ *	Secondly all the timings depend on the clock for the chip which we must
+ *	detect and look up
+ *
+ *	This is the known chip mappings. It may be missing a couple of later
+ *	releases.
+ *
+ *	Chip version		PCI		Rev	Notes
+ *	HPT366			4 (HPT366)	0	Other driver
+ *	HPT366			4 (HPT366)	1	Other driver
+ *	HPT368			4 (HPT366)	2	Other driver
+ *	HPT370			4 (HPT366)	3	UDMA100
+ *	HPT370A			4 (HPT366)	4	UDMA100
+ *	HPT372			4 (HPT366)	5	UDMA133 (1)
+ *	HPT372N			4 (HPT366)	6	Other driver
+ *	HPT372A			5 (HPT372)	1	UDMA133 (1)
+ *	HPT372N			5 (HPT372)	2	Other driver
+ *	HPT302			6 (HPT302)	1	UDMA133
+ *	HPT302N			6 (HPT302)	2	Other driver
+ *	HPT371			7 (HPT371)	*	UDMA133
+ *	HPT374			8 (HPT374)	*	UDMA133 4 channel
+ *	HPT372N			9 (HPT372N)	*	Other driver
+ *
+ *	(1) UDMA133 support depends on the bus clock
+ */
+ 
+static int hpt37x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	/* HPT370 - UDMA100 */
+	static struct ata_port_info info_hpt370 = {
+		.sht = &hpt37x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &hpt370_port_ops
+	};
+	/* HPT370A - UDMA100 */
+	static struct ata_port_info info_hpt370a = {
+		.sht = &hpt37x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &hpt370a_port_ops
+	};
+	/* HPT371, 372 and friends - UDMA133 */
+	static struct ata_port_info info_hpt372 = {
+		.sht = &hpt37x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &hpt372_port_ops
+	};
+	/* HPT371, 372 and friends - UDMA100 at 50MHz clock */
+	static struct ata_port_info info_hpt372_50 = {
+		.sht = &hpt37x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &hpt372_port_ops
+	};
+	/* HPT374 - UDMA133 */
+	static struct ata_port_info info_hpt374 = {
+		.sht = &hpt37x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &hpt374_port_ops
+	};
+
+	static const int MHz[4] = { 33, 40, 50, 66 };
+
+	struct ata_port_info *port_info[2];
+	struct ata_port_info *port;
+
+	u8 irqmask;
+	u32 class_rev;
+	u32 freq;
+
+	const struct hpt_chip *chip_table;
+	int clock_slot;
+
+	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class_rev);
+	class_rev &= 0xFF;
+	
+	if (dev->device == PCI_DEVICE_ID_TTI_HPT366) {
+		/* May be a later chip in disguise. Check */
+		/* Older chips are in the HPT366 driver. Ignore them */
+		if (class_rev < 3)
+			return -ENODEV;
+		/* N series chips have their own driver. Ignore */
+		if (class_rev == 6)
+			return -ENODEV;
+
+		switch(class_rev) {			
+			case 3:
+				port = &info_hpt370;
+				chip_table = &hpt370;
+				break;
+			case 4:
+				port = &info_hpt370a;
+				chip_table = &hpt370a;
+				break;
+			case 5:
+				port = &info_hpt372;
+				chip_table = &hpt372;
+				break;
+			default:
+				printk(KERN_ERR "pata_hpt37x: Unknown HPT366 subtype please report (%d).\n", class_rev);
+				return -ENODEV;
+		}
+	} else {
+		switch(dev->device) {
+			case PCI_DEVICE_ID_TTI_HPT372:
+				/* 372N if rev >= 2*/
+				if (class_rev >= 2)
+					return -ENODEV;
+				port = &info_hpt372;
+				chip_table = &hpt372a;
+				break;
+			case PCI_DEVICE_ID_TTI_HPT302:
+				/* 302N if rev > 1 */
+				if (class_rev > 1)
+					return -ENODEV;
+				port = &info_hpt372;
+				/* Check this */
+				chip_table = &hpt302;
+				break;
+			case PCI_DEVICE_ID_TTI_HPT371:
+				port = &info_hpt372;
+				chip_table = &hpt371;
+				break;
+			case PCI_DEVICE_ID_TTI_HPT374:
+				chip_table = &hpt374;
+				port = &info_hpt374;
+				break;
+			default:
+				printk(KERN_ERR "pata_hpt37x: PCI table is bogus please report (%d).\n", dev->device);
+				return -ENODEV;
+		}
+	}
+	/* Ok so this is a chip we support */
+
+	pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
+	pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
+	pci_write_config_byte(dev, PCI_MIN_GNT, 0x08);
+	pci_write_config_byte(dev, PCI_MAX_LAT, 0x08);
+
+	pci_read_config_byte(dev, 0x5A, &irqmask);
+	irqmask &= ~0x10;
+	pci_write_config_byte(dev, 0x5a, irqmask);
+
+	/*
+	 * default to pci clock. make sure MA15/16 are set to output
+	 * to prevent drives having problems with 40-pin cables. Needed
+	 * for some drives such as IBM-DTLA which will not enter ready
+	 * state on reset when PDIAG is a input.
+	 */
+
+	pci_write_config_byte(dev, 0x5b, 0x23);	
+	
+	pci_read_config_dword(dev, 0x70, &freq);
+	if ((freq >> 12) != 0xABCDE) {
+		int i;
+		u8 sr;
+		u32 total = 0;
+		
+		printk(KERN_WARNING "pata_hpt37x: BIOS has not set timing clocks.\n");
+		
+		/* This is the process the HPT371 BIOS is reported to use */
+		for(i = 0; i < 128; i++) {
+			pci_read_config_byte(dev, 0x78, &sr);
+			total += sr;
+			udelay(15);
+		}
+		freq = total / 128;
+	}
+	freq &= 0x1FF;
+	
+	/*
+	 *	Turn the frequency check into a band and then find a timing
+	 *	table to match it.
+	 */
+	 
+	clock_slot = hpt37x_clock_slot(freq, chip_table->base);
+	if (chip_table->clocks[clock_slot] == NULL) {
+		/*
+		 *	We need to try PLL mode instead
+		 */
+		unsigned int f_low = (MHz[clock_slot] * chip_table->base) / 192;
+		unsigned int f_high = f_low + 2;
+		int adjust;
+		
+		for(adjust = 0; adjust < 8; adjust++) {
+			if (hpt37x_calibrate_dpll(dev))
+				break;
+			/* See if it'll settle at a fractionally different clock */
+			if ((adjust & 3) == 3) {
+				f_low --;
+				f_high ++;
+			}
+			pci_write_config_dword(dev, 0x5C, (f_high << 16) | f_low);
+		}
+		if (adjust == 8) {
+			printk(KERN_WARNING "hpt37x: DPLL did not stabilize.\n");
+			return -ENODEV;
+		}
+		/* Check if this works for all cases */
+		port->private_data = (void *)hpt370_timings_66;
+		
+		printk(KERN_INFO "hpt37x: Bus clock %dMHz, using DPLL.\n", MHz[clock_slot]);
+	} else {
+		port->private_data = (void *)chip_table->clocks[clock_slot];
+		/*
+		 *	Perform a final fixup. The 371 and 372 clock determines
+		 *	if UDMA133 is available.
+		 */
+	 
+		if (clock_slot == 2 && chip_table == &hpt372) {	/* 50Mhz */
+			printk(KERN_WARNING "pata_hpt37x: No UDMA133 support available with 50MHz bus clock.\n");
+			if (port == &info_hpt372)
+				port = &info_hpt372_50;
+			else BUG();
+		}
+		printk(KERN_INFO "hpt37x: %s: Bus clock %dMHz.\n", chip_table->name, MHz[clock_slot]);
+	}
+	port_info[0] = port_info[1] = port;	
+	/* Now kick off ATA set up */
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id hpt37x[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT366), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT371), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT372), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT374), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT302), },
+	{ 0, },
+};
+
+static struct pci_driver hpt37x_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= hpt37x,
+	.probe 		= hpt37x_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init hpt37x_init(void)
+{
+	return pci_register_driver(&hpt37x_pci_driver);
+}
+
+
+static void __exit hpt37x_exit(void)
+{
+	pci_unregister_driver(&hpt37x_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the Highpoint HPT37x/30x");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, hpt37x);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(hpt37x_init);
+module_exit(hpt37x_exit);
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
new file mode 100644
index 000000000000..40fcda62c7a2
--- /dev/null
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -0,0 +1,597 @@
+/*
+ * Libata driver for the highpoint 372N and 302N UDMA66 ATA controllers.
+ *
+ * This driver is heavily based upon:
+ *
+ * linux/drivers/ide/pci/hpt366.c		Version 0.36	April 25, 2003
+ *
+ * Copyright (C) 1999-2003		Andre Hedrick <andre@linux-ide.org>
+ * Portions Copyright (C) 2001	        Sun Microsystems, Inc.
+ * Portions Copyright (C) 2003		Red Hat Inc
+ *
+ *
+ * TODO
+ *	371N
+ *	Work out best PLL policy
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"pata_hpt3x2n"
+#define DRV_VERSION	"0.3"
+
+enum {
+	HPT_PCI_FAST	=	(1 << 31),
+	PCI66		=	(1 << 1),
+	USE_DPLL	=	(1 << 0)
+};
+
+struct hpt_clock {
+	u8	xfer_speed;
+	u32	timing;
+};
+
+struct hpt_chip {
+	const char *name;
+	struct hpt_clock *clocks[3];
+};
+
+/* key for bus clock timings
+ * bit
+ * 0:3    data_high_time. inactive time of DIOW_/DIOR_ for PIO and MW
+ *        DMA. cycles = value + 1
+ * 4:8    data_low_time. active time of DIOW_/DIOR_ for PIO and MW
+ *        DMA. cycles = value + 1
+ * 9:12   cmd_high_time. inactive time of DIOW_/DIOR_ during task file
+ *        register access.
+ * 13:17  cmd_low_time. active time of DIOW_/DIOR_ during task file
+ *        register access.
+ * 18:21  udma_cycle_time. clock freq and clock cycles for UDMA xfer.
+ *        during task file register access.
+ * 22:24  pre_high_time. time to initialize 1st cycle for PIO and MW DMA
+ *        xfer.
+ * 25:27  cmd_pre_high_time. time to initialize 1st PIO cycle for task
+ *        register access.
+ * 28     UDMA enable
+ * 29     DMA enable
+ * 30     PIO_MST enable. if set, the chip is in bus master mode during
+ *        PIO.
+ * 31     FIFO enable.
+ */
+ 
+/* 66MHz DPLL clocks */
+
+static struct hpt_clock hpt3x2n_clocks[] = {
+	{	XFER_UDMA_7,	0x1c869c62	},
+	{	XFER_UDMA_6,	0x1c869c62	},
+	{	XFER_UDMA_5,	0x1c8a9c62	},
+	{	XFER_UDMA_4,	0x1c8a9c62	},
+	{	XFER_UDMA_3,	0x1c8e9c62	},
+	{	XFER_UDMA_2,	0x1c929c62	},
+	{	XFER_UDMA_1,	0x1c9a9c62	},
+	{	XFER_UDMA_0,	0x1c829c62	},
+
+	{	XFER_MW_DMA_2,	0x2c829c62	},
+	{	XFER_MW_DMA_1,	0x2c829c66	},
+	{	XFER_MW_DMA_0,	0x2c829d2c	},
+
+	{	XFER_PIO_4,	0x0c829c62	},
+	{	XFER_PIO_3,	0x0c829c84	},
+	{	XFER_PIO_2,	0x0c829ca6	},
+	{	XFER_PIO_1,	0x0d029d26	},
+	{	XFER_PIO_0,	0x0d029d5e	},
+	{	0,		0x0d029d5e	}
+};
+
+/**
+ *	hpt3x2n_find_mode	-	reset the hpt3x2n bus
+ *	@ap: ATA port
+ *	@speed: transfer mode
+ *
+ *	Return the 32bit register programming information for this channel
+ *	that matches the speed provided. For the moment the clocks table
+ *	is hard coded but easy to change. This will be needed if we use
+ *	different DPLLs
+ */
+ 
+static u32 hpt3x2n_find_mode(struct ata_port *ap, int speed)
+{
+	struct hpt_clock *clocks = hpt3x2n_clocks;
+	
+	while(clocks->xfer_speed) {
+		if (clocks->xfer_speed == speed)
+			return clocks->timing;
+		clocks++;
+	}
+	BUG();
+	return 0xffffffffU;	/* silence compiler warning */
+}
+
+/**
+ *	hpt3x2n_pre_reset	-	reset the hpt3x2n bus
+ *	@ap: ATA port to reset
+ *
+ *	Perform the initial reset handling for the 3x2n series controllers.
+ *	Reset the hardware and state machine, obtain the cable type.
+ */
+ 
+static int hpt3xn_pre_reset(struct ata_port *ap)
+{
+	u8 scr2, ata66;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	pci_read_config_byte(pdev, 0x5B, &scr2);
+	pci_write_config_byte(pdev, 0x5B, scr2 & ~0x01);
+	/* Cable register now active */
+	pci_read_config_byte(pdev, 0x5A, &ata66);
+	/* Restore state */
+	pci_write_config_byte(pdev, 0x5B, scr2);
+	
+	if (ata66 & (1 << ap->port_no))
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+
+	/* Reset the state machine */
+	pci_write_config_byte(pdev, 0x50, 0x37); 
+	pci_write_config_byte(pdev, 0x54, 0x37); 
+	udelay(100);
+
+	return ata_std_prereset(ap);
+}
+	
+/**
+ *	hpt3x2n_error_handler	-	probe the hpt3x2n bus
+ *	@ap: ATA port to reset
+ *
+ *	Perform the probe reset handling for the 3x2N
+ */
+ 
+static void hpt3x2n_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, hpt3xn_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+				     
+/**
+ *	hpt3x2n_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Perform PIO mode setup. 
+ */
+ 
+static void hpt3x2n_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	fast &= ~0x07;
+	pci_write_config_byte(pdev, addr2, fast);
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt3x2n_find_mode(ap, adev->pio_mode);
+	mode &= ~0x8000000;	/* No FIFO in PIO */
+	mode &= ~0x30070000;	/* Leave config bits alone */
+	reg &= 0x30070000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt3x2n_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	Set up the channel for MWDMA or UDMA modes. Much the same as with
+ *	PIO, load the mode number and then set MWDMA or UDMA flag.
+ */
+ 
+static void hpt3x2n_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 addr1, addr2;
+	u32 reg;
+	u32 mode;
+	u8 fast;
+
+	addr1 = 0x40 + 4 * (adev->devno + 2 * ap->port_no);
+	addr2 = 0x51 + 4 * ap->port_no;
+	
+	/* Fast interrupt prediction disable, hold off interrupt disable */
+	pci_read_config_byte(pdev, addr2, &fast);
+	fast &= ~0x07;
+	pci_write_config_byte(pdev, addr2, fast);
+	
+	pci_read_config_dword(pdev, addr1, &reg);
+	mode = hpt3x2n_find_mode(ap, adev->dma_mode);
+	mode |= 0x8000000;	/* FIFO in MWDMA or UDMA */
+	mode &= ~0xC0000000;	/* Leave config bits alone */
+	reg &= 0xC0000000;	/* Strip timing bits */
+	pci_write_config_dword(pdev, addr1, reg | mode);
+}
+
+/**
+ *	hpt3x2n_bmdma_end		-	DMA engine stop
+ *	@qc: ATA command
+ *
+ *	Clean up after the HPT3x2n and later DMA engine
+ */
+ 
+static void hpt3x2n_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int mscreg = 0x50 + 2 * ap->port_no;
+	u8 bwsr_stat, msc_stat;
+	
+	pci_read_config_byte(pdev, 0x6A, &bwsr_stat);
+	pci_read_config_byte(pdev, mscreg, &msc_stat);
+	if (bwsr_stat & (1 << ap->port_no))
+		pci_write_config_byte(pdev, mscreg, msc_stat | 0x30);
+	ata_bmdma_stop(qc);
+}
+
+/**
+ *	hpt3x2n_set_clock	-	clock control
+ *	@ap: ATA port
+ *	@source: 0x21 or 0x23 for PLL or PCI sourced clock
+ *
+ *	Switch the ATA bus clock between the PLL and PCI clock sources
+ *	while correctly isolating the bus and resetting internal logic
+ *
+ *	We must use the DPLL for
+ *	-	writing
+ *	-	second channel UDMA7 (SATA ports) or higher
+ *	-	66MHz PCI
+ *	
+ *	or we will underclock the device and get reduced performance.
+ */
+ 
+static void hpt3x2n_set_clock(struct ata_port *ap, int source)
+{
+	unsigned long bmdma = ap->ioaddr.bmdma_addr;
+	
+	/* Tristate the bus */
+	outb(0x80, bmdma+0x73);
+	outb(0x80, bmdma+0x77);
+	
+	/* Switch clock and reset channels */
+	outb(source, bmdma+0x7B);
+	outb(0xC0, bmdma+0x79);
+	
+	/* Reset state machines */
+	outb(0x37, bmdma+0x70);
+	outb(0x37, bmdma+0x74);
+	
+	/* Complete reset */
+	outb(0x00, bmdma+0x79);
+	
+	/* Reconnect channels to bus */
+	outb(0x00, bmdma+0x73);
+	outb(0x00, bmdma+0x77);
+}
+
+/* Check if our partner interface is busy */
+
+static int hpt3x2n_pair_idle(struct ata_port *ap)
+{
+	struct ata_host *host = ap->host;
+	struct ata_port *pair = host->ports[ap->port_no ^ 1];
+	
+	if (pair->hsm_task_state == HSM_ST_IDLE)
+		return 1;
+	return 0;
+}
+
+static int hpt3x2n_use_dpll(struct ata_port *ap, int reading)
+{
+	long flags = (long)ap->host->private_data;
+	/* See if we should use the DPLL */
+	if (reading == 0)
+		return USE_DPLL;	/* Needed for write */
+	if (flags & PCI66)
+		return USE_DPLL;	/* Needed at 66Mhz */
+	return 0;	
+}
+
+static unsigned int hpt3x2n_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_taskfile *tf = &qc->tf;
+	struct ata_port *ap = qc->ap;
+	int flags = (long)ap->host->private_data;
+	
+	if (hpt3x2n_pair_idle(ap)) {
+		int dpll = hpt3x2n_use_dpll(ap, (tf->flags & ATA_TFLAG_WRITE));
+		if ((flags & USE_DPLL) != dpll) {
+			if (dpll == 1)
+				hpt3x2n_set_clock(ap, 0x21);
+			else
+				hpt3x2n_set_clock(ap, 0x23);
+		}
+	}
+	return ata_qc_issue_prot(qc);
+}
+
+static struct scsi_host_template hpt3x2n_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+/*
+ *	Configuration for HPT3x2n.
+ */
+ 
+static struct ata_port_operations hpt3x2n_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt3x2n_set_piomode,
+	.set_dmamode	= hpt3x2n_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt3x2n_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= hpt3x2n_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= hpt3x2n_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/**
+ *	hpt3xn_calibrate_dpll		-	Calibrate the DPLL loop
+ *	@dev: PCI device 
+ *
+ *	Perform a calibration cycle on the HPT3xN DPLL. Returns 1 if this
+ *	succeeds
+ */
+
+static int hpt3xn_calibrate_dpll(struct pci_dev *dev)
+{
+	u8 reg5b;
+	u32 reg5c;
+	int tries;
+ 
+	for(tries = 0; tries < 0x5000; tries++) {
+		udelay(50);
+		pci_read_config_byte(dev, 0x5b, &reg5b);
+		if (reg5b & 0x80) {
+			/* See if it stays set */
+			for(tries = 0; tries < 0x1000; tries ++) {
+				pci_read_config_byte(dev, 0x5b, &reg5b);
+				/* Failed ? */
+				if ((reg5b & 0x80) == 0)
+					return 0;
+			}
+			/* Turn off tuning, we have the DPLL set */
+			pci_read_config_dword(dev, 0x5c, &reg5c);
+			pci_write_config_dword(dev, 0x5c, reg5c & ~ 0x100);
+			return 1;
+		}
+	}
+	/* Never went stable */
+	return 0;
+}
+
+static int hpt3x2n_pci_clock(struct pci_dev *pdev)
+{
+	unsigned long freq;
+	u32 fcnt;
+	
+	pci_read_config_dword(pdev, 0x70/*CHECKME*/, &fcnt);
+	if ((fcnt >> 12) != 0xABCDE) {
+		printk(KERN_WARNING "hpt3xn: BIOS clock data not set.\n");
+		return 33;	/* Not BIOS set */
+	}
+	fcnt &= 0x1FF;
+	
+	freq = (fcnt * 77) / 192;
+	
+	/* Clamp to bands */
+	if (freq < 40)
+		return 33;
+	if (freq < 45)
+		return 40;
+	if (freq < 55)
+		return 50;
+	return 66;
+}
+
+/**
+ *	hpt3x2n_init_one		-	Initialise an HPT37X/302
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Initialise an HPT3x2n device. There are some interesting complications
+ *	here. Firstly the chip may report 366 and be one of several variants.
+ *	Secondly all the timings depend on the clock for the chip which we must
+ *	detect and look up
+ *
+ *	This is the known chip mappings. It may be missing a couple of later
+ *	releases.
+ *
+ *	Chip version		PCI		Rev	Notes
+ *	HPT372			4 (HPT366)	5	Other driver
+ *	HPT372N			4 (HPT366)	6	UDMA133
+ *	HPT372			5 (HPT372)	1	Other driver
+ *	HPT372N			5 (HPT372)	2	UDMA133
+ *	HPT302			6 (HPT302)	*	Other driver
+ *	HPT302N			6 (HPT302)	> 1	UDMA133
+ *	HPT371			7 (HPT371)	*	Other driver
+ *	HPT371N			7 (HPT371)	> 1	UDMA133
+ *	HPT374			8 (HPT374)	*	Other driver
+ *	HPT372N			9 (HPT372N)	*	UDMA133
+ *
+ *	(1) UDMA133 support depends on the bus clock
+ *
+ *	To pin down		HPT371N
+ */
+ 
+static int hpt3x2n_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	/* HPT372N and friends - UDMA133 */
+	static struct ata_port_info info = {
+		.sht = &hpt3x2n_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &hpt3x2n_port_ops
+	};
+	struct ata_port_info *port_info[2];
+	struct ata_port_info *port = &info;
+
+	u8 irqmask;
+	u32 class_rev;
+	
+	unsigned int pci_mhz;
+	unsigned int f_low, f_high;
+	int adjust;
+	
+	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class_rev);
+	class_rev &= 0xFF;
+	
+	switch(dev->device) {
+		case PCI_DEVICE_ID_TTI_HPT366:
+			if (class_rev < 6)
+				return -ENODEV;
+			break;
+		case PCI_DEVICE_ID_TTI_HPT372:
+			/* 372N if rev >= 1*/
+			if (class_rev == 0)
+				return -ENODEV;
+			break;
+		case PCI_DEVICE_ID_TTI_HPT302:
+			if (class_rev < 2)
+				return -ENODEV;
+			break;
+		case PCI_DEVICE_ID_TTI_HPT372N:
+			break;
+		default:
+			printk(KERN_ERR "pata_hpt3x2n: PCI table is bogus please report (%d).\n", dev->device);
+			return -ENODEV;
+	}
+
+	/* Ok so this is a chip we support */
+
+	pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
+	pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
+	pci_write_config_byte(dev, PCI_MIN_GNT, 0x08);
+	pci_write_config_byte(dev, PCI_MAX_LAT, 0x08);
+
+	pci_read_config_byte(dev, 0x5A, &irqmask);
+	irqmask &= ~0x10;
+	pci_write_config_byte(dev, 0x5a, irqmask);
+
+	/* Tune the PLL. HPT recommend using 75 for SATA, 66 for UDMA133 or
+	   50 for UDMA100. Right now we always use 66 */
+	   
+	pci_mhz = hpt3x2n_pci_clock(dev);
+	
+	f_low = (pci_mhz * 48) / 66;	/* PCI Mhz for 66Mhz DPLL */
+	f_high = f_low + 2;		/* Tolerance */
+	
+	pci_write_config_dword(dev, 0x5C, (f_high << 16) | f_low | 0x100);
+	/* PLL clock */
+	pci_write_config_byte(dev, 0x5B, 0x21);
+	
+	/* Unlike the 37x we don't try jiggling the frequency */
+	for(adjust = 0; adjust < 8; adjust++) {
+		if (hpt3xn_calibrate_dpll(dev))
+			break;
+		pci_write_config_dword(dev, 0x5C, (f_high << 16) | f_low);
+	}
+	if (adjust == 8)
+		printk(KERN_WARNING "hpt3xn: DPLL did not stabilize.\n");
+
+	/* Set our private data up. We only need a few flags so we use
+	   it directly */
+	port->private_data = NULL;
+	if (pci_mhz > 60)
+		port->private_data = (void *)PCI66;
+		
+	/* Now kick off ATA set up */
+	port_info[0] = port_info[1] = port;
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id hpt3x2n[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT366), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT372), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT302), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT372N), },
+	{ 0, },
+};
+
+static struct pci_driver hpt3x2n_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= hpt3x2n,
+	.probe 		= hpt3x2n_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init hpt3x2n_init(void)
+{
+	return pci_register_driver(&hpt3x2n_pci_driver);
+}
+
+
+static void __exit hpt3x2n_exit(void)
+{
+	pci_unregister_driver(&hpt3x2n_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the Highpoint HPT3x2n/30x");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, hpt3x2n);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(hpt3x2n_init);
+module_exit(hpt3x2n_exit);
diff --git a/drivers/ata/pata_hpt3x3.c b/drivers/ata/pata_hpt3x3.c
new file mode 100644
index 000000000000..c93406ebf6ed
--- /dev/null
+++ b/drivers/ata/pata_hpt3x3.c
@@ -0,0 +1,226 @@
+/*
+ *	pata_hpt3x3		-	HPT3x3 driver
+ *	(c) Copyright 2005-2006 Red Hat
+ *
+ *	Was pata_hpt34x but the naming was confusing as it supported the
+ *	343 and 363 so it has been renamed.
+ *
+ *	Based on:
+ *	linux/drivers/ide/pci/hpt34x.c		Version 0.40	Sept 10, 2002
+ *	Copyright (C) 1998-2000	Andre Hedrick <andre@linux-ide.org>
+ *
+ *	May be copied or modified under the terms of the GNU General Public
+ *	License
+ */
+ 
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"pata_hpt3x3"
+#define DRV_VERSION	"0.4.1"
+
+static int hpt3x3_probe_init(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	hpt3x3_probe_reset	-	reset the hpt3x3 bus
+ *	@ap: ATA port to reset
+ *
+ *	Perform the housekeeping when doing an ATA bus reeset. We just
+ *	need to force the cable type.
+ */
+
+static void hpt3x3_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, hpt3x3_probe_init, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	hpt3x3_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Set our PIO requirements. This is fairly simple on the HPT3x3 as
+ *	all we have to do is clear the MWDMA and UDMA bits then load the
+ *	mode number.
+ */
+
+static void hpt3x3_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 r1, r2;
+	int dn = 2 * ap->port_no + adev->devno;
+
+	pci_read_config_dword(pdev, 0x44, &r1);
+	pci_read_config_dword(pdev, 0x48, &r2);
+	/* Load the PIO timing number */
+	r1 &= ~(7 << (3 * dn));
+	r1 |= (adev->pio_mode - XFER_PIO_0) << (3 * dn);
+	r2 &= ~(0x11 << dn);	/* Clear MWDMA and UDMA bits */
+
+	pci_write_config_dword(pdev, 0x44, r1);
+	pci_write_config_dword(pdev, 0x48, r2);
+}
+
+/**
+ *	hpt3x3_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	Set up the channel for MWDMA or UDMA modes. Much the same as with
+ *	PIO, load the mode number and then set MWDMA or UDMA flag.
+ */
+
+static void hpt3x3_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 r1, r2;
+	int dn = 2 * ap->port_no + adev->devno;
+	int mode_num = adev->dma_mode & 0x0F;
+
+	pci_read_config_dword(pdev, 0x44, &r1);
+	pci_read_config_dword(pdev, 0x48, &r2);
+	/* Load the timing number */
+	r1 &= ~(7 << (3 * dn));
+	r1 |= (mode_num << (3 * dn));
+	r2 &= ~(0x11 << dn);	/* Clear MWDMA and UDMA bits */
+
+	if (adev->dma_mode >= XFER_UDMA_0)
+		r2 |= 0x01 << dn;	/* Ultra mode */
+	else
+		r2 |= 0x10 << dn;	/* MWDMA */
+
+	pci_write_config_dword(pdev, 0x44, r1);
+	pci_write_config_dword(pdev, 0x48, r2);
+}
+
+static struct scsi_host_template hpt3x3_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations hpt3x3_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= hpt3x3_set_piomode,
+	.set_dmamode	= hpt3x3_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= hpt3x3_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	hpt3x3_init_one		-	Initialise an HPT343/363
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Perform basic initialisation. The chip has a quirk that it won't
+ *	function unless it is at XX00. The old ATA driver touched this up
+ *	but we leave it for pci quirks to do properly.
+ */
+
+static int hpt3x3_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &hpt3x3_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x07,
+		.port_ops = &hpt3x3_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+	u16 cmd;
+
+	/* Initialize the board */
+	pci_write_config_word(dev, 0x80, 0x00);
+	/* Check if it is a 343 or a 363. 363 has COMMAND_MEMORY set */
+	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	if (cmd & PCI_COMMAND_MEMORY)
+		pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0xF0);
+	else
+		pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x20);
+
+	/* Now kick off ATA set up */
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id hpt3x3[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_TTI, PCI_DEVICE_ID_TTI_HPT343), },
+	{ 0, },
+};
+
+static struct pci_driver hpt3x3_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= hpt3x3,
+	.probe 		= hpt3x3_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init hpt3x3_init(void)
+{
+	return pci_register_driver(&hpt3x3_pci_driver);
+}
+
+
+static void __exit hpt3x3_exit(void)
+{
+	pci_unregister_driver(&hpt3x3_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the Highpoint HPT343/363");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, hpt3x3);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(hpt3x3_init);
+module_exit(hpt3x3_exit);
diff --git a/drivers/ata/pata_isapnp.c b/drivers/ata/pata_isapnp.c
new file mode 100644
index 000000000000..73948c8b7270
--- /dev/null
+++ b/drivers/ata/pata_isapnp.c
@@ -0,0 +1,156 @@
+
+/*
+ *   pata-isapnp.c - ISA PnP PATA controller driver.
+ *   Copyright 2005/2006 Red Hat Inc <alan@redhat.com>, all rights reserved.
+ *
+ *   Based in part on ide-pnp.c by Andrey Panin <pazke@donpac.ru>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/isapnp.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/ata.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_isapnp"
+#define DRV_VERSION "0.1.5"
+
+static struct scsi_host_template isapnp_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations isapnp_port_ops = {
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ata_bmdma_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	isapnp_init_one		-	attach an isapnp interface
+ *	@idev: PnP device
+ *	@dev_id: matching detect line
+ *
+ *	Register an ISA bus IDE interface. Such interfaces are PIO 0 and
+ *	non shared IRQ.
+ */
+
+static int isapnp_init_one(struct pnp_dev *idev, const struct pnp_device_id *dev_id)
+{
+	struct ata_probe_ent ae;
+
+	if (pnp_port_valid(idev, 0) == 0)
+		return -ENODEV;
+
+	/* FIXME: Should selected polled PIO here not fail */
+	if (pnp_irq_valid(idev, 0) == 0)
+		return -ENODEV;
+
+	memset(&ae, 0, sizeof(struct ata_probe_ent));
+	INIT_LIST_HEAD(&ae.node);
+	ae.dev = &idev->dev;
+	ae.port_ops = &isapnp_port_ops;
+	ae.sht = &isapnp_sht;
+	ae.n_ports = 1;
+	ae.pio_mask = 1;		/* ISA so PIO 0 cycles */
+	ae.irq = pnp_irq(idev, 0);
+	ae.irq_flags = 0;
+	ae.port_flags = ATA_FLAG_SLAVE_POSS;
+	ae.port[0].cmd_addr = pnp_port_start(idev, 0);
+
+	if (pnp_port_valid(idev, 1) == 0) {
+		ae.port[0].altstatus_addr = pnp_port_start(idev, 1);
+		ae.port[0].ctl_addr = pnp_port_start(idev, 1);
+		ae.port_flags |= ATA_FLAG_SRST;
+	}
+	ata_std_ports(&ae.port[0]);
+
+	if (ata_device_add(&ae) == 0)
+		return -ENODEV;
+	return 0;
+}
+
+/**
+ *	isapnp_remove_one	-	unplug an isapnp interface
+ *	@idev: PnP device
+ *
+ *	Remove a previously configured PnP ATA port. Called only on module
+ *	unload events as the core does not currently deal with ISAPnP docking.
+ */
+
+static void isapnp_remove_one(struct pnp_dev *idev)
+{
+	struct device *dev = &idev->dev;
+	struct ata_host *host = dev_get_drvdata(dev);
+
+	ata_host_remove(host);
+	dev_set_drvdata(dev, NULL);
+}
+
+static struct pnp_device_id isapnp_devices[] = {
+  	/* Generic ESDI/IDE/ATA compatible hard disk controller */
+	{.id = "PNP0600", .driver_data = 0},
+	{.id = ""}
+};
+
+static struct pnp_driver isapnp_driver = {
+	.name		= DRV_NAME,
+	.id_table	= isapnp_devices,
+	.probe		= isapnp_init_one,
+	.remove		= isapnp_remove_one,
+};
+
+static int __init isapnp_init(void)
+{
+	return pnp_register_driver(&isapnp_driver);
+}
+
+static void __exit isapnp_exit(void)
+{
+	pnp_unregister_driver(&isapnp_driver);
+}
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for ISA PnP ATA");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+
+module_init(isapnp_init);
+module_exit(isapnp_exit);
diff --git a/drivers/ata/pata_it8172.c b/drivers/ata/pata_it8172.c
new file mode 100644
index 000000000000..f8367191fc7b
--- /dev/null
+++ b/drivers/ata/pata_it8172.c
@@ -0,0 +1,288 @@
+/*
+ * pata_it8172.c 	- IT8172 PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based heavily on
+ *
+ * BRIEF MODULE DESCRIPTION
+ *      IT8172 IDE controller support
+ *
+ * Copyright 2000 MontaVista Software Inc.
+ * Author: MontaVista Software, Inc.
+ *              stevel@mvista.com or source@mvista.com
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
+ *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
+ *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
+ *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
+ *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
+ *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
+ *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *  You should have received a copy of the  GNU General Public License along
+ *  with this program; if not, write  to the Free Software Foundation, Inc.,
+ *  675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ *	TODO
+ *	Check for errata
+ *	See if we really need to force native mode
+ *	PIO timings (also lacking in original)
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_it8172"
+#define DRV_VERSION "0.3.1"
+
+static int it8172_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits it8172_enable_bits[] = {
+		{ 0x00, 0, 0x00, 0x00 },
+		{ 0x40, 1, 0x00, 0x01 }
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	if (ap->port_no && !pci_test_config_bits(pdev, &it8172_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+static void it8172_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, it8172_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	it8172_set_pio_timing	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called by both the pio and dma setup functions to set the controller
+ *	timings for PIO transfers. We must load both the mode number and
+ *	timing values into the controller.
+ */
+
+static void it8172_set_pio_timing(struct ata_port *ap, struct ata_device *adev, int pio)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u16 reg40;
+
+	pci_read_config_word(pdev, 0x40, &reg40);
+
+	/*
+	 * FIX! The DIOR/DIOW pulse width and recovery times in port 0x44
+	 * are being left at the default values of 8 PCI clocks (242 nsec
+	 * for a 33 MHz clock). These can be safely shortened at higher
+	 * PIO modes. The DIOR/DIOW pulse width and recovery times only
+	 * apply to PIO modes, not to the DMA modes.
+	 */
+
+	/*
+	 * Enable port 0x44. The IT8172G spec is confused; it calls
+	 * this register the "Slave IDE Timing Register", but in fact,
+	 * it controls timing for both master and slave drives.
+	 */
+
+	reg40 |= 0x4000;
+	if (adev->devno) {
+		reg40 &= 0xC006;
+		if (pio > 1)
+			/* Enable prefetch and IORDY sample-point */
+			reg40 |= 0x0060;
+	} else {
+		reg40 &= 0xC060;
+		if (pio > 1)
+			/* Enable prefetch and IORDY sample-point */
+			reg40 |= 0x0006;
+	}
+	/* Write back the enables */
+	pci_write_config_word(pdev, 0x40, reg40);
+}
+
+/**
+ *	it8172_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup. We use a shared helper for this
+ *	as the DMA setup must also adjust the PIO timing information.
+ */
+
+static void it8172_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	it8172_set_pio_timing(ap, adev, adev->pio_mode - XFER_PIO_0);
+}
+
+/**
+ *	it8172_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the DMA mode setup. We must tune an appropriate PIO
+ *	mode to match.
+ */
+
+static void it8172_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int dn = (2 * ap->port_no) + adev->devno;
+	u8 reg48, reg4a;
+	int pio;
+
+	static const int pio_map[] = { 1, 3, 4};
+	/*
+	 * Setting the DMA cycle time to 2 or 3 PCI clocks (60 and 91 nsec
+	 * at 33 MHz PCI clock) seems to cause BadCRC errors during DMA
+	 * transfers on some drives, even though both numbers meet the minimum
+	 * ATAPI-4 spec of 73 and 54 nsec for UDMA 1 and 2 respectively.
+	 * So the faster times are just commented out here. The good news is
+	 * that the slower cycle time has very little affect on transfer
+	 * performance.
+	 */
+
+	pci_read_config_byte(pdev, 0x48, &reg48);
+	pci_read_config_byte(pdev, 0x4A, &reg4a);
+
+ 	reg4a &= ~(3 << (4 * dn));
+
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		reg48 |= 1 << dn;
+#ifdef UDMA_TIMING_SET
+		reg4a |= ((adev->dma_mode - XFER_UDMA_0) << (4 * dn));
+#endif
+	 	pio = 4;
+	} else  {
+		pio = pio_map[adev->dma_mode - XFER_MW_DMA_0];
+		reg48 &= ~ (1 << dn);
+	}
+	pci_write_config_byte(pdev, 0x48, reg48);
+	pci_write_config_byte(pdev, 0x4A, reg4a);
+ 	it8172_set_pio_timing(ap, adev, pio);
+
+}
+
+static struct scsi_host_template it8172_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations it8172_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= it8172_set_piomode,
+	.set_dmamode	= it8172_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= it8172_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int it8172_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &it8172_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x06,	/* No MWDMA0 support */
+		.udma_mask = 0x7,
+		.port_ops = &it8172_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+
+        if ((!(PCI_FUNC(dev->devfn) & 1) ||
+            (!((dev->class >> 8) == PCI_CLASS_STORAGE_IDE))))
+		return -ENODEV; /* IT8172 is more than an IDE controller */
+
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id it8172[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_IT8172G), },
+	{ 0, },
+};
+
+static struct pci_driver it8172_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= it8172,
+	.probe 		= it8172_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init it8172_init(void)
+{
+	return pci_register_driver(&it8172_pci_driver);
+}
+
+
+static void __exit it8172_exit(void)
+{
+	pci_unregister_driver(&it8172_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for ITE IT8172");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, it8172);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(it8172_init);
+module_exit(it8172_exit);
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
new file mode 100644
index 000000000000..c8a7798f8ce5
--- /dev/null
+++ b/drivers/ata/pata_it821x.c
@@ -0,0 +1,847 @@
+/*
+ * ata-it821x.c 	- IT821x PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based upon
+ *
+ * it821x.c
+ * 
+ * linux/drivers/ide/pci/it821x.c		Version 0.09	December 2004
+ *
+ * Copyright (C) 2004		Red Hat <alan@redhat.com>
+ *
+ *  May be copied or modified under the terms of the GNU General Public License
+ *  Based in part on the ITE vendor provided SCSI driver.
+ *
+ *  Documentation available from
+ * 	http://www.ite.com.tw/pc/IT8212F_V04.pdf
+ *  Some other documents are NDA.
+ *
+ *  The ITE8212 isn't exactly a standard IDE controller. It has two
+ *  modes. In pass through mode then it is an IDE controller. In its smart
+ *  mode its actually quite a capable hardware raid controller disguised
+ *  as an IDE controller. Smart mode only understands DMA read/write and
+ *  identify, none of the fancier commands apply. The IT8211 is identical
+ *  in other respects but lacks the raid mode.
+ *
+ *  Errata:
+ *  o	Rev 0x10 also requires master/slave hold the same DMA timings and
+ *	cannot do ATAPI MWDMA.
+ *  o	The identify data for raid volumes lacks CHS info (technically ok)
+ *	but also fails to set the LBA28 and other bits. We fix these in
+ *	the IDE probe quirk code.
+ *  o	If you write LBA48 sized I/O's (ie > 256 sector) in smart mode
+ *	raid then the controller firmware dies
+ *  o	Smart mode without RAID doesn't clear all the necessary identify
+ *	bits to reduce the command set to the one used
+ *
+ *  This has a few impacts on the driver
+ *  - In pass through mode we do all the work you would expect
+ *  - In smart mode the clocking set up is done by the controller generally
+ *    but we must watch the other limits and filter.
+ *  - There are a few extra vendor commands that actually talk to the
+ *    controller but only work PIO with no IRQ.
+ *
+ *  Vendor areas of the identify block in smart mode are used for the
+ *  timing and policy set up. Each HDD in raid mode also has a serial
+ *  block on the disk. The hardware extra commands are get/set chip status,
+ *  rebuild, get rebuild status.
+ *
+ *  In Linux the driver supports pass through mode as if the device was
+ *  just another IDE controller. If the smart mode is running then
+ *  volumes are managed by the controller firmware and each IDE "disk"
+ *  is a raid volume. Even more cute - the controller can do automated
+ *  hotplug and rebuild.
+ *
+ *  The pass through controller itself is a little demented. It has a
+ *  flaw that it has a single set of PIO/MWDMA timings per channel so
+ *  non UDMA devices restrict each others performance. It also has a
+ *  single clock source per channel so mixed UDMA100/133 performance
+ *  isn't perfect and we have to pick a clock. Thankfully none of this
+ *  matters in smart mode. ATAPI DMA is not currently supported.
+ *
+ *  It seems the smart mode is a win for RAID1/RAID10 but otherwise not.
+ *
+ *  TODO
+ *	-	ATAPI and other speed filtering
+ *	-	Command filter in smart mode
+ *	-	RAID configuration ioctls
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+
+#define DRV_NAME "pata_it821x"
+#define DRV_VERSION "0.3.2"
+
+struct it821x_dev
+{
+	unsigned int smart:1,		/* Are we in smart raid mode */
+		timing10:1;		/* Rev 0x10 */
+	u8	clock_mode;		/* 0, ATA_50 or ATA_66 */
+	u8	want[2][2];		/* Mode/Pri log for master slave */
+	/* We need these for switching the clock when DMA goes on/off
+	   The high byte is the 66Mhz timing */
+	u16	pio[2];			/* Cached PIO values */
+	u16	mwdma[2];		/* Cached MWDMA values */
+	u16	udma[2];		/* Cached UDMA values (per drive) */
+	u16	last_device;		/* Master or slave loaded ? */
+};
+
+#define ATA_66		0
+#define ATA_50		1
+#define ATA_ANY		2
+
+#define UDMA_OFF	0
+#define MWDMA_OFF	0
+
+/*
+ *	We allow users to force the card into non raid mode without
+ *	flashing the alternative BIOS. This is also neccessary right now
+ *	for embedded platforms that cannot run a PC BIOS but are using this
+ *	device.
+ */
+
+static int it8212_noraid;
+
+/**
+ *	it821x_pre_reset	-	probe
+ *	@ap: ATA port
+ *
+ *	Set the cable type
+ */
+ 
+static int it821x_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	it821x_error_handler	-	probe/reset
+ *	@ap: ATA port
+ *
+ *	Set the cable type and trigger a probe
+ */
+ 
+static void it821x_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, it821x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	it821x_program	-	program the PIO/MWDMA registers
+ *	@ap: ATA port
+ *	@adev: Device to program
+ *	@timing: Timing value (66Mhz in top 8bits, 50 in the low 8)
+ *
+ *	Program the PIO/MWDMA timing for this channel according to the
+ *	current clock. These share the same register so are managed by
+ *	the DMA start/stop sequence as with the old driver.
+ */
+
+static void it821x_program(struct ata_port *ap, struct ata_device *adev, u16 timing)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct it821x_dev *itdev = ap->private_data;
+	int channel = ap->port_no;
+	u8 conf;
+
+	/* Program PIO/MWDMA timing bits */
+	if (itdev->clock_mode == ATA_66)
+		conf = timing >> 8;
+	else
+		conf = timing & 0xFF;
+	pci_write_config_byte(pdev, 0x54 + 4 * channel, conf);
+}
+
+
+/**
+ *	it821x_program_udma	-	program the UDMA registers
+ *	@ap: ATA port
+ *	@adev: ATA device to update
+ *	@timing: Timing bits. Top 8 are for 66Mhz bottom for 50Mhz
+ *
+ *	Program the UDMA timing for this drive according to the
+ *	current clock. Handles the dual clocks and also knows about
+ *	the errata on the 0x10 revision. The UDMA errata is partly handled
+ *	here and partly in start_dma.
+ */
+
+static void it821x_program_udma(struct ata_port *ap, struct ata_device *adev, u16 timing)
+{
+	struct it821x_dev *itdev = ap->private_data;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int channel = ap->port_no;
+	int unit = adev->devno;
+	u8 conf;
+
+	/* Program UDMA timing bits */
+	if (itdev->clock_mode == ATA_66)
+		conf = timing >> 8;
+	else
+		conf = timing & 0xFF;
+	if (itdev->timing10 == 0)
+		pci_write_config_byte(pdev, 0x56 + 4 * channel + unit, conf);
+	else {
+		/* Early revision must be programmed for both together */
+		pci_write_config_byte(pdev, 0x56 + 4 * channel, conf);
+		pci_write_config_byte(pdev, 0x56 + 4 * channel + 1, conf);
+	}
+}
+
+/**
+ *	it821x_clock_strategy
+ *	@ap: ATA interface
+ *	@adev: ATA device being updated
+ *
+ *	Select between the 50 and 66Mhz base clocks to get the best
+ *	results for this interface.
+ */
+
+static void it821x_clock_strategy(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct it821x_dev *itdev = ap->private_data;
+	u8 unit = adev->devno;
+	struct ata_device *pair = ata_dev_pair(adev);
+
+	int clock, altclock;
+	u8 v;
+	int sel = 0;
+
+	/* Look for the most wanted clocking */
+	if (itdev->want[0][0] > itdev->want[1][0]) {
+		clock = itdev->want[0][1];
+		altclock = itdev->want[1][1];
+	} else {
+		clock = itdev->want[1][1];
+		altclock = itdev->want[0][1];
+	}
+
+	/* Master doesn't care does the slave ? */
+	if (clock == ATA_ANY)
+		clock = altclock;
+
+	/* Nobody cares - keep the same clock */
+	if (clock == ATA_ANY)
+		return;
+	/* No change */
+	if (clock == itdev->clock_mode)
+		return;
+
+	/* Load this into the controller */
+	if (clock == ATA_66)
+		itdev->clock_mode = ATA_66;
+	else {
+		itdev->clock_mode = ATA_50;
+		sel = 1;
+	}
+	pci_read_config_byte(pdev, 0x50, &v);
+	v &= ~(1 << (1 + ap->port_no));
+	v |= sel << (1 + ap->port_no);
+	pci_write_config_byte(pdev, 0x50, v);
+
+	/*
+	 *	Reprogram the UDMA/PIO of the pair drive for the switch
+	 *	MWDMA will be dealt with by the dma switcher
+	 */
+	if (pair && itdev->udma[1-unit] != UDMA_OFF) {
+		it821x_program_udma(ap, pair, itdev->udma[1-unit]);
+		it821x_program(ap, pair, itdev->pio[1-unit]);
+	}
+	/*
+	 *	Reprogram the UDMA/PIO of our drive for the switch.
+	 *	MWDMA will be dealt with by the dma switcher
+	 */
+	if (itdev->udma[unit] != UDMA_OFF) {
+		it821x_program_udma(ap, adev, itdev->udma[unit]);
+		it821x_program(ap, adev, itdev->pio[unit]);
+	}
+}
+
+/**
+ *	it821x_passthru_set_piomode	-	set PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Configure for PIO mode. This is complicated as the register is
+ *	shared by PIO and MWDMA and for both channels.
+ */
+
+static void it821x_passthru_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	/* Spec says 89 ref driver uses 88 */
+	static const u16 pio[]	= { 0xAA88, 0xA382, 0xA181, 0x3332, 0x3121 };
+	static const u8 pio_want[]    = { ATA_66, ATA_66, ATA_66, ATA_66, ATA_ANY };
+
+	struct it821x_dev *itdev = ap->private_data;
+	int unit = adev->devno;
+	int mode_wanted = adev->pio_mode - XFER_PIO_0;
+	
+	/* We prefer 66Mhz clock for PIO 0-3, don't care for PIO4 */
+	itdev->want[unit][1] = pio_want[mode_wanted];
+	itdev->want[unit][0] = 1;	/* PIO is lowest priority */
+	itdev->pio[unit] = pio[mode_wanted];
+	it821x_clock_strategy(ap, adev);
+	it821x_program(ap, adev, itdev->pio[unit]);
+}
+
+/**
+ *	it821x_passthru_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Set up the DMA modes. The actions taken depend heavily on the mode
+ *	to use. If UDMA is used as is hopefully the usual case then the 
+ *	timing register is private and we need only consider the clock. If
+ *	we are using MWDMA then we have to manage the setting ourself as
+ *	we switch devices and mode.
+ */
+
+static void it821x_passthru_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u16 dma[]	= 	{ 0x8866, 0x3222, 0x3121 };
+	static const u8 mwdma_want[] =  { ATA_ANY, ATA_66, ATA_ANY };
+	static const u16 udma[]	= 	{ 0x4433, 0x4231, 0x3121, 0x2121, 0x1111, 0x2211, 0x1111 };
+	static const u8 udma_want[] =   { ATA_ANY, ATA_50, ATA_ANY, ATA_66, ATA_66, ATA_50, ATA_66 };
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct it821x_dev *itdev = ap->private_data;
+	int channel = ap->port_no;
+	int unit = adev->devno;
+	u8 conf;
+
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		int mode_wanted = adev->dma_mode - XFER_UDMA_0;
+		
+		itdev->want[unit][1] = udma_want[mode_wanted];
+		itdev->want[unit][0] = 3;	/* UDMA is high priority */
+		itdev->mwdma[unit] = MWDMA_OFF;
+		itdev->udma[unit] = udma[mode_wanted];
+		if (mode_wanted >= 5)
+			itdev->udma[unit] |= 0x8080;	/* UDMA 5/6 select on */
+
+		/* UDMA on. Again revision 0x10 must do the pair */
+		pci_read_config_byte(pdev, 0x50, &conf);
+		if (itdev->timing10)
+			conf &= channel ? 0x9F: 0xE7;
+		else
+			conf &= ~ (1 << (3 + 2 * channel + unit));
+		pci_write_config_byte(pdev, 0x50, conf);
+		it821x_clock_strategy(ap, adev);
+		it821x_program_udma(ap, adev, itdev->udma[unit]);
+	} else {
+		int mode_wanted = adev->dma_mode - XFER_MW_DMA_0;
+		
+		itdev->want[unit][1] = mwdma_want[mode_wanted];
+		itdev->want[unit][0] = 2;	/* MWDMA is low priority */
+		itdev->mwdma[unit] = dma[mode_wanted];
+		itdev->udma[unit] = UDMA_OFF;
+
+		/* UDMA bits off - Revision 0x10 do them in pairs */
+		pci_read_config_byte(pdev, 0x50, &conf);
+		if (itdev->timing10)
+			conf |= channel ? 0x60: 0x18;
+		else
+			conf |= 1 << (3 + 2 * channel + unit);
+		pci_write_config_byte(pdev, 0x50, conf);
+		it821x_clock_strategy(ap, adev);
+	}
+}
+
+/**
+ *	it821x_passthru_dma_start	-	DMA start callback
+ *	@qc: Command in progress
+ *
+ *	Usually drivers set the DMA timing at the point the set_dmamode call
+ *	is made. IT821x however requires we load new timings on the 
+ *	transitions in some cases.
+ */
+
+static void it821x_passthru_bmdma_start(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct it821x_dev *itdev = ap->private_data;
+	int unit = adev->devno;
+
+	if (itdev->mwdma[unit] != MWDMA_OFF)
+		it821x_program(ap, adev, itdev->mwdma[unit]);
+	else if (itdev->udma[unit] != UDMA_OFF && itdev->timing10)
+		it821x_program_udma(ap, adev, itdev->udma[unit]);
+	ata_bmdma_start(qc);
+}
+
+/**
+ *	it821x_passthru_dma_stop	-	DMA stop callback
+ *	@qc: ATA command
+ *
+ *	We loaded new timings in dma_start, as a result we need to restore
+ *	the PIO timings in dma_stop so that the next command issue gets the
+ *	right clock values.
+ */
+
+static void it821x_passthru_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct it821x_dev *itdev = ap->private_data;
+	int unit = adev->devno;
+
+	ata_bmdma_stop(qc);
+	if (itdev->mwdma[unit] != MWDMA_OFF)
+		it821x_program(ap, adev, itdev->pio[unit]);
+}
+
+
+/**
+ *	it821x_passthru_dev_select	-	Select master/slave
+ *	@ap: ATA port
+ *	@device: Device number (not pointer)
+ *
+ *	Device selection hook. If neccessary perform clock switching
+ */
+ 
+static void it821x_passthru_dev_select(struct ata_port *ap,
+				       unsigned int device)
+{
+	struct it821x_dev *itdev = ap->private_data;
+	if (itdev && device != itdev->last_device) {
+		struct ata_device *adev = &ap->device[device];
+		it821x_program(ap, adev, itdev->pio[adev->devno]);
+		itdev->last_device = device;
+	}
+	ata_std_dev_select(ap, device);
+}
+
+/**
+ *	it821x_smart_qc_issue_prot	-	wrap qc issue prot
+ *	@qc: command
+ *
+ *	Wrap the command issue sequence for the IT821x. We need to
+ *	perform out own device selection timing loads before the
+ *	usual happenings kick off
+ */
+ 
+static unsigned int it821x_smart_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	switch(qc->tf.command)
+	{
+		/* Commands the firmware supports */
+		case ATA_CMD_READ:
+		case ATA_CMD_READ_EXT:
+		case ATA_CMD_WRITE:
+		case ATA_CMD_WRITE_EXT:
+		case ATA_CMD_PIO_READ:
+		case ATA_CMD_PIO_READ_EXT:
+		case ATA_CMD_PIO_WRITE:
+		case ATA_CMD_PIO_WRITE_EXT:
+		case ATA_CMD_READ_MULTI:
+		case ATA_CMD_READ_MULTI_EXT:
+		case ATA_CMD_WRITE_MULTI:
+		case ATA_CMD_WRITE_MULTI_EXT:
+		case ATA_CMD_ID_ATA:
+		/* Arguably should just no-op this one */
+		case ATA_CMD_SET_FEATURES:
+			return ata_qc_issue_prot(qc);
+	}
+	printk(KERN_DEBUG "it821x: can't process command 0x%02X\n", qc->tf.command);
+	return AC_ERR_INVALID;
+}
+
+/**
+ *	it821x_passthru_qc_issue_prot	-	wrap qc issue prot
+ *	@qc: command
+ *
+ *	Wrap the command issue sequence for the IT821x. We need to
+ *	perform out own device selection timing loads before the
+ *	usual happenings kick off
+ */
+ 
+static unsigned int it821x_passthru_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	it821x_passthru_dev_select(qc->ap, qc->dev->devno);
+	return ata_qc_issue_prot(qc);
+}
+
+/**
+ *	it821x_smart_set_mode	-	mode setting
+ *	@ap: interface to set up
+ *
+ *	Use a non standard set_mode function. We don't want to be tuned.
+ *	The BIOS configured everything. Our job is not to fiddle. We
+ *	read the dma enabled bits from the PCI configuration of the device
+ *	and respect them. 
+ */
+ 
+static void it821x_smart_set_mode(struct ata_port *ap)
+{
+	int dma_enabled = 0;
+	int i;
+
+	/* Bits 5 and 6 indicate if DMA is active on master/slave */
+	/* It is possible that BMDMA isn't allocated */
+	if (ap->ioaddr.bmdma_addr)
+		dma_enabled = inb(ap->ioaddr.bmdma_addr + ATA_DMA_CMD);
+	
+	for (i = 0; i < ATA_MAX_DEVICES; i++) {
+		struct ata_device *dev = &ap->device[i];
+		if (ata_dev_enabled(dev)) {
+			/* We don't really care */
+			dev->pio_mode = XFER_PIO_0;
+			dev->dma_mode = XFER_MW_DMA_0;
+			/* We do need the right mode information for DMA or PIO 
+			   and this comes from the current configuration flags */
+			if (dma_enabled & (1 << (5 + i))) {
+				dev->xfer_mode = XFER_MW_DMA_0;
+				dev->xfer_shift = ATA_SHIFT_MWDMA;
+				dev->flags &= ~ATA_DFLAG_PIO;
+			} else {
+				dev->xfer_mode = XFER_PIO_0;
+				dev->xfer_shift = ATA_SHIFT_PIO;
+				dev->flags |= ATA_DFLAG_PIO;
+			}
+		}
+	}
+}
+
+/**
+ *	it821x_dev_config	-	Called each device identify
+ *	@ap: ATA port
+ *	@adev: Device that has just been identified
+ *
+ *	Perform the initial setup needed for each device that is chip
+ *	special. In our case we need to lock the sector count to avoid
+ *	blowing the brains out of the firmware with large LBA48 requests
+ *
+ *	FIXME: When FUA appears we need to block FUA too. And SMART and
+ *	basically we need to filter commands for this chip.
+ */
+ 
+static void it821x_dev_config(struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned char model_num[40];
+	char *s;
+	unsigned int len;
+
+	/* This block ought to be a library routine as it is in several
+	   drivers now */
+	   
+	ata_id_string(adev->id, model_num, ATA_ID_PROD_OFS,
+			  sizeof(model_num));
+	s = &model_num[0];
+	len = strnlen(s, sizeof(model_num));
+
+	/* ATAPI specifies that empty space is blank-filled; remove blanks */
+	while ((len > 0) && (s[len - 1] == ' ')) {
+		len--;
+		s[len] = 0;
+	}
+
+	if (adev->max_sectors > 255)
+		adev->max_sectors = 255;
+	
+	if (strstr(model_num, "Integrated Technology Express")) {
+		/* RAID mode */
+		printk(KERN_INFO "IT821x %sRAID%d volume",
+			adev->id[147]?"Bootable ":"",
+			adev->id[129]);
+		if (adev->id[129] != 1)
+			printk("(%dK stripe)", adev->id[146]);
+		printk(".\n");
+	}
+}
+
+
+/**
+ *	it821x_check_atapi_dma	-	ATAPI DMA handler
+ *	@qc: Command we are about to issue
+ *
+ *	Decide if this ATAPI command can be issued by DMA on this
+ *	controller. Return 0 if it can be.
+ */
+ 
+static int it821x_check_atapi_dma(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct it821x_dev *itdev = ap->private_data;
+	
+	/* No ATAPI DMA in smart mode */
+	if (itdev->smart)
+		return -EOPNOTSUPP;
+	/* No ATAPI DMA on rev 10 */
+	if (itdev->timing10)
+		return -EOPNOTSUPP;
+	/* Cool */
+	return 0;
+}
+		
+
+/**
+ *	it821x_port_start	-	port setup
+ *	@ap: ATA port being set up
+ *
+ *	The it821x needs to maintain private data structures and also to
+ *	use the standard PCI interface which lacks support for this
+ *	functionality. We instead set up the private data on the port 
+ *	start hook, and tear it down on port stop
+ */
+ 
+static int it821x_port_start(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct it821x_dev *itdev;
+	u8 conf;
+
+	int ret = ata_port_start(ap);
+	if (ret < 0)
+		return ret;
+		
+	ap->private_data = kmalloc(sizeof(struct it821x_dev), GFP_KERNEL);
+	if (ap->private_data == NULL) {
+		ata_port_stop(ap);
+		return -ENOMEM;
+	}
+	
+	itdev = ap->private_data;
+	memset(itdev, 0, sizeof(struct it821x_dev));
+
+	pci_read_config_byte(pdev, 0x50, &conf);
+
+	if (conf & 1) {
+		itdev->smart = 1;
+		/* Long I/O's although allowed in LBA48 space cause the
+		   onboard firmware to enter the twighlight zone */
+		/* No ATAPI DMA in this mode either */
+	}
+	/* Pull the current clocks from 0x50 */
+	if (conf & (1 << (1 + ap->port_no)))
+		itdev->clock_mode = ATA_50;
+	else
+		itdev->clock_mode = ATA_66;
+
+	itdev->want[0][1] = ATA_ANY;
+	itdev->want[1][1] = ATA_ANY;
+	itdev->last_device = -1;
+
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &conf);
+	if (conf == 0x10) {
+		itdev->timing10 = 1;
+		/* Need to disable ATAPI DMA for this case */
+		if (!itdev->smart)
+			printk(KERN_WARNING DRV_NAME": Revision 0x10, workarounds activated.\n");
+	}
+
+	return 0;
+}
+
+/**
+ *	it821x_port_stop	-	port shutdown
+ *	@ap: ATA port being removed
+ *
+ *	Release the private objects we added in it821x_port_start
+ */
+ 
+static void it821x_port_stop(struct ata_port *ap) {
+	kfree(ap->private_data);
+	ap->private_data = NULL;	/* We want an OOPS if we reuse this
+					   too late! */
+	ata_port_stop(ap);
+}
+
+static struct scsi_host_template it821x_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	/* 255 sectors to begin with. This is locked in smart mode but not
+	   in pass through */
+	.max_sectors		= 255,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations it821x_smart_port_ops = {
+	.set_mode	= it821x_smart_set_mode,
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.check_status 	= ata_check_status,
+	.check_atapi_dma= it821x_check_atapi_dma,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+	.dev_config	= it821x_dev_config,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= it821x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= it821x_smart_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= it821x_port_start,
+	.port_stop	= it821x_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static struct ata_port_operations it821x_passthru_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= it821x_passthru_set_piomode,
+	.set_dmamode	= it821x_passthru_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.check_atapi_dma= it821x_check_atapi_dma,
+	.dev_select 	= it821x_passthru_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= it821x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= it821x_passthru_bmdma_start,
+	.bmdma_stop	= it821x_passthru_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= it821x_passthru_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_clear	= ata_bmdma_irq_clear,
+	.irq_handler	= ata_interrupt,
+
+	.port_start	= it821x_port_start,
+	.port_stop	= it821x_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static void __devinit it821x_disable_raid(struct pci_dev *pdev)
+{
+	/* Reset local CPU, and set BIOS not ready */
+	pci_write_config_byte(pdev, 0x5E, 0x01);
+
+	/* Set to bypass mode, and reset PCI bus */
+	pci_write_config_byte(pdev, 0x50, 0x00);
+	pci_write_config_word(pdev, PCI_COMMAND,
+			      PCI_COMMAND_PARITY | PCI_COMMAND_IO |
+			      PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER);
+	pci_write_config_word(pdev, 0x40, 0xA0F3);
+
+	pci_write_config_dword(pdev,0x4C, 0x02040204);
+	pci_write_config_byte(pdev, 0x42, 0x36);
+	pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x20);
+}
+
+	
+static int it821x_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	u8 conf;
+	
+	static struct ata_port_info info_smart = {
+		.sht = &it821x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &it821x_smart_port_ops
+	};
+	static struct ata_port_info info_passthru = {
+		.sht = &it821x_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &it821x_passthru_port_ops
+	};
+	static struct ata_port_info *port_info[2];
+	
+	static char *mode[2] = { "pass through", "smart" };
+
+	/* Force the card into bypass mode if so requested */
+	if (it8212_noraid) {
+		printk(KERN_INFO DRV_NAME ": forcing bypass mode.\n");
+		it821x_disable_raid(pdev);
+	}
+	pci_read_config_byte(pdev, 0x50, &conf);
+	conf &= 1;
+	
+	printk(KERN_INFO DRV_NAME ": controller in %s mode.\n", mode[conf]);
+	if (conf == 0)
+		port_info[0] = port_info[1] = &info_passthru;
+	else
+		port_info[0] = port_info[1] = &info_smart;
+	
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static struct pci_device_id it821x[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_8211), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ITE, PCI_DEVICE_ID_ITE_8212), },
+	{ 0, },
+};
+
+static struct pci_driver it821x_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= it821x,
+	.probe 		= it821x_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init it821x_init(void)
+{
+	return pci_register_driver(&it821x_pci_driver);
+}
+
+
+static void __exit it821x_exit(void)
+{
+	pci_unregister_driver(&it821x_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for the IT8211/IT8212 IDE RAID controller");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, it821x);
+MODULE_VERSION(DRV_VERSION);
+
+
+module_param_named(noraid, it8212_noraid, int, S_IRUGO);
+MODULE_PARM_DESC(it8212_noraid, "Force card into bypass mode");
+
+module_init(it821x_init);
+module_exit(it821x_exit);
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
new file mode 100644
index 000000000000..6832a643a9eb
--- /dev/null
+++ b/drivers/ata/pata_jmicron.c
@@ -0,0 +1,266 @@
+/*
+ *    pata_jmicron.c - JMicron ATA driver for non AHCI mode. This drives the
+ *			PATA port of the controller. The SATA ports are
+ *			driven by AHCI in the usual configuration although
+ *			this driver can handle other setups if we need it.
+ *
+ *	(c) 2006 Red Hat  <alan@redhat.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_jmicron"
+#define DRV_VERSION	"0.1.2"
+
+typedef enum {
+	PORT_PATA0 = 0,
+	PORT_PATA1 = 1,
+	PORT_SATA = 2,
+} port_type;
+
+/**
+ *	jmicron_pre_reset	-	check for 40/80 pin
+ *	@ap: Port
+ *
+ *	Perform the PATA port setup we need.
+
+ *	On the Jmicron 361/363 there is a single PATA port that can be mapped
+ *	either as primary or secondary (or neither). We don't do any policy
+ *	and setup here. We assume that has been done by init_one and the
+ *	BIOS.
+ */
+
+static int jmicron_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 control;
+	u32 control5;
+	int port_mask = 1<< (4 * ap->port_no);
+	int port = ap->port_no;
+	port_type port_map[2];
+
+	/* Check if our port is enabled */
+	pci_read_config_dword(pdev, 0x40, &control);
+	if ((control & port_mask) == 0)
+		return 0;
+
+	/* There are two basic mappings. One has the two SATA ports merged
+	   as master/slave and the secondary as PATA, the other has only the
+	   SATA port mapped */
+	if (control & (1 << 23)) {
+		port_map[0] = PORT_SATA;
+		port_map[1] = PORT_PATA0;
+	} else {
+		port_map[0] = PORT_SATA;
+		port_map[1] = PORT_SATA;
+	}
+
+	/* The 365/366 may have this bit set to map the second PATA port
+	   as the internal primary channel */
+	pci_read_config_dword(pdev, 0x80, &control5);
+	if (control5 & (1<<24))
+		port_map[0] = PORT_PATA1;
+
+	/* The two ports may then be logically swapped by the firmware */
+	if (control & (1 << 22))
+		port = port ^ 1;
+
+	/*
+	 *	Now we know which physical port we are talking about we can
+	 *	actually do our cable checking etc. Thankfully we don't need
+	 *	to do the plumbing for other cases.
+	 */
+	switch (port_map[port])
+	{
+	case PORT_PATA0:
+		if (control & (1 << 5))
+			return 0;
+		if (control & (1 << 3))	/* 40/80 pin primary */
+			ap->cbl = ATA_CBL_PATA40;
+		else
+			ap->cbl = ATA_CBL_PATA80;
+		break;
+	case PORT_PATA1:
+		/* Bit 21 is set if the port is enabled */
+		if ((control5 & (1 << 21)) == 0)
+			return 0;
+		if (control5 & (1 << 19))	/* 40/80 pin secondary */
+			ap->cbl = ATA_CBL_PATA40;
+		else
+			ap->cbl = ATA_CBL_PATA80;
+		break;
+	case PORT_SATA:
+		ap->cbl = ATA_CBL_SATA;
+		break;
+	}
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	jmicron_error_handler - Setup and error handler
+ *	@ap: Port to handle
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void jmicron_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, jmicron_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/* No PIO or DMA methods needed for this device */
+
+static struct scsi_host_template jmicron_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	/* Special handling needed if you have sector or LBA48 limits */
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	/* Use standard CHS mapping rules */
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations jmicron_ops = {
+	.port_disable		= ata_port_disable,
+
+	/* Task file is PCI ATA format, use helpers */
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= jmicron_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	/* BMDMA handling is PCI ATA format, use helpers */
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	/* Timeout handling. Special recovery hooks here */
+	.eng_timeout		= ata_eng_timeout,
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	/* Generic PATA PCI ATA helpers */
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+
+/**
+ *	jmicron_init_one - Register Jmicron ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in jmicron_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht		= &jmicron_sht,
+		.flags	= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+
+		.pio_mask	= 0x1f,
+		.mwdma_mask	= 0x07,
+		.udma_mask 	= 0x3f,
+
+		.port_ops	= &jmicron_ops,
+	};
+	struct ata_port_info *port_info[2] = { &info, &info };
+
+	u32 reg;
+
+	if (id->driver_data != 368) {
+		/* Put the controller into AHCI mode in case the AHCI driver
+		   has not yet been loaded. This can be done with either
+		   function present */
+
+		/* FIXME: We may want a way to override this in future */
+		pci_write_config_byte(pdev, 0x41, 0xa1);
+	}
+
+	/* PATA controller is fn 1, AHCI is fn 0 */
+	if (PCI_FUNC(pdev->devfn) != 1)
+		return -ENODEV;
+
+	if ( id->driver_data == 365 || id->driver_data == 366) {
+		/* The 365/66 have two PATA channels, redirect the second */
+		pci_read_config_dword(pdev, 0x80, &reg);
+		reg |= (1 << 24);	/* IDE1 to PATA IDE secondary */
+		pci_write_config_dword(pdev, 0x80, reg);
+	}
+
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id jmicron_pci_tbl[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361), 361},
+	{ PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363), 363},
+	{ PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365), 365},
+	{ PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366), 366},
+	{ PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368), 368},
+	{ }	/* terminate list */
+};
+
+static struct pci_driver jmicron_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= jmicron_pci_tbl,
+	.probe			= jmicron_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init jmicron_init(void)
+{
+	return pci_register_driver(&jmicron_pci_driver);
+}
+
+static void __exit jmicron_exit(void)
+{
+	pci_unregister_driver(&jmicron_pci_driver);
+}
+
+module_init(jmicron_init);
+module_exit(jmicron_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for Jmicron PATA ports");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, jmicron_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
new file mode 100644
index 000000000000..eb510660b424
--- /dev/null
+++ b/drivers/ata/pata_legacy.c
@@ -0,0 +1,949 @@
+/*
+ *   pata-legacy.c - Legacy port PATA/SATA controller driver.
+ *   Copyright 2005/2006 Red Hat <alan@redhat.com>, all rights reserved.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ *   An ATA driver for the legacy ATA ports.
+ *
+ *   Data Sources:
+ *	Opti 82C465/82C611 support: Data sheets at opti-inc.com
+ *	HT6560 series:
+ *	Promise 20230/20620:
+ *		http://www.ryston.cz/petr/vlb/pdc20230b.html
+ *		http://www.ryston.cz/petr/vlb/pdc20230c.html
+ *		http://www.ryston.cz/petr/vlb/pdc20630.html
+ *
+ *  Unsupported but docs exist:
+ *	Appian/Adaptec AIC25VL01/Cirrus Logic PD7220
+ *	Winbond W83759A
+ *
+ *  This driver handles legacy (that is "ISA/VLB side") IDE ports found
+ *  on PC class systems. There are three hybrid devices that are exceptions
+ *  The Cyrix 5510/5520 where a pre SFF ATA device is on the bridge and
+ *  the MPIIX where the tuning is PCI side but the IDE is "ISA side".
+ *
+ *  Specific support is included for the ht6560a/ht6560b/opti82c611a/
+ *  opti82c465mv/promise 20230c/20630
+ *
+ *  Use the autospeed and pio_mask options with:
+ *	Appian ADI/2 aka CLPD7220 or AIC25VL01.
+ *  Use the jumpers, autospeed and set pio_mask to the mode on the jumpers with
+ *	Goldstar GM82C711, PIC-1288A-125, UMC 82C871F, Winbond W83759,
+ *	Winbond W83759A, Promise PDC20230-B
+ *
+ *  For now use autospeed and pio_mask as above with the W83759A. This may
+ *  change.
+ *
+ *  TODO
+ *	Merge existing pata_qdi driver
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/ata.h>
+#include <linux/libata.h>
+#include <linux/platform_device.h>
+
+#define DRV_NAME "pata_legacy"
+#define DRV_VERSION "0.5.3"
+
+#define NR_HOST 6
+
+static int legacy_port[NR_HOST] = { 0x1f0, 0x170, 0x1e8, 0x168, 0x1e0, 0x160 };
+static int legacy_irq[NR_HOST] = { 15, 14, 11, 10, 8, 12 };
+
+struct legacy_data {
+	unsigned long timing;
+	u8 clock[2];
+	u8 last;
+	int fast;
+	struct platform_device *platform_dev;
+
+};
+
+static struct legacy_data legacy_data[NR_HOST];
+static struct ata_host *legacy_host[NR_HOST];
+static int nr_legacy_host;
+
+
+static int probe_all;			/* Set to check all ISA port ranges */
+static int ht6560a;			/* HT 6560A on primary 1, secondary 2, both 3 */
+static int ht6560b;			/* HT 6560A on primary 1, secondary 2, both 3 */
+static int opti82c611a;			/* Opti82c611A on primary 1, secondary 2, both 3 */
+static int opti82c46x;		/* Opti 82c465MV present (pri/sec autodetect) */
+static int autospeed;			/* Chip present which snoops speed changes */
+static int pio_mask = 0x1F;		/* PIO range for autospeed devices */
+
+/**
+ *	legacy_set_mode		-	mode setting
+ *	@ap: IDE interface
+ *
+ *	Use a non standard set_mode function. We don't want to be tuned.
+ *
+ *	The BIOS configured everything. Our job is not to fiddle. Just use
+ *	whatever PIO the hardware is using and leave it at that. When we
+ *	get some kind of nice user driven API for control then we can
+ *	expand on this as per hdparm in the base kernel.
+ */
+
+static void legacy_set_mode(struct ata_port *ap)
+{
+	int i;
+
+	for (i = 0; i < ATA_MAX_DEVICES; i++) {
+		struct ata_device *dev = &ap->device[i];
+		if (ata_dev_enabled(dev)) {
+			dev->pio_mode = XFER_PIO_0;
+			dev->xfer_mode = XFER_PIO_0;
+			dev->xfer_shift = ATA_SHIFT_PIO;
+			dev->flags |= ATA_DFLAG_PIO;
+		}
+	}
+}
+
+static struct scsi_host_template legacy_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+/*
+ *	These ops are used if the user indicates the hardware
+ *	snoops the commands to decide on the mode and handles the
+ *	mode selection "magically" itself. Several legacy controllers
+ *	do this. The mode range can be set if it is not 0x1F by setting
+ *	pio_mask as well.
+ */
+
+static struct ata_port_operations simple_port_ops = {
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ata_bmdma_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer_noirq,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations legacy_port_ops = {
+	.set_mode	= legacy_set_mode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer_noirq,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Promise 20230C and 20620 support
+ *
+ *	This controller supports PIO0 to PIO2. We set PIO timings conservatively to
+ *	allow for 50MHz Vesa Local Bus. The 20620 DMA support is weird being DMA to
+ *	controller and PIO'd to the host and not supported.
+ */
+
+static void pdc20230_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	int tries = 5;
+	int pio = adev->pio_mode - XFER_PIO_0;
+	u8 rt;
+	unsigned long flags;
+	
+	/* Safe as UP only. Force I/Os to occur together */
+	
+	local_irq_save(flags);
+	
+	/* Unlock the control interface */
+	do
+	{
+		inb(0x1F5);
+		outb(inb(0x1F2) | 0x80, 0x1F2);
+		inb(0x1F2);
+		inb(0x3F6);
+		inb(0x3F6);
+		inb(0x1F2);
+		inb(0x1F2);
+	}
+	while((inb(0x1F2) & 0x80) && --tries);
+
+	local_irq_restore(flags);
+	
+	outb(inb(0x1F4) & 0x07, 0x1F4);
+
+	rt = inb(0x1F3);
+	rt &= 0x07 << (3 * adev->devno);
+	if (pio)
+		rt |= (1 + 3 * pio) << (3 * adev->devno);
+
+	udelay(100);
+	outb(inb(0x1F2) | 0x01, 0x1F2);
+	udelay(100);
+	inb(0x1F5);
+
+}
+
+static void pdc_data_xfer_vlb(struct ata_device *adev, unsigned char *buf, unsigned int buflen, int write_data)
+{
+	struct ata_port *ap = adev->ap;
+	int slop = buflen & 3;
+	unsigned long flags;
+
+	if (ata_id_has_dword_io(adev->id)) {
+		local_irq_save(flags);
+
+		/* Perform the 32bit I/O synchronization sequence */
+		inb(ap->ioaddr.nsect_addr);
+		inb(ap->ioaddr.nsect_addr);
+		inb(ap->ioaddr.nsect_addr);
+
+		/* Now the data */
+
+		if (write_data)
+			outsl(ap->ioaddr.data_addr, buf, buflen >> 2);
+		else
+			insl(ap->ioaddr.data_addr, buf, buflen >> 2);
+
+		if (unlikely(slop)) {
+			u32 pad;
+			if (write_data) {
+				memcpy(&pad, buf + buflen - slop, slop);
+				outl(le32_to_cpu(pad), ap->ioaddr.data_addr);
+			} else {
+				pad = cpu_to_le16(inl(ap->ioaddr.data_addr));
+				memcpy(buf + buflen - slop, &pad, slop);
+			}
+		}
+		local_irq_restore(flags);
+	}
+	else
+		ata_pio_data_xfer_noirq(adev, buf, buflen, write_data);
+}
+
+static struct ata_port_operations pdc20230_port_ops = {
+	.set_piomode	= pdc20230_set_piomode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= pdc_data_xfer_vlb,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Holtek 6560A support
+ *
+ *	This controller supports PIO0 to PIO2 (no IORDY even though higher timings
+ *	can be loaded).
+ */
+
+static void ht6560a_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	u8 active, recover;
+	struct ata_timing t;
+
+	/* Get the timing data in cycles. For now play safe at 50Mhz */
+	ata_timing_compute(adev, adev->pio_mode, &t, 20000, 1000);
+
+	active = FIT(t.active, 2, 15);
+	recover = FIT(t.recover, 4, 15);
+
+	inb(0x3E6);
+	inb(0x3E6);
+	inb(0x3E6);
+	inb(0x3E6);
+
+	outb(recover << 4 | active, ap->ioaddr.device_addr);
+	inb(ap->ioaddr.status_addr);
+}
+
+static struct ata_port_operations ht6560a_port_ops = {
+	.set_piomode	= ht6560a_set_piomode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,	/* Check vlb/noirq */
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Holtek 6560B support
+ *
+ *	This controller supports PIO0 to PIO4. We honour the BIOS/jumper FIFO setting
+ *	unless we see an ATAPI device in which case we force it off.
+ *
+ *	FIXME: need to implement 2nd channel support.
+ */
+
+static void ht6560b_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	u8 active, recover;
+	struct ata_timing t;
+
+	/* Get the timing data in cycles. For now play safe at 50Mhz */
+	ata_timing_compute(adev, adev->pio_mode, &t, 20000, 1000);
+
+	active = FIT(t.active, 2, 15);
+	recover = FIT(t.recover, 2, 16);
+	recover &= 0x15;
+
+	inb(0x3E6);
+	inb(0x3E6);
+	inb(0x3E6);
+	inb(0x3E6);
+
+	outb(recover << 4 | active, ap->ioaddr.device_addr);
+
+	if (adev->class != ATA_DEV_ATA) {
+		u8 rconf = inb(0x3E6);
+		if (rconf & 0x24) {
+			rconf &= ~ 0x24;
+			outb(rconf, 0x3E6);
+		}
+	}
+	inb(ap->ioaddr.status_addr);
+}
+
+static struct ata_port_operations ht6560b_port_ops = {
+	.set_piomode	= ht6560b_set_piomode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,	/* FIXME: Check 32bit and noirq */
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Opti core chipset helpers
+ */
+ 
+/**
+ *	opti_syscfg	-	read OPTI chipset configuration
+ *	@reg: Configuration register to read
+ *
+ *	Returns the value of an OPTI system board configuration register.
+ */
+
+static u8 opti_syscfg(u8 reg)
+{
+	unsigned long flags;
+	u8 r;
+	
+	/* Uniprocessor chipset and must force cycles adjancent */
+	local_irq_save(flags);
+	outb(reg, 0x22);
+	r = inb(0x24);
+	local_irq_restore(flags);
+	return r;
+}
+
+/*
+ *	Opti 82C611A
+ *
+ *	This controller supports PIO0 to PIO3.
+ */
+
+static void opti82c611a_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	u8 active, recover, setup;
+	struct ata_timing t;
+	struct ata_device *pair = ata_dev_pair(adev);
+	int clock;
+	int khz[4] = { 50000, 40000, 33000, 25000 };
+	u8 rc;
+
+	/* Enter configuration mode */
+	inw(ap->ioaddr.error_addr);
+	inw(ap->ioaddr.error_addr);
+	outb(3, ap->ioaddr.nsect_addr);
+
+	/* Read VLB clock strapping */
+	clock = 1000000000 / khz[inb(ap->ioaddr.lbah_addr) & 0x03];
+
+	/* Get the timing data in cycles */
+	ata_timing_compute(adev, adev->pio_mode, &t, clock, 1000);
+
+	/* Setup timing is shared */
+	if (pair) {
+		struct ata_timing tp;
+		ata_timing_compute(pair, pair->pio_mode, &tp, clock, 1000);
+
+		ata_timing_merge(&t, &tp, &t, ATA_TIMING_SETUP);
+	}
+
+	active = FIT(t.active, 2, 17) - 2;
+	recover = FIT(t.recover, 1, 16) - 1;
+	setup = FIT(t.setup, 1, 4) - 1;
+
+	/* Select the right timing bank for write timing */
+	rc = inb(ap->ioaddr.lbal_addr);
+	rc &= 0x7F;
+	rc |= (adev->devno << 7);
+	outb(rc, ap->ioaddr.lbal_addr);
+
+	/* Write the timings */
+	outb(active << 4 | recover, ap->ioaddr.error_addr);
+
+	/* Select the right bank for read timings, also
+	   load the shared timings for address */
+	rc = inb(ap->ioaddr.device_addr);
+	rc &= 0xC0;
+	rc |= adev->devno;	/* Index select */
+	rc |= (setup << 4) | 0x04;
+	outb(rc, ap->ioaddr.device_addr);
+
+	/* Load the read timings */
+	outb(active << 4 | recover, ap->ioaddr.data_addr);
+
+	/* Ensure the timing register mode is right */
+	rc = inb (ap->ioaddr.lbal_addr);
+	rc &= 0x73;
+	rc |= 0x84;
+	outb(rc, ap->ioaddr.lbal_addr);
+
+	/* Exit command mode */
+	outb(0x83,  ap->ioaddr.nsect_addr);
+}
+
+
+static struct ata_port_operations opti82c611a_port_ops = {
+	.set_piomode	= opti82c611a_set_piomode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/*
+ *	Opti 82C465MV
+ *
+ *	This controller supports PIO0 to PIO3. Unlike the 611A the MVB
+ *	version is dual channel but doesn't have a lot of unique registers.
+ */
+
+static void opti82c46x_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	u8 active, recover, setup;
+	struct ata_timing t;
+	struct ata_device *pair = ata_dev_pair(adev);
+	int clock;
+	int khz[4] = { 50000, 40000, 33000, 25000 };
+	u8 rc;
+	u8 sysclk;
+
+	/* Get the clock */
+	sysclk = opti_syscfg(0xAC) & 0xC0;	/* BIOS set */
+
+	/* Enter configuration mode */
+	inw(ap->ioaddr.error_addr);
+	inw(ap->ioaddr.error_addr);
+	outb(3, ap->ioaddr.nsect_addr);
+
+	/* Read VLB clock strapping */
+	clock = 1000000000 / khz[sysclk];
+
+	/* Get the timing data in cycles */
+	ata_timing_compute(adev, adev->pio_mode, &t, clock, 1000);
+
+	/* Setup timing is shared */
+	if (pair) {
+		struct ata_timing tp;
+		ata_timing_compute(pair, pair->pio_mode, &tp, clock, 1000);
+
+		ata_timing_merge(&t, &tp, &t, ATA_TIMING_SETUP);
+	}
+
+	active = FIT(t.active, 2, 17) - 2;
+	recover = FIT(t.recover, 1, 16) - 1;
+	setup = FIT(t.setup, 1, 4) - 1;
+
+	/* Select the right timing bank for write timing */
+	rc = inb(ap->ioaddr.lbal_addr);
+	rc &= 0x7F;
+	rc |= (adev->devno << 7);
+	outb(rc, ap->ioaddr.lbal_addr);
+
+	/* Write the timings */
+	outb(active << 4 | recover, ap->ioaddr.error_addr);
+
+	/* Select the right bank for read timings, also
+	   load the shared timings for address */
+	rc = inb(ap->ioaddr.device_addr);
+	rc &= 0xC0;
+	rc |= adev->devno;	/* Index select */
+	rc |= (setup << 4) | 0x04;
+	outb(rc, ap->ioaddr.device_addr);
+
+	/* Load the read timings */
+	outb(active << 4 | recover, ap->ioaddr.data_addr);
+
+	/* Ensure the timing register mode is right */
+	rc = inb (ap->ioaddr.lbal_addr);
+	rc &= 0x73;
+	rc |= 0x84;
+	outb(rc, ap->ioaddr.lbal_addr);
+
+	/* Exit command mode */
+	outb(0x83,  ap->ioaddr.nsect_addr);
+
+	/* We need to know this for quad device on the MVB */
+	ap->host->private_data = ap;
+}
+
+/**
+ *	opt82c465mv_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings. The
+ *	MVB has a single set of timing registers and these are shared
+ *	across channels. As there are two registers we really ought to
+ *	track the last two used values as a sort of register window. For
+ *	now we just reload on a channel switch. On the single channel
+ *	setup this condition never fires so we do nothing extra.
+ *
+ *	FIXME: dual channel needs ->serialize support
+ */
+
+static unsigned int opti82c46x_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+
+	/* If timings are set and for the wrong channel (2nd test is
+	   due to a libata shortcoming and will eventually go I hope) */
+	if (ap->host->private_data != ap->host
+	    && ap->host->private_data != NULL)
+		opti82c46x_set_piomode(ap, adev);
+
+	return ata_qc_issue_prot(qc);
+}
+
+static struct ata_port_operations opti82c46x_port_ops = {
+	.set_piomode	= opti82c46x_set_piomode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= opti82c46x_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+
+/**
+ *	legacy_init_one		-	attach a legacy interface
+ *	@port: port number
+ *	@io: I/O port start
+ *	@ctrl: control port
+ *	@irq: interrupt line
+ *
+ *	Register an ISA bus IDE interface. Such interfaces are PIO and we
+ *	assume do not support IRQ sharing.
+ */
+
+static __init int legacy_init_one(int port, unsigned long io, unsigned long ctrl, int irq)
+{
+	struct legacy_data *ld = &legacy_data[nr_legacy_host];
+	struct ata_probe_ent ae;
+	struct platform_device *pdev;
+	int ret = -EBUSY;
+	struct ata_port_operations *ops = &legacy_port_ops;
+	int pio_modes = pio_mask;
+	u32 mask = (1 << port);
+
+	if (request_region(io, 8, "pata_legacy") == NULL)
+		return -EBUSY;
+	if (request_region(ctrl, 1, "pata_legacy") == NULL)
+		goto fail_io;
+
+	pdev = platform_device_register_simple(DRV_NAME, nr_legacy_host, NULL, 0);
+	if (pdev == NULL)
+		goto fail_dev;
+
+	if (ht6560a & mask) {
+		ops = &ht6560a_port_ops;
+		pio_modes = 0x07;
+	}
+	if (ht6560b & mask) {
+		ops = &ht6560b_port_ops;
+		pio_modes = 0x1F;
+	}
+	if (opti82c611a & mask) {
+		ops = &opti82c611a_port_ops;
+		pio_modes = 0x0F;
+	}
+	if (opti82c46x & mask) {
+		ops = &opti82c46x_port_ops;
+		pio_modes = 0x0F;
+	}
+
+	/* Probe for automatically detectable controllers */
+ 
+	if (io == 0x1F0 && ops == &legacy_port_ops) {
+		unsigned long flags;
+
+		local_irq_save(flags);
+
+		/* Probes */
+		inb(0x1F5);
+		outb(inb(0x1F2) | 0x80, 0x1F2);
+		inb(0x1F2);
+		inb(0x3F6);
+		inb(0x3F6);
+		inb(0x1F2);
+		inb(0x1F2);
+
+		if ((inb(0x1F2) & 0x80) == 0) {
+			/* PDC20230c or 20630 ? */
+			printk(KERN_INFO "PDC20230-C/20630 VLB ATA controller detected.\n");
+				pio_modes = 0x07;
+			ops = &pdc20230_port_ops;
+			udelay(100);
+			inb(0x1F5);
+		} else {
+			outb(0x55, 0x1F2);
+			inb(0x1F2);
+			inb(0x1F2);
+			if (inb(0x1F2) == 0x00) {
+				printk(KERN_INFO "PDC20230-B VLB ATA controller detected.\n");
+			}
+		}
+		local_irq_restore(flags);
+	}
+
+
+	/* Chip does mode setting by command snooping */
+	if (ops == &legacy_port_ops && (autospeed & mask))
+		ops = &simple_port_ops;
+	memset(&ae, 0, sizeof(struct ata_probe_ent));
+	INIT_LIST_HEAD(&ae.node);
+	ae.dev = &pdev->dev;
+	ae.port_ops = ops;
+	ae.sht = &legacy_sht;
+	ae.n_ports = 1;
+	ae.pio_mask = pio_modes;
+	ae.irq = irq;
+	ae.irq_flags = 0;
+	ae.port_flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST;
+	ae.port[0].cmd_addr = io;
+	ae.port[0].altstatus_addr = ctrl;
+	ae.port[0].ctl_addr =	ctrl;
+	ata_std_ports(&ae.port[0]);
+	ae.private_data = ld;
+
+	ret = ata_device_add(&ae);
+	if (ret == 0) {
+		ret = -ENODEV;
+		goto fail;
+	}
+	legacy_host[nr_legacy_host++] = dev_get_drvdata(&pdev->dev);
+	ld->platform_dev = pdev;
+	return 0;
+
+fail:
+	platform_device_unregister(pdev);
+fail_dev:
+	release_region(ctrl, 1);
+fail_io:
+	release_region(io, 8);
+	return ret;
+}
+
+/**
+ *	legacy_check_special_cases	-	ATA special cases
+ *	@p: PCI device to check
+ *	@master: set this if we find an ATA master
+ *	@master: set this if we find an ATA secondary
+ *
+ *	A small number of vendors implemented early PCI ATA interfaces on bridge logic
+ *	without the ATA interface being PCI visible. Where we have a matching PCI driver
+ *	we must skip the relevant device here. If we don't know about it then the legacy
+ *	driver is the right driver anyway.
+ */
+
+static void legacy_check_special_cases(struct pci_dev *p, int *primary, int *secondary)
+{
+	/* Cyrix CS5510 pre SFF MWDMA ATA on the bridge */
+	if (p->vendor == 0x1078 && p->device == 0x0000) {
+		*primary = *secondary = 1;
+		return;
+	}
+	/* Cyrix CS5520 pre SFF MWDMA ATA on the bridge */
+	if (p->vendor == 0x1078 && p->device == 0x0002) {
+		*primary = *secondary = 1;
+		return;
+	}
+	/* Intel MPIIX - PIO ATA on non PCI side of bridge */
+	if (p->vendor == 0x8086 && p->device == 0x1234) {
+		u16 r;
+		pci_read_config_word(p, 0x6C, &r);
+		if (r & 0x8000) {	/* ATA port enabled */
+			if (r & 0x4000)
+				*secondary = 1;
+			else
+				*primary = 1;
+		}
+		return;
+	}
+}
+
+
+/**
+ *	legacy_init		-	attach legacy interfaces
+ *
+ *	Attach legacy IDE interfaces by scanning the usual IRQ/port suspects.
+ *	Right now we do not scan the ide0 and ide1 address but should do so
+ *	for non PCI systems or systems with no PCI IDE legacy mode devices.
+ *	If you fix that note there are special cases to consider like VLB
+ *	drivers and CS5510/20.
+ */
+
+static __init int legacy_init(void)
+{
+	int i;
+	int ct = 0;
+	int primary = 0;
+	int secondary = 0;
+	int last_port = NR_HOST;
+
+	struct pci_dev *p = NULL;
+
+	for_each_pci_dev(p) {
+		int r;
+		/* Check for any overlap of the system ATA mappings. Native mode controllers
+		   stuck on these addresses or some devices in 'raid' mode won't be found by
+		   the storage class test */
+		for (r = 0; r < 6; r++) {
+			if (pci_resource_start(p, r) == 0x1f0)
+				primary = 1;
+			if (pci_resource_start(p, r) == 0x170)
+				secondary = 1;
+		}
+		/* Check for special cases */
+		legacy_check_special_cases(p, &primary, &secondary);
+
+		/* If PCI bus is present then don't probe for tertiary legacy ports */
+		if (probe_all == 0)
+			last_port = 2;
+	}
+
+	/* If an OPTI 82C46X is present find out where the channels are */	
+	if (opti82c46x) {
+		static const char *optis[4] = {
+			"3/463MV", "5MV",
+			"5MVA", "5MVB"
+		};
+		u8 chans = 1;
+		u8 ctrl = (opti_syscfg(0x30) & 0xC0) >> 6;
+		
+		opti82c46x = 3;	/* Assume master and slave first */
+		printk(KERN_INFO DRV_NAME ": Opti 82C46%s chipset support.\n", optis[ctrl]);
+		if (ctrl == 3)
+			chans = (opti_syscfg(0x3F) & 0x20) ? 2 : 1;
+		ctrl = opti_syscfg(0xAC);
+		/* Check enabled and this port is the 465MV port. On the
+		   MVB we may have two channels */
+		if (ctrl & 8) {
+			if (ctrl & 4)
+				opti82c46x = 2;	/* Slave */
+			else
+				opti82c46x = 1;	/* Master */
+			if (chans == 2)
+				opti82c46x = 3; /* Master and Slave */
+		}	/* Slave only */
+		else if (chans == 1)
+			opti82c46x = 1;
+	}
+
+	for (i = 0; i < last_port; i++) {
+		/* Skip primary if we have seen a PCI one */
+		if (i == 0 && primary == 1)
+			continue;
+		/* Skip secondary if we have seen a PCI one */
+		if (i == 1 && secondary == 1)
+			continue;
+		if (legacy_init_one(i, legacy_port[i],
+				   legacy_port[i] + 0x0206,
+				   legacy_irq[i]) == 0)
+			ct++;
+	}
+	if (ct != 0)
+		return 0;
+	return -ENODEV;
+}
+
+static __exit void legacy_exit(void)
+{
+	int i;
+
+	for (i = 0; i < nr_legacy_host; i++) {
+		struct legacy_data *ld = &legacy_data[i];
+		struct ata_port *ap =legacy_host[i]->ports[0];
+		unsigned long io = ap->ioaddr.cmd_addr;
+		unsigned long ctrl = ap->ioaddr.ctl_addr;
+		ata_host_remove(legacy_host[i]);
+		platform_device_unregister(ld->platform_dev);
+		if (ld->timing)
+			release_region(ld->timing, 2);
+		release_region(io, 8);
+		release_region(ctrl, 1);
+	}
+}
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for legacy ATA");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+
+module_param(probe_all, int, 0);
+module_param(autospeed, int, 0);
+module_param(ht6560a, int, 0);
+module_param(ht6560b, int, 0);
+module_param(opti82c611a, int, 0);
+module_param(opti82c46x, int, 0);
+module_param(pio_mask, int, 0);
+
+module_init(legacy_init);
+module_exit(legacy_exit);
+
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
new file mode 100644
index 000000000000..1958c4ed09a8
--- /dev/null
+++ b/drivers/ata/pata_mpiix.c
@@ -0,0 +1,313 @@
+/*
+ * pata_mpiix.c 	- Intel MPIIX PATA for new ATA layer
+ *			  (C) 2005-2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * The MPIIX is different enough to the PIIX4 and friends that we give it
+ * a separate driver. The old ide/pci code handles this by just not tuning
+ * MPIIX at all.
+ *
+ * The MPIIX also differs in another important way from the majority of PIIX
+ * devices. The chip is a bridge (pardon the pun) between the old world of
+ * ISA IDE and PCI IDE. Although the ATA timings are PCI configured the actual
+ * IDE controller is not decoded in PCI space and the chip does not claim to
+ * be IDE class PCI. This requires slightly non-standard probe logic compared
+ * with PCI IDE and also that we do not disable the device when our driver is
+ * unloaded (as it has many other functions).
+ *
+ * The driver conciously keeps this logic internally to avoid pushing quirky
+ * PATA history into the clean libata layer.
+ *
+ * Thinkpad specific note: If you boot an MPIIX using thinkpad with a PCMCIA
+ * hard disk present this driver will not detect it. This is not a bug. In this
+ * configuration the secondary port of the MPIIX is disabled and the addresses
+ * are decoded by the PCMCIA bridge and therefore are for a generic IDE driver
+ * to operate.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_mpiix"
+#define DRV_VERSION "0.7.1"
+
+enum {
+	IDETIM = 0x6C,		/* IDE control register */
+	IORDY = (1 << 1),
+	PPE = (1 << 2),
+	FTIM = (1 << 0),
+	ENABLED = (1 << 15),
+	SECONDARY = (1 << 14)
+};
+
+static int mpiix_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static const struct pci_bits mpiix_enable_bits[] = {
+		{ 0x6D, 1, 0x80, 0x80 },
+		{ 0x6F, 1, 0x80, 0x80 }
+	};
+
+	if (!pci_test_config_bits(pdev, &mpiix_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	mpiix_error_handler		-	probe reset
+ *	@ap: ATA port
+ *
+ *	Perform the ATA probe and bus reset sequence plus specific handling
+ *	for this hardware. The MPIIX has the enable bits in a different place
+ *	to PIIX4 and friends. As a pure PIO device it has no cable detect
+ */
+
+static void mpiix_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, mpiix_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	mpiix_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup. The MPIIX allows us to program the
+ *	IORDY sample point (2-5 clocks), recovery 1-4 clocks and whether
+ *	prefetching or iordy are used.
+ *
+ *	This would get very ugly because we can only program timing for one
+ *	device at a time, the other gets PIO0. Fortunately libata calls
+ *	our qc_issue_prot command before a command is issued so we can
+ *	flip the timings back and forth to reduce the pain.
+ */
+
+static void mpiix_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	int control = 0;
+	int pio = adev->pio_mode - XFER_PIO_0;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u16 idetim;
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 0 },
+			    { 2, 1 },
+			    { 2, 3 }, };
+
+	pci_read_config_word(pdev, IDETIM, &idetim);
+	/* Mask the IORDY/TIME/PPE0 bank for this device */
+	if (adev->class == ATA_DEV_ATA)
+		control |= PPE;		/* PPE enable for disk */
+	if (ata_pio_need_iordy(adev))
+		control |= IORDY;	/* IORDY */
+	if (pio > 0)
+		control |= FTIM;	/* This drive is on the fast timing bank */
+
+	/* Mask out timing and clear both TIME bank selects */
+	idetim &= 0xCCEE;
+	idetim &= ~(0x07  << (2 * adev->devno));
+	idetim |= (control << (2 * adev->devno));
+
+	idetim |= (timings[pio][0] << 12) | (timings[pio][1] << 8);
+	pci_write_config_word(pdev, IDETIM, idetim);
+
+	/* We use ap->private_data as a pointer to the device currently
+	   loaded for timing */
+	ap->private_data = adev;
+}
+
+/**
+ *	mpiix_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings if
+ *	neccessary. Our logic also clears TIME0/TIME1 for the other device so
+ *	that, even if we get this wrong, cycles to the other device will
+ *	be made PIO0.
+ */
+
+static unsigned int mpiix_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+
+	/* If modes have been configured and the channel data is not loaded
+	   then load it. We have to check if pio_mode is set as the core code
+	   does not set adev->pio_mode to XFER_PIO_0 while probing as would be
+	   logical */
+
+	if (adev->pio_mode && adev != ap->private_data)
+		mpiix_set_piomode(ap, adev);
+
+	return ata_qc_issue_prot(qc);
+}
+
+static struct scsi_host_template mpiix_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations mpiix_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= mpiix_set_piomode,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= mpiix_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= mpiix_qc_issue_prot,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int mpiix_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	/* Single threaded by the PCI probe logic */
+	static struct ata_probe_ent probe[2];
+	static int printed_version;
+	u16 idetim;
+	int enabled;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
+
+	/* MPIIX has many functions which can be turned on or off according
+	   to other devices present. Make sure IDE is enabled before we try
+	   and use it */
+
+	pci_read_config_word(dev, IDETIM, &idetim);
+	if (!(idetim & ENABLED))
+		return -ENODEV;
+
+	/* We do our own plumbing to avoid leaking special cases for whacko
+	   ancient hardware into the core code. There are two issues to
+	   worry about.  #1 The chip is a bridge so if in legacy mode and
+	   without BARs set fools the setup.  #2 If you pci_disable_device
+	   the MPIIX your box goes castors up */
+
+	INIT_LIST_HEAD(&probe[0].node);
+	probe[0].dev = pci_dev_to_dev(dev);
+	probe[0].port_ops = &mpiix_port_ops;
+	probe[0].sht = &mpiix_sht;
+	probe[0].pio_mask = 0x1F;
+	probe[0].irq = 14;
+	probe[0].irq_flags = SA_SHIRQ;
+	probe[0].port_flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST;
+	probe[0].n_ports = 1;
+	probe[0].port[0].cmd_addr = 0x1F0;
+	probe[0].port[0].ctl_addr = 0x3F6;
+	probe[0].port[0].altstatus_addr = 0x3F6;
+
+	/* The secondary lurks at different addresses but is otherwise
+	   the same beastie */
+
+	INIT_LIST_HEAD(&probe[1].node);
+	probe[1] = probe[0];
+	probe[1].irq = 15;
+	probe[1].port[0].cmd_addr = 0x170;
+	probe[1].port[0].ctl_addr = 0x376;
+	probe[1].port[0].altstatus_addr = 0x376;
+
+	/* Let libata fill in the port details */
+	ata_std_ports(&probe[0].port[0]);
+	ata_std_ports(&probe[1].port[0]);
+
+	/* Now add the port that is active */
+	enabled = (idetim & SECONDARY) ? 1 : 0;
+
+	if (ata_device_add(&probe[enabled]))
+		return 0;
+	return -ENODEV;
+}
+
+/**
+ *	mpiix_remove_one	-	device unload
+ *	@pdev: PCI device being removed
+ *
+ *	Handle an unplug/unload event for a PCI device. Unload the
+ *	PCI driver but do not use the default handler as we *MUST NOT*
+ *	disable the device as it has other functions.
+ */
+
+static void __devexit mpiix_remove_one(struct pci_dev *pdev)
+{
+	struct device *dev = pci_dev_to_dev(pdev);
+	struct ata_host *host = dev_get_drvdata(dev);
+
+	ata_host_remove(host);
+	dev_set_drvdata(dev, NULL);
+}
+
+
+
+static const struct pci_device_id mpiix[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82371MX), },
+	{ 0, },
+};
+
+static struct pci_driver mpiix_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= mpiix,
+	.probe 		= mpiix_init_one,
+	.remove		= mpiix_remove_one
+};
+
+static int __init mpiix_init(void)
+{
+	return pci_register_driver(&mpiix_pci_driver);
+}
+
+
+static void __exit mpiix_exit(void)
+{
+	pci_unregister_driver(&mpiix_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Intel MPIIX");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, mpiix);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(mpiix_init);
+module_exit(mpiix_exit);
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
new file mode 100644
index 000000000000..e3a85778cd4f
--- /dev/null
+++ b/drivers/ata/pata_netcell.c
@@ -0,0 +1,175 @@
+/*
+ *    pata_netcell.c - Netcell PATA driver
+ *
+ *	(c) 2006 Red Hat  <alan@redhat.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_netcell"
+#define DRV_VERSION	"0.1.5"
+
+/**
+ *	netcell_probe_init	-	check for 40/80 pin
+ *	@ap: Port
+ *
+ *	Cables are handled by the RAID controller. Report 80 pin.
+ */
+ 
+static int netcell_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	netcell_probe_reset - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void netcell_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, netcell_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/* No PIO or DMA methods needed for this device */
+
+static struct scsi_host_template netcell_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	/* Special handling needed if you have sector or LBA48 limits */
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	/* Use standard CHS mapping rules */
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations netcell_ops = {
+	.port_disable		= ata_port_disable,
+
+	/* Task file is PCI ATA format, use helpers */
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= netcell_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	/* BMDMA handling is PCI ATA format, use helpers */
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	/* Timeout handling. Special recovery hooks here */
+	.eng_timeout		= ata_eng_timeout,
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	/* Generic PATA PCI ATA helpers */
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+
+/**
+ *	netcell_init_one - Register Netcell ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in netcell_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int netcell_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	static struct ata_port_info info = {
+		.sht		= &netcell_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		/* Actually we don't really care about these as the
+		   firmware deals with it */
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma0-2 */
+		.udma_mask 	= 0x3f, /* UDMA 133 */
+		.port_ops	= &netcell_ops,
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev,
+			   "version " DRV_VERSION "\n");
+
+	/* Any chip specific setup/optimisation/messages here */
+	ata_pci_clear_simplex(pdev);
+	
+	/* And let the library code do the work */
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id netcell_pci_tbl[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_NETCELL, PCI_DEVICE_ID_REVOLUTION), },
+	{ }	/* terminate list */
+};
+
+static struct pci_driver netcell_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= netcell_pci_tbl,
+	.probe			= netcell_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init netcell_init(void)
+{
+	return pci_register_driver(&netcell_pci_driver);
+}
+
+static void __exit netcell_exit(void)
+{
+	pci_unregister_driver(&netcell_pci_driver);
+}
+
+module_init(netcell_init);
+module_exit(netcell_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for Netcell PATA RAID");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, netcell_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
new file mode 100644
index 000000000000..93d6646d2954
--- /dev/null
+++ b/drivers/ata/pata_ns87410.c
@@ -0,0 +1,236 @@
+/*
+ * pata_ns87410.c 	- National Semiconductor 87410 PATA for new ATA layer
+ *			  (C) 2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_ns87410"
+#define DRV_VERSION "0.4.2"
+
+/**
+ *	ns87410_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+
+static int ns87410_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static const struct pci_bits ns87410_enable_bits[] = {
+		{ 0x43, 1, 0x08, 0x08 },
+		{ 0x47, 1, 0x08, 0x08 }
+	};
+
+	if (!pci_test_config_bits(pdev, &ns87410_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	ns87410_error_handler		-	probe reset
+ *	@ap: ATA port
+ *
+ *	Perform the ATA probe and bus reset sequence plus specific handling
+ *	for this hardware. The MPIIX has the enable bits in a different place
+ *	to PIIX4 and friends. As a pure PIO device it has no cable detect
+ */
+
+static void ns87410_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, ns87410_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	ns87410_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program timing data. This is kept per channel not per device,
+ *	and only affects the data port.
+ */
+
+static void ns87410_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int port = 0x40 + 4 * ap->port_no;
+	u8 idetcr, idefr;
+	struct ata_timing at;
+
+	static const u8 activebits[15] = {
+		0, 1, 2, 3, 4,
+		5, 5, 6, 6, 6,
+		6, 7, 7, 7, 7
+	};
+
+	static const u8 recoverbits[12] = {
+		0, 1, 2, 3, 4, 5, 6, 6, 7, 7, 7, 7
+	};
+
+	pci_read_config_byte(pdev, port + 3, &idefr);
+
+	if (ata_pio_need_iordy(adev))
+		idefr |= 0x04;	/* IORDY enable */
+	else
+		idefr &= ~0x04;
+
+	if (ata_timing_compute(adev, adev->pio_mode, &at, 30303, 1) < 0) {
+		dev_printk(KERN_ERR, &pdev->dev, "unknown mode %d.\n", adev->pio_mode);
+		return;
+	}
+
+	at.active = FIT(at.active, 2, 16) - 2;
+	at.setup = FIT(at.setup, 1, 4) - 1;
+	at.recover = FIT(at.recover, 1, 12) - 1;
+
+	idetcr = (at.setup << 6) | (recoverbits[at.recover] << 3) | activebits[at.active];
+
+	pci_write_config_byte(pdev, port, idetcr);
+	pci_write_config_byte(pdev, port + 3, idefr);
+	/* We use ap->private_data as a pointer to the device currently
+	   loaded for timing */
+	ap->private_data = adev;
+}
+
+/**
+ *	ns87410_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings if
+ *	neccessary.
+ */
+
+static unsigned int ns87410_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+
+	/* If modes have been configured and the channel data is not loaded
+	   then load it. We have to check if pio_mode is set as the core code
+	   does not set adev->pio_mode to XFER_PIO_0 while probing as would be
+	   logical */
+
+	if (adev->pio_mode && adev != ap->private_data)
+		ns87410_set_piomode(ap, adev);
+
+	return ata_qc_issue_prot(qc);
+}
+
+static struct scsi_host_template ns87410_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations ns87410_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= ns87410_set_piomode,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ns87410_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ns87410_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int ns87410_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &ns87410_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x0F,
+		.port_ops = &ns87410_port_ops
+	};
+	static struct ata_port_info *port_info[2] = {&info, &info};
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static const struct pci_device_id ns87410[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_87410), },
+	{ 0, },
+};
+
+static struct pci_driver ns87410_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= ns87410,
+	.probe 		= ns87410_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init ns87410_init(void)
+{
+	return pci_register_driver(&ns87410_pci_driver);
+}
+
+
+static void __exit ns87410_exit(void)
+{
+	pci_unregister_driver(&ns87410_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Nat Semi 87410");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, ns87410);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(ns87410_init);
+module_exit(ns87410_exit);
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
new file mode 100644
index 000000000000..04c618a2664b
--- /dev/null
+++ b/drivers/ata/pata_oldpiix.c
@@ -0,0 +1,339 @@
+/*
+ *    pata_oldpiix.c - Intel PATA/SATA controllers
+ *
+ *	(C) 2005 Red Hat <alan@redhat.com>
+ *
+ *    Some parts based on ata_piix.c by Jeff Garzik and others.
+ *
+ *    Early PIIX differs significantly from the later PIIX as it lacks
+ *    SITRE and the slave timing registers. This means that you have to
+ *    set timing per channel, or be clever. Libata tells us whenever it
+ *    does drive selection and we use this to reload the timings.
+ *
+ *    Because of these behaviour differences PIIX gets its own driver module.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_oldpiix"
+#define DRV_VERSION	"0.5.1"
+
+/**
+ *	oldpiix_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+
+static int oldpiix_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static const struct pci_bits oldpiix_enable_bits[] = {
+		{ 0x41U, 1U, 0x80UL, 0x80UL },	/* port 0 */
+		{ 0x43U, 1U, 0x80UL, 0x80UL },	/* port 1 */
+	};
+
+	if (!pci_test_config_bits(pdev, &oldpiix_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	oldpiix_pata_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *	@classes:
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void oldpiix_pata_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, oldpiix_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	oldpiix_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: um
+ *
+ *	Set PIO mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void oldpiix_set_piomode (struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int pio	= adev->pio_mode - XFER_PIO_0;
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
+	unsigned int idetm_port= ap->port_no ? 0x42 : 0x40;
+	u16 idetm_data;
+	int control = 0;
+
+	/*
+	 *	See Intel Document 298600-004 for the timing programing rules
+	 *	for PIIX/ICH. Note that the early PIIX does not have the slave
+	 *	timing port at 0x44.
+	 */
+
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 0 },
+			    { 2, 1 },
+			    { 2, 3 }, };
+
+	if (pio > 2)
+		control |= 1;	/* TIME1 enable */
+	if (ata_pio_need_iordy(adev))
+		control |= 2;	/* IE IORDY */
+
+	/* Intel specifies that the PPE functionality is for disk only */
+	if (adev->class == ATA_DEV_ATA)
+		control |= 4;	/* PPE enable */
+
+	pci_read_config_word(dev, idetm_port, &idetm_data);
+
+	/* Enable PPE, IE and TIME as appropriate. Clear the other
+	   drive timing bits */
+	if (adev->devno == 0) {
+		idetm_data &= 0xCCE0;
+		idetm_data |= control;
+	} else {
+		idetm_data &= 0xCC0E;
+		idetm_data |= (control << 4);
+	}
+	idetm_data |= (timings[pio][0] << 12) |
+			(timings[pio][1] << 8);
+	pci_write_config_word(dev, idetm_port, idetm_data);
+
+	/* Track which port is configured */
+	ap->private_data = adev;
+}
+
+/**
+ *	oldpiix_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *	@isich: True if the device is an ICH and has IOCFG registers
+ *
+ *	Set MWDMA mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void oldpiix_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
+	u8 idetm_port		= ap->port_no ? 0x42 : 0x40;
+	u16 idetm_data;
+
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 0 },
+			    { 2, 1 },
+			    { 2, 3 }, };
+
+	/*
+	 * MWDMA is driven by the PIO timings. We must also enable
+	 * IORDY unconditionally along with TIME1. PPE has already
+	 * been set when the PIO timing was set.
+	 */
+
+	unsigned int mwdma	= adev->dma_mode - XFER_MW_DMA_0;
+	unsigned int control;
+	const unsigned int needed_pio[3] = {
+		XFER_PIO_0, XFER_PIO_3, XFER_PIO_4
+	};
+	int pio = needed_pio[mwdma] - XFER_PIO_0;
+
+	pci_read_config_word(dev, idetm_port, &idetm_data);
+
+	control = 3;	/* IORDY|TIME0 */
+	/* Intel specifies that the PPE functionality is for disk only */
+	if (adev->class == ATA_DEV_ATA)
+		control |= 4;	/* PPE enable */
+
+	/* If the drive MWDMA is faster than it can do PIO then
+	   we must force PIO into PIO0 */
+
+	if (adev->pio_mode < needed_pio[mwdma])
+		/* Enable DMA timing only */
+		control |= 8;	/* PIO cycles in PIO0 */
+
+	/* Mask out the relevant control and timing bits we will load. Also
+	   clear the other drive TIME register as a precaution */
+	if (adev->devno == 0) {
+		idetm_data &= 0xCCE0;
+		idetm_data |= control;
+	} else {
+		idetm_data &= 0xCC0E;
+		idetm_data |= (control << 4);
+	}
+	idetm_data |= (timings[pio][0] << 12) | (timings[pio][1] << 8);
+	pci_write_config_word(dev, idetm_port, idetm_data);
+
+	/* Track which port is configured */
+	ap->private_data = adev;
+}
+
+/**
+ *	oldpiix_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings if
+ *	neccessary. Our logic also clears TIME0/TIME1 for the other device so
+ *	that, even if we get this wrong, cycles to the other device will
+ *	be made PIO0.
+ */
+
+static unsigned int oldpiix_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+
+	if (adev != ap->private_data) {
+		if (adev->dma_mode)
+			oldpiix_set_dmamode(ap, adev);
+		else if (adev->pio_mode)
+			oldpiix_set_piomode(ap, adev);
+	}
+	return ata_qc_issue_prot(qc);
+}
+
+
+static struct scsi_host_template oldpiix_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations oldpiix_pata_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= oldpiix_set_piomode,
+	.set_dmamode		= oldpiix_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= oldpiix_pata_error_handler,
+	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= oldpiix_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+
+/**
+ *	oldpiix_init_one - Register PIIX ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in oldpiix_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.  We probe for combined mode (sigh),
+ *	and then hand over control to libata, for it to do the rest.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int oldpiix_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	static struct ata_port_info info = {
+		.sht		= &oldpiix_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma1-2 */
+		.port_ops	= &oldpiix_pata_ops,
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev,
+			   "version " DRV_VERSION "\n");
+
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id oldpiix_pci_tbl[] = {
+	{ PCI_DEVICE(0x8086, 0x1230), },
+	{ }	/* terminate list */
+};
+
+static struct pci_driver oldpiix_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= oldpiix_pci_tbl,
+	.probe			= oldpiix_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init oldpiix_init(void)
+{
+	return pci_register_driver(&oldpiix_pci_driver);
+}
+
+static void __exit oldpiix_exit(void)
+{
+	pci_unregister_driver(&oldpiix_pci_driver);
+}
+
+
+module_init(oldpiix_init);
+module_exit(oldpiix_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for early PIIX series controllers");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, oldpiix_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
new file mode 100644
index 000000000000..c3d01325e0e2
--- /dev/null
+++ b/drivers/ata/pata_opti.c
@@ -0,0 +1,292 @@
+/*
+ * pata_opti.c 	- ATI PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based on
+ *  linux/drivers/ide/pci/opti621.c		Version 0.7	Sept 10, 2002
+ *
+ *  Copyright (C) 1996-1998  Linus Torvalds & authors (see below)
+ *
+ * Authors:
+ * Jaromir Koutek <miri@punknet.cz>,
+ * Jan Harkes <jaharkes@cwi.nl>,
+ * Mark Lord <mlord@pobox.com>
+ * Some parts of code are from ali14xx.c and from rz1000.c.
+ *
+ * Also consulted the FreeBSD prototype driver by Kevin Day to try
+ * and resolve some confusions. Further documentation can be found in
+ * Ralf Brown's interrupt list
+ *
+ * If you have other variants of the Opti range (Viper/Vendetta) please
+ * try this driver with those PCI idents and report back. For the later
+ * chips see the pata_optidma driver
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_opti"
+#define DRV_VERSION "0.2.4"
+
+enum {
+	READ_REG	= 0,	/* index of Read cycle timing register */
+	WRITE_REG 	= 1,	/* index of Write cycle timing register */
+	CNTRL_REG 	= 3,	/* index of Control register */
+	STRAP_REG 	= 5,	/* index of Strap register */
+	MISC_REG 	= 6	/* index of Miscellaneous register */
+};
+
+/**
+ *	opti_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+
+static int opti_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static const struct pci_bits opti_enable_bits[] = {
+		{ 0x45, 1, 0x80, 0x00 },
+		{ 0x40, 1, 0x08, 0x00 }
+	};
+
+	if (!pci_test_config_bits(pdev, &opti_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	opti_probe_reset		-	probe reset
+ *	@ap: ATA port
+ *
+ *	Perform the ATA probe and bus reset sequence plus specific handling
+ *	for this hardware. The Opti needs little handling - we have no UDMA66
+ *	capability that needs cable detection. All we must do is check the port
+ *	is enabled.
+ */
+
+static void opti_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, opti_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	opti_write_reg		-	control register setup
+ *	@ap: ATA port
+ *	@value: value
+ *	@reg: control register number
+ *
+ *	The Opti uses magic 'trapdoor' register accesses to do configuration
+ *	rather than using PCI space as other controllers do. The double inw
+ *	on the error register activates configuration mode. We can then write
+ *	the control register
+ */
+
+static void opti_write_reg(struct ata_port *ap, u8 val, int reg)
+{
+	unsigned long regio = ap->ioaddr.cmd_addr;
+
+	/* These 3 unlock the control register access */
+	inw(regio + 1);
+	inw(regio + 1);
+	outb(3, regio + 2);
+
+	/* Do the I/O */
+	outb(val, regio + reg);
+
+	/* Relock */
+	outb(0x83, regio + 2);
+}
+
+#if 0
+/**
+ *	opti_read_reg		-	control register read
+ *	@ap: ATA port
+ *	@reg: control register number
+ *
+ *	The Opti uses magic 'trapdoor' register accesses to do configuration
+ *	rather than using PCI space as other controllers do. The double inw
+ *	on the error register activates configuration mode. We can then read
+ *	the control register
+ */
+
+static u8 opti_read_reg(struct ata_port *ap, int reg)
+{
+	unsigned long regio = ap->ioaddr.cmd_addr;
+	u8 ret;
+	inw(regio + 1);
+	inw(regio + 1);
+	outb(3, regio + 2);
+	ret = inb(regio + reg);
+	outb(0x83, regio + 2);
+}
+#endif
+
+/**
+ *	opti_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup. Timing numbers are taken from
+ *	the FreeBSD driver then pre computed to keep the code clean. There
+ *	are two tables depending on the hardware clock speed.
+ */
+
+static void opti_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct ata_device *pair = ata_dev_pair(adev);
+	int clock;
+	int pio = adev->pio_mode - XFER_PIO_0;
+	unsigned long regio = ap->ioaddr.cmd_addr;
+	u8 addr;
+
+	/* Address table precomputed with prefetch off and a DCLK of 2 */
+	static const u8 addr_timing[2][5] = {
+		{ 0x30, 0x20, 0x20, 0x10, 0x10 },
+		{ 0x20, 0x20, 0x10, 0x10, 0x10 }
+	};
+	static const u8 data_rec_timing[2][5] = {
+		{ 0x6B, 0x56, 0x42, 0x32, 0x31 },
+		{ 0x58, 0x44, 0x32, 0x22, 0x21 }
+	};
+
+	outb(0xff, regio + 5);
+	clock = inw(regio + 5) & 1;
+
+	/*
+ 	 *	As with many controllers the address setup time is shared
+ 	 *	and must suit both devices if present.
+	 */
+
+	addr = addr_timing[clock][pio];
+	if (pair) {
+		/* Hardware constraint */
+		u8 pair_addr = addr_timing[clock][pair->pio_mode - XFER_PIO_0];
+		if (pair_addr > addr)
+			addr = pair_addr;
+	}
+
+	/* Commence primary programming sequence */
+	opti_write_reg(ap, adev->devno, MISC_REG);
+	opti_write_reg(ap, data_rec_timing[clock][pio], READ_REG);
+	opti_write_reg(ap, data_rec_timing[clock][pio], WRITE_REG);
+	opti_write_reg(ap, addr, MISC_REG);
+
+	/* Programming sequence complete, override strapping */
+	opti_write_reg(ap, 0x85, CNTRL_REG);
+}
+
+static struct scsi_host_template opti_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations opti_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= opti_set_piomode,
+/*	.set_dmamode	= opti_set_dmamode, */
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= opti_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int opti_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &opti_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.port_ops = &opti_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+	static int printed_version;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
+
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static const struct pci_device_id opti[] = {
+	{ PCI_VENDOR_ID_OPTI, PCI_DEVICE_ID_OPTI_82C621, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{ PCI_VENDOR_ID_OPTI, PCI_DEVICE_ID_OPTI_82C825, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
+	{ 0, },
+};
+
+static struct pci_driver opti_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= opti,
+	.probe 		= opti_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init opti_init(void)
+{
+	return pci_register_driver(&opti_pci_driver);
+}
+
+
+static void __exit opti_exit(void)
+{
+	pci_unregister_driver(&opti_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Opti 621/621X");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, opti);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(opti_init);
+module_exit(opti_exit);
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
new file mode 100644
index 000000000000..800aea7b9444
--- /dev/null
+++ b/drivers/ata/pata_optidma.c
@@ -0,0 +1,547 @@
+/*
+ * pata_optidma.c 	- Opti DMA PATA for new ATA layer
+ *			  (C) 2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ *	The Opti DMA controllers are related to the older PIO PCI controllers
+ *	and indeed the VLB ones. The main differences are that the timing
+ *	numbers are now based off PCI clocks not VLB and differ, and that
+ *	MWDMA is supported.
+ *
+ *	This driver should support Viper-N+, FireStar, FireStar Plus.
+ *
+ *	These devices support virtual DMA for read (aka the CS5520). Later
+ *	chips support UDMA33, but only if the rest of the board logic does,
+ *	so you have to get this right. We don't support the virtual DMA
+ *	but we do handle UDMA.
+ *
+ *	Bits that are worth knowing
+ *		Most control registers are shadowed into I/O registers
+ *		0x1F5 bit 0 tells you if the PCI/VLB clock is 33 or 25Mhz
+ *		Virtual DMA registers *move* between rev 0x02 and rev 0x10
+ *		UDMA requires a 66MHz FSB
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_optidma"
+#define DRV_VERSION "0.2.1"
+
+enum {
+	READ_REG	= 0,	/* index of Read cycle timing register */
+	WRITE_REG 	= 1,	/* index of Write cycle timing register */
+	CNTRL_REG 	= 3,	/* index of Control register */
+	STRAP_REG 	= 5,	/* index of Strap register */
+	MISC_REG 	= 6	/* index of Miscellaneous register */
+};
+
+static int pci_clock;	/* 0 = 33 1 = 25 */
+
+/**
+ *	optidma_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+ 
+static int optidma_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static const struct pci_bits optidma_enable_bits = { 
+		0x40, 1, 0x08, 0x00
+	};
+
+	if (ap->port_no && !pci_test_config_bits(pdev, &optidma_enable_bits)) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	optidma_probe_reset		-	probe reset
+ *	@ap: ATA port
+ *
+ *	Perform the ATA probe and bus reset sequence plus specific handling
+ *	for this hardware. The Opti needs little handling - we have no UDMA66
+ *	capability that needs cable detection. All we must do is check the port
+ *	is enabled.
+ */
+
+static void optidma_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, optidma_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	optidma_unlock		-	unlock control registers
+ *	@ap: ATA port
+ *
+ *	Unlock the control register block for this adapter. Registers must not
+ *	be unlocked in a situation where libata might look at them.
+ */
+ 
+static void optidma_unlock(struct ata_port *ap)
+{
+	unsigned long regio = ap->ioaddr.cmd_addr;
+	
+	/* These 3 unlock the control register access */
+	inw(regio + 1);
+	inw(regio + 1);
+	outb(3, regio + 2);
+}
+
+/**
+ *	optidma_lock		-	issue temporary relock
+ *	@ap: ATA port
+ *
+ *	Re-lock the configuration register settings.
+ */
+ 
+static void optidma_lock(struct ata_port *ap)
+{
+	unsigned long regio = ap->ioaddr.cmd_addr;
+	
+	/* Relock */
+	outb(0x83, regio + 2);
+}
+
+/**
+ *	optidma_set_mode	-	set mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *	@mode: Mode to set
+ *
+ *	Called to do the DMA or PIO mode setup. Timing numbers are all
+ *	pre computed to keep the code clean. There are two tables depending
+ *	on the hardware clock speed.
+ *
+ *	WARNING: While we do this the IDE registers vanish. If we take an
+ *	IRQ here we depend on the host set locking to avoid catastrophe.
+ */
+
+static void optidma_set_mode(struct ata_port *ap, struct ata_device *adev, u8 mode)
+{
+	struct ata_device *pair = ata_dev_pair(adev);
+	int pio = adev->pio_mode - XFER_PIO_0;
+	int dma = adev->dma_mode - XFER_MW_DMA_0;
+	unsigned long regio = ap->ioaddr.cmd_addr;
+	u8 addr;
+
+	/* Address table precomputed with a DCLK of 2 */
+	static const u8 addr_timing[2][5] = {
+		{ 0x30, 0x20, 0x20, 0x10, 0x10 },
+		{ 0x20, 0x20, 0x10, 0x10, 0x10 }
+	};
+	static const u8 data_rec_timing[2][5] = {
+		{ 0x59, 0x46, 0x30, 0x20, 0x20 },
+		{ 0x46, 0x32, 0x20, 0x20, 0x10 }
+	};
+	static const u8 dma_data_rec_timing[2][3] = {
+		{ 0x76, 0x20, 0x20 },
+		{ 0x54, 0x20, 0x10 }
+	};
+
+	/* Switch from IDE to control mode */
+	optidma_unlock(ap);
+	
+
+	/*
+ 	 *	As with many controllers the address setup time is shared
+ 	 *	and must suit both devices if present. FIXME: Check if we
+ 	 *	need to look at slowest of PIO/DMA mode of either device
+	 */
+
+	if (mode >= XFER_MW_DMA_0)
+		addr = 0;
+	else
+		addr = addr_timing[pci_clock][pio];
+	
+	if (pair) {
+		u8 pair_addr;
+		/* Hardware constraint */
+		if (pair->dma_mode)
+			pair_addr = 0;
+		else
+			pair_addr = addr_timing[pci_clock][pair->pio_mode - XFER_PIO_0];
+		if (pair_addr > addr)
+			addr = pair_addr;
+	}
+	
+	/* Commence primary programming sequence */
+	/* First we load the device number into the timing select */
+	outb(adev->devno, regio + MISC_REG);
+	/* Now we load the data timings into read data/write data */
+	if (mode < XFER_MW_DMA_0) {
+		outb(data_rec_timing[pci_clock][pio], regio + READ_REG);
+		outb(data_rec_timing[pci_clock][pio], regio + WRITE_REG);
+	} else if (mode < XFER_UDMA_0) {
+		outb(dma_data_rec_timing[pci_clock][dma], regio + READ_REG);
+		outb(dma_data_rec_timing[pci_clock][dma], regio + WRITE_REG);
+	}
+	/* Finally we load the address setup into the misc register */
+	outb(addr | adev->devno, regio + MISC_REG);
+
+	/* Programming sequence complete, timing 0 dev 0, timing 1 dev 1 */
+	outb(0x85, regio + CNTRL_REG);
+	
+	/* Switch back to IDE mode */
+	optidma_lock(ap);
+	
+	/* Note: at this point our programming is incomplete. We are
+	   not supposed to program PCI 0x43 "things we hacked onto the chip"
+	   until we've done both sets of PIO/DMA timings */
+}
+
+/**
+ *	optiplus_set_mode	-	DMA setup for Firestar Plus
+ *	@ap: ATA port
+ *	@adev: device
+ *	@mode: desired mode
+ *
+ *	The Firestar plus has additional UDMA functionality for UDMA0-2 and
+ *	requires we do some additional work. Because the base work we must do
+ *	is mostly shared we wrap the Firestar setup functionality in this
+ *	one
+ */
+
+static void optiplus_set_mode(struct ata_port *ap, struct ata_device *adev, u8 mode)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 udcfg;
+	u8 udslave;
+	int dev2 = 2 * adev->devno;
+	int unit = 2 * ap->port_no + adev->devno;
+	int udma = mode - XFER_UDMA_0;
+	
+	pci_read_config_byte(pdev, 0x44, &udcfg);
+	if (mode <= XFER_UDMA_0) {
+		udcfg &= ~(1 << unit);
+		optidma_set_mode(ap, adev, adev->dma_mode);
+	} else {
+		udcfg |=  (1 << unit);
+		if (ap->port_no) {
+			pci_read_config_byte(pdev, 0x45, &udslave);
+			udslave &= ~(0x03 << dev2);
+			udslave |= (udma << dev2);
+			pci_write_config_byte(pdev, 0x45, udslave);
+		} else {
+			udcfg &= ~(0x30 << dev2);
+			udcfg |= (udma << dev2);
+		}
+	}
+	pci_write_config_byte(pdev, 0x44, udcfg);
+}
+
+/**
+ *	optidma_set_pio_mode	-	PIO setup callback
+ *	@ap: ATA port
+ *	@adev: Device
+ *
+ *	The libata core provides separate functions for handling PIO and
+ *	DMA programming. The architecture of the Firestar makes it easier
+ *	for us to have a common function so we provide wrappers
+ */
+ 
+static void optidma_set_pio_mode(struct ata_port *ap, struct ata_device *adev)
+{
+	optidma_set_mode(ap, adev, adev->pio_mode);
+}
+
+/**
+ *	optidma_set_dma_mode	-	DMA setup callback
+ *	@ap: ATA port
+ *	@adev: Device
+ *
+ *	The libata core provides separate functions for handling PIO and
+ *	DMA programming. The architecture of the Firestar makes it easier
+ *	for us to have a common function so we provide wrappers
+ */
+ 
+static void optidma_set_dma_mode(struct ata_port *ap, struct ata_device *adev)
+{
+	optidma_set_mode(ap, adev, adev->dma_mode);
+}
+
+/**
+ *	optiplus_set_pio_mode	-	PIO setup callback
+ *	@ap: ATA port
+ *	@adev: Device
+ *
+ *	The libata core provides separate functions for handling PIO and
+ *	DMA programming. The architecture of the Firestar makes it easier
+ *	for us to have a common function so we provide wrappers
+ */
+ 
+static void optiplus_set_pio_mode(struct ata_port *ap, struct ata_device *adev)
+{
+	optiplus_set_mode(ap, adev, adev->pio_mode);
+}
+
+/**
+ *	optiplus_set_dma_mode	-	DMA setup callback
+ *	@ap: ATA port
+ *	@adev: Device
+ *
+ *	The libata core provides separate functions for handling PIO and
+ *	DMA programming. The architecture of the Firestar makes it easier
+ *	for us to have a common function so we provide wrappers
+ */
+ 
+static void optiplus_set_dma_mode(struct ata_port *ap, struct ata_device *adev)
+{
+	optiplus_set_mode(ap, adev, adev->dma_mode);
+}
+
+/**
+ *	optidma_make_bits	-	PCI setup helper
+ *	@adev: ATA device
+ *
+ *	Turn the ATA device setup into PCI configuration bits
+ *	for register 0x43 and return the two bits needed.
+ */
+ 
+static u8 optidma_make_bits43(struct ata_device *adev)
+{
+	static const u8 bits43[5] = {
+		0, 0, 0, 1, 2
+	};
+	if (!ata_dev_enabled(adev))
+		return 0;
+	if (adev->dma_mode)
+		return adev->dma_mode - XFER_MW_DMA_0;
+	return bits43[adev->pio_mode - XFER_PIO_0];
+}
+
+/**
+ *	optidma_post_set_mode	-	finalize PCI setup
+ *	@ap: port to set up
+ *
+ *	Finalise the configuration by writing the nibble of extra bits
+ *	of data into the chip.
+ */
+ 
+static void optidma_post_set_mode(struct ata_port *ap)
+{
+	u8 r;
+	int nybble = 4 * ap->port_no;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	pci_read_config_byte(pdev, 0x43, &r);
+	
+	r &= (0x0F << nybble);
+	r |= (optidma_make_bits43(&ap->device[0]) + 
+	     (optidma_make_bits43(&ap->device[0]) << 2)) << nybble;
+
+	pci_write_config_byte(pdev, 0x43, r);
+}
+
+static struct scsi_host_template optidma_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations optidma_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= optidma_set_pio_mode,
+	.set_dmamode	= optidma_set_dma_mode,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.error_handler	= optidma_error_handler,
+	.post_set_mode	= optidma_post_set_mode,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations optiplus_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= optiplus_set_pio_mode,
+	.set_dmamode	= optiplus_set_dma_mode,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	.error_handler	= optidma_error_handler,
+	.post_set_mode	= optidma_post_set_mode,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	optiplus_with_udma	-	Look for UDMA capable setup
+ *	@pdev; ATA controller
+ */
+ 
+static int optiplus_with_udma(struct pci_dev *pdev)
+{
+	u8 r;
+	int ret = 0;
+	int ioport = 0x22;
+	struct pci_dev *dev1;
+	
+	/* Find function 1 */
+	dev1 = pci_get_device(0x1045, 0xC701, NULL);
+	if(dev1 == NULL)
+		return 0;
+	
+	/* Rev must be >= 0x10 */
+	pci_read_config_byte(dev1, 0x08, &r);
+	if (r < 0x10)
+		goto done_nomsg;
+	/* Read the chipset system configuration to check our mode */
+	pci_read_config_byte(dev1, 0x5F, &r);
+	ioport |= (r << 8);
+	outb(0x10, ioport);
+	/* Must be 66Mhz sync */
+	if ((inb(ioport + 2) & 1) == 0)
+		goto done;
+
+	/* Check the ATA arbitration/timing is suitable */
+	pci_read_config_byte(pdev, 0x42, &r);
+	if ((r & 0x36) != 0x36)
+		goto done;
+	pci_read_config_byte(dev1, 0x52, &r);
+	if (r & 0x80)	/* IDEDIR disabled */
+		ret = 1;
+done:		
+	printk(KERN_WARNING "UDMA not supported in this configuration.\n");
+done_nomsg:		/* Wrong chip revision */
+	pci_dev_put(dev1);
+	return ret;
+}
+
+static int optidma_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info_82c700 = {
+		.sht = &optidma_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &optidma_port_ops
+	};
+	static struct ata_port_info info_82c700_udma = {
+		.sht = &optidma_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x07,
+		.port_ops = &optiplus_port_ops
+	};
+	static struct ata_port_info *port_info[2];
+	struct ata_port_info *info = &info_82c700;
+	static int printed_version;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
+
+	/* Fixed location chipset magic */
+	inw(0x1F1);
+	inw(0x1F1);
+	pci_clock = inb(0x1F5) & 1;		/* 0 = 33Mhz, 1 = 25Mhz */
+	
+	if (optiplus_with_udma(dev))
+		info = &info_82c700_udma;
+
+	port_info[0] = port_info[1] = info;
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static const struct pci_device_id optidma[] = {
+	{ PCI_DEVICE(0x1045, 0xD568), },	/* Opti 82C700 */
+	{ 0, },
+};
+
+static struct pci_driver optidma_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= optidma,
+	.probe 		= optidma_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init optidma_init(void)
+{
+	return pci_register_driver(&optidma_pci_driver);
+}
+
+
+static void __exit optidma_exit(void)
+{
+	pci_unregister_driver(&optidma_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Opti Firestar/Firestar Plus");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, optidma);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(optidma_init);
+module_exit(optidma_exit);
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
new file mode 100644
index 000000000000..2abe0a3bbb92
--- /dev/null
+++ b/drivers/ata/pata_pcmcia.c
@@ -0,0 +1,393 @@
+/*
+ *   pata_pcmcia.c - PCMCIA PATA controller driver.
+ *   Copyright 2005-2006 Red Hat Inc <alan@redhat.com>, all rights reserved.
+ *   PCMCIA ident update Copyright 2006 Marcin Juszkiewicz 
+ *						<openembedded@hrw.one.pl>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ *   Heavily based upon ide-cs.c
+ *   The initial developer of the original code is David A. Hinds
+ *   <dahinds@users.sourceforge.net>.  Portions created by David A. Hinds
+ *   are Copyright (C) 1999 David A. Hinds.  All Rights Reserved.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/ata.h>
+#include <linux/libata.h>
+
+#include <pcmcia/cs_types.h>
+#include <pcmcia/cs.h>
+#include <pcmcia/cistpl.h>
+#include <pcmcia/ds.h>
+#include <pcmcia/cisreg.h>
+#include <pcmcia/ciscode.h>
+
+
+#define DRV_NAME "pata_pcmcia"
+#define DRV_VERSION "0.2.9"
+
+/*
+ *	Private data structure to glue stuff together
+ */
+
+struct ata_pcmcia_info {
+	struct pcmcia_device *pdev;
+	int		ndev;
+	dev_node_t	node;
+};
+
+static struct scsi_host_template pcmcia_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations pcmcia_port_ops = {
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ata_bmdma_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer_noirq,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+#define CS_CHECK(fn, ret) \
+do { last_fn = (fn); if ((last_ret = (ret)) != 0) goto cs_failed; } while (0)
+
+/**
+ *	pcmcia_init_one		-	attach a PCMCIA interface
+ *	@pdev: pcmcia device
+ *
+ *	Register a PCMCIA IDE interface. Such interfaces are PIO 0 and
+ *	shared IRQ.
+ */
+
+static int pcmcia_init_one(struct pcmcia_device *pdev)
+{
+	struct ata_probe_ent ae;
+	struct ata_pcmcia_info *info;
+	tuple_t tuple;
+	struct {
+		unsigned short buf[128];
+		cisparse_t parse;
+		config_info_t conf;
+		cistpl_cftable_entry_t dflt;
+	} *stk = NULL;
+	cistpl_cftable_entry_t *cfg;
+	int pass, last_ret = 0, last_fn = 0, is_kme = 0, ret = -ENOMEM;
+	unsigned long io_base, ctl_base;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (info == NULL)
+		return -ENOMEM;
+
+	/* Glue stuff together. FIXME: We may be able to get rid of info with care */
+	info->pdev = pdev;
+	pdev->priv = info;
+
+	/* Set up attributes in order to probe card and get resources */
+	pdev->io.Attributes1 = IO_DATA_PATH_WIDTH_AUTO;
+	pdev->io.Attributes2 = IO_DATA_PATH_WIDTH_8;
+	pdev->io.IOAddrLines = 3;
+	pdev->irq.Attributes = IRQ_TYPE_DYNAMIC_SHARING;
+	pdev->irq.IRQInfo1 = IRQ_LEVEL_ID;
+	pdev->conf.Attributes = CONF_ENABLE_IRQ;
+	pdev->conf.IntType = INT_MEMORY_AND_IO;
+
+	/* Allocate resoure probing structures */
+
+	stk = kzalloc(sizeof(*stk), GFP_KERNEL);
+	if (!stk)
+		goto out1;
+
+	cfg = &stk->parse.cftable_entry;
+
+	/* Tuples we are walking */
+	tuple.TupleData = (cisdata_t *)&stk->buf;
+	tuple.TupleOffset = 0;
+	tuple.TupleDataMax = 255;
+	tuple.Attributes = 0;
+	tuple.DesiredTuple = CISTPL_CONFIG;
+
+	CS_CHECK(GetFirstTuple, pcmcia_get_first_tuple(pdev, &tuple));
+	CS_CHECK(GetTupleData, pcmcia_get_tuple_data(pdev, &tuple));
+	CS_CHECK(ParseTuple, pcmcia_parse_tuple(pdev, &tuple, &stk->parse));
+	pdev->conf.ConfigBase = stk->parse.config.base;
+	pdev->conf.Present = stk->parse.config.rmask[0];
+
+	/* See if we have a manufacturer identifier. Use it to set is_kme for
+	   vendor quirks */
+	tuple.DesiredTuple = CISTPL_MANFID;
+	if (!pcmcia_get_first_tuple(pdev, &tuple) && !pcmcia_get_tuple_data(pdev, &tuple) && !pcmcia_parse_tuple(pdev, &tuple, &stk->parse))
+			is_kme = ((stk->parse.manfid.manf == MANFID_KME) && ((stk->parse.manfid.card == PRODID_KME_KXLC005_A) || (stk->parse.manfid.card == PRODID_KME_KXLC005_B)));
+
+	/* Not sure if this is right... look up the current Vcc */
+	CS_CHECK(GetConfigurationInfo, pcmcia_get_configuration_info(pdev, &stk->conf));
+/*	link->conf.Vcc = stk->conf.Vcc; */
+
+	pass = io_base = ctl_base = 0;
+	tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY;
+	tuple.Attributes = 0;
+	CS_CHECK(GetFirstTuple, pcmcia_get_first_tuple(pdev, &tuple));
+
+	/* Now munch the resources looking for a suitable set */
+	while (1) {
+		if (pcmcia_get_tuple_data(pdev, &tuple) != 0)
+			goto next_entry;
+		if (pcmcia_parse_tuple(pdev, &tuple, &stk->parse) != 0)
+			goto next_entry;
+		/* Check for matching Vcc, unless we're desperate */
+		if (!pass) {
+			if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
+				if (stk->conf.Vcc != cfg->vcc.param[CISTPL_POWER_VNOM] / 10000)
+					goto next_entry;
+			} else if (stk->dflt.vcc.present & (1 << CISTPL_POWER_VNOM)) {
+				if (stk->conf.Vcc != stk->dflt.vcc.param[CISTPL_POWER_VNOM] / 10000)
+					goto next_entry;
+			}
+		}
+
+		if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
+			pdev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
+		else if (stk->dflt.vpp1.present & (1 << CISTPL_POWER_VNOM))
+			pdev->conf.Vpp = stk->dflt.vpp1.param[CISTPL_POWER_VNOM] / 10000;
+
+		if ((cfg->io.nwin > 0) || (stk->dflt.io.nwin > 0)) {
+			cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &stk->dflt.io;
+			pdev->conf.ConfigIndex = cfg->index;
+			pdev->io.BasePort1 = io->win[0].base;
+			pdev->io.IOAddrLines = io->flags & CISTPL_IO_LINES_MASK;
+			if (!(io->flags & CISTPL_IO_16BIT))
+				pdev->io.Attributes1 = IO_DATA_PATH_WIDTH_8;
+			if (io->nwin == 2) {
+				pdev->io.NumPorts1 = 8;
+				pdev->io.BasePort2 = io->win[1].base;
+				pdev->io.NumPorts2 = (is_kme) ? 2 : 1;
+				if (pcmcia_request_io(pdev, &pdev->io) != 0)
+					goto next_entry;
+				io_base = pdev->io.BasePort1;
+				ctl_base = pdev->io.BasePort2;
+			} else if ((io->nwin == 1) && (io->win[0].len >= 16)) {
+				pdev->io.NumPorts1 = io->win[0].len;
+				pdev->io.NumPorts2 = 0;
+				if (pcmcia_request_io(pdev, &pdev->io) != 0)
+					goto next_entry;
+				io_base = pdev->io.BasePort1;
+				ctl_base = pdev->io.BasePort1 + 0x0e;
+			} else goto next_entry;
+			/* If we've got this far, we're done */
+			break;
+		}
+next_entry:
+		if (cfg->flags & CISTPL_CFTABLE_DEFAULT)
+			memcpy(&stk->dflt, cfg, sizeof(stk->dflt));
+		if (pass) {
+			CS_CHECK(GetNextTuple, pcmcia_get_next_tuple(pdev, &tuple));
+		} else if (pcmcia_get_next_tuple(pdev, &tuple) != 0) {
+			CS_CHECK(GetFirstTuple, pcmcia_get_first_tuple(pdev, &tuple));
+			memset(&stk->dflt, 0, sizeof(stk->dflt));
+			pass++;
+		}
+	}
+
+	CS_CHECK(RequestIRQ, pcmcia_request_irq(pdev, &pdev->irq));
+	CS_CHECK(RequestConfiguration, pcmcia_request_configuration(pdev, &pdev->conf));
+
+	/* Success. Disable the IRQ nIEN line, do quirks */
+	outb(0x02, ctl_base);
+	if (is_kme)
+		outb(0x81, ctl_base + 0x01);
+
+	/* FIXME: Could be more ports at base + 0x10 but we only deal with
+	   one right now */
+	if (pdev->io.NumPorts1 >= 0x20)
+		printk(KERN_WARNING DRV_NAME ": second channel not yet supported.\n");
+
+	/*
+ 	 *	Having done the PCMCIA plumbing the ATA side is relatively
+ 	 *	sane.
+	 */
+
+	memset(&ae, 0, sizeof(struct ata_probe_ent));
+	INIT_LIST_HEAD(&ae.node);
+	ae.dev = &pdev->dev;
+	ae.port_ops = &pcmcia_port_ops;
+	ae.sht = &pcmcia_sht;
+	ae.n_ports = 1;
+	ae.pio_mask = 1;		/* ISA so PIO 0 cycles */
+	ae.irq = pdev->irq.AssignedIRQ;
+	ae.irq_flags = SA_SHIRQ;
+	ae.port_flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST;
+	ae.port[0].cmd_addr = io_base;
+	ae.port[0].altstatus_addr = ctl_base;
+	ae.port[0].ctl_addr = ctl_base;
+	ata_std_ports(&ae.port[0]);
+
+	if (ata_device_add(&ae) == 0)
+		goto failed;
+
+	info->ndev = 1;
+	kfree(stk);
+	return 0;
+
+cs_failed:
+	cs_error(pdev, last_fn, last_ret);
+failed:
+	kfree(stk);
+	info->ndev = 0;
+	pcmcia_disable_device(pdev);
+out1:
+	kfree(info);
+	return ret;
+}
+
+/**
+ *	pcmcia_remove_one	-	unplug an pcmcia interface
+ *	@pdev: pcmcia device
+ *
+ *	A PCMCIA ATA device has been unplugged. Perform the needed
+ *	cleanup. Also called on module unload for any active devices.
+ */
+
+static void pcmcia_remove_one(struct pcmcia_device *pdev)
+{
+	struct ata_pcmcia_info *info = pdev->priv;
+	struct device *dev = &pdev->dev;
+
+	if (info != NULL) {
+		/* If we have attached the device to the ATA layer, detach it */
+		if (info->ndev) {
+			struct ata_host *host = dev_get_drvdata(dev);
+			ata_host_remove(host);
+			dev_set_drvdata(dev, NULL);
+		}
+		info->ndev = 0;
+		pdev->priv = NULL;
+	}
+	pcmcia_disable_device(pdev);
+	kfree(info);
+}
+
+static struct pcmcia_device_id pcmcia_devices[] = {
+	PCMCIA_DEVICE_FUNC_ID(4),
+	PCMCIA_DEVICE_MANF_CARD(0x0007, 0x0000),	/* Hitachi */
+	PCMCIA_DEVICE_MANF_CARD(0x0032, 0x0704),
+	PCMCIA_DEVICE_MANF_CARD(0x0045, 0x0401),
+	PCMCIA_DEVICE_MANF_CARD(0x0098, 0x0000),	/* Toshiba */
+	PCMCIA_DEVICE_MANF_CARD(0x00a4, 0x002d),
+	PCMCIA_DEVICE_MANF_CARD(0x00ce, 0x0000),	/* Samsung */
+ 	PCMCIA_DEVICE_MANF_CARD(0x0319, 0x0000),	/* Hitachi */
+	PCMCIA_DEVICE_MANF_CARD(0x2080, 0x0001),
+	PCMCIA_DEVICE_MANF_CARD(0x4e01, 0x0200),	/* Lexar */
+	PCMCIA_DEVICE_PROD_ID123("Caravelle", "PSC-IDE ", "PSC000", 0x8c36137c, 0xd0693ab8, 0x2768a9f0),
+	PCMCIA_DEVICE_PROD_ID123("CDROM", "IDE", "MCD-601p", 0x1b9179ca, 0xede88951, 0x0d902f74),
+	PCMCIA_DEVICE_PROD_ID123("PCMCIA", "IDE CARD", "F1", 0x281f1c5d, 0x1907960c, 0xf7fde8b9),
+	PCMCIA_DEVICE_PROD_ID12("ARGOSY", "CD-ROM", 0x78f308dc, 0x66536591),
+	PCMCIA_DEVICE_PROD_ID12("ARGOSY", "PnPIDE", 0x78f308dc, 0x0c694728),
+	PCMCIA_DEVICE_PROD_ID12("CNF CD-M", "CD-ROM", 0x7d93b852, 0x66536591),
+	PCMCIA_DEVICE_PROD_ID12("Creative Technology Ltd.", "PCMCIA CD-ROM Interface Card", 0xff8c8a45, 0xfe8020c4),
+	PCMCIA_DEVICE_PROD_ID12("Digital Equipment Corporation.", "Digital Mobile Media CD-ROM", 0x17692a66, 0xef1dcbde),
+	PCMCIA_DEVICE_PROD_ID12("EXP", "CD+GAME", 0x6f58c983, 0x63c13aaf),
+	PCMCIA_DEVICE_PROD_ID12("EXP   ", "CD-ROM", 0x0a5c52fd, 0x66536591),
+	PCMCIA_DEVICE_PROD_ID12("EXP   ", "PnPIDE", 0x0a5c52fd, 0x0c694728),
+	PCMCIA_DEVICE_PROD_ID12("FREECOM", "PCCARD-IDE", 0x5714cbf7, 0x48e0ab8e),
+	PCMCIA_DEVICE_PROD_ID12("HITACHI", "FLASH", 0xf4f43949, 0x9eb86aae),
+	PCMCIA_DEVICE_PROD_ID12("HITACHI", "microdrive", 0xf4f43949, 0xa6d76178),
+	PCMCIA_DEVICE_PROD_ID12("IBM", "microdrive", 0xb569a6e5, 0xa6d76178),
+	PCMCIA_DEVICE_PROD_ID12("IBM", "IBM17JSSFP20", 0xb569a6e5, 0xf2508753),
+	PCMCIA_DEVICE_PROD_ID12("IO DATA", "CBIDE2      ", 0x547e66dc, 0x8671043b),
+	PCMCIA_DEVICE_PROD_ID12("IO DATA", "PCIDE", 0x547e66dc, 0x5c5ab149),
+	PCMCIA_DEVICE_PROD_ID12("IO DATA", "PCIDEII", 0x547e66dc, 0xb3662674),
+	PCMCIA_DEVICE_PROD_ID12("LOOKMEET", "CBIDE2      ", 0xe37be2b5, 0x8671043b),
+	PCMCIA_DEVICE_PROD_ID12("M-Systems", "CF500", 0x7ed2ad87, 0x7a13045c),
+	PCMCIA_DEVICE_PROD_ID2("NinjaATA-", 0xebe0bd79),
+	PCMCIA_DEVICE_PROD_ID12("PCMCIA", "CD-ROM", 0x281f1c5d, 0x66536591),
+	PCMCIA_DEVICE_PROD_ID12("PCMCIA", "PnPIDE", 0x281f1c5d, 0x0c694728),
+	PCMCIA_DEVICE_PROD_ID12("SHUTTLE TECHNOLOGY LTD.", "PCCARD-IDE/ATAPI Adapter", 0x4a3f0ba0, 0x322560e1),
+	PCMCIA_DEVICE_PROD_ID12("SEAGATE", "ST1", 0x87c1b330, 0xe1f30883),
+	PCMCIA_DEVICE_PROD_ID12("SAMSUNG", "04/05/06", 0x43d74cb4, 0x6a22777d),
+	PCMCIA_DEVICE_PROD_ID12("SMI VENDOR", "SMI PRODUCT", 0x30896c92, 0x703cc5f6),
+	PCMCIA_DEVICE_PROD_ID12("TOSHIBA", "MK2001MPL", 0xb4585a1a, 0x3489e003),
+	PCMCIA_DEVICE_PROD_ID12("WIT", "IDE16", 0x244e5994, 0x3e232852),
+	PCMCIA_DEVICE_PROD_ID1("STI Flash", 0xe4a13209),
+	PCMCIA_DEVICE_PROD_ID12("STI", "Flash 5.0", 0xbf2df18d, 0x8cb57a0e),
+	PCMCIA_MFC_DEVICE_PROD_ID12(1, "SanDisk", "ConnectPlus", 0x7a954bd9, 0x74be00c6),
+	PCMCIA_DEVICE_NULL,
+};
+
+MODULE_DEVICE_TABLE(pcmcia, pcmcia_devices);
+
+static struct pcmcia_driver pcmcia_driver = {
+	.owner		= THIS_MODULE,
+	.drv = {
+		.name		= DRV_NAME,
+	},
+	.id_table	= pcmcia_devices,
+	.probe		= pcmcia_init_one,
+	.remove		= pcmcia_remove_one,
+};
+
+static int __init pcmcia_init(void)
+{
+	return pcmcia_register_driver(&pcmcia_driver);
+}
+
+static void __exit pcmcia_exit(void)
+{
+	pcmcia_unregister_driver(&pcmcia_driver);
+}
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for PCMCIA ATA");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+
+module_init(pcmcia_init);
+module_exit(pcmcia_exit);
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
new file mode 100644
index 000000000000..56b8c1ee2937
--- /dev/null
+++ b/drivers/ata/pata_pdc2027x.c
@@ -0,0 +1,869 @@
+/*
+ *  Promise PATA TX2/TX4/TX2000/133 IDE driver for pdc20268 to pdc20277.
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ *
+ *  Ported to libata by:
+ *  Albert Lee <albertcc@tw.ibm.com> IBM Corporation
+ *
+ *  Copyright (C) 1998-2002		Andre Hedrick <andre@linux-ide.org>
+ *  Portions Copyright (C) 1999 Promise Technology, Inc.
+ *
+ *  Author: Frank Tiernan (frankt@promise.com)
+ *  Released under terms of General Public License
+ *
+ *
+ *  libata documentation is available via 'make {ps|pdf}docs',
+ *  as Documentation/DocBook/libata.*
+ *
+ *  Hardware information only available under NDA.
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_cmnd.h>
+#include <linux/libata.h>
+#include <asm/io.h>
+
+#define DRV_NAME	"pata_pdc2027x"
+#define DRV_VERSION	"0.74-ac3"
+#undef PDC_DEBUG
+
+#ifdef PDC_DEBUG
+#define PDPRINTK(fmt, args...) printk(KERN_ERR "%s: " fmt, __FUNCTION__, ## args)
+#else
+#define PDPRINTK(fmt, args...)
+#endif
+
+enum {
+	PDC_UDMA_100		= 0,
+	PDC_UDMA_133		= 1,
+
+	PDC_100_MHZ		= 100000000,
+	PDC_133_MHZ		= 133333333,
+
+	PDC_SYS_CTL		= 0x1100,
+	PDC_ATA_CTL		= 0x1104,
+	PDC_GLOBAL_CTL		= 0x1108,
+	PDC_CTCR0		= 0x110C,
+	PDC_CTCR1		= 0x1110,
+	PDC_BYTE_COUNT		= 0x1120,
+	PDC_PLL_CTL		= 0x1202,
+};
+
+static int pdc2027x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
+static void pdc2027x_remove_one(struct pci_dev *pdev);
+static void pdc2027x_error_handler(struct ata_port *ap);
+static void pdc2027x_set_piomode(struct ata_port *ap, struct ata_device *adev);
+static void pdc2027x_set_dmamode(struct ata_port *ap, struct ata_device *adev);
+static void pdc2027x_post_set_mode(struct ata_port *ap);
+static int pdc2027x_check_atapi_dma(struct ata_queued_cmd *qc);
+
+/*
+ * ATA Timing Tables based on 133MHz controller clock.
+ * These tables are only used when the controller is in 133MHz clock.
+ * If the controller is in 100MHz clock, the ASIC hardware will
+ * set the timing registers automatically when "set feature" command
+ * is issued to the device. However, if the controller clock is 133MHz,
+ * the following tables must be used.
+ */
+static struct pdc2027x_pio_timing {
+	u8 value0, value1, value2;
+} pdc2027x_pio_timing_tbl [] = {
+	{ 0xfb, 0x2b, 0xac }, /* PIO mode 0 */
+	{ 0x46, 0x29, 0xa4 }, /* PIO mode 1 */
+	{ 0x23, 0x26, 0x64 }, /* PIO mode 2 */
+	{ 0x27, 0x0d, 0x35 }, /* PIO mode 3, IORDY on, Prefetch off */
+	{ 0x23, 0x09, 0x25 }, /* PIO mode 4, IORDY on, Prefetch off */
+};
+
+static struct pdc2027x_mdma_timing {
+	u8 value0, value1;
+} pdc2027x_mdma_timing_tbl [] = {
+	{ 0xdf, 0x5f }, /* MDMA mode 0 */
+	{ 0x6b, 0x27 }, /* MDMA mode 1 */
+	{ 0x69, 0x25 }, /* MDMA mode 2 */
+};
+
+static struct pdc2027x_udma_timing {
+	u8 value0, value1, value2;
+} pdc2027x_udma_timing_tbl [] = {
+	{ 0x4a, 0x0f, 0xd5 }, /* UDMA mode 0 */
+	{ 0x3a, 0x0a, 0xd0 }, /* UDMA mode 1 */
+	{ 0x2a, 0x07, 0xcd }, /* UDMA mode 2 */
+	{ 0x1a, 0x05, 0xcd }, /* UDMA mode 3 */
+	{ 0x1a, 0x03, 0xcd }, /* UDMA mode 4 */
+	{ 0x1a, 0x02, 0xcb }, /* UDMA mode 5 */
+	{ 0x1a, 0x01, 0xcb }, /* UDMA mode 6 */
+};
+
+static const struct pci_device_id pdc2027x_pci_tbl[] = {
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20268, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_100 },
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20269, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_133 },
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20270, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_100 },
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20271, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_133 },
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20275, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_133 },
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20276, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_133 },
+	{ PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20277, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PDC_UDMA_133 },
+	{ }	/* terminate list */
+};
+
+static struct pci_driver pdc2027x_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= pdc2027x_pci_tbl,
+	.probe			= pdc2027x_init_one,
+	.remove			= __devexit_p(pdc2027x_remove_one),
+};
+
+static struct scsi_host_template pdc2027x_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations pdc2027x_pata100_ops = {
+	.port_disable		= ata_port_disable,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.check_atapi_dma	= pdc2027x_check_atapi_dma,
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_mmio_data_xfer,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= pdc2027x_error_handler,
+	.post_internal_cmd 	= ata_bmdma_post_internal_cmd,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_pci_host_stop,
+};
+
+static struct ata_port_operations pdc2027x_pata133_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= pdc2027x_set_piomode,
+	.set_dmamode		= pdc2027x_set_dmamode,
+	.post_set_mode		= pdc2027x_post_set_mode,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.check_atapi_dma	= pdc2027x_check_atapi_dma,
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_mmio_data_xfer,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= pdc2027x_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_pci_host_stop,
+};
+
+static struct ata_port_info pdc2027x_port_info[] = {
+	/* PDC_UDMA_100 */
+	{
+		.sht		= &pdc2027x_sht,
+		.flags		= ATA_FLAG_NO_LEGACY | ATA_FLAG_SLAVE_POSS |
+		                  ATA_FLAG_MMIO,
+		.pio_mask	= 0x1f, /* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma0-2 */
+		.udma_mask	= ATA_UDMA5, /* udma0-5 */
+		.port_ops	= &pdc2027x_pata100_ops,
+	},
+	/* PDC_UDMA_133 */
+	{
+		.sht		= &pdc2027x_sht,
+		.flags		= ATA_FLAG_NO_LEGACY | ATA_FLAG_SLAVE_POSS |
+                        	  ATA_FLAG_MMIO,
+		.pio_mask	= 0x1f, /* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma0-2 */
+		.udma_mask	= ATA_UDMA6, /* udma0-6 */
+		.port_ops	= &pdc2027x_pata133_ops,
+	},
+};
+
+MODULE_AUTHOR("Andre Hedrick, Frank Tiernan, Albert Lee");
+MODULE_DESCRIPTION("libata driver module for Promise PDC20268 to PDC20277");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+MODULE_DEVICE_TABLE(pci, pdc2027x_pci_tbl);
+
+/**
+ *	port_mmio - Get the MMIO address of PDC2027x extended registers
+ *	@ap: Port
+ *	@offset: offset from mmio base
+ */
+static inline void* port_mmio(struct ata_port *ap, unsigned int offset)
+{
+	return ap->host->mmio_base + ap->port_no * 0x100 + offset;
+}
+
+/**
+ *	dev_mmio - Get the MMIO address of PDC2027x extended registers
+ *	@ap: Port
+ *	@adev: device
+ *	@offset: offset from mmio base
+ */
+static inline void* dev_mmio(struct ata_port *ap, struct ata_device *adev, unsigned int offset)
+{
+	u8 adj = (adev->devno) ? 0x08 : 0x00;
+	return port_mmio(ap, offset) + adj;
+}
+
+/**
+ *	pdc2027x_pata_cbl_detect - Probe host controller cable detect info
+ *	@ap: Port for which cable detect info is desired
+ *
+ *	Read 80c cable indicator from Promise extended register.
+ *      This register is latched when the system is reset.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+static void pdc2027x_cbl_detect(struct ata_port *ap)
+{
+	u32 cgcr;
+
+	/* check cable detect results */
+	cgcr = readl(port_mmio(ap, PDC_GLOBAL_CTL));
+	if (cgcr & (1 << 26))
+		goto cbl40;
+
+	PDPRINTK("No cable or 80-conductor cable on port %d\n", ap->port_no);
+
+	ap->cbl = ATA_CBL_PATA80;
+	return;
+
+cbl40:
+	printk(KERN_INFO DRV_NAME ": 40-conductor cable detected on port %d\n", ap->port_no);
+	ap->cbl = ATA_CBL_PATA40;
+	ap->udma_mask &= ATA_UDMA_MASK_40C;
+}
+
+/**
+ * pdc2027x_port_enabled - Check PDC ATA control register to see whether the port is enabled.
+ * @ap: Port to check
+ */
+static inline int pdc2027x_port_enabled(struct ata_port *ap)
+{
+	return readb(port_mmio(ap, PDC_ATA_CTL)) & 0x02;
+}
+
+/**
+ *	pdc2027x_prereset - prereset for PATA host controller
+ *	@ap: Target port
+ *
+ *	Probeinit including cable detection.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static int pdc2027x_prereset(struct ata_port *ap)
+{
+	/* Check whether port enabled */
+	if (!pdc2027x_port_enabled(ap)) {
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	pdc2027x_cbl_detect(ap);
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	pdc2027x_error_handler - Perform reset on PATA port and classify
+ *	@ap: Port to reset
+ *
+ *	Reset PATA phy and classify attached devices.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void pdc2027x_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, pdc2027x_prereset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	pdc2027x_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port to configure
+ *	@adev: um
+ *	@pio: PIO mode, 0 - 4
+ *
+ *	Set PIO mode for device.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void pdc2027x_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int pio = adev->pio_mode - XFER_PIO_0;
+	u32 ctcr0, ctcr1;
+
+	PDPRINTK("adev->pio_mode[%X]\n", adev->pio_mode);
+
+	/* Sanity check */
+	if (pio > 4) {
+		printk(KERN_ERR DRV_NAME ": Unknown pio mode [%d] ignored\n", pio);
+		return;
+
+	}
+
+	/* Set the PIO timing registers using value table for 133MHz */
+	PDPRINTK("Set pio regs... \n");
+
+	ctcr0 = readl(dev_mmio(ap, adev, PDC_CTCR0));
+	ctcr0 &= 0xffff0000;
+	ctcr0 |= pdc2027x_pio_timing_tbl[pio].value0 |
+		(pdc2027x_pio_timing_tbl[pio].value1 << 8);
+	writel(ctcr0, dev_mmio(ap, adev, PDC_CTCR0));
+
+	ctcr1 = readl(dev_mmio(ap, adev, PDC_CTCR1));
+	ctcr1 &= 0x00ffffff;
+	ctcr1 |= (pdc2027x_pio_timing_tbl[pio].value2 << 24);
+	writel(ctcr1, dev_mmio(ap, adev, PDC_CTCR1));
+
+	PDPRINTK("Set pio regs done\n");
+
+	PDPRINTK("Set to pio mode[%u] \n", pio);
+}
+
+/**
+ *	pdc2027x_set_dmamode - Initialize host controller PATA UDMA timings
+ *	@ap: Port to configure
+ *	@adev: um
+ *	@udma: udma mode, XFER_UDMA_0 to XFER_UDMA_6
+ *
+ *	Set UDMA mode for device.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+static void pdc2027x_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int dma_mode = adev->dma_mode;
+	u32 ctcr0, ctcr1;
+
+	if ((dma_mode >= XFER_UDMA_0) &&
+	   (dma_mode <= XFER_UDMA_6)) {
+		/* Set the UDMA timing registers with value table for 133MHz */
+		unsigned int udma_mode = dma_mode & 0x07;
+
+		if (dma_mode == XFER_UDMA_2) {
+			/*
+			 * Turn off tHOLD.
+			 * If tHOLD is '1', the hardware will add half clock for data hold time.
+			 * This code segment seems to be no effect. tHOLD will be overwritten below.
+			 */
+			ctcr1 = readl(dev_mmio(ap, adev, PDC_CTCR1));
+			writel(ctcr1 & ~(1 << 7), dev_mmio(ap, adev, PDC_CTCR1));
+		}
+
+		PDPRINTK("Set udma regs... \n");
+
+		ctcr1 = readl(dev_mmio(ap, adev, PDC_CTCR1));
+		ctcr1 &= 0xff000000;
+		ctcr1 |= pdc2027x_udma_timing_tbl[udma_mode].value0 |
+			(pdc2027x_udma_timing_tbl[udma_mode].value1 << 8) |
+			(pdc2027x_udma_timing_tbl[udma_mode].value2 << 16);
+		writel(ctcr1, dev_mmio(ap, adev, PDC_CTCR1));
+
+		PDPRINTK("Set udma regs done\n");
+
+		PDPRINTK("Set to udma mode[%u] \n", udma_mode);
+
+	} else  if ((dma_mode >= XFER_MW_DMA_0) &&
+		   (dma_mode <= XFER_MW_DMA_2)) {
+		/* Set the MDMA timing registers with value table for 133MHz */
+		unsigned int mdma_mode = dma_mode & 0x07;
+
+		PDPRINTK("Set mdma regs... \n");
+		ctcr0 = readl(dev_mmio(ap, adev, PDC_CTCR0));
+
+		ctcr0 &= 0x0000ffff;
+		ctcr0 |= (pdc2027x_mdma_timing_tbl[mdma_mode].value0 << 16) |
+			(pdc2027x_mdma_timing_tbl[mdma_mode].value1 << 24);
+
+		writel(ctcr0, dev_mmio(ap, adev, PDC_CTCR0));
+		PDPRINTK("Set mdma regs done\n");
+
+		PDPRINTK("Set to mdma mode[%u] \n", mdma_mode);
+	} else {
+		printk(KERN_ERR DRV_NAME ": Unknown dma mode [%u] ignored\n", dma_mode);
+	}
+}
+
+/**
+ *	pdc2027x_post_set_mode - Set the timing registers back to correct values.
+ *	@ap: Port to configure
+ *
+ *	The pdc2027x hardware will look at "SET FEATURES" and change the timing registers
+ *	automatically. The values set by the hardware might be incorrect, under 133Mhz PLL.
+ *	This function overwrites the possibly incorrect values set by the hardware to be correct.
+ */
+static void pdc2027x_post_set_mode(struct ata_port *ap)
+{
+	int i;
+
+	for (i = 0; i < ATA_MAX_DEVICES; i++) {
+		struct ata_device *dev = &ap->device[i];
+
+		if (ata_dev_enabled(dev)) {
+
+			pdc2027x_set_piomode(ap, dev);
+
+			/*
+			 * Enable prefetch if the device support PIO only.
+			 */
+			if (dev->xfer_shift == ATA_SHIFT_PIO) {
+				u32 ctcr1 = readl(dev_mmio(ap, dev, PDC_CTCR1));
+				ctcr1 |= (1 << 25);
+				writel(ctcr1, dev_mmio(ap, dev, PDC_CTCR1));
+
+				PDPRINTK("Turn on prefetch\n");
+			} else {
+				pdc2027x_set_dmamode(ap, dev);
+			}
+		}
+	}
+}
+
+/**
+ *	pdc2027x_check_atapi_dma - Check whether ATAPI DMA can be supported for this command
+ *	@qc: Metadata associated with taskfile to check
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ *
+ *	RETURNS: 0 when ATAPI DMA can be used
+ *		 1 otherwise
+ */
+static int pdc2027x_check_atapi_dma(struct ata_queued_cmd *qc)
+{
+	struct scsi_cmnd *cmd = qc->scsicmd;
+	u8 *scsicmd = cmd->cmnd;
+	int rc = 1; /* atapi dma off by default */
+
+	/*
+	 * This workaround is from Promise's GPL driver.
+	 * If ATAPI DMA is used for commands not in the
+	 * following white list, say MODE_SENSE and REQUEST_SENSE,
+	 * pdc2027x might hit the irq lost problem.
+	 */
+	switch (scsicmd[0]) {
+	case READ_10:
+	case WRITE_10:
+	case READ_12:
+	case WRITE_12:
+	case READ_6:
+	case WRITE_6:
+	case 0xad: /* READ_DVD_STRUCTURE */
+	case 0xbe: /* READ_CD */
+		/* ATAPI DMA is ok */
+		rc = 0;
+		break;
+	default:
+		;
+	}
+
+	return rc;
+}
+
+/**
+ * pdc_read_counter - Read the ctr counter
+ * @probe_ent: for the port address
+ */
+
+static long pdc_read_counter(struct ata_probe_ent *probe_ent)
+{
+	long counter;
+	int retry = 1;
+	u32 bccrl, bccrh, bccrlv, bccrhv;
+
+retry:
+	bccrl = readl(probe_ent->mmio_base + PDC_BYTE_COUNT) & 0xffff;
+	bccrh = readl(probe_ent->mmio_base + PDC_BYTE_COUNT + 0x100) & 0xffff;
+	rmb();
+
+	/* Read the counter values again for verification */
+	bccrlv = readl(probe_ent->mmio_base + PDC_BYTE_COUNT) & 0xffff;
+	bccrhv = readl(probe_ent->mmio_base + PDC_BYTE_COUNT + 0x100) & 0xffff;
+	rmb();
+
+	counter = (bccrh << 15) | bccrl;
+
+	PDPRINTK("bccrh [%X] bccrl [%X]\n", bccrh,  bccrl);
+	PDPRINTK("bccrhv[%X] bccrlv[%X]\n", bccrhv, bccrlv);
+
+	/*
+	 * The 30-bit decreasing counter are read by 2 pieces.
+	 * Incorrect value may be read when both bccrh and bccrl are changing.
+	 * Ex. When 7900 decrease to 78FF, wrong value 7800 might be read.
+	 */
+	if (retry && !(bccrh == bccrhv && bccrl >= bccrlv)) {
+		retry--;
+		PDPRINTK("rereading counter\n");
+		goto retry;
+	}
+
+	return counter;
+}
+
+/**
+ * adjust_pll - Adjust the PLL input clock in Hz.
+ *
+ * @pdc_controller: controller specific information
+ * @probe_ent: For the port address
+ * @pll_clock: The input of PLL in HZ
+ */
+static void pdc_adjust_pll(struct ata_probe_ent *probe_ent, long pll_clock, unsigned int board_idx)
+{
+
+	u16 pll_ctl;
+	long pll_clock_khz = pll_clock / 1000;
+	long pout_required = board_idx? PDC_133_MHZ:PDC_100_MHZ;
+	long ratio = pout_required / pll_clock_khz;
+	int F, R;
+
+	/* Sanity check */
+	if (unlikely(pll_clock_khz < 5000L || pll_clock_khz > 70000L)) {
+		printk(KERN_ERR DRV_NAME ": Invalid PLL input clock %ldkHz, give up!\n", pll_clock_khz);
+		return;
+	}
+
+#ifdef PDC_DEBUG
+	PDPRINTK("pout_required is %ld\n", pout_required);
+
+	/* Show the current clock value of PLL control register
+	 * (maybe already configured by the firmware)
+	 */
+	pll_ctl = readw(probe_ent->mmio_base + PDC_PLL_CTL);
+
+	PDPRINTK("pll_ctl[%X]\n", pll_ctl);
+#endif
+
+	/*
+	 * Calculate the ratio of F, R and OD
+	 * POUT = (F + 2) / (( R + 2) * NO)
+	 */
+	if (ratio < 8600L) { /* 8.6x */
+		/* Using NO = 0x01, R = 0x0D */
+		R = 0x0d;
+	} else if (ratio < 12900L) { /* 12.9x */
+		/* Using NO = 0x01, R = 0x08 */
+		R = 0x08;
+	} else if (ratio < 16100L) { /* 16.1x */
+		/* Using NO = 0x01, R = 0x06 */
+		R = 0x06;
+	} else if (ratio < 64000L) { /* 64x */
+		R = 0x00;
+	} else {
+		/* Invalid ratio */
+		printk(KERN_ERR DRV_NAME ": Invalid ratio %ld, give up!\n", ratio);
+		return;
+	}
+
+	F = (ratio * (R+2)) / 1000 - 2;
+
+	if (unlikely(F < 0 || F > 127)) {
+		/* Invalid F */
+		printk(KERN_ERR DRV_NAME ": F[%d] invalid!\n", F);
+		return;
+	}
+
+	PDPRINTK("F[%d] R[%d] ratio*1000[%ld]\n", F, R, ratio);
+
+	pll_ctl = (R << 8) | F;
+
+	PDPRINTK("Writing pll_ctl[%X]\n", pll_ctl);
+
+	writew(pll_ctl, probe_ent->mmio_base + PDC_PLL_CTL);
+	readw(probe_ent->mmio_base + PDC_PLL_CTL); /* flush */
+
+	/* Wait the PLL circuit to be stable */
+	mdelay(30);
+
+#ifdef PDC_DEBUG
+	/*
+	 *  Show the current clock value of PLL control register
+	 * (maybe configured by the firmware)
+	 */
+	pll_ctl = readw(probe_ent->mmio_base + PDC_PLL_CTL);
+
+	PDPRINTK("pll_ctl[%X]\n", pll_ctl);
+#endif
+
+	return;
+}
+
+/**
+ * detect_pll_input_clock - Detect the PLL input clock in Hz.
+ * @probe_ent: for the port address
+ * Ex. 16949000 on 33MHz PCI bus for pdc20275.
+ *     Half of the PCI clock.
+ */
+static long pdc_detect_pll_input_clock(struct ata_probe_ent *probe_ent)
+{
+	u32 scr;
+	long start_count, end_count;
+	long pll_clock;
+
+	/* Read current counter value */
+	start_count = pdc_read_counter(probe_ent);
+
+	/* Start the test mode */
+	scr = readl(probe_ent->mmio_base + PDC_SYS_CTL);
+	PDPRINTK("scr[%X]\n", scr);
+	writel(scr | (0x01 << 14), probe_ent->mmio_base + PDC_SYS_CTL);
+	readl(probe_ent->mmio_base + PDC_SYS_CTL); /* flush */
+
+	/* Let the counter run for 100 ms. */
+	mdelay(100);
+
+	/* Read the counter values again */
+	end_count = pdc_read_counter(probe_ent);
+
+	/* Stop the test mode */
+	scr = readl(probe_ent->mmio_base + PDC_SYS_CTL);
+	PDPRINTK("scr[%X]\n", scr);
+	writel(scr & ~(0x01 << 14), probe_ent->mmio_base + PDC_SYS_CTL);
+	readl(probe_ent->mmio_base + PDC_SYS_CTL); /* flush */
+
+	/* calculate the input clock in Hz */
+	pll_clock = (start_count - end_count) * 10;
+
+	PDPRINTK("start[%ld] end[%ld] \n", start_count, end_count);
+	PDPRINTK("PLL input clock[%ld]Hz\n", pll_clock);
+
+	return pll_clock;
+}
+
+/**
+ * pdc_hardware_init - Initialize the hardware.
+ * @pdev: instance of pci_dev found
+ * @pdc_controller: controller specific information
+ * @pe:  for the port address
+ */
+static int pdc_hardware_init(struct pci_dev *pdev, struct ata_probe_ent *pe, unsigned int board_idx)
+{
+	long pll_clock;
+
+	/*
+	 * Detect PLL input clock rate.
+	 * On some system, where PCI bus is running at non-standard clock rate.
+	 * Ex. 25MHz or 40MHz, we have to adjust the cycle_time.
+	 * The pdc20275 controller employs PLL circuit to help correct timing registers setting.
+	 */
+	pll_clock = pdc_detect_pll_input_clock(pe);
+
+	if (pll_clock < 0) /* counter overflow? Try again. */
+		pll_clock = pdc_detect_pll_input_clock(pe);
+
+	dev_printk(KERN_INFO, &pdev->dev, "PLL input clock %ld kHz\n", pll_clock/1000);
+
+	/* Adjust PLL control register */
+	pdc_adjust_pll(pe, pll_clock, board_idx);
+
+	return 0;
+}
+
+/**
+ * pdc_ata_setup_port - setup the mmio address
+ * @port: ata ioports to setup
+ * @base: base address
+ */
+static void pdc_ata_setup_port(struct ata_ioports *port, unsigned long base)
+{
+	port->cmd_addr		=
+	port->data_addr		= base;
+	port->feature_addr	=
+	port->error_addr	= base + 0x05;
+	port->nsect_addr	= base + 0x0a;
+	port->lbal_addr		= base + 0x0f;
+	port->lbam_addr		= base + 0x10;
+	port->lbah_addr		= base + 0x15;
+	port->device_addr	= base + 0x1a;
+	port->command_addr	=
+	port->status_addr	= base + 0x1f;
+	port->altstatus_addr	=
+	port->ctl_addr		= base + 0x81a;
+}
+
+/**
+ * pdc2027x_init_one - PCI probe function
+ * Called when an instance of PCI adapter is inserted.
+ * This function checks whether the hardware is supported,
+ * initialize hardware and register an instance of ata_host to
+ * libata by providing struct ata_probe_ent and ata_device_add().
+ * (implements struct pci_driver.probe() )
+ *
+ * @pdev: instance of pci_dev found
+ * @ent:  matching entry in the id_tbl[]
+ */
+static int __devinit pdc2027x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	unsigned int board_idx = (unsigned int) ent->driver_data;
+
+	struct ata_probe_ent *probe_ent = NULL;
+	unsigned long base;
+	void *mmio_base;
+	int rc;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev, "version " DRV_VERSION "\n");
+
+	rc = pci_enable_device(pdev);
+	if (rc)
+		return rc;
+
+	rc = pci_request_regions(pdev, DRV_NAME);
+	if (rc)
+		goto err_out;
+
+	rc = pci_set_dma_mask(pdev, ATA_DMA_MASK);
+	if (rc)
+		goto err_out_regions;
+
+	rc = pci_set_consistent_dma_mask(pdev, ATA_DMA_MASK);
+	if (rc)
+		goto err_out_regions;
+
+	/* Prepare the probe entry */
+	probe_ent = kzalloc(sizeof(*probe_ent), GFP_KERNEL);
+	if (probe_ent == NULL) {
+		rc = -ENOMEM;
+		goto err_out_regions;
+	}
+
+	probe_ent->dev = pci_dev_to_dev(pdev);
+	INIT_LIST_HEAD(&probe_ent->node);
+
+	mmio_base = pci_iomap(pdev, 5, 0);
+	if (!mmio_base) {
+		rc = -ENOMEM;
+		goto err_out_free_ent;
+	}
+
+	base = (unsigned long) mmio_base;
+
+	probe_ent->sht		= pdc2027x_port_info[board_idx].sht;
+	probe_ent->port_flags	= pdc2027x_port_info[board_idx].flags;
+	probe_ent->pio_mask	= pdc2027x_port_info[board_idx].pio_mask;
+	probe_ent->mwdma_mask	= pdc2027x_port_info[board_idx].mwdma_mask;
+	probe_ent->udma_mask	= pdc2027x_port_info[board_idx].udma_mask;
+	probe_ent->port_ops	= pdc2027x_port_info[board_idx].port_ops;
+
+       	probe_ent->irq = pdev->irq;
+       	probe_ent->irq_flags = SA_SHIRQ;
+	probe_ent->mmio_base = mmio_base;
+
+	pdc_ata_setup_port(&probe_ent->port[0], base + 0x17c0);
+	probe_ent->port[0].bmdma_addr = base + 0x1000;
+	pdc_ata_setup_port(&probe_ent->port[1], base + 0x15c0);
+	probe_ent->port[1].bmdma_addr = base + 0x1008;
+
+	probe_ent->n_ports = 2;
+
+	pci_set_master(pdev);
+	//pci_enable_intx(pdev);
+
+	/* initialize adapter */
+	if (pdc_hardware_init(pdev, probe_ent, board_idx) != 0)
+		goto err_out_free_ent;
+
+	ata_device_add(probe_ent);
+	kfree(probe_ent);
+
+	return 0;
+
+err_out_free_ent:
+	kfree(probe_ent);
+err_out_regions:
+	pci_release_regions(pdev);
+err_out:
+	pci_disable_device(pdev);
+	return rc;
+}
+
+/**
+ * pdc2027x_remove_one - Called to remove a single instance of the
+ * adapter.
+ *
+ * @dev: The PCI device to remove.
+ * FIXME: module load/unload not working yet
+ */
+static void __devexit pdc2027x_remove_one(struct pci_dev *pdev)
+{
+	ata_pci_remove_one(pdev);
+}
+
+/**
+ * pdc2027x_init - Called after this module is loaded into the kernel.
+ */
+static int __init pdc2027x_init(void)
+{
+	return pci_module_init(&pdc2027x_pci_driver);
+}
+
+/**
+ * pdc2027x_exit - Called before this module unloaded from the kernel
+ */
+static void __exit pdc2027x_exit(void)
+{
+	pci_unregister_driver(&pdc2027x_pci_driver);
+}
+
+module_init(pdc2027x_init);
+module_exit(pdc2027x_exit);
diff --git a/drivers/ata/pata_pdc202xx_old.c b/drivers/ata/pata_pdc202xx_old.c
new file mode 100644
index 000000000000..6cb52b0e7696
--- /dev/null
+++ b/drivers/ata/pata_pdc202xx_old.c
@@ -0,0 +1,423 @@
+/*
+ * pata_pdc202xx_old.c 	- Promise PDC202xx PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based in part on linux/drivers/ide/pci/pdc202xx_old.c
+ *
+ * First cut with LBA48/ATAPI
+ *
+ * TODO:
+ *	Channel interlock/reset on both required ?
+ */
+ 
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_pdc202xx_old"
+#define DRV_VERSION "0.2.1"
+
+/**
+ *	pdc2024x_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+ 
+static int pdc2024x_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+
+static void pdc2024x_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, pdc2024x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+
+static int pdc2026x_pre_reset(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u16 cis;
+	
+	pci_read_config_word(pdev, 0x50, &cis);
+	if (cis & (1 << (10 + ap->port_no)))
+		ap->cbl = ATA_CBL_PATA80;
+	else
+		ap->cbl = ATA_CBL_PATA40;
+
+	return ata_std_prereset(ap);
+}
+
+static void pdc2026x_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, pdc2026x_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	pdc_configure_piomode	-	set chip PIO timing
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *	@pio: PIO mode
+ *
+ *	Called to do the PIO mode setup. Our timing registers are shared
+ *	so a configure_dmamode call will undo any work we do here and vice
+ *	versa
+ */
+ 
+static void pdc_configure_piomode(struct ata_port *ap, struct ata_device *adev, int pio)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int port = 0x60 + 4 * ap->port_no + 2 * adev->devno;
+	static u16 pio_timing[5] = {
+		0x0913, 0x050C , 0x0308, 0x0206, 0x0104
+	};
+	u8 r_ap, r_bp;
+
+	pci_read_config_byte(pdev, port, &r_ap);
+	pci_read_config_byte(pdev, port + 1, &r_bp);
+	r_ap &= ~0x3F;	/* Preserve ERRDY_EN, SYNC_IN */
+	r_bp &= ~0x07;
+	r_ap |= (pio_timing[pio] >> 8);
+	r_bp |= (pio_timing[pio] & 0xFF);
+	
+	if (ata_pio_need_iordy(adev))
+		r_ap |= 0x20;	/* IORDY enable */
+	if (adev->class == ATA_DEV_ATA)
+		r_ap |= 0x10;	/* FIFO enable */
+	pci_write_config_byte(pdev, port, r_ap);
+	pci_write_config_byte(pdev, port + 1, r_bp);
+}
+
+/**
+ *	pdc_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup. Our timing registers are shared
+ *	but we want to set the PIO timing by default.
+ */
+ 
+static void pdc_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	pdc_configure_piomode(ap, adev, adev->pio_mode - XFER_PIO_0);
+}
+
+/**
+ *	pdc_configure_dmamode	-	set DMA mode in chip
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Load DMA cycle times into the chip ready for a DMA transfer
+ *	to occur.
+ */
+ 
+static void pdc_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int port = 0x60 + 4 * ap->port_no + 2 * adev->devno;
+	static u8 udma_timing[6][2] = {
+		{ 0x60, 0x03 },	/* 33 Mhz Clock */
+		{ 0x40, 0x02 },
+		{ 0x20, 0x01 },
+		{ 0x40, 0x02 },	/* 66 Mhz Clock */
+		{ 0x20, 0x01 },
+		{ 0x20, 0x01 }		
+	};
+	u8 r_bp, r_cp;
+	
+	pci_read_config_byte(pdev, port + 1, &r_bp);
+	pci_read_config_byte(pdev, port + 2, &r_cp);
+	
+	r_bp &= ~0xF0;
+	r_cp &= ~0x0F;
+	
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		int speed = adev->dma_mode - XFER_UDMA_0;
+		r_bp |= udma_timing[speed][0];
+		r_cp |= udma_timing[speed][1];
+		
+	} else {
+		int speed = adev->dma_mode - XFER_MW_DMA_0;
+		r_bp |= 0x60;
+		r_cp |= (5 - speed);
+	}
+	pci_write_config_byte(pdev, port + 1, r_bp);
+	pci_write_config_byte(pdev, port + 2, r_cp);
+	
+}
+
+/**
+ *	pdc2026x_bmdma_start		-	DMA engine begin
+ *	@qc: ATA command
+ *
+ *	In UDMA3 or higher we have to clock switch for the duration of the
+ *	DMA transfer sequence.
+ */
+ 
+static void pdc2026x_bmdma_start(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct ata_taskfile *tf = &qc->tf;
+	int sel66 = ap->port_no ? 0x08: 0x02;
+
+	unsigned long master = ap->host->ports[0]->ioaddr.bmdma_addr;
+	unsigned long clock = master + 0x11;
+	unsigned long atapi_reg = master + 0x20 + (4 * ap->port_no);
+	
+	u32 len;
+	
+	/* Check we keep host level locking here */
+	if (adev->dma_mode >= XFER_UDMA_2)
+		outb(inb(clock) | sel66, clock);
+	else
+		outb(inb(clock) & ~sel66, clock);
+
+	/* The DMA clocks may have been trashed by a reset. FIXME: make conditional 
+	   and move to qc_issue ? */
+	pdc_set_dmamode(ap, qc->dev);
+
+	/* Cases the state machine will not complete correctly without help */
+	if ((tf->flags & ATA_TFLAG_LBA48) ||  tf->protocol == ATA_PROT_ATAPI_DMA)
+	{
+		if (tf->flags & ATA_TFLAG_LBA48)
+			len = qc->nsect * 512;
+		else
+			len = qc->nbytes;
+		
+		if (tf->flags & ATA_TFLAG_WRITE)
+			len |= 0x06000000;
+		else
+			len |= 0x05000000;
+			
+		outl(len, atapi_reg);
+	}
+	
+	/* Activate DMA */	
+	ata_bmdma_start(qc);
+}
+
+/**
+ *	pdc2026x_bmdma_end		-	DMA engine stop
+ *	@qc: ATA command
+ *
+ *	After a DMA completes we need to put the clock back to 33MHz for
+ *	PIO timings.
+ */
+ 
+static void pdc2026x_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct ata_taskfile *tf = &qc->tf;
+	
+	int sel66 = ap->port_no ? 0x08: 0x02;
+	/* The clock bits are in the same register for both channels */
+	unsigned long master = ap->host->ports[0]->ioaddr.bmdma_addr;
+	unsigned long clock = master + 0x11;
+	unsigned long atapi_reg = master + 0x20 + (4 * ap->port_no);
+	
+	/* Cases the state machine will not complete correctly */
+	if (tf->protocol == ATA_PROT_ATAPI_DMA || ( tf->flags & ATA_TFLAG_LBA48)) {
+		outl(0, atapi_reg);
+		outb(inb(clock) & ~sel66, clock);
+	}
+	/* Check we keep host level locking here */
+	/* Flip back to 33Mhz for PIO */
+	if (adev->dma_mode >= XFER_UDMA_2)
+		outb(inb(clock) & ~sel66, clock);
+
+	ata_bmdma_stop(qc);
+}
+
+/**
+ *	pdc2026x_dev_config	-	device setup hook
+ *	@ap: ATA port
+ *	@adev: newly found device
+ *
+ *	Perform chip specific early setup. We need to lock the transfer
+ *	sizes to 8bit to avoid making the state engine on the 2026x cards
+ *	barf.
+ */
+ 
+static void pdc2026x_dev_config(struct ata_port *ap, struct ata_device *adev)
+{
+	adev->max_sectors = 256;
+}
+
+static struct scsi_host_template pdc_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations pdc2024x_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= pdc_set_piomode,
+	.set_dmamode	= pdc_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= pdc2024x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static struct ata_port_operations pdc2026x_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= pdc_set_piomode,
+	.set_dmamode	= pdc_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+	.dev_config	= pdc2026x_dev_config,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= pdc2026x_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= pdc2026x_bmdma_start,
+	.bmdma_stop	= pdc2026x_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static int pdc_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info[3] = {
+		{
+			.sht = &pdc_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = ATA_UDMA2,
+			.port_ops = &pdc2024x_port_ops
+		}, 
+		{
+			.sht = &pdc_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = ATA_UDMA4,
+			.port_ops = &pdc2026x_port_ops
+		},
+		{
+			.sht = &pdc_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = ATA_UDMA5,
+			.port_ops = &pdc2026x_port_ops
+		}
+		
+	};
+	static struct ata_port_info *port_info[2];
+
+	port_info[0] = port_info[1] = &info[id->driver_data];
+	
+	if (dev->device == PCI_DEVICE_ID_PROMISE_20265) {
+		struct pci_dev *bridge = dev->bus->self;
+		/* Don't grab anything behind a Promise I2O RAID */
+		if (bridge && bridge->vendor == PCI_VENDOR_ID_INTEL) {
+			if( bridge->device == PCI_DEVICE_ID_INTEL_I960)
+				return -ENODEV;
+			if( bridge->device == PCI_DEVICE_ID_INTEL_I960RM)
+				return -ENODEV;
+		}
+	}
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static struct pci_device_id pdc[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20246), 0},
+	{ PCI_DEVICE(PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20262), 1},
+	{ PCI_DEVICE(PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20263), 1},
+	{ PCI_DEVICE(PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20265), 2},
+	{ PCI_DEVICE(PCI_VENDOR_ID_PROMISE, PCI_DEVICE_ID_PROMISE_20267), 2},
+	{ 0, },
+};
+
+static struct pci_driver pdc_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= pdc,
+	.probe 		= pdc_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init pdc_init(void)
+{
+	return pci_register_driver(&pdc_pci_driver);
+}
+
+
+static void __exit pdc_exit(void)
+{
+	pci_unregister_driver(&pdc_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Promise 2024x and 20262-20267");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, pdc);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(pdc_init);
+module_exit(pdc_exit);
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
new file mode 100644
index 000000000000..ededb40e084d
--- /dev/null
+++ b/drivers/ata/pata_qdi.c
@@ -0,0 +1,403 @@
+/*
+ *    pata_qdi.c - QDI VLB ATA controllers
+ *	(C) 2006 Red Hat <alan@redhat.com>
+ *
+ * This driver mostly exists as a proof of concept for non PCI devices under
+ * libata. While the QDI6580 was 'neat' in 1993 it is no longer terribly
+ * useful.
+ *
+ * Tuning code written from the documentation at
+ * http://www.ryston.cz/petr/vlb/qd6500.html
+ * http://www.ryston.cz/petr/vlb/qd6580.html
+ *
+ * Probe code based on drivers/ide/legacy/qd65xx.c
+ * Rewritten from the work of Colten Edwards <pje120@cs.usask.ca> by
+ * Samuel Thibault <samuel.thibault@fnac.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/platform_device.h>
+
+#define DRV_NAME "pata_qdi"
+#define DRV_VERSION "0.2.4"
+
+#define NR_HOST 4	/* Two 6580s */
+
+struct qdi_data {
+	unsigned long timing;
+	u8 clock[2];
+	u8 last;
+	int fast;
+	struct platform_device *platform_dev;
+	
+};
+
+static struct ata_host *qdi_host[NR_HOST];
+static struct qdi_data qdi_data[NR_HOST];
+static int nr_qdi_host;
+
+#ifdef MODULE
+static int probe_qdi = 1;
+#else
+static int probe_qdi;
+#endif
+
+static void qdi6500_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct ata_timing t;
+	struct qdi_data *qdi = ap->host->private_data;
+	int active, recovery;
+	u8 timing;
+
+	/* Get the timing data in cycles */
+	ata_timing_compute(adev, adev->pio_mode, &t, 30303, 1000);
+	
+	if (qdi->fast) {
+		active = 8 - FIT(t.active, 1, 8);
+		recovery = 18 - FIT(t.recover, 3, 18);
+	} else {
+		active = 9 - FIT(t.active, 2, 9);
+		recovery = 15 - FIT(t.recover, 0, 15);
+	}
+	timing = (recovery << 4) | active | 0x08;
+	
+	qdi->clock[adev->devno] = timing;
+
+	outb(timing, qdi->timing);	
+}
+
+static void qdi6580_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct ata_timing t;
+	struct qdi_data *qdi = ap->host->private_data;
+	int active, recovery;
+	u8 timing;
+
+	/* Get the timing data in cycles */
+	ata_timing_compute(adev, adev->pio_mode, &t, 30303, 1000);
+	
+	if (qdi->fast) {
+		active = 8 - FIT(t.active, 1, 8);
+		recovery = 18 - FIT(t.recover, 3, 18);
+	} else {
+		active = 9 - FIT(t.active, 2, 9);
+		recovery = 15 - FIT(t.recover, 0, 15);
+	}
+	timing = (recovery << 4) | active | 0x08;
+	
+	qdi->clock[adev->devno] = timing;
+
+	outb(timing, qdi->timing);
+	
+	/* Clear the FIFO */
+	if (adev->class != ATA_DEV_ATA)
+		outb(0x5F, (qdi->timing & 0xFFF0) + 3);
+}
+
+/**
+ *	qdi_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings.
+ */
+
+static unsigned int qdi_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct qdi_data *qdi = ap->host->private_data;
+
+	if (qdi->clock[adev->devno] != qdi->last) {
+		if (adev->pio_mode) {
+			qdi->last = qdi->clock[adev->devno];
+			outb(qdi->clock[adev->devno], qdi->timing);
+		}
+	}
+	return ata_qc_issue_prot(qc);
+}
+
+static void qdi_data_xfer(struct ata_device *adev, unsigned char *buf, unsigned int buflen, int write_data)
+{
+	struct ata_port *ap = adev->ap;
+	int slop = buflen & 3;
+	
+	if (ata_id_has_dword_io(adev->id)) {
+		if (write_data)
+			outsl(ap->ioaddr.data_addr, buf, buflen >> 2);
+		else
+			insl(ap->ioaddr.data_addr, buf, buflen >> 2);
+	
+		if (unlikely(slop)) {
+			u32 pad;
+			if (write_data) {
+				memcpy(&pad, buf + buflen - slop, slop);
+				outl(le32_to_cpu(pad), ap->ioaddr.data_addr);
+			} else {
+				pad = cpu_to_le16(inl(ap->ioaddr.data_addr));
+				memcpy(buf + buflen - slop, &pad, slop);
+			}
+		}
+	} else
+		ata_pio_data_xfer(adev, buf, buflen, write_data);
+}
+
+static struct scsi_host_template qdi_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations qdi6500_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= qdi6500_set_piomode,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ata_bmdma_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= qdi_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= qdi_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static struct ata_port_operations qdi6580_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= qdi6580_set_piomode,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= ata_bmdma_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+	
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= qdi_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= qdi_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/**
+ *	qdi_init_one		-	attach a qdi interface
+ *	@type: Type to display
+ *	@io: I/O port start
+ *	@irq: interrupt line
+ *	@fast: True if on a > 33Mhz VLB
+ *
+ *	Register an ISA bus IDE interface. Such interfaces are PIO and we
+ *	assume do not support IRQ sharing.
+ */
+ 
+static __init int qdi_init_one(unsigned long port, int type, unsigned long io, int irq, int fast)
+{
+	struct ata_probe_ent ae;
+	struct platform_device *pdev;
+	int ret;
+
+	unsigned long ctrl = io + 0x206;
+
+	/*
+	 *	Fill in a probe structure first of all
+	 */
+
+	pdev = platform_device_register_simple(DRV_NAME, nr_qdi_host, NULL, 0);
+	if (pdev == NULL)
+		return -ENOMEM;
+	
+	memset(&ae, 0, sizeof(struct ata_probe_ent));
+	INIT_LIST_HEAD(&ae.node);
+	ae.dev = &pdev->dev;
+	
+	if (type == 6580) {
+		ae.port_ops = &qdi6580_port_ops;
+		ae.pio_mask = 0x1F;
+	} else {
+		ae.port_ops = &qdi6500_port_ops;
+		ae.pio_mask = 0x07;	/* Actually PIO3 !IORDY is possible */
+	}
+	
+	ae.sht = &qdi_sht;
+	ae.n_ports = 1;
+	ae.irq = irq;
+	ae.irq_flags = 0;
+	ae.port_flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST;
+	ae.port[0].cmd_addr = io;
+	ae.port[0].altstatus_addr = ctrl;
+	ae.port[0].ctl_addr =	ctrl;
+	ata_std_ports(&ae.port[0]);
+
+	/*
+	 *	Hook in a private data structure per channel
+	 */
+	ae.private_data = &qdi_data[nr_qdi_host];
+
+	qdi_data[nr_qdi_host].timing = port;
+	qdi_data[nr_qdi_host].fast = fast;
+	qdi_data[nr_qdi_host].platform_dev = pdev;
+
+	printk(KERN_INFO DRV_NAME": qd%d at 0x%lx.\n", type, io);
+	ret = ata_device_add(&ae);
+	if (ret == 0) {
+		platform_device_unregister(pdev);
+		return -ENODEV;
+	}
+		
+	qdi_host[nr_qdi_host++] = dev_get_drvdata(&pdev->dev);
+	return 0;	
+}
+
+/**
+ *	qdi_init		-	attach qdi interfaces
+ *
+ *	Attach qdi IDE interfaces by scanning the ports it may occupy.
+ */
+ 
+static __init int qdi_init(void)
+{
+	unsigned long flags;
+	static const unsigned long qd_port[2] = { 0x30, 0xB0 };
+	static const unsigned long ide_port[2] = { 0x170, 0x1F0 };
+	static const int ide_irq[2] = { 14, 15 };
+	
+	int ct = 0;
+	int i;
+	
+	if (probe_qdi == 0)
+		return -ENODEV;
+	
+	/*
+ 	 *	Check each possible QD65xx base address
+	 */
+	
+	for (i = 0; i < 2; i++) {
+		unsigned long port = qd_port[i];
+		u8 r, res;
+		
+		
+		if (request_region(port, 2, "pata_qdi")) {
+			/* Check for a card */
+			local_irq_save(flags);
+			r = inb_p(port);
+			outb_p(0x19, port);
+			res = inb_p(port);
+			outb_p(r, port);
+			local_irq_restore(flags);
+		
+			/* Fail */
+			if (res == 0x19)
+			{
+				release_region(port, 2);
+				continue;
+			}
+			
+			/* Passes the presence test */
+			r = inb_p(port + 1);	/* Check port agrees with port set */
+			if ((r & 2) >> 1 != i) {
+				release_region(port, 2);
+				continue;
+			}
+
+			/* Check card type */			
+			if ((r & 0xF0) == 0xC0) {
+				/* QD6500: single channel */
+				if (r & 8) {
+					/* Disabled ? */
+					release_region(port, 2);
+					continue;
+				}
+				ct += qdi_init_one(port, 6500, ide_port[r & 0x01], ide_irq[r & 0x01], r & 0x04);
+			}
+			if (((r & 0xF0) == 0xA0) || (r & 0xF0) == 0x50) {
+				/* QD6580: dual channel */
+				if (!request_region(port + 2 , 2, "pata_qdi"))
+				{
+					release_region(port, 2);
+					continue;
+				}
+				res = inb(port + 3);
+				if (res & 1) {
+					/* Single channel mode */
+					ct += qdi_init_one(port, 6580, ide_port[r & 0x01], ide_irq[r & 0x01], r & 0x04);
+				} else {
+					/* Dual channel mode */
+					ct += qdi_init_one(port, 6580, 0x1F0, 14, r & 0x04);
+					ct += qdi_init_one(port + 2, 6580, 0x170, 15, r & 0x04);
+				}
+			}
+		}
+	}
+	if (ct != 0)
+		return 0;
+	return -ENODEV;
+}
+
+static __exit void qdi_exit(void)
+{
+	int i;
+
+	for (i = 0; i < nr_qdi_host; i++) {
+		ata_host_remove(qdi_host[i]);
+		/* Free the control resource. The 6580 dual channel has the resources
+		 * claimed as a pair of 2 byte resources so we need no special cases...
+		 */
+		release_region(qdi_data[i].timing, 2);
+		platform_device_unregister(qdi_data[i].platform_dev);
+	}	
+}
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for qdi ATA");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+
+module_init(qdi_init);
+module_exit(qdi_exit);
+
+module_param(probe_qdi, int, 0);
+
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
new file mode 100644
index 000000000000..6f7d0527265f
--- /dev/null
+++ b/drivers/ata/pata_radisys.c
@@ -0,0 +1,335 @@
+/*
+ *    pata_radisys.c - Intel PATA/SATA controllers
+ *
+ *	(C) 2006 Red Hat <alan@redhat.com>
+ *
+ *    Some parts based on ata_piix.c by Jeff Garzik and others.
+ *
+ *    A PIIX relative, this device has a single ATA channel and no
+ *    slave timings, SITRE or PPE. In that sense it is a close relative
+ *    of the original PIIX. It does however support UDMA 33/66 per channel
+ *    although no other modes/timings. Also lacking is 32bit I/O on the ATA
+ *    port.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_radisys"
+#define DRV_VERSION	"0.4.1"
+
+/**
+ *	radisys_probe_init		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+ 
+static int radisys_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA80;
+	return ata_std_prereset(ap);
+}
+
+
+/**
+ *	radisys_pata_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *	@classes:
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void radisys_pata_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, radisys_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	radisys_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: um
+ *
+ *	Set PIO mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void radisys_set_piomode (struct ata_port *ap, struct ata_device *adev)
+{
+	unsigned int pio	= adev->pio_mode - XFER_PIO_0;
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
+	u16 idetm_data;
+	int control = 0;
+
+	/*
+	 *	See Intel Document 298600-004 for the timing programing rules
+	 *	for PIIX/ICH. Note that the early PIIX does not have the slave
+	 *	timing port at 0x44. The Radisys is a relative of the PIIX
+	 *	but not the same so be careful.
+	 */
+
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },	/* Check me */
+			    { 0, 0 },
+			    { 1, 1 },
+			    { 2, 2 },
+			    { 3, 3 }, };
+
+	if (pio > 0)
+		control |= 1;	/* TIME1 enable */
+	if (ata_pio_need_iordy(adev))
+		control |= 2;	/* IE IORDY */
+
+	pci_read_config_word(dev, 0x40, &idetm_data);
+
+	/* Enable IE and TIME as appropriate. Clear the other
+	   drive timing bits */
+	idetm_data &= 0xCCCC;
+	idetm_data |= (control << (4 * adev->devno));
+	idetm_data |= (timings[pio][0] << 12) |
+			(timings[pio][1] << 8);
+	pci_write_config_word(dev, 0x40, idetm_data);
+
+	/* Track which port is configured */
+	ap->private_data = adev;
+}
+
+/**
+ *	radisys_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *	@isich: True if the device is an ICH and has IOCFG registers
+ *
+ *	Set MWDMA mode for device, in host controller PCI config space.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void radisys_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *dev	= to_pci_dev(ap->host->dev);
+	u16 idetm_data;
+	u8 udma_enable;
+	
+	static const	 /* ISP  RTC */
+	u8 timings[][2]	= { { 0, 0 },
+			    { 0, 0 },
+			    { 1, 1 },
+			    { 2, 2 },
+			    { 3, 3 }, };
+
+	/*
+	 * MWDMA is driven by the PIO timings. We must also enable
+	 * IORDY unconditionally.
+	 */
+
+	pci_read_config_word(dev, 0x40, &idetm_data);
+	pci_read_config_byte(dev, 0x48, &udma_enable);
+
+	if (adev->dma_mode < XFER_UDMA_0) {
+		unsigned int mwdma	= adev->dma_mode - XFER_MW_DMA_0;
+		const unsigned int needed_pio[3] = {
+			XFER_PIO_0, XFER_PIO_3, XFER_PIO_4
+		};
+		int pio = needed_pio[mwdma] - XFER_PIO_0;
+		int control = 3;	/* IORDY|TIME0 */
+
+		/* If the drive MWDMA is faster than it can do PIO then
+		   we must force PIO0 for PIO cycles. */
+
+		if (adev->pio_mode < needed_pio[mwdma])
+			control = 1;
+
+		/* Mask out the relevant control and timing bits we will load. Also
+		   clear the other drive TIME register as a precaution */
+		   
+		idetm_data &= 0xCCCC;
+		idetm_data |= control << (4 * adev->devno);
+		idetm_data |= (timings[pio][0] << 12) | (timings[pio][1] << 8);
+
+		udma_enable &= ~(1 << adev->devno);
+	} else {
+		u8 udma_mode;
+		
+		/* UDMA66 on: UDMA 33 and 66 are switchable via register 0x4A */
+		
+		pci_read_config_byte(dev, 0x4A, &udma_mode);
+		
+		if (adev->xfer_mode == XFER_UDMA_2)
+			udma_mode &= ~ (1 << adev->devno);
+		else /* UDMA 4 */
+			udma_mode |= (1 << adev->devno);
+			
+		pci_write_config_byte(dev, 0x4A, udma_mode);
+		
+		udma_enable |= (1 << adev->devno);
+	}
+	pci_write_config_word(dev, 0x40, idetm_data);
+	pci_write_config_byte(dev, 0x48, udma_enable);
+
+	/* Track which port is configured */
+	ap->private_data = adev;
+}
+
+/**
+ *	radisys_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings if
+ *	neccessary. Our logic also clears TIME0/TIME1 for the other device so
+ *	that, even if we get this wrong, cycles to the other device will
+ *	be made PIO0.
+ */
+
+static unsigned int radisys_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+
+	if (adev != ap->private_data) {
+		/* UDMA timing is not shared */
+		if (adev->dma_mode < XFER_UDMA_0) {
+			if (adev->dma_mode)
+				radisys_set_dmamode(ap, adev);
+			else if (adev->pio_mode)
+				radisys_set_piomode(ap, adev);
+		}
+	}
+	return ata_qc_issue_prot(qc);
+}
+
+
+static struct scsi_host_template radisys_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations radisys_pata_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= radisys_set_piomode,
+	.set_dmamode		= radisys_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= radisys_pata_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= radisys_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+
+/**
+ *	radisys_init_one - Register PIIX ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in radisys_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.  We probe for combined mode (sigh),
+ *	and then hand over control to libata, for it to do the rest.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int radisys_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	static struct ata_port_info info = {
+		.sht		= &radisys_sht,
+		.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask	= 0x1f,	/* pio0-4 */
+		.mwdma_mask	= 0x07, /* mwdma1-2 */
+		.udma_mask	= 0x14, /* UDMA33/66 only */
+		.port_ops	= &radisys_pata_ops,
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev,
+			   "version " DRV_VERSION "\n");
+
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id radisys_pci_tbl[] = {
+	{ 0x1331, 0x8201, PCI_ANY_ID, PCI_ANY_ID, },
+	{ }	/* terminate list */
+};
+
+static struct pci_driver radisys_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= radisys_pci_tbl,
+	.probe			= radisys_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init radisys_init(void)
+{
+	return pci_register_driver(&radisys_pci_driver);
+}
+
+static void __exit radisys_exit(void)
+{
+	pci_unregister_driver(&radisys_pci_driver);
+}
+
+
+module_init(radisys_init);
+module_exit(radisys_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for Radisys R82600 controllers");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, radisys_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
new file mode 100644
index 000000000000..3c6d84fd4312
--- /dev/null
+++ b/drivers/ata/pata_rz1000.c
@@ -0,0 +1,205 @@
+/*
+ *  RZ1000/1001 driver based upon
+ *
+ *  linux/drivers/ide/pci/rz1000.c	Version 0.06	January 12, 2003
+ *  Copyright (C) 1995-1998  Linus Torvalds & author (see below)
+ *  Principal Author:  mlord@pobox.com (Mark Lord)
+ *
+ *  See linux/MAINTAINERS for address of current maintainer.
+ *
+ *  This file provides support for disabling the buggy read-ahead
+ *  mode of the RZ1000 IDE chipset, commonly used on Intel motherboards.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"pata_rz1000"
+#define DRV_VERSION	"0.2.2"
+
+
+/**
+ *	rz1000_prereset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generics
+ */
+
+static int rz1000_prereset(struct ata_port *ap)
+{
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	rz1000_error_handler		-	probe reset
+ *	@ap: ATA port
+ *
+ *	Perform the ATA standard reset sequence
+ */
+
+static void rz1000_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, rz1000_prereset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	rz1000_set_mode		-	mode setting function
+ *	@ap: ATA interface
+ *
+ *	Use a non standard set_mode function. We don't want to be tuned. We
+ *	would prefer to be BIOS generic but for the fact our hardware is
+ *	whacked out.
+ */
+
+static void rz1000_set_mode(struct ata_port *ap)
+{
+	int i;
+
+	for (i = 0; i < ATA_MAX_DEVICES; i++) {
+		struct ata_device *dev = &ap->device[i];
+		if (ata_dev_enabled(dev)) {
+			/* We don't really care */
+			dev->pio_mode = XFER_PIO_0;
+			dev->xfer_mode = XFER_PIO_0;
+			dev->xfer_shift = ATA_SHIFT_PIO;
+			dev->flags |= ATA_DFLAG_PIO;
+		}
+	}
+}
+
+
+static struct scsi_host_template rz1000_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations rz1000_port_ops = {
+	.set_mode	= rz1000_set_mode,
+
+	.port_disable	= ata_port_disable,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= rz1000_error_handler,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= rz1000_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	rz1000_init_one - Register RZ1000 ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in rz1000_pci_tbl matching with @pdev
+ *
+ *	Configure an RZ1000 interface. This doesn't require much special
+ *	handling except that we *MUST* kill the chipset readahead or the
+ *	user may experience data corruption.
+ */
+
+static int rz1000_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	struct ata_port_info *port_info[2];
+	u16 reg;
+	static struct ata_port_info info = {
+		.sht = &rz1000_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.port_ops = &rz1000_port_ops
+	};
+
+	if (!printed_version++)
+		printk(KERN_DEBUG DRV_NAME " version " DRV_VERSION "\n");
+
+	/* Be exceptionally paranoid as we must be sure to apply the fix */
+	if (pci_read_config_word(pdev, 0x40, &reg) != 0)
+		goto fail;
+	reg &= 0xDFFF;
+	if (pci_write_config_word(pdev, 0x40, reg) != 0)
+		goto fail;
+	printk(KERN_INFO DRV_NAME ": disabled chipset readahead.\n");
+
+	port_info[0] = &info;
+	port_info[1] = &info;
+	return ata_pci_init_one(pdev, port_info, 2);
+fail:
+	printk(KERN_ERR DRV_NAME ": failed to disable read-ahead on chipset..\n");
+	/* Not safe to use so skip */
+	return -ENODEV;
+}
+
+static struct pci_device_id pata_rz1000[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_PCTECH, PCI_DEVICE_ID_PCTECH_RZ1000), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_PCTECH, PCI_DEVICE_ID_PCTECH_RZ1001), },
+	{ 0, },
+};
+
+static struct pci_driver rz1000_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= pata_rz1000,
+	.probe 		= rz1000_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+
+static int __init rz1000_init(void)
+{
+	return pci_register_driver(&rz1000_pci_driver);
+}
+
+static void __exit rz1000_exit(void)
+{
+	pci_unregister_driver(&rz1000_pci_driver);
+}
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for RZ1000 PCI ATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, pata_rz1000);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(rz1000_init);
+module_exit(rz1000_exit);
+
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
new file mode 100644
index 000000000000..9b42b9a52423
--- /dev/null
+++ b/drivers/ata/pata_sc1200.c
@@ -0,0 +1,287 @@
+/*
+ * New ATA layer SC1200 driver		Alan Cox <alan@redhat.com>
+ *
+ * TODO: Mode selection filtering
+ * TODO: Can't enable second channel until ATA core has serialize
+ * TODO: Needs custom DMA cleanup code
+ *
+ * Based very heavily on
+ *
+ * linux/drivers/ide/pci/sc1200.c		Version 0.91	28-Jan-2003
+ *
+ * Copyright (C) 2000-2002		Mark Lord <mlord@pobox.com>
+ * May be copied or modified under the terms of the GNU General Public License
+ *
+ * Development of this chipset driver was funded
+ * by the nice folks at National Semiconductor.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ * 
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME	"sc1200"
+#define DRV_VERSION	"0.2.3"
+
+#define SC1200_REV_A	0x00
+#define SC1200_REV_B1	0x01
+#define SC1200_REV_B3	0x02
+#define SC1200_REV_C1	0x03
+#define SC1200_REV_D1	0x04
+
+/**
+ *	sc1200_clock	-	PCI clock
+ *
+ *	Return the PCI bus clocking for the SC1200 chipset configuration
+ *	in use. We return 0 for 33MHz 1 for 48MHz and 2 for 66Mhz
+ */
+ 
+static int sc1200_clock(void)
+{
+	/* Magic registers that give us the chipset data */
+	u8 chip_id = inb(0x903C);
+	u8 silicon_rev = inb(0x903D);
+	u16 pci_clock;
+	
+	if (chip_id == 0x04 && silicon_rev < SC1200_REV_B1)
+		return 0;	/* 33 MHz mode */
+
+	/* Clock generator configuration 0x901E its 8/9 are the PCI clocking
+	   0/3 is 33Mhz 1 is 48 2 is 66 */
+
+	pci_clock = inw(0x901E);
+	pci_clock >>= 8;
+	pci_clock &= 0x03;
+	if (pci_clock == 3)
+		pci_clock = 0;
+	return pci_clock;
+}
+
+/**
+ *	sc1200_set_piomode		-	PIO setup
+ *	@ap: ATA interface
+ *	@adev: device on the interface
+ *
+ *	Set our PIO requirements. This is fairly simple on the SC1200
+ */
+ 
+static void sc1200_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u32 pio_timings[4][5] = {
+		{0x00009172, 0x00012171, 0x00020080, 0x00032010, 0x00040010},	// format0  33Mhz
+		{0xd1329172, 0x71212171, 0x30200080, 0x20102010, 0x00100010},	// format1, 33Mhz
+		{0xfaa3f4f3, 0xc23232b2, 0x513101c1, 0x31213121, 0x10211021},	// format1, 48Mhz
+		{0xfff4fff4, 0xf35353d3, 0x814102f1, 0x42314231, 0x11311131}	// format1, 66Mhz
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 format;
+	unsigned int reg = 0x40 + 0x10 * ap->port_no;
+	int mode = adev->pio_mode - XFER_PIO_0;
+	
+	pci_read_config_dword(pdev, reg + 4, &format);
+	format >>= 31;
+	format += sc1200_clock();
+	pci_write_config_dword(pdev, reg + 8 * adev->devno, 
+				pio_timings[format][mode]);
+}
+
+/**
+ *	sc1200_set_dmamode		-	DMA timing setup
+ *	@ap: ATA interface
+ *	@adev: Device being configured
+ *
+ *	We cannot mix MWDMA and UDMA without reloading timings each switch
+ *	master to slave.
+ */
+ 
+static void sc1200_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u32 udma_timing[3][3] = {
+		{ 0x00921250, 0x00911140, 0x00911030 },
+		{ 0x00932470, 0x00922260, 0x00922140 },
+		{ 0x009436A1, 0x00933481, 0x00923261 }
+	};
+	
+	static const u32 mwdma_timing[3][3] = {
+		{ 0x00077771, 0x00012121, 0x00002020 },
+		{ 0x000BBBB2, 0x00024241, 0x00013131 },
+		{ 0x000FFFF3, 0x00035352, 0x00015151 }
+	};
+	
+	int clock = sc1200_clock();
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	unsigned int reg = 0x40 + 0x10 * ap->port_no;
+	int mode = adev->dma_mode;
+	u32 format;
+
+	if (mode >= XFER_UDMA_0)
+		format = udma_timing[clock][mode - XFER_UDMA_0];
+	else
+		format = mwdma_timing[clock][mode - XFER_MW_DMA_0];
+	
+	if (adev->devno == 0) {
+		u32 timings;
+		
+		pci_read_config_dword(pdev, reg + 4, &timings);
+		timings &= 0x80000000UL;
+		timings |= format;
+		pci_write_config_dword(pdev, reg + 4, timings);
+	} else
+		pci_write_config_dword(pdev, reg + 12, format);
+}
+
+/**
+ *	sc1200_qc_issue_prot	-	command issue
+ *	@qc: command pending
+ *
+ *	Called when the libata layer is about to issue a command. We wrap
+ *	this interface so that we can load the correct ATA timings if
+ *	neccessary.  Specifically we have a problem that there is only
+ *	one MWDMA/UDMA bit.
+ */
+
+static unsigned int sc1200_qc_issue_prot(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+	struct ata_device *adev = qc->dev;
+	struct ata_device *prev = ap->private_data;
+
+	/* See if the DMA settings could be wrong */
+	if (adev->dma_mode != 0 && adev != prev && prev != NULL) {
+		/* Maybe, but do the channels match MWDMA/UDMA ? */
+		if ((adev->dma_mode >= XFER_UDMA_0 && prev->dma_mode < XFER_UDMA_0) ||
+		    (adev->dma_mode < XFER_UDMA_0 && prev->dma_mode >= XFER_UDMA_0))
+		    	/* Switch the mode bits */
+		    	sc1200_set_dmamode(ap, adev);
+	}
+
+	return ata_qc_issue_prot(qc);
+}
+
+static struct scsi_host_template sc1200_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations sc1200_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= sc1200_set_piomode,
+	.set_dmamode	= sc1200_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= ata_bmdma_error_handler,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= sc1200_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/**
+ *	sc1200_init_one		-	Initialise an SC1200
+ *	@dev: PCI device
+ *	@id: Entry in match table
+ *
+ *	Just throw the needed data at the libata helper and it does all
+ *	our work.
+ */
+ 
+static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &sc1200_sht,
+		.flags = ATA_FLAG_SLAVE_POSS|ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x07,
+		.port_ops = &sc1200_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+	
+	/* Can't enable port 2 yet, see top comments */
+	return ata_pci_init_one(dev, port_info, 1);
+}
+
+static struct pci_device_id sc1200[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_SCx200_IDE), },
+	{ 0, },
+};
+
+static struct pci_driver sc1200_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= sc1200,
+	.probe 		= sc1200_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init sc1200_init(void)
+{
+	return pci_register_driver(&sc1200_pci_driver);
+}
+
+
+static void __exit sc1200_exit(void)
+{
+	pci_unregister_driver(&sc1200_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox, Mark Lord");
+MODULE_DESCRIPTION("low-level driver for the NS/AMD SC1200");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, sc1200);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(sc1200_init);
+module_exit(sc1200_exit);
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
new file mode 100644
index 000000000000..3ec30036c978
--- /dev/null
+++ b/drivers/ata/pata_serverworks.c
@@ -0,0 +1,587 @@
+/*
+ * ata-serverworks.c 	- Serverworks PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based upon
+ *
+ * serverworks.c
+ * 
+ * Copyright (C) 1998-2000 Michel Aubry
+ * Copyright (C) 1998-2000 Andrzej Krzysztofowicz
+ * Copyright (C) 1998-2000 Andre Hedrick <andre@linux-ide.org>
+ * Portions copyright (c) 2001 Sun Microsystems
+ *
+ *
+ * RCC/ServerWorks IDE driver for Linux
+ *
+ *   OSB4: `Open South Bridge' IDE Interface (fn 1)
+ *         supports UDMA mode 2 (33 MB/s)
+ *
+ *   CSB5: `Champion South Bridge' IDE Interface (fn 1)
+ *         all revisions support UDMA mode 4 (66 MB/s)
+ *         revision A2.0 and up support UDMA mode 5 (100 MB/s)
+ *
+ *         *** The CSB5 does not provide ANY register ***
+ *         *** to detect 80-conductor cable presence. ***
+ *
+ *   CSB6: `Champion South Bridge' IDE Interface (optional: third channel)
+ *
+ * Documentation:
+ *	Available under NDA only. Errata info very hard to get.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_serverworks"
+#define DRV_VERSION "0.3.6"
+
+#define SVWKS_CSB5_REVISION_NEW	0x92 /* min PCI_REVISION_ID for UDMA5 (A2.0) */
+#define SVWKS_CSB6_REVISION	0xa0 /* min PCI_REVISION_ID for UDMA4 (A1.0) */
+
+/* Seagate Barracuda ATA IV Family drives in UDMA mode 5
+ * can overrun their FIFOs when used with the CSB5 */
+
+static const char *csb_bad_ata100[] = {
+	"ST320011A",
+	"ST340016A",
+	"ST360021A",
+	"ST380021A",
+	NULL
+};
+
+/**
+ *	dell_cable	-	Dell serverworks cable detection
+ *	@ap: ATA port to do cable detect
+ *
+ *	Dell hide the 40/80 pin select for their interfaces in the top two
+ *	bits of the subsystem ID. 
+ */
+ 
+static int dell_cable(struct ata_port *ap) {
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	if (pdev->subsystem_device & (1 << (ap->port_no + 14)))
+		return ATA_CBL_PATA80;
+	return ATA_CBL_PATA40;
+}
+
+/**
+ *	sun_cable	-	Sun Cobalt 'Alpine' cable detection
+ *	@ap: ATA port to do cable select
+ *
+ *	Cobalt CSB5 IDE hides the 40/80pin in the top two bits of the
+ *	subsystem ID the same as dell. We could use one function but we may
+ *	need to extend the Dell one in future
+ */
+ 
+static int sun_cable(struct ata_port *ap) {
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	if (pdev->subsystem_device & (1 << (ap->port_no + 14)))
+		return ATA_CBL_PATA80;
+	return ATA_CBL_PATA40;
+}
+
+/**
+ *	osb4_cable	-	OSB4 cable detect
+ *	@ap: ATA port to check
+ *
+ *	The OSB4 isn't UDMA66 capable so this is easy
+ */
+
+static int osb4_cable(struct ata_port *ap) {
+	return ATA_CBL_PATA40;
+}
+
+/**
+ *	csb4_cable	-	CSB5/6 cable detect
+ *	@ap: ATA port to check
+ *
+ *	Serverworks default arrangement is to use the drive side detection
+ *	only.
+ */
+
+static int csb_cable(struct ata_port *ap) {
+	return ATA_CBL_PATA80;
+}
+
+struct sv_cable_table {
+	int device;
+	int subvendor;
+	int (*cable_detect)(struct ata_port *ap);
+};
+
+/*
+ *	Note that we don't copy the old serverworks code because the old
+ *	code contains obvious mistakes
+ */
+ 
+static struct sv_cable_table cable_detect[] = {
+	{ PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, PCI_VENDOR_ID_DELL, dell_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_CSB6IDE, PCI_VENDOR_ID_DELL, dell_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, PCI_VENDOR_ID_SUN,  sun_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_OSB4, PCI_ANY_ID, osb4_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, PCI_ANY_ID, csb_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_CSB6IDE, PCI_ANY_ID, csb_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2, PCI_ANY_ID, csb_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_HT1000IDE, PCI_ANY_ID, csb_cable },
+	{ }
+};
+
+/**
+ *	serverworks_pre_reset		-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection according to the device and subvendor 
+ *	identifications
+ */
+ 
+static int serverworks_pre_reset(struct ata_port *ap) {
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct sv_cable_table *cb = cable_detect;
+
+	while(cb->device) {
+		if (cb->device == pdev->device && 
+		    (cb->subvendor == pdev->subsystem_vendor ||
+		      cb->subvendor == PCI_ANY_ID)) {
+			ap->cbl = cb->cable_detect(ap);
+			return ata_std_prereset(ap);
+		}
+		cb++;
+	}
+
+	BUG();
+	return -1;	/* kill compiler warning */
+}
+
+static void serverworks_error_handler(struct ata_port *ap)
+{
+	return ata_bmdma_drive_eh(ap, serverworks_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	serverworks_is_csb	-	Check for CSB or OSB
+ *	@pdev: PCI device to check
+ *
+ *	Returns true if the device being checked is known to be a CSB
+ *	series device.
+ */
+ 
+static u8 serverworks_is_csb(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+		case PCI_DEVICE_ID_SERVERWORKS_CSB5IDE:
+		case PCI_DEVICE_ID_SERVERWORKS_CSB6IDE:
+		case PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2:
+		case PCI_DEVICE_ID_SERVERWORKS_HT1000IDE:
+			return 1;
+		default:
+			break;
+	}
+	return 0;
+}
+
+/**
+ *	serverworks_osb4_filter	-	mode selection filter
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Filter the offered modes for the device to apply controller
+ *	specific rules. OSB4 requires no UDMA for disks due to a FIFO
+ *	bug we hit.
+ */
+ 
+static unsigned long serverworks_osb4_filter(const struct ata_port *ap, struct ata_device *adev, unsigned long mask)
+{
+	if (adev->class == ATA_DEV_ATA)
+		mask &= ~ATA_MASK_UDMA;
+	return ata_pci_default_filter(ap, adev, mask);
+}
+
+
+/**
+ *	serverworks_csb_filter	-	mode selection filter
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Check the blacklist and disable UDMA5 if matched
+ */
+
+static unsigned long serverworks_csb_filter(const struct ata_port *ap, struct ata_device *adev, unsigned long mask)
+{
+	const char *p;
+	char model_num[40];
+	int len, i;
+
+	/* Disk, UDMA */	
+	if (adev->class != ATA_DEV_ATA)
+		return ata_pci_default_filter(ap, adev, mask);
+
+	/* Actually do need to check */
+	ata_id_string(adev->id, model_num, ATA_ID_PROD_OFS, sizeof(model_num));
+	/* Precuationary - why not do this in the libata core ?? */
+	
+	len = strlen(model_num);
+	while ((len > 0) && (model_num[len - 1] == ' ')) {
+		len--;
+		model_num[len] = 0;
+	}
+
+	for(i = 0; (p = csb_bad_ata100[i]) != NULL; i++) {
+		if (!strncmp(p, model_num, len))
+			mask &= ~(0x1F << ATA_SHIFT_UDMA);
+	}
+	return ata_pci_default_filter(ap, adev, mask);
+}
+
+
+/**
+ *	serverworks_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the OSB4/CSB5 timing registers for PIO. The PIO register
+ *	load is done as a simple lookup.
+ */
+static void serverworks_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u8 pio_mode[] = { 0x5d, 0x47, 0x34, 0x22, 0x20 };
+	int offset = 1 + (2 * ap->port_no) - adev->devno;
+	int devbits = (2 * ap->port_no + adev->devno) * 4;
+	u16 csb5_pio;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int pio = adev->pio_mode - XFER_PIO_0;
+
+	pci_write_config_byte(pdev, 0x40 + offset, pio_mode[pio]);
+	
+	/* The OSB4 just requires the timing but the CSB series want the
+	   mode number as well */
+	if (serverworks_is_csb(pdev)) {
+		pci_read_config_word(pdev, 0x4A, &csb5_pio);
+		csb5_pio &= ~(0x0F << devbits);
+		pci_write_config_byte(pdev, 0x4A, csb5_pio | (pio << devbits));
+	}
+}
+
+/**
+ *	serverworks_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the MWDMA/UDMA modes for the serverworks OSB4/CSB5
+ *	chipset. The MWDMA mode values are pulled from a lookup table
+ *	while the chipset uses mode number for UDMA.
+ */
+
+static void serverworks_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static const u8 dma_mode[] = { 0x77, 0x21, 0x20 };
+	int offset = 1 + 2 * ap->port_no - adev->devno;
+	int devbits = (2 * ap->port_no + adev->devno);
+	u8 ultra;
+	u8 ultra_cfg;
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+
+	pci_read_config_byte(pdev, 0x54, &ultra_cfg);
+
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		pci_write_config_byte(pdev, 0x44 + offset,  0x20);
+
+		pci_read_config_byte(pdev, 0x56 + ap->port_no, &ultra);
+		ultra &= ~(0x0F << (ap->port_no * 4));
+		ultra |= (adev->dma_mode - XFER_UDMA_0)
+					<< (ap->port_no * 4);
+		pci_write_config_byte(pdev, 0x56 + ap->port_no, ultra);
+
+		ultra_cfg |=  (1 << devbits);
+	} else {
+		pci_write_config_byte(pdev, 0x44 + offset, 
+			dma_mode[adev->dma_mode - XFER_MW_DMA_0]);
+		ultra_cfg &= ~(1 << devbits);
+	}
+	pci_write_config_byte(pdev, 0x54, ultra_cfg);
+}
+
+static struct scsi_host_template serverworks_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations serverworks_osb4_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= serverworks_set_piomode,
+	.set_dmamode	= serverworks_set_dmamode,
+	.mode_filter	= serverworks_osb4_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= serverworks_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+	
+	.irq_handler	= ata_interrupt,
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static struct ata_port_operations serverworks_csb_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= serverworks_set_piomode,
+	.set_dmamode	= serverworks_set_dmamode,
+	.mode_filter	= serverworks_csb_filter,
+	
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= serverworks_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+	
+	.irq_handler	= ata_interrupt,
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+static int serverworks_fixup_osb4(struct pci_dev *pdev)
+{
+	u32 reg;
+	struct pci_dev *isa_dev = pci_get_device(PCI_VENDOR_ID_SERVERWORKS,
+		  PCI_DEVICE_ID_SERVERWORKS_OSB4, NULL);
+	if (isa_dev) {
+		pci_read_config_dword(isa_dev, 0x64, &reg);
+		reg &= ~0x00002000; /* disable 600ns interrupt mask */
+		if (!(reg & 0x00004000))
+			printk(KERN_DEBUG DRV_NAME ": UDMA not BIOS enabled.\n");
+		reg |=  0x00004000; /* enable UDMA/33 support */
+		pci_write_config_dword(isa_dev, 0x64, reg);
+		pci_dev_put(isa_dev);
+		return 0;
+	}
+	printk(KERN_WARNING "ata_serverworks: Unable to find bridge.\n");
+	return -ENODEV;
+}
+
+static int serverworks_fixup_csb(struct pci_dev *pdev)
+{
+	u8 rev;
+	u8 btr;
+	
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &rev);
+
+	/* Third Channel Test */
+	if (!(PCI_FUNC(pdev->devfn) & 1)) {
+		struct pci_dev * findev = NULL;
+		u32 reg4c = 0;
+		findev = pci_get_device(PCI_VENDOR_ID_SERVERWORKS,
+			PCI_DEVICE_ID_SERVERWORKS_CSB5, NULL);
+		if (findev) {
+			pci_read_config_dword(findev, 0x4C, &reg4c);
+			reg4c &= ~0x000007FF;
+			reg4c |=  0x00000040;
+			reg4c |=  0x00000020;
+			pci_write_config_dword(findev, 0x4C, reg4c);
+			pci_dev_put(findev);
+		}
+	} else {
+		struct pci_dev * findev = NULL;
+		u8 reg41 = 0;
+
+		findev = pci_get_device(PCI_VENDOR_ID_SERVERWORKS,
+				PCI_DEVICE_ID_SERVERWORKS_CSB6, NULL);
+		if (findev) {
+			pci_read_config_byte(findev, 0x41, &reg41);
+			reg41 &= ~0x40;
+			pci_write_config_byte(findev, 0x41, reg41);
+			pci_dev_put(findev);
+		}
+	}
+	/* setup the UDMA Control register
+	 *
+	 * 1. clear bit 6 to enable DMA
+	 * 2. enable DMA modes with bits 0-1
+	 * 	00 : legacy
+	 * 	01 : udma2
+	 * 	10 : udma2/udma4
+	 * 	11 : udma2/udma4/udma5
+	 */
+	pci_read_config_byte(pdev, 0x5A, &btr);
+	btr &= ~0x40;
+	if (!(PCI_FUNC(pdev->devfn) & 1))
+		btr |= 0x2;
+	else
+		btr |= (rev >= SVWKS_CSB5_REVISION_NEW) ? 0x3 : 0x2;
+	pci_write_config_byte(pdev, 0x5A, btr);
+	
+	return btr;
+}
+
+static void serverworks_fixup_ht1000(struct pci_dev *pdev)
+{
+	u8 btr;
+	/* Setup HT1000 SouthBridge Controller - Single Channel Only */
+	pci_read_config_byte(pdev, 0x5A, &btr);
+	btr &= ~0x40;
+	btr |= 0x3;
+	pci_write_config_byte(pdev, 0x5A, btr);
+}
+
+
+static int serverworks_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	int ports = 2;
+	static struct ata_port_info info[4] = {
+		{ /* OSB4 */
+			.sht = &serverworks_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x07,
+			.port_ops = &serverworks_osb4_port_ops
+		}, { /* OSB4 no UDMA */
+			.sht = &serverworks_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x00,
+			.port_ops = &serverworks_osb4_port_ops
+		}, { /* CSB5 */
+			.sht = &serverworks_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x1f,
+			.port_ops = &serverworks_csb_port_ops
+		}, { /* CSB5 - later revisions*/
+			.sht = &serverworks_sht,
+			.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+			.pio_mask = 0x1f,
+			.mwdma_mask = 0x07,
+			.udma_mask = 0x3f,
+			.port_ops = &serverworks_csb_port_ops
+		}
+	};
+	static struct ata_port_info *port_info[2];
+	struct ata_port_info *devinfo = &info[id->driver_data];
+	
+	/* Force master latency timer to 64 PCI clocks */
+	pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x40);
+
+	/* OSB4 : South Bridge and IDE */
+	if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_OSB4IDE) {
+		/* Select non UDMA capable OSB4 if we can't do fixups */
+		if ( serverworks_fixup_osb4(pdev) < 0)
+			devinfo = &info[1];
+	}
+	/* setup CSB5/CSB6 : South Bridge and IDE option RAID */
+	else if ((pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB5IDE) ||
+		 (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB6IDE) ||
+		 (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2)) {
+		 
+		 /* If the returned btr is the newer revision then
+		    select the right info block */
+		 if (serverworks_fixup_csb(pdev) == 3)
+		 	devinfo = &info[3];
+		 	
+		/* Is this the 3rd channel CSB6 IDE ? */
+		if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2)
+			ports = 1;
+	}
+	/* setup HT1000E */
+	else if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_HT1000IDE)
+		serverworks_fixup_ht1000(pdev);
+		
+	if (pdev->device == PCI_DEVICE_ID_SERVERWORKS_CSB5IDE)
+		ata_pci_clear_simplex(pdev);
+		
+	port_info[0] = port_info[1] = devinfo;
+	return ata_pci_init_one(pdev, port_info, ports);
+}
+
+static struct pci_device_id serverworks[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_OSB4IDE), 0},
+	{ PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB5IDE), 2},
+	{ PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB6IDE), 2},
+	{ PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2), 2},
+	{ PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_HT1000IDE), 2},
+	{ 0, },
+};
+
+static struct pci_driver serverworks_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= serverworks,
+	.probe 		= serverworks_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init serverworks_init(void)
+{
+	return pci_register_driver(&serverworks_pci_driver);
+}
+
+
+static void __exit serverworks_exit(void)
+{
+	pci_unregister_driver(&serverworks_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Serverworks OSB4/CSB5/CSB6");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, serverworks);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(serverworks_init);
+module_exit(serverworks_exit);
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
new file mode 100644
index 000000000000..8f7db9638d0a
--- /dev/null
+++ b/drivers/ata/pata_sil680.c
@@ -0,0 +1,381 @@
+/*
+ * pata_sil680.c 	- SIL680 PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based upon
+ *
+ * linux/drivers/ide/pci/siimage.c		Version 1.07	Nov 30, 2003
+ *
+ * Copyright (C) 2001-2002	Andre Hedrick <andre@linux-ide.org>
+ * Copyright (C) 2003		Red Hat <alan@redhat.com>
+ *
+ *  May be copied or modified under the terms of the GNU General Public License
+ *
+ *  Documentation publically available.
+ *
+ *	If you have strange problems with nVidia chipset systems please
+ *	see the SI support documentation and update your system BIOS
+ *	if neccessary
+ *
+ * TODO
+ *	If we know all our devices are LBA28 (or LBA28 sized)  we could use
+ *	the command fifo mode.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_sil680"
+#define DRV_VERSION "0.3.2"
+
+/**
+ *	sil680_selreg		-	return register base
+ *	@hwif: interface
+ *	@r: config offset
+ *
+ *	Turn a config register offset into the right address in either
+ *	PCI space or MMIO space to access the control register in question
+ *	Thankfully this is a configuration operation so isnt performance
+ *	criticial.
+ */
+
+static unsigned long sil680_selreg(struct ata_port *ap, int r)
+{
+	unsigned long base = 0xA0 + r;
+	base += (ap->port_no << 4);
+	return base;
+}
+
+/**
+ *	sil680_seldev		-	return register base
+ *	@hwif: interface
+ *	@r: config offset
+ *
+ *	Turn a config register offset into the right address in either
+ *	PCI space or MMIO space to access the control register in question
+ *	including accounting for the unit shift.
+ */
+
+static unsigned long sil680_seldev(struct ata_port *ap, struct ata_device *adev, int r)
+{
+	unsigned long base = 0xA0 + r;
+	base += (ap->port_no << 4);
+	base |= adev->devno ? 2 : 0;
+	return base;
+}
+
+
+/**
+ *	sil680_cable_detect	-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection. The SIL680 stores this in PCI config
+ *	space for us.
+ */
+
+static int sil680_cable_detect(struct ata_port *ap) {
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	unsigned long addr = sil680_selreg(ap, 0);
+	u8 ata66;
+	pci_read_config_byte(pdev, addr, &ata66);
+	if (ata66 & 1)
+		return ATA_CBL_PATA80;
+	else
+		return ATA_CBL_PATA40;
+}
+
+static int sil680_pre_reset(struct ata_port *ap)
+{
+	ap->cbl = sil680_cable_detect(ap);
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	sil680_bus_reset	-	reset the SIL680 bus
+ *	@ap: ATA port to reset
+ *
+ *	Perform the SIL680 housekeeping when doing an ATA bus reset
+ */
+
+static int sil680_bus_reset(struct ata_port *ap,unsigned int *classes)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	unsigned long addr = sil680_selreg(ap, 0);
+	u8 reset;
+
+	pci_read_config_byte(pdev, addr, &reset);
+	pci_write_config_byte(pdev, addr, reset | 0x03);
+	udelay(25);
+	pci_write_config_byte(pdev, addr, reset);
+	return ata_std_softreset(ap, classes);
+}
+
+static void sil680_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, sil680_pre_reset, sil680_bus_reset, NULL, ata_std_postreset);
+}
+
+/**
+ *	sil680_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the SIL680 registers for PIO mode. Note that the task speed
+ *	registers are shared between the devices so we must pick the lowest
+ *	mode for command work.
+ */
+
+static void sil680_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	static u16 speed_p[5] = { 0x328A, 0x2283, 0x1104, 0x10C3, 0x10C1 };
+	static u16 speed_t[5] = { 0x328A, 0x1281, 0x1281, 0x10C3, 0x10C1 };
+
+	unsigned long tfaddr = sil680_selreg(ap, 0x02);
+	unsigned long addr = sil680_seldev(ap, adev, 0x04);
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	int pio = adev->pio_mode - XFER_PIO_0;
+	int lowest_pio = pio;
+	u16 reg;
+
+	struct ata_device *pair = ata_dev_pair(adev);
+
+	if (pair != NULL && adev->pio_mode > pair->pio_mode)
+		lowest_pio = pair->pio_mode - XFER_PIO_0;
+
+	pci_write_config_word(pdev, addr, speed_p[pio]);
+	pci_write_config_word(pdev, tfaddr, speed_t[lowest_pio]);
+
+	pci_read_config_word(pdev, tfaddr-2, &reg);
+	reg &= ~0x0200;			/* Clear IORDY */
+	if (ata_pio_need_iordy(adev))
+		reg |= 0x0200;		/* Enable IORDY */
+	pci_write_config_word(pdev, tfaddr-2, reg);
+}
+
+/**
+ *	sil680_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Program the MWDMA/UDMA modes for the sil680 k
+ *	chipset. The MWDMA mode values are pulled from a lookup table
+ *	while the chipset uses mode number for UDMA.
+ */
+
+static void sil680_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	static u8 ultra_table[2][7] = {
+		{ 0x0C, 0x07, 0x05, 0x04, 0x02, 0x01, 0xFF },	/* 100MHz */
+		{ 0x0F, 0x0B, 0x07, 0x05, 0x03, 0x02, 0x01 },	/* 133Mhz */
+	};
+	static u16 dma_table[3] = { 0x2208, 0x10C2, 0x10C1 };
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	unsigned long ma = sil680_seldev(ap, adev, 0x08);
+	unsigned long ua = sil680_seldev(ap, adev, 0x0C);
+	unsigned long addr_mask = 0x80 + 4 * ap->port_no;
+	int port_shift = adev->devno * 4;
+	u8 scsc, mode;
+	u16 multi, ultra;
+
+	pci_read_config_byte(pdev, 0x8A, &scsc);
+	pci_read_config_byte(pdev, addr_mask, &mode);
+	pci_read_config_word(pdev, ma, &multi);
+	pci_read_config_word(pdev, ua, &ultra);
+
+	/* Mask timing bits */
+	ultra &= ~0x3F;
+	mode &= ~(0x03 << port_shift);
+
+	/* Extract scsc */
+	scsc = (scsc & 0x30) ? 1: 0;
+
+	if (adev->dma_mode >= XFER_UDMA_0) {
+		multi = 0x10C1;
+		ultra |= ultra_table[scsc][adev->dma_mode - XFER_UDMA_0];
+		mode |= (0x03 << port_shift);
+	} else {
+		multi = dma_table[adev->dma_mode - XFER_MW_DMA_0];
+		mode |= (0x02 << port_shift);
+	}
+	pci_write_config_byte(pdev, addr_mask, mode);
+	pci_write_config_word(pdev, ma, multi);
+	pci_write_config_word(pdev, ua, ultra);
+}
+
+static struct scsi_host_template sil680_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations sil680_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= sil680_set_piomode,
+	.set_dmamode	= sil680_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= sil680_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int sil680_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &sil680_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,
+		.port_ops = &sil680_port_ops
+	};
+	static struct ata_port_info info_slow = {
+		.sht = &sil680_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &sil680_port_ops
+	};
+	static struct ata_port_info *port_info[2] = {&info, &info};
+	static int printed_version;
+	u32 class_rev	= 0;
+	u8 tmpbyte	= 0;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev, "version " DRV_VERSION "\n");
+
+        pci_read_config_dword(pdev, PCI_CLASS_REVISION, &class_rev);
+        class_rev &= 0xff;
+        /* FIXME: double check */
+	pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE, (class_rev) ? 1 : 255);
+
+	pci_write_config_byte(pdev, 0x80, 0x00);
+	pci_write_config_byte(pdev, 0x84, 0x00);
+
+	pci_read_config_byte(pdev, 0x8A, &tmpbyte);
+
+	printk(KERN_INFO "sil680: BA5_EN = %d clock = %02X\n",
+			tmpbyte & 1, tmpbyte & 0x30);
+
+	switch(tmpbyte & 0x30) {
+		case 0x00:
+			/* 133 clock attempt to force it on */
+			pci_write_config_byte(pdev, 0x8A, tmpbyte|0x10);
+			break;
+		case 0x30:
+			/* if clocking is disabled */
+			/* 133 clock attempt to force it on */
+			pci_write_config_byte(pdev, 0x8A, tmpbyte & ~0x20);
+			break;
+		case 0x10:
+			/* 133 already */
+			break;
+		case 0x20:
+			/* BIOS set PCI x2 clocking */
+			break;
+	}
+
+	pci_read_config_byte(pdev,   0x8A, &tmpbyte);
+	printk(KERN_INFO "sil680: BA5_EN = %d clock = %02X\n",
+			tmpbyte & 1, tmpbyte & 0x30);
+	if ((tmpbyte & 0x30) == 0)
+		port_info[0] = port_info[1] = &info_slow;
+
+	pci_write_config_byte(pdev,  0xA1, 0x72);
+	pci_write_config_word(pdev,  0xA2, 0x328A);
+	pci_write_config_dword(pdev, 0xA4, 0x62DD62DD);
+	pci_write_config_dword(pdev, 0xA8, 0x43924392);
+	pci_write_config_dword(pdev, 0xAC, 0x40094009);
+	pci_write_config_byte(pdev,  0xB1, 0x72);
+	pci_write_config_word(pdev,  0xB2, 0x328A);
+	pci_write_config_dword(pdev, 0xB4, 0x62DD62DD);
+	pci_write_config_dword(pdev, 0xB8, 0x43924392);
+	pci_write_config_dword(pdev, 0xBC, 0x40094009);
+
+	switch(tmpbyte & 0x30) {
+		case 0x00: printk(KERN_INFO "sil680: 100MHz clock.\n");break;
+		case 0x10: printk(KERN_INFO "sil680: 133MHz clock.\n");break;
+		case 0x20: printk(KERN_INFO "sil680: Using PCI clock.\n");break;
+		/* This last case is _NOT_ ok */
+		case 0x30: printk(KERN_ERR "sil680: Clock disabled ?\n");
+			return -EIO;
+	}
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id sil680[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_CMD, PCI_DEVICE_ID_SII_680), },
+	{ 0, },
+};
+
+static struct pci_driver sil680_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= sil680,
+	.probe 		= sil680_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init sil680_init(void)
+{
+	return pci_register_driver(&sil680_pci_driver);
+}
+
+
+static void __exit sil680_exit(void)
+{
+	pci_unregister_driver(&sil680_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for SI680 PATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, sil680);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(sil680_init);
+module_exit(sil680_exit);
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
new file mode 100644
index 000000000000..4ad7fb44bfa6
--- /dev/null
+++ b/drivers/ata/pata_sis.c
@@ -0,0 +1,1030 @@
+/*
+ *    pata_sis.c - SiS ATA driver
+ *
+ *	(C) 2005 Red Hat <alan@redhat.com>
+ *
+ *    Based upon linux/drivers/ide/pci/sis5513.c
+ * Copyright (C) 1999-2000	Andre Hedrick <andre@linux-ide.org>
+ * Copyright (C) 2002		Lionel Bouton <Lionel.Bouton@inet6.fr>, Maintainer
+ * Copyright (C) 2003		Vojtech Pavlik <vojtech@suse.cz>
+ * SiS Taiwan		: for direct support and hardware.
+ * Daniela Engert	: for initial ATA100 advices and numerous others.
+ * John Fremlin, Manfred Spraul, Dave Morgan, Peter Kjellerstedt	:
+ *			  for checking code correctness, providing patches.
+ * Original tests and design on the SiS620 chipset.
+ * ATA100 tests and design on the SiS735 chipset.
+ * ATA16/33 support from specs
+ * ATA133 support for SiS961/962 by L.C. Chang <lcchang@sis.com.tw>
+ *
+ *
+ *	TODO
+ *	Check MWDMA on drives that don't support MWDMA speed pio cycles ?
+ *	More Testing
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+#include <linux/ata.h>
+
+#define DRV_NAME	"pata_sis"
+#define DRV_VERSION	"0.4.2"
+
+struct sis_chipset {
+	u16 device;			/* PCI host ID */
+	struct ata_port_info *info;	/* Info block */
+	/* Probably add family, cable detect type etc here to clean
+	   up code later */
+};
+
+/**
+ *	sis_port_base		-	return PCI configuration base for dev
+ *	@adev: device
+ *
+ *	Returns the base of the PCI configuration registers for this port
+ *	number.
+ */
+
+static int sis_port_base(struct ata_device *adev)
+{
+	return  0x40 + (4 * adev->ap->port_no) +  (2 * adev->devno);
+}
+
+/**
+ *	sis_133_pre_reset	-	check for 40/80 pin
+ *	@ap: Port
+ *
+ *	Perform cable detection for the later UDMA133 capable
+ *	SiS chipset.
+ */
+
+static int sis_133_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits sis_enable_bits[] = {
+		{ 0x4aU, 1U, 0x02UL, 0x02UL },	/* port 0 */
+		{ 0x4aU, 1U, 0x04UL, 0x04UL },	/* port 1 */
+	};
+	
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u16 tmp;
+
+	if (!pci_test_config_bits(pdev, &sis_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	/* The top bit of this register is the cable detect bit */
+	pci_read_config_word(pdev, 0x50 + 2 * ap->port_no, &tmp);
+	if (tmp & 0x8000)
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	sis_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_133_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, sis_133_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+
+/**
+ *	sis_66_pre_reset	-	check for 40/80 pin
+ *	@ap: Port
+ *
+ *	Perform cable detection on the UDMA66, UDMA100 and early UDMA133
+ *	SiS IDE controllers.
+ */
+
+static int sis_66_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits sis_enable_bits[] = {
+		{ 0x4aU, 1U, 0x02UL, 0x02UL },	/* port 0 */
+		{ 0x4aU, 1U, 0x04UL, 0x04UL },	/* port 1 */
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 tmp;
+
+	if (!pci_test_config_bits(pdev, &sis_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	/* Older chips keep cable detect in bits 4/5 of reg 0x48 */
+	pci_read_config_byte(pdev, 0x48, &tmp);
+	tmp >>= ap->port_no;
+	if (tmp & 0x10)
+		ap->cbl = ATA_CBL_PATA40;
+	else
+		ap->cbl = ATA_CBL_PATA80;
+
+	return ata_std_prereset(ap);
+}
+
+/**
+ *	sis_66_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *	@classes:
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_66_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, sis_66_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	sis_old_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+
+static int sis_old_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits sis_enable_bits[] = {
+		{ 0x4aU, 1U, 0x02UL, 0x02UL },	/* port 0 */
+		{ 0x4aU, 1U, 0x04UL, 0x04UL },	/* port 1 */
+	};
+	
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+
+	if (!pci_test_config_bits(pdev, &sis_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+
+/**
+ *	sis_old_error_handler - Probe specified port on PATA host controller
+ *	@ap: Port to probe
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_old_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, sis_old_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	sis_set_fifo	-	Set RWP fifo bits for this device
+ *	@ap: Port
+ *	@adev: Device
+ *
+ *	SIS chipsets implement prefetch/postwrite bits for each device
+ *	on both channels. This functionality is not ATAPI compatible and
+ *	must be configured according to the class of device present
+ */
+
+static void sis_set_fifo(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u8 fifoctrl;
+	u8 mask = 0x11;
+
+	mask <<= (2 * ap->port_no);
+	mask <<= adev->devno;
+
+	/* This holds various bits including the FIFO control */
+	pci_read_config_byte(pdev, 0x4B, &fifoctrl);
+	fifoctrl &= ~mask;
+
+	/* Enable for ATA (disk) only */
+	if (adev->class == ATA_DEV_ATA)
+		fifoctrl |= mask;
+	pci_write_config_byte(pdev, 0x4B, fifoctrl);
+}
+
+/**
+ *	sis_old_set_piomode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring for.
+ *
+ *	Set PIO mode for device, in host controller PCI config space. This
+ *	function handles PIO set up for all chips that are pre ATA100 and
+ *	also early ATA100 devices.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_old_set_piomode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int port = sis_port_base(adev);
+	u8 t1, t2;
+	int speed = adev->pio_mode - XFER_PIO_0;
+
+	const u8 active[]   = { 0x00, 0x07, 0x04, 0x03, 0x01 };
+	const u8 recovery[] = { 0x00, 0x06, 0x04, 0x03, 0x03 };
+
+	sis_set_fifo(ap, adev);
+
+	pci_read_config_byte(pdev, port, &t1);
+	pci_read_config_byte(pdev, port + 1, &t2);
+
+	t1 &= ~0x0F;	/* Clear active/recovery timings */
+	t2 &= ~0x07;
+
+	t1 |= active[speed];
+	t2 |= recovery[speed];
+
+	pci_write_config_byte(pdev, port, t1);
+	pci_write_config_byte(pdev, port + 1, t2);
+}
+
+/**
+ *	sis_100_set_pioode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring for.
+ *
+ *	Set PIO mode for device, in host controller PCI config space. This
+ *	function handles PIO set up for ATA100 devices and early ATA133.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_100_set_piomode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int port = sis_port_base(adev);
+	int speed = adev->pio_mode - XFER_PIO_0;
+
+	const u8 actrec[] = { 0x00, 0x67, 0x44, 0x33, 0x31 };
+
+	sis_set_fifo(ap, adev);
+
+	pci_write_config_byte(pdev, port, actrec[speed]);
+}
+
+/**
+ *	sis_133_set_pioode - Initialize host controller PATA PIO timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device we are configuring for.
+ *
+ *	Set PIO mode for device, in host controller PCI config space. This
+ *	function handles PIO set up for the later ATA133 devices.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_133_set_piomode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int port = 0x40;
+	u32 t1;
+	u32 reg54;
+	int speed = adev->pio_mode - XFER_PIO_0;
+
+	const u32 timing133[] = {
+		0x28269000,	/* Recovery << 24 | Act << 16 | Ini << 12 */
+		0x0C266000,
+		0x04263000,
+		0x0C0A3000,
+		0x05093000
+	};
+	const u32 timing100[] = {
+		0x1E1C6000,	/* Recovery << 24 | Act << 16 | Ini << 12 */
+		0x091C4000,
+		0x031C2000,
+		0x09072000,
+		0x04062000
+	};
+
+	sis_set_fifo(ap, adev);
+
+	/* If bit 14 is set then the registers are mapped at 0x70 not 0x40 */
+	pci_read_config_dword(pdev, 0x54, &reg54);
+	if (reg54 & 0x40000000)
+		port = 0x70;
+	port += 8 * ap->port_no +  4 * adev->devno;
+
+	pci_read_config_dword(pdev, port, &t1);
+	t1 &= 0xC0C00FFF;	/* Mask out timing */
+
+	if (t1 & 0x08)		/* 100 or 133 ? */
+		t1 |= timing133[speed];
+	else
+		t1 |= timing100[speed];
+	pci_write_config_byte(pdev, port, t1);
+}
+
+/**
+ *	sis_old_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *
+ *	Set UDMA/MWDMA mode for device, in host controller PCI config space.
+ *	Handles pre UDMA and UDMA33 devices. Supports MWDMA as well unlike
+ *	the old ide/pci driver.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_old_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int speed = adev->dma_mode - XFER_MW_DMA_0;
+	int drive_pci = sis_port_base(adev);
+	u16 timing;
+
+	const u16 mwdma_bits[] = { 0x707, 0x202, 0x202 };
+	const u16 udma_bits[]  = { 0xE000, 0xC000, 0xA000 };
+
+	pci_read_config_word(pdev, drive_pci, &timing);
+
+	if (adev->dma_mode < XFER_UDMA_0) {
+		/* bits 3-0 hold recovery timing bits 8-10 active timing and
+		   the higer bits are dependant on the device */
+		timing &= ~ 0x870F;
+		timing |= mwdma_bits[speed];
+		pci_write_config_word(pdev, drive_pci, timing);
+	} else {
+		/* Bit 15 is UDMA on/off, bit 13-14 are cycle time */
+		speed = adev->dma_mode - XFER_UDMA_0;
+		timing &= ~0x6000;
+		timing |= udma_bits[speed];
+	}
+}
+
+/**
+ *	sis_66_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *
+ *	Set UDMA/MWDMA mode for device, in host controller PCI config space.
+ *	Handles UDMA66 and early UDMA100 devices. Supports MWDMA as well unlike
+ *	the old ide/pci driver.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_66_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int speed = adev->dma_mode - XFER_MW_DMA_0;
+	int drive_pci = sis_port_base(adev);
+	u16 timing;
+
+	const u16 mwdma_bits[] = { 0x707, 0x202, 0x202 };
+	const u16 udma_bits[]  = { 0xF000, 0xD000, 0xB000, 0xA000, 0x9000};
+
+	pci_read_config_word(pdev, drive_pci, &timing);
+
+	if (adev->dma_mode < XFER_UDMA_0) {
+		/* bits 3-0 hold recovery timing bits 8-10 active timing and
+		   the higer bits are dependant on the device, bit 15 udma */
+		timing &= ~ 0x870F;
+		timing |= mwdma_bits[speed];
+	} else {
+		/* Bit 15 is UDMA on/off, bit 12-14 are cycle time */
+		speed = adev->dma_mode - XFER_UDMA_0;
+		timing &= ~0x6000;
+		timing |= udma_bits[speed];
+	}
+	pci_write_config_word(pdev, drive_pci, timing);
+}
+
+/**
+ *	sis_100_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *
+ *	Set UDMA/MWDMA mode for device, in host controller PCI config space.
+ *	Handles UDMA66 and early UDMA100 devices.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_100_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int speed = adev->dma_mode - XFER_MW_DMA_0;
+	int drive_pci = sis_port_base(adev);
+	u16 timing;
+
+	const u16 udma_bits[]  = { 0x8B00, 0x8700, 0x8500, 0x8300, 0x8200, 0x8100};
+
+	pci_read_config_word(pdev, drive_pci, &timing);
+
+	if (adev->dma_mode < XFER_UDMA_0) {
+		/* NOT SUPPORTED YET: NEED DATA SHEET. DITTO IN OLD DRIVER */
+	} else {
+		/* Bit 15 is UDMA on/off, bit 12-14 are cycle time */
+		speed = adev->dma_mode - XFER_UDMA_0;
+		timing &= ~0x0F00;
+		timing |= udma_bits[speed];
+	}
+	pci_write_config_word(pdev, drive_pci, timing);
+}
+
+/**
+ *	sis_133_early_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *
+ *	Set UDMA/MWDMA mode for device, in host controller PCI config space.
+ *	Handles early SiS 961 bridges. Supports MWDMA as well unlike
+ *	the old ide/pci driver.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_133_early_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int speed = adev->dma_mode - XFER_MW_DMA_0;
+	int drive_pci = sis_port_base(adev);
+	u16 timing;
+
+	const u16 udma_bits[]  = { 0x8F00, 0x8A00, 0x8700, 0x8500, 0x8300, 0x8200, 0x8100};
+
+	pci_read_config_word(pdev, drive_pci, &timing);
+
+	if (adev->dma_mode < XFER_UDMA_0) {
+		/* NOT SUPPORTED YET: NEED DATA SHEET. DITTO IN OLD DRIVER */
+	} else {
+		/* Bit 15 is UDMA on/off, bit 12-14 are cycle time */
+		speed = adev->dma_mode - XFER_UDMA_0;
+		timing &= ~0x0F00;
+		timing |= udma_bits[speed];
+	}
+	pci_write_config_word(pdev, drive_pci, timing);
+}
+
+/**
+ *	sis_133_set_dmamode - Initialize host controller PATA DMA timings
+ *	@ap: Port whose timings we are configuring
+ *	@adev: Device to program
+ *
+ *	Set UDMA/MWDMA mode for device, in host controller PCI config space.
+ *	Handles early SiS 961 bridges. Supports MWDMA as well unlike
+ *	the old ide/pci driver.
+ *
+ *	LOCKING:
+ *	None (inherited from caller).
+ */
+
+static void sis_133_set_dmamode (struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev	= to_pci_dev(ap->host->dev);
+	int speed = adev->dma_mode - XFER_MW_DMA_0;
+	int port = 0x40;
+	u32 t1;
+	u32 reg54;
+
+	/* bits 4- cycle time 8 - cvs time */
+	const u32 timing_u100[] = { 0x6B0, 0x470, 0x350, 0x140, 0x120, 0x110, 0x000 };
+	const u32 timing_u133[] = { 0x9F0, 0x6A0, 0x470, 0x250, 0x230, 0x220, 0x210 };
+
+	/* If bit 14 is set then the registers are mapped at 0x70 not 0x40 */
+	pci_read_config_dword(pdev, 0x54, &reg54);
+	if (reg54 & 0x40000000)
+		port = 0x70;
+	port += (8 * ap->port_no) +  (4 * adev->devno);
+
+	pci_read_config_dword(pdev, port, &t1);
+
+	if (adev->dma_mode < XFER_UDMA_0) {
+		t1 &= ~0x00000004;
+		/* FIXME: need data sheet to add MWDMA here. Also lacking on
+		   ide/pci driver */
+	} else {
+		speed = adev->dma_mode - XFER_UDMA_0;
+		/* if & 8 no UDMA133 - need info for ... */
+		t1 &= ~0x00000FF0;
+		t1 |= 0x00000004;
+		if (t1 & 0x08)
+			t1 |= timing_u133[speed];
+		else
+			t1 |= timing_u100[speed];
+	}
+	pci_write_config_dword(pdev, port, t1);
+}
+
+static struct scsi_host_template sis_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static const struct ata_port_operations sis_133_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= sis_133_set_piomode,
+	.set_dmamode		= sis_133_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= sis_133_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static const struct ata_port_operations sis_133_early_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= sis_100_set_piomode,
+	.set_dmamode		= sis_133_early_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= sis_66_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+	
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static const struct ata_port_operations sis_100_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= sis_100_set_piomode,
+	.set_dmamode		= sis_100_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= sis_66_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+	
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static const struct ata_port_operations sis_66_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= sis_old_set_piomode,
+	.set_dmamode		= sis_66_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= sis_66_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static const struct ata_port_operations sis_old_ops = {
+	.port_disable		= ata_port_disable,
+	.set_piomode		= sis_old_set_piomode,
+	.set_dmamode		= sis_old_set_dmamode,
+	.mode_filter		= ata_pci_default_filter,
+
+	.tf_load		= ata_tf_load,
+	.tf_read		= ata_tf_read,
+	.check_status		= ata_check_status,
+	.exec_command		= ata_exec_command,
+	.dev_select		= ata_std_dev_select,
+
+	.freeze			= ata_bmdma_freeze,
+	.thaw			= ata_bmdma_thaw,
+	.error_handler		= sis_old_error_handler,
+	.post_internal_cmd	= ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup		= ata_bmdma_setup,
+	.bmdma_start		= ata_bmdma_start,
+	.bmdma_stop		= ata_bmdma_stop,
+	.bmdma_status		= ata_bmdma_status,
+	.qc_prep		= ata_qc_prep,
+	.qc_issue		= ata_qc_issue_prot,
+	.data_xfer		= ata_pio_data_xfer,
+
+	.eng_timeout		= ata_eng_timeout,
+
+	.irq_handler		= ata_interrupt,
+	.irq_clear		= ata_bmdma_irq_clear,
+
+	.port_start		= ata_port_start,
+	.port_stop		= ata_port_stop,
+	.host_stop		= ata_host_stop,
+};
+
+static struct ata_port_info sis_info = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.mwdma_mask	= 0x07,
+	.udma_mask	= 0,
+	.port_ops	= &sis_old_ops,
+};
+static struct ata_port_info sis_info33 = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.mwdma_mask	= 0x07,
+	.udma_mask	= ATA_UDMA2,	/* UDMA 33 */
+	.port_ops	= &sis_old_ops,
+};
+static struct ata_port_info sis_info66 = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.udma_mask	= ATA_UDMA4,	/* UDMA 66 */
+	.port_ops	= &sis_66_ops,
+};
+static struct ata_port_info sis_info100 = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.udma_mask	= ATA_UDMA5,
+	.port_ops	= &sis_100_ops,
+};
+static struct ata_port_info sis_info100_early = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.udma_mask	= ATA_UDMA5,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.port_ops	= &sis_66_ops,
+};
+static struct ata_port_info sis_info133 = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.udma_mask	= ATA_UDMA6,
+	.port_ops	= &sis_133_ops,
+};
+static struct ata_port_info sis_info133_early = {
+	.sht		= &sis_sht,
+	.flags		= ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+	.pio_mask	= 0x1f,	/* pio0-4 */
+	.udma_mask	= ATA_UDMA6,
+	.port_ops	= &sis_133_early_ops,
+};
+
+
+static void sis_fixup(struct pci_dev *pdev, struct sis_chipset *sis)
+{
+	u16 regw;
+	u8 reg;
+
+	if (sis->info == &sis_info133) {
+		pci_read_config_word(pdev, 0x50, &regw);
+		if (regw & 0x08)
+			pci_write_config_word(pdev, 0x50, regw & ~0x08);
+		pci_read_config_word(pdev, 0x52, &regw);
+		if (regw & 0x08)
+			pci_write_config_word(pdev, 0x52, regw & ~0x08);
+		return;
+	}
+
+	if (sis->info == &sis_info133_early || sis->info == &sis_info100) {
+		/* Fix up latency */
+		pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x80);
+		/* Set compatibility bit */
+		pci_read_config_byte(pdev, 0x49, &reg);
+		if (!(reg & 0x01))
+			pci_write_config_byte(pdev, 0x49, reg | 0x01);
+		return;
+	}
+
+	if (sis->info == &sis_info66 || sis->info == &sis_info100_early) {
+		/* Fix up latency */
+		pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x80);
+		/* Set compatibility bit */
+		pci_read_config_byte(pdev, 0x52, &reg);
+		if (!(reg & 0x04))
+			pci_write_config_byte(pdev, 0x52, reg | 0x04);
+		return;
+	}
+
+	if (sis->info == &sis_info33) {
+		pci_read_config_byte(pdev, PCI_CLASS_PROG, &reg);
+		if (( reg & 0x0F ) != 0x00)
+			pci_write_config_byte(pdev, PCI_CLASS_PROG, reg & 0xF0);
+		/* Fall through to ATA16 fixup below */
+	}
+
+	if (sis->info == &sis_info || sis->info == &sis_info33) {
+		/* force per drive recovery and active timings
+		   needed on ATA_33 and below chips */
+		pci_read_config_byte(pdev, 0x52, &reg);
+		if (!(reg & 0x08))
+			pci_write_config_byte(pdev, 0x52, reg|0x08);
+		return;
+	}
+
+	BUG();
+}
+
+/**
+ *	sis_init_one - Register SiS ATA PCI device with kernel services
+ *	@pdev: PCI device to register
+ *	@ent: Entry in sis_pci_tbl matching with @pdev
+ *
+ *	Called from kernel PCI layer.  We probe for combined mode (sigh),
+ *	and then hand over control to libata, for it to do the rest.
+ *
+ *	LOCKING:
+ *	Inherited from PCI layer (may sleep).
+ *
+ *	RETURNS:
+ *	Zero on success, or -ERRNO value.
+ */
+
+static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	static int printed_version;
+	static struct ata_port_info *port_info[2];
+	struct ata_port_info *port;
+	struct pci_dev *host = NULL;
+	struct sis_chipset *chipset = NULL;
+
+	static struct sis_chipset sis_chipsets[] = {
+		{ 0x0745, &sis_info100 },
+		{ 0x0735, &sis_info100 },
+		{ 0x0733, &sis_info100 },
+		{ 0x0635, &sis_info100 },
+		{ 0x0633, &sis_info100 },
+
+		{ 0x0730, &sis_info100_early },	/* 100 with ATA 66 layout */
+		{ 0x0550, &sis_info100_early },	/* 100 with ATA 66 layout */
+
+		{ 0x0640, &sis_info66 },
+		{ 0x0630, &sis_info66 },
+		{ 0x0620, &sis_info66 },
+		{ 0x0540, &sis_info66 },
+		{ 0x0530, &sis_info66 },
+
+		{ 0x5600, &sis_info33 },
+		{ 0x5598, &sis_info33 },
+		{ 0x5597, &sis_info33 },
+		{ 0x5591, &sis_info33 },
+		{ 0x5582, &sis_info33 },
+		{ 0x5581, &sis_info33 },
+
+		{ 0x5596, &sis_info },
+		{ 0x5571, &sis_info },
+		{ 0x5517, &sis_info },
+		{ 0x5511, &sis_info },
+
+		{0}
+	};
+	static struct sis_chipset sis133_early = {
+		0x0, &sis_info133_early
+	};
+	static struct sis_chipset sis133 = {
+		0x0, &sis_info133
+	};
+	static struct sis_chipset sis100_early = {
+		0x0, &sis_info100_early
+	};
+	static struct sis_chipset sis100 = {
+		0x0, &sis_info100
+	};
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev,
+			   "version " DRV_VERSION "\n");
+
+	/* We have to find the bridge first */
+
+	for (chipset = &sis_chipsets[0]; chipset->device; chipset++) {
+		host = pci_get_device(PCI_VENDOR_ID_SI, chipset->device, NULL);
+		if (host != NULL) {
+			if (chipset->device == 0x630) {	/* SIS630 */
+				u8 host_rev;
+				pci_read_config_byte(host, PCI_REVISION_ID, &host_rev);
+				if (host_rev >= 0x30)	/* 630 ET */
+					chipset = &sis100_early;
+			}
+			break;
+		}
+	}
+
+	/* Look for concealed bridges */
+	if (host == NULL) {
+		/* Second check */
+		u32 idemisc;
+		u16 trueid;
+
+		/* Disable ID masking and register remapping then
+		   see what the real ID is */
+
+		pci_read_config_dword(pdev, 0x54, &idemisc);
+		pci_write_config_dword(pdev, 0x54, idemisc & 0x7fffffff);
+		pci_read_config_word(pdev, PCI_DEVICE_ID, &trueid);
+		pci_write_config_dword(pdev, 0x54, idemisc);
+
+		switch(trueid) {
+		case 0x5518:	/* SIS 962/963 */
+			chipset = &sis133;
+			if ((idemisc & 0x40000000) == 0) {
+				pci_write_config_dword(pdev, 0x54, idemisc | 0x40000000);
+				printk(KERN_INFO "SIS5513: Switching to 5513 register mapping\n");
+			}
+			break;
+		case 0x0180:	/* SIS 965/965L */
+			chipset =  &sis133;
+			break;
+		case 0x1180:	/* SIS 966/966L */
+			chipset =  &sis133;
+			break;
+		}
+	}
+
+	/* Further check */
+	if (chipset == NULL) {
+		struct pci_dev *lpc_bridge;
+		u16 trueid;
+		u8 prefctl;
+		u8 idecfg;
+		u8 sbrev;
+
+		/* Try the second unmasking technique */
+		pci_read_config_byte(pdev, 0x4a, &idecfg);
+		pci_write_config_byte(pdev, 0x4a, idecfg | 0x10);
+		pci_read_config_word(pdev, PCI_DEVICE_ID, &trueid);
+		pci_write_config_byte(pdev, 0x4a, idecfg);
+
+		switch(trueid) {
+		case 0x5517:
+			lpc_bridge = pci_get_slot(pdev->bus, 0x10); /* Bus 0 Dev 2 Fn 0 */
+			if (lpc_bridge == NULL)
+				break;
+			pci_read_config_byte(lpc_bridge, PCI_REVISION_ID, &sbrev);
+			pci_read_config_byte(pdev, 0x49, &prefctl);
+			pci_dev_put(lpc_bridge);
+
+			if (sbrev == 0x10 && (prefctl & 0x80)) {
+				chipset = &sis133_early;
+				break;
+			}
+			chipset = &sis100;
+			break;
+		}
+	}
+	pci_dev_put(host);
+
+	/* No chipset info, no support */
+	if (chipset == NULL)
+		return -ENODEV;
+
+	port = chipset->info;
+	port->private_data = chipset;
+
+	sis_fixup(pdev, chipset);
+
+	port_info[0] = port_info[1] = port;
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id sis_pci_tbl[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_SI, 0x5513), },	/* SiS 5513 */
+	{ PCI_DEVICE(PCI_VENDOR_ID_SI, 0x5518), },	/* SiS 5518 */
+	{ }
+};
+
+static struct pci_driver sis_pci_driver = {
+	.name			= DRV_NAME,
+	.id_table		= sis_pci_tbl,
+	.probe			= sis_init_one,
+	.remove			= ata_pci_remove_one,
+};
+
+static int __init sis_init(void)
+{
+	return pci_register_driver(&sis_pci_driver);
+}
+
+static void __exit sis_exit(void)
+{
+	pci_unregister_driver(&sis_pci_driver);
+}
+
+
+module_init(sis_init);
+module_exit(sis_exit);
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("SCSI low-level driver for SiS ATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, sis_pci_tbl);
+MODULE_VERSION(DRV_VERSION);
+
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
new file mode 100644
index 000000000000..47b290b5604a
--- /dev/null
+++ b/drivers/ata/pata_sl82c105.c
@@ -0,0 +1,388 @@
+/*
+ * pata_sl82c105.c 	- SL82C105 PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * Based in part on linux/drivers/ide/pci/sl82c105.c
+ * 		SL82C105/Winbond 553 IDE driver
+ *
+ * and in part on the documentation and errata sheet
+ */
+ 
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_sl82c105"
+#define DRV_VERSION "0.2.2"
+
+enum {
+	/*
+	 * SL82C105 PCI config register 0x40 bits.
+	 */
+	CTRL_IDE_IRQB	=	(1 << 30),
+	CTRL_IDE_IRQA   =	(1 << 28),
+	CTRL_LEGIRQ     =	(1 << 11),
+	CTRL_P1F16      =	(1 << 5),
+	CTRL_P1EN       =	(1 << 4),
+	CTRL_P0F16      =	(1 << 1),
+	CTRL_P0EN       =	(1 << 0)
+};
+
+/**
+ *	sl82c105_pre_reset		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+ 
+static int sl82c105_pre_reset(struct ata_port *ap)
+{
+	static const struct pci_bits sl82c105_enable_bits[] = {
+		{ 0x40, 1, 0x01, 0x01 },
+		{ 0x40, 1, 0x10, 0x10 }
+	};
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+
+	if (ap->port_no && !pci_test_config_bits(pdev, &sl82c105_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		dev_printk(KERN_INFO, &pdev->dev, "port disabled. ignoring.\n");
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+
+static void sl82c105_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, sl82c105_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+
+/**
+ *	sl82c105_configure_piomode	-	set chip PIO timing
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *	@pio: PIO mode
+ *
+ *	Called to do the PIO mode setup. Our timing registers are shared
+ *	so a configure_dmamode call will undo any work we do here and vice
+ *	versa
+ */
+ 
+static void sl82c105_configure_piomode(struct ata_port *ap, struct ata_device *adev, int pio)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static u16 pio_timing[5] = {
+		0x50D, 0x407, 0x304, 0x242, 0x240
+	};
+	u16 dummy;
+	int timing = 0x44 + (8 * ap->port_no) + (4 * adev->devno);
+	
+	pci_write_config_word(pdev, timing, pio_timing[pio]);
+	/* Can we lose this oddity of the old driver */
+	pci_read_config_word(pdev, timing, &dummy);
+}
+
+/**
+ *	sl82c105_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the PIO mode setup. Our timing registers are shared
+ *	but we want to set the PIO timing by default.
+ */
+ 
+static void sl82c105_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	sl82c105_configure_piomode(ap, adev, adev->pio_mode - XFER_PIO_0);
+}
+
+/**
+ *	sl82c105_configure_dmamode	-	set DMA mode in chip
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Load DMA cycle times into the chip ready for a DMA transfer
+ *	to occur.
+ */
+ 
+static void sl82c105_configure_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	static u16 dma_timing[3] = {
+		0x707, 0x201, 0x200
+	};
+	u16 dummy;
+	int timing = 0x44 + (8 * ap->port_no) + (4 * adev->devno);
+	int dma = adev->dma_mode - XFER_MW_DMA_0;
+	
+	pci_write_config_word(pdev, timing, dma_timing[dma]);
+	/* Can we lose this oddity of the old driver */
+	pci_read_config_word(pdev, timing, &dummy);
+}
+
+/**
+ *	sl82c105_set_dmamode	-	set initial DMA mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Called to do the DMA mode setup. This replaces the PIO timings
+ *	for the device in question. Set appropriate PIO timings not DMA
+ *	timings at this point.
+ */
+ 
+static void sl82c105_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	switch(adev->dma_mode) {
+		case XFER_MW_DMA_0:
+			sl82c105_configure_piomode(ap, adev, 1);
+			break;
+		case XFER_MW_DMA_1:
+			sl82c105_configure_piomode(ap, adev, 3);
+			break;
+		case XFER_MW_DMA_2:
+			sl82c105_configure_piomode(ap, adev, 3);
+			break;
+		default:
+			BUG();
+	}			
+}
+
+/**
+ *	sl82c105_reset_engine	-	Reset the DMA engine
+ *	@ap: ATA interface
+ *
+ *	The sl82c105 has some serious problems with the DMA engine
+ *	when transfers don't run as expected or ATAPI is used. The 
+ *	recommended fix is to reset the engine each use using a chip
+ *	test register.
+ */
+ 
+static void sl82c105_reset_engine(struct ata_port *ap)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u16 val;
+	
+	pci_read_config_word(pdev, 0x7E, &val);
+	pci_write_config_word(pdev, 0x7E, val | 4);
+	pci_write_config_word(pdev, 0x7E, val & ~4);
+}
+
+/**
+ *	sl82c105_bmdma_start		-	DMA engine begin
+ *	@qc: ATA command
+ *
+ *	Reset the DMA engine each use as recommended by the errata
+ *	document. 
+ *
+ *	FIXME: if we switch clock at BMDMA start/end we might get better
+ *	PIO performance on DMA capable devices.
+ */
+ 
+static void sl82c105_bmdma_start(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+
+	sl82c105_reset_engine(ap);
+	
+	/* Set the clocks for DMA */
+	sl82c105_configure_dmamode(ap, qc->dev);
+	/* Activate DMA */	
+	ata_bmdma_start(qc);
+}
+
+/**
+ *	sl82c105_bmdma_end		-	DMA engine stop
+ *	@qc: ATA command
+ *
+ *	Reset the DMA engine each use as recommended by the errata
+ *	document.
+ *
+ *	This function is also called to turn off DMA when a timeout occurs
+ *	during DMA operation. In both cases we need to reset the engine,
+ *	so no actual eng_timeout handler is required.
+ *
+ *	We assume bmdma_stop is always called if bmdma_start as called. If
+ *	not then we may need to wrap qc_issue.
+ */
+ 
+static void sl82c105_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	struct ata_port *ap = qc->ap;
+
+	ata_bmdma_stop(qc);
+	sl82c105_reset_engine(ap);
+	
+	/* This will redo the initial setup of the DMA device to matching
+	   PIO timings */
+	sl82c105_set_dmamode(ap, qc->dev);
+}
+
+static struct scsi_host_template sl82c105_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations sl82c105_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= sl82c105_set_piomode,
+	.set_dmamode	= sl82c105_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.error_handler	= sl82c105_error_handler,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= sl82c105_bmdma_start,
+	.bmdma_stop	= sl82c105_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+	
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};	
+
+/**
+ *	sl82c105_bridge_revision	-	find bridge version
+ *	@pdev: PCI device for the ATA function
+ *
+ *	Locates the PCI bridge associated with the ATA function and
+ *	providing it is a Winbond 553 reports the revision. If it cannot
+ *	find a revision or the right device it returns -1
+ */
+ 
+static int sl82c105_bridge_revision(struct pci_dev *pdev)
+{
+	struct pci_dev *bridge;
+	u8 rev;
+
+	/*
+	 * The bridge should be part of the same device, but function 0.
+	 */
+	bridge = pci_get_slot(pdev->bus,
+			       PCI_DEVFN(PCI_SLOT(pdev->devfn), 0));
+	if (!bridge)
+		return -1;
+
+	/*
+	 * Make sure it is a Winbond 553 and is an ISA bridge.
+	 */
+	if (bridge->vendor != PCI_VENDOR_ID_WINBOND ||
+	    bridge->device != PCI_DEVICE_ID_WINBOND_83C553 ||
+	    bridge->class >> 8 != PCI_CLASS_BRIDGE_ISA) {
+	    	pci_dev_put(bridge);
+		return -1;
+	}
+	/*
+	 * We need to find function 0's revision, not function 1
+	 */
+	pci_read_config_byte(bridge, PCI_REVISION_ID, &rev);
+
+	pci_dev_put(bridge);
+	return rev;
+}
+
+	
+static int sl82c105_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info_dma = {
+		.sht = &sl82c105_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &sl82c105_port_ops
+	};
+	static struct ata_port_info info_early = {
+		.sht = &sl82c105_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.port_ops = &sl82c105_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info_early, &info_early };
+	u32 val;
+	int rev;
+
+	rev = sl82c105_bridge_revision(dev);
+	
+	if (rev == -1)
+		dev_printk(KERN_WARNING, &dev->dev, "pata_sl82c105: Unable to find bridge, disabling DMA.\n");
+	else if (rev <= 5)
+		dev_printk(KERN_WARNING, &dev->dev, "pata_sl82c105: Early bridge revision, no DMA available.\n");
+	else {
+		port_info[0] = &info_dma;
+		port_info[1] = &info_dma;
+	}
+	
+	pci_read_config_dword(dev, 0x40, &val);
+	val |= CTRL_P0EN | CTRL_P0F16 | CTRL_P1F16;
+	pci_write_config_dword(dev, 0x40, val);
+
+
+	return ata_pci_init_one(dev, port_info, 1); /* For now */
+}
+
+static struct pci_device_id sl82c105[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_WINBOND, PCI_DEVICE_ID_WINBOND_82C105), },
+	{ 0, },
+};
+
+static struct pci_driver sl82c105_pci_driver = {
+	.name 		= DRV_NAME,
+	.id_table	= sl82c105,
+	.probe 		= sl82c105_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init sl82c105_init(void)
+{
+	return pci_register_driver(&sl82c105_pci_driver);
+}
+
+
+static void __exit sl82c105_exit(void)
+{
+	pci_unregister_driver(&sl82c105_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Sl82c105");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, sl82c105);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(sl82c105_init);
+module_exit(sl82c105_exit);
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
new file mode 100644
index 000000000000..7dd6ea3a21e4
--- /dev/null
+++ b/drivers/ata/pata_triflex.c
@@ -0,0 +1,285 @@
+/*
+ * pata_triflex.c 	- Compaq PATA for new ATA layer
+ *			  (C) 2005 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ * based upon
+ *
+ * triflex.c
+ *
+ * IDE Chipset driver for the Compaq TriFlex IDE controller.
+ *
+ * Known to work with the Compaq Workstation 5x00 series.
+ *
+ * Copyright (C) 2002 Hewlett-Packard Development Group, L.P.
+ * Author: Torben Mathiasen <torben.mathiasen@hp.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Loosely based on the piix & svwks drivers.
+ *
+ * Documentation:
+ *	Not publically available.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_triflex"
+#define DRV_VERSION "0.2.5"
+
+/**
+ *	triflex_probe_init		-	probe begin
+ *	@ap: ATA port
+ *
+ *	Set up cable type and use generic probe init
+ */
+
+static int triflex_probe_init(struct ata_port *ap)
+{
+	static const struct pci_bits triflex_enable_bits[] = {
+		{ 0x80, 1, 0x01, 0x01 },
+		{ 0x80, 1, 0x02, 0x02 }
+	};
+
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	
+	if (!pci_test_config_bits(pdev, &triflex_enable_bits[ap->port_no])) {
+		ata_port_disable(ap);
+		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+		return 0;
+	}
+	ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+
+
+static void triflex_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, triflex_probe_init, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	triflex_load_timing		-	timing configuration
+ *	@ap: ATA interface
+ *	@adev: Device on the bus
+ *	@speed: speed to configure
+ *
+ *	The Triflex has one set of timings per device per channel. This
+ *	means we must do some switching. As the PIO and DMA timings don't
+ *	match we have to do some reloading unlike PIIX devices where tuning
+ *	tricks can avoid it.
+ */
+
+static void triflex_load_timing(struct ata_port *ap, struct ata_device *adev, int speed)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 timing = 0;
+	u32 triflex_timing, old_triflex_timing;
+	int channel_offset = ap->port_no ? 0x74: 0x70;
+	unsigned int is_slave	= (adev->devno != 0);
+
+
+	pci_read_config_dword(pdev, channel_offset, &old_triflex_timing);
+	triflex_timing = old_triflex_timing;
+
+	switch(speed)
+	{
+		case XFER_MW_DMA_2:
+			timing = 0x0103;break;
+		case XFER_MW_DMA_1:
+			timing = 0x0203;break;
+		case XFER_MW_DMA_0:
+			timing = 0x0808;break;
+		case XFER_SW_DMA_2:
+		case XFER_SW_DMA_1:
+		case XFER_SW_DMA_0:
+			timing = 0x0F0F;break;
+		case XFER_PIO_4:
+			timing = 0x0202;break;
+		case XFER_PIO_3:
+			timing = 0x0204;break;
+		case XFER_PIO_2:
+			timing = 0x0404;break;
+		case XFER_PIO_1:
+			timing = 0x0508;break;
+		case XFER_PIO_0:
+			timing = 0x0808;break;
+		default:
+			BUG();
+	}
+	triflex_timing &= ~ (0xFFFF << (16 * is_slave));
+	triflex_timing |= (timing << (16 * is_slave));
+
+	if (triflex_timing != old_triflex_timing)
+		pci_write_config_dword(pdev, channel_offset, triflex_timing);
+}
+
+/**
+ *	triflex_set_piomode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	Use the timing loader to set up the PIO mode. We have to do this
+ *	because DMA start/stop will only be called once DMA occurs. If there
+ *	has been no DMA then the PIO timings are still needed.
+ */
+static void triflex_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	triflex_load_timing(ap, adev, adev->pio_mode);
+}
+
+/**
+ *	triflex_dma_start	-	DMA start callback
+ *	@qc: Command in progress
+ *
+ *	Usually drivers set the DMA timing at the point the set_dmamode call
+ *	is made. Triflex however requires we load new timings on the
+ *	transition or keep matching PIO/DMA pairs (ie MWDMA2/PIO4 etc).
+ *	We load the DMA timings just before starting DMA and then restore
+ *	the PIO timing when the DMA is finished.
+ */
+
+static void triflex_bmdma_start(struct ata_queued_cmd *qc)
+{
+	triflex_load_timing(qc->ap, qc->dev, qc->dev->dma_mode);
+	ata_bmdma_start(qc);
+}
+
+/**
+ *	triflex_dma_stop	-	DMA stop callback
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *
+ *	We loaded new timings in dma_start, as a result we need to restore
+ *	the PIO timings in dma_stop so that the next command issue gets the
+ *	right clock values.
+ */
+
+static void triflex_bmdma_stop(struct ata_queued_cmd *qc)
+{
+	ata_bmdma_stop(qc);
+	triflex_load_timing(qc->ap, qc->dev, qc->dev->pio_mode);
+}
+
+static struct scsi_host_template triflex_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations triflex_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= triflex_set_piomode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= triflex_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= triflex_bmdma_start,
+	.bmdma_stop	= triflex_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static int triflex_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	static struct ata_port_info info = {
+		.sht = &triflex_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &triflex_port_ops
+	};
+	static struct ata_port_info *port_info[2] = { &info, &info };
+	static int printed_version;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &dev->dev, "version " DRV_VERSION "\n");
+
+	return ata_pci_init_one(dev, port_info, 2);
+}
+
+static const struct pci_device_id triflex[] = {
+	{ PCI_VENDOR_ID_COMPAQ, PCI_DEVICE_ID_COMPAQ_TRIFLEX_IDE,
+				  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
+	{ 0, },
+};
+
+static struct pci_driver triflex_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= triflex,
+	.probe 		= triflex_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init triflex_init(void)
+{
+	return pci_register_driver(&triflex_pci_driver);
+}
+
+
+static void __exit triflex_exit(void)
+{
+	pci_unregister_driver(&triflex_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for Compaq Triflex");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, triflex);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(triflex_init);
+module_exit(triflex_exit);
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
new file mode 100644
index 000000000000..2580e1606d9e
--- /dev/null
+++ b/drivers/ata/pata_via.c
@@ -0,0 +1,568 @@
+/*
+ * pata_via.c 	- VIA PATA for new ATA layer
+ *			  (C) 2005-2006 Red Hat Inc
+ *			  Alan Cox <alan@redhat.com>
+ *
+ *  Documentation
+ *	Most chipset documentation available under NDA only
+ *
+ *  VIA version guide
+ *	VIA VT82C561	-	early design, uses ata_generic currently
+ *	VIA VT82C576	-	MWDMA, 33Mhz
+ *	VIA VT82C586	-	MWDMA, 33Mhz
+ *	VIA VT82C586a	-	Added UDMA to 33Mhz
+ *	VIA VT82C586b	-	UDMA33
+ *	VIA VT82C596a	-	Nonfunctional UDMA66
+ *	VIA VT82C596b	-	Working UDMA66
+ *	VIA VT82C686	-	Nonfunctional UDMA66
+ *	VIA VT82C686a	-	Working UDMA66
+ *	VIA VT82C686b	-	Updated to UDMA100
+ *	VIA VT8231	-	UDMA100
+ *	VIA VT8233	-	UDMA100
+ *	VIA VT8233a	-	UDMA133
+ *	VIA VT8233c	-	UDMA100
+ *	VIA VT8235	-	UDMA133
+ *	VIA VT8237	-	UDMA133
+ *
+ *	Most registers remain compatible across chips. Others start reserved
+ *	and acquire sensible semantics if set to 1 (eg cable detect). A few
+ *	exceptions exist, notably around the FIFO settings.
+ *
+ *	One additional quirk of the VIA design is that like ALi they use few
+ *	PCI IDs for a lot of chips.
+ *
+ *	Based heavily on:
+ *
+ * Version 3.38
+ *
+ * VIA IDE driver for Linux. Supported southbridges:
+ *
+ *   vt82c576, vt82c586, vt82c586a, vt82c586b, vt82c596a, vt82c596b,
+ *   vt82c686, vt82c686a, vt82c686b, vt8231, vt8233, vt8233c, vt8233a,
+ *   vt8235, vt8237
+ *
+ * Copyright (c) 2000-2002 Vojtech Pavlik
+ *
+ * Based on the work of:
+ *	Michel Aubry
+ *	Jeff Garzik
+ *	Andre Hedrick
+
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <linux/delay.h>
+#include <scsi/scsi_host.h>
+#include <linux/libata.h>
+
+#define DRV_NAME "pata_via"
+#define DRV_VERSION "0.1.13"
+
+/*
+ *	The following comes directly from Vojtech Pavlik's ide/pci/via82cxxx
+ *	driver.
+ */
+
+enum {
+	VIA_UDMA	= 0x007,
+	VIA_UDMA_NONE	= 0x000,
+	VIA_UDMA_33	= 0x001,
+	VIA_UDMA_66	= 0x002,
+	VIA_UDMA_100	= 0x003,
+	VIA_UDMA_133	= 0x004,
+	VIA_BAD_PREQ	= 0x010, /* Crashes if PREQ# till DDACK# set */
+	VIA_BAD_CLK66	= 0x020, /* 66 MHz clock doesn't work correctly */
+	VIA_SET_FIFO	= 0x040, /* Needs to have FIFO split set */
+	VIA_NO_UNMASK	= 0x080, /* Doesn't work with IRQ unmasking on */
+	VIA_BAD_ID	= 0x100, /* Has wrong vendor ID (0x1107) */
+	VIA_BAD_AST	= 0x200, /* Don't touch Address Setup Timing */
+	VIA_NO_ENABLES	= 0x400, /* Has no enablebits */
+};
+
+/*
+ * VIA SouthBridge chips.
+ */
+
+static const struct via_isa_bridge {
+	const char *name;
+	u16 id;
+	u8 rev_min;
+	u8 rev_max;
+	u16 flags;
+} via_isa_bridges[] = {
+	{ "cx700",	PCI_DEVICE_ID_VIA_CX700,    0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
+	{ "vt6410",	PCI_DEVICE_ID_VIA_6410,     0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_NO_ENABLES},
+	{ "vt8237a",	PCI_DEVICE_ID_VIA_8237A,    0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
+	{ "vt8237",	PCI_DEVICE_ID_VIA_8237,     0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
+	{ "vt8235",	PCI_DEVICE_ID_VIA_8235,     0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
+	{ "vt8233a",	PCI_DEVICE_ID_VIA_8233A,    0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
+	{ "vt8233c",	PCI_DEVICE_ID_VIA_8233C_0,  0x00, 0x2f, VIA_UDMA_100 },
+	{ "vt8233",	PCI_DEVICE_ID_VIA_8233_0,   0x00, 0x2f, VIA_UDMA_100 },
+	{ "vt8231",	PCI_DEVICE_ID_VIA_8231,     0x00, 0x2f, VIA_UDMA_100 },
+	{ "vt82c686b",	PCI_DEVICE_ID_VIA_82C686,   0x40, 0x4f, VIA_UDMA_100 },
+	{ "vt82c686a",	PCI_DEVICE_ID_VIA_82C686,   0x10, 0x2f, VIA_UDMA_66 },
+	{ "vt82c686",	PCI_DEVICE_ID_VIA_82C686,   0x00, 0x0f, VIA_UDMA_33 | VIA_BAD_CLK66 },
+	{ "vt82c596b",	PCI_DEVICE_ID_VIA_82C596,   0x10, 0x2f, VIA_UDMA_66 },
+	{ "vt82c596a",	PCI_DEVICE_ID_VIA_82C596,   0x00, 0x0f, VIA_UDMA_33 | VIA_BAD_CLK66 },
+	{ "vt82c586b",	PCI_DEVICE_ID_VIA_82C586_0, 0x47, 0x4f, VIA_UDMA_33 | VIA_SET_FIFO },
+	{ "vt82c586b",	PCI_DEVICE_ID_VIA_82C586_0, 0x40, 0x46, VIA_UDMA_33 | VIA_SET_FIFO | VIA_BAD_PREQ },
+	{ "vt82c586b",	PCI_DEVICE_ID_VIA_82C586_0, 0x30, 0x3f, VIA_UDMA_33 | VIA_SET_FIFO },
+	{ "vt82c586a",	PCI_DEVICE_ID_VIA_82C586_0, 0x20, 0x2f, VIA_UDMA_33 | VIA_SET_FIFO },
+	{ "vt82c586",	PCI_DEVICE_ID_VIA_82C586_0, 0x00, 0x0f, VIA_UDMA_NONE | VIA_SET_FIFO },
+	{ "vt82c576",	PCI_DEVICE_ID_VIA_82C576,   0x00, 0x2f, VIA_UDMA_NONE | VIA_SET_FIFO | VIA_NO_UNMASK },
+	{ "vt82c576",	PCI_DEVICE_ID_VIA_82C576,   0x00, 0x2f, VIA_UDMA_NONE | VIA_SET_FIFO | VIA_NO_UNMASK | VIA_BAD_ID },
+	{ NULL }
+};
+
+/**
+ *	via_cable_detect	-	cable detection
+ *	@ap: ATA port
+ *
+ *	Perform cable detection. Actually for the VIA case the BIOS
+ *	already did this for us. We read the values provided by the
+ *	BIOS. If you are using an 8235 in a non-PC configuration you
+ *	may need to update this code.
+ *
+ *	Hotplug also impacts on this.
+ */
+
+static int via_cable_detect(struct ata_port *ap) {
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	u32 ata66;
+
+	pci_read_config_dword(pdev, 0x50, &ata66);
+	/* Check both the drive cable reporting bits, we might not have
+	   two drives */
+	if (ata66 & (0x10100000 >> (16 * ap->port_no)))
+		return ATA_CBL_PATA80;
+	else
+		return ATA_CBL_PATA40;
+}
+
+static int via_pre_reset(struct ata_port *ap)
+{
+	const struct via_isa_bridge *config = ap->host->private_data;
+
+	if (!(config->flags & VIA_NO_ENABLES)) {
+		static const struct pci_bits via_enable_bits[] = {
+			{ 0x40, 1, 0x02, 0x02 },
+			{ 0x40, 1, 0x01, 0x01 }
+		};
+
+		struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+		
+		if (!pci_test_config_bits(pdev, &via_enable_bits[ap->port_no])) {
+			ata_port_disable(ap);
+			printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
+			return 0;
+		}
+	}
+
+	if ((config->flags & VIA_UDMA) >= VIA_UDMA_66)
+		ap->cbl = via_cable_detect(ap);
+	else
+		ap->cbl = ATA_CBL_PATA40;
+	return ata_std_prereset(ap);
+}
+
+
+/**
+ *	via_error_handler		-	reset for VIA chips
+ *	@ap: ATA port
+ *
+ *	Handle the reset callback for the later chips with cable detect
+ */
+
+static void via_error_handler(struct ata_port *ap)
+{
+	ata_bmdma_drive_eh(ap, via_pre_reset, ata_std_softreset, NULL, ata_std_postreset);
+}
+
+/**
+ *	via_do_set_mode	-	set initial PIO mode data
+ *	@ap: ATA interface
+ *	@adev: ATA device
+ *	@mode: ATA mode being programmed
+ *	@tdiv: Clocks per PCI clock
+ *	@set_ast: Set to program address setup
+ *	@udma_type: UDMA mode/format of registers
+ *
+ *	Program the VIA registers for DMA and PIO modes. Uses the ata timing
+ *	support in order to compute modes.
+ *
+ *	FIXME: Hotplug will require we serialize multiple mode changes
+ *	on the two channels.
+ */
+
+static void via_do_set_mode(struct ata_port *ap, struct ata_device *adev, int mode, int tdiv, int set_ast, int udma_type)
+{
+	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
+	struct ata_device *peer = ata_dev_pair(adev);
+	struct ata_timing t, p;
+	static int via_clock = 33333;	/* Bus clock in kHZ - ought to be tunable one day */
+	unsigned long T =  1000000000 / via_clock;
+	unsigned long UT = T/tdiv;
+	int ut;
+	int offset = 3 - (2*ap->port_no) - adev->devno;
+
+
+	/* Calculate the timing values we require */
+	ata_timing_compute(adev, mode, &t, T, UT);
+
+	/* We share 8bit timing so we must merge the constraints */
+	if (peer) {
+		if (peer->pio_mode) {
+			ata_timing_compute(peer, peer->pio_mode, &p, T, UT);
+			ata_timing_merge(&p, &t, &t, ATA_TIMING_8BIT);
+		}
+	}
+
+	/* Address setup is programmable but breaks on UDMA133 setups */
+	if (set_ast) {
+		u8 setup;	/* 2 bits per drive */
+		int shift = 2 * offset;
+
+		pci_read_config_byte(pdev, 0x4C, &setup);
+		setup &= ~(3 << shift);
+		setup |= FIT(t.setup, 1, 4) << shift;	/* 1,4 or 1,4 - 1  FIXME */
+		pci_write_config_byte(pdev, 0x4C, setup);
+	}
+
+	/* Load the PIO mode bits */
+	pci_write_config_byte(pdev, 0x4F - ap->port_no,
+		((FIT(t.act8b, 1, 16) - 1) << 4) | (FIT(t.rec8b, 1, 16) - 1));
+	pci_write_config_byte(pdev, 0x48 + offset,
+		((FIT(t.active, 1, 16) - 1) << 4) | (FIT(t.recover, 1, 16) - 1));
+
+	/* Load the UDMA bits according to type */
+	switch(udma_type) {
+		default:
+			/* BUG() ? */
+			/* fall through */
+		case 33:
+			ut = t.udma ? (0xe0 | (FIT(t.udma, 2, 5) - 2)) : 0x03;
+			break;
+		case 66:
+			ut = t.udma ? (0xe8 | (FIT(t.udma, 2, 9) - 2)) : 0x0f;
+			break;
+		case 100:
+			ut = t.udma ? (0xe0 | (FIT(t.udma, 2, 9) - 2)) : 0x07;
+			break;
+		case 133:
+			ut = t.udma ? (0xe0 | (FIT(t.udma, 2, 9) - 2)) : 0x07;
+			break;
+	}
+	/* Set UDMA unless device is not UDMA capable */
+	if (udma_type)
+		pci_write_config_byte(pdev, 0x50 + offset, ut);
+}
+
+static void via_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	const struct via_isa_bridge *config = ap->host->private_data;
+	int set_ast = (config->flags & VIA_BAD_AST) ? 0 : 1;
+	int mode = config->flags & VIA_UDMA;
+	static u8 tclock[5] = { 1, 1, 2, 3, 4 };
+	static u8 udma[5] = { 0, 33, 66, 100, 133 };
+
+	via_do_set_mode(ap, adev, adev->pio_mode, tclock[mode], set_ast, udma[mode]);
+}
+
+static void via_set_dmamode(struct ata_port *ap, struct ata_device *adev)
+{
+	const struct via_isa_bridge *config = ap->host->private_data;
+	int set_ast = (config->flags & VIA_BAD_AST) ? 0 : 1;
+	int mode = config->flags & VIA_UDMA;
+	static u8 tclock[5] = { 1, 1, 2, 3, 4 };
+	static u8 udma[5] = { 0, 33, 66, 100, 133 };
+
+	via_do_set_mode(ap, adev, adev->dma_mode, tclock[mode], set_ast, udma[mode]);
+}
+
+static struct scsi_host_template via_sht = {
+	.module			= THIS_MODULE,
+	.name			= DRV_NAME,
+	.ioctl			= ata_scsi_ioctl,
+	.queuecommand		= ata_scsi_queuecmd,
+	.can_queue		= ATA_DEF_QUEUE,
+	.this_id		= ATA_SHT_THIS_ID,
+	.sg_tablesize		= LIBATA_MAX_PRD,
+	.max_sectors		= ATA_MAX_SECTORS,
+	.cmd_per_lun		= ATA_SHT_CMD_PER_LUN,
+	.emulated		= ATA_SHT_EMULATED,
+	.use_clustering		= ATA_SHT_USE_CLUSTERING,
+	.proc_name		= DRV_NAME,
+	.dma_boundary		= ATA_DMA_BOUNDARY,
+	.slave_configure	= ata_scsi_slave_config,
+	.bios_param		= ata_std_bios_param,
+};
+
+static struct ata_port_operations via_port_ops = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= via_set_piomode,
+	.set_dmamode	= via_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= via_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+static struct ata_port_operations via_port_ops_noirq = {
+	.port_disable	= ata_port_disable,
+	.set_piomode	= via_set_piomode,
+	.set_dmamode	= via_set_dmamode,
+	.mode_filter	= ata_pci_default_filter,
+
+	.tf_load	= ata_tf_load,
+	.tf_read	= ata_tf_read,
+	.check_status 	= ata_check_status,
+	.exec_command	= ata_exec_command,
+	.dev_select 	= ata_std_dev_select,
+
+	.freeze		= ata_bmdma_freeze,
+	.thaw		= ata_bmdma_thaw,
+	.error_handler	= via_error_handler,
+	.post_internal_cmd = ata_bmdma_post_internal_cmd,
+
+	.bmdma_setup 	= ata_bmdma_setup,
+	.bmdma_start 	= ata_bmdma_start,
+	.bmdma_stop	= ata_bmdma_stop,
+	.bmdma_status 	= ata_bmdma_status,
+
+	.qc_prep 	= ata_qc_prep,
+	.qc_issue	= ata_qc_issue_prot,
+	.eng_timeout	= ata_eng_timeout,
+	.data_xfer	= ata_pio_data_xfer_noirq,
+
+	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
+	.port_start	= ata_port_start,
+	.port_stop	= ata_port_stop,
+	.host_stop	= ata_host_stop
+};
+
+/**
+ *	via_init_one		-	discovery callback
+ *	@pdev: PCI device ID
+ *	@id: PCI table info
+ *
+ *	A VIA IDE interface has been discovered. Figure out what revision
+ *	and perform configuration work before handing it to the ATA layer
+ */
+
+static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	/* Early VIA without UDMA support */
+	static struct ata_port_info via_mwdma_info = {
+		.sht = &via_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &via_port_ops
+	};
+	/* Ditto with IRQ masking required */
+	static struct ata_port_info via_mwdma_info_borked = {
+		.sht = &via_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.port_ops = &via_port_ops_noirq,
+	};
+	/* VIA UDMA 33 devices (and borked 66) */
+	static struct ata_port_info via_udma33_info = {
+		.sht = &via_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7,
+		.port_ops = &via_port_ops
+	};
+	/* VIA UDMA 66 devices */
+	static struct ata_port_info via_udma66_info = {
+		.sht = &via_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x1f,
+		.port_ops = &via_port_ops
+	};
+	/* VIA UDMA 100 devices */
+	static struct ata_port_info via_udma100_info = {
+		.sht = &via_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x3f,
+		.port_ops = &via_port_ops
+	};
+	/* UDMA133 with bad AST (All current 133) */
+	static struct ata_port_info via_udma133_info = {
+		.sht = &via_sht,
+		.flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST,
+		.pio_mask = 0x1f,
+		.mwdma_mask = 0x07,
+		.udma_mask = 0x7f,	/* FIXME: should check north bridge */
+		.port_ops = &via_port_ops
+	};
+	struct ata_port_info *port_info[2], *type;
+	struct pci_dev *isa = NULL;
+	const struct via_isa_bridge *config;
+	static int printed_version;
+	u8 t;
+	u8 enable;
+	u32 timing;
+
+	if (!printed_version++)
+		dev_printk(KERN_DEBUG, &pdev->dev, "version " DRV_VERSION "\n");
+
+	/* To find out how the IDE will behave and what features we
+	   actually have to look at the bridge not the IDE controller */
+	for (config = via_isa_bridges; config->id; config++)
+		if ((isa = pci_get_device(PCI_VENDOR_ID_VIA +
+			!!(config->flags & VIA_BAD_ID),
+			config->id, NULL))) {
+
+			pci_read_config_byte(isa, PCI_REVISION_ID, &t);
+			if (t >= config->rev_min &&
+			    t <= config->rev_max)
+				break;
+			pci_dev_put(isa);
+		}
+
+	if (!config->id) {
+		printk(KERN_WARNING "via: Unknown VIA SouthBridge, disabling.\n");
+		return -ENODEV;
+	}
+	pci_dev_put(isa);
+
+	/* 0x40 low bits indicate enabled channels */
+	pci_read_config_byte(pdev, 0x40 , &enable);
+	enable &= 3;
+	if (enable == 0) {
+		return -ENODEV;
+	}
+
+	/* Initialise the FIFO for the enabled channels. */
+	if (config->flags & VIA_SET_FIFO) {
+		u8 fifo_setting[4] = {0x00, 0x60, 0x00, 0x20};
+		u8 fifo;
+
+		pci_read_config_byte(pdev, 0x43, &fifo);
+
+		/* Clear PREQ# until DDACK# for errata */
+		if (config->flags & VIA_BAD_PREQ)
+			fifo &= 0x7F;
+		else
+			fifo &= 0x9f;
+		/* Turn on FIFO for enabled channels */
+		fifo |= fifo_setting[enable];
+		pci_write_config_byte(pdev, 0x43, fifo);
+	}
+	/* Clock set up */
+	switch(config->flags & VIA_UDMA) {
+		case VIA_UDMA_NONE:
+			if (config->flags & VIA_NO_UNMASK)
+				type = &via_mwdma_info_borked;
+			else
+				type = &via_mwdma_info;
+			break;
+		case VIA_UDMA_33:
+			type = &via_udma33_info;
+			break;
+		case VIA_UDMA_66:
+			type = &via_udma66_info;
+			/* The 66 MHz devices require we enable the clock */
+			pci_read_config_dword(pdev, 0x50, &timing);
+			timing |= 0x80008;
+			pci_write_config_dword(pdev, 0x50, timing);
+			break;
+		case VIA_UDMA_100:
+			type = &via_udma100_info;
+			break;
+		case VIA_UDMA_133:
+			type = &via_udma133_info;
+			break;
+		default:
+			WARN_ON(1);
+			return -ENODEV;
+	}
+
+	if (config->flags & VIA_BAD_CLK66) {
+		/* Disable the 66MHz clock on problem devices */
+		pci_read_config_dword(pdev, 0x50, &timing);
+		timing &= ~0x80008;
+		pci_write_config_dword(pdev, 0x50, timing);
+	}
+
+	/* We have established the device type, now fire it up */
+	type->private_data = (void *)config;
+
+	port_info[0] = port_info[1] = type;
+	return ata_pci_init_one(pdev, port_info, 2);
+}
+
+static const struct pci_device_id via[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C576_1), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_1), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_6410), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_SATA_EIDE), },
+	{ 0, },
+};
+
+static struct pci_driver via_pci_driver = {
+        .name 		= DRV_NAME,
+	.id_table	= via,
+	.probe 		= via_init_one,
+	.remove		= ata_pci_remove_one
+};
+
+static int __init via_init(void)
+{
+	return pci_register_driver(&via_pci_driver);
+}
+
+
+static void __exit via_exit(void)
+{
+	pci_unregister_driver(&via_pci_driver);
+}
+
+
+MODULE_AUTHOR("Alan Cox");
+MODULE_DESCRIPTION("low-level driver for VIA PATA");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, via);
+MODULE_VERSION(DRV_VERSION);
+
+module_init(via_init);
+module_exit(via_exit);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 563885cb0995..a6d818148ae0 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -46,7 +46,7 @@
 #undef ATA_VERBOSE_DEBUG	/* yet more debugging output */
 #undef ATA_IRQ_TRAP		/* define to ack screaming irqs */
 #undef ATA_NDEBUG		/* define to disable quick runtime checks */
-#undef ATA_ENABLE_PATA		/* define to enable PATA support in some
+#define ATA_ENABLE_PATA		/* define to enable PATA support in some
 				 * low-level drivers */
 
 
-- 
cgit v1.2.3


From 187afbed1814ea0851bf30bacbf807217dd7864b Mon Sep 17 00:00:00 2001
From: Jon Masters <jcm@redhat.com>
Date: Mon, 28 Aug 2006 17:08:21 -0500
Subject: [SCSI] MODULE_FIRMWARE for binary firmware(s)

Right now, various kernel modules are being migrated over to use
request_firmware in order to pull in binary firmware blobs from userland
when the module is loaded. This makes sense.

However, there is right now little mechanism in place to automatically
determine which binary firmware blobs must be included with a kernel in
order to satisfy the prerequisites of these drivers. This affects
vendors, but also regular users to a certain extent too.

The attached patch introduces MODULE_FIRMWARE as a mechanism for
advertising that a particular firmware file is to be loaded - it will
then show up via modinfo and could be used e.g. when packaging a kernel.

Signed-off-by: Jon Masters <jcm@redhat.com>

Comments added in line with all the other MODULE_ tag

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/module.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/linux/module.h b/include/linux/module.h
index 0dfb794c52d3..d4486cc2e7fe 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -156,6 +156,11 @@ extern struct module __this_module;
 */
 #define MODULE_VERSION(_version) MODULE_INFO(version, _version)
 
+/* Optional firmware file (or files) needed by the module
+ * format is simply firmware file name.  Multiple firmware
+ * files require multiple MODULE_FIRMWARE() specifiers */
+#define MODULE_FIRMWARE(_firmware) MODULE_INFO(firmware, _firmware)
+
 /* Given an address, look for it in the exception tables */
 const struct exception_table_entry *search_exception_tables(unsigned long add);
 
-- 
cgit v1.2.3


From 9bec2e38527a9f2497b3d976715c672d08d6160d Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Thu, 31 Aug 2006 00:02:15 -0400
Subject: [libata] Trim trailing whitespace.

---
 drivers/ata/ahci.c     | 2 +-
 drivers/ata/pdc_adma.c | 2 +-
 include/linux/ata.h    | 4 ++--
 include/linux/libata.h | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 3f1106fdaed1..1aabc81d82f1 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1075,7 +1075,7 @@ static void ahci_host_intr(struct ata_port *ap)
 		return;
 
 	/* ignore interim PIO setup fis interrupts */
-	if (ata_tag_valid(ap->active_tag) && (status & PORT_IRQ_PIOS_FIS)) 
+	if (ata_tag_valid(ap->active_tag) && (status & PORT_IRQ_PIOS_FIS))
 		return;
 
 	if (ata_ratelimit())
diff --git a/drivers/ata/pdc_adma.c b/drivers/ata/pdc_adma.c
index 912211ada816..0e23ecb77bc2 100644
--- a/drivers/ata/pdc_adma.c
+++ b/drivers/ata/pdc_adma.c
@@ -497,7 +497,7 @@ static inline unsigned int adma_intr_mmio(struct ata_host *host)
 					continue;
 				DPRINTK("ata%u: protocol %d (dev_stat 0x%X)\n",
 					ap->id, qc->tf.protocol, status);
-		
+
 				/* complete taskfile transaction */
 				pp->state = adma_state_idle;
 				qc->err_mask |= ac_err_mask(status);
diff --git a/include/linux/ata.h b/include/linux/ata.h
index 991b858acc30..d89441907024 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -320,8 +320,8 @@ static inline unsigned int ata_id_major_version(const u16 *id)
 
 static inline int ata_id_current_chs_valid(const u16 *id)
 {
-	/* For ATA-1 devices, if the INITIALIZE DEVICE PARAMETERS command 
-	   has not been issued to the device then the values of 
+	/* For ATA-1 devices, if the INITIALIZE DEVICE PARAMETERS command
+	   has not been issued to the device then the values of
 	   id[54] to id[56] are vendor specific. */
 	return (id[53] & 0x01) && /* Current translation valid */
 		id[54] &&  /* cylinders in current translation */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 563885cb0995..0ddf16c17b93 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -198,7 +198,7 @@ enum {
 
 	/* host set flags */
 	ATA_HOST_SIMPLEX	= (1 << 0),	/* Host is simplex, one DMA channel per host only */
-	
+
 	/* various lengths of time */
 	ATA_TMOUT_BOOT		= 30 * HZ,	/* heuristic */
 	ATA_TMOUT_BOOT_QUICK	= 7 * HZ,	/* heuristic */
-- 
cgit v1.2.3


From 492dfb489658dfe4a755fa29dd0e34e9c8bd8fb8 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Wed, 30 Aug 2006 15:48:45 -0400
Subject: [SCSI] block: add support for shared tag maps

The current block queue implementation already contains most of the
machinery for shared tag maps.  The only remaining pieces are a way to
allocate and destroy a tag map independently of the queues (so that
the maps can be managed on the life cycle of the overseeing entity)

Acked-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 block/ll_rw_blk.c      | 109 +++++++++++++++++++++++++++++++++++++++----------
 include/linux/blkdev.h |   2 +
 2 files changed, 90 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index ddd9253f9d55..556a3d354eab 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -848,21 +848,18 @@ struct request *blk_queue_find_tag(request_queue_t *q, int tag)
 EXPORT_SYMBOL(blk_queue_find_tag);
 
 /**
- * __blk_queue_free_tags - release tag maintenance info
- * @q:  the request queue for the device
+ * __blk_free_tags - release a given set of tag maintenance info
+ * @bqt:	the tag map to free
  *
- *  Notes:
- *    blk_cleanup_queue() will take care of calling this function, if tagging
- *    has been used. So there's no need to call this directly.
- **/
-static void __blk_queue_free_tags(request_queue_t *q)
+ * Tries to free the specified @bqt@.  Returns true if it was
+ * actually freed and false if there are still references using it
+ */
+static int __blk_free_tags(struct blk_queue_tag *bqt)
 {
-	struct blk_queue_tag *bqt = q->queue_tags;
-
-	if (!bqt)
-		return;
+	int retval;
 
-	if (atomic_dec_and_test(&bqt->refcnt)) {
+	retval = atomic_dec_and_test(&bqt->refcnt);
+	if (retval) {
 		BUG_ON(bqt->busy);
 		BUG_ON(!list_empty(&bqt->busy_list));
 
@@ -873,12 +870,49 @@ static void __blk_queue_free_tags(request_queue_t *q)
 		bqt->tag_map = NULL;
 
 		kfree(bqt);
+
 	}
 
+	return retval;
+}
+
+/**
+ * __blk_queue_free_tags - release tag maintenance info
+ * @q:  the request queue for the device
+ *
+ *  Notes:
+ *    blk_cleanup_queue() will take care of calling this function, if tagging
+ *    has been used. So there's no need to call this directly.
+ **/
+static void __blk_queue_free_tags(request_queue_t *q)
+{
+	struct blk_queue_tag *bqt = q->queue_tags;
+
+	if (!bqt)
+		return;
+
+	__blk_free_tags(bqt);
+
 	q->queue_tags = NULL;
 	q->queue_flags &= ~(1 << QUEUE_FLAG_QUEUED);
 }
 
+
+/**
+ * blk_free_tags - release a given set of tag maintenance info
+ * @bqt:	the tag map to free
+ *
+ * For externally managed @bqt@ frees the map.  Callers of this
+ * function must guarantee to have released all the queues that
+ * might have been using this tag map.
+ */
+void blk_free_tags(struct blk_queue_tag *bqt)
+{
+	if (unlikely(!__blk_free_tags(bqt)))
+		BUG();
+}
+EXPORT_SYMBOL(blk_free_tags);
+
 /**
  * blk_queue_free_tags - release tag maintenance info
  * @q:  the request queue for the device
@@ -901,7 +935,7 @@ init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth)
 	unsigned long *tag_map;
 	int nr_ulongs;
 
-	if (depth > q->nr_requests * 2) {
+	if (q && depth > q->nr_requests * 2) {
 		depth = q->nr_requests * 2;
 		printk(KERN_ERR "%s: adjusted depth to %d\n",
 				__FUNCTION__, depth);
@@ -927,6 +961,38 @@ fail:
 	return -ENOMEM;
 }
 
+static struct blk_queue_tag *__blk_queue_init_tags(struct request_queue *q,
+						   int depth)
+{
+	struct blk_queue_tag *tags;
+
+	tags = kmalloc(sizeof(struct blk_queue_tag), GFP_ATOMIC);
+	if (!tags)
+		goto fail;
+
+	if (init_tag_map(q, tags, depth))
+		goto fail;
+
+	INIT_LIST_HEAD(&tags->busy_list);
+	tags->busy = 0;
+	atomic_set(&tags->refcnt, 1);
+	return tags;
+fail:
+	kfree(tags);
+	return NULL;
+}
+
+/**
+ * blk_init_tags - initialize the tag info for an external tag map
+ * @depth:	the maximum queue depth supported
+ * @tags: the tag to use
+ **/
+struct blk_queue_tag *blk_init_tags(int depth)
+{
+	return __blk_queue_init_tags(NULL, depth);
+}
+EXPORT_SYMBOL(blk_init_tags);
+
 /**
  * blk_queue_init_tags - initialize the queue tag info
  * @q:  the request queue for the device
@@ -941,16 +1007,10 @@ int blk_queue_init_tags(request_queue_t *q, int depth,
 	BUG_ON(tags && q->queue_tags && tags != q->queue_tags);
 
 	if (!tags && !q->queue_tags) {
-		tags = kmalloc(sizeof(struct blk_queue_tag), GFP_ATOMIC);
-		if (!tags)
-			goto fail;
+		tags = __blk_queue_init_tags(q, depth);
 
-		if (init_tag_map(q, tags, depth))
+		if (!tags)
 			goto fail;
-
-		INIT_LIST_HEAD(&tags->busy_list);
-		tags->busy = 0;
-		atomic_set(&tags->refcnt, 1);
 	} else if (q->queue_tags) {
 		if ((rc = blk_queue_resize_tags(q, depth)))
 			return rc;
@@ -1001,6 +1061,13 @@ int blk_queue_resize_tags(request_queue_t *q, int new_depth)
 		return 0;
 	}
 
+	/*
+	 * Currently cannot replace a shared tag map with a new
+	 * one, so error out if this is the case
+	 */
+	if (atomic_read(&bqt->refcnt) != 1)
+		return -EBUSY;
+
 	/*
 	 * save the old state info, so we can copy it back
 	 */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index aafe82788b4e..427b0d61be6c 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -746,6 +746,8 @@ extern void blk_queue_free_tags(request_queue_t *);
 extern int blk_queue_resize_tags(request_queue_t *, int);
 extern void blk_queue_invalidate_tags(request_queue_t *);
 extern long blk_congestion_wait(int rw, long timeout);
+extern struct blk_queue_tag *blk_init_tags(int);
+extern void blk_free_tags(struct blk_queue_tag *);
 
 extern void blk_rq_bio_prep(request_queue_t *, struct request *, struct bio *);
 extern int blkdev_issue_flush(struct block_device *, sector_t *);
-- 
cgit v1.2.3


From 86e33a296c2c9ed6eece0bfff4ac776f42040504 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Wed, 30 Aug 2006 09:45:51 -0400
Subject: [SCSI] add shared tag map helpers

This patch adds support for sharing tag maps at the host level
(i.e. either every queue [LUN] has its own tag map or there's a single
one for the entire host).  This formulation is primarily intended to
help single issue queue hardware, like the aic7xxx

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/hosts.c     |  3 +++
 include/scsi/scsi_host.h |  7 +++++++
 include/scsi/scsi_tcq.h  | 14 +++++++++++++-
 3 files changed, 23 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index f244d4f6597a..68ef1636678d 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -265,6 +265,9 @@ static void scsi_host_dev_release(struct device *dev)
 		destroy_workqueue(shost->work_q);
 
 	scsi_destroy_command_freelist(shost);
+	if (shost->bqt)
+		blk_free_tags(shost->bqt);
+
 	kfree(shost->shost_data);
 
 	if (parent)
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index b3dd90f3e858..39c6f8cc20c3 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -16,6 +16,7 @@ struct scsi_target;
 struct Scsi_Host;
 struct scsi_host_cmd_pool;
 struct scsi_transport_template;
+struct blk_queue_tags;
 
 
 /*
@@ -465,6 +466,12 @@ struct Scsi_Host {
 	struct scsi_host_template *hostt;
 	struct scsi_transport_template *transportt;
 
+	/*
+	 * area to keep a shared tag map (if needed, will be
+	 * NULL if not)
+	 */
+	struct blk_queue_tag	*bqt;
+
 	/*
 	 * The following two fields are protected with host_lock;
 	 * however, eh routines can safely access during eh processing
diff --git a/include/scsi/scsi_tcq.h b/include/scsi/scsi_tcq.h
index e47e36a4ef49..4eea254b1ce9 100644
--- a/include/scsi/scsi_tcq.h
+++ b/include/scsi/scsi_tcq.h
@@ -4,6 +4,7 @@
 #include <linux/blkdev.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_device.h>
+#include <scsi/scsi_host.h>
 
 
 #define MSG_SIMPLE_TAG	0x20
@@ -66,7 +67,8 @@ static inline void scsi_activate_tcq(struct scsi_device *sdev, int depth)
 		return;
 
 	if (!blk_queue_tagged(sdev->request_queue))
-		blk_queue_init_tags(sdev->request_queue, depth, NULL);
+		blk_queue_init_tags(sdev->request_queue, depth,
+				    sdev->host->bqt);
 
 	scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth);
 }
@@ -131,4 +133,14 @@ static inline struct scsi_cmnd *scsi_find_tag(struct scsi_device *sdev, int tag)
 	return sdev->current_cmnd;
 }
 
+/**
+ * scsi_init_shared_tag_map - create a shared tag map
+ * @shost:	the host to share the tag map among all devices
+ * @depth:	the total depth of the map
+ */
+static inline void scsi_init_shared_tag_map(struct Scsi_Host *shost, int depth)
+{
+	shost->bqt = blk_init_tags(depth);
+}
+
 #endif /* _SCSI_SCSI_TCQ_H */
-- 
cgit v1.2.3


From ffd0436ed2e5a741c8d30062b489b989acf0a526 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Thu, 31 Aug 2006 18:09:24 -0400
Subject: [SCSI] libiscsi, iscsi_tcp, iscsi_iser: check that burst lengths are
 valid.

iSCSI RFC states that the first burst length must be smaller than the
max burst length. We currently assume targets will be good, but that may
not be the case, so this patch adds a check.

This patch also moves the unsol data out offset to the lib so the LLDs
do not have to track it.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/infiniband/ulp/iser/iscsi_iser.c | 18 ++++--------------
 drivers/infiniband/ulp/iser/iscsi_iser.h |  1 -
 drivers/scsi/iscsi_tcp.c                 | 30 +++++++-----------------------
 drivers/scsi/iscsi_tcp.h                 |  1 -
 drivers/scsi/libiscsi.c                  | 25 +++++++++++++++++--------
 include/scsi/libiscsi.h                  |  5 +++--
 6 files changed, 31 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index 1437d7ee3b19..101e407eaa43 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -141,18 +141,11 @@ iscsi_iser_cmd_init(struct iscsi_cmd_task *ctask)
 
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
 		BUG_ON(ctask->total_length == 0);
-		/* bytes to be sent via RDMA operations */
-		iser_ctask->rdma_data_count = ctask->total_length -
-					 ctask->imm_count -
-					 ctask->unsol_count;
 
-		debug_scsi("cmd [itt %x total %d imm %d unsol_data %d "
-			   "rdma_data %d]\n",
+		debug_scsi("cmd [itt %x total %d imm %d unsol_data %d\n",
 			   ctask->itt, ctask->total_length, ctask->imm_count,
-			   ctask->unsol_count, iser_ctask->rdma_data_count);
-	} else
-		/* bytes to be sent via RDMA operations */
-		iser_ctask->rdma_data_count = ctask->total_length;
+			   ctask->unsol_count);
+	}
 
 	iser_ctask_rdma_init(iser_ctask);
 }
@@ -196,13 +189,10 @@ iscsi_iser_ctask_xmit_unsol_data(struct iscsi_conn *conn,
 {
 	struct iscsi_data  hdr;
 	int error = 0;
-	struct iscsi_iser_cmd_task *iser_ctask = ctask->dd_data;
 
 	/* Send data-out PDUs while there's still unsolicited data to send */
 	while (ctask->unsol_count > 0) {
-		iscsi_prep_unsolicit_data_pdu(ctask, &hdr,
-					      iser_ctask->rdma_data_count);
-
+		iscsi_prep_unsolicit_data_pdu(ctask, &hdr);
 		debug_scsi("Sending data-out: itt 0x%x, data count %d\n",
 			   hdr.itt, ctask->data_count);
 
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h
index 3350ba690cfe..7c3d0c96d889 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.h
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.h
@@ -257,7 +257,6 @@ struct iscsi_iser_conn {
 struct iscsi_iser_cmd_task {
 	struct iser_desc             desc;
 	struct iscsi_iser_conn	     *iser_conn;
-	int			     rdma_data_count;/* RDMA bytes           */
 	enum iser_task_status 	     status;
 	int                          command_sent;  /* set if command  sent  */
 	int                          dir[ISER_DIRS_NUM];      /* set if dir use*/
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 058f094f945a..a97a3a4e99eb 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -1264,19 +1264,6 @@ iscsi_solicit_data_cont(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 			    r2t->data_count);
 }
 
-static void
-iscsi_unsolicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
-{
-	struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data;
-	struct iscsi_data_task *dtask;
-
-	dtask = tcp_ctask->dtask = &tcp_ctask->unsol_dtask;
-	iscsi_prep_unsolicit_data_pdu(ctask, &dtask->hdr,
-				      tcp_ctask->r2t_data_count);
-	iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)&dtask->hdr,
-			   sizeof(struct iscsi_hdr));
-}
-
 /**
  * iscsi_tcp_cmd_init - Initialize iSCSI SCSI_READ or SCSI_WRITE commands
  * @conn: iscsi connection
@@ -1326,14 +1313,11 @@ iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask)
 		if (ctask->unsol_count)
 			tcp_ctask->xmstate |= XMSTATE_UNS_HDR |
 						XMSTATE_UNS_INIT;
-		tcp_ctask->r2t_data_count = ctask->total_length -
-				    ctask->imm_count -
-				    ctask->unsol_count;
 
-		debug_scsi("cmd [itt 0x%x total %d imm %d imm_data %d "
-			   "r2t_data %d]\n",
+		debug_scsi("cmd [itt 0x%x total %d imm_data %d "
+			   "unsol count %d, unsol offset %d]\n",
 			   ctask->itt, ctask->total_length, ctask->imm_count,
-			   ctask->unsol_count, tcp_ctask->r2t_data_count);
+			   ctask->unsol_count, ctask->unsol_offset);
 	} else
 		tcp_ctask->xmstate = XMSTATE_R_HDR;
 
@@ -1531,8 +1515,10 @@ handle_xmstate_uns_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 
 	tcp_ctask->xmstate |= XMSTATE_UNS_DATA;
 	if (tcp_ctask->xmstate & XMSTATE_UNS_INIT) {
-		iscsi_unsolicit_data_init(conn, ctask);
-		dtask = tcp_ctask->dtask;
+		dtask = tcp_ctask->dtask = &tcp_ctask->unsol_dtask;
+		iscsi_prep_unsolicit_data_pdu(ctask, &dtask->hdr);
+		iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)&dtask->hdr,
+				   sizeof(struct iscsi_hdr));
 		if (conn->hdrdgst_en)
 			iscsi_hdr_digest(conn, &tcp_ctask->headbuf,
 					(u8*)dtask->hdrext);
@@ -1720,7 +1706,6 @@ data_out_done:
 	 * Done with this R2T. Check if there are more
 	 * outstanding R2Ts ready to be processed.
 	 */
-	BUG_ON(tcp_ctask->r2t_data_count - r2t->data_length < 0);
 	if (conn->datadgst_en) {
 		rc = iscsi_digest_final_send(conn, ctask, &dtask->digestbuf,
 					    &dtask->digest, 1);
@@ -1732,7 +1717,6 @@ data_out_done:
 		debug_tcp("r2t done dout digest 0x%x\n", dtask->digest);
 	}
 
-	tcp_ctask->r2t_data_count -= r2t->data_length;
 	tcp_ctask->r2t = NULL;
 	spin_lock_bh(&session->lock);
 	__kfifo_put(tcp_ctask->r2tpool.queue, (void*)&r2t, sizeof(void*));
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index 6a4ee704e46e..aace8f70dfd7 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -157,7 +157,6 @@ struct iscsi_tcp_cmd_task {
 	struct scatterlist	*bad_sg;		/* assert statement */
 	int			sg_count;		/* SG's to process  */
 	uint32_t		exp_r2tsn;
-	int			r2t_data_count;		/* R2T Data-Out bytes */
 	int			data_offset;
 	struct iscsi_r2t_info	*r2t;			/* in progress R2T    */
 	struct iscsi_queue	r2tpool;
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 5884cd26d53a..a7c6e70f4ef8 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -68,8 +68,7 @@ iscsi_check_assign_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr)
 EXPORT_SYMBOL_GPL(iscsi_check_assign_cmdsn);
 
 void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *ctask,
-				   struct iscsi_data *hdr,
-				   int transport_data_cnt)
+				   struct iscsi_data *hdr)
 {
 	struct iscsi_conn *conn = ctask->conn;
 
@@ -82,14 +81,12 @@ void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *ctask,
 
 	hdr->itt = ctask->hdr->itt;
 	hdr->exp_statsn = cpu_to_be32(conn->exp_statsn);
-
-	hdr->offset = cpu_to_be32(ctask->total_length -
-				  transport_data_cnt -
-				  ctask->unsol_count);
+	hdr->offset = cpu_to_be32(ctask->unsol_offset);
 
 	if (ctask->unsol_count > conn->max_xmit_dlength) {
 		hton24(hdr->dlength, conn->max_xmit_dlength);
 		ctask->data_count = conn->max_xmit_dlength;
+		ctask->unsol_offset += ctask->data_count;
 		hdr->flags = 0;
 	} else {
 		hton24(hdr->dlength, ctask->unsol_count);
@@ -125,6 +122,7 @@ static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
         memcpy(hdr->cdb, sc->cmnd, sc->cmd_len);
         memset(&hdr->cdb[sc->cmd_len], 0, MAX_COMMAND_SIZE - sc->cmd_len);
 
+	ctask->data_count = 0;
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
 		hdr->flags |= ISCSI_FLAG_CMD_WRITE;
 		/*
@@ -143,6 +141,7 @@ static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 		 */
 		ctask->imm_count = 0;
 		ctask->unsol_count = 0;
+		ctask->unsol_offset = 0;
 		ctask->unsol_datasn = 0;
 
 		if (session->imm_data_en) {
@@ -156,9 +155,12 @@ static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask)
 		} else
 			zero_data(ctask->hdr->dlength);
 
-		if (!session->initial_r2t_en)
+		if (!session->initial_r2t_en) {
 			ctask->unsol_count = min(session->first_burst,
 				ctask->total_length) - ctask->imm_count;
+			ctask->unsol_offset = ctask->imm_count;
+		}
+
 		if (!ctask->unsol_count)
 			/* No unsolicit Data-Out's */
 			ctask->hdr->flags |= ISCSI_FLAG_CMD_FINAL;
@@ -1520,11 +1522,18 @@ int iscsi_conn_start(struct iscsi_cls_conn *cls_conn)
 	struct iscsi_conn *conn = cls_conn->dd_data;
 	struct iscsi_session *session = conn->session;
 
-	if (session == NULL) {
+	if (!session) {
 		printk(KERN_ERR "iscsi: can't start unbound connection\n");
 		return -EPERM;
 	}
 
+	if (session->first_burst > session->max_burst) {
+		printk("iscsi: invalid burst lengths: "
+		       "first_burst %d max_burst %d\n",
+		       session->first_burst, session->max_burst);
+		return -EINVAL;
+	}
+
 	spin_lock_bh(&session->lock);
 	conn->c_stage = ISCSI_CONN_STARTED;
 	session->state = ISCSI_STATE_LOGGED_IN;
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index 41904f611d12..4900650bd081 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -102,6 +102,8 @@ struct iscsi_cmd_task {
 	uint32_t		unsol_datasn;
 	int			imm_count;	/* imm-data (bytes)   */
 	int			unsol_count;	/* unsolicited (bytes)*/
+	/* offset in unsolicited stream (bytes); */
+	int			unsol_offset;
 	int			data_count;	/* remaining Data-Out */
 	struct scsi_cmnd	*sc;		/* associated SCSI cmd*/
 	int			total_length;
@@ -290,8 +292,7 @@ extern int iscsi_conn_get_param(struct iscsi_cls_conn *cls_conn,
 extern int iscsi_check_assign_cmdsn(struct iscsi_session *,
 				    struct iscsi_nopin *);
 extern void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *,
-					struct iscsi_data *hdr,
-					int transport_data_cnt);
+					struct iscsi_data *hdr);
 extern int iscsi_conn_send_pdu(struct iscsi_cls_conn *, struct iscsi_hdr *,
 				char *, uint32_t);
 extern int iscsi_complete_pdu(struct iscsi_conn *, struct iscsi_hdr *,
-- 
cgit v1.2.3


From 60ecebf5a10e42f5e2d6e07eb9e24bdee8500b81 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Thu, 31 Aug 2006 18:09:25 -0400
Subject: [SCSI] add refcouting around ctask usage in main IO patch

It is possible that a ctask could be completing and getting
cleaned up at the same time, we are finishing up the last
data transfer. This could then result in the data transfer
code using stale or invalid values. This patch adds a refcount
to the ctask. When the count goes to zero then we know the
transmit thread and recv thread or softirq are not touching
it and we can safely release it.

The eh should not need to grab a reference because it only cleans
up a task if it has both the xmit mutex and recv lock (or recv
side suspended).

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/libiscsi.c | 59 +++++++++++++++++++++++++++++++++++++++++--------
 include/scsi/libiscsi.h |  1 +
 2 files changed, 51 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index a7c6e70f4ef8..9584cbc082fe 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -179,16 +179,15 @@ EXPORT_SYMBOL_GPL(iscsi_prep_scsi_cmd_pdu);
 
 /**
  * iscsi_complete_command - return command back to scsi-ml
- * @session: iscsi session
  * @ctask: iscsi cmd task
  *
  * Must be called with session lock.
  * This function returns the scsi command to scsi-ml and returns
  * the cmd task to the pool of available cmd tasks.
  */
-static void iscsi_complete_command(struct iscsi_session *session,
-				   struct iscsi_cmd_task *ctask)
+static void iscsi_complete_command(struct iscsi_cmd_task *ctask)
 {
+	struct iscsi_session *session = ctask->conn->session;
 	struct scsi_cmnd *sc = ctask->sc;
 
 	ctask->state = ISCSI_TASK_COMPLETED;
@@ -198,6 +197,35 @@ static void iscsi_complete_command(struct iscsi_session *session,
 	sc->scsi_done(sc);
 }
 
+static void __iscsi_get_ctask(struct iscsi_cmd_task *ctask)
+{
+	atomic_inc(&ctask->refcount);
+}
+
+static void iscsi_get_ctask(struct iscsi_cmd_task *ctask)
+{
+	spin_lock_bh(&ctask->conn->session->lock);
+	__iscsi_get_ctask(ctask);
+	spin_unlock_bh(&ctask->conn->session->lock);
+}
+
+static void __iscsi_put_ctask(struct iscsi_cmd_task *ctask)
+{
+	struct iscsi_conn *conn = ctask->conn;
+
+	if (atomic_dec_and_test(&ctask->refcount)) {
+		conn->session->tt->cleanup_cmd_task(conn, ctask);
+		iscsi_complete_command(ctask);
+	}
+}
+
+static void iscsi_put_ctask(struct iscsi_cmd_task *ctask)
+{
+	spin_lock_bh(&ctask->conn->session->lock);
+	__iscsi_put_ctask(ctask);
+	spin_unlock_bh(&ctask->conn->session->lock);
+}
+
 /**
  * iscsi_cmd_rsp - SCSI Command Response processing
  * @conn: iscsi connection
@@ -274,7 +302,7 @@ out:
 		   (long)sc, sc->result, ctask->itt);
 	conn->scsirsp_pdus_cnt++;
 
-	iscsi_complete_command(conn->session, ctask);
+	__iscsi_put_ctask(ctask);
 	return rc;
 }
 
@@ -338,7 +366,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 			BUG_ON((void*)ctask != ctask->sc->SCp.ptr);
 			if (hdr->flags & ISCSI_FLAG_DATA_STATUS) {
 				conn->scsirsp_pdus_cnt++;
-				iscsi_complete_command(session, ctask);
+				__iscsi_put_ctask(ctask);
 			}
 			break;
 		case ISCSI_OP_R2T:
@@ -563,7 +591,9 @@ static int iscsi_data_xmit(struct iscsi_conn *conn)
 	BUG_ON(conn->ctask && conn->mtask);
 
 	if (conn->ctask) {
+		iscsi_get_ctask(conn->ctask);
 		rc = tt->xmit_cmd_task(conn, conn->ctask);
+		iscsi_put_ctask(conn->ctask);
 		if (rc)
 			goto again;
 		/* done with this in-progress ctask */
@@ -604,12 +634,19 @@ static int iscsi_data_xmit(struct iscsi_conn *conn)
 					 struct iscsi_cmd_task, running);
 		conn->ctask->state = ISCSI_TASK_RUNNING;
 		list_move_tail(conn->xmitqueue.next, &conn->run_list);
+		__iscsi_get_ctask(conn->ctask);
 		spin_unlock_bh(&conn->session->lock);
 
 		rc = tt->xmit_cmd_task(conn, conn->ctask);
 		if (rc)
 			goto again;
+
 		spin_lock_bh(&conn->session->lock);
+		__iscsi_put_ctask(conn->ctask);
+		if (rc) {
+			spin_unlock_bh(&conn->session->lock);
+			goto again;
+		}
 	}
 	spin_unlock_bh(&conn->session->lock);
 	/* done with this ctask */
@@ -659,6 +696,7 @@ enum {
 	FAILURE_SESSION_FAILED,
 	FAILURE_SESSION_FREED,
 	FAILURE_WINDOW_CLOSED,
+	FAILURE_OOM,
 	FAILURE_SESSION_TERMINATE,
 	FAILURE_SESSION_IN_RECOVERY,
 	FAILURE_SESSION_RECOVERY_TIMEOUT,
@@ -717,10 +755,15 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *))
 
 	conn = session->leadconn;
 
-	__kfifo_get(session->cmdpool.queue, (void*)&ctask, sizeof(void*));
+	if (!__kfifo_get(session->cmdpool.queue, (void*)&ctask,
+			 sizeof(void*))) {
+		reason = FAILURE_OOM;
+		goto reject;
+	}
 	sc->SCp.phase = session->age;
 	sc->SCp.ptr = (char *)ctask;
 
+	atomic_set(&ctask->refcount, 1);
 	ctask->state = ISCSI_TASK_PENDING;
 	ctask->mtask = NULL;
 	ctask->conn = conn;
@@ -1057,13 +1100,11 @@ static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask,
 	sc = ctask->sc;
 	if (!sc)
 		return;
-
-	conn->session->tt->cleanup_cmd_task(conn, ctask);
 	iscsi_ctask_mtask_cleanup(ctask);
 
 	sc->result = err;
 	sc->resid = sc->request_bufflen;
-	iscsi_complete_command(conn->session, ctask);
+	__iscsi_put_ctask(ctask);
 }
 
 int iscsi_eh_abort(struct scsi_cmnd *sc)
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index 4900650bd081..401192e56e50 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -112,6 +112,7 @@ struct iscsi_cmd_task {
 
 	/* state set/tested under session->lock */
 	int			state;
+	atomic_t		refcount;
 	struct list_head	running;	/* running cmd list */
 	void			*dd_data;	/* driver/transport data */
 };
-- 
cgit v1.2.3


From deb81d80ba27da8dfabc29ccb5977db8f4942a0a Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Fri, 1 Sep 2006 09:28:48 -0400
Subject: [SCSI] add failure return to scsi_init_shared_tag_map()

And use it in the stex driver.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/stex.c     | 5 ++---
 include/scsi/scsi_tcq.h | 3 ++-
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/stex.c b/drivers/scsi/stex.c
index 15fb99f224ee..3cf3106a29b8 100644
--- a/drivers/scsi/stex.c
+++ b/drivers/scsi/stex.c
@@ -1108,9 +1108,8 @@ stex_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto out_free_irq;
 
-	scsi_init_shared_tag_map(host, ST_CAN_QUEUE);
-	if (host->bqt == NULL) {
-		err = -ENOMEM;
+	err = scsi_init_shared_tag_map(host, ST_CAN_QUEUE);
+	if (err) {
 		printk(KERN_ERR DRV_NAME "(%s): init shared queue failed\n",
 			pci_name(pdev));
 		goto out_free_irq;
diff --git a/include/scsi/scsi_tcq.h b/include/scsi/scsi_tcq.h
index 4eea254b1ce9..d04d05adfa9b 100644
--- a/include/scsi/scsi_tcq.h
+++ b/include/scsi/scsi_tcq.h
@@ -138,9 +138,10 @@ static inline struct scsi_cmnd *scsi_find_tag(struct scsi_device *sdev, int tag)
  * @shost:	the host to share the tag map among all devices
  * @depth:	the total depth of the map
  */
-static inline void scsi_init_shared_tag_map(struct Scsi_Host *shost, int depth)
+static inline int scsi_init_shared_tag_map(struct Scsi_Host *shost, int depth)
 {
 	shost->bqt = blk_init_tags(depth);
+	return shost->bqt ? 0 : -ENOMEM;
 }
 
 #endif /* _SCSI_SCSI_TCQ_H */
-- 
cgit v1.2.3


From 84314fd4740ad73550c76dee4a9578979d84af48 Mon Sep 17 00:00:00 2001
From: James Smart <James.Smart@Emulex.Com>
Date: Fri, 18 Aug 2006 17:30:09 -0400
Subject: [SCSI] SCSI and FC Transport: add netlink support for posting of
 transport events

This patch formally adds support for the posting of FC events via netlink.
It is a followup to the original RFC at:
  http://marc.theaimsgroup.com/?l=linux-scsi&m=114530667923464&w=2
and the initial posting at:
  http://marc.theaimsgroup.com/?l=linux-scsi&m=115507374832500&w=2

The patch has been updated to optimize the send path, per the discussions
in the initial posting.

Per discussions at the Storage Summit and at OLS, we are to use netlink for
async events from transports. Also per discussions, to avoid a netlink
protocol per transport, I've create a single NETLINK_SCSITRANSPORT protocol,
which can then be used by all transports.

This patch:
- Creates new files scsi_netlink.c and scsi_netlink.h, which contains the
  single and shared definitions for the SCSI Transport. It is tied into the
  base SCSI subsystem intialization.
  Contains a single interface routine, scsi_send_transport_event(), for a
  transport to send an event (via multicast to a protocol specific group).
- Creates a new scsi_netlink_fc.h file, which contains the FC netlink event
  messages
- Adds 3 new routines to the fc transport:
   fc_get_event_number() -  to get a FC event #
   fc_host_post_event()  -  to send a simple FC event (32 bits of data)
   fc_host_post_vendor_event() - to send a Vendor unique event, with
                                 arbitrary amounts of data.

   Note: the separation of event number allows for a LLD to send a standard
     event, followed by vendor-specific data for the event.

Note: This patch assumes 2 prior fc transport patches have been installed:
   http://marc.theaimsgroup.com/?l=linux-scsi&m=115555807316329&w=2
   http://marc.theaimsgroup.com/?l=linux-scsi&m=115581614930261&w=2

   Sorry - next time I'll do something like making these individual
   patches of the same posting when I know they'll be posted closely
   together.

Signed-off-by: James Smart <James.Smart@emulex.com>

Tidy up configuration not to make SCSI always select NET

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/Kconfig             |   6 ++
 drivers/scsi/Makefile            |   1 +
 drivers/scsi/scsi.c              |   3 +
 drivers/scsi/scsi_netlink.c      | 199 ++++++++++++++++++++++++++++++++++++++
 drivers/scsi/scsi_priv.h         |  11 +++
 drivers/scsi/scsi_transport_fc.c | 200 ++++++++++++++++++++++++++++++++++++++-
 include/linux/netlink.h          |   2 +
 include/scsi/scsi_netlink.h      |  86 +++++++++++++++++
 include/scsi/scsi_netlink_fc.h   |  71 ++++++++++++++
 include/scsi/scsi_transport_fc.h |  34 +++++++
 10 files changed, 612 insertions(+), 1 deletion(-)
 create mode 100644 drivers/scsi/scsi_netlink.c
 create mode 100644 include/scsi/scsi_netlink.h
 create mode 100644 include/scsi/scsi_netlink_fc.h

(limited to 'include')

diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index c8c606589ea6..4d1998d23f0f 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -27,6 +27,11 @@ config SCSI
 	  However, do not compile this as a module if your root file system
 	  (the one containing the directory /) is located on a SCSI device.
 
+config SCSI_NETLINK
+	tristate
+	default	n
+	select NET
+
 config SCSI_PROC_FS
 	bool "legacy /proc/scsi/ support"
 	depends on SCSI && PROC_FS
@@ -222,6 +227,7 @@ config SCSI_SPI_ATTRS
 config SCSI_FC_ATTRS
 	tristate "FiberChannel Transport Attributes"
 	depends on SCSI
+	select SCSI_NETLINK
 	help
 	  If you wish to export transport-specific information about
 	  each attached FiberChannel device to sysfs, say Y.
diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index fd9aeb1ba07f..8fc2c594b537 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -159,6 +159,7 @@ scsi_mod-y			+= scsi.o hosts.o scsi_ioctl.o constants.o \
 				   scsicam.o scsi_error.o scsi_lib.o \
 				   scsi_scan.o scsi_sysfs.o \
 				   scsi_devinfo.o
+scsi_mod-$(CONFIG_SCSI_NETLINK)	+= scsi_netlink.o
 scsi_mod-$(CONFIG_SYSCTL)	+= scsi_sysctl.o
 scsi_mod-$(CONFIG_SCSI_PROC_FS)	+= scsi_proc.o
 
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 37843927e47f..eedfd059b82b 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -1118,6 +1118,8 @@ static int __init init_scsi(void)
 	for_each_possible_cpu(i)
 		INIT_LIST_HEAD(&per_cpu(scsi_done_q, i));
 
+	scsi_netlink_init();
+
 	printk(KERN_NOTICE "SCSI subsystem initialized\n");
 	return 0;
 
@@ -1138,6 +1140,7 @@ cleanup_queue:
 
 static void __exit exit_scsi(void)
 {
+	scsi_netlink_exit();
 	scsi_sysfs_unregister();
 	scsi_exit_sysctl();
 	scsi_exit_hosts();
diff --git a/drivers/scsi/scsi_netlink.c b/drivers/scsi/scsi_netlink.c
new file mode 100644
index 000000000000..1b59b27e887f
--- /dev/null
+++ b/drivers/scsi/scsi_netlink.c
@@ -0,0 +1,199 @@
+/*
+ *  scsi_netlink.c  - SCSI Transport Netlink Interface
+ *
+ *  Copyright (C) 2006   James Smart, Emulex Corporation
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+#include <linux/time.h>
+#include <linux/jiffies.h>
+#include <linux/security.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+
+#include <scsi/scsi_netlink.h>
+#include "scsi_priv.h"
+
+struct sock *scsi_nl_sock = NULL;
+EXPORT_SYMBOL_GPL(scsi_nl_sock);
+
+
+/**
+ * scsi_nl_rcv_msg -
+ *    Receive message handler. Extracts message from a receive buffer.
+ *    Validates message header and calls appropriate transport message handler
+ *
+ * @skb:		socket receive buffer
+ *
+ **/
+static void
+scsi_nl_rcv_msg(struct sk_buff *skb)
+{
+	struct nlmsghdr *nlh;
+	struct scsi_nl_hdr *hdr;
+	uint32_t rlen;
+	int err;
+
+	while (skb->len >= NLMSG_SPACE(0)) {
+		err = 0;
+
+		nlh = (struct nlmsghdr *) skb->data;
+		if ((nlh->nlmsg_len < (sizeof(*nlh) + sizeof(*hdr))) ||
+		    (skb->len < nlh->nlmsg_len)) {
+			printk(KERN_WARNING "%s: discarding partial skb\n",
+				 __FUNCTION__);
+			return;
+		}
+
+		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
+		if (rlen > skb->len)
+			rlen = skb->len;
+
+		if (nlh->nlmsg_type != SCSI_TRANSPORT_MSG) {
+			err = -EBADMSG;
+			goto next_msg;
+		}
+
+		hdr = NLMSG_DATA(nlh);
+		if ((hdr->version != SCSI_NL_VERSION) ||
+		    (hdr->magic != SCSI_NL_MAGIC)) {
+			err = -EPROTOTYPE;
+			goto next_msg;
+		}
+
+		if (security_netlink_recv(skb, CAP_SYS_ADMIN)) {
+			err = -EPERM;
+			goto next_msg;
+		}
+
+		if (nlh->nlmsg_len < (sizeof(*nlh) + hdr->msglen)) {
+			printk(KERN_WARNING "%s: discarding partial message\n",
+				 __FUNCTION__);
+			return;
+		}
+
+		/*
+		 * We currently don't support anyone sending us a message
+		 */
+
+next_msg:
+		if ((err) || (nlh->nlmsg_flags & NLM_F_ACK))
+			netlink_ack(skb, nlh, err);
+
+		skb_pull(skb, rlen);
+	}
+}
+
+
+/**
+ * scsi_nl_rcv_msg -
+ *    Receive handler for a socket. Extracts a received message buffer from
+ *    the socket, and starts message processing.
+ *
+ * @sk:		socket
+ * @len:	unused
+ *
+ **/
+static void
+scsi_nl_rcv(struct sock *sk, int len)
+{
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&sk->sk_receive_queue))) {
+		scsi_nl_rcv_msg(skb);
+		kfree_skb(skb);
+	}
+}
+
+
+/**
+ * scsi_nl_rcv_event -
+ *    Event handler for a netlink socket.
+ *
+ * @this:		event notifier block
+ * @event:		event type
+ * @ptr:		event payload
+ *
+ **/
+static int
+scsi_nl_rcv_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+	struct netlink_notify *n = ptr;
+
+	if (n->protocol != NETLINK_SCSITRANSPORT)
+		return NOTIFY_DONE;
+
+	/*
+	 * Currently, we are not tracking PID's, etc. There is nothing
+	 * to handle.
+	 */
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block scsi_netlink_notifier = {
+	.notifier_call  = scsi_nl_rcv_event,
+};
+
+
+/**
+ * scsi_netlink_init -
+ *    Called by SCSI subsystem to intialize the SCSI transport netlink
+ *    interface
+ *
+ **/
+void
+scsi_netlink_init(void)
+{
+	int error;
+
+	error = netlink_register_notifier(&scsi_netlink_notifier);
+	if (error) {
+		printk(KERN_ERR "%s: register of event handler failed - %d\n",
+				__FUNCTION__, error);
+		return;
+	}
+
+	scsi_nl_sock = netlink_kernel_create(NETLINK_SCSITRANSPORT,
+				SCSI_NL_GRP_CNT, scsi_nl_rcv, THIS_MODULE);
+	if (!scsi_nl_sock) {
+		printk(KERN_ERR "%s: register of recieve handler failed\n",
+				__FUNCTION__);
+		netlink_unregister_notifier(&scsi_netlink_notifier);
+	}
+
+	return;
+}
+
+
+/**
+ * scsi_netlink_exit -
+ *    Called by SCSI subsystem to disable the SCSI transport netlink
+ *    interface
+ *
+ **/
+void
+scsi_netlink_exit(void)
+{
+	if (scsi_nl_sock) {
+		sock_release(scsi_nl_sock->sk_socket);
+		netlink_unregister_notifier(&scsi_netlink_notifier);
+	}
+
+	return;
+}
+
+
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index ae24c85aaeea..5d023d44e5e7 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -8,6 +8,7 @@ struct scsi_cmnd;
 struct scsi_device;
 struct scsi_host_template;
 struct Scsi_Host;
+struct scsi_nl_hdr;
 
 
 /*
@@ -110,6 +111,16 @@ extern void __scsi_remove_device(struct scsi_device *);
 
 extern struct bus_type scsi_bus_type;
 
+/* scsi_netlink.c */
+#ifdef CONFIG_SCSI_NETLINK
+extern void scsi_netlink_init(void);
+extern void scsi_netlink_exit(void);
+extern struct sock *scsi_nl_sock;
+#else
+static inline void scsi_netlink_init(void) {}
+static inline void scsi_netlink_exit(void) {}
+#endif
+
 /* 
  * internal scsi timeout functions: for use by mid-layer and transport
  * classes.
diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 79d31ca2b741..05989f130554 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -32,6 +32,9 @@
 #include <scsi/scsi_transport.h>
 #include <scsi/scsi_transport_fc.h>
 #include <scsi/scsi_cmnd.h>
+#include <linux/netlink.h>
+#include <net/netlink.h>
+#include <scsi/scsi_netlink_fc.h>
 #include "scsi_priv.h"
 
 static int fc_queue_work(struct Scsi_Host *, struct work_struct *);
@@ -93,6 +96,29 @@ fc_enum_name_search(port_type, fc_port_type, fc_port_type_names)
 #define FC_PORTTYPE_MAX_NAMELEN		50
 
 
+/* Convert fc_host_event_code values to ascii string name */
+static const struct {
+	enum fc_host_event_code		value;
+	char				*name;
+} fc_host_event_code_names[] = {
+	{ FCH_EVT_LIP,			"lip" },
+	{ FCH_EVT_LINKUP,		"link_up" },
+	{ FCH_EVT_LINKDOWN,		"link_down" },
+	{ FCH_EVT_LIPRESET,		"lip_reset" },
+	{ FCH_EVT_RSCN,			"rscn" },
+	{ FCH_EVT_ADAPTER_CHANGE,	"adapter_chg" },
+	{ FCH_EVT_PORT_UNKNOWN,		"port_unknown" },
+	{ FCH_EVT_PORT_ONLINE,		"port_online" },
+	{ FCH_EVT_PORT_OFFLINE,		"port_offline" },
+	{ FCH_EVT_PORT_FABRIC,		"port_fabric" },
+	{ FCH_EVT_LINK_UNKNOWN,		"link_unknown" },
+	{ FCH_EVT_VENDOR_UNIQUE,	"vendor_unique" },
+};
+fc_enum_name_search(host_event_code, fc_host_event_code,
+		fc_host_event_code_names)
+#define FC_HOST_EVENT_CODE_MAX_NAMELEN	30
+
+
 /* Convert fc_port_state values to ascii string name */
 static struct {
 	enum fc_port_state	value;
@@ -377,10 +403,182 @@ MODULE_PARM_DESC(dev_loss_tmo,
 		 " exceeded, the scsi target is removed. Value should be"
 		 " between 1 and SCSI_DEVICE_BLOCK_MAX_TIMEOUT.");
 
+/**
+ * Netlink Infrastructure
+ **/
+
+static atomic_t fc_event_seq;
+
+/**
+ * fc_get_event_number - Obtain the next sequential FC event number
+ *
+ * Notes:
+ *   We could have inline'd this, but it would have required fc_event_seq to
+ *   be exposed. For now, live with the subroutine call.
+ *   Atomic used to avoid lock/unlock...
+ **/
+u32
+fc_get_event_number(void)
+{
+	return atomic_add_return(1, &fc_event_seq);
+}
+EXPORT_SYMBOL(fc_get_event_number);
+
+
+/**
+ * fc_host_post_event - called to post an even on an fc_host.
+ *
+ * @shost:		host the event occurred on
+ * @event_number:	fc event number obtained from get_fc_event_number()
+ * @event_code:		fc_host event being posted
+ * @event_data:		32bits of data for the event being posted
+ *
+ * Notes:
+ *	This routine assumes no locks are held on entry.
+ **/
+void
+fc_host_post_event(struct Scsi_Host *shost, u32 event_number,
+		enum fc_host_event_code event_code, u32 event_data)
+{
+	struct sk_buff *skb;
+	struct nlmsghdr	*nlh;
+	struct fc_nl_event *event;
+	const char *name;
+	u32 len, skblen;
+	int err;
+
+	if (!scsi_nl_sock) {
+		err = -ENOENT;
+		goto send_fail;
+	}
+
+	len = FC_NL_MSGALIGN(sizeof(*event));
+	skblen = NLMSG_SPACE(len);
+
+	skb = alloc_skb(skblen, GFP_KERNEL);
+	if (!skb) {
+		err = -ENOBUFS;
+		goto send_fail;
+	}
+
+	nlh = nlmsg_put(skb, 0, 0, SCSI_TRANSPORT_MSG,
+				skblen - sizeof(*nlh), 0);
+	if (!nlh) {
+		err = -ENOBUFS;
+		goto send_fail_skb;
+	}
+	event = NLMSG_DATA(nlh);
+
+	INIT_SCSI_NL_HDR(&event->snlh, SCSI_NL_TRANSPORT_FC,
+				FC_NL_ASYNC_EVENT, len);
+	event->seconds = get_seconds();
+	event->vendor_id = 0;
+	event->host_no = shost->host_no;
+	event->event_datalen = sizeof(u32);	/* bytes */
+	event->event_num = event_number;
+	event->event_code = event_code;
+	event->event_data = event_data;
+
+	err = nlmsg_multicast(scsi_nl_sock, skb, 0, SCSI_NL_GRP_FC_EVENTS);
+	if (err && (err != -ESRCH))	/* filter no recipient errors */
+		/* nlmsg_multicast already kfree_skb'd */
+		goto send_fail;
+
+	return;
+
+send_fail_skb:
+	kfree_skb(skb);
+send_fail:
+	name = get_fc_host_event_code_name(event_code);
+	printk(KERN_WARNING
+		"%s: Dropped Event : host %d %s data 0x%08x - err %d\n",
+		__FUNCTION__, shost->host_no,
+		(name) ? name : "<unknown>", event_data, err);
+	return;
+}
+EXPORT_SYMBOL(fc_host_post_event);
+
+
+/**
+ * fc_host_post_vendor_event - called to post a vendor unique event on
+ *                             a fc_host
+ *
+ * @shost:		host the event occurred on
+ * @event_number:	fc event number obtained from get_fc_event_number()
+ * @data_len:		amount, in bytes, of vendor unique data
+ * @data_buf:		pointer to vendor unique data
+ *
+ * Notes:
+ *	This routine assumes no locks are held on entry.
+ **/
+void
+fc_host_post_vendor_event(struct Scsi_Host *shost, u32 event_number,
+		u32 data_len, char * data_buf, u32 vendor_id)
+{
+	struct sk_buff *skb;
+	struct nlmsghdr	*nlh;
+	struct fc_nl_event *event;
+	u32 len, skblen;
+	int err;
+
+	if (!scsi_nl_sock) {
+		err = -ENOENT;
+		goto send_vendor_fail;
+	}
+
+	len = FC_NL_MSGALIGN(sizeof(*event) + data_len);
+	skblen = NLMSG_SPACE(len);
+
+	skb = alloc_skb(skblen, GFP_KERNEL);
+	if (!skb) {
+		err = -ENOBUFS;
+		goto send_vendor_fail;
+	}
+
+	nlh = nlmsg_put(skb, 0, 0, SCSI_TRANSPORT_MSG,
+				skblen - sizeof(*nlh), 0);
+	if (!nlh) {
+		err = -ENOBUFS;
+		goto send_vendor_fail_skb;
+	}
+	event = NLMSG_DATA(nlh);
+
+	INIT_SCSI_NL_HDR(&event->snlh, SCSI_NL_TRANSPORT_FC,
+				FC_NL_ASYNC_EVENT, len);
+	event->seconds = get_seconds();
+	event->vendor_id = vendor_id;
+	event->host_no = shost->host_no;
+	event->event_datalen = data_len;	/* bytes */
+	event->event_num = event_number;
+	event->event_code = FCH_EVT_VENDOR_UNIQUE;
+	memcpy(&event->event_data, data_buf, data_len);
+
+	err = nlmsg_multicast(scsi_nl_sock, skb, 0, SCSI_NL_GRP_FC_EVENTS);
+	if (err && (err != -ESRCH))	/* filter no recipient errors */
+		/* nlmsg_multicast already kfree_skb'd */
+		goto send_vendor_fail;
+
+	return;
+
+send_vendor_fail_skb:
+	kfree_skb(skb);
+send_vendor_fail:
+	printk(KERN_WARNING
+		"%s: Dropped Event : host %d vendor_unique - err %d\n",
+		__FUNCTION__, shost->host_no, err);
+	return;
+}
+EXPORT_SYMBOL(fc_host_post_vendor_event);
+
+
 
 static __init int fc_transport_init(void)
 {
-	int error = transport_class_register(&fc_host_class);
+	int error;
+
+	atomic_set(&fc_event_seq, 0);
+
+	error = transport_class_register(&fc_host_class);
 	if (error)
 		return error;
 	error = transport_class_register(&fc_rport_class);
diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 855b44668caa..66411622e06e 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -21,6 +21,8 @@
 #define NETLINK_DNRTMSG		14	/* DECnet routing messages */
 #define NETLINK_KOBJECT_UEVENT	15	/* Kernel messages to userspace */
 #define NETLINK_GENERIC		16
+/* leave room for NETLINK_DM (DM Events) */
+#define NETLINK_SCSITRANSPORT	18	/* SCSI Transports */
 
 #define MAX_LINKS 32		
 
diff --git a/include/scsi/scsi_netlink.h b/include/scsi/scsi_netlink.h
new file mode 100644
index 000000000000..7a3a20e640c0
--- /dev/null
+++ b/include/scsi/scsi_netlink.h
@@ -0,0 +1,86 @@
+/*
+ *  SCSI Transport Netlink Interface
+ *    Used for the posting of outbound SCSI transport events
+ *
+ *  Copyright (C) 2006   James Smart, Emulex Corporation
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+#ifndef SCSI_NETLINK_H
+#define SCSI_NETLINK_H
+
+/*
+ * This file intended to be included by both kernel and user space
+ */
+
+/* Single Netlink Message type to send all SCSI Transport messages */
+#define SCSI_TRANSPORT_MSG		NLMSG_MIN_TYPE + 1
+
+/* SCSI Transport Broadcast Groups */
+	/* leaving groups 0 and 1 unassigned */
+#define SCSI_NL_GRP_FC_EVENTS		(1<<2)		/* Group 2 */
+#define SCSI_NL_GRP_CNT			3
+
+
+/* SCSI_TRANSPORT_MSG event message header */
+struct scsi_nl_hdr {
+	uint8_t version;
+	uint8_t transport;
+	uint16_t magic;
+	uint16_t msgtype;
+	uint16_t msglen;
+} __attribute__((aligned(sizeof(uint64_t))));
+
+/* scsi_nl_hdr->version value */
+#define SCSI_NL_VERSION				1
+
+/* scsi_nl_hdr->magic value */
+#define SCSI_NL_MAGIC				0xA1B2
+
+/* scsi_nl_hdr->transport value */
+#define SCSI_NL_TRANSPORT			0
+#define SCSI_NL_TRANSPORT_FC			1
+#define SCSI_NL_MAX_TRANSPORTS			2
+
+/* scsi_nl_hdr->msgtype values are defined in each transport */
+
+
+/*
+ * Vendor ID:
+ *   If transports post vendor-unique events, they must pass a well-known
+ *   32-bit vendor identifier. This identifier consists of 8 bits indicating
+ *   the "type" of identifier contained, and 24 bits of id data.
+ *
+ *   Identifiers for each type:
+ *    PCI :  ID data is the 16 bit PCI Registered Vendor ID
+ */
+#define SCSI_NL_VID_ID_MASK			0x00FFFFFF
+#define SCSI_NL_VID_TYPE_MASK			0xFF000000
+#define SCSI_NL_VID_TYPE_PCI			0x01000000
+
+
+#define INIT_SCSI_NL_HDR(hdr, t, mtype, mlen)			\
+	{							\
+	(hdr)->version = SCSI_NL_VERSION;			\
+	(hdr)->transport = t;					\
+	(hdr)->magic = SCSI_NL_MAGIC;				\
+	(hdr)->msgtype = mtype;					\
+	(hdr)->msglen = mlen;					\
+	}
+
+
+#endif /* SCSI_NETLINK_H */
+
diff --git a/include/scsi/scsi_netlink_fc.h b/include/scsi/scsi_netlink_fc.h
new file mode 100644
index 000000000000..b213d2909fed
--- /dev/null
+++ b/include/scsi/scsi_netlink_fc.h
@@ -0,0 +1,71 @@
+/*
+ *  FC Transport Netlink Interface
+ *
+ *  Copyright (C) 2006   James Smart, Emulex Corporation
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+#ifndef SCSI_NETLINK_FC_H
+#define SCSI_NETLINK_FC_H
+
+#include <scsi/scsi_netlink.h>
+
+/*
+ * This file intended to be included by both kernel and user space
+ */
+
+/*
+ * FC Transport Message Types
+ */
+	/* kernel -> user */
+#define FC_NL_ASYNC_EVENT			0x0100
+	/* user -> kernel */
+/* none */
+
+
+/*
+ * Message Structures :
+ */
+
+/* macro to round up message lengths to 8byte boundary */
+#define FC_NL_MSGALIGN(len)		(((len) + 7) & ~7)
+
+
+/*
+ * FC Transport Broadcast Event Message :
+ *   FC_NL_ASYNC_EVENT
+ *
+ * Note: if Vendor Unique message, &event_data will be  start of
+ * 	 vendor unique payload, and the length of the payload is
+ *       per event_datalen
+ *
+ * Note: When specifying vendor_id, be sure to read the Vendor Type and ID
+ *   formatting requirements specified in scsi_netlink.h
+ */
+struct fc_nl_event {
+	struct scsi_nl_hdr snlh;		/* must be 1st element ! */
+	uint64_t seconds;
+	uint32_t vendor_id;
+	uint16_t host_no;
+	uint16_t event_datalen;
+	uint32_t event_num;
+	uint32_t event_code;
+	uint32_t event_data;
+} __attribute__((aligned(sizeof(uint64_t))));
+
+
+#endif /* SCSI_NETLINK_FC_H */
+
diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h
index c74be5dabfeb..f91c5358af3a 100644
--- a/include/scsi/scsi_transport_fc.h
+++ b/include/scsi/scsi_transport_fc.h
@@ -29,6 +29,7 @@
 
 #include <linux/sched.h>
 #include <scsi/scsi.h>
+#include <scsi/scsi_netlink.h>
 
 struct scsi_transport_template;
 
@@ -283,6 +284,30 @@ struct fc_host_statistics {
 };
 
 
+/*
+ * FC Event Codes - Polled and Async, following FC HBAAPI v2.0 guidelines
+ */
+
+/*
+ * fc_host_event_code: If you alter this, you also need to alter
+ * scsi_transport_fc.c (for the ascii descriptions).
+ */
+enum fc_host_event_code  {
+	FCH_EVT_LIP			= 0x1,
+	FCH_EVT_LINKUP			= 0x2,
+	FCH_EVT_LINKDOWN		= 0x3,
+	FCH_EVT_LIPRESET		= 0x4,
+	FCH_EVT_RSCN			= 0x5,
+	FCH_EVT_ADAPTER_CHANGE		= 0x103,
+	FCH_EVT_PORT_UNKNOWN		= 0x200,
+	FCH_EVT_PORT_OFFLINE		= 0x201,
+	FCH_EVT_PORT_ONLINE		= 0x202,
+	FCH_EVT_PORT_FABRIC		= 0x204,
+	FCH_EVT_LINK_UNKNOWN		= 0x500,
+	FCH_EVT_VENDOR_UNIQUE		= 0xffff,
+};
+
+
 /*
  * FC Local Port (Host) Attributes
  *
@@ -526,5 +551,14 @@ struct fc_rport *fc_remote_port_add(struct Scsi_Host *shost,
 void fc_remote_port_delete(struct fc_rport  *rport);
 void fc_remote_port_rolechg(struct fc_rport  *rport, u32 roles);
 int scsi_is_fc_rport(const struct device *);
+u32 fc_get_event_number(void);
+void fc_host_post_event(struct Scsi_Host *shost, u32 event_number,
+		enum fc_host_event_code event_code, u32 event_data);
+void fc_host_post_vendor_event(struct Scsi_Host *shost, u32 event_number,
+		u32 data_len, char * data_buf, u32 vendor_id);
+	/* Note: when specifying vendor_id to fc_host_post_vendor_event()
+	 *   be sure to read the Vendor Type and ID formatting requirements
+	 *   specified in scsi_netlink.h
+	 */
 
 #endif /* SCSI_TRANSPORT_FC_H */
-- 
cgit v1.2.3


From f14e2e29cdd07f80de6dec168dc2bb39de37eec3 Mon Sep 17 00:00:00 2001
From: James Smart <James.Smart@Emulex.Com>
Date: Tue, 22 Aug 2006 09:55:23 -0400
Subject: [SCSI] SCSI & FC transport: extend event vendor id's to 64bits

During discussions with Mike Christie, I became convinced that we needed
a larger vendor id. This patch extends the id from 32 to 64 bits.

This applies on top of the prior patches that add SCSI transport events
via netlink.

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_fc.c | 2 +-
 include/scsi/scsi_netlink.h      | 7 ++++---
 include/scsi/scsi_netlink_fc.h   | 2 +-
 include/scsi/scsi_transport_fc.h | 2 +-
 4 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 05989f130554..293188cbff8c 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -513,7 +513,7 @@ EXPORT_SYMBOL(fc_host_post_event);
  **/
 void
 fc_host_post_vendor_event(struct Scsi_Host *shost, u32 event_number,
-		u32 data_len, char * data_buf, u32 vendor_id)
+		u32 data_len, char * data_buf, u64 vendor_id)
 {
 	struct sk_buff *skb;
 	struct nlmsghdr	*nlh;
diff --git a/include/scsi/scsi_netlink.h b/include/scsi/scsi_netlink.h
index 7a3a20e640c0..8c1470cc8209 100644
--- a/include/scsi/scsi_netlink.h
+++ b/include/scsi/scsi_netlink.h
@@ -67,9 +67,10 @@ struct scsi_nl_hdr {
  *   Identifiers for each type:
  *    PCI :  ID data is the 16 bit PCI Registered Vendor ID
  */
-#define SCSI_NL_VID_ID_MASK			0x00FFFFFF
-#define SCSI_NL_VID_TYPE_MASK			0xFF000000
-#define SCSI_NL_VID_TYPE_PCI			0x01000000
+#define SCSI_NL_VID_TYPE_SHIFT		56
+#define SCSI_NL_VID_TYPE_MASK		((u64)0xFF << SCSI_NL_VID_TYPE_SHIFT)
+#define SCSI_NL_VID_TYPE_PCI		((u64)0x01 << SCSI_NL_VID_TYPE_SHIFT)
+#define SCSI_NL_VID_ID_MASK		(~ SCSI_NL_VID_TYPE_MASK)
 
 
 #define INIT_SCSI_NL_HDR(hdr, t, mtype, mlen)			\
diff --git a/include/scsi/scsi_netlink_fc.h b/include/scsi/scsi_netlink_fc.h
index b213d2909fed..cbf76e479761 100644
--- a/include/scsi/scsi_netlink_fc.h
+++ b/include/scsi/scsi_netlink_fc.h
@@ -58,7 +58,7 @@
 struct fc_nl_event {
 	struct scsi_nl_hdr snlh;		/* must be 1st element ! */
 	uint64_t seconds;
-	uint32_t vendor_id;
+	uint64_t vendor_id;
 	uint16_t host_no;
 	uint16_t event_datalen;
 	uint32_t event_num;
diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h
index f91c5358af3a..0b11eff989e0 100644
--- a/include/scsi/scsi_transport_fc.h
+++ b/include/scsi/scsi_transport_fc.h
@@ -555,7 +555,7 @@ u32 fc_get_event_number(void);
 void fc_host_post_event(struct Scsi_Host *shost, u32 event_number,
 		enum fc_host_event_code event_code, u32 event_data);
 void fc_host_post_vendor_event(struct Scsi_Host *shost, u32 event_number,
-		u32 data_len, char * data_buf, u32 vendor_id);
+		u32 data_len, char * data_buf, u64 vendor_id);
 	/* Note: when specifying vendor_id to fc_host_post_vendor_event()
 	 *   be sure to read the Vendor Type and ID formatting requirements
 	 *   specified in scsi_netlink.h
-- 
cgit v1.2.3


From 0f29b966d60e9a4f5ecff9f3832257b38aea4f13 Mon Sep 17 00:00:00 2001
From: James Smart <James.Smart@Emulex.Com>
Date: Fri, 18 Aug 2006 17:33:29 -0400
Subject: [SCSI] FC transport: Add dev_loss_tmo callbacks, and new
 fast_io_fail_tmo w/ callback

This patch adds the following functionality to the FC transport:

- dev_loss_tmo LLDD callback :
  Called to essentially confirm the deletion of an rport. Thus, it is
  called whenever the dev_loss_tmo fires, or when the rport is deleted
  due to other circumstances (module unload, etc).  It is expected that
  the callback will initiate the termination of any outstanding i/o on
  the rport.

- fast_io_fail_tmo and LLD callback:
  There are some cases where it may take a long while to truly determine
  device loss, but the system is in a multipathing configuration that if
  the i/o was failed quickly (faster than dev_loss_tmo), it could be
  redirected to a different path and completed sooner.

Many thanks to Mike Reed who cleaned up the initial RFC in support
of this post.

The original RFC is at:
http://marc.theaimsgroup.com/?l=linux-scsi&m=115505981027246&w=2

Signed-off-by: James Smart <James.Smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_fc.c | 134 +++++++++++++++++++++++++++++++++++----
 include/scsi/scsi_transport_fc.h |   5 ++
 2 files changed, 128 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 293188cbff8c..4ab176ed480d 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -242,6 +242,7 @@ fc_bitfield_name_search(remote_port_roles, fc_remote_port_role_names)
 
 
 static void fc_timeout_deleted_rport(void *data);
+static void fc_timeout_fail_rport_io(void *data);
 static void fc_scsi_scan_rport(void *data);
 
 /*
@@ -249,7 +250,7 @@ static void fc_scsi_scan_rport(void *data);
  * Increase these values if you add attributes
  */
 #define FC_STARGET_NUM_ATTRS 	3
-#define FC_RPORT_NUM_ATTRS	9
+#define FC_RPORT_NUM_ATTRS	10
 #define FC_HOST_NUM_ATTRS	17
 
 struct fc_internal {
@@ -622,11 +623,14 @@ store_fc_rport_##field(struct class_device *cdev, const char *buf,	\
 	struct fc_rport *rport = transport_class_to_rport(cdev);	\
 	struct Scsi_Host *shost = rport_to_shost(rport);		\
 	struct fc_internal *i = to_fc_internal(shost->transportt);	\
+	char *cp;							\
 	if ((rport->port_state == FC_PORTSTATE_BLOCKED) ||		\
 	    (rport->port_state == FC_PORTSTATE_DELETED) ||		\
 	    (rport->port_state == FC_PORTSTATE_NOTPRESENT))		\
 		return -EBUSY;						\
-	val = simple_strtoul(buf, NULL, 0);				\
+	val = simple_strtoul(buf, &cp, 0);				\
+	if (*cp && (*cp != '\n'))					\
+		return -EINVAL;						\
 	i->f->set_rport_##field(rport, val);				\
 	return count;							\
 }
@@ -708,6 +712,13 @@ static FC_CLASS_DEVICE_ATTR(rport, title, S_IRUGO,			\
 	if (i->f->show_rport_##field)					\
 		count++
 
+#define SETUP_PRIVATE_RPORT_ATTRIBUTE_RW(field)				\
+{									\
+	i->private_rport_attrs[count] = class_device_attr_rport_##field; \
+	i->rport_attrs[count] = &i->private_rport_attrs[count];		\
+	count++;							\
+}
+
 
 /* The FC Transport Remote Port Attributes: */
 
@@ -740,12 +751,14 @@ store_fc_rport_dev_loss_tmo(struct class_device *cdev, const char *buf,
 	struct fc_rport *rport = transport_class_to_rport(cdev);
 	struct Scsi_Host *shost = rport_to_shost(rport);
 	struct fc_internal *i = to_fc_internal(shost->transportt);
+	char *cp;
 	if ((rport->port_state == FC_PORTSTATE_BLOCKED) ||
 	    (rport->port_state == FC_PORTSTATE_DELETED) ||
 	    (rport->port_state == FC_PORTSTATE_NOTPRESENT))
 		return -EBUSY;
-	val = simple_strtoul(buf, NULL, 0);
-	if ((val < 0) || (val > SCSI_DEVICE_BLOCK_MAX_TIMEOUT))
+	val = simple_strtoul(buf, &cp, 0);
+	if ((*cp && (*cp != '\n')) ||
+	    (val < 0) || (val > SCSI_DEVICE_BLOCK_MAX_TIMEOUT))
 		return -EINVAL;
 	i->f->set_rport_dev_loss_tmo(rport, val);
 	return count;
@@ -795,6 +808,44 @@ static FC_CLASS_DEVICE_ATTR(rport, roles, S_IRUGO,
 fc_private_rport_rd_enum_attr(port_state, FC_PORTSTATE_MAX_NAMELEN);
 fc_private_rport_rd_attr(scsi_target_id, "%d\n", 20);
 
+/*
+ * fast_io_fail_tmo attribute
+ */
+static ssize_t
+show_fc_rport_fast_io_fail_tmo (struct class_device *cdev, char *buf)
+{
+	struct fc_rport *rport = transport_class_to_rport(cdev);
+
+	if (rport->fast_io_fail_tmo == -1)
+		return snprintf(buf, 5, "off\n");
+	return snprintf(buf, 20, "%d\n", rport->fast_io_fail_tmo);
+}
+
+static ssize_t
+store_fc_rport_fast_io_fail_tmo(struct class_device *cdev, const char *buf,
+			   size_t count)
+{
+	int val;
+	char *cp;
+	struct fc_rport *rport = transport_class_to_rport(cdev);
+
+	if ((rport->port_state == FC_PORTSTATE_BLOCKED) ||
+	    (rport->port_state == FC_PORTSTATE_DELETED) ||
+	    (rport->port_state == FC_PORTSTATE_NOTPRESENT))
+		return -EBUSY;
+	if (strncmp(buf, "off", 3) == 0)
+		rport->fast_io_fail_tmo = -1;
+	else {
+		val = simple_strtoul(buf, &cp, 0);
+		if ((*cp && (*cp != '\n')) ||
+		    (val < 0) || (val >= rport->dev_loss_tmo))
+			return -EINVAL;
+		rport->fast_io_fail_tmo = val;
+	}
+	return count;
+}
+static FC_CLASS_DEVICE_ATTR(rport, fast_io_fail_tmo, S_IRUGO | S_IWUSR,
+	show_fc_rport_fast_io_fail_tmo, store_fc_rport_fast_io_fail_tmo);
 
 
 /*
@@ -880,8 +931,11 @@ store_fc_host_##field(struct class_device *cdev, const char *buf,	\
 	int val;							\
 	struct Scsi_Host *shost = transport_class_to_shost(cdev);	\
 	struct fc_internal *i = to_fc_internal(shost->transportt);	\
+	char *cp;							\
 									\
-	val = simple_strtoul(buf, NULL, 0);				\
+	val = simple_strtoul(buf, &cp, 0);				\
+	if (*cp && (*cp != '\n'))					\
+		return -EINVAL;						\
 	i->f->set_host_##field(shost, val);				\
 	return count;							\
 }
@@ -1481,6 +1535,8 @@ fc_attach_transport(struct fc_function_template *ft)
 	SETUP_PRIVATE_RPORT_ATTRIBUTE_RD(roles);
 	SETUP_PRIVATE_RPORT_ATTRIBUTE_RD(port_state);
 	SETUP_PRIVATE_RPORT_ATTRIBUTE_RD(scsi_target_id);
+	if (ft->terminate_rport_io)
+		SETUP_PRIVATE_RPORT_ATTRIBUTE_RW(fast_io_fail_tmo);
 
 	BUG_ON(count > FC_RPORT_NUM_ATTRS);
 
@@ -1552,7 +1608,7 @@ fc_flush_work(struct Scsi_Host *shost)
  * @delay:	jiffies to delay the work queuing
  *
  * Return value:
- * 	0 on success / != 0 for error
+ * 	1 on success / 0 already queued / < 0 for error
  **/
 static int
 fc_queue_devloss_work(struct Scsi_Host *shost, struct work_struct *work,
@@ -1567,6 +1623,9 @@ fc_queue_devloss_work(struct Scsi_Host *shost, struct work_struct *work,
 		return -EINVAL;
 	}
 
+	if (delay == 0)
+		return queue_work(fc_host_devloss_work_q(shost), work);
+
 	return queue_delayed_work(fc_host_devloss_work_q(shost), work, delay);
 }
 
@@ -1659,10 +1718,23 @@ fc_starget_delete(void *data)
 	struct fc_rport *rport = (struct fc_rport *)data;
 	struct Scsi_Host *shost = rport_to_shost(rport);
 	unsigned long flags;
+	struct fc_internal *i = to_fc_internal(shost->transportt);
+
+	/*
+	 * Involve the LLDD if possible. All io on the rport is to
+	 * be terminated, either as part of the dev_loss_tmo callback
+	 * processing, or via the terminate_rport_io function.
+	 */
+	if (i->f->dev_loss_tmo_callbk)
+		i->f->dev_loss_tmo_callbk(rport);
+	else if (i->f->terminate_rport_io)
+		i->f->terminate_rport_io(rport);
 
 	spin_lock_irqsave(shost->host_lock, flags);
 	if (rport->flags & FC_RPORT_DEVLOSS_PENDING) {
 		spin_unlock_irqrestore(shost->host_lock, flags);
+		if (!cancel_delayed_work(&rport->fail_io_work))
+			fc_flush_devloss(shost);
 		if (!cancel_delayed_work(&rport->dev_loss_work))
 			fc_flush_devloss(shost);
 		spin_lock_irqsave(shost->host_lock, flags);
@@ -1685,10 +1757,7 @@ fc_rport_final_delete(void *data)
 	struct fc_rport *rport = (struct fc_rport *)data;
 	struct device *dev = &rport->dev;
 	struct Scsi_Host *shost = rport_to_shost(rport);
-
-	/* Delete SCSI target and sdevs */
-	if (rport->scsi_target_id != -1)
-		fc_starget_delete(data);
+	struct fc_internal *i = to_fc_internal(shost->transportt);
 
 	/*
 	 * if a scan is pending, flush the SCSI Host work_q so that 
@@ -1697,6 +1766,14 @@ fc_rport_final_delete(void *data)
 	if (rport->flags & FC_RPORT_SCAN_PENDING)
 		scsi_flush_work(shost);
 
+	/* Delete SCSI target and sdevs */
+	if (rport->scsi_target_id != -1)
+		fc_starget_delete(data);
+	else if (i->f->dev_loss_tmo_callbk)
+		i->f->dev_loss_tmo_callbk(rport);
+	else if (i->f->terminate_rport_io)
+		i->f->terminate_rport_io(rport);
+
 	transport_remove_device(dev);
 	device_del(dev);
 	transport_destroy_device(dev);
@@ -1748,8 +1825,10 @@ fc_rport_create(struct Scsi_Host *shost, int channel,
 	if (fci->f->dd_fcrport_size)
 		rport->dd_data = &rport[1];
 	rport->channel = channel;
+	rport->fast_io_fail_tmo = -1;
 
 	INIT_WORK(&rport->dev_loss_work, fc_timeout_deleted_rport, rport);
+	INIT_WORK(&rport->fail_io_work, fc_timeout_fail_rport_io, rport);
 	INIT_WORK(&rport->scan_work, fc_scsi_scan_rport, rport);
 	INIT_WORK(&rport->stgt_delete_work, fc_starget_delete, rport);
 	INIT_WORK(&rport->rport_delete_work, fc_rport_final_delete, rport);
@@ -1913,11 +1992,13 @@ fc_remote_port_add(struct Scsi_Host *shost, int channel,
 				/* restart the target */
 
 				/*
-				 * Stop the target timer first. Take no action
+				 * Stop the target timers first. Take no action
 				 * on the del_timer failure as the state
 				 * machine state change will validate the
 				 * transaction.
 				 */
+				if (!cancel_delayed_work(&rport->fail_io_work))
+					fc_flush_devloss(shost);
 				if (!cancel_delayed_work(work))
 					fc_flush_devloss(shost);
 
@@ -2061,6 +2142,7 @@ void
 fc_remote_port_delete(struct fc_rport  *rport)
 {
 	struct Scsi_Host *shost = rport_to_shost(rport);
+	struct fc_internal *i = to_fc_internal(shost->transportt);
 	int timeout = rport->dev_loss_tmo;
 	unsigned long flags;
 
@@ -2091,6 +2173,12 @@ fc_remote_port_delete(struct fc_rport  *rport)
 
 	scsi_target_block(&rport->dev);
 
+	/* see if we need to kill io faster than waiting for device loss */
+	if ((rport->fast_io_fail_tmo != -1) &&
+	    (rport->fast_io_fail_tmo < timeout) && (i->f->terminate_rport_io))
+		fc_queue_devloss_work(shost, &rport->fail_io_work,
+					rport->fast_io_fail_tmo * HZ);
+
 	/* cap the length the devices can be blocked until they are deleted */
 	fc_queue_devloss_work(shost, &rport->dev_loss_work, timeout * HZ);
 }
@@ -2150,6 +2238,8 @@ fc_remote_port_rolechg(struct fc_rport  *rport, u32 roles)
 		 * machine state change will validate the
 		 * transaction.
 		 */
+		if (!cancel_delayed_work(&rport->fail_io_work))
+			fc_flush_devloss(shost);
 		if (!cancel_delayed_work(&rport->dev_loss_work))
 			fc_flush_devloss(shost);
 
@@ -2270,6 +2360,28 @@ fc_timeout_deleted_rport(void  *data)
 	fc_queue_work(shost, &rport->stgt_delete_work);
 }
 
+/**
+ * fc_timeout_fail_rport_io - Timeout handler for a fast io failing on a
+ *                       disconnected SCSI target.
+ *
+ * @data:	rport to terminate io on.
+ *
+ * Notes: Only requests the failure of the io, not that all are flushed
+ *    prior to returning.
+ **/
+static void
+fc_timeout_fail_rport_io(void  *data)
+{
+	struct fc_rport *rport = (struct fc_rport *)data;
+	struct Scsi_Host *shost = rport_to_shost(rport);
+	struct fc_internal *i = to_fc_internal(shost->transportt);
+
+	if (rport->port_state != FC_PORTSTATE_BLOCKED)
+		return;
+
+	i->f->terminate_rport_io(rport);
+}
+
 /**
  * fc_scsi_scan_rport - called to perform a scsi scan on a remote port.
  *
diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h
index 0b11eff989e0..fd352323378b 100644
--- a/include/scsi/scsi_transport_fc.h
+++ b/include/scsi/scsi_transport_fc.h
@@ -195,6 +195,7 @@ struct fc_rport {	/* aka fc_starget_attrs */
 	u32 roles;
 	enum fc_port_state port_state;	/* Will only be ONLINE or UNKNOWN */
 	u32 scsi_target_id;
+	u32 fast_io_fail_tmo;
 
 	/* exported data */
 	void *dd_data;			/* Used for driver-specific storage */
@@ -207,6 +208,7 @@ struct fc_rport {	/* aka fc_starget_attrs */
 	struct device dev;
  	struct work_struct dev_loss_work;
  	struct work_struct scan_work;
+ 	struct work_struct fail_io_work;
  	struct work_struct stgt_delete_work;
 	struct work_struct rport_delete_work;
 } __attribute__((aligned(sizeof(unsigned long))));
@@ -445,6 +447,9 @@ struct fc_function_template {
 
 	int	(*issue_fc_host_lip)(struct Scsi_Host *);
 
+	void    (*dev_loss_tmo_callbk)(struct fc_rport *);
+	void	(*terminate_rport_io)(struct fc_rport *);
+
 	/* allocation lengths for host-specific data */
 	u32	 			dd_fcrport_size;
 
-- 
cgit v1.2.3


From 88edf74610bd894b93438f389688bc8b4a2d3414 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Wed, 6 Sep 2006 17:36:13 -0500
Subject: [SCSI] SAS: consolidate linkspeed definitions

At the moment we have two separate linkspeed enumerations covering
roughly the same values.  This patch consolidates on a single one enum
sas_linkspeed in scsi_transport_sas.h and uses it everywhere in the
aic94xx driver.  Eventually I'll get around to removing the duplicated
fields in asd_sas_phy and sas_phy ...

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/aic94xx/aic94xx_hwi.c  |  2 +-
 drivers/scsi/aic94xx/aic94xx_init.c | 12 ++++++++----
 drivers/scsi/aic94xx/aic94xx_scb.c  | 26 +++++++++++++-------------
 drivers/scsi/libsas/sas_expander.c  | 27 +++++++--------------------
 drivers/scsi/libsas/sas_internal.h  |  2 +-
 include/scsi/libsas.h               | 14 +++++++-------
 include/scsi/sas.h                  | 14 --------------
 include/scsi/scsi_transport_sas.h   | 26 +++++++++++++++++---------
 8 files changed, 54 insertions(+), 69 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c
index 075cea85b56b..a24201351108 100644
--- a/drivers/scsi/aic94xx/aic94xx_hwi.c
+++ b/drivers/scsi/aic94xx/aic94xx_hwi.c
@@ -96,7 +96,7 @@ static int asd_init_phy(struct asd_phy *phy)
 	sas_phy->type = PHY_TYPE_PHYSICAL;
 	sas_phy->role = PHY_ROLE_INITIATOR;
 	sas_phy->oob_mode = OOB_NOT_CONNECTED;
-	sas_phy->linkrate = PHY_LINKRATE_NONE;
+	sas_phy->linkrate = SAS_LINK_RATE_UNKNOWN;
 
 	phy->id_frm_tok = asd_alloc_coherent(asd_ha,
 					     sizeof(*phy->identify_frame),
diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c
index 69aa70887530..302b54fddf3c 100644
--- a/drivers/scsi/aic94xx/aic94xx_init.c
+++ b/drivers/scsi/aic94xx/aic94xx_init.c
@@ -240,10 +240,14 @@ static int __devinit asd_common_setup(struct asd_ha_struct *asd_ha)
 	/* All phys are enabled, by default. */
 	asd_ha->hw_prof.enabled_phys = 0xFF;
 	for (i = 0; i < ASD_MAX_PHYS; i++) {
-		asd_ha->hw_prof.phy_desc[i].max_sas_lrate = PHY_LINKRATE_3;
-		asd_ha->hw_prof.phy_desc[i].min_sas_lrate = PHY_LINKRATE_1_5;
-		asd_ha->hw_prof.phy_desc[i].max_sata_lrate= PHY_LINKRATE_1_5;
-		asd_ha->hw_prof.phy_desc[i].min_sata_lrate= PHY_LINKRATE_1_5;
+		asd_ha->hw_prof.phy_desc[i].max_sas_lrate =
+			SAS_LINK_RATE_3_0_GBPS;
+		asd_ha->hw_prof.phy_desc[i].min_sas_lrate =
+			SAS_LINK_RATE_1_5_GBPS;
+		asd_ha->hw_prof.phy_desc[i].max_sata_lrate =
+			SAS_LINK_RATE_1_5_GBPS;
+		asd_ha->hw_prof.phy_desc[i].min_sata_lrate =
+			SAS_LINK_RATE_1_5_GBPS;
 	}
 
 	return 0;
diff --git a/drivers/scsi/aic94xx/aic94xx_scb.c b/drivers/scsi/aic94xx/aic94xx_scb.c
index fc1b7438a913..ef8ca08b545f 100644
--- a/drivers/scsi/aic94xx/aic94xx_scb.c
+++ b/drivers/scsi/aic94xx/aic94xx_scb.c
@@ -55,15 +55,15 @@ static inline void get_lrate_mode(struct asd_phy *phy, u8 oob_mode)
 	switch (oob_mode & 7) {
 	case PHY_SPEED_60:
 		/* FIXME: sas transport class doesn't have this */
-		phy->sas_phy.linkrate = PHY_LINKRATE_6;
+		phy->sas_phy.linkrate = SAS_LINK_RATE_6_0_GBPS;
 		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_6_0_GBPS;
 		break;
 	case PHY_SPEED_30:
-		phy->sas_phy.linkrate = PHY_LINKRATE_3;
+		phy->sas_phy.linkrate = SAS_LINK_RATE_3_0_GBPS;
 		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_3_0_GBPS;
 		break;
 	case PHY_SPEED_15:
-		phy->sas_phy.linkrate = PHY_LINKRATE_1_5;
+		phy->sas_phy.linkrate = SAS_LINK_RATE_1_5_GBPS;
 		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_1_5_GBPS;
 		break;
 	}
@@ -540,39 +540,39 @@ static inline void set_speed_mask(u8 *speed_mask, struct asd_phy_desc *pd)
 		| SATA_SPEED_30_DIS | SATA_SPEED_15_DIS;
 
 	switch (pd->max_sas_lrate) {
-	case PHY_LINKRATE_6:
+	case SAS_LINK_RATE_6_0_GBPS:
 		*speed_mask &= ~SAS_SPEED_60_DIS;
 	default:
-	case PHY_LINKRATE_3:
+	case SAS_LINK_RATE_3_0_GBPS:
 		*speed_mask &= ~SAS_SPEED_30_DIS;
-	case PHY_LINKRATE_1_5:
+	case SAS_LINK_RATE_1_5_GBPS:
 		*speed_mask &= ~SAS_SPEED_15_DIS;
 	}
 
 	switch (pd->min_sas_lrate) {
-	case PHY_LINKRATE_6:
+	case SAS_LINK_RATE_6_0_GBPS:
 		*speed_mask |= SAS_SPEED_30_DIS;
-	case PHY_LINKRATE_3:
+	case SAS_LINK_RATE_3_0_GBPS:
 		*speed_mask |= SAS_SPEED_15_DIS;
 	default:
-	case PHY_LINKRATE_1_5:
+	case SAS_LINK_RATE_1_5_GBPS:
 		/* nothing to do */
 		;
 	}
 
 	switch (pd->max_sata_lrate) {
-	case PHY_LINKRATE_3:
+	case SAS_LINK_RATE_3_0_GBPS:
 		*speed_mask &= ~SATA_SPEED_30_DIS;
 	default:
-	case PHY_LINKRATE_1_5:
+	case SAS_LINK_RATE_1_5_GBPS:
 		*speed_mask &= ~SATA_SPEED_15_DIS;
 	}
 
 	switch (pd->min_sata_lrate) {
-	case PHY_LINKRATE_3:
+	case SAS_LINK_RATE_3_0_GBPS:
 		*speed_mask |= SATA_SPEED_15_DIS;
 	default:
-	case PHY_LINKRATE_1_5:
+	case SAS_LINK_RATE_1_5_GBPS:
 		/* nothing to do */
 		;
 	}
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index b653a263f76a..02e796ee027e 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -191,20 +191,7 @@ static void sas_set_ex_phy(struct domain_device *dev, int phy_id,
 	phy->phy->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS;
 	phy->phy->minimum_linkrate = SAS_LINK_RATE_1_5_GBPS;
 	phy->phy->maximum_linkrate = SAS_LINK_RATE_3_0_GBPS;
-	switch (phy->linkrate) {
-	case PHY_LINKRATE_1_5:
-		phy->phy->negotiated_linkrate = SAS_LINK_RATE_1_5_GBPS;
-		break;
-	case PHY_LINKRATE_3:
-		phy->phy->negotiated_linkrate = SAS_LINK_RATE_3_0_GBPS;
-		break;
-	case PHY_LINKRATE_6:
-		phy->phy->negotiated_linkrate = SAS_LINK_RATE_6_0_GBPS;
-		break;
-	default:
-		phy->phy->negotiated_linkrate = SAS_LINK_RATE_UNKNOWN;
-		break;
-	}
+	phy->phy->negotiated_linkrate = phy->linkrate;
 
 	if (!rediscover)
 		sas_phy_add(phy->phy);
@@ -450,7 +437,7 @@ static void sas_ex_disable_phy(struct domain_device *dev, int phy_id)
 	struct ex_phy *phy = &ex->ex_phy[phy_id];
 
 	sas_smp_phy_control(dev, phy_id, PHY_FUNC_DISABLE);
-	phy->linkrate = PHY_DISABLED;
+	phy->linkrate = SAS_PHY_DISABLED;
 }
 
 static void sas_ex_disable_port(struct domain_device *dev, u8 *sas_addr)
@@ -743,7 +730,7 @@ static int sas_ex_discover_dev(struct domain_device *dev, int phy_id)
 	int res = 0;
 
 	/* Phy state */
-	if (ex_phy->linkrate == PHY_SPINUP_HOLD) {
+	if (ex_phy->linkrate == SAS_SATA_SPINUP_HOLD) {
 		if (!sas_smp_phy_control(dev, phy_id, PHY_FUNC_LINK_RESET))
 			res = sas_ex_phy_discover(dev, phy_id);
 		if (res)
@@ -773,7 +760,7 @@ static int sas_ex_discover_dev(struct domain_device *dev, int phy_id)
 			sas_configure_routing(dev, ex_phy->attached_sas_addr);
 		}
 		return 0;
-	} else if (ex_phy->linkrate == PHY_LINKRATE_UNKNOWN)
+	} else if (ex_phy->linkrate == SAS_LINK_RATE_UNKNOWN)
 		return 0;
 
 	if (ex_phy->attached_dev_type != SAS_END_DEV &&
@@ -922,9 +909,9 @@ static int sas_ex_discover_devices(struct domain_device *dev, int single)
 			continue;
 
 		switch (ex_phy->linkrate) {
-		case PHY_DISABLED:
-		case PHY_RESET_PROBLEM:
-		case PHY_PORT_SELECTOR:
+		case SAS_PHY_DISABLED:
+		case SAS_PHY_RESET_PROBLEM:
+		case SAS_SATA_PORT_SELECTOR:
 			continue;
 		default:
 			res = sas_ex_discover_dev(dev, i);
diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h
index 89c397680846..0d69ede4b944 100644
--- a/drivers/scsi/libsas/sas_internal.h
+++ b/drivers/scsi/libsas/sas_internal.h
@@ -43,7 +43,7 @@ void sas_scsi_recover_host(struct Scsi_Host *shost);
 
 int sas_show_class(enum sas_class class, char *buf);
 int sas_show_proto(enum sas_proto proto, char *buf);
-int sas_show_linkrate(enum sas_phy_linkrate linkrate, char *buf);
+int sas_show_linkrate(enum sas_linkrate linkrate, char *buf);
 int sas_show_oob_mode(enum sas_oob_mode oob_mode, char *buf);
 
 int  sas_register_phys(struct sas_ha_struct *sas_ha);
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index 72acdabe7f80..8d91313dd888 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -114,7 +114,7 @@ struct ex_phy {
 	enum ex_phy_state phy_state;
 
 	enum sas_dev_type attached_dev_type;
-	enum sas_phy_linkrate linkrate;
+	enum sas_linkrate linkrate;
 
 	u8   attached_sata_host:1;
 	u8   attached_sata_dev:1;
@@ -170,9 +170,9 @@ struct sata_device {
 struct domain_device {
         enum sas_dev_type dev_type;
 
-        enum sas_phy_linkrate linkrate;
-        enum sas_phy_linkrate min_linkrate;
-        enum sas_phy_linkrate max_linkrate;
+        enum sas_linkrate linkrate;
+        enum sas_linkrate min_linkrate;
+        enum sas_linkrate max_linkrate;
 
         int  pathways;
 
@@ -220,7 +220,7 @@ struct asd_sas_port {
 	struct domain_device *port_dev;
 	spinlock_t dev_list_lock;
 	struct list_head dev_list;
-	enum   sas_phy_linkrate linkrate;
+	enum   sas_linkrate linkrate;
 
 	struct sas_phy *phy;
 	struct work_struct work;
@@ -276,7 +276,7 @@ struct asd_sas_phy {
 	enum sas_phy_type  type;
 	enum sas_phy_role  role;
 	enum sas_oob_mode  oob_mode;
-	enum sas_phy_linkrate linkrate;
+	enum sas_linkrate linkrate;
 
 	u8   *sas_addr;		  /* must be set */
 	u8   attached_sas_addr[SAS_ADDR_SIZE]; /* class:RO, driver: R/W */
@@ -368,7 +368,7 @@ void sas_hash_addr(u8 *hashed, const u8 *sas_addr);
 static inline void sas_phy_disconnected(struct asd_sas_phy *phy)
 {
 	phy->oob_mode = OOB_NOT_CONNECTED;
-	phy->linkrate = PHY_LINKRATE_NONE;
+	phy->linkrate = SAS_LINK_RATE_UNKNOWN;
 }
 
 /* ---------- Tasks ---------- */
diff --git a/include/scsi/sas.h b/include/scsi/sas.h
index 752853a113dc..9c8a5b91ae64 100644
--- a/include/scsi/sas.h
+++ b/include/scsi/sas.h
@@ -102,20 +102,6 @@ enum sas_dev_type {
 	SATA_PM_PORT= 8,
 };
 
-enum sas_phy_linkrate {
-	PHY_LINKRATE_NONE = 0,
-	PHY_LINKRATE_UNKNOWN = 0,
-	PHY_DISABLED,
-	PHY_RESET_PROBLEM,
-	PHY_SPINUP_HOLD,
-	PHY_PORT_SELECTOR,
-	PHY_LINKRATE_1_5 = 0x08,
-	PHY_LINKRATE_G1  = PHY_LINKRATE_1_5,
-	PHY_LINKRATE_3   = 0x09,
-	PHY_LINKRATE_G2  = PHY_LINKRATE_3,
-	PHY_LINKRATE_6   = 0x0A,
-};
-
 /* Partly from IDENTIFY address frame. */
 enum sas_proto {
 	SATA_PROTO    = 1,
diff --git a/include/scsi/scsi_transport_sas.h b/include/scsi/scsi_transport_sas.h
index eeb2200de855..87de518960c1 100644
--- a/include/scsi/scsi_transport_sas.h
+++ b/include/scsi/scsi_transport_sas.h
@@ -24,15 +24,23 @@ enum sas_protocol {
 };
 
 enum sas_linkrate {
-	SAS_LINK_RATE_UNKNOWN,
-	SAS_PHY_DISABLED,
-	SAS_LINK_RATE_FAILED,
-	SAS_SATA_SPINUP_HOLD,
-	SAS_SATA_PORT_SELECTOR,
-	SAS_LINK_RATE_1_5_GBPS,
-	SAS_LINK_RATE_3_0_GBPS,
-	SAS_LINK_RATE_6_0_GBPS,
-	SAS_LINK_VIRTUAL,
+	/* These Values are defined in the SAS standard */
+	SAS_LINK_RATE_UNKNOWN = 0,
+	SAS_PHY_DISABLED = 1,
+	SAS_PHY_RESET_PROBLEM = 2,
+	SAS_SATA_SPINUP_HOLD = 3,
+	SAS_SATA_PORT_SELECTOR = 4,
+	SAS_PHY_RESET_IN_PROGRESS = 5,
+	SAS_LINK_RATE_1_5_GBPS = 8,
+	SAS_LINK_RATE_G1 = SAS_LINK_RATE_1_5_GBPS,
+	SAS_LINK_RATE_3_0_GBPS = 9,
+	SAS_LINK_RATE_G2 = SAS_LINK_RATE_3_0_GBPS,
+	SAS_LINK_RATE_6_0_GBPS = 10,
+	/* These are virtual to the transport class and may never
+	 * be signalled normally since the standard defined field
+	 * is only 4 bits */
+	SAS_LINK_RATE_FAILED = 0x10,
+	SAS_PHY_VIRTUAL = 0x11,
 };
 
 struct sas_identify {
-- 
cgit v1.2.3


From d24e1eeb3a16e4944288c2f3bf082e1513f4b425 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Wed, 6 Sep 2006 19:25:22 -0500
Subject: [SCSI] scsi_transport_sas: make minimum and maximum linkrate settable
 quantities

According to SPEC, the minimum_linkrate and maximum_linkrate should be
settable by the user.  This patch introduces a callback that allows the
sas class to pass these settings on to the driver.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_sas.c | 73 +++++++++++++++++++++++++++++++++++----
 include/scsi/scsi_transport_sas.h |  6 ++++
 2 files changed, 73 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
index d518c1207fb4..b5b0c2cba96b 100644
--- a/drivers/scsi/scsi_transport_sas.c
+++ b/drivers/scsi/scsi_transport_sas.c
@@ -77,6 +77,24 @@ get_sas_##title##_names(u32 table_key, char *buf)		\
 	return len;						\
 }
 
+#define sas_bitfield_name_set(title, table)			\
+static ssize_t							\
+set_sas_##title##_names(u32 *table_key, const char *buf)	\
+{								\
+	ssize_t len = 0;					\
+	int i;							\
+								\
+	for (i = 0; i < ARRAY_SIZE(table); i++) {		\
+		len = strlen(table[i].name);			\
+		if (strncmp(buf, table[i].name, len) == 0 &&	\
+		    (buf[len] == '\n' || buf[len] == '\0')) {	\
+			*table_key = table[i].value;		\
+			return 0;				\
+		}						\
+	}							\
+	return -EINVAL;						\
+}
+
 #define sas_bitfield_name_search(title, table)			\
 static ssize_t							\
 get_sas_##title##_names(u32 table_key, char *buf)		\
@@ -131,7 +149,7 @@ static struct {
 	{ SAS_LINK_RATE_6_0_GBPS,	"6.0 Gbit" },
 };
 sas_bitfield_name_search(linkspeed, sas_linkspeed_names)
-
+sas_bitfield_name_set(linkspeed, sas_linkspeed_names)
 
 /*
  * SAS host attributes
@@ -253,10 +271,39 @@ show_sas_phy_##field(struct class_device *cdev, char *buf)		\
 	return get_sas_linkspeed_names(phy->field, buf);		\
 }
 
+/* Fudge to tell if we're minimum or maximum */
+#define sas_phy_store_linkspeed(field)					\
+static ssize_t								\
+store_sas_phy_##field(struct class_device *cdev, const char *buf,	\
+		      size_t count)					\
+{									\
+	struct sas_phy *phy = transport_class_to_phy(cdev);		\
+	struct Scsi_Host *shost = dev_to_shost(phy->dev.parent);	\
+	struct sas_internal *i = to_sas_internal(shost->transportt);	\
+	u32 value;							\
+	struct sas_phy_linkrates rates = {0};				\
+	int error;							\
+									\
+	error = set_sas_linkspeed_names(&value, buf);			\
+	if (error)							\
+		return error;						\
+	rates.field = value;						\
+	error = i->f->set_phy_speed(phy, &rates);			\
+									\
+	return error ? error : count;					\
+}
+
+#define sas_phy_linkspeed_rw_attr(field)				\
+	sas_phy_show_linkspeed(field)					\
+	sas_phy_store_linkspeed(field)					\
+static CLASS_DEVICE_ATTR(field, S_IRUGO, show_sas_phy_##field,		\
+	store_sas_phy_##field)
+
 #define sas_phy_linkspeed_attr(field)					\
 	sas_phy_show_linkspeed(field)					\
 static CLASS_DEVICE_ATTR(field, S_IRUGO, show_sas_phy_##field, NULL)
 
+
 #define sas_phy_show_linkerror(field)					\
 static ssize_t								\
 show_sas_phy_##field(struct class_device *cdev, char *buf)		\
@@ -326,9 +373,9 @@ sas_phy_simple_attr(identify.phy_identifier, phy_identifier, "%d\n", u8);
 //sas_phy_simple_attr(port_identifier, port_identifier, "%d\n", int);
 sas_phy_linkspeed_attr(negotiated_linkrate);
 sas_phy_linkspeed_attr(minimum_linkrate_hw);
-sas_phy_linkspeed_attr(minimum_linkrate);
+sas_phy_linkspeed_rw_attr(minimum_linkrate);
 sas_phy_linkspeed_attr(maximum_linkrate_hw);
-sas_phy_linkspeed_attr(maximum_linkrate);
+sas_phy_linkspeed_rw_attr(maximum_linkrate);
 sas_phy_linkerror_attr(invalid_dword_count);
 sas_phy_linkerror_attr(running_disparity_error_count);
 sas_phy_linkerror_attr(loss_of_dword_sync_count);
@@ -1310,13 +1357,23 @@ static int sas_user_scan(struct Scsi_Host *shost, uint channel,
  * Setup / Teardown code
  */
 
-#define SETUP_TEMPLATE(attrb, field, perm, test)				\
+#define SETUP_TEMPLATE(attrb, field, perm, test)			\
 	i->private_##attrb[count] = class_device_attr_##field;		\
 	i->private_##attrb[count].attr.mode = perm;			\
 	i->attrb[count] = &i->private_##attrb[count];			\
 	if (test)							\
 		count++
 
+#define SETUP_TEMPLATE_RW(attrb, field, perm, test, ro_test, ro_perm)	\
+	i->private_##attrb[count] = class_device_attr_##field;		\
+	i->private_##attrb[count].attr.mode = perm;			\
+	if (ro_test) {							\
+		i->private_##attrb[count].attr.mode = ro_perm;		\
+		i->private_##attrb[count].store = NULL;			\
+	}								\
+	i->attrb[count] = &i->private_##attrb[count];			\
+	if (test)							\
+		count++
 
 #define SETUP_RPORT_ATTRIBUTE(field) 					\
 	SETUP_TEMPLATE(rphy_attrs, field, S_IRUGO, 1)
@@ -1327,6 +1384,10 @@ static int sas_user_scan(struct Scsi_Host *shost, uint channel,
 #define SETUP_PHY_ATTRIBUTE(field)					\
 	SETUP_TEMPLATE(phy_attrs, field, S_IRUGO, 1)
 
+#define SETUP_PHY_ATTRIBUTE_RW(field)					\
+	SETUP_TEMPLATE_RW(phy_attrs, field, S_IRUGO | S_IWUSR, 1,	\
+			!i->f->set_phy_speed, S_IRUGO)
+
 #define SETUP_PORT_ATTRIBUTE(field)					\
 	SETUP_TEMPLATE(port_attrs, field, S_IRUGO, 1)
 
@@ -1407,9 +1468,9 @@ sas_attach_transport(struct sas_function_template *ft)
 	//SETUP_PHY_ATTRIBUTE(port_identifier);
 	SETUP_PHY_ATTRIBUTE(negotiated_linkrate);
 	SETUP_PHY_ATTRIBUTE(minimum_linkrate_hw);
-	SETUP_PHY_ATTRIBUTE(minimum_linkrate);
+	SETUP_PHY_ATTRIBUTE_RW(minimum_linkrate);
 	SETUP_PHY_ATTRIBUTE(maximum_linkrate_hw);
-	SETUP_PHY_ATTRIBUTE(maximum_linkrate);
+	SETUP_PHY_ATTRIBUTE_RW(maximum_linkrate);
 
 	SETUP_PHY_ATTRIBUTE(invalid_dword_count);
 	SETUP_PHY_ATTRIBUTE(running_disparity_error_count);
diff --git a/include/scsi/scsi_transport_sas.h b/include/scsi/scsi_transport_sas.h
index 87de518960c1..53024377f3b8 100644
--- a/include/scsi/scsi_transport_sas.h
+++ b/include/scsi/scsi_transport_sas.h
@@ -150,12 +150,18 @@ struct sas_port {
 #define transport_class_to_sas_port(cdev) \
 	dev_to_sas_port((cdev)->dev)
 
+struct sas_phy_linkrates {
+	enum sas_linkrate maximum_linkrate;
+	enum sas_linkrate minimum_linkrate;
+};
+
 /* The functions by which the transport class and the driver communicate */
 struct sas_function_template {
 	int (*get_linkerrors)(struct sas_phy *);
 	int (*get_enclosure_identifier)(struct sas_rphy *, u64 *);
 	int (*get_bay_identifier)(struct sas_rphy *);
 	int (*phy_reset)(struct sas_phy *, int);
+	int (*set_phy_speed)(struct sas_phy *, struct sas_phy_linkrates *);
 };
 
 
-- 
cgit v1.2.3


From a01e70e570a72b8a8c9a58062e4f5bdcd3986222 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Wed, 6 Sep 2006 19:28:07 -0500
Subject: [SCSI] aci94xx: implement link rate setting

This patch implements the ability to set the minimum and maximum
linkrates for both libsas (for expanders) and aic94xx (for the host
phys).  It also tidies up the setting of the hardware min and max to
make sure they're updated when the expander emits a change broadcast.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/aic94xx/aic94xx.h     |  2 +-
 drivers/scsi/aic94xx/aic94xx_scb.c | 30 ++++++++++++++++++++++++--
 drivers/scsi/libsas/sas_expander.c | 20 +++++++++++------
 drivers/scsi/libsas/sas_init.c     | 44 ++++++++++++++++++++++++++++++++++++--
 drivers/scsi/libsas/sas_internal.h |  2 +-
 drivers/scsi/libsas/sas_phy.c      | 15 +++++++------
 include/scsi/libsas.h              |  2 +-
 include/scsi/sas.h                 |  1 +
 8 files changed, 95 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/aic94xx/aic94xx.h b/drivers/scsi/aic94xx/aic94xx.h
index cb7caf1c9ce1..1bd5b4ecf3d5 100644
--- a/drivers/scsi/aic94xx/aic94xx.h
+++ b/drivers/scsi/aic94xx/aic94xx.h
@@ -109,6 +109,6 @@ int  asd_clear_nexus_port(struct asd_sas_port *port);
 int  asd_clear_nexus_ha(struct sas_ha_struct *sas_ha);
 
 /* ---------- Phy Management ---------- */
-int  asd_control_phy(struct asd_sas_phy *phy, enum phy_func func);
+int  asd_control_phy(struct asd_sas_phy *phy, enum phy_func func, void *arg);
 
 #endif
diff --git a/drivers/scsi/aic94xx/aic94xx_scb.c b/drivers/scsi/aic94xx/aic94xx_scb.c
index ef8ca08b545f..7ee49b51b724 100644
--- a/drivers/scsi/aic94xx/aic94xx_scb.c
+++ b/drivers/scsi/aic94xx/aic94xx_scb.c
@@ -52,6 +52,8 @@
 
 static inline void get_lrate_mode(struct asd_phy *phy, u8 oob_mode)
 {
+	struct sas_phy *sas_phy = phy->sas_phy.phy;
+
 	switch (oob_mode & 7) {
 	case PHY_SPEED_60:
 		/* FIXME: sas transport class doesn't have this */
@@ -67,6 +69,12 @@ static inline void get_lrate_mode(struct asd_phy *phy, u8 oob_mode)
 		phy->sas_phy.phy->negotiated_linkrate = SAS_LINK_RATE_1_5_GBPS;
 		break;
 	}
+	sas_phy->negotiated_linkrate = phy->sas_phy.linkrate;
+	sas_phy->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS;
+	sas_phy->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS;
+	sas_phy->maximum_linkrate = phy->phy_desc->max_sas_lrate;
+	sas_phy->minimum_linkrate = phy->phy_desc->min_sas_lrate;
+
 	if (oob_mode & SAS_MODE)
 		phy->sas_phy.oob_mode = SAS_OOB_MODE;
 	else if (oob_mode & SATA_MODE)
@@ -710,14 +718,32 @@ static const int phy_func_table[] = {
 	[PHY_FUNC_RELEASE_SPINUP_HOLD] = RELEASE_SPINUP_HOLD,
 };
 
-int asd_control_phy(struct asd_sas_phy *phy, enum phy_func func)
+int asd_control_phy(struct asd_sas_phy *phy, enum phy_func func, void *arg)
 {
 	struct asd_ha_struct *asd_ha = phy->ha->lldd_ha;
+	struct asd_phy_desc *pd = asd_ha->phys[phy->id].phy_desc;
 	struct asd_ascb *ascb;
+	struct sas_phy_linkrates *rates;
 	int res = 1;
 
-	if (func == PHY_FUNC_CLEAR_ERROR_LOG)
+	switch (func) {
+	case PHY_FUNC_CLEAR_ERROR_LOG:
 		return -ENOSYS;
+	case PHY_FUNC_SET_LINK_RATE:
+		rates = arg;
+		if (rates->minimum_linkrate) {
+			pd->min_sas_lrate = rates->minimum_linkrate;
+			pd->min_sata_lrate = rates->minimum_linkrate;
+		}
+		if (rates->maximum_linkrate) {
+			pd->max_sas_lrate = rates->maximum_linkrate;
+			pd->max_sata_lrate = rates->maximum_linkrate;
+		}
+		func = PHY_FUNC_LINK_RESET;
+		break;
+	default:
+		break;
+	}
 
 	ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL);
 	if (!ascb)
diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c
index 02e796ee027e..30b8014bcc7a 100644
--- a/drivers/scsi/libsas/sas_expander.c
+++ b/drivers/scsi/libsas/sas_expander.c
@@ -187,10 +187,10 @@ static void sas_set_ex_phy(struct domain_device *dev, int phy_id,
 	phy->phy->identify.initiator_port_protocols = phy->attached_iproto;
 	phy->phy->identify.target_port_protocols = phy->attached_tproto;
 	phy->phy->identify.phy_identifier = phy_id;
-	phy->phy->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS;
-	phy->phy->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS;
-	phy->phy->minimum_linkrate = SAS_LINK_RATE_1_5_GBPS;
-	phy->phy->maximum_linkrate = SAS_LINK_RATE_3_0_GBPS;
+	phy->phy->minimum_linkrate_hw = dr->hmin_linkrate;
+	phy->phy->maximum_linkrate_hw = dr->hmax_linkrate;
+	phy->phy->minimum_linkrate = dr->pmin_linkrate;
+	phy->phy->maximum_linkrate = dr->pmax_linkrate;
 	phy->phy->negotiated_linkrate = phy->linkrate;
 
 	if (!rediscover)
@@ -404,7 +404,8 @@ out:
 #define PC_RESP_SIZE 8
 
 int sas_smp_phy_control(struct domain_device *dev, int phy_id,
-			enum phy_func phy_func)
+			enum phy_func phy_func,
+			struct sas_phy_linkrates *rates)
 {
 	u8 *pc_req;
 	u8 *pc_resp;
@@ -423,6 +424,10 @@ int sas_smp_phy_control(struct domain_device *dev, int phy_id,
 	pc_req[1] = SMP_PHY_CONTROL;
 	pc_req[9] = phy_id;
 	pc_req[10]= phy_func;
+	if (rates) {
+		pc_req[32] = rates->minimum_linkrate << 4;
+		pc_req[33] = rates->maximum_linkrate << 4;
+	}
 
 	res = smp_execute_task(dev, pc_req, PC_REQ_SIZE, pc_resp,PC_RESP_SIZE);
 
@@ -436,7 +441,7 @@ static void sas_ex_disable_phy(struct domain_device *dev, int phy_id)
 	struct expander_device *ex = &dev->ex_dev;
 	struct ex_phy *phy = &ex->ex_phy[phy_id];
 
-	sas_smp_phy_control(dev, phy_id, PHY_FUNC_DISABLE);
+	sas_smp_phy_control(dev, phy_id, PHY_FUNC_DISABLE, NULL);
 	phy->linkrate = SAS_PHY_DISABLED;
 }
 
@@ -731,7 +736,7 @@ static int sas_ex_discover_dev(struct domain_device *dev, int phy_id)
 
 	/* Phy state */
 	if (ex_phy->linkrate == SAS_SATA_SPINUP_HOLD) {
-		if (!sas_smp_phy_control(dev, phy_id, PHY_FUNC_LINK_RESET))
+		if (!sas_smp_phy_control(dev, phy_id, PHY_FUNC_LINK_RESET, NULL))
 			res = sas_ex_phy_discover(dev, phy_id);
 		if (res)
 			return res;
@@ -1706,6 +1711,7 @@ static int sas_rediscover_dev(struct domain_device *dev, int phy_id)
 		   SAS_ADDR(phy->attached_sas_addr)) {
 		SAS_DPRINTK("ex %016llx phy 0x%x broadcast flutter\n",
 			    SAS_ADDR(dev->sas_addr), phy_id);
+		sas_ex_phy_discover(dev, phy_id);
 	} else
 		res = sas_discover_new(dev, phy_id);
 out:
diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c
index b961664b8106..c836a237fb79 100644
--- a/drivers/scsi/libsas/sas_init.c
+++ b/drivers/scsi/libsas/sas_init.c
@@ -159,17 +159,57 @@ static int sas_phy_reset(struct sas_phy *phy, int hard_reset)
 		struct sas_internal *i =
 			to_sas_internal(sas_ha->core.shost->transportt);
 
-		ret = i->dft->lldd_control_phy(asd_phy, reset_type);
+		ret = i->dft->lldd_control_phy(asd_phy, reset_type, NULL);
 	} else {
 		struct sas_rphy *rphy = dev_to_rphy(phy->dev.parent);
 		struct domain_device *ddev = sas_find_dev_by_rphy(rphy);
-		ret = sas_smp_phy_control(ddev, phy->number, reset_type);
+		ret = sas_smp_phy_control(ddev, phy->number, reset_type, NULL);
 	}
 	return ret;
 }
 
+static int sas_set_phy_speed(struct sas_phy *phy,
+			     struct sas_phy_linkrates *rates)
+{
+	int ret;
+
+	if ((rates->minimum_linkrate &&
+	     rates->minimum_linkrate > phy->maximum_linkrate) ||
+	    (rates->maximum_linkrate &&
+	     rates->maximum_linkrate < phy->minimum_linkrate))
+		return -EINVAL;
+
+	if (rates->minimum_linkrate &&
+	    rates->minimum_linkrate < phy->minimum_linkrate_hw)
+		rates->minimum_linkrate = phy->minimum_linkrate_hw;
+
+	if (rates->maximum_linkrate &&
+	    rates->maximum_linkrate > phy->maximum_linkrate_hw)
+		rates->maximum_linkrate = phy->maximum_linkrate_hw;
+
+	if (scsi_is_sas_phy_local(phy)) {
+		struct Scsi_Host *shost = dev_to_shost(phy->dev.parent);
+		struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost);
+		struct asd_sas_phy *asd_phy = sas_ha->sas_phy[phy->number];
+		struct sas_internal *i =
+			to_sas_internal(sas_ha->core.shost->transportt);
+
+		ret = i->dft->lldd_control_phy(asd_phy, PHY_FUNC_SET_LINK_RATE,
+					       rates);
+	} else {
+		struct sas_rphy *rphy = dev_to_rphy(phy->dev.parent);
+		struct domain_device *ddev = sas_find_dev_by_rphy(rphy);
+		ret = sas_smp_phy_control(ddev, phy->number,
+					  PHY_FUNC_LINK_RESET, rates);
+
+	}
+
+	return ret;
+}
+
 static struct sas_function_template sft = {
 	.phy_reset = sas_phy_reset,
+	.set_phy_speed = sas_set_phy_speed,
 	.get_linkerrors = sas_get_linkerrors,
 };
 
diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h
index 0d69ede4b944..bffcee474921 100644
--- a/drivers/scsi/libsas/sas_internal.h
+++ b/drivers/scsi/libsas/sas_internal.h
@@ -70,7 +70,7 @@ int sas_notify_lldd_dev_found(struct domain_device *);
 void sas_notify_lldd_dev_gone(struct domain_device *);
 
 int sas_smp_phy_control(struct domain_device *dev, int phy_id,
-			enum phy_func phy_func);
+			enum phy_func phy_func, struct sas_phy_linkrates *);
 int sas_smp_get_phy_events(struct sas_phy *phy);
 
 struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy);
diff --git a/drivers/scsi/libsas/sas_phy.c b/drivers/scsi/libsas/sas_phy.c
index 024ab00e70d2..9340cdbae4a3 100644
--- a/drivers/scsi/libsas/sas_phy.c
+++ b/drivers/scsi/libsas/sas_phy.c
@@ -67,13 +67,14 @@ static void sas_phye_oob_error(void *data)
 		switch (phy->error) {
 		case 1:
 		case 2:
-			i->dft->lldd_control_phy(phy, PHY_FUNC_HARD_RESET);
+			i->dft->lldd_control_phy(phy, PHY_FUNC_HARD_RESET,
+						 NULL);
 			break;
 		case 3:
 		default:
 			phy->error = 0;
 			phy->enabled = 0;
-			i->dft->lldd_control_phy(phy, PHY_FUNC_DISABLE);
+			i->dft->lldd_control_phy(phy, PHY_FUNC_DISABLE, NULL);
 			break;
 		}
 	}
@@ -90,7 +91,7 @@ static void sas_phye_spinup_hold(void *data)
 			&phy->phy_events_pending);
 
 	phy->error = 0;
-	i->dft->lldd_control_phy(phy, PHY_FUNC_RELEASE_SPINUP_HOLD);
+	i->dft->lldd_control_phy(phy, PHY_FUNC_RELEASE_SPINUP_HOLD, NULL);
 }
 
 /* ---------- Phy class registration ---------- */
@@ -144,10 +145,10 @@ int sas_register_phys(struct sas_ha_struct *sas_ha)
 		phy->phy->identify.target_port_protocols = phy->tproto;
 		phy->phy->identify.sas_address = SAS_ADDR(sas_ha->sas_addr);
 		phy->phy->identify.phy_identifier = i;
-		phy->phy->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS;
-		phy->phy->maximum_linkrate_hw = SAS_LINK_RATE_3_0_GBPS;
-		phy->phy->minimum_linkrate = SAS_LINK_RATE_1_5_GBPS;
-		phy->phy->maximum_linkrate = SAS_LINK_RATE_3_0_GBPS;
+		phy->phy->minimum_linkrate_hw = SAS_LINK_RATE_UNKNOWN;
+		phy->phy->maximum_linkrate_hw = SAS_LINK_RATE_UNKNOWN;
+		phy->phy->minimum_linkrate = SAS_LINK_RATE_UNKNOWN;
+		phy->phy->maximum_linkrate = SAS_LINK_RATE_UNKNOWN;
 		phy->phy->negotiated_linkrate = SAS_LINK_RATE_UNKNOWN;
 
 		sas_phy_add(phy->phy);
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index 8d91313dd888..8e39982fc3db 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -586,7 +586,7 @@ struct sas_domain_function_template {
 	int (*lldd_clear_nexus_ha)(struct sas_ha_struct *);
 
 	/* Phy management */
-	int (*lldd_control_phy)(struct asd_sas_phy *, enum phy_func);
+	int (*lldd_control_phy)(struct asd_sas_phy *, enum phy_func, void *);
 };
 
 extern int sas_register_ha(struct sas_ha_struct *);
diff --git a/include/scsi/sas.h b/include/scsi/sas.h
index 9c8a5b91ae64..2f4b6afa34fc 100644
--- a/include/scsi/sas.h
+++ b/include/scsi/sas.h
@@ -121,6 +121,7 @@ enum phy_func {
 	PHY_FUNC_CLEAR_AFFIL,
 	PHY_FUNC_TX_SATA_PS_SIGNAL,
 	PHY_FUNC_RELEASE_SPINUP_HOLD = 0x10, /* LOCAL PORT ONLY! */
+	PHY_FUNC_SET_LINK_RATE,
 };
 
 /* SAS LLDD would need to report only _very_few_ of those, like BROADCAST.
-- 
cgit v1.2.3


From 0024300000769eadcb4a4fcdff531d45ee7735d4 Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 6 Sep 2006 14:35:19 -0500
Subject: [POWERPC] powerpc: Divorce CPU_FTR_CTRL from
 CPU_FTR_PPCAS_ARCH_V2_BASE

The performance monitor implementation (including CTRL register behaviour)
is just included in PPC v2 as an example, it's not truly part of the base.

It's actually a somewhat misleading feature, but I'll leave that be for
now: The presence of the register is not what the feature bit is used
for, but instead it's used to determine if it contains the runlatch
bit for idle reporting of the performance monitor. For alternative
implementations, the register might still exist but the bit might have
different meaning (or no meaning at all).

For now, split it off and don't include it in CPU_FTR_PPCAS_ARCH_V2_BASE.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/cputable.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 748bc1805da9..3608259c49cf 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -148,7 +148,7 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 
 #define CPU_FTR_PPCAS_ARCH_V2_BASE (CPU_FTR_SLB | \
 					CPU_FTR_TLBIEL | CPU_FTR_NOEXECUTE | \
-					CPU_FTR_NODSISRALIGN | CPU_FTR_CTRL)
+					CPU_FTR_NODSISRALIGN)
 
 /* iSeries doesn't support large pages */
 #ifdef CONFIG_PPC_ISERIES
@@ -313,24 +313,25 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_IABR | \
 	    CPU_FTR_MMCRA | CPU_FTR_CTRL)
 #define CPU_FTRS_POWER4	(CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
-	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA)
+	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \
+	    CPU_FTR_MMCRA)
 #define CPU_FTRS_PPC970	(CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
-	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | \
+	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \
 	    CPU_FTR_ALTIVEC_COMP | CPU_FTR_CAN_NAP | CPU_FTR_MMCRA)
 #define CPU_FTRS_POWER5	(CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
-	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | \
+	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \
 	    CPU_FTR_MMCRA | CPU_FTR_SMT | \
 	    CPU_FTR_COHERENT_ICACHE | CPU_FTR_LOCKLESS_TLBIE | \
 	    CPU_FTR_PURR)
 #define CPU_FTRS_POWER6 (CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
-	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | \
+	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \
 	    CPU_FTR_MMCRA | CPU_FTR_SMT | \
 	    CPU_FTR_COHERENT_ICACHE | CPU_FTR_LOCKLESS_TLBIE | \
 	    CPU_FTR_PURR | CPU_FTR_CI_LARGE_PAGE | CPU_FTR_REAL_LE)
 #define CPU_FTRS_CELL	(CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
-	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | \
+	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \
 	    CPU_FTR_ALTIVEC_COMP | CPU_FTR_MMCRA | CPU_FTR_SMT | \
-	    CPU_FTR_CTRL | CPU_FTR_PAUSE_ZERO | CPU_FTR_CI_LARGE_PAGE)
+	    CPU_FTR_PAUSE_ZERO | CPU_FTR_CI_LARGE_PAGE)
 #define CPU_FTRS_COMPATIBLE	(CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2)
 #endif
-- 
cgit v1.2.3


From b3ebd1d862d6c23caa58e40d341eefc426f835e1 Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 6 Sep 2006 14:35:57 -0500
Subject: [POWERPC] powerpc: PA6T cputable entry, PVR value

Introduce PWRficient PA6T cputable entries and feature bits.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/cputable.c | 14 ++++++++++++++
 include/asm-powerpc/cputable.h |  9 +++++++--
 include/asm-powerpc/reg.h      |  1 +
 3 files changed, 22 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 306da4cd37a0..db65c9f6559a 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -58,6 +58,9 @@ extern void __restore_cpu_ppc970(void);
 #define COMMON_USER_POWER6	(COMMON_USER_PPC64 | PPC_FEATURE_ARCH_2_05 |\
 				 PPC_FEATURE_SMT | PPC_FEATURE_ICACHE_SNOOP | \
 				 PPC_FEATURE_TRUE_LE)
+#define COMMON_USER_PA6T	(COMMON_USER_PPC64 | PPC_FEATURE_PA6T |\
+				 PPC_FEATURE_TRUE_LE | \
+				 PPC_FEATURE_HAS_ALTIVEC_COMP)
 #define COMMON_USER_BOOKE	(PPC_FEATURE_32 | PPC_FEATURE_HAS_MMU | \
 				 PPC_FEATURE_BOOKE)
 
@@ -286,6 +289,17 @@ struct cpu_spec	cpu_specs[] = {
 		.dcache_bsize		= 128,
 		.platform		= "ppc-cell-be",
 	},
+	{	/* PA Semi PA6T */
+		.pvr_mask		= 0x7fff0000,
+		.pvr_value		= 0x00900000,
+		.cpu_name		= "PA6T",
+		.cpu_features		= CPU_FTRS_PA6T,
+		.cpu_user_features	= COMMON_USER_PA6T,
+		.icache_bsize		= 64,
+		.dcache_bsize		= 64,
+		.num_pmcs		= 6,
+		.platform		= "pa6t",
+	},
 	{	/* default match */
 		.pvr_mask		= 0x00000000,
 		.pvr_value		= 0x00000000,
diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 3608259c49cf..12707ab9dc98 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -23,6 +23,7 @@
 #define PPC_FEATURE_SMT			0x00004000
 #define PPC_FEATURE_ICACHE_SNOOP	0x00002000
 #define PPC_FEATURE_ARCH_2_05		0x00001000
+#define PPC_FEATURE_PA6T		0x00000800
 
 #define PPC_FEATURE_TRUE_LE		0x00000002
 #define PPC_FEATURE_PPC_LE		0x00000001
@@ -332,6 +333,10 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \
 	    CPU_FTR_ALTIVEC_COMP | CPU_FTR_MMCRA | CPU_FTR_SMT | \
 	    CPU_FTR_PAUSE_ZERO | CPU_FTR_CI_LARGE_PAGE)
+#define CPU_FTRS_PA6T (CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
+	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | \
+	    CPU_FTR_ALTIVEC_COMP | CPU_FTR_CI_LARGE_PAGE | \
+	    CPU_FTR_PURR | CPU_FTR_REAL_LE)
 #define CPU_FTRS_COMPATIBLE	(CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2)
 #endif
@@ -340,7 +345,7 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #define CPU_FTRS_POSSIBLE	\
 	    (CPU_FTRS_POWER3 | CPU_FTRS_RS64 | CPU_FTRS_POWER4 |	\
 	    CPU_FTRS_PPC970 | CPU_FTRS_POWER5 | CPU_FTRS_POWER6 |	\
-	    CPU_FTRS_CELL | CPU_FTR_CI_LARGE_PAGE)
+	    CPU_FTRS_CELL | CPU_FTRS_PA6T)
 #else
 enum {
 	CPU_FTRS_POSSIBLE =
@@ -379,7 +384,7 @@ enum {
 #define CPU_FTRS_ALWAYS		\
 	    (CPU_FTRS_POWER3 & CPU_FTRS_RS64 & CPU_FTRS_POWER4 &	\
 	    CPU_FTRS_PPC970 & CPU_FTRS_POWER5 & CPU_FTRS_POWER6 &	\
-	    CPU_FTRS_CELL & CPU_FTRS_POSSIBLE)
+	    CPU_FTRS_CELL & CPU_FTRS_PA6T & CPU_FTRS_POSSIBLE)
 #else
 enum {
 	CPU_FTRS_ALWAYS =
diff --git a/include/asm-powerpc/reg.h b/include/asm-powerpc/reg.h
index cf73475a0c69..3a9fcc15811b 100644
--- a/include/asm-powerpc/reg.h
+++ b/include/asm-powerpc/reg.h
@@ -592,6 +592,7 @@
 #define PV_630p	0x0041
 #define PV_970MP	0x0044
 #define PV_BE		0x0070
+#define PV_PA6T		0x0090
 
 /*
  * Number of entries in the SLB. If this ever changes we should handle
-- 
cgit v1.2.3


From 57852a853b0d6761f270be0961d5d8387e98c8bb Mon Sep 17 00:00:00 2001
From: Mike Kravetz <kravetz@us.ibm.com>
Date: Wed, 6 Sep 2006 16:23:12 -0700
Subject: [POWERPC] powerpc: Instrument Hypervisor Calls

Add instrumentation for hypervisor calls on pseries.  Call statistics
include number of calls, wall time and cpu cycles (if available) and
are made available via debugfs.  Instrumentation code is behind the
HCALL_STATS config option and has no impact if not enabled.

Signed-off-by: Mike Kravetz <kravetz@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/Kconfig.debug                   |  14 +++
 arch/powerpc/kernel/asm-offsets.c            |   7 ++
 arch/powerpc/platforms/pseries/Makefile      |   1 +
 arch/powerpc/platforms/pseries/hvCall.S      |  72 +++++++++++++++
 arch/powerpc/platforms/pseries/hvCall_inst.c | 129 +++++++++++++++++++++++++++
 include/asm-powerpc/hvcall.h                 |  12 ++-
 6 files changed, 234 insertions(+), 1 deletion(-)
 create mode 100644 arch/powerpc/platforms/pseries/hvCall_inst.c

(limited to 'include')

diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug
index e29ef77d3b00..d7b2aedd89aa 100644
--- a/arch/powerpc/Kconfig.debug
+++ b/arch/powerpc/Kconfig.debug
@@ -18,6 +18,20 @@ config DEBUG_STACK_USAGE
 
 	  This option will slow down process creation somewhat.
 
+config HCALL_STATS
+	bool "Hypervisor call instrumentation"
+	depends on PPC_PSERIES && DEBUG_FS
+	help
+	  Adds code to keep track of the number of hypervisor calls made and
+	  the amount of time spent in hypervisor callsr.  Wall time spent in
+	  each call is always calculated, and if available CPU cycles spent
+	  are also calculated.  A directory named hcall_inst is added at the
+	  root of the debugfs filesystem.  Within the hcall_inst directory
+	  are files that contain CPU specific call statistics.
+
+	  This option will add a small amount of overhead to all hypervisor
+	  calls.
+
 config DEBUGGER
 	bool "Enable debugger hooks"
 	depends on DEBUG_KERNEL
diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index c53acd2a6dfc..c578e7ab8173 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -137,6 +137,7 @@ int main(void)
 	DEFINE(PACA_USER_TIME, offsetof(struct paca_struct, user_time));
 	DEFINE(PACA_SYSTEM_TIME, offsetof(struct paca_struct, system_time));
 	DEFINE(PACA_SLBSHADOWPTR, offsetof(struct paca_struct, slb_shadow_ptr));
+	DEFINE(PACA_DATA_OFFSET, offsetof(struct paca_struct, data_offset));
 
 	DEFINE(SLBSHADOW_STACKVSID,
 	       offsetof(struct slb_shadow, save_area[SLB_NUM_BOLTED - 1].vsid));
@@ -165,6 +166,12 @@ int main(void)
 	/* Create extra stack space for SRR0 and SRR1 when calling prom/rtas. */
 	DEFINE(PROM_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs) + 16);
 	DEFINE(RTAS_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs) + 16);
+
+	/* hcall statistics */
+	DEFINE(HCALL_STAT_SIZE, sizeof(struct hcall_stats));
+	DEFINE(HCALL_STAT_CALLS, offsetof(struct hcall_stats, num_calls));
+	DEFINE(HCALL_STAT_TB, offsetof(struct hcall_stats, tb_total));
+	DEFINE(HCALL_STAT_PURR, offsetof(struct hcall_stats, purr_total));
 #endif /* CONFIG_PPC64 */
 	DEFINE(GPR0, STACK_FRAME_OVERHEAD+offsetof(struct pt_regs, gpr[0]));
 	DEFINE(GPR1, STACK_FRAME_OVERHEAD+offsetof(struct pt_regs, gpr[1]));
diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile
index e5e0ff466904..997243a91be8 100644
--- a/arch/powerpc/platforms/pseries/Makefile
+++ b/arch/powerpc/platforms/pseries/Makefile
@@ -12,3 +12,4 @@ obj-$(CONFIG_EEH)	+= eeh.o eeh_cache.o eeh_driver.o eeh_event.o
 
 obj-$(CONFIG_HVC_CONSOLE)	+= hvconsole.o
 obj-$(CONFIG_HVCS)		+= hvcserver.o
+obj-$(CONFIG_HCALL_STATS)	+= hvCall_inst.o
diff --git a/arch/powerpc/platforms/pseries/hvCall.S b/arch/powerpc/platforms/pseries/hvCall.S
index 9a99b056bd27..c00cfed7af2c 100644
--- a/arch/powerpc/platforms/pseries/hvCall.S
+++ b/arch/powerpc/platforms/pseries/hvCall.S
@@ -10,9 +10,69 @@
 #include <asm/hvcall.h>
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
 	
 #define STK_PARM(i)     (48 + ((i)-3)*8)
 
+#ifdef CONFIG_HCALL_STATS
+/*
+ * precall must preserve all registers.  use unused STK_PARM()
+ * areas to save snapshots and opcode.
+ */
+#define HCALL_INST_PRECALL					\
+	std	r3,STK_PARM(r3)(r1);	/* save opcode */	\
+	mftb	r0;			/* get timebase and */	\
+	std     r0,STK_PARM(r5)(r1);	/* save for later */	\
+BEGIN_FTR_SECTION;						\
+	mfspr	r0,SPRN_PURR;		/* get PURR and */	\
+	std	r0,STK_PARM(r6)(r1);	/* save for later */	\
+END_FTR_SECTION_IFCLR(CPU_FTR_PURR);
+	
+/*
+ * postcall is performed immediately before function return which
+ * allows liberal use of volatile registers.
+ */
+#define HCALL_INST_POSTCALL					\
+	ld	r4,STK_PARM(r3)(r1);	/* validate opcode */	\
+	cmpldi	cr7,r4,MAX_HCALL_OPCODE;			\
+	bgt-	cr7,1f;						\
+								\
+	/* get time and PURR snapshots after hcall */		\
+	mftb	r7;			/* timebase after */	\
+BEGIN_FTR_SECTION;						\
+	mfspr	r8,SPRN_PURR;		/* PURR after */	\
+	ld	r6,STK_PARM(r6)(r1);	/* PURR before */	\
+	subf	r6,r6,r8;		/* delta */		\
+END_FTR_SECTION_IFCLR(CPU_FTR_PURR);				\
+	ld	r5,STK_PARM(r5)(r1);	/* timebase before */	\
+	subf	r5,r5,r7;		/* time delta */	\
+								\
+	/* calculate address of stat structure r4 = opcode */	\
+	srdi	r4,r4,2;		/* index into array */	\
+	mulli	r4,r4,HCALL_STAT_SIZE;				\
+	LOAD_REG_ADDR(r7, per_cpu__hcall_stats);		\
+	add	r4,r4,r7;					\
+	ld	r7,PACA_DATA_OFFSET(r13); /* per cpu offset */	\
+	add	r4,r4,r7;					\
+								\
+	/* update stats	*/					\
+	ld	r7,HCALL_STAT_CALLS(r4); /* count */		\
+	addi	r7,r7,1;					\
+	std	r7,HCALL_STAT_CALLS(r4);			\
+	ld      r7,HCALL_STAT_TB(r4);	/* timebase */		\
+	add	r7,r7,r5;					\
+	std	r7,HCALL_STAT_TB(r4);				\
+BEGIN_FTR_SECTION;						\
+	ld	r7,HCALL_STAT_PURR(r4);	/* PURR */		\
+	add	r7,r7,r6;					\
+	std	r7,HCALL_STAT_PURR(r4);				\
+END_FTR_SECTION_IFCLR(CPU_FTR_PURR);				\
+1:
+#else
+#define HCALL_INST_PRECALL
+#define HCALL_INST_POSTCALL
+#endif
+
 	.text
 
 _GLOBAL(plpar_hcall_norets)
@@ -21,8 +81,12 @@ _GLOBAL(plpar_hcall_norets)
 	mfcr	r0
 	stw	r0,8(r1)
 
+	HCALL_INST_PRECALL
+
 	HVSC				/* invoke the hypervisor */
 
+	HCALL_INST_POSTCALL
+
 	lwz	r0,8(r1)
 	mtcrf	0xff,r0
 	blr				/* return r3 = status */
@@ -33,6 +97,8 @@ _GLOBAL(plpar_hcall)
 	mfcr	r0
 	stw	r0,8(r1)
 
+	HCALL_INST_PRECALL
+
 	std     r4,STK_PARM(r4)(r1)     /* Save ret buffer */
 
 	mr	r4,r5
@@ -50,6 +116,8 @@ _GLOBAL(plpar_hcall)
 	std	r6, 16(r12)
 	std	r7, 24(r12)
 
+	HCALL_INST_POSTCALL
+
 	lwz	r0,8(r1)
 	mtcrf	0xff,r0
 
@@ -61,6 +129,8 @@ _GLOBAL(plpar_hcall9)
 	mfcr	r0
 	stw	r0,8(r1)
 
+	HCALL_INST_PRECALL
+
 	std     r4,STK_PARM(r4)(r1)     /* Save ret buffer */
 
 	mr	r4,r5
@@ -86,6 +156,8 @@ _GLOBAL(plpar_hcall9)
 	std	r11,56(r12)
 	std	r12,64(r12)
 
+	HCALL_INST_POSTCALL
+
 	lwz	r0,8(r1)
 	mtcrf	0xff,r0
 
diff --git a/arch/powerpc/platforms/pseries/hvCall_inst.c b/arch/powerpc/platforms/pseries/hvCall_inst.c
new file mode 100644
index 000000000000..641e6511cf06
--- /dev/null
+++ b/arch/powerpc/platforms/pseries/hvCall_inst.c
@@ -0,0 +1,129 @@
+/*
+ * Copyright (C) 2006 Mike Kravetz IBM Corporation
+ *
+ * Hypervisor Call Instrumentation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/cpumask.h>
+#include <asm/hvcall.h>
+#include <asm/firmware.h>
+#include <asm/cputable.h>
+
+DEFINE_PER_CPU(struct hcall_stats[HCALL_STAT_ARRAY_SIZE], hcall_stats);
+
+/*
+ * Routines for displaying the statistics in debugfs
+ */
+static void *hc_start(struct seq_file *m, loff_t *pos)
+{
+	if ((int)*pos < HCALL_STAT_ARRAY_SIZE)
+		return (void *)(unsigned long)(*pos + 1);
+
+	return NULL;
+}
+
+static void *hc_next(struct seq_file *m, void *p, loff_t * pos)
+{
+	++*pos;
+
+	return hc_start(m, pos);
+}
+
+static void hc_stop(struct seq_file *m, void *p)
+{
+}
+
+static int hc_show(struct seq_file *m, void *p)
+{
+	unsigned long h_num = (unsigned long)p;
+	struct hcall_stats *hs = (struct hcall_stats *)m->private;
+
+	if (hs[h_num].num_calls) {
+		if (!cpu_has_feature(CPU_FTR_PURR))
+			seq_printf(m, "%lu %lu %lu %lu\n", h_num<<2,
+				   hs[h_num].num_calls,
+				   hs[h_num].tb_total,
+				   hs[h_num].purr_total);
+		else
+			seq_printf(m, "%lu %lu %lu\n", h_num<<2,
+				   hs[h_num].num_calls,
+				   hs[h_num].tb_total);
+	}
+
+	return 0;
+}
+
+static struct seq_operations hcall_inst_seq_ops = {
+        .start = hc_start,
+        .next  = hc_next,
+        .stop  = hc_stop,
+        .show  = hc_show
+};
+
+static int hcall_inst_seq_open(struct inode *inode, struct file *file)
+{
+	int rc;
+	struct seq_file *seq;
+
+	rc = seq_open(file, &hcall_inst_seq_ops);
+	seq = file->private_data;
+	seq->private = file->f_dentry->d_inode->u.generic_ip;
+
+	return rc;
+}
+
+static struct file_operations hcall_inst_seq_fops = {
+	.open = hcall_inst_seq_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = seq_release,
+};
+
+#define	HCALL_ROOT_DIR		"hcall_inst"
+#define CPU_NAME_BUF_SIZE	32
+
+static int __init hcall_inst_init(void)
+{
+	struct dentry *hcall_root;
+	struct dentry *hcall_file;
+	char cpu_name_buf[CPU_NAME_BUF_SIZE];
+	int cpu;
+
+	if (!firmware_has_feature(FW_FEATURE_LPAR))
+		return 0;
+
+	hcall_root = debugfs_create_dir(HCALL_ROOT_DIR, NULL);
+	if (!hcall_root)
+		return -ENOMEM;
+
+	for_each_possible_cpu(cpu) {
+		snprintf(cpu_name_buf, CPU_NAME_BUF_SIZE, "cpu%d", cpu);
+		hcall_file = debugfs_create_file(cpu_name_buf, S_IRUGO,
+						 hcall_root,
+						 per_cpu(hcall_stats, cpu),
+						 &hcall_inst_seq_fops);
+		if (!hcall_file)
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+__initcall(hcall_inst_init);
diff --git a/include/asm-powerpc/hvcall.h b/include/asm-powerpc/hvcall.h
index 63ce1ac8c1f4..257d1cecb8c9 100644
--- a/include/asm-powerpc/hvcall.h
+++ b/include/asm-powerpc/hvcall.h
@@ -208,7 +208,7 @@
 #define H_JOIN			0x298
 #define H_VASI_STATE            0x2A4
 #define H_ENABLE_CRQ		0x2B0
-#define MAX_HCALL_OPCODES	(H_ENABLE_CRQ >> 2)
+#define MAX_HCALL_OPCODE	H_ENABLE_CRQ
 
 #ifndef __ASSEMBLY__
 
@@ -246,6 +246,16 @@ long plpar_hcall(unsigned long opcode, unsigned long *retbuf, ...);
 #define PLPAR_HCALL9_BUFSIZE 9
 long plpar_hcall9(unsigned long opcode, unsigned long *retbuf, ...);
 
+/* For hcall instrumentation.  One structure per-hcall, per-CPU */
+struct hcall_stats {
+	unsigned long	num_calls;	/* number of calls (on this CPU) */
+	unsigned long	tb_total;	/* total wall time (mftb) of calls. */
+	unsigned long	purr_total;	/* total cpu time (PURR) of calls. */
+};
+void update_hcall_stats(unsigned long opcode, unsigned long tb_delta,
+			unsigned long purr_delta);
+#define HCALL_STAT_ARRAY_SIZE	((MAX_HCALL_OPCODE >> 2) + 1)
+
 #endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_HVCALL_H */
-- 
cgit v1.2.3


From 76fd85937097a0c2ec8ab23bf21dc10992d1c398 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Fri, 8 Sep 2006 11:16:13 -0700
Subject: [PATCH] ethtool: allow const ethtool_ops

The ethtool_ops structure is immutable, it expected to be setup
by the driver and is never changed. This patch allows drivers to
declare there ethtool_ops structure read-only.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/linux/netdevice.h |  2 +-
 net/core/ethtool.c        | 14 +++++++-------
 2 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 50a4719512ed..a2e747353367 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -342,7 +342,7 @@ struct net_device
 	/* Instance data managed by the core of Wireless Extensions. */
 	struct iw_public_data *	wireless_data;
 
-	struct ethtool_ops *ethtool_ops;
+	const struct ethtool_ops *ethtool_ops;
 
 	/*
 	 * This marks the end of the "visible" part of the structure. All
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 2797e2815418..e0ca04f38cef 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -143,7 +143,7 @@ static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
 static int ethtool_get_drvinfo(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_drvinfo info;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 
 	if (!ops->get_drvinfo)
 		return -EOPNOTSUPP;
@@ -169,7 +169,7 @@ static int ethtool_get_drvinfo(struct net_device *dev, void __user *useraddr)
 static int ethtool_get_regs(struct net_device *dev, char __user *useraddr)
 {
 	struct ethtool_regs regs;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 	void *regbuf;
 	int reglen, ret;
 
@@ -282,7 +282,7 @@ static int ethtool_get_link(struct net_device *dev, void __user *useraddr)
 static int ethtool_get_eeprom(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_eeprom eeprom;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 	u8 *data;
 	int ret;
 
@@ -327,7 +327,7 @@ static int ethtool_get_eeprom(struct net_device *dev, void __user *useraddr)
 static int ethtool_set_eeprom(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_eeprom eeprom;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 	u8 *data;
 	int ret;
 
@@ -640,7 +640,7 @@ static int ethtool_set_gso(struct net_device *dev, char __user *useraddr)
 static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
 {
 	struct ethtool_test test;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 	u64 *data;
 	int ret;
 
@@ -673,7 +673,7 @@ static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
 static int ethtool_get_strings(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_gstrings gstrings;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 	u8 *data;
 	int ret;
 
@@ -733,7 +733,7 @@ static int ethtool_phys_id(struct net_device *dev, void __user *useraddr)
 static int ethtool_get_stats(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_stats stats;
-	struct ethtool_ops *ops = dev->ethtool_ops;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
 	u64 *data;
 	int ret;
 
-- 
cgit v1.2.3


From 132919ba80ad207755fe271277bfefff865a54fe Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 27 Aug 2006 13:56:52 +0100
Subject: [MMC] Remove data->blksz_bits member

data->blksz_bits is unused now - remove it.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mmc/mmc.c       | 1 -
 drivers/mmc/mmc_block.c | 1 -
 include/linux/mmc/mmc.h | 1 -
 3 files changed, 3 deletions(-)

(limited to 'include')

diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c
index 74eaaee66de0..5b9caa7978d3 100644
--- a/drivers/mmc/mmc.c
+++ b/drivers/mmc/mmc.c
@@ -996,7 +996,6 @@ static void mmc_read_scrs(struct mmc_host *host)
 
 		mmc_set_data_timeout(&data, card, 0);
 
-		data.blksz_bits = 3;
 		data.blksz = 1 << 3;
 		data.blocks = 1;
 		data.flags = MMC_DATA_READ;
diff --git a/drivers/mmc/mmc_block.c b/drivers/mmc/mmc_block.c
index a0e0dad1b419..f3a99dd8df8f 100644
--- a/drivers/mmc/mmc_block.c
+++ b/drivers/mmc/mmc_block.c
@@ -172,7 +172,6 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 
 		brq.cmd.arg = req->sector << 9;
 		brq.cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC;
-		brq.data.blksz_bits = md->block_bits;
 		brq.data.blksz = 1 << md->block_bits;
 		brq.data.blocks = req->nr_sectors >> (md->block_bits - 9);
 		brq.stop.opcode = MMC_STOP_TRANSMISSION;
diff --git a/include/linux/mmc/mmc.h b/include/linux/mmc/mmc.h
index 627e2c08ce41..a3594dfd6963 100644
--- a/include/linux/mmc/mmc.h
+++ b/include/linux/mmc/mmc.h
@@ -68,7 +68,6 @@ struct mmc_command {
 struct mmc_data {
 	unsigned int		timeout_ns;	/* data timeout (in ns, max 80ms) */
 	unsigned int		timeout_clks;	/* data timeout (in clocks) */
-	unsigned int		blksz_bits;	/* data block size */
 	unsigned int		blksz;		/* data block size */
 	unsigned int		blocks;		/* number of blocks */
 	unsigned int		error;		/* data error */
-- 
cgit v1.2.3


From db53f28b3a6d9338cca1b7e917dc063ac99e1871 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 30 Aug 2006 15:14:56 +0100
Subject: [MMC] Add multi block-write capability

Add a capability flag for drivers to set when they can perform multi-
block transfers to cards _and_ correctly report the number of bytes
transferred should an error occur.

The last point is very important - if a driver reports more bytes than
were actually accepted by the card and an error occurs, there is the
possibility for data loss.

Pierre Ossman provided the patch for wbsd and sdhci.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mmc/mmc_block.c  | 27 ++++++++++++++++++++-------
 drivers/mmc/mmci.c       |  1 +
 drivers/mmc/sdhci.c      |  2 +-
 drivers/mmc/wbsd.c       |  2 +-
 include/linux/mmc/host.h |  1 +
 5 files changed, 24 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/mmc/mmc_block.c b/drivers/mmc/mmc_block.c
index f3a99dd8df8f..8d18b87bfd34 100644
--- a/drivers/mmc/mmc_block.c
+++ b/drivers/mmc/mmc_block.c
@@ -32,6 +32,7 @@
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/protocol.h>
+#include <linux/mmc/host.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -165,6 +166,7 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 	do {
 		struct mmc_blk_request brq;
 		struct mmc_command cmd;
+		u32 readcmd, writecmd;
 
 		memset(&brq, 0, sizeof(struct mmc_blk_request));
 		brq.mrq.cmd = &brq.cmd;
@@ -180,20 +182,31 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 
 		mmc_set_data_timeout(&brq.data, card, rq_data_dir(req) != READ);
 
-		if (rq_data_dir(req) == READ) {
-			brq.cmd.opcode = brq.data.blocks > 1 ? MMC_READ_MULTIPLE_BLOCK : MMC_READ_SINGLE_BLOCK;
-			brq.data.flags |= MMC_DATA_READ;
-		} else {
-			brq.cmd.opcode = MMC_WRITE_BLOCK;
-			brq.data.flags |= MMC_DATA_WRITE;
+		/*
+		 * If the host doesn't support multiple block writes, force
+		 * block writes to single block.
+		 */
+		if (rq_data_dir(req) != READ &&
+		    !(card->host->caps & MMC_CAP_MULTIWRITE))
 			brq.data.blocks = 1;
-		}
 
 		if (brq.data.blocks > 1) {
 			brq.data.flags |= MMC_DATA_MULTI;
 			brq.mrq.stop = &brq.stop;
+			readcmd = MMC_READ_MULTIPLE_BLOCK;
+			writecmd = MMC_WRITE_MULTIPLE_BLOCK;
 		} else {
 			brq.mrq.stop = NULL;
+			readcmd = MMC_READ_SINGLE_BLOCK;
+			writecmd = MMC_WRITE_BLOCK;
+		}
+
+		if (rq_data_dir(req) == READ) {
+			brq.cmd.opcode = readcmd;
+			brq.data.flags |= MMC_DATA_READ;
+		} else {
+			brq.cmd.opcode = writecmd;
+			brq.data.flags |= MMC_DATA_WRITE;
 		}
 
 		brq.data.sg = mq->sg;
diff --git a/drivers/mmc/mmci.c b/drivers/mmc/mmci.c
index 8419489e7744..2b5a0cc9ea56 100644
--- a/drivers/mmc/mmci.c
+++ b/drivers/mmc/mmci.c
@@ -509,6 +509,7 @@ static int mmci_probe(struct amba_device *dev, void *id)
 	mmc->f_min = (host->mclk + 511) / 512;
 	mmc->f_max = min(host->mclk, fmax);
 	mmc->ocr_avail = plat->ocr_mask;
+	mmc->caps = MMC_CAP_MULTIWRITE;
 
 	/*
 	 * We can do SGIO
diff --git a/drivers/mmc/sdhci.c b/drivers/mmc/sdhci.c
index 4e21b3b9d330..dea4edd1c434 100644
--- a/drivers/mmc/sdhci.c
+++ b/drivers/mmc/sdhci.c
@@ -1262,7 +1262,7 @@ static int __devinit sdhci_probe_slot(struct pci_dev *pdev, int slot)
 	mmc->ops = &sdhci_ops;
 	mmc->f_min = host->max_clk / 256;
 	mmc->f_max = host->max_clk;
-	mmc->caps = MMC_CAP_4_BIT_DATA;
+	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
 
 	mmc->ocr_avail = 0;
 	if (caps & SDHCI_CAN_VDD_330)
diff --git a/drivers/mmc/wbsd.c b/drivers/mmc/wbsd.c
index c351c6d1a18a..4a6617d9a49f 100644
--- a/drivers/mmc/wbsd.c
+++ b/drivers/mmc/wbsd.c
@@ -1323,7 +1323,7 @@ static int __devinit wbsd_alloc_mmc(struct device *dev)
 	mmc->f_min = 375000;
 	mmc->f_max = 24000000;
 	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps = MMC_CAP_4_BIT_DATA;
+	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
 
 	spin_lock_init(&host->lock);
 
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index ba095aebedff..b282ec9bba08 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -85,6 +85,7 @@ struct mmc_host {
 	unsigned long		caps;		/* Host capabilities */
 
 #define MMC_CAP_4_BIT_DATA	(1 << 0)	/* Can the host do 4 bit transfers */
+#define MMC_CAP_MULTIWRITE	(1 << 1)	/* Can accurately report bytes sent to card on error */
 
 	/* host specific block data */
 	unsigned int		max_seg_size;	/* see blk_queue_max_segment_size */
-- 
cgit v1.2.3


From fea63e38013ec628ab3f7fddc4c2148064b7910a Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sat, 16 Sep 2006 03:04:15 +0900
Subject: [PATCH] libata: fix non-uniform ports handling

Non-uniform ports handling got broken while updating libata to handle
those in the same host.  Only separate irq for the non-uniform
secondary port was implemented while all other fields (host flags,
transfer mode...) of the secondary port simply shared those of the
first.

For ata_piix combined mode, which ATM is the only user of non-uniform
ports, this causes the secondary port assume the wrong type.  This can
cause PATA port to use SATA ops, which results in bogus check on PCS
and detection failure.

This patch adds ata_probe_ent->pinfo2 which points to optional
port_info for the secondary port.  For the time being, this seems to
be the simplest solution.  This workaround will be removed together
with ata_probe_ent itself after init model is updated to allow more
flexibility.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Nelson A. de Oliveira <naoliv@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 18 +++++++++++++-----
 drivers/ata/libata-sff.c  |  2 ++
 include/linux/libata.h    |  8 ++++++++
 3 files changed, 23 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 1c9315401f7a..bb66a12c84e5 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5269,11 +5269,19 @@ void ata_port_init(struct ata_port *ap, struct ata_host *host,
 	ap->host = host;
 	ap->dev = ent->dev;
 	ap->port_no = port_no;
-	ap->pio_mask = ent->pio_mask;
-	ap->mwdma_mask = ent->mwdma_mask;
-	ap->udma_mask = ent->udma_mask;
-	ap->flags |= ent->port_flags;
-	ap->ops = ent->port_ops;
+	if (port_no == 1 && ent->pinfo2) {
+		ap->pio_mask = ent->pinfo2->pio_mask;
+		ap->mwdma_mask = ent->pinfo2->mwdma_mask;
+		ap->udma_mask = ent->pinfo2->udma_mask;
+		ap->flags |= ent->pinfo2->flags;
+		ap->ops = ent->pinfo2->port_ops;
+	} else {
+		ap->pio_mask = ent->pio_mask;
+		ap->mwdma_mask = ent->mwdma_mask;
+		ap->udma_mask = ent->udma_mask;
+		ap->flags |= ent->port_flags;
+		ap->ops = ent->port_ops;
+	}
 	ap->hw_sata_spd_limit = UINT_MAX;
 	ap->active_tag = ATA_TAG_POISON;
 	ap->last_ctl = 0xFF;
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index d51dc41fa195..688bb55e197a 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -858,6 +858,7 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int
 			probe_ent->port[p].bmdma_addr = bmdma;
 		}
 		ata_std_ports(&probe_ent->port[p]);
+		probe_ent->pinfo2 = port[1];
 		p++;
 	}
 
@@ -907,6 +908,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 				probe_ent->_host_flags |= ATA_HOST_SIMPLEX;
 		}
 		ata_std_ports(&probe_ent->port[1]);
+		probe_ent->pinfo2 = port[1];
 	} else
 		probe_ent->dummy_port_mask |= ATA_PORT_SECONDARY;
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 8715305f611f..ff67e7524fe9 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -361,6 +361,14 @@ struct ata_probe_ent {
 	unsigned long		_host_flags;
 	void __iomem		*mmio_base;
 	void			*private_data;
+
+	/* port_info for the secondary port.  Together with irq2, it's
+	 * used to implement non-uniform secondary port.  Currently,
+	 * the only user is ata_piix combined mode.  This workaround
+	 * will be removed together with ata_probe_ent when init model
+	 * is updated.
+	 */
+	const struct ata_port_info *pinfo2;
 };
 
 struct ata_host {
-- 
cgit v1.2.3


From 93590859884784520a1850767f86296abc2cdc6d Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 12 Sep 2006 16:55:12 +0100
Subject: [PATCH] libata: improve handling of diagostic fail (and hardware that
 misreports it)

Our ATA probe code checks that a device is not reporting a diagnostic
failure during start up. Unfortunately at least one device seems to like
doing this - the Gigabyte iRAM.

This is only done for the master right now (which is fine for the iRAM
as it is SATA), as with PATA some combinations of ATAPI device seem to
fool the check into seeing a drive that isn't there if it is applied to
the slave.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 19 +++++++++++++++++--
 include/linux/libata.h    |  6 ++++++
 2 files changed, 23 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index e85c2f8cf193..753b0152afd1 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -616,8 +616,11 @@ ata_dev_try_classify(struct ata_port *ap, unsigned int device, u8 *r_err)
 	if (r_err)
 		*r_err = err;
 
-	/* see if device passed diags */
-	if (err == 1)
+	/* see if device passed diags: if master then continue and warn later */
+	if (err == 0 && device == 0)
+		/* diagnostic fail : do nothing _YET_ */
+		ap->device[device].horkage |= ATA_HORKAGE_DIAGNOSTIC;
+	else if (err == 1)
 		/* do nothing */ ;
 	else if ((device == 0) && (err == 0x81))
 		/* do nothing */ ;
@@ -1523,6 +1526,18 @@ int ata_dev_configure(struct ata_device *dev, int print_info)
 				       cdb_intr_string);
 	}
 
+	if (dev->horkage & ATA_HORKAGE_DIAGNOSTIC) {
+		/* Let the user know. We don't want to disallow opens for
+		   rescue purposes, or in case the vendor is just a blithering
+		   idiot */
+                if (print_info) {
+			ata_dev_printk(dev, KERN_WARNING,
+"Drive reports diagnostics failure. This may indicate a drive\n");
+			ata_dev_printk(dev, KERN_WARNING,
+"fault or invalid emulation. Contact drive vendor for information.\n");
+		}
+	}
+
 	ata_set_port_max_cmd_len(ap);
 
 	/* limit bridge transfers to udma5, 200 sectors */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index ff67e7524fe9..1ef3d3901b47 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -289,6 +289,11 @@ enum {
 	 * most devices.
 	 */
 	ATA_SPINUP_WAIT		= 8000,
+	
+	/* Horkage types. May be set by libata or controller on drives
+	   (some horkage may be drive/controller pair dependant */
+
+	ATA_HORKAGE_DIAGNOSTIC	= (1 << 0),	/* Failed boot diag */
 };
 
 enum hsm_task_states {
@@ -476,6 +481,7 @@ struct ata_device {
 
 	/* error history */
 	struct ata_ering	ering;
+	unsigned int		horkage;	/* List of broken features */
 };
 
 /* Offset into struct ata_device.  Fields above it are maintained
-- 
cgit v1.2.3


From fadcfa33b6319a5faf8af2287f08bf93a7f926b6 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Tue, 19 Sep 2006 12:43:58 +0100
Subject: [HEADERS] One line per header in Kbuild files to reduce conflicts

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/Kbuild                        |  11 +-
 include/asm-alpha/Kbuild              |  10 +-
 include/asm-generic/Kbuild            |  15 +-
 include/asm-generic/Kbuild.asm        |  38 +++-
 include/asm-i386/Kbuild               |   9 +-
 include/asm-ia64/Kbuild               |  18 +-
 include/asm-powerpc/Kbuild            |  45 +++-
 include/asm-s390/Kbuild               |  11 +-
 include/asm-sparc/Kbuild              |  24 +-
 include/asm-sparc64/Kbuild            |  27 ++-
 include/asm-x86_64/Kbuild             |  18 +-
 include/linux/Kbuild                  | 400 +++++++++++++++++++++++++++++-----
 include/linux/byteorder/Kbuild        |   9 +-
 include/linux/dvb/Kbuild              |  11 +-
 include/linux/netfilter/Kbuild        |  47 +++-
 include/linux/netfilter_arp/Kbuild    |   5 +-
 include/linux/netfilter_bridge/Kbuild |  21 +-
 include/linux/netfilter_ipv4/Kbuild   |  82 +++++--
 include/linux/netfilter_ipv6/Kbuild   |  27 ++-
 include/linux/nfsd/Kbuild             |   9 +-
 include/linux/raid/Kbuild             |   3 +-
 include/linux/sunrpc/Kbuild           |   2 +-
 include/linux/tc_act/Kbuild           |   5 +-
 include/linux/tc_ematch/Kbuild        |   5 +-
 include/mtd/Kbuild                    |   8 +-
 include/rdma/Kbuild                   |   2 +-
 include/scsi/Kbuild                   |   4 +-
 include/sound/Kbuild                  |  12 +-
 include/video/Kbuild                  |   2 +-
 29 files changed, 721 insertions(+), 159 deletions(-)

(limited to 'include')

diff --git a/include/Kbuild b/include/Kbuild
index cb2534800b19..2d03f995865f 100644
--- a/include/Kbuild
+++ b/include/Kbuild
@@ -1,2 +1,9 @@
-header-y += asm-generic/ linux/ scsi/ sound/ mtd/ rdma/ video/
-header-y += asm-$(ARCH)/ 
+header-y += asm-generic/
+header-y += linux/
+header-y += scsi/
+header-y += sound/
+header-y += mtd/
+header-y += rdma/
+header-y += video/
+
+header-y += asm-$(ARCH)/
diff --git a/include/asm-alpha/Kbuild b/include/asm-alpha/Kbuild
index 2b06b3bad5ff..b7c8f188b313 100644
--- a/include/asm-alpha/Kbuild
+++ b/include/asm-alpha/Kbuild
@@ -1,5 +1,11 @@
 include include/asm-generic/Kbuild.asm
 
-unifdef-y += console.h fpu.h sysinfo.h compiler.h
+header-y += gentrap.h
+header-y += regdef.h
+header-y += pal.h
+header-y += reg.h
 
-header-y += gentrap.h regdef.h pal.h reg.h
+unifdef-y += console.h
+unifdef-y += fpu.h
+unifdef-y += sysinfo.h
+unifdef-y += compiler.h
diff --git a/include/asm-generic/Kbuild b/include/asm-generic/Kbuild
index 70594b275a6e..3c06be381701 100644
--- a/include/asm-generic/Kbuild
+++ b/include/asm-generic/Kbuild
@@ -1,3 +1,12 @@
-header-y += atomic.h errno-base.h errno.h fcntl.h ioctl.h ipc.h mman.h \
-	signal.h statfs.h
-unifdef-y := resource.h siginfo.h
+header-y += atomic.h
+header-y += errno-base.h
+header-y += errno.h
+header-y += fcntl.h
+header-y += ioctl.h
+header-y += ipc.h
+header-y += mman.h
+header-y += signal.h
+header-y += statfs.h
+
+unifdef-y += resource.h
+unifdef-y += siginfo.h
diff --git a/include/asm-generic/Kbuild.asm b/include/asm-generic/Kbuild.asm
index c00de6028fa8..a84c3d88a189 100644
--- a/include/asm-generic/Kbuild.asm
+++ b/include/asm-generic/Kbuild.asm
@@ -1,8 +1,34 @@
-unifdef-y += a.out.h auxvec.h byteorder.h errno.h fcntl.h ioctl.h	\
-	ioctls.h ipcbuf.h mman.h msgbuf.h param.h poll.h		\
-	posix_types.h ptrace.h resource.h sembuf.h shmbuf.h shmparam.h	\
-	sigcontext.h siginfo.h signal.h socket.h sockios.h stat.h	\
-	statfs.h termbits.h termios.h types.h unistd.h user.h
+unifdef-y += a.out.h
+unifdef-y += auxvec.h
+unifdef-y += byteorder.h
+unifdef-y += errno.h
+unifdef-y += fcntl.h
+unifdef-y += ioctl.h
+unifdef-y += ioctls.h
+unifdef-y += ipcbuf.h
+unifdef-y += mman.h
+unifdef-y += msgbuf.h
+unifdef-y += param.h
+unifdef-y += poll.h
+unifdef-y += posix_types.h
+unifdef-y += ptrace.h
+unifdef-y += resource.h
+unifdef-y += sembuf.h
+unifdef-y += shmbuf.h
+unifdef-y += sigcontext.h
+unifdef-y += siginfo.h
+unifdef-y += signal.h
+unifdef-y += socket.h
+unifdef-y += sockios.h
+unifdef-y += stat.h
+unifdef-y += statfs.h
+unifdef-y += termbits.h
+unifdef-y += termios.h
+unifdef-y += types.h
+unifdef-y += unistd.h
+unifdef-y += user.h
 
 # These probably shouldn't be exported
-unifdef-y += elf.h page.h
+unifdef-y += shmparam.h
+unifdef-y += elf.h
+unifdef-y += page.h
diff --git a/include/asm-i386/Kbuild b/include/asm-i386/Kbuild
index 2308190321da..b75a348d0c1c 100644
--- a/include/asm-i386/Kbuild
+++ b/include/asm-i386/Kbuild
@@ -1,5 +1,10 @@
 include include/asm-generic/Kbuild.asm
 
-header-y += boot.h debugreg.h ldt.h ucontext.h
+header-y += boot.h
+header-y += debugreg.h
+header-y += ldt.h
+header-y += ucontext.h
 
-unifdef-y += mtrr.h setup.h vm86.h
+unifdef-y += mtrr.h
+unifdef-y += setup.h
+unifdef-y += vm86.h
diff --git a/include/asm-ia64/Kbuild b/include/asm-ia64/Kbuild
index f1cb00f39c22..15818a18bc52 100644
--- a/include/asm-ia64/Kbuild
+++ b/include/asm-ia64/Kbuild
@@ -1,7 +1,17 @@
 include include/asm-generic/Kbuild.asm
 
-header-y += break.h fpu.h fpswa.h gcc_intrin.h ia64regs.h		\
-	 intel_intrin.h intrinsics.h perfmon_default_smpl.h	\
-	 ptrace_offsets.h rse.h setup.h ucontext.h
+header-y += break.h
+header-y += fpu.h
+header-y += fpswa.h
+header-y += gcc_intrin.h
+header-y += ia64regs.h
+header-y += intel_intrin.h
+header-y += intrinsics.h
+header-y += perfmon_default_smpl.h
+header-y += ptrace_offsets.h
+header-y += rse.h
+header-y += setup.h
+header-y += ucontext.h
 
-unifdef-y += perfmon.h ustack.h
+unifdef-y += perfmon.h
+unifdef-y += ustack.h
diff --git a/include/asm-powerpc/Kbuild b/include/asm-powerpc/Kbuild
index ac61d7eb6021..9827849953a3 100644
--- a/include/asm-powerpc/Kbuild
+++ b/include/asm-powerpc/Kbuild
@@ -1,10 +1,41 @@
 include include/asm-generic/Kbuild.asm
 
-unifdef-y += a.out.h asm-compat.h bootx.h byteorder.h cputable.h elf.h	\
-	nvram.h param.h posix_types.h ptrace.h seccomp.h signal.h	\
-	termios.h types.h unistd.h
+header-y += auxvec.h
+header-y += ioctls.h
+header-y += mman.h
+header-y += sembuf.h
+header-y += siginfo.h
+header-y += stat.h
+header-y += errno.h
+header-y += ipcbuf.h
+header-y += msgbuf.h
+header-y += shmbuf.h
+header-y += socket.h
+header-y += termbits.h
+header-y += fcntl.h
+header-y += ipc.h
+header-y += poll.h
+header-y += shmparam.h
+header-y += sockios.h
+header-y += ucontext.h
+header-y += ioctl.h
+header-y += linkage.h
+header-y += resource.h
+header-y += sigcontext.h
+header-y += statfs.h
 
-header-y += auxvec.h ioctls.h mman.h sembuf.h siginfo.h stat.h errno.h	\
-	ipcbuf.h msgbuf.h shmbuf.h socket.h termbits.h fcntl.h ipc.h	\
-	poll.h shmparam.h sockios.h ucontext.h ioctl.h linkage.h	\
-	resource.h sigcontext.h statfs.h
+unifdef-y += a.out.h
+unifdef-y += asm-compat.h
+unifdef-y += bootx.h
+unifdef-y += byteorder.h
+unifdef-y += cputable.h
+unifdef-y += elf.h
+unifdef-y += nvram.h
+unifdef-y += param.h
+unifdef-y += posix_types.h
+unifdef-y += ptrace.h
+unifdef-y += seccomp.h
+unifdef-y += signal.h
+unifdef-y += termios.h
+unifdef-y += types.h
+unifdef-y += unistd.h
diff --git a/include/asm-s390/Kbuild b/include/asm-s390/Kbuild
index ed8955f49e47..14158a4a9c87 100644
--- a/include/asm-s390/Kbuild
+++ b/include/asm-s390/Kbuild
@@ -1,4 +1,11 @@
 include include/asm-generic/Kbuild.asm
 
-unifdef-y += cmb.h debug.h
-header-y += dasd.h qeth.h tape390.h ucontext.h vtoc.h z90crypt.h
+header-y += dasd.h
+header-y += qeth.h
+header-y += tape390.h
+header-y += ucontext.h
+header-y += vtoc.h
+header-y += z90crypt.h
+
+unifdef-y += cmb.h
+unifdef-y += debug.h
diff --git a/include/asm-sparc/Kbuild b/include/asm-sparc/Kbuild
index e2a57fd7abfa..b22b67a64ecc 100644
--- a/include/asm-sparc/Kbuild
+++ b/include/asm-sparc/Kbuild
@@ -1,6 +1,22 @@
 include include/asm-generic/Kbuild.asm
 
-unifdef-y += fbio.h perfctr.h psr.h
-header-y += apc.h asi.h auxio.h bpp.h head.h ipc.h jsflash.h	\
-	openpromio.h pbm.h pconf.h pgtsun4.h reg.h traps.h	\
-	turbosparc.h vfc_ioctls.h winmacro.h
+header-y += apc.h
+header-y += asi.h
+header-y += auxio.h
+header-y += bpp.h
+header-y += head.h
+header-y += ipc.h
+header-y += jsflash.h
+header-y += openpromio.h
+header-y += pbm.h
+header-y += pconf.h
+header-y += pgtsun4.h
+header-y += reg.h
+header-y += traps.h
+header-y += turbosparc.h
+header-y += vfc_ioctls.h
+header-y += winmacro.h
+
+unifdef-y += fbio.h
+unifdef-y += perfctr.h
+unifdef-y += psr.h
diff --git a/include/asm-sparc64/Kbuild b/include/asm-sparc64/Kbuild
index 9284c3cb27ec..4b59ce46cc2d 100644
--- a/include/asm-sparc64/Kbuild
+++ b/include/asm-sparc64/Kbuild
@@ -4,7 +4,26 @@ ALTARCH := sparc
 ARCHDEF := defined __sparc__ && defined __arch64__
 ALTARCHDEF := defined __sparc__ && !defined __arch64__
 
-unifdef-y += fbio.h perfctr.h
-header-y += apb.h asi.h bbc.h bpp.h display7seg.h envctrl.h floppy.h	\
-	ipc.h kdebug.h mostek.h openprom.h openpromio.h parport.h	\
-	pconf.h psrcompat.h pstate.h reg.h uctx.h utrap.h watchdog.h
+header-y += apb.h
+header-y += asi.h
+header-y += bbc.h
+header-y += bpp.h
+header-y += display7seg.h
+header-y += envctrl.h
+header-y += floppy.h
+header-y += ipc.h
+header-y += kdebug.h
+header-y += mostek.h
+header-y += openprom.h
+header-y += openpromio.h
+header-y += parport.h
+header-y += pconf.h
+header-y += psrcompat.h
+header-y += pstate.h
+header-y += reg.h
+header-y += uctx.h
+header-y += utrap.h
+header-y += watchdog.h
+
+unifdef-y += fbio.h
+unifdef-y += perfctr.h
diff --git a/include/asm-x86_64/Kbuild b/include/asm-x86_64/Kbuild
index dc4d101e8a16..40f2f13fe174 100644
--- a/include/asm-x86_64/Kbuild
+++ b/include/asm-x86_64/Kbuild
@@ -4,8 +4,18 @@ ALTARCH := i386
 ARCHDEF := defined __x86_64__
 ALTARCHDEF := defined __i386__
 
-header-y += boot.h bootsetup.h cpufeature.h debugreg.h ldt.h \
-	 msr.h prctl.h setup.h sigcontext32.h ucontext.h \
-	 vsyscall32.h
+header-y += boot.h
+header-y += bootsetup.h
+header-y += cpufeature.h
+header-y += debugreg.h
+header-y += ldt.h
+header-y += msr.h
+header-y += prctl.h
+header-y += setup.h
+header-y += sigcontext32.h
+header-y += ucontext.h
+header-y += vsyscall32.h
 
-unifdef-y += mce.h mtrr.h vsyscall.h
+unifdef-y += mce.h
+unifdef-y += mtrr.h
+unifdef-y += vsyscall.h
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 2b8a7d68fae3..7d076d97b2f7 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -1,63 +1,343 @@
-header-y := byteorder/ dvb/ hdlc/ isdn/ nfsd/ raid/ sunrpc/ tc_act/	\
-	netfilter/ netfilter_arp/ netfilter_bridge/ netfilter_ipv4/	\
-	netfilter_ipv6/
+header-y += byteorder/
+header-y += dvb/
+header-y += hdlc/
+header-y += isdn/
+header-y += nfsd/
+header-y += raid/
+header-y += sunrpc/
+header-y += tc_act/
+header-y += netfilter/
+header-y += netfilter_arp/
+header-y += netfilter_bridge/
+header-y += netfilter_ipv4/
+header-y += netfilter_ipv6/
 
-header-y += affs_fs.h affs_hardblocks.h aio_abi.h a.out.h arcfb.h	\
-	atmapi.h atmbr2684.h atmclip.h atm_eni.h atm_he.h		\
-	atm_idt77105.h atmioc.h atmlec.h atmmpc.h atm_nicstar.h		\
-	atmppp.h atmsap.h atmsvc.h atm_zatm.h auto_fs4.h auxvec.h	\
-	awe_voice.h ax25.h b1lli.h baycom.h bfs_fs.h blkpg.h		\
-	bpqether.h cdk.h chio.h coda_psdev.h coff.h comstats.h		\
-	consolemap.h cycx_cfm.h dm-ioctl.h dn.h dqblk_v1.h		\
-	dqblk_v2.h dqblk_xfs.h efs_fs_sb.h elf-fdpic.h elf.h elf-em.h	\
-	fadvise.h fd.h fdreg.h ftape-header-segment.h ftape-vendors.h	\
-	fuse.h futex.h genetlink.h gen_stats.h gigaset_dev.h hdsmart.h	\
-	hpfs_fs.h hysdn_if.h i2c-dev.h i8k.h icmp.h			\
-	if_arcnet.h if_arp.h if_bonding.h if_cablemodem.h if_fc.h	\
-	if_fddi.h if.h if_hippi.h if_infiniband.h if_packet.h		\
-	if_plip.h if_ppp.h if_slip.h if_strip.h if_tunnel.h in6.h	\
-	in_route.h ioctl.h ip.h ipmi_msgdefs.h ip_mp_alg.h ipsec.h	\
-	ipx.h irda.h isdn_divertif.h iso_fs.h ite_gpio.h ixjuser.h	\
-	jffs2.h keyctl.h limits.h major.h matroxfb.h meye.h minix_fs.h	\
-	mmtimer.h mqueue.h mtio.h ncp_no.h netfilter_arp.h netrom.h	\
-	nfs2.h nfs4_mount.h nfs_mount.h openprom_fs.h param.h		\
-	pci_ids.h pci_regs.h personality.h pfkeyv2.h pg.h pkt_cls.h	\
-	pkt_sched.h posix_types.h ppdev.h prctl.h ps2esdi.h qic117.h	\
-	qnxtypes.h quotaio_v1.h quotaio_v2.h radeonfb.h raw.h		\
-	resource.h rose.h sctp.h smbno.h snmp.h sockios.h som.h		\
-	sound.h stddef.h synclink.h telephony.h termios.h ticable.h	\
-	times.h tiocl.h tipc.h toshiba.h ultrasound.h un.h utime.h	\
-	utsname.h video_decoder.h video_encoder.h videotext.h vt.h	\
-	wavefront.h wireless.h xattr.h x25.h zorro_ids.h
+header-y += affs_fs.h
+header-y += affs_hardblocks.h
+header-y += aio_abi.h
+header-y += a.out.h
+header-y += arcfb.h
+header-y += atmapi.h
+header-y += atmbr2684.h
+header-y += atmclip.h
+header-y += atm_eni.h
+header-y += atm_he.h
+header-y += atm_idt77105.h
+header-y += atmioc.h
+header-y += atmlec.h
+header-y += atmmpc.h
+header-y += atm_nicstar.h
+header-y += atmppp.h
+header-y += atmsap.h
+header-y += atmsvc.h
+header-y += atm_zatm.h
+header-y += auto_fs4.h
+header-y += auxvec.h
+header-y += awe_voice.h
+header-y += ax25.h
+header-y += b1lli.h
+header-y += baycom.h
+header-y += bfs_fs.h
+header-y += blkpg.h
+header-y += bpqether.h
+header-y += cdk.h
+header-y += chio.h
+header-y += coda_psdev.h
+header-y += coff.h
+header-y += comstats.h
+header-y += consolemap.h
+header-y += cycx_cfm.h
+header-y += dm-ioctl.h
+header-y += dn.h
+header-y += dqblk_v1.h
+header-y += dqblk_v2.h
+header-y += dqblk_xfs.h
+header-y += efs_fs_sb.h
+header-y += elf-fdpic.h
+header-y += elf.h
+header-y += elf-em.h
+header-y += fadvise.h
+header-y += fd.h
+header-y += fdreg.h
+header-y += ftape-header-segment.h
+header-y += ftape-vendors.h
+header-y += fuse.h
+header-y += futex.h
+header-y += genetlink.h
+header-y += gen_stats.h
+header-y += gigaset_dev.h
+header-y += hdsmart.h
+header-y += hpfs_fs.h
+header-y += hysdn_if.h
+header-y += i2c-dev.h
+header-y += i8k.h
+header-y += icmp.h
+header-y += if_arcnet.h
+header-y += if_arp.h
+header-y += if_bonding.h
+header-y += if_cablemodem.h
+header-y += if_fc.h
+header-y += if_fddi.h
+header-y += if.h
+header-y += if_hippi.h
+header-y += if_infiniband.h
+header-y += if_packet.h
+header-y += if_plip.h
+header-y += if_ppp.h
+header-y += if_slip.h
+header-y += if_strip.h
+header-y += if_tunnel.h
+header-y += in6.h
+header-y += in_route.h
+header-y += ioctl.h
+header-y += ip.h
+header-y += ipmi_msgdefs.h
+header-y += ip_mp_alg.h
+header-y += ipsec.h
+header-y += ipx.h
+header-y += irda.h
+header-y += isdn_divertif.h
+header-y += iso_fs.h
+header-y += ite_gpio.h
+header-y += ixjuser.h
+header-y += jffs2.h
+header-y += keyctl.h
+header-y += limits.h
+header-y += major.h
+header-y += matroxfb.h
+header-y += meye.h
+header-y += minix_fs.h
+header-y += mmtimer.h
+header-y += mqueue.h
+header-y += mtio.h
+header-y += ncp_no.h
+header-y += netfilter_arp.h
+header-y += netrom.h
+header-y += nfs2.h
+header-y += nfs4_mount.h
+header-y += nfs_mount.h
+header-y += openprom_fs.h
+header-y += param.h
+header-y += pci_ids.h
+header-y += pci_regs.h
+header-y += personality.h
+header-y += pfkeyv2.h
+header-y += pg.h
+header-y += pkt_cls.h
+header-y += pkt_sched.h
+header-y += posix_types.h
+header-y += ppdev.h
+header-y += prctl.h
+header-y += ps2esdi.h
+header-y += qic117.h
+header-y += qnxtypes.h
+header-y += quotaio_v1.h
+header-y += quotaio_v2.h
+header-y += radeonfb.h
+header-y += raw.h
+header-y += resource.h
+header-y += rose.h
+header-y += sctp.h
+header-y += smbno.h
+header-y += snmp.h
+header-y += sockios.h
+header-y += som.h
+header-y += sound.h
+header-y += stddef.h
+header-y += synclink.h
+header-y += telephony.h
+header-y += termios.h
+header-y += ticable.h
+header-y += times.h
+header-y += tiocl.h
+header-y += tipc.h
+header-y += toshiba.h
+header-y += ultrasound.h
+header-y += un.h
+header-y += utime.h
+header-y += utsname.h
+header-y += video_decoder.h
+header-y += video_encoder.h
+header-y += videotext.h
+header-y += vt.h
+header-y += wavefront.h
+header-y += wireless.h
+header-y += xattr.h
+header-y += x25.h
+header-y += zorro_ids.h
 
-unifdef-y += acct.h adb.h adfs_fs.h agpgart.h apm_bios.h atalk.h	\
-	atmarp.h atmdev.h atm.h atm_tcp.h audit.h auto_fs.h binfmts.h	\
-	capability.h capi.h cciss_ioctl.h cdrom.h cm4000_cs.h		\
-	cn_proc.h coda.h connector.h cramfs_fs.h cuda.h cyclades.h	\
-	dccp.h dirent.h divert.h elfcore.h errno.h errqueue.h		\
-	ethtool.h eventpoll.h ext2_fs.h ext3_fs.h fb.h fcntl.h		\
-	filter.h flat.h fs.h ftape.h gameport.h generic_serial.h	\
-	genhd.h hayesesp.h hdlcdrv.h hdlc.h hdreg.h hiddev.h hpet.h	\
-	i2c.h i2o-dev.h icmpv6.h if_bridge.h if_ec.h			\
-	if_eql.h if_ether.h if_frad.h if_ltalk.h if_pppox.h		\
-	if_shaper.h if_tr.h if_tun.h if_vlan.h if_wanpipe.h igmp.h	\
-	inet_diag.h in.h inotify.h input.h ipc.h ipmi.h ipv6.h		\
-	ipv6_route.h isdn.h isdnif.h isdn_ppp.h isicom.h jbd.h		\
-	joystick.h kdev_t.h kd.h kernelcapi.h kernel.h keyboard.h	\
-	llc.h loop.h lp.h mempolicy.h mii.h mman.h mroute.h msdos_fs.h	\
-	msg.h nbd.h ncp_fs.h ncp.h ncp_mount.h netdevice.h		\
-	netfilter_bridge.h netfilter_decnet.h netfilter.h		\
-	netfilter_ipv4.h netfilter_ipv6.h netfilter_logging.h net.h	\
-	netlink.h nfs3.h nfs4.h nfsacl.h nfs_fs.h nfs.h nfs_idmap.h	\
-	n_r3964.h nubus.h nvram.h parport.h patchkey.h pci.h pktcdvd.h	\
-	pmu.h poll.h ppp_defs.h ppp-comp.h ptrace.h qnx4_fs.h quota.h	\
-	random.h reboot.h reiserfs_fs.h reiserfs_xattr.h romfs_fs.h	\
-	route.h rtc.h rtnetlink.h scc.h sched.h sdla.h			\
-	selinux_netlink.h sem.h serial_core.h serial.h serio.h shm.h	\
-	signal.h smb_fs.h smb.h smb_mount.h socket.h sonet.h sonypi.h	\
-	soundcard.h stat.h sysctl.h tcp.h time.h timex.h tty.h types.h	\
-	udf_fs_i.h udp.h uinput.h uio.h unistd.h usb_ch9.h		\
-	usbdevice_fs.h user.h videodev2.h videodev.h wait.h		\
-	wanrouter.h watchdog.h xfrm.h zftape.h
+unifdef-y += acct.h
+unifdef-y += adb.h
+unifdef-y += adfs_fs.h
+unifdef-y += agpgart.h
+unifdef-y += apm_bios.h
+unifdef-y += atalk.h
+unifdef-y += atmarp.h
+unifdef-y += atmdev.h
+unifdef-y += atm.h
+unifdef-y += atm_tcp.h
+unifdef-y += audit.h
+unifdef-y += auto_fs.h
+unifdef-y += binfmts.h
+unifdef-y += capability.h
+unifdef-y += capi.h
+unifdef-y += cciss_ioctl.h
+unifdef-y += cdrom.h
+unifdef-y += cm4000_cs.h
+unifdef-y += cn_proc.h
+unifdef-y += coda.h
+unifdef-y += connector.h
+unifdef-y += cramfs_fs.h
+unifdef-y += cuda.h
+unifdef-y += cyclades.h
+unifdef-y += dccp.h
+unifdef-y += dirent.h
+unifdef-y += divert.h
+unifdef-y += elfcore.h
+unifdef-y += errno.h
+unifdef-y += errqueue.h
+unifdef-y += ethtool.h
+unifdef-y += eventpoll.h
+unifdef-y += ext2_fs.h
+unifdef-y += ext3_fs.h
+unifdef-y += fb.h
+unifdef-y += fcntl.h
+unifdef-y += filter.h
+unifdef-y += flat.h
+unifdef-y += fs.h
+unifdef-y += ftape.h
+unifdef-y += gameport.h
+unifdef-y += generic_serial.h
+unifdef-y += genhd.h
+unifdef-y += hayesesp.h
+unifdef-y += hdlcdrv.h
+unifdef-y += hdlc.h
+unifdef-y += hdreg.h
+unifdef-y += hiddev.h
+unifdef-y += hpet.h
+unifdef-y += i2c.h
+unifdef-y += i2o-dev.h
+unifdef-y += icmpv6.h
+unifdef-y += if_bridge.h
+unifdef-y += if_ec.h
+unifdef-y += if_eql.h
+unifdef-y += if_ether.h
+unifdef-y += if_frad.h
+unifdef-y += if_ltalk.h
+unifdef-y += if_pppox.h
+unifdef-y += if_shaper.h
+unifdef-y += if_tr.h
+unifdef-y += if_tun.h
+unifdef-y += if_vlan.h
+unifdef-y += if_wanpipe.h
+unifdef-y += igmp.h
+unifdef-y += inet_diag.h
+unifdef-y += in.h
+unifdef-y += inotify.h
+unifdef-y += input.h
+unifdef-y += ipc.h
+unifdef-y += ipmi.h
+unifdef-y += ipv6.h
+unifdef-y += ipv6_route.h
+unifdef-y += isdn.h
+unifdef-y += isdnif.h
+unifdef-y += isdn_ppp.h
+unifdef-y += isicom.h
+unifdef-y += jbd.h
+unifdef-y += joystick.h
+unifdef-y += kdev_t.h
+unifdef-y += kd.h
+unifdef-y += kernelcapi.h
+unifdef-y += kernel.h
+unifdef-y += keyboard.h
+unifdef-y += llc.h
+unifdef-y += loop.h
+unifdef-y += lp.h
+unifdef-y += mempolicy.h
+unifdef-y += mii.h
+unifdef-y += mman.h
+unifdef-y += mroute.h
+unifdef-y += msdos_fs.h
+unifdef-y += msg.h
+unifdef-y += nbd.h
+unifdef-y += ncp_fs.h
+unifdef-y += ncp.h
+unifdef-y += ncp_mount.h
+unifdef-y += netdevice.h
+unifdef-y += netfilter_bridge.h
+unifdef-y += netfilter_decnet.h
+unifdef-y += netfilter.h
+unifdef-y += netfilter_ipv4.h
+unifdef-y += netfilter_ipv6.h
+unifdef-y += netfilter_logging.h
+unifdef-y += net.h
+unifdef-y += netlink.h
+unifdef-y += nfs3.h
+unifdef-y += nfs4.h
+unifdef-y += nfsacl.h
+unifdef-y += nfs_fs.h
+unifdef-y += nfs.h
+unifdef-y += nfs_idmap.h
+unifdef-y += n_r3964.h
+unifdef-y += nubus.h
+unifdef-y += nvram.h
+unifdef-y += parport.h
+unifdef-y += patchkey.h
+unifdef-y += pci.h
+unifdef-y += pktcdvd.h
+unifdef-y += pmu.h
+unifdef-y += poll.h
+unifdef-y += ppp_defs.h
+unifdef-y += ppp-comp.h
+unifdef-y += ptrace.h
+unifdef-y += qnx4_fs.h
+unifdef-y += quota.h
+unifdef-y += random.h
+unifdef-y += reboot.h
+unifdef-y += reiserfs_fs.h
+unifdef-y += reiserfs_xattr.h
+unifdef-y += romfs_fs.h
+unifdef-y += route.h
+unifdef-y += rtc.h
+unifdef-y += rtnetlink.h
+unifdef-y += scc.h
+unifdef-y += sched.h
+unifdef-y += sdla.h
+unifdef-y += selinux_netlink.h
+unifdef-y += sem.h
+unifdef-y += serial_core.h
+unifdef-y += serial.h
+unifdef-y += serio.h
+unifdef-y += shm.h
+unifdef-y += signal.h
+unifdef-y += smb_fs.h
+unifdef-y += smb.h
+unifdef-y += smb_mount.h
+unifdef-y += socket.h
+unifdef-y += sonet.h
+unifdef-y += sonypi.h
+unifdef-y += soundcard.h
+unifdef-y += stat.h
+unifdef-y += sysctl.h
+unifdef-y += tcp.h
+unifdef-y += time.h
+unifdef-y += timex.h
+unifdef-y += tty.h
+unifdef-y += types.h
+unifdef-y += udf_fs_i.h
+unifdef-y += udp.h
+unifdef-y += uinput.h
+unifdef-y += uio.h
+unifdef-y += unistd.h
+unifdef-y += usb_ch9.h
+unifdef-y += usbdevice_fs.h
+unifdef-y += user.h
+unifdef-y += videodev2.h
+unifdef-y += videodev.h
+unifdef-y += wait.h
+unifdef-y += wanrouter.h
+unifdef-y += watchdog.h
+unifdef-y += xfrm.h
+unifdef-y += zftape.h
 
-objhdr-y := version.h
+objhdr-y += version.h
diff --git a/include/linux/byteorder/Kbuild b/include/linux/byteorder/Kbuild
index 84a57d4fb212..56499ab9e32e 100644
--- a/include/linux/byteorder/Kbuild
+++ b/include/linux/byteorder/Kbuild
@@ -1,2 +1,7 @@
-unifdef-y += generic.h swabb.h swab.h
-header-y += big_endian.h little_endian.h pdp_endian.h
+header-y += big_endian.h
+header-y += little_endian.h
+header-y += pdp_endian.h
+
+unifdef-y += generic.h
+unifdef-y += swabb.h
+unifdef-y += swab.h
diff --git a/include/linux/dvb/Kbuild b/include/linux/dvb/Kbuild
index 63973af72fd5..d97b3a51e227 100644
--- a/include/linux/dvb/Kbuild
+++ b/include/linux/dvb/Kbuild
@@ -1,2 +1,9 @@
-header-y += ca.h frontend.h net.h osd.h version.h
-unifdef-y := audio.h dmx.h video.h
+header-y += ca.h
+header-y += frontend.h
+header-y += net.h
+header-y += osd.h
+header-y += version.h
+
+unifdef-y += audio.h
+unifdef-y += dmx.h
+unifdef-y += video.h
diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index 1d3a14e2da6e..9a285cecf249 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -1,11 +1,38 @@
-header-y := nf_conntrack_sctp.h nf_conntrack_tuple_common.h		\
-	    nfnetlink_conntrack.h nfnetlink_log.h nfnetlink_queue.h	\
-	    xt_CLASSIFY.h xt_comment.h xt_connbytes.h xt_connmark.h	\
-	    xt_CONNMARK.h xt_conntrack.h xt_dccp.h xt_esp.h		\
-	    xt_helper.h xt_length.h xt_limit.h xt_mac.h xt_mark.h	\
-	    xt_MARK.h xt_multiport.h xt_NFQUEUE.h xt_pkttype.h		\
-	    xt_policy.h xt_realm.h xt_sctp.h xt_state.h xt_string.h	\
-	    xt_tcpmss.h xt_tcpudp.h xt_SECMARK.h xt_CONNSECMARK.h
+header-y += nf_conntrack_sctp.h
+header-y += nf_conntrack_tuple_common.h
+header-y += nfnetlink_conntrack.h
+header-y += nfnetlink_log.h
+header-y += nfnetlink_queue.h
+header-y += xt_CLASSIFY.h
+header-y += xt_comment.h
+header-y += xt_connbytes.h
+header-y += xt_connmark.h
+header-y += xt_CONNMARK.h
+header-y += xt_conntrack.h
+header-y += xt_dccp.h
+header-y += xt_esp.h
+header-y += xt_helper.h
+header-y += xt_length.h
+header-y += xt_limit.h
+header-y += xt_mac.h
+header-y += xt_mark.h
+header-y += xt_MARK.h
+header-y += xt_multiport.h
+header-y += xt_NFQUEUE.h
+header-y += xt_pkttype.h
+header-y += xt_policy.h
+header-y += xt_realm.h
+header-y += xt_sctp.h
+header-y += xt_state.h
+header-y += xt_string.h
+header-y += xt_tcpmss.h
+header-y += xt_tcpudp.h
+header-y += xt_SECMARK.h
+header-y += xt_CONNSECMARK.h
 
-unifdef-y := nf_conntrack_common.h nf_conntrack_ftp.h		\
-	nf_conntrack_tcp.h nfnetlink.h x_tables.h xt_physdev.h
+unifdef-y += nf_conntrack_common.h
+unifdef-y += nf_conntrack_ftp.h
+unifdef-y += nf_conntrack_tcp.h
+unifdef-y += nfnetlink.h
+unifdef-y += x_tables.h
+unifdef-y += xt_physdev.h
diff --git a/include/linux/netfilter_arp/Kbuild b/include/linux/netfilter_arp/Kbuild
index 198ec5e7b17d..4f13dfcb92ea 100644
--- a/include/linux/netfilter_arp/Kbuild
+++ b/include/linux/netfilter_arp/Kbuild
@@ -1,2 +1,3 @@
-header-y := arpt_mangle.h
-unifdef-y := arp_tables.h
+header-y += arpt_mangle.h
+
+unifdef-y += arp_tables.h
diff --git a/include/linux/netfilter_bridge/Kbuild b/include/linux/netfilter_bridge/Kbuild
index 5b1aba6abbad..76ff4c47d8c4 100644
--- a/include/linux/netfilter_bridge/Kbuild
+++ b/include/linux/netfilter_bridge/Kbuild
@@ -1,4 +1,17 @@
-header-y += ebt_among.h ebt_arp.h ebt_arpreply.h ebt_ip.h ebt_limit.h	\
-	ebt_log.h ebt_mark_m.h ebt_mark_t.h ebt_nat.h ebt_pkttype.h	\
-	ebt_redirect.h ebt_stp.h ebt_ulog.h ebt_vlan.h
-unifdef-y := ebtables.h ebt_802_3.h
+header-y += ebt_among.h
+header-y += ebt_arp.h
+header-y += ebt_arpreply.h
+header-y += ebt_ip.h
+header-y += ebt_limit.h
+header-y += ebt_log.h
+header-y += ebt_mark_m.h
+header-y += ebt_mark_t.h
+header-y += ebt_nat.h
+header-y += ebt_pkttype.h
+header-y += ebt_redirect.h
+header-y += ebt_stp.h
+header-y += ebt_ulog.h
+header-y += ebt_vlan.h
+
+unifdef-y += ebtables.h
+unifdef-y += ebt_802_3.h
diff --git a/include/linux/netfilter_ipv4/Kbuild b/include/linux/netfilter_ipv4/Kbuild
index 04e4d2721689..591c1a809c00 100644
--- a/include/linux/netfilter_ipv4/Kbuild
+++ b/include/linux/netfilter_ipv4/Kbuild
@@ -1,21 +1,63 @@
+header-y += ip_conntrack_helper.h
+header-y += ip_conntrack_helper_h323_asn1.h
+header-y += ip_conntrack_helper_h323_types.h
+header-y += ip_conntrack_protocol.h
+header-y += ip_conntrack_sctp.h
+header-y += ip_conntrack_tcp.h
+header-y += ip_conntrack_tftp.h
+header-y += ip_nat_pptp.h
+header-y += ipt_addrtype.h
+header-y += ipt_ah.h
+header-y += ipt_CLASSIFY.h
+header-y += ipt_CLUSTERIP.h
+header-y += ipt_comment.h
+header-y += ipt_connbytes.h
+header-y += ipt_connmark.h
+header-y += ipt_CONNMARK.h
+header-y += ipt_conntrack.h
+header-y += ipt_dccp.h
+header-y += ipt_dscp.h
+header-y += ipt_DSCP.h
+header-y += ipt_ecn.h
+header-y += ipt_ECN.h
+header-y += ipt_esp.h
+header-y += ipt_hashlimit.h
+header-y += ipt_helper.h
+header-y += ipt_iprange.h
+header-y += ipt_length.h
+header-y += ipt_limit.h
+header-y += ipt_LOG.h
+header-y += ipt_mac.h
+header-y += ipt_mark.h
+header-y += ipt_MARK.h
+header-y += ipt_multiport.h
+header-y += ipt_NFQUEUE.h
+header-y += ipt_owner.h
+header-y += ipt_physdev.h
+header-y += ipt_pkttype.h
+header-y += ipt_policy.h
+header-y += ipt_realm.h
+header-y += ipt_recent.h
+header-y += ipt_REJECT.h
+header-y += ipt_SAME.h
+header-y += ipt_sctp.h
+header-y += ipt_state.h
+header-y += ipt_string.h
+header-y += ipt_tcpmss.h
+header-y += ipt_TCPMSS.h
+header-y += ipt_tos.h
+header-y += ipt_TOS.h
+header-y += ipt_ttl.h
+header-y += ipt_TTL.h
+header-y += ipt_ULOG.h
 
-header-y := ip_conntrack_helper.h ip_conntrack_helper_h323_asn1.h	\
-	    ip_conntrack_helper_h323_types.h ip_conntrack_protocol.h	\
-	    ip_conntrack_sctp.h ip_conntrack_tcp.h ip_conntrack_tftp.h	\
-	    ip_nat_pptp.h ipt_addrtype.h ipt_ah.h	\
-	    ipt_CLASSIFY.h ipt_CLUSTERIP.h ipt_comment.h		\
-	    ipt_connbytes.h ipt_connmark.h ipt_CONNMARK.h		\
-	    ipt_conntrack.h ipt_dccp.h ipt_dscp.h ipt_DSCP.h ipt_ecn.h	\
-	    ipt_ECN.h ipt_esp.h ipt_hashlimit.h ipt_helper.h		\
-	    ipt_iprange.h ipt_length.h ipt_limit.h ipt_LOG.h ipt_mac.h	\
-	    ipt_mark.h ipt_MARK.h ipt_multiport.h ipt_NFQUEUE.h		\
-	    ipt_owner.h ipt_physdev.h ipt_pkttype.h ipt_policy.h	\
-	    ipt_realm.h ipt_recent.h ipt_REJECT.h ipt_SAME.h		\
-	    ipt_sctp.h ipt_state.h ipt_string.h ipt_tcpmss.h		\
-	    ipt_TCPMSS.h ipt_tos.h ipt_TOS.h ipt_ttl.h ipt_TTL.h	\
-	    ipt_ULOG.h
-
-unifdef-y := ip_conntrack.h ip_conntrack_h323.h ip_conntrack_irc.h	\
-	ip_conntrack_pptp.h ip_conntrack_proto_gre.h			\
-	ip_conntrack_tuple.h ip_nat.h ip_nat_rule.h ip_queue.h		\
-	ip_tables.h
+unifdef-y += ip_conntrack.h
+unifdef-y += ip_conntrack_h323.h
+unifdef-y += ip_conntrack_irc.h
+unifdef-y += ip_conntrack_pptp.h
+unifdef-y += ip_conntrack_proto_gre.h
+unifdef-y += ip_conntrack_tuple.h
+unifdef-y += ip_nat.h
+unifdef-y += ip_nat_rule.h
+unifdef-y += ip_queue.h
+unifdef-y += ip_tables.h
diff --git a/include/linux/netfilter_ipv6/Kbuild b/include/linux/netfilter_ipv6/Kbuild
index 913ddbf55b4b..9dd978d149ff 100644
--- a/include/linux/netfilter_ipv6/Kbuild
+++ b/include/linux/netfilter_ipv6/Kbuild
@@ -1,6 +1,21 @@
-header-y += ip6t_HL.h ip6t_LOG.h ip6t_MARK.h ip6t_REJECT.h ip6t_ah.h	\
-	ip6t_esp.h ip6t_frag.h ip6t_hl.h ip6t_ipv6header.h		\
-	ip6t_length.h ip6t_limit.h ip6t_mac.h ip6t_mark.h		\
-	ip6t_multiport.h ip6t_opts.h ip6t_owner.h ip6t_policy.h		\
-	ip6t_physdev.h ip6t_rt.h
-unifdef-y := ip6_tables.h
+header-y += ip6t_HL.h
+header-y += ip6t_LOG.h
+header-y += ip6t_MARK.h
+header-y += ip6t_REJECT.h
+header-y += ip6t_ah.h
+header-y += ip6t_esp.h
+header-y += ip6t_frag.h
+header-y += ip6t_hl.h
+header-y += ip6t_ipv6header.h
+header-y += ip6t_length.h
+header-y += ip6t_limit.h
+header-y += ip6t_mac.h
+header-y += ip6t_mark.h
+header-y += ip6t_multiport.h
+header-y += ip6t_opts.h
+header-y += ip6t_owner.h
+header-y += ip6t_policy.h
+header-y += ip6t_physdev.h
+header-y += ip6t_rt.h
+
+unifdef-y += ip6_tables.h
diff --git a/include/linux/nfsd/Kbuild b/include/linux/nfsd/Kbuild
index c8c545665885..d9c5455808e5 100644
--- a/include/linux/nfsd/Kbuild
+++ b/include/linux/nfsd/Kbuild
@@ -1,2 +1,7 @@
-unifdef-y := const.h export.h stats.h syscall.h nfsfh.h debug.h auth.h
-
+unifdef-y += const.h
+unifdef-y += export.h
+unifdef-y += stats.h
+unifdef-y += syscall.h
+unifdef-y += nfsfh.h
+unifdef-y += debug.h
+unifdef-y += auth.h
diff --git a/include/linux/raid/Kbuild b/include/linux/raid/Kbuild
index 73fa27a8d552..2415a64c5e51 100644
--- a/include/linux/raid/Kbuild
+++ b/include/linux/raid/Kbuild
@@ -1 +1,2 @@
-header-y += md_p.h md_u.h
+header-y += md_p.h
+header-y += md_u.h
diff --git a/include/linux/sunrpc/Kbuild b/include/linux/sunrpc/Kbuild
index 0d1d768a27bf..fb438f158eee 100644
--- a/include/linux/sunrpc/Kbuild
+++ b/include/linux/sunrpc/Kbuild
@@ -1 +1 @@
-unifdef-y := debug.h
+unifdef-y += debug.h
diff --git a/include/linux/tc_act/Kbuild b/include/linux/tc_act/Kbuild
index 5251a505b2f1..78dfbac36375 100644
--- a/include/linux/tc_act/Kbuild
+++ b/include/linux/tc_act/Kbuild
@@ -1 +1,4 @@
-header-y += tc_gact.h tc_ipt.h tc_mirred.h tc_pedit.h
+header-y += tc_gact.h
+header-y += tc_ipt.h
+header-y += tc_mirred.h
+header-y += tc_pedit.h
diff --git a/include/linux/tc_ematch/Kbuild b/include/linux/tc_ematch/Kbuild
index 381e93018df6..4a58a1c32a00 100644
--- a/include/linux/tc_ematch/Kbuild
+++ b/include/linux/tc_ematch/Kbuild
@@ -1 +1,4 @@
-headers-y := tc_em_cmp.h tc_em_meta.h tc_em_nbyte.h tc_em_text.h
+header-y += tc_em_cmp.h
+header-y += tc_em_meta.h
+header-y += tc_em_nbyte.h
+header-y += tc_em_text.h
diff --git a/include/mtd/Kbuild b/include/mtd/Kbuild
index e1da2a5b2a57..13e7a3c6d794 100644
--- a/include/mtd/Kbuild
+++ b/include/mtd/Kbuild
@@ -1,2 +1,6 @@
-unifdef-y := mtd-abi.h
-header-y := inftl-user.h jffs2-user.h mtd-user.h nftl-user.h
+header-y += inftl-user.h
+header-y += jffs2-user.h
+header-y += mtd-user.h
+header-y += nftl-user.h
+
+unifdef-y += mtd-abi.h
diff --git a/include/rdma/Kbuild b/include/rdma/Kbuild
index eb710ba9b1a0..e7c043216558 100644
--- a/include/rdma/Kbuild
+++ b/include/rdma/Kbuild
@@ -1 +1 @@
-header-y := ib_user_mad.h
+header-y += ib_user_mad.h
diff --git a/include/scsi/Kbuild b/include/scsi/Kbuild
index 14a033d73314..744f85011f1e 100644
--- a/include/scsi/Kbuild
+++ b/include/scsi/Kbuild
@@ -1,2 +1,4 @@
 header-y += scsi.h
-unifdef-y := scsi_ioctl.h sg.h
+
+unifdef-y += scsi_ioctl.h
+unifdef-y += sg.h
diff --git a/include/sound/Kbuild b/include/sound/Kbuild
index 3a5a3df61496..fd054a344324 100644
--- a/include/sound/Kbuild
+++ b/include/sound/Kbuild
@@ -1,2 +1,10 @@
-header-y := asound_fm.h hdsp.h hdspm.h sfnt_info.h sscape_ioctl.h
-unifdef-y := asequencer.h asound.h emu10k1.h sb16_csp.h 
+header-y += asound_fm.h
+header-y += hdsp.h
+header-y += hdspm.h
+header-y += sfnt_info.h
+header-y += sscape_ioctl.h
+
+unifdef-y += asequencer.h
+unifdef-y += asound.h
+unifdef-y += emu10k1.h
+unifdef-y += sb16_csp.h
diff --git a/include/video/Kbuild b/include/video/Kbuild
index 76a60737cc15..a14f9c045b8c 100644
--- a/include/video/Kbuild
+++ b/include/video/Kbuild
@@ -1 +1 @@
-unifdef-y := sisfb.h
+unifdef-y += sisfb.h
-- 
cgit v1.2.3


From fa053d2f008cb73fa768b8e171486d8c0b33312b Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 19 Sep 2006 14:51:40 +1000
Subject: [POWERPC] remove unused io accessors

The io accessors insw_ns, outsw_ns, insl_ns and outsl_ns are unused
(except for one unnecessary use in drivers/net/3c509.c that is addressed
in a previous patch) and are only defined in powerpc/ppc, so remove them.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/ppc/syslib/m8260_pci_erratum9.c | 16 ----------------
 include/asm-powerpc/io.h             | 13 +------------
 include/asm-ppc/io.h                 | 10 ----------
 include/asm-ppc/mpc8260_pci9.h       |  4 ----
 4 files changed, 1 insertion(+), 42 deletions(-)

(limited to 'include')

diff --git a/arch/ppc/syslib/m8260_pci_erratum9.c b/arch/ppc/syslib/m8260_pci_erratum9.c
index 974581ea4849..5475709ce07b 100644
--- a/arch/ppc/syslib/m8260_pci_erratum9.c
+++ b/arch/ppc/syslib/m8260_pci_erratum9.c
@@ -339,20 +339,6 @@ void insl(unsigned port, void *buf, int nl)
 	idma_pci9_read((u8 *)buf, (u8 *)addr, nl*sizeof(u32), sizeof(u32), 0);
 }
 
-void insw_ns(unsigned port, void *buf, int ns)
-{
-	u8 *addr = (u8 *)(port + _IO_BASE);
-
-	idma_pci9_read((u8 *)buf, (u8 *)addr, ns*sizeof(u16), sizeof(u16), 0);
-}
-
-void insl_ns(unsigned port, void *buf, int nl)
-{
-	u8 *addr = (u8 *)(port + _IO_BASE);
-
-	idma_pci9_read((u8 *)buf, (u8 *)addr, nl*sizeof(u32), sizeof(u32), 0);
-}
-
 void *memcpy_fromio(void *dest, unsigned long src, size_t count)
 {
 	unsigned long pa = iopa((unsigned long) src);
@@ -373,8 +359,6 @@ EXPORT_SYMBOL(inl);
 EXPORT_SYMBOL(insb);
 EXPORT_SYMBOL(insw);
 EXPORT_SYMBOL(insl);
-EXPORT_SYMBOL(insw_ns);
-EXPORT_SYMBOL(insl_ns);
 EXPORT_SYMBOL(memcpy_fromio);
 
 #endif	/* ifdef CONFIG_8260_PCI9 */
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 212428db0d8b..9aaced542624 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -76,8 +76,7 @@ extern unsigned long pci_io_base;
 #define insb(port, buf, ns)	_insb((u8 __iomem *)((port)+pci_io_base), (buf), (ns))
 #define insw(port, buf, ns)	_insw_ns((u8 __iomem *)((port)+pci_io_base), (buf), (ns))
 #define insl(port, buf, nl)	_insl_ns((u8 __iomem *)((port)+pci_io_base), (buf), (nl))
-#define insw_ns(port, buf, ns)	_insw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns))
-#define insl_ns(port, buf, nl)	_insl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl))
+
 #else
 
 static inline unsigned char __raw_readb(const volatile void __iomem *addr)
@@ -138,8 +137,6 @@ static inline void __raw_writeq(unsigned long v, volatile void __iomem *addr)
 #define insb(port, buf, ns)	eeh_insb((port), (buf), (ns))
 #define insw(port, buf, ns)	eeh_insw_ns((port), (buf), (ns))
 #define insl(port, buf, nl)	eeh_insl_ns((port), (buf), (nl))
-#define insw_ns(port, buf, ns)	eeh_insw_ns((port), (buf), (ns))
-#define insl_ns(port, buf, nl)	eeh_insl_ns((port), (buf), (nl))
 
 #endif
 
@@ -180,14 +177,6 @@ static inline void mmiowb(void)
 #define inl_p(port)             inl(port)
 #define outl_p(val, port)       (udelay(1), outl((val), (port)))
 
-/*
- * The *_ns versions below don't do byte-swapping.
- * Neither do the standard versions now, these are just here
- * for older code.
- */
-#define outsw_ns(port, buf, ns)	_outsw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns))
-#define outsl_ns(port, buf, nl)	_outsl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl))
-
 
 #define IO_SPACE_LIMIT ~(0UL)
 
diff --git a/include/asm-ppc/io.h b/include/asm-ppc/io.h
index 680555be22ec..fb0a8fcc51cf 100644
--- a/include/asm-ppc/io.h
+++ b/include/asm-ppc/io.h
@@ -338,16 +338,6 @@ extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, int ns);
 extern void _insl_ns(volatile u32 __iomem *port, void *buf, int nl);
 extern void _outsl_ns(volatile u32 __iomem *port, const void *buf, int nl);
 
-/*
- * The *_ns versions below don't do byte-swapping.
- * Neither do the standard versions now, these are just here
- * for older code.
- */
-#define insw_ns(port, buf, ns)	_insw_ns((port)+___IO_BASE, (buf), (ns))
-#define outsw_ns(port, buf, ns)	_outsw_ns((port)+___IO_BASE, (buf), (ns))
-#define insl_ns(port, buf, nl)	_insl_ns((port)+___IO_BASE, (buf), (nl))
-#define outsl_ns(port, buf, nl)	_outsl_ns((port)+___IO_BASE, (buf), (nl))
-
 
 #define IO_SPACE_LIMIT ~0
 
diff --git a/include/asm-ppc/mpc8260_pci9.h b/include/asm-ppc/mpc8260_pci9.h
index 26b3f6e787bc..9f7176881c56 100644
--- a/include/asm-ppc/mpc8260_pci9.h
+++ b/include/asm-ppc/mpc8260_pci9.h
@@ -30,8 +30,6 @@
 #undef inb
 #undef inw
 #undef inl
-#undef insw_ns
-#undef insl_ns
 #undef memcpy_fromio
 
 extern int readb(volatile unsigned char *addr);
@@ -43,8 +41,6 @@ extern void insl(unsigned port, void *buf, int nl);
 extern int inb(unsigned port);
 extern int inw(unsigned port);
 extern unsigned inl(unsigned port);
-extern void insw_ns(unsigned port, void *buf, int ns);
-extern void insl_ns(unsigned port, void *buf, int nl);
 extern void *memcpy_fromio(void *dest, unsigned long src, size_t count);
 
 #endif /* !__CONFIG_8260_PCI9_DEFS */
-- 
cgit v1.2.3


From 661f1cdb8b3e3c2c44e97df122c1d5643c054ce8 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 19 Sep 2006 16:52:55 +1000
Subject: [POWERPC] remove unused asm routines

_insw, _outsw, _insl amd _outsl are all unused, so remove them.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/kernel/misc.S      | 52 -------------------------
 arch/powerpc/kernel/ppc_ksyms.c |  4 --
 arch/ppc/kernel/misc.S          | 84 -----------------------------------------
 arch/ppc/kernel/ppc_ksyms.c     |  4 --
 include/asm-powerpc/io.h        |  4 --
 include/asm-ppc/io.h            |  4 --
 6 files changed, 152 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/misc.S b/arch/powerpc/kernel/misc.S
index f770805f1215..dd5f8e429196 100644
--- a/arch/powerpc/kernel/misc.S
+++ b/arch/powerpc/kernel/misc.S
@@ -86,58 +86,6 @@ _GLOBAL(_outsb)
 	sync
 	blr
 
-_GLOBAL(_insw)
-	sync
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,2
-	blelr-
-00:	lhbrx	r5,0,r3
-	eieio
-	sthu	r5,2(r4)
-	bdnz	00b
-	twi	0,r5,0
-	isync
-	blr
-
-_GLOBAL(_outsw)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,2
-	blelr-
-	sync
-00:	lhzu	r5,2(r4)
-	sthbrx	r5,0,r3
-	bdnz	00b
-	sync
-	blr
-
-_GLOBAL(_insl)
-	sync
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,4
-	blelr-
-00:	lwbrx	r5,0,r3
-	eieio
-	stwu	r5,4(r4)
-	bdnz	00b
-	twi	0,r5,0
-	isync
-	blr
-
-_GLOBAL(_outsl)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,4
-	blelr-
-	sync
-00:	lwzu	r5,4(r4)
-	stwbrx	r5,0,r3
-	bdnz	00b
-	sync
-	blr
-
 #ifdef CONFIG_PPC32
 _GLOBAL(__ide_mm_insw)
 #endif
diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c
index b2edac8ddf0a..314d6114e6ec 100644
--- a/arch/powerpc/kernel/ppc_ksyms.c
+++ b/arch/powerpc/kernel/ppc_ksyms.c
@@ -104,10 +104,6 @@ EXPORT_SYMBOL(__ide_mm_outsl);
 
 EXPORT_SYMBOL(_insb);
 EXPORT_SYMBOL(_outsb);
-EXPORT_SYMBOL(_insw);
-EXPORT_SYMBOL(_outsw);
-EXPORT_SYMBOL(_insl);
-EXPORT_SYMBOL(_outsl);
 EXPORT_SYMBOL(_insw_ns);
 EXPORT_SYMBOL(_outsw_ns);
 EXPORT_SYMBOL(_insl_ns);
diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S
index 2fa0075f2b5f..44700bbfe7bc 100644
--- a/arch/ppc/kernel/misc.S
+++ b/arch/ppc/kernel/misc.S
@@ -768,90 +768,6 @@ _GLOBAL(_outsb)
 	bdnz	00b
 	blr
 
-_GLOBAL(_insw)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,2
-	blelr-
-00:	lhbrx	r5,0,r3
-01:	eieio
-02:	sthu	r5,2(r4)
-	ISYNC_8xx
-	.section .fixup,"ax"
-03:	blr
-	.text
-	.section __ex_table, "a"
-		.align 2
-		.long 00b, 03b
-		.long 01b, 03b
-		.long 02b, 03b
-	.text
-	bdnz	00b
-	blr
-
-_GLOBAL(_outsw)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,2
-	blelr-
-00:	lhzu	r5,2(r4)
-01:	eieio
-02:	sthbrx	r5,0,r3
-	ISYNC_8xx
-	.section .fixup,"ax"
-03:	blr
-	.text
-	.section __ex_table, "a"
-		.align 2
-		.long 00b, 03b
-		.long 01b, 03b
-		.long 02b, 03b
-	.text
-	bdnz	00b
-	blr
-
-_GLOBAL(_insl)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,4
-	blelr-
-00:	lwbrx	r5,0,r3
-01:	eieio
-02:	stwu	r5,4(r4)
-	ISYNC_8xx
-	.section .fixup,"ax"
-03:	blr
-	.text
-	.section __ex_table, "a"
-		.align 2
-		.long 00b, 03b
-		.long 01b, 03b
-		.long 02b, 03b
-	.text
-	bdnz	00b
-	blr
-
-_GLOBAL(_outsl)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,4
-	blelr-
-00:	lwzu	r5,4(r4)
-01:	stwbrx	r5,0,r3
-02:	eieio
-	ISYNC_8xx
-	.section .fixup,"ax"
-03:	blr
-	.text
-	.section __ex_table, "a"
-		.align 2
-		.long 00b, 03b
-		.long 01b, 03b
-		.long 02b, 03b
-	.text
-	bdnz	00b
-	blr
-
 _GLOBAL(__ide_mm_insw)
 _GLOBAL(_insw_ns)
 	cmpwi	0,r5,0
diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c
index d1735401384c..2bd1f7353f56 100644
--- a/arch/ppc/kernel/ppc_ksyms.c
+++ b/arch/ppc/kernel/ppc_ksyms.c
@@ -122,10 +122,6 @@ EXPORT_SYMBOL(__ide_mm_outsl);
 
 EXPORT_SYMBOL(_insb);
 EXPORT_SYMBOL(_outsb);
-EXPORT_SYMBOL(_insw);
-EXPORT_SYMBOL(_outsw);
-EXPORT_SYMBOL(_insl);
-EXPORT_SYMBOL(_outsl);
 EXPORT_SYMBOL(_insw_ns);
 EXPORT_SYMBOL(_outsw_ns);
 EXPORT_SYMBOL(_insl_ns);
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 9aaced542624..0ee48436b1e3 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -151,10 +151,6 @@ static inline void __raw_writeq(unsigned long v, volatile void __iomem *addr)
 
 extern void _insb(volatile u8 __iomem *port, void *buf, int ns);
 extern void _outsb(volatile u8 __iomem *port, const void *buf, int ns);
-extern void _insw(volatile u16 __iomem *port, void *buf, int ns);
-extern void _outsw(volatile u16 __iomem *port, const void *buf, int ns);
-extern void _insl(volatile u32 __iomem *port, void *buf, int nl);
-extern void _outsl(volatile u32 __iomem *port, const void *buf, int nl);
 extern void _insw_ns(volatile u16 __iomem *port, void *buf, int ns);
 extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, int ns);
 extern void _insl_ns(volatile u32 __iomem *port, void *buf, int nl);
diff --git a/include/asm-ppc/io.h b/include/asm-ppc/io.h
index fb0a8fcc51cf..9fac420f1648 100644
--- a/include/asm-ppc/io.h
+++ b/include/asm-ppc/io.h
@@ -329,10 +329,6 @@ __do_out_asm(outl, "stwbrx")
 
 extern void _insb(volatile u8 __iomem *port, void *buf, int ns);
 extern void _outsb(volatile u8 __iomem *port, const void *buf, int ns);
-extern void _insw(volatile u16 __iomem *port, void *buf, int ns);
-extern void _outsw(volatile u16 __iomem *port, const void *buf, int ns);
-extern void _insl(volatile u32 __iomem *port, void *buf, int nl);
-extern void _outsl(volatile u32 __iomem *port, const void *buf, int nl);
 extern void _insw_ns(volatile u16 __iomem *port, void *buf, int ns);
 extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, int ns);
 extern void _insl_ns(volatile u32 __iomem *port, void *buf, int nl);
-- 
cgit v1.2.3


From 73ea9e1bcb8eea4f3b2052fe7ccd7ee4b5a271a0 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 19 Sep 2006 17:30:20 +1000
Subject: [POWERPC] clean up ide io accessors

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/kernel/misc.S      | 12 ------------
 arch/powerpc/kernel/ppc_ksyms.c |  7 -------
 arch/ppc/kernel/misc.S          |  4 ----
 arch/ppc/kernel/ppc_ksyms.c     |  5 -----
 include/asm-powerpc/ide.h       | 12 ++++++------
 include/asm-powerpc/io.h        |  6 ------
 6 files changed, 6 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/misc.S b/arch/powerpc/kernel/misc.S
index dd5f8e429196..6feb391422ec 100644
--- a/arch/powerpc/kernel/misc.S
+++ b/arch/powerpc/kernel/misc.S
@@ -86,9 +86,6 @@ _GLOBAL(_outsb)
 	sync
 	blr
 
-#ifdef CONFIG_PPC32
-_GLOBAL(__ide_mm_insw)
-#endif
 _GLOBAL(_insw_ns)
 	sync
 	cmpwi	0,r5,0
@@ -103,9 +100,6 @@ _GLOBAL(_insw_ns)
 	isync
 	blr
 
-#ifdef CONFIG_PPC32
-_GLOBAL(__ide_mm_outsw)
-#endif
 _GLOBAL(_outsw_ns)
 	cmpwi	0,r5,0
 	mtctr	r5
@@ -118,9 +112,6 @@ _GLOBAL(_outsw_ns)
 	sync
 	blr
 
-#ifdef CONFIG_PPC32
-_GLOBAL(__ide_mm_insl)
-#endif
 _GLOBAL(_insl_ns)
 	sync
 	cmpwi	0,r5,0
@@ -135,9 +126,6 @@ _GLOBAL(_insl_ns)
 	isync
 	blr
 
-#ifdef CONFIG_PPC32
-_GLOBAL(__ide_mm_outsl)
-#endif
 _GLOBAL(_outsl_ns)
 	cmpwi	0,r5,0
 	mtctr	r5
diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c
index 314d6114e6ec..75429e580518 100644
--- a/arch/powerpc/kernel/ppc_ksyms.c
+++ b/arch/powerpc/kernel/ppc_ksyms.c
@@ -95,13 +95,6 @@ EXPORT_SYMBOL(__strnlen_user);
 EXPORT_SYMBOL(copy_4K_page);
 #endif
 
-#ifndef  __powerpc64__
-EXPORT_SYMBOL(__ide_mm_insl);
-EXPORT_SYMBOL(__ide_mm_outsw);
-EXPORT_SYMBOL(__ide_mm_insw);
-EXPORT_SYMBOL(__ide_mm_outsl);
-#endif
-
 EXPORT_SYMBOL(_insb);
 EXPORT_SYMBOL(_outsb);
 EXPORT_SYMBOL(_insw_ns);
diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S
index 44700bbfe7bc..50b4bbd06804 100644
--- a/arch/ppc/kernel/misc.S
+++ b/arch/ppc/kernel/misc.S
@@ -768,7 +768,6 @@ _GLOBAL(_outsb)
 	bdnz	00b
 	blr
 
-_GLOBAL(__ide_mm_insw)
 _GLOBAL(_insw_ns)
 	cmpwi	0,r5,0
 	mtctr	r5
@@ -790,7 +789,6 @@ _GLOBAL(_insw_ns)
 	bdnz	00b
 	blr
 
-_GLOBAL(__ide_mm_outsw)
 _GLOBAL(_outsw_ns)
 	cmpwi	0,r5,0
 	mtctr	r5
@@ -812,7 +810,6 @@ _GLOBAL(_outsw_ns)
 	bdnz	00b
 	blr
 
-_GLOBAL(__ide_mm_insl)
 _GLOBAL(_insl_ns)
 	cmpwi	0,r5,0
 	mtctr	r5
@@ -834,7 +831,6 @@ _GLOBAL(_insl_ns)
 	bdnz	00b
 	blr
 
-_GLOBAL(__ide_mm_outsl)
 _GLOBAL(_outsl_ns)
 	cmpwi	0,r5,0
 	mtctr	r5
diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c
index 2bd1f7353f56..c8b65ca8a350 100644
--- a/arch/ppc/kernel/ppc_ksyms.c
+++ b/arch/ppc/kernel/ppc_ksyms.c
@@ -115,11 +115,6 @@ EXPORT_SYMBOL(outw);
 EXPORT_SYMBOL(outl);
 EXPORT_SYMBOL(outsl);*/
 
-EXPORT_SYMBOL(__ide_mm_insl);
-EXPORT_SYMBOL(__ide_mm_outsw);
-EXPORT_SYMBOL(__ide_mm_insw);
-EXPORT_SYMBOL(__ide_mm_outsl);
-
 EXPORT_SYMBOL(_insb);
 EXPORT_SYMBOL(_outsb);
 EXPORT_SYMBOL(_insw_ns);
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index b09b42af6a1e..c8390f9485de 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -12,6 +12,7 @@
 #include <linux/sched.h>
 #include <asm/mpc8xx.h>
 #endif
+#include <asm/io.h>
 
 #ifndef MAX_HWIFS
 #ifdef __powerpc64__
@@ -21,15 +22,14 @@
 #endif
 #endif
 
+#define __ide_mm_insw(p, a, c)	_insw_ns((volatile u16 __iomem *)(p), (a), (c))
+#define __ide_mm_insl(p, a, c)	_insl_ns((volatile u32 __iomem *)(p), (a), (c))
+#define __ide_mm_outsw(p, a, c)	_outsw_ns((volatile u16 __iomem *)(p), (a), (c))
+#define __ide_mm_outsl(p, a, c)	_outsl_ns((volatile u32 __iomem *)(p), (a), (c))
+
 #ifndef  __powerpc64__
 #include <linux/hdreg.h>
 #include <linux/ioport.h>
-#include <asm/io.h>
-
-extern void __ide_mm_insw(void __iomem *port, void *addr, u32 count);
-extern void __ide_mm_outsw(void __iomem *port, void *addr, u32 count);
-extern void __ide_mm_insl(void __iomem *port, void *addr, u32 count);
-extern void __ide_mm_outsl(void __iomem *port, void *addr, u32 count);
 
 struct ide_machdep_calls {
         int         (*default_irq)(unsigned long base);
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 0ee48436b1e3..51a598747367 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -28,12 +28,6 @@ extern int check_legacy_ioport(unsigned long base_port);
 
 #include <asm-generic/iomap.h>
 
-#define __ide_mm_insw(p, a, c) _insw_ns((volatile u16 __iomem *)(p), (a), (c))
-#define __ide_mm_insl(p, a, c) _insl_ns((volatile u32 __iomem *)(p), (a), (c))
-#define __ide_mm_outsw(p, a, c) _outsw_ns((volatile u16 __iomem *)(p), (a), (c))
-#define __ide_mm_outsl(p, a, c) _outsl_ns((volatile u32 __iomem *)(p), (a), (c))
-
-
 #define SIO_CONFIG_RA	0x398
 #define SIO_CONFIG_RD	0x399
 
-- 
cgit v1.2.3


From 5adcaf50cf697aa4d0c731107003c1383b59b214 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 19 Sep 2006 22:17:49 +1000
Subject: [POWERPC] convert string i/o operations to C

This produces essentially the same code and will make the iSeries i/o
consolidation easier.

The count parameter is changed to long since that will produce the same
(better) code on 32 and 64 bit builds.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/powerpc/kernel/Makefile    |   2 +-
 arch/powerpc/kernel/io.c        | 117 ++++++++++++++++++++++++++++++++++++++++
 arch/powerpc/kernel/misc.S      |  95 --------------------------------
 arch/powerpc/kernel/ppc_ksyms.c |   7 ---
 include/asm-powerpc/io.h        |  12 ++---
 include/asm-ppc/io.h            |  12 ++---
 6 files changed, 130 insertions(+), 115 deletions(-)
 create mode 100644 arch/powerpc/kernel/io.c

(limited to 'include')

diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index 8b3f4faf5768..8b133afbdc20 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -51,7 +51,7 @@ extra-$(CONFIG_8xx)		:= head_8xx.o
 extra-y				+= vmlinux.lds
 
 obj-y				+= time.o prom.o traps.o setup-common.o \
-				   udbg.o misc.o
+				   udbg.o misc.o io.o
 obj-$(CONFIG_PPC32)		+= entry_32.o setup_32.o misc_32.o
 obj-$(CONFIG_PPC64)		+= misc_64.o dma_64.o iommu.o
 obj-$(CONFIG_PPC_MULTIPLATFORM)	+= prom_init.o
diff --git a/arch/powerpc/kernel/io.c b/arch/powerpc/kernel/io.c
new file mode 100644
index 000000000000..80a3209acef4
--- /dev/null
+++ b/arch/powerpc/kernel/io.c
@@ -0,0 +1,117 @@
+/*
+ * I/O string operations
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *    Copyright (C) 2006 IBM Corporation
+ *
+ * Largely rewritten by Cort Dougan (cort@cs.nmt.edu)
+ * and Paul Mackerras.
+ *
+ * Adapted for iSeries by Mike Corrigan (mikejc@us.ibm.com)
+ * PPC64 updates by Dave Engebretsen (engebret@us.ibm.com)
+ *
+ * Rewritten in C by Stephen Rothwell.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/compiler.h>
+#include <linux/module.h>
+
+#include <asm/io.h>
+
+void _insb(volatile u8 __iomem *port, void *buf, long count)
+{
+	u8 *tbuf = buf;
+	u8 tmp;
+
+	if (unlikely(count <= 0))
+		return;
+	asm volatile("sync");
+	do {
+		tmp = *port;
+		asm volatile("eieio");
+		*tbuf++ = tmp;
+	} while (--count != 0);
+	asm volatile("twi 0,%0,0; isync" : : "r" (tmp));
+}
+EXPORT_SYMBOL(_insb);
+
+void _outsb(volatile u8 __iomem *port, const void *buf, long count)
+{
+	const u8 *tbuf = buf;
+
+	if (unlikely(count <= 0))
+		return;
+	asm volatile("sync");
+	do {
+		*port = *tbuf++;
+	} while (--count != 0);
+	asm volatile("sync");
+}
+EXPORT_SYMBOL(_outsb);
+
+void _insw_ns(volatile u16 __iomem *port, void *buf, long count)
+{
+	u16 *tbuf = buf;
+	u16 tmp;
+
+	if (unlikely(count <= 0))
+		return;
+	asm volatile("sync");
+	do {
+		tmp = *port;
+		asm volatile("eieio");
+		*tbuf++ = tmp;
+	} while (--count != 0);
+	asm volatile("twi 0,%0,0; isync" : : "r" (tmp));
+}
+EXPORT_SYMBOL(_insw_ns);
+
+void _outsw_ns(volatile u16 __iomem *port, const void *buf, long count)
+{
+	const u16 *tbuf = buf;
+
+	if (unlikely(count <= 0))
+		return;
+	asm volatile("sync");
+	do {
+		*port = *tbuf++;
+	} while (--count != 0);
+	asm volatile("sync");
+}
+EXPORT_SYMBOL(_outsw_ns);
+
+void _insl_ns(volatile u32 __iomem *port, void *buf, long count)
+{
+	u32 *tbuf = buf;
+	u32 tmp;
+
+	if (unlikely(count <= 0))
+		return;
+	asm volatile("sync");
+	do {
+		tmp = *port;
+		asm volatile("eieio");
+		*tbuf++ = tmp;
+	} while (--count != 0);
+	asm volatile("twi 0,%0,0; isync" : : "r" (tmp));
+}
+EXPORT_SYMBOL(_insl_ns);
+
+void _outsl_ns(volatile u32 __iomem *port, const void *buf, long count)
+{
+	const u32 *tbuf = buf;
+
+	if (unlikely(count <= 0))
+		return;
+	asm volatile("sync");
+	do {
+		*port = *tbuf++;
+	} while (--count != 0);
+	asm volatile("sync");
+}
+EXPORT_SYMBOL(_outsl_ns);
diff --git a/arch/powerpc/kernel/misc.S b/arch/powerpc/kernel/misc.S
index 6feb391422ec..330c9dc7db86 100644
--- a/arch/powerpc/kernel/misc.S
+++ b/arch/powerpc/kernel/misc.S
@@ -43,98 +43,3 @@ _GLOBAL(add_reloc_offset)
 	add	r3,r3,r5
 	mtlr	r0
 	blr
-
-/*
- * I/O string operations
- *
- * insb(port, buf, len)
- * outsb(port, buf, len)
- * insw(port, buf, len)
- * outsw(port, buf, len)
- * insl(port, buf, len)
- * outsl(port, buf, len)
- * insw_ns(port, buf, len)
- * outsw_ns(port, buf, len)
- * insl_ns(port, buf, len)
- * outsl_ns(port, buf, len)
- *
- * The *_ns versions don't do byte-swapping.
- */
-_GLOBAL(_insb)
-	sync
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,1
-	blelr-
-00:	lbz	r5,0(r3)
-	eieio
-	stbu	r5,1(r4)
-	bdnz	00b
-	twi	0,r5,0
-	isync
-	blr
-
-_GLOBAL(_outsb)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,1
-	blelr-
-	sync
-00:	lbzu	r5,1(r4)
-	stb	r5,0(r3)
-	bdnz	00b
-	sync
-	blr
-
-_GLOBAL(_insw_ns)
-	sync
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,2
-	blelr-
-00:	lhz	r5,0(r3)
-	eieio
-	sthu	r5,2(r4)
-	bdnz	00b
-	twi	0,r5,0
-	isync
-	blr
-
-_GLOBAL(_outsw_ns)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,2
-	blelr-
-	sync
-00:	lhzu	r5,2(r4)
-	sth	r5,0(r3)
-	bdnz	00b
-	sync
-	blr
-
-_GLOBAL(_insl_ns)
-	sync
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,4
-	blelr-
-00:	lwz	r5,0(r3)
-	eieio
-	stwu	r5,4(r4)
-	bdnz	00b
-	twi	0,r5,0
-	isync
-	blr
-
-_GLOBAL(_outsl_ns)
-	cmpwi	0,r5,0
-	mtctr	r5
-	subi	r4,r4,4
-	blelr-
-	sync
-00:	lwzu	r5,4(r4)
-	stw	r5,0(r3)
-	bdnz	00b
-	sync
-	blr
-
diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c
index 75429e580518..807193a3c784 100644
--- a/arch/powerpc/kernel/ppc_ksyms.c
+++ b/arch/powerpc/kernel/ppc_ksyms.c
@@ -95,13 +95,6 @@ EXPORT_SYMBOL(__strnlen_user);
 EXPORT_SYMBOL(copy_4K_page);
 #endif
 
-EXPORT_SYMBOL(_insb);
-EXPORT_SYMBOL(_outsb);
-EXPORT_SYMBOL(_insw_ns);
-EXPORT_SYMBOL(_outsw_ns);
-EXPORT_SYMBOL(_insl_ns);
-EXPORT_SYMBOL(_outsl_ns);
-
 #if defined(CONFIG_PPC32) && (defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE))
 EXPORT_SYMBOL(ppc_ide_md);
 #endif
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 51a598747367..57e7d14d6563 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -143,12 +143,12 @@ static inline void __raw_writeq(unsigned long v, volatile void __iomem *addr)
 #define readl_relaxed(addr) readl(addr)
 #define readq_relaxed(addr) readq(addr)
 
-extern void _insb(volatile u8 __iomem *port, void *buf, int ns);
-extern void _outsb(volatile u8 __iomem *port, const void *buf, int ns);
-extern void _insw_ns(volatile u16 __iomem *port, void *buf, int ns);
-extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, int ns);
-extern void _insl_ns(volatile u32 __iomem *port, void *buf, int nl);
-extern void _outsl_ns(volatile u32 __iomem *port, const void *buf, int nl);
+extern void _insb(volatile u8 __iomem *port, void *buf, long count);
+extern void _outsb(volatile u8 __iomem *port, const void *buf, long count);
+extern void _insw_ns(volatile u16 __iomem *port, void *buf, long count);
+extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, long count);
+extern void _insl_ns(volatile u32 __iomem *port, void *buf, long count);
+extern void _outsl_ns(volatile u32 __iomem *port, const void *buf, long count);
 
 static inline void mmiowb(void)
 {
diff --git a/include/asm-ppc/io.h b/include/asm-ppc/io.h
index 9fac420f1648..3d9a9e6f3321 100644
--- a/include/asm-ppc/io.h
+++ b/include/asm-ppc/io.h
@@ -327,12 +327,12 @@ __do_out_asm(outl, "stwbrx")
 #define inl_p(port)		inl((port))
 #define outl_p(val, port)	outl((val), (port))
 
-extern void _insb(volatile u8 __iomem *port, void *buf, int ns);
-extern void _outsb(volatile u8 __iomem *port, const void *buf, int ns);
-extern void _insw_ns(volatile u16 __iomem *port, void *buf, int ns);
-extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, int ns);
-extern void _insl_ns(volatile u32 __iomem *port, void *buf, int nl);
-extern void _outsl_ns(volatile u32 __iomem *port, const void *buf, int nl);
+extern void _insb(volatile u8 __iomem *port, void *buf, long count);
+extern void _outsb(volatile u8 __iomem *port, const void *buf, long count);
+extern void _insw_ns(volatile u16 __iomem *port, void *buf, long count);
+extern void _outsw_ns(volatile u16 __iomem *port, const void *buf, long count);
+extern void _insl_ns(volatile u32 __iomem *port, void *buf, long count);
+extern void _outsl_ns(volatile u32 __iomem *port, const void *buf, long count);
 
 
 #define IO_SPACE_LIMIT ~0
-- 
cgit v1.2.3


From 19e59df4dc2e6f7b46190ee77ce7093769f597a7 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 14 Sep 2006 14:55:36 +1000
Subject: [POWERPC] iseries: eliminate a couple of warnings

Copy and paste bug in io.h

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 include/asm-powerpc/io.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 57e7d14d6563..174fb89d5eda 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -68,8 +68,8 @@ extern unsigned long pci_io_base;
  * for older code.
  */
 #define insb(port, buf, ns)	_insb((u8 __iomem *)((port)+pci_io_base), (buf), (ns))
-#define insw(port, buf, ns)	_insw_ns((u8 __iomem *)((port)+pci_io_base), (buf), (ns))
-#define insl(port, buf, nl)	_insl_ns((u8 __iomem *)((port)+pci_io_base), (buf), (nl))
+#define insw(port, buf, ns)	_insw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns))
+#define insl(port, buf, nl)	_insl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl))
 
 #else
 
-- 
cgit v1.2.3


From a4dc7ff08915a2035aa6d6decc53fa1deaa410bb Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Tue, 19 Sep 2006 14:06:27 +1000
Subject: [POWERPC] Define of_read_ulong helper

There are various places where we want to extract an unsigned long
value from a device-tree property that can be 1 or 2 cells in length.
This replaces some open-coded calculations, and one place where we
assumed without checking that properties were the length we wanted,
with a little of_read_ulong() helper.

Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/prom.c         | 19 ++-----------------
 arch/powerpc/kernel/setup-common.c | 13 ++++++++-----
 arch/powerpc/kernel/time.c         |  4 +---
 include/asm-powerpc/prom.h         | 12 +++++++++++-
 4 files changed, 22 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index bf2005b2feb6..eb913f80bfb1 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -757,24 +757,9 @@ static int __init early_init_dt_scan_root(unsigned long node,
 static unsigned long __init dt_mem_next_cell(int s, cell_t **cellp)
 {
 	cell_t *p = *cellp;
-	unsigned long r;
 
-	/* Ignore more than 2 cells */
-	while (s > sizeof(unsigned long) / 4) {
-		p++;
-		s--;
-	}
-	r = *p++;
-#ifdef CONFIG_PPC64
-	if (s > 1) {
-		r <<= 32;
-		r |= *(p++);
-		s--;
-	}
-#endif
-
-	*cellp = p;
-	return r;
+	*cellp = p + s;
+	return of_read_ulong(p, s);
 }
 
 
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index 465e7435efbc..0af3fc1bdcc9 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -304,18 +304,21 @@ struct seq_operations cpuinfo_op = {
 void __init check_for_initrd(void)
 {
 #ifdef CONFIG_BLK_DEV_INITRD
-	const unsigned long *prop;
+	const unsigned int *prop;
+	int len;
 
 	DBG(" -> check_for_initrd()\n");
 
 	if (of_chosen) {
-		prop = get_property(of_chosen, "linux,initrd-start", NULL);
+		prop = get_property(of_chosen, "linux,initrd-start", &len);
 		if (prop != NULL) {
-			initrd_start = (unsigned long)__va(*prop);
+			initrd_start = (unsigned long)
+				__va(of_read_ulong(prop, len / 4));
 			prop = get_property(of_chosen,
-					"linux,initrd-end", NULL);
+					"linux,initrd-end", &len);
 			if (prop != NULL) {
-				initrd_end = (unsigned long)__va(*prop);
+				initrd_end = (unsigned long)
+					__va(of_read_ulong(prop, len / 4));
 				initrd_below_start_ok = 1;
 			} else
 				initrd_start = 0;
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index b9a2061cfdb7..7a3c3f791ade 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -870,9 +870,7 @@ static int __init get_freq(char *name, int cells, unsigned long *val)
 		fp = get_property(cpu, name, NULL);
 		if (fp) {
 			found = 1;
-			*val = 0;
-			while (cells--)
-				*val = (*val << 32) | *fp++;
+			*val = of_read_ulong(fp, cells);
 		}
 
 		of_node_put(cpu);
diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h
index c15e66a2e681..524629769336 100644
--- a/include/asm-powerpc/prom.h
+++ b/include/asm-powerpc/prom.h
@@ -197,7 +197,7 @@ extern int release_OF_resource(struct device_node* node, int index);
  */
 
 
-/* Helper to read a big number */
+/* Helper to read a big number; size is in cells (not bytes) */
 static inline u64 of_read_number(const u32 *cell, int size)
 {
 	u64 r = 0;
@@ -206,6 +206,16 @@ static inline u64 of_read_number(const u32 *cell, int size)
 	return r;
 }
 
+/* Like of_read_number, but we want an unsigned long result */
+#ifdef CONFIG_PPC32
+static inline unsigned long of_read_ulong(const u32 *cell, int size)
+{
+	return cell[size-1];
+}
+#else
+#define of_read_ulong(cell, size)	of_read_number(cell, size)
+#endif
+
 /* Translate an OF address block into a CPU physical address
  */
 #define OF_BAD_ADDR	((u64)-1)
-- 
cgit v1.2.3


From de1a3f1ce6c4c3b2b14cf9157a22d6b4c64f708e Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:20 +0200
Subject: [S390] EX_TABLE macro.

Add EX_TABLE helper macro to simplify creation of inline assembly
exception table entries.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/processor.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

(limited to 'include')

diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index 5b71d3731723..a3a4e5fd30d7 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -339,4 +339,21 @@ int unregister_idle_notifier(struct notifier_block *nb);
 
 #endif
 
+/*
+ * Helper macro for exception table entries
+ */
+#ifndef __s390x__
+#define EX_TABLE(_fault,_target)			\
+	".section __ex_table,\"a\"\n"			\
+	"	.align 4\n"				\
+	"	.long  " #_fault "," #_target "\n"	\
+	".previous\n"
+#else
+#define EX_TABLE(_fault,_target)			\
+	".section __ex_table,\"a\"\n"			\
+	"	.align 8\n"				\
+	"	.quad  " #_fault "," #_target "\n"	\
+	".previous\n"
+#endif
+
 #endif                                 /* __ASM_S390_PROCESSOR_H           */
-- 
cgit v1.2.3


From 7561b974e0cbbdca1bb880b55200afd9a1a20737 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:22 +0200
Subject: [S390] remove old z90crypt driver.

The z90crypt driver has served its term. It is replaced by the shiny
new zcrypt device driver.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/Kconfig              |    9 -
 drivers/s390/crypto/Makefile      |    2 -
 drivers/s390/crypto/z90common.h   |  166 --
 drivers/s390/crypto/z90crypt.h    |   71 -
 drivers/s390/crypto/z90hardware.c | 2531 ---------------------------
 drivers/s390/crypto/z90main.c     | 3379 -------------------------------------
 include/asm-s390/z90crypt.h       |  212 ---
 7 files changed, 6370 deletions(-)
 delete mode 100644 drivers/s390/crypto/z90common.h
 delete mode 100644 drivers/s390/crypto/z90crypt.h
 delete mode 100644 drivers/s390/crypto/z90hardware.c
 delete mode 100644 drivers/s390/crypto/z90main.c
 delete mode 100644 include/asm-s390/z90crypt.h

(limited to 'include')

diff --git a/drivers/s390/Kconfig b/drivers/s390/Kconfig
index 4d36208ff8de..f0ea550d39bc 100644
--- a/drivers/s390/Kconfig
+++ b/drivers/s390/Kconfig
@@ -217,13 +217,4 @@ endmenu
 
 menu "Cryptographic devices"
 
-config Z90CRYPT
-	tristate "Support for PCI-attached cryptographic adapters"
-        default "m"
-        help
-	  Select this option if you want to use a PCI-attached cryptographic
-	  adapter like the PCI Cryptographic Accelerator (PCICA) or the PCI
-	  Cryptographic Coprocessor (PCICC).  This option is also available
-	  as a module called z90crypt.ko.
-
 endmenu
diff --git a/drivers/s390/crypto/Makefile b/drivers/s390/crypto/Makefile
index 15edebbead7f..67e75be8e4e4 100644
--- a/drivers/s390/crypto/Makefile
+++ b/drivers/s390/crypto/Makefile
@@ -2,5 +2,3 @@
 # S/390 crypto devices
 #
 
-z90crypt-objs := z90main.o z90hardware.o
-obj-$(CONFIG_Z90CRYPT) += z90crypt.o
diff --git a/drivers/s390/crypto/z90common.h b/drivers/s390/crypto/z90common.h
deleted file mode 100644
index dbbcda3c846a..000000000000
--- a/drivers/s390/crypto/z90common.h
+++ /dev/null
@@ -1,166 +0,0 @@
-/*
- *  linux/drivers/s390/crypto/z90common.h
- *
- *  z90crypt 1.3.3
- *
- *  Copyright (C)  2001, 2005 IBM Corporation
- *  Author(s): Robert Burroughs (burrough@us.ibm.com)
- *             Eric Rossman (edrossma@us.ibm.com)
- *
- *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#ifndef _Z90COMMON_H_
-#define _Z90COMMON_H_
-
-
-#define RESPBUFFSIZE 256
-#define PCI_FUNC_KEY_DECRYPT 0x5044
-#define PCI_FUNC_KEY_ENCRYPT 0x504B
-extern int ext_bitlens;
-
-enum devstat {
-	DEV_GONE,
-	DEV_ONLINE,
-	DEV_QUEUE_FULL,
-	DEV_EMPTY,
-	DEV_NO_WORK,
-	DEV_BAD_MESSAGE,
-	DEV_TSQ_EXCEPTION,
-	DEV_RSQ_EXCEPTION,
-	DEV_SEN_EXCEPTION,
-	DEV_REC_EXCEPTION
-};
-
-enum hdstat {
-	HD_NOT_THERE,
-	HD_BUSY,
-	HD_DECONFIGURED,
-	HD_CHECKSTOPPED,
-	HD_ONLINE,
-	HD_TSQ_EXCEPTION
-};
-
-#define Z90C_NO_DEVICES		1
-#define Z90C_AMBIGUOUS_DOMAIN	2
-#define Z90C_INCORRECT_DOMAIN	3
-#define ENOTINIT		4
-
-#define SEN_BUSY	 7
-#define SEN_USER_ERROR	 8
-#define SEN_QUEUE_FULL	11
-#define SEN_NOT_AVAIL	16
-#define SEN_PAD_ERROR	17
-#define SEN_RETRY	18
-#define SEN_RELEASED	24
-
-#define REC_EMPTY	 4
-#define REC_BUSY	 6
-#define REC_OPERAND_INV	 8
-#define REC_OPERAND_SIZE 9
-#define REC_EVEN_MOD	10
-#define REC_NO_WORK	11
-#define REC_HARDWAR_ERR	12
-#define REC_NO_RESPONSE	13
-#define REC_RETRY_DEV	14
-#define REC_USER_GONE	15
-#define REC_BAD_MESSAGE	16
-#define REC_INVALID_PAD	17
-#define REC_USE_PCICA	18
-
-#define WRONG_DEVICE_TYPE 20
-
-#define REC_FATAL_ERROR 32
-#define SEN_FATAL_ERROR 33
-#define TSQ_FATAL_ERROR 34
-#define RSQ_FATAL_ERROR 35
-
-#define Z90CRYPT_NUM_TYPES	6
-#define PCICA		0
-#define PCICC		1
-#define PCIXCC_MCL2	2
-#define PCIXCC_MCL3	3
-#define CEX2C		4
-#define CEX2A		5
-#define NILDEV		-1
-#define ANYDEV		-1
-#define PCIXCC_UNK	-2
-
-enum hdevice_type {
-	PCICC_HW  = 3,
-	PCICA_HW  = 4,
-	PCIXCC_HW = 5,
-	CEX2A_HW  = 6,
-	CEX2C_HW  = 7
-};
-
-struct CPRBX {
-	unsigned short cprb_len;
-	unsigned char  cprb_ver_id;
-	unsigned char  pad_000[3];
-	unsigned char  func_id[2];
-	unsigned char  cprb_flags[4];
-	unsigned int   req_parml;
-	unsigned int   req_datal;
-	unsigned int   rpl_msgbl;
-	unsigned int   rpld_parml;
-	unsigned int   rpl_datal;
-	unsigned int   rpld_datal;
-	unsigned int   req_extbl;
-	unsigned char  pad_001[4];
-	unsigned int   rpld_extbl;
-	unsigned char  req_parmb[16];
-	unsigned char  req_datab[16];
-	unsigned char  rpl_parmb[16];
-	unsigned char  rpl_datab[16];
-	unsigned char  req_extb[16];
-	unsigned char  rpl_extb[16];
-	unsigned short ccp_rtcode;
-	unsigned short ccp_rscode;
-	unsigned int   mac_data_len;
-	unsigned char  logon_id[8];
-	unsigned char  mac_value[8];
-	unsigned char  mac_content_flgs;
-	unsigned char  pad_002;
-	unsigned short domain;
-	unsigned char  pad_003[12];
-	unsigned char  pad_004[36];
-};
-
-#ifndef DEV_NAME
-#define DEV_NAME	"z90crypt"
-#endif
-#define PRINTK(fmt, args...) \
-	printk(KERN_DEBUG DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
-#define PRINTKN(fmt, args...) \
-	printk(KERN_DEBUG DEV_NAME ": " fmt, ## args)
-#define PRINTKW(fmt, args...) \
-	printk(KERN_WARNING DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
-#define PRINTKC(fmt, args...) \
-	printk(KERN_CRIT DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
-
-#ifdef Z90CRYPT_DEBUG
-#define PDEBUG(fmt, args...) \
-	printk(KERN_DEBUG DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
-#else
-#define PDEBUG(fmt, args...) do {} while (0)
-#endif
-
-#define UMIN(a,b) ((a) < (b) ? (a) : (b))
-#define IS_EVEN(x) ((x) == (2 * ((x) / 2)))
-
-#endif
diff --git a/drivers/s390/crypto/z90crypt.h b/drivers/s390/crypto/z90crypt.h
deleted file mode 100644
index 0ca1d126ccb6..000000000000
--- a/drivers/s390/crypto/z90crypt.h
+++ /dev/null
@@ -1,71 +0,0 @@
-/*
- *  linux/drivers/s390/crypto/z90crypt.h
- *
- *  z90crypt 1.3.3 (kernel-private header)
- *
- *  Copyright (C)  2001, 2005 IBM Corporation
- *  Author(s): Robert Burroughs (burrough@us.ibm.com)
- *             Eric Rossman (edrossma@us.ibm.com)
- *
- *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#ifndef _Z90CRYPT_H_
-#define _Z90CRYPT_H_
-
-#include <asm/z90crypt.h>
-
-/**
- * local errno definitions
- */
-#define ENOBUFF	  129	// filp->private_data->...>work_elem_p->buffer is NULL
-#define EWORKPEND 130	// user issues ioctl while another pending
-#define ERELEASED 131	// user released while ioctl pending
-#define EQUIESCE  132	// z90crypt quiescing (no more work allowed)
-#define ETIMEOUT  133	// request timed out
-#define EUNKNOWN  134	// some unrecognized error occured (retry may succeed)
-#define EGETBUFF  135	// Error getting buffer or hardware lacks capability
-			// (retry in software)
-
-/**
- * DEPRECATED STRUCTURES
- */
-
-/**
- * This structure is DEPRECATED and the corresponding ioctl() has been
- * replaced with individual ioctl()s for each piece of data!
- * This structure will NOT survive past version 1.3.1, so switch to the
- * new ioctl()s.
- */
-#define MASK_LENGTH 64 // mask length
-struct ica_z90_status {
-	int totalcount;
-	int leedslitecount; // PCICA
-	int leeds2count;    // PCICC
-	// int PCIXCCCount; is not in struct for backward compatibility
-	int requestqWaitCount;
-	int pendingqWaitCount;
-	int totalOpenCount;
-	int cryptoDomain;
-	// status: 0=not there, 1=PCICA, 2=PCICC, 3=PCIXCC_MCL2, 4=PCIXCC_MCL3,
-	//         5=CEX2C
-	unsigned char status[MASK_LENGTH];
-	// qdepth: # work elements waiting for each device
-	unsigned char qdepth[MASK_LENGTH];
-};
-
-#endif /* _Z90CRYPT_H_ */
diff --git a/drivers/s390/crypto/z90hardware.c b/drivers/s390/crypto/z90hardware.c
deleted file mode 100644
index be60795f4a74..000000000000
--- a/drivers/s390/crypto/z90hardware.c
+++ /dev/null
@@ -1,2531 +0,0 @@
-/*
- *  linux/drivers/s390/crypto/z90hardware.c
- *
- *  z90crypt 1.3.3
- *
- *  Copyright (C)  2001, 2005 IBM Corporation
- *  Author(s): Robert Burroughs (burrough@us.ibm.com)
- *             Eric Rossman (edrossma@us.ibm.com)
- *
- *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#include <asm/uaccess.h>
-#include <linux/compiler.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/module.h>
-#include "z90crypt.h"
-#include "z90common.h"
-
-struct cca_token_hdr {
-	unsigned char  token_identifier;
-	unsigned char  version;
-	unsigned short token_length;
-	unsigned char  reserved[4];
-};
-
-#define CCA_TKN_HDR_ID_EXT 0x1E
-
-struct cca_private_ext_ME_sec {
-	unsigned char  section_identifier;
-	unsigned char  version;
-	unsigned short section_length;
-	unsigned char  private_key_hash[20];
-	unsigned char  reserved1[4];
-	unsigned char  key_format;
-	unsigned char  reserved2;
-	unsigned char  key_name_hash[20];
-	unsigned char  key_use_flags[4];
-	unsigned char  reserved3[6];
-	unsigned char  reserved4[24];
-	unsigned char  confounder[24];
-	unsigned char  exponent[128];
-	unsigned char  modulus[128];
-};
-
-#define CCA_PVT_USAGE_ALL 0x80
-
-struct cca_public_sec {
-	unsigned char  section_identifier;
-	unsigned char  version;
-	unsigned short section_length;
-	unsigned char  reserved[2];
-	unsigned short exponent_len;
-	unsigned short modulus_bit_len;
-	unsigned short modulus_byte_len;
-	unsigned char  exponent[3];
-};
-
-struct cca_private_ext_ME {
-	struct cca_token_hdr	      pvtMEHdr;
-	struct cca_private_ext_ME_sec pvtMESec;
-	struct cca_public_sec	      pubMESec;
-};
-
-struct cca_public_key {
-	struct cca_token_hdr  pubHdr;
-	struct cca_public_sec pubSec;
-};
-
-struct cca_pvt_ext_CRT_sec {
-	unsigned char  section_identifier;
-	unsigned char  version;
-	unsigned short section_length;
-	unsigned char  private_key_hash[20];
-	unsigned char  reserved1[4];
-	unsigned char  key_format;
-	unsigned char  reserved2;
-	unsigned char  key_name_hash[20];
-	unsigned char  key_use_flags[4];
-	unsigned short p_len;
-	unsigned short q_len;
-	unsigned short dp_len;
-	unsigned short dq_len;
-	unsigned short u_len;
-	unsigned short mod_len;
-	unsigned char  reserved3[4];
-	unsigned short pad_len;
-	unsigned char  reserved4[52];
-	unsigned char  confounder[8];
-};
-
-#define CCA_PVT_EXT_CRT_SEC_ID_PVT 0x08
-#define CCA_PVT_EXT_CRT_SEC_FMT_CL 0x40
-
-struct cca_private_ext_CRT {
-	struct cca_token_hdr	   pvtCrtHdr;
-	struct cca_pvt_ext_CRT_sec pvtCrtSec;
-	struct cca_public_sec	   pubCrtSec;
-};
-
-struct ap_status_word {
-	unsigned char q_stat_flags;
-	unsigned char response_code;
-	unsigned char reserved[2];
-};
-
-#define AP_Q_STATUS_EMPTY		0x80
-#define AP_Q_STATUS_REPLIES_WAITING	0x40
-#define AP_Q_STATUS_ARRAY_FULL		0x20
-
-#define AP_RESPONSE_NORMAL		0x00
-#define AP_RESPONSE_Q_NOT_AVAIL		0x01
-#define AP_RESPONSE_RESET_IN_PROGRESS	0x02
-#define AP_RESPONSE_DECONFIGURED	0x03
-#define AP_RESPONSE_CHECKSTOPPED	0x04
-#define AP_RESPONSE_BUSY		0x05
-#define AP_RESPONSE_Q_FULL		0x10
-#define AP_RESPONSE_NO_PENDING_REPLY	0x10
-#define AP_RESPONSE_INDEX_TOO_BIG	0x11
-#define AP_RESPONSE_NO_FIRST_PART	0x13
-#define AP_RESPONSE_MESSAGE_TOO_BIG	0x15
-
-#define AP_MAX_CDX_BITL		4
-#define AP_RQID_RESERVED_BITL	4
-#define SKIP_BITL		(AP_MAX_CDX_BITL + AP_RQID_RESERVED_BITL)
-
-struct type4_hdr {
-	unsigned char  reserved1;
-	unsigned char  msg_type_code;
-	unsigned short msg_len;
-	unsigned char  request_code;
-	unsigned char  msg_fmt;
-	unsigned short reserved2;
-};
-
-#define TYPE4_TYPE_CODE 0x04
-#define TYPE4_REQU_CODE 0x40
-
-#define TYPE4_SME_LEN 0x0188
-#define TYPE4_LME_LEN 0x0308
-#define TYPE4_SCR_LEN 0x01E0
-#define TYPE4_LCR_LEN 0x03A0
-
-#define TYPE4_SME_FMT 0x00
-#define TYPE4_LME_FMT 0x10
-#define TYPE4_SCR_FMT 0x40
-#define TYPE4_LCR_FMT 0x50
-
-struct type4_sme {
-	struct type4_hdr header;
-	unsigned char	 message[128];
-	unsigned char	 exponent[128];
-	unsigned char	 modulus[128];
-};
-
-struct type4_lme {
-	struct type4_hdr header;
-	unsigned char	 message[256];
-	unsigned char	 exponent[256];
-	unsigned char	 modulus[256];
-};
-
-struct type4_scr {
-	struct type4_hdr header;
-	unsigned char	 message[128];
-	unsigned char	 dp[72];
-	unsigned char	 dq[64];
-	unsigned char	 p[72];
-	unsigned char	 q[64];
-	unsigned char	 u[72];
-};
-
-struct type4_lcr {
-	struct type4_hdr header;
-	unsigned char	 message[256];
-	unsigned char	 dp[136];
-	unsigned char	 dq[128];
-	unsigned char	 p[136];
-	unsigned char	 q[128];
-	unsigned char	 u[136];
-};
-
-union type4_msg {
-	struct type4_sme sme;
-	struct type4_lme lme;
-	struct type4_scr scr;
-	struct type4_lcr lcr;
-};
-
-struct type84_hdr {
-	unsigned char  reserved1;
-	unsigned char  code;
-	unsigned short len;
-	unsigned char  reserved2[4];
-};
-
-#define TYPE84_RSP_CODE 0x84
-
-struct type6_hdr {
-	unsigned char reserved1;
-	unsigned char type;
-	unsigned char reserved2[2];
-	unsigned char right[4];
-	unsigned char reserved3[2];
-	unsigned char reserved4[2];
-	unsigned char apfs[4];
-	unsigned int  offset1;
-	unsigned int  offset2;
-	unsigned int  offset3;
-	unsigned int  offset4;
-	unsigned char agent_id[16];
-	unsigned char rqid[2];
-	unsigned char reserved5[2];
-	unsigned char function_code[2];
-	unsigned char reserved6[2];
-	unsigned int  ToCardLen1;
-	unsigned int  ToCardLen2;
-	unsigned int  ToCardLen3;
-	unsigned int  ToCardLen4;
-	unsigned int  FromCardLen1;
-	unsigned int  FromCardLen2;
-	unsigned int  FromCardLen3;
-	unsigned int  FromCardLen4;
-};
-
-struct CPRB {
-	unsigned char cprb_len[2];
-	unsigned char cprb_ver_id;
-	unsigned char pad_000;
-	unsigned char srpi_rtcode[4];
-	unsigned char srpi_verb;
-	unsigned char flags;
-	unsigned char func_id[2];
-	unsigned char checkpoint_flag;
-	unsigned char resv2;
-	unsigned char req_parml[2];
-	unsigned char req_parmp[4];
-	unsigned char req_datal[4];
-	unsigned char req_datap[4];
-	unsigned char rpl_parml[2];
-	unsigned char pad_001[2];
-	unsigned char rpl_parmp[4];
-	unsigned char rpl_datal[4];
-	unsigned char rpl_datap[4];
-	unsigned char ccp_rscode[2];
-	unsigned char ccp_rtcode[2];
-	unsigned char repd_parml[2];
-	unsigned char mac_data_len[2];
-	unsigned char repd_datal[4];
-	unsigned char req_pc[2];
-	unsigned char res_origin[8];
-	unsigned char mac_value[8];
-	unsigned char logon_id[8];
-	unsigned char usage_domain[2];
-	unsigned char resv3[18];
-	unsigned char svr_namel[2];
-	unsigned char svr_name[8];
-};
-
-struct type6_msg {
-	struct type6_hdr header;
-	struct CPRB	 CPRB;
-};
-
-struct type86_hdr {
-	unsigned char reserved1;
-	unsigned char type;
-	unsigned char format;
-	unsigned char reserved2;
-	unsigned char reply_code;
-	unsigned char reserved3[3];
-};
-
-#define TYPE86_RSP_CODE 0x86
-#define TYPE86_FMT2	0x02
-
-struct type86_fmt2_msg {
-	struct type86_hdr header;
-	unsigned char	  reserved[4];
-	unsigned char	  apfs[4];
-	unsigned int	  count1;
-	unsigned int	  offset1;
-	unsigned int	  count2;
-	unsigned int	  offset2;
-	unsigned int	  count3;
-	unsigned int	  offset3;
-	unsigned int	  count4;
-	unsigned int	  offset4;
-};
-
-static struct type6_hdr static_type6_hdr = {
-	0x00,
-	0x06,
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	0x00000058,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	{0x01,0x00,0x43,0x43,0x41,0x2D,0x41,0x50,
-	 0x50,0x4C,0x20,0x20,0x20,0x01,0x01,0x01},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x50,0x44},
-	{0x00,0x00},
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000
-};
-
-static struct type6_hdr static_type6_hdrX = {
-	0x00,
-	0x06,
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	0x00000058,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	{0x43,0x41,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x50,0x44},
-	{0x00,0x00},
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000
-};
-
-static struct CPRB static_cprb = {
-	{0x70,0x00},
-	0x41,
-	0x00,
-	{0x00,0x00,0x00,0x00},
-	0x00,
-	0x00,
-	{0x54,0x32},
-	0x01,
-	0x00,
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00},
-	{0x08,0x00},
-	{0x49,0x43,0x53,0x46,0x20,0x20,0x20,0x20}
-};
-
-struct function_and_rules_block {
-	unsigned char function_code[2];
-	unsigned char ulen[2];
-	unsigned char only_rule[8];
-};
-
-static struct function_and_rules_block static_pkd_function_and_rules = {
-	{0x50,0x44},
-	{0x0A,0x00},
-	{'P','K','C','S','-','1','.','2'}
-};
-
-static struct function_and_rules_block static_pke_function_and_rules = {
-	{0x50,0x4B},
-	{0x0A,0x00},
-	{'P','K','C','S','-','1','.','2'}
-};
-
-struct T6_keyBlock_hdr {
-	unsigned char blen[2];
-	unsigned char ulen[2];
-	unsigned char flags[2];
-};
-
-static struct T6_keyBlock_hdr static_T6_keyBlock_hdr = {
-	{0x89,0x01},
-	{0x87,0x01},
-	{0x00}
-};
-
-static struct CPRBX static_cprbx = {
-	0x00DC,
-	0x02,
-	{0x00,0x00,0x00},
-	{0x54,0x32},
-	{0x00,0x00,0x00,0x00},
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	{0x00,0x00,0x00,0x00},
-	0x00000000,
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	0x0000,
-	0x0000,
-	0x00000000,
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	0x00,
-	0x00,
-	0x0000,
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-	{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-	 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00}
-};
-
-static struct function_and_rules_block static_pkd_function_and_rulesX_MCL2 = {
-	{0x50,0x44},
-	{0x00,0x0A},
-	{'P','K','C','S','-','1','.','2'}
-};
-
-static struct function_and_rules_block static_pke_function_and_rulesX_MCL2 = {
-	{0x50,0x4B},
-	{0x00,0x0A},
-	{'Z','E','R','O','-','P','A','D'}
-};
-
-static struct function_and_rules_block static_pkd_function_and_rulesX = {
-	{0x50,0x44},
-	{0x00,0x0A},
-	{'Z','E','R','O','-','P','A','D'}
-};
-
-static struct function_and_rules_block static_pke_function_and_rulesX = {
-	{0x50,0x4B},
-	{0x00,0x0A},
-	{'M','R','P',' ',' ',' ',' ',' '}
-};
-
-static unsigned char static_PKE_function_code[2] = {0x50, 0x4B};
-
-struct T6_keyBlock_hdrX {
-	unsigned short blen;
-	unsigned short ulen;
-	unsigned char flags[2];
-};
-
-static unsigned char static_pad[256] = {
-0x1B,0x7B,0x5D,0xB5,0x75,0x01,0x3D,0xFD,0x8D,0xD1,0xC7,0x03,0x2D,0x09,0x23,0x57,
-0x89,0x49,0xB9,0x3F,0xBB,0x99,0x41,0x5B,0x75,0x21,0x7B,0x9D,0x3B,0x6B,0x51,0x39,
-0xBB,0x0D,0x35,0xB9,0x89,0x0F,0x93,0xA5,0x0B,0x47,0xF1,0xD3,0xBB,0xCB,0xF1,0x9D,
-0x23,0x73,0x71,0xFF,0xF3,0xF5,0x45,0xFB,0x61,0x29,0x23,0xFD,0xF1,0x29,0x3F,0x7F,
-0x17,0xB7,0x1B,0xA9,0x19,0xBD,0x57,0xA9,0xD7,0x95,0xA3,0xCB,0xED,0x1D,0xDB,0x45,
-0x7D,0x11,0xD1,0x51,0x1B,0xED,0x71,0xE9,0xB1,0xD1,0xAB,0xAB,0x21,0x2B,0x1B,0x9F,
-0x3B,0x9F,0xF7,0xF7,0xBD,0x63,0xEB,0xAD,0xDF,0xB3,0x6F,0x5B,0xDB,0x8D,0xA9,0x5D,
-0xE3,0x7D,0x77,0x49,0x47,0xF5,0xA7,0xFD,0xAB,0x2F,0x27,0x35,0x77,0xD3,0x49,0xC9,
-0x09,0xEB,0xB1,0xF9,0xBF,0x4B,0xCB,0x2B,0xEB,0xEB,0x05,0xFF,0x7D,0xC7,0x91,0x8B,
-0x09,0x83,0xB9,0xB9,0x69,0x33,0x39,0x6B,0x79,0x75,0x19,0xBF,0xBB,0x07,0x1D,0xBD,
-0x29,0xBF,0x39,0x95,0x93,0x1D,0x35,0xC7,0xC9,0x4D,0xE5,0x97,0x0B,0x43,0x9B,0xF1,
-0x16,0x93,0x03,0x1F,0xA5,0xFB,0xDB,0xF3,0x27,0x4F,0x27,0x61,0x05,0x1F,0xB9,0x23,
-0x2F,0xC3,0x81,0xA9,0x23,0x71,0x55,0x55,0xEB,0xED,0x41,0xE5,0xF3,0x11,0xF1,0x43,
-0x69,0x03,0xBD,0x0B,0x37,0x0F,0x51,0x8F,0x0B,0xB5,0x89,0x5B,0x67,0xA9,0xD9,0x4F,
-0x01,0xF9,0x21,0x77,0x37,0x73,0x79,0xC5,0x7F,0x51,0xC1,0xCF,0x97,0xA1,0x75,0xAD,
-0x35,0x9D,0xD3,0xD3,0xA7,0x9D,0x5D,0x41,0x6F,0x65,0x1B,0xCF,0xA9,0x87,0x91,0x09
-};
-
-static struct cca_private_ext_ME static_pvt_me_key = {
-	{
-		0x1E,
-		0x00,
-		0x0183,
-		{0x00,0x00,0x00,0x00}
-	},
-
-	{
-		0x02,
-		0x00,
-		0x016C,
-		{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00},
-		{0x00,0x00,0x00,0x00},
-		0x00,
-		0x00,
-		{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00},
-		{0x80,0x00,0x00,0x00},
-		{0x00,0x00,0x00,0x00,0x00,0x00},
-		{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-		{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-		{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00},
-		{0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-		 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00}
-	},
-
-	{
-		0x04,
-		0x00,
-		0x000F,
-		{0x00,0x00},
-		0x0003,
-		0x0000,
-		0x0000,
-		{0x01,0x00,0x01}
-	}
-};
-
-static struct cca_public_key static_public_key = {
-	{
-		0x1E,
-		0x00,
-		0x0000,
-		{0x00,0x00,0x00,0x00}
-	},
-
-	{
-		0x04,
-		0x00,
-		0x0000,
-		{0x00,0x00},
-		0x0000,
-		0x0000,
-		0x0000,
-		{0x01,0x00,0x01}
-	}
-};
-
-#define FIXED_TYPE6_ME_LEN 0x0000025F
-
-#define FIXED_TYPE6_ME_EN_LEN 0x000000F0
-
-#define FIXED_TYPE6_ME_LENX 0x000002CB
-
-#define FIXED_TYPE6_ME_EN_LENX 0x0000015C
-
-static struct cca_public_sec static_cca_pub_sec = {
-	0x04,
-	0x00,
-	0x000f,
-	{0x00,0x00},
-	0x0003,
-	0x0000,
-	0x0000,
-	{0x01,0x00,0x01}
-};
-
-#define FIXED_TYPE6_CR_LEN 0x00000177
-
-#define FIXED_TYPE6_CR_LENX 0x000001E3
-
-#define MAX_RESPONSE_SIZE 0x00000710
-
-#define MAX_RESPONSEX_SIZE 0x0000077C
-
-#define RESPONSE_CPRB_SIZE  0x000006B8
-#define RESPONSE_CPRBX_SIZE 0x00000724
-
-struct type50_hdr {
-	u8    reserved1;
-	u8    msg_type_code;
-	u16   msg_len;
-	u8    reserved2;
-	u8    ignored;
-	u16   reserved3;
-};
-
-#define TYPE50_TYPE_CODE 0x50
-
-#define TYPE50_MEB1_LEN (sizeof(struct type50_meb1_msg))
-#define TYPE50_MEB2_LEN (sizeof(struct type50_meb2_msg))
-#define TYPE50_CRB1_LEN (sizeof(struct type50_crb1_msg))
-#define TYPE50_CRB2_LEN (sizeof(struct type50_crb2_msg))
-
-#define TYPE50_MEB1_FMT 0x0001
-#define TYPE50_MEB2_FMT 0x0002
-#define TYPE50_CRB1_FMT 0x0011
-#define TYPE50_CRB2_FMT 0x0012
-
-struct type50_meb1_msg {
-	struct type50_hdr	header;
-	u16			keyblock_type;
-	u8			reserved[6];
-	u8			exponent[128];
-	u8			modulus[128];
-	u8			message[128];
-};
-
-struct type50_meb2_msg {
-	struct type50_hdr	header;
-	u16			keyblock_type;
-	u8			reserved[6];
-	u8			exponent[256];
-	u8			modulus[256];
-	u8			message[256];
-};
-
-struct type50_crb1_msg {
-	struct type50_hdr	header;
-	u16			keyblock_type;
-	u8			reserved[6];
-	u8			p[64];
-	u8			q[64];
-	u8			dp[64];
-	u8			dq[64];
-	u8			u[64];
-	u8			message[128];
-};
-
-struct type50_crb2_msg {
-	struct type50_hdr	header;
-	u16			keyblock_type;
-	u8			reserved[6];
-	u8			p[128];
-	u8			q[128];
-	u8			dp[128];
-	u8			dq[128];
-	u8			u[128];
-	u8			message[256];
-};
-
-union type50_msg {
-	struct type50_meb1_msg meb1;
-	struct type50_meb2_msg meb2;
-	struct type50_crb1_msg crb1;
-	struct type50_crb2_msg crb2;
-};
-
-struct type80_hdr {
-	u8	reserved1;
-	u8	type;
-	u16	len;
-	u8	code;
-	u8	reserved2[3];
-	u8	reserved3[8];
-};
-
-#define TYPE80_RSP_CODE 0x80
-
-struct error_hdr {
-	unsigned char reserved1;
-	unsigned char type;
-	unsigned char reserved2[2];
-	unsigned char reply_code;
-	unsigned char reserved3[3];
-};
-
-#define TYPE82_RSP_CODE 0x82
-#define TYPE88_RSP_CODE 0x88
-
-#define REP82_ERROR_MACHINE_FAILURE  0x10
-#define REP82_ERROR_PREEMPT_FAILURE  0x12
-#define REP82_ERROR_CHECKPT_FAILURE  0x14
-#define REP82_ERROR_MESSAGE_TYPE     0x20
-#define REP82_ERROR_INVALID_COMM_CD  0x21
-#define REP82_ERROR_INVALID_MSG_LEN  0x23
-#define REP82_ERROR_RESERVD_FIELD    0x24
-#define REP82_ERROR_FORMAT_FIELD     0x29
-#define REP82_ERROR_INVALID_COMMAND  0x30
-#define REP82_ERROR_MALFORMED_MSG    0x40
-#define REP82_ERROR_RESERVED_FIELDO  0x50
-#define REP82_ERROR_WORD_ALIGNMENT   0x60
-#define REP82_ERROR_MESSAGE_LENGTH   0x80
-#define REP82_ERROR_OPERAND_INVALID  0x82
-#define REP82_ERROR_OPERAND_SIZE     0x84
-#define REP82_ERROR_EVEN_MOD_IN_OPND 0x85
-#define REP82_ERROR_RESERVED_FIELD   0x88
-#define REP82_ERROR_TRANSPORT_FAIL   0x90
-#define REP82_ERROR_PACKET_TRUNCATED 0xA0
-#define REP82_ERROR_ZERO_BUFFER_LEN  0xB0
-
-#define REP88_ERROR_MODULE_FAILURE   0x10
-#define REP88_ERROR_MODULE_TIMEOUT   0x11
-#define REP88_ERROR_MODULE_NOTINIT   0x13
-#define REP88_ERROR_MODULE_NOTAVAIL  0x14
-#define REP88_ERROR_MODULE_DISABLED  0x15
-#define REP88_ERROR_MODULE_IN_DIAGN  0x17
-#define REP88_ERROR_FASTPATH_DISABLD 0x19
-#define REP88_ERROR_MESSAGE_TYPE     0x20
-#define REP88_ERROR_MESSAGE_MALFORMD 0x22
-#define REP88_ERROR_MESSAGE_LENGTH   0x23
-#define REP88_ERROR_RESERVED_FIELD   0x24
-#define REP88_ERROR_KEY_TYPE         0x34
-#define REP88_ERROR_INVALID_KEY      0x82
-#define REP88_ERROR_OPERAND          0x84
-#define REP88_ERROR_OPERAND_EVEN_MOD 0x85
-
-#define CALLER_HEADER 12
-
-static inline int
-testq(int q_nr, int *q_depth, int *dev_type, struct ap_status_word *stat)
-{
-	int ccode;
-
-	asm volatile
-#ifdef CONFIG_64BIT
-	("	llgfr	0,%4		\n"
-	 "	slgr	1,1		\n"
-	 "	lgr	2,1		\n"
-	 "0:	.long	0xb2af0000	\n"
-	 "1:	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	iihh	%0,0		\n"
-	 "	iihl	%0,0		\n"
-	 "	lgr	%1,1		\n"
-	 "	lgr	%3,2		\n"
-	 "	srl	%3,24		\n"
-	 "	sll	2,24		\n"
-	 "	srl	2,24		\n"
-	 "	lgr	%2,2		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi	%0,%h5		\n"
-	 "	jg	2b		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "	.align	8		\n"
-	 "	.quad	0b,3b		\n"
-	 "	.quad	1b,3b		\n"
-	 ".previous"
-	 :"=d" (ccode),"=d" (*stat),"=d" (*q_depth), "=d" (*dev_type)
-	 :"d" (q_nr), "K" (DEV_TSQ_EXCEPTION)
-	 :"cc","0","1","2","memory");
-#else
-	("	lr	0,%4		\n"
-	 "	slr	1,1		\n"
-	 "	lr	2,1		\n"
-	 "0:	.long	0xb2af0000	\n"
-	 "1:	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	lr	%1,1		\n"
-	 "	lr	%3,2		\n"
-	 "	srl	%3,24		\n"
-	 "	sll	2,24		\n"
-	 "	srl	2,24		\n"
-	 "	lr	%2,2		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi	%0,%h5		\n"
-	 "	bras	1,4f		\n"
-	 "	.long	2b		\n"
-	 "4:				\n"
-	 "	l	1,0(1)		\n"
-	 "	br	1		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "	.align	4		\n"
-	 "	.long	0b,3b		\n"
-	 "	.long	1b,3b		\n"
-	 ".previous"
-	 :"=d" (ccode),"=d" (*stat),"=d" (*q_depth), "=d" (*dev_type)
-	 :"d" (q_nr), "K" (DEV_TSQ_EXCEPTION)
-	 :"cc","0","1","2","memory");
-#endif
-	return ccode;
-}
-
-static inline int
-resetq(int q_nr, struct ap_status_word *stat_p)
-{
-	int ccode;
-
-	asm volatile
-#ifdef CONFIG_64BIT
-	("	llgfr	0,%2		\n"
-	 "	lghi	1,1		\n"
-	 "	sll	1,24		\n"
-	 "	or	0,1		\n"
-	 "	slgr	1,1		\n"
-	 "	lgr	2,1		\n"
-	 "0:	.long	0xb2af0000	\n"
-	 "1:	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	iihh	%0,0		\n"
-	 "	iihl	%0,0		\n"
-	 "	lgr	%1,1		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi	%0,%h3		\n"
-	 "	jg	2b		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "	.align	8		\n"
-	 "	.quad	0b,3b		\n"
-	 "	.quad	1b,3b		\n"
-	 ".previous"
-	 :"=d" (ccode),"=d" (*stat_p)
-	 :"d" (q_nr), "K" (DEV_RSQ_EXCEPTION)
-	 :"cc","0","1","2","memory");
-#else
-	("	lr	0,%2		\n"
-	 "	lhi	1,1		\n"
-	 "	sll	1,24		\n"
-	 "	or	0,1		\n"
-	 "	slr	1,1		\n"
-	 "	lr	2,1		\n"
-	 "0:	.long	0xb2af0000	\n"
-	 "1:	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	lr	%1,1		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi	%0,%h3		\n"
-	 "	bras	1,4f		\n"
-	 "	.long	2b		\n"
-	 "4:				\n"
-	 "	l	1,0(1)		\n"
-	 "	br	1		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "	.align	4		\n"
-	 "	.long	0b,3b		\n"
-	 "	.long	1b,3b		\n"
-	 ".previous"
-	 :"=d" (ccode),"=d" (*stat_p)
-	 :"d" (q_nr), "K" (DEV_RSQ_EXCEPTION)
-	 :"cc","0","1","2","memory");
-#endif
-	return ccode;
-}
-
-static inline int
-sen(int msg_len, unsigned char *msg_ext, struct ap_status_word *stat)
-{
-	int ccode;
-
-	asm volatile
-#ifdef CONFIG_64BIT
-	("	lgr	6,%3		\n"
-	 "	llgfr	7,%2		\n"
-	 "	llgt	0,0(6)		\n"
-	 "	lghi	1,64		\n"
-	 "	sll	1,24		\n"
-	 "	or	0,1		\n"
-	 "	la	6,4(6)		\n"
-	 "	llgt	2,0(6)		\n"
-	 "	llgt	3,4(6)		\n"
-	 "	la	6,8(6)		\n"
-	 "	slr	1,1		\n"
-	 "0:	.long	0xb2ad0026	\n"
-	 "1:	brc	2,0b		\n"
-	 "	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	iihh	%0,0		\n"
-	 "	iihl	%0,0		\n"
-	 "	lgr	%1,1		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi	%0,%h4		\n"
-	 "	jg	2b		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "	.align	8		\n"
-	 "	.quad	0b,3b		\n"
-	 "	.quad	1b,3b		\n"
-	 ".previous"
-	 :"=d" (ccode),"=d" (*stat)
-	 :"d" (msg_len),"a" (msg_ext), "K" (DEV_SEN_EXCEPTION)
-	 :"cc","0","1","2","3","6","7","memory");
-#else
-	("	lr	6,%3		\n"
-	 "	lr	7,%2		\n"
-	 "	l	0,0(6)		\n"
-	 "	lhi	1,64		\n"
-	 "	sll	1,24		\n"
-	 "	or	0,1		\n"
-	 "	la	6,4(6)		\n"
-	 "	l	2,0(6)		\n"
-	 "	l	3,4(6)		\n"
-	 "	la	6,8(6)		\n"
-	 "	slr	1,1		\n"
-	 "0:	.long	0xb2ad0026	\n"
-	 "1:	brc	2,0b		\n"
-	 "	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	lr	%1,1		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi	%0,%h4		\n"
-	 "	bras	1,4f		\n"
-	 "	.long	2b		\n"
-	 "4:				\n"
-	 "	l	1,0(1)		\n"
-	 "	br	1		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "	.align	4		\n"
-	 "	.long	0b,3b		\n"
-	 "	.long	1b,3b		\n"
-	 ".previous"
-	 :"=d" (ccode),"=d" (*stat)
-	 :"d" (msg_len),"a" (msg_ext), "K" (DEV_SEN_EXCEPTION)
-	 :"cc","0","1","2","3","6","7","memory");
-#endif
-	return ccode;
-}
-
-static inline int
-rec(int q_nr, int buff_l, unsigned char *rsp, unsigned char *id,
-    struct ap_status_word *st)
-{
-	int ccode;
-
-	asm volatile
-#ifdef CONFIG_64BIT
-	("	llgfr	0,%2		\n"
-	 "	lgr	3,%4		\n"
-	 "	lgr	6,%3		\n"
-	 "	llgfr	7,%5		\n"
-	 "	lghi	1,128		\n"
-	 "	sll	1,24		\n"
-	 "	or	0,1		\n"
-	 "	slgr	1,1		\n"
-	 "	lgr	2,1		\n"
-	 "	lgr	4,1		\n"
-	 "	lgr	5,1		\n"
-	 "0:	.long	0xb2ae0046	\n"
-	 "1:	brc	2,0b		\n"
-	 "	brc	4,0b		\n"
-	 "	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	iihh	%0,0		\n"
-	 "	iihl	%0,0		\n"
-	 "	lgr	%1,1		\n"
-	 "	st	4,0(3)		\n"
-	 "	st	5,4(3)		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi   %0,%h6		\n"
-	 "	jg    2b		\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "   .align	8		\n"
-	 "   .quad	0b,3b		\n"
-	 "   .quad	1b,3b		\n"
-	 ".previous"
-	 :"=d"(ccode),"=d"(*st)
-	 :"d" (q_nr), "d" (rsp), "d" (id), "d" (buff_l), "K" (DEV_REC_EXCEPTION)
-	 :"cc","0","1","2","3","4","5","6","7","memory");
-#else
-	("	lr	0,%2		\n"
-	 "	lr	3,%4		\n"
-	 "	lr	6,%3		\n"
-	 "	lr	7,%5		\n"
-	 "	lhi	1,128		\n"
-	 "	sll	1,24		\n"
-	 "	or	0,1		\n"
-	 "	slr	1,1		\n"
-	 "	lr	2,1		\n"
-	 "	lr	4,1		\n"
-	 "	lr	5,1		\n"
-	 "0:	.long	0xb2ae0046	\n"
-	 "1:	brc	2,0b		\n"
-	 "	brc	4,0b		\n"
-	 "	ipm	%0		\n"
-	 "	srl	%0,28		\n"
-	 "	lr	%1,1		\n"
-	 "	st	4,0(3)		\n"
-	 "	st	5,4(3)		\n"
-	 "2:				\n"
-	 ".section .fixup,\"ax\"	\n"
-	 "3:				\n"
-	 "	lhi   %0,%h6		\n"
-	 "	bras  1,4f		\n"
-	 "	.long 2b		\n"
-	 "4:				\n"
-	 "	l     1,0(1)		\n"
-	 "	br    1			\n"
-	 ".previous			\n"
-	 ".section __ex_table,\"a\"	\n"
-	 "   .align	4		\n"
-	 "   .long	0b,3b		\n"
-	 "   .long	1b,3b		\n"
-	 ".previous"
-	 :"=d"(ccode),"=d"(*st)
-	 :"d" (q_nr), "d" (rsp), "d" (id), "d" (buff_l), "K" (DEV_REC_EXCEPTION)
-	 :"cc","0","1","2","3","4","5","6","7","memory");
-#endif
-	return ccode;
-}
-
-static inline void
-itoLe2(int *i_p, unsigned char *lechars)
-{
-	*lechars       = *((unsigned char *) i_p + sizeof(int) - 1);
-	*(lechars + 1) = *((unsigned char *) i_p + sizeof(int) - 2);
-}
-
-static inline void
-le2toI(unsigned char *lechars, int *i_p)
-{
-	unsigned char *ic_p;
-	*i_p = 0;
-	ic_p = (unsigned char *) i_p;
-	*(ic_p + 2) = *(lechars + 1);
-	*(ic_p + 3) = *(lechars);
-}
-
-static inline int
-is_empty(unsigned char *ptr, int len)
-{
-	return !memcmp(ptr, (unsigned char *) &static_pvt_me_key+60, len);
-}
-
-enum hdstat
-query_online(int deviceNr, int cdx, int resetNr, int *q_depth, int *dev_type)
-{
-	int q_nr, i, t_depth, t_dev_type;
-	enum devstat ccode;
-	struct ap_status_word stat_word;
-	enum hdstat stat;
-	int break_out;
-
-	q_nr = (deviceNr << SKIP_BITL) + cdx;
-	stat = HD_BUSY;
-	ccode = testq(q_nr, &t_depth, &t_dev_type, &stat_word);
-	PDEBUG("ccode %d response_code %02X\n", ccode, stat_word.response_code);
-	break_out = 0;
-	for (i = 0; i < resetNr; i++) {
-		if (ccode > 3) {
-			PRINTKC("Exception testing device %d\n", i);
-			return HD_TSQ_EXCEPTION;
-		}
-		switch (ccode) {
-		case 0:
-			PDEBUG("t_dev_type %d\n", t_dev_type);
-			break_out = 1;
-			stat = HD_ONLINE;
-			*q_depth = t_depth + 1;
-			switch (t_dev_type) {
-			case PCICA_HW:
-				*dev_type = PCICA;
-				break;
-			case PCICC_HW:
-				*dev_type = PCICC;
-				break;
-			case PCIXCC_HW:
-				*dev_type = PCIXCC_UNK;
-				break;
-			case CEX2C_HW:
-				*dev_type = CEX2C;
-				break;
-			case CEX2A_HW:
-				*dev_type = CEX2A;
-				break;
-			default:
-				*dev_type = NILDEV;
-				break;
-			}
-			PDEBUG("available device %d: Q depth = %d, dev "
-			       "type = %d, stat = %02X%02X%02X%02X\n",
-			       deviceNr, *q_depth, *dev_type,
-			       stat_word.q_stat_flags,
-			       stat_word.response_code,
-			       stat_word.reserved[0],
-			       stat_word.reserved[1]);
-			break;
-		case 3:
-			switch (stat_word.response_code) {
-			case AP_RESPONSE_NORMAL:
-				stat = HD_ONLINE;
-				break_out = 1;
-				*q_depth = t_depth + 1;
-				*dev_type = t_dev_type;
-				PDEBUG("cc3, available device "
-				       "%d: Q depth = %d, dev "
-				       "type = %d, stat = "
-				       "%02X%02X%02X%02X\n",
-				       deviceNr, *q_depth,
-				       *dev_type,
-				       stat_word.q_stat_flags,
-				       stat_word.response_code,
-				       stat_word.reserved[0],
-				       stat_word.reserved[1]);
-				break;
-			case AP_RESPONSE_Q_NOT_AVAIL:
-				stat = HD_NOT_THERE;
-				break_out = 1;
-				break;
-			case AP_RESPONSE_RESET_IN_PROGRESS:
-				PDEBUG("device %d in reset\n",
-				       deviceNr);
-				break;
-			case AP_RESPONSE_DECONFIGURED:
-				stat = HD_DECONFIGURED;
-				break_out = 1;
-				break;
-			case AP_RESPONSE_CHECKSTOPPED:
-				stat = HD_CHECKSTOPPED;
-				break_out = 1;
-				break;
-			case AP_RESPONSE_BUSY:
-				PDEBUG("device %d busy\n",
-				       deviceNr);
-				break;
-			default:
-				break;
-			}
-			break;
-		default:
-			stat = HD_NOT_THERE;
-			break_out = 1;
-			break;
-		}
-		if (break_out)
-			break;
-
-		udelay(5);
-
-		ccode = testq(q_nr, &t_depth, &t_dev_type, &stat_word);
-	}
-	return stat;
-}
-
-enum devstat
-reset_device(int deviceNr, int cdx, int resetNr)
-{
-	int q_nr, ccode = 0, dummy_qdepth, dummy_devType, i;
-	struct ap_status_word stat_word;
-	enum devstat stat;
-	int break_out;
-
-	q_nr = (deviceNr << SKIP_BITL) + cdx;
-	stat = DEV_GONE;
-	ccode = resetq(q_nr, &stat_word);
-	if (ccode > 3)
-		return DEV_RSQ_EXCEPTION;
-
-	break_out = 0;
-	for (i = 0; i < resetNr; i++) {
-		switch (ccode) {
-		case 0:
-			stat = DEV_ONLINE;
-			if (stat_word.q_stat_flags & AP_Q_STATUS_EMPTY)
-				break_out = 1;
-			break;
-		case 3:
-			switch (stat_word.response_code) {
-			case AP_RESPONSE_NORMAL:
-				stat = DEV_ONLINE;
-				if (stat_word.q_stat_flags & AP_Q_STATUS_EMPTY)
-					break_out = 1;
-				break;
-			case AP_RESPONSE_Q_NOT_AVAIL:
-			case AP_RESPONSE_DECONFIGURED:
-			case AP_RESPONSE_CHECKSTOPPED:
-				stat = DEV_GONE;
-				break_out = 1;
-				break;
-			case AP_RESPONSE_RESET_IN_PROGRESS:
-			case AP_RESPONSE_BUSY:
-			default:
-				break;
-			}
-			break;
-		default:
-			stat = DEV_GONE;
-			break_out = 1;
-			break;
-		}
-		if (break_out == 1)
-			break;
-		udelay(5);
-
-		ccode = testq(q_nr, &dummy_qdepth, &dummy_devType, &stat_word);
-		if (ccode > 3) {
-			stat = DEV_TSQ_EXCEPTION;
-			break;
-		}
-	}
-	PDEBUG("Number of testq's needed for reset: %d\n", i);
-
-	if (i >= resetNr) {
-	  stat = DEV_GONE;
-	}
-
-	return stat;
-}
-
-#ifdef DEBUG_HYDRA_MSGS
-static inline void
-print_buffer(unsigned char *buffer, int bufflen)
-{
-	int i;
-	for (i = 0; i < bufflen; i += 16) {
-		PRINTK("%04X: %02X%02X%02X%02X %02X%02X%02X%02X "
-		       "%02X%02X%02X%02X %02X%02X%02X%02X\n", i,
-		       buffer[i+0], buffer[i+1], buffer[i+2], buffer[i+3],
-		       buffer[i+4], buffer[i+5], buffer[i+6], buffer[i+7],
-		       buffer[i+8], buffer[i+9], buffer[i+10], buffer[i+11],
-		       buffer[i+12], buffer[i+13], buffer[i+14], buffer[i+15]);
-	}
-}
-#endif
-
-enum devstat
-send_to_AP(int dev_nr, int cdx, int msg_len, unsigned char *msg_ext)
-{
-	struct ap_status_word stat_word;
-	enum devstat stat;
-	int ccode;
-	u32 *q_nr_p = (u32 *)msg_ext;
-
-	*q_nr_p = (dev_nr << SKIP_BITL) + cdx;
-	PDEBUG("msg_len passed to sen: %d\n", msg_len);
-	PDEBUG("q number passed to sen: %02x%02x%02x%02x\n",
-	       msg_ext[0], msg_ext[1], msg_ext[2], msg_ext[3]);
-	stat = DEV_GONE;
-
-#ifdef DEBUG_HYDRA_MSGS
-	PRINTK("Request header: %02X%02X%02X%02X %02X%02X%02X%02X "
-	       "%02X%02X%02X%02X\n",
-	       msg_ext[0], msg_ext[1], msg_ext[2], msg_ext[3],
-	       msg_ext[4], msg_ext[5], msg_ext[6], msg_ext[7],
-	       msg_ext[8], msg_ext[9], msg_ext[10], msg_ext[11]);
-	print_buffer(msg_ext+CALLER_HEADER, msg_len);
-#endif
-
-	ccode = sen(msg_len, msg_ext, &stat_word);
-	if (ccode > 3)
-		return DEV_SEN_EXCEPTION;
-
-	PDEBUG("nq cc: %u, st: %02x%02x%02x%02x\n",
-	       ccode, stat_word.q_stat_flags, stat_word.response_code,
-	       stat_word.reserved[0], stat_word.reserved[1]);
-	switch (ccode) {
-	case 0:
-		stat = DEV_ONLINE;
-		break;
-	case 1:
-		stat = DEV_GONE;
-		break;
-	case 3:
-		switch (stat_word.response_code) {
-		case AP_RESPONSE_NORMAL:
-			stat = DEV_ONLINE;
-			break;
-		case AP_RESPONSE_Q_FULL:
-			stat = DEV_QUEUE_FULL;
-			break;
-		default:
-			stat = DEV_GONE;
-			break;
-		}
-		break;
-	default:
-		stat = DEV_GONE;
-		break;
-	}
-
-	return stat;
-}
-
-enum devstat
-receive_from_AP(int dev_nr, int cdx, int resplen, unsigned char *resp,
-		unsigned char *psmid)
-{
-	int ccode;
-	struct ap_status_word stat_word;
-	enum devstat stat;
-
-	memset(resp, 0x00, 8);
-
-	ccode = rec((dev_nr << SKIP_BITL) + cdx, resplen, resp, psmid,
-		    &stat_word);
-	if (ccode > 3)
-		return DEV_REC_EXCEPTION;
-
-	PDEBUG("dq cc: %u, st: %02x%02x%02x%02x\n",
-	       ccode, stat_word.q_stat_flags, stat_word.response_code,
-	       stat_word.reserved[0], stat_word.reserved[1]);
-
-	stat = DEV_GONE;
-	switch (ccode) {
-	case 0:
-		stat = DEV_ONLINE;
-#ifdef DEBUG_HYDRA_MSGS
-		print_buffer(resp, resplen);
-#endif
-		break;
-	case 3:
-		switch (stat_word.response_code) {
-		case AP_RESPONSE_NORMAL:
-			stat = DEV_ONLINE;
-			break;
-		case AP_RESPONSE_NO_PENDING_REPLY:
-			if (stat_word.q_stat_flags & AP_Q_STATUS_EMPTY)
-				stat = DEV_EMPTY;
-			else
-				stat = DEV_NO_WORK;
-			break;
-		case AP_RESPONSE_INDEX_TOO_BIG:
-		case AP_RESPONSE_NO_FIRST_PART:
-		case AP_RESPONSE_MESSAGE_TOO_BIG:
-			stat = DEV_BAD_MESSAGE;
-			break;
-		default:
-			break;
-		}
-		break;
-	default:
-		break;
-	}
-
-	return stat;
-}
-
-static inline int
-pad_msg(unsigned char *buffer, int  totalLength, int msgLength)
-{
-	int pad_len;
-
-	for (pad_len = 0; pad_len < (totalLength - msgLength); pad_len++)
-		if (buffer[pad_len] != 0x00)
-			break;
-	pad_len -= 3;
-	if (pad_len < 8)
-		return SEN_PAD_ERROR;
-
-	buffer[0] = 0x00;
-	buffer[1] = 0x02;
-
-	memcpy(buffer+2, static_pad, pad_len);
-
-	buffer[pad_len + 2] = 0x00;
-
-	return 0;
-}
-
-static inline int
-is_common_public_key(unsigned char *key, int len)
-{
-	int i;
-
-	for (i = 0; i < len; i++)
-		if (key[i])
-			break;
-	key += i;
-	len -= i;
-	if (((len == 1) && (key[0] == 3)) ||
-	    ((len == 3) && (key[0] == 1) && (key[1] == 0) && (key[2] == 1)))
-		return 1;
-
-	return 0;
-}
-
-static int
-ICAMEX_msg_to_type4MEX_msg(struct ica_rsa_modexpo *icaMex_p, int *z90cMsg_l_p,
-			   union type4_msg *z90cMsg_p)
-{
-	int mod_len, msg_size, mod_tgt_len, exp_tgt_len, inp_tgt_len;
-	unsigned char *mod_tgt, *exp_tgt, *inp_tgt;
-	union type4_msg *tmp_type4_msg;
-
-	mod_len = icaMex_p->inputdatalength;
-
-	msg_size = ((mod_len <= 128) ? TYPE4_SME_LEN : TYPE4_LME_LEN) +
-		    CALLER_HEADER;
-
-	memset(z90cMsg_p, 0, msg_size);
-
-	tmp_type4_msg = (union type4_msg *)
-		((unsigned char *) z90cMsg_p + CALLER_HEADER);
-
-	tmp_type4_msg->sme.header.msg_type_code = TYPE4_TYPE_CODE;
-	tmp_type4_msg->sme.header.request_code = TYPE4_REQU_CODE;
-
-	if (mod_len <= 128) {
-		tmp_type4_msg->sme.header.msg_fmt = TYPE4_SME_FMT;
-		tmp_type4_msg->sme.header.msg_len = TYPE4_SME_LEN;
-		mod_tgt = tmp_type4_msg->sme.modulus;
-		mod_tgt_len = sizeof(tmp_type4_msg->sme.modulus);
-		exp_tgt = tmp_type4_msg->sme.exponent;
-		exp_tgt_len = sizeof(tmp_type4_msg->sme.exponent);
-		inp_tgt = tmp_type4_msg->sme.message;
-		inp_tgt_len = sizeof(tmp_type4_msg->sme.message);
-	} else {
-		tmp_type4_msg->lme.header.msg_fmt = TYPE4_LME_FMT;
-		tmp_type4_msg->lme.header.msg_len = TYPE4_LME_LEN;
-		mod_tgt = tmp_type4_msg->lme.modulus;
-		mod_tgt_len = sizeof(tmp_type4_msg->lme.modulus);
-		exp_tgt = tmp_type4_msg->lme.exponent;
-		exp_tgt_len = sizeof(tmp_type4_msg->lme.exponent);
-		inp_tgt = tmp_type4_msg->lme.message;
-		inp_tgt_len = sizeof(tmp_type4_msg->lme.message);
-	}
-
-	mod_tgt += (mod_tgt_len - mod_len);
-	if (copy_from_user(mod_tgt, icaMex_p->n_modulus, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(mod_tgt, mod_len))
-		return SEN_USER_ERROR;
-	exp_tgt += (exp_tgt_len - mod_len);
-	if (copy_from_user(exp_tgt, icaMex_p->b_key, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(exp_tgt, mod_len))
-		return SEN_USER_ERROR;
-	inp_tgt += (inp_tgt_len - mod_len);
-	if (copy_from_user(inp_tgt, icaMex_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(inp_tgt, mod_len))
-		return SEN_USER_ERROR;
-
-	*z90cMsg_l_p = msg_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICACRT_msg_to_type4CRT_msg(struct ica_rsa_modexpo_crt *icaMsg_p,
-			   int *z90cMsg_l_p, union type4_msg *z90cMsg_p)
-{
-	int mod_len, short_len, long_len, tmp_size, p_tgt_len, q_tgt_len,
-	    dp_tgt_len, dq_tgt_len, u_tgt_len, inp_tgt_len;
-	unsigned char *p_tgt, *q_tgt, *dp_tgt, *dq_tgt, *u_tgt, *inp_tgt;
-	union type4_msg *tmp_type4_msg;
-
-	mod_len = icaMsg_p->inputdatalength;
-	short_len = mod_len / 2;
-	long_len = mod_len / 2 + 8;
-
-	tmp_size = ((mod_len <= 128) ? TYPE4_SCR_LEN : TYPE4_LCR_LEN) +
-		    CALLER_HEADER;
-
-	memset(z90cMsg_p, 0, tmp_size);
-
-	tmp_type4_msg = (union type4_msg *)
-		((unsigned char *) z90cMsg_p + CALLER_HEADER);
-
-	tmp_type4_msg->scr.header.msg_type_code = TYPE4_TYPE_CODE;
-	tmp_type4_msg->scr.header.request_code = TYPE4_REQU_CODE;
-	if (mod_len <= 128) {
-		tmp_type4_msg->scr.header.msg_fmt = TYPE4_SCR_FMT;
-		tmp_type4_msg->scr.header.msg_len = TYPE4_SCR_LEN;
-		p_tgt = tmp_type4_msg->scr.p;
-		p_tgt_len = sizeof(tmp_type4_msg->scr.p);
-		q_tgt = tmp_type4_msg->scr.q;
-		q_tgt_len = sizeof(tmp_type4_msg->scr.q);
-		dp_tgt = tmp_type4_msg->scr.dp;
-		dp_tgt_len = sizeof(tmp_type4_msg->scr.dp);
-		dq_tgt = tmp_type4_msg->scr.dq;
-		dq_tgt_len = sizeof(tmp_type4_msg->scr.dq);
-		u_tgt = tmp_type4_msg->scr.u;
-		u_tgt_len = sizeof(tmp_type4_msg->scr.u);
-		inp_tgt = tmp_type4_msg->scr.message;
-		inp_tgt_len = sizeof(tmp_type4_msg->scr.message);
-	} else {
-		tmp_type4_msg->lcr.header.msg_fmt = TYPE4_LCR_FMT;
-		tmp_type4_msg->lcr.header.msg_len = TYPE4_LCR_LEN;
-		p_tgt = tmp_type4_msg->lcr.p;
-		p_tgt_len = sizeof(tmp_type4_msg->lcr.p);
-		q_tgt = tmp_type4_msg->lcr.q;
-		q_tgt_len = sizeof(tmp_type4_msg->lcr.q);
-		dp_tgt = tmp_type4_msg->lcr.dp;
-		dp_tgt_len = sizeof(tmp_type4_msg->lcr.dp);
-		dq_tgt = tmp_type4_msg->lcr.dq;
-		dq_tgt_len = sizeof(tmp_type4_msg->lcr.dq);
-		u_tgt = tmp_type4_msg->lcr.u;
-		u_tgt_len = sizeof(tmp_type4_msg->lcr.u);
-		inp_tgt = tmp_type4_msg->lcr.message;
-		inp_tgt_len = sizeof(tmp_type4_msg->lcr.message);
-	}
-
-	p_tgt += (p_tgt_len - long_len);
-	if (copy_from_user(p_tgt, icaMsg_p->np_prime, long_len))
-		return SEN_RELEASED;
-	if (is_empty(p_tgt, long_len))
-		return SEN_USER_ERROR;
-	q_tgt += (q_tgt_len - short_len);
-	if (copy_from_user(q_tgt, icaMsg_p->nq_prime, short_len))
-		return SEN_RELEASED;
-	if (is_empty(q_tgt, short_len))
-		return SEN_USER_ERROR;
-	dp_tgt += (dp_tgt_len - long_len);
-	if (copy_from_user(dp_tgt, icaMsg_p->bp_key, long_len))
-		return SEN_RELEASED;
-	if (is_empty(dp_tgt, long_len))
-		return SEN_USER_ERROR;
-	dq_tgt += (dq_tgt_len - short_len);
-	if (copy_from_user(dq_tgt, icaMsg_p->bq_key, short_len))
-		return SEN_RELEASED;
-	if (is_empty(dq_tgt, short_len))
-		return SEN_USER_ERROR;
-	u_tgt += (u_tgt_len - long_len);
-	if (copy_from_user(u_tgt, icaMsg_p->u_mult_inv, long_len))
-		return SEN_RELEASED;
-	if (is_empty(u_tgt, long_len))
-		return SEN_USER_ERROR;
-	inp_tgt += (inp_tgt_len - mod_len);
-	if (copy_from_user(inp_tgt, icaMsg_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(inp_tgt, mod_len))
-		return SEN_USER_ERROR;
-
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICAMEX_msg_to_type6MEX_de_msg(struct ica_rsa_modexpo *icaMsg_p, int cdx,
-			      int *z90cMsg_l_p, struct type6_msg *z90cMsg_p)
-{
-	int mod_len, vud_len, tmp_size, total_CPRB_len, parmBlock_l;
-	unsigned char *temp;
-	struct type6_hdr *tp6Hdr_p;
-	struct CPRB *cprb_p;
-	struct cca_private_ext_ME *key_p;
-	static int deprecated_msg_count = 0;
-
-	mod_len = icaMsg_p->inputdatalength;
-	tmp_size = FIXED_TYPE6_ME_LEN + mod_len;
-	total_CPRB_len = tmp_size - sizeof(struct type6_hdr);
-	parmBlock_l = total_CPRB_len - sizeof(struct CPRB);
-	tmp_size = 4*((tmp_size + 3)/4) + CALLER_HEADER;
-
-	memset(z90cMsg_p, 0, tmp_size);
-
-	temp = (unsigned char *)z90cMsg_p + CALLER_HEADER;
-	memcpy(temp, &static_type6_hdr, sizeof(struct type6_hdr));
-	tp6Hdr_p = (struct type6_hdr *)temp;
-	tp6Hdr_p->ToCardLen1 = 4*((total_CPRB_len+3)/4);
-	tp6Hdr_p->FromCardLen1 = RESPONSE_CPRB_SIZE;
-
-	temp += sizeof(struct type6_hdr);
-	memcpy(temp, &static_cprb, sizeof(struct CPRB));
-	cprb_p = (struct CPRB *) temp;
-	cprb_p->usage_domain[0]= (unsigned char)cdx;
-	itoLe2(&parmBlock_l, cprb_p->req_parml);
-	itoLe2((int *)&(tp6Hdr_p->FromCardLen1), cprb_p->rpl_parml);
-
-	temp += sizeof(struct CPRB);
-	memcpy(temp, &static_pkd_function_and_rules,
-	       sizeof(struct function_and_rules_block));
-
-	temp += sizeof(struct function_and_rules_block);
-	vud_len = 2 + icaMsg_p->inputdatalength;
-	itoLe2(&vud_len, temp);
-
-	temp += 2;
-	if (copy_from_user(temp, icaMsg_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(temp, mod_len))
-		return SEN_USER_ERROR;
-
-	temp += mod_len;
-	memcpy(temp, &static_T6_keyBlock_hdr, sizeof(struct T6_keyBlock_hdr));
-
-	temp += sizeof(struct T6_keyBlock_hdr);
-	memcpy(temp, &static_pvt_me_key, sizeof(struct cca_private_ext_ME));
-	key_p = (struct cca_private_ext_ME *)temp;
-	temp = key_p->pvtMESec.exponent + sizeof(key_p->pvtMESec.exponent)
-	       - mod_len;
-	if (copy_from_user(temp, icaMsg_p->b_key, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(temp, mod_len))
-		return SEN_USER_ERROR;
-
-	if (is_common_public_key(temp, mod_len)) {
-		if (deprecated_msg_count < 20) {
-			PRINTK("Common public key used for modex decrypt\n");
-			deprecated_msg_count++;
-			if (deprecated_msg_count == 20)
-				PRINTK("No longer issuing messages about common"
-				       " public key for modex decrypt.\n");
-		}
-		return SEN_NOT_AVAIL;
-	}
-
-	temp = key_p->pvtMESec.modulus + sizeof(key_p->pvtMESec.modulus)
-	       - mod_len;
-	if (copy_from_user(temp, icaMsg_p->n_modulus, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(temp, mod_len))
-		return SEN_USER_ERROR;
-
-	key_p->pubMESec.modulus_bit_len = 8 * mod_len;
-
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICAMEX_msg_to_type6MEX_en_msg(struct ica_rsa_modexpo *icaMsg_p, int cdx,
-			      int *z90cMsg_l_p, struct type6_msg *z90cMsg_p)
-{
-	int mod_len, vud_len, exp_len, key_len;
-	int pad_len, tmp_size, total_CPRB_len, parmBlock_l, i;
-	unsigned char *temp_exp, *exp_p, *temp;
-	struct type6_hdr *tp6Hdr_p;
-	struct CPRB *cprb_p;
-	struct cca_public_key *key_p;
-	struct T6_keyBlock_hdr *keyb_p;
-
-	temp_exp = kmalloc(256, GFP_KERNEL);
-	if (!temp_exp)
-		return EGETBUFF;
-	mod_len = icaMsg_p->inputdatalength;
-	if (copy_from_user(temp_exp, icaMsg_p->b_key, mod_len)) {
-		kfree(temp_exp);
-		return SEN_RELEASED;
-	}
-	if (is_empty(temp_exp, mod_len)) {
-		kfree(temp_exp);
-		return SEN_USER_ERROR;
-	}
-
-	exp_p = temp_exp;
-	for (i = 0; i < mod_len; i++)
-		if (exp_p[i])
-			break;
-	if (i >= mod_len) {
-		kfree(temp_exp);
-		return SEN_USER_ERROR;
-	}
-
-	exp_len = mod_len - i;
-	exp_p += i;
-
-	PDEBUG("exp_len after computation: %08x\n", exp_len);
-	tmp_size = FIXED_TYPE6_ME_EN_LEN + 2 * mod_len + exp_len;
-	total_CPRB_len = tmp_size - sizeof(struct type6_hdr);
-	parmBlock_l = total_CPRB_len - sizeof(struct CPRB);
-	tmp_size = 4*((tmp_size + 3)/4) + CALLER_HEADER;
-
-	vud_len = 2 + mod_len;
-	memset(z90cMsg_p, 0, tmp_size);
-
-	temp = (unsigned char *)z90cMsg_p + CALLER_HEADER;
-	memcpy(temp, &static_type6_hdr, sizeof(struct type6_hdr));
-	tp6Hdr_p = (struct type6_hdr *)temp;
-	tp6Hdr_p->ToCardLen1 = 4*((total_CPRB_len+3)/4);
-	tp6Hdr_p->FromCardLen1 = RESPONSE_CPRB_SIZE;
-	memcpy(tp6Hdr_p->function_code, static_PKE_function_code,
-	       sizeof(static_PKE_function_code));
-	temp += sizeof(struct type6_hdr);
-	memcpy(temp, &static_cprb, sizeof(struct CPRB));
-	cprb_p = (struct CPRB *) temp;
-	cprb_p->usage_domain[0]= (unsigned char)cdx;
-	itoLe2((int *)&(tp6Hdr_p->FromCardLen1), cprb_p->rpl_parml);
-	temp += sizeof(struct CPRB);
-	memcpy(temp, &static_pke_function_and_rules,
-		 sizeof(struct function_and_rules_block));
-	temp += sizeof(struct function_and_rules_block);
-	temp += 2;
-	if (copy_from_user(temp, icaMsg_p->inputdata, mod_len)) {
-		kfree(temp_exp);
-		return SEN_RELEASED;
-	}
-	if (is_empty(temp, mod_len)) {
-		kfree(temp_exp);
-		return SEN_USER_ERROR;
-	}
-	if ((temp[0] != 0x00) || (temp[1] != 0x02)) {
-		kfree(temp_exp);
-		return SEN_NOT_AVAIL;
-	}
-	for (i = 2; i < mod_len; i++)
-		if (temp[i] == 0x00)
-			break;
-	if ((i < 9) || (i > (mod_len - 2))) {
-		kfree(temp_exp);
-		return SEN_NOT_AVAIL;
-	}
-	pad_len = i + 1;
-	vud_len = mod_len - pad_len;
-	memmove(temp, temp+pad_len, vud_len);
-	temp -= 2;
-	vud_len += 2;
-	itoLe2(&vud_len, temp);
-	temp += (vud_len);
-	keyb_p = (struct T6_keyBlock_hdr *)temp;
-	temp += sizeof(struct T6_keyBlock_hdr);
-	memcpy(temp, &static_public_key, sizeof(static_public_key));
-	key_p = (struct cca_public_key *)temp;
-	temp = key_p->pubSec.exponent;
-	memcpy(temp, exp_p, exp_len);
-	kfree(temp_exp);
-	temp += exp_len;
-	if (copy_from_user(temp, icaMsg_p->n_modulus, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(temp, mod_len))
-		return SEN_USER_ERROR;
-	key_p->pubSec.modulus_bit_len = 8 * mod_len;
-	key_p->pubSec.modulus_byte_len = mod_len;
-	key_p->pubSec.exponent_len = exp_len;
-	key_p->pubSec.section_length = CALLER_HEADER + mod_len + exp_len;
-	key_len = key_p->pubSec.section_length + sizeof(struct cca_token_hdr);
-	key_p->pubHdr.token_length = key_len;
-	key_len += 4;
-	itoLe2(&key_len, keyb_p->ulen);
-	key_len += 2;
-	itoLe2(&key_len, keyb_p->blen);
-	parmBlock_l -= pad_len;
-	itoLe2(&parmBlock_l, cprb_p->req_parml);
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICACRT_msg_to_type6CRT_msg(struct ica_rsa_modexpo_crt *icaMsg_p, int cdx,
-			   int *z90cMsg_l_p, struct type6_msg *z90cMsg_p)
-{
-	int mod_len, vud_len, tmp_size, total_CPRB_len, parmBlock_l, short_len;
-	int long_len, pad_len, keyPartsLen, tmp_l;
-	unsigned char *tgt_p, *temp;
-	struct type6_hdr *tp6Hdr_p;
-	struct CPRB *cprb_p;
-	struct cca_token_hdr *keyHdr_p;
-	struct cca_pvt_ext_CRT_sec *pvtSec_p;
-	struct cca_public_sec *pubSec_p;
-
-	mod_len = icaMsg_p->inputdatalength;
-	short_len = mod_len / 2;
-	long_len = 8 + short_len;
-	keyPartsLen = 3 * long_len + 2 * short_len;
-	pad_len = (8 - (keyPartsLen % 8)) % 8;
-	keyPartsLen += pad_len + mod_len;
-	tmp_size = FIXED_TYPE6_CR_LEN + keyPartsLen + mod_len;
-	total_CPRB_len = tmp_size -  sizeof(struct type6_hdr);
-	parmBlock_l = total_CPRB_len - sizeof(struct CPRB);
-	vud_len = 2 + mod_len;
-	tmp_size = 4*((tmp_size + 3)/4) + CALLER_HEADER;
-
-	memset(z90cMsg_p, 0, tmp_size);
-	tgt_p = (unsigned char *)z90cMsg_p + CALLER_HEADER;
-	memcpy(tgt_p, &static_type6_hdr, sizeof(struct type6_hdr));
-	tp6Hdr_p = (struct type6_hdr *)tgt_p;
-	tp6Hdr_p->ToCardLen1 = 4*((total_CPRB_len+3)/4);
-	tp6Hdr_p->FromCardLen1 = RESPONSE_CPRB_SIZE;
-	tgt_p += sizeof(struct type6_hdr);
-	cprb_p = (struct CPRB *) tgt_p;
-	memcpy(tgt_p, &static_cprb, sizeof(struct CPRB));
-	cprb_p->usage_domain[0]= *((unsigned char *)(&(cdx))+3);
-	itoLe2(&parmBlock_l, cprb_p->req_parml);
-	memcpy(cprb_p->rpl_parml, cprb_p->req_parml,
-	       sizeof(cprb_p->req_parml));
-	tgt_p += sizeof(struct CPRB);
-	memcpy(tgt_p, &static_pkd_function_and_rules,
-	       sizeof(struct function_and_rules_block));
-	tgt_p += sizeof(struct function_and_rules_block);
-	itoLe2(&vud_len, tgt_p);
-	tgt_p += 2;
-	if (copy_from_user(tgt_p, icaMsg_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, mod_len))
-		return SEN_USER_ERROR;
-	tgt_p += mod_len;
-	tmp_l = sizeof(struct T6_keyBlock_hdr) + sizeof(struct cca_token_hdr) +
-		sizeof(struct cca_pvt_ext_CRT_sec) + 0x0F + keyPartsLen;
-	itoLe2(&tmp_l, tgt_p);
-	temp = tgt_p + 2;
-	tmp_l -= 2;
-	itoLe2(&tmp_l, temp);
-	tgt_p += sizeof(struct T6_keyBlock_hdr);
-	keyHdr_p = (struct cca_token_hdr *)tgt_p;
-	keyHdr_p->token_identifier = CCA_TKN_HDR_ID_EXT;
-	tmp_l -= 4;
-	keyHdr_p->token_length = tmp_l;
-	tgt_p += sizeof(struct cca_token_hdr);
-	pvtSec_p = (struct cca_pvt_ext_CRT_sec *)tgt_p;
-	pvtSec_p->section_identifier = CCA_PVT_EXT_CRT_SEC_ID_PVT;
-	pvtSec_p->section_length =
-		sizeof(struct cca_pvt_ext_CRT_sec) + keyPartsLen;
-	pvtSec_p->key_format = CCA_PVT_EXT_CRT_SEC_FMT_CL;
-	pvtSec_p->key_use_flags[0] = CCA_PVT_USAGE_ALL;
-	pvtSec_p->p_len = long_len;
-	pvtSec_p->q_len = short_len;
-	pvtSec_p->dp_len = long_len;
-	pvtSec_p->dq_len = short_len;
-	pvtSec_p->u_len = long_len;
-	pvtSec_p->mod_len = mod_len;
-	pvtSec_p->pad_len = pad_len;
-	tgt_p += sizeof(struct cca_pvt_ext_CRT_sec);
-	if (copy_from_user(tgt_p, icaMsg_p->np_prime, long_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, long_len))
-		return SEN_USER_ERROR;
-	tgt_p += long_len;
-	if (copy_from_user(tgt_p, icaMsg_p->nq_prime, short_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, short_len))
-		return SEN_USER_ERROR;
-	tgt_p += short_len;
-	if (copy_from_user(tgt_p, icaMsg_p->bp_key, long_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, long_len))
-		return SEN_USER_ERROR;
-	tgt_p += long_len;
-	if (copy_from_user(tgt_p, icaMsg_p->bq_key, short_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, short_len))
-		return SEN_USER_ERROR;
-	tgt_p += short_len;
-	if (copy_from_user(tgt_p, icaMsg_p->u_mult_inv, long_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, long_len))
-		return SEN_USER_ERROR;
-	tgt_p += long_len;
-	tgt_p += pad_len;
-	memset(tgt_p, 0xFF, mod_len);
-	tgt_p += mod_len;
-	memcpy(tgt_p, &static_cca_pub_sec, sizeof(struct cca_public_sec));
-	pubSec_p = (struct cca_public_sec *) tgt_p;
-	pubSec_p->modulus_bit_len = 8 * mod_len;
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICAMEX_msg_to_type6MEX_msgX(struct ica_rsa_modexpo *icaMsg_p, int cdx,
-			    int *z90cMsg_l_p, struct type6_msg *z90cMsg_p,
-			    int dev_type)
-{
-	int mod_len, exp_len, vud_len, tmp_size, total_CPRB_len, parmBlock_l;
-	int key_len, i;
-	unsigned char *temp_exp, *tgt_p, *temp, *exp_p;
-	struct type6_hdr *tp6Hdr_p;
-	struct CPRBX *cprbx_p;
-	struct cca_public_key *key_p;
-	struct T6_keyBlock_hdrX *keyb_p;
-
-	temp_exp = kmalloc(256, GFP_KERNEL);
-	if (!temp_exp)
-		return EGETBUFF;
-	mod_len = icaMsg_p->inputdatalength;
-	if (copy_from_user(temp_exp, icaMsg_p->b_key, mod_len)) {
-		kfree(temp_exp);
-		return SEN_RELEASED;
-	}
-	if (is_empty(temp_exp, mod_len)) {
-		kfree(temp_exp);
-		return SEN_USER_ERROR;
-	}
-	exp_p = temp_exp;
-	for (i = 0; i < mod_len; i++)
-		if (exp_p[i])
-			break;
-	if (i >= mod_len) {
-		kfree(temp_exp);
-		return SEN_USER_ERROR;
-	}
-	exp_len = mod_len - i;
-	exp_p += i;
-	PDEBUG("exp_len after computation: %08x\n", exp_len);
-	tmp_size = FIXED_TYPE6_ME_EN_LENX + 2 * mod_len + exp_len;
-	total_CPRB_len = tmp_size - sizeof(struct type6_hdr);
-	parmBlock_l = total_CPRB_len - sizeof(struct CPRBX);
-	tmp_size = tmp_size + CALLER_HEADER;
-	vud_len = 2 + mod_len;
-	memset(z90cMsg_p, 0, tmp_size);
-	tgt_p = (unsigned char *)z90cMsg_p + CALLER_HEADER;
-	memcpy(tgt_p, &static_type6_hdrX, sizeof(struct type6_hdr));
-	tp6Hdr_p = (struct type6_hdr *)tgt_p;
-	tp6Hdr_p->ToCardLen1 = total_CPRB_len;
-	tp6Hdr_p->FromCardLen1 = RESPONSE_CPRBX_SIZE;
-	memcpy(tp6Hdr_p->function_code, static_PKE_function_code,
-	       sizeof(static_PKE_function_code));
-	tgt_p += sizeof(struct type6_hdr);
-	memcpy(tgt_p, &static_cprbx, sizeof(struct CPRBX));
-	cprbx_p = (struct CPRBX *) tgt_p;
-	cprbx_p->domain = (unsigned short)cdx;
-	cprbx_p->rpl_msgbl = RESPONSE_CPRBX_SIZE;
-	tgt_p += sizeof(struct CPRBX);
-	if (dev_type == PCIXCC_MCL2)
-		memcpy(tgt_p, &static_pke_function_and_rulesX_MCL2,
-		       sizeof(struct function_and_rules_block));
-	else
-		memcpy(tgt_p, &static_pke_function_and_rulesX,
-		       sizeof(struct function_and_rules_block));
-	tgt_p += sizeof(struct function_and_rules_block);
-
-	tgt_p += 2;
-	if (copy_from_user(tgt_p, icaMsg_p->inputdata, mod_len)) {
-		kfree(temp_exp);
-		return SEN_RELEASED;
-	}
-	if (is_empty(tgt_p, mod_len)) {
-		kfree(temp_exp);
-		return SEN_USER_ERROR;
-	}
-	tgt_p -= 2;
-	*((short *)tgt_p) = (short) vud_len;
-	tgt_p += vud_len;
-	keyb_p = (struct T6_keyBlock_hdrX *)tgt_p;
-	tgt_p += sizeof(struct T6_keyBlock_hdrX);
-	memcpy(tgt_p, &static_public_key, sizeof(static_public_key));
-	key_p = (struct cca_public_key *)tgt_p;
-	temp = key_p->pubSec.exponent;
-	memcpy(temp, exp_p, exp_len);
-	kfree(temp_exp);
-	temp += exp_len;
-	if (copy_from_user(temp, icaMsg_p->n_modulus, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(temp, mod_len))
-		return SEN_USER_ERROR;
-	key_p->pubSec.modulus_bit_len = 8 * mod_len;
-	key_p->pubSec.modulus_byte_len = mod_len;
-	key_p->pubSec.exponent_len = exp_len;
-	key_p->pubSec.section_length = CALLER_HEADER + mod_len + exp_len;
-	key_len = key_p->pubSec.section_length + sizeof(struct cca_token_hdr);
-	key_p->pubHdr.token_length = key_len;
-	key_len += 4;
-	keyb_p->ulen = (unsigned short)key_len;
-	key_len += 2;
-	keyb_p->blen = (unsigned short)key_len;
-	cprbx_p->req_parml = parmBlock_l;
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICACRT_msg_to_type6CRT_msgX(struct ica_rsa_modexpo_crt *icaMsg_p, int cdx,
-			    int *z90cMsg_l_p, struct type6_msg *z90cMsg_p,
-			    int dev_type)
-{
-	int mod_len, vud_len, tmp_size, total_CPRB_len, parmBlock_l, short_len;
-	int long_len, pad_len, keyPartsLen, tmp_l;
-	unsigned char *tgt_p, *temp;
-	struct type6_hdr *tp6Hdr_p;
-	struct CPRBX *cprbx_p;
-	struct cca_token_hdr *keyHdr_p;
-	struct cca_pvt_ext_CRT_sec *pvtSec_p;
-	struct cca_public_sec *pubSec_p;
-
-	mod_len = icaMsg_p->inputdatalength;
-	short_len = mod_len / 2;
-	long_len = 8 + short_len;
-	keyPartsLen = 3 * long_len + 2 * short_len;
-	pad_len = (8 - (keyPartsLen % 8)) % 8;
-	keyPartsLen += pad_len + mod_len;
-	tmp_size = FIXED_TYPE6_CR_LENX + keyPartsLen + mod_len;
-	total_CPRB_len = tmp_size -  sizeof(struct type6_hdr);
-	parmBlock_l = total_CPRB_len - sizeof(struct CPRBX);
-	vud_len = 2 + mod_len;
-	tmp_size = tmp_size + CALLER_HEADER;
-	memset(z90cMsg_p, 0, tmp_size);
-	tgt_p = (unsigned char *)z90cMsg_p + CALLER_HEADER;
-	memcpy(tgt_p, &static_type6_hdrX, sizeof(struct type6_hdr));
-	tp6Hdr_p = (struct type6_hdr *)tgt_p;
-	tp6Hdr_p->ToCardLen1 = total_CPRB_len;
-	tp6Hdr_p->FromCardLen1 = RESPONSE_CPRBX_SIZE;
-	tgt_p += sizeof(struct type6_hdr);
-	cprbx_p = (struct CPRBX *) tgt_p;
-	memcpy(tgt_p, &static_cprbx, sizeof(struct CPRBX));
-	cprbx_p->domain = (unsigned short)cdx;
-	cprbx_p->req_parml = parmBlock_l;
-	cprbx_p->rpl_msgbl = parmBlock_l;
-	tgt_p += sizeof(struct CPRBX);
-	if (dev_type == PCIXCC_MCL2)
-		memcpy(tgt_p, &static_pkd_function_and_rulesX_MCL2,
-		       sizeof(struct function_and_rules_block));
-	else
-		memcpy(tgt_p, &static_pkd_function_and_rulesX,
-		       sizeof(struct function_and_rules_block));
-	tgt_p += sizeof(struct function_and_rules_block);
-	*((short *)tgt_p) = (short) vud_len;
-	tgt_p += 2;
-	if (copy_from_user(tgt_p, icaMsg_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, mod_len))
-		return SEN_USER_ERROR;
-	tgt_p += mod_len;
-	tmp_l = sizeof(struct T6_keyBlock_hdr) + sizeof(struct cca_token_hdr) +
-		sizeof(struct cca_pvt_ext_CRT_sec) + 0x0F + keyPartsLen;
-	*((short *)tgt_p) = (short) tmp_l;
-	temp = tgt_p + 2;
-	tmp_l -= 2;
-	*((short *)temp) = (short) tmp_l;
-	tgt_p += sizeof(struct T6_keyBlock_hdr);
-	keyHdr_p = (struct cca_token_hdr *)tgt_p;
-	keyHdr_p->token_identifier = CCA_TKN_HDR_ID_EXT;
-	tmp_l -= 4;
-	keyHdr_p->token_length = tmp_l;
-	tgt_p += sizeof(struct cca_token_hdr);
-	pvtSec_p = (struct cca_pvt_ext_CRT_sec *)tgt_p;
-	pvtSec_p->section_identifier = CCA_PVT_EXT_CRT_SEC_ID_PVT;
-	pvtSec_p->section_length =
-		sizeof(struct cca_pvt_ext_CRT_sec) + keyPartsLen;
-	pvtSec_p->key_format = CCA_PVT_EXT_CRT_SEC_FMT_CL;
-	pvtSec_p->key_use_flags[0] = CCA_PVT_USAGE_ALL;
-	pvtSec_p->p_len = long_len;
-	pvtSec_p->q_len = short_len;
-	pvtSec_p->dp_len = long_len;
-	pvtSec_p->dq_len = short_len;
-	pvtSec_p->u_len = long_len;
-	pvtSec_p->mod_len = mod_len;
-	pvtSec_p->pad_len = pad_len;
-	tgt_p += sizeof(struct cca_pvt_ext_CRT_sec);
-	if (copy_from_user(tgt_p, icaMsg_p->np_prime, long_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, long_len))
-		return SEN_USER_ERROR;
-	tgt_p += long_len;
-	if (copy_from_user(tgt_p, icaMsg_p->nq_prime, short_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, short_len))
-		return SEN_USER_ERROR;
-	tgt_p += short_len;
-	if (copy_from_user(tgt_p, icaMsg_p->bp_key, long_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, long_len))
-		return SEN_USER_ERROR;
-	tgt_p += long_len;
-	if (copy_from_user(tgt_p, icaMsg_p->bq_key, short_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, short_len))
-		return SEN_USER_ERROR;
-	tgt_p += short_len;
-	if (copy_from_user(tgt_p, icaMsg_p->u_mult_inv, long_len))
-		return SEN_RELEASED;
-	if (is_empty(tgt_p, long_len))
-		return SEN_USER_ERROR;
-	tgt_p += long_len;
-	tgt_p += pad_len;
-	memset(tgt_p, 0xFF, mod_len);
-	tgt_p += mod_len;
-	memcpy(tgt_p, &static_cca_pub_sec, sizeof(struct cca_public_sec));
-	pubSec_p = (struct cca_public_sec *) tgt_p;
-	pubSec_p->modulus_bit_len = 8 * mod_len;
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICAMEX_msg_to_type50MEX_msg(struct ica_rsa_modexpo *icaMex_p, int *z90cMsg_l_p,
-			    union type50_msg *z90cMsg_p)
-{
-	int mod_len, msg_size, mod_tgt_len, exp_tgt_len, inp_tgt_len;
-	unsigned char *mod_tgt, *exp_tgt, *inp_tgt;
-	union type50_msg *tmp_type50_msg;
-
-	mod_len = icaMex_p->inputdatalength;
-
-	msg_size = ((mod_len <= 128) ? TYPE50_MEB1_LEN : TYPE50_MEB2_LEN) +
-		    CALLER_HEADER;
-
-	memset(z90cMsg_p, 0, msg_size);
-
-	tmp_type50_msg = (union type50_msg *)
-		((unsigned char *) z90cMsg_p + CALLER_HEADER);
-
-	tmp_type50_msg->meb1.header.msg_type_code = TYPE50_TYPE_CODE;
-
-	if (mod_len <= 128) {
-		tmp_type50_msg->meb1.header.msg_len = TYPE50_MEB1_LEN;
-		tmp_type50_msg->meb1.keyblock_type = TYPE50_MEB1_FMT;
-		mod_tgt = tmp_type50_msg->meb1.modulus;
-		mod_tgt_len = sizeof(tmp_type50_msg->meb1.modulus);
-		exp_tgt = tmp_type50_msg->meb1.exponent;
-		exp_tgt_len = sizeof(tmp_type50_msg->meb1.exponent);
-		inp_tgt = tmp_type50_msg->meb1.message;
-		inp_tgt_len = sizeof(tmp_type50_msg->meb1.message);
-	} else {
-		tmp_type50_msg->meb2.header.msg_len = TYPE50_MEB2_LEN;
-		tmp_type50_msg->meb2.keyblock_type = TYPE50_MEB2_FMT;
-		mod_tgt = tmp_type50_msg->meb2.modulus;
-		mod_tgt_len = sizeof(tmp_type50_msg->meb2.modulus);
-		exp_tgt = tmp_type50_msg->meb2.exponent;
-		exp_tgt_len = sizeof(tmp_type50_msg->meb2.exponent);
-		inp_tgt = tmp_type50_msg->meb2.message;
-		inp_tgt_len = sizeof(tmp_type50_msg->meb2.message);
-	}
-
-	mod_tgt += (mod_tgt_len - mod_len);
-	if (copy_from_user(mod_tgt, icaMex_p->n_modulus, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(mod_tgt, mod_len))
-		return SEN_USER_ERROR;
-	exp_tgt += (exp_tgt_len - mod_len);
-	if (copy_from_user(exp_tgt, icaMex_p->b_key, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(exp_tgt, mod_len))
-		return SEN_USER_ERROR;
-	inp_tgt += (inp_tgt_len - mod_len);
-	if (copy_from_user(inp_tgt, icaMex_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(inp_tgt, mod_len))
-		return SEN_USER_ERROR;
-
-	*z90cMsg_l_p = msg_size - CALLER_HEADER;
-
-	return 0;
-}
-
-static int
-ICACRT_msg_to_type50CRT_msg(struct ica_rsa_modexpo_crt *icaMsg_p,
-			    int *z90cMsg_l_p, union type50_msg *z90cMsg_p)
-{
-	int mod_len, short_len, long_len, tmp_size, p_tgt_len, q_tgt_len,
-	    dp_tgt_len, dq_tgt_len, u_tgt_len, inp_tgt_len, long_offset;
-	unsigned char *p_tgt, *q_tgt, *dp_tgt, *dq_tgt, *u_tgt, *inp_tgt,
-		      temp[8];
-	union type50_msg *tmp_type50_msg;
-
-	mod_len = icaMsg_p->inputdatalength;
-	short_len = mod_len / 2;
-	long_len = mod_len / 2 + 8;
-	long_offset = 0;
-
-	if (long_len > 128) {
-		memset(temp, 0x00, sizeof(temp));
-		if (copy_from_user(temp, icaMsg_p->np_prime, long_len-128))
-			return SEN_RELEASED;
-		if (!is_empty(temp, 8))
-			return SEN_NOT_AVAIL;
-		if (copy_from_user(temp, icaMsg_p->bp_key, long_len-128))
-			return SEN_RELEASED;
-		if (!is_empty(temp, 8))
-			return SEN_NOT_AVAIL;
-		if (copy_from_user(temp, icaMsg_p->u_mult_inv, long_len-128))
-			return SEN_RELEASED;
-		if (!is_empty(temp, 8))
-			return SEN_NOT_AVAIL;
-		long_offset = long_len - 128;
-		long_len = 128;
-	}
-
-	tmp_size = ((long_len <= 64) ? TYPE50_CRB1_LEN : TYPE50_CRB2_LEN) +
-		    CALLER_HEADER;
-
-	memset(z90cMsg_p, 0, tmp_size);
-
-	tmp_type50_msg = (union type50_msg *)
-		((unsigned char *) z90cMsg_p + CALLER_HEADER);
-
-	tmp_type50_msg->crb1.header.msg_type_code = TYPE50_TYPE_CODE;
-	if (long_len <= 64) {
-		tmp_type50_msg->crb1.header.msg_len = TYPE50_CRB1_LEN;
-		tmp_type50_msg->crb1.keyblock_type = TYPE50_CRB1_FMT;
-		p_tgt = tmp_type50_msg->crb1.p;
-		p_tgt_len = sizeof(tmp_type50_msg->crb1.p);
-		q_tgt = tmp_type50_msg->crb1.q;
-		q_tgt_len = sizeof(tmp_type50_msg->crb1.q);
-		dp_tgt = tmp_type50_msg->crb1.dp;
-		dp_tgt_len = sizeof(tmp_type50_msg->crb1.dp);
-		dq_tgt = tmp_type50_msg->crb1.dq;
-		dq_tgt_len = sizeof(tmp_type50_msg->crb1.dq);
-		u_tgt = tmp_type50_msg->crb1.u;
-		u_tgt_len = sizeof(tmp_type50_msg->crb1.u);
-		inp_tgt = tmp_type50_msg->crb1.message;
-		inp_tgt_len = sizeof(tmp_type50_msg->crb1.message);
-	} else {
-		tmp_type50_msg->crb2.header.msg_len = TYPE50_CRB2_LEN;
-		tmp_type50_msg->crb2.keyblock_type = TYPE50_CRB2_FMT;
-		p_tgt = tmp_type50_msg->crb2.p;
-		p_tgt_len = sizeof(tmp_type50_msg->crb2.p);
-		q_tgt = tmp_type50_msg->crb2.q;
-		q_tgt_len = sizeof(tmp_type50_msg->crb2.q);
-		dp_tgt = tmp_type50_msg->crb2.dp;
-		dp_tgt_len = sizeof(tmp_type50_msg->crb2.dp);
-		dq_tgt = tmp_type50_msg->crb2.dq;
-		dq_tgt_len = sizeof(tmp_type50_msg->crb2.dq);
-		u_tgt = tmp_type50_msg->crb2.u;
-		u_tgt_len = sizeof(tmp_type50_msg->crb2.u);
-		inp_tgt = tmp_type50_msg->crb2.message;
-		inp_tgt_len = sizeof(tmp_type50_msg->crb2.message);
-	}
-
-	p_tgt += (p_tgt_len - long_len);
-	if (copy_from_user(p_tgt, icaMsg_p->np_prime + long_offset, long_len))
-		return SEN_RELEASED;
-	if (is_empty(p_tgt, long_len))
-		return SEN_USER_ERROR;
-	q_tgt += (q_tgt_len - short_len);
-	if (copy_from_user(q_tgt, icaMsg_p->nq_prime, short_len))
-		return SEN_RELEASED;
-	if (is_empty(q_tgt, short_len))
-		return SEN_USER_ERROR;
-	dp_tgt += (dp_tgt_len - long_len);
-	if (copy_from_user(dp_tgt, icaMsg_p->bp_key + long_offset, long_len))
-		return SEN_RELEASED;
-	if (is_empty(dp_tgt, long_len))
-		return SEN_USER_ERROR;
-	dq_tgt += (dq_tgt_len - short_len);
-	if (copy_from_user(dq_tgt, icaMsg_p->bq_key, short_len))
-		return SEN_RELEASED;
-	if (is_empty(dq_tgt, short_len))
-		return SEN_USER_ERROR;
-	u_tgt += (u_tgt_len - long_len);
-	if (copy_from_user(u_tgt, icaMsg_p->u_mult_inv + long_offset, long_len))
-		return SEN_RELEASED;
-	if (is_empty(u_tgt, long_len))
-		return SEN_USER_ERROR;
-	inp_tgt += (inp_tgt_len - mod_len);
-	if (copy_from_user(inp_tgt, icaMsg_p->inputdata, mod_len))
-		return SEN_RELEASED;
-	if (is_empty(inp_tgt, mod_len))
-		return SEN_USER_ERROR;
-
-	*z90cMsg_l_p = tmp_size - CALLER_HEADER;
-
-	return 0;
-}
-
-int
-convert_request(unsigned char *buffer, int func, unsigned short function,
-		int cdx, int dev_type, int *msg_l_p, unsigned char *msg_p)
-{
-	if (dev_type == PCICA) {
-		if (func == ICARSACRT)
-			return ICACRT_msg_to_type4CRT_msg(
-				(struct ica_rsa_modexpo_crt *) buffer,
-				msg_l_p, (union type4_msg *) msg_p);
-		else
-			return ICAMEX_msg_to_type4MEX_msg(
-				(struct ica_rsa_modexpo *) buffer,
-				msg_l_p, (union type4_msg *) msg_p);
-	}
-	if (dev_type == PCICC) {
-		if (func == ICARSACRT)
-			return ICACRT_msg_to_type6CRT_msg(
-				(struct ica_rsa_modexpo_crt *) buffer,
-				cdx, msg_l_p, (struct type6_msg *)msg_p);
-		if (function == PCI_FUNC_KEY_ENCRYPT)
-			return ICAMEX_msg_to_type6MEX_en_msg(
-				(struct ica_rsa_modexpo *) buffer,
-				cdx, msg_l_p, (struct type6_msg *) msg_p);
-		else
-			return ICAMEX_msg_to_type6MEX_de_msg(
-				(struct ica_rsa_modexpo *) buffer,
-				cdx, msg_l_p, (struct type6_msg *) msg_p);
-	}
-	if ((dev_type == PCIXCC_MCL2) ||
-	    (dev_type == PCIXCC_MCL3) ||
-	    (dev_type == CEX2C)) {
-		if (func == ICARSACRT)
-			return ICACRT_msg_to_type6CRT_msgX(
-				(struct ica_rsa_modexpo_crt *) buffer,
-				cdx, msg_l_p, (struct type6_msg *) msg_p,
-				dev_type);
-		else
-			return ICAMEX_msg_to_type6MEX_msgX(
-				(struct ica_rsa_modexpo *) buffer,
-				cdx, msg_l_p, (struct type6_msg *) msg_p,
-				dev_type);
-	}
-	if (dev_type == CEX2A) {
-		if (func == ICARSACRT)
-			return ICACRT_msg_to_type50CRT_msg(
-				(struct ica_rsa_modexpo_crt *) buffer,
-				msg_l_p, (union type50_msg *) msg_p);
-		else
-			return ICAMEX_msg_to_type50MEX_msg(
-				(struct ica_rsa_modexpo *) buffer,
-				msg_l_p, (union type50_msg *) msg_p);
-	}
-
-	return 0;
-}
-
-int ext_bitlens_msg_count = 0;
-static inline void
-unset_ext_bitlens(void)
-{
-	if (!ext_bitlens_msg_count) {
-		PRINTK("Unable to use coprocessors for extended bitlengths. "
-		       "Using PCICAs/CEX2As (if present) for extended "
-		       "bitlengths. This is not an error.\n");
-		ext_bitlens_msg_count++;
-	}
-	ext_bitlens = 0;
-}
-
-int
-convert_response(unsigned char *response, unsigned char *buffer,
-		 int *respbufflen_p, unsigned char *resp_buff)
-{
-	struct ica_rsa_modexpo *icaMsg_p = (struct ica_rsa_modexpo *) buffer;
-	struct error_hdr *errh_p = (struct error_hdr *) response;
-	struct type80_hdr *t80h_p = (struct type80_hdr *) response;
-	struct type84_hdr *t84h_p = (struct type84_hdr *) response;
-	struct type86_fmt2_msg *t86m_p =  (struct type86_fmt2_msg *) response;
-	int reply_code, service_rc, service_rs, src_l;
-	unsigned char *src_p, *tgt_p;
-	struct CPRB *cprb_p;
-	struct CPRBX *cprbx_p;
-
-	src_p = 0;
-	reply_code = 0;
-	service_rc = 0;
-	service_rs = 0;
-	src_l = 0;
-	switch (errh_p->type) {
-	case TYPE82_RSP_CODE:
-	case TYPE88_RSP_CODE:
-		reply_code = errh_p->reply_code;
-		src_p = (unsigned char *)errh_p;
-		PRINTK("Hardware error: Type %02X Message Header: "
-		       "%02x%02x%02x%02x%02x%02x%02x%02x\n",
-		       errh_p->type,
-		       src_p[0], src_p[1], src_p[2], src_p[3],
-		       src_p[4], src_p[5], src_p[6], src_p[7]);
-		break;
-	case TYPE80_RSP_CODE:
-		src_l = icaMsg_p->outputdatalength;
-		src_p = response + (int)t80h_p->len - src_l;
-		break;
-	case TYPE84_RSP_CODE:
-		src_l = icaMsg_p->outputdatalength;
-		src_p = response + (int)t84h_p->len - src_l;
-		break;
-	case TYPE86_RSP_CODE:
-		reply_code = t86m_p->header.reply_code;
-		if (reply_code != 0)
-			break;
-		cprb_p = (struct CPRB *)
-			(response + sizeof(struct type86_fmt2_msg));
-		cprbx_p = (struct CPRBX *) cprb_p;
-		if (cprb_p->cprb_ver_id != 0x02) {
-			le2toI(cprb_p->ccp_rtcode, &service_rc);
-			if (service_rc != 0) {
-				le2toI(cprb_p->ccp_rscode, &service_rs);
-				if ((service_rc == 8) && (service_rs == 66))
-					PDEBUG("Bad block format on PCICC\n");
-				else if ((service_rc == 8) && (service_rs == 65))
-					PDEBUG("Probably an even modulus on "
-					       "PCICC\n");
-				else if ((service_rc == 8) && (service_rs == 770)) {
-					PDEBUG("Invalid key length on PCICC\n");
-					unset_ext_bitlens();
-					return REC_USE_PCICA;
-				}
-				else if ((service_rc == 8) && (service_rs == 783)) {
-					PDEBUG("Extended bitlengths not enabled"
-					       "on PCICC\n");
-					unset_ext_bitlens();
-					return REC_USE_PCICA;
-				}
-				else
-					PRINTK("service rc/rs (PCICC): %d/%d\n",
-					       service_rc, service_rs);
-				return REC_OPERAND_INV;
-			}
-			src_p = (unsigned char *)cprb_p + sizeof(struct CPRB);
-			src_p += 4;
-			le2toI(src_p, &src_l);
-			src_l -= 2;
-			src_p += 2;
-		} else {
-			service_rc = (int)cprbx_p->ccp_rtcode;
-			if (service_rc != 0) {
-				service_rs = (int) cprbx_p->ccp_rscode;
-				if ((service_rc == 8) && (service_rs == 66))
-					PDEBUG("Bad block format on PCIXCC\n");
-				else if ((service_rc == 8) && (service_rs == 65))
-					PDEBUG("Probably an even modulus on "
-					       "PCIXCC\n");
-				else if ((service_rc == 8) && (service_rs == 770)) {
-					PDEBUG("Invalid key length on PCIXCC\n");
-					unset_ext_bitlens();
-					return REC_USE_PCICA;
-				}
-				else if ((service_rc == 8) && (service_rs == 783)) {
-					PDEBUG("Extended bitlengths not enabled"
-					       "on PCIXCC\n");
-					unset_ext_bitlens();
-					return REC_USE_PCICA;
-				}
-				else
-					PRINTK("service rc/rs (PCIXCC): %d/%d\n",
-					       service_rc, service_rs);
-				return REC_OPERAND_INV;
-			}
-			src_p = (unsigned char *)
-				cprbx_p + sizeof(struct CPRBX);
-			src_p += 4;
-			src_l = (int)(*((short *) src_p));
-			src_l -= 2;
-			src_p += 2;
-		}
-		break;
-	default:
-		src_p = (unsigned char *)errh_p;
-		PRINTK("Unrecognized Message Header: "
-		       "%02x%02x%02x%02x%02x%02x%02x%02x\n",
-		       src_p[0], src_p[1], src_p[2], src_p[3],
-		       src_p[4], src_p[5], src_p[6], src_p[7]);
-		return REC_BAD_MESSAGE;
-	}
-
-	if (reply_code)
-		switch (reply_code) {
-		case REP82_ERROR_MACHINE_FAILURE:
-			if (errh_p->type == TYPE82_RSP_CODE)
-				PRINTKW("Machine check failure\n");
-			else
-				PRINTKW("Module failure\n");
-			return REC_HARDWAR_ERR;
-		case REP82_ERROR_OPERAND_INVALID:
-			return REC_OPERAND_INV;
-		case REP88_ERROR_MESSAGE_MALFORMD:
-			PRINTKW("Message malformed\n");
-			return REC_OPERAND_INV;
-		case REP82_ERROR_OPERAND_SIZE:
-			return REC_OPERAND_SIZE;
-		case REP82_ERROR_EVEN_MOD_IN_OPND:
-			return REC_EVEN_MOD;
-		case REP82_ERROR_MESSAGE_TYPE:
-			return WRONG_DEVICE_TYPE;
-		case REP82_ERROR_TRANSPORT_FAIL:
-			PRINTKW("Transport failed (APFS = %02X%02X%02X%02X)\n",
-				t86m_p->apfs[0], t86m_p->apfs[1],
-				t86m_p->apfs[2], t86m_p->apfs[3]);
-			return REC_HARDWAR_ERR;
-		default:
-			PRINTKW("reply code = %d\n", reply_code);
-			return REC_HARDWAR_ERR;
-		}
-
-	if (service_rc != 0)
-		return REC_OPERAND_INV;
-
-	if ((src_l > icaMsg_p->outputdatalength) ||
-	    (src_l > RESPBUFFSIZE) ||
-	    (src_l <= 0))
-		return REC_OPERAND_SIZE;
-
-	PDEBUG("Length returned = %d\n", src_l);
-	tgt_p = resp_buff + icaMsg_p->outputdatalength - src_l;
-	memcpy(tgt_p, src_p, src_l);
-	if ((errh_p->type == TYPE86_RSP_CODE) && (resp_buff < tgt_p)) {
-		memset(resp_buff, 0, icaMsg_p->outputdatalength - src_l);
-		if (pad_msg(resp_buff, icaMsg_p->outputdatalength, src_l))
-			return REC_INVALID_PAD;
-	}
-	*respbufflen_p = icaMsg_p->outputdatalength;
-	if (*respbufflen_p == 0)
-		PRINTK("Zero *respbufflen_p\n");
-
-	return 0;
-}
-
diff --git a/drivers/s390/crypto/z90main.c b/drivers/s390/crypto/z90main.c
deleted file mode 100644
index b2f20ab8431a..000000000000
--- a/drivers/s390/crypto/z90main.c
+++ /dev/null
@@ -1,3379 +0,0 @@
-/*
- *  linux/drivers/s390/crypto/z90main.c
- *
- *  z90crypt 1.3.3
- *
- *  Copyright (C)  2001, 2005 IBM Corporation
- *  Author(s): Robert Burroughs (burrough@us.ibm.com)
- *             Eric Rossman (edrossma@us.ibm.com)
- *
- *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#include <asm/uaccess.h>       // copy_(from|to)_user
-#include <linux/compat.h>
-#include <linux/compiler.h>
-#include <linux/delay.h>       // mdelay
-#include <linux/init.h>
-#include <linux/interrupt.h>   // for tasklets
-#include <linux/miscdevice.h>
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/proc_fs.h>
-#include <linux/syscalls.h>
-#include "z90crypt.h"
-#include "z90common.h"
-
-/**
- * Defaults that may be modified.
- */
-
-/**
- * You can specify a different minor at compile time.
- */
-#ifndef Z90CRYPT_MINOR
-#define Z90CRYPT_MINOR	MISC_DYNAMIC_MINOR
-#endif
-
-/**
- * You can specify a different domain at compile time or on the insmod
- * command line.
- */
-#ifndef DOMAIN_INDEX
-#define DOMAIN_INDEX	-1
-#endif
-
-/**
- * This is the name under which the device is registered in /proc/modules.
- */
-#define REG_NAME	"z90crypt"
-
-/**
- * Cleanup should run every CLEANUPTIME seconds and should clean up requests
- * older than CLEANUPTIME seconds in the past.
- */
-#ifndef CLEANUPTIME
-#define CLEANUPTIME 15
-#endif
-
-/**
- * Config should run every CONFIGTIME seconds
- */
-#ifndef CONFIGTIME
-#define CONFIGTIME 30
-#endif
-
-/**
- * The first execution of the config task should take place
- * immediately after initialization
- */
-#ifndef INITIAL_CONFIGTIME
-#define INITIAL_CONFIGTIME 1
-#endif
-
-/**
- * Reader should run every READERTIME milliseconds
- * With the 100Hz patch for s390, z90crypt can lock the system solid while
- * under heavy load. We'll try to avoid that.
- */
-#ifndef READERTIME
-#if HZ > 1000
-#define READERTIME 2
-#else
-#define READERTIME 10
-#endif
-#endif
-
-/**
- * turn long device array index into device pointer
- */
-#define LONG2DEVPTR(ndx) (z90crypt.device_p[(ndx)])
-
-/**
- * turn short device array index into long device array index
- */
-#define SHRT2LONG(ndx) (z90crypt.overall_device_x.device_index[(ndx)])
-
-/**
- * turn short device array index into device pointer
- */
-#define SHRT2DEVPTR(ndx) LONG2DEVPTR(SHRT2LONG(ndx))
-
-/**
- * Status for a work-element
- */
-#define STAT_DEFAULT	0x00 // request has not been processed
-
-#define STAT_ROUTED	0x80 // bit 7: requests get routed to specific device
-			     //	       else, device is determined each write
-#define STAT_FAILED	0x40 // bit 6: this bit is set if the request failed
-			     //	       before being sent to the hardware.
-#define STAT_WRITTEN	0x30 // bits 5-4: work to be done, not sent to device
-//			0x20 // UNUSED state
-#define STAT_READPEND	0x10 // bits 5-4: work done, we're returning data now
-#define STAT_NOWORK	0x00 // bits off: no work on any queue
-#define STAT_RDWRMASK	0x30 // mask for bits 5-4
-
-/**
- * Macros to check the status RDWRMASK
- */
-#define CHK_RDWRMASK(statbyte) ((statbyte) & STAT_RDWRMASK)
-#define SET_RDWRMASK(statbyte, newval) \
-	{(statbyte) &= ~STAT_RDWRMASK; (statbyte) |= newval;}
-
-/**
- * Audit Trail.	 Progress of a Work element
- * audit[0]: Unless noted otherwise, these bits are all set by the process
- */
-#define FP_COPYFROM	0x80 // Caller's buffer has been copied to work element
-#define FP_BUFFREQ	0x40 // Low Level buffer requested
-#define FP_BUFFGOT	0x20 // Low Level buffer obtained
-#define FP_SENT		0x10 // Work element sent to a crypto device
-			     // (may be set by process or by reader task)
-#define FP_PENDING	0x08 // Work element placed on pending queue
-			     // (may be set by process or by reader task)
-#define FP_REQUEST	0x04 // Work element placed on request queue
-#define FP_ASLEEP	0x02 // Work element about to sleep
-#define FP_AWAKE	0x01 // Work element has been awakened
-
-/**
- * audit[1]: These bits are set by the reader task and/or the cleanup task
- */
-#define FP_NOTPENDING	  0x80 // Work element removed from pending queue
-#define FP_AWAKENING	  0x40 // Caller about to be awakened
-#define FP_TIMEDOUT	  0x20 // Caller timed out
-#define FP_RESPSIZESET	  0x10 // Response size copied to work element
-#define FP_RESPADDRCOPIED 0x08 // Response address copied to work element
-#define FP_RESPBUFFCOPIED 0x04 // Response buffer copied to work element
-#define FP_REMREQUEST	  0x02 // Work element removed from request queue
-#define FP_SIGNALED	  0x01 // Work element was awakened by a signal
-
-/**
- * audit[2]: unused
- */
-
-/**
- * state of the file handle in private_data.status
- */
-#define STAT_OPEN 0
-#define STAT_CLOSED 1
-
-/**
- * PID() expands to the process ID of the current process
- */
-#define PID() (current->pid)
-
-/**
- * Selected Constants.	The number of APs and the number of devices
- */
-#ifndef Z90CRYPT_NUM_APS
-#define Z90CRYPT_NUM_APS 64
-#endif
-#ifndef Z90CRYPT_NUM_DEVS
-#define Z90CRYPT_NUM_DEVS Z90CRYPT_NUM_APS
-#endif
-
-/**
- * Buffer size for receiving responses. The maximum Response Size
- * is actually the maximum request size, since in an error condition
- * the request itself may be returned unchanged.
- */
-#define MAX_RESPONSE_SIZE 0x0000077C
-
-/**
- * A count and status-byte mask
- */
-struct status {
-	int	      st_count;		    // # of enabled devices
-	int	      disabled_count;	    // # of disabled devices
-	int	      user_disabled_count;  // # of devices disabled via proc fs
-	unsigned char st_mask[Z90CRYPT_NUM_APS]; // current status mask
-};
-
-/**
- * The array of device indexes is a mechanism for fast indexing into
- * a long (and sparse) array.  For instance, if APs 3, 9 and 47 are
- * installed, z90CDeviceIndex[0] is 3, z90CDeviceIndex[1] is 9, and
- * z90CDeviceIndex[2] is 47.
- */
-struct device_x {
-	int device_index[Z90CRYPT_NUM_DEVS];
-};
-
-/**
- * All devices are arranged in a single array: 64 APs
- */
-struct device {
-	int		 dev_type;	    // PCICA, PCICC, PCIXCC_MCL2,
-					    // PCIXCC_MCL3, CEX2C, CEX2A
-	enum devstat	 dev_stat;	    // current device status
-	int		 dev_self_x;	    // Index in array
-	int		 disabled;	    // Set when device is in error
-	int		 user_disabled;	    // Set when device is disabled by user
-	int		 dev_q_depth;	    // q depth
-	unsigned char *	 dev_resp_p;	    // Response buffer address
-	int		 dev_resp_l;	    // Response Buffer length
-	int		 dev_caller_count;  // Number of callers
-	int		 dev_total_req_cnt; // # requests for device since load
-	struct list_head dev_caller_list;   // List of callers
-};
-
-/**
- * There's a struct status and a struct device_x for each device type.
- */
-struct hdware_block {
-	struct status	hdware_mask;
-	struct status	type_mask[Z90CRYPT_NUM_TYPES];
-	struct device_x type_x_addr[Z90CRYPT_NUM_TYPES];
-	unsigned char	device_type_array[Z90CRYPT_NUM_APS];
-};
-
-/**
- * z90crypt is the topmost data structure in the hierarchy.
- */
-struct z90crypt {
-	int		     max_count;		// Nr of possible crypto devices
-	struct status	     mask;
-	int		     q_depth_array[Z90CRYPT_NUM_DEVS];
-	int		     dev_type_array[Z90CRYPT_NUM_DEVS];
-	struct device_x	     overall_device_x;	// array device indexes
-	struct device *	     device_p[Z90CRYPT_NUM_DEVS];
-	int		     terminating;
-	int		     domain_established;// TRUE:  domain has been found
-	int		     cdx;		// Crypto Domain Index
-	int		     len;		// Length of this data structure
-	struct hdware_block *hdware_info;
-};
-
-/**
- * An array of these structures is pointed to from dev_caller
- * The length of the array depends on the device type. For APs,
- * there are 8.
- *
- * The caller buffer is allocated to the user at OPEN. At WRITE,
- * it contains the request; at READ, the response. The function
- * send_to_crypto_device converts the request to device-dependent
- * form and use the caller's OPEN-allocated buffer for the response.
- *
- * For the contents of caller_dev_dep_req and caller_dev_dep_req_p
- * because that points to it, see the discussion in z90hardware.c.
- * Search for "extended request message block".
- */
-struct caller {
-	int		 caller_buf_l;		 // length of original request
-	unsigned char *	 caller_buf_p;		 // Original request on WRITE
-	int		 caller_dev_dep_req_l;	 // len device dependent request
-	unsigned char *	 caller_dev_dep_req_p;	 // Device dependent form
-	unsigned char	 caller_id[8];		 // caller-supplied message id
-	struct list_head caller_liste;
-	unsigned char	 caller_dev_dep_req[MAX_RESPONSE_SIZE];
-};
-
-/**
- * Function prototypes from z90hardware.c
- */
-enum hdstat query_online(int deviceNr, int cdx, int resetNr, int *q_depth,
-			 int *dev_type);
-enum devstat reset_device(int deviceNr, int cdx, int resetNr);
-enum devstat send_to_AP(int dev_nr, int cdx, int msg_len, unsigned char *msg_ext);
-enum devstat receive_from_AP(int dev_nr, int cdx, int resplen,
-			     unsigned char *resp, unsigned char *psmid);
-int convert_request(unsigned char *buffer, int func, unsigned short function,
-		    int cdx, int dev_type, int *msg_l_p, unsigned char *msg_p);
-int convert_response(unsigned char *response, unsigned char *buffer,
-		     int *respbufflen_p, unsigned char *resp_buff);
-
-/**
- * Low level function prototypes
- */
-static int create_z90crypt(int *cdx_p);
-static int refresh_z90crypt(int *cdx_p);
-static int find_crypto_devices(struct status *deviceMask);
-static int create_crypto_device(int index);
-static int destroy_crypto_device(int index);
-static void destroy_z90crypt(void);
-static int refresh_index_array(struct status *status_str,
-			       struct device_x *index_array);
-static int probe_device_type(struct device *devPtr);
-static int probe_PCIXCC_type(struct device *devPtr);
-
-/**
- * proc fs definitions
- */
-static struct proc_dir_entry *z90crypt_entry;
-
-/**
- * data structures
- */
-
-/**
- * work_element.opener points back to this structure
- */
-struct priv_data {
-	pid_t	opener_pid;
-	unsigned char	status;		// 0: open  1: closed
-};
-
-/**
- * A work element is allocated for each request
- */
-struct work_element {
-	struct priv_data *priv_data;
-	pid_t		  pid;
-	int		  devindex;	  // index of device processing this w_e
-					  // (If request did not specify device,
-					  // -1 until placed onto a queue)
-	int		  devtype;
-	struct list_head  liste;	  // used for requestq and pendingq
-	char		  buffer[128];	  // local copy of user request
-	int		  buff_size;	  // size of the buffer for the request
-	char		  resp_buff[RESPBUFFSIZE];
-	int		  resp_buff_size;
-	char __user *	  resp_addr;	  // address of response in user space
-	unsigned int	  funccode;	  // function code of request
-	wait_queue_head_t waitq;
-	unsigned long	  requestsent;	  // time at which the request was sent
-	atomic_t	  alarmrung;	  // wake-up signal
-	unsigned char	  caller_id[8];	  // pid + counter, for this w_e
-	unsigned char	  status[1];	  // bits to mark status of the request
-	unsigned char	  audit[3];	  // record of work element's progress
-	unsigned char *	  requestptr;	  // address of request buffer
-	int		  retcode;	  // return code of request
-};
-
-/**
- * High level function prototypes
- */
-static int z90crypt_open(struct inode *, struct file *);
-static int z90crypt_release(struct inode *, struct file *);
-static ssize_t z90crypt_read(struct file *, char __user *, size_t, loff_t *);
-static ssize_t z90crypt_write(struct file *, const char __user *,
-							size_t, loff_t *);
-static long z90crypt_unlocked_ioctl(struct file *, unsigned int, unsigned long);
-static long z90crypt_compat_ioctl(struct file *, unsigned int, unsigned long);
-
-static void z90crypt_reader_task(unsigned long);
-static void z90crypt_schedule_reader_task(unsigned long);
-static void z90crypt_config_task(unsigned long);
-static void z90crypt_cleanup_task(unsigned long);
-
-static int z90crypt_status(char *, char **, off_t, int, int *, void *);
-static int z90crypt_status_write(struct file *, const char __user *,
-				 unsigned long, void *);
-
-/**
- * Storage allocated at initialization and used throughout the life of
- * this insmod
- */
-static int domain = DOMAIN_INDEX;
-static struct z90crypt z90crypt;
-static int quiesce_z90crypt;
-static spinlock_t queuespinlock;
-static struct list_head request_list;
-static int requestq_count;
-static struct list_head pending_list;
-static int pendingq_count;
-
-static struct tasklet_struct reader_tasklet;
-static struct timer_list reader_timer;
-static struct timer_list config_timer;
-static struct timer_list cleanup_timer;
-static atomic_t total_open;
-static atomic_t z90crypt_step;
-
-static struct file_operations z90crypt_fops = {
-	.owner		= THIS_MODULE,
-	.read		= z90crypt_read,
-	.write		= z90crypt_write,
-	.unlocked_ioctl	= z90crypt_unlocked_ioctl,
-#ifdef CONFIG_COMPAT
-	.compat_ioctl	= z90crypt_compat_ioctl,
-#endif
-	.open		= z90crypt_open,
-	.release	= z90crypt_release
-};
-
-static struct miscdevice z90crypt_misc_device = {
-	.minor	    = Z90CRYPT_MINOR,
-	.name	    = DEV_NAME,
-	.fops	    = &z90crypt_fops,
-};
-
-/**
- * Documentation values.
- */
-MODULE_AUTHOR("zSeries Linux Crypto Team: Robert H. Burroughs, Eric D. Rossman"
-	      "and Jochen Roehrig");
-MODULE_DESCRIPTION("zSeries Linux Cryptographic Coprocessor device driver, "
-		   "Copyright 2001, 2005 IBM Corporation");
-MODULE_LICENSE("GPL");
-module_param(domain, int, 0);
-MODULE_PARM_DESC(domain, "domain index for device");
-
-#ifdef CONFIG_COMPAT
-/**
- * ioctl32 conversion routines
- */
-struct ica_rsa_modexpo_32 { // For 32-bit callers
-	compat_uptr_t	inputdata;
-	unsigned int	inputdatalength;
-	compat_uptr_t	outputdata;
-	unsigned int	outputdatalength;
-	compat_uptr_t	b_key;
-	compat_uptr_t	n_modulus;
-};
-
-static long
-trans_modexpo32(struct file *filp, unsigned int cmd, unsigned long arg)
-{
-	struct ica_rsa_modexpo_32 __user *mex32u = compat_ptr(arg);
-	struct ica_rsa_modexpo_32  mex32k;
-	struct ica_rsa_modexpo __user *mex64;
-	long ret = 0;
-	unsigned int i;
-
-	if (!access_ok(VERIFY_WRITE, mex32u, sizeof(struct ica_rsa_modexpo_32)))
-		return -EFAULT;
-	mex64 = compat_alloc_user_space(sizeof(struct ica_rsa_modexpo));
-	if (!access_ok(VERIFY_WRITE, mex64, sizeof(struct ica_rsa_modexpo)))
-		return -EFAULT;
-	if (copy_from_user(&mex32k, mex32u, sizeof(struct ica_rsa_modexpo_32)))
-		return -EFAULT;
-	if (__put_user(compat_ptr(mex32k.inputdata), &mex64->inputdata)   ||
-	    __put_user(mex32k.inputdatalength, &mex64->inputdatalength)   ||
-	    __put_user(compat_ptr(mex32k.outputdata), &mex64->outputdata) ||
-	    __put_user(mex32k.outputdatalength, &mex64->outputdatalength) ||
-	    __put_user(compat_ptr(mex32k.b_key), &mex64->b_key)           ||
-	    __put_user(compat_ptr(mex32k.n_modulus), &mex64->n_modulus))
-		return -EFAULT;
-	ret = z90crypt_unlocked_ioctl(filp, cmd, (unsigned long)mex64);
-	if (!ret)
-		if (__get_user(i, &mex64->outputdatalength) ||
-		    __put_user(i, &mex32u->outputdatalength))
-			ret = -EFAULT;
-	return ret;
-}
-
-struct ica_rsa_modexpo_crt_32 { // For 32-bit callers
-	compat_uptr_t	inputdata;
-	unsigned int	inputdatalength;
-	compat_uptr_t	outputdata;
-	unsigned int	outputdatalength;
-	compat_uptr_t	bp_key;
-	compat_uptr_t	bq_key;
-	compat_uptr_t	np_prime;
-	compat_uptr_t	nq_prime;
-	compat_uptr_t	u_mult_inv;
-};
-
-static long
-trans_modexpo_crt32(struct file *filp, unsigned int cmd, unsigned long arg)
-{
-	struct ica_rsa_modexpo_crt_32 __user *crt32u = compat_ptr(arg);
-	struct ica_rsa_modexpo_crt_32  crt32k;
-	struct ica_rsa_modexpo_crt __user *crt64;
-	long ret = 0;
-	unsigned int i;
-
-	if (!access_ok(VERIFY_WRITE, crt32u,
-		       sizeof(struct ica_rsa_modexpo_crt_32)))
-		return -EFAULT;
-	crt64 = compat_alloc_user_space(sizeof(struct ica_rsa_modexpo_crt));
-	if (!access_ok(VERIFY_WRITE, crt64, sizeof(struct ica_rsa_modexpo_crt)))
-		return -EFAULT;
-	if (copy_from_user(&crt32k, crt32u,
-			   sizeof(struct ica_rsa_modexpo_crt_32)))
-		return -EFAULT;
-	if (__put_user(compat_ptr(crt32k.inputdata), &crt64->inputdata)   ||
-	    __put_user(crt32k.inputdatalength, &crt64->inputdatalength)   ||
-	    __put_user(compat_ptr(crt32k.outputdata), &crt64->outputdata) ||
-	    __put_user(crt32k.outputdatalength, &crt64->outputdatalength) ||
-	    __put_user(compat_ptr(crt32k.bp_key), &crt64->bp_key)         ||
-	    __put_user(compat_ptr(crt32k.bq_key), &crt64->bq_key)         ||
-	    __put_user(compat_ptr(crt32k.np_prime), &crt64->np_prime)     ||
-	    __put_user(compat_ptr(crt32k.nq_prime), &crt64->nq_prime)     ||
-	    __put_user(compat_ptr(crt32k.u_mult_inv), &crt64->u_mult_inv))
-		return -EFAULT;
-	ret = z90crypt_unlocked_ioctl(filp, cmd, (unsigned long)crt64);
-	if (!ret)
-		if (__get_user(i, &crt64->outputdatalength) ||
-		    __put_user(i, &crt32u->outputdatalength))
-			ret = -EFAULT;
-	return ret;
-}
-
-static long
-z90crypt_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
-{
-	switch (cmd) {
-	case ICAZ90STATUS:
-	case Z90QUIESCE:
-	case Z90STAT_TOTALCOUNT:
-	case Z90STAT_PCICACOUNT:
-	case Z90STAT_PCICCCOUNT:
-	case Z90STAT_PCIXCCCOUNT:
-	case Z90STAT_PCIXCCMCL2COUNT:
-	case Z90STAT_PCIXCCMCL3COUNT:
-	case Z90STAT_CEX2CCOUNT:
-	case Z90STAT_REQUESTQ_COUNT:
-	case Z90STAT_PENDINGQ_COUNT:
-	case Z90STAT_TOTALOPEN_COUNT:
-	case Z90STAT_DOMAIN_INDEX:
-	case Z90STAT_STATUS_MASK:
-	case Z90STAT_QDEPTH_MASK:
-	case Z90STAT_PERDEV_REQCNT:
-		return z90crypt_unlocked_ioctl(filp, cmd, arg);
-	case ICARSAMODEXPO:
-		return trans_modexpo32(filp, cmd, arg);
-	case ICARSACRT:
-		return trans_modexpo_crt32(filp, cmd, arg);
-	default:
-		return -ENOIOCTLCMD;
-  	}
-}
-#endif
-
-/**
- * The module initialization code.
- */
-static int __init
-z90crypt_init_module(void)
-{
-	int result, nresult;
-	struct proc_dir_entry *entry;
-
-	PDEBUG("PID %d\n", PID());
-
-	if ((domain < -1) || (domain > 15)) {
-		PRINTKW("Invalid param: domain = %d.  Not loading.\n", domain);
-		return -EINVAL;
-	}
-
-	/* Register as misc device with given minor (or get a dynamic one). */
-	result = misc_register(&z90crypt_misc_device);
-	if (result < 0) {
-		PRINTKW(KERN_ERR "misc_register (minor %d) failed with %d\n",
-			z90crypt_misc_device.minor, result);
-		return result;
-	}
-
-	PDEBUG("Registered " DEV_NAME " with result %d\n", result);
-
-	result = create_z90crypt(&domain);
-	if (result != 0) {
-		PRINTKW("create_z90crypt (domain index %d) failed with %d.\n",
-			domain, result);
-		result = -ENOMEM;
-		goto init_module_cleanup;
-	}
-
-	if (result == 0) {
-		PRINTKN("Version %d.%d.%d loaded, built on %s %s\n",
-			z90crypt_VERSION, z90crypt_RELEASE, z90crypt_VARIANT,
-			__DATE__, __TIME__);
-		PDEBUG("create_z90crypt (domain index %d) successful.\n",
-		       domain);
-	} else
-		PRINTK("No devices at startup\n");
-
-	/* Initialize globals. */
-	spin_lock_init(&queuespinlock);
-
-	INIT_LIST_HEAD(&pending_list);
-	pendingq_count = 0;
-
-	INIT_LIST_HEAD(&request_list);
-	requestq_count = 0;
-
-	quiesce_z90crypt = 0;
-
-	atomic_set(&total_open, 0);
-	atomic_set(&z90crypt_step, 0);
-
-	/* Set up the cleanup task. */
-	init_timer(&cleanup_timer);
-	cleanup_timer.function = z90crypt_cleanup_task;
-	cleanup_timer.data = 0;
-	cleanup_timer.expires = jiffies + (CLEANUPTIME * HZ);
-	add_timer(&cleanup_timer);
-
-	/* Set up the proc file system */
-	entry = create_proc_entry("driver/z90crypt", 0644, 0);
-	if (entry) {
-		entry->nlink = 1;
-		entry->data = 0;
-		entry->read_proc = z90crypt_status;
-		entry->write_proc = z90crypt_status_write;
-	}
-	else
-		PRINTK("Couldn't create z90crypt proc entry\n");
-	z90crypt_entry = entry;
-
-	/* Set up the configuration task. */
-	init_timer(&config_timer);
-	config_timer.function = z90crypt_config_task;
-	config_timer.data = 0;
-	config_timer.expires = jiffies + (INITIAL_CONFIGTIME * HZ);
-	add_timer(&config_timer);
-
-	/* Set up the reader task */
-	tasklet_init(&reader_tasklet, z90crypt_reader_task, 0);
-	init_timer(&reader_timer);
-	reader_timer.function = z90crypt_schedule_reader_task;
-	reader_timer.data = 0;
-	reader_timer.expires = jiffies + (READERTIME * HZ / 1000);
-	add_timer(&reader_timer);
-
-	return 0; // success
-
-init_module_cleanup:
-	if ((nresult = misc_deregister(&z90crypt_misc_device)))
-		PRINTK("misc_deregister failed with %d.\n", nresult);
-	else
-		PDEBUG("misc_deregister successful.\n");
-
-	return result; // failure
-}
-
-/**
- * The module termination code
- */
-static void __exit
-z90crypt_cleanup_module(void)
-{
-	int nresult;
-
-	PDEBUG("PID %d\n", PID());
-
-	remove_proc_entry("driver/z90crypt", 0);
-
-	if ((nresult = misc_deregister(&z90crypt_misc_device)))
-		PRINTK("misc_deregister failed with %d.\n", nresult);
-	else
-		PDEBUG("misc_deregister successful.\n");
-
-	/* Remove the tasks */
-	tasklet_kill(&reader_tasklet);
-	del_timer(&reader_timer);
-	del_timer(&config_timer);
-	del_timer(&cleanup_timer);
-
-	destroy_z90crypt();
-
-	PRINTKN("Unloaded.\n");
-}
-
-/**
- * Functions running under a process id
- *
- * The I/O functions:
- *     z90crypt_open
- *     z90crypt_release
- *     z90crypt_read
- *     z90crypt_write
- *     z90crypt_unlocked_ioctl
- *     z90crypt_status
- *     z90crypt_status_write
- *	 disable_card
- *	 enable_card
- *
- * Helper functions:
- *     z90crypt_rsa
- *	 z90crypt_prepare
- *	 z90crypt_send
- *	 z90crypt_process_results
- *
- */
-static int
-z90crypt_open(struct inode *inode, struct file *filp)
-{
-	struct priv_data *private_data_p;
-
-	if (quiesce_z90crypt)
-		return -EQUIESCE;
-
-	private_data_p = kzalloc(sizeof(struct priv_data), GFP_KERNEL);
-	if (!private_data_p) {
-		PRINTK("Memory allocate failed\n");
-		return -ENOMEM;
-	}
-
-	private_data_p->status = STAT_OPEN;
-	private_data_p->opener_pid = PID();
-	filp->private_data = private_data_p;
-	atomic_inc(&total_open);
-
-	return 0;
-}
-
-static int
-z90crypt_release(struct inode *inode, struct file *filp)
-{
-	struct priv_data *private_data_p = filp->private_data;
-
-	PDEBUG("PID %d (filp %p)\n", PID(), filp);
-
-	private_data_p->status = STAT_CLOSED;
-	memset(private_data_p, 0, sizeof(struct priv_data));
-	kfree(private_data_p);
-	atomic_dec(&total_open);
-
-	return 0;
-}
-
-/*
- * there are two read functions, of which compile options will choose one
- * without USE_GET_RANDOM_BYTES
- *   => read() always returns -EPERM;
- * otherwise
- *   => read() uses get_random_bytes() kernel function
- */
-#ifndef USE_GET_RANDOM_BYTES
-/**
- * z90crypt_read will not be supported beyond z90crypt 1.3.1
- */
-static ssize_t
-z90crypt_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos)
-{
-	PDEBUG("filp %p (PID %d)\n", filp, PID());
-	return -EPERM;
-}
-#else // we want to use get_random_bytes
-/**
- * read() just returns a string of random bytes.  Since we have no way
- * to generate these cryptographically, we just execute get_random_bytes
- * for the length specified.
- */
-#include <linux/random.h>
-static ssize_t
-z90crypt_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos)
-{
-	unsigned char *temp_buff;
-
-	PDEBUG("filp %p (PID %d)\n", filp, PID());
-
-	if (quiesce_z90crypt)
-		return -EQUIESCE;
-	if (count < 0) {
-		PRINTK("Requested random byte count negative: %ld\n", count);
-		return -EINVAL;
-	}
-	if (count > RESPBUFFSIZE) {
-		PDEBUG("count[%d] > RESPBUFFSIZE", count);
-		return -EINVAL;
-	}
-	if (count == 0)
-		return 0;
-	temp_buff = kmalloc(RESPBUFFSIZE, GFP_KERNEL);
-	if (!temp_buff) {
-		PRINTK("Memory allocate failed\n");
-		return -ENOMEM;
-	}
-	get_random_bytes(temp_buff, count);
-
-	if (copy_to_user(buf, temp_buff, count) != 0) {
-		kfree(temp_buff);
-		return -EFAULT;
-	}
-	kfree(temp_buff);
-	return count;
-}
-#endif
-
-/**
- * Write is is not allowed
- */
-static ssize_t
-z90crypt_write(struct file *filp, const char __user *buf, size_t count, loff_t *f_pos)
-{
-	PDEBUG("filp %p (PID %d)\n", filp, PID());
-	return -EPERM;
-}
-
-/**
- * New status functions
- */
-static inline int
-get_status_totalcount(void)
-{
-	return z90crypt.hdware_info->hdware_mask.st_count;
-}
-
-static inline int
-get_status_PCICAcount(void)
-{
-	return z90crypt.hdware_info->type_mask[PCICA].st_count;
-}
-
-static inline int
-get_status_PCICCcount(void)
-{
-	return z90crypt.hdware_info->type_mask[PCICC].st_count;
-}
-
-static inline int
-get_status_PCIXCCcount(void)
-{
-	return z90crypt.hdware_info->type_mask[PCIXCC_MCL2].st_count +
-	       z90crypt.hdware_info->type_mask[PCIXCC_MCL3].st_count;
-}
-
-static inline int
-get_status_PCIXCCMCL2count(void)
-{
-	return z90crypt.hdware_info->type_mask[PCIXCC_MCL2].st_count;
-}
-
-static inline int
-get_status_PCIXCCMCL3count(void)
-{
-	return z90crypt.hdware_info->type_mask[PCIXCC_MCL3].st_count;
-}
-
-static inline int
-get_status_CEX2Ccount(void)
-{
-	return z90crypt.hdware_info->type_mask[CEX2C].st_count;
-}
-
-static inline int
-get_status_CEX2Acount(void)
-{
-	return z90crypt.hdware_info->type_mask[CEX2A].st_count;
-}
-
-static inline int
-get_status_requestq_count(void)
-{
-	return requestq_count;
-}
-
-static inline int
-get_status_pendingq_count(void)
-{
-	return pendingq_count;
-}
-
-static inline int
-get_status_totalopen_count(void)
-{
-	return atomic_read(&total_open);
-}
-
-static inline int
-get_status_domain_index(void)
-{
-	return z90crypt.cdx;
-}
-
-static inline unsigned char *
-get_status_status_mask(unsigned char status[Z90CRYPT_NUM_APS])
-{
-	int i, ix;
-
-	memcpy(status, z90crypt.hdware_info->device_type_array,
-	       Z90CRYPT_NUM_APS);
-
-	for (i = 0; i < get_status_totalcount(); i++) {
-		ix = SHRT2LONG(i);
-		if (LONG2DEVPTR(ix)->user_disabled)
-			status[ix] = 0x0d;
-	}
-
-	return status;
-}
-
-static inline unsigned char *
-get_status_qdepth_mask(unsigned char qdepth[Z90CRYPT_NUM_APS])
-{
-	int i, ix;
-
-	memset(qdepth, 0, Z90CRYPT_NUM_APS);
-
-	for (i = 0; i < get_status_totalcount(); i++) {
-		ix = SHRT2LONG(i);
-		qdepth[ix] = LONG2DEVPTR(ix)->dev_caller_count;
-	}
-
-	return qdepth;
-}
-
-static inline unsigned int *
-get_status_perdevice_reqcnt(unsigned int reqcnt[Z90CRYPT_NUM_APS])
-{
-	int i, ix;
-
-	memset(reqcnt, 0, Z90CRYPT_NUM_APS * sizeof(int));
-
-	for (i = 0; i < get_status_totalcount(); i++) {
-		ix = SHRT2LONG(i);
-		reqcnt[ix] = LONG2DEVPTR(ix)->dev_total_req_cnt;
-	}
-
-	return reqcnt;
-}
-
-static inline void
-init_work_element(struct work_element *we_p,
-		  struct priv_data *priv_data, pid_t pid)
-{
-	int step;
-
-	we_p->requestptr = (unsigned char *)we_p + sizeof(struct work_element);
-	/* Come up with a unique id for this caller. */
-	step = atomic_inc_return(&z90crypt_step);
-	memcpy(we_p->caller_id+0, (void *) &pid, sizeof(pid));
-	memcpy(we_p->caller_id+4, (void *) &step, sizeof(step));
-	we_p->pid = pid;
-	we_p->priv_data = priv_data;
-	we_p->status[0] = STAT_DEFAULT;
-	we_p->audit[0] = 0x00;
-	we_p->audit[1] = 0x00;
-	we_p->audit[2] = 0x00;
-	we_p->resp_buff_size = 0;
-	we_p->retcode = 0;
-	we_p->devindex = -1;
-	we_p->devtype = -1;
-	atomic_set(&we_p->alarmrung, 0);
-	init_waitqueue_head(&we_p->waitq);
-	INIT_LIST_HEAD(&(we_p->liste));
-}
-
-static inline int
-allocate_work_element(struct work_element **we_pp,
-		      struct priv_data *priv_data_p, pid_t pid)
-{
-	struct work_element *we_p;
-
-	we_p = (struct work_element *) get_zeroed_page(GFP_KERNEL);
-	if (!we_p)
-		return -ENOMEM;
-	init_work_element(we_p, priv_data_p, pid);
-	*we_pp = we_p;
-	return 0;
-}
-
-static inline void
-remove_device(struct device *device_p)
-{
-	if (!device_p || (device_p->disabled != 0))
-		return;
-	device_p->disabled = 1;
-	z90crypt.hdware_info->type_mask[device_p->dev_type].disabled_count++;
-	z90crypt.hdware_info->hdware_mask.disabled_count++;
-}
-
-/**
- * Bitlength limits for each card
- *
- * There are new MCLs which allow more bitlengths. See the table for details.
- * The MCL must be applied and the newer bitlengths enabled for these to work.
- *
- * Card Type    Old limit    New limit
- * PCICA          ??-2048     same (the lower limit is less than 128 bit...)
- * PCICC         512-1024     512-2048
- * PCIXCC_MCL2   512-2048     ----- (applying any GA LIC will make an MCL3 card)
- * PCIXCC_MCL3   -----        128-2048
- * CEX2C         512-2048     128-2048
- * CEX2A          ??-2048     same (the lower limit is less than 128 bit...)
- *
- * ext_bitlens (extended bitlengths) is a global, since you should not apply an
- * MCL to just one card in a machine. We assume, at first, that all cards have
- * these capabilities.
- */
-int ext_bitlens = 1; // This is global
-#define PCIXCC_MIN_MOD_SIZE	 16	//  128 bits
-#define OLD_PCIXCC_MIN_MOD_SIZE	 64	//  512 bits
-#define PCICC_MIN_MOD_SIZE	 64	//  512 bits
-#define OLD_PCICC_MAX_MOD_SIZE	128	// 1024 bits
-#define MAX_MOD_SIZE		256	// 2048 bits
-
-static inline int
-select_device_type(int *dev_type_p, int bytelength)
-{
-	static int count = 0;
-	int PCICA_avail, PCIXCC_MCL3_avail, CEX2C_avail, CEX2A_avail,
-	    index_to_use;
-	struct status *stat;
-	if ((*dev_type_p != PCICC) && (*dev_type_p != PCICA) &&
-	    (*dev_type_p != PCIXCC_MCL2) && (*dev_type_p != PCIXCC_MCL3) &&
-	    (*dev_type_p != CEX2C) && (*dev_type_p != CEX2A) &&
-	    (*dev_type_p != ANYDEV))
-		return -1;
-	if (*dev_type_p != ANYDEV) {
-		stat = &z90crypt.hdware_info->type_mask[*dev_type_p];
-		if (stat->st_count >
-		    (stat->disabled_count + stat->user_disabled_count))
-			return 0;
-		return -1;
-	}
-
-	/**
-	 * Assumption: PCICA, PCIXCC_MCL3, CEX2C, and CEX2A are all similar in
-	 * speed.
-	 *
-	 * PCICA and CEX2A do NOT co-exist, so it would be either one or the
-	 * other present.
-	 */
-	stat = &z90crypt.hdware_info->type_mask[PCICA];
-	PCICA_avail = stat->st_count -
-			(stat->disabled_count + stat->user_disabled_count);
-	stat = &z90crypt.hdware_info->type_mask[PCIXCC_MCL3];
-	PCIXCC_MCL3_avail = stat->st_count -
-			(stat->disabled_count + stat->user_disabled_count);
-	stat = &z90crypt.hdware_info->type_mask[CEX2C];
-	CEX2C_avail = stat->st_count -
-			(stat->disabled_count + stat->user_disabled_count);
-	stat = &z90crypt.hdware_info->type_mask[CEX2A];
-	CEX2A_avail = stat->st_count -
-			(stat->disabled_count + stat->user_disabled_count);
-	if (PCICA_avail || PCIXCC_MCL3_avail || CEX2C_avail || CEX2A_avail) {
-		/**
-		 * bitlength is a factor, PCICA or CEX2A are the most capable,
-		 * even with the new MCL for PCIXCC.
-		 */
-		if ((bytelength < PCIXCC_MIN_MOD_SIZE) ||
-		    (!ext_bitlens && (bytelength < OLD_PCIXCC_MIN_MOD_SIZE))) {
-			if (PCICA_avail) {
-				*dev_type_p = PCICA;
-				return 0;
-			}
-			if (CEX2A_avail) {
-				*dev_type_p = CEX2A;
-				return 0;
-			}
-			return -1;
-		}
-
-		index_to_use = count % (PCICA_avail + PCIXCC_MCL3_avail +
-					CEX2C_avail + CEX2A_avail);
-		if (index_to_use < PCICA_avail)
-			*dev_type_p = PCICA;
-		else if (index_to_use < (PCICA_avail + PCIXCC_MCL3_avail))
-			*dev_type_p = PCIXCC_MCL3;
-		else if (index_to_use < (PCICA_avail + PCIXCC_MCL3_avail +
-					 CEX2C_avail))
-			*dev_type_p = CEX2C;
-		else
-			*dev_type_p = CEX2A;
-		count++;
-		return 0;
-	}
-
-	/* Less than OLD_PCIXCC_MIN_MOD_SIZE cannot go to a PCIXCC_MCL2 */
-	if (bytelength < OLD_PCIXCC_MIN_MOD_SIZE)
-		return -1;
-	stat = &z90crypt.hdware_info->type_mask[PCIXCC_MCL2];
-	if (stat->st_count >
-	    (stat->disabled_count + stat->user_disabled_count)) {
-		*dev_type_p = PCIXCC_MCL2;
-		return 0;
-	}
-
-	/**
-	 * Less than PCICC_MIN_MOD_SIZE or more than OLD_PCICC_MAX_MOD_SIZE
-	 * (if we don't have the MCL applied and the newer bitlengths enabled)
-	 * cannot go to a PCICC
-	 */
-	if ((bytelength < PCICC_MIN_MOD_SIZE) ||
-	    (!ext_bitlens && (bytelength > OLD_PCICC_MAX_MOD_SIZE))) {
-		return -1;
-	}
-	stat = &z90crypt.hdware_info->type_mask[PCICC];
-	if (stat->st_count >
-	    (stat->disabled_count + stat->user_disabled_count)) {
-		*dev_type_p = PCICC;
-		return 0;
-	}
-
-	return -1;
-}
-
-/**
- * Try the selected number, then the selected type (can be ANYDEV)
- */
-static inline int
-select_device(int *dev_type_p, int *device_nr_p, int bytelength)
-{
-	int i, indx, devTp, low_count, low_indx;
-	struct device_x *index_p;
-	struct device *dev_ptr;
-
-	PDEBUG("device type = %d, index = %d\n", *dev_type_p, *device_nr_p);
-	if ((*device_nr_p >= 0) && (*device_nr_p < Z90CRYPT_NUM_DEVS)) {
-		PDEBUG("trying index = %d\n", *device_nr_p);
-		dev_ptr = z90crypt.device_p[*device_nr_p];
-
-		if (dev_ptr &&
-		    (dev_ptr->dev_stat != DEV_GONE) &&
-		    (dev_ptr->disabled == 0) &&
-		    (dev_ptr->user_disabled == 0)) {
-			PDEBUG("selected by number, index = %d\n",
-			       *device_nr_p);
-			*dev_type_p = dev_ptr->dev_type;
-			return *device_nr_p;
-		}
-	}
-	*device_nr_p = -1;
-	PDEBUG("trying type = %d\n", *dev_type_p);
-	devTp = *dev_type_p;
-	if (select_device_type(&devTp, bytelength) == -1) {
-		PDEBUG("failed to select by type\n");
-		return -1;
-	}
-	PDEBUG("selected type = %d\n", devTp);
-	index_p = &z90crypt.hdware_info->type_x_addr[devTp];
-	low_count = 0x0000FFFF;
-	low_indx = -1;
-	for (i = 0; i < z90crypt.hdware_info->type_mask[devTp].st_count; i++) {
-		indx = index_p->device_index[i];
-		dev_ptr = z90crypt.device_p[indx];
-		if (dev_ptr &&
-		    (dev_ptr->dev_stat != DEV_GONE) &&
-		    (dev_ptr->disabled == 0) &&
-		    (dev_ptr->user_disabled == 0) &&
-		    (devTp == dev_ptr->dev_type) &&
-		    (low_count > dev_ptr->dev_caller_count)) {
-			low_count = dev_ptr->dev_caller_count;
-			low_indx = indx;
-		}
-	}
-	*device_nr_p = low_indx;
-	return low_indx;
-}
-
-static inline int
-send_to_crypto_device(struct work_element *we_p)
-{
-	struct caller *caller_p;
-	struct device *device_p;
-	int dev_nr;
-	int bytelen = ((struct ica_rsa_modexpo *)we_p->buffer)->inputdatalength;
-
-	if (!we_p->requestptr)
-		return SEN_FATAL_ERROR;
-	caller_p = (struct caller *)we_p->requestptr;
-	dev_nr = we_p->devindex;
-	if (select_device(&we_p->devtype, &dev_nr, bytelen) == -1) {
-		if (z90crypt.hdware_info->hdware_mask.st_count != 0)
-			return SEN_RETRY;
-		else
-			return SEN_NOT_AVAIL;
-	}
-	we_p->devindex = dev_nr;
-	device_p = z90crypt.device_p[dev_nr];
-	if (!device_p)
-		return SEN_NOT_AVAIL;
-	if (device_p->dev_type != we_p->devtype)
-		return SEN_RETRY;
-	if (device_p->dev_caller_count >= device_p->dev_q_depth)
-		return SEN_QUEUE_FULL;
-	PDEBUG("device number prior to send: %d\n", dev_nr);
-	switch (send_to_AP(dev_nr, z90crypt.cdx,
-			   caller_p->caller_dev_dep_req_l,
-			   caller_p->caller_dev_dep_req_p)) {
-	case DEV_SEN_EXCEPTION:
-		PRINTKC("Exception during send to device %d\n", dev_nr);
-		z90crypt.terminating = 1;
-		return SEN_FATAL_ERROR;
-	case DEV_GONE:
-		PRINTK("Device %d not available\n", dev_nr);
-		remove_device(device_p);
-		return SEN_NOT_AVAIL;
-	case DEV_EMPTY:
-		return SEN_NOT_AVAIL;
-	case DEV_NO_WORK:
-		return SEN_FATAL_ERROR;
-	case DEV_BAD_MESSAGE:
-		return SEN_USER_ERROR;
-	case DEV_QUEUE_FULL:
-		return SEN_QUEUE_FULL;
-	default:
-	case DEV_ONLINE:
-		break;
-	}
-	list_add_tail(&(caller_p->caller_liste), &(device_p->dev_caller_list));
-	device_p->dev_caller_count++;
-	return 0;
-}
-
-/**
- * Send puts the user's work on one of two queues:
- *   the pending queue if the send was successful
- *   the request queue if the send failed because device full or busy
- */
-static inline int
-z90crypt_send(struct work_element *we_p, const char *buf)
-{
-	int rv;
-
-	PDEBUG("PID %d\n", PID());
-
-	if (CHK_RDWRMASK(we_p->status[0]) != STAT_NOWORK) {
-		PDEBUG("PID %d tried to send more work but has outstanding "
-		       "work.\n", PID());
-		return -EWORKPEND;
-	}
-	we_p->devindex = -1; // Reset device number
-	spin_lock_irq(&queuespinlock);
-	rv = send_to_crypto_device(we_p);
-	switch (rv) {
-	case 0:
-		we_p->requestsent = jiffies;
-		we_p->audit[0] |= FP_SENT;
-		list_add_tail(&we_p->liste, &pending_list);
-		++pendingq_count;
-		we_p->audit[0] |= FP_PENDING;
-		break;
-	case SEN_BUSY:
-	case SEN_QUEUE_FULL:
-		rv = 0;
-		we_p->devindex = -1; // any device will do
-		we_p->requestsent = jiffies;
-		list_add_tail(&we_p->liste, &request_list);
-		++requestq_count;
-		we_p->audit[0] |= FP_REQUEST;
-		break;
-	case SEN_RETRY:
-		rv = -ERESTARTSYS;
-		break;
-	case SEN_NOT_AVAIL:
-		PRINTK("*** No devices available.\n");
-		rv = we_p->retcode = -ENODEV;
-		we_p->status[0] |= STAT_FAILED;
-		break;
-	case REC_OPERAND_INV:
-	case REC_OPERAND_SIZE:
-	case REC_EVEN_MOD:
-	case REC_INVALID_PAD:
-		rv = we_p->retcode = -EINVAL;
-		we_p->status[0] |= STAT_FAILED;
-		break;
-	default:
-		we_p->retcode = rv;
-		we_p->status[0] |= STAT_FAILED;
-		break;
-	}
-	if (rv != -ERESTARTSYS)
-		SET_RDWRMASK(we_p->status[0], STAT_WRITTEN);
-	spin_unlock_irq(&queuespinlock);
-	if (rv == 0)
-		tasklet_schedule(&reader_tasklet);
-	return rv;
-}
-
-/**
- * process_results copies the user's work from kernel space.
- */
-static inline int
-z90crypt_process_results(struct work_element *we_p, char __user *buf)
-{
-	int rv;
-
-	PDEBUG("we_p %p (PID %d)\n", we_p, PID());
-
-	LONG2DEVPTR(we_p->devindex)->dev_total_req_cnt++;
-	SET_RDWRMASK(we_p->status[0], STAT_READPEND);
-
-	rv = 0;
-	if (!we_p->buffer) {
-		PRINTK("we_p %p PID %d in STAT_READPEND: buffer NULL.\n",
-			we_p, PID());
-		rv = -ENOBUFF;
-	}
-
-	if (!rv)
-		if ((rv = copy_to_user(buf, we_p->buffer, we_p->buff_size))) {
-			PDEBUG("copy_to_user failed: rv = %d\n", rv);
-			rv = -EFAULT;
-		}
-
-	if (!rv)
-		rv = we_p->retcode;
-	if (!rv)
-		if (we_p->resp_buff_size
-		    &&	copy_to_user(we_p->resp_addr, we_p->resp_buff,
-				     we_p->resp_buff_size))
-			rv = -EFAULT;
-
-	SET_RDWRMASK(we_p->status[0], STAT_NOWORK);
-	return rv;
-}
-
-static unsigned char NULL_psmid[8] =
-{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
-
-/**
- * Used in device configuration functions
- */
-#define MAX_RESET 90
-
-/**
- * This is used only for PCICC support
- */
-static inline int
-is_PKCS11_padded(unsigned char *buffer, int length)
-{
-	int i;
-	if ((buffer[0] != 0x00) || (buffer[1] != 0x01))
-		return 0;
-	for (i = 2; i < length; i++)
-		if (buffer[i] != 0xFF)
-			break;
-	if ((i < 10) || (i == length))
-		return 0;
-	if (buffer[i] != 0x00)
-		return 0;
-	return 1;
-}
-
-/**
- * This is used only for PCICC support
- */
-static inline int
-is_PKCS12_padded(unsigned char *buffer, int length)
-{
-	int i;
-	if ((buffer[0] != 0x00) || (buffer[1] != 0x02))
-		return 0;
-	for (i = 2; i < length; i++)
-		if (buffer[i] == 0x00)
-			break;
-	if ((i < 10) || (i == length))
-		return 0;
-	if (buffer[i] != 0x00)
-		return 0;
-	return 1;
-}
-
-/**
- * builds struct caller and converts message from generic format to
- * device-dependent format
- * func is ICARSAMODEXPO or ICARSACRT
- * function is PCI_FUNC_KEY_ENCRYPT or PCI_FUNC_KEY_DECRYPT
- */
-static inline int
-build_caller(struct work_element *we_p, short function)
-{
-	int rv;
-	struct caller *caller_p = (struct caller *)we_p->requestptr;
-
-	if ((we_p->devtype != PCICC) && (we_p->devtype != PCICA) &&
-	    (we_p->devtype != PCIXCC_MCL2) && (we_p->devtype != PCIXCC_MCL3) &&
-	    (we_p->devtype != CEX2C) && (we_p->devtype != CEX2A))
-		return SEN_NOT_AVAIL;
-
-	memcpy(caller_p->caller_id, we_p->caller_id,
-	       sizeof(caller_p->caller_id));
-	caller_p->caller_dev_dep_req_p = caller_p->caller_dev_dep_req;
-	caller_p->caller_dev_dep_req_l = MAX_RESPONSE_SIZE;
-	caller_p->caller_buf_p = we_p->buffer;
-	INIT_LIST_HEAD(&(caller_p->caller_liste));
-
-	rv = convert_request(we_p->buffer, we_p->funccode, function,
-			     z90crypt.cdx, we_p->devtype,
-			     &caller_p->caller_dev_dep_req_l,
-			     caller_p->caller_dev_dep_req_p);
-	if (rv) {
-		if (rv == SEN_NOT_AVAIL)
-			PDEBUG("request can't be processed on hdwr avail\n");
-		else
-			PRINTK("Error from convert_request: %d\n", rv);
-	}
-	else
-		memcpy(&(caller_p->caller_dev_dep_req_p[4]), we_p->caller_id,8);
-	return rv;
-}
-
-static inline void
-unbuild_caller(struct device *device_p, struct caller *caller_p)
-{
-	if (!caller_p)
-		return;
-	if (caller_p->caller_liste.next && caller_p->caller_liste.prev)
-		if (!list_empty(&caller_p->caller_liste)) {
-			list_del_init(&caller_p->caller_liste);
-			device_p->dev_caller_count--;
-		}
-	memset(caller_p->caller_id, 0, sizeof(caller_p->caller_id));
-}
-
-static inline int
-get_crypto_request_buffer(struct work_element *we_p)
-{
-	struct ica_rsa_modexpo *mex_p;
-	struct ica_rsa_modexpo_crt *crt_p;
-	unsigned char *temp_buffer;
-	short function;
-	int rv;
-
-	mex_p =	(struct ica_rsa_modexpo *) we_p->buffer;
-	crt_p = (struct ica_rsa_modexpo_crt *) we_p->buffer;
-
-	PDEBUG("device type input = %d\n", we_p->devtype);
-
-	if (z90crypt.terminating)
-		return REC_NO_RESPONSE;
-	if (memcmp(we_p->caller_id, NULL_psmid, 8) == 0) {
-		PRINTK("psmid zeroes\n");
-		return SEN_FATAL_ERROR;
-	}
-	if (!we_p->buffer) {
-		PRINTK("buffer pointer NULL\n");
-		return SEN_USER_ERROR;
-	}
-	if (!we_p->requestptr) {
-		PRINTK("caller pointer NULL\n");
-		return SEN_USER_ERROR;
-	}
-
-	if ((we_p->devtype != PCICA) && (we_p->devtype != PCICC) &&
-	    (we_p->devtype != PCIXCC_MCL2) && (we_p->devtype != PCIXCC_MCL3) &&
-	    (we_p->devtype != CEX2C) && (we_p->devtype != CEX2A) &&
-	    (we_p->devtype != ANYDEV)) {
-		PRINTK("invalid device type\n");
-		return SEN_USER_ERROR;
-	}
-
-	if ((mex_p->inputdatalength < 1) ||
-	    (mex_p->inputdatalength > MAX_MOD_SIZE)) {
-		PRINTK("inputdatalength[%d] is not valid\n",
-		       mex_p->inputdatalength);
-		return SEN_USER_ERROR;
-	}
-
-	if (mex_p->outputdatalength < mex_p->inputdatalength) {
-		PRINTK("outputdatalength[%d] < inputdatalength[%d]\n",
-		       mex_p->outputdatalength, mex_p->inputdatalength);
-		return SEN_USER_ERROR;
-	}
-
-	if (!mex_p->inputdata || !mex_p->outputdata) {
-		PRINTK("inputdata[%p] or outputdata[%p] is NULL\n",
-		       mex_p->outputdata, mex_p->inputdata);
-		return SEN_USER_ERROR;
-	}
-
-	/**
-	 * As long as outputdatalength is big enough, we can set the
-	 * outputdatalength equal to the inputdatalength, since that is the
-	 * number of bytes we will copy in any case
-	 */
-	mex_p->outputdatalength = mex_p->inputdatalength;
-
-	rv = 0;
-	switch (we_p->funccode) {
-	case ICARSAMODEXPO:
-		if (!mex_p->b_key || !mex_p->n_modulus)
-			rv = SEN_USER_ERROR;
-		break;
-	case ICARSACRT:
-		if (!IS_EVEN(crt_p->inputdatalength)) {
-			PRINTK("inputdatalength[%d] is odd, CRT form\n",
-			       crt_p->inputdatalength);
-			rv = SEN_USER_ERROR;
-			break;
-		}
-		if (!crt_p->bp_key ||
-		    !crt_p->bq_key ||
-		    !crt_p->np_prime ||
-		    !crt_p->nq_prime ||
-		    !crt_p->u_mult_inv) {
-			PRINTK("CRT form, bad data: %p/%p/%p/%p/%p\n",
-			       crt_p->bp_key, crt_p->bq_key,
-			       crt_p->np_prime, crt_p->nq_prime,
-			       crt_p->u_mult_inv);
-			rv = SEN_USER_ERROR;
-		}
-		break;
-	default:
-		PRINTK("bad func = %d\n", we_p->funccode);
-		rv = SEN_USER_ERROR;
-		break;
-	}
-	if (rv != 0)
-		return rv;
-
-	if (select_device_type(&we_p->devtype, mex_p->inputdatalength) < 0)
-		return SEN_NOT_AVAIL;
-
-	temp_buffer = (unsigned char *)we_p + sizeof(struct work_element) +
-		      sizeof(struct caller);
-	if (copy_from_user(temp_buffer, mex_p->inputdata,
-			   mex_p->inputdatalength) != 0)
-		return SEN_RELEASED;
-
-	function = PCI_FUNC_KEY_ENCRYPT;
-	switch (we_p->devtype) {
-	/* PCICA and CEX2A do everything with a simple RSA mod-expo operation */
-	case PCICA:
-	case CEX2A:
-		function = PCI_FUNC_KEY_ENCRYPT;
-		break;
-	/**
-	 * PCIXCC_MCL2 does all Mod-Expo form with a simple RSA mod-expo
-	 * operation, and all CRT forms with a PKCS-1.2 format decrypt.
-	 * PCIXCC_MCL3 and CEX2C do all Mod-Expo and CRT forms with a simple RSA
-	 * mod-expo operation
-	 */
-	case PCIXCC_MCL2:
-		if (we_p->funccode == ICARSAMODEXPO)
-			function = PCI_FUNC_KEY_ENCRYPT;
-		else
-			function = PCI_FUNC_KEY_DECRYPT;
-		break;
-	case PCIXCC_MCL3:
-	case CEX2C:
-		if (we_p->funccode == ICARSAMODEXPO)
-			function = PCI_FUNC_KEY_ENCRYPT;
-		else
-			function = PCI_FUNC_KEY_DECRYPT;
-		break;
-	/**
-	 * PCICC does everything as a PKCS-1.2 format request
-	 */
-	case PCICC:
-		/* PCICC cannot handle input that is is PKCS#1.1 padded */
-		if (is_PKCS11_padded(temp_buffer, mex_p->inputdatalength)) {
-			return SEN_NOT_AVAIL;
-		}
-		if (we_p->funccode == ICARSAMODEXPO) {
-			if (is_PKCS12_padded(temp_buffer,
-					     mex_p->inputdatalength))
-				function = PCI_FUNC_KEY_ENCRYPT;
-			else
-				function = PCI_FUNC_KEY_DECRYPT;
-		} else
-			/* all CRT forms are decrypts */
-			function = PCI_FUNC_KEY_DECRYPT;
-		break;
-	}
-	PDEBUG("function: %04x\n", function);
-	rv = build_caller(we_p, function);
-	PDEBUG("rv from build_caller = %d\n", rv);
-	return rv;
-}
-
-static inline int
-z90crypt_prepare(struct work_element *we_p, unsigned int funccode,
-		 const char __user *buffer)
-{
-	int rv;
-
-	we_p->devindex = -1;
-	if (funccode == ICARSAMODEXPO)
-		we_p->buff_size = sizeof(struct ica_rsa_modexpo);
-	else
-		we_p->buff_size = sizeof(struct ica_rsa_modexpo_crt);
-
-	if (copy_from_user(we_p->buffer, buffer, we_p->buff_size))
-		return -EFAULT;
-
-	we_p->audit[0] |= FP_COPYFROM;
-	SET_RDWRMASK(we_p->status[0], STAT_WRITTEN);
-	we_p->funccode = funccode;
-	we_p->devtype = -1;
-	we_p->audit[0] |= FP_BUFFREQ;
-	rv = get_crypto_request_buffer(we_p);
-	switch (rv) {
-	case 0:
-		we_p->audit[0] |= FP_BUFFGOT;
-		break;
-	case SEN_USER_ERROR:
-		rv = -EINVAL;
-		break;
-	case SEN_QUEUE_FULL:
-		rv = 0;
-		break;
-	case SEN_RELEASED:
-		rv = -EFAULT;
-		break;
-	case REC_NO_RESPONSE:
-		rv = -ENODEV;
-		break;
-	case SEN_NOT_AVAIL:
-	case EGETBUFF:
-		rv = -EGETBUFF;
-		break;
-	default:
-		PRINTK("rv = %d\n", rv);
-		rv = -EGETBUFF;
-		break;
-	}
-	if (CHK_RDWRMASK(we_p->status[0]) == STAT_WRITTEN)
-		SET_RDWRMASK(we_p->status[0], STAT_DEFAULT);
-	return rv;
-}
-
-static inline void
-purge_work_element(struct work_element *we_p)
-{
-	struct list_head *lptr;
-
-	spin_lock_irq(&queuespinlock);
-	list_for_each(lptr, &request_list) {
-		if (lptr == &we_p->liste) {
-			list_del_init(lptr);
-			requestq_count--;
-			break;
-		}
-	}
-	list_for_each(lptr, &pending_list) {
-		if (lptr == &we_p->liste) {
-			list_del_init(lptr);
-			pendingq_count--;
-			break;
-		}
-	}
-	spin_unlock_irq(&queuespinlock);
-}
-
-/**
- * Build the request and send it.
- */
-static inline int
-z90crypt_rsa(struct priv_data *private_data_p, pid_t pid,
-	     unsigned int cmd, unsigned long arg)
-{
-	struct work_element *we_p;
-	int rv;
-
-	if ((rv = allocate_work_element(&we_p, private_data_p, pid))) {
-		PDEBUG("PID %d: allocate_work_element returned ENOMEM\n", pid);
-		return rv;
-	}
-	if ((rv = z90crypt_prepare(we_p, cmd, (const char __user *)arg)))
-		PDEBUG("PID %d: rv = %d from z90crypt_prepare\n", pid, rv);
-	if (!rv)
-		if ((rv = z90crypt_send(we_p, (const char *)arg)))
-			PDEBUG("PID %d: rv %d from z90crypt_send.\n", pid, rv);
-	if (!rv) {
-		we_p->audit[0] |= FP_ASLEEP;
-		wait_event(we_p->waitq, atomic_read(&we_p->alarmrung));
-		we_p->audit[0] |= FP_AWAKE;
-		rv = we_p->retcode;
-	}
-	if (!rv)
-		rv = z90crypt_process_results(we_p, (char __user *)arg);
-
-	if ((we_p->status[0] & STAT_FAILED)) {
-		switch (rv) {
-		/**
-		 * EINVAL *after* receive is almost always a padding error or
-		 * length error issued by a coprocessor (not an accelerator).
-		 * We convert this return value to -EGETBUFF which should
-		 * trigger a fallback to software.
-		 */
-		case -EINVAL:
-			if ((we_p->devtype != PCICA) &&
-			    (we_p->devtype != CEX2A))
-				rv = -EGETBUFF;
-			break;
-		case -ETIMEOUT:
-			if (z90crypt.mask.st_count > 0)
-				rv = -ERESTARTSYS; // retry with another
-			else
-				rv = -ENODEV; // no cards left
-		/* fall through to clean up request queue */
-		case -ERESTARTSYS:
-		case -ERELEASED:
-			switch (CHK_RDWRMASK(we_p->status[0])) {
-			case STAT_WRITTEN:
-				purge_work_element(we_p);
-				break;
-			case STAT_READPEND:
-			case STAT_NOWORK:
-			default:
-				break;
-			}
-			break;
-		default:
-			we_p->status[0] ^= STAT_FAILED;
-			break;
-		}
-	}
-	free_page((long)we_p);
-	return rv;
-}
-
-/**
- * This function is a little long, but it's really just one large switch
- * statement.
- */
-static long
-z90crypt_unlocked_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
-{
-	struct priv_data *private_data_p = filp->private_data;
-	unsigned char *status;
-	unsigned char *qdepth;
-	unsigned int *reqcnt;
-	struct ica_z90_status *pstat;
-	int ret, i, loopLim, tempstat;
-	static int deprecated_msg_count1 = 0;
-	static int deprecated_msg_count2 = 0;
-
-	PDEBUG("filp %p (PID %d), cmd 0x%08X\n", filp, PID(), cmd);
-	PDEBUG("cmd 0x%08X: dir %s, size 0x%04X, type 0x%02X, nr 0x%02X\n",
-		cmd,
-		!_IOC_DIR(cmd) ? "NO"
-		: ((_IOC_DIR(cmd) == (_IOC_READ|_IOC_WRITE)) ? "RW"
-		: ((_IOC_DIR(cmd) == _IOC_READ) ? "RD"
-		: "WR")),
-		_IOC_SIZE(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd));
-
-	if (_IOC_TYPE(cmd) != Z90_IOCTL_MAGIC) {
-		PRINTK("cmd 0x%08X contains bad magic\n", cmd);
-		return -ENOTTY;
-	}
-
-	ret = 0;
-	switch (cmd) {
-	case ICARSAMODEXPO:
-	case ICARSACRT:
-		if (quiesce_z90crypt) {
-			ret = -EQUIESCE;
-			break;
-		}
-		ret = -ENODEV; // Default if no devices
-		loopLim = z90crypt.hdware_info->hdware_mask.st_count -
-			(z90crypt.hdware_info->hdware_mask.disabled_count +
-			 z90crypt.hdware_info->hdware_mask.user_disabled_count);
-		for (i = 0; i < loopLim; i++) {
-			ret = z90crypt_rsa(private_data_p, PID(), cmd, arg);
-			if (ret != -ERESTARTSYS)
-				break;
-		}
-		if (ret == -ERESTARTSYS)
-			ret = -ENODEV;
-		break;
-
-	case Z90STAT_TOTALCOUNT:
-		tempstat = get_status_totalcount();
-		if (copy_to_user((int __user *)arg, &tempstat,sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_PCICACOUNT:
-		tempstat = get_status_PCICAcount();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_PCICCCOUNT:
-		tempstat = get_status_PCICCcount();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_PCIXCCMCL2COUNT:
-		tempstat = get_status_PCIXCCMCL2count();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_PCIXCCMCL3COUNT:
-		tempstat = get_status_PCIXCCMCL3count();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_CEX2CCOUNT:
-		tempstat = get_status_CEX2Ccount();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_CEX2ACOUNT:
-		tempstat = get_status_CEX2Acount();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_REQUESTQ_COUNT:
-		tempstat = get_status_requestq_count();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_PENDINGQ_COUNT:
-		tempstat = get_status_pendingq_count();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_TOTALOPEN_COUNT:
-		tempstat = get_status_totalopen_count();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_DOMAIN_INDEX:
-		tempstat = get_status_domain_index();
-		if (copy_to_user((int __user *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90STAT_STATUS_MASK:
-		status = kmalloc(Z90CRYPT_NUM_APS, GFP_KERNEL);
-		if (!status) {
-			PRINTK("kmalloc for status failed!\n");
-			ret = -ENOMEM;
-			break;
-		}
-		get_status_status_mask(status);
-		if (copy_to_user((char __user *) arg, status, Z90CRYPT_NUM_APS)
-									!= 0)
-			ret = -EFAULT;
-		kfree(status);
-		break;
-
-	case Z90STAT_QDEPTH_MASK:
-		qdepth = kmalloc(Z90CRYPT_NUM_APS, GFP_KERNEL);
-		if (!qdepth) {
-			PRINTK("kmalloc for qdepth failed!\n");
-			ret = -ENOMEM;
-			break;
-		}
-		get_status_qdepth_mask(qdepth);
-		if (copy_to_user((char __user *) arg, qdepth, Z90CRYPT_NUM_APS) != 0)
-			ret = -EFAULT;
-		kfree(qdepth);
-		break;
-
-	case Z90STAT_PERDEV_REQCNT:
-		reqcnt = kmalloc(sizeof(int) * Z90CRYPT_NUM_APS, GFP_KERNEL);
-		if (!reqcnt) {
-			PRINTK("kmalloc for reqcnt failed!\n");
-			ret = -ENOMEM;
-			break;
-		}
-		get_status_perdevice_reqcnt(reqcnt);
-		if (copy_to_user((char __user *) arg, reqcnt,
-				 Z90CRYPT_NUM_APS * sizeof(int)) != 0)
-			ret = -EFAULT;
-		kfree(reqcnt);
-		break;
-
-		/* THIS IS DEPRECATED.	USE THE NEW STATUS CALLS */
-	case ICAZ90STATUS:
-		if (deprecated_msg_count1 < 20) {
-			PRINTK("deprecated call to ioctl (ICAZ90STATUS)!\n");
-			deprecated_msg_count1++;
-			if (deprecated_msg_count1 == 20)
-				PRINTK("No longer issuing messages related to "
-				       "deprecated call to ICAZ90STATUS.\n");
-		}
-
-		pstat = kmalloc(sizeof(struct ica_z90_status), GFP_KERNEL);
-		if (!pstat) {
-			PRINTK("kmalloc for pstat failed!\n");
-			ret = -ENOMEM;
-			break;
-		}
-
-		pstat->totalcount	 = get_status_totalcount();
-		pstat->leedslitecount	 = get_status_PCICAcount();
-		pstat->leeds2count	 = get_status_PCICCcount();
-		pstat->requestqWaitCount = get_status_requestq_count();
-		pstat->pendingqWaitCount = get_status_pendingq_count();
-		pstat->totalOpenCount	 = get_status_totalopen_count();
-		pstat->cryptoDomain	 = get_status_domain_index();
-		get_status_status_mask(pstat->status);
-		get_status_qdepth_mask(pstat->qdepth);
-
-		if (copy_to_user((struct ica_z90_status __user *) arg, pstat,
-				 sizeof(struct ica_z90_status)) != 0)
-			ret = -EFAULT;
-		kfree(pstat);
-		break;
-
-		/* THIS IS DEPRECATED.	USE THE NEW STATUS CALLS */
-	case Z90STAT_PCIXCCCOUNT:
-		if (deprecated_msg_count2 < 20) {
-			PRINTK("deprecated ioctl (Z90STAT_PCIXCCCOUNT)!\n");
-			deprecated_msg_count2++;
-			if (deprecated_msg_count2 == 20)
-				PRINTK("No longer issuing messages about depre"
-				       "cated ioctl Z90STAT_PCIXCCCOUNT.\n");
-		}
-
-		tempstat = get_status_PCIXCCcount();
-		if (copy_to_user((int *)arg, &tempstat, sizeof(int)) != 0)
-			ret = -EFAULT;
-		break;
-
-	case Z90QUIESCE:
-		if (current->euid != 0) {
-			PRINTK("QUIESCE fails: euid %d\n",
-			       current->euid);
-			ret = -EACCES;
-		} else {
-			PRINTK("QUIESCE device from PID %d\n", PID());
-			quiesce_z90crypt = 1;
-		}
-		break;
-
-	default:
-		/* user passed an invalid IOCTL number */
-		PDEBUG("cmd 0x%08X contains invalid ioctl code\n", cmd);
-		ret = -ENOTTY;
-		break;
-	}
-
-	return ret;
-}
-
-static inline int
-sprintcl(unsigned char *outaddr, unsigned char *addr, unsigned int len)
-{
-	int hl, i;
-
-	hl = 0;
-	for (i = 0; i < len; i++)
-		hl += sprintf(outaddr+hl, "%01x", (unsigned int) addr[i]);
-	hl += sprintf(outaddr+hl, " ");
-
-	return hl;
-}
-
-static inline int
-sprintrw(unsigned char *outaddr, unsigned char *addr, unsigned int len)
-{
-	int hl, inl, c, cx;
-
-	hl = sprintf(outaddr, "	   ");
-	inl = 0;
-	for (c = 0; c < (len / 16); c++) {
-		hl += sprintcl(outaddr+hl, addr+inl, 16);
-		inl += 16;
-	}
-
-	cx = len%16;
-	if (cx) {
-		hl += sprintcl(outaddr+hl, addr+inl, cx);
-		inl += cx;
-	}
-
-	hl += sprintf(outaddr+hl, "\n");
-
-	return hl;
-}
-
-static inline int
-sprinthx(unsigned char *title, unsigned char *outaddr,
-	 unsigned char *addr, unsigned int len)
-{
-	int hl, inl, r, rx;
-
-	hl = sprintf(outaddr, "\n%s\n", title);
-	inl = 0;
-	for (r = 0; r < (len / 64); r++) {
-		hl += sprintrw(outaddr+hl, addr+inl, 64);
-		inl += 64;
-	}
-	rx = len % 64;
-	if (rx) {
-		hl += sprintrw(outaddr+hl, addr+inl, rx);
-		inl += rx;
-	}
-
-	hl += sprintf(outaddr+hl, "\n");
-
-	return hl;
-}
-
-static inline int
-sprinthx4(unsigned char *title, unsigned char *outaddr,
-	  unsigned int *array, unsigned int len)
-{
-	int hl, r;
-
-	hl = sprintf(outaddr, "\n%s\n", title);
-
-	for (r = 0; r < len; r++) {
-		if ((r % 8) == 0)
-			hl += sprintf(outaddr+hl, "    ");
-		hl += sprintf(outaddr+hl, "%08X ", array[r]);
-		if ((r % 8) == 7)
-			hl += sprintf(outaddr+hl, "\n");
-	}
-
-	hl += sprintf(outaddr+hl, "\n");
-
-	return hl;
-}
-
-static int
-z90crypt_status(char *resp_buff, char **start, off_t offset,
-		int count, int *eof, void *data)
-{
-	unsigned char *workarea;
-	int len;
-
-	/* resp_buff is a page. Use the right half for a work area */
-	workarea = resp_buff+2000;
-	len = 0;
-	len += sprintf(resp_buff+len, "\nz90crypt version: %d.%d.%d\n",
-		z90crypt_VERSION, z90crypt_RELEASE, z90crypt_VARIANT);
-	len += sprintf(resp_buff+len, "Cryptographic domain: %d\n",
-		get_status_domain_index());
-	len += sprintf(resp_buff+len, "Total device count: %d\n",
-		get_status_totalcount());
-	len += sprintf(resp_buff+len, "PCICA count: %d\n",
-		get_status_PCICAcount());
-	len += sprintf(resp_buff+len, "PCICC count: %d\n",
-		get_status_PCICCcount());
-	len += sprintf(resp_buff+len, "PCIXCC MCL2 count: %d\n",
-		get_status_PCIXCCMCL2count());
-	len += sprintf(resp_buff+len, "PCIXCC MCL3 count: %d\n",
-		get_status_PCIXCCMCL3count());
-	len += sprintf(resp_buff+len, "CEX2C count: %d\n",
-		get_status_CEX2Ccount());
-	len += sprintf(resp_buff+len, "CEX2A count: %d\n",
-		get_status_CEX2Acount());
-	len += sprintf(resp_buff+len, "requestq count: %d\n",
-		get_status_requestq_count());
-	len += sprintf(resp_buff+len, "pendingq count: %d\n",
-		get_status_pendingq_count());
-	len += sprintf(resp_buff+len, "Total open handles: %d\n\n",
-		get_status_totalopen_count());
-	len += sprinthx(
-		"Online devices: 1=PCICA 2=PCICC 3=PCIXCC(MCL2) "
-		"4=PCIXCC(MCL3) 5=CEX2C 6=CEX2A",
-		resp_buff+len,
-		get_status_status_mask(workarea),
-		Z90CRYPT_NUM_APS);
-	len += sprinthx("Waiting work element counts",
-		resp_buff+len,
-		get_status_qdepth_mask(workarea),
-		Z90CRYPT_NUM_APS);
-	len += sprinthx4(
-		"Per-device successfully completed request counts",
-		resp_buff+len,
-		get_status_perdevice_reqcnt((unsigned int *)workarea),
-		Z90CRYPT_NUM_APS);
-	*eof = 1;
-	memset(workarea, 0, Z90CRYPT_NUM_APS * sizeof(unsigned int));
-	return len;
-}
-
-static inline void
-disable_card(int card_index)
-{
-	struct device *devp;
-
-	devp = LONG2DEVPTR(card_index);
-	if (!devp || devp->user_disabled)
-		return;
-	devp->user_disabled = 1;
-	z90crypt.hdware_info->hdware_mask.user_disabled_count++;
-	if (devp->dev_type == -1)
-		return;
-	z90crypt.hdware_info->type_mask[devp->dev_type].user_disabled_count++;
-}
-
-static inline void
-enable_card(int card_index)
-{
-	struct device *devp;
-
-	devp = LONG2DEVPTR(card_index);
-	if (!devp || !devp->user_disabled)
-		return;
-	devp->user_disabled = 0;
-	z90crypt.hdware_info->hdware_mask.user_disabled_count--;
-	if (devp->dev_type == -1)
-		return;
-	z90crypt.hdware_info->type_mask[devp->dev_type].user_disabled_count--;
-}
-
-static int
-z90crypt_status_write(struct file *file, const char __user *buffer,
-		      unsigned long count, void *data)
-{
-	int j, eol;
-	unsigned char *lbuf, *ptr;
-	unsigned int local_count;
-
-#define LBUFSIZE 1200
-	lbuf = kmalloc(LBUFSIZE, GFP_KERNEL);
-	if (!lbuf) {
-		PRINTK("kmalloc failed!\n");
-		return 0;
-	}
-
-	if (count <= 0)
-		return 0;
-
-	local_count = UMIN((unsigned int)count, LBUFSIZE-1);
-
-	if (copy_from_user(lbuf, buffer, local_count) != 0) {
-		kfree(lbuf);
-		return -EFAULT;
-	}
-
-	lbuf[local_count] = '\0';
-
-	ptr = strstr(lbuf, "Online devices");
-	if (ptr == 0) {
-		PRINTK("Unable to parse data (missing \"Online devices\")\n");
-		kfree(lbuf);
-		return count;
-	}
-
-	ptr = strstr(ptr, "\n");
-	if (ptr == 0) {
-		PRINTK("Unable to parse data (missing newline after \"Online devices\")\n");
-		kfree(lbuf);
-		return count;
-	}
-	ptr++;
-
-	if (strstr(ptr, "Waiting work element counts") == NULL) {
-		PRINTK("Unable to parse data (missing \"Waiting work element counts\")\n");
-		kfree(lbuf);
-		return count;
-	}
-
-	j = 0;
-	eol = 0;
-	while ((j < 64) && (*ptr != '\0')) {
-		switch (*ptr) {
-		case '\t':
-		case ' ':
-			break;
-		case '\n':
-		default:
-			eol = 1;
-			break;
-		case '0':	// no device
-		case '1':	// PCICA
-		case '2':	// PCICC
-		case '3':	// PCIXCC_MCL2
-		case '4':	// PCIXCC_MCL3
-		case '5':	// CEX2C
-		case '6':       // CEX2A
-			j++;
-			break;
-		case 'd':
-		case 'D':
-			disable_card(j);
-			j++;
-			break;
-		case 'e':
-		case 'E':
-			enable_card(j);
-			j++;
-			break;
-		}
-		if (eol)
-			break;
-		ptr++;
-	}
-
-	kfree(lbuf);
-	return count;
-}
-
-/**
- * Functions that run under a timer, with no process id
- *
- * The task functions:
- *     z90crypt_reader_task
- *	 helper_send_work
- *	 helper_handle_work_element
- *	 helper_receive_rc
- *     z90crypt_config_task
- *     z90crypt_cleanup_task
- *
- * Helper functions:
- *     z90crypt_schedule_reader_timer
- *     z90crypt_schedule_reader_task
- *     z90crypt_schedule_config_task
- *     z90crypt_schedule_cleanup_task
- */
-static inline int
-receive_from_crypto_device(int index, unsigned char *psmid, int *buff_len_p,
-			   unsigned char *buff, unsigned char __user **dest_p_p)
-{
-	int dv, rv;
-	struct device *dev_ptr;
-	struct caller *caller_p;
-	struct ica_rsa_modexpo *icaMsg_p;
-	struct list_head *ptr, *tptr;
-
-	memcpy(psmid, NULL_psmid, sizeof(NULL_psmid));
-
-	if (z90crypt.terminating)
-		return REC_FATAL_ERROR;
-
-	caller_p = 0;
-	dev_ptr = z90crypt.device_p[index];
-	rv = 0;
-	do {
-		if (!dev_ptr || dev_ptr->disabled) {
-			rv = REC_NO_WORK; // a disabled device can't return work
-			break;
-		}
-		if (dev_ptr->dev_self_x != index) {
-			PRINTKC("Corrupt dev ptr\n");
-			z90crypt.terminating = 1;
-			rv = REC_FATAL_ERROR;
-			break;
-		}
-		if (!dev_ptr->dev_resp_l || !dev_ptr->dev_resp_p) {
-			dv = DEV_REC_EXCEPTION;
-			PRINTK("dev_resp_l = %d, dev_resp_p = %p\n",
-			       dev_ptr->dev_resp_l, dev_ptr->dev_resp_p);
-		} else {
-			PDEBUG("Dequeue called for device %d\n", index);
-			dv = receive_from_AP(index, z90crypt.cdx,
-					     dev_ptr->dev_resp_l,
-					     dev_ptr->dev_resp_p, psmid);
-		}
-		switch (dv) {
-		case DEV_REC_EXCEPTION:
-			rv = REC_FATAL_ERROR;
-			z90crypt.terminating = 1;
-			PRINTKC("Exception in receive from device %d\n",
-				index);
-			break;
-		case DEV_ONLINE:
-			rv = 0;
-			break;
-		case DEV_EMPTY:
-			rv = REC_EMPTY;
-			break;
-		case DEV_NO_WORK:
-			rv = REC_NO_WORK;
-			break;
-		case DEV_BAD_MESSAGE:
-		case DEV_GONE:
-		case REC_HARDWAR_ERR:
-		default:
-			rv = REC_NO_RESPONSE;
-			break;
-		}
-		if (rv)
-			break;
-		if (dev_ptr->dev_caller_count <= 0) {
-			rv = REC_USER_GONE;
-			break;
-	        }
-
-		list_for_each_safe(ptr, tptr, &dev_ptr->dev_caller_list) {
-			caller_p = list_entry(ptr, struct caller, caller_liste);
-			if (!memcmp(caller_p->caller_id, psmid,
-				    sizeof(caller_p->caller_id))) {
-				if (!list_empty(&caller_p->caller_liste)) {
-					list_del_init(ptr);
-					dev_ptr->dev_caller_count--;
-					break;
-				}
-			}
-			caller_p = 0;
-		}
-		if (!caller_p) {
-			PRINTKW("Unable to locate PSMID %02X%02X%02X%02X%02X"
-				"%02X%02X%02X in device list\n",
-				psmid[0], psmid[1], psmid[2], psmid[3],
-				psmid[4], psmid[5], psmid[6], psmid[7]);
-			rv = REC_USER_GONE;
-			break;
-		}
-
-		PDEBUG("caller_p after successful receive: %p\n", caller_p);
-		rv = convert_response(dev_ptr->dev_resp_p,
-				      caller_p->caller_buf_p, buff_len_p, buff);
-		switch (rv) {
-		case REC_USE_PCICA:
-			break;
-		case REC_OPERAND_INV:
-		case REC_OPERAND_SIZE:
-		case REC_EVEN_MOD:
-		case REC_INVALID_PAD:
-			PDEBUG("device %d: 'user error' %d\n", index, rv);
-			break;
-		case WRONG_DEVICE_TYPE:
-		case REC_HARDWAR_ERR:
-		case REC_BAD_MESSAGE:
-			PRINTKW("device %d: hardware error %d\n", index, rv);
-			rv = REC_NO_RESPONSE;
-			break;
-		default:
-			PDEBUG("device %d: rv = %d\n", index, rv);
-			break;
-		}
-	} while (0);
-
-	switch (rv) {
-	case 0:
-		PDEBUG("Successful receive from device %d\n", index);
-		icaMsg_p = (struct ica_rsa_modexpo *)caller_p->caller_buf_p;
-		*dest_p_p = icaMsg_p->outputdata;
-		if (*buff_len_p == 0)
-			PRINTK("Zero *buff_len_p\n");
-		break;
-	case REC_NO_RESPONSE:
-		PRINTKW("Removing device %d from availability\n", index);
-		remove_device(dev_ptr);
-		break;
-	}
-
-	if (caller_p)
-		unbuild_caller(dev_ptr, caller_p);
-
-	return rv;
-}
-
-static inline void
-helper_send_work(int index)
-{
-	struct work_element *rq_p;
-	int rv;
-
-	if (list_empty(&request_list))
-		return;
-	requestq_count--;
-	rq_p = list_entry(request_list.next, struct work_element, liste);
-	list_del_init(&rq_p->liste);
-	rq_p->audit[1] |= FP_REMREQUEST;
-	if (rq_p->devtype == SHRT2DEVPTR(index)->dev_type) {
-		rq_p->devindex = SHRT2LONG(index);
-		rv = send_to_crypto_device(rq_p);
-		if (rv == 0) {
-			rq_p->requestsent = jiffies;
-			rq_p->audit[0] |= FP_SENT;
-			list_add_tail(&rq_p->liste, &pending_list);
-			++pendingq_count;
-			rq_p->audit[0] |= FP_PENDING;
-		} else {
-			switch (rv) {
-			case REC_OPERAND_INV:
-			case REC_OPERAND_SIZE:
-			case REC_EVEN_MOD:
-			case REC_INVALID_PAD:
-				rq_p->retcode = -EINVAL;
-				break;
-			case SEN_NOT_AVAIL:
-			case SEN_RETRY:
-			case REC_NO_RESPONSE:
-			default:
-				if (z90crypt.mask.st_count > 1)
-					rq_p->retcode =
-						-ERESTARTSYS;
-				else
-					rq_p->retcode = -ENODEV;
-				break;
-			}
-			rq_p->status[0] |= STAT_FAILED;
-			rq_p->audit[1] |= FP_AWAKENING;
-			atomic_set(&rq_p->alarmrung, 1);
-			wake_up(&rq_p->waitq);
-		}
-	} else {
-		if (z90crypt.mask.st_count > 1)
-			rq_p->retcode = -ERESTARTSYS;
-		else
-			rq_p->retcode = -ENODEV;
-		rq_p->status[0] |= STAT_FAILED;
-		rq_p->audit[1] |= FP_AWAKENING;
-		atomic_set(&rq_p->alarmrung, 1);
-		wake_up(&rq_p->waitq);
-	}
-}
-
-static inline void
-helper_handle_work_element(int index, unsigned char psmid[8], int rc,
-			   int buff_len, unsigned char *buff,
-			   unsigned char __user *resp_addr)
-{
-	struct work_element *pq_p;
-	struct list_head *lptr, *tptr;
-
-	pq_p = 0;
-	list_for_each_safe(lptr, tptr, &pending_list) {
-		pq_p = list_entry(lptr, struct work_element, liste);
-		if (!memcmp(pq_p->caller_id, psmid, sizeof(pq_p->caller_id))) {
-			list_del_init(lptr);
-			pendingq_count--;
-			pq_p->audit[1] |= FP_NOTPENDING;
-			break;
-		}
-		pq_p = 0;
-	}
-
-	if (!pq_p) {
-		PRINTK("device %d has work but no caller exists on pending Q\n",
-		       SHRT2LONG(index));
-		return;
-	}
-
-	switch (rc) {
-		case 0:
-			pq_p->resp_buff_size = buff_len;
-			pq_p->audit[1] |= FP_RESPSIZESET;
-			if (buff_len) {
-				pq_p->resp_addr = resp_addr;
-				pq_p->audit[1] |= FP_RESPADDRCOPIED;
-				memcpy(pq_p->resp_buff, buff, buff_len);
-				pq_p->audit[1] |= FP_RESPBUFFCOPIED;
-			}
-			break;
-		case REC_OPERAND_INV:
-		case REC_OPERAND_SIZE:
-		case REC_EVEN_MOD:
-		case REC_INVALID_PAD:
-			PDEBUG("-EINVAL after application error %d\n", rc);
-			pq_p->retcode = -EINVAL;
-			pq_p->status[0] |= STAT_FAILED;
-			break;
-		case REC_USE_PCICA:
-			pq_p->retcode = -ERESTARTSYS;
-			pq_p->status[0] |= STAT_FAILED;
-			break;
-		case REC_NO_RESPONSE:
-		default:
-			if (z90crypt.mask.st_count > 1)
-				pq_p->retcode = -ERESTARTSYS;
-			else
-				pq_p->retcode = -ENODEV;
-			pq_p->status[0] |= STAT_FAILED;
-			break;
-	}
-	if ((pq_p->status[0] != STAT_FAILED) || (pq_p->retcode != -ERELEASED)) {
-		pq_p->audit[1] |= FP_AWAKENING;
-		atomic_set(&pq_p->alarmrung, 1);
-		wake_up(&pq_p->waitq);
-	}
-}
-
-/**
- * return TRUE if the work element should be removed from the queue
- */
-static inline int
-helper_receive_rc(int index, int *rc_p)
-{
-	switch (*rc_p) {
-	case 0:
-	case REC_OPERAND_INV:
-	case REC_OPERAND_SIZE:
-	case REC_EVEN_MOD:
-	case REC_INVALID_PAD:
-	case REC_USE_PCICA:
-		break;
-
-	case REC_BUSY:
-	case REC_NO_WORK:
-	case REC_EMPTY:
-	case REC_RETRY_DEV:
-	case REC_FATAL_ERROR:
-		return 0;
-
-	case REC_NO_RESPONSE:
-		break;
-
-	default:
-		PRINTK("rc %d, device %d converted to REC_NO_RESPONSE\n",
-		       *rc_p, SHRT2LONG(index));
-		*rc_p = REC_NO_RESPONSE;
-		break;
-	}
-	return 1;
-}
-
-static inline void
-z90crypt_schedule_reader_timer(void)
-{
-	if (timer_pending(&reader_timer))
-		return;
-	if (mod_timer(&reader_timer, jiffies+(READERTIME*HZ/1000)) != 0)
-		PRINTK("Timer pending while modifying reader timer\n");
-}
-
-static void
-z90crypt_reader_task(unsigned long ptr)
-{
-	int workavail, index, rc, buff_len;
-	unsigned char	psmid[8];
-	unsigned char __user *resp_addr;
-	static unsigned char buff[1024];
-
-	/**
-	 * we use workavail = 2 to ensure 2 passes with nothing dequeued before
-	 * exiting the loop. If (pendingq_count+requestq_count) == 0 after the
-	 * loop, there is no work remaining on the queues.
-	 */
-	resp_addr = 0;
-	workavail = 2;
-	buff_len = 0;
-	while (workavail) {
-		workavail--;
-		rc = 0;
-		spin_lock_irq(&queuespinlock);
-		memset(buff, 0x00, sizeof(buff));
-
-		/* Dequeue once from each device in round robin. */
-		for (index = 0; index < z90crypt.mask.st_count; index++) {
-			PDEBUG("About to receive.\n");
-			rc = receive_from_crypto_device(SHRT2LONG(index),
-							psmid,
-							&buff_len,
-							buff,
-							&resp_addr);
-			PDEBUG("Dequeued: rc = %d.\n", rc);
-
-			if (helper_receive_rc(index, &rc)) {
-				if (rc != REC_NO_RESPONSE) {
-					helper_send_work(index);
-					workavail = 2;
-				}
-
-				helper_handle_work_element(index, psmid, rc,
-							   buff_len, buff,
-							   resp_addr);
-			}
-
-			if (rc == REC_FATAL_ERROR)
-				PRINTKW("REC_FATAL_ERROR from device %d!\n",
-					SHRT2LONG(index));
-		}
-		spin_unlock_irq(&queuespinlock);
-	}
-
-	if (pendingq_count + requestq_count)
-		z90crypt_schedule_reader_timer();
-}
-
-static inline void
-z90crypt_schedule_config_task(unsigned int expiration)
-{
-	if (timer_pending(&config_timer))
-		return;
-	if (mod_timer(&config_timer, jiffies+(expiration*HZ)) != 0)
-		PRINTK("Timer pending while modifying config timer\n");
-}
-
-static void
-z90crypt_config_task(unsigned long ptr)
-{
-	int rc;
-
-	PDEBUG("jiffies %ld\n", jiffies);
-
-	if ((rc = refresh_z90crypt(&z90crypt.cdx)))
-		PRINTK("Error %d detected in refresh_z90crypt.\n", rc);
-	/* If return was fatal, don't bother reconfiguring */
-	if ((rc != TSQ_FATAL_ERROR) && (rc != RSQ_FATAL_ERROR))
-		z90crypt_schedule_config_task(CONFIGTIME);
-}
-
-static inline void
-z90crypt_schedule_cleanup_task(void)
-{
-	if (timer_pending(&cleanup_timer))
-		return;
-	if (mod_timer(&cleanup_timer, jiffies+(CLEANUPTIME*HZ)) != 0)
-		PRINTK("Timer pending while modifying cleanup timer\n");
-}
-
-static inline void
-helper_drain_queues(void)
-{
-	struct work_element *pq_p;
-	struct list_head *lptr, *tptr;
-
-	list_for_each_safe(lptr, tptr, &pending_list) {
-		pq_p = list_entry(lptr, struct work_element, liste);
-		pq_p->retcode = -ENODEV;
-		pq_p->status[0] |= STAT_FAILED;
-		unbuild_caller(LONG2DEVPTR(pq_p->devindex),
-			       (struct caller *)pq_p->requestptr);
-		list_del_init(lptr);
-		pendingq_count--;
-		pq_p->audit[1] |= FP_NOTPENDING;
-		pq_p->audit[1] |= FP_AWAKENING;
-		atomic_set(&pq_p->alarmrung, 1);
-		wake_up(&pq_p->waitq);
-	}
-
-	list_for_each_safe(lptr, tptr, &request_list) {
-		pq_p = list_entry(lptr, struct work_element, liste);
-		pq_p->retcode = -ENODEV;
-		pq_p->status[0] |= STAT_FAILED;
-		list_del_init(lptr);
-		requestq_count--;
-		pq_p->audit[1] |= FP_REMREQUEST;
-		pq_p->audit[1] |= FP_AWAKENING;
-		atomic_set(&pq_p->alarmrung, 1);
-		wake_up(&pq_p->waitq);
-	}
-}
-
-static inline void
-helper_timeout_requests(void)
-{
-	struct work_element *pq_p;
-	struct list_head *lptr, *tptr;
-	long timelimit;
-
-	timelimit = jiffies - (CLEANUPTIME * HZ);
-	/* The list is in strict chronological order */
-	list_for_each_safe(lptr, tptr, &pending_list) {
-		pq_p = list_entry(lptr, struct work_element, liste);
-		if (pq_p->requestsent >= timelimit)
-			break;
-		PRINTKW("Purging(PQ) PSMID %02X%02X%02X%02X%02X%02X%02X%02X\n",
-		       ((struct caller *)pq_p->requestptr)->caller_id[0],
-		       ((struct caller *)pq_p->requestptr)->caller_id[1],
-		       ((struct caller *)pq_p->requestptr)->caller_id[2],
-		       ((struct caller *)pq_p->requestptr)->caller_id[3],
-		       ((struct caller *)pq_p->requestptr)->caller_id[4],
-		       ((struct caller *)pq_p->requestptr)->caller_id[5],
-		       ((struct caller *)pq_p->requestptr)->caller_id[6],
-		       ((struct caller *)pq_p->requestptr)->caller_id[7]);
-		pq_p->retcode = -ETIMEOUT;
-		pq_p->status[0] |= STAT_FAILED;
-		/* get this off any caller queue it may be on */
-		unbuild_caller(LONG2DEVPTR(pq_p->devindex),
-			       (struct caller *) pq_p->requestptr);
-		list_del_init(lptr);
-		pendingq_count--;
-		pq_p->audit[1] |= FP_TIMEDOUT;
-		pq_p->audit[1] |= FP_NOTPENDING;
-		pq_p->audit[1] |= FP_AWAKENING;
-		atomic_set(&pq_p->alarmrung, 1);
-		wake_up(&pq_p->waitq);
-	}
-
-	/**
-	 * If pending count is zero, items left on the request queue may
-	 * never be processed.
-	 */
-	if (pendingq_count <= 0) {
-		list_for_each_safe(lptr, tptr, &request_list) {
-			pq_p = list_entry(lptr, struct work_element, liste);
-			if (pq_p->requestsent >= timelimit)
-				break;
-		PRINTKW("Purging(RQ) PSMID %02X%02X%02X%02X%02X%02X%02X%02X\n",
-		       ((struct caller *)pq_p->requestptr)->caller_id[0],
-		       ((struct caller *)pq_p->requestptr)->caller_id[1],
-		       ((struct caller *)pq_p->requestptr)->caller_id[2],
-		       ((struct caller *)pq_p->requestptr)->caller_id[3],
-		       ((struct caller *)pq_p->requestptr)->caller_id[4],
-		       ((struct caller *)pq_p->requestptr)->caller_id[5],
-		       ((struct caller *)pq_p->requestptr)->caller_id[6],
-		       ((struct caller *)pq_p->requestptr)->caller_id[7]);
-			pq_p->retcode = -ETIMEOUT;
-			pq_p->status[0] |= STAT_FAILED;
-			list_del_init(lptr);
-			requestq_count--;
-			pq_p->audit[1] |= FP_TIMEDOUT;
-			pq_p->audit[1] |= FP_REMREQUEST;
-			pq_p->audit[1] |= FP_AWAKENING;
-			atomic_set(&pq_p->alarmrung, 1);
-			wake_up(&pq_p->waitq);
-		}
-	}
-}
-
-static void
-z90crypt_cleanup_task(unsigned long ptr)
-{
-	PDEBUG("jiffies %ld\n", jiffies);
-	spin_lock_irq(&queuespinlock);
-	if (z90crypt.mask.st_count <= 0) // no devices!
-		helper_drain_queues();
-	else
-		helper_timeout_requests();
-	spin_unlock_irq(&queuespinlock);
-	z90crypt_schedule_cleanup_task();
-}
-
-static void
-z90crypt_schedule_reader_task(unsigned long ptr)
-{
-	tasklet_schedule(&reader_tasklet);
-}
-
-/**
- * Lowlevel Functions:
- *
- *   create_z90crypt:  creates and initializes basic data structures
- *   refresh_z90crypt:	re-initializes basic data structures
- *   find_crypto_devices: returns a count and mask of hardware status
- *   create_crypto_device:  builds the descriptor for a device
- *   destroy_crypto_device:  unallocates the descriptor for a device
- *   destroy_z90crypt:	drains all work, unallocates structs
- */
-
-/**
- * build the z90crypt root structure using the given domain index
- */
-static int
-create_z90crypt(int *cdx_p)
-{
-	struct hdware_block *hdware_blk_p;
-
-	memset(&z90crypt, 0x00, sizeof(struct z90crypt));
-	z90crypt.domain_established = 0;
-	z90crypt.len = sizeof(struct z90crypt);
-	z90crypt.max_count = Z90CRYPT_NUM_DEVS;
-	z90crypt.cdx = *cdx_p;
-
-	hdware_blk_p = kzalloc(sizeof(struct hdware_block), GFP_ATOMIC);
-	if (!hdware_blk_p) {
-		PDEBUG("kmalloc for hardware block failed\n");
-		return ENOMEM;
-	}
-	z90crypt.hdware_info = hdware_blk_p;
-
-	return 0;
-}
-
-static inline int
-helper_scan_devices(int cdx_array[16], int *cdx_p, int *correct_cdx_found)
-{
-	enum hdstat hd_stat;
-	int q_depth, dev_type;
-	int indx, chkdom, numdomains;
-
-	q_depth = dev_type = numdomains = 0;
-	for (chkdom = 0; chkdom <= 15; cdx_array[chkdom++] = -1);
-	for (indx = 0; indx < z90crypt.max_count; indx++) {
-		hd_stat = HD_NOT_THERE;
-		numdomains = 0;
-		for (chkdom = 0; chkdom <= 15; chkdom++) {
-			hd_stat = query_online(indx, chkdom, MAX_RESET,
-					       &q_depth, &dev_type);
-			if (hd_stat == HD_TSQ_EXCEPTION) {
-				z90crypt.terminating = 1;
-				PRINTKC("exception taken!\n");
-				break;
-			}
-			if (hd_stat == HD_ONLINE) {
-				cdx_array[numdomains++] = chkdom;
-				if (*cdx_p == chkdom) {
-					*correct_cdx_found  = 1;
-					break;
-				}
-			}
-		}
-		if ((*correct_cdx_found == 1) || (numdomains != 0))
-			break;
-		if (z90crypt.terminating)
-			break;
-	}
-	return numdomains;
-}
-
-static inline int
-probe_crypto_domain(int *cdx_p)
-{
-	int cdx_array[16];
-	char cdx_array_text[53], temp[5];
-	int correct_cdx_found, numdomains;
-
-	correct_cdx_found = 0;
-	numdomains = helper_scan_devices(cdx_array, cdx_p, &correct_cdx_found);
-
-	if (z90crypt.terminating)
-		return TSQ_FATAL_ERROR;
-
-	if (correct_cdx_found)
-		return 0;
-
-	if (numdomains == 0) {
-		PRINTKW("Unable to find crypto domain: No devices found\n");
-		return Z90C_NO_DEVICES;
-	}
-
-	if (numdomains == 1) {
-		if (*cdx_p == -1) {
-			*cdx_p = cdx_array[0];
-			return 0;
-		}
-		PRINTKW("incorrect domain: specified = %d, found = %d\n",
-		       *cdx_p, cdx_array[0]);
-		return Z90C_INCORRECT_DOMAIN;
-	}
-
-	numdomains--;
-	sprintf(cdx_array_text, "%d", cdx_array[numdomains]);
-	while (numdomains) {
-		numdomains--;
-		sprintf(temp, ", %d", cdx_array[numdomains]);
-		strcat(cdx_array_text, temp);
-	}
-
-	PRINTKW("ambiguous domain detected: specified = %d, found array = %s\n",
-		*cdx_p, cdx_array_text);
-	return Z90C_AMBIGUOUS_DOMAIN;
-}
-
-static int
-refresh_z90crypt(int *cdx_p)
-{
-	int i, j, indx, rv;
-	static struct status local_mask;
-	struct device *devPtr;
-	unsigned char oldStat, newStat;
-	int return_unchanged;
-
-	if (z90crypt.len != sizeof(z90crypt))
-		return ENOTINIT;
-	if (z90crypt.terminating)
-		return TSQ_FATAL_ERROR;
-	rv = 0;
-	if (!z90crypt.hdware_info->hdware_mask.st_count &&
-	    !z90crypt.domain_established) {
-		rv = probe_crypto_domain(cdx_p);
-		if (z90crypt.terminating)
-			return TSQ_FATAL_ERROR;
-		if (rv == Z90C_NO_DEVICES)
-			return 0; // try later
-		if (rv)
-			return rv;
-		z90crypt.cdx = *cdx_p;
-		z90crypt.domain_established = 1;
-	}
-	rv = find_crypto_devices(&local_mask);
-	if (rv) {
-		PRINTK("find crypto devices returned %d\n", rv);
-		return rv;
-	}
-	if (!memcmp(&local_mask, &z90crypt.hdware_info->hdware_mask,
-		    sizeof(struct status))) {
-		return_unchanged = 1;
-		for (i = 0; i < Z90CRYPT_NUM_TYPES; i++) {
-			/**
-			 * Check for disabled cards.  If any device is marked
-			 * disabled, destroy it.
-			 */
-			for (j = 0;
-			     j < z90crypt.hdware_info->type_mask[i].st_count;
-			     j++) {
-				indx = z90crypt.hdware_info->type_x_addr[i].
-								device_index[j];
-				devPtr = z90crypt.device_p[indx];
-				if (devPtr && devPtr->disabled) {
-					local_mask.st_mask[indx] = HD_NOT_THERE;
-					return_unchanged = 0;
-				}
-			}
-		}
-		if (return_unchanged == 1)
-			return 0;
-	}
-
-	spin_lock_irq(&queuespinlock);
-	for (i = 0; i < z90crypt.max_count; i++) {
-		oldStat = z90crypt.hdware_info->hdware_mask.st_mask[i];
-		newStat = local_mask.st_mask[i];
-		if ((oldStat == HD_ONLINE) && (newStat != HD_ONLINE))
-			destroy_crypto_device(i);
-		else if ((oldStat != HD_ONLINE) && (newStat == HD_ONLINE)) {
-			rv = create_crypto_device(i);
-			if (rv >= REC_FATAL_ERROR)
-				return rv;
-			if (rv != 0) {
-				local_mask.st_mask[i] = HD_NOT_THERE;
-				local_mask.st_count--;
-			}
-		}
-	}
-	memcpy(z90crypt.hdware_info->hdware_mask.st_mask, local_mask.st_mask,
-	       sizeof(local_mask.st_mask));
-	z90crypt.hdware_info->hdware_mask.st_count = local_mask.st_count;
-	z90crypt.hdware_info->hdware_mask.disabled_count =
-						      local_mask.disabled_count;
-	refresh_index_array(&z90crypt.mask, &z90crypt.overall_device_x);
-	for (i = 0; i < Z90CRYPT_NUM_TYPES; i++)
-		refresh_index_array(&(z90crypt.hdware_info->type_mask[i]),
-				    &(z90crypt.hdware_info->type_x_addr[i]));
-	spin_unlock_irq(&queuespinlock);
-
-	return rv;
-}
-
-static int
-find_crypto_devices(struct status *deviceMask)
-{
-	int i, q_depth, dev_type;
-	enum hdstat hd_stat;
-
-	deviceMask->st_count = 0;
-	deviceMask->disabled_count = 0;
-	deviceMask->user_disabled_count = 0;
-
-	for (i = 0; i < z90crypt.max_count; i++) {
-		hd_stat = query_online(i, z90crypt.cdx, MAX_RESET, &q_depth,
-				       &dev_type);
-		if (hd_stat == HD_TSQ_EXCEPTION) {
-			z90crypt.terminating = 1;
-			PRINTKC("Exception during probe for crypto devices\n");
-			return TSQ_FATAL_ERROR;
-		}
-		deviceMask->st_mask[i] = hd_stat;
-		if (hd_stat == HD_ONLINE) {
-			PDEBUG("Got an online crypto!: %d\n", i);
-			PDEBUG("Got a queue depth of %d\n", q_depth);
-			PDEBUG("Got a device type of %d\n", dev_type);
-			if (q_depth <= 0)
-				return TSQ_FATAL_ERROR;
-			deviceMask->st_count++;
-			z90crypt.q_depth_array[i] = q_depth;
-			z90crypt.dev_type_array[i] = dev_type;
-		}
-	}
-
-	return 0;
-}
-
-static int
-refresh_index_array(struct status *status_str, struct device_x *index_array)
-{
-	int i, count;
-	enum devstat stat;
-
-	i = -1;
-	count = 0;
-	do {
-		stat = status_str->st_mask[++i];
-		if (stat == DEV_ONLINE)
-			index_array->device_index[count++] = i;
-	} while ((i < Z90CRYPT_NUM_DEVS) && (count < status_str->st_count));
-
-	return count;
-}
-
-static int
-create_crypto_device(int index)
-{
-	int rv, devstat, total_size;
-	struct device *dev_ptr;
-	struct status *type_str_p;
-	int deviceType;
-
-	dev_ptr = z90crypt.device_p[index];
-	if (!dev_ptr) {
-		total_size = sizeof(struct device) +
-			     z90crypt.q_depth_array[index] * sizeof(int);
-
-		dev_ptr = kzalloc(total_size, GFP_ATOMIC);
-		if (!dev_ptr) {
-			PRINTK("kmalloc device %d failed\n", index);
-			return ENOMEM;
-		}
-		dev_ptr->dev_resp_p = kmalloc(MAX_RESPONSE_SIZE, GFP_ATOMIC);
-		if (!dev_ptr->dev_resp_p) {
-			kfree(dev_ptr);
-			PRINTK("kmalloc device %d rec buffer failed\n", index);
-			return ENOMEM;
-		}
-		dev_ptr->dev_resp_l = MAX_RESPONSE_SIZE;
-		INIT_LIST_HEAD(&(dev_ptr->dev_caller_list));
-	}
-
-	devstat = reset_device(index, z90crypt.cdx, MAX_RESET);
-	if (devstat == DEV_RSQ_EXCEPTION) {
-		PRINTK("exception during reset device %d\n", index);
-		kfree(dev_ptr->dev_resp_p);
-		kfree(dev_ptr);
-		return RSQ_FATAL_ERROR;
-	}
-	if (devstat == DEV_ONLINE) {
-		dev_ptr->dev_self_x = index;
-		dev_ptr->dev_type = z90crypt.dev_type_array[index];
-		if (dev_ptr->dev_type == NILDEV) {
-			rv = probe_device_type(dev_ptr);
-			if (rv) {
-				PRINTK("rv = %d from probe_device_type %d\n",
-				       rv, index);
-				kfree(dev_ptr->dev_resp_p);
-				kfree(dev_ptr);
-				return rv;
-			}
-		}
-		if (dev_ptr->dev_type == PCIXCC_UNK) {
-			rv = probe_PCIXCC_type(dev_ptr);
-			if (rv) {
-				PRINTK("rv = %d from probe_PCIXCC_type %d\n",
-				       rv, index);
-				kfree(dev_ptr->dev_resp_p);
-				kfree(dev_ptr);
-				return rv;
-			}
-		}
-		deviceType = dev_ptr->dev_type;
-		z90crypt.dev_type_array[index] = deviceType;
-		if (deviceType == PCICA)
-			z90crypt.hdware_info->device_type_array[index] = 1;
-		else if (deviceType == PCICC)
-			z90crypt.hdware_info->device_type_array[index] = 2;
-		else if (deviceType == PCIXCC_MCL2)
-			z90crypt.hdware_info->device_type_array[index] = 3;
-		else if (deviceType == PCIXCC_MCL3)
-			z90crypt.hdware_info->device_type_array[index] = 4;
-		else if (deviceType == CEX2C)
-			z90crypt.hdware_info->device_type_array[index] = 5;
-		else if (deviceType == CEX2A)
-			z90crypt.hdware_info->device_type_array[index] = 6;
-		else // No idea how this would happen.
-			z90crypt.hdware_info->device_type_array[index] = -1;
-	}
-
-	/**
-	 * 'q_depth' returned by the hardware is one less than
-	 * the actual depth
-	 */
-	dev_ptr->dev_q_depth = z90crypt.q_depth_array[index];
-	dev_ptr->dev_type = z90crypt.dev_type_array[index];
-	dev_ptr->dev_stat = devstat;
-	dev_ptr->disabled = 0;
-	z90crypt.device_p[index] = dev_ptr;
-
-	if (devstat == DEV_ONLINE) {
-		if (z90crypt.mask.st_mask[index] != DEV_ONLINE) {
-			z90crypt.mask.st_mask[index] = DEV_ONLINE;
-			z90crypt.mask.st_count++;
-		}
-		deviceType = dev_ptr->dev_type;
-		type_str_p = &z90crypt.hdware_info->type_mask[deviceType];
-		if (type_str_p->st_mask[index] != DEV_ONLINE) {
-			type_str_p->st_mask[index] = DEV_ONLINE;
-			type_str_p->st_count++;
-		}
-	}
-
-	return 0;
-}
-
-static int
-destroy_crypto_device(int index)
-{
-	struct device *dev_ptr;
-	int t, disabledFlag;
-
-	dev_ptr = z90crypt.device_p[index];
-
-	/* remember device type; get rid of device struct */
-	if (dev_ptr) {
-		disabledFlag = dev_ptr->disabled;
-		t = dev_ptr->dev_type;
-		kfree(dev_ptr->dev_resp_p);
-		kfree(dev_ptr);
-	} else {
-		disabledFlag = 0;
-		t = -1;
-	}
-	z90crypt.device_p[index] = 0;
-
-	/* if the type is valid, remove the device from the type_mask */
-	if ((t != -1) && z90crypt.hdware_info->type_mask[t].st_mask[index]) {
-		  z90crypt.hdware_info->type_mask[t].st_mask[index] = 0x00;
-		  z90crypt.hdware_info->type_mask[t].st_count--;
-		  if (disabledFlag == 1)
-			z90crypt.hdware_info->type_mask[t].disabled_count--;
-	}
-	if (z90crypt.mask.st_mask[index] != DEV_GONE) {
-		z90crypt.mask.st_mask[index] = DEV_GONE;
-		z90crypt.mask.st_count--;
-	}
-	z90crypt.hdware_info->device_type_array[index] = 0;
-
-	return 0;
-}
-
-static void
-destroy_z90crypt(void)
-{
-	int i;
-
-	for (i = 0; i < z90crypt.max_count; i++)
-		if (z90crypt.device_p[i])
-			destroy_crypto_device(i);
-	kfree(z90crypt.hdware_info);
-	memset((void *)&z90crypt, 0, sizeof(z90crypt));
-}
-
-static unsigned char static_testmsg[384] = {
-0x00,0x00,0x00,0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x00,0x06,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x58,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x43,0x43,
-0x41,0x2d,0x41,0x50,0x50,0x4c,0x20,0x20,0x20,0x01,0x01,0x01,0x00,0x00,0x00,0x00,
-0x50,0x4b,0x00,0x00,0x00,0x00,0x01,0x1c,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x05,0xb8,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x70,0x00,0x41,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x54,0x32,
-0x01,0x00,0xa0,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0xb8,0x05,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x0a,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x08,0x00,0x49,0x43,0x53,0x46,
-0x20,0x20,0x20,0x20,0x50,0x4b,0x0a,0x00,0x50,0x4b,0x43,0x53,0x2d,0x31,0x2e,0x32,
-0x37,0x00,0x11,0x22,0x33,0x44,0x55,0x66,0x77,0x88,0x99,0x00,0x11,0x22,0x33,0x44,
-0x55,0x66,0x77,0x88,0x99,0x00,0x11,0x22,0x33,0x44,0x55,0x66,0x77,0x88,0x99,0x00,
-0x11,0x22,0x33,0x44,0x55,0x66,0x77,0x88,0x99,0x00,0x11,0x22,0x33,0x44,0x55,0x66,
-0x77,0x88,0x99,0x00,0x11,0x22,0x33,0x5d,0x00,0x5b,0x00,0x77,0x88,0x1e,0x00,0x00,
-0x57,0x00,0x00,0x00,0x00,0x04,0x00,0x00,0x4f,0x00,0x00,0x00,0x03,0x02,0x00,0x00,
-0x40,0x01,0x00,0x01,0xce,0x02,0x68,0x2d,0x5f,0xa9,0xde,0x0c,0xf6,0xd2,0x7b,0x58,
-0x4b,0xf9,0x28,0x68,0x3d,0xb4,0xf4,0xef,0x78,0xd5,0xbe,0x66,0x63,0x42,0xef,0xf8,
-0xfd,0xa4,0xf8,0xb0,0x8e,0x29,0xc2,0xc9,0x2e,0xd8,0x45,0xb8,0x53,0x8c,0x6f,0x4e,
-0x72,0x8f,0x6c,0x04,0x9c,0x88,0xfc,0x1e,0xc5,0x83,0x55,0x57,0xf7,0xdd,0xfd,0x4f,
-0x11,0x36,0x95,0x5d,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00
-};
-
-static int
-probe_device_type(struct device *devPtr)
-{
-	int rv, dv, i, index, length;
-	unsigned char psmid[8];
-	static unsigned char loc_testmsg[sizeof(static_testmsg)];
-
-	index = devPtr->dev_self_x;
-	rv = 0;
-	do {
-		memcpy(loc_testmsg, static_testmsg, sizeof(static_testmsg));
-		length = sizeof(static_testmsg) - 24;
-		/* the -24 allows for the header */
-		dv = send_to_AP(index, z90crypt.cdx, length, loc_testmsg);
-		if (dv) {
-			PDEBUG("dv returned by send during probe: %d\n", dv);
-			if (dv == DEV_SEN_EXCEPTION) {
-				rv = SEN_FATAL_ERROR;
-				PRINTKC("exception in send to AP %d\n", index);
-				break;
-			}
-			PDEBUG("return value from send_to_AP: %d\n", rv);
-			switch (dv) {
-			case DEV_GONE:
-				PDEBUG("dev %d not available\n", index);
-				rv = SEN_NOT_AVAIL;
-				break;
-			case DEV_ONLINE:
-				rv = 0;
-				break;
-			case DEV_EMPTY:
-				rv = SEN_NOT_AVAIL;
-				break;
-			case DEV_NO_WORK:
-				rv = SEN_FATAL_ERROR;
-				break;
-			case DEV_BAD_MESSAGE:
-				rv = SEN_USER_ERROR;
-				break;
-			case DEV_QUEUE_FULL:
-				rv = SEN_QUEUE_FULL;
-				break;
-			default:
-				PRINTK("unknown dv=%d for dev %d\n", dv, index);
-				rv = SEN_NOT_AVAIL;
-				break;
-			}
-		}
-
-		if (rv)
-			break;
-
-		for (i = 0; i < 6; i++) {
-			mdelay(300);
-			dv = receive_from_AP(index, z90crypt.cdx,
-					     devPtr->dev_resp_l,
-					     devPtr->dev_resp_p, psmid);
-			PDEBUG("dv returned by DQ = %d\n", dv);
-			if (dv == DEV_REC_EXCEPTION) {
-				rv = REC_FATAL_ERROR;
-				PRINTKC("exception in dequeue %d\n",
-					index);
-				break;
-			}
-			switch (dv) {
-			case DEV_ONLINE:
-				rv = 0;
-				break;
-			case DEV_EMPTY:
-				rv = REC_EMPTY;
-				break;
-			case DEV_NO_WORK:
-				rv = REC_NO_WORK;
-				break;
-			case DEV_BAD_MESSAGE:
-			case DEV_GONE:
-			default:
-				rv = REC_NO_RESPONSE;
-				break;
-			}
-			if ((rv != 0) && (rv != REC_NO_WORK))
-				break;
-			if (rv == 0)
-				break;
-		}
-		if (rv)
-			break;
-		rv = (devPtr->dev_resp_p[0] == 0x00) &&
-		     (devPtr->dev_resp_p[1] == 0x86);
-		if (rv)
-			devPtr->dev_type = PCICC;
-		else
-			devPtr->dev_type = PCICA;
-		rv = 0;
-	} while (0);
-	/* In a general error case, the card is not marked online */
-	return rv;
-}
-
-static unsigned char MCL3_testmsg[] = {
-0x00,0x00,0x00,0x00,0xEE,0xEE,0xEE,0xEE,0xEE,0xEE,0xEE,0xEE,
-0x00,0x06,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x58,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x43,0x41,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x50,0x4B,0x00,0x00,0x00,0x00,0x01,0xC4,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x07,0x24,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0xDC,0x02,0x00,0x00,0x00,0x54,0x32,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0xE8,0x00,0x00,0x00,0x00,0x00,0x00,0x07,0x24,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x04,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-0x00,0x00,0x00,0x00,0x50,0x4B,0x00,0x0A,0x4D,0x52,0x50,0x20,0x20,0x20,0x20,0x20,
-0x00,0x42,0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,
-0x0E,0x0F,0x00,0x11,0x22,0x33,0x44,0x55,0x66,0x77,0x88,0x99,0xAA,0xBB,0xCC,0xDD,
-0xEE,0xFF,0xFF,0xEE,0xDD,0xCC,0xBB,0xAA,0x99,0x88,0x77,0x66,0x55,0x44,0x33,0x22,
-0x11,0x00,0x01,0x23,0x45,0x67,0x89,0xAB,0xCD,0xEF,0xFE,0xDC,0xBA,0x98,0x76,0x54,
-0x32,0x10,0x00,0x9A,0x00,0x98,0x00,0x00,0x1E,0x00,0x00,0x94,0x00,0x00,0x00,0x00,
-0x04,0x00,0x00,0x8C,0x00,0x00,0x00,0x40,0x02,0x00,0x00,0x40,0xBA,0xE8,0x23,0x3C,
-0x75,0xF3,0x91,0x61,0xD6,0x73,0x39,0xCF,0x7B,0x6D,0x8E,0x61,0x97,0x63,0x9E,0xD9,
-0x60,0x55,0xD6,0xC7,0xEF,0xF8,0x1E,0x63,0x95,0x17,0xCC,0x28,0x45,0x60,0x11,0xC5,
-0xC4,0x4E,0x66,0xC6,0xE6,0xC3,0xDE,0x8A,0x19,0x30,0xCF,0x0E,0xD7,0xAA,0xDB,0x01,
-0xD8,0x00,0xBB,0x8F,0x39,0x9F,0x64,0x28,0xF5,0x7A,0x77,0x49,0xCC,0x6B,0xA3,0x91,
-0x97,0x70,0xE7,0x60,0x1E,0x39,0xE1,0xE5,0x33,0xE1,0x15,0x63,0x69,0x08,0x80,0x4C,
-0x67,0xC4,0x41,0x8F,0x48,0xDF,0x26,0x98,0xF1,0xD5,0x8D,0x88,0xD9,0x6A,0xA4,0x96,
-0xC5,0x84,0xD9,0x30,0x49,0x67,0x7D,0x19,0xB1,0xB3,0x45,0x4D,0xB2,0x53,0x9A,0x47,
-0x3C,0x7C,0x55,0xBF,0xCC,0x85,0x00,0x36,0xF1,0x3D,0x93,0x53
-};
-
-static int
-probe_PCIXCC_type(struct device *devPtr)
-{
-	int rv, dv, i, index, length;
-	unsigned char psmid[8];
-	static unsigned char loc_testmsg[548];
-	struct CPRBX *cprbx_p;
-
-	index = devPtr->dev_self_x;
-	rv = 0;
-	do {
-		memcpy(loc_testmsg, MCL3_testmsg, sizeof(MCL3_testmsg));
-		length = sizeof(MCL3_testmsg) - 0x0C;
-		dv = send_to_AP(index, z90crypt.cdx, length, loc_testmsg);
-		if (dv) {
-			PDEBUG("dv returned = %d\n", dv);
-			if (dv == DEV_SEN_EXCEPTION) {
-				rv = SEN_FATAL_ERROR;
-				PRINTKC("exception in send to AP %d\n", index);
-				break;
-			}
-			PDEBUG("return value from send_to_AP: %d\n", rv);
-			switch (dv) {
-			case DEV_GONE:
-				PDEBUG("dev %d not available\n", index);
-				rv = SEN_NOT_AVAIL;
-				break;
-			case DEV_ONLINE:
-				rv = 0;
-				break;
-			case DEV_EMPTY:
-				rv = SEN_NOT_AVAIL;
-				break;
-			case DEV_NO_WORK:
-				rv = SEN_FATAL_ERROR;
-				break;
-			case DEV_BAD_MESSAGE:
-				rv = SEN_USER_ERROR;
-				break;
-			case DEV_QUEUE_FULL:
-				rv = SEN_QUEUE_FULL;
-				break;
-			default:
-				PRINTK("unknown dv=%d for dev %d\n", dv, index);
-				rv = SEN_NOT_AVAIL;
-				break;
-			}
-		}
-
-		if (rv)
-			break;
-
-		for (i = 0; i < 6; i++) {
-			mdelay(300);
-			dv = receive_from_AP(index, z90crypt.cdx,
-					     devPtr->dev_resp_l,
-					     devPtr->dev_resp_p, psmid);
-			PDEBUG("dv returned by DQ = %d\n", dv);
-			if (dv == DEV_REC_EXCEPTION) {
-				rv = REC_FATAL_ERROR;
-				PRINTKC("exception in dequeue %d\n",
-					index);
-				break;
-			}
-			switch (dv) {
-			case DEV_ONLINE:
-				rv = 0;
-				break;
-			case DEV_EMPTY:
-				rv = REC_EMPTY;
-				break;
-			case DEV_NO_WORK:
-				rv = REC_NO_WORK;
-				break;
-			case DEV_BAD_MESSAGE:
-			case DEV_GONE:
-			default:
-				rv = REC_NO_RESPONSE;
-				break;
-			}
-			if ((rv != 0) && (rv != REC_NO_WORK))
-				break;
-			if (rv == 0)
-				break;
-		}
-		if (rv)
-			break;
-		cprbx_p = (struct CPRBX *) (devPtr->dev_resp_p + 48);
-		if ((cprbx_p->ccp_rtcode == 8) && (cprbx_p->ccp_rscode == 33)) {
-			devPtr->dev_type = PCIXCC_MCL2;
-			PDEBUG("device %d is MCL2\n", index);
-		} else {
-			devPtr->dev_type = PCIXCC_MCL3;
-			PDEBUG("device %d is MCL3\n", index);
-		}
-	} while (0);
-	/* In a general error case, the card is not marked online */
-	return rv;
-}
-
-module_init(z90crypt_init_module);
-module_exit(z90crypt_cleanup_module);
diff --git a/include/asm-s390/z90crypt.h b/include/asm-s390/z90crypt.h
deleted file mode 100644
index 31a2439b07bd..000000000000
--- a/include/asm-s390/z90crypt.h
+++ /dev/null
@@ -1,212 +0,0 @@
-/*
- *  include/asm-s390/z90crypt.h
- *
- *  z90crypt 1.3.3 (user-visible header)
- *
- *  Copyright (C)  2001, 2005 IBM Corporation
- *  Author(s): Robert Burroughs
- *             Eric Rossman (edrossma@us.ibm.com)
- *
- *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#ifndef __ASM_S390_Z90CRYPT_H
-#define __ASM_S390_Z90CRYPT_H
-#include <linux/ioctl.h>
-
-#define z90crypt_VERSION 1
-#define z90crypt_RELEASE 3	// 2 = PCIXCC, 3 = rewrite for coding standards
-#define z90crypt_VARIANT 3	// 3 = CEX2A support
-
-/**
- * struct ica_rsa_modexpo
- *
- * Requirements:
- * - outputdatalength is at least as large as inputdatalength.
- * - All key parts are right justified in their fields, padded on
- *   the left with zeroes.
- * - length(b_key) = inputdatalength
- * - length(n_modulus) = inputdatalength
- */
-struct ica_rsa_modexpo {
-	char __user *	inputdata;
-	unsigned int	inputdatalength;
-	char __user *	outputdata;
-	unsigned int	outputdatalength;
-	char __user *	b_key;
-	char __user *	n_modulus;
-};
-
-/**
- * struct ica_rsa_modexpo_crt
- *
- * Requirements:
- * - inputdatalength is even.
- * - outputdatalength is at least as large as inputdatalength.
- * - All key parts are right justified in their fields, padded on
- *   the left with zeroes.
- * - length(bp_key)	= inputdatalength/2 + 8
- * - length(bq_key)	= inputdatalength/2
- * - length(np_key)	= inputdatalength/2 + 8
- * - length(nq_key)	= inputdatalength/2
- * - length(u_mult_inv) = inputdatalength/2 + 8
- */
-struct ica_rsa_modexpo_crt {
-	char __user *	inputdata;
-	unsigned int	inputdatalength;
-	char __user *	outputdata;
-	unsigned int	outputdatalength;
-	char __user *	bp_key;
-	char __user *	bq_key;
-	char __user *	np_prime;
-	char __user *	nq_prime;
-	char __user *	u_mult_inv;
-};
-
-#define Z90_IOCTL_MAGIC 'z'  // NOTE:  Need to allocate from linux folks
-
-/**
- * Interface notes:
- *
- * The ioctl()s which are implemented (along with relevant details)
- * are:
- *
- *   ICARSAMODEXPO
- *     Perform an RSA operation using a Modulus-Exponent pair
- *     This takes an ica_rsa_modexpo struct as its arg.
- *
- *     NOTE: please refer to the comments preceding this structure
- *           for the implementation details for the contents of the
- *           block
- *
- *   ICARSACRT
- *     Perform an RSA operation using a Chinese-Remainder Theorem key
- *     This takes an ica_rsa_modexpo_crt struct as its arg.
- *
- *     NOTE: please refer to the comments preceding this structure
- *           for the implementation details for the contents of the
- *           block
- *
- *   Z90STAT_TOTALCOUNT
- *     Return an integer count of all device types together.
- *
- *   Z90STAT_PCICACOUNT
- *     Return an integer count of all PCICAs.
- *
- *   Z90STAT_PCICCCOUNT
- *     Return an integer count of all PCICCs.
- *
- *   Z90STAT_PCIXCCMCL2COUNT
- *     Return an integer count of all MCL2 PCIXCCs.
- *
- *   Z90STAT_PCIXCCMCL3COUNT
- *     Return an integer count of all MCL3 PCIXCCs.
- *
- *   Z90STAT_CEX2CCOUNT
- *     Return an integer count of all CEX2Cs.
- *
- *   Z90STAT_CEX2ACOUNT
- *     Return an integer count of all CEX2As.
- *
- *   Z90STAT_REQUESTQ_COUNT
- *     Return an integer count of the number of entries waiting to be
- *     sent to a device.
- *
- *   Z90STAT_PENDINGQ_COUNT
- *     Return an integer count of the number of entries sent to a
- *     device awaiting the reply.
- *
- *   Z90STAT_TOTALOPEN_COUNT
- *     Return an integer count of the number of open file handles.
- *
- *   Z90STAT_DOMAIN_INDEX
- *     Return the integer value of the Cryptographic Domain.
- *
- *   Z90STAT_STATUS_MASK
- *     Return an 64 element array of unsigned chars for the status of
- *     all devices.
- *       0x01: PCICA
- *       0x02: PCICC
- *       0x03: PCIXCC_MCL2
- *       0x04: PCIXCC_MCL3
- *       0x05: CEX2C
- *       0x06: CEX2A
- *       0x0d: device is disabled via the proc filesystem
- *
- *   Z90STAT_QDEPTH_MASK
- *     Return an 64 element array of unsigned chars for the queue
- *     depth of all devices.
- *
- *   Z90STAT_PERDEV_REQCNT
- *     Return an 64 element array of unsigned integers for the number
- *     of successfully completed requests per device since the device
- *     was detected and made available.
- *
- *   ICAZ90STATUS (deprecated)
- *     Return some device driver status in a ica_z90_status struct
- *     This takes an ica_z90_status struct as its arg.
- *
- *     NOTE: this ioctl() is deprecated, and has been replaced with
- *           single ioctl()s for each type of status being requested
- *
- *   Z90STAT_PCIXCCCOUNT (deprecated)
- *     Return an integer count of all PCIXCCs (MCL2 + MCL3).
- *     This is DEPRECATED now that MCL3 PCIXCCs are treated differently from
- *     MCL2 PCIXCCs.
- *
- *   Z90QUIESCE (not recommended)
- *     Quiesce the driver.  This is intended to stop all new
- *     requests from being processed.  Its use is NOT recommended,
- *     except in circumstances where there is no other way to stop
- *     callers from accessing the driver.  Its original use was to
- *     allow the driver to be "drained" of work in preparation for
- *     a system shutdown.
- *
- *     NOTE: once issued, this ban on new work cannot be undone
- *           except by unloading and reloading the driver.
- */
-
-/**
- * Supported ioctl calls
- */
-#define ICARSAMODEXPO	_IOC(_IOC_READ|_IOC_WRITE, Z90_IOCTL_MAGIC, 0x05, 0)
-#define ICARSACRT	_IOC(_IOC_READ|_IOC_WRITE, Z90_IOCTL_MAGIC, 0x06, 0)
-
-/* DEPRECATED status calls (bound for removal at some point) */
-#define ICAZ90STATUS	_IOR(Z90_IOCTL_MAGIC, 0x10, struct ica_z90_status)
-#define Z90STAT_PCIXCCCOUNT	_IOR(Z90_IOCTL_MAGIC, 0x43, int)
-
-/* unrelated to ICA callers */
-#define Z90QUIESCE	_IO(Z90_IOCTL_MAGIC, 0x11)
-
-/* New status calls */
-#define Z90STAT_TOTALCOUNT	_IOR(Z90_IOCTL_MAGIC, 0x40, int)
-#define Z90STAT_PCICACOUNT	_IOR(Z90_IOCTL_MAGIC, 0x41, int)
-#define Z90STAT_PCICCCOUNT	_IOR(Z90_IOCTL_MAGIC, 0x42, int)
-#define Z90STAT_PCIXCCMCL2COUNT	_IOR(Z90_IOCTL_MAGIC, 0x4b, int)
-#define Z90STAT_PCIXCCMCL3COUNT	_IOR(Z90_IOCTL_MAGIC, 0x4c, int)
-#define Z90STAT_CEX2CCOUNT	_IOR(Z90_IOCTL_MAGIC, 0x4d, int)
-#define Z90STAT_CEX2ACOUNT	_IOR(Z90_IOCTL_MAGIC, 0x4e, int)
-#define Z90STAT_REQUESTQ_COUNT	_IOR(Z90_IOCTL_MAGIC, 0x44, int)
-#define Z90STAT_PENDINGQ_COUNT	_IOR(Z90_IOCTL_MAGIC, 0x45, int)
-#define Z90STAT_TOTALOPEN_COUNT _IOR(Z90_IOCTL_MAGIC, 0x46, int)
-#define Z90STAT_DOMAIN_INDEX	_IOR(Z90_IOCTL_MAGIC, 0x47, int)
-#define Z90STAT_STATUS_MASK	_IOR(Z90_IOCTL_MAGIC, 0x48, char[64])
-#define Z90STAT_QDEPTH_MASK	_IOR(Z90_IOCTL_MAGIC, 0x49, char[64])
-#define Z90STAT_PERDEV_REQCNT	_IOR(Z90_IOCTL_MAGIC, 0x4a, int[64])
-
-#endif /* __ASM_S390_Z90CRYPT_H */
-- 
cgit v1.2.3


From 1534c3820c26aca4e2567f97b8add8bea40e7e2b Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:25 +0200
Subject: [S390] zcrypt adjunct processor bus.

Add a bus for the adjunct processor interface. Up to 64 devices can
be connect to the ap bus interface, each device with 16 domains. That
makes 1024 message queues. The interface is asynchronous, the answer
to a message sent to a queue needs to be received at some later point
in time. Unfortunately the interface does not provide interrupts when
a message reply is pending. So the ap bus needs to implement some
fancy polling, each active queue is polled once per 1/HZ second or
continuously if an idle cpus exsists and the poll thread is activ
(see poll_thread parameter).

The ap bus uses the sysfs path /sys/bus/ap and has two bus attributes,
ap_domain and config_time. The ap_domain selects one of the 16 domains
to be used for this system. This limits the maximum number of ap devices
to 64. The config_time attribute contains the number of seconds between
two ap bus scans to find new devices.

The ap bus uses the modalias entries of the form "ap:tN" to autoload
the ap driver for hardware type N. Currently known types are:
3 - PCICC, 4 - PCICA, 5 - PCIXCC, 6 - CEX2A and 7 - CEX2C.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Ralph Wuerthner <rwuerthn@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/crypto/ap_bus.c    | 1221 +++++++++++++++++++++++++++++++++++++++
 drivers/s390/crypto/ap_bus.h    |  158 +++++
 include/linux/mod_devicetable.h |   11 +
 scripts/mod/file2alias.c        |   12 +
 4 files changed, 1402 insertions(+)
 create mode 100644 drivers/s390/crypto/ap_bus.c
 create mode 100644 drivers/s390/crypto/ap_bus.h

(limited to 'include')

diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c
new file mode 100644
index 000000000000..6ed0985c0c91
--- /dev/null
+++ b/drivers/s390/crypto/ap_bus.c
@@ -0,0 +1,1221 @@
+/*
+ * linux/drivers/s390/crypto/ap_bus.c
+ *
+ * Copyright (C) 2006 IBM Corporation
+ * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com>
+ *	      Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *	      Ralph Wuerthner <rwuerthn@de.ibm.com>
+ *
+ * Adjunct processor bus.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/workqueue.h>
+#include <linux/notifier.h>
+#include <linux/kthread.h>
+#include <linux/mutex.h>
+#include <asm/s390_rdev.h>
+
+#include "ap_bus.h"
+
+/* Some prototypes. */
+static void ap_scan_bus(void *);
+static void ap_poll_all(unsigned long);
+static void ap_poll_timeout(unsigned long);
+static int ap_poll_thread_start(void);
+static void ap_poll_thread_stop(void);
+
+/**
+ * Module description.
+ */
+MODULE_AUTHOR("IBM Corporation");
+MODULE_DESCRIPTION("Adjunct Processor Bus driver, "
+		   "Copyright 2006 IBM Corporation");
+MODULE_LICENSE("GPL");
+
+/**
+ * Module parameter
+ */
+int ap_domain_index = -1;	/* Adjunct Processor Domain Index */
+module_param_named(domain, ap_domain_index, int, 0000);
+MODULE_PARM_DESC(domain, "domain index for ap devices");
+EXPORT_SYMBOL(ap_domain_index);
+
+static int ap_thread_flag = 1;
+module_param_named(poll_thread, ap_thread_flag, int, 0000);
+MODULE_PARM_DESC(poll_thread, "Turn on/off poll thread, default is 1 (on).");
+
+static struct device *ap_root_device = NULL;
+
+/**
+ * Workqueue & timer for bus rescan.
+ */
+static struct workqueue_struct *ap_work_queue;
+static struct timer_list ap_config_timer;
+static int ap_config_time = AP_CONFIG_TIME;
+static DECLARE_WORK(ap_config_work, ap_scan_bus, NULL);
+
+/**
+ * Tasklet & timer for AP request polling.
+ */
+static struct timer_list ap_poll_timer = TIMER_INITIALIZER(ap_poll_timeout,0,0);
+static DECLARE_TASKLET(ap_tasklet, ap_poll_all, 0);
+static atomic_t ap_poll_requests = ATOMIC_INIT(0);
+static DECLARE_WAIT_QUEUE_HEAD(ap_poll_wait);
+static struct task_struct *ap_poll_kthread = NULL;
+static DEFINE_MUTEX(ap_poll_thread_mutex);
+
+/**
+ * Test if ap instructions are available.
+ *
+ * Returns 0 if the ap instructions are installed.
+ */
+static inline int ap_instructions_available(void)
+{
+	register unsigned long reg0 asm ("0") = AP_MKQID(0,0);
+	register unsigned long reg1 asm ("1") = -ENODEV;
+	register unsigned long reg2 asm ("2") = 0UL;
+
+	asm volatile(
+		"   .long 0xb2af0000\n"		/* PQAP(TAPQ) */
+		"0: la    %1,0\n"
+		"1:\n"
+		EX_TABLE(0b, 1b)
+		: "+d" (reg0), "+d" (reg1), "+d" (reg2) : : "cc" );
+	return reg1;
+}
+
+/**
+ * Test adjunct processor queue.
+ * @qid: the ap queue number
+ * @queue_depth: pointer to queue depth value
+ * @device_type: pointer to device type value
+ *
+ * Returns ap queue status structure.
+ */
+static inline struct ap_queue_status
+ap_test_queue(ap_qid_t qid, int *queue_depth, int *device_type)
+{
+	register unsigned long reg0 asm ("0") = qid;
+	register struct ap_queue_status reg1 asm ("1");
+	register unsigned long reg2 asm ("2") = 0UL;
+
+	asm volatile(".long 0xb2af0000"		/* PQAP(TAPQ) */
+		     : "+d" (reg0), "=d" (reg1), "+d" (reg2) : : "cc");
+	*device_type = (int) (reg2 >> 24);
+	*queue_depth = (int) (reg2 & 0xff);
+	return reg1;
+}
+
+/**
+ * Reset adjunct processor queue.
+ * @qid: the ap queue number
+ *
+ * Returns ap queue status structure.
+ */
+static inline struct ap_queue_status ap_reset_queue(ap_qid_t qid)
+{
+	register unsigned long reg0 asm ("0") = qid | 0x01000000UL;
+	register struct ap_queue_status reg1 asm ("1");
+	register unsigned long reg2 asm ("2") = 0UL;
+
+	asm volatile(
+		".long 0xb2af0000"		/* PQAP(RAPQ) */
+		: "+d" (reg0), "=d" (reg1), "+d" (reg2) : : "cc");
+	return reg1;
+}
+
+/**
+ * Send message to adjunct processor queue.
+ * @qid: the ap queue number
+ * @psmid: the program supplied message identifier
+ * @msg: the message text
+ * @length: the message length
+ *
+ * Returns ap queue status structure.
+ *
+ * Condition code 1 on NQAP can't happen because the L bit is 1.
+ *
+ * Condition code 2 on NQAP also means the send is incomplete,
+ * because a segment boundary was reached. The NQAP is repeated.
+ */
+static inline struct ap_queue_status
+__ap_send(ap_qid_t qid, unsigned long long psmid, void *msg, size_t length)
+{
+	typedef struct { char _[length]; } msgblock;
+	register unsigned long reg0 asm ("0") = qid | 0x40000000UL;
+	register struct ap_queue_status reg1 asm ("1");
+	register unsigned long reg2 asm ("2") = (unsigned long) msg;
+	register unsigned long reg3 asm ("3") = (unsigned long) length;
+	register unsigned long reg4 asm ("4") = (unsigned int) (psmid >> 32);
+	register unsigned long reg5 asm ("5") = (unsigned int) psmid;
+
+	asm volatile (
+		"0: .long 0xb2ad0042\n"		/* DQAP */
+		"   brc   2,0b"
+		: "+d" (reg0), "=d" (reg1), "+d" (reg2), "+d" (reg3)
+		: "d" (reg4), "d" (reg5), "m" (*(msgblock *) msg)
+		: "cc" );
+	return reg1;
+}
+
+int ap_send(ap_qid_t qid, unsigned long long psmid, void *msg, size_t length)
+{
+	struct ap_queue_status status;
+
+	status = __ap_send(qid, psmid, msg, length);
+	switch (status.response_code) {
+	case AP_RESPONSE_NORMAL:
+		return 0;
+	case AP_RESPONSE_Q_FULL:
+		return -EBUSY;
+	default:	/* Device is gone. */
+		return -ENODEV;
+	}
+}
+EXPORT_SYMBOL(ap_send);
+
+/*
+ * Receive message from adjunct processor queue.
+ * @qid: the ap queue number
+ * @psmid: pointer to program supplied message identifier
+ * @msg: the message text
+ * @length: the message length
+ *
+ * Returns ap queue status structure.
+ *
+ * Condition code 1 on DQAP means the receive has taken place
+ * but only partially.	The response is incomplete, hence the
+ * DQAP is repeated.
+ *
+ * Condition code 2 on DQAP also means the receive is incomplete,
+ * this time because a segment boundary was reached. Again, the
+ * DQAP is repeated.
+ *
+ * Note that gpr2 is used by the DQAP instruction to keep track of
+ * any 'residual' length, in case the instruction gets interrupted.
+ * Hence it gets zeroed before the instruction.
+ */
+static inline struct ap_queue_status
+__ap_recv(ap_qid_t qid, unsigned long long *psmid, void *msg, size_t length)
+{
+	typedef struct { char _[length]; } msgblock;
+	register unsigned long reg0 asm("0") = qid | 0x80000000UL;
+	register struct ap_queue_status reg1 asm ("1");
+	register unsigned long reg2 asm("2") = 0UL;
+	register unsigned long reg4 asm("4") = (unsigned long) msg;
+	register unsigned long reg5 asm("5") = (unsigned long) length;
+	register unsigned long reg6 asm("6") = 0UL;
+	register unsigned long reg7 asm("7") = 0UL;
+
+
+	asm volatile(
+		"0: .long 0xb2ae0064\n"
+		"   brc   6,0b\n"
+		: "+d" (reg0), "=d" (reg1), "+d" (reg2),
+		"+d" (reg4), "+d" (reg5), "+d" (reg6), "+d" (reg7),
+		"=m" (*(msgblock *) msg) : : "cc" );
+	*psmid = (((unsigned long long) reg6) << 32) + reg7;
+	return reg1;
+}
+
+int ap_recv(ap_qid_t qid, unsigned long long *psmid, void *msg, size_t length)
+{
+	struct ap_queue_status status;
+
+	status = __ap_recv(qid, psmid, msg, length);
+	switch (status.response_code) {
+	case AP_RESPONSE_NORMAL:
+		return 0;
+	case AP_RESPONSE_NO_PENDING_REPLY:
+		if (status.queue_empty)
+			return -ENOENT;
+		return -EBUSY;
+	default:
+		return -ENODEV;
+	}
+}
+EXPORT_SYMBOL(ap_recv);
+
+/**
+ * Check if an AP queue is available. The test is repeated for
+ * AP_MAX_RESET times.
+ * @qid: the ap queue number
+ * @queue_depth: pointer to queue depth value
+ * @device_type: pointer to device type value
+ */
+static int ap_query_queue(ap_qid_t qid, int *queue_depth, int *device_type)
+{
+	struct ap_queue_status status;
+	int t_depth, t_device_type, rc, i;
+
+	rc = -EBUSY;
+	for (i = 0; i < AP_MAX_RESET; i++) {
+		status = ap_test_queue(qid, &t_depth, &t_device_type);
+		switch (status.response_code) {
+		case AP_RESPONSE_NORMAL:
+			*queue_depth = t_depth + 1;
+			*device_type = t_device_type;
+			rc = 0;
+			break;
+		case AP_RESPONSE_Q_NOT_AVAIL:
+			rc = -ENODEV;
+			break;
+		case AP_RESPONSE_RESET_IN_PROGRESS:
+			break;
+		case AP_RESPONSE_DECONFIGURED:
+			rc = -ENODEV;
+			break;
+		case AP_RESPONSE_CHECKSTOPPED:
+			rc = -ENODEV;
+			break;
+		case AP_RESPONSE_BUSY:
+			break;
+		default:
+			BUG();
+		}
+		if (rc != -EBUSY)
+			break;
+		if (i < AP_MAX_RESET - 1)
+			udelay(5);
+	}
+	return rc;
+}
+
+/**
+ * Reset an AP queue and wait for it to become available again.
+ * @qid: the ap queue number
+ */
+static int ap_init_queue(ap_qid_t qid)
+{
+	struct ap_queue_status status;
+	int rc, dummy, i;
+
+	rc = -ENODEV;
+	status = ap_reset_queue(qid);
+	for (i = 0; i < AP_MAX_RESET; i++) {
+		switch (status.response_code) {
+		case AP_RESPONSE_NORMAL:
+			if (status.queue_empty)
+				rc = 0;
+			break;
+		case AP_RESPONSE_Q_NOT_AVAIL:
+		case AP_RESPONSE_DECONFIGURED:
+		case AP_RESPONSE_CHECKSTOPPED:
+			i = AP_MAX_RESET;	/* return with -ENODEV */
+			break;
+		case AP_RESPONSE_RESET_IN_PROGRESS:
+		case AP_RESPONSE_BUSY:
+		default:
+			break;
+		}
+		if (rc != -ENODEV)
+			break;
+		if (i < AP_MAX_RESET - 1) {
+			udelay(5);
+			status = ap_test_queue(qid, &dummy, &dummy);
+		}
+	}
+	return rc;
+}
+
+/**
+ * AP device related attributes.
+ */
+static ssize_t ap_hwtype_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	return snprintf(buf, PAGE_SIZE, "%d\n", ap_dev->device_type);
+}
+static DEVICE_ATTR(hwtype, 0444, ap_hwtype_show, NULL);
+
+static ssize_t ap_depth_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	return snprintf(buf, PAGE_SIZE, "%d\n", ap_dev->queue_depth);
+}
+static DEVICE_ATTR(depth, 0444, ap_depth_show, NULL);
+
+static ssize_t ap_request_count_show(struct device *dev,
+				     struct device_attribute *attr,
+				     char *buf)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	int rc;
+
+	spin_lock_bh(&ap_dev->lock);
+	rc = snprintf(buf, PAGE_SIZE, "%d\n", ap_dev->total_request_count);
+	spin_unlock_bh(&ap_dev->lock);
+	return rc;
+}
+
+static DEVICE_ATTR(request_count, 0444, ap_request_count_show, NULL);
+
+static ssize_t ap_modalias_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "ap:t%02X", to_ap_dev(dev)->device_type);
+}
+
+static DEVICE_ATTR(modalias, 0444, ap_modalias_show, NULL);
+
+static struct attribute *ap_dev_attrs[] = {
+	&dev_attr_hwtype.attr,
+	&dev_attr_depth.attr,
+	&dev_attr_request_count.attr,
+	&dev_attr_modalias.attr,
+	NULL
+};
+static struct attribute_group ap_dev_attr_group = {
+	.attrs = ap_dev_attrs
+};
+
+/**
+ * AP bus driver registration/unregistration.
+ */
+static int ap_bus_match(struct device *dev, struct device_driver *drv)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	struct ap_driver *ap_drv = to_ap_drv(drv);
+	struct ap_device_id *id;
+
+	/**
+	 * Compare device type of the device with the list of
+	 * supported types of the device_driver.
+	 */
+	for (id = ap_drv->ids; id->match_flags; id++) {
+		if ((id->match_flags & AP_DEVICE_ID_MATCH_DEVICE_TYPE) &&
+		    (id->dev_type != ap_dev->device_type))
+			continue;
+		return 1;
+	}
+	return 0;
+}
+
+/**
+ * uevent function for AP devices. It sets up a single environment
+ * variable DEV_TYPE which contains the hardware device type.
+ */
+static int ap_uevent (struct device *dev, char **envp, int num_envp,
+		       char *buffer, int buffer_size)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	int length;
+
+	if (!ap_dev)
+		return -ENODEV;
+
+	/* Set up DEV_TYPE environment variable. */
+	envp[0] = buffer;
+	length = scnprintf(buffer, buffer_size, "DEV_TYPE=%04X",
+			   ap_dev->device_type);
+	if (buffer_size - length <= 0)
+		return -ENOMEM;
+	envp[1] = 0;
+	return 0;
+}
+
+static struct bus_type ap_bus_type = {
+	.name = "ap",
+	.match = &ap_bus_match,
+	.uevent = &ap_uevent,
+};
+
+static int ap_device_probe(struct device *dev)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	struct ap_driver *ap_drv = to_ap_drv(dev->driver);
+	int rc;
+
+	ap_dev->drv = ap_drv;
+	rc = ap_drv->probe ? ap_drv->probe(ap_dev) : -ENODEV;
+	if (rc)
+		ap_dev->unregistered = 1;
+	return rc;
+}
+
+/**
+ * Flush all requests from the request/pending queue of an AP device.
+ * @ap_dev: pointer to the AP device.
+ */
+static inline void __ap_flush_queue(struct ap_device *ap_dev)
+{
+	struct ap_message *ap_msg, *next;
+
+	list_for_each_entry_safe(ap_msg, next, &ap_dev->pendingq, list) {
+		list_del_init(&ap_msg->list);
+		ap_dev->pendingq_count--;
+		ap_dev->drv->receive(ap_dev, ap_msg, ERR_PTR(-ENODEV));
+	}
+	list_for_each_entry_safe(ap_msg, next, &ap_dev->requestq, list) {
+		list_del_init(&ap_msg->list);
+		ap_dev->requestq_count--;
+		ap_dev->drv->receive(ap_dev, ap_msg, ERR_PTR(-ENODEV));
+	}
+}
+
+void ap_flush_queue(struct ap_device *ap_dev)
+{
+	spin_lock_bh(&ap_dev->lock);
+	__ap_flush_queue(ap_dev);
+	spin_unlock_bh(&ap_dev->lock);
+}
+EXPORT_SYMBOL(ap_flush_queue);
+
+static int ap_device_remove(struct device *dev)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	struct ap_driver *ap_drv = ap_dev->drv;
+
+	spin_lock_bh(&ap_dev->lock);
+	__ap_flush_queue(ap_dev);
+	/**
+	 * set ->unregistered to 1 while holding the lock. This prevents
+	 * new messages to be put on the queue from now on.
+	 */
+	ap_dev->unregistered = 1;
+	spin_unlock_bh(&ap_dev->lock);
+	if (ap_drv->remove)
+		ap_drv->remove(ap_dev);
+	return 0;
+}
+
+int ap_driver_register(struct ap_driver *ap_drv, struct module *owner,
+		       char *name)
+{
+	struct device_driver *drv = &ap_drv->driver;
+
+	drv->bus = &ap_bus_type;
+	drv->probe = ap_device_probe;
+	drv->remove = ap_device_remove;
+	drv->owner = owner;
+	drv->name = name;
+	return driver_register(drv);
+}
+EXPORT_SYMBOL(ap_driver_register);
+
+void ap_driver_unregister(struct ap_driver *ap_drv)
+{
+	driver_unregister(&ap_drv->driver);
+}
+EXPORT_SYMBOL(ap_driver_unregister);
+
+/**
+ * AP bus attributes.
+ */
+static ssize_t ap_domain_show(struct bus_type *bus, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", ap_domain_index);
+}
+
+static BUS_ATTR(ap_domain, 0444, ap_domain_show, NULL);
+
+static ssize_t ap_config_time_show(struct bus_type *bus, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", ap_config_time);
+}
+
+static ssize_t ap_config_time_store(struct bus_type *bus,
+				    const char *buf, size_t count)
+{
+	int time;
+
+	if (sscanf(buf, "%d\n", &time) != 1 || time < 5 || time > 120)
+		return -EINVAL;
+	ap_config_time = time;
+	if (!timer_pending(&ap_config_timer) ||
+	    !mod_timer(&ap_config_timer, jiffies + ap_config_time * HZ)) {
+		ap_config_timer.expires = jiffies + ap_config_time * HZ;
+		add_timer(&ap_config_timer);
+	}
+	return count;
+}
+
+static BUS_ATTR(config_time, 0644, ap_config_time_show, ap_config_time_store);
+
+static ssize_t ap_poll_thread_show(struct bus_type *bus, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", ap_poll_kthread ? 1 : 0);
+}
+
+static ssize_t ap_poll_thread_store(struct bus_type *bus,
+				    const char *buf, size_t count)
+{
+	int flag, rc;
+
+	if (sscanf(buf, "%d\n", &flag) != 1)
+		return -EINVAL;
+	if (flag) {
+		rc = ap_poll_thread_start();
+		if (rc)
+			return rc;
+	}
+	else
+		ap_poll_thread_stop();
+	return count;
+}
+
+static BUS_ATTR(poll_thread, 0644, ap_poll_thread_show, ap_poll_thread_store);
+
+static struct bus_attribute *const ap_bus_attrs[] = {
+	&bus_attr_ap_domain,
+	&bus_attr_config_time,
+	&bus_attr_poll_thread,
+	NULL
+};
+
+/**
+ * Pick one of the 16 ap domains.
+ */
+static inline int ap_select_domain(void)
+{
+	int queue_depth, device_type, count, max_count, best_domain;
+	int rc, i, j;
+
+	/**
+	 * We want to use a single domain. Either the one specified with
+	 * the "domain=" parameter or the domain with the maximum number
+	 * of devices.
+	 */
+	if (ap_domain_index >= 0 && ap_domain_index < AP_DOMAINS)
+		/* Domain has already been selected. */
+		return 0;
+	best_domain = -1;
+	max_count = 0;
+	for (i = 0; i < AP_DOMAINS; i++) {
+		count = 0;
+		for (j = 0; j < AP_DEVICES; j++) {
+			ap_qid_t qid = AP_MKQID(j, i);
+			rc = ap_query_queue(qid, &queue_depth, &device_type);
+			if (rc)
+				continue;
+			count++;
+		}
+		if (count > max_count) {
+			max_count = count;
+			best_domain = i;
+		}
+	}
+	if (best_domain >= 0){
+		ap_domain_index = best_domain;
+		return 0;
+	}
+	return -ENODEV;
+}
+
+/**
+ * Find the device type if query queue returned a device type of 0.
+ * @ap_dev: pointer to the AP device.
+ */
+static int ap_probe_device_type(struct ap_device *ap_dev)
+{
+	static unsigned char msg[] = {
+		0x00,0x06,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x58,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x01,0x00,0x43,0x43,0x41,0x2d,0x41,0x50,
+		0x50,0x4c,0x20,0x20,0x20,0x01,0x01,0x01,
+		0x00,0x00,0x00,0x00,0x50,0x4b,0x00,0x00,
+		0x00,0x00,0x01,0x1c,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x05,0xb8,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x70,0x00,0x41,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x54,0x32,0x01,0x00,0xa0,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0xb8,0x05,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x0a,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
+		0x00,0x00,0x00,0x00,0x00,0x00,0x08,0x00,
+		0x49,0x43,0x53,0x46,0x20,0x20,0x20,0x20,
+		0x50,0x4b,0x0a,0x00,0x50,0x4b,0x43,0x53,
+		0x2d,0x31,0x2e,0x32,0x37,0x00,0x11,0x22,
+		0x33,0x44,0x55,0x66,0x77,0x88,0x99,0x00,
+		0x11,0x22,0x33,0x44,0x55,0x66,0x77,0x88,
+		0x99,0x00,0x11,0x22,0x33,0x44,0x55,0x66,
+		0x77,0x88,0x99,0x00,0x11,0x22,0x33,0x44,
+		0x55,0x66,0x77,0x88,0x99,0x00,0x11,0x22,
+		0x33,0x44,0x55,0x66,0x77,0x88,0x99,0x00,
+		0x11,0x22,0x33,0x5d,0x00,0x5b,0x00,0x77,
+		0x88,0x1e,0x00,0x00,0x57,0x00,0x00,0x00,
+		0x00,0x04,0x00,0x00,0x4f,0x00,0x00,0x00,
+		0x03,0x02,0x00,0x00,0x40,0x01,0x00,0x01,
+		0xce,0x02,0x68,0x2d,0x5f,0xa9,0xde,0x0c,
+		0xf6,0xd2,0x7b,0x58,0x4b,0xf9,0x28,0x68,
+		0x3d,0xb4,0xf4,0xef,0x78,0xd5,0xbe,0x66,
+		0x63,0x42,0xef,0xf8,0xfd,0xa4,0xf8,0xb0,
+		0x8e,0x29,0xc2,0xc9,0x2e,0xd8,0x45,0xb8,
+		0x53,0x8c,0x6f,0x4e,0x72,0x8f,0x6c,0x04,
+		0x9c,0x88,0xfc,0x1e,0xc5,0x83,0x55,0x57,
+		0xf7,0xdd,0xfd,0x4f,0x11,0x36,0x95,0x5d,
+	};
+	struct ap_queue_status status;
+	unsigned long long psmid;
+	char *reply;
+	int rc, i;
+
+	reply = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!reply) {
+		rc = -ENOMEM;
+		goto out;
+	}
+
+	status = __ap_send(ap_dev->qid, 0x0102030405060708ULL,
+			   msg, sizeof(msg));
+	if (status.response_code != AP_RESPONSE_NORMAL) {
+		rc = -ENODEV;
+		goto out_free;
+	}
+
+	/* Wait for the test message to complete. */
+	for (i = 0; i < 6; i++) {
+		mdelay(300);
+		status = __ap_recv(ap_dev->qid, &psmid, reply, 4096);
+		if (status.response_code == AP_RESPONSE_NORMAL &&
+		    psmid == 0x0102030405060708ULL)
+			break;
+	}
+	if (i < 6) {
+		/* Got an answer. */
+		if (reply[0] == 0x00 && reply[1] == 0x86)
+			ap_dev->device_type = AP_DEVICE_TYPE_PCICC;
+		else
+			ap_dev->device_type = AP_DEVICE_TYPE_PCICA;
+		rc = 0;
+	} else
+		rc = -ENODEV;
+
+out_free:
+	free_page((unsigned long) reply);
+out:
+	return rc;
+}
+
+/**
+ * Scan the ap bus for new devices.
+ */
+static int __ap_scan_bus(struct device *dev, void *data)
+{
+	return to_ap_dev(dev)->qid == (ap_qid_t)(unsigned long) data;
+}
+
+static void ap_device_release(struct device *dev)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+
+	kfree(ap_dev);
+}
+
+static void ap_scan_bus(void *data)
+{
+	struct ap_device *ap_dev;
+	struct device *dev;
+	ap_qid_t qid;
+	int queue_depth, device_type;
+	int rc, i;
+
+	if (ap_select_domain() != 0)
+		return;
+	for (i = 0; i < AP_DEVICES; i++) {
+		qid = AP_MKQID(i, ap_domain_index);
+		dev = bus_find_device(&ap_bus_type, NULL,
+				      (void *)(unsigned long)qid,
+				      __ap_scan_bus);
+		if (dev) {
+			put_device(dev);
+			continue;
+		}
+		rc = ap_query_queue(qid, &queue_depth, &device_type);
+		if (rc)
+			continue;
+		rc = ap_init_queue(qid);
+		if (rc)
+			continue;
+		ap_dev = kzalloc(sizeof(*ap_dev), GFP_KERNEL);
+		if (!ap_dev)
+			break;
+		ap_dev->qid = qid;
+		ap_dev->queue_depth = queue_depth;
+		spin_lock_init(&ap_dev->lock);
+		INIT_LIST_HEAD(&ap_dev->pendingq);
+		INIT_LIST_HEAD(&ap_dev->requestq);
+		if (device_type == 0)
+			ap_probe_device_type(ap_dev);
+		else
+			ap_dev->device_type = device_type;
+
+		ap_dev->device.bus = &ap_bus_type;
+		ap_dev->device.parent = ap_root_device;
+		snprintf(ap_dev->device.bus_id, BUS_ID_SIZE, "card%02x",
+			 AP_QID_DEVICE(ap_dev->qid));
+		ap_dev->device.release = ap_device_release;
+		rc = device_register(&ap_dev->device);
+		if (rc) {
+			kfree(ap_dev);
+			continue;
+		}
+		/* Add device attributes. */
+		rc = sysfs_create_group(&ap_dev->device.kobj,
+					&ap_dev_attr_group);
+		if (rc)
+			device_unregister(&ap_dev->device);
+	}
+}
+
+static void
+ap_config_timeout(unsigned long ptr)
+{
+	queue_work(ap_work_queue, &ap_config_work);
+	ap_config_timer.expires = jiffies + ap_config_time * HZ;
+	add_timer(&ap_config_timer);
+}
+
+/**
+ * Set up the timer to run the poll tasklet
+ */
+static inline void ap_schedule_poll_timer(void)
+{
+	if (timer_pending(&ap_poll_timer))
+		return;
+	mod_timer(&ap_poll_timer, jiffies + AP_POLL_TIME);
+}
+
+/**
+ * Receive pending reply messages from an AP device.
+ * @ap_dev: pointer to the AP device
+ * @flags: pointer to control flags, bit 2^0 is set if another poll is
+ *	   required, bit 2^1 is set if the poll timer needs to get armed
+ * Returns 0 if the device is still present, -ENODEV if not.
+ */
+static inline int ap_poll_read(struct ap_device *ap_dev, unsigned long *flags)
+{
+	struct ap_queue_status status;
+	struct ap_message *ap_msg;
+
+	if (ap_dev->queue_count <= 0)
+		return 0;
+	status = __ap_recv(ap_dev->qid, &ap_dev->reply->psmid,
+			   ap_dev->reply->message, ap_dev->reply->length);
+	switch (status.response_code) {
+	case AP_RESPONSE_NORMAL:
+		atomic_dec(&ap_poll_requests);
+		ap_dev->queue_count--;
+		list_for_each_entry(ap_msg, &ap_dev->pendingq, list) {
+			if (ap_msg->psmid != ap_dev->reply->psmid)
+				continue;
+			list_del_init(&ap_msg->list);
+			ap_dev->pendingq_count--;
+			ap_dev->drv->receive(ap_dev, ap_msg, ap_dev->reply);
+			break;
+		}
+		if (ap_dev->queue_count > 0)
+			*flags |= 1;
+		break;
+	case AP_RESPONSE_NO_PENDING_REPLY:
+		if (status.queue_empty) {
+			/* The card shouldn't forget requests but who knows. */
+			ap_dev->queue_count = 0;
+			list_splice_init(&ap_dev->pendingq, &ap_dev->requestq);
+			ap_dev->requestq_count += ap_dev->pendingq_count;
+			ap_dev->pendingq_count = 0;
+		} else
+			*flags |= 2;
+		break;
+	default:
+		return -ENODEV;
+	}
+	return 0;
+}
+
+/**
+ * Send messages from the request queue to an AP device.
+ * @ap_dev: pointer to the AP device
+ * @flags: pointer to control flags, bit 2^0 is set if another poll is
+ *	   required, bit 2^1 is set if the poll timer needs to get armed
+ * Returns 0 if the device is still present, -ENODEV if not.
+ */
+static inline int ap_poll_write(struct ap_device *ap_dev, unsigned long *flags)
+{
+	struct ap_queue_status status;
+	struct ap_message *ap_msg;
+
+	if (ap_dev->requestq_count <= 0 ||
+	    ap_dev->queue_count >= ap_dev->queue_depth)
+		return 0;
+	/* Start the next request on the queue. */
+	ap_msg = list_entry(ap_dev->requestq.next, struct ap_message, list);
+	status = __ap_send(ap_dev->qid, ap_msg->psmid,
+			   ap_msg->message, ap_msg->length);
+	switch (status.response_code) {
+	case AP_RESPONSE_NORMAL:
+		atomic_inc(&ap_poll_requests);
+		ap_dev->queue_count++;
+		list_move_tail(&ap_msg->list, &ap_dev->pendingq);
+		ap_dev->requestq_count--;
+		ap_dev->pendingq_count++;
+		if (ap_dev->queue_count < ap_dev->queue_depth &&
+		    ap_dev->requestq_count > 0)
+			*flags |= 1;
+		*flags |= 2;
+		break;
+	case AP_RESPONSE_Q_FULL:
+		*flags |= 2;
+		break;
+	case AP_RESPONSE_MESSAGE_TOO_BIG:
+		return -EINVAL;
+	default:
+		return -ENODEV;
+	}
+	return 0;
+}
+
+/**
+ * Poll AP device for pending replies and send new messages. If either
+ * ap_poll_read or ap_poll_write returns -ENODEV unregister the device.
+ * @ap_dev: pointer to the bus device
+ * @flags: pointer to control flags, bit 2^0 is set if another poll is
+ *	   required, bit 2^1 is set if the poll timer needs to get armed
+ * Returns 0.
+ */
+static inline int ap_poll_queue(struct ap_device *ap_dev, unsigned long *flags)
+{
+	int rc;
+
+	rc = ap_poll_read(ap_dev, flags);
+	if (rc)
+		return rc;
+	return ap_poll_write(ap_dev, flags);
+}
+
+/**
+ * Queue a message to a device.
+ * @ap_dev: pointer to the AP device
+ * @ap_msg: the message to be queued
+ */
+static int __ap_queue_message(struct ap_device *ap_dev, struct ap_message *ap_msg)
+{
+	struct ap_queue_status status;
+
+	if (list_empty(&ap_dev->requestq) &&
+	    ap_dev->queue_count < ap_dev->queue_depth) {
+		status = __ap_send(ap_dev->qid, ap_msg->psmid,
+				   ap_msg->message, ap_msg->length);
+		switch (status.response_code) {
+		case AP_RESPONSE_NORMAL:
+			list_add_tail(&ap_msg->list, &ap_dev->pendingq);
+			atomic_inc(&ap_poll_requests);
+			ap_dev->pendingq_count++;
+			ap_dev->queue_count++;
+			ap_dev->total_request_count++;
+			break;
+		case AP_RESPONSE_Q_FULL:
+			list_add_tail(&ap_msg->list, &ap_dev->requestq);
+			ap_dev->requestq_count++;
+			ap_dev->total_request_count++;
+			return -EBUSY;
+		case AP_RESPONSE_MESSAGE_TOO_BIG:
+			ap_dev->drv->receive(ap_dev, ap_msg, ERR_PTR(-EINVAL));
+			return -EINVAL;
+		default:	/* Device is gone. */
+			ap_dev->drv->receive(ap_dev, ap_msg, ERR_PTR(-ENODEV));
+			return -ENODEV;
+		}
+	} else {
+		list_add_tail(&ap_msg->list, &ap_dev->requestq);
+		ap_dev->requestq_count++;
+		ap_dev->total_request_count++;
+		return -EBUSY;
+	}
+	ap_schedule_poll_timer();
+	return 0;
+}
+
+void ap_queue_message(struct ap_device *ap_dev, struct ap_message *ap_msg)
+{
+	unsigned long flags;
+	int rc;
+
+	spin_lock_bh(&ap_dev->lock);
+	if (!ap_dev->unregistered) {
+		/* Make room on the queue by polling for finished requests. */
+		rc = ap_poll_queue(ap_dev, &flags);
+		if (!rc)
+			rc = __ap_queue_message(ap_dev, ap_msg);
+		if (!rc)
+			wake_up(&ap_poll_wait);
+	} else {
+		ap_dev->drv->receive(ap_dev, ap_msg, ERR_PTR(-ENODEV));
+		rc = 0;
+	}
+	spin_unlock_bh(&ap_dev->lock);
+	if (rc == -ENODEV)
+		device_unregister(&ap_dev->device);
+}
+EXPORT_SYMBOL(ap_queue_message);
+
+/**
+ * Cancel a crypto request. This is done by removing the request
+ * from the devive pendingq or requestq queue. Note that the
+ * request stays on the AP queue. When it finishes the message
+ * reply will be discarded because the psmid can't be found.
+ * @ap_dev: AP device that has the message queued
+ * @ap_msg: the message that is to be removed
+ */
+void ap_cancel_message(struct ap_device *ap_dev, struct ap_message *ap_msg)
+{
+	struct ap_message *tmp;
+
+	spin_lock_bh(&ap_dev->lock);
+	if (!list_empty(&ap_msg->list)) {
+		list_for_each_entry(tmp, &ap_dev->pendingq, list)
+			if (tmp->psmid == ap_msg->psmid) {
+				ap_dev->pendingq_count--;
+				goto found;
+			}
+		ap_dev->requestq_count--;
+	found:
+		list_del_init(&ap_msg->list);
+	}
+	spin_unlock_bh(&ap_dev->lock);
+}
+EXPORT_SYMBOL(ap_cancel_message);
+
+/**
+ * AP receive polling for finished AP requests
+ */
+static void ap_poll_timeout(unsigned long unused)
+{
+	tasklet_schedule(&ap_tasklet);
+}
+
+/**
+ * Poll all AP devices on the bus in a round robin fashion. Continue
+ * polling until bit 2^0 of the control flags is not set. If bit 2^1
+ * of the control flags has been set arm the poll timer.
+ */
+static int __ap_poll_all(struct device *dev, void *data)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	int rc;
+
+	spin_lock(&ap_dev->lock);
+	if (!ap_dev->unregistered) {
+		rc = ap_poll_queue(to_ap_dev(dev), (unsigned long *) data);
+	} else
+		rc = 0;
+	spin_unlock(&ap_dev->lock);
+	if (rc)
+		device_unregister(&ap_dev->device);
+	return 0;
+}
+
+static void ap_poll_all(unsigned long dummy)
+{
+	unsigned long flags;
+
+	do {
+		flags = 0;
+		bus_for_each_dev(&ap_bus_type, NULL, &flags, __ap_poll_all);
+	} while (flags & 1);
+	if (flags & 2)
+		ap_schedule_poll_timer();
+}
+
+/**
+ * AP bus poll thread. The purpose of this thread is to poll for
+ * finished requests in a loop if there is a "free" cpu - that is
+ * a cpu that doesn't have anything better to do. The polling stops
+ * as soon as there is another task or if all messages have been
+ * delivered.
+ */
+static int ap_poll_thread(void *data)
+{
+	DECLARE_WAITQUEUE(wait, current);
+	unsigned long flags;
+	int requests;
+
+	set_user_nice(current, -20);
+	while (1) {
+		if (need_resched()) {
+			schedule();
+			continue;
+		}
+		add_wait_queue(&ap_poll_wait, &wait);
+		set_current_state(TASK_INTERRUPTIBLE);
+		if (kthread_should_stop())
+			break;
+		requests = atomic_read(&ap_poll_requests);
+		if (requests <= 0)
+			schedule();
+		set_current_state(TASK_RUNNING);
+		remove_wait_queue(&ap_poll_wait, &wait);
+
+		local_bh_disable();
+		flags = 0;
+		bus_for_each_dev(&ap_bus_type, NULL, &flags, __ap_poll_all);
+		local_bh_enable();
+	}
+	set_current_state(TASK_RUNNING);
+	remove_wait_queue(&ap_poll_wait, &wait);
+	return 0;
+}
+
+static int ap_poll_thread_start(void)
+{
+	int rc;
+
+	mutex_lock(&ap_poll_thread_mutex);
+	if (!ap_poll_kthread) {
+		ap_poll_kthread = kthread_run(ap_poll_thread, NULL, "appoll");
+		rc = IS_ERR(ap_poll_kthread) ? PTR_ERR(ap_poll_kthread) : 0;
+		if (rc)
+			ap_poll_kthread = NULL;
+	}
+	else
+		rc = 0;
+	mutex_unlock(&ap_poll_thread_mutex);
+	return rc;
+}
+
+static void ap_poll_thread_stop(void)
+{
+	mutex_lock(&ap_poll_thread_mutex);
+	if (ap_poll_kthread) {
+		kthread_stop(ap_poll_kthread);
+		ap_poll_kthread = NULL;
+	}
+	mutex_unlock(&ap_poll_thread_mutex);
+}
+
+/**
+ * The module initialization code.
+ */
+int __init ap_module_init(void)
+{
+	int rc, i;
+
+	if (ap_domain_index < -1 || ap_domain_index >= AP_DOMAINS) {
+		printk(KERN_WARNING "Invalid param: domain = %d. "
+		       " Not loading.\n", ap_domain_index);
+		return -EINVAL;
+	}
+	if (ap_instructions_available() != 0) {
+		printk(KERN_WARNING "AP instructions not installed.\n");
+		return -ENODEV;
+	}
+
+	/* Create /sys/bus/ap. */
+	rc = bus_register(&ap_bus_type);
+	if (rc)
+		goto out;
+	for (i = 0; ap_bus_attrs[i]; i++) {
+		rc = bus_create_file(&ap_bus_type, ap_bus_attrs[i]);
+		if (rc)
+			goto out_bus;
+	}
+
+	/* Create /sys/devices/ap. */
+	ap_root_device = s390_root_dev_register("ap");
+	rc = IS_ERR(ap_root_device) ? PTR_ERR(ap_root_device) : 0;
+	if (rc)
+		goto out_bus;
+
+	ap_work_queue = create_singlethread_workqueue("kapwork");
+	if (!ap_work_queue) {
+		rc = -ENOMEM;
+		goto out_root;
+	}
+
+	if (ap_select_domain() == 0)
+		ap_scan_bus(NULL);
+
+	/* Setup the ap bus rescan timer. */
+	init_timer(&ap_config_timer);
+	ap_config_timer.function = ap_config_timeout;
+	ap_config_timer.data = 0;
+	ap_config_timer.expires = jiffies + ap_config_time * HZ;
+	add_timer(&ap_config_timer);
+
+	/* Start the low priority AP bus poll thread. */
+	if (ap_thread_flag) {
+		rc = ap_poll_thread_start();
+		if (rc)
+			goto out_work;
+	}
+
+	return 0;
+
+out_work:
+	del_timer_sync(&ap_config_timer);
+	del_timer_sync(&ap_poll_timer);
+	destroy_workqueue(ap_work_queue);
+out_root:
+	s390_root_dev_unregister(ap_root_device);
+out_bus:
+	while (i--)
+		bus_remove_file(&ap_bus_type, ap_bus_attrs[i]);
+	bus_unregister(&ap_bus_type);
+out:
+	return rc;
+}
+
+static int __ap_match_all(struct device *dev, void *data)
+{
+	return 1;
+}
+
+/**
+ * The module termination code
+ */
+void ap_module_exit(void)
+{
+	int i;
+	struct device *dev;
+
+	ap_poll_thread_stop();
+	del_timer_sync(&ap_config_timer);
+	del_timer_sync(&ap_poll_timer);
+	destroy_workqueue(ap_work_queue);
+	s390_root_dev_unregister(ap_root_device);
+	while ((dev = bus_find_device(&ap_bus_type, NULL, NULL,
+		    __ap_match_all)))
+	{
+		device_unregister(dev);
+		put_device(dev);
+	}
+	for (i = 0; ap_bus_attrs[i]; i++)
+		bus_remove_file(&ap_bus_type, ap_bus_attrs[i]);
+	bus_unregister(&ap_bus_type);
+}
+
+#ifndef CONFIG_ZCRYPT_MONOLITHIC
+module_init(ap_module_init);
+module_exit(ap_module_exit);
+#endif
diff --git a/drivers/s390/crypto/ap_bus.h b/drivers/s390/crypto/ap_bus.h
new file mode 100644
index 000000000000..83b69c01cd6e
--- /dev/null
+++ b/drivers/s390/crypto/ap_bus.h
@@ -0,0 +1,158 @@
+/*
+ * linux/drivers/s390/crypto/ap_bus.h
+ *
+ * Copyright (C) 2006 IBM Corporation
+ * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com>
+ *	      Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *	      Ralph Wuerthner <rwuerthn@de.ibm.com>
+ *
+ * Adjunct processor bus header file.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _AP_BUS_H_
+#define _AP_BUS_H_
+
+#include <linux/device.h>
+#include <linux/mod_devicetable.h>
+#include <linux/types.h>
+
+#define AP_DEVICES 64		/* Number of AP devices. */
+#define AP_DOMAINS 16		/* Number of AP domains. */
+#define AP_MAX_RESET 90		/* Maximum number of resets. */
+#define AP_CONFIG_TIME 30	/* Time in seconds between AP bus rescans. */
+#define AP_POLL_TIME 1		/* Time in ticks between receive polls. */
+
+extern int ap_domain_index;
+
+/**
+ * The ap_qid_t identifier of an ap queue. It contains a
+ * 6 bit device index and a 4 bit queue index (domain).
+ */
+typedef unsigned int ap_qid_t;
+
+#define AP_MKQID(_device,_queue) (((_device) & 63) << 8 | ((_queue) & 15))
+#define AP_QID_DEVICE(_qid) (((_qid) >> 8) & 63)
+#define AP_QID_QUEUE(_qid) ((_qid) & 15)
+
+/**
+ * The ap queue status word is returned by all three AP functions
+ * (PQAP, NQAP and DQAP).  There's a set of flags in the first
+ * byte, followed by a 1 byte response code.
+ */
+struct ap_queue_status {
+	unsigned int queue_empty	: 1;
+	unsigned int replies_waiting	: 1;
+	unsigned int queue_full		: 1;
+	unsigned int pad1		: 5;
+	unsigned int response_code	: 8;
+	unsigned int pad2		: 16;
+};
+
+#define AP_RESPONSE_NORMAL		0x00
+#define AP_RESPONSE_Q_NOT_AVAIL		0x01
+#define AP_RESPONSE_RESET_IN_PROGRESS	0x02
+#define AP_RESPONSE_DECONFIGURED	0x03
+#define AP_RESPONSE_CHECKSTOPPED	0x04
+#define AP_RESPONSE_BUSY		0x05
+#define AP_RESPONSE_Q_FULL		0x10
+#define AP_RESPONSE_NO_PENDING_REPLY	0x10
+#define AP_RESPONSE_INDEX_TOO_BIG	0x11
+#define AP_RESPONSE_NO_FIRST_PART	0x13
+#define AP_RESPONSE_MESSAGE_TOO_BIG	0x15
+
+/**
+ * Known device types
+ */
+#define AP_DEVICE_TYPE_PCICC	3
+#define AP_DEVICE_TYPE_PCICA	4
+#define AP_DEVICE_TYPE_PCIXCC	5
+#define AP_DEVICE_TYPE_CEX2A	6
+#define AP_DEVICE_TYPE_CEX2C	7
+
+struct ap_device;
+struct ap_message;
+
+struct ap_driver {
+	struct device_driver driver;
+	struct ap_device_id *ids;
+
+	int (*probe)(struct ap_device *);
+	void (*remove)(struct ap_device *);
+	/* receive is called from tasklet context */
+	void (*receive)(struct ap_device *, struct ap_message *,
+			struct ap_message *);
+};
+
+#define to_ap_drv(x) container_of((x), struct ap_driver, driver)
+
+int ap_driver_register(struct ap_driver *, struct module *, char *);
+void ap_driver_unregister(struct ap_driver *);
+
+struct ap_device {
+	struct device device;
+	struct ap_driver *drv;		/* Pointer to AP device driver. */
+	spinlock_t lock;		/* Per device lock. */
+
+	ap_qid_t qid;			/* AP queue id. */
+	int queue_depth;		/* AP queue depth.*/
+	int device_type;		/* AP device type. */
+	int unregistered;		/* marks AP device as unregistered */
+
+	int queue_count;		/* # messages currently on AP queue. */
+
+	struct list_head pendingq;	/* List of message sent to AP queue. */
+	int pendingq_count;		/* # requests on pendingq list. */
+	struct list_head requestq;	/* List of message yet to be sent. */
+	int requestq_count;		/* # requests on requestq list. */
+	int total_request_count;	/* # requests ever for this AP device. */
+
+	struct ap_message *reply;	/* Per device reply message. */
+
+	void *private;			/* ap driver private pointer. */
+};
+
+#define to_ap_dev(x) container_of((x), struct ap_device, device)
+
+struct ap_message {
+	struct list_head list;		/* Request queueing. */
+	unsigned long long psmid;	/* Message id. */
+	void *message;			/* Pointer to message buffer. */
+	size_t length;			/* Message length. */
+
+	void *private;			/* ap driver private pointer. */
+};
+
+#define AP_DEVICE(dt)					\
+	.dev_type=(dt),					\
+	.match_flags=AP_DEVICE_ID_MATCH_DEVICE_TYPE,
+
+/**
+ * Note: don't use ap_send/ap_recv after using ap_queue_message
+ * for the first time. Otherwise the ap message queue will get
+ * confused.
+ */
+int ap_send(ap_qid_t, unsigned long long, void *, size_t);
+int ap_recv(ap_qid_t, unsigned long long *, void *, size_t);
+
+void ap_queue_message(struct ap_device *ap_dev, struct ap_message *ap_msg);
+void ap_cancel_message(struct ap_device *ap_dev, struct ap_message *ap_msg);
+void ap_flush_queue(struct ap_device *ap_dev);
+
+int ap_module_init(void);
+void ap_module_exit(void);
+
+#endif /* _AP_BUS_H_ */
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index f6977708585c..f7ca0b09075d 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -148,6 +148,17 @@ struct ccw_device_id {
 #define CCW_DEVICE_ID_MATCH_DEVICE_TYPE		0x04
 #define CCW_DEVICE_ID_MATCH_DEVICE_MODEL	0x08
 
+/* s390 AP bus devices */
+struct ap_device_id {
+	__u16 match_flags;	/* which fields to match against */
+	__u8 dev_type;		/* device type */
+	__u8 pad1;
+	__u32 pad2;
+	kernel_ulong_t driver_info;
+};
+
+#define AP_DEVICE_ID_MATCH_DEVICE_TYPE		0x01
+
 
 #define PNP_ID_LEN	8
 #define PNP_MAX_DEVICES	8
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index e2de650d3dbf..de76da80443f 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -265,6 +265,14 @@ static int do_ccw_entry(const char *filename,
 	return 1;
 }
 
+/* looks like: "ap:tN" */
+static int do_ap_entry(const char *filename,
+		       struct ap_device_id *id, char *alias)
+{
+	sprintf(alias, "ap:t%02X", id->dev_type);
+	return 1;
+}
+
 /* Looks like: "serio:tyNprNidNexN" */
 static int do_serio_entry(const char *filename,
 			  struct serio_device_id *id, char *alias)
@@ -503,6 +511,10 @@ void handle_moddevtable(struct module *mod, struct elf_info *info,
 		do_table(symval, sym->st_size,
 			 sizeof(struct ccw_device_id), "ccw",
 			 do_ccw_entry, mod);
+	else if (sym_is(symname, "__mod_ap_device_table"))
+		do_table(symval, sym->st_size,
+			 sizeof(struct ap_device_id), "ap",
+			 do_ap_entry, mod);
 	else if (sym_is(symname, "__mod_serio_device_table"))
 		do_table(symval, sym->st_size,
 			 sizeof(struct serio_device_id), "serio",
-- 
cgit v1.2.3


From 2dbc2418bac32a18a372ae9aec386f0fe9174389 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:27 +0200
Subject: [S390] zcrypt user space interface.

The user space interface of the zcrypt device driver implements the old
user space interface as defined by the old z90crypt driver. Everything
is there, the /dev/z90crypt misc character device, all the lovely ioctls
and the /proc file. Even writing to the z90crypt proc file to configure
the crypto device still works. It stands to reason to remove the proc
write function someday since a much cleaner configuration via the sysfs
is now available.

The ap bus device drivers register crypto cards to the zcrypt user
space interface. The request router of the user space interface
picks one of the registered cards based on the predicted latency
for the request and calls the driver via a callback found in the
zcrypt_ops of the device. The request router only knows which
operations the card can do and the minimum / maximum number of bits
a request can have.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Ralph Wuerthner <rwuerthn@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/crypto/zcrypt_api.c | 981 +++++++++++++++++++++++++++++++++++++++
 drivers/s390/crypto/zcrypt_api.h | 140 ++++++
 include/asm-s390/zcrypt.h        | 207 +++++++++
 3 files changed, 1328 insertions(+)
 create mode 100644 drivers/s390/crypto/zcrypt_api.c
 create mode 100644 drivers/s390/crypto/zcrypt_api.h
 create mode 100644 include/asm-s390/zcrypt.h

(limited to 'include')

diff --git a/drivers/s390/crypto/zcrypt_api.c b/drivers/s390/crypto/zcrypt_api.c
new file mode 100644
index 000000000000..b3fe003b3d2d
--- /dev/null
+++ b/drivers/s390/crypto/zcrypt_api.c
@@ -0,0 +1,981 @@
+/*
+ *  linux/drivers/s390/crypto/zcrypt_api.c
+ *
+ *  zcrypt 2.0.0
+ *
+ *  Copyright (C)  2001, 2006 IBM Corporation
+ *  Author(s): Robert Burroughs
+ *	       Eric Rossman (edrossma@us.ibm.com)
+ *	       Cornelia Huck <cornelia.huck@de.ibm.com>
+ *
+ *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
+ *  Major cleanup & driver split: Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *				  Ralph Wuerthner <rwuerthn@de.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/miscdevice.h>
+#include <linux/fs.h>
+#include <linux/proc_fs.h>
+#include <linux/compat.h>
+#include <asm/atomic.h>
+#include <asm/uaccess.h>
+
+#include "zcrypt_api.h"
+
+/**
+ * Module description.
+ */
+MODULE_AUTHOR("IBM Corporation");
+MODULE_DESCRIPTION("Cryptographic Coprocessor interface, "
+		   "Copyright 2001, 2006 IBM Corporation");
+MODULE_LICENSE("GPL");
+
+static DEFINE_SPINLOCK(zcrypt_device_lock);
+static LIST_HEAD(zcrypt_device_list);
+static int zcrypt_device_count = 0;
+static atomic_t zcrypt_open_count = ATOMIC_INIT(0);
+
+/**
+ * Device attributes common for all crypto devices.
+ */
+static ssize_t zcrypt_type_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct zcrypt_device *zdev = to_ap_dev(dev)->private;
+	return snprintf(buf, PAGE_SIZE, "%s\n", zdev->type_string);
+}
+
+static DEVICE_ATTR(type, 0444, zcrypt_type_show, NULL);
+
+static ssize_t zcrypt_online_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct zcrypt_device *zdev = to_ap_dev(dev)->private;
+	return snprintf(buf, PAGE_SIZE, "%d\n", zdev->online);
+}
+
+static ssize_t zcrypt_online_store(struct device *dev,
+				   struct device_attribute *attr,
+				   const char *buf, size_t count)
+{
+	struct zcrypt_device *zdev = to_ap_dev(dev)->private;
+	int online;
+
+	if (sscanf(buf, "%d\n", &online) != 1 || online < 0 || online > 1)
+		return -EINVAL;
+	zdev->online = online;
+	if (!online)
+		ap_flush_queue(zdev->ap_dev);
+	return count;
+}
+
+static DEVICE_ATTR(online, 0644, zcrypt_online_show, zcrypt_online_store);
+
+static struct attribute * zcrypt_device_attrs[] = {
+	&dev_attr_type.attr,
+	&dev_attr_online.attr,
+	NULL,
+};
+
+static struct attribute_group zcrypt_device_attr_group = {
+	.attrs = zcrypt_device_attrs,
+};
+
+/**
+ * Move the device towards the head of the device list.
+ * Need to be called while holding the zcrypt device list lock.
+ * Note: cards with speed_rating of 0 are kept at the end of the list.
+ */
+static void __zcrypt_increase_preference(struct zcrypt_device *zdev)
+{
+	struct zcrypt_device *tmp;
+	struct list_head *l;
+
+	if (zdev->speed_rating == 0)
+		return;
+	for (l = zdev->list.prev; l != &zcrypt_device_list; l = l->prev) {
+		tmp = list_entry(l, struct zcrypt_device, list);
+		if ((tmp->request_count + 1) * tmp->speed_rating <=
+		    (zdev->request_count + 1) * zdev->speed_rating &&
+		    tmp->speed_rating != 0)
+			break;
+	}
+	if (l == zdev->list.prev)
+		return;
+	/* Move zdev behind l */
+	list_del(&zdev->list);
+	list_add(&zdev->list, l);
+}
+
+/**
+ * Move the device towards the tail of the device list.
+ * Need to be called while holding the zcrypt device list lock.
+ * Note: cards with speed_rating of 0 are kept at the end of the list.
+ */
+static void __zcrypt_decrease_preference(struct zcrypt_device *zdev)
+{
+	struct zcrypt_device *tmp;
+	struct list_head *l;
+
+	if (zdev->speed_rating == 0)
+		return;
+	for (l = zdev->list.next; l != &zcrypt_device_list; l = l->next) {
+		tmp = list_entry(l, struct zcrypt_device, list);
+		if ((tmp->request_count + 1) * tmp->speed_rating >
+		    (zdev->request_count + 1) * zdev->speed_rating ||
+		    tmp->speed_rating == 0)
+			break;
+	}
+	if (l == zdev->list.next)
+		return;
+	/* Move zdev before l */
+	list_del(&zdev->list);
+	list_add_tail(&zdev->list, l);
+}
+
+static void zcrypt_device_release(struct kref *kref)
+{
+	struct zcrypt_device *zdev =
+		container_of(kref, struct zcrypt_device, refcount);
+	zcrypt_device_free(zdev);
+}
+
+void zcrypt_device_get(struct zcrypt_device *zdev)
+{
+	kref_get(&zdev->refcount);
+}
+EXPORT_SYMBOL(zcrypt_device_get);
+
+int zcrypt_device_put(struct zcrypt_device *zdev)
+{
+	return kref_put(&zdev->refcount, zcrypt_device_release);
+}
+EXPORT_SYMBOL(zcrypt_device_put);
+
+struct zcrypt_device *zcrypt_device_alloc(size_t max_response_size)
+{
+	struct zcrypt_device *zdev;
+
+	zdev = kzalloc(sizeof(struct zcrypt_device), GFP_KERNEL);
+	if (!zdev)
+		return NULL;
+	zdev->reply.message = kmalloc(max_response_size, GFP_KERNEL);
+	if (!zdev->reply.message)
+		goto out_free;
+	zdev->reply.length = max_response_size;
+	spin_lock_init(&zdev->lock);
+	INIT_LIST_HEAD(&zdev->list);
+	return zdev;
+
+out_free:
+	kfree(zdev);
+	return NULL;
+}
+EXPORT_SYMBOL(zcrypt_device_alloc);
+
+void zcrypt_device_free(struct zcrypt_device *zdev)
+{
+	kfree(zdev->reply.message);
+	kfree(zdev);
+}
+EXPORT_SYMBOL(zcrypt_device_free);
+
+/**
+ * Register a crypto device.
+ */
+int zcrypt_device_register(struct zcrypt_device *zdev)
+{
+	int rc;
+
+	rc = sysfs_create_group(&zdev->ap_dev->device.kobj,
+				&zcrypt_device_attr_group);
+	if (rc)
+		goto out;
+	get_device(&zdev->ap_dev->device);
+	kref_init(&zdev->refcount);
+	spin_lock_bh(&zcrypt_device_lock);
+	zdev->online = 1;	/* New devices are online by default. */
+	list_add_tail(&zdev->list, &zcrypt_device_list);
+	__zcrypt_increase_preference(zdev);
+	zcrypt_device_count++;
+	spin_unlock_bh(&zcrypt_device_lock);
+out:
+	return rc;
+}
+EXPORT_SYMBOL(zcrypt_device_register);
+
+/**
+ * Unregister a crypto device.
+ */
+void zcrypt_device_unregister(struct zcrypt_device *zdev)
+{
+	spin_lock_bh(&zcrypt_device_lock);
+	zcrypt_device_count--;
+	list_del_init(&zdev->list);
+	spin_unlock_bh(&zcrypt_device_lock);
+	sysfs_remove_group(&zdev->ap_dev->device.kobj,
+			   &zcrypt_device_attr_group);
+	put_device(&zdev->ap_dev->device);
+	zcrypt_device_put(zdev);
+}
+EXPORT_SYMBOL(zcrypt_device_unregister);
+
+/**
+ * zcrypt_read is not be supported beyond zcrypt 1.3.1
+ */
+static ssize_t zcrypt_read(struct file *filp, char __user *buf,
+			   size_t count, loff_t *f_pos)
+{
+	return -EPERM;
+}
+
+/**
+ * Write is is not allowed
+ */
+static ssize_t zcrypt_write(struct file *filp, const char __user *buf,
+			    size_t count, loff_t *f_pos)
+{
+	return -EPERM;
+}
+
+/**
+ * Device open/close functions to count number of users.
+ */
+static int zcrypt_open(struct inode *inode, struct file *filp)
+{
+	atomic_inc(&zcrypt_open_count);
+	return 0;
+}
+
+static int zcrypt_release(struct inode *inode, struct file *filp)
+{
+	atomic_dec(&zcrypt_open_count);
+	return 0;
+}
+
+/**
+ * zcrypt ioctls.
+ */
+static long zcrypt_rsa_modexpo(struct ica_rsa_modexpo *mex)
+{
+	struct zcrypt_device *zdev;
+	int rc;
+
+	if (mex->outputdatalength < mex->inputdatalength)
+		return -EINVAL;
+	/**
+	 * As long as outputdatalength is big enough, we can set the
+	 * outputdatalength equal to the inputdatalength, since that is the
+	 * number of bytes we will copy in any case
+	 */
+	mex->outputdatalength = mex->inputdatalength;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		if (!zdev->online ||
+		    !zdev->ops->rsa_modexpo ||
+		    zdev->min_mod_size > mex->inputdatalength ||
+		    zdev->max_mod_size < mex->inputdatalength)
+			continue;
+		zcrypt_device_get(zdev);
+		get_device(&zdev->ap_dev->device);
+		zdev->request_count++;
+		__zcrypt_decrease_preference(zdev);
+		spin_unlock_bh(&zcrypt_device_lock);
+		if (try_module_get(zdev->ap_dev->drv->driver.owner)) {
+			rc = zdev->ops->rsa_modexpo(zdev, mex);
+			module_put(zdev->ap_dev->drv->driver.owner);
+		}
+		else
+			rc = -EAGAIN;
+		spin_lock_bh(&zcrypt_device_lock);
+		zdev->request_count--;
+		__zcrypt_increase_preference(zdev);
+		put_device(&zdev->ap_dev->device);
+		zcrypt_device_put(zdev);
+		spin_unlock_bh(&zcrypt_device_lock);
+		return rc;
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+	return -ENODEV;
+}
+
+static long zcrypt_rsa_crt(struct ica_rsa_modexpo_crt *crt)
+{
+	struct zcrypt_device *zdev;
+	unsigned long long z1, z2, z3;
+	int rc, copied;
+
+	if (crt->outputdatalength < crt->inputdatalength ||
+	    (crt->inputdatalength & 1))
+		return -EINVAL;
+	/**
+	 * As long as outputdatalength is big enough, we can set the
+	 * outputdatalength equal to the inputdatalength, since that is the
+	 * number of bytes we will copy in any case
+	 */
+	crt->outputdatalength = crt->inputdatalength;
+
+	copied = 0;
+ restart:
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		if (!zdev->online ||
+		    !zdev->ops->rsa_modexpo_crt ||
+		    zdev->min_mod_size > crt->inputdatalength ||
+		    zdev->max_mod_size < crt->inputdatalength)
+			continue;
+		if (zdev->short_crt && crt->inputdatalength > 240) {
+			/**
+			 * Check inputdata for leading zeros for cards
+			 * that can't handle np_prime, bp_key, or
+			 * u_mult_inv > 128 bytes.
+			 */
+			if (copied == 0) {
+				int len;
+				spin_unlock_bh(&zcrypt_device_lock);
+				/* len is max 256 / 2 - 120 = 8 */
+				len = crt->inputdatalength / 2 - 120;
+				z1 = z2 = z3 = 0;
+				if (copy_from_user(&z1, crt->np_prime, len) ||
+				    copy_from_user(&z2, crt->bp_key, len) ||
+				    copy_from_user(&z3, crt->u_mult_inv, len))
+					return -EFAULT;
+				copied = 1;
+				/**
+				 * We have to restart device lookup -
+				 * the device list may have changed by now.
+				 */
+				goto restart;
+			}
+			if (z1 != 0ULL || z2 != 0ULL || z3 != 0ULL)
+				/* The device can't handle this request. */
+				continue;
+		}
+		zcrypt_device_get(zdev);
+		get_device(&zdev->ap_dev->device);
+		zdev->request_count++;
+		__zcrypt_decrease_preference(zdev);
+		spin_unlock_bh(&zcrypt_device_lock);
+		if (try_module_get(zdev->ap_dev->drv->driver.owner)) {
+			rc = zdev->ops->rsa_modexpo_crt(zdev, crt);
+			module_put(zdev->ap_dev->drv->driver.owner);
+		}
+		else
+			rc = -EAGAIN;
+		spin_lock_bh(&zcrypt_device_lock);
+		zdev->request_count--;
+		__zcrypt_increase_preference(zdev);
+		put_device(&zdev->ap_dev->device);
+		zcrypt_device_put(zdev);
+		spin_unlock_bh(&zcrypt_device_lock);
+		return rc;
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+	return -ENODEV;
+}
+
+static void zcrypt_status_mask(char status[AP_DEVICES])
+{
+	struct zcrypt_device *zdev;
+
+	memset(status, 0, sizeof(char) * AP_DEVICES);
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list)
+		status[AP_QID_DEVICE(zdev->ap_dev->qid)] =
+			zdev->online ? zdev->user_space_type : 0x0d;
+	spin_unlock_bh(&zcrypt_device_lock);
+}
+
+static void zcrypt_qdepth_mask(char qdepth[AP_DEVICES])
+{
+	struct zcrypt_device *zdev;
+
+	memset(qdepth, 0, sizeof(char)	* AP_DEVICES);
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		spin_lock(&zdev->ap_dev->lock);
+		qdepth[AP_QID_DEVICE(zdev->ap_dev->qid)] =
+			zdev->ap_dev->pendingq_count +
+			zdev->ap_dev->requestq_count;
+		spin_unlock(&zdev->ap_dev->lock);
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+}
+
+static void zcrypt_perdev_reqcnt(int reqcnt[AP_DEVICES])
+{
+	struct zcrypt_device *zdev;
+
+	memset(reqcnt, 0, sizeof(int) * AP_DEVICES);
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		spin_lock(&zdev->ap_dev->lock);
+		reqcnt[AP_QID_DEVICE(zdev->ap_dev->qid)] =
+			zdev->ap_dev->total_request_count;
+		spin_unlock(&zdev->ap_dev->lock);
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+}
+
+static int zcrypt_pendingq_count(void)
+{
+	struct zcrypt_device *zdev;
+	int pendingq_count = 0;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		spin_lock(&zdev->ap_dev->lock);
+		pendingq_count += zdev->ap_dev->pendingq_count;
+		spin_unlock(&zdev->ap_dev->lock);
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+	return pendingq_count;
+}
+
+static int zcrypt_requestq_count(void)
+{
+	struct zcrypt_device *zdev;
+	int requestq_count = 0;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		spin_lock(&zdev->ap_dev->lock);
+		requestq_count += zdev->ap_dev->requestq_count;
+		spin_unlock(&zdev->ap_dev->lock);
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+	return requestq_count;
+}
+
+static int zcrypt_count_type(int type)
+{
+	struct zcrypt_device *zdev;
+	int device_count = 0;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list)
+		if (zdev->user_space_type == type)
+			device_count++;
+	spin_unlock_bh(&zcrypt_device_lock);
+	return device_count;
+}
+
+/**
+ * Old, deprecated combi status call.
+ */
+static long zcrypt_ica_status(struct file *filp, unsigned long arg)
+{
+	struct ica_z90_status *pstat;
+	int ret;
+
+	pstat = kzalloc(sizeof(*pstat), GFP_KERNEL);
+	if (!pstat)
+		return -ENOMEM;
+	pstat->totalcount = zcrypt_device_count;
+	pstat->leedslitecount = zcrypt_count_type(ZCRYPT_PCICA);
+	pstat->leeds2count = zcrypt_count_type(ZCRYPT_PCICC);
+	pstat->requestqWaitCount = zcrypt_requestq_count();
+	pstat->pendingqWaitCount = zcrypt_pendingq_count();
+	pstat->totalOpenCount = atomic_read(&zcrypt_open_count);
+	pstat->cryptoDomain = ap_domain_index;
+	zcrypt_status_mask(pstat->status);
+	zcrypt_qdepth_mask(pstat->qdepth);
+	ret = 0;
+	if (copy_to_user((void __user *) arg, pstat, sizeof(*pstat)))
+		ret = -EFAULT;
+	kfree(pstat);
+	return ret;
+}
+
+static long zcrypt_unlocked_ioctl(struct file *filp, unsigned int cmd,
+				  unsigned long arg)
+{
+	int rc;
+
+	switch (cmd) {
+	case ICARSAMODEXPO: {
+		struct ica_rsa_modexpo __user *umex = (void __user *) arg;
+		struct ica_rsa_modexpo mex;
+		if (copy_from_user(&mex, umex, sizeof(mex)))
+			return -EFAULT;
+		do {
+			rc = zcrypt_rsa_modexpo(&mex);
+		} while (rc == -EAGAIN);
+		if (rc)
+			return rc;
+		return put_user(mex.outputdatalength, &umex->outputdatalength);
+	}
+	case ICARSACRT: {
+		struct ica_rsa_modexpo_crt __user *ucrt = (void __user *) arg;
+		struct ica_rsa_modexpo_crt crt;
+		if (copy_from_user(&crt, ucrt, sizeof(crt)))
+			return -EFAULT;
+		do {
+			rc = zcrypt_rsa_crt(&crt);
+		} while (rc == -EAGAIN);
+		if (rc)
+			return rc;
+		return put_user(crt.outputdatalength, &ucrt->outputdatalength);
+	}
+	case Z90STAT_STATUS_MASK: {
+		char status[AP_DEVICES];
+		zcrypt_status_mask(status);
+		if (copy_to_user((char __user *) arg, status,
+				 sizeof(char) * AP_DEVICES))
+			return -EFAULT;
+		return 0;
+	}
+	case Z90STAT_QDEPTH_MASK: {
+		char qdepth[AP_DEVICES];
+		zcrypt_qdepth_mask(qdepth);
+		if (copy_to_user((char __user *) arg, qdepth,
+				 sizeof(char) * AP_DEVICES))
+			return -EFAULT;
+		return 0;
+	}
+	case Z90STAT_PERDEV_REQCNT: {
+		int reqcnt[AP_DEVICES];
+		zcrypt_perdev_reqcnt(reqcnt);
+		if (copy_to_user((int __user *) arg, reqcnt,
+				 sizeof(int) * AP_DEVICES))
+			return -EFAULT;
+		return 0;
+	}
+	case Z90STAT_REQUESTQ_COUNT:
+		return put_user(zcrypt_requestq_count(), (int __user *) arg);
+	case Z90STAT_PENDINGQ_COUNT:
+		return put_user(zcrypt_pendingq_count(), (int __user *) arg);
+	case Z90STAT_TOTALOPEN_COUNT:
+		return put_user(atomic_read(&zcrypt_open_count),
+				(int __user *) arg);
+	case Z90STAT_DOMAIN_INDEX:
+		return put_user(ap_domain_index, (int __user *) arg);
+	/**
+	 * Deprecated ioctls. Don't add another device count ioctl,
+	 * you can count them yourself in the user space with the
+	 * output of the Z90STAT_STATUS_MASK ioctl.
+	 */
+	case ICAZ90STATUS:
+		return zcrypt_ica_status(filp, arg);
+	case Z90STAT_TOTALCOUNT:
+		return put_user(zcrypt_device_count, (int __user *) arg);
+	case Z90STAT_PCICACOUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_PCICA),
+				(int __user *) arg);
+	case Z90STAT_PCICCCOUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_PCICC),
+				(int __user *) arg);
+	case Z90STAT_PCIXCCMCL2COUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_PCIXCC_MCL2),
+				(int __user *) arg);
+	case Z90STAT_PCIXCCMCL3COUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_PCIXCC_MCL3),
+				(int __user *) arg);
+	case Z90STAT_PCIXCCCOUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_PCIXCC_MCL2) +
+				zcrypt_count_type(ZCRYPT_PCIXCC_MCL3),
+				(int __user *) arg);
+	case Z90STAT_CEX2CCOUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_CEX2C),
+				(int __user *) arg);
+	case Z90STAT_CEX2ACOUNT:
+		return put_user(zcrypt_count_type(ZCRYPT_CEX2A),
+				(int __user *) arg);
+	default:
+		/* unknown ioctl number */
+		return -ENOIOCTLCMD;
+	}
+}
+
+#ifdef CONFIG_COMPAT
+/**
+ * ioctl32 conversion routines
+ */
+struct compat_ica_rsa_modexpo {
+	compat_uptr_t	inputdata;
+	unsigned int	inputdatalength;
+	compat_uptr_t	outputdata;
+	unsigned int	outputdatalength;
+	compat_uptr_t	b_key;
+	compat_uptr_t	n_modulus;
+};
+
+static long trans_modexpo32(struct file *filp, unsigned int cmd,
+			    unsigned long arg)
+{
+	struct compat_ica_rsa_modexpo __user *umex32 = compat_ptr(arg);
+	struct compat_ica_rsa_modexpo mex32;
+	struct ica_rsa_modexpo mex64;
+	long rc;
+
+	if (copy_from_user(&mex32, umex32, sizeof(mex32)))
+		return -EFAULT;
+	mex64.inputdata = compat_ptr(mex32.inputdata);
+	mex64.inputdatalength = mex32.inputdatalength;
+	mex64.outputdata = compat_ptr(mex32.outputdata);
+	mex64.outputdatalength = mex32.outputdatalength;
+	mex64.b_key = compat_ptr(mex32.b_key);
+	mex64.n_modulus = compat_ptr(mex32.n_modulus);
+	do {
+		rc = zcrypt_rsa_modexpo(&mex64);
+	} while (rc == -EAGAIN);
+	if (!rc)
+		rc = put_user(mex64.outputdatalength,
+			      &umex32->outputdatalength);
+	return rc;
+}
+
+struct compat_ica_rsa_modexpo_crt {
+	compat_uptr_t	inputdata;
+	unsigned int	inputdatalength;
+	compat_uptr_t	outputdata;
+	unsigned int	outputdatalength;
+	compat_uptr_t	bp_key;
+	compat_uptr_t	bq_key;
+	compat_uptr_t	np_prime;
+	compat_uptr_t	nq_prime;
+	compat_uptr_t	u_mult_inv;
+};
+
+static long trans_modexpo_crt32(struct file *filp, unsigned int cmd,
+				unsigned long arg)
+{
+	struct compat_ica_rsa_modexpo_crt __user *ucrt32 = compat_ptr(arg);
+	struct compat_ica_rsa_modexpo_crt crt32;
+	struct ica_rsa_modexpo_crt crt64;
+	long rc;
+
+	if (copy_from_user(&crt32, ucrt32, sizeof(crt32)))
+		return -EFAULT;
+	crt64.inputdata = compat_ptr(crt32.inputdata);
+	crt64.inputdatalength = crt32.inputdatalength;
+	crt64.outputdata=  compat_ptr(crt32.outputdata);
+	crt64.outputdatalength = crt32.outputdatalength;
+	crt64.bp_key = compat_ptr(crt32.bp_key);
+	crt64.bq_key = compat_ptr(crt32.bq_key);
+	crt64.np_prime = compat_ptr(crt32.np_prime);
+	crt64.nq_prime = compat_ptr(crt32.nq_prime);
+	crt64.u_mult_inv = compat_ptr(crt32.u_mult_inv);
+	do {
+		rc = zcrypt_rsa_crt(&crt64);
+	} while (rc == -EAGAIN);
+	if (!rc)
+		rc = put_user(crt64.outputdatalength,
+			      &ucrt32->outputdatalength);
+	return rc;
+}
+
+long zcrypt_compat_ioctl(struct file *filp, unsigned int cmd,
+			 unsigned long arg)
+{
+	if (cmd == ICARSAMODEXPO)
+		return trans_modexpo32(filp, cmd, arg);
+	if (cmd == ICARSACRT)
+		return trans_modexpo_crt32(filp, cmd, arg);
+	return zcrypt_unlocked_ioctl(filp, cmd, arg);
+}
+#endif
+
+/**
+ * Misc device file operations.
+ */
+static struct file_operations zcrypt_fops = {
+	.owner		= THIS_MODULE,
+	.read		= zcrypt_read,
+	.write		= zcrypt_write,
+	.unlocked_ioctl	= zcrypt_unlocked_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= zcrypt_compat_ioctl,
+#endif
+	.open		= zcrypt_open,
+	.release	= zcrypt_release
+};
+
+/**
+ * Misc device.
+ */
+static struct miscdevice zcrypt_misc_device = {
+	.minor	    = MISC_DYNAMIC_MINOR,
+	.name	    = "z90crypt",
+	.fops	    = &zcrypt_fops,
+};
+
+/**
+ * Deprecated /proc entry support.
+ */
+static struct proc_dir_entry *zcrypt_entry;
+
+static inline int sprintcl(unsigned char *outaddr, unsigned char *addr,
+			   unsigned int len)
+{
+	int hl, i;
+
+	hl = 0;
+	for (i = 0; i < len; i++)
+		hl += sprintf(outaddr+hl, "%01x", (unsigned int) addr[i]);
+	hl += sprintf(outaddr+hl, " ");
+	return hl;
+}
+
+static inline int sprintrw(unsigned char *outaddr, unsigned char *addr,
+			   unsigned int len)
+{
+	int hl, inl, c, cx;
+
+	hl = sprintf(outaddr, "	   ");
+	inl = 0;
+	for (c = 0; c < (len / 16); c++) {
+		hl += sprintcl(outaddr+hl, addr+inl, 16);
+		inl += 16;
+	}
+	cx = len%16;
+	if (cx) {
+		hl += sprintcl(outaddr+hl, addr+inl, cx);
+		inl += cx;
+	}
+	hl += sprintf(outaddr+hl, "\n");
+	return hl;
+}
+
+static inline int sprinthx(unsigned char *title, unsigned char *outaddr,
+			   unsigned char *addr, unsigned int len)
+{
+	int hl, inl, r, rx;
+
+	hl = sprintf(outaddr, "\n%s\n", title);
+	inl = 0;
+	for (r = 0; r < (len / 64); r++) {
+		hl += sprintrw(outaddr+hl, addr+inl, 64);
+		inl += 64;
+	}
+	rx = len % 64;
+	if (rx) {
+		hl += sprintrw(outaddr+hl, addr+inl, rx);
+		inl += rx;
+	}
+	hl += sprintf(outaddr+hl, "\n");
+	return hl;
+}
+
+static inline int sprinthx4(unsigned char *title, unsigned char *outaddr,
+			    unsigned int *array, unsigned int len)
+{
+	int hl, r;
+
+	hl = sprintf(outaddr, "\n%s\n", title);
+	for (r = 0; r < len; r++) {
+		if ((r % 8) == 0)
+			hl += sprintf(outaddr+hl, "    ");
+		hl += sprintf(outaddr+hl, "%08X ", array[r]);
+		if ((r % 8) == 7)
+			hl += sprintf(outaddr+hl, "\n");
+	}
+	hl += sprintf(outaddr+hl, "\n");
+	return hl;
+}
+
+static int zcrypt_status_read(char *resp_buff, char **start, off_t offset,
+			      int count, int *eof, void *data)
+{
+	unsigned char *workarea;
+	int len;
+
+	len = 0;
+
+	/* resp_buff is a page. Use the right half for a work area */
+	workarea = resp_buff + 2000;
+	len += sprintf(resp_buff + len, "\nzcrypt version: %d.%d.%d\n",
+		ZCRYPT_VERSION, ZCRYPT_RELEASE, ZCRYPT_VARIANT);
+	len += sprintf(resp_buff + len, "Cryptographic domain: %d\n",
+		       ap_domain_index);
+	len += sprintf(resp_buff + len, "Total device count: %d\n",
+		       zcrypt_device_count);
+	len += sprintf(resp_buff + len, "PCICA count: %d\n",
+		       zcrypt_count_type(ZCRYPT_PCICA));
+	len += sprintf(resp_buff + len, "PCICC count: %d\n",
+		       zcrypt_count_type(ZCRYPT_PCICC));
+	len += sprintf(resp_buff + len, "PCIXCC MCL2 count: %d\n",
+		       zcrypt_count_type(ZCRYPT_PCIXCC_MCL2));
+	len += sprintf(resp_buff + len, "PCIXCC MCL3 count: %d\n",
+		       zcrypt_count_type(ZCRYPT_PCIXCC_MCL3));
+	len += sprintf(resp_buff + len, "CEX2C count: %d\n",
+		       zcrypt_count_type(ZCRYPT_CEX2C));
+	len += sprintf(resp_buff + len, "CEX2A count: %d\n",
+		       zcrypt_count_type(ZCRYPT_CEX2A));
+	len += sprintf(resp_buff + len, "requestq count: %d\n",
+		       zcrypt_requestq_count());
+	len += sprintf(resp_buff + len, "pendingq count: %d\n",
+		       zcrypt_pendingq_count());
+	len += sprintf(resp_buff + len, "Total open handles: %d\n\n",
+		       atomic_read(&zcrypt_open_count));
+	zcrypt_status_mask(workarea);
+	len += sprinthx("Online devices: 1=PCICA 2=PCICC 3=PCIXCC(MCL2) "
+			"4=PCIXCC(MCL3) 5=CEX2C 6=CEX2A",
+			resp_buff+len, workarea, AP_DEVICES);
+	zcrypt_qdepth_mask(workarea);
+	len += sprinthx("Waiting work element counts",
+			resp_buff+len, workarea, AP_DEVICES);
+	zcrypt_perdev_reqcnt((unsigned int *) workarea);
+	len += sprinthx4("Per-device successfully completed request counts",
+			 resp_buff+len,(unsigned int *) workarea, AP_DEVICES);
+	*eof = 1;
+	memset((void *) workarea, 0x00, AP_DEVICES * sizeof(unsigned int));
+	return len;
+}
+
+static void zcrypt_disable_card(int index)
+{
+	struct zcrypt_device *zdev;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list)
+		if (AP_QID_DEVICE(zdev->ap_dev->qid) == index) {
+			zdev->online = 0;
+			ap_flush_queue(zdev->ap_dev);
+			break;
+		}
+	spin_unlock_bh(&zcrypt_device_lock);
+}
+
+static void zcrypt_enable_card(int index)
+{
+	struct zcrypt_device *zdev;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list)
+		if (AP_QID_DEVICE(zdev->ap_dev->qid) == index) {
+			zdev->online = 1;
+			break;
+		}
+	spin_unlock_bh(&zcrypt_device_lock);
+}
+
+static int zcrypt_status_write(struct file *file, const char __user *buffer,
+			       unsigned long count, void *data)
+{
+	unsigned char *lbuf, *ptr;
+	unsigned long local_count;
+	int j;
+
+	if (count <= 0)
+		return 0;
+
+#define LBUFSIZE 1200UL
+	lbuf = kmalloc(LBUFSIZE, GFP_KERNEL);
+	if (!lbuf) {
+		PRINTK("kmalloc failed!\n");
+		return 0;
+	}
+
+	local_count = min(LBUFSIZE - 1, count);
+	if (copy_from_user(lbuf, buffer, local_count) != 0) {
+		kfree(lbuf);
+		return -EFAULT;
+	}
+	lbuf[local_count] = '\0';
+
+	ptr = strstr(lbuf, "Online devices");
+	if (!ptr) {
+		PRINTK("Unable to parse data (missing \"Online devices\")\n");
+		goto out;
+	}
+	ptr = strstr(ptr, "\n");
+	if (!ptr) {
+		PRINTK("Unable to parse data (missing newline "
+		       "after \"Online devices\")\n");
+		goto out;
+	}
+	ptr++;
+
+	if (strstr(ptr, "Waiting work element counts") == NULL) {
+		PRINTK("Unable to parse data (missing "
+		       "\"Waiting work element counts\")\n");
+		goto out;
+	}
+
+	for (j = 0; j < 64 && *ptr; ptr++) {
+		/**
+		 * '0' for no device, '1' for PCICA, '2' for PCICC,
+		 * '3' for PCIXCC_MCL2, '4' for PCIXCC_MCL3,
+		 * '5' for CEX2C and '6' for CEX2A'
+		 */
+		if (*ptr >= '0' && *ptr <= '6')
+			j++;
+		else if (*ptr == 'd' || *ptr == 'D')
+			zcrypt_disable_card(j++);
+		else if (*ptr == 'e' || *ptr == 'E')
+			zcrypt_enable_card(j++);
+		else if (*ptr != ' ' && *ptr != '\t')
+			break;
+	}
+out:
+	kfree(lbuf);
+	return count;
+}
+
+/**
+ * The module initialization code.
+ */
+int __init zcrypt_api_init(void)
+{
+	int rc;
+
+	/* Register the request sprayer. */
+	rc = misc_register(&zcrypt_misc_device);
+	if (rc < 0) {
+		PRINTKW(KERN_ERR "misc_register (minor %d) failed with %d\n",
+			zcrypt_misc_device.minor, rc);
+		goto out;
+	}
+
+	/* Set up the proc file system */
+	zcrypt_entry = create_proc_entry("driver/z90crypt", 0644, NULL);
+	if (!zcrypt_entry) {
+		PRINTK("Couldn't create z90crypt proc entry\n");
+		rc = -ENOMEM;
+		goto out_misc;
+	}
+	zcrypt_entry->nlink = 1;
+	zcrypt_entry->data = NULL;
+	zcrypt_entry->read_proc = zcrypt_status_read;
+	zcrypt_entry->write_proc = zcrypt_status_write;
+
+	return 0;
+
+out_misc:
+	misc_deregister(&zcrypt_misc_device);
+out:
+	return rc;
+}
+
+/**
+ * The module termination code.
+ */
+void zcrypt_api_exit(void)
+{
+	remove_proc_entry("driver/z90crypt", NULL);
+	misc_deregister(&zcrypt_misc_device);
+}
+
+#ifndef CONFIG_ZCRYPT_MONOLITHIC
+module_init(zcrypt_api_init);
+module_exit(zcrypt_api_exit);
+#endif
diff --git a/drivers/s390/crypto/zcrypt_api.h b/drivers/s390/crypto/zcrypt_api.h
new file mode 100644
index 000000000000..1f0e61f2e9b4
--- /dev/null
+++ b/drivers/s390/crypto/zcrypt_api.h
@@ -0,0 +1,140 @@
+/*
+ *  linux/drivers/s390/crypto/zcrypt_api.h
+ *
+ *  zcrypt 2.0.0
+ *
+ *  Copyright (C)  2001, 2006 IBM Corporation
+ *  Author(s): Robert Burroughs
+ *	       Eric Rossman (edrossma@us.ibm.com)
+ *	       Cornelia Huck <cornelia.huck@de.ibm.com>
+ *
+ *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
+ *  Major cleanup & driver split: Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *				  Ralph Wuerthner <rwuerthn@de.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _ZCRYPT_API_H_
+#define _ZCRYPT_API_H_
+
+/**
+ * Macro definitions
+ *
+ * PDEBUG debugs in the form "zcrypt: function_name -> message"
+ *
+ * PRINTK is like PDEBUG, except that it is always enabled
+ * PRINTKN is like PRINTK, except that it does not include the function name
+ * PRINTKW is like PRINTK, except that it uses KERN_WARNING
+ * PRINTKC is like PRINTK, except that it uses KERN_CRIT
+ */
+#define DEV_NAME	"zcrypt"
+
+#define PRINTK(fmt, args...) \
+	printk(KERN_DEBUG DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
+#define PRINTKN(fmt, args...) \
+	printk(KERN_DEBUG DEV_NAME ": " fmt, ## args)
+#define PRINTKW(fmt, args...) \
+	printk(KERN_WARNING DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
+#define PRINTKC(fmt, args...) \
+	printk(KERN_CRIT DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
+
+#ifdef ZCRYPT_DEBUG
+#define PDEBUG(fmt, args...) \
+	printk(KERN_DEBUG DEV_NAME ": %s -> " fmt, __FUNCTION__ , ## args)
+#else
+#define PDEBUG(fmt, args...) do {} while (0)
+#endif
+
+#include "ap_bus.h"
+#include <asm/zcrypt.h>
+
+/* deprecated status calls */
+#define ICAZ90STATUS		_IOR(ZCRYPT_IOCTL_MAGIC, 0x10, struct ica_z90_status)
+#define Z90STAT_PCIXCCCOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x43, int)
+
+/**
+ * This structure is deprecated and the corresponding ioctl() has been
+ * replaced with individual ioctl()s for each piece of data!
+ */
+struct ica_z90_status {
+	int totalcount;
+	int leedslitecount; // PCICA
+	int leeds2count;    // PCICC
+	// int PCIXCCCount; is not in struct for backward compatibility
+	int requestqWaitCount;
+	int pendingqWaitCount;
+	int totalOpenCount;
+	int cryptoDomain;
+	// status: 0=not there, 1=PCICA, 2=PCICC, 3=PCIXCC_MCL2, 4=PCIXCC_MCL3,
+	//	   5=CEX2C
+	unsigned char status[64];
+	// qdepth: # work elements waiting for each device
+	unsigned char qdepth[64];
+};
+
+/**
+ * device type for an actual device is either PCICA, PCICC, PCIXCC_MCL2,
+ * PCIXCC_MCL3, CEX2C, or CEX2A
+ *
+ * NOTE: PCIXCC_MCL3 refers to a PCIXCC with May 2004 version of Licensed
+ *	 Internal Code (LIC) (EC J12220 level 29).
+ *	 PCIXCC_MCL2 refers to any LIC before this level.
+ */
+#define ZCRYPT_PCICA		1
+#define ZCRYPT_PCICC		2
+#define ZCRYPT_PCIXCC_MCL2	3
+#define ZCRYPT_PCIXCC_MCL3	4
+#define ZCRYPT_CEX2C		5
+#define ZCRYPT_CEX2A		6
+
+struct zcrypt_device;
+
+struct zcrypt_ops {
+	long (*rsa_modexpo)(struct zcrypt_device *, struct ica_rsa_modexpo *);
+	long (*rsa_modexpo_crt)(struct zcrypt_device *,
+				struct ica_rsa_modexpo_crt *);
+};
+
+struct zcrypt_device {
+	struct list_head list;		/* Device list. */
+	spinlock_t lock;		/* Per device lock. */
+	struct kref refcount;		/* device refcounting */
+	struct ap_device *ap_dev;	/* The "real" ap device. */
+	struct zcrypt_ops *ops;		/* Crypto operations. */
+	int online;			/* User online/offline */
+
+	int user_space_type;		/* User space device id. */
+	char *type_string;		/* User space device name. */
+	int min_mod_size;		/* Min number of bits. */
+	int max_mod_size;		/* Max number of bits. */
+	int short_crt;			/* Card has crt length restriction. */
+	int speed_rating;		/* Speed of the crypto device. */
+
+	int request_count;		/* # current requests. */
+
+	struct ap_message reply;	/* Per-device reply structure. */
+};
+
+struct zcrypt_device *zcrypt_device_alloc(size_t);
+void zcrypt_device_free(struct zcrypt_device *);
+void zcrypt_device_get(struct zcrypt_device *);
+int zcrypt_device_put(struct zcrypt_device *);
+int zcrypt_device_register(struct zcrypt_device *);
+void zcrypt_device_unregister(struct zcrypt_device *);
+int zcrypt_api_init(void);
+void zcrypt_api_exit(void);
+
+#endif /* _ZCRYPT_API_H_ */
diff --git a/include/asm-s390/zcrypt.h b/include/asm-s390/zcrypt.h
new file mode 100644
index 000000000000..0d6a3e2a3349
--- /dev/null
+++ b/include/asm-s390/zcrypt.h
@@ -0,0 +1,207 @@
+/*
+ *  include/asm-s390/zcrypt.h
+ *
+ *  zcrypt 2.0.0 (user-visible header)
+ *
+ *  Copyright (C)  2001, 2006 IBM Corporation
+ *  Author(s): Robert Burroughs
+ *	       Eric Rossman (edrossma@us.ibm.com)
+ *
+ *  Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef __ASM_S390_ZCRYPT_H
+#define __ASM_S390_ZCRYPT_H
+
+#define ZCRYPT_VERSION 2
+#define ZCRYPT_RELEASE 1
+#define ZCRYPT_VARIANT 0
+
+#include <linux/ioctl.h>
+#include <linux/compiler.h>
+
+/**
+ * struct ica_rsa_modexpo
+ *
+ * Requirements:
+ * - outputdatalength is at least as large as inputdatalength.
+ * - All key parts are right justified in their fields, padded on
+ *   the left with zeroes.
+ * - length(b_key) = inputdatalength
+ * - length(n_modulus) = inputdatalength
+ */
+struct ica_rsa_modexpo {
+	char __user *	inputdata;
+	unsigned int	inputdatalength;
+	char __user *	outputdata;
+	unsigned int	outputdatalength;
+	char __user *	b_key;
+	char __user *	n_modulus;
+};
+
+/**
+ * struct ica_rsa_modexpo_crt
+ *
+ * Requirements:
+ * - inputdatalength is even.
+ * - outputdatalength is at least as large as inputdatalength.
+ * - All key parts are right justified in their fields, padded on
+ *   the left with zeroes.
+ * - length(bp_key)	= inputdatalength/2 + 8
+ * - length(bq_key)	= inputdatalength/2
+ * - length(np_key)	= inputdatalength/2 + 8
+ * - length(nq_key)	= inputdatalength/2
+ * - length(u_mult_inv) = inputdatalength/2 + 8
+ */
+struct ica_rsa_modexpo_crt {
+	char __user *	inputdata;
+	unsigned int	inputdatalength;
+	char __user *	outputdata;
+	unsigned int	outputdatalength;
+	char __user *	bp_key;
+	char __user *	bq_key;
+	char __user *	np_prime;
+	char __user *	nq_prime;
+	char __user *	u_mult_inv;
+};
+
+#define ZCRYPT_IOCTL_MAGIC 'z'
+
+/**
+ * Interface notes:
+ *
+ * The ioctl()s which are implemented (along with relevant details)
+ * are:
+ *
+ *   ICARSAMODEXPO
+ *     Perform an RSA operation using a Modulus-Exponent pair
+ *     This takes an ica_rsa_modexpo struct as its arg.
+ *
+ *     NOTE: please refer to the comments preceding this structure
+ *	     for the implementation details for the contents of the
+ *	     block
+ *
+ *   ICARSACRT
+ *     Perform an RSA operation using a Chinese-Remainder Theorem key
+ *     This takes an ica_rsa_modexpo_crt struct as its arg.
+ *
+ *     NOTE: please refer to the comments preceding this structure
+ *	     for the implementation details for the contents of the
+ *	     block
+ *
+ *   Z90STAT_TOTALCOUNT
+ *     Return an integer count of all device types together.
+ *
+ *   Z90STAT_PCICACOUNT
+ *     Return an integer count of all PCICAs.
+ *
+ *   Z90STAT_PCICCCOUNT
+ *     Return an integer count of all PCICCs.
+ *
+ *   Z90STAT_PCIXCCMCL2COUNT
+ *     Return an integer count of all MCL2 PCIXCCs.
+ *
+ *   Z90STAT_PCIXCCMCL3COUNT
+ *     Return an integer count of all MCL3 PCIXCCs.
+ *
+ *   Z90STAT_CEX2CCOUNT
+ *     Return an integer count of all CEX2Cs.
+ *
+ *   Z90STAT_CEX2ACOUNT
+ *     Return an integer count of all CEX2As.
+ *
+ *   Z90STAT_REQUESTQ_COUNT
+ *     Return an integer count of the number of entries waiting to be
+ *     sent to a device.
+ *
+ *   Z90STAT_PENDINGQ_COUNT
+ *     Return an integer count of the number of entries sent to a
+ *     device awaiting the reply.
+ *
+ *   Z90STAT_TOTALOPEN_COUNT
+ *     Return an integer count of the number of open file handles.
+ *
+ *   Z90STAT_DOMAIN_INDEX
+ *     Return the integer value of the Cryptographic Domain.
+ *
+ *   Z90STAT_STATUS_MASK
+ *     Return an 64 element array of unsigned chars for the status of
+ *     all devices.
+ *	 0x01: PCICA
+ *	 0x02: PCICC
+ *	 0x03: PCIXCC_MCL2
+ *	 0x04: PCIXCC_MCL3
+ *	 0x05: CEX2C
+ *	 0x06: CEX2A
+ *	 0x0d: device is disabled via the proc filesystem
+ *
+ *   Z90STAT_QDEPTH_MASK
+ *     Return an 64 element array of unsigned chars for the queue
+ *     depth of all devices.
+ *
+ *   Z90STAT_PERDEV_REQCNT
+ *     Return an 64 element array of unsigned integers for the number
+ *     of successfully completed requests per device since the device
+ *     was detected and made available.
+ *
+ *   ICAZ90STATUS (deprecated)
+ *     Return some device driver status in a ica_z90_status struct
+ *     This takes an ica_z90_status struct as its arg.
+ *
+ *     NOTE: this ioctl() is deprecated, and has been replaced with
+ *	     single ioctl()s for each type of status being requested
+ *
+ *   Z90STAT_PCIXCCCOUNT (deprecated)
+ *     Return an integer count of all PCIXCCs (MCL2 + MCL3).
+ *     This is DEPRECATED now that MCL3 PCIXCCs are treated differently from
+ *     MCL2 PCIXCCs.
+ *
+ *   Z90QUIESCE (not recommended)
+ *     Quiesce the driver.  This is intended to stop all new
+ *     requests from being processed.  Its use is NOT recommended,
+ *     except in circumstances where there is no other way to stop
+ *     callers from accessing the driver.  Its original use was to
+ *     allow the driver to be "drained" of work in preparation for
+ *     a system shutdown.
+ *
+ *     NOTE: once issued, this ban on new work cannot be undone
+ *	     except by unloading and reloading the driver.
+ */
+
+/**
+ * Supported ioctl calls
+ */
+#define ICARSAMODEXPO	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x05, 0)
+#define ICARSACRT	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x06, 0)
+
+/* New status calls */
+#define Z90STAT_TOTALCOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x40, int)
+#define Z90STAT_PCICACOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x41, int)
+#define Z90STAT_PCICCCOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x42, int)
+#define Z90STAT_PCIXCCMCL2COUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x4b, int)
+#define Z90STAT_PCIXCCMCL3COUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x4c, int)
+#define Z90STAT_CEX2CCOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x4d, int)
+#define Z90STAT_CEX2ACOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x4e, int)
+#define Z90STAT_REQUESTQ_COUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x44, int)
+#define Z90STAT_PENDINGQ_COUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x45, int)
+#define Z90STAT_TOTALOPEN_COUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x46, int)
+#define Z90STAT_DOMAIN_INDEX	_IOR(ZCRYPT_IOCTL_MAGIC, 0x47, int)
+#define Z90STAT_STATUS_MASK	_IOR(ZCRYPT_IOCTL_MAGIC, 0x48, char[64])
+#define Z90STAT_QDEPTH_MASK	_IOR(ZCRYPT_IOCTL_MAGIC, 0x49, char[64])
+#define Z90STAT_PERDEV_REQCNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x4a, int[64])
+
+#endif /* __ASM_S390_ZCRYPT_H */
-- 
cgit v1.2.3


From 1b2e2b73b4c84c918686c04a00724197036c0847 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 21 Aug 2006 17:06:38 +0100
Subject: [ARM] Cleanup arch/arm/mm a little

Move top_pmd into arch/arm/mm/mm.h - nothing outside arch/arm/mm
references it.

Move the repeated definition of TOP_PTE into mm/mm.h, as well as
a few function prototypes.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/copypage-v4mc.c   |  4 ++--
 arch/arm/mm/copypage-v6.c     |  4 ++--
 arch/arm/mm/copypage-xscale.c |  4 ++--
 arch/arm/mm/flush.c           |  4 ++--
 arch/arm/mm/init.c            | 20 +++++++-------------
 arch/arm/mm/mm-armv.c         | 14 ++------------
 arch/arm/mm/mm.h              | 19 +++++++++++++++++++
 include/asm-arm/page.h        |  3 ---
 8 files changed, 36 insertions(+), 36 deletions(-)
 create mode 100644 arch/arm/mm/mm.h

(limited to 'include')

diff --git a/arch/arm/mm/copypage-v4mc.c b/arch/arm/mm/copypage-v4mc.c
index fc69dccdace1..df1645e14b4c 100644
--- a/arch/arm/mm/copypage-v4mc.c
+++ b/arch/arm/mm/copypage-v4mc.c
@@ -20,6 +20,8 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 
+#include "mm.h"
+
 /*
  * 0xffff8000 to 0xffffffff is reserved for any ARM architecture
  * specific hacks for copying pages efficiently.
@@ -27,8 +29,6 @@
 #define minicache_pgprot __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | \
 				  L_PTE_CACHEABLE)
 
-#define TOP_PTE(x)	pte_offset_kernel(top_pmd, x)
-
 static DEFINE_SPINLOCK(minicache_lock);
 
 /*
diff --git a/arch/arm/mm/copypage-v6.c b/arch/arm/mm/copypage-v6.c
index 269ce6913ee9..3d0d3a963d20 100644
--- a/arch/arm/mm/copypage-v6.c
+++ b/arch/arm/mm/copypage-v6.c
@@ -17,6 +17,8 @@
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
 
+#include "mm.h"
+
 #if SHMLBA > 16384
 #error FIX ME
 #endif
@@ -24,8 +26,6 @@
 #define from_address	(0xffff8000)
 #define to_address	(0xffffc000)
 
-#define TOP_PTE(x)	pte_offset_kernel(top_pmd, x)
-
 static DEFINE_SPINLOCK(v6_lock);
 
 /*
diff --git a/arch/arm/mm/copypage-xscale.c b/arch/arm/mm/copypage-xscale.c
index 42a6ee255ce0..84ebe0aa379e 100644
--- a/arch/arm/mm/copypage-xscale.c
+++ b/arch/arm/mm/copypage-xscale.c
@@ -20,6 +20,8 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 
+#include "mm.h"
+
 /*
  * 0xffff8000 to 0xffffffff is reserved for any ARM architecture
  * specific hacks for copying pages efficiently.
@@ -29,8 +31,6 @@
 #define minicache_pgprot __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | \
 				  L_PTE_CACHEABLE)
 
-#define TOP_PTE(x)	pte_offset_kernel(top_pmd, x)
-
 static DEFINE_SPINLOCK(minicache_lock);
 
 /*
diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
index d438ce41cdd5..1efb05c64db3 100644
--- a/arch/arm/mm/flush.c
+++ b/arch/arm/mm/flush.c
@@ -15,12 +15,12 @@
 #include <asm/system.h>
 #include <asm/tlbflush.h>
 
+#include "mm.h"
+
 #ifdef CONFIG_CPU_CACHE_VIPT
 
 #define ALIAS_FLUSH_START	0xffff4000
 
-#define TOP_PTE(x)	pte_offset_kernel(top_pmd, x)
-
 static void flush_pfn_alias(unsigned long pfn, unsigned long vaddr)
 {
 	unsigned long to = ALIAS_FLUSH_START + (CACHE_COLOUR(vaddr) << PAGE_SHIFT);
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index fe3f7f625008..1099af6d470a 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -25,6 +25,8 @@
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
 
+#include "mm.h"
+
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
 extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
@@ -44,6 +46,11 @@ static struct meminfo meminfo __initdata = { 0, };
  */
 struct page *empty_zero_page;
 
+/*
+ * The pmd table for the upper-most set of pages.
+ */
+pmd_t *top_pmd;
+
 void show_mem(void)
 {
 	int free = 0, total = 0, reserved = 0;
@@ -83,16 +90,6 @@ void show_mem(void)
 	printk("%d pages swap cached\n", cached);
 }
 
-static inline pmd_t *pmd_off(pgd_t *pgd, unsigned long virt)
-{
-	return pmd_offset(pgd, virt);
-}
-
-static inline pmd_t *pmd_off_k(unsigned long virt)
-{
-	return pmd_off(pgd_offset_k(virt), virt);
-}
-
 #define for_each_nodebank(iter,mi,no)			\
 	for (iter = 0; iter < mi->nr_banks; iter++)	\
 		if (mi->bank[iter].node == no)
@@ -229,9 +226,6 @@ static __init void reserve_node_zero(pg_data_t *pgdat)
 		reserve_bootmem_node(pgdat, PHYS_OFFSET, res_size);
 }
 
-void __init build_mem_type_table(void);
-void __init create_mapping(struct map_desc *md);
-
 static unsigned long __init
 bootmem_init_node(int node, int initrd_node, struct meminfo *mi)
 {
diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c
index 38769f5862bc..ee9647823fad 100644
--- a/arch/arm/mm/mm-armv.c
+++ b/arch/arm/mm/mm-armv.c
@@ -23,6 +23,8 @@
 
 #include <asm/mach/map.h>
 
+#include "mm.h"
+
 #define CPOLICY_UNCACHED	0
 #define CPOLICY_BUFFERED	1
 #define CPOLICY_WRITETHROUGH	2
@@ -35,8 +37,6 @@ pgprot_t pgprot_kernel;
 
 EXPORT_SYMBOL(pgprot_kernel);
 
-pmd_t *top_pmd;
-
 struct cachepolicy {
 	const char	policy[16];
 	unsigned int	cr_mask;
@@ -142,16 +142,6 @@ __setup("noalign", noalign_setup);
 
 #define FIRST_KERNEL_PGD_NR	(FIRST_USER_PGD_NR + USER_PTRS_PER_PGD)
 
-static inline pmd_t *pmd_off(pgd_t *pgd, unsigned long virt)
-{
-	return pmd_offset(pgd, virt);
-}
-
-static inline pmd_t *pmd_off_k(unsigned long virt)
-{
-	return pmd_off(pgd_offset_k(virt), virt);
-}
-
 /*
  * need to get a 16k page for level 1
  */
diff --git a/arch/arm/mm/mm.h b/arch/arm/mm/mm.h
new file mode 100644
index 000000000000..8d73ffbce8df
--- /dev/null
+++ b/arch/arm/mm/mm.h
@@ -0,0 +1,19 @@
+/* the upper-most page table pointer */
+extern pmd_t *top_pmd;
+
+#define TOP_PTE(x)	pte_offset_kernel(top_pmd, x)
+
+static inline pmd_t *pmd_off(pgd_t *pgd, unsigned long virt)
+{
+	return pmd_offset(pgd, virt);
+}
+
+static inline pmd_t *pmd_off_k(unsigned long virt)
+{
+	return pmd_off(pgd_offset_k(virt), virt);
+}
+
+struct map_desc;
+
+void __init build_mem_type_table(void);
+void __init create_mapping(struct map_desc *md);
diff --git a/include/asm-arm/page.h b/include/asm-arm/page.h
index b721270b9986..af9c3fe7588c 100644
--- a/include/asm-arm/page.h
+++ b/include/asm-arm/page.h
@@ -174,9 +174,6 @@ typedef unsigned long pgprot_t;
 
 #endif /* STRICT_MM_TYPECHECKS */
 
-/* the upper-most page table pointer */
-extern pmd_t *top_pmd;
-
 #endif /* CONFIG_MMU */
 
 #include <asm/memory.h>
-- 
cgit v1.2.3


From 7ad1bcb25c5623f1f87c50fdf2272f58ff91db5a Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 27 Aug 2006 12:07:02 +0100
Subject: [ARM] Add ARM irqtrace support

This adds support for irqtrace for lockdep on ARM.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig             |   4 ++
 arch/arm/kernel/entry-armv.S |  13 +++++
 include/asm-arm/irqflags.h   | 132 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/system.h     | 125 +---------------------------------------
 4 files changed, 150 insertions(+), 124 deletions(-)
 create mode 100644 include/asm-arm/irqflags.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index f81a62380add..d9376f048ed7 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -51,6 +51,10 @@ config GENERIC_HARDIRQS
 	bool
 	default y
 
+config TRACE_IRQFLAGS_SUPPORT
+	bool
+	default y
+
 config HARDIRQS_SW_RESEND
 	bool
 	default y
diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index de4e33137901..bd623b73445f 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -191,6 +191,9 @@ __dabt_svc:
 __irq_svc:
 	svc_entry
 
+#ifdef CONFIG_TRACE_IRQFLAGS
+	bl	trace_hardirqs_off
+#endif
 #ifdef CONFIG_PREEMPT
 	get_thread_info tsk
 	ldr	r8, [tsk, #TI_PREEMPT]		@ get preempt count
@@ -211,6 +214,10 @@ preempt_return:
 #endif
 	ldr	r0, [sp, #S_PSR]		@ irqs are already disabled
 	msr	spsr_cxsf, r0
+#ifdef CONFIG_TRACE_IRQFLAGS
+	tst	r0, #PSR_I_BIT
+	bleq	trace_hardirqs_on
+#endif
 	ldmia	sp, {r0 - pc}^			@ load r0 - pc, cpsr
 
 	.ltorg
@@ -398,6 +405,9 @@ __dabt_usr:
 __irq_usr:
 	usr_entry
 
+#ifdef CONFIG_TRACE_IRQFLAGS
+	bl	trace_hardirqs_off
+#endif
 	get_thread_info tsk
 #ifdef CONFIG_PREEMPT
 	ldr	r8, [tsk, #TI_PREEMPT]		@ get preempt count
@@ -412,6 +422,9 @@ __irq_usr:
 	teq	r0, r7
 	strne	r0, [r0, -r0]
 #endif
+#ifdef CONFIG_TRACE_IRQFLAGS
+	bl	trace_hardirqs_on
+#endif
 
 	mov	why, #0
 	b	ret_to_user
diff --git a/include/asm-arm/irqflags.h b/include/asm-arm/irqflags.h
new file mode 100644
index 000000000000..6d09974e6646
--- /dev/null
+++ b/include/asm-arm/irqflags.h
@@ -0,0 +1,132 @@
+#ifndef __ASM_ARM_IRQFLAGS_H
+#define __ASM_ARM_IRQFLAGS_H
+
+#ifdef __KERNEL__
+
+#include <asm/ptrace.h>
+
+/*
+ * CPU interrupt mask handling.
+ */
+#if __LINUX_ARM_ARCH__ >= 6
+
+#define raw_local_irq_save(x)					\
+	({							\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ local_irq_save\n"	\
+	"cpsid	i"						\
+	: "=r" (x) : : "memory", "cc");				\
+	})
+
+#define raw_local_irq_enable()  __asm__("cpsie i	@ __sti" : : : "memory", "cc")
+#define raw_local_irq_disable() __asm__("cpsid i	@ __cli" : : : "memory", "cc")
+#define local_fiq_enable()  __asm__("cpsie f	@ __stf" : : : "memory", "cc")
+#define local_fiq_disable() __asm__("cpsid f	@ __clf" : : : "memory", "cc")
+
+#else
+
+/*
+ * Save the current interrupt enable state & disable IRQs
+ */
+#define raw_local_irq_save(x)					\
+	({							\
+		unsigned long temp;				\
+		(void) (&temp == &x);				\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ local_irq_save\n"	\
+"	orr	%1, %0, #128\n"					\
+"	msr	cpsr_c, %1"					\
+	: "=r" (x), "=r" (temp)					\
+	:							\
+	: "memory", "cc");					\
+	})
+	
+/*
+ * Enable IRQs
+ */
+#define raw_local_irq_enable()					\
+	({							\
+		unsigned long temp;				\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ local_irq_enable\n"	\
+"	bic	%0, %0, #128\n"					\
+"	msr	cpsr_c, %0"					\
+	: "=r" (temp)						\
+	:							\
+	: "memory", "cc");					\
+	})
+
+/*
+ * Disable IRQs
+ */
+#define raw_local_irq_disable()					\
+	({							\
+		unsigned long temp;				\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ local_irq_disable\n"	\
+"	orr	%0, %0, #128\n"					\
+"	msr	cpsr_c, %0"					\
+	: "=r" (temp)						\
+	:							\
+	: "memory", "cc");					\
+	})
+
+/*
+ * Enable FIQs
+ */
+#define local_fiq_enable()					\
+	({							\
+		unsigned long temp;				\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ stf\n"		\
+"	bic	%0, %0, #64\n"					\
+"	msr	cpsr_c, %0"					\
+	: "=r" (temp)						\
+	:							\
+	: "memory", "cc");					\
+	})
+
+/*
+ * Disable FIQs
+ */
+#define local_fiq_disable()					\
+	({							\
+		unsigned long temp;				\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ clf\n"		\
+"	orr	%0, %0, #64\n"					\
+"	msr	cpsr_c, %0"					\
+	: "=r" (temp)						\
+	:							\
+	: "memory", "cc");					\
+	})
+
+#endif
+
+/*
+ * Save the current interrupt enable state.
+ */
+#define raw_local_save_flags(x)					\
+	({							\
+	__asm__ __volatile__(					\
+	"mrs	%0, cpsr		@ local_save_flags"	\
+	: "=r" (x) : : "memory", "cc");				\
+	})
+
+/*
+ * restore saved IRQ & FIQ state
+ */
+#define raw_local_irq_restore(x)				\
+	__asm__ __volatile__(					\
+	"msr	cpsr_c, %0		@ local_irq_restore\n"	\
+	:							\
+	: "r" (x)						\
+	: "memory", "cc")
+
+#define raw_irqs_disabled_flags(flags)	\
+({					\
+	(int)((flags) & PSR_I_BIT);	\
+})
+
+#endif
+#endif
diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index 0947cbf9b69a..174ff52661b0 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -207,130 +207,7 @@ static inline void sched_cacheflush(void)
 {
 }
 
-/*
- * CPU interrupt mask handling.
- */
-#if __LINUX_ARM_ARCH__ >= 6
-
-#define local_irq_save(x)					\
-	({							\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_save\n"	\
-	"cpsid	i"						\
-	: "=r" (x) : : "memory", "cc");				\
-	})
-
-#define local_irq_enable()  __asm__("cpsie i	@ __sti" : : : "memory", "cc")
-#define local_irq_disable() __asm__("cpsid i	@ __cli" : : : "memory", "cc")
-#define local_fiq_enable()  __asm__("cpsie f	@ __stf" : : : "memory", "cc")
-#define local_fiq_disable() __asm__("cpsid f	@ __clf" : : : "memory", "cc")
-
-#else
-
-/*
- * Save the current interrupt enable state & disable IRQs
- */
-#define local_irq_save(x)					\
-	({							\
-		unsigned long temp;				\
-		(void) (&temp == &x);				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_save\n"	\
-"	orr	%1, %0, #128\n"					\
-"	msr	cpsr_c, %1"					\
-	: "=r" (x), "=r" (temp)					\
-	:							\
-	: "memory", "cc");					\
-	})
-	
-/*
- * Enable IRQs
- */
-#define local_irq_enable()					\
-	({							\
-		unsigned long temp;				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_enable\n"	\
-"	bic	%0, %0, #128\n"					\
-"	msr	cpsr_c, %0"					\
-	: "=r" (temp)						\
-	:							\
-	: "memory", "cc");					\
-	})
-
-/*
- * Disable IRQs
- */
-#define local_irq_disable()					\
-	({							\
-		unsigned long temp;				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_disable\n"	\
-"	orr	%0, %0, #128\n"					\
-"	msr	cpsr_c, %0"					\
-	: "=r" (temp)						\
-	:							\
-	: "memory", "cc");					\
-	})
-
-/*
- * Enable FIQs
- */
-#define local_fiq_enable()					\
-	({							\
-		unsigned long temp;				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ stf\n"		\
-"	bic	%0, %0, #64\n"					\
-"	msr	cpsr_c, %0"					\
-	: "=r" (temp)						\
-	:							\
-	: "memory", "cc");					\
-	})
-
-/*
- * Disable FIQs
- */
-#define local_fiq_disable()					\
-	({							\
-		unsigned long temp;				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ clf\n"		\
-"	orr	%0, %0, #64\n"					\
-"	msr	cpsr_c, %0"					\
-	: "=r" (temp)						\
-	:							\
-	: "memory", "cc");					\
-	})
-
-#endif
-
-/*
- * Save the current interrupt enable state.
- */
-#define local_save_flags(x)					\
-	({							\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_save_flags"	\
-	: "=r" (x) : : "memory", "cc");				\
-	})
-
-/*
- * restore saved IRQ & FIQ state
- */
-#define local_irq_restore(x)					\
-	__asm__ __volatile__(					\
-	"msr	cpsr_c, %0		@ local_irq_restore\n"	\
-	:							\
-	: "r" (x)						\
-	: "memory", "cc")
-
-#define irqs_disabled()			\
-({					\
-	unsigned long flags;		\
-	local_save_flags(flags);	\
-	(int)(flags & PSR_I_BIT);	\
-})
+#include <linux/irqflags.h>
 
 #ifdef CONFIG_SMP
 
-- 
cgit v1.2.3


From 6a39dd6222dda5ee2414a1b42e8e62118742a49e Mon Sep 17 00:00:00 2001
From: Daniel Jacobowitz <drow@false.org>
Date: Wed, 30 Aug 2006 15:02:08 +0100
Subject: [ARM] 3759/2: Remove uses of %?

Patch from Daniel Jacobowitz

The ARM kernel has several uses of asm("foo%?").  %? is a GCC internal
modifier used to output conditional execution predicates.  However, no
version of GCC supports conditionalizing asm statements.  GCC 4.2 will
correctly expand %? to the empty string in user asms.  Earlier versions may
reuse the condition from the previous instruction.  In 'if (foo) asm
("bar%?");' this is somewhat likely to be right... but not reliable.

So, the only safe thing to do is to remove the uses of %?.  I believe
the tlbflush.h occurances were supposed to be removed before, based
on the comment about %? not working at the top of that file.

Old versions of GCC could omit branches around user asms if the asm didn't
mark the condition codes as clobbered.  This problem hasn't been seen on any
recent (3.x or 4.x) GCC, but it could theoretically happen.  So, where
%? was removed a cc clobber was added.

Signed-off-by: Daniel Jacobowitz <dan@codesourcery.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/traps.c            |  2 +-
 arch/arm/mach-footbridge/dc21285.c | 27 ++++++++------
 arch/arm/vfp/vfpinstr.h            |  8 ++--
 include/asm-arm/arch-l7200/io.h    |  8 ++--
 include/asm-arm/tlbflush.h         | 76 +++++++++++++++++++-------------------
 5 files changed, 62 insertions(+), 59 deletions(-)

(limited to 'include')

diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index aeeed806f991..bede380c07a9 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -191,7 +191,7 @@ void show_stack(struct task_struct *tsk, unsigned long *sp)
 	if (tsk != current)
 		fp = thread_saved_fp(tsk);
 	else
-		asm("mov%? %0, fp" : "=r" (fp));
+		asm("mov %0, fp" : "=r" (fp) : : "cc");
 
 	c_backtrace(fp, 0x10);
 	barrier();
diff --git a/arch/arm/mach-footbridge/dc21285.c b/arch/arm/mach-footbridge/dc21285.c
index 823e25d4547e..a1ae49df5c3b 100644
--- a/arch/arm/mach-footbridge/dc21285.c
+++ b/arch/arm/mach-footbridge/dc21285.c
@@ -69,16 +69,16 @@ dc21285_read_config(struct pci_bus *bus, unsigned int devfn, int where,
 	if (addr)
 		switch (size) {
 		case 1:
-			asm("ldr%?b	%0, [%1, %2]"
-				: "=r" (v) : "r" (addr), "r" (where));
+			asm("ldrb	%0, [%1, %2]"
+				: "=r" (v) : "r" (addr), "r" (where) : "cc");
 			break;
 		case 2:
-			asm("ldr%?h	%0, [%1, %2]"
-				: "=r" (v) : "r" (addr), "r" (where));
+			asm("ldrh	%0, [%1, %2]"
+				: "=r" (v) : "r" (addr), "r" (where) : "cc");
 			break;
 		case 4:
-			asm("ldr%?	%0, [%1, %2]"
-				: "=r" (v) : "r" (addr), "r" (where));
+			asm("ldr	%0, [%1, %2]"
+				: "=r" (v) : "r" (addr), "r" (where) : "cc");
 			break;
 		}
 
@@ -103,16 +103,19 @@ dc21285_write_config(struct pci_bus *bus, unsigned int devfn, int where,
 	if (addr)
 		switch (size) {
 		case 1:
-			asm("str%?b	%0, [%1, %2]"
-				: : "r" (value), "r" (addr), "r" (where));
+			asm("strb	%0, [%1, %2]"
+				: : "r" (value), "r" (addr), "r" (where)
+				: "cc");
 			break;
 		case 2:
-			asm("str%?h	%0, [%1, %2]"
-				: : "r" (value), "r" (addr), "r" (where));
+			asm("strh	%0, [%1, %2]"
+				: : "r" (value), "r" (addr), "r" (where)
+				: "cc");
 			break;
 		case 4:
-			asm("str%?	%0, [%1, %2]"
-				: : "r" (value), "r" (addr), "r" (where));
+			asm("str	%0, [%1, %2]"
+				: : "r" (value), "r" (addr), "r" (where)
+				: "cc");
 			break;
 		}
 
diff --git a/arch/arm/vfp/vfpinstr.h b/arch/arm/vfp/vfpinstr.h
index 6c819aeae006..7f343a4beca0 100644
--- a/arch/arm/vfp/vfpinstr.h
+++ b/arch/arm/vfp/vfpinstr.h
@@ -73,14 +73,14 @@
 
 #define fmrx(_vfp_) ({			\
 	u32 __v;			\
-	asm("mrc%? p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmrx	%0, " #_vfp_	\
-	    : "=r" (__v));		\
+	asm("mrc p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmrx	%0, " #_vfp_	\
+	    : "=r" (__v) : : "cc");	\
 	__v;				\
  })
 
 #define fmxr(_vfp_,_var_)		\
-	asm("mcr%? p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmxr	" #_vfp_ ", %0"	\
-	   : : "r" (_var_))
+	asm("mcr p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmxr	" #_vfp_ ", %0"	\
+	   : : "r" (_var_) : "cc")
 
 u32 vfp_single_cpdo(u32 inst, u32 fpscr);
 u32 vfp_single_cprt(u32 inst, u32 fpscr, struct pt_regs *regs);
diff --git a/include/asm-arm/arch-l7200/io.h b/include/asm-arm/arch-l7200/io.h
index cd080d8384d9..d744d97c18a5 100644
--- a/include/asm-arm/arch-l7200/io.h
+++ b/include/asm-arm/arch-l7200/io.h
@@ -31,9 +31,9 @@
 static inline unsigned int __arch_getw(unsigned long a)
 {
 	unsigned int value;
-	__asm__ __volatile__("ldr%?h    %0, [%1, #0]    @ getw"
+	__asm__ __volatile__("ldrh    %0, [%1, #0]    @ getw"
 		: "=&r" (value)
-		: "r" (a));
+		: "r" (a) : "cc");
 	return value;
 }
 
@@ -42,8 +42,8 @@ static inline unsigned int __arch_getw(unsigned long a)
 
 static inline void __arch_putw(unsigned int value, unsigned long a)
 {
-        __asm__ __volatile__("str%?h    %0, [%1, #0]    @ putw"
-                : : "r" (value), "r" (a));
+        __asm__ __volatile__("strh    %0, [%1, #0]    @ putw"
+                : : "r" (value), "r" (a) : "cc");
 }
 
 /*
diff --git a/include/asm-arm/tlbflush.h b/include/asm-arm/tlbflush.h
index d97fc76189a5..cd10a0b5f8ae 100644
--- a/include/asm-arm/tlbflush.h
+++ b/include/asm-arm/tlbflush.h
@@ -247,16 +247,16 @@ static inline void local_flush_tlb_all(void)
 	const unsigned int __tlb_flag = __cpu_tlb_flags;
 
 	if (tlb_flag(TLB_WB))
-		asm("mcr%? p15, 0, %0, c7, c10, 4" : : "r" (zero));
+		asm("mcr p15, 0, %0, c7, c10, 4" : : "r" (zero) : "cc");
 
 	if (tlb_flag(TLB_V3_FULL))
-		asm("mcr%? p15, 0, %0, c6, c0, 0" : : "r" (zero));
+		asm("mcr p15, 0, %0, c6, c0, 0" : : "r" (zero) : "cc");
 	if (tlb_flag(TLB_V4_U_FULL | TLB_V6_U_FULL))
-		asm("mcr%? p15, 0, %0, c8, c7, 0" : : "r" (zero));
+		asm("mcr p15, 0, %0, c8, c7, 0" : : "r" (zero) : "cc");
 	if (tlb_flag(TLB_V4_D_FULL | TLB_V6_D_FULL))
-		asm("mcr%? p15, 0, %0, c8, c6, 0" : : "r" (zero));
+		asm("mcr p15, 0, %0, c8, c6, 0" : : "r" (zero) : "cc");
 	if (tlb_flag(TLB_V4_I_FULL | TLB_V6_I_FULL))
-		asm("mcr%? p15, 0, %0, c8, c5, 0" : : "r" (zero));
+		asm("mcr p15, 0, %0, c8, c5, 0" : : "r" (zero) : "cc");
 }
 
 static inline void local_flush_tlb_mm(struct mm_struct *mm)
@@ -266,25 +266,25 @@ static inline void local_flush_tlb_mm(struct mm_struct *mm)
 	const unsigned int __tlb_flag = __cpu_tlb_flags;
 
 	if (tlb_flag(TLB_WB))
-		asm("mcr%? p15, 0, %0, c7, c10, 4" : : "r" (zero));
+		asm("mcr p15, 0, %0, c7, c10, 4" : : "r" (zero) : "cc");
 
 	if (cpu_isset(smp_processor_id(), mm->cpu_vm_mask)) {
 		if (tlb_flag(TLB_V3_FULL))
-			asm("mcr%? p15, 0, %0, c6, c0, 0" : : "r" (zero));
+			asm("mcr p15, 0, %0, c6, c0, 0" : : "r" (zero) : "cc");
 		if (tlb_flag(TLB_V4_U_FULL))
-			asm("mcr%? p15, 0, %0, c8, c7, 0" : : "r" (zero));
+			asm("mcr p15, 0, %0, c8, c7, 0" : : "r" (zero) : "cc");
 		if (tlb_flag(TLB_V4_D_FULL))
-			asm("mcr%? p15, 0, %0, c8, c6, 0" : : "r" (zero));
+			asm("mcr p15, 0, %0, c8, c6, 0" : : "r" (zero) : "cc");
 		if (tlb_flag(TLB_V4_I_FULL))
-			asm("mcr%? p15, 0, %0, c8, c5, 0" : : "r" (zero));
+			asm("mcr p15, 0, %0, c8, c5, 0" : : "r" (zero) : "cc");
 	}
 
 	if (tlb_flag(TLB_V6_U_ASID))
-		asm("mcr%? p15, 0, %0, c8, c7, 2" : : "r" (asid));
+		asm("mcr p15, 0, %0, c8, c7, 2" : : "r" (asid) : "cc");
 	if (tlb_flag(TLB_V6_D_ASID))
-		asm("mcr%? p15, 0, %0, c8, c6, 2" : : "r" (asid));
+		asm("mcr p15, 0, %0, c8, c6, 2" : : "r" (asid) : "cc");
 	if (tlb_flag(TLB_V6_I_ASID))
-		asm("mcr%? p15, 0, %0, c8, c5, 2" : : "r" (asid));
+		asm("mcr p15, 0, %0, c8, c5, 2" : : "r" (asid) : "cc");
 }
 
 static inline void
@@ -296,27 +296,27 @@ local_flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr)
 	uaddr = (uaddr & PAGE_MASK) | ASID(vma->vm_mm);
 
 	if (tlb_flag(TLB_WB))
-		asm("mcr%? p15, 0, %0, c7, c10, 4" : : "r" (zero));
+		asm("mcr p15, 0, %0, c7, c10, 4" : : "r" (zero));
 
 	if (cpu_isset(smp_processor_id(), vma->vm_mm->cpu_vm_mask)) {
 		if (tlb_flag(TLB_V3_PAGE))
-			asm("mcr%? p15, 0, %0, c6, c0, 0" : : "r" (uaddr));
+			asm("mcr p15, 0, %0, c6, c0, 0" : : "r" (uaddr) : "cc");
 		if (tlb_flag(TLB_V4_U_PAGE))
-			asm("mcr%? p15, 0, %0, c8, c7, 1" : : "r" (uaddr));
+			asm("mcr p15, 0, %0, c8, c7, 1" : : "r" (uaddr) : "cc");
 		if (tlb_flag(TLB_V4_D_PAGE))
-			asm("mcr%? p15, 0, %0, c8, c6, 1" : : "r" (uaddr));
+			asm("mcr p15, 0, %0, c8, c6, 1" : : "r" (uaddr) : "cc");
 		if (tlb_flag(TLB_V4_I_PAGE))
-			asm("mcr%? p15, 0, %0, c8, c5, 1" : : "r" (uaddr));
+			asm("mcr p15, 0, %0, c8, c5, 1" : : "r" (uaddr) : "cc");
 		if (!tlb_flag(TLB_V4_I_PAGE) && tlb_flag(TLB_V4_I_FULL))
-			asm("mcr%? p15, 0, %0, c8, c5, 0" : : "r" (zero));
+			asm("mcr p15, 0, %0, c8, c5, 0" : : "r" (zero) : "cc");
 	}
 
 	if (tlb_flag(TLB_V6_U_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c7, 1" : : "r" (uaddr));
+		asm("mcr p15, 0, %0, c8, c7, 1" : : "r" (uaddr) : "cc");
 	if (tlb_flag(TLB_V6_D_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c6, 1" : : "r" (uaddr));
+		asm("mcr p15, 0, %0, c8, c6, 1" : : "r" (uaddr) : "cc");
 	if (tlb_flag(TLB_V6_I_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c5, 1" : : "r" (uaddr));
+		asm("mcr p15, 0, %0, c8, c5, 1" : : "r" (uaddr) : "cc");
 }
 
 static inline void local_flush_tlb_kernel_page(unsigned long kaddr)
@@ -327,31 +327,31 @@ static inline void local_flush_tlb_kernel_page(unsigned long kaddr)
 	kaddr &= PAGE_MASK;
 
 	if (tlb_flag(TLB_WB))
-		asm("mcr%? p15, 0, %0, c7, c10, 4" : : "r" (zero));
+		asm("mcr p15, 0, %0, c7, c10, 4" : : "r" (zero) : "cc");
 
 	if (tlb_flag(TLB_V3_PAGE))
-		asm("mcr%? p15, 0, %0, c6, c0, 0" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c6, c0, 0" : : "r" (kaddr) : "cc");
 	if (tlb_flag(TLB_V4_U_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c7, 1" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c8, c7, 1" : : "r" (kaddr) : "cc");
 	if (tlb_flag(TLB_V4_D_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c6, 1" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c8, c6, 1" : : "r" (kaddr) : "cc");
 	if (tlb_flag(TLB_V4_I_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c5, 1" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c8, c5, 1" : : "r" (kaddr) : "cc");
 	if (!tlb_flag(TLB_V4_I_PAGE) && tlb_flag(TLB_V4_I_FULL))
-		asm("mcr%? p15, 0, %0, c8, c5, 0" : : "r" (zero));
+		asm("mcr p15, 0, %0, c8, c5, 0" : : "r" (zero) : "cc");
 
 	if (tlb_flag(TLB_V6_U_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c7, 1" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c8, c7, 1" : : "r" (kaddr) : "cc");
 	if (tlb_flag(TLB_V6_D_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c6, 1" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c8, c6, 1" : : "r" (kaddr) : "cc");
 	if (tlb_flag(TLB_V6_I_PAGE))
-		asm("mcr%? p15, 0, %0, c8, c5, 1" : : "r" (kaddr));
+		asm("mcr p15, 0, %0, c8, c5, 1" : : "r" (kaddr) : "cc");
 
 	/* The ARM ARM states that the completion of a TLB maintenance
 	 * operation is only guaranteed by a DSB instruction
 	 */
 	if (tlb_flag(TLB_V6_U_PAGE | TLB_V6_D_PAGE | TLB_V6_I_PAGE))
-		asm("mcr%? p15, 0, %0, c7, c10, 4" : : "r" (zero));
+		asm("mcr p15, 0, %0, c7, c10, 4" : : "r" (zero) : "cc");
 }
 
 /*
@@ -373,11 +373,11 @@ static inline void flush_pmd_entry(pmd_t *pmd)
 	const unsigned int __tlb_flag = __cpu_tlb_flags;
 
 	if (tlb_flag(TLB_DCLEAN))
-		asm("mcr%?	p15, 0, %0, c7, c10, 1	@ flush_pmd"
-			: : "r" (pmd));
+		asm("mcr	p15, 0, %0, c7, c10, 1	@ flush_pmd"
+			: : "r" (pmd) : "cc");
 	if (tlb_flag(TLB_WB))
-		asm("mcr%?	p15, 0, %0, c7, c10, 4	@ flush_pmd"
-			: : "r" (zero));
+		asm("mcr	p15, 0, %0, c7, c10, 4	@ flush_pmd"
+			: : "r" (zero) : "cc");
 }
 
 static inline void clean_pmd_entry(pmd_t *pmd)
@@ -385,8 +385,8 @@ static inline void clean_pmd_entry(pmd_t *pmd)
 	const unsigned int __tlb_flag = __cpu_tlb_flags;
 
 	if (tlb_flag(TLB_DCLEAN))
-		asm("mcr%?	p15, 0, %0, c7, c10, 1	@ flush_pmd"
-			: : "r" (pmd));
+		asm("mcr	p15, 0, %0, c7, c10, 1	@ flush_pmd"
+			: : "r" (pmd) : "cc");
 }
 
 #undef tlb_flag
-- 
cgit v1.2.3


From acc46c0144b6d1cf0d77bb8b4d1b7dcd5dc28d71 Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@snapgear.com>
Date: Thu, 14 Sep 2006 00:28:26 +1000
Subject: [ARM] nommu: create flat.h to support uClinux flat binaries

Create header with uClinux flat format binary support macros for ARM
platforms. Derived from the m68knommu flat.h.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/flat.h | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
 create mode 100644 include/asm-arm/flat.h

(limited to 'include')

diff --git a/include/asm-arm/flat.h b/include/asm-arm/flat.h
new file mode 100644
index 000000000000..966946478589
--- /dev/null
+++ b/include/asm-arm/flat.h
@@ -0,0 +1,16 @@
+/*
+ * include/asm-arm/flat.h -- uClinux flat-format executables
+ */
+
+#ifndef __ARM_FLAT_H__
+#define __ARM_FLAT_H__
+
+#define	flat_stack_align(sp)			/* nothing needed */
+#define	flat_argvp_envp_on_stack()		1
+#define	flat_old_ram_flag(flags)		(flags)
+#define	flat_reloc_valid(reloc, size)		((reloc) <= (size))
+#define	flat_get_addr_from_rp(rp, relval, flags) get_unaligned(rp)
+#define	flat_put_addr_at_rp(rp, val, relval)	put_unaligned(val,rp)
+#define	flat_get_relocate_addr(rel)		(rel)
+
+#endif /* __ARM_FLAT_H__ */
-- 
cgit v1.2.3


From 5432114baf0300286a6ca1b0aea549492a379432 Mon Sep 17 00:00:00 2001
From: Ralph Wuerthner <rwuerthn@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:36 +0200
Subject: [S390] zcrypt secure key cryptography extension.

Allow the user space to send extended cprb messages directly to the
PCIXCC / CEX2C cards. This allows the CCA library to construct special
crypto requests that use "secure" keys that are stored on the card.

Signed-off-by: Ralph Wuerthner <rwuerthn@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/crypto/zcrypt_api.c     | 112 ++++++++++++++-
 drivers/s390/crypto/zcrypt_api.h     |   3 +-
 drivers/s390/crypto/zcrypt_cca_key.h |   2 +-
 drivers/s390/crypto/zcrypt_cex2a.c   |   2 +-
 drivers/s390/crypto/zcrypt_cex2a.h   |   2 +-
 drivers/s390/crypto/zcrypt_error.h   |   2 +-
 drivers/s390/crypto/zcrypt_mono.c    |   2 +-
 drivers/s390/crypto/zcrypt_pcica.c   |   2 +-
 drivers/s390/crypto/zcrypt_pcica.h   |   2 +-
 drivers/s390/crypto/zcrypt_pcicc.c   |   2 +-
 drivers/s390/crypto/zcrypt_pcicc.h   |   2 +-
 drivers/s390/crypto/zcrypt_pcixcc.c  | 263 +++++++++++++++++++++++++++++++++--
 drivers/s390/crypto/zcrypt_pcixcc.h  |   2 +-
 include/asm-s390/zcrypt.h            |  80 ++++++++++-
 14 files changed, 452 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/drivers/s390/crypto/zcrypt_api.c b/drivers/s390/crypto/zcrypt_api.c
index b3fe003b3d2d..1edc10a7a6f2 100644
--- a/drivers/s390/crypto/zcrypt_api.c
+++ b/drivers/s390/crypto/zcrypt_api.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_api.c
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
@@ -392,6 +392,41 @@ static long zcrypt_rsa_crt(struct ica_rsa_modexpo_crt *crt)
 	return -ENODEV;
 }
 
+static long zcrypt_send_cprb(struct ica_xcRB *xcRB)
+{
+	struct zcrypt_device *zdev;
+	int rc;
+
+	spin_lock_bh(&zcrypt_device_lock);
+	list_for_each_entry(zdev, &zcrypt_device_list, list) {
+		if (!zdev->online || !zdev->ops->send_cprb ||
+		    (xcRB->user_defined != AUTOSELECT &&
+			AP_QID_DEVICE(zdev->ap_dev->qid) != xcRB->user_defined)
+		    )
+			continue;
+		zcrypt_device_get(zdev);
+		get_device(&zdev->ap_dev->device);
+		zdev->request_count++;
+		__zcrypt_decrease_preference(zdev);
+		spin_unlock_bh(&zcrypt_device_lock);
+		if (try_module_get(zdev->ap_dev->drv->driver.owner)) {
+			rc = zdev->ops->send_cprb(zdev, xcRB);
+			module_put(zdev->ap_dev->drv->driver.owner);
+		}
+		else
+			rc = -EAGAIN;
+		spin_lock_bh(&zcrypt_device_lock);
+		zdev->request_count--;
+		__zcrypt_increase_preference(zdev);
+		put_device(&zdev->ap_dev->device);
+		zcrypt_device_put(zdev);
+		spin_unlock_bh(&zcrypt_device_lock);
+		return rc;
+	}
+	spin_unlock_bh(&zcrypt_device_lock);
+	return -ENODEV;
+}
+
 static void zcrypt_status_mask(char status[AP_DEVICES])
 {
 	struct zcrypt_device *zdev;
@@ -535,6 +570,18 @@ static long zcrypt_unlocked_ioctl(struct file *filp, unsigned int cmd,
 			return rc;
 		return put_user(crt.outputdatalength, &ucrt->outputdatalength);
 	}
+	case ZSECSENDCPRB: {
+		struct ica_xcRB __user *uxcRB = (void __user *) arg;
+		struct ica_xcRB xcRB;
+		if (copy_from_user(&xcRB, uxcRB, sizeof(xcRB)))
+			return -EFAULT;
+		do {
+			rc = zcrypt_send_cprb(&xcRB);
+		} while (rc == -EAGAIN);
+		if (copy_to_user(uxcRB, &xcRB, sizeof(xcRB)))
+			return -EFAULT;
+		return rc;
+	}
 	case Z90STAT_STATUS_MASK: {
 		char status[AP_DEVICES];
 		zcrypt_status_mask(status);
@@ -683,6 +730,67 @@ static long trans_modexpo_crt32(struct file *filp, unsigned int cmd,
 	return rc;
 }
 
+struct compat_ica_xcRB {
+	unsigned short	agent_ID;
+	unsigned int	user_defined;
+	unsigned short	request_ID;
+	unsigned int	request_control_blk_length;
+	unsigned char	padding1[16 - sizeof (compat_uptr_t)];
+	compat_uptr_t	request_control_blk_addr;
+	unsigned int	request_data_length;
+	char		padding2[16 - sizeof (compat_uptr_t)];
+	compat_uptr_t	request_data_address;
+	unsigned int	reply_control_blk_length;
+	char		padding3[16 - sizeof (compat_uptr_t)];
+	compat_uptr_t	reply_control_blk_addr;
+	unsigned int	reply_data_length;
+	char		padding4[16 - sizeof (compat_uptr_t)];
+	compat_uptr_t	reply_data_addr;
+	unsigned short	priority_window;
+	unsigned int	status;
+} __attribute__((packed));
+
+static long trans_xcRB32(struct file *filp, unsigned int cmd,
+			 unsigned long arg)
+{
+	struct compat_ica_xcRB __user *uxcRB32 = compat_ptr(arg);
+	struct compat_ica_xcRB xcRB32;
+	struct ica_xcRB xcRB64;
+	long rc;
+
+	if (copy_from_user(&xcRB32, uxcRB32, sizeof(xcRB32)))
+		return -EFAULT;
+	xcRB64.agent_ID = xcRB32.agent_ID;
+	xcRB64.user_defined = xcRB32.user_defined;
+	xcRB64.request_ID = xcRB32.request_ID;
+	xcRB64.request_control_blk_length =
+		xcRB32.request_control_blk_length;
+	xcRB64.request_control_blk_addr =
+		compat_ptr(xcRB32.request_control_blk_addr);
+	xcRB64.request_data_length =
+		xcRB32.request_data_length;
+	xcRB64.request_data_address =
+		compat_ptr(xcRB32.request_data_address);
+	xcRB64.reply_control_blk_length =
+		xcRB32.reply_control_blk_length;
+	xcRB64.reply_control_blk_addr =
+		compat_ptr(xcRB32.reply_control_blk_addr);
+	xcRB64.reply_data_length = xcRB32.reply_data_length;
+	xcRB64.reply_data_addr =
+		compat_ptr(xcRB32.reply_data_addr);
+	xcRB64.priority_window = xcRB32.priority_window;
+	xcRB64.status = xcRB32.status;
+	do {
+		rc = zcrypt_send_cprb(&xcRB64);
+	} while (rc == -EAGAIN);
+	xcRB32.reply_control_blk_length = xcRB64.reply_control_blk_length;
+	xcRB32.reply_data_length = xcRB64.reply_data_length;
+	xcRB32.status = xcRB64.status;
+	if (copy_to_user(uxcRB32, &xcRB32, sizeof(xcRB32)))
+			return -EFAULT;
+	return rc;
+}
+
 long zcrypt_compat_ioctl(struct file *filp, unsigned int cmd,
 			 unsigned long arg)
 {
@@ -690,6 +798,8 @@ long zcrypt_compat_ioctl(struct file *filp, unsigned int cmd,
 		return trans_modexpo32(filp, cmd, arg);
 	if (cmd == ICARSACRT)
 		return trans_modexpo_crt32(filp, cmd, arg);
+	if (cmd == ZSECSENDCPRB)
+		return trans_xcRB32(filp, cmd, arg);
 	return zcrypt_unlocked_ioctl(filp, cmd, arg);
 }
 #endif
diff --git a/drivers/s390/crypto/zcrypt_api.h b/drivers/s390/crypto/zcrypt_api.h
index 1f0e61f2e9b4..de4877ee618f 100644
--- a/drivers/s390/crypto/zcrypt_api.h
+++ b/drivers/s390/crypto/zcrypt_api.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_api.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
@@ -106,6 +106,7 @@ struct zcrypt_ops {
 	long (*rsa_modexpo)(struct zcrypt_device *, struct ica_rsa_modexpo *);
 	long (*rsa_modexpo_crt)(struct zcrypt_device *,
 				struct ica_rsa_modexpo_crt *);
+	long (*send_cprb)(struct zcrypt_device *, struct ica_xcRB *);
 };
 
 struct zcrypt_device {
diff --git a/drivers/s390/crypto/zcrypt_cca_key.h b/drivers/s390/crypto/zcrypt_cca_key.h
index c80f40d44197..8dbcf0eef3e5 100644
--- a/drivers/s390/crypto/zcrypt_cca_key.h
+++ b/drivers/s390/crypto/zcrypt_cca_key.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_cca_key.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_cex2a.c b/drivers/s390/crypto/zcrypt_cex2a.c
index 350248e5cd93..a62b00083d0c 100644
--- a/drivers/s390/crypto/zcrypt_cex2a.c
+++ b/drivers/s390/crypto/zcrypt_cex2a.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_cex2a.c
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_cex2a.h b/drivers/s390/crypto/zcrypt_cex2a.h
index 61a78c32dce4..8f69d1dacab8 100644
--- a/drivers/s390/crypto/zcrypt_cex2a.h
+++ b/drivers/s390/crypto/zcrypt_cex2a.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_cex2a.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_error.h b/drivers/s390/crypto/zcrypt_error.h
index b22bd055a03b..2cb616ba8bec 100644
--- a/drivers/s390/crypto/zcrypt_error.h
+++ b/drivers/s390/crypto/zcrypt_error.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_error.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_mono.c b/drivers/s390/crypto/zcrypt_mono.c
index f48b61a6126c..2a9349ad68b7 100644
--- a/drivers/s390/crypto/zcrypt_mono.c
+++ b/drivers/s390/crypto/zcrypt_mono.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_mono.c
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_pcica.c b/drivers/s390/crypto/zcrypt_pcica.c
index 0ff56e86caae..b6a4ecdc8025 100644
--- a/drivers/s390/crypto/zcrypt_pcica.c
+++ b/drivers/s390/crypto/zcrypt_pcica.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_pcica.c
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_pcica.h b/drivers/s390/crypto/zcrypt_pcica.h
index a08a4f8c33c9..3be11187f6df 100644
--- a/drivers/s390/crypto/zcrypt_pcica.h
+++ b/drivers/s390/crypto/zcrypt_pcica.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_pcica.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_pcicc.c b/drivers/s390/crypto/zcrypt_pcicc.c
index 900362983fec..f295a403b29a 100644
--- a/drivers/s390/crypto/zcrypt_pcicc.c
+++ b/drivers/s390/crypto/zcrypt_pcicc.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_pcicc.c
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_pcicc.h b/drivers/s390/crypto/zcrypt_pcicc.h
index 027bafc7312a..6d4454846c8f 100644
--- a/drivers/s390/crypto/zcrypt_pcicc.h
+++ b/drivers/s390/crypto/zcrypt_pcicc.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_pcicc.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/drivers/s390/crypto/zcrypt_pcixcc.c b/drivers/s390/crypto/zcrypt_pcixcc.c
index 6064cf58be43..2da8b9381407 100644
--- a/drivers/s390/crypto/zcrypt_pcixcc.c
+++ b/drivers/s390/crypto/zcrypt_pcixcc.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_pcixcc.c
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
@@ -60,6 +60,15 @@
 
 #define PCIXCC_CLEANUP_TIME	(15*HZ)
 
+#define CEIL4(x) ((((x)+3)/4)*4)
+
+struct response_type {
+	struct completion work;
+	int type;
+};
+#define PCIXCC_RESPONSE_TYPE_ICA  0
+#define PCIXCC_RESPONSE_TYPE_XCRB 1
+
 static struct ap_device_id zcrypt_pcixcc_ids[] = {
 	{ AP_DEVICE(AP_DEVICE_TYPE_PCIXCC) },
 	{ AP_DEVICE(AP_DEVICE_TYPE_CEX2C) },
@@ -243,6 +252,108 @@ static int ICACRT_msg_to_type6CRT_msgX(struct zcrypt_device *zdev,
 	return 0;
 }
 
+/**
+ * Convert a XCRB message to a type6 CPRB message.
+ *
+ * @zdev: crypto device pointer
+ * @ap_msg: pointer to AP message
+ * @xcRB: pointer to user input data
+ *
+ * Returns 0 on success or -EFAULT.
+ */
+struct type86_fmt2_msg {
+	struct type86_hdr hdr;
+	struct type86_fmt2_ext fmt2;
+} __attribute__((packed));
+
+static int XCRB_msg_to_type6CPRB_msgX(struct zcrypt_device *zdev,
+				       struct ap_message *ap_msg,
+				       struct ica_xcRB *xcRB)
+{
+	static struct type6_hdr static_type6_hdrX = {
+		.type		=  0x06,
+		.offset1	=  0x00000058,
+	};
+	struct {
+		struct type6_hdr hdr;
+		struct ica_CPRBX cprbx;
+	} __attribute__((packed)) *msg = ap_msg->message;
+
+	int rcblen = CEIL4(xcRB->request_control_blk_length);
+	int replylen;
+	char *req_data = ap_msg->message + sizeof(struct type6_hdr) + rcblen;
+	char *function_code;
+
+	/* length checks */
+	ap_msg->length = sizeof(struct type6_hdr) +
+		CEIL4(xcRB->request_control_blk_length) +
+		xcRB->request_data_length;
+	if (ap_msg->length > PCIXCC_MAX_XCRB_MESSAGE_SIZE) {
+		PRINTK("Combined message is too large (%ld/%d/%d).\n",
+		    sizeof(struct type6_hdr),
+		    xcRB->request_control_blk_length,
+		    xcRB->request_data_length);
+		return -EFAULT;
+	}
+	if (CEIL4(xcRB->reply_control_blk_length) >
+	    PCIXCC_MAX_XCRB_REPLY_SIZE) {
+		PDEBUG("Reply CPRB length is too large (%d).\n",
+		    xcRB->request_control_blk_length);
+		return -EFAULT;
+	}
+	if (CEIL4(xcRB->reply_data_length) > PCIXCC_MAX_XCRB_DATA_SIZE) {
+		PDEBUG("Reply data block length is too large (%d).\n",
+		    xcRB->reply_data_length);
+		return -EFAULT;
+	}
+	replylen = CEIL4(xcRB->reply_control_blk_length) +
+		CEIL4(xcRB->reply_data_length) +
+		sizeof(struct type86_fmt2_msg);
+	if (replylen > PCIXCC_MAX_XCRB_RESPONSE_SIZE) {
+		PDEBUG("Reply CPRB + data block > PCIXCC_MAX_XCRB_RESPONSE_SIZE"
+		       " (%d/%d/%d).\n",
+		       sizeof(struct type86_fmt2_msg),
+		       xcRB->reply_control_blk_length,
+		       xcRB->reply_data_length);
+		xcRB->reply_control_blk_length = PCIXCC_MAX_XCRB_RESPONSE_SIZE -
+			(sizeof(struct type86_fmt2_msg) +
+			    CEIL4(xcRB->reply_data_length));
+		PDEBUG("Capping Reply CPRB length at %d\n",
+		       xcRB->reply_control_blk_length);
+	}
+
+	/* prepare type6 header */
+	msg->hdr = static_type6_hdrX;
+	memcpy(msg->hdr.agent_id , &(xcRB->agent_ID), sizeof(xcRB->agent_ID));
+	msg->hdr.ToCardLen1 = xcRB->request_control_blk_length;
+	if (xcRB->request_data_length) {
+		msg->hdr.offset2 = msg->hdr.offset1 + rcblen;
+		msg->hdr.ToCardLen2 = xcRB->request_data_length;
+	}
+	msg->hdr.FromCardLen1 = xcRB->reply_control_blk_length;
+	msg->hdr.FromCardLen2 = xcRB->reply_data_length;
+
+	/* prepare CPRB */
+	if (copy_from_user(&(msg->cprbx), xcRB->request_control_blk_addr,
+		    xcRB->request_control_blk_length))
+		return -EFAULT;
+	if (msg->cprbx.cprb_len + sizeof(msg->hdr.function_code) >
+	    xcRB->request_control_blk_length) {
+		PDEBUG("cprb_len too large (%d/%d)\n", msg->cprbx.cprb_len,
+		    xcRB->request_control_blk_length);
+		return -EFAULT;
+	}
+	function_code = ((unsigned char *)&msg->cprbx) + msg->cprbx.cprb_len;
+	memcpy(msg->hdr.function_code, function_code, sizeof(msg->hdr.function_code));
+
+	/* copy data block */
+	if (xcRB->request_data_length &&
+	    copy_from_user(req_data, xcRB->request_data_address,
+		xcRB->request_data_length))
+		return -EFAULT;
+	return 0;
+}
+
 /**
  * Copy results from a type 86 ICA reply message back to user space.
  *
@@ -363,6 +474,37 @@ static int convert_type86_ica(struct zcrypt_device *zdev,
 	return 0;
 }
 
+/**
+ * Copy results from a type 86 XCRB reply message back to user space.
+ *
+ * @zdev: crypto device pointer
+ * @reply: reply AP message.
+ * @xcRB: pointer to XCRB
+ *
+ * Returns 0 on success or -EINVAL, -EFAULT, -EAGAIN in case of an error.
+ */
+static int convert_type86_xcrb(struct zcrypt_device *zdev,
+			       struct ap_message *reply,
+			       struct ica_xcRB *xcRB)
+{
+	struct type86_fmt2_msg *msg = reply->message;
+	char *data = reply->message;
+
+	/* Copy CPRB to user */
+	if (copy_to_user(xcRB->reply_control_blk_addr,
+		data + msg->fmt2.offset1, msg->fmt2.count1))
+		return -EFAULT;
+	xcRB->reply_control_blk_length = msg->fmt2.count1;
+
+	/* Copy data buffer to user */
+	if (msg->fmt2.count2)
+		if (copy_to_user(xcRB->reply_data_addr,
+			data + msg->fmt2.offset2, msg->fmt2.count2))
+			return -EFAULT;
+	xcRB->reply_data_length = msg->fmt2.count2;
+	return 0;
+}
+
 static int convert_response_ica(struct zcrypt_device *zdev,
 			    struct ap_message *reply,
 			    char __user *outputdata,
@@ -391,6 +533,36 @@ static int convert_response_ica(struct zcrypt_device *zdev,
 	}
 }
 
+static int convert_response_xcrb(struct zcrypt_device *zdev,
+			    struct ap_message *reply,
+			    struct ica_xcRB *xcRB)
+{
+	struct type86x_reply *msg = reply->message;
+
+	/* Response type byte is the second byte in the response. */
+	switch (((unsigned char *) reply->message)[1]) {
+	case TYPE82_RSP_CODE:
+	case TYPE88_RSP_CODE:
+		xcRB->status = 0x0008044DL; /* HDD_InvalidParm */
+		return convert_error(zdev, reply);
+	case TYPE86_RSP_CODE:
+		if (msg->hdr.reply_code) {
+			memcpy(&(xcRB->status), msg->fmt2.apfs, sizeof(u32));
+			return convert_error(zdev, reply);
+		}
+		if (msg->cprbx.cprb_ver_id == 0x02)
+			return convert_type86_xcrb(zdev, reply, xcRB);
+		/* no break, incorrect cprb version is an unknown response */
+	default: /* Unknown response type, this should NEVER EVER happen */
+		PRINTK("Unrecognized Message Header: %08x%08x\n",
+		       *(unsigned int *) reply->message,
+		       *(unsigned int *) (reply->message+4));
+		xcRB->status = 0x0008044DL; /* HDD_InvalidParm */
+		zdev->online = 0;
+		return -EAGAIN;	/* repeat the request on a different device. */
+	}
+}
+
 /**
  * This function is called from the AP bus code after a crypto request
  * "msg" has finished with the reply message "reply".
@@ -407,6 +579,8 @@ static void zcrypt_pcixcc_receive(struct ap_device *ap_dev,
 		.type = TYPE82_RSP_CODE,
 		.reply_code = REP82_ERROR_MACHINE_FAILURE,
 	};
+	struct response_type *resp_type =
+		(struct response_type *) msg->private;
 	struct type86x_reply *t86r = reply->message;
 	int length;
 
@@ -415,12 +589,27 @@ static void zcrypt_pcixcc_receive(struct ap_device *ap_dev,
 		memcpy(msg->message, &error_reply, sizeof(error_reply));
 	else if (t86r->hdr.type == TYPE86_RSP_CODE &&
 		 t86r->cprbx.cprb_ver_id == 0x02) {
-		length = sizeof(struct type86x_reply) + t86r->length - 2;
-		length = min(PCIXCC_MAX_ICA_RESPONSE_SIZE, length);
-		memcpy(msg->message, reply->message, length);
+		switch (resp_type->type) {
+		case PCIXCC_RESPONSE_TYPE_ICA:
+			length = sizeof(struct type86x_reply)
+				+ t86r->length - 2;
+			length = min(PCIXCC_MAX_ICA_RESPONSE_SIZE, length);
+			memcpy(msg->message, reply->message, length);
+			break;
+		case PCIXCC_RESPONSE_TYPE_XCRB:
+			length = t86r->fmt2.offset2 + t86r->fmt2.count2;
+			length = min(PCIXCC_MAX_XCRB_RESPONSE_SIZE, length);
+			memcpy(msg->message, reply->message, length);
+			break;
+		default:
+			PRINTK("Invalid internal response type: %i\n",
+			    resp_type->type);
+			memcpy(msg->message, &error_reply,
+			    sizeof error_reply);
+		}
 	} else
 		memcpy(msg->message, reply->message, sizeof error_reply);
-	complete((struct completion *) msg->private);
+	complete(&(resp_type->work));
 }
 
 static atomic_t zcrypt_step = ATOMIC_INIT(0);
@@ -436,7 +625,9 @@ static long zcrypt_pcixcc_modexpo(struct zcrypt_device *zdev,
 				  struct ica_rsa_modexpo *mex)
 {
 	struct ap_message ap_msg;
-	struct completion work;
+	struct response_type resp_type = {
+		.type = PCIXCC_RESPONSE_TYPE_ICA,
+	};
 	int rc;
 
 	ap_msg.message = (void *) get_zeroed_page(GFP_KERNEL);
@@ -444,14 +635,14 @@ static long zcrypt_pcixcc_modexpo(struct zcrypt_device *zdev,
 		return -ENOMEM;
 	ap_msg.psmid = (((unsigned long long) current->pid) << 32) +
 				atomic_inc_return(&zcrypt_step);
-	ap_msg.private = &work;
+	ap_msg.private = &resp_type;
 	rc = ICAMEX_msg_to_type6MEX_msgX(zdev, &ap_msg, mex);
 	if (rc)
 		goto out_free;
-	init_completion(&work);
+	init_completion(&resp_type.work);
 	ap_queue_message(zdev->ap_dev, &ap_msg);
 	rc = wait_for_completion_interruptible_timeout(
-				&work, PCIXCC_CLEANUP_TIME);
+				&resp_type.work, PCIXCC_CLEANUP_TIME);
 	if (rc > 0)
 		rc = convert_response_ica(zdev, &ap_msg, mex->outputdata,
 					  mex->outputdatalength);
@@ -478,7 +669,9 @@ static long zcrypt_pcixcc_modexpo_crt(struct zcrypt_device *zdev,
 				      struct ica_rsa_modexpo_crt *crt)
 {
 	struct ap_message ap_msg;
-	struct completion work;
+	struct response_type resp_type = {
+		.type = PCIXCC_RESPONSE_TYPE_ICA,
+	};
 	int rc;
 
 	ap_msg.message = (void *) get_zeroed_page(GFP_KERNEL);
@@ -486,14 +679,14 @@ static long zcrypt_pcixcc_modexpo_crt(struct zcrypt_device *zdev,
 		return -ENOMEM;
 	ap_msg.psmid = (((unsigned long long) current->pid) << 32) +
 				atomic_inc_return(&zcrypt_step);
-	ap_msg.private = &work;
+	ap_msg.private = &resp_type;
 	rc = ICACRT_msg_to_type6CRT_msgX(zdev, &ap_msg, crt);
 	if (rc)
 		goto out_free;
-	init_completion(&work);
+	init_completion(&resp_type.work);
 	ap_queue_message(zdev->ap_dev, &ap_msg);
 	rc = wait_for_completion_interruptible_timeout(
-				&work, PCIXCC_CLEANUP_TIME);
+				&resp_type.work, PCIXCC_CLEANUP_TIME);
 	if (rc > 0)
 		rc = convert_response_ica(zdev, &ap_msg, crt->outputdata,
 					  crt->outputdatalength);
@@ -509,12 +702,56 @@ out_free:
 	return rc;
 }
 
+/**
+ * The request distributor calls this function if it picked the PCIXCC/CEX2C
+ * device to handle a send_cprb request.
+ * @zdev: pointer to zcrypt_device structure that identifies the
+ *	  PCIXCC/CEX2C device to the request distributor
+ * @xcRB: pointer to the send_cprb request buffer
+ */
+long zcrypt_pcixcc_send_cprb(struct zcrypt_device *zdev, struct ica_xcRB *xcRB)
+{
+	struct ap_message ap_msg;
+	struct response_type resp_type = {
+		.type = PCIXCC_RESPONSE_TYPE_XCRB,
+	};
+	int rc;
+
+	ap_msg.message = (void *) kmalloc(PCIXCC_MAX_XCRB_MESSAGE_SIZE, GFP_KERNEL);
+	if (!ap_msg.message)
+		return -ENOMEM;
+	ap_msg.psmid = (((unsigned long long) current->pid) << 32) +
+				atomic_inc_return(&zcrypt_step);
+	ap_msg.private = &resp_type;
+	rc = XCRB_msg_to_type6CPRB_msgX(zdev, &ap_msg, xcRB);
+	if (rc)
+		goto out_free;
+	init_completion(&resp_type.work);
+	ap_queue_message(zdev->ap_dev, &ap_msg);
+	rc = wait_for_completion_interruptible_timeout(
+				&resp_type.work, PCIXCC_CLEANUP_TIME);
+	if (rc > 0)
+		rc = convert_response_xcrb(zdev, &ap_msg, xcRB);
+	else {
+		/* Signal pending or message timed out. */
+		ap_cancel_message(zdev->ap_dev, &ap_msg);
+		if (rc == 0)
+			/* Message timed out. */
+			rc = -ETIME;
+	}
+out_free:
+	memset(ap_msg.message, 0x0, ap_msg.length);
+	kfree(ap_msg.message);
+	return rc;
+}
+
 /**
  * The crypto operations for a PCIXCC/CEX2C card.
  */
 static struct zcrypt_ops zcrypt_pcixcc_ops = {
 	.rsa_modexpo = zcrypt_pcixcc_modexpo,
 	.rsa_modexpo_crt = zcrypt_pcixcc_modexpo_crt,
+	.send_cprb = zcrypt_pcixcc_send_cprb,
 };
 
 /**
diff --git a/drivers/s390/crypto/zcrypt_pcixcc.h b/drivers/s390/crypto/zcrypt_pcixcc.h
index d4c44c4d7ad0..a78ff307fd19 100644
--- a/drivers/s390/crypto/zcrypt_pcixcc.h
+++ b/drivers/s390/crypto/zcrypt_pcixcc.h
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/s390/crypto/zcrypt_pcixcc.h
  *
- *  zcrypt 2.0.0
+ *  zcrypt 2.1.0
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
diff --git a/include/asm-s390/zcrypt.h b/include/asm-s390/zcrypt.h
index 0d6a3e2a3349..7244c68464f2 100644
--- a/include/asm-s390/zcrypt.h
+++ b/include/asm-s390/zcrypt.h
@@ -1,7 +1,7 @@
 /*
  *  include/asm-s390/zcrypt.h
  *
- *  zcrypt 2.0.0 (user-visible header)
+ *  zcrypt 2.1.0 (user-visible header)
  *
  *  Copyright (C)  2001, 2006 IBM Corporation
  *  Author(s): Robert Burroughs
@@ -79,6 +79,83 @@ struct ica_rsa_modexpo_crt {
 	char __user *	u_mult_inv;
 };
 
+/**
+ * CPRBX
+ *	  Note that all shorts and ints are big-endian.
+ *	  All pointer fields are 16 bytes long, and mean nothing.
+ *
+ *	  A request CPRB is followed by a request_parameter_block.
+ *
+ *	  The request (or reply) parameter block is organized thus:
+ *	    function code
+ *	    VUD block
+ *	    key block
+ */
+struct ica_CPRBX {
+	unsigned short	cprb_len;	/* CPRB length	      220	 */
+	unsigned char	cprb_ver_id;	/* CPRB version id.   0x02	 */
+	unsigned char	pad_000[3];	/* Alignment pad bytes		 */
+	unsigned char	func_id[2];	/* function id	      0x5432	 */
+	unsigned char	cprb_flags[4];	/* Flags			 */
+	unsigned int	req_parml;	/* request parameter buffer len	 */
+	unsigned int	req_datal;	/* request data buffer		 */
+	unsigned int	rpl_msgbl;	/* reply  message block length	 */
+	unsigned int	rpld_parml;	/* replied parameter block len	 */
+	unsigned int	rpl_datal;	/* reply data block len		 */
+	unsigned int	rpld_datal;	/* replied data block len	 */
+	unsigned int	req_extbl;	/* request extension block len	 */
+	unsigned char	pad_001[4];	/* reserved			 */
+	unsigned int	rpld_extbl;	/* replied extension block len	 */
+	unsigned char	padx000[16 - sizeof (char *)];
+	unsigned char *	req_parmb;	/* request parm block 'address'	 */
+	unsigned char	padx001[16 - sizeof (char *)];
+	unsigned char *	req_datab;	/* request data block 'address'	 */
+	unsigned char	padx002[16 - sizeof (char *)];
+	unsigned char *	rpl_parmb;	/* reply parm block 'address'	 */
+	unsigned char	padx003[16 - sizeof (char *)];
+	unsigned char *	rpl_datab;	/* reply data block 'address'	 */
+	unsigned char	padx004[16 - sizeof (char *)];
+	unsigned char *	req_extb;	/* request extension block 'addr'*/
+	unsigned char	padx005[16 - sizeof (char *)];
+	unsigned char *	rpl_extb;	/* reply extension block 'addres'*/
+	unsigned short	ccp_rtcode;	/* server return code		 */
+	unsigned short	ccp_rscode;	/* server reason code		 */
+	unsigned int	mac_data_len;	/* Mac Data Length		 */
+	unsigned char	logon_id[8];	/* Logon Identifier		 */
+	unsigned char	mac_value[8];	/* Mac Value			 */
+	unsigned char	mac_content_flgs;/* Mac content flag byte	 */
+	unsigned char	pad_002;	/* Alignment			 */
+	unsigned short	domain;		/* Domain			 */
+	unsigned char	usage_domain[4];/* Usage domain			 */
+	unsigned char	cntrl_domain[4];/* Control domain		 */
+	unsigned char	S390enf_mask[4];/* S/390 enforcement mask	 */
+	unsigned char	pad_004[36];	/* reserved			 */
+};
+
+/**
+ * xcRB
+ */
+struct ica_xcRB {
+	unsigned short	agent_ID;
+	unsigned int	user_defined;
+	unsigned short	request_ID;
+	unsigned int	request_control_blk_length;
+	unsigned char	padding1[16 - sizeof (char *)];
+	char __user *	request_control_blk_addr;
+	unsigned int	request_data_length;
+	char		padding2[16 - sizeof (char *)];
+	char __user *	request_data_address;
+	unsigned int	reply_control_blk_length;
+	char		padding3[16 - sizeof (char *)];
+	char __user *	reply_control_blk_addr;
+	unsigned int	reply_data_length;
+	char		padding4[16 - sizeof (char *)];
+	char __user *	reply_data_addr;
+	unsigned short	priority_window;
+	unsigned int	status;
+} __attribute__((packed));
+#define AUTOSELECT ((unsigned int)0xFFFFFFFF)
+
 #define ZCRYPT_IOCTL_MAGIC 'z'
 
 /**
@@ -187,6 +264,7 @@ struct ica_rsa_modexpo_crt {
  */
 #define ICARSAMODEXPO	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x05, 0)
 #define ICARSACRT	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x06, 0)
+#define ZSECSENDCPRB	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x81, 0)
 
 /* New status calls */
 #define Z90STAT_TOTALCOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x40, int)
-- 
cgit v1.2.3


From 4ba069b802c29eee066385f9826e2d83716626b4 Mon Sep 17 00:00:00 2001
From: Michael Grundy <grundym@us.ibm.com>
Date: Wed, 20 Sep 2006 15:58:39 +0200
Subject: [S390] add kprobes support.

Signed-off-by: Michael Grundy <grundym@us.ibm.com>
Signed-off-by: David Wilder <dwilder@us.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig              |  14 +
 arch/s390/kernel/Makefile      |   1 +
 arch/s390/kernel/entry.S       |  12 +
 arch/s390/kernel/entry64.S     |  12 +
 arch/s390/kernel/kprobes.c     | 657 +++++++++++++++++++++++++++++++++++++++++
 arch/s390/kernel/traps.c       |  31 +-
 arch/s390/kernel/vmlinux.lds.S |   1 +
 arch/s390/mm/fault.c           |  40 ++-
 include/asm-s390/kdebug.h      |  59 ++++
 include/asm-s390/kprobes.h     | 114 +++++++
 10 files changed, 937 insertions(+), 4 deletions(-)
 create mode 100644 arch/s390/kernel/kprobes.c
 create mode 100644 include/asm-s390/kdebug.h
 create mode 100644 include/asm-s390/kprobes.h

(limited to 'include')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 2f4f70c4dbb2..76122ce1e6cb 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -487,8 +487,22 @@ source "drivers/net/Kconfig"
 
 source "fs/Kconfig"
 
+menu "Instrumentation Support"
+
 source "arch/s390/oprofile/Kconfig"
 
+config KPROBES
+	bool "Kprobes (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && MODULES
+	help
+	  Kprobes allows you to trap at almost any kernel address and
+	  execute a callback function.	register_kprobe() establishes
+	  a probepoint and specifies the callback.  Kprobes is useful
+	  for kernel debugging, non-intrusive instrumentation and testing.
+	  If in doubt, say "N".
+
+endmenu
+
 source "arch/s390/Kconfig.debug"
 
 source "security/Kconfig"
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 9a33ed6ca696..33a5069c0e16 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -24,6 +24,7 @@ obj-$(CONFIG_COMPAT)		+= compat_linux.o compat_signal.o \
 
 obj-$(CONFIG_VIRT_TIMER)	+= vtime.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o
 
 # Kexec part
 S390_KEXEC_OBJS := machine_kexec.o crash.o
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 5b5799ac8f83..0c712b78a7e8 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -505,6 +505,8 @@ pgm_no_vtime2:
 	mvc	__THREAD_per+__PER_address(4,%r1),__LC_PER_ADDRESS
 	mvc	__THREAD_per+__PER_access_id(1,%r1),__LC_PER_ACCESS_ID
 	oi	__TI_flags+3(%r9),_TIF_SINGLE_STEP # set TIF_SINGLE_STEP
+	tm	SP_PSW+1(%r15),0x01	# kernel per event ?
+	bz	BASED(kernel_per)
 	l	%r3,__LC_PGM_ILC	 # load program interruption code
 	la	%r8,0x7f
 	nr	%r8,%r3                  # clear per-event-bit and ilc
@@ -536,6 +538,16 @@ pgm_no_vtime3:
 	stosm	__SF_EMPTY(%r15),0x03	# reenable interrupts
 	b	BASED(sysc_do_svc)
 
+#
+# per was called from kernel, must be kprobes
+#
+kernel_per:
+	mvi	SP_TRAP+1(%r15),0x28	# set trap indication to pgm check
+	la	%r2,SP_PTREGS(%r15)	# address of register-save area
+	l	%r1,BASED(.Lhandle_per)	# load adr. of per handler
+	la	%r14,BASED(sysc_leave)	# load adr. of system return
+	br	%r1			# branch to do_single_step
+
 /*
  * IO interrupt handler routine
  */
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index 56f5f613b868..8b956d1538f5 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -518,6 +518,8 @@ pgm_no_vtime2:
 #endif
 	lg	%r9,__LC_THREAD_INFO	# load pointer to thread_info struct
 	lg	%r1,__TI_task(%r9)
+	tm	SP_PSW+1(%r15),0x01	# kernel per event ?
+	jz	kernel_per
 	mvc	__THREAD_per+__PER_atmid(2,%r1),__LC_PER_ATMID
 	mvc	__THREAD_per+__PER_address(8,%r1),__LC_PER_ADDRESS
 	mvc	__THREAD_per+__PER_access_id(1,%r1),__LC_PER_ACCESS_ID
@@ -553,6 +555,16 @@ pgm_no_vtime3:
 	stosm	__SF_EMPTY(%r15),0x03	# reenable interrupts
 	j	sysc_do_svc
 
+#
+# per was called from kernel, must be kprobes
+#
+kernel_per:
+	lhi	%r0,__LC_PGM_OLD_PSW
+	sth	%r0,SP_TRAP(%r15)	# set trap indication to pgm check
+	la	%r2,SP_PTREGS(%r15)	# address of register-save area
+	larl	%r14,sysc_leave		# load adr. of system ret, no work
+	jg	do_single_step		# branch to do_single_step
+
 /*
  * IO interrupt handler routine
  */
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
new file mode 100644
index 000000000000..ca28fb0b3790
--- /dev/null
+++ b/arch/s390/kernel/kprobes.c
@@ -0,0 +1,657 @@
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2006
+ *
+ * s390 port, used ppc64 as template. Mike Grundy <grundym@us.ibm.com>
+ */
+
+#include <linux/config.h>
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+#include <linux/preempt.h>
+#include <linux/stop_machine.h>
+#include <asm/cacheflush.h>
+#include <asm/kdebug.h>
+#include <asm/sections.h>
+#include <asm/uaccess.h>
+#include <linux/module.h>
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	/* Make sure the probe isn't going on a difficult instruction */
+	if (is_prohibited_opcode((kprobe_opcode_t *) p->addr))
+		return -EINVAL;
+
+	if ((unsigned long)p->addr & 0x01) {
+		printk("Attempt to register kprobe at an unaligned address\n");
+		return -EINVAL;
+		}
+
+	/* Use the get_insn_slot() facility for correctness */
+	if (!(p->ainsn.insn = get_insn_slot()))
+		return -ENOMEM;
+
+	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+
+	get_instruction_type(&p->ainsn);
+	p->opcode = *p->addr;
+	return 0;
+}
+
+int __kprobes is_prohibited_opcode(kprobe_opcode_t *instruction)
+{
+	switch (*(__u8 *) instruction) {
+	case 0x0c:	/* bassm */
+	case 0x0b:	/* bsm	 */
+	case 0x83:	/* diag  */
+	case 0x44:	/* ex	 */
+		return -EINVAL;
+	}
+	switch (*(__u16 *) instruction) {
+	case 0x0101:	/* pr	 */
+	case 0xb25a:	/* bsa	 */
+	case 0xb240:	/* bakr  */
+	case 0xb258:	/* bsg	 */
+	case 0xb218:	/* pc	 */
+	case 0xb228:	/* pt	 */
+		return -EINVAL;
+	}
+	return 0;
+}
+
+void __kprobes get_instruction_type(struct arch_specific_insn *ainsn)
+{
+	/* default fixup method */
+	ainsn->fixup = FIXUP_PSW_NORMAL;
+
+	/* save r1 operand */
+	ainsn->reg = (*ainsn->insn & 0xf0) >> 4;
+
+	/* save the instruction length (pop 5-5) in bytes */
+	switch (*(__u8 *) (ainsn->insn) >> 4) {
+	case 0:
+		ainsn->ilen = 2;
+		break;
+	case 1:
+	case 2:
+		ainsn->ilen = 4;
+		break;
+	case 3:
+		ainsn->ilen = 6;
+		break;
+	}
+
+	switch (*(__u8 *) ainsn->insn) {
+	case 0x05:	/* balr	*/
+	case 0x0d:	/* basr */
+		ainsn->fixup = FIXUP_RETURN_REGISTER;
+		/* if r2 = 0, no branch will be taken */
+		if ((*ainsn->insn & 0x0f) == 0)
+			ainsn->fixup |= FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0x06:	/* bctr	*/
+	case 0x07:	/* bcr	*/
+		ainsn->fixup = FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0x45:	/* bal	*/
+	case 0x4d:	/* bas	*/
+		ainsn->fixup = FIXUP_RETURN_REGISTER;
+		break;
+	case 0x47:	/* bc	*/
+	case 0x46:	/* bct	*/
+	case 0x86:	/* bxh	*/
+	case 0x87:	/* bxle	*/
+		ainsn->fixup = FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0x82:	/* lpsw	*/
+		ainsn->fixup = FIXUP_NOT_REQUIRED;
+		break;
+	case 0xb2:	/* lpswe */
+		if (*(((__u8 *) ainsn->insn) + 1) == 0xb2) {
+			ainsn->fixup = FIXUP_NOT_REQUIRED;
+		}
+		break;
+	case 0xa7:	/* bras	*/
+		if ((*ainsn->insn & 0x0f) == 0x05) {
+			ainsn->fixup |= FIXUP_RETURN_REGISTER;
+		}
+		break;
+	case 0xc0:
+		if ((*ainsn->insn & 0x0f) == 0x00  /* larl  */
+			|| (*ainsn->insn & 0x0f) == 0x05) /* brasl */
+		ainsn->fixup |= FIXUP_RETURN_REGISTER;
+		break;
+	case 0xeb:
+		if (*(((__u8 *) ainsn->insn) + 5 ) == 0x44 ||	/* bxhg  */
+			*(((__u8 *) ainsn->insn) + 5) == 0x45) {/* bxleg */
+			ainsn->fixup = FIXUP_BRANCH_NOT_TAKEN;
+		}
+		break;
+	case 0xe3:	/* bctg	*/
+		if (*(((__u8 *) ainsn->insn) + 5) == 0x46) {
+			ainsn->fixup = FIXUP_BRANCH_NOT_TAKEN;
+		}
+		break;
+	}
+}
+
+static int __kprobes swap_instruction(void *aref)
+{
+	struct ins_replace_args *args = aref;
+	int err = -EFAULT;
+
+	asm volatile(
+		"0: mvc  0(2,%2),0(%3)\n"
+		"1: la   %0,0\n"
+		"2:\n"
+		EX_TABLE(0b,2b)
+		: "+d" (err), "=m" (*args->ptr)
+		: "a" (args->ptr), "a" (&args->new), "m" (args->new));
+	return err;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	unsigned long status = kcb->kprobe_status;
+	struct ins_replace_args args;
+
+	args.ptr = p->addr;
+	args.old = p->opcode;
+	args.new = BREAKPOINT_INSTRUCTION;
+
+	kcb->kprobe_status = KPROBE_SWAP_INST;
+	stop_machine_run(swap_instruction, &args, NR_CPUS);
+	kcb->kprobe_status = status;
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	unsigned long status = kcb->kprobe_status;
+	struct ins_replace_args args;
+
+	args.ptr = p->addr;
+	args.old = BREAKPOINT_INSTRUCTION;
+	args.new = p->opcode;
+
+	kcb->kprobe_status = KPROBE_SWAP_INST;
+	stop_machine_run(swap_instruction, &args, NR_CPUS);
+	kcb->kprobe_status = status;
+}
+
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	mutex_lock(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	mutex_unlock(&kprobe_mutex);
+}
+
+static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+{
+	per_cr_bits kprobe_per_regs[1];
+
+	memset(kprobe_per_regs, 0, sizeof(per_cr_bits));
+	regs->psw.addr = (unsigned long)p->ainsn.insn | PSW_ADDR_AMODE;
+
+	/* Set up the per control reg info, will pass to lctl */
+	kprobe_per_regs[0].em_instruction_fetch = 1;
+	kprobe_per_regs[0].starting_addr = (unsigned long)p->ainsn.insn;
+	kprobe_per_regs[0].ending_addr = (unsigned long)p->ainsn.insn + 1;
+
+	/* Set the PER control regs, turns on single step for this address */
+	__ctl_load(kprobe_per_regs, 9, 11);
+	regs->psw.mask |= PSW_MASK_PER;
+	regs->psw.mask &= ~(PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK);
+}
+
+static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	kcb->prev_kprobe.kp = kprobe_running();
+	kcb->prev_kprobe.status = kcb->kprobe_status;
+	kcb->prev_kprobe.kprobe_saved_imask = kcb->kprobe_saved_imask;
+	memcpy(kcb->prev_kprobe.kprobe_saved_ctl, kcb->kprobe_saved_ctl,
+					sizeof(kcb->kprobe_saved_ctl));
+}
+
+static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	__get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
+	kcb->kprobe_status = kcb->prev_kprobe.status;
+	kcb->kprobe_saved_imask = kcb->prev_kprobe.kprobe_saved_imask;
+	memcpy(kcb->kprobe_saved_ctl, kcb->prev_kprobe.kprobe_saved_ctl,
+					sizeof(kcb->kprobe_saved_ctl));
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
+						struct kprobe_ctlblk *kcb)
+{
+	__get_cpu_var(current_kprobe) = p;
+	/* Save the interrupt and per flags */
+	kcb->kprobe_saved_imask = regs->psw.mask &
+	    (PSW_MASK_PER | PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK);
+	/* Save the control regs that govern PER */
+	__ctl_store(kcb->kprobe_saved_ctl, 9, 11);
+}
+
+/* Called with kretprobe_lock held */
+void __kprobes arch_prepare_kretprobe(struct kretprobe *rp,
+					struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *) regs->gprs[14];
+
+		/* Replace the return addr with trampoline addr */
+		regs->gprs[14] = (unsigned long)&kretprobe_trampoline;
+
+		add_rp_inst(ri);
+	} else {
+		rp->nmissed++;
+	}
+}
+
+static int __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p;
+	int ret = 0;
+	unsigned long *addr = (unsigned long *)
+		((regs->psw.addr & PSW_ADDR_INSN) - 2);
+	struct kprobe_ctlblk *kcb;
+
+	/*
+	 * We don't want to be preempted for the entire
+	 * duration of kprobe processing
+	 */
+	preempt_disable();
+	kcb = get_kprobe_ctlblk();
+
+	/* Check we're not actually recursing */
+	if (kprobe_running()) {
+		p = get_kprobe(addr);
+		if (p) {
+			if (kcb->kprobe_status == KPROBE_HIT_SS &&
+			    *p->ainsn.insn == BREAKPOINT_INSTRUCTION) {
+				regs->psw.mask &= ~PSW_MASK_PER;
+				regs->psw.mask |= kcb->kprobe_saved_imask;
+				goto no_kprobe;
+			}
+			/* We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe(kcb);
+			set_current_kprobe(p, regs, kcb);
+			kprobes_inc_nmissed_count(p);
+			prepare_singlestep(p, regs);
+			kcb->kprobe_status = KPROBE_REENTER;
+			return 1;
+		} else {
+			p = __get_cpu_var(current_kprobe);
+			if (p->break_handler && p->break_handler(p, regs)) {
+				goto ss_probe;
+			}
+		}
+		goto no_kprobe;
+	}
+
+	p = get_kprobe(addr);
+	if (!p) {
+		if (*addr != BREAKPOINT_INSTRUCTION) {
+			/*
+			 * The breakpoint instruction was removed right
+			 * after we hit it.  Another cpu has removed
+			 * either a probepoint or a debugger breakpoint
+			 * at this address.  In either case, no further
+			 * handling of this interrupt is appropriate.
+			 *
+			 */
+			ret = 1;
+		}
+		/* Not one of ours: let kernel handle it */
+		goto no_kprobe;
+	}
+
+	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+	set_current_kprobe(p, regs, kcb);
+	if (p->pre_handler && p->pre_handler(p, regs))
+		/* handler has already set things up, so skip ss setup */
+		return 1;
+
+ss_probe:
+	prepare_singlestep(p, regs);
+	kcb->kprobe_status = KPROBE_HIT_SS;
+	return 1;
+
+no_kprobe:
+	preempt_enable_no_resched();
+	return ret;
+}
+
+/*
+ * Function return probe trampoline:
+ *	- init_kprobes() establishes a probepoint here
+ *	- When the probed function returns, this probe
+ *		causes the handlers to fire
+ */
+void __kprobes kretprobe_trampoline_holder(void)
+{
+	asm volatile(".global kretprobe_trampoline\n"
+		     "kretprobe_trampoline: bcr 0,0\n");
+}
+
+/*
+ * Called when the probe at kretprobe trampoline is hit
+ */
+int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri = NULL;
+	struct hlist_head *head;
+	struct hlist_node *node, *tmp;
+	unsigned long flags, orig_ret_address = 0;
+	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
+
+	spin_lock_irqsave(&kretprobe_lock, flags);
+	head = kretprobe_inst_table_head(current);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+	 *	 function, the first instance's ret_addr will point to the
+	 *	 real return address, and all the rest will point to
+	 *	 kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+		if (ri->task != current)
+			/* another task is sharing our hash bucket */
+			continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address) {
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+		}
+	}
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->psw.addr = orig_ret_address | PSW_ADDR_AMODE;
+
+	reset_current_kprobe();
+	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	preempt_enable_no_resched();
+
+	/*
+	 * By returning a non-zero value, we are telling
+	 * kprobe_handler() that we don't want the post_handler
+	 * to run (and have re-enabled preemption)
+	 */
+	return 1;
+}
+
+/*
+ * Called after single-stepping.  p->addr is the address of the
+ * instruction whose first byte has been replaced by the "breakpoint"
+ * instruction.  To avoid the SMP problems that can occur when we
+ * temporarily put back the original opcode to single-step, we
+ * single-stepped a copy of the instruction.  The address of this
+ * copy is p->ainsn.insn.
+ */
+static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	regs->psw.addr &= PSW_ADDR_INSN;
+
+	if (p->ainsn.fixup & FIXUP_PSW_NORMAL)
+		regs->psw.addr = (unsigned long)p->addr +
+				((unsigned long)regs->psw.addr -
+				 (unsigned long)p->ainsn.insn);
+
+	if (p->ainsn.fixup & FIXUP_BRANCH_NOT_TAKEN)
+		if ((unsigned long)regs->psw.addr -
+		    (unsigned long)p->ainsn.insn == p->ainsn.ilen)
+			regs->psw.addr = (unsigned long)p->addr + p->ainsn.ilen;
+
+	if (p->ainsn.fixup & FIXUP_RETURN_REGISTER)
+		regs->gprs[p->ainsn.reg] = ((unsigned long)p->addr +
+						(regs->gprs[p->ainsn.reg] -
+						(unsigned long)p->ainsn.insn))
+						| PSW_ADDR_AMODE;
+
+	regs->psw.addr |= PSW_ADDR_AMODE;
+	/* turn off PER mode */
+	regs->psw.mask &= ~PSW_MASK_PER;
+	/* Restore the original per control regs */
+	__ctl_load(kcb->kprobe_saved_ctl, 9, 11);
+	regs->psw.mask |= kcb->kprobe_saved_imask;
+}
+
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	if (!cur)
+		return 0;
+
+	if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
+		kcb->kprobe_status = KPROBE_HIT_SSDONE;
+		cur->post_handler(cur, regs, 0);
+	}
+
+	resume_execution(cur, regs);
+
+	/*Restore back the original saved kprobes variables and continue. */
+	if (kcb->kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe(kcb);
+		goto out;
+	}
+	reset_current_kprobe();
+out:
+	preempt_enable_no_resched();
+
+	/*
+	 * if somebody else is singlestepping across a probe point, psw mask
+	 * will have PER set, in which case, continue the remaining processing
+	 * of do_single_step, as if this is not a probe hit.
+	 */
+	if (regs->psw.mask & PSW_MASK_PER) {
+		return 0;
+	}
+
+	return 1;
+}
+
+static int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	const struct exception_table_entry *entry;
+
+	switch(kcb->kprobe_status) {
+	case KPROBE_SWAP_INST:
+		/* We are here because the instruction replacement failed */
+		return 0;
+	case KPROBE_HIT_SS:
+	case KPROBE_REENTER:
+		/*
+		 * We are here because the instruction being single
+		 * stepped caused a page fault. We reset the current
+		 * kprobe and the nip points back to the probe address
+		 * and allow the page fault handler to continue as a
+		 * normal page fault.
+		 */
+		regs->psw.addr = (unsigned long)cur->addr | PSW_ADDR_AMODE;
+		regs->psw.mask &= ~PSW_MASK_PER;
+		regs->psw.mask |= kcb->kprobe_saved_imask;
+		if (kcb->kprobe_status == KPROBE_REENTER)
+			restore_previous_kprobe(kcb);
+		else
+			reset_current_kprobe();
+		preempt_enable_no_resched();
+		break;
+	case KPROBE_HIT_ACTIVE:
+	case KPROBE_HIT_SSDONE:
+		/*
+		 * We increment the nmissed count for accounting,
+		 * we can also use npre/npostfault count for accouting
+		 * these specific fault cases.
+		 */
+		kprobes_inc_nmissed_count(cur);
+
+		/*
+		 * We come here because instructions in the pre/post
+		 * handler caused the page_fault, this could happen
+		 * if handler tries to access user space by
+		 * copy_from_user(), get_user() etc. Let the
+		 * user-specified handler try to fix it first.
+		 */
+		if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
+			return 1;
+
+		/*
+		 * In case the user-specified fault handler returned
+		 * zero, try to fix up.
+		 */
+		entry = search_exception_tables(regs->psw.addr & PSW_ADDR_INSN);
+		if (entry) {
+			regs->psw.addr = entry->fixup | PSW_ADDR_AMODE;
+			return 1;
+		}
+
+		/*
+		 * fixup_exception() could not handle it,
+		 * Let do_page_fault() fix it.
+		 */
+		break;
+	default:
+		break;
+	}
+	return 0;
+}
+
+/*
+ * Wrapper routine to for handling exceptions.
+ */
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	switch (val) {
+	case DIE_BPT:
+		if (kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_SSTEP:
+		if (post_kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_TRAP:
+	case DIE_PAGE_FAULT:
+		/* kprobe_running() needs smp_processor_id() */
+		preempt_disable();
+		if (kprobe_running() &&
+		    kprobe_fault_handler(args->regs, args->trapnr))
+			ret = NOTIFY_STOP;
+		preempt_enable();
+		break;
+	default:
+		break;
+	}
+	return ret;
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+	unsigned long addr;
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	memcpy(&kcb->jprobe_saved_regs, regs, sizeof(struct pt_regs));
+
+	/* setup return addr to the jprobe handler routine */
+	regs->psw.addr = (unsigned long)(jp->entry) | PSW_ADDR_AMODE;
+
+	/* r14 is the function return address */
+	kcb->jprobe_saved_r14 = (unsigned long)regs->gprs[14];
+	/* r15 is the stack pointer */
+	kcb->jprobe_saved_r15 = (unsigned long)regs->gprs[15];
+	addr = (unsigned long)kcb->jprobe_saved_r15;
+
+	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *) addr,
+	       MIN_STACK_SIZE(addr));
+	return 1;
+}
+
+void __kprobes jprobe_return(void)
+{
+	asm volatile(".word 0x0002");
+}
+
+void __kprobes jprobe_return_end(void)
+{
+	asm volatile("bcr 0,0");
+}
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	unsigned long stack_addr = (unsigned long)(kcb->jprobe_saved_r15);
+
+	/* Put the regs back */
+	memcpy(regs, &kcb->jprobe_saved_regs, sizeof(struct pt_regs));
+	/* put the stack back */
+	memcpy((kprobe_opcode_t *) stack_addr, kcb->jprobes_stack,
+	       MIN_STACK_SIZE(stack_addr));
+	preempt_enable_no_resched();
+	return 1;
+}
+
+static struct kprobe trampoline_p = {
+	.addr = (kprobe_opcode_t *) & kretprobe_trampoline,
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init_kprobes(void)
+{
+	return register_kprobe(&trampoline_p);
+}
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index bde1d1d59858..c4982c963424 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 #include <linux/reboot.h>
+#include <linux/kprobes.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -39,6 +40,7 @@
 #include <asm/s390_ext.h>
 #include <asm/lowcore.h>
 #include <asm/debug.h>
+#include <asm/kdebug.h>
 
 /* Called from entry.S only */
 extern void handle_per_exception(struct pt_regs *regs);
@@ -74,6 +76,20 @@ static int kstack_depth_to_print = 12;
 static int kstack_depth_to_print = 20;
 #endif /* CONFIG_64BIT */
 
+ATOMIC_NOTIFIER_HEAD(s390die_chain);
+
+int register_die_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_register(&s390die_chain, nb);
+}
+EXPORT_SYMBOL(register_die_notifier);
+
+int unregister_die_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_unregister(&s390die_chain, nb);
+}
+EXPORT_SYMBOL(unregister_die_notifier);
+
 /*
  * For show_trace we have tree different stack to consider:
  *   - the panic stack which is used if the kernel stack has overflown
@@ -305,8 +321,9 @@ report_user_fault(long interruption_code, struct pt_regs *regs)
 #endif
 }
 
-static void inline do_trap(long interruption_code, int signr, char *str,
-                           struct pt_regs *regs, siginfo_t *info)
+static void __kprobes inline do_trap(long interruption_code, int signr,
+					char *str, struct pt_regs *regs,
+					siginfo_t *info)
 {
 	/*
 	 * We got all needed information from the lowcore and can
@@ -315,6 +332,10 @@ static void inline do_trap(long interruption_code, int signr, char *str,
         if (regs->psw.mask & PSW_MASK_PSTATE)
 		local_irq_enable();
 
+	if (notify_die(DIE_TRAP, str, regs, interruption_code,
+				interruption_code, signr) == NOTIFY_STOP)
+		return;
+
         if (regs->psw.mask & PSW_MASK_PSTATE) {
                 struct task_struct *tsk = current;
 
@@ -336,8 +357,12 @@ static inline void __user *get_check_address(struct pt_regs *regs)
 	return (void __user *)((regs->psw.addr-S390_lowcore.pgm_ilc) & PSW_ADDR_INSN);
 }
 
-void do_single_step(struct pt_regs *regs)
+void __kprobes do_single_step(struct pt_regs *regs)
 {
+	if (notify_die(DIE_SSTEP, "sstep", regs, 0, 0,
+					SIGTRAP) == NOTIFY_STOP){
+		return;
+	}
 	if ((current->ptrace & PT_PTRACED) != 0)
 		force_sig(SIGTRAP, current);
 }
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index ff5f7bb34f75..df0c16ab8e92 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -24,6 +24,7 @@ SECTIONS
 	*(.text)
 	SCHED_TEXT
 	LOCK_TEXT
+	KPROBES_TEXT
 	*(.fixup)
 	*(.gnu.warning)
 	} = 0x0700
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 7cd82575813d..44f0cda7e72e 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -25,10 +25,12 @@
 #include <linux/console.h>
 #include <linux/module.h>
 #include <linux/hardirq.h>
+#include <linux/kprobes.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
+#include <asm/kdebug.h>
 
 #ifndef CONFIG_64BIT
 #define __FAIL_ADDR_MASK 0x7ffff000
@@ -48,6 +50,38 @@ extern int sysctl_userprocess_debug;
 
 extern void die(const char *,struct pt_regs *,long);
 
+#ifdef CONFIG_KPROBES
+ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+int register_page_fault_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
+}
+
+int unregister_page_fault_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
+}
+
+static inline int notify_page_fault(enum die_val val, const char *str,
+			struct pt_regs *regs, long err, int trap, int sig)
+{
+	struct die_args args = {
+		.regs = regs,
+		.str = str,
+		.err = err,
+		.trapnr = trap,
+		.signr = sig
+	};
+	return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
+}
+#else
+static inline int notify_page_fault(enum die_val val, const char *str,
+			struct pt_regs *regs, long err, int trap, int sig)
+{
+	return NOTIFY_DONE;
+}
+#endif
+
 extern spinlock_t timerlist_lock;
 
 /*
@@ -159,7 +193,7 @@ static void do_sigsegv(struct pt_regs *regs, unsigned long error_code,
  *   11       Page translation     ->  Not present       (nullification)
  *   3b       Region third trans.  ->  Not present       (nullification)
  */
-static inline void
+static inline void __kprobes
 do_exception(struct pt_regs *regs, unsigned long error_code, int is_protection)
 {
         struct task_struct *tsk;
@@ -173,6 +207,10 @@ do_exception(struct pt_regs *regs, unsigned long error_code, int is_protection)
         tsk = current;
         mm = tsk->mm;
 	
+	if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+					SIGSEGV) == NOTIFY_STOP)
+		return;
+
 	/* 
          * Check for low-address protection.  This needs to be treated
 	 * as a special case because the translation exception code 
diff --git a/include/asm-s390/kdebug.h b/include/asm-s390/kdebug.h
new file mode 100644
index 000000000000..40cc68025e01
--- /dev/null
+++ b/include/asm-s390/kdebug.h
@@ -0,0 +1,59 @@
+#ifndef _S390_KDEBUG_H
+#define _S390_KDEBUG_H
+
+/*
+ * Feb 2006 Ported to s390 <grundym@us.ibm.com>
+ */
+#include <linux/notifier.h>
+
+struct pt_regs;
+
+struct die_args {
+	struct pt_regs *regs;
+	const char *str;
+	long err;
+	int trapnr;
+	int signr;
+};
+
+/* Note - you should never unregister because that can race with NMIs.
+ * If you really want to do it first unregister - then synchronize_sched
+ *  - then free.
+ */
+extern int register_die_notifier(struct notifier_block *);
+extern int unregister_die_notifier(struct notifier_block *);
+extern int register_page_fault_notifier(struct notifier_block *);
+extern int unregister_page_fault_notifier(struct notifier_block *);
+extern struct atomic_notifier_head s390die_chain;
+
+
+enum die_val {
+	DIE_OOPS = 1,
+	DIE_BPT,
+	DIE_SSTEP,
+	DIE_PANIC,
+	DIE_NMI,
+	DIE_DIE,
+	DIE_NMIWATCHDOG,
+	DIE_KERNELDEBUG,
+	DIE_TRAP,
+	DIE_GPF,
+	DIE_CALL,
+	DIE_NMI_IPI,
+	DIE_PAGE_FAULT,
+};
+
+static inline int notify_die(enum die_val val, const char *str,
+			struct pt_regs *regs, long err, int trap, int sig)
+{
+	struct die_args args = {
+		.regs = regs,
+		.str = str,
+		.err = err,
+		.trapnr = trap,
+		.signr = sig
+	};
+	return atomic_notifier_call_chain(&s390die_chain, val, &args);
+}
+
+#endif
diff --git a/include/asm-s390/kprobes.h b/include/asm-s390/kprobes.h
new file mode 100644
index 000000000000..b847ff0ec3fa
--- /dev/null
+++ b/include/asm-s390/kprobes.h
@@ -0,0 +1,114 @@
+#ifndef _ASM_S390_KPROBES_H
+#define _ASM_S390_KPROBES_H
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2006
+ *
+ * 2002-Oct	Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel
+ *		Probes initial implementation ( includes suggestions from
+ *		Rusty Russell).
+ * 2004-Nov	Modified for PPC64 by Ananth N Mavinakayanahalli
+ *		<ananth@in.ibm.com>
+ * 2005-Dec	Used as a template for s390 by Mike Grundy
+ *		<grundym@us.ibm.com>
+ */
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/percpu.h>
+
+#define  __ARCH_WANT_KPROBES_INSN_SLOT
+struct pt_regs;
+struct kprobe;
+
+typedef u16 kprobe_opcode_t;
+#define BREAKPOINT_INSTRUCTION	0x0002
+
+/* Maximum instruction size is 3 (16bit) halfwords: */
+#define MAX_INSN_SIZE		0x0003
+#define MAX_STACK_SIZE		64
+#define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
+	(((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) \
+	? (MAX_STACK_SIZE) \
+	: (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
+
+#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)(pentry)
+
+#define ARCH_SUPPORTS_KRETPROBES
+#define ARCH_INACTIVE_KPROBE_COUNT 0
+
+#define KPROBE_SWAP_INST	0x10
+
+#define FIXUP_PSW_NORMAL	0x08
+#define FIXUP_BRANCH_NOT_TAKEN	0x04
+#define FIXUP_RETURN_REGISTER	0x02
+#define FIXUP_NOT_REQUIRED	0x01
+
+/* Architecture specific copy of original instruction */
+struct arch_specific_insn {
+	/* copy of original instruction */
+	kprobe_opcode_t *insn;
+	int fixup;
+	int ilen;
+	int reg;
+};
+
+struct ins_replace_args {
+	kprobe_opcode_t *ptr;
+	kprobe_opcode_t old;
+	kprobe_opcode_t new;
+};
+struct prev_kprobe {
+	struct kprobe *kp;
+	unsigned long status;
+	unsigned long saved_psw;
+	unsigned long kprobe_saved_imask;
+	unsigned long kprobe_saved_ctl[3];
+};
+
+/* per-cpu kprobe control block */
+struct kprobe_ctlblk {
+	unsigned long kprobe_status;
+	unsigned long kprobe_saved_imask;
+	unsigned long kprobe_saved_ctl[3];
+	struct pt_regs jprobe_saved_regs;
+	unsigned long jprobe_saved_r14;
+	unsigned long jprobe_saved_r15;
+	struct prev_kprobe prev_kprobe;
+	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
+};
+
+void arch_remove_kprobe(struct kprobe *p);
+void kretprobe_trampoline(void);
+int  is_prohibited_opcode(kprobe_opcode_t *instruction);
+void get_instruction_type(struct arch_specific_insn *ainsn);
+
+#define flush_insn_slot(p)	do { } while (0)
+
+#endif	/* _ASM_S390_KPROBES_H */
+
+#ifdef CONFIG_KPROBES
+
+extern int kprobe_exceptions_notify(struct notifier_block *self,
+					unsigned long val, void *data);
+#else	/* !CONFIG_KPROBES */
+static inline int kprobe_exceptions_notify(struct notifier_block *self,
+						unsigned long val, void *data)
+{
+	return 0;
+}
+#endif
-- 
cgit v1.2.3


From 65912a84c0f33304fa5ea004c7b6ee58d5f5572e Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:41 +0200
Subject: [S390] initrd vs. bootmem bitmap.

Move initrd if the bitmap of the bootmem allocator would overwrite it.
In addition this patch sets the default size and address of the initrd to 0.
Therefore all boot loaders must set the initrd size and address correctly.
This is especially relevant for ftp boot via HMC/SE, where this change
requires a special patch file entry in the .ins file which sets these two
values contained at address 0x10408 and 0x10410.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/head.S   | 10 ++++++----
 arch/s390/kernel/head31.S |  4 ++--
 arch/s390/kernel/head64.S |  4 ++--
 arch/s390/kernel/setup.c  | 41 ++++++++++++++++++++++++++++++++++++++---
 include/asm-s390/setup.h  |  2 --
 5 files changed, 48 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index adad8863ee2f..a6e9bdb53591 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -272,7 +272,7 @@ iplstart:
 # load parameter file from ipl device
 #
 .Lagain1:
- 	l     %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12) # ramdisk loc. is temp
+	l     %r2,.Linitrd		       # ramdisk loc. is temp
         bas   %r14,.Lloader                    # load parameter file
         ltr   %r2,%r2                          # got anything ?
         bz    .Lnopf
@@ -280,7 +280,7 @@ iplstart:
 	bnh   .Lnotrunc
 	la    %r2,895
 .Lnotrunc:
-	l     %r4,INITRD_START+ARCH_OFFSET-PARMAREA(%r12)
+	l     %r4,.Linitrd
 	clc   0(3,%r4),.L_hdr		       # if it is HDRx
 	bz    .Lagain1			       # skip dataset header
 	clc   0(3,%r4),.L_eof		       # if it is EOFx
@@ -323,14 +323,15 @@ iplstart:
 # load ramdisk from ipl device
 #	
 .Lagain2:
- 	l     %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12) # addr of ramdisk
+	l     %r2,.Linitrd		       # addr of ramdisk
+	st    %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12)
         bas   %r14,.Lloader                    # load ramdisk
  	st    %r2,INITRD_SIZE+ARCH_OFFSET-PARMAREA(%r12) # store size of ramdisk
         ltr   %r2,%r2
         bnz   .Lrdcont
         st    %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12) # no ramdisk found
 .Lrdcont:
-	l     %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12)
+	l     %r2,.Linitrd
 
 	clc   0(3,%r2),.L_hdr		       # skip HDRx and EOFx 
 	bz    .Lagain2
@@ -379,6 +380,7 @@ iplstart:
         l     %r1,.Lstartup
         br    %r1
 
+.Linitrd:.long _end + 0x400000		       # default address of initrd
 .Lparm:	.long  PARMAREA
 .Lstartup: .long startup
 .Lcvtab:.long  _ebcasc                         # ebcdic to ascii table
diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S
index a4dc61f3285e..0e46077d7140 100644
--- a/arch/s390/kernel/head31.S
+++ b/arch/s390/kernel/head31.S
@@ -26,8 +26,8 @@ startup:basr	%r13,0			# get base
 #
 	.org	PARMAREA
 	.long	0,0			# IPL_DEVICE
-	.long	0,RAMDISK_ORIGIN	# INITRD_START
-	.long	0,RAMDISK_SIZE		# INITRD_SIZE
+	.long	0,0			# INITRD_START
+	.long	0,0			# INITRD_SIZE
 
 	.org	COMMAND_LINE
 	.byte	"root=/dev/ram0 ro"
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 9d80c5b1ef95..3e0341acd04e 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -26,8 +26,8 @@ startup:basr  %r13,0			 # get base
 #
 	.org   PARMAREA
 	.quad  0			# IPL_DEVICE
-	.quad  RAMDISK_ORIGIN		# INITRD_START
-	.quad  RAMDISK_SIZE		# INITRD_SIZE
+	.quad  0			# INITRD_START
+	.quad  0			# INITRD_SIZE
 
 	.org   COMMAND_LINE
 	.byte  "root=/dev/ram0 ro"
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index c902f059c7aa..89051e8a5d8d 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -37,6 +37,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/device.h>
 #include <linux/notifier.h>
+#include <linux/pfn.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -501,13 +502,47 @@ setup_memory(void)
 	 * partially used pages are not usable - thus
 	 * we are rounding upwards:
 	 */
-	start_pfn = (__pa(&_end) + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	end_pfn = max_pfn = memory_end >> PAGE_SHIFT;
+	start_pfn = PFN_UP(__pa(&_end));
+	end_pfn = max_pfn = PFN_DOWN(memory_end);
 
 	/* Initialize storage key for kernel pages */
 	for (init_pfn = 0 ; init_pfn < start_pfn; init_pfn++)
 		page_set_storage_key(init_pfn << PAGE_SHIFT, PAGE_DEFAULT_KEY);
 
+#ifdef CONFIG_BLK_DEV_INITRD
+	/*
+	 * Move the initrd in case the bitmap of the bootmem allocater
+	 * would overwrite it.
+	 */
+
+	if (INITRD_START && INITRD_SIZE) {
+		unsigned long bmap_size;
+		unsigned long start;
+
+		bmap_size = bootmem_bootmap_pages(end_pfn - start_pfn + 1);
+		bmap_size = PFN_PHYS(bmap_size);
+
+		if (PFN_PHYS(start_pfn) + bmap_size > INITRD_START) {
+			start = PFN_PHYS(start_pfn) + bmap_size + PAGE_SIZE;
+
+			if (start + INITRD_SIZE > memory_end) {
+				printk("initrd extends beyond end of memory "
+				       "(0x%08lx > 0x%08lx)\n"
+				       "disabling initrd\n",
+				       start + INITRD_SIZE, memory_end);
+				INITRD_START = INITRD_SIZE = 0;
+			} else {
+				printk("Moving initrd (0x%08lx -> 0x%08lx, "
+				       "size: %ld)\n",
+				       INITRD_START, start, INITRD_SIZE);
+				memmove((void *) start, (void *) INITRD_START,
+					INITRD_SIZE);
+				INITRD_START = start;
+			}
+		}
+	}
+#endif
+
 	/*
 	 * Initialize the boot-time allocator (with low memory only):
 	 */
@@ -559,7 +594,7 @@ setup_memory(void)
 	reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size);
 
 #ifdef CONFIG_BLK_DEV_INITRD
-	if (INITRD_START) {
+	if (INITRD_START && INITRD_SIZE) {
 		if (INITRD_START + INITRD_SIZE <= memory_end) {
 			reserve_bootmem(INITRD_START, INITRD_SIZE);
 			initrd_start = INITRD_START;
diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h
index 19e31979309a..02c96d57f0cf 100644
--- a/include/asm-s390/setup.h
+++ b/include/asm-s390/setup.h
@@ -14,8 +14,6 @@
 
 #define PARMAREA		0x10400
 #define COMMAND_LINE_SIZE 	896
-#define RAMDISK_ORIGIN		0x800000
-#define RAMDISK_SIZE		0x800000
 #define MEMORY_CHUNKS		16	/* max 0x7fff */
 #define IPL_PARMBLOCK_ORIGIN	0x2000
 
-- 
cgit v1.2.3


From ff6b8ea68f4b7353f88b97024f28127e2148aa00 Mon Sep 17 00:00:00 2001
From: Michael Holzheu <holzheu@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:49 +0200
Subject: [S390] ipl/dump on panic.

It is now possible to specify a ccw/fcp dump device which is used to
automatically create a system dump in case of a kernel panic. The dump
device can be configured under /sys/firmware/dump.
In addition it is now possible to specify a ccw/fcp device which is used
for the next reboot of Linux. The reipl device can be configured under
/sys/firmware/reipl.

Signed-off-by: Michael Holzheu <holzheu@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/Makefile     |   2 +-
 arch/s390/kernel/head31.S     |   3 +
 arch/s390/kernel/head64.S     |   3 +
 arch/s390/kernel/ipl.c        | 942 ++++++++++++++++++++++++++++++++++++++++++
 arch/s390/kernel/reipl.S      |  33 +-
 arch/s390/kernel/reipl64.S    |  34 +-
 arch/s390/kernel/reipl_diag.c |  39 --
 arch/s390/kernel/setup.c      | 220 +---------
 arch/s390/kernel/smp.c        |  10 +-
 drivers/s390/cio/cio.c        |  50 ++-
 include/asm-s390/cio.h        |   7 +
 include/asm-s390/lowcore.h    |  13 +-
 include/asm-s390/setup.h      |  54 ++-
 13 files changed, 1099 insertions(+), 311 deletions(-)
 create mode 100644 arch/s390/kernel/ipl.c
 delete mode 100644 arch/s390/kernel/reipl_diag.c

(limited to 'include')

diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 33a5069c0e16..aa978978d3d1 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -6,7 +6,7 @@ EXTRA_AFLAGS	:= -traditional
 
 obj-y	:=  bitmap.o traps.o time.o process.o \
             setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \
-            semaphore.o s390_ext.o debug.o profile.o irq.o reipl_diag.o
+	    semaphore.o s390_ext.o debug.o profile.o irq.o ipl.o
 
 obj-y	+= $(if $(CONFIG_64BIT),entry64.o,entry.o)
 obj-y	+= $(if $(CONFIG_64BIT),reipl64.o,reipl.o)
diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S
index 0e46077d7140..d8bb68a72527 100644
--- a/arch/s390/kernel/head31.S
+++ b/arch/s390/kernel/head31.S
@@ -38,6 +38,7 @@ startup:basr	%r13,0			# get base
 startup_continue:
 	basr	%r13,0			# get base
 .LPG1:	GET_IPL_DEVICE
+	mvi	__LC_AR_MODE_ID,0	# set ESA flag (mode 0)
 	lctl	%c0,%c15,.Lctl-.LPG1(%r13) # load control registers
 	l	%r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area
 					# move IPL device to lowcore
@@ -274,6 +275,8 @@ startup_continue:
 .Lparmaddr: .long PARMAREA
 .Lsccbaddr: .long .Lsccb
 	.org	0x12000
+.globl s390_readinfo_sccb
+s390_readinfo_sccb:
 .Lsccb:
 	.hword	0x1000			# length, one page
 	.byte	0x00,0x00,0x00
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 3e0341acd04e..c2005101fee1 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -41,6 +41,7 @@ startup_continue:
         srl   %r13,1
 	GET_IPL_DEVICE
         lhi   %r1,1                      # mode 1 = esame
+	mvi   __LC_AR_MODE_ID,1		 # set esame flag
         slr   %r0,%r0                    # set cpuid to zero
         sigp  %r1,%r0,0x12               # switch to esame mode
 	sam64				 # switch to 64 bit mode
@@ -269,6 +270,8 @@ startup_continue:
 	.quad	PARMAREA
 
 	.org	0x12000
+.globl s390_readinfo_sccb
+s390_readinfo_sccb:
 .Lsccb:
 	.hword 0x1000			# length, one page
 	.byte 0x00,0x00,0x00
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
new file mode 100644
index 000000000000..105ee15a2b31
--- /dev/null
+++ b/arch/s390/kernel/ipl.c
@@ -0,0 +1,942 @@
+/*
+ *  arch/s390/kernel/ipl.c
+ *    ipl/reipl/dump support for Linux on s390.
+ *
+ *    Copyright (C) IBM Corp. 2005,2006
+ *    Author(s): Michael Holzheu <holzheu@de.ibm.com>
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>
+ *		 Volker Sameske <sameske@de.ibm.com>
+ */
+
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/delay.h>
+#include <linux/reboot.h>
+#include <asm/smp.h>
+#include <asm/setup.h>
+#include <asm/cpcmd.h>
+#include <asm/cio.h>
+
+#define IPL_PARM_BLOCK_VERSION 0
+
+enum ipl_type {
+	IPL_TYPE_NONE	 = 1,
+	IPL_TYPE_UNKNOWN = 2,
+	IPL_TYPE_CCW	 = 4,
+	IPL_TYPE_FCP	 = 8,
+};
+
+#define IPL_NONE_STR	 "none"
+#define IPL_UNKNOWN_STR  "unknown"
+#define IPL_CCW_STR	 "ccw"
+#define IPL_FCP_STR	 "fcp"
+
+static char *ipl_type_str(enum ipl_type type)
+{
+	switch (type) {
+	case IPL_TYPE_NONE:
+		return IPL_NONE_STR;
+	case IPL_TYPE_CCW:
+		return IPL_CCW_STR;
+	case IPL_TYPE_FCP:
+		return IPL_FCP_STR;
+	case IPL_TYPE_UNKNOWN:
+	default:
+		return IPL_UNKNOWN_STR;
+	}
+}
+
+enum ipl_method {
+	IPL_METHOD_NONE,
+	IPL_METHOD_CCW_CIO,
+	IPL_METHOD_CCW_DIAG,
+	IPL_METHOD_CCW_VM,
+	IPL_METHOD_FCP_RO_DIAG,
+	IPL_METHOD_FCP_RW_DIAG,
+	IPL_METHOD_FCP_RO_VM,
+};
+
+enum shutdown_action {
+	SHUTDOWN_REIPL,
+	SHUTDOWN_DUMP,
+	SHUTDOWN_STOP,
+};
+
+#define SHUTDOWN_REIPL_STR "reipl"
+#define SHUTDOWN_DUMP_STR  "dump"
+#define SHUTDOWN_STOP_STR  "stop"
+
+static char *shutdown_action_str(enum shutdown_action action)
+{
+	switch (action) {
+	case SHUTDOWN_REIPL:
+		return SHUTDOWN_REIPL_STR;
+	case SHUTDOWN_DUMP:
+		return SHUTDOWN_DUMP_STR;
+	case SHUTDOWN_STOP:
+		return SHUTDOWN_STOP_STR;
+	default:
+		BUG();
+	}
+}
+
+enum diag308_subcode  {
+	DIAG308_IPL   = 3,
+	DIAG308_DUMP  = 4,
+	DIAG308_SET   = 5,
+	DIAG308_STORE = 6,
+};
+
+enum diag308_ipl_type {
+	DIAG308_IPL_TYPE_FCP = 0,
+	DIAG308_IPL_TYPE_CCW = 2,
+};
+
+enum diag308_opt {
+	DIAG308_IPL_OPT_IPL  = 0x10,
+	DIAG308_IPL_OPT_DUMP = 0x20,
+};
+
+enum diag308_rc {
+	DIAG308_RC_OK = 1,
+};
+
+static int diag308_set_works = 0;
+
+static int reipl_capabilities = IPL_TYPE_UNKNOWN;
+static enum ipl_type reipl_type = IPL_TYPE_UNKNOWN;
+static enum ipl_method reipl_method = IPL_METHOD_NONE;
+static struct ipl_parameter_block *reipl_block_fcp;
+static struct ipl_parameter_block *reipl_block_ccw;
+
+static int dump_capabilities = IPL_TYPE_NONE;
+static enum ipl_type dump_type = IPL_TYPE_NONE;
+static enum ipl_method dump_method = IPL_METHOD_NONE;
+static struct ipl_parameter_block *dump_block_fcp;
+static struct ipl_parameter_block *dump_block_ccw;
+
+static enum shutdown_action on_panic_action = SHUTDOWN_STOP;
+
+static int diag308(unsigned long subcode, void *addr)
+{
+	register unsigned long _addr asm("0") = (unsigned long)addr;
+	register unsigned long _rc asm("1") = 0;
+
+	asm volatile (
+		"   diag %0,%2,0x308\n"
+		"0: \n"
+		".section __ex_table,\"a\"\n"
+#ifdef CONFIG_64BIT
+		"   .align 8\n"
+		"   .quad 0b, 0b\n"
+#else
+		"   .align 4\n"
+		"   .long 0b, 0b\n"
+#endif
+		".previous\n"
+		: "+d" (_addr), "+d" (_rc)
+		: "d" (subcode) : "cc", "memory" );
+
+	return _rc;
+}
+
+/* SYSFS */
+
+#define DEFINE_IPL_ATTR_RO(_prefix, _name, _format, _value)		\
+static ssize_t sys_##_prefix##_##_name##_show(struct subsystem *subsys,	\
+		char *page)						\
+{									\
+	return sprintf(page, _format, _value);				\
+}									\
+static struct subsys_attribute sys_##_prefix##_##_name##_attr =		\
+	__ATTR(_name, S_IRUGO, sys_##_prefix##_##_name##_show, NULL);
+
+#define DEFINE_IPL_ATTR_RW(_prefix, _name, _fmt_out, _fmt_in, _value)	\
+static ssize_t sys_##_prefix##_##_name##_show(struct subsystem *subsys,	\
+		char *page)						\
+{									\
+	return sprintf(page, _fmt_out,					\
+			(unsigned long long) _value);			\
+}									\
+static ssize_t sys_##_prefix##_##_name##_store(struct subsystem *subsys,\
+		const char *buf, size_t len)				\
+{									\
+	unsigned long long value;					\
+	if (sscanf(buf, _fmt_in, &value) != 1)				\
+		return -EINVAL;						\
+	_value = value;							\
+	return len;							\
+}									\
+static struct subsys_attribute sys_##_prefix##_##_name##_attr =		\
+	__ATTR(_name,(S_IRUGO | S_IWUSR),				\
+			sys_##_prefix##_##_name##_show,			\
+			sys_##_prefix##_##_name##_store);
+
+static void make_attrs_ro(struct attribute **attrs)
+{
+	while (*attrs) {
+		(*attrs)->mode = S_IRUGO;
+		attrs++;
+	}
+}
+
+/*
+ * ipl section
+ */
+
+static enum ipl_type ipl_get_type(void)
+{
+	struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START;
+
+	if (!IPL_DEVNO_VALID)
+		return IPL_TYPE_UNKNOWN;
+	if (!IPL_PARMBLOCK_VALID)
+		return IPL_TYPE_CCW;
+	if (ipl->hdr.version > IPL_MAX_SUPPORTED_VERSION)
+		return IPL_TYPE_UNKNOWN;
+	if (ipl->hdr.pbt != DIAG308_IPL_TYPE_FCP)
+		return IPL_TYPE_UNKNOWN;
+	return IPL_TYPE_FCP;
+}
+
+static ssize_t ipl_type_show(struct subsystem *subsys, char *page)
+{
+	return sprintf(page, "%s\n", ipl_type_str(ipl_get_type()));
+}
+
+static struct subsys_attribute sys_ipl_type_attr = __ATTR_RO(ipl_type);
+
+static ssize_t sys_ipl_device_show(struct subsystem *subsys, char *page)
+{
+	struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START;
+
+	switch (ipl_get_type()) {
+	case IPL_TYPE_CCW:
+		return sprintf(page, "0.0.%04x\n", ipl_devno);
+	case IPL_TYPE_FCP:
+		return sprintf(page, "0.0.%04x\n", ipl->ipl_info.fcp.devno);
+	default:
+		return 0;
+	}
+}
+
+static struct subsys_attribute sys_ipl_device_attr =
+	__ATTR(device, S_IRUGO, sys_ipl_device_show, NULL);
+
+static ssize_t ipl_parameter_read(struct kobject *kobj, char *buf, loff_t off,
+				  size_t count)
+{
+	unsigned int size = IPL_PARMBLOCK_SIZE;
+
+	if (off > size)
+		return 0;
+	if (off + count > size)
+		count = size - off;
+	memcpy(buf, (void *)IPL_PARMBLOCK_START + off, count);
+	return count;
+}
+
+static struct bin_attribute ipl_parameter_attr = {
+	.attr = {
+		.name = "binary_parameter",
+		.mode = S_IRUGO,
+		.owner = THIS_MODULE,
+	},
+	.size = PAGE_SIZE,
+	.read = &ipl_parameter_read,
+};
+
+static ssize_t ipl_scp_data_read(struct kobject *kobj, char *buf, loff_t off,
+	size_t count)
+{
+	unsigned int size = IPL_PARMBLOCK_START->ipl_info.fcp.scp_data_len;
+	void *scp_data = &IPL_PARMBLOCK_START->ipl_info.fcp.scp_data;
+
+	if (off > size)
+		return 0;
+	if (off + count > size)
+		count = size - off;
+	memcpy(buf, scp_data + off, count);
+	return count;
+}
+
+static struct bin_attribute ipl_scp_data_attr = {
+	.attr = {
+		.name = "scp_data",
+		.mode = S_IRUGO,
+		.owner = THIS_MODULE,
+	},
+	.size = PAGE_SIZE,
+	.read = &ipl_scp_data_read,
+};
+
+/* FCP ipl device attributes */
+
+DEFINE_IPL_ATTR_RO(ipl_fcp, wwpn, "0x%016llx\n", (unsigned long long)
+		   IPL_PARMBLOCK_START->ipl_info.fcp.wwpn);
+DEFINE_IPL_ATTR_RO(ipl_fcp, lun, "0x%016llx\n", (unsigned long long)
+		   IPL_PARMBLOCK_START->ipl_info.fcp.lun);
+DEFINE_IPL_ATTR_RO(ipl_fcp, bootprog, "%lld\n", (unsigned long long)
+		   IPL_PARMBLOCK_START->ipl_info.fcp.bootprog);
+DEFINE_IPL_ATTR_RO(ipl_fcp, br_lba, "%lld\n", (unsigned long long)
+		   IPL_PARMBLOCK_START->ipl_info.fcp.br_lba);
+
+static struct attribute *ipl_fcp_attrs[] = {
+	&sys_ipl_type_attr.attr,
+	&sys_ipl_device_attr.attr,
+	&sys_ipl_fcp_wwpn_attr.attr,
+	&sys_ipl_fcp_lun_attr.attr,
+	&sys_ipl_fcp_bootprog_attr.attr,
+	&sys_ipl_fcp_br_lba_attr.attr,
+	NULL,
+};
+
+static struct attribute_group ipl_fcp_attr_group = {
+	.attrs = ipl_fcp_attrs,
+};
+
+/* CCW ipl device attributes */
+
+static struct attribute *ipl_ccw_attrs[] = {
+	&sys_ipl_type_attr.attr,
+	&sys_ipl_device_attr.attr,
+	NULL,
+};
+
+static struct attribute_group ipl_ccw_attr_group = {
+	.attrs = ipl_ccw_attrs,
+};
+
+/* UNKNOWN ipl device attributes */
+
+static struct attribute *ipl_unknown_attrs[] = {
+	&sys_ipl_type_attr.attr,
+	NULL,
+};
+
+static struct attribute_group ipl_unknown_attr_group = {
+	.attrs = ipl_unknown_attrs,
+};
+
+static decl_subsys(ipl, NULL, NULL);
+
+/*
+ * reipl section
+ */
+
+/* FCP reipl device attributes */
+
+DEFINE_IPL_ATTR_RW(reipl_fcp, wwpn, "0x%016llx\n", "%016llx\n",
+		   reipl_block_fcp->ipl_info.fcp.wwpn);
+DEFINE_IPL_ATTR_RW(reipl_fcp, lun, "0x%016llx\n", "%016llx\n",
+		   reipl_block_fcp->ipl_info.fcp.lun);
+DEFINE_IPL_ATTR_RW(reipl_fcp, bootprog, "%lld\n", "%lld\n",
+		   reipl_block_fcp->ipl_info.fcp.bootprog);
+DEFINE_IPL_ATTR_RW(reipl_fcp, br_lba, "%lld\n", "%lld\n",
+		   reipl_block_fcp->ipl_info.fcp.br_lba);
+DEFINE_IPL_ATTR_RW(reipl_fcp, device, "0.0.%04llx\n", "0.0.%llx\n",
+		   reipl_block_fcp->ipl_info.fcp.devno);
+
+static struct attribute *reipl_fcp_attrs[] = {
+	&sys_reipl_fcp_device_attr.attr,
+	&sys_reipl_fcp_wwpn_attr.attr,
+	&sys_reipl_fcp_lun_attr.attr,
+	&sys_reipl_fcp_bootprog_attr.attr,
+	&sys_reipl_fcp_br_lba_attr.attr,
+	NULL,
+};
+
+static struct attribute_group reipl_fcp_attr_group = {
+	.name  = IPL_FCP_STR,
+	.attrs = reipl_fcp_attrs,
+};
+
+/* CCW reipl device attributes */
+
+DEFINE_IPL_ATTR_RW(reipl_ccw, device, "0.0.%04llx\n", "0.0.%llx\n",
+	reipl_block_ccw->ipl_info.ccw.devno);
+
+static struct attribute *reipl_ccw_attrs[] = {
+	&sys_reipl_ccw_device_attr.attr,
+	NULL,
+};
+
+static struct attribute_group reipl_ccw_attr_group = {
+	.name  = IPL_CCW_STR,
+	.attrs = reipl_ccw_attrs,
+};
+
+/* reipl type */
+
+static int reipl_set_type(enum ipl_type type)
+{
+	if (!(reipl_capabilities & type))
+		return -EINVAL;
+
+	switch(type) {
+	case IPL_TYPE_CCW:
+		if (MACHINE_IS_VM)
+			reipl_method = IPL_METHOD_CCW_VM;
+		else
+			reipl_method = IPL_METHOD_CCW_CIO;
+		break;
+	case IPL_TYPE_FCP:
+		if (diag308_set_works)
+			reipl_method = IPL_METHOD_FCP_RW_DIAG;
+		else if (MACHINE_IS_VM)
+			reipl_method = IPL_METHOD_FCP_RO_VM;
+		else
+			reipl_method = IPL_METHOD_FCP_RO_DIAG;
+		break;
+	default:
+		reipl_method = IPL_METHOD_NONE;
+	}
+	reipl_type = type;
+	return 0;
+}
+
+static ssize_t reipl_type_show(struct subsystem *subsys, char *page)
+{
+	return sprintf(page, "%s\n", ipl_type_str(reipl_type));
+}
+
+static ssize_t reipl_type_store(struct subsystem *subsys, const char *buf,
+				size_t len)
+{
+	int rc = -EINVAL;
+
+	if (strncmp(buf, IPL_CCW_STR, strlen(IPL_CCW_STR)) == 0)
+		rc = reipl_set_type(IPL_TYPE_CCW);
+	else if (strncmp(buf, IPL_FCP_STR, strlen(IPL_FCP_STR)) == 0)
+		rc = reipl_set_type(IPL_TYPE_FCP);
+	return (rc != 0) ? rc : len;
+}
+
+static struct subsys_attribute reipl_type_attr =
+		__ATTR(reipl_type, 0644, reipl_type_show, reipl_type_store);
+
+static decl_subsys(reipl, NULL, NULL);
+
+/*
+ * dump section
+ */
+
+/* FCP dump device attributes */
+
+DEFINE_IPL_ATTR_RW(dump_fcp, wwpn, "0x%016llx\n", "%016llx\n",
+		   dump_block_fcp->ipl_info.fcp.wwpn);
+DEFINE_IPL_ATTR_RW(dump_fcp, lun, "0x%016llx\n", "%016llx\n",
+		   dump_block_fcp->ipl_info.fcp.lun);
+DEFINE_IPL_ATTR_RW(dump_fcp, bootprog, "%lld\n", "%lld\n",
+		   dump_block_fcp->ipl_info.fcp.bootprog);
+DEFINE_IPL_ATTR_RW(dump_fcp, br_lba, "%lld\n", "%lld\n",
+		   dump_block_fcp->ipl_info.fcp.br_lba);
+DEFINE_IPL_ATTR_RW(dump_fcp, device, "0.0.%04llx\n", "0.0.%llx\n",
+		   dump_block_fcp->ipl_info.fcp.devno);
+
+static struct attribute *dump_fcp_attrs[] = {
+	&sys_dump_fcp_device_attr.attr,
+	&sys_dump_fcp_wwpn_attr.attr,
+	&sys_dump_fcp_lun_attr.attr,
+	&sys_dump_fcp_bootprog_attr.attr,
+	&sys_dump_fcp_br_lba_attr.attr,
+	NULL,
+};
+
+static struct attribute_group dump_fcp_attr_group = {
+	.name  = IPL_FCP_STR,
+	.attrs = dump_fcp_attrs,
+};
+
+/* CCW dump device attributes */
+
+DEFINE_IPL_ATTR_RW(dump_ccw, device, "0.0.%04llx\n", "0.0.%llx\n",
+		   dump_block_ccw->ipl_info.ccw.devno);
+
+static struct attribute *dump_ccw_attrs[] = {
+	&sys_dump_ccw_device_attr.attr,
+	NULL,
+};
+
+static struct attribute_group dump_ccw_attr_group = {
+	.name  = IPL_CCW_STR,
+	.attrs = dump_ccw_attrs,
+};
+
+/* dump type */
+
+static int dump_set_type(enum ipl_type type)
+{
+	if (!(dump_capabilities & type))
+		return -EINVAL;
+	switch(type) {
+	case IPL_TYPE_CCW:
+		if (MACHINE_IS_VM)
+			dump_method = IPL_METHOD_CCW_VM;
+		else
+			dump_method = IPL_METHOD_CCW_CIO;
+		break;
+	case IPL_TYPE_FCP:
+		dump_method = IPL_METHOD_FCP_RW_DIAG;
+		break;
+	default:
+		dump_method = IPL_METHOD_NONE;
+	}
+	dump_type = type;
+	return 0;
+}
+
+static ssize_t dump_type_show(struct subsystem *subsys, char *page)
+{
+	return sprintf(page, "%s\n", ipl_type_str(dump_type));
+}
+
+static ssize_t dump_type_store(struct subsystem *subsys, const char *buf,
+			       size_t len)
+{
+	int rc = -EINVAL;
+
+	if (strncmp(buf, IPL_NONE_STR, strlen(IPL_NONE_STR)) == 0)
+		rc = dump_set_type(IPL_TYPE_NONE);
+	else if (strncmp(buf, IPL_CCW_STR, strlen(IPL_CCW_STR)) == 0)
+		rc = dump_set_type(IPL_TYPE_CCW);
+	else if (strncmp(buf, IPL_FCP_STR, strlen(IPL_FCP_STR)) == 0)
+		rc = dump_set_type(IPL_TYPE_FCP);
+	return (rc != 0) ? rc : len;
+}
+
+static struct subsys_attribute dump_type_attr =
+		__ATTR(dump_type, 0644, dump_type_show, dump_type_store);
+
+static decl_subsys(dump, NULL, NULL);
+
+#ifdef CONFIG_SMP
+static void dump_smp_stop_all(void)
+{
+	int cpu;
+	preempt_disable();
+	for_each_online_cpu(cpu) {
+		if (cpu == smp_processor_id())
+			continue;
+		while (signal_processor(cpu, sigp_stop) == sigp_busy)
+			udelay(10);
+	}
+	preempt_enable();
+}
+#else
+#define dump_smp_stop_all() do { } while (0)
+#endif
+
+/*
+ * Shutdown actions section
+ */
+
+static decl_subsys(shutdown_actions, NULL, NULL);
+
+/* on panic */
+
+static ssize_t on_panic_show(struct subsystem *subsys, char *page)
+{
+	return sprintf(page, "%s\n", shutdown_action_str(on_panic_action));
+}
+
+static ssize_t on_panic_store(struct subsystem *subsys, const char *buf,
+			      size_t len)
+{
+	if (strncmp(buf, SHUTDOWN_REIPL_STR, strlen(SHUTDOWN_REIPL_STR)) == 0)
+		on_panic_action = SHUTDOWN_REIPL;
+	else if (strncmp(buf, SHUTDOWN_DUMP_STR,
+			 strlen(SHUTDOWN_DUMP_STR)) == 0)
+		on_panic_action = SHUTDOWN_DUMP;
+	else if (strncmp(buf, SHUTDOWN_STOP_STR,
+			 strlen(SHUTDOWN_STOP_STR)) == 0)
+		on_panic_action = SHUTDOWN_STOP;
+	else
+		return -EINVAL;
+
+	return len;
+}
+
+static struct subsys_attribute on_panic_attr =
+		__ATTR(on_panic, 0644, on_panic_show, on_panic_store);
+
+static void print_fcp_block(struct ipl_parameter_block *fcp_block)
+{
+	printk(KERN_EMERG "wwpn:      %016llx\n",
+		(unsigned long long)fcp_block->ipl_info.fcp.wwpn);
+	printk(KERN_EMERG "lun:       %016llx\n",
+		(unsigned long long)fcp_block->ipl_info.fcp.lun);
+	printk(KERN_EMERG "bootprog:  %lld\n",
+		(unsigned long long)fcp_block->ipl_info.fcp.bootprog);
+	printk(KERN_EMERG "br_lba:    %lld\n",
+		(unsigned long long)fcp_block->ipl_info.fcp.br_lba);
+	printk(KERN_EMERG "device:    %llx\n",
+		(unsigned long long)fcp_block->ipl_info.fcp.devno);
+	printk(KERN_EMERG "opt:       %x\n", fcp_block->ipl_info.fcp.opt);
+}
+
+void do_reipl(void)
+{
+	struct ccw_dev_id devid;
+	static char buf[100];
+
+	switch (reipl_type) {
+	case IPL_TYPE_CCW:
+		printk(KERN_EMERG "reboot on ccw device: 0.0.%04x\n",
+			reipl_block_ccw->ipl_info.ccw.devno);
+		break;
+	case IPL_TYPE_FCP:
+		printk(KERN_EMERG "reboot on fcp device:\n");
+		print_fcp_block(reipl_block_fcp);
+		break;
+	default:
+		break;
+	}
+
+	switch (reipl_method) {
+	case IPL_METHOD_CCW_CIO:
+		devid.devno = reipl_block_ccw->ipl_info.ccw.devno;
+		devid.ssid  = 0;
+		reipl_ccw_dev(&devid);
+		break;
+	case IPL_METHOD_CCW_VM:
+		sprintf(buf, "IPL %X", reipl_block_ccw->ipl_info.ccw.devno);
+		cpcmd(buf, NULL, 0, NULL);
+		break;
+	case IPL_METHOD_CCW_DIAG:
+		diag308(DIAG308_SET, reipl_block_ccw);
+		diag308(DIAG308_IPL, NULL);
+		break;
+	case IPL_METHOD_FCP_RW_DIAG:
+		diag308(DIAG308_SET, reipl_block_fcp);
+		diag308(DIAG308_IPL, NULL);
+		break;
+	case IPL_METHOD_FCP_RO_DIAG:
+		diag308(DIAG308_IPL, NULL);
+		break;
+	case IPL_METHOD_FCP_RO_VM:
+		cpcmd("IPL", NULL, 0, NULL);
+		break;
+	case IPL_METHOD_NONE:
+	default:
+		if (MACHINE_IS_VM)
+			cpcmd("IPL", NULL, 0, NULL);
+		diag308(DIAG308_IPL, NULL);
+		break;
+	}
+	panic("reipl failed!\n");
+}
+
+static void do_dump(void)
+{
+	struct ccw_dev_id devid;
+	static char buf[100];
+
+	switch (dump_type) {
+	case IPL_TYPE_CCW:
+		printk(KERN_EMERG "Automatic dump on ccw device: 0.0.%04x\n",
+		       dump_block_ccw->ipl_info.ccw.devno);
+		break;
+	case IPL_TYPE_FCP:
+		printk(KERN_EMERG "Automatic dump on fcp device:\n");
+		print_fcp_block(dump_block_fcp);
+		break;
+	default:
+		return;
+	}
+
+	switch (dump_method) {
+	case IPL_METHOD_CCW_CIO:
+		dump_smp_stop_all();
+		devid.devno = dump_block_ccw->ipl_info.ccw.devno;
+		devid.ssid  = 0;
+		reipl_ccw_dev(&devid);
+		break;
+	case IPL_METHOD_CCW_VM:
+		dump_smp_stop_all();
+		sprintf(buf, "STORE STATUS");
+		cpcmd(buf, NULL, 0, NULL);
+		sprintf(buf, "IPL %X", dump_block_ccw->ipl_info.ccw.devno);
+		cpcmd(buf, NULL, 0, NULL);
+		break;
+	case IPL_METHOD_CCW_DIAG:
+		diag308(DIAG308_SET, dump_block_ccw);
+		diag308(DIAG308_DUMP, NULL);
+		break;
+	case IPL_METHOD_FCP_RW_DIAG:
+		diag308(DIAG308_SET, dump_block_fcp);
+		diag308(DIAG308_DUMP, NULL);
+		break;
+	case IPL_METHOD_NONE:
+	default:
+		return;
+	}
+	printk(KERN_EMERG "Dump failed!\n");
+}
+
+/* init functions */
+
+static int __init ipl_register_fcp_files(void)
+{
+	int rc;
+
+	rc = sysfs_create_group(&ipl_subsys.kset.kobj,
+				&ipl_fcp_attr_group);
+	if (rc)
+		goto out;
+	rc = sysfs_create_bin_file(&ipl_subsys.kset.kobj,
+				   &ipl_parameter_attr);
+	if (rc)
+		goto out_ipl_parm;
+	rc = sysfs_create_bin_file(&ipl_subsys.kset.kobj,
+				   &ipl_scp_data_attr);
+	if (!rc)
+		goto out;
+
+	sysfs_remove_bin_file(&ipl_subsys.kset.kobj, &ipl_parameter_attr);
+
+out_ipl_parm:
+	sysfs_remove_group(&ipl_subsys.kset.kobj, &ipl_fcp_attr_group);
+out:
+	return rc;
+}
+
+static int __init ipl_init(void)
+{
+	int rc;
+
+	rc = firmware_register(&ipl_subsys);
+	if (rc)
+		return rc;
+	switch (ipl_get_type()) {
+	case IPL_TYPE_CCW:
+		rc = sysfs_create_group(&ipl_subsys.kset.kobj,
+					&ipl_ccw_attr_group);
+		break;
+	case IPL_TYPE_FCP:
+		rc = ipl_register_fcp_files();
+		break;
+	default:
+		rc = sysfs_create_group(&ipl_subsys.kset.kobj,
+					&ipl_unknown_attr_group);
+		break;
+	}
+	if (rc)
+		firmware_unregister(&ipl_subsys);
+	return rc;
+}
+
+static void __init reipl_probe(void)
+{
+	void *buffer;
+
+	buffer = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!buffer)
+		return;
+	if (diag308(DIAG308_STORE, buffer) == DIAG308_RC_OK)
+		diag308_set_works = 1;
+	free_page((unsigned long)buffer);
+}
+
+static int __init reipl_ccw_init(void)
+{
+	int rc;
+
+	reipl_block_ccw = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!reipl_block_ccw)
+		return -ENOMEM;
+	rc = sysfs_create_group(&reipl_subsys.kset.kobj, &reipl_ccw_attr_group);
+	if (rc) {
+		free_page((unsigned long)reipl_block_ccw);
+		return rc;
+	}
+	reipl_block_ccw->hdr.len = IPL_PARM_BLK_CCW_LEN;
+	reipl_block_ccw->hdr.version = IPL_PARM_BLOCK_VERSION;
+	reipl_block_ccw->hdr.blk0_len = sizeof(reipl_block_ccw->ipl_info.ccw);
+	reipl_block_ccw->hdr.pbt = DIAG308_IPL_TYPE_CCW;
+	if (ipl_get_type() == IPL_TYPE_CCW)
+		reipl_block_ccw->ipl_info.ccw.devno = ipl_devno;
+	reipl_capabilities |= IPL_TYPE_CCW;
+	return 0;
+}
+
+static int __init reipl_fcp_init(void)
+{
+	int rc;
+
+	if ((!diag308_set_works) && (ipl_get_type() != IPL_TYPE_FCP))
+		return 0;
+	if ((!diag308_set_works) && (ipl_get_type() == IPL_TYPE_FCP))
+		make_attrs_ro(reipl_fcp_attrs);
+
+	reipl_block_fcp = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!reipl_block_fcp)
+		return -ENOMEM;
+	rc = sysfs_create_group(&reipl_subsys.kset.kobj, &reipl_fcp_attr_group);
+	if (rc) {
+		free_page((unsigned long)reipl_block_fcp);
+		return rc;
+	}
+	if (ipl_get_type() == IPL_TYPE_FCP) {
+		memcpy(reipl_block_fcp, IPL_PARMBLOCK_START, PAGE_SIZE);
+	} else {
+		reipl_block_fcp->hdr.len = IPL_PARM_BLK_FCP_LEN;
+		reipl_block_fcp->hdr.version = IPL_PARM_BLOCK_VERSION;
+		reipl_block_fcp->hdr.blk0_len =
+			sizeof(reipl_block_fcp->ipl_info.fcp);
+		reipl_block_fcp->hdr.pbt = DIAG308_IPL_TYPE_FCP;
+		reipl_block_fcp->ipl_info.fcp.opt = DIAG308_IPL_OPT_IPL;
+	}
+	reipl_capabilities |= IPL_TYPE_FCP;
+	return 0;
+}
+
+static int __init reipl_init(void)
+{
+	int rc;
+
+	rc = firmware_register(&reipl_subsys);
+	if (rc)
+		return rc;
+	rc = subsys_create_file(&reipl_subsys, &reipl_type_attr);
+	if (rc) {
+		firmware_unregister(&reipl_subsys);
+		return rc;
+	}
+	rc = reipl_ccw_init();
+	if (rc)
+		return rc;
+	rc = reipl_fcp_init();
+	if (rc)
+		return rc;
+	rc = reipl_set_type(ipl_get_type());
+	if (rc)
+		return rc;
+	return 0;
+}
+
+static int __init dump_ccw_init(void)
+{
+	int rc;
+
+	dump_block_ccw = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!dump_block_ccw)
+		return -ENOMEM;
+	rc = sysfs_create_group(&dump_subsys.kset.kobj, &dump_ccw_attr_group);
+	if (rc) {
+		free_page((unsigned long)dump_block_ccw);
+		return rc;
+	}
+	dump_block_ccw->hdr.len = IPL_PARM_BLK_CCW_LEN;
+	dump_block_ccw->hdr.version = IPL_PARM_BLOCK_VERSION;
+	dump_block_ccw->hdr.blk0_len = sizeof(reipl_block_ccw->ipl_info.ccw);
+	dump_block_ccw->hdr.pbt = DIAG308_IPL_TYPE_CCW;
+	dump_capabilities |= IPL_TYPE_CCW;
+	return 0;
+}
+
+extern char s390_readinfo_sccb[];
+
+static int __init dump_fcp_init(void)
+{
+	int rc;
+
+	if(!(s390_readinfo_sccb[91] & 0x2))
+		return 0; /* LDIPL DUMP is not installed */
+	if (!diag308_set_works)
+		return 0;
+	dump_block_fcp = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!dump_block_fcp)
+		return -ENOMEM;
+	rc = sysfs_create_group(&dump_subsys.kset.kobj, &dump_fcp_attr_group);
+	if (rc) {
+		free_page((unsigned long)dump_block_fcp);
+		return rc;
+	}
+	dump_block_fcp->hdr.len = IPL_PARM_BLK_FCP_LEN;
+	dump_block_fcp->hdr.version = IPL_PARM_BLOCK_VERSION;
+	dump_block_fcp->hdr.blk0_len = sizeof(dump_block_fcp->ipl_info.fcp);
+	dump_block_fcp->hdr.pbt = DIAG308_IPL_TYPE_FCP;
+	dump_block_fcp->ipl_info.fcp.opt = DIAG308_IPL_OPT_DUMP;
+	dump_capabilities |= IPL_TYPE_FCP;
+	return 0;
+}
+
+#define SHUTDOWN_ON_PANIC_PRIO 0
+
+static int shutdown_on_panic_notify(struct notifier_block *self,
+				    unsigned long event, void *data)
+{
+	if (on_panic_action == SHUTDOWN_DUMP)
+		do_dump();
+	else if (on_panic_action == SHUTDOWN_REIPL)
+		do_reipl();
+	return NOTIFY_OK;
+}
+
+static struct notifier_block shutdown_on_panic_nb = {
+	.notifier_call = shutdown_on_panic_notify,
+	.priority = SHUTDOWN_ON_PANIC_PRIO
+};
+
+static int __init dump_init(void)
+{
+	int rc;
+
+	rc = firmware_register(&dump_subsys);
+	if (rc)
+		return rc;
+	rc = subsys_create_file(&dump_subsys, &dump_type_attr);
+	if (rc) {
+		firmware_unregister(&dump_subsys);
+		return rc;
+	}
+	rc = dump_ccw_init();
+	if (rc)
+		return rc;
+	rc = dump_fcp_init();
+	if (rc)
+		return rc;
+	dump_set_type(IPL_TYPE_NONE);
+	return 0;
+}
+
+static int __init shutdown_actions_init(void)
+{
+	int rc;
+
+	rc = firmware_register(&shutdown_actions_subsys);
+	if (rc)
+		return rc;
+	rc = subsys_create_file(&shutdown_actions_subsys, &on_panic_attr);
+	if (rc) {
+		firmware_unregister(&shutdown_actions_subsys);
+		return rc;
+	}
+	atomic_notifier_chain_register(&panic_notifier_list,
+				       &shutdown_on_panic_nb);
+	return 0;
+}
+
+static int __init s390_ipl_init(void)
+{
+	int rc;
+
+	reipl_probe();
+	rc = ipl_init();
+	if (rc)
+		return rc;
+	rc = reipl_init();
+	if (rc)
+		return rc;
+	rc = dump_init();
+	if (rc)
+		return rc;
+	rc = shutdown_actions_init();
+	if (rc)
+		return rc;
+	return 0;
+}
+
+__initcall(s390_ipl_init);
diff --git a/arch/s390/kernel/reipl.S b/arch/s390/kernel/reipl.S
index 658e5ac484f9..4562cdbce8eb 100644
--- a/arch/s390/kernel/reipl.S
+++ b/arch/s390/kernel/reipl.S
@@ -8,13 +8,30 @@
 
 #include <asm/lowcore.h>
 
-		.globl	do_reipl
-do_reipl:	basr	%r13,0
+		.globl	do_reipl_asm
+do_reipl_asm:	basr	%r13,0
 .Lpg0:		lpsw	.Lnewpsw-.Lpg0(%r13)
-.Lpg1:		lctl	%c6,%c6,.Lall-.Lpg0(%r13)
-                stctl   %c0,%c0,.Lctlsave-.Lpg0(%r13)
-                ni      .Lctlsave-.Lpg0(%r13),0xef
-                lctl    %c0,%c0,.Lctlsave-.Lpg0(%r13)
+
+		# switch off lowcore protection
+
+.Lpg1:		stctl	%c0,%c0,.Lctlsave1-.Lpg0(%r13)
+		stctl	%c0,%c0,.Lctlsave2-.Lpg0(%r13)
+		ni	.Lctlsave1-.Lpg0(%r13),0xef
+		lctl	%c0,%c0,.Lctlsave1-.Lpg0(%r13)
+
+		# do store status of all registers
+
+		stm	%r0,%r15,__LC_GPREGS_SAVE_AREA
+		stctl	%c0,%c15,__LC_CREGS_SAVE_AREA
+		mvc	__LC_CREGS_SAVE_AREA(4),.Lctlsave2-.Lpg0(%r13)
+		stam	%a0,%a15,__LC_AREGS_SAVE_AREA
+		stpx	__LC_PREFIX_SAVE_AREA
+		stckc	.Lclkcmp-.Lpg0(%r13)
+		mvc	__LC_CLOCK_COMP_SAVE_AREA(8),.Lclkcmp-.Lpg0(%r13)
+		stpt	__LC_CPU_TIMER_SAVE_AREA
+		st	%r13, __LC_PSW_SAVE_AREA+4
+
+		lctl	%c6,%c6,.Lall-.Lpg0(%r13)
                 lr      %r1,%r2
         	mvc     __LC_PGM_NEW_PSW(8),.Lpcnew-.Lpg0(%r13)
                 stsch   .Lschib-.Lpg0(%r13)                                    
@@ -46,9 +63,11 @@ do_reipl:	basr	%r13,0
 .Ldisab:	st      %r14,.Ldispsw+4-.Lpg0(%r13)
 		lpsw	.Ldispsw-.Lpg0(%r13)
                 .align 	8
+.Lclkcmp:	.quad	0x0000000000000000
 .Lall:		.long	0xff000000
 .Lnull:		.long   0x00000000
-.Lctlsave:      .long   0x00000000
+.Lctlsave1:	.long	0x00000000
+.Lctlsave2:	.long	0x00000000
                 .align 	8
 .Lnewpsw:	.long   0x00080000,0x80000000+.Lpg1
 .Lpcnew:  	.long   0x00080000,0x80000000+.Lecs
diff --git a/arch/s390/kernel/reipl64.S b/arch/s390/kernel/reipl64.S
index 4d090d60f3ef..95bd1e234f63 100644
--- a/arch/s390/kernel/reipl64.S
+++ b/arch/s390/kernel/reipl64.S
@@ -8,13 +8,30 @@
  */
 
 #include <asm/lowcore.h>
-		.globl	do_reipl
-do_reipl:	basr	%r13,0
-.Lpg0:		lpswe   .Lnewpsw-.Lpg0(%r13)
+		.globl	do_reipl_asm
+do_reipl_asm:	basr	%r13,0
+
+		# do store status of all registers
+
+.Lpg0:		stg	%r1,.Lregsave-.Lpg0(%r13)
+		lghi	%r1,0x1000
+		stmg	%r0,%r15,__LC_GPREGS_SAVE_AREA-0x1000(%r1)
+		lg	%r0,.Lregsave-.Lpg0(%r13)
+		stg	%r0,__LC_GPREGS_SAVE_AREA-0x1000+8(%r1)
+		stctg	%c0,%c15,__LC_CREGS_SAVE_AREA-0x1000(%r1)
+		stam	%a0,%a15,__LC_AREGS_SAVE_AREA-0x1000(%r1)
+		stpx	__LC_PREFIX_SAVE_AREA-0x1000(%r1)
+		stfpc	__LC_FP_CREG_SAVE_AREA-0x1000(%r1)
+		stckc	.Lclkcmp-.Lpg0(%r13)
+		mvc	__LC_CLOCK_COMP_SAVE_AREA-0x1000(8,%r1),.Lclkcmp-.Lpg0(%r13)
+		stpt	__LC_CPU_TIMER_SAVE_AREA-0x1000(%r1)
+		stg	%r13, __LC_PSW_SAVE_AREA-0x1000+8(%r1)
+
+		lpswe	.Lnewpsw-.Lpg0(%r13)
 .Lpg1:		lctlg	%c6,%c6,.Lall-.Lpg0(%r13)
-                stctg   %c0,%c0,.Lctlsave-.Lpg0(%r13)
-                ni      .Lctlsave+4-.Lpg0(%r13),0xef
-                lctlg   %c0,%c0,.Lctlsave-.Lpg0(%r13)
+		stctg	%c0,%c0,.Lregsave-.Lpg0(%r13)
+		ni	.Lregsave+4-.Lpg0(%r13),0xef
+		lctlg	%c0,%c0,.Lregsave-.Lpg0(%r13)
                 lgr     %r1,%r2
         	mvc     __LC_PGM_NEW_PSW(16),.Lpcnew-.Lpg0(%r13)
                 stsch   .Lschib-.Lpg0(%r13)                                    
@@ -50,8 +67,9 @@ do_reipl:	basr	%r13,0
 		st     %r14,.Ldispsw+12-.Lpg0(%r13)
 		lpswe	.Ldispsw-.Lpg0(%r13)
                 .align 	8
+.Lclkcmp:	.quad	0x0000000000000000
 .Lall:		.quad	0x00000000ff000000
-.Lctlsave:      .quad   0x0000000000000000
+.Lregsave:	.quad	0x0000000000000000
 .Lnull:		.long   0x0000000000000000
                 .align 	16
 /*
@@ -92,5 +110,3 @@ do_reipl:	basr	%r13,0
 		.long	0x00000000,0x00000000
 		.long	0x00000000,0x00000000
 	
-
-	
diff --git a/arch/s390/kernel/reipl_diag.c b/arch/s390/kernel/reipl_diag.c
deleted file mode 100644
index 1f33951ba439..000000000000
--- a/arch/s390/kernel/reipl_diag.c
+++ /dev/null
@@ -1,39 +0,0 @@
-/*
- * This file contains the implementation of the
- * Linux re-IPL support
- *
- * (C) Copyright IBM Corp. 2005
- *
- * Author(s): Volker Sameske (sameske@de.ibm.com)
- *
- */
-
-#include <linux/kernel.h>
-
-static unsigned int reipl_diag_rc1;
-static unsigned int reipl_diag_rc2;
-
-/*
- * re-IPL the system using the last used IPL parameters
- */
-void reipl_diag(void)
-{
-        asm volatile (
-		"   la   %%r4,0\n"
-		"   la   %%r5,0\n"
-                "   diag %%r4,%2,0x308\n"
-                "0:\n"
-		"   st   %%r4,%0\n"
-		"   st   %%r5,%1\n"
-                ".section __ex_table,\"a\"\n"
-#ifdef CONFIG_64BIT
-                "   .align 8\n"
-                "   .quad 0b, 0b\n"
-#else
-                "   .align 4\n"
-                "   .long 0b, 0b\n"
-#endif
-                ".previous\n"
-                : "=m" (reipl_diag_rc1), "=m" (reipl_diag_rc2)
-		: "d" (3) : "cc", "4", "5" );
-}
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 89051e8a5d8d..f2a9165ca4f8 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -285,16 +285,9 @@ void (*_machine_power_off)(void) = machine_power_off_smp;
 /*
  * Reboot, halt and power_off routines for non SMP.
  */
-extern void reipl(unsigned long devno);
-extern void reipl_diag(void);
 static void do_machine_restart_nonsmp(char * __unused)
 {
-	reipl_diag();
-
-	if (MACHINE_IS_VM)
-		cpcmd ("IPL", NULL, 0, NULL);
-	else
-		reipl (0x10000 | S390_lowcore.ipl_device);
+	do_reipl();
 }
 
 static void do_machine_halt_nonsmp(void)
@@ -755,214 +748,3 @@ struct seq_operations cpuinfo_op = {
 	.show	= show_cpuinfo,
 };
 
-#define DEFINE_IPL_ATTR(_name, _format, _value)			\
-static ssize_t ipl_##_name##_show(struct subsystem *subsys,	\
-		char *page)					\
-{								\
-	return sprintf(page, _format, _value);			\
-}								\
-static struct subsys_attribute ipl_##_name##_attr =		\
-	__ATTR(_name, S_IRUGO, ipl_##_name##_show, NULL);
-
-DEFINE_IPL_ATTR(wwpn, "0x%016llx\n", (unsigned long long)
-		IPL_PARMBLOCK_START->fcp.wwpn);
-DEFINE_IPL_ATTR(lun, "0x%016llx\n", (unsigned long long)
-		IPL_PARMBLOCK_START->fcp.lun);
-DEFINE_IPL_ATTR(bootprog, "%lld\n", (unsigned long long)
-		IPL_PARMBLOCK_START->fcp.bootprog);
-DEFINE_IPL_ATTR(br_lba, "%lld\n", (unsigned long long)
-		IPL_PARMBLOCK_START->fcp.br_lba);
-
-enum ipl_type_type {
-	ipl_type_unknown,
-	ipl_type_ccw,
-	ipl_type_fcp,
-};
-
-static enum ipl_type_type
-get_ipl_type(void)
-{
-	struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START;
-
-	if (!IPL_DEVNO_VALID)
-		return ipl_type_unknown;
-	if (!IPL_PARMBLOCK_VALID)
-		return ipl_type_ccw;
-	if (ipl->hdr.header.version > IPL_MAX_SUPPORTED_VERSION)
-		return ipl_type_unknown;
-	if (ipl->fcp.pbt != IPL_TYPE_FCP)
-		return ipl_type_unknown;
-	return ipl_type_fcp;
-}
-
-static ssize_t
-ipl_type_show(struct subsystem *subsys, char *page)
-{
-	switch (get_ipl_type()) {
-	case ipl_type_ccw:
-		return sprintf(page, "ccw\n");
-	case ipl_type_fcp:
-		return sprintf(page, "fcp\n");
-	default:
-		return sprintf(page, "unknown\n");
-	}
-}
-
-static struct subsys_attribute ipl_type_attr = __ATTR_RO(ipl_type);
-
-static ssize_t
-ipl_device_show(struct subsystem *subsys, char *page)
-{
-	struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START;
-
-	switch (get_ipl_type()) {
-	case ipl_type_ccw:
-		return sprintf(page, "0.0.%04x\n", ipl_devno);
-	case ipl_type_fcp:
-		return sprintf(page, "0.0.%04x\n", ipl->fcp.devno);
-	default:
-		return 0;
-	}
-}
-
-static struct subsys_attribute ipl_device_attr =
-	__ATTR(device, S_IRUGO, ipl_device_show, NULL);
-
-static struct attribute *ipl_fcp_attrs[] = {
-	&ipl_type_attr.attr,
-	&ipl_device_attr.attr,
-	&ipl_wwpn_attr.attr,
-	&ipl_lun_attr.attr,
-	&ipl_bootprog_attr.attr,
-	&ipl_br_lba_attr.attr,
-	NULL,
-};
-
-static struct attribute_group ipl_fcp_attr_group = {
-	.attrs = ipl_fcp_attrs,
-};
-
-static struct attribute *ipl_ccw_attrs[] = {
-	&ipl_type_attr.attr,
-	&ipl_device_attr.attr,
-	NULL,
-};
-
-static struct attribute_group ipl_ccw_attr_group = {
-	.attrs = ipl_ccw_attrs,
-};
-
-static struct attribute *ipl_unknown_attrs[] = {
-	&ipl_type_attr.attr,
-	NULL,
-};
-
-static struct attribute_group ipl_unknown_attr_group = {
-	.attrs = ipl_unknown_attrs,
-};
-
-static ssize_t
-ipl_parameter_read(struct kobject *kobj, char *buf, loff_t off, size_t count)
-{
-	unsigned int size = IPL_PARMBLOCK_SIZE;
-
-	if (off > size)
-		return 0;
-	if (off + count > size)
-		count = size - off;
-
-	memcpy(buf, (void *) IPL_PARMBLOCK_START + off, count);
-	return count;
-}
-
-static struct bin_attribute ipl_parameter_attr = {
-	.attr = {
-		.name = "binary_parameter",
-		.mode = S_IRUGO,
-		.owner = THIS_MODULE,
-	},
-	.size = PAGE_SIZE,
-	.read = &ipl_parameter_read,
-};
-
-static ssize_t
-ipl_scp_data_read(struct kobject *kobj, char *buf, loff_t off, size_t count)
-{
-	unsigned int size =  IPL_PARMBLOCK_START->fcp.scp_data_len;
-	void *scp_data = &IPL_PARMBLOCK_START->fcp.scp_data;
-
-	if (off > size)
-		return 0;
-	if (off + count > size)
-		count = size - off;
-
-	memcpy(buf, scp_data + off, count);
-	return count;
-}
-
-static struct bin_attribute ipl_scp_data_attr = {
-	.attr = {
-		.name = "scp_data",
-		.mode = S_IRUGO,
-		.owner = THIS_MODULE,
-	},
-	.size = PAGE_SIZE,
-	.read = &ipl_scp_data_read,
-};
-
-static decl_subsys(ipl, NULL, NULL);
-
-static int ipl_register_fcp_files(void)
-{
-	int rc;
-
-	rc = sysfs_create_group(&ipl_subsys.kset.kobj,
-				&ipl_fcp_attr_group);
-	if (rc)
-		goto out;
-	rc = sysfs_create_bin_file(&ipl_subsys.kset.kobj,
-				   &ipl_parameter_attr);
-	if (rc)
-		goto out_ipl_parm;
-	rc = sysfs_create_bin_file(&ipl_subsys.kset.kobj,
-				   &ipl_scp_data_attr);
-	if (!rc)
-		goto out;
-
-	sysfs_remove_bin_file(&ipl_subsys.kset.kobj, &ipl_parameter_attr);
-
-out_ipl_parm:
-	sysfs_remove_group(&ipl_subsys.kset.kobj, &ipl_fcp_attr_group);
-out:
-	return rc;
-}
-
-static int __init
-ipl_device_sysfs_register(void) {
-	int rc;
-
-	rc = firmware_register(&ipl_subsys);
-	if (rc)
-		goto out;
-
-	switch (get_ipl_type()) {
-	case ipl_type_ccw:
-		rc = sysfs_create_group(&ipl_subsys.kset.kobj,
-					&ipl_ccw_attr_group);
-		break;
-	case ipl_type_fcp:
-		rc = ipl_register_fcp_files();
-		break;
-	default:
-		rc = sysfs_create_group(&ipl_subsys.kset.kobj,
-					&ipl_unknown_attr_group);
-		break;
-	}
-
-	if (rc)
-		firmware_unregister(&ipl_subsys);
-out:
-	return rc;
-}
-
-__initcall(ipl_device_sysfs_register);
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 8e03219eea76..b2e6f4c8d382 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -59,9 +59,6 @@ static struct task_struct *current_set[NR_CPUS];
 extern char vmhalt_cmd[];
 extern char vmpoff_cmd[];
 
-extern void reipl(unsigned long devno);
-extern void reipl_diag(void);
-
 static void smp_ext_bitcall(int, ec_bit_sig);
 static void smp_ext_bitcall_others(ec_bit_sig);
 
@@ -279,12 +276,7 @@ static void do_machine_restart(void * __unused)
 	 * interrupted by an external interrupt and s390irq
 	 * locks are always held disabled).
 	 */
-	reipl_diag();
-
-	if (MACHINE_IS_VM)
-		cpcmd ("IPL", NULL, 0, NULL);
-	else
-		reipl (0x10000 | S390_lowcore.ipl_device);
+	do_reipl();
 }
 
 void machine_restart_smp(char * __unused) 
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 89320c1ad825..050963f15802 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -841,14 +841,26 @@ __clear_subchannel_easy(struct subchannel_id schid)
 	return -EBUSY;
 }
 
-extern void do_reipl(unsigned long devno);
-static int
-__shutdown_subchannel_easy(struct subchannel_id schid, void *data)
+struct sch_match_id {
+	struct subchannel_id schid;
+	struct ccw_dev_id devid;
+	int rc;
+};
+
+static int __shutdown_subchannel_easy_and_match(struct subchannel_id schid,
+	void *data)
 {
 	struct schib schib;
+	struct sch_match_id *match_id = data;
 
 	if (stsch_err(schid, &schib))
 		return -ENXIO;
+	if (match_id && schib.pmcw.dnv &&
+		(schib.pmcw.dev == match_id->devid.devno) &&
+		(schid.ssid == match_id->devid.ssid)) {
+		match_id->schid = schid;
+		match_id->rc = 0;
+	}
 	if (!schib.pmcw.ena)
 		return 0;
 	switch(__disable_subchannel_easy(schid, &schib)) {
@@ -864,18 +876,36 @@ __shutdown_subchannel_easy(struct subchannel_id schid, void *data)
 	return 0;
 }
 
-void
-clear_all_subchannels(void)
+static int clear_all_subchannels_and_match(struct ccw_dev_id *devid,
+	struct subchannel_id *schid)
 {
+	struct sch_match_id match_id;
+
+	match_id.devid = *devid;
+	match_id.rc = -ENODEV;
 	local_irq_disable();
-	for_each_subchannel(__shutdown_subchannel_easy, NULL);
+	for_each_subchannel(__shutdown_subchannel_easy_and_match, &match_id);
+	if (match_id.rc == 0)
+		*schid = match_id.schid;
+	return match_id.rc;
 }
 
+
+void clear_all_subchannels(void)
+{
+	local_irq_disable();
+	for_each_subchannel(__shutdown_subchannel_easy_and_match, NULL);
+}
+
+extern void do_reipl_asm(__u32 schid);
+
 /* Make sure all subchannels are quiet before we re-ipl an lpar. */
-void
-reipl(unsigned long devno)
+void reipl_ccw_dev(struct ccw_dev_id *devid)
 {
-	clear_all_subchannels();
+	struct subchannel_id schid;
+
+	if (clear_all_subchannels_and_match(devid, &schid))
+		panic("IPL Device not found\n");
 	cio_reset_channel_paths();
-	do_reipl(devno);
+	do_reipl_asm(*((__u32*)&schid));
 }
diff --git a/include/asm-s390/cio.h b/include/asm-s390/cio.h
index 28fdd6e2b8ba..da063cd5f0a0 100644
--- a/include/asm-s390/cio.h
+++ b/include/asm-s390/cio.h
@@ -270,6 +270,11 @@ struct diag210 {
 	__u32 vrdccrft : 8;    /* real device feature (output) */
 } __attribute__ ((packed,aligned(4)));
 
+struct ccw_dev_id {
+	u8 ssid;
+	u16 devno;
+};
+
 extern int diag210(struct diag210 *addr);
 
 extern void wait_cons_dev(void);
@@ -280,6 +285,8 @@ extern void cio_reset_channel_paths(void);
 
 extern void css_schedule_reprobe(void);
 
+extern void reipl_ccw_dev(struct ccw_dev_id *id);
+
 #endif
 
 #endif
diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index 596c8b172104..2e3d4cca5e21 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -47,6 +47,7 @@
 #define __LC_PER_ATMID			0x096
 #define __LC_PER_ADDRESS		0x098
 #define __LC_PER_ACCESS_ID		0x0A1
+#define __LC_AR_MODE_ID			0x0A3
 
 #define __LC_SUBCHANNEL_ID              0x0B8
 #define __LC_SUBCHANNEL_NR              0x0BA
@@ -106,18 +107,28 @@
 #define __LC_INT_CLOCK			0xDE8
 #endif /* __s390x__ */
 
-#define __LC_PANIC_MAGIC                0xE00
 
+#define __LC_PANIC_MAGIC		0xE00
 #ifndef __s390x__
 #define __LC_PFAULT_INTPARM             0x080
 #define __LC_CPU_TIMER_SAVE_AREA        0x0D8
+#define __LC_CLOCK_COMP_SAVE_AREA	0x0E0
+#define __LC_PSW_SAVE_AREA		0x100
+#define __LC_PREFIX_SAVE_AREA		0x108
 #define __LC_AREGS_SAVE_AREA            0x120
+#define __LC_FPREGS_SAVE_AREA		0x160
 #define __LC_GPREGS_SAVE_AREA           0x180
 #define __LC_CREGS_SAVE_AREA            0x1C0
 #else /* __s390x__ */
 #define __LC_PFAULT_INTPARM             0x11B8
+#define __LC_FPREGS_SAVE_AREA		0x1200
 #define __LC_GPREGS_SAVE_AREA           0x1280
+#define __LC_PSW_SAVE_AREA		0x1300
+#define __LC_PREFIX_SAVE_AREA		0x1318
+#define __LC_FP_CREG_SAVE_AREA		0x131C
+#define __LC_TODREG_SAVE_AREA		0x1324
 #define __LC_CPU_TIMER_SAVE_AREA        0x1328
+#define __LC_CLOCK_COMP_SAVE_AREA	0x1331
 #define __LC_AREGS_SAVE_AREA            0x1340
 #define __LC_CREGS_SAVE_AREA            0x1380
 #endif /* __s390x__ */
diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h
index 02c96d57f0cf..4a1126d8439a 100644
--- a/include/asm-s390/setup.h
+++ b/include/asm-s390/setup.h
@@ -68,39 +68,59 @@ extern unsigned int console_irq;
 #define SET_CONSOLE_3215	do { console_mode = 2; } while (0)
 #define SET_CONSOLE_3270	do { console_mode = 3; } while (0)
 
-struct ipl_list_header {
-	u32 length;
-	u8  reserved[3];
+
+struct ipl_list_hdr {
+	u32 len;
+	u8  reserved1[3];
 	u8  version;
+	u32 blk0_len;
+	u8  pbt;
+	u8  flags;
+	u16 reserved2;
 } __attribute__((packed));
 
 struct ipl_block_fcp {
-	u32 length;
-	u8  pbt;
-	u8  reserved1[322-1];
+	u8  reserved1[313-1];
+	u8  opt;
+	u8  reserved2[3];
+	u16 reserved3;
 	u16 devno;
-	u8  reserved2[4];
+	u8  reserved4[4];
 	u64 wwpn;
 	u64 lun;
 	u32 bootprog;
-	u8  reserved3[12];
+	u8  reserved5[12];
 	u64 br_lba;
 	u32 scp_data_len;
-	u8  reserved4[260];
+	u8  reserved6[260];
 	u8  scp_data[];
 } __attribute__((packed));
 
+struct ipl_block_ccw {
+	u8  load_param[8];
+	u8  reserved1[84];
+	u8  reserved2[2];
+	u16 devno;
+	u8  vm_flags;
+	u8  reserved3[3];
+	u32 vm_parm_len;
+} __attribute__((packed));
+
 struct ipl_parameter_block {
+	struct ipl_list_hdr hdr;
 	union {
-		u32 length;
-		struct ipl_list_header header;
-	} hdr;
-	struct ipl_block_fcp fcp;
+		struct ipl_block_fcp fcp;
+		struct ipl_block_ccw ccw;
+	} ipl_info;
 } __attribute__((packed));
 
-#define IPL_MAX_SUPPORTED_VERSION (0)
+#define IPL_PARM_BLK_FCP_LEN (sizeof(struct ipl_list_hdr) + \
+			      sizeof(struct ipl_block_fcp))
 
-#define IPL_TYPE_FCP (0)
+#define IPL_PARM_BLK_CCW_LEN (sizeof(struct ipl_list_hdr) + \
+			      sizeof(struct ipl_block_ccw))
+
+#define IPL_MAX_SUPPORTED_VERSION (0)
 
 /*
  * IPL validity flags and parameters as detected in head.S
@@ -108,12 +128,14 @@ struct ipl_parameter_block {
 extern u32 ipl_parameter_flags;
 extern u16 ipl_devno;
 
+void do_reipl(void);
+
 #define IPL_DEVNO_VALID		(ipl_parameter_flags & 1)
 #define IPL_PARMBLOCK_VALID	(ipl_parameter_flags & 2)
 
 #define IPL_PARMBLOCK_START	((struct ipl_parameter_block *) \
 				 IPL_PARMBLOCK_ORIGIN)
-#define IPL_PARMBLOCK_SIZE	(IPL_PARMBLOCK_START->hdr.length)
+#define IPL_PARMBLOCK_SIZE	(IPL_PARMBLOCK_START->hdr.len)
 
 #else /* __ASSEMBLY__ */
 
-- 
cgit v1.2.3


From 39b083fe1c3c7b88939f6fa1b0b96e579f12e96f Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:51 +0200
Subject: [S390] empty function defines.

Use do { } while (0) constructs instead of empty defines to avoid
subtle compile bugs.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/qdio.h      | 16 ++++++++--------
 drivers/s390/scsi/zfcp_def.h |  8 ++++----
 include/asm-s390/dma.h       |  2 +-
 include/asm-s390/io.h        |  2 +-
 include/asm-s390/smp.h       |  2 +-
 5 files changed, 15 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h
index ceb3ab31ee08..124569362f02 100644
--- a/drivers/s390/cio/qdio.h
+++ b/drivers/s390/cio/qdio.h
@@ -191,49 +191,49 @@ enum qdio_irq_states {
 #if QDIO_VERBOSE_LEVEL>8
 #define QDIO_PRINT_STUPID(x...) printk( KERN_DEBUG QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_STUPID(x...)
+#define QDIO_PRINT_STUPID(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>7
 #define QDIO_PRINT_ALL(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_ALL(x...)
+#define QDIO_PRINT_ALL(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>6
 #define QDIO_PRINT_INFO(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_INFO(x...)
+#define QDIO_PRINT_INFO(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>5
 #define QDIO_PRINT_WARN(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_WARN(x...)
+#define QDIO_PRINT_WARN(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>4
 #define QDIO_PRINT_ERR(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_ERR(x...)
+#define QDIO_PRINT_ERR(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>3
 #define QDIO_PRINT_CRIT(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_CRIT(x...)
+#define QDIO_PRINT_CRIT(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>2
 #define QDIO_PRINT_ALERT(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_ALERT(x...)
+#define QDIO_PRINT_ALERT(x...) do { } while (0)
 #endif
 
 #if QDIO_VERBOSE_LEVEL>1
 #define QDIO_PRINT_EMERG(x...) printk( QDIO_PRINTK_HEADER x)
 #else
-#define QDIO_PRINT_EMERG(x...)
+#define QDIO_PRINT_EMERG(x...) do { } while (0)
 #endif
 
 #define HEXDUMP16(importance,header,ptr) \
diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h
index 94d1b74db356..7c84b3d4bd94 100644
--- a/drivers/s390/scsi/zfcp_def.h
+++ b/drivers/s390/scsi/zfcp_def.h
@@ -543,7 +543,7 @@ do { \
 } while (0)
 	
 #if ZFCP_LOG_LEVEL_LIMIT < ZFCP_LOG_LEVEL_NORMAL
-# define ZFCP_LOG_NORMAL(fmt, args...)
+# define ZFCP_LOG_NORMAL(fmt, args...)	do { } while (0)
 #else
 # define ZFCP_LOG_NORMAL(fmt, args...) \
 do { \
@@ -553,7 +553,7 @@ do { \
 #endif
 
 #if ZFCP_LOG_LEVEL_LIMIT < ZFCP_LOG_LEVEL_INFO
-# define ZFCP_LOG_INFO(fmt, args...)
+# define ZFCP_LOG_INFO(fmt, args...)	do { } while (0)
 #else
 # define ZFCP_LOG_INFO(fmt, args...) \
 do { \
@@ -563,14 +563,14 @@ do { \
 #endif
 
 #if ZFCP_LOG_LEVEL_LIMIT < ZFCP_LOG_LEVEL_DEBUG
-# define ZFCP_LOG_DEBUG(fmt, args...)
+# define ZFCP_LOG_DEBUG(fmt, args...)	do { } while (0)
 #else
 # define ZFCP_LOG_DEBUG(fmt, args...) \
 	ZFCP_LOG(ZFCP_LOG_LEVEL_DEBUG, fmt , ##args)
 #endif
 
 #if ZFCP_LOG_LEVEL_LIMIT < ZFCP_LOG_LEVEL_TRACE
-# define ZFCP_LOG_TRACE(fmt, args...)
+# define ZFCP_LOG_TRACE(fmt, args...)	do { } while (0)
 #else
 # define ZFCP_LOG_TRACE(fmt, args...) \
 	ZFCP_LOG(ZFCP_LOG_LEVEL_TRACE, fmt , ##args)
diff --git a/include/asm-s390/dma.h b/include/asm-s390/dma.h
index 02720c449cd8..7425c6af6cd4 100644
--- a/include/asm-s390/dma.h
+++ b/include/asm-s390/dma.h
@@ -11,6 +11,6 @@
 
 #define MAX_DMA_ADDRESS         0x80000000
 
-#define free_dma(x)
+#define free_dma(x)	do { } while (0)
 
 #endif /* _ASM_DMA_H */
diff --git a/include/asm-s390/io.h b/include/asm-s390/io.h
index d4614b35f423..a6cc27e77007 100644
--- a/include/asm-s390/io.h
+++ b/include/asm-s390/io.h
@@ -116,7 +116,7 @@ extern void iounmap(void *addr);
 #define outb(x,addr) ((void) writeb(x,addr))
 #define outb_p(x,addr) outb(x,addr)
 
-#define mmiowb()
+#define mmiowb()	do { } while (0)
 
 /*
  * Convert a physical pointer to a virtual kernel pointer for /dev/mem
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 657646054c5e..9fb02e9779c9 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -104,7 +104,7 @@ smp_call_function_on(void (*func) (void *info), void *info,
 #define smp_cpu_not_running(cpu)	1
 #define smp_get_cpu(cpu) ({ 0; })
 #define smp_put_cpu(cpu) ({ 0; })
-#define smp_setup_cpu_possible_map()
+#define smp_setup_cpu_possible_map()	do { } while (0)
 #endif
 
 #endif
-- 
cgit v1.2.3


From ba8ce5c6f0a15f08eae39880a0de296007f4a4e7 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 20 Sep 2006 15:58:56 +0200
Subject: [S390] #undef in unistd.h

Avoid using #undef in unistd.h.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/unistd.h | 170 ++++++++++++++++------------------------------
 1 file changed, 59 insertions(+), 111 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index aa7a243862e1..02b942d85c37 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -25,17 +25,12 @@
 #define __NR_unlink              10
 #define __NR_execve              11
 #define __NR_chdir               12
-#define __NR_time                13
 #define __NR_mknod               14
 #define __NR_chmod               15
-#define __NR_lchown              16
 #define __NR_lseek               19
 #define __NR_getpid              20
 #define __NR_mount               21
 #define __NR_umount              22
-#define __NR_setuid              23
-#define __NR_getuid              24
-#define __NR_stime               25
 #define __NR_ptrace              26
 #define __NR_alarm               27
 #define __NR_pause               29
@@ -51,11 +46,7 @@
 #define __NR_pipe                42
 #define __NR_times               43
 #define __NR_brk                 45
-#define __NR_setgid              46
-#define __NR_getgid              47
 #define __NR_signal              48
-#define __NR_geteuid             49
-#define __NR_getegid             50
 #define __NR_acct                51
 #define __NR_umount2             52
 #define __NR_ioctl               54
@@ -69,18 +60,13 @@
 #define __NR_getpgrp             65
 #define __NR_setsid              66
 #define __NR_sigaction           67
-#define __NR_setreuid            70
-#define __NR_setregid            71
 #define __NR_sigsuspend          72
 #define __NR_sigpending          73
 #define __NR_sethostname         74
 #define __NR_setrlimit           75
-#define __NR_getrlimit           76
 #define __NR_getrusage           77
 #define __NR_gettimeofday        78
 #define __NR_settimeofday        79
-#define __NR_getgroups           80
-#define __NR_setgroups           81
 #define __NR_symlink             83
 #define __NR_readlink            85
 #define __NR_uselib              86
@@ -92,12 +78,10 @@
 #define __NR_truncate            92
 #define __NR_ftruncate           93
 #define __NR_fchmod              94
-#define __NR_fchown              95
 #define __NR_getpriority         96
 #define __NR_setpriority         97
 #define __NR_statfs              99
 #define __NR_fstatfs            100
-#define __NR_ioperm             101
 #define __NR_socketcall         102
 #define __NR_syslog             103
 #define __NR_setitimer          104
@@ -131,11 +115,7 @@
 #define __NR_sysfs              135
 #define __NR_personality        136
 #define __NR_afs_syscall        137 /* Syscall for Andrew File System */
-#define __NR_setfsuid           138
-#define __NR_setfsgid           139
-#define __NR__llseek            140
 #define __NR_getdents           141
-#define __NR__newselect         142
 #define __NR_flock              143
 #define __NR_msync              144
 #define __NR_readv              145
@@ -157,13 +137,9 @@
 #define __NR_sched_rr_get_interval      161
 #define __NR_nanosleep          162
 #define __NR_mremap             163
-#define __NR_setresuid          164
-#define __NR_getresuid          165
 #define __NR_query_module       167
 #define __NR_poll               168
 #define __NR_nfsservctl         169
-#define __NR_setresgid          170
-#define __NR_getresgid          171
 #define __NR_prctl              172
 #define __NR_rt_sigreturn       173
 #define __NR_rt_sigaction       174
@@ -174,7 +150,6 @@
 #define __NR_rt_sigsuspend      179
 #define __NR_pread64            180
 #define __NR_pwrite64           181
-#define __NR_chown              182
 #define __NR_getcwd             183
 #define __NR_capget             184
 #define __NR_capset             185
@@ -183,39 +158,11 @@
 #define __NR_getpmsg		188
 #define __NR_putpmsg		189
 #define __NR_vfork		190
-#define __NR_ugetrlimit		191	/* SuS compliant getrlimit */
-#define __NR_mmap2		192
-#define __NR_truncate64		193
-#define __NR_ftruncate64	194
-#define __NR_stat64		195
-#define __NR_lstat64		196
-#define __NR_fstat64		197
-#define __NR_lchown32		198
-#define __NR_getuid32		199
-#define __NR_getgid32		200
-#define __NR_geteuid32		201
-#define __NR_getegid32		202
-#define __NR_setreuid32		203
-#define __NR_setregid32		204
-#define __NR_getgroups32	205
-#define __NR_setgroups32	206
-#define __NR_fchown32		207
-#define __NR_setresuid32	208
-#define __NR_getresuid32	209
-#define __NR_setresgid32	210
-#define __NR_getresgid32	211
-#define __NR_chown32		212
-#define __NR_setuid32		213
-#define __NR_setgid32		214
-#define __NR_setfsuid32		215
-#define __NR_setfsgid32		216
 #define __NR_pivot_root         217
 #define __NR_mincore            218
 #define __NR_madvise            219
 #define __NR_getdents64		220
-#define __NR_fcntl64		221
 #define __NR_readahead		222
-#define __NR_sendfile64		223
 #define __NR_setxattr		224
 #define __NR_lsetxattr		225
 #define __NR_fsetxattr		226
@@ -256,7 +203,6 @@
 #define __NR_clock_getres	(__NR_timer_create+7)
 #define __NR_clock_nanosleep	(__NR_timer_create+8)
 /* Number 263 is reserved for vserver */
-#define __NR_fadvise64_64	264
 #define __NR_statfs64		265
 #define __NR_fstatfs64		266
 #define __NR_remap_file_pages	267
@@ -285,7 +231,6 @@
 #define __NR_mknodat		290
 #define __NR_fchownat		291
 #define __NR_futimesat		292
-#define __NR_fstatat64		293
 #define __NR_unlinkat		294
 #define __NR_renameat		295
 #define __NR_linkat		296
@@ -310,62 +255,65 @@
  * have a different name although they do the same (e.g. __NR_chown32
  * is __NR_chown on 64 bit).
  */
-#ifdef __s390x__
-#undef  __NR_time
-#undef  __NR_lchown
-#undef  __NR_setuid
-#undef  __NR_getuid
-#undef  __NR_stime
-#undef  __NR_setgid
-#undef  __NR_getgid
-#undef  __NR_geteuid
-#undef  __NR_getegid
-#undef  __NR_setreuid
-#undef  __NR_setregid
-#undef  __NR_getrlimit
-#undef  __NR_getgroups
-#undef  __NR_setgroups
-#undef  __NR_fchown
-#undef  __NR_ioperm
-#undef  __NR_setfsuid
-#undef  __NR_setfsgid
-#undef  __NR__llseek
-#undef  __NR__newselect
-#undef  __NR_setresuid
-#undef  __NR_getresuid
-#undef  __NR_setresgid
-#undef  __NR_getresgid
-#undef  __NR_chown
-#undef  __NR_ugetrlimit
-#undef  __NR_mmap2
-#undef  __NR_truncate64
-#undef  __NR_ftruncate64
-#undef  __NR_stat64
-#undef  __NR_lstat64
-#undef  __NR_fstat64
-#undef  __NR_lchown32
-#undef  __NR_getuid32
-#undef  __NR_getgid32
-#undef  __NR_geteuid32
-#undef  __NR_getegid32
-#undef  __NR_setreuid32
-#undef  __NR_setregid32
-#undef  __NR_getgroups32
-#undef  __NR_setgroups32
-#undef  __NR_fchown32
-#undef  __NR_setresuid32
-#undef  __NR_getresuid32
-#undef  __NR_setresgid32
-#undef  __NR_getresgid32
-#undef  __NR_chown32
-#undef  __NR_setuid32
-#undef  __NR_setgid32
-#undef  __NR_setfsuid32
-#undef  __NR_setfsgid32
-#undef  __NR_fcntl64
-#undef  __NR_sendfile64
-#undef  __NR_fadvise64_64
-#undef  __NR_fstatat64
+#ifndef __s390x__
+
+#define __NR_time		 13
+#define __NR_lchown		 16
+#define __NR_setuid		 23
+#define __NR_getuid		 24
+#define __NR_stime		 25
+#define __NR_setgid		 46
+#define __NR_getgid		 47
+#define __NR_geteuid		 49
+#define __NR_getegid		 50
+#define __NR_setreuid		 70
+#define __NR_setregid		 71
+#define __NR_getrlimit		 76
+#define __NR_getgroups		 80
+#define __NR_setgroups		 81
+#define __NR_fchown		 95
+#define __NR_ioperm		101
+#define __NR_setfsuid		138
+#define __NR_setfsgid		139
+#define __NR__llseek		140
+#define __NR__newselect 	142
+#define __NR_setresuid		164
+#define __NR_getresuid		165
+#define __NR_setresgid		170
+#define __NR_getresgid		171
+#define __NR_chown		182
+#define __NR_ugetrlimit		191	/* SuS compliant getrlimit */
+#define __NR_mmap2		192
+#define __NR_truncate64		193
+#define __NR_ftruncate64	194
+#define __NR_stat64		195
+#define __NR_lstat64		196
+#define __NR_fstat64		197
+#define __NR_lchown32		198
+#define __NR_getuid32		199
+#define __NR_getgid32		200
+#define __NR_geteuid32		201
+#define __NR_getegid32		202
+#define __NR_setreuid32		203
+#define __NR_setregid32		204
+#define __NR_getgroups32	205
+#define __NR_setgroups32	206
+#define __NR_fchown32		207
+#define __NR_setresuid32	208
+#define __NR_getresuid32	209
+#define __NR_setresgid32	210
+#define __NR_getresgid32	211
+#define __NR_chown32		212
+#define __NR_setuid32		213
+#define __NR_setgid32		214
+#define __NR_setfsuid32		215
+#define __NR_setfsgid32		216
+#define __NR_fcntl64		221
+#define __NR_sendfile64		223
+#define __NR_fadvise64_64	264
+#define __NR_fstatat64		293
+
+#else
 
 #define __NR_select		142
 #define __NR_getrlimit		191	/* SuS compliant getrlimit */
-- 
cgit v1.2.3


From e87bfe51b5ca2db99dd680bbb1e8fe3c94b607df Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 20 Sep 2006 15:59:15 +0200
Subject: [S390] convert some assembler to C.

Convert GET_IPL_DEVICE assembler macro to C function.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/head.S    | 59 ----------------------------------------------
 arch/s390/kernel/head31.S  | 43 +++++++++++++++++++++++----------
 arch/s390/kernel/head64.S  | 39 +++++++++++++++++++-----------
 arch/s390/kernel/ipl.c     |  4 ++--
 drivers/s390/cio/cio.c     | 38 +++++++++++++++++++++++++++--
 include/asm-s390/lowcore.h |  1 +
 include/asm-s390/setup.h   |  8 ++++---
 7 files changed, 99 insertions(+), 93 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index a6e9bdb53591..0f1db268a8a9 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -481,65 +481,6 @@ start:
 	.byte 0xf0,0xf1,0xf2,0xf3,0xf4,0xf5,0xf6,0xf7 
 	.byte 0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff
 
-.macro GET_IPL_DEVICE
-.Lget_ipl_device:
-	l     %r1,0xb8			# get sid
-	sll   %r1,15			# test if subchannel is enabled
-	srl   %r1,31
-	ltr   %r1,%r1
-	bz    2f-.LPG1(%r13)		# subchannel disabled
-	l     %r1,0xb8
-	la    %r5,.Lipl_schib-.LPG1(%r13)
-	stsch 0(%r5)		        # get schib of subchannel
-	bnz   2f-.LPG1(%r13)		# schib not available
-	tm    5(%r5),0x01		# devno valid?
-	bno   2f-.LPG1(%r13)
-	la    %r6,ipl_parameter_flags-.LPG1(%r13)
-	oi    3(%r6),0x01		# set flag
-	la    %r2,ipl_devno-.LPG1(%r13)
-	mvc   0(2,%r2),6(%r5)		# store devno
-	tm    4(%r5),0x80		# qdio capable device?
-	bno   2f-.LPG1(%r13)
-	oi    3(%r6),0x02		# set flag
-
-	# copy ipl parameters
-
-	lhi   %r0,4096
-	l     %r2,20(%r0)		# get address of parameter list
-	lhi   %r3,IPL_PARMBLOCK_ORIGIN
-	st    %r3,20(%r0)
-	lhi   %r4,1
-	cr    %r2,%r3			# start parameters < destination ?
-	jl    0f
-	lhi   %r1,1			# copy direction is upwards
-	j     1f
-0:	lhi   %r1,-1			# copy direction is downwards
-	ar    %r2,%r0
-	ar    %r3,%r0
-	ar    %r2,%r1
-	ar    %r3,%r1
-1:	mvc   0(1,%r3),0(%r2)		# finally copy ipl parameters
-	ar    %r3,%r1
-	ar    %r2,%r1
-	sr    %r0,%r4
-	jne   1b
-	b     2f-.LPG1(%r13)
-
-	.align 4
-.Lipl_schib:
-	.rept 13
-	.long 0
-	.endr
-
-	.globl ipl_parameter_flags
-ipl_parameter_flags:
-	.long 0
-	.globl ipl_devno
-ipl_devno:
-	.word 0
-2:
-.endm
-
 #ifdef CONFIG_64BIT
 #include "head64.S"
 #else
diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S
index d8bb68a72527..1fa9fa1ca740 100644
--- a/arch/s390/kernel/head31.S
+++ b/arch/s390/kernel/head31.S
@@ -37,13 +37,23 @@ startup:basr	%r13,0			# get base
 
 startup_continue:
 	basr	%r13,0			# get base
-.LPG1:	GET_IPL_DEVICE
-	mvi	__LC_AR_MODE_ID,0	# set ESA flag (mode 0)
+.LPG1:	mvi	__LC_AR_MODE_ID,0	# set ESA flag (mode 0)
 	lctl	%c0,%c15,.Lctl-.LPG1(%r13) # load control registers
 	l	%r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area
 					# move IPL device to lowcore
 	mvc	__LC_IPLDEV(4),IPL_DEVICE-PARMAREA(%r12)
+#
+# Setup stack
+#
+	l	%r15,.Linittu-.LPG1(%r13)
+	mvc	__LC_CURRENT(4),__TI_task(%r15)
+	ahi	%r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union+THREAD_SIZE
+	st	%r15,__LC_KERNEL_STACK	# set end of kernel stack
+	ahi	%r15,-96
+	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # clear backchain
 
+	l	%r14,.Lipl_save_parameters-.LPG1(%r13)
+	basr	%r14,%r14
 #
 # clear bss memory
 #
@@ -115,6 +125,10 @@ startup_continue:
 	b	.Lfchunk-.LPG1(%r13)
 
 	.align 4
+.Lipl_save_parameters:
+	.long	ipl_save_parameters
+.Linittu:
+	.long	init_thread_union
 .Lpmask:
 	.byte	0
 .align 8
@@ -274,6 +288,20 @@ startup_continue:
 .Lbss_end:  .long _end
 .Lparmaddr: .long PARMAREA
 .Lsccbaddr: .long .Lsccb
+
+	.globl ipl_schib
+ipl_schib:
+	.rept 13
+	.long 0
+	.endr
+
+	.globl ipl_flags
+ipl_flags:
+	.long 0
+	.globl ipl_devno
+ipl_devno:
+	.word 0
+
 	.org	0x12000
 .globl s390_readinfo_sccb
 s390_readinfo_sccb:
@@ -305,16 +333,6 @@ s390_readinfo_sccb:
 	.globl	_stext
 _stext:	basr	%r13,0			# get base
 .LPG3:
-#
-# Setup stack
-#
-	l	%r15,.Linittu-.LPG3(%r13)
-	mvc	__LC_CURRENT(4),__TI_task(%r15)
-	ahi	%r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union+THREAD_SIZE
-	st	%r15,__LC_KERNEL_STACK	# set end of kernel stack
-	ahi	%r15,-96
-	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # clear backchain
-
 # check control registers
 	stctl	%c0,%c15,0(%r15)
 	oi	2(%r15),0x40		# enable sigp emergency signal
@@ -333,6 +351,5 @@ _stext:	basr	%r13,0			# get base
 #
 	.align	8
 .Ldw:	.long	0x000a0000,0x00000000
-.Linittu:.long	init_thread_union
 .Lstart:.long	start_kernel
 .Laregs:.long	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index c2005101fee1..1ebaa338aa7e 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -39,7 +39,6 @@ startup_continue:
 	basr  %r13,0			 # get base
 .LPG1:  sll   %r13,1                     # remove high order bit
         srl   %r13,1
-	GET_IPL_DEVICE
         lhi   %r1,1                      # mode 1 = esame
 	mvi   __LC_AR_MODE_ID,1		 # set esame flag
         slr   %r0,%r0                    # set cpuid to zero
@@ -49,7 +48,18 @@ startup_continue:
 	lg    %r12,.Lparmaddr-.LPG1(%r13)# pointer to parameter area
 					 # move IPL device to lowcore
         mvc   __LC_IPLDEV(4),IPL_DEVICE+4-PARMAREA(%r12)
+#
+# Setup stack
+#
+	larl  %r15,init_thread_union
+	lg    %r14,__TI_task(%r15)	# cache current in lowcore
+	stg   %r14,__LC_CURRENT
+	aghi  %r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union + THREAD_SIZE
+	stg   %r15,__LC_KERNEL_STACK	# set end of kernel stack
+	aghi  %r15,-160
+	xc    __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # clear backchain
 
+	brasl %r14,ipl_save_parameters
 #
 # clear bss memory
 #
@@ -269,6 +279,19 @@ startup_continue:
 .Lparmaddr:
 	.quad	PARMAREA
 
+	.globl ipl_schib
+ipl_schib:
+	.rept 13
+	.long 0
+	.endr
+
+	.globl ipl_flags
+ipl_flags:
+	.long 0
+	.globl ipl_devno
+ipl_devno:
+	.word 0
+
 	.org	0x12000
 .globl s390_readinfo_sccb
 s390_readinfo_sccb:
@@ -300,24 +323,12 @@ s390_readinfo_sccb:
         .globl _stext
 _stext:	basr  %r13,0                    # get base
 .LPG3:
-#
-# Setup stack
-#
-	larl  %r15,init_thread_union
-	lg    %r14,__TI_task(%r15)      # cache current in lowcore
-	stg   %r14,__LC_CURRENT
-        aghi  %r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union + THREAD_SIZE
-        stg   %r15,__LC_KERNEL_STACK    # set end of kernel stack
-        aghi  %r15,-160
-        xc    __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # clear backchain
-
 # check control registers
         stctg  %c0,%c15,0(%r15)
 	oi     6(%r15),0x40             # enable sigp emergency signal
 	oi     4(%r15),0x10             # switch on low address proctection
         lctlg  %c0,%c15,0(%r15)
 
-#
         lam    0,15,.Laregs-.LPG3(%r13) # load access regs needed by uaccess
         brasl  %r14,start_kernel        # go to C code
 #
@@ -325,7 +336,7 @@ _stext:	basr  %r13,0                    # get base
 #
         basr  %r13,0
 	lpswe .Ldw-.(%r13)           # load disabled wait psw
-#
+
             .align 8
 .Ldw:       .quad  0x0002000180000000,0x0000000000000000
 .Laregs:    .long  0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 105ee15a2b31..6555cc48e28f 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -189,9 +189,9 @@ static enum ipl_type ipl_get_type(void)
 {
 	struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START;
 
-	if (!IPL_DEVNO_VALID)
+	if (!(ipl_flags & IPL_DEVNO_VALID))
 		return IPL_TYPE_UNKNOWN;
-	if (!IPL_PARMBLOCK_VALID)
+	if (!(ipl_flags & IPL_PARMBLOCK_VALID))
 		return IPL_TYPE_CCW;
 	if (ipl->hdr.version > IPL_MAX_SUPPORTED_VERSION)
 		return IPL_TYPE_UNKNOWN;
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 050963f15802..61eb7caa1567 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -16,11 +16,10 @@
 #include <linux/device.h>
 #include <linux/kernel_stat.h>
 #include <linux/interrupt.h>
-
 #include <asm/cio.h>
 #include <asm/delay.h>
 #include <asm/irq.h>
-
+#include <asm/setup.h>
 #include "airq.h"
 #include "cio.h"
 #include "css.h"
@@ -909,3 +908,38 @@ void reipl_ccw_dev(struct ccw_dev_id *devid)
 	cio_reset_channel_paths();
 	do_reipl_asm(*((__u32*)&schid));
 }
+
+extern struct schib ipl_schib;
+
+/*
+ * ipl_save_parameters gets called very early. It is not allowed to access
+ * anything in the bss section at all. The bss section is not cleared yet,
+ * but may contain some ipl parameters written by the firmware.
+ * These parameters (if present) are copied to 0x2000.
+ * To avoid corruption of the ipl parameters, all variables used by this
+ * function must reside on the stack or in the data section.
+ */
+void ipl_save_parameters(void)
+{
+	struct subchannel_id schid;
+	unsigned int *ipl_ptr;
+	void *src, *dst;
+
+	schid = *(struct subchannel_id *)__LC_SUBCHANNEL_ID;
+	if (!schid.one)
+		return;
+	if (stsch(schid, &ipl_schib))
+		return;
+	if (!ipl_schib.pmcw.dnv)
+		return;
+	ipl_devno = ipl_schib.pmcw.dev;
+	ipl_flags |= IPL_DEVNO_VALID;
+	if (!ipl_schib.pmcw.qf)
+		return;
+	ipl_flags |= IPL_PARMBLOCK_VALID;
+	ipl_ptr = (unsigned int *)__LC_IPL_PARMBLOCK_PTR;
+	src = (void *)(unsigned long)*ipl_ptr;
+	dst = (void *)IPL_PARMBLOCK_ORIGIN;
+	memmove(dst, src, PAGE_SIZE);
+	*ipl_ptr = IPL_PARMBLOCK_ORIGIN;
+}
diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index 2e3d4cca5e21..18695d10dedf 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -35,6 +35,7 @@
 #define __LC_IO_NEW_PSW                 0x01f0
 #endif /* !__s390x__ */
 
+#define __LC_IPL_PARMBLOCK_PTR		0x014
 #define __LC_EXT_PARAMS                 0x080
 #define __LC_CPU_ADDRESS                0x084
 #define __LC_EXT_INT_CODE               0x086
diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h
index 4a1126d8439a..00c03e46689b 100644
--- a/include/asm-s390/setup.h
+++ b/include/asm-s390/setup.h
@@ -125,13 +125,15 @@ struct ipl_parameter_block {
 /*
  * IPL validity flags and parameters as detected in head.S
  */
-extern u32 ipl_parameter_flags;
+extern u32 ipl_flags;
 extern u16 ipl_devno;
 
 void do_reipl(void);
 
-#define IPL_DEVNO_VALID		(ipl_parameter_flags & 1)
-#define IPL_PARMBLOCK_VALID	(ipl_parameter_flags & 2)
+enum {
+	IPL_DEVNO_VALID	= 1,
+	IPL_PARMBLOCK_VALID = 2,
+};
 
 #define IPL_PARMBLOCK_START	((struct ipl_parameter_block *) \
 				 IPL_PARMBLOCK_ORIGIN)
-- 
cgit v1.2.3


From 1f38d61347203055b55e34083cce7a9cd8c529a9 Mon Sep 17 00:00:00 2001
From: Gerald Schaefer <geraldsc@de.ibm.com>
Date: Wed, 20 Sep 2006 15:59:26 +0200
Subject: [S390] cleanup appldata.

Introduce asm header that contains the appldata data structures and
the diag inline assembly.

Signed-off-by: Gerald Schaefer <geraldsc@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/appldata/appldata.h      | 16 -------
 arch/s390/appldata/appldata_base.c | 81 +++++-----------------------------
 arch/s390/appldata/appldata_os.c   |  1 +
 include/asm-s390/appldata.h        | 90 ++++++++++++++++++++++++++++++++++++++
 4 files changed, 103 insertions(+), 85 deletions(-)
 create mode 100644 include/asm-s390/appldata.h

(limited to 'include')

diff --git a/arch/s390/appldata/appldata.h b/arch/s390/appldata/appldata.h
index 71d65eb30650..0429481dea63 100644
--- a/arch/s390/appldata/appldata.h
+++ b/arch/s390/appldata/appldata.h
@@ -29,22 +29,6 @@
 #define CTL_APPLDATA_NET_SUM	2125
 #define CTL_APPLDATA_PROC	2126
 
-#ifndef CONFIG_64BIT
-
-#define APPLDATA_START_INTERVAL_REC 0x00	/* Function codes for */
-#define APPLDATA_STOP_REC	    0x01	/* DIAG 0xDC	  */
-#define APPLDATA_GEN_EVENT_RECORD   0x02
-#define APPLDATA_START_CONFIG_REC   0x03
-
-#else
-
-#define APPLDATA_START_INTERVAL_REC 0x80
-#define APPLDATA_STOP_REC	    0x81
-#define APPLDATA_GEN_EVENT_RECORD   0x82
-#define APPLDATA_START_CONFIG_REC   0x83
-
-#endif /* CONFIG_64BIT */
-
 #define P_INFO(x...)	printk(KERN_INFO MY_PRINT_NAME " info: " x)
 #define P_ERROR(x...)	printk(KERN_ERR MY_PRINT_NAME " error: " x)
 #define P_WARNING(x...)	printk(KERN_WARNING MY_PRINT_NAME " status: " x)
diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c
index a0a94e0ef8d1..b69ed742f981 100644
--- a/arch/s390/appldata/appldata_base.c
+++ b/arch/s390/appldata/appldata_base.c
@@ -14,20 +14,20 @@
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/errno.h>
-#include <asm/uaccess.h>
-#include <asm/io.h>
-#include <asm/smp.h>
 #include <linux/interrupt.h>
 #include <linux/proc_fs.h>
 #include <linux/page-flags.h>
 #include <linux/swap.h>
 #include <linux/pagemap.h>
 #include <linux/sysctl.h>
-#include <asm/timer.h>
-//#include <linux/kernel_stat.h>
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/workqueue.h>
+#include <asm/appldata.h>
+#include <asm/timer.h>
+#include <asm/uaccess.h>
+#include <asm/io.h>
+#include <asm/smp.h>
 
 #include "appldata.h"
 
@@ -39,34 +39,6 @@
 
 #define TOD_MICRO	0x01000			/* nr. of TOD clock units
 						   for 1 microsecond */
-
-/*
- * Parameter list for DIAGNOSE X'DC'
- */
-#ifndef CONFIG_64BIT
-struct appldata_parameter_list {
-	u16 diag;		/* The DIAGNOSE code X'00DC'          */
-	u8  function;		/* The function code for the DIAGNOSE */
-	u8  parlist_length;	/* Length of the parameter list       */
-	u32 product_id_addr;	/* Address of the 16-byte product ID  */
-	u16 reserved;
-	u16 buffer_length;	/* Length of the application data buffer  */
-	u32 buffer_addr;	/* Address of the application data buffer */
-};
-#else
-struct appldata_parameter_list {
-	u16 diag;
-	u8  function;
-	u8  parlist_length;
-	u32 unused01;
-	u16 reserved;
-	u16 buffer_length;
-	u32 unused02;
-	u64 product_id_addr;
-	u64 buffer_addr;
-};
-#endif /* CONFIG_64BIT */
-
 /*
  * /proc entries (sysctl)
  */
@@ -181,46 +153,17 @@ static void appldata_work_fn(void *data)
 int appldata_diag(char record_nr, u16 function, unsigned long buffer,
 			u16 length, char *mod_lvl)
 {
-	unsigned long ry;
-	struct appldata_product_id {
-		char prod_nr[7];			/* product nr. */
-		char prod_fn[2];			/* product function */
-		char record_nr;				/* record nr. */
-		char version_nr[2];			/* version */
-		char release_nr[2];			/* release */
-		char mod_lvl[2];			/* modification lvl. */
-	} appldata_product_id = {
-	/* all strings are EBCDIC, record_nr is byte */
+	struct appldata_product_id id = {
 		.prod_nr    = {0xD3, 0xC9, 0xD5, 0xE4,
-				0xE7, 0xD2, 0xD9},	/* "LINUXKR" */
-		.prod_fn    = {0xD5, 0xD3},		/* "NL" */
+			       0xE7, 0xD2, 0xD9},	/* "LINUXKR" */
+		.prod_fn    = 0xD5D3,			/* "NL" */
 		.record_nr  = record_nr,
-		.version_nr = {0xF2, 0xF6},		/* "26" */
-		.release_nr = {0xF0, 0xF1},		/* "01" */
-		.mod_lvl    = {mod_lvl[0], mod_lvl[1]},
-	};
-	struct appldata_parameter_list appldata_parameter_list = {
-				.diag = 0xDC,
-				.function = function,
-				.parlist_length =
-					sizeof(appldata_parameter_list),
-				.buffer_length = length,
-				.product_id_addr =
-					(unsigned long) &appldata_product_id,
-				.buffer_addr = virt_to_phys((void *) buffer)
+		.version_nr = 0xF2F6,			/* "26" */
+		.release_nr = 0xF0F1,			/* "01" */
+		.mod_lvl    = (mod_lvl[0]) << 8 | mod_lvl[1],
 	};
 
-	if (!MACHINE_IS_VM)
-		return -ENOSYS;
-	ry = -1;
-	asm volatile(
-			"diag %1,%0,0xDC\n\t"
-			: "=d" (ry)
-			: "d" (&appldata_parameter_list),
-			  "m" (appldata_parameter_list),
-			  "m" (appldata_product_id)
-			: "cc");
-	return (int) ry;
+	return appldata_asm(&id, function, (void *) buffer, length);
 }
 /************************ timer, work, DIAG <END> ****************************/
 
diff --git a/arch/s390/appldata/appldata_os.c b/arch/s390/appldata/appldata_os.c
index 161acc5c8a1b..76a15523ae9e 100644
--- a/arch/s390/appldata/appldata_os.c
+++ b/arch/s390/appldata/appldata_os.c
@@ -16,6 +16,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/netdevice.h>
 #include <linux/sched.h>
+#include <asm/appldata.h>
 #include <asm/smp.h>
 
 #include "appldata.h"
diff --git a/include/asm-s390/appldata.h b/include/asm-s390/appldata.h
new file mode 100644
index 000000000000..b1770703b706
--- /dev/null
+++ b/include/asm-s390/appldata.h
@@ -0,0 +1,90 @@
+/*
+ * include/asm-s390/appldata.h
+ *
+ * Copyright (C) IBM Corp. 2006
+ *
+ * Author(s): Melissa Howland <melissah@us.ibm.com>
+ */
+
+#ifndef _ASM_S390_APPLDATA_H
+#define _ASM_S390_APPLDATA_H
+
+#include <asm/io.h>
+
+#ifndef CONFIG_64BIT
+
+#define APPLDATA_START_INTERVAL_REC	0x00	/* Function codes for */
+#define APPLDATA_STOP_REC		0x01	/* DIAG 0xDC	      */
+#define APPLDATA_GEN_EVENT_REC		0x02
+#define APPLDATA_START_CONFIG_REC	0x03
+
+/*
+ * Parameter list for DIAGNOSE X'DC'
+ */
+struct appldata_parameter_list {
+	u16 diag;		/* The DIAGNOSE code X'00DC'	      */
+	u8  function;		/* The function code for the DIAGNOSE */
+	u8  parlist_length;	/* Length of the parameter list       */
+	u32 product_id_addr;	/* Address of the 16-byte product ID  */
+	u16 reserved;
+	u16 buffer_length;	/* Length of the application data buffer  */
+	u32 buffer_addr;	/* Address of the application data buffer */
+} __attribute__ ((packed));
+
+#else /* CONFIG_64BIT */
+
+#define APPLDATA_START_INTERVAL_REC	0x80
+#define APPLDATA_STOP_REC		0x81
+#define APPLDATA_GEN_EVENT_REC		0x82
+#define APPLDATA_START_CONFIG_REC	0x83
+
+/*
+ * Parameter list for DIAGNOSE X'DC'
+ */
+struct appldata_parameter_list {
+	u16 diag;
+	u8  function;
+	u8  parlist_length;
+	u32 unused01;
+	u16 reserved;
+	u16 buffer_length;
+	u32 unused02;
+	u64 product_id_addr;
+	u64 buffer_addr;
+} __attribute__ ((packed));
+
+#endif /* CONFIG_64BIT */
+
+struct appldata_product_id {
+	char prod_nr[7];	/* product number */
+	u16  prod_fn;		/* product function */
+	u8   record_nr; 	/* record number */
+	u16  version_nr;	/* version */
+	u16  release_nr;	/* release */
+	u16  mod_lvl;		/* modification level */
+} __attribute__ ((packed));
+
+static inline int appldata_asm(struct appldata_product_id *id,
+			       unsigned short fn, void *buffer,
+			       unsigned short length)
+{
+	struct appldata_parameter_list parm_list;
+	int ry;
+
+	if (!MACHINE_IS_VM)
+		return -ENOSYS;
+	parm_list.diag = 0xdc;
+	parm_list.function = fn;
+	parm_list.parlist_length = sizeof(parm_list);
+	parm_list.buffer_length = length;
+	parm_list.product_id_addr = (unsigned long) id;
+	parm_list.buffer_addr = virt_to_phys(buffer);
+	asm volatile(
+		"diag %1,%0,0xdc"
+		: "=d" (ry)
+		: "d" (&parm_list), "m" (parm_list), "m" (*id)
+		: "cc");
+	return ry;
+}
+
+#endif /* _ASM_S390_APPLDATA_H */
-- 
cgit v1.2.3


From 31b58088292c7f00f0b81088bfb557285b0b6247 Mon Sep 17 00:00:00 2001
From: Melissa Howland <melissah@us.ibm.com>
Date: Wed, 20 Sep 2006 15:59:34 +0200
Subject: [S390] Linux API for writing z/VM APPLDATA Monitor records.

This patch delivers a new Linux API in the form of a misc char
device that is useable from user space and allows write access
to the z/VM APPLDATA Monitor Records collected by the *MONITOR
System Service of z/VM.

Signed-off-by: Melissa Howland <melissah@us.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/defconfig           |   1 +
 drivers/s390/Kconfig          |   6 +
 drivers/s390/char/Makefile    |   1 +
 drivers/s390/char/monwriter.c | 292 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-s390/Kbuild       |   2 +-
 include/asm-s390/monwriter.h  |  33 +++++
 6 files changed, 334 insertions(+), 1 deletion(-)
 create mode 100644 drivers/s390/char/monwriter.c
 create mode 100644 include/asm-s390/monwriter.h

(limited to 'include')

diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index f1d4591eddbb..35da53986b1b 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -428,6 +428,7 @@ CONFIG_S390_TAPE_34XX=m
 # CONFIG_VMLOGRDR is not set
 # CONFIG_VMCP is not set
 # CONFIG_MONREADER is not set
+CONFIG_MONWRITER=m
 
 #
 # Cryptographic devices
diff --git a/drivers/s390/Kconfig b/drivers/s390/Kconfig
index bc4261e8b606..ae89b9b88743 100644
--- a/drivers/s390/Kconfig
+++ b/drivers/s390/Kconfig
@@ -213,6 +213,12 @@ config MONREADER
 	help
 	  Character device driver for reading z/VM monitor service records
 
+config MONWRITER
+	tristate "API for writing z/VM monitor service records"
+	default "m"
+	help
+	  Character device driver for writing z/VM monitor service records
+
 endmenu
 
 menu "Cryptographic devices"
diff --git a/drivers/s390/char/Makefile b/drivers/s390/char/Makefile
index 0c0162ff6c0c..c3e97b4fc186 100644
--- a/drivers/s390/char/Makefile
+++ b/drivers/s390/char/Makefile
@@ -28,3 +28,4 @@ obj-$(CONFIG_S390_TAPE) += tape.o tape_class.o
 obj-$(CONFIG_S390_TAPE_34XX) += tape_34xx.o
 obj-$(CONFIG_S390_TAPE_3590) += tape_3590.o
 obj-$(CONFIG_MONREADER) += monreader.o
+obj-$(CONFIG_MONWRITER) += monwriter.o
diff --git a/drivers/s390/char/monwriter.c b/drivers/s390/char/monwriter.c
new file mode 100644
index 000000000000..1e3939aeb8ab
--- /dev/null
+++ b/drivers/s390/char/monwriter.c
@@ -0,0 +1,292 @@
+/*
+ * drivers/s390/char/monwriter.c
+ *
+ * Character device driver for writing z/VM *MONITOR service records.
+ *
+ * Copyright (C) IBM Corp. 2006
+ *
+ * Author(s): Melissa Howland <Melissa.Howland@us.ibm.com>
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/ctype.h>
+#include <linux/poll.h>
+#include <asm/uaccess.h>
+#include <asm/ebcdic.h>
+#include <asm/io.h>
+#include <asm/appldata.h>
+#include <asm/monwriter.h>
+
+#define MONWRITE_MAX_DATALEN	4024
+
+static int mon_max_bufs = 255;
+
+struct mon_buf {
+	struct list_head list;
+	struct monwrite_hdr hdr;
+	int diag_done;
+	char *data;
+};
+
+struct mon_private {
+	struct list_head list;
+	struct monwrite_hdr hdr;
+	size_t hdr_to_read;
+	size_t data_to_read;
+	struct mon_buf *current_buf;
+	int mon_buf_count;
+};
+
+/*
+ * helper functions
+ */
+
+static int monwrite_diag(struct monwrite_hdr *myhdr, char *buffer, int fcn)
+{
+	struct appldata_product_id id;
+	int rc;
+
+	strcpy(id.prod_nr, "LNXAPPL");
+	id.prod_fn = myhdr->applid;
+	id.record_nr = myhdr->record_num;
+	id.version_nr = myhdr->version;
+	id.release_nr = myhdr->release;
+	id.mod_lvl = myhdr->mod_level;
+	rc = appldata_asm(&id, fcn, (void *) buffer, myhdr->datalen);
+	if (rc <= 0)
+		return rc;
+	if (rc == 5)
+		return -EPERM;
+	printk("DIAG X'DC' error with return code: %i\n", rc);
+	return -EINVAL;
+}
+
+static inline struct mon_buf *monwrite_find_hdr(struct mon_private *monpriv,
+						struct monwrite_hdr *monhdr)
+{
+	struct mon_buf *entry, *next;
+
+	list_for_each_entry_safe(entry, next, &monpriv->list, list)
+		if (entry->hdr.applid == monhdr->applid &&
+		    entry->hdr.record_num == monhdr->record_num &&
+		    entry->hdr.version == monhdr->version &&
+		    entry->hdr.release == monhdr->release &&
+		    entry->hdr.mod_level == monhdr->mod_level)
+			return entry;
+	return NULL;
+}
+
+static int monwrite_new_hdr(struct mon_private *monpriv)
+{
+	struct monwrite_hdr *monhdr = &monpriv->hdr;
+	struct mon_buf *monbuf;
+	int rc;
+
+	if (monhdr->datalen > MONWRITE_MAX_DATALEN ||
+	    monhdr->mon_function > MONWRITE_START_CONFIG ||
+	    monhdr->hdrlen != sizeof(struct monwrite_hdr))
+		return -EINVAL;
+	monbuf = monwrite_find_hdr(monpriv, monhdr);
+	if (monbuf) {
+		if (monhdr->mon_function == MONWRITE_STOP_INTERVAL) {
+			monhdr->datalen = monbuf->hdr.datalen;
+			rc = monwrite_diag(monhdr, monbuf->data,
+					   APPLDATA_STOP_REC);
+			list_del(&monbuf->list);
+			monpriv->mon_buf_count--;
+			kfree(monbuf->data);
+			kfree(monbuf);
+			monbuf = NULL;
+		}
+	} else {
+		if (monpriv->mon_buf_count >= mon_max_bufs)
+			return -ENOSPC;
+		monbuf = kzalloc(sizeof(struct mon_buf), GFP_KERNEL);
+		if (!monbuf)
+			return -ENOMEM;
+		monbuf->data = kzalloc(monbuf->hdr.datalen,
+				       GFP_KERNEL | GFP_DMA);
+		if (!monbuf->data) {
+			kfree(monbuf);
+			return -ENOMEM;
+		}
+		monbuf->hdr = *monhdr;
+		list_add_tail(&monbuf->list, &monpriv->list);
+		monpriv->mon_buf_count++;
+	}
+	monpriv->current_buf = monbuf;
+	return 0;
+}
+
+static int monwrite_new_data(struct mon_private *monpriv)
+{
+	struct monwrite_hdr *monhdr = &monpriv->hdr;
+	struct mon_buf *monbuf = monpriv->current_buf;
+	int rc = 0;
+
+	switch (monhdr->mon_function) {
+	case MONWRITE_START_INTERVAL:
+		if (!monbuf->diag_done) {
+			rc = monwrite_diag(monhdr, monbuf->data,
+					   APPLDATA_START_INTERVAL_REC);
+			monbuf->diag_done = 1;
+		}
+		break;
+	case MONWRITE_START_CONFIG:
+		if (!monbuf->diag_done) {
+			rc = monwrite_diag(monhdr, monbuf->data,
+					   APPLDATA_START_CONFIG_REC);
+			monbuf->diag_done = 1;
+		}
+		break;
+	case MONWRITE_GEN_EVENT:
+		rc = monwrite_diag(monhdr, monbuf->data,
+				   APPLDATA_GEN_EVENT_REC);
+		list_del(&monpriv->current_buf->list);
+		kfree(monpriv->current_buf->data);
+		kfree(monpriv->current_buf);
+		monpriv->current_buf = NULL;
+		break;
+	default:
+		/* monhdr->mon_function is checked in monwrite_new_hdr */
+		BUG();
+	}
+	return rc;
+}
+
+/*
+ * file operations
+ */
+
+static int monwrite_open(struct inode *inode, struct file *filp)
+{
+	struct mon_private *monpriv;
+
+	monpriv = kzalloc(sizeof(struct mon_private), GFP_KERNEL);
+	if (!monpriv)
+		return -ENOMEM;
+	INIT_LIST_HEAD(&monpriv->list);
+	monpriv->hdr_to_read = sizeof(monpriv->hdr);
+	filp->private_data = monpriv;
+	return nonseekable_open(inode, filp);
+}
+
+static int monwrite_close(struct inode *inode, struct file *filp)
+{
+	struct mon_private *monpriv = filp->private_data;
+	struct mon_buf *entry, *next;
+
+	list_for_each_entry_safe(entry, next, &monpriv->list, list) {
+		if (entry->hdr.mon_function != MONWRITE_GEN_EVENT)
+			monwrite_diag(&entry->hdr, entry->data,
+				      APPLDATA_STOP_REC);
+		monpriv->mon_buf_count--;
+		list_del(&entry->list);
+		kfree(entry->data);
+		kfree(entry);
+	}
+	kfree(monpriv);
+	return 0;
+}
+
+static ssize_t monwrite_write(struct file *filp, const char __user *data,
+			      size_t count, loff_t *ppos)
+{
+	struct mon_private *monpriv = filp->private_data;
+	size_t len, written;
+	void *to;
+	int rc;
+
+	for (written = 0; written < count; ) {
+		if (monpriv->hdr_to_read) {
+			len = min(count - written, monpriv->hdr_to_read);
+			to = (char *) &monpriv->hdr +
+				sizeof(monpriv->hdr) - monpriv->hdr_to_read;
+			if (copy_from_user(to, data + written, len)) {
+				rc = -EFAULT;
+				goto out_error;
+			}
+			monpriv->hdr_to_read -= len;
+			written += len;
+			if (monpriv->hdr_to_read > 0)
+				continue;
+			rc = monwrite_new_hdr(monpriv);
+			if (rc)
+				goto out_error;
+			monpriv->data_to_read = monpriv->current_buf ?
+				monpriv->current_buf->hdr.datalen : 0;
+		}
+
+		if (monpriv->data_to_read) {
+			len = min(count - written, monpriv->data_to_read);
+			to = monpriv->current_buf->data +
+				monpriv->hdr.datalen - monpriv->data_to_read;
+			if (copy_from_user(to, data + written, len)) {
+				rc = -EFAULT;
+				goto out_error;
+			}
+			monpriv->data_to_read -= len;
+			written += len;
+			if (monpriv->data_to_read > 0)
+				continue;
+			rc = monwrite_new_data(monpriv);
+			if (rc)
+				goto out_error;
+		}
+		monpriv->hdr_to_read = sizeof(monpriv->hdr);
+	}
+	return written;
+
+out_error:
+	monpriv->data_to_read = 0;
+	monpriv->hdr_to_read = sizeof(struct monwrite_hdr);
+	return rc;
+}
+
+static struct file_operations monwrite_fops = {
+	.owner	 = THIS_MODULE,
+	.open	 = &monwrite_open,
+	.release = &monwrite_close,
+	.write	 = &monwrite_write,
+};
+
+static struct miscdevice mon_dev = {
+	.name	= "monwriter",
+	.fops	= &monwrite_fops,
+	.minor	= MISC_DYNAMIC_MINOR,
+};
+
+/*
+ * module init/exit
+ */
+
+static int __init mon_init(void)
+{
+	if (MACHINE_IS_VM)
+		return misc_register(&mon_dev);
+	else
+		return -ENODEV;
+}
+
+static void __exit mon_exit(void)
+{
+	WARN_ON(misc_deregister(&mon_dev) != 0);
+}
+
+module_init(mon_init);
+module_exit(mon_exit);
+
+module_param_named(max_bufs, mon_max_bufs, int, 0644);
+MODULE_PARM_DESC(max_bufs, "Maximum number of sample monitor data buffers"
+		 "that can be active at one time");
+
+MODULE_AUTHOR("Melissa Howland <Melissa.Howland@us.ibm.com>");
+MODULE_DESCRIPTION("Character device driver for writing z/VM "
+		   "APPLDATA monitor records.");
+MODULE_LICENSE("GPL");
diff --git a/include/asm-s390/Kbuild b/include/asm-s390/Kbuild
index ed8955f49e47..979145026a29 100644
--- a/include/asm-s390/Kbuild
+++ b/include/asm-s390/Kbuild
@@ -1,4 +1,4 @@
 include include/asm-generic/Kbuild.asm
 
 unifdef-y += cmb.h debug.h
-header-y += dasd.h qeth.h tape390.h ucontext.h vtoc.h z90crypt.h
+header-y += dasd.h monwriter.h qeth.h tape390.h ucontext.h vtoc.h z90crypt.h
diff --git a/include/asm-s390/monwriter.h b/include/asm-s390/monwriter.h
new file mode 100644
index 000000000000..f0cbf96c52e6
--- /dev/null
+++ b/include/asm-s390/monwriter.h
@@ -0,0 +1,33 @@
+/*
+ * include/asm-s390/monwriter.h
+ *
+ * Copyright (C) IBM Corp. 2006
+ * Character device driver for writing z/VM APPLDATA monitor records
+ * Version 1.0
+ * Author(s): Melissa Howland <melissah@us.ibm.com>
+ *
+ */
+
+#ifndef _ASM_390_MONWRITER_H
+#define _ASM_390_MONWRITER_H
+
+/* mon_function values */
+#define MONWRITE_START_INTERVAL	0x00 /* start interval recording */
+#define MONWRITE_STOP_INTERVAL	0x01 /* stop interval or config recording */
+#define MONWRITE_GEN_EVENT	0x02 /* generate event record */
+#define MONWRITE_START_CONFIG	0x03 /* start configuration recording */
+
+/* the header the app uses in its write() data */
+struct monwrite_hdr {
+	unsigned char mon_function;
+	unsigned short applid;
+	unsigned char record_num;
+	unsigned short version;
+	unsigned short release;
+	unsigned short mod_level;
+	unsigned short datalen;
+	unsigned char hdrlen;
+
+} __attribute__((packed));
+
+#endif /* _ASM_390_MONWRITER_H */
-- 
cgit v1.2.3


From 9282ed929758b82f448a40d3c17319d794970624 Mon Sep 17 00:00:00 2001
From: Gerald Schaefer <geraldsc@de.ibm.com>
Date: Wed, 20 Sep 2006 15:59:37 +0200
Subject: [S390] Cleanup in page table related code.

Changed and simplified some page table related #defines and code.

Signed-off-by: Gerald Schaefer <geraldsc@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/mm/init.c        |  36 +++++--------
 include/asm-s390/pgalloc.h |  67 ++++++++++++------------
 include/asm-s390/pgtable.h | 124 +++++++++++++++++++++------------------------
 3 files changed, 106 insertions(+), 121 deletions(-)

(limited to 'include')

diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index 6e6b6de77770..cfd9b8f7a523 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -108,16 +108,23 @@ void __init paging_init(void)
         unsigned long pgdir_k = (__pa(swapper_pg_dir) & PAGE_MASK) | _KERNSEG_TABLE;
         static const int ssm_mask = 0x04000000L;
 	unsigned long ro_start_pfn, ro_end_pfn;
+	unsigned long zones_size[MAX_NR_ZONES];
 
 	ro_start_pfn = PFN_DOWN((unsigned long)&__start_rodata);
 	ro_end_pfn = PFN_UP((unsigned long)&__end_rodata);
 
+	memset(zones_size, 0, sizeof(zones_size));
+	zones_size[ZONE_DMA] = max_low_pfn;
+	free_area_init_node(0, &contig_page_data, zones_size,
+			    __pa(PAGE_OFFSET) >> PAGE_SHIFT,
+			    zholes_size);
+
 	/* unmap whole virtual address space */
 	
         pg_dir = swapper_pg_dir;
 
-	for (i=0;i<KERNEL_PGD_PTRS;i++) 
-	        pmd_clear((pmd_t*)pg_dir++);
+	for (i = 0; i < PTRS_PER_PGD; i++)
+		pmd_clear((pmd_t *) pg_dir++);
 		
 	/*
 	 * map whole physical memory to virtual memory (identity mapping) 
@@ -131,10 +138,7 @@ void __init paging_init(void)
                  */
 		pg_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
 
-                pg_dir->pgd0 =  (_PAGE_TABLE | __pa(pg_table));
-                pg_dir->pgd1 =  (_PAGE_TABLE | (__pa(pg_table)+1024));
-                pg_dir->pgd2 =  (_PAGE_TABLE | (__pa(pg_table)+2048));
-                pg_dir->pgd3 =  (_PAGE_TABLE | (__pa(pg_table)+3072));
+		pmd_populate_kernel(&init_mm, (pmd_t *) pg_dir, pg_table);
                 pg_dir++;
 
                 for (tmp = 0 ; tmp < PTRS_PER_PTE ; tmp++,pg_table++) {
@@ -143,8 +147,8 @@ void __init paging_init(void)
 			else
 				pte = pfn_pte(pfn, PAGE_KERNEL);
                         if (pfn >= max_low_pfn)
-                                pte_clear(&init_mm, 0, &pte);
-                        set_pte(pg_table, pte);
+				pte_val(pte) = _PAGE_TYPE_EMPTY;
+			set_pte(pg_table, pte);
                         pfn++;
                 }
         }
@@ -159,16 +163,6 @@ void __init paging_init(void)
 			     : : "m" (pgdir_k), "m" (ssm_mask));
 
         local_flush_tlb();
-
-	{
-		unsigned long zones_size[MAX_NR_ZONES];
-
-		memset(zones_size, 0, sizeof(zones_size));
-		zones_size[ZONE_DMA] = max_low_pfn;
-		free_area_init_node(0, &contig_page_data, zones_size,
-				    __pa(PAGE_OFFSET) >> PAGE_SHIFT,
-				    zholes_size);
-	}
         return;
 }
 
@@ -236,10 +230,8 @@ void __init paging_init(void)
 					pte = pfn_pte(pfn, __pgprot(_PAGE_RO));
 				else
 					pte = pfn_pte(pfn, PAGE_KERNEL);
-                                if (pfn >= max_low_pfn) {
-                                        pte_clear(&init_mm, 0, &pte); 
-                                        continue;
-                                }
+				if (pfn >= max_low_pfn)
+					pte_val(pte) = _PAGE_TYPE_EMPTY;
                                 set_pte(pt_dir, pte);
                                 pfn++;
                         }
diff --git a/include/asm-s390/pgalloc.h b/include/asm-s390/pgalloc.h
index a78e853e0dd5..803bc7064418 100644
--- a/include/asm-s390/pgalloc.h
+++ b/include/asm-s390/pgalloc.h
@@ -21,6 +21,16 @@
 
 extern void diag10(unsigned long addr);
 
+/*
+ * Page allocation orders.
+ */
+#ifndef __s390x__
+# define PGD_ALLOC_ORDER	1
+#else /* __s390x__ */
+# define PMD_ALLOC_ORDER	2
+# define PGD_ALLOC_ORDER	2
+#endif /* __s390x__ */
+
 /*
  * Allocate and free page tables. The xxx_kernel() versions are
  * used to allocate a kernel page table - this turns on ASN bits
@@ -29,30 +39,23 @@ extern void diag10(unsigned long addr);
 
 static inline pgd_t *pgd_alloc(struct mm_struct *mm)
 {
-	pgd_t *pgd;
+	pgd_t *pgd = (pgd_t *) __get_free_pages(GFP_KERNEL, PGD_ALLOC_ORDER);
 	int i;
 
+	if (!pgd)
+		return NULL;
+	for (i = 0; i < PTRS_PER_PGD; i++)
 #ifndef __s390x__
-	pgd = (pgd_t *) __get_free_pages(GFP_KERNEL,1);
-        if (pgd != NULL)
-		for (i = 0; i < USER_PTRS_PER_PGD; i++)
-			pmd_clear(pmd_offset(pgd + i, i*PGDIR_SIZE));
-#else /* __s390x__ */
-	pgd = (pgd_t *) __get_free_pages(GFP_KERNEL,2);
-        if (pgd != NULL)
-		for (i = 0; i < PTRS_PER_PGD; i++)
-			pgd_clear(pgd + i);
-#endif /* __s390x__ */
+		pmd_clear(pmd_offset(pgd + i, i*PGDIR_SIZE));
+#else
+		pgd_clear(pgd + i);
+#endif
 	return pgd;
 }
 
 static inline void pgd_free(pgd_t *pgd)
 {
-#ifndef __s390x__
-        free_pages((unsigned long) pgd, 1);
-#else /* __s390x__ */
-        free_pages((unsigned long) pgd, 2);
-#endif /* __s390x__ */
+	free_pages((unsigned long) pgd, PGD_ALLOC_ORDER);
 }
 
 #ifndef __s390x__
@@ -68,20 +71,19 @@ static inline void pgd_free(pgd_t *pgd)
 #else /* __s390x__ */
 static inline pmd_t * pmd_alloc_one(struct mm_struct *mm, unsigned long vmaddr)
 {
-	pmd_t *pmd;
-        int i;
+	pmd_t *pmd = (pmd_t *) __get_free_pages(GFP_KERNEL, PMD_ALLOC_ORDER);
+	int i;
 
-	pmd = (pmd_t *) __get_free_pages(GFP_KERNEL, 2);
-	if (pmd != NULL) {
-		for (i=0; i < PTRS_PER_PMD; i++)
-			pmd_clear(pmd+i);
-	}
+	if (!pmd)
+		return NULL;
+	for (i=0; i < PTRS_PER_PMD; i++)
+		pmd_clear(pmd + i);
 	return pmd;
 }
 
 static inline void pmd_free (pmd_t *pmd)
 {
-	free_pages((unsigned long) pmd, 2);
+	free_pages((unsigned long) pmd, PMD_ALLOC_ORDER);
 }
 
 #define __pmd_free_tlb(tlb,pmd)			\
@@ -123,15 +125,14 @@ pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *page)
 static inline pte_t *
 pte_alloc_one_kernel(struct mm_struct *mm, unsigned long vmaddr)
 {
-	pte_t *pte;
-        int i;
-
-	pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT);
-	if (pte != NULL) {
-		for (i=0; i < PTRS_PER_PTE; i++) {
-			pte_clear(mm, vmaddr, pte+i);
-			vmaddr += PAGE_SIZE;
-		}
+	pte_t *pte = (pte_t *) __get_free_page(GFP_KERNEL|__GFP_REPEAT);
+	int i;
+
+	if (!pte)
+		return NULL;
+	for (i=0; i < PTRS_PER_PTE; i++) {
+		pte_clear(mm, vmaddr, pte + i);
+		vmaddr += PAGE_SIZE;
 	}
 	return pte;
 }
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index 24312387fa24..1a07028d575e 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -89,19 +89,6 @@ extern char empty_zero_page[PAGE_SIZE];
 # define PTRS_PER_PGD    2048
 #endif /* __s390x__ */
 
-/*
- * pgd entries used up by user/kernel:
- */
-#ifndef __s390x__
-# define USER_PTRS_PER_PGD  512
-# define USER_PGD_PTRS      512
-# define KERNEL_PGD_PTRS    512
-#else /* __s390x__ */
-# define USER_PTRS_PER_PGD  2048
-# define USER_PGD_PTRS      2048
-# define KERNEL_PGD_PTRS    2048
-#endif /* __s390x__ */
-
 #define FIRST_USER_ADDRESS  0
 
 #define pte_ERROR(e) \
@@ -216,12 +203,14 @@ extern char empty_zero_page[PAGE_SIZE];
 #define _PAGE_RO        0x200          /* HW read-only                     */
 #define _PAGE_INVALID   0x400          /* HW invalid                       */
 
-/* Mask and four different kinds of invalid pages. */
-#define _PAGE_INVALID_MASK	0x601
-#define _PAGE_INVALID_EMPTY	0x400
-#define _PAGE_INVALID_NONE	0x401
-#define _PAGE_INVALID_SWAP	0x600
-#define _PAGE_INVALID_FILE	0x601
+/* Mask and six different types of pages. */
+#define _PAGE_TYPE_MASK		0x601
+#define _PAGE_TYPE_EMPTY	0x400
+#define _PAGE_TYPE_NONE		0x401
+#define _PAGE_TYPE_SWAP		0x600
+#define _PAGE_TYPE_FILE		0x601
+#define _PAGE_TYPE_RO		0x200
+#define _PAGE_TYPE_RW		0x000
 
 #ifndef __s390x__
 
@@ -280,15 +269,14 @@ extern char empty_zero_page[PAGE_SIZE];
 #endif /* __s390x__ */
 
 /*
- * No mapping available
+ * Page protection definitions.
  */
-#define PAGE_NONE_SHARED  __pgprot(_PAGE_INVALID_NONE)
-#define PAGE_NONE_PRIVATE __pgprot(_PAGE_INVALID_NONE)
-#define PAGE_RO_SHARED	  __pgprot(_PAGE_RO)
-#define PAGE_RO_PRIVATE	  __pgprot(_PAGE_RO)
-#define PAGE_COPY	  __pgprot(_PAGE_RO)
-#define PAGE_SHARED	  __pgprot(0)
-#define PAGE_KERNEL	  __pgprot(0)
+#define PAGE_NONE	__pgprot(_PAGE_TYPE_NONE)
+#define PAGE_RO		__pgprot(_PAGE_TYPE_RO)
+#define PAGE_RW		__pgprot(_PAGE_TYPE_RW)
+
+#define PAGE_KERNEL	PAGE_RW
+#define PAGE_COPY	PAGE_RO
 
 /*
  * The S390 can't do page protection for execute, and considers that the
@@ -296,23 +284,23 @@ extern char empty_zero_page[PAGE_SIZE];
  * the closest we can get..
  */
          /*xwr*/
-#define __P000  PAGE_NONE_PRIVATE
-#define __P001  PAGE_RO_PRIVATE
-#define __P010  PAGE_COPY
-#define __P011  PAGE_COPY
-#define __P100  PAGE_RO_PRIVATE
-#define __P101  PAGE_RO_PRIVATE
-#define __P110  PAGE_COPY
-#define __P111  PAGE_COPY
-
-#define __S000  PAGE_NONE_SHARED
-#define __S001  PAGE_RO_SHARED
-#define __S010  PAGE_SHARED
-#define __S011  PAGE_SHARED
-#define __S100  PAGE_RO_SHARED
-#define __S101  PAGE_RO_SHARED
-#define __S110  PAGE_SHARED
-#define __S111  PAGE_SHARED
+#define __P000	PAGE_NONE
+#define __P001	PAGE_RO
+#define __P010	PAGE_RO
+#define __P011	PAGE_RO
+#define __P100	PAGE_RO
+#define __P101	PAGE_RO
+#define __P110	PAGE_RO
+#define __P111	PAGE_RO
+
+#define __S000	PAGE_NONE
+#define __S001	PAGE_RO
+#define __S010	PAGE_RW
+#define __S011	PAGE_RW
+#define __S100	PAGE_RO
+#define __S101	PAGE_RO
+#define __S110	PAGE_RW
+#define __S111	PAGE_RW
 
 /*
  * Certain architectures need to do special things when PTEs
@@ -377,18 +365,18 @@ static inline int pmd_bad(pmd_t pmd)
 
 static inline int pte_none(pte_t pte)
 {
-	return (pte_val(pte) & _PAGE_INVALID_MASK) == _PAGE_INVALID_EMPTY;
+	return (pte_val(pte) & _PAGE_TYPE_MASK) == _PAGE_TYPE_EMPTY;
 }
 
 static inline int pte_present(pte_t pte)
 {
 	return !(pte_val(pte) & _PAGE_INVALID) ||
-		(pte_val(pte) & _PAGE_INVALID_MASK) == _PAGE_INVALID_NONE;
+		(pte_val(pte) & _PAGE_TYPE_MASK) == _PAGE_TYPE_NONE;
 }
 
 static inline int pte_file(pte_t pte)
 {
-	return (pte_val(pte) & _PAGE_INVALID_MASK) == _PAGE_INVALID_FILE;
+	return (pte_val(pte) & _PAGE_TYPE_MASK) == _PAGE_TYPE_FILE;
 }
 
 #define pte_same(a,b)	(pte_val(a) == pte_val(b))
@@ -461,7 +449,7 @@ static inline void pmd_clear(pmd_t * pmdp)
 
 static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-	pte_val(*ptep) = _PAGE_INVALID_EMPTY;
+	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
 }
 
 /*
@@ -477,7 +465,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 static inline pte_t pte_wrprotect(pte_t pte)
 {
-	/* Do not clobber _PAGE_INVALID_NONE pages!  */
+	/* Do not clobber _PAGE_TYPE_NONE pages!  */
 	if (!(pte_val(pte) & _PAGE_INVALID))
 		pte_val(pte) |= _PAGE_RO;
 	return pte;
@@ -556,26 +544,30 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 	return pte;
 }
 
-static inline pte_t
-ptep_clear_flush(struct vm_area_struct *vma,
-		 unsigned long address, pte_t *ptep)
+static inline void __ptep_ipte(unsigned long address, pte_t *ptep)
 {
-	pte_t pte = *ptep;
+	if (!(pte_val(*ptep) & _PAGE_INVALID)) {
 #ifndef __s390x__
-	if (!(pte_val(pte) & _PAGE_INVALID)) {
 		/* S390 has 1mb segments, we are emulating 4MB segments */
 		pte_t *pto = (pte_t *) (((unsigned long) ptep) & 0x7ffffc00);
-		__asm__ __volatile__ ("ipte %2,%3"
-				      : "=m" (*ptep) : "m" (*ptep),
-				        "a" (pto), "a" (address) );
+#else
+		/* ipte in zarch mode can do the math */
+		pte_t *pto = ptep;
+#endif
+		asm volatile ("ipte %2,%3"
+			      : "=m" (*ptep) : "m" (*ptep),
+				"a" (pto), "a" (address) );
 	}
-#else /* __s390x__ */
-	if (!(pte_val(pte) & _PAGE_INVALID)) 
-		__asm__ __volatile__ ("ipte %2,%3"
-				      : "=m" (*ptep) : "m" (*ptep),
-				        "a" (ptep), "a" (address) );
-#endif /* __s390x__ */
-	pte_val(*ptep) = _PAGE_INVALID_EMPTY;
+	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
+}
+
+static inline pte_t
+ptep_clear_flush(struct vm_area_struct *vma,
+		 unsigned long address, pte_t *ptep)
+{
+	pte_t pte = *ptep;
+
+	__ptep_ipte(address, ptep);
 	return pte;
 }
 
@@ -755,7 +747,7 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
 {
 	pte_t pte;
 	offset &= __SWP_OFFSET_MASK;
-	pte_val(pte) = _PAGE_INVALID_SWAP | ((type & 0x1f) << 2) |
+	pte_val(pte) = _PAGE_TYPE_SWAP | ((type & 0x1f) << 2) |
 		((offset & 1UL) << 7) | ((offset & ~1UL) << 11);
 	return pte;
 }
@@ -778,7 +770,7 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
 
 #define pgoff_to_pte(__off) \
 	((pte_t) { ((((__off) & 0x7f) << 1) + (((__off) >> 7) << 12)) \
-		   | _PAGE_INVALID_FILE })
+		   | _PAGE_TYPE_FILE })
 
 #endif /* !__ASSEMBLY__ */
 
-- 
cgit v1.2.3


From d02765d1af743567398eb6d523dea0ba5e5e7e8e Mon Sep 17 00:00:00 2001
From: Gerald Schaefer <geraldsc@de.ibm.com>
Date: Wed, 20 Sep 2006 15:59:42 +0200
Subject: [S390] Make user-copy operations run-time configurable.

Introduces a struct uaccess_ops which allows setting user-copy
operations at run-time.

Signed-off-by: Gerald Schaefer <geraldsc@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/s390_ksyms.c |   6 -
 arch/s390/kernel/setup.c      |   7 +
 arch/s390/lib/Makefile        |   3 +-
 arch/s390/lib/uaccess.S       | 211 --------------------------
 arch/s390/lib/uaccess64.S     | 207 -------------------------
 arch/s390/lib/uaccess_std.c   | 340 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-s390/futex.h      |  87 +----------
 include/asm-s390/uaccess.h    | 171 +++++++--------------
 8 files changed, 411 insertions(+), 621 deletions(-)
 delete mode 100644 arch/s390/lib/uaccess.S
 delete mode 100644 arch/s390/lib/uaccess64.S
 create mode 100644 arch/s390/lib/uaccess_std.c

(limited to 'include')

diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c
index c73a45467fa4..9f19e833a562 100644
--- a/arch/s390/kernel/s390_ksyms.c
+++ b/arch/s390/kernel/s390_ksyms.c
@@ -25,12 +25,6 @@ EXPORT_SYMBOL(_oi_bitmap);
 EXPORT_SYMBOL(_ni_bitmap);
 EXPORT_SYMBOL(_zb_findmap);
 EXPORT_SYMBOL(_sb_findmap);
-EXPORT_SYMBOL(__copy_from_user_asm);
-EXPORT_SYMBOL(__copy_to_user_asm);
-EXPORT_SYMBOL(__copy_in_user_asm);
-EXPORT_SYMBOL(__clear_user_asm);
-EXPORT_SYMBOL(__strncpy_from_user_asm);
-EXPORT_SYMBOL(__strnlen_user_asm);
 EXPORT_SYMBOL(diag10);
 
 /*
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index f2a9165ca4f8..e229af59976c 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -50,6 +50,12 @@
 #include <asm/ptrace.h>
 #include <asm/sections.h>
 
+/*
+ * User copy operations.
+ */
+struct uaccess_ops uaccess;
+EXPORT_SYMBOL_GPL(uaccess);
+
 /*
  * Machine setup..
  */
@@ -641,6 +647,7 @@ setup_arch(char **cmdline_p)
 
 	memory_end = memory_size;
 
+	memcpy(&uaccess, &uaccess_std, sizeof(uaccess));
 	parse_early_param();
 
 #ifndef CONFIG_64BIT
diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile
index e05d087a6eae..96c82424d88b 100644
--- a/arch/s390/lib/Makefile
+++ b/arch/s390/lib/Makefile
@@ -4,6 +4,5 @@
 
 EXTRA_AFLAGS := -traditional
 
-lib-y += delay.o string.o
-lib-y += $(if $(CONFIG_64BIT),uaccess64.o,uaccess.o)
+lib-y += delay.o string.o uaccess_std.o
 lib-$(CONFIG_SMP) += spinlock.o
diff --git a/arch/s390/lib/uaccess.S b/arch/s390/lib/uaccess.S
deleted file mode 100644
index 837275284d9f..000000000000
--- a/arch/s390/lib/uaccess.S
+++ /dev/null
@@ -1,211 +0,0 @@
-/*
- *  arch/s390/lib/uaccess.S
- *    __copy_{from|to}_user functions.
- *
- *  s390
- *    Copyright (C) 2000,2002 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *    Authors(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
- *
- *  These functions have standard call interface
- */
-
-#include <linux/errno.h>
-#include <asm/lowcore.h>
-#include <asm/asm-offsets.h>
-
-        .text
-        .align 4
-        .globl __copy_from_user_asm
-	# %r2 = to, %r3 = n, %r4 = from
-__copy_from_user_asm:
-	slr	%r0,%r0
-0:	mvcp	0(%r3,%r2),0(%r4),%r0
-	jnz	1f
-	slr	%r2,%r2
-	br	%r14
-1:	la	%r2,256(%r2)
-	la	%r4,256(%r4)
-	ahi	%r3,-256
-2:	mvcp	0(%r3,%r2),0(%r4),%r0
-	jnz	1b
-3:	slr	%r2,%r2
-	br	%r14
-4:	lhi	%r0,-4096
-	lr	%r5,%r4
-	slr	%r5,%r0
-	nr	%r5,%r0		# %r5 = (%r4 + 4096) & -4096
-	slr	%r5,%r4		# %r5 = #bytes to next user page boundary
-	clr	%r3,%r5		# copy crosses next page boundary ?
-	jnh	6f		# no, the current page faulted
-	# move with the reduced length which is < 256
-5:	mvcp	0(%r5,%r2),0(%r4),%r0
-	slr	%r3,%r5
-6:	lr	%r2,%r3
-	br	%r14
-        .section __ex_table,"a"
-	.long	0b,4b
-	.long	2b,4b
-	.long	5b,6b
-        .previous
-
-        .align 4
-        .text
-        .globl __copy_to_user_asm
-	# %r2 = from, %r3 = n, %r4 = to
-__copy_to_user_asm:
-	slr	%r0,%r0
-0:	mvcs	0(%r3,%r4),0(%r2),%r0
-	jnz	1f
-	slr	%r2,%r2
-	br	%r14
-1:	la	%r2,256(%r2)
-	la	%r4,256(%r4)
-	ahi	%r3,-256
-2:	mvcs	0(%r3,%r4),0(%r2),%r0
-	jnz	1b
-3:	slr	%r2,%r2
-	br	%r14
-4:	lhi	%r0,-4096
-	lr	%r5,%r4
-	slr	%r5,%r0
-	nr	%r5,%r0		# %r5 = (%r4 + 4096) & -4096
-	slr	%r5,%r4		# %r5 = #bytes to next user page boundary
-	clr	%r3,%r5		# copy crosses next page boundary ?
-	jnh	6f		# no, the current page faulted
-	# move with the reduced length which is < 256
-5:	mvcs	0(%r5,%r4),0(%r2),%r0
-	slr	%r3,%r5
-6:	lr	%r2,%r3
-	br	%r14
-        .section __ex_table,"a"
-	.long	0b,4b
-	.long	2b,4b
-	.long	5b,6b
-        .previous
-
-        .align 4
-        .text
-        .globl __copy_in_user_asm
-	# %r2 = from, %r3 = n, %r4 = to
-__copy_in_user_asm:
-	ahi	%r3,-1
-	jo	6f
-	sacf	256
-	bras	%r1,4f
-0:	ahi	%r3,257
-1:	mvc	0(1,%r4),0(%r2)
-	la	%r2,1(%r2)
-	la	%r4,1(%r4)
-	ahi	%r3,-1
-	jnz	1b
-2:	lr	%r2,%r3
-	br	%r14
-3:	mvc	0(256,%r4),0(%r2)
-	la	%r2,256(%r2)
-	la	%r4,256(%r4)
-4:	ahi	%r3,-256
-	jnm	3b
-5:	ex	%r3,4(%r1)
-	sacf	0
-6:	slr	%r2,%r2
-	br	%r14
-        .section __ex_table,"a"
-	.long	1b,2b
-	.long	3b,0b
-	.long	5b,0b
-        .previous
-
-        .align 4
-        .text
-        .globl __clear_user_asm
-	# %r2 = to, %r3 = n
-__clear_user_asm:
-	bras	%r5,0f
-	.long	empty_zero_page
-0:	l	%r5,0(%r5)
-	slr	%r0,%r0
-1:	mvcs	0(%r3,%r2),0(%r5),%r0
-	jnz	2f
-	slr	%r2,%r2
-	br	%r14
-2:	la	%r2,256(%r2)
-	ahi	%r3,-256
-3:	mvcs	0(%r3,%r2),0(%r5),%r0
-	jnz	2b
-4:	slr	%r2,%r2
-	br	%r14
-5:	lhi	%r0,-4096
-	lr	%r4,%r2
-	slr	%r4,%r0
-	nr	%r4,%r0		# %r4 = (%r2 + 4096) & -4096
-	slr	%r4,%r2		# %r4 = #bytes to next user page boundary
-	clr	%r3,%r4		# clear crosses next page boundary ?
-	jnh	7f		# no, the current page faulted
-	# clear with the reduced length which is < 256
-6:	mvcs	0(%r4,%r2),0(%r5),%r0
-	slr	%r3,%r4
-7:	lr	%r2,%r3
-	br	%r14
-        .section __ex_table,"a"
-	.long	1b,5b
-	.long	3b,5b
-	.long	6b,7b
-        .previous
-
-        .align 4
-        .text
-        .globl __strncpy_from_user_asm
-	# %r2 = count, %r3 = dst, %r4 = src
-__strncpy_from_user_asm:
-	lhi	%r0,0
-	lr	%r1,%r4
-	la	%r4,0(%r4)	# clear high order bit from %r4
-	la	%r2,0(%r2,%r4)	# %r2 points to first byte after string
-	sacf	256
-0:	srst	%r2,%r1
-	jo	0b
-	sacf	0
-	lr	%r1,%r2
-	jh	1f		# \0 found in string ?
-	ahi	%r1,1		# include \0 in copy
-1:	slr	%r1,%r4		# %r1 = copy length (without \0)
-	slr	%r2,%r4		# %r2 = return length (including \0)
-2:	mvcp	0(%r1,%r3),0(%r4),%r0
-	jnz	3f
-	br	%r14
-3:	la	%r3,256(%r3)
-	la	%r4,256(%r4)
-	ahi	%r1,-256
-	mvcp	0(%r1,%r3),0(%r4),%r0
-	jnz	3b
-	br	%r14
-4:	sacf	0
-	lhi	%r2,-EFAULT
-	br	%r14
-	.section __ex_table,"a"
-	.long	0b,4b
-	.previous
-
-        .align 4
-        .text
-        .globl __strnlen_user_asm
-	# %r2 = count, %r3 = src
-__strnlen_user_asm:
-	lhi	%r0,0
-	lr	%r1,%r3
-	la	%r3,0(%r3)	# clear high order bit from %r4
-	la	%r2,0(%r2,%r3)	# %r2 points to first byte after string
-	sacf	256
-0:	srst	%r2,%r1
-	jo	0b
-	sacf	0
-	ahi	%r2,1		# strnlen_user result includes the \0
-				# or return count+1 if \0 not found
-	slr	%r2,%r3
-	br	%r14
-2:	sacf	0
-	slr	%r2,%r2		# return 0 on exception
-	br	%r14
-	.section __ex_table,"a"
-	.long	0b,2b
-	.previous
diff --git a/arch/s390/lib/uaccess64.S b/arch/s390/lib/uaccess64.S
deleted file mode 100644
index 1f755be22f92..000000000000
--- a/arch/s390/lib/uaccess64.S
+++ /dev/null
@@ -1,207 +0,0 @@
-/*
- *  arch/s390x/lib/uaccess.S
- *    __copy_{from|to}_user functions.
- *
- *  s390
- *    Copyright (C) 2000,2002 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *    Authors(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
- *
- *  These functions have standard call interface
- */
-
-#include <linux/errno.h>
-#include <asm/lowcore.h>
-#include <asm/asm-offsets.h>
-
-        .text
-        .align 4
-        .globl __copy_from_user_asm
-	# %r2 = to, %r3 = n, %r4 = from
-__copy_from_user_asm:
-	slgr	%r0,%r0
-0:	mvcp	0(%r3,%r2),0(%r4),%r0
-	jnz	1f
-	slgr	%r2,%r2
-	br	%r14
-1:	la	%r2,256(%r2)
-	la	%r4,256(%r4)
-	aghi	%r3,-256
-2:	mvcp	0(%r3,%r2),0(%r4),%r0
-	jnz	1b
-3:	slgr	%r2,%r2
-	br	%r14
-4:	lghi	%r0,-4096
-	lgr	%r5,%r4
-	slgr	%r5,%r0
-	ngr	%r5,%r0		# %r5 = (%r4 + 4096) & -4096
-	slgr	%r5,%r4		# %r5 = #bytes to next user page boundary
-	clgr	%r3,%r5		# copy crosses next page boundary ?
-	jnh	6f		# no, the current page faulted
-	# move with the reduced length which is < 256
-5:	mvcp	0(%r5,%r2),0(%r4),%r0
-	slgr	%r3,%r5
-6:	lgr	%r2,%r3
-	br	%r14
-        .section __ex_table,"a"
-	.quad	0b,4b
-	.quad	2b,4b
-	.quad	5b,6b
-        .previous
-
-        .align 4
-        .text
-        .globl __copy_to_user_asm
-	# %r2 = from, %r3 = n, %r4 = to
-__copy_to_user_asm:
-	slgr	%r0,%r0
-0:	mvcs	0(%r3,%r4),0(%r2),%r0
-	jnz	1f
-	slgr	%r2,%r2
-	br	%r14
-1:	la	%r2,256(%r2)
-	la	%r4,256(%r4)
-	aghi	%r3,-256
-2:	mvcs	0(%r3,%r4),0(%r2),%r0
-	jnz	1b
-3:	slgr	%r2,%r2
-	br	%r14
-4:	lghi	%r0,-4096
-	lgr	%r5,%r4
-	slgr	%r5,%r0
-	ngr	%r5,%r0		# %r5 = (%r4 + 4096) & -4096
-	slgr	%r5,%r4		# %r5 = #bytes to next user page boundary
-	clgr	%r3,%r5		# copy crosses next page boundary ?
-	jnh	6f		# no, the current page faulted
-	# move with the reduced length which is < 256
-5:	mvcs	0(%r5,%r4),0(%r2),%r0
-	slgr	%r3,%r5
-6:	lgr	%r2,%r3
-	br	%r14
-        .section __ex_table,"a"
-	.quad	0b,4b
-	.quad	2b,4b
-	.quad	5b,6b
-        .previous
-
-        .align 4
-        .text
-        .globl __copy_in_user_asm
-	# %r2 = from, %r3 = n, %r4 = to
-__copy_in_user_asm:
-	aghi	%r3,-1
-	jo	6f
-	sacf	256
-	bras	%r1,4f
-0:	aghi	%r3,257
-1:	mvc	0(1,%r4),0(%r2)
-	la	%r2,1(%r2)
-	la	%r4,1(%r4)
-	aghi	%r3,-1
-	jnz	1b
-2:	lgr	%r2,%r3
-	br	%r14
-3:	mvc	0(256,%r4),0(%r2)
-	la	%r2,256(%r2)
-	la	%r4,256(%r4)
-4:	aghi	%r3,-256
-	jnm	3b
-5:	ex	%r3,4(%r1)
-	sacf	0
-6:	slgr	%r2,%r2
-	br	14
-        .section __ex_table,"a"
-	.quad	1b,2b
-	.quad	3b,0b
-	.quad	5b,0b
-        .previous
-
-        .align 4
-        .text
-        .globl __clear_user_asm
-	# %r2 = to, %r3 = n
-__clear_user_asm:
-	slgr	%r0,%r0
-	larl	%r5,empty_zero_page
-1:	mvcs	0(%r3,%r2),0(%r5),%r0
-	jnz	2f
-	slgr	%r2,%r2
-	br	%r14
-2:	la	%r2,256(%r2)
-	aghi	%r3,-256
-3:	mvcs	0(%r3,%r2),0(%r5),%r0
-	jnz	2b
-4:	slgr	%r2,%r2
-	br	%r14
-5:	lghi	%r0,-4096
-	lgr	%r4,%r2
-	slgr	%r4,%r0
-	ngr	%r4,%r0		# %r4 = (%r2 + 4096) & -4096
-	slgr	%r4,%r2		# %r4 = #bytes to next user page boundary
-	clgr	%r3,%r4		# clear crosses next page boundary ?
-	jnh	7f		# no, the current page faulted
-	# clear with the reduced length which is < 256
-6:	mvcs	0(%r4,%r2),0(%r5),%r0
-	slgr	%r3,%r4
-7:	lgr	%r2,%r3
-	br	%r14
-        .section __ex_table,"a"
-	.quad	1b,5b
-	.quad	3b,5b
-	.quad	6b,7b
-        .previous
-
-        .align 4
-        .text
-        .globl __strncpy_from_user_asm
-	# %r2 = count, %r3 = dst, %r4 = src
-__strncpy_from_user_asm:
-	lghi	%r0,0
-	lgr	%r1,%r4
-	la	%r2,0(%r2,%r4)	# %r2 points to first byte after string
-	sacf	256
-0:	srst	%r2,%r1
-	jo	0b
-	sacf	0
-	lgr	%r1,%r2
-	jh	1f		# \0 found in string ?
-	aghi	%r1,1		# include \0 in copy
-1:	slgr	%r1,%r4		# %r1 = copy length (without \0)
-	slgr	%r2,%r4		# %r2 = return length (including \0)
-2:	mvcp	0(%r1,%r3),0(%r4),%r0
-	jnz	3f
-	br	%r14
-3:	la	%r3,256(%r3)
-	la	%r4,256(%r4)
-	aghi	%r1,-256
-	mvcp	0(%r1,%r3),0(%r4),%r0
-	jnz	3b
-	br	%r14
-4:	sacf	0
-	lghi	%r2,-EFAULT
-	br	%r14
-	.section __ex_table,"a"
-	.quad	0b,4b
-	.previous
-
-        .align 4
-        .text
-        .globl __strnlen_user_asm
-	# %r2 = count, %r3 = src
-__strnlen_user_asm:
-	lghi	%r0,0
-	lgr	%r1,%r3
-	la	%r2,0(%r2,%r3)	# %r2 points to first byte after string
-	sacf	256
-0:	srst	%r2,%r1
-	jo	0b
-	sacf	0
-	aghi	%r2,1		# strnlen_user result includes the \0
-				# or return count+1 if \0 not found
-	slgr	%r2,%r3
-	br	%r14
-2:	sacf	0
-	slgr	%r2,%r2		# return 0 on exception
-	br	%r14
-	.section __ex_table,"a"
-	.quad	0b,2b
-	.previous
diff --git a/arch/s390/lib/uaccess_std.c b/arch/s390/lib/uaccess_std.c
new file mode 100644
index 000000000000..9a4d4a29ea79
--- /dev/null
+++ b/arch/s390/lib/uaccess_std.c
@@ -0,0 +1,340 @@
+/*
+ *  arch/s390/lib/uaccess_std.c
+ *
+ *  Standard user space access functions based on mvcp/mvcs and doing
+ *  interesting things in the secondary space mode.
+ *
+ *    Copyright (C) IBM Corp. 2006
+ *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
+ *		 Gerald Schaefer (gerald.schaefer@de.ibm.com)
+ */
+
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <asm/uaccess.h>
+#include <asm/futex.h>
+
+#ifndef __s390x__
+#define AHI	"ahi"
+#define ALR	"alr"
+#define CLR	"clr"
+#define LHI	"lhi"
+#define SLR	"slr"
+#else
+#define AHI	"aghi"
+#define ALR	"algr"
+#define CLR	"clgr"
+#define LHI	"lghi"
+#define SLR	"slgr"
+#endif
+
+size_t copy_from_user_std(size_t size, const void __user *ptr, void *x)
+{
+	unsigned long tmp1, tmp2;
+
+	tmp1 = -256UL;
+	asm volatile(
+		"0: mvcp  0(%0,%2),0(%1),%3\n"
+		"   jz    5f\n"
+		"1:"ALR"  %0,%3\n"
+		"   la    %1,256(%1)\n"
+		"   la    %2,256(%2)\n"
+		"2: mvcp  0(%0,%2),0(%1),%3\n"
+		"   jnz   1b\n"
+		"   j     5f\n"
+		"3: la    %4,255(%1)\n"	/* %4 = ptr + 255 */
+		"  "LHI"  %3,-4096\n"
+		"   nr    %4,%3\n"	/* %4 = (ptr + 255) & -4096 */
+		"  "SLR"  %4,%1\n"
+		"  "CLR"  %0,%4\n"	/* copy crosses next page boundary? */
+		"   jnh   6f\n"
+		"4: mvcp  0(%4,%2),0(%1),%3\n"
+		"  "SLR"  %0,%4\n"
+		"   j     6f\n"
+		"5:"SLR"  %0,%0\n"
+		"6: \n"
+		EX_TABLE(0b,3b) EX_TABLE(2b,3b) EX_TABLE(4b,6b)
+		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
+		: : "cc", "memory");
+	return size;
+}
+
+size_t copy_from_user_std_small(size_t size, const void __user *ptr, void *x)
+{
+	unsigned long tmp1, tmp2;
+
+	tmp1 = 0UL;
+	asm volatile(
+		"0: mvcp  0(%0,%2),0(%1),%3\n"
+		"  "SLR"  %0,%0\n"
+		"   j     3f\n"
+		"1: la    %4,255(%1)\n" /* %4 = ptr + 255 */
+		"  "LHI"  %3,-4096\n"
+		"   nr    %4,%3\n"	/* %4 = (ptr + 255) & -4096 */
+		"  "SLR"  %4,%1\n"
+		"  "CLR"  %0,%4\n"	/* copy crosses next page boundary? */
+		"   jnh   3f\n"
+		"2: mvcp  0(%4,%2),0(%1),%3\n"
+		"  "SLR"  %0,%4\n"
+		"3:\n"
+		EX_TABLE(0b,1b) EX_TABLE(2b,3b)
+		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
+		: : "cc", "memory");
+	return size;
+}
+
+size_t copy_to_user_std(size_t size, void __user *ptr, const void *x)
+{
+	unsigned long tmp1, tmp2;
+
+	tmp1 = -256UL;
+	asm volatile(
+		"0: mvcs  0(%0,%1),0(%2),%3\n"
+		"   jz    5f\n"
+		"1:"ALR"  %0,%3\n"
+		"   la    %1,256(%1)\n"
+		"   la    %2,256(%2)\n"
+		"2: mvcs  0(%0,%1),0(%2),%3\n"
+		"   jnz   1b\n"
+		"   j     5f\n"
+		"3: la    %4,255(%1)\n" /* %4 = ptr + 255 */
+		"  "LHI"  %3,-4096\n"
+		"   nr    %4,%3\n"	/* %4 = (ptr + 255) & -4096 */
+		"  "SLR"  %4,%1\n"
+		"  "CLR"  %0,%4\n"	/* copy crosses next page boundary? */
+		"   jnh   6f\n"
+		"4: mvcs  0(%4,%1),0(%2),%3\n"
+		"  "SLR"  %0,%4\n"
+		"   j     6f\n"
+		"5:"SLR"  %0,%0\n"
+		"6: \n"
+		EX_TABLE(0b,3b) EX_TABLE(2b,3b) EX_TABLE(4b,6b)
+		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
+		: : "cc", "memory");
+	return size;
+}
+
+size_t copy_to_user_std_small(size_t size, void __user *ptr, const void *x)
+{
+	unsigned long tmp1, tmp2;
+
+	tmp1 = 0UL;
+	asm volatile(
+		"0: mvcs  0(%0,%1),0(%2),%3\n"
+		"  "SLR"  %0,%0\n"
+		"   j     3f\n"
+		"1: la    %4,255(%1)\n" /* ptr + 255 */
+		"  "LHI"  %3,-4096\n"
+		"   nr    %4,%3\n"	/* (ptr + 255) & -4096UL */
+		"  "SLR"  %4,%1\n"
+		"  "CLR"  %0,%4\n"	/* copy crosses next page boundary? */
+		"   jnh   3f\n"
+		"2: mvcs  0(%4,%1),0(%2),%3\n"
+		"  "SLR"  %0,%4\n"
+		"3:\n"
+		EX_TABLE(0b,1b) EX_TABLE(2b,3b)
+		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
+		: : "cc", "memory");
+	return size;
+}
+
+size_t copy_in_user_std(size_t size, void __user *to, const void __user *from)
+{
+	unsigned long tmp1;
+
+	asm volatile(
+		"  "AHI"  %0,-1\n"
+		"   jo    5f\n"
+		"   sacf  256\n"
+		"   bras  %3,3f\n"
+		"0:"AHI"  %0,257\n"
+		"1: mvc   0(1,%1),0(%2)\n"
+		"   la    %1,1(%1)\n"
+		"   la    %2,1(%2)\n"
+		"  "AHI"  %0,-1\n"
+		"   jnz   1b\n"
+		"   j     5f\n"
+		"2: mvc   0(256,%1),0(%2)\n"
+		"   la    %1,256(%1)\n"
+		"   la    %2,256(%2)\n"
+		"3:"AHI"  %0,-256\n"
+		"   jnm   2b\n"
+		"4: ex    %0,1b-0b(%3)\n"
+		"   sacf  0\n"
+		"5: "SLR"  %0,%0\n"
+		"6:\n"
+		EX_TABLE(1b,6b) EX_TABLE(2b,0b) EX_TABLE(4b,0b)
+		: "+a" (size), "+a" (to), "+a" (from), "=a" (tmp1)
+		: : "cc", "memory");
+	return size;
+}
+
+size_t clear_user_std(size_t size, void __user *to)
+{
+	unsigned long tmp1, tmp2;
+
+	asm volatile(
+		"  "AHI"  %0,-1\n"
+		"   jo    5f\n"
+		"   sacf  256\n"
+		"   bras  %3,3f\n"
+		"   xc    0(1,%1),0(%1)\n"
+		"0:"AHI"  %0,257\n"
+		"   la    %2,255(%1)\n" /* %2 = ptr + 255 */
+		"   srl   %2,12\n"
+		"   sll   %2,12\n"	/* %2 = (ptr + 255) & -4096 */
+		"  "SLR"  %2,%1\n"
+		"  "CLR"  %0,%2\n"	/* clear crosses next page boundary? */
+		"   jnh   5f\n"
+		"  "AHI"  %2,-1\n"
+		"1: ex    %2,0(%3)\n"
+		"  "AHI"  %2,1\n"
+		"  "SLR"  %0,%2\n"
+		"   j     5f\n"
+		"2: xc    0(256,%1),0(%1)\n"
+		"   la    %1,256(%1)\n"
+		"3:"AHI"  %0,-256\n"
+		"   jnm   2b\n"
+		"4: ex    %0,0(%3)\n"
+		"   sacf  0\n"
+		"5: "SLR"  %0,%0\n"
+		"6:\n"
+		EX_TABLE(1b,6b) EX_TABLE(2b,0b) EX_TABLE(4b,0b)
+		: "+a" (size), "+a" (to), "=a" (tmp1), "=a" (tmp2)
+		: : "cc", "memory");
+	return size;
+}
+
+size_t strnlen_user_std(size_t size, const char __user *src)
+{
+	register unsigned long reg0 asm("0") = 0UL;
+	unsigned long tmp1, tmp2;
+
+	asm volatile(
+		"   la    %2,0(%1)\n"
+		"   la    %3,0(%0,%1)\n"
+		"  "SLR"  %0,%0\n"
+		"   sacf  256\n"
+		"0: srst  %3,%2\n"
+		"   jo    0b\n"
+		"   la    %0,1(%3)\n"	/* strnlen_user results includes \0 */
+		"  "SLR"  %0,%1\n"
+		"1: sacf  0\n"
+		EX_TABLE(0b,1b)
+		: "+a" (size), "+a" (src), "=a" (tmp1), "=a" (tmp2)
+		: "d" (reg0) : "cc", "memory");
+	return size;
+}
+
+size_t strncpy_from_user_std(size_t size, const char __user *src, char *dst)
+{
+	register unsigned long reg0 asm("0") = 0UL;
+	unsigned long tmp1, tmp2;
+
+	asm volatile(
+		"   la    %3,0(%1)\n"
+		"   la    %4,0(%0,%1)\n"
+		"   sacf  256\n"
+		"0: srst  %4,%3\n"
+		"   jo    0b\n"
+		"   sacf  0\n"
+		"   la    %0,0(%4)\n"
+		"   jh    1f\n"		/* found \0 in string ? */
+		"  "AHI"  %4,1\n"	/* include \0 in copy */
+		"1:"SLR"  %0,%1\n"	/* %0 = return length (without \0) */
+		"  "SLR"  %4,%1\n"	/* %4 = copy length (including \0) */
+		"2: mvcp  0(%4,%2),0(%1),%5\n"
+		"   jz    9f\n"
+		"3:"AHI"  %4,-256\n"
+		"   la    %1,256(%1)\n"
+		"   la    %2,256(%2)\n"
+		"4: mvcp  0(%4,%2),0(%1),%5\n"
+		"   jnz   3b\n"
+		"   j     9f\n"
+		"7: sacf  0\n"
+		"8:"LHI"  %0,%6\n"
+		"9:\n"
+		EX_TABLE(0b,7b) EX_TABLE(2b,8b) EX_TABLE(4b,8b)
+		: "+a" (size), "+a" (src), "+d" (dst), "=a" (tmp1), "=a" (tmp2)
+		: "d" (reg0), "K" (-EFAULT) : "cc", "memory");
+	return size;
+}
+
+#define __futex_atomic_op(insn, ret, oldval, newval, uaddr, oparg)	\
+	asm volatile(							\
+		"   sacf  256\n"					\
+		"0: l     %1,0(%6)\n"					\
+		"1:"insn						\
+		"2: cs    %1,%2,0(%6)\n"				\
+		"3: jl    1b\n"						\
+		"   lhi   %0,0\n"					\
+		"4: sacf  0\n"						\
+		EX_TABLE(0b,4b) EX_TABLE(2b,4b) EX_TABLE(3b,4b)		\
+		: "=d" (ret), "=&d" (oldval), "=&d" (newval),		\
+		  "=m" (*uaddr)						\
+		: "0" (-EFAULT), "d" (oparg), "a" (uaddr),		\
+		  "m" (*uaddr) : "cc");
+
+int futex_atomic_op(int op, int __user *uaddr, int oparg, int *old)
+{
+	int oldval = 0, newval, ret;
+
+	inc_preempt_count();
+
+	switch (op) {
+	case FUTEX_OP_SET:
+		__futex_atomic_op("lr %2,%5\n",
+				  ret, oldval, newval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ADD:
+		__futex_atomic_op("lr %2,%1\nar %2,%5\n",
+				  ret, oldval, newval, uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_atomic_op("lr %2,%1\nor %2,%5\n",
+				  ret, oldval, newval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_atomic_op("lr %2,%1\nnr %2,%5\n",
+				  ret, oldval, newval, uaddr, oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_atomic_op("lr %2,%1\nxr %2,%5\n",
+				  ret, oldval, newval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
+	}
+	dec_preempt_count();
+	*old = oldval;
+	return ret;
+}
+
+int futex_atomic_cmpxchg(int __user *uaddr, int oldval, int newval)
+{
+	int ret;
+
+	asm volatile(
+		"   sacf 256\n"
+		"   cs   %1,%4,0(%5)\n"
+		"0: lr   %0,%1\n"
+		"1: sacf 0\n"
+		EX_TABLE(0b,1b)
+		: "=d" (ret), "+d" (oldval), "=m" (*uaddr)
+		: "0" (-EFAULT), "d" (newval), "a" (uaddr), "m" (*uaddr)
+		: "cc", "memory" );
+	return ret;
+}
+
+struct uaccess_ops uaccess_std = {
+	.copy_from_user = copy_from_user_std,
+	.copy_from_user_small = copy_from_user_std_small,
+	.copy_to_user = copy_to_user_std,
+	.copy_to_user_small = copy_to_user_std_small,
+	.copy_in_user = copy_in_user_std,
+	.clear_user = clear_user_std,
+	.strnlen_user = strnlen_user_std,
+	.strncpy_from_user = strncpy_from_user_std,
+	.futex_atomic_op = futex_atomic_op,
+	.futex_atomic_cmpxchg = futex_atomic_cmpxchg,
+};
diff --git a/include/asm-s390/futex.h b/include/asm-s390/futex.h
index ffedf14f89f6..5e261e1de671 100644
--- a/include/asm-s390/futex.h
+++ b/include/asm-s390/futex.h
@@ -7,75 +7,21 @@
 #include <asm/errno.h>
 #include <asm/uaccess.h>
 
-#ifndef __s390x__
-#define __futex_atomic_fixup \
-		     ".section __ex_table,\"a\"\n"			\
-		     "   .align 4\n"					\
-		     "   .long  0b,4b,2b,4b,3b,4b\n"			\
-		     ".previous"
-#else /* __s390x__ */
-#define __futex_atomic_fixup \
-		     ".section __ex_table,\"a\"\n"			\
-		     "   .align 8\n"					\
-		     "   .quad  0b,4b,2b,4b,3b,4b\n"			\
-		     ".previous"
-#endif /* __s390x__ */
-
-#define __futex_atomic_op(insn, ret, oldval, newval, uaddr, oparg)	\
-	asm volatile("   sacf 256\n"					\
-		     "0: l   %1,0(%6)\n"				\
-		     "1: " insn						\
-		     "2: cs  %1,%2,0(%6)\n"				\
-		     "3: jl  1b\n"					\
-		     "   lhi %0,0\n"					\
-		     "4: sacf 0\n"					\
-		     __futex_atomic_fixup				\
-		     : "=d" (ret), "=&d" (oldval), "=&d" (newval),	\
-		       "=m" (*uaddr)					\
-		     : "0" (-EFAULT), "d" (oparg), "a" (uaddr),		\
-		       "m" (*uaddr) : "cc" );
-
 static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
 {
 	int op = (encoded_op >> 28) & 7;
 	int cmp = (encoded_op >> 24) & 15;
 	int oparg = (encoded_op << 8) >> 20;
 	int cmparg = (encoded_op << 20) >> 20;
-	int oldval = 0, newval, ret;
+	int oldval, ret;
+
 	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
 		oparg = 1 << oparg;
 
 	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
 		return -EFAULT;
 
-	inc_preempt_count();
-
-	switch (op) {
-	case FUTEX_OP_SET:
-		__futex_atomic_op("lr %2,%5\n",
-				  ret, oldval, newval, uaddr, oparg);
-		break;
-	case FUTEX_OP_ADD:
-		__futex_atomic_op("lr %2,%1\nar %2,%5\n",
-				  ret, oldval, newval, uaddr, oparg);
-		break;
-	case FUTEX_OP_OR:
-		__futex_atomic_op("lr %2,%1\nor %2,%5\n",
-				  ret, oldval, newval, uaddr, oparg);
-		break;
-	case FUTEX_OP_ANDN:
-		__futex_atomic_op("lr %2,%1\nnr %2,%5\n",
-				  ret, oldval, newval, uaddr, oparg);
-		break;
-	case FUTEX_OP_XOR:
-		__futex_atomic_op("lr %2,%1\nxr %2,%5\n",
-				  ret, oldval, newval, uaddr, oparg);
-		break;
-	default:
-		ret = -ENOSYS;
-	}
-
-	dec_preempt_count();
+	ret = uaccess.futex_atomic_op(op, uaddr, oparg, &oldval);
 
 	if (!ret) {
 		switch (cmp) {
@@ -91,32 +37,13 @@ static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
 	return ret;
 }
 
-static inline int
-futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr,
+						int oldval, int newval)
 {
-	int ret;
-
 	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
 		return -EFAULT;
-	asm volatile("   sacf 256\n"
-		     "   cs   %1,%4,0(%5)\n"
-		     "0: lr   %0,%1\n"
-		     "1: sacf 0\n"
-#ifndef __s390x__
-		     ".section __ex_table,\"a\"\n"
-		     "   .align 4\n"
-		     "   .long  0b,1b\n"
-		     ".previous"
-#else /* __s390x__ */
-		     ".section __ex_table,\"a\"\n"
-		     "   .align 8\n"
-		     "   .quad  0b,1b\n"
-		     ".previous"
-#endif /* __s390x__ */
-		     : "=d" (ret), "+d" (oldval), "=m" (*uaddr)
-		     : "0" (-EFAULT), "d" (newval), "a" (uaddr), "m" (*uaddr)
-		     : "cc", "memory" );
-	return oldval;
+
+	return uaccess.futex_atomic_cmpxchg(uaddr, oldval, newval);
 }
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-s390/uaccess.h b/include/asm-s390/uaccess.h
index 0b7c0ca4c3d7..39a2716ae188 100644
--- a/include/asm-s390/uaccess.h
+++ b/include/asm-s390/uaccess.h
@@ -47,7 +47,7 @@
 		S390_lowcore.user_asce : S390_lowcore.kernel_asce;	\
 	asm volatile ("lctlg 7,7,%0" : : "m" (__pto) );			\
 })
-#else
+#else /* __s390x__ */
 #define set_fs(x) \
 ({									\
 	unsigned long __pto;						\
@@ -56,7 +56,7 @@
 		S390_lowcore.user_asce : S390_lowcore.kernel_asce;	\
 	asm volatile ("lctl  7,7,%0" : : "m" (__pto) );			\
 })
-#endif
+#endif /* __s390x__ */
 
 #define segment_eq(a,b) ((a).ar4 == (b).ar4)
 
@@ -85,76 +85,50 @@ struct exception_table_entry
         unsigned long insn, fixup;
 };
 
-#ifndef __s390x__
-#define __uaccess_fixup \
-	".section .fixup,\"ax\"\n"	\
-	"2: lhi    %0,%4\n"		\
-	"   bras   1,3f\n"		\
-	"   .long  1b\n"		\
-	"3: l      1,0(1)\n"		\
-	"   br     1\n"			\
-	".previous\n"			\
-	".section __ex_table,\"a\"\n"	\
-	"   .align 4\n"			\
-	"   .long  0b,2b\n"		\
-	".previous"
-#define __uaccess_clobber "cc", "1"
-#else /* __s390x__ */
-#define __uaccess_fixup \
-	".section .fixup,\"ax\"\n"	\
-	"2: lghi   %0,%4\n"		\
-	"   jg     1b\n"		\
-	".previous\n"			\
-	".section __ex_table,\"a\"\n"	\
-	"   .align 8\n"			\
-	"   .quad  0b,2b\n"		\
-	".previous"
-#define __uaccess_clobber "cc"
-#endif /* __s390x__ */
+struct uaccess_ops {
+	size_t (*copy_from_user)(size_t, const void __user *, void *);
+	size_t (*copy_from_user_small)(size_t, const void __user *, void *);
+	size_t (*copy_to_user)(size_t, void __user *, const void *);
+	size_t (*copy_to_user_small)(size_t, void __user *, const void *);
+	size_t (*copy_in_user)(size_t, void __user *, const void __user *);
+	size_t (*clear_user)(size_t, void __user *);
+	size_t (*strnlen_user)(size_t, const char __user *);
+	size_t (*strncpy_from_user)(size_t, const char __user *, char *);
+	int (*futex_atomic_op)(int op, int __user *, int oparg, int *old);
+	int (*futex_atomic_cmpxchg)(int __user *, int old, int new);
+};
+
+extern struct uaccess_ops uaccess;
+extern struct uaccess_ops uaccess_std;
+
+static inline int __put_user_fn(size_t size, void __user *ptr, void *x)
+{
+	size = uaccess.copy_to_user_small(size, ptr, x);
+	return size ? -EFAULT : size;
+}
+
+static inline int __get_user_fn(size_t size, const void __user *ptr, void *x)
+{
+	size = uaccess.copy_from_user_small(size, ptr, x);
+	return size ? -EFAULT : size;
+}
 
 /*
  * These are the main single-value transfer routines.  They automatically
  * use the right size if we just have the right pointer type.
  */
-#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
-#define __put_user_asm(x, ptr, err) \
-({								\
-	err = 0;						\
-	asm volatile(						\
-		"0: mvcs  0(%1,%2),%3,%0\n"			\
-		"1:\n"						\
-		__uaccess_fixup					\
-		: "+&d" (err)					\
-		: "d" (sizeof(*(ptr))), "a" (ptr), "Q" (x),	\
-		  "K" (-EFAULT)					\
-		: __uaccess_clobber );				\
-})
-#else
-#define __put_user_asm(x, ptr, err) \
-({								\
-	err = 0;						\
-	asm volatile(						\
-		"0: mvcs  0(%1,%2),0(%3),%0\n"			\
-		"1:\n"						\
-		__uaccess_fixup					\
-		: "+&d" (err)					\
-		: "d" (sizeof(*(ptr))), "a" (ptr), "a" (&(x)),	\
-		  "K" (-EFAULT), "m" (x)			\
-		: __uaccess_clobber );				\
-})
-#endif
-
 #define __put_user(x, ptr) \
 ({								\
 	__typeof__(*(ptr)) __x = (x);				\
-	int __pu_err;						\
+	int __pu_err = -EFAULT;					\
         __chk_user_ptr(ptr);                                    \
 	switch (sizeof (*(ptr))) {				\
 	case 1:							\
 	case 2:							\
 	case 4:							\
 	case 8:							\
-		__put_user_asm(__x, ptr, __pu_err);		\
+		__pu_err = __put_user_fn(sizeof (*(ptr)),	\
+					 ptr, &__x);		\
 		break;						\
 	default:						\
 		__put_user_bad();				\
@@ -172,60 +146,36 @@ struct exception_table_entry
 
 extern int __put_user_bad(void) __attribute__((noreturn));
 
-#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
-#define __get_user_asm(x, ptr, err) \
-({								\
-	err = 0;						\
-	asm volatile (						\
-		"0: mvcp  %O1(%2,%R1),0(%3),%0\n"		\
-		"1:\n"						\
-		__uaccess_fixup					\
-		: "+&d" (err), "=Q" (x)				\
-		: "d" (sizeof(*(ptr))), "a" (ptr),		\
-		  "K" (-EFAULT)					\
-		: __uaccess_clobber );				\
-})
-#else
-#define __get_user_asm(x, ptr, err) \
-({								\
-	err = 0;						\
-	asm volatile (						\
-		"0: mvcp  0(%2,%5),0(%3),%0\n"			\
-		"1:\n"						\
-		__uaccess_fixup					\
-		: "+&d" (err), "=m" (x)				\
-		: "d" (sizeof(*(ptr))), "a" (ptr),		\
-		  "K" (-EFAULT), "a" (&(x))			\
-		: __uaccess_clobber );				\
-})
-#endif
-
 #define __get_user(x, ptr)					\
 ({								\
-	int __gu_err;						\
-        __chk_user_ptr(ptr);                                    \
+	int __gu_err = -EFAULT;					\
+	__chk_user_ptr(ptr);					\
 	switch (sizeof(*(ptr))) {				\
 	case 1: {						\
 		unsigned char __x;				\
-		__get_user_asm(__x, ptr, __gu_err);		\
+		__gu_err = __get_user_fn(sizeof (*(ptr)),	\
+					 ptr, &__x);		\
 		(x) = *(__force __typeof__(*(ptr)) *) &__x;	\
 		break;						\
 	};							\
 	case 2: {						\
 		unsigned short __x;				\
-		__get_user_asm(__x, ptr, __gu_err);		\
+		__gu_err = __get_user_fn(sizeof (*(ptr)),	\
+					 ptr, &__x);		\
 		(x) = *(__force __typeof__(*(ptr)) *) &__x;	\
 		break;						\
 	};							\
 	case 4: {						\
 		unsigned int __x;				\
-		__get_user_asm(__x, ptr, __gu_err);		\
+		__gu_err = __get_user_fn(sizeof (*(ptr)),	\
+					 ptr, &__x);		\
 		(x) = *(__force __typeof__(*(ptr)) *) &__x;	\
 		break;						\
 	};							\
 	case 8: {						\
 		unsigned long long __x;				\
-		__get_user_asm(__x, ptr, __gu_err);		\
+		__gu_err = __get_user_fn(sizeof (*(ptr)),	\
+					 ptr, &__x);		\
 		(x) = *(__force __typeof__(*(ptr)) *) &__x;	\
 		break;						\
 	};							\
@@ -247,8 +197,6 @@ extern int __get_user_bad(void) __attribute__((noreturn));
 #define __put_user_unaligned __put_user
 #define __get_user_unaligned __get_user
 
-extern long __copy_to_user_asm(const void *from, long n, void __user *to);
-
 /**
  * __copy_to_user: - Copy a block of data into user space, with less checking.
  * @to:   Destination address, in user space.
@@ -266,7 +214,10 @@ extern long __copy_to_user_asm(const void *from, long n, void __user *to);
 static inline unsigned long
 __copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	return __copy_to_user_asm(from, n, to);
+	if (__builtin_constant_p(n) && (n <= 256))
+		return uaccess.copy_to_user_small(n, to, from);
+	else
+		return uaccess.copy_to_user(n, to, from);
 }
 
 #define __copy_to_user_inatomic __copy_to_user
@@ -294,8 +245,6 @@ copy_to_user(void __user *to, const void *from, unsigned long n)
 	return n;
 }
 
-extern long __copy_from_user_asm(void *to, long n, const void __user *from);
-
 /**
  * __copy_from_user: - Copy a block of data from user space, with less checking.
  * @to:   Destination address, in kernel space.
@@ -316,7 +265,10 @@ extern long __copy_from_user_asm(void *to, long n, const void __user *from);
 static inline unsigned long
 __copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	return __copy_from_user_asm(to, n, from);
+	if (__builtin_constant_p(n) && (n <= 256))
+		return uaccess.copy_from_user_small(n, from, to);
+	else
+		return uaccess.copy_from_user(n, from, to);
 }
 
 /**
@@ -346,13 +298,10 @@ copy_from_user(void *to, const void __user *from, unsigned long n)
 	return n;
 }
 
-extern unsigned long __copy_in_user_asm(const void __user *from, long n,
-							void __user *to);
-
 static inline unsigned long
 __copy_in_user(void __user *to, const void __user *from, unsigned long n)
 {
-	return __copy_in_user_asm(from, n, to);
+	return uaccess.copy_in_user(n, to, from);
 }
 
 static inline unsigned long
@@ -360,34 +309,28 @@ copy_in_user(void __user *to, const void __user *from, unsigned long n)
 {
 	might_sleep();
 	if (__access_ok(from,n) && __access_ok(to,n))
-		n = __copy_in_user_asm(from, n, to);
+		n = __copy_in_user(to, from, n);
 	return n;
 }
 
 /*
  * Copy a null terminated string from userspace.
  */
-extern long __strncpy_from_user_asm(long count, char *dst,
-					const char __user *src);
-
 static inline long
 strncpy_from_user(char *dst, const char __user *src, long count)
 {
         long res = -EFAULT;
         might_sleep();
         if (access_ok(VERIFY_READ, src, 1))
-                res = __strncpy_from_user_asm(count, dst, src);
+		res = uaccess.strncpy_from_user(count, src, dst);
         return res;
 }
 
-
-extern long __strnlen_user_asm(long count, const char __user *src);
-
 static inline unsigned long
 strnlen_user(const char __user * src, unsigned long n)
 {
 	might_sleep();
-	return __strnlen_user_asm(n, src);
+	return uaccess.strnlen_user(n, src);
 }
 
 /**
@@ -410,12 +353,10 @@ strnlen_user(const char __user * src, unsigned long n)
  * Zero Userspace
  */
 
-extern long __clear_user_asm(void __user *to, long n);
-
 static inline unsigned long
 __clear_user(void __user *to, unsigned long n)
 {
-	return __clear_user_asm(to, n);
+	return uaccess.clear_user(n, to);
 }
 
 static inline unsigned long
@@ -423,7 +364,7 @@ clear_user(void __user *to, unsigned long n)
 {
 	might_sleep();
 	if (access_ok(VERIFY_WRITE, to, n))
-		n = __clear_user_asm(to, n);
+		n = uaccess.clear_user(n, to);
 	return n;
 }
 
-- 
cgit v1.2.3


From 6c2a9e6df60478e712f3c3d98b5047778a82a3d7 Mon Sep 17 00:00:00 2001
From: Gerald Schaefer <geraldsc@de.ibm.com>
Date: Wed, 20 Sep 2006 15:59:44 +0200
Subject: [S390] Use alternative user-copy operations for new hardware.

This introduces new user-copy operations which are optimized for
copying more than 256 Bytes on new hardware.

Signed-off-by: Gerald Schaefer <geraldsc@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/head64.S     |  13 ++++
 arch/s390/kernel/setup.c      |   6 +-
 arch/s390/lib/Makefile        |   1 +
 arch/s390/lib/uaccess_mvcos.c | 156 ++++++++++++++++++++++++++++++++++++++++++
 include/asm-s390/setup.h      |   2 +
 include/asm-s390/uaccess.h    |   1 +
 6 files changed, 178 insertions(+), 1 deletion(-)
 create mode 100644 arch/s390/lib/uaccess_mvcos.c

(limited to 'include')

diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 1ebaa338aa7e..a8bdd96494c7 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -250,6 +250,19 @@ startup_continue:
 	oi	7(%r12),0x80		# set IDTE flag
 0:
 
+#
+# find out if we have the MVCOS instruction
+#
+	la	%r1,0f-.LPG1(%r13)	# set program check address
+	stg	%r1,__LC_PGM_NEW_PSW+8
+	.short	0xc800			# mvcos 0(%r0),0(%r0),%r0
+	.short	0x0000
+	.short	0x0000
+0:	tm	0x8f,0x13		# special-operation exception?
+	bno	1f-.LPG1(%r13)		# if yes, MVCOS is present
+	oi	6(%r12),2		# set MVCOS flag
+1:
+
         lpswe .Lentry-.LPG1(13)         # jump to _stext in primary-space,
                                         # virtual and never return ...
         .align 16
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index e229af59976c..e3d9325f6022 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -647,7 +647,11 @@ setup_arch(char **cmdline_p)
 
 	memory_end = memory_size;
 
-	memcpy(&uaccess, &uaccess_std, sizeof(uaccess));
+	if (MACHINE_HAS_MVCOS)
+		memcpy(&uaccess, &uaccess_mvcos, sizeof(uaccess));
+	else
+		memcpy(&uaccess, &uaccess_std, sizeof(uaccess));
+
 	parse_early_param();
 
 #ifndef CONFIG_64BIT
diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile
index 96c82424d88b..c42ffedfdb49 100644
--- a/arch/s390/lib/Makefile
+++ b/arch/s390/lib/Makefile
@@ -5,4 +5,5 @@
 EXTRA_AFLAGS := -traditional
 
 lib-y += delay.o string.o uaccess_std.o
+lib-$(CONFIG_64BIT) += uaccess_mvcos.o
 lib-$(CONFIG_SMP) += spinlock.o
diff --git a/arch/s390/lib/uaccess_mvcos.c b/arch/s390/lib/uaccess_mvcos.c
new file mode 100644
index 000000000000..86c96d6c191a
--- /dev/null
+++ b/arch/s390/lib/uaccess_mvcos.c
@@ -0,0 +1,156 @@
+/*
+ *  arch/s390/lib/uaccess_mvcos.c
+ *
+ *  Optimized user space space access functions based on mvcos.
+ *
+ *    Copyright (C) IBM Corp. 2006
+ *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
+ *		 Gerald Schaefer (gerald.schaefer@de.ibm.com)
+ */
+
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <asm/uaccess.h>
+#include <asm/futex.h>
+
+#ifndef __s390x__
+#define AHI	"ahi"
+#define ALR	"alr"
+#define CLR	"clr"
+#define LHI	"lhi"
+#define SLR	"slr"
+#else
+#define AHI	"aghi"
+#define ALR	"algr"
+#define CLR	"clgr"
+#define LHI	"lghi"
+#define SLR	"slgr"
+#endif
+
+size_t copy_from_user_mvcos(size_t size, const void __user *ptr, void *x)
+{
+	register unsigned long reg0 asm("0") = 0x81UL;
+	unsigned long tmp1, tmp2;
+
+	tmp1 = -4096UL;
+	asm volatile(
+		"0: .insn ss,0xc80000000000,0(%0,%2),0(%1),0\n"
+		"   jz    4f\n"
+		"1:"ALR"  %0,%3\n"
+		"  "SLR"  %1,%3\n"
+		"  "SLR"  %2,%3\n"
+		"   j     0b\n"
+		"2: la    %4,4095(%1)\n"/* %4 = ptr + 4095 */
+		"   nr    %4,%3\n"	/* %4 = (ptr + 4095) & -4096 */
+		"  "SLR"  %4,%1\n"
+		"  "CLR"  %0,%4\n"	/* copy crosses next page boundary? */
+		"   jnh   5f\n"
+		"3: .insn ss,0xc80000000000,0(%4,%2),0(%1),0\n"
+		"  "SLR"  %0,%4\n"
+		"   j     5f\n"
+		"4:"SLR"  %0,%0\n"
+		"5: \n"
+		EX_TABLE(0b,2b) EX_TABLE(3b,5b)
+		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
+		: "d" (reg0) : "cc", "memory");
+	return size;
+}
+
+size_t copy_to_user_mvcos(size_t size, void __user *ptr, const void *x)
+{
+	register unsigned long reg0 asm("0") = 0x810000UL;
+	unsigned long tmp1, tmp2;
+
+	tmp1 = -4096UL;
+	asm volatile(
+		"0: .insn ss,0xc80000000000,0(%0,%1),0(%2),0\n"
+		"   jz    4f\n"
+		"1:"ALR"  %0,%3\n"
+		"  "SLR"  %1,%3\n"
+		"  "SLR"  %2,%3\n"
+		"   j     0b\n"
+		"2: la    %4,4095(%1)\n"/* %4 = ptr + 4095 */
+		"   nr    %4,%3\n"	/* %4 = (ptr + 4095) & -4096 */
+		"  "SLR"  %4,%1\n"
+		"  "CLR"  %0,%4\n"	/* copy crosses next page boundary? */
+		"   jnh   5f\n"
+		"3: .insn ss,0xc80000000000,0(%4,%1),0(%2),0\n"
+		"  "SLR"  %0,%4\n"
+		"   j     5f\n"
+		"4:"SLR"  %0,%0\n"
+		"5: \n"
+		EX_TABLE(0b,2b) EX_TABLE(3b,5b)
+		: "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2)
+		: "d" (reg0) : "cc", "memory");
+	return size;
+}
+
+size_t copy_in_user_mvcos(size_t size, void __user *to, const void __user *from)
+{
+	register unsigned long reg0 asm("0") = 0x810081UL;
+	unsigned long tmp1, tmp2;
+
+	tmp1 = -4096UL;
+	/* FIXME: copy with reduced length. */
+	asm volatile(
+		"0: .insn ss,0xc80000000000,0(%0,%1),0(%2),0\n"
+		"   jz    2f\n"
+		"1:"ALR"  %0,%3\n"
+		"  "SLR"  %1,%3\n"
+		"  "SLR"  %2,%3\n"
+		"   j     0b\n"
+		"2:"SLR"  %0,%0\n"
+		"3: \n"
+		EX_TABLE(0b,3b)
+		: "+a" (size), "+a" (to), "+a" (from), "+a" (tmp1), "=a" (tmp2)
+		: "d" (reg0) : "cc", "memory");
+	return size;
+}
+
+size_t clear_user_mvcos(size_t size, void __user *to)
+{
+	register unsigned long reg0 asm("0") = 0x810000UL;
+	unsigned long tmp1, tmp2;
+
+	tmp1 = -4096UL;
+	asm volatile(
+		"0: .insn ss,0xc80000000000,0(%0,%1),0(%4),0\n"
+		"   jz    4f\n"
+		"1:"ALR"  %0,%2\n"
+		"  "SLR"  %1,%2\n"
+		"   j     0b\n"
+		"2: la    %3,4095(%1)\n"/* %4 = to + 4095 */
+		"   nr    %3,%2\n"	/* %4 = (to + 4095) & -4096 */
+		"  "SLR"  %3,%1\n"
+		"  "CLR"  %0,%3\n"	/* copy crosses next page boundary? */
+		"   jnh   5f\n"
+		"3: .insn ss,0xc80000000000,0(%3,%1),0(%4),0\n"
+		"  "SLR"  %0,%3\n"
+		"   j     5f\n"
+		"4:"SLR"  %0,%0\n"
+		"5: \n"
+		EX_TABLE(0b,2b) EX_TABLE(3b,5b)
+		: "+a" (size), "+a" (to), "+a" (tmp1), "=a" (tmp2)
+		: "a" (empty_zero_page), "d" (reg0) : "cc", "memory");
+	return size;
+}
+
+extern size_t copy_from_user_std_small(size_t, const void __user *, void *);
+extern size_t copy_to_user_std_small(size_t, void __user *, const void *);
+extern size_t strnlen_user_std(size_t, const char __user *);
+extern size_t strncpy_from_user_std(size_t, const char __user *, char *);
+extern int futex_atomic_op(int, int __user *, int, int *);
+extern int futex_atomic_cmpxchg(int __user *, int, int);
+
+struct uaccess_ops uaccess_mvcos = {
+	.copy_from_user = copy_from_user_mvcos,
+	.copy_from_user_small = copy_from_user_std_small,
+	.copy_to_user = copy_to_user_mvcos,
+	.copy_to_user_small = copy_to_user_std_small,
+	.copy_in_user = copy_in_user_mvcos,
+	.clear_user = clear_user_mvcos,
+	.strnlen_user = strnlen_user_std,
+	.strncpy_from_user = strncpy_from_user_std,
+	.futex_atomic_op = futex_atomic_op,
+	.futex_atomic_cmpxchg = futex_atomic_cmpxchg,
+};
diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h
index 00c03e46689b..f1959732b6fd 100644
--- a/include/asm-s390/setup.h
+++ b/include/asm-s390/setup.h
@@ -44,10 +44,12 @@ extern unsigned long machine_flags;
 #define MACHINE_HAS_IEEE	(machine_flags & 2)
 #define MACHINE_HAS_CSP		(machine_flags & 8)
 #define MACHINE_HAS_DIAG44	(1)
+#define MACHINE_HAS_MVCOS	(0)
 #else /* __s390x__ */
 #define MACHINE_HAS_IEEE	(1)
 #define MACHINE_HAS_CSP		(1)
 #define MACHINE_HAS_DIAG44	(machine_flags & 32)
+#define MACHINE_HAS_MVCOS	(machine_flags & 512)
 #endif /* __s390x__ */
 
 
diff --git a/include/asm-s390/uaccess.h b/include/asm-s390/uaccess.h
index 39a2716ae188..e2047b0c9092 100644
--- a/include/asm-s390/uaccess.h
+++ b/include/asm-s390/uaccess.h
@@ -100,6 +100,7 @@ struct uaccess_ops {
 
 extern struct uaccess_ops uaccess;
 extern struct uaccess_ops uaccess_std;
+extern struct uaccess_ops uaccess_mvcos;
 
 static inline int __put_user_fn(size_t size, void __user *ptr, void *x)
 {
-- 
cgit v1.2.3


From 799111020c66c41aef621a3b53ad112543754124 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 21 Aug 2006 21:03:52 +1000
Subject: [CRYPTO] api: Fixed crypto_tfm context alignment

Previously the __aligned__ attribute was added to the crypto_tfm context
member to ensure it is alinged correctly on architectures such as arm.
Unfortunately kmalloc does not use the same minimum alignment rules as
gcc so this is useless.

This patch changes it to use kmalloc's minimum alignment.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/crypto.h | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 7f946241b879..cb1e6631b132 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -21,8 +21,9 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/list.h>
+#include <linux/slab.h>
 #include <linux/string.h>
-#include <asm/page.h>
+#include <linux/uaccess.h>
 
 /*
  * Algorithm masks and types.
@@ -61,6 +62,26 @@
 #define CRYPTO_DIR_ENCRYPT		1
 #define CRYPTO_DIR_DECRYPT		0
 
+/*
+ * The macro CRYPTO_MINALIGN_ATTR (along with the void * type in the actual
+ * declaration) is used to ensure that the crypto_tfm context structure is
+ * aligned correctly for the given architecture so that there are no alignment
+ * faults for C data types.  In particular, this is required on platforms such
+ * as arm where pointers are 32-bit aligned but there are data types such as
+ * u64 which require 64-bit alignment.
+ */
+#if defined(ARCH_KMALLOC_MINALIGN)
+#define CRYPTO_MINALIGN ARCH_KMALLOC_MINALIGN
+#elif defined(ARCH_SLAB_MINALIGN)
+#define CRYPTO_MINALIGN ARCH_SLAB_MINALIGN
+#endif
+
+#ifdef CRYPTO_MINALIGN
+#define CRYPTO_MINALIGN_ATTR __attribute__ ((__aligned__(CRYPTO_MINALIGN)))
+#else
+#define CRYPTO_MINALIGN_ATTR
+#endif
+
 struct scatterlist;
 struct crypto_tfm;
 
@@ -231,7 +252,7 @@ struct crypto_tfm {
 	
 	struct crypto_alg *__crt_alg;
 
-	char __crt_ctx[] __attribute__ ((__aligned__));
+	void *__crt_ctx[] CRYPTO_MINALIGN_ATTR;
 };
 
 /* 
-- 
cgit v1.2.3


From 2729bb427f686e47970406d6bde6b11892885f29 Mon Sep 17 00:00:00 2001
From: Joachim Fritschi <jfritschi@freenet.de>
Date: Tue, 20 Jun 2006 20:37:23 +1000
Subject: [CRYPTO] twofish: Split out common c code

This patch splits up the twofish crypto routine into a common part ( key
setup  ) which will be uses by all twofish crypto modules ( generic-c , i586
assembler and x86_64 assembler ) and generic-c part. It also creates a new
header file which will be used by all 3 modules.

This eliminates all code duplication.

Correctness was verified with the tcrypt module and automated test scripts.

Signed-off-by: Joachim Fritschi <jfritschi@freenet.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Kconfig           |   8 +
 crypto/Makefile          |   1 +
 crypto/twofish.c         | 698 +-------------------------------------------
 crypto/twofish_common.c  | 744 +++++++++++++++++++++++++++++++++++++++++++++++
 include/crypto/twofish.h |  23 ++
 5 files changed, 777 insertions(+), 697 deletions(-)
 create mode 100644 crypto/twofish_common.c
 create mode 100644 include/crypto/twofish.h

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index ba133d557045..5472f693e6ec 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -131,6 +131,7 @@ config CRYPTO_BLOWFISH
 config CRYPTO_TWOFISH
 	tristate "Twofish cipher algorithm"
 	depends on CRYPTO
+	select CRYPTO_TWOFISH_COMMON
 	help
 	  Twofish cipher algorithm.
 	  
@@ -142,6 +143,13 @@ config CRYPTO_TWOFISH
 	  See also:
 	  <http://www.schneier.com/twofish.html>
 
+config CRYPTO_TWOFISH_COMMON
+	tristate
+	depends on CRYPTO
+	help
+	  Common parts of the Twofish cipher algorithm shared by the
+	  generic c and the assembler implementations.
+
 config CRYPTO_SERPENT
 	tristate "Serpent cipher algorithm"
 	depends on CRYPTO
diff --git a/crypto/Makefile b/crypto/Makefile
index d287b9e60c47..fe934f1001c6 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_CRYPTO_TGR192) += tgr192.o
 obj-$(CONFIG_CRYPTO_DES) += des.o
 obj-$(CONFIG_CRYPTO_BLOWFISH) += blowfish.o
 obj-$(CONFIG_CRYPTO_TWOFISH) += twofish.o
+obj-$(CONFIG_CRYPTO_TWOFISH_COMMON) += twofish_common.o
 obj-$(CONFIG_CRYPTO_SERPENT) += serpent.o
 obj-$(CONFIG_CRYPTO_AES) += aes.o
 obj-$(CONFIG_CRYPTO_CAST5) += cast5.o
diff --git a/crypto/twofish.c b/crypto/twofish.c
index ec2488242e2d..e3b3a0a6cb4d 100644
--- a/crypto/twofish.c
+++ b/crypto/twofish.c
@@ -39,6 +39,7 @@
  */
 
 #include <asm/byteorder.h>
+#include <crypto/twofish.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/types.h>
@@ -46,534 +47,6 @@
 #include <linux/crypto.h>
 #include <linux/bitops.h>
 
-
-/* The large precomputed tables for the Twofish cipher (twofish.c)
- * Taken from the same source as twofish.c
- * Marc Mutz <Marc@Mutz.com>
- */
-
-/* These two tables are the q0 and q1 permutations, exactly as described in
- * the Twofish paper. */
-
-static const u8 q0[256] = {
-   0xA9, 0x67, 0xB3, 0xE8, 0x04, 0xFD, 0xA3, 0x76, 0x9A, 0x92, 0x80, 0x78,
-   0xE4, 0xDD, 0xD1, 0x38, 0x0D, 0xC6, 0x35, 0x98, 0x18, 0xF7, 0xEC, 0x6C,
-   0x43, 0x75, 0x37, 0x26, 0xFA, 0x13, 0x94, 0x48, 0xF2, 0xD0, 0x8B, 0x30,
-   0x84, 0x54, 0xDF, 0x23, 0x19, 0x5B, 0x3D, 0x59, 0xF3, 0xAE, 0xA2, 0x82,
-   0x63, 0x01, 0x83, 0x2E, 0xD9, 0x51, 0x9B, 0x7C, 0xA6, 0xEB, 0xA5, 0xBE,
-   0x16, 0x0C, 0xE3, 0x61, 0xC0, 0x8C, 0x3A, 0xF5, 0x73, 0x2C, 0x25, 0x0B,
-   0xBB, 0x4E, 0x89, 0x6B, 0x53, 0x6A, 0xB4, 0xF1, 0xE1, 0xE6, 0xBD, 0x45,
-   0xE2, 0xF4, 0xB6, 0x66, 0xCC, 0x95, 0x03, 0x56, 0xD4, 0x1C, 0x1E, 0xD7,
-   0xFB, 0xC3, 0x8E, 0xB5, 0xE9, 0xCF, 0xBF, 0xBA, 0xEA, 0x77, 0x39, 0xAF,
-   0x33, 0xC9, 0x62, 0x71, 0x81, 0x79, 0x09, 0xAD, 0x24, 0xCD, 0xF9, 0xD8,
-   0xE5, 0xC5, 0xB9, 0x4D, 0x44, 0x08, 0x86, 0xE7, 0xA1, 0x1D, 0xAA, 0xED,
-   0x06, 0x70, 0xB2, 0xD2, 0x41, 0x7B, 0xA0, 0x11, 0x31, 0xC2, 0x27, 0x90,
-   0x20, 0xF6, 0x60, 0xFF, 0x96, 0x5C, 0xB1, 0xAB, 0x9E, 0x9C, 0x52, 0x1B,
-   0x5F, 0x93, 0x0A, 0xEF, 0x91, 0x85, 0x49, 0xEE, 0x2D, 0x4F, 0x8F, 0x3B,
-   0x47, 0x87, 0x6D, 0x46, 0xD6, 0x3E, 0x69, 0x64, 0x2A, 0xCE, 0xCB, 0x2F,
-   0xFC, 0x97, 0x05, 0x7A, 0xAC, 0x7F, 0xD5, 0x1A, 0x4B, 0x0E, 0xA7, 0x5A,
-   0x28, 0x14, 0x3F, 0x29, 0x88, 0x3C, 0x4C, 0x02, 0xB8, 0xDA, 0xB0, 0x17,
-   0x55, 0x1F, 0x8A, 0x7D, 0x57, 0xC7, 0x8D, 0x74, 0xB7, 0xC4, 0x9F, 0x72,
-   0x7E, 0x15, 0x22, 0x12, 0x58, 0x07, 0x99, 0x34, 0x6E, 0x50, 0xDE, 0x68,
-   0x65, 0xBC, 0xDB, 0xF8, 0xC8, 0xA8, 0x2B, 0x40, 0xDC, 0xFE, 0x32, 0xA4,
-   0xCA, 0x10, 0x21, 0xF0, 0xD3, 0x5D, 0x0F, 0x00, 0x6F, 0x9D, 0x36, 0x42,
-   0x4A, 0x5E, 0xC1, 0xE0
-};
-
-static const u8 q1[256] = {
-   0x75, 0xF3, 0xC6, 0xF4, 0xDB, 0x7B, 0xFB, 0xC8, 0x4A, 0xD3, 0xE6, 0x6B,
-   0x45, 0x7D, 0xE8, 0x4B, 0xD6, 0x32, 0xD8, 0xFD, 0x37, 0x71, 0xF1, 0xE1,
-   0x30, 0x0F, 0xF8, 0x1B, 0x87, 0xFA, 0x06, 0x3F, 0x5E, 0xBA, 0xAE, 0x5B,
-   0x8A, 0x00, 0xBC, 0x9D, 0x6D, 0xC1, 0xB1, 0x0E, 0x80, 0x5D, 0xD2, 0xD5,
-   0xA0, 0x84, 0x07, 0x14, 0xB5, 0x90, 0x2C, 0xA3, 0xB2, 0x73, 0x4C, 0x54,
-   0x92, 0x74, 0x36, 0x51, 0x38, 0xB0, 0xBD, 0x5A, 0xFC, 0x60, 0x62, 0x96,
-   0x6C, 0x42, 0xF7, 0x10, 0x7C, 0x28, 0x27, 0x8C, 0x13, 0x95, 0x9C, 0xC7,
-   0x24, 0x46, 0x3B, 0x70, 0xCA, 0xE3, 0x85, 0xCB, 0x11, 0xD0, 0x93, 0xB8,
-   0xA6, 0x83, 0x20, 0xFF, 0x9F, 0x77, 0xC3, 0xCC, 0x03, 0x6F, 0x08, 0xBF,
-   0x40, 0xE7, 0x2B, 0xE2, 0x79, 0x0C, 0xAA, 0x82, 0x41, 0x3A, 0xEA, 0xB9,
-   0xE4, 0x9A, 0xA4, 0x97, 0x7E, 0xDA, 0x7A, 0x17, 0x66, 0x94, 0xA1, 0x1D,
-   0x3D, 0xF0, 0xDE, 0xB3, 0x0B, 0x72, 0xA7, 0x1C, 0xEF, 0xD1, 0x53, 0x3E,
-   0x8F, 0x33, 0x26, 0x5F, 0xEC, 0x76, 0x2A, 0x49, 0x81, 0x88, 0xEE, 0x21,
-   0xC4, 0x1A, 0xEB, 0xD9, 0xC5, 0x39, 0x99, 0xCD, 0xAD, 0x31, 0x8B, 0x01,
-   0x18, 0x23, 0xDD, 0x1F, 0x4E, 0x2D, 0xF9, 0x48, 0x4F, 0xF2, 0x65, 0x8E,
-   0x78, 0x5C, 0x58, 0x19, 0x8D, 0xE5, 0x98, 0x57, 0x67, 0x7F, 0x05, 0x64,
-   0xAF, 0x63, 0xB6, 0xFE, 0xF5, 0xB7, 0x3C, 0xA5, 0xCE, 0xE9, 0x68, 0x44,
-   0xE0, 0x4D, 0x43, 0x69, 0x29, 0x2E, 0xAC, 0x15, 0x59, 0xA8, 0x0A, 0x9E,
-   0x6E, 0x47, 0xDF, 0x34, 0x35, 0x6A, 0xCF, 0xDC, 0x22, 0xC9, 0xC0, 0x9B,
-   0x89, 0xD4, 0xED, 0xAB, 0x12, 0xA2, 0x0D, 0x52, 0xBB, 0x02, 0x2F, 0xA9,
-   0xD7, 0x61, 0x1E, 0xB4, 0x50, 0x04, 0xF6, 0xC2, 0x16, 0x25, 0x86, 0x56,
-   0x55, 0x09, 0xBE, 0x91
-};
-
-/* These MDS tables are actually tables of MDS composed with q0 and q1,
- * because it is only ever used that way and we can save some time by
- * precomputing.  Of course the main saving comes from precomputing the
- * GF(2^8) multiplication involved in the MDS matrix multiply; by looking
- * things up in these tables we reduce the matrix multiply to four lookups
- * and three XORs.  Semi-formally, the definition of these tables is:
- * mds[0][i] = MDS (q1[i] 0 0 0)^T  mds[1][i] = MDS (0 q0[i] 0 0)^T
- * mds[2][i] = MDS (0 0 q1[i] 0)^T  mds[3][i] = MDS (0 0 0 q0[i])^T
- * where ^T means "transpose", the matrix multiply is performed in GF(2^8)
- * represented as GF(2)[x]/v(x) where v(x)=x^8+x^6+x^5+x^3+1 as described
- * by Schneier et al, and I'm casually glossing over the byte/word
- * conversion issues. */
-
-static const u32 mds[4][256] = {
-   {0xBCBC3275, 0xECEC21F3, 0x202043C6, 0xB3B3C9F4, 0xDADA03DB, 0x02028B7B,
-    0xE2E22BFB, 0x9E9EFAC8, 0xC9C9EC4A, 0xD4D409D3, 0x18186BE6, 0x1E1E9F6B,
-    0x98980E45, 0xB2B2387D, 0xA6A6D2E8, 0x2626B74B, 0x3C3C57D6, 0x93938A32,
-    0x8282EED8, 0x525298FD, 0x7B7BD437, 0xBBBB3771, 0x5B5B97F1, 0x474783E1,
-    0x24243C30, 0x5151E20F, 0xBABAC6F8, 0x4A4AF31B, 0xBFBF4887, 0x0D0D70FA,
-    0xB0B0B306, 0x7575DE3F, 0xD2D2FD5E, 0x7D7D20BA, 0x666631AE, 0x3A3AA35B,
-    0x59591C8A, 0x00000000, 0xCDCD93BC, 0x1A1AE09D, 0xAEAE2C6D, 0x7F7FABC1,
-    0x2B2BC7B1, 0xBEBEB90E, 0xE0E0A080, 0x8A8A105D, 0x3B3B52D2, 0x6464BAD5,
-    0xD8D888A0, 0xE7E7A584, 0x5F5FE807, 0x1B1B1114, 0x2C2CC2B5, 0xFCFCB490,
-    0x3131272C, 0x808065A3, 0x73732AB2, 0x0C0C8173, 0x79795F4C, 0x6B6B4154,
-    0x4B4B0292, 0x53536974, 0x94948F36, 0x83831F51, 0x2A2A3638, 0xC4C49CB0,
-    0x2222C8BD, 0xD5D5F85A, 0xBDBDC3FC, 0x48487860, 0xFFFFCE62, 0x4C4C0796,
-    0x4141776C, 0xC7C7E642, 0xEBEB24F7, 0x1C1C1410, 0x5D5D637C, 0x36362228,
-    0x6767C027, 0xE9E9AF8C, 0x4444F913, 0x1414EA95, 0xF5F5BB9C, 0xCFCF18C7,
-    0x3F3F2D24, 0xC0C0E346, 0x7272DB3B, 0x54546C70, 0x29294CCA, 0xF0F035E3,
-    0x0808FE85, 0xC6C617CB, 0xF3F34F11, 0x8C8CE4D0, 0xA4A45993, 0xCACA96B8,
-    0x68683BA6, 0xB8B84D83, 0x38382820, 0xE5E52EFF, 0xADAD569F, 0x0B0B8477,
-    0xC8C81DC3, 0x9999FFCC, 0x5858ED03, 0x19199A6F, 0x0E0E0A08, 0x95957EBF,
-    0x70705040, 0xF7F730E7, 0x6E6ECF2B, 0x1F1F6EE2, 0xB5B53D79, 0x09090F0C,
-    0x616134AA, 0x57571682, 0x9F9F0B41, 0x9D9D803A, 0x111164EA, 0x2525CDB9,
-    0xAFAFDDE4, 0x4545089A, 0xDFDF8DA4, 0xA3A35C97, 0xEAEAD57E, 0x353558DA,
-    0xEDEDD07A, 0x4343FC17, 0xF8F8CB66, 0xFBFBB194, 0x3737D3A1, 0xFAFA401D,
-    0xC2C2683D, 0xB4B4CCF0, 0x32325DDE, 0x9C9C71B3, 0x5656E70B, 0xE3E3DA72,
-    0x878760A7, 0x15151B1C, 0xF9F93AEF, 0x6363BFD1, 0x3434A953, 0x9A9A853E,
-    0xB1B1428F, 0x7C7CD133, 0x88889B26, 0x3D3DA65F, 0xA1A1D7EC, 0xE4E4DF76,
-    0x8181942A, 0x91910149, 0x0F0FFB81, 0xEEEEAA88, 0x161661EE, 0xD7D77321,
-    0x9797F5C4, 0xA5A5A81A, 0xFEFE3FEB, 0x6D6DB5D9, 0x7878AEC5, 0xC5C56D39,
-    0x1D1DE599, 0x7676A4CD, 0x3E3EDCAD, 0xCBCB6731, 0xB6B6478B, 0xEFEF5B01,
-    0x12121E18, 0x6060C523, 0x6A6AB0DD, 0x4D4DF61F, 0xCECEE94E, 0xDEDE7C2D,
-    0x55559DF9, 0x7E7E5A48, 0x2121B24F, 0x03037AF2, 0xA0A02665, 0x5E5E198E,
-    0x5A5A6678, 0x65654B5C, 0x62624E58, 0xFDFD4519, 0x0606F48D, 0x404086E5,
-    0xF2F2BE98, 0x3333AC57, 0x17179067, 0x05058E7F, 0xE8E85E05, 0x4F4F7D64,
-    0x89896AAF, 0x10109563, 0x74742FB6, 0x0A0A75FE, 0x5C5C92F5, 0x9B9B74B7,
-    0x2D2D333C, 0x3030D6A5, 0x2E2E49CE, 0x494989E9, 0x46467268, 0x77775544,
-    0xA8A8D8E0, 0x9696044D, 0x2828BD43, 0xA9A92969, 0xD9D97929, 0x8686912E,
-    0xD1D187AC, 0xF4F44A15, 0x8D8D1559, 0xD6D682A8, 0xB9B9BC0A, 0x42420D9E,
-    0xF6F6C16E, 0x2F2FB847, 0xDDDD06DF, 0x23233934, 0xCCCC6235, 0xF1F1C46A,
-    0xC1C112CF, 0x8585EBDC, 0x8F8F9E22, 0x7171A1C9, 0x9090F0C0, 0xAAAA539B,
-    0x0101F189, 0x8B8BE1D4, 0x4E4E8CED, 0x8E8E6FAB, 0xABABA212, 0x6F6F3EA2,
-    0xE6E6540D, 0xDBDBF252, 0x92927BBB, 0xB7B7B602, 0x6969CA2F, 0x3939D9A9,
-    0xD3D30CD7, 0xA7A72361, 0xA2A2AD1E, 0xC3C399B4, 0x6C6C4450, 0x07070504,
-    0x04047FF6, 0x272746C2, 0xACACA716, 0xD0D07625, 0x50501386, 0xDCDCF756,
-    0x84841A55, 0xE1E15109, 0x7A7A25BE, 0x1313EF91},
-
-   {0xA9D93939, 0x67901717, 0xB3719C9C, 0xE8D2A6A6, 0x04050707, 0xFD985252,
-    0xA3658080, 0x76DFE4E4, 0x9A084545, 0x92024B4B, 0x80A0E0E0, 0x78665A5A,
-    0xE4DDAFAF, 0xDDB06A6A, 0xD1BF6363, 0x38362A2A, 0x0D54E6E6, 0xC6432020,
-    0x3562CCCC, 0x98BEF2F2, 0x181E1212, 0xF724EBEB, 0xECD7A1A1, 0x6C774141,
-    0x43BD2828, 0x7532BCBC, 0x37D47B7B, 0x269B8888, 0xFA700D0D, 0x13F94444,
-    0x94B1FBFB, 0x485A7E7E, 0xF27A0303, 0xD0E48C8C, 0x8B47B6B6, 0x303C2424,
-    0x84A5E7E7, 0x54416B6B, 0xDF06DDDD, 0x23C56060, 0x1945FDFD, 0x5BA33A3A,
-    0x3D68C2C2, 0x59158D8D, 0xF321ECEC, 0xAE316666, 0xA23E6F6F, 0x82165757,
-    0x63951010, 0x015BEFEF, 0x834DB8B8, 0x2E918686, 0xD9B56D6D, 0x511F8383,
-    0x9B53AAAA, 0x7C635D5D, 0xA63B6868, 0xEB3FFEFE, 0xA5D63030, 0xBE257A7A,
-    0x16A7ACAC, 0x0C0F0909, 0xE335F0F0, 0x6123A7A7, 0xC0F09090, 0x8CAFE9E9,
-    0x3A809D9D, 0xF5925C5C, 0x73810C0C, 0x2C273131, 0x2576D0D0, 0x0BE75656,
-    0xBB7B9292, 0x4EE9CECE, 0x89F10101, 0x6B9F1E1E, 0x53A93434, 0x6AC4F1F1,
-    0xB499C3C3, 0xF1975B5B, 0xE1834747, 0xE66B1818, 0xBDC82222, 0x450E9898,
-    0xE26E1F1F, 0xF4C9B3B3, 0xB62F7474, 0x66CBF8F8, 0xCCFF9999, 0x95EA1414,
-    0x03ED5858, 0x56F7DCDC, 0xD4E18B8B, 0x1C1B1515, 0x1EADA2A2, 0xD70CD3D3,
-    0xFB2BE2E2, 0xC31DC8C8, 0x8E195E5E, 0xB5C22C2C, 0xE9894949, 0xCF12C1C1,
-    0xBF7E9595, 0xBA207D7D, 0xEA641111, 0x77840B0B, 0x396DC5C5, 0xAF6A8989,
-    0x33D17C7C, 0xC9A17171, 0x62CEFFFF, 0x7137BBBB, 0x81FB0F0F, 0x793DB5B5,
-    0x0951E1E1, 0xADDC3E3E, 0x242D3F3F, 0xCDA47676, 0xF99D5555, 0xD8EE8282,
-    0xE5864040, 0xC5AE7878, 0xB9CD2525, 0x4D049696, 0x44557777, 0x080A0E0E,
-    0x86135050, 0xE730F7F7, 0xA1D33737, 0x1D40FAFA, 0xAA346161, 0xED8C4E4E,
-    0x06B3B0B0, 0x706C5454, 0xB22A7373, 0xD2523B3B, 0x410B9F9F, 0x7B8B0202,
-    0xA088D8D8, 0x114FF3F3, 0x3167CBCB, 0xC2462727, 0x27C06767, 0x90B4FCFC,
-    0x20283838, 0xF67F0404, 0x60784848, 0xFF2EE5E5, 0x96074C4C, 0x5C4B6565,
-    0xB1C72B2B, 0xAB6F8E8E, 0x9E0D4242, 0x9CBBF5F5, 0x52F2DBDB, 0x1BF34A4A,
-    0x5FA63D3D, 0x9359A4A4, 0x0ABCB9B9, 0xEF3AF9F9, 0x91EF1313, 0x85FE0808,
-    0x49019191, 0xEE611616, 0x2D7CDEDE, 0x4FB22121, 0x8F42B1B1, 0x3BDB7272,
-    0x47B82F2F, 0x8748BFBF, 0x6D2CAEAE, 0x46E3C0C0, 0xD6573C3C, 0x3E859A9A,
-    0x6929A9A9, 0x647D4F4F, 0x2A948181, 0xCE492E2E, 0xCB17C6C6, 0x2FCA6969,
-    0xFCC3BDBD, 0x975CA3A3, 0x055EE8E8, 0x7AD0EDED, 0xAC87D1D1, 0x7F8E0505,
-    0xD5BA6464, 0x1AA8A5A5, 0x4BB72626, 0x0EB9BEBE, 0xA7608787, 0x5AF8D5D5,
-    0x28223636, 0x14111B1B, 0x3FDE7575, 0x2979D9D9, 0x88AAEEEE, 0x3C332D2D,
-    0x4C5F7979, 0x02B6B7B7, 0xB896CACA, 0xDA583535, 0xB09CC4C4, 0x17FC4343,
-    0x551A8484, 0x1FF64D4D, 0x8A1C5959, 0x7D38B2B2, 0x57AC3333, 0xC718CFCF,
-    0x8DF40606, 0x74695353, 0xB7749B9B, 0xC4F59797, 0x9F56ADAD, 0x72DAE3E3,
-    0x7ED5EAEA, 0x154AF4F4, 0x229E8F8F, 0x12A2ABAB, 0x584E6262, 0x07E85F5F,
-    0x99E51D1D, 0x34392323, 0x6EC1F6F6, 0x50446C6C, 0xDE5D3232, 0x68724646,
-    0x6526A0A0, 0xBC93CDCD, 0xDB03DADA, 0xF8C6BABA, 0xC8FA9E9E, 0xA882D6D6,
-    0x2BCF6E6E, 0x40507070, 0xDCEB8585, 0xFE750A0A, 0x328A9393, 0xA48DDFDF,
-    0xCA4C2929, 0x10141C1C, 0x2173D7D7, 0xF0CCB4B4, 0xD309D4D4, 0x5D108A8A,
-    0x0FE25151, 0x00000000, 0x6F9A1919, 0x9DE01A1A, 0x368F9494, 0x42E6C7C7,
-    0x4AECC9C9, 0x5EFDD2D2, 0xC1AB7F7F, 0xE0D8A8A8},
-
-   {0xBC75BC32, 0xECF3EC21, 0x20C62043, 0xB3F4B3C9, 0xDADBDA03, 0x027B028B,
-    0xE2FBE22B, 0x9EC89EFA, 0xC94AC9EC, 0xD4D3D409, 0x18E6186B, 0x1E6B1E9F,
-    0x9845980E, 0xB27DB238, 0xA6E8A6D2, 0x264B26B7, 0x3CD63C57, 0x9332938A,
-    0x82D882EE, 0x52FD5298, 0x7B377BD4, 0xBB71BB37, 0x5BF15B97, 0x47E14783,
-    0x2430243C, 0x510F51E2, 0xBAF8BAC6, 0x4A1B4AF3, 0xBF87BF48, 0x0DFA0D70,
-    0xB006B0B3, 0x753F75DE, 0xD25ED2FD, 0x7DBA7D20, 0x66AE6631, 0x3A5B3AA3,
-    0x598A591C, 0x00000000, 0xCDBCCD93, 0x1A9D1AE0, 0xAE6DAE2C, 0x7FC17FAB,
-    0x2BB12BC7, 0xBE0EBEB9, 0xE080E0A0, 0x8A5D8A10, 0x3BD23B52, 0x64D564BA,
-    0xD8A0D888, 0xE784E7A5, 0x5F075FE8, 0x1B141B11, 0x2CB52CC2, 0xFC90FCB4,
-    0x312C3127, 0x80A38065, 0x73B2732A, 0x0C730C81, 0x794C795F, 0x6B546B41,
-    0x4B924B02, 0x53745369, 0x9436948F, 0x8351831F, 0x2A382A36, 0xC4B0C49C,
-    0x22BD22C8, 0xD55AD5F8, 0xBDFCBDC3, 0x48604878, 0xFF62FFCE, 0x4C964C07,
-    0x416C4177, 0xC742C7E6, 0xEBF7EB24, 0x1C101C14, 0x5D7C5D63, 0x36283622,
-    0x672767C0, 0xE98CE9AF, 0x441344F9, 0x149514EA, 0xF59CF5BB, 0xCFC7CF18,
-    0x3F243F2D, 0xC046C0E3, 0x723B72DB, 0x5470546C, 0x29CA294C, 0xF0E3F035,
-    0x088508FE, 0xC6CBC617, 0xF311F34F, 0x8CD08CE4, 0xA493A459, 0xCAB8CA96,
-    0x68A6683B, 0xB883B84D, 0x38203828, 0xE5FFE52E, 0xAD9FAD56, 0x0B770B84,
-    0xC8C3C81D, 0x99CC99FF, 0x580358ED, 0x196F199A, 0x0E080E0A, 0x95BF957E,
-    0x70407050, 0xF7E7F730, 0x6E2B6ECF, 0x1FE21F6E, 0xB579B53D, 0x090C090F,
-    0x61AA6134, 0x57825716, 0x9F419F0B, 0x9D3A9D80, 0x11EA1164, 0x25B925CD,
-    0xAFE4AFDD, 0x459A4508, 0xDFA4DF8D, 0xA397A35C, 0xEA7EEAD5, 0x35DA3558,
-    0xED7AEDD0, 0x431743FC, 0xF866F8CB, 0xFB94FBB1, 0x37A137D3, 0xFA1DFA40,
-    0xC23DC268, 0xB4F0B4CC, 0x32DE325D, 0x9CB39C71, 0x560B56E7, 0xE372E3DA,
-    0x87A78760, 0x151C151B, 0xF9EFF93A, 0x63D163BF, 0x345334A9, 0x9A3E9A85,
-    0xB18FB142, 0x7C337CD1, 0x8826889B, 0x3D5F3DA6, 0xA1ECA1D7, 0xE476E4DF,
-    0x812A8194, 0x91499101, 0x0F810FFB, 0xEE88EEAA, 0x16EE1661, 0xD721D773,
-    0x97C497F5, 0xA51AA5A8, 0xFEEBFE3F, 0x6DD96DB5, 0x78C578AE, 0xC539C56D,
-    0x1D991DE5, 0x76CD76A4, 0x3EAD3EDC, 0xCB31CB67, 0xB68BB647, 0xEF01EF5B,
-    0x1218121E, 0x602360C5, 0x6ADD6AB0, 0x4D1F4DF6, 0xCE4ECEE9, 0xDE2DDE7C,
-    0x55F9559D, 0x7E487E5A, 0x214F21B2, 0x03F2037A, 0xA065A026, 0x5E8E5E19,
-    0x5A785A66, 0x655C654B, 0x6258624E, 0xFD19FD45, 0x068D06F4, 0x40E54086,
-    0xF298F2BE, 0x335733AC, 0x17671790, 0x057F058E, 0xE805E85E, 0x4F644F7D,
-    0x89AF896A, 0x10631095, 0x74B6742F, 0x0AFE0A75, 0x5CF55C92, 0x9BB79B74,
-    0x2D3C2D33, 0x30A530D6, 0x2ECE2E49, 0x49E94989, 0x46684672, 0x77447755,
-    0xA8E0A8D8, 0x964D9604, 0x284328BD, 0xA969A929, 0xD929D979, 0x862E8691,
-    0xD1ACD187, 0xF415F44A, 0x8D598D15, 0xD6A8D682, 0xB90AB9BC, 0x429E420D,
-    0xF66EF6C1, 0x2F472FB8, 0xDDDFDD06, 0x23342339, 0xCC35CC62, 0xF16AF1C4,
-    0xC1CFC112, 0x85DC85EB, 0x8F228F9E, 0x71C971A1, 0x90C090F0, 0xAA9BAA53,
-    0x018901F1, 0x8BD48BE1, 0x4EED4E8C, 0x8EAB8E6F, 0xAB12ABA2, 0x6FA26F3E,
-    0xE60DE654, 0xDB52DBF2, 0x92BB927B, 0xB702B7B6, 0x692F69CA, 0x39A939D9,
-    0xD3D7D30C, 0xA761A723, 0xA21EA2AD, 0xC3B4C399, 0x6C506C44, 0x07040705,
-    0x04F6047F, 0x27C22746, 0xAC16ACA7, 0xD025D076, 0x50865013, 0xDC56DCF7,
-    0x8455841A, 0xE109E151, 0x7ABE7A25, 0x139113EF},
-
-   {0xD939A9D9, 0x90176790, 0x719CB371, 0xD2A6E8D2, 0x05070405, 0x9852FD98,
-    0x6580A365, 0xDFE476DF, 0x08459A08, 0x024B9202, 0xA0E080A0, 0x665A7866,
-    0xDDAFE4DD, 0xB06ADDB0, 0xBF63D1BF, 0x362A3836, 0x54E60D54, 0x4320C643,
-    0x62CC3562, 0xBEF298BE, 0x1E12181E, 0x24EBF724, 0xD7A1ECD7, 0x77416C77,
-    0xBD2843BD, 0x32BC7532, 0xD47B37D4, 0x9B88269B, 0x700DFA70, 0xF94413F9,
-    0xB1FB94B1, 0x5A7E485A, 0x7A03F27A, 0xE48CD0E4, 0x47B68B47, 0x3C24303C,
-    0xA5E784A5, 0x416B5441, 0x06DDDF06, 0xC56023C5, 0x45FD1945, 0xA33A5BA3,
-    0x68C23D68, 0x158D5915, 0x21ECF321, 0x3166AE31, 0x3E6FA23E, 0x16578216,
-    0x95106395, 0x5BEF015B, 0x4DB8834D, 0x91862E91, 0xB56DD9B5, 0x1F83511F,
-    0x53AA9B53, 0x635D7C63, 0x3B68A63B, 0x3FFEEB3F, 0xD630A5D6, 0x257ABE25,
-    0xA7AC16A7, 0x0F090C0F, 0x35F0E335, 0x23A76123, 0xF090C0F0, 0xAFE98CAF,
-    0x809D3A80, 0x925CF592, 0x810C7381, 0x27312C27, 0x76D02576, 0xE7560BE7,
-    0x7B92BB7B, 0xE9CE4EE9, 0xF10189F1, 0x9F1E6B9F, 0xA93453A9, 0xC4F16AC4,
-    0x99C3B499, 0x975BF197, 0x8347E183, 0x6B18E66B, 0xC822BDC8, 0x0E98450E,
-    0x6E1FE26E, 0xC9B3F4C9, 0x2F74B62F, 0xCBF866CB, 0xFF99CCFF, 0xEA1495EA,
-    0xED5803ED, 0xF7DC56F7, 0xE18BD4E1, 0x1B151C1B, 0xADA21EAD, 0x0CD3D70C,
-    0x2BE2FB2B, 0x1DC8C31D, 0x195E8E19, 0xC22CB5C2, 0x8949E989, 0x12C1CF12,
-    0x7E95BF7E, 0x207DBA20, 0x6411EA64, 0x840B7784, 0x6DC5396D, 0x6A89AF6A,
-    0xD17C33D1, 0xA171C9A1, 0xCEFF62CE, 0x37BB7137, 0xFB0F81FB, 0x3DB5793D,
-    0x51E10951, 0xDC3EADDC, 0x2D3F242D, 0xA476CDA4, 0x9D55F99D, 0xEE82D8EE,
-    0x8640E586, 0xAE78C5AE, 0xCD25B9CD, 0x04964D04, 0x55774455, 0x0A0E080A,
-    0x13508613, 0x30F7E730, 0xD337A1D3, 0x40FA1D40, 0x3461AA34, 0x8C4EED8C,
-    0xB3B006B3, 0x6C54706C, 0x2A73B22A, 0x523BD252, 0x0B9F410B, 0x8B027B8B,
-    0x88D8A088, 0x4FF3114F, 0x67CB3167, 0x4627C246, 0xC06727C0, 0xB4FC90B4,
-    0x28382028, 0x7F04F67F, 0x78486078, 0x2EE5FF2E, 0x074C9607, 0x4B655C4B,
-    0xC72BB1C7, 0x6F8EAB6F, 0x0D429E0D, 0xBBF59CBB, 0xF2DB52F2, 0xF34A1BF3,
-    0xA63D5FA6, 0x59A49359, 0xBCB90ABC, 0x3AF9EF3A, 0xEF1391EF, 0xFE0885FE,
-    0x01914901, 0x6116EE61, 0x7CDE2D7C, 0xB2214FB2, 0x42B18F42, 0xDB723BDB,
-    0xB82F47B8, 0x48BF8748, 0x2CAE6D2C, 0xE3C046E3, 0x573CD657, 0x859A3E85,
-    0x29A96929, 0x7D4F647D, 0x94812A94, 0x492ECE49, 0x17C6CB17, 0xCA692FCA,
-    0xC3BDFCC3, 0x5CA3975C, 0x5EE8055E, 0xD0ED7AD0, 0x87D1AC87, 0x8E057F8E,
-    0xBA64D5BA, 0xA8A51AA8, 0xB7264BB7, 0xB9BE0EB9, 0x6087A760, 0xF8D55AF8,
-    0x22362822, 0x111B1411, 0xDE753FDE, 0x79D92979, 0xAAEE88AA, 0x332D3C33,
-    0x5F794C5F, 0xB6B702B6, 0x96CAB896, 0x5835DA58, 0x9CC4B09C, 0xFC4317FC,
-    0x1A84551A, 0xF64D1FF6, 0x1C598A1C, 0x38B27D38, 0xAC3357AC, 0x18CFC718,
-    0xF4068DF4, 0x69537469, 0x749BB774, 0xF597C4F5, 0x56AD9F56, 0xDAE372DA,
-    0xD5EA7ED5, 0x4AF4154A, 0x9E8F229E, 0xA2AB12A2, 0x4E62584E, 0xE85F07E8,
-    0xE51D99E5, 0x39233439, 0xC1F66EC1, 0x446C5044, 0x5D32DE5D, 0x72466872,
-    0x26A06526, 0x93CDBC93, 0x03DADB03, 0xC6BAF8C6, 0xFA9EC8FA, 0x82D6A882,
-    0xCF6E2BCF, 0x50704050, 0xEB85DCEB, 0x750AFE75, 0x8A93328A, 0x8DDFA48D,
-    0x4C29CA4C, 0x141C1014, 0x73D72173, 0xCCB4F0CC, 0x09D4D309, 0x108A5D10,
-    0xE2510FE2, 0x00000000, 0x9A196F9A, 0xE01A9DE0, 0x8F94368F, 0xE6C742E6,
-    0xECC94AEC, 0xFDD25EFD, 0xAB7FC1AB, 0xD8A8E0D8}
-};
-
-/* The exp_to_poly and poly_to_exp tables are used to perform efficient
- * operations in GF(2^8) represented as GF(2)[x]/w(x) where
- * w(x)=x^8+x^6+x^3+x^2+1.  We care about doing that because it's part of the
- * definition of the RS matrix in the key schedule.  Elements of that field
- * are polynomials of degree not greater than 7 and all coefficients 0 or 1,
- * which can be represented naturally by bytes (just substitute x=2).  In that
- * form, GF(2^8) addition is the same as bitwise XOR, but GF(2^8)
- * multiplication is inefficient without hardware support.  To multiply
- * faster, I make use of the fact x is a generator for the nonzero elements,
- * so that every element p of GF(2)[x]/w(x) is either 0 or equal to (x)^n for
- * some n in 0..254.  Note that that caret is exponentiation in GF(2^8),
- * *not* polynomial notation.  So if I want to compute pq where p and q are
- * in GF(2^8), I can just say:
- *    1. if p=0 or q=0 then pq=0
- *    2. otherwise, find m and n such that p=x^m and q=x^n
- *    3. pq=(x^m)(x^n)=x^(m+n), so add m and n and find pq
- * The translations in steps 2 and 3 are looked up in the tables
- * poly_to_exp (for step 2) and exp_to_poly (for step 3).  To see this
- * in action, look at the CALC_S macro.  As additional wrinkles, note that
- * one of my operands is always a constant, so the poly_to_exp lookup on it
- * is done in advance; I included the original values in the comments so
- * readers can have some chance of recognizing that this *is* the RS matrix
- * from the Twofish paper.  I've only included the table entries I actually
- * need; I never do a lookup on a variable input of zero and the biggest
- * exponents I'll ever see are 254 (variable) and 237 (constant), so they'll
- * never sum to more than 491.	I'm repeating part of the exp_to_poly table
- * so that I don't have to do mod-255 reduction in the exponent arithmetic.
- * Since I know my constant operands are never zero, I only have to worry
- * about zero values in the variable operand, and I do it with a simple
- * conditional branch.	I know conditionals are expensive, but I couldn't
- * see a non-horrible way of avoiding them, and I did manage to group the
- * statements so that each if covers four group multiplications. */
-
-static const u8 poly_to_exp[255] = {
-   0x00, 0x01, 0x17, 0x02, 0x2E, 0x18, 0x53, 0x03, 0x6A, 0x2F, 0x93, 0x19,
-   0x34, 0x54, 0x45, 0x04, 0x5C, 0x6B, 0xB6, 0x30, 0xA6, 0x94, 0x4B, 0x1A,
-   0x8C, 0x35, 0x81, 0x55, 0xAA, 0x46, 0x0D, 0x05, 0x24, 0x5D, 0x87, 0x6C,
-   0x9B, 0xB7, 0xC1, 0x31, 0x2B, 0xA7, 0xA3, 0x95, 0x98, 0x4C, 0xCA, 0x1B,
-   0xE6, 0x8D, 0x73, 0x36, 0xCD, 0x82, 0x12, 0x56, 0x62, 0xAB, 0xF0, 0x47,
-   0x4F, 0x0E, 0xBD, 0x06, 0xD4, 0x25, 0xD2, 0x5E, 0x27, 0x88, 0x66, 0x6D,
-   0xD6, 0x9C, 0x79, 0xB8, 0x08, 0xC2, 0xDF, 0x32, 0x68, 0x2C, 0xFD, 0xA8,
-   0x8A, 0xA4, 0x5A, 0x96, 0x29, 0x99, 0x22, 0x4D, 0x60, 0xCB, 0xE4, 0x1C,
-   0x7B, 0xE7, 0x3B, 0x8E, 0x9E, 0x74, 0xF4, 0x37, 0xD8, 0xCE, 0xF9, 0x83,
-   0x6F, 0x13, 0xB2, 0x57, 0xE1, 0x63, 0xDC, 0xAC, 0xC4, 0xF1, 0xAF, 0x48,
-   0x0A, 0x50, 0x42, 0x0F, 0xBA, 0xBE, 0xC7, 0x07, 0xDE, 0xD5, 0x78, 0x26,
-   0x65, 0xD3, 0xD1, 0x5F, 0xE3, 0x28, 0x21, 0x89, 0x59, 0x67, 0xFC, 0x6E,
-   0xB1, 0xD7, 0xF8, 0x9D, 0xF3, 0x7A, 0x3A, 0xB9, 0xC6, 0x09, 0x41, 0xC3,
-   0xAE, 0xE0, 0xDB, 0x33, 0x44, 0x69, 0x92, 0x2D, 0x52, 0xFE, 0x16, 0xA9,
-   0x0C, 0x8B, 0x80, 0xA5, 0x4A, 0x5B, 0xB5, 0x97, 0xC9, 0x2A, 0xA2, 0x9A,
-   0xC0, 0x23, 0x86, 0x4E, 0xBC, 0x61, 0xEF, 0xCC, 0x11, 0xE5, 0x72, 0x1D,
-   0x3D, 0x7C, 0xEB, 0xE8, 0xE9, 0x3C, 0xEA, 0x8F, 0x7D, 0x9F, 0xEC, 0x75,
-   0x1E, 0xF5, 0x3E, 0x38, 0xF6, 0xD9, 0x3F, 0xCF, 0x76, 0xFA, 0x1F, 0x84,
-   0xA0, 0x70, 0xED, 0x14, 0x90, 0xB3, 0x7E, 0x58, 0xFB, 0xE2, 0x20, 0x64,
-   0xD0, 0xDD, 0x77, 0xAD, 0xDA, 0xC5, 0x40, 0xF2, 0x39, 0xB0, 0xF7, 0x49,
-   0xB4, 0x0B, 0x7F, 0x51, 0x15, 0x43, 0x91, 0x10, 0x71, 0xBB, 0xEE, 0xBF,
-   0x85, 0xC8, 0xA1
-};
-
-static const u8 exp_to_poly[492] = {
-   0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x4D, 0x9A, 0x79, 0xF2,
-   0xA9, 0x1F, 0x3E, 0x7C, 0xF8, 0xBD, 0x37, 0x6E, 0xDC, 0xF5, 0xA7, 0x03,
-   0x06, 0x0C, 0x18, 0x30, 0x60, 0xC0, 0xCD, 0xD7, 0xE3, 0x8B, 0x5B, 0xB6,
-   0x21, 0x42, 0x84, 0x45, 0x8A, 0x59, 0xB2, 0x29, 0x52, 0xA4, 0x05, 0x0A,
-   0x14, 0x28, 0x50, 0xA0, 0x0D, 0x1A, 0x34, 0x68, 0xD0, 0xED, 0x97, 0x63,
-   0xC6, 0xC1, 0xCF, 0xD3, 0xEB, 0x9B, 0x7B, 0xF6, 0xA1, 0x0F, 0x1E, 0x3C,
-   0x78, 0xF0, 0xAD, 0x17, 0x2E, 0x5C, 0xB8, 0x3D, 0x7A, 0xF4, 0xA5, 0x07,
-   0x0E, 0x1C, 0x38, 0x70, 0xE0, 0x8D, 0x57, 0xAE, 0x11, 0x22, 0x44, 0x88,
-   0x5D, 0xBA, 0x39, 0x72, 0xE4, 0x85, 0x47, 0x8E, 0x51, 0xA2, 0x09, 0x12,
-   0x24, 0x48, 0x90, 0x6D, 0xDA, 0xF9, 0xBF, 0x33, 0x66, 0xCC, 0xD5, 0xE7,
-   0x83, 0x4B, 0x96, 0x61, 0xC2, 0xC9, 0xDF, 0xF3, 0xAB, 0x1B, 0x36, 0x6C,
-   0xD8, 0xFD, 0xB7, 0x23, 0x46, 0x8C, 0x55, 0xAA, 0x19, 0x32, 0x64, 0xC8,
-   0xDD, 0xF7, 0xA3, 0x0B, 0x16, 0x2C, 0x58, 0xB0, 0x2D, 0x5A, 0xB4, 0x25,
-   0x4A, 0x94, 0x65, 0xCA, 0xD9, 0xFF, 0xB3, 0x2B, 0x56, 0xAC, 0x15, 0x2A,
-   0x54, 0xA8, 0x1D, 0x3A, 0x74, 0xE8, 0x9D, 0x77, 0xEE, 0x91, 0x6F, 0xDE,
-   0xF1, 0xAF, 0x13, 0x26, 0x4C, 0x98, 0x7D, 0xFA, 0xB9, 0x3F, 0x7E, 0xFC,
-   0xB5, 0x27, 0x4E, 0x9C, 0x75, 0xEA, 0x99, 0x7F, 0xFE, 0xB1, 0x2F, 0x5E,
-   0xBC, 0x35, 0x6A, 0xD4, 0xE5, 0x87, 0x43, 0x86, 0x41, 0x82, 0x49, 0x92,
-   0x69, 0xD2, 0xE9, 0x9F, 0x73, 0xE6, 0x81, 0x4F, 0x9E, 0x71, 0xE2, 0x89,
-   0x5F, 0xBE, 0x31, 0x62, 0xC4, 0xC5, 0xC7, 0xC3, 0xCB, 0xDB, 0xFB, 0xBB,
-   0x3B, 0x76, 0xEC, 0x95, 0x67, 0xCE, 0xD1, 0xEF, 0x93, 0x6B, 0xD6, 0xE1,
-   0x8F, 0x53, 0xA6, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x4D,
-   0x9A, 0x79, 0xF2, 0xA9, 0x1F, 0x3E, 0x7C, 0xF8, 0xBD, 0x37, 0x6E, 0xDC,
-   0xF5, 0xA7, 0x03, 0x06, 0x0C, 0x18, 0x30, 0x60, 0xC0, 0xCD, 0xD7, 0xE3,
-   0x8B, 0x5B, 0xB6, 0x21, 0x42, 0x84, 0x45, 0x8A, 0x59, 0xB2, 0x29, 0x52,
-   0xA4, 0x05, 0x0A, 0x14, 0x28, 0x50, 0xA0, 0x0D, 0x1A, 0x34, 0x68, 0xD0,
-   0xED, 0x97, 0x63, 0xC6, 0xC1, 0xCF, 0xD3, 0xEB, 0x9B, 0x7B, 0xF6, 0xA1,
-   0x0F, 0x1E, 0x3C, 0x78, 0xF0, 0xAD, 0x17, 0x2E, 0x5C, 0xB8, 0x3D, 0x7A,
-   0xF4, 0xA5, 0x07, 0x0E, 0x1C, 0x38, 0x70, 0xE0, 0x8D, 0x57, 0xAE, 0x11,
-   0x22, 0x44, 0x88, 0x5D, 0xBA, 0x39, 0x72, 0xE4, 0x85, 0x47, 0x8E, 0x51,
-   0xA2, 0x09, 0x12, 0x24, 0x48, 0x90, 0x6D, 0xDA, 0xF9, 0xBF, 0x33, 0x66,
-   0xCC, 0xD5, 0xE7, 0x83, 0x4B, 0x96, 0x61, 0xC2, 0xC9, 0xDF, 0xF3, 0xAB,
-   0x1B, 0x36, 0x6C, 0xD8, 0xFD, 0xB7, 0x23, 0x46, 0x8C, 0x55, 0xAA, 0x19,
-   0x32, 0x64, 0xC8, 0xDD, 0xF7, 0xA3, 0x0B, 0x16, 0x2C, 0x58, 0xB0, 0x2D,
-   0x5A, 0xB4, 0x25, 0x4A, 0x94, 0x65, 0xCA, 0xD9, 0xFF, 0xB3, 0x2B, 0x56,
-   0xAC, 0x15, 0x2A, 0x54, 0xA8, 0x1D, 0x3A, 0x74, 0xE8, 0x9D, 0x77, 0xEE,
-   0x91, 0x6F, 0xDE, 0xF1, 0xAF, 0x13, 0x26, 0x4C, 0x98, 0x7D, 0xFA, 0xB9,
-   0x3F, 0x7E, 0xFC, 0xB5, 0x27, 0x4E, 0x9C, 0x75, 0xEA, 0x99, 0x7F, 0xFE,
-   0xB1, 0x2F, 0x5E, 0xBC, 0x35, 0x6A, 0xD4, 0xE5, 0x87, 0x43, 0x86, 0x41,
-   0x82, 0x49, 0x92, 0x69, 0xD2, 0xE9, 0x9F, 0x73, 0xE6, 0x81, 0x4F, 0x9E,
-   0x71, 0xE2, 0x89, 0x5F, 0xBE, 0x31, 0x62, 0xC4, 0xC5, 0xC7, 0xC3, 0xCB
-};
-
-
-/* The table constants are indices of
- * S-box entries, preprocessed through q0 and q1. */
-static const u8 calc_sb_tbl[512] = {
-    0xA9, 0x75, 0x67, 0xF3, 0xB3, 0xC6, 0xE8, 0xF4,
-    0x04, 0xDB, 0xFD, 0x7B, 0xA3, 0xFB, 0x76, 0xC8,
-    0x9A, 0x4A, 0x92, 0xD3, 0x80, 0xE6, 0x78, 0x6B,
-    0xE4, 0x45, 0xDD, 0x7D, 0xD1, 0xE8, 0x38, 0x4B,
-    0x0D, 0xD6, 0xC6, 0x32, 0x35, 0xD8, 0x98, 0xFD,
-    0x18, 0x37, 0xF7, 0x71, 0xEC, 0xF1, 0x6C, 0xE1,
-    0x43, 0x30, 0x75, 0x0F, 0x37, 0xF8, 0x26, 0x1B,
-    0xFA, 0x87, 0x13, 0xFA, 0x94, 0x06, 0x48, 0x3F,
-    0xF2, 0x5E, 0xD0, 0xBA, 0x8B, 0xAE, 0x30, 0x5B,
-    0x84, 0x8A, 0x54, 0x00, 0xDF, 0xBC, 0x23, 0x9D,
-    0x19, 0x6D, 0x5B, 0xC1, 0x3D, 0xB1, 0x59, 0x0E,
-    0xF3, 0x80, 0xAE, 0x5D, 0xA2, 0xD2, 0x82, 0xD5,
-    0x63, 0xA0, 0x01, 0x84, 0x83, 0x07, 0x2E, 0x14,
-    0xD9, 0xB5, 0x51, 0x90, 0x9B, 0x2C, 0x7C, 0xA3,
-    0xA6, 0xB2, 0xEB, 0x73, 0xA5, 0x4C, 0xBE, 0x54,
-    0x16, 0x92, 0x0C, 0x74, 0xE3, 0x36, 0x61, 0x51,
-    0xC0, 0x38, 0x8C, 0xB0, 0x3A, 0xBD, 0xF5, 0x5A,
-    0x73, 0xFC, 0x2C, 0x60, 0x25, 0x62, 0x0B, 0x96,
-    0xBB, 0x6C, 0x4E, 0x42, 0x89, 0xF7, 0x6B, 0x10,
-    0x53, 0x7C, 0x6A, 0x28, 0xB4, 0x27, 0xF1, 0x8C,
-    0xE1, 0x13, 0xE6, 0x95, 0xBD, 0x9C, 0x45, 0xC7,
-    0xE2, 0x24, 0xF4, 0x46, 0xB6, 0x3B, 0x66, 0x70,
-    0xCC, 0xCA, 0x95, 0xE3, 0x03, 0x85, 0x56, 0xCB,
-    0xD4, 0x11, 0x1C, 0xD0, 0x1E, 0x93, 0xD7, 0xB8,
-    0xFB, 0xA6, 0xC3, 0x83, 0x8E, 0x20, 0xB5, 0xFF,
-    0xE9, 0x9F, 0xCF, 0x77, 0xBF, 0xC3, 0xBA, 0xCC,
-    0xEA, 0x03, 0x77, 0x6F, 0x39, 0x08, 0xAF, 0xBF,
-    0x33, 0x40, 0xC9, 0xE7, 0x62, 0x2B, 0x71, 0xE2,
-    0x81, 0x79, 0x79, 0x0C, 0x09, 0xAA, 0xAD, 0x82,
-    0x24, 0x41, 0xCD, 0x3A, 0xF9, 0xEA, 0xD8, 0xB9,
-    0xE5, 0xE4, 0xC5, 0x9A, 0xB9, 0xA4, 0x4D, 0x97,
-    0x44, 0x7E, 0x08, 0xDA, 0x86, 0x7A, 0xE7, 0x17,
-    0xA1, 0x66, 0x1D, 0x94, 0xAA, 0xA1, 0xED, 0x1D,
-    0x06, 0x3D, 0x70, 0xF0, 0xB2, 0xDE, 0xD2, 0xB3,
-    0x41, 0x0B, 0x7B, 0x72, 0xA0, 0xA7, 0x11, 0x1C,
-    0x31, 0xEF, 0xC2, 0xD1, 0x27, 0x53, 0x90, 0x3E,
-    0x20, 0x8F, 0xF6, 0x33, 0x60, 0x26, 0xFF, 0x5F,
-    0x96, 0xEC, 0x5C, 0x76, 0xB1, 0x2A, 0xAB, 0x49,
-    0x9E, 0x81, 0x9C, 0x88, 0x52, 0xEE, 0x1B, 0x21,
-    0x5F, 0xC4, 0x93, 0x1A, 0x0A, 0xEB, 0xEF, 0xD9,
-    0x91, 0xC5, 0x85, 0x39, 0x49, 0x99, 0xEE, 0xCD,
-    0x2D, 0xAD, 0x4F, 0x31, 0x8F, 0x8B, 0x3B, 0x01,
-    0x47, 0x18, 0x87, 0x23, 0x6D, 0xDD, 0x46, 0x1F,
-    0xD6, 0x4E, 0x3E, 0x2D, 0x69, 0xF9, 0x64, 0x48,
-    0x2A, 0x4F, 0xCE, 0xF2, 0xCB, 0x65, 0x2F, 0x8E,
-    0xFC, 0x78, 0x97, 0x5C, 0x05, 0x58, 0x7A, 0x19,
-    0xAC, 0x8D, 0x7F, 0xE5, 0xD5, 0x98, 0x1A, 0x57,
-    0x4B, 0x67, 0x0E, 0x7F, 0xA7, 0x05, 0x5A, 0x64,
-    0x28, 0xAF, 0x14, 0x63, 0x3F, 0xB6, 0x29, 0xFE,
-    0x88, 0xF5, 0x3C, 0xB7, 0x4C, 0x3C, 0x02, 0xA5,
-    0xB8, 0xCE, 0xDA, 0xE9, 0xB0, 0x68, 0x17, 0x44,
-    0x55, 0xE0, 0x1F, 0x4D, 0x8A, 0x43, 0x7D, 0x69,
-    0x57, 0x29, 0xC7, 0x2E, 0x8D, 0xAC, 0x74, 0x15,
-    0xB7, 0x59, 0xC4, 0xA8, 0x9F, 0x0A, 0x72, 0x9E,
-    0x7E, 0x6E, 0x15, 0x47, 0x22, 0xDF, 0x12, 0x34,
-    0x58, 0x35, 0x07, 0x6A, 0x99, 0xCF, 0x34, 0xDC,
-    0x6E, 0x22, 0x50, 0xC9, 0xDE, 0xC0, 0x68, 0x9B,
-    0x65, 0x89, 0xBC, 0xD4, 0xDB, 0xED, 0xF8, 0xAB,
-    0xC8, 0x12, 0xA8, 0xA2, 0x2B, 0x0D, 0x40, 0x52,
-    0xDC, 0xBB, 0xFE, 0x02, 0x32, 0x2F, 0xA4, 0xA9,
-    0xCA, 0xD7, 0x10, 0x61, 0x21, 0x1E, 0xF0, 0xB4,
-    0xD3, 0x50, 0x5D, 0x04, 0x0F, 0xF6, 0x00, 0xC2,
-    0x6F, 0x16, 0x9D, 0x25, 0x36, 0x86, 0x42, 0x56,
-    0x4A, 0x55, 0x5E, 0x09, 0xC1, 0xBE, 0xE0, 0x91
-};
-
-/* Macro to perform one column of the RS matrix multiplication.  The
- * parameters a, b, c, and d are the four bytes of output; i is the index
- * of the key bytes, and w, x, y, and z, are the column of constants from
- * the RS matrix, preprocessed through the poly_to_exp table. */
-
-#define CALC_S(a, b, c, d, i, w, x, y, z) \
-   if (key[i]) { \
-      tmp = poly_to_exp[key[i] - 1]; \
-      (a) ^= exp_to_poly[tmp + (w)]; \
-      (b) ^= exp_to_poly[tmp + (x)]; \
-      (c) ^= exp_to_poly[tmp + (y)]; \
-      (d) ^= exp_to_poly[tmp + (z)]; \
-   }
-
-/* Macros to calculate the key-dependent S-boxes for a 128-bit key using
- * the S vector from CALC_S.  CALC_SB_2 computes a single entry in all
- * four S-boxes, where i is the index of the entry to compute, and a and b
- * are the index numbers preprocessed through the q0 and q1 tables
- * respectively. */
-
-#define CALC_SB_2(i, a, b) \
-   ctx->s[0][i] = mds[0][q0[(a) ^ sa] ^ se]; \
-   ctx->s[1][i] = mds[1][q0[(b) ^ sb] ^ sf]; \
-   ctx->s[2][i] = mds[2][q1[(a) ^ sc] ^ sg]; \
-   ctx->s[3][i] = mds[3][q1[(b) ^ sd] ^ sh]
-
-/* Macro exactly like CALC_SB_2, but for 192-bit keys. */
-
-#define CALC_SB192_2(i, a, b) \
-   ctx->s[0][i] = mds[0][q0[q0[(b) ^ sa] ^ se] ^ si]; \
-   ctx->s[1][i] = mds[1][q0[q1[(b) ^ sb] ^ sf] ^ sj]; \
-   ctx->s[2][i] = mds[2][q1[q0[(a) ^ sc] ^ sg] ^ sk]; \
-   ctx->s[3][i] = mds[3][q1[q1[(a) ^ sd] ^ sh] ^ sl];
-
-/* Macro exactly like CALC_SB_2, but for 256-bit keys. */
-
-#define CALC_SB256_2(i, a, b) \
-   ctx->s[0][i] = mds[0][q0[q0[q1[(b) ^ sa] ^ se] ^ si] ^ sm]; \
-   ctx->s[1][i] = mds[1][q0[q1[q1[(a) ^ sb] ^ sf] ^ sj] ^ sn]; \
-   ctx->s[2][i] = mds[2][q1[q0[q0[(a) ^ sc] ^ sg] ^ sk] ^ so]; \
-   ctx->s[3][i] = mds[3][q1[q1[q0[(b) ^ sd] ^ sh] ^ sl] ^ sp];
-
-/* Macros to calculate the whitening and round subkeys.  CALC_K_2 computes the
- * last two stages of the h() function for a given index (either 2i or 2i+1).
- * a, b, c, and d are the four bytes going into the last two stages.  For
- * 128-bit keys, this is the entire h() function and a and c are the index
- * preprocessed through q0 and q1 respectively; for longer keys they are the
- * output of previous stages.  j is the index of the first key byte to use.
- * CALC_K computes a pair of subkeys for 128-bit Twofish, by calling CALC_K_2
- * twice, doing the Pseudo-Hadamard Transform, and doing the necessary
- * rotations.  Its parameters are: a, the array to write the results into,
- * j, the index of the first output entry, k and l, the preprocessed indices
- * for index 2i, and m and n, the preprocessed indices for index 2i+1.
- * CALC_K192_2 expands CALC_K_2 to handle 192-bit keys, by doing an
- * additional lookup-and-XOR stage.  The parameters a, b, c and d are the
- * four bytes going into the last three stages.  For 192-bit keys, c = d
- * are the index preprocessed through q0, and a = b are the index
- * preprocessed through q1; j is the index of the first key byte to use.
- * CALC_K192 is identical to CALC_K but for using the CALC_K192_2 macro
- * instead of CALC_K_2.
- * CALC_K256_2 expands CALC_K192_2 to handle 256-bit keys, by doing an
- * additional lookup-and-XOR stage.  The parameters a and b are the index
- * preprocessed through q0 and q1 respectively; j is the index of the first
- * key byte to use.  CALC_K256 is identical to CALC_K but for using the
- * CALC_K256_2 macro instead of CALC_K_2. */
-
-#define CALC_K_2(a, b, c, d, j) \
-     mds[0][q0[a ^ key[(j) + 8]] ^ key[j]] \
-   ^ mds[1][q0[b ^ key[(j) + 9]] ^ key[(j) + 1]] \
-   ^ mds[2][q1[c ^ key[(j) + 10]] ^ key[(j) + 2]] \
-   ^ mds[3][q1[d ^ key[(j) + 11]] ^ key[(j) + 3]]
-
-#define CALC_K(a, j, k, l, m, n) \
-   x = CALC_K_2 (k, l, k, l, 0); \
-   y = CALC_K_2 (m, n, m, n, 4); \
-   y = rol32(y, 8); \
-   x += y; y += x; ctx->a[j] = x; \
-   ctx->a[(j) + 1] = rol32(y, 9)
-
-#define CALC_K192_2(a, b, c, d, j) \
-   CALC_K_2 (q0[a ^ key[(j) + 16]], \
-	     q1[b ^ key[(j) + 17]], \
-	     q0[c ^ key[(j) + 18]], \
-	     q1[d ^ key[(j) + 19]], j)
-
-#define CALC_K192(a, j, k, l, m, n) \
-   x = CALC_K192_2 (l, l, k, k, 0); \
-   y = CALC_K192_2 (n, n, m, m, 4); \
-   y = rol32(y, 8); \
-   x += y; y += x; ctx->a[j] = x; \
-   ctx->a[(j) + 1] = rol32(y, 9)
-
-#define CALC_K256_2(a, b, j) \
-   CALC_K192_2 (q1[b ^ key[(j) + 24]], \
-	        q1[a ^ key[(j) + 25]], \
-	        q0[a ^ key[(j) + 26]], \
-	        q0[b ^ key[(j) + 27]], j)
-
-#define CALC_K256(a, j, k, l, m, n) \
-   x = CALC_K256_2 (k, l, 0); \
-   y = CALC_K256_2 (m, n, 4); \
-   y = rol32(y, 8); \
-   x += y; y += x; ctx->a[j] = x; \
-   ctx->a[(j) + 1] = rol32(y, 9)
-
-
 /* Macros to compute the g() function in the encryption and decryption
  * rounds.  G1 is the straight g() function; G2 includes the 8-bit
  * rotation for the high 32-bit word. */
@@ -630,176 +103,7 @@ static const u8 calc_sb_tbl[512] = {
    x ^= ctx->w[m]; \
    dst[n] = cpu_to_le32(x)
 
-#define TF_MIN_KEY_SIZE 16
-#define TF_MAX_KEY_SIZE 32
-#define TF_BLOCK_SIZE 16
-
-/* Structure for an expanded Twofish key.  s contains the key-dependent
- * S-boxes composed with the MDS matrix; w contains the eight "whitening"
- * subkeys, K[0] through K[7].	k holds the remaining, "round" subkeys.  Note
- * that k[i] corresponds to what the Twofish paper calls K[i+8]. */
-struct twofish_ctx {
-   u32 s[4][256], w[8], k[32];
-};
-
-/* Perform the key setup. */
-static int twofish_setkey(struct crypto_tfm *tfm, const u8 *key,
-			  unsigned int key_len, u32 *flags)
-{
-	
-	struct twofish_ctx *ctx = crypto_tfm_ctx(tfm);
 
-	int i, j, k;
-
-	/* Temporaries for CALC_K. */
-	u32 x, y;
-
-	/* The S vector used to key the S-boxes, split up into individual bytes.
-	 * 128-bit keys use only sa through sh; 256-bit use all of them. */
-	u8 sa = 0, sb = 0, sc = 0, sd = 0, se = 0, sf = 0, sg = 0, sh = 0;
-	u8 si = 0, sj = 0, sk = 0, sl = 0, sm = 0, sn = 0, so = 0, sp = 0;
-
-	/* Temporary for CALC_S. */
-	u8 tmp;
-
-	/* Check key length. */
-	if (key_len != 16 && key_len != 24 && key_len != 32)
-	{
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL; /* unsupported key length */
-	}
-
-	/* Compute the first two words of the S vector.  The magic numbers are
-	 * the entries of the RS matrix, preprocessed through poly_to_exp. The
-	 * numbers in the comments are the original (polynomial form) matrix
-	 * entries. */
-	CALC_S (sa, sb, sc, sd, 0, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
-	CALC_S (sa, sb, sc, sd, 1, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
-	CALC_S (sa, sb, sc, sd, 2, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
-	CALC_S (sa, sb, sc, sd, 3, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
-	CALC_S (sa, sb, sc, sd, 4, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
-	CALC_S (sa, sb, sc, sd, 5, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
-	CALC_S (sa, sb, sc, sd, 6, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
-	CALC_S (sa, sb, sc, sd, 7, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
-	CALC_S (se, sf, sg, sh, 8, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
-	CALC_S (se, sf, sg, sh, 9, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
-	CALC_S (se, sf, sg, sh, 10, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
-	CALC_S (se, sf, sg, sh, 11, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
-	CALC_S (se, sf, sg, sh, 12, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
-	CALC_S (se, sf, sg, sh, 13, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
-	CALC_S (se, sf, sg, sh, 14, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
-	CALC_S (se, sf, sg, sh, 15, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
-
-	if (key_len == 24 || key_len == 32) { /* 192- or 256-bit key */
-		/* Calculate the third word of the S vector */
-		CALC_S (si, sj, sk, sl, 16, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
-		CALC_S (si, sj, sk, sl, 17, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
-		CALC_S (si, sj, sk, sl, 18, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
-		CALC_S (si, sj, sk, sl, 19, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
-		CALC_S (si, sj, sk, sl, 20, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
-		CALC_S (si, sj, sk, sl, 21, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
-		CALC_S (si, sj, sk, sl, 22, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
-		CALC_S (si, sj, sk, sl, 23, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
-	}
-
-	if (key_len == 32) { /* 256-bit key */
-		/* Calculate the fourth word of the S vector */
-		CALC_S (sm, sn, so, sp, 24, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
-		CALC_S (sm, sn, so, sp, 25, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
-		CALC_S (sm, sn, so, sp, 26, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
-		CALC_S (sm, sn, so, sp, 27, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
-		CALC_S (sm, sn, so, sp, 28, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
-		CALC_S (sm, sn, so, sp, 29, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
-		CALC_S (sm, sn, so, sp, 30, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
-		CALC_S (sm, sn, so, sp, 31, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
-
-		/* Compute the S-boxes. */
-		for ( i = j = 0, k = 1; i < 256; i++, j += 2, k += 2 ) {
-			CALC_SB256_2( i, calc_sb_tbl[j], calc_sb_tbl[k] );
-		}
-
-		/* Calculate whitening and round subkeys.  The constants are
-		 * indices of subkeys, preprocessed through q0 and q1. */
-		CALC_K256 (w, 0, 0xA9, 0x75, 0x67, 0xF3);
-		CALC_K256 (w, 2, 0xB3, 0xC6, 0xE8, 0xF4);
-		CALC_K256 (w, 4, 0x04, 0xDB, 0xFD, 0x7B);
-		CALC_K256 (w, 6, 0xA3, 0xFB, 0x76, 0xC8);
-		CALC_K256 (k, 0, 0x9A, 0x4A, 0x92, 0xD3);
-		CALC_K256 (k, 2, 0x80, 0xE6, 0x78, 0x6B);
-		CALC_K256 (k, 4, 0xE4, 0x45, 0xDD, 0x7D);
-		CALC_K256 (k, 6, 0xD1, 0xE8, 0x38, 0x4B);
-		CALC_K256 (k, 8, 0x0D, 0xD6, 0xC6, 0x32);
-		CALC_K256 (k, 10, 0x35, 0xD8, 0x98, 0xFD);
-		CALC_K256 (k, 12, 0x18, 0x37, 0xF7, 0x71);
-		CALC_K256 (k, 14, 0xEC, 0xF1, 0x6C, 0xE1);
-		CALC_K256 (k, 16, 0x43, 0x30, 0x75, 0x0F);
-		CALC_K256 (k, 18, 0x37, 0xF8, 0x26, 0x1B);
-		CALC_K256 (k, 20, 0xFA, 0x87, 0x13, 0xFA);
-		CALC_K256 (k, 22, 0x94, 0x06, 0x48, 0x3F);
-		CALC_K256 (k, 24, 0xF2, 0x5E, 0xD0, 0xBA);
-		CALC_K256 (k, 26, 0x8B, 0xAE, 0x30, 0x5B);
-		CALC_K256 (k, 28, 0x84, 0x8A, 0x54, 0x00);
-		CALC_K256 (k, 30, 0xDF, 0xBC, 0x23, 0x9D);
-	} else if (key_len == 24) { /* 192-bit key */
-		/* Compute the S-boxes. */
-		for ( i = j = 0, k = 1; i < 256; i++, j += 2, k += 2 ) {
-		        CALC_SB192_2( i, calc_sb_tbl[j], calc_sb_tbl[k] );
-		}
-
-		/* Calculate whitening and round subkeys.  The constants are
-		 * indices of subkeys, preprocessed through q0 and q1. */
-		CALC_K192 (w, 0, 0xA9, 0x75, 0x67, 0xF3);
-		CALC_K192 (w, 2, 0xB3, 0xC6, 0xE8, 0xF4);
-		CALC_K192 (w, 4, 0x04, 0xDB, 0xFD, 0x7B);
-		CALC_K192 (w, 6, 0xA3, 0xFB, 0x76, 0xC8);
-		CALC_K192 (k, 0, 0x9A, 0x4A, 0x92, 0xD3);
-		CALC_K192 (k, 2, 0x80, 0xE6, 0x78, 0x6B);
-		CALC_K192 (k, 4, 0xE4, 0x45, 0xDD, 0x7D);
-		CALC_K192 (k, 6, 0xD1, 0xE8, 0x38, 0x4B);
-		CALC_K192 (k, 8, 0x0D, 0xD6, 0xC6, 0x32);
-		CALC_K192 (k, 10, 0x35, 0xD8, 0x98, 0xFD);
-		CALC_K192 (k, 12, 0x18, 0x37, 0xF7, 0x71);
-		CALC_K192 (k, 14, 0xEC, 0xF1, 0x6C, 0xE1);
-		CALC_K192 (k, 16, 0x43, 0x30, 0x75, 0x0F);
-		CALC_K192 (k, 18, 0x37, 0xF8, 0x26, 0x1B);
-		CALC_K192 (k, 20, 0xFA, 0x87, 0x13, 0xFA);
-		CALC_K192 (k, 22, 0x94, 0x06, 0x48, 0x3F);
-		CALC_K192 (k, 24, 0xF2, 0x5E, 0xD0, 0xBA);
-		CALC_K192 (k, 26, 0x8B, 0xAE, 0x30, 0x5B);
-		CALC_K192 (k, 28, 0x84, 0x8A, 0x54, 0x00);
-		CALC_K192 (k, 30, 0xDF, 0xBC, 0x23, 0x9D);
-	} else { /* 128-bit key */
-		/* Compute the S-boxes. */
-		for ( i = j = 0, k = 1; i < 256; i++, j += 2, k += 2 ) {
-			CALC_SB_2( i, calc_sb_tbl[j], calc_sb_tbl[k] );
-		}
-
-		/* Calculate whitening and round subkeys.  The constants are
-		 * indices of subkeys, preprocessed through q0 and q1. */
-		CALC_K (w, 0, 0xA9, 0x75, 0x67, 0xF3);
-		CALC_K (w, 2, 0xB3, 0xC6, 0xE8, 0xF4);
-		CALC_K (w, 4, 0x04, 0xDB, 0xFD, 0x7B);
-		CALC_K (w, 6, 0xA3, 0xFB, 0x76, 0xC8);
-		CALC_K (k, 0, 0x9A, 0x4A, 0x92, 0xD3);
-		CALC_K (k, 2, 0x80, 0xE6, 0x78, 0x6B);
-		CALC_K (k, 4, 0xE4, 0x45, 0xDD, 0x7D);
-		CALC_K (k, 6, 0xD1, 0xE8, 0x38, 0x4B);
-		CALC_K (k, 8, 0x0D, 0xD6, 0xC6, 0x32);
-		CALC_K (k, 10, 0x35, 0xD8, 0x98, 0xFD);
-		CALC_K (k, 12, 0x18, 0x37, 0xF7, 0x71);
-		CALC_K (k, 14, 0xEC, 0xF1, 0x6C, 0xE1);
-		CALC_K (k, 16, 0x43, 0x30, 0x75, 0x0F);
-		CALC_K (k, 18, 0x37, 0xF8, 0x26, 0x1B);
-		CALC_K (k, 20, 0xFA, 0x87, 0x13, 0xFA);
-		CALC_K (k, 22, 0x94, 0x06, 0x48, 0x3F);
-		CALC_K (k, 24, 0xF2, 0x5E, 0xD0, 0xBA);
-		CALC_K (k, 26, 0x8B, 0xAE, 0x30, 0x5B);
-		CALC_K (k, 28, 0x84, 0x8A, 0x54, 0x00);
-		CALC_K (k, 30, 0xDF, 0xBC, 0x23, 0x9D);
-	}
-
-	return 0;
-}
 
 /* Encrypt one block.  in and out may be the same. */
 static void twofish_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
diff --git a/crypto/twofish_common.c b/crypto/twofish_common.c
new file mode 100644
index 000000000000..1ae0280c2513
--- /dev/null
+++ b/crypto/twofish_common.c
@@ -0,0 +1,744 @@
+/*
+ * Common Twofish algorithm parts shared between the c and assembler
+ * implementations
+ *
+ * Originally Twofish for GPG
+ * By Matthew Skala <mskala@ansuz.sooke.bc.ca>, July 26, 1998
+ * 256-bit key length added March 20, 1999
+ * Some modifications to reduce the text size by Werner Koch, April, 1998
+ * Ported to the kerneli patch by Marc Mutz <Marc@Mutz.com>
+ * Ported to CryptoAPI by Colin Slater <hoho@tacomeat.net>
+ *
+ * The original author has disclaimed all copyright interest in this
+ * code and thus put it in the public domain. The subsequent authors
+ * have put this under the GNU General Public License.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307
+ * USA
+ *
+ * This code is a "clean room" implementation, written from the paper
+ * _Twofish: A 128-Bit Block Cipher_ by Bruce Schneier, John Kelsey,
+ * Doug Whiting, David Wagner, Chris Hall, and Niels Ferguson, available
+ * through http://www.counterpane.com/twofish.html
+ *
+ * For background information on multiplication in finite fields, used for
+ * the matrix operations in the key schedule, see the book _Contemporary
+ * Abstract Algebra_ by Joseph A. Gallian, especially chapter 22 in the
+ * Third Edition.
+ */
+
+#include <crypto/twofish.h>
+#include <linux/bitops.h>
+#include <linux/crypto.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/types.h>
+
+
+/* The large precomputed tables for the Twofish cipher (twofish.c)
+ * Taken from the same source as twofish.c
+ * Marc Mutz <Marc@Mutz.com>
+ */
+
+/* These two tables are the q0 and q1 permutations, exactly as described in
+ * the Twofish paper. */
+
+static const u8 q0[256] = {
+	0xA9, 0x67, 0xB3, 0xE8, 0x04, 0xFD, 0xA3, 0x76, 0x9A, 0x92, 0x80, 0x78,
+	0xE4, 0xDD, 0xD1, 0x38, 0x0D, 0xC6, 0x35, 0x98, 0x18, 0xF7, 0xEC, 0x6C,
+	0x43, 0x75, 0x37, 0x26, 0xFA, 0x13, 0x94, 0x48, 0xF2, 0xD0, 0x8B, 0x30,
+	0x84, 0x54, 0xDF, 0x23, 0x19, 0x5B, 0x3D, 0x59, 0xF3, 0xAE, 0xA2, 0x82,
+	0x63, 0x01, 0x83, 0x2E, 0xD9, 0x51, 0x9B, 0x7C, 0xA6, 0xEB, 0xA5, 0xBE,
+	0x16, 0x0C, 0xE3, 0x61, 0xC0, 0x8C, 0x3A, 0xF5, 0x73, 0x2C, 0x25, 0x0B,
+	0xBB, 0x4E, 0x89, 0x6B, 0x53, 0x6A, 0xB4, 0xF1, 0xE1, 0xE6, 0xBD, 0x45,
+	0xE2, 0xF4, 0xB6, 0x66, 0xCC, 0x95, 0x03, 0x56, 0xD4, 0x1C, 0x1E, 0xD7,
+	0xFB, 0xC3, 0x8E, 0xB5, 0xE9, 0xCF, 0xBF, 0xBA, 0xEA, 0x77, 0x39, 0xAF,
+	0x33, 0xC9, 0x62, 0x71, 0x81, 0x79, 0x09, 0xAD, 0x24, 0xCD, 0xF9, 0xD8,
+	0xE5, 0xC5, 0xB9, 0x4D, 0x44, 0x08, 0x86, 0xE7, 0xA1, 0x1D, 0xAA, 0xED,
+	0x06, 0x70, 0xB2, 0xD2, 0x41, 0x7B, 0xA0, 0x11, 0x31, 0xC2, 0x27, 0x90,
+	0x20, 0xF6, 0x60, 0xFF, 0x96, 0x5C, 0xB1, 0xAB, 0x9E, 0x9C, 0x52, 0x1B,
+	0x5F, 0x93, 0x0A, 0xEF, 0x91, 0x85, 0x49, 0xEE, 0x2D, 0x4F, 0x8F, 0x3B,
+	0x47, 0x87, 0x6D, 0x46, 0xD6, 0x3E, 0x69, 0x64, 0x2A, 0xCE, 0xCB, 0x2F,
+	0xFC, 0x97, 0x05, 0x7A, 0xAC, 0x7F, 0xD5, 0x1A, 0x4B, 0x0E, 0xA7, 0x5A,
+	0x28, 0x14, 0x3F, 0x29, 0x88, 0x3C, 0x4C, 0x02, 0xB8, 0xDA, 0xB0, 0x17,
+	0x55, 0x1F, 0x8A, 0x7D, 0x57, 0xC7, 0x8D, 0x74, 0xB7, 0xC4, 0x9F, 0x72,
+	0x7E, 0x15, 0x22, 0x12, 0x58, 0x07, 0x99, 0x34, 0x6E, 0x50, 0xDE, 0x68,
+	0x65, 0xBC, 0xDB, 0xF8, 0xC8, 0xA8, 0x2B, 0x40, 0xDC, 0xFE, 0x32, 0xA4,
+	0xCA, 0x10, 0x21, 0xF0, 0xD3, 0x5D, 0x0F, 0x00, 0x6F, 0x9D, 0x36, 0x42,
+	0x4A, 0x5E, 0xC1, 0xE0
+};
+
+static const u8 q1[256] = {
+	0x75, 0xF3, 0xC6, 0xF4, 0xDB, 0x7B, 0xFB, 0xC8, 0x4A, 0xD3, 0xE6, 0x6B,
+	0x45, 0x7D, 0xE8, 0x4B, 0xD6, 0x32, 0xD8, 0xFD, 0x37, 0x71, 0xF1, 0xE1,
+	0x30, 0x0F, 0xF8, 0x1B, 0x87, 0xFA, 0x06, 0x3F, 0x5E, 0xBA, 0xAE, 0x5B,
+	0x8A, 0x00, 0xBC, 0x9D, 0x6D, 0xC1, 0xB1, 0x0E, 0x80, 0x5D, 0xD2, 0xD5,
+	0xA0, 0x84, 0x07, 0x14, 0xB5, 0x90, 0x2C, 0xA3, 0xB2, 0x73, 0x4C, 0x54,
+	0x92, 0x74, 0x36, 0x51, 0x38, 0xB0, 0xBD, 0x5A, 0xFC, 0x60, 0x62, 0x96,
+	0x6C, 0x42, 0xF7, 0x10, 0x7C, 0x28, 0x27, 0x8C, 0x13, 0x95, 0x9C, 0xC7,
+	0x24, 0x46, 0x3B, 0x70, 0xCA, 0xE3, 0x85, 0xCB, 0x11, 0xD0, 0x93, 0xB8,
+	0xA6, 0x83, 0x20, 0xFF, 0x9F, 0x77, 0xC3, 0xCC, 0x03, 0x6F, 0x08, 0xBF,
+	0x40, 0xE7, 0x2B, 0xE2, 0x79, 0x0C, 0xAA, 0x82, 0x41, 0x3A, 0xEA, 0xB9,
+	0xE4, 0x9A, 0xA4, 0x97, 0x7E, 0xDA, 0x7A, 0x17, 0x66, 0x94, 0xA1, 0x1D,
+	0x3D, 0xF0, 0xDE, 0xB3, 0x0B, 0x72, 0xA7, 0x1C, 0xEF, 0xD1, 0x53, 0x3E,
+	0x8F, 0x33, 0x26, 0x5F, 0xEC, 0x76, 0x2A, 0x49, 0x81, 0x88, 0xEE, 0x21,
+	0xC4, 0x1A, 0xEB, 0xD9, 0xC5, 0x39, 0x99, 0xCD, 0xAD, 0x31, 0x8B, 0x01,
+	0x18, 0x23, 0xDD, 0x1F, 0x4E, 0x2D, 0xF9, 0x48, 0x4F, 0xF2, 0x65, 0x8E,
+	0x78, 0x5C, 0x58, 0x19, 0x8D, 0xE5, 0x98, 0x57, 0x67, 0x7F, 0x05, 0x64,
+	0xAF, 0x63, 0xB6, 0xFE, 0xF5, 0xB7, 0x3C, 0xA5, 0xCE, 0xE9, 0x68, 0x44,
+	0xE0, 0x4D, 0x43, 0x69, 0x29, 0x2E, 0xAC, 0x15, 0x59, 0xA8, 0x0A, 0x9E,
+	0x6E, 0x47, 0xDF, 0x34, 0x35, 0x6A, 0xCF, 0xDC, 0x22, 0xC9, 0xC0, 0x9B,
+	0x89, 0xD4, 0xED, 0xAB, 0x12, 0xA2, 0x0D, 0x52, 0xBB, 0x02, 0x2F, 0xA9,
+	0xD7, 0x61, 0x1E, 0xB4, 0x50, 0x04, 0xF6, 0xC2, 0x16, 0x25, 0x86, 0x56,
+	0x55, 0x09, 0xBE, 0x91
+};
+
+/* These MDS tables are actually tables of MDS composed with q0 and q1,
+ * because it is only ever used that way and we can save some time by
+ * precomputing.  Of course the main saving comes from precomputing the
+ * GF(2^8) multiplication involved in the MDS matrix multiply; by looking
+ * things up in these tables we reduce the matrix multiply to four lookups
+ * and three XORs.  Semi-formally, the definition of these tables is:
+ * mds[0][i] = MDS (q1[i] 0 0 0)^T  mds[1][i] = MDS (0 q0[i] 0 0)^T
+ * mds[2][i] = MDS (0 0 q1[i] 0)^T  mds[3][i] = MDS (0 0 0 q0[i])^T
+ * where ^T means "transpose", the matrix multiply is performed in GF(2^8)
+ * represented as GF(2)[x]/v(x) where v(x)=x^8+x^6+x^5+x^3+1 as described
+ * by Schneier et al, and I'm casually glossing over the byte/word
+ * conversion issues. */
+
+static const u32 mds[4][256] = {
+	{
+	0xBCBC3275, 0xECEC21F3, 0x202043C6, 0xB3B3C9F4, 0xDADA03DB, 0x02028B7B,
+	0xE2E22BFB, 0x9E9EFAC8, 0xC9C9EC4A, 0xD4D409D3, 0x18186BE6, 0x1E1E9F6B,
+	0x98980E45, 0xB2B2387D, 0xA6A6D2E8, 0x2626B74B, 0x3C3C57D6, 0x93938A32,
+	0x8282EED8, 0x525298FD, 0x7B7BD437, 0xBBBB3771, 0x5B5B97F1, 0x474783E1,
+	0x24243C30, 0x5151E20F, 0xBABAC6F8, 0x4A4AF31B, 0xBFBF4887, 0x0D0D70FA,
+	0xB0B0B306, 0x7575DE3F, 0xD2D2FD5E, 0x7D7D20BA, 0x666631AE, 0x3A3AA35B,
+	0x59591C8A, 0x00000000, 0xCDCD93BC, 0x1A1AE09D, 0xAEAE2C6D, 0x7F7FABC1,
+	0x2B2BC7B1, 0xBEBEB90E, 0xE0E0A080, 0x8A8A105D, 0x3B3B52D2, 0x6464BAD5,
+	0xD8D888A0, 0xE7E7A584, 0x5F5FE807, 0x1B1B1114, 0x2C2CC2B5, 0xFCFCB490,
+	0x3131272C, 0x808065A3, 0x73732AB2, 0x0C0C8173, 0x79795F4C, 0x6B6B4154,
+	0x4B4B0292, 0x53536974, 0x94948F36, 0x83831F51, 0x2A2A3638, 0xC4C49CB0,
+	0x2222C8BD, 0xD5D5F85A, 0xBDBDC3FC, 0x48487860, 0xFFFFCE62, 0x4C4C0796,
+	0x4141776C, 0xC7C7E642, 0xEBEB24F7, 0x1C1C1410, 0x5D5D637C, 0x36362228,
+	0x6767C027, 0xE9E9AF8C, 0x4444F913, 0x1414EA95, 0xF5F5BB9C, 0xCFCF18C7,
+	0x3F3F2D24, 0xC0C0E346, 0x7272DB3B, 0x54546C70, 0x29294CCA, 0xF0F035E3,
+	0x0808FE85, 0xC6C617CB, 0xF3F34F11, 0x8C8CE4D0, 0xA4A45993, 0xCACA96B8,
+	0x68683BA6, 0xB8B84D83, 0x38382820, 0xE5E52EFF, 0xADAD569F, 0x0B0B8477,
+	0xC8C81DC3, 0x9999FFCC, 0x5858ED03, 0x19199A6F, 0x0E0E0A08, 0x95957EBF,
+	0x70705040, 0xF7F730E7, 0x6E6ECF2B, 0x1F1F6EE2, 0xB5B53D79, 0x09090F0C,
+	0x616134AA, 0x57571682, 0x9F9F0B41, 0x9D9D803A, 0x111164EA, 0x2525CDB9,
+	0xAFAFDDE4, 0x4545089A, 0xDFDF8DA4, 0xA3A35C97, 0xEAEAD57E, 0x353558DA,
+	0xEDEDD07A, 0x4343FC17, 0xF8F8CB66, 0xFBFBB194, 0x3737D3A1, 0xFAFA401D,
+	0xC2C2683D, 0xB4B4CCF0, 0x32325DDE, 0x9C9C71B3, 0x5656E70B, 0xE3E3DA72,
+	0x878760A7, 0x15151B1C, 0xF9F93AEF, 0x6363BFD1, 0x3434A953, 0x9A9A853E,
+	0xB1B1428F, 0x7C7CD133, 0x88889B26, 0x3D3DA65F, 0xA1A1D7EC, 0xE4E4DF76,
+	0x8181942A, 0x91910149, 0x0F0FFB81, 0xEEEEAA88, 0x161661EE, 0xD7D77321,
+	0x9797F5C4, 0xA5A5A81A, 0xFEFE3FEB, 0x6D6DB5D9, 0x7878AEC5, 0xC5C56D39,
+	0x1D1DE599, 0x7676A4CD, 0x3E3EDCAD, 0xCBCB6731, 0xB6B6478B, 0xEFEF5B01,
+	0x12121E18, 0x6060C523, 0x6A6AB0DD, 0x4D4DF61F, 0xCECEE94E, 0xDEDE7C2D,
+	0x55559DF9, 0x7E7E5A48, 0x2121B24F, 0x03037AF2, 0xA0A02665, 0x5E5E198E,
+	0x5A5A6678, 0x65654B5C, 0x62624E58, 0xFDFD4519, 0x0606F48D, 0x404086E5,
+	0xF2F2BE98, 0x3333AC57, 0x17179067, 0x05058E7F, 0xE8E85E05, 0x4F4F7D64,
+	0x89896AAF, 0x10109563, 0x74742FB6, 0x0A0A75FE, 0x5C5C92F5, 0x9B9B74B7,
+	0x2D2D333C, 0x3030D6A5, 0x2E2E49CE, 0x494989E9, 0x46467268, 0x77775544,
+	0xA8A8D8E0, 0x9696044D, 0x2828BD43, 0xA9A92969, 0xD9D97929, 0x8686912E,
+	0xD1D187AC, 0xF4F44A15, 0x8D8D1559, 0xD6D682A8, 0xB9B9BC0A, 0x42420D9E,
+	0xF6F6C16E, 0x2F2FB847, 0xDDDD06DF, 0x23233934, 0xCCCC6235, 0xF1F1C46A,
+	0xC1C112CF, 0x8585EBDC, 0x8F8F9E22, 0x7171A1C9, 0x9090F0C0, 0xAAAA539B,
+	0x0101F189, 0x8B8BE1D4, 0x4E4E8CED, 0x8E8E6FAB, 0xABABA212, 0x6F6F3EA2,
+	0xE6E6540D, 0xDBDBF252, 0x92927BBB, 0xB7B7B602, 0x6969CA2F, 0x3939D9A9,
+	0xD3D30CD7, 0xA7A72361, 0xA2A2AD1E, 0xC3C399B4, 0x6C6C4450, 0x07070504,
+	0x04047FF6, 0x272746C2, 0xACACA716, 0xD0D07625, 0x50501386, 0xDCDCF756,
+	0x84841A55, 0xE1E15109, 0x7A7A25BE, 0x1313EF91},
+
+	{
+	0xA9D93939, 0x67901717, 0xB3719C9C, 0xE8D2A6A6, 0x04050707, 0xFD985252,
+	0xA3658080, 0x76DFE4E4, 0x9A084545, 0x92024B4B, 0x80A0E0E0, 0x78665A5A,
+	0xE4DDAFAF, 0xDDB06A6A, 0xD1BF6363, 0x38362A2A, 0x0D54E6E6, 0xC6432020,
+	0x3562CCCC, 0x98BEF2F2, 0x181E1212, 0xF724EBEB, 0xECD7A1A1, 0x6C774141,
+	0x43BD2828, 0x7532BCBC, 0x37D47B7B, 0x269B8888, 0xFA700D0D, 0x13F94444,
+	0x94B1FBFB, 0x485A7E7E, 0xF27A0303, 0xD0E48C8C, 0x8B47B6B6, 0x303C2424,
+	0x84A5E7E7, 0x54416B6B, 0xDF06DDDD, 0x23C56060, 0x1945FDFD, 0x5BA33A3A,
+	0x3D68C2C2, 0x59158D8D, 0xF321ECEC, 0xAE316666, 0xA23E6F6F, 0x82165757,
+	0x63951010, 0x015BEFEF, 0x834DB8B8, 0x2E918686, 0xD9B56D6D, 0x511F8383,
+	0x9B53AAAA, 0x7C635D5D, 0xA63B6868, 0xEB3FFEFE, 0xA5D63030, 0xBE257A7A,
+	0x16A7ACAC, 0x0C0F0909, 0xE335F0F0, 0x6123A7A7, 0xC0F09090, 0x8CAFE9E9,
+	0x3A809D9D, 0xF5925C5C, 0x73810C0C, 0x2C273131, 0x2576D0D0, 0x0BE75656,
+	0xBB7B9292, 0x4EE9CECE, 0x89F10101, 0x6B9F1E1E, 0x53A93434, 0x6AC4F1F1,
+	0xB499C3C3, 0xF1975B5B, 0xE1834747, 0xE66B1818, 0xBDC82222, 0x450E9898,
+	0xE26E1F1F, 0xF4C9B3B3, 0xB62F7474, 0x66CBF8F8, 0xCCFF9999, 0x95EA1414,
+	0x03ED5858, 0x56F7DCDC, 0xD4E18B8B, 0x1C1B1515, 0x1EADA2A2, 0xD70CD3D3,
+	0xFB2BE2E2, 0xC31DC8C8, 0x8E195E5E, 0xB5C22C2C, 0xE9894949, 0xCF12C1C1,
+	0xBF7E9595, 0xBA207D7D, 0xEA641111, 0x77840B0B, 0x396DC5C5, 0xAF6A8989,
+	0x33D17C7C, 0xC9A17171, 0x62CEFFFF, 0x7137BBBB, 0x81FB0F0F, 0x793DB5B5,
+	0x0951E1E1, 0xADDC3E3E, 0x242D3F3F, 0xCDA47676, 0xF99D5555, 0xD8EE8282,
+	0xE5864040, 0xC5AE7878, 0xB9CD2525, 0x4D049696, 0x44557777, 0x080A0E0E,
+	0x86135050, 0xE730F7F7, 0xA1D33737, 0x1D40FAFA, 0xAA346161, 0xED8C4E4E,
+	0x06B3B0B0, 0x706C5454, 0xB22A7373, 0xD2523B3B, 0x410B9F9F, 0x7B8B0202,
+	0xA088D8D8, 0x114FF3F3, 0x3167CBCB, 0xC2462727, 0x27C06767, 0x90B4FCFC,
+	0x20283838, 0xF67F0404, 0x60784848, 0xFF2EE5E5, 0x96074C4C, 0x5C4B6565,
+	0xB1C72B2B, 0xAB6F8E8E, 0x9E0D4242, 0x9CBBF5F5, 0x52F2DBDB, 0x1BF34A4A,
+	0x5FA63D3D, 0x9359A4A4, 0x0ABCB9B9, 0xEF3AF9F9, 0x91EF1313, 0x85FE0808,
+	0x49019191, 0xEE611616, 0x2D7CDEDE, 0x4FB22121, 0x8F42B1B1, 0x3BDB7272,
+	0x47B82F2F, 0x8748BFBF, 0x6D2CAEAE, 0x46E3C0C0, 0xD6573C3C, 0x3E859A9A,
+	0x6929A9A9, 0x647D4F4F, 0x2A948181, 0xCE492E2E, 0xCB17C6C6, 0x2FCA6969,
+	0xFCC3BDBD, 0x975CA3A3, 0x055EE8E8, 0x7AD0EDED, 0xAC87D1D1, 0x7F8E0505,
+	0xD5BA6464, 0x1AA8A5A5, 0x4BB72626, 0x0EB9BEBE, 0xA7608787, 0x5AF8D5D5,
+	0x28223636, 0x14111B1B, 0x3FDE7575, 0x2979D9D9, 0x88AAEEEE, 0x3C332D2D,
+	0x4C5F7979, 0x02B6B7B7, 0xB896CACA, 0xDA583535, 0xB09CC4C4, 0x17FC4343,
+	0x551A8484, 0x1FF64D4D, 0x8A1C5959, 0x7D38B2B2, 0x57AC3333, 0xC718CFCF,
+	0x8DF40606, 0x74695353, 0xB7749B9B, 0xC4F59797, 0x9F56ADAD, 0x72DAE3E3,
+	0x7ED5EAEA, 0x154AF4F4, 0x229E8F8F, 0x12A2ABAB, 0x584E6262, 0x07E85F5F,
+	0x99E51D1D, 0x34392323, 0x6EC1F6F6, 0x50446C6C, 0xDE5D3232, 0x68724646,
+	0x6526A0A0, 0xBC93CDCD, 0xDB03DADA, 0xF8C6BABA, 0xC8FA9E9E, 0xA882D6D6,
+	0x2BCF6E6E, 0x40507070, 0xDCEB8585, 0xFE750A0A, 0x328A9393, 0xA48DDFDF,
+	0xCA4C2929, 0x10141C1C, 0x2173D7D7, 0xF0CCB4B4, 0xD309D4D4, 0x5D108A8A,
+	0x0FE25151, 0x00000000, 0x6F9A1919, 0x9DE01A1A, 0x368F9494, 0x42E6C7C7,
+	0x4AECC9C9, 0x5EFDD2D2, 0xC1AB7F7F, 0xE0D8A8A8},
+
+	{
+	0xBC75BC32, 0xECF3EC21, 0x20C62043, 0xB3F4B3C9, 0xDADBDA03, 0x027B028B,
+	0xE2FBE22B, 0x9EC89EFA, 0xC94AC9EC, 0xD4D3D409, 0x18E6186B, 0x1E6B1E9F,
+	0x9845980E, 0xB27DB238, 0xA6E8A6D2, 0x264B26B7, 0x3CD63C57, 0x9332938A,
+	0x82D882EE, 0x52FD5298, 0x7B377BD4, 0xBB71BB37, 0x5BF15B97, 0x47E14783,
+	0x2430243C, 0x510F51E2, 0xBAF8BAC6, 0x4A1B4AF3, 0xBF87BF48, 0x0DFA0D70,
+	0xB006B0B3, 0x753F75DE, 0xD25ED2FD, 0x7DBA7D20, 0x66AE6631, 0x3A5B3AA3,
+	0x598A591C, 0x00000000, 0xCDBCCD93, 0x1A9D1AE0, 0xAE6DAE2C, 0x7FC17FAB,
+	0x2BB12BC7, 0xBE0EBEB9, 0xE080E0A0, 0x8A5D8A10, 0x3BD23B52, 0x64D564BA,
+	0xD8A0D888, 0xE784E7A5, 0x5F075FE8, 0x1B141B11, 0x2CB52CC2, 0xFC90FCB4,
+	0x312C3127, 0x80A38065, 0x73B2732A, 0x0C730C81, 0x794C795F, 0x6B546B41,
+	0x4B924B02, 0x53745369, 0x9436948F, 0x8351831F, 0x2A382A36, 0xC4B0C49C,
+	0x22BD22C8, 0xD55AD5F8, 0xBDFCBDC3, 0x48604878, 0xFF62FFCE, 0x4C964C07,
+	0x416C4177, 0xC742C7E6, 0xEBF7EB24, 0x1C101C14, 0x5D7C5D63, 0x36283622,
+	0x672767C0, 0xE98CE9AF, 0x441344F9, 0x149514EA, 0xF59CF5BB, 0xCFC7CF18,
+	0x3F243F2D, 0xC046C0E3, 0x723B72DB, 0x5470546C, 0x29CA294C, 0xF0E3F035,
+	0x088508FE, 0xC6CBC617, 0xF311F34F, 0x8CD08CE4, 0xA493A459, 0xCAB8CA96,
+	0x68A6683B, 0xB883B84D, 0x38203828, 0xE5FFE52E, 0xAD9FAD56, 0x0B770B84,
+	0xC8C3C81D, 0x99CC99FF, 0x580358ED, 0x196F199A, 0x0E080E0A, 0x95BF957E,
+	0x70407050, 0xF7E7F730, 0x6E2B6ECF, 0x1FE21F6E, 0xB579B53D, 0x090C090F,
+	0x61AA6134, 0x57825716, 0x9F419F0B, 0x9D3A9D80, 0x11EA1164, 0x25B925CD,
+	0xAFE4AFDD, 0x459A4508, 0xDFA4DF8D, 0xA397A35C, 0xEA7EEAD5, 0x35DA3558,
+	0xED7AEDD0, 0x431743FC, 0xF866F8CB, 0xFB94FBB1, 0x37A137D3, 0xFA1DFA40,
+	0xC23DC268, 0xB4F0B4CC, 0x32DE325D, 0x9CB39C71, 0x560B56E7, 0xE372E3DA,
+	0x87A78760, 0x151C151B, 0xF9EFF93A, 0x63D163BF, 0x345334A9, 0x9A3E9A85,
+	0xB18FB142, 0x7C337CD1, 0x8826889B, 0x3D5F3DA6, 0xA1ECA1D7, 0xE476E4DF,
+	0x812A8194, 0x91499101, 0x0F810FFB, 0xEE88EEAA, 0x16EE1661, 0xD721D773,
+	0x97C497F5, 0xA51AA5A8, 0xFEEBFE3F, 0x6DD96DB5, 0x78C578AE, 0xC539C56D,
+	0x1D991DE5, 0x76CD76A4, 0x3EAD3EDC, 0xCB31CB67, 0xB68BB647, 0xEF01EF5B,
+	0x1218121E, 0x602360C5, 0x6ADD6AB0, 0x4D1F4DF6, 0xCE4ECEE9, 0xDE2DDE7C,
+	0x55F9559D, 0x7E487E5A, 0x214F21B2, 0x03F2037A, 0xA065A026, 0x5E8E5E19,
+	0x5A785A66, 0x655C654B, 0x6258624E, 0xFD19FD45, 0x068D06F4, 0x40E54086,
+	0xF298F2BE, 0x335733AC, 0x17671790, 0x057F058E, 0xE805E85E, 0x4F644F7D,
+	0x89AF896A, 0x10631095, 0x74B6742F, 0x0AFE0A75, 0x5CF55C92, 0x9BB79B74,
+	0x2D3C2D33, 0x30A530D6, 0x2ECE2E49, 0x49E94989, 0x46684672, 0x77447755,
+	0xA8E0A8D8, 0x964D9604, 0x284328BD, 0xA969A929, 0xD929D979, 0x862E8691,
+	0xD1ACD187, 0xF415F44A, 0x8D598D15, 0xD6A8D682, 0xB90AB9BC, 0x429E420D,
+	0xF66EF6C1, 0x2F472FB8, 0xDDDFDD06, 0x23342339, 0xCC35CC62, 0xF16AF1C4,
+	0xC1CFC112, 0x85DC85EB, 0x8F228F9E, 0x71C971A1, 0x90C090F0, 0xAA9BAA53,
+	0x018901F1, 0x8BD48BE1, 0x4EED4E8C, 0x8EAB8E6F, 0xAB12ABA2, 0x6FA26F3E,
+	0xE60DE654, 0xDB52DBF2, 0x92BB927B, 0xB702B7B6, 0x692F69CA, 0x39A939D9,
+	0xD3D7D30C, 0xA761A723, 0xA21EA2AD, 0xC3B4C399, 0x6C506C44, 0x07040705,
+	0x04F6047F, 0x27C22746, 0xAC16ACA7, 0xD025D076, 0x50865013, 0xDC56DCF7,
+	0x8455841A, 0xE109E151, 0x7ABE7A25, 0x139113EF},
+
+	{
+	0xD939A9D9, 0x90176790, 0x719CB371, 0xD2A6E8D2, 0x05070405, 0x9852FD98,
+	0x6580A365, 0xDFE476DF, 0x08459A08, 0x024B9202, 0xA0E080A0, 0x665A7866,
+	0xDDAFE4DD, 0xB06ADDB0, 0xBF63D1BF, 0x362A3836, 0x54E60D54, 0x4320C643,
+	0x62CC3562, 0xBEF298BE, 0x1E12181E, 0x24EBF724, 0xD7A1ECD7, 0x77416C77,
+	0xBD2843BD, 0x32BC7532, 0xD47B37D4, 0x9B88269B, 0x700DFA70, 0xF94413F9,
+	0xB1FB94B1, 0x5A7E485A, 0x7A03F27A, 0xE48CD0E4, 0x47B68B47, 0x3C24303C,
+	0xA5E784A5, 0x416B5441, 0x06DDDF06, 0xC56023C5, 0x45FD1945, 0xA33A5BA3,
+	0x68C23D68, 0x158D5915, 0x21ECF321, 0x3166AE31, 0x3E6FA23E, 0x16578216,
+	0x95106395, 0x5BEF015B, 0x4DB8834D, 0x91862E91, 0xB56DD9B5, 0x1F83511F,
+	0x53AA9B53, 0x635D7C63, 0x3B68A63B, 0x3FFEEB3F, 0xD630A5D6, 0x257ABE25,
+	0xA7AC16A7, 0x0F090C0F, 0x35F0E335, 0x23A76123, 0xF090C0F0, 0xAFE98CAF,
+	0x809D3A80, 0x925CF592, 0x810C7381, 0x27312C27, 0x76D02576, 0xE7560BE7,
+	0x7B92BB7B, 0xE9CE4EE9, 0xF10189F1, 0x9F1E6B9F, 0xA93453A9, 0xC4F16AC4,
+	0x99C3B499, 0x975BF197, 0x8347E183, 0x6B18E66B, 0xC822BDC8, 0x0E98450E,
+	0x6E1FE26E, 0xC9B3F4C9, 0x2F74B62F, 0xCBF866CB, 0xFF99CCFF, 0xEA1495EA,
+	0xED5803ED, 0xF7DC56F7, 0xE18BD4E1, 0x1B151C1B, 0xADA21EAD, 0x0CD3D70C,
+	0x2BE2FB2B, 0x1DC8C31D, 0x195E8E19, 0xC22CB5C2, 0x8949E989, 0x12C1CF12,
+	0x7E95BF7E, 0x207DBA20, 0x6411EA64, 0x840B7784, 0x6DC5396D, 0x6A89AF6A,
+	0xD17C33D1, 0xA171C9A1, 0xCEFF62CE, 0x37BB7137, 0xFB0F81FB, 0x3DB5793D,
+	0x51E10951, 0xDC3EADDC, 0x2D3F242D, 0xA476CDA4, 0x9D55F99D, 0xEE82D8EE,
+	0x8640E586, 0xAE78C5AE, 0xCD25B9CD, 0x04964D04, 0x55774455, 0x0A0E080A,
+	0x13508613, 0x30F7E730, 0xD337A1D3, 0x40FA1D40, 0x3461AA34, 0x8C4EED8C,
+	0xB3B006B3, 0x6C54706C, 0x2A73B22A, 0x523BD252, 0x0B9F410B, 0x8B027B8B,
+	0x88D8A088, 0x4FF3114F, 0x67CB3167, 0x4627C246, 0xC06727C0, 0xB4FC90B4,
+	0x28382028, 0x7F04F67F, 0x78486078, 0x2EE5FF2E, 0x074C9607, 0x4B655C4B,
+	0xC72BB1C7, 0x6F8EAB6F, 0x0D429E0D, 0xBBF59CBB, 0xF2DB52F2, 0xF34A1BF3,
+	0xA63D5FA6, 0x59A49359, 0xBCB90ABC, 0x3AF9EF3A, 0xEF1391EF, 0xFE0885FE,
+	0x01914901, 0x6116EE61, 0x7CDE2D7C, 0xB2214FB2, 0x42B18F42, 0xDB723BDB,
+	0xB82F47B8, 0x48BF8748, 0x2CAE6D2C, 0xE3C046E3, 0x573CD657, 0x859A3E85,
+	0x29A96929, 0x7D4F647D, 0x94812A94, 0x492ECE49, 0x17C6CB17, 0xCA692FCA,
+	0xC3BDFCC3, 0x5CA3975C, 0x5EE8055E, 0xD0ED7AD0, 0x87D1AC87, 0x8E057F8E,
+	0xBA64D5BA, 0xA8A51AA8, 0xB7264BB7, 0xB9BE0EB9, 0x6087A760, 0xF8D55AF8,
+	0x22362822, 0x111B1411, 0xDE753FDE, 0x79D92979, 0xAAEE88AA, 0x332D3C33,
+	0x5F794C5F, 0xB6B702B6, 0x96CAB896, 0x5835DA58, 0x9CC4B09C, 0xFC4317FC,
+	0x1A84551A, 0xF64D1FF6, 0x1C598A1C, 0x38B27D38, 0xAC3357AC, 0x18CFC718,
+	0xF4068DF4, 0x69537469, 0x749BB774, 0xF597C4F5, 0x56AD9F56, 0xDAE372DA,
+	0xD5EA7ED5, 0x4AF4154A, 0x9E8F229E, 0xA2AB12A2, 0x4E62584E, 0xE85F07E8,
+	0xE51D99E5, 0x39233439, 0xC1F66EC1, 0x446C5044, 0x5D32DE5D, 0x72466872,
+	0x26A06526, 0x93CDBC93, 0x03DADB03, 0xC6BAF8C6, 0xFA9EC8FA, 0x82D6A882,
+	0xCF6E2BCF, 0x50704050, 0xEB85DCEB, 0x750AFE75, 0x8A93328A, 0x8DDFA48D,
+	0x4C29CA4C, 0x141C1014, 0x73D72173, 0xCCB4F0CC, 0x09D4D309, 0x108A5D10,
+	0xE2510FE2, 0x00000000, 0x9A196F9A, 0xE01A9DE0, 0x8F94368F, 0xE6C742E6,
+	0xECC94AEC, 0xFDD25EFD, 0xAB7FC1AB, 0xD8A8E0D8}
+};
+
+/* The exp_to_poly and poly_to_exp tables are used to perform efficient
+ * operations in GF(2^8) represented as GF(2)[x]/w(x) where
+ * w(x)=x^8+x^6+x^3+x^2+1.  We care about doing that because it's part of the
+ * definition of the RS matrix in the key schedule.  Elements of that field
+ * are polynomials of degree not greater than 7 and all coefficients 0 or 1,
+ * which can be represented naturally by bytes (just substitute x=2).  In that
+ * form, GF(2^8) addition is the same as bitwise XOR, but GF(2^8)
+ * multiplication is inefficient without hardware support.  To multiply
+ * faster, I make use of the fact x is a generator for the nonzero elements,
+ * so that every element p of GF(2)[x]/w(x) is either 0 or equal to (x)^n for
+ * some n in 0..254.  Note that that caret is exponentiation in GF(2^8),
+ * *not* polynomial notation.  So if I want to compute pq where p and q are
+ * in GF(2^8), I can just say:
+ *    1. if p=0 or q=0 then pq=0
+ *    2. otherwise, find m and n such that p=x^m and q=x^n
+ *    3. pq=(x^m)(x^n)=x^(m+n), so add m and n and find pq
+ * The translations in steps 2 and 3 are looked up in the tables
+ * poly_to_exp (for step 2) and exp_to_poly (for step 3).  To see this
+ * in action, look at the CALC_S macro.  As additional wrinkles, note that
+ * one of my operands is always a constant, so the poly_to_exp lookup on it
+ * is done in advance; I included the original values in the comments so
+ * readers can have some chance of recognizing that this *is* the RS matrix
+ * from the Twofish paper.  I've only included the table entries I actually
+ * need; I never do a lookup on a variable input of zero and the biggest
+ * exponents I'll ever see are 254 (variable) and 237 (constant), so they'll
+ * never sum to more than 491.	I'm repeating part of the exp_to_poly table
+ * so that I don't have to do mod-255 reduction in the exponent arithmetic.
+ * Since I know my constant operands are never zero, I only have to worry
+ * about zero values in the variable operand, and I do it with a simple
+ * conditional branch.	I know conditionals are expensive, but I couldn't
+ * see a non-horrible way of avoiding them, and I did manage to group the
+ * statements so that each if covers four group multiplications. */
+
+static const u8 poly_to_exp[255] = {
+	0x00, 0x01, 0x17, 0x02, 0x2E, 0x18, 0x53, 0x03, 0x6A, 0x2F, 0x93, 0x19,
+	0x34, 0x54, 0x45, 0x04, 0x5C, 0x6B, 0xB6, 0x30, 0xA6, 0x94, 0x4B, 0x1A,
+	0x8C, 0x35, 0x81, 0x55, 0xAA, 0x46, 0x0D, 0x05, 0x24, 0x5D, 0x87, 0x6C,
+	0x9B, 0xB7, 0xC1, 0x31, 0x2B, 0xA7, 0xA3, 0x95, 0x98, 0x4C, 0xCA, 0x1B,
+	0xE6, 0x8D, 0x73, 0x36, 0xCD, 0x82, 0x12, 0x56, 0x62, 0xAB, 0xF0, 0x47,
+	0x4F, 0x0E, 0xBD, 0x06, 0xD4, 0x25, 0xD2, 0x5E, 0x27, 0x88, 0x66, 0x6D,
+	0xD6, 0x9C, 0x79, 0xB8, 0x08, 0xC2, 0xDF, 0x32, 0x68, 0x2C, 0xFD, 0xA8,
+	0x8A, 0xA4, 0x5A, 0x96, 0x29, 0x99, 0x22, 0x4D, 0x60, 0xCB, 0xE4, 0x1C,
+	0x7B, 0xE7, 0x3B, 0x8E, 0x9E, 0x74, 0xF4, 0x37, 0xD8, 0xCE, 0xF9, 0x83,
+	0x6F, 0x13, 0xB2, 0x57, 0xE1, 0x63, 0xDC, 0xAC, 0xC4, 0xF1, 0xAF, 0x48,
+	0x0A, 0x50, 0x42, 0x0F, 0xBA, 0xBE, 0xC7, 0x07, 0xDE, 0xD5, 0x78, 0x26,
+	0x65, 0xD3, 0xD1, 0x5F, 0xE3, 0x28, 0x21, 0x89, 0x59, 0x67, 0xFC, 0x6E,
+	0xB1, 0xD7, 0xF8, 0x9D, 0xF3, 0x7A, 0x3A, 0xB9, 0xC6, 0x09, 0x41, 0xC3,
+	0xAE, 0xE0, 0xDB, 0x33, 0x44, 0x69, 0x92, 0x2D, 0x52, 0xFE, 0x16, 0xA9,
+	0x0C, 0x8B, 0x80, 0xA5, 0x4A, 0x5B, 0xB5, 0x97, 0xC9, 0x2A, 0xA2, 0x9A,
+	0xC0, 0x23, 0x86, 0x4E, 0xBC, 0x61, 0xEF, 0xCC, 0x11, 0xE5, 0x72, 0x1D,
+	0x3D, 0x7C, 0xEB, 0xE8, 0xE9, 0x3C, 0xEA, 0x8F, 0x7D, 0x9F, 0xEC, 0x75,
+	0x1E, 0xF5, 0x3E, 0x38, 0xF6, 0xD9, 0x3F, 0xCF, 0x76, 0xFA, 0x1F, 0x84,
+	0xA0, 0x70, 0xED, 0x14, 0x90, 0xB3, 0x7E, 0x58, 0xFB, 0xE2, 0x20, 0x64,
+	0xD0, 0xDD, 0x77, 0xAD, 0xDA, 0xC5, 0x40, 0xF2, 0x39, 0xB0, 0xF7, 0x49,
+	0xB4, 0x0B, 0x7F, 0x51, 0x15, 0x43, 0x91, 0x10, 0x71, 0xBB, 0xEE, 0xBF,
+	0x85, 0xC8, 0xA1
+};
+
+static const u8 exp_to_poly[492] = {
+	0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x4D, 0x9A, 0x79, 0xF2,
+	0xA9, 0x1F, 0x3E, 0x7C, 0xF8, 0xBD, 0x37, 0x6E, 0xDC, 0xF5, 0xA7, 0x03,
+	0x06, 0x0C, 0x18, 0x30, 0x60, 0xC0, 0xCD, 0xD7, 0xE3, 0x8B, 0x5B, 0xB6,
+	0x21, 0x42, 0x84, 0x45, 0x8A, 0x59, 0xB2, 0x29, 0x52, 0xA4, 0x05, 0x0A,
+	0x14, 0x28, 0x50, 0xA0, 0x0D, 0x1A, 0x34, 0x68, 0xD0, 0xED, 0x97, 0x63,
+	0xC6, 0xC1, 0xCF, 0xD3, 0xEB, 0x9B, 0x7B, 0xF6, 0xA1, 0x0F, 0x1E, 0x3C,
+	0x78, 0xF0, 0xAD, 0x17, 0x2E, 0x5C, 0xB8, 0x3D, 0x7A, 0xF4, 0xA5, 0x07,
+	0x0E, 0x1C, 0x38, 0x70, 0xE0, 0x8D, 0x57, 0xAE, 0x11, 0x22, 0x44, 0x88,
+	0x5D, 0xBA, 0x39, 0x72, 0xE4, 0x85, 0x47, 0x8E, 0x51, 0xA2, 0x09, 0x12,
+	0x24, 0x48, 0x90, 0x6D, 0xDA, 0xF9, 0xBF, 0x33, 0x66, 0xCC, 0xD5, 0xE7,
+	0x83, 0x4B, 0x96, 0x61, 0xC2, 0xC9, 0xDF, 0xF3, 0xAB, 0x1B, 0x36, 0x6C,
+	0xD8, 0xFD, 0xB7, 0x23, 0x46, 0x8C, 0x55, 0xAA, 0x19, 0x32, 0x64, 0xC8,
+	0xDD, 0xF7, 0xA3, 0x0B, 0x16, 0x2C, 0x58, 0xB0, 0x2D, 0x5A, 0xB4, 0x25,
+	0x4A, 0x94, 0x65, 0xCA, 0xD9, 0xFF, 0xB3, 0x2B, 0x56, 0xAC, 0x15, 0x2A,
+	0x54, 0xA8, 0x1D, 0x3A, 0x74, 0xE8, 0x9D, 0x77, 0xEE, 0x91, 0x6F, 0xDE,
+	0xF1, 0xAF, 0x13, 0x26, 0x4C, 0x98, 0x7D, 0xFA, 0xB9, 0x3F, 0x7E, 0xFC,
+	0xB5, 0x27, 0x4E, 0x9C, 0x75, 0xEA, 0x99, 0x7F, 0xFE, 0xB1, 0x2F, 0x5E,
+	0xBC, 0x35, 0x6A, 0xD4, 0xE5, 0x87, 0x43, 0x86, 0x41, 0x82, 0x49, 0x92,
+	0x69, 0xD2, 0xE9, 0x9F, 0x73, 0xE6, 0x81, 0x4F, 0x9E, 0x71, 0xE2, 0x89,
+	0x5F, 0xBE, 0x31, 0x62, 0xC4, 0xC5, 0xC7, 0xC3, 0xCB, 0xDB, 0xFB, 0xBB,
+	0x3B, 0x76, 0xEC, 0x95, 0x67, 0xCE, 0xD1, 0xEF, 0x93, 0x6B, 0xD6, 0xE1,
+	0x8F, 0x53, 0xA6, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x4D,
+	0x9A, 0x79, 0xF2, 0xA9, 0x1F, 0x3E, 0x7C, 0xF8, 0xBD, 0x37, 0x6E, 0xDC,
+	0xF5, 0xA7, 0x03, 0x06, 0x0C, 0x18, 0x30, 0x60, 0xC0, 0xCD, 0xD7, 0xE3,
+	0x8B, 0x5B, 0xB6, 0x21, 0x42, 0x84, 0x45, 0x8A, 0x59, 0xB2, 0x29, 0x52,
+	0xA4, 0x05, 0x0A, 0x14, 0x28, 0x50, 0xA0, 0x0D, 0x1A, 0x34, 0x68, 0xD0,
+	0xED, 0x97, 0x63, 0xC6, 0xC1, 0xCF, 0xD3, 0xEB, 0x9B, 0x7B, 0xF6, 0xA1,
+	0x0F, 0x1E, 0x3C, 0x78, 0xF0, 0xAD, 0x17, 0x2E, 0x5C, 0xB8, 0x3D, 0x7A,
+	0xF4, 0xA5, 0x07, 0x0E, 0x1C, 0x38, 0x70, 0xE0, 0x8D, 0x57, 0xAE, 0x11,
+	0x22, 0x44, 0x88, 0x5D, 0xBA, 0x39, 0x72, 0xE4, 0x85, 0x47, 0x8E, 0x51,
+	0xA2, 0x09, 0x12, 0x24, 0x48, 0x90, 0x6D, 0xDA, 0xF9, 0xBF, 0x33, 0x66,
+	0xCC, 0xD5, 0xE7, 0x83, 0x4B, 0x96, 0x61, 0xC2, 0xC9, 0xDF, 0xF3, 0xAB,
+	0x1B, 0x36, 0x6C, 0xD8, 0xFD, 0xB7, 0x23, 0x46, 0x8C, 0x55, 0xAA, 0x19,
+	0x32, 0x64, 0xC8, 0xDD, 0xF7, 0xA3, 0x0B, 0x16, 0x2C, 0x58, 0xB0, 0x2D,
+	0x5A, 0xB4, 0x25, 0x4A, 0x94, 0x65, 0xCA, 0xD9, 0xFF, 0xB3, 0x2B, 0x56,
+	0xAC, 0x15, 0x2A, 0x54, 0xA8, 0x1D, 0x3A, 0x74, 0xE8, 0x9D, 0x77, 0xEE,
+	0x91, 0x6F, 0xDE, 0xF1, 0xAF, 0x13, 0x26, 0x4C, 0x98, 0x7D, 0xFA, 0xB9,
+	0x3F, 0x7E, 0xFC, 0xB5, 0x27, 0x4E, 0x9C, 0x75, 0xEA, 0x99, 0x7F, 0xFE,
+	0xB1, 0x2F, 0x5E, 0xBC, 0x35, 0x6A, 0xD4, 0xE5, 0x87, 0x43, 0x86, 0x41,
+	0x82, 0x49, 0x92, 0x69, 0xD2, 0xE9, 0x9F, 0x73, 0xE6, 0x81, 0x4F, 0x9E,
+	0x71, 0xE2, 0x89, 0x5F, 0xBE, 0x31, 0x62, 0xC4, 0xC5, 0xC7, 0xC3, 0xCB
+};
+
+
+/* The table constants are indices of
+ * S-box entries, preprocessed through q0 and q1. */
+static const u8 calc_sb_tbl[512] = {
+	0xA9, 0x75, 0x67, 0xF3, 0xB3, 0xC6, 0xE8, 0xF4,
+	0x04, 0xDB, 0xFD, 0x7B, 0xA3, 0xFB, 0x76, 0xC8,
+	0x9A, 0x4A, 0x92, 0xD3, 0x80, 0xE6, 0x78, 0x6B,
+	0xE4, 0x45, 0xDD, 0x7D, 0xD1, 0xE8, 0x38, 0x4B,
+	0x0D, 0xD6, 0xC6, 0x32, 0x35, 0xD8, 0x98, 0xFD,
+	0x18, 0x37, 0xF7, 0x71, 0xEC, 0xF1, 0x6C, 0xE1,
+	0x43, 0x30, 0x75, 0x0F, 0x37, 0xF8, 0x26, 0x1B,
+	0xFA, 0x87, 0x13, 0xFA, 0x94, 0x06, 0x48, 0x3F,
+	0xF2, 0x5E, 0xD0, 0xBA, 0x8B, 0xAE, 0x30, 0x5B,
+	0x84, 0x8A, 0x54, 0x00, 0xDF, 0xBC, 0x23, 0x9D,
+	0x19, 0x6D, 0x5B, 0xC1, 0x3D, 0xB1, 0x59, 0x0E,
+	0xF3, 0x80, 0xAE, 0x5D, 0xA2, 0xD2, 0x82, 0xD5,
+	0x63, 0xA0, 0x01, 0x84, 0x83, 0x07, 0x2E, 0x14,
+	0xD9, 0xB5, 0x51, 0x90, 0x9B, 0x2C, 0x7C, 0xA3,
+	0xA6, 0xB2, 0xEB, 0x73, 0xA5, 0x4C, 0xBE, 0x54,
+	0x16, 0x92, 0x0C, 0x74, 0xE3, 0x36, 0x61, 0x51,
+	0xC0, 0x38, 0x8C, 0xB0, 0x3A, 0xBD, 0xF5, 0x5A,
+	0x73, 0xFC, 0x2C, 0x60, 0x25, 0x62, 0x0B, 0x96,
+	0xBB, 0x6C, 0x4E, 0x42, 0x89, 0xF7, 0x6B, 0x10,
+	0x53, 0x7C, 0x6A, 0x28, 0xB4, 0x27, 0xF1, 0x8C,
+	0xE1, 0x13, 0xE6, 0x95, 0xBD, 0x9C, 0x45, 0xC7,
+	0xE2, 0x24, 0xF4, 0x46, 0xB6, 0x3B, 0x66, 0x70,
+	0xCC, 0xCA, 0x95, 0xE3, 0x03, 0x85, 0x56, 0xCB,
+	0xD4, 0x11, 0x1C, 0xD0, 0x1E, 0x93, 0xD7, 0xB8,
+	0xFB, 0xA6, 0xC3, 0x83, 0x8E, 0x20, 0xB5, 0xFF,
+	0xE9, 0x9F, 0xCF, 0x77, 0xBF, 0xC3, 0xBA, 0xCC,
+	0xEA, 0x03, 0x77, 0x6F, 0x39, 0x08, 0xAF, 0xBF,
+	0x33, 0x40, 0xC9, 0xE7, 0x62, 0x2B, 0x71, 0xE2,
+	0x81, 0x79, 0x79, 0x0C, 0x09, 0xAA, 0xAD, 0x82,
+	0x24, 0x41, 0xCD, 0x3A, 0xF9, 0xEA, 0xD8, 0xB9,
+	0xE5, 0xE4, 0xC5, 0x9A, 0xB9, 0xA4, 0x4D, 0x97,
+	0x44, 0x7E, 0x08, 0xDA, 0x86, 0x7A, 0xE7, 0x17,
+	0xA1, 0x66, 0x1D, 0x94, 0xAA, 0xA1, 0xED, 0x1D,
+	0x06, 0x3D, 0x70, 0xF0, 0xB2, 0xDE, 0xD2, 0xB3,
+	0x41, 0x0B, 0x7B, 0x72, 0xA0, 0xA7, 0x11, 0x1C,
+	0x31, 0xEF, 0xC2, 0xD1, 0x27, 0x53, 0x90, 0x3E,
+	0x20, 0x8F, 0xF6, 0x33, 0x60, 0x26, 0xFF, 0x5F,
+	0x96, 0xEC, 0x5C, 0x76, 0xB1, 0x2A, 0xAB, 0x49,
+	0x9E, 0x81, 0x9C, 0x88, 0x52, 0xEE, 0x1B, 0x21,
+	0x5F, 0xC4, 0x93, 0x1A, 0x0A, 0xEB, 0xEF, 0xD9,
+	0x91, 0xC5, 0x85, 0x39, 0x49, 0x99, 0xEE, 0xCD,
+	0x2D, 0xAD, 0x4F, 0x31, 0x8F, 0x8B, 0x3B, 0x01,
+	0x47, 0x18, 0x87, 0x23, 0x6D, 0xDD, 0x46, 0x1F,
+	0xD6, 0x4E, 0x3E, 0x2D, 0x69, 0xF9, 0x64, 0x48,
+	0x2A, 0x4F, 0xCE, 0xF2, 0xCB, 0x65, 0x2F, 0x8E,
+	0xFC, 0x78, 0x97, 0x5C, 0x05, 0x58, 0x7A, 0x19,
+	0xAC, 0x8D, 0x7F, 0xE5, 0xD5, 0x98, 0x1A, 0x57,
+	0x4B, 0x67, 0x0E, 0x7F, 0xA7, 0x05, 0x5A, 0x64,
+	0x28, 0xAF, 0x14, 0x63, 0x3F, 0xB6, 0x29, 0xFE,
+	0x88, 0xF5, 0x3C, 0xB7, 0x4C, 0x3C, 0x02, 0xA5,
+	0xB8, 0xCE, 0xDA, 0xE9, 0xB0, 0x68, 0x17, 0x44,
+	0x55, 0xE0, 0x1F, 0x4D, 0x8A, 0x43, 0x7D, 0x69,
+	0x57, 0x29, 0xC7, 0x2E, 0x8D, 0xAC, 0x74, 0x15,
+	0xB7, 0x59, 0xC4, 0xA8, 0x9F, 0x0A, 0x72, 0x9E,
+	0x7E, 0x6E, 0x15, 0x47, 0x22, 0xDF, 0x12, 0x34,
+	0x58, 0x35, 0x07, 0x6A, 0x99, 0xCF, 0x34, 0xDC,
+	0x6E, 0x22, 0x50, 0xC9, 0xDE, 0xC0, 0x68, 0x9B,
+	0x65, 0x89, 0xBC, 0xD4, 0xDB, 0xED, 0xF8, 0xAB,
+	0xC8, 0x12, 0xA8, 0xA2, 0x2B, 0x0D, 0x40, 0x52,
+	0xDC, 0xBB, 0xFE, 0x02, 0x32, 0x2F, 0xA4, 0xA9,
+	0xCA, 0xD7, 0x10, 0x61, 0x21, 0x1E, 0xF0, 0xB4,
+	0xD3, 0x50, 0x5D, 0x04, 0x0F, 0xF6, 0x00, 0xC2,
+	0x6F, 0x16, 0x9D, 0x25, 0x36, 0x86, 0x42, 0x56,
+	0x4A, 0x55, 0x5E, 0x09, 0xC1, 0xBE, 0xE0, 0x91
+};
+
+/* Macro to perform one column of the RS matrix multiplication.  The
+ * parameters a, b, c, and d are the four bytes of output; i is the index
+ * of the key bytes, and w, x, y, and z, are the column of constants from
+ * the RS matrix, preprocessed through the poly_to_exp table. */
+
+#define CALC_S(a, b, c, d, i, w, x, y, z) \
+   if (key[i]) { \
+      tmp = poly_to_exp[key[i] - 1]; \
+      (a) ^= exp_to_poly[tmp + (w)]; \
+      (b) ^= exp_to_poly[tmp + (x)]; \
+      (c) ^= exp_to_poly[tmp + (y)]; \
+      (d) ^= exp_to_poly[tmp + (z)]; \
+   }
+
+/* Macros to calculate the key-dependent S-boxes for a 128-bit key using
+ * the S vector from CALC_S.  CALC_SB_2 computes a single entry in all
+ * four S-boxes, where i is the index of the entry to compute, and a and b
+ * are the index numbers preprocessed through the q0 and q1 tables
+ * respectively. */
+
+#define CALC_SB_2(i, a, b) \
+   ctx->s[0][i] = mds[0][q0[(a) ^ sa] ^ se]; \
+   ctx->s[1][i] = mds[1][q0[(b) ^ sb] ^ sf]; \
+   ctx->s[2][i] = mds[2][q1[(a) ^ sc] ^ sg]; \
+   ctx->s[3][i] = mds[3][q1[(b) ^ sd] ^ sh]
+
+/* Macro exactly like CALC_SB_2, but for 192-bit keys. */
+
+#define CALC_SB192_2(i, a, b) \
+   ctx->s[0][i] = mds[0][q0[q0[(b) ^ sa] ^ se] ^ si]; \
+   ctx->s[1][i] = mds[1][q0[q1[(b) ^ sb] ^ sf] ^ sj]; \
+   ctx->s[2][i] = mds[2][q1[q0[(a) ^ sc] ^ sg] ^ sk]; \
+   ctx->s[3][i] = mds[3][q1[q1[(a) ^ sd] ^ sh] ^ sl];
+
+/* Macro exactly like CALC_SB_2, but for 256-bit keys. */
+
+#define CALC_SB256_2(i, a, b) \
+   ctx->s[0][i] = mds[0][q0[q0[q1[(b) ^ sa] ^ se] ^ si] ^ sm]; \
+   ctx->s[1][i] = mds[1][q0[q1[q1[(a) ^ sb] ^ sf] ^ sj] ^ sn]; \
+   ctx->s[2][i] = mds[2][q1[q0[q0[(a) ^ sc] ^ sg] ^ sk] ^ so]; \
+   ctx->s[3][i] = mds[3][q1[q1[q0[(b) ^ sd] ^ sh] ^ sl] ^ sp];
+
+/* Macros to calculate the whitening and round subkeys.  CALC_K_2 computes the
+ * last two stages of the h() function for a given index (either 2i or 2i+1).
+ * a, b, c, and d are the four bytes going into the last two stages.  For
+ * 128-bit keys, this is the entire h() function and a and c are the index
+ * preprocessed through q0 and q1 respectively; for longer keys they are the
+ * output of previous stages.  j is the index of the first key byte to use.
+ * CALC_K computes a pair of subkeys for 128-bit Twofish, by calling CALC_K_2
+ * twice, doing the Pseudo-Hadamard Transform, and doing the necessary
+ * rotations.  Its parameters are: a, the array to write the results into,
+ * j, the index of the first output entry, k and l, the preprocessed indices
+ * for index 2i, and m and n, the preprocessed indices for index 2i+1.
+ * CALC_K192_2 expands CALC_K_2 to handle 192-bit keys, by doing an
+ * additional lookup-and-XOR stage.  The parameters a, b, c and d are the
+ * four bytes going into the last three stages.  For 192-bit keys, c = d
+ * are the index preprocessed through q0, and a = b are the index
+ * preprocessed through q1; j is the index of the first key byte to use.
+ * CALC_K192 is identical to CALC_K but for using the CALC_K192_2 macro
+ * instead of CALC_K_2.
+ * CALC_K256_2 expands CALC_K192_2 to handle 256-bit keys, by doing an
+ * additional lookup-and-XOR stage.  The parameters a and b are the index
+ * preprocessed through q0 and q1 respectively; j is the index of the first
+ * key byte to use.  CALC_K256 is identical to CALC_K but for using the
+ * CALC_K256_2 macro instead of CALC_K_2. */
+
+#define CALC_K_2(a, b, c, d, j) \
+     mds[0][q0[a ^ key[(j) + 8]] ^ key[j]] \
+   ^ mds[1][q0[b ^ key[(j) + 9]] ^ key[(j) + 1]] \
+   ^ mds[2][q1[c ^ key[(j) + 10]] ^ key[(j) + 2]] \
+   ^ mds[3][q1[d ^ key[(j) + 11]] ^ key[(j) + 3]]
+
+#define CALC_K(a, j, k, l, m, n) \
+   x = CALC_K_2 (k, l, k, l, 0); \
+   y = CALC_K_2 (m, n, m, n, 4); \
+   y = rol32(y, 8); \
+   x += y; y += x; ctx->a[j] = x; \
+   ctx->a[(j) + 1] = rol32(y, 9)
+
+#define CALC_K192_2(a, b, c, d, j) \
+   CALC_K_2 (q0[a ^ key[(j) + 16]], \
+	     q1[b ^ key[(j) + 17]], \
+	     q0[c ^ key[(j) + 18]], \
+	     q1[d ^ key[(j) + 19]], j)
+
+#define CALC_K192(a, j, k, l, m, n) \
+   x = CALC_K192_2 (l, l, k, k, 0); \
+   y = CALC_K192_2 (n, n, m, m, 4); \
+   y = rol32(y, 8); \
+   x += y; y += x; ctx->a[j] = x; \
+   ctx->a[(j) + 1] = rol32(y, 9)
+
+#define CALC_K256_2(a, b, j) \
+   CALC_K192_2 (q1[b ^ key[(j) + 24]], \
+	        q1[a ^ key[(j) + 25]], \
+	        q0[a ^ key[(j) + 26]], \
+	        q0[b ^ key[(j) + 27]], j)
+
+#define CALC_K256(a, j, k, l, m, n) \
+   x = CALC_K256_2 (k, l, 0); \
+   y = CALC_K256_2 (m, n, 4); \
+   y = rol32(y, 8); \
+   x += y; y += x; ctx->a[j] = x; \
+   ctx->a[(j) + 1] = rol32(y, 9)
+
+/* Perform the key setup. */
+int twofish_setkey(struct crypto_tfm *tfm, const u8 *key,
+		   unsigned int key_len, u32 *flags)
+{
+
+	struct twofish_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	int i, j, k;
+
+	/* Temporaries for CALC_K. */
+	u32 x, y;
+
+	/* The S vector used to key the S-boxes, split up into individual bytes.
+	 * 128-bit keys use only sa through sh; 256-bit use all of them. */
+	u8 sa = 0, sb = 0, sc = 0, sd = 0, se = 0, sf = 0, sg = 0, sh = 0;
+	u8 si = 0, sj = 0, sk = 0, sl = 0, sm = 0, sn = 0, so = 0, sp = 0;
+
+	/* Temporary for CALC_S. */
+	u8 tmp;
+
+	/* Check key length. */
+	if (key_len != 16 && key_len != 24 && key_len != 32)
+	{
+		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
+		return -EINVAL; /* unsupported key length */
+	}
+
+	/* Compute the first two words of the S vector.  The magic numbers are
+	 * the entries of the RS matrix, preprocessed through poly_to_exp. The
+	 * numbers in the comments are the original (polynomial form) matrix
+	 * entries. */
+	CALC_S (sa, sb, sc, sd, 0, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
+	CALC_S (sa, sb, sc, sd, 1, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
+	CALC_S (sa, sb, sc, sd, 2, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
+	CALC_S (sa, sb, sc, sd, 3, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
+	CALC_S (sa, sb, sc, sd, 4, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
+	CALC_S (sa, sb, sc, sd, 5, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
+	CALC_S (sa, sb, sc, sd, 6, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
+	CALC_S (sa, sb, sc, sd, 7, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
+	CALC_S (se, sf, sg, sh, 8, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
+	CALC_S (se, sf, sg, sh, 9, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
+	CALC_S (se, sf, sg, sh, 10, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
+	CALC_S (se, sf, sg, sh, 11, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
+	CALC_S (se, sf, sg, sh, 12, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
+	CALC_S (se, sf, sg, sh, 13, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
+	CALC_S (se, sf, sg, sh, 14, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
+	CALC_S (se, sf, sg, sh, 15, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
+
+	if (key_len == 24 || key_len == 32) { /* 192- or 256-bit key */
+		/* Calculate the third word of the S vector */
+		CALC_S (si, sj, sk, sl, 16, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
+		CALC_S (si, sj, sk, sl, 17, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
+		CALC_S (si, sj, sk, sl, 18, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
+		CALC_S (si, sj, sk, sl, 19, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
+		CALC_S (si, sj, sk, sl, 20, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
+		CALC_S (si, sj, sk, sl, 21, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
+		CALC_S (si, sj, sk, sl, 22, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
+		CALC_S (si, sj, sk, sl, 23, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
+	}
+
+	if (key_len == 32) { /* 256-bit key */
+		/* Calculate the fourth word of the S vector */
+		CALC_S (sm, sn, so, sp, 24, 0x00, 0x2D, 0x01, 0x2D); /* 01 A4 02 A4 */
+		CALC_S (sm, sn, so, sp, 25, 0x2D, 0xA4, 0x44, 0x8A); /* A4 56 A1 55 */
+		CALC_S (sm, sn, so, sp, 26, 0x8A, 0xD5, 0xBF, 0xD1); /* 55 82 FC 87 */
+		CALC_S (sm, sn, so, sp, 27, 0xD1, 0x7F, 0x3D, 0x99); /* 87 F3 C1 5A */
+		CALC_S (sm, sn, so, sp, 28, 0x99, 0x46, 0x66, 0x96); /* 5A 1E 47 58 */
+		CALC_S (sm, sn, so, sp, 29, 0x96, 0x3C, 0x5B, 0xED); /* 58 C6 AE DB */
+		CALC_S (sm, sn, so, sp, 30, 0xED, 0x37, 0x4F, 0xE0); /* DB 68 3D 9E */
+		CALC_S (sm, sn, so, sp, 31, 0xE0, 0xD0, 0x8C, 0x17); /* 9E E5 19 03 */
+
+		/* Compute the S-boxes. */
+		for ( i = j = 0, k = 1; i < 256; i++, j += 2, k += 2 ) {
+			CALC_SB256_2( i, calc_sb_tbl[j], calc_sb_tbl[k] );
+		}
+
+		/* Calculate whitening and round subkeys.  The constants are
+		 * indices of subkeys, preprocessed through q0 and q1. */
+		CALC_K256 (w, 0, 0xA9, 0x75, 0x67, 0xF3);
+		CALC_K256 (w, 2, 0xB3, 0xC6, 0xE8, 0xF4);
+		CALC_K256 (w, 4, 0x04, 0xDB, 0xFD, 0x7B);
+		CALC_K256 (w, 6, 0xA3, 0xFB, 0x76, 0xC8);
+		CALC_K256 (k, 0, 0x9A, 0x4A, 0x92, 0xD3);
+		CALC_K256 (k, 2, 0x80, 0xE6, 0x78, 0x6B);
+		CALC_K256 (k, 4, 0xE4, 0x45, 0xDD, 0x7D);
+		CALC_K256 (k, 6, 0xD1, 0xE8, 0x38, 0x4B);
+		CALC_K256 (k, 8, 0x0D, 0xD6, 0xC6, 0x32);
+		CALC_K256 (k, 10, 0x35, 0xD8, 0x98, 0xFD);
+		CALC_K256 (k, 12, 0x18, 0x37, 0xF7, 0x71);
+		CALC_K256 (k, 14, 0xEC, 0xF1, 0x6C, 0xE1);
+		CALC_K256 (k, 16, 0x43, 0x30, 0x75, 0x0F);
+		CALC_K256 (k, 18, 0x37, 0xF8, 0x26, 0x1B);
+		CALC_K256 (k, 20, 0xFA, 0x87, 0x13, 0xFA);
+		CALC_K256 (k, 22, 0x94, 0x06, 0x48, 0x3F);
+		CALC_K256 (k, 24, 0xF2, 0x5E, 0xD0, 0xBA);
+		CALC_K256 (k, 26, 0x8B, 0xAE, 0x30, 0x5B);
+		CALC_K256 (k, 28, 0x84, 0x8A, 0x54, 0x00);
+		CALC_K256 (k, 30, 0xDF, 0xBC, 0x23, 0x9D);
+	} else if (key_len == 24) { /* 192-bit key */
+		/* Compute the S-boxes. */
+		for ( i = j = 0, k = 1; i < 256; i++, j += 2, k += 2 ) {
+		        CALC_SB192_2( i, calc_sb_tbl[j], calc_sb_tbl[k] );
+		}
+
+		/* Calculate whitening and round subkeys.  The constants are
+		 * indices of subkeys, preprocessed through q0 and q1. */
+		CALC_K192 (w, 0, 0xA9, 0x75, 0x67, 0xF3);
+		CALC_K192 (w, 2, 0xB3, 0xC6, 0xE8, 0xF4);
+		CALC_K192 (w, 4, 0x04, 0xDB, 0xFD, 0x7B);
+		CALC_K192 (w, 6, 0xA3, 0xFB, 0x76, 0xC8);
+		CALC_K192 (k, 0, 0x9A, 0x4A, 0x92, 0xD3);
+		CALC_K192 (k, 2, 0x80, 0xE6, 0x78, 0x6B);
+		CALC_K192 (k, 4, 0xE4, 0x45, 0xDD, 0x7D);
+		CALC_K192 (k, 6, 0xD1, 0xE8, 0x38, 0x4B);
+		CALC_K192 (k, 8, 0x0D, 0xD6, 0xC6, 0x32);
+		CALC_K192 (k, 10, 0x35, 0xD8, 0x98, 0xFD);
+		CALC_K192 (k, 12, 0x18, 0x37, 0xF7, 0x71);
+		CALC_K192 (k, 14, 0xEC, 0xF1, 0x6C, 0xE1);
+		CALC_K192 (k, 16, 0x43, 0x30, 0x75, 0x0F);
+		CALC_K192 (k, 18, 0x37, 0xF8, 0x26, 0x1B);
+		CALC_K192 (k, 20, 0xFA, 0x87, 0x13, 0xFA);
+		CALC_K192 (k, 22, 0x94, 0x06, 0x48, 0x3F);
+		CALC_K192 (k, 24, 0xF2, 0x5E, 0xD0, 0xBA);
+		CALC_K192 (k, 26, 0x8B, 0xAE, 0x30, 0x5B);
+		CALC_K192 (k, 28, 0x84, 0x8A, 0x54, 0x00);
+		CALC_K192 (k, 30, 0xDF, 0xBC, 0x23, 0x9D);
+	} else { /* 128-bit key */
+		/* Compute the S-boxes. */
+		for ( i = j = 0, k = 1; i < 256; i++, j += 2, k += 2 ) {
+			CALC_SB_2( i, calc_sb_tbl[j], calc_sb_tbl[k] );
+		}
+
+		/* Calculate whitening and round subkeys.  The constants are
+		 * indices of subkeys, preprocessed through q0 and q1. */
+		CALC_K (w, 0, 0xA9, 0x75, 0x67, 0xF3);
+		CALC_K (w, 2, 0xB3, 0xC6, 0xE8, 0xF4);
+		CALC_K (w, 4, 0x04, 0xDB, 0xFD, 0x7B);
+		CALC_K (w, 6, 0xA3, 0xFB, 0x76, 0xC8);
+		CALC_K (k, 0, 0x9A, 0x4A, 0x92, 0xD3);
+		CALC_K (k, 2, 0x80, 0xE6, 0x78, 0x6B);
+		CALC_K (k, 4, 0xE4, 0x45, 0xDD, 0x7D);
+		CALC_K (k, 6, 0xD1, 0xE8, 0x38, 0x4B);
+		CALC_K (k, 8, 0x0D, 0xD6, 0xC6, 0x32);
+		CALC_K (k, 10, 0x35, 0xD8, 0x98, 0xFD);
+		CALC_K (k, 12, 0x18, 0x37, 0xF7, 0x71);
+		CALC_K (k, 14, 0xEC, 0xF1, 0x6C, 0xE1);
+		CALC_K (k, 16, 0x43, 0x30, 0x75, 0x0F);
+		CALC_K (k, 18, 0x37, 0xF8, 0x26, 0x1B);
+		CALC_K (k, 20, 0xFA, 0x87, 0x13, 0xFA);
+		CALC_K (k, 22, 0x94, 0x06, 0x48, 0x3F);
+		CALC_K (k, 24, 0xF2, 0x5E, 0xD0, 0xBA);
+		CALC_K (k, 26, 0x8B, 0xAE, 0x30, 0x5B);
+		CALC_K (k, 28, 0x84, 0x8A, 0x54, 0x00);
+		CALC_K (k, 30, 0xDF, 0xBC, 0x23, 0x9D);
+	}
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(twofish_setkey);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Twofish cipher common functions");
diff --git a/include/crypto/twofish.h b/include/crypto/twofish.h
new file mode 100644
index 000000000000..e4328cfaaf64
--- /dev/null
+++ b/include/crypto/twofish.h
@@ -0,0 +1,23 @@
+#ifndef _CRYPTO_TWOFISH_H
+#define _CRYPTO_TWOFISH_H
+
+#include <linux/types.h>
+
+#define TF_MIN_KEY_SIZE 16
+#define TF_MAX_KEY_SIZE 32
+#define TF_BLOCK_SIZE 16
+
+struct crypto_tfm;
+
+/* Structure for an expanded Twofish key.  s contains the key-dependent
+ * S-boxes composed with the MDS matrix; w contains the eight "whitening"
+ * subkeys, K[0] through K[7].	k holds the remaining, "round" subkeys.  Note
+ * that k[i] corresponds to what the Twofish paper calls K[i+8]. */
+struct twofish_ctx {
+	u32 s[4][256], w[8], k[32];
+};
+
+int twofish_setkey(struct crypto_tfm *tfm, const u8 *key,
+		   unsigned int key_len, u32 *flags);
+
+#endif
-- 
cgit v1.2.3


From 6521f30273fbec65146a0f16de74b7b402b0f7b0 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 6 Aug 2006 20:28:44 +1000
Subject: [CRYPTO] api: Add crypto_alg reference counting

Up until now we've relied on module reference counting to ensure that the
crypto_alg structures don't disappear from under us.  This was good enough
as long as each crypto_alg came from exactly one module.

However, with parameterised crypto algorithms a crypto_alg object may need
two or more modules to operate.  This means that we need to count the
references to the crypto_alg object directly.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/api.c           | 32 ++++++++++++++++++++++++++------
 crypto/proc.c          |  3 +++
 include/linux/crypto.h |  3 +++
 3 files changed, 32 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/crypto/api.c b/crypto/api.c
index 8c2743a05f90..5994a58ef954 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -29,13 +29,26 @@
 LIST_HEAD(crypto_alg_list);
 DECLARE_RWSEM(crypto_alg_sem);
 
-static inline int crypto_mod_get(struct crypto_alg *alg)
+static inline struct crypto_alg *crypto_alg_get(struct crypto_alg *alg)
 {
-	return try_module_get(alg->cra_module);
+	atomic_inc(&alg->cra_refcnt);
+	return alg;
+}
+
+static inline void crypto_alg_put(struct crypto_alg *alg)
+{
+	if (atomic_dec_and_test(&alg->cra_refcnt) && alg->cra_destroy)
+		alg->cra_destroy(alg);
+}
+
+static struct crypto_alg *crypto_mod_get(struct crypto_alg *alg)
+{
+	return try_module_get(alg->cra_module) ? crypto_alg_get(alg) : NULL;
 }
 
-static inline void crypto_mod_put(struct crypto_alg *alg)
+static void crypto_mod_put(struct crypto_alg *alg)
 {
+	crypto_alg_put(alg);
 	module_put(alg->cra_module);
 }
 
@@ -274,6 +287,7 @@ int crypto_register_alg(struct crypto_alg *alg)
 	}
 	
 	list_add(&alg->cra_list, &crypto_alg_list);
+	atomic_set(&alg->cra_refcnt, 1);
 out:	
 	up_write(&crypto_alg_sem);
 	return ret;
@@ -284,8 +298,6 @@ int crypto_unregister_alg(struct crypto_alg *alg)
 	int ret = -ENOENT;
 	struct crypto_alg *q;
 	
-	BUG_ON(!alg->cra_module);
-	
 	down_write(&crypto_alg_sem);
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
 		if (alg == q) {
@@ -296,7 +308,15 @@ int crypto_unregister_alg(struct crypto_alg *alg)
 	}
 out:	
 	up_write(&crypto_alg_sem);
-	return ret;
+
+	if (ret)
+		return ret;
+
+	BUG_ON(atomic_read(&alg->cra_refcnt) != 1);
+	if (alg->cra_destroy)
+		alg->cra_destroy(alg);
+
+	return 0;
 }
 
 int crypto_alg_available(const char *name, u32 flags)
diff --git a/crypto/proc.c b/crypto/proc.c
index c0a5dd7ce2cc..8543b7a157d6 100644
--- a/crypto/proc.c
+++ b/crypto/proc.c
@@ -12,6 +12,8 @@
  * any later version.
  *
  */
+
+#include <asm/atomic.h>
 #include <linux/init.h>
 #include <linux/crypto.h>
 #include <linux/rwsem.h>
@@ -54,6 +56,7 @@ static int c_show(struct seq_file *m, void *p)
 	seq_printf(m, "driver       : %s\n", alg->cra_driver_name);
 	seq_printf(m, "module       : %s\n", module_name(alg->cra_module));
 	seq_printf(m, "priority     : %d\n", alg->cra_priority);
+	seq_printf(m, "refcnt       : %d\n", atomic_read(&alg->cra_refcnt));
 	
 	switch (alg->cra_flags & CRYPTO_ALG_TYPE_MASK) {
 	case CRYPTO_ALG_TYPE_CIPHER:
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index cb1e6631b132..7f57ff8ec975 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -17,6 +17,7 @@
 #ifndef _LINUX_CRYPTO_H
 #define _LINUX_CRYPTO_H
 
+#include <asm/atomic.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
@@ -148,6 +149,7 @@ struct crypto_alg {
 	unsigned int cra_alignmask;
 
 	int cra_priority;
+	atomic_t cra_refcnt;
 
 	char cra_name[CRYPTO_MAX_ALG_NAME];
 	char cra_driver_name[CRYPTO_MAX_ALG_NAME];
@@ -160,6 +162,7 @@ struct crypto_alg {
 
 	int (*cra_init)(struct crypto_tfm *tfm);
 	void (*cra_exit)(struct crypto_tfm *tfm);
+	void (*cra_destroy)(struct crypto_alg *alg);
 	
 	struct module *cra_module;
 };
-- 
cgit v1.2.3


From 9409f38a0c8773c04bff8dda8c552d7ea013d956 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 6 Aug 2006 19:49:12 +1000
Subject: [IPSEC]: Move linux/crypto.h inclusion out of net/xfrm.h

The header file linux/crypto.h is only needed by a few files so including
it in net/xfrm.h (which is included by half of the networking stack) is a
waste.  This patch moves it out of net/xfrm.h and into the specific header
files that actually need it.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/net/ah.h     | 1 +
 include/net/esp.h    | 1 +
 include/net/ipcomp.h | 4 ++++
 include/net/xfrm.h   | 2 +-
 net/xfrm/xfrm_user.c | 1 +
 5 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ah.h b/include/net/ah.h
index ceff00afae09..8e27c9ba8b84 100644
--- a/include/net/ah.h
+++ b/include/net/ah.h
@@ -1,6 +1,7 @@
 #ifndef _NET_AH_H
 #define _NET_AH_H
 
+#include <linux/crypto.h>
 #include <net/xfrm.h>
 
 /* This is the maximum truncated ICV length that we know of. */
diff --git a/include/net/esp.h b/include/net/esp.h
index 90cd94fad7d9..6eb837973c84 100644
--- a/include/net/esp.h
+++ b/include/net/esp.h
@@ -1,6 +1,7 @@
 #ifndef _NET_ESP_H
 #define _NET_ESP_H
 
+#include <linux/crypto.h>
 #include <net/xfrm.h>
 #include <asm/scatterlist.h>
 
diff --git a/include/net/ipcomp.h b/include/net/ipcomp.h
index e651a57ecdd5..b94e3047b4d9 100644
--- a/include/net/ipcomp.h
+++ b/include/net/ipcomp.h
@@ -1,8 +1,12 @@
 #ifndef _NET_IPCOMP_H
 #define _NET_IPCOMP_H
 
+#include <linux/types.h>
+
 #define IPCOMP_SCRATCH_SIZE     65400
 
+struct crypto_tfm;
+
 struct ipcomp_data {
 	u16 threshold;
 	struct crypto_tfm **tfms;
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 9c5ee9f20b65..10396b4bde14 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -8,7 +8,6 @@
 #include <linux/list.h>
 #include <linux/skbuff.h>
 #include <linux/socket.h>
-#include <linux/crypto.h>
 #include <linux/pfkeyv2.h>
 #include <linux/in6.h>
 #include <linux/mutex.h>
@@ -985,6 +984,7 @@ extern struct xfrm_algo_desc *xfrm_ealg_get_byname(char *name, int probe);
 extern struct xfrm_algo_desc *xfrm_calg_get_byname(char *name, int probe);
 
 struct crypto_tfm;
+struct scatterlist;
 typedef void (icv_update_fn_t)(struct crypto_tfm *, struct scatterlist *, unsigned int);
 
 extern void skb_icv_walk(const struct sk_buff *skb, struct crypto_tfm *tfm,
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 3e6a722d072e..7d18ca03c80d 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -10,6 +10,7 @@
  *
  */
 
+#include <linux/crypto.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
-- 
cgit v1.2.3


From cce9e06d100df19a327b19f23adad76e7bf63edd Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 21 Aug 2006 21:08:13 +1000
Subject: [CRYPTO] api: Split out low-level API

The crypto API is made up of the part facing users such as IPsec and the
low-level part which is used by cryptographic entities such as algorithms.
This patch splits out the latter so that the two APIs are more clearly
delineated.  As a bonus the low-level API can now be modularised if all
algorithms are built as modules.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Kconfig          |  84 ++++++++++++++++++++--------------
 crypto/Makefile         |   7 +--
 crypto/algapi.c         | 118 ++++++++++++++++++++++++++++++++++++++++++++++++
 crypto/api.c            |  97 +--------------------------------------
 crypto/internal.h       |   6 ++-
 crypto/proc.c           |   5 ++
 drivers/crypto/Kconfig  |   3 +-
 include/crypto/algapi.h |  18 ++++++++
 8 files changed, 204 insertions(+), 134 deletions(-)
 create mode 100644 crypto/algapi.c
 create mode 100644 include/crypto/algapi.h

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index fa927a287a1d..aabc63195222 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -9,47 +9,54 @@ config CRYPTO
 	help
 	  This option provides the core Cryptographic API.
 
+if CRYPTO
+
+config CRYPTO_ALGAPI
+	tristate
+	help
+	  This option provides the API for cryptographic algorithms.
+
 config CRYPTO_HMAC
 	bool "HMAC support"
-	depends on CRYPTO
 	help
 	  HMAC: Keyed-Hashing for Message Authentication (RFC2104).
 	  This is required for IPSec.
 
 config CRYPTO_NULL
 	tristate "Null algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  These are 'Null' algorithms, used by IPsec, which do nothing.
 
 config CRYPTO_MD4
 	tristate "MD4 digest algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  MD4 message digest algorithm (RFC1320).
 
 config CRYPTO_MD5
 	tristate "MD5 digest algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  MD5 message digest algorithm (RFC1321).
 
 config CRYPTO_SHA1
 	tristate "SHA1 digest algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2).
 
 config CRYPTO_SHA1_S390
 	tristate "SHA1 digest algorithm (s390)"
-	depends on CRYPTO && S390
+	depends on S390
+	select CRYPTO_ALGAPI
 	help
 	  This is the s390 hardware accelerated implementation of the
 	  SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2).
 
 config CRYPTO_SHA256
 	tristate "SHA256 digest algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  SHA256 secure hash standard (DFIPS 180-2).
 	  
@@ -58,7 +65,8 @@ config CRYPTO_SHA256
 
 config CRYPTO_SHA256_S390
 	tristate "SHA256 digest algorithm (s390)"
-	depends on CRYPTO && S390
+	depends on S390
+	select CRYPTO_ALGAPI
 	help
 	  This is the s390 hardware accelerated implementation of the
 	  SHA256 secure hash standard (DFIPS 180-2).
@@ -68,7 +76,7 @@ config CRYPTO_SHA256_S390
 
 config CRYPTO_SHA512
 	tristate "SHA384 and SHA512 digest algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  SHA512 secure hash standard (DFIPS 180-2).
 	  
@@ -80,7 +88,7 @@ config CRYPTO_SHA512
 
 config CRYPTO_WP512
 	tristate "Whirlpool digest algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Whirlpool hash algorithm 512, 384 and 256-bit hashes
 
@@ -92,7 +100,7 @@ config CRYPTO_WP512
 
 config CRYPTO_TGR192
 	tristate "Tiger digest algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Tiger hash algorithm 192, 160 and 128-bit hashes
 
@@ -105,19 +113,20 @@ config CRYPTO_TGR192
 
 config CRYPTO_DES
 	tristate "DES and Triple DES EDE cipher algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  DES cipher algorithm (FIPS 46-2), and Triple DES EDE (FIPS 46-3).
 
 config CRYPTO_DES_S390
 	tristate "DES and Triple DES cipher algorithms (s390)"
-	depends on CRYPTO && S390
+	depends on S390
+	select CRYPTO_ALGAPI
 	help
 	  DES cipher algorithm (FIPS 46-2), and Triple DES EDE (FIPS 46-3).
 
 config CRYPTO_BLOWFISH
 	tristate "Blowfish cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Blowfish cipher algorithm, by Bruce Schneier.
 	  
@@ -130,7 +139,7 @@ config CRYPTO_BLOWFISH
 
 config CRYPTO_TWOFISH
 	tristate "Twofish cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	select CRYPTO_TWOFISH_COMMON
 	help
 	  Twofish cipher algorithm.
@@ -145,14 +154,14 @@ config CRYPTO_TWOFISH
 
 config CRYPTO_TWOFISH_COMMON
 	tristate
-	depends on CRYPTO
 	help
 	  Common parts of the Twofish cipher algorithm shared by the
 	  generic c and the assembler implementations.
 
 config CRYPTO_TWOFISH_586
 	tristate "Twofish cipher algorithms (i586)"
-	depends on CRYPTO && ((X86 || UML_X86) && !64BIT)
+	depends on (X86 || UML_X86) && !64BIT
+	select CRYPTO_ALGAPI
 	select CRYPTO_TWOFISH_COMMON
 	help
 	  Twofish cipher algorithm.
@@ -167,7 +176,8 @@ config CRYPTO_TWOFISH_586
 
 config CRYPTO_TWOFISH_X86_64
 	tristate "Twofish cipher algorithm (x86_64)"
-	depends on CRYPTO && ((X86 || UML_X86) && 64BIT)
+	depends on (X86 || UML_X86) && 64BIT
+	select CRYPTO_ALGAPI
 	select CRYPTO_TWOFISH_COMMON
 	help
 	  Twofish cipher algorithm (x86_64).
@@ -182,7 +192,7 @@ config CRYPTO_TWOFISH_X86_64
 
 config CRYPTO_SERPENT
 	tristate "Serpent cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Serpent cipher algorithm, by Anderson, Biham & Knudsen.
 
@@ -195,7 +205,7 @@ config CRYPTO_SERPENT
 
 config CRYPTO_AES
 	tristate "AES cipher algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  AES cipher algorithms (FIPS-197). AES uses the Rijndael 
 	  algorithm.
@@ -215,7 +225,8 @@ config CRYPTO_AES
 
 config CRYPTO_AES_586
 	tristate "AES cipher algorithms (i586)"
-	depends on CRYPTO && ((X86 || UML_X86) && !64BIT)
+	depends on (X86 || UML_X86) && !64BIT
+	select CRYPTO_ALGAPI
 	help
 	  AES cipher algorithms (FIPS-197). AES uses the Rijndael 
 	  algorithm.
@@ -235,7 +246,8 @@ config CRYPTO_AES_586
 
 config CRYPTO_AES_X86_64
 	tristate "AES cipher algorithms (x86_64)"
-	depends on CRYPTO && ((X86 || UML_X86) && 64BIT)
+	depends on (X86 || UML_X86) && 64BIT
+	select CRYPTO_ALGAPI
 	help
 	  AES cipher algorithms (FIPS-197). AES uses the Rijndael 
 	  algorithm.
@@ -255,7 +267,8 @@ config CRYPTO_AES_X86_64
 
 config CRYPTO_AES_S390
 	tristate "AES cipher algorithms (s390)"
-	depends on CRYPTO && S390
+	depends on S390
+	select CRYPTO_ALGAPI
 	help
 	  This is the s390 hardware accelerated implementation of the
 	  AES cipher algorithms (FIPS-197). AES uses the Rijndael
@@ -275,21 +288,21 @@ config CRYPTO_AES_S390
 
 config CRYPTO_CAST5
 	tristate "CAST5 (CAST-128) cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  The CAST5 encryption algorithm (synonymous with CAST-128) is
 	  described in RFC2144.
 
 config CRYPTO_CAST6
 	tristate "CAST6 (CAST-256) cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  The CAST6 encryption algorithm (synonymous with CAST-256) is
 	  described in RFC2612.
 
 config CRYPTO_TEA
 	tristate "TEA, XTEA and XETA cipher algorithms"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  TEA cipher algorithm.
 
@@ -306,7 +319,7 @@ config CRYPTO_TEA
 
 config CRYPTO_ARC4
 	tristate "ARC4 cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  ARC4 cipher algorithm.
 
@@ -317,7 +330,7 @@ config CRYPTO_ARC4
 
 config CRYPTO_KHAZAD
 	tristate "Khazad cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Khazad cipher algorithm.
 
@@ -330,7 +343,7 @@ config CRYPTO_KHAZAD
 
 config CRYPTO_ANUBIS
 	tristate "Anubis cipher algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Anubis cipher algorithm.
 
@@ -345,7 +358,7 @@ config CRYPTO_ANUBIS
 
 config CRYPTO_DEFLATE
 	tristate "Deflate compression algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	select ZLIB_INFLATE
 	select ZLIB_DEFLATE
 	help
@@ -356,7 +369,7 @@ config CRYPTO_DEFLATE
 
 config CRYPTO_MICHAEL_MIC
 	tristate "Michael MIC keyed digest algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	help
 	  Michael MIC is used for message integrity protection in TKIP
 	  (IEEE 802.11i). This algorithm is required for TKIP, but it
@@ -365,7 +378,7 @@ config CRYPTO_MICHAEL_MIC
 
 config CRYPTO_CRC32C
 	tristate "CRC32c CRC algorithm"
-	depends on CRYPTO
+	select CRYPTO_ALGAPI
 	select LIBCRC32C
 	help
 	  Castagnoli, et al Cyclic Redundancy-Check Algorithm.  Used
@@ -375,10 +388,13 @@ config CRYPTO_CRC32C
 
 config CRYPTO_TEST
 	tristate "Testing module"
-	depends on CRYPTO && m
+	depends on m
+	select CRYPTO_ALGAPI
 	help
 	  Quick & dirty crypto test module.
 
 source "drivers/crypto/Kconfig"
-endmenu
 
+endif	# if CRYPTO
+
+endmenu
diff --git a/crypto/Makefile b/crypto/Makefile
index fe934f1001c6..6d51f80753a1 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -2,10 +2,11 @@
 # Cryptographic API
 #
 
-proc-crypto-$(CONFIG_PROC_FS) = proc.o
+obj-$(CONFIG_CRYPTO) += api.o scatterwalk.o cipher.o digest.o compress.o
 
-obj-$(CONFIG_CRYPTO) += api.o scatterwalk.o cipher.o digest.o compress.o \
-			$(proc-crypto-y)
+crypto_algapi-$(CONFIG_PROC_FS) += proc.o
+crypto_algapi-objs := algapi.o $(crypto_algapi-y)
+obj-$(CONFIG_CRYPTO_ALGAPI) += crypto_algapi.o
 
 obj-$(CONFIG_CRYPTO_HMAC) += hmac.o
 obj-$(CONFIG_CRYPTO_NULL) += crypto_null.o
diff --git a/crypto/algapi.c b/crypto/algapi.c
new file mode 100644
index 000000000000..a65c6ccfbe17
--- /dev/null
+++ b/crypto/algapi.c
@@ -0,0 +1,118 @@
+/*
+ * Cryptographic API for algorithms (i.e., low-level API).
+ *
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/string.h>
+
+#include "internal.h"
+
+static inline int crypto_set_driver_name(struct crypto_alg *alg)
+{
+	static const char suffix[] = "-generic";
+	char *driver_name = alg->cra_driver_name;
+	int len;
+
+	if (*driver_name)
+		return 0;
+
+	len = strlcpy(driver_name, alg->cra_name, CRYPTO_MAX_ALG_NAME);
+	if (len + sizeof(suffix) > CRYPTO_MAX_ALG_NAME)
+		return -ENAMETOOLONG;
+
+	memcpy(driver_name + len, suffix, sizeof(suffix));
+	return 0;
+}
+
+int crypto_register_alg(struct crypto_alg *alg)
+{
+	int ret;
+	struct crypto_alg *q;
+
+	if (alg->cra_alignmask & (alg->cra_alignmask + 1))
+		return -EINVAL;
+
+	if (alg->cra_alignmask & alg->cra_blocksize)
+		return -EINVAL;
+
+	if (alg->cra_blocksize > PAGE_SIZE / 8)
+		return -EINVAL;
+
+	if (alg->cra_priority < 0)
+		return -EINVAL;
+	
+	ret = crypto_set_driver_name(alg);
+	if (unlikely(ret))
+		return ret;
+
+	down_write(&crypto_alg_sem);
+	
+	list_for_each_entry(q, &crypto_alg_list, cra_list) {
+		if (q == alg) {
+			ret = -EEXIST;
+			goto out;
+		}
+	}
+	
+	list_add(&alg->cra_list, &crypto_alg_list);
+	atomic_set(&alg->cra_refcnt, 1);
+out:	
+	up_write(&crypto_alg_sem);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(crypto_register_alg);
+
+int crypto_unregister_alg(struct crypto_alg *alg)
+{
+	int ret = -ENOENT;
+	struct crypto_alg *q;
+	
+	down_write(&crypto_alg_sem);
+	list_for_each_entry(q, &crypto_alg_list, cra_list) {
+		if (alg == q) {
+			list_del(&alg->cra_list);
+			ret = 0;
+			goto out;
+		}
+	}
+out:	
+	up_write(&crypto_alg_sem);
+
+	if (ret)
+		return ret;
+
+	BUG_ON(atomic_read(&alg->cra_refcnt) != 1);
+	if (alg->cra_destroy)
+		alg->cra_destroy(alg);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(crypto_unregister_alg);
+
+static int __init crypto_algapi_init(void)
+{
+	crypto_init_proc();
+	return 0;
+}
+
+static void __exit crypto_algapi_exit(void)
+{
+	crypto_exit_proc();
+}
+
+module_init(crypto_algapi_init);
+module_exit(crypto_algapi_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Cryptographic algorithms API");
diff --git a/crypto/api.c b/crypto/api.c
index 5994a58ef954..c922090b4842 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -15,19 +15,17 @@
  *
  */
 
-#include <linux/compiler.h>
-#include <linux/init.h>
-#include <linux/crypto.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/kmod.h>
-#include <linux/rwsem.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include "internal.h"
 
 LIST_HEAD(crypto_alg_list);
+EXPORT_SYMBOL_GPL(crypto_alg_list);
 DECLARE_RWSEM(crypto_alg_sem);
+EXPORT_SYMBOL_GPL(crypto_alg_sem);
 
 static inline struct crypto_alg *crypto_alg_get(struct crypto_alg *alg)
 {
@@ -239,86 +237,6 @@ void crypto_free_tfm(struct crypto_tfm *tfm)
 	kfree(tfm);
 }
 
-static inline int crypto_set_driver_name(struct crypto_alg *alg)
-{
-	static const char suffix[] = "-generic";
-	char *driver_name = alg->cra_driver_name;
-	int len;
-
-	if (*driver_name)
-		return 0;
-
-	len = strlcpy(driver_name, alg->cra_name, CRYPTO_MAX_ALG_NAME);
-	if (len + sizeof(suffix) > CRYPTO_MAX_ALG_NAME)
-		return -ENAMETOOLONG;
-
-	memcpy(driver_name + len, suffix, sizeof(suffix));
-	return 0;
-}
-
-int crypto_register_alg(struct crypto_alg *alg)
-{
-	int ret;
-	struct crypto_alg *q;
-
-	if (alg->cra_alignmask & (alg->cra_alignmask + 1))
-		return -EINVAL;
-
-	if (alg->cra_alignmask & alg->cra_blocksize)
-		return -EINVAL;
-
-	if (alg->cra_blocksize > PAGE_SIZE / 8)
-		return -EINVAL;
-
-	if (alg->cra_priority < 0)
-		return -EINVAL;
-	
-	ret = crypto_set_driver_name(alg);
-	if (unlikely(ret))
-		return ret;
-
-	down_write(&crypto_alg_sem);
-	
-	list_for_each_entry(q, &crypto_alg_list, cra_list) {
-		if (q == alg) {
-			ret = -EEXIST;
-			goto out;
-		}
-	}
-	
-	list_add(&alg->cra_list, &crypto_alg_list);
-	atomic_set(&alg->cra_refcnt, 1);
-out:	
-	up_write(&crypto_alg_sem);
-	return ret;
-}
-
-int crypto_unregister_alg(struct crypto_alg *alg)
-{
-	int ret = -ENOENT;
-	struct crypto_alg *q;
-	
-	down_write(&crypto_alg_sem);
-	list_for_each_entry(q, &crypto_alg_list, cra_list) {
-		if (alg == q) {
-			list_del(&alg->cra_list);
-			ret = 0;
-			goto out;
-		}
-	}
-out:	
-	up_write(&crypto_alg_sem);
-
-	if (ret)
-		return ret;
-
-	BUG_ON(atomic_read(&alg->cra_refcnt) != 1);
-	if (alg->cra_destroy)
-		alg->cra_destroy(alg);
-
-	return 0;
-}
-
 int crypto_alg_available(const char *name, u32 flags)
 {
 	int ret = 0;
@@ -332,17 +250,6 @@ int crypto_alg_available(const char *name, u32 flags)
 	return ret;
 }
 
-static int __init init_crypto(void)
-{
-	printk(KERN_INFO "Initializing Cryptographic API\n");
-	crypto_init_proc();
-	return 0;
-}
-
-__initcall(init_crypto);
-
-EXPORT_SYMBOL_GPL(crypto_register_alg);
-EXPORT_SYMBOL_GPL(crypto_unregister_alg);
 EXPORT_SYMBOL_GPL(crypto_alloc_tfm);
 EXPORT_SYMBOL_GPL(crypto_free_tfm);
 EXPORT_SYMBOL_GPL(crypto_alg_available);
diff --git a/crypto/internal.h b/crypto/internal.h
index 959e602909a6..26f47d331551 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -12,7 +12,8 @@
  */
 #ifndef _CRYPTO_INTERNAL_H
 #define _CRYPTO_INTERNAL_H
-#include <linux/crypto.h>
+
+#include <crypto/algapi.h>
 #include <linux/mm.h>
 #include <linux/highmem.h>
 #include <linux/interrupt.h>
@@ -64,9 +65,12 @@ static inline void crypto_free_hmac_block(struct crypto_tfm *tfm)
 
 #ifdef CONFIG_PROC_FS
 void __init crypto_init_proc(void);
+void __exit crypto_exit_proc(void);
 #else
 static inline void crypto_init_proc(void)
 { }
+static inline void crypto_exit_proc(void)
+{ }
 #endif
 
 static inline unsigned int crypto_digest_ctxsize(struct crypto_alg *alg,
diff --git a/crypto/proc.c b/crypto/proc.c
index 8543b7a157d6..9e573b17e887 100644
--- a/crypto/proc.c
+++ b/crypto/proc.c
@@ -113,3 +113,8 @@ void __init crypto_init_proc(void)
 	if (proc)
 		proc->proc_fops = &proc_crypto_ops;
 }
+
+void __exit crypto_exit_proc(void)
+{
+	remove_proc_entry("crypto", NULL);
+}
diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig
index 4263935443cc..ba23683ab8c4 100644
--- a/drivers/crypto/Kconfig
+++ b/drivers/crypto/Kconfig
@@ -2,7 +2,8 @@ menu "Hardware crypto devices"
 
 config CRYPTO_DEV_PADLOCK
 	tristate "Support for VIA PadLock ACE"
-	depends on CRYPTO && X86_32
+	depends on X86_32
+	select CRYPTO_ALGAPI
 	help
 	  Some VIA processors come with an integrated crypto engine
 	  (so called VIA PadLock ACE, Advanced Cryptography Engine)
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
new file mode 100644
index 000000000000..ed68d494b364
--- /dev/null
+++ b/include/crypto/algapi.h
@@ -0,0 +1,18 @@
+/*
+ * Cryptographic API for algorithms (i.e., low-level API).
+ *
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+#ifndef _CRYPTO_ALGAPI_H
+#define _CRYPTO_ALGAPI_H
+
+#include <linux/crypto.h>
+
+#endif	/* _CRYPTO_ALGAPI_H */
+
-- 
cgit v1.2.3


From 4cc7720cd165273b08a72b4193146dffee58e34b Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 6 Aug 2006 21:16:34 +1000
Subject: [CRYPTO] api: Add template registration

A crypto_template generates a crypto_alg object when given a set of
parameters.  this patch adds the basic data structure fo templates
and code to handle their registration/deregistration.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/algapi.c         | 156 +++++++++++++++++++++++++++++++++++++++++-------
 crypto/internal.h       |  17 ++++++
 include/crypto/algapi.h |  31 ++++++++++
 3 files changed, 182 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/crypto/algapi.c b/crypto/algapi.c
index a65c6ccfbe17..232b37d81613 100644
--- a/crypto/algapi.c
+++ b/crypto/algapi.c
@@ -13,11 +13,14 @@
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/module.h>
 #include <linux/string.h>
 
 #include "internal.h"
 
+static LIST_HEAD(crypto_template_list);
+
 static inline int crypto_set_driver_name(struct crypto_alg *alg)
 {
 	static const char suffix[] = "-generic";
@@ -35,11 +38,8 @@ static inline int crypto_set_driver_name(struct crypto_alg *alg)
 	return 0;
 }
 
-int crypto_register_alg(struct crypto_alg *alg)
+static int crypto_check_alg(struct crypto_alg *alg)
 {
-	int ret;
-	struct crypto_alg *q;
-
 	if (alg->cra_alignmask & (alg->cra_alignmask + 1))
 		return -EINVAL;
 
@@ -51,42 +51,52 @@ int crypto_register_alg(struct crypto_alg *alg)
 
 	if (alg->cra_priority < 0)
 		return -EINVAL;
-	
-	ret = crypto_set_driver_name(alg);
-	if (unlikely(ret))
-		return ret;
 
-	down_write(&crypto_alg_sem);
-	
+	return crypto_set_driver_name(alg);
+}
+
+static int __crypto_register_alg(struct crypto_alg *alg)
+{
+	struct crypto_alg *q;
+	int ret = -EEXIST;
+
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
-		if (q == alg) {
-			ret = -EEXIST;
+		if (q == alg)
 			goto out;
-		}
 	}
 	
 	list_add(&alg->cra_list, &crypto_alg_list);
 	atomic_set(&alg->cra_refcnt, 1);
+	ret = 0;
 out:	
-	up_write(&crypto_alg_sem);
 	return ret;
 }
+
+int crypto_register_alg(struct crypto_alg *alg)
+{
+	int err;
+
+	err = crypto_check_alg(alg);
+	if (err)
+		return err;
+
+	down_write(&crypto_alg_sem);
+	err = __crypto_register_alg(alg);
+	up_write(&crypto_alg_sem);
+
+	return err;
+}
 EXPORT_SYMBOL_GPL(crypto_register_alg);
 
 int crypto_unregister_alg(struct crypto_alg *alg)
 {
 	int ret = -ENOENT;
-	struct crypto_alg *q;
 	
 	down_write(&crypto_alg_sem);
-	list_for_each_entry(q, &crypto_alg_list, cra_list) {
-		if (alg == q) {
-			list_del(&alg->cra_list);
-			ret = 0;
-			goto out;
-		}
+	if (likely(!list_empty(&alg->cra_list))) {
+		list_del_init(&alg->cra_list);
+		ret = 0;
 	}
-out:	
 	up_write(&crypto_alg_sem);
 
 	if (ret)
@@ -100,6 +110,108 @@ out:
 }
 EXPORT_SYMBOL_GPL(crypto_unregister_alg);
 
+int crypto_register_template(struct crypto_template *tmpl)
+{
+	struct crypto_template *q;
+	int err = -EEXIST;
+
+	down_write(&crypto_alg_sem);
+
+	list_for_each_entry(q, &crypto_template_list, list) {
+		if (q == tmpl)
+			goto out;
+	}
+
+	list_add(&tmpl->list, &crypto_template_list);
+	err = 0;
+out:
+	up_write(&crypto_alg_sem);
+	return err;
+}
+EXPORT_SYMBOL_GPL(crypto_register_template);
+
+void crypto_unregister_template(struct crypto_template *tmpl)
+{
+	struct crypto_instance *inst;
+	struct hlist_node *p, *n;
+	struct hlist_head *list;
+
+	down_write(&crypto_alg_sem);
+
+	BUG_ON(list_empty(&tmpl->list));
+	list_del_init(&tmpl->list);
+
+	list = &tmpl->instances;
+	hlist_for_each_entry(inst, p, list, list) {
+		BUG_ON(list_empty(&inst->alg.cra_list));
+		list_del_init(&inst->alg.cra_list);
+	}
+
+	up_write(&crypto_alg_sem);
+
+	hlist_for_each_entry_safe(inst, p, n, list, list) {
+		BUG_ON(atomic_read(&inst->alg.cra_refcnt) != 1);
+		tmpl->free(inst);
+	}
+}
+EXPORT_SYMBOL_GPL(crypto_unregister_template);
+
+static struct crypto_template *__crypto_lookup_template(const char *name)
+{
+	struct crypto_template *q, *tmpl = NULL;
+
+	down_read(&crypto_alg_sem);
+	list_for_each_entry(q, &crypto_template_list, list) {
+		if (strcmp(q->name, name))
+			continue;
+		if (unlikely(!crypto_tmpl_get(q)))
+			continue;
+
+		tmpl = q;
+		break;
+	}
+	up_read(&crypto_alg_sem);
+
+	return tmpl;
+}
+
+struct crypto_template *crypto_lookup_template(const char *name)
+{
+	return try_then_request_module(__crypto_lookup_template(name), name);
+}
+EXPORT_SYMBOL_GPL(crypto_lookup_template);
+
+int crypto_register_instance(struct crypto_template *tmpl,
+			     struct crypto_instance *inst)
+{
+	int err = -EINVAL;
+
+	if (inst->alg.cra_destroy)
+		goto err;
+
+	err = crypto_check_alg(&inst->alg);
+	if (err)
+		goto err;
+
+	inst->alg.cra_module = tmpl->module;
+
+	down_write(&crypto_alg_sem);
+
+	err = __crypto_register_alg(&inst->alg);
+	if (err)
+		goto unlock;
+
+	hlist_add_head(&inst->list, &tmpl->instances);
+	inst->tmpl = tmpl;
+
+unlock:
+	up_write(&crypto_alg_sem);
+
+err:
+	return err;
+}
+EXPORT_SYMBOL_GPL(crypto_register_instance);
+
 static int __init crypto_algapi_init(void)
 {
 	crypto_init_proc();
diff --git a/crypto/internal.h b/crypto/internal.h
index 26f47d331551..c3ab4a950f30 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -19,11 +19,15 @@
 #include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/list.h>
+#include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/rwsem.h>
 #include <linux/slab.h>
 #include <asm/kmap_types.h>
 
+struct crypto_instance;
+struct crypto_template;
+
 extern struct list_head crypto_alg_list;
 extern struct rw_semaphore crypto_alg_sem;
 
@@ -112,5 +116,18 @@ void crypto_exit_digest_ops(struct crypto_tfm *tfm);
 void crypto_exit_cipher_ops(struct crypto_tfm *tfm);
 void crypto_exit_compress_ops(struct crypto_tfm *tfm);
 
+int crypto_register_instance(struct crypto_template *tmpl,
+			     struct crypto_instance *inst);
+
+static inline int crypto_tmpl_get(struct crypto_template *tmpl)
+{
+	return try_module_get(tmpl->module);
+}
+
+static inline void crypto_tmpl_put(struct crypto_template *tmpl)
+{
+	module_put(tmpl->module);
+}
+
 #endif	/* _CRYPTO_INTERNAL_H */
 
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index ed68d494b364..ffec530d52fb 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -14,5 +14,36 @@
 
 #include <linux/crypto.h>
 
+struct module;
+
+struct crypto_instance {
+	struct crypto_alg alg;
+
+	struct crypto_template *tmpl;
+	struct hlist_node list;
+
+	void *__ctx[] CRYPTO_MINALIGN_ATTR;
+};
+
+struct crypto_template {
+	struct list_head list;
+	struct hlist_head instances;
+	struct module *module;
+
+	struct crypto_instance *(*alloc)(void *param, unsigned int len);
+	void (*free)(struct crypto_instance *inst);
+
+	char name[CRYPTO_MAX_ALG_NAME];
+};
+
+int crypto_register_template(struct crypto_template *tmpl);
+void crypto_unregister_template(struct crypto_template *tmpl);
+struct crypto_template *crypto_lookup_template(const char *name);
+
+static inline void *crypto_instance_ctx(struct crypto_instance *inst)
+{
+	return inst->__ctx;
+}
+
 #endif	/* _CRYPTO_ALGAPI_H */
 
-- 
cgit v1.2.3


From 2825982d9d66ebba4b532a07391dfbb357f71c5f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 6 Aug 2006 21:23:26 +1000
Subject: [CRYPTO] api: Added event notification

This patch adds a notifier chain for algorithm/template registration events.
This will be used to register compound algorithms such as cbc(aes).  In
future this will also be passed onto user-space through netlink.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/algapi.c        |  50 +++++++++++++++++++-
 crypto/api.c           | 122 ++++++++++++++++++++++++++++++++++++++++++++-----
 crypto/internal.h      |  37 +++++++++++++++
 include/linux/crypto.h |   4 +-
 4 files changed, 199 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/crypto/algapi.c b/crypto/algapi.c
index 232b37d81613..f0df85fc1f50 100644
--- a/crypto/algapi.c
+++ b/crypto/algapi.c
@@ -21,6 +21,24 @@
 
 static LIST_HEAD(crypto_template_list);
 
+void crypto_larval_error(const char *name)
+{
+	struct crypto_alg *alg;
+
+	down_read(&crypto_alg_sem);
+	alg = __crypto_alg_lookup(name);
+	up_read(&crypto_alg_sem);
+
+	if (alg) {
+		if (crypto_is_larval(alg)) {
+			struct crypto_larval *larval = (void *)alg;
+			complete(&larval->completion);
+		}
+		crypto_mod_put(alg);
+	}
+}
+EXPORT_SYMBOL_GPL(crypto_larval_error);
+
 static inline int crypto_set_driver_name(struct crypto_alg *alg)
 {
 	static const char suffix[] = "-generic";
@@ -60,14 +78,27 @@ static int __crypto_register_alg(struct crypto_alg *alg)
 	struct crypto_alg *q;
 	int ret = -EEXIST;
 
+	atomic_set(&alg->cra_refcnt, 1);
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
 		if (q == alg)
 			goto out;
+		if (crypto_is_larval(q) &&
+		    (!strcmp(alg->cra_name, q->cra_name) ||
+		     !strcmp(alg->cra_driver_name, q->cra_name))) {
+			struct crypto_larval *larval = (void *)q;
+
+			if (!crypto_mod_get(alg))
+				continue;
+			larval->adult = alg;
+			complete(&larval->completion);
+		}
 	}
 	
 	list_add(&alg->cra_list, &crypto_alg_list);
-	atomic_set(&alg->cra_refcnt, 1);
+
+	crypto_notify(CRYPTO_MSG_ALG_REGISTER, alg);
 	ret = 0;
+
 out:	
 	return ret;
 }
@@ -97,6 +128,7 @@ int crypto_unregister_alg(struct crypto_alg *alg)
 		list_del_init(&alg->cra_list);
 		ret = 0;
 	}
+	crypto_notify(CRYPTO_MSG_ALG_UNREGISTER, alg);
 	up_write(&crypto_alg_sem);
 
 	if (ret)
@@ -123,6 +155,7 @@ int crypto_register_template(struct crypto_template *tmpl)
 	}
 
 	list_add(&tmpl->list, &crypto_template_list);
+	crypto_notify(CRYPTO_MSG_TMPL_REGISTER, tmpl);
 	err = 0;
 out:
 	up_write(&crypto_alg_sem);
@@ -145,8 +178,11 @@ void crypto_unregister_template(struct crypto_template *tmpl)
 	hlist_for_each_entry(inst, p, list, list) {
 		BUG_ON(list_empty(&inst->alg.cra_list));
 		list_del_init(&inst->alg.cra_list);
+		crypto_notify(CRYPTO_MSG_ALG_UNREGISTER, &inst->alg);
 	}
 
+	crypto_notify(CRYPTO_MSG_TMPL_UNREGISTER, tmpl);
+
 	up_write(&crypto_alg_sem);
 
 	hlist_for_each_entry_safe(inst, p, n, list, list) {
@@ -212,6 +248,18 @@ err:
 }
 EXPORT_SYMBOL_GPL(crypto_register_instance);
 
+int crypto_register_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&crypto_chain, nb);
+}
+EXPORT_SYMBOL_GPL(crypto_register_notifier);
+
+int crypto_unregister_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&crypto_chain, nb);
+}
+EXPORT_SYMBOL_GPL(crypto_unregister_notifier);
+
 static int __init crypto_algapi_init(void)
 {
 	crypto_init_proc();
diff --git a/crypto/api.c b/crypto/api.c
index c922090b4842..5a0d6a17cfd7 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -18,6 +18,7 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/kmod.h>
+#include <linux/param.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include "internal.h"
@@ -27,6 +28,9 @@ EXPORT_SYMBOL_GPL(crypto_alg_list);
 DECLARE_RWSEM(crypto_alg_sem);
 EXPORT_SYMBOL_GPL(crypto_alg_sem);
 
+BLOCKING_NOTIFIER_HEAD(crypto_chain);
+EXPORT_SYMBOL_GPL(crypto_chain);
+
 static inline struct crypto_alg *crypto_alg_get(struct crypto_alg *alg)
 {
 	atomic_inc(&alg->cra_refcnt);
@@ -39,27 +43,24 @@ static inline void crypto_alg_put(struct crypto_alg *alg)
 		alg->cra_destroy(alg);
 }
 
-static struct crypto_alg *crypto_mod_get(struct crypto_alg *alg)
+struct crypto_alg *crypto_mod_get(struct crypto_alg *alg)
 {
 	return try_module_get(alg->cra_module) ? crypto_alg_get(alg) : NULL;
 }
+EXPORT_SYMBOL_GPL(crypto_mod_get);
 
-static void crypto_mod_put(struct crypto_alg *alg)
+void crypto_mod_put(struct crypto_alg *alg)
 {
 	crypto_alg_put(alg);
 	module_put(alg->cra_module);
 }
+EXPORT_SYMBOL_GPL(crypto_mod_put);
 
-static struct crypto_alg *crypto_alg_lookup(const char *name)
+struct crypto_alg *__crypto_alg_lookup(const char *name)
 {
 	struct crypto_alg *q, *alg = NULL;
-	int best = -1;
+	int best = -2;
 
-	if (!name)
-		return NULL;
-	
-	down_read(&crypto_alg_sem);
-	
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
 		int exact, fuzzy;
 
@@ -79,16 +80,113 @@ static struct crypto_alg *crypto_alg_lookup(const char *name)
 		if (exact)
 			break;
 	}
-	
+
+	return alg;
+}
+EXPORT_SYMBOL_GPL(__crypto_alg_lookup);
+
+static void crypto_larval_destroy(struct crypto_alg *alg)
+{
+	struct crypto_larval *larval = (void *)alg;
+
+	BUG_ON(!crypto_is_larval(alg));
+	if (larval->adult)
+		crypto_mod_put(larval->adult);
+	kfree(larval);
+}
+
+static struct crypto_alg *crypto_larval_alloc(const char *name)
+{
+	struct crypto_alg *alg;
+	struct crypto_larval *larval;
+
+	larval = kzalloc(sizeof(*larval), GFP_KERNEL);
+	if (!larval)
+		return NULL;
+
+	larval->alg.cra_flags = CRYPTO_ALG_LARVAL;
+	larval->alg.cra_priority = -1;
+	larval->alg.cra_destroy = crypto_larval_destroy;
+
+	atomic_set(&larval->alg.cra_refcnt, 2);
+	strlcpy(larval->alg.cra_name, name, CRYPTO_MAX_ALG_NAME);
+	init_completion(&larval->completion);
+
+	down_write(&crypto_alg_sem);
+	alg = __crypto_alg_lookup(name);
+	if (!alg) {
+		alg = &larval->alg;
+		list_add(&alg->cra_list, &crypto_alg_list);
+	}
+	up_write(&crypto_alg_sem);
+
+	if (alg != &larval->alg)
+		kfree(larval);
+
+	return alg;
+}
+
+static void crypto_larval_kill(struct crypto_alg *alg)
+{
+	struct crypto_larval *larval = (void *)alg;
+
+	down_write(&crypto_alg_sem);
+	list_del(&alg->cra_list);
+	up_write(&crypto_alg_sem);
+	complete(&larval->completion);
+	crypto_alg_put(alg);
+}
+
+static struct crypto_alg *crypto_larval_wait(struct crypto_alg *alg)
+{
+	struct crypto_larval *larval = (void *)alg;
+
+	wait_for_completion_interruptible_timeout(&larval->completion, 60 * HZ);
+	alg = larval->adult;
+	if (alg && !crypto_mod_get(alg))
+		alg = NULL;
+	crypto_mod_put(&larval->alg);
+
+	return alg;
+}
+
+static struct crypto_alg *crypto_alg_lookup(const char *name)
+{
+	struct crypto_alg *alg;
+
+	if (!name)
+		return NULL;
+
+	down_read(&crypto_alg_sem);
+	alg = __crypto_alg_lookup(name);
 	up_read(&crypto_alg_sem);
+
 	return alg;
 }
 
 /* A far more intelligent version of this is planned.  For now, just
  * try an exact match on the name of the algorithm. */
-static inline struct crypto_alg *crypto_alg_mod_lookup(const char *name)
+static struct crypto_alg *crypto_alg_mod_lookup(const char *name)
 {
-	return try_then_request_module(crypto_alg_lookup(name), name);
+	struct crypto_alg *alg;
+	struct crypto_alg *larval;
+
+	alg = try_then_request_module(crypto_alg_lookup(name), name);
+	if (alg)
+		return crypto_is_larval(alg) ? crypto_larval_wait(alg) : alg;
+
+	larval = crypto_larval_alloc(name);
+	if (!larval || !crypto_is_larval(larval))
+		return larval;
+
+	if (crypto_notify(CRYPTO_MSG_ALG_REQUEST, larval) == NOTIFY_STOP)
+		alg = crypto_larval_wait(larval);
+	else {
+		crypto_mod_put(larval);
+		alg = NULL;
+	}
+	crypto_larval_kill(larval);
+	return alg;
 }
 
 static int crypto_init_flags(struct crypto_tfm *tfm, u32 flags)
diff --git a/crypto/internal.h b/crypto/internal.h
index c3ab4a950f30..3a08d25fba45 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -14,6 +14,7 @@
 #define _CRYPTO_INTERNAL_H
 
 #include <crypto/algapi.h>
+#include <linux/completion.h>
 #include <linux/mm.h>
 #include <linux/highmem.h>
 #include <linux/interrupt.h>
@@ -21,15 +22,32 @@
 #include <linux/list.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
+#include <linux/notifier.h>
 #include <linux/rwsem.h>
 #include <linux/slab.h>
 #include <asm/kmap_types.h>
 
+/* Crypto notification events. */
+enum {
+	CRYPTO_MSG_ALG_REQUEST,
+	CRYPTO_MSG_ALG_REGISTER,
+	CRYPTO_MSG_ALG_UNREGISTER,
+	CRYPTO_MSG_TMPL_REGISTER,
+	CRYPTO_MSG_TMPL_UNREGISTER,
+};
+
 struct crypto_instance;
 struct crypto_template;
 
+struct crypto_larval {
+	struct crypto_alg alg;
+	struct crypto_alg *adult;
+	struct completion completion;
+};
+
 extern struct list_head crypto_alg_list;
 extern struct rw_semaphore crypto_alg_sem;
+extern struct blocking_notifier_head crypto_chain;
 
 extern enum km_type crypto_km_types[];
 
@@ -104,6 +122,10 @@ static inline unsigned int crypto_compress_ctxsize(struct crypto_alg *alg,
 	return alg->cra_ctxsize;
 }
 
+struct crypto_alg *crypto_mod_get(struct crypto_alg *alg);
+void crypto_mod_put(struct crypto_alg *alg);
+struct crypto_alg *__crypto_alg_lookup(const char *name);
+
 int crypto_init_digest_flags(struct crypto_tfm *tfm, u32 flags);
 int crypto_init_cipher_flags(struct crypto_tfm *tfm, u32 flags);
 int crypto_init_compress_flags(struct crypto_tfm *tfm, u32 flags);
@@ -116,9 +138,14 @@ void crypto_exit_digest_ops(struct crypto_tfm *tfm);
 void crypto_exit_cipher_ops(struct crypto_tfm *tfm);
 void crypto_exit_compress_ops(struct crypto_tfm *tfm);
 
+void crypto_larval_error(const char *name);
+
 int crypto_register_instance(struct crypto_template *tmpl,
 			     struct crypto_instance *inst);
 
+int crypto_register_notifier(struct notifier_block *nb);
+int crypto_unregister_notifier(struct notifier_block *nb);
+
 static inline int crypto_tmpl_get(struct crypto_template *tmpl)
 {
 	return try_module_get(tmpl->module);
@@ -129,5 +156,15 @@ static inline void crypto_tmpl_put(struct crypto_template *tmpl)
 	module_put(tmpl->module);
 }
 
+static inline int crypto_is_larval(struct crypto_alg *alg)
+{
+	return alg->cra_flags & CRYPTO_ALG_LARVAL;
+}
+
+static inline int crypto_notify(unsigned long val, void *v)
+{
+	return blocking_notifier_call_chain(&crypto_chain, val, v);
+}
+
 #endif	/* _CRYPTO_INTERNAL_H */
 
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 7f57ff8ec975..3e3e95aff133 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -29,11 +29,13 @@
 /*
  * Algorithm masks and types.
  */
-#define CRYPTO_ALG_TYPE_MASK		0x000000ff
+#define CRYPTO_ALG_TYPE_MASK		0x0000000f
 #define CRYPTO_ALG_TYPE_CIPHER		0x00000001
 #define CRYPTO_ALG_TYPE_DIGEST		0x00000002
 #define CRYPTO_ALG_TYPE_COMPRESS	0x00000004
 
+#define CRYPTO_ALG_LARVAL		0x00000010
+
 /*
  * Transform masks and values (for crt_flags).
  */
-- 
cgit v1.2.3


From 2b8c19dbdc692e81243a328725a02efb77b144a5 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 21 Sep 2006 11:31:44 +1000
Subject: [CRYPTO] api: Add cryptomgr

The cryptomgr module is a simple manager of crypto algorithm instances.
It ensures that parameterised algorithms of the type tmpl(alg) (e.g.,
cbc(aes)) are always created.

This is meant to satisfy the needs for most users.  For more complex
cases such as deeper combinations or multiple parameters, a netlink
module will be created which allows arbitrary expressions to be parsed
in user-space.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/Kconfig         |   8 +++
 crypto/Makefile        |   1 +
 crypto/api.c           |  10 +++-
 crypto/cryptomgr.c     | 146 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/crypto.h |   9 +++
 5 files changed, 173 insertions(+), 1 deletion(-)
 create mode 100644 crypto/cryptomgr.c

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index aabc63195222..4ce509dba329 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -16,6 +16,14 @@ config CRYPTO_ALGAPI
 	help
 	  This option provides the API for cryptographic algorithms.
 
+config CRYPTO_MANAGER
+	tristate "Cryptographic algorithm manager"
+	select CRYPTO_ALGAPI
+	default m
+	help
+	  Create default cryptographic template instantiations such as
+	  cbc(aes).
+
 config CRYPTO_HMAC
 	bool "HMAC support"
 	help
diff --git a/crypto/Makefile b/crypto/Makefile
index 6d51f80753a1..b8745f3d3595 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -8,6 +8,7 @@ crypto_algapi-$(CONFIG_PROC_FS) += proc.o
 crypto_algapi-objs := algapi.o $(crypto_algapi-y)
 obj-$(CONFIG_CRYPTO_ALGAPI) += crypto_algapi.o
 
+obj-$(CONFIG_CRYPTO_MANAGER) += cryptomgr.o
 obj-$(CONFIG_CRYPTO_HMAC) += hmac.o
 obj-$(CONFIG_CRYPTO_NULL) += crypto_null.o
 obj-$(CONFIG_CRYPTO_MD4) += md4.o
diff --git a/crypto/api.c b/crypto/api.c
index 5a0d6a17cfd7..67cd6f87b74a 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -18,6 +18,7 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/kmod.h>
+#include <linux/module.h>
 #include <linux/param.h>
 #include <linux/slab.h>
 #include <linux/string.h>
@@ -170,6 +171,7 @@ static struct crypto_alg *crypto_alg_mod_lookup(const char *name)
 {
 	struct crypto_alg *alg;
 	struct crypto_alg *larval;
+	int ok;
 
 	alg = try_then_request_module(crypto_alg_lookup(name), name);
 	if (alg)
@@ -179,7 +181,13 @@ static struct crypto_alg *crypto_alg_mod_lookup(const char *name)
 	if (!larval || !crypto_is_larval(larval))
 		return larval;
 
-	if (crypto_notify(CRYPTO_MSG_ALG_REQUEST, larval) == NOTIFY_STOP)
+	ok = crypto_notify(CRYPTO_MSG_ALG_REQUEST, larval);
+	if (ok == NOTIFY_DONE) {
+		request_module("cryptomgr");
+		ok = crypto_notify(CRYPTO_MSG_ALG_REQUEST, larval);
+	}
+
+	if (ok == NOTIFY_STOP)
 		alg = crypto_larval_wait(larval);
 	else {
 		crypto_mod_put(larval);
diff --git a/crypto/cryptomgr.c b/crypto/cryptomgr.c
new file mode 100644
index 000000000000..e0ebe1b44b99
--- /dev/null
+++ b/crypto/cryptomgr.c
@@ -0,0 +1,146 @@
+/*
+ * Create default crypto algorithm instances.
+ *
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <linux/crypto.h>
+#include <linux/ctype.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
+#include <linux/rtnetlink.h>
+#include <linux/string.h>
+#include <linux/workqueue.h>
+
+#include "internal.h"
+
+struct cryptomgr_param {
+	struct work_struct work;
+
+	struct {
+		struct rtattr attr;
+		struct crypto_attr_alg data;
+	} alg;
+
+	struct {
+		char name[CRYPTO_MAX_ALG_NAME];
+	} larval;
+
+	char template[CRYPTO_MAX_ALG_NAME];
+};
+
+static void cryptomgr_probe(void *data)
+{
+	struct cryptomgr_param *param = data;
+	struct crypto_template *tmpl;
+	struct crypto_instance *inst;
+
+	tmpl = crypto_lookup_template(param->template);
+	if (!tmpl)
+		goto err;
+
+	inst = tmpl->alloc(&param->alg, sizeof(param->alg));
+	if (IS_ERR(inst))
+		goto err;
+	else if ((err = crypto_register_instance(tmpl, inst))) {
+		tmpl->free(inst);
+		goto err;
+	}
+
+	crypto_tmpl_put(tmpl);
+
+out:
+	kfree(param);
+	return;
+
+err:
+	crypto_larval_error(param->larval.name);
+	goto out;
+}
+
+static int cryptomgr_schedule_probe(struct crypto_larval *larval)
+{
+	struct cryptomgr_param *param;
+	const char *name = larval->alg.cra_name;
+	const char *p;
+	unsigned int len;
+
+	param = kmalloc(sizeof(*param), GFP_KERNEL);
+	if (!param)
+		goto err;
+
+	for (p = name; isalnum(*p) || *p == '-' || *p == '_'; p++)
+		;
+
+	len = p - name;
+	if (!len || *p != '(')
+		goto err_free_param;
+
+	memcpy(param->template, name, len);
+	param->template[len] = 0;
+
+	name = p + 1;
+	for (p = name; isalnum(*p) || *p == '-' || *p == '_'; p++)
+		;
+
+	len = p - name;
+	if (!len || *p != ')' || p[1])
+		goto err_free_param;
+
+	param->alg.attr.rta_len = sizeof(param->alg);
+	param->alg.attr.rta_type = CRYPTOA_ALG;
+	memcpy(param->alg.data.name, name, len);
+	param->alg.data.name[len] = 0;
+
+	memcpy(param->larval.name, larval->alg.cra_name, CRYPTO_MAX_ALG_NAME);
+
+	INIT_WORK(&param->work, cryptomgr_probe, param);
+	schedule_work(&param->work);
+
+	return NOTIFY_STOP;
+
+err_free_param:
+	kfree(param);
+err:
+	return NOTIFY_OK;
+}
+
+static int cryptomgr_notify(struct notifier_block *this, unsigned long msg,
+			    void *data)
+{
+	switch (msg) {
+	case CRYPTO_MSG_ALG_REQUEST:
+		return cryptomgr_schedule_probe(data);
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block cryptomgr_notifier = {
+	.notifier_call = cryptomgr_notify,
+};
+
+static int __init cryptomgr_init(void)
+{
+	return crypto_register_notifier(&cryptomgr_notifier);
+}
+
+static void __exit cryptomgr_exit(void)
+{
+	int err = crypto_unregister_notifier(&cryptomgr_notifier);
+	BUG_ON(err);
+}
+
+module_init(cryptomgr_init);
+module_exit(cryptomgr_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Crypto Algorithm Manager");
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 3e3e95aff133..85f73c381913 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -260,6 +260,15 @@ struct crypto_tfm {
 	void *__crt_ctx[] CRYPTO_MINALIGN_ATTR;
 };
 
+enum {
+	CRYPTOA_UNSPEC,
+	CRYPTOA_ALG,
+};
+
+struct crypto_attr_alg {
+	char name[CRYPTO_MAX_ALG_NAME];
+};
+
 /* 
  * Transform user interface.
  */
-- 
cgit v1.2.3


From 6bfd48096ff8ecabf955958b51ddfa7988eb0a14 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 21 Sep 2006 11:39:29 +1000
Subject: [CRYPTO] api: Added spawns

Spawns lock a specific crypto algorithm in place.  They can then be used
with crypto_spawn_tfm to allocate a tfm for that algorithm.  When the base
algorithm of a spawn is deregistered, all its spawns will be automatically
removed.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/algapi.c         | 185 +++++++++++++++++++++++++++++++++++++++++++-----
 crypto/api.c            |  95 +++++++++++++++++--------
 crypto/cryptomgr.c      |  19 +++--
 crypto/internal.h       |  19 +++++
 include/crypto/algapi.h |  11 +++
 include/linux/crypto.h  |   4 ++
 6 files changed, 280 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/crypto/algapi.c b/crypto/algapi.c
index acea250677c0..36c4f1bdb521 100644
--- a/crypto/algapi.c
+++ b/crypto/algapi.c
@@ -10,6 +10,7 @@
  *
  */
 
+#include <linux/err.h>
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -73,27 +74,96 @@ static int crypto_check_alg(struct crypto_alg *alg)
 	return crypto_set_driver_name(alg);
 }
 
-static int __crypto_register_alg(struct crypto_alg *alg)
+static void crypto_destroy_instance(struct crypto_alg *alg)
+{
+	struct crypto_instance *inst = (void *)alg;
+	struct crypto_template *tmpl = inst->tmpl;
+
+	tmpl->free(inst);
+	crypto_tmpl_put(tmpl);
+}
+
+static void crypto_remove_spawns(struct list_head *spawns,
+				 struct list_head *list)
+{
+	struct crypto_spawn *spawn, *n;
+
+	list_for_each_entry_safe(spawn, n, spawns, list) {
+		struct crypto_instance *inst = spawn->inst;
+		struct crypto_template *tmpl = inst->tmpl;
+
+		list_del_init(&spawn->list);
+		spawn->alg = NULL;
+
+		if (crypto_is_dead(&inst->alg))
+			continue;
+
+		inst->alg.cra_flags |= CRYPTO_ALG_DEAD;
+		if (!tmpl || !crypto_tmpl_get(tmpl))
+			continue;
+
+		crypto_notify(CRYPTO_MSG_ALG_UNREGISTER, &inst->alg);
+		list_move(&inst->alg.cra_list, list);
+		hlist_del(&inst->list);
+		inst->alg.cra_destroy = crypto_destroy_instance;
+
+		if (!list_empty(&inst->alg.cra_users)) {
+			if (&n->list == spawns)
+				n = list_entry(inst->alg.cra_users.next,
+					       typeof(*n), list);
+			__list_splice(&inst->alg.cra_users, spawns->prev);
+		}
+	}
+}
+
+static int __crypto_register_alg(struct crypto_alg *alg,
+				 struct list_head *list)
 {
 	struct crypto_alg *q;
-	int ret = -EEXIST;
+	int ret = -EAGAIN;
+
+	if (crypto_is_dead(alg))
+		goto out;
+
+	INIT_LIST_HEAD(&alg->cra_users);
+
+	ret = -EEXIST;
 
 	atomic_set(&alg->cra_refcnt, 1);
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
 		if (q == alg)
 			goto out;
-		if (crypto_is_larval(q) &&
-		    (!strcmp(alg->cra_name, q->cra_name) ||
-		     !strcmp(alg->cra_driver_name, q->cra_name))) {
+
+		if (crypto_is_moribund(q))
+			continue;
+
+		if (crypto_is_larval(q)) {
 			struct crypto_larval *larval = (void *)q;
 
+			if (strcmp(alg->cra_name, q->cra_name) &&
+			    strcmp(alg->cra_driver_name, q->cra_name))
+				continue;
+
+			if (larval->adult)
+				continue;
 			if ((q->cra_flags ^ alg->cra_flags) & larval->mask)
 				continue;
 			if (!crypto_mod_get(alg))
 				continue;
+
 			larval->adult = alg;
 			complete(&larval->completion);
+			continue;
 		}
+
+		if (strcmp(alg->cra_name, q->cra_name))
+			continue;
+
+		if (strcmp(alg->cra_driver_name, q->cra_driver_name) &&
+		    q->cra_priority > alg->cra_priority)
+			continue;
+
+		crypto_remove_spawns(&q->cra_users, list);
 	}
 	
 	list_add(&alg->cra_list, &crypto_alg_list);
@@ -105,8 +175,20 @@ out:
 	return ret;
 }
 
+static void crypto_remove_final(struct list_head *list)
+{
+	struct crypto_alg *alg;
+	struct crypto_alg *n;
+
+	list_for_each_entry_safe(alg, n, list, cra_list) {
+		list_del_init(&alg->cra_list);
+		crypto_alg_put(alg);
+	}
+}
+
 int crypto_register_alg(struct crypto_alg *alg)
 {
+	LIST_HEAD(list);
 	int err;
 
 	err = crypto_check_alg(alg);
@@ -114,23 +196,35 @@ int crypto_register_alg(struct crypto_alg *alg)
 		return err;
 
 	down_write(&crypto_alg_sem);
-	err = __crypto_register_alg(alg);
+	err = __crypto_register_alg(alg, &list);
 	up_write(&crypto_alg_sem);
 
+	crypto_remove_final(&list);
 	return err;
 }
 EXPORT_SYMBOL_GPL(crypto_register_alg);
 
+static int crypto_remove_alg(struct crypto_alg *alg, struct list_head *list)
+{
+	if (unlikely(list_empty(&alg->cra_list)))
+		return -ENOENT;
+
+	alg->cra_flags |= CRYPTO_ALG_DEAD;
+
+	crypto_notify(CRYPTO_MSG_ALG_UNREGISTER, alg);
+	list_del_init(&alg->cra_list);
+	crypto_remove_spawns(&alg->cra_users, list);
+
+	return 0;
+}
+
 int crypto_unregister_alg(struct crypto_alg *alg)
 {
-	int ret = -ENOENT;
+	int ret;
+	LIST_HEAD(list);
 	
 	down_write(&crypto_alg_sem);
-	if (likely(!list_empty(&alg->cra_list))) {
-		list_del_init(&alg->cra_list);
-		ret = 0;
-	}
-	crypto_notify(CRYPTO_MSG_ALG_UNREGISTER, alg);
+	ret = crypto_remove_alg(alg, &list);
 	up_write(&crypto_alg_sem);
 
 	if (ret)
@@ -140,6 +234,7 @@ int crypto_unregister_alg(struct crypto_alg *alg)
 	if (alg->cra_destroy)
 		alg->cra_destroy(alg);
 
+	crypto_remove_final(&list);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(crypto_unregister_alg);
@@ -170,6 +265,7 @@ void crypto_unregister_template(struct crypto_template *tmpl)
 	struct crypto_instance *inst;
 	struct hlist_node *p, *n;
 	struct hlist_head *list;
+	LIST_HEAD(users);
 
 	down_write(&crypto_alg_sem);
 
@@ -178,9 +274,8 @@ void crypto_unregister_template(struct crypto_template *tmpl)
 
 	list = &tmpl->instances;
 	hlist_for_each_entry(inst, p, list, list) {
-		BUG_ON(list_empty(&inst->alg.cra_list));
-		list_del_init(&inst->alg.cra_list);
-		crypto_notify(CRYPTO_MSG_ALG_UNREGISTER, &inst->alg);
+		int err = crypto_remove_alg(&inst->alg, &users);
+		BUG_ON(err);
 	}
 
 	crypto_notify(CRYPTO_MSG_TMPL_UNREGISTER, tmpl);
@@ -191,6 +286,7 @@ void crypto_unregister_template(struct crypto_template *tmpl)
 		BUG_ON(atomic_read(&inst->alg.cra_refcnt) != 1);
 		tmpl->free(inst);
 	}
+	crypto_remove_final(&users);
 }
 EXPORT_SYMBOL_GPL(crypto_unregister_template);
 
@@ -222,6 +318,7 @@ EXPORT_SYMBOL_GPL(crypto_lookup_template);
 int crypto_register_instance(struct crypto_template *tmpl,
 			     struct crypto_instance *inst)
 {
+	LIST_HEAD(list);
 	int err = -EINVAL;
 
 	if (inst->alg.cra_destroy)
@@ -235,7 +332,7 @@ int crypto_register_instance(struct crypto_template *tmpl,
 
 	down_write(&crypto_alg_sem);
 
-	err = __crypto_register_alg(&inst->alg);
+	err = __crypto_register_alg(&inst->alg, &list);
 	if (err)
 		goto unlock;
 
@@ -245,11 +342,67 @@ int crypto_register_instance(struct crypto_template *tmpl,
 unlock:
 	up_write(&crypto_alg_sem);
 
+	crypto_remove_final(&list);
+
 err:
 	return err;
 }
 EXPORT_SYMBOL_GPL(crypto_register_instance);
 
+int crypto_init_spawn(struct crypto_spawn *spawn, struct crypto_alg *alg,
+		      struct crypto_instance *inst)
+{
+	int err = -EAGAIN;
+
+	spawn->inst = inst;
+
+	down_write(&crypto_alg_sem);
+	if (!crypto_is_moribund(alg)) {
+		list_add(&spawn->list, &alg->cra_users);
+		spawn->alg = alg;
+		err = 0;
+	}
+	up_write(&crypto_alg_sem);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(crypto_init_spawn);
+
+void crypto_drop_spawn(struct crypto_spawn *spawn)
+{
+	down_write(&crypto_alg_sem);
+	list_del(&spawn->list);
+	up_write(&crypto_alg_sem);
+}
+EXPORT_SYMBOL_GPL(crypto_drop_spawn);
+
+struct crypto_tfm *crypto_spawn_tfm(struct crypto_spawn *spawn)
+{
+	struct crypto_alg *alg;
+	struct crypto_alg *alg2;
+	struct crypto_tfm *tfm;
+
+	down_read(&crypto_alg_sem);
+	alg = spawn->alg;
+	alg2 = alg;
+	if (alg2)
+		alg2 = crypto_mod_get(alg2);
+	up_read(&crypto_alg_sem);
+
+	if (!alg2) {
+		if (alg)
+			crypto_shoot_alg(alg);
+		return ERR_PTR(-EAGAIN);
+	}
+
+	tfm = __crypto_alloc_tfm(alg, 0);
+	if (IS_ERR(tfm))
+		crypto_mod_put(alg);
+
+	return tfm;
+}
+EXPORT_SYMBOL_GPL(crypto_spawn_tfm);
+
 int crypto_register_notifier(struct notifier_block *nb)
 {
 	return blocking_notifier_chain_register(&crypto_chain, nb);
diff --git a/crypto/api.c b/crypto/api.c
index ddf6a767acdd..7e5522cf856e 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -15,11 +15,13 @@
  *
  */
 
+#include <linux/err.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/kmod.h>
 #include <linux/module.h>
 #include <linux/param.h>
+#include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include "internal.h"
@@ -38,12 +40,6 @@ static inline struct crypto_alg *crypto_alg_get(struct crypto_alg *alg)
 	return alg;
 }
 
-static inline void crypto_alg_put(struct crypto_alg *alg)
-{
-	if (atomic_dec_and_test(&alg->cra_refcnt) && alg->cra_destroy)
-		alg->cra_destroy(alg);
-}
-
 struct crypto_alg *crypto_mod_get(struct crypto_alg *alg)
 {
 	return try_module_get(alg->cra_module) ? crypto_alg_get(alg) : NULL;
@@ -65,6 +61,9 @@ struct crypto_alg *__crypto_alg_lookup(const char *name, u32 type, u32 mask)
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
 		int exact, fuzzy;
 
+		if (crypto_is_moribund(q))
+			continue;
+
 		if ((q->cra_flags ^ type) & mask)
 			continue;
 
@@ -111,7 +110,7 @@ static struct crypto_alg *crypto_larval_alloc(const char *name, u32 type,
 
 	larval = kzalloc(sizeof(*larval), GFP_KERNEL);
 	if (!larval)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	larval->mask = mask;
 	larval->alg.cra_flags = CRYPTO_ALG_LARVAL | type;
@@ -153,8 +152,11 @@ static struct crypto_alg *crypto_larval_wait(struct crypto_alg *alg)
 
 	wait_for_completion_interruptible_timeout(&larval->completion, 60 * HZ);
 	alg = larval->adult;
-	if (alg && !crypto_mod_get(alg))
-		alg = NULL;
+	if (alg) {
+		if (!crypto_mod_get(alg))
+			alg = ERR_PTR(-EAGAIN);
+	} else
+		alg = ERR_PTR(-ENOENT);
 	crypto_mod_put(&larval->alg);
 
 	return alg;
@@ -165,9 +167,6 @@ static struct crypto_alg *crypto_alg_lookup(const char *name, u32 type,
 {
 	struct crypto_alg *alg;
 
-	if (!name)
-		return NULL;
-
 	down_read(&crypto_alg_sem);
 	alg = __crypto_alg_lookup(name, type, mask);
 	up_read(&crypto_alg_sem);
@@ -181,7 +180,10 @@ struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask)
 	struct crypto_alg *larval;
 	int ok;
 
-	mask &= ~CRYPTO_ALG_LARVAL;
+	if (!name)
+		return ERR_PTR(-ENOENT);
+
+	mask &= ~(CRYPTO_ALG_LARVAL | CRYPTO_ALG_DEAD);
 	type &= mask;
 
 	alg = try_then_request_module(crypto_alg_lookup(name, type, mask),
@@ -190,7 +192,7 @@ struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask)
 		return crypto_is_larval(alg) ? crypto_larval_wait(alg) : alg;
 
 	larval = crypto_larval_alloc(name, type, mask);
-	if (!larval || !crypto_is_larval(larval))
+	if (IS_ERR(larval) || !crypto_is_larval(larval))
 		return larval;
 
 	ok = crypto_notify(CRYPTO_MSG_ALG_REQUEST, larval);
@@ -203,7 +205,7 @@ struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask)
 		alg = crypto_larval_wait(larval);
 	else {
 		crypto_mod_put(larval);
-		alg = NULL;
+		alg = ERR_PTR(-ENOENT);
 	}
 	crypto_larval_kill(larval);
 	return alg;
@@ -298,31 +300,40 @@ static unsigned int crypto_ctxsize(struct crypto_alg *alg, int flags)
 	return len + (alg->cra_alignmask & ~(crypto_tfm_ctx_alignment() - 1));
 }
 
-struct crypto_tfm *crypto_alloc_tfm(const char *name, u32 flags)
+void crypto_shoot_alg(struct crypto_alg *alg)
+{
+	down_write(&crypto_alg_sem);
+	alg->cra_flags |= CRYPTO_ALG_DYING;
+	up_write(&crypto_alg_sem);
+}
+EXPORT_SYMBOL_GPL(crypto_shoot_alg);
+
+struct crypto_tfm *__crypto_alloc_tfm(struct crypto_alg *alg, u32 flags)
 {
 	struct crypto_tfm *tfm = NULL;
-	struct crypto_alg *alg;
 	unsigned int tfm_size;
-
-	alg = crypto_alg_mod_lookup(name, 0, 0);
-	if (alg == NULL)
-		goto out;
+	int err = -ENOMEM;
 
 	tfm_size = sizeof(*tfm) + crypto_ctxsize(alg, flags);
 	tfm = kzalloc(tfm_size, GFP_KERNEL);
 	if (tfm == NULL)
-		goto out_put;
+		goto out;
 
 	tfm->__crt_alg = alg;
-	
-	if (crypto_init_flags(tfm, flags))
+
+	err = crypto_init_flags(tfm, flags);
+	if (err)
 		goto out_free_tfm;
 		
-	if (crypto_init_ops(tfm))
+	err = crypto_init_ops(tfm);
+	if (err)
 		goto out_free_tfm;
 
-	if (alg->cra_init && alg->cra_init(tfm))
+	if (alg->cra_init && (err = alg->cra_init(tfm))) {
+		if (err == -EAGAIN)
+			crypto_shoot_alg(alg);
 		goto cra_init_failed;
+	}
 
 	goto out;
 
@@ -330,12 +341,36 @@ cra_init_failed:
 	crypto_exit_ops(tfm);
 out_free_tfm:
 	kfree(tfm);
-	tfm = NULL;
-out_put:
-	crypto_mod_put(alg);
+	tfm = ERR_PTR(err);
 out:
 	return tfm;
 }
+EXPORT_SYMBOL_GPL(__crypto_alloc_tfm);
+
+struct crypto_tfm *crypto_alloc_tfm(const char *name, u32 flags)
+{
+	struct crypto_tfm *tfm = NULL;
+	int err;
+
+	do {
+		struct crypto_alg *alg;
+
+		alg = crypto_alg_mod_lookup(name, 0, 0);
+		err = PTR_ERR(alg);
+		if (IS_ERR(alg))
+			continue;
+
+		tfm = __crypto_alloc_tfm(alg, flags);
+		err = 0;
+		if (IS_ERR(tfm)) {
+			crypto_mod_put(alg);
+			err = PTR_ERR(tfm);
+			tfm = NULL;
+		}
+	} while (err == -EAGAIN && !signal_pending(current));
+
+	return tfm;
+}
 
 void crypto_free_tfm(struct crypto_tfm *tfm)
 {
@@ -361,7 +396,7 @@ int crypto_alg_available(const char *name, u32 flags)
 	int ret = 0;
 	struct crypto_alg *alg = crypto_alg_mod_lookup(name, 0, 0);
 	
-	if (alg) {
+	if (!IS_ERR(alg)) {
 		crypto_mod_put(alg);
 		ret = 1;
 	}
diff --git a/crypto/cryptomgr.c b/crypto/cryptomgr.c
index ae54942e3b31..9b5b15601068 100644
--- a/crypto/cryptomgr.c
+++ b/crypto/cryptomgr.c
@@ -17,6 +17,7 @@
 #include <linux/module.h>
 #include <linux/notifier.h>
 #include <linux/rtnetlink.h>
+#include <linux/sched.h>
 #include <linux/string.h>
 #include <linux/workqueue.h>
 
@@ -44,21 +45,25 @@ static void cryptomgr_probe(void *data)
 	struct cryptomgr_param *param = data;
 	struct crypto_template *tmpl;
 	struct crypto_instance *inst;
+	int err;
 
 	tmpl = crypto_lookup_template(param->template);
 	if (!tmpl)
 		goto err;
 
-	inst = tmpl->alloc(&param->alg, sizeof(param->alg));
-	if (IS_ERR(inst))
-		goto err;
-	else if ((err = crypto_register_instance(tmpl, inst))) {
-		tmpl->free(inst);
-		goto err;
-	}
+	do {
+		inst = tmpl->alloc(&param->alg, sizeof(param->alg));
+		if (IS_ERR(inst))
+			err = PTR_ERR(inst);
+		else if ((err = crypto_register_instance(tmpl, inst)))
+			tmpl->free(inst);
+	} while (err == -EAGAIN && !signal_pending(current));
 
 	crypto_tmpl_put(tmpl);
 
+	if (err)
+		goto err;
+
 out:
 	kfree(param);
 	return;
diff --git a/crypto/internal.h b/crypto/internal.h
index c08d93bdadc4..03c00b0e6b60 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -142,12 +142,21 @@ void crypto_exit_compress_ops(struct crypto_tfm *tfm);
 
 void crypto_larval_error(const char *name, u32 type, u32 mask);
 
+void crypto_shoot_alg(struct crypto_alg *alg);
+struct crypto_tfm *__crypto_alloc_tfm(struct crypto_alg *alg, u32 flags);
+
 int crypto_register_instance(struct crypto_template *tmpl,
 			     struct crypto_instance *inst);
 
 int crypto_register_notifier(struct notifier_block *nb);
 int crypto_unregister_notifier(struct notifier_block *nb);
 
+static inline void crypto_alg_put(struct crypto_alg *alg)
+{
+	if (atomic_dec_and_test(&alg->cra_refcnt) && alg->cra_destroy)
+		alg->cra_destroy(alg);
+}
+
 static inline int crypto_tmpl_get(struct crypto_template *tmpl)
 {
 	return try_module_get(tmpl->module);
@@ -163,6 +172,16 @@ static inline int crypto_is_larval(struct crypto_alg *alg)
 	return alg->cra_flags & CRYPTO_ALG_LARVAL;
 }
 
+static inline int crypto_is_dead(struct crypto_alg *alg)
+{
+	return alg->cra_flags & CRYPTO_ALG_DEAD;
+}
+
+static inline int crypto_is_moribund(struct crypto_alg *alg)
+{
+	return alg->cra_flags & (CRYPTO_ALG_DEAD | CRYPTO_ALG_DYING);
+}
+
 static inline int crypto_notify(unsigned long val, void *v)
 {
 	return blocking_notifier_call_chain(&crypto_chain, val, v);
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index ffec530d52fb..b20f4bdb23ba 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -36,10 +36,21 @@ struct crypto_template {
 	char name[CRYPTO_MAX_ALG_NAME];
 };
 
+struct crypto_spawn {
+	struct list_head list;
+	struct crypto_alg *alg;
+	struct crypto_instance *inst;
+};
+
 int crypto_register_template(struct crypto_template *tmpl);
 void crypto_unregister_template(struct crypto_template *tmpl);
 struct crypto_template *crypto_lookup_template(const char *name);
 
+int crypto_init_spawn(struct crypto_spawn *spawn, struct crypto_alg *alg,
+		      struct crypto_instance *inst);
+void crypto_drop_spawn(struct crypto_spawn *spawn);
+struct crypto_tfm *crypto_spawn_tfm(struct crypto_spawn *spawn);
+
 static inline void *crypto_instance_ctx(struct crypto_instance *inst)
 {
 	return inst->__ctx;
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 85f73c381913..40a6330abc8d 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -35,6 +35,8 @@
 #define CRYPTO_ALG_TYPE_COMPRESS	0x00000004
 
 #define CRYPTO_ALG_LARVAL		0x00000010
+#define CRYPTO_ALG_DEAD			0x00000020
+#define CRYPTO_ALG_DYING		0x00000040
 
 /*
  * Transform masks and values (for crt_flags).
@@ -145,6 +147,8 @@ struct compress_alg {
 
 struct crypto_alg {
 	struct list_head cra_list;
+	struct list_head cra_users;
+
 	u32 cra_flags;
 	unsigned int cra_blocksize;
 	unsigned int cra_ctxsize;
-- 
cgit v1.2.3


From b14cdd6704c96474ba5c74b5959487beaa5ee1cd Mon Sep 17 00:00:00 2001
From: Michal Ludvig <michal@logix.cz>
Date: Sun, 9 Jul 2006 09:02:24 +1000
Subject: [CRYPTO] api: Add missing accessors for new crypto_alg fields

Add missing accessors for cra_driver_name and cra_priority.

Signed-off-by: Michal Ludvig <michal@logix.cz>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/crypto.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 40a6330abc8d..d6e184c876b5 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -297,6 +297,16 @@ static inline const char *crypto_tfm_alg_name(struct crypto_tfm *tfm)
 	return tfm->__crt_alg->cra_name;
 }
 
+static inline const char *crypto_tfm_alg_driver_name(struct crypto_tfm *tfm)
+{
+	return tfm->__crt_alg->cra_driver_name;
+}
+
+static inline int crypto_tfm_alg_priority(struct crypto_tfm *tfm)
+{
+	return tfm->__crt_alg->cra_priority;
+}
+
 static inline const char *crypto_tfm_alg_modname(struct crypto_tfm *tfm)
 {
 	return module_name(tfm->__crt_alg->cra_module);
-- 
cgit v1.2.3


From 560c06ae1ab7c677002ea3b6ac83521bf12ee07d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 13 Aug 2006 14:16:39 +1000
Subject: [CRYPTO] api: Get rid of flags argument to setkey

Now that the tfm is passed directly to setkey instead of the ctx, we no
longer need to pass the &tfm->crt_flags pointer.

This patch also gets rid of a few unnecessary checks on the key length
for ciphers as the cipher layer guarantees that the key length is within
the bounds specified by the algorithm.

Rather than testing dia_setkey every time, this patch does it only once
during crypto_alloc_tfm.  The redundant check from crypto_digest_setkey
is also removed.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 arch/i386/crypto/aes.c       |  3 ++-
 arch/s390/crypto/aes_s390.c  |  3 ++-
 arch/s390/crypto/des_s390.c  | 13 ++++++++-----
 arch/x86_64/crypto/aes.c     |  5 +++--
 crypto/aes.c                 |  5 +++--
 crypto/anubis.c              |  3 ++-
 crypto/arc4.c                |  2 +-
 crypto/blowfish.c            |  3 +--
 crypto/cast5.c               |  8 +-------
 crypto/cast6.c               |  5 +++--
 crypto/cipher.c              |  4 ++--
 crypto/crc32c.c              |  5 ++---
 crypto/crypto_null.c         |  2 +-
 crypto/des.c                 |  6 ++++--
 crypto/digest.c              | 15 ++++++++++-----
 crypto/khazad.c              |  8 +-------
 crypto/michael_mic.c         |  5 ++---
 crypto/serpent.c             | 19 +++----------------
 crypto/tcrypt.c              |  5 +----
 crypto/tea.c                 | 16 ++--------------
 crypto/twofish_common.c      |  6 +++---
 drivers/crypto/padlock-aes.c |  5 +++--
 include/crypto/twofish.h     |  3 +--
 include/linux/crypto.h       |  6 ++----
 24 files changed, 63 insertions(+), 92 deletions(-)

(limited to 'include')

diff --git a/arch/i386/crypto/aes.c b/arch/i386/crypto/aes.c
index d3806daa3de3..49aad9397f10 100644
--- a/arch/i386/crypto/aes.c
+++ b/arch/i386/crypto/aes.c
@@ -379,12 +379,13 @@ static void gen_tabs(void)
 }
 
 static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-		       unsigned int key_len, u32 *flags)
+		       unsigned int key_len)
 {
 	int i;
 	u32 ss[8];
 	struct aes_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
+	u32 *flags = &tfm->crt_flags;
 
 	/* encryption schedule */
 	
diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c
index 5713c7e5bd16..c7c43c9de0d9 100644
--- a/arch/s390/crypto/aes_s390.c
+++ b/arch/s390/crypto/aes_s390.c
@@ -38,9 +38,10 @@ struct s390_aes_ctx {
 };
 
 static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-		       unsigned int key_len, u32 *flags)
+		       unsigned int key_len)
 {
 	struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm);
+	u32 *flags = &tfm->crt_flags;
 
 	switch (key_len) {
 	case 16:
diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c
index b3f7496a79b4..170757b3451d 100644
--- a/arch/s390/crypto/des_s390.c
+++ b/arch/s390/crypto/des_s390.c
@@ -45,9 +45,10 @@ struct crypt_s390_des3_192_ctx {
 };
 
 static int des_setkey(struct crypto_tfm *tfm, const u8 *key,
-		      unsigned int keylen, u32 *flags)
+		      unsigned int keylen)
 {
 	struct crypt_s390_des_ctx *dctx = crypto_tfm_ctx(tfm);
+	u32 *flags = &tfm->crt_flags;
 	int ret;
 
 	/* test if key is valid (not a weak key) */
@@ -167,11 +168,12 @@ static struct crypto_alg des_alg = {
  *
  */
 static int des3_128_setkey(struct crypto_tfm *tfm, const u8 *key,
-			   unsigned int keylen, u32 *flags)
+			   unsigned int keylen)
 {
 	int i, ret;
 	struct crypt_s390_des3_128_ctx *dctx = crypto_tfm_ctx(tfm);
-	const u8* temp_key = key;
+	const u8 *temp_key = key;
+	u32 *flags = &tfm->crt_flags;
 
 	if (!(memcmp(key, &key[DES_KEY_SIZE], DES_KEY_SIZE))) {
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_SCHED;
@@ -303,11 +305,12 @@ static struct crypto_alg des3_128_alg = {
  *
  */
 static int des3_192_setkey(struct crypto_tfm *tfm, const u8 *key,
-			   unsigned int keylen, u32 *flags)
+			   unsigned int keylen)
 {
 	int i, ret;
 	struct crypt_s390_des3_192_ctx *dctx = crypto_tfm_ctx(tfm);
-	const u8* temp_key = key;
+	const u8 *temp_key = key;
+	u32 *flags = &tfm->crt_flags;
 
 	if (!(memcmp(key, &key[DES_KEY_SIZE], DES_KEY_SIZE) &&
 	    memcmp(&key[DES_KEY_SIZE], &key[DES_KEY_SIZE * 2],
diff --git a/arch/x86_64/crypto/aes.c b/arch/x86_64/crypto/aes.c
index 68866fab37aa..5cdb13ea5cc2 100644
--- a/arch/x86_64/crypto/aes.c
+++ b/arch/x86_64/crypto/aes.c
@@ -228,13 +228,14 @@ static void __init gen_tabs(void)
 }
 
 static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-		       unsigned int key_len, u32 *flags)
+		       unsigned int key_len)
 {
 	struct aes_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
+	u32 *flags = &tfm->crt_flags;
 	u32 i, j, t, u, v, w;
 
-	if (key_len != 16 && key_len != 24 && key_len != 32) {
+	if (key_len % 8) {
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	}
diff --git a/crypto/aes.c b/crypto/aes.c
index a038711831e7..e2440773878c 100644
--- a/crypto/aes.c
+++ b/crypto/aes.c
@@ -249,13 +249,14 @@ gen_tabs (void)
 }
 
 static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-		       unsigned int key_len, u32 *flags)
+		       unsigned int key_len)
 {
 	struct aes_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
+	u32 *flags = &tfm->crt_flags;
 	u32 i, t, u, v, w;
 
-	if (key_len != 16 && key_len != 24 && key_len != 32) {
+	if (key_len % 8) {
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	}
diff --git a/crypto/anubis.c b/crypto/anubis.c
index 7e2e1a29800e..1c771f7f4dc5 100644
--- a/crypto/anubis.c
+++ b/crypto/anubis.c
@@ -461,10 +461,11 @@ static const u32 rc[] = {
 };
 
 static int anubis_setkey(struct crypto_tfm *tfm, const u8 *in_key,
-			 unsigned int key_len, u32 *flags)
+			 unsigned int key_len)
 {
 	struct anubis_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __be32 *key = (const __be32 *)in_key;
+	u32 *flags = &tfm->crt_flags;
 	int N, R, i, r;
 	u32 kappa[ANUBIS_MAX_N];
 	u32 inter[ANUBIS_MAX_N];
diff --git a/crypto/arc4.c b/crypto/arc4.c
index 5edc6a65b987..8be47e13a9e3 100644
--- a/crypto/arc4.c
+++ b/crypto/arc4.c
@@ -25,7 +25,7 @@ struct arc4_ctx {
 };
 
 static int arc4_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-			unsigned int key_len, u32 *flags)
+			unsigned int key_len)
 {
 	struct arc4_ctx *ctx = crypto_tfm_ctx(tfm);
 	int i, j = 0, k = 0;
diff --git a/crypto/blowfish.c b/crypto/blowfish.c
index 490265f42b3b..55238c4e37f0 100644
--- a/crypto/blowfish.c
+++ b/crypto/blowfish.c
@@ -399,8 +399,7 @@ static void bf_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 /* 
  * Calculates the blowfish S and P boxes for encryption and decryption.
  */
-static int bf_setkey(struct crypto_tfm *tfm, const u8 *key,
-		     unsigned int keylen, u32 *flags)
+static int bf_setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
 {
 	struct bf_ctx *ctx = crypto_tfm_ctx(tfm);
 	u32 *P = ctx->p;
diff --git a/crypto/cast5.c b/crypto/cast5.c
index 08eef58c1d3d..13ea60abc19a 100644
--- a/crypto/cast5.c
+++ b/crypto/cast5.c
@@ -769,8 +769,7 @@ static void key_schedule(u32 * x, u32 * z, u32 * k)
 }
 
 
-static int cast5_setkey(struct crypto_tfm *tfm, const u8 *key,
-			unsigned key_len, u32 *flags)
+static int cast5_setkey(struct crypto_tfm *tfm, const u8 *key, unsigned key_len)
 {
 	struct cast5_ctx *c = crypto_tfm_ctx(tfm);
 	int i;
@@ -778,11 +777,6 @@ static int cast5_setkey(struct crypto_tfm *tfm, const u8 *key,
 	u32 z[4];
 	u32 k[16];
 	__be32 p_key[4];
-	
-	if (key_len < 5 || key_len > 16) {
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
 
 	c->rr = key_len <= 10 ? 1 : 0;
 
diff --git a/crypto/cast6.c b/crypto/cast6.c
index 08e33bfc3ad1..136ab6dfe8c5 100644
--- a/crypto/cast6.c
+++ b/crypto/cast6.c
@@ -382,14 +382,15 @@ static inline void W(u32 *key, unsigned int i) {
 }
 
 static int cast6_setkey(struct crypto_tfm *tfm, const u8 *in_key,
-			unsigned key_len, u32 *flags)
+			unsigned key_len)
 {
 	int i;
 	u32 key[8];
 	__be32 p_key[8]; /* padded key */
 	struct cast6_ctx *c = crypto_tfm_ctx(tfm);
+	u32 *flags = &tfm->crt_flags;
 
-	if (key_len < 16 || key_len > 32 || key_len % 4 != 0) {
+	if (key_len % 4 != 0) {
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	}	
diff --git a/crypto/cipher.c b/crypto/cipher.c
index b899eb97abd7..56406a4a88d4 100644
--- a/crypto/cipher.c
+++ b/crypto/cipher.c
@@ -264,12 +264,12 @@ static int setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
 {
 	struct cipher_alg *cia = &tfm->__crt_alg->cra_cipher;
 	
+	tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK;
 	if (keylen < cia->cia_min_keysize || keylen > cia->cia_max_keysize) {
 		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	} else
-		return cia->cia_setkey(tfm, key, keylen,
-		                       &tfm->crt_flags);
+		return cia->cia_setkey(tfm, key, keylen);
 }
 
 static int ecb_encrypt(struct crypto_tfm *tfm,
diff --git a/crypto/crc32c.c b/crypto/crc32c.c
index 91ecd895e957..0fa744392a4c 100644
--- a/crypto/crc32c.c
+++ b/crypto/crc32c.c
@@ -44,13 +44,12 @@ static void chksum_init(struct crypto_tfm *tfm)
  * the seed.
  */
 static int chksum_setkey(struct crypto_tfm *tfm, const u8 *key,
-			 unsigned int keylen, u32 *flags)
+			 unsigned int keylen)
 {
 	struct chksum_ctx *mctx = crypto_tfm_ctx(tfm);
 
 	if (keylen != sizeof(mctx->crc)) {
-		if (flags)
-			*flags = CRYPTO_TFM_RES_BAD_KEY_LEN;
+		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	}
 	mctx->key = le32_to_cpu(*(__le32 *)key);
diff --git a/crypto/crypto_null.c b/crypto/crypto_null.c
index a0d956b52949..24dbb5d8617e 100644
--- a/crypto/crypto_null.c
+++ b/crypto/crypto_null.c
@@ -48,7 +48,7 @@ static void null_final(struct crypto_tfm *tfm, u8 *out)
 { }
 
 static int null_setkey(struct crypto_tfm *tfm, const u8 *key,
-		       unsigned int keylen, u32 *flags)
+		       unsigned int keylen)
 { return 0; }
 
 static void null_crypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
diff --git a/crypto/des.c b/crypto/des.c
index a9d3c235a6af..1df3a714fa47 100644
--- a/crypto/des.c
+++ b/crypto/des.c
@@ -784,9 +784,10 @@ static void dkey(u32 *pe, const u8 *k)
 }
 
 static int des_setkey(struct crypto_tfm *tfm, const u8 *key,
-		      unsigned int keylen, u32 *flags)
+		      unsigned int keylen)
 {
 	struct des_ctx *dctx = crypto_tfm_ctx(tfm);
+	u32 *flags = &tfm->crt_flags;
 	u32 tmp[DES_EXPKEY_WORDS];
 	int ret;
 
@@ -864,11 +865,12 @@ static void des_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
  *
  */
 static int des3_ede_setkey(struct crypto_tfm *tfm, const u8 *key,
-			   unsigned int keylen, u32 *flags)
+			   unsigned int keylen)
 {
 	const u32 *K = (const u32 *)key;
 	struct des3_ede_ctx *dctx = crypto_tfm_ctx(tfm);
 	u32 *expkey = dctx->expkey;
+	u32 *flags = &tfm->crt_flags;
 
 	if (unlikely(!((K[0] ^ K[2]) | (K[1] ^ K[3])) ||
 		     !((K[2] ^ K[4]) | (K[3] ^ K[5]))))
diff --git a/crypto/digest.c b/crypto/digest.c
index 603006a7bef2..0df7f392a56a 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -76,12 +76,16 @@ static void final(struct crypto_tfm *tfm, u8 *out)
 		tfm->__crt_alg->cra_digest.dia_final(tfm, out);
 }
 
+static int nosetkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
+{
+	tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK;
+	return -ENOSYS;
+}
+
 static int setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
 {
-	u32 flags;
-	if (tfm->__crt_alg->cra_digest.dia_setkey == NULL)
-		return -ENOSYS;
-	return tfm->__crt_alg->cra_digest.dia_setkey(tfm, key, keylen, &flags);
+	tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK;
+	return tfm->__crt_alg->cra_digest.dia_setkey(tfm, key, keylen);
 }
 
 static void digest(struct crypto_tfm *tfm,
@@ -100,12 +104,13 @@ int crypto_init_digest_flags(struct crypto_tfm *tfm, u32 flags)
 int crypto_init_digest_ops(struct crypto_tfm *tfm)
 {
 	struct digest_tfm *ops = &tfm->crt_digest;
+	struct digest_alg *dalg = &tfm->__crt_alg->cra_digest;
 	
 	ops->dit_init	= init;
 	ops->dit_update	= update;
 	ops->dit_final	= final;
 	ops->dit_digest	= digest;
-	ops->dit_setkey	= setkey;
+	ops->dit_setkey	= dalg->dia_setkey ? setkey : nosetkey;
 	
 	return crypto_alloc_hmac_block(tfm);
 }
diff --git a/crypto/khazad.c b/crypto/khazad.c
index d4c9d3657b36..9fa24a2dd6ff 100644
--- a/crypto/khazad.c
+++ b/crypto/khazad.c
@@ -755,19 +755,13 @@ static const u64 c[KHAZAD_ROUNDS + 1] = {
 };
 
 static int khazad_setkey(struct crypto_tfm *tfm, const u8 *in_key,
-			 unsigned int key_len, u32 *flags)
+			 unsigned int key_len)
 {
 	struct khazad_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __be32 *key = (const __be32 *)in_key;
 	int r;
 	const u64 *S = T7;
 	u64 K2, K1;
-	
-	if (key_len != 16)
-	{
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
 
 	/* key is supposed to be 32-bit aligned */
 	K2 = ((u64)be32_to_cpu(key[0]) << 32) | be32_to_cpu(key[1]);
diff --git a/crypto/michael_mic.c b/crypto/michael_mic.c
index d061da21cfda..094397b48849 100644
--- a/crypto/michael_mic.c
+++ b/crypto/michael_mic.c
@@ -123,14 +123,13 @@ static void michael_final(struct crypto_tfm *tfm, u8 *out)
 
 
 static int michael_setkey(struct crypto_tfm *tfm, const u8 *key,
-			  unsigned int keylen, u32 *flags)
+			  unsigned int keylen)
 {
 	struct michael_mic_ctx *mctx = crypto_tfm_ctx(tfm);
 	const __le32 *data = (const __le32 *)key;
 
 	if (keylen != 8) {
-		if (flags)
-			*flags = CRYPTO_TFM_RES_BAD_KEY_LEN;
+		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	}
 
diff --git a/crypto/serpent.c b/crypto/serpent.c
index de60cdddbf4a..465d091cd3ec 100644
--- a/crypto/serpent.c
+++ b/crypto/serpent.c
@@ -216,7 +216,7 @@ struct serpent_ctx {
 
 
 static int serpent_setkey(struct crypto_tfm *tfm, const u8 *key,
-			  unsigned int keylen, u32 *flags)
+			  unsigned int keylen)
 {
 	struct serpent_ctx *ctx = crypto_tfm_ctx(tfm);
 	u32 *k = ctx->expkey;
@@ -224,13 +224,6 @@ static int serpent_setkey(struct crypto_tfm *tfm, const u8 *key,
 	u32 r0,r1,r2,r3,r4;
 	int i;
 
-	if ((keylen < SERPENT_MIN_KEY_SIZE)
-			|| (keylen > SERPENT_MAX_KEY_SIZE))
-	{
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
-
 	/* Copy key, add padding */
 
 	for (i = 0; i < keylen; ++i)
@@ -497,21 +490,15 @@ static struct crypto_alg serpent_alg = {
 };
 
 static int tnepres_setkey(struct crypto_tfm *tfm, const u8 *key,
-			  unsigned int keylen, u32 *flags)
+			  unsigned int keylen)
 {
 	u8 rev_key[SERPENT_MAX_KEY_SIZE];
 	int i;
 
-	if ((keylen < SERPENT_MIN_KEY_SIZE)
-	    || (keylen > SERPENT_MAX_KEY_SIZE)) {
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	} 
-
 	for (i = 0; i < keylen; ++i)
 		rev_key[keylen - i - 1] = key[i];
  
-	return serpent_setkey(tfm, rev_key, keylen, flags);
+	return serpent_setkey(tfm, rev_key, keylen);
 }
 
 static void tnepres_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index bed225e83231..606777074671 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -118,10 +118,7 @@ static void test_hash(char *algo, struct hash_testvec *template,
 		sg_set_buf(&sg[0], hash_tv[i].plaintext, hash_tv[i].psize);
 
 		crypto_digest_init(tfm);
-		if (tfm->crt_u.digest.dit_setkey) {
-			crypto_digest_setkey(tfm, hash_tv[i].key,
-					     hash_tv[i].ksize);
-		}
+		crypto_digest_setkey(tfm, hash_tv[i].key, hash_tv[i].ksize);
 		crypto_digest_update(tfm, sg, 1);
 		crypto_digest_final(tfm, result);
 
diff --git a/crypto/tea.c b/crypto/tea.c
index 5367adc82fc9..1c54e26fa529 100644
--- a/crypto/tea.c
+++ b/crypto/tea.c
@@ -46,16 +46,10 @@ struct xtea_ctx {
 };
 
 static int tea_setkey(struct crypto_tfm *tfm, const u8 *in_key,
-		      unsigned int key_len, u32 *flags)
+		      unsigned int key_len)
 {
 	struct tea_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
-	
-	if (key_len != 16)
-	{
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
 
 	ctx->KEY[0] = le32_to_cpu(key[0]);
 	ctx->KEY[1] = le32_to_cpu(key[1]);
@@ -125,16 +119,10 @@ static void tea_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 }
 
 static int xtea_setkey(struct crypto_tfm *tfm, const u8 *in_key,
-		       unsigned int key_len, u32 *flags)
+		       unsigned int key_len)
 {
 	struct xtea_ctx *ctx = crypto_tfm_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
-	
-	if (key_len != 16)
-	{
-		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
 
 	ctx->KEY[0] = le32_to_cpu(key[0]);
 	ctx->KEY[1] = le32_to_cpu(key[1]);
diff --git a/crypto/twofish_common.c b/crypto/twofish_common.c
index 1ae0280c2513..b4b9c0c3f4ae 100644
--- a/crypto/twofish_common.c
+++ b/crypto/twofish_common.c
@@ -580,11 +580,11 @@ static const u8 calc_sb_tbl[512] = {
    ctx->a[(j) + 1] = rol32(y, 9)
 
 /* Perform the key setup. */
-int twofish_setkey(struct crypto_tfm *tfm, const u8 *key,
-		   unsigned int key_len, u32 *flags)
+int twofish_setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int key_len)
 {
 
 	struct twofish_ctx *ctx = crypto_tfm_ctx(tfm);
+	u32 *flags = &tfm->crt_flags;
 
 	int i, j, k;
 
@@ -600,7 +600,7 @@ int twofish_setkey(struct crypto_tfm *tfm, const u8 *key,
 	u8 tmp;
 
 	/* Check key length. */
-	if (key_len != 16 && key_len != 24 && key_len != 32)
+	if (key_len % 8)
 	{
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL; /* unsupported key length */
diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c
index 3a2a71108d35..3e683709243e 100644
--- a/drivers/crypto/padlock-aes.c
+++ b/drivers/crypto/padlock-aes.c
@@ -308,15 +308,16 @@ static inline struct aes_ctx *aes_ctx(struct crypto_tfm *tfm)
 }
 
 static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-		       unsigned int key_len, u32 *flags)
+		       unsigned int key_len)
 {
 	struct aes_ctx *ctx = aes_ctx(tfm);
 	const __le32 *key = (const __le32 *)in_key;
+	u32 *flags = &tfm->crt_flags;
 	uint32_t i, t, u, v, w;
 	uint32_t P[AES_EXTENDED_KEY_SIZE];
 	uint32_t rounds;
 
-	if (key_len != 16 && key_len != 24 && key_len != 32) {
+	if (key_len % 8) {
 		*flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
 		return -EINVAL;
 	}
diff --git a/include/crypto/twofish.h b/include/crypto/twofish.h
index e4328cfaaf64..c408522595c6 100644
--- a/include/crypto/twofish.h
+++ b/include/crypto/twofish.h
@@ -17,7 +17,6 @@ struct twofish_ctx {
 	u32 s[4][256], w[8], k[32];
 };
 
-int twofish_setkey(struct crypto_tfm *tfm, const u8 *key,
-		   unsigned int key_len, u32 *flags);
+int twofish_setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int key_len);
 
 #endif
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index d6e184c876b5..053bfab43e8d 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -106,7 +106,7 @@ struct cipher_alg {
 	unsigned int cia_min_keysize;
 	unsigned int cia_max_keysize;
 	int (*cia_setkey)(struct crypto_tfm *tfm, const u8 *key,
-	                  unsigned int keylen, u32 *flags);
+	                  unsigned int keylen);
 	void (*cia_encrypt)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
 	void (*cia_decrypt)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
 
@@ -131,7 +131,7 @@ struct digest_alg {
 			   unsigned int len);
 	void (*dia_final)(struct crypto_tfm *tfm, u8 *out);
 	int (*dia_setkey)(struct crypto_tfm *tfm, const u8 *key,
-	                  unsigned int keylen, u32 *flags);
+	                  unsigned int keylen);
 };
 
 struct compress_alg {
@@ -397,8 +397,6 @@ static inline int crypto_digest_setkey(struct crypto_tfm *tfm,
                                        const u8 *key, unsigned int keylen)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-	if (tfm->crt_digest.dit_setkey == NULL)
-		return -ENOSYS;
 	return tfm->crt_digest.dit_setkey(tfm, key, keylen);
 }
 
-- 
cgit v1.2.3


From df89820ebd5bbf4f3c6b5f8ee7d9e983107f6a91 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 14 Jul 2006 10:42:27 +1000
Subject: [CRYPTO] cipher: Removed special IV checks for ECB

This patch makes IV operations on ECB fail through nocrypt_iv rather than
calling BUG().  This is needed to generalise CBC/ECB using the template
mechanism.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/cipher.c        | 2 ++
 include/linux/crypto.h | 2 --
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/crypto/cipher.c b/crypto/cipher.c
index 56406a4a88d4..aebc4a2adc80 100644
--- a/crypto/cipher.c
+++ b/crypto/cipher.c
@@ -399,6 +399,8 @@ int crypto_init_cipher_ops(struct crypto_tfm *tfm)
 	case CRYPTO_TFM_MODE_ECB:
 		ops->cit_encrypt = ecb_encrypt;
 		ops->cit_decrypt = ecb_decrypt;
+		ops->cit_encrypt_iv = nocrypt_iv;
+		ops->cit_decrypt_iv = nocrypt_iv;
 		break;
 		
 	case CRYPTO_TFM_MODE_CBC:
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 053bfab43e8d..dbdfc7c79367 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -422,7 +422,6 @@ static inline int crypto_cipher_encrypt_iv(struct crypto_tfm *tfm,
                                            unsigned int nbytes, u8 *iv)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
-	BUG_ON(tfm->crt_cipher.cit_mode == CRYPTO_TFM_MODE_ECB);
 	return tfm->crt_cipher.cit_encrypt_iv(tfm, dst, src, nbytes, iv);
 }                                        
 
@@ -441,7 +440,6 @@ static inline int crypto_cipher_decrypt_iv(struct crypto_tfm *tfm,
                                            unsigned int nbytes, u8 *iv)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
-	BUG_ON(tfm->crt_cipher.cit_mode == CRYPTO_TFM_MODE_ECB);
 	return tfm->crt_cipher.cit_decrypt_iv(tfm, dst, src, nbytes, iv);
 }
 
-- 
cgit v1.2.3


From 7fed0bf271b374be4c98a5880faed4b1128e78e9 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 6 Aug 2006 23:10:45 +1000
Subject: [CRYPTO] api: Add common instance initialisation code

This patch adds the helpers crypto_get_attr_alg and crypto_alloc_instance
which can be used by simple one-argument templates like hmac to process
input parameters and allocate instances.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/algapi.c         | 53 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/crypto/algapi.h |  5 +++++
 2 files changed, 58 insertions(+)

(limited to 'include')

diff --git a/crypto/algapi.c b/crypto/algapi.c
index 36c4f1bdb521..c91530021e9c 100644
--- a/crypto/algapi.c
+++ b/crypto/algapi.c
@@ -16,6 +16,7 @@
 #include <linux/kernel.h>
 #include <linux/list.h>
 #include <linux/module.h>
+#include <linux/rtnetlink.h>
 #include <linux/string.h>
 
 #include "internal.h"
@@ -415,6 +416,58 @@ int crypto_unregister_notifier(struct notifier_block *nb)
 }
 EXPORT_SYMBOL_GPL(crypto_unregister_notifier);
 
+struct crypto_alg *crypto_get_attr_alg(void *param, unsigned int len,
+				       u32 type, u32 mask)
+{
+	struct rtattr *rta = param;
+	struct crypto_attr_alg *alga;
+
+	if (!RTA_OK(rta, len))
+		return ERR_PTR(-EBADR);
+	if (rta->rta_type != CRYPTOA_ALG || RTA_PAYLOAD(rta) < sizeof(*alga))
+		return ERR_PTR(-EINVAL);
+
+	alga = RTA_DATA(rta);
+	alga->name[CRYPTO_MAX_ALG_NAME - 1] = 0;
+
+	return crypto_alg_mod_lookup(alga->name, type, mask);
+}
+EXPORT_SYMBOL_GPL(crypto_get_attr_alg);
+
+struct crypto_instance *crypto_alloc_instance(const char *name,
+					      struct crypto_alg *alg)
+{
+	struct crypto_instance *inst;
+	struct crypto_spawn *spawn;
+	int err;
+
+	inst = kzalloc(sizeof(*inst) + sizeof(*spawn), GFP_KERNEL);
+	if (!inst)
+		return ERR_PTR(-ENOMEM);
+
+	err = -ENAMETOOLONG;
+	if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME, "%s(%s)", name,
+		     alg->cra_name) >= CRYPTO_MAX_ALG_NAME)
+		goto err_free_inst;
+
+	if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME, "%s(%s)",
+		     name, alg->cra_driver_name) >= CRYPTO_MAX_ALG_NAME)
+		goto err_free_inst;
+
+	spawn = crypto_instance_ctx(inst);
+	err = crypto_init_spawn(spawn, alg, inst);
+
+	if (err)
+		goto err_free_inst;
+
+	return inst;
+
+err_free_inst:
+	kfree(inst);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(crypto_alloc_instance);
+
 static int __init crypto_algapi_init(void)
 {
 	crypto_init_proc();
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index b20f4bdb23ba..1a598f829417 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -51,6 +51,11 @@ int crypto_init_spawn(struct crypto_spawn *spawn, struct crypto_alg *alg,
 void crypto_drop_spawn(struct crypto_spawn *spawn);
 struct crypto_tfm *crypto_spawn_tfm(struct crypto_spawn *spawn);
 
+struct crypto_alg *crypto_get_attr_alg(void *param, unsigned int len,
+				       u32 type, u32 mask);
+struct crypto_instance *crypto_alloc_instance(const char *name,
+					      struct crypto_alg *alg);
+
 static inline void *crypto_instance_ctx(struct crypto_instance *inst)
 {
 	return inst->__ctx;
-- 
cgit v1.2.3


From f3f632d61ae9af85d436706ee8e33af1a7fb9c28 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 6 Aug 2006 23:12:59 +1000
Subject: [CRYPTO] api: Added asynchronous flag

This patch adds the asynchronous flag and changes all existing users to
only look up algorithms that are synchronous.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/api.c           | 5 +++--
 include/linux/crypto.h | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/crypto/api.c b/crypto/api.c
index 7e5522cf856e..1e4692a13474 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -355,7 +355,7 @@ struct crypto_tfm *crypto_alloc_tfm(const char *name, u32 flags)
 	do {
 		struct crypto_alg *alg;
 
-		alg = crypto_alg_mod_lookup(name, 0, 0);
+		alg = crypto_alg_mod_lookup(name, 0, CRYPTO_ALG_ASYNC);
 		err = PTR_ERR(alg);
 		if (IS_ERR(alg))
 			continue;
@@ -394,7 +394,8 @@ void crypto_free_tfm(struct crypto_tfm *tfm)
 int crypto_alg_available(const char *name, u32 flags)
 {
 	int ret = 0;
-	struct crypto_alg *alg = crypto_alg_mod_lookup(name, 0, 0);
+	struct crypto_alg *alg = crypto_alg_mod_lookup(name, 0,
+						       CRYPTO_ALG_ASYNC);
 	
 	if (!IS_ERR(alg)) {
 		crypto_mod_put(alg);
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index dbdfc7c79367..530dc4bf363c 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -37,6 +37,7 @@
 #define CRYPTO_ALG_LARVAL		0x00000010
 #define CRYPTO_ALG_DEAD			0x00000020
 #define CRYPTO_ALG_DYING		0x00000040
+#define CRYPTO_ALG_ASYNC		0x00000080
 
 /*
  * Transform masks and values (for crt_flags).
-- 
cgit v1.2.3


From 6d7d684d635ac5a345f075015f2c84169c111c6a Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 30 Jul 2006 11:53:01 +1000
Subject: [CRYPTO] api: Added crypto_alloc_base

Up until now all crypto transforms have been of the same type, struct
crypto_tfm, regardless of whether they are ciphers, digests, or other
types.  As a result of that, we check the types at run-time before
each crypto operation.

This is rather cumbersome.  We could instead use different C types for
each crypto type to ensure that the correct types are used at compile
time.  That is, we would have crypto_cipher/crypto_digest instead of
just crypto_tfm.  The appropriate type would then be required for the
actual operations such as crypto_digest_digest.

Now that we have the type/mask fields when looking up algorithms, it
is easy to request for an algorithm of the precise type that the user
wants.  However, crypto_alloc_tfm currently does not expose these new
attributes.

This patch introduces the function crypto_alloc_base which will carry
these new parameters.  It will be renamed to crypto_alloc_tfm once
all existing users have been converted.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/api.c           | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/crypto.h | 14 +++---------
 2 files changed, 63 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/crypto/api.c b/crypto/api.c
index 1e4692a13474..bc4b7901acdf 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -372,6 +372,66 @@ struct crypto_tfm *crypto_alloc_tfm(const char *name, u32 flags)
 	return tfm;
 }
 
+/*
+ *	crypto_alloc_base - Locate algorithm and allocate transform
+ *	@alg_name: Name of algorithm
+ *	@type: Type of algorithm
+ *	@mask: Mask for type comparison
+ *
+ *	crypto_alloc_base() will first attempt to locate an already loaded
+ *	algorithm.  If that fails and the kernel supports dynamically loadable
+ *	modules, it will then attempt to load a module of the same name or
+ *	alias.  If that fails it will send a query to any loaded crypto manager
+ *	to construct an algorithm on the fly.  A refcount is grabbed on the
+ *	algorithm which is then associated with the new transform.
+ *
+ *	The returned transform is of a non-determinate type.  Most people
+ *	should use one of the more specific allocation functions such as
+ *	crypto_alloc_blkcipher.
+ *
+ *	In case of error the return value is an error pointer.
+ */
+struct crypto_tfm *crypto_alloc_base(const char *alg_name, u32 type, u32 mask)
+{
+	struct crypto_tfm *tfm;
+	int err;
+
+	for (;;) {
+		struct crypto_alg *alg;
+
+		alg = crypto_alg_mod_lookup(alg_name, type, mask);
+		err = PTR_ERR(alg);
+		tfm = ERR_PTR(err);
+		if (IS_ERR(alg))
+			goto err;
+
+		tfm = __crypto_alloc_tfm(alg, 0);
+		if (!IS_ERR(tfm))
+			break;
+
+		crypto_mod_put(alg);
+		err = PTR_ERR(tfm);
+
+err:
+		if (err != -EAGAIN)
+			break;
+		if (signal_pending(current)) {
+			err = -EINTR;
+			break;
+		}
+	};
+
+	return tfm;
+}
+EXPORT_SYMBOL_GPL(crypto_alloc_base);
+ 
+/*
+ *	crypto_free_tfm - Free crypto transform
+ *	@tfm: Transform to free
+ *
+ *	crypto_free_tfm() frees up the transform and any associated resources,
+ *	then drops the refcount on the associated algorithm.
+ */
 void crypto_free_tfm(struct crypto_tfm *tfm)
 {
 	struct crypto_alg *alg;
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 530dc4bf363c..6847ab0ea30e 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -194,8 +194,8 @@ static inline int crypto_alg_available(const char *name, u32 flags)
 
 /*
  * Transforms: user-instantiated objects which encapsulate algorithms
- * and core processing logic.  Managed via crypto_alloc_tfm() and
- * crypto_free_tfm(), as well as the various helpers below.
+ * and core processing logic.  Managed via crypto_alloc_*() and
+ * crypto_free_*(), as well as the various helpers below.
  */
 
 struct cipher_tfm {
@@ -278,16 +278,8 @@ struct crypto_attr_alg {
  * Transform user interface.
  */
  
-/*
- * crypto_alloc_tfm() will first attempt to locate an already loaded algorithm.
- * If that fails and the kernel supports dynamically loadable modules, it
- * will then attempt to load a module of the same name or alias.  A refcount
- * is grabbed on the algorithm which is then associated with the new transform.
- *
- * crypto_free_tfm() frees up the transform and any associated resources,
- * then drops the refcount on the associated algorithm.
- */
 struct crypto_tfm *crypto_alloc_tfm(const char *alg_name, u32 tfm_flags);
+struct crypto_tfm *crypto_alloc_base(const char *alg_name, u32 type, u32 mask);
 void crypto_free_tfm(struct crypto_tfm *tfm);
 
 /*
-- 
cgit v1.2.3


From e853c3cfa8cc24869ecd2526e589bcb176bc12e9 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 22 Aug 2006 00:06:54 +1000
Subject: [CRYPTO] api: Added crypto_type support

This patch adds the crypto_type structure which will be used for all new
crypto algorithm types, beginning with block ciphers.

The primary purpose of this abstraction is to allow different crypto_type
objects for crypto algorithms of the same type, in particular, there will
be a different crypto_type objects for asynchronous algorithms.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/api.c            | 32 +++++++++++++++++++++++---------
 crypto/proc.c           |  5 ++++-
 include/crypto/algapi.h |  8 ++++++++
 include/linux/crypto.h  |  3 +++
 4 files changed, 38 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/crypto/api.c b/crypto/api.c
index bc4b7901acdf..edaa843d8e83 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -226,17 +226,18 @@ static int crypto_init_flags(struct crypto_tfm *tfm, u32 flags)
 		
 	case CRYPTO_ALG_TYPE_COMPRESS:
 		return crypto_init_compress_flags(tfm, flags);
-	
-	default:
-		break;
 	}
 	
-	BUG();
-	return -EINVAL;
+	return 0;
 }
 
 static int crypto_init_ops(struct crypto_tfm *tfm)
 {
+	const struct crypto_type *type = tfm->__crt_alg->cra_type;
+
+	if (type)
+		return type->init(tfm);
+
 	switch (crypto_tfm_alg_type(tfm)) {
 	case CRYPTO_ALG_TYPE_CIPHER:
 		return crypto_init_cipher_ops(tfm);
@@ -257,6 +258,14 @@ static int crypto_init_ops(struct crypto_tfm *tfm)
 
 static void crypto_exit_ops(struct crypto_tfm *tfm)
 {
+	const struct crypto_type *type = tfm->__crt_alg->cra_type;
+
+	if (type) {
+		if (type->exit)
+			type->exit(tfm);
+		return;
+	}
+
 	switch (crypto_tfm_alg_type(tfm)) {
 	case CRYPTO_ALG_TYPE_CIPHER:
 		crypto_exit_cipher_ops(tfm);
@@ -278,26 +287,31 @@ static void crypto_exit_ops(struct crypto_tfm *tfm)
 
 static unsigned int crypto_ctxsize(struct crypto_alg *alg, int flags)
 {
+	const struct crypto_type *type = alg->cra_type;
 	unsigned int len;
 
+	len = alg->cra_alignmask & ~(crypto_tfm_ctx_alignment() - 1);
+	if (type)
+		return len + type->ctxsize(alg);
+
 	switch (alg->cra_flags & CRYPTO_ALG_TYPE_MASK) {
 	default:
 		BUG();
 
 	case CRYPTO_ALG_TYPE_CIPHER:
-		len = crypto_cipher_ctxsize(alg, flags);
+		len += crypto_cipher_ctxsize(alg, flags);
 		break;
 		
 	case CRYPTO_ALG_TYPE_DIGEST:
-		len = crypto_digest_ctxsize(alg, flags);
+		len += crypto_digest_ctxsize(alg, flags);
 		break;
 		
 	case CRYPTO_ALG_TYPE_COMPRESS:
-		len = crypto_compress_ctxsize(alg, flags);
+		len += crypto_compress_ctxsize(alg, flags);
 		break;
 	}
 
-	return len + (alg->cra_alignmask & ~(crypto_tfm_ctx_alignment() - 1));
+	return len;
 }
 
 void crypto_shoot_alg(struct crypto_alg *alg)
diff --git a/crypto/proc.c b/crypto/proc.c
index 9e573b17e887..dabce0676f63 100644
--- a/crypto/proc.c
+++ b/crypto/proc.c
@@ -78,7 +78,10 @@ static int c_show(struct seq_file *m, void *p)
 		seq_printf(m, "type         : compression\n");
 		break;
 	default:
-		seq_printf(m, "type         : unknown\n");
+		if (alg->cra_type && alg->cra_type->show)
+			alg->cra_type->show(m, alg);
+		else
+			seq_printf(m, "type         : unknown\n");
 		break;
 	}
 
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 1a598f829417..c533c0a291af 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -15,6 +15,14 @@
 #include <linux/crypto.h>
 
 struct module;
+struct seq_file;
+
+struct crypto_type {
+	unsigned int (*ctxsize)(struct crypto_alg *alg);
+	int (*init)(struct crypto_tfm *tfm);
+	void (*exit)(struct crypto_tfm *tfm);
+	void (*show)(struct seq_file *m, struct crypto_alg *alg);
+};
 
 struct crypto_instance {
 	struct crypto_alg alg;
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 6847ab0ea30e..8e9c407b00d2 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -90,6 +90,7 @@
 
 struct scatterlist;
 struct crypto_tfm;
+struct crypto_type;
 
 struct cipher_desc {
 	struct crypto_tfm *tfm;
@@ -161,6 +162,8 @@ struct crypto_alg {
 	char cra_name[CRYPTO_MAX_ALG_NAME];
 	char cra_driver_name[CRYPTO_MAX_ALG_NAME];
 
+	const struct crypto_type *cra_type;
+
 	union {
 		struct cipher_alg cipher;
 		struct digest_alg digest;
-- 
cgit v1.2.3


From f28776a369b12f9a03a822a8e1090ed670a41f4f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 13 Aug 2006 20:58:18 +1000
Subject: [CRYPTO] cipher: Added encrypt_one/decrypt_one

This patch adds two new operations for the simple cipher that encrypts or
decrypts a single block at a time.  This will be the main interface after
the existing block operations have moved over to the new block ciphers.

It also adds the crypto_cipher type which is currently only used on the
new operations but will be extended to setkey as well once existing users
have been converted to use block ciphers where applicable.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/cipher.c         | 48 +++++++++++++++++++++++++
 include/crypto/algapi.h |  5 +++
 include/linux/crypto.h  | 96 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 149 insertions(+)

(limited to 'include')

diff --git a/crypto/cipher.c b/crypto/cipher.c
index f573c59ed9dc..d8ca0ec8d0be 100644
--- a/crypto/cipher.c
+++ b/crypto/cipher.c
@@ -388,12 +388,60 @@ int crypto_init_cipher_flags(struct crypto_tfm *tfm, u32 flags)
 	return 0;
 }
 
+static void cipher_crypt_unaligned(void (*fn)(struct crypto_tfm *, u8 *,
+					      const u8 *),
+				   struct crypto_tfm *tfm,
+				   u8 *dst, const u8 *src)
+{
+	unsigned long alignmask = crypto_tfm_alg_alignmask(tfm);
+	unsigned int size = crypto_tfm_alg_blocksize(tfm);
+	u8 buffer[size + alignmask];
+	u8 *tmp = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1);
+
+	memcpy(tmp, src, size);
+	fn(tfm, tmp, tmp);
+	memcpy(dst, tmp, size);
+}
+
+static void cipher_encrypt_unaligned(struct crypto_tfm *tfm,
+				     u8 *dst, const u8 *src)
+{
+	unsigned long alignmask = crypto_tfm_alg_alignmask(tfm);
+	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+
+	if (unlikely(((unsigned long)dst | (unsigned long)src) & alignmask)) {
+		cipher_crypt_unaligned(cipher->cia_encrypt, tfm, dst, src);
+		return;
+	}
+
+	cipher->cia_encrypt(tfm, dst, src);
+}
+
+static void cipher_decrypt_unaligned(struct crypto_tfm *tfm,
+				     u8 *dst, const u8 *src)
+{
+	unsigned long alignmask = crypto_tfm_alg_alignmask(tfm);
+	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+
+	if (unlikely(((unsigned long)dst | (unsigned long)src) & alignmask)) {
+		cipher_crypt_unaligned(cipher->cia_decrypt, tfm, dst, src);
+		return;
+	}
+
+	cipher->cia_decrypt(tfm, dst, src);
+}
+
 int crypto_init_cipher_ops(struct crypto_tfm *tfm)
 {
 	int ret = 0;
 	struct cipher_tfm *ops = &tfm->crt_cipher;
+	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
 
 	ops->cit_setkey = setkey;
+	ops->cit_encrypt_one = crypto_tfm_alg_alignmask(tfm) ?
+		cipher_encrypt_unaligned : cipher->cia_encrypt;
+	ops->cit_decrypt_one = crypto_tfm_alg_alignmask(tfm) ?
+		cipher_decrypt_unaligned : cipher->cia_decrypt;
 
 	switch (tfm->crt_cipher.cit_mode) {
 	case CRYPTO_TFM_MODE_ECB:
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index c533c0a291af..6f9fb27b2071 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -69,5 +69,10 @@ static inline void *crypto_instance_ctx(struct crypto_instance *inst)
 	return inst->__ctx;
 }
 
+static inline struct cipher_alg *crypto_cipher_alg(struct crypto_cipher *tfm)
+{
+	return &crypto_cipher_tfm(tfm)->__crt_alg->cra_cipher;
+}
+
 #endif	/* _CRYPTO_ALGAPI_H */
 
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 8e9c407b00d2..fdecee83878c 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -224,6 +224,8 @@ struct cipher_tfm {
 			   struct scatterlist *src,
 			   unsigned int nbytes, u8 *iv);
 	void (*cit_xor_block)(u8 *dst, const u8 *src);
+	void (*cit_encrypt_one)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
+	void (*cit_decrypt_one)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
 };
 
 struct digest_tfm {
@@ -268,6 +270,8 @@ struct crypto_tfm {
 	void *__crt_ctx[] CRYPTO_MINALIGN_ATTR;
 };
 
+#define crypto_cipher crypto_tfm
+
 enum {
 	CRYPTOA_UNSPEC,
 	CRYPTOA_ALG,
@@ -347,6 +351,21 @@ static inline unsigned int crypto_tfm_alg_alignmask(struct crypto_tfm *tfm)
 	return tfm->__crt_alg->cra_alignmask;
 }
 
+static inline u32 crypto_tfm_get_flags(struct crypto_tfm *tfm)
+{
+	return tfm->crt_flags;
+}
+
+static inline void crypto_tfm_set_flags(struct crypto_tfm *tfm, u32 flags)
+{
+	tfm->crt_flags |= flags;
+}
+
+static inline void crypto_tfm_clear_flags(struct crypto_tfm *tfm, u32 flags)
+{
+	tfm->crt_flags &= ~flags;
+}
+
 static inline void *crypto_tfm_ctx(struct crypto_tfm *tfm)
 {
 	return tfm->__crt_ctx;
@@ -361,6 +380,83 @@ static inline unsigned int crypto_tfm_ctx_alignment(void)
 /*
  * API wrappers.
  */
+static inline struct crypto_cipher *__crypto_cipher_cast(struct crypto_tfm *tfm)
+{
+	return (struct crypto_cipher *)tfm;
+}
+
+static inline struct crypto_cipher *crypto_cipher_cast(struct crypto_tfm *tfm)
+{
+	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
+	return __crypto_cipher_cast(tfm);
+}
+
+static inline struct crypto_cipher *crypto_alloc_cipher(const char *alg_name,
+							u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_CIPHER;
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	return __crypto_cipher_cast(crypto_alloc_base(alg_name, type, mask));
+}
+
+static inline struct crypto_tfm *crypto_cipher_tfm(struct crypto_cipher *tfm)
+{
+	return tfm;
+}
+
+static inline void crypto_free_cipher(struct crypto_cipher *tfm)
+{
+	crypto_free_tfm(crypto_cipher_tfm(tfm));
+}
+
+static inline struct cipher_tfm *crypto_cipher_crt(struct crypto_cipher *tfm)
+{
+	return &crypto_cipher_tfm(tfm)->crt_cipher;
+}
+
+static inline unsigned int crypto_cipher_blocksize(struct crypto_cipher *tfm)
+{
+	return crypto_tfm_alg_blocksize(crypto_cipher_tfm(tfm));
+}
+
+static inline unsigned int crypto_cipher_alignmask(struct crypto_cipher *tfm)
+{
+	return crypto_tfm_alg_alignmask(crypto_cipher_tfm(tfm));
+}
+
+static inline u32 crypto_cipher_get_flags(struct crypto_cipher *tfm)
+{
+	return crypto_tfm_get_flags(crypto_cipher_tfm(tfm));
+}
+
+static inline void crypto_cipher_set_flags(struct crypto_cipher *tfm,
+					   u32 flags)
+{
+	crypto_tfm_set_flags(crypto_cipher_tfm(tfm), flags);
+}
+
+static inline void crypto_cipher_clear_flags(struct crypto_cipher *tfm,
+					     u32 flags)
+{
+	crypto_tfm_clear_flags(crypto_cipher_tfm(tfm), flags);
+}
+
+static inline void crypto_cipher_encrypt_one(struct crypto_cipher *tfm,
+					     u8 *dst, const u8 *src)
+{
+	crypto_cipher_crt(tfm)->cit_encrypt_one(crypto_cipher_tfm(tfm),
+						dst, src);
+}
+
+static inline void crypto_cipher_decrypt_one(struct crypto_cipher *tfm,
+					     u8 *dst, const u8 *src)
+{
+	crypto_cipher_crt(tfm)->cit_decrypt_one(crypto_cipher_tfm(tfm),
+						dst, src);
+}
+
 static inline void crypto_digest_init(struct crypto_tfm *tfm)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-- 
cgit v1.2.3


From 5c64097aa0f6dc4f27718ef47ca9a12538d62860 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 12 Aug 2006 21:56:17 +1000
Subject: [CRYPTO] scatterwalk: Prepare for block ciphers

This patch prepares the scatterwalk code for use by the new block cipher
type.

Firstly it halves the size of scatter_walk on 32-bit platforms.  This
is important as we allocate at least two of these objects on the stack
for each block cipher operation.

It also exports the symbols since the block cipher code can be built as
a module.

Finally there is a hack in scatterwalk_unmap that relies on progress
being made.  Unfortunately, for hardware crypto we can't guarantee
progress to be made since the hardware can fail.

So this also gets rid of the hack by not advancing the address returned
by scatterwalk_map.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/cipher.c         | 27 ++++++++-------
 crypto/scatterwalk.c    | 89 ++++++++++++++++++++++---------------------------
 crypto/scatterwalk.h    | 48 +++++++++++++++-----------
 include/crypto/algapi.h |  5 +++
 4 files changed, 87 insertions(+), 82 deletions(-)

(limited to 'include')

diff --git a/crypto/cipher.c b/crypto/cipher.c
index d8ca0ec8d0be..326461780673 100644
--- a/crypto/cipher.c
+++ b/crypto/cipher.c
@@ -45,15 +45,10 @@ static unsigned int crypt_slow(const struct cipher_desc *desc,
 	u8 buffer[bsize * 2 + alignmask];
 	u8 *src = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1);
 	u8 *dst = src + bsize;
-	unsigned int n;
-
-	n = scatterwalk_copychunks(src, in, bsize, 0);
-	scatterwalk_advance(in, n);
 
+	scatterwalk_copychunks(src, in, bsize, 0);
 	desc->prfn(desc, dst, src, bsize);
-
-	n = scatterwalk_copychunks(dst, out, bsize, 1);
-	scatterwalk_advance(out, n);
+	scatterwalk_copychunks(dst, out, bsize, 1);
 
 	return bsize;
 }
@@ -64,12 +59,16 @@ static inline unsigned int crypt_fast(const struct cipher_desc *desc,
 				      unsigned int nbytes, u8 *tmp)
 {
 	u8 *src, *dst;
+	u8 *real_src, *real_dst;
+
+	real_src = scatterwalk_map(in, 0);
+	real_dst = scatterwalk_map(out, 1);
 
-	src = in->data;
-	dst = scatterwalk_samebuf(in, out) ? src : out->data;
+	src = real_src;
+	dst = scatterwalk_samebuf(in, out) ? src : real_dst;
 
 	if (tmp) {
-		memcpy(tmp, in->data, nbytes);
+		memcpy(tmp, src, nbytes);
 		src = tmp;
 		dst = tmp;
 	}
@@ -77,7 +76,10 @@ static inline unsigned int crypt_fast(const struct cipher_desc *desc,
 	nbytes = desc->prfn(desc, dst, src, nbytes);
 
 	if (tmp)
-		memcpy(out->data, tmp, nbytes);
+		memcpy(real_dst, tmp, nbytes);
+
+	scatterwalk_unmap(real_src, 0);
+	scatterwalk_unmap(real_dst, 1);
 
 	scatterwalk_advance(in, nbytes);
 	scatterwalk_advance(out, nbytes);
@@ -126,9 +128,6 @@ static int crypt(const struct cipher_desc *desc,
 			tmp = (u8 *)buffer;
 		}
 
-		scatterwalk_map(&walk_in, 0);
-		scatterwalk_map(&walk_out, 1);
-
 		n = scatterwalk_clamp(&walk_in, n);
 		n = scatterwalk_clamp(&walk_out, n);
 
diff --git a/crypto/scatterwalk.c b/crypto/scatterwalk.c
index 2953e2cc56f0..35172d3f043b 100644
--- a/crypto/scatterwalk.c
+++ b/crypto/scatterwalk.c
@@ -15,9 +15,11 @@
  */
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/module.h>
 #include <linux/pagemap.h>
 #include <linux/highmem.h>
-#include <asm/scatterlist.h>
+#include <linux/scatterlist.h>
+
 #include "internal.h"
 #include "scatterwalk.h"
 
@@ -27,88 +29,77 @@ enum km_type crypto_km_types[] = {
 	KM_SOFTIRQ0,
 	KM_SOFTIRQ1,
 };
+EXPORT_SYMBOL_GPL(crypto_km_types);
 
-static void memcpy_dir(void *buf, void *sgdata, size_t nbytes, int out)
+static inline void memcpy_dir(void *buf, void *sgdata, size_t nbytes, int out)
 {
-	if (out)
-		memcpy(sgdata, buf, nbytes);
-	else
-		memcpy(buf, sgdata, nbytes);
+	void *src = out ? buf : sgdata;
+	void *dst = out ? sgdata : buf;
+
+	memcpy(dst, src, nbytes);
 }
 
 void scatterwalk_start(struct scatter_walk *walk, struct scatterlist *sg)
 {
-	unsigned int rest_of_page;
-
 	walk->sg = sg;
 
-	walk->page = sg->page;
-	walk->len_this_segment = sg->length;
-
 	BUG_ON(!sg->length);
 
-	rest_of_page = PAGE_CACHE_SIZE - (sg->offset & (PAGE_CACHE_SIZE - 1));
-	walk->len_this_page = min(sg->length, rest_of_page);
 	walk->offset = sg->offset;
 }
+EXPORT_SYMBOL_GPL(scatterwalk_start);
 
-void scatterwalk_map(struct scatter_walk *walk, int out)
-{
-	walk->data = crypto_kmap(walk->page, out) + walk->offset;
-}
-
-static inline void scatterwalk_unmap(struct scatter_walk *walk, int out)
+void *scatterwalk_map(struct scatter_walk *walk, int out)
 {
-	/* walk->data may be pointing the first byte of the next page;
-	   however, we know we transfered at least one byte.  So,
-	   walk->data - 1 will be a virtual address in the mapped page. */
-	crypto_kunmap(walk->data - 1, out);
+	return crypto_kmap(scatterwalk_page(walk), out) +
+	       offset_in_page(walk->offset);
 }
+EXPORT_SYMBOL_GPL(scatterwalk_map);
 
 static void scatterwalk_pagedone(struct scatter_walk *walk, int out,
 				 unsigned int more)
 {
 	if (out)
-		flush_dcache_page(walk->page);
+		flush_dcache_page(scatterwalk_page(walk));
 
 	if (more) {
-		walk->len_this_segment -= walk->len_this_page;
-
-		if (walk->len_this_segment) {
-			walk->page++;
-			walk->len_this_page = min(walk->len_this_segment,
-						  (unsigned)PAGE_CACHE_SIZE);
-			walk->offset = 0;
-		}
-		else
+		walk->offset += PAGE_SIZE - 1;
+		walk->offset &= PAGE_MASK;
+		if (walk->offset >= walk->sg->offset + walk->sg->length)
 			scatterwalk_start(walk, sg_next(walk->sg));
 	}
 }
 
 void scatterwalk_done(struct scatter_walk *walk, int out, int more)
 {
-	scatterwalk_unmap(walk, out);
-	if (walk->len_this_page == 0 || !more)
+	if (!offset_in_page(walk->offset) || !more)
 		scatterwalk_pagedone(walk, out, more);
 }
+EXPORT_SYMBOL_GPL(scatterwalk_done);
 
-/*
- * Do not call this unless the total length of all of the fragments
- * has been verified as multiple of the block size.
- */
-int scatterwalk_copychunks(void *buf, struct scatter_walk *walk,
-			   size_t nbytes, int out)
+void scatterwalk_copychunks(void *buf, struct scatter_walk *walk,
+			    size_t nbytes, int out)
 {
-	while (nbytes > walk->len_this_page) {
-		memcpy_dir(buf, walk->data, walk->len_this_page, out);
-		buf += walk->len_this_page;
-		nbytes -= walk->len_this_page;
+	for (;;) {
+		unsigned int len_this_page = scatterwalk_pagelen(walk);
+		u8 *vaddr;
+
+		if (len_this_page > nbytes)
+			len_this_page = nbytes;
+
+		vaddr = scatterwalk_map(walk, out);
+		memcpy_dir(buf, vaddr, len_this_page, out);
+		scatterwalk_unmap(vaddr, out);
+
+		if (nbytes == len_this_page)
+			break;
+
+		buf += len_this_page;
+		nbytes -= len_this_page;
 
-		scatterwalk_unmap(walk, out);
 		scatterwalk_pagedone(walk, out, 1);
-		scatterwalk_map(walk, out);
 	}
 
-	memcpy_dir(buf, walk->data, nbytes, out);
-	return nbytes;
+	scatterwalk_advance(walk, nbytes);
 }
+EXPORT_SYMBOL_GPL(scatterwalk_copychunks);
diff --git a/crypto/scatterwalk.h b/crypto/scatterwalk.h
index e79925c474a3..ace595a2e119 100644
--- a/crypto/scatterwalk.h
+++ b/crypto/scatterwalk.h
@@ -14,17 +14,11 @@
 
 #ifndef _CRYPTO_SCATTERWALK_H
 #define _CRYPTO_SCATTERWALK_H
+
 #include <linux/mm.h>
-#include <asm/scatterlist.h>
+#include <linux/scatterlist.h>
 
-struct scatter_walk {
-	struct scatterlist	*sg;
-	struct page		*page;
-	void			*data;
-	unsigned int		len_this_page;
-	unsigned int		len_this_segment;
-	unsigned int		offset;
-};
+#include "internal.h"
 
 /* Define sg_next is an inline routine now in case we want to change
    scatterlist to a linked list later. */
@@ -33,26 +27,31 @@ static inline struct scatterlist *sg_next(struct scatterlist *sg)
 	return sg + 1;
 }
 
-static inline int scatterwalk_samebuf(struct scatter_walk *walk_in,
-				      struct scatter_walk *walk_out)
+static inline unsigned long scatterwalk_samebuf(struct scatter_walk *walk_in,
+						struct scatter_walk *walk_out)
 {
-	return walk_in->page == walk_out->page &&
-	       walk_in->offset == walk_out->offset;
+	return !(((walk_in->sg->page - walk_out->sg->page) << PAGE_SHIFT) +
+		 (int)(walk_in->offset - walk_out->offset));
+}
+
+static inline unsigned int scatterwalk_pagelen(struct scatter_walk *walk)
+{
+	unsigned int len = walk->sg->offset + walk->sg->length - walk->offset;
+	unsigned int len_this_page = offset_in_page(~walk->offset) + 1;
+	return len_this_page > len ? len : len_this_page;
 }
 
 static inline unsigned int scatterwalk_clamp(struct scatter_walk *walk,
 					     unsigned int nbytes)
 {
-	return nbytes > walk->len_this_page ? walk->len_this_page : nbytes;
+	unsigned int len_this_page = scatterwalk_pagelen(walk);
+	return nbytes > len_this_page ? len_this_page : nbytes;
 }
 
 static inline void scatterwalk_advance(struct scatter_walk *walk,
 				       unsigned int nbytes)
 {
-	walk->data += nbytes;
 	walk->offset += nbytes;
-	walk->len_this_page -= nbytes;
-	walk->len_this_segment -= nbytes;
 }
 
 static inline unsigned int scatterwalk_aligned(struct scatter_walk *walk,
@@ -61,9 +60,20 @@ static inline unsigned int scatterwalk_aligned(struct scatter_walk *walk,
 	return !(walk->offset & alignmask);
 }
 
+static inline struct page *scatterwalk_page(struct scatter_walk *walk)
+{
+	return walk->sg->page + (walk->offset >> PAGE_SHIFT);
+}
+
+static inline void scatterwalk_unmap(void *vaddr, int out)
+{
+	crypto_kunmap(vaddr, out);
+}
+
 void scatterwalk_start(struct scatter_walk *walk, struct scatterlist *sg);
-int scatterwalk_copychunks(void *buf, struct scatter_walk *walk, size_t nbytes, int out);
-void scatterwalk_map(struct scatter_walk *walk, int out);
+void scatterwalk_copychunks(void *buf, struct scatter_walk *walk,
+			    size_t nbytes, int out);
+void *scatterwalk_map(struct scatter_walk *walk, int out);
 void scatterwalk_done(struct scatter_walk *walk, int out, int more);
 
 #endif  /* _CRYPTO_SCATTERWALK_H */
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 6f9fb27b2071..f21ae672e8a8 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -50,6 +50,11 @@ struct crypto_spawn {
 	struct crypto_instance *inst;
 };
 
+struct scatter_walk {
+	struct scatterlist *sg;
+	unsigned int offset;
+};
+
 int crypto_register_template(struct crypto_template *tmpl);
 void crypto_unregister_template(struct crypto_template *tmpl);
 struct crypto_template *crypto_lookup_template(const char *name);
-- 
cgit v1.2.3


From 5cde0af2a9825dd1edaca233bd9590566579ef21 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 22 Aug 2006 00:07:53 +1000
Subject: [CRYPTO] cipher: Added block cipher type

This patch adds the new type of block ciphers.  Unlike current cipher
algorithms which operate on a single block at a time, block ciphers
operate on an arbitrarily long linear area of data.  As it is block-based,
it will skip any data remaining at the end which cannot form a block.

The block cipher has one major difference when compared to the existing
block cipher implementation.  The sg walking is now performed by the
algorithm rather than the cipher mid-layer.  This is needed for drivers
that directly support sg lists.  It also improves performance for all
algorithms as it reduces the total number of indirect calls by one.

In future the existing cipher algorithm will be converted to only have
a single-block interface.  This will be done after all existing users
have switched over to the new block cipher type.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Kconfig          |   4 +
 crypto/Makefile         |   2 +
 crypto/blkcipher.c      | 405 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/crypto/algapi.h |  65 ++++++++
 include/linux/crypto.h  | 179 +++++++++++++++++++++
 5 files changed, 655 insertions(+)
 create mode 100644 crypto/blkcipher.c

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index 4ce509dba329..68790ad7308d 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -16,6 +16,10 @@ config CRYPTO_ALGAPI
 	help
 	  This option provides the API for cryptographic algorithms.
 
+config CRYPTO_BLKCIPHER
+	tristate
+	select CRYPTO_ALGAPI
+
 config CRYPTO_MANAGER
 	tristate "Cryptographic algorithm manager"
 	select CRYPTO_ALGAPI
diff --git a/crypto/Makefile b/crypto/Makefile
index b8745f3d3595..b5051951c636 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -8,6 +8,8 @@ crypto_algapi-$(CONFIG_PROC_FS) += proc.o
 crypto_algapi-objs := algapi.o $(crypto_algapi-y)
 obj-$(CONFIG_CRYPTO_ALGAPI) += crypto_algapi.o
 
+obj-$(CONFIG_CRYPTO_BLKCIPHER) += blkcipher.o
+
 obj-$(CONFIG_CRYPTO_MANAGER) += cryptomgr.o
 obj-$(CONFIG_CRYPTO_HMAC) += hmac.o
 obj-$(CONFIG_CRYPTO_NULL) += crypto_null.o
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
new file mode 100644
index 000000000000..034c939bf91a
--- /dev/null
+++ b/crypto/blkcipher.c
@@ -0,0 +1,405 @@
+/*
+ * Block chaining cipher operations.
+ * 
+ * Generic encrypt/decrypt wrapper for ciphers, handles operations across
+ * multiple page boundaries by using temporary blocks.  In user context,
+ * the kernel is given a chance to schedule us once per page.
+ *
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+
+#include <linux/crypto.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/scatterlist.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+
+#include "internal.h"
+#include "scatterwalk.h"
+
+enum {
+	BLKCIPHER_WALK_PHYS = 1 << 0,
+	BLKCIPHER_WALK_SLOW = 1 << 1,
+	BLKCIPHER_WALK_COPY = 1 << 2,
+	BLKCIPHER_WALK_DIFF = 1 << 3,
+};
+
+static int blkcipher_walk_next(struct blkcipher_desc *desc,
+			       struct blkcipher_walk *walk);
+static int blkcipher_walk_first(struct blkcipher_desc *desc,
+				struct blkcipher_walk *walk);
+
+static inline void blkcipher_map_src(struct blkcipher_walk *walk)
+{
+	walk->src.virt.addr = scatterwalk_map(&walk->in, 0);
+}
+
+static inline void blkcipher_map_dst(struct blkcipher_walk *walk)
+{
+	walk->dst.virt.addr = scatterwalk_map(&walk->out, 1);
+}
+
+static inline void blkcipher_unmap_src(struct blkcipher_walk *walk)
+{
+	scatterwalk_unmap(walk->src.virt.addr, 0);
+}
+
+static inline void blkcipher_unmap_dst(struct blkcipher_walk *walk)
+{
+	scatterwalk_unmap(walk->dst.virt.addr, 1);
+}
+
+static inline u8 *blkcipher_get_spot(u8 *start, unsigned int len)
+{
+	if (offset_in_page(start + len) < len)
+		return (u8 *)((unsigned long)(start + len) & PAGE_MASK);
+	return start;
+}
+
+static inline unsigned int blkcipher_done_slow(struct crypto_blkcipher *tfm,
+					       struct blkcipher_walk *walk,
+					       unsigned int bsize)
+{
+	u8 *addr;
+	unsigned int alignmask = crypto_blkcipher_alignmask(tfm);
+
+	addr = (u8 *)ALIGN((unsigned long)walk->buffer, alignmask + 1);
+	addr = blkcipher_get_spot(addr, bsize);
+	scatterwalk_copychunks(addr, &walk->out, bsize, 1);
+	return bsize;
+}
+
+static inline unsigned int blkcipher_done_fast(struct blkcipher_walk *walk,
+					       unsigned int n)
+{
+	n = walk->nbytes - n;
+
+	if (walk->flags & BLKCIPHER_WALK_COPY) {
+		blkcipher_map_dst(walk);
+		memcpy(walk->dst.virt.addr, walk->page, n);
+		blkcipher_unmap_dst(walk);
+	} else if (!(walk->flags & BLKCIPHER_WALK_PHYS)) {
+		blkcipher_unmap_src(walk);
+		if (walk->flags & BLKCIPHER_WALK_DIFF)
+			blkcipher_unmap_dst(walk);
+	}
+
+	scatterwalk_advance(&walk->in, n);
+	scatterwalk_advance(&walk->out, n);
+
+	return n;
+}
+
+int blkcipher_walk_done(struct blkcipher_desc *desc,
+			struct blkcipher_walk *walk, int err)
+{
+	struct crypto_blkcipher *tfm = desc->tfm;
+	unsigned int nbytes = 0;
+
+	if (likely(err >= 0)) {
+		unsigned int bsize = crypto_blkcipher_blocksize(tfm);
+		unsigned int n;
+
+		if (likely(!(walk->flags & BLKCIPHER_WALK_SLOW)))
+			n = blkcipher_done_fast(walk, err);
+		else
+			n = blkcipher_done_slow(tfm, walk, bsize);
+
+		nbytes = walk->total - n;
+		err = 0;
+	}
+
+	scatterwalk_done(&walk->in, 0, nbytes);
+	scatterwalk_done(&walk->out, 1, nbytes);
+
+	walk->total = nbytes;
+	walk->nbytes = nbytes;
+
+	if (nbytes) {
+		crypto_yield(desc->flags);
+		return blkcipher_walk_next(desc, walk);
+	}
+
+	if (walk->iv != desc->info)
+		memcpy(desc->info, walk->iv, crypto_blkcipher_ivsize(tfm));
+	if (walk->buffer != walk->page)
+		kfree(walk->buffer);
+	if (walk->page)
+		free_page((unsigned long)walk->page);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(blkcipher_walk_done);
+
+static inline int blkcipher_next_slow(struct blkcipher_desc *desc,
+				      struct blkcipher_walk *walk,
+				      unsigned int bsize,
+				      unsigned int alignmask)
+{
+	unsigned int n;
+
+	if (walk->buffer)
+		goto ok;
+
+	walk->buffer = walk->page;
+	if (walk->buffer)
+		goto ok;
+
+	n = bsize * 2 + (alignmask & ~(crypto_tfm_ctx_alignment() - 1));
+	walk->buffer = kmalloc(n, GFP_ATOMIC);
+	if (!walk->buffer)
+		return blkcipher_walk_done(desc, walk, -ENOMEM);
+
+ok:
+	walk->dst.virt.addr = (u8 *)ALIGN((unsigned long)walk->buffer,
+					  alignmask + 1);
+	walk->dst.virt.addr = blkcipher_get_spot(walk->dst.virt.addr, bsize);
+	walk->src.virt.addr = blkcipher_get_spot(walk->dst.virt.addr + bsize,
+						 bsize);
+
+	scatterwalk_copychunks(walk->src.virt.addr, &walk->in, bsize, 0);
+
+	walk->nbytes = bsize;
+	walk->flags |= BLKCIPHER_WALK_SLOW;
+
+	return 0;
+}
+
+static inline int blkcipher_next_copy(struct blkcipher_walk *walk)
+{
+	u8 *tmp = walk->page;
+
+	blkcipher_map_src(walk);
+	memcpy(tmp, walk->src.virt.addr, walk->nbytes);
+	blkcipher_unmap_src(walk);
+
+	walk->src.virt.addr = tmp;
+	walk->dst.virt.addr = tmp;
+
+	return 0;
+}
+
+static inline int blkcipher_next_fast(struct blkcipher_desc *desc,
+				      struct blkcipher_walk *walk)
+{
+	unsigned long diff;
+
+	walk->src.phys.page = scatterwalk_page(&walk->in);
+	walk->src.phys.offset = offset_in_page(walk->in.offset);
+	walk->dst.phys.page = scatterwalk_page(&walk->out);
+	walk->dst.phys.offset = offset_in_page(walk->out.offset);
+
+	if (walk->flags & BLKCIPHER_WALK_PHYS)
+		return 0;
+
+	diff = walk->src.phys.offset - walk->dst.phys.offset;
+	diff |= walk->src.virt.page - walk->dst.virt.page;
+
+	blkcipher_map_src(walk);
+	walk->dst.virt.addr = walk->src.virt.addr;
+
+	if (diff) {
+		walk->flags |= BLKCIPHER_WALK_DIFF;
+		blkcipher_map_dst(walk);
+	}
+
+	return 0;
+}
+
+static int blkcipher_walk_next(struct blkcipher_desc *desc,
+			       struct blkcipher_walk *walk)
+{
+	struct crypto_blkcipher *tfm = desc->tfm;
+	unsigned int alignmask = crypto_blkcipher_alignmask(tfm);
+	unsigned int bsize = crypto_blkcipher_blocksize(tfm);
+	unsigned int n;
+	int err;
+
+	n = walk->total;
+	if (unlikely(n < bsize)) {
+		desc->flags |= CRYPTO_TFM_RES_BAD_BLOCK_LEN;
+		return blkcipher_walk_done(desc, walk, -EINVAL);
+	}
+
+	walk->flags &= ~(BLKCIPHER_WALK_SLOW | BLKCIPHER_WALK_COPY |
+			 BLKCIPHER_WALK_DIFF);
+	if (!scatterwalk_aligned(&walk->in, alignmask) ||
+	    !scatterwalk_aligned(&walk->out, alignmask)) {
+		walk->flags |= BLKCIPHER_WALK_COPY;
+		if (!walk->page) {
+			walk->page = (void *)__get_free_page(GFP_ATOMIC);
+			if (!walk->page)
+				n = 0;
+		}
+	}
+
+	n = scatterwalk_clamp(&walk->in, n);
+	n = scatterwalk_clamp(&walk->out, n);
+
+	if (unlikely(n < bsize)) {
+		err = blkcipher_next_slow(desc, walk, bsize, alignmask);
+		goto set_phys_lowmem;
+	}
+
+	walk->nbytes = n;
+	if (walk->flags & BLKCIPHER_WALK_COPY) {
+		err = blkcipher_next_copy(walk);
+		goto set_phys_lowmem;
+	}
+
+	return blkcipher_next_fast(desc, walk);
+
+set_phys_lowmem:
+	if (walk->flags & BLKCIPHER_WALK_PHYS) {
+		walk->src.phys.page = virt_to_page(walk->src.virt.addr);
+		walk->dst.phys.page = virt_to_page(walk->dst.virt.addr);
+		walk->src.phys.offset &= PAGE_SIZE - 1;
+		walk->dst.phys.offset &= PAGE_SIZE - 1;
+	}
+	return err;
+}
+
+static inline int blkcipher_copy_iv(struct blkcipher_walk *walk,
+				    struct crypto_blkcipher *tfm,
+				    unsigned int alignmask)
+{
+	unsigned bs = crypto_blkcipher_blocksize(tfm);
+	unsigned int ivsize = crypto_blkcipher_ivsize(tfm);
+	unsigned int size = bs * 2 + ivsize + max(bs, ivsize) - (alignmask + 1);
+	u8 *iv;
+
+	size += alignmask & ~(crypto_tfm_ctx_alignment() - 1);
+	walk->buffer = kmalloc(size, GFP_ATOMIC);
+	if (!walk->buffer)
+		return -ENOMEM;
+
+	iv = (u8 *)ALIGN((unsigned long)walk->buffer, alignmask + 1);
+	iv = blkcipher_get_spot(iv, bs) + bs;
+	iv = blkcipher_get_spot(iv, bs) + bs;
+	iv = blkcipher_get_spot(iv, ivsize);
+
+	walk->iv = memcpy(iv, walk->iv, ivsize);
+	return 0;
+}
+
+int blkcipher_walk_virt(struct blkcipher_desc *desc,
+			struct blkcipher_walk *walk)
+{
+	walk->flags &= ~BLKCIPHER_WALK_PHYS;
+	return blkcipher_walk_first(desc, walk);
+}
+EXPORT_SYMBOL_GPL(blkcipher_walk_virt);
+
+int blkcipher_walk_phys(struct blkcipher_desc *desc,
+			struct blkcipher_walk *walk)
+{
+	walk->flags |= BLKCIPHER_WALK_PHYS;
+	return blkcipher_walk_first(desc, walk);
+}
+EXPORT_SYMBOL_GPL(blkcipher_walk_phys);
+
+static int blkcipher_walk_first(struct blkcipher_desc *desc,
+				struct blkcipher_walk *walk)
+{
+	struct crypto_blkcipher *tfm = desc->tfm;
+	unsigned int alignmask = crypto_blkcipher_alignmask(tfm);
+
+	walk->nbytes = walk->total;
+	if (unlikely(!walk->total))
+		return 0;
+
+	walk->buffer = NULL;
+	walk->iv = desc->info;
+	if (unlikely(((unsigned long)walk->iv & alignmask))) {
+		int err = blkcipher_copy_iv(walk, tfm, alignmask);
+		if (err)
+			return err;
+	}
+
+	scatterwalk_start(&walk->in, walk->in.sg);
+	scatterwalk_start(&walk->out, walk->out.sg);
+	walk->page = NULL;
+
+	return blkcipher_walk_next(desc, walk);
+}
+
+static int setkey(struct crypto_tfm *tfm, const u8 *key,
+		  unsigned int keylen)
+{
+	struct blkcipher_alg *cipher = &tfm->__crt_alg->cra_blkcipher;
+
+	if (keylen < cipher->min_keysize || keylen > cipher->max_keysize) {
+		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
+		return -EINVAL;
+	}
+
+	return cipher->setkey(tfm, key, keylen);
+}
+
+static unsigned int crypto_blkcipher_ctxsize(struct crypto_alg *alg)
+{
+	struct blkcipher_alg *cipher = &alg->cra_blkcipher;
+	unsigned int len = alg->cra_ctxsize;
+
+	if (cipher->ivsize) {
+		len = ALIGN(len, (unsigned long)alg->cra_alignmask + 1);
+		len += cipher->ivsize;
+	}
+
+	return len;
+}
+
+static int crypto_init_blkcipher_ops(struct crypto_tfm *tfm)
+{
+	struct blkcipher_tfm *crt = &tfm->crt_blkcipher;
+	struct blkcipher_alg *alg = &tfm->__crt_alg->cra_blkcipher;
+	unsigned long align = crypto_tfm_alg_alignmask(tfm) + 1;
+	unsigned long addr;
+
+	if (alg->ivsize > PAGE_SIZE / 8)
+		return -EINVAL;
+
+	crt->setkey = setkey;
+	crt->encrypt = alg->encrypt;
+	crt->decrypt = alg->decrypt;
+
+	addr = (unsigned long)crypto_tfm_ctx(tfm);
+	addr = ALIGN(addr, align);
+	addr += ALIGN(tfm->__crt_alg->cra_ctxsize, align);
+	crt->iv = (void *)addr;
+
+	return 0;
+}
+
+static void crypto_blkcipher_show(struct seq_file *m, struct crypto_alg *alg)
+	__attribute_used__;
+static void crypto_blkcipher_show(struct seq_file *m, struct crypto_alg *alg)
+{
+	seq_printf(m, "type         : blkcipher\n");
+	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
+	seq_printf(m, "min keysize  : %u\n", alg->cra_blkcipher.min_keysize);
+	seq_printf(m, "max keysize  : %u\n", alg->cra_blkcipher.max_keysize);
+	seq_printf(m, "ivsize       : %u\n", alg->cra_blkcipher.ivsize);
+}
+
+const struct crypto_type crypto_blkcipher_type = {
+	.ctxsize = crypto_blkcipher_ctxsize,
+	.init = crypto_init_blkcipher_ops,
+#ifdef CONFIG_PROC_FS
+	.show = crypto_blkcipher_show,
+#endif
+};
+EXPORT_SYMBOL_GPL(crypto_blkcipher_type);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Generic block chaining cipher type");
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index f21ae672e8a8..f3946baf0c07 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -55,6 +55,34 @@ struct scatter_walk {
 	unsigned int offset;
 };
 
+struct blkcipher_walk {
+	union {
+		struct {
+			struct page *page;
+			unsigned long offset;
+		} phys;
+
+		struct {
+			u8 *page;
+			u8 *addr;
+		} virt;
+	} src, dst;
+
+	struct scatter_walk in;
+	unsigned int nbytes;
+
+	struct scatter_walk out;
+	unsigned int total;
+
+	void *page;
+	u8 *buffer;
+	u8 *iv;
+
+	int flags;
+};
+
+extern const struct crypto_type crypto_blkcipher_type;
+
 int crypto_register_template(struct crypto_template *tmpl);
 void crypto_unregister_template(struct crypto_template *tmpl);
 struct crypto_template *crypto_lookup_template(const char *name);
@@ -69,15 +97,52 @@ struct crypto_alg *crypto_get_attr_alg(void *param, unsigned int len,
 struct crypto_instance *crypto_alloc_instance(const char *name,
 					      struct crypto_alg *alg);
 
+int blkcipher_walk_done(struct blkcipher_desc *desc,
+			struct blkcipher_walk *walk, int err);
+int blkcipher_walk_virt(struct blkcipher_desc *desc,
+			struct blkcipher_walk *walk);
+int blkcipher_walk_phys(struct blkcipher_desc *desc,
+			struct blkcipher_walk *walk);
+
+static inline void *crypto_tfm_ctx_aligned(struct crypto_tfm *tfm)
+{
+	unsigned long addr = (unsigned long)crypto_tfm_ctx(tfm);
+	unsigned long align = crypto_tfm_alg_alignmask(tfm);
+
+	if (align <= crypto_tfm_ctx_alignment())
+		align = 1;
+	return (void *)ALIGN(addr, align);
+}
+
 static inline void *crypto_instance_ctx(struct crypto_instance *inst)
 {
 	return inst->__ctx;
 }
 
+static inline void *crypto_blkcipher_ctx(struct crypto_blkcipher *tfm)
+{
+	return crypto_tfm_ctx(&tfm->base);
+}
+
+static inline void *crypto_blkcipher_ctx_aligned(struct crypto_blkcipher *tfm)
+{
+	return crypto_tfm_ctx_aligned(&tfm->base);
+}
+
 static inline struct cipher_alg *crypto_cipher_alg(struct crypto_cipher *tfm)
 {
 	return &crypto_cipher_tfm(tfm)->__crt_alg->cra_cipher;
 }
 
+static inline void blkcipher_walk_init(struct blkcipher_walk *walk,
+				       struct scatterlist *dst,
+				       struct scatterlist *src,
+				       unsigned int nbytes)
+{
+	walk->in.sg = src;
+	walk->out.sg = dst;
+	walk->total = nbytes;
+}
+
 #endif	/* _CRYPTO_ALGAPI_H */
 
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index fdecee83878c..5a5466d518e8 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -32,6 +32,7 @@
 #define CRYPTO_ALG_TYPE_MASK		0x0000000f
 #define CRYPTO_ALG_TYPE_CIPHER		0x00000001
 #define CRYPTO_ALG_TYPE_DIGEST		0x00000002
+#define CRYPTO_ALG_TYPE_BLKCIPHER	0x00000003
 #define CRYPTO_ALG_TYPE_COMPRESS	0x00000004
 
 #define CRYPTO_ALG_LARVAL		0x00000010
@@ -89,9 +90,16 @@
 #endif
 
 struct scatterlist;
+struct crypto_blkcipher;
 struct crypto_tfm;
 struct crypto_type;
 
+struct blkcipher_desc {
+	struct crypto_blkcipher *tfm;
+	void *info;
+	u32 flags;
+};
+
 struct cipher_desc {
 	struct crypto_tfm *tfm;
 	void (*crfn)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
@@ -104,6 +112,21 @@ struct cipher_desc {
  * Algorithms: modular crypto algorithm implementations, managed
  * via crypto_register_alg() and crypto_unregister_alg().
  */
+struct blkcipher_alg {
+	int (*setkey)(struct crypto_tfm *tfm, const u8 *key,
+	              unsigned int keylen);
+	int (*encrypt)(struct blkcipher_desc *desc,
+		       struct scatterlist *dst, struct scatterlist *src,
+		       unsigned int nbytes);
+	int (*decrypt)(struct blkcipher_desc *desc,
+		       struct scatterlist *dst, struct scatterlist *src,
+		       unsigned int nbytes);
+
+	unsigned int min_keysize;
+	unsigned int max_keysize;
+	unsigned int ivsize;
+};
+
 struct cipher_alg {
 	unsigned int cia_min_keysize;
 	unsigned int cia_max_keysize;
@@ -143,6 +166,7 @@ struct compress_alg {
 			      unsigned int slen, u8 *dst, unsigned int *dlen);
 };
 
+#define cra_blkcipher	cra_u.blkcipher
 #define cra_cipher	cra_u.cipher
 #define cra_digest	cra_u.digest
 #define cra_compress	cra_u.compress
@@ -165,6 +189,7 @@ struct crypto_alg {
 	const struct crypto_type *cra_type;
 
 	union {
+		struct blkcipher_alg blkcipher;
 		struct cipher_alg cipher;
 		struct digest_alg digest;
 		struct compress_alg compress;
@@ -201,6 +226,16 @@ static inline int crypto_alg_available(const char *name, u32 flags)
  * crypto_free_*(), as well as the various helpers below.
  */
 
+struct blkcipher_tfm {
+	void *iv;
+	int (*setkey)(struct crypto_tfm *tfm, const u8 *key,
+		      unsigned int keylen);
+	int (*encrypt)(struct blkcipher_desc *desc, struct scatterlist *dst,
+		       struct scatterlist *src, unsigned int nbytes);
+	int (*decrypt)(struct blkcipher_desc *desc, struct scatterlist *dst,
+		       struct scatterlist *src, unsigned int nbytes);
+};
+
 struct cipher_tfm {
 	void *cit_iv;
 	unsigned int cit_ivsize;
@@ -251,6 +286,7 @@ struct compress_tfm {
 	                      u8 *dst, unsigned int *dlen);
 };
 
+#define crt_blkcipher	crt_u.blkcipher
 #define crt_cipher	crt_u.cipher
 #define crt_digest	crt_u.digest
 #define crt_compress	crt_u.compress
@@ -260,6 +296,7 @@ struct crypto_tfm {
 	u32 crt_flags;
 	
 	union {
+		struct blkcipher_tfm blkcipher;
 		struct cipher_tfm cipher;
 		struct digest_tfm digest;
 		struct compress_tfm compress;
@@ -272,6 +309,10 @@ struct crypto_tfm {
 
 #define crypto_cipher crypto_tfm
 
+struct crypto_blkcipher {
+	struct crypto_tfm base;
+};
+
 enum {
 	CRYPTOA_UNSPEC,
 	CRYPTOA_ALG,
@@ -380,6 +421,144 @@ static inline unsigned int crypto_tfm_ctx_alignment(void)
 /*
  * API wrappers.
  */
+static inline struct crypto_blkcipher *__crypto_blkcipher_cast(
+	struct crypto_tfm *tfm)
+{
+	return (struct crypto_blkcipher *)tfm;
+}
+
+static inline struct crypto_blkcipher *crypto_blkcipher_cast(
+	struct crypto_tfm *tfm)
+{
+	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_BLKCIPHER);
+	return __crypto_blkcipher_cast(tfm);
+}
+
+static inline struct crypto_blkcipher *crypto_alloc_blkcipher(
+	const char *alg_name, u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	return __crypto_blkcipher_cast(crypto_alloc_base(alg_name, type, mask));
+}
+
+static inline struct crypto_tfm *crypto_blkcipher_tfm(
+	struct crypto_blkcipher *tfm)
+{
+	return &tfm->base;
+}
+
+static inline void crypto_free_blkcipher(struct crypto_blkcipher *tfm)
+{
+	crypto_free_tfm(crypto_blkcipher_tfm(tfm));
+}
+
+static inline const char *crypto_blkcipher_name(struct crypto_blkcipher *tfm)
+{
+	return crypto_tfm_alg_name(crypto_blkcipher_tfm(tfm));
+}
+
+static inline struct blkcipher_tfm *crypto_blkcipher_crt(
+	struct crypto_blkcipher *tfm)
+{
+	return &crypto_blkcipher_tfm(tfm)->crt_blkcipher;
+}
+
+static inline struct blkcipher_alg *crypto_blkcipher_alg(
+	struct crypto_blkcipher *tfm)
+{
+	return &crypto_blkcipher_tfm(tfm)->__crt_alg->cra_blkcipher;
+}
+
+static inline unsigned int crypto_blkcipher_ivsize(struct crypto_blkcipher *tfm)
+{
+	return crypto_blkcipher_alg(tfm)->ivsize;
+}
+
+static inline unsigned int crypto_blkcipher_blocksize(
+	struct crypto_blkcipher *tfm)
+{
+	return crypto_tfm_alg_blocksize(crypto_blkcipher_tfm(tfm));
+}
+
+static inline unsigned int crypto_blkcipher_alignmask(
+	struct crypto_blkcipher *tfm)
+{
+	return crypto_tfm_alg_alignmask(crypto_blkcipher_tfm(tfm));
+}
+
+static inline u32 crypto_blkcipher_get_flags(struct crypto_blkcipher *tfm)
+{
+	return crypto_tfm_get_flags(crypto_blkcipher_tfm(tfm));
+}
+
+static inline void crypto_blkcipher_set_flags(struct crypto_blkcipher *tfm,
+					      u32 flags)
+{
+	crypto_tfm_set_flags(crypto_blkcipher_tfm(tfm), flags);
+}
+
+static inline void crypto_blkcipher_clear_flags(struct crypto_blkcipher *tfm,
+						u32 flags)
+{
+	crypto_tfm_clear_flags(crypto_blkcipher_tfm(tfm), flags);
+}
+
+static inline int crypto_blkcipher_setkey(struct crypto_blkcipher *tfm,
+					  const u8 *key, unsigned int keylen)
+{
+	return crypto_blkcipher_crt(tfm)->setkey(crypto_blkcipher_tfm(tfm),
+						 key, keylen);
+}
+
+static inline int crypto_blkcipher_encrypt(struct blkcipher_desc *desc,
+					   struct scatterlist *dst,
+					   struct scatterlist *src,
+					   unsigned int nbytes)
+{
+	desc->info = crypto_blkcipher_crt(desc->tfm)->iv;
+	return crypto_blkcipher_crt(desc->tfm)->encrypt(desc, dst, src, nbytes);
+}
+
+static inline int crypto_blkcipher_encrypt_iv(struct blkcipher_desc *desc,
+					      struct scatterlist *dst,
+					      struct scatterlist *src,
+					      unsigned int nbytes)
+{
+	return crypto_blkcipher_crt(desc->tfm)->encrypt(desc, dst, src, nbytes);
+}
+
+static inline int crypto_blkcipher_decrypt(struct blkcipher_desc *desc,
+					   struct scatterlist *dst,
+					   struct scatterlist *src,
+					   unsigned int nbytes)
+{
+	desc->info = crypto_blkcipher_crt(desc->tfm)->iv;
+	return crypto_blkcipher_crt(desc->tfm)->decrypt(desc, dst, src, nbytes);
+}
+
+static inline int crypto_blkcipher_decrypt_iv(struct blkcipher_desc *desc,
+					      struct scatterlist *dst,
+					      struct scatterlist *src,
+					      unsigned int nbytes)
+{
+	return crypto_blkcipher_crt(desc->tfm)->decrypt(desc, dst, src, nbytes);
+}
+
+static inline void crypto_blkcipher_set_iv(struct crypto_blkcipher *tfm,
+					   const u8 *src, unsigned int len)
+{
+	memcpy(crypto_blkcipher_crt(tfm)->iv, src, len);
+}
+
+static inline void crypto_blkcipher_get_iv(struct crypto_blkcipher *tfm,
+					   u8 *dst, unsigned int len)
+{
+	memcpy(dst, crypto_blkcipher_crt(tfm)->iv, len);
+}
+
 static inline struct crypto_cipher *__crypto_cipher_cast(struct crypto_tfm *tfm)
 {
 	return (struct crypto_cipher *)tfm;
-- 
cgit v1.2.3


From db131ef9084110d9e82549c0a627e157e8bb99d7 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 21 Sep 2006 11:44:08 +1000
Subject: [CRYPTO] cipher: Added block ciphers for CBC/ECB

This patch adds two block cipher algorithms, CBC and ECB.  These
are implemented as templates on top of existing single-block cipher
algorithms.  They invoke the single-block cipher through the new
encrypt_one/decrypt_one interface.

This also optimises the in-place encryption and decryption to remove
the cost of an IV copy each round.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Kconfig          |  17 +++
 crypto/Makefile         |   2 +
 crypto/cbc.c            | 344 ++++++++++++++++++++++++++++++++++++++++++++++++
 crypto/ecb.c            | 181 +++++++++++++++++++++++++
 crypto/internal.h       |   1 -
 include/crypto/algapi.h |   2 +
 6 files changed, 546 insertions(+), 1 deletion(-)
 create mode 100644 crypto/cbc.c
 create mode 100644 crypto/ecb.c

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index 68790ad7308d..90d467c99c2c 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -123,6 +123,23 @@ config CRYPTO_TGR192
 	  See also:
 	  <http://www.cs.technion.ac.il/~biham/Reports/Tiger/>.
 
+config CRYPTO_ECB
+	tristate "ECB support"
+	select CRYPTO_BLKCIPHER
+	default m
+	help
+	  ECB: Electronic CodeBook mode
+	  This is the simplest block cipher algorithm.  It simply encrypts
+	  the input block by block.
+
+config CRYPTO_CBC
+	tristate "CBC support"
+	select CRYPTO_BLKCIPHER
+	default m
+	help
+	  CBC: Cipher Block Chaining mode
+	  This block cipher algorithm is required for IPSec.
+
 config CRYPTO_DES
 	tristate "DES and Triple DES EDE cipher algorithms"
 	select CRYPTO_ALGAPI
diff --git a/crypto/Makefile b/crypto/Makefile
index b5051951c636..5e1ff4e0b1fc 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -20,6 +20,8 @@ obj-$(CONFIG_CRYPTO_SHA256) += sha256.o
 obj-$(CONFIG_CRYPTO_SHA512) += sha512.o
 obj-$(CONFIG_CRYPTO_WP512) += wp512.o
 obj-$(CONFIG_CRYPTO_TGR192) += tgr192.o
+obj-$(CONFIG_CRYPTO_ECB) += ecb.o
+obj-$(CONFIG_CRYPTO_CBC) += cbc.o
 obj-$(CONFIG_CRYPTO_DES) += des.o
 obj-$(CONFIG_CRYPTO_BLOWFISH) += blowfish.o
 obj-$(CONFIG_CRYPTO_TWOFISH) += twofish.o
diff --git a/crypto/cbc.c b/crypto/cbc.c
new file mode 100644
index 000000000000..f5542b4db387
--- /dev/null
+++ b/crypto/cbc.c
@@ -0,0 +1,344 @@
+/*
+ * CBC: Cipher Block Chaining mode
+ *
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <crypto/algapi.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/scatterlist.h>
+#include <linux/slab.h>
+
+struct crypto_cbc_ctx {
+	struct crypto_cipher *child;
+	void (*xor)(u8 *dst, const u8 *src, unsigned int bs);
+};
+
+static int crypto_cbc_setkey(struct crypto_tfm *parent, const u8 *key,
+			     unsigned int keylen)
+{
+	struct crypto_cbc_ctx *ctx = crypto_tfm_ctx(parent);
+	struct crypto_cipher *child = ctx->child;
+	int err;
+
+	crypto_cipher_clear_flags(child, CRYPTO_TFM_REQ_MASK);
+	crypto_cipher_set_flags(child, crypto_tfm_get_flags(parent) &
+				       CRYPTO_TFM_REQ_MASK);
+	err = crypto_cipher_setkey(child, key, keylen);
+	crypto_tfm_set_flags(parent, crypto_cipher_get_flags(child) &
+				     CRYPTO_TFM_RES_MASK);
+	return err;
+}
+
+static int crypto_cbc_encrypt_segment(struct blkcipher_desc *desc,
+				      struct blkcipher_walk *walk,
+				      struct crypto_cipher *tfm,
+				      void (*xor)(u8 *, const u8 *,
+						  unsigned int))
+{
+	void (*fn)(struct crypto_tfm *, u8 *, const u8 *) =
+		crypto_cipher_alg(tfm)->cia_encrypt;
+	int bsize = crypto_cipher_blocksize(tfm);
+	unsigned int nbytes = walk->nbytes;
+	u8 *src = walk->src.virt.addr;
+	u8 *dst = walk->dst.virt.addr;
+	u8 *iv = walk->iv;
+
+	do {
+		xor(iv, src, bsize);
+		fn(crypto_cipher_tfm(tfm), dst, iv);
+		memcpy(iv, dst, bsize);
+
+		src += bsize;
+		dst += bsize;
+	} while ((nbytes -= bsize) >= bsize);
+
+	return nbytes;
+}
+
+static int crypto_cbc_encrypt_inplace(struct blkcipher_desc *desc,
+				      struct blkcipher_walk *walk,
+				      struct crypto_cipher *tfm,
+				      void (*xor)(u8 *, const u8 *,
+						  unsigned int))
+{
+	void (*fn)(struct crypto_tfm *, u8 *, const u8 *) =
+		crypto_cipher_alg(tfm)->cia_encrypt;
+	int bsize = crypto_cipher_blocksize(tfm);
+	unsigned int nbytes = walk->nbytes;
+	u8 *src = walk->src.virt.addr;
+	u8 *iv = walk->iv;
+
+	do {
+		xor(src, iv, bsize);
+		fn(crypto_cipher_tfm(tfm), src, src);
+		iv = src;
+
+		src += bsize;
+	} while ((nbytes -= bsize) >= bsize);
+
+	memcpy(walk->iv, iv, bsize);
+
+	return nbytes;
+}
+
+static int crypto_cbc_encrypt(struct blkcipher_desc *desc,
+			      struct scatterlist *dst, struct scatterlist *src,
+			      unsigned int nbytes)
+{
+	struct blkcipher_walk walk;
+	struct crypto_blkcipher *tfm = desc->tfm;
+	struct crypto_cbc_ctx *ctx = crypto_blkcipher_ctx(tfm);
+	struct crypto_cipher *child = ctx->child;
+	void (*xor)(u8 *, const u8 *, unsigned int bs) = ctx->xor;
+	int err;
+
+	blkcipher_walk_init(&walk, dst, src, nbytes);
+	err = blkcipher_walk_virt(desc, &walk);
+
+	while ((nbytes = walk.nbytes)) {
+		if (walk.src.virt.addr == walk.dst.virt.addr)
+			nbytes = crypto_cbc_encrypt_inplace(desc, &walk, child,
+							    xor);
+		else
+			nbytes = crypto_cbc_encrypt_segment(desc, &walk, child,
+							    xor);
+		err = blkcipher_walk_done(desc, &walk, nbytes);
+	}
+
+	return err;
+}
+
+static int crypto_cbc_decrypt_segment(struct blkcipher_desc *desc,
+				      struct blkcipher_walk *walk,
+				      struct crypto_cipher *tfm,
+				      void (*xor)(u8 *, const u8 *,
+						  unsigned int))
+{
+	void (*fn)(struct crypto_tfm *, u8 *, const u8 *) =
+		crypto_cipher_alg(tfm)->cia_decrypt;
+	int bsize = crypto_cipher_blocksize(tfm);
+	unsigned int nbytes = walk->nbytes;
+	u8 *src = walk->src.virt.addr;
+	u8 *dst = walk->dst.virt.addr;
+	u8 *iv = walk->iv;
+
+	do {
+		fn(crypto_cipher_tfm(tfm), dst, src);
+		xor(dst, iv, bsize);
+		iv = src;
+
+		src += bsize;
+		dst += bsize;
+	} while ((nbytes -= bsize) >= bsize);
+
+	memcpy(walk->iv, iv, bsize);
+
+	return nbytes;
+}
+
+static int crypto_cbc_decrypt_inplace(struct blkcipher_desc *desc,
+				      struct blkcipher_walk *walk,
+				      struct crypto_cipher *tfm,
+				      void (*xor)(u8 *, const u8 *,
+						  unsigned int))
+{
+	void (*fn)(struct crypto_tfm *, u8 *, const u8 *) =
+		crypto_cipher_alg(tfm)->cia_decrypt;
+	int bsize = crypto_cipher_blocksize(tfm);
+	unsigned long alignmask = crypto_cipher_alignmask(tfm);
+	unsigned int nbytes = walk->nbytes;
+	u8 *src = walk->src.virt.addr;
+	u8 stack[bsize + alignmask];
+	u8 *first_iv = (u8 *)ALIGN((unsigned long)stack, alignmask + 1);
+
+	memcpy(first_iv, walk->iv, bsize);
+
+	/* Start of the last block. */
+	src += nbytes - nbytes % bsize - bsize;
+	memcpy(walk->iv, src, bsize);
+
+	for (;;) {
+		fn(crypto_cipher_tfm(tfm), src, src);
+		if ((nbytes -= bsize) < bsize)
+			break;
+		xor(src, src - bsize, bsize);
+		src -= bsize;
+	}
+
+	xor(src, first_iv, bsize);
+
+	return nbytes;
+}
+
+static int crypto_cbc_decrypt(struct blkcipher_desc *desc,
+			      struct scatterlist *dst, struct scatterlist *src,
+			      unsigned int nbytes)
+{
+	struct blkcipher_walk walk;
+	struct crypto_blkcipher *tfm = desc->tfm;
+	struct crypto_cbc_ctx *ctx = crypto_blkcipher_ctx(tfm);
+	struct crypto_cipher *child = ctx->child;
+	void (*xor)(u8 *, const u8 *, unsigned int bs) = ctx->xor;
+	int err;
+
+	blkcipher_walk_init(&walk, dst, src, nbytes);
+	err = blkcipher_walk_virt(desc, &walk);
+
+	while ((nbytes = walk.nbytes)) {
+		if (walk.src.virt.addr == walk.dst.virt.addr)
+			nbytes = crypto_cbc_decrypt_inplace(desc, &walk, child,
+							    xor);
+		else
+			nbytes = crypto_cbc_decrypt_segment(desc, &walk, child,
+							    xor);
+		err = blkcipher_walk_done(desc, &walk, nbytes);
+	}
+
+	return err;
+}
+
+static void xor_byte(u8 *a, const u8 *b, unsigned int bs)
+{
+	do {
+		*a++ ^= *b++;
+	} while (--bs);
+}
+
+static void xor_quad(u8 *dst, const u8 *src, unsigned int bs)
+{
+	u32 *a = (u32 *)dst;
+	u32 *b = (u32 *)src;
+
+	do {
+		*a++ ^= *b++;
+	} while ((bs -= 4));
+}
+
+static void xor_64(u8 *a, const u8 *b, unsigned int bs)
+{
+	((u32 *)a)[0] ^= ((u32 *)b)[0];
+	((u32 *)a)[1] ^= ((u32 *)b)[1];
+}
+
+static void xor_128(u8 *a, const u8 *b, unsigned int bs)
+{
+	((u32 *)a)[0] ^= ((u32 *)b)[0];
+	((u32 *)a)[1] ^= ((u32 *)b)[1];
+	((u32 *)a)[2] ^= ((u32 *)b)[2];
+	((u32 *)a)[3] ^= ((u32 *)b)[3];
+}
+
+static int crypto_cbc_init_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_instance *inst = (void *)tfm->__crt_alg;
+	struct crypto_spawn *spawn = crypto_instance_ctx(inst);
+	struct crypto_cbc_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	switch (crypto_tfm_alg_blocksize(tfm)) {
+	case 8:
+		ctx->xor = xor_64;
+		break;
+
+	case 16:
+		ctx->xor = xor_128;
+		break;
+
+	default:
+		if (crypto_tfm_alg_blocksize(tfm) % 4)
+			ctx->xor = xor_byte;
+		else
+			ctx->xor = xor_quad;
+	}
+
+	tfm = crypto_spawn_tfm(spawn);
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	ctx->child = crypto_cipher_cast(tfm);
+	return 0;
+}
+
+static void crypto_cbc_exit_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_cbc_ctx *ctx = crypto_tfm_ctx(tfm);
+	crypto_free_cipher(ctx->child);
+}
+
+static struct crypto_instance *crypto_cbc_alloc(void *param, unsigned int len)
+{
+	struct crypto_instance *inst;
+	struct crypto_alg *alg;
+
+	alg = crypto_get_attr_alg(param, len, CRYPTO_ALG_TYPE_CIPHER,
+				  CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
+	if (IS_ERR(alg))
+		return ERR_PTR(PTR_ERR(alg));
+
+	inst = crypto_alloc_instance("cbc", alg);
+	if (IS_ERR(inst))
+		goto out_put_alg;
+
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER;
+	inst->alg.cra_priority = alg->cra_priority;
+	inst->alg.cra_blocksize = alg->cra_blocksize;
+	inst->alg.cra_alignmask = alg->cra_alignmask;
+	inst->alg.cra_type = &crypto_blkcipher_type;
+
+	if (!(alg->cra_blocksize % 4))
+		inst->alg.cra_alignmask |= 3;
+	inst->alg.cra_blkcipher.ivsize = alg->cra_blocksize;
+	inst->alg.cra_blkcipher.min_keysize = alg->cra_cipher.cia_min_keysize;
+	inst->alg.cra_blkcipher.max_keysize = alg->cra_cipher.cia_max_keysize;
+
+	inst->alg.cra_ctxsize = sizeof(struct crypto_cbc_ctx);
+
+	inst->alg.cra_init = crypto_cbc_init_tfm;
+	inst->alg.cra_exit = crypto_cbc_exit_tfm;
+
+	inst->alg.cra_blkcipher.setkey = crypto_cbc_setkey;
+	inst->alg.cra_blkcipher.encrypt = crypto_cbc_encrypt;
+	inst->alg.cra_blkcipher.decrypt = crypto_cbc_decrypt;
+
+out_put_alg:
+	crypto_mod_put(alg);
+	return inst;
+}
+
+static void crypto_cbc_free(struct crypto_instance *inst)
+{
+	crypto_drop_spawn(crypto_instance_ctx(inst));
+	kfree(inst);
+}
+
+static struct crypto_template crypto_cbc_tmpl = {
+	.name = "cbc",
+	.alloc = crypto_cbc_alloc,
+	.free = crypto_cbc_free,
+	.module = THIS_MODULE,
+};
+
+static int __init crypto_cbc_module_init(void)
+{
+	return crypto_register_template(&crypto_cbc_tmpl);
+}
+
+static void __exit crypto_cbc_module_exit(void)
+{
+	crypto_unregister_template(&crypto_cbc_tmpl);
+}
+
+module_init(crypto_cbc_module_init);
+module_exit(crypto_cbc_module_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("CBC block cipher algorithm");
diff --git a/crypto/ecb.c b/crypto/ecb.c
new file mode 100644
index 000000000000..f239aa9c4017
--- /dev/null
+++ b/crypto/ecb.c
@@ -0,0 +1,181 @@
+/*
+ * ECB: Electronic CodeBook mode
+ *
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <crypto/algapi.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/scatterlist.h>
+#include <linux/slab.h>
+
+struct crypto_ecb_ctx {
+	struct crypto_cipher *child;
+};
+
+static int crypto_ecb_setkey(struct crypto_tfm *parent, const u8 *key,
+			     unsigned int keylen)
+{
+	struct crypto_ecb_ctx *ctx = crypto_tfm_ctx(parent);
+	struct crypto_cipher *child = ctx->child;
+	int err;
+
+	crypto_cipher_clear_flags(child, CRYPTO_TFM_REQ_MASK);
+	crypto_cipher_set_flags(child, crypto_tfm_get_flags(parent) &
+				       CRYPTO_TFM_REQ_MASK);
+	err = crypto_cipher_setkey(child, key, keylen);
+	crypto_tfm_set_flags(parent, crypto_cipher_get_flags(child) &
+				     CRYPTO_TFM_RES_MASK);
+	return err;
+}
+
+static int crypto_ecb_crypt(struct blkcipher_desc *desc,
+			    struct blkcipher_walk *walk,
+			    struct crypto_cipher *tfm,
+			    void (*fn)(struct crypto_tfm *, u8 *, const u8 *))
+{
+	int bsize = crypto_cipher_blocksize(tfm);
+	unsigned int nbytes;
+	int err;
+
+	err = blkcipher_walk_virt(desc, walk);
+
+	while ((nbytes = walk->nbytes)) {
+		u8 *wsrc = walk->src.virt.addr;
+		u8 *wdst = walk->dst.virt.addr;
+
+		do {
+			fn(crypto_cipher_tfm(tfm), wdst, wsrc);
+	
+			wsrc += bsize;
+			wdst += bsize;
+		} while ((nbytes -= bsize) >= bsize);
+
+		err = blkcipher_walk_done(desc, walk, nbytes);
+	}
+
+	return err;
+}
+
+static int crypto_ecb_encrypt(struct blkcipher_desc *desc,
+			      struct scatterlist *dst, struct scatterlist *src,
+			      unsigned int nbytes)
+{
+	struct blkcipher_walk walk;
+	struct crypto_blkcipher *tfm = desc->tfm;
+	struct crypto_ecb_ctx *ctx = crypto_blkcipher_ctx(tfm);
+	struct crypto_cipher *child = ctx->child;
+
+	blkcipher_walk_init(&walk, dst, src, nbytes);
+	return crypto_ecb_crypt(desc, &walk, child,
+				crypto_cipher_alg(child)->cia_encrypt);
+}
+
+static int crypto_ecb_decrypt(struct blkcipher_desc *desc,
+			      struct scatterlist *dst, struct scatterlist *src,
+			      unsigned int nbytes)
+{
+	struct blkcipher_walk walk;
+	struct crypto_blkcipher *tfm = desc->tfm;
+	struct crypto_ecb_ctx *ctx = crypto_blkcipher_ctx(tfm);
+	struct crypto_cipher *child = ctx->child;
+
+	blkcipher_walk_init(&walk, dst, src, nbytes);
+	return crypto_ecb_crypt(desc, &walk, child,
+				crypto_cipher_alg(child)->cia_decrypt);
+}
+
+static int crypto_ecb_init_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_instance *inst = (void *)tfm->__crt_alg;
+	struct crypto_spawn *spawn = crypto_instance_ctx(inst);
+	struct crypto_ecb_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	tfm = crypto_spawn_tfm(spawn);
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	ctx->child = crypto_cipher_cast(tfm);
+	return 0;
+}
+
+static void crypto_ecb_exit_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_ecb_ctx *ctx = crypto_tfm_ctx(tfm);
+	crypto_free_cipher(ctx->child);
+}
+
+static struct crypto_instance *crypto_ecb_alloc(void *param, unsigned int len)
+{
+	struct crypto_instance *inst;
+	struct crypto_alg *alg;
+
+	alg = crypto_get_attr_alg(param, len, CRYPTO_ALG_TYPE_CIPHER,
+				  CRYPTO_ALG_TYPE_MASK | CRYPTO_ALG_ASYNC);
+	if (IS_ERR(alg))
+		return ERR_PTR(PTR_ERR(alg));
+
+	inst = crypto_alloc_instance("ecb", alg);
+	if (IS_ERR(inst))
+		goto out_put_alg;
+
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER;
+	inst->alg.cra_priority = alg->cra_priority;
+	inst->alg.cra_blocksize = alg->cra_blocksize;
+	inst->alg.cra_alignmask = alg->cra_alignmask;
+	inst->alg.cra_type = &crypto_blkcipher_type;
+
+	inst->alg.cra_blkcipher.min_keysize = alg->cra_cipher.cia_min_keysize;
+	inst->alg.cra_blkcipher.max_keysize = alg->cra_cipher.cia_max_keysize;
+
+	inst->alg.cra_ctxsize = sizeof(struct crypto_ecb_ctx);
+
+	inst->alg.cra_init = crypto_ecb_init_tfm;
+	inst->alg.cra_exit = crypto_ecb_exit_tfm;
+
+	inst->alg.cra_blkcipher.setkey = crypto_ecb_setkey;
+	inst->alg.cra_blkcipher.encrypt = crypto_ecb_encrypt;
+	inst->alg.cra_blkcipher.decrypt = crypto_ecb_decrypt;
+
+out_put_alg:
+	crypto_mod_put(alg);
+	return inst;
+}
+
+static void crypto_ecb_free(struct crypto_instance *inst)
+{
+	crypto_drop_spawn(crypto_instance_ctx(inst));
+	kfree(inst);
+}
+
+static struct crypto_template crypto_ecb_tmpl = {
+	.name = "ecb",
+	.alloc = crypto_ecb_alloc,
+	.free = crypto_ecb_free,
+	.module = THIS_MODULE,
+};
+
+static int __init crypto_ecb_module_init(void)
+{
+	return crypto_register_template(&crypto_ecb_tmpl);
+}
+
+static void __exit crypto_ecb_module_exit(void)
+{
+	crypto_unregister_template(&crypto_ecb_tmpl);
+}
+
+module_init(crypto_ecb_module_init);
+module_exit(crypto_ecb_module_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("ECB block cipher algorithm");
diff --git a/crypto/internal.h b/crypto/internal.h
index 7dc04efb55c6..93d9b10ff914 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -131,7 +131,6 @@ static inline unsigned int crypto_compress_ctxsize(struct crypto_alg *alg,
 }
 
 struct crypto_alg *crypto_mod_get(struct crypto_alg *alg);
-void crypto_mod_put(struct crypto_alg *alg);
 struct crypto_alg *__crypto_alg_lookup(const char *name, u32 type, u32 mask);
 struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask);
 
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index f3946baf0c07..444f602724db 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -83,6 +83,8 @@ struct blkcipher_walk {
 
 extern const struct crypto_type crypto_blkcipher_type;
 
+void crypto_mod_put(struct crypto_alg *alg);
+
 int crypto_register_template(struct crypto_template *tmpl);
 void crypto_unregister_template(struct crypto_template *tmpl);
 struct crypto_template *crypto_lookup_template(const char *name);
-- 
cgit v1.2.3


From 04ff12609445c7b462d7fc7f2d30dad442c922f3 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 13 Aug 2006 08:50:00 +1000
Subject: [IPSEC]: Add compatibility algorithm name support

This patch adds a compatibility name field for each IPsec algorithm.  This
is needed when parameterised algorithms are used.  For example, "md5" will
become "hmac(md5)", and "aes" will become "cbc(aes)".

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/net/xfrm.h   | 1 +
 net/xfrm/xfrm_algo.c | 3 ++-
 net/xfrm/xfrm_user.c | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 10396b4bde14..e9114e41affc 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -854,6 +854,7 @@ struct xfrm_algo_comp_info {
 
 struct xfrm_algo_desc {
 	char *name;
+	char *compat;
 	u8 available:1;
 	union {
 		struct xfrm_algo_auth_info auth;
diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index 04e1aea58bc9..b68974b38741 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -359,7 +359,8 @@ static struct xfrm_algo_desc *xfrm_get_byname(struct xfrm_algo_desc *list,
 		return NULL;
 
 	for (i = 0; i < entries; i++) {
-		if (strcmp(name, list[i].name))
+		if (strcmp(name, list[i].name) &&
+		    (!list[i].compat || strcmp(name, list[i].compat)))
 			continue;
 
 		if (list[i].available)
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 7d18ca03c80d..fa79ddc4239e 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -213,6 +213,7 @@ static int attach_one_algo(struct xfrm_algo **algpp, u8 *props,
 		return -ENOMEM;
 
 	memcpy(p, ualg, len);
+	strcpy(p->alg_name, algo->name);
 	*algpp = p;
 	return 0;
 }
-- 
cgit v1.2.3


From 6b7326c8497f954c2cfcb4c49fe42be5b80887bc Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 30 Jul 2006 15:41:01 +1000
Subject: [IPSEC] ESP: Use block ciphers where applicable

This patch converts IPSec/ESP to use the new block cipher type where
applicable.  Similar to the HMAC conversion, existing algorithm names
have been kept for compatibility.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/net/esp.h    |  2 +-
 net/ipv4/Kconfig     |  1 +
 net/ipv4/esp4.c      | 49 +++++++++++++++++++++++++++++--------------------
 net/ipv6/Kconfig     |  1 +
 net/ipv6/esp6.c      | 48 ++++++++++++++++++++++++++++--------------------
 net/xfrm/xfrm_algo.c | 24 ++++++++++++++++--------
 6 files changed, 76 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/net/esp.h b/include/net/esp.h
index 6eb837973c84..af2ff18700c7 100644
--- a/include/net/esp.h
+++ b/include/net/esp.h
@@ -22,7 +22,7 @@ struct esp_data
 		 * >= crypto_tfm_alg_ivsize(tfm). */
 		int			ivlen;
 		int			padlen;		/* 0..255 */
-		struct crypto_tfm	*tfm;		/* crypto handle */
+		struct crypto_blkcipher	*tfm;		/* crypto handle */
 	} conf;
 
 	/* Integrity. It is active when icv_full_len != 0 */
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 8514106761b0..3b5d504a74be 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -386,6 +386,7 @@ config INET_ESP
 	select CRYPTO
 	select CRYPTO_HMAC
 	select CRYPTO_MD5
+	select CRYPTO_CBC
 	select CRYPTO_SHA1
 	select CRYPTO_DES
 	---help---
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index fc2f8ce441de..7c63ae494742 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -1,3 +1,4 @@
+#include <linux/err.h>
 #include <linux/module.h>
 #include <net/ip.h>
 #include <net/xfrm.h>
@@ -16,7 +17,8 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 	int err;
 	struct iphdr *top_iph;
 	struct ip_esp_hdr *esph;
-	struct crypto_tfm *tfm;
+	struct crypto_blkcipher *tfm;
+	struct blkcipher_desc desc;
 	struct esp_data *esp;
 	struct sk_buff *trailer;
 	int blksize;
@@ -36,7 +38,9 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 	esp = x->data;
 	alen = esp->auth.icv_trunc_len;
 	tfm = esp->conf.tfm;
-	blksize = ALIGN(crypto_tfm_alg_blocksize(tfm), 4);
+	desc.tfm = tfm;
+	desc.flags = 0;
+	blksize = ALIGN(crypto_blkcipher_blocksize(tfm), 4);
 	clen = ALIGN(clen + 2, blksize);
 	if (esp->conf.padlen)
 		clen = ALIGN(clen, esp->conf.padlen);
@@ -92,7 +96,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 	xfrm_aevent_doreplay(x);
 
 	if (esp->conf.ivlen)
-		crypto_cipher_set_iv(tfm, esp->conf.ivec, crypto_tfm_alg_ivsize(tfm));
+		crypto_blkcipher_set_iv(tfm, esp->conf.ivec, esp->conf.ivlen);
 
 	do {
 		struct scatterlist *sg = &esp->sgbuf[0];
@@ -103,14 +107,17 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 				goto error;
 		}
 		skb_to_sgvec(skb, sg, esph->enc_data+esp->conf.ivlen-skb->data, clen);
-		crypto_cipher_encrypt(tfm, sg, sg, clen);
+		err = crypto_blkcipher_encrypt(&desc, sg, sg, clen);
 		if (unlikely(sg != &esp->sgbuf[0]))
 			kfree(sg);
 	} while (0);
 
+	if (unlikely(err))
+		goto error;
+
 	if (esp->conf.ivlen) {
-		memcpy(esph->enc_data, esp->conf.ivec, crypto_tfm_alg_ivsize(tfm));
-		crypto_cipher_get_iv(tfm, esp->conf.ivec, crypto_tfm_alg_ivsize(tfm));
+		memcpy(esph->enc_data, esp->conf.ivec, esp->conf.ivlen);
+		crypto_blkcipher_get_iv(tfm, esp->conf.ivec, esp->conf.ivlen);
 	}
 
 	if (esp->auth.icv_full_len) {
@@ -121,8 +128,6 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 
 	ip_send_check(top_iph);
 
-	err = 0;
-
 error:
 	return err;
 }
@@ -137,8 +142,10 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 	struct iphdr *iph;
 	struct ip_esp_hdr *esph;
 	struct esp_data *esp = x->data;
+	struct crypto_blkcipher *tfm = esp->conf.tfm;
+	struct blkcipher_desc desc = { .tfm = tfm };
 	struct sk_buff *trailer;
-	int blksize = ALIGN(crypto_tfm_alg_blocksize(esp->conf.tfm), 4);
+	int blksize = ALIGN(crypto_blkcipher_blocksize(tfm), 4);
 	int alen = esp->auth.icv_trunc_len;
 	int elen = skb->len - sizeof(struct ip_esp_hdr) - esp->conf.ivlen - alen;
 	int nfrags;
@@ -146,6 +153,7 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 	u8 nexthdr[2];
 	struct scatterlist *sg;
 	int padlen;
+	int err;
 
 	if (!pskb_may_pull(skb, sizeof(struct ip_esp_hdr)))
 		goto out;
@@ -178,7 +186,7 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 
 	/* Get ivec. This can be wrong, check against another impls. */
 	if (esp->conf.ivlen)
-		crypto_cipher_set_iv(esp->conf.tfm, esph->enc_data, crypto_tfm_alg_ivsize(esp->conf.tfm));
+		crypto_blkcipher_set_iv(tfm, esph->enc_data, esp->conf.ivlen);
 
 	sg = &esp->sgbuf[0];
 
@@ -188,9 +196,11 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 			goto out;
 	}
 	skb_to_sgvec(skb, sg, sizeof(struct ip_esp_hdr) + esp->conf.ivlen, elen);
-	crypto_cipher_decrypt(esp->conf.tfm, sg, sg, elen);
+	err = crypto_blkcipher_decrypt(&desc, sg, sg, elen);
 	if (unlikely(sg != &esp->sgbuf[0]))
 		kfree(sg);
+	if (unlikely(err))
+		return err;
 
 	if (skb_copy_bits(skb, skb->len-alen-2, nexthdr, 2))
 		BUG();
@@ -254,7 +264,7 @@ out:
 static u32 esp4_get_max_size(struct xfrm_state *x, int mtu)
 {
 	struct esp_data *esp = x->data;
-	u32 blksize = ALIGN(crypto_tfm_alg_blocksize(esp->conf.tfm), 4);
+	u32 blksize = ALIGN(crypto_blkcipher_blocksize(esp->conf.tfm), 4);
 
 	if (x->props.mode) {
 		mtu = ALIGN(mtu + 2, blksize);
@@ -293,7 +303,7 @@ static void esp_destroy(struct xfrm_state *x)
 	if (!esp)
 		return;
 
-	crypto_free_tfm(esp->conf.tfm);
+	crypto_free_blkcipher(esp->conf.tfm);
 	esp->conf.tfm = NULL;
 	kfree(esp->conf.ivec);
 	esp->conf.ivec = NULL;
@@ -307,6 +317,7 @@ static void esp_destroy(struct xfrm_state *x)
 static int esp_init_state(struct xfrm_state *x)
 {
 	struct esp_data *esp = NULL;
+	struct crypto_blkcipher *tfm;
 
 	/* null auth and encryption can have zero length keys */
 	if (x->aalg) {
@@ -351,13 +362,11 @@ static int esp_init_state(struct xfrm_state *x)
 	}
 	esp->conf.key = x->ealg->alg_key;
 	esp->conf.key_len = (x->ealg->alg_key_len+7)/8;
-	if (x->props.ealgo == SADB_EALG_NULL)
-		esp->conf.tfm = crypto_alloc_tfm(x->ealg->alg_name, CRYPTO_TFM_MODE_ECB);
-	else
-		esp->conf.tfm = crypto_alloc_tfm(x->ealg->alg_name, CRYPTO_TFM_MODE_CBC);
-	if (esp->conf.tfm == NULL)
+	tfm = crypto_alloc_blkcipher(x->ealg->alg_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(tfm))
 		goto error;
-	esp->conf.ivlen = crypto_tfm_alg_ivsize(esp->conf.tfm);
+	esp->conf.tfm = tfm;
+	esp->conf.ivlen = crypto_blkcipher_ivsize(tfm);
 	esp->conf.padlen = 0;
 	if (esp->conf.ivlen) {
 		esp->conf.ivec = kmalloc(esp->conf.ivlen, GFP_KERNEL);
@@ -365,7 +374,7 @@ static int esp_init_state(struct xfrm_state *x)
 			goto error;
 		get_random_bytes(esp->conf.ivec, esp->conf.ivlen);
 	}
-	if (crypto_cipher_setkey(esp->conf.tfm, esp->conf.key, esp->conf.key_len))
+	if (crypto_blkcipher_setkey(tfm, esp->conf.key, esp->conf.key_len))
 		goto error;
 	x->props.header_len = sizeof(struct ip_esp_hdr) + esp->conf.ivlen;
 	if (x->props.mode)
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index e923d4dea418..0ba06c0c5d39 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -77,6 +77,7 @@ config INET6_ESP
 	select CRYPTO
 	select CRYPTO_HMAC
 	select CRYPTO_MD5
+	select CRYPTO_CBC
 	select CRYPTO_SHA1
 	select CRYPTO_DES
 	---help---
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index a278d5e862fe..46a7e687948e 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -24,6 +24,7 @@
  * 	This file is derived from net/ipv4/esp.c
  */
 
+#include <linux/err.h>
 #include <linux/module.h>
 #include <net/ip.h>
 #include <net/xfrm.h>
@@ -44,7 +45,8 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 	int hdr_len;
 	struct ipv6hdr *top_iph;
 	struct ipv6_esp_hdr *esph;
-	struct crypto_tfm *tfm;
+	struct crypto_blkcipher *tfm;
+	struct blkcipher_desc desc;
 	struct esp_data *esp;
 	struct sk_buff *trailer;
 	int blksize;
@@ -67,7 +69,9 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 
 	alen = esp->auth.icv_trunc_len;
 	tfm = esp->conf.tfm;
-	blksize = ALIGN(crypto_tfm_alg_blocksize(tfm), 4);
+	desc.tfm = tfm;
+	desc.flags = 0;
+	blksize = ALIGN(crypto_blkcipher_blocksize(tfm), 4);
 	clen = ALIGN(clen + 2, blksize);
 	if (esp->conf.padlen)
 		clen = ALIGN(clen, esp->conf.padlen);
@@ -96,7 +100,7 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 	xfrm_aevent_doreplay(x);
 
 	if (esp->conf.ivlen)
-		crypto_cipher_set_iv(tfm, esp->conf.ivec, crypto_tfm_alg_ivsize(tfm));
+		crypto_blkcipher_set_iv(tfm, esp->conf.ivec, esp->conf.ivlen);
 
 	do {
 		struct scatterlist *sg = &esp->sgbuf[0];
@@ -107,14 +111,17 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 				goto error;
 		}
 		skb_to_sgvec(skb, sg, esph->enc_data+esp->conf.ivlen-skb->data, clen);
-		crypto_cipher_encrypt(tfm, sg, sg, clen);
+		err = crypto_blkcipher_encrypt(&desc, sg, sg, clen);
 		if (unlikely(sg != &esp->sgbuf[0]))
 			kfree(sg);
 	} while (0);
 
+	if (unlikely(err))
+		goto error;
+
 	if (esp->conf.ivlen) {
-		memcpy(esph->enc_data, esp->conf.ivec, crypto_tfm_alg_ivsize(tfm));
-		crypto_cipher_get_iv(tfm, esp->conf.ivec, crypto_tfm_alg_ivsize(tfm));
+		memcpy(esph->enc_data, esp->conf.ivec, esp->conf.ivlen);
+		crypto_blkcipher_get_iv(tfm, esp->conf.ivec, esp->conf.ivlen);
 	}
 
 	if (esp->auth.icv_full_len) {
@@ -123,8 +130,6 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 		pskb_put(skb, trailer, alen);
 	}
 
-	err = 0;
-
 error:
 	return err;
 }
@@ -134,8 +139,10 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb)
 	struct ipv6hdr *iph;
 	struct ipv6_esp_hdr *esph;
 	struct esp_data *esp = x->data;
+	struct crypto_blkcipher *tfm = esp->conf.tfm;
+	struct blkcipher_desc desc = { .tfm = tfm };
 	struct sk_buff *trailer;
-	int blksize = ALIGN(crypto_tfm_alg_blocksize(esp->conf.tfm), 4);
+	int blksize = ALIGN(crypto_blkcipher_blocksize(tfm), 4);
 	int alen = esp->auth.icv_trunc_len;
 	int elen = skb->len - sizeof(struct ipv6_esp_hdr) - esp->conf.ivlen - alen;
 
@@ -182,7 +189,7 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb)
 
 	/* Get ivec. This can be wrong, check against another impls. */
 	if (esp->conf.ivlen)
-		crypto_cipher_set_iv(esp->conf.tfm, esph->enc_data, crypto_tfm_alg_ivsize(esp->conf.tfm));
+		crypto_blkcipher_set_iv(tfm, esph->enc_data, esp->conf.ivlen);
 
         {
 		u8 nexthdr[2];
@@ -197,9 +204,11 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb)
 			}
 		}
 		skb_to_sgvec(skb, sg, sizeof(struct ipv6_esp_hdr) + esp->conf.ivlen, elen);
-		crypto_cipher_decrypt(esp->conf.tfm, sg, sg, elen);
+		ret = crypto_blkcipher_decrypt(&desc, sg, sg, elen);
 		if (unlikely(sg != &esp->sgbuf[0]))
 			kfree(sg);
+		if (unlikely(ret))
+			goto out;
 
 		if (skb_copy_bits(skb, skb->len-alen-2, nexthdr, 2))
 			BUG();
@@ -225,7 +234,7 @@ out:
 static u32 esp6_get_max_size(struct xfrm_state *x, int mtu)
 {
 	struct esp_data *esp = x->data;
-	u32 blksize = ALIGN(crypto_tfm_alg_blocksize(esp->conf.tfm), 4);
+	u32 blksize = ALIGN(crypto_blkcipher_blocksize(esp->conf.tfm), 4);
 
 	if (x->props.mode) {
 		mtu = ALIGN(mtu + 2, blksize);
@@ -266,7 +275,7 @@ static void esp6_destroy(struct xfrm_state *x)
 	if (!esp)
 		return;
 
-	crypto_free_tfm(esp->conf.tfm);
+	crypto_free_blkcipher(esp->conf.tfm);
 	esp->conf.tfm = NULL;
 	kfree(esp->conf.ivec);
 	esp->conf.ivec = NULL;
@@ -280,6 +289,7 @@ static void esp6_destroy(struct xfrm_state *x)
 static int esp6_init_state(struct xfrm_state *x)
 {
 	struct esp_data *esp = NULL;
+	struct crypto_blkcipher *tfm;
 
 	/* null auth and encryption can have zero length keys */
 	if (x->aalg) {
@@ -327,13 +337,11 @@ static int esp6_init_state(struct xfrm_state *x)
 	}
 	esp->conf.key = x->ealg->alg_key;
 	esp->conf.key_len = (x->ealg->alg_key_len+7)/8;
-	if (x->props.ealgo == SADB_EALG_NULL)
-		esp->conf.tfm = crypto_alloc_tfm(x->ealg->alg_name, CRYPTO_TFM_MODE_ECB);
-	else
-		esp->conf.tfm = crypto_alloc_tfm(x->ealg->alg_name, CRYPTO_TFM_MODE_CBC);
-	if (esp->conf.tfm == NULL)
+	tfm = crypto_alloc_blkcipher(x->ealg->alg_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(tfm))
 		goto error;
-	esp->conf.ivlen = crypto_tfm_alg_ivsize(esp->conf.tfm);
+	esp->conf.tfm = tfm;
+	esp->conf.ivlen = crypto_blkcipher_ivsize(tfm);
 	esp->conf.padlen = 0;
 	if (esp->conf.ivlen) {
 		esp->conf.ivec = kmalloc(esp->conf.ivlen, GFP_KERNEL);
@@ -341,7 +349,7 @@ static int esp6_init_state(struct xfrm_state *x)
 			goto error;
 		get_random_bytes(esp->conf.ivec, esp->conf.ivlen);
 	}
-	if (crypto_cipher_setkey(esp->conf.tfm, esp->conf.key, esp->conf.key_len))
+	if (crypto_blkcipher_setkey(tfm, esp->conf.key, esp->conf.key_len))
 		goto error;
 	x->props.header_len = sizeof(struct ipv6_esp_hdr) + esp->conf.ivlen;
 	if (x->props.mode)
diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index b68974b38741..9b03d8497fba 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -118,7 +118,8 @@ static struct xfrm_algo_desc aalg_list[] = {
 
 static struct xfrm_algo_desc ealg_list[] = {
 {
-	.name = "cipher_null",
+	.name = "ecb(cipher_null)",
+	.compat = "cipher_null",
 	
 	.uinfo = {
 		.encr = {
@@ -135,7 +136,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-	.name = "des",
+	.name = "cbc(des)",
+	.compat = "des",
 
 	.uinfo = {
 		.encr = {
@@ -152,7 +154,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-	.name = "des3_ede",
+	.name = "cbc(des3_ede)",
+	.compat = "des3_ede",
 
 	.uinfo = {
 		.encr = {
@@ -169,7 +172,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-	.name = "cast128",
+	.name = "cbc(cast128)",
+	.compat = "cast128",
 
 	.uinfo = {
 		.encr = {
@@ -186,7 +190,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-	.name = "blowfish",
+	.name = "cbc(blowfish)",
+	.compat = "blowfish",
 
 	.uinfo = {
 		.encr = {
@@ -203,7 +208,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-	.name = "aes",
+	.name = "cbc(aes)",
+	.compat = "aes",
 
 	.uinfo = {
 		.encr = {
@@ -220,7 +226,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-        .name = "serpent",
+        .name = "cbc(serpent)",
+        .compat = "serpent",
 
         .uinfo = {
                 .encr = {
@@ -237,7 +244,8 @@ static struct xfrm_algo_desc ealg_list[] = {
         }
 },
 {
-        .name = "twofish",
+        .name = "cbc(twofish)",
+        .compat = "twofish",
                  
         .uinfo = {
                 .encr = {
-- 
cgit v1.2.3


From 378c6697a282c383d89428380a3405bf95189347 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 22 Aug 2006 20:33:54 +1000
Subject: [SUNRPC] GSS: Use block ciphers where applicable

This patch converts SUNRPC/GSS to use the new block cipher type where
applicable.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/sunrpc/gss_krb5.h       | 19 ++++++------
 include/linux/sunrpc/gss_spkm3.h      |  4 +--
 net/sunrpc/auth_gss/gss_krb5_crypto.c | 57 ++++++++++++++++++++---------------
 net/sunrpc/auth_gss/gss_krb5_mech.c   | 24 ++++++++-------
 net/sunrpc/auth_gss/gss_krb5_seqnum.c |  4 +--
 net/sunrpc/auth_gss/gss_krb5_wrap.c   |  4 +--
 net/sunrpc/auth_gss/gss_spkm3_mech.c  | 29 +++++++++---------
 7 files changed, 76 insertions(+), 65 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h
index 1279280d7196..e30ba201910a 100644
--- a/include/linux/sunrpc/gss_krb5.h
+++ b/include/linux/sunrpc/gss_krb5.h
@@ -46,8 +46,8 @@ struct krb5_ctx {
 	unsigned char		seed[16];
 	int			signalg;
 	int			sealalg;
-	struct crypto_tfm	*enc;
-	struct crypto_tfm	*seq;
+	struct crypto_blkcipher	*enc;
+	struct crypto_blkcipher	*seq;
 	s32			endtime;
 	u32			seq_send;
 	struct xdr_netobj	mech_used;
@@ -136,26 +136,27 @@ gss_unwrap_kerberos(struct gss_ctx *ctx_id, int offset,
 
 
 u32
-krb5_encrypt(struct crypto_tfm * key,
+krb5_encrypt(struct crypto_blkcipher *key,
 	     void *iv, void *in, void *out, int length);
 
 u32
-krb5_decrypt(struct crypto_tfm * key,
+krb5_decrypt(struct crypto_blkcipher *key,
 	     void *iv, void *in, void *out, int length); 
 
 int
-gss_encrypt_xdr_buf(struct crypto_tfm *tfm, struct xdr_buf *outbuf, int offset,
-		struct page **pages);
+gss_encrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *outbuf,
+		    int offset, struct page **pages);
 
 int
-gss_decrypt_xdr_buf(struct crypto_tfm *tfm, struct xdr_buf *inbuf, int offset);
+gss_decrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *inbuf,
+		    int offset);
 
 s32
-krb5_make_seq_num(struct crypto_tfm * key,
+krb5_make_seq_num(struct crypto_blkcipher *key,
 		int direction,
 		s32 seqnum, unsigned char *cksum, unsigned char *buf);
 
 s32
-krb5_get_seq_num(struct crypto_tfm * key,
+krb5_get_seq_num(struct crypto_blkcipher *key,
 	       unsigned char *cksum,
 	       unsigned char *buf, int *direction, s32 * seqnum);
diff --git a/include/linux/sunrpc/gss_spkm3.h b/include/linux/sunrpc/gss_spkm3.h
index 336e218c2782..2cf3fbb40b4f 100644
--- a/include/linux/sunrpc/gss_spkm3.h
+++ b/include/linux/sunrpc/gss_spkm3.h
@@ -19,9 +19,9 @@ struct spkm3_ctx {
 	unsigned int		req_flags ;
 	struct xdr_netobj	share_key;
 	int			conf_alg;
-	struct crypto_tfm*	derived_conf_key;
+	struct crypto_blkcipher	*derived_conf_key;
 	int			intg_alg;
-	struct crypto_tfm*	derived_integ_key;
+	struct crypto_blkcipher	*derived_integ_key;
 	int			keyestb_alg;   /* alg used to get share_key */
 	int			owf_alg;   /* one way function */
 };
diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c
index 76b969e6904f..57192dfe3065 100644
--- a/net/sunrpc/auth_gss/gss_krb5_crypto.c
+++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c
@@ -49,7 +49,7 @@
 
 u32
 krb5_encrypt(
-	struct crypto_tfm *tfm,
+	struct crypto_blkcipher *tfm,
 	void * iv,
 	void * in,
 	void * out,
@@ -58,26 +58,27 @@ krb5_encrypt(
 	u32 ret = -EINVAL;
         struct scatterlist sg[1];
 	u8 local_iv[16] = {0};
+	struct blkcipher_desc desc = { .tfm = tfm, .info = local_iv };
 
 	dprintk("RPC:      krb5_encrypt: input data:\n");
 	print_hexl((u32 *)in, length, 0);
 
-	if (length % crypto_tfm_alg_blocksize(tfm) != 0)
+	if (length % crypto_blkcipher_blocksize(tfm) != 0)
 		goto out;
 
-	if (crypto_tfm_alg_ivsize(tfm) > 16) {
+	if (crypto_blkcipher_ivsize(tfm) > 16) {
 		dprintk("RPC:      gss_k5encrypt: tfm iv size to large %d\n",
-		         crypto_tfm_alg_ivsize(tfm));
+		         crypto_blkcipher_ivsize(tfm));
 		goto out;
 	}
 
 	if (iv)
-		memcpy(local_iv, iv, crypto_tfm_alg_ivsize(tfm));
+		memcpy(local_iv, iv, crypto_blkcipher_ivsize(tfm));
 
 	memcpy(out, in, length);
 	sg_set_buf(sg, out, length);
 
-	ret = crypto_cipher_encrypt_iv(tfm, sg, sg, length, local_iv);
+	ret = crypto_blkcipher_encrypt_iv(&desc, sg, sg, length);
 
 	dprintk("RPC:      krb5_encrypt: output data:\n");
 	print_hexl((u32 *)out, length, 0);
@@ -90,7 +91,7 @@ EXPORT_SYMBOL(krb5_encrypt);
 
 u32
 krb5_decrypt(
-     struct crypto_tfm *tfm,
+     struct crypto_blkcipher *tfm,
      void * iv,
      void * in,
      void * out,
@@ -99,25 +100,26 @@ krb5_decrypt(
 	u32 ret = -EINVAL;
 	struct scatterlist sg[1];
 	u8 local_iv[16] = {0};
+	struct blkcipher_desc desc = { .tfm = tfm, .info = local_iv };
 
 	dprintk("RPC:      krb5_decrypt: input data:\n");
 	print_hexl((u32 *)in, length, 0);
 
-	if (length % crypto_tfm_alg_blocksize(tfm) != 0)
+	if (length % crypto_blkcipher_blocksize(tfm) != 0)
 		goto out;
 
-	if (crypto_tfm_alg_ivsize(tfm) > 16) {
+	if (crypto_blkcipher_ivsize(tfm) > 16) {
 		dprintk("RPC:      gss_k5decrypt: tfm iv size to large %d\n",
-			crypto_tfm_alg_ivsize(tfm));
+			crypto_blkcipher_ivsize(tfm));
 		goto out;
 	}
 	if (iv)
-		memcpy(local_iv,iv, crypto_tfm_alg_ivsize(tfm));
+		memcpy(local_iv,iv, crypto_blkcipher_ivsize(tfm));
 
 	memcpy(out, in, length);
 	sg_set_buf(sg, out, length);
 
-	ret = crypto_cipher_decrypt_iv(tfm, sg, sg, length, local_iv);
+	ret = crypto_blkcipher_decrypt_iv(&desc, sg, sg, length);
 
 	dprintk("RPC:      krb5_decrypt: output_data:\n");
 	print_hexl((u32 *)out, length, 0);
@@ -240,7 +242,7 @@ EXPORT_SYMBOL(make_checksum);
 
 struct encryptor_desc {
 	u8 iv[8]; /* XXX hard-coded blocksize */
-	struct crypto_tfm *tfm;
+	struct blkcipher_desc desc;
 	int pos;
 	struct xdr_buf *outbuf;
 	struct page **pages;
@@ -285,8 +287,8 @@ encryptor(struct scatterlist *sg, void *data)
 	if (thislen == 0)
 		return 0;
 
-	ret = crypto_cipher_encrypt_iv(desc->tfm, desc->outfrags, desc->infrags,
-					thislen, desc->iv);
+	ret = crypto_blkcipher_encrypt_iv(&desc->desc, desc->outfrags,
+					  desc->infrags, thislen);
 	if (ret)
 		return ret;
 	if (fraglen) {
@@ -305,16 +307,18 @@ encryptor(struct scatterlist *sg, void *data)
 }
 
 int
-gss_encrypt_xdr_buf(struct crypto_tfm *tfm, struct xdr_buf *buf, int offset,
-		struct page **pages)
+gss_encrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *buf,
+		    int offset, struct page **pages)
 {
 	int ret;
 	struct encryptor_desc desc;
 
-	BUG_ON((buf->len - offset) % crypto_tfm_alg_blocksize(tfm) != 0);
+	BUG_ON((buf->len - offset) % crypto_blkcipher_blocksize(tfm) != 0);
 
 	memset(desc.iv, 0, sizeof(desc.iv));
-	desc.tfm = tfm;
+	desc.desc.tfm = tfm;
+	desc.desc.info = desc.iv;
+	desc.desc.flags = 0;
 	desc.pos = offset;
 	desc.outbuf = buf;
 	desc.pages = pages;
@@ -329,7 +333,7 @@ EXPORT_SYMBOL(gss_encrypt_xdr_buf);
 
 struct decryptor_desc {
 	u8 iv[8]; /* XXX hard-coded blocksize */
-	struct crypto_tfm *tfm;
+	struct blkcipher_desc desc;
 	struct scatterlist frags[4];
 	int fragno;
 	int fraglen;
@@ -355,8 +359,8 @@ decryptor(struct scatterlist *sg, void *data)
 	if (thislen == 0)
 		return 0;
 
-	ret = crypto_cipher_decrypt_iv(desc->tfm, desc->frags, desc->frags,
-					thislen, desc->iv);
+	ret = crypto_blkcipher_decrypt_iv(&desc->desc, desc->frags,
+					  desc->frags, thislen);
 	if (ret)
 		return ret;
 	if (fraglen) {
@@ -373,15 +377,18 @@ decryptor(struct scatterlist *sg, void *data)
 }
 
 int
-gss_decrypt_xdr_buf(struct crypto_tfm *tfm, struct xdr_buf *buf, int offset)
+gss_decrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *buf,
+		    int offset)
 {
 	struct decryptor_desc desc;
 
 	/* XXXJBF: */
-	BUG_ON((buf->len - offset) % crypto_tfm_alg_blocksize(tfm) != 0);
+	BUG_ON((buf->len - offset) % crypto_blkcipher_blocksize(tfm) != 0);
 
 	memset(desc.iv, 0, sizeof(desc.iv));
-	desc.tfm = tfm;
+	desc.desc.tfm = tfm;
+	desc.desc.info = desc.iv;
+	desc.desc.flags = 0;
 	desc.fragno = 0;
 	desc.fraglen = 0;
 	return process_xdr_buf(buf, offset, buf->len - offset, decryptor, &desc);
diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c
index 70e1e53a632b..325e72e4fd31 100644
--- a/net/sunrpc/auth_gss/gss_krb5_mech.c
+++ b/net/sunrpc/auth_gss/gss_krb5_mech.c
@@ -34,6 +34,7 @@
  *
  */
 
+#include <linux/err.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/types.h>
@@ -78,10 +79,10 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *res)
 }
 
 static inline const void *
-get_key(const void *p, const void *end, struct crypto_tfm **res)
+get_key(const void *p, const void *end, struct crypto_blkcipher **res)
 {
 	struct xdr_netobj	key;
-	int			alg, alg_mode;
+	int			alg;
 	char			*alg_name;
 
 	p = simple_get_bytes(p, end, &alg, sizeof(alg));
@@ -93,18 +94,19 @@ get_key(const void *p, const void *end, struct crypto_tfm **res)
 
 	switch (alg) {
 		case ENCTYPE_DES_CBC_RAW:
-			alg_name = "des";
-			alg_mode = CRYPTO_TFM_MODE_CBC;
+			alg_name = "cbc(des)";
 			break;
 		default:
 			printk("gss_kerberos_mech: unsupported algorithm %d\n", alg);
 			goto out_err_free_key;
 	}
-	if (!(*res = crypto_alloc_tfm(alg_name, alg_mode))) {
+	*res = crypto_alloc_blkcipher(alg_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(*res)) {
 		printk("gss_kerberos_mech: unable to initialize crypto algorithm %s\n", alg_name);
+		*res = NULL;
 		goto out_err_free_key;
 	}
-	if (crypto_cipher_setkey(*res, key.data, key.len)) {
+	if (crypto_blkcipher_setkey(*res, key.data, key.len)) {
 		printk("gss_kerberos_mech: error setting key for crypto algorithm %s\n", alg_name);
 		goto out_err_free_tfm;
 	}
@@ -113,7 +115,7 @@ get_key(const void *p, const void *end, struct crypto_tfm **res)
 	return p;
 
 out_err_free_tfm:
-	crypto_free_tfm(*res);
+	crypto_free_blkcipher(*res);
 out_err_free_key:
 	kfree(key.data);
 	p = ERR_PTR(-EINVAL);
@@ -172,9 +174,9 @@ gss_import_sec_context_kerberos(const void *p,
 	return 0;
 
 out_err_free_key2:
-	crypto_free_tfm(ctx->seq);
+	crypto_free_blkcipher(ctx->seq);
 out_err_free_key1:
-	crypto_free_tfm(ctx->enc);
+	crypto_free_blkcipher(ctx->enc);
 out_err_free_mech:
 	kfree(ctx->mech_used.data);
 out_err_free_ctx:
@@ -187,8 +189,8 @@ static void
 gss_delete_sec_context_kerberos(void *internal_ctx) {
 	struct krb5_ctx *kctx = internal_ctx;
 
-	crypto_free_tfm(kctx->seq);
-	crypto_free_tfm(kctx->enc);
+	crypto_free_blkcipher(kctx->seq);
+	crypto_free_blkcipher(kctx->enc);
 	kfree(kctx->mech_used.data);
 	kfree(kctx);
 }
diff --git a/net/sunrpc/auth_gss/gss_krb5_seqnum.c b/net/sunrpc/auth_gss/gss_krb5_seqnum.c
index c53ead39118d..c604baf3a5f6 100644
--- a/net/sunrpc/auth_gss/gss_krb5_seqnum.c
+++ b/net/sunrpc/auth_gss/gss_krb5_seqnum.c
@@ -41,7 +41,7 @@
 #endif
 
 s32
-krb5_make_seq_num(struct crypto_tfm *key,
+krb5_make_seq_num(struct crypto_blkcipher *key,
 		int direction,
 		s32 seqnum,
 		unsigned char *cksum, unsigned char *buf)
@@ -62,7 +62,7 @@ krb5_make_seq_num(struct crypto_tfm *key,
 }
 
 s32
-krb5_get_seq_num(struct crypto_tfm *key,
+krb5_get_seq_num(struct crypto_blkcipher *key,
 	       unsigned char *cksum,
 	       unsigned char *buf,
 	       int *direction, s32 * seqnum)
diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c
index 89d1f3e14128..f179415d0c38 100644
--- a/net/sunrpc/auth_gss/gss_krb5_wrap.c
+++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c
@@ -149,7 +149,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 		goto out_err;
 	}
 
-	blocksize = crypto_tfm_alg_blocksize(kctx->enc);
+	blocksize = crypto_blkcipher_blocksize(kctx->enc);
 	gss_krb5_add_padding(buf, offset, blocksize);
 	BUG_ON((buf->len - offset) % blocksize);
 	plainlen = blocksize + buf->len - offset;
@@ -346,7 +346,7 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 	/* Copy the data back to the right position.  XXX: Would probably be
 	 * better to copy and encrypt at the same time. */
 
-	blocksize = crypto_tfm_alg_blocksize(kctx->enc);
+	blocksize = crypto_blkcipher_blocksize(kctx->enc);
 	data_start = ptr + 22 + blocksize;
 	orig_start = buf->head[0].iov_base + offset;
 	data_len = (buf->head[0].iov_base + buf->head[0].iov_len) - data_start;
diff --git a/net/sunrpc/auth_gss/gss_spkm3_mech.c b/net/sunrpc/auth_gss/gss_spkm3_mech.c
index 88dcb52d171b..bdedf456bc17 100644
--- a/net/sunrpc/auth_gss/gss_spkm3_mech.c
+++ b/net/sunrpc/auth_gss/gss_spkm3_mech.c
@@ -34,6 +34,7 @@
  *
  */
 
+#include <linux/err.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/types.h>
@@ -83,10 +84,11 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *res)
 }
 
 static inline const void *
-get_key(const void *p, const void *end, struct crypto_tfm **res, int *resalg)
+get_key(const void *p, const void *end, struct crypto_blkcipher **res,
+	int *resalg)
 {
 	struct xdr_netobj	key = { 0 };
-	int			alg_mode,setkey = 0;
+	int			setkey = 0;
 	char			*alg_name;
 
 	p = simple_get_bytes(p, end, resalg, sizeof(*resalg));
@@ -98,14 +100,12 @@ get_key(const void *p, const void *end, struct crypto_tfm **res, int *resalg)
 
 	switch (*resalg) {
 		case NID_des_cbc:
-			alg_name = "des";
-			alg_mode = CRYPTO_TFM_MODE_CBC;
+			alg_name = "cbc(des)";
 			setkey = 1;
 			break;
 		case NID_cast5_cbc:
 			/* XXXX here in name only, not used */
-			alg_name = "cast5";
-			alg_mode = CRYPTO_TFM_MODE_CBC;
+			alg_name = "cbc(cast5)";
 			setkey = 0; /* XXX will need to set to 1 */
 			break;
 		case NID_md5:
@@ -113,19 +113,20 @@ get_key(const void *p, const void *end, struct crypto_tfm **res, int *resalg)
 				dprintk("RPC: SPKM3 get_key: NID_md5 zero Key length\n");
 			}
 			alg_name = "md5";
-			alg_mode = 0;
 			setkey = 0;
 			break;
 		default:
 			dprintk("gss_spkm3_mech: unsupported algorithm %d\n", *resalg);
 			goto out_err_free_key;
 	}
-	if (!(*res = crypto_alloc_tfm(alg_name, alg_mode))) {
+	*res = crypto_alloc_blkcipher(alg_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(*res)) {
 		printk("gss_spkm3_mech: unable to initialize crypto algorthm %s\n", alg_name);
+		*res = NULL;
 		goto out_err_free_key;
 	}
 	if (setkey) {
-		if (crypto_cipher_setkey(*res, key.data, key.len)) {
+		if (crypto_blkcipher_setkey(*res, key.data, key.len)) {
 			printk("gss_spkm3_mech: error setting key for crypto algorthm %s\n", alg_name);
 			goto out_err_free_tfm;
 		}
@@ -136,7 +137,7 @@ get_key(const void *p, const void *end, struct crypto_tfm **res, int *resalg)
 	return p;
 
 out_err_free_tfm:
-	crypto_free_tfm(*res);
+	crypto_free_blkcipher(*res);
 out_err_free_key:
 	if(key.len > 0)
 		kfree(key.data);
@@ -204,9 +205,9 @@ gss_import_sec_context_spkm3(const void *p, size_t len,
 	return 0;
 
 out_err_free_key2:
-	crypto_free_tfm(ctx->derived_integ_key);
+	crypto_free_blkcipher(ctx->derived_integ_key);
 out_err_free_key1:
-	crypto_free_tfm(ctx->derived_conf_key);
+	crypto_free_blkcipher(ctx->derived_conf_key);
 out_err_free_s_key:
 	kfree(ctx->share_key.data);
 out_err_free_mech:
@@ -223,8 +224,8 @@ static void
 gss_delete_sec_context_spkm3(void *internal_ctx) {
 	struct spkm3_ctx *sctx = internal_ctx;
 
-	crypto_free_tfm(sctx->derived_integ_key);
-	crypto_free_tfm(sctx->derived_conf_key);
+	crypto_free_blkcipher(sctx->derived_integ_key);
+	crypto_free_blkcipher(sctx->derived_conf_key);
 	kfree(sctx->share_key.data);
 	kfree(sctx->mech_used.data);
 	kfree(sctx);
-- 
cgit v1.2.3


From 03fd9cee7f46dddcd2562bc175d2c348502ce281 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 14 Aug 2006 23:11:53 +1000
Subject: [PATCH] scatterlist: Add const to sg_set_buf/sg_init_one pointer
 argument

This patch adds a const modifier to the buf argument of sg_set_buf and
sg_init_one.  This lets people call it with pointers that are const.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/scatterlist.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index 66ff545552f7..4efbd9c445f5 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -5,7 +5,7 @@
 #include <linux/mm.h>
 #include <linux/string.h>
 
-static inline void sg_set_buf(struct scatterlist *sg, void *buf,
+static inline void sg_set_buf(struct scatterlist *sg, const void *buf,
 			      unsigned int buflen)
 {
 	sg->page = virt_to_page(buf);
@@ -13,7 +13,7 @@ static inline void sg_set_buf(struct scatterlist *sg, void *buf,
 	sg->length = buflen;
 }
 
-static inline void sg_init_one(struct scatterlist *sg, void *buf,
+static inline void sg_init_one(struct scatterlist *sg, const void *buf,
 			       unsigned int buflen)
 {
 	memset(sg, 0, sizeof(*sg));
-- 
cgit v1.2.3


From 7226bc877a22244e8003924031435a4bffd52654 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 21 Aug 2006 21:40:49 +1000
Subject: [CRYPTO] api: Mark parts of cipher interface as deprecated

Mark the parts of the cipher interface that have been replaced by
block ciphers as deprecated.  Thanks to Andrew Morton for suggesting
doing this before removing them completely.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/cipher.c        | 34 ++++++++++++++++++++++++++++------
 include/linux/crypto.h | 48 ++++++++++++++++++++++++++++++++++++------------
 2 files changed, 64 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/crypto/cipher.c b/crypto/cipher.c
index 326461780673..9e03701cfdcc 100644
--- a/crypto/cipher.c
+++ b/crypto/cipher.c
@@ -23,6 +23,28 @@
 #include "internal.h"
 #include "scatterwalk.h"
 
+struct cipher_alg_compat {
+	unsigned int cia_min_keysize;
+	unsigned int cia_max_keysize;
+	int (*cia_setkey)(struct crypto_tfm *tfm, const u8 *key,
+	                  unsigned int keylen);
+	void (*cia_encrypt)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
+	void (*cia_decrypt)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
+
+	unsigned int (*cia_encrypt_ecb)(const struct cipher_desc *desc,
+					u8 *dst, const u8 *src,
+					unsigned int nbytes);
+	unsigned int (*cia_decrypt_ecb)(const struct cipher_desc *desc,
+					u8 *dst, const u8 *src,
+					unsigned int nbytes);
+	unsigned int (*cia_encrypt_cbc)(const struct cipher_desc *desc,
+					u8 *dst, const u8 *src,
+					unsigned int nbytes);
+	unsigned int (*cia_decrypt_cbc)(const struct cipher_desc *desc,
+					u8 *dst, const u8 *src,
+					unsigned int nbytes);
+};
+
 static inline void xor_64(u8 *a, const u8 *b)
 {
 	((u32 *)a)[0] ^= ((u32 *)b)[0];
@@ -276,7 +298,7 @@ static int ecb_encrypt(struct crypto_tfm *tfm,
                        struct scatterlist *src, unsigned int nbytes)
 {
 	struct cipher_desc desc;
-	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+	struct cipher_alg_compat *cipher = (void *)&tfm->__crt_alg->cra_cipher;
 
 	desc.tfm = tfm;
 	desc.crfn = cipher->cia_encrypt;
@@ -291,7 +313,7 @@ static int ecb_decrypt(struct crypto_tfm *tfm,
 		       unsigned int nbytes)
 {
 	struct cipher_desc desc;
-	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+	struct cipher_alg_compat *cipher = (void *)&tfm->__crt_alg->cra_cipher;
 
 	desc.tfm = tfm;
 	desc.crfn = cipher->cia_decrypt;
@@ -306,7 +328,7 @@ static int cbc_encrypt(struct crypto_tfm *tfm,
 		       unsigned int nbytes)
 {
 	struct cipher_desc desc;
-	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+	struct cipher_alg_compat *cipher = (void *)&tfm->__crt_alg->cra_cipher;
 
 	desc.tfm = tfm;
 	desc.crfn = cipher->cia_encrypt;
@@ -322,7 +344,7 @@ static int cbc_encrypt_iv(struct crypto_tfm *tfm,
                           unsigned int nbytes, u8 *iv)
 {
 	struct cipher_desc desc;
-	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+	struct cipher_alg_compat *cipher = (void *)&tfm->__crt_alg->cra_cipher;
 
 	desc.tfm = tfm;
 	desc.crfn = cipher->cia_encrypt;
@@ -338,7 +360,7 @@ static int cbc_decrypt(struct crypto_tfm *tfm,
 		       unsigned int nbytes)
 {
 	struct cipher_desc desc;
-	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+	struct cipher_alg_compat *cipher = (void *)&tfm->__crt_alg->cra_cipher;
 
 	desc.tfm = tfm;
 	desc.crfn = cipher->cia_decrypt;
@@ -354,7 +376,7 @@ static int cbc_decrypt_iv(struct crypto_tfm *tfm,
                           unsigned int nbytes, u8 *iv)
 {
 	struct cipher_desc desc;
-	struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher;
+	struct cipher_alg_compat *cipher = (void *)&tfm->__crt_alg->cra_cipher;
 
 	desc.tfm = tfm;
 	desc.crfn = cipher->cia_decrypt;
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 5a5466d518e8..0be666b50463 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -20,7 +20,6 @@
 #include <asm/atomic.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/types.h>
 #include <linux/list.h>
 #include <linux/slab.h>
 #include <linux/string.h>
@@ -137,16 +136,16 @@ struct cipher_alg {
 
 	unsigned int (*cia_encrypt_ecb)(const struct cipher_desc *desc,
 					u8 *dst, const u8 *src,
-					unsigned int nbytes);
+					unsigned int nbytes) __deprecated;
 	unsigned int (*cia_decrypt_ecb)(const struct cipher_desc *desc,
 					u8 *dst, const u8 *src,
-					unsigned int nbytes);
+					unsigned int nbytes) __deprecated;
 	unsigned int (*cia_encrypt_cbc)(const struct cipher_desc *desc,
 					u8 *dst, const u8 *src,
-					unsigned int nbytes);
+					unsigned int nbytes) __deprecated;
 	unsigned int (*cia_decrypt_cbc)(const struct cipher_desc *desc,
 					u8 *dst, const u8 *src,
-					unsigned int nbytes);
+					unsigned int nbytes) __deprecated;
 };
 
 struct digest_alg {
@@ -358,18 +357,23 @@ static inline u32 crypto_tfm_alg_type(struct crypto_tfm *tfm)
 	return tfm->__crt_alg->cra_flags & CRYPTO_ALG_TYPE_MASK;
 }
 
+static unsigned int crypto_tfm_alg_min_keysize(struct crypto_tfm *tfm)
+	__deprecated;
 static inline unsigned int crypto_tfm_alg_min_keysize(struct crypto_tfm *tfm)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
 	return tfm->__crt_alg->cra_cipher.cia_min_keysize;
 }
 
+static unsigned int crypto_tfm_alg_max_keysize(struct crypto_tfm *tfm)
+	__deprecated;
 static inline unsigned int crypto_tfm_alg_max_keysize(struct crypto_tfm *tfm)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
 	return tfm->__crt_alg->cra_cipher.cia_max_keysize;
 }
 
+static unsigned int crypto_tfm_alg_ivsize(struct crypto_tfm *tfm) __deprecated;
 static inline unsigned int crypto_tfm_alg_ivsize(struct crypto_tfm *tfm)
 {
 	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
@@ -622,6 +626,13 @@ static inline void crypto_cipher_clear_flags(struct crypto_cipher *tfm,
 	crypto_tfm_clear_flags(crypto_cipher_tfm(tfm), flags);
 }
 
+static inline int crypto_cipher_setkey(struct crypto_cipher *tfm,
+                                       const u8 *key, unsigned int keylen)
+{
+	return crypto_cipher_crt(tfm)->cit_setkey(crypto_cipher_tfm(tfm),
+						  key, keylen);
+}
+
 static inline void crypto_cipher_encrypt_one(struct crypto_cipher *tfm,
 					     u8 *dst, const u8 *src)
 {
@@ -671,13 +682,10 @@ static inline int crypto_digest_setkey(struct crypto_tfm *tfm,
 	return tfm->crt_digest.dit_setkey(tfm, key, keylen);
 }
 
-static inline int crypto_cipher_setkey(struct crypto_tfm *tfm,
-                                       const u8 *key, unsigned int keylen)
-{
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_CIPHER);
-	return tfm->crt_cipher.cit_setkey(tfm, key, keylen);
-}
-
+static int crypto_cipher_encrypt(struct crypto_tfm *tfm,
+				 struct scatterlist *dst,
+				 struct scatterlist *src,
+				 unsigned int nbytes) __deprecated;
 static inline int crypto_cipher_encrypt(struct crypto_tfm *tfm,
                                         struct scatterlist *dst,
                                         struct scatterlist *src,
@@ -687,6 +695,10 @@ static inline int crypto_cipher_encrypt(struct crypto_tfm *tfm,
 	return tfm->crt_cipher.cit_encrypt(tfm, dst, src, nbytes);
 }                                        
 
+static int crypto_cipher_encrypt_iv(struct crypto_tfm *tfm,
+				    struct scatterlist *dst,
+				    struct scatterlist *src,
+				    unsigned int nbytes, u8 *iv) __deprecated;
 static inline int crypto_cipher_encrypt_iv(struct crypto_tfm *tfm,
                                            struct scatterlist *dst,
                                            struct scatterlist *src,
@@ -696,6 +708,10 @@ static inline int crypto_cipher_encrypt_iv(struct crypto_tfm *tfm,
 	return tfm->crt_cipher.cit_encrypt_iv(tfm, dst, src, nbytes, iv);
 }                                        
 
+static int crypto_cipher_decrypt(struct crypto_tfm *tfm,
+				 struct scatterlist *dst,
+				 struct scatterlist *src,
+				 unsigned int nbytes) __deprecated;
 static inline int crypto_cipher_decrypt(struct crypto_tfm *tfm,
                                         struct scatterlist *dst,
                                         struct scatterlist *src,
@@ -705,6 +721,10 @@ static inline int crypto_cipher_decrypt(struct crypto_tfm *tfm,
 	return tfm->crt_cipher.cit_decrypt(tfm, dst, src, nbytes);
 }
 
+static int crypto_cipher_decrypt_iv(struct crypto_tfm *tfm,
+				    struct scatterlist *dst,
+				    struct scatterlist *src,
+				    unsigned int nbytes, u8 *iv) __deprecated;
 static inline int crypto_cipher_decrypt_iv(struct crypto_tfm *tfm,
                                            struct scatterlist *dst,
                                            struct scatterlist *src,
@@ -714,6 +734,8 @@ static inline int crypto_cipher_decrypt_iv(struct crypto_tfm *tfm,
 	return tfm->crt_cipher.cit_decrypt_iv(tfm, dst, src, nbytes, iv);
 }
 
+static void crypto_cipher_set_iv(struct crypto_tfm *tfm,
+				 const u8 *src, unsigned int len) __deprecated;
 static inline void crypto_cipher_set_iv(struct crypto_tfm *tfm,
                                         const u8 *src, unsigned int len)
 {
@@ -721,6 +743,8 @@ static inline void crypto_cipher_set_iv(struct crypto_tfm *tfm,
 	memcpy(tfm->crt_cipher.cit_iv, src, len);
 }
 
+static void crypto_cipher_get_iv(struct crypto_tfm *tfm,
+				 u8 *dst, unsigned int len) __deprecated;
 static inline void crypto_cipher_get_iv(struct crypto_tfm *tfm,
                                         u8 *dst, unsigned int len)
 {
-- 
cgit v1.2.3


From 055bcee3102dc35f019b69df9c2618e9d6dd1c09 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 19 Aug 2006 22:24:23 +1000
Subject: [CRYPTO] digest: Added user API for new hash type

The existing digest user interface is inadequate for support asynchronous
operations.  For one it doesn't return a value to indicate success or
failure, nor does it take a per-operation descriptor which is essential
for the issuing of requests while other requests are still outstanding.

This patch is the first in a series of steps to remodel the interface
for asynchronous operations.

For the ease of transition the new interface will be known as "hash"
while the old one will remain as "digest".

This patch also changes sg_next to allow chaining.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Kconfig          |   4 ++
 crypto/Makefile         |   3 +
 crypto/digest.c         | 129 ++++++++++++++++++++++++++++--------
 crypto/hash.c           |  61 +++++++++++++++++
 crypto/hmac.c           |  12 ++--
 crypto/scatterwalk.h    |   4 +-
 include/crypto/algapi.h |   6 ++
 include/linux/crypto.h  | 172 ++++++++++++++++++++++++++++++++++++++----------
 8 files changed, 321 insertions(+), 70 deletions(-)
 create mode 100644 crypto/hash.c

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index be5eb0cb7c30..69c5f992bcd4 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -20,6 +20,10 @@ config CRYPTO_BLKCIPHER
 	tristate
 	select CRYPTO_ALGAPI
 
+config CRYPTO_HASH
+	tristate
+	select CRYPTO_ALGAPI
+
 config CRYPTO_MANAGER
 	tristate "Cryptographic algorithm manager"
 	select CRYPTO_ALGAPI
diff --git a/crypto/Makefile b/crypto/Makefile
index 5e1ff4e0b1fc..72366208e291 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -10,6 +10,9 @@ obj-$(CONFIG_CRYPTO_ALGAPI) += crypto_algapi.o
 
 obj-$(CONFIG_CRYPTO_BLKCIPHER) += blkcipher.o
 
+crypto_hash-objs := hash.o
+obj-$(CONFIG_CRYPTO_HASH) += crypto_hash.o
+
 obj-$(CONFIG_CRYPTO_MANAGER) += cryptomgr.o
 obj-$(CONFIG_CRYPTO_HMAC) += hmac.o
 obj-$(CONFIG_CRYPTO_NULL) += crypto_null.o
diff --git a/crypto/digest.c b/crypto/digest.c
index 96244a528844..5873063db840 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -11,29 +11,89 @@
  * any later version.
  *
  */
-#include <linux/crypto.h>
+
 #include <linux/mm.h>
 #include <linux/errno.h>
 #include <linux/highmem.h>
-#include <asm/scatterlist.h>
+#include <linux/module.h>
+#include <linux/scatterlist.h>
+
 #include "internal.h"
+#include "scatterwalk.h"
 
-static void init(struct crypto_tfm *tfm)
+void crypto_digest_init(struct crypto_tfm *tfm)
 {
-	tfm->__crt_alg->cra_digest.dia_init(tfm);
+	struct crypto_hash *hash = crypto_hash_cast(tfm);
+	struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags };
+
+	crypto_hash_init(&desc);
 }
+EXPORT_SYMBOL_GPL(crypto_digest_init);
 
-static void update(struct crypto_tfm *tfm,
-                   struct scatterlist *sg, unsigned int nsg)
+void crypto_digest_update(struct crypto_tfm *tfm,
+			  struct scatterlist *sg, unsigned int nsg)
 {
+	struct crypto_hash *hash = crypto_hash_cast(tfm);
+	struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags };
+	unsigned int nbytes = 0;
 	unsigned int i;
+
+	for (i = 0; i < nsg; i++)
+		nbytes += sg[i].length;
+
+	crypto_hash_update(&desc, sg, nbytes);
+}
+EXPORT_SYMBOL_GPL(crypto_digest_update);
+
+void crypto_digest_final(struct crypto_tfm *tfm, u8 *out)
+{
+	struct crypto_hash *hash = crypto_hash_cast(tfm);
+	struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags };
+
+	crypto_hash_final(&desc, out);
+}
+EXPORT_SYMBOL_GPL(crypto_digest_final);
+
+void crypto_digest_digest(struct crypto_tfm *tfm,
+			  struct scatterlist *sg, unsigned int nsg, u8 *out)
+{
+	struct crypto_hash *hash = crypto_hash_cast(tfm);
+	struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags };
+	unsigned int nbytes = 0;
+	unsigned int i;
+
+	for (i = 0; i < nsg; i++)
+		nbytes += sg[i].length;
+
+	crypto_hash_digest(&desc, sg, nbytes, out);
+}
+EXPORT_SYMBOL_GPL(crypto_digest_digest);
+
+static int init(struct hash_desc *desc)
+{
+	struct crypto_tfm *tfm = crypto_hash_tfm(desc->tfm);
+
+	tfm->__crt_alg->cra_digest.dia_init(tfm);
+	return 0;
+}
+
+static int update(struct hash_desc *desc,
+		  struct scatterlist *sg, unsigned int nbytes)
+{
+	struct crypto_tfm *tfm = crypto_hash_tfm(desc->tfm);
 	unsigned int alignmask = crypto_tfm_alg_alignmask(tfm);
 
-	for (i = 0; i < nsg; i++) {
+	if (!nbytes)
+		return 0;
+
+	for (;;) {
+		struct page *pg = sg->page;
+		unsigned int offset = sg->offset;
+		unsigned int l = sg->length;
 
-		struct page *pg = sg[i].page;
-		unsigned int offset = sg[i].offset;
-		unsigned int l = sg[i].length;
+		if (unlikely(l > nbytes))
+			l = nbytes;
+		nbytes -= l;
 
 		do {
 			unsigned int bytes_from_page = min(l, ((unsigned int)
@@ -55,16 +115,23 @@ static void update(struct crypto_tfm *tfm,
 			tfm->__crt_alg->cra_digest.dia_update(tfm, p,
 							      bytes_from_page);
 			crypto_kunmap(src, 0);
-			crypto_yield(tfm->crt_flags);
+			crypto_yield(desc->flags);
 			offset = 0;
 			pg++;
 			l -= bytes_from_page;
 		} while (l > 0);
+
+		if (!nbytes)
+			break;
+		sg = sg_next(sg);
 	}
+
+	return 0;
 }
 
-static void final(struct crypto_tfm *tfm, u8 *out)
+static int final(struct hash_desc *desc, u8 *out)
 {
+	struct crypto_tfm *tfm = crypto_hash_tfm(desc->tfm);
 	unsigned long alignmask = crypto_tfm_alg_alignmask(tfm);
 	struct digest_alg *digest = &tfm->__crt_alg->cra_digest;
 
@@ -78,26 +145,30 @@ static void final(struct crypto_tfm *tfm, u8 *out)
 		memcpy(out, dst, digest->dia_digestsize);
 	} else
 		digest->dia_final(tfm, out);
+
+	return 0;
 }
 
-static int nosetkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
+static int nosetkey(struct crypto_hash *tfm, const u8 *key, unsigned int keylen)
 {
-	tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK;
+	crypto_hash_clear_flags(tfm, CRYPTO_TFM_RES_MASK);
 	return -ENOSYS;
 }
 
-static int setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
+static int setkey(struct crypto_hash *hash, const u8 *key, unsigned int keylen)
 {
-	tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK;
+	struct crypto_tfm *tfm = crypto_hash_tfm(hash);
+
+	crypto_hash_clear_flags(hash, CRYPTO_TFM_RES_MASK);
 	return tfm->__crt_alg->cra_digest.dia_setkey(tfm, key, keylen);
 }
 
-static void digest(struct crypto_tfm *tfm,
-                   struct scatterlist *sg, unsigned int nsg, u8 *out)
+static int digest(struct hash_desc *desc,
+		  struct scatterlist *sg, unsigned int nbytes, u8 *out)
 {
-	init(tfm);
-	update(tfm, sg, nsg);
-	final(tfm, out);
+	init(desc);
+	update(desc, sg, nbytes);
+	return final(desc, out);
 }
 
 int crypto_init_digest_flags(struct crypto_tfm *tfm, u32 flags)
@@ -107,14 +178,18 @@ int crypto_init_digest_flags(struct crypto_tfm *tfm, u32 flags)
 
 int crypto_init_digest_ops(struct crypto_tfm *tfm)
 {
-	struct digest_tfm *ops = &tfm->crt_digest;
+	struct hash_tfm *ops = &tfm->crt_hash;
 	struct digest_alg *dalg = &tfm->__crt_alg->cra_digest;
+
+	if (dalg->dia_digestsize > crypto_tfm_alg_blocksize(tfm))
+		return -EINVAL;
 	
-	ops->dit_init	= init;
-	ops->dit_update	= update;
-	ops->dit_final	= final;
-	ops->dit_digest	= digest;
-	ops->dit_setkey	= dalg->dia_setkey ? setkey : nosetkey;
+	ops->init	= init;
+	ops->update	= update;
+	ops->final	= final;
+	ops->digest	= digest;
+	ops->setkey	= dalg->dia_setkey ? setkey : nosetkey;
+	ops->digestsize	= dalg->dia_digestsize;
 	
 	return crypto_alloc_hmac_block(tfm);
 }
diff --git a/crypto/hash.c b/crypto/hash.c
new file mode 100644
index 000000000000..cdec23d885fe
--- /dev/null
+++ b/crypto/hash.c
@@ -0,0 +1,61 @@
+/*
+ * Cryptographic Hash operations.
+ * 
+ * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ */
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/seq_file.h>
+
+#include "internal.h"
+
+static unsigned int crypto_hash_ctxsize(struct crypto_alg *alg)
+{
+	return alg->cra_ctxsize;
+}
+
+static int crypto_init_hash_ops(struct crypto_tfm *tfm)
+{
+	struct hash_tfm *crt = &tfm->crt_hash;
+	struct hash_alg *alg = &tfm->__crt_alg->cra_hash;
+
+	if (alg->digestsize > crypto_tfm_alg_blocksize(tfm))
+		return -EINVAL;
+
+	crt->init = alg->init;
+	crt->update = alg->update;
+	crt->final = alg->final;
+	crt->digest = alg->digest;
+	crt->setkey = alg->setkey;
+	crt->digestsize = alg->digestsize;
+
+	return 0;
+}
+
+static void crypto_hash_show(struct seq_file *m, struct crypto_alg *alg)
+	__attribute_used__;
+static void crypto_hash_show(struct seq_file *m, struct crypto_alg *alg)
+{
+	seq_printf(m, "type         : hash\n");
+	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
+	seq_printf(m, "digestsize   : %u\n", alg->cra_hash.digestsize);
+}
+
+const struct crypto_type crypto_hash_type = {
+	.ctxsize = crypto_hash_ctxsize,
+	.init = crypto_init_hash_ops,
+#ifdef CONFIG_PROC_FS
+	.show = crypto_hash_show,
+#endif
+};
+EXPORT_SYMBOL_GPL(crypto_hash_type);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Generic cryptographic hash type");
diff --git a/crypto/hmac.c b/crypto/hmac.c
index 46120dee5ada..ecf7b0a95b56 100644
--- a/crypto/hmac.c
+++ b/crypto/hmac.c
@@ -35,9 +35,9 @@ int crypto_alloc_hmac_block(struct crypto_tfm *tfm)
 
 	BUG_ON(!crypto_tfm_alg_blocksize(tfm));
 	
-	tfm->crt_digest.dit_hmac_block = kmalloc(crypto_tfm_alg_blocksize(tfm),
-	                                         GFP_KERNEL);
-	if (tfm->crt_digest.dit_hmac_block == NULL)
+	tfm->crt_hash.hmac_block = kmalloc(crypto_tfm_alg_blocksize(tfm),
+					   GFP_KERNEL);
+	if (tfm->crt_hash.hmac_block == NULL)
 		ret = -ENOMEM;
 
 	return ret;
@@ -46,14 +46,14 @@ int crypto_alloc_hmac_block(struct crypto_tfm *tfm)
 
 void crypto_free_hmac_block(struct crypto_tfm *tfm)
 {
-	kfree(tfm->crt_digest.dit_hmac_block);
+	kfree(tfm->crt_hash.hmac_block);
 }
 
 void crypto_hmac_init(struct crypto_tfm *tfm, u8 *key, unsigned int *keylen)
 {
 	unsigned int i;
 	struct scatterlist tmp;
-	char *ipad = tfm->crt_digest.dit_hmac_block;
+	char *ipad = tfm->crt_hash.hmac_block;
 	
 	if (*keylen > crypto_tfm_alg_blocksize(tfm)) {
 		hash_key(tfm, key, *keylen);
@@ -83,7 +83,7 @@ void crypto_hmac_final(struct crypto_tfm *tfm, u8 *key,
 {
 	unsigned int i;
 	struct scatterlist tmp;
-	char *opad = tfm->crt_digest.dit_hmac_block;
+	char *opad = tfm->crt_hash.hmac_block;
 	
 	if (*keylen > crypto_tfm_alg_blocksize(tfm)) {
 		hash_key(tfm, key, *keylen);
diff --git a/crypto/scatterwalk.h b/crypto/scatterwalk.h
index ace595a2e119..f1592cc2d0f4 100644
--- a/crypto/scatterwalk.h
+++ b/crypto/scatterwalk.h
@@ -20,11 +20,9 @@
 
 #include "internal.h"
 
-/* Define sg_next is an inline routine now in case we want to change
-   scatterlist to a linked list later. */
 static inline struct scatterlist *sg_next(struct scatterlist *sg)
 {
-	return sg + 1;
+	return (++sg)->length ? sg : (void *)sg->page;
 }
 
 static inline unsigned long scatterwalk_samebuf(struct scatter_walk *walk_in,
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 444f602724db..5748aecdb414 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -82,6 +82,7 @@ struct blkcipher_walk {
 };
 
 extern const struct crypto_type crypto_blkcipher_type;
+extern const struct crypto_type crypto_hash_type;
 
 void crypto_mod_put(struct crypto_alg *alg);
 
@@ -136,6 +137,11 @@ static inline struct cipher_alg *crypto_cipher_alg(struct crypto_cipher *tfm)
 	return &crypto_cipher_tfm(tfm)->__crt_alg->cra_cipher;
 }
 
+static inline void *crypto_hash_ctx_aligned(struct crypto_hash *tfm)
+{
+	return crypto_tfm_ctx_aligned(&tfm->base);
+}
+
 static inline void blkcipher_walk_init(struct blkcipher_walk *walk,
 				       struct scatterlist *dst,
 				       struct scatterlist *src,
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 0be666b50463..40c0aab8ad4c 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -31,8 +31,11 @@
 #define CRYPTO_ALG_TYPE_MASK		0x0000000f
 #define CRYPTO_ALG_TYPE_CIPHER		0x00000001
 #define CRYPTO_ALG_TYPE_DIGEST		0x00000002
-#define CRYPTO_ALG_TYPE_BLKCIPHER	0x00000003
-#define CRYPTO_ALG_TYPE_COMPRESS	0x00000004
+#define CRYPTO_ALG_TYPE_HASH		0x00000003
+#define CRYPTO_ALG_TYPE_BLKCIPHER	0x00000004
+#define CRYPTO_ALG_TYPE_COMPRESS	0x00000005
+
+#define CRYPTO_ALG_TYPE_HASH_MASK	0x0000000e
 
 #define CRYPTO_ALG_LARVAL		0x00000010
 #define CRYPTO_ALG_DEAD			0x00000020
@@ -90,6 +93,7 @@
 
 struct scatterlist;
 struct crypto_blkcipher;
+struct crypto_hash;
 struct crypto_tfm;
 struct crypto_type;
 
@@ -107,6 +111,11 @@ struct cipher_desc {
 	void *info;
 };
 
+struct hash_desc {
+	struct crypto_hash *tfm;
+	u32 flags;
+};
+
 /*
  * Algorithms: modular crypto algorithm implementations, managed
  * via crypto_register_alg() and crypto_unregister_alg().
@@ -158,6 +167,19 @@ struct digest_alg {
 	                  unsigned int keylen);
 };
 
+struct hash_alg {
+	int (*init)(struct hash_desc *desc);
+	int (*update)(struct hash_desc *desc, struct scatterlist *sg,
+		      unsigned int nbytes);
+	int (*final)(struct hash_desc *desc, u8 *out);
+	int (*digest)(struct hash_desc *desc, struct scatterlist *sg,
+		      unsigned int nbytes, u8 *out);
+	int (*setkey)(struct crypto_hash *tfm, const u8 *key,
+		      unsigned int keylen);
+
+	unsigned int digestsize;
+};
+
 struct compress_alg {
 	int (*coa_compress)(struct crypto_tfm *tfm, const u8 *src,
 			    unsigned int slen, u8 *dst, unsigned int *dlen);
@@ -168,6 +190,7 @@ struct compress_alg {
 #define cra_blkcipher	cra_u.blkcipher
 #define cra_cipher	cra_u.cipher
 #define cra_digest	cra_u.digest
+#define cra_hash	cra_u.hash
 #define cra_compress	cra_u.compress
 
 struct crypto_alg {
@@ -191,6 +214,7 @@ struct crypto_alg {
 		struct blkcipher_alg blkcipher;
 		struct cipher_alg cipher;
 		struct digest_alg digest;
+		struct hash_alg hash;
 		struct compress_alg compress;
 	} cra_u;
 
@@ -262,18 +286,19 @@ struct cipher_tfm {
 	void (*cit_decrypt_one)(struct crypto_tfm *tfm, u8 *dst, const u8 *src);
 };
 
-struct digest_tfm {
-	void (*dit_init)(struct crypto_tfm *tfm);
-	void (*dit_update)(struct crypto_tfm *tfm,
-	                   struct scatterlist *sg, unsigned int nsg);
-	void (*dit_final)(struct crypto_tfm *tfm, u8 *out);
-	void (*dit_digest)(struct crypto_tfm *tfm, struct scatterlist *sg,
-	                   unsigned int nsg, u8 *out);
-	int (*dit_setkey)(struct crypto_tfm *tfm,
-	                  const u8 *key, unsigned int keylen);
+struct hash_tfm {
+	int (*init)(struct hash_desc *desc);
+	int (*update)(struct hash_desc *desc,
+		      struct scatterlist *sg, unsigned int nsg);
+	int (*final)(struct hash_desc *desc, u8 *out);
+	int (*digest)(struct hash_desc *desc, struct scatterlist *sg,
+		      unsigned int nsg, u8 *out);
+	int (*setkey)(struct crypto_hash *tfm, const u8 *key,
+		      unsigned int keylen);
 #ifdef CONFIG_CRYPTO_HMAC
-	void *dit_hmac_block;
+	void *hmac_block;
 #endif
+	unsigned int digestsize;
 };
 
 struct compress_tfm {
@@ -287,7 +312,7 @@ struct compress_tfm {
 
 #define crt_blkcipher	crt_u.blkcipher
 #define crt_cipher	crt_u.cipher
-#define crt_digest	crt_u.digest
+#define crt_hash	crt_u.hash
 #define crt_compress	crt_u.compress
 
 struct crypto_tfm {
@@ -297,7 +322,7 @@ struct crypto_tfm {
 	union {
 		struct blkcipher_tfm blkcipher;
 		struct cipher_tfm cipher;
-		struct digest_tfm digest;
+		struct hash_tfm hash;
 		struct compress_tfm compress;
 	} crt_u;
 	
@@ -312,6 +337,10 @@ struct crypto_blkcipher {
 	struct crypto_tfm base;
 };
 
+struct crypto_hash {
+	struct crypto_tfm base;
+};
+
 enum {
 	CRYPTOA_UNSPEC,
 	CRYPTOA_ALG,
@@ -647,39 +676,114 @@ static inline void crypto_cipher_decrypt_one(struct crypto_cipher *tfm,
 						dst, src);
 }
 
-static inline void crypto_digest_init(struct crypto_tfm *tfm)
+void crypto_digest_init(struct crypto_tfm *tfm);
+void crypto_digest_update(struct crypto_tfm *tfm,
+			  struct scatterlist *sg, unsigned int nsg);
+void crypto_digest_final(struct crypto_tfm *tfm, u8 *out);
+void crypto_digest_digest(struct crypto_tfm *tfm,
+			  struct scatterlist *sg, unsigned int nsg, u8 *out);
+
+static inline struct crypto_hash *__crypto_hash_cast(struct crypto_tfm *tfm)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-	tfm->crt_digest.dit_init(tfm);
+	return (struct crypto_hash *)tfm;
 }
 
-static inline void crypto_digest_update(struct crypto_tfm *tfm,
-                                        struct scatterlist *sg,
-                                        unsigned int nsg)
+static inline struct crypto_hash *crypto_hash_cast(struct crypto_tfm *tfm)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-	tfm->crt_digest.dit_update(tfm, sg, nsg);
+	BUG_ON((crypto_tfm_alg_type(tfm) ^ CRYPTO_ALG_TYPE_HASH) &
+	       CRYPTO_ALG_TYPE_HASH_MASK);
+	return __crypto_hash_cast(tfm);
 }
 
-static inline void crypto_digest_final(struct crypto_tfm *tfm, u8 *out)
+static inline int crypto_digest_setkey(struct crypto_tfm *tfm,
+                                       const u8 *key, unsigned int keylen)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-	tfm->crt_digest.dit_final(tfm, out);
+	return tfm->crt_hash.setkey(crypto_hash_cast(tfm), key, keylen);
 }
 
-static inline void crypto_digest_digest(struct crypto_tfm *tfm,
-                                        struct scatterlist *sg,
-                                        unsigned int nsg, u8 *out)
+static inline struct crypto_hash *crypto_alloc_hash(const char *alg_name,
+						    u32 type, u32 mask)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-	tfm->crt_digest.dit_digest(tfm, sg, nsg, out);
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_HASH;
+	mask |= CRYPTO_ALG_TYPE_HASH_MASK;
+
+	return __crypto_hash_cast(crypto_alloc_base(alg_name, type, mask));
 }
 
-static inline int crypto_digest_setkey(struct crypto_tfm *tfm,
-                                       const u8 *key, unsigned int keylen)
+static inline struct crypto_tfm *crypto_hash_tfm(struct crypto_hash *tfm)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_DIGEST);
-	return tfm->crt_digest.dit_setkey(tfm, key, keylen);
+	return &tfm->base;
+}
+
+static inline void crypto_free_hash(struct crypto_hash *tfm)
+{
+	crypto_free_tfm(crypto_hash_tfm(tfm));
+}
+
+static inline struct hash_tfm *crypto_hash_crt(struct crypto_hash *tfm)
+{
+	return &crypto_hash_tfm(tfm)->crt_hash;
+}
+
+static inline unsigned int crypto_hash_blocksize(struct crypto_hash *tfm)
+{
+	return crypto_tfm_alg_blocksize(crypto_hash_tfm(tfm));
+}
+
+static inline unsigned int crypto_hash_alignmask(struct crypto_hash *tfm)
+{
+	return crypto_tfm_alg_alignmask(crypto_hash_tfm(tfm));
+}
+
+static inline unsigned int crypto_hash_digestsize(struct crypto_hash *tfm)
+{
+	return crypto_hash_crt(tfm)->digestsize;
+}
+
+static inline u32 crypto_hash_get_flags(struct crypto_hash *tfm)
+{
+	return crypto_tfm_get_flags(crypto_hash_tfm(tfm));
+}
+
+static inline void crypto_hash_set_flags(struct crypto_hash *tfm, u32 flags)
+{
+	crypto_tfm_set_flags(crypto_hash_tfm(tfm), flags);
+}
+
+static inline void crypto_hash_clear_flags(struct crypto_hash *tfm, u32 flags)
+{
+	crypto_tfm_clear_flags(crypto_hash_tfm(tfm), flags);
+}
+
+static inline int crypto_hash_init(struct hash_desc *desc)
+{
+	return crypto_hash_crt(desc->tfm)->init(desc);
+}
+
+static inline int crypto_hash_update(struct hash_desc *desc,
+				     struct scatterlist *sg,
+				     unsigned int nbytes)
+{
+	return crypto_hash_crt(desc->tfm)->update(desc, sg, nbytes);
+}
+
+static inline int crypto_hash_final(struct hash_desc *desc, u8 *out)
+{
+	return crypto_hash_crt(desc->tfm)->final(desc, out);
+}
+
+static inline int crypto_hash_digest(struct hash_desc *desc,
+				     struct scatterlist *sg,
+				     unsigned int nbytes, u8 *out)
+{
+	return crypto_hash_crt(desc->tfm)->digest(desc, sg, nbytes, out);
+}
+
+static inline int crypto_hash_setkey(struct crypto_hash *hash,
+				     const u8 *key, unsigned int keylen)
+{
+	return crypto_hash_crt(hash)->setkey(hash, key, keylen);
 }
 
 static int crypto_cipher_encrypt(struct crypto_tfm *tfm,
-- 
cgit v1.2.3


From 07d4ee583e21830ec5604d31f65cdc60a6eca19e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 20 Aug 2006 14:24:50 +1000
Subject: [IPSEC]: Use HMAC template and hash interface

This patch converts IPsec to use the new HMAC template.  The names of
existing simple digest algorithms may still be used to refer to their
HMAC composites.

The same structure can be used by other MACs such as AES-XCBC-MAC.

This patch also switches from the digest interface to hash.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ah.h     | 29 ++++++++++++++++++-----------
 include/net/esp.h    | 28 ++++++++++++++++------------
 include/net/xfrm.h   |  9 +++++----
 net/ipv4/ah4.c       | 36 ++++++++++++++++++++++++------------
 net/ipv4/esp4.c      | 36 +++++++++++++++++++++---------------
 net/ipv6/ah6.c       | 35 +++++++++++++++++++++++------------
 net/ipv6/esp6.c      | 42 ++++++++++++++++++++++++------------------
 net/xfrm/xfrm_algo.c | 40 +++++++++++++++++++++++++++-------------
 8 files changed, 158 insertions(+), 97 deletions(-)

(limited to 'include')

diff --git a/include/net/ah.h b/include/net/ah.h
index 8e27c9ba8b84..8f257c159902 100644
--- a/include/net/ah.h
+++ b/include/net/ah.h
@@ -15,22 +15,29 @@ struct ah_data
 	int			icv_full_len;
 	int			icv_trunc_len;
 
-	void			(*icv)(struct ah_data*,
-	                               struct sk_buff *skb, u8 *icv);
-
-	struct crypto_tfm	*tfm;
+	struct crypto_hash	*tfm;
 };
 
-static inline void
-ah_hmac_digest(struct ah_data *ahp, struct sk_buff *skb, u8 *auth_data)
+static inline int ah_mac_digest(struct ah_data *ahp, struct sk_buff *skb,
+				u8 *auth_data)
 {
-	struct crypto_tfm *tfm = ahp->tfm;
+	struct hash_desc desc;
+	int err;
+
+	desc.tfm = ahp->tfm;
+	desc.flags = 0;
 
 	memset(auth_data, 0, ahp->icv_trunc_len);
-	crypto_hmac_init(tfm, ahp->key, &ahp->key_len);
-	skb_icv_walk(skb, tfm, 0, skb->len, crypto_hmac_update);
-	crypto_hmac_final(tfm, ahp->key, &ahp->key_len, ahp->work_icv);
-	memcpy(auth_data, ahp->work_icv, ahp->icv_trunc_len);
+	err = crypto_hash_init(&desc);
+	if (unlikely(err))
+		goto out;
+	err = skb_icv_walk(skb, &desc, 0, skb->len, crypto_hash_update);
+	if (unlikely(err))
+		goto out;
+	err = crypto_hash_final(&desc, ahp->work_icv);
+
+out:
+	return err;
 }
 
 #endif
diff --git a/include/net/esp.h b/include/net/esp.h
index af2ff18700c7..064366d66eea 100644
--- a/include/net/esp.h
+++ b/include/net/esp.h
@@ -35,7 +35,7 @@ struct esp_data
 		void			(*icv)(struct esp_data*,
 		                               struct sk_buff *skb,
 		                               int offset, int len, u8 *icv);
-		struct crypto_tfm	*tfm;
+		struct crypto_hash	*tfm;
 	} auth;
 };
 
@@ -43,18 +43,22 @@ extern int skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset,
 extern int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer);
 extern void *pskb_put(struct sk_buff *skb, struct sk_buff *tail, int len);
 
-static inline void
-esp_hmac_digest(struct esp_data *esp, struct sk_buff *skb, int offset,
-                int len, u8 *auth_data)
+static inline int esp_mac_digest(struct esp_data *esp, struct sk_buff *skb,
+				 int offset, int len)
 {
-	struct crypto_tfm *tfm = esp->auth.tfm;
-	char *icv = esp->auth.work_icv;
-
-	memset(auth_data, 0, esp->auth.icv_trunc_len);
-	crypto_hmac_init(tfm, esp->auth.key, &esp->auth.key_len);
-	skb_icv_walk(skb, tfm, offset, len, crypto_hmac_update);
-	crypto_hmac_final(tfm, esp->auth.key, &esp->auth.key_len, icv);
-	memcpy(auth_data, icv, esp->auth.icv_trunc_len);
+	struct hash_desc desc;
+	int err;
+
+	desc.tfm = esp->auth.tfm;
+	desc.flags = 0;
+
+	err = crypto_hash_init(&desc);
+	if (unlikely(err))
+		return err;
+	err = skb_icv_walk(skb, &desc, offset, len, crypto_hash_update);
+	if (unlikely(err))
+		return err;
+	return crypto_hash_final(&desc, esp->auth.work_icv);
 }
 
 #endif
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index e9114e41affc..3ecd9fa1ed4b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -984,12 +984,13 @@ extern struct xfrm_algo_desc *xfrm_aalg_get_byname(char *name, int probe);
 extern struct xfrm_algo_desc *xfrm_ealg_get_byname(char *name, int probe);
 extern struct xfrm_algo_desc *xfrm_calg_get_byname(char *name, int probe);
 
-struct crypto_tfm;
+struct hash_desc;
 struct scatterlist;
-typedef void (icv_update_fn_t)(struct crypto_tfm *, struct scatterlist *, unsigned int);
+typedef int (icv_update_fn_t)(struct hash_desc *, struct scatterlist *,
+			      unsigned int);
 
-extern void skb_icv_walk(const struct sk_buff *skb, struct crypto_tfm *tfm,
-			 int offset, int len, icv_update_fn_t icv_update);
+extern int skb_icv_walk(const struct sk_buff *skb, struct hash_desc *tfm,
+			int offset, int len, icv_update_fn_t icv_update);
 
 static inline int xfrm_addr_cmp(xfrm_address_t *a, xfrm_address_t *b,
 				int family)
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index 1366bc6ce6a5..2b98943e6b02 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -1,3 +1,4 @@
+#include <linux/err.h>
 #include <linux/module.h>
 #include <net/ip.h>
 #include <net/xfrm.h>
@@ -97,7 +98,10 @@ static int ah_output(struct xfrm_state *x, struct sk_buff *skb)
 	ah->spi = x->id.spi;
 	ah->seq_no = htonl(++x->replay.oseq);
 	xfrm_aevent_doreplay(x);
-	ahp->icv(ahp, skb, ah->auth_data);
+	err = ah_mac_digest(ahp, skb, ah->auth_data);
+	if (err)
+		goto error;
+	memcpy(ah->auth_data, ahp->work_icv, ahp->icv_trunc_len);
 
 	top_iph->tos = iph->tos;
 	top_iph->ttl = iph->ttl;
@@ -119,6 +123,7 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb)
 {
 	int ah_hlen;
 	int ihl;
+	int err = -EINVAL;
 	struct iphdr *iph;
 	struct ip_auth_hdr *ah;
 	struct ah_data *ahp;
@@ -166,8 +171,11 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb)
 		
 		memcpy(auth_data, ah->auth_data, ahp->icv_trunc_len);
 		skb_push(skb, ihl);
-		ahp->icv(ahp, skb, ah->auth_data);
-		if (memcmp(ah->auth_data, auth_data, ahp->icv_trunc_len)) {
+		err = ah_mac_digest(ahp, skb, ah->auth_data);
+		if (err)
+			goto out;
+		err = -EINVAL;
+		if (memcmp(ahp->work_icv, auth_data, ahp->icv_trunc_len)) {
 			x->stats.integrity_failed++;
 			goto out;
 		}
@@ -179,7 +187,7 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb)
 	return 0;
 
 out:
-	return -EINVAL;
+	return err;
 }
 
 static void ah4_err(struct sk_buff *skb, u32 info)
@@ -204,6 +212,7 @@ static int ah_init_state(struct xfrm_state *x)
 {
 	struct ah_data *ahp = NULL;
 	struct xfrm_algo_desc *aalg_desc;
+	struct crypto_hash *tfm;
 
 	if (!x->aalg)
 		goto error;
@@ -221,24 +230,27 @@ static int ah_init_state(struct xfrm_state *x)
 
 	ahp->key = x->aalg->alg_key;
 	ahp->key_len = (x->aalg->alg_key_len+7)/8;
-	ahp->tfm = crypto_alloc_tfm(x->aalg->alg_name, 0);
-	if (!ahp->tfm)
+	tfm = crypto_alloc_hash(x->aalg->alg_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(tfm))
+		goto error;
+
+	ahp->tfm = tfm;
+	if (crypto_hash_setkey(tfm, ahp->key, ahp->key_len))
 		goto error;
-	ahp->icv = ah_hmac_digest;
 	
 	/*
 	 * Lookup the algorithm description maintained by xfrm_algo,
 	 * verify crypto transform properties, and store information
 	 * we need for AH processing.  This lookup cannot fail here
-	 * after a successful crypto_alloc_tfm().
+	 * after a successful crypto_alloc_hash().
 	 */
 	aalg_desc = xfrm_aalg_get_byname(x->aalg->alg_name, 0);
 	BUG_ON(!aalg_desc);
 
 	if (aalg_desc->uinfo.auth.icv_fullbits/8 !=
-	    crypto_tfm_alg_digestsize(ahp->tfm)) {
+	    crypto_hash_digestsize(tfm)) {
 		printk(KERN_INFO "AH: %s digestsize %u != %hu\n",
-		       x->aalg->alg_name, crypto_tfm_alg_digestsize(ahp->tfm),
+		       x->aalg->alg_name, crypto_hash_digestsize(tfm),
 		       aalg_desc->uinfo.auth.icv_fullbits/8);
 		goto error;
 	}
@@ -262,7 +274,7 @@ static int ah_init_state(struct xfrm_state *x)
 error:
 	if (ahp) {
 		kfree(ahp->work_icv);
-		crypto_free_tfm(ahp->tfm);
+		crypto_free_hash(ahp->tfm);
 		kfree(ahp);
 	}
 	return -EINVAL;
@@ -277,7 +289,7 @@ static void ah_destroy(struct xfrm_state *x)
 
 	kfree(ahp->work_icv);
 	ahp->work_icv = NULL;
-	crypto_free_tfm(ahp->tfm);
+	crypto_free_hash(ahp->tfm);
 	ahp->tfm = NULL;
 	kfree(ahp);
 }
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 7c63ae494742..b428489f6ccd 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -121,9 +121,9 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 	}
 
 	if (esp->auth.icv_full_len) {
-		esp->auth.icv(esp, skb, (u8*)esph-skb->data,
-		              sizeof(struct ip_esp_hdr) + esp->conf.ivlen+clen, trailer->tail);
-		pskb_put(skb, trailer, alen);
+		err = esp_mac_digest(esp, skb, (u8 *)esph - skb->data,
+				     sizeof(*esph) + esp->conf.ivlen + clen);
+		memcpy(pskb_put(skb, trailer, alen), esp->auth.work_icv, alen);
 	}
 
 	ip_send_check(top_iph);
@@ -163,15 +163,16 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 
 	/* If integrity check is required, do this. */
 	if (esp->auth.icv_full_len) {
-		u8 sum[esp->auth.icv_full_len];
-		u8 sum1[alen];
-		
-		esp->auth.icv(esp, skb, 0, skb->len-alen, sum);
+		u8 sum[alen];
 
-		if (skb_copy_bits(skb, skb->len-alen, sum1, alen))
+		err = esp_mac_digest(esp, skb, 0, skb->len - alen);
+		if (err)
+			goto out;
+
+		if (skb_copy_bits(skb, skb->len - alen, sum, alen))
 			BUG();
 
-		if (unlikely(memcmp(sum, sum1, alen))) {
+		if (unlikely(memcmp(esp->auth.work_icv, sum, alen))) {
 			x->stats.integrity_failed++;
 			goto out;
 		}
@@ -307,7 +308,7 @@ static void esp_destroy(struct xfrm_state *x)
 	esp->conf.tfm = NULL;
 	kfree(esp->conf.ivec);
 	esp->conf.ivec = NULL;
-	crypto_free_tfm(esp->auth.tfm);
+	crypto_free_hash(esp->auth.tfm);
 	esp->auth.tfm = NULL;
 	kfree(esp->auth.work_icv);
 	esp->auth.work_icv = NULL;
@@ -333,22 +334,27 @@ static int esp_init_state(struct xfrm_state *x)
 
 	if (x->aalg) {
 		struct xfrm_algo_desc *aalg_desc;
+		struct crypto_hash *hash;
 
 		esp->auth.key = x->aalg->alg_key;
 		esp->auth.key_len = (x->aalg->alg_key_len+7)/8;
-		esp->auth.tfm = crypto_alloc_tfm(x->aalg->alg_name, 0);
-		if (esp->auth.tfm == NULL)
+		hash = crypto_alloc_hash(x->aalg->alg_name, 0,
+					 CRYPTO_ALG_ASYNC);
+		if (IS_ERR(hash))
+			goto error;
+
+		esp->auth.tfm = hash;
+		if (crypto_hash_setkey(hash, esp->auth.key, esp->auth.key_len))
 			goto error;
-		esp->auth.icv = esp_hmac_digest;
 
 		aalg_desc = xfrm_aalg_get_byname(x->aalg->alg_name, 0);
 		BUG_ON(!aalg_desc);
 
 		if (aalg_desc->uinfo.auth.icv_fullbits/8 !=
-		    crypto_tfm_alg_digestsize(esp->auth.tfm)) {
+		    crypto_hash_digestsize(hash)) {
 			NETDEBUG(KERN_INFO "ESP: %s digestsize %u != %hu\n",
 				 x->aalg->alg_name,
-				 crypto_tfm_alg_digestsize(esp->auth.tfm),
+				 crypto_hash_digestsize(hash),
 				 aalg_desc->uinfo.auth.icv_fullbits/8);
 			goto error;
 		}
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 9d4831bd4335..00ffa7bc6c9f 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -213,7 +213,10 @@ static int ah6_output(struct xfrm_state *x, struct sk_buff *skb)
 	ah->spi = x->id.spi;
 	ah->seq_no = htonl(++x->replay.oseq);
 	xfrm_aevent_doreplay(x);
-	ahp->icv(ahp, skb, ah->auth_data);
+	err = ah_mac_digest(ahp, skb, ah->auth_data);
+	if (err)
+		goto error_free_iph;
+	memcpy(ah->auth_data, ahp->work_icv, ahp->icv_trunc_len);
 
 	err = 0;
 
@@ -251,6 +254,7 @@ static int ah6_input(struct xfrm_state *x, struct sk_buff *skb)
 	u16 hdr_len;
 	u16 ah_hlen;
 	int nexthdr;
+	int err = -EINVAL;
 
 	if (!pskb_may_pull(skb, sizeof(struct ip_auth_hdr)))
 		goto out;
@@ -292,8 +296,11 @@ static int ah6_input(struct xfrm_state *x, struct sk_buff *skb)
 		memcpy(auth_data, ah->auth_data, ahp->icv_trunc_len);
 		memset(ah->auth_data, 0, ahp->icv_trunc_len);
 		skb_push(skb, hdr_len);
-		ahp->icv(ahp, skb, ah->auth_data);
-		if (memcmp(ah->auth_data, auth_data, ahp->icv_trunc_len)) {
+		err = ah_mac_digest(ahp, skb, ah->auth_data);
+		if (err)
+			goto free_out;
+		err = -EINVAL;
+		if (memcmp(ahp->work_icv, auth_data, ahp->icv_trunc_len)) {
 			LIMIT_NETDEBUG(KERN_WARNING "ipsec ah authentication error\n");
 			x->stats.integrity_failed++;
 			goto free_out;
@@ -310,7 +317,7 @@ static int ah6_input(struct xfrm_state *x, struct sk_buff *skb)
 free_out:
 	kfree(tmp_hdr);
 out:
-	return -EINVAL;
+	return err;
 }
 
 static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, 
@@ -338,6 +345,7 @@ static int ah6_init_state(struct xfrm_state *x)
 {
 	struct ah_data *ahp = NULL;
 	struct xfrm_algo_desc *aalg_desc;
+	struct crypto_hash *tfm;
 
 	if (!x->aalg)
 		goto error;
@@ -355,24 +363,27 @@ static int ah6_init_state(struct xfrm_state *x)
 
 	ahp->key = x->aalg->alg_key;
 	ahp->key_len = (x->aalg->alg_key_len+7)/8;
-	ahp->tfm = crypto_alloc_tfm(x->aalg->alg_name, 0);
-	if (!ahp->tfm)
+	tfm = crypto_alloc_hash(x->aalg->alg_name, 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(tfm))
+		goto error;
+
+	ahp->tfm = tfm;
+	if (crypto_hash_setkey(tfm, ahp->key, ahp->key_len))
 		goto error;
-	ahp->icv = ah_hmac_digest;
 	
 	/*
 	 * Lookup the algorithm description maintained by xfrm_algo,
 	 * verify crypto transform properties, and store information
 	 * we need for AH processing.  This lookup cannot fail here
-	 * after a successful crypto_alloc_tfm().
+	 * after a successful crypto_alloc_hash().
 	 */
 	aalg_desc = xfrm_aalg_get_byname(x->aalg->alg_name, 0);
 	BUG_ON(!aalg_desc);
 
 	if (aalg_desc->uinfo.auth.icv_fullbits/8 !=
-	    crypto_tfm_alg_digestsize(ahp->tfm)) {
+	    crypto_hash_digestsize(tfm)) {
 		printk(KERN_INFO "AH: %s digestsize %u != %hu\n",
-		       x->aalg->alg_name, crypto_tfm_alg_digestsize(ahp->tfm),
+		       x->aalg->alg_name, crypto_hash_digestsize(tfm),
 		       aalg_desc->uinfo.auth.icv_fullbits/8);
 		goto error;
 	}
@@ -396,7 +407,7 @@ static int ah6_init_state(struct xfrm_state *x)
 error:
 	if (ahp) {
 		kfree(ahp->work_icv);
-		crypto_free_tfm(ahp->tfm);
+		crypto_free_hash(ahp->tfm);
 		kfree(ahp);
 	}
 	return -EINVAL;
@@ -411,7 +422,7 @@ static void ah6_destroy(struct xfrm_state *x)
 
 	kfree(ahp->work_icv);
 	ahp->work_icv = NULL;
-	crypto_free_tfm(ahp->tfm);
+	crypto_free_hash(ahp->tfm);
 	ahp->tfm = NULL;
 	kfree(ahp);
 }
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 46a7e687948e..2ebfd281e721 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -125,9 +125,9 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 	}
 
 	if (esp->auth.icv_full_len) {
-		esp->auth.icv(esp, skb, (u8*)esph-skb->data,
-			sizeof(struct ipv6_esp_hdr) + esp->conf.ivlen+clen, trailer->tail);
-		pskb_put(skb, trailer, alen);
+		err = esp_mac_digest(esp, skb, (u8 *)esph - skb->data,
+				     sizeof(*esph) + esp->conf.ivlen + clen);
+		memcpy(pskb_put(skb, trailer, alen), esp->auth.work_icv, alen);
 	}
 
 error:
@@ -162,15 +162,16 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb)
 
 	/* If integrity check is required, do this. */
         if (esp->auth.icv_full_len) {
-		u8 sum[esp->auth.icv_full_len];
-		u8 sum1[alen];
+		u8 sum[alen];
 
-		esp->auth.icv(esp, skb, 0, skb->len-alen, sum);
+		ret = esp_mac_digest(esp, skb, 0, skb->len - alen);
+		if (ret)
+			goto out;
 
-		if (skb_copy_bits(skb, skb->len-alen, sum1, alen))
+		if (skb_copy_bits(skb, skb->len - alen, sum, alen))
 			BUG();
 
-		if (unlikely(memcmp(sum, sum1, alen))) {
+		if (unlikely(memcmp(esp->auth.work_icv, sum, alen))) {
 			x->stats.integrity_failed++;
 			ret = -EINVAL;
 			goto out;
@@ -279,7 +280,7 @@ static void esp6_destroy(struct xfrm_state *x)
 	esp->conf.tfm = NULL;
 	kfree(esp->conf.ivec);
 	esp->conf.ivec = NULL;
-	crypto_free_tfm(esp->auth.tfm);
+	crypto_free_hash(esp->auth.tfm);
 	esp->auth.tfm = NULL;
 	kfree(esp->auth.work_icv);
 	esp->auth.work_icv = NULL;
@@ -308,24 +309,29 @@ static int esp6_init_state(struct xfrm_state *x)
 
 	if (x->aalg) {
 		struct xfrm_algo_desc *aalg_desc;
+		struct crypto_hash *hash;
 
 		esp->auth.key = x->aalg->alg_key;
 		esp->auth.key_len = (x->aalg->alg_key_len+7)/8;
-		esp->auth.tfm = crypto_alloc_tfm(x->aalg->alg_name, 0);
-		if (esp->auth.tfm == NULL)
+		hash = crypto_alloc_hash(x->aalg->alg_name, 0,
+					 CRYPTO_ALG_ASYNC);
+		if (IS_ERR(hash))
+			goto error;
+
+		esp->auth.tfm = hash;
+		if (crypto_hash_setkey(hash, esp->auth.key, esp->auth.key_len))
 			goto error;
-		esp->auth.icv = esp_hmac_digest;
  
 		aalg_desc = xfrm_aalg_get_byname(x->aalg->alg_name, 0);
 		BUG_ON(!aalg_desc);
  
 		if (aalg_desc->uinfo.auth.icv_fullbits/8 !=
-			crypto_tfm_alg_digestsize(esp->auth.tfm)) {
-				printk(KERN_INFO "ESP: %s digestsize %u != %hu\n",
-					x->aalg->alg_name,
-					crypto_tfm_alg_digestsize(esp->auth.tfm),
-					aalg_desc->uinfo.auth.icv_fullbits/8);
-				goto error;
+		    crypto_hash_digestsize(hash)) {
+			NETDEBUG(KERN_INFO "ESP: %s digestsize %u != %hu\n",
+				 x->aalg->alg_name,
+				 crypto_hash_digestsize(hash),
+				 aalg_desc->uinfo.auth.icv_fullbits/8);
+			goto error;
 		}
  
 		esp->auth.icv_full_len = aalg_desc->uinfo.auth.icv_fullbits/8;
diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index 9b03d8497fba..87918f281bb4 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -30,7 +30,8 @@
  */
 static struct xfrm_algo_desc aalg_list[] = {
 {
-	.name = "digest_null",
+	.name = "hmac(digest_null)",
+	.compat = "digest_null",
 	
 	.uinfo = {
 		.auth = {
@@ -47,7 +48,8 @@ static struct xfrm_algo_desc aalg_list[] = {
 	}
 },
 {
-	.name = "md5",
+	.name = "hmac(md5)",
+	.compat = "md5",
 
 	.uinfo = {
 		.auth = {
@@ -64,7 +66,8 @@ static struct xfrm_algo_desc aalg_list[] = {
 	}
 },
 {
-	.name = "sha1",
+	.name = "hmac(sha1)",
+	.compat = "sha1",
 
 	.uinfo = {
 		.auth = {
@@ -81,7 +84,8 @@ static struct xfrm_algo_desc aalg_list[] = {
 	}
 },
 {
-	.name = "sha256",
+	.name = "hmac(sha256)",
+	.compat = "sha256",
 
 	.uinfo = {
 		.auth = {
@@ -98,7 +102,8 @@ static struct xfrm_algo_desc aalg_list[] = {
 	}
 },
 {
-	.name = "ripemd160",
+	.name = "hmac(ripemd160)",
+	.compat = "ripemd160",
 
 	.uinfo = {
 		.auth = {
@@ -480,11 +485,12 @@ EXPORT_SYMBOL_GPL(xfrm_count_enc_supported);
 
 /* Move to common area: it is shared with AH. */
 
-void skb_icv_walk(const struct sk_buff *skb, struct crypto_tfm *tfm,
-		  int offset, int len, icv_update_fn_t icv_update)
+int skb_icv_walk(const struct sk_buff *skb, struct hash_desc *desc,
+		 int offset, int len, icv_update_fn_t icv_update)
 {
 	int start = skb_headlen(skb);
 	int i, copy = start - offset;
+	int err;
 	struct scatterlist sg;
 
 	/* Checksum header. */
@@ -496,10 +502,12 @@ void skb_icv_walk(const struct sk_buff *skb, struct crypto_tfm *tfm,
 		sg.offset = (unsigned long)(skb->data + offset) % PAGE_SIZE;
 		sg.length = copy;
 		
-		icv_update(tfm, &sg, 1);
+		err = icv_update(desc, &sg, copy);
+		if (unlikely(err))
+			return err;
 		
 		if ((len -= copy) == 0)
-			return;
+			return 0;
 		offset += copy;
 	}
 
@@ -519,10 +527,12 @@ void skb_icv_walk(const struct sk_buff *skb, struct crypto_tfm *tfm,
 			sg.offset = frag->page_offset + offset-start;
 			sg.length = copy;
 			
-			icv_update(tfm, &sg, 1);
+			err = icv_update(desc, &sg, copy);
+			if (unlikely(err))
+				return err;
 
 			if (!(len -= copy))
-				return;
+				return 0;
 			offset += copy;
 		}
 		start = end;
@@ -540,15 +550,19 @@ void skb_icv_walk(const struct sk_buff *skb, struct crypto_tfm *tfm,
 			if ((copy = end - offset) > 0) {
 				if (copy > len)
 					copy = len;
-				skb_icv_walk(list, tfm, offset-start, copy, icv_update);
+				err = skb_icv_walk(list, desc, offset-start,
+						   copy, icv_update);
+				if (unlikely(err))
+					return err;
 				if ((len -= copy) == 0)
-					return;
+					return 0;
 				offset += copy;
 			}
 			start = end;
 		}
 	}
 	BUG_ON(len);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(skb_icv_walk);
 
-- 
cgit v1.2.3


From 1b489e11d4df82514792f9f981f31976f8a94ddf Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 20 Aug 2006 15:07:14 +1000
Subject: [SCTP]: Use HMAC template and hash interface

This patch converts SCTP to use the new HMAC template and hash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/constants.h |  4 ++--
 include/net/sctp/sctp.h      | 11 -----------
 include/net/sctp/structs.h   |  3 ++-
 net/sctp/endpointola.c       |  2 +-
 net/sctp/sm_make_chunk.c     | 37 +++++++++++++++++++++++++++----------
 net/sctp/socket.c            |  6 +++---
 6 files changed, 35 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h
index c51541ee0247..57166bfdf8eb 100644
--- a/include/net/sctp/constants.h
+++ b/include/net/sctp/constants.h
@@ -312,9 +312,9 @@ enum { SCTP_MAX_GABS = 16 };
 				 */
 
 #if defined (CONFIG_SCTP_HMAC_MD5)
-#define SCTP_COOKIE_HMAC_ALG "md5"
+#define SCTP_COOKIE_HMAC_ALG "hmac(md5)"
 #elif defined (CONFIG_SCTP_HMAC_SHA1)
-#define SCTP_COOKIE_HMAC_ALG "sha1"
+#define SCTP_COOKIE_HMAC_ALG "hmac(sha1)"
 #else
 #define SCTP_COOKIE_HMAC_ALG NULL
 #endif
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 92eae0e0f3f1..1c1abce5f6b6 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -330,17 +330,6 @@ static inline void sctp_v6_exit(void) { return; }
 
 #endif /* #if defined(CONFIG_IPV6) */
 
-/* Some wrappers, in case crypto not available. */
-#if defined (CONFIG_CRYPTO_HMAC)
-#define sctp_crypto_alloc_tfm crypto_alloc_tfm
-#define sctp_crypto_free_tfm crypto_free_tfm
-#define sctp_crypto_hmac crypto_hmac
-#else
-#define sctp_crypto_alloc_tfm(x...) NULL
-#define sctp_crypto_free_tfm(x...)
-#define sctp_crypto_hmac(x...)
-#endif
-
 
 /* Map an association to an assoc_id. */
 static inline sctp_assoc_t sctp_assoc2id(const struct sctp_association *asoc)
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index e5aa7ff1f5b5..0412e730c765 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -87,6 +87,7 @@ struct sctp_bind_addr;
 struct sctp_ulpq;
 struct sctp_ep_common;
 struct sctp_ssnmap;
+struct crypto_hash;
 
 
 #include <net/sctp/tsnmap.h>
@@ -264,7 +265,7 @@ struct sctp_sock {
 	struct sctp_pf *pf;
 
 	/* Access to HMAC transform. */
-	struct crypto_tfm *hmac;
+	struct crypto_hash *hmac;
 
 	/* What is our base endpointer? */
 	struct sctp_endpoint *ep;
diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c
index ffda1d680529..35c49ff2d062 100644
--- a/net/sctp/endpointola.c
+++ b/net/sctp/endpointola.c
@@ -173,7 +173,7 @@ static void sctp_endpoint_destroy(struct sctp_endpoint *ep)
 	SCTP_ASSERT(ep->base.dead, "Endpoint is not dead", return);
 
 	/* Free up the HMAC transform. */
-	sctp_crypto_free_tfm(sctp_sk(ep->base.sk)->hmac);
+	crypto_free_hash(sctp_sk(ep->base.sk)->hmac);
 
 	/* Cleanup. */
 	sctp_inq_free(&ep->base.inqueue);
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 17b509282cf2..7745bdea7817 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1282,10 +1282,8 @@ static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep,
 
 	retval = kmalloc(*cookie_len, GFP_ATOMIC);
 
-	if (!retval) {
-		*cookie_len = 0;
+	if (!retval)
 		goto nodata;
-	}
 
 	/* Clear this memory since we are sending this data structure
 	 * out on the network.
@@ -1321,19 +1319,29 @@ static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep,
 	       ntohs(init_chunk->chunk_hdr->length), raw_addrs, addrs_len);
 
   	if (sctp_sk(ep->base.sk)->hmac) {
+		struct hash_desc desc;
+
 		/* Sign the message.  */
 		sg.page = virt_to_page(&cookie->c);
 		sg.offset = (unsigned long)(&cookie->c) % PAGE_SIZE;
 		sg.length = bodysize;
 		keylen = SCTP_SECRET_SIZE;
 		key = (char *)ep->secret_key[ep->current_key];
+  		desc.tfm = sctp_sk(ep->base.sk)->hmac;
+  		desc.flags = 0;
 
-		sctp_crypto_hmac(sctp_sk(ep->base.sk)->hmac, key, &keylen,
-				 &sg, 1, cookie->signature);
+		if (crypto_hash_setkey(desc.tfm, key, keylen) ||
+		    crypto_hash_digest(&desc, &sg, bodysize, cookie->signature))
+			goto free_cookie;
 	}
 
-nodata:
 	return retval;
+
+free_cookie:
+	kfree(retval);
+nodata:
+	*cookie_len = 0;
+	return NULL;
 }
 
 /* Unpack the cookie from COOKIE ECHO chunk, recreating the association.  */
@@ -1354,6 +1362,7 @@ struct sctp_association *sctp_unpack_cookie(
 	sctp_scope_t scope;
 	struct sk_buff *skb = chunk->skb;
 	struct timeval tv;
+	struct hash_desc desc;
 
 	/* Header size is static data prior to the actual cookie, including
 	 * any padding.
@@ -1389,17 +1398,25 @@ struct sctp_association *sctp_unpack_cookie(
 	sg.offset = (unsigned long)(bear_cookie) % PAGE_SIZE;
 	sg.length = bodysize;
 	key = (char *)ep->secret_key[ep->current_key];
+	desc.tfm = sctp_sk(ep->base.sk)->hmac;
+	desc.flags = 0;
 
 	memset(digest, 0x00, SCTP_SIGNATURE_SIZE);
-	sctp_crypto_hmac(sctp_sk(ep->base.sk)->hmac, key, &keylen, &sg,
-			 1, digest);
+	if (crypto_hash_setkey(desc.tfm, key, keylen) ||
+	    crypto_hash_digest(&desc, &sg, bodysize, digest)) {
+		*error = -SCTP_IERROR_NOMEM;
+		goto fail;
+	}
 
 	if (memcmp(digest, cookie->signature, SCTP_SIGNATURE_SIZE)) {
 		/* Try the previous key. */
 		key = (char *)ep->secret_key[ep->last_key];
 		memset(digest, 0x00, SCTP_SIGNATURE_SIZE);
-		sctp_crypto_hmac(sctp_sk(ep->base.sk)->hmac, key, &keylen,
-				 &sg, 1, digest);
+		if (crypto_hash_setkey(desc.tfm, key, keylen) ||
+		    crypto_hash_digest(&desc, &sg, bodysize, digest)) {
+			*error = -SCTP_IERROR_NOMEM;
+			goto fail;
+		}
 
 		if (memcmp(digest, cookie->signature, SCTP_SIGNATURE_SIZE)) {
 			/* Yikes!  Still bad signature! */
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index dab15949958e..85caf7963886 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -4898,7 +4898,7 @@ SCTP_STATIC int sctp_stream_listen(struct sock *sk, int backlog)
 int sctp_inet_listen(struct socket *sock, int backlog)
 {
 	struct sock *sk = sock->sk;
-	struct crypto_tfm *tfm=NULL;
+	struct crypto_hash *tfm = NULL;
 	int err = -EINVAL;
 
 	if (unlikely(backlog < 0))
@@ -4911,7 +4911,7 @@ int sctp_inet_listen(struct socket *sock, int backlog)
 
 	/* Allocate HMAC for generating cookie. */
 	if (sctp_hmac_alg) {
-		tfm = sctp_crypto_alloc_tfm(sctp_hmac_alg, 0);
+		tfm = crypto_alloc_hash(sctp_hmac_alg, 0, CRYPTO_ALG_ASYNC);
 		if (!tfm) {
 			err = -ENOSYS;
 			goto out;
@@ -4937,7 +4937,7 @@ out:
 	sctp_release_sock(sk);
 	return err;
 cleanup:
-	sctp_crypto_free_tfm(tfm);
+	crypto_free_hash(tfm);
 	goto out;
 }
 
-- 
cgit v1.2.3


From 8425165dfed27945e8509c141cea245d1739e372 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 20 Aug 2006 15:25:22 +1000
Subject: [CRYPTO] digest: Remove old HMAC implementation

This patch removes the old HMAC implementation now that nobody uses it
anymore.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/Kconfig         |   2 +-
 crypto/digest.c        |   3 +-
 crypto/hmac.c          | 101 -------------------------------------------------
 crypto/internal.h      |  13 -------
 include/linux/crypto.h |  16 --------
 5 files changed, 2 insertions(+), 133 deletions(-)

(limited to 'include')

diff --git a/crypto/Kconfig b/crypto/Kconfig
index f07d9237950f..1e2f39c21180 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -33,7 +33,7 @@ config CRYPTO_MANAGER
 	  cbc(aes).
 
 config CRYPTO_HMAC
-	bool "HMAC support"
+	tristate "HMAC support"
 	select CRYPTO_HASH
 	help
 	  HMAC: Keyed-Hashing for Message Authentication (RFC2104).
diff --git a/crypto/digest.c b/crypto/digest.c
index 5873063db840..0155a94e4b15 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -191,10 +191,9 @@ int crypto_init_digest_ops(struct crypto_tfm *tfm)
 	ops->setkey	= dalg->dia_setkey ? setkey : nosetkey;
 	ops->digestsize	= dalg->dia_digestsize;
 	
-	return crypto_alloc_hmac_block(tfm);
+	return 0;
 }
 
 void crypto_exit_digest_ops(struct crypto_tfm *tfm)
 {
-	crypto_free_hmac_block(tfm);
 }
diff --git a/crypto/hmac.c b/crypto/hmac.c
index eac77e294740..f403b6946047 100644
--- a/crypto/hmac.c
+++ b/crypto/hmac.c
@@ -29,107 +29,6 @@ struct hmac_ctx {
 	struct crypto_hash *child;
 };
 
-static void hash_key(struct crypto_tfm *tfm, u8 *key, unsigned int keylen)
-{
-	struct scatterlist tmp;
-	
-	sg_set_buf(&tmp, key, keylen);
-	crypto_digest_digest(tfm, &tmp, 1, key);
-}
-
-int crypto_alloc_hmac_block(struct crypto_tfm *tfm)
-{
-	int ret = 0;
-
-	BUG_ON(!crypto_tfm_alg_blocksize(tfm));
-	
-	tfm->crt_hash.hmac_block = kmalloc(crypto_tfm_alg_blocksize(tfm),
-					   GFP_KERNEL);
-	if (tfm->crt_hash.hmac_block == NULL)
-		ret = -ENOMEM;
-
-	return ret;
-		
-}
-
-void crypto_free_hmac_block(struct crypto_tfm *tfm)
-{
-	kfree(tfm->crt_hash.hmac_block);
-}
-
-void crypto_hmac_init(struct crypto_tfm *tfm, u8 *key, unsigned int *keylen)
-{
-	unsigned int i;
-	struct scatterlist tmp;
-	char *ipad = tfm->crt_hash.hmac_block;
-	
-	if (*keylen > crypto_tfm_alg_blocksize(tfm)) {
-		hash_key(tfm, key, *keylen);
-		*keylen = crypto_tfm_alg_digestsize(tfm);
-	}
-
-	memset(ipad, 0, crypto_tfm_alg_blocksize(tfm));
-	memcpy(ipad, key, *keylen);
-
-	for (i = 0; i < crypto_tfm_alg_blocksize(tfm); i++)
-		ipad[i] ^= 0x36;
-
-	sg_set_buf(&tmp, ipad, crypto_tfm_alg_blocksize(tfm));
-	
-	crypto_digest_init(tfm);
-	crypto_digest_update(tfm, &tmp, 1);
-}
-
-void crypto_hmac_update(struct crypto_tfm *tfm,
-                        struct scatterlist *sg, unsigned int nsg)
-{
-	crypto_digest_update(tfm, sg, nsg);
-}
-
-void crypto_hmac_final(struct crypto_tfm *tfm, u8 *key,
-                       unsigned int *keylen, u8 *out)
-{
-	unsigned int i;
-	struct scatterlist tmp;
-	char *opad = tfm->crt_hash.hmac_block;
-	
-	if (*keylen > crypto_tfm_alg_blocksize(tfm)) {
-		hash_key(tfm, key, *keylen);
-		*keylen = crypto_tfm_alg_digestsize(tfm);
-	}
-
-	crypto_digest_final(tfm, out);
-
-	memset(opad, 0, crypto_tfm_alg_blocksize(tfm));
-	memcpy(opad, key, *keylen);
-		
-	for (i = 0; i < crypto_tfm_alg_blocksize(tfm); i++)
-		opad[i] ^= 0x5c;
-
-	sg_set_buf(&tmp, opad, crypto_tfm_alg_blocksize(tfm));
-
-	crypto_digest_init(tfm);
-	crypto_digest_update(tfm, &tmp, 1);
-	
-	sg_set_buf(&tmp, out, crypto_tfm_alg_digestsize(tfm));
-	
-	crypto_digest_update(tfm, &tmp, 1);
-	crypto_digest_final(tfm, out);
-}
-
-void crypto_hmac(struct crypto_tfm *tfm, u8 *key, unsigned int *keylen,
-                 struct scatterlist *sg, unsigned int nsg, u8 *out)
-{
-	crypto_hmac_init(tfm, key, keylen);
-	crypto_hmac_update(tfm, sg, nsg);
-	crypto_hmac_final(tfm, key, keylen, out);
-}
-
-EXPORT_SYMBOL_GPL(crypto_hmac_init);
-EXPORT_SYMBOL_GPL(crypto_hmac_update);
-EXPORT_SYMBOL_GPL(crypto_hmac_final);
-EXPORT_SYMBOL_GPL(crypto_hmac);
-
 static inline void *align_ptr(void *p, unsigned int align)
 {
 	return (void *)ALIGN((unsigned long)p, align);
diff --git a/crypto/internal.h b/crypto/internal.h
index 93d9b10ff914..2da6ad4f3593 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -73,19 +73,6 @@ static inline void crypto_yield(u32 flags)
 		cond_resched();
 }
 
-#ifdef CONFIG_CRYPTO_HMAC
-int crypto_alloc_hmac_block(struct crypto_tfm *tfm);
-void crypto_free_hmac_block(struct crypto_tfm *tfm);
-#else
-static inline int crypto_alloc_hmac_block(struct crypto_tfm *tfm)
-{
-	return 0;
-}
-
-static inline void crypto_free_hmac_block(struct crypto_tfm *tfm)
-{ }
-#endif
-
 #ifdef CONFIG_PROC_FS
 void __init crypto_init_proc(void);
 void __exit crypto_exit_proc(void);
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 40c0aab8ad4c..929fb9ad1314 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -295,9 +295,6 @@ struct hash_tfm {
 		      unsigned int nsg, u8 *out);
 	int (*setkey)(struct crypto_hash *tfm, const u8 *key,
 		      unsigned int keylen);
-#ifdef CONFIG_CRYPTO_HMAC
-	void *hmac_block;
-#endif
 	unsigned int digestsize;
 };
 
@@ -872,18 +869,5 @@ static inline int crypto_comp_decompress(struct crypto_tfm *tfm,
 	return tfm->crt_compress.cot_decompress(tfm, src, slen, dst, dlen);
 }
 
-/*
- * HMAC support.
- */
-#ifdef CONFIG_CRYPTO_HMAC
-void crypto_hmac_init(struct crypto_tfm *tfm, u8 *key, unsigned int *keylen);
-void crypto_hmac_update(struct crypto_tfm *tfm,
-                        struct scatterlist *sg, unsigned int nsg);
-void crypto_hmac_final(struct crypto_tfm *tfm, u8 *key,
-                       unsigned int *keylen, u8 *out);
-void crypto_hmac(struct crypto_tfm *tfm, u8 *key, unsigned int *keylen,
-                 struct scatterlist *sg, unsigned int nsg, u8 *out);
-#endif	/* CONFIG_CRYPTO_HMAC */
-
 #endif	/* _LINUX_CRYPTO_H */
 
-- 
cgit v1.2.3


From fce32d70ba834129b164c40c2d4260e5a7a7d850 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 26 Aug 2006 17:35:45 +1000
Subject: [CRYPTO] api: Add crypto_comp and crypto_has_*

This patch adds the crypto_comp type to complete the compile-time checking
conversion.  The functions crypto_has_alg and crypto_has_cipher, etc. are
also added to replace crypto_alg_available.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/api.c           | 14 ++++++++
 include/linux/crypto.h | 90 ++++++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 98 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/crypto/api.c b/crypto/api.c
index edaa843d8e83..2e84d4b54790 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -482,3 +482,17 @@ int crypto_alg_available(const char *name, u32 flags)
 EXPORT_SYMBOL_GPL(crypto_alloc_tfm);
 EXPORT_SYMBOL_GPL(crypto_free_tfm);
 EXPORT_SYMBOL_GPL(crypto_alg_available);
+
+int crypto_has_alg(const char *name, u32 type, u32 mask)
+{
+	int ret = 0;
+	struct crypto_alg *alg = crypto_alg_mod_lookup(name, type, mask);
+	
+	if (!IS_ERR(alg)) {
+		crypto_mod_put(alg);
+		ret = 1;
+	}
+	
+	return ret;
+}
+EXPORT_SYMBOL_GPL(crypto_has_alg);
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 929fb9ad1314..cf91c4c0638b 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -236,11 +236,17 @@ int crypto_unregister_alg(struct crypto_alg *alg);
  */
 #ifdef CONFIG_CRYPTO
 int crypto_alg_available(const char *name, u32 flags);
+int crypto_has_alg(const char *name, u32 type, u32 mask);
 #else
 static inline int crypto_alg_available(const char *name, u32 flags)
 {
 	return 0;
 }
+
+static inline int crypto_has_alg(const char *name, u32 type, u32 mask)
+{
+	return 0;
+}
 #endif
 
 /*
@@ -329,6 +335,7 @@ struct crypto_tfm {
 };
 
 #define crypto_cipher crypto_tfm
+#define crypto_comp crypto_tfm
 
 struct crypto_blkcipher {
 	struct crypto_tfm base;
@@ -485,6 +492,15 @@ static inline void crypto_free_blkcipher(struct crypto_blkcipher *tfm)
 	crypto_free_tfm(crypto_blkcipher_tfm(tfm));
 }
 
+static inline int crypto_has_blkcipher(const char *alg_name, u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_BLKCIPHER;
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	return crypto_has_alg(alg_name, type, mask);
+}
+
 static inline const char *crypto_blkcipher_name(struct crypto_blkcipher *tfm)
 {
 	return crypto_tfm_alg_name(crypto_blkcipher_tfm(tfm));
@@ -620,6 +636,15 @@ static inline void crypto_free_cipher(struct crypto_cipher *tfm)
 	crypto_free_tfm(crypto_cipher_tfm(tfm));
 }
 
+static inline int crypto_has_cipher(const char *alg_name, u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_CIPHER;
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	return crypto_has_alg(alg_name, type, mask);
+}
+
 static inline struct cipher_tfm *crypto_cipher_crt(struct crypto_cipher *tfm)
 {
 	return &crypto_cipher_tfm(tfm)->crt_cipher;
@@ -718,6 +743,15 @@ static inline void crypto_free_hash(struct crypto_hash *tfm)
 	crypto_free_tfm(crypto_hash_tfm(tfm));
 }
 
+static inline int crypto_has_hash(const char *alg_name, u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_HASH;
+	mask |= CRYPTO_ALG_TYPE_HASH_MASK;
+
+	return crypto_has_alg(alg_name, type, mask);
+}
+
 static inline struct hash_tfm *crypto_hash_crt(struct crypto_hash *tfm)
 {
 	return &crypto_hash_tfm(tfm)->crt_hash;
@@ -853,20 +887,64 @@ static inline void crypto_cipher_get_iv(struct crypto_tfm *tfm,
 	memcpy(dst, tfm->crt_cipher.cit_iv, len);
 }
 
-static inline int crypto_comp_compress(struct crypto_tfm *tfm,
+static inline struct crypto_comp *__crypto_comp_cast(struct crypto_tfm *tfm)
+{
+	return (struct crypto_comp *)tfm;
+}
+
+static inline struct crypto_comp *crypto_comp_cast(struct crypto_tfm *tfm)
+{
+	BUG_ON((crypto_tfm_alg_type(tfm) ^ CRYPTO_ALG_TYPE_COMPRESS) &
+	       CRYPTO_ALG_TYPE_MASK);
+	return __crypto_comp_cast(tfm);
+}
+
+static inline struct crypto_comp *crypto_alloc_comp(const char *alg_name,
+						    u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_COMPRESS;
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	return __crypto_comp_cast(crypto_alloc_base(alg_name, type, mask));
+}
+
+static inline struct crypto_tfm *crypto_comp_tfm(struct crypto_comp *tfm)
+{
+	return tfm;
+}
+
+static inline void crypto_free_comp(struct crypto_comp *tfm)
+{
+	crypto_free_tfm(crypto_comp_tfm(tfm));
+}
+
+static inline int crypto_has_comp(const char *alg_name, u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_COMPRESS;
+	mask |= CRYPTO_ALG_TYPE_MASK;
+
+	return crypto_has_alg(alg_name, type, mask);
+}
+
+static inline struct compress_tfm *crypto_comp_crt(struct crypto_comp *tfm)
+{
+	return &crypto_comp_tfm(tfm)->crt_compress;
+}
+
+static inline int crypto_comp_compress(struct crypto_comp *tfm,
                                        const u8 *src, unsigned int slen,
                                        u8 *dst, unsigned int *dlen)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_COMPRESS);
-	return tfm->crt_compress.cot_compress(tfm, src, slen, dst, dlen);
+	return crypto_comp_crt(tfm)->cot_compress(tfm, src, slen, dst, dlen);
 }
 
-static inline int crypto_comp_decompress(struct crypto_tfm *tfm,
+static inline int crypto_comp_decompress(struct crypto_comp *tfm,
                                          const u8 *src, unsigned int slen,
                                          u8 *dst, unsigned int *dlen)
 {
-	BUG_ON(crypto_tfm_alg_type(tfm) != CRYPTO_ALG_TYPE_COMPRESS);
-	return tfm->crt_compress.cot_decompress(tfm, src, slen, dst, dlen);
+	return crypto_comp_crt(tfm)->cot_decompress(tfm, src, slen, dst, dlen);
 }
 
 #endif	/* _LINUX_CRYPTO_H */
-- 
cgit v1.2.3


From e4d5b79c661c7cfca9d8d5afd040a295f128d3cb Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 26 Aug 2006 18:12:40 +1000
Subject: [CRYPTO] users: Use crypto_comp and crypto_has_*

This patch converts all users to use the new crypto_comp type and the
crypto_has_* functions.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/tcrypt.c          |  8 ++++----
 drivers/crypto/padlock.c |  6 +++---
 drivers/net/ppp_mppe.c   |  4 ++--
 include/linux/crypto.h   |  5 +++++
 include/net/ipcomp.h     |  5 ++---
 net/ipv4/ipcomp.c        | 25 +++++++++++++------------
 net/ipv6/ipcomp6.c       | 25 +++++++++++++------------
 net/xfrm/xfrm_algo.c     | 27 ++++++++++++++++++---------
 8 files changed, 60 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index 840ab8be0b96..83307420d31c 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -749,7 +749,7 @@ static void test_deflate(void)
 {
 	unsigned int i;
 	char result[COMP_BUF_SIZE];
-	struct crypto_tfm *tfm;
+	struct crypto_comp *tfm;
 	struct comp_testvec *tv;
 	unsigned int tsize;
 
@@ -821,7 +821,7 @@ static void test_deflate(void)
 		       ilen, dlen);
 	}
 out:
-	crypto_free_tfm(tfm);
+	crypto_free_comp(tfm);
 }
 
 static void test_available(void)
@@ -830,8 +830,8 @@ static void test_available(void)
 
 	while (*name) {
 		printk("alg %s ", *name);
-		printk((crypto_alg_available(*name, 0)) ?
-			"found\n" : "not found\n");
+		printk(crypto_has_alg(*name, 0, CRYPTO_ALG_ASYNC) ?
+		       "found\n" : "not found\n");
 		name++;
 	}
 }
diff --git a/drivers/crypto/padlock.c b/drivers/crypto/padlock.c
index ce581684f4b4..d6d7dd5bb98c 100644
--- a/drivers/crypto/padlock.c
+++ b/drivers/crypto/padlock.c
@@ -26,13 +26,13 @@ static int __init padlock_init(void)
 {
 	int success = 0;
 
-	if (crypto_alg_available("aes-padlock", 0))
+	if (crypto_has_cipher("aes-padlock", 0, 0))
 		success++;
 
-	if (crypto_alg_available("sha1-padlock", 0))
+	if (crypto_has_hash("sha1-padlock", 0, 0))
 		success++;
 
-	if (crypto_alg_available("sha256-padlock", 0))
+	if (crypto_has_hash("sha256-padlock", 0, 0))
 		success++;
 
 	if (!success) {
diff --git a/drivers/net/ppp_mppe.c b/drivers/net/ppp_mppe.c
index e7a0eb4fca60..f3655fd772f5 100644
--- a/drivers/net/ppp_mppe.c
+++ b/drivers/net/ppp_mppe.c
@@ -710,8 +710,8 @@ static struct compressor ppp_mppe = {
 static int __init ppp_mppe_init(void)
 {
 	int answer;
-	if (!(crypto_alg_available("ecb(arc4)", 0) &&
-	      crypto_alg_available("sha1", 0)))
+	if (!(crypto_has_blkcipher("ecb(arc4)", 0, CRYPTO_ALG_ASYNC) &&
+	      crypto_has_hash("sha1", 0, CRYPTO_ALG_ASYNC)))
 		return -ENODEV;
 
 	sha_pad = kmalloc(sizeof(struct sha_pad), GFP_KERNEL);
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index cf91c4c0638b..d4f9948b64b1 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -928,6 +928,11 @@ static inline int crypto_has_comp(const char *alg_name, u32 type, u32 mask)
 	return crypto_has_alg(alg_name, type, mask);
 }
 
+static inline const char *crypto_comp_name(struct crypto_comp *tfm)
+{
+	return crypto_tfm_alg_name(crypto_comp_tfm(tfm));
+}
+
 static inline struct compress_tfm *crypto_comp_crt(struct crypto_comp *tfm)
 {
 	return &crypto_comp_tfm(tfm)->crt_compress;
diff --git a/include/net/ipcomp.h b/include/net/ipcomp.h
index b94e3047b4d9..87c1af3e5e82 100644
--- a/include/net/ipcomp.h
+++ b/include/net/ipcomp.h
@@ -1,15 +1,14 @@
 #ifndef _NET_IPCOMP_H
 #define _NET_IPCOMP_H
 
+#include <linux/crypto.h>
 #include <linux/types.h>
 
 #define IPCOMP_SCRATCH_SIZE     65400
 
-struct crypto_tfm;
-
 struct ipcomp_data {
 	u16 threshold;
-	struct crypto_tfm **tfms;
+	struct crypto_comp **tfms;
 };
 
 #endif
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index a0c28b2b756e..5bb9c9f03fb6 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -32,7 +32,7 @@
 
 struct ipcomp_tfms {
 	struct list_head list;
-	struct crypto_tfm **tfms;
+	struct crypto_comp **tfms;
 	int users;
 };
 
@@ -46,7 +46,7 @@ static int ipcomp_decompress(struct xfrm_state *x, struct sk_buff *skb)
 	int err, plen, dlen;
 	struct ipcomp_data *ipcd = x->data;
 	u8 *start, *scratch;
-	struct crypto_tfm *tfm;
+	struct crypto_comp *tfm;
 	int cpu;
 	
 	plen = skb->len;
@@ -107,7 +107,7 @@ static int ipcomp_compress(struct xfrm_state *x, struct sk_buff *skb)
 	struct iphdr *iph = skb->nh.iph;
 	struct ipcomp_data *ipcd = x->data;
 	u8 *start, *scratch;
-	struct crypto_tfm *tfm;
+	struct crypto_comp *tfm;
 	int cpu;
 	
 	ihlen = iph->ihl * 4;
@@ -302,7 +302,7 @@ static void **ipcomp_alloc_scratches(void)
 	return scratches;
 }
 
-static void ipcomp_free_tfms(struct crypto_tfm **tfms)
+static void ipcomp_free_tfms(struct crypto_comp **tfms)
 {
 	struct ipcomp_tfms *pos;
 	int cpu;
@@ -324,28 +324,28 @@ static void ipcomp_free_tfms(struct crypto_tfm **tfms)
 		return;
 
 	for_each_possible_cpu(cpu) {
-		struct crypto_tfm *tfm = *per_cpu_ptr(tfms, cpu);
-		crypto_free_tfm(tfm);
+		struct crypto_comp *tfm = *per_cpu_ptr(tfms, cpu);
+		crypto_free_comp(tfm);
 	}
 	free_percpu(tfms);
 }
 
-static struct crypto_tfm **ipcomp_alloc_tfms(const char *alg_name)
+static struct crypto_comp **ipcomp_alloc_tfms(const char *alg_name)
 {
 	struct ipcomp_tfms *pos;
-	struct crypto_tfm **tfms;
+	struct crypto_comp **tfms;
 	int cpu;
 
 	/* This can be any valid CPU ID so we don't need locking. */
 	cpu = raw_smp_processor_id();
 
 	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
-		struct crypto_tfm *tfm;
+		struct crypto_comp *tfm;
 
 		tfms = pos->tfms;
 		tfm = *per_cpu_ptr(tfms, cpu);
 
-		if (!strcmp(crypto_tfm_alg_name(tfm), alg_name)) {
+		if (!strcmp(crypto_comp_name(tfm), alg_name)) {
 			pos->users++;
 			return tfms;
 		}
@@ -359,12 +359,13 @@ static struct crypto_tfm **ipcomp_alloc_tfms(const char *alg_name)
 	INIT_LIST_HEAD(&pos->list);
 	list_add(&pos->list, &ipcomp_tfms_list);
 
-	pos->tfms = tfms = alloc_percpu(struct crypto_tfm *);
+	pos->tfms = tfms = alloc_percpu(struct crypto_comp *);
 	if (!tfms)
 		goto error;
 
 	for_each_possible_cpu(cpu) {
-		struct crypto_tfm *tfm = crypto_alloc_tfm(alg_name, 0);
+		struct crypto_comp *tfm = crypto_alloc_comp(alg_name, 0,
+							    CRYPTO_ALG_ASYNC);
 		if (!tfm)
 			goto error;
 		*per_cpu_ptr(tfms, cpu) = tfm;
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index 7e4d1c17bfbc..a81e9e9d93bd 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -53,7 +53,7 @@
 
 struct ipcomp6_tfms {
 	struct list_head list;
-	struct crypto_tfm **tfms;
+	struct crypto_comp **tfms;
 	int users;
 };
 
@@ -70,7 +70,7 @@ static int ipcomp6_input(struct xfrm_state *x, struct sk_buff *skb)
 	int plen, dlen;
 	struct ipcomp_data *ipcd = x->data;
 	u8 *start, *scratch;
-	struct crypto_tfm *tfm;
+	struct crypto_comp *tfm;
 	int cpu;
 
 	if (skb_linearize_cow(skb))
@@ -129,7 +129,7 @@ static int ipcomp6_output(struct xfrm_state *x, struct sk_buff *skb)
 	struct ipcomp_data *ipcd = x->data;
 	int plen, dlen;
 	u8 *start, *scratch;
-	struct crypto_tfm *tfm;
+	struct crypto_comp *tfm;
 	int cpu;
 
 	hdr_len = skb->h.raw - skb->data;
@@ -301,7 +301,7 @@ static void **ipcomp6_alloc_scratches(void)
 	return scratches;
 }
 
-static void ipcomp6_free_tfms(struct crypto_tfm **tfms)
+static void ipcomp6_free_tfms(struct crypto_comp **tfms)
 {
 	struct ipcomp6_tfms *pos;
 	int cpu;
@@ -323,28 +323,28 @@ static void ipcomp6_free_tfms(struct crypto_tfm **tfms)
 		return;
 
 	for_each_possible_cpu(cpu) {
-		struct crypto_tfm *tfm = *per_cpu_ptr(tfms, cpu);
-		crypto_free_tfm(tfm);
+		struct crypto_comp *tfm = *per_cpu_ptr(tfms, cpu);
+		crypto_free_comp(tfm);
 	}
 	free_percpu(tfms);
 }
 
-static struct crypto_tfm **ipcomp6_alloc_tfms(const char *alg_name)
+static struct crypto_comp **ipcomp6_alloc_tfms(const char *alg_name)
 {
 	struct ipcomp6_tfms *pos;
-	struct crypto_tfm **tfms;
+	struct crypto_comp **tfms;
 	int cpu;
 
 	/* This can be any valid CPU ID so we don't need locking. */
 	cpu = raw_smp_processor_id();
 
 	list_for_each_entry(pos, &ipcomp6_tfms_list, list) {
-		struct crypto_tfm *tfm;
+		struct crypto_comp *tfm;
 
 		tfms = pos->tfms;
 		tfm = *per_cpu_ptr(tfms, cpu);
 
-		if (!strcmp(crypto_tfm_alg_name(tfm), alg_name)) {
+		if (!strcmp(crypto_comp_name(tfm), alg_name)) {
 			pos->users++;
 			return tfms;
 		}
@@ -358,12 +358,13 @@ static struct crypto_tfm **ipcomp6_alloc_tfms(const char *alg_name)
 	INIT_LIST_HEAD(&pos->list);
 	list_add(&pos->list, &ipcomp6_tfms_list);
 
-	pos->tfms = tfms = alloc_percpu(struct crypto_tfm *);
+	pos->tfms = tfms = alloc_percpu(struct crypto_comp *);
 	if (!tfms)
 		goto error;
 
 	for_each_possible_cpu(cpu) {
-		struct crypto_tfm *tfm = crypto_alloc_tfm(alg_name, 0);
+		struct crypto_comp *tfm = crypto_alloc_comp(alg_name, 0,
+							    CRYPTO_ALG_ASYNC);
 		if (!tfm)
 			goto error;
 		*per_cpu_ptr(tfms, cpu) = tfm;
diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index 87918f281bb4..5a0dbeb6bbe8 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -363,8 +363,8 @@ struct xfrm_algo_desc *xfrm_calg_get_byid(int alg_id)
 EXPORT_SYMBOL_GPL(xfrm_calg_get_byid);
 
 static struct xfrm_algo_desc *xfrm_get_byname(struct xfrm_algo_desc *list,
-					      int entries, char *name,
-					      int probe)
+					      int entries, u32 type, u32 mask,
+					      char *name, int probe)
 {
 	int i, status;
 
@@ -382,7 +382,7 @@ static struct xfrm_algo_desc *xfrm_get_byname(struct xfrm_algo_desc *list,
 		if (!probe)
 			break;
 
-		status = crypto_alg_available(name, 0);
+		status = crypto_has_alg(name, type, mask | CRYPTO_ALG_ASYNC);
 		if (!status)
 			break;
 
@@ -394,19 +394,25 @@ static struct xfrm_algo_desc *xfrm_get_byname(struct xfrm_algo_desc *list,
 
 struct xfrm_algo_desc *xfrm_aalg_get_byname(char *name, int probe)
 {
-	return xfrm_get_byname(aalg_list, aalg_entries(), name, probe);
+	return xfrm_get_byname(aalg_list, aalg_entries(),
+			       CRYPTO_ALG_TYPE_HASH, CRYPTO_ALG_TYPE_HASH_MASK,
+			       name, probe);
 }
 EXPORT_SYMBOL_GPL(xfrm_aalg_get_byname);
 
 struct xfrm_algo_desc *xfrm_ealg_get_byname(char *name, int probe)
 {
-	return xfrm_get_byname(ealg_list, ealg_entries(), name, probe);
+	return xfrm_get_byname(ealg_list, ealg_entries(),
+			       CRYPTO_ALG_TYPE_BLKCIPHER, CRYPTO_ALG_TYPE_MASK,
+			       name, probe);
 }
 EXPORT_SYMBOL_GPL(xfrm_ealg_get_byname);
 
 struct xfrm_algo_desc *xfrm_calg_get_byname(char *name, int probe)
 {
-	return xfrm_get_byname(calg_list, calg_entries(), name, probe);
+	return xfrm_get_byname(calg_list, calg_entries(),
+			       CRYPTO_ALG_TYPE_COMPRESS, CRYPTO_ALG_TYPE_MASK,
+			       name, probe);
 }
 EXPORT_SYMBOL_GPL(xfrm_calg_get_byname);
 
@@ -441,19 +447,22 @@ void xfrm_probe_algs(void)
 	BUG_ON(in_softirq());
 
 	for (i = 0; i < aalg_entries(); i++) {
-		status = crypto_alg_available(aalg_list[i].name, 0);
+		status = crypto_has_hash(aalg_list[i].name, 0,
+					 CRYPTO_ALG_ASYNC);
 		if (aalg_list[i].available != status)
 			aalg_list[i].available = status;
 	}
 	
 	for (i = 0; i < ealg_entries(); i++) {
-		status = crypto_alg_available(ealg_list[i].name, 0);
+		status = crypto_has_blkcipher(ealg_list[i].name, 0,
+					      CRYPTO_ALG_ASYNC);
 		if (ealg_list[i].available != status)
 			ealg_list[i].available = status;
 	}
 	
 	for (i = 0; i < calg_entries(); i++) {
-		status = crypto_alg_available(calg_list[i].name, 0);
+		status = crypto_has_comp(calg_list[i].name, 0,
+					 CRYPTO_ALG_ASYNC);
 		if (calg_list[i].available != status)
 			calg_list[i].available = status;
 	}
-- 
cgit v1.2.3


From 6010439f47e6b308c031dad7d99686030ef942dd Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 26 Aug 2006 18:34:10 +1000
Subject: [CRYPTO] padlock: Convert padlock-sha to use crypto_hash

This patch converts padlock-sha to use crypto_hash for its fallback.
It also changes the fallback selection to use selection by type instead
of name.  This is done through the new CRYPTO_ALG_NEED_FALLBACK bit,
which is set if and only if an algorithm needs a fallback of the same
type.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 drivers/crypto/padlock-sha.c | 91 +++++++++++++++++---------------------------
 include/linux/crypto.h       |  6 +++
 2 files changed, 41 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/drivers/crypto/padlock-sha.c b/drivers/crypto/padlock-sha.c
index b028db61c301..a781fd23b607 100644
--- a/drivers/crypto/padlock-sha.c
+++ b/drivers/crypto/padlock-sha.c
@@ -12,10 +12,11 @@
  *
  */
 
+#include <crypto/algapi.h>
+#include <linux/err.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/errno.h>
-#include <linux/crypto.h>
 #include <linux/cryptohash.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
@@ -30,28 +31,17 @@
 #define SHA256_DIGEST_SIZE      32
 #define SHA256_HMAC_BLOCK_SIZE  64
 
-static char *sha1_fallback = SHA1_DEFAULT_FALLBACK;
-static char *sha256_fallback = SHA256_DEFAULT_FALLBACK;
-
-module_param(sha1_fallback, charp, 0644);
-module_param(sha256_fallback, charp, 0644);
-
-MODULE_PARM_DESC(sha1_fallback, "Fallback driver for SHA1. Default is "
-		 SHA1_DEFAULT_FALLBACK);
-MODULE_PARM_DESC(sha256_fallback, "Fallback driver for SHA256. Default is "
-		 SHA256_DEFAULT_FALLBACK);
-
 struct padlock_sha_ctx {
 	char		*data;
 	size_t		used;
 	int		bypass;
 	void (*f_sha_padlock)(const char *in, char *out, int count);
-	struct crypto_tfm *fallback_tfm;
+	struct hash_desc fallback;
 };
 
 static inline struct padlock_sha_ctx *ctx(struct crypto_tfm *tfm)
 {
-	return (struct padlock_sha_ctx *)(crypto_tfm_ctx(tfm));
+	return crypto_tfm_ctx(tfm);
 }
 
 /* We'll need aligned address on the stack */
@@ -65,14 +55,12 @@ static void padlock_sha_bypass(struct crypto_tfm *tfm)
 	if (ctx(tfm)->bypass)
 		return;
 
-	BUG_ON(!ctx(tfm)->fallback_tfm);
-
-	crypto_digest_init(ctx(tfm)->fallback_tfm);
+	crypto_hash_init(&ctx(tfm)->fallback);
 	if (ctx(tfm)->data && ctx(tfm)->used) {
 		struct scatterlist sg;
 
 		sg_set_buf(&sg, ctx(tfm)->data, ctx(tfm)->used);
-		crypto_digest_update(ctx(tfm)->fallback_tfm, &sg, 1);
+		crypto_hash_update(&ctx(tfm)->fallback, &sg, sg.length);
 	}
 
 	ctx(tfm)->used = 0;
@@ -95,9 +83,8 @@ static void padlock_sha_update(struct crypto_tfm *tfm,
 
 	if (unlikely(ctx(tfm)->bypass)) {
 		struct scatterlist sg;
-		BUG_ON(!ctx(tfm)->fallback_tfm);
 		sg_set_buf(&sg, (uint8_t *)data, length);
-		crypto_digest_update(ctx(tfm)->fallback_tfm, &sg, 1);
+		crypto_hash_update(&ctx(tfm)->fallback, &sg, length);
 		return;
 	}
 
@@ -160,8 +147,7 @@ static void padlock_do_sha256(const char *in, char *out, int count)
 static void padlock_sha_final(struct crypto_tfm *tfm, uint8_t *out)
 {
 	if (unlikely(ctx(tfm)->bypass)) {
-		BUG_ON(!ctx(tfm)->fallback_tfm);
-		crypto_digest_final(ctx(tfm)->fallback_tfm, out);
+		crypto_hash_final(&ctx(tfm)->fallback, out);
 		ctx(tfm)->bypass = 0;
 		return;
 	}
@@ -172,8 +158,11 @@ static void padlock_sha_final(struct crypto_tfm *tfm, uint8_t *out)
 	ctx(tfm)->used = 0;
 }
 
-static int padlock_cra_init(struct crypto_tfm *tfm, const char *fallback_driver_name)
+static int padlock_cra_init(struct crypto_tfm *tfm)
 {
+	const char *fallback_driver_name = tfm->__crt_alg->cra_name;
+	struct crypto_hash *fallback_tfm;
+
 	/* For now we'll allocate one page. This
 	 * could eventually be configurable one day. */
 	ctx(tfm)->data = (char *)__get_free_page(GFP_KERNEL);
@@ -181,14 +170,17 @@ static int padlock_cra_init(struct crypto_tfm *tfm, const char *fallback_driver_
 		return -ENOMEM;
 
 	/* Allocate a fallback and abort if it failed. */
-	ctx(tfm)->fallback_tfm = crypto_alloc_tfm(fallback_driver_name, 0);
-	if (!ctx(tfm)->fallback_tfm) {
+	fallback_tfm = crypto_alloc_hash(fallback_driver_name, 0,
+					 CRYPTO_ALG_ASYNC |
+					 CRYPTO_ALG_NEED_FALLBACK);
+	if (IS_ERR(fallback_tfm)) {
 		printk(KERN_WARNING PFX "Fallback driver '%s' could not be loaded!\n",
 		       fallback_driver_name);
 		free_page((unsigned long)(ctx(tfm)->data));
-		return -ENOENT;
+		return PTR_ERR(fallback_tfm);
 	}
 
+	ctx(tfm)->fallback.tfm = fallback_tfm;
 	return 0;
 }
 
@@ -196,14 +188,14 @@ static int padlock_sha1_cra_init(struct crypto_tfm *tfm)
 {
 	ctx(tfm)->f_sha_padlock = padlock_do_sha1;
 
-	return padlock_cra_init(tfm, sha1_fallback);
+	return padlock_cra_init(tfm);
 }
 
 static int padlock_sha256_cra_init(struct crypto_tfm *tfm)
 {
 	ctx(tfm)->f_sha_padlock = padlock_do_sha256;
 
-	return padlock_cra_init(tfm, sha256_fallback);
+	return padlock_cra_init(tfm);
 }
 
 static void padlock_cra_exit(struct crypto_tfm *tfm)
@@ -213,16 +205,16 @@ static void padlock_cra_exit(struct crypto_tfm *tfm)
 		ctx(tfm)->data = NULL;
 	}
 
-	BUG_ON(!ctx(tfm)->fallback_tfm);
-	crypto_free_tfm(ctx(tfm)->fallback_tfm);
-	ctx(tfm)->fallback_tfm = NULL;
+	crypto_free_hash(ctx(tfm)->fallback.tfm);
+	ctx(tfm)->fallback.tfm = NULL;
 }
 
 static struct crypto_alg sha1_alg = {
 	.cra_name		=	"sha1",
 	.cra_driver_name	=	"sha1-padlock",
 	.cra_priority		=	PADLOCK_CRA_PRIORITY,
-	.cra_flags		=	CRYPTO_ALG_TYPE_DIGEST,
+	.cra_flags		=	CRYPTO_ALG_TYPE_DIGEST |
+					CRYPTO_ALG_NEED_FALLBACK,
 	.cra_blocksize		=	SHA1_HMAC_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof(struct padlock_sha_ctx),
 	.cra_module		=	THIS_MODULE,
@@ -243,7 +235,8 @@ static struct crypto_alg sha256_alg = {
 	.cra_name		=	"sha256",
 	.cra_driver_name	=	"sha256-padlock",
 	.cra_priority		=	PADLOCK_CRA_PRIORITY,
-	.cra_flags		=	CRYPTO_ALG_TYPE_DIGEST,
+	.cra_flags		=	CRYPTO_ALG_TYPE_DIGEST |
+					CRYPTO_ALG_NEED_FALLBACK,
 	.cra_blocksize		=	SHA256_HMAC_BLOCK_SIZE,
 	.cra_ctxsize		=	sizeof(struct padlock_sha_ctx),
 	.cra_module		=	THIS_MODULE,
@@ -262,29 +255,15 @@ static struct crypto_alg sha256_alg = {
 
 static void __init padlock_sha_check_fallbacks(void)
 {
-	struct crypto_tfm *tfm;
-
-	/* We'll try to allocate one TFM for each fallback
-	 * to test that the modules are available. */
-	tfm = crypto_alloc_tfm(sha1_fallback, 0);
-	if (!tfm) {
-		printk(KERN_WARNING PFX "Couldn't load fallback module for '%s'. Tried '%s'.\n",
-		       sha1_alg.cra_name, sha1_fallback);
-	} else {
-		printk(KERN_NOTICE PFX "Fallback for '%s' is driver '%s' (prio=%d)\n", sha1_alg.cra_name,
-		       crypto_tfm_alg_driver_name(tfm), crypto_tfm_alg_priority(tfm));
-		crypto_free_tfm(tfm);
-	}
-
-	tfm = crypto_alloc_tfm(sha256_fallback, 0);
-	if (!tfm) {
-		printk(KERN_WARNING PFX "Couldn't load fallback module for '%s'. Tried '%s'.\n",
-		       sha256_alg.cra_name, sha256_fallback);
-	} else {
-		printk(KERN_NOTICE PFX "Fallback for '%s' is driver '%s' (prio=%d)\n", sha256_alg.cra_name,
-		       crypto_tfm_alg_driver_name(tfm), crypto_tfm_alg_priority(tfm));
-		crypto_free_tfm(tfm);
-	}
+	if (!crypto_has_hash("sha1", 0, CRYPTO_ALG_ASYNC |
+					CRYPTO_ALG_NEED_FALLBACK))
+		printk(KERN_WARNING PFX
+		       "Couldn't load fallback module for sha1.\n");
+
+	if (!crypto_has_hash("sha256", 0, CRYPTO_ALG_ASYNC |
+					CRYPTO_ALG_NEED_FALLBACK))
+		printk(KERN_WARNING PFX
+		       "Couldn't load fallback module for sha256.\n");
 }
 
 static int __init padlock_init(void)
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index d4f9948b64b1..187c6ea91959 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -42,6 +42,12 @@
 #define CRYPTO_ALG_DYING		0x00000040
 #define CRYPTO_ALG_ASYNC		0x00000080
 
+/*
+ * Set this bit if and only if the algorithm requires another algorithm of
+ * the same type to handle corner cases.
+ */
+#define CRYPTO_ALG_NEED_FALLBACK	0x00000100
+
 /*
  * Transform masks and values (for crt_flags).
  */
-- 
cgit v1.2.3


From 3ad819c61f5f8347f39cdcbe652b3c60ec615888 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sat, 26 Aug 2006 18:44:31 +1000
Subject: [CRYPTO] api: Deprecate crypto_digest_* and crypto_alg_available

This patch marks the crypto_digest_* functions and crypto_alg_available
as deprecated.  They've been replaced by crypto_hash_* and crypto_has_*
respectively.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 include/linux/crypto.h | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 187c6ea91959..8f2ffa4caabf 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -241,9 +241,12 @@ int crypto_unregister_alg(struct crypto_alg *alg);
  * Algorithm query interface.
  */
 #ifdef CONFIG_CRYPTO
-int crypto_alg_available(const char *name, u32 flags);
+int crypto_alg_available(const char *name, u32 flags)
+	__deprecated_for_modules;
 int crypto_has_alg(const char *name, u32 type, u32 mask);
 #else
+static int crypto_alg_available(const char *name, u32 flags);
+	__deprecated_for_modules;
 static inline int crypto_alg_available(const char *name, u32 flags)
 {
 	return 0;
@@ -704,12 +707,15 @@ static inline void crypto_cipher_decrypt_one(struct crypto_cipher *tfm,
 						dst, src);
 }
 
-void crypto_digest_init(struct crypto_tfm *tfm);
+void crypto_digest_init(struct crypto_tfm *tfm) __deprecated_for_modules;
 void crypto_digest_update(struct crypto_tfm *tfm,
-			  struct scatterlist *sg, unsigned int nsg);
-void crypto_digest_final(struct crypto_tfm *tfm, u8 *out);
+			  struct scatterlist *sg, unsigned int nsg)
+	__deprecated_for_modules;
+void crypto_digest_final(struct crypto_tfm *tfm, u8 *out)
+	__deprecated_for_modules;
 void crypto_digest_digest(struct crypto_tfm *tfm,
-			  struct scatterlist *sg, unsigned int nsg, u8 *out);
+			  struct scatterlist *sg, unsigned int nsg, u8 *out)
+	__deprecated_for_modules;
 
 static inline struct crypto_hash *__crypto_hash_cast(struct crypto_tfm *tfm)
 {
@@ -723,6 +729,8 @@ static inline struct crypto_hash *crypto_hash_cast(struct crypto_tfm *tfm)
 	return __crypto_hash_cast(tfm);
 }
 
+static int crypto_digest_setkey(struct crypto_tfm *tfm, const u8 *key,
+				unsigned int keylen) __deprecated;
 static inline int crypto_digest_setkey(struct crypto_tfm *tfm,
                                        const u8 *key, unsigned int keylen)
 {
-- 
cgit v1.2.3


From 6ff6340abeaaf1a15587c87dac3e56754778cc7a Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 08:34:39 +0100
Subject: [HEADERS] Fix ARM 'make headers_check'

Sanitise the ARM headers exported to userspace.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-arm/elf.h  | 18 ++++++++----------
 include/asm-arm/page.h |  4 ++--
 2 files changed, 10 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/elf.h b/include/asm-arm/elf.h
index ae7baa6c73f7..17f0c656d272 100644
--- a/include/asm-arm/elf.h
+++ b/include/asm-arm/elf.h
@@ -8,9 +8,6 @@
 
 #include <asm/ptrace.h>
 #include <asm/user.h>
-#ifdef __KERNEL
-#include <asm/procinfo.h>
-#endif
 
 typedef unsigned long elf_greg_t;
 typedef unsigned long elf_freg_t[3];
@@ -31,11 +28,6 @@ typedef elf_greg_t elf_gregset_t[ELF_NGREG];
 
 typedef struct user_fp elf_fpregset_t;
 
-/*
- * This is used to ensure we don't load something for the wrong architecture.
- */
-#define elf_check_arch(x) ( ((x)->e_machine == EM_ARM) && (ELF_PROC_OK((x))) )
-
 /*
  * These are used to set parameters in the core dumps.
  */
@@ -47,6 +39,14 @@ typedef struct user_fp elf_fpregset_t;
 #endif
 #define ELF_ARCH	EM_ARM
 
+#ifdef __KERNEL__
+#include <asm/procinfo.h>
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x) ( ((x)->e_machine == EM_ARM) && (ELF_PROC_OK((x))) )
+
 #define USE_ELF_CORE_DUMP
 #define ELF_EXEC_PAGESIZE	4096
 
@@ -83,8 +83,6 @@ typedef struct user_fp elf_fpregset_t;
 extern char elf_platform[];
 #define ELF_PLATFORM	(elf_platform)
 
-#ifdef __KERNEL__
-
 /*
  * 32-bit code is always OK.  Some cpus can do 26-bit, some can't.
  */
diff --git a/include/asm-arm/page.h b/include/asm-arm/page.h
index b721270b9986..02bd3ee935b0 100644
--- a/include/asm-arm/page.h
+++ b/include/asm-arm/page.h
@@ -11,13 +11,13 @@
 #define _ASMARM_PAGE_H
 
 
+#ifdef __KERNEL__
+
 /* PAGE_SHIFT determines the page size */
 #define PAGE_SHIFT		12
 #define PAGE_SIZE		(1UL << PAGE_SHIFT)
 #define PAGE_MASK		(~(PAGE_SIZE-1))
 
-#ifdef __KERNEL__
-
 /* to align the pointer to the (next) page boundary */
 #define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
 
-- 
cgit v1.2.3


From b5233d0704c9a6147ebbfabc576d1638b3ac5274 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 20 Sep 2006 03:25:34 +0900
Subject: Fix 'make headers_check' on sh

Cleanup for user headers, as noted:

asm-sh/page.h requires asm-generic/memory_model.h, which does not exist in exported headers
asm-sh/ptrace.h requires asm/ubc.h, which does not exist in exported headers

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 arch/sh/kernel/process.c | 1 +
 include/asm-sh/page.h    | 3 +--
 include/asm-sh/ptrace.h  | 2 --
 3 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 22dc9c21201d..f2031314cb2b 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -26,6 +26,7 @@
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
 #include <asm/elf.h>
+#include <asm/ubc.h>
 
 static int hlt_counter=0;
 
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 5a057b00f19a..6f7eb8a3aba5 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -112,9 +112,8 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
 #endif /* __ASM_SH_PAGE_H */
diff --git a/include/asm-sh/ptrace.h b/include/asm-sh/ptrace.h
index 792fc35bd624..ed358a376e6e 100644
--- a/include/asm-sh/ptrace.h
+++ b/include/asm-sh/ptrace.h
@@ -1,8 +1,6 @@
 #ifndef __ASM_SH_PTRACE_H
 #define __ASM_SH_PTRACE_H
 
-#include <asm/ubc.h>
-
 /*
  * Copyright (C) 1999, 2000  Niibe Yutaka
  *
-- 
cgit v1.2.3


From 029669da25efa18ee4b8911e694fdcf4a11c8cbe Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 20 Sep 2006 03:27:17 +0900
Subject: Fix 'make headers_check' on sh64

Cleanup for user headers, as noted:

asm-sh64/page.h requires asm-generic/memory_model.h, which does not exist in exported headers
asm-sh64/shmparam.h requires asm/cache.h, which does not exist in exported headers
asm-sh64/signal.h requires asm/processor.h, which does not exist in exported headers
asm-sh64/user.h requires asm/processor.h, which does not exist in exported headers

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-sh64/page.h     |  3 +--
 include/asm-sh64/shmparam.h | 16 ++++------------
 include/asm-sh64/signal.h   |  1 -
 include/asm-sh64/user.h     |  1 -
 4 files changed, 5 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh64/page.h b/include/asm-sh64/page.h
index 34fb34754ae6..472089aefc60 100644
--- a/include/asm-sh64/page.h
+++ b/include/asm-sh64/page.h
@@ -112,9 +112,8 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
 #endif /* __ASM_SH64_PAGE_H */
diff --git a/include/asm-sh64/shmparam.h b/include/asm-sh64/shmparam.h
index d3a99a4dc0e3..1bb820c833ee 100644
--- a/include/asm-sh64/shmparam.h
+++ b/include/asm-sh64/shmparam.h
@@ -2,19 +2,11 @@
 #define __ASM_SH64_SHMPARAM_H
 
 /*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * include/asm-sh64/shmparam.h
- *
- * Copyright (C) 2000, 2001  Paolo Alberelli
- *
+ * Set this to a sensible safe default, we'll work out the specifics for the
+ * align mask from the cache descriptor at run-time.
  */
+#define	SHMLBA	0x4000
 
-#include <asm/cache.h>
-
-/* attach addr a multiple of this */
-#define	SHMLBA	(cpu_data->dcache.sets * L1_CACHE_BYTES)
+#define __ARCH_FORCE_SHMLBA
 
 #endif /* __ASM_SH64_SHMPARAM_H */
diff --git a/include/asm-sh64/signal.h b/include/asm-sh64/signal.h
index a5a28203cb3b..244e134730d9 100644
--- a/include/asm-sh64/signal.h
+++ b/include/asm-sh64/signal.h
@@ -13,7 +13,6 @@
  */
 
 #include <linux/types.h>
-#include <asm/processor.h>
 
 /* Avoid too many header ordering problems.  */
 struct siginfo;
diff --git a/include/asm-sh64/user.h b/include/asm-sh64/user.h
index 8f32f39a8ca9..eb3b33edd73e 100644
--- a/include/asm-sh64/user.h
+++ b/include/asm-sh64/user.h
@@ -13,7 +13,6 @@
  */
 
 #include <linux/types.h>
-#include <asm/processor.h>
 #include <asm/ptrace.h>
 #include <asm/page.h>
 
-- 
cgit v1.2.3


From 47dbec79d1b9ce9e80bed932f345adc92049f05d Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Sun, 17 Sep 2006 08:39:39 +0100
Subject: Fix 'make headers_check' on m32r

> asm-m32r/page.h requires asm-generic/memory_model.h, which does not exist
> asm-m32r/ptrace.h requires asm/m32r.h, which does not exist
> asm-m32r/signal.h requires linux/linkage.h, which does not exist
> asm-m32r/unistd.h requires asm/syscall.h, which does not exist
> asm-m32r/user.h requires asm/processor.h, which does not exist

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-m32r/page.h   | 3 +--
 include/asm-m32r/ptrace.h | 4 ++--
 include/asm-m32r/signal.h | 1 -
 include/asm-m32r/unistd.h | 4 ++--
 include/asm-m32r/user.h   | 1 -
 5 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-m32r/page.h b/include/asm-m32r/page.h
index 9688be003620..404a4c24007b 100644
--- a/include/asm-m32r/page.h
+++ b/include/asm-m32r/page.h
@@ -87,10 +87,9 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 
 #define devmem_is_allowed(x) 1
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
 #endif /* _ASM_M32R_PAGE_H */
 
diff --git a/include/asm-m32r/ptrace.h b/include/asm-m32r/ptrace.h
index a07fa90314d2..2d2a6c97331e 100644
--- a/include/asm-m32r/ptrace.h
+++ b/include/asm-m32r/ptrace.h
@@ -12,8 +12,6 @@
  *   Copyright (C) 2001-2002, 2004  Hirokazu Takata <takata at linux-m32r.org>
  */
 
-#include <asm/m32r.h>		/* M32R_PSW_BSM, M32R_PSW_BPM */
-
 /* 0 - 13 are integer registers (general purpose registers).  */
 #define PT_R4		0
 #define PT_R5		1
@@ -140,6 +138,8 @@ struct pt_regs {
 
 #ifdef __KERNEL__
 
+#include <asm/m32r.h>		/* M32R_PSW_BSM, M32R_PSW_BPM */
+
 #define __ARCH_SYS_PTRACE	1
 
 #if defined(CONFIG_ISA_M32R2) || defined(CONFIG_CHIP_VDEC2)
diff --git a/include/asm-m32r/signal.h b/include/asm-m32r/signal.h
index e750045164d4..65423bed32b1 100644
--- a/include/asm-m32r/signal.h
+++ b/include/asm-m32r/signal.h
@@ -6,7 +6,6 @@
 /* orig : i386 2.4.18 */
 
 #include <linux/types.h>
-#include <linux/linkage.h>
 #include <linux/time.h>
 #include <linux/compiler.h>
 
diff --git a/include/asm-m32r/unistd.h b/include/asm-m32r/unistd.h
index cc31790d8077..89f376e6229f 100644
--- a/include/asm-m32r/unistd.h
+++ b/include/asm-m32r/unistd.h
@@ -3,8 +3,6 @@
 
 /* $Id$ */
 
-#include <asm/syscall.h>	/* SYSCALL_* */
-
 /*
  * This file contains the system call numbers.
  */
@@ -303,6 +301,8 @@
  * <asm-m32r/errno.h>
  */
 
+#include <asm/syscall.h>	/* SYSCALL_* */
+
 #define __syscall_return(type, res) \
 do { \
 	if ((unsigned long)(res) >= (unsigned long)(-(124 + 1))) { \
diff --git a/include/asm-m32r/user.h b/include/asm-m32r/user.h
index 2ffd0c65a782..1ad4ded8483b 100644
--- a/include/asm-m32r/user.h
+++ b/include/asm-m32r/user.h
@@ -8,7 +8,6 @@
  */
 
 #include <linux/types.h>
-#include <asm/processor.h>
 #include <asm/ptrace.h>
 #include <asm/page.h>
 
-- 
cgit v1.2.3


From 09087a1a8722fac30b1969a4a542cde064af13f8 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 08:48:27 +0100
Subject: Fix exported headers for SPARC, SPARC64

Mostly removing files which have no business being used in userspace.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/asm-sparc/Kbuild       | 7 -------
 include/asm-sparc/page.h       | 8 ++++----
 include/asm-sparc64/Kbuild     | 5 +----
 include/asm-sparc64/page.h     | 9 ++++-----
 include/asm-sparc64/shmparam.h | 2 ++
 5 files changed, 11 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/asm-sparc/Kbuild b/include/asm-sparc/Kbuild
index b22b67a64ecc..c6a55cf0d337 100644
--- a/include/asm-sparc/Kbuild
+++ b/include/asm-sparc/Kbuild
@@ -2,20 +2,13 @@ include include/asm-generic/Kbuild.asm
 
 header-y += apc.h
 header-y += asi.h
-header-y += auxio.h
 header-y += bpp.h
-header-y += head.h
-header-y += ipc.h
 header-y += jsflash.h
 header-y += openpromio.h
-header-y += pbm.h
 header-y += pconf.h
-header-y += pgtsun4.h
 header-y += reg.h
 header-y += traps.h
-header-y += turbosparc.h
 header-y += vfc_ioctls.h
-header-y += winmacro.h
 
 unifdef-y += fbio.h
 unifdef-y += perfctr.h
diff --git a/include/asm-sparc/page.h b/include/asm-sparc/page.h
index 5bab8a7c25ce..ff57648eb8f8 100644
--- a/include/asm-sparc/page.h
+++ b/include/asm-sparc/page.h
@@ -8,6 +8,8 @@
 #ifndef _SPARC_PAGE_H
 #define _SPARC_PAGE_H
 
+#ifdef __KERNEL__
+
 #ifdef CONFIG_SUN4
 #define PAGE_SHIFT   13
 #else
@@ -21,8 +23,6 @@
 #endif
 #define PAGE_MASK    (~(PAGE_SIZE-1))
 
-#ifdef __KERNEL__
-
 #include <asm/btfixup.h>
 
 #ifndef __ASSEMBLY__
@@ -160,9 +160,9 @@ extern unsigned long pfn_base;
 #define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
+
 #endif /* _SPARC_PAGE_H */
diff --git a/include/asm-sparc64/Kbuild b/include/asm-sparc64/Kbuild
index 4b59ce46cc2d..a7f44408c93b 100644
--- a/include/asm-sparc64/Kbuild
+++ b/include/asm-sparc64/Kbuild
@@ -8,15 +8,12 @@ header-y += apb.h
 header-y += asi.h
 header-y += bbc.h
 header-y += bpp.h
+header-y += const.h
 header-y += display7seg.h
 header-y += envctrl.h
-header-y += floppy.h
 header-y += ipc.h
-header-y += kdebug.h
-header-y += mostek.h
 header-y += openprom.h
 header-y += openpromio.h
-header-y += parport.h
 header-y += pconf.h
 header-y += psrcompat.h
 header-y += pstate.h
diff --git a/include/asm-sparc64/page.h b/include/asm-sparc64/page.h
index fdf0ceb76028..ff736eafa64d 100644
--- a/include/asm-sparc64/page.h
+++ b/include/asm-sparc64/page.h
@@ -3,6 +3,8 @@
 #ifndef _SPARC64_PAGE_H
 #define _SPARC64_PAGE_H
 
+#ifdef __KERNEL__
+
 #include <asm/const.h>
 
 #if defined(CONFIG_SPARC64_PAGE_SIZE_8KB)
@@ -27,8 +29,6 @@
 #define DCACHE_ALIASING_POSSIBLE
 #endif
 
-#ifdef __KERNEL__
-
 #if defined(CONFIG_HUGETLB_PAGE_SIZE_4MB)
 #define HPAGE_SHIFT		22
 #elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K)
@@ -141,8 +141,7 @@ typedef unsigned long pgprot_t;
 #define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-#endif /* !(__KERNEL__) */
-
 #include <asm-generic/page.h>
 
-#endif /* !(_SPARC64_PAGE_H) */
+#endif /* __KERNEL__ */
+#endif /* _SPARC64_PAGE_H */
diff --git a/include/asm-sparc64/shmparam.h b/include/asm-sparc64/shmparam.h
index 8c66fded8a32..911d0427de6d 100644
--- a/include/asm-sparc64/shmparam.h
+++ b/include/asm-sparc64/shmparam.h
@@ -1,6 +1,7 @@
 /* $Id: shmparam.h,v 1.5 2001/09/24 21:17:57 kanoj Exp $ */
 #ifndef _ASMSPARC64_SHMPARAM_H
 #define _ASMSPARC64_SHMPARAM_H
+#ifdef __KERNEL__
 
 #include <asm/spitfire.h>
 
@@ -8,4 +9,5 @@
 /* attach addr a multiple of this */
 #define	SHMLBA	((PAGE_SIZE > L1DCACHE_SIZE) ? PAGE_SIZE : L1DCACHE_SIZE)
 
+#endif /* __KERNEL__ */
 #endif /* _ASMSPARC64_SHMPARAM_H */
-- 
cgit v1.2.3


From 5daf3033aa4fc5ec0535816d8a8e94874160482e Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 08:50:36 +0100
Subject: Fix m68knommu exported headers

Just clean up asm/page.h

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-m68knommu/page.h | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-m68knommu/page.h b/include/asm-m68knommu/page.h
index a22bf5a88160..2a1b8bdcb29c 100644
--- a/include/asm-m68knommu/page.h
+++ b/include/asm-m68knommu/page.h
@@ -1,6 +1,7 @@
 #ifndef _M68KNOMMU_PAGE_H
 #define _M68KNOMMU_PAGE_H
 
+#ifdef __KERNEL__
 
 /* PAGE_SHIFT determines the page size */
 
@@ -8,8 +9,6 @@
 #define PAGE_SIZE	(1UL << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 
-#ifdef __KERNEL__
-
 #include <asm/setup.h>
 
 #ifndef __ASSEMBLY__
@@ -76,8 +75,8 @@ extern unsigned long memory_end;
 
 #endif /* __ASSEMBLY__ */
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
+
 #endif /* _M68KNOMMU_PAGE_H */
-- 
cgit v1.2.3


From 0000f0b1bc9aa340356a8dcdb9e8b1dbbaf0afcb Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 08:51:43 +0100
Subject: Fix H8300 exported headers.

Just clean up asm/page.h

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-h8300/page.h | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-h8300/page.h b/include/asm-h8300/page.h
index d673077d2fd6..3b4f2903f91d 100644
--- a/include/asm-h8300/page.h
+++ b/include/asm-h8300/page.h
@@ -1,6 +1,7 @@
 #ifndef _H8300_PAGE_H
 #define _H8300_PAGE_H
 
+#ifdef __KERNEL__
 
 /* PAGE_SHIFT determines the page size */
 
@@ -8,8 +9,6 @@
 #define PAGE_SIZE	(1UL << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 
-#ifdef __KERNEL__
-
 #include <asm/setup.h>
 
 #ifndef __ASSEMBLY__
@@ -76,9 +75,9 @@ extern unsigned long memory_end;
 
 #endif /* __ASSEMBLY__ */
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
+
 #endif /* _H8300_PAGE_H */
-- 
cgit v1.2.3


From cb16da6a71924772dc4580c16541dece2b9a8472 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 08:54:44 +0100
Subject: Remove ARM26 header export.

We ought to be able to use ARM headers; no need for special ARM26 version.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-arm26/Kbuild | 1 -
 1 file changed, 1 deletion(-)
 delete mode 100644 include/asm-arm26/Kbuild

(limited to 'include')

diff --git a/include/asm-arm26/Kbuild b/include/asm-arm26/Kbuild
deleted file mode 100644
index c68e1680da01..000000000000
--- a/include/asm-arm26/Kbuild
+++ /dev/null
@@ -1 +0,0 @@
-include include/asm-generic/Kbuild.asm
-- 
cgit v1.2.3


From d28d1f10f99f8ab2fe2bd06c3d22397d0ba08687 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 08:55:28 +0100
Subject: Remove UML header export

No need for UML to export headers for userspace to build against.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-um/Kbuild | 1 -
 1 file changed, 1 deletion(-)
 delete mode 100644 include/asm-um/Kbuild

(limited to 'include')

diff --git a/include/asm-um/Kbuild b/include/asm-um/Kbuild
deleted file mode 100644
index c68e1680da01..000000000000
--- a/include/asm-um/Kbuild
+++ /dev/null
@@ -1 +0,0 @@
-include include/asm-generic/Kbuild.asm
-- 
cgit v1.2.3


From 2a1b181eff32f497f285fcfc1e771ec469205908 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 09:05:25 +0100
Subject: Fix v850 exported headers

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-v850/page.h  | 7 ++++---
 include/asm-v850/param.h | 4 ++--
 2 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-v850/page.h b/include/asm-v850/page.h
index ad03c46a1f92..d693ffb1364d 100644
--- a/include/asm-v850/page.h
+++ b/include/asm-v850/page.h
@@ -14,6 +14,8 @@
 #ifndef __V850_PAGE_H__
 #define __V850_PAGE_H__
 
+#ifdef __KERNEL__
+
 #include <asm/machdep.h>
 
 
@@ -32,7 +34,6 @@
 #endif
 
 
-#ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
 #define STRICT_MM_TYPECHECKS
@@ -122,9 +123,9 @@ typedef unsigned long pgprot_t;
 #define __va(x)		     ((void *)__phys_to_virt ((unsigned long)(x)))
 
 
-#endif /* KERNEL */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* KERNEL */
+
 #endif /* __V850_PAGE_H__ */
diff --git a/include/asm-v850/param.h b/include/asm-v850/param.h
index 8d796e4bff52..3c65bd573782 100644
--- a/include/asm-v850/param.h
+++ b/include/asm-v850/param.h
@@ -14,8 +14,6 @@
 #ifndef __V850_PARAM_H__
 #define __V850_PARAM_H__
 
-#include <asm/machdep.h>	/* For HZ */
-
 #define EXEC_PAGESIZE	4096
 
 #ifndef NOGROUP
@@ -25,6 +23,8 @@
 #define MAXHOSTNAMELEN	64	/* max length of hostname */
 
 #ifdef __KERNEL__
+#include <asm/machdep.h>	/* For HZ */
+
 # define USER_HZ	100
 # define CLOCKS_PER_SEC	USER_HZ
 #endif
-- 
cgit v1.2.3


From ed9526b0d342a1c08a19f880c3f0c1d0fec4e8db Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Thu, 21 Sep 2006 09:33:55 +0100
Subject: Clean up exported headers on CRIS

This fixes most of the issues with exported headers on CRIS, although
we do still need to deal with the asm/arch symlink.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-cris/Kbuild          | 4 ++++
 include/asm-cris/arch-v10/Kbuild | 2 ++
 include/asm-cris/arch-v32/Kbuild | 2 ++
 include/asm-cris/byteorder.h     | 3 ++-
 include/asm-cris/elf.h           | 8 +++++---
 include/asm-cris/page.h          | 8 ++++----
 include/asm-cris/posix_types.h   | 9 +++------
 include/asm-cris/unistd.h        | 4 +---
 8 files changed, 23 insertions(+), 17 deletions(-)
 create mode 100644 include/asm-cris/arch-v10/Kbuild
 create mode 100644 include/asm-cris/arch-v32/Kbuild

(limited to 'include')

diff --git a/include/asm-cris/Kbuild b/include/asm-cris/Kbuild
index c68e1680da01..14498d5a2f65 100644
--- a/include/asm-cris/Kbuild
+++ b/include/asm-cris/Kbuild
@@ -1 +1,5 @@
 include include/asm-generic/Kbuild.asm
+
+header-y += arch-v10/ arch-v32/
+
+unifdef-y += rs485.h
diff --git a/include/asm-cris/arch-v10/Kbuild b/include/asm-cris/arch-v10/Kbuild
new file mode 100644
index 000000000000..d7f27dc0941a
--- /dev/null
+++ b/include/asm-cris/arch-v10/Kbuild
@@ -0,0 +1,2 @@
+header-y += ptrace.h
+header-y += user.h
diff --git a/include/asm-cris/arch-v32/Kbuild b/include/asm-cris/arch-v32/Kbuild
new file mode 100644
index 000000000000..d7f27dc0941a
--- /dev/null
+++ b/include/asm-cris/arch-v32/Kbuild
@@ -0,0 +1,2 @@
+header-y += ptrace.h
+header-y += user.h
diff --git a/include/asm-cris/byteorder.h b/include/asm-cris/byteorder.h
index a1a222adaa9f..0cd9db1cc888 100644
--- a/include/asm-cris/byteorder.h
+++ b/include/asm-cris/byteorder.h
@@ -3,14 +3,15 @@
 
 #ifdef __GNUC__
 
+#ifdef __KERNEL__
 #include <asm/arch/byteorder.h>
 
 /* defines are necessary because the other files detect the presence
  * of a defined __arch_swab32, not an inline
  */
-
 #define __arch__swab32(x) ___arch__swab32(x)
 #define __arch__swab16(x) ___arch__swab16(x)
+#endif /* __KERNEL__ */
 
 #if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
 #  define __BYTEORDER_HAS_U64__
diff --git a/include/asm-cris/elf.h b/include/asm-cris/elf.h
index 87a60bd8e667..96a40c1de57c 100644
--- a/include/asm-cris/elf.h
+++ b/include/asm-cris/elf.h
@@ -5,7 +5,6 @@
  * ELF register definitions..
  */
 
-#include <asm/arch/elf.h>
 #include <asm/user.h>
 
 #define R_CRIS_NONE             0
@@ -46,6 +45,9 @@ typedef unsigned long elf_fpregset_t;
 #define ELF_DATA	ELFDATA2LSB
 #define ELF_ARCH	EM_CRIS
 
+#ifdef __KERNEL__
+#include <asm/arch/elf.h>
+
 /* The master for these definitions is {binutils}/include/elf/cris.h:  */
 /* User symbols in this file have a leading underscore.  */
 #define EF_CRIS_UNDERSCORE		0x00000001
@@ -87,8 +89,8 @@ typedef unsigned long elf_fpregset_t;
 
 #define ELF_PLATFORM  (NULL)
 
-#ifdef __KERNEL__
 #define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX)
-#endif
+
+#endif /* __KERNEL__ */
 
 #endif
diff --git a/include/asm-cris/page.h b/include/asm-cris/page.h
index 81832e9e157f..9f13c32552bf 100644
--- a/include/asm-cris/page.h
+++ b/include/asm-cris/page.h
@@ -1,6 +1,8 @@
 #ifndef _CRIS_PAGE_H
 #define _CRIS_PAGE_H
 
+#ifdef __KERNEL__
+
 #include <asm/arch/page.h>
 
 /* PAGE_SHIFT determines the page size */
@@ -12,8 +14,6 @@
 #endif
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 
-#ifdef __KERNEL__
-
 #define clear_page(page)        memset((void *)(page), 0, PAGE_SIZE)
 #define copy_page(to,from)      memcpy((void *)(to), (void *)(from), PAGE_SIZE)
 
@@ -73,10 +73,10 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-#endif /* __KERNEL__ */
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+#endif /* __KERNEL__ */
+
 #endif /* _CRIS_PAGE_H */
 
diff --git a/include/asm-cris/posix_types.h b/include/asm-cris/posix_types.h
index 6d26fee4a614..7b9ed22ab5dd 100644
--- a/include/asm-cris/posix_types.h
+++ b/include/asm-cris/posix_types.h
@@ -6,8 +6,6 @@
 #ifndef __ARCH_CRIS_POSIX_TYPES_H
 #define __ARCH_CRIS_POSIX_TYPES_H
 
-#include <asm/bitops.h>
-
 /*
  * This file is generally used by user-level software, so you need to
  * be a little careful about namespace pollution etc.  Also, we cannot
@@ -53,9 +51,8 @@ typedef struct {
 #endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
 } __kernel_fsid_t;
 
-/* should this ifdef be here ?  */
-
-#if defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2)
+#ifdef __KERNEL__
+#include <asm/bitops.h>
 
 #undef	__FD_SET
 #define __FD_SET(fd,fdsetp) set_bit(fd, (void *)(fdsetp))
@@ -69,6 +66,6 @@ typedef struct {
 #undef	__FD_ZERO
 #define __FD_ZERO(fdsetp) memset((void *)(fdsetp), 0, __FDSET_LONGS << 2)
 
-#endif /* defined(__KERNEL__) || !defined(__GLIBC__) || (__GLIBC__ < 2) */
+#endif /* __KERNEL__ */
 
 #endif /* __ARCH_CRIS_POSIX_TYPES_H */
diff --git a/include/asm-cris/unistd.h b/include/asm-cris/unistd.h
index c2954e90aa24..7372efae0516 100644
--- a/include/asm-cris/unistd.h
+++ b/include/asm-cris/unistd.h
@@ -1,8 +1,6 @@
 #ifndef _ASM_CRIS_UNISTD_H_
 #define _ASM_CRIS_UNISTD_H_
 
-#include <asm/arch/unistd.h>
-
 /*
  * This file contains the system call numbers, and stub macros for libc.
  */
@@ -299,6 +297,7 @@
 
 #define NR_syscalls 289
 
+#include <asm/arch/unistd.h>
 
 #define __ARCH_WANT_IPC_PARSE_VERSION
 #define __ARCH_WANT_OLD_READDIR
@@ -322,7 +321,6 @@
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
-#endif
 
 #ifdef __KERNEL_SYSCALLS__
 
-- 
cgit v1.2.3


From 47b5c838af92d3504e99633bf568578203b7305f Mon Sep 17 00:00:00 2001
From: Linas Vepstas <linas@austin.ibm.com>
Date: Fri, 15 Sep 2006 18:57:42 -0500
Subject: [POWERPC] EEH: enable MMIO/DMA on frozen slot

Add wrapper around the rtas call to enable MMIO or DMA on a frozen pci
slot.

Signed-off-by: Linas Vepstas <linas@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/pseries/eeh.c | 29 +++++++++++++++++++++++++++++
 include/asm-powerpc/ppc-pci.h        | 11 +++++++++++
 2 files changed, 40 insertions(+)

(limited to 'include')

diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c
index fb91842fc819..4534886e3b4e 100644
--- a/arch/powerpc/platforms/pseries/eeh.c
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -481,6 +481,35 @@ eeh_slot_availability(struct pci_dn *pdn)
 	return -1;
 }
 
+/**
+ * rtas_pci_enable - enable MMIO or DMA transfers for this slot
+ * @pdn pci device node
+ */
+
+int
+rtas_pci_enable(struct pci_dn *pdn, int function)
+{
+	int config_addr;
+	int rc;
+
+	/* Use PE configuration address, if present */
+	config_addr = pdn->eeh_config_addr;
+	if (pdn->eeh_pe_config_addr)
+		config_addr = pdn->eeh_pe_config_addr;
+
+	rc = rtas_call(ibm_set_eeh_option, 4, 1, NULL,
+	               config_addr,
+	               BUID_HI(pdn->phb->buid),
+	               BUID_LO(pdn->phb->buid),
+		            function);
+
+	if (rc)
+		printk(KERN_WARNING "EEH: Cannot enable function %d, err=%d dn=%s\n",
+		        function, rc, pdn->node->full_name);
+
+	return rc;
+}
+
 /**
  * rtas_pci_slot_reset - raises/lowers the pci #RST line
  * @pdn pci device node
diff --git a/include/asm-powerpc/ppc-pci.h b/include/asm-powerpc/ppc-pci.h
index cf79bc7ebb55..1115756c79f9 100644
--- a/include/asm-powerpc/ppc-pci.h
+++ b/include/asm-powerpc/ppc-pci.h
@@ -68,6 +68,17 @@ struct pci_dev *pci_get_device_by_addr(unsigned long addr);
  */
 void eeh_slot_error_detail (struct pci_dn *pdn, int severity);
 
+/**
+ * rtas_pci_enableo - enable IO transfers for this slot
+ * @pdn:       pci device node
+ * @function:  either EEH_THAW_MMIO or EEH_THAW_DMA 
+ *
+ * Enable I/O transfers to this slot 
+ */
+#define EEH_THAW_MMIO 2
+#define EEH_THAW_DMA  3
+int rtas_pci_enable(struct pci_dn *pdn, int function);
+
 /**
  * rtas_set_slot_reset -- unfreeze a frozen slot
  *
-- 
cgit v1.2.3


From 7da8a2e5c1fd2ee513fdeac8d13c4f3623838fd0 Mon Sep 17 00:00:00 2001
From: Josh Boyer <jdub@us.ibm.com>
Date: Wed, 20 Sep 2006 09:11:59 -0500
Subject: [POWERPC] 40x: Fix debug status register defines

This fixes some debug register defines on PPC 40x that were incorrect.

Signed-off-by: Josh Boyer <jdub@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-ppc/reg_booke.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-ppc/reg_booke.h b/include/asm-ppc/reg_booke.h
index 4944c0fb8bea..602fbadeaf48 100644
--- a/include/asm-ppc/reg_booke.h
+++ b/include/asm-ppc/reg_booke.h
@@ -300,14 +300,14 @@ do {						\
 #define DBSR_IC		0x80000000	/* Instruction Completion */
 #define DBSR_BT		0x40000000	/* Branch taken */
 #define DBSR_TIE	0x10000000	/* Trap Instruction debug Event */
-#define DBSR_IAC1	0x00800000	/* Instruction Address Compare 1 Event */
-#define DBSR_IAC2	0x00400000	/* Instruction Address Compare 2 Event */
-#define DBSR_IAC3	0x00200000	/* Instruction Address Compare 3 Event */
-#define DBSR_IAC4	0x00100000	/* Instruction Address Compare 4 Event */
-#define DBSR_DAC1R	0x00080000	/* Data Address Compare 1 Read Event */
-#define DBSR_DAC1W	0x00040000	/* Data Address Compare 1 Write Event */
-#define DBSR_DAC2R	0x00020000	/* Data Address Compare 2 Read Event */
-#define DBSR_DAC2W	0x00010000	/* Data Address Compare 2 Write Event */
+#define DBSR_IAC1	0x04000000	/* Instruction Address Compare 1 Event */
+#define DBSR_IAC2	0x02000000	/* Instruction Address Compare 2 Event */
+#define DBSR_IAC3	0x00080000	/* Instruction Address Compare 3 Event */
+#define DBSR_IAC4	0x00040000	/* Instruction Address Compare 4 Event */
+#define DBSR_DAC1R	0x01000000	/* Data Address Compare 1 Read Event */
+#define DBSR_DAC1W	0x00800000	/* Data Address Compare 1 Write Event */
+#define DBSR_DAC2R	0x00400000	/* Data Address Compare 2 Read Event */
+#define DBSR_DAC2W	0x00200000	/* Data Address Compare 2 Write Event */
 #endif
 
 /* Bit definitions related to the ESR. */
-- 
cgit v1.2.3


From caf81329c39b5c48f6cc0d78fa159b5a587e37f9 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 21 Sep 2006 18:00:00 +1000
Subject: [POWERPC] Merge iSeries i/o operations with the rest

This patch changes the io operations so that they are out of line if
CONFIG_PPC_ISERIES is set and includes a firmware feature check in
that case.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/io.c                 |  14 ++
 arch/powerpc/platforms/iseries/pci.c     | 280 +++++++++++++++++++++++++------
 include/asm-powerpc/io.h                 | 148 ++++++++--------
 include/asm-powerpc/iseries/iseries_io.h |  60 -------
 4 files changed, 319 insertions(+), 183 deletions(-)
 delete mode 100644 include/asm-powerpc/iseries/iseries_io.h

(limited to 'include')

diff --git a/arch/powerpc/kernel/io.c b/arch/powerpc/kernel/io.c
index 80a3209acef4..e98180686b35 100644
--- a/arch/powerpc/kernel/io.c
+++ b/arch/powerpc/kernel/io.c
@@ -22,12 +22,16 @@
 #include <linux/module.h>
 
 #include <asm/io.h>
+#include <asm/firmware.h>
+#include <asm/bug.h>
 
 void _insb(volatile u8 __iomem *port, void *buf, long count)
 {
 	u8 *tbuf = buf;
 	u8 tmp;
 
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
 	if (unlikely(count <= 0))
 		return;
 	asm volatile("sync");
@@ -44,6 +48,8 @@ void _outsb(volatile u8 __iomem *port, const void *buf, long count)
 {
 	const u8 *tbuf = buf;
 
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
 	if (unlikely(count <= 0))
 		return;
 	asm volatile("sync");
@@ -59,6 +65,8 @@ void _insw_ns(volatile u16 __iomem *port, void *buf, long count)
 	u16 *tbuf = buf;
 	u16 tmp;
 
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
 	if (unlikely(count <= 0))
 		return;
 	asm volatile("sync");
@@ -75,6 +83,8 @@ void _outsw_ns(volatile u16 __iomem *port, const void *buf, long count)
 {
 	const u16 *tbuf = buf;
 
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
 	if (unlikely(count <= 0))
 		return;
 	asm volatile("sync");
@@ -90,6 +100,8 @@ void _insl_ns(volatile u32 __iomem *port, void *buf, long count)
 	u32 *tbuf = buf;
 	u32 tmp;
 
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
 	if (unlikely(count <= 0))
 		return;
 	asm volatile("sync");
@@ -106,6 +118,8 @@ void _outsl_ns(volatile u32 __iomem *port, const void *buf, long count)
 {
 	const u32 *tbuf = buf;
 
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
 	if (unlikely(count <= 0))
 		return;
 	asm volatile("sync");
diff --git a/arch/powerpc/platforms/iseries/pci.c b/arch/powerpc/platforms/iseries/pci.c
index f4d427a7bb2d..3eb12065df23 100644
--- a/arch/powerpc/platforms/iseries/pci.c
+++ b/arch/powerpc/platforms/iseries/pci.c
@@ -34,6 +34,7 @@
 #include <asm/pci-bridge.h>
 #include <asm/iommu.h>
 #include <asm/abs_addr.h>
+#include <asm/firmware.h>
 
 #include <asm/iseries/hv_call_xm.h>
 #include <asm/iseries/mf.h>
@@ -269,46 +270,6 @@ void pcibios_fixup_resources(struct pci_dev *pdev)
 {
 }
 
-/*
- * I/0 Memory copy MUST use mmio commands on iSeries
- * To do; For performance, include the hv call directly
- */
-void iSeries_memset_io(volatile void __iomem *dest, char c, size_t Count)
-{
-	u8 ByteValue = c;
-	long NumberOfBytes = Count;
-
-	while (NumberOfBytes > 0) {
-		iSeries_Write_Byte(ByteValue, dest++);
-		-- NumberOfBytes;
-	}
-}
-EXPORT_SYMBOL(iSeries_memset_io);
-
-void iSeries_memcpy_toio(volatile void __iomem *dest, void *source, size_t count)
-{
-	char *src = source;
-	long NumberOfBytes = count;
-
-	while (NumberOfBytes > 0) {
-		iSeries_Write_Byte(*src++, dest++);
-		-- NumberOfBytes;
-	}
-}
-EXPORT_SYMBOL(iSeries_memcpy_toio);
-
-void iSeries_memcpy_fromio(void *dest, const volatile void __iomem *src, size_t count)
-{
-	char *dst = dest;
-	long NumberOfBytes = count;
-
-	while (NumberOfBytes > 0) {
-		*dst++ = iSeries_Read_Byte(src++);
-		-- NumberOfBytes;
-	}
-}
-EXPORT_SYMBOL(iSeries_memcpy_fromio);
-
 /*
  * Look down the chain to find the matching Device Device
  */
@@ -491,7 +452,7 @@ static inline struct device_node *xlate_iomm_address(
  * iSeries_Read_Word = Read Word  (16 bit)
  * iSeries_Read_Long = Read Long  (32 bit)
  */
-u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress)
+static u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
@@ -518,9 +479,8 @@ u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress)
 
 	return (u8)ret.value;
 }
-EXPORT_SYMBOL(iSeries_Read_Byte);
 
-u16 iSeries_Read_Word(const volatile void __iomem *IoAddress)
+static u16 iSeries_Read_Word(const volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
@@ -548,9 +508,8 @@ u16 iSeries_Read_Word(const volatile void __iomem *IoAddress)
 
 	return swab16((u16)ret.value);
 }
-EXPORT_SYMBOL(iSeries_Read_Word);
 
-u32 iSeries_Read_Long(const volatile void __iomem *IoAddress)
+static u32 iSeries_Read_Long(const volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
@@ -578,7 +537,6 @@ u32 iSeries_Read_Long(const volatile void __iomem *IoAddress)
 
 	return swab32((u32)ret.value);
 }
-EXPORT_SYMBOL(iSeries_Read_Long);
 
 /*
  * Write MM I/O Instructions for the iSeries
@@ -587,7 +545,7 @@ EXPORT_SYMBOL(iSeries_Read_Long);
  * iSeries_Write_Word = Write Word(16 bit)
  * iSeries_Write_Long = Write Long(32 bit)
  */
-void iSeries_Write_Byte(u8 data, volatile void __iomem *IoAddress)
+static void iSeries_Write_Byte(u8 data, volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
@@ -612,9 +570,8 @@ void iSeries_Write_Byte(u8 data, volatile void __iomem *IoAddress)
 		rc = HvCall4(HvCallPciBarStore8, dsa, BarOffset, data, 0);
 	} while (CheckReturnCode("WWB", DevNode, &retry, rc) != 0);
 }
-EXPORT_SYMBOL(iSeries_Write_Byte);
 
-void iSeries_Write_Word(u16 data, volatile void __iomem *IoAddress)
+static void iSeries_Write_Word(u16 data, volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
@@ -639,9 +596,8 @@ void iSeries_Write_Word(u16 data, volatile void __iomem *IoAddress)
 		rc = HvCall4(HvCallPciBarStore16, dsa, BarOffset, swab16(data), 0);
 	} while (CheckReturnCode("WWW", DevNode, &retry, rc) != 0);
 }
-EXPORT_SYMBOL(iSeries_Write_Word);
 
-void iSeries_Write_Long(u32 data, volatile void __iomem *IoAddress)
+static void iSeries_Write_Long(u32 data, volatile void __iomem *IoAddress)
 {
 	u64 BarOffset;
 	u64 dsa;
@@ -666,4 +622,224 @@ void iSeries_Write_Long(u32 data, volatile void __iomem *IoAddress)
 		rc = HvCall4(HvCallPciBarStore32, dsa, BarOffset, swab32(data), 0);
 	} while (CheckReturnCode("WWL", DevNode, &retry, rc) != 0);
 }
-EXPORT_SYMBOL(iSeries_Write_Long);
+
+extern unsigned char __raw_readb(const volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return *(volatile unsigned char __force *)addr;
+}
+EXPORT_SYMBOL(__raw_readb);
+
+extern unsigned short __raw_readw(const volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return *(volatile unsigned short __force *)addr;
+}
+EXPORT_SYMBOL(__raw_readw);
+
+extern unsigned int __raw_readl(const volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return *(volatile unsigned int __force *)addr;
+}
+EXPORT_SYMBOL(__raw_readl);
+
+extern unsigned long __raw_readq(const volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return *(volatile unsigned long __force *)addr;
+}
+EXPORT_SYMBOL(__raw_readq);
+
+extern void __raw_writeb(unsigned char v, volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	*(volatile unsigned char __force *)addr = v;
+}
+EXPORT_SYMBOL(__raw_writeb);
+
+extern void __raw_writew(unsigned short v, volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	*(volatile unsigned short __force *)addr = v;
+}
+EXPORT_SYMBOL(__raw_writew);
+
+extern void __raw_writel(unsigned int v, volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	*(volatile unsigned int __force *)addr = v;
+}
+EXPORT_SYMBOL(__raw_writel);
+
+extern void __raw_writeq(unsigned long v, volatile void __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	*(volatile unsigned long __force *)addr = v;
+}
+EXPORT_SYMBOL(__raw_writeq);
+
+int in_8(const volatile unsigned char __iomem *addr)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES))
+		return iSeries_Read_Byte(addr);
+	return __in_8(addr);
+}
+EXPORT_SYMBOL(in_8);
+
+void out_8(volatile unsigned char __iomem *addr, int val)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES))
+		iSeries_Write_Byte(val, addr);
+	else
+		__out_8(addr, val);
+}
+EXPORT_SYMBOL(out_8);
+
+int in_le16(const volatile unsigned short __iomem *addr)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES))
+		return iSeries_Read_Word(addr);
+	return __in_le16(addr);
+}
+EXPORT_SYMBOL(in_le16);
+
+int in_be16(const volatile unsigned short __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return __in_be16(addr);
+}
+EXPORT_SYMBOL(in_be16);
+
+void out_le16(volatile unsigned short __iomem *addr, int val)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES))
+		iSeries_Write_Word(val, addr);
+	else
+		__out_le16(addr, val);
+}
+EXPORT_SYMBOL(out_le16);
+
+void out_be16(volatile unsigned short __iomem *addr, int val)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	__out_be16(addr, val);
+}
+EXPORT_SYMBOL(out_be16);
+
+unsigned in_le32(const volatile unsigned __iomem *addr)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES))
+		return iSeries_Read_Long(addr);
+	return __in_le32(addr);
+}
+EXPORT_SYMBOL(in_le32);
+
+unsigned in_be32(const volatile unsigned __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return __in_be32(addr);
+}
+EXPORT_SYMBOL(in_be32);
+
+void out_le32(volatile unsigned __iomem *addr, int val)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES))
+		iSeries_Write_Long(val, addr);
+	else
+		__out_le32(addr, val);
+}
+EXPORT_SYMBOL(out_le32);
+
+void out_be32(volatile unsigned __iomem *addr, int val)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	__out_be32(addr, val);
+}
+EXPORT_SYMBOL(out_be32);
+
+unsigned long in_le64(const volatile unsigned long __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return __in_le64(addr);
+}
+EXPORT_SYMBOL(in_le64);
+
+unsigned long in_be64(const volatile unsigned long __iomem *addr)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	return __in_be64(addr);
+}
+EXPORT_SYMBOL(in_be64);
+
+void out_le64(volatile unsigned long __iomem *addr, unsigned long val)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	__out_le64(addr, val);
+}
+EXPORT_SYMBOL(out_le64);
+
+void out_be64(volatile unsigned long __iomem *addr, unsigned long val)
+{
+	BUG_ON(firmware_has_feature(FW_FEATURE_ISERIES));
+
+	__out_be64(addr, val);
+}
+EXPORT_SYMBOL(out_be64);
+
+void memset_io(volatile void __iomem *addr, int c, unsigned long n)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES)) {
+		volatile char __iomem *d = addr;
+
+		while (n-- > 0) {
+			iSeries_Write_Byte(c, d++);
+		}
+	} else
+		eeh_memset_io(addr, c, n);
+}
+EXPORT_SYMBOL(memset_io);
+
+void memcpy_fromio(void *dest, const volatile void __iomem *src,
+                                 unsigned long n)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES)) {
+		char *d = dest;
+		const volatile char __iomem *s = src;
+
+		while (n-- > 0) {
+			*d++ = iSeries_Read_Byte(s++);
+		}
+	} else
+		eeh_memcpy_fromio(dest, src, n);
+}
+EXPORT_SYMBOL(memcpy_fromio);
+
+void memcpy_toio(volatile void __iomem *dest, const void *src, unsigned long n)
+{
+	if (firmware_has_feature(FW_FEATURE_ISERIES)) {
+		const char *s = src;
+		volatile char __iomem *d = dest;
+
+		while (n-- > 0) {
+			iSeries_Write_Byte(*s++, d++);
+		}
+	} else
+		eeh_memcpy_toio(dest, src, n);
+}
+EXPORT_SYMBOL(memcpy_toio);
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 174fb89d5eda..46bae1cf385b 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -20,9 +20,6 @@ extern int check_legacy_ioport(unsigned long base_port);
 #include <asm/page.h>
 #include <asm/byteorder.h>
 #include <asm/paca.h>
-#ifdef CONFIG_PPC_ISERIES 
-#include <asm/iseries/iseries_io.h>
-#endif  
 #include <asm/synch.h>
 #include <asm/delay.h>
 
@@ -37,41 +34,53 @@ extern unsigned long isa_io_base;
 extern unsigned long pci_io_base;
 
 #ifdef CONFIG_PPC_ISERIES
-/* __raw_* accessors aren't supported on iSeries */
-#define __raw_readb(addr)	{ BUG(); 0; }
-#define __raw_readw(addr)       { BUG(); 0; }
-#define __raw_readl(addr)       { BUG(); 0; }
-#define __raw_readq(addr)       { BUG(); 0; }
-#define __raw_writeb(v, addr)   { BUG(); 0; }
-#define __raw_writew(v, addr)   { BUG(); 0; }
-#define __raw_writel(v, addr)   { BUG(); 0; }
-#define __raw_writeq(v, addr)   { BUG(); 0; }
-#define readb(addr)		iSeries_Read_Byte(addr)
-#define readw(addr)		iSeries_Read_Word(addr)
-#define readl(addr)		iSeries_Read_Long(addr)
-#define writeb(data, addr)	iSeries_Write_Byte((data),(addr))
-#define writew(data, addr)	iSeries_Write_Word((data),(addr))
-#define writel(data, addr)	iSeries_Write_Long((data),(addr))
-#define memset_io(a,b,c)	iSeries_memset_io((a),(b),(c))
-#define memcpy_fromio(a,b,c)	iSeries_memcpy_fromio((a), (b), (c))
-#define memcpy_toio(a,b,c)	iSeries_memcpy_toio((a), (b), (c))
-
-#define inb(addr)		readb(((void __iomem *)(long)(addr)))
-#define inw(addr)		readw(((void __iomem *)(long)(addr)))
-#define inl(addr)		readl(((void __iomem *)(long)(addr)))
-#define outb(data,addr)		writeb(data,((void __iomem *)(long)(addr)))
-#define outw(data,addr)		writew(data,((void __iomem *)(long)(addr)))
-#define outl(data,addr)		writel(data,((void __iomem *)(long)(addr)))
-/*
- * The *_ns versions below don't do byte-swapping.
- * Neither do the standard versions now, these are just here
- * for older code.
- */
-#define insb(port, buf, ns)	_insb((u8 __iomem *)((port)+pci_io_base), (buf), (ns))
-#define insw(port, buf, ns)	_insw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns))
-#define insl(port, buf, nl)	_insl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl))
 
-#else
+extern int in_8(const volatile unsigned char __iomem *addr);
+extern void out_8(volatile unsigned char __iomem *addr, int val);
+extern int in_le16(const volatile unsigned short __iomem *addr);
+extern int in_be16(const volatile unsigned short __iomem *addr);
+extern void out_le16(volatile unsigned short __iomem *addr, int val);
+extern void out_be16(volatile unsigned short __iomem *addr, int val);
+extern unsigned in_le32(const volatile unsigned __iomem *addr);
+extern unsigned in_be32(const volatile unsigned __iomem *addr);
+extern void out_le32(volatile unsigned __iomem *addr, int val);
+extern void out_be32(volatile unsigned __iomem *addr, int val);
+extern unsigned long in_le64(const volatile unsigned long __iomem *addr);
+extern unsigned long in_be64(const volatile unsigned long __iomem *addr);
+extern void out_le64(volatile unsigned long __iomem *addr, unsigned long val);
+extern void out_be64(volatile unsigned long __iomem *addr, unsigned long val);
+
+extern unsigned char __raw_readb(const volatile void __iomem *addr);
+extern unsigned short __raw_readw(const volatile void __iomem *addr);
+extern unsigned int __raw_readl(const volatile void __iomem *addr);
+extern unsigned long __raw_readq(const volatile void __iomem *addr);
+extern void __raw_writeb(unsigned char v, volatile void __iomem *addr);
+extern void __raw_writew(unsigned short v, volatile void __iomem *addr);
+extern void __raw_writel(unsigned int v, volatile void __iomem *addr);
+extern void __raw_writeq(unsigned long v, volatile void __iomem *addr);
+
+extern void memset_io(volatile void __iomem *addr, int c, unsigned long n);
+extern void memcpy_fromio(void *dest, const volatile void __iomem *src,
+                                 unsigned long n);
+extern void memcpy_toio(volatile void __iomem *dest, const void *src,
+                                 unsigned long n);
+
+#else /* CONFIG_PPC_ISERIES */
+
+#define in_8(addr)		__in_8((addr))
+#define out_8(addr, val)	__out_8((addr), (val))
+#define in_le16(addr)		__in_le16((addr))
+#define in_be16(addr)		__in_be16((addr))
+#define out_le16(addr, val)	__out_le16((addr), (val))
+#define out_be16(addr, val)	__out_be16((addr), (val))
+#define in_le32(addr)		__in_le32((addr))
+#define in_be32(addr)		__in_be32((addr))
+#define out_le32(addr, val)	__out_le32((addr), (val))
+#define out_be32(addr, val)	__out_be32((addr), (val))
+#define in_le64(addr)		__in_le64((addr))
+#define in_be64(addr)		__in_be64((addr))
+#define out_le64(addr, val)	__out_le64((addr), (val))
+#define out_be64(addr, val)	__out_be64((addr), (val))
 
 static inline unsigned char __raw_readb(const volatile void __iomem *addr)
 {
@@ -105,23 +114,11 @@ static inline void __raw_writeq(unsigned long v, volatile void __iomem *addr)
 {
 	*(volatile unsigned long __force *)addr = v;
 }
-#define readb(addr)		eeh_readb(addr)
-#define readw(addr)		eeh_readw(addr)
-#define readl(addr)		eeh_readl(addr)
-#define readq(addr)		eeh_readq(addr)
-#define writeb(data, addr)	eeh_writeb((data), (addr))
-#define writew(data, addr)	eeh_writew((data), (addr))
-#define writel(data, addr)	eeh_writel((data), (addr))
-#define writeq(data, addr)	eeh_writeq((data), (addr))
 #define memset_io(a,b,c)	eeh_memset_io((a),(b),(c))
 #define memcpy_fromio(a,b,c)	eeh_memcpy_fromio((a),(b),(c))
 #define memcpy_toio(a,b,c)	eeh_memcpy_toio((a),(b),(c))
-#define inb(port)		eeh_inb((unsigned long)port)
-#define outb(val, port)		eeh_outb(val, (unsigned long)port)
-#define inw(port)		eeh_inw((unsigned long)port)
-#define outw(val, port)		eeh_outw(val, (unsigned long)port)
-#define inl(port)		eeh_inl((unsigned long)port)
-#define outl(val, port)		eeh_outl(val, (unsigned long)port)
+
+#endif /* CONFIG_PPC_ISERIES */
 
 /*
  * The insw/outsw/insl/outsl macros don't do byte-swapping.
@@ -132,12 +129,25 @@ static inline void __raw_writeq(unsigned long v, volatile void __iomem *addr)
 #define insw(port, buf, ns)	eeh_insw_ns((port), (buf), (ns))
 #define insl(port, buf, nl)	eeh_insl_ns((port), (buf), (nl))
 
-#endif
-
 #define outsb(port, buf, ns)  _outsb((u8 __iomem *)((port)+pci_io_base), (buf), (ns))
 #define outsw(port, buf, ns)  _outsw_ns((u16 __iomem *)((port)+pci_io_base), (buf), (ns))
 #define outsl(port, buf, nl)  _outsl_ns((u32 __iomem *)((port)+pci_io_base), (buf), (nl))
 
+#define readb(addr)		eeh_readb(addr)
+#define readw(addr)		eeh_readw(addr)
+#define readl(addr)		eeh_readl(addr)
+#define readq(addr)		eeh_readq(addr)
+#define writeb(data, addr)	eeh_writeb((data), (addr))
+#define writew(data, addr)	eeh_writew((data), (addr))
+#define writel(data, addr)	eeh_writel((data), (addr))
+#define writeq(data, addr)	eeh_writeq((data), (addr))
+#define inb(port)		eeh_inb((unsigned long)port)
+#define outb(val, port)		eeh_outb(val, (unsigned long)port)
+#define inw(port)		eeh_inw((unsigned long)port)
+#define outw(val, port)		eeh_outw(val, (unsigned long)port)
+#define inl(port)		eeh_inl((unsigned long)port)
+#define outl(val, port)		eeh_outl(val, (unsigned long)port)
+
 #define readb_relaxed(addr) readb(addr)
 #define readw_relaxed(addr) readw(addr)
 #define readl_relaxed(addr) readl(addr)
@@ -258,7 +268,7 @@ static inline void iosync(void)
  * and should not be used directly by device drivers.  Use inb/readb
  * instead.
  */
-static inline int in_8(const volatile unsigned char __iomem *addr)
+static inline int __in_8(const volatile unsigned char __iomem *addr)
 {
 	int ret;
 
@@ -267,14 +277,14 @@ static inline int in_8(const volatile unsigned char __iomem *addr)
 	return ret;
 }
 
-static inline void out_8(volatile unsigned char __iomem *addr, int val)
+static inline void __out_8(volatile unsigned char __iomem *addr, int val)
 {
 	__asm__ __volatile__("sync; stb%U0%X0 %1,%0"
 			     : "=m" (*addr) : "r" (val));
 	get_paca()->io_sync = 1;
 }
 
-static inline int in_le16(const volatile unsigned short __iomem *addr)
+static inline int __in_le16(const volatile unsigned short __iomem *addr)
 {
 	int ret;
 
@@ -283,7 +293,7 @@ static inline int in_le16(const volatile unsigned short __iomem *addr)
 	return ret;
 }
 
-static inline int in_be16(const volatile unsigned short __iomem *addr)
+static inline int __in_be16(const volatile unsigned short __iomem *addr)
 {
 	int ret;
 
@@ -292,21 +302,21 @@ static inline int in_be16(const volatile unsigned short __iomem *addr)
 	return ret;
 }
 
-static inline void out_le16(volatile unsigned short __iomem *addr, int val)
+static inline void __out_le16(volatile unsigned short __iomem *addr, int val)
 {
 	__asm__ __volatile__("sync; sthbrx %1,0,%2"
 			     : "=m" (*addr) : "r" (val), "r" (addr));
 	get_paca()->io_sync = 1;
 }
 
-static inline void out_be16(volatile unsigned short __iomem *addr, int val)
+static inline void __out_be16(volatile unsigned short __iomem *addr, int val)
 {
 	__asm__ __volatile__("sync; sth%U0%X0 %1,%0"
 			     : "=m" (*addr) : "r" (val));
 	get_paca()->io_sync = 1;
 }
 
-static inline unsigned in_le32(const volatile unsigned __iomem *addr)
+static inline unsigned __in_le32(const volatile unsigned __iomem *addr)
 {
 	unsigned ret;
 
@@ -315,7 +325,7 @@ static inline unsigned in_le32(const volatile unsigned __iomem *addr)
 	return ret;
 }
 
-static inline unsigned in_be32(const volatile unsigned __iomem *addr)
+static inline unsigned __in_be32(const volatile unsigned __iomem *addr)
 {
 	unsigned ret;
 
@@ -324,21 +334,21 @@ static inline unsigned in_be32(const volatile unsigned __iomem *addr)
 	return ret;
 }
 
-static inline void out_le32(volatile unsigned __iomem *addr, int val)
+static inline void __out_le32(volatile unsigned __iomem *addr, int val)
 {
 	__asm__ __volatile__("sync; stwbrx %1,0,%2" : "=m" (*addr)
 			     : "r" (val), "r" (addr));
 	get_paca()->io_sync = 1;
 }
 
-static inline void out_be32(volatile unsigned __iomem *addr, int val)
+static inline void __out_be32(volatile unsigned __iomem *addr, int val)
 {
 	__asm__ __volatile__("sync; stw%U0%X0 %1,%0"
 			     : "=m" (*addr) : "r" (val));
 	get_paca()->io_sync = 1;
 }
 
-static inline unsigned long in_le64(const volatile unsigned long __iomem *addr)
+static inline unsigned long __in_le64(const volatile unsigned long __iomem *addr)
 {
 	unsigned long tmp, ret;
 
@@ -358,7 +368,7 @@ static inline unsigned long in_le64(const volatile unsigned long __iomem *addr)
 	return ret;
 }
 
-static inline unsigned long in_be64(const volatile unsigned long __iomem *addr)
+static inline unsigned long __in_be64(const volatile unsigned long __iomem *addr)
 {
 	unsigned long ret;
 
@@ -367,7 +377,7 @@ static inline unsigned long in_be64(const volatile unsigned long __iomem *addr)
 	return ret;
 }
 
-static inline void out_le64(volatile unsigned long __iomem *addr, unsigned long val)
+static inline void __out_le64(volatile unsigned long __iomem *addr, unsigned long val)
 {
 	unsigned long tmp;
 
@@ -385,15 +395,13 @@ static inline void out_le64(volatile unsigned long __iomem *addr, unsigned long
 	get_paca()->io_sync = 1;
 }
 
-static inline void out_be64(volatile unsigned long __iomem *addr, unsigned long val)
+static inline void __out_be64(volatile unsigned long __iomem *addr, unsigned long val)
 {
 	__asm__ __volatile__("sync; std%U0%X0 %1,%0" : "=m" (*addr) : "r" (val));
 	get_paca()->io_sync = 1;
 }
 
-#ifndef CONFIG_PPC_ISERIES 
 #include <asm/eeh.h>
-#endif
 
 /**
  *	check_signature		-	find BIOS signatures
@@ -409,7 +417,6 @@ static inline int check_signature(const volatile void __iomem * io_addr,
 	const unsigned char *signature, int length)
 {
 	int retval = 0;
-#ifndef CONFIG_PPC_ISERIES 
 	do {
 		if (readb(io_addr) != *signature)
 			goto out;
@@ -419,7 +426,6 @@ static inline int check_signature(const volatile void __iomem * io_addr,
 	} while (length);
 	retval = 1;
 out:
-#endif
 	return retval;
 }
 
diff --git a/include/asm-powerpc/iseries/iseries_io.h b/include/asm-powerpc/iseries/iseries_io.h
deleted file mode 100644
index f29009bd63c9..000000000000
--- a/include/asm-powerpc/iseries/iseries_io.h
+++ /dev/null
@@ -1,60 +0,0 @@
-#ifndef _ASM_POWERPC_ISERIES_ISERIES_IO_H
-#define _ASM_POWERPC_ISERIES_ISERIES_IO_H
-
-
-#ifdef CONFIG_PPC_ISERIES
-#include <linux/types.h>
-/*
- * Created by Allan Trautman on Thu Dec 28 2000.
- *
- * Remaps the io.h for the iSeries Io
- * Copyright (C) 2000  Allan H Trautman, IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the:
- * Free Software Foundation, Inc.,
- * 59 Temple Place, Suite 330,
- * Boston, MA  02111-1307  USA
- *
- * Change Activity:
- *   Created December 28, 2000
- * End Change Activity
- */
-
-#ifdef CONFIG_PCI
-extern u8   iSeries_Read_Byte(const volatile void __iomem * IoAddress);
-extern u16  iSeries_Read_Word(const volatile void __iomem * IoAddress);
-extern u32  iSeries_Read_Long(const volatile void __iomem * IoAddress);
-extern void iSeries_Write_Byte(u8  IoData, volatile void __iomem * IoAddress);
-extern void iSeries_Write_Word(u16 IoData, volatile void __iomem * IoAddress);
-extern void iSeries_Write_Long(u32 IoData, volatile void __iomem * IoAddress);
-
-extern void iSeries_memset_io(volatile void __iomem *dest, char x, size_t n);
-extern void iSeries_memcpy_toio(volatile void __iomem *dest, void *source,
-		size_t n);
-extern void iSeries_memcpy_fromio(void *dest,
-		const volatile void __iomem *source, size_t n);
-#else
-static inline u8 iSeries_Read_Byte(const volatile void __iomem *IoAddress)
-{
-	return 0xff;
-}
-
-static inline void iSeries_Write_Byte(u8 IoData,
-		volatile void __iomem *IoAddress)
-{
-}
-#endif	/* CONFIG_PCI */
-
-#endif /* CONFIG_PPC_ISERIES */
-#endif /* _ASM_POWERPC_ISERIES_ISERIES_IO_H */
-- 
cgit v1.2.3


From 2954da897c40de0f3abdd6a100f2978f30d04068 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 21 Sep 2006 18:21:35 +1000
Subject: [POWERPC] Remove DISCONTIGMEM cruft from page.h

This looks like cruft to me, these functions don't exist AFAICT,
and I can't see that it's possible to even enable DISCONTIGMEM on
powerpc anymore. CC'ing some folks who might know better, based on
the who-touched-it-last principle.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/page.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-powerpc/page.h b/include/asm-powerpc/page.h
index fb597b37c2a2..b4d38b0b15f8 100644
--- a/include/asm-powerpc/page.h
+++ b/include/asm-powerpc/page.h
@@ -55,12 +55,6 @@
 #define PAGE_OFFSET     ASM_CONST(CONFIG_KERNEL_START)
 #define KERNELBASE      (PAGE_OFFSET + PHYSICAL_START)
 
-#ifdef CONFIG_DISCONTIGMEM
-#define page_to_pfn(page)	discontigmem_page_to_pfn(page)
-#define pfn_to_page(pfn)	discontigmem_pfn_to_page(pfn)
-#define pfn_valid(pfn)		discontigmem_pfn_valid(pfn)
-#endif
-
 #ifdef CONFIG_FLATMEM
 #define pfn_valid(pfn)		((pfn) < max_mapnr)
 #endif
-- 
cgit v1.2.3


From 7d452c326c2ac879aced884411a0fe3ba75d9c87 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Date: Thu, 21 Sep 2006 12:29:51 +0200
Subject: [POWERPC] powerpc: fix building gdb against asm/ptrace.h

Ulrich Weigand found a bug with the current version of the
asm-powerpc/ptrace.h that prevents building at least the
SPU target version of gdb, since some ptrace opcodes are
not defined.

The problem seems to have originated in the merging of 32 and
64 bit versions of that file, the problem is that some opcodes
are only valid on 64 bit kernels, but are also used by 32 bit
programs, so they can't depends on the __powerpc64__ symbol.

Signed-off-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/ptrace.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-powerpc/ptrace.h b/include/asm-powerpc/ptrace.h
index dc4cb9cc73a1..4435efe85d0e 100644
--- a/include/asm-powerpc/ptrace.h
+++ b/include/asm-powerpc/ptrace.h
@@ -215,12 +215,10 @@ do {									      \
 #define PTRACE_GETVRREGS	18
 #define PTRACE_SETVRREGS	19
 
-#ifndef __powerpc64__
 /* Get/set all the upper 32-bits of the SPE registers, accumulator, and
  * spefscr, in one go */
 #define PTRACE_GETEVRREGS	20
 #define PTRACE_SETEVRREGS	21
-#endif /* __powerpc64__ */
 
 /*
  * Get or set a debug register. The first 16 are DABR registers and the
@@ -235,7 +233,6 @@ do {									      \
 #define PPC_PTRACE_GETFPREGS	0x97	/* Get FPRs 0 - 31 */
 #define PPC_PTRACE_SETFPREGS	0x96	/* Set FPRs 0 - 31 */
 
-#ifdef __powerpc64__
 /* Calls to trace a 64bit program from a 32bit program */
 #define PPC_PTRACE_PEEKTEXT_3264 0x95
 #define PPC_PTRACE_PEEKDATA_3264 0x94
@@ -243,6 +240,5 @@ do {									      \
 #define PPC_PTRACE_POKEDATA_3264 0x92
 #define PPC_PTRACE_PEEKUSR_3264  0x91
 #define PPC_PTRACE_POKEUSR_3264  0x90
-#endif /* __powerpc64__ */
 
 #endif /* _ASM_POWERPC_PTRACE_H */
-- 
cgit v1.2.3


From 1694176a210189312e31b083bac1e1688981219a Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Fri, 22 Sep 2006 08:00:42 +0100
Subject: Remove offsetof() from user-visible <linux/stddef.h>

It's not used by anything user-visible, and it make g++ unhappy.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/linux/Kbuild   | 2 +-
 include/linux/stddef.h | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 7d076d97b2f7..10d2ca07562a 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -143,7 +143,6 @@ header-y += snmp.h
 header-y += sockios.h
 header-y += som.h
 header-y += sound.h
-header-y += stddef.h
 header-y += synclink.h
 header-y += telephony.h
 header-y += termios.h
@@ -318,6 +317,7 @@ unifdef-y += sonet.h
 unifdef-y += sonypi.h
 unifdef-y += soundcard.h
 unifdef-y += stat.h
+unifdef-y += stddef.h
 unifdef-y += sysctl.h
 unifdef-y += tcp.h
 unifdef-y += time.h
diff --git a/include/linux/stddef.h b/include/linux/stddef.h
index b3a2cadf90f2..ea65dfb60cd8 100644
--- a/include/linux/stddef.h
+++ b/include/linux/stddef.h
@@ -10,11 +10,13 @@
 #define NULL ((void *)0)
 #endif
 
+#ifdef __KERNEL__
 #undef offsetof
 #ifdef __compiler_offsetof
 #define offsetof(TYPE,MEMBER) __compiler_offsetof(TYPE,MEMBER)
 #else
 #define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
 #endif
+#endif /* __KERNEL__ */
 
 #endif
-- 
cgit v1.2.3


From 892c141e62982272b9c738b5520ad0e5e1ad7b42 Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Fri, 4 Aug 2006 23:08:56 -0700
Subject: [MLSXFRM]: Add security sid to sock

This adds security for IP sockets at the sock level. Security at the
sock level is needed to enforce the SELinux security policy for
security associations even when a sock is orphaned (such as in the TCP
LAST_ACK state).

This will also be used to enforce SELinux controls over data arriving
at or leaving a child socket while it's still waiting to be accepted.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h          | 12 ++++++++++++
 include/net/sock.h                | 13 +++++++++++++
 net/core/sock.c                   |  2 +-
 security/dummy.c                  |  5 +++++
 security/selinux/hooks.c          | 38 +++++++++++++++++++++-----------------
 security/selinux/include/objsec.h |  1 +
 6 files changed, 53 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 6bc2aad494ff..4d7fb59996b0 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -812,6 +812,8 @@ struct swap_info_struct;
  *      which is used to copy security attributes between local stream sockets.
  * @sk_free_security:
  *	Deallocate security structure.
+ * @sk_clone_security:
+ *	Clone/copy security structure.
  * @sk_getsid:
  *	Retrieve the LSM-specific sid for the sock to enable caching of network
  *	authorizations.
@@ -1332,6 +1334,7 @@ struct security_operations {
 	int (*socket_getpeersec_dgram) (struct socket *sock, struct sk_buff *skb, u32 *secid);
 	int (*sk_alloc_security) (struct sock *sk, int family, gfp_t priority);
 	void (*sk_free_security) (struct sock *sk);
+	void (*sk_clone_security) (const struct sock *sk, struct sock *newsk);
 	unsigned int (*sk_getsid) (struct sock *sk, struct flowi *fl, u8 dir);
 #endif	/* CONFIG_SECURITY_NETWORK */
 
@@ -2885,6 +2888,11 @@ static inline void security_sk_free(struct sock *sk)
 	return security_ops->sk_free_security(sk);
 }
 
+static inline void security_sk_clone(const struct sock *sk, struct sock *newsk)
+{
+	return security_ops->sk_clone_security(sk, newsk);
+}
+
 static inline unsigned int security_sk_sid(struct sock *sk, struct flowi *fl, u8 dir)
 {
 	return security_ops->sk_getsid(sk, fl, dir);
@@ -3011,6 +3019,10 @@ static inline void security_sk_free(struct sock *sk)
 {
 }
 
+static inline void security_sk_clone(const struct sock *sk, struct sock *newsk)
+{
+}
+
 static inline unsigned int security_sk_sid(struct sock *sk, struct flowi *fl, u8 dir)
 {
 	return 0;
diff --git a/include/net/sock.h b/include/net/sock.h
index 324b3ea233d6..91cdceb3c028 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -972,6 +972,19 @@ static inline void sock_graft(struct sock *sk, struct socket *parent)
 	write_unlock_bh(&sk->sk_callback_lock);
 }
 
+static inline void sock_copy(struct sock *nsk, const struct sock *osk)
+{
+#ifdef CONFIG_SECURITY_NETWORK
+	void *sptr = nsk->sk_security;
+#endif
+
+	memcpy(nsk, osk, osk->sk_prot->obj_size);
+#ifdef CONFIG_SECURITY_NETWORK
+	nsk->sk_security = sptr;
+	security_sk_clone(osk, nsk);
+#endif
+}
+
 extern int sock_i_uid(struct sock *sk);
 extern unsigned long sock_i_ino(struct sock *sk);
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 51fcfbc041a7..b67d868649cd 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -911,7 +911,7 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 	if (newsk != NULL) {
 		struct sk_filter *filter;
 
-		memcpy(newsk, sk, sk->sk_prot->obj_size);
+		sock_copy(newsk, sk);
 
 		/* SANITY */
 		sk_node_init(&newsk->sk_node);
diff --git a/security/dummy.c b/security/dummy.c
index 58c6d399c844..bd3bc5faa9a8 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -805,6 +805,10 @@ static inline void dummy_sk_free_security (struct sock *sk)
 {
 }
 
+static inline void dummy_sk_clone_security (const struct sock *sk, struct sock *newsk)
+{
+}
+
 static unsigned int dummy_sk_getsid(struct sock *sk, struct flowi *fl, u8 dir)
 {
 	return 0;
@@ -1060,6 +1064,7 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, socket_getpeersec_dgram);
 	set_to_dummy_if_null(ops, sk_alloc_security);
 	set_to_dummy_if_null(ops, sk_free_security);
+	set_to_dummy_if_null(ops, sk_clone_security);
 	set_to_dummy_if_null(ops, sk_getsid);
  #endif	/* CONFIG_SECURITY_NETWORK */
 #ifdef  CONFIG_SECURITY_NETWORK_XFRM
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 5d1b8c733199..d67abf77584a 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -269,15 +269,13 @@ static int sk_alloc_security(struct sock *sk, int family, gfp_t priority)
 {
 	struct sk_security_struct *ssec;
 
-	if (family != PF_UNIX)
-		return 0;
-
 	ssec = kzalloc(sizeof(*ssec), priority);
 	if (!ssec)
 		return -ENOMEM;
 
 	ssec->sk = sk;
 	ssec->peer_sid = SECINITSID_UNLABELED;
+	ssec->sid = SECINITSID_UNLABELED;
 	sk->sk_security = ssec;
 
 	return 0;
@@ -287,9 +285,6 @@ static void sk_free_security(struct sock *sk)
 {
 	struct sk_security_struct *ssec = sk->sk_security;
 
-	if (sk->sk_family != PF_UNIX)
-		return;
-
 	sk->sk_security = NULL;
 	kfree(ssec);
 }
@@ -3068,6 +3063,7 @@ static void selinux_socket_post_create(struct socket *sock, int family,
 {
 	struct inode_security_struct *isec;
 	struct task_security_struct *tsec;
+	struct sk_security_struct *sksec;
 	u32 newsid;
 
 	isec = SOCK_INODE(sock)->i_security;
@@ -3078,6 +3074,11 @@ static void selinux_socket_post_create(struct socket *sock, int family,
 	isec->sid = kern ? SECINITSID_KERNEL : newsid;
 	isec->initialized = 1;
 
+	if (sock->sk) {
+		sksec = sock->sk->sk_security;
+		sksec->sid = isec->sid;
+	}
+
 	return;
 }
 
@@ -3551,22 +3552,24 @@ static void selinux_sk_free_security(struct sock *sk)
 	sk_free_security(sk);
 }
 
-static unsigned int selinux_sk_getsid_security(struct sock *sk, struct flowi *fl, u8 dir)
+static void selinux_sk_clone_security(const struct sock *sk, struct sock *newsk)
 {
-	struct inode_security_struct *isec;
-	u32 sock_sid = SECINITSID_ANY_SOCKET;
+	struct sk_security_struct *ssec = sk->sk_security;
+	struct sk_security_struct *newssec = newsk->sk_security;
 
+	newssec->sid = ssec->sid;
+	newssec->peer_sid = ssec->peer_sid;
+}
+
+static unsigned int selinux_sk_getsid_security(struct sock *sk, struct flowi *fl, u8 dir)
+{
 	if (!sk)
 		return selinux_no_sk_sid(fl);
+	else {
+		struct sk_security_struct *sksec = sk->sk_security;
 
-	read_lock_bh(&sk->sk_callback_lock);
-	isec = get_sock_isec(sk);
-
-	if (isec)
-		sock_sid = isec->sid;
-
-	read_unlock_bh(&sk->sk_callback_lock);
-	return sock_sid;
+		return sksec->sid;
+	}
 }
 
 static int selinux_nlmsg_perm(struct sock *sk, struct sk_buff *skb)
@@ -4618,6 +4621,7 @@ static struct security_operations selinux_ops = {
 	.socket_getpeersec_dgram =	selinux_socket_getpeersec_dgram,
 	.sk_alloc_security =		selinux_sk_alloc_security,
 	.sk_free_security =		selinux_sk_free_security,
+	.sk_clone_security =		selinux_sk_clone_security,
 	.sk_getsid = 			selinux_sk_getsid_security,
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 940178865fc7..79b9e0af19a0 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -99,6 +99,7 @@ struct netif_security_struct {
 
 struct sk_security_struct {
 	struct sock *sk;		/* back pointer to sk object */
+	u32 sid;			/* SID of this object */
 	u32 peer_sid;			/* SID of peer */
 };
 
-- 
cgit v1.2.3


From b6340fcd761acf9249b3acbc95c4dc555d9beb07 Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Mon, 24 Jul 2006 23:28:37 -0700
Subject: [MLSXFRM]: Add security sid to flowi

This adds security to flow key for labeling of flows as also to allow
for making flow cache lookups based on the security label seemless.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/secid.txt | 14 ++++++++++++++
 include/net/flow.h                 |  1 +
 2 files changed, 15 insertions(+)
 create mode 100644 Documentation/networking/secid.txt

(limited to 'include')

diff --git a/Documentation/networking/secid.txt b/Documentation/networking/secid.txt
new file mode 100644
index 000000000000..95ea06784333
--- /dev/null
+++ b/Documentation/networking/secid.txt
@@ -0,0 +1,14 @@
+flowi structure:
+
+The secid member in the flow structure is used in LSMs (e.g. SELinux) to indicate
+the label of the flow. This label of the flow is currently used in selecting
+matching labeled xfrm(s).
+
+If this is an outbound flow, the label is derived from the socket, if any, or
+the incoming packet this flow is being generated as a response to (e.g. tcp
+resets, timewait ack, etc.). It is also conceivable that the label could be
+derived from other sources such as process context, device, etc., in special
+cases, as may be appropriate.
+
+If this is an inbound flow, the label is derived from the IPSec security
+associations, if any, used by the packet.
diff --git a/include/net/flow.h b/include/net/flow.h
index 04d89f763451..1cee5a83433a 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -78,6 +78,7 @@ struct flowi {
 #define fl_icmp_type	uli_u.icmpt.type
 #define fl_icmp_code	uli_u.icmpt.code
 #define fl_ipsec_spi	uli_u.spi
+	__u32           secid;	/* used by xfrm; see secid.txt */
 } __attribute__((__aligned__(BITS_PER_LONG/8)));
 
 #define FLOW_DIR_IN	0
-- 
cgit v1.2.3


From e0d1caa7b0d5f02e4f34aa09c695d04251310c6c Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Mon, 24 Jul 2006 23:29:07 -0700
Subject: [MLSXFRM]: Flow based matching of xfrm policy and state

This implements a seemless mechanism for xfrm policy selection and
state matching based on the flow sid. This also includes the necessary
SELinux enforcement pieces.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h        | 106 +++++++++++++++++----
 include/net/flow.h              |   4 +-
 net/core/flow.c                 |   7 +-
 net/xfrm/xfrm_policy.c          |  28 +++---
 net/xfrm/xfrm_state.c           |  12 ++-
 security/dummy.c                |  23 ++++-
 security/selinux/hooks.c        |   7 +-
 security/selinux/include/xfrm.h |  23 +++--
 security/selinux/xfrm.c         | 199 +++++++++++++++++++++++++++++++++-------
 9 files changed, 329 insertions(+), 80 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 4d7fb59996b0..2c4921d79d19 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -31,6 +31,7 @@
 #include <linux/msg.h>
 #include <linux/sched.h>
 #include <linux/key.h>
+#include <linux/xfrm.h>
 
 struct ctl_table;
 
@@ -825,9 +826,8 @@ struct swap_info_struct;
  *	used by the XFRM system.
  *	@sec_ctx contains the security context information being provided by
  *	the user-level policy update program (e.g., setkey).
- *	Allocate a security structure to the xp->security field.
- *	The security field is initialized to NULL when the xfrm_policy is
- *	allocated.
+ *	Allocate a security structure to the xp->security field; the security
+ *	field is initialized to NULL when the xfrm_policy is allocated.
  *	Return 0 if operation was successful (memory to allocate, legal context)
  * @xfrm_policy_clone_security:
  *	@old contains an existing xfrm_policy in the SPD.
@@ -846,9 +846,14 @@ struct swap_info_struct;
  *	Database by the XFRM system.
  *	@sec_ctx contains the security context information being provided by
  *	the user-level SA generation program (e.g., setkey or racoon).
- *	Allocate a security structure to the x->security field.  The
- *	security field is initialized to NULL when the xfrm_state is
- *	allocated.
+ *	@polsec contains the security context information associated with a xfrm
+ *	policy rule from which to take the base context. polsec must be NULL
+ *	when sec_ctx is specified.
+ *	@secid contains the secid from which to take the mls portion of the context.
+ *	Allocate a security structure to the x->security field; the security
+ *	field is initialized to NULL when the xfrm_state is allocated. Set the
+ *	context to correspond to either sec_ctx or polsec, with the mls portion
+ *	taken from secid in the latter case.
  *	Return 0 if operation was successful (memory to allocate, legal context).
  * @xfrm_state_free_security:
  *	@x contains the xfrm_state.
@@ -859,13 +864,26 @@ struct swap_info_struct;
  * @xfrm_policy_lookup:
  *	@xp contains the xfrm_policy for which the access control is being
  *	checked.
- *	@sk_sid contains the sock security label that is used to authorize
+ *	@fl_secid contains the flow security label that is used to authorize
  *	access to the policy xp.
  *	@dir contains the direction of the flow (input or output).
- *	Check permission when a sock selects a xfrm_policy for processing
+ *	Check permission when a flow selects a xfrm_policy for processing
  *	XFRMs on a packet.  The hook is called when selecting either a
  *	per-socket policy or a generic xfrm policy.
  *	Return 0 if permission is granted.
+ * @xfrm_state_pol_flow_match:
+ *	@x contains the state to match.
+ *	@xp contains the policy to check for a match.
+ *	@fl contains the flow to check for a match.
+ *	Return 1 if there is a match.
+ * @xfrm_flow_state_match:
+ *	@fl contains the flow key to match.
+ *	@xfrm points to the xfrm_state to match.
+ *	Return 1 if there is a match.
+ * @xfrm_decode_session:
+ *	@skb points to skb to decode.
+ *	@fl points to the flow key to set.
+ *	Return 0 if successful decoding.
  *
  * Security hooks affecting all Key Management operations
  *
@@ -1343,10 +1361,16 @@ struct security_operations {
 	int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new);
 	void (*xfrm_policy_free_security) (struct xfrm_policy *xp);
 	int (*xfrm_policy_delete_security) (struct xfrm_policy *xp);
-	int (*xfrm_state_alloc_security) (struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx);
+	int (*xfrm_state_alloc_security) (struct xfrm_state *x,
+		struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *polsec,
+		u32 secid);
 	void (*xfrm_state_free_security) (struct xfrm_state *x);
 	int (*xfrm_state_delete_security) (struct xfrm_state *x);
-	int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 sk_sid, u8 dir);
+	int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 fl_secid, u8 dir);
+	int (*xfrm_state_pol_flow_match)(struct xfrm_state *x,
+			struct xfrm_policy *xp, struct flowi *fl);
+	int (*xfrm_flow_state_match)(struct flowi *fl, struct xfrm_state *xfrm);
+	int (*xfrm_decode_session)(struct sk_buff *skb, struct flowi *fl);
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
 	/* key management security hooks */
@@ -3050,9 +3074,18 @@ static inline int security_xfrm_policy_delete(struct xfrm_policy *xp)
 	return security_ops->xfrm_policy_delete_security(xp);
 }
 
-static inline int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx)
+static inline int security_xfrm_state_alloc(struct xfrm_state *x,
+			struct xfrm_user_sec_ctx *sec_ctx)
+{
+	return security_ops->xfrm_state_alloc_security(x, sec_ctx, NULL, 0);
+}
+
+static inline int security_xfrm_state_alloc_acquire(struct xfrm_state *x,
+				struct xfrm_sec_ctx *polsec, u32 secid)
 {
-	return security_ops->xfrm_state_alloc_security(x, sec_ctx);
+	if (!polsec)
+		return 0;
+	return security_ops->xfrm_state_alloc_security(x, NULL, polsec, secid);
 }
 
 static inline int security_xfrm_state_delete(struct xfrm_state *x)
@@ -3065,9 +3098,25 @@ static inline void security_xfrm_state_free(struct xfrm_state *x)
 	security_ops->xfrm_state_free_security(x);
 }
 
-static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
+static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
+{
+	return security_ops->xfrm_policy_lookup(xp, fl_secid, dir);
+}
+
+static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x,
+			struct xfrm_policy *xp, struct flowi *fl)
+{
+	return security_ops->xfrm_state_pol_flow_match(x, xp, fl);
+}
+
+static inline int security_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm)
+{
+	return security_ops->xfrm_flow_state_match(fl, xfrm);
+}
+
+static inline int security_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
 {
-	return security_ops->xfrm_policy_lookup(xp, sk_sid, dir);
+	return security_ops->xfrm_decode_session(skb, fl);
 }
 #else	/* CONFIG_SECURITY_NETWORK_XFRM */
 static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx)
@@ -3089,7 +3138,14 @@ static inline int security_xfrm_policy_delete(struct xfrm_policy *xp)
 	return 0;
 }
 
-static inline int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx)
+static inline int security_xfrm_state_alloc(struct xfrm_state *x,
+					struct xfrm_user_sec_ctx *sec_ctx)
+{
+	return 0;
+}
+
+static inline int security_xfrm_state_alloc_acquire(struct xfrm_state *x,
+					struct xfrm_sec_ctx *polsec, u32 secid)
 {
 	return 0;
 }
@@ -3103,10 +3159,28 @@ static inline int security_xfrm_state_delete(struct xfrm_state *x)
 	return 0;
 }
 
-static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
+static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
 {
 	return 0;
 }
+
+static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x,
+			struct xfrm_policy *xp, struct flowi *fl)
+{
+	return 1;
+}
+
+static inline int security_xfrm_flow_state_match(struct flowi *fl,
+                                struct xfrm_state *xfrm)
+{
+	return 1;
+}
+
+static inline int security_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+{
+	return 0;
+}
+
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
 #ifdef CONFIG_KEYS
diff --git a/include/net/flow.h b/include/net/flow.h
index 1cee5a83433a..21d988b2058a 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -86,10 +86,10 @@ struct flowi {
 #define FLOW_DIR_FWD	2
 
 struct sock;
-typedef void (*flow_resolve_t)(struct flowi *key, u32 sk_sid, u16 family, u8 dir,
+typedef void (*flow_resolve_t)(struct flowi *key, u16 family, u8 dir,
 			       void **objp, atomic_t **obj_refp);
 
-extern void *flow_cache_lookup(struct flowi *key, u32 sk_sid, u16 family, u8 dir,
+extern void *flow_cache_lookup(struct flowi *key, u16 family, u8 dir,
 	 		       flow_resolve_t resolver);
 extern void flow_cache_flush(void);
 extern atomic_t flow_cache_genid;
diff --git a/net/core/flow.c b/net/core/flow.c
index 2191af5f26ac..645241165e6c 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -32,7 +32,6 @@ struct flow_cache_entry {
 	u8			dir;
 	struct flowi		key;
 	u32			genid;
-	u32			sk_sid;
 	void			*object;
 	atomic_t		*object_ref;
 };
@@ -165,7 +164,7 @@ static int flow_key_compare(struct flowi *key1, struct flowi *key2)
 	return 0;
 }
 
-void *flow_cache_lookup(struct flowi *key, u32 sk_sid, u16 family, u8 dir,
+void *flow_cache_lookup(struct flowi *key, u16 family, u8 dir,
 			flow_resolve_t resolver)
 {
 	struct flow_cache_entry *fle, **head;
@@ -189,7 +188,6 @@ void *flow_cache_lookup(struct flowi *key, u32 sk_sid, u16 family, u8 dir,
 	for (fle = *head; fle; fle = fle->next) {
 		if (fle->family == family &&
 		    fle->dir == dir &&
-		    fle->sk_sid == sk_sid &&
 		    flow_key_compare(key, &fle->key) == 0) {
 			if (fle->genid == atomic_read(&flow_cache_genid)) {
 				void *ret = fle->object;
@@ -214,7 +212,6 @@ void *flow_cache_lookup(struct flowi *key, u32 sk_sid, u16 family, u8 dir,
 			*head = fle;
 			fle->family = family;
 			fle->dir = dir;
-			fle->sk_sid = sk_sid;
 			memcpy(&fle->key, key, sizeof(*key));
 			fle->object = NULL;
 			flow_count(cpu)++;
@@ -226,7 +223,7 @@ nocache:
 		void *obj;
 		atomic_t *obj_ref;
 
-		resolver(key, sk_sid, family, dir, &obj, &obj_ref);
+		resolver(key, family, dir, &obj, &obj_ref);
 
 		if (fle) {
 			fle->genid = atomic_read(&flow_cache_genid);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 3da67ca2c3ce..79405daadc52 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -597,7 +597,7 @@ EXPORT_SYMBOL(xfrm_policy_walk);
 
 /* Find policy to apply to this flow. */
 
-static void xfrm_policy_lookup(struct flowi *fl, u32 sk_sid, u16 family, u8 dir,
+static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir,
 			       void **objp, atomic_t **obj_refp)
 {
 	struct xfrm_policy *pol;
@@ -613,7 +613,7 @@ static void xfrm_policy_lookup(struct flowi *fl, u32 sk_sid, u16 family, u8 dir,
 		match = xfrm_selector_match(sel, fl, family);
 
 		if (match) {
- 			if (!security_xfrm_policy_lookup(pol, sk_sid, dir)) {
+ 			if (!security_xfrm_policy_lookup(pol, fl->secid, dir)) {
 				xfrm_pol_hold(pol);
 				break;
 			}
@@ -641,7 +641,7 @@ static inline int policy_to_flow_dir(int dir)
 	};
 }
 
-static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struct flowi *fl, u32 sk_sid)
+static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struct flowi *fl)
 {
 	struct xfrm_policy *pol;
 
@@ -652,7 +652,7 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc
  		int err = 0;
 
 		if (match)
-		  err = security_xfrm_policy_lookup(pol, sk_sid, policy_to_flow_dir(dir));
+		  err = security_xfrm_policy_lookup(pol, fl->secid, policy_to_flow_dir(dir));
 
  		if (match && !err)
 			xfrm_pol_hold(pol);
@@ -862,19 +862,20 @@ int xfrm_lookup(struct dst_entry **dst_p, struct flowi *fl,
 	u32 genid;
 	u16 family;
 	u8 dir = policy_to_flow_dir(XFRM_POLICY_OUT);
-	u32 sk_sid = security_sk_sid(sk, fl, dir);
+
+	fl->secid = security_sk_sid(sk, fl, dir);
 restart:
 	genid = atomic_read(&flow_cache_genid);
 	policy = NULL;
 	if (sk && sk->sk_policy[1])
-		policy = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl, sk_sid);
+		policy = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl);
 
 	if (!policy) {
 		/* To accelerate a bit...  */
 		if ((dst_orig->flags & DST_NOXFRM) || !xfrm_policy_list[XFRM_POLICY_OUT])
 			return 0;
 
-		policy = flow_cache_lookup(fl, sk_sid, dst_orig->ops->family,
+		policy = flow_cache_lookup(fl, dst_orig->ops->family,
 					   dir, xfrm_policy_lookup);
 	}
 
@@ -1032,13 +1033,15 @@ int
 xfrm_decode_session(struct sk_buff *skb, struct flowi *fl, unsigned short family)
 {
 	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
+	int err;
 
 	if (unlikely(afinfo == NULL))
 		return -EAFNOSUPPORT;
 
 	afinfo->decode_session(skb, fl);
+	err = security_xfrm_decode_session(skb, fl);
 	xfrm_policy_put_afinfo(afinfo);
-	return 0;
+	return err;
 }
 EXPORT_SYMBOL(xfrm_decode_session);
 
@@ -1058,14 +1061,11 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 	struct xfrm_policy *pol;
 	struct flowi fl;
 	u8 fl_dir = policy_to_flow_dir(dir);
-	u32 sk_sid;
 
 	if (xfrm_decode_session(skb, &fl, family) < 0)
 		return 0;
 	nf_nat_decode_session(skb, &fl, family);
 
-	sk_sid = security_sk_sid(sk, &fl, fl_dir);
-
 	/* First, check used SA against their selectors. */
 	if (skb->sp) {
 		int i;
@@ -1079,10 +1079,10 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 
 	pol = NULL;
 	if (sk && sk->sk_policy[dir])
-		pol = xfrm_sk_policy_lookup(sk, dir, &fl, sk_sid);
+		pol = xfrm_sk_policy_lookup(sk, dir, &fl);
 
 	if (!pol)
-		pol = flow_cache_lookup(&fl, sk_sid, family, fl_dir,
+		pol = flow_cache_lookup(&fl, family, fl_dir,
 					xfrm_policy_lookup);
 
 	if (!pol)
@@ -1298,6 +1298,8 @@ int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family)
 
 		if (fl && !xfrm_selector_match(&dst->xfrm->sel, fl, family))
 			return 0;
+		if (fl && !security_xfrm_flow_state_match(fl, dst->xfrm))
+			return 0;
 		if (dst->xfrm->km.state != XFRM_STATE_VALID)
 			return 0;
 
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 0021aad5db43..be02bd981d12 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -367,7 +367,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 			 */
 			if (x->km.state == XFRM_STATE_VALID) {
 				if (!xfrm_selector_match(&x->sel, fl, family) ||
-				    !xfrm_sec_ctx_match(pol->security, x->security))
+				    !security_xfrm_state_pol_flow_match(x, pol, fl))
 					continue;
 				if (!best ||
 				    best->km.dying > x->km.dying ||
@@ -379,7 +379,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 			} else if (x->km.state == XFRM_STATE_ERROR ||
 				   x->km.state == XFRM_STATE_EXPIRED) {
  				if (xfrm_selector_match(&x->sel, fl, family) &&
-				    xfrm_sec_ctx_match(pol->security, x->security))
+				    security_xfrm_state_pol_flow_match(x, pol, fl))
 					error = -ESRCH;
 			}
 		}
@@ -403,6 +403,14 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 		 * to current session. */
 		xfrm_init_tempsel(x, fl, tmpl, daddr, saddr, family);
 
+		error = security_xfrm_state_alloc_acquire(x, pol->security, fl->secid);
+		if (error) {
+			x->km.state = XFRM_STATE_DEAD;
+			xfrm_state_put(x);
+			x = NULL;
+			goto out;
+		}
+
 		if (km_query(x, tmpl, pol) == 0) {
 			x->km.state = XFRM_STATE_ACQ;
 			list_add_tail(&x->bydst, xfrm_state_bydst+h);
diff --git a/security/dummy.c b/security/dummy.c
index bd3bc5faa9a8..c1f10654871e 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -835,7 +835,8 @@ static int dummy_xfrm_policy_delete_security(struct xfrm_policy *xp)
 	return 0;
 }
 
-static int dummy_xfrm_state_alloc_security(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx)
+static int dummy_xfrm_state_alloc_security(struct xfrm_state *x,
+	struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *pol, u32 secid)
 {
 	return 0;
 }
@@ -853,6 +854,23 @@ static int dummy_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
 {
 	return 0;
 }
+
+static int dummy_xfrm_state_pol_flow_match(struct xfrm_state *x,
+				struct xfrm_policy *xp, struct flowi *fl)
+{
+	return 1;
+}
+
+static int dummy_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm)
+{
+	return 1;
+}
+
+static int dummy_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+{
+	return 0;
+}
+
 #endif /* CONFIG_SECURITY_NETWORK_XFRM */
 static int dummy_register_security (const char *name, struct security_operations *ops)
 {
@@ -1076,6 +1094,9 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, xfrm_state_free_security);
 	set_to_dummy_if_null(ops, xfrm_state_delete_security);
 	set_to_dummy_if_null(ops, xfrm_policy_lookup);
+	set_to_dummy_if_null(ops, xfrm_state_pol_flow_match);
+	set_to_dummy_if_null(ops, xfrm_flow_state_match);
+	set_to_dummy_if_null(ops, xfrm_decode_session);
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 #ifdef CONFIG_KEYS
 	set_to_dummy_if_null(ops, key_alloc);
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index d67abf77584a..5c189da07bc9 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3468,7 +3468,7 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (err)
 		goto out;
 
-	err = selinux_xfrm_sock_rcv_skb(sock_sid, skb);
+	err = selinux_xfrm_sock_rcv_skb(sock_sid, skb, &ad);
 out:	
 	return err;
 }
@@ -3720,7 +3720,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum,
 	if (err)
 		goto out;
 
-	err = selinux_xfrm_postroute_last(isec->sid, skb);
+	err = selinux_xfrm_postroute_last(isec->sid, skb, &ad);
 out:
 	return err ? NF_DROP : NF_ACCEPT;
 }
@@ -4633,6 +4633,9 @@ static struct security_operations selinux_ops = {
 	.xfrm_state_free_security =	selinux_xfrm_state_free,
 	.xfrm_state_delete_security =	selinux_xfrm_state_delete,
 	.xfrm_policy_lookup = 		selinux_xfrm_policy_lookup,
+	.xfrm_state_pol_flow_match =	selinux_xfrm_state_pol_flow_match,
+	.xfrm_flow_state_match =	selinux_xfrm_flow_state_match,
+	.xfrm_decode_session =		selinux_xfrm_decode_session,
 #endif
 
 #ifdef CONFIG_KEYS
diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h
index c96498a10eb8..f51a3e84bd9b 100644
--- a/security/selinux/include/xfrm.h
+++ b/security/selinux/include/xfrm.h
@@ -2,6 +2,7 @@
  * SELinux support for the XFRM LSM hooks
  *
  * Author : Trent Jaeger, <jaegert@us.ibm.com>
+ * Updated : Venkat Yekkirala, <vyekkirala@TrustedCS.com>
  */
 #ifndef _SELINUX_XFRM_H_
 #define _SELINUX_XFRM_H_
@@ -10,10 +11,16 @@ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *
 int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new);
 void selinux_xfrm_policy_free(struct xfrm_policy *xp);
 int selinux_xfrm_policy_delete(struct xfrm_policy *xp);
-int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx);
+int selinux_xfrm_state_alloc(struct xfrm_state *x,
+	struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *pol, u32 secid);
 void selinux_xfrm_state_free(struct xfrm_state *x);
 int selinux_xfrm_state_delete(struct xfrm_state *x);
-int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir);
+int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir);
+int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x,
+			struct xfrm_policy *xp, struct flowi *fl);
+int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm);
+int selinux_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl);
+
 
 /*
  * Extract the security blob from the sock (it's actually on the socket)
@@ -39,17 +46,21 @@ static inline u32 selinux_no_sk_sid(struct flowi *fl)
 }
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
-int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb);
-int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb);
+int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb,
+			struct avc_audit_data *ad);
+int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb,
+			struct avc_audit_data *ad);
 u32 selinux_socket_getpeer_stream(struct sock *sk);
 u32 selinux_socket_getpeer_dgram(struct sk_buff *skb);
 #else
-static inline int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb)
+static inline int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb,
+			struct avc_audit_data *ad)
 {
 	return 0;
 }
 
-static inline int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb)
+static inline int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb,
+			struct avc_audit_data *ad)
 {
 	return 0;
 }
diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c
index 6c985ced8102..a502b0540e3d 100644
--- a/security/selinux/xfrm.c
+++ b/security/selinux/xfrm.c
@@ -6,7 +6,12 @@
  *  Authors:  Serge Hallyn <sergeh@us.ibm.com>
  *	      Trent Jaeger <jaegert@us.ibm.com>
  *
+ *  Updated: Venkat Yekkirala <vyekkirala@TrustedCS.com>
+ *
+ *           Granular IPSec Associations for use in MLS environments.
+ *
  *  Copyright (C) 2005 International Business Machines Corporation
+ *  Copyright (C) 2006 Trusted Computer Solutions, Inc.
  *
  *	This program is free software; you can redistribute it and/or modify
  *	it under the terms of the GNU General Public License version 2,
@@ -67,10 +72,10 @@ static inline int selinux_authorizable_xfrm(struct xfrm_state *x)
 }
 
 /*
- * LSM hook implementation that authorizes that a socket can be used
- * with the corresponding xfrm_sec_ctx and direction.
+ * LSM hook implementation that authorizes that a flow can use
+ * a xfrm policy rule.
  */
-int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
+int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir)
 {
 	int rc = 0;
 	u32 sel_sid = SECINITSID_UNLABELED;
@@ -84,27 +89,129 @@ int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
 		sel_sid = ctx->ctx_sid;
 	}
 
-	rc = avc_has_perm(sk_sid, sel_sid, SECCLASS_ASSOCIATION,
-			  ((dir == FLOW_DIR_IN) ? ASSOCIATION__RECVFROM :
-			   ((dir == FLOW_DIR_OUT) ?  ASSOCIATION__SENDTO :
-			    (ASSOCIATION__SENDTO | ASSOCIATION__RECVFROM))),
+	rc = avc_has_perm(fl_secid, sel_sid, SECCLASS_ASSOCIATION,
+			  ASSOCIATION__POLMATCH,
 			  NULL);
 
 	return rc;
 }
 
+/*
+ * LSM hook implementation that authorizes that a state matches
+ * the given policy, flow combo.
+ */
+
+int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy *xp,
+			struct flowi *fl)
+{
+	u32 state_sid;
+	u32 pol_sid;
+	int err;
+
+	if (x->security)
+		state_sid = x->security->ctx_sid;
+	else
+		state_sid = SECINITSID_UNLABELED;
+
+	if (xp->security)
+		pol_sid = xp->security->ctx_sid;
+	else
+		pol_sid = SECINITSID_UNLABELED;
+
+	err = avc_has_perm(state_sid, pol_sid, SECCLASS_ASSOCIATION,
+			  ASSOCIATION__POLMATCH,
+			  NULL);
+
+	if (err)
+		return 0;
+
+	return selinux_xfrm_flow_state_match(fl, x);
+}
+
+/*
+ * LSM hook implementation that authorizes that a particular outgoing flow
+ * can use a given security association.
+ */
+
+int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm)
+{
+	int rc = 0;
+	u32 sel_sid = SECINITSID_UNLABELED;
+	struct xfrm_sec_ctx *ctx;
+
+	/* Context sid is either set to label or ANY_ASSOC */
+	if ((ctx = xfrm->security)) {
+		if (!selinux_authorizable_ctx(ctx))
+			return 0;
+
+		sel_sid = ctx->ctx_sid;
+	}
+
+	rc = avc_has_perm(fl->secid, sel_sid, SECCLASS_ASSOCIATION,
+			  ASSOCIATION__SENDTO,
+			  NULL)? 0:1;
+
+	return rc;
+}
+
+/*
+ * LSM hook implementation that determines the sid for the session.
+ */
+
+int selinux_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+{
+	struct sec_path *sp;
+
+	fl->secid = SECSID_NULL;
+
+	if (skb == NULL)
+		return 0;
+
+	sp = skb->sp;
+	if (sp) {
+		int i, sid_set = 0;
+
+		for (i = sp->len-1; i >= 0; i--) {
+			struct xfrm_state *x = sp->xvec[i];
+			if (selinux_authorizable_xfrm(x)) {
+				struct xfrm_sec_ctx *ctx = x->security;
+
+				if (!sid_set) {
+					fl->secid = ctx->ctx_sid;
+					sid_set = 1;
+				}
+				else if (fl->secid != ctx->ctx_sid)
+					return -EINVAL;
+			}
+		}
+	}
+
+	return 0;
+}
+
 /*
  * Security blob allocation for xfrm_policy and xfrm_state
  * CTX does not have a meaningful value on input
  */
-static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *uctx)
+static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp,
+	struct xfrm_user_sec_ctx *uctx, struct xfrm_sec_ctx *pol, u32 sid)
 {
 	int rc = 0;
 	struct task_security_struct *tsec = current->security;
-	struct xfrm_sec_ctx *ctx;
+	struct xfrm_sec_ctx *ctx = NULL;
+	char *ctx_str = NULL;
+	u32 str_len;
+	u32 ctx_sid;
+
+	BUG_ON(uctx && pol);
+
+	if (pol)
+		goto from_policy;
 
 	BUG_ON(!uctx);
-	BUG_ON(uctx->ctx_doi != XFRM_SC_ALG_SELINUX);
+
+	if (uctx->ctx_doi != XFRM_SC_ALG_SELINUX)
+		return -EINVAL;
 
 	if (uctx->ctx_len >= PAGE_SIZE)
 		return -ENOMEM;
@@ -141,9 +248,41 @@ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_us
 
 	return rc;
 
+from_policy:
+	BUG_ON(!pol);
+	rc = security_sid_mls_copy(pol->ctx_sid, sid, &ctx_sid);
+	if (rc)
+		goto out;
+
+	rc = security_sid_to_context(ctx_sid, &ctx_str, &str_len);
+	if (rc)
+		goto out;
+
+	*ctxp = ctx = kmalloc(sizeof(*ctx) +
+			      str_len,
+			      GFP_ATOMIC);
+
+	if (!ctx) {
+		rc = -ENOMEM;
+		goto out;
+	}
+
+
+	ctx->ctx_doi = XFRM_SC_DOI_LSM;
+	ctx->ctx_alg = XFRM_SC_ALG_SELINUX;
+	ctx->ctx_sid = ctx_sid;
+	ctx->ctx_len = str_len;
+	memcpy(ctx->ctx_str,
+	       ctx_str,
+	       str_len);
+
+	goto out2;
+
 out:
 	*ctxp = NULL;
 	kfree(ctx);
+out2:
+	kfree(ctx_str);
 	return rc;
 }
 
@@ -157,7 +296,7 @@ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *
 
 	BUG_ON(!xp);
 
-	err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx);
+	err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, NULL, 0);
 	return err;
 }
 
@@ -217,13 +356,14 @@ int selinux_xfrm_policy_delete(struct xfrm_policy *xp)
  * LSM hook implementation that allocs and transfers sec_ctx spec to
  * xfrm_state.
  */
-int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *uctx)
+int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *uctx,
+		struct xfrm_sec_ctx *pol, u32 secid)
 {
 	int err;
 
 	BUG_ON(!x);
 
-	err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx);
+	err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx, pol, secid);
 	return err;
 }
 
@@ -329,38 +469,30 @@ int selinux_xfrm_state_delete(struct xfrm_state *x)
  * we need to check for unlabelled access since this may not have
  * gone thru the IPSec process.
  */
-int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb)
+int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb,
+				struct avc_audit_data *ad)
 {
 	int i, rc = 0;
 	struct sec_path *sp;
+	u32 sel_sid = SECINITSID_UNLABELED;
 
 	sp = skb->sp;
 
 	if (sp) {
-		/*
-		 * __xfrm_policy_check does not approve unless xfrm_policy_ok
-		 * says that spi's match for policy and the socket.
-		 *
-		 *  Only need to verify the existence of an authorizable sp.
-		 */
 		for (i = 0; i < sp->len; i++) {
 			struct xfrm_state *x = sp->xvec[i];
 
-			if (x && selinux_authorizable_xfrm(x))
-				goto accept;
+			if (x && selinux_authorizable_xfrm(x)) {
+				struct xfrm_sec_ctx *ctx = x->security;
+				sel_sid = ctx->ctx_sid;
+				break;
+			}
 		}
 	}
 
-	/* check SELinux sock for unlabelled access */
-	rc = avc_has_perm(isec_sid, SECINITSID_UNLABELED, SECCLASS_ASSOCIATION,
-			  ASSOCIATION__RECVFROM, NULL);
-	if (rc)
-		goto drop;
-
-accept:
-	return 0;
+	rc = avc_has_perm(isec_sid, sel_sid, SECCLASS_ASSOCIATION,
+			  ASSOCIATION__RECVFROM, ad);
 
-drop:
 	return rc;
 }
 
@@ -371,7 +503,8 @@ drop:
  * If we do have a authorizable security association, then it has already been
  * checked in xfrm_policy_lookup hook.
  */
-int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb)
+int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb,
+					struct avc_audit_data *ad)
 {
 	struct dst_entry *dst;
 	int rc = 0;
@@ -391,7 +524,7 @@ int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb)
 	}
 
 	rc = avc_has_perm(isec_sid, SECINITSID_UNLABELED, SECCLASS_ASSOCIATION,
-			  ASSOCIATION__SENDTO, NULL);
+			  ASSOCIATION__SENDTO, ad);
 out:
 	return rc;
 }
-- 
cgit v1.2.3


From beb8d13bed80f8388f1a9a107d07ddd342e627e8 Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Fri, 4 Aug 2006 23:12:42 -0700
Subject: [MLSXFRM]: Add flow labeling

This labels the flows that could utilize IPSec xfrms at the points the
flows are defined so that IPSec policy and SAs at the right label can
be used.

The following protos are currently not handled, but they should
continue to be able to use single-labeled IPSec like they currently
do.

ipmr
ip_gre
ipip
igmp
sit
sctp
ip6_tunnel (IPv6 over IPv6 tunnel device)
decnet

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h         | 38 +++++++++++++++++++++++++-------------
 include/net/route.h              |  3 +++
 net/dccp/ipv4.c                  |  1 +
 net/dccp/ipv6.c                  |  6 ++++++
 net/ipv4/af_inet.c               |  1 +
 net/ipv4/icmp.c                  |  2 ++
 net/ipv4/inet_connection_sock.c  |  1 +
 net/ipv4/ip_output.c             |  2 ++
 net/ipv4/netfilter/ipt_REJECT.c  |  1 +
 net/ipv4/raw.c                   |  1 +
 net/ipv4/syncookies.c            |  1 +
 net/ipv4/udp.c                   |  1 +
 net/ipv6/af_inet6.c              |  1 +
 net/ipv6/datagram.c              |  2 ++
 net/ipv6/icmp.c                  |  2 ++
 net/ipv6/inet6_connection_sock.c |  1 +
 net/ipv6/ndisc.c                 |  1 +
 net/ipv6/netfilter/ip6t_REJECT.c |  1 +
 net/ipv6/raw.c                   |  1 +
 net/ipv6/tcp_ipv6.c              |  7 +++++++
 net/ipv6/udp.c                   |  2 ++
 net/xfrm/xfrm_policy.c           |  3 +--
 security/dummy.c                 |  7 +++----
 security/selinux/hooks.c         |  8 ++++----
 security/selinux/include/xfrm.h  | 14 +-------------
 security/selinux/xfrm.c          | 11 +++++++----
 26 files changed, 79 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 2c4921d79d19..f3909d189fe0 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -32,6 +32,7 @@
 #include <linux/sched.h>
 #include <linux/key.h>
 #include <linux/xfrm.h>
+#include <net/flow.h>
 
 struct ctl_table;
 
@@ -815,8 +816,8 @@ struct swap_info_struct;
  *	Deallocate security structure.
  * @sk_clone_security:
  *	Clone/copy security structure.
- * @sk_getsid:
- *	Retrieve the LSM-specific sid for the sock to enable caching of network
+ * @sk_getsecid:
+ *	Retrieve the LSM-specific secid for the sock to enable caching of network
  *	authorizations.
  *
  * Security hooks for XFRM operations.
@@ -882,8 +883,9 @@ struct swap_info_struct;
  *	Return 1 if there is a match.
  * @xfrm_decode_session:
  *	@skb points to skb to decode.
- *	@fl points to the flow key to set.
- *	Return 0 if successful decoding.
+ *	@secid points to the flow key secid to set.
+ *	@ckall says if all xfrms used should be checked for same secid.
+ *	Return 0 if ckall is zero or all xfrms used have the same secid.
  *
  * Security hooks affecting all Key Management operations
  *
@@ -1353,7 +1355,7 @@ struct security_operations {
 	int (*sk_alloc_security) (struct sock *sk, int family, gfp_t priority);
 	void (*sk_free_security) (struct sock *sk);
 	void (*sk_clone_security) (const struct sock *sk, struct sock *newsk);
-	unsigned int (*sk_getsid) (struct sock *sk, struct flowi *fl, u8 dir);
+	void (*sk_getsecid) (struct sock *sk, u32 *secid);
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
@@ -1370,7 +1372,7 @@ struct security_operations {
 	int (*xfrm_state_pol_flow_match)(struct xfrm_state *x,
 			struct xfrm_policy *xp, struct flowi *fl);
 	int (*xfrm_flow_state_match)(struct flowi *fl, struct xfrm_state *xfrm);
-	int (*xfrm_decode_session)(struct sk_buff *skb, struct flowi *fl);
+	int (*xfrm_decode_session)(struct sk_buff *skb, u32 *secid, int ckall);
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
 	/* key management security hooks */
@@ -2917,9 +2919,9 @@ static inline void security_sk_clone(const struct sock *sk, struct sock *newsk)
 	return security_ops->sk_clone_security(sk, newsk);
 }
 
-static inline unsigned int security_sk_sid(struct sock *sk, struct flowi *fl, u8 dir)
+static inline void security_sk_classify_flow(struct sock *sk, struct flowi *fl)
 {
-	return security_ops->sk_getsid(sk, fl, dir);
+	security_ops->sk_getsecid(sk, &fl->secid);
 }
 #else	/* CONFIG_SECURITY_NETWORK */
 static inline int security_unix_stream_connect(struct socket * sock,
@@ -3047,9 +3049,8 @@ static inline void security_sk_clone(const struct sock *sk, struct sock *newsk)
 {
 }
 
-static inline unsigned int security_sk_sid(struct sock *sk, struct flowi *fl, u8 dir)
+static inline void security_sk_classify_flow(struct sock *sk, struct flowi *fl)
 {
-	return 0;
 }
 #endif	/* CONFIG_SECURITY_NETWORK */
 
@@ -3114,9 +3115,16 @@ static inline int security_xfrm_flow_state_match(struct flowi *fl, struct xfrm_s
 	return security_ops->xfrm_flow_state_match(fl, xfrm);
 }
 
-static inline int security_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid)
+{
+	return security_ops->xfrm_decode_session(skb, secid, 1);
+}
+
+static inline void security_skb_classify_flow(struct sk_buff *skb, struct flowi *fl)
 {
-	return security_ops->xfrm_decode_session(skb, fl);
+	int rc = security_ops->xfrm_decode_session(skb, &fl->secid, 0);
+
+	BUG_ON(rc);
 }
 #else	/* CONFIG_SECURITY_NETWORK_XFRM */
 static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx)
@@ -3176,11 +3184,15 @@ static inline int security_xfrm_flow_state_match(struct flowi *fl,
 	return 1;
 }
 
-static inline int security_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid)
 {
 	return 0;
 }
 
+static inline void security_skb_classify_flow(struct sk_buff *skb, struct flowi *fl)
+{
+}
+
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
 #ifdef CONFIG_KEYS
diff --git a/include/net/route.h b/include/net/route.h
index c4a068692dcc..7f93ac0e0899 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -32,6 +32,7 @@
 #include <linux/route.h>
 #include <linux/ip.h>
 #include <linux/cache.h>
+#include <linux/security.h>
 
 #ifndef __KERNEL__
 #warning This file is not supposed to be used outside of kernel.
@@ -166,6 +167,7 @@ static inline int ip_route_connect(struct rtable **rp, u32 dst,
 		ip_rt_put(*rp);
 		*rp = NULL;
 	}
+	security_sk_classify_flow(sk, &fl);
 	return ip_route_output_flow(rp, &fl, sk, 0);
 }
 
@@ -182,6 +184,7 @@ static inline int ip_route_newports(struct rtable **rp, u8 protocol,
 		fl.proto = protocol;
 		ip_rt_put(*rp);
 		*rp = NULL;
+		security_sk_classify_flow(sk, &fl);
 		return ip_route_output_flow(rp, &fl, sk, 0);
 	}
 	return 0;
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 7f56f7e8f571..386498053b1c 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -678,6 +678,7 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk,
 			   	     }
 			  };
 
+	security_skb_classify_flow(skb, &fl);
 	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 610c722ac27f..53d255c01431 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -201,6 +201,7 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	fl.oif = sk->sk_bound_dev_if;
 	fl.fl_ip_dport = usin->sin6_port;
 	fl.fl_ip_sport = inet->sport;
+	security_sk_classify_flow(sk, &fl);
 
 	if (np->opt != NULL && np->opt->srcrt != NULL) {
 		const struct rt0_hdr *rt0 = (struct rt0_hdr *)np->opt->srcrt;
@@ -322,6 +323,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			fl.oif = sk->sk_bound_dev_if;
 			fl.fl_ip_dport = inet->dport;
 			fl.fl_ip_sport = inet->sport;
+			security_sk_classify_flow(sk, &fl);
 
 			err = ip6_dst_lookup(sk, &dst, &fl);
 			if (err) {
@@ -422,6 +424,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req,
 	fl.oif = ireq6->iif;
 	fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 	fl.fl_ip_sport = inet_sk(sk)->sport;
+	security_sk_classify_flow(sk, &fl);
 
 	if (dst == NULL) {
 		opt = np->opt;
@@ -566,6 +569,7 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb)
 	fl.oif = inet6_iif(rxskb);
 	fl.fl_ip_dport = dh->dccph_dport;
 	fl.fl_ip_sport = dh->dccph_sport;
+	security_skb_classify_flow(rxskb, &fl);
 
 	/* sk = NULL, but it is safe for now. RST socket required. */
 	if (!ip6_dst_lookup(NULL, &skb->dst, &fl)) {
@@ -622,6 +626,7 @@ static void dccp_v6_reqsk_send_ack(struct sk_buff *rxskb,
 	fl.oif = inet6_iif(rxskb);
 	fl.fl_ip_dport = dh->dccph_dport;
 	fl.fl_ip_sport = dh->dccph_sport;
+	security_skb_classify_flow(rxskb, &fl);
 
 	if (!ip6_dst_lookup(NULL, &skb->dst, &fl)) {
 		if (xfrm_lookup(&skb->dst, &fl, NULL, 0) >= 0) {
@@ -842,6 +847,7 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk,
 		fl.oif = sk->sk_bound_dev_if;
 		fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 		fl.fl_ip_sport = inet_sk(sk)->sport;
+		security_sk_classify_flow(sk, &fl);
 
 		if (ip6_dst_lookup(sk, &dst, &fl))
 			goto out;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index c84a32070f8d..fc40da3b6d39 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1074,6 +1074,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 		},
 	};
 						
+	security_sk_classify_flow(sk, &fl);
 	err = ip_route_output_flow(&rt, &fl, sk, 0);
 }
 	if (!err)
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 4c86ac3d882d..6ad797c14163 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -406,6 +406,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 						.saddr = rt->rt_spec_dst,
 						.tos = RT_TOS(skb->nh.iph->tos) } },
 				    .proto = IPPROTO_ICMP };
+		security_skb_classify_flow(skb, &fl);
 		if (ip_route_output_key(&rt, &fl))
 			goto out_unlock;
 	}
@@ -560,6 +561,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, u32 info)
 				}
 			}
 		};
+		security_skb_classify_flow(skb_in, &fl);
 		if (ip_route_output_key(&rt, &fl))
 			goto out_unlock;
 	}
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index e50a1bfd7ccc..772b4eac78bc 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -327,6 +327,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 				       { .sport = inet_sk(sk)->sport,
 					 .dport = ireq->rmt_port } } };
 
+	security_sk_classify_flow(sk, &fl);
 	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index a2ede167e045..308bdeac3455 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -328,6 +328,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok)
 			 * keep trying until route appears or the connection times
 			 * itself out.
 			 */
+			security_sk_classify_flow(sk, &fl);
 			if (ip_route_output_flow(&rt, &fl, sk, 0))
 				goto no_route;
 		}
@@ -1366,6 +1367,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 					       { .sport = skb->h.th->dest,
 					         .dport = skb->h.th->source } },
 				    .proto = sk->sk_protocol };
+		security_skb_classify_flow(skb, &fl);
 		if (ip_route_output_key(&rt, &fl))
 			return;
 	}
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index 269bc2067cb8..7f905bf2bde5 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -90,6 +90,7 @@ static inline struct rtable *route_reverse(struct sk_buff *skb,
 	fl.proto = IPPROTO_TCP;
 	fl.fl_ip_sport = tcph->dest;
 	fl.fl_ip_dport = tcph->source;
+	security_skb_classify_flow(skb, &fl);
 
 	xfrm_lookup((struct dst_entry **)&rt, &fl, NULL, 0);
 
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 62b2762a2420..fe44cb50a1c5 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -484,6 +484,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		if (!inet->hdrincl)
 			raw_probe_proto_opt(&fl, msg);
 
+		security_sk_classify_flow(sk, &fl);
 		err = ip_route_output_flow(&rt, &fl, sk, !(msg->msg_flags&MSG_DONTWAIT));
 	}
 	if (err)
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index e20be3331f67..307dc3c0d635 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -259,6 +259,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 				    .uli_u = { .ports =
 					       { .sport = skb->h.th->dest,
 						 .dport = skb->h.th->source } } };
+		security_sk_classify_flow(sk, &fl);
 		if (ip_route_output_key(&rt, &fl)) {
 			reqsk_free(req);
 			goto out; 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index f136cec96d95..a4d005eccc7f 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -603,6 +603,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 				    .uli_u = { .ports =
 					       { .sport = inet->sport,
 						 .dport = dport } } };
+		security_sk_classify_flow(sk, &fl);
 		err = ip_route_output_flow(&rt, &fl, sk, !(msg->msg_flags&MSG_DONTWAIT));
 		if (err)
 			goto out;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index ac85e9c532c2..82a1b1a328db 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -637,6 +637,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 		fl.oif = sk->sk_bound_dev_if;
 		fl.fl_ip_dport = inet->dport;
 		fl.fl_ip_sport = inet->sport;
+		security_sk_classify_flow(sk, &fl);
 
 		if (np->opt && np->opt->srcrt) {
 			struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index 3b55b4c8e2d1..c73508e090a6 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -156,6 +156,8 @@ ipv4_connected:
 	if (!fl.oif && (addr_type&IPV6_ADDR_MULTICAST))
 		fl.oif = np->mcast_oif;
 
+	security_sk_classify_flow(sk, &fl);
+
 	if (flowlabel) {
 		if (flowlabel->opt && flowlabel->opt->srcrt) {
 			struct rt0_hdr *rt0 = (struct rt0_hdr *) flowlabel->opt->srcrt;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 356a8a7ef22a..dbfce089e916 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -358,6 +358,7 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 	fl.oif = iif;
 	fl.fl_icmp_type = type;
 	fl.fl_icmp_code = code;
+	security_skb_classify_flow(skb, &fl);
 
 	if (icmpv6_xmit_lock())
 		return;
@@ -472,6 +473,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
 		ipv6_addr_copy(&fl.fl6_src, saddr);
 	fl.oif = skb->dev->ifindex;
 	fl.fl_icmp_type = ICMPV6_ECHO_REPLY;
+	security_skb_classify_flow(skb, &fl);
 
 	if (icmpv6_xmit_lock())
 		return;
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index bf491077b822..7a51a258615d 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -157,6 +157,7 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok)
 	fl.oif = sk->sk_bound_dev_if;
 	fl.fl_ip_sport = inet->sport;
 	fl.fl_ip_dport = inet->dport;
+	security_sk_classify_flow(sk, &fl);
 
 	if (np->opt && np->opt->srcrt) {
 		struct rt0_hdr *rt0 = (struct rt0_hdr *)np->opt->srcrt;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index b50055b9278d..67cfc3813c32 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -419,6 +419,7 @@ static inline void ndisc_flow_init(struct flowi *fl, u8 type,
 	fl->proto	 	= IPPROTO_ICMPV6;
 	fl->fl_icmp_type	= type;
 	fl->fl_icmp_code	= 0;
+	security_sk_classify_flow(ndisc_socket->sk, fl);
 }
 
 static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index 8629ba195d2d..c4eba1aeb323 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -96,6 +96,7 @@ static void send_reset(struct sk_buff *oldskb)
 	ipv6_addr_copy(&fl.fl6_dst, &oip6h->saddr);
 	fl.fl_ip_sport = otcph.dest;
 	fl.fl_ip_dport = otcph.source;
+	security_skb_classify_flow(oldskb, &fl);
 	dst = ip6_route_output(NULL, &fl);
 	if (dst == NULL)
 		return;
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 15b862d8acab..d5040e172292 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -759,6 +759,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 
 	if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst))
 		fl.oif = np->mcast_oif;
+	security_sk_classify_flow(sk, &fl);
 
 	err = ip6_dst_lookup(sk, &dst, &fl);
 	if (err)
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 802a1a6b1037..46922e57e311 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -251,6 +251,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 		final_p = &final;
 	}
 
+	security_sk_classify_flow(sk, &fl);
+
 	err = ip6_dst_lookup(sk, &dst, &fl);
 	if (err)
 		goto failure;
@@ -374,6 +376,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			fl.oif = sk->sk_bound_dev_if;
 			fl.fl_ip_dport = inet->dport;
 			fl.fl_ip_sport = inet->sport;
+			security_skb_classify_flow(skb, &fl);
 
 			if ((err = ip6_dst_lookup(sk, &dst, &fl))) {
 				sk->sk_err_soft = -err;
@@ -467,6 +470,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
 	fl.oif = treq->iif;
 	fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 	fl.fl_ip_sport = inet_sk(sk)->sport;
+	security_sk_classify_flow(sk, &fl);
 
 	if (dst == NULL) {
 		opt = np->opt;
@@ -625,6 +629,7 @@ static void tcp_v6_send_reset(struct sk_buff *skb)
 	fl.oif = inet6_iif(skb);
 	fl.fl_ip_dport = t1->dest;
 	fl.fl_ip_sport = t1->source;
+	security_skb_classify_flow(skb, &fl);
 
 	/* sk = NULL, but it is safe for now. RST socket required. */
 	if (!ip6_dst_lookup(NULL, &buff->dst, &fl)) {
@@ -691,6 +696,7 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32
 	fl.oif = inet6_iif(skb);
 	fl.fl_ip_dport = t1->dest;
 	fl.fl_ip_sport = t1->source;
+	security_skb_classify_flow(skb, &fl);
 
 	if (!ip6_dst_lookup(NULL, &buff->dst, &fl)) {
 		if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
@@ -923,6 +929,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 		fl.oif = sk->sk_bound_dev_if;
 		fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 		fl.fl_ip_sport = inet_sk(sk)->sport;
+		security_sk_classify_flow(sk, &fl);
 
 		if (ip6_dst_lookup(sk, &dst, &fl))
 			goto out;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 3d54f246411e..82c7c9cde2a8 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -782,6 +782,8 @@ do_udp_sendmsg:
 		connected = 0;
 	}
 
+	security_sk_classify_flow(sk, fl);
+
 	err = ip6_sk_dst_lookup(sk, &dst, fl);
 	if (err)
 		goto out;
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 79405daadc52..32c963c90573 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -863,7 +863,6 @@ int xfrm_lookup(struct dst_entry **dst_p, struct flowi *fl,
 	u16 family;
 	u8 dir = policy_to_flow_dir(XFRM_POLICY_OUT);
 
-	fl->secid = security_sk_sid(sk, fl, dir);
 restart:
 	genid = atomic_read(&flow_cache_genid);
 	policy = NULL;
@@ -1039,7 +1038,7 @@ xfrm_decode_session(struct sk_buff *skb, struct flowi *fl, unsigned short family
 		return -EAFNOSUPPORT;
 
 	afinfo->decode_session(skb, fl);
-	err = security_xfrm_decode_session(skb, fl);
+	err = security_xfrm_decode_session(skb, &fl->secid);
 	xfrm_policy_put_afinfo(afinfo);
 	return err;
 }
diff --git a/security/dummy.c b/security/dummy.c
index c1f10654871e..c0ff6b9bfd7d 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -809,9 +809,8 @@ static inline void dummy_sk_clone_security (const struct sock *sk, struct sock *
 {
 }
 
-static unsigned int dummy_sk_getsid(struct sock *sk, struct flowi *fl, u8 dir)
+static inline void dummy_sk_getsecid(struct sock *sk, u32 *secid)
 {
-	return 0;
 }
 #endif	/* CONFIG_SECURITY_NETWORK */
 
@@ -866,7 +865,7 @@ static int dummy_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm
 	return 1;
 }
 
-static int dummy_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+static int dummy_xfrm_decode_session(struct sk_buff *skb, u32 *fl, int ckall)
 {
 	return 0;
 }
@@ -1083,7 +1082,7 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, sk_alloc_security);
 	set_to_dummy_if_null(ops, sk_free_security);
 	set_to_dummy_if_null(ops, sk_clone_security);
-	set_to_dummy_if_null(ops, sk_getsid);
+	set_to_dummy_if_null(ops, sk_getsecid);
  #endif	/* CONFIG_SECURITY_NETWORK */
 #ifdef  CONFIG_SECURITY_NETWORK_XFRM
 	set_to_dummy_if_null(ops, xfrm_policy_alloc_security);
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 5c189da07bc9..4e5989d584ce 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3561,14 +3561,14 @@ static void selinux_sk_clone_security(const struct sock *sk, struct sock *newsk)
 	newssec->peer_sid = ssec->peer_sid;
 }
 
-static unsigned int selinux_sk_getsid_security(struct sock *sk, struct flowi *fl, u8 dir)
+static void selinux_sk_getsecid(struct sock *sk, u32 *secid)
 {
 	if (!sk)
-		return selinux_no_sk_sid(fl);
+		*secid = SECINITSID_ANY_SOCKET;
 	else {
 		struct sk_security_struct *sksec = sk->sk_security;
 
-		return sksec->sid;
+		*secid = sksec->sid;
 	}
 }
 
@@ -4622,7 +4622,7 @@ static struct security_operations selinux_ops = {
 	.sk_alloc_security =		selinux_sk_alloc_security,
 	.sk_free_security =		selinux_sk_free_security,
 	.sk_clone_security =		selinux_sk_clone_security,
-	.sk_getsid = 			selinux_sk_getsid_security,
+	.sk_getsecid = 			selinux_sk_getsecid,
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 	.xfrm_policy_alloc_security =	selinux_xfrm_policy_alloc,
diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h
index f51a3e84bd9b..8e45c1d588a8 100644
--- a/security/selinux/include/xfrm.h
+++ b/security/selinux/include/xfrm.h
@@ -19,7 +19,7 @@ int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir);
 int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x,
 			struct xfrm_policy *xp, struct flowi *fl);
 int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm);
-int selinux_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl);
+int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *fl, int ckall);
 
 
 /*
@@ -33,18 +33,6 @@ static inline struct inode_security_struct *get_sock_isec(struct sock *sk)
 	return SOCK_INODE(sk->sk_socket)->i_security;
 }
 
-
-static inline u32 selinux_no_sk_sid(struct flowi *fl)
-{
-	/* NOTE: no sock occurs on ICMP reply, forwards, ... */
-	/* icmp_reply: authorize as kernel packet */
-	if (fl && fl->proto == IPPROTO_ICMP) {
-		return SECINITSID_KERNEL;
-	}
-
-	return SECINITSID_ANY_SOCKET;
-}
-
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb,
 			struct avc_audit_data *ad);
diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c
index a502b0540e3d..c750ef7af66f 100644
--- a/security/selinux/xfrm.c
+++ b/security/selinux/xfrm.c
@@ -158,11 +158,11 @@ int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm)
  * LSM hook implementation that determines the sid for the session.
  */
 
-int selinux_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
+int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall)
 {
 	struct sec_path *sp;
 
-	fl->secid = SECSID_NULL;
+	*sid = SECSID_NULL;
 
 	if (skb == NULL)
 		return 0;
@@ -177,10 +177,13 @@ int selinux_xfrm_decode_session(struct sk_buff *skb, struct flowi *fl)
 				struct xfrm_sec_ctx *ctx = x->security;
 
 				if (!sid_set) {
-					fl->secid = ctx->ctx_sid;
+					*sid = ctx->ctx_sid;
 					sid_set = 1;
+
+					if (!ckall)
+						break;
 				}
-				else if (fl->secid != ctx->ctx_sid)
+				else if (*sid != ctx->ctx_sid)
 					return -EINVAL;
 			}
 		}
-- 
cgit v1.2.3


From cb969f072b6d67770b559617f14e767f47e77ece Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Mon, 24 Jul 2006 23:32:20 -0700
Subject: [MLSXFRM]: Default labeling of socket specific IPSec policies

This defaults the label of socket-specific IPSec policies to be the
same as the socket they are set on.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h        | 19 ++++++++++++++++---
 include/net/xfrm.h              |  2 +-
 net/key/af_key.c                | 15 +++++++++++----
 net/xfrm/xfrm_state.c           |  2 +-
 net/xfrm/xfrm_user.c            | 13 +++++++++++--
 security/dummy.c                |  3 ++-
 security/selinux/include/xfrm.h |  3 ++-
 security/selinux/xfrm.c         | 33 ++++++++++++++++++++++-----------
 8 files changed, 66 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index f3909d189fe0..8e3dc6c51a6d 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -827,8 +827,10 @@ struct swap_info_struct;
  *	used by the XFRM system.
  *	@sec_ctx contains the security context information being provided by
  *	the user-level policy update program (e.g., setkey).
+ *	@sk refers to the sock from which to derive the security context.
  *	Allocate a security structure to the xp->security field; the security
- *	field is initialized to NULL when the xfrm_policy is allocated.
+ *	field is initialized to NULL when the xfrm_policy is allocated. Only
+ *	one of sec_ctx or sock can be specified.
  *	Return 0 if operation was successful (memory to allocate, legal context)
  * @xfrm_policy_clone_security:
  *	@old contains an existing xfrm_policy in the SPD.
@@ -1359,7 +1361,8 @@ struct security_operations {
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
-	int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx);
+	int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp,
+			struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk);
 	int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new);
 	void (*xfrm_policy_free_security) (struct xfrm_policy *xp);
 	int (*xfrm_policy_delete_security) (struct xfrm_policy *xp);
@@ -3057,7 +3060,12 @@ static inline void security_sk_classify_flow(struct sock *sk, struct flowi *fl)
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx)
 {
-	return security_ops->xfrm_policy_alloc_security(xp, sec_ctx);
+	return security_ops->xfrm_policy_alloc_security(xp, sec_ctx, NULL);
+}
+
+static inline int security_xfrm_sock_policy_alloc(struct xfrm_policy *xp, struct sock *sk)
+{
+	return security_ops->xfrm_policy_alloc_security(xp, NULL, sk);
 }
 
 static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new)
@@ -3132,6 +3140,11 @@ static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm
 	return 0;
 }
 
+static inline int security_xfrm_sock_policy_alloc(struct xfrm_policy *xp, struct sock *sk)
+{
+	return 0;
+}
+
 static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new)
 {
 	return 0;
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 3ecd9fa1ed4b..00bf86e6e82b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -362,7 +362,7 @@ struct xfrm_mgr
 	char			*id;
 	int			(*notify)(struct xfrm_state *x, struct km_event *c);
 	int			(*acquire)(struct xfrm_state *x, struct xfrm_tmpl *, struct xfrm_policy *xp, int dir);
-	struct xfrm_policy	*(*compile_policy)(u16 family, int opt, u8 *data, int len, int *dir);
+	struct xfrm_policy	*(*compile_policy)(struct sock *sk, int opt, u8 *data, int len, int *dir);
 	int			(*new_mapping)(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport);
 	int			(*notify_policy)(struct xfrm_policy *x, int dir, struct km_event *c);
 };
diff --git a/net/key/af_key.c b/net/key/af_key.c
index a065e1a67773..797c744a8438 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -2843,14 +2843,14 @@ static int pfkey_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *t, struct
 	return pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_REGISTERED, NULL);
 }
 
-static struct xfrm_policy *pfkey_compile_policy(u16 family, int opt,
+static struct xfrm_policy *pfkey_compile_policy(struct sock *sk, int opt,
                                                 u8 *data, int len, int *dir)
 {
 	struct xfrm_policy *xp;
 	struct sadb_x_policy *pol = (struct sadb_x_policy*)data;
 	struct sadb_x_sec_ctx *sec_ctx;
 
-	switch (family) {
+	switch (sk->sk_family) {
 	case AF_INET:
 		if (opt != IP_IPSEC_POLICY) {
 			*dir = -EOPNOTSUPP;
@@ -2891,7 +2891,7 @@ static struct xfrm_policy *pfkey_compile_policy(u16 family, int opt,
 	xp->lft.hard_byte_limit = XFRM_INF;
 	xp->lft.soft_packet_limit = XFRM_INF;
 	xp->lft.hard_packet_limit = XFRM_INF;
-	xp->family = family;
+	xp->family = sk->sk_family;
 
 	xp->xfrm_nr = 0;
 	if (pol->sadb_x_policy_type == IPSEC_POLICY_IPSEC &&
@@ -2907,8 +2907,10 @@ static struct xfrm_policy *pfkey_compile_policy(u16 family, int opt,
 		p += pol->sadb_x_policy_len*8;
 		sec_ctx = (struct sadb_x_sec_ctx *)p;
 		if (len < pol->sadb_x_policy_len*8 +
-		    sec_ctx->sadb_x_sec_len)
+		    sec_ctx->sadb_x_sec_len) {
+			*dir = -EINVAL;
 			goto out;
+		}
 		if ((*dir = verify_sec_ctx_len(p)))
 			goto out;
 		uctx = pfkey_sadb2xfrm_user_sec_ctx(sec_ctx);
@@ -2918,6 +2920,11 @@ static struct xfrm_policy *pfkey_compile_policy(u16 family, int opt,
 		if (*dir)
 			goto out;
 	}
+	else {
+		*dir = security_xfrm_sock_policy_alloc(xp, sk);
+		if (*dir)
+			goto out;
+	}
 
 	*dir = pol->sadb_x_policy_dir-1;
 	return xp;
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index be02bd981d12..1c796087ee78 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1026,7 +1026,7 @@ int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen
 	err = -EINVAL;
 	read_lock(&xfrm_km_lock);
 	list_for_each_entry(km, &xfrm_km_list, list) {
-		pol = km->compile_policy(sk->sk_family, optname, data,
+		pol = km->compile_policy(sk, optname, data,
 					 optlen, &err);
 		if (err >= 0)
 			break;
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index dac8db1088bc..f70e158874d2 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -1757,7 +1757,7 @@ static int xfrm_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *xt,
 /* User gives us xfrm_user_policy_info followed by an array of 0
  * or more templates.
  */
-static struct xfrm_policy *xfrm_compile_policy(u16 family, int opt,
+static struct xfrm_policy *xfrm_compile_policy(struct sock *sk, int opt,
 					       u8 *data, int len, int *dir)
 {
 	struct xfrm_userpolicy_info *p = (struct xfrm_userpolicy_info *)data;
@@ -1765,7 +1765,7 @@ static struct xfrm_policy *xfrm_compile_policy(u16 family, int opt,
 	struct xfrm_policy *xp;
 	int nr;
 
-	switch (family) {
+	switch (sk->sk_family) {
 	case AF_INET:
 		if (opt != IP_XFRM_POLICY) {
 			*dir = -EOPNOTSUPP;
@@ -1807,6 +1807,15 @@ static struct xfrm_policy *xfrm_compile_policy(u16 family, int opt,
 	copy_from_user_policy(xp, p);
 	copy_templates(xp, ut, nr);
 
+	if (!xp->security) {
+		int err = security_xfrm_sock_policy_alloc(xp, sk);
+		if (err) {
+			kfree(xp);
+			*dir = err;
+			return NULL;
+		}
+	}
+
 	*dir = p->dir;
 
 	return xp;
diff --git a/security/dummy.c b/security/dummy.c
index c0ff6b9bfd7d..66cc06404930 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -815,7 +815,8 @@ static inline void dummy_sk_getsecid(struct sock *sk, u32 *secid)
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
-static int dummy_xfrm_policy_alloc_security(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx)
+static int dummy_xfrm_policy_alloc_security(struct xfrm_policy *xp,
+		struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk)
 {
 	return 0;
 }
diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h
index 8e45c1d588a8..1822c73e5085 100644
--- a/security/selinux/include/xfrm.h
+++ b/security/selinux/include/xfrm.h
@@ -7,7 +7,8 @@
 #ifndef _SELINUX_XFRM_H_
 #define _SELINUX_XFRM_H_
 
-int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx);
+int selinux_xfrm_policy_alloc(struct xfrm_policy *xp,
+		struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk);
 int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new);
 void selinux_xfrm_policy_free(struct xfrm_policy *xp);
 int selinux_xfrm_policy_delete(struct xfrm_policy *xp);
diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c
index c750ef7af66f..d3690f985135 100644
--- a/security/selinux/xfrm.c
+++ b/security/selinux/xfrm.c
@@ -208,10 +208,8 @@ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp,
 
 	BUG_ON(uctx && pol);
 
-	if (pol)
-		goto from_policy;
-
-	BUG_ON(!uctx);
+	if (!uctx)
+		goto not_from_user;
 
 	if (uctx->ctx_doi != XFRM_SC_ALG_SELINUX)
 		return -EINVAL;
@@ -251,11 +249,14 @@ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp,
 
 	return rc;
 
-from_policy:
-	BUG_ON(!pol);
-	rc = security_sid_mls_copy(pol->ctx_sid, sid, &ctx_sid);
-	if (rc)
-		goto out;
+not_from_user:
+	if (pol) {
+		rc = security_sid_mls_copy(pol->ctx_sid, sid, &ctx_sid);
+		if (rc)
+			goto out;
+	}
+	else
+		ctx_sid = sid;
 
 	rc = security_sid_to_context(ctx_sid, &ctx_str, &str_len);
 	if (rc)
@@ -293,13 +294,23 @@ out2:
  * LSM hook implementation that allocs and transfers uctx spec to
  * xfrm_policy.
  */
-int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *uctx)
+int selinux_xfrm_policy_alloc(struct xfrm_policy *xp,
+		struct xfrm_user_sec_ctx *uctx, struct sock *sk)
 {
 	int err;
+	u32 sid;
 
 	BUG_ON(!xp);
+	BUG_ON(uctx && sk);
+
+	if (sk) {
+		struct sk_security_struct *ssec = sk->sk_security;
+		sid = ssec->sid;
+	}
+	else
+		sid = SECSID_NULL;
 
-	err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, NULL, 0);
+	err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, NULL, sid);
 	return err;
 }
 
-- 
cgit v1.2.3


From 4237c75c0a35535d7f9f2bfeeb4b4df1e068a0bf Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Mon, 24 Jul 2006 23:32:50 -0700
Subject: [MLSXFRM]: Auto-labeling of child sockets

This automatically labels the TCP, Unix stream, and dccp child sockets
as well as openreqs to be at the same MLS level as the peer. This will
result in the selection of appropriately labeled IPSec Security
Associations.

This also uses the sock's sid (as opposed to the isec sid) in SELinux
enforcement of secmark in rcv_skb and postroute_last hooks.

Signed-off-by: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h        |  55 ++++++++++++++++
 include/net/request_sock.h      |   1 +
 include/net/sock.h              |   1 +
 net/dccp/ipv4.c                 |   3 +
 net/dccp/ipv6.c                 |   7 +-
 net/ipv4/inet_connection_sock.c |   4 +-
 net/ipv4/syncookies.c           |   6 +-
 net/ipv4/tcp_ipv4.c             |   3 +
 net/ipv6/tcp_ipv6.c             |   6 +-
 security/dummy.c                |  24 +++++++
 security/selinux/hooks.c        | 137 +++++++++++++++++++++++++++-------------
 security/selinux/xfrm.c         |   1 -
 12 files changed, 197 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 8e3dc6c51a6d..bb4c80fdfe7a 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -90,6 +90,7 @@ extern int cap_netlink_recv(struct sk_buff *skb, int cap);
 struct nfsctl_arg;
 struct sched_param;
 struct swap_info_struct;
+struct request_sock;
 
 /* bprm_apply_creds unsafe reasons */
 #define LSM_UNSAFE_SHARE	1
@@ -819,6 +820,14 @@ struct swap_info_struct;
  * @sk_getsecid:
  *	Retrieve the LSM-specific secid for the sock to enable caching of network
  *	authorizations.
+ * @sock_graft:
+ *	Sets the socket's isec sid to the sock's sid.
+ * @inet_conn_request:
+ *	Sets the openreq's sid to socket's sid with MLS portion taken from peer sid.
+ * @inet_csk_clone:
+ *	Sets the new child socket's sid to the openreq sid.
+ * @req_classify_flow:
+ *	Sets the flow's sid to the openreq sid.
  *
  * Security hooks for XFRM operations.
  *
@@ -1358,6 +1367,11 @@ struct security_operations {
 	void (*sk_free_security) (struct sock *sk);
 	void (*sk_clone_security) (const struct sock *sk, struct sock *newsk);
 	void (*sk_getsecid) (struct sock *sk, u32 *secid);
+	void (*sock_graft)(struct sock* sk, struct socket *parent);
+	int (*inet_conn_request)(struct sock *sk, struct sk_buff *skb,
+					struct request_sock *req);
+	void (*inet_csk_clone)(struct sock *newsk, const struct request_sock *req);
+	void (*req_classify_flow)(const struct request_sock *req, struct flowi *fl);
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
@@ -2926,6 +2940,28 @@ static inline void security_sk_classify_flow(struct sock *sk, struct flowi *fl)
 {
 	security_ops->sk_getsecid(sk, &fl->secid);
 }
+
+static inline void security_req_classify_flow(const struct request_sock *req, struct flowi *fl)
+{
+	security_ops->req_classify_flow(req, fl);
+}
+
+static inline void security_sock_graft(struct sock* sk, struct socket *parent)
+{
+	security_ops->sock_graft(sk, parent);
+}
+
+static inline int security_inet_conn_request(struct sock *sk,
+			struct sk_buff *skb, struct request_sock *req)
+{
+	return security_ops->inet_conn_request(sk, skb, req);
+}
+
+static inline void security_inet_csk_clone(struct sock *newsk,
+			const struct request_sock *req)
+{
+	security_ops->inet_csk_clone(newsk, req);
+}
 #else	/* CONFIG_SECURITY_NETWORK */
 static inline int security_unix_stream_connect(struct socket * sock,
 					       struct socket * other, 
@@ -3055,6 +3091,25 @@ static inline void security_sk_clone(const struct sock *sk, struct sock *newsk)
 static inline void security_sk_classify_flow(struct sock *sk, struct flowi *fl)
 {
 }
+
+static inline void security_req_classify_flow(const struct request_sock *req, struct flowi *fl)
+{
+}
+
+static inline void security_sock_graft(struct sock* sk, struct socket *parent)
+{
+}
+
+static inline int security_inet_conn_request(struct sock *sk,
+			struct sk_buff *skb, struct request_sock *req)
+{
+	return 0;
+}
+
+static inline void security_inet_csk_clone(struct sock *newsk,
+			const struct request_sock *req)
+{
+}
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index c5d7f920c352..8e165ca16bd8 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -53,6 +53,7 @@ struct request_sock {
 	unsigned long			expires;
 	struct request_sock_ops		*rsk_ops;
 	struct sock			*sk;
+	u32				secid;
 };
 
 static inline struct request_sock *reqsk_alloc(struct request_sock_ops *ops)
diff --git a/include/net/sock.h b/include/net/sock.h
index 91cdceb3c028..337ebec84c70 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -969,6 +969,7 @@ static inline void sock_graft(struct sock *sk, struct socket *parent)
 	sk->sk_sleep = &parent->wait;
 	parent->sk = sk;
 	sk->sk_socket = parent;
+	security_sock_graft(sk, parent);
 	write_unlock_bh(&sk->sk_callback_lock);
 }
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 386498053b1c..171d363876ee 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -501,6 +501,9 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	dccp_openreq_init(req, &dp, skb);
 
+	if (security_inet_conn_request(sk, skb, req))
+		goto drop_and_free;
+
 	ireq = inet_rsk(req);
 	ireq->loc_addr = daddr;
 	ireq->rmt_addr = saddr;
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 53d255c01431..231bc7c7e749 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -424,7 +424,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req,
 	fl.oif = ireq6->iif;
 	fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 	fl.fl_ip_sport = inet_sk(sk)->sport;
-	security_sk_classify_flow(sk, &fl);
+	security_req_classify_flow(req, &fl);
 
 	if (dst == NULL) {
 		opt = np->opt;
@@ -626,7 +626,7 @@ static void dccp_v6_reqsk_send_ack(struct sk_buff *rxskb,
 	fl.oif = inet6_iif(rxskb);
 	fl.fl_ip_dport = dh->dccph_dport;
 	fl.fl_ip_sport = dh->dccph_sport;
-	security_skb_classify_flow(rxskb, &fl);
+	security_req_classify_flow(req, &fl);
 
 	if (!ip6_dst_lookup(NULL, &skb->dst, &fl)) {
 		if (xfrm_lookup(&skb->dst, &fl, NULL, 0) >= 0) {
@@ -709,6 +709,9 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	dccp_openreq_init(req, &dp, skb);
 
+	if (security_inet_conn_request(sk, skb, req))
+		goto drop_and_free;
+
 	ireq6 = inet6_rsk(req);
 	ireq = inet_rsk(req);
 	ipv6_addr_copy(&ireq6->rmt_addr, &skb->nh.ipv6h->saddr);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 772b4eac78bc..07204391d083 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -327,7 +327,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 				       { .sport = inet_sk(sk)->sport,
 					 .dport = ireq->rmt_port } } };
 
-	security_sk_classify_flow(sk, &fl);
+	security_req_classify_flow(req, &fl);
 	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
 		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
 		return NULL;
@@ -510,6 +510,8 @@ struct sock *inet_csk_clone(struct sock *sk, const struct request_sock *req,
 
 		/* Deinitialize accept_queue to trap illegal accesses. */
 		memset(&newicsk->icsk_accept_queue, 0, sizeof(newicsk->icsk_accept_queue));
+
+		security_inet_csk_clone(newsk, req);
 	}
 	return newsk;
 }
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 307dc3c0d635..661e0a4bca72 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -214,6 +214,10 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	if (!req)
 		goto out;
 
+	if (security_inet_conn_request(sk, skb, req)) {
+		reqsk_free(req);
+		goto out;
+	}
 	ireq = inet_rsk(req);
 	treq = tcp_rsk(req);
 	treq->rcv_isn		= htonl(skb->h.th->seq) - 1;
@@ -259,7 +263,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 				    .uli_u = { .ports =
 					       { .sport = skb->h.th->dest,
 						 .dport = skb->h.th->source } } };
-		security_sk_classify_flow(sk, &fl);
+		security_req_classify_flow(req, &fl);
 		if (ip_route_output_key(&rt, &fl)) {
 			reqsk_free(req);
 			goto out; 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 4b04c3edd4a9..43f6740244f8 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -798,6 +798,9 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	tcp_openreq_init(req, &tmp_opt, skb);
 
+	if (security_inet_conn_request(sk, skb, req))
+		goto drop_and_free;
+
 	ireq = inet_rsk(req);
 	ireq->loc_addr = daddr;
 	ireq->rmt_addr = saddr;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 46922e57e311..302786a11cd6 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -470,7 +470,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
 	fl.oif = treq->iif;
 	fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 	fl.fl_ip_sport = inet_sk(sk)->sport;
-	security_sk_classify_flow(sk, &fl);
+	security_req_classify_flow(req, &fl);
 
 	if (dst == NULL) {
 		opt = np->opt;
@@ -826,6 +826,8 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
 
 	tcp_rsk(req)->snt_isn = isn;
 
+	security_inet_conn_request(sk, skb, req);
+
 	if (tcp_v6_send_synack(sk, req, NULL))
 		goto drop;
 
@@ -929,7 +931,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 		fl.oif = sk->sk_bound_dev_if;
 		fl.fl_ip_dport = inet_rsk(req)->rmt_port;
 		fl.fl_ip_sport = inet_sk(sk)->sport;
-		security_sk_classify_flow(sk, &fl);
+		security_req_classify_flow(req, &fl);
 
 		if (ip6_dst_lookup(sk, &dst, &fl))
 			goto out;
diff --git a/security/dummy.c b/security/dummy.c
index 66cc06404930..1c45f8e4aad1 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -812,6 +812,26 @@ static inline void dummy_sk_clone_security (const struct sock *sk, struct sock *
 static inline void dummy_sk_getsecid(struct sock *sk, u32 *secid)
 {
 }
+
+static inline void dummy_sock_graft(struct sock* sk, struct socket *parent)
+{
+}
+
+static inline int dummy_inet_conn_request(struct sock *sk,
+			struct sk_buff *skb, struct request_sock *req)
+{
+	return 0;
+}
+
+static inline void dummy_inet_csk_clone(struct sock *newsk,
+			const struct request_sock *req)
+{
+}
+
+static inline void dummy_req_classify_flow(const struct request_sock *req,
+			struct flowi *fl)
+{
+}
 #endif	/* CONFIG_SECURITY_NETWORK */
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
@@ -1084,6 +1104,10 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, sk_free_security);
 	set_to_dummy_if_null(ops, sk_clone_security);
 	set_to_dummy_if_null(ops, sk_getsecid);
+	set_to_dummy_if_null(ops, sock_graft);
+	set_to_dummy_if_null(ops, inet_conn_request);
+	set_to_dummy_if_null(ops, inet_csk_clone);
+	set_to_dummy_if_null(ops, req_classify_flow);
  #endif	/* CONFIG_SECURITY_NETWORK */
 #ifdef  CONFIG_SECURITY_NETWORK_XFRM
 	set_to_dummy_if_null(ops, xfrm_policy_alloc_security);
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 4e5989d584ce..1dc935f7b919 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3328,8 +3328,9 @@ static int selinux_socket_unix_stream_connect(struct socket *sock,
 	/* server child socket */
 	ssec = newsk->sk_security;
 	ssec->peer_sid = isec->sid;
-	
-	return 0;
+	err = security_sid_mls_copy(other_isec->sid, ssec->peer_sid, &ssec->sid);
+
+	return err;
 }
 
 static int selinux_socket_unix_may_send(struct socket *sock,
@@ -3355,11 +3356,29 @@ static int selinux_socket_unix_may_send(struct socket *sock,
 }
 
 static int selinux_sock_rcv_skb_compat(struct sock *sk, struct sk_buff *skb,
-		struct avc_audit_data *ad, u32 sock_sid, u16 sock_class,
-		u16 family, char *addrp, int len)
+		struct avc_audit_data *ad, u16 family, char *addrp, int len)
 {
 	int err = 0;
 	u32 netif_perm, node_perm, node_sid, if_sid, recv_perm = 0;
+	struct socket *sock;
+	u16 sock_class = 0;
+	u32 sock_sid = 0;
+
+ 	read_lock_bh(&sk->sk_callback_lock);
+ 	sock = sk->sk_socket;
+ 	if (sock) {
+ 		struct inode *inode;
+ 		inode = SOCK_INODE(sock);
+ 		if (inode) {
+ 			struct inode_security_struct *isec;
+ 			isec = inode->i_security;
+ 			sock_sid = isec->sid;
+ 			sock_class = isec->sclass;
+ 		}
+ 	}
+ 	read_unlock_bh(&sk->sk_callback_lock);
+ 	if (!sock_sid)
+  		goto out;
 
 	if (!skb->dev)
 		goto out;
@@ -3419,12 +3438,10 @@ out:
 static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 {
 	u16 family;
-	u16 sock_class = 0;
 	char *addrp;
 	int len, err = 0;
-	u32 sock_sid = 0;
-	struct socket *sock;
 	struct avc_audit_data ad;
+	struct sk_security_struct *sksec = sk->sk_security;
 
 	family = sk->sk_family;
 	if (family != PF_INET && family != PF_INET6)
@@ -3434,22 +3451,6 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (family == PF_INET6 && skb->protocol == ntohs(ETH_P_IP))
 		family = PF_INET;
 
- 	read_lock_bh(&sk->sk_callback_lock);
- 	sock = sk->sk_socket;
- 	if (sock) {
- 		struct inode *inode;
- 		inode = SOCK_INODE(sock);
- 		if (inode) {
- 			struct inode_security_struct *isec;
- 			isec = inode->i_security;
- 			sock_sid = isec->sid;
- 			sock_class = isec->sclass;
- 		}
- 	}
- 	read_unlock_bh(&sk->sk_callback_lock);
- 	if (!sock_sid)
-  		goto out;
-
 	AVC_AUDIT_DATA_INIT(&ad, NET);
 	ad.u.net.netif = skb->dev ? skb->dev->name : "[unknown]";
 	ad.u.net.family = family;
@@ -3459,16 +3460,15 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 		goto out;
 
 	if (selinux_compat_net)
-		err = selinux_sock_rcv_skb_compat(sk, skb, &ad, sock_sid,
-						  sock_class, family,
+		err = selinux_sock_rcv_skb_compat(sk, skb, &ad, family,
 						  addrp, len);
 	else
-		err = avc_has_perm(sock_sid, skb->secmark, SECCLASS_PACKET,
+		err = avc_has_perm(sksec->sid, skb->secmark, SECCLASS_PACKET,
 				   PACKET__RECV, &ad);
 	if (err)
 		goto out;
 
-	err = selinux_xfrm_sock_rcv_skb(sock_sid, skb, &ad);
+	err = selinux_xfrm_sock_rcv_skb(sksec->sid, skb, &ad);
 out:	
 	return err;
 }
@@ -3572,6 +3572,49 @@ static void selinux_sk_getsecid(struct sock *sk, u32 *secid)
 	}
 }
 
+void selinux_sock_graft(struct sock* sk, struct socket *parent)
+{
+	struct inode_security_struct *isec = SOCK_INODE(parent)->i_security;
+	struct sk_security_struct *sksec = sk->sk_security;
+
+	isec->sid = sksec->sid;
+}
+
+int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb,
+					   struct request_sock *req)
+{
+	struct sk_security_struct *sksec = sk->sk_security;
+	int err;
+	u32 newsid = 0;
+	u32 peersid;
+
+	err = selinux_xfrm_decode_session(skb, &peersid, 0);
+	BUG_ON(err);
+
+	err = security_sid_mls_copy(sksec->sid, peersid, &newsid);
+	if (err)
+		return err;
+
+	req->secid = newsid;
+	return 0;
+}
+
+void selinux_inet_csk_clone(struct sock *newsk, const struct request_sock *req)
+{
+	struct sk_security_struct *newsksec = newsk->sk_security;
+
+	newsksec->sid = req->secid;
+	/* NOTE: Ideally, we should also get the isec->sid for the
+	   new socket in sync, but we don't have the isec available yet.
+	   So we will wait until sock_graft to do it, by which
+	   time it will have been created and available. */
+}
+
+void selinux_req_classify_flow(const struct request_sock *req, struct flowi *fl)
+{
+	fl->secid = req->secid;
+}
+
 static int selinux_nlmsg_perm(struct sock *sk, struct sk_buff *skb)
 {
 	int err = 0;
@@ -3611,12 +3654,24 @@ out:
 #ifdef CONFIG_NETFILTER
 
 static int selinux_ip_postroute_last_compat(struct sock *sk, struct net_device *dev,
-					    struct inode_security_struct *isec,
 					    struct avc_audit_data *ad,
 					    u16 family, char *addrp, int len)
 {
-	int err;
+	int err = 0;
 	u32 netif_perm, node_perm, node_sid, if_sid, send_perm = 0;
+	struct socket *sock;
+	struct inode *inode;
+	struct inode_security_struct *isec;
+
+	sock = sk->sk_socket;
+	if (!sock)
+		goto out;
+
+	inode = SOCK_INODE(sock);
+	if (!inode)
+		goto out;
+
+	isec = inode->i_security;
 	
 	err = sel_netif_sids(dev, &if_sid, NULL);
 	if (err)
@@ -3681,26 +3736,16 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum,
 	char *addrp;
 	int len, err = 0;
 	struct sock *sk;
-	struct socket *sock;
-	struct inode *inode;
 	struct sk_buff *skb = *pskb;
-	struct inode_security_struct *isec;
 	struct avc_audit_data ad;
 	struct net_device *dev = (struct net_device *)out;
+	struct sk_security_struct *sksec;
 
 	sk = skb->sk;
 	if (!sk)
 		goto out;
 
-	sock = sk->sk_socket;
-	if (!sock)
-		goto out;
-
-	inode = SOCK_INODE(sock);
-	if (!inode)
-		goto out;
-
-	isec = inode->i_security;
+	sksec = sk->sk_security;
 
 	AVC_AUDIT_DATA_INIT(&ad, NET);
 	ad.u.net.netif = dev->name;
@@ -3711,16 +3756,16 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum,
 		goto out;
 
 	if (selinux_compat_net)
-		err = selinux_ip_postroute_last_compat(sk, dev, isec, &ad,
+		err = selinux_ip_postroute_last_compat(sk, dev, &ad,
 						       family, addrp, len);
 	else
-		err = avc_has_perm(isec->sid, skb->secmark, SECCLASS_PACKET,
+		err = avc_has_perm(sksec->sid, skb->secmark, SECCLASS_PACKET,
 				   PACKET__SEND, &ad);
 
 	if (err)
 		goto out;
 
-	err = selinux_xfrm_postroute_last(isec->sid, skb, &ad);
+	err = selinux_xfrm_postroute_last(sksec->sid, skb, &ad);
 out:
 	return err ? NF_DROP : NF_ACCEPT;
 }
@@ -4623,6 +4668,10 @@ static struct security_operations selinux_ops = {
 	.sk_free_security =		selinux_sk_free_security,
 	.sk_clone_security =		selinux_sk_clone_security,
 	.sk_getsecid = 			selinux_sk_getsecid,
+	.sock_graft =			selinux_sock_graft,
+	.inet_conn_request =		selinux_inet_conn_request,
+	.inet_csk_clone =		selinux_inet_csk_clone,
+	.req_classify_flow =		selinux_req_classify_flow,
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 	.xfrm_policy_alloc_security =	selinux_xfrm_policy_alloc,
diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c
index d3690f985135..3e742b850af6 100644
--- a/security/selinux/xfrm.c
+++ b/security/selinux/xfrm.c
@@ -271,7 +271,6 @@ not_from_user:
 		goto out;
 	}
 
-
 	ctx->ctx_doi = XFRM_SC_DOI_LSM;
 	ctx->ctx_alg = XFRM_SC_ALG_SELINUX;
 	ctx->ctx_sid = ctx_sid;
-- 
cgit v1.2.3


From 11a03f78fbf15a866ba3bf6359a75cdfd1ced703 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Thu, 3 Aug 2006 16:46:20 -0700
Subject: [NetLabel]: core network changes

Changes to the core network stack to support the NetLabel subsystem.  This
includes changes to the IPv4 option handling to support CIPSO labels.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ip.h       |   1 +
 include/net/cipso_ipv4.h | 250 ++++++++++++++++++++++++++++++++++++++++
 include/net/inet_sock.h  |   2 +-
 include/net/netlabel.h   | 291 +++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/ah4.c           |   2 +-
 net/ipv4/ip_options.c    |  19 ++++
 6 files changed, 563 insertions(+), 2 deletions(-)
 create mode 100644 include/net/cipso_ipv4.h
 create mode 100644 include/net/netlabel.h

(limited to 'include')

diff --git a/include/linux/ip.h b/include/linux/ip.h
index 4b55cf1df732..2f4600146f83 100644
--- a/include/linux/ip.h
+++ b/include/linux/ip.h
@@ -57,6 +57,7 @@
 #define IPOPT_SEC	(2 |IPOPT_CONTROL|IPOPT_COPY)
 #define IPOPT_LSRR	(3 |IPOPT_CONTROL|IPOPT_COPY)
 #define IPOPT_TIMESTAMP	(4 |IPOPT_MEASUREMENT)
+#define IPOPT_CIPSO	(6 |IPOPT_CONTROL|IPOPT_COPY)
 #define IPOPT_RR	(7 |IPOPT_CONTROL)
 #define IPOPT_SID	(8 |IPOPT_CONTROL|IPOPT_COPY)
 #define IPOPT_SSRR	(9 |IPOPT_CONTROL|IPOPT_COPY)
diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
new file mode 100644
index 000000000000..c7175e725804
--- /dev/null
+++ b/include/net/cipso_ipv4.h
@@ -0,0 +1,250 @@
+/*
+ * CIPSO - Commercial IP Security Option
+ *
+ * This is an implementation of the CIPSO 2.2 protocol as specified in
+ * draft-ietf-cipso-ipsecurity-01.txt with additional tag types as found in
+ * FIPS-188, copies of both documents can be found in the Documentation
+ * directory.  While CIPSO never became a full IETF RFC standard many vendors
+ * have chosen to adopt the protocol and over the years it has become a
+ * de-facto standard for labeled networking.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program;  if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _CIPSO_IPV4_H
+#define _CIPSO_IPV4_H
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <net/netlabel.h>
+
+/* known doi values */
+#define CIPSO_V4_DOI_UNKNOWN          0x00000000
+
+/* tag types */
+#define CIPSO_V4_TAG_INVALID          0
+#define CIPSO_V4_TAG_RBITMAP          1
+#define CIPSO_V4_TAG_ENUM             2
+#define CIPSO_V4_TAG_RANGE            5
+#define CIPSO_V4_TAG_PBITMAP          6
+#define CIPSO_V4_TAG_FREEFORM         7
+
+/* doi mapping types */
+#define CIPSO_V4_MAP_UNKNOWN          0
+#define CIPSO_V4_MAP_STD              1
+#define CIPSO_V4_MAP_PASS             2
+
+/* limits */
+#define CIPSO_V4_MAX_REM_LVLS         256
+#define CIPSO_V4_INV_LVL              0x80000000
+#define CIPSO_V4_MAX_LOC_LVLS         (CIPSO_V4_INV_LVL - 1)
+#define CIPSO_V4_MAX_REM_CATS         65536
+#define CIPSO_V4_INV_CAT              0x80000000
+#define CIPSO_V4_MAX_LOC_CATS         (CIPSO_V4_INV_CAT - 1)
+
+/*
+ * CIPSO DOI definitions
+ */
+
+/* DOI definition struct */
+#define CIPSO_V4_TAG_MAXCNT           5
+struct cipso_v4_doi {
+	u32 doi;
+	u32 type;
+	union {
+		struct cipso_v4_std_map_tbl *std;
+	} map;
+	u8 tags[CIPSO_V4_TAG_MAXCNT];
+
+	u32 valid;
+	struct list_head list;
+	struct rcu_head rcu;
+	struct list_head dom_list;
+};
+
+/* Standard CIPSO mapping table */
+/* NOTE: the highest order bit (i.e. 0x80000000) is an 'invalid' flag, if the
+ *       bit is set then consider that value as unspecified, meaning the
+ *       mapping for that particular level/category is invalid */
+struct cipso_v4_std_map_tbl {
+	struct {
+		u32 *cipso;
+		u32 *local;
+		u32 cipso_size;
+		u32 local_size;
+	} lvl;
+	struct {
+		u32 *cipso;
+		u32 *local;
+		u32 cipso_size;
+		u32 local_size;
+	} cat;
+};
+
+/*
+ * Sysctl Variables
+ */
+
+#ifdef CONFIG_NETLABEL
+extern int cipso_v4_cache_enabled;
+extern int cipso_v4_cache_bucketsize;
+extern int cipso_v4_rbm_optfmt;
+extern int cipso_v4_rbm_strictvalid;
+#endif
+
+/*
+ * Helper Functions
+ */
+
+#define CIPSO_V4_OPTEXIST(x) (IPCB(x)->opt.cipso != 0)
+#define CIPSO_V4_OPTPTR(x) ((x)->nh.raw + IPCB(x)->opt.cipso)
+
+/*
+ * DOI List Functions
+ */
+
+#ifdef CONFIG_NETLABEL
+int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
+int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head));
+struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi);
+struct sk_buff *cipso_v4_doi_dump_all(size_t headroom);
+struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom);
+int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain);
+int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
+			       const char *domain);
+#else
+static inline int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
+{
+	return -ENOSYS;
+}
+
+static inline int cipso_v4_doi_remove(u32 doi,
+				    void (*callback) (struct rcu_head * head))
+{
+	return 0;
+}
+
+static inline struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
+{
+	return NULL;
+}
+
+static inline struct sk_buff *cipso_v4_doi_dump_all(size_t headroom)
+{
+	return NULL;
+}
+
+static inline struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom)
+{
+	return NULL;
+}
+
+static inline int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def,
+					  const char *domain)
+{
+	return -ENOSYS;
+}
+
+static inline int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
+					     const char *domain)
+{
+	return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * Label Mapping Cache Functions
+ */
+
+#ifdef CONFIG_NETLABEL
+void cipso_v4_cache_invalidate(void);
+int cipso_v4_cache_add(const struct sk_buff *skb,
+		       const struct netlbl_lsm_secattr *secattr);
+#else
+static inline void cipso_v4_cache_invalidate(void)
+{
+	return;
+}
+
+static inline int cipso_v4_cache_add(const struct sk_buff *skb,
+				     const struct netlbl_lsm_secattr *secattr)
+{
+	return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * Protocol Handling Functions
+ */
+
+#ifdef CONFIG_NETLABEL
+void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway);
+int cipso_v4_socket_setopt(struct socket *sock,
+			   unsigned char *opt,
+			   u32 opt_len);
+int cipso_v4_socket_setattr(const struct socket *sock,
+			    const struct cipso_v4_doi *doi_def,
+			    const struct netlbl_lsm_secattr *secattr);
+int cipso_v4_socket_getopt(const struct socket *sock,
+			   unsigned char **opt,
+			   u32 *opt_len);
+int cipso_v4_socket_getattr(const struct socket *sock,
+			    struct netlbl_lsm_secattr *secattr);
+int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
+			    struct netlbl_lsm_secattr *secattr);
+int cipso_v4_validate(unsigned char **option);
+#else
+static inline void cipso_v4_error(struct sk_buff *skb,
+				  int error,
+				  u32 gateway)
+{
+	return;
+}
+
+static inline int cipso_v4_socket_setattr(const struct socket *sock,
+				  const struct cipso_v4_doi *doi_def,
+				  const struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline int cipso_v4_socket_getattr(const struct socket *sock,
+					  struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
+					  struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline int cipso_v4_validate(unsigned char **option)
+{
+	return -ENOSYS;
+}
+#endif /* CONFIG_NETLABEL */
+
+#endif /* _CIPSO_IPV4_H */
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 1f4a9a60d4cc..f4caad56cd03 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -51,7 +51,7 @@ struct ip_options {
 			ts_needtime:1,
 			ts_needaddr:1;
 	unsigned char	router_alert;
-	unsigned char	__pad1;
+	unsigned char	cipso;
 	unsigned char	__pad2;
 	unsigned char	__data[0];
 };
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
new file mode 100644
index 000000000000..7cae730832c7
--- /dev/null
+++ b/include/net/netlabel.h
@@ -0,0 +1,291 @@
+/*
+ * NetLabel System
+ *
+ * The NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program;  if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_H
+#define _NETLABEL_H
+
+#include <linux/types.h>
+#include <linux/skbuff.h>
+#include <net/netlink.h>
+
+/*
+ * NetLabel - A management interface for maintaining network packet label
+ *            mapping tables for explicit packet labling protocols.
+ *
+ * Network protocols such as CIPSO and RIPSO require a label translation layer
+ * to convert the label on the packet into something meaningful on the host
+ * machine.  In the current Linux implementation these mapping tables live
+ * inside the kernel; NetLabel provides a mechanism for user space applications
+ * to manage these mapping tables.
+ *
+ * NetLabel makes use of the Generic NETLINK mechanism as a transport layer to
+ * send messages between kernel and user space.  The general format of a
+ * NetLabel message is shown below:
+ *
+ *  +-----------------+-------------------+--------- --- -- -
+ *  | struct nlmsghdr | struct genlmsghdr | payload
+ *  +-----------------+-------------------+--------- --- -- -
+ *
+ * The 'nlmsghdr' and 'genlmsghdr' structs should be dealt with like normal.
+ * The payload is dependent on the subsystem specified in the
+ * 'nlmsghdr->nlmsg_type' and should be defined below, supporting functions
+ * should be defined in the corresponding net/netlabel/netlabel_<subsys>.h|c
+ * file.  All of the fields in the NetLabel payload are NETLINK attributes, the
+ * length of each field is the length of the NETLINK attribute payload, see
+ * include/net/netlink.h for more information on NETLINK attributes.
+ *
+ */
+
+/*
+ * NetLabel NETLINK protocol
+ */
+
+#define NETLBL_PROTO_VERSION            1
+
+/* NetLabel NETLINK types/families */
+#define NETLBL_NLTYPE_NONE              0
+#define NETLBL_NLTYPE_MGMT              1
+#define NETLBL_NLTYPE_MGMT_NAME         "NLBL_MGMT"
+#define NETLBL_NLTYPE_RIPSO             2
+#define NETLBL_NLTYPE_RIPSO_NAME        "NLBL_RIPSO"
+#define NETLBL_NLTYPE_CIPSOV4           3
+#define NETLBL_NLTYPE_CIPSOV4_NAME      "NLBL_CIPSOv4"
+#define NETLBL_NLTYPE_CIPSOV6           4
+#define NETLBL_NLTYPE_CIPSOV6_NAME      "NLBL_CIPSOv6"
+#define NETLBL_NLTYPE_UNLABELED         5
+#define NETLBL_NLTYPE_UNLABELED_NAME    "NLBL_UNLBL"
+
+/* NetLabel return codes */
+#define NETLBL_E_OK                     0
+
+/*
+ * Helper functions
+ */
+
+#define NETLBL_LEN_U8                   nla_total_size(sizeof(u8))
+#define NETLBL_LEN_U16                  nla_total_size(sizeof(u16))
+#define NETLBL_LEN_U32                  nla_total_size(sizeof(u32))
+
+/**
+ * netlbl_netlink_alloc_skb - Allocate a NETLINK message buffer
+ * @head: the amount of headroom in bytes
+ * @body: the desired size (minus headroom) in bytes
+ * @gfp_flags: the alloc flags to pass to alloc_skb()
+ *
+ * Description:
+ * Allocate a NETLINK message buffer based on the sizes given in @head and
+ * @body.  If @head is greater than zero skb_reserve() is called to reserve
+ * @head bytes at the start of the buffer.  Returns a valid sk_buff pointer on
+ * success, NULL on failure.
+ *
+ */
+static inline struct sk_buff *netlbl_netlink_alloc_skb(size_t head,
+						       size_t body,
+						       int gfp_flags)
+{
+	struct sk_buff *skb;
+
+	skb = alloc_skb(NLMSG_ALIGN(head + body), gfp_flags);
+	if (skb == NULL)
+		return NULL;
+	if (head > 0) {
+		skb_reserve(skb, head);
+		if (skb_tailroom(skb) < body) {
+			kfree_skb(skb);
+			return NULL;
+		}
+	}
+
+	return skb;
+}
+
+/*
+ * NetLabel - Kernel API for accessing the network packet label mappings.
+ *
+ * The following functions are provided for use by other kernel modules,
+ * specifically kernel LSM modules, to provide a consistent, transparent API
+ * for dealing with explicit packet labeling protocols such as CIPSO and
+ * RIPSO.  The functions defined here are implemented in the
+ * net/netlabel/netlabel_kapi.c file.
+ *
+ */
+
+/* Domain mapping definition struct */
+struct netlbl_dom_map;
+
+/* Domain mapping operations */
+int netlbl_domhsh_remove(const char *domain);
+
+/* LSM security attributes */
+struct netlbl_lsm_cache {
+	void (*free) (const void *data);
+	void *data;
+};
+struct netlbl_lsm_secattr {
+	char *domain;
+
+	u32 mls_lvl;
+	u32 mls_lvl_vld;
+	unsigned char *mls_cat;
+	size_t mls_cat_len;
+
+	struct netlbl_lsm_cache cache;
+};
+
+/*
+ * LSM security attribute operations
+ */
+
+
+/**
+ * netlbl_secattr_init - Initialize a netlbl_lsm_secattr struct
+ * @secattr: the struct to initialize
+ *
+ * Description:
+ * Initialize an already allocated netlbl_lsm_secattr struct.  Returns zero on
+ * success, negative values on error.
+ *
+ */
+static inline int netlbl_secattr_init(struct netlbl_lsm_secattr *secattr)
+{
+	memset(secattr, 0, sizeof(*secattr));
+	return 0;
+}
+
+/**
+ * netlbl_secattr_destroy - Clears a netlbl_lsm_secattr struct
+ * @secattr: the struct to clear
+ * @clear_cache: cache clear flag
+ *
+ * Description:
+ * Destroys the @secattr struct, including freeing all of the internal buffers.
+ * If @clear_cache is true then free the cache fields, otherwise leave them
+ * intact.  The struct must be reset with a call to netlbl_secattr_init()
+ * before reuse.
+ *
+ */
+static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr,
+					  u32 clear_cache)
+{
+	if (clear_cache && secattr->cache.data != NULL && secattr->cache.free)
+		secattr->cache.free(secattr->cache.data);
+	kfree(secattr->domain);
+	kfree(secattr->mls_cat);
+}
+
+/**
+ * netlbl_secattr_alloc - Allocate and initialize a netlbl_lsm_secattr struct
+ * @flags: the memory allocation flags
+ *
+ * Description:
+ * Allocate and initialize a netlbl_lsm_secattr struct.  Returns a valid
+ * pointer on success, or NULL on failure.
+ *
+ */
+static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(int flags)
+{
+	return kzalloc(sizeof(struct netlbl_lsm_secattr), flags);
+}
+
+/**
+ * netlbl_secattr_free - Frees a netlbl_lsm_secattr struct
+ * @secattr: the struct to free
+ * @clear_cache: cache clear flag
+ *
+ * Description:
+ * Frees @secattr including all of the internal buffers.  If @clear_cache is
+ * true then free the cache fields, otherwise leave them intact.
+ *
+ */
+static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr,
+				       u32 clear_cache)
+{
+	netlbl_secattr_destroy(secattr, clear_cache);
+	kfree(secattr);
+}
+
+/*
+ * LSM protocol operations
+ */
+
+#ifdef CONFIG_NETLABEL
+int netlbl_socket_setattr(const struct socket *sock,
+			  const struct netlbl_lsm_secattr *secattr);
+int netlbl_socket_getattr(const struct socket *sock,
+			  struct netlbl_lsm_secattr *secattr);
+int netlbl_skbuff_getattr(const struct sk_buff *skb,
+			  struct netlbl_lsm_secattr *secattr);
+void netlbl_skbuff_err(struct sk_buff *skb, int error);
+#else
+static inline int netlbl_socket_setattr(const struct socket *sock,
+				     const struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline int netlbl_socket_getattr(const struct socket *sock,
+					struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline int netlbl_skbuff_getattr(const struct sk_buff *skb,
+					struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline void netlbl_skbuff_err(struct sk_buff *skb, int error)
+{
+	return;
+}
+#endif /* CONFIG_NETLABEL */
+
+/*
+ * LSM label mapping cache operations
+ */
+
+#ifdef CONFIG_NETLABEL
+void netlbl_cache_invalidate(void);
+int netlbl_cache_add(const struct sk_buff *skb,
+		     const struct netlbl_lsm_secattr *secattr);
+#else
+static inline void netlbl_cache_invalidate(void)
+{
+	return;
+}
+
+static inline int netlbl_cache_add(const struct sk_buff *skb,
+				   const struct netlbl_lsm_secattr *secattr)
+{
+	return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+#endif /* _NETLABEL_H */
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index 2b98943e6b02..008e69d2e423 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -35,7 +35,7 @@ static int ip_clear_mutable_options(struct iphdr *iph, u32 *daddr)
 		switch (*optptr) {
 		case IPOPT_SEC:
 		case 0x85:	/* Some "Extended Security" crap. */
-		case 0x86:	/* Another "Commercial Security" crap. */
+		case IPOPT_CIPSO:
 		case IPOPT_RA:
 		case 0x80|21:	/* RFC1770 */
 			break;
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 406056edc02b..e0a93b4fa8cc 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -24,6 +24,7 @@
 #include <net/ip.h>
 #include <net/icmp.h>
 #include <net/route.h>
+#include <net/cipso_ipv4.h>
 
 /* 
  * Write options to IP header, record destination address to
@@ -194,6 +195,13 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 			dopt->is_strictroute = sopt->is_strictroute;
 		}
 	}
+	if (sopt->cipso) {
+		optlen  = sptr[sopt->cipso+1];
+		dopt->cipso = dopt->optlen+sizeof(struct iphdr);
+		memcpy(dptr, sptr+sopt->cipso, optlen);
+		dptr += optlen;
+		dopt->optlen += optlen;
+	}
 	while (dopt->optlen & 3) {
 		*dptr++ = IPOPT_END;
 		dopt->optlen++;
@@ -434,6 +442,17 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
 			if (optptr[2] == 0 && optptr[3] == 0)
 				opt->router_alert = optptr - iph;
 			break;
+		      case IPOPT_CIPSO:
+		        if (opt->cipso) {
+				pp_ptr = optptr;
+				goto error;
+			}
+			opt->cipso = optptr - iph;
+		        if (cipso_v4_validate(&optptr)) {
+				pp_ptr = optptr;
+				goto error;
+			}
+			break;
 		      case IPOPT_SEC:
 		      case IPOPT_SID:
 		      default:
-- 
cgit v1.2.3


From 446fda4f26822b2d42ab3396aafcedf38a9ff2b6 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Thu, 3 Aug 2006 16:48:06 -0700
Subject: [NetLabel]: CIPSOv4 engine

Add support for the Commercial IP Security Option (CIPSO) to the IPv4
network stack.  CIPSO has become a de-facto standard for
trusted/labeled networking amongst existing Trusted Operating Systems
such as Trusted Solaris, HP-UX CMW, etc.  This implementation is
designed to be used with the NetLabel subsystem to provide explicit
packet labeling to LSM developers.

The CIPSO/IPv4 packet labeling works by the LSM calling a NetLabel API
function which attaches a CIPSO label (IPv4 option) to a given socket;
this in turn attaches the CIPSO label to every packet leaving the
socket without any extra processing on the outbound side.  On the
inbound side the individual packet's sk_buff is examined through a
call to a NetLabel API function to determine if a CIPSO/IPv4 label is
present and if so the security attributes of the CIPSO label are
returned to the caller of the NetLabel API function.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sysctl.h     |    4 +
 net/ipv4/Makefile          |    1 +
 net/ipv4/cipso_ipv4.c      | 1607 ++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/sysctl_net_ipv4.c |   35 +
 4 files changed, 1647 insertions(+)
 create mode 100644 net/ipv4/cipso_ipv4.c

(limited to 'include')

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index e4b1a4d4dcf3..af61d9235409 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -411,6 +411,10 @@ enum
 	NET_IPV4_TCP_WORKAROUND_SIGNED_WINDOWS=115,
 	NET_TCP_DMA_COPYBREAK=116,
 	NET_TCP_SLOW_START_AFTER_IDLE=117,
+	NET_CIPSOV4_CACHE_ENABLE=118,
+	NET_CIPSOV4_CACHE_BUCKET_SIZE=119,
+	NET_CIPSOV4_RBM_OPTFMT=120,
+	NET_CIPSOV4_RBM_STRICTVALID=121,
 };
 
 enum {
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 4878fc5be85f..f66049e28aeb 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -47,6 +47,7 @@ obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o
 obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o
 obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
 obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
+obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
 		      xfrm4_output.o
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
new file mode 100644
index 000000000000..b82a101c95c5
--- /dev/null
+++ b/net/ipv4/cipso_ipv4.c
@@ -0,0 +1,1607 @@
+/*
+ * CIPSO - Commercial IP Security Option
+ *
+ * This is an implementation of the CIPSO 2.2 protocol as specified in
+ * draft-ietf-cipso-ipsecurity-01.txt with additional tag types as found in
+ * FIPS-188, copies of both documents can be found in the Documentation
+ * directory.  While CIPSO never became a full IETF RFC standard many vendors
+ * have chosen to adopt the protocol and over the years it has become a
+ * de-facto standard for labeled networking.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program;  if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/jhash.h>
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/tcp.h>
+#include <net/netlabel.h>
+#include <net/cipso_ipv4.h>
+#include <asm/bug.h>
+
+struct cipso_v4_domhsh_entry {
+	char *domain;
+	u32 valid;
+	struct list_head list;
+	struct rcu_head rcu;
+};
+
+/* List of available DOI definitions */
+/* XXX - Updates should be minimal so having a single lock for the
+ * cipso_v4_doi_list and the cipso_v4_doi_list->dom_list should be
+ * okay. */
+/* XXX - This currently assumes a minimal number of different DOIs in use,
+ * if in practice there are a lot of different DOIs this list should
+ * probably be turned into a hash table or something similar so we
+ * can do quick lookups. */
+DEFINE_SPINLOCK(cipso_v4_doi_list_lock);
+static struct list_head cipso_v4_doi_list = LIST_HEAD_INIT(cipso_v4_doi_list);
+
+/* Label mapping cache */
+int cipso_v4_cache_enabled = 1;
+int cipso_v4_cache_bucketsize = 10;
+#define CIPSO_V4_CACHE_BUCKETBITS     7
+#define CIPSO_V4_CACHE_BUCKETS        (1 << CIPSO_V4_CACHE_BUCKETBITS)
+#define CIPSO_V4_CACHE_REORDERLIMIT   10
+struct cipso_v4_map_cache_bkt {
+	spinlock_t lock;
+	u32 size;
+	struct list_head list;
+};
+struct cipso_v4_map_cache_entry {
+	u32 hash;
+	unsigned char *key;
+	size_t key_len;
+
+	struct netlbl_lsm_cache lsm_data;
+
+	u32 activity;
+	struct list_head list;
+};
+static struct cipso_v4_map_cache_bkt *cipso_v4_cache = NULL;
+
+/* Restricted bitmap (tag #1) flags */
+int cipso_v4_rbm_optfmt = 0;
+int cipso_v4_rbm_strictvalid = 1;
+
+/*
+ * Helper Functions
+ */
+
+/**
+ * cipso_v4_bitmap_walk - Walk a bitmap looking for a bit
+ * @bitmap: the bitmap
+ * @bitmap_len: length in bits
+ * @offset: starting offset
+ * @state: if non-zero, look for a set (1) bit else look for a cleared (0) bit
+ *
+ * Description:
+ * Starting at @offset, walk the bitmap from left to right until either the
+ * desired bit is found or we reach the end.  Return the bit offset, -1 if
+ * not found, or -2 if error.
+ */
+static int cipso_v4_bitmap_walk(const unsigned char *bitmap,
+				u32 bitmap_len,
+				u32 offset,
+				u8 state)
+{
+	u32 bit_spot;
+	u32 byte_offset;
+	unsigned char bitmask;
+	unsigned char byte;
+
+	/* gcc always rounds to zero when doing integer division */
+	byte_offset = offset / 8;
+	byte = bitmap[byte_offset];
+	bit_spot = offset;
+	bitmask = 0x80 >> (offset % 8);
+
+	while (bit_spot < bitmap_len) {
+		if ((state && (byte & bitmask) == bitmask) ||
+		    (state == 0 && (byte & bitmask) == 0))
+			return bit_spot;
+
+		bit_spot++;
+		bitmask >>= 1;
+		if (bitmask == 0) {
+			byte = bitmap[++byte_offset];
+			bitmask = 0x80;
+		}
+	}
+
+	return -1;
+}
+
+/**
+ * cipso_v4_bitmap_setbit - Sets a single bit in a bitmap
+ * @bitmap: the bitmap
+ * @bit: the bit
+ * @state: if non-zero, set the bit (1) else clear the bit (0)
+ *
+ * Description:
+ * Set a single bit in the bitmask.  Returns zero on success, negative values
+ * on error.
+ */
+static void cipso_v4_bitmap_setbit(unsigned char *bitmap,
+				   u32 bit,
+				   u8 state)
+{
+	u32 byte_spot;
+	u8 bitmask;
+
+	/* gcc always rounds to zero when doing integer division */
+	byte_spot = bit / 8;
+	bitmask = 0x80 >> (bit % 8);
+	if (state)
+		bitmap[byte_spot] |= bitmask;
+	else
+		bitmap[byte_spot] &= ~bitmask;
+}
+
+/**
+ * cipso_v4_doi_domhsh_free - Frees a domain list entry
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that the memory allocated to a domain list entry can be released
+ * safely.
+ *
+ */
+static void cipso_v4_doi_domhsh_free(struct rcu_head *entry)
+{
+	struct cipso_v4_domhsh_entry *ptr;
+
+	ptr = container_of(entry, struct cipso_v4_domhsh_entry, rcu);
+	kfree(ptr->domain);
+	kfree(ptr);
+}
+
+/**
+ * cipso_v4_cache_entry_free - Frees a cache entry
+ * @entry: the entry to free
+ *
+ * Description:
+ * This function frees the memory associated with a cache entry.
+ *
+ */
+static void cipso_v4_cache_entry_free(struct cipso_v4_map_cache_entry *entry)
+{
+	if (entry->lsm_data.free)
+		entry->lsm_data.free(entry->lsm_data.data);
+	kfree(entry->key);
+	kfree(entry);
+}
+
+/**
+ * cipso_v4_map_cache_hash - Hashing function for the CIPSO cache
+ * @key: the hash key
+ * @key_len: the length of the key in bytes
+ *
+ * Description:
+ * The CIPSO tag hashing function.  Returns a 32-bit hash value.
+ *
+ */
+static u32 cipso_v4_map_cache_hash(const unsigned char *key, u32 key_len)
+{
+	return jhash(key, key_len, 0);
+}
+
+/*
+ * Label Mapping Cache Functions
+ */
+
+/**
+ * cipso_v4_cache_init - Initialize the CIPSO cache
+ *
+ * Description:
+ * Initializes the CIPSO label mapping cache, this function should be called
+ * before any of the other functions defined in this file.  Returns zero on
+ * success, negative values on error.
+ *
+ */
+static int cipso_v4_cache_init(void)
+{
+	u32 iter;
+
+	cipso_v4_cache = kcalloc(CIPSO_V4_CACHE_BUCKETS,
+				 sizeof(struct cipso_v4_map_cache_bkt),
+				 GFP_KERNEL);
+	if (cipso_v4_cache == NULL)
+		return -ENOMEM;
+
+	for (iter = 0; iter < CIPSO_V4_CACHE_BUCKETS; iter++) {
+		spin_lock_init(&cipso_v4_cache[iter].lock);
+		cipso_v4_cache[iter].size = 0;
+		INIT_LIST_HEAD(&cipso_v4_cache[iter].list);
+	}
+
+	return 0;
+}
+
+/**
+ * cipso_v4_cache_invalidate - Invalidates the current CIPSO cache
+ *
+ * Description:
+ * Invalidates and frees any entries in the CIPSO cache.  Returns zero on
+ * success and negative values on failure.
+ *
+ */
+void cipso_v4_cache_invalidate(void)
+{
+	struct cipso_v4_map_cache_entry *entry, *tmp_entry;
+	u32 iter;
+
+	for (iter = 0; iter < CIPSO_V4_CACHE_BUCKETS; iter++) {
+		spin_lock(&cipso_v4_cache[iter].lock);
+		list_for_each_entry_safe(entry,
+					 tmp_entry,
+					 &cipso_v4_cache[iter].list, list) {
+			list_del(&entry->list);
+			cipso_v4_cache_entry_free(entry);
+		}
+		cipso_v4_cache[iter].size = 0;
+		spin_unlock(&cipso_v4_cache[iter].lock);
+	}
+
+	return;
+}
+
+/**
+ * cipso_v4_cache_check - Check the CIPSO cache for a label mapping
+ * @key: the buffer to check
+ * @key_len: buffer length in bytes
+ * @secattr: the security attribute struct to use
+ *
+ * Description:
+ * This function checks the cache to see if a label mapping already exists for
+ * the given key.  If there is a match then the cache is adjusted and the
+ * @secattr struct is populated with the correct LSM security attributes.  The
+ * cache is adjusted in the following manner if the entry is not already the
+ * first in the cache bucket:
+ *
+ *  1. The cache entry's activity counter is incremented
+ *  2. The previous (higher ranking) entry's activity counter is decremented
+ *  3. If the difference between the two activity counters is geater than
+ *     CIPSO_V4_CACHE_REORDERLIMIT the two entries are swapped
+ *
+ * Returns zero on success, -ENOENT for a cache miss, and other negative values
+ * on error.
+ *
+ */
+static int cipso_v4_cache_check(const unsigned char *key,
+				u32 key_len,
+				struct netlbl_lsm_secattr *secattr)
+{
+	u32 bkt;
+	struct cipso_v4_map_cache_entry *entry;
+	struct cipso_v4_map_cache_entry *prev_entry = NULL;
+	u32 hash;
+
+	if (!cipso_v4_cache_enabled)
+		return -ENOENT;
+
+	hash = cipso_v4_map_cache_hash(key, key_len);
+	bkt = hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
+	spin_lock(&cipso_v4_cache[bkt].lock);
+	list_for_each_entry(entry, &cipso_v4_cache[bkt].list, list) {
+		if (entry->hash == hash &&
+		    entry->key_len == key_len &&
+		    memcmp(entry->key, key, key_len) == 0) {
+			entry->activity += 1;
+			secattr->cache.free = entry->lsm_data.free;
+			secattr->cache.data = entry->lsm_data.data;
+			if (prev_entry == NULL) {
+				spin_unlock(&cipso_v4_cache[bkt].lock);
+				return 0;
+			}
+
+			if (prev_entry->activity > 0)
+				prev_entry->activity -= 1;
+			if (entry->activity > prev_entry->activity &&
+			    entry->activity - prev_entry->activity >
+			    CIPSO_V4_CACHE_REORDERLIMIT) {
+				__list_del(entry->list.prev, entry->list.next);
+				__list_add(&entry->list,
+					   prev_entry->list.prev,
+					   &prev_entry->list);
+			}
+
+			spin_unlock(&cipso_v4_cache[bkt].lock);
+			return 0;
+		}
+		prev_entry = entry;
+	}
+	spin_unlock(&cipso_v4_cache[bkt].lock);
+
+	return -ENOENT;
+}
+
+/**
+ * cipso_v4_cache_add - Add an entry to the CIPSO cache
+ * @skb: the packet
+ * @secattr: the packet's security attributes
+ *
+ * Description:
+ * Add a new entry into the CIPSO label mapping cache.  Add the new entry to
+ * head of the cache bucket's list, if the cache bucket is out of room remove
+ * the last entry in the list first.  It is important to note that there is
+ * currently no checking for duplicate keys.  Returns zero on success,
+ * negative values on failure.
+ *
+ */
+int cipso_v4_cache_add(const struct sk_buff *skb,
+		       const struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val = -EPERM;
+	u32 bkt;
+	struct cipso_v4_map_cache_entry *entry = NULL;
+	struct cipso_v4_map_cache_entry *old_entry = NULL;
+	unsigned char *cipso_ptr;
+	u32 cipso_ptr_len;
+
+	if (!cipso_v4_cache_enabled || cipso_v4_cache_bucketsize <= 0)
+		return 0;
+
+	cipso_ptr = CIPSO_V4_OPTPTR(skb);
+	cipso_ptr_len = cipso_ptr[1];
+
+	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
+	if (entry == NULL)
+		return -ENOMEM;
+	entry->key = kmalloc(cipso_ptr_len, GFP_ATOMIC);
+	if (entry->key == NULL) {
+		ret_val = -ENOMEM;
+		goto cache_add_failure;
+	}
+	memcpy(entry->key, cipso_ptr, cipso_ptr_len);
+	entry->key_len = cipso_ptr_len;
+	entry->hash = cipso_v4_map_cache_hash(cipso_ptr, cipso_ptr_len);
+	entry->lsm_data.free = secattr->cache.free;
+	entry->lsm_data.data = secattr->cache.data;
+
+	bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
+	spin_lock(&cipso_v4_cache[bkt].lock);
+	if (cipso_v4_cache[bkt].size < cipso_v4_cache_bucketsize) {
+		list_add(&entry->list, &cipso_v4_cache[bkt].list);
+		cipso_v4_cache[bkt].size += 1;
+	} else {
+		old_entry = list_entry(cipso_v4_cache[bkt].list.prev,
+				       struct cipso_v4_map_cache_entry, list);
+		list_del(&old_entry->list);
+		list_add(&entry->list, &cipso_v4_cache[bkt].list);
+		cipso_v4_cache_entry_free(old_entry);
+	}
+	spin_unlock(&cipso_v4_cache[bkt].lock);
+
+	return 0;
+
+cache_add_failure:
+	if (entry)
+		cipso_v4_cache_entry_free(entry);
+	return ret_val;
+}
+
+/*
+ * DOI List Functions
+ */
+
+/**
+ * cipso_v4_doi_search - Searches for a DOI definition
+ * @doi: the DOI to search for
+ *
+ * Description:
+ * Search the DOI definition list for a DOI definition with a DOI value that
+ * matches @doi.  The caller is responsibile for calling rcu_read_[un]lock().
+ * Returns a pointer to the DOI definition on success and NULL on failure.
+ */
+static struct cipso_v4_doi *cipso_v4_doi_search(u32 doi)
+{
+	struct cipso_v4_doi *iter;
+
+	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
+		if (iter->doi == doi && iter->valid)
+			return iter;
+	return NULL;
+}
+
+/**
+ * cipso_v4_doi_add - Add a new DOI to the CIPSO protocol engine
+ * @doi_def: the DOI structure
+ *
+ * Description:
+ * The caller defines a new DOI for use by the CIPSO engine and calls this
+ * function to add it to the list of acceptable domains.  The caller must
+ * ensure that the mapping table specified in @doi_def->map meets all of the
+ * requirements of the mapping type (see cipso_ipv4.h for details).  Returns
+ * zero on success and non-zero on failure.
+ *
+ */
+int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
+{
+	if (doi_def == NULL || doi_def->doi == CIPSO_V4_DOI_UNKNOWN)
+		return -EINVAL;
+
+	doi_def->valid = 1;
+	INIT_RCU_HEAD(&doi_def->rcu);
+	INIT_LIST_HEAD(&doi_def->dom_list);
+
+	rcu_read_lock();
+	if (cipso_v4_doi_search(doi_def->doi) != NULL)
+		goto doi_add_failure_rlock;
+	spin_lock(&cipso_v4_doi_list_lock);
+	if (cipso_v4_doi_search(doi_def->doi) != NULL)
+		goto doi_add_failure_slock;
+	list_add_tail_rcu(&doi_def->list, &cipso_v4_doi_list);
+	spin_unlock(&cipso_v4_doi_list_lock);
+	rcu_read_unlock();
+
+	return 0;
+
+doi_add_failure_slock:
+	spin_unlock(&cipso_v4_doi_list_lock);
+doi_add_failure_rlock:
+	rcu_read_unlock();
+	return -EEXIST;
+}
+
+/**
+ * cipso_v4_doi_remove - Remove an existing DOI from the CIPSO protocol engine
+ * @doi: the DOI value
+ * @callback: the DOI cleanup/free callback
+ *
+ * Description:
+ * Removes a DOI definition from the CIPSO engine, @callback is called to
+ * free any memory.  The NetLabel routines will be called to release their own
+ * LSM domain mappings as well as our own domain list.  Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head))
+{
+	struct cipso_v4_doi *doi_def;
+	struct cipso_v4_domhsh_entry *dom_iter;
+
+	rcu_read_lock();
+	if (cipso_v4_doi_search(doi) != NULL) {
+		spin_lock(&cipso_v4_doi_list_lock);
+		doi_def = cipso_v4_doi_search(doi);
+		if (doi_def == NULL) {
+			spin_unlock(&cipso_v4_doi_list_lock);
+			rcu_read_unlock();
+			return -ENOENT;
+		}
+		doi_def->valid = 0;
+		list_del_rcu(&doi_def->list);
+		spin_unlock(&cipso_v4_doi_list_lock);
+		list_for_each_entry_rcu(dom_iter, &doi_def->dom_list, list)
+			if (dom_iter->valid)
+				netlbl_domhsh_remove(dom_iter->domain);
+		cipso_v4_cache_invalidate();
+		rcu_read_unlock();
+
+		call_rcu(&doi_def->rcu, callback);
+		return 0;
+	}
+	rcu_read_unlock();
+
+	return -ENOENT;
+}
+
+/**
+ * cipso_v4_doi_getdef - Returns a pointer to a valid DOI definition
+ * @doi: the DOI value
+ *
+ * Description:
+ * Searches for a valid DOI definition and if one is found it is returned to
+ * the caller.  Otherwise NULL is returned.  The caller must ensure that
+ * rcu_read_lock() is held while accessing the returned definition.
+ *
+ */
+struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
+{
+	return cipso_v4_doi_search(doi);
+}
+
+/**
+ * cipso_v4_doi_dump_all - Dump all the CIPSO DOI definitions into a sk_buff
+ * @headroom: the amount of headroom to allocate for the sk_buff
+ *
+ * Description:
+ * Dump a list of all the configured DOI values into a sk_buff.  The returned
+ * sk_buff has room at the front of the sk_buff for @headroom bytes.  See
+ * net/netlabel/netlabel_cipso_v4.h for the LISTALL message format.  This
+ * function may fail if another process is changing the DOI list at the same
+ * time.  Returns a pointer to a sk_buff on success, NULL on error.
+ *
+ */
+struct sk_buff *cipso_v4_doi_dump_all(size_t headroom)
+{
+	struct sk_buff *skb = NULL;
+	struct cipso_v4_doi *iter;
+	u32 doi_cnt = 0;
+	ssize_t buf_len;
+
+	buf_len = NETLBL_LEN_U32;
+	rcu_read_lock();
+	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
+		if (iter->valid) {
+			doi_cnt += 1;
+			buf_len += 2 * NETLBL_LEN_U32;
+		}
+
+	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
+	if (skb == NULL)
+		goto doi_dump_all_failure;
+
+	if (nla_put_u32(skb, NLA_U32, doi_cnt) != 0)
+		goto doi_dump_all_failure;
+	buf_len -= NETLBL_LEN_U32;
+	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
+		if (iter->valid) {
+			if (buf_len < 2 * NETLBL_LEN_U32)
+				goto doi_dump_all_failure;
+			if (nla_put_u32(skb, NLA_U32, iter->doi) != 0)
+				goto doi_dump_all_failure;
+			if (nla_put_u32(skb, NLA_U32, iter->type) != 0)
+				goto doi_dump_all_failure;
+			buf_len -= 2 * NETLBL_LEN_U32;
+		}
+	rcu_read_unlock();
+
+	return skb;
+
+doi_dump_all_failure:
+	rcu_read_unlock();
+	kfree(skb);
+	return NULL;
+}
+
+/**
+ * cipso_v4_doi_dump - Dump a CIPSO DOI definition into a sk_buff
+ * @doi: the DOI value
+ * @headroom: the amount of headroom to allocate for the sk_buff
+ *
+ * Description:
+ * Lookup the DOI definition matching @doi and dump it's contents into a
+ * sk_buff.  The returned sk_buff has room at the front of the sk_buff for
+ * @headroom bytes.  See net/netlabel/netlabel_cipso_v4.h for the LIST message
+ * format.  This function may fail if another process is changing the DOI list
+ * at the same time.  Returns a pointer to a sk_buff on success, NULL on error.
+ *
+ */
+struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom)
+{
+	struct sk_buff *skb = NULL;
+	struct cipso_v4_doi *iter;
+	u32 tag_cnt = 0;
+	u32 lvl_cnt = 0;
+	u32 cat_cnt = 0;
+	ssize_t buf_len;
+	ssize_t tmp;
+
+	rcu_read_lock();
+	iter = cipso_v4_doi_getdef(doi);
+	if (iter == NULL)
+		goto doi_dump_failure;
+	buf_len = NETLBL_LEN_U32;
+	switch (iter->type) {
+	case CIPSO_V4_MAP_PASS:
+		buf_len += NETLBL_LEN_U32;
+		while(tag_cnt < CIPSO_V4_TAG_MAXCNT &&
+		      iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
+			tag_cnt += 1;
+			buf_len += NETLBL_LEN_U8;
+		}
+		break;
+	case CIPSO_V4_MAP_STD:
+		buf_len += 3 * NETLBL_LEN_U32;
+		while (tag_cnt < CIPSO_V4_TAG_MAXCNT &&
+		       iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
+			tag_cnt += 1;
+			buf_len += NETLBL_LEN_U8;
+		}
+		for (tmp = 0; tmp < iter->map.std->lvl.local_size; tmp++)
+			if (iter->map.std->lvl.local[tmp] !=
+			    CIPSO_V4_INV_LVL) {
+				lvl_cnt += 1;
+				buf_len += NETLBL_LEN_U32 + NETLBL_LEN_U8;
+			}
+		for (tmp = 0; tmp < iter->map.std->cat.local_size; tmp++)
+			if (iter->map.std->cat.local[tmp] !=
+			    CIPSO_V4_INV_CAT) {
+				cat_cnt += 1;
+				buf_len += NETLBL_LEN_U32 + NETLBL_LEN_U16;
+			}
+		break;
+	}
+
+	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
+	if (skb == NULL)
+		goto doi_dump_failure;
+
+	if (nla_put_u32(skb, NLA_U32, iter->type) != 0)
+		goto doi_dump_failure;
+	buf_len -= NETLBL_LEN_U32;
+	if (iter != cipso_v4_doi_getdef(doi))
+		goto doi_dump_failure;
+	switch (iter->type) {
+	case CIPSO_V4_MAP_PASS:
+		if (nla_put_u32(skb, NLA_U32, tag_cnt) != 0)
+			goto doi_dump_failure;
+		buf_len -= NETLBL_LEN_U32;
+		for (tmp = 0;
+		     tmp < CIPSO_V4_TAG_MAXCNT &&
+			     iter->tags[tmp] != CIPSO_V4_TAG_INVALID;
+		     tmp++) {
+			if (buf_len < NETLBL_LEN_U8)
+				goto doi_dump_failure;
+			if (nla_put_u8(skb, NLA_U8, iter->tags[tmp]) != 0)
+				goto doi_dump_failure;
+			buf_len -= NETLBL_LEN_U8;
+		}
+		break;
+	case CIPSO_V4_MAP_STD:
+		if (nla_put_u32(skb, NLA_U32, tag_cnt) != 0)
+			goto doi_dump_failure;
+		if (nla_put_u32(skb, NLA_U32, lvl_cnt) != 0)
+			goto doi_dump_failure;
+		if (nla_put_u32(skb, NLA_U32, cat_cnt) != 0)
+			goto doi_dump_failure;
+		buf_len -= 3 * NETLBL_LEN_U32;
+		for (tmp = 0;
+		     tmp < CIPSO_V4_TAG_MAXCNT &&
+			     iter->tags[tmp] != CIPSO_V4_TAG_INVALID;
+		     tmp++) {
+			if (buf_len < NETLBL_LEN_U8)
+				goto doi_dump_failure;
+			if (nla_put_u8(skb, NLA_U8, iter->tags[tmp]) != 0)
+				goto doi_dump_failure;
+			buf_len -= NETLBL_LEN_U8;
+		}
+		for (tmp = 0; tmp < iter->map.std->lvl.local_size; tmp++)
+			if (iter->map.std->lvl.local[tmp] !=
+			    CIPSO_V4_INV_LVL) {
+				if (buf_len < NETLBL_LEN_U32 + NETLBL_LEN_U8)
+					goto doi_dump_failure;
+				if (nla_put_u32(skb, NLA_U32, tmp) != 0)
+					goto doi_dump_failure;
+				if (nla_put_u8(skb,
+					   NLA_U8,
+					   iter->map.std->lvl.local[tmp]) != 0)
+					goto doi_dump_failure;
+				buf_len -= NETLBL_LEN_U32 + NETLBL_LEN_U8;
+			}
+		for (tmp = 0; tmp < iter->map.std->cat.local_size; tmp++)
+			if (iter->map.std->cat.local[tmp] !=
+			    CIPSO_V4_INV_CAT) {
+				if (buf_len < NETLBL_LEN_U32 + NETLBL_LEN_U16)
+					goto doi_dump_failure;
+				if (nla_put_u32(skb, NLA_U32, tmp) != 0)
+					goto doi_dump_failure;
+				if (nla_put_u16(skb,
+					   NLA_U16,
+					   iter->map.std->cat.local[tmp]) != 0)
+					goto doi_dump_failure;
+				buf_len -= NETLBL_LEN_U32 + NETLBL_LEN_U16;
+			}
+		break;
+	}
+	rcu_read_unlock();
+
+	return skb;
+
+doi_dump_failure:
+	rcu_read_unlock();
+	kfree(skb);
+	return NULL;
+}
+
+/**
+ * cipso_v4_doi_domhsh_add - Adds a domain entry to a DOI definition
+ * @doi_def: the DOI definition
+ * @domain: the domain to add
+ *
+ * Description:
+ * Adds the @domain to the the DOI specified by @doi_def, this function
+ * should only be called by external functions (i.e. NetLabel).  This function
+ * does allocate memory.  Returns zero on success, negative values on failure.
+ *
+ */
+int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain)
+{
+	struct cipso_v4_domhsh_entry *iter;
+	struct cipso_v4_domhsh_entry *new_dom;
+
+	new_dom = kzalloc(sizeof(*new_dom), GFP_KERNEL);
+	if (new_dom == NULL)
+		return -ENOMEM;
+	if (domain) {
+		new_dom->domain = kstrdup(domain, GFP_KERNEL);
+		if (new_dom->domain == NULL) {
+			kfree(new_dom);
+			return -ENOMEM;
+		}
+	}
+	new_dom->valid = 1;
+	INIT_RCU_HEAD(&new_dom->rcu);
+
+	rcu_read_lock();
+	spin_lock(&cipso_v4_doi_list_lock);
+	list_for_each_entry_rcu(iter, &doi_def->dom_list, list)
+		if (iter->valid &&
+		    ((domain != NULL && iter->domain != NULL &&
+		      strcmp(iter->domain, domain) == 0) ||
+		     (domain == NULL && iter->domain == NULL))) {
+			spin_unlock(&cipso_v4_doi_list_lock);
+			rcu_read_unlock();
+			kfree(new_dom->domain);
+			kfree(new_dom);
+			return -EEXIST;
+		}
+	list_add_tail_rcu(&new_dom->list, &doi_def->dom_list);
+	spin_unlock(&cipso_v4_doi_list_lock);
+	rcu_read_unlock();
+
+	return 0;
+}
+
+/**
+ * cipso_v4_doi_domhsh_remove - Removes a domain entry from a DOI definition
+ * @doi_def: the DOI definition
+ * @domain: the domain to remove
+ *
+ * Description:
+ * Removes the @domain from the DOI specified by @doi_def, this function
+ * should only be called by external functions (i.e. NetLabel).   Returns zero
+ * on success and negative values on error.
+ *
+ */
+int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
+			       const char *domain)
+{
+	struct cipso_v4_domhsh_entry *iter;
+
+	rcu_read_lock();
+	spin_lock(&cipso_v4_doi_list_lock);
+	list_for_each_entry_rcu(iter, &doi_def->dom_list, list)
+		if (iter->valid &&
+		    ((domain != NULL && iter->domain != NULL &&
+		      strcmp(iter->domain, domain) == 0) ||
+		     (domain == NULL && iter->domain == NULL))) {
+			iter->valid = 0;
+			list_del_rcu(&iter->list);
+			spin_unlock(&cipso_v4_doi_list_lock);
+			rcu_read_unlock();
+			call_rcu(&iter->rcu, cipso_v4_doi_domhsh_free);
+
+			return 0;
+		}
+	spin_unlock(&cipso_v4_doi_list_lock);
+	rcu_read_unlock();
+
+	return -ENOENT;
+}
+
+/*
+ * Label Mapping Functions
+ */
+
+/**
+ * cipso_v4_map_lvl_valid - Checks to see if the given level is understood
+ * @doi_def: the DOI definition
+ * @level: the level to check
+ *
+ * Description:
+ * Checks the given level against the given DOI definition and returns a
+ * negative value if the level does not have a valid mapping and a zero value
+ * if the level is defined by the DOI.
+ *
+ */
+static int cipso_v4_map_lvl_valid(const struct cipso_v4_doi *doi_def, u8 level)
+{
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_PASS:
+		return 0;
+	case CIPSO_V4_MAP_STD:
+		if (doi_def->map.std->lvl.cipso[level] < CIPSO_V4_INV_LVL)
+			return 0;
+		break;
+	}
+
+	return -EFAULT;
+}
+
+/**
+ * cipso_v4_map_lvl_hton - Perform a level mapping from the host to the network
+ * @doi_def: the DOI definition
+ * @host_lvl: the host MLS level
+ * @net_lvl: the network/CIPSO MLS level
+ *
+ * Description:
+ * Perform a label mapping to translate a local MLS level to the correct
+ * CIPSO level using the given DOI definition.  Returns zero on success,
+ * negative values otherwise.
+ *
+ */
+static int cipso_v4_map_lvl_hton(const struct cipso_v4_doi *doi_def,
+				 u32 host_lvl,
+				 u32 *net_lvl)
+{
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_PASS:
+		*net_lvl = host_lvl;
+		return 0;
+	case CIPSO_V4_MAP_STD:
+		if (host_lvl < doi_def->map.std->lvl.local_size) {
+			*net_lvl = doi_def->map.std->lvl.local[host_lvl];
+			return 0;
+		}
+		break;
+	}
+
+	return -EINVAL;
+}
+
+/**
+ * cipso_v4_map_lvl_ntoh - Perform a level mapping from the network to the host
+ * @doi_def: the DOI definition
+ * @net_lvl: the network/CIPSO MLS level
+ * @host_lvl: the host MLS level
+ *
+ * Description:
+ * Perform a label mapping to translate a CIPSO level to the correct local MLS
+ * level using the given DOI definition.  Returns zero on success, negative
+ * values otherwise.
+ *
+ */
+static int cipso_v4_map_lvl_ntoh(const struct cipso_v4_doi *doi_def,
+				 u32 net_lvl,
+				 u32 *host_lvl)
+{
+	struct cipso_v4_std_map_tbl *map_tbl;
+
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_PASS:
+		*host_lvl = net_lvl;
+		return 0;
+	case CIPSO_V4_MAP_STD:
+		map_tbl = doi_def->map.std;
+		if (net_lvl < map_tbl->lvl.cipso_size &&
+		    map_tbl->lvl.cipso[net_lvl] < CIPSO_V4_INV_LVL) {
+			*host_lvl = doi_def->map.std->lvl.cipso[net_lvl];
+			return 0;
+		}
+		break;
+	}
+
+	return -EINVAL;
+}
+
+/**
+ * cipso_v4_map_cat_rbm_valid - Checks to see if the category bitmap is valid
+ * @doi_def: the DOI definition
+ * @bitmap: category bitmap
+ * @bitmap_len: bitmap length in bytes
+ *
+ * Description:
+ * Checks the given category bitmap against the given DOI definition and
+ * returns a negative value if any of the categories in the bitmap do not have
+ * a valid mapping and a zero value if all of the categories are valid.
+ *
+ */
+static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def,
+				      const unsigned char *bitmap,
+				      u32 bitmap_len)
+{
+	int cat = -1;
+	u32 bitmap_len_bits = bitmap_len * 8;
+	u32 cipso_cat_size = doi_def->map.std->cat.cipso_size;
+	u32 *cipso_array = doi_def->map.std->cat.cipso;
+
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_PASS:
+		return 0;
+	case CIPSO_V4_MAP_STD:
+		for (;;) {
+			cat = cipso_v4_bitmap_walk(bitmap,
+						   bitmap_len_bits,
+						   cat + 1,
+						   1);
+			if (cat < 0)
+				break;
+			if (cat >= cipso_cat_size ||
+			    cipso_array[cat] >= CIPSO_V4_INV_CAT)
+				return -EFAULT;
+		}
+
+		if (cat == -1)
+			return 0;
+		break;
+	}
+
+	return -EFAULT;
+}
+
+/**
+ * cipso_v4_map_cat_rbm_hton - Perform a category mapping from host to network
+ * @doi_def: the DOI definition
+ * @host_cat: the category bitmap in host format
+ * @host_cat_len: the length of the host's category bitmap in bytes
+ * @net_cat: the zero'd out category bitmap in network/CIPSO format
+ * @net_cat_len: the length of the CIPSO bitmap in bytes
+ *
+ * Description:
+ * Perform a label mapping to translate a local MLS category bitmap to the
+ * correct CIPSO bitmap using the given DOI definition.  Returns the minimum
+ * size in bytes of the network bitmap on success, negative values otherwise.
+ *
+ */
+static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def,
+				     const unsigned char *host_cat,
+				     u32 host_cat_len,
+				     unsigned char *net_cat,
+				     u32 net_cat_len)
+{
+	int host_spot = -1;
+	u32 net_spot;
+	u32 net_spot_max = 0;
+	u32 host_clen_bits = host_cat_len * 8;
+	u32 net_clen_bits = net_cat_len * 8;
+	u32 host_cat_size = doi_def->map.std->cat.local_size;
+	u32 *host_cat_array = doi_def->map.std->cat.local;
+
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_PASS:
+		net_spot_max = host_cat_len - 1;
+		while (net_spot_max > 0 && host_cat[net_spot_max] == 0)
+			net_spot_max--;
+		if (net_spot_max > net_cat_len)
+			return -EINVAL;
+		memcpy(net_cat, host_cat, net_spot_max);
+		return net_spot_max;
+	case CIPSO_V4_MAP_STD:
+		for (;;) {
+			host_spot = cipso_v4_bitmap_walk(host_cat,
+							 host_clen_bits,
+							 host_spot + 1,
+							 1);
+			if (host_spot < 0)
+				break;
+			if (host_spot >= host_cat_size)
+				return -EPERM;
+
+			net_spot = host_cat_array[host_spot];
+			if (net_spot >= net_clen_bits)
+				return -ENOSPC;
+			cipso_v4_bitmap_setbit(net_cat, net_spot, 1);
+
+			if (net_spot > net_spot_max)
+				net_spot_max = net_spot;
+		}
+
+		if (host_spot == -2)
+			return -EFAULT;
+
+		if (++net_spot_max % 8)
+			return net_spot_max / 8 + 1;
+		return net_spot_max / 8;
+	}
+
+	return -EINVAL;
+}
+
+/**
+ * cipso_v4_map_cat_rbm_ntoh - Perform a category mapping from network to host
+ * @doi_def: the DOI definition
+ * @net_cat: the category bitmap in network/CIPSO format
+ * @net_cat_len: the length of the CIPSO bitmap in bytes
+ * @host_cat: the zero'd out category bitmap in host format
+ * @host_cat_len: the length of the host's category bitmap in bytes
+ *
+ * Description:
+ * Perform a label mapping to translate a CIPSO bitmap to the correct local
+ * MLS category bitmap using the given DOI definition.  Returns the minimum
+ * size in bytes of the host bitmap on success, negative values otherwise.
+ *
+ */
+static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def,
+				     const unsigned char *net_cat,
+				     u32 net_cat_len,
+				     unsigned char *host_cat,
+				     u32 host_cat_len)
+{
+	u32 host_spot;
+	u32 host_spot_max = 0;
+	int net_spot = -1;
+	u32 net_clen_bits = net_cat_len * 8;
+	u32 host_clen_bits = host_cat_len * 8;
+	u32 net_cat_size = doi_def->map.std->cat.cipso_size;
+	u32 *net_cat_array = doi_def->map.std->cat.cipso;
+
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_PASS:
+		if (net_cat_len > host_cat_len)
+			return -EINVAL;
+		memcpy(host_cat, net_cat, net_cat_len);
+		return net_cat_len;
+	case CIPSO_V4_MAP_STD:
+		for (;;) {
+			net_spot = cipso_v4_bitmap_walk(net_cat,
+							net_clen_bits,
+							net_spot + 1,
+							1);
+			if (net_spot < 0)
+				break;
+			if (net_spot >= net_cat_size ||
+			    net_cat_array[net_spot] >= CIPSO_V4_INV_CAT)
+				return -EPERM;
+
+			host_spot = net_cat_array[net_spot];
+			if (host_spot >= host_clen_bits)
+				return -ENOSPC;
+			cipso_v4_bitmap_setbit(host_cat, host_spot, 1);
+
+			if (host_spot > host_spot_max)
+				host_spot_max = host_spot;
+		}
+
+		if (net_spot == -2)
+			return -EFAULT;
+
+		if (++host_spot_max % 8)
+			return host_spot_max / 8 + 1;
+		return host_spot_max / 8;
+	}
+
+	return -EINVAL;
+}
+
+/*
+ * Protocol Handling Functions
+ */
+
+#define CIPSO_V4_HDR_LEN              6
+
+/**
+ * cipso_v4_gentag_hdr - Generate a CIPSO option header
+ * @doi_def: the DOI definition
+ * @len: the total tag length in bytes
+ * @buf: the CIPSO option buffer
+ *
+ * Description:
+ * Write a CIPSO header into the beginning of @buffer.  Return zero on success,
+ * negative values on failure.
+ *
+ */
+static int cipso_v4_gentag_hdr(const struct cipso_v4_doi *doi_def,
+			       u32 len,
+			       unsigned char *buf)
+{
+	if (CIPSO_V4_HDR_LEN + len > 40)
+		return -ENOSPC;
+
+	buf[0] = IPOPT_CIPSO;
+	buf[1] = CIPSO_V4_HDR_LEN + len;
+	*(u32 *)&buf[2] = htonl(doi_def->doi);
+
+	return 0;
+}
+
+#define CIPSO_V4_TAG1_CAT_LEN         30
+
+/**
+ * cipso_v4_gentag_rbm - Generate a CIPSO restricted bitmap tag (type #1)
+ * @doi_def: the DOI definition
+ * @secattr: the security attributes
+ * @buffer: the option buffer
+ * @buffer_len: length of buffer in bytes
+ *
+ * Description:
+ * Generate a CIPSO option using the restricted bitmap tag, tag type #1.  The
+ * actual buffer length may be larger than the indicated size due to
+ * translation between host and network category bitmaps.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def,
+			       const struct netlbl_lsm_secattr *secattr,
+			       unsigned char **buffer,
+			       u32 *buffer_len)
+{
+	int ret_val = -EPERM;
+	unsigned char *buf = NULL;
+	u32 buf_len;
+	u32 level;
+
+	if (secattr->mls_cat) {
+		buf = kzalloc(CIPSO_V4_HDR_LEN + 4 + CIPSO_V4_TAG1_CAT_LEN,
+			      GFP_ATOMIC);
+		if (buf == NULL)
+			return -ENOMEM;
+
+		ret_val = cipso_v4_map_cat_rbm_hton(doi_def,
+						    secattr->mls_cat,
+						    secattr->mls_cat_len,
+						    &buf[CIPSO_V4_HDR_LEN + 4],
+						    CIPSO_V4_TAG1_CAT_LEN);
+		if (ret_val < 0)
+			goto gentag_failure;
+
+		/* This will send packets using the "optimized" format when
+		 * possibile as specified in  section 3.4.2.6 of the
+		 * CIPSO draft. */
+		if (cipso_v4_rbm_optfmt && (ret_val > 0 && ret_val < 10))
+			ret_val = 10;
+
+		buf_len = 4 + ret_val;
+	} else {
+		buf = kzalloc(CIPSO_V4_HDR_LEN + 4, GFP_ATOMIC);
+		if (buf == NULL)
+			return -ENOMEM;
+		buf_len = 4;
+	}
+
+	ret_val = cipso_v4_map_lvl_hton(doi_def, secattr->mls_lvl, &level);
+	if (ret_val != 0)
+		goto gentag_failure;
+
+	ret_val = cipso_v4_gentag_hdr(doi_def, buf_len, buf);
+	if (ret_val != 0)
+		goto gentag_failure;
+
+	buf[CIPSO_V4_HDR_LEN] = 0x01;
+	buf[CIPSO_V4_HDR_LEN + 1] = buf_len;
+	buf[CIPSO_V4_HDR_LEN + 3] = level;
+
+	*buffer = buf;
+	*buffer_len = CIPSO_V4_HDR_LEN + buf_len;
+
+	return 0;
+
+gentag_failure:
+	kfree(buf);
+	return ret_val;
+}
+
+/**
+ * cipso_v4_parsetag_rbm - Parse a CIPSO restricted bitmap tag
+ * @doi_def: the DOI definition
+ * @tag: the CIPSO tag
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Parse a CIPSO restricted bitmap tag (tag type #1) and return the security
+ * attributes in @secattr.  Return zero on success, negatives values on
+ * failure.
+ *
+ */
+static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def,
+				 const unsigned char *tag,
+				 struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+	u8 tag_len = tag[1];
+	u32 level;
+
+	ret_val = cipso_v4_map_lvl_ntoh(doi_def, tag[3], &level);
+	if (ret_val != 0)
+		return ret_val;
+	secattr->mls_lvl = level;
+	secattr->mls_lvl_vld = 1;
+
+	if (tag_len > 4) {
+		switch (doi_def->type) {
+		case CIPSO_V4_MAP_PASS:
+			secattr->mls_cat_len = tag_len - 4;
+			break;
+		case CIPSO_V4_MAP_STD:
+			secattr->mls_cat_len =
+				doi_def->map.std->cat.local_size;
+			break;
+		}
+		secattr->mls_cat = kzalloc(secattr->mls_cat_len, GFP_ATOMIC);
+		if (secattr->mls_cat == NULL)
+			return -ENOMEM;
+
+		ret_val = cipso_v4_map_cat_rbm_ntoh(doi_def,
+						    &tag[4],
+						    tag_len - 4,
+						    secattr->mls_cat,
+						    secattr->mls_cat_len);
+		if (ret_val < 0) {
+			kfree(secattr->mls_cat);
+			return ret_val;
+		}
+		secattr->mls_cat_len = ret_val;
+	}
+
+	return 0;
+}
+
+/**
+ * cipso_v4_validate - Validate a CIPSO option
+ * @option: the start of the option, on error it is set to point to the error
+ *
+ * Description:
+ * This routine is called to validate a CIPSO option, it checks all of the
+ * fields to ensure that they are at least valid, see the draft snippet below
+ * for details.  If the option is valid then a zero value is returned and
+ * the value of @option is unchanged.  If the option is invalid then a
+ * non-zero value is returned and @option is adjusted to point to the
+ * offending portion of the option.  From the IETF draft ...
+ *
+ *  "If any field within the CIPSO options, such as the DOI identifier, is not
+ *   recognized the IP datagram is discarded and an ICMP 'parameter problem'
+ *   (type 12) is generated and returned.  The ICMP code field is set to 'bad
+ *   parameter' (code 0) and the pointer is set to the start of the CIPSO field
+ *   that is unrecognized."
+ *
+ */
+int cipso_v4_validate(unsigned char **option)
+{
+	unsigned char *opt = *option;
+	unsigned char *tag;
+	unsigned char opt_iter;
+	unsigned char err_offset = 0;
+	u8 opt_len;
+	u8 tag_len;
+	struct cipso_v4_doi *doi_def = NULL;
+	u32 tag_iter;
+
+	/* caller already checks for length values that are too large */
+	opt_len = opt[1];
+	if (opt_len < 8) {
+		err_offset = 1;
+		goto validate_return;
+	}
+
+	rcu_read_lock();
+	doi_def = cipso_v4_doi_getdef(ntohl(*((u32 *)&opt[2])));
+	if (doi_def == NULL) {
+		err_offset = 2;
+		goto validate_return_locked;
+	}
+
+	opt_iter = 6;
+	tag = opt + opt_iter;
+	while (opt_iter < opt_len) {
+		for (tag_iter = 0; doi_def->tags[tag_iter] != tag[0];)
+			if (doi_def->tags[tag_iter] == CIPSO_V4_TAG_INVALID ||
+			    ++tag_iter == CIPSO_V4_TAG_MAXCNT) {
+				err_offset = opt_iter;
+				goto validate_return_locked;
+			}
+
+		tag_len = tag[1];
+		if (tag_len > (opt_len - opt_iter)) {
+			err_offset = opt_iter + 1;
+			goto validate_return_locked;
+		}
+
+		switch (tag[0]) {
+		case CIPSO_V4_TAG_RBITMAP:
+			if (tag_len < 4) {
+				err_offset = opt_iter + 1;
+				goto validate_return_locked;
+			}
+
+			/* We are already going to do all the verification
+			 * necessary at the socket layer so from our point of
+			 * view it is safe to turn these checks off (and less
+			 * work), however, the CIPSO draft says we should do
+			 * all the CIPSO validations here but it doesn't
+			 * really specify _exactly_ what we need to validate
+			 * ... so, just make it a sysctl tunable. */
+			if (cipso_v4_rbm_strictvalid) {
+				if (cipso_v4_map_lvl_valid(doi_def,
+							   tag[3]) < 0) {
+					err_offset = opt_iter + 3;
+					goto validate_return_locked;
+				}
+				if (tag_len > 4 &&
+				    cipso_v4_map_cat_rbm_valid(doi_def,
+							    &tag[4],
+							    tag_len - 4) < 0) {
+					err_offset = opt_iter + 4;
+					goto validate_return_locked;
+				}
+			}
+			break;
+		default:
+			err_offset = opt_iter;
+			goto validate_return_locked;
+		}
+
+		tag += tag_len;
+		opt_iter += tag_len;
+	}
+
+validate_return_locked:
+	rcu_read_unlock();
+validate_return:
+	*option = opt + err_offset;
+	return err_offset;
+}
+
+/**
+ * cipso_v4_error - Send the correct reponse for a bad packet
+ * @skb: the packet
+ * @error: the error code
+ * @gateway: CIPSO gateway flag
+ *
+ * Description:
+ * Based on the error code given in @error, send an ICMP error message back to
+ * the originating host.  From the IETF draft ...
+ *
+ *  "If the contents of the CIPSO [option] are valid but the security label is
+ *   outside of the configured host or port label range, the datagram is
+ *   discarded and an ICMP 'destination unreachable' (type 3) is generated and
+ *   returned.  The code field of the ICMP is set to 'communication with
+ *   destination network administratively prohibited' (code 9) or to
+ *   'communication with destination host administratively prohibited'
+ *   (code 10).  The value of the code is dependent on whether the originator
+ *   of the ICMP message is acting as a CIPSO host or a CIPSO gateway.  The
+ *   recipient of the ICMP message MUST be able to handle either value.  The
+ *   same procedure is performed if a CIPSO [option] can not be added to an
+ *   IP packet because it is too large to fit in the IP options area."
+ *
+ *  "If the error is triggered by receipt of an ICMP message, the message is
+ *   discarded and no response is permitted (consistent with general ICMP
+ *   processing rules)."
+ *
+ */
+void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway)
+{
+	if (skb->nh.iph->protocol == IPPROTO_ICMP || error != -EACCES)
+		return;
+
+	if (gateway)
+		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_NET_ANO, 0);
+	else
+		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_ANO, 0);
+}
+
+/**
+ * cipso_v4_socket_setattr - Add a CIPSO option to a socket
+ * @sock: the socket
+ * @doi_def: the CIPSO DOI to use
+ * @secattr: the specific security attributes of the socket
+ *
+ * Description:
+ * Set the CIPSO option on the given socket using the DOI definition and
+ * security attributes passed to the function.  This function requires
+ * exclusive access to @sock->sk, which means it either needs to be in the
+ * process of being created or locked via lock_sock(sock->sk).  Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_socket_setattr(const struct socket *sock,
+			    const struct cipso_v4_doi *doi_def,
+			    const struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val = -EPERM;
+	u32 iter;
+	unsigned char *buf = NULL;
+	u32 buf_len = 0;
+	u32 opt_len;
+	struct ip_options *opt = NULL;
+	struct sock *sk;
+	struct inet_sock *sk_inet;
+	struct inet_connection_sock *sk_conn;
+
+	/* In the case of sock_create_lite(), the sock->sk field is not
+	 * defined yet but it is not a problem as the only users of these
+	 * "lite" PF_INET sockets are functions which do an accept() call
+	 * afterwards so we will label the socket as part of the accept(). */
+	sk = sock->sk;
+	if (sk == NULL)
+		return 0;
+
+	/* XXX - This code assumes only one tag per CIPSO option which isn't
+	 * really a good assumption to make but since we only support the MAC
+	 * tags right now it is a safe assumption. */
+	iter = 0;
+	do {
+		switch (doi_def->tags[iter]) {
+		case CIPSO_V4_TAG_RBITMAP:
+			ret_val = cipso_v4_gentag_rbm(doi_def,
+						      secattr,
+						      &buf,
+						      &buf_len);
+			break;
+		default:
+			ret_val = -EPERM;
+			goto socket_setattr_failure;
+		}
+
+		iter++;
+	} while (ret_val != 0 &&
+		 iter < CIPSO_V4_TAG_MAXCNT &&
+		 doi_def->tags[iter] != CIPSO_V4_TAG_INVALID);
+	if (ret_val != 0)
+		goto socket_setattr_failure;
+
+	/* We can't use ip_options_get() directly because it makes a call to
+	 * ip_options_get_alloc() which allocates memory with GFP_KERNEL and
+	 * we can't block here. */
+	opt_len = (buf_len + 3) & ~3;
+	opt = kzalloc(sizeof(*opt) + opt_len, GFP_ATOMIC);
+	if (opt == NULL) {
+		ret_val = -ENOMEM;
+		goto socket_setattr_failure;
+	}
+	memcpy(opt->__data, buf, buf_len);
+	opt->optlen = opt_len;
+	opt->is_data = 1;
+	kfree(buf);
+	buf = NULL;
+	ret_val = ip_options_compile(opt, NULL);
+	if (ret_val != 0)
+		goto socket_setattr_failure;
+
+	sk_inet = inet_sk(sk);
+	if (sk_inet->is_icsk) {
+		sk_conn = inet_csk(sk);
+		if (sk_inet->opt)
+			sk_conn->icsk_ext_hdr_len -= sk_inet->opt->optlen;
+		sk_conn->icsk_ext_hdr_len += opt->optlen;
+		sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie);
+	}
+	opt = xchg(&sk_inet->opt, opt);
+	kfree(opt);
+
+	return 0;
+
+socket_setattr_failure:
+	kfree(buf);
+	kfree(opt);
+	return ret_val;
+}
+
+/**
+ * cipso_v4_socket_getattr - Get the security attributes from a socket
+ * @sock: the socket
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Query @sock to see if there is a CIPSO option attached to the socket and if
+ * there is return the CIPSO security attributes in @secattr.  Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_socket_getattr(const struct socket *sock,
+			    struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val = -ENOMSG;
+	struct sock *sk;
+	struct inet_sock *sk_inet;
+	unsigned char *cipso_ptr;
+	u32 doi;
+	struct cipso_v4_doi *doi_def;
+
+	sk = sock->sk;
+	lock_sock(sk);
+	sk_inet = inet_sk(sk);
+	if (sk_inet->opt == NULL || sk_inet->opt->cipso == 0)
+		goto socket_getattr_return;
+	cipso_ptr = sk_inet->opt->__data + sk_inet->opt->cipso -
+		sizeof(struct iphdr);
+	ret_val = cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr);
+	if (ret_val == 0)
+		goto socket_getattr_return;
+
+	doi = ntohl(*(u32 *)&cipso_ptr[2]);
+	rcu_read_lock();
+	doi_def = cipso_v4_doi_getdef(doi);
+	if (doi_def == NULL) {
+		rcu_read_unlock();
+		goto socket_getattr_return;
+	}
+	switch (cipso_ptr[6]) {
+	case CIPSO_V4_TAG_RBITMAP:
+		ret_val = cipso_v4_parsetag_rbm(doi_def,
+						&cipso_ptr[6],
+						secattr);
+		break;
+	}
+	rcu_read_unlock();
+
+socket_getattr_return:
+	release_sock(sk);
+	return ret_val;
+}
+
+/**
+ * cipso_v4_skbuff_getattr - Get the security attributes from the CIPSO option
+ * @skb: the packet
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Parse the given packet's CIPSO option and return the security attributes.
+ * Returns zero on success and negative values on failure.
+ *
+ */
+int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
+			    struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val = -ENOMSG;
+	unsigned char *cipso_ptr;
+	u32 doi;
+	struct cipso_v4_doi *doi_def;
+
+	if (!CIPSO_V4_OPTEXIST(skb))
+		return -ENOMSG;
+	cipso_ptr = CIPSO_V4_OPTPTR(skb);
+	if (cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr) == 0)
+		return 0;
+
+	doi = ntohl(*(u32 *)&cipso_ptr[2]);
+	rcu_read_lock();
+	doi_def = cipso_v4_doi_getdef(doi);
+	if (doi_def == NULL)
+		goto skbuff_getattr_return;
+	switch (cipso_ptr[6]) {
+	case CIPSO_V4_TAG_RBITMAP:
+		ret_val = cipso_v4_parsetag_rbm(doi_def,
+						&cipso_ptr[6],
+						secattr);
+		break;
+	}
+
+skbuff_getattr_return:
+	rcu_read_unlock();
+	return ret_val;
+}
+
+/*
+ * Setup Functions
+ */
+
+/**
+ * cipso_v4_init - Initialize the CIPSO module
+ *
+ * Description:
+ * Initialize the CIPSO module and prepare it for use.  Returns zero on success
+ * and negative values on failure.
+ *
+ */
+static int __init cipso_v4_init(void)
+{
+	int ret_val;
+
+	ret_val = cipso_v4_cache_init();
+	if (ret_val != 0)
+		panic("Failed to initialize the CIPSO/IPv4 cache (%d)\n",
+		      ret_val);
+
+	return 0;
+}
+
+subsys_initcall(cipso_v4_init);
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 70cea9d08a38..19b2071ff319 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -17,6 +17,7 @@
 #include <net/ip.h>
 #include <net/route.h>
 #include <net/tcp.h>
+#include <net/cipso_ipv4.h>
 
 /* From af_inet.c */
 extern int sysctl_ip_nonlocal_bind;
@@ -697,6 +698,40 @@ ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
+#ifdef CONFIG_NETLABEL
+	{
+		.ctl_name	= NET_CIPSOV4_CACHE_ENABLE,
+		.procname	= "cipso_cache_enable",
+		.data		= &cipso_v4_cache_enabled,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= NET_CIPSOV4_CACHE_BUCKET_SIZE,
+		.procname	= "cipso_cache_bucket_size",
+		.data		= &cipso_v4_cache_bucketsize,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= NET_CIPSOV4_RBM_OPTFMT,
+		.procname	= "cipso_rbm_optfmt",
+		.data		= &cipso_v4_rbm_optfmt,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= NET_CIPSOV4_RBM_STRICTVALID,
+		.procname	= "cipso_rbm_strictvalid",
+		.data		= &cipso_v4_rbm_strictvalid,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+#endif /* CONFIG_NETLABEL */
 	{ .ctl_name = 0 }
 };
 
-- 
cgit v1.2.3


From 7420ed23a4f77480b5b7b3245e5da30dd24b7575 Mon Sep 17 00:00:00 2001
From: Venkat Yekkirala <vyekkirala@TrustedCS.com>
Date: Fri, 4 Aug 2006 23:17:57 -0700
Subject: [NetLabel]: SELinux support

Add NetLabel support to the SELinux LSM and modify the
socket_post_create() LSM hook to return an error code.  The most
significant part of this patch is the addition of NetLabel hooks into
the following SELinux LSM hooks:

 * selinux_file_permission()
 * selinux_socket_sendmsg()
 * selinux_socket_post_create()
 * selinux_socket_sock_rcv_skb()
 * selinux_socket_getpeersec_stream()
 * selinux_socket_getpeersec_dgram()
 * selinux_sock_graft()
 * selinux_inet_conn_request()

The basic reasoning behind this patch is that outgoing packets are
"NetLabel'd" by labeling their socket and the NetLabel security
attributes are checked via the additional hook in
selinux_socket_sock_rcv_skb().  NetLabel itself is only a labeling
mechanism, similar to filesystem extended attributes, it is up to the
SELinux enforcement mechanism to perform the actual access checks.

In addition to the changes outlined above this patch also includes
some changes to the extended bitmap (ebitmap) and multi-level security
(mls) code to import and export SELinux TE/MLS attributes into and out
of NetLabel.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/security.h                    |  25 +-
 net/socket.c                                |  13 +-
 security/dummy.c                            |   6 +-
 security/selinux/hooks.c                    |  56 +++-
 security/selinux/include/objsec.h           |   8 +
 security/selinux/include/selinux_netlabel.h | 125 +++++++
 security/selinux/ss/ebitmap.c               | 144 ++++++++
 security/selinux/ss/ebitmap.h               |   6 +
 security/selinux/ss/mls.c                   | 156 +++++++++
 security/selinux/ss/mls.h                   |  21 ++
 security/selinux/ss/services.c              | 488 ++++++++++++++++++++++++++++
 11 files changed, 1020 insertions(+), 28 deletions(-)
 create mode 100644 security/selinux/include/selinux_netlabel.h

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index bb4c80fdfe7a..9f56fb8a4a6c 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1341,8 +1341,8 @@ struct security_operations {
 	int (*unix_may_send) (struct socket * sock, struct socket * other);
 
 	int (*socket_create) (int family, int type, int protocol, int kern);
-	void (*socket_post_create) (struct socket * sock, int family,
-				    int type, int protocol, int kern);
+	int (*socket_post_create) (struct socket * sock, int family,
+				   int type, int protocol, int kern);
 	int (*socket_bind) (struct socket * sock,
 			    struct sockaddr * address, int addrlen);
 	int (*socket_connect) (struct socket * sock,
@@ -2824,13 +2824,13 @@ static inline int security_socket_create (int family, int type,
 	return security_ops->socket_create(family, type, protocol, kern);
 }
 
-static inline void security_socket_post_create(struct socket * sock, 
-					       int family,
-					       int type, 
-					       int protocol, int kern)
+static inline int security_socket_post_create(struct socket * sock,
+					      int family,
+					      int type,
+					      int protocol, int kern)
 {
-	security_ops->socket_post_create(sock, family, type,
-					 protocol, kern);
+	return security_ops->socket_post_create(sock, family, type,
+						protocol, kern);
 }
 
 static inline int security_socket_bind(struct socket * sock, 
@@ -2982,11 +2982,12 @@ static inline int security_socket_create (int family, int type,
 	return 0;
 }
 
-static inline void security_socket_post_create(struct socket * sock, 
-					       int family,
-					       int type, 
-					       int protocol, int kern)
+static inline int security_socket_post_create(struct socket * sock,
+					      int family,
+					      int type,
+					      int protocol, int kern)
 {
+	return 0;
 }
 
 static inline int security_socket_bind(struct socket * sock, 
diff --git a/net/socket.c b/net/socket.c
index 6d261bf206fc..6756e57e1ff0 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -973,11 +973,18 @@ int sock_create_lite(int family, int type, int protocol, struct socket **res)
 		goto out;
 	}
 
-	security_socket_post_create(sock, family, type, protocol, 1);
 	sock->type = type;
+	err = security_socket_post_create(sock, family, type, protocol, 1);
+	if (err)
+		goto out_release;
+
 out:
 	*res = sock;
 	return err;
+out_release:
+	sock_release(sock);
+	sock = NULL;
+	goto out;
 }
 
 /* No kernel lock held - perfect */
@@ -1214,7 +1221,9 @@ static int __sock_create(int family, int type, int protocol, struct socket **res
 	 */
 	module_put(net_families[family]->owner);
 	*res = sock;
-	security_socket_post_create(sock, family, type, protocol, kern);
+	err = security_socket_post_create(sock, family, type, protocol, kern);
+	if (err)
+		goto out_release;
 
 out:
 	net_family_read_unlock();
diff --git a/security/dummy.c b/security/dummy.c
index 1c45f8e4aad1..aeee70565509 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -709,10 +709,10 @@ static int dummy_socket_create (int family, int type,
 	return 0;
 }
 
-static void dummy_socket_post_create (struct socket *sock, int family, int type,
-				      int protocol, int kern)
+static int dummy_socket_post_create (struct socket *sock, int family, int type,
+				     int protocol, int kern)
 {
-	return;
+	return 0;
 }
 
 static int dummy_socket_bind (struct socket *sock, struct sockaddr *address,
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 33028b3b19ce..2a6bbb921e1e 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -12,6 +12,8 @@
  *  Copyright (C) 2003 Red Hat, Inc., James Morris <jmorris@redhat.com>
  *  Copyright (C) 2004-2005 Trusted Computer Solutions, Inc.
  *                          <dgoeddel@trustedcs.com>
+ *  Copyright (C) 2006 Hewlett-Packard Development Company, L.P.
+ *                     Paul Moore, <paul.moore@hp.com>
  *
  *	This program is free software; you can redistribute it and/or modify
  *	it under the terms of the GNU General Public License version 2,
@@ -74,6 +76,7 @@
 #include "objsec.h"
 #include "netif.h"
 #include "xfrm.h"
+#include "selinux_netlabel.h"
 
 #define XATTR_SELINUX_SUFFIX "selinux"
 #define XATTR_NAME_SELINUX XATTR_SECURITY_PREFIX XATTR_SELINUX_SUFFIX
@@ -2395,6 +2398,7 @@ static int selinux_inode_listsecurity(struct inode *inode, char *buffer, size_t
 
 static int selinux_file_permission(struct file *file, int mask)
 {
+	int rc;
 	struct inode *inode = file->f_dentry->d_inode;
 
 	if (!mask) {
@@ -2406,8 +2410,12 @@ static int selinux_file_permission(struct file *file, int mask)
 	if ((file->f_flags & O_APPEND) && (mask & MAY_WRITE))
 		mask |= MAY_APPEND;
 
-	return file_has_perm(current, file,
-			     file_mask_to_av(inode->i_mode, mask));
+	rc = file_has_perm(current, file,
+			   file_mask_to_av(inode->i_mode, mask));
+	if (rc)
+		return rc;
+
+	return selinux_netlbl_inode_permission(inode, mask);
 }
 
 static int selinux_file_alloc_security(struct file *file)
@@ -3058,9 +3066,10 @@ out:
 	return err;
 }
 
-static void selinux_socket_post_create(struct socket *sock, int family,
-				       int type, int protocol, int kern)
+static int selinux_socket_post_create(struct socket *sock, int family,
+				      int type, int protocol, int kern)
 {
+	int err = 0;
 	struct inode_security_struct *isec;
 	struct task_security_struct *tsec;
 	struct sk_security_struct *sksec;
@@ -3077,9 +3086,12 @@ static void selinux_socket_post_create(struct socket *sock, int family,
 	if (sock->sk) {
 		sksec = sock->sk->sk_security;
 		sksec->sid = isec->sid;
+		err = selinux_netlbl_socket_post_create(sock,
+							family,
+							isec->sid);
 	}
 
-	return;
+	return err;
 }
 
 /* Range of port numbers used to automatically bind.
@@ -3260,7 +3272,13 @@ static int selinux_socket_accept(struct socket *sock, struct socket *newsock)
 static int selinux_socket_sendmsg(struct socket *sock, struct msghdr *msg,
  				  int size)
 {
-	return socket_has_perm(current, sock, SOCKET__WRITE);
+	int rc;
+
+	rc = socket_has_perm(current, sock, SOCKET__WRITE);
+	if (rc)
+		return rc;
+
+	return selinux_netlbl_inode_permission(SOCK_INODE(sock), MAY_WRITE);
 }
 
 static int selinux_socket_recvmsg(struct socket *sock, struct msghdr *msg,
@@ -3468,6 +3486,10 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (err)
 		goto out;
 
+	err = selinux_netlbl_sock_rcv_skb(sksec, skb, &ad);
+	if (err)
+		goto out;
+
 	err = selinux_xfrm_sock_rcv_skb(sksec->sid, skb, &ad);
 out:	
 	return err;
@@ -3491,8 +3513,9 @@ static int selinux_socket_getpeersec_stream(struct socket *sock, char __user *op
 		peer_sid = ssec->peer_sid;
 	}
 	else if (isec->sclass == SECCLASS_TCP_SOCKET) {
-		peer_sid = selinux_socket_getpeer_stream(sock->sk);
-
+		peer_sid = selinux_netlbl_socket_getpeersec_stream(sock);
+		if (peer_sid == SECSID_NULL)
+			peer_sid = selinux_socket_getpeer_stream(sock->sk);
 		if (peer_sid == SECSID_NULL) {
 			err = -ENOPROTOOPT;
 			goto out;
@@ -3532,8 +3555,11 @@ static int selinux_socket_getpeersec_dgram(struct socket *sock, struct sk_buff *
 
 	if (sock && (sock->sk->sk_family == PF_UNIX))
 		selinux_get_inode_sid(SOCK_INODE(sock), &peer_secid);
-	else if (skb)
-		peer_secid = selinux_socket_getpeer_dgram(skb);
+	else if (skb) {
+		peer_secid = selinux_netlbl_socket_getpeersec_dgram(skb);
+		if (peer_secid == SECSID_NULL)
+			peer_secid = selinux_socket_getpeer_dgram(skb);
+	}
 
 	if (peer_secid == SECSID_NULL)
 		err = -EINVAL;
@@ -3578,6 +3604,8 @@ void selinux_sock_graft(struct sock* sk, struct socket *parent)
 	struct sk_security_struct *sksec = sk->sk_security;
 
 	isec->sid = sksec->sid;
+
+	selinux_netlbl_sock_graft(sk, parent);
 }
 
 int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb,
@@ -3585,9 +3613,15 @@ int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb,
 {
 	struct sk_security_struct *sksec = sk->sk_security;
 	int err;
-	u32 newsid = 0;
+	u32 newsid;
 	u32 peersid;
 
+	newsid = selinux_netlbl_inet_conn_request(skb, sksec->sid);
+	if (newsid != SECSID_NULL) {
+		req->secid = newsid;
+		return 0;
+	}
+
 	err = selinux_xfrm_decode_session(skb, &peersid, 0);
 	BUG_ON(err);
 
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 79b9e0af19a0..0a39bfd1319f 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -101,6 +101,14 @@ struct sk_security_struct {
 	struct sock *sk;		/* back pointer to sk object */
 	u32 sid;			/* SID of this object */
 	u32 peer_sid;			/* SID of peer */
+#ifdef CONFIG_NETLABEL
+	u16 sclass;			/* sock security class */
+	enum {				/* NetLabel state */
+		NLBL_UNSET = 0,
+		NLBL_REQUIRE,
+		NLBL_LABELED,
+	} nlbl_state;
+#endif
 };
 
 struct key_security_struct {
diff --git a/security/selinux/include/selinux_netlabel.h b/security/selinux/include/selinux_netlabel.h
new file mode 100644
index 000000000000..88c463eef1e1
--- /dev/null
+++ b/security/selinux/include/selinux_netlabel.h
@@ -0,0 +1,125 @@
+/*
+ * SELinux interface to the NetLabel subsystem
+ *
+ * Author : Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program;  if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _SELINUX_NETLABEL_H_
+#define _SELINUX_NETLABEL_H_
+
+#ifdef CONFIG_NETLABEL
+void selinux_netlbl_cache_invalidate(void);
+int selinux_netlbl_socket_post_create(struct socket *sock,
+				      int sock_family,
+				      u32 sid);
+void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock);
+u32 selinux_netlbl_inet_conn_request(struct sk_buff *skb, u32 sock_sid);
+int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
+				struct sk_buff *skb,
+				struct avc_audit_data *ad);
+u32 selinux_netlbl_socket_getpeersec_stream(struct socket *sock);
+u32 selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb);
+
+int __selinux_netlbl_inode_permission(struct inode *inode, int mask);
+/**
+ * selinux_netlbl_inode_permission - Verify the socket is NetLabel labeled
+ * @inode: the file descriptor's inode
+ * @mask: the permission mask
+ *
+ * Description:
+ * Looks at a file's inode and if it is marked as a socket protected by
+ * NetLabel then verify that the socket has been labeled, if not try to label
+ * the socket now with the inode's SID.  Returns zero on success, negative
+ * values on failure.
+ *
+ */
+static inline int selinux_netlbl_inode_permission(struct inode *inode,
+						  int mask)
+{
+	int rc = 0;
+	struct inode_security_struct *isec;
+	struct sk_security_struct *sksec;
+
+	if (!S_ISSOCK(inode->i_mode))
+		return 0;
+
+	isec = inode->i_security;
+	sksec = SOCKET_I(inode)->sk->sk_security;
+	down(&isec->sem);
+	if (unlikely(sksec->nlbl_state == NLBL_REQUIRE &&
+		     (mask & (MAY_WRITE | MAY_APPEND))))
+		rc = __selinux_netlbl_inode_permission(inode, mask);
+	up(&isec->sem);
+
+	return rc;
+}
+#else
+static inline void selinux_netlbl_cache_invalidate(void)
+{
+	return;
+}
+
+static inline int selinux_netlbl_socket_post_create(struct socket *sock,
+						    int sock_family,
+						    u32 sid)
+{
+	return 0;
+}
+
+static inline void selinux_netlbl_sock_graft(struct sock *sk,
+					     struct socket *sock)
+{
+	return;
+}
+
+static inline u32 selinux_netlbl_inet_conn_request(struct sk_buff *skb,
+						   u32 sock_sid)
+{
+	return SECSID_NULL;
+}
+
+static inline int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
+					      struct sk_buff *skb,
+					      struct avc_audit_data *ad)
+{
+	return 0;
+}
+
+static inline u32 selinux_netlbl_socket_getpeersec_stream(struct socket *sock)
+{
+	return SECSID_NULL;
+}
+
+static inline u32 selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb)
+{
+	return SECSID_NULL;
+}
+
+static inline int selinux_netlbl_inode_permission(struct inode *inode,
+						  int mask)
+{
+	return 0;
+}
+#endif /* CONFIG_NETLABEL */
+
+#endif
diff --git a/security/selinux/ss/ebitmap.c b/security/selinux/ss/ebitmap.c
index 47024a6e1844..4b915eb60c45 100644
--- a/security/selinux/ss/ebitmap.c
+++ b/security/selinux/ss/ebitmap.c
@@ -3,6 +3,14 @@
  *
  * Author : Stephen Smalley, <sds@epoch.ncsc.mil>
  */
+/*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ *      Added ebitmap_export() and ebitmap_import()
+ *
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
+
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/errno.h>
@@ -59,6 +67,142 @@ int ebitmap_cpy(struct ebitmap *dst, struct ebitmap *src)
 	return 0;
 }
 
+/**
+ * ebitmap_export - Export an ebitmap to a unsigned char bitmap string
+ * @src: the ebitmap to export
+ * @dst: the resulting bitmap string
+ * @dst_len: length of dst in bytes
+ *
+ * Description:
+ * Allocate a buffer at least src->highbit bits long and export the extensible
+ * bitmap into the buffer.  The bitmap string will be in little endian format,
+ * i.e. LSB first.  The value returned in dst_len may not the true size of the
+ * buffer as the length of the buffer is rounded up to a multiple of MAPTYPE.
+ * The caller must free the buffer when finished. Returns zero on success,
+ * negative values on failure.
+ *
+ */
+int ebitmap_export(const struct ebitmap *src,
+		   unsigned char **dst,
+		   size_t *dst_len)
+{
+	size_t bitmap_len;
+	unsigned char *bitmap;
+	struct ebitmap_node *iter_node;
+	MAPTYPE node_val;
+	size_t bitmap_byte;
+	unsigned char bitmask;
+
+	bitmap_len = src->highbit / 8;
+	if (src->highbit % 7)
+		bitmap_len += 1;
+	if (bitmap_len == 0)
+		return -EINVAL;
+
+	bitmap = kzalloc((bitmap_len & ~(sizeof(MAPTYPE) - 1)) +
+			 sizeof(MAPTYPE),
+			 GFP_ATOMIC);
+	if (bitmap == NULL)
+		return -ENOMEM;
+
+	iter_node = src->node;
+	do {
+		bitmap_byte = iter_node->startbit / 8;
+		bitmask = 0x80;
+		node_val = iter_node->map;
+		do {
+			if (bitmask == 0) {
+				bitmap_byte++;
+				bitmask = 0x80;
+			}
+			if (node_val & (MAPTYPE)0x01)
+				bitmap[bitmap_byte] |= bitmask;
+			node_val >>= 1;
+			bitmask >>= 1;
+		} while (node_val > 0);
+		iter_node = iter_node->next;
+	} while (iter_node);
+
+	*dst = bitmap;
+	*dst_len = bitmap_len;
+	return 0;
+}
+
+/**
+ * ebitmap_import - Import an unsigned char bitmap string into an ebitmap
+ * @src: the bitmap string
+ * @src_len: the bitmap length in bytes
+ * @dst: the empty ebitmap
+ *
+ * Description:
+ * This function takes a little endian bitmap string in src and imports it into
+ * the ebitmap pointed to by dst.  Returns zero on success, negative values on
+ * failure.
+ *
+ */
+int ebitmap_import(const unsigned char *src,
+		   size_t src_len,
+		   struct ebitmap *dst)
+{
+	size_t src_off = 0;
+	struct ebitmap_node *node_new;
+	struct ebitmap_node *node_last = NULL;
+	size_t iter;
+	size_t iter_bit;
+	size_t iter_limit;
+	unsigned char src_byte;
+
+	do {
+		iter_limit = src_len - src_off;
+		if (iter_limit >= sizeof(MAPTYPE)) {
+			if (*(MAPTYPE *)&src[src_off] == 0) {
+				src_off += sizeof(MAPTYPE);
+				continue;
+			}
+			iter_limit = sizeof(MAPTYPE);
+		} else {
+			iter = src_off;
+			src_byte = 0;
+			do {
+				src_byte |= src[iter++];
+			} while (iter < src_len && src_byte == 0);
+			if (src_byte == 0)
+				break;
+		}
+
+		node_new = kzalloc(sizeof(*node_new), GFP_ATOMIC);
+		if (unlikely(node_new == NULL)) {
+			ebitmap_destroy(dst);
+			return -ENOMEM;
+		}
+		node_new->startbit = src_off * 8;
+		iter = 0;
+		do {
+			src_byte = src[src_off++];
+			iter_bit = iter++ * 8;
+			while (src_byte != 0) {
+				if (src_byte & 0x80)
+					node_new->map |= MAPBIT << iter_bit;
+				iter_bit++;
+				src_byte <<= 1;
+			}
+		} while (iter < iter_limit);
+
+		if (node_last != NULL)
+			node_last->next = node_new;
+		else
+			dst->node = node_new;
+		node_last = node_new;
+	} while (src_off < src_len);
+
+	if (likely(node_last != NULL))
+		dst->highbit = node_last->startbit + MAPSIZE;
+	else
+		ebitmap_init(dst);
+
+	return 0;
+}
+
 int ebitmap_contains(struct ebitmap *e1, struct ebitmap *e2)
 {
 	struct ebitmap_node *n1, *n2;
diff --git a/security/selinux/ss/ebitmap.h b/security/selinux/ss/ebitmap.h
index 8bf41055a6cb..da2d4651b10d 100644
--- a/security/selinux/ss/ebitmap.h
+++ b/security/selinux/ss/ebitmap.h
@@ -69,6 +69,12 @@ static inline int ebitmap_node_get_bit(struct ebitmap_node * n,
 
 int ebitmap_cmp(struct ebitmap *e1, struct ebitmap *e2);
 int ebitmap_cpy(struct ebitmap *dst, struct ebitmap *src);
+int ebitmap_export(const struct ebitmap *src,
+		   unsigned char **dst,
+		   size_t *dst_len);
+int ebitmap_import(const unsigned char *src,
+		   size_t src_len,
+		   struct ebitmap *dst);
 int ebitmap_contains(struct ebitmap *e1, struct ebitmap *e2);
 int ebitmap_get_bit(struct ebitmap *e, unsigned long bit);
 int ebitmap_set_bit(struct ebitmap *e, unsigned long bit, int value);
diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c
index e15f7e0399b8..119bd6078ba1 100644
--- a/security/selinux/ss/mls.c
+++ b/security/selinux/ss/mls.c
@@ -10,6 +10,13 @@
  *
  * Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
  */
+/*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ *      Added support to import/export the MLS label
+ *
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
 
 #include <linux/kernel.h>
 #include <linux/slab.h>
@@ -565,3 +572,152 @@ int mls_compute_sid(struct context *scontext,
 	return -EINVAL;
 }
 
+/**
+ * mls_export_lvl - Export the MLS sensitivity levels
+ * @context: the security context
+ * @low: the low sensitivity level
+ * @high: the high sensitivity level
+ *
+ * Description:
+ * Given the security context copy the low MLS sensitivity level into lvl_low
+ * and the high sensitivity level in lvl_high.  The MLS levels are only
+ * exported if the pointers are not NULL, if they are NULL then that level is
+ * not exported.
+ *
+ */
+void mls_export_lvl(const struct context *context, u32 *low, u32 *high)
+{
+	if (!selinux_mls_enabled)
+		return;
+
+	if (low != NULL)
+		*low = context->range.level[0].sens - 1;
+	if (high != NULL)
+		*high = context->range.level[1].sens - 1;
+}
+
+/**
+ * mls_import_lvl - Import the MLS sensitivity levels
+ * @context: the security context
+ * @low: the low sensitivity level
+ * @high: the high sensitivity level
+ *
+ * Description:
+ * Given the security context and the two sensitivty levels, set the MLS levels
+ * in the context according the two given as parameters.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+void mls_import_lvl(struct context *context, u32 low, u32 high)
+{
+	if (!selinux_mls_enabled)
+		return;
+
+	context->range.level[0].sens = low + 1;
+	context->range.level[1].sens = high + 1;
+}
+
+/**
+ * mls_export_cat - Export the MLS categories
+ * @context: the security context
+ * @low: the low category
+ * @low_len: length of the cat_low bitmap in bytes
+ * @high: the high category
+ * @high_len: length of the cat_high bitmap in bytes
+ *
+ * Description:
+ * Given the security context export the low MLS category bitmap into cat_low
+ * and the high category bitmap into cat_high.  The MLS categories are only
+ * exported if the pointers are not NULL, if they are NULL then that level is
+ * not exported.  The caller is responsibile for freeing the memory when
+ * finished.  Returns zero on success, negative values on failure.
+ *
+ */
+int mls_export_cat(const struct context *context,
+		   unsigned char **low,
+		   size_t *low_len,
+		   unsigned char **high,
+		   size_t *high_len)
+{
+	int rc = -EPERM;
+
+	if (!selinux_mls_enabled)
+		return 0;
+
+	if (low != NULL) {
+		rc = ebitmap_export(&context->range.level[0].cat,
+				    low,
+				    low_len);
+		if (rc != 0)
+			goto export_cat_failure;
+	}
+	if (high != NULL) {
+		rc = ebitmap_export(&context->range.level[1].cat,
+				    high,
+				    high_len);
+		if (rc != 0)
+			goto export_cat_failure;
+	}
+
+	return 0;
+
+export_cat_failure:
+	if (low != NULL)
+		kfree(*low);
+	if (high != NULL)
+		kfree(*high);
+	return rc;
+}
+
+/**
+ * mls_import_cat - Import the MLS categories
+ * @context: the security context
+ * @low: the low category
+ * @low_len: length of the cat_low bitmap in bytes
+ * @high: the high category
+ * @high_len: length of the cat_high bitmap in bytes
+ *
+ * Description:
+ * Given the security context and the two category bitmap strings import the
+ * categories into the security context.  The MLS categories are only imported
+ * if the pointers are not NULL, if they are NULL they are skipped.  Returns
+ * zero on success, negative values on failure.
+ *
+ */
+int mls_import_cat(struct context *context,
+		   const unsigned char *low,
+		   size_t low_len,
+		   const unsigned char *high,
+		   size_t high_len)
+{
+	int rc = -EPERM;
+
+	if (!selinux_mls_enabled)
+		return 0;
+
+	if (low != NULL) {
+		rc = ebitmap_import(low,
+				    low_len,
+				    &context->range.level[0].cat);
+		if (rc != 0)
+			goto import_cat_failure;
+	}
+	if (high != NULL) {
+		if (high == low)
+			rc = ebitmap_cpy(&context->range.level[1].cat,
+					 &context->range.level[0].cat);
+		else
+			rc = ebitmap_import(high,
+					    high_len,
+					    &context->range.level[1].cat);
+		if (rc != 0)
+			goto import_cat_failure;
+	}
+
+	return 0;
+
+import_cat_failure:
+	ebitmap_destroy(&context->range.level[0].cat);
+	ebitmap_destroy(&context->range.level[1].cat);
+	return rc;
+}
diff --git a/security/selinux/ss/mls.h b/security/selinux/ss/mls.h
index 90c5e88987fa..df6032c6d492 100644
--- a/security/selinux/ss/mls.h
+++ b/security/selinux/ss/mls.h
@@ -10,6 +10,13 @@
  *
  * Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
  */
+/*
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ *      Added support to import/export the MLS label
+ *
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
 
 #ifndef _SS_MLS_H_
 #define _SS_MLS_H_
@@ -62,5 +69,19 @@ int mls_compute_sid(struct context *scontext,
 int mls_setup_user_range(struct context *fromcon, struct user_datum *user,
                          struct context *usercon);
 
+void mls_export_lvl(const struct context *context, u32 *low, u32 *high);
+void mls_import_lvl(struct context *context, u32 low, u32 high);
+
+int mls_export_cat(const struct context *context,
+		   unsigned char **low,
+		   size_t *low_len,
+		   unsigned char **high,
+		   size_t *high_len);
+int mls_import_cat(struct context *context,
+		   const unsigned char *low,
+		   size_t low_len,
+		   const unsigned char *high,
+		   size_t high_len);
+
 #endif	/* _SS_MLS_H */
 
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index b00ec69f0ffd..910afa1ffc31 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -13,6 +13,11 @@
  *
  * 	Added conditional policy language extensions
  *
+ * Updated: Hewlett-Packard <paul.moore@hp.com>
+ *
+ *      Added support for NetLabel
+ *
+ * Copyright (C) 2006 Hewlett-Packard Development Company, L.P.
  * Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
  * Copyright (C) 2003 - 2004 Tresys Technology, LLC
  * Copyright (C) 2003 Red Hat, Inc., James Morris <jmorris@redhat.com>
@@ -29,6 +34,8 @@
 #include <linux/sched.h>
 #include <linux/audit.h>
 #include <linux/mutex.h>
+#include <net/sock.h>
+#include <net/netlabel.h>
 
 #include "flask.h"
 #include "avc.h"
@@ -40,6 +47,8 @@
 #include "services.h"
 #include "conditional.h"
 #include "mls.h"
+#include "objsec.h"
+#include "selinux_netlabel.h"
 
 extern void selnl_notify_policyload(u32 seqno);
 unsigned int policydb_loaded_version;
@@ -1241,6 +1250,7 @@ int security_load_policy(void *data, size_t len)
 		selinux_complete_init();
 		avc_ss_reset(seqno);
 		selnl_notify_policyload(seqno);
+		selinux_netlbl_cache_invalidate();
 		return 0;
 	}
 
@@ -1295,6 +1305,7 @@ int security_load_policy(void *data, size_t len)
 
 	avc_ss_reset(seqno);
 	selnl_notify_policyload(seqno);
+	selinux_netlbl_cache_invalidate();
 
 	return 0;
 
@@ -2133,3 +2144,480 @@ void selinux_audit_set_callback(int (*callback)(void))
 {
 	aurule_callback = callback;
 }
+
+#ifdef CONFIG_NETLABEL
+/*
+ * This is the structure we store inside the NetLabel cache block.
+ */
+#define NETLBL_CACHE(x)           ((struct netlbl_cache *)(x))
+#define NETLBL_CACHE_T_NONE       0
+#define NETLBL_CACHE_T_SID        1
+#define NETLBL_CACHE_T_MLS        2
+struct netlbl_cache {
+	u32 type;
+	union {
+		u32 sid;
+		struct mls_range mls_label;
+	} data;
+};
+
+/**
+ * selinux_netlbl_cache_free - Free the NetLabel cached data
+ * @data: the data to free
+ *
+ * Description:
+ * This function is intended to be used as the free() callback inside the
+ * netlbl_lsm_cache structure.
+ *
+ */
+static void selinux_netlbl_cache_free(const void *data)
+{
+	struct netlbl_cache *cache = NETLBL_CACHE(data);
+	switch (cache->type) {
+	case NETLBL_CACHE_T_MLS:
+		ebitmap_destroy(&cache->data.mls_label.level[0].cat);
+		break;
+	}
+	kfree(data);
+}
+
+/**
+ * selinux_netlbl_cache_add - Add an entry to the NetLabel cache
+ * @skb: the packet
+ * @ctx: the SELinux context
+ *
+ * Description:
+ * Attempt to cache the context in @ctx, which was derived from the packet in
+ * @skb, in the NetLabel subsystem cache.
+ *
+ */
+static void selinux_netlbl_cache_add(struct sk_buff *skb, struct context *ctx)
+{
+	struct netlbl_cache *cache = NULL;
+	struct netlbl_lsm_secattr secattr;
+
+	netlbl_secattr_init(&secattr);
+
+	cache = kzalloc(sizeof(*cache),	GFP_ATOMIC);
+	if (cache == NULL)
+		goto netlbl_cache_add_failure;
+	secattr.cache.free = selinux_netlbl_cache_free;
+	secattr.cache.data = (void *)cache;
+
+	cache->type = NETLBL_CACHE_T_MLS;
+	if (ebitmap_cpy(&cache->data.mls_label.level[0].cat,
+			&ctx->range.level[0].cat) != 0)
+		goto netlbl_cache_add_failure;
+	cache->data.mls_label.level[1].cat.highbit =
+		cache->data.mls_label.level[0].cat.highbit;
+	cache->data.mls_label.level[1].cat.node =
+		cache->data.mls_label.level[0].cat.node;
+	cache->data.mls_label.level[0].sens = ctx->range.level[0].sens;
+	cache->data.mls_label.level[1].sens = ctx->range.level[0].sens;
+
+	if (netlbl_cache_add(skb, &secattr) != 0)
+		goto netlbl_cache_add_failure;
+
+	return;
+
+netlbl_cache_add_failure:
+	netlbl_secattr_destroy(&secattr, 1);
+}
+
+/**
+ * selinux_netlbl_cache_invalidate - Invalidate the NetLabel cache
+ *
+ * Description:
+ * Invalidate the NetLabel security attribute mapping cache.
+ *
+ */
+void selinux_netlbl_cache_invalidate(void)
+{
+	netlbl_cache_invalidate();
+}
+
+/**
+ * selinux_netlbl_secattr_to_sid - Convert a NetLabel secattr to a SELinux SID
+ * @skb: the network packet
+ * @secattr: the NetLabel packet security attributes
+ * @base_sid: the SELinux SID to use as a context for MLS only attributes
+ * @sid: the SELinux SID
+ *
+ * Description:
+ * Convert the given NetLabel packet security attributes in @secattr into a
+ * SELinux SID.  If the @secattr field does not contain a full SELinux
+ * SID/context then use the context in @base_sid as the foundation.  If @skb
+ * is not NULL attempt to cache as much data as possibile.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb,
+					 struct netlbl_lsm_secattr *secattr,
+					 u32 base_sid,
+					 u32 *sid)
+{
+	int rc = -EIDRM;
+	struct context *ctx;
+	struct context ctx_new;
+	struct netlbl_cache *cache;
+
+	POLICY_RDLOCK;
+
+	if (secattr->cache.data) {
+		cache = NETLBL_CACHE(secattr->cache.data);
+		switch (cache->type) {
+		case NETLBL_CACHE_T_SID:
+			*sid = cache->data.sid;
+			rc = 0;
+			break;
+		case NETLBL_CACHE_T_MLS:
+			ctx = sidtab_search(&sidtab, base_sid);
+			if (ctx == NULL)
+				goto netlbl_secattr_to_sid_return;
+
+			ctx_new.user = ctx->user;
+			ctx_new.role = ctx->role;
+			ctx_new.type = ctx->type;
+			ctx_new.range.level[0].sens =
+				cache->data.mls_label.level[0].sens;
+			ctx_new.range.level[0].cat.highbit =
+				cache->data.mls_label.level[0].cat.highbit;
+			ctx_new.range.level[0].cat.node =
+				cache->data.mls_label.level[0].cat.node;
+			ctx_new.range.level[1].sens =
+				cache->data.mls_label.level[1].sens;
+			ctx_new.range.level[1].cat.highbit =
+				cache->data.mls_label.level[1].cat.highbit;
+			ctx_new.range.level[1].cat.node =
+				cache->data.mls_label.level[1].cat.node;
+
+			rc = sidtab_context_to_sid(&sidtab, &ctx_new, sid);
+			break;
+		default:
+			goto netlbl_secattr_to_sid_return;
+		}
+	} else if (secattr->mls_lvl_vld) {
+		ctx = sidtab_search(&sidtab, base_sid);
+		if (ctx == NULL)
+			goto netlbl_secattr_to_sid_return;
+
+		ctx_new.user = ctx->user;
+		ctx_new.role = ctx->role;
+		ctx_new.type = ctx->type;
+		mls_import_lvl(&ctx_new, secattr->mls_lvl, secattr->mls_lvl);
+		if (secattr->mls_cat) {
+			if (mls_import_cat(&ctx_new,
+					   secattr->mls_cat,
+					   secattr->mls_cat_len,
+					   NULL,
+					   0) != 0)
+				goto netlbl_secattr_to_sid_return;
+			ctx_new.range.level[1].cat.highbit =
+				ctx_new.range.level[0].cat.highbit;
+			ctx_new.range.level[1].cat.node =
+				ctx_new.range.level[0].cat.node;
+		} else {
+			ebitmap_init(&ctx_new.range.level[0].cat);
+			ebitmap_init(&ctx_new.range.level[1].cat);
+		}
+		if (mls_context_isvalid(&policydb, &ctx_new) != 1)
+			goto netlbl_secattr_to_sid_return_cleanup;
+
+		rc = sidtab_context_to_sid(&sidtab, &ctx_new, sid);
+		if (rc != 0)
+			goto netlbl_secattr_to_sid_return_cleanup;
+
+		if (skb != NULL)
+			selinux_netlbl_cache_add(skb, &ctx_new);
+		ebitmap_destroy(&ctx_new.range.level[0].cat);
+	} else {
+		*sid = SECINITSID_UNLABELED;
+		rc = 0;
+	}
+
+netlbl_secattr_to_sid_return:
+	POLICY_RDUNLOCK;
+	return rc;
+netlbl_secattr_to_sid_return_cleanup:
+	ebitmap_destroy(&ctx_new.range.level[0].cat);
+	goto netlbl_secattr_to_sid_return;
+}
+
+/**
+ * selinux_netlbl_skbuff_getsid - Get the sid of a packet using NetLabel
+ * @skb: the packet
+ * @base_sid: the SELinux SID to use as a context for MLS only attributes
+ * @sid: the SID
+ *
+ * Description:
+ * Call the NetLabel mechanism to get the security attributes of the given
+ * packet and use those attributes to determine the correct context/SID to
+ * assign to the packet.  Returns zero on success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
+					u32 base_sid,
+					u32 *sid)
+{
+	int rc;
+	struct netlbl_lsm_secattr secattr;
+
+	netlbl_secattr_init(&secattr);
+	rc = netlbl_skbuff_getattr(skb, &secattr);
+	if (rc == 0)
+		rc = selinux_netlbl_secattr_to_sid(skb,
+						   &secattr,
+						   base_sid,
+						   sid);
+	netlbl_secattr_destroy(&secattr, 0);
+
+	return rc;
+}
+
+/**
+ * selinux_netlbl_socket_setsid - Label a socket using the NetLabel mechanism
+ * @sock: the socket to label
+ * @sid: the SID to use
+ *
+ * Description:
+ * Attempt to label a socket using the NetLabel mechanism using the given
+ * SID.  Returns zero values on success, negative values on failure.
+ *
+ */
+static int selinux_netlbl_socket_setsid(struct socket *sock, u32 sid)
+{
+	int rc = -ENOENT;
+	struct sk_security_struct *sksec = sock->sk->sk_security;
+	struct netlbl_lsm_secattr secattr;
+	struct context *ctx;
+
+	if (!ss_initialized)
+		return 0;
+
+	POLICY_RDLOCK;
+
+	ctx = sidtab_search(&sidtab, sid);
+	if (ctx == NULL)
+		goto netlbl_socket_setsid_return;
+
+	netlbl_secattr_init(&secattr);
+	secattr.domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1],
+				 GFP_ATOMIC);
+	mls_export_lvl(ctx, &secattr.mls_lvl, NULL);
+	secattr.mls_lvl_vld = 1;
+	mls_export_cat(ctx,
+		       &secattr.mls_cat,
+		       &secattr.mls_cat_len,
+		       NULL,
+		       NULL);
+
+	rc = netlbl_socket_setattr(sock, &secattr);
+	if (rc == 0)
+		sksec->nlbl_state = NLBL_LABELED;
+
+	netlbl_secattr_destroy(&secattr, 0);
+
+netlbl_socket_setsid_return:
+	POLICY_RDUNLOCK;
+	return rc;
+}
+
+/**
+ * selinux_netlbl_socket_post_create - Label a socket using NetLabel
+ * @sock: the socket to label
+ * @sock_family: the socket family
+ * @sid: the SID to use
+ *
+ * Description:
+ * Attempt to label a socket using the NetLabel mechanism using the given
+ * SID.  Returns zero values on success, negative values on failure.
+ *
+ */
+int selinux_netlbl_socket_post_create(struct socket *sock,
+				      int sock_family,
+				      u32 sid)
+{
+	struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
+	struct sk_security_struct *sksec = sock->sk->sk_security;
+
+	if (sock_family != PF_INET)
+		return 0;
+
+	sksec->sclass = isec->sclass;
+	sksec->nlbl_state = NLBL_REQUIRE;
+	return selinux_netlbl_socket_setsid(sock, sid);
+}
+
+/**
+ * selinux_netlbl_sock_graft - Netlabel the new socket
+ * @sk: the new connection
+ * @sock: the new socket
+ *
+ * Description:
+ * The connection represented by @sk is being grafted onto @sock so set the
+ * socket's NetLabel to match the SID of @sk.
+ *
+ */
+void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock)
+{
+	struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
+	struct sk_security_struct *sksec = sk->sk_security;
+
+	if (sk->sk_family != PF_INET)
+		return;
+
+	sksec->nlbl_state = NLBL_REQUIRE;
+	sksec->peer_sid = sksec->sid;
+	sksec->sclass = isec->sclass;
+
+	/* Try to set the NetLabel on the socket to save time later, if we fail
+	 * here we will pick up the pieces in later calls to
+	 * selinux_netlbl_inode_permission(). */
+	selinux_netlbl_socket_setsid(sock, sksec->sid);
+}
+
+/**
+ * selinux_netlbl_inet_conn_request - Handle a new connection request
+ * @skb: the packet
+ * @sock_sid: the SID of the parent socket
+ *
+ * Description:
+ * If present, use the security attributes of the packet in @skb and the
+ * parent sock's SID to arrive at a SID for the new child sock.  Returns the
+ * SID of the connection or SECSID_NULL on failure.
+ *
+ */
+u32 selinux_netlbl_inet_conn_request(struct sk_buff *skb, u32 sock_sid)
+{
+	int rc;
+	u32 peer_sid;
+
+	rc = selinux_netlbl_skbuff_getsid(skb, sock_sid, &peer_sid);
+	if (rc != 0)
+		return SECSID_NULL;
+
+	if (peer_sid == SECINITSID_UNLABELED)
+		return SECSID_NULL;
+
+	return peer_sid;
+}
+
+/**
+ * __selinux_netlbl_inode_permission - Label a socket using NetLabel
+ * @inode: the file descriptor's inode
+ * @mask: the permission mask
+ *
+ * Description:
+ * Try to label a socket with the inode's SID using NetLabel.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+int __selinux_netlbl_inode_permission(struct inode *inode, int mask)
+{
+	int rc;
+	struct socket *sock = SOCKET_I(inode);
+	struct sk_security_struct *sksec = sock->sk->sk_security;
+
+	lock_sock(sock->sk);
+	rc = selinux_netlbl_socket_setsid(sock, sksec->sid);
+	release_sock(sock->sk);
+
+	return rc;
+}
+
+/**
+ * selinux_netlbl_sock_rcv_skb - Do an inbound access check using NetLabel
+ * @sksec: the sock's sk_security_struct
+ * @skb: the packet
+ * @ad: the audit data
+ *
+ * Description:
+ * Fetch the NetLabel security attributes from @skb and perform an access check
+ * against the receiving socket.  Returns zero on success, negative values on
+ * error.
+ *
+ */
+int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
+				struct sk_buff *skb,
+				struct avc_audit_data *ad)
+{
+	int rc;
+	u32 netlbl_sid;
+	u32 recv_perm;
+
+	rc = selinux_netlbl_skbuff_getsid(skb, sksec->sid, &netlbl_sid);
+	if (rc != 0)
+		return rc;
+
+	if (netlbl_sid == SECINITSID_UNLABELED)
+		return 0;
+
+	switch (sksec->sclass) {
+	case SECCLASS_UDP_SOCKET:
+		recv_perm = UDP_SOCKET__RECV_MSG;
+		break;
+	case SECCLASS_TCP_SOCKET:
+		recv_perm = TCP_SOCKET__RECV_MSG;
+		break;
+	default:
+		recv_perm = RAWIP_SOCKET__RECV_MSG;
+	}
+
+	rc = avc_has_perm(sksec->sid,
+			  netlbl_sid,
+			  sksec->sclass,
+			  recv_perm,
+			  ad);
+	if (rc == 0)
+		return 0;
+
+	netlbl_skbuff_err(skb, rc);
+	return rc;
+}
+
+/**
+ * selinux_netlbl_socket_peersid - Return the peer SID of a connected socket
+ * @sock: the socket
+ *
+ * Description:
+ * Examine @sock to find the connected peer's SID.  Returns the SID on success
+ * or SECSID_NULL on error.
+ *
+ */
+u32 selinux_netlbl_socket_getpeersec_stream(struct socket *sock)
+{
+	struct sk_security_struct *sksec = sock->sk->sk_security;
+
+	if (sksec->peer_sid == SECINITSID_UNLABELED)
+		return SECSID_NULL;
+
+	return sksec->peer_sid;
+}
+
+/**
+ * selinux_netlbl_socket_getpeersec_dgram - Return the SID of a NetLabel packet
+ * @skb: the packet
+ *
+ * Description:
+ * Examine @skb to find the SID assigned to it by NetLabel.  Returns the SID on
+ * success, SECSID_NULL on error.
+ *
+ */
+u32 selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb)
+{
+	int peer_sid;
+	struct sock *sk = skb->sk;
+	struct inode_security_struct *isec;
+
+	if (sk == NULL || sk->sk_socket == NULL)
+		return SECSID_NULL;
+
+	isec = SOCK_INODE(sk->sk_socket)->i_security;
+	if (selinux_netlbl_skbuff_getsid(skb, isec->sid, &peer_sid) != 0)
+		return SECSID_NULL;
+	if (peer_sid == SECINITSID_UNLABELED)
+		return SECSID_NULL;
+
+	return peer_sid;
+}
+#endif /* CONFIG_NETLABEL */
-- 
cgit v1.2.3


From c71099acce933455123ee505cc75964610a209ad Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 23:20:06 -0700
Subject: [IPV6]: Multiple Routing Tables

Adds the framework to support multiple IPv6 routing tables.
Currently all automatically generated routes are put into the
same table. This could be changed at a later point after
considering the produced locking overhead.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h   |  39 ++++-
 include/net/ip6_route.h |   3 +-
 net/ipv6/Kconfig        |   6 +
 net/ipv6/addrconf.c     |   6 +-
 net/ipv6/ip6_fib.c      | 144 +++++++++++++++++-
 net/ipv6/route.c        | 380 ++++++++++++++++++++++++++++++++----------------
 6 files changed, 441 insertions(+), 137 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index a66e9de16a6c..818411519c89 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -51,6 +51,8 @@ struct rt6key
 	int		plen;
 };
 
+struct fib6_table;
+
 struct rt6_info
 {
 	union {
@@ -71,6 +73,7 @@ struct rt6_info
 	u32				rt6i_flags;
 	u32				rt6i_metric;
 	atomic_t			rt6i_ref;
+	struct fib6_table		*rt6i_table;
 
 	struct rt6key			rt6i_dst;
 	struct rt6key			rt6i_src;
@@ -143,12 +146,43 @@ struct rt6_statistics {
 
 typedef void			(*f_pnode)(struct fib6_node *fn, void *);
 
-extern struct fib6_node		ip6_routing_table;
+struct fib6_table {
+	struct hlist_node	tb6_hlist;
+	u32			tb6_id;
+	rwlock_t		tb6_lock;
+	struct fib6_node	tb6_root;
+};
+
+#define RT6_TABLE_UNSPEC	RT_TABLE_UNSPEC
+#define RT6_TABLE_MAIN		RT_TABLE_MAIN
+#define RT6_TABLE_LOCAL		RT6_TABLE_MAIN
+#define RT6_TABLE_DFLT		RT6_TABLE_MAIN
+#define RT6_TABLE_INFO		RT6_TABLE_MAIN
+#define RT6_TABLE_PREFIX	RT6_TABLE_MAIN
+
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+#define FIB6_TABLE_MIN		1
+#define FIB6_TABLE_MAX		RT_TABLE_MAX
+#else
+#define FIB6_TABLE_MIN		RT_TABLE_MAIN
+#define FIB6_TABLE_MAX		FIB6_TABLE_MIN
+#endif
+
+#define RT6_F_STRICT		1
+#define RT6_F_HAS_SADDR		2
+
+typedef struct rt6_info *(*pol_lookup_t)(struct fib6_table *,
+					 struct flowi *, int);
 
 /*
  *	exported functions
  */
 
+extern struct fib6_table *	fib6_get_table(u32 id);
+extern struct fib6_table *	fib6_new_table(u32 id);
+extern struct dst_entry *	fib6_rule_lookup(struct flowi *fl, int flags,
+						 pol_lookup_t lookup);
+
 extern struct fib6_node		*fib6_lookup(struct fib6_node *root,
 					     struct in6_addr *daddr,
 					     struct in6_addr *saddr);
@@ -161,6 +195,9 @@ extern void			fib6_clean_tree(struct fib6_node *root,
 						int (*func)(struct rt6_info *, void *arg),
 						int prune, void *arg);
 
+extern void			fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
+					       int prune, void *arg);
+
 extern int			fib6_walk(struct fib6_walker_t *w);
 extern int			fib6_walk_continue(struct fib6_walker_t *w);
 
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 96b0e66406ec..d49c8c90eb68 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -58,7 +58,8 @@ extern int			ipv6_route_ioctl(unsigned int cmd, void __user *arg);
 extern int			ip6_route_add(struct in6_rtmsg *rtmsg,
 					      struct nlmsghdr *,
 					      void *rtattr,
-					      struct netlink_skb_parms *req);
+					      struct netlink_skb_parms *req,
+					      u32 table_id);
 extern int			ip6_ins_rt(struct rt6_info *,
 					   struct nlmsghdr *,
 					   void *rtattr,
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 0ba06c0c5d39..159c63d99c81 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -136,3 +136,9 @@ config IPV6_TUNNEL
 
 	  If unsure, say N.
 
+config IPV6_MULTIPLE_TABLES
+	bool "IPv6: Multiple Routing Tables"
+	depends on IPV6 && EXPERIMENTAL
+	---help---
+	  Support multiple routing tables.
+
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index c7852b38e03e..318767fcefdc 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1525,7 +1525,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
 	if (dev->type == ARPHRD_SIT && (dev->flags&IFF_POINTOPOINT))
 		rtmsg.rtmsg_flags |= RTF_NONEXTHOP;
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_PREFIX);
 }
 
 /* Create "default" multicast route to the interface */
@@ -1542,7 +1542,7 @@ static void addrconf_add_mroute(struct net_device *dev)
 	rtmsg.rtmsg_ifindex = dev->ifindex;
 	rtmsg.rtmsg_flags = RTF_UP;
 	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
-	ip6_route_add(&rtmsg, NULL, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_LOCAL);
 }
 
 static void sit_route_add(struct net_device *dev)
@@ -1559,7 +1559,7 @@ static void sit_route_add(struct net_device *dev)
 	rtmsg.rtmsg_flags	= RTF_UP|RTF_NONEXTHOP;
 	rtmsg.rtmsg_ifindex	= dev->ifindex;
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_MAIN);
 }
 
 static void addrconf_add_lroute(struct net_device *dev)
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 764221220afd..fcd7da830aca 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -26,6 +26,7 @@
 #include <linux/netdevice.h>
 #include <linux/in6.h>
 #include <linux/init.h>
+#include <linux/list.h>
 
 #ifdef 	CONFIG_PROC_FS
 #include <linux/proc_fs.h>
@@ -147,6 +148,126 @@ static __inline__ void rt6_release(struct rt6_info *rt)
 		dst_free(&rt->u.dst);
 }
 
+static struct fib6_table fib6_main_tbl = {
+	.tb6_id		= RT6_TABLE_MAIN,
+	.tb6_lock	= RW_LOCK_UNLOCKED,
+	.tb6_root	= {
+		.leaf		= &ip6_null_entry,
+		.fn_flags	= RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO,
+	},
+};
+
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+
+#define FIB_TABLE_HASHSZ 256
+static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
+
+static struct fib6_table *fib6_alloc_table(u32 id)
+{
+	struct fib6_table *table;
+
+	table = kzalloc(sizeof(*table), GFP_ATOMIC);
+	if (table != NULL) {
+		table->tb6_id = id;
+		table->tb6_lock = RW_LOCK_UNLOCKED;
+		table->tb6_root.leaf = &ip6_null_entry;
+		table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO;
+	}
+
+	return table;
+}
+
+static void fib6_link_table(struct fib6_table *tb)
+{
+	unsigned int h;
+
+	h = tb->tb6_id & (FIB_TABLE_HASHSZ - 1);
+
+	/*
+	 * No protection necessary, this is the only list mutatation
+	 * operation, tables never disappear once they exist.
+	 */
+	hlist_add_head_rcu(&tb->tb6_hlist, &fib_table_hash[h]);
+}
+
+struct fib6_table *fib6_new_table(u32 id)
+{
+	struct fib6_table *tb;
+
+	if (id == 0)
+		id = RT6_TABLE_MAIN;
+	tb = fib6_get_table(id);
+	if (tb)
+		return tb;
+
+	tb = fib6_alloc_table(id);
+	if (tb != NULL)
+		fib6_link_table(tb);
+
+	return tb;
+}
+
+struct fib6_table *fib6_get_table(u32 id)
+{
+	struct fib6_table *tb;
+	struct hlist_node *node;
+	unsigned int h;
+
+	if (id == 0)
+		id = RT6_TABLE_MAIN;
+	h = id & (FIB_TABLE_HASHSZ - 1);
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(tb, node, &fib_table_hash[h], tb6_hlist) {
+		if (tb->tb6_id == id) {
+			rcu_read_unlock();
+			return tb;
+		}
+	}
+	rcu_read_unlock();
+
+	return NULL;
+}
+
+struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
+				   pol_lookup_t lookup)
+{
+	/*
+	 * TODO: Add rule lookup
+	 */
+	struct fib6_table *table = fib6_get_table(RT6_TABLE_MAIN);
+
+	return (struct dst_entry *) lookup(table, fl, flags);
+}
+
+static void __init fib6_tables_init(void)
+{
+	fib6_link_table(&fib6_main_tbl);
+}
+
+#else
+
+struct fib6_table *fib6_new_table(u32 id)
+{
+	return fib6_get_table(id);
+}
+
+struct fib6_table *fib6_get_table(u32 id)
+{
+	return &fib6_main_tbl;
+}
+
+struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
+				   pol_lookup_t lookup)
+{
+	return (struct dst_entry *) lookup(&fib6_main_tbl, fl, flags);
+}
+
+static void __init fib6_tables_init(void)
+{
+}
+
+#endif
+
 
 /*
  *	Routing Table
@@ -1064,6 +1185,22 @@ void fib6_clean_tree(struct fib6_node *root,
 	fib6_walk(&c.w);
 }
 
+void fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
+		    int prune, void *arg)
+{
+	int i;
+	struct fib6_table *table;
+
+	for (i = FIB6_TABLE_MIN; i <= FIB6_TABLE_MAX; i++) {
+		table = fib6_get_table(i);
+		if (table != NULL) {
+			write_lock_bh(&table->tb6_lock);
+			fib6_clean_tree(&table->tb6_root, func, prune, arg);
+			write_unlock_bh(&table->tb6_lock);
+		}
+	}
+}
+
 static int fib6_prune_clone(struct rt6_info *rt, void *arg)
 {
 	if (rt->rt6i_flags & RTF_CACHE) {
@@ -1142,11 +1279,8 @@ void fib6_run_gc(unsigned long dummy)
 	}
 	gc_args.more = 0;
 
-
-	write_lock_bh(&rt6_lock);
 	ndisc_dst_gc(&gc_args.more);
-	fib6_clean_tree(&ip6_routing_table, fib6_age, 0, NULL);
-	write_unlock_bh(&rt6_lock);
+	fib6_clean_all(fib6_age, 0, NULL);
 
 	if (gc_args.more)
 		mod_timer(&ip6_fib_timer, jiffies + ip6_rt_gc_interval);
@@ -1165,6 +1299,8 @@ void __init fib6_init(void)
 					   NULL, NULL);
 	if (!fib6_node_kmem)
 		panic("cannot create fib6_nodes cache");
+
+	fib6_tables_init();
 }
 
 void fib6_gc_cleanup(void)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ce1f49b595b0..73efdadb9ab8 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -140,16 +140,6 @@ struct rt6_info ip6_null_entry = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
-struct fib6_node ip6_routing_table = {
-	.leaf		= &ip6_null_entry,
-	.fn_flags	= RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO,
-};
-
-/* Protects all the ip6 fib */
-
-DEFINE_RWLOCK(rt6_lock);
-
-
 /* allocate dst with ip6_dst_ops */
 static __inline__ struct rt6_info *ip6_dst_alloc(void)
 {
@@ -188,8 +178,14 @@ static __inline__ int rt6_check_expired(const struct rt6_info *rt)
 		time_after(jiffies, rt->rt6i_expires));
 }
 
+static inline int rt6_need_strict(struct in6_addr *daddr)
+{
+	return (ipv6_addr_type(daddr) &
+		(IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL));
+}
+
 /*
- *	Route lookup. Any rt6_lock is implied.
+ *	Route lookup. Any table->tb6_lock is implied.
  */
 
 static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt,
@@ -441,27 +437,66 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 }
 #endif
 
-struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
-			    int oif, int strict)
+#define BACKTRACK() \
+if (rt == &ip6_null_entry && flags & RT6_F_STRICT) { \
+	while ((fn = fn->parent) != NULL) { \
+		if (fn->fn_flags & RTN_TL_ROOT) { \
+			dst_hold(&rt->u.dst); \
+			goto out; \
+		} \
+		if (fn->fn_flags & RTN_RTINFO) \
+			goto restart; \
+	} \
+}
+
+static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table,
+					     struct flowi *fl, int flags)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt;
 
-	read_lock_bh(&rt6_lock);
-	fn = fib6_lookup(&ip6_routing_table, daddr, saddr);
-	rt = rt6_device_match(fn->leaf, oif, strict);
+	read_lock_bh(&table->tb6_lock);
+	fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src);
+restart:
+	rt = fn->leaf;
+	rt = rt6_device_match(rt, fl->oif, flags & RT6_F_STRICT);
+	BACKTRACK();
 	dst_hold(&rt->u.dst);
-	rt->u.dst.__use++;
-	read_unlock_bh(&rt6_lock);
+out:
+	read_unlock_bh(&table->tb6_lock);
 
 	rt->u.dst.lastuse = jiffies;
-	if (rt->u.dst.error == 0)
-		return rt;
-	dst_release(&rt->u.dst);
+	rt->u.dst.__use++;
+
+	return rt;
+
+}
+
+struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
+			    int oif, int strict)
+{
+	struct flowi fl = {
+		.oif = oif,
+		.nl_u = {
+			.ip6_u = {
+				.daddr = *daddr,
+				/* TODO: saddr */
+			},
+		},
+	};
+	struct dst_entry *dst;
+	int flags = strict ? RT6_F_STRICT : 0;
+
+	dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup);
+	if (dst->error == 0)
+		return (struct rt6_info *) dst;
+
+	dst_release(dst);
+
 	return NULL;
 }
 
-/* ip6_ins_rt is called with FREE rt6_lock.
+/* ip6_ins_rt is called with FREE table->tb6_lock.
    It takes new route entry, the addition fails by any reason the
    route is freed. In any case, if caller does not hold it, it may
    be destroyed.
@@ -471,10 +506,12 @@ int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
 		void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
+	struct fib6_table *table;
 
-	write_lock_bh(&rt6_lock);
-	err = fib6_add(&ip6_routing_table, rt, nlh, _rtattr, req);
-	write_unlock_bh(&rt6_lock);
+	table = rt->rt6i_table;
+	write_lock_bh(&table->tb6_lock);
+	err = fib6_add(&table->tb6_root, rt, nlh, _rtattr, req);
+	write_unlock_bh(&table->tb6_lock);
 
 	return err;
 }
@@ -532,51 +569,40 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d
 	return rt;
 }
 
-#define BACKTRACK() \
-if (rt == &ip6_null_entry) { \
-       while ((fn = fn->parent) != NULL) { \
-		if (fn->fn_flags & RTN_ROOT) { \
-			goto out; \
-		} \
-		if (fn->fn_flags & RTN_RTINFO) \
-			goto restart; \
-	} \
-}
-
-
-void ip6_route_input(struct sk_buff *skb)
+struct rt6_info *ip6_pol_route_input(struct fib6_table *table, struct flowi *fl,
+				     int flags)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt, *nrt;
-	int strict;
+	int strict = 0;
 	int attempts = 3;
 	int err;
 	int reachable = RT6_SELECT_F_REACHABLE;
 
-	strict = ipv6_addr_type(&skb->nh.ipv6h->daddr) & (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL) ? RT6_SELECT_F_IFACE : 0;
+	if (flags & RT6_F_STRICT)
+		strict = RT6_SELECT_F_IFACE;
 
 relookup:
-	read_lock_bh(&rt6_lock);
+	read_lock_bh(&table->tb6_lock);
 
 restart_2:
-	fn = fib6_lookup(&ip6_routing_table, &skb->nh.ipv6h->daddr,
-			 &skb->nh.ipv6h->saddr);
+	fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src);
 
 restart:
-	rt = rt6_select(&fn->leaf, skb->dev->ifindex, strict | reachable);
+	rt = rt6_select(&fn->leaf, fl->iif, strict | reachable);
 	BACKTRACK();
 	if (rt == &ip6_null_entry ||
 	    rt->rt6i_flags & RTF_CACHE)
 		goto out;
 
 	dst_hold(&rt->u.dst);
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 
 	if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP))
-		nrt = rt6_alloc_cow(rt, &skb->nh.ipv6h->daddr, &skb->nh.ipv6h->saddr);
+		nrt = rt6_alloc_cow(rt, &fl->fl6_dst, &fl->fl6_src);
 	else {
 #if CLONE_OFFLINK_ROUTE
-		nrt = rt6_alloc_clone(rt, &skb->nh.ipv6h->daddr);
+		nrt = rt6_alloc_clone(rt, &fl->fl6_dst);
 #else
 		goto out2;
 #endif
@@ -587,7 +613,7 @@ restart:
 
 	dst_hold(&rt->u.dst);
 	if (nrt) {
-		err = ip6_ins_rt(nrt, NULL, NULL, &NETLINK_CB(skb));
+		err = ip6_ins_rt(nrt, NULL, NULL, NULL);
 		if (!err)
 			goto out2;
 	}
@@ -596,7 +622,7 @@ restart:
 		goto out2;
 
 	/*
-	 * Race condition! In the gap, when rt6_lock was
+	 * Race condition! In the gap, when table->tb6_lock was
 	 * released someone could insert this route.  Relookup.
 	 */
 	dst_release(&rt->u.dst);
@@ -608,30 +634,54 @@ out:
 		goto restart_2;
 	}
 	dst_hold(&rt->u.dst);
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 out2:
 	rt->u.dst.lastuse = jiffies;
 	rt->u.dst.__use++;
-	skb->dst = (struct dst_entry *) rt;
-	return;
+
+	return rt;
 }
 
-struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
+void ip6_route_input(struct sk_buff *skb)
+{
+	struct ipv6hdr *iph = skb->nh.ipv6h;
+	struct flowi fl = {
+		.iif = skb->dev->ifindex,
+		.nl_u = {
+			.ip6_u = {
+				.daddr = iph->daddr,
+				.saddr = iph->saddr,
+				.flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK,
+			},
+		},
+		.proto = iph->nexthdr,
+	};
+	int flags = 0;
+
+	if (rt6_need_strict(&iph->daddr))
+		flags |= RT6_F_STRICT;
+
+	skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input);
+}
+
+static struct rt6_info *ip6_pol_route_output(struct fib6_table *table,
+					     struct flowi *fl, int flags)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt, *nrt;
-	int strict;
+	int strict = 0;
 	int attempts = 3;
 	int err;
 	int reachable = RT6_SELECT_F_REACHABLE;
 
-	strict = ipv6_addr_type(&fl->fl6_dst) & (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL) ? RT6_SELECT_F_IFACE : 0;
+	if (flags & RT6_F_STRICT)
+		strict = RT6_SELECT_F_IFACE;
 
 relookup:
-	read_lock_bh(&rt6_lock);
+	read_lock_bh(&table->tb6_lock);
 
 restart_2:
-	fn = fib6_lookup(&ip6_routing_table, &fl->fl6_dst, &fl->fl6_src);
+	fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src);
 
 restart:
 	rt = rt6_select(&fn->leaf, fl->oif, strict | reachable);
@@ -641,7 +691,7 @@ restart:
 		goto out;
 
 	dst_hold(&rt->u.dst);
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 
 	if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP))
 		nrt = rt6_alloc_cow(rt, &fl->fl6_dst, &fl->fl6_src);
@@ -667,7 +717,7 @@ restart:
 		goto out2;
 
 	/*
-	 * Race condition! In the gap, when rt6_lock was
+	 * Race condition! In the gap, when table->tb6_lock was
 	 * released someone could insert this route.  Relookup.
 	 */
 	dst_release(&rt->u.dst);
@@ -679,11 +729,21 @@ out:
 		goto restart_2;
 	}
 	dst_hold(&rt->u.dst);
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 out2:
 	rt->u.dst.lastuse = jiffies;
 	rt->u.dst.__use++;
-	return &rt->u.dst;
+	return rt;
+}
+
+struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
+{
+	int flags = 0;
+
+	if (rt6_need_strict(&fl->fl6_dst))
+		flags |= RT6_F_STRICT;
+
+	return fib6_rule_lookup(fl, flags, ip6_pol_route_output);
 }
 
 
@@ -906,7 +966,8 @@ int ipv6_get_hoplimit(struct net_device *dev)
  */
 
 int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, 
-		void *_rtattr, struct netlink_skb_parms *req)
+		  void *_rtattr, struct netlink_skb_parms *req,
+		  u32 table_id)
 {
 	int err;
 	struct rtmsg *r;
@@ -914,6 +975,7 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 	struct rt6_info *rt = NULL;
 	struct net_device *dev = NULL;
 	struct inet6_dev *idev = NULL;
+	struct fib6_table *table;
 	int addr_type;
 
 	rta = (struct rtattr **) _rtattr;
@@ -937,6 +999,12 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 	if (rtmsg->rtmsg_metric == 0)
 		rtmsg->rtmsg_metric = IP6_RT_PRIO_USER;
 
+	table = fib6_new_table(table_id);
+	if (table == NULL) {
+		err = -ENOBUFS;
+		goto out;
+	}
+
 	rt = ip6_dst_alloc();
 
 	if (rt == NULL) {
@@ -1093,6 +1161,7 @@ install_route:
 		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst));
 	rt->u.dst.dev = dev;
 	rt->rt6i_idev = idev;
+	rt->rt6i_table = table;
 	return ip6_ins_rt(rt, nlh, _rtattr, req);
 
 out:
@@ -1108,26 +1177,35 @@ out:
 int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
+	struct fib6_table *table;
 
-	write_lock_bh(&rt6_lock);
+	table = rt->rt6i_table;
+	write_lock_bh(&table->tb6_lock);
 
 	err = fib6_del(rt, nlh, _rtattr, req);
 	dst_release(&rt->u.dst);
 
-	write_unlock_bh(&rt6_lock);
+	write_unlock_bh(&table->tb6_lock);
 
 	return err;
 }
 
-static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
+static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
+			 void *_rtattr, struct netlink_skb_parms *req,
+			 u32 table_id)
 {
+	struct fib6_table *table;
 	struct fib6_node *fn;
 	struct rt6_info *rt;
 	int err = -ESRCH;
 
-	read_lock_bh(&rt6_lock);
+	table = fib6_get_table(table_id);
+	if (table == NULL)
+		return err;
+
+	read_lock_bh(&table->tb6_lock);
 
-	fn = fib6_locate(&ip6_routing_table,
+	fn = fib6_locate(&table->tb6_root,
 			 &rtmsg->rtmsg_dst, rtmsg->rtmsg_dst_len,
 			 &rtmsg->rtmsg_src, rtmsg->rtmsg_src_len);
 	
@@ -1144,12 +1222,12 @@ static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_r
 			    rtmsg->rtmsg_metric != rt->rt6i_metric)
 				continue;
 			dst_hold(&rt->u.dst);
-			read_unlock_bh(&rt6_lock);
+			read_unlock_bh(&table->tb6_lock);
 
 			return ip6_del_rt(rt, nlh, _rtattr, req);
 		}
 	}
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 
 	return err;
 }
@@ -1161,10 +1239,15 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr,
 		  struct neighbour *neigh, u8 *lladdr, int on_link)
 {
 	struct rt6_info *rt, *nrt = NULL;
-	int strict;
 	struct fib6_node *fn;
+	struct fib6_table *table;
 	struct netevent_redirect netevent;
 
+	/* TODO: Very lazy, might need to check all tables */
+	table = fib6_get_table(RT6_TABLE_MAIN);
+	if (table == NULL)
+		return;
+
 	/*
 	 * Get the "current" route for this destination and
 	 * check if the redirect has come from approriate router.
@@ -1175,10 +1258,9 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr,
 	 * is a bit fuzzy and one might need to check all possible
 	 * routes.
 	 */
-	strict = ipv6_addr_type(dest) & (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL);
 
-	read_lock_bh(&rt6_lock);
-	fn = fib6_lookup(&ip6_routing_table, dest, NULL);
+	read_lock_bh(&table->tb6_lock);
+	fn = fib6_lookup(&table->tb6_root, dest, NULL);
 restart:
 	for (rt = fn->leaf; rt; rt = rt->u.next) {
 		/*
@@ -1201,7 +1283,7 @@ restart:
 	}
 	if (rt)
 		dst_hold(&rt->u.dst);
-	else if (strict) {
+	else if (rt6_need_strict(dest)) {
 		while ((fn = fn->parent) != NULL) {
 			if (fn->fn_flags & RTN_ROOT)
 				break;
@@ -1209,7 +1291,7 @@ restart:
 				goto restart;
 		}
 	}
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 
 	if (!rt) {
 		if (net_ratelimit())
@@ -1384,6 +1466,7 @@ static struct rt6_info * ip6_rt_copy(struct rt6_info *ort)
 #ifdef CONFIG_IPV6_SUBTREES
 		memcpy(&rt->rt6i_src, &ort->rt6i_src, sizeof(struct rt6key));
 #endif
+		rt->rt6i_table = ort->rt6i_table;
 	}
 	return rt;
 }
@@ -1394,9 +1477,14 @@ static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixle
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt = NULL;
+	struct fib6_table *table;
+
+	table = fib6_get_table(RT6_TABLE_INFO);
+	if (table == NULL)
+		return NULL;
 
-	write_lock_bh(&rt6_lock);
-	fn = fib6_locate(&ip6_routing_table, prefix ,prefixlen, NULL, 0);
+	write_lock_bh(&table->tb6_lock);
+	fn = fib6_locate(&table->tb6_root, prefix ,prefixlen, NULL, 0);
 	if (!fn)
 		goto out;
 
@@ -1411,7 +1499,7 @@ static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixle
 		break;
 	}
 out:
-	write_unlock_bh(&rt6_lock);
+	write_unlock_bh(&table->tb6_lock);
 	return rt;
 }
 
@@ -1433,7 +1521,7 @@ static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixle
 		rtmsg.rtmsg_flags |= RTF_DEFAULT;
 	rtmsg.rtmsg_ifindex = ifindex;
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_INFO);
 
 	return rt6_get_route_info(prefix, prefixlen, gwaddr, ifindex);
 }
@@ -1442,12 +1530,14 @@ static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixle
 struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *dev)
 {	
 	struct rt6_info *rt;
-	struct fib6_node *fn;
+	struct fib6_table *table;
 
-	fn = &ip6_routing_table;
+	table = fib6_get_table(RT6_TABLE_DFLT);
+	if (table == NULL)
+		return NULL;
 
-	write_lock_bh(&rt6_lock);
-	for (rt = fn->leaf; rt; rt=rt->u.next) {
+	write_lock_bh(&table->tb6_lock);
+	for (rt = table->tb6_root.leaf; rt; rt=rt->u.next) {
 		if (dev == rt->rt6i_dev &&
 		    ((rt->rt6i_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) &&
 		    ipv6_addr_equal(&rt->rt6i_gateway, addr))
@@ -1455,7 +1545,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 	}
 	if (rt)
 		dst_hold(&rt->u.dst);
-	write_unlock_bh(&rt6_lock);
+	write_unlock_bh(&table->tb6_lock);
 	return rt;
 }
 
@@ -1474,28 +1564,31 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 
 	rtmsg.rtmsg_ifindex = dev->ifindex;
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL);
+	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_DFLT);
 	return rt6_get_dflt_router(gwaddr, dev);
 }
 
 void rt6_purge_dflt_routers(void)
 {
 	struct rt6_info *rt;
+	struct fib6_table *table;
+
+	/* NOTE: Keep consistent with rt6_get_dflt_router */
+	table = fib6_get_table(RT6_TABLE_DFLT);
+	if (table == NULL)
+		return;
 
 restart:
-	read_lock_bh(&rt6_lock);
-	for (rt = ip6_routing_table.leaf; rt; rt = rt->u.next) {
+	read_lock_bh(&table->tb6_lock);
+	for (rt = table->tb6_root.leaf; rt; rt = rt->u.next) {
 		if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) {
 			dst_hold(&rt->u.dst);
-
-			read_unlock_bh(&rt6_lock);
-
+			read_unlock_bh(&table->tb6_lock);
 			ip6_del_rt(rt, NULL, NULL, NULL);
-
 			goto restart;
 		}
 	}
-	read_unlock_bh(&rt6_lock);
+	read_unlock_bh(&table->tb6_lock);
 }
 
 int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
@@ -1516,10 +1609,12 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
 		rtnl_lock();
 		switch (cmd) {
 		case SIOCADDRT:
-			err = ip6_route_add(&rtmsg, NULL, NULL, NULL);
+			err = ip6_route_add(&rtmsg, NULL, NULL, NULL,
+					    RT6_TABLE_MAIN);
 			break;
 		case SIOCDELRT:
-			err = ip6_route_del(&rtmsg, NULL, NULL, NULL);
+			err = ip6_route_del(&rtmsg, NULL, NULL, NULL,
+					    RT6_TABLE_MAIN);
 			break;
 		default:
 			err = -EINVAL;
@@ -1593,6 +1688,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 
 	ipv6_addr_copy(&rt->rt6i_dst.addr, addr);
 	rt->rt6i_dst.plen = 128;
+	rt->rt6i_table = fib6_get_table(RT6_TABLE_LOCAL);
 
 	atomic_set(&rt->u.dst.__refcnt, 1);
 
@@ -1611,9 +1707,7 @@ static int fib6_ifdown(struct rt6_info *rt, void *arg)
 
 void rt6_ifdown(struct net_device *dev)
 {
-	write_lock_bh(&rt6_lock);
-	fib6_clean_tree(&ip6_routing_table, fib6_ifdown, 0, dev);
-	write_unlock_bh(&rt6_lock);
+	fib6_clean_all(fib6_ifdown, 0, dev);
 }
 
 struct rt6_mtu_change_arg
@@ -1663,13 +1757,12 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
 
 void rt6_mtu_change(struct net_device *dev, unsigned mtu)
 {
-	struct rt6_mtu_change_arg arg;
+	struct rt6_mtu_change_arg arg = {
+		.dev = dev,
+		.mtu = mtu,
+	};
 
-	arg.dev = dev;
-	arg.mtu = mtu;
-	read_lock_bh(&rt6_lock);
-	fib6_clean_tree(&ip6_routing_table, rt6_mtu_change_route, 0, &arg);
-	read_unlock_bh(&rt6_lock);
+	fib6_clean_all(rt6_mtu_change_route, 0, &arg);
 }
 
 static int inet6_rtm_to_rtmsg(struct rtmsg *r, struct rtattr **rta,
@@ -1719,7 +1812,7 @@ int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
 		return -EINVAL;
-	return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb));
+	return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table);
 }
 
 int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
@@ -1729,7 +1822,7 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
 		return -EINVAL;
-	return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb));
+	return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table);
 }
 
 struct rt6_rtnl_dump_arg
@@ -1761,6 +1854,10 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 	rtm->rtm_dst_len = rt->rt6i_dst.plen;
 	rtm->rtm_src_len = rt->rt6i_src.plen;
 	rtm->rtm_tos = 0;
+	if (rt->rt6i_table)
+		rtm->rtm_table = rt->rt6i_table->tb6_id;
+	else
+		rtm->rtm_table = RT6_TABLE_UNSPEC;
 	rtm->rtm_table = RT_TABLE_MAIN;
 	if (rt->rt6i_flags&RTF_REJECT)
 		rtm->rtm_type = RTN_UNREACHABLE;
@@ -1868,7 +1965,6 @@ static void fib6_dump_end(struct netlink_callback *cb)
 
 	if (w) {
 		cb->args[0] = 0;
-		fib6_walker_unlink(w);
 		kfree(w);
 	}
 	cb->done = (void*)cb->args[1];
@@ -1883,13 +1979,20 @@ static int fib6_dump_done(struct netlink_callback *cb)
 
 int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	struct fib6_table *table;
 	struct rt6_rtnl_dump_arg arg;
 	struct fib6_walker_t *w;
-	int res;
+	int i, res = 0;
 
 	arg.skb = skb;
 	arg.cb = cb;
 
+	/*
+	 * cb->args[0] = pointer to walker structure
+	 * cb->args[1] = saved cb->done() pointer
+	 * cb->args[2] = current table being dumped
+	 */
+
 	w = (void*)cb->args[0];
 	if (w == NULL) {
 		/* New dump:
@@ -1905,24 +2008,48 @@ int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 		w = kzalloc(sizeof(*w), GFP_ATOMIC);
 		if (w == NULL)
 			return -ENOMEM;
-		RT6_TRACE("dump<%p", w);
-		w->root = &ip6_routing_table;
 		w->func = fib6_dump_node;
 		w->args = &arg;
 		cb->args[0] = (long)w;
-		read_lock_bh(&rt6_lock);
-		res = fib6_walk(w);
-		read_unlock_bh(&rt6_lock);
+		cb->args[2] = FIB6_TABLE_MIN;
 	} else {
 		w->args = &arg;
-		read_lock_bh(&rt6_lock);
-		res = fib6_walk_continue(w);
-		read_unlock_bh(&rt6_lock);
+		i = cb->args[2];
+		if (i > FIB6_TABLE_MAX)
+			goto end;
+
+		table = fib6_get_table(i);
+		if (table != NULL) {
+			read_lock_bh(&table->tb6_lock);
+			w->root = &table->tb6_root;
+			res = fib6_walk_continue(w);
+			read_unlock_bh(&table->tb6_lock);
+			if (res != 0) {
+				if (res < 0)
+					fib6_walker_unlink(w);
+				goto end;
+			}
+		}
+
+		fib6_walker_unlink(w);
+		cb->args[2] = ++i;
 	}
-#if RT6_DEBUG >= 3
-	if (res <= 0 && skb->len == 0)
-		RT6_TRACE("%p>dump end\n", w);
-#endif
+
+	for (i = cb->args[2]; i <= FIB6_TABLE_MAX; i++) {
+		table = fib6_get_table(i);
+		if (table == NULL)
+			continue;
+
+		read_lock_bh(&table->tb6_lock);
+		w->root = &table->tb6_root;
+		res = fib6_walk(w);
+		read_unlock_bh(&table->tb6_lock);
+		if (res)
+			break;
+	}
+end:
+	cb->args[2] = i;
+
 	res = res < 0 ? res : skb->len;
 	/* res < 0 is an error. (really, impossible)
 	   res == 0 means that dump is complete, but skb still can contain data.
@@ -2102,16 +2229,13 @@ static int rt6_info_route(struct rt6_info *rt, void *p_arg)
 
 static int rt6_proc_info(char *buffer, char **start, off_t offset, int length)
 {
-	struct rt6_proc_arg arg;
-	arg.buffer = buffer;
-	arg.offset = offset;
-	arg.length = length;
-	arg.skip = 0;
-	arg.len = 0;
+	struct rt6_proc_arg arg = {
+		.buffer = buffer,
+		.offset = offset,
+		.length = length,
+	};
 
-	read_lock_bh(&rt6_lock);
-	fib6_clean_tree(&ip6_routing_table, rt6_info_route, 0, &arg);
-	read_unlock_bh(&rt6_lock);
+	fib6_clean_all(rt6_info_route, 0, &arg);
 
 	*start = buffer;
 	if (offset)
-- 
cgit v1.2.3


From 14c0b97ddfc2944982d078b8e33b088840068976 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 03:38:38 -0700
Subject: [NET]: Protocol Independant Policy Routing Rules Framework

Derived from net/ipv/fib_rules.c

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/fib_rules.h |  60 +++++++
 include/net/fib_rules.h   |  90 ++++++++++
 net/Kconfig               |   3 +
 net/core/Makefile         |   1 +
 net/core/fib_rules.c      | 416 ++++++++++++++++++++++++++++++++++++++++++++++
 net/core/rtnetlink.c      |   9 +-
 6 files changed, 577 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/fib_rules.h
 create mode 100644 include/net/fib_rules.h
 create mode 100644 net/core/fib_rules.c

(limited to 'include')

diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h
new file mode 100644
index 000000000000..5e503f0ca6e4
--- /dev/null
+++ b/include/linux/fib_rules.h
@@ -0,0 +1,60 @@
+#ifndef __LINUX_FIB_RULES_H
+#define __LINUX_FIB_RULES_H
+
+#include <linux/types.h>
+#include <linux/rtnetlink.h>
+
+/* rule is permanent, and cannot be deleted */
+#define FIB_RULE_PERMANENT	1
+
+struct fib_rule_hdr
+{
+	__u8		family;
+	__u8		dst_len;
+	__u8		src_len;
+	__u8		tos;
+
+	__u8		table;
+	__u8		res1;	/* reserved */
+	__u8		res2;	/* reserved */
+	__u8		action;
+
+	__u32		flags;
+};
+
+enum
+{
+	FRA_UNSPEC,
+	FRA_DST,	/* destination address */
+	FRA_SRC,	/* source address */
+	FRA_IFNAME,	/* interface name */
+	FRA_UNUSED1,
+	FRA_UNUSED2,
+	FRA_PRIORITY,	/* priority/preference */
+	FRA_UNUSED3,
+	FRA_UNUSED4,
+	FRA_UNUSED5,
+	FRA_FWMARK,	/* netfilter mark (IPv4) */
+	FRA_FLOW,	/* flow/class id */
+	__FRA_MAX
+};
+
+#define FRA_MAX (__FRA_MAX - 1)
+
+enum
+{
+	FR_ACT_UNSPEC,
+	FR_ACT_TO_TBL,		/* Pass to fixed table */
+	FR_ACT_RES1,
+	FR_ACT_RES2,
+	FR_ACT_RES3,
+	FR_ACT_RES4,
+	FR_ACT_BLACKHOLE,	/* Drop without notification */
+	FR_ACT_UNREACHABLE,	/* Drop with ENETUNREACH */
+	FR_ACT_PROHIBIT,	/* Drop with EACCES */
+	__FR_ACT_MAX,
+};
+
+#define FR_ACT_MAX (__FR_ACT_MAX - 1)
+
+#endif
diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
new file mode 100644
index 000000000000..61375d9e53f8
--- /dev/null
+++ b/include/net/fib_rules.h
@@ -0,0 +1,90 @@
+#ifndef __NET_FIB_RULES_H
+#define __NET_FIB_RULES_H
+
+#include <linux/types.h>
+#include <linux/netdevice.h>
+#include <linux/fib_rules.h>
+#include <net/flow.h>
+#include <net/netlink.h>
+
+struct fib_rule
+{
+	struct list_head	list;
+	atomic_t		refcnt;
+	int			ifindex;
+	char			ifname[IFNAMSIZ];
+	u32			pref;
+	u32			flags;
+	u32			table;
+	u8			action;
+	struct rcu_head		rcu;
+};
+
+struct fib_lookup_arg
+{
+	void			*lookup_ptr;
+	void			*result;
+	struct fib_rule		*rule;
+};
+
+struct fib_rules_ops
+{
+	int			family;
+	struct list_head	list;
+	int			rule_size;
+
+	int			(*action)(struct fib_rule *,
+					  struct flowi *, int,
+					  struct fib_lookup_arg *);
+	int			(*match)(struct fib_rule *,
+					 struct flowi *, int);
+	int			(*configure)(struct fib_rule *,
+					     struct sk_buff *,
+					     struct nlmsghdr *,
+					     struct fib_rule_hdr *,
+					     struct nlattr **);
+	int			(*compare)(struct fib_rule *,
+					   struct fib_rule_hdr *,
+					   struct nlattr **);
+	int			(*fill)(struct fib_rule *, struct sk_buff *,
+					struct nlmsghdr *,
+					struct fib_rule_hdr *);
+	u32			(*default_pref)(void);
+
+	int			nlgroup;
+	struct nla_policy	*policy;
+	struct list_head	*rules_list;
+	struct module		*owner;
+};
+
+static inline void fib_rule_get(struct fib_rule *rule)
+{
+	atomic_inc(&rule->refcnt);
+}
+
+static inline void fib_rule_put_rcu(struct rcu_head *head)
+{
+	struct fib_rule *rule = container_of(head, struct fib_rule, rcu);
+	kfree(rule);
+}
+
+static inline void fib_rule_put(struct fib_rule *rule)
+{
+	if (atomic_dec_and_test(&rule->refcnt))
+		call_rcu(&rule->rcu, fib_rule_put_rcu);
+}
+
+extern int			fib_rules_register(struct fib_rules_ops *);
+extern int			fib_rules_unregister(struct fib_rules_ops *);
+
+extern int			fib_rules_lookup(struct fib_rules_ops *,
+						 struct flowi *, int flags,
+						 struct fib_lookup_arg *);
+
+extern int			fib_nl_newrule(struct sk_buff *,
+					       struct nlmsghdr *, void *);
+extern int			fib_nl_delrule(struct sk_buff *,
+					       struct nlmsghdr *, void *);
+extern int			fib_rules_dump(struct sk_buff *,
+					       struct netlink_callback *, int);
+#endif
diff --git a/net/Kconfig b/net/Kconfig
index eb855b7fa642..6528a935622c 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -251,6 +251,9 @@ config WIRELESS_EXT
 
 source "net/netlabel/Kconfig"
 
+config FIB_RULES
+	bool
+
 endif   # if NET
 endmenu # Networking
 
diff --git a/net/core/Makefile b/net/core/Makefile
index 2645ba428d48..119568077dab 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -17,3 +17,4 @@ obj-$(CONFIG_NET_PKTGEN) += pktgen.o
 obj-$(CONFIG_WIRELESS_EXT) += wireless.o
 obj-$(CONFIG_NETPOLL) += netpoll.o
 obj-$(CONFIG_NET_DMA) += user_dma.o
+obj-$(CONFIG_FIB_RULES) += fib_rules.o
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
new file mode 100644
index 000000000000..6cdad24038e2
--- /dev/null
+++ b/net/core/fib_rules.c
@@ -0,0 +1,416 @@
+/*
+ * net/core/fib_rules.c		Generic Routing Rules
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License as
+ *	published by the Free Software Foundation, version 2.
+ *
+ * Authors:	Thomas Graf <tgraf@suug.ch>
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <net/fib_rules.h>
+
+static LIST_HEAD(rules_ops);
+static DEFINE_SPINLOCK(rules_mod_lock);
+
+static void notify_rule_change(int event, struct fib_rule *rule,
+			       struct fib_rules_ops *ops);
+
+static struct fib_rules_ops *lookup_rules_ops(int family)
+{
+	struct fib_rules_ops *ops;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(ops, &rules_ops, list) {
+		if (ops->family == family) {
+			if (!try_module_get(ops->owner))
+				ops = NULL;
+			rcu_read_unlock();
+			return ops;
+		}
+	}
+	rcu_read_unlock();
+
+	return NULL;
+}
+
+static void rules_ops_put(struct fib_rules_ops *ops)
+{
+	if (ops)
+		module_put(ops->owner);
+}
+
+int fib_rules_register(struct fib_rules_ops *ops)
+{
+	int err = -EEXIST;
+	struct fib_rules_ops *o;
+
+	if (ops->rule_size < sizeof(struct fib_rule))
+		return -EINVAL;
+
+	if (ops->match == NULL || ops->configure == NULL ||
+	    ops->compare == NULL || ops->fill == NULL ||
+	    ops->action == NULL)
+		return -EINVAL;
+
+	spin_lock(&rules_mod_lock);
+	list_for_each_entry(o, &rules_ops, list)
+		if (ops->family == o->family)
+			goto errout;
+
+	list_add_tail_rcu(&ops->list, &rules_ops);
+	err = 0;
+errout:
+	spin_unlock(&rules_mod_lock);
+
+	return err;
+}
+
+EXPORT_SYMBOL_GPL(fib_rules_register);
+
+static void cleanup_ops(struct fib_rules_ops *ops)
+{
+	struct fib_rule *rule, *tmp;
+
+	list_for_each_entry_safe(rule, tmp, ops->rules_list, list) {
+		list_del_rcu(&rule->list);
+		fib_rule_put(rule);
+	}
+}
+
+int fib_rules_unregister(struct fib_rules_ops *ops)
+{
+	int err = 0;
+	struct fib_rules_ops *o;
+
+	spin_lock(&rules_mod_lock);
+	list_for_each_entry(o, &rules_ops, list) {
+		if (o == ops) {
+			list_del_rcu(&o->list);
+			cleanup_ops(ops);
+			goto out;
+		}
+	}
+
+	err = -ENOENT;
+out:
+	spin_unlock(&rules_mod_lock);
+
+	synchronize_rcu();
+
+	return err;
+}
+
+EXPORT_SYMBOL_GPL(fib_rules_unregister);
+
+int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl,
+		     int flags, struct fib_lookup_arg *arg)
+{
+	struct fib_rule *rule;
+	int err;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(rule, ops->rules_list, list) {
+		if (rule->ifindex && (rule->ifindex != fl->iif))
+			continue;
+
+		if (!ops->match(rule, fl, flags))
+			continue;
+
+		err = ops->action(rule, fl, flags, arg);
+		if (err != -EAGAIN) {
+			fib_rule_get(rule);
+			arg->rule = rule;
+			goto out;
+		}
+	}
+
+	err = -ENETUNREACH;
+out:
+	rcu_read_unlock();
+
+	return err;
+}
+
+EXPORT_SYMBOL_GPL(fib_rules_lookup);
+
+int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
+{
+	struct fib_rule_hdr *frh = nlmsg_data(nlh);
+	struct fib_rules_ops *ops = NULL;
+	struct fib_rule *rule, *r, *last = NULL;
+	struct nlattr *tb[FRA_MAX+1];
+	int err = -EINVAL;
+
+	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
+		goto errout;
+
+	ops = lookup_rules_ops(frh->family);
+	if (ops == NULL) {
+		err = EAFNOSUPPORT;
+		goto errout;
+	}
+
+	err = nlmsg_parse(nlh, sizeof(*frh), tb, FRA_MAX, ops->policy);
+	if (err < 0)
+		goto errout;
+
+	if (tb[FRA_IFNAME] && nla_len(tb[FRA_IFNAME]) > IFNAMSIZ)
+		goto errout;
+
+	rule = kzalloc(ops->rule_size, GFP_KERNEL);
+	if (rule == NULL) {
+		err = -ENOMEM;
+		goto errout;
+	}
+
+	if (tb[FRA_PRIORITY])
+		rule->pref = nla_get_u32(tb[FRA_PRIORITY]);
+
+	if (tb[FRA_IFNAME]) {
+		struct net_device *dev;
+
+		rule->ifindex = -1;
+		if (nla_strlcpy(rule->ifname, tb[FRA_IFNAME],
+				IFNAMSIZ) >= IFNAMSIZ)
+			goto errout_free;
+
+		dev = __dev_get_by_name(rule->ifname);
+		if (dev)
+			rule->ifindex = dev->ifindex;
+	}
+
+	rule->action = frh->action;
+	rule->flags = frh->flags;
+	rule->table = frh->table;
+
+	if (!rule->pref && ops->default_pref)
+		rule->pref = ops->default_pref();
+
+	err = ops->configure(rule, skb, nlh, frh, tb);
+	if (err < 0)
+		goto errout_free;
+
+	list_for_each_entry(r, ops->rules_list, list) {
+		if (r->pref > rule->pref)
+			break;
+		last = r;
+	}
+
+	fib_rule_get(rule);
+
+	if (last)
+		list_add_rcu(&rule->list, &last->list);
+	else
+		list_add_rcu(&rule->list, ops->rules_list);
+
+	notify_rule_change(RTM_NEWRULE, rule, ops);
+	rules_ops_put(ops);
+	return 0;
+
+errout_free:
+	kfree(rule);
+errout:
+	rules_ops_put(ops);
+	return err;
+}
+
+int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
+{
+	struct fib_rule_hdr *frh = nlmsg_data(nlh);
+	struct fib_rules_ops *ops = NULL;
+	struct fib_rule *rule;
+	struct nlattr *tb[FRA_MAX+1];
+	int err = -EINVAL;
+
+	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
+		goto errout;
+
+	ops = lookup_rules_ops(frh->family);
+	if (ops == NULL) {
+		err = EAFNOSUPPORT;
+		goto errout;
+	}
+
+	err = nlmsg_parse(nlh, sizeof(*frh), tb, FRA_MAX, ops->policy);
+	if (err < 0)
+		goto errout;
+
+	list_for_each_entry(rule, ops->rules_list, list) {
+		if (frh->action && (frh->action != rule->action))
+			continue;
+
+		if (frh->table && (frh->table != rule->table))
+			continue;
+
+		if (tb[FRA_PRIORITY] &&
+		    (rule->pref != nla_get_u32(tb[FRA_PRIORITY])))
+			continue;
+
+		if (tb[FRA_IFNAME] &&
+		    nla_strcmp(tb[FRA_IFNAME], rule->ifname))
+			continue;
+
+		if (!ops->compare(rule, frh, tb))
+			continue;
+
+		if (rule->flags & FIB_RULE_PERMANENT) {
+			err = -EPERM;
+			goto errout;
+		}
+
+		list_del_rcu(&rule->list);
+		synchronize_rcu();
+		notify_rule_change(RTM_DELRULE, rule, ops);
+		fib_rule_put(rule);
+		rules_ops_put(ops);
+		return 0;
+	}
+
+	err = -ENOENT;
+errout:
+	rules_ops_put(ops);
+	return err;
+}
+
+static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule,
+			    u32 pid, u32 seq, int type, int flags,
+			    struct fib_rules_ops *ops)
+{
+	struct nlmsghdr *nlh;
+	struct fib_rule_hdr *frh;
+
+	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*frh), flags);
+	if (nlh == NULL)
+		return -1;
+
+	frh = nlmsg_data(nlh);
+	frh->table = rule->table;
+	frh->res1 = 0;
+	frh->res2 = 0;
+	frh->action = rule->action;
+	frh->flags = rule->flags;
+
+	if (rule->ifname[0])
+		NLA_PUT_STRING(skb, FRA_IFNAME, rule->ifname);
+
+	if (rule->pref)
+		NLA_PUT_U32(skb, FRA_PRIORITY, rule->pref);
+
+	if (ops->fill(rule, skb, nlh, frh) < 0)
+		goto nla_put_failure;
+
+	return nlmsg_end(skb, nlh);
+
+nla_put_failure:
+	return nlmsg_cancel(skb, nlh);
+}
+
+int fib_rules_dump(struct sk_buff *skb, struct netlink_callback *cb, int family)
+{
+	int idx = 0;
+	struct fib_rule *rule;
+	struct fib_rules_ops *ops;
+
+	ops = lookup_rules_ops(family);
+	if (ops == NULL)
+		return -EAFNOSUPPORT;
+
+	rcu_read_lock();
+	list_for_each_entry(rule, ops->rules_list, list) {
+		if (idx < cb->args[0])
+			goto skip;
+
+		if (fib_nl_fill_rule(skb, rule, NETLINK_CB(cb->skb).pid,
+				     cb->nlh->nlmsg_seq, RTM_NEWRULE,
+				     NLM_F_MULTI, ops) < 0)
+			break;
+skip:
+		idx++;
+	}
+	rcu_read_unlock();
+	cb->args[0] = idx;
+	rules_ops_put(ops);
+
+	return skb->len;
+}
+
+EXPORT_SYMBOL_GPL(fib_rules_dump);
+
+static void notify_rule_change(int event, struct fib_rule *rule,
+			       struct fib_rules_ops *ops)
+{
+	int size = nlmsg_total_size(sizeof(struct fib_rule_hdr) + 128);
+	struct sk_buff *skb = alloc_skb(size, GFP_KERNEL);
+
+	if (skb == NULL)
+		netlink_set_err(rtnl, 0, ops->nlgroup, ENOBUFS);
+	else if (fib_nl_fill_rule(skb, rule, 0, 0, event, 0, ops) < 0) {
+		kfree_skb(skb);
+		netlink_set_err(rtnl, 0, ops->nlgroup, EINVAL);
+	} else
+		netlink_broadcast(rtnl, skb, 0, ops->nlgroup, GFP_KERNEL);
+}
+
+static void attach_rules(struct list_head *rules, struct net_device *dev)
+{
+	struct fib_rule *rule;
+
+	list_for_each_entry(rule, rules, list) {
+		if (rule->ifindex == -1 &&
+		    strcmp(dev->name, rule->ifname) == 0)
+			rule->ifindex = dev->ifindex;
+	}
+}
+
+static void detach_rules(struct list_head *rules, struct net_device *dev)
+{
+	struct fib_rule *rule;
+
+	list_for_each_entry(rule, rules, list)
+		if (rule->ifindex == dev->ifindex)
+			rule->ifindex = -1;
+}
+
+
+static int fib_rules_event(struct notifier_block *this, unsigned long event,
+			    void *ptr)
+{
+	struct net_device *dev = ptr;
+	struct fib_rules_ops *ops;
+
+	ASSERT_RTNL();
+	rcu_read_lock();
+
+	switch (event) {
+	case NETDEV_REGISTER:
+		list_for_each_entry(ops, &rules_ops, list)
+			attach_rules(ops->rules_list, dev);
+		break;
+
+	case NETDEV_UNREGISTER:
+		list_for_each_entry(ops, &rules_ops, list)
+			detach_rules(ops->rules_list, dev);
+		break;
+	}
+
+	rcu_read_unlock();
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block fib_rules_notifier = {
+	.notifier_call = fib_rules_event,
+};
+
+static int __init fib_rules_init(void)
+{
+	return register_netdevice_notifier(&fib_rules_notifier);
+}
+
+subsys_initcall(fib_rules_init);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 30cc1ba6ed5c..aa7cff2257b1 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -49,6 +49,7 @@
 #include <net/udp.h>
 #include <net/sock.h>
 #include <net/pkt_sched.h>
+#include <net/fib_rules.h>
 #include <net/netlink.h>
 #ifdef CONFIG_NET_WIRELESS_RTNETLINK
 #include <linux/wireless.h>
@@ -103,7 +104,7 @@ static const int rtm_min[RTM_NR_FAMILIES] =
 	[RTM_FAM(RTM_NEWADDR)]      = NLMSG_LENGTH(sizeof(struct ifaddrmsg)),
 	[RTM_FAM(RTM_NEWROUTE)]     = NLMSG_LENGTH(sizeof(struct rtmsg)),
 	[RTM_FAM(RTM_NEWNEIGH)]     = NLMSG_LENGTH(sizeof(struct ndmsg)),
-	[RTM_FAM(RTM_NEWRULE)]      = NLMSG_LENGTH(sizeof(struct rtmsg)),
+	[RTM_FAM(RTM_NEWRULE)]      = NLMSG_LENGTH(sizeof(struct fib_rule_hdr)),
 	[RTM_FAM(RTM_NEWQDISC)]     = NLMSG_LENGTH(sizeof(struct tcmsg)),
 	[RTM_FAM(RTM_NEWTCLASS)]    = NLMSG_LENGTH(sizeof(struct tcmsg)),
 	[RTM_FAM(RTM_NEWTFILTER)]   = NLMSG_LENGTH(sizeof(struct tcmsg)),
@@ -120,7 +121,7 @@ static const int rta_max[RTM_NR_FAMILIES] =
 	[RTM_FAM(RTM_NEWADDR)]      = IFA_MAX,
 	[RTM_FAM(RTM_NEWROUTE)]     = RTA_MAX,
 	[RTM_FAM(RTM_NEWNEIGH)]     = NDA_MAX,
-	[RTM_FAM(RTM_NEWRULE)]      = RTA_MAX,
+	[RTM_FAM(RTM_NEWRULE)]      = FRA_MAX,
 	[RTM_FAM(RTM_NEWQDISC)]     = TCA_MAX,
 	[RTM_FAM(RTM_NEWTCLASS)]    = TCA_MAX,
 	[RTM_FAM(RTM_NEWTFILTER)]   = TCA_MAX,
@@ -757,6 +758,10 @@ static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] =
 	[RTM_NEWNEIGH    - RTM_BASE] = { .doit   = neigh_add		 },
 	[RTM_DELNEIGH    - RTM_BASE] = { .doit   = neigh_delete		 },
 	[RTM_GETNEIGH    - RTM_BASE] = { .dumpit = neigh_dump_info	 },
+#ifdef CONFIG_FIB_RULES
+	[RTM_NEWRULE     - RTM_BASE] = { .doit   = fib_nl_newrule	 },
+	[RTM_DELRULE     - RTM_BASE] = { .doit   = fib_nl_delrule	 },
+#endif
 	[RTM_GETRULE     - RTM_BASE] = { .dumpit = rtnetlink_dump_all	 },
 	[RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info	 },
 	[RTM_SETNEIGHTBL - RTM_BASE] = { .doit   = neightbl_set		 },
-- 
cgit v1.2.3


From 101367c2f8c464ea96643192673aa18d88e6336d Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 03:39:02 -0700
Subject: [IPV6]: Policy Routing Rules

Adds support for policy routing rules including a new
local table for routes with a local destination.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h |   2 +
 include/net/ip6_fib.h     |   9 +-
 include/net/ip6_route.h   |   5 +
 net/ipv6/Kconfig          |   1 +
 net/ipv6/Makefile         |   1 +
 net/ipv6/addrconf.c       |   1 +
 net/ipv6/fib6_rules.c     | 251 ++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/ip6_fib.c        |  21 ++--
 net/ipv6/route.c          |  50 +++++++++
 9 files changed, 329 insertions(+), 12 deletions(-)
 create mode 100644 net/ipv6/fib6_rules.c

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index facd9ee37b76..bf353538ae93 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -889,6 +889,8 @@ enum rtnetlink_groups {
 	RTNLGRP_NOP4,
 	RTNLGRP_IPV6_PREFIX,
 #define RTNLGRP_IPV6_PREFIX	RTNLGRP_IPV6_PREFIX
+	RTNLGRP_IPV6_RULE,
+#define RTNLGRP_IPV6_RULE	RTNLGRP_IPV6_RULE
 	__RTNLGRP_MAX
 };
 #define RTNLGRP_MAX	(__RTNLGRP_MAX - 1)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 818411519c89..7b47e8d5a765 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -155,7 +155,6 @@ struct fib6_table {
 
 #define RT6_TABLE_UNSPEC	RT_TABLE_UNSPEC
 #define RT6_TABLE_MAIN		RT_TABLE_MAIN
-#define RT6_TABLE_LOCAL		RT6_TABLE_MAIN
 #define RT6_TABLE_DFLT		RT6_TABLE_MAIN
 #define RT6_TABLE_INFO		RT6_TABLE_MAIN
 #define RT6_TABLE_PREFIX	RT6_TABLE_MAIN
@@ -163,9 +162,11 @@ struct fib6_table {
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 #define FIB6_TABLE_MIN		1
 #define FIB6_TABLE_MAX		RT_TABLE_MAX
+#define RT6_TABLE_LOCAL		RT_TABLE_LOCAL
 #else
 #define FIB6_TABLE_MIN		RT_TABLE_MAIN
 #define FIB6_TABLE_MAX		FIB6_TABLE_MIN
+#define RT6_TABLE_LOCAL		RT6_TABLE_MAIN
 #endif
 
 #define RT6_F_STRICT		1
@@ -221,5 +222,11 @@ extern void			fib6_run_gc(unsigned long dummy);
 extern void			fib6_gc_cleanup(void);
 
 extern void			fib6_init(void);
+
+extern void			fib6_rules_init(void);
+extern void			fib6_rules_cleanup(void);
+extern int			fib6_rules_dump(struct sk_buff *,
+						struct netlink_callback *);
+
 #endif
 #endif
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index d49c8c90eb68..9bfa3cc6cedb 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -41,6 +41,11 @@ struct pol_chain {
 
 extern struct rt6_info	ip6_null_entry;
 
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+extern struct rt6_info	ip6_prohibit_entry;
+extern struct rt6_info	ip6_blk_hole_entry;
+#endif
+
 extern int ip6_rt_gc_interval;
 
 extern void			ip6_route_input(struct sk_buff *skb);
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 159c63d99c81..36a6c2b79889 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -139,6 +139,7 @@ config IPV6_TUNNEL
 config IPV6_MULTIPLE_TABLES
 	bool "IPv6: Multiple Routing Tables"
 	depends on IPV6 && EXPERIMENTAL
+	select FIB_RULES
 	---help---
 	  Support multiple routing tables.
 
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 386e0a626948..9eebf6091279 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -13,6 +13,7 @@ ipv6-objs :=	af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o sit.o \
 ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \
 	xfrm6_output.o
 ipv6-$(CONFIG_NETFILTER) += netfilter.o
+ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o
 ipv6-objs += $(ipv6-y)
 
 obj-$(CONFIG_INET6_AH) += ah6.o
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 318767fcefdc..ed766eebc022 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3528,6 +3528,7 @@ static struct rtnetlink_link inet6_rtnetlink_table[RTM_NR_MSGTYPES] = {
 	[RTM_DELROUTE - RTM_BASE] = { .doit	= inet6_rtm_delroute, },
 	[RTM_GETROUTE - RTM_BASE] = { .doit	= inet6_rtm_getroute,
 				      .dumpit	= inet6_dump_fib, },
+	[RTM_GETRULE  - RTM_BASE] = { .dumpit   = fib6_rules_dump,   },
 };
 
 static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
new file mode 100644
index 000000000000..c3c8195744ee
--- /dev/null
+++ b/net/ipv6/fib6_rules.c
@@ -0,0 +1,251 @@
+/*
+ * net/ipv6/fib6_rules.c	IPv6 Routing Policy Rules
+ *
+ * Copyright (C)2003-2006 Helsinki University of Technology
+ * Copyright (C)2003-2006 USAGI/WIDE Project
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License as
+ *	published by the Free Software Foundation, version 2.
+ *
+ * Authors
+ *	Thomas Graf		<tgraf@suug.ch>
+ *	Ville Nuorvala		<vnuorval@tcs.hut.fi>
+ */
+
+#include <linux/config.h>
+#include <linux/netdevice.h>
+
+#include <net/fib_rules.h>
+#include <net/ipv6.h>
+#include <net/ip6_route.h>
+#include <net/netlink.h>
+
+struct fib6_rule
+{
+	struct fib_rule		common;
+	struct rt6key		src;
+	struct rt6key		dst;
+	u8			tclass;
+};
+
+static struct fib_rules_ops fib6_rules_ops;
+
+static struct fib6_rule main_rule = {
+	.common = {
+		.refcnt =	ATOMIC_INIT(2),
+		.pref =		0x7FFE,
+		.action =	FR_ACT_TO_TBL,
+		.table =	RT6_TABLE_MAIN,
+	},
+};
+
+static struct fib6_rule local_rule = {
+	.common = {
+		.refcnt =	ATOMIC_INIT(2),
+		.pref =		0,
+		.action =	FR_ACT_TO_TBL,
+		.table =	RT6_TABLE_LOCAL,
+		.flags =	FIB_RULE_PERMANENT,
+	},
+};
+
+static LIST_HEAD(fib6_rules);
+
+struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
+				   pol_lookup_t lookup)
+{
+	struct fib_lookup_arg arg = {
+		.lookup_ptr = lookup,
+	};
+
+	fib_rules_lookup(&fib6_rules_ops, fl, flags, &arg);
+	if (arg.rule)
+		fib_rule_put(arg.rule);
+
+	return (struct dst_entry *) arg.result;
+}
+
+int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
+		     int flags, struct fib_lookup_arg *arg)
+{
+	struct rt6_info *rt = NULL;
+	struct fib6_table *table;
+	pol_lookup_t lookup = arg->lookup_ptr;
+
+	switch (rule->action) {
+	case FR_ACT_TO_TBL:
+		break;
+	case FR_ACT_UNREACHABLE:
+		rt = &ip6_null_entry;
+		goto discard_pkt;
+	default:
+	case FR_ACT_BLACKHOLE:
+		rt = &ip6_blk_hole_entry;
+		goto discard_pkt;
+	case FR_ACT_PROHIBIT:
+		rt = &ip6_prohibit_entry;
+		goto discard_pkt;
+	}
+
+	table = fib6_get_table(rule->table);
+	if (table)
+		rt = lookup(table, flp, flags);
+
+	if (rt != &ip6_null_entry)
+		goto out;
+
+	dst_release(&rt->u.dst);
+discard_pkt:
+	dst_hold(&rt->u.dst);
+out:
+	arg->result = rt;
+	return rt == NULL ? -EAGAIN : 0;
+}
+
+
+static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
+{
+	struct fib6_rule *r = (struct fib6_rule *) rule;
+
+	if (!ipv6_prefix_equal(&fl->fl6_dst, &r->dst.addr, r->dst.plen))
+		return 0;
+
+	if ((flags & RT6_F_HAS_SADDR) &&
+	    !ipv6_prefix_equal(&fl->fl6_src, &r->src.addr, r->src.plen))
+		return 0;
+
+	return 1;
+}
+
+static struct nla_policy fib6_rule_policy[RTA_MAX+1] __read_mostly = {
+	[FRA_IFNAME]	= { .type = NLA_STRING },
+	[FRA_PRIORITY]	= { .type = NLA_U32 },
+	[FRA_SRC]	= { .minlen = sizeof(struct in6_addr) },
+	[FRA_DST]	= { .minlen = sizeof(struct in6_addr) },
+};
+
+static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
+			       struct nlmsghdr *nlh, struct fib_rule_hdr *frh,
+			       struct nlattr **tb)
+{
+	int err = -EINVAL;
+	struct fib6_rule *rule6 = (struct fib6_rule *) rule;
+
+	if (frh->src_len > 128 || frh->dst_len > 128 ||
+	    (frh->tos & ~IPV6_FLOWINFO_MASK))
+		goto errout;
+
+	if (rule->action == FR_ACT_TO_TBL) {
+		if (rule->table == RT6_TABLE_UNSPEC)
+			goto errout;
+
+		if (fib6_new_table(rule->table) == NULL) {
+			err = -ENOBUFS;
+			goto errout;
+		}
+	}
+
+	if (tb[FRA_SRC])
+		nla_memcpy(&rule6->src.addr, tb[FRA_SRC],
+			   sizeof(struct in6_addr));
+
+	if (tb[FRA_DST])
+		nla_memcpy(&rule6->dst.addr, tb[FRA_DST],
+			   sizeof(struct in6_addr));
+
+	rule6->src.plen = frh->src_len;
+	rule6->dst.plen = frh->dst_len;
+	rule6->tclass = frh->tos;
+
+	err = 0;
+errout:
+	return err;
+}
+
+static int fib6_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
+			     struct nlattr **tb)
+{
+	struct fib6_rule *rule6 = (struct fib6_rule *) rule;
+
+	if (frh->src_len && (rule6->src.plen != frh->src_len))
+		return 0;
+
+	if (frh->dst_len && (rule6->dst.plen != frh->dst_len))
+		return 0;
+
+	if (frh->tos && (rule6->tclass != frh->tos))
+		return 0;
+
+	if (tb[FRA_SRC] &&
+	    nla_memcmp(tb[FRA_SRC], &rule6->src.addr, sizeof(struct in6_addr)))
+		return 0;
+
+	if (tb[FRA_DST] &&
+	    nla_memcmp(tb[FRA_DST], &rule6->dst.addr, sizeof(struct in6_addr)))
+		return 0;
+
+	return 1;
+}
+
+static int fib6_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
+			  struct nlmsghdr *nlh, struct fib_rule_hdr *frh)
+{
+	struct fib6_rule *rule6 = (struct fib6_rule *) rule;
+
+	frh->family = AF_INET6;
+	frh->dst_len = rule6->dst.plen;
+	frh->src_len = rule6->src.plen;
+	frh->tos = rule6->tclass;
+
+	if (rule6->dst.plen)
+		NLA_PUT(skb, FRA_DST, sizeof(struct in6_addr),
+			&rule6->dst.addr);
+
+	if (rule6->src.plen)
+		NLA_PUT(skb, FRA_SRC, sizeof(struct in6_addr),
+			&rule6->src.addr);
+
+	return 0;
+
+nla_put_failure:
+	return -ENOBUFS;
+}
+
+int fib6_rules_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	return fib_rules_dump(skb, cb, AF_INET6);
+}
+
+static u32 fib6_rule_default_pref(void)
+{
+	return 0x3FFF;
+}
+
+static struct fib_rules_ops fib6_rules_ops = {
+	.family			= AF_INET6,
+	.rule_size		= sizeof(struct fib6_rule),
+	.action			= fib6_rule_action,
+	.match			= fib6_rule_match,
+	.configure		= fib6_rule_configure,
+	.compare		= fib6_rule_compare,
+	.fill			= fib6_rule_fill,
+	.default_pref		= fib6_rule_default_pref,
+	.nlgroup		= RTNLGRP_IPV6_RULE,
+	.policy			= fib6_rule_policy,
+	.rules_list		= &fib6_rules,
+	.owner			= THIS_MODULE,
+};
+
+void __init fib6_rules_init(void)
+{
+	list_add_tail(&local_rule.common.list, &fib6_rules);
+	list_add_tail(&main_rule.common.list, &fib6_rules);
+
+	fib_rules_register(&fib6_rules_ops);
+}
+
+void fib6_rules_cleanup(void)
+{
+	fib_rules_unregister(&fib6_rules_ops);
+}
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index fcd7da830aca..ce226c14bef5 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -159,6 +159,15 @@ static struct fib6_table fib6_main_tbl = {
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 
+static struct fib6_table fib6_local_tbl = {
+	.tb6_id		= RT6_TABLE_LOCAL,
+	.tb6_lock	= RW_LOCK_UNLOCKED,
+	.tb6_root 	= {
+		.leaf		= &ip6_null_entry,
+		.fn_flags	= RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO,
+	},
+};
+
 #define FIB_TABLE_HASHSZ 256
 static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
@@ -228,20 +237,10 @@ struct fib6_table *fib6_get_table(u32 id)
 	return NULL;
 }
 
-struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
-				   pol_lookup_t lookup)
-{
-	/*
-	 * TODO: Add rule lookup
-	 */
-	struct fib6_table *table = fib6_get_table(RT6_TABLE_MAIN);
-
-	return (struct dst_entry *) lookup(table, fl, flags);
-}
-
 static void __init fib6_tables_init(void)
 {
 	fib6_link_table(&fib6_main_tbl);
+	fib6_link_table(&fib6_local_tbl);
 }
 
 #else
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 73efdadb9ab8..438977e2085d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -140,6 +140,50 @@ struct rt6_info ip6_null_entry = {
 	.rt6i_ref	= ATOMIC_INIT(1),
 };
 
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+
+struct rt6_info ip6_prohibit_entry = {
+	.u = {
+		.dst = {
+			.__refcnt	= ATOMIC_INIT(1),
+			.__use		= 1,
+			.dev		= &loopback_dev,
+			.obsolete	= -1,
+			.error		= -EACCES,
+			.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
+			.input		= ip6_pkt_discard,
+			.output		= ip6_pkt_discard_out,
+			.ops		= &ip6_dst_ops,
+			.path		= (struct dst_entry*)&ip6_prohibit_entry,
+		}
+	},
+	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
+	.rt6i_metric	= ~(u32) 0,
+	.rt6i_ref	= ATOMIC_INIT(1),
+};
+
+struct rt6_info ip6_blk_hole_entry = {
+	.u = {
+		.dst = {
+			.__refcnt	= ATOMIC_INIT(1),
+			.__use		= 1,
+			.dev		= &loopback_dev,
+			.obsolete	= -1,
+			.error		= -EINVAL,
+			.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
+			.input		= ip6_pkt_discard,
+			.output		= ip6_pkt_discard_out,
+			.ops		= &ip6_dst_ops,
+			.path		= (struct dst_entry*)&ip6_blk_hole_entry,
+		}
+	},
+	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
+	.rt6i_metric	= ~(u32) 0,
+	.rt6i_ref	= ATOMIC_INIT(1),
+};
+
+#endif
+
 /* allocate dst with ip6_dst_ops */
 static __inline__ struct rt6_info *ip6_dst_alloc(void)
 {
@@ -2408,10 +2452,16 @@ void __init ip6_route_init(void)
 #ifdef CONFIG_XFRM
 	xfrm6_init();
 #endif
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	fib6_rules_init();
+#endif
 }
 
 void ip6_route_cleanup(void)
 {
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
+	fib6_rules_cleanup();
+#endif
 #ifdef CONFIG_PROC_FS
 	proc_net_remove("ipv6_route");
 	proc_net_remove("rt6_stats");
-- 
cgit v1.2.3


From e1ef4bf23b1ced0bf78a1c98289f746486e5c912 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 03:39:22 -0700
Subject: [IPV4]: Use Protocol Independant Policy Routing Rules Framework

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    |  14 +-
 net/ipv4/Kconfig        |   1 +
 net/ipv4/devinet.c      |   4 +-
 net/ipv4/fib_frontend.c |   2 +-
 net/ipv4/fib_rules.c    | 605 +++++++++++++++++++-----------------------------
 5 files changed, 249 insertions(+), 377 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index a095d1dec7a4..14c82e611c95 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -18,6 +18,7 @@
 
 #include <net/flow.h>
 #include <linux/seq_file.h>
+#include <net/fib_rules.h>
 
 /* WARNING: The ordering of these elements must match ordering
  *          of RTA_* rtnetlink attribute numbers.
@@ -203,9 +204,8 @@ static inline void fib_select_default(const struct flowi *flp, struct fib_result
 #define ip_fib_main_table (fib_tables[RT_TABLE_MAIN])
 
 extern struct fib_table * fib_tables[RT_TABLE_MAX+1];
-extern int fib_lookup(const struct flowi *flp, struct fib_result *res);
+extern int fib_lookup(struct flowi *flp, struct fib_result *res);
 extern struct fib_table *__fib_new_table(int id);
-extern void fib_rule_put(struct fib_rule *r);
 
 static inline struct fib_table *fib_get_table(int id)
 {
@@ -251,15 +251,15 @@ extern u32  __fib_res_prefsrc(struct fib_result *res);
 extern struct fib_table *fib_hash_init(int id);
 
 #ifdef CONFIG_IP_MULTIPLE_TABLES
-/* Exported by fib_rules.c */
+extern int fib4_rules_dump(struct sk_buff *skb, struct netlink_callback *cb);
+
+extern void __init fib4_rules_init(void);
+extern void __exit fib4_rules_cleanup(void);
 
-extern int inet_rtm_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
-extern int inet_rtm_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
-extern int inet_dump_rules(struct sk_buff *skb, struct netlink_callback *cb);
 #ifdef CONFIG_NET_CLS_ROUTE
 extern u32 fib_rules_tclass(struct fib_result *res);
 #endif
-extern void fib_rules_init(void);
+
 #endif
 
 static inline void fib_combine_itag(u32 *itag, struct fib_result *res)
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 3b5d504a74be..1650b64415aa 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -88,6 +88,7 @@ config IP_FIB_HASH
 config IP_MULTIPLE_TABLES
 	bool "IP: policy routing"
 	depends on IP_ADVANCED_ROUTER
+	select FIB_RULES
 	---help---
 	  Normally, a router decides what to do with a received packet based
 	  solely on the packet's final destination address. If you say Y here,
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index a6cc31d911eb..9f3ffbec3296 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1151,9 +1151,7 @@ static struct rtnetlink_link inet_rtnetlink_table[RTM_NR_MSGTYPES] = {
 	[RTM_GETROUTE - RTM_BASE] = { .doit	= inet_rtm_getroute,
 				      .dumpit	= inet_dump_fib,	},
 #ifdef CONFIG_IP_MULTIPLE_TABLES
-	[RTM_NEWRULE  - RTM_BASE] = { .doit	= inet_rtm_newrule,	},
-	[RTM_DELRULE  - RTM_BASE] = { .doit	= inet_rtm_delrule,	},
-	[RTM_GETRULE  - RTM_BASE] = { .dumpit	= inet_dump_rules,	},
+	[RTM_GETRULE  - RTM_BASE] = { .dumpit	= fib4_rules_dump,	},
 #endif
 };
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index ba2a70745a63..fe4a53d4d10d 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -656,7 +656,7 @@ void __init ip_fib_init(void)
 	ip_fib_local_table = fib_hash_init(RT_TABLE_LOCAL);
 	ip_fib_main_table  = fib_hash_init(RT_TABLE_MAIN);
 #else
-	fib_rules_init();
+	fib4_rules_init();
 #endif
 
 	register_netdevice_notifier(&fib_netdev_notifier);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 79b04718bdfd..23ec6ae1a0f6 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -5,9 +5,8 @@
  *
  *		IPv4 Forwarding Information Base: policy rules.
  *
- * Version:	$Id: fib_rules.c,v 1.17 2001/10/31 21:55:54 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
+ * 		Thomas Graf <tgraf@suug.ch>
  *
  *		This program is free software; you can redistribute it and/or
  *		modify it under the terms of the GNU General Public License
@@ -19,129 +18,154 @@
  *		Marc Boucher	:	routing by fwmark
  */
 
-#include <asm/uaccess.h>
-#include <asm/system.h>
-#include <linux/bitops.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/socket.h>
-#include <linux/sockios.h>
-#include <linux/errno.h>
-#include <linux/in.h>
-#include <linux/inet.h>
-#include <linux/inetdevice.h>
 #include <linux/netdevice.h>
-#include <linux/if_arp.h>
-#include <linux/proc_fs.h>
-#include <linux/skbuff.h>
 #include <linux/netlink.h>
+#include <linux/inetdevice.h>
 #include <linux/init.h>
 #include <linux/list.h>
 #include <linux/rcupdate.h>
-
 #include <net/ip.h>
-#include <net/protocol.h>
 #include <net/route.h>
 #include <net/tcp.h>
-#include <net/sock.h>
 #include <net/ip_fib.h>
+#include <net/fib_rules.h>
 
-#define FRprintk(a...)
+static struct fib_rules_ops fib4_rules_ops;
 
-struct fib_rule
+struct fib4_rule
 {
-	struct hlist_node hlist;
-	atomic_t	r_clntref;
-	u32		r_preference;
-	unsigned char	r_table;
-	unsigned char	r_action;
-	unsigned char	r_dst_len;
-	unsigned char	r_src_len;
-	u32		r_src;
-	u32		r_srcmask;
-	u32		r_dst;
-	u32		r_dstmask;
-	u32		r_srcmap;
-	u8		r_flags;
-	u8		r_tos;
+	struct fib_rule		common;
+	u8			dst_len;
+	u8			src_len;
+	u8			tos;
+	u32			src;
+	u32			srcmask;
+	u32			dst;
+	u32			dstmask;
 #ifdef CONFIG_IP_ROUTE_FWMARK
-	u32		r_fwmark;
+	u32			fwmark;
 #endif
-	int		r_ifindex;
 #ifdef CONFIG_NET_CLS_ROUTE
-	__u32		r_tclassid;
+	u32			tclassid;
 #endif
-	char		r_ifname[IFNAMSIZ];
-	int		r_dead;
-	struct		rcu_head rcu;
 };
 
-static struct fib_rule default_rule = {
-	.r_clntref =	ATOMIC_INIT(2),
-	.r_preference =	0x7FFF,
-	.r_table =	RT_TABLE_DEFAULT,
-	.r_action =	RTN_UNICAST,
+static struct fib4_rule default_rule = {
+	.common = {
+		.refcnt =	ATOMIC_INIT(2),
+		.pref =		0x7FFF,
+		.table =	RT_TABLE_DEFAULT,
+		.action =	FR_ACT_TO_TBL,
+	},
 };
 
-static struct fib_rule main_rule = {
-	.r_clntref =	ATOMIC_INIT(2),
-	.r_preference =	0x7FFE,
-	.r_table =	RT_TABLE_MAIN,
-	.r_action =	RTN_UNICAST,
+static struct fib4_rule main_rule = {
+	.common = {
+		.refcnt =	ATOMIC_INIT(2),
+		.pref =		0x7FFE,
+		.table =	RT_TABLE_MAIN,
+		.action =	FR_ACT_TO_TBL,
+	},
 };
 
-static struct fib_rule local_rule = {
-	.r_clntref =	ATOMIC_INIT(2),
-	.r_table =	RT_TABLE_LOCAL,
-	.r_action =	RTN_UNICAST,
+static struct fib4_rule local_rule = {
+	.common = {
+		.refcnt =	ATOMIC_INIT(2),
+		.table =	RT_TABLE_LOCAL,
+		.action =	FR_ACT_TO_TBL,
+		.flags =	FIB_RULE_PERMANENT,
+	},
 };
 
-static struct hlist_head fib_rules;
+static LIST_HEAD(fib4_rules);
+
+#ifdef CONFIG_NET_CLS_ROUTE
+u32 fib_rules_tclass(struct fib_result *res)
+{
+	return res->r ? ((struct fib4_rule *) res->r)->tclassid : 0;
+}
+#endif
 
-/* writer func called from netlink -- rtnl_sem hold*/
+int fib_lookup(struct flowi *flp, struct fib_result *res)
+{
+	struct fib_lookup_arg arg = {
+		.result = res,
+	};
+	int err;
 
-static void rtmsg_rule(int, struct fib_rule *);
+	err = fib_rules_lookup(&fib4_rules_ops, flp, 0, &arg);
+	res->r = arg.rule;
 
-int inet_rtm_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
+	return err;
+}
+
+int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, int flags,
+		     struct fib_lookup_arg *arg)
 {
-	struct rtattr **rta = arg;
-	struct rtmsg *rtm = NLMSG_DATA(nlh);
-	struct fib_rule *r;
-	struct hlist_node *node;
-	int err = -ESRCH;
-
-	hlist_for_each_entry(r, node, &fib_rules, hlist) {
-		if ((!rta[RTA_SRC-1] || memcmp(RTA_DATA(rta[RTA_SRC-1]), &r->r_src, 4) == 0) &&
-		    rtm->rtm_src_len == r->r_src_len &&
-		    rtm->rtm_dst_len == r->r_dst_len &&
-		    (!rta[RTA_DST-1] || memcmp(RTA_DATA(rta[RTA_DST-1]), &r->r_dst, 4) == 0) &&
-		    rtm->rtm_tos == r->r_tos &&
-#ifdef CONFIG_IP_ROUTE_FWMARK
-		    (!rta[RTA_PROTOINFO-1] || memcmp(RTA_DATA(rta[RTA_PROTOINFO-1]), &r->r_fwmark, 4) == 0) &&
-#endif
-		    (!rtm->rtm_type || rtm->rtm_type == r->r_action) &&
-		    (!rta[RTA_PRIORITY-1] || memcmp(RTA_DATA(rta[RTA_PRIORITY-1]), &r->r_preference, 4) == 0) &&
-		    (!rta[RTA_IIF-1] || rtattr_strcmp(rta[RTA_IIF-1], r->r_ifname) == 0) &&
-		    (!rtm->rtm_table || (r && rtm->rtm_table == r->r_table))) {
-			err = -EPERM;
-			if (r == &local_rule)
-				break;
-
-			hlist_del_rcu(&r->hlist);
-			r->r_dead = 1;
-			rtmsg_rule(RTM_DELRULE, r);
-			fib_rule_put(r);
-			err = 0;
-			break;
-		}
+	int err = -EAGAIN;
+	struct fib_table *tbl;
+
+	switch (rule->action) {
+	case FR_ACT_TO_TBL:
+		break;
+
+	case FR_ACT_UNREACHABLE:
+		err = -ENETUNREACH;
+		goto errout;
+
+	case FR_ACT_PROHIBIT:
+		err = -EACCES;
+		goto errout;
+
+	case FR_ACT_BLACKHOLE:
+	default:
+		err = -EINVAL;
+		goto errout;
 	}
+
+	if ((tbl = fib_get_table(rule->table)) == NULL)
+		goto errout;
+
+	err = tbl->tb_lookup(tbl, flp, (struct fib_result *) arg->result);
+	if (err > 0)
+		err = -EAGAIN;
+errout:
 	return err;
 }
 
-/* Allocate new unique table id */
+
+void fib_select_default(const struct flowi *flp, struct fib_result *res)
+{
+	if (res->r && res->r->action == FR_ACT_TO_TBL &&
+	    FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK) {
+		struct fib_table *tb;
+		if ((tb = fib_get_table(res->r->table)) != NULL)
+			tb->tb_select_default(tb, flp, res);
+	}
+}
+
+static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
+{
+	struct fib4_rule *r = (struct fib4_rule *) rule;
+	u32 daddr = fl->fl4_dst;
+	u32 saddr = fl->fl4_src;
+
+	if (((saddr ^ r->src) & r->srcmask) ||
+	    ((daddr ^ r->dst) & r->dstmask))
+		return 0;
+
+	if (r->tos && (r->tos != fl->fl4_tos))
+		return 0;
+
+#ifdef CONFIG_IP_ROUTE_FWMARK
+	if (r->fwmark && (r->fwmark != fl->fl4_fwmark))
+		return 0;
+#endif
+
+	return 1;
+}
 
 static struct fib_table *fib_empty_table(void)
 {
@@ -153,329 +177,178 @@ static struct fib_table *fib_empty_table(void)
 	return NULL;
 }
 
-static inline void fib_rule_put_rcu(struct rcu_head *head)
-{
-	struct fib_rule *r = container_of(head, struct fib_rule, rcu);
-	kfree(r);
-}
+static struct nla_policy fib4_rule_policy[FRA_MAX+1] __read_mostly = {
+	[FRA_IFNAME]	= { .type = NLA_STRING },
+	[FRA_PRIORITY]	= { .type = NLA_U32 },
+	[FRA_SRC]	= { .type = NLA_U32 },
+	[FRA_DST]	= { .type = NLA_U32 },
+	[FRA_FWMARK]	= { .type = NLA_U32 },
+	[FRA_FLOW]	= { .type = NLA_U32 },
+};
 
-void fib_rule_put(struct fib_rule *r)
+static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
+			       struct nlmsghdr *nlh, struct fib_rule_hdr *frh,
+			       struct nlattr **tb)
 {
-	if (atomic_dec_and_test(&r->r_clntref)) {
-		if (r->r_dead)
-			call_rcu(&r->rcu, fib_rule_put_rcu);
-		else
-			printk("Freeing alive rule %p\n", r);
-	}
-}
+	int err = -EINVAL;
+	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
-/* writer func called from netlink -- rtnl_sem hold*/
+	if (frh->src_len > 32 || frh->dst_len > 32 ||
+	    (frh->tos & ~IPTOS_TOS_MASK))
+		goto errout;
 
-int inet_rtm_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
-{
-	struct rtattr **rta = arg;
-	struct rtmsg *rtm = NLMSG_DATA(nlh);
-	struct fib_rule *r, *new_r, *last = NULL;
-	struct hlist_node *node = NULL;
-	unsigned char table_id;
-
-	if (rtm->rtm_src_len > 32 || rtm->rtm_dst_len > 32 ||
-	    (rtm->rtm_tos & ~IPTOS_TOS_MASK))
-		return -EINVAL;
-
-	if (rta[RTA_IIF-1] && RTA_PAYLOAD(rta[RTA_IIF-1]) > IFNAMSIZ)
-		return -EINVAL;
-
-	table_id = rtm->rtm_table;
-	if (table_id == RT_TABLE_UNSPEC) {
-		struct fib_table *table;
-		if (rtm->rtm_type == RTN_UNICAST) {
-			if ((table = fib_empty_table()) == NULL)
-				return -ENOBUFS;
-			table_id = table->tb_id;
-		}
-	}
+	if (rule->table == RT_TABLE_UNSPEC) {
+		if (rule->action == FR_ACT_TO_TBL) {
+			struct fib_table *table;
 
-	new_r = kzalloc(sizeof(*new_r), GFP_KERNEL);
-	if (!new_r)
-		return -ENOMEM;
-
-	if (rta[RTA_SRC-1])
-		memcpy(&new_r->r_src, RTA_DATA(rta[RTA_SRC-1]), 4);
-	if (rta[RTA_DST-1])
-		memcpy(&new_r->r_dst, RTA_DATA(rta[RTA_DST-1]), 4);
-	if (rta[RTA_GATEWAY-1])
-		memcpy(&new_r->r_srcmap, RTA_DATA(rta[RTA_GATEWAY-1]), 4);
-	new_r->r_src_len = rtm->rtm_src_len;
-	new_r->r_dst_len = rtm->rtm_dst_len;
-	new_r->r_srcmask = inet_make_mask(rtm->rtm_src_len);
-	new_r->r_dstmask = inet_make_mask(rtm->rtm_dst_len);
-	new_r->r_tos = rtm->rtm_tos;
-#ifdef CONFIG_IP_ROUTE_FWMARK
-	if (rta[RTA_PROTOINFO-1])
-		memcpy(&new_r->r_fwmark, RTA_DATA(rta[RTA_PROTOINFO-1]), 4);
-#endif
-	new_r->r_action = rtm->rtm_type;
-	new_r->r_flags = rtm->rtm_flags;
-	if (rta[RTA_PRIORITY-1])
-		memcpy(&new_r->r_preference, RTA_DATA(rta[RTA_PRIORITY-1]), 4);
-	new_r->r_table = table_id;
-	if (rta[RTA_IIF-1]) {
-		struct net_device *dev;
-		rtattr_strlcpy(new_r->r_ifname, rta[RTA_IIF-1], IFNAMSIZ);
-		new_r->r_ifindex = -1;
-		dev = __dev_get_by_name(new_r->r_ifname);
-		if (dev)
-			new_r->r_ifindex = dev->ifindex;
-	}
-#ifdef CONFIG_NET_CLS_ROUTE
-	if (rta[RTA_FLOW-1])
-		memcpy(&new_r->r_tclassid, RTA_DATA(rta[RTA_FLOW-1]), 4);
-#endif
-	r = container_of(fib_rules.first, struct fib_rule, hlist);
+			table = fib_empty_table();
+			if (table == NULL) {
+				err = -ENOBUFS;
+				goto errout;
+			}
 
-	if (!new_r->r_preference) {
-		if (r && r->hlist.next != NULL) {
-			r = container_of(r->hlist.next, struct fib_rule, hlist);
-			if (r->r_preference)
-				new_r->r_preference = r->r_preference - 1;
+			rule->table = table->tb_id;
 		}
 	}
 
-	hlist_for_each_entry(r, node, &fib_rules, hlist) {
-		if (r->r_preference > new_r->r_preference)
-			break;
-		last = r;
-	}
-	atomic_inc(&new_r->r_clntref);
+	if (tb[FRA_SRC])
+		rule4->src = nla_get_u32(tb[FRA_SRC]);
 
-	if (last)
-		hlist_add_after_rcu(&last->hlist, &new_r->hlist);
-	else
-		hlist_add_before_rcu(&new_r->hlist, &r->hlist);
+	if (tb[FRA_DST])
+		rule4->dst = nla_get_u32(tb[FRA_DST]);
 
-	rtmsg_rule(RTM_NEWRULE, new_r);
-	return 0;
-}
+#ifdef CONFIG_IP_ROUTE_FWMARK
+	if (tb[FRA_FWMARK])
+		rule4->fwmark = nla_get_u32(tb[FRA_FWMARK]);
+#endif
 
 #ifdef CONFIG_NET_CLS_ROUTE
-u32 fib_rules_tclass(struct fib_result *res)
-{
-	if (res->r)
-		return res->r->r_tclassid;
-	return 0;
-}
+	if (tb[FRA_FLOW])
+		rule4->tclassid = nla_get_u32(tb[FRA_FLOW]);
 #endif
 
-/* callers should hold rtnl semaphore */
-
-static void fib_rules_detach(struct net_device *dev)
-{
-	struct hlist_node *node;
-	struct fib_rule *r;
-
-	hlist_for_each_entry(r, node, &fib_rules, hlist) {
-		if (r->r_ifindex == dev->ifindex)
-			r->r_ifindex = -1;
+	rule4->src_len = frh->src_len;
+	rule4->srcmask = inet_make_mask(rule4->src_len);
+	rule4->dst_len = frh->dst_len;
+	rule4->dstmask = inet_make_mask(rule4->dst_len);
+	rule4->tos = frh->tos;
 
-	}
-}
-
-/* callers should hold rtnl semaphore */
-
-static void fib_rules_attach(struct net_device *dev)
-{
-	struct hlist_node *node;
-	struct fib_rule *r;
-
-	hlist_for_each_entry(r, node, &fib_rules, hlist) {
-		if (r->r_ifindex == -1 && strcmp(dev->name, r->r_ifname) == 0)
-			r->r_ifindex = dev->ifindex;
-	}
+	err = 0;
+errout:
+	return err;
 }
 
-int fib_lookup(const struct flowi *flp, struct fib_result *res)
+static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
+			     struct nlattr **tb)
 {
-	int err;
-	struct fib_rule *r, *policy;
-	struct fib_table *tb;
-	struct hlist_node *node;
+	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
-	u32 daddr = flp->fl4_dst;
-	u32 saddr = flp->fl4_src;
+	if (frh->src_len && (rule4->src_len != frh->src_len))
+		return 0;
 
-FRprintk("Lookup: %u.%u.%u.%u <- %u.%u.%u.%u ",
-	NIPQUAD(flp->fl4_dst), NIPQUAD(flp->fl4_src));
+	if (frh->dst_len && (rule4->dst_len != frh->dst_len))
+		return 0;
 
-	rcu_read_lock();
+	if (frh->tos && (rule4->tos != frh->tos))
+		return 0;
 
-	hlist_for_each_entry_rcu(r, node, &fib_rules, hlist) {
-		if (((saddr^r->r_src) & r->r_srcmask) ||
-		    ((daddr^r->r_dst) & r->r_dstmask) ||
-		    (r->r_tos && r->r_tos != flp->fl4_tos) ||
 #ifdef CONFIG_IP_ROUTE_FWMARK
-		    (r->r_fwmark && r->r_fwmark != flp->fl4_fwmark) ||
+	if (tb[FRA_FWMARK] && (rule4->fwmark != nla_get_u32(tb[FRA_FWMARK])))
+		return 0;
 #endif
-		    (r->r_ifindex && r->r_ifindex != flp->iif))
-			continue;
-
-FRprintk("tb %d r %d ", r->r_table, r->r_action);
-		switch (r->r_action) {
-		case RTN_UNICAST:
-			policy = r;
-			break;
-		case RTN_UNREACHABLE:
-			rcu_read_unlock();
-			return -ENETUNREACH;
-		default:
-		case RTN_BLACKHOLE:
-			rcu_read_unlock();
-			return -EINVAL;
-		case RTN_PROHIBIT:
-			rcu_read_unlock();
-			return -EACCES;
-		}
 
-		if ((tb = fib_get_table(r->r_table)) == NULL)
-			continue;
-		err = tb->tb_lookup(tb, flp, res);
-		if (err == 0) {
-			res->r = policy;
-			if (policy)
-				atomic_inc(&policy->r_clntref);
-			rcu_read_unlock();
-			return 0;
-		}
-		if (err < 0 && err != -EAGAIN) {
-			rcu_read_unlock();
-			return err;
-		}
-	}
-FRprintk("FAILURE\n");
-	rcu_read_unlock();
-	return -ENETUNREACH;
-}
+#ifdef CONFIG_NET_CLS_ROUTE
+	if (tb[FRA_FLOW] && (rule4->tclassid != nla_get_u32(tb[FRA_FLOW])))
+		return 0;
+#endif
 
-void fib_select_default(const struct flowi *flp, struct fib_result *res)
-{
-	if (res->r && res->r->r_action == RTN_UNICAST &&
-	    FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK) {
-		struct fib_table *tb;
-		if ((tb = fib_get_table(res->r->r_table)) != NULL)
-			tb->tb_select_default(tb, flp, res);
-	}
-}
+	if (tb[FRA_SRC] && (rule4->src != nla_get_u32(tb[FRA_SRC])))
+		return 0;
 
-static int fib_rules_event(struct notifier_block *this, unsigned long event, void *ptr)
-{
-	struct net_device *dev = ptr;
+	if (tb[FRA_DST] && (rule4->dst != nla_get_u32(tb[FRA_DST])))
+		return 0;
 
-	if (event == NETDEV_UNREGISTER)
-		fib_rules_detach(dev);
-	else if (event == NETDEV_REGISTER)
-		fib_rules_attach(dev);
-	return NOTIFY_DONE;
+	return 1;
 }
 
+static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
+			  struct nlmsghdr *nlh, struct fib_rule_hdr *frh)
+{
+	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
-static struct notifier_block fib_rules_notifier = {
-	.notifier_call =fib_rules_event,
-};
+	frh->family = AF_INET;
+	frh->dst_len = rule4->dst_len;
+	frh->src_len = rule4->src_len;
+	frh->tos = rule4->tos;
 
-static __inline__ int inet_fill_rule(struct sk_buff *skb,
-				     struct fib_rule *r,
-				     u32 pid, u32 seq, int event,
-				     unsigned int flags)
-{
-	struct rtmsg *rtm;
-	struct nlmsghdr  *nlh;
-	unsigned char	 *b = skb->tail;
-
-	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*rtm), flags);
-	rtm = NLMSG_DATA(nlh);
-	rtm->rtm_family = AF_INET;
-	rtm->rtm_dst_len = r->r_dst_len;
-	rtm->rtm_src_len = r->r_src_len;
-	rtm->rtm_tos = r->r_tos;
 #ifdef CONFIG_IP_ROUTE_FWMARK
-	if (r->r_fwmark)
-		RTA_PUT(skb, RTA_PROTOINFO, 4, &r->r_fwmark);
+	if (rule4->fwmark)
+		NLA_PUT_U32(skb, FRA_FWMARK, rule4->fwmark);
 #endif
-	rtm->rtm_table = r->r_table;
-	rtm->rtm_protocol = 0;
-	rtm->rtm_scope = 0;
-	rtm->rtm_type = r->r_action;
-	rtm->rtm_flags = r->r_flags;
-
-	if (r->r_dst_len)
-		RTA_PUT(skb, RTA_DST, 4, &r->r_dst);
-	if (r->r_src_len)
-		RTA_PUT(skb, RTA_SRC, 4, &r->r_src);
-	if (r->r_ifname[0])
-		RTA_PUT(skb, RTA_IIF, IFNAMSIZ, &r->r_ifname);
-	if (r->r_preference)
-		RTA_PUT(skb, RTA_PRIORITY, 4, &r->r_preference);
-	if (r->r_srcmap)
-		RTA_PUT(skb, RTA_GATEWAY, 4, &r->r_srcmap);
+
+	if (rule4->dst_len)
+		NLA_PUT_U32(skb, FRA_DST, rule4->dst);
+
+	if (rule4->src_len)
+		NLA_PUT_U32(skb, FRA_SRC, rule4->src);
+
 #ifdef CONFIG_NET_CLS_ROUTE
-	if (r->r_tclassid)
-		RTA_PUT(skb, RTA_FLOW, 4, &r->r_tclassid);
+	if (rule4->tclassid)
+		NLA_PUT_U32(skb, FRA_FLOW, rule4->tclassid);
 #endif
-	nlh->nlmsg_len = skb->tail - b;
-	return skb->len;
+	return 0;
 
-nlmsg_failure:
-rtattr_failure:
-	skb_trim(skb, b - skb->data);
-	return -1;
+nla_put_failure:
+	return -ENOBUFS;
 }
 
-/* callers should hold rtnl semaphore */
+int fib4_rules_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	return fib_rules_dump(skb, cb, AF_INET);
+}
 
-static void rtmsg_rule(int event, struct fib_rule *r)
+static u32 fib4_rule_default_pref(void)
 {
-	int size = NLMSG_SPACE(sizeof(struct rtmsg) + 128);
-	struct sk_buff *skb = alloc_skb(size, GFP_KERNEL);
-
-	if (!skb)
-		netlink_set_err(rtnl, 0, RTNLGRP_IPV4_RULE, ENOBUFS);
-	else if (inet_fill_rule(skb, r, 0, 0, event, 0) < 0) {
-		kfree_skb(skb);
-		netlink_set_err(rtnl, 0, RTNLGRP_IPV4_RULE, EINVAL);
-	} else {
-		netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV4_RULE, GFP_KERNEL);
+	struct list_head *pos;
+	struct fib_rule *rule;
+
+	if (!list_empty(&fib4_rules)) {
+		pos = fib4_rules.next;
+		if (pos->next != &fib4_rules) {
+			rule = list_entry(pos->next, struct fib_rule, list);
+			if (rule->pref)
+				return rule->pref - 1;
+		}
 	}
+
+	return 0;
 }
 
-int inet_dump_rules(struct sk_buff *skb, struct netlink_callback *cb)
+static struct fib_rules_ops fib4_rules_ops = {
+	.family		= AF_INET,
+	.rule_size	= sizeof(struct fib4_rule),
+	.action		= fib4_rule_action,
+	.match		= fib4_rule_match,
+	.configure	= fib4_rule_configure,
+	.compare	= fib4_rule_compare,
+	.fill		= fib4_rule_fill,
+	.default_pref	= fib4_rule_default_pref,
+	.nlgroup	= RTNLGRP_IPV4_RULE,
+	.policy		= fib4_rule_policy,
+	.rules_list	= &fib4_rules,
+	.owner		= THIS_MODULE,
+};
+
+void __init fib4_rules_init(void)
 {
-	int idx = 0;
-	int s_idx = cb->args[0];
-	struct fib_rule *r;
-	struct hlist_node *node;
-
-	rcu_read_lock();
-	hlist_for_each_entry(r, node, &fib_rules, hlist) {
-		if (idx < s_idx)
-			goto next;
-		if (inet_fill_rule(skb, r, NETLINK_CB(cb->skb).pid,
-				   cb->nlh->nlmsg_seq,
-				   RTM_NEWRULE, NLM_F_MULTI) < 0)
-			break;
-next:
-		idx++;
-	}
-	rcu_read_unlock();
-	cb->args[0] = idx;
+	list_add_tail(&local_rule.common.list, &fib4_rules);
+	list_add_tail(&main_rule.common.list, &fib4_rules);
+	list_add_tail(&default_rule.common.list, &fib4_rules);
 
-	return skb->len;
+	fib_rules_register(&fib4_rules_ops);
 }
 
-void __init fib_rules_init(void)
+void __exit fib4_rules_cleanup(void)
 {
-	INIT_HLIST_HEAD(&fib_rules);
-	hlist_add_head(&local_rule.hlist, &fib_rules);
-	hlist_add_after(&local_rule.hlist, &main_rule.hlist);
-	hlist_add_after(&main_rule.hlist, &default_rule.hlist);
-	register_netdevice_notifier(&fib_rules_notifier);
+	fib_rules_unregister(&fib4_rules_ops);
 }
-- 
cgit v1.2.3


From fe4944e59c357f945f81bc67edb7ed1392e875ad Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 23:03:05 -0700
Subject: [NETLINK]: Extend netlink messaging interface

Adds:
 nlmsg_get_pos()                 return current position in message
 nlmsg_trim()                    trim part of message
 nla_reserve_nohdr(skb, len)     reserve room for an attribute w/o hdr
 nla_put_nohdr(skb, len, data)   add attribute w/o hdr
 nla_find_nested()               find attribute in nested attributes

Fixes nlmsg_new() to take allocation flags and consider size.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h   | 74 +++++++++++++++++++++++++++++++++++++++++-------
 kernel/taskstats.c      |  2 +-
 net/netlink/attr.c      | 75 +++++++++++++++++++++++++++++++++++++++++++++++++
 net/netlink/genetlink.c |  2 +-
 4 files changed, 141 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 640c26a90cf1..3a5e40b1e045 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -35,6 +35,8 @@
  *   nlmsg_put()			add a netlink message to an skb
  *   nlmsg_put_answer()			callback based nlmsg_put()
  *   nlmsg_end()			finanlize netlink message
+ *   nlmsg_get_pos()			return current position in message
+ *   nlmsg_trim()			trim part of message
  *   nlmsg_cancel()			cancel message construction
  *   nlmsg_free()			free a netlink message
  *
@@ -80,8 +82,10 @@
  *   struct nlattr			netlink attribtue header
  *
  * Attribute Construction:
- *   nla_reserve(skb, type, len)	reserve skb tailroom for an attribute
+ *   nla_reserve(skb, type, len)	reserve room for an attribute
+ *   nla_reserve_nohdr(skb, len)	reserve room for an attribute w/o hdr
  *   nla_put(skb, type, len, data)	add attribute to skb
+ *   nla_put_nohdr(skb, len, data)	add attribute w/o hdr
  *
  * Attribute Construction for Basic Types:
  *   nla_put_u8(skb, type, value)	add u8 attribute to skb
@@ -139,6 +143,7 @@
  *   nla_next(nla, remaining)		get next netlink attribute
  *   nla_validate()			validate a stream of attributes
  *   nla_find()				find attribute in stream of attributes
+ *   nla_find_nested()			find attribute in nested attributes
  *   nla_parse()			parse and validate stream of attrs
  *   nla_parse_nested()			parse nested attribuets
  *   nla_for_each_attr()		loop over all attributes
@@ -203,12 +208,18 @@ extern int		nla_memcmp(const struct nlattr *nla, const void *data,
 extern int		nla_strcmp(const struct nlattr *nla, const char *str);
 extern struct nlattr *	__nla_reserve(struct sk_buff *skb, int attrtype,
 				      int attrlen);
+extern void *		__nla_reserve_nohdr(struct sk_buff *skb, int attrlen);
 extern struct nlattr *	nla_reserve(struct sk_buff *skb, int attrtype,
 				    int attrlen);
+extern void *		nla_reserve_nohdr(struct sk_buff *skb, int attrlen);
 extern void		__nla_put(struct sk_buff *skb, int attrtype,
 				  int attrlen, const void *data);
+extern void		__nla_put_nohdr(struct sk_buff *skb, int attrlen,
+					const void *data);
 extern int		nla_put(struct sk_buff *skb, int attrtype,
 				int attrlen, const void *data);
+extern int		nla_put_nohdr(struct sk_buff *skb, int attrlen,
+				      const void *data);
 
 /**************************************************************************
  * Netlink Messages
@@ -453,12 +464,13 @@ static inline struct nlmsghdr *nlmsg_put_answer(struct sk_buff *skb,
 /**
  * nlmsg_new - Allocate a new netlink message
  * @size: maximum size of message
+ * @flags: the type of memory to allocate.
  *
  * Use NLMSG_GOODSIZE if size isn't know and you need a good default size.
  */
-static inline struct sk_buff *nlmsg_new(int size)
+static inline struct sk_buff *nlmsg_new(int size, gfp_t flags)
 {
-	return alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	return alloc_skb(size, flags);
 }
 
 /**
@@ -479,6 +491,32 @@ static inline int nlmsg_end(struct sk_buff *skb, struct nlmsghdr *nlh)
 	return skb->len;
 }
 
+/**
+ * nlmsg_get_pos - return current position in netlink message
+ * @skb: socket buffer the message is stored in
+ *
+ * Returns a pointer to the current tail of the message.
+ */
+static inline void *nlmsg_get_pos(struct sk_buff *skb)
+{
+	return skb->tail;
+}
+
+/**
+ * nlmsg_trim - Trim message to a mark
+ * @skb: socket buffer the message is stored in
+ * @mark: mark to trim to
+ *
+ * Trims the message to the provided mark. Returns -1.
+ */
+static inline int nlmsg_trim(struct sk_buff *skb, void *mark)
+{
+	if (mark)
+		skb_trim(skb, (unsigned char *) mark - skb->data);
+
+	return -1;
+}
+
 /**
  * nlmsg_cancel - Cancel construction of a netlink message
  * @skb: socket buffer the message is stored in
@@ -489,9 +527,7 @@ static inline int nlmsg_end(struct sk_buff *skb, struct nlmsghdr *nlh)
  */
 static inline int nlmsg_cancel(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
-	skb_trim(skb, (unsigned char *) nlh - skb->data);
-
-	return -1;
+	return nlmsg_trim(skb, nlh);
 }
 
 /**
@@ -630,6 +666,18 @@ static inline struct nlattr *nla_next(const struct nlattr *nla, int *remaining)
 	return (struct nlattr *) ((char *) nla + totlen);
 }
 
+/**
+ * nla_find_nested - find attribute in a set of nested attributes
+ * @nla: attribute containing the nested attributes
+ * @attrtype: type of attribute to look for
+ *
+ * Returns the first attribute which matches the specified type.
+ */
+static inline struct nlattr *nla_find_nested(struct nlattr *nla, int attrtype)
+{
+	return nla_find(nla_data(nla), nla_len(nla), attrtype);
+}
+
 /**
  * nla_parse_nested - parse nested attributes
  * @tb: destination array with maxtype+1 elements
@@ -862,10 +910,7 @@ static inline int nla_nest_end(struct sk_buff *skb, struct nlattr *start)
  */
 static inline int nla_nest_cancel(struct sk_buff *skb, struct nlattr *start)
 {
-	if (start)
-		skb_trim(skb, (unsigned char *) start - skb->data);
-
-	return -1;
+	return nlmsg_trim(skb, start);
 }
 
 /**
@@ -880,4 +925,13 @@ static inline int nla_nest_cancel(struct sk_buff *skb, struct nlattr *start)
 	     nla_ok(pos, rem); \
 	     pos = nla_next(pos, &(rem)))
 
+/**
+ * nla_for_each_nested - iterate over nested attributes
+ * @pos: loop counter, set to current attribute
+ * @nla: attribute containing the nested attributes
+ * @rem: initialized to len, holds bytes currently remaining in stream
+ */
+#define nla_for_each_nested(pos, nla, rem) \
+	nla_for_each_attr(pos, nla_data(nla), nla_len(nla), rem)
+
 #endif
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index e78187657330..2ed4040d0dc5 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -75,7 +75,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp,
 	/*
 	 * If new attributes are added, please revisit this allocation
 	 */
-	skb = nlmsg_new(size);
+	skb = nlmsg_new(size, GFP_KERNEL);
 	if (!skb)
 		return -ENOMEM;
 
diff --git a/net/netlink/attr.c b/net/netlink/attr.c
index dddbd15135a8..136e529e5780 100644
--- a/net/netlink/attr.c
+++ b/net/netlink/attr.c
@@ -254,6 +254,26 @@ struct nlattr *__nla_reserve(struct sk_buff *skb, int attrtype, int attrlen)
 	return nla;
 }
 
+/**
+ * __nla_reserve_nohdr - reserve room for attribute without header
+ * @skb: socket buffer to reserve room on
+ * @attrlen: length of attribute payload
+ *
+ * Reserves room for attribute payload without a header.
+ *
+ * The caller is responsible to ensure that the skb provides enough
+ * tailroom for the payload.
+ */
+void *__nla_reserve_nohdr(struct sk_buff *skb, int attrlen)
+{
+	void *start;
+
+	start = skb_put(skb, NLA_ALIGN(attrlen));
+	memset(start, 0, NLA_ALIGN(attrlen));
+
+	return start;
+}
+
 /**
  * nla_reserve - reserve room for attribute on the skb
  * @skb: socket buffer to reserve room on
@@ -274,6 +294,24 @@ struct nlattr *nla_reserve(struct sk_buff *skb, int attrtype, int attrlen)
 	return __nla_reserve(skb, attrtype, attrlen);
 }
 
+/**
+ * nla_reserve - reserve room for attribute without header
+ * @skb: socket buffer to reserve room on
+ * @len: length of attribute payload
+ *
+ * Reserves room for attribute payload without a header.
+ *
+ * Returns NULL if the tailroom of the skb is insufficient to store
+ * the attribute payload.
+ */
+void *nla_reserve_nohdr(struct sk_buff *skb, int attrlen)
+{
+	if (unlikely(skb_tailroom(skb) < NLA_ALIGN(attrlen)))
+		return NULL;
+
+	return __nla_reserve_nohdr(skb, attrlen);
+}
+
 /**
  * __nla_put - Add a netlink attribute to a socket buffer
  * @skb: socket buffer to add attribute to
@@ -293,6 +331,22 @@ void __nla_put(struct sk_buff *skb, int attrtype, int attrlen,
 	memcpy(nla_data(nla), data, attrlen);
 }
 
+/**
+ * __nla_put_nohdr - Add a netlink attribute without header
+ * @skb: socket buffer to add attribute to
+ * @attrlen: length of attribute payload
+ * @data: head of attribute payload
+ *
+ * The caller is responsible to ensure that the skb provides enough
+ * tailroom for the attribute payload.
+ */
+void __nla_put_nohdr(struct sk_buff *skb, int attrlen, const void *data)
+{
+	void *start;
+
+	start = __nla_reserve_nohdr(skb, attrlen);
+	memcpy(start, data, attrlen);
+}
 
 /**
  * nla_put - Add a netlink attribute to a socket buffer
@@ -313,15 +367,36 @@ int nla_put(struct sk_buff *skb, int attrtype, int attrlen, const void *data)
 	return 0;
 }
 
+/**
+ * nla_put_nohdr - Add a netlink attribute without header
+ * @skb: socket buffer to add attribute to
+ * @attrlen: length of attribute payload
+ * @data: head of attribute payload
+ *
+ * Returns -1 if the tailroom of the skb is insufficient to store
+ * the attribute payload.
+ */
+int nla_put_nohdr(struct sk_buff *skb, int attrlen, const void *data)
+{
+	if (unlikely(skb_tailroom(skb) < NLA_ALIGN(attrlen)))
+		return -1;
+
+	__nla_put_nohdr(skb, attrlen, data);
+	return 0;
+}
 
 EXPORT_SYMBOL(nla_validate);
 EXPORT_SYMBOL(nla_parse);
 EXPORT_SYMBOL(nla_find);
 EXPORT_SYMBOL(nla_strlcpy);
 EXPORT_SYMBOL(__nla_reserve);
+EXPORT_SYMBOL(__nla_reserve_nohdr);
 EXPORT_SYMBOL(nla_reserve);
+EXPORT_SYMBOL(nla_reserve_nohdr);
 EXPORT_SYMBOL(__nla_put);
+EXPORT_SYMBOL(__nla_put_nohdr);
 EXPORT_SYMBOL(nla_put);
+EXPORT_SYMBOL(nla_put_nohdr);
 EXPORT_SYMBOL(nla_memcpy);
 EXPORT_SYMBOL(nla_memcmp);
 EXPORT_SYMBOL(nla_strcmp);
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index a298f77cc3e3..75bb47a898dd 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -440,7 +440,7 @@ static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid,
 	struct sk_buff *skb;
 	int err;
 
-	skb = nlmsg_new(NLMSG_GOODSIZE);
+	skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (skb == NULL)
 		return ERR_PTR(-ENOBUFS);
 
-- 
cgit v1.2.3


From 1823730fbc89fadde72a7bb3b7bdf03cc7b8835c Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 23:04:54 -0700
Subject: [IPv4]: Move interface address bits to linux/if_addr.h

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_addr.h   | 53 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/rtnetlink.h | 56 -----------------------------------------------
 net/core/rtnetlink.c      |  1 +
 net/decnet/dn_dev.c       |  1 +
 net/ipv4/devinet.c        |  1 +
 net/ipv4/fib_frontend.c   |  1 +
 net/ipv6/addrconf.c       |  1 +
 net/ipv6/ndisc.c          |  1 +
 8 files changed, 59 insertions(+), 56 deletions(-)
 create mode 100644 include/linux/if_addr.h

(limited to 'include')

diff --git a/include/linux/if_addr.h b/include/linux/if_addr.h
new file mode 100644
index 000000000000..e1590454db59
--- /dev/null
+++ b/include/linux/if_addr.h
@@ -0,0 +1,53 @@
+#ifndef __LINUX_IF_ADDR_H
+#define __LINUX_IF_ADDR_H
+
+#include <linux/netlink.h>
+
+struct ifaddrmsg
+{
+	__u8		ifa_family;
+	__u8		ifa_prefixlen;	/* The prefix length		*/
+	__u8		ifa_flags;	/* Flags			*/
+	__u8		ifa_scope;	/* Address scope		*/
+	__u32		ifa_index;	/* Link index			*/
+};
+
+/*
+ * Important comment:
+ * IFA_ADDRESS is prefix address, rather than local interface address.
+ * It makes no difference for normally configured broadcast interfaces,
+ * but for point-to-point IFA_ADDRESS is DESTINATION address,
+ * local address is supplied in IFA_LOCAL attribute.
+ */
+enum
+{
+	IFA_UNSPEC,
+	IFA_ADDRESS,
+	IFA_LOCAL,
+	IFA_LABEL,
+	IFA_BROADCAST,
+	IFA_ANYCAST,
+	IFA_CACHEINFO,
+	IFA_MULTICAST,
+	__IFA_MAX,
+};
+
+#define IFA_MAX (__IFA_MAX - 1)
+
+/* ifa_flags */
+#define IFA_F_SECONDARY		0x01
+#define IFA_F_TEMPORARY		IFA_F_SECONDARY
+
+#define IFA_F_DEPRECATED	0x20
+#define IFA_F_TENTATIVE		0x40
+#define IFA_F_PERMANENT		0x80
+
+struct ifa_cacheinfo
+{
+	__u32	ifa_prefered;
+	__u32	ifa_valid;
+	__u32	cstamp; /* created timestamp, hundredths of seconds */
+	__u32	tstamp; /* updated timestamp, hundredths of seconds */
+};
+
+#endif
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index bf353538ae93..890c4d4038b6 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -384,62 +384,6 @@ struct rta_session
 };
 
 
-/*********************************************************
- *		Interface address.
- ****/
-
-struct ifaddrmsg
-{
-	unsigned char	ifa_family;
-	unsigned char	ifa_prefixlen;	/* The prefix length		*/
-	unsigned char	ifa_flags;	/* Flags			*/
-	unsigned char	ifa_scope;	/* See above			*/
-	int		ifa_index;	/* Link index			*/
-};
-
-enum
-{
-	IFA_UNSPEC,
-	IFA_ADDRESS,
-	IFA_LOCAL,
-	IFA_LABEL,
-	IFA_BROADCAST,
-	IFA_ANYCAST,
-	IFA_CACHEINFO,
-	IFA_MULTICAST,
-	__IFA_MAX
-};
-
-#define IFA_MAX (__IFA_MAX - 1)
-
-/* ifa_flags */
-
-#define IFA_F_SECONDARY		0x01
-#define IFA_F_TEMPORARY		IFA_F_SECONDARY
-
-#define IFA_F_DEPRECATED	0x20
-#define IFA_F_TENTATIVE		0x40
-#define IFA_F_PERMANENT		0x80
-
-struct ifa_cacheinfo
-{
-	__u32	ifa_prefered;
-	__u32	ifa_valid;
-	__u32	cstamp; /* created timestamp, hundredths of seconds */
-	__u32	tstamp; /* updated timestamp, hundredths of seconds */
-};
-
-
-#define IFA_RTA(r)  ((struct rtattr*)(((char*)(r)) + NLMSG_ALIGN(sizeof(struct ifaddrmsg))))
-#define IFA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ifaddrmsg))
-
-/*
-   Important comment:
-   IFA_ADDRESS is prefix address, rather than local interface address.
-   It makes no difference for normally configured broadcast interfaces,
-   but for point-to-point IFA_ADDRESS is DESTINATION address,
-   local address is supplied in IFA_LOCAL attribute.
- */
 
 /**************************************************************
  *		Neighbour discovery.
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index aa7cff2257b1..35712031e2c3 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -35,6 +35,7 @@
 #include <linux/init.h>
 #include <linux/security.h>
 #include <linux/mutex.h>
+#include <linux/if_addr.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c
index 476455fbdb03..632c5a90b589 100644
--- a/net/decnet/dn_dev.c
+++ b/net/decnet/dn_dev.c
@@ -34,6 +34,7 @@
 #include <linux/seq_file.h>
 #include <linux/timer.h>
 #include <linux/string.h>
+#include <linux/if_addr.h>
 #include <linux/if_arp.h>
 #include <linux/if_ether.h>
 #include <linux/skbuff.h>
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 80bf5b2ea2e6..398e7b9ca66b 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -43,6 +43,7 @@
 #include <linux/in.h>
 #include <linux/errno.h>
 #include <linux/interrupt.h>
+#include <linux/if_addr.h>
 #include <linux/if_ether.h>
 #include <linux/inet.h>
 #include <linux/netdevice.h>
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index fe4a53d4d10d..a83f1aa8034e 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -32,6 +32,7 @@
 #include <linux/inet.h>
 #include <linux/inetdevice.h>
 #include <linux/netdevice.h>
+#include <linux/if_addr.h>
 #include <linux/if_arp.h>
 #include <linux/skbuff.h>
 #include <linux/netlink.h>
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index ed766eebc022..c2a4db843e51 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -48,6 +48,7 @@
 #include <linux/net.h>
 #include <linux/in6.h>
 #include <linux/netdevice.h>
+#include <linux/if_addr.h>
 #include <linux/if_arp.h>
 #include <linux/if_arcnet.h>
 #include <linux/if_infiniband.h>
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 67cfc3813c32..5743e8bffefd 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -62,6 +62,7 @@
 #include <linux/sysctl.h>
 #endif
 
+#include <linux/if_addr.h>
 #include <linux/if_arp.h>
 #include <linux/ipv6.h>
 #include <linux/icmpv6.h>
-- 
cgit v1.2.3


From 0844565fb8a9418f5a860aa480c1aef70319c9a2 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Fri, 4 Aug 2006 23:05:56 -0700
Subject: [NET]: Move netlink interface bits to linux/if.h

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if.h        | 129 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/rtnetlink.h | 133 +---------------------------------------------
 2 files changed, 130 insertions(+), 132 deletions(-)

(limited to 'include')

diff --git a/include/linux/if.h b/include/linux/if.h
index 374e20ad8b0d..cd080d765324 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -212,5 +212,134 @@ struct ifconf
 #define	ifc_buf	ifc_ifcu.ifcu_buf		/* buffer address	*/
 #define	ifc_req	ifc_ifcu.ifcu_req		/* array of structures	*/
 
+/* The struct should be in sync with struct net_device_stats */
+struct rtnl_link_stats
+{
+	__u32	rx_packets;		/* total packets received	*/
+	__u32	tx_packets;		/* total packets transmitted	*/
+	__u32	rx_bytes;		/* total bytes received 	*/
+	__u32	tx_bytes;		/* total bytes transmitted	*/
+	__u32	rx_errors;		/* bad packets received		*/
+	__u32	tx_errors;		/* packet transmit problems	*/
+	__u32	rx_dropped;		/* no space in linux buffers	*/
+	__u32	tx_dropped;		/* no space available in linux	*/
+	__u32	multicast;		/* multicast packets received	*/
+	__u32	collisions;
+
+	/* detailed rx_errors: */
+	__u32	rx_length_errors;
+	__u32	rx_over_errors;		/* receiver ring buff overflow	*/
+	__u32	rx_crc_errors;		/* recved pkt with crc error	*/
+	__u32	rx_frame_errors;	/* recv'd frame alignment error */
+	__u32	rx_fifo_errors;		/* recv'r fifo overrun		*/
+	__u32	rx_missed_errors;	/* receiver missed packet	*/
+
+	/* detailed tx_errors */
+	__u32	tx_aborted_errors;
+	__u32	tx_carrier_errors;
+	__u32	tx_fifo_errors;
+	__u32	tx_heartbeat_errors;
+	__u32	tx_window_errors;
+
+	/* for cslip etc */
+	__u32	rx_compressed;
+	__u32	tx_compressed;
+};
+
+/* The struct should be in sync with struct ifmap */
+struct rtnl_link_ifmap
+{
+	__u64	mem_start;
+	__u64	mem_end;
+	__u64	base_addr;
+	__u16	irq;
+	__u8	dma;
+	__u8	port;
+};
+
+enum
+{
+	IFLA_UNSPEC,
+	IFLA_ADDRESS,
+	IFLA_BROADCAST,
+	IFLA_IFNAME,
+	IFLA_MTU,
+	IFLA_LINK,
+	IFLA_QDISC,
+	IFLA_STATS,
+	IFLA_COST,
+#define IFLA_COST IFLA_COST
+	IFLA_PRIORITY,
+#define IFLA_PRIORITY IFLA_PRIORITY
+	IFLA_MASTER,
+#define IFLA_MASTER IFLA_MASTER
+	IFLA_WIRELESS,		/* Wireless Extension event - see wireless.h */
+#define IFLA_WIRELESS IFLA_WIRELESS
+	IFLA_PROTINFO,		/* Protocol specific information for a link */
+#define IFLA_PROTINFO IFLA_PROTINFO
+	IFLA_TXQLEN,
+#define IFLA_TXQLEN IFLA_TXQLEN
+	IFLA_MAP,
+#define IFLA_MAP IFLA_MAP
+	IFLA_WEIGHT,
+#define IFLA_WEIGHT IFLA_WEIGHT
+	IFLA_OPERSTATE,
+	IFLA_LINKMODE,
+	__IFLA_MAX
+};
+
+
+#define IFLA_MAX (__IFLA_MAX - 1)
+
+/* ifi_flags.
+
+   IFF_* flags.
+
+   The only change is:
+   IFF_LOOPBACK, IFF_BROADCAST and IFF_POINTOPOINT are
+   more not changeable by user. They describe link media
+   characteristics and set by device driver.
+
+   Comments:
+   - Combination IFF_BROADCAST|IFF_POINTOPOINT is invalid
+   - If neither of these three flags are set;
+     the interface is NBMA.
+
+   - IFF_MULTICAST does not mean anything special:
+   multicasts can be used on all not-NBMA links.
+   IFF_MULTICAST means that this media uses special encapsulation
+   for multicast frames. Apparently, all IFF_POINTOPOINT and
+   IFF_BROADCAST devices are able to use multicasts too.
+ */
+
+/* IFLA_LINK.
+   For usual devices it is equal ifi_index.
+   If it is a "virtual interface" (f.e. tunnel), ifi_link
+   can point to real physical interface (f.e. for bandwidth calculations),
+   or maybe 0, what means, that real media is unknown (usual
+   for IPIP tunnels, when route to endpoint is allowed to change)
+ */
+
+/* Subtype attributes for IFLA_PROTINFO */
+enum
+{
+	IFLA_INET6_UNSPEC,
+	IFLA_INET6_FLAGS,	/* link flags			*/
+	IFLA_INET6_CONF,	/* sysctl parameters		*/
+	IFLA_INET6_STATS,	/* statistics			*/
+	IFLA_INET6_MCAST,	/* MC things. What of them?	*/
+	IFLA_INET6_CACHEINFO,	/* time values and max reasm size */
+	__IFLA_INET6_MAX
+};
+
+#define IFLA_INET6_MAX	(__IFLA_INET6_MAX - 1)
+
+struct ifla_cacheinfo
+{
+	__u32	max_reasm_len;
+	__u32	tstamp;		/* ipv6InterfaceTable updated timestamp */
+	__u32	reachable_time;
+	__u32	retrans_time;
+};
 
 #endif /* _LINUX_IF_H */
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 890c4d4038b6..84f3eb426da2 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -2,6 +2,7 @@
 #define __LINUX_RTNETLINK_H
 
 #include <linux/netlink.h>
+#include <linux/if.h>
 
 /****
  *		Routing/neighbour discovery messages.
@@ -607,138 +608,6 @@ struct prefix_cacheinfo
 	__u32	valid_time;
 };
 
-/* The struct should be in sync with struct net_device_stats */
-struct rtnl_link_stats
-{
-	__u32	rx_packets;		/* total packets received	*/
-	__u32	tx_packets;		/* total packets transmitted	*/
-	__u32	rx_bytes;		/* total bytes received 	*/
-	__u32	tx_bytes;		/* total bytes transmitted	*/
-	__u32	rx_errors;		/* bad packets received		*/
-	__u32	tx_errors;		/* packet transmit problems	*/
-	__u32	rx_dropped;		/* no space in linux buffers	*/
-	__u32	tx_dropped;		/* no space available in linux	*/
-	__u32	multicast;		/* multicast packets received	*/
-	__u32	collisions;
-
-	/* detailed rx_errors: */
-	__u32	rx_length_errors;
-	__u32	rx_over_errors;		/* receiver ring buff overflow	*/
-	__u32	rx_crc_errors;		/* recved pkt with crc error	*/
-	__u32	rx_frame_errors;	/* recv'd frame alignment error */
-	__u32	rx_fifo_errors;		/* recv'r fifo overrun		*/
-	__u32	rx_missed_errors;	/* receiver missed packet	*/
-
-	/* detailed tx_errors */
-	__u32	tx_aborted_errors;
-	__u32	tx_carrier_errors;
-	__u32	tx_fifo_errors;
-	__u32	tx_heartbeat_errors;
-	__u32	tx_window_errors;
-	
-	/* for cslip etc */
-	__u32	rx_compressed;
-	__u32	tx_compressed;
-};
-
-/* The struct should be in sync with struct ifmap */
-struct rtnl_link_ifmap
-{
-	__u64	mem_start;
-	__u64	mem_end;
-	__u64	base_addr;
-	__u16	irq;
-	__u8	dma;
-	__u8	port;
-};
-
-enum
-{
-	IFLA_UNSPEC,
-	IFLA_ADDRESS,
-	IFLA_BROADCAST,
-	IFLA_IFNAME,
-	IFLA_MTU,
-	IFLA_LINK,
-	IFLA_QDISC,
-	IFLA_STATS,
-	IFLA_COST,
-#define IFLA_COST IFLA_COST
-	IFLA_PRIORITY,
-#define IFLA_PRIORITY IFLA_PRIORITY
-	IFLA_MASTER,
-#define IFLA_MASTER IFLA_MASTER
-	IFLA_WIRELESS,		/* Wireless Extension event - see wireless.h */
-#define IFLA_WIRELESS IFLA_WIRELESS
-	IFLA_PROTINFO,		/* Protocol specific information for a link */
-#define IFLA_PROTINFO IFLA_PROTINFO
-	IFLA_TXQLEN,
-#define IFLA_TXQLEN IFLA_TXQLEN
-	IFLA_MAP,
-#define IFLA_MAP IFLA_MAP
-	IFLA_WEIGHT,
-#define IFLA_WEIGHT IFLA_WEIGHT
-	IFLA_OPERSTATE,
-	IFLA_LINKMODE,
-	__IFLA_MAX
-};
-
-
-#define IFLA_MAX (__IFLA_MAX - 1)
-
-#define IFLA_RTA(r)  ((struct rtattr*)(((char*)(r)) + NLMSG_ALIGN(sizeof(struct ifinfomsg))))
-#define IFLA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ifinfomsg))
-
-/* ifi_flags.
-
-   IFF_* flags.
-
-   The only change is:
-   IFF_LOOPBACK, IFF_BROADCAST and IFF_POINTOPOINT are
-   more not changeable by user. They describe link media
-   characteristics and set by device driver.
-
-   Comments:
-   - Combination IFF_BROADCAST|IFF_POINTOPOINT is invalid
-   - If neither of these three flags are set;
-     the interface is NBMA.
-
-   - IFF_MULTICAST does not mean anything special:
-   multicasts can be used on all not-NBMA links.
-   IFF_MULTICAST means that this media uses special encapsulation
-   for multicast frames. Apparently, all IFF_POINTOPOINT and
-   IFF_BROADCAST devices are able to use multicasts too.
- */
-
-/* IFLA_LINK.
-   For usual devices it is equal ifi_index.
-   If it is a "virtual interface" (f.e. tunnel), ifi_link
-   can point to real physical interface (f.e. for bandwidth calculations),
-   or maybe 0, what means, that real media is unknown (usual
-   for IPIP tunnels, when route to endpoint is allowed to change)
- */
-
-/* Subtype attributes for IFLA_PROTINFO */
-enum
-{
-	IFLA_INET6_UNSPEC,
-	IFLA_INET6_FLAGS,	/* link flags			*/
-	IFLA_INET6_CONF,	/* sysctl parameters		*/
-	IFLA_INET6_STATS,	/* statistics			*/
-	IFLA_INET6_MCAST,	/* MC things. What of them?	*/
-	IFLA_INET6_CACHEINFO,	/* time values and max reasm size */
-	__IFLA_INET6_MAX
-};
-
-#define IFLA_INET6_MAX	(__IFLA_INET6_MAX - 1)
-
-struct ifla_cacheinfo
-{
-	__u32	max_reasm_len;
-	__u32	tstamp;		/* ipv6InterfaceTable updated timestamp */
-	__u32	reachable_time;
-	__u32	retrans_time;
-};
 
 /*****************************************************************
  *		Traffic control messages.
-- 
cgit v1.2.3


From 84fa7933a33f806bbbaae6775e87459b1ec584c0 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 29 Aug 2006 16:44:56 -0700
Subject: [NET]: Replace CHECKSUM_HW by CHECKSUM_PARTIAL/CHECKSUM_COMPLETE

Replace CHECKSUM_HW by CHECKSUM_PARTIAL (for outgoing packets, whose
checksum still needs to be completed) and CHECKSUM_COMPLETE (for
incoming packets, device supplied full checksum).

Patch originally from Herbert Xu, updated by myself for 2.6.18-rc3.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/atm/he.c                            |  2 +-
 drivers/net/3c59x.c                         |  2 +-
 drivers/net/8139cp.c                        |  6 +++---
 drivers/net/acenic.c                        |  8 ++++----
 drivers/net/bnx2.c                          |  2 +-
 drivers/net/cassini.c                       |  4 ++--
 drivers/net/chelsio/sge.c                   | 10 +++++-----
 drivers/net/dl2k.c                          |  2 +-
 drivers/net/e1000/e1000_main.c              |  8 ++++----
 drivers/net/forcedeth.c                     |  3 ++-
 drivers/net/gianfar.c                       |  2 +-
 drivers/net/hamachi.c                       |  2 +-
 drivers/net/ibm_emac/ibm_emac_core.c        |  2 +-
 drivers/net/ioc3-eth.c                      |  2 +-
 drivers/net/ixgb/ixgb_main.c                |  2 +-
 drivers/net/mv643xx_eth.c                   |  2 +-
 drivers/net/myri10ge/myri10ge.c             |  8 ++++----
 drivers/net/ns83820.c                       |  2 +-
 drivers/net/r8169.c                         |  2 +-
 drivers/net/s2io.c                          |  2 +-
 drivers/net/sk98lin/skge.c                  |  6 +++---
 drivers/net/skge.c                          |  4 ++--
 drivers/net/sky2.c                          |  6 +++---
 drivers/net/starfire.c                      |  6 +++---
 drivers/net/sungem.c                        |  4 ++--
 drivers/net/sunhme.c                        |  6 +++---
 drivers/net/tg3.c                           |  6 +++---
 drivers/net/typhoon.c                       |  2 +-
 drivers/net/via-rhine.c                     |  2 +-
 drivers/net/via-velocity.c                  |  2 +-
 include/linux/netdevice.h                   |  4 ++--
 include/linux/skbuff.h                      | 17 +++++++++--------
 net/core/datagram.c                         |  4 ++--
 net/core/dev.c                              | 12 ++++++------
 net/core/netpoll.c                          |  2 +-
 net/core/skbuff.c                           | 14 +++++++-------
 net/ipv4/icmp.c                             |  2 +-
 net/ipv4/igmp.c                             |  2 +-
 net/ipv4/ip_fragment.c                      |  2 +-
 net/ipv4/ip_gre.c                           |  4 ++--
 net/ipv4/ip_output.c                        |  6 +++---
 net/ipv4/ipvs/ip_vs_proto_tcp.c             |  8 ++++----
 net/ipv4/ipvs/ip_vs_proto_udp.c             |  8 ++++----
 net/ipv4/netfilter.c                        |  2 +-
 net/ipv4/netfilter/ip_conntrack_proto_tcp.c |  3 +--
 net/ipv4/netfilter/ip_conntrack_proto_udp.c |  3 +--
 net/ipv4/netfilter/ip_nat_standalone.c      |  5 +++--
 net/ipv4/netfilter/ip_queue.c               |  6 +++---
 net/ipv4/netfilter/ipt_ECN.c                |  9 +++++----
 net/ipv4/netfilter/ipt_TCPMSS.c             |  5 +++--
 net/ipv4/tcp.c                              |  8 ++++----
 net/ipv4/tcp_ipv4.c                         |  6 +++---
 net/ipv4/tcp_output.c                       | 18 ++++++++----------
 net/ipv4/udp.c                              |  6 +++---
 net/ipv4/xfrm4_output.c                     |  4 ++--
 net/ipv6/exthdrs.c                          |  2 +-
 net/ipv6/icmp.c                             |  2 +-
 net/ipv6/ip6_output.c                       |  2 +-
 net/ipv6/netfilter.c                        |  2 +-
 net/ipv6/netfilter/ip6_queue.c              |  6 +++---
 net/ipv6/netfilter/nf_conntrack_reasm.c     |  6 +++---
 net/ipv6/raw.c                              |  2 +-
 net/ipv6/reassembly.c                       |  6 +++---
 net/ipv6/tcp_ipv6.c                         |  6 +++---
 net/ipv6/udp.c                              |  2 +-
 net/ipv6/xfrm6_output.c                     |  4 ++--
 net/netfilter/nf_conntrack_proto_tcp.c      |  3 +--
 net/netfilter/nf_conntrack_proto_udp.c      |  3 +--
 net/netfilter/nfnetlink_queue.c             |  6 +++---
 net/packet/af_packet.c                      |  2 +-
 net/sched/sch_netem.c                       |  4 ++--
 net/sunrpc/socklib.c                        |  2 +-
 72 files changed, 168 insertions(+), 169 deletions(-)

(limited to 'include')

diff --git a/drivers/atm/he.c b/drivers/atm/he.c
index ffcb9fd31c38..41e052fecd7f 100644
--- a/drivers/atm/he.c
+++ b/drivers/atm/he.c
@@ -1912,7 +1912,7 @@ he_service_rbrq(struct he_dev *he_dev, int group)
 				skb->tail = skb->data + skb->len;
 #ifdef USE_CHECKSUM_HW
 				if (vcc->vpi == 0 && vcc->vci >= ATM_NOT_RSV_VCI) {
-					skb->ip_summed = CHECKSUM_HW;
+					skb->ip_summed = CHECKSUM_COMPLETE;
 					skb->csum = TCP_CKSUM(skb->data,
 							he_vcc->pdu_len);
 				}
diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c
index 80e8ca013e44..29dede2eaa85 100644
--- a/drivers/net/3c59x.c
+++ b/drivers/net/3c59x.c
@@ -2077,7 +2077,7 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	vp->tx_ring[entry].next = 0;
 #if DO_ZEROCOPY
-	if (skb->ip_summed != CHECKSUM_HW)
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
 			vp->tx_ring[entry].status = cpu_to_le32(skb->len | TxIntrUploaded);
 	else
 			vp->tx_ring[entry].status = cpu_to_le32(skb->len | TxIntrUploaded | AddTCPChksum | AddUDPChksum);
diff --git a/drivers/net/8139cp.c b/drivers/net/8139cp.c
index 1428bb7715af..a48b211c489d 100644
--- a/drivers/net/8139cp.c
+++ b/drivers/net/8139cp.c
@@ -813,7 +813,7 @@ static int cp_start_xmit (struct sk_buff *skb, struct net_device *dev)
 
 		if (mss)
 			flags |= LargeSend | ((mss & MSSMask) << MSSShift);
-		else if (skb->ip_summed == CHECKSUM_HW) {
+		else if (skb->ip_summed == CHECKSUM_PARTIAL) {
 			const struct iphdr *ip = skb->nh.iph;
 			if (ip->protocol == IPPROTO_TCP)
 				flags |= IPCS | TCPCS;
@@ -867,7 +867,7 @@ static int cp_start_xmit (struct sk_buff *skb, struct net_device *dev)
 			if (mss)
 				ctrl |= LargeSend |
 					((mss & MSSMask) << MSSShift);
-			else if (skb->ip_summed == CHECKSUM_HW) {
+			else if (skb->ip_summed == CHECKSUM_PARTIAL) {
 				if (ip->protocol == IPPROTO_TCP)
 					ctrl |= IPCS | TCPCS;
 				else if (ip->protocol == IPPROTO_UDP)
@@ -898,7 +898,7 @@ static int cp_start_xmit (struct sk_buff *skb, struct net_device *dev)
 		txd->addr = cpu_to_le64(first_mapping);
 		wmb();
 
-		if (skb->ip_summed == CHECKSUM_HW) {
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
 			if (ip->protocol == IPPROTO_TCP)
 				txd->opts1 = cpu_to_le32(first_eor | first_len |
 							 FirstFrag | DescOwn |
diff --git a/drivers/net/acenic.c b/drivers/net/acenic.c
index 1c01e9b3d07c..826548644d7b 100644
--- a/drivers/net/acenic.c
+++ b/drivers/net/acenic.c
@@ -2040,7 +2040,7 @@ static void ace_rx_int(struct net_device *dev, u32 rxretprd, u32 rxretcsm)
 		 */
 		if (bd_flags & BD_FLG_TCP_UDP_SUM) {
 			skb->csum = htons(csum);
-			skb->ip_summed = CHECKSUM_HW;
+			skb->ip_summed = CHECKSUM_COMPLETE;
 		} else {
 			skb->ip_summed = CHECKSUM_NONE;
 		}
@@ -2511,7 +2511,7 @@ restart:
 
 		mapping = ace_map_tx_skb(ap, skb, skb, idx);
 		flagsize = (skb->len << 16) | (BD_FLG_END);
-		if (skb->ip_summed == CHECKSUM_HW)
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
 			flagsize |= BD_FLG_TCP_UDP_SUM;
 #if ACENIC_DO_VLAN
 		if (vlan_tx_tag_present(skb)) {
@@ -2534,7 +2534,7 @@ restart:
 
 		mapping = ace_map_tx_skb(ap, skb, NULL, idx);
 		flagsize = (skb_headlen(skb) << 16);
-		if (skb->ip_summed == CHECKSUM_HW)
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
 			flagsize |= BD_FLG_TCP_UDP_SUM;
 #if ACENIC_DO_VLAN
 		if (vlan_tx_tag_present(skb)) {
@@ -2560,7 +2560,7 @@ restart:
 					       PCI_DMA_TODEVICE);
 
 			flagsize = (frag->size << 16);
-			if (skb->ip_summed == CHECKSUM_HW)
+			if (skb->ip_summed == CHECKSUM_PARTIAL)
 				flagsize |= BD_FLG_TCP_UDP_SUM;
 			idx = (idx + 1) % ACE_TX_RING_ENTRIES(ap);
 
diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 652eb05a6c2d..7857b4630124 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -4423,7 +4423,7 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	ring_prod = TX_RING_IDX(prod);
 
 	vlan_tag_flags = 0;
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		vlan_tag_flags |= TX_BD_FLAGS_TCP_UDP_CKSUM;
 	}
 
diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c
index a31544ccb3c4..558fdb8ad2dc 100644
--- a/drivers/net/cassini.c
+++ b/drivers/net/cassini.c
@@ -2167,7 +2167,7 @@ end_copy_pkt:
 			cas_page_unmap(addr);
 	}
 	skb->csum = ntohs(i ^ 0xffff);
-	skb->ip_summed = CHECKSUM_HW;
+	skb->ip_summed = CHECKSUM_COMPLETE;
 	skb->protocol = eth_type_trans(skb, cp->dev);
 	return len;
 }
@@ -2821,7 +2821,7 @@ static inline int cas_xmit_tx_ringN(struct cas *cp, int ring,
 	}
 
 	ctrl = 0;
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		u64 csum_start_off, csum_stuff_off;
 
 		csum_start_off = (u64) (skb->h.raw - skb->data);
diff --git a/drivers/net/chelsio/sge.c b/drivers/net/chelsio/sge.c
index 61b3754f50ff..ddd0bdb498f4 100644
--- a/drivers/net/chelsio/sge.c
+++ b/drivers/net/chelsio/sge.c
@@ -1470,9 +1470,9 @@ int t1_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 
 		if (!(adapter->flags & UDP_CSUM_CAPABLE) &&
-		    skb->ip_summed == CHECKSUM_HW &&
+		    skb->ip_summed == CHECKSUM_PARTIAL &&
 		    skb->nh.iph->protocol == IPPROTO_UDP)
-			if (unlikely(skb_checksum_help(skb, 0))) {
+			if (unlikely(skb_checksum_help(skb))) {
 				dev_kfree_skb_any(skb);
 				return NETDEV_TX_OK;
 			}
@@ -1495,11 +1495,11 @@ int t1_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		cpl = (struct cpl_tx_pkt *)__skb_push(skb, sizeof(*cpl));
 		cpl->opcode = CPL_TX_PKT;
 		cpl->ip_csum_dis = 1;    /* SW calculates IP csum */
-		cpl->l4_csum_dis = skb->ip_summed == CHECKSUM_HW ? 0 : 1;
+		cpl->l4_csum_dis = skb->ip_summed == CHECKSUM_PARTIAL ? 0 : 1;
 		/* the length field isn't used so don't bother setting it */
 
-		st->tx_cso += (skb->ip_summed == CHECKSUM_HW);
-		sge->stats.tx_do_cksum += (skb->ip_summed == CHECKSUM_HW);
+		st->tx_cso += (skb->ip_summed == CHECKSUM_PARTIAL);
+		sge->stats.tx_do_cksum += (skb->ip_summed == CHECKSUM_PARTIAL);
 		sge->stats.tx_reg_pkts++;
 	}
 	cpl->iff = dev->if_port;
diff --git a/drivers/net/dl2k.c b/drivers/net/dl2k.c
index 402961e68c89..b74e67654764 100644
--- a/drivers/net/dl2k.c
+++ b/drivers/net/dl2k.c
@@ -611,7 +611,7 @@ start_xmit (struct sk_buff *skb, struct net_device *dev)
 	txdesc = &np->tx_ring[entry];
 
 #if 0
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		txdesc->status |=
 		    cpu_to_le64 (TCPChecksumEnable | UDPChecksumEnable |
 				 IPChecksumEnable);
diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 98ef9f85482f..2ab9f96f5dab 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -2600,7 +2600,7 @@ e1000_tx_csum(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring,
 	unsigned int i;
 	uint8_t css;
 
-	if (likely(skb->ip_summed == CHECKSUM_HW)) {
+	if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
 		css = skb->h.raw - skb->data;
 
 		i = tx_ring->next_to_use;
@@ -2927,11 +2927,11 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 	}
 
 	/* reserve a descriptor for the offload context */
-	if ((mss) || (skb->ip_summed == CHECKSUM_HW))
+	if ((mss) || (skb->ip_summed == CHECKSUM_PARTIAL))
 		count++;
 	count++;
 #else
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
 		count++;
 #endif
 
@@ -3608,7 +3608,7 @@ e1000_rx_checksum(struct e1000_adapter *adapter,
 		 */
 		csum = ntohl(csum ^ 0xFFFF);
 		skb->csum = csum;
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_COMPLETE;
 	}
 	adapter->hw_csum_good++;
 }
diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
index 11b8f1b43dd5..32cacf115f75 100644
--- a/drivers/net/forcedeth.c
+++ b/drivers/net/forcedeth.c
@@ -1503,7 +1503,8 @@ static int nv_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		tx_flags_extra = NV_TX2_TSO | (skb_shinfo(skb)->gso_size << NV_TX2_TSO_SHIFT);
 	else
 #endif
-	tx_flags_extra = (skb->ip_summed == CHECKSUM_HW ? (NV_TX2_CHECKSUM_L3|NV_TX2_CHECKSUM_L4) : 0);
+	tx_flags_extra = skb->ip_summed == CHECKSUM_PARTIAL ?
+			 NV_TX2_CHECKSUM_L3 | NV_TX2_CHECKSUM_L4 : 0;
 
 	/* vlan tag */
 	if (np->vlangrp && vlan_tx_tag_present(skb)) {
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index ebbbd6ca6204..ba960913c034 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -947,7 +947,7 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/* Set up checksumming */
 	if (likely((dev->features & NETIF_F_IP_CSUM)
-			&& (CHECKSUM_HW == skb->ip_summed))) {
+			&& (CHECKSUM_PARTIAL == skb->ip_summed))) {
 		fcb = gfar_add_fcb(skb, txbdp);
 		status |= TXBD_TOE;
 		gfar_tx_checksum(skb, fcb);
diff --git a/drivers/net/hamachi.c b/drivers/net/hamachi.c
index 409c6aab0411..763373ae9666 100644
--- a/drivers/net/hamachi.c
+++ b/drivers/net/hamachi.c
@@ -1648,7 +1648,7 @@ static int hamachi_rx(struct net_device *dev)
 						* could do the pseudo myself and return
 						* CHECKSUM_UNNECESSARY
 						*/
-						skb->ip_summed = CHECKSUM_HW;
+						skb->ip_summed = CHECKSUM_COMPLETE;
 					}
 				}	
 			}
diff --git a/drivers/net/ibm_emac/ibm_emac_core.c b/drivers/net/ibm_emac/ibm_emac_core.c
index 82468e2dc799..57e214d85e9a 100644
--- a/drivers/net/ibm_emac/ibm_emac_core.c
+++ b/drivers/net/ibm_emac/ibm_emac_core.c
@@ -1036,7 +1036,7 @@ static inline u16 emac_tx_csum(struct ocp_enet_private *dev,
 			       struct sk_buff *skb)
 {
 #if defined(CONFIG_IBM_EMAC_TAH)
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		++dev->stats.tx_packets_csum;
 		return EMAC_TX_CTRL_TAH_CSUM;
 	}
diff --git a/drivers/net/ioc3-eth.c b/drivers/net/ioc3-eth.c
index 68d8af7df08e..65f897ddb920 100644
--- a/drivers/net/ioc3-eth.c
+++ b/drivers/net/ioc3-eth.c
@@ -1387,7 +1387,7 @@ static int ioc3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 * MAC header which should not be summed and the TCP/UDP pseudo headers
 	 * manually.
 	 */
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		int proto = ntohs(skb->nh.iph->protocol);
 		unsigned int csoff;
 		struct iphdr *ih = skb->nh.iph;
diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c
index 7bbd447289b5..9405b44f3214 100644
--- a/drivers/net/ixgb/ixgb_main.c
+++ b/drivers/net/ixgb/ixgb_main.c
@@ -1232,7 +1232,7 @@ ixgb_tx_csum(struct ixgb_adapter *adapter, struct sk_buff *skb)
 	unsigned int i;
 	uint8_t css, cso;
 
-	if(likely(skb->ip_summed == CHECKSUM_HW)) {
+	if(likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
 		css = skb->h.raw - skb->data;
 		cso = (skb->h.raw + skb->csum) - skb->data;
 
diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index eeab1df5bef3..38df58fdb358 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -1147,7 +1147,7 @@ static void eth_tx_submit_descs_for_skb(struct mv643xx_private *mp,
 	desc->byte_cnt = length;
 	desc->buf_ptr = dma_map_single(NULL, skb->data, length, DMA_TO_DEVICE);
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		BUG_ON(skb->protocol != ETH_P_IP);
 
 		cmd_sts |= ETH_GEN_TCP_UDP_CHECKSUM |
diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c
index 9bdd43ab3573..9f16681d0e7e 100644
--- a/drivers/net/myri10ge/myri10ge.c
+++ b/drivers/net/myri10ge/myri10ge.c
@@ -930,7 +930,7 @@ static inline void myri10ge_vlan_ip_csum(struct sk_buff *skb, u16 hw_csum)
 	    (vh->h_vlan_encapsulated_proto == htons(ETH_P_IP) ||
 	     vh->h_vlan_encapsulated_proto == htons(ETH_P_IPV6))) {
 		skb->csum = hw_csum;
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_COMPLETE;
 	}
 }
 
@@ -973,7 +973,7 @@ myri10ge_rx_done(struct myri10ge_priv *mgp, struct myri10ge_rx_buf *rx,
 		if ((skb->protocol == ntohs(ETH_P_IP)) ||
 		    (skb->protocol == ntohs(ETH_P_IPV6))) {
 			skb->csum = ntohs((u16) csum);
-			skb->ip_summed = CHECKSUM_HW;
+			skb->ip_summed = CHECKSUM_COMPLETE;
 		} else
 			myri10ge_vlan_ip_csum(skb, ntohs((u16) csum));
 	}
@@ -1897,13 +1897,13 @@ again:
 	pseudo_hdr_offset = 0;
 	odd_flag = 0;
 	flags = (MXGEFW_FLAGS_NO_TSO | MXGEFW_FLAGS_FIRST);
-	if (likely(skb->ip_summed == CHECKSUM_HW)) {
+	if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
 		cksum_offset = (skb->h.raw - skb->data);
 		pseudo_hdr_offset = (skb->h.raw + skb->csum) - skb->data;
 		/* If the headers are excessively large, then we must
 		 * fall back to a software checksum */
 		if (unlikely(cksum_offset > 255 || pseudo_hdr_offset > 127)) {
-			if (skb_checksum_help(skb, 0))
+			if (skb_checksum_help(skb))
 				goto drop;
 			cksum_offset = 0;
 			pseudo_hdr_offset = 0;
diff --git a/drivers/net/ns83820.c b/drivers/net/ns83820.c
index 0e76859c90a2..5143f5dbb2e5 100644
--- a/drivers/net/ns83820.c
+++ b/drivers/net/ns83820.c
@@ -1153,7 +1153,7 @@ again:
 	if (!nr_frags)
 		frag = NULL;
 	extsts = 0;
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		extsts |= EXTSTS_IPPKT;
 		if (IPPROTO_TCP == skb->nh.iph->protocol)
 			extsts |= EXTSTS_TCPPKT;
diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 4c2f575faad7..d9b960aa9b0d 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -2169,7 +2169,7 @@ static inline u32 rtl8169_tso_csum(struct sk_buff *skb, struct net_device *dev)
 		if (mss)
 			return LargeSend | ((mss & MSSMask) << MSSShift);
 	}
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		const struct iphdr *ip = skb->nh.iph;
 
 		if (ip->protocol == IPPROTO_TCP)
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index e72e0e099060..5b3713f622d7 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -3893,7 +3893,7 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev)
 		txdp->Control_1 |= TXD_TCP_LSO_MSS(s2io_tcp_mss(skb));
 	}
 #endif
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		txdp->Control_2 |=
 		    (TXD_TX_CKO_IPV4_EN | TXD_TX_CKO_TCP_EN |
 		     TXD_TX_CKO_UDP_EN);
diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c
index ee62845d3ac9..eb3b35180c2f 100644
--- a/drivers/net/sk98lin/skge.c
+++ b/drivers/net/sk98lin/skge.c
@@ -1559,7 +1559,7 @@ struct sk_buff	*pMessage)	/* pointer to send-message              */
 	pTxd->VDataHigh = (SK_U32) (PhysAddr >> 32);
 	pTxd->pMBuf     = pMessage;
 
-	if (pMessage->ip_summed == CHECKSUM_HW) {
+	if (pMessage->ip_summed == CHECKSUM_PARTIAL) {
 		u16 hdrlen = pMessage->h.raw - pMessage->data;
 		u16 offset = hdrlen + pMessage->csum;
 
@@ -1678,7 +1678,7 @@ struct sk_buff	*pMessage)	/* pointer to send-message              */
 	/* 
 	** Does the HW need to evaluate checksum for TCP or UDP packets? 
 	*/
-	if (pMessage->ip_summed == CHECKSUM_HW) {
+	if (pMessage->ip_summed == CHECKSUM_PARTIAL) {
 		u16 hdrlen = pMessage->h.raw - pMessage->data;
 		u16 offset = hdrlen + pMessage->csum;
 
@@ -2158,7 +2158,7 @@ rx_start:
 
 #ifdef USE_SK_RX_CHECKSUM
 		pMsg->csum = pRxd->TcpSums & 0xffff;
-		pMsg->ip_summed = CHECKSUM_HW;
+		pMsg->ip_summed = CHECKSUM_COMPLETE;
 #else
 		pMsg->ip_summed = CHECKSUM_NONE;
 #endif
diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index ad878dfddef4..b3d6fa3d6df4 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -2338,7 +2338,7 @@ static int skge_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 	td->dma_lo = map;
 	td->dma_hi = map >> 32;
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		int offset = skb->h.raw - skb->data;
 
 		/* This seems backwards, but it is what the sk98lin
@@ -2642,7 +2642,7 @@ static inline struct sk_buff *skge_rx_get(struct skge_port *skge,
 	skb->dev = skge->netdev;
 	if (skge->rx_csum) {
 		skb->csum = csum;
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_COMPLETE;
 	}
 
 	skb->protocol = eth_type_trans(skb, skge->netdev);
diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 933e87f1cc68..8e92566b587e 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -1163,7 +1163,7 @@ static unsigned tx_le_req(const struct sk_buff *skb)
 	if (skb_is_gso(skb))
 		++count;
 
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
 		++count;
 
 	return count;
@@ -1272,7 +1272,7 @@ static int sky2_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 #endif
 
 	/* Handle TCP checksum offload */
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		u16 hdr = skb->h.raw - skb->data;
 		u16 offset = hdr + skb->csum;
 
@@ -2000,7 +2000,7 @@ static int sky2_status_intr(struct sky2_hw *hw, int to_do)
 #endif
 		case OP_RXCHKS:
 			skb = sky2->rx_ring[sky2->rx_next].skb;
-			skb->ip_summed = CHECKSUM_HW;
+			skb->ip_summed = CHECKSUM_COMPLETE;
 			skb->csum = le16_to_cpu(status);
 			break;
 
diff --git a/drivers/net/starfire.c b/drivers/net/starfire.c
index c0a62b00ffc8..2607aa51d8e0 100644
--- a/drivers/net/starfire.c
+++ b/drivers/net/starfire.c
@@ -1230,7 +1230,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev)
 	}
 
 #if defined(ZEROCOPY) && defined(HAS_BROKEN_FIRMWARE)
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		if (skb_padto(skb, (skb->len + PADDING_MASK) & ~PADDING_MASK))
 			return NETDEV_TX_OK;
 	}
@@ -1252,7 +1252,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev)
 				status |= TxDescIntr;
 				np->reap_tx = 0;
 			}
-			if (skb->ip_summed == CHECKSUM_HW) {
+			if (skb->ip_summed == CHECKSUM_PARTIAL) {
 				status |= TxCalTCP;
 				np->stats.tx_compressed++;
 			}
@@ -1499,7 +1499,7 @@ static int __netdev_rx(struct net_device *dev, int *quota)
 		 * Until then, the printk stays. :-) -Ion
 		 */
 		else if (le16_to_cpu(desc->status2) & 0x0040) {
-			skb->ip_summed = CHECKSUM_HW;
+			skb->ip_summed = CHECKSUM_COMPLETE;
 			skb->csum = le16_to_cpu(desc->csum);
 			printk(KERN_DEBUG "%s: checksum_hw, status2 = %#x\n", dev->name, le16_to_cpu(desc->status2));
 		}
diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c
index d7b1d1882cab..b388651b7836 100644
--- a/drivers/net/sungem.c
+++ b/drivers/net/sungem.c
@@ -855,7 +855,7 @@ static int gem_rx(struct gem *gp, int work_to_do)
 		}
 
 		skb->csum = ntohs((status & RXDCTRL_TCPCSUM) ^ 0xffff);
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_COMPLETE;
 		skb->protocol = eth_type_trans(skb, gp->dev);
 
 		netif_receive_skb(skb);
@@ -1026,7 +1026,7 @@ static int gem_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	unsigned long flags;
 
 	ctrl = 0;
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		u64 csum_start_off, csum_stuff_off;
 
 		csum_start_off = (u64) (skb->h.raw - skb->data);
diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c
index c6f5bc3c042f..17981da22730 100644
--- a/drivers/net/sunhme.c
+++ b/drivers/net/sunhme.c
@@ -1207,7 +1207,7 @@ static void happy_meal_transceiver_check(struct happy_meal *hp, void __iomem *tr
  * flags, thus:
  *
  * 	skb->csum = rxd->rx_flags & 0xffff;
- * 	skb->ip_summed = CHECKSUM_HW;
+ * 	skb->ip_summed = CHECKSUM_COMPLETE;
  *
  * before sending off the skb to the protocols, and we are good as gold.
  */
@@ -2074,7 +2074,7 @@ static void happy_meal_rx(struct happy_meal *hp, struct net_device *dev)
 
 		/* This card is _fucking_ hot... */
 		skb->csum = ntohs(csum ^ 0xffff);
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_COMPLETE;
 
 		RXD(("len=%d csum=%4x]", len, csum));
 		skb->protocol = eth_type_trans(skb, dev);
@@ -2268,7 +2268,7 @@ static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev)
  	u32 tx_flags;
 
 	tx_flags = TXFLAG_OWN;
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		u32 csum_start_off, csum_stuff_off;
 
 		csum_start_off = (u32) (skb->h.raw - skb->data);
diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index eafabb253f08..6f5d3a38c582 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -3851,11 +3851,11 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		skb->h.th->check = 0;
 
 	}
-	else if (skb->ip_summed == CHECKSUM_HW)
+	else if (skb->ip_summed == CHECKSUM_PARTIAL)
 		base_flags |= TXD_FLAG_TCPUDP_CSUM;
 #else
 	mss = 0;
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
 		base_flags |= TXD_FLAG_TCPUDP_CSUM;
 #endif
 #if TG3_VLAN_TAG_USED
@@ -3981,7 +3981,7 @@ static int tg3_start_xmit_dma_bug(struct sk_buff *skb, struct net_device *dev)
 
 	entry = tp->tx_prod;
 	base_flags = 0;
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
 		base_flags |= TXD_FLAG_TCPUDP_CSUM;
 #if TG3_TSO_SUPPORT != 0
 	mss = 0;
diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c
index 4103c37172f9..c6e601dc6bbc 100644
--- a/drivers/net/typhoon.c
+++ b/drivers/net/typhoon.c
@@ -830,7 +830,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev)
 	first_txd->addrHi = (u64)((unsigned long) skb) >> 32;
 	first_txd->processFlags = 0;
 
-	if(skb->ip_summed == CHECKSUM_HW) {
+	if(skb->ip_summed == CHECKSUM_PARTIAL) {
 		/* The 3XP will figure out if this is UDP/TCP */
 		first_txd->processFlags |= TYPHOON_TX_PF_TCP_CHKSUM;
 		first_txd->processFlags |= TYPHOON_TX_PF_UDP_CHKSUM;
diff --git a/drivers/net/via-rhine.c b/drivers/net/via-rhine.c
index ae971080e2e4..66547159bfd9 100644
--- a/drivers/net/via-rhine.c
+++ b/drivers/net/via-rhine.c
@@ -1230,7 +1230,7 @@ static int rhine_start_tx(struct sk_buff *skb, struct net_device *dev)
 	rp->tx_skbuff[entry] = skb;
 
 	if ((rp->quirks & rqRhineI) &&
-	    (((unsigned long)skb->data & 3) || skb_shinfo(skb)->nr_frags != 0 || skb->ip_summed == CHECKSUM_HW)) {
+	    (((unsigned long)skb->data & 3) || skb_shinfo(skb)->nr_frags != 0 || skb->ip_summed == CHECKSUM_PARTIAL)) {
 		/* Must use alignment buffer. */
 		if (skb->len > PKT_BUF_SZ) {
 			/* packet too long, drop it */
diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c
index aa9cd92f46b2..f1e0c746a388 100644
--- a/drivers/net/via-velocity.c
+++ b/drivers/net/via-velocity.c
@@ -2002,7 +2002,7 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev)
 	 *	Handle hardware checksum
 	 */
 	if ((vptr->flags & VELOCITY_FLAGS_TX_CSUM)
-				 && (skb->ip_summed == CHECKSUM_HW)) {
+				 && (skb->ip_summed == CHECKSUM_PARTIAL)) {
 		struct iphdr *ip = skb->nh.iph;
 		if (ip->protocol == IPPROTO_TCP)
 			td_ptr->tdesc1.TCR |= TCR0_TCPCK;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 50a4719512ed..4f2c2b6beb5e 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -976,7 +976,7 @@ extern void		dev_mcast_init(void);
 extern int		netdev_max_backlog;
 extern int		weight_p;
 extern int		netdev_set_master(struct net_device *dev, struct net_device *master);
-extern int skb_checksum_help(struct sk_buff *skb, int inward);
+extern int skb_checksum_help(struct sk_buff *skb);
 extern struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features);
 #ifdef CONFIG_BUG
 extern void netdev_rx_csum_fault(struct net_device *dev);
@@ -1012,7 +1012,7 @@ static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
 {
 	return skb_is_gso(skb) &&
 	       (!skb_gso_ok(skb, dev->features) ||
-		unlikely(skb->ip_summed != CHECKSUM_HW));
+		unlikely(skb->ip_summed != CHECKSUM_PARTIAL));
 }
 
 /* On bonding slaves other than the currently active slave, suppress
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 755e9cddac47..85577a4ffa61 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -34,8 +34,9 @@
 #define HAVE_ALIGNABLE_SKB	/* Ditto 8)		   */
 
 #define CHECKSUM_NONE 0
-#define CHECKSUM_HW 1
+#define CHECKSUM_PARTIAL 1
 #define CHECKSUM_UNNECESSARY 2
+#define CHECKSUM_COMPLETE 3
 
 #define SKB_DATA_ALIGN(X)	(((X) + (SMP_CACHE_BYTES - 1)) & \
 				 ~(SMP_CACHE_BYTES - 1))
@@ -56,17 +57,17 @@
  *	      Apparently with secret goal to sell you new device, when you
  *	      will add new protocol to your host. F.e. IPv6. 8)
  *
- *	HW: the most generic way. Device supplied checksum of _all_
+ *	COMPLETE: the most generic way. Device supplied checksum of _all_
  *	    the packet as seen by netif_rx in skb->csum.
  *	    NOTE: Even if device supports only some protocols, but
- *	    is able to produce some skb->csum, it MUST use HW,
+ *	    is able to produce some skb->csum, it MUST use COMPLETE,
  *	    not UNNECESSARY.
  *
  * B. Checksumming on output.
  *
  *	NONE: skb is checksummed by protocol or csum is not required.
  *
- *	HW: device is required to csum packet as seen by hard_start_xmit
+ *	PARTIAL: device is required to csum packet as seen by hard_start_xmit
  *	from skb->h.raw to the end and to record the checksum
  *	at skb->h.raw+skb->csum.
  *
@@ -1261,14 +1262,14 @@ static inline int skb_linearize_cow(struct sk_buff *skb)
  *	@len: length of data pulled
  *
  *	After doing a pull on a received packet, you need to call this to
- *	update the CHECKSUM_HW checksum, or set ip_summed to CHECKSUM_NONE
- *	so that it can be recomputed from scratch.
+ *	update the CHECKSUM_COMPLETE checksum, or set ip_summed to
+ *	CHECKSUM_NONE so that it can be recomputed from scratch.
  */
 
 static inline void skb_postpull_rcsum(struct sk_buff *skb,
 				      const void *start, unsigned int len)
 {
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
 }
 
@@ -1287,7 +1288,7 @@ static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len)
 {
 	if (likely(len >= skb->len))
 		return 0;
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->ip_summed = CHECKSUM_NONE;
 	return __pskb_trim(skb, len);
 }
diff --git a/net/core/datagram.c b/net/core/datagram.c
index aecddcc30401..f558c61aecc7 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -417,7 +417,7 @@ unsigned int __skb_checksum_complete(struct sk_buff *skb)
 
 	sum = (u16)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum));
 	if (likely(!sum)) {
-		if (unlikely(skb->ip_summed == CHECKSUM_HW))
+		if (unlikely(skb->ip_summed == CHECKSUM_COMPLETE))
 			netdev_rx_csum_fault(skb->dev);
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
@@ -462,7 +462,7 @@ int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb,
 			goto fault;
 		if ((unsigned short)csum_fold(csum))
 			goto csum_error;
-		if (unlikely(skb->ip_summed == CHECKSUM_HW))
+		if (unlikely(skb->ip_summed == CHECKSUM_COMPLETE))
 			netdev_rx_csum_fault(skb->dev);
 		iov->iov_len -= chunk;
 		iov->iov_base += chunk;
diff --git a/net/core/dev.c b/net/core/dev.c
index d4a1ec3bded5..fc82f6f6e1c1 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1166,12 +1166,12 @@ EXPORT_SYMBOL(netif_device_attach);
  * Invalidate hardware checksum when packet is to be mangled, and
  * complete checksum manually on outgoing path.
  */
-int skb_checksum_help(struct sk_buff *skb, int inward)
+int skb_checksum_help(struct sk_buff *skb)
 {
 	unsigned int csum;
 	int ret = 0, offset = skb->h.raw - skb->data;
 
-	if (inward)
+	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		goto out_set_summed;
 
 	if (unlikely(skb_shinfo(skb)->gso_size)) {
@@ -1223,7 +1223,7 @@ struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features)
 	skb->mac_len = skb->nh.raw - skb->data;
 	__skb_pull(skb, skb->mac_len);
 
-	if (unlikely(skb->ip_summed != CHECKSUM_HW)) {
+	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) {
 		if (skb_header_cloned(skb) &&
 		    (err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
 			return ERR_PTR(err);
@@ -1232,7 +1232,7 @@ struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features)
 	rcu_read_lock();
 	list_for_each_entry_rcu(ptype, &ptype_base[ntohs(type) & 15], list) {
 		if (ptype->type == type && !ptype->dev && ptype->gso_segment) {
-			if (unlikely(skb->ip_summed != CHECKSUM_HW)) {
+			if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) {
 				err = ptype->gso_send_check(skb);
 				segs = ERR_PTR(err);
 				if (err || skb_gso_ok(skb, features))
@@ -1444,11 +1444,11 @@ int dev_queue_xmit(struct sk_buff *skb)
 	/* If packet is not checksummed and device does not support
 	 * checksumming for this protocol, complete checksumming here.
 	 */
-	if (skb->ip_summed == CHECKSUM_HW &&
+	if (skb->ip_summed == CHECKSUM_PARTIAL &&
 	    (!(dev->features & NETIF_F_GEN_CSUM) &&
 	     (!(dev->features & NETIF_F_IP_CSUM) ||
 	      skb->protocol != htons(ETH_P_IP))))
-	      	if (skb_checksum_help(skb, 0))
+	      	if (skb_checksum_help(skb))
 	      		goto out_kfree_skb;
 
 gso:
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 471da451cd48..ead5920c26d6 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -110,7 +110,7 @@ static int checksum_udp(struct sk_buff *skb, struct udphdr *uh,
 
 	psum = csum_tcpudp_nofold(saddr, daddr, ulen, IPPROTO_UDP, 0);
 
-	if (skb->ip_summed == CHECKSUM_HW &&
+	if (skb->ip_summed == CHECKSUM_COMPLETE &&
 	    !(u16)csum_fold(csum_add(psum, skb->csum)))
 		return 0;
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index c54f3664bce5..8a476f1956e5 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1397,7 +1397,7 @@ void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to)
 	unsigned int csum;
 	long csstart;
 
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
 		csstart = skb->h.raw - skb->data;
 	else
 		csstart = skb_headlen(skb);
@@ -1411,7 +1411,7 @@ void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to)
 		csum = skb_copy_and_csum_bits(skb, csstart, to + csstart,
 					      skb->len - csstart, 0);
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		long csstuff = csstart + skb->csum;
 
 		*((unsigned short *)(to + csstuff)) = csum_fold(csum);
@@ -1898,10 +1898,10 @@ int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb,
  *	@len: length of data pulled
  *
  *	This function performs an skb_pull on the packet and updates
- *	update the CHECKSUM_HW checksum.  It should be used on receive
- *	path processing instead of skb_pull unless you know that the
- *	checksum difference is zero (e.g., a valid IP header) or you
- *	are setting ip_summed to CHECKSUM_NONE.
+ *	update the CHECKSUM_COMPLETE checksum.  It should be used on
+ *	receive path processing instead of skb_pull unless you know
+ *	that the checksum difference is zero (e.g., a valid IP header)
+ *	or you are setting ip_summed to CHECKSUM_NONE.
  */
 unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len)
 {
@@ -1994,7 +1994,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features)
 		frag = skb_shinfo(nskb)->frags;
 		k = 0;
 
-		nskb->ip_summed = CHECKSUM_HW;
+		nskb->ip_summed = CHECKSUM_PARTIAL;
 		nskb->csum = skb->csum;
 		memcpy(skb_put(nskb, hsize), skb->data + offset, hsize);
 
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 6ad797c14163..6d223e5c6741 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -930,7 +930,7 @@ int icmp_rcv(struct sk_buff *skb)
 	ICMP_INC_STATS_BH(ICMP_MIB_INMSGS);
 
 	switch (skb->ip_summed) {
-	case CHECKSUM_HW:
+	case CHECKSUM_COMPLETE:
 		if (!(u16)csum_fold(skb->csum))
 			break;
 		/* fall through */
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 8e8117c19e4d..7003e763d970 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -931,7 +931,7 @@ int igmp_rcv(struct sk_buff *skb)
 		goto drop;
 
 	switch (skb->ip_summed) {
-	case CHECKSUM_HW:
+	case CHECKSUM_COMPLETE:
 		if (!(u16)csum_fold(skb->csum))
 			break;
 		/* fall through */
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index b84b53a47526..8d7f107c2eef 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -665,7 +665,7 @@ static struct sk_buff *ip_frag_reasm(struct ipq *qp, struct net_device *dev)
 		head->len += fp->len;
 		if (head->ip_summed != fp->ip_summed)
 			head->ip_summed = CHECKSUM_NONE;
-		else if (head->ip_summed == CHECKSUM_HW)
+		else if (head->ip_summed == CHECKSUM_COMPLETE)
 			head->csum = csum_add(head->csum, fp->csum);
 		head->truesize += fp->truesize;
 		atomic_sub(fp->truesize, &ip_frag_mem);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 0f9b3a31997b..e66f6ff2e198 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -576,7 +576,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 
 		if (flags&GRE_CSUM) {
 			switch (skb->ip_summed) {
-			case CHECKSUM_HW:
+			case CHECKSUM_COMPLETE:
 				csum = (u16)csum_fold(skb->csum);
 				if (!csum)
 					break;
@@ -584,7 +584,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 			case CHECKSUM_NONE:
 				skb->csum = 0;
 				csum = __skb_checksum_complete(skb);
-				skb->ip_summed = CHECKSUM_HW;
+				skb->ip_summed = CHECKSUM_COMPLETE;
 			}
 			offset += 4;
 		}
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 308bdeac3455..1b9b6742ef77 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -680,7 +680,7 @@ ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk
 {
 	struct iovec *iov = from;
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		if (memcpy_fromiovecend(to, iov, offset, len) < 0)
 			return -EFAULT;
 	} else {
@@ -736,7 +736,7 @@ static inline int ip_ufo_append_data(struct sock *sk,
 		/* initialize protocol header pointer */
 		skb->h.raw = skb->data + fragheaderlen;
 
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum = 0;
 		sk->sk_sndmsg_off = 0;
 	}
@@ -844,7 +844,7 @@ int ip_append_data(struct sock *sk,
 	    length + fragheaderlen <= mtu &&
 	    rt->u.dst.dev->features & NETIF_F_ALL_CSUM &&
 	    !exthdrlen)
-		csummode = CHECKSUM_HW;
+		csummode = CHECKSUM_PARTIAL;
 
 	inet->cork.length += length;
 	if (((length > mtu) && (sk->sk_protocol == IPPROTO_UDP)) &&
diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c
index bc28b1160a3a..820e8318d10d 100644
--- a/net/ipv4/ipvs/ip_vs_proto_tcp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c
@@ -151,7 +151,7 @@ tcp_snat_handler(struct sk_buff **pskb,
 		/* Only port and addr are changed, do fast csum update */
 		tcp_fast_csum_update(tcph, cp->daddr, cp->vaddr,
 				     cp->dport, cp->vport);
-		if ((*pskb)->ip_summed == CHECKSUM_HW)
+		if ((*pskb)->ip_summed == CHECKSUM_COMPLETE)
 			(*pskb)->ip_summed = CHECKSUM_NONE;
 	} else {
 		/* full checksum calculation */
@@ -204,7 +204,7 @@ tcp_dnat_handler(struct sk_buff **pskb,
 		/* Only port and addr are changed, do fast csum update */
 		tcp_fast_csum_update(tcph, cp->vaddr, cp->daddr,
 				     cp->vport, cp->dport);
-		if ((*pskb)->ip_summed == CHECKSUM_HW)
+		if ((*pskb)->ip_summed == CHECKSUM_COMPLETE)
 			(*pskb)->ip_summed = CHECKSUM_NONE;
 	} else {
 		/* full checksum calculation */
@@ -229,7 +229,7 @@ tcp_csum_check(struct sk_buff *skb, struct ip_vs_protocol *pp)
 	switch (skb->ip_summed) {
 	case CHECKSUM_NONE:
 		skb->csum = skb_checksum(skb, tcphoff, skb->len - tcphoff, 0);
-	case CHECKSUM_HW:
+	case CHECKSUM_COMPLETE:
 		if (csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr,
 				      skb->len - tcphoff,
 				      skb->nh.iph->protocol, skb->csum)) {
@@ -239,7 +239,7 @@ tcp_csum_check(struct sk_buff *skb, struct ip_vs_protocol *pp)
 		}
 		break;
 	default:
-		/* CHECKSUM_UNNECESSARY */
+		/* No need to checksum. */
 		break;
 	}
 
diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c
index 89d9175d8f28..90c8166c0ec1 100644
--- a/net/ipv4/ipvs/ip_vs_proto_udp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_udp.c
@@ -161,7 +161,7 @@ udp_snat_handler(struct sk_buff **pskb,
 		/* Only port and addr are changed, do fast csum update */
 		udp_fast_csum_update(udph, cp->daddr, cp->vaddr,
 				     cp->dport, cp->vport);
-		if ((*pskb)->ip_summed == CHECKSUM_HW)
+		if ((*pskb)->ip_summed == CHECKSUM_COMPLETE)
 			(*pskb)->ip_summed = CHECKSUM_NONE;
 	} else {
 		/* full checksum calculation */
@@ -216,7 +216,7 @@ udp_dnat_handler(struct sk_buff **pskb,
 		/* Only port and addr are changed, do fast csum update */
 		udp_fast_csum_update(udph, cp->vaddr, cp->daddr,
 				     cp->vport, cp->dport);
-		if ((*pskb)->ip_summed == CHECKSUM_HW)
+		if ((*pskb)->ip_summed == CHECKSUM_COMPLETE)
 			(*pskb)->ip_summed = CHECKSUM_NONE;
 	} else {
 		/* full checksum calculation */
@@ -250,7 +250,7 @@ udp_csum_check(struct sk_buff *skb, struct ip_vs_protocol *pp)
 		case CHECKSUM_NONE:
 			skb->csum = skb_checksum(skb, udphoff,
 						 skb->len - udphoff, 0);
-		case CHECKSUM_HW:
+		case CHECKSUM_COMPLETE:
 			if (csum_tcpudp_magic(skb->nh.iph->saddr,
 					      skb->nh.iph->daddr,
 					      skb->len - udphoff,
@@ -262,7 +262,7 @@ udp_csum_check(struct sk_buff *skb, struct ip_vs_protocol *pp)
 			}
 			break;
 		default:
-			/* CHECKSUM_UNNECESSARY */
+			/* No need to checksum. */
 			break;
 		}
 	}
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 6a9e34b794bc..f88347de21a9 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -168,7 +168,7 @@ unsigned int nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
 	unsigned int csum = 0;
 
 	switch (skb->ip_summed) {
-	case CHECKSUM_HW:
+	case CHECKSUM_COMPLETE:
 		if (hook != NF_IP_PRE_ROUTING && hook != NF_IP_LOCAL_IN)
 			break;
 		if ((protocol == 0 && !(u16)csum_fold(skb->csum)) ||
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
index fb920e76ec10..9de81ff645d5 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
@@ -865,8 +865,7 @@ static int tcp_error(struct sk_buff *skb,
   
 	/* Checksum invalid? Ignore.
 	 * We skip checking packets on the outgoing path
-	 * because the semantic of CHECKSUM_HW is different there 
-	 * and moreover root might send raw packets.
+	 * because it is assumed to be correct.
 	 */
 	/* FIXME: Source route IP option packets --RR */
 	if (ip_conntrack_checksum && hooknum == NF_IP_PRE_ROUTING &&
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_udp.c b/net/ipv4/netfilter/ip_conntrack_proto_udp.c
index 9b2c16b4d2ff..e58e52f14553 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_udp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_udp.c
@@ -117,8 +117,7 @@ static int udp_error(struct sk_buff *skb, enum ip_conntrack_info *ctinfo,
 
 	/* Checksum invalid? Ignore.
 	 * We skip checking packets on the outgoing path
-	 * because the semantic of CHECKSUM_HW is different there 
-	 * and moreover root might send raw packets.
+	 * because the checksum is assumed to be correct.
 	 * FIXME: Source route IP option packets --RR */
 	if (ip_conntrack_checksum && hooknum == NF_IP_PRE_ROUTING &&
 	    nf_ip_checksum(skb, hooknum, iph->ihl * 4, IPPROTO_UDP)) {
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index 17de077a7901..f4f00c816d87 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -111,8 +111,9 @@ ip_nat_fn(unsigned int hooknum,
 		       & htons(IP_MF|IP_OFFSET)));
 
 	/* If we had a hardware checksum before, it's now invalid */
-	if ((*pskb)->ip_summed == CHECKSUM_HW)
-		if (skb_checksum_help(*pskb, (out == NULL)))
+	if ((*pskb)->ip_summed == CHECKSUM_PARTIAL ||
+	    (*pskb)->ip_summed == CHECKSUM_COMPLETE)
+		if (skb_checksum_help(*pskb))
 			return NF_DROP;
 
 	ct = ip_conntrack_get(*pskb, &ctinfo);
diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c
index 198ac36db861..276a964ee6cf 100644
--- a/net/ipv4/netfilter/ip_queue.c
+++ b/net/ipv4/netfilter/ip_queue.c
@@ -208,9 +208,9 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
 		break;
 	
 	case IPQ_COPY_PACKET:
-		if (entry->skb->ip_summed == CHECKSUM_HW &&
-		    (*errp = skb_checksum_help(entry->skb,
-		                               entry->info->outdev == NULL))) {
+		if ((entry->skb->ip_summed == CHECKSUM_PARTIAL ||
+		     entry->skb->ip_summed == CHECKSUM_COMPLETE) &&
+		    (*errp = skb_checksum_help(entry->skb))) {
 			read_unlock_bh(&queue_lock);
 			return NULL;
 		}
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index 4adf5c9d34f5..4ec43f98fe49 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -49,7 +49,7 @@ set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 
 /* Return 0 if there was an error. */
 static inline int
-set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo, int inward)
+set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 {
 	struct tcphdr _tcph, *tcph;
 	u_int16_t diffs[2];
@@ -70,8 +70,9 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo, int inward)
 		return 0;
 	tcph = (void *)(*pskb)->nh.iph + (*pskb)->nh.iph->ihl*4;
 
-	if ((*pskb)->ip_summed == CHECKSUM_HW &&
-	    skb_checksum_help(*pskb, inward))
+	if (((*pskb)->ip_summed == CHECKSUM_PARTIAL ||
+	     (*pskb)->ip_summed == CHECKSUM_COMPLETE) &&
+	    skb_checksum_help(*pskb))
 		return 0;
 
 	diffs[0] = ((u_int16_t *)tcph)[6];
@@ -106,7 +107,7 @@ target(struct sk_buff **pskb,
 
 	if (einfo->operation & (IPT_ECN_OP_SET_ECE | IPT_ECN_OP_SET_CWR)
 	    && (*pskb)->nh.iph->protocol == IPPROTO_TCP)
-		if (!set_ect_tcp(pskb, einfo, (out == NULL)))
+		if (!set_ect_tcp(pskb, einfo))
 			return NF_DROP;
 
 	return IPT_CONTINUE;
diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c
index ef2fe5b3f0d8..c998dc0fcd15 100644
--- a/net/ipv4/netfilter/ipt_TCPMSS.c
+++ b/net/ipv4/netfilter/ipt_TCPMSS.c
@@ -62,8 +62,9 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	if (!skb_make_writable(pskb, (*pskb)->len))
 		return NF_DROP;
 
-	if ((*pskb)->ip_summed == CHECKSUM_HW &&
-	    skb_checksum_help(*pskb, out == NULL))
+	if (((*pskb)->ip_summed == CHECKSUM_PARTIAL ||
+	     (*pskb)->ip_summed == CHECKSUM_COMPLETE) &&
+	    skb_checksum_help(*pskb))
 		return NF_DROP;
 
 	iph = (*pskb)->nh.iph;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 934396bb1376..b0124e69ab38 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -568,7 +568,7 @@ new_segment:
 		skb->truesize += copy;
 		sk->sk_wmem_queued += copy;
 		sk->sk_forward_alloc -= copy;
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_PARTIAL;
 		tp->write_seq += copy;
 		TCP_SKB_CB(skb)->end_seq += copy;
 		skb_shinfo(skb)->gso_segs = 0;
@@ -723,7 +723,7 @@ new_segment:
 				 * Check whether we can use HW checksum.
 				 */
 				if (sk->sk_route_caps & NETIF_F_ALL_CSUM)
-					skb->ip_summed = CHECKSUM_HW;
+					skb->ip_summed = CHECKSUM_PARTIAL;
 
 				skb_entail(sk, tp, skb);
 				copy = size_goal;
@@ -2205,7 +2205,7 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features)
 		th->fin = th->psh = 0;
 
 		th->check = ~csum_fold(th->check + delta);
-		if (skb->ip_summed != CHECKSUM_HW)
+		if (skb->ip_summed != CHECKSUM_PARTIAL)
 			th->check = csum_fold(csum_partial(skb->h.raw, thlen,
 							   skb->csum));
 
@@ -2219,7 +2219,7 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features)
 
 	delta = htonl(oldlen + (skb->tail - skb->h.raw) + skb->data_len);
 	th->check = ~csum_fold(th->check + delta);
-	if (skb->ip_summed != CHECKSUM_HW)
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
 		th->check = csum_fold(csum_partial(skb->h.raw, thlen,
 						   skb->csum));
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 43f6740244f8..b2aa512a30e9 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -484,7 +484,7 @@ void tcp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb)
 	struct inet_sock *inet = inet_sk(sk);
 	struct tcphdr *th = skb->h.th;
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		th->check = ~tcp_v4_check(th, len, inet->saddr, inet->daddr, 0);
 		skb->csum = offsetof(struct tcphdr, check);
 	} else {
@@ -509,7 +509,7 @@ int tcp_v4_gso_send_check(struct sk_buff *skb)
 	th->check = 0;
 	th->check = ~tcp_v4_check(th, skb->len, iph->saddr, iph->daddr, 0);
 	skb->csum = offsetof(struct tcphdr, check);
-	skb->ip_summed = CHECKSUM_HW;
+	skb->ip_summed = CHECKSUM_PARTIAL;
 	return 0;
 }
 
@@ -973,7 +973,7 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
 
 static int tcp_v4_checksum_init(struct sk_buff *skb)
 {
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_COMPLETE) {
 		if (!tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr,
 				  skb->nh.iph->daddr, skb->csum)) {
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index b4f3ffe1b3b4..9252a50c4b49 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -577,7 +577,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 	TCP_SKB_CB(buff)->sacked = TCP_SKB_CB(skb)->sacked;
 	TCP_SKB_CB(skb)->sacked &= ~TCPCB_AT_TAIL;
 
-	if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_HW) {
+	if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_PARTIAL) {
 		/* Copy and checksum data tail into the new buffer. */
 		buff->csum = csum_partial_copy_nocheck(skb->data + len, skb_put(buff, nsize),
 						       nsize, 0);
@@ -586,7 +586,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
 
 		skb->csum = csum_block_sub(skb->csum, buff->csum, len);
 	} else {
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb_split(skb, buff, len);
 	}
 
@@ -689,7 +689,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
 		__pskb_trim_head(skb, len - skb_headlen(skb));
 
 	TCP_SKB_CB(skb)->seq += len;
-	skb->ip_summed = CHECKSUM_HW;
+	skb->ip_summed = CHECKSUM_PARTIAL;
 
 	skb->truesize	     -= len;
 	sk->sk_wmem_queued   -= len;
@@ -1062,7 +1062,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
 	/* This packet was never sent out yet, so no SACK bits. */
 	TCP_SKB_CB(buff)->sacked = 0;
 
-	buff->ip_summed = skb->ip_summed = CHECKSUM_HW;
+	buff->ip_summed = skb->ip_summed = CHECKSUM_PARTIAL;
 	skb_split(skb, buff, len);
 
 	/* Fix up tso_factor for both original and new SKB.  */
@@ -1206,8 +1206,7 @@ static int tcp_mtu_probe(struct sock *sk)
 	TCP_SKB_CB(nskb)->flags = TCPCB_FLAG_ACK;
 	TCP_SKB_CB(nskb)->sacked = 0;
 	nskb->csum = 0;
-	if (skb->ip_summed == CHECKSUM_HW)
-		nskb->ip_summed = CHECKSUM_HW;
+	nskb->ip_summed = skb->ip_summed;
 
 	len = 0;
 	while (len < probe_size) {
@@ -1231,7 +1230,7 @@ static int tcp_mtu_probe(struct sock *sk)
 			                           ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
 			if (!skb_shinfo(skb)->nr_frags) {
 				skb_pull(skb, copy);
-				if (skb->ip_summed != CHECKSUM_HW)
+				if (skb->ip_summed != CHECKSUM_PARTIAL)
 					skb->csum = csum_partial(skb->data, skb->len, 0);
 			} else {
 				__pskb_trim_head(skb, copy);
@@ -1572,10 +1571,9 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m
 
 		memcpy(skb_put(skb, next_skb_size), next_skb->data, next_skb_size);
 
-		if (next_skb->ip_summed == CHECKSUM_HW)
-			skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = next_skb->ip_summed;
 
-		if (skb->ip_summed != CHECKSUM_HW)
+		if (skb->ip_summed != CHECKSUM_PARTIAL)
 			skb->csum = csum_block_add(skb->csum, next_skb->csum, skb_size);
 
 		/* Update sequence range on original skb. */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index a4d005eccc7f..87152510980c 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -429,7 +429,7 @@ static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up)
 		/*
 		 * Only one fragment on the socket.
 		 */
-		if (skb->ip_summed == CHECKSUM_HW) {
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
 			skb->csum = offsetof(struct udphdr, check);
 			uh->check = ~csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst,
 					up->len, IPPROTO_UDP, 0);
@@ -448,7 +448,7 @@ static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up)
 		 * fragments on the socket so that all csums of sk_buffs
 		 * should be together.
 		 */
-		if (skb->ip_summed == CHECKSUM_HW) {
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
 			int offset = (unsigned char *)uh - skb->data;
 			skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
 
@@ -1088,7 +1088,7 @@ static void udp_checksum_init(struct sk_buff *skb, struct udphdr *uh,
 {
 	if (uh->check == 0) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
-	} else if (skb->ip_summed == CHECKSUM_HW) {
+	} else if (skb->ip_summed == CHECKSUM_COMPLETE) {
 		if (!udp_check(uh, ulen, saddr, daddr, skb->csum))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index d16f863cf687..4a96a9e3ef3b 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -48,8 +48,8 @@ static int xfrm4_output_one(struct sk_buff *skb)
 	struct xfrm_state *x = dst->xfrm;
 	int err;
 	
-	if (skb->ip_summed == CHECKSUM_HW) {
-		err = skb_checksum_help(skb, 0);
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		err = skb_checksum_help(skb);
 		if (err)
 			goto error_nolock;
 	}
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 86dac106873b..05afa6b1912b 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -294,7 +294,7 @@ looped_back:
 		hdr = (struct ipv6_rt_hdr *) skb2->h.raw;
 	}
 
-	if (skb->ip_summed == CHECKSUM_HW)
+	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->ip_summed = CHECKSUM_NONE;
 
 	i = n - --hdr->segments_left;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index dbfce089e916..103055107674 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -606,7 +606,7 @@ static int icmpv6_rcv(struct sk_buff **pskb)
 
 	/* Perform checksum. */
 	switch (skb->ip_summed) {
-	case CHECKSUM_HW:
+	case CHECKSUM_COMPLETE:
 		if (!csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6,
 				     skb->csum))
 			break;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 4fb47a252913..65514f21c186 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -866,7 +866,7 @@ static inline int ip6_ufo_append_data(struct sock *sk,
 		/* initialize protocol header pointer */
 		skb->h.raw = skb->data + fragheaderlen;
 
-		skb->ip_summed = CHECKSUM_HW;
+		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum = 0;
 		sk->sk_sndmsg_off = 0;
 	}
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 395a417ba955..580b1aba6722 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -87,7 +87,7 @@ unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 	unsigned int csum = 0;
 
 	switch (skb->ip_summed) {
-	case CHECKSUM_HW:
+	case CHECKSUM_COMPLETE:
 		if (hook != NF_IP6_PRE_ROUTING && hook != NF_IP6_LOCAL_IN)
 			break;
 		if (!csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr,
diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c
index 968a14be0d05..c01c126224e2 100644
--- a/net/ipv6/netfilter/ip6_queue.c
+++ b/net/ipv6/netfilter/ip6_queue.c
@@ -206,9 +206,9 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp)
 		break;
 	
 	case IPQ_COPY_PACKET:
-		if (entry->skb->ip_summed == CHECKSUM_HW &&
-		    (*errp = skb_checksum_help(entry->skb,
-		                               entry->info->outdev == NULL))) {
+		if ((entry->skb->ip_summed == CHECKSUM_PARTIAL ||
+		     entry->skb->ip_summed == CHECKSUM_COMPLETE) &&
+		    (*errp = skb_checksum_help(entry->skb))) {
 			read_unlock_bh(&queue_lock);
 			return NULL;
 		}
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 00d5583807f7..7a4e4c2e3197 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -408,7 +408,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
  		return -1;
 	}
 
- 	if (skb->ip_summed == CHECKSUM_HW)
+ 	if (skb->ip_summed == CHECKSUM_COMPLETE)
  		skb->csum = csum_sub(skb->csum,
  				     csum_partial(skb->nh.raw,
 						  (u8*)(fhdr + 1) - skb->nh.raw,
@@ -640,7 +640,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 		head->len += fp->len;
 		if (head->ip_summed != fp->ip_summed)
 			head->ip_summed = CHECKSUM_NONE;
-		else if (head->ip_summed == CHECKSUM_HW)
+		else if (head->ip_summed == CHECKSUM_COMPLETE)
 			head->csum = csum_add(head->csum, fp->csum);
 		head->truesize += fp->truesize;
 		atomic_sub(fp->truesize, &nf_ct_frag6_mem);
@@ -652,7 +652,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 	head->nh.ipv6h->payload_len = htons(payload_len);
 
 	/* Yes, and fold redundant checksum back. 8) */
-	if (head->ip_summed == CHECKSUM_HW)
+	if (head->ip_summed == CHECKSUM_COMPLETE)
 		head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum);
 
 	fq->fragments = NULL;
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index d5040e172292..d4af1cb5e19f 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -334,7 +334,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
 	if (!rp->checksum)
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_COMPLETE) {
 		skb_postpull_rcsum(skb, skb->nh.raw,
 		                   skb->h.raw - skb->nh.raw);
 		if (!csum_ipv6_magic(&skb->nh.ipv6h->saddr,
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 4e299c69e1c6..a8623d2b0879 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -433,7 +433,7 @@ static void ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
  		return;
 	}
 
- 	if (skb->ip_summed == CHECKSUM_HW)
+ 	if (skb->ip_summed == CHECKSUM_COMPLETE)
  		skb->csum = csum_sub(skb->csum,
  				     csum_partial(skb->nh.raw, (u8*)(fhdr+1)-skb->nh.raw, 0));
 
@@ -647,7 +647,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff **skb_in,
 		head->len += fp->len;
 		if (head->ip_summed != fp->ip_summed)
 			head->ip_summed = CHECKSUM_NONE;
-		else if (head->ip_summed == CHECKSUM_HW)
+		else if (head->ip_summed == CHECKSUM_COMPLETE)
 			head->csum = csum_add(head->csum, fp->csum);
 		head->truesize += fp->truesize;
 		atomic_sub(fp->truesize, &ip6_frag_mem);
@@ -662,7 +662,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff **skb_in,
 	*skb_in = head;
 
 	/* Yes, and fold redundant checksum back. 8) */
-	if (head->ip_summed == CHECKSUM_HW)
+	if (head->ip_summed == CHECKSUM_COMPLETE)
 		head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum);
 
 	IP6_INC_STATS_BH(IPSTATS_MIB_REASMOKS);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 302786a11cd6..7f1b660493b7 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -545,7 +545,7 @@ static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb)
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct tcphdr *th = skb->h.th;
 
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		th->check = ~csum_ipv6_magic(&np->saddr, &np->daddr, len, IPPROTO_TCP,  0);
 		skb->csum = offsetof(struct tcphdr, check);
 	} else {
@@ -570,7 +570,7 @@ static int tcp_v6_gso_send_check(struct sk_buff *skb)
 	th->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, skb->len,
 				     IPPROTO_TCP, 0);
 	skb->csum = offsetof(struct tcphdr, check);
-	skb->ip_summed = CHECKSUM_HW;
+	skb->ip_summed = CHECKSUM_PARTIAL;
 	return 0;
 }
 
@@ -1033,7 +1033,7 @@ out:
 
 static int tcp_v6_checksum_init(struct sk_buff *skb)
 {
-	if (skb->ip_summed == CHECKSUM_HW) {
+	if (skb->ip_summed == CHECKSUM_COMPLETE) {
 		if (!tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr,
 				  &skb->nh.ipv6h->daddr,skb->csum)) {
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 82c7c9cde2a8..780b89f6dfcc 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -475,7 +475,7 @@ static int udpv6_rcv(struct sk_buff **pskb)
 		uh = skb->h.uh;
 	}
 
-	if (skb->ip_summed == CHECKSUM_HW &&
+	if (skb->ip_summed == CHECKSUM_COMPLETE &&
 	    !csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum))
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index c8c8b44a0f58..6d111743e508 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -41,8 +41,8 @@ static int xfrm6_output_one(struct sk_buff *skb)
 	struct xfrm_state *x = dst->xfrm;
 	int err;
 	
-	if (skb->ip_summed == CHECKSUM_HW) {
-		err = skb_checksum_help(skb, 0);
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		err = skb_checksum_help(skb);
 		if (err)
 			goto error_nolock;
 	}
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index af8adcba23a7..308d2abd7ee5 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -823,8 +823,7 @@ static int tcp_error(struct sk_buff *skb,
   
 	/* Checksum invalid? Ignore.
 	 * We skip checking packets on the outgoing path
-	 * because the semantic of CHECKSUM_HW is different there 
-	 * and moreover root might send raw packets.
+	 * because the checksum is assumed to be correct.
 	 */
 	/* FIXME: Source route IP option packets --RR */
 	if (nf_conntrack_checksum &&
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index ae07ebe3ab37..d36e03139e8b 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -131,8 +131,7 @@ static int udp_error(struct sk_buff *skb, unsigned int dataoff,
 
 	/* Checksum invalid? Ignore.
 	 * We skip checking packets on the outgoing path
-	 * because the semantic of CHECKSUM_HW is different there
-	 * and moreover root might send raw packets.
+	 * because the checksum is assumed to be correct.
 	 * FIXME: Source route IP option packets --RR */
 	if (nf_conntrack_checksum &&
 	    ((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) ||
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 49ef41e34c48..eddfbe4441a2 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -377,9 +377,9 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue,
 		break;
 	
 	case NFQNL_COPY_PACKET:
-		if (entskb->ip_summed == CHECKSUM_HW &&
-		    (*errp = skb_checksum_help(entskb,
-		                               outdev == NULL))) {
+		if ((entskb->ip_summed == CHECKSUM_PARTIAL ||
+		     entskb->ip_summed == CHECKSUM_COMPLETE) &&
+		    (*errp = skb_checksum_help(entskb))) {
 			spin_unlock_bh(&queue->lock);
 			return NULL;
 		}
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 4172a5235916..300215bdbf46 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -586,7 +586,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		else if (skb->pkt_type == PACKET_OUTGOING) {
 			/* Special case: outgoing packets have ll header at head */
 			skb_pull(skb, skb->nh.raw - skb->data);
-			if (skb->ip_summed == CHECKSUM_HW)
+			if (skb->ip_summed == CHECKSUM_PARTIAL)
 				status |= TP_STATUS_CSUMNOTREADY;
 		}
 	}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index a08ec4c7c55d..45939bafbdf8 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -192,8 +192,8 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	 */
 	if (q->corrupt && q->corrupt >= get_crandom(&q->corrupt_cor)) {
 		if (!(skb = skb_unshare(skb, GFP_ATOMIC))
-		    || (skb->ip_summed == CHECKSUM_HW
-			&& skb_checksum_help(skb, 0))) {
+		    || (skb->ip_summed == CHECKSUM_PARTIAL
+			&& skb_checksum_help(skb))) {
 			sch->qstats.drops++;
 			return NET_XMIT_DROP;
 		}
diff --git a/net/sunrpc/socklib.c b/net/sunrpc/socklib.c
index eb330d4f66d6..6f17527b9e69 100644
--- a/net/sunrpc/socklib.c
+++ b/net/sunrpc/socklib.c
@@ -168,7 +168,7 @@ int csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb)
 		return -1;
 	if ((unsigned short)csum_fold(desc.csum))
 		return -1;
-	if (unlikely(skb->ip_summed == CHECKSUM_HW))
+	if (unlikely(skb->ip_summed == CHECKSUM_COMPLETE))
 		netdev_rx_csum_fault(skb->dev);
 	return 0;
 no_checksum:
-- 
cgit v1.2.3


From 4cf411de49c65140b3c259748629b561c0d3340f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Aug 2006 00:58:33 -0700
Subject: [NETFILTER]: Get rid of HW checksum invalidation

Update hardware checksums incrementally to avoid breaking GSO.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h                  |  6 +++
 include/linux/netfilter_ipv4/ip_nat.h      |  4 --
 include/linux/netfilter_ipv4/ip_nat_core.h |  8 ++--
 net/ipv4/netfilter/ip_nat_core.c           | 52 ++++++++++----------------
 net/ipv4/netfilter/ip_nat_helper.c         | 59 +++++++++++++++++++++---------
 net/ipv4/netfilter/ip_nat_proto_gre.c      |  5 ++-
 net/ipv4/netfilter/ip_nat_proto_icmp.c     |  8 ++--
 net/ipv4/netfilter/ip_nat_proto_tcp.c      |  7 ++--
 net/ipv4/netfilter/ip_nat_proto_udp.c      | 15 +++++---
 net/ipv4/netfilter/ip_nat_standalone.c     | 10 +----
 net/ipv4/netfilter/ipt_ECN.c               | 19 +++-------
 net/ipv4/netfilter/ipt_REJECT.c            |  1 +
 net/ipv4/netfilter/ipt_TCPMSS.c            | 39 ++++++++------------
 net/netfilter/core.c                       | 22 +++++++++++
 14 files changed, 138 insertions(+), 117 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 10168e26a846..b7e67d1d4382 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -282,6 +282,12 @@ extern void nf_invalidate_cache(int pf);
    Returns true or false. */
 extern int skb_make_writable(struct sk_buff **pskb, unsigned int writable_len);
 
+extern u_int16_t nf_csum_update(u_int32_t oldval, u_int32_t newval,
+				u_int32_t csum);
+extern u_int16_t nf_proto_csum_update(struct sk_buff *skb,
+				      u_int32_t oldval, u_int32_t newval,
+				      u_int16_t csum, int pseudohdr);
+
 struct nf_afinfo {
 	unsigned short	family;
 	unsigned int	(*checksum)(struct sk_buff *skb, unsigned int hook,
diff --git a/include/linux/netfilter_ipv4/ip_nat.h b/include/linux/netfilter_ipv4/ip_nat.h
index e9f5ed1d9f68..98f8407e4cb5 100644
--- a/include/linux/netfilter_ipv4/ip_nat.h
+++ b/include/linux/netfilter_ipv4/ip_nat.h
@@ -72,10 +72,6 @@ extern unsigned int ip_nat_setup_info(struct ip_conntrack *conntrack,
 extern int ip_nat_used_tuple(const struct ip_conntrack_tuple *tuple,
 			     const struct ip_conntrack *ignored_conntrack);
 
-/* Calculate relative checksum. */
-extern u_int16_t ip_nat_cheat_check(u_int32_t oldvalinv,
-				    u_int32_t newval,
-				    u_int16_t oldcheck);
 #else  /* !__KERNEL__: iptables wants this to compile. */
 #define ip_nat_multi_range ip_nat_multi_range_compat
 #endif /*__KERNEL__*/
diff --git a/include/linux/netfilter_ipv4/ip_nat_core.h b/include/linux/netfilter_ipv4/ip_nat_core.h
index 30db23f06b03..60566f9fd7b3 100644
--- a/include/linux/netfilter_ipv4/ip_nat_core.h
+++ b/include/linux/netfilter_ipv4/ip_nat_core.h
@@ -11,8 +11,8 @@ extern unsigned int ip_nat_packet(struct ip_conntrack *ct,
 			       unsigned int hooknum,
 			       struct sk_buff **pskb);
 
-extern int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
-					 struct ip_conntrack *ct,
-					 enum ip_nat_manip_type manip,
-					 enum ip_conntrack_dir dir);
+extern int ip_nat_icmp_reply_translation(struct ip_conntrack *ct,
+					 enum ip_conntrack_info ctinfo,
+					 unsigned int hooknum,
+					 struct sk_buff **pskb);
 #endif /* _IP_NAT_CORE_H */
diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c
index 1741d555ad0d..4c540d03d48e 100644
--- a/net/ipv4/netfilter/ip_nat_core.c
+++ b/net/ipv4/netfilter/ip_nat_core.c
@@ -101,18 +101,6 @@ static void ip_nat_cleanup_conntrack(struct ip_conntrack *conn)
 	write_unlock_bh(&ip_nat_lock);
 }
 
-/* We do checksum mangling, so if they were wrong before they're still
- * wrong.  Also works for incomplete packets (eg. ICMP dest
- * unreachables.) */
-u_int16_t
-ip_nat_cheat_check(u_int32_t oldvalinv, u_int32_t newval, u_int16_t oldcheck)
-{
-	u_int32_t diffs[] = { oldvalinv, newval };
-	return csum_fold(csum_partial((char *)diffs, sizeof(diffs),
-				      oldcheck^0xFFFF));
-}
-EXPORT_SYMBOL(ip_nat_cheat_check);
-
 /* Is this tuple already taken? (not by us) */
 int
 ip_nat_used_tuple(const struct ip_conntrack_tuple *tuple,
@@ -378,12 +366,12 @@ manip_pkt(u_int16_t proto,
 	iph = (void *)(*pskb)->data + iphdroff;
 
 	if (maniptype == IP_NAT_MANIP_SRC) {
-		iph->check = ip_nat_cheat_check(~iph->saddr, target->src.ip,
-						iph->check);
+		iph->check = nf_csum_update(~iph->saddr, target->src.ip,
+					    iph->check);
 		iph->saddr = target->src.ip;
 	} else {
-		iph->check = ip_nat_cheat_check(~iph->daddr, target->dst.ip,
-						iph->check);
+		iph->check = nf_csum_update(~iph->daddr, target->dst.ip,
+					    iph->check);
 		iph->daddr = target->dst.ip;
 	}
 	return 1;
@@ -423,10 +411,10 @@ unsigned int ip_nat_packet(struct ip_conntrack *ct,
 EXPORT_SYMBOL_GPL(ip_nat_packet);
 
 /* Dir is direction ICMP is coming from (opposite to packet it contains) */
-int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
-				  struct ip_conntrack *ct,
-				  enum ip_nat_manip_type manip,
-				  enum ip_conntrack_dir dir)
+int ip_nat_icmp_reply_translation(struct ip_conntrack *ct,
+				  enum ip_conntrack_info ctinfo,
+				  unsigned int hooknum,
+				  struct sk_buff **pskb)
 {
 	struct {
 		struct icmphdr icmp;
@@ -434,7 +422,9 @@ int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
 	} *inside;
 	struct ip_conntrack_tuple inner, target;
 	int hdrlen = (*pskb)->nh.iph->ihl * 4;
+	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	unsigned long statusbit;
+	enum ip_nat_manip_type manip = HOOK2MANIP(hooknum);
 
 	if (!skb_make_writable(pskb, hdrlen + sizeof(*inside)))
 		return 0;
@@ -443,12 +433,8 @@ int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
 
 	/* We're actually going to mangle it beyond trivial checksum
 	   adjustment, so make sure the current checksum is correct. */
-	if ((*pskb)->ip_summed != CHECKSUM_UNNECESSARY) {
-		hdrlen = (*pskb)->nh.iph->ihl * 4;
-		if ((u16)csum_fold(skb_checksum(*pskb, hdrlen,
-						(*pskb)->len - hdrlen, 0)))
-			return 0;
-	}
+	if (nf_ip_checksum(*pskb, hooknum, hdrlen, 0))
+		return 0;
 
 	/* Must be RELATED */
 	IP_NF_ASSERT((*pskb)->nfctinfo == IP_CT_RELATED ||
@@ -487,12 +473,14 @@ int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
 		       !manip))
 		return 0;
 
-	/* Reloading "inside" here since manip_pkt inner. */
-	inside = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4;
-	inside->icmp.checksum = 0;
-	inside->icmp.checksum = csum_fold(skb_checksum(*pskb, hdrlen,
-						       (*pskb)->len - hdrlen,
-						       0));
+	if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) {
+		/* Reloading "inside" here since manip_pkt inner. */
+		inside = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4;
+		inside->icmp.checksum = 0;
+		inside->icmp.checksum = csum_fold(skb_checksum(*pskb, hdrlen,
+							       (*pskb)->len - hdrlen,
+							       0));
+	}
 
 	/* Change outer to look the reply to an incoming packet
 	 * (proto 0 means don't invert per-proto part). */
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index cbcaa45370ae..021c3daae3ed 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -165,7 +165,7 @@ ip_nat_mangle_tcp_packet(struct sk_buff **pskb,
 {
 	struct iphdr *iph;
 	struct tcphdr *tcph;
-	int datalen;
+	int oldlen, datalen;
 
 	if (!skb_make_writable(pskb, (*pskb)->len))
 		return 0;
@@ -180,13 +180,22 @@ ip_nat_mangle_tcp_packet(struct sk_buff **pskb,
 	iph = (*pskb)->nh.iph;
 	tcph = (void *)iph + iph->ihl*4;
 
+	oldlen = (*pskb)->len - iph->ihl*4;
 	mangle_contents(*pskb, iph->ihl*4 + tcph->doff*4,
 			match_offset, match_len, rep_buffer, rep_len);
 
 	datalen = (*pskb)->len - iph->ihl*4;
-	tcph->check = 0;
-	tcph->check = tcp_v4_check(tcph, datalen, iph->saddr, iph->daddr,
-				   csum_partial((char *)tcph, datalen, 0));
+	if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) {
+		tcph->check = 0;
+		tcph->check = tcp_v4_check(tcph, datalen,
+					   iph->saddr, iph->daddr,
+					   csum_partial((char *)tcph,
+					   		datalen, 0));
+	} else
+		tcph->check = nf_proto_csum_update(*pskb,
+						   htons(oldlen) ^ 0xFFFF,
+						   htons(datalen),
+						   tcph->check, 1);
 
 	if (rep_len != match_len) {
 		set_bit(IPS_SEQ_ADJUST_BIT, &ct->status);
@@ -221,6 +230,7 @@ ip_nat_mangle_udp_packet(struct sk_buff **pskb,
 {
 	struct iphdr *iph;
 	struct udphdr *udph;
+	int datalen, oldlen;
 
 	/* UDP helpers might accidentally mangle the wrong packet */
 	iph = (*pskb)->nh.iph;
@@ -238,22 +248,32 @@ ip_nat_mangle_udp_packet(struct sk_buff **pskb,
 
 	iph = (*pskb)->nh.iph;
 	udph = (void *)iph + iph->ihl*4;
+
+	oldlen = (*pskb)->len - iph->ihl*4;
 	mangle_contents(*pskb, iph->ihl*4 + sizeof(*udph),
 			match_offset, match_len, rep_buffer, rep_len);
 
 	/* update the length of the UDP packet */
-	udph->len = htons((*pskb)->len - iph->ihl*4);
+	datalen = (*pskb)->len - iph->ihl*4;
+	udph->len = htons(datalen);
 
 	/* fix udp checksum if udp checksum was previously calculated */
-	if (udph->check) {
-		int datalen = (*pskb)->len - iph->ihl * 4;
+	if (!udph->check && (*pskb)->ip_summed != CHECKSUM_PARTIAL)
+		return 1;
+
+	if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) {
 		udph->check = 0;
 		udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
 		                                datalen, IPPROTO_UDP,
 		                                csum_partial((char *)udph,
 		                                             datalen, 0));
-	}
-
+		if (!udph->check)
+			udph->check = -1;
+	} else
+		udph->check = nf_proto_csum_update(*pskb,
+						   htons(oldlen) ^ 0xFFFF,
+						   htons(datalen),
+						   udph->check, 1);
 	return 1;
 }
 EXPORT_SYMBOL(ip_nat_mangle_udp_packet);
@@ -293,11 +313,14 @@ sack_adjust(struct sk_buff *skb,
 			ntohl(sack->start_seq), new_start_seq,
 			ntohl(sack->end_seq), new_end_seq);
 
-		tcph->check = 
-			ip_nat_cheat_check(~sack->start_seq, new_start_seq,
-					   ip_nat_cheat_check(~sack->end_seq, 
-						   	      new_end_seq,
-							      tcph->check));
+		tcph->check = nf_proto_csum_update(skb,
+						   ~sack->start_seq,
+						   new_start_seq,
+						   tcph->check, 0);
+		tcph->check = nf_proto_csum_update(skb,
+						   ~sack->end_seq,
+						   new_end_seq,
+						   tcph->check, 0);
 		sack->start_seq = new_start_seq;
 		sack->end_seq = new_end_seq;
 		sackoff += sizeof(*sack);
@@ -381,10 +404,10 @@ ip_nat_seq_adjust(struct sk_buff **pskb,
 		newack = ntohl(tcph->ack_seq) - other_way->offset_before;
 	newack = htonl(newack);
 
-	tcph->check = ip_nat_cheat_check(~tcph->seq, newseq,
-					 ip_nat_cheat_check(~tcph->ack_seq, 
-					 		    newack, 
-							    tcph->check));
+	tcph->check = nf_proto_csum_update(*pskb, ~tcph->seq, newseq,
+					   tcph->check, 0);
+	tcph->check = nf_proto_csum_update(*pskb, ~tcph->ack_seq, newack,
+					   tcph->check, 0);
 
 	DEBUGP("Adjusting sequence number from %u->%u, ack from %u->%u\n",
 		ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq),
diff --git a/net/ipv4/netfilter/ip_nat_proto_gre.c b/net/ipv4/netfilter/ip_nat_proto_gre.c
index 38acfdf540eb..70a65372225a 100644
--- a/net/ipv4/netfilter/ip_nat_proto_gre.c
+++ b/net/ipv4/netfilter/ip_nat_proto_gre.c
@@ -130,9 +130,10 @@ gre_manip_pkt(struct sk_buff **pskb,
 			if (greh->csum) {
 				/* FIXME: Never tested this code... */
 				*(gre_csum(greh)) = 
-					ip_nat_cheat_check(~*(gre_key(greh)),
+					nf_proto_csum_update(*pskb,
+							~*(gre_key(greh)),
 							tuple->dst.u.gre.key,
-							*(gre_csum(greh)));
+							*(gre_csum(greh)), 0);
 			}
 			*(gre_key(greh)) = tuple->dst.u.gre.key;
 			break;
diff --git a/net/ipv4/netfilter/ip_nat_proto_icmp.c b/net/ipv4/netfilter/ip_nat_proto_icmp.c
index 31a3f4ccb99c..ec50cc295317 100644
--- a/net/ipv4/netfilter/ip_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/ip_nat_proto_icmp.c
@@ -66,10 +66,10 @@ icmp_manip_pkt(struct sk_buff **pskb,
 		return 0;
 
 	hdr = (struct icmphdr *)((*pskb)->data + hdroff);
-
-	hdr->checksum = ip_nat_cheat_check(hdr->un.echo.id ^ 0xFFFF,
-					    tuple->src.u.icmp.id,
-					    hdr->checksum);
+	hdr->checksum = nf_proto_csum_update(*pskb,
+					     hdr->un.echo.id ^ 0xFFFF,
+					     tuple->src.u.icmp.id,
+					     hdr->checksum, 0);
 	hdr->un.echo.id = tuple->src.u.icmp.id;
 	return 1;
 }
diff --git a/net/ipv4/netfilter/ip_nat_proto_tcp.c b/net/ipv4/netfilter/ip_nat_proto_tcp.c
index a3d14079eba6..72a6307bd2db 100644
--- a/net/ipv4/netfilter/ip_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_nat_proto_tcp.c
@@ -129,10 +129,9 @@ tcp_manip_pkt(struct sk_buff **pskb,
 	if (hdrsize < sizeof(*hdr))
 		return 1;
 
-	hdr->check = ip_nat_cheat_check(~oldip, newip,
-					ip_nat_cheat_check(oldport ^ 0xFFFF,
-							   newport,
-							   hdr->check));
+	hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip, hdr->check, 1);
+	hdr->check = nf_proto_csum_update(*pskb, oldport ^ 0xFFFF, newport,
+					  hdr->check, 0);
 	return 1;
 }
 
diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c
index ec6053fdc867..5da196ae758c 100644
--- a/net/ipv4/netfilter/ip_nat_proto_udp.c
+++ b/net/ipv4/netfilter/ip_nat_proto_udp.c
@@ -113,11 +113,16 @@ udp_manip_pkt(struct sk_buff **pskb,
 		newport = tuple->dst.u.udp.port;
 		portptr = &hdr->dest;
 	}
-	if (hdr->check) /* 0 is a special case meaning no checksum */
-		hdr->check = ip_nat_cheat_check(~oldip, newip,
-					ip_nat_cheat_check(*portptr ^ 0xFFFF,
-							   newport,
-							   hdr->check));
+
+	if (hdr->check || (*pskb)->ip_summed == CHECKSUM_PARTIAL) {
+		hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip,
+						  hdr->check, 1);
+		hdr->check = nf_proto_csum_update(*pskb,
+						  *portptr ^ 0xFFFF, newport,
+						  hdr->check, 0);
+		if (!hdr->check)
+			hdr->check = -1;
+	}
 	*portptr = newport;
 	return 1;
 }
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index f4f00c816d87..f3b778355432 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -110,12 +110,6 @@ ip_nat_fn(unsigned int hooknum,
 	IP_NF_ASSERT(!((*pskb)->nh.iph->frag_off
 		       & htons(IP_MF|IP_OFFSET)));
 
-	/* If we had a hardware checksum before, it's now invalid */
-	if ((*pskb)->ip_summed == CHECKSUM_PARTIAL ||
-	    (*pskb)->ip_summed == CHECKSUM_COMPLETE)
-		if (skb_checksum_help(*pskb))
-			return NF_DROP;
-
 	ct = ip_conntrack_get(*pskb, &ctinfo);
 	/* Can't track?  It's not due to stress, or conntrack would
 	   have dropped it.  Hence it's the user's responsibilty to
@@ -146,8 +140,8 @@ ip_nat_fn(unsigned int hooknum,
 	case IP_CT_RELATED:
 	case IP_CT_RELATED+IP_CT_IS_REPLY:
 		if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) {
-			if (!ip_nat_icmp_reply_translation(pskb, ct, maniptype,
-							   CTINFO2DIR(ctinfo)))
+			if (!ip_nat_icmp_reply_translation(ct, ctinfo,
+							   hooknum, pskb))
 				return NF_DROP;
 			else
 				return NF_ACCEPT;
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index 4ec43f98fe49..35916c74fe4e 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -52,7 +52,7 @@ static inline int
 set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 {
 	struct tcphdr _tcph, *tcph;
-	u_int16_t diffs[2];
+	u_int16_t oldval;
 
 	/* Not enought header? */
 	tcph = skb_header_pointer(*pskb, (*pskb)->nh.iph->ihl*4,
@@ -70,23 +70,16 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 		return 0;
 	tcph = (void *)(*pskb)->nh.iph + (*pskb)->nh.iph->ihl*4;
 
-	if (((*pskb)->ip_summed == CHECKSUM_PARTIAL ||
-	     (*pskb)->ip_summed == CHECKSUM_COMPLETE) &&
-	    skb_checksum_help(*pskb))
-		return 0;
-
-	diffs[0] = ((u_int16_t *)tcph)[6];
+	oldval = ((u_int16_t *)tcph)[6];
 	if (einfo->operation & IPT_ECN_OP_SET_ECE)
 		tcph->ece = einfo->proto.tcp.ece;
 	if (einfo->operation & IPT_ECN_OP_SET_CWR)
 		tcph->cwr = einfo->proto.tcp.cwr;
-	diffs[1] = ((u_int16_t *)tcph)[6];
-	diffs[0] = diffs[0] ^ 0xFFFF;
 
-	if ((*pskb)->ip_summed != CHECKSUM_UNNECESSARY)
-		tcph->check = csum_fold(csum_partial((char *)diffs,
-						     sizeof(diffs),
-						     tcph->check^0xFFFF));
+	tcph->check = nf_proto_csum_update((*pskb),
+					   oldval ^ 0xFFFF,
+					   ((u_int16_t *)tcph)[6],
+					   tcph->check, 0);
 	return 1;
 }
 
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index 7f905bf2bde5..95c6662b663c 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -185,6 +185,7 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 	tcph->urg_ptr = 0;
 
 	/* Adjust TCP checksum */
+	nskb->ip_summed = CHECKSUM_NONE;
 	tcph->check = 0;
 	tcph->check = tcp_v4_check(tcph, sizeof(struct tcphdr),
 				   nskb->nh.iph->saddr,
diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c
index c998dc0fcd15..0fce85e05507 100644
--- a/net/ipv4/netfilter/ipt_TCPMSS.c
+++ b/net/ipv4/netfilter/ipt_TCPMSS.c
@@ -27,14 +27,6 @@ MODULE_DESCRIPTION("iptables TCP MSS modification module");
 #define DEBUGP(format, args...)
 #endif
 
-static u_int16_t
-cheat_check(u_int32_t oldvalinv, u_int32_t newval, u_int16_t oldcheck)
-{
-	u_int32_t diffs[] = { oldvalinv, newval };
-	return csum_fold(csum_partial((char *)diffs, sizeof(diffs),
-                                      oldcheck^0xFFFF));
-}
-
 static inline unsigned int
 optlen(const u_int8_t *opt, unsigned int offset)
 {
@@ -62,11 +54,6 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	if (!skb_make_writable(pskb, (*pskb)->len))
 		return NF_DROP;
 
-	if (((*pskb)->ip_summed == CHECKSUM_PARTIAL ||
-	     (*pskb)->ip_summed == CHECKSUM_COMPLETE) &&
-	    skb_checksum_help(*pskb))
-		return NF_DROP;
-
 	iph = (*pskb)->nh.iph;
 	tcplen = (*pskb)->len - iph->ihl*4;
 
@@ -120,9 +107,10 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 			opt[i+2] = (newmss & 0xff00) >> 8;
 			opt[i+3] = (newmss & 0x00ff);
 
-			tcph->check = cheat_check(htons(oldmss)^0xFFFF,
-						  htons(newmss),
-						  tcph->check);
+			tcph->check = nf_proto_csum_update(*pskb,
+							   htons(oldmss)^0xFFFF,
+							   htons(newmss),
+							   tcph->check, 0);
 
 			DEBUGP(KERN_INFO "ipt_tcpmss_target: %u.%u.%u.%u:%hu"
 			       "->%u.%u.%u.%u:%hu changed TCP MSS option"
@@ -162,8 +150,10 @@ ipt_tcpmss_target(struct sk_buff **pskb,
  	opt = (u_int8_t *)tcph + sizeof(struct tcphdr);
 	memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr));
 
-	tcph->check = cheat_check(htons(tcplen) ^ 0xFFFF,
-				  htons(tcplen + TCPOLEN_MSS), tcph->check);
+	tcph->check = nf_proto_csum_update(*pskb,
+					   htons(tcplen) ^ 0xFFFF,
+				           htons(tcplen + TCPOLEN_MSS),
+					   tcph->check, 1);
 	tcplen += TCPOLEN_MSS;
 
 	opt[0] = TCPOPT_MSS;
@@ -171,16 +161,19 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	opt[2] = (newmss & 0xff00) >> 8;
 	opt[3] = (newmss & 0x00ff);
 
-	tcph->check = cheat_check(~0, *((u_int32_t *)opt), tcph->check);
+	tcph->check = nf_proto_csum_update(*pskb, ~0, *((u_int32_t *)opt),
+					   tcph->check, 0);
 
 	oldval = ((u_int16_t *)tcph)[6];
 	tcph->doff += TCPOLEN_MSS/4;
-	tcph->check = cheat_check(oldval ^ 0xFFFF,
-				  ((u_int16_t *)tcph)[6], tcph->check);
+	tcph->check = nf_proto_csum_update(*pskb,
+					   oldval ^ 0xFFFF,
+					   ((u_int16_t *)tcph)[6],
+					   tcph->check, 0);
 
 	newtotlen = htons(ntohs(iph->tot_len) + TCPOLEN_MSS);
-	iph->check = cheat_check(iph->tot_len ^ 0xFFFF,
-				 newtotlen, iph->check);
+	iph->check = nf_csum_update(iph->tot_len ^ 0xFFFF,
+				    newtotlen, iph->check);
 	iph->tot_len = newtotlen;
 
 	DEBUGP(KERN_INFO "ipt_tcpmss_target: %u.%u.%u.%u:%hu"
diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index 5d29d5e23624..27f639f3ac2a 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -222,6 +222,28 @@ copy_skb:
 }
 EXPORT_SYMBOL(skb_make_writable);
 
+u_int16_t nf_csum_update(u_int32_t oldval, u_int32_t newval, u_int32_t csum)
+{
+	u_int32_t diff[] = { oldval, newval };
+
+	return csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum));
+}
+EXPORT_SYMBOL(nf_csum_update);
+
+u_int16_t nf_proto_csum_update(struct sk_buff *skb,
+			       u_int32_t oldval, u_int32_t newval,
+			       u_int16_t csum, int pseudohdr)
+{
+	if (skb->ip_summed != CHECKSUM_PARTIAL) {
+		csum = nf_csum_update(oldval, newval, csum);
+		if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr)
+			skb->csum = nf_csum_update(oldval, newval, skb->csum);
+	} else if (pseudohdr)
+		csum = ~nf_csum_update(oldval, newval, ~csum);
+
+	return csum;
+}
+EXPORT_SYMBOL(nf_proto_csum_update);
 
 /* This does not belong here, but locally generated errors need it if connection
    tracking in use: without this, connection may not be in hash table, and hence
-- 
cgit v1.2.3


From 9067c722cf6930adf1df2d169de9094dd90b0c33 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Mon, 7 Aug 2006 17:57:44 -0700
Subject: [NEIGH]: Move netlink neighbour bits to linux/neighbour.h

Moves netlink neighbour bits to linux/neighbour.h. Also
moves bits to be exported to userspace from net/neighbour.h
to linux/neighbour.h and removes __KERNEL__ guards, userspace
is not supposed to be using it.

rtnetlink_rcv_msg() is not longer required to parse attributes
for the neighbour layer, remove dependency on obsolete and
buggy rta_buf.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/neighbour.h | 65 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/rtnetlink.h | 63 ---------------------------------------------
 include/net/neighbour.h   | 39 ++--------------------------
 net/core/rtnetlink.c      |  2 --
 4 files changed, 67 insertions(+), 102 deletions(-)
 create mode 100644 include/linux/neighbour.h

(limited to 'include')

diff --git a/include/linux/neighbour.h b/include/linux/neighbour.h
new file mode 100644
index 000000000000..8e8293d86fbb
--- /dev/null
+++ b/include/linux/neighbour.h
@@ -0,0 +1,65 @@
+#ifndef __LINUX_NEIGHBOUR_H
+#define __LINUX_NEIGHBOUR_H
+
+#include <linux/netlink.h>
+
+struct ndmsg
+{
+	__u8		ndm_family;
+	__u8		ndm_pad1;
+	__u16		ndm_pad2;
+	__s32		ndm_ifindex;
+	__u16		ndm_state;
+	__u8		ndm_flags;
+	__u8		ndm_type;
+};
+
+enum
+{
+	NDA_UNSPEC,
+	NDA_DST,
+	NDA_LLADDR,
+	NDA_CACHEINFO,
+	NDA_PROBES,
+	__NDA_MAX
+};
+
+#define NDA_MAX (__NDA_MAX - 1)
+
+/*
+ *	Neighbor Cache Entry Flags
+ */
+
+#define NTF_PROXY	0x08	/* == ATF_PUBL */
+#define NTF_ROUTER	0x80
+
+/*
+ *	Neighbor Cache Entry States.
+ */
+
+#define NUD_INCOMPLETE	0x01
+#define NUD_REACHABLE	0x02
+#define NUD_STALE	0x04
+#define NUD_DELAY	0x08
+#define NUD_PROBE	0x10
+#define NUD_FAILED	0x20
+
+/* Dummy states */
+#define NUD_NOARP	0x40
+#define NUD_PERMANENT	0x80
+#define NUD_NONE	0x00
+
+/* NUD_NOARP & NUD_PERMANENT are pseudostates, they never change
+   and make no address resolution or NUD.
+   NUD_PERMANENT is also cannot be deleted by garbage collectors.
+ */
+
+struct nda_cacheinfo
+{
+	__u32		ndm_confirmed;
+	__u32		ndm_used;
+	__u32		ndm_updated;
+	__u32		ndm_refcnt;
+};
+
+#endif
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 84f3eb426da2..9750f0214c22 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -386,69 +386,6 @@ struct rta_session
 
 
 
-/**************************************************************
- *		Neighbour discovery.
- ****/
-
-struct ndmsg
-{
-	unsigned char	ndm_family;
-	unsigned char	ndm_pad1;
-	unsigned short	ndm_pad2;
-	int		ndm_ifindex;	/* Link index			*/
-	__u16		ndm_state;
-	__u8		ndm_flags;
-	__u8		ndm_type;
-};
-
-enum
-{
-	NDA_UNSPEC,
-	NDA_DST,
-	NDA_LLADDR,
-	NDA_CACHEINFO,
-	NDA_PROBES,
-	__NDA_MAX
-};
-
-#define NDA_MAX (__NDA_MAX - 1)
-
-#define NDA_RTA(r)  ((struct rtattr*)(((char*)(r)) + NLMSG_ALIGN(sizeof(struct ndmsg))))
-#define NDA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ndmsg))
-
-/*
- *	Neighbor Cache Entry Flags
- */
-
-#define NTF_PROXY	0x08	/* == ATF_PUBL */
-#define NTF_ROUTER	0x80
-
-/*
- *	Neighbor Cache Entry States.
- */
-
-#define NUD_INCOMPLETE	0x01
-#define NUD_REACHABLE	0x02
-#define NUD_STALE	0x04
-#define NUD_DELAY	0x08
-#define NUD_PROBE	0x10
-#define NUD_FAILED	0x20
-
-/* Dummy states */
-#define NUD_NOARP	0x40
-#define NUD_PERMANENT	0x80
-#define NUD_NONE	0x00
-
-
-struct nda_cacheinfo
-{
-	__u32		ndm_confirmed;
-	__u32		ndm_used;
-	__u32		ndm_updated;
-	__u32		ndm_refcnt;
-};
-
-
 /*****************************************************************
  *		Neighbour tables specific messages.
  *
diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 4901ee446879..74c4b6ff8a5c 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -1,6 +1,8 @@
 #ifndef _NET_NEIGHBOUR_H
 #define _NET_NEIGHBOUR_H
 
+#include <linux/neighbour.h>
+
 /*
  *	Generic neighbour manipulation
  *
@@ -14,40 +16,6 @@
  *		- Add neighbour cache statistics like rtstat
  */
 
-/* The following flags & states are exported to user space,
-   so that they should be moved to include/linux/ directory.
- */
-
-/*
- *	Neighbor Cache Entry Flags
- */
-
-#define NTF_PROXY	0x08	/* == ATF_PUBL */
-#define NTF_ROUTER	0x80
-
-/*
- *	Neighbor Cache Entry States.
- */
-
-#define NUD_INCOMPLETE	0x01
-#define NUD_REACHABLE	0x02
-#define NUD_STALE	0x04
-#define NUD_DELAY	0x08
-#define NUD_PROBE	0x10
-#define NUD_FAILED	0x20
-
-/* Dummy states */
-#define NUD_NOARP	0x40
-#define NUD_PERMANENT	0x80
-#define NUD_NONE	0x00
-
-/* NUD_NOARP & NUD_PERMANENT are pseudostates, they never change
-   and make no address resolution or NUD.
-   NUD_PERMANENT is also cannot be deleted by garbage collectors.
- */
-
-#ifdef __KERNEL__
-
 #include <asm/atomic.h>
 #include <linux/netdevice.h>
 #include <linux/skbuff.h>
@@ -374,6 +342,3 @@ struct neighbour_cb {
 #define NEIGH_CB(skb)	((struct neighbour_cb *)(skb)->cb)
 
 #endif
-#endif
-
-
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 93ba04fb8444..78ccbd4c4e37 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -104,7 +104,6 @@ static const int rtm_min[RTM_NR_FAMILIES] =
 	[RTM_FAM(RTM_NEWLINK)]      = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
 	[RTM_FAM(RTM_NEWADDR)]      = NLMSG_LENGTH(sizeof(struct ifaddrmsg)),
 	[RTM_FAM(RTM_NEWROUTE)]     = NLMSG_LENGTH(sizeof(struct rtmsg)),
-	[RTM_FAM(RTM_NEWNEIGH)]     = NLMSG_LENGTH(sizeof(struct ndmsg)),
 	[RTM_FAM(RTM_NEWRULE)]      = NLMSG_LENGTH(sizeof(struct fib_rule_hdr)),
 	[RTM_FAM(RTM_NEWQDISC)]     = NLMSG_LENGTH(sizeof(struct tcmsg)),
 	[RTM_FAM(RTM_NEWTCLASS)]    = NLMSG_LENGTH(sizeof(struct tcmsg)),
@@ -121,7 +120,6 @@ static const int rta_max[RTM_NR_FAMILIES] =
 	[RTM_FAM(RTM_NEWLINK)]      = IFLA_MAX,
 	[RTM_FAM(RTM_NEWADDR)]      = IFA_MAX,
 	[RTM_FAM(RTM_NEWROUTE)]     = RTA_MAX,
-	[RTM_FAM(RTM_NEWNEIGH)]     = NDA_MAX,
 	[RTM_FAM(RTM_NEWRULE)]      = FRA_MAX,
 	[RTM_FAM(RTM_NEWQDISC)]     = TCA_MAX,
 	[RTM_FAM(RTM_NEWTCLASS)]    = TCA_MAX,
-- 
cgit v1.2.3


From b63bbc5006a0a62fabc81c4f77e95f16ff16f340 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Mon, 7 Aug 2006 18:00:57 -0700
Subject: [NEIGH]: Move netlink neighbour table bits to linux/neighbour.h

rtnetlink_rcv_msg() is not longer required to parse attributes
for the neighbour tables layer, remove dependency on obsolete and
buggy rta_buf.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/neighbour.h |  94 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/rtnetlink.h | 101 ----------------------------------------------
 net/core/rtnetlink.c      |   2 -
 3 files changed, 94 insertions(+), 103 deletions(-)

(limited to 'include')

diff --git a/include/linux/neighbour.h b/include/linux/neighbour.h
index 8e8293d86fbb..bd3bbf668cdb 100644
--- a/include/linux/neighbour.h
+++ b/include/linux/neighbour.h
@@ -62,4 +62,98 @@ struct nda_cacheinfo
 	__u32		ndm_refcnt;
 };
 
+/*****************************************************************
+ *		Neighbour tables specific messages.
+ *
+ * To retrieve the neighbour tables send RTM_GETNEIGHTBL with the
+ * NLM_F_DUMP flag set. Every neighbour table configuration is
+ * spread over multiple messages to avoid running into message
+ * size limits on systems with many interfaces. The first message
+ * in the sequence transports all not device specific data such as
+ * statistics, configuration, and the default parameter set.
+ * This message is followed by 0..n messages carrying device
+ * specific parameter sets.
+ * Although the ordering should be sufficient, NDTA_NAME can be
+ * used to identify sequences. The initial message can be identified
+ * by checking for NDTA_CONFIG. The device specific messages do
+ * not contain this TLV but have NDTPA_IFINDEX set to the
+ * corresponding interface index.
+ *
+ * To change neighbour table attributes, send RTM_SETNEIGHTBL
+ * with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3],
+ * NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked
+ * otherwise. Device specific parameter sets can be changed by
+ * setting NDTPA_IFINDEX to the interface index of the corresponding
+ * device.
+ ****/
+
+struct ndt_stats
+{
+	__u64		ndts_allocs;
+	__u64		ndts_destroys;
+	__u64		ndts_hash_grows;
+	__u64		ndts_res_failed;
+	__u64		ndts_lookups;
+	__u64		ndts_hits;
+	__u64		ndts_rcv_probes_mcast;
+	__u64		ndts_rcv_probes_ucast;
+	__u64		ndts_periodic_gc_runs;
+	__u64		ndts_forced_gc_runs;
+};
+
+enum {
+	NDTPA_UNSPEC,
+	NDTPA_IFINDEX,			/* u32, unchangeable */
+	NDTPA_REFCNT,			/* u32, read-only */
+	NDTPA_REACHABLE_TIME,		/* u64, read-only, msecs */
+	NDTPA_BASE_REACHABLE_TIME,	/* u64, msecs */
+	NDTPA_RETRANS_TIME,		/* u64, msecs */
+	NDTPA_GC_STALETIME,		/* u64, msecs */
+	NDTPA_DELAY_PROBE_TIME,		/* u64, msecs */
+	NDTPA_QUEUE_LEN,		/* u32 */
+	NDTPA_APP_PROBES,		/* u32 */
+	NDTPA_UCAST_PROBES,		/* u32 */
+	NDTPA_MCAST_PROBES,		/* u32 */
+	NDTPA_ANYCAST_DELAY,		/* u64, msecs */
+	NDTPA_PROXY_DELAY,		/* u64, msecs */
+	NDTPA_PROXY_QLEN,		/* u32 */
+	NDTPA_LOCKTIME,			/* u64, msecs */
+	__NDTPA_MAX
+};
+#define NDTPA_MAX (__NDTPA_MAX - 1)
+
+struct ndtmsg
+{
+	__u8		ndtm_family;
+	__u8		ndtm_pad1;
+	__u16		ndtm_pad2;
+};
+
+struct ndt_config
+{
+	__u16		ndtc_key_len;
+	__u16		ndtc_entry_size;
+	__u32		ndtc_entries;
+	__u32		ndtc_last_flush;	/* delta to now in msecs */
+	__u32		ndtc_last_rand;		/* delta to now in msecs */
+	__u32		ndtc_hash_rnd;
+	__u32		ndtc_hash_mask;
+	__u32		ndtc_hash_chain_gc;
+	__u32		ndtc_proxy_qlen;
+};
+
+enum {
+	NDTA_UNSPEC,
+	NDTA_NAME,			/* char *, unchangeable */
+	NDTA_THRESH1,			/* u32 */
+	NDTA_THRESH2,			/* u32 */
+	NDTA_THRESH3,			/* u32 */
+	NDTA_CONFIG,			/* struct ndt_config, read-only */
+	NDTA_PARMS,			/* nested TLV NDTPA_* */
+	NDTA_STATS,			/* struct ndt_stats, read-only */
+	NDTA_GC_INTERVAL,		/* u64, msecs */
+	__NDTA_MAX
+};
+#define NDTA_MAX (__NDTA_MAX - 1)
+
 #endif
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 9750f0214c22..784a1a29490e 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -384,107 +384,6 @@ struct rta_session
 	} u;
 };
 
-
-
-/*****************************************************************
- *		Neighbour tables specific messages.
- *
- * To retrieve the neighbour tables send RTM_GETNEIGHTBL with the
- * NLM_F_DUMP flag set. Every neighbour table configuration is
- * spread over multiple messages to avoid running into message
- * size limits on systems with many interfaces. The first message
- * in the sequence transports all not device specific data such as
- * statistics, configuration, and the default parameter set.
- * This message is followed by 0..n messages carrying device
- * specific parameter sets.
- * Although the ordering should be sufficient, NDTA_NAME can be
- * used to identify sequences. The initial message can be identified
- * by checking for NDTA_CONFIG. The device specific messages do
- * not contain this TLV but have NDTPA_IFINDEX set to the
- * corresponding interface index.
- *
- * To change neighbour table attributes, send RTM_SETNEIGHTBL
- * with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3],
- * NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked
- * otherwise. Device specific parameter sets can be changed by
- * setting NDTPA_IFINDEX to the interface index of the corresponding
- * device.
- ****/
-
-struct ndt_stats
-{
-	__u64		ndts_allocs;
-	__u64		ndts_destroys;
-	__u64		ndts_hash_grows;
-	__u64		ndts_res_failed;
-	__u64		ndts_lookups;
-	__u64		ndts_hits;
-	__u64		ndts_rcv_probes_mcast;
-	__u64		ndts_rcv_probes_ucast;
-	__u64		ndts_periodic_gc_runs;
-	__u64		ndts_forced_gc_runs;
-};
-
-enum {
-	NDTPA_UNSPEC,
-	NDTPA_IFINDEX,			/* u32, unchangeable */
-	NDTPA_REFCNT,			/* u32, read-only */
-	NDTPA_REACHABLE_TIME,		/* u64, read-only, msecs */
-	NDTPA_BASE_REACHABLE_TIME,	/* u64, msecs */
-	NDTPA_RETRANS_TIME,		/* u64, msecs */
-	NDTPA_GC_STALETIME,		/* u64, msecs */
-	NDTPA_DELAY_PROBE_TIME,		/* u64, msecs */
-	NDTPA_QUEUE_LEN,		/* u32 */
-	NDTPA_APP_PROBES,		/* u32 */
-	NDTPA_UCAST_PROBES,		/* u32 */
-	NDTPA_MCAST_PROBES,		/* u32 */
-	NDTPA_ANYCAST_DELAY,		/* u64, msecs */
-	NDTPA_PROXY_DELAY,		/* u64, msecs */
-	NDTPA_PROXY_QLEN,		/* u32 */
-	NDTPA_LOCKTIME,			/* u64, msecs */
-	__NDTPA_MAX
-};
-#define NDTPA_MAX (__NDTPA_MAX - 1)
-
-struct ndtmsg
-{
-	__u8		ndtm_family;
-	__u8		ndtm_pad1;
-	__u16		ndtm_pad2;
-};
-
-struct ndt_config
-{
-	__u16		ndtc_key_len;
-	__u16		ndtc_entry_size;
-	__u32		ndtc_entries;
-	__u32		ndtc_last_flush;	/* delta to now in msecs */
-	__u32		ndtc_last_rand;		/* delta to now in msecs */
-	__u32		ndtc_hash_rnd;
-	__u32		ndtc_hash_mask;
-	__u32		ndtc_hash_chain_gc;
-	__u32		ndtc_proxy_qlen;
-};
-
-enum {
-	NDTA_UNSPEC,
-	NDTA_NAME,			/* char *, unchangeable */
-	NDTA_THRESH1,			/* u32 */
-	NDTA_THRESH2,			/* u32 */
-	NDTA_THRESH3,			/* u32 */
-	NDTA_CONFIG,			/* struct ndt_config, read-only */
-	NDTA_PARMS,			/* nested TLV NDTPA_* */
-	NDTA_STATS,			/* struct ndt_stats, read-only */
-	NDTA_GC_INTERVAL,		/* u64, msecs */
-	__NDTA_MAX
-};
-#define NDTA_MAX (__NDTA_MAX - 1)
-
-#define NDTA_RTA(r) ((struct rtattr*)(((char*)(r)) + \
-		     NLMSG_ALIGN(sizeof(struct ndtmsg))))
-#define NDTA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ndtmsg))
-
-
 /****
  *		General form of address family dependent message.
  ****/
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 78ccbd4c4e37..a1b783a6afc6 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -112,7 +112,6 @@ static const int rtm_min[RTM_NR_FAMILIES] =
 	[RTM_FAM(RTM_NEWPREFIX)]    = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
 	[RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
 	[RTM_FAM(RTM_GETANYCAST)]   = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
-	[RTM_FAM(RTM_NEWNEIGHTBL)]  = NLMSG_LENGTH(sizeof(struct ndtmsg)),
 };
 
 static const int rta_max[RTM_NR_FAMILIES] =
@@ -125,7 +124,6 @@ static const int rta_max[RTM_NR_FAMILIES] =
 	[RTM_FAM(RTM_NEWTCLASS)]    = TCA_MAX,
 	[RTM_FAM(RTM_NEWTFILTER)]   = TCA_MAX,
 	[RTM_FAM(RTM_NEWACTION)]    = TCAA_MAX,
-	[RTM_FAM(RTM_NEWNEIGHTBL)]  = NDTA_MAX,
 };
 
 void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data)
-- 
cgit v1.2.3


From ac5a488ef252ed673cb067843e411f8cc43f7ab9 Mon Sep 17 00:00:00 2001
From: Sridhar Samudrala <sri@us.ibm.com>
Date: Mon, 7 Aug 2006 20:57:31 -0700
Subject: [NET]: Round out in-kernel sockets API

This patch implements wrapper functions that provide a convenient way
to access the sockets API for in-kernel users like sunrpc, cifs &
ocfs2 etc and any future users.

Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h |  19 +++++++++
 net/socket.c        | 113 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 132 insertions(+)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index b20c53c74413..19da2c08d7b6 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -208,6 +208,25 @@ extern int   	     kernel_recvmsg(struct socket *sock, struct msghdr *msg,
 				    struct kvec *vec, size_t num,
 				    size_t len, int flags);
 
+extern int kernel_bind(struct socket *sock, struct sockaddr *addr,
+		       int addrlen);
+extern int kernel_listen(struct socket *sock, int backlog);
+extern int kernel_accept(struct socket *sock, struct socket **newsock,
+			 int flags);
+extern int kernel_connect(struct socket *sock, struct sockaddr *addr,
+			  int addrlen, int flags);
+extern int kernel_getsockname(struct socket *sock, struct sockaddr *addr,
+			      int *addrlen);
+extern int kernel_getpeername(struct socket *sock, struct sockaddr *addr,
+			      int *addrlen);
+extern int kernel_getsockopt(struct socket *sock, int level, int optname,
+			     char *optval, int *optlen);
+extern int kernel_setsockopt(struct socket *sock, int level, int optname,
+			     char *optval, int optlen);
+extern int kernel_sendpage(struct socket *sock, struct page *page, int offset,
+			   size_t size, int flags);
+extern int kernel_sock_ioctl(struct socket *sock, int cmd, unsigned long arg);
+
 #ifndef CONFIG_SMP
 #define SOCKOPS_WRAPPED(name) name
 #define SOCKOPS_WRAP(name, fam)
diff --git a/net/socket.c b/net/socket.c
index 6756e57e1ff0..2eaebf934a1a 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2170,6 +2170,109 @@ static long compat_sock_ioctl(struct file *file, unsigned cmd,
 }
 #endif
 
+int kernel_bind(struct socket *sock, struct sockaddr *addr, int addrlen)
+{
+	return sock->ops->bind(sock, addr, addrlen);
+}
+
+int kernel_listen(struct socket *sock, int backlog)
+{
+	return sock->ops->listen(sock, backlog);
+}
+
+int kernel_accept(struct socket *sock, struct socket **newsock, int flags)
+{
+	struct sock *sk = sock->sk;
+	int err;
+
+	err = sock_create_lite(sk->sk_family, sk->sk_type, sk->sk_protocol,
+			       newsock);
+	if (err < 0)
+		goto done;
+
+	err = sock->ops->accept(sock, *newsock, flags);
+	if (err < 0) {
+		sock_release(*newsock);
+		goto done;
+	}
+
+	(*newsock)->ops = sock->ops;
+
+done:
+	return err;
+}
+
+int kernel_connect(struct socket *sock, struct sockaddr *addr, int addrlen,
+                   int flags)
+{
+	return sock->ops->connect(sock, addr, addrlen, flags);
+}
+
+int kernel_getsockname(struct socket *sock, struct sockaddr *addr,
+			 int *addrlen)
+{
+	return sock->ops->getname(sock, addr, addrlen, 0);
+}
+
+int kernel_getpeername(struct socket *sock, struct sockaddr *addr,
+			 int *addrlen)
+{
+	return sock->ops->getname(sock, addr, addrlen, 1);
+}
+
+int kernel_getsockopt(struct socket *sock, int level, int optname,
+			char *optval, int *optlen)
+{
+	mm_segment_t oldfs = get_fs();
+	int err;
+
+	set_fs(KERNEL_DS);
+	if (level == SOL_SOCKET)
+		err = sock_getsockopt(sock, level, optname, optval, optlen);
+	else
+		err = sock->ops->getsockopt(sock, level, optname, optval,
+					    optlen);
+	set_fs(oldfs);
+	return err;
+}
+
+int kernel_setsockopt(struct socket *sock, int level, int optname,
+			char *optval, int optlen)
+{
+	mm_segment_t oldfs = get_fs();
+	int err;
+
+	set_fs(KERNEL_DS);
+	if (level == SOL_SOCKET)
+		err = sock_setsockopt(sock, level, optname, optval, optlen);
+	else
+		err = sock->ops->setsockopt(sock, level, optname, optval,
+					    optlen);
+	set_fs(oldfs);
+	return err;
+}
+
+int kernel_sendpage(struct socket *sock, struct page *page, int offset,
+		    size_t size, int flags)
+{
+	if (sock->ops->sendpage)
+		return sock->ops->sendpage(sock, page, offset, size, flags);
+
+	return sock_no_sendpage(sock, page, offset, size, flags);
+}
+
+int kernel_sock_ioctl(struct socket *sock, int cmd, unsigned long arg)
+{
+	mm_segment_t oldfs = get_fs();
+	int err;
+
+	set_fs(KERNEL_DS);
+	err = sock->ops->ioctl(sock, cmd, arg);
+	set_fs(oldfs);
+
+	return err;
+}
+
 /* ABI emulation layers need these two */
 EXPORT_SYMBOL(move_addr_to_kernel);
 EXPORT_SYMBOL(move_addr_to_user);
@@ -2186,3 +2289,13 @@ EXPORT_SYMBOL(sock_wake_async);
 EXPORT_SYMBOL(sockfd_lookup);
 EXPORT_SYMBOL(kernel_sendmsg);
 EXPORT_SYMBOL(kernel_recvmsg);
+EXPORT_SYMBOL(kernel_bind);
+EXPORT_SYMBOL(kernel_listen);
+EXPORT_SYMBOL(kernel_accept);
+EXPORT_SYMBOL(kernel_connect);
+EXPORT_SYMBOL(kernel_getsockname);
+EXPORT_SYMBOL(kernel_getpeername);
+EXPORT_SYMBOL(kernel_getsockopt);
+EXPORT_SYMBOL(kernel_setsockopt);
+EXPORT_SYMBOL(kernel_sendpage);
+EXPORT_SYMBOL(kernel_sock_ioctl);
-- 
cgit v1.2.3


From 8ce11e6a9faf1f1c849b77104adc1642c46aee95 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 7 Aug 2006 21:50:48 -0700
Subject: [NET]: Make code static.

This patch makes needlessly global code static.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h              | 4 ----
 net/ipv4/cipso_ipv4.c              | 2 +-
 net/ipv4/fib_rules.c               | 4 ++--
 net/ipv6/fib6_rules.c              | 4 ++--
 net/ipv6/ip6_fib.c                 | 6 +++---
 net/ipv6/route.c                   | 6 +++---
 net/netlabel/netlabel_domainhash.c | 4 ++--
 7 files changed, 13 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 7b47e8d5a765..c0660cea9a2f 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -192,10 +192,6 @@ struct fib6_node		*fib6_locate(struct fib6_node *root,
 					     struct in6_addr *daddr, int dst_len,
 					     struct in6_addr *saddr, int src_len);
 
-extern void			fib6_clean_tree(struct fib6_node *root,
-						int (*func)(struct rt6_info *, void *arg),
-						int prune, void *arg);
-
 extern void			fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
 					       int prune, void *arg);
 
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index b82a101c95c5..80a2a0911b49 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -60,7 +60,7 @@ struct cipso_v4_domhsh_entry {
  * if in practice there are a lot of different DOIs this list should
  * probably be turned into a hash table or something similar so we
  * can do quick lookups. */
-DEFINE_SPINLOCK(cipso_v4_doi_list_lock);
+static DEFINE_SPINLOCK(cipso_v4_doi_list_lock);
 static struct list_head cipso_v4_doi_list = LIST_HEAD_INIT(cipso_v4_doi_list);
 
 /* Label mapping cache */
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 23ec6ae1a0f6..03d1e8a43a48 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -101,8 +101,8 @@ int fib_lookup(struct flowi *flp, struct fib_result *res)
 	return err;
 }
 
-int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, int flags,
-		     struct fib_lookup_arg *arg)
+static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp,
+			    int flags, struct fib_lookup_arg *arg)
 {
 	int err = -EAGAIN;
 	struct fib_table *tbl;
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 94a46ec967a4..bf9bba83b852 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -66,8 +66,8 @@ struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
 	return (struct dst_entry *) arg.result;
 }
 
-int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
-		     int flags, struct fib_lookup_arg *arg)
+static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
+			    int flags, struct fib_lookup_arg *arg)
 {
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index ce226c14bef5..1f2316187ca4 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1169,9 +1169,9 @@ static int fib6_clean_node(struct fib6_walker_t *w)
  *	ignoring pure split nodes) will be scanned.
  */
 
-void fib6_clean_tree(struct fib6_node *root,
-		     int (*func)(struct rt6_info *, void *arg),
-		     int prune, void *arg)
+static void fib6_clean_tree(struct fib6_node *root,
+			    int (*func)(struct rt6_info *, void *arg),
+			    int prune, void *arg)
 {
 	struct fib6_cleaner_t c;
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 41c5905d3191..e08d84063c1f 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -613,8 +613,8 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d
 	return rt;
 }
 
-struct rt6_info *ip6_pol_route_input(struct fib6_table *table, struct flowi *fl,
-				     int flags)
+static struct rt6_info *ip6_pol_route_input(struct fib6_table *table,
+					    struct flowi *fl, int flags)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt, *nrt;
@@ -872,7 +872,7 @@ static inline unsigned int ipv6_advmss(unsigned int mtu)
 }
 
 static struct dst_entry *ndisc_dst_gc_list;
-DEFINE_SPINLOCK(ndisc_lock);
+static DEFINE_SPINLOCK(ndisc_lock);
 
 struct dst_entry *ndisc_dst_alloc(struct net_device *dev, 
 				  struct neighbour *neigh,
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index 5bb3fad4a115..0489a1378101 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -50,11 +50,11 @@ struct netlbl_domhsh_tbl {
 /* Domain hash table */
 /* XXX - updates should be so rare that having one spinlock for the entire
  * hash table should be okay */
-DEFINE_SPINLOCK(netlbl_domhsh_lock);
+static DEFINE_SPINLOCK(netlbl_domhsh_lock);
 static struct netlbl_domhsh_tbl *netlbl_domhsh = NULL;
 
 /* Default domain mapping */
-DEFINE_SPINLOCK(netlbl_domhsh_def_lock);
+static DEFINE_SPINLOCK(netlbl_domhsh_def_lock);
 static struct netlbl_dom_map *netlbl_domhsh_def = NULL;
 
 /*
-- 
cgit v1.2.3


From 0298f36a579b5bd7f10f6f6d57e5929977a865a1 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Mon, 7 Aug 2006 21:56:52 -0700
Subject: [IPV4]: Kill fib4_rules_clean().

As noted by Adrian Bunk this function is totally unused.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h | 1 -
 net/ipv4/fib_rules.c | 5 -----
 2 files changed, 6 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 14c82e611c95..adf73586bc05 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -254,7 +254,6 @@ extern struct fib_table *fib_hash_init(int id);
 extern int fib4_rules_dump(struct sk_buff *skb, struct netlink_callback *cb);
 
 extern void __init fib4_rules_init(void);
-extern void __exit fib4_rules_cleanup(void);
 
 #ifdef CONFIG_NET_CLS_ROUTE
 extern u32 fib_rules_tclass(struct fib_result *res);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 03d1e8a43a48..d242e5291fcc 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -347,8 +347,3 @@ void __init fib4_rules_init(void)
 
 	fib_rules_register(&fib4_rules_ops);
 }
-
-void __exit fib4_rules_cleanup(void)
-{
-	fib_rules_unregister(&fib4_rules_ops);
-}
-- 
cgit v1.2.3


From 1a01912ae0a5666c4c24eaae2b4821711e2ad79a Mon Sep 17 00:00:00 2001
From: Louis Nyffenegger <louis.nyffenegger@gmail.com>
Date: Tue, 8 Aug 2006 00:56:11 -0700
Subject: [INET]: Remove is_setbyuser patch

The value is_setbyuser from struct ip_options is never used and set
only one time (http://linux-net.osdl.org/index.php/TODO#IPV4).
This little patch removes it from the kernel source.

Signed-off-by: Louis Nyffenegger <louis.nyffenegger@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 4 +---
 net/ipv4/ip_options.c   | 1 -
 2 files changed, 1 insertion(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index f4caad56cd03..f6242710f2ff 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -27,7 +27,6 @@
 /** struct ip_options - IP Options
  *
  * @faddr - Saved first hop address
- * @is_setbyuser - Set by setsockopt?
  * @is_data - Options in __data, rather than skb
  * @is_strictroute - Strict source route
  * @srr_is_hit - Packet destination addr was our one
@@ -42,8 +41,7 @@ struct ip_options {
 	unsigned char	srr;
 	unsigned char	rr;
 	unsigned char	ts;
-	unsigned char	is_setbyuser:1,
-			is_data:1,
+	unsigned char	is_data:1,
 			is_strictroute:1,
 			srr_is_hit:1,
 			is_changed:1,
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index e0a93b4fa8cc..e7437c091326 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -525,7 +525,6 @@ static int ip_options_get_finish(struct ip_options **optp,
 		opt->__data[optlen++] = IPOPT_END;
 	opt->optlen = optlen;
 	opt->is_data = 1;
-	opt->is_setbyuser = 1;
 	if (optlen && ip_options_compile(opt, NULL)) {
 		kfree(opt);
 		return -EINVAL;
-- 
cgit v1.2.3


From 99a92ff50424146ba01a222248fd47a1cd55b78f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 8 Aug 2006 02:18:10 -0700
Subject: [IPV4]: Uninline inet_lookup_listener

By modern standards this function is way too big to be inlined.  It's
even bigger than __inet_lookup_listener :)

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 33 ++-------------------------------
 net/ipv4/inet_hashtables.c    | 35 ++++++++++++++++++++++++++++++++---
 2 files changed, 34 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 98e0bb3014fe..bd513f3b9c7e 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -271,39 +271,10 @@ static inline int inet_iif(const struct sk_buff *skb)
 	return ((struct rtable *)skb->dst)->rt_iif;
 }
 
-extern struct sock *__inet_lookup_listener(const struct hlist_head *head,
-					   const u32 daddr,
-					   const unsigned short hnum,
-					   const int dif);
-
-/* Optimize the common listener case. */
-static inline struct sock *
+extern struct sock *
 		inet_lookup_listener(struct inet_hashinfo *hashinfo,
 				     const u32 daddr,
-				     const unsigned short hnum, const int dif)
-{
-	struct sock *sk = NULL;
-	const struct hlist_head *head;
-
-	read_lock(&hashinfo->lhash_lock);
-	head = &hashinfo->listening_hash[inet_lhashfn(hnum)];
-	if (!hlist_empty(head)) {
-		const struct inet_sock *inet = inet_sk((sk = __sk_head(head)));
-
-		if (inet->num == hnum && !sk->sk_node.next &&
-		    (!inet->rcv_saddr || inet->rcv_saddr == daddr) &&
-		    (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) &&
-		    !sk->sk_bound_dev_if)
-			goto sherry_cache;
-		sk = __inet_lookup_listener(head, daddr, hnum, dif);
-	}
-	if (sk) {
-sherry_cache:
-		sock_hold(sk);
-	}
-	read_unlock(&hashinfo->lhash_lock);
-	return sk;
-}
+				     const unsigned short hnum, const int dif);
 
 /* Socket demux engine toys. */
 #ifdef __BIG_ENDIAN
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 95fac5532994..bfc39066e730 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -124,8 +124,10 @@ EXPORT_SYMBOL(inet_listen_wlock);
  * remote address for the connection. So always assume those are both
  * wildcarded during the search since they can never be otherwise.
  */
-struct sock *__inet_lookup_listener(const struct hlist_head *head, const u32 daddr,
-				    const unsigned short hnum, const int dif)
+static struct sock *__inet_lookup_listener(const struct hlist_head *head,
+					   const u32 daddr,
+					   const unsigned short hnum,
+					   const int dif)
 {
 	struct sock *result = NULL, *sk;
 	const struct hlist_node *node;
@@ -159,7 +161,34 @@ struct sock *__inet_lookup_listener(const struct hlist_head *head, const u32 dad
 	return result;
 }
 
-EXPORT_SYMBOL_GPL(__inet_lookup_listener);
+/* Optimize the common listener case. */
+struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
+				  const u32 daddr, const unsigned short hnum,
+				  const int dif)
+{
+	struct sock *sk = NULL;
+	const struct hlist_head *head;
+
+	read_lock(&hashinfo->lhash_lock);
+	head = &hashinfo->listening_hash[inet_lhashfn(hnum)];
+	if (!hlist_empty(head)) {
+		const struct inet_sock *inet = inet_sk((sk = __sk_head(head)));
+
+		if (inet->num == hnum && !sk->sk_node.next &&
+		    (!inet->rcv_saddr || inet->rcv_saddr == daddr) &&
+		    (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) &&
+		    !sk->sk_bound_dev_if)
+			goto sherry_cache;
+		sk = __inet_lookup_listener(head, daddr, hnum, dif);
+	}
+	if (sk) {
+sherry_cache:
+		sock_hold(sk);
+	}
+	read_unlock(&hashinfo->lhash_lock);
+	return sk;
+}
+EXPORT_SYMBOL_GPL(inet_lookup_listener);
 
 /* called with local bh disabled */
 static int __inet_check_established(struct inet_timewait_death_row *death_row,
-- 
cgit v1.2.3


From 8f491069b40be5d627007a343f99759e9da6a178 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 9 Aug 2006 15:47:12 -0700
Subject: [IPV4]: Use network-order dport for all visible inet_lookup_*

Right now most inet_lookup_* functions take a host-order hnum instead
of a network-order dport because that's how it is represented
internally.

This means that users of these functions have to be careful about
using the right byte-order.  To add more confusion, inet_lookup takes
a network-order dport unlike all other functions.

So this patch changes all visible inet_lookup functions to take a
dport and move all dport->hnum conversion inside them.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 31 ++++++++++++++++++++++++-------
 net/dccp/ipv4.c               | 10 +++++-----
 net/ipv4/inet_hashtables.c    | 18 +++++++++---------
 net/ipv4/tcp_ipv4.c           | 10 +++++-----
 4 files changed, 43 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index bd513f3b9c7e..b4491c9e2a5a 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -271,10 +271,16 @@ static inline int inet_iif(const struct sk_buff *skb)
 	return ((struct rtable *)skb->dst)->rt_iif;
 }
 
-extern struct sock *
-		inet_lookup_listener(struct inet_hashinfo *hashinfo,
-				     const u32 daddr,
-				     const unsigned short hnum, const int dif);
+extern struct sock *__inet_lookup_listener(struct inet_hashinfo *hashinfo,
+					   const u32 daddr,
+					   const unsigned short hnum,
+					   const int dif);
+
+static inline struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
+						u32 daddr, u16 dport, int dif)
+{
+	return __inet_lookup_listener(hashinfo, daddr, ntohs(dport), dif);
+}
 
 /* Socket demux engine toys. */
 #ifdef __BIG_ENDIAN
@@ -362,14 +368,25 @@ hit:
 	goto out;
 }
 
+static inline struct sock *
+	inet_lookup_established(struct inet_hashinfo *hashinfo,
+				const u32 saddr, const u16 sport,
+				const u32 daddr, const u16 dport,
+				const int dif)
+{
+	return __inet_lookup_established(hashinfo, saddr, sport, daddr,
+					 ntohs(dport), dif);
+}
+
 static inline struct sock *__inet_lookup(struct inet_hashinfo *hashinfo,
 					 const u32 saddr, const u16 sport,
-					 const u32 daddr, const u16 hnum,
+					 const u32 daddr, const u16 dport,
 					 const int dif)
 {
+	u16 hnum = ntohs(dport);
 	struct sock *sk = __inet_lookup_established(hashinfo, saddr, sport, daddr,
 						    hnum, dif);
-	return sk ? : inet_lookup_listener(hashinfo, daddr, hnum, dif);
+	return sk ? : __inet_lookup_listener(hashinfo, daddr, hnum, dif);
 }
 
 static inline struct sock *inet_lookup(struct inet_hashinfo *hashinfo,
@@ -380,7 +397,7 @@ static inline struct sock *inet_lookup(struct inet_hashinfo *hashinfo,
 	struct sock *sk;
 
 	local_bh_disable();
-	sk = __inet_lookup(hashinfo, saddr, sport, daddr, ntohs(dport), dif);
+	sk = __inet_lookup(hashinfo, saddr, sport, daddr, dport, dif);
 	local_bh_enable();
 
 	return sk;
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 171d363876ee..9a1a76a7dc41 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -608,10 +608,10 @@ static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
 	if (req != NULL)
 		return dccp_check_req(sk, skb, req, prev);
 
-	nsk = __inet_lookup_established(&dccp_hashinfo,
-					iph->saddr, dh->dccph_sport,
-					iph->daddr, ntohs(dh->dccph_dport),
-					inet_iif(skb));
+	nsk = inet_lookup_established(&dccp_hashinfo,
+				      iph->saddr, dh->dccph_sport,
+				      iph->daddr, dh->dccph_dport,
+				      inet_iif(skb));
 	if (nsk != NULL) {
 		if (nsk->sk_state != DCCP_TIME_WAIT) {
 			bh_lock_sock(nsk);
@@ -925,7 +925,7 @@ static int dccp_v4_rcv(struct sk_buff *skb)
 	 * 	Look up flow ID in table and get corresponding socket */
 	sk = __inet_lookup(&dccp_hashinfo,
 			   skb->nh.iph->saddr, dh->dccph_sport,
-			   skb->nh.iph->daddr, ntohs(dh->dccph_dport),
+			   skb->nh.iph->daddr, dh->dccph_dport,
 			   inet_iif(skb));
 
 	/* 
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index bfc39066e730..fb296c9a7f3f 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -124,10 +124,10 @@ EXPORT_SYMBOL(inet_listen_wlock);
  * remote address for the connection. So always assume those are both
  * wildcarded during the search since they can never be otherwise.
  */
-static struct sock *__inet_lookup_listener(const struct hlist_head *head,
-					   const u32 daddr,
-					   const unsigned short hnum,
-					   const int dif)
+static struct sock *inet_lookup_listener_slow(const struct hlist_head *head,
+					      const u32 daddr,
+					      const unsigned short hnum,
+					      const int dif)
 {
 	struct sock *result = NULL, *sk;
 	const struct hlist_node *node;
@@ -162,9 +162,9 @@ static struct sock *__inet_lookup_listener(const struct hlist_head *head,
 }
 
 /* Optimize the common listener case. */
-struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
-				  const u32 daddr, const unsigned short hnum,
-				  const int dif)
+struct sock *__inet_lookup_listener(struct inet_hashinfo *hashinfo,
+				    const u32 daddr, const unsigned short hnum,
+				    const int dif)
 {
 	struct sock *sk = NULL;
 	const struct hlist_head *head;
@@ -179,7 +179,7 @@ struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
 		    (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) &&
 		    !sk->sk_bound_dev_if)
 			goto sherry_cache;
-		sk = __inet_lookup_listener(head, daddr, hnum, dif);
+		sk = inet_lookup_listener_slow(head, daddr, hnum, dif);
 	}
 	if (sk) {
 sherry_cache:
@@ -188,7 +188,7 @@ sherry_cache:
 	read_unlock(&hashinfo->lhash_lock);
 	return sk;
 }
-EXPORT_SYMBOL_GPL(inet_lookup_listener);
+EXPORT_SYMBOL_GPL(__inet_lookup_listener);
 
 /* called with local bh disabled */
 static int __inet_check_established(struct inet_timewait_death_row *death_row,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index b2aa512a30e9..2973dee0a489 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -951,9 +951,9 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
 	if (req)
 		return tcp_check_req(sk, skb, req, prev);
 
-	nsk = __inet_lookup_established(&tcp_hashinfo, skb->nh.iph->saddr,
-					th->source, skb->nh.iph->daddr,
-					ntohs(th->dest), inet_iif(skb));
+	nsk = inet_lookup_established(&tcp_hashinfo, skb->nh.iph->saddr,
+				      th->source, skb->nh.iph->daddr,
+				      th->dest, inet_iif(skb));
 
 	if (nsk) {
 		if (nsk->sk_state != TCP_TIME_WAIT) {
@@ -1090,7 +1090,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 	TCP_SKB_CB(skb)->sacked	 = 0;
 
 	sk = __inet_lookup(&tcp_hashinfo, skb->nh.iph->saddr, th->source,
-			   skb->nh.iph->daddr, ntohs(th->dest),
+			   skb->nh.iph->daddr, th->dest,
 			   inet_iif(skb));
 
 	if (!sk)
@@ -1168,7 +1168,7 @@ do_time_wait:
 	case TCP_TW_SYN: {
 		struct sock *sk2 = inet_lookup_listener(&tcp_hashinfo,
 							skb->nh.iph->daddr,
-							ntohs(th->dest),
+							th->dest,
 							inet_iif(skb));
 		if (sk2) {
 			inet_twsk_deschedule((struct inet_timewait_sock *)sk,
-- 
cgit v1.2.3


From a8731cbf61c8768ea129780b70dc7dfc6795aad4 Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <steve@chygwyn.com>
Date: Wed, 9 Aug 2006 15:56:46 -0700
Subject: [DECNET]: Covert rules to use generic code

This patch converts the DECnet rules code to use the generic
rules system created by Thomas Graf <tgraf@suug.ch>.

Signed-off-by: Steven Whitehouse <steve@chygwyn.com>
Acked-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h |   3 +-
 include/net/dn_fib.h      |   8 +-
 net/decnet/Kconfig        |   1 +
 net/decnet/af_decnet.c    |   1 +
 net/decnet/dn_dev.c       |   3 +-
 net/decnet/dn_fib.c       |   1 +
 net/decnet/dn_route.c     |   3 +-
 net/decnet/dn_rules.c     | 494 +++++++++++++++++-----------------------------
 net/decnet/dn_table.c     |   1 +
 9 files changed, 196 insertions(+), 319 deletions(-)

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 784a1a29490e..0aaffa2ae666 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -534,7 +534,8 @@ enum rtnetlink_groups {
 	RTNLGRP_NOP2,
 	RTNLGRP_DECnet_ROUTE,
 #define RTNLGRP_DECnet_ROUTE	RTNLGRP_DECnet_ROUTE
-	RTNLGRP_NOP3,
+	RTNLGRP_DECnet_RULE,
+#define RTNLGRP_DECnet_RULE	RTNLGRP_DECnet_RULE
 	RTNLGRP_NOP4,
 	RTNLGRP_IPV6_PREFIX,
 #define RTNLGRP_IPV6_PREFIX	RTNLGRP_IPV6_PREFIX
diff --git a/include/net/dn_fib.h b/include/net/dn_fib.h
index a15dcf0d5c1e..32bc8ce5c5ce 100644
--- a/include/net/dn_fib.h
+++ b/include/net/dn_fib.h
@@ -22,7 +22,7 @@ struct dn_kern_rta
 };
 
 struct dn_fib_res {
-	struct dn_fib_rule *r;
+	struct fib_rule *r;
 	struct dn_fib_info *fi;
 	unsigned char prefixlen;
 	unsigned char nh_sel;
@@ -147,10 +147,8 @@ extern void dn_fib_table_cleanup(void);
  */
 extern void dn_fib_rules_init(void);
 extern void dn_fib_rules_cleanup(void);
-extern void dn_fib_rule_put(struct dn_fib_rule *);
-extern __le16 dn_fib_rules_policy(__le16 saddr, struct dn_fib_res *res, unsigned *flags);
 extern unsigned dnet_addr_type(__le16 addr);
-extern int dn_fib_lookup(const struct flowi *fl, struct dn_fib_res *res);
+extern int dn_fib_lookup(struct flowi *fl, struct dn_fib_res *res);
 
 /*
  * rtnetlink interface
@@ -176,7 +174,7 @@ static inline void dn_fib_res_put(struct dn_fib_res *res)
 	if (res->fi)
 		dn_fib_info_put(res->fi);
 	if (res->r)
-		dn_fib_rule_put(res->r);
+		fib_rule_put(res->r);
 }
 
 extern struct dn_fib_table *dn_fib_tables[];
diff --git a/net/decnet/Kconfig b/net/decnet/Kconfig
index 92f2ec46fd22..36e72cb145b0 100644
--- a/net/decnet/Kconfig
+++ b/net/decnet/Kconfig
@@ -27,6 +27,7 @@ config DECNET
 config DECNET_ROUTER
 	bool "DECnet: router support (EXPERIMENTAL)"
 	depends on DECNET && EXPERIMENTAL
+	select FIB_RULES
 	---help---
 	  Add support for turning your DECnet Endnode into a level 1 or 2
 	  router.  This is an experimental, but functional option.  If you
diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index 5486247735f6..70e027375682 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -130,6 +130,7 @@ Version 0.0.6    2.1.110   07-aug-98   Eduardo Marcelo Serrat
 #include <linux/poll.h>
 #include <net/neighbour.h>
 #include <net/dst.h>
+#include <net/fib_rules.h>
 #include <net/dn.h>
 #include <net/dn_nsp.h>
 #include <net/dn_dev.h>
diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c
index 632c5a90b589..88ea7a13bb24 100644
--- a/net/decnet/dn_dev.c
+++ b/net/decnet/dn_dev.c
@@ -46,6 +46,7 @@
 #include <net/neighbour.h>
 #include <net/dst.h>
 #include <net/flow.h>
+#include <net/fib_rules.h>
 #include <net/dn.h>
 #include <net/dn_dev.h>
 #include <net/dn_route.h>
@@ -1418,8 +1419,6 @@ static struct rtnetlink_link dnet_rtnetlink_table[RTM_NR_MSGTYPES] =
 	[RTM_DELROUTE - RTM_BASE] = { .doit	= dn_fib_rtm_delroute,	},
 	[RTM_GETROUTE - RTM_BASE] = { .doit	= dn_cache_getroute,
 				      .dumpit	= dn_fib_dump,		},
-	[RTM_NEWRULE  - RTM_BASE] = { .doit	= dn_fib_rtm_newrule,	},
-	[RTM_DELRULE  - RTM_BASE] = { .doit	= dn_fib_rtm_delrule,	},
 	[RTM_GETRULE  - RTM_BASE] = { .dumpit	= dn_fib_dump_rules,	},
 #else
 	[RTM_GETROUTE - RTM_BASE] = { .doit	= dn_cache_getroute,
diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c
index fa20e2efcfc1..846df3954a63 100644
--- a/net/decnet/dn_fib.c
+++ b/net/decnet/dn_fib.c
@@ -34,6 +34,7 @@
 #include <net/neighbour.h>
 #include <net/dst.h>
 #include <net/flow.h>
+#include <net/fib_rules.h>
 #include <net/dn.h>
 #include <net/dn_route.h>
 #include <net/dn_fib.h>
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 743e9fcf7c5a..5e6f4616ca10 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -80,6 +80,7 @@
 #include <net/neighbour.h>
 #include <net/dst.h>
 #include <net/flow.h>
+#include <net/fib_rules.h>
 #include <net/dn.h>
 #include <net/dn_dev.h>
 #include <net/dn_nsp.h>
@@ -1284,7 +1285,7 @@ static int dn_route_input_slow(struct sk_buff *skb)
 		dev_hold(out_dev);
 
 		if (res.r)
-			src_map = dn_fib_rules_policy(fl.fld_src, &res, &flags);
+			src_map = fl.fld_src; /* no NAT support for now */
 
 		gateway = DN_FIB_RES_GW(res);
 		if (res.type == RTN_NAT) {
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index 6986be754ef2..096f1273e714 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -11,259 +11,198 @@
  *
  *
  * Changes:
+ *              Steve Whitehouse <steve@chygwyn.com>
+ *              Updated for Thomas Graf's generic rules
  *
  */
-#include <linux/string.h>
 #include <linux/net.h>
-#include <linux/socket.h>
-#include <linux/sockios.h>
 #include <linux/init.h>
-#include <linux/skbuff.h>
 #include <linux/netlink.h>
 #include <linux/rtnetlink.h>
-#include <linux/proc_fs.h>
 #include <linux/netdevice.h>
-#include <linux/timer.h>
 #include <linux/spinlock.h>
-#include <linux/in_route.h>
 #include <linux/list.h>
 #include <linux/rcupdate.h>
-#include <asm/atomic.h>
-#include <asm/uaccess.h>
 #include <net/neighbour.h>
 #include <net/dst.h>
 #include <net/flow.h>
+#include <net/fib_rules.h>
 #include <net/dn.h>
 #include <net/dn_fib.h>
 #include <net/dn_neigh.h>
 #include <net/dn_dev.h>
 
+static struct fib_rules_ops dn_fib_rules_ops;
+
 struct dn_fib_rule
 {
-	struct hlist_node	r_hlist;
-	atomic_t		r_clntref;
-	u32			r_preference;
-	unsigned char		r_table;
-	unsigned char		r_action;
-	unsigned char		r_dst_len;
-	unsigned char		r_src_len;
-	__le16			r_src;
-	__le16			r_srcmask;
-	__le16			r_dst;
-	__le16			r_dstmask;
-	__le16			r_srcmap;
-	u8			r_flags;
+	struct fib_rule		common;
+	unsigned char		dst_len;
+	unsigned char		src_len;
+	__le16			src;
+	__le16			srcmask;
+	__le16			dst;
+	__le16			dstmask;
+	__le16			srcmap;
+	u8			flags;
 #ifdef CONFIG_DECNET_ROUTE_FWMARK
-	u32			r_fwmark;
+	u32			fwmark;
 #endif
-	int			r_ifindex;
-	char			r_ifname[IFNAMSIZ];
-	int			r_dead;
-	struct rcu_head		rcu;
 };
 
 static struct dn_fib_rule default_rule = {
-	.r_clntref =		ATOMIC_INIT(2),
-	.r_preference =		0x7fff,
-	.r_table =		RT_TABLE_MAIN,
-	.r_action =		RTN_UNICAST
+	.common = {
+		.refcnt =		ATOMIC_INIT(2),
+		.pref =			0x7fff,
+		.table =		RT_TABLE_MAIN,
+		.action =		FR_ACT_TO_TBL,
+	},
 };
 
-static struct hlist_head dn_fib_rules;
+static LIST_HEAD(dn_fib_rules);
+
 
-int dn_fib_rtm_delrule(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
+int dn_fib_lookup(struct flowi *flp, struct dn_fib_res *res)
 {
-	struct rtattr **rta = arg;
-	struct rtmsg *rtm = NLMSG_DATA(nlh);
-	struct dn_fib_rule *r;
-	struct hlist_node *node;
-	int err = -ESRCH;
-
-	hlist_for_each_entry(r, node, &dn_fib_rules, r_hlist) {
-		if ((!rta[RTA_SRC-1] || memcmp(RTA_DATA(rta[RTA_SRC-1]), &r->r_src, 2) == 0) &&
-			rtm->rtm_src_len == r->r_src_len &&
-			rtm->rtm_dst_len == r->r_dst_len &&
-			(!rta[RTA_DST-1] || memcmp(RTA_DATA(rta[RTA_DST-1]), &r->r_dst, 2) == 0) &&
-#ifdef CONFIG_DECNET_ROUTE_FWMARK
-			(!rta[RTA_PROTOINFO-1] || memcmp(RTA_DATA(rta[RTA_PROTOINFO-1]), &r->r_fwmark, 4) == 0) &&
-#endif
-			(!rtm->rtm_type || rtm->rtm_type == r->r_action) &&
-			(!rta[RTA_PRIORITY-1] || memcmp(RTA_DATA(rta[RTA_PRIORITY-1]), &r->r_preference, 4) == 0) &&
-			(!rta[RTA_IIF-1] || rtattr_strcmp(rta[RTA_IIF-1], r->r_ifname) == 0) &&
-			(!rtm->rtm_table || (r && rtm->rtm_table == r->r_table))) {
-
-			err = -EPERM;
-			if (r == &default_rule)
-				break;
-
-			hlist_del_rcu(&r->r_hlist);
-			r->r_dead = 1;
-			dn_fib_rule_put(r);
-			err = 0;
-			break;
-		}
-	}
+	struct fib_lookup_arg arg = {
+		.result = res,
+	};
+	int err;
+
+	err = fib_rules_lookup(&dn_fib_rules_ops, flp, 0, &arg);
+	res->r = arg.rule;
 
 	return err;
 }
 
-static inline void dn_fib_rule_put_rcu(struct rcu_head *head)
+int dn_fib_rule_action(struct fib_rule *rule, struct flowi *flp, int flags,
+		       struct fib_lookup_arg *arg)
 {
-	struct dn_fib_rule *r = container_of(head, struct dn_fib_rule, rcu);
-	kfree(r);
-}
+	int err = -EAGAIN;
+	struct dn_fib_table *tbl;
 
-void dn_fib_rule_put(struct dn_fib_rule *r)
-{
-	if (atomic_dec_and_test(&r->r_clntref)) {
-		if (r->r_dead)
-			call_rcu(&r->rcu, dn_fib_rule_put_rcu);
-		else
-			printk(KERN_DEBUG "Attempt to free alive dn_fib_rule\n");
+	switch(rule->action) {
+	case FR_ACT_TO_TBL:
+		break;
+
+	case FR_ACT_UNREACHABLE:
+		err = -ENETUNREACH;
+		goto errout;
+
+	case FR_ACT_PROHIBIT:
+		err = -EACCES;
+		goto errout;
+
+	case FR_ACT_BLACKHOLE:
+	default:
+		err = -EINVAL;
+		goto errout;
 	}
+
+	tbl = dn_fib_get_table(rule->table, 0);
+	if (tbl == NULL)
+		goto errout;
+
+	err = tbl->lookup(tbl, flp, (struct dn_fib_res *)arg->result);
+	if (err > 0)
+		err = -EAGAIN;
+errout:
+	return err;
 }
 
+static struct nla_policy dn_fib_rule_policy[FRA_MAX+1] __read_mostly = {
+	[FRA_IFNAME]	= { .type = NLA_STRING },
+	[FRA_PRIORITY]	= { .type = NLA_U32 },
+	[FRA_SRC]	= { .type = NLA_U16 },
+	[FRA_DST]	= { .type = NLA_U16 },
+	[FRA_FWMARK]	= { .type = NLA_U32 },
+};
 
-int dn_fib_rtm_newrule(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
+static int dn_fib_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 {
-	struct rtattr **rta = arg;
-	struct rtmsg *rtm = NLMSG_DATA(nlh);
-	struct dn_fib_rule *r, *new_r, *last = NULL;
-	struct hlist_node *node = NULL;
-	unsigned char table_id;
-
-	if (rtm->rtm_src_len > 16 || rtm->rtm_dst_len > 16)
-		return -EINVAL;
-
-	if (rta[RTA_IIF-1] && RTA_PAYLOAD(rta[RTA_IIF-1]) > IFNAMSIZ)
-		return -EINVAL;
-
-	if (rtm->rtm_type == RTN_NAT)
-		return -EINVAL;
-
-	table_id = rtm->rtm_table;
-	if (table_id == RT_TABLE_UNSPEC) {
-		struct dn_fib_table *tb;
-		if (rtm->rtm_type == RTN_UNICAST) {
-			if ((tb = dn_fib_empty_table()) == NULL)
-				return -ENOBUFS;
-			table_id = tb->n;
-		}
-	}
+	struct dn_fib_rule *r = (struct dn_fib_rule *)rule;
+	u16 daddr = fl->fld_dst;
+	u16 saddr = fl->fld_src;
+
+	if (((saddr ^ r->src) & r->srcmask) ||
+	    ((daddr ^ r->dst) & r->dstmask))
+		return 0;
 
-	new_r = kzalloc(sizeof(*new_r), GFP_KERNEL);
-	if (!new_r)
-		return -ENOMEM;
-
-	if (rta[RTA_SRC-1])
-		memcpy(&new_r->r_src, RTA_DATA(rta[RTA_SRC-1]), 2);
-	if (rta[RTA_DST-1])
-		memcpy(&new_r->r_dst, RTA_DATA(rta[RTA_DST-1]), 2);
-	if (rta[RTA_GATEWAY-1])
-		memcpy(&new_r->r_srcmap, RTA_DATA(rta[RTA_GATEWAY-1]), 2);
-	new_r->r_src_len = rtm->rtm_src_len;
-	new_r->r_dst_len = rtm->rtm_dst_len;
-	new_r->r_srcmask = dnet_make_mask(rtm->rtm_src_len);
-	new_r->r_dstmask = dnet_make_mask(rtm->rtm_dst_len);
 #ifdef CONFIG_DECNET_ROUTE_FWMARK
-	if (rta[RTA_PROTOINFO-1])
-		memcpy(&new_r->r_fwmark, RTA_DATA(rta[RTA_PROTOINFO-1]), 4);
+	if (r->fwmark && (r->fwmark != fl->fld_fwmark))
+		return 0;
 #endif
-	new_r->r_action = rtm->rtm_type;
-	new_r->r_flags = rtm->rtm_flags;
-	if (rta[RTA_PRIORITY-1])
-		memcpy(&new_r->r_preference, RTA_DATA(rta[RTA_PRIORITY-1]), 4);
-	new_r->r_table = table_id;
-	if (rta[RTA_IIF-1]) {
-		struct net_device *dev;
-		rtattr_strlcpy(new_r->r_ifname, rta[RTA_IIF-1], IFNAMSIZ);
-		new_r->r_ifindex = -1;
-		dev = dev_get_by_name(new_r->r_ifname);
-		if (dev) {
-			new_r->r_ifindex = dev->ifindex;
-			dev_put(dev);
-		}
-	}
 
-	r = container_of(dn_fib_rules.first, struct dn_fib_rule, r_hlist);
-	if (!new_r->r_preference) {
-		if (r && r->r_hlist.next != NULL) {
-			r = container_of(r->r_hlist.next, struct dn_fib_rule, r_hlist);
-			if (r->r_preference)
-				new_r->r_preference = r->r_preference - 1;
+	return 1;
+}
+
+static int dn_fib_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
+				 struct nlmsghdr *nlh, struct fib_rule_hdr *frh,
+				 struct nlattr **tb)
+{
+	int err = -EINVAL;
+	struct dn_fib_rule *r = (struct dn_fib_rule *)rule;
+
+	if (frh->src_len > 16 || frh->dst_len > 16 || frh->tos)
+		goto  errout;
+
+	if (rule->table == RT_TABLE_UNSPEC) {
+		if (rule->action == FR_ACT_TO_TBL) {
+			struct dn_fib_table *table;
+
+			table = dn_fib_empty_table();
+			if (table == NULL) {
+				err = -ENOBUFS;
+				goto errout;
+			}
+
+			rule->table = table->n;
 		}
 	}
 
-	hlist_for_each_entry(r, node, &dn_fib_rules, r_hlist) {
-		if (r->r_preference > new_r->r_preference)
-			break;
-		last = r;
-	}
-	atomic_inc(&new_r->r_clntref);
+	if (tb[FRA_SRC])
+		r->src = nla_get_u16(tb[FRA_SRC]);
 
-	if (last)
-		hlist_add_after_rcu(&last->r_hlist, &new_r->r_hlist);
-	else
-		hlist_add_before_rcu(&new_r->r_hlist, &r->r_hlist);
-	return 0;
-}
+	if (tb[FRA_DST])
+		r->dst = nla_get_u16(tb[FRA_DST]);
 
+#ifdef CONFIG_DECNET_ROUTE_FWMARK
+	if (tb[FRA_FWMARK])
+		r->fwmark = nla_get_u32(tb[FRA_FWMARK]);
+#endif
+
+	r->src_len = frh->src_len;
+	r->srcmask = dnet_make_mask(r->src_len);
+	r->dst_len = frh->dst_len;
+	r->dstmask = dnet_make_mask(r->dst_len);
+	err = 0;
+errout:
+	return err;
+}
 
-int dn_fib_lookup(const struct flowi *flp, struct dn_fib_res *res)
+static int dn_fib_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
+			       struct nlattr **tb)
 {
-	struct dn_fib_rule *r, *policy;
-	struct dn_fib_table *tb;
-	__le16 saddr = flp->fld_src;
-	__le16 daddr = flp->fld_dst;
-	struct hlist_node *node;
-	int err;
+	struct dn_fib_rule *r = (struct dn_fib_rule *)rule;
+
+	if (frh->src_len && (r->src_len != frh->src_len))
+		return 0;
 
-	rcu_read_lock();
+	if (frh->dst_len && (r->dst_len != frh->dst_len))
+		return 0;
 
-	hlist_for_each_entry_rcu(r, node, &dn_fib_rules, r_hlist) {
-		if (((saddr^r->r_src) & r->r_srcmask) ||
-		    ((daddr^r->r_dst) & r->r_dstmask) ||
 #ifdef CONFIG_DECNET_ROUTE_FWMARK
-		    (r->r_fwmark && r->r_fwmark != flp->fld_fwmark) ||
+	if (tb[FRA_FWMARK] && (r->fwmark != nla_get_u32(tb[FRA_FWMARK])))
+		return 0;
 #endif
-		    (r->r_ifindex && r->r_ifindex != flp->iif))
-			continue;
-
-		switch(r->r_action) {
-			case RTN_UNICAST:
-			case RTN_NAT:
-				policy = r;
-				break;
-			case RTN_UNREACHABLE:
-				rcu_read_unlock();
-				return -ENETUNREACH;
-			default:
-			case RTN_BLACKHOLE:
-				rcu_read_unlock();
-				return -EINVAL;
-			case RTN_PROHIBIT:
-				rcu_read_unlock();
-				return -EACCES;
-		}
 
-		if ((tb = dn_fib_get_table(r->r_table, 0)) == NULL)
-			continue;
-		err = tb->lookup(tb, flp, res);
-		if (err == 0) {
-			res->r = policy;
-			if (policy)
-				atomic_inc(&policy->r_clntref);
-			rcu_read_unlock();
-			return 0;
-		}
-		if (err < 0 && err != -EAGAIN) {
-			rcu_read_unlock();
-			return err;
-		}
-	}
+	if (tb[FRA_SRC] && (r->src != nla_get_u32(tb[FRA_SRC])))
+		return 0;
+
+	if (tb[FRA_DST] && (r->dst != nla_get_u32(tb[FRA_DST])))
+		return 0;
 
-	rcu_read_unlock();
-	return -ESRCH;
+	return 1;
 }
 
 unsigned dnet_addr_type(__le16 addr)
@@ -284,142 +223,77 @@ unsigned dnet_addr_type(__le16 addr)
 	return ret;
 }
 
-__le16 dn_fib_rules_policy(__le16 saddr, struct dn_fib_res *res, unsigned *flags)
+static int dn_fib_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
+			    struct nlmsghdr *nlh, struct fib_rule_hdr *frh)
 {
-	struct dn_fib_rule *r = res->r;
-
-	if (r->r_action == RTN_NAT) {
-		int addrtype = dnet_addr_type(r->r_srcmap);
-
-		if (addrtype == RTN_NAT) {
-			saddr = (saddr&~r->r_srcmask)|r->r_srcmap;
-			*flags |= RTCF_SNAT;
-		} else if (addrtype == RTN_LOCAL || r->r_srcmap == 0) {
-			saddr = r->r_srcmap;
-			*flags |= RTCF_MASQ;
-		}
-	}
-	return saddr;
-}
+	struct dn_fib_rule *r = (struct dn_fib_rule *)rule;
 
-static void dn_fib_rules_detach(struct net_device *dev)
-{
-	struct hlist_node *node;
-	struct dn_fib_rule *r;
+	frh->family = AF_DECnet;
+	frh->dst_len = r->dst_len;
+	frh->src_len = r->src_len;
+	frh->tos = 0;
 
-	hlist_for_each_entry(r, node, &dn_fib_rules, r_hlist) {
-		if (r->r_ifindex == dev->ifindex)
-			r->r_ifindex = -1;
-	}
-}
+#ifdef CONFIG_DECNET_ROUTE_FWMARK
+	if (r->fwmark)
+		NLA_PUT_U32(skb, FRA_FWMARK, r->fwmark);
+#endif
+	if (r->dst_len)
+		NLA_PUT_U16(skb, FRA_DST, r->dst);
+	if (r->src_len)
+		NLA_PUT_U16(skb, FRA_SRC, r->src);
 
-static void dn_fib_rules_attach(struct net_device *dev)
-{
-	struct hlist_node *node;
-	struct dn_fib_rule *r;
+	return 0;
 
-	hlist_for_each_entry(r, node, &dn_fib_rules, r_hlist) {
-		if (r->r_ifindex == -1 && strcmp(dev->name, r->r_ifname) == 0)
-			r->r_ifindex = dev->ifindex;
-	}
+nla_put_failure:
+	return -ENOBUFS;
 }
 
-static int dn_fib_rules_event(struct notifier_block *this, unsigned long event, void *ptr)
+static u32 dn_fib_rule_default_pref(void)
 {
-	struct net_device *dev = ptr;
-
-	switch(event) {
-		case NETDEV_UNREGISTER:
-			dn_fib_rules_detach(dev);
-			dn_fib_sync_down(0, dev, 1);
-		case NETDEV_REGISTER:
-			dn_fib_rules_attach(dev);
-			dn_fib_sync_up(dev);
+	struct list_head *pos;
+	struct fib_rule *rule;
+
+	if (!list_empty(&dn_fib_rules)) {
+		pos = dn_fib_rules.next;
+		if (pos->next != &dn_fib_rules) {
+			rule = list_entry(pos->next, struct fib_rule, list);
+			if (rule->pref)
+				return rule->pref - 1;
+		}
 	}
 
-	return NOTIFY_DONE;
-}
-
-
-static struct notifier_block dn_fib_rules_notifier = {
-	.notifier_call =	dn_fib_rules_event,
-};
-
-static int dn_fib_fill_rule(struct sk_buff *skb, struct dn_fib_rule *r,
-			    struct netlink_callback *cb, unsigned int flags)
-{
-	struct rtmsg *rtm;
-	struct nlmsghdr *nlh;
-	unsigned char *b = skb->tail;
-
-
-	nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWRULE, sizeof(*rtm), flags);
-	rtm = NLMSG_DATA(nlh);
-	rtm->rtm_family = AF_DECnet;
-	rtm->rtm_dst_len = r->r_dst_len;
-	rtm->rtm_src_len = r->r_src_len;
-	rtm->rtm_tos = 0;
-#ifdef CONFIG_DECNET_ROUTE_FWMARK
-	if (r->r_fwmark)
-		RTA_PUT(skb, RTA_PROTOINFO, 4, &r->r_fwmark);
-#endif
-	rtm->rtm_table = r->r_table;
-	rtm->rtm_protocol = 0;
-	rtm->rtm_scope = 0;
-	rtm->rtm_type = r->r_action;
-	rtm->rtm_flags = r->r_flags;
-
-	if (r->r_dst_len)
-		RTA_PUT(skb, RTA_DST, 2, &r->r_dst);
-	if (r->r_src_len)
-		RTA_PUT(skb, RTA_SRC, 2, &r->r_src);
-	if (r->r_ifname[0])
-		RTA_PUT(skb, RTA_IIF, IFNAMSIZ, &r->r_ifname);
-	if (r->r_preference)
-		RTA_PUT(skb, RTA_PRIORITY, 4, &r->r_preference);
-	if (r->r_srcmap)
-		RTA_PUT(skb, RTA_GATEWAY, 2, &r->r_srcmap);
-	nlh->nlmsg_len = skb->tail - b;
-	return skb->len;
-
-nlmsg_failure:
-rtattr_failure:
-	skb_trim(skb, b - skb->data);
-	return -1;
+	return 0;
 }
 
 int dn_fib_dump_rules(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	int idx = 0;
-	int s_idx = cb->args[0];
-	struct dn_fib_rule *r;
-	struct hlist_node *node;
-
-	rcu_read_lock();
-	hlist_for_each_entry(r, node, &dn_fib_rules, r_hlist) {
-		if (idx < s_idx)
-			goto next;
-		if (dn_fib_fill_rule(skb, r, cb, NLM_F_MULTI) < 0)
-			break;
-next:
-		idx++;
-	}
-	rcu_read_unlock();
-	cb->args[0] = idx;
-
-	return skb->len;
+	return fib_rules_dump(skb, cb, AF_DECnet);
 }
 
+static struct fib_rules_ops dn_fib_rules_ops = {
+	.family		= AF_DECnet,
+	.rule_size	= sizeof(struct dn_fib_rule),
+	.action		= dn_fib_rule_action,
+	.match		= dn_fib_rule_match,
+	.configure	= dn_fib_rule_configure,
+	.compare	= dn_fib_rule_compare,
+	.fill		= dn_fib_rule_fill,
+	.default_pref	= dn_fib_rule_default_pref,
+	.nlgroup	= RTNLGRP_DECnet_RULE,
+	.policy		= dn_fib_rule_policy,
+	.rules_list	= &dn_fib_rules,
+	.owner		= THIS_MODULE,
+};
+
 void __init dn_fib_rules_init(void)
 {
-	INIT_HLIST_HEAD(&dn_fib_rules);
-	hlist_add_head(&default_rule.r_hlist, &dn_fib_rules);
-	register_netdevice_notifier(&dn_fib_rules_notifier);
+	list_add_tail(&default_rule.common.list, &dn_fib_rules);
+	fib_rules_register(&dn_fib_rules_ops);
 }
 
 void __exit dn_fib_rules_cleanup(void)
 {
-	unregister_netdevice_notifier(&dn_fib_rules_notifier);
+	fib_rules_unregister(&dn_fib_rules_ops);
 }
 
 
diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c
index e926c952e363..2e01b67398c8 100644
--- a/net/decnet/dn_table.c
+++ b/net/decnet/dn_table.c
@@ -30,6 +30,7 @@
 #include <net/neighbour.h>
 #include <net/dst.h>
 #include <net/flow.h>
+#include <net/fib_rules.h>
 #include <net/dn.h>
 #include <net/dn_route.h>
 #include <net/dn_fib.h>
-- 
cgit v1.2.3


From 757dbb494be3309fe41ce4c62f8057d8b41d8897 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 9 Aug 2006 20:50:00 -0700
Subject: [NET]: drop unused elements from net_proto_family

Three values in net_proto_family are defined but never used.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index 19da2c08d7b6..1bd76327ee2b 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -169,11 +169,6 @@ struct proto_ops {
 struct net_proto_family {
 	int		family;
 	int		(*create)(struct socket *sock, int protocol);
-	/* These are counters for the number of different methods of
-	   each we support */
-	short		authentication;
-	short		encryption;
-	short		encrypt_net;
 	struct module	*owner;
 };
 
-- 
cgit v1.2.3


From f0fd27d42e39b91f85e1840ec49b072fd6c545b8 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 9 Aug 2006 21:03:17 -0700
Subject: [NET]: sock_register interface changes

The sock_register() doesn't change the family, so the protocols can
define it read-only.  No caller ever checks return value from
sock_unregister()

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h |  4 ++--
 net/socket.c        | 10 ++++------
 2 files changed, 6 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/net.h b/include/linux/net.h
index 1bd76327ee2b..c257f716e00f 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -176,8 +176,8 @@ struct iovec;
 struct kvec;
 
 extern int	     sock_wake_async(struct socket *sk, int how, int band);
-extern int	     sock_register(struct net_proto_family *fam);
-extern int	     sock_unregister(int family);
+extern int	     sock_register(const struct net_proto_family *fam);
+extern void	     sock_unregister(int family);
 extern int	     sock_create(int family, int type, int proto,
 				 struct socket **res);
 extern int	     sock_create_kern(int family, int type, int proto,
diff --git a/net/socket.c b/net/socket.c
index b5a3fcb9ed6d..4147fe4bf41d 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -147,7 +147,7 @@ static struct file_operations socket_file_ops = {
  */
 
 static DEFINE_SPINLOCK(net_family_lock);
-static const struct net_proto_family *net_families[NPROTO];
+static const struct net_proto_family *net_families[NPROTO] __read_mostly;
 
 /*
  *	Statistics counters of the socket lists
@@ -2080,7 +2080,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
  *	socket interface. The value ops->family coresponds to the
  *	socket system call protocol family.
  */
-int sock_register(struct net_proto_family *ops)
+int sock_register(const struct net_proto_family *ops)
 {
 	int err;
 
@@ -2116,10 +2116,9 @@ int sock_register(struct net_proto_family *ops)
  *	a module then it needs to provide its own protection in
  *	the ops->create routine.
  */
-int sock_unregister(int family)
+void sock_unregister(int family)
 {
-	if (family < 0 || family >= NPROTO)
-		return -EINVAL;
+	BUG_ON(family < 0 || family >= NPROTO);
 
 	spin_lock(&net_family_lock);
 	net_families[family] = NULL;
@@ -2128,7 +2127,6 @@ int sock_unregister(int family)
 	synchronize_rcu();
 
 	printk(KERN_INFO "NET: Unregistered protocol family %d\n", family);
-	return 0;
 }
 
 static int __init sock_init(void)
-- 
cgit v1.2.3


From d924424aaed116b362c6d0e667d912b77e655085 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 10 Aug 2006 23:03:23 -0700
Subject: [NEIGHBOUR]: Use ALIGN() macro.

Rather than opencoding the mask, it looks better to use ALIGN()
macro from kernel.h.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/neighbour.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index 74c4b6ff8a5c..bd187daffdb9 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -101,7 +101,7 @@ struct neighbour
 	__u8			dead;
 	atomic_t		probes;
 	rwlock_t		lock;
-	unsigned char		ha[(MAX_ADDR_LEN+sizeof(unsigned long)-1)&~(sizeof(unsigned long)-1)];
+	unsigned char		ha[ALIGN(MAX_ADDR_LEN, sizeof(unsigned long))];
 	struct hh_cache		*hh;
 	atomic_t		refcnt;
 	int			(*output)(struct sk_buff *skb);
-- 
cgit v1.2.3


From 2dfe55b47e3d66ded5a84caf71e0da5710edf48b Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Aug 2006 23:08:33 -0700
Subject: [NET]: Use u32 for routing table IDs

Use u32 for routing table IDs in net/ipv4 and net/decnet in preparation of
support for a larger number of routing tables. net/ipv6 already uses u32
everywhere and needs no further changes. No functional changes are made by
this patch.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dn_fib.h     |  4 ++--
 include/net/ip_fib.h     | 14 +++++++-------
 net/decnet/dn_fib.c      |  6 +++---
 net/decnet/dn_table.c    | 10 +++++-----
 net/ipv4/fib_frontend.c  |  8 ++++----
 net/ipv4/fib_hash.c      |  4 ++--
 net/ipv4/fib_lookup.h    |  4 ++--
 net/ipv4/fib_rules.c     |  2 +-
 net/ipv4/fib_semantics.c |  4 ++--
 net/ipv4/fib_trie.c      |  6 +++---
 10 files changed, 31 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/dn_fib.h b/include/net/dn_fib.h
index 32bc8ce5c5ce..cd9c3782f838 100644
--- a/include/net/dn_fib.h
+++ b/include/net/dn_fib.h
@@ -94,7 +94,7 @@ struct dn_fib_node {
 
 
 struct dn_fib_table {
-	int n;
+	u32 n;
 
 	int (*insert)(struct dn_fib_table *t, struct rtmsg *r, 
 			struct dn_kern_rta *rta, struct nlmsghdr *n, 
@@ -137,7 +137,7 @@ extern int dn_fib_sync_up(struct net_device *dev);
 /*
  * dn_tables.c
  */
-extern struct dn_fib_table *dn_fib_get_table(int n, int creat);
+extern struct dn_fib_table *dn_fib_get_table(u32 n, int creat);
 extern struct dn_fib_table *dn_fib_empty_table(void);
 extern void dn_fib_table_init(void);
 extern void dn_fib_table_cleanup(void);
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index adf73586bc05..0dcbf166eb94 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -150,7 +150,7 @@ struct fib_result_nl {
 #endif /* CONFIG_IP_ROUTE_MULTIPATH_WRANDOM */
 
 struct fib_table {
-	unsigned char	tb_id;
+	u32		tb_id;
 	unsigned	tb_stamp;
 	int		(*tb_lookup)(struct fib_table *tb, const struct flowi *flp, struct fib_result *res);
 	int		(*tb_insert)(struct fib_table *table, struct rtmsg *r,
@@ -173,14 +173,14 @@ struct fib_table {
 extern struct fib_table *ip_fib_local_table;
 extern struct fib_table *ip_fib_main_table;
 
-static inline struct fib_table *fib_get_table(int id)
+static inline struct fib_table *fib_get_table(u32 id)
 {
 	if (id != RT_TABLE_LOCAL)
 		return ip_fib_main_table;
 	return ip_fib_local_table;
 }
 
-static inline struct fib_table *fib_new_table(int id)
+static inline struct fib_table *fib_new_table(u32 id)
 {
 	return fib_get_table(id);
 }
@@ -205,9 +205,9 @@ static inline void fib_select_default(const struct flowi *flp, struct fib_result
 
 extern struct fib_table * fib_tables[RT_TABLE_MAX+1];
 extern int fib_lookup(struct flowi *flp, struct fib_result *res);
-extern struct fib_table *__fib_new_table(int id);
+extern struct fib_table *__fib_new_table(u32 id);
 
-static inline struct fib_table *fib_get_table(int id)
+static inline struct fib_table *fib_get_table(u32 id)
 {
 	if (id == 0)
 		id = RT_TABLE_MAIN;
@@ -215,7 +215,7 @@ static inline struct fib_table *fib_get_table(int id)
 	return fib_tables[id];
 }
 
-static inline struct fib_table *fib_new_table(int id)
+static inline struct fib_table *fib_new_table(u32 id)
 {
 	if (id == 0)
 		id = RT_TABLE_MAIN;
@@ -248,7 +248,7 @@ extern int fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm,
 extern u32  __fib_res_prefsrc(struct fib_result *res);
 
 /* Exported by fib_hash.c */
-extern struct fib_table *fib_hash_init(int id);
+extern struct fib_table *fib_hash_init(u32 id);
 
 #ifdef CONFIG_IP_MULTIPLE_TABLES
 extern int fib4_rules_dump(struct sk_buff *skb, struct netlink_callback *cb);
diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c
index ed5fb5c3eab5..7b3bf5c3d720 100644
--- a/net/decnet/dn_fib.c
+++ b/net/decnet/dn_fib.c
@@ -534,8 +534,8 @@ int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 
 int dn_fib_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	int t;
-	int s_t;
+	u32 t;
+	u32 s_t;
 	struct dn_fib_table *tb;
 
 	if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) &&
@@ -765,7 +765,7 @@ void dn_fib_flush(void)
 {
         int flushed = 0;
         struct dn_fib_table *tb;
-        int id;
+        u32 id;
 
         for(id = RT_TABLE_MAX; id > 0; id--) {
                 if ((tb = dn_fib_get_table(id, 0)) == NULL)
diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c
index 2e01b67398c8..1601ee5406a8 100644
--- a/net/decnet/dn_table.c
+++ b/net/decnet/dn_table.c
@@ -264,7 +264,7 @@ static int dn_fib_nh_match(struct rtmsg *r, struct nlmsghdr *nlh, struct dn_kern
 }
 
 static int dn_fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-                        u8 tb_id, u8 type, u8 scope, void *dst, int dst_len,
+                        u32 tb_id, u8 type, u8 scope, void *dst, int dst_len,
                         struct dn_fib_info *fi, unsigned int flags)
 {
         struct rtmsg *rtm;
@@ -327,7 +327,7 @@ rtattr_failure:
 }
 
 
-static void dn_rtmsg_fib(int event, struct dn_fib_node *f, int z, int tb_id,
+static void dn_rtmsg_fib(int event, struct dn_fib_node *f, int z, u32 tb_id,
                         struct nlmsghdr *nlh, struct netlink_skb_parms *req)
 {
         struct sk_buff *skb;
@@ -740,7 +740,7 @@ out:
 }
 
 
-struct dn_fib_table *dn_fib_get_table(int n, int create)
+struct dn_fib_table *dn_fib_get_table(u32 n, int create)
 {
         struct dn_fib_table *t;
 
@@ -777,7 +777,7 @@ struct dn_fib_table *dn_fib_get_table(int n, int create)
         return t;
 }
 
-static void dn_fib_del_tree(int n)
+static void dn_fib_del_tree(u32 n)
 {
 	struct dn_fib_table *t;
 
@@ -791,7 +791,7 @@ static void dn_fib_del_tree(int n)
 
 struct dn_fib_table *dn_fib_empty_table(void)
 {
-        int id;
+        u32 id;
 
         for(id = RT_TABLE_MIN; id <= RT_TABLE_MAX; id++)
                 if (dn_fib_tables[id] == NULL)
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index a83f1aa8034e..06f4b23f6f57 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -62,7 +62,7 @@ struct fib_table *ip_fib_main_table;
 
 struct fib_table *fib_tables[RT_TABLE_MAX+1];
 
-struct fib_table *__fib_new_table(int id)
+struct fib_table *__fib_new_table(u32 id)
 {
 	struct fib_table *tb;
 
@@ -82,7 +82,7 @@ static void fib_flush(void)
 	int flushed = 0;
 #ifdef CONFIG_IP_MULTIPLE_TABLES
 	struct fib_table *tb;
-	int id;
+	u32 id;
 
 	for (id = RT_TABLE_MAX; id>0; id--) {
 		if ((tb = fib_get_table(id))==NULL)
@@ -333,8 +333,8 @@ int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	int t;
-	int s_t;
+	u32 t;
+	u32 s_t;
 	struct fib_table *tb;
 
 	if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) &&
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 72c633b357cf..f8d5c8024ccb 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -765,9 +765,9 @@ static int fn_hash_dump(struct fib_table *tb, struct sk_buff *skb, struct netlin
 }
 
 #ifdef CONFIG_IP_MULTIPLE_TABLES
-struct fib_table * fib_hash_init(int id)
+struct fib_table * fib_hash_init(u32 id)
 #else
-struct fib_table * __init fib_hash_init(int id)
+struct fib_table * __init fib_hash_init(u32 id)
 #endif
 {
 	struct fib_table *tb;
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index ef6609ea0eb7..ddd52496b451 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -30,11 +30,11 @@ extern struct fib_info *fib_create_info(const struct rtmsg *r,
 extern int fib_nh_match(struct rtmsg *r, struct nlmsghdr *,
 			struct kern_rta *rta, struct fib_info *fi);
 extern int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-			 u8 tb_id, u8 type, u8 scope, void *dst,
+			 u32 tb_id, u8 type, u8 scope, void *dst,
 			 int dst_len, u8 tos, struct fib_info *fi,
 			 unsigned int);
 extern void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
-		      int z, int tb_id,
+		      int z, u32 tb_id,
 		      struct nlmsghdr *n, struct netlink_skb_parms *req);
 extern struct fib_alias *fib_find_alias(struct list_head *fah,
 					u8 tos, u32 prio);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index d242e5291fcc..58fb91b00fdf 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -169,7 +169,7 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 
 static struct fib_table *fib_empty_table(void)
 {
-	int id;
+	u32 id;
 
 	for (id = 1; id <= RT_TABLE_MAX; id++)
 		if (fib_tables[id] == NULL)
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 38bca473c7e2..c7a112b5a185 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -273,7 +273,7 @@ int ip_fib_check_default(u32 gw, struct net_device *dev)
 }
 
 void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
-	       int z, int tb_id,
+	       int z, u32 tb_id,
 	       struct nlmsghdr *n, struct netlink_skb_parms *req)
 {
 	struct sk_buff *skb;
@@ -939,7 +939,7 @@ u32 __fib_res_prefsrc(struct fib_result *res)
 
 int
 fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-	      u8 tb_id, u8 type, u8 scope, void *dst, int dst_len, u8 tos,
+	      u32 tb_id, u8 type, u8 scope, void *dst, int dst_len, u8 tos,
 	      struct fib_info *fi, unsigned int flags)
 {
 	struct rtmsg *rtm;
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 01801c0f885d..4a27b2d573a3 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1148,7 +1148,7 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 
 	key = ntohl(key);
 
-	pr_debug("Insert table=%d %08x/%d\n", tb->tb_id, key, plen);
+	pr_debug("Insert table=%u %08x/%d\n", tb->tb_id, key, plen);
 
 	mask = ntohl(inet_make_mask(plen));
 
@@ -1943,9 +1943,9 @@ out:
 /* Fix more generic FIB names for init later */
 
 #ifdef CONFIG_IP_MULTIPLE_TABLES
-struct fib_table * fib_hash_init(int id)
+struct fib_table * fib_hash_init(u32 id)
 #else
-struct fib_table * __init fib_hash_init(int id)
+struct fib_table * __init fib_hash_init(u32 id)
 #endif
 {
 	struct fib_table *tb;
-- 
cgit v1.2.3


From 9e762a4a89b302cb3b26a1f9bb33eff459eaeca9 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Aug 2006 23:09:48 -0700
Subject: [NET]: Introduce RTA_TABLE/FRA_TABLE attributes

Introduce RTA_TABLE route attribute and FRA_TABLE routing rule attribute
to hold 32 bit routing table IDs. Usespace compatibility is provided by
continuing to accept and send the rtm_table field, but because of its
limited size it can only carry the low 8 bits of the table ID. This
implies that if larger IDs are used, _all_ userspace programs using them
need to use RTA_TABLE.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/fib_rules.h |  4 ++++
 include/linux/rtnetlink.h |  8 ++++++++
 include/net/fib_rules.h   |  7 +++++++
 net/core/fib_rules.c      |  5 +++--
 net/decnet/dn_fib.c       |  7 ++++---
 net/decnet/dn_route.c     |  1 +
 net/decnet/dn_table.c     |  1 +
 net/ipv4/fib_frontend.c   |  7 ++++---
 net/ipv4/fib_rules.c      |  1 +
 net/ipv4/fib_semantics.c  |  1 +
 net/ipv4/route.c          |  1 +
 net/ipv6/fib6_rules.c     |  1 +
 net/ipv6/route.c          | 13 +++++++++----
 13 files changed, 45 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h
index 5e503f0ca6e4..19a82b6c1c1f 100644
--- a/include/linux/fib_rules.h
+++ b/include/linux/fib_rules.h
@@ -36,6 +36,10 @@ enum
 	FRA_UNUSED5,
 	FRA_FWMARK,	/* netfilter mark (IPv4) */
 	FRA_FLOW,	/* flow/class id */
+	FRA_UNUSED6,
+	FRA_UNUSED7,
+	FRA_UNUSED8,
+	FRA_TABLE,	/* Extended table id */
 	__FRA_MAX
 };
 
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 0aaffa2ae666..ea422a539a03 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -264,6 +264,7 @@ enum rtattr_type_t
 	RTA_CACHEINFO,
 	RTA_SESSION,
 	RTA_MP_ALGO,
+	RTA_TABLE,
 	__RTA_MAX
 };
 
@@ -717,6 +718,13 @@ extern void __rtnl_unlock(void);
 	} \
 } while(0)
 
+static inline u32 rtm_get_table(struct rtattr **rta, u8 table)
+{
+	return RTA_GET_U32(rta[RTA_TABLE-1]);
+rtattr_failure:
+	return table;
+}
+
 #endif /* __KERNEL__ */
 
 
diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 61375d9e53f8..8e2f473d3e82 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -74,6 +74,13 @@ static inline void fib_rule_put(struct fib_rule *rule)
 		call_rcu(&rule->rcu, fib_rule_put_rcu);
 }
 
+static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla)
+{
+	if (nla[FRA_TABLE])
+		return nla_get_u32(nla[FRA_TABLE]);
+	return frh->table;
+}
+
 extern int			fib_rules_register(struct fib_rules_ops *);
 extern int			fib_rules_unregister(struct fib_rules_ops *);
 
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 6cdad24038e2..873b04d5df81 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -187,7 +187,7 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	rule->action = frh->action;
 	rule->flags = frh->flags;
-	rule->table = frh->table;
+	rule->table = frh_get_table(frh, tb);
 
 	if (!rule->pref && ops->default_pref)
 		rule->pref = ops->default_pref();
@@ -245,7 +245,7 @@ int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		if (frh->action && (frh->action != rule->action))
 			continue;
 
-		if (frh->table && (frh->table != rule->table))
+		if (frh->table && (frh_get_table(frh, tb) != rule->table))
 			continue;
 
 		if (tb[FRA_PRIORITY] &&
@@ -291,6 +291,7 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule,
 
 	frh = nlmsg_data(nlh);
 	frh->table = rule->table;
+	NLA_PUT_U32(skb, FRA_TABLE, rule->table);
 	frh->res1 = 0;
 	frh->res2 = 0;
 	frh->action = rule->action;
diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c
index 7b3bf5c3d720..fb596373daa8 100644
--- a/net/decnet/dn_fib.c
+++ b/net/decnet/dn_fib.c
@@ -491,7 +491,8 @@ static int dn_fib_check_attr(struct rtmsg *r, struct rtattr **rta)
 		if (attr) {
 			if (RTA_PAYLOAD(attr) < 4 && RTA_PAYLOAD(attr) != 2)
 				return -EINVAL;
-			if (i != RTA_MULTIPATH && i != RTA_METRICS)
+			if (i != RTA_MULTIPATH && i != RTA_METRICS &&
+			    i != RTA_TABLE)
 				rta[i-1] = (struct rtattr *)RTA_DATA(attr);
 		}
 	}
@@ -508,7 +509,7 @@ int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	if (dn_fib_check_attr(r, rta))
 		return -EINVAL;
 
-	tb = dn_fib_get_table(r->rtm_table, 0);
+	tb = dn_fib_get_table(rtm_get_table(rta, r->rtm_table), 0);
 	if (tb)
 		return tb->delete(tb, r, (struct dn_kern_rta *)rta, nlh, &NETLINK_CB(skb));
 
@@ -524,7 +525,7 @@ int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	if (dn_fib_check_attr(r, rta))
 		return -EINVAL;
 
-	tb = dn_fib_get_table(r->rtm_table, 1);
+	tb = dn_fib_get_table(rtm_get_table(rta, r->rtm_table), 1);
 	if (tb) 
 		return tb->insert(tb, r, (struct dn_kern_rta *)rta, nlh, &NETLINK_CB(skb));
 
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 5e6f4616ca10..4c963213fba5 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -1486,6 +1486,7 @@ static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq,
 	r->rtm_src_len = 0;
 	r->rtm_tos = 0;
 	r->rtm_table = RT_TABLE_MAIN;
+	RTA_PUT_U32(skb, RTA_TABLE, RT_TABLE_MAIN);
 	r->rtm_type = rt->rt_type;
 	r->rtm_flags = (rt->rt_flags & ~0xFFFF) | RTM_F_CLONED;
 	r->rtm_scope = RT_SCOPE_UNIVERSE;
diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c
index 1601ee5406a8..eca7c1e10c80 100644
--- a/net/decnet/dn_table.c
+++ b/net/decnet/dn_table.c
@@ -278,6 +278,7 @@ static int dn_fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
         rtm->rtm_src_len = 0;
         rtm->rtm_tos = 0;
         rtm->rtm_table = tb_id;
+	RTA_PUT_U32(skb, RTA_TABLE, tb_id);
         rtm->rtm_flags = fi->fib_flags;
         rtm->rtm_scope = scope;
 	rtm->rtm_type  = type;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 06f4b23f6f57..2696ede52de2 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -294,7 +294,8 @@ static int inet_check_attr(struct rtmsg *r, struct rtattr **rta)
 		if (attr) {
 			if (RTA_PAYLOAD(attr) < 4)
 				return -EINVAL;
-			if (i != RTA_MULTIPATH && i != RTA_METRICS)
+			if (i != RTA_MULTIPATH && i != RTA_METRICS &&
+			    i != RTA_TABLE)
 				*rta = (struct rtattr*)RTA_DATA(attr);
 		}
 	}
@@ -310,7 +311,7 @@ int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	if (inet_check_attr(r, rta))
 		return -EINVAL;
 
-	tb = fib_get_table(r->rtm_table);
+	tb = fib_get_table(rtm_get_table(rta, r->rtm_table));
 	if (tb)
 		return tb->tb_delete(tb, r, (struct kern_rta*)rta, nlh, &NETLINK_CB(skb));
 	return -ESRCH;
@@ -325,7 +326,7 @@ int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 	if (inet_check_attr(r, rta))
 		return -EINVAL;
 
-	tb = fib_new_table(r->rtm_table);
+	tb = fib_new_table(rtm_get_table(rta, r->rtm_table));
 	if (tb)
 		return tb->tb_insert(tb, r, (struct kern_rta*)rta, nlh, &NETLINK_CB(skb));
 	return -ENOBUFS;
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 58fb91b00fdf..0330b9cc4b58 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -184,6 +184,7 @@ static struct nla_policy fib4_rule_policy[FRA_MAX+1] __read_mostly = {
 	[FRA_DST]	= { .type = NLA_U32 },
 	[FRA_FWMARK]	= { .type = NLA_U32 },
 	[FRA_FLOW]	= { .type = NLA_U32 },
+	[FRA_TABLE]	= { .type = NLA_U32 },
 };
 
 static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index c7a112b5a185..ab753df20a39 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -953,6 +953,7 @@ fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
 	rtm->rtm_src_len = 0;
 	rtm->rtm_tos = tos;
 	rtm->rtm_table = tb_id;
+	RTA_PUT_U32(skb, RTA_TABLE, tb_id);
 	rtm->rtm_type = type;
 	rtm->rtm_flags = fi->fib_flags;
 	rtm->rtm_scope = scope;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index b873cbcdd0b8..12128b82c9dc 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2652,6 +2652,7 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
 	r->rtm_src_len	= 0;
 	r->rtm_tos	= rt->fl.fl4_tos;
 	r->rtm_table	= RT_TABLE_MAIN;
+	RTA_PUT_U32(skb, RTA_TABLE, RT_TABLE_MAIN);
 	r->rtm_type	= rt->rt_type;
 	r->rtm_scope	= RT_SCOPE_UNIVERSE;
 	r->rtm_protocol = RTPROT_UNSPEC;
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 22a2fdb09831..2c4fbc855e6c 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -129,6 +129,7 @@ static struct nla_policy fib6_rule_policy[RTA_MAX+1] __read_mostly = {
 	[FRA_PRIORITY]	= { .type = NLA_U32 },
 	[FRA_SRC]	= { .minlen = sizeof(struct in6_addr) },
 	[FRA_DST]	= { .minlen = sizeof(struct in6_addr) },
+	[FRA_TABLE]	= { .type = NLA_U32 },
 };
 
 static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index e08d84063c1f..843c5509fced 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1859,7 +1859,8 @@ int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
 		return -EINVAL;
-	return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table);
+	return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb),
+			     rtm_get_table(arg, r->rtm_table));
 }
 
 int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
@@ -1869,7 +1870,8 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
 		return -EINVAL;
-	return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table);
+	return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb),
+			     rtm_get_table(arg, r->rtm_table));
 }
 
 struct rt6_rtnl_dump_arg
@@ -1887,6 +1889,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 	struct nlmsghdr  *nlh;
 	unsigned char	 *b = skb->tail;
 	struct rta_cacheinfo ci;
+	u32 table;
 
 	if (prefix) {	/* user wants prefix routes only */
 		if (!(rt->rt6i_flags & RTF_PREFIX_RT)) {
@@ -1902,9 +1905,11 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 	rtm->rtm_src_len = rt->rt6i_src.plen;
 	rtm->rtm_tos = 0;
 	if (rt->rt6i_table)
-		rtm->rtm_table = rt->rt6i_table->tb6_id;
+		table = rt->rt6i_table->tb6_id;
 	else
-		rtm->rtm_table = RT6_TABLE_UNSPEC;
+		table = RT6_TABLE_UNSPEC;
+	rtm->rtm_table = table;
+	RTA_PUT_U32(skb, RTA_TABLE, table);
 	if (rt->rt6i_flags&RTF_REJECT)
 		rtm->rtm_type = RTN_UNREACHABLE;
 	else if (rt->rt6i_dev && (rt->rt6i_dev->flags&IFF_LOOPBACK))
-- 
cgit v1.2.3


From 1af5a8c4a11cfed0c9a7f30fcfb689981750599c Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Aug 2006 23:10:46 -0700
Subject: [IPV4]: Increase number of possible routing tables to 2^32

Increase the number of possible routing tables to 2^32 by replacing the
fixed sized array of pointers by a hash table and replacing iterations
over all possible table IDs by hash table walking.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    |  25 +++---------
 net/ipv4/fib_frontend.c | 102 ++++++++++++++++++++++++++++++++----------------
 net/ipv4/fib_hash.c     |  26 ++++++------
 net/ipv4/fib_rules.c    |   4 +-
 net/ipv4/fib_trie.c     |  26 ++++++------
 5 files changed, 101 insertions(+), 82 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 0dcbf166eb94..8e9ba563d342 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -150,6 +150,7 @@ struct fib_result_nl {
 #endif /* CONFIG_IP_ROUTE_MULTIPATH_WRANDOM */
 
 struct fib_table {
+	struct hlist_node tb_hlist;
 	u32		tb_id;
 	unsigned	tb_stamp;
 	int		(*tb_lookup)(struct fib_table *tb, const struct flowi *flp, struct fib_result *res);
@@ -200,29 +201,13 @@ static inline void fib_select_default(const struct flowi *flp, struct fib_result
 }
 
 #else /* CONFIG_IP_MULTIPLE_TABLES */
-#define ip_fib_local_table (fib_tables[RT_TABLE_LOCAL])
-#define ip_fib_main_table (fib_tables[RT_TABLE_MAIN])
+#define ip_fib_local_table fib_get_table(RT_TABLE_LOCAL)
+#define ip_fib_main_table fib_get_table(RT_TABLE_MAIN)
 
-extern struct fib_table * fib_tables[RT_TABLE_MAX+1];
 extern int fib_lookup(struct flowi *flp, struct fib_result *res);
-extern struct fib_table *__fib_new_table(u32 id);
-
-static inline struct fib_table *fib_get_table(u32 id)
-{
-	if (id == 0)
-		id = RT_TABLE_MAIN;
-
-	return fib_tables[id];
-}
-
-static inline struct fib_table *fib_new_table(u32 id)
-{
-	if (id == 0)
-		id = RT_TABLE_MAIN;
-
-	return fib_tables[id] ? : __fib_new_table(id);
-}
 
+extern struct fib_table *fib_new_table(u32 id);
+extern struct fib_table *fib_get_table(u32 id);
 extern void fib_select_default(const struct flowi *flp, struct fib_result *res);
 
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 2696ede52de2..ad4c14f968a1 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -37,6 +37,7 @@
 #include <linux/skbuff.h>
 #include <linux/netlink.h>
 #include <linux/init.h>
+#include <linux/list.h>
 
 #include <net/ip.h>
 #include <net/protocol.h>
@@ -51,48 +52,67 @@
 
 #ifndef CONFIG_IP_MULTIPLE_TABLES
 
-#define RT_TABLE_MIN RT_TABLE_MAIN
-
 struct fib_table *ip_fib_local_table;
 struct fib_table *ip_fib_main_table;
 
-#else
+#define FIB_TABLE_HASHSZ 1
+static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
-#define RT_TABLE_MIN 1
+#else
 
-struct fib_table *fib_tables[RT_TABLE_MAX+1];
+#define FIB_TABLE_HASHSZ 256
+static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
 
-struct fib_table *__fib_new_table(u32 id)
+struct fib_table *fib_new_table(u32 id)
 {
 	struct fib_table *tb;
+	unsigned int h;
 
+	if (id == 0)
+		id = RT_TABLE_MAIN;
+	tb = fib_get_table(id);
+	if (tb)
+		return tb;
 	tb = fib_hash_init(id);
 	if (!tb)
 		return NULL;
-	fib_tables[id] = tb;
+	h = id & (FIB_TABLE_HASHSZ - 1);
+	hlist_add_head_rcu(&tb->tb_hlist, &fib_table_hash[h]);
 	return tb;
 }
 
+struct fib_table *fib_get_table(u32 id)
+{
+	struct fib_table *tb;
+	struct hlist_node *node;
+	unsigned int h;
 
+	if (id == 0)
+		id = RT_TABLE_MAIN;
+	h = id & (FIB_TABLE_HASHSZ - 1);
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(tb, node, &fib_table_hash[h], tb_hlist) {
+		if (tb->tb_id == id) {
+			rcu_read_unlock();
+			return tb;
+		}
+	}
+	rcu_read_unlock();
+	return NULL;
+}
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
-
 static void fib_flush(void)
 {
 	int flushed = 0;
-#ifdef CONFIG_IP_MULTIPLE_TABLES
 	struct fib_table *tb;
-	u32 id;
+	struct hlist_node *node;
+	unsigned int h;
 
-	for (id = RT_TABLE_MAX; id>0; id--) {
-		if ((tb = fib_get_table(id))==NULL)
-			continue;
-		flushed += tb->tb_flush(tb);
+	for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
+		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb_hlist)
+			flushed += tb->tb_flush(tb);
 	}
-#else /* CONFIG_IP_MULTIPLE_TABLES */
-	flushed += ip_fib_main_table->tb_flush(ip_fib_main_table);
-	flushed += ip_fib_local_table->tb_flush(ip_fib_local_table);
-#endif /* CONFIG_IP_MULTIPLE_TABLES */
 
 	if (flushed)
 		rt_cache_flush(-1);
@@ -334,29 +354,37 @@ int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 
 int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	u32 t;
-	u32 s_t;
+	unsigned int h, s_h;
+	unsigned int e = 0, s_e;
 	struct fib_table *tb;
+	struct hlist_node *node;
+	int dumped = 0;
 
 	if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) &&
 	    ((struct rtmsg*)NLMSG_DATA(cb->nlh))->rtm_flags&RTM_F_CLONED)
 		return ip_rt_dump(skb, cb);
 
-	s_t = cb->args[0];
-	if (s_t == 0)
-		s_t = cb->args[0] = RT_TABLE_MIN;
-
-	for (t=s_t; t<=RT_TABLE_MAX; t++) {
-		if (t < s_t) continue;
-		if (t > s_t)
-			memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
-		if ((tb = fib_get_table(t))==NULL)
-			continue;
-		if (tb->tb_dump(tb, skb, cb) < 0) 
-			break;
+	s_h = cb->args[0];
+	s_e = cb->args[1];
+
+	for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) {
+		e = 0;
+		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb_hlist) {
+			if (e < s_e)
+				goto next;
+			if (dumped)
+				memset(&cb->args[2], 0, sizeof(cb->args) -
+				                 2 * sizeof(cb->args[0]));
+			if (tb->tb_dump(tb, skb, cb) < 0)
+				goto out;
+			dumped = 1;
+next:
+			e++;
+		}
 	}
-
-	cb->args[0] = t;
+out:
+	cb->args[1] = e;
+	cb->args[0] = h;
 
 	return skb->len;
 }
@@ -654,9 +682,15 @@ static struct notifier_block fib_netdev_notifier = {
 
 void __init ip_fib_init(void)
 {
+	unsigned int i;
+
+	for (i = 0; i < FIB_TABLE_HASHSZ; i++)
+		INIT_HLIST_HEAD(&fib_table_hash[i]);
 #ifndef CONFIG_IP_MULTIPLE_TABLES
 	ip_fib_local_table = fib_hash_init(RT_TABLE_LOCAL);
+	hlist_add_head_rcu(&ip_fib_local_table->tb_hlist, &fib_table_hash[0]);
 	ip_fib_main_table  = fib_hash_init(RT_TABLE_MAIN);
+	hlist_add_head_rcu(&ip_fib_main_table->tb_hlist, &fib_table_hash[0]);
 #else
 	fib4_rules_init();
 #endif
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index f8d5c8024ccb..b5bee1a71e5c 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -684,7 +684,7 @@ fn_hash_dump_bucket(struct sk_buff *skb, struct netlink_callback *cb,
 	struct fib_node *f;
 	int i, s_i;
 
-	s_i = cb->args[3];
+	s_i = cb->args[4];
 	i = 0;
 	hlist_for_each_entry(f, node, head, fn_hash) {
 		struct fib_alias *fa;
@@ -704,14 +704,14 @@ fn_hash_dump_bucket(struct sk_buff *skb, struct netlink_callback *cb,
 					  fa->fa_tos,
 					  fa->fa_info,
 					  NLM_F_MULTI) < 0) {
-				cb->args[3] = i;
+				cb->args[4] = i;
 				return -1;
 			}
 		next:
 			i++;
 		}
 	}
-	cb->args[3] = i;
+	cb->args[4] = i;
 	return skb->len;
 }
 
@@ -722,21 +722,21 @@ fn_hash_dump_zone(struct sk_buff *skb, struct netlink_callback *cb,
 {
 	int h, s_h;
 
-	s_h = cb->args[2];
+	s_h = cb->args[3];
 	for (h=0; h < fz->fz_divisor; h++) {
 		if (h < s_h) continue;
 		if (h > s_h)
-			memset(&cb->args[3], 0,
-			       sizeof(cb->args) - 3*sizeof(cb->args[0]));
+			memset(&cb->args[4], 0,
+			       sizeof(cb->args) - 4*sizeof(cb->args[0]));
 		if (fz->fz_hash == NULL ||
 		    hlist_empty(&fz->fz_hash[h]))
 			continue;
 		if (fn_hash_dump_bucket(skb, cb, tb, fz, &fz->fz_hash[h])<0) {
-			cb->args[2] = h;
+			cb->args[3] = h;
 			return -1;
 		}
 	}
-	cb->args[2] = h;
+	cb->args[3] = h;
 	return skb->len;
 }
 
@@ -746,21 +746,21 @@ static int fn_hash_dump(struct fib_table *tb, struct sk_buff *skb, struct netlin
 	struct fn_zone *fz;
 	struct fn_hash *table = (struct fn_hash*)tb->tb_data;
 
-	s_m = cb->args[1];
+	s_m = cb->args[2];
 	read_lock(&fib_hash_lock);
 	for (fz = table->fn_zone_list, m=0; fz; fz = fz->fz_next, m++) {
 		if (m < s_m) continue;
 		if (m > s_m)
-			memset(&cb->args[2], 0,
-			       sizeof(cb->args) - 2*sizeof(cb->args[0]));
+			memset(&cb->args[3], 0,
+			       sizeof(cb->args) - 3*sizeof(cb->args[0]));
 		if (fn_hash_dump_zone(skb, cb, tb, fz) < 0) {
-			cb->args[1] = m;
+			cb->args[2] = m;
 			read_unlock(&fib_hash_lock);
 			return -1;
 		}
 	}
 	read_unlock(&fib_hash_lock);
-	cb->args[1] = m;
+	cb->args[2] = m;
 	return skb->len;
 }
 
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 0330b9cc4b58..ce185ac6f260 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -172,8 +172,8 @@ static struct fib_table *fib_empty_table(void)
 	u32 id;
 
 	for (id = 1; id <= RT_TABLE_MAX; id++)
-		if (fib_tables[id] == NULL)
-			return __fib_new_table(id);
+		if (fib_get_table(id) == NULL)
+			return fib_new_table(id);
 	return NULL;
 }
 
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 4a27b2d573a3..2a580eb2579b 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1848,7 +1848,7 @@ static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fi
 
 	u32 xkey = htonl(key);
 
-	s_i = cb->args[3];
+	s_i = cb->args[4];
 	i = 0;
 
 	/* rcu_read_lock is hold by caller */
@@ -1870,12 +1870,12 @@ static int fn_trie_dump_fa(t_key key, int plen, struct list_head *fah, struct fi
 				  plen,
 				  fa->fa_tos,
 				  fa->fa_info, 0) < 0) {
-			cb->args[3] = i;
+			cb->args[4] = i;
 			return -1;
 		}
 		i++;
 	}
-	cb->args[3] = i;
+	cb->args[4] = i;
 	return skb->len;
 }
 
@@ -1886,14 +1886,14 @@ static int fn_trie_dump_plen(struct trie *t, int plen, struct fib_table *tb, str
 	struct list_head *fa_head;
 	struct leaf *l = NULL;
 
-	s_h = cb->args[2];
+	s_h = cb->args[3];
 
 	for (h = 0; (l = nextleaf(t, l)) != NULL; h++) {
 		if (h < s_h)
 			continue;
 		if (h > s_h)
-			memset(&cb->args[3], 0,
-			       sizeof(cb->args) - 3*sizeof(cb->args[0]));
+			memset(&cb->args[4], 0,
+			       sizeof(cb->args) - 4*sizeof(cb->args[0]));
 
 		fa_head = get_fa_head(l, plen);
 
@@ -1904,11 +1904,11 @@ static int fn_trie_dump_plen(struct trie *t, int plen, struct fib_table *tb, str
 			continue;
 
 		if (fn_trie_dump_fa(l->key, plen, fa_head, tb, skb, cb)<0) {
-			cb->args[2] = h;
+			cb->args[3] = h;
 			return -1;
 		}
 	}
-	cb->args[2] = h;
+	cb->args[3] = h;
 	return skb->len;
 }
 
@@ -1917,23 +1917,23 @@ static int fn_trie_dump(struct fib_table *tb, struct sk_buff *skb, struct netlin
 	int m, s_m;
 	struct trie *t = (struct trie *) tb->tb_data;
 
-	s_m = cb->args[1];
+	s_m = cb->args[2];
 
 	rcu_read_lock();
 	for (m = 0; m <= 32; m++) {
 		if (m < s_m)
 			continue;
 		if (m > s_m)
-			memset(&cb->args[2], 0,
-				sizeof(cb->args) - 2*sizeof(cb->args[0]));
+			memset(&cb->args[3], 0,
+				sizeof(cb->args) - 3*sizeof(cb->args[0]));
 
 		if (fn_trie_dump_plen(t, 32-m, tb, skb, cb)<0) {
-			cb->args[1] = m;
+			cb->args[2] = m;
 			goto out;
 		}
 	}
 	rcu_read_unlock();
-	cb->args[1] = m;
+	cb->args[2] = m;
 	return skb->len;
 out:
 	rcu_read_unlock();
-- 
cgit v1.2.3


From 1b43af5480c351dbcb2eef478bafe179cbeb6e83 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Aug 2006 23:11:17 -0700
Subject: [IPV6]: Increase number of possible routing tables to 2^32

Increase number of possible routing tables to 2^32 by replacing iterations
over all possible table IDs by hash table walking.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |   7 ++
 net/ipv6/ip6_fib.c      | 171 ++++++++++++++++++++++++++++++++++++++++++------
 net/ipv6/route.c        | 128 +-----------------------------------
 3 files changed, 159 insertions(+), 147 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 9bfa3cc6cedb..01bfe404784f 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -137,6 +137,13 @@ extern int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *a
 extern int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
 extern int inet6_rtm_getroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
 
+struct rt6_rtnl_dump_arg
+{
+	struct sk_buff *skb;
+	struct netlink_callback *cb;
+};
+
+extern int rt6_dump_route(struct rt6_info *rt, void *p_arg);
 extern void rt6_ifdown(struct net_device *dev);
 extern void rt6_mtu_change(struct net_device *dev, unsigned mtu);
 
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 1f2316187ca4..bececbe9dd2c 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -158,7 +158,26 @@ static struct fib6_table fib6_main_tbl = {
 };
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
+#define FIB_TABLE_HASHSZ 256
+#else
+#define FIB_TABLE_HASHSZ 1
+#endif
+static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
+
+static void fib6_link_table(struct fib6_table *tb)
+{
+	unsigned int h;
+
+	h = tb->tb6_id & (FIB_TABLE_HASHSZ - 1);
 
+	/*
+	 * No protection necessary, this is the only list mutatation
+	 * operation, tables never disappear once they exist.
+	 */
+	hlist_add_head_rcu(&tb->tb6_hlist, &fib_table_hash[h]);
+}
+
+#ifdef CONFIG_IPV6_MULTIPLE_TABLES
 static struct fib6_table fib6_local_tbl = {
 	.tb6_id		= RT6_TABLE_LOCAL,
 	.tb6_lock	= RW_LOCK_UNLOCKED,
@@ -168,9 +187,6 @@ static struct fib6_table fib6_local_tbl = {
 	},
 };
 
-#define FIB_TABLE_HASHSZ 256
-static struct hlist_head fib_table_hash[FIB_TABLE_HASHSZ];
-
 static struct fib6_table *fib6_alloc_table(u32 id)
 {
 	struct fib6_table *table;
@@ -186,19 +202,6 @@ static struct fib6_table *fib6_alloc_table(u32 id)
 	return table;
 }
 
-static void fib6_link_table(struct fib6_table *tb)
-{
-	unsigned int h;
-
-	h = tb->tb6_id & (FIB_TABLE_HASHSZ - 1);
-
-	/*
-	 * No protection necessary, this is the only list mutatation
-	 * operation, tables never disappear once they exist.
-	 */
-	hlist_add_head_rcu(&tb->tb6_hlist, &fib_table_hash[h]);
-}
-
 struct fib6_table *fib6_new_table(u32 id)
 {
 	struct fib6_table *tb;
@@ -263,10 +266,135 @@ struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags,
 
 static void __init fib6_tables_init(void)
 {
+	fib6_link_table(&fib6_main_tbl);
 }
 
 #endif
 
+static int fib6_dump_node(struct fib6_walker_t *w)
+{
+	int res;
+	struct rt6_info *rt;
+
+	for (rt = w->leaf; rt; rt = rt->u.next) {
+		res = rt6_dump_route(rt, w->args);
+		if (res < 0) {
+			/* Frame is full, suspend walking */
+			w->leaf = rt;
+			return 1;
+		}
+		BUG_TRAP(res!=0);
+	}
+	w->leaf = NULL;
+	return 0;
+}
+
+static void fib6_dump_end(struct netlink_callback *cb)
+{
+	struct fib6_walker_t *w = (void*)cb->args[2];
+
+	if (w) {
+		cb->args[2] = 0;
+		kfree(w);
+	}
+	cb->done = (void*)cb->args[3];
+	cb->args[1] = 3;
+}
+
+static int fib6_dump_done(struct netlink_callback *cb)
+{
+	fib6_dump_end(cb);
+	return cb->done ? cb->done(cb) : 0;
+}
+
+static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb,
+			   struct netlink_callback *cb)
+{
+	struct fib6_walker_t *w;
+	int res;
+
+	w = (void *)cb->args[2];
+	w->root = &table->tb6_root;
+
+	if (cb->args[4] == 0) {
+		read_lock_bh(&table->tb6_lock);
+		res = fib6_walk(w);
+		read_unlock_bh(&table->tb6_lock);
+		if (res > 0)
+			cb->args[4] = 1;
+	} else {
+		read_lock_bh(&table->tb6_lock);
+		res = fib6_walk_continue(w);
+		read_unlock_bh(&table->tb6_lock);
+		if (res != 0) {
+			if (res < 0)
+				fib6_walker_unlink(w);
+			goto end;
+		}
+		fib6_walker_unlink(w);
+		cb->args[4] = 0;
+	}
+end:
+	return res;
+}
+
+int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	unsigned int h, s_h;
+	unsigned int e = 0, s_e;
+	struct rt6_rtnl_dump_arg arg;
+	struct fib6_walker_t *w;
+	struct fib6_table *tb;
+	struct hlist_node *node;
+	int res = 0;
+
+	s_h = cb->args[0];
+	s_e = cb->args[1];
+
+	w = (void *)cb->args[2];
+	if (w == NULL) {
+		/* New dump:
+		 *
+		 * 1. hook callback destructor.
+		 */
+		cb->args[3] = (long)cb->done;
+		cb->done = fib6_dump_done;
+
+		/*
+		 * 2. allocate and initialize walker.
+		 */
+		w = kzalloc(sizeof(*w), GFP_ATOMIC);
+		if (w == NULL)
+			return -ENOMEM;
+		w->func = fib6_dump_node;
+		cb->args[2] = (long)w;
+	}
+
+	arg.skb = skb;
+	arg.cb = cb;
+	w->args = &arg;
+
+	for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) {
+		e = 0;
+		hlist_for_each_entry(tb, node, &fib_table_hash[h], tb6_hlist) {
+			if (e < s_e)
+				goto next;
+			res = fib6_dump_table(tb, skb, cb);
+			if (res != 0)
+				goto out;
+next:
+			e++;
+		}
+	}
+out:
+	cb->args[1] = e;
+	cb->args[0] = h;
+
+	res = res < 0 ? res : skb->len;
+	if (res <= 0)
+		fib6_dump_end(cb);
+	return res;
+}
 
 /*
  *	Routing Table
@@ -1187,17 +1315,20 @@ static void fib6_clean_tree(struct fib6_node *root,
 void fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
 		    int prune, void *arg)
 {
-	int i;
 	struct fib6_table *table;
+	struct hlist_node *node;
+	unsigned int h;
 
-	for (i = FIB6_TABLE_MIN; i <= FIB6_TABLE_MAX; i++) {
-		table = fib6_get_table(i);
-		if (table != NULL) {
+	rcu_read_lock();
+	for (h = 0; h < FIB_TABLE_HASHSZ; h++) {
+		hlist_for_each_entry_rcu(table, node, &fib_table_hash[h],
+					 tb6_hlist) {
 			write_lock_bh(&table->tb6_lock);
 			fib6_clean_tree(&table->tb6_root, func, prune, arg);
 			write_unlock_bh(&table->tb6_lock);
 		}
 	}
+	rcu_read_unlock();
 }
 
 static int fib6_prune_clone(struct rt6_info *rt, void *arg)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 843c5509fced..9ce28277f47f 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1874,12 +1874,6 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 			     rtm_get_table(arg, r->rtm_table));
 }
 
-struct rt6_rtnl_dump_arg
-{
-	struct sk_buff *skb;
-	struct netlink_callback *cb;
-};
-
 static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
 			 struct in6_addr *dst, struct in6_addr *src,
 			 int iif, int type, u32 pid, u32 seq,
@@ -1976,7 +1970,7 @@ rtattr_failure:
 	return -1;
 }
 
-static int rt6_dump_route(struct rt6_info *rt, void *p_arg)
+int rt6_dump_route(struct rt6_info *rt, void *p_arg)
 {
 	struct rt6_rtnl_dump_arg *arg = (struct rt6_rtnl_dump_arg *) p_arg;
 	int prefix;
@@ -1992,126 +1986,6 @@ static int rt6_dump_route(struct rt6_info *rt, void *p_arg)
 		     prefix, NLM_F_MULTI);
 }
 
-static int fib6_dump_node(struct fib6_walker_t *w)
-{
-	int res;
-	struct rt6_info *rt;
-
-	for (rt = w->leaf; rt; rt = rt->u.next) {
-		res = rt6_dump_route(rt, w->args);
-		if (res < 0) {
-			/* Frame is full, suspend walking */
-			w->leaf = rt;
-			return 1;
-		}
-		BUG_TRAP(res!=0);
-	}
-	w->leaf = NULL;
-	return 0;
-}
-
-static void fib6_dump_end(struct netlink_callback *cb)
-{
-	struct fib6_walker_t *w = (void*)cb->args[0];
-
-	if (w) {
-		cb->args[0] = 0;
-		kfree(w);
-	}
-	cb->done = (void*)cb->args[1];
-	cb->args[1] = 0;
-}
-
-static int fib6_dump_done(struct netlink_callback *cb)
-{
-	fib6_dump_end(cb);
-	return cb->done ? cb->done(cb) : 0;
-}
-
-int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
-{
-	struct fib6_table *table;
-	struct rt6_rtnl_dump_arg arg;
-	struct fib6_walker_t *w;
-	int i, res = 0;
-
-	arg.skb = skb;
-	arg.cb = cb;
-
-	/*
-	 * cb->args[0] = pointer to walker structure
-	 * cb->args[1] = saved cb->done() pointer
-	 * cb->args[2] = current table being dumped
-	 */
-
-	w = (void*)cb->args[0];
-	if (w == NULL) {
-		/* New dump:
-		 * 
-		 * 1. hook callback destructor.
-		 */
-		cb->args[1] = (long)cb->done;
-		cb->done = fib6_dump_done;
-
-		/*
-		 * 2. allocate and initialize walker.
-		 */
-		w = kzalloc(sizeof(*w), GFP_ATOMIC);
-		if (w == NULL)
-			return -ENOMEM;
-		w->func = fib6_dump_node;
-		w->args = &arg;
-		cb->args[0] = (long)w;
-		cb->args[2] = FIB6_TABLE_MIN;
-	} else {
-		w->args = &arg;
-		i = cb->args[2];
-		if (i > FIB6_TABLE_MAX)
-			goto end;
-
-		table = fib6_get_table(i);
-		if (table != NULL) {
-			read_lock_bh(&table->tb6_lock);
-			w->root = &table->tb6_root;
-			res = fib6_walk_continue(w);
-			read_unlock_bh(&table->tb6_lock);
-			if (res != 0) {
-				if (res < 0)
-					fib6_walker_unlink(w);
-				goto end;
-			}
-		}
-
-		fib6_walker_unlink(w);
-		cb->args[2] = ++i;
-	}
-
-	for (i = cb->args[2]; i <= FIB6_TABLE_MAX; i++) {
-		table = fib6_get_table(i);
-		if (table == NULL)
-			continue;
-
-		read_lock_bh(&table->tb6_lock);
-		w->root = &table->tb6_root;
-		res = fib6_walk(w);
-		read_unlock_bh(&table->tb6_lock);
-		if (res)
-			break;
-	}
-end:
-	cb->args[2] = i;
-
-	res = res < 0 ? res : skb->len;
-	/* res < 0 is an error. (really, impossible)
-	   res == 0 means that dump is complete, but skb still can contain data.
-	   res > 0 dump is not complete, but frame is full.
-	 */
-	/* Destroy walker, if dump of this table is complete. */
-	if (res <= 0)
-		fib6_dump_end(cb);
-	return res;
-}
-
 int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 {
 	struct rtattr **rta = arg;
-- 
cgit v1.2.3


From abcab268303c22d24fc89fedd35d82271d20f5da Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Aug 2006 23:11:47 -0700
Subject: [DECNET]: Increase number of possible routing tables to 2^32

Increase the number of possible routing tables to 2^32 by replacing the
fixed sized array of pointers by a hash table and replacing iterations
over all possible table IDs by hash table walking.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dn_fib.h  |   3 +-
 net/decnet/dn_fib.c   |  49 --------------------
 net/decnet/dn_rules.c |   2 +-
 net/decnet/dn_table.c | 125 ++++++++++++++++++++++++++++++++++++--------------
 4 files changed, 93 insertions(+), 86 deletions(-)

(limited to 'include')

diff --git a/include/net/dn_fib.h b/include/net/dn_fib.h
index cd9c3782f838..d97aa10c463f 100644
--- a/include/net/dn_fib.h
+++ b/include/net/dn_fib.h
@@ -94,6 +94,7 @@ struct dn_fib_node {
 
 
 struct dn_fib_table {
+	struct hlist_node hlist;
 	u32 n;
 
 	int (*insert)(struct dn_fib_table *t, struct rtmsg *r, 
@@ -177,8 +178,6 @@ static inline void dn_fib_res_put(struct dn_fib_res *res)
 		fib_rule_put(res->r);
 }
 
-extern struct dn_fib_table *dn_fib_tables[];
-
 #else /* Endnode */
 
 #define dn_fib_init()  do { } while(0)
diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c
index fb596373daa8..5ccca3ed53bd 100644
--- a/net/decnet/dn_fib.c
+++ b/net/decnet/dn_fib.c
@@ -532,39 +532,6 @@ int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	return -ENOBUFS;
 }
 
-
-int dn_fib_dump(struct sk_buff *skb, struct netlink_callback *cb)
-{
-	u32 t;
-	u32 s_t;
-	struct dn_fib_table *tb;
-
-	if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) &&
-		((struct rtmsg *)NLMSG_DATA(cb->nlh))->rtm_flags&RTM_F_CLONED)
-			return dn_cache_dump(skb, cb);
-
-	s_t = cb->args[0];
-	if (s_t == 0)
-		s_t = cb->args[0] = RT_MIN_TABLE;
-
-	for(t = s_t; t <= RT_TABLE_MAX; t++) {
-		if (t < s_t)
-			continue;
-		if (t > s_t)
-			memset(&cb->args[1], 0,
-			       sizeof(cb->args) - sizeof(cb->args[0]));
-		tb = dn_fib_get_table(t, 0);
-		if (tb == NULL)
-			continue;
-		if (tb->dump(tb, skb, cb) < 0)
-			break;
-	}
-
-	cb->args[0] = t;
-
-	return skb->len;
-}
-
 static void fib_magic(int cmd, int type, __le16 dst, int dst_len, struct dn_ifaddr *ifa)
 {
 	struct dn_fib_table *tb;
@@ -762,22 +729,6 @@ int dn_fib_sync_up(struct net_device *dev)
         return ret;
 }
 
-void dn_fib_flush(void)
-{
-        int flushed = 0;
-        struct dn_fib_table *tb;
-        u32 id;
-
-        for(id = RT_TABLE_MAX; id > 0; id--) {
-                if ((tb = dn_fib_get_table(id, 0)) == NULL)
-                        continue;
-                flushed += tb->flush(tb);
-        }
-
-        if (flushed)
-                dn_rt_cache_flush(-1);
-}
-
 static struct notifier_block dn_fib_dnaddr_notifier = {
 	.notifier_call = dn_fib_dnaddr_event,
 };
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index 096f1273e714..878312ff34ec 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -210,7 +210,7 @@ unsigned dnet_addr_type(__le16 addr)
 	struct flowi fl = { .nl_u = { .dn_u = { .daddr = addr } } };
 	struct dn_fib_res res;
 	unsigned ret = RTN_UNICAST;
-	struct dn_fib_table *tb = dn_fib_tables[RT_TABLE_LOCAL];
+	struct dn_fib_table *tb = dn_fib_get_table(RT_TABLE_LOCAL, 0);
 
 	res.r = NULL;
 
diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c
index eca7c1e10c80..10e87262b6fb 100644
--- a/net/decnet/dn_table.c
+++ b/net/decnet/dn_table.c
@@ -75,9 +75,9 @@ for( ; ((f) = *(fp)) != NULL; (fp) = &(f)->fn_next)
 for( ; ((f) = *(fp)) != NULL && dn_key_eq((f)->fn_key, (key)); (fp) = &(f)->fn_next)
 
 #define RT_TABLE_MIN 1
-
+#define DN_FIB_TABLE_HASHSZ 256
+static struct hlist_head dn_fib_table_hash[DN_FIB_TABLE_HASHSZ];
 static DEFINE_RWLOCK(dn_fib_tables_lock);
-struct dn_fib_table *dn_fib_tables[RT_TABLE_MAX + 1];
 
 static kmem_cache_t *dn_hash_kmem __read_mostly;
 static int dn_fib_hash_zombies;
@@ -361,7 +361,7 @@ static __inline__ int dn_hash_dump_bucket(struct sk_buff *skb,
 {
 	int i, s_i;
 
-	s_i = cb->args[3];
+	s_i = cb->args[4];
 	for(i = 0; f; i++, f = f->fn_next) {
 		if (i < s_i)
 			continue;
@@ -374,11 +374,11 @@ static __inline__ int dn_hash_dump_bucket(struct sk_buff *skb,
 				(f->fn_state & DN_S_ZOMBIE) ? 0 : f->fn_type,
 				f->fn_scope, &f->fn_key, dz->dz_order, 
 				f->fn_info, NLM_F_MULTI) < 0) {
-			cb->args[3] = i;
+			cb->args[4] = i;
 			return -1;
 		}
 	}
-	cb->args[3] = i;
+	cb->args[4] = i;
 	return skb->len;
 }
 
@@ -389,20 +389,20 @@ static __inline__ int dn_hash_dump_zone(struct sk_buff *skb,
 {
 	int h, s_h;
 
-	s_h = cb->args[2];
+	s_h = cb->args[3];
 	for(h = 0; h < dz->dz_divisor; h++) {
 		if (h < s_h)
 			continue;
 		if (h > s_h)
-			memset(&cb->args[3], 0, sizeof(cb->args) - 3*sizeof(cb->args[0]));
+			memset(&cb->args[4], 0, sizeof(cb->args) - 4*sizeof(cb->args[0]));
 		if (dz->dz_hash == NULL || dz->dz_hash[h] == NULL)
 			continue;
 		if (dn_hash_dump_bucket(skb, cb, tb, dz, dz->dz_hash[h]) < 0) {
-			cb->args[2] = h;
+			cb->args[3] = h;
 			return -1;
 		}
 	}
-	cb->args[2] = h;
+	cb->args[3] = h;
 	return skb->len;
 }
 
@@ -413,26 +413,63 @@ static int dn_fib_table_dump(struct dn_fib_table *tb, struct sk_buff *skb,
 	struct dn_zone *dz;
 	struct dn_hash *table = (struct dn_hash *)tb->data;
 
-	s_m = cb->args[1];
+	s_m = cb->args[2];
 	read_lock(&dn_fib_tables_lock);
 	for(dz = table->dh_zone_list, m = 0; dz; dz = dz->dz_next, m++) {
 		if (m < s_m)
 			continue;
 		if (m > s_m)
-			memset(&cb->args[2], 0, sizeof(cb->args) - 2*sizeof(cb->args[0]));
+			memset(&cb->args[3], 0, sizeof(cb->args) - 3*sizeof(cb->args[0]));
 
 		if (dn_hash_dump_zone(skb, cb, tb, dz) < 0) {
-			cb->args[1] = m;
+			cb->args[2] = m;
 			read_unlock(&dn_fib_tables_lock);
 			return -1;
 		}
 	}
 	read_unlock(&dn_fib_tables_lock);
-	cb->args[1] = m;
+	cb->args[2] = m;
 
         return skb->len;
 }
 
+int dn_fib_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	unsigned int h, s_h;
+	unsigned int e = 0, s_e;
+	struct dn_fib_table *tb;
+	struct hlist_node *node;
+	int dumped = 0;
+
+	if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) &&
+		((struct rtmsg *)NLMSG_DATA(cb->nlh))->rtm_flags&RTM_F_CLONED)
+			return dn_cache_dump(skb, cb);
+
+	s_h = cb->args[0];
+	s_e = cb->args[1];
+
+	for (h = s_h; h < DN_FIB_TABLE_HASHSZ; h++, s_h = 0) {
+		e = 0;
+		hlist_for_each_entry(tb, node, &dn_fib_table_hash[h], hlist) {
+			if (e < s_e)
+				goto next;
+			if (dumped)
+				memset(&cb->args[2], 0, sizeof(cb->args) -
+				                 2 * sizeof(cb->args[0]));
+			if (tb->dump(tb, skb, cb) < 0)
+				goto out;
+			dumped = 1;
+next:
+			e++;
+		}
+	}
+out:
+	cb->args[1] = e;
+	cb->args[0] = h;
+
+	return skb->len;
+}
+
 static int dn_fib_table_insert(struct dn_fib_table *tb, struct rtmsg *r, struct dn_kern_rta *rta, struct nlmsghdr *n, struct netlink_skb_parms *req)
 {
 	struct dn_hash *table = (struct dn_hash *)tb->data;
@@ -744,6 +781,8 @@ out:
 struct dn_fib_table *dn_fib_get_table(u32 n, int create)
 {
         struct dn_fib_table *t;
+	struct hlist_node *node;
+	unsigned int h;
 
         if (n < RT_TABLE_MIN)
                 return NULL;
@@ -751,8 +790,15 @@ struct dn_fib_table *dn_fib_get_table(u32 n, int create)
         if (n > RT_TABLE_MAX)
                 return NULL;
 
-        if (dn_fib_tables[n]) 
-                return dn_fib_tables[n];
+	h = n & (DN_FIB_TABLE_HASHSZ - 1);
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(t, node, &dn_fib_table_hash[h], hlist) {
+		if (t->n == n) {
+			rcu_read_unlock();
+			return t;
+		}
+	}
+	rcu_read_unlock();
 
         if (!create)
                 return NULL;
@@ -773,33 +819,37 @@ struct dn_fib_table *dn_fib_get_table(u32 n, int create)
         t->flush  = dn_fib_table_flush;
         t->dump = dn_fib_table_dump;
 	memset(t->data, 0, sizeof(struct dn_hash));
-        dn_fib_tables[n] = t;
+	hlist_add_head_rcu(&t->hlist, &dn_fib_table_hash[h]);
 
         return t;
 }
 
-static void dn_fib_del_tree(u32 n)
-{
-	struct dn_fib_table *t;
-
-	write_lock(&dn_fib_tables_lock);
-	t = dn_fib_tables[n];
-	dn_fib_tables[n] = NULL;
-	write_unlock(&dn_fib_tables_lock);
-
-	kfree(t);
-}
-
 struct dn_fib_table *dn_fib_empty_table(void)
 {
         u32 id;
 
         for(id = RT_TABLE_MIN; id <= RT_TABLE_MAX; id++)
-                if (dn_fib_tables[id] == NULL)
+		if (dn_fib_get_table(id, 0) == NULL)
                         return dn_fib_get_table(id, 1);
         return NULL;
 }
 
+void dn_fib_flush(void)
+{
+        int flushed = 0;
+        struct dn_fib_table *tb;
+	struct hlist_node *node;
+	unsigned int h;
+
+	for (h = 0; h < DN_FIB_TABLE_HASHSZ; h++) {
+		hlist_for_each_entry(tb, node, &dn_fib_table_hash[h], hlist)
+	                flushed += tb->flush(tb);
+        }
+
+        if (flushed)
+                dn_rt_cache_flush(-1);
+}
+
 void __init dn_fib_table_init(void)
 {
 	dn_hash_kmem = kmem_cache_create("dn_fib_info_cache",
@@ -810,10 +860,17 @@ void __init dn_fib_table_init(void)
 
 void __exit dn_fib_table_cleanup(void)
 {
-	int i;
-
-	for (i = RT_TABLE_MIN; i <= RT_TABLE_MAX; ++i)
-		dn_fib_del_tree(i);
+	struct dn_fib_table *t;
+	struct hlist_node *node, *next;
+	unsigned int h;
 
-	return;
+	write_lock(&dn_fib_tables_lock);
+	for (h = 0; h < DN_FIB_TABLE_HASHSZ; h++) {
+		hlist_for_each_entry_safe(t, node, next, &dn_fib_table_hash[h],
+		                          hlist) {
+			hlist_del(&t->hlist);
+			kfree(t);
+		}
+	}
+	write_unlock(&dn_fib_tables_lock);
 }
-- 
cgit v1.2.3


From b801f54917b7c6e8540f877ee562cd0725e62ebd Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Thu, 10 Aug 2006 23:12:34 -0700
Subject: [NET]: Increate RT_TABLE_MAX to 2^32

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index ea422a539a03..7e4aa48680a7 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -239,10 +239,8 @@ enum rt_class_t
 	RT_TABLE_DEFAULT=253,
 	RT_TABLE_MAIN=254,
 	RT_TABLE_LOCAL=255,
-	__RT_TABLE_MAX
+	RT_TABLE_MAX=0xFFFFFFFF
 };
-#define RT_TABLE_MAX (__RT_TABLE_MAX - 1)
-
 
 
 /* Routing message attributes */
-- 
cgit v1.2.3


From 90d41122f79c8c3687d965dde4c6d30a6e0cac4c Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 14 Aug 2006 23:49:16 -0700
Subject: [IPV6] ip6_fib.c: make code static

Make the following needlessly global code static:
- fib6_walker_lock
- struct fib6_walker_list
- fib6_walk_continue()
- fib6_walk()

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h | 25 -------------------------
 net/ipv6/ip6_fib.c    | 29 ++++++++++++++++++++++++-----
 2 files changed, 24 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index c0660cea9a2f..69c444209781 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -92,28 +92,6 @@ struct fib6_walker_t
 	void *args;
 };
 
-extern struct fib6_walker_t fib6_walker_list;
-extern rwlock_t fib6_walker_lock;
-
-static inline void fib6_walker_link(struct fib6_walker_t *w)
-{
-	write_lock_bh(&fib6_walker_lock);
-	w->next = fib6_walker_list.next;
-	w->prev = &fib6_walker_list;
-	w->next->prev = w;
-	w->prev->next = w;
-	write_unlock_bh(&fib6_walker_lock);
-}
-
-static inline void fib6_walker_unlink(struct fib6_walker_t *w)
-{
-	write_lock_bh(&fib6_walker_lock);
-	w->next->prev = w->prev;
-	w->prev->next = w->next;
-	w->prev = w->next = w;
-	write_unlock_bh(&fib6_walker_lock);
-}
-
 struct rt6_statistics {
 	__u32		fib_nodes;
 	__u32		fib_route_nodes;
@@ -195,9 +173,6 @@ struct fib6_node		*fib6_locate(struct fib6_node *root,
 extern void			fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
 					       int prune, void *arg);
 
-extern int			fib6_walk(struct fib6_walker_t *w);
-extern int			fib6_walk_continue(struct fib6_walker_t *w);
-
 extern int			fib6_add(struct fib6_node *root,
 					 struct rt6_info *rt,
 					 struct nlmsghdr *nlh,
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index bececbe9dd2c..be36f4acda94 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -69,8 +69,7 @@ struct fib6_cleaner_t
 	void *arg;
 };
 
-DEFINE_RWLOCK(fib6_walker_lock);
-
+static DEFINE_RWLOCK(fib6_walker_lock);
 
 #ifdef CONFIG_IPV6_SUBTREES
 #define FWS_INIT FWS_S
@@ -82,6 +81,8 @@ DEFINE_RWLOCK(fib6_walker_lock);
 
 static void fib6_prune_clones(struct fib6_node *fn, struct rt6_info *rt);
 static struct fib6_node * fib6_repair_tree(struct fib6_node *fn);
+static int fib6_walk(struct fib6_walker_t *w);
+static int fib6_walk_continue(struct fib6_walker_t *w);
 
 /*
  *	A routing update causes an increase of the serial number on the
@@ -94,13 +95,31 @@ static __u32 rt_sernum;
 
 static DEFINE_TIMER(ip6_fib_timer, fib6_run_gc, 0, 0);
 
-struct fib6_walker_t fib6_walker_list = {
+static struct fib6_walker_t fib6_walker_list = {
 	.prev	= &fib6_walker_list,
 	.next	= &fib6_walker_list, 
 };
 
 #define FOR_WALKERS(w) for ((w)=fib6_walker_list.next; (w) != &fib6_walker_list; (w)=(w)->next)
 
+static inline void fib6_walker_link(struct fib6_walker_t *w)
+{
+	write_lock_bh(&fib6_walker_lock);
+	w->next = fib6_walker_list.next;
+	w->prev = &fib6_walker_list;
+	w->next->prev = w;
+	w->prev->next = w;
+	write_unlock_bh(&fib6_walker_lock);
+}
+
+static inline void fib6_walker_unlink(struct fib6_walker_t *w)
+{
+	write_lock_bh(&fib6_walker_lock);
+	w->next->prev = w->prev;
+	w->prev->next = w->next;
+	w->prev = w->next = w;
+	write_unlock_bh(&fib6_walker_lock);
+}
 static __inline__ u32 fib6_new_sernum(void)
 {
 	u32 n = ++rt_sernum;
@@ -1173,7 +1192,7 @@ int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct ne
  *	<0  -> walk is terminated by an error.
  */
 
-int fib6_walk_continue(struct fib6_walker_t *w)
+static int fib6_walk_continue(struct fib6_walker_t *w)
 {
 	struct fib6_node *fn, *pn;
 
@@ -1247,7 +1266,7 @@ int fib6_walk_continue(struct fib6_walker_t *w)
 	}
 }
 
-int fib6_walk(struct fib6_walker_t *w)
+static int fib6_walk(struct fib6_walker_t *w)
 {
 	int res;
 
-- 
cgit v1.2.3


From 2aa7f36cdb332a32849afbf25fcbf35dce5b1940 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 14 Aug 2006 23:55:20 -0700
Subject: [DECNET]: cleanups

- make the following needlessly global functions static:
  - dn_fib.c: dn_fib_sync_down()
  - dn_fib.c: dn_fib_sync_up()
  - dn_rules.c: dn_fib_rule_action()
- remove the following unneeded prototype:
  - dn_fib.c: dn_cache_dump()

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dn_fib.h  | 3 ---
 net/decnet/dn_fib.c   | 9 +++++----
 net/decnet/dn_rules.c | 4 ++--
 3 files changed, 7 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/net/dn_fib.h b/include/net/dn_fib.h
index d97aa10c463f..f01626cbbed6 100644
--- a/include/net/dn_fib.h
+++ b/include/net/dn_fib.h
@@ -131,9 +131,6 @@ extern __le16 dn_fib_get_attr16(struct rtattr *attr, int attrlen, int type);
 extern void dn_fib_flush(void);
 extern void dn_fib_select_multipath(const struct flowi *fl,
 					struct dn_fib_res *res);
-extern int dn_fib_sync_down(__le16 local, struct net_device *dev,
-				int force);
-extern int dn_fib_sync_up(struct net_device *dev);
 
 /*
  * dn_tables.c
diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c
index 5ccca3ed53bd..1cf010124ec5 100644
--- a/net/decnet/dn_fib.c
+++ b/net/decnet/dn_fib.c
@@ -55,8 +55,6 @@
 
 #define endfor_nexthops(fi) }
 
-extern int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb);
-
 static DEFINE_SPINLOCK(dn_fib_multipath_lock);
 static struct dn_fib_info *dn_fib_info_list;
 static DEFINE_SPINLOCK(dn_fib_info_lock);
@@ -80,6 +78,9 @@ static struct
 	[RTN_XRESOLVE] =    { .error = -EINVAL, .scope = RT_SCOPE_NOWHERE },
 };
 
+static int dn_fib_sync_down(__le16 local, struct net_device *dev, int force);
+static int dn_fib_sync_up(struct net_device *dev);
+
 void dn_fib_free_info(struct dn_fib_info *fi)
 {
 	if (fi->fib_dead == 0) {
@@ -651,7 +652,7 @@ static int dn_fib_dnaddr_event(struct notifier_block *this, unsigned long event,
 	return NOTIFY_DONE;
 }
 
-int dn_fib_sync_down(__le16 local, struct net_device *dev, int force)
+static int dn_fib_sync_down(__le16 local, struct net_device *dev, int force)
 {
         int ret = 0;
         int scope = RT_SCOPE_NOWHERE;
@@ -695,7 +696,7 @@ int dn_fib_sync_down(__le16 local, struct net_device *dev, int force)
 }
 
 
-int dn_fib_sync_up(struct net_device *dev)
+static int dn_fib_sync_up(struct net_device *dev)
 {
         int ret = 0;
 
diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c
index 977bb56c3ce4..50e819edf8c7 100644
--- a/net/decnet/dn_rules.c
+++ b/net/decnet/dn_rules.c
@@ -75,8 +75,8 @@ int dn_fib_lookup(struct flowi *flp, struct dn_fib_res *res)
 	return err;
 }
 
-int dn_fib_rule_action(struct fib_rule *rule, struct flowi *flp, int flags,
-		       struct fib_lookup_arg *arg)
+static int dn_fib_rule_action(struct fib_rule *rule, struct flowi *flp,
+			      int flags, struct fib_lookup_arg *arg)
 {
 	int err = -EAGAIN;
 	struct dn_fib_table *tbl;
-- 
cgit v1.2.3


From 81aa646cc4df3779bcbf9d18cc2c0813ee9b3262 Mon Sep 17 00:00:00 2001
From: Martin Bligh <mbligh@google.com>
Date: Mon, 14 Aug 2006 23:57:10 -0700
Subject: [IPV4]: add the UdpSndbufErrors and UdpRcvbufErrors MIBs

Signed-off-by: Martin Bligh <mbligh@google.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 include/linux/snmp.h |  2 ++
 net/ipv4/proc.c      |  2 ++
 net/ipv4/udp.c       | 16 +++++++++++++++-
 3 files changed, 19 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/snmp.h b/include/linux/snmp.h
index 4db25d5c7cd1..30156556f78d 100644
--- a/include/linux/snmp.h
+++ b/include/linux/snmp.h
@@ -155,6 +155,8 @@ enum
 	UDP_MIB_NOPORTS,			/* NoPorts */
 	UDP_MIB_INERRORS,			/* InErrors */
 	UDP_MIB_OUTDATAGRAMS,			/* OutDatagrams */
+	UDP_MIB_RCVBUFERRORS,			/* RcvbufErrors */
+	UDP_MIB_SNDBUFERRORS,			/* SndbufErrors */
 	__UDP_MIB_MAX
 };
 
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index d61e2a9d394d..9c6cbe3d9fb8 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -173,6 +173,8 @@ static const struct snmp_mib snmp4_udp_list[] = {
 	SNMP_MIB_ITEM("NoPorts", UDP_MIB_NOPORTS),
 	SNMP_MIB_ITEM("InErrors", UDP_MIB_INERRORS),
 	SNMP_MIB_ITEM("OutDatagrams", UDP_MIB_OUTDATAGRAMS),
+	SNMP_MIB_ITEM("RcvbufErrors", UDP_MIB_RCVBUFERRORS),
+	SNMP_MIB_ITEM("SndbufErrors", UDP_MIB_SNDBUFERRORS),
 	SNMP_MIB_SENTINEL
 };
 
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 87152510980c..514c1e9ae810 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -662,6 +662,16 @@ out:
 		UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS);
 		return len;
 	}
+	/*
+	 * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
+	 * ENOBUFS might not be good (it's not tunable per se), but otherwise
+	 * we don't have a good statistic (IpOutDiscards but it can be too many
+	 * things).  We could add another new stat but at least for now that
+	 * seems like overkill.
+	 */
+	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
+		UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS);
+	}
 	return err;
 
 do_confirm:
@@ -981,6 +991,7 @@ static int udp_encap_rcv(struct sock * sk, struct sk_buff *skb)
 static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 {
 	struct udp_sock *up = udp_sk(sk);
+	int rc;
 
 	/*
 	 *	Charge it to the socket, dropping if the queue is full.
@@ -1027,7 +1038,10 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 
-	if (sock_queue_rcv_skb(sk,skb)<0) {
+	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
+		/* Note that an ENOMEM error is charged twice */
+		if (rc == -ENOMEM)
+			UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS);
 		UDP_INC_STATS_BH(UDP_MIB_INERRORS);
 		kfree_skb(skb);
 		return -1;
-- 
cgit v1.2.3


From 2942e90050569525628a9f34e0daaa9b661b49cc Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 15 Aug 2006 00:30:25 -0700
Subject: [RTNETLINK]: Use rtnl_unicast() for rtnetlink unicasts

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h |  1 +
 net/core/rtnetlink.c      | 10 +++++++---
 net/decnet/dn_route.c     |  4 +---
 net/ipv4/ipmr.c           |  7 ++++---
 net/ipv4/route.c          |  7 +++----
 net/ipv6/addrconf.c       |  4 +---
 net/ipv6/route.c          |  4 +---
 net/sched/act_api.c       |  7 ++-----
 8 files changed, 20 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 7e4aa48680a7..0e4f478e2cb5 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -584,6 +584,7 @@ struct rtnetlink_link
 
 extern struct rtnetlink_link * rtnetlink_links[NPROTO];
 extern int rtnetlink_send(struct sk_buff *skb, u32 pid, u32 group, int echo);
+extern int rtnl_unicast(struct sk_buff *skb, u32 pid);
 extern int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics);
 
 extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index a1b783a6afc6..e02fa6a33f42 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -166,6 +166,11 @@ int rtnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo)
 	return err;
 }
 
+int rtnl_unicast(struct sk_buff *skb, u32 pid)
+{
+	return nlmsg_unicast(rtnl, skb, pid);
+}
+
 int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
 {
 	struct rtattr *mx = (struct rtattr*)skb->tail;
@@ -574,9 +579,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 		goto errout;
 	}
 
-	err = netlink_unicast(rtnl, skb, NETLINK_CB(skb).pid, MSG_DONTWAIT);
-	if (err > 0)
-		err = 0;
+	err = rtnl_unicast(skb, NETLINK_CB(skb).pid);
 errout:
 	kfree(iw_buf);
 	dev_put(dev);
@@ -825,3 +828,4 @@ EXPORT_SYMBOL(rtnl);
 EXPORT_SYMBOL(rtnl_lock);
 EXPORT_SYMBOL(rtnl_trylock);
 EXPORT_SYMBOL(rtnl_unlock);
+EXPORT_SYMBOL(rtnl_unicast);
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 4c963213fba5..c5daf3557c1f 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -1611,9 +1611,7 @@ int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg)
 		goto out_free;
 	}
 
-	err = netlink_unicast(rtnl, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT);
-
-	return err;
+	return rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
 
 out_free:
 	kfree_skb(skb);
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 85893eef6b16..98f0aa0d4216 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -312,7 +312,8 @@ static void ipmr_destroy_unres(struct mfc_cache *c)
 			e = NLMSG_DATA(nlh);
 			e->error = -ETIMEDOUT;
 			memset(&e->msg, 0, sizeof(e->msg));
-			netlink_unicast(rtnl, skb, NETLINK_CB(skb).dst_pid, MSG_DONTWAIT);
+
+			rtnl_unicast(skb, NETLINK_CB(skb).pid);
 		} else
 			kfree_skb(skb);
 	}
@@ -512,7 +513,6 @@ static void ipmr_cache_resolve(struct mfc_cache *uc, struct mfc_cache *c)
 
 	while((skb=__skb_dequeue(&uc->mfc_un.unres.unresolved))) {
 		if (skb->nh.iph->version == 0) {
-			int err;
 			struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct iphdr));
 
 			if (ipmr_fill_mroute(skb, c, NLMSG_DATA(nlh)) > 0) {
@@ -525,7 +525,8 @@ static void ipmr_cache_resolve(struct mfc_cache *uc, struct mfc_cache *c)
 				e->error = -EMSGSIZE;
 				memset(&e->msg, 0, sizeof(e->msg));
 			}
-			err = netlink_unicast(rtnl, skb, NETLINK_CB(skb).dst_pid, MSG_DONTWAIT);
+
+			rtnl_unicast(skb, NETLINK_CB(skb).pid);
 		} else
 			ip_mr_forward(skb, c, 0);
 	}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 12128b82c9dc..b8f6cadc5b3a 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2809,10 +2809,9 @@ int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 		goto out_free;
 	}
 
-	err = netlink_unicast(rtnl, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT);
-	if (err > 0)
-		err = 0;
-out:	return err;
+	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
+out:
+	return err;
 
 out_free:
 	kfree_skb(skb);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 9ba1e811ba50..4f991a2234d0 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3268,9 +3268,7 @@ static int inet6_rtm_getaddr(struct sk_buff *in_skb,
 		goto out_free;
 	}
 
-	err = netlink_unicast(rtnl, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT);
-	if (err > 0)
-		err = 0;
+	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
 out:
 	in6_ifa_put(ifa);
 	return err;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 9ce28277f47f..024c8e26c2ec 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2044,9 +2044,7 @@ int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 		goto out_free;
 	}
 
-	err = netlink_unicast(rtnl, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT);
-	if (err > 0)
-		err = 0;
+	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
 out:
 	return err;
 out_free:
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index a2587b52e531..6990747d6d5a 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -459,7 +459,6 @@ static int
 act_get_notify(u32 pid, struct nlmsghdr *n, struct tc_action *a, int event)
 {
 	struct sk_buff *skb;
-	int err = 0;
 
 	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (!skb)
@@ -468,10 +467,8 @@ act_get_notify(u32 pid, struct nlmsghdr *n, struct tc_action *a, int event)
 		kfree_skb(skb);
 		return -EINVAL;
 	}
-	err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
-	if (err > 0)
-		err = 0;
-	return err;
+
+	return rtnl_unicast(skb, pid);
 }
 
 static struct tc_action *
-- 
cgit v1.2.3


From d387f6ad10764fc2174373b4a1cca443adee36e3 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 15 Aug 2006 00:31:06 -0700
Subject: [NETLINK]: Add notification message sending interface

Adds nlmsg_notify() implementing proper notification logic. The
message is multicasted to all listeners in the group. The
applications the requests orignates from can request a unicast
back report in which case said socket will be excluded from the
multicast to avoid duplicated notifications.

nlmsg_multicast() is extended to take allocation flags to
allow notification in atomic contexts.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/genetlink.h      |  5 +++--
 include/net/netlink.h        |  6 ++++--
 net/netlabel/netlabel_user.c |  2 +-
 net/netlink/af_netlink.c     | 34 +++++++++++++++++++++++++++++++++-
 net/netlink/genetlink.c      |  2 +-
 5 files changed, 42 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/genetlink.h b/include/net/genetlink.h
index 8c2287264266..97d6d3aba9d2 100644
--- a/include/net/genetlink.h
+++ b/include/net/genetlink.h
@@ -133,11 +133,12 @@ static inline int genlmsg_cancel(struct sk_buff *skb, void *hdr)
  * @skb: netlink message as socket buffer
  * @pid: own netlink pid to avoid sending to yourself
  * @group: multicast group id
+ * @flags: allocation flags
  */
 static inline int genlmsg_multicast(struct sk_buff *skb, u32 pid,
-				    unsigned int group)
+				    unsigned int group, gfp_t flags)
 {
-	return nlmsg_multicast(genl_sock, skb, pid, group);
+	return nlmsg_multicast(genl_sock, skb, pid, group, flags);
 }
 
 /**
diff --git a/include/net/netlink.h b/include/net/netlink.h
index 3a5e40b1e045..b154b81d9a7a 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -43,6 +43,7 @@
  * Message Sending:
  *   nlmsg_multicast()			multicast message to several groups
  *   nlmsg_unicast()			unicast a message to a single socket
+ *   nlmsg_notify()			send notification message
  *
  * Message Length Calculations:
  *   nlmsg_msg_size(payload)		length of message w/o padding
@@ -545,15 +546,16 @@ static inline void nlmsg_free(struct sk_buff *skb)
  * @skb: netlink message as socket buffer
  * @pid: own netlink pid to avoid sending to yourself
  * @group: multicast group id
+ * @flags: allocation flags
  */
 static inline int nlmsg_multicast(struct sock *sk, struct sk_buff *skb,
-				  u32 pid, unsigned int group)
+				  u32 pid, unsigned int group, gfp_t flags)
 {
 	int err;
 
 	NETLINK_CB(skb).dst_group = group;
 
-	err = netlink_broadcast(sk, skb, pid, group, GFP_KERNEL);
+	err = netlink_broadcast(sk, skb, pid, group, flags);
 	if (err > 0)
 		err = 0;
 
diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c
index 80022221b0a7..73cbe66e42ff 100644
--- a/net/netlabel/netlabel_user.c
+++ b/net/netlabel/netlabel_user.c
@@ -154,5 +154,5 @@ int netlbl_netlink_snd(struct sk_buff *skb, u32 pid)
  */
 int netlbl_netlink_snd_multicast(struct sk_buff *skb, u32 pid, u32 group)
 {
-	return genlmsg_multicast(skb, pid, group);
+	return genlmsg_multicast(skb, pid, group, GFP_KERNEL);
 }
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 0f36ddc0b72d..a80e4456e204 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1549,6 +1549,38 @@ void netlink_queue_skip(struct nlmsghdr *nlh, struct sk_buff *skb)
 	skb_pull(skb, msglen);
 }
 
+/**
+ * nlmsg_notify - send a notification netlink message
+ * @sk: netlink socket to use
+ * @skb: notification message
+ * @pid: destination netlink pid for reports or 0
+ * @group: destination multicast group or 0
+ * @report: 1 to report back, 0 to disable
+ * @flags: allocation flags
+ */
+int nlmsg_notify(struct sock *sk, struct sk_buff *skb, u32 pid,
+		 unsigned int group, int report, gfp_t flags)
+{
+	int err = 0;
+
+	if (group) {
+		int exclude_pid = 0;
+
+		if (report) {
+			atomic_inc(&skb->users);
+			exclude_pid = pid;
+		}
+
+		/* errors reported via destination sk->sk_err */
+		nlmsg_multicast(sk, skb, exclude_pid, group, flags);
+	}
+
+	if (report)
+		err = nlmsg_unicast(sk, skb, pid);
+
+	return err;
+}
+
 #ifdef CONFIG_PROC_FS
 struct nl_seq_iter {
 	int link;
@@ -1802,4 +1834,4 @@ EXPORT_SYMBOL(netlink_set_err);
 EXPORT_SYMBOL(netlink_set_nonroot);
 EXPORT_SYMBOL(netlink_unicast);
 EXPORT_SYMBOL(netlink_unregister_notifier);
-
+EXPORT_SYMBOL(nlmsg_notify);
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index 75bb47a898dd..d32599116c56 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -510,7 +510,7 @@ static int genl_ctrl_event(int event, void *data)
 		if (IS_ERR(msg))
 			return PTR_ERR(msg);
 
-		genlmsg_multicast(msg, 0, GENL_ID_CTRL);
+		genlmsg_multicast(msg, 0, GENL_ID_CTRL, GFP_KERNEL);
 		break;
 	}
 
-- 
cgit v1.2.3


From 97676b6b5538b3e059d33b8338e7d5cc41c5f1f1 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 15 Aug 2006 00:31:41 -0700
Subject: [RTNETLINK]: Add rtnetlink notification interface

Adds rtnl_notify() to send rtnetlink notification messages and
rtnl_set_sk_err() to report notification errors as socket
errors in order to indicate the need of a resync due to loss
of events.

nlmsg_report() is added to properly document the meaning of
NLM_F_ECHO.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h |  3 +++
 include/net/netlink.h     | 17 +++++++++++++++++
 net/core/rtnetlink.c      | 18 ++++++++++++++++++
 3 files changed, 38 insertions(+)

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 0e4f478e2cb5..ecbe0349060c 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -585,6 +585,9 @@ struct rtnetlink_link
 extern struct rtnetlink_link * rtnetlink_links[NPROTO];
 extern int rtnetlink_send(struct sk_buff *skb, u32 pid, u32 group, int echo);
 extern int rtnl_unicast(struct sk_buff *skb, u32 pid);
+extern int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group,
+		       struct nlmsghdr *nlh, gfp_t flags);
+extern void rtnl_set_sk_err(u32 group, int error);
 extern int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics);
 
 extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data);
diff --git a/include/net/netlink.h b/include/net/netlink.h
index b154b81d9a7a..bf593eb59e1b 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -65,6 +65,9 @@
  *   nlmsg_validate()			validate netlink message incl. attrs
  *   nlmsg_for_each_attr()		loop over all attributes
  *
+ * Misc:
+ *   nlmsg_report()			report back to application?
+ *
  * ------------------------------------------------------------------------
  *                          Attributes Interface
  * ------------------------------------------------------------------------
@@ -194,6 +197,9 @@ extern void		netlink_run_queue(struct sock *sk, unsigned int *qlen,
 						    struct nlmsghdr *, int *));
 extern void		netlink_queue_skip(struct nlmsghdr *nlh,
 					   struct sk_buff *skb);
+extern int		nlmsg_notify(struct sock *sk, struct sk_buff *skb,
+				     u32 pid, unsigned int group, int report,
+				     gfp_t flags);
 
 extern int		nla_validate(struct nlattr *head, int len, int maxtype,
 				     struct nla_policy *policy);
@@ -375,6 +381,17 @@ static inline int nlmsg_validate(struct nlmsghdr *nlh, int hdrlen, int maxtype,
 			    nlmsg_attrlen(nlh, hdrlen), maxtype, policy);
 }
 
+/**
+ * nlmsg_report - need to report back to application?
+ * @nlh: netlink message header
+ *
+ * Returns 1 if a report back to the application is requested.
+ */
+static inline int nlmsg_report(struct nlmsghdr *nlh)
+{
+	return !!(nlh->nlmsg_flags & NLM_F_ECHO);
+}
+
 /**
  * nlmsg_for_each_attr - iterate over a stream of attributes
  * @pos: loop counter, set to current attribute
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index e02fa6a33f42..2b1af17e6389 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -171,6 +171,22 @@ int rtnl_unicast(struct sk_buff *skb, u32 pid)
 	return nlmsg_unicast(rtnl, skb, pid);
 }
 
+int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group,
+		struct nlmsghdr *nlh, gfp_t flags)
+{
+	int report = 0;
+
+	if (nlh)
+		report = nlmsg_report(nlh);
+
+	return nlmsg_notify(rtnl, skb, pid, group, report, flags);
+}
+
+void rtnl_set_sk_err(u32 group, int error)
+{
+	netlink_set_err(rtnl, 0, group, error);
+}
+
 int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
 {
 	struct rtattr *mx = (struct rtattr*)skb->tail;
@@ -829,3 +845,5 @@ EXPORT_SYMBOL(rtnl_lock);
 EXPORT_SYMBOL(rtnl_trylock);
 EXPORT_SYMBOL(rtnl_unlock);
 EXPORT_SYMBOL(rtnl_unicast);
+EXPORT_SYMBOL(rtnl_notify);
+EXPORT_SYMBOL(rtnl_set_sk_err);
-- 
cgit v1.2.3


From 56fc85ac961e2c20dcb5ef07e2628b3f93de2e49 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 15 Aug 2006 00:37:29 -0700
Subject: [RTNETLINK]: Unexport rtnl socket

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h | 2 --
 net/core/rtnetlink.c      | 4 +---
 2 files changed, 1 insertion(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index ecbe0349060c..9c92dc8b9a08 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -574,8 +574,6 @@ extern int rtattr_parse(struct rtattr *tb[], int maxattr, struct rtattr *rta, in
 #define rtattr_parse_nested(tb, max, rta) \
 	rtattr_parse((tb), (max), RTA_DATA((rta)), RTA_PAYLOAD((rta)))
 
-extern struct sock *rtnl;
-
 struct rtnetlink_link
 {
 	int (*doit)(struct sk_buff *, struct nlmsghdr*, void *attr);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index f5300b5dd0fd..dfc58269240a 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -58,6 +58,7 @@
 #endif	/* CONFIG_NET_WIRELESS_RTNETLINK */
 
 static DEFINE_MUTEX(rtnl_mutex);
+static struct sock *rtnl;
 
 void rtnl_lock(void)
 {
@@ -95,8 +96,6 @@ int rtattr_parse(struct rtattr *tb[], int maxattr, struct rtattr *rta, int len)
 	return 0;
 }
 
-struct sock *rtnl;
-
 struct rtnetlink_link * rtnetlink_links[NPROTO];
 
 static const int rtm_min[RTM_NR_FAMILIES] =
@@ -842,7 +841,6 @@ EXPORT_SYMBOL(rtattr_strlcpy);
 EXPORT_SYMBOL(rtattr_parse);
 EXPORT_SYMBOL(rtnetlink_links);
 EXPORT_SYMBOL(rtnetlink_put_metrics);
-EXPORT_SYMBOL(rtnl);
 EXPORT_SYMBOL(rtnl_lock);
 EXPORT_SYMBOL(rtnl_trylock);
 EXPORT_SYMBOL(rtnl_unlock);
-- 
cgit v1.2.3


From 4e902c57417c4c285b98ba2722468d1c3ed83d1b Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 17 Aug 2006 18:14:52 -0700
Subject: [IPv4]: FIB configuration using struct fib_config

Introduces struct fib_config replacing the ugly struct kern_rta
prone to ordering issues. Avoids creating faked netlink messages
for auto generated routes or requests via ioctl.

A new interface net/nexthop.h is added to help navigate through
nexthop configuration arrays.

A new struct nl_info will be used to carry the necessary netlink
information to be used for notifications later on.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     |  55 +++----
 include/net/netlink.h    |  10 ++
 include/net/nexthop.h    |  33 ++++
 net/ipv4/fib_frontend.c  | 364 +++++++++++++++++++++++++++++++++++---------
 net/ipv4/fib_hash.c      |  94 ++++++------
 net/ipv4/fib_lookup.h    |  11 +-
 net/ipv4/fib_semantics.c | 385 ++++++++++++++---------------------------------
 net/ipv4/fib_trie.c      |  76 ++++------
 8 files changed, 560 insertions(+), 468 deletions(-)
 create mode 100644 include/net/nexthop.h

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 8e9ba563d342..42ed96fab3f5 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -20,25 +20,32 @@
 #include <linux/seq_file.h>
 #include <net/fib_rules.h>
 
-/* WARNING: The ordering of these elements must match ordering
- *          of RTA_* rtnetlink attribute numbers.
- */
-struct kern_rta {
-	void		*rta_dst;
-	void		*rta_src;
-	int		*rta_iif;
-	int		*rta_oif;
-	void		*rta_gw;
-	u32		*rta_priority;
-	void		*rta_prefsrc;
-	struct rtattr	*rta_mx;
-	struct rtattr	*rta_mp;
-	unsigned char	*rta_protoinfo;
-	u32		*rta_flow;
-	struct rta_cacheinfo *rta_ci;
-	struct rta_session *rta_sess;
-	u32		*rta_mp_alg;
-};
+struct fib_config {
+	u8			fc_family;
+	u8			fc_dst_len;
+	u8			fc_src_len;
+	u8			fc_tos;
+	u8			fc_protocol;
+	u8			fc_scope;
+	u8			fc_type;
+	/* 1 byte unused */
+	u32			fc_table;
+	u32			fc_dst;
+	u32			fc_src;
+	u32			fc_gw;
+	int			fc_oif;
+	u32			fc_flags;
+	u32			fc_priority;
+	u32			fc_prefsrc;
+	struct nlattr		*fc_mx;
+	struct rtnexthop	*fc_mp;
+	int			fc_mx_len;
+	int			fc_mp_len;
+	u32			fc_flow;
+	u32			fc_mp_alg;
+	u32			fc_nlflags;
+	struct nl_info		fc_nlinfo;
+ };
 
 struct fib_info;
 
@@ -154,12 +161,8 @@ struct fib_table {
 	u32		tb_id;
 	unsigned	tb_stamp;
 	int		(*tb_lookup)(struct fib_table *tb, const struct flowi *flp, struct fib_result *res);
-	int		(*tb_insert)(struct fib_table *table, struct rtmsg *r,
-				     struct kern_rta *rta, struct nlmsghdr *n,
-				     struct netlink_skb_parms *req);
-	int		(*tb_delete)(struct fib_table *table, struct rtmsg *r,
-				     struct kern_rta *rta, struct nlmsghdr *n,
-				     struct netlink_skb_parms *req);
+	int		(*tb_insert)(struct fib_table *, struct fib_config *);
+	int		(*tb_delete)(struct fib_table *, struct fib_config *);
 	int		(*tb_dump)(struct fib_table *table, struct sk_buff *skb,
 				     struct netlink_callback *cb);
 	int		(*tb_flush)(struct fib_table *table);
@@ -228,8 +231,6 @@ struct rtentry;
 extern int ip_fib_check_default(u32 gw, struct net_device *dev);
 extern int fib_sync_down(u32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
-extern int fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm,
-			       struct kern_rta *rta, struct rtentry *r);
 extern u32  __fib_res_prefsrc(struct fib_result *res);
 
 /* Exported by fib_hash.c */
diff --git a/include/net/netlink.h b/include/net/netlink.h
index bf593eb59e1b..47044da167c5 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -192,6 +192,16 @@ struct nla_policy {
 	u16		minlen;
 };
 
+/**
+ * struct nl_info - netlink source information
+ * @nlh: Netlink message header of original request
+ * @pid: Netlink PID of requesting application
+ */
+struct nl_info {
+	struct nlmsghdr		*nlh;
+	u32			pid;
+};
+
 extern void		netlink_run_queue(struct sock *sk, unsigned int *qlen,
 					  int (*cb)(struct sk_buff *,
 						    struct nlmsghdr *, int *));
diff --git a/include/net/nexthop.h b/include/net/nexthop.h
new file mode 100644
index 000000000000..3334dbfa5aa4
--- /dev/null
+++ b/include/net/nexthop.h
@@ -0,0 +1,33 @@
+#ifndef __NET_NEXTHOP_H
+#define __NET_NEXTHOP_H
+
+#include <linux/rtnetlink.h>
+#include <net/netlink.h>
+
+static inline int rtnh_ok(const struct rtnexthop *rtnh, int remaining)
+{
+	return remaining >= sizeof(*rtnh) &&
+	       rtnh->rtnh_len >= sizeof(*rtnh) &&
+	       rtnh->rtnh_len <= remaining;
+}
+
+static inline struct rtnexthop *rtnh_next(const struct rtnexthop *rtnh,
+                                         int *remaining)
+{
+	int totlen = NLA_ALIGN(rtnh->rtnh_len);
+
+	*remaining -= totlen;
+	return (struct rtnexthop *) ((char *) rtnh + totlen);
+}
+
+static inline struct nlattr *rtnh_attrs(const struct rtnexthop *rtnh)
+{
+	return (struct nlattr *) ((char *) rtnh + NLA_ALIGN(sizeof(*rtnh)));
+}
+
+static inline int rtnh_attrlen(const struct rtnexthop *rtnh)
+{
+	return rtnh->rtnh_len - NLA_ALIGN(sizeof(*rtnh));
+}
+
+#endif
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index ad4c14f968a1..acc18bdf2dee 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -253,42 +253,190 @@ e_inval:
 
 #ifndef CONFIG_IP_NOSIOCRT
 
+static inline u32 sk_extract_addr(struct sockaddr *addr)
+{
+	return ((struct sockaddr_in *) addr)->sin_addr.s_addr;
+}
+
+static int put_rtax(struct nlattr *mx, int len, int type, u32 value)
+{
+	struct nlattr *nla;
+
+	nla = (struct nlattr *) ((char *) mx + len);
+	nla->nla_type = type;
+	nla->nla_len = nla_attr_size(4);
+	*(u32 *) nla_data(nla) = value;
+
+	return len + nla_total_size(4);
+}
+
+static int rtentry_to_fib_config(int cmd, struct rtentry *rt,
+				 struct fib_config *cfg)
+{
+	u32 addr;
+	int plen;
+
+	memset(cfg, 0, sizeof(*cfg));
+
+	if (rt->rt_dst.sa_family != AF_INET)
+		return -EAFNOSUPPORT;
+
+	/*
+	 * Check mask for validity:
+	 * a) it must be contiguous.
+	 * b) destination must have all host bits clear.
+	 * c) if application forgot to set correct family (AF_INET),
+	 *    reject request unless it is absolutely clear i.e.
+	 *    both family and mask are zero.
+	 */
+	plen = 32;
+	addr = sk_extract_addr(&rt->rt_dst);
+	if (!(rt->rt_flags & RTF_HOST)) {
+		u32 mask = sk_extract_addr(&rt->rt_genmask);
+
+		if (rt->rt_genmask.sa_family != AF_INET) {
+			if (mask || rt->rt_genmask.sa_family)
+				return -EAFNOSUPPORT;
+		}
+
+		if (bad_mask(mask, addr))
+			return -EINVAL;
+
+		plen = inet_mask_len(mask);
+	}
+
+	cfg->fc_dst_len = plen;
+	cfg->fc_dst = addr;
+
+	if (cmd != SIOCDELRT) {
+		cfg->fc_nlflags = NLM_F_CREATE;
+		cfg->fc_protocol = RTPROT_BOOT;
+	}
+
+	if (rt->rt_metric)
+		cfg->fc_priority = rt->rt_metric - 1;
+
+	if (rt->rt_flags & RTF_REJECT) {
+		cfg->fc_scope = RT_SCOPE_HOST;
+		cfg->fc_type = RTN_UNREACHABLE;
+		return 0;
+	}
+
+	cfg->fc_scope = RT_SCOPE_NOWHERE;
+	cfg->fc_type = RTN_UNICAST;
+
+	if (rt->rt_dev) {
+		char *colon;
+		struct net_device *dev;
+		char devname[IFNAMSIZ];
+
+		if (copy_from_user(devname, rt->rt_dev, IFNAMSIZ-1))
+			return -EFAULT;
+
+		devname[IFNAMSIZ-1] = 0;
+		colon = strchr(devname, ':');
+		if (colon)
+			*colon = 0;
+		dev = __dev_get_by_name(devname);
+		if (!dev)
+			return -ENODEV;
+		cfg->fc_oif = dev->ifindex;
+		if (colon) {
+			struct in_ifaddr *ifa;
+			struct in_device *in_dev = __in_dev_get_rtnl(dev);
+			if (!in_dev)
+				return -ENODEV;
+			*colon = ':';
+			for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next)
+				if (strcmp(ifa->ifa_label, devname) == 0)
+					break;
+			if (ifa == NULL)
+				return -ENODEV;
+			cfg->fc_prefsrc = ifa->ifa_local;
+		}
+	}
+
+	addr = sk_extract_addr(&rt->rt_gateway);
+	if (rt->rt_gateway.sa_family == AF_INET && addr) {
+		cfg->fc_gw = addr;
+		if (rt->rt_flags & RTF_GATEWAY &&
+		    inet_addr_type(addr) == RTN_UNICAST)
+			cfg->fc_scope = RT_SCOPE_UNIVERSE;
+	}
+
+	if (cmd == SIOCDELRT)
+		return 0;
+
+	if (rt->rt_flags & RTF_GATEWAY && !cfg->fc_gw)
+		return -EINVAL;
+
+	if (cfg->fc_scope == RT_SCOPE_NOWHERE)
+		cfg->fc_scope = RT_SCOPE_LINK;
+
+	if (rt->rt_flags & (RTF_MTU | RTF_WINDOW | RTF_IRTT)) {
+		struct nlattr *mx;
+		int len = 0;
+
+		mx = kzalloc(3 * nla_total_size(4), GFP_KERNEL);
+ 		if (mx == NULL)
+			return -ENOMEM;
+
+		if (rt->rt_flags & RTF_MTU)
+			len = put_rtax(mx, len, RTAX_ADVMSS, rt->rt_mtu - 40);
+
+		if (rt->rt_flags & RTF_WINDOW)
+			len = put_rtax(mx, len, RTAX_WINDOW, rt->rt_window);
+
+		if (rt->rt_flags & RTF_IRTT)
+			len = put_rtax(mx, len, RTAX_RTT, rt->rt_irtt << 3);
+
+		cfg->fc_mx = mx;
+		cfg->fc_mx_len = len;
+	}
+
+	return 0;
+}
+
 /*
  *	Handle IP routing ioctl calls. These are used to manipulate the routing tables
  */
  
 int ip_rt_ioctl(unsigned int cmd, void __user *arg)
 {
+	struct fib_config cfg;
+	struct rtentry rt;
 	int err;
-	struct kern_rta rta;
-	struct rtentry  r;
-	struct {
-		struct nlmsghdr nlh;
-		struct rtmsg	rtm;
-	} req;
 
 	switch (cmd) {
 	case SIOCADDRT:		/* Add a route */
 	case SIOCDELRT:		/* Delete a route */
 		if (!capable(CAP_NET_ADMIN))
 			return -EPERM;
-		if (copy_from_user(&r, arg, sizeof(struct rtentry)))
+
+		if (copy_from_user(&rt, arg, sizeof(rt)))
 			return -EFAULT;
+
 		rtnl_lock();
-		err = fib_convert_rtentry(cmd, &req.nlh, &req.rtm, &rta, &r);
+		err = rtentry_to_fib_config(cmd, &rt, &cfg);
 		if (err == 0) {
+			struct fib_table *tb;
+
 			if (cmd == SIOCDELRT) {
-				struct fib_table *tb = fib_get_table(req.rtm.rtm_table);
-				err = -ESRCH;
+				tb = fib_get_table(cfg.fc_table);
 				if (tb)
-					err = tb->tb_delete(tb, &req.rtm, &rta, &req.nlh, NULL);
+					err = tb->tb_delete(tb, &cfg);
+				else
+					err = -ESRCH;
 			} else {
-				struct fib_table *tb = fib_new_table(req.rtm.rtm_table);
-				err = -ENOBUFS;
+				tb = fib_new_table(cfg.fc_table);
 				if (tb)
-					err = tb->tb_insert(tb, &req.rtm, &rta, &req.nlh, NULL);
+					err = tb->tb_insert(tb, &cfg);
+				else
+					err = -ENOBUFS;
 			}
-			kfree(rta.rta_mx);
+
+			/* allocated by rtentry_to_fib_config() */
+			kfree(cfg.fc_mx);
 		}
 		rtnl_unlock();
 		return err;
@@ -305,51 +453,134 @@ int ip_rt_ioctl(unsigned int cmd, void *arg)
 
 #endif
 
-static int inet_check_attr(struct rtmsg *r, struct rtattr **rta)
+static struct nla_policy rtm_ipv4_policy[RTA_MAX+1] __read_mostly = {
+	[RTA_DST]		= { .type = NLA_U32 },
+	[RTA_SRC]		= { .type = NLA_U32 },
+	[RTA_IIF]		= { .type = NLA_U32 },
+	[RTA_OIF]		= { .type = NLA_U32 },
+	[RTA_GATEWAY]		= { .type = NLA_U32 },
+	[RTA_PRIORITY]		= { .type = NLA_U32 },
+	[RTA_PREFSRC]		= { .type = NLA_U32 },
+	[RTA_METRICS]		= { .type = NLA_NESTED },
+	[RTA_MULTIPATH]		= { .minlen = sizeof(struct rtnexthop) },
+	[RTA_PROTOINFO]		= { .type = NLA_U32 },
+	[RTA_FLOW]		= { .type = NLA_U32 },
+	[RTA_MP_ALGO]		= { .type = NLA_U32 },
+};
+
+static int rtm_to_fib_config(struct sk_buff *skb, struct nlmsghdr *nlh,
+			     struct fib_config *cfg)
 {
-	int i;
-
-	for (i=1; i<=RTA_MAX; i++, rta++) {
-		struct rtattr *attr = *rta;
-		if (attr) {
-			if (RTA_PAYLOAD(attr) < 4)
-				return -EINVAL;
-			if (i != RTA_MULTIPATH && i != RTA_METRICS &&
-			    i != RTA_TABLE)
-				*rta = (struct rtattr*)RTA_DATA(attr);
+	struct nlattr *attr;
+	int err, remaining;
+	struct rtmsg *rtm;
+
+	err = nlmsg_validate(nlh, sizeof(*rtm), RTA_MAX, rtm_ipv4_policy);
+	if (err < 0)
+		goto errout;
+
+	memset(cfg, 0, sizeof(*cfg));
+
+	rtm = nlmsg_data(nlh);
+	cfg->fc_family = rtm->rtm_family;
+	cfg->fc_dst_len = rtm->rtm_dst_len;
+	cfg->fc_src_len = rtm->rtm_src_len;
+	cfg->fc_tos = rtm->rtm_tos;
+	cfg->fc_table = rtm->rtm_table;
+	cfg->fc_protocol = rtm->rtm_protocol;
+	cfg->fc_scope = rtm->rtm_scope;
+	cfg->fc_type = rtm->rtm_type;
+	cfg->fc_flags = rtm->rtm_flags;
+	cfg->fc_nlflags = nlh->nlmsg_flags;
+
+	cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid;
+	cfg->fc_nlinfo.nlh = nlh;
+
+	nlmsg_for_each_attr(attr, nlh, sizeof(struct rtmsg), remaining) {
+		switch (attr->nla_type) {
+		case RTA_DST:
+			cfg->fc_dst = nla_get_u32(attr);
+			break;
+		case RTA_SRC:
+			cfg->fc_src = nla_get_u32(attr);
+			break;
+		case RTA_OIF:
+			cfg->fc_oif = nla_get_u32(attr);
+			break;
+		case RTA_GATEWAY:
+			cfg->fc_gw = nla_get_u32(attr);
+			break;
+		case RTA_PRIORITY:
+			cfg->fc_priority = nla_get_u32(attr);
+			break;
+		case RTA_PREFSRC:
+			cfg->fc_prefsrc = nla_get_u32(attr);
+			break;
+		case RTA_METRICS:
+			cfg->fc_mx = nla_data(attr);
+			cfg->fc_mx_len = nla_len(attr);
+			break;
+		case RTA_MULTIPATH:
+			cfg->fc_mp = nla_data(attr);
+			cfg->fc_mp_len = nla_len(attr);
+			break;
+		case RTA_FLOW:
+			cfg->fc_flow = nla_get_u32(attr);
+			break;
+		case RTA_MP_ALGO:
+			cfg->fc_mp_alg = nla_get_u32(attr);
+			break;
+		case RTA_TABLE:
+			cfg->fc_table = nla_get_u32(attr);
+			break;
 		}
 	}
+
 	return 0;
+errout:
+	return err;
 }
 
 int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct fib_table * tb;
-	struct rtattr **rta = arg;
-	struct rtmsg *r = NLMSG_DATA(nlh);
+	struct fib_config cfg;
+	struct fib_table *tb;
+	int err;
 
-	if (inet_check_attr(r, rta))
-		return -EINVAL;
+	err = rtm_to_fib_config(skb, nlh, &cfg);
+	if (err < 0)
+		goto errout;
 
-	tb = fib_get_table(rtm_get_table(rta, r->rtm_table));
-	if (tb)
-		return tb->tb_delete(tb, r, (struct kern_rta*)rta, nlh, &NETLINK_CB(skb));
-	return -ESRCH;
+	tb = fib_get_table(cfg.fc_table);
+	if (tb == NULL) {
+		err = -ESRCH;
+		goto errout;
+	}
+
+	err = tb->tb_delete(tb, &cfg);
+errout:
+	return err;
 }
 
 int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct fib_table * tb;
-	struct rtattr **rta = arg;
-	struct rtmsg *r = NLMSG_DATA(nlh);
+	struct fib_config cfg;
+	struct fib_table *tb;
+	int err;
 
-	if (inet_check_attr(r, rta))
-		return -EINVAL;
+	err = rtm_to_fib_config(skb, nlh, &cfg);
+	if (err < 0)
+		goto errout;
 
-	tb = fib_new_table(rtm_get_table(rta, r->rtm_table));
-	if (tb)
-		return tb->tb_insert(tb, r, (struct kern_rta*)rta, nlh, &NETLINK_CB(skb));
-	return -ENOBUFS;
+	tb = fib_new_table(cfg.fc_table);
+	if (tb == NULL) {
+		err = -ENOBUFS;
+		goto errout;
+	}
+
+	err = tb->tb_insert(tb, &cfg);
+errout:
+	return err;
 }
 
 int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
@@ -396,17 +627,19 @@ out:
    only when netlink is already locked.
  */
 
-static void fib_magic(int cmd, int type, u32 dst, int dst_len, struct in_ifaddr *ifa)
+static void fib_magic(int cmd, int type, u32 dst, int dst_len,
+		      struct in_ifaddr *ifa)
 {
-	struct fib_table * tb;
-	struct {
-		struct nlmsghdr	nlh;
-		struct rtmsg	rtm;
-	} req;
-	struct kern_rta rta;
-
-	memset(&req.rtm, 0, sizeof(req.rtm));
-	memset(&rta, 0, sizeof(rta));
+	struct fib_table *tb;
+	struct fib_config cfg = {
+		.fc_protocol = RTPROT_KERNEL,
+		.fc_type = type,
+		.fc_dst = dst,
+		.fc_dst_len = dst_len,
+		.fc_prefsrc = ifa->ifa_local,
+		.fc_oif = ifa->ifa_dev->dev->ifindex,
+		.fc_nlflags = NLM_F_CREATE | NLM_F_APPEND,
+	};
 
 	if (type == RTN_UNICAST)
 		tb = fib_new_table(RT_TABLE_MAIN);
@@ -416,26 +649,17 @@ static void fib_magic(int cmd, int type, u32 dst, int dst_len, struct in_ifaddr
 	if (tb == NULL)
 		return;
 
-	req.nlh.nlmsg_len = sizeof(req);
-	req.nlh.nlmsg_type = cmd;
-	req.nlh.nlmsg_flags = NLM_F_REQUEST|NLM_F_CREATE|NLM_F_APPEND;
-	req.nlh.nlmsg_pid = 0;
-	req.nlh.nlmsg_seq = 0;
+	cfg.fc_table = tb->tb_id;
 
-	req.rtm.rtm_dst_len = dst_len;
-	req.rtm.rtm_table = tb->tb_id;
-	req.rtm.rtm_protocol = RTPROT_KERNEL;
-	req.rtm.rtm_scope = (type != RTN_LOCAL ? RT_SCOPE_LINK : RT_SCOPE_HOST);
-	req.rtm.rtm_type = type;
-
-	rta.rta_dst = &dst;
-	rta.rta_prefsrc = &ifa->ifa_local;
-	rta.rta_oif = &ifa->ifa_dev->dev->ifindex;
+	if (type != RTN_LOCAL)
+		cfg.fc_scope = RT_SCOPE_LINK;
+	else
+		cfg.fc_scope = RT_SCOPE_HOST;
 
 	if (cmd == RTM_NEWROUTE)
-		tb->tb_insert(tb, &req.rtm, &rta, &req.nlh, NULL);
+		tb->tb_insert(tb, &cfg);
 	else
-		tb->tb_delete(tb, &req.rtm, &rta, &req.nlh, NULL);
+		tb->tb_delete(tb, &cfg);
 }
 
 void fib_add_ifaddr(struct in_ifaddr *ifa)
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index b5bee1a71e5c..357557549ce5 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -379,42 +379,39 @@ static struct fib_node *fib_find_node(struct fn_zone *fz, u32 key)
 	return NULL;
 }
 
-static int
-fn_hash_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
-	       struct nlmsghdr *n, struct netlink_skb_parms *req)
+static int fn_hash_insert(struct fib_table *tb, struct fib_config *cfg)
 {
 	struct fn_hash *table = (struct fn_hash *) tb->tb_data;
 	struct fib_node *new_f, *f;
 	struct fib_alias *fa, *new_fa;
 	struct fn_zone *fz;
 	struct fib_info *fi;
-	int z = r->rtm_dst_len;
-	int type = r->rtm_type;
-	u8 tos = r->rtm_tos;
+	u8 tos = cfg->fc_tos;
 	u32 key;
 	int err;
 
-	if (z > 32)
+	if (cfg->fc_dst_len > 32)
 		return -EINVAL;
-	fz = table->fn_zones[z];
-	if (!fz && !(fz = fn_new_zone(table, z)))
+
+	fz = table->fn_zones[cfg->fc_dst_len];
+	if (!fz && !(fz = fn_new_zone(table, cfg->fc_dst_len)))
 		return -ENOBUFS;
 
 	key = 0;
-	if (rta->rta_dst) {
-		u32 dst;
-		memcpy(&dst, rta->rta_dst, 4);
-		if (dst & ~FZ_MASK(fz))
+	if (cfg->fc_dst) {
+		if (cfg->fc_dst & ~FZ_MASK(fz))
 			return -EINVAL;
-		key = fz_key(dst, fz);
+		key = fz_key(cfg->fc_dst, fz);
 	}
 
-	if  ((fi = fib_create_info(r, rta, n, &err)) == NULL)
-		return err;
+	fi = fib_create_info(cfg);
+	if (IS_ERR(fi))
+		return PTR_ERR(fi);
 
 	if (fz->fz_nent > (fz->fz_divisor<<1) &&
 	    fz->fz_divisor < FZ_MAX_DIVISOR &&
-	    (z==32 || (1<<z) > fz->fz_divisor))
+	    (cfg->fc_dst_len == 32 ||
+	     (1 << cfg->fc_dst_len) > fz->fz_divisor))
 		fn_rehash_zone(fz);
 
 	f = fib_find_node(fz, key);
@@ -440,18 +437,18 @@ fn_hash_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 		struct fib_alias *fa_orig;
 
 		err = -EEXIST;
-		if (n->nlmsg_flags & NLM_F_EXCL)
+		if (cfg->fc_nlflags & NLM_F_EXCL)
 			goto out;
 
-		if (n->nlmsg_flags & NLM_F_REPLACE) {
+		if (cfg->fc_nlflags & NLM_F_REPLACE) {
 			struct fib_info *fi_drop;
 			u8 state;
 
 			write_lock_bh(&fib_hash_lock);
 			fi_drop = fa->fa_info;
 			fa->fa_info = fi;
-			fa->fa_type = type;
-			fa->fa_scope = r->rtm_scope;
+			fa->fa_type = cfg->fc_type;
+			fa->fa_scope = cfg->fc_scope;
 			state = fa->fa_state;
 			fa->fa_state &= ~FA_S_ACCESSED;
 			fib_hash_genid++;
@@ -474,17 +471,17 @@ fn_hash_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 				break;
 			if (fa->fa_info->fib_priority != fi->fib_priority)
 				break;
-			if (fa->fa_type == type &&
-			    fa->fa_scope == r->rtm_scope &&
+			if (fa->fa_type == cfg->fc_type &&
+			    fa->fa_scope == cfg->fc_scope &&
 			    fa->fa_info == fi)
 				goto out;
 		}
-		if (!(n->nlmsg_flags & NLM_F_APPEND))
+		if (!(cfg->fc_nlflags & NLM_F_APPEND))
 			fa = fa_orig;
 	}
 
 	err = -ENOENT;
-	if (!(n->nlmsg_flags&NLM_F_CREATE))
+	if (!(cfg->fc_nlflags & NLM_F_CREATE))
 		goto out;
 
 	err = -ENOBUFS;
@@ -506,8 +503,8 @@ fn_hash_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 
 	new_fa->fa_info = fi;
 	new_fa->fa_tos = tos;
-	new_fa->fa_type = type;
-	new_fa->fa_scope = r->rtm_scope;
+	new_fa->fa_type = cfg->fc_type;
+	new_fa->fa_scope = cfg->fc_scope;
 	new_fa->fa_state = 0;
 
 	/*
@@ -526,7 +523,8 @@ fn_hash_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 		fz->fz_nent++;
 	rt_cache_flush(-1);
 
-	rtmsg_fib(RTM_NEWROUTE, key, new_fa, z, tb->tb_id, n, req);
+	rtmsg_fib(RTM_NEWROUTE, key, new_fa, cfg->fc_dst_len, tb->tb_id,
+		  &cfg->fc_nlinfo);
 	return 0;
 
 out_free_new_fa:
@@ -537,30 +535,25 @@ out:
 }
 
 
-static int
-fn_hash_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
-	       struct nlmsghdr *n, struct netlink_skb_parms *req)
+static int fn_hash_delete(struct fib_table *tb, struct fib_config *cfg)
 {
 	struct fn_hash *table = (struct fn_hash*)tb->tb_data;
 	struct fib_node *f;
 	struct fib_alias *fa, *fa_to_delete;
-	int z = r->rtm_dst_len;
 	struct fn_zone *fz;
 	u32 key;
-	u8 tos = r->rtm_tos;
 
-	if (z > 32)
+	if (cfg->fc_dst_len > 32)
 		return -EINVAL;
-	if ((fz  = table->fn_zones[z]) == NULL)
+
+	if ((fz  = table->fn_zones[cfg->fc_dst_len]) == NULL)
 		return -ESRCH;
 
 	key = 0;
-	if (rta->rta_dst) {
-		u32 dst;
-		memcpy(&dst, rta->rta_dst, 4);
-		if (dst & ~FZ_MASK(fz))
+	if (cfg->fc_dst) {
+		if (cfg->fc_dst & ~FZ_MASK(fz))
 			return -EINVAL;
-		key = fz_key(dst, fz);
+		key = fz_key(cfg->fc_dst, fz);
 	}
 
 	f = fib_find_node(fz, key);
@@ -568,7 +561,7 @@ fn_hash_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 	if (!f)
 		fa = NULL;
 	else
-		fa = fib_find_alias(&f->fn_alias, tos, 0);
+		fa = fib_find_alias(&f->fn_alias, cfg->fc_tos, 0);
 	if (!fa)
 		return -ESRCH;
 
@@ -577,16 +570,16 @@ fn_hash_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 	list_for_each_entry_continue(fa, &f->fn_alias, fa_list) {
 		struct fib_info *fi = fa->fa_info;
 
-		if (fa->fa_tos != tos)
+		if (fa->fa_tos != cfg->fc_tos)
 			break;
 
-		if ((!r->rtm_type ||
-		     fa->fa_type == r->rtm_type) &&
-		    (r->rtm_scope == RT_SCOPE_NOWHERE ||
-		     fa->fa_scope == r->rtm_scope) &&
-		    (!r->rtm_protocol ||
-		     fi->fib_protocol == r->rtm_protocol) &&
-		    fib_nh_match(r, n, rta, fi) == 0) {
+		if ((!cfg->fc_type ||
+		     fa->fa_type == cfg->fc_type) &&
+		    (cfg->fc_scope == RT_SCOPE_NOWHERE ||
+		     fa->fa_scope == cfg->fc_scope) &&
+		    (!cfg->fc_protocol ||
+		     fi->fib_protocol == cfg->fc_protocol) &&
+		    fib_nh_match(cfg, fi) == 0) {
 			fa_to_delete = fa;
 			break;
 		}
@@ -596,7 +589,8 @@ fn_hash_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 		int kill_fn;
 
 		fa = fa_to_delete;
-		rtmsg_fib(RTM_DELROUTE, key, fa, z, tb->tb_id, n, req);
+		rtmsg_fib(RTM_DELROUTE, key, fa, cfg->fc_dst_len,
+			  tb->tb_id, &cfg->fc_nlinfo);
 
 		kill_fn = 0;
 		write_lock_bh(&fib_hash_lock);
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index ddd52496b451..d6d1a89e4003 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -23,19 +23,14 @@ extern int fib_semantic_match(struct list_head *head,
 			      struct fib_result *res, __u32 zone, __u32 mask,
 				int prefixlen);
 extern void fib_release_info(struct fib_info *);
-extern struct fib_info *fib_create_info(const struct rtmsg *r,
-					struct kern_rta *rta,
-					const struct nlmsghdr *,
-					int *err);
-extern int fib_nh_match(struct rtmsg *r, struct nlmsghdr *,
-			struct kern_rta *rta, struct fib_info *fi);
+extern struct fib_info *fib_create_info(struct fib_config *cfg);
+extern int fib_nh_match(struct fib_config *cfg, struct fib_info *fi);
 extern int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
 			 u32 tb_id, u8 type, u8 scope, void *dst,
 			 int dst_len, u8 tos, struct fib_info *fi,
 			 unsigned int);
 extern void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
-		      int z, u32 tb_id,
-		      struct nlmsghdr *n, struct netlink_skb_parms *req);
+		      int dst_len, u32 tb_id, struct nl_info *info);
 extern struct fib_alias *fib_find_alias(struct list_head *fah,
 					u8 tos, u32 prio);
 extern int fib_detect_death(struct fib_info *fi, int order,
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 5dfdad5cbcd4..340f9db389e5 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -44,6 +44,7 @@
 #include <net/ip_fib.h>
 #include <net/ip_mp_alg.h>
 #include <net/netlink.h>
+#include <net/nexthop.h>
 
 #include "fib_lookup.h"
 
@@ -273,27 +274,27 @@ int ip_fib_check_default(u32 gw, struct net_device *dev)
 }
 
 void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
-	       int z, u32 tb_id,
-	       struct nlmsghdr *n, struct netlink_skb_parms *req)
+	       int dst_len, u32 tb_id, struct nl_info *info)
 {
 	struct sk_buff *skb;
-	u32 pid = req ? req->pid : n->nlmsg_pid;
 	int payload = sizeof(struct rtmsg) + 256;
+	u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0;
 	int err = -ENOBUFS;
 
 	skb = nlmsg_new(nlmsg_total_size(payload), GFP_KERNEL);
 	if (skb == NULL)
 		goto errout;
 
-	err = fib_dump_info(skb, pid, n->nlmsg_seq, event, tb_id,
-			    fa->fa_type, fa->fa_scope, &key, z, fa->fa_tos,
-			    fa->fa_info, 0);
+	err = fib_dump_info(skb, info->pid, seq, event, tb_id,
+			    fa->fa_type, fa->fa_scope, &key, dst_len,
+			    fa->fa_tos, fa->fa_info, 0);
 	if (err < 0) {
 		kfree_skb(skb);
 		goto errout;
 	}
 
-	err = rtnl_notify(skb, pid, RTNLGRP_IPV4_ROUTE, n, GFP_KERNEL);
+	err = rtnl_notify(skb, info->pid, RTNLGRP_IPV4_ROUTE,
+			  info->nlh, GFP_KERNEL);
 errout:
 	if (err < 0)
 		rtnl_set_sk_err(RTNLGRP_IPV4_ROUTE, err);
@@ -342,102 +343,100 @@ int fib_detect_death(struct fib_info *fi, int order,
 
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 
-static u32 fib_get_attr32(struct rtattr *attr, int attrlen, int type)
-{
-	while (RTA_OK(attr,attrlen)) {
-		if (attr->rta_type == type)
-			return *(u32*)RTA_DATA(attr);
-		attr = RTA_NEXT(attr, attrlen);
-	}
-	return 0;
-}
-
-static int
-fib_count_nexthops(struct rtattr *rta)
+static int fib_count_nexthops(struct rtnexthop *rtnh, int remaining)
 {
 	int nhs = 0;
-	struct rtnexthop *nhp = RTA_DATA(rta);
-	int nhlen = RTA_PAYLOAD(rta);
 
-	while (nhlen >= (int)sizeof(struct rtnexthop)) {
-		if ((nhlen -= nhp->rtnh_len) < 0)
-			return 0;
+	while (rtnh_ok(rtnh, remaining)) {
 		nhs++;
-		nhp = RTNH_NEXT(nhp);
-	};
-	return nhs;
+		rtnh = rtnh_next(rtnh, &remaining);
+	}
+
+	/* leftover implies invalid nexthop configuration, discard it */
+	return remaining > 0 ? 0 : nhs;
 }
 
-static int
-fib_get_nhs(struct fib_info *fi, const struct rtattr *rta, const struct rtmsg *r)
+static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
+		       int remaining, struct fib_config *cfg)
 {
-	struct rtnexthop *nhp = RTA_DATA(rta);
-	int nhlen = RTA_PAYLOAD(rta);
-
 	change_nexthops(fi) {
-		int attrlen = nhlen - sizeof(struct rtnexthop);
-		if (attrlen < 0 || (nhlen -= nhp->rtnh_len) < 0)
+		int attrlen;
+
+		if (!rtnh_ok(rtnh, remaining))
 			return -EINVAL;
-		nh->nh_flags = (r->rtm_flags&~0xFF) | nhp->rtnh_flags;
-		nh->nh_oif = nhp->rtnh_ifindex;
-		nh->nh_weight = nhp->rtnh_hops + 1;
-		if (attrlen) {
-			nh->nh_gw = fib_get_attr32(RTNH_DATA(nhp), attrlen, RTA_GATEWAY);
+
+		nh->nh_flags = (cfg->fc_flags & ~0xFF) | rtnh->rtnh_flags;
+		nh->nh_oif = rtnh->rtnh_ifindex;
+		nh->nh_weight = rtnh->rtnh_hops + 1;
+
+		attrlen = rtnh_attrlen(rtnh);
+		if (attrlen > 0) {
+			struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
+
+			nla = nla_find(attrs, attrlen, RTA_GATEWAY);
+			nh->nh_gw = nla ? nla_get_u32(nla) : 0;
 #ifdef CONFIG_NET_CLS_ROUTE
-			nh->nh_tclassid = fib_get_attr32(RTNH_DATA(nhp), attrlen, RTA_FLOW);
+			nla = nla_find(attrs, attrlen, RTA_FLOW);
+			nh->nh_tclassid = nla ? nla_get_u32(nla) : 0;
 #endif
 		}
-		nhp = RTNH_NEXT(nhp);
+
+		rtnh = rtnh_next(rtnh, &remaining);
 	} endfor_nexthops(fi);
+
 	return 0;
 }
 
 #endif
 
-int fib_nh_match(struct rtmsg *r, struct nlmsghdr *nlh, struct kern_rta *rta,
-		 struct fib_info *fi)
+int fib_nh_match(struct fib_config *cfg, struct fib_info *fi)
 {
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
-	struct rtnexthop *nhp;
-	int nhlen;
+	struct rtnexthop *rtnh;
+	int remaining;
 #endif
 
-	if (rta->rta_priority &&
-	    *rta->rta_priority != fi->fib_priority)
+	if (cfg->fc_priority && cfg->fc_priority != fi->fib_priority)
 		return 1;
 
-	if (rta->rta_oif || rta->rta_gw) {
-		if ((!rta->rta_oif || *rta->rta_oif == fi->fib_nh->nh_oif) &&
-		    (!rta->rta_gw  || memcmp(rta->rta_gw, &fi->fib_nh->nh_gw, 4) == 0))
+	if (cfg->fc_oif || cfg->fc_gw) {
+		if ((!cfg->fc_oif || cfg->fc_oif == fi->fib_nh->nh_oif) &&
+		    (!cfg->fc_gw  || cfg->fc_gw == fi->fib_nh->nh_gw))
 			return 0;
 		return 1;
 	}
 
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
-	if (rta->rta_mp == NULL)
+	if (cfg->fc_mp == NULL)
 		return 0;
-	nhp = RTA_DATA(rta->rta_mp);
-	nhlen = RTA_PAYLOAD(rta->rta_mp);
+
+	rtnh = cfg->fc_mp;
+	remaining = cfg->fc_mp_len;
 	
 	for_nexthops(fi) {
-		int attrlen = nhlen - sizeof(struct rtnexthop);
-		u32 gw;
+		int attrlen;
 
-		if (attrlen < 0 || (nhlen -= nhp->rtnh_len) < 0)
+		if (!rtnh_ok(rtnh, remaining))
 			return -EINVAL;
-		if (nhp->rtnh_ifindex && nhp->rtnh_ifindex != nh->nh_oif)
+
+		if (rtnh->rtnh_ifindex && rtnh->rtnh_ifindex != nh->nh_oif)
 			return 1;
-		if (attrlen) {
-			gw = fib_get_attr32(RTNH_DATA(nhp), attrlen, RTA_GATEWAY);
-			if (gw && gw != nh->nh_gw)
+
+		attrlen = rtnh_attrlen(rtnh);
+		if (attrlen < 0) {
+			struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
+
+			nla = nla_find(attrs, attrlen, RTA_GATEWAY);
+			if (nla && nla_get_u32(nla) != nh->nh_gw)
 				return 1;
 #ifdef CONFIG_NET_CLS_ROUTE
-			gw = fib_get_attr32(RTNH_DATA(nhp), attrlen, RTA_FLOW);
-			if (gw && gw != nh->nh_tclassid)
+			nla = nla_find(attrs, attrlen, RTA_FLOW);
+			if (nla && nla_get_u32(nla) != nh->nh_tclassid)
 				return 1;
 #endif
 		}
-		nhp = RTNH_NEXT(nhp);
+
+		rtnh = rtnh_next(rtnh, &remaining);
 	} endfor_nexthops(fi);
 #endif
 	return 0;
@@ -488,7 +487,8 @@ int fib_nh_match(struct rtmsg *r, struct nlmsghdr *nlh, struct kern_rta *rta,
 						|-> {local prefix} (terminal node)
  */
 
-static int fib_check_nh(const struct rtmsg *r, struct fib_info *fi, struct fib_nh *nh)
+static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi,
+			struct fib_nh *nh)
 {
 	int err;
 
@@ -502,7 +502,7 @@ static int fib_check_nh(const struct rtmsg *r, struct fib_info *fi, struct fib_n
 		if (nh->nh_flags&RTNH_F_ONLINK) {
 			struct net_device *dev;
 
-			if (r->rtm_scope >= RT_SCOPE_LINK)
+			if (cfg->fc_scope >= RT_SCOPE_LINK)
 				return -EINVAL;
 			if (inet_addr_type(nh->nh_gw) != RTN_UNICAST)
 				return -EINVAL;
@@ -516,10 +516,15 @@ static int fib_check_nh(const struct rtmsg *r, struct fib_info *fi, struct fib_n
 			return 0;
 		}
 		{
-			struct flowi fl = { .nl_u = { .ip4_u =
-						      { .daddr = nh->nh_gw,
-							.scope = r->rtm_scope + 1 } },
-					    .oif = nh->nh_oif };
+			struct flowi fl = {
+				.nl_u = {
+					.ip4_u = {
+						.daddr = nh->nh_gw,
+						.scope = cfg->fc_scope + 1,
+					},
+				},
+				.oif = nh->nh_oif,
+			};
 
 			/* It is not necessary, but requires a bit of thinking */
 			if (fl.fl4_scope < RT_SCOPE_LINK)
@@ -646,39 +651,28 @@ static void fib_hash_move(struct hlist_head *new_info_hash,
 	fib_hash_free(old_laddrhash, bytes);
 }
 
-struct fib_info *
-fib_create_info(const struct rtmsg *r, struct kern_rta *rta,
-		const struct nlmsghdr *nlh, int *errp)
+struct fib_info *fib_create_info(struct fib_config *cfg)
 {
 	int err;
 	struct fib_info *fi = NULL;
 	struct fib_info *ofi;
-#ifdef CONFIG_IP_ROUTE_MULTIPATH
 	int nhs = 1;
-#else
-	const int nhs = 1;
-#endif
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
-	u32 mp_alg = IP_MP_ALG_NONE;
-#endif
 
 	/* Fast check to catch the most weird cases */
-	if (fib_props[r->rtm_type].scope > r->rtm_scope)
+	if (fib_props[cfg->fc_type].scope > cfg->fc_scope)
 		goto err_inval;
 
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
-	if (rta->rta_mp) {
-		nhs = fib_count_nexthops(rta->rta_mp);
+	if (cfg->fc_mp) {
+		nhs = fib_count_nexthops(cfg->fc_mp, cfg->fc_mp_len);
 		if (nhs == 0)
 			goto err_inval;
 	}
 #endif
 #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
-	if (rta->rta_mp_alg) {
-		mp_alg = *rta->rta_mp_alg;
-
-		if (mp_alg < IP_MP_ALG_NONE ||
-		    mp_alg > IP_MP_ALG_MAX)
+	if (cfg->fc_mp_alg) {
+		if (cfg->fc_mp_alg < IP_MP_ALG_NONE ||
+		    cfg->fc_mp_alg > IP_MP_ALG_MAX)
 			goto err_inval;
 	}
 #endif
@@ -714,43 +708,42 @@ fib_create_info(const struct rtmsg *r, struct kern_rta *rta,
 		goto failure;
 	fib_info_cnt++;
 
-	fi->fib_protocol = r->rtm_protocol;
+	fi->fib_protocol = cfg->fc_protocol;
+	fi->fib_flags = cfg->fc_flags;
+	fi->fib_priority = cfg->fc_priority;
+	fi->fib_prefsrc = cfg->fc_prefsrc;
 
 	fi->fib_nhs = nhs;
 	change_nexthops(fi) {
 		nh->nh_parent = fi;
 	} endfor_nexthops(fi)
 
-	fi->fib_flags = r->rtm_flags;
-	if (rta->rta_priority)
-		fi->fib_priority = *rta->rta_priority;
-	if (rta->rta_mx) {
-		int attrlen = RTA_PAYLOAD(rta->rta_mx);
-		struct rtattr *attr = RTA_DATA(rta->rta_mx);
-
-		while (RTA_OK(attr, attrlen)) {
-			unsigned flavor = attr->rta_type;
-			if (flavor) {
-				if (flavor > RTAX_MAX)
+	if (cfg->fc_mx) {
+		struct nlattr *nla;
+		int remaining;
+
+		nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) {
+			int type = nla->nla_type;
+
+			if (type) {
+				if (type > RTAX_MAX)
 					goto err_inval;
-				fi->fib_metrics[flavor-1] = *(unsigned*)RTA_DATA(attr);
+				fi->fib_metrics[type - 1] = nla_get_u32(nla);
 			}
-			attr = RTA_NEXT(attr, attrlen);
 		}
 	}
-	if (rta->rta_prefsrc)
-		memcpy(&fi->fib_prefsrc, rta->rta_prefsrc, 4);
 
-	if (rta->rta_mp) {
+	if (cfg->fc_mp) {
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
-		if ((err = fib_get_nhs(fi, rta->rta_mp, r)) != 0)
+		err = fib_get_nhs(fi, cfg->fc_mp, cfg->fc_mp_len, cfg);
+		if (err != 0)
 			goto failure;
-		if (rta->rta_oif && fi->fib_nh->nh_oif != *rta->rta_oif)
+		if (cfg->fc_oif && fi->fib_nh->nh_oif != cfg->fc_oif)
 			goto err_inval;
-		if (rta->rta_gw && memcmp(&fi->fib_nh->nh_gw, rta->rta_gw, 4))
+		if (cfg->fc_gw && fi->fib_nh->nh_gw != cfg->fc_gw)
 			goto err_inval;
 #ifdef CONFIG_NET_CLS_ROUTE
-		if (rta->rta_flow && memcmp(&fi->fib_nh->nh_tclassid, rta->rta_flow, 4))
+		if (cfg->fc_flow && fi->fib_nh->nh_tclassid != cfg->fc_flow)
 			goto err_inval;
 #endif
 #else
@@ -758,34 +751,32 @@ fib_create_info(const struct rtmsg *r, struct kern_rta *rta,
 #endif
 	} else {
 		struct fib_nh *nh = fi->fib_nh;
-		if (rta->rta_oif)
-			nh->nh_oif = *rta->rta_oif;
-		if (rta->rta_gw)
-			memcpy(&nh->nh_gw, rta->rta_gw, 4);
+
+		nh->nh_oif = cfg->fc_oif;
+		nh->nh_gw = cfg->fc_gw;
+		nh->nh_flags = cfg->fc_flags;
 #ifdef CONFIG_NET_CLS_ROUTE
-		if (rta->rta_flow)
-			memcpy(&nh->nh_tclassid, rta->rta_flow, 4);
+		nh->nh_tclassid = cfg->fc_flow;
 #endif
-		nh->nh_flags = r->rtm_flags;
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 		nh->nh_weight = 1;
 #endif
 	}
 
 #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
-	fi->fib_mp_alg = mp_alg;
+	fi->fib_mp_alg = cfg->fc_mp_alg;
 #endif
 
-	if (fib_props[r->rtm_type].error) {
-		if (rta->rta_gw || rta->rta_oif || rta->rta_mp)
+	if (fib_props[cfg->fc_type].error) {
+		if (cfg->fc_gw || cfg->fc_oif || cfg->fc_mp)
 			goto err_inval;
 		goto link_it;
 	}
 
-	if (r->rtm_scope > RT_SCOPE_HOST)
+	if (cfg->fc_scope > RT_SCOPE_HOST)
 		goto err_inval;
 
-	if (r->rtm_scope == RT_SCOPE_HOST) {
+	if (cfg->fc_scope == RT_SCOPE_HOST) {
 		struct fib_nh *nh = fi->fib_nh;
 
 		/* Local address is added. */
@@ -798,14 +789,14 @@ fib_create_info(const struct rtmsg *r, struct kern_rta *rta,
 			goto failure;
 	} else {
 		change_nexthops(fi) {
-			if ((err = fib_check_nh(r, fi, nh)) != 0)
+			if ((err = fib_check_nh(cfg, fi, nh)) != 0)
 				goto failure;
 		} endfor_nexthops(fi)
 	}
 
 	if (fi->fib_prefsrc) {
-		if (r->rtm_type != RTN_LOCAL || rta->rta_dst == NULL ||
-		    memcmp(&fi->fib_prefsrc, rta->rta_dst, 4))
+		if (cfg->fc_type != RTN_LOCAL || !cfg->fc_dst ||
+		    fi->fib_prefsrc != cfg->fc_dst)
 			if (inet_addr_type(fi->fib_prefsrc) != RTN_LOCAL)
 				goto err_inval;
 	}
@@ -846,12 +837,12 @@ err_inval:
 	err = -EINVAL;
 
 failure:
-        *errp = err;
         if (fi) {
 		fi->fib_dead = 1;
 		free_fib_info(fi);
 	}
-	return NULL;
+
+	return ERR_PTR(err);
 }
 
 /* Note! fib_semantic_match intentionally uses  RCU list functions. */
@@ -1012,150 +1003,6 @@ rtattr_failure:
 	return -1;
 }
 
-#ifndef CONFIG_IP_NOSIOCRT
-
-int
-fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm,
-		    struct kern_rta *rta, struct rtentry *r)
-{
-	int    plen;
-	u32    *ptr;
-
-	memset(rtm, 0, sizeof(*rtm));
-	memset(rta, 0, sizeof(*rta));
-
-	if (r->rt_dst.sa_family != AF_INET)
-		return -EAFNOSUPPORT;
-
-	/* Check mask for validity:
-	   a) it must be contiguous.
-	   b) destination must have all host bits clear.
-	   c) if application forgot to set correct family (AF_INET),
-	      reject request unless it is absolutely clear i.e.
-	      both family and mask are zero.
-	 */
-	plen = 32;
-	ptr = &((struct sockaddr_in*)&r->rt_dst)->sin_addr.s_addr;
-	if (!(r->rt_flags&RTF_HOST)) {
-		u32 mask = ((struct sockaddr_in*)&r->rt_genmask)->sin_addr.s_addr;
-		if (r->rt_genmask.sa_family != AF_INET) {
-			if (mask || r->rt_genmask.sa_family)
-				return -EAFNOSUPPORT;
-		}
-		if (bad_mask(mask, *ptr))
-			return -EINVAL;
-		plen = inet_mask_len(mask);
-	}
-
-	nl->nlmsg_flags = NLM_F_REQUEST;
-	nl->nlmsg_pid = 0;
-	nl->nlmsg_seq = 0;
-	nl->nlmsg_len = NLMSG_LENGTH(sizeof(*rtm));
-	if (cmd == SIOCDELRT) {
-		nl->nlmsg_type = RTM_DELROUTE;
-		nl->nlmsg_flags = 0;
-	} else {
-		nl->nlmsg_type = RTM_NEWROUTE;
-		nl->nlmsg_flags = NLM_F_REQUEST|NLM_F_CREATE;
-		rtm->rtm_protocol = RTPROT_BOOT;
-	}
-
-	rtm->rtm_dst_len = plen;
-	rta->rta_dst = ptr;
-
-	if (r->rt_metric) {
-		*(u32*)&r->rt_pad3 = r->rt_metric - 1;
-		rta->rta_priority = (u32*)&r->rt_pad3;
-	}
-	if (r->rt_flags&RTF_REJECT) {
-		rtm->rtm_scope = RT_SCOPE_HOST;
-		rtm->rtm_type = RTN_UNREACHABLE;
-		return 0;
-	}
-	rtm->rtm_scope = RT_SCOPE_NOWHERE;
-	rtm->rtm_type = RTN_UNICAST;
-
-	if (r->rt_dev) {
-		char *colon;
-		struct net_device *dev;
-		char   devname[IFNAMSIZ];
-
-		if (copy_from_user(devname, r->rt_dev, IFNAMSIZ-1))
-			return -EFAULT;
-		devname[IFNAMSIZ-1] = 0;
-		colon = strchr(devname, ':');
-		if (colon)
-			*colon = 0;
-		dev = __dev_get_by_name(devname);
-		if (!dev)
-			return -ENODEV;
-		rta->rta_oif = &dev->ifindex;
-		if (colon) {
-			struct in_ifaddr *ifa;
-			struct in_device *in_dev = __in_dev_get_rtnl(dev);
-			if (!in_dev)
-				return -ENODEV;
-			*colon = ':';
-			for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next)
-				if (strcmp(ifa->ifa_label, devname) == 0)
-					break;
-			if (ifa == NULL)
-				return -ENODEV;
-			rta->rta_prefsrc = &ifa->ifa_local;
-		}
-	}
-
-	ptr = &((struct sockaddr_in*)&r->rt_gateway)->sin_addr.s_addr;
-	if (r->rt_gateway.sa_family == AF_INET && *ptr) {
-		rta->rta_gw = ptr;
-		if (r->rt_flags&RTF_GATEWAY && inet_addr_type(*ptr) == RTN_UNICAST)
-			rtm->rtm_scope = RT_SCOPE_UNIVERSE;
-	}
-
-	if (cmd == SIOCDELRT)
-		return 0;
-
-	if (r->rt_flags&RTF_GATEWAY && rta->rta_gw == NULL)
-		return -EINVAL;
-
-	if (rtm->rtm_scope == RT_SCOPE_NOWHERE)
-		rtm->rtm_scope = RT_SCOPE_LINK;
-
-	if (r->rt_flags&(RTF_MTU|RTF_WINDOW|RTF_IRTT)) {
-		struct rtattr *rec;
-		struct rtattr *mx = kmalloc(RTA_LENGTH(3*RTA_LENGTH(4)), GFP_KERNEL);
-		if (mx == NULL)
-			return -ENOMEM;
-		rta->rta_mx = mx;
-		mx->rta_type = RTA_METRICS;
-		mx->rta_len  = RTA_LENGTH(0);
-		if (r->rt_flags&RTF_MTU) {
-			rec = (void*)((char*)mx + RTA_ALIGN(mx->rta_len));
-			rec->rta_type = RTAX_ADVMSS;
-			rec->rta_len = RTA_LENGTH(4);
-			mx->rta_len += RTA_LENGTH(4);
-			*(u32*)RTA_DATA(rec) = r->rt_mtu - 40;
-		}
-		if (r->rt_flags&RTF_WINDOW) {
-			rec = (void*)((char*)mx + RTA_ALIGN(mx->rta_len));
-			rec->rta_type = RTAX_WINDOW;
-			rec->rta_len = RTA_LENGTH(4);
-			mx->rta_len += RTA_LENGTH(4);
-			*(u32*)RTA_DATA(rec) = r->rt_window;
-		}
-		if (r->rt_flags&RTF_IRTT) {
-			rec = (void*)((char*)mx + RTA_ALIGN(mx->rta_len));
-			rec->rta_type = RTAX_RTT;
-			rec->rta_len = RTA_LENGTH(4);
-			mx->rta_len += RTA_LENGTH(4);
-			*(u32*)RTA_DATA(rec) = r->rt_irtt<<3;
-		}
-	}
-	return 0;
-}
-
-#endif
-
 /*
    Update FIB if:
    - local address disappeared -> we must delete all the entries
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 2a580eb2579b..41bef0a88ab6 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1124,17 +1124,14 @@ err:
 	return fa_head;
 }
 
-static int
-fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
-	       struct nlmsghdr *nlhdr, struct netlink_skb_parms *req)
+static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg)
 {
 	struct trie *t = (struct trie *) tb->tb_data;
 	struct fib_alias *fa, *new_fa;
 	struct list_head *fa_head = NULL;
 	struct fib_info *fi;
-	int plen = r->rtm_dst_len;
-	int type = r->rtm_type;
-	u8 tos = r->rtm_tos;
+	int plen = cfg->fc_dst_len;
+	u8 tos = cfg->fc_tos;
 	u32 key, mask;
 	int err;
 	struct leaf *l;
@@ -1142,11 +1139,7 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 	if (plen > 32)
 		return -EINVAL;
 
-	key = 0;
-	if (rta->rta_dst)
-		memcpy(&key, rta->rta_dst, 4);
-
-	key = ntohl(key);
+	key = ntohl(cfg->fc_dst);
 
 	pr_debug("Insert table=%u %08x/%d\n", tb->tb_id, key, plen);
 
@@ -1157,10 +1150,11 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 
 	key = key & mask;
 
-	fi = fib_create_info(r, rta, nlhdr, &err);
-
-	if (!fi)
+	fi = fib_create_info(cfg);
+	if (IS_ERR(fi)) {
+		err = PTR_ERR(fi);
 		goto err;
+	}
 
 	l = fib_find_node(t, key);
 	fa = NULL;
@@ -1185,10 +1179,10 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 		struct fib_alias *fa_orig;
 
 		err = -EEXIST;
-		if (nlhdr->nlmsg_flags & NLM_F_EXCL)
+		if (cfg->fc_nlflags & NLM_F_EXCL)
 			goto out;
 
-		if (nlhdr->nlmsg_flags & NLM_F_REPLACE) {
+		if (cfg->fc_nlflags & NLM_F_REPLACE) {
 			struct fib_info *fi_drop;
 			u8 state;
 
@@ -1200,8 +1194,8 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 			fi_drop = fa->fa_info;
 			new_fa->fa_tos = fa->fa_tos;
 			new_fa->fa_info = fi;
-			new_fa->fa_type = type;
-			new_fa->fa_scope = r->rtm_scope;
+			new_fa->fa_type = cfg->fc_type;
+			new_fa->fa_scope = cfg->fc_scope;
 			state = fa->fa_state;
 			new_fa->fa_state &= ~FA_S_ACCESSED;
 
@@ -1224,17 +1218,17 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 				break;
 			if (fa->fa_info->fib_priority != fi->fib_priority)
 				break;
-			if (fa->fa_type == type &&
-			    fa->fa_scope == r->rtm_scope &&
+			if (fa->fa_type == cfg->fc_type &&
+			    fa->fa_scope == cfg->fc_scope &&
 			    fa->fa_info == fi) {
 				goto out;
 			}
 		}
-		if (!(nlhdr->nlmsg_flags & NLM_F_APPEND))
+		if (!(cfg->fc_nlflags & NLM_F_APPEND))
 			fa = fa_orig;
 	}
 	err = -ENOENT;
-	if (!(nlhdr->nlmsg_flags & NLM_F_CREATE))
+	if (!(cfg->fc_nlflags & NLM_F_CREATE))
 		goto out;
 
 	err = -ENOBUFS;
@@ -1244,8 +1238,8 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 
 	new_fa->fa_info = fi;
 	new_fa->fa_tos = tos;
-	new_fa->fa_type = type;
-	new_fa->fa_scope = r->rtm_scope;
+	new_fa->fa_type = cfg->fc_type;
+	new_fa->fa_scope = cfg->fc_scope;
 	new_fa->fa_state = 0;
 	/*
 	 * Insert new entry to the list.
@@ -1262,7 +1256,8 @@ fn_trie_insert(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 			  (fa ? &fa->fa_list : fa_head));
 
 	rt_cache_flush(-1);
-	rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id, nlhdr, req);
+	rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id,
+		  &cfg->fc_nlinfo);
 succeeded:
 	return 0;
 
@@ -1548,28 +1543,21 @@ static int trie_leaf_remove(struct trie *t, t_key key)
 	return 1;
 }
 
-static int
-fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
-		struct nlmsghdr *nlhdr, struct netlink_skb_parms *req)
+static int fn_trie_delete(struct fib_table *tb, struct fib_config *cfg)
 {
 	struct trie *t = (struct trie *) tb->tb_data;
 	u32 key, mask;
-	int plen = r->rtm_dst_len;
-	u8 tos = r->rtm_tos;
+	int plen = cfg->fc_dst_len;
+	u8 tos = cfg->fc_tos;
 	struct fib_alias *fa, *fa_to_delete;
 	struct list_head *fa_head;
 	struct leaf *l;
 	struct leaf_info *li;
 
-
 	if (plen > 32)
 		return -EINVAL;
 
-	key = 0;
-	if (rta->rta_dst)
-		memcpy(&key, rta->rta_dst, 4);
-
-	key = ntohl(key);
+	key = ntohl(cfg->fc_dst);
 	mask = ntohl(inet_make_mask(plen));
 
 	if (key & ~mask)
@@ -1598,13 +1586,12 @@ fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 		if (fa->fa_tos != tos)
 			break;
 
-		if ((!r->rtm_type ||
-		     fa->fa_type == r->rtm_type) &&
-		    (r->rtm_scope == RT_SCOPE_NOWHERE ||
-		     fa->fa_scope == r->rtm_scope) &&
-		    (!r->rtm_protocol ||
-		     fi->fib_protocol == r->rtm_protocol) &&
-		    fib_nh_match(r, nlhdr, rta, fi) == 0) {
+		if ((!cfg->fc_type || fa->fa_type == cfg->fc_type) &&
+		    (cfg->fc_scope == RT_SCOPE_NOWHERE ||
+		     fa->fa_scope == cfg->fc_scope) &&
+		    (!cfg->fc_protocol ||
+		     fi->fib_protocol == cfg->fc_protocol) &&
+		    fib_nh_match(cfg, fi) == 0) {
 			fa_to_delete = fa;
 			break;
 		}
@@ -1614,7 +1601,8 @@ fn_trie_delete(struct fib_table *tb, struct rtmsg *r, struct kern_rta *rta,
 		return -ESRCH;
 
 	fa = fa_to_delete;
-	rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id, nlhdr, req);
+	rtmsg_fib(RTM_DELROUTE, htonl(key), fa, plen, tb->tb_id,
+		  &cfg->fc_nlinfo);
 
 	l = fib_find_node(t, key);
 	li = find_leaf_info(l, plen);
-- 
cgit v1.2.3


From d889ce3b29e55b91257964b4c9aac70b91fedd91 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Thu, 17 Aug 2006 18:15:44 -0700
Subject: [IPv4]: Convert route get to new netlink api

Fixes various unvalidated netlink attributes causing memory
corruptions when left empty by userspace applications.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    |  1 +
 net/ipv4/fib_frontend.c |  2 +-
 net/ipv4/route.c        | 84 ++++++++++++++++++++++++++-----------------------
 3 files changed, 47 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 42ed96fab3f5..fcc159a4ac17 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -216,6 +216,7 @@ extern void fib_select_default(const struct flowi *flp, struct fib_result *res);
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
 /* Exported by fib_frontend.c */
+extern struct nla_policy rtm_ipv4_policy[];
 extern void		ip_fib_init(void);
 extern int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
 extern int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index d537c933abe3..d0abeab16e66 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -453,7 +453,7 @@ int ip_rt_ioctl(unsigned int cmd, void *arg)
 
 #endif
 
-static struct nla_policy rtm_ipv4_policy[RTA_MAX+1] __read_mostly = {
+struct nla_policy rtm_ipv4_policy[RTA_MAX+1] __read_mostly = {
 	[RTA_DST]		= { .type = NLA_U32 },
 	[RTA_SRC]		= { .type = NLA_U32 },
 	[RTA_IIF]		= { .type = NLA_U32 },
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 31b67059ac29..a4d4cb85a16c 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2737,18 +2737,24 @@ nla_put_failure:
 
 int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct rtattr **rta = arg;
-	struct rtmsg *rtm = NLMSG_DATA(nlh);
+	struct rtmsg *rtm;
+	struct nlattr *tb[RTA_MAX+1];
 	struct rtable *rt = NULL;
-	u32 dst = 0;
-	u32 src = 0;
-	int iif = 0;
-	int err = -ENOBUFS;
+	u32 dst, src, iif;
+	int err;
 	struct sk_buff *skb;
 
+	err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv4_policy);
+	if (err < 0)
+		goto errout;
+
+	rtm = nlmsg_data(nlh);
+
 	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
-	if (!skb)
-		goto out;
+	if (skb == NULL) {
+		err = -ENOBUFS;
+		goto errout;
+	}
 
 	/* Reserve room for dummy headers, this skb can pass
 	   through good chunk of routing engine.
@@ -2759,61 +2765,61 @@ int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 	skb->nh.iph->protocol = IPPROTO_ICMP;
 	skb_reserve(skb, MAX_HEADER + sizeof(struct iphdr));
 
-	if (rta[RTA_SRC - 1])
-		memcpy(&src, RTA_DATA(rta[RTA_SRC - 1]), 4);
-	if (rta[RTA_DST - 1])
-		memcpy(&dst, RTA_DATA(rta[RTA_DST - 1]), 4);
-	if (rta[RTA_IIF - 1])
-		memcpy(&iif, RTA_DATA(rta[RTA_IIF - 1]), sizeof(int));
+	src = tb[RTA_SRC] ? nla_get_u32(tb[RTA_SRC]) : 0;
+	dst = tb[RTA_DST] ? nla_get_u32(tb[RTA_DST]) : 0;
+	iif = tb[RTA_IIF] ? nla_get_u32(tb[RTA_IIF]) : 0;
 
 	if (iif) {
-		struct net_device *dev = __dev_get_by_index(iif);
-		err = -ENODEV;
-		if (!dev)
-			goto out_free;
+		struct net_device *dev;
+
+		dev = __dev_get_by_index(iif);
+		if (dev == NULL) {
+			err = -ENODEV;
+			goto errout_free;
+		}
+
 		skb->protocol	= htons(ETH_P_IP);
 		skb->dev	= dev;
 		local_bh_disable();
 		err = ip_route_input(skb, dst, src, rtm->rtm_tos, dev);
 		local_bh_enable();
-		rt = (struct rtable*)skb->dst;
-		if (!err && rt->u.dst.error)
+
+		rt = (struct rtable*) skb->dst;
+		if (err == 0 && rt->u.dst.error)
 			err = -rt->u.dst.error;
 	} else {
-		struct flowi fl = { .nl_u = { .ip4_u = { .daddr = dst,
-							 .saddr = src,
-							 .tos = rtm->rtm_tos } } };
-		int oif = 0;
-		if (rta[RTA_OIF - 1])
-			memcpy(&oif, RTA_DATA(rta[RTA_OIF - 1]), sizeof(int));
-		fl.oif = oif;
+		struct flowi fl = {
+			.nl_u = {
+				.ip4_u = {
+					.daddr = dst,
+					.saddr = src,
+					.tos = rtm->rtm_tos,
+				},
+			},
+			.oif = tb[RTA_OIF] ? nla_get_u32(tb[RTA_OIF]) : 0,
+		};
 		err = ip_route_output_key(&rt, &fl);
 	}
+
 	if (err)
-		goto out_free;
+		goto errout_free;
 
 	skb->dst = &rt->u.dst;
 	if (rtm->rtm_flags & RTM_F_NOTIFY)
 		rt->rt_flags |= RTCF_NOTIFY;
 
-	NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid;
-
 	err = rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq,
 				RTM_NEWROUTE, 0, 0);
-	if (!err)
-		goto out_free;
-	if (err < 0) {
-		err = -EMSGSIZE;
-		goto out_free;
-	}
+	if (err <= 0)
+		goto errout_free;
 
 	err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid);
-out:
+errout:
 	return err;
 
-out_free:
+errout_free:
 	kfree_skb(skb);
-	goto out;
+	goto errout;
 }
 
 int ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb)
-- 
cgit v1.2.3


From e9ce1cd3cf6cf35b21d0ce990f2e738f35907386 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Mon, 21 Aug 2006 23:54:55 -0700
Subject: [PKT_SCHED]: Kill pkt_act.h inlining.

This was simply making templates of functions and mostly causing a lot
of code duplication in the classifier action modules.

We solve this more cleanly by having a common "struct tcf_common" that
hash worker functions contained once in act_api.c can work with.

Callers work with real action objects that have the common struct
plus their module specific struct members.  You go from a common
object to the higher level one using a "to_foo()" macro which makes
use of container_of() to do the dirty work.

This also kills off act_generic.h which was only used by act_simple.c
and keeping it around was more work than the it's value.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/act_api.h          | 136 +++++++----
 include/net/act_generic.h      | 142 ------------
 include/net/pkt_act.h          | 273 ----------------------
 include/net/tc_act/tc_defact.h |  13 +-
 include/net/tc_act/tc_gact.h   |  18 +-
 include/net/tc_act/tc_ipt.h    |  15 +-
 include/net/tc_act/tc_mirred.h |  17 +-
 include/net/tc_act/tc_pedit.h  |  15 +-
 net/sched/act_api.c            | 246 ++++++++++++++++++--
 net/sched/act_gact.c           | 142 ++++++------
 net/sched/act_ipt.c            | 175 +++++++-------
 net/sched/act_mirred.c         | 159 ++++++-------
 net/sched/act_pedit.c          | 166 ++++++--------
 net/sched/act_police.c         | 508 +++++++++++++++++++++--------------------
 net/sched/act_simple.c         | 183 ++++++++++++---
 15 files changed, 1061 insertions(+), 1147 deletions(-)
 delete mode 100644 include/net/act_generic.h
 delete mode 100644 include/net/pkt_act.h

(limited to 'include')

diff --git a/include/net/act_api.h b/include/net/act_api.h
index 11e9eaf79f5a..8b06c2f3657f 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -8,70 +8,110 @@
 #include <net/sch_generic.h>
 #include <net/pkt_sched.h>
 
-#define tca_gen(name) \
-struct tcf_##name *next; \
-	u32 index; \
-	int refcnt; \
-	int bindcnt; \
-	u32 capab; \
-	int action; \
-	struct tcf_t tm; \
-	struct gnet_stats_basic bstats; \
-	struct gnet_stats_queue qstats; \
-	struct gnet_stats_rate_est rate_est; \
-	spinlock_t *stats_lock; \
-	spinlock_t lock
-
-struct tcf_police
-{
-	tca_gen(police);
-	int		result;
-	u32		ewma_rate;
-	u32		burst;
-	u32		mtu;
-	u32		toks;
-	u32		ptoks;
-	psched_time_t	t_c;
-	struct qdisc_rate_table *R_tab;
-	struct qdisc_rate_table *P_tab;
+struct tcf_common {
+	struct tcf_common		*tcfc_next;
+	u32				tcfc_index;
+	int				tcfc_refcnt;
+	int				tcfc_bindcnt;
+	u32				tcfc_capab;
+	int				tcfc_action;
+	struct tcf_t			tcfc_tm;
+	struct gnet_stats_basic		tcfc_bstats;
+	struct gnet_stats_queue		tcfc_qstats;
+	struct gnet_stats_rate_est	tcfc_rate_est;
+	spinlock_t			*tcfc_stats_lock;
+	spinlock_t			tcfc_lock;
+};
+#define tcf_next	common.tcfc_next
+#define tcf_index	common.tcfc_index
+#define tcf_refcnt	common.tcfc_refcnt
+#define tcf_bindcnt	common.tcfc_bindcnt
+#define tcf_capab	common.tcfc_capab
+#define tcf_action	common.tcfc_action
+#define tcf_tm		common.tcfc_tm
+#define tcf_bstats	common.tcfc_bstats
+#define tcf_qstats	common.tcfc_qstats
+#define tcf_rate_est	common.tcfc_rate_est
+#define tcf_stats_lock	common.tcfc_stats_lock
+#define tcf_lock	common.tcfc_lock
+
+struct tcf_police {
+	struct tcf_common	common;
+	int			tcfp_result;
+	u32			tcfp_ewma_rate;
+	u32			tcfp_burst;
+	u32			tcfp_mtu;
+	u32			tcfp_toks;
+	u32			tcfp_ptoks;
+	psched_time_t		tcfp_t_c;
+	struct qdisc_rate_table	*tcfp_R_tab;
+	struct qdisc_rate_table	*tcfp_P_tab;
 };
+#define to_police(pc)	\
+	container_of(pc, struct tcf_police, common)
+
+struct tcf_hashinfo {
+	struct tcf_common	**htab;
+	unsigned int		hmask;
+	rwlock_t		*lock;
+};
+
+static inline unsigned int tcf_hash(u32 index, unsigned int hmask)
+{
+	return index & hmask;
+}
 
 #ifdef CONFIG_NET_CLS_ACT
 
 #define ACT_P_CREATED 1
 #define ACT_P_DELETED 1
 
-struct tcf_act_hdr
-{
-	tca_gen(act_hdr);
+struct tcf_act_hdr {
+	struct tcf_common	common;
 };
 
-struct tc_action
-{
-	void *priv;
-	struct tc_action_ops *ops;
-	__u32   type;   /* for backward compat(TCA_OLD_COMPAT) */
-	__u32   order; 
-	struct tc_action *next;
+struct tc_action {
+	void			*priv;
+	struct tc_action_ops	*ops;
+	__u32			type; /* for backward compat(TCA_OLD_COMPAT) */
+	__u32			order;
+	struct tc_action	*next;
 };
 
 #define TCA_CAP_NONE 0
-struct tc_action_ops
-{
+struct tc_action_ops {
 	struct tc_action_ops *next;
+	struct tcf_hashinfo *hinfo;
 	char    kind[IFNAMSIZ];
 	__u32   type; /* TBD to match kind */
 	__u32 	capab;  /* capabilities includes 4 bit version */
 	struct module		*owner;
 	int     (*act)(struct sk_buff *, struct tc_action *, struct tcf_result *);
 	int     (*get_stats)(struct sk_buff *, struct tc_action *);
-	int     (*dump)(struct sk_buff *, struct tc_action *,int , int);
+	int     (*dump)(struct sk_buff *, struct tc_action *, int, int);
 	int     (*cleanup)(struct tc_action *, int bind);
-	int     (*lookup)(struct tc_action *, u32 );
-	int     (*init)(struct rtattr *,struct rtattr *,struct tc_action *, int , int );
-	int     (*walk)(struct sk_buff *, struct netlink_callback *, int , struct tc_action *);
+	int     (*lookup)(struct tc_action *, u32);
+	int     (*init)(struct rtattr *, struct rtattr *, struct tc_action *, int , int);
+	int     (*walk)(struct sk_buff *, struct netlink_callback *, int, struct tc_action *);
 };
 
+extern struct tcf_common *tcf_hash_lookup(u32 index,
+					  struct tcf_hashinfo *hinfo);
+extern void tcf_hash_destroy(struct tcf_common *p, struct tcf_hashinfo *hinfo);
+extern int tcf_hash_release(struct tcf_common *p, int bind,
+			    struct tcf_hashinfo *hinfo);
+extern int tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb,
+			      int type, struct tc_action *a);
+extern u32 tcf_hash_new_index(u32 *idx_gen, struct tcf_hashinfo *hinfo);
+extern int tcf_hash_search(struct tc_action *a, u32 index);
+extern struct tcf_common *tcf_hash_check(u32 index, struct tc_action *a,
+					 int bind, struct tcf_hashinfo *hinfo);
+extern struct tcf_common *tcf_hash_create(u32 index, struct rtattr *est,
+					  struct tc_action *a, int size,
+					  int bind, u32 *idx_gen,
+					  struct tcf_hashinfo *hinfo);
+extern void tcf_hash_insert(struct tcf_common *p, struct tcf_hashinfo *hinfo);
+
 extern int tcf_register_action(struct tc_action_ops *a);
 extern int tcf_unregister_action(struct tc_action_ops *a);
 extern void tcf_action_destroy(struct tc_action *a, int bind);
@@ -96,17 +136,17 @@ tcf_police_release(struct tcf_police *p, int bind)
 	int ret = 0;
 #ifdef CONFIG_NET_CLS_ACT
 	if (p) {
-		if (bind) {
-			 p->bindcnt--;
-		}
-		p->refcnt--;
-		if (p->refcnt <= 0 && !p->bindcnt) {
+		if (bind)
+			p->tcf_bindcnt--;
+
+		p->tcf_refcnt--;
+		if (p->tcf_refcnt <= 0 && !p->tcf_bindcnt) {
 			tcf_police_destroy(p);
 			ret = 1;
 		}
 	}
 #else
-	if (p && --p->refcnt == 0)
+	if (p && --p->tcf_refcnt == 0)
 		tcf_police_destroy(p);
 
 #endif /* CONFIG_NET_CLS_ACT */
diff --git a/include/net/act_generic.h b/include/net/act_generic.h
deleted file mode 100644
index c9daa7e52300..000000000000
--- a/include/net/act_generic.h
+++ /dev/null
@@ -1,142 +0,0 @@
-/*
- * include/net/act_generic.h
- *
-*/
-#ifndef _NET_ACT_GENERIC_H
-#define _NET_ACT_GENERIC_H
-static inline int tcf_defact_release(struct tcf_defact *p, int bind)
-{
-	int ret = 0;
-	if (p) {
-		if (bind) {
-			p->bindcnt--;
-		}
-		p->refcnt--;
-		if (p->bindcnt <= 0 && p->refcnt <= 0) {
-			kfree(p->defdata);
-			tcf_hash_destroy(p);
-			ret = 1;
-		}
-	}
-	return ret;
-}
-
-static inline int
-alloc_defdata(struct tcf_defact *p, u32 datalen, void *defdata)
-{
-	p->defdata = kmalloc(datalen, GFP_KERNEL);
-	if (p->defdata == NULL)
-		return -ENOMEM;
-	p->datalen = datalen;
-	memcpy(p->defdata, defdata, datalen);
-	return 0;
-}
-
-static inline int
-realloc_defdata(struct tcf_defact *p, u32 datalen, void *defdata)
-{
-	/* safer to be just brute force for now */
-	kfree(p->defdata);
-	return alloc_defdata(p, datalen, defdata);
-}
-
-static inline int
-tcf_defact_init(struct rtattr *rta, struct rtattr *est,
-		struct tc_action *a, int ovr, int bind)
-{
-	struct rtattr *tb[TCA_DEF_MAX];
-	struct tc_defact *parm;
-	struct tcf_defact *p;
-	void *defdata;
-	u32 datalen = 0;
-	int ret = 0;
-
-	if (rta == NULL || rtattr_parse_nested(tb, TCA_DEF_MAX, rta) < 0)
-		return -EINVAL;
-
-	if (tb[TCA_DEF_PARMS - 1] == NULL || 
-	    RTA_PAYLOAD(tb[TCA_DEF_PARMS - 1]) < sizeof(*parm))
-		return -EINVAL;
-
-	parm = RTA_DATA(tb[TCA_DEF_PARMS - 1]);
-	defdata = RTA_DATA(tb[TCA_DEF_DATA - 1]);
-	if (defdata == NULL)
-		return -EINVAL;
-
-	datalen = RTA_PAYLOAD(tb[TCA_DEF_DATA - 1]);
-	if (datalen <= 0)
-		return -EINVAL;
-
-	p = tcf_hash_check(parm->index, a, ovr, bind);
-	if (p == NULL) {
-		p = tcf_hash_create(parm->index, est, a, sizeof(*p), ovr, bind);
-		if (p == NULL)
-			return -ENOMEM;
-
-		ret = alloc_defdata(p, datalen, defdata);
-		if (ret < 0) {
-			kfree(p);
-			return ret;
-		}
-		ret = ACT_P_CREATED;
-	} else {
-		if (!ovr) {
-			tcf_defact_release(p, bind);
-			return -EEXIST;
-		}
-		realloc_defdata(p, datalen, defdata);
-	}
-
-	spin_lock_bh(&p->lock);
-	p->action = parm->action;
-	spin_unlock_bh(&p->lock);
-	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(p);
-	return ret;
-}
-
-static inline int tcf_defact_cleanup(struct tc_action *a, int bind)
-{
-	struct tcf_defact *p = PRIV(a, defact);
-
-	if (p != NULL)
-		return tcf_defact_release(p, bind);
-	return 0;
-}
-
-static inline int
-tcf_defact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
-{
-	unsigned char *b = skb->tail;
-	struct tc_defact opt;
-	struct tcf_defact *p = PRIV(a, defact);
-	struct tcf_t t;
-
-	opt.index = p->index;
-	opt.refcnt = p->refcnt - ref;
-	opt.bindcnt = p->bindcnt - bind;
-	opt.action = p->action;
-	RTA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt);
-	RTA_PUT(skb, TCA_DEF_DATA, p->datalen, p->defdata);
-	t.install = jiffies_to_clock_t(jiffies - p->tm.install);
-	t.lastuse = jiffies_to_clock_t(jiffies - p->tm.lastuse);
-	t.expires = jiffies_to_clock_t(p->tm.expires);
-	RTA_PUT(skb, TCA_DEF_TM, sizeof(t), &t);
-	return skb->len;
-
-rtattr_failure:
-	skb_trim(skb, b - skb->data);
-	return -1;
-}
-
-#define tca_use_default_ops \
-	.dump           =       tcf_defact_dump, \
-	.cleanup        =       tcf_defact_cleanup, \
-	.init           =       tcf_defact_init, \
-	.walk           =       tcf_generic_walker, \
-
-#define tca_use_default_defines(name) \
-	static u32 idx_gen; \
-	static struct tcf_defact *tcf_##name_ht[MY_TAB_SIZE]; \
-	static DEFINE_RWLOCK(##name_lock);
-#endif /* _NET_ACT_GENERIC_H */
diff --git a/include/net/pkt_act.h b/include/net/pkt_act.h
deleted file mode 100644
index cf5e4d2e4c21..000000000000
--- a/include/net/pkt_act.h
+++ /dev/null
@@ -1,273 +0,0 @@
-#ifndef __NET_PKT_ACT_H
-#define __NET_PKT_ACT_H
-
-#include <asm/uaccess.h>
-#include <asm/system.h>
-#include <linux/bitops.h>
-#include <linux/types.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/string.h>
-#include <linux/mm.h>
-#include <linux/socket.h>
-#include <linux/sockios.h>
-#include <linux/in.h>
-#include <linux/errno.h>
-#include <linux/interrupt.h>
-#include <linux/skbuff.h>
-#include <linux/rtnetlink.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/proc_fs.h>
-#include <net/sock.h>
-#include <net/pkt_sched.h>
-
-#define tca_st(val) (struct tcf_##val *)
-#define PRIV(a,name) ( tca_st(name) (a)->priv)
-
-#if 0 /* control */
-#define DPRINTK(format,args...) printk(KERN_DEBUG format,##args)
-#else
-#define DPRINTK(format,args...)
-#endif
-
-#if 0 /* data */
-#define D2PRINTK(format,args...) printk(KERN_DEBUG format,##args)
-#else
-#define D2PRINTK(format,args...)
-#endif
-
-static __inline__ unsigned
-tcf_hash(u32 index)
-{
-	return index & MY_TAB_MASK;
-}
-
-/* probably move this from being inline
- * and put into act_generic
-*/
-static inline void
-tcf_hash_destroy(struct tcf_st *p)
-{
-	unsigned h = tcf_hash(p->index);
-	struct tcf_st **p1p;
-
-	for (p1p = &tcf_ht[h]; *p1p; p1p = &(*p1p)->next) {
-		if (*p1p == p) {
-			write_lock_bh(&tcf_t_lock);
-			*p1p = p->next;
-			write_unlock_bh(&tcf_t_lock);
-#ifdef CONFIG_NET_ESTIMATOR
-			gen_kill_estimator(&p->bstats, &p->rate_est);
-#endif
-			kfree(p);
-			return;
-		}
-	}
-	BUG_TRAP(0);
-}
-
-static inline int
-tcf_hash_release(struct tcf_st *p, int bind )
-{
-	int ret = 0;
-	if (p) {
-		if (bind) {
-			p->bindcnt--;
-		}
-		p->refcnt--;
-	       	if(p->bindcnt <=0 && p->refcnt <= 0) {
-			tcf_hash_destroy(p);
-			ret = 1;
-		}
-	}
-	return ret;
-}
-
-static __inline__ int
-tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb,
-		struct tc_action *a)
-{
-	struct tcf_st *p;
-	int err =0, index =  -1,i= 0, s_i = 0, n_i = 0;
-	struct rtattr *r ;
-
-	read_lock(&tcf_t_lock);
-
-	s_i = cb->args[0];
-
-	for (i = 0; i < MY_TAB_SIZE; i++) {
-		p = tcf_ht[tcf_hash(i)];
-
-		for (; p; p = p->next) {
-			index++;
-			if (index < s_i)
-				continue;
-			a->priv = p;
-			a->order = n_i;
-			r = (struct rtattr*) skb->tail;
-			RTA_PUT(skb, a->order, 0, NULL);
-			err = tcf_action_dump_1(skb, a, 0, 0);
-			if (0 > err) {
-				index--;
-				skb_trim(skb, (u8*)r - skb->data);
-				goto done;
-			}
-			r->rta_len = skb->tail - (u8*)r;
-			n_i++;
-			if (n_i >= TCA_ACT_MAX_PRIO) {
-				goto done;
-			}
-		}
-	}
-done:
-	read_unlock(&tcf_t_lock);
-	if (n_i)
-		cb->args[0] += n_i;
-	return n_i;
-
-rtattr_failure:
-	skb_trim(skb, (u8*)r - skb->data);
-	goto done;
-}
-
-static __inline__ int
-tcf_del_walker(struct sk_buff *skb, struct tc_action *a)
-{
-	struct tcf_st *p, *s_p;
-	struct rtattr *r ;
-	int i= 0, n_i = 0;
-
-	r = (struct rtattr*) skb->tail;
-	RTA_PUT(skb, a->order, 0, NULL);
-	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, a->ops->kind);
-	for (i = 0; i < MY_TAB_SIZE; i++) {
-		p = tcf_ht[tcf_hash(i)];
-
-		while (p != NULL) {
-			s_p = p->next;
-			if (ACT_P_DELETED == tcf_hash_release(p, 0)) {
-				 module_put(a->ops->owner);
-			}
-			n_i++;
-			p = s_p;
-		}
-	}
-	RTA_PUT(skb, TCA_FCNT, 4, &n_i);
-	r->rta_len = skb->tail - (u8*)r;
-
-	return n_i;
-rtattr_failure:
-	skb_trim(skb, (u8*)r - skb->data);
-	return -EINVAL;
-}
-
-static __inline__ int
-tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb, int type,
-		struct tc_action *a)
-{
-		if (type == RTM_DELACTION) {
-			return tcf_del_walker(skb,a);
-		} else if (type == RTM_GETACTION) {
-			return tcf_dump_walker(skb,cb,a);
-		} else {
-			printk("tcf_generic_walker: unknown action %d\n",type);
-			return -EINVAL;
-		}
-}
-
-static __inline__ struct tcf_st *
-tcf_hash_lookup(u32 index)
-{
-	struct tcf_st *p;
-
-	read_lock(&tcf_t_lock);
-	for (p = tcf_ht[tcf_hash(index)]; p; p = p->next) {
-		if (p->index == index)
-			break;
-	}
-	read_unlock(&tcf_t_lock);
-	return p;
-}
-
-static __inline__ u32
-tcf_hash_new_index(void)
-{
-	do {
-		if (++idx_gen == 0)
-			idx_gen = 1;
-	} while (tcf_hash_lookup(idx_gen));
-
-	return idx_gen;
-}
-
-
-static inline int
-tcf_hash_search(struct tc_action *a, u32 index)
-{
-	struct tcf_st *p = tcf_hash_lookup(index);
-
-	if (p != NULL) {
-		a->priv = p;
-		return 1;
-	}
-	return 0;
-}
-
-#ifdef CONFIG_NET_ACT_INIT
-static inline struct tcf_st *
-tcf_hash_check(u32 index, struct tc_action *a, int ovr, int bind)
-{
-	struct tcf_st *p = NULL;
-	if (index && (p = tcf_hash_lookup(index)) != NULL) {
-		if (bind) {
-			p->bindcnt++;
-			p->refcnt++;
-		}
-		a->priv = p;
-	}
-	return p;
-}
-
-static inline struct tcf_st *
-tcf_hash_create(u32 index, struct rtattr *est, struct tc_action *a, int size, int ovr, int bind)
-{
-	struct tcf_st *p = NULL;
-
-	p = kmalloc(size, GFP_KERNEL);
-	if (p == NULL)
-		return p;
-
-	memset(p, 0, size);
-	p->refcnt = 1;
-
-	if (bind) {
-		p->bindcnt = 1;
-	}
-
-	spin_lock_init(&p->lock);
-	p->stats_lock = &p->lock;
-	p->index = index ? : tcf_hash_new_index();
-	p->tm.install = jiffies;
-	p->tm.lastuse = jiffies;
-#ifdef CONFIG_NET_ESTIMATOR
-	if (est)
-		gen_new_estimator(&p->bstats, &p->rate_est, p->stats_lock, est);
-#endif
-	a->priv = (void *) p;
-	return p;
-}
-
-static inline void tcf_hash_insert(struct tcf_st *p)
-{
-	unsigned h = tcf_hash(p->index);
-
-	write_lock_bh(&tcf_t_lock);
-	p->next = tcf_ht[h];
-	tcf_ht[h] = p;
-	write_unlock_bh(&tcf_t_lock);
-}
-
-#endif
-
-#endif
diff --git a/include/net/tc_act/tc_defact.h b/include/net/tc_act/tc_defact.h
index 463aa671f95d..65f024b80958 100644
--- a/include/net/tc_act/tc_defact.h
+++ b/include/net/tc_act/tc_defact.h
@@ -3,11 +3,12 @@
 
 #include <net/act_api.h>
 
-struct tcf_defact
-{
-	tca_gen(defact);
-	u32     datalen;
-	void    *defdata;
+struct tcf_defact {
+	struct tcf_common	common;
+	u32     		tcfd_datalen;
+	void    		*tcfd_defdata;
 };
+#define to_defact(pc) \
+	container_of(pc, struct tcf_defact, common)
 
-#endif
+#endif /* __NET_TC_DEF_H */
diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h
index 59f0d9628ad1..9e3f6767b80e 100644
--- a/include/net/tc_act/tc_gact.h
+++ b/include/net/tc_act/tc_gact.h
@@ -3,15 +3,15 @@
 
 #include <net/act_api.h>
 
-struct tcf_gact
-{
-        tca_gen(gact);
+struct tcf_gact {
+	struct tcf_common	common;
 #ifdef CONFIG_GACT_PROB
-        u16                 ptype;
-        u16                 pval;
-        int                 paction;
+        u16			tcfg_ptype;
+        u16			tcfg_pval;
+        int			tcfg_paction;
 #endif
-                                                                                
 };
-                                                                                
-#endif
+#define to_gact(pc) \
+	container_of(pc, struct tcf_gact, common)
+
+#endif /* __NET_TC_GACT_H */
diff --git a/include/net/tc_act/tc_ipt.h b/include/net/tc_act/tc_ipt.h
index cb37ad08427f..f7d25dfcc4b7 100644
--- a/include/net/tc_act/tc_ipt.h
+++ b/include/net/tc_act/tc_ipt.h
@@ -5,12 +5,13 @@
 
 struct xt_entry_target;
 
-struct tcf_ipt
-{
-	tca_gen(ipt);
-	u32 hook;
-	char *tname;
-	struct xt_entry_target *t;
+struct tcf_ipt {
+	struct tcf_common	common;
+	u32			tcfi_hook;
+	char			*tcfi_tname;
+	struct xt_entry_target	*tcfi_t;
 };
+#define to_ipt(pc) \
+	container_of(pc, struct tcf_ipt, common)
 
-#endif
+#endif /* __NET_TC_IPT_H */
diff --git a/include/net/tc_act/tc_mirred.h b/include/net/tc_act/tc_mirred.h
index b5c32f65c12c..ceac661cdfd5 100644
--- a/include/net/tc_act/tc_mirred.h
+++ b/include/net/tc_act/tc_mirred.h
@@ -3,13 +3,14 @@
 
 #include <net/act_api.h>
 
-struct tcf_mirred
-{
-	tca_gen(mirred);
-	int eaction;
-	int ifindex;
-	int ok_push;
-	struct net_device *dev;
+struct tcf_mirred {
+	struct tcf_common	common;
+	int			tcfm_eaction;
+	int			tcfm_ifindex;
+	int			tcfm_ok_push;
+	struct net_device	*tcfm_dev;
 };
+#define to_mirred(pc) \
+	container_of(pc, struct tcf_mirred, common)
 
-#endif
+#endif /* __NET_TC_MIR_H */
diff --git a/include/net/tc_act/tc_pedit.h b/include/net/tc_act/tc_pedit.h
index eb21689d759d..e6f6e15956f5 100644
--- a/include/net/tc_act/tc_pedit.h
+++ b/include/net/tc_act/tc_pedit.h
@@ -3,12 +3,13 @@
 
 #include <net/act_api.h>
 
-struct tcf_pedit
-{
-	tca_gen(pedit);
-	unsigned char           nkeys;
-	unsigned char           flags;
-	struct tc_pedit_key     *keys;
+struct tcf_pedit {
+	struct tcf_common	common;
+	unsigned char		tcfp_nkeys;
+	unsigned char		tcfp_flags;
+	struct tc_pedit_key	*tcfp_keys;
 };
+#define to_pedit(pc) \
+	container_of(pc, struct tcf_pedit, common)
 
-#endif
+#endif /* __NET_TC_PED_H */
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 6990747d6d5a..835070e9169c 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -33,16 +33,230 @@
 #include <net/sch_generic.h>
 #include <net/act_api.h>
 
-#if 0 /* control */
-#define DPRINTK(format, args...) printk(KERN_DEBUG format, ##args)
-#else
-#define DPRINTK(format, args...)
+void tcf_hash_destroy(struct tcf_common *p, struct tcf_hashinfo *hinfo)
+{
+	unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask);
+	struct tcf_common **p1p;
+
+	for (p1p = &hinfo->htab[h]; *p1p; p1p = &(*p1p)->tcfc_next) {
+		if (*p1p == p) {
+			write_lock_bh(hinfo->lock);
+			*p1p = p->tcfc_next;
+			write_unlock_bh(hinfo->lock);
+#ifdef CONFIG_NET_ESTIMATOR
+			gen_kill_estimator(&p->tcfc_bstats,
+					   &p->tcfc_rate_est);
 #endif
-#if 0 /* data */
-#define D2PRINTK(format, args...) printk(KERN_DEBUG format, ##args)
-#else
-#define D2PRINTK(format, args...)
+			kfree(p);
+			return;
+		}
+	}
+	BUG_TRAP(0);
+}
+EXPORT_SYMBOL(tcf_hash_destroy);
+
+int tcf_hash_release(struct tcf_common *p, int bind,
+		     struct tcf_hashinfo *hinfo)
+{
+	int ret = 0;
+
+	if (p) {
+		if (bind)
+			p->tcfc_bindcnt--;
+
+		p->tcfc_refcnt--;
+	       	if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) {
+			tcf_hash_destroy(p, hinfo);
+			ret = 1;
+		}
+	}
+	return ret;
+}
+EXPORT_SYMBOL(tcf_hash_release);
+
+static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb,
+			   struct tc_action *a, struct tcf_hashinfo *hinfo)
+{
+	struct tcf_common *p;
+	int err = 0, index = -1,i = 0, s_i = 0, n_i = 0;
+	struct rtattr *r ;
+
+	read_lock(hinfo->lock);
+
+	s_i = cb->args[0];
+
+	for (i = 0; i < (hinfo->hmask + 1); i++) {
+		p = hinfo->htab[tcf_hash(i, hinfo->hmask)];
+
+		for (; p; p = p->tcfc_next) {
+			index++;
+			if (index < s_i)
+				continue;
+			a->priv = p;
+			a->order = n_i;
+			r = (struct rtattr*) skb->tail;
+			RTA_PUT(skb, a->order, 0, NULL);
+			err = tcf_action_dump_1(skb, a, 0, 0);
+			if (err < 0) {
+				index--;
+				skb_trim(skb, (u8*)r - skb->data);
+				goto done;
+			}
+			r->rta_len = skb->tail - (u8*)r;
+			n_i++;
+			if (n_i >= TCA_ACT_MAX_PRIO)
+				goto done;
+		}
+	}
+done:
+	read_unlock(hinfo->lock);
+	if (n_i)
+		cb->args[0] += n_i;
+	return n_i;
+
+rtattr_failure:
+	skb_trim(skb, (u8*)r - skb->data);
+	goto done;
+}
+
+static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a,
+			  struct tcf_hashinfo *hinfo)
+{
+	struct tcf_common *p, *s_p;
+	struct rtattr *r ;
+	int i= 0, n_i = 0;
+
+	r = (struct rtattr*) skb->tail;
+	RTA_PUT(skb, a->order, 0, NULL);
+	RTA_PUT(skb, TCA_KIND, IFNAMSIZ, a->ops->kind);
+	for (i = 0; i < (hinfo->hmask + 1); i++) {
+		p = hinfo->htab[tcf_hash(i, hinfo->hmask)];
+
+		while (p != NULL) {
+			s_p = p->tcfc_next;
+			if (ACT_P_DELETED == tcf_hash_release(p, 0, hinfo))
+				 module_put(a->ops->owner);
+			n_i++;
+			p = s_p;
+		}
+	}
+	RTA_PUT(skb, TCA_FCNT, 4, &n_i);
+	r->rta_len = skb->tail - (u8*)r;
+
+	return n_i;
+rtattr_failure:
+	skb_trim(skb, (u8*)r - skb->data);
+	return -EINVAL;
+}
+
+int tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb,
+		       int type, struct tc_action *a)
+{
+	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+
+	if (type == RTM_DELACTION) {
+		return tcf_del_walker(skb, a, hinfo);
+	} else if (type == RTM_GETACTION) {
+		return tcf_dump_walker(skb, cb, a, hinfo);
+	} else {
+		printk("tcf_generic_walker: unknown action %d\n", type);
+		return -EINVAL;
+	}
+}
+EXPORT_SYMBOL(tcf_generic_walker);
+
+struct tcf_common *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo)
+{
+	struct tcf_common *p;
+
+	read_lock(hinfo->lock);
+	for (p = hinfo->htab[tcf_hash(index, hinfo->hmask)]; p;
+	     p = p->tcfc_next) {
+		if (p->tcfc_index == index)
+			break;
+	}
+	read_unlock(hinfo->lock);
+
+	return p;
+}
+EXPORT_SYMBOL(tcf_hash_lookup);
+
+u32 tcf_hash_new_index(u32 *idx_gen, struct tcf_hashinfo *hinfo)
+{
+	u32 val = *idx_gen;
+
+	do {
+		if (++val == 0)
+			val = 1;
+	} while (tcf_hash_lookup(val, hinfo));
+
+	return (*idx_gen = val);
+}
+EXPORT_SYMBOL(tcf_hash_new_index);
+
+int tcf_hash_search(struct tc_action *a, u32 index)
+{
+	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+	struct tcf_common *p = tcf_hash_lookup(index, hinfo);
+
+	if (p) {
+		a->priv = p;
+		return 1;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(tcf_hash_search);
+
+struct tcf_common *tcf_hash_check(u32 index, struct tc_action *a, int bind,
+				  struct tcf_hashinfo *hinfo)
+{
+	struct tcf_common *p = NULL;
+	if (index && (p = tcf_hash_lookup(index, hinfo)) != NULL) {
+		if (bind) {
+			p->tcfc_bindcnt++;
+			p->tcfc_refcnt++;
+		}
+		a->priv = p;
+	}
+	return p;
+}
+EXPORT_SYMBOL(tcf_hash_check);
+
+struct tcf_common *tcf_hash_create(u32 index, struct rtattr *est, struct tc_action *a, int size, int bind, u32 *idx_gen, struct tcf_hashinfo *hinfo)
+{
+	struct tcf_common *p = kzalloc(size, GFP_KERNEL);
+
+	if (unlikely(!p))
+		return p;
+	p->tcfc_refcnt = 1;
+	if (bind)
+		p->tcfc_bindcnt = 1;
+
+	spin_lock_init(&p->tcfc_lock);
+	p->tcfc_stats_lock = &p->tcfc_lock;
+	p->tcfc_index = index ? index : tcf_hash_new_index(idx_gen, hinfo);
+	p->tcfc_tm.install = jiffies;
+	p->tcfc_tm.lastuse = jiffies;
+#ifdef CONFIG_NET_ESTIMATOR
+	if (est)
+		gen_new_estimator(&p->tcfc_bstats, &p->tcfc_rate_est,
+				  p->tcfc_stats_lock, est);
 #endif
+	a->priv = (void *) p;
+	return p;
+}
+EXPORT_SYMBOL(tcf_hash_create);
+
+void tcf_hash_insert(struct tcf_common *p, struct tcf_hashinfo *hinfo)
+{
+	unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask);
+
+	write_lock_bh(hinfo->lock);
+	p->tcfc_next = hinfo->htab[h];
+	hinfo->htab[h] = p;
+	write_unlock_bh(hinfo->lock);
+}
+EXPORT_SYMBOL(tcf_hash_insert);
 
 static struct tc_action_ops *act_base = NULL;
 static DEFINE_RWLOCK(act_mod_lock);
@@ -155,9 +369,6 @@ int tcf_action_exec(struct sk_buff *skb, struct tc_action *act,
 
 	if (skb->tc_verd & TC_NCLS) {
 		skb->tc_verd = CLR_TC_NCLS(skb->tc_verd);
-		D2PRINTK("(%p)tcf_action_exec: cleared TC_NCLS in %s out %s\n",
-		         skb, skb->input_dev ? skb->input_dev->name : "xxx",
-		         skb->dev->name);
 		ret = TC_ACT_OK;
 		goto exec_done;
 	}
@@ -187,8 +398,6 @@ void tcf_action_destroy(struct tc_action *act, int bind)
 
 	for (a = act; a; a = act) {
 		if (a->ops && a->ops->cleanup) {
-			DPRINTK("tcf_action_destroy destroying %p next %p\n",
-			        a, a->next);
 			if (a->ops->cleanup(a, bind) == ACT_P_DELETED)
 				module_put(a->ops->owner);
 			act = act->next;
@@ -331,7 +540,6 @@ struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est,
 	if (*err != ACT_P_CREATED)
 		module_put(a_o->owner);
 	a->ops = a_o;
-	DPRINTK("tcf_action_init_1: successfull %s\n", act_name);
 
 	*err = 0;
 	return a;
@@ -392,12 +600,12 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
 	if (compat_mode) {
 		if (a->type == TCA_OLD_COMPAT)
 			err = gnet_stats_start_copy_compat(skb, 0,
-				TCA_STATS, TCA_XSTATS, h->stats_lock, &d);
+				TCA_STATS, TCA_XSTATS, h->tcf_stats_lock, &d);
 		else
 			return 0;
 	} else
 		err = gnet_stats_start_copy(skb, TCA_ACT_STATS,
-			h->stats_lock, &d);
+			h->tcf_stats_lock, &d);
 
 	if (err < 0)
 		goto errout;
@@ -406,11 +614,11 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,
 		if (a->ops->get_stats(skb, a) < 0)
 			goto errout;
 
-	if (gnet_stats_copy_basic(&d, &h->bstats) < 0 ||
+	if (gnet_stats_copy_basic(&d, &h->tcf_bstats) < 0 ||
 #ifdef CONFIG_NET_ESTIMATOR
-	    gnet_stats_copy_rate_est(&d, &h->rate_est) < 0 ||
+	    gnet_stats_copy_rate_est(&d, &h->tcf_rate_est) < 0 ||
 #endif
-	    gnet_stats_copy_queue(&d, &h->qstats) < 0)
+	    gnet_stats_copy_queue(&d, &h->tcf_qstats) < 0)
 		goto errout;
 
 	if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c
index e75a147ad60f..6cff56696a81 100644
--- a/net/sched/act_gact.c
+++ b/net/sched/act_gact.c
@@ -34,48 +34,43 @@
 #include <linux/tc_act/tc_gact.h>
 #include <net/tc_act/tc_gact.h>
 
-/* use generic hash table */
-#define MY_TAB_SIZE	16
-#define MY_TAB_MASK	15
-
-static u32 idx_gen;
-static struct tcf_gact *tcf_gact_ht[MY_TAB_SIZE];
+#define GACT_TAB_MASK	15
+static struct tcf_common *tcf_gact_ht[GACT_TAB_MASK + 1];
+static u32 gact_idx_gen;
 static DEFINE_RWLOCK(gact_lock);
 
-/* ovewrride the defaults */
-#define tcf_st		tcf_gact
-#define tc_st		tc_gact
-#define tcf_t_lock	gact_lock
-#define tcf_ht		tcf_gact_ht
-
-#define CONFIG_NET_ACT_INIT 1
-#include <net/pkt_act.h>
+static struct tcf_hashinfo gact_hash_info = {
+	.htab	=	tcf_gact_ht,
+	.hmask	=	GACT_TAB_MASK,
+	.lock	=	&gact_lock,
+};
 
 #ifdef CONFIG_GACT_PROB
-static int gact_net_rand(struct tcf_gact *p)
+static int gact_net_rand(struct tcf_gact *gact)
 {
-	if (net_random()%p->pval)
-		return p->action;
-	return p->paction;
+	if (net_random() % gact->tcfg_pval)
+		return gact->tcf_action;
+	return gact->tcfg_paction;
 }
 
-static int gact_determ(struct tcf_gact *p)
+static int gact_determ(struct tcf_gact *gact)
 {
-	if (p->bstats.packets%p->pval)
-		return p->action;
-	return p->paction;
+	if (gact->tcf_bstats.packets % gact->tcfg_pval)
+		return gact->tcf_action;
+	return gact->tcfg_paction;
 }
 
-typedef int (*g_rand)(struct tcf_gact *p);
+typedef int (*g_rand)(struct tcf_gact *gact);
 static g_rand gact_rand[MAX_RAND]= { NULL, gact_net_rand, gact_determ };
-#endif
+#endif /* CONFIG_GACT_PROB */
 
 static int tcf_gact_init(struct rtattr *rta, struct rtattr *est,
                          struct tc_action *a, int ovr, int bind)
 {
 	struct rtattr *tb[TCA_GACT_MAX];
 	struct tc_gact *parm;
-	struct tcf_gact *p;
+	struct tcf_gact *gact;
+	struct tcf_common *pc;
 	int ret = 0;
 
 	if (rta == NULL || rtattr_parse_nested(tb, TCA_GACT_MAX, rta) < 0)
@@ -94,105 +89,106 @@ static int tcf_gact_init(struct rtattr *rta, struct rtattr *est,
 		return -EOPNOTSUPP;
 #endif
 
-	p = tcf_hash_check(parm->index, a, ovr, bind);
-	if (p == NULL) {
-		p = tcf_hash_create(parm->index, est, a, sizeof(*p), ovr, bind);
-		if (p == NULL)
+	pc = tcf_hash_check(parm->index, a, bind, &gact_hash_info);
+	if (!pc) {
+		pc = tcf_hash_create(parm->index, est, a, sizeof(*gact),
+				     bind, &gact_idx_gen, &gact_hash_info);
+		if (unlikely(!pc))
 			return -ENOMEM;
 		ret = ACT_P_CREATED;
 	} else {
 		if (!ovr) {
-			tcf_hash_release(p, bind);
+			tcf_hash_release(pc, bind, &gact_hash_info);
 			return -EEXIST;
 		}
 	}
 
-	spin_lock_bh(&p->lock);
-	p->action = parm->action;
+	gact = to_gact(pc);
+
+	spin_lock_bh(&gact->tcf_lock);
+	gact->tcf_action = parm->action;
 #ifdef CONFIG_GACT_PROB
 	if (tb[TCA_GACT_PROB-1] != NULL) {
 		struct tc_gact_p *p_parm = RTA_DATA(tb[TCA_GACT_PROB-1]);
-		p->paction = p_parm->paction;
-		p->pval    = p_parm->pval;
-		p->ptype   = p_parm->ptype;
+		gact->tcfg_paction = p_parm->paction;
+		gact->tcfg_pval    = p_parm->pval;
+		gact->tcfg_ptype   = p_parm->ptype;
 	}
 #endif
-	spin_unlock_bh(&p->lock);
+	spin_unlock_bh(&gact->tcf_lock);
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(p);
+		tcf_hash_insert(pc, &gact_hash_info);
 	return ret;
 }
 
-static int
-tcf_gact_cleanup(struct tc_action *a, int bind)
+static int tcf_gact_cleanup(struct tc_action *a, int bind)
 {
-	struct tcf_gact *p = PRIV(a, gact);
+	struct tcf_gact *gact = a->priv;
 
-	if (p != NULL)
-		return tcf_hash_release(p, bind);
+	if (gact)
+		return tcf_hash_release(&gact->common, bind, &gact_hash_info);
 	return 0;
 }
 
-static int
-tcf_gact(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
+static int tcf_gact(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 {
-	struct tcf_gact *p = PRIV(a, gact);
+	struct tcf_gact *gact = a->priv;
 	int action = TC_ACT_SHOT;
 
-	spin_lock(&p->lock);
+	spin_lock(&gact->tcf_lock);
 #ifdef CONFIG_GACT_PROB
-	if (p->ptype && gact_rand[p->ptype] != NULL)
-		action = gact_rand[p->ptype](p);
+	if (gact->tcfg_ptype && gact_rand[gact->tcfg_ptype] != NULL)
+		action = gact_rand[gact->tcfg_ptype](gact);
 	else
-		action = p->action;
+		action = gact->tcf_action;
 #else
-	action = p->action;
+	action = gact->tcf_action;
 #endif
-	p->bstats.bytes += skb->len;
-	p->bstats.packets++;
+	gact->tcf_bstats.bytes += skb->len;
+	gact->tcf_bstats.packets++;
 	if (action == TC_ACT_SHOT)
-		p->qstats.drops++;
-	p->tm.lastuse = jiffies;
-	spin_unlock(&p->lock);
+		gact->tcf_qstats.drops++;
+	gact->tcf_tm.lastuse = jiffies;
+	spin_unlock(&gact->tcf_lock);
 
 	return action;
 }
 
-static int
-tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
+static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 {
 	unsigned char *b = skb->tail;
 	struct tc_gact opt;
-	struct tcf_gact *p = PRIV(a, gact);
+	struct tcf_gact *gact = a->priv;
 	struct tcf_t t;
 
-	opt.index = p->index;
-	opt.refcnt = p->refcnt - ref;
-	opt.bindcnt = p->bindcnt - bind;
-	opt.action = p->action;
+	opt.index = gact->tcf_index;
+	opt.refcnt = gact->tcf_refcnt - ref;
+	opt.bindcnt = gact->tcf_bindcnt - bind;
+	opt.action = gact->tcf_action;
 	RTA_PUT(skb, TCA_GACT_PARMS, sizeof(opt), &opt);
 #ifdef CONFIG_GACT_PROB
-	if (p->ptype) {
+	if (gact->tcfg_ptype) {
 		struct tc_gact_p p_opt;
-		p_opt.paction = p->paction;
-		p_opt.pval = p->pval;
-		p_opt.ptype = p->ptype;
+		p_opt.paction = gact->tcfg_paction;
+		p_opt.pval = gact->tcfg_pval;
+		p_opt.ptype = gact->tcfg_ptype;
 		RTA_PUT(skb, TCA_GACT_PROB, sizeof(p_opt), &p_opt);
 	}
 #endif
-	t.install = jiffies_to_clock_t(jiffies - p->tm.install);
-	t.lastuse = jiffies_to_clock_t(jiffies - p->tm.lastuse);
-	t.expires = jiffies_to_clock_t(p->tm.expires);
+	t.install = jiffies_to_clock_t(jiffies - gact->tcf_tm.install);
+	t.lastuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.lastuse);
+	t.expires = jiffies_to_clock_t(gact->tcf_tm.expires);
 	RTA_PUT(skb, TCA_GACT_TM, sizeof(t), &t);
 	return skb->len;
 
-      rtattr_failure:
+rtattr_failure:
 	skb_trim(skb, b - skb->data);
 	return -1;
 }
 
 static struct tc_action_ops act_gact_ops = {
 	.kind		=	"gact",
+	.hinfo		=	&gact_hash_info,
 	.type		=	TCA_ACT_GACT,
 	.capab		=	TCA_CAP_NONE,
 	.owner		=	THIS_MODULE,
@@ -208,8 +204,7 @@ MODULE_AUTHOR("Jamal Hadi Salim(2002-4)");
 MODULE_DESCRIPTION("Generic Classifier actions");
 MODULE_LICENSE("GPL");
 
-static int __init
-gact_init_module(void)
+static int __init gact_init_module(void)
 {
 #ifdef CONFIG_GACT_PROB
 	printk("GACT probability on\n");
@@ -219,8 +214,7 @@ gact_init_module(void)
 	return tcf_register_action(&act_gact_ops);
 }
 
-static void __exit
-gact_cleanup_module(void)
+static void __exit gact_cleanup_module(void)
 {
 	tcf_unregister_action(&act_gact_ops);
 }
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index d799e01248c4..224c078a398e 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -38,25 +38,19 @@
 
 #include <linux/netfilter_ipv4/ip_tables.h>
 
-/* use generic hash table */
-#define MY_TAB_SIZE     16
-#define MY_TAB_MASK     15
 
-static u32 idx_gen;
-static struct tcf_ipt *tcf_ipt_ht[MY_TAB_SIZE];
-/* ipt hash table lock */
+#define IPT_TAB_MASK     15
+static struct tcf_common *tcf_ipt_ht[IPT_TAB_MASK + 1];
+static u32 ipt_idx_gen;
 static DEFINE_RWLOCK(ipt_lock);
 
-/* ovewrride the defaults */
-#define tcf_st		tcf_ipt
-#define tcf_t_lock	ipt_lock
-#define tcf_ht		tcf_ipt_ht
-
-#define CONFIG_NET_ACT_INIT
-#include <net/pkt_act.h>
+static struct tcf_hashinfo ipt_hash_info = {
+	.htab	=	tcf_ipt_ht,
+	.hmask	=	IPT_TAB_MASK,
+	.lock	=	&ipt_lock,
+};
 
-static int
-ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook)
+static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook)
 {
 	struct ipt_target *target;
 	int ret = 0;
@@ -65,7 +59,6 @@ ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook)
 	if (!target)
 		return -ENOENT;
 
-	DPRINTK("ipt_init_target: found %s\n", target->name);
 	t->u.kernel.target = target;
 
 	ret = xt_check_target(target, AF_INET, t->u.target_size - sizeof(*t),
@@ -78,8 +71,6 @@ ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook)
 		    			       t->u.kernel.target, t->data,
 					       t->u.target_size - sizeof(*t),
 					       hook)) {
-		DPRINTK("ipt_init_target: check failed for `%s'.\n",
-			t->u.kernel.target->name);
 		module_put(t->u.kernel.target->me);
 		ret = -EINVAL;
 	}
@@ -87,8 +78,7 @@ ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook)
 	return ret;
 }
 
-static void
-ipt_destroy_target(struct ipt_entry_target *t)
+static void ipt_destroy_target(struct ipt_entry_target *t)
 {
 	if (t->u.kernel.target->destroy)
 		t->u.kernel.target->destroy(t->u.kernel.target, t->data,
@@ -96,31 +86,30 @@ ipt_destroy_target(struct ipt_entry_target *t)
         module_put(t->u.kernel.target->me);
 }
 
-static int
-tcf_ipt_release(struct tcf_ipt *p, int bind)
+static int tcf_ipt_release(struct tcf_ipt *ipt, int bind)
 {
 	int ret = 0;
-	if (p) {
+	if (ipt) {
 		if (bind)
-			p->bindcnt--;
-		p->refcnt--;
-		if (p->bindcnt <= 0 && p->refcnt <= 0) {
-			ipt_destroy_target(p->t);
-			kfree(p->tname);
-			kfree(p->t);
-			tcf_hash_destroy(p);
+			ipt->tcf_bindcnt--;
+		ipt->tcf_refcnt--;
+		if (ipt->tcf_bindcnt <= 0 && ipt->tcf_refcnt <= 0) {
+			ipt_destroy_target(ipt->tcfi_t);
+			kfree(ipt->tcfi_tname);
+			kfree(ipt->tcfi_t);
+			tcf_hash_destroy(&ipt->common, &ipt_hash_info);
 			ret = ACT_P_DELETED;
 		}
 	}
 	return ret;
 }
 
-static int
-tcf_ipt_init(struct rtattr *rta, struct rtattr *est, struct tc_action *a,
-             int ovr, int bind)
+static int tcf_ipt_init(struct rtattr *rta, struct rtattr *est,
+			struct tc_action *a, int ovr, int bind)
 {
 	struct rtattr *tb[TCA_IPT_MAX];
-	struct tcf_ipt *p;
+	struct tcf_ipt *ipt;
+	struct tcf_common *pc;
 	struct ipt_entry_target *td, *t;
 	char *tname;
 	int ret = 0, err;
@@ -144,49 +133,51 @@ tcf_ipt_init(struct rtattr *rta, struct rtattr *est, struct tc_action *a,
 	    RTA_PAYLOAD(tb[TCA_IPT_INDEX-1]) >= sizeof(u32))
 		index = *(u32 *)RTA_DATA(tb[TCA_IPT_INDEX-1]);
 
-	p = tcf_hash_check(index, a, ovr, bind);
-	if (p == NULL) {
-		p = tcf_hash_create(index, est, a, sizeof(*p), ovr, bind);
-		if (p == NULL)
+	pc = tcf_hash_check(index, a, bind, &ipt_hash_info);
+	if (!pc) {
+		pc = tcf_hash_create(index, est, a, sizeof(*ipt), bind,
+				     &ipt_idx_gen, &ipt_hash_info);
+		if (unlikely(!pc))
 			return -ENOMEM;
 		ret = ACT_P_CREATED;
 	} else {
 		if (!ovr) {
-			tcf_ipt_release(p, bind);
+			tcf_ipt_release(to_ipt(pc), bind);
 			return -EEXIST;
 		}
 	}
+	ipt = to_ipt(pc);
 
 	hook = *(u32 *)RTA_DATA(tb[TCA_IPT_HOOK-1]);
 
 	err = -ENOMEM;
 	tname = kmalloc(IFNAMSIZ, GFP_KERNEL);
-	if (tname == NULL)
+	if (unlikely(!tname))
 		goto err1;
 	if (tb[TCA_IPT_TABLE - 1] == NULL ||
 	    rtattr_strlcpy(tname, tb[TCA_IPT_TABLE-1], IFNAMSIZ) >= IFNAMSIZ)
 		strcpy(tname, "mangle");
 
 	t = kmalloc(td->u.target_size, GFP_KERNEL);
-	if (t == NULL)
+	if (unlikely(!t))
 		goto err2;
 	memcpy(t, td, td->u.target_size);
 
 	if ((err = ipt_init_target(t, tname, hook)) < 0)
 		goto err3;
 
-	spin_lock_bh(&p->lock);
+	spin_lock_bh(&ipt->tcf_lock);
 	if (ret != ACT_P_CREATED) {
-		ipt_destroy_target(p->t);
-		kfree(p->tname);
-		kfree(p->t);
+		ipt_destroy_target(ipt->tcfi_t);
+		kfree(ipt->tcfi_tname);
+		kfree(ipt->tcfi_t);
 	}
-	p->tname = tname;
-	p->t     = t;
-	p->hook  = hook;
-	spin_unlock_bh(&p->lock);
+	ipt->tcfi_tname = tname;
+	ipt->tcfi_t     = t;
+	ipt->tcfi_hook  = hook;
+	spin_unlock_bh(&ipt->tcf_lock);
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(p);
+		tcf_hash_insert(pc, &ipt_hash_info);
 	return ret;
 
 err3:
@@ -194,33 +185,32 @@ err3:
 err2:
 	kfree(tname);
 err1:
-	kfree(p);
+	kfree(pc);
 	return err;
 }
 
-static int
-tcf_ipt_cleanup(struct tc_action *a, int bind)
+static int tcf_ipt_cleanup(struct tc_action *a, int bind)
 {
-	struct tcf_ipt *p = PRIV(a, ipt);
-	return tcf_ipt_release(p, bind);
+	struct tcf_ipt *ipt = a->priv;
+	return tcf_ipt_release(ipt, bind);
 }
 
-static int
-tcf_ipt(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
+static int tcf_ipt(struct sk_buff *skb, struct tc_action *a,
+		   struct tcf_result *res)
 {
 	int ret = 0, result = 0;
-	struct tcf_ipt *p = PRIV(a, ipt);
+	struct tcf_ipt *ipt = a->priv;
 
 	if (skb_cloned(skb)) {
 		if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
 			return TC_ACT_UNSPEC;
 	}
 
-	spin_lock(&p->lock);
+	spin_lock(&ipt->tcf_lock);
 
-	p->tm.lastuse = jiffies;
-	p->bstats.bytes += skb->len;
-	p->bstats.packets++;
+	ipt->tcf_tm.lastuse = jiffies;
+	ipt->tcf_bstats.bytes += skb->len;
+	ipt->tcf_bstats.packets++;
 
 	/* yes, we have to worry about both in and out dev
 	 worry later - danger - this API seems to have changed
@@ -229,16 +219,17 @@ tcf_ipt(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 	/* iptables targets take a double skb pointer in case the skb
 	 * needs to be replaced. We don't own the skb, so this must not
 	 * happen. The pskb_expand_head above should make sure of this */
-	ret = p->t->u.kernel.target->target(&skb, skb->dev, NULL, p->hook,
-					    p->t->u.kernel.target, p->t->data,
-					    NULL);
+	ret = ipt->tcfi_t->u.kernel.target->target(&skb, skb->dev, NULL,
+						   ipt->tcfi_hook,
+						   ipt->tcfi_t->u.kernel.target,
+						   ipt->tcfi_t->data, NULL);
 	switch (ret) {
 	case NF_ACCEPT:
 		result = TC_ACT_OK;
 		break;
 	case NF_DROP:
 		result = TC_ACT_SHOT;
-		p->qstats.drops++;
+		ipt->tcf_qstats.drops++;
 		break;
 	case IPT_CONTINUE:
 		result = TC_ACT_PIPE;
@@ -249,53 +240,46 @@ tcf_ipt(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 		result = TC_POLICE_OK;
 		break;
 	}
-	spin_unlock(&p->lock);
+	spin_unlock(&ipt->tcf_lock);
 	return result;
 
 }
 
-static int
-tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
+static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 {
+	unsigned char *b = skb->tail;
+	struct tcf_ipt *ipt = a->priv;
 	struct ipt_entry_target *t;
 	struct tcf_t tm;
 	struct tc_cnt c;
-	unsigned char *b = skb->tail;
-	struct tcf_ipt *p = PRIV(a, ipt);
 
 	/* for simple targets kernel size == user size
 	** user name = target name
 	** for foolproof you need to not assume this
 	*/
 
-	t = kmalloc(p->t->u.user.target_size, GFP_ATOMIC);
-	if (t == NULL)
+	t = kmalloc(ipt->tcfi_t->u.user.target_size, GFP_ATOMIC);
+	if (unlikely(!t))
 		goto rtattr_failure;
 
-	c.bindcnt = p->bindcnt - bind;
-	c.refcnt = p->refcnt - ref;
-	memcpy(t, p->t, p->t->u.user.target_size);
-	strcpy(t->u.user.name, p->t->u.kernel.target->name);
-
-	DPRINTK("\ttcf_ipt_dump tablename %s length %d\n", p->tname,
-		strlen(p->tname));
-	DPRINTK("\tdump target name %s size %d size user %d "
-	        "data[0] %x data[1] %x\n", p->t->u.kernel.target->name,
-	        p->t->u.target_size, p->t->u.user.target_size,
-	        p->t->data[0], p->t->data[1]);
-	RTA_PUT(skb, TCA_IPT_TARG, p->t->u.user.target_size, t);
-	RTA_PUT(skb, TCA_IPT_INDEX, 4, &p->index);
-	RTA_PUT(skb, TCA_IPT_HOOK, 4, &p->hook);
+	c.bindcnt = ipt->tcf_bindcnt - bind;
+	c.refcnt = ipt->tcf_refcnt - ref;
+	memcpy(t, ipt->tcfi_t, ipt->tcfi_t->u.user.target_size);
+	strcpy(t->u.user.name, ipt->tcfi_t->u.kernel.target->name);
+
+	RTA_PUT(skb, TCA_IPT_TARG, ipt->tcfi_t->u.user.target_size, t);
+	RTA_PUT(skb, TCA_IPT_INDEX, 4, &ipt->tcf_index);
+	RTA_PUT(skb, TCA_IPT_HOOK, 4, &ipt->tcfi_hook);
 	RTA_PUT(skb, TCA_IPT_CNT, sizeof(struct tc_cnt), &c);
-	RTA_PUT(skb, TCA_IPT_TABLE, IFNAMSIZ, p->tname);
-	tm.install = jiffies_to_clock_t(jiffies - p->tm.install);
-	tm.lastuse = jiffies_to_clock_t(jiffies - p->tm.lastuse);
-	tm.expires = jiffies_to_clock_t(p->tm.expires);
+	RTA_PUT(skb, TCA_IPT_TABLE, IFNAMSIZ, ipt->tcfi_tname);
+	tm.install = jiffies_to_clock_t(jiffies - ipt->tcf_tm.install);
+	tm.lastuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.lastuse);
+	tm.expires = jiffies_to_clock_t(ipt->tcf_tm.expires);
 	RTA_PUT(skb, TCA_IPT_TM, sizeof (tm), &tm);
 	kfree(t);
 	return skb->len;
 
-      rtattr_failure:
+rtattr_failure:
 	skb_trim(skb, b - skb->data);
 	kfree(t);
 	return -1;
@@ -303,6 +287,7 @@ tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 
 static struct tc_action_ops act_ipt_ops = {
 	.kind		=	"ipt",
+	.hinfo		=	&ipt_hash_info,
 	.type		=	TCA_ACT_IPT,
 	.capab		=	TCA_CAP_NONE,
 	.owner		=	THIS_MODULE,
@@ -318,14 +303,12 @@ MODULE_AUTHOR("Jamal Hadi Salim(2002-4)");
 MODULE_DESCRIPTION("Iptables target actions");
 MODULE_LICENSE("GPL");
 
-static int __init
-ipt_init_module(void)
+static int __init ipt_init_module(void)
 {
 	return tcf_register_action(&act_ipt_ops);
 }
 
-static void __exit
-ipt_cleanup_module(void)
+static void __exit ipt_cleanup_module(void)
 {
 	tcf_unregister_action(&act_ipt_ops);
 }
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index fc562047ecc5..483897271f15 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -39,46 +39,39 @@
 #include <linux/etherdevice.h>
 #include <linux/if_arp.h>
 
-
-/* use generic hash table */
-#define MY_TAB_SIZE     8
-#define MY_TAB_MASK     (MY_TAB_SIZE - 1)
-static u32 idx_gen;
-static struct tcf_mirred *tcf_mirred_ht[MY_TAB_SIZE];
+#define MIRRED_TAB_MASK     7
+static struct tcf_common *tcf_mirred_ht[MIRRED_TAB_MASK + 1];
+static u32 mirred_idx_gen;
 static DEFINE_RWLOCK(mirred_lock);
 
-/* ovewrride the defaults */
-#define tcf_st		tcf_mirred
-#define tc_st		tc_mirred
-#define tcf_t_lock	mirred_lock
-#define tcf_ht		tcf_mirred_ht
-
-#define CONFIG_NET_ACT_INIT 1
-#include <net/pkt_act.h>
+static struct tcf_hashinfo mirred_hash_info = {
+	.htab	=	tcf_mirred_ht,
+	.hmask	=	MIRRED_TAB_MASK,
+	.lock	=	&mirred_lock,
+};
 
-static inline int
-tcf_mirred_release(struct tcf_mirred *p, int bind)
+static inline int tcf_mirred_release(struct tcf_mirred *m, int bind)
 {
-	if (p) {
+	if (m) {
 		if (bind)
-			p->bindcnt--;
-		p->refcnt--;
-		if(!p->bindcnt && p->refcnt <= 0) {
-			dev_put(p->dev);
-			tcf_hash_destroy(p);
+			m->tcf_bindcnt--;
+		m->tcf_refcnt--;
+		if(!m->tcf_bindcnt && m->tcf_refcnt <= 0) {
+			dev_put(m->tcfm_dev);
+			tcf_hash_destroy(&m->common, &mirred_hash_info);
 			return 1;
 		}
 	}
 	return 0;
 }
 
-static int
-tcf_mirred_init(struct rtattr *rta, struct rtattr *est, struct tc_action *a,
-                int ovr, int bind)
+static int tcf_mirred_init(struct rtattr *rta, struct rtattr *est,
+			   struct tc_action *a, int ovr, int bind)
 {
 	struct rtattr *tb[TCA_MIRRED_MAX];
 	struct tc_mirred *parm;
-	struct tcf_mirred *p;
+	struct tcf_mirred *m;
+	struct tcf_common *pc;
 	struct net_device *dev = NULL;
 	int ret = 0;
 	int ok_push = 0;
@@ -110,64 +103,62 @@ tcf_mirred_init(struct rtattr *rta, struct rtattr *est, struct tc_action *a,
 		}
 	}
 
-	p = tcf_hash_check(parm->index, a, ovr, bind);
-	if (p == NULL) {
+	pc = tcf_hash_check(parm->index, a, bind, &mirred_hash_info);
+	if (!pc) {
 		if (!parm->ifindex)
 			return -EINVAL;
-		p = tcf_hash_create(parm->index, est, a, sizeof(*p), ovr, bind);
-		if (p == NULL)
+		pc = tcf_hash_create(parm->index, est, a, sizeof(*m), bind,
+				     &mirred_idx_gen, &mirred_hash_info);
+		if (unlikely(!pc))
 			return -ENOMEM;
 		ret = ACT_P_CREATED;
 	} else {
 		if (!ovr) {
-			tcf_mirred_release(p, bind);
+			tcf_mirred_release(to_mirred(pc), bind);
 			return -EEXIST;
 		}
 	}
+	m = to_mirred(pc);
 
-	spin_lock_bh(&p->lock);
-	p->action = parm->action;
-	p->eaction = parm->eaction;
+	spin_lock_bh(&m->tcf_lock);
+	m->tcf_action = parm->action;
+	m->tcfm_eaction = parm->eaction;
 	if (parm->ifindex) {
-		p->ifindex = parm->ifindex;
+		m->tcfm_ifindex = parm->ifindex;
 		if (ret != ACT_P_CREATED)
-			dev_put(p->dev);
-		p->dev = dev;
+			dev_put(m->tcfm_dev);
+		m->tcfm_dev = dev;
 		dev_hold(dev);
-		p->ok_push = ok_push;
+		m->tcfm_ok_push = ok_push;
 	}
-	spin_unlock_bh(&p->lock);
+	spin_unlock_bh(&m->tcf_lock);
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(p);
+		tcf_hash_insert(pc, &mirred_hash_info);
 
-	DPRINTK("tcf_mirred_init index %d action %d eaction %d device %s "
-	        "ifindex %d\n", parm->index, parm->action, parm->eaction,
-	        dev->name, parm->ifindex);
 	return ret;
 }
 
-static int
-tcf_mirred_cleanup(struct tc_action *a, int bind)
+static int tcf_mirred_cleanup(struct tc_action *a, int bind)
 {
-	struct tcf_mirred *p = PRIV(a, mirred);
+	struct tcf_mirred *m = a->priv;
 
-	if (p != NULL)
-		return tcf_mirred_release(p, bind);
+	if (m)
+		return tcf_mirred_release(m, bind);
 	return 0;
 }
 
-static int
-tcf_mirred(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
+static int tcf_mirred(struct sk_buff *skb, struct tc_action *a,
+		      struct tcf_result *res)
 {
-	struct tcf_mirred *p = PRIV(a, mirred);
+	struct tcf_mirred *m = a->priv;
 	struct net_device *dev;
 	struct sk_buff *skb2 = NULL;
 	u32 at = G_TC_AT(skb->tc_verd);
 
-	spin_lock(&p->lock);
+	spin_lock(&m->tcf_lock);
 
-	dev = p->dev;
-	p->tm.lastuse = jiffies;
+	dev = m->tcfm_dev;
+	m->tcf_tm.lastuse = jiffies;
 
 	if (!(dev->flags&IFF_UP) ) {
 		if (net_ratelimit())
@@ -176,10 +167,10 @@ tcf_mirred(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 bad_mirred:
 		if (skb2 != NULL)
 			kfree_skb(skb2);
-		p->qstats.overlimits++;
-		p->bstats.bytes += skb->len;
-		p->bstats.packets++;
-		spin_unlock(&p->lock);
+		m->tcf_qstats.overlimits++;
+		m->tcf_bstats.bytes += skb->len;
+		m->tcf_bstats.packets++;
+		spin_unlock(&m->tcf_lock);
 		/* should we be asking for packet to be dropped?
 		 * may make sense for redirect case only
 		*/
@@ -189,59 +180,59 @@ bad_mirred:
 	skb2 = skb_clone(skb, GFP_ATOMIC);
 	if (skb2 == NULL)
 		goto bad_mirred;
-	if (p->eaction != TCA_EGRESS_MIRROR && p->eaction != TCA_EGRESS_REDIR) {
+	if (m->tcfm_eaction != TCA_EGRESS_MIRROR &&
+	    m->tcfm_eaction != TCA_EGRESS_REDIR) {
 		if (net_ratelimit())
-			printk("tcf_mirred unknown action %d\n", p->eaction);
+			printk("tcf_mirred unknown action %d\n",
+			       m->tcfm_eaction);
 		goto bad_mirred;
 	}
 
-	p->bstats.bytes += skb2->len;
-	p->bstats.packets++;
+	m->tcf_bstats.bytes += skb2->len;
+	m->tcf_bstats.packets++;
 	if (!(at & AT_EGRESS))
-		if (p->ok_push)
+		if (m->tcfm_ok_push)
 			skb_push(skb2, skb2->dev->hard_header_len);
 
 	/* mirror is always swallowed */
-	if (p->eaction != TCA_EGRESS_MIRROR)
+	if (m->tcfm_eaction != TCA_EGRESS_MIRROR)
 		skb2->tc_verd = SET_TC_FROM(skb2->tc_verd, at);
 
 	skb2->dev = dev;
 	skb2->input_dev = skb->dev;
 	dev_queue_xmit(skb2);
-	spin_unlock(&p->lock);
-	return p->action;
+	spin_unlock(&m->tcf_lock);
+	return m->tcf_action;
 }
 
-static int
-tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
+static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 {
 	unsigned char *b = skb->tail;
+	struct tcf_mirred *m = a->priv;
 	struct tc_mirred opt;
-	struct tcf_mirred *p = PRIV(a, mirred);
 	struct tcf_t t;
 
-	opt.index = p->index;
-	opt.action = p->action;
-	opt.refcnt = p->refcnt - ref;
-	opt.bindcnt = p->bindcnt - bind;
-	opt.eaction = p->eaction;
-	opt.ifindex = p->ifindex;
-	DPRINTK("tcf_mirred_dump index %d action %d eaction %d ifindex %d\n",
-	         p->index, p->action, p->eaction, p->ifindex);
+	opt.index = m->tcf_index;
+	opt.action = m->tcf_action;
+	opt.refcnt = m->tcf_refcnt - ref;
+	opt.bindcnt = m->tcf_bindcnt - bind;
+	opt.eaction = m->tcfm_eaction;
+	opt.ifindex = m->tcfm_ifindex;
 	RTA_PUT(skb, TCA_MIRRED_PARMS, sizeof(opt), &opt);
-	t.install = jiffies_to_clock_t(jiffies - p->tm.install);
-	t.lastuse = jiffies_to_clock_t(jiffies - p->tm.lastuse);
-	t.expires = jiffies_to_clock_t(p->tm.expires);
+	t.install = jiffies_to_clock_t(jiffies - m->tcf_tm.install);
+	t.lastuse = jiffies_to_clock_t(jiffies - m->tcf_tm.lastuse);
+	t.expires = jiffies_to_clock_t(m->tcf_tm.expires);
 	RTA_PUT(skb, TCA_MIRRED_TM, sizeof(t), &t);
 	return skb->len;
 
-      rtattr_failure:
+rtattr_failure:
 	skb_trim(skb, b - skb->data);
 	return -1;
 }
 
 static struct tc_action_ops act_mirred_ops = {
 	.kind		=	"mirred",
+	.hinfo		=	&mirred_hash_info,
 	.type		=	TCA_ACT_MIRRED,
 	.capab		=	TCA_CAP_NONE,
 	.owner		=	THIS_MODULE,
@@ -257,15 +248,13 @@ MODULE_AUTHOR("Jamal Hadi Salim(2002)");
 MODULE_DESCRIPTION("Device Mirror/redirect actions");
 MODULE_LICENSE("GPL");
 
-static int __init
-mirred_init_module(void)
+static int __init mirred_init_module(void)
 {
 	printk("Mirror/redirect action on\n");
 	return tcf_register_action(&act_mirred_ops);
 }
 
-static void __exit
-mirred_cleanup_module(void)
+static void __exit mirred_cleanup_module(void)
 {
 	tcf_unregister_action(&act_mirred_ops);
 }
diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index f257475e0e0c..8ac65c219b98 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -33,32 +33,25 @@
 #include <linux/tc_act/tc_pedit.h>
 #include <net/tc_act/tc_pedit.h>
 
-
-#define PEDIT_DEB 1
-
-/* use generic hash table */
-#define MY_TAB_SIZE     16
-#define MY_TAB_MASK     15
-static u32 idx_gen;
-static struct tcf_pedit *tcf_pedit_ht[MY_TAB_SIZE];
+#define PEDIT_TAB_MASK	15
+static struct tcf_common *tcf_pedit_ht[PEDIT_TAB_MASK + 1];
+static u32 pedit_idx_gen;
 static DEFINE_RWLOCK(pedit_lock);
 
-#define tcf_st		tcf_pedit
-#define tc_st		tc_pedit
-#define tcf_t_lock	pedit_lock
-#define tcf_ht		tcf_pedit_ht
-
-#define CONFIG_NET_ACT_INIT 1
-#include <net/pkt_act.h>
+static struct tcf_hashinfo pedit_hash_info = {
+	.htab	=	tcf_pedit_ht,
+	.hmask	=	PEDIT_TAB_MASK,
+	.lock	=	&pedit_lock,
+};
 
-static int
-tcf_pedit_init(struct rtattr *rta, struct rtattr *est, struct tc_action *a,
-               int ovr, int bind)
+static int tcf_pedit_init(struct rtattr *rta, struct rtattr *est,
+			  struct tc_action *a, int ovr, int bind)
 {
 	struct rtattr *tb[TCA_PEDIT_MAX];
 	struct tc_pedit *parm;
 	int ret = 0;
 	struct tcf_pedit *p;
+	struct tcf_common *pc;
 	struct tc_pedit_key *keys = NULL;
 	int ksize;
 
@@ -73,54 +66,56 @@ tcf_pedit_init(struct rtattr *rta, struct rtattr *est, struct tc_action *a,
 	if (RTA_PAYLOAD(tb[TCA_PEDIT_PARMS-1]) < sizeof(*parm) + ksize)
 		return -EINVAL;
 
-	p = tcf_hash_check(parm->index, a, ovr, bind);
-	if (p == NULL) {
+	pc = tcf_hash_check(parm->index, a, bind, &pedit_hash_info);
+	if (!pc) {
 		if (!parm->nkeys)
 			return -EINVAL;
-		p = tcf_hash_create(parm->index, est, a, sizeof(*p), ovr, bind);
-		if (p == NULL)
+		pc = tcf_hash_create(parm->index, est, a, sizeof(*p), bind,
+				     &pedit_idx_gen, &pedit_hash_info);
+		if (unlikely(!pc))
 			return -ENOMEM;
+		p = to_pedit(pc);
 		keys = kmalloc(ksize, GFP_KERNEL);
 		if (keys == NULL) {
-			kfree(p);
+			kfree(pc);
 			return -ENOMEM;
 		}
 		ret = ACT_P_CREATED;
 	} else {
+		p = to_pedit(pc);
 		if (!ovr) {
-			tcf_hash_release(p, bind);
+			tcf_hash_release(pc, bind, &pedit_hash_info);
 			return -EEXIST;
 		}
-		if (p->nkeys && p->nkeys != parm->nkeys) {
+		if (p->tcfp_nkeys && p->tcfp_nkeys != parm->nkeys) {
 			keys = kmalloc(ksize, GFP_KERNEL);
 			if (keys == NULL)
 				return -ENOMEM;
 		}
 	}
 
-	spin_lock_bh(&p->lock);
-	p->flags = parm->flags;
-	p->action = parm->action;
+	spin_lock_bh(&p->tcf_lock);
+	p->tcfp_flags = parm->flags;
+	p->tcf_action = parm->action;
 	if (keys) {
-		kfree(p->keys);
-		p->keys = keys;
-		p->nkeys = parm->nkeys;
+		kfree(p->tcfp_keys);
+		p->tcfp_keys = keys;
+		p->tcfp_nkeys = parm->nkeys;
 	}
-	memcpy(p->keys, parm->keys, ksize);
-	spin_unlock_bh(&p->lock);
+	memcpy(p->tcfp_keys, parm->keys, ksize);
+	spin_unlock_bh(&p->tcf_lock);
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(p);
+		tcf_hash_insert(pc, &pedit_hash_info);
 	return ret;
 }
 
-static int
-tcf_pedit_cleanup(struct tc_action *a, int bind)
+static int tcf_pedit_cleanup(struct tc_action *a, int bind)
 {
-	struct tcf_pedit *p = PRIV(a, pedit);
+	struct tcf_pedit *p = a->priv;
 
-	if (p != NULL) {
-		struct tc_pedit_key *keys = p->keys;
-		if (tcf_hash_release(p, bind)) {
+	if (p) {
+		struct tc_pedit_key *keys = p->tcfp_keys;
+		if (tcf_hash_release(&p->common, bind, &pedit_hash_info)) {
 			kfree(keys);
 			return 1;
 		}
@@ -128,30 +123,30 @@ tcf_pedit_cleanup(struct tc_action *a, int bind)
 	return 0;
 }
 
-static int
-tcf_pedit(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
+static int tcf_pedit(struct sk_buff *skb, struct tc_action *a,
+		     struct tcf_result *res)
 {
-	struct tcf_pedit *p = PRIV(a, pedit);
+	struct tcf_pedit *p = a->priv;
 	int i, munged = 0;
 	u8 *pptr;
 
 	if (!(skb->tc_verd & TC_OK2MUNGE)) {
 		/* should we set skb->cloned? */
 		if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) {
-			return p->action;
+			return p->tcf_action;
 		}
 	}
 
 	pptr = skb->nh.raw;
 
-	spin_lock(&p->lock);
+	spin_lock(&p->tcf_lock);
 
-	p->tm.lastuse = jiffies;
+	p->tcf_tm.lastuse = jiffies;
 
-	if (p->nkeys > 0) {
-		struct tc_pedit_key *tkey = p->keys;
+	if (p->tcfp_nkeys > 0) {
+		struct tc_pedit_key *tkey = p->tcfp_keys;
 
-		for (i = p->nkeys; i > 0; i--, tkey++) {
+		for (i = p->tcfp_nkeys; i > 0; i--, tkey++) {
 			u32 *ptr;
 			int offset = tkey->off;
 
@@ -169,7 +164,8 @@ tcf_pedit(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 				printk("offset must be on 32 bit boundaries\n");
 				goto bad;
 			}
-			if (skb->len < 0 || (offset > 0 && offset > skb->len)) {
+			if (skb->len < 0 ||
+			    (offset > 0 && offset > skb->len)) {
 				printk("offset %d cant exceed pkt length %d\n",
 				       offset, skb->len);
 				goto bad;
@@ -185,63 +181,47 @@ tcf_pedit(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 			skb->tc_verd = SET_TC_MUNGED(skb->tc_verd);
 		goto done;
 	} else {
-		printk("pedit BUG: index %d\n",p->index);
+		printk("pedit BUG: index %d\n", p->tcf_index);
 	}
 
 bad:
-	p->qstats.overlimits++;
+	p->tcf_qstats.overlimits++;
 done:
-	p->bstats.bytes += skb->len;
-	p->bstats.packets++;
-	spin_unlock(&p->lock);
-	return p->action;
+	p->tcf_bstats.bytes += skb->len;
+	p->tcf_bstats.packets++;
+	spin_unlock(&p->tcf_lock);
+	return p->tcf_action;
 }
 
-static int
-tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a,int bind, int ref)
+static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a,
+			  int bind, int ref)
 {
 	unsigned char *b = skb->tail;
+	struct tcf_pedit *p = a->priv;
 	struct tc_pedit *opt;
-	struct tcf_pedit *p = PRIV(a, pedit);
 	struct tcf_t t;
 	int s; 
 		
-	s = sizeof(*opt) + p->nkeys * sizeof(struct tc_pedit_key);
+	s = sizeof(*opt) + p->tcfp_nkeys * sizeof(struct tc_pedit_key);
 
 	/* netlink spinlocks held above us - must use ATOMIC */
 	opt = kzalloc(s, GFP_ATOMIC);
-	if (opt == NULL)
+	if (unlikely(!opt))
 		return -ENOBUFS;
 
-	memcpy(opt->keys, p->keys, p->nkeys * sizeof(struct tc_pedit_key));
-	opt->index = p->index;
-	opt->nkeys = p->nkeys;
-	opt->flags = p->flags;
-	opt->action = p->action;
-	opt->refcnt = p->refcnt - ref;
-	opt->bindcnt = p->bindcnt - bind;
-
-
-#ifdef PEDIT_DEB
-	{                
-		/* Debug - get rid of later */
-		int i;
-		struct tc_pedit_key *key = opt->keys;
-
-		for (i=0; i<opt->nkeys; i++, key++) {
-			printk( "\n key #%d",i);
-			printk( "  at %d: val %08x mask %08x",
-			(unsigned int)key->off,
-			(unsigned int)key->val,
-			(unsigned int)key->mask);
-		}
-	}
-#endif
+	memcpy(opt->keys, p->tcfp_keys,
+	       p->tcfp_nkeys * sizeof(struct tc_pedit_key));
+	opt->index = p->tcf_index;
+	opt->nkeys = p->tcfp_nkeys;
+	opt->flags = p->tcfp_flags;
+	opt->action = p->tcf_action;
+	opt->refcnt = p->tcf_refcnt - ref;
+	opt->bindcnt = p->tcf_bindcnt - bind;
 
 	RTA_PUT(skb, TCA_PEDIT_PARMS, s, opt);
-	t.install = jiffies_to_clock_t(jiffies - p->tm.install);
-	t.lastuse = jiffies_to_clock_t(jiffies - p->tm.lastuse);
-	t.expires = jiffies_to_clock_t(p->tm.expires);
+	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
+	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
+	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
 	RTA_PUT(skb, TCA_PEDIT_TM, sizeof(t), &t);
 	kfree(opt);
 	return skb->len;
@@ -252,9 +232,9 @@ rtattr_failure:
 	return -1;
 }
 
-static
-struct tc_action_ops act_pedit_ops = {
+static struct tc_action_ops act_pedit_ops = {
 	.kind		=	"pedit",
+	.hinfo		=	&pedit_hash_info,
 	.type		=	TCA_ACT_PEDIT,
 	.capab		=	TCA_CAP_NONE,
 	.owner		=	THIS_MODULE,
@@ -270,14 +250,12 @@ MODULE_AUTHOR("Jamal Hadi Salim(2002-4)");
 MODULE_DESCRIPTION("Generic Packet Editor actions");
 MODULE_LICENSE("GPL");
 
-static int __init
-pedit_init_module(void)
+static int __init pedit_init_module(void)
 {
 	return tcf_register_action(&act_pedit_ops);
 }
 
-static void __exit
-pedit_cleanup_module(void)
+static void __exit pedit_cleanup_module(void)
 {
 	tcf_unregister_action(&act_pedit_ops);
 }
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index da905d7b4b40..fed47b658837 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -32,43 +32,27 @@
 #include <net/sock.h>
 #include <net/act_api.h>
 
-#define L2T(p,L)   ((p)->R_tab->data[(L)>>(p)->R_tab->rate.cell_log])
-#define L2T_P(p,L) ((p)->P_tab->data[(L)>>(p)->P_tab->rate.cell_log])
-#define PRIV(a) ((struct tcf_police *) (a)->priv)
-
-/* use generic hash table */
-#define MY_TAB_SIZE     16
-#define MY_TAB_MASK     15
-static u32 idx_gen;
-static struct tcf_police *tcf_police_ht[MY_TAB_SIZE];
-/* Policer hash table lock */
-static DEFINE_RWLOCK(police_lock);
-
-/* Each policer is serialized by its individual spinlock */
+#define L2T(p,L)   ((p)->tcfp_R_tab->data[(L)>>(p)->tcfp_R_tab->rate.cell_log])
+#define L2T_P(p,L) ((p)->tcfp_P_tab->data[(L)>>(p)->tcfp_P_tab->rate.cell_log])
 
-static __inline__ unsigned tcf_police_hash(u32 index)
-{
-	return index&0xF;
-}
+#define POL_TAB_MASK     15
+static struct tcf_common *tcf_police_ht[POL_TAB_MASK + 1];
+static u32 police_idx_gen;
+static DEFINE_RWLOCK(police_lock);
 
-static __inline__ struct tcf_police * tcf_police_lookup(u32 index)
-{
-	struct tcf_police *p;
+static struct tcf_hashinfo police_hash_info = {
+	.htab	=	tcf_police_ht,
+	.hmask	=	POL_TAB_MASK,
+	.lock	=	&police_lock,
+};
 
-	read_lock(&police_lock);
-	for (p = tcf_police_ht[tcf_police_hash(index)]; p; p = p->next) {
-		if (p->index == index)
-			break;
-	}
-	read_unlock(&police_lock);
-	return p;
-}
+/* Each policer is serialized by its individual spinlock */
 
 #ifdef CONFIG_NET_CLS_ACT
 static int tcf_act_police_walker(struct sk_buff *skb, struct netlink_callback *cb,
                               int type, struct tc_action *a)
 {
-	struct tcf_police *p;
+	struct tcf_common *p;
 	int err = 0, index = -1, i = 0, s_i = 0, n_i = 0;
 	struct rtattr *r;
 
@@ -76,10 +60,10 @@ static int tcf_act_police_walker(struct sk_buff *skb, struct netlink_callback *c
 
 	s_i = cb->args[0];
 
-	for (i = 0; i < MY_TAB_SIZE; i++) {
-		p = tcf_police_ht[tcf_police_hash(i)];
+	for (i = 0; i < (POL_TAB_MASK + 1); i++) {
+		p = tcf_police_ht[tcf_hash(i, POL_TAB_MASK)];
 
-		for (; p; p = p->next) {
+		for (; p; p = p->tcfc_next) {
 			index++;
 			if (index < s_i)
 				continue;
@@ -110,48 +94,26 @@ rtattr_failure:
 	skb_trim(skb, (u8*)r - skb->data);
 	goto done;
 }
-
-static inline int
-tcf_act_police_hash_search(struct tc_action *a, u32 index)
-{
-	struct tcf_police *p = tcf_police_lookup(index);
-
-	if (p != NULL) {
-		a->priv = p;
-		return 1;
-	} else {
-		return 0;
-	}
-}
 #endif
 
-static inline u32 tcf_police_new_index(void)
-{
-	do {
-		if (++idx_gen == 0)
-			idx_gen = 1;
-	} while (tcf_police_lookup(idx_gen));
-
-	return idx_gen;
-}
-
 void tcf_police_destroy(struct tcf_police *p)
 {
-	unsigned h = tcf_police_hash(p->index);
-	struct tcf_police **p1p;
+	unsigned int h = tcf_hash(p->tcf_index, POL_TAB_MASK);
+	struct tcf_common **p1p;
 	
-	for (p1p = &tcf_police_ht[h]; *p1p; p1p = &(*p1p)->next) {
-		if (*p1p == p) {
+	for (p1p = &tcf_police_ht[h]; *p1p; p1p = &(*p1p)->tcfc_next) {
+		if (*p1p == &p->common) {
 			write_lock_bh(&police_lock);
-			*p1p = p->next;
+			*p1p = p->tcf_next;
 			write_unlock_bh(&police_lock);
 #ifdef CONFIG_NET_ESTIMATOR
-			gen_kill_estimator(&p->bstats, &p->rate_est);
+			gen_kill_estimator(&p->tcf_bstats,
+					   &p->tcf_rate_est);
 #endif
-			if (p->R_tab)
-				qdisc_put_rtab(p->R_tab);
-			if (p->P_tab)
-				qdisc_put_rtab(p->P_tab);
+			if (p->tcfp_R_tab)
+				qdisc_put_rtab(p->tcfp_R_tab);
+			if (p->tcfp_P_tab)
+				qdisc_put_rtab(p->tcfp_P_tab);
 			kfree(p);
 			return;
 		}
@@ -167,7 +129,7 @@ static int tcf_act_police_locate(struct rtattr *rta, struct rtattr *est,
 	int ret = 0, err;
 	struct rtattr *tb[TCA_POLICE_MAX];
 	struct tc_police *parm;
-	struct tcf_police *p;
+	struct tcf_police *police;
 	struct qdisc_rate_table *R_tab = NULL, *P_tab = NULL;
 
 	if (rta == NULL || rtattr_parse_nested(tb, TCA_POLICE_MAX, rta) < 0)
@@ -185,27 +147,32 @@ static int tcf_act_police_locate(struct rtattr *rta, struct rtattr *est,
 	    RTA_PAYLOAD(tb[TCA_POLICE_RESULT-1]) != sizeof(u32))
 		return -EINVAL;
 
-	if (parm->index && (p = tcf_police_lookup(parm->index)) != NULL) {
-		a->priv = p;
-		if (bind) {
-			p->bindcnt += 1;
-			p->refcnt += 1;
+	if (parm->index) {
+		struct tcf_common *pc;
+
+		pc = tcf_hash_lookup(parm->index, &police_hash_info);
+		if (pc != NULL) {
+			a->priv = pc;
+			police = to_police(pc);
+			if (bind) {
+				police->tcf_bindcnt += 1;
+				police->tcf_refcnt += 1;
+			}
+			if (ovr)
+				goto override;
+			return ret;
 		}
-		if (ovr)
-			goto override;
-		return ret;
 	}
 
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
-	if (p == NULL)
+	police = kzalloc(sizeof(*police), GFP_KERNEL);
+	if (police == NULL)
 		return -ENOMEM;
-
 	ret = ACT_P_CREATED;
-	p->refcnt = 1;
-	spin_lock_init(&p->lock);
-	p->stats_lock = &p->lock;
+	police->tcf_refcnt = 1;
+	spin_lock_init(&police->tcf_lock);
+	police->tcf_stats_lock = &police->tcf_lock;
 	if (bind)
-		p->bindcnt = 1;
+		police->tcf_bindcnt = 1;
 override:
 	if (parm->rate.rate) {
 		err = -ENOMEM;
@@ -215,67 +182,71 @@ override:
 		if (parm->peakrate.rate) {
 			P_tab = qdisc_get_rtab(&parm->peakrate,
 					       tb[TCA_POLICE_PEAKRATE-1]);
-			if (p->P_tab == NULL) {
+			if (P_tab == NULL) {
 				qdisc_put_rtab(R_tab);
 				goto failure;
 			}
 		}
 	}
 	/* No failure allowed after this point */
-	spin_lock_bh(&p->lock);
+	spin_lock_bh(&police->tcf_lock);
 	if (R_tab != NULL) {
-		qdisc_put_rtab(p->R_tab);
-		p->R_tab = R_tab;
+		qdisc_put_rtab(police->tcfp_R_tab);
+		police->tcfp_R_tab = R_tab;
 	}
 	if (P_tab != NULL) {
-		qdisc_put_rtab(p->P_tab);
-		p->P_tab = P_tab;
+		qdisc_put_rtab(police->tcfp_P_tab);
+		police->tcfp_P_tab = P_tab;
 	}
 
 	if (tb[TCA_POLICE_RESULT-1])
-		p->result = *(u32*)RTA_DATA(tb[TCA_POLICE_RESULT-1]);
-	p->toks = p->burst = parm->burst;
-	p->mtu = parm->mtu;
-	if (p->mtu == 0) {
-		p->mtu = ~0;
-		if (p->R_tab)
-			p->mtu = 255<<p->R_tab->rate.cell_log;
+		police->tcfp_result = *(u32*)RTA_DATA(tb[TCA_POLICE_RESULT-1]);
+	police->tcfp_toks = police->tcfp_burst = parm->burst;
+	police->tcfp_mtu = parm->mtu;
+	if (police->tcfp_mtu == 0) {
+		police->tcfp_mtu = ~0;
+		if (police->tcfp_R_tab)
+			police->tcfp_mtu = 255<<police->tcfp_R_tab->rate.cell_log;
 	}
-	if (p->P_tab)
-		p->ptoks = L2T_P(p, p->mtu);
-	p->action = parm->action;
+	if (police->tcfp_P_tab)
+		police->tcfp_ptoks = L2T_P(police, police->tcfp_mtu);
+	police->tcf_action = parm->action;
 
 #ifdef CONFIG_NET_ESTIMATOR
 	if (tb[TCA_POLICE_AVRATE-1])
-		p->ewma_rate = *(u32*)RTA_DATA(tb[TCA_POLICE_AVRATE-1]);
+		police->tcfp_ewma_rate =
+			*(u32*)RTA_DATA(tb[TCA_POLICE_AVRATE-1]);
 	if (est)
-		gen_replace_estimator(&p->bstats, &p->rate_est, p->stats_lock, est);
+		gen_replace_estimator(&police->tcf_bstats,
+				      &police->tcf_rate_est,
+				      police->tcf_stats_lock, est);
 #endif
 
-	spin_unlock_bh(&p->lock);
+	spin_unlock_bh(&police->tcf_lock);
 	if (ret != ACT_P_CREATED)
 		return ret;
 
-	PSCHED_GET_TIME(p->t_c);
-	p->index = parm->index ? : tcf_police_new_index();
-	h = tcf_police_hash(p->index);
+	PSCHED_GET_TIME(police->tcfp_t_c);
+	police->tcf_index = parm->index ? parm->index :
+		tcf_hash_new_index(&police_idx_gen, &police_hash_info);
+	h = tcf_hash(police->tcf_index, POL_TAB_MASK);
 	write_lock_bh(&police_lock);
-	p->next = tcf_police_ht[h];
-	tcf_police_ht[h] = p;
+	police->tcf_next = tcf_police_ht[h];
+	tcf_police_ht[h] = &police->common;
 	write_unlock_bh(&police_lock);
 
-	a->priv = p;
+	a->priv = police;
 	return ret;
 
 failure:
 	if (ret == ACT_P_CREATED)
-		kfree(p);
+		kfree(police);
 	return err;
 }
 
 static int tcf_act_police_cleanup(struct tc_action *a, int bind)
 {
-	struct tcf_police *p = PRIV(a);
+	struct tcf_police *p = a->priv;
 
 	if (p != NULL)
 		return tcf_police_release(p, bind);
@@ -285,86 +256,87 @@ static int tcf_act_police_cleanup(struct tc_action *a, int bind)
 static int tcf_act_police(struct sk_buff *skb, struct tc_action *a,
                           struct tcf_result *res)
 {
+	struct tcf_police *police = a->priv;
 	psched_time_t now;
-	struct tcf_police *p = PRIV(a);
 	long toks;
 	long ptoks = 0;
 
-	spin_lock(&p->lock);
+	spin_lock(&police->tcf_lock);
 
-	p->bstats.bytes += skb->len;
-	p->bstats.packets++;
+	police->tcf_bstats.bytes += skb->len;
+	police->tcf_bstats.packets++;
 
 #ifdef CONFIG_NET_ESTIMATOR
-	if (p->ewma_rate && p->rate_est.bps >= p->ewma_rate) {
-		p->qstats.overlimits++;
-		spin_unlock(&p->lock);
-		return p->action;
+	if (police->tcfp_ewma_rate &&
+	    police->tcf_rate_est.bps >= police->tcfp_ewma_rate) {
+		police->tcf_qstats.overlimits++;
+		spin_unlock(&police->tcf_lock);
+		return police->tcf_action;
 	}
 #endif
 
-	if (skb->len <= p->mtu) {
-		if (p->R_tab == NULL) {
-			spin_unlock(&p->lock);
-			return p->result;
+	if (skb->len <= police->tcfp_mtu) {
+		if (police->tcfp_R_tab == NULL) {
+			spin_unlock(&police->tcf_lock);
+			return police->tcfp_result;
 		}
 
 		PSCHED_GET_TIME(now);
 
-		toks = PSCHED_TDIFF_SAFE(now, p->t_c, p->burst);
-
-		if (p->P_tab) {
-			ptoks = toks + p->ptoks;
-			if (ptoks > (long)L2T_P(p, p->mtu))
-				ptoks = (long)L2T_P(p, p->mtu);
-			ptoks -= L2T_P(p, skb->len);
+		toks = PSCHED_TDIFF_SAFE(now, police->tcfp_t_c,
+					 police->tcfp_burst);
+		if (police->tcfp_P_tab) {
+			ptoks = toks + police->tcfp_ptoks;
+			if (ptoks > (long)L2T_P(police, police->tcfp_mtu))
+				ptoks = (long)L2T_P(police, police->tcfp_mtu);
+			ptoks -= L2T_P(police, skb->len);
 		}
-		toks += p->toks;
-		if (toks > (long)p->burst)
-			toks = p->burst;
-		toks -= L2T(p, skb->len);
-
+		toks += police->tcfp_toks;
+		if (toks > (long)police->tcfp_burst)
+			toks = police->tcfp_burst;
+		toks -= L2T(police, skb->len);
 		if ((toks|ptoks) >= 0) {
-			p->t_c = now;
-			p->toks = toks;
-			p->ptoks = ptoks;
-			spin_unlock(&p->lock);
-			return p->result;
+			police->tcfp_t_c = now;
+			police->tcfp_toks = toks;
+			police->tcfp_ptoks = ptoks;
+			spin_unlock(&police->tcf_lock);
+			return police->tcfp_result;
 		}
 	}
 
-	p->qstats.overlimits++;
-	spin_unlock(&p->lock);
-	return p->action;
+	police->tcf_qstats.overlimits++;
+	spin_unlock(&police->tcf_lock);
+	return police->tcf_action;
 }
 
 static int
 tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref)
 {
 	unsigned char	 *b = skb->tail;
+	struct tcf_police *police = a->priv;
 	struct tc_police opt;
-	struct tcf_police *p = PRIV(a);
-
-	opt.index = p->index;
-	opt.action = p->action;
-	opt.mtu = p->mtu;
-	opt.burst = p->burst;
-	opt.refcnt = p->refcnt - ref;
-	opt.bindcnt = p->bindcnt - bind;
-	if (p->R_tab)
-		opt.rate = p->R_tab->rate;
+
+	opt.index = police->tcf_index;
+	opt.action = police->tcf_action;
+	opt.mtu = police->tcfp_mtu;
+	opt.burst = police->tcfp_burst;
+	opt.refcnt = police->tcf_refcnt - ref;
+	opt.bindcnt = police->tcf_bindcnt - bind;
+	if (police->tcfp_R_tab)
+		opt.rate = police->tcfp_R_tab->rate;
 	else
 		memset(&opt.rate, 0, sizeof(opt.rate));
-	if (p->P_tab)
-		opt.peakrate = p->P_tab->rate;
+	if (police->tcfp_P_tab)
+		opt.peakrate = police->tcfp_P_tab->rate;
 	else
 		memset(&opt.peakrate, 0, sizeof(opt.peakrate));
 	RTA_PUT(skb, TCA_POLICE_TBF, sizeof(opt), &opt);
-	if (p->result)
-		RTA_PUT(skb, TCA_POLICE_RESULT, sizeof(int), &p->result);
+	if (police->tcfp_result)
+		RTA_PUT(skb, TCA_POLICE_RESULT, sizeof(int),
+			&police->tcfp_result);
 #ifdef CONFIG_NET_ESTIMATOR
-	if (p->ewma_rate)
-		RTA_PUT(skb, TCA_POLICE_AVRATE, 4, &p->ewma_rate);
+	if (police->tcfp_ewma_rate)
+		RTA_PUT(skb, TCA_POLICE_AVRATE, 4, &police->tcfp_ewma_rate);
 #endif
 	return skb->len;
 
@@ -379,13 +351,14 @@ MODULE_LICENSE("GPL");
 
 static struct tc_action_ops act_police_ops = {
 	.kind		=	"police",
+	.hinfo		=	&police_hash_info,
 	.type		=	TCA_ID_POLICE,
 	.capab		=	TCA_CAP_NONE,
 	.owner		=	THIS_MODULE,
 	.act		=	tcf_act_police,
 	.dump		=	tcf_act_police_dump,
 	.cleanup	=	tcf_act_police_cleanup,
-	.lookup		=	tcf_act_police_hash_search,
+	.lookup		=	tcf_hash_search,
 	.init		=	tcf_act_police_locate,
 	.walk		=	tcf_act_police_walker
 };
@@ -407,10 +380,39 @@ module_exit(police_cleanup_module);
 
 #else /* CONFIG_NET_CLS_ACT */
 
-struct tcf_police * tcf_police_locate(struct rtattr *rta, struct rtattr *est)
+static struct tcf_common *tcf_police_lookup(u32 index)
 {
-	unsigned h;
-	struct tcf_police *p;
+	struct tcf_hashinfo *hinfo = &police_hash_info;
+	struct tcf_common *p;
+
+	read_lock(hinfo->lock);
+	for (p = hinfo->htab[tcf_hash(index, hinfo->hmask)]; p;
+	     p = p->tcfc_next) {
+		if (p->tcfc_index == index)
+			break;
+	}
+	read_unlock(hinfo->lock);
+
+	return p;
+}
+
+static u32 tcf_police_new_index(void)
+{
+	u32 *idx_gen = &police_idx_gen;
+	u32 val = *idx_gen;
+
+	do {
+		if (++val == 0)
+			val = 1;
+	} while (tcf_police_lookup(val));
+
+	return (*idx_gen = val);
+}
+
+struct tcf_police *tcf_police_locate(struct rtattr *rta, struct rtattr *est)
+{
+	unsigned int h;
+	struct tcf_police *police;
 	struct rtattr *tb[TCA_POLICE_MAX];
 	struct tc_police *parm;
 
@@ -423,149 +425,158 @@ struct tcf_police * tcf_police_locate(struct rtattr *rta, struct rtattr *est)
 
 	parm = RTA_DATA(tb[TCA_POLICE_TBF-1]);
 
-	if (parm->index && (p = tcf_police_lookup(parm->index)) != NULL) {
-		p->refcnt++;
-		return p;
-	}
+	if (parm->index) {
+		struct tcf_common *pc;
 
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
-	if (p == NULL)
+		pc = tcf_police_lookup(parm->index);
+		if (pc) {
+			police = to_police(pc);
+			police->tcf_refcnt++;
+			return police;
+		}
+	}
+	police = kzalloc(sizeof(*police), GFP_KERNEL);
+	if (unlikely(!police))
 		return NULL;
 
-	p->refcnt = 1;
-	spin_lock_init(&p->lock);
-	p->stats_lock = &p->lock;
+	police->tcf_refcnt = 1;
+	spin_lock_init(&police->tcf_lock);
+	police->tcf_stats_lock = &police->tcf_lock;
 	if (parm->rate.rate) {
-		p->R_tab = qdisc_get_rtab(&parm->rate, tb[TCA_POLICE_RATE-1]);
-		if (p->R_tab == NULL)
+		police->tcfp_R_tab =
+			qdisc_get_rtab(&parm->rate, tb[TCA_POLICE_RATE-1]);
+		if (police->tcfp_R_tab == NULL)
 			goto failure;
 		if (parm->peakrate.rate) {
-			p->P_tab = qdisc_get_rtab(&parm->peakrate,
-			                          tb[TCA_POLICE_PEAKRATE-1]);
-			if (p->P_tab == NULL)
+			police->tcfp_P_tab =
+				qdisc_get_rtab(&parm->peakrate,
+					       tb[TCA_POLICE_PEAKRATE-1]);
+			if (police->tcfp_P_tab == NULL)
 				goto failure;
 		}
 	}
 	if (tb[TCA_POLICE_RESULT-1]) {
 		if (RTA_PAYLOAD(tb[TCA_POLICE_RESULT-1]) != sizeof(u32))
 			goto failure;
-		p->result = *(u32*)RTA_DATA(tb[TCA_POLICE_RESULT-1]);
+		police->tcfp_result = *(u32*)RTA_DATA(tb[TCA_POLICE_RESULT-1]);
 	}
 #ifdef CONFIG_NET_ESTIMATOR
 	if (tb[TCA_POLICE_AVRATE-1]) {
 		if (RTA_PAYLOAD(tb[TCA_POLICE_AVRATE-1]) != sizeof(u32))
 			goto failure;
-		p->ewma_rate = *(u32*)RTA_DATA(tb[TCA_POLICE_AVRATE-1]);
+		police->tcfp_ewma_rate =
+			*(u32*)RTA_DATA(tb[TCA_POLICE_AVRATE-1]);
 	}
 #endif
-	p->toks = p->burst = parm->burst;
-	p->mtu = parm->mtu;
-	if (p->mtu == 0) {
-		p->mtu = ~0;
-		if (p->R_tab)
-			p->mtu = 255<<p->R_tab->rate.cell_log;
+	police->tcfp_toks = police->tcfp_burst = parm->burst;
+	police->tcfp_mtu = parm->mtu;
+	if (police->tcfp_mtu == 0) {
+		police->tcfp_mtu = ~0;
+		if (police->tcfp_R_tab)
+			police->tcfp_mtu = 255<<police->tcfp_R_tab->rate.cell_log;
 	}
-	if (p->P_tab)
-		p->ptoks = L2T_P(p, p->mtu);
-	PSCHED_GET_TIME(p->t_c);
-	p->index = parm->index ? : tcf_police_new_index();
-	p->action = parm->action;
+	if (police->tcfp_P_tab)
+		police->tcfp_ptoks = L2T_P(police, police->tcfp_mtu);
+	PSCHED_GET_TIME(police->tcfp_t_c);
+	police->tcf_index = parm->index ? parm->index :
+		tcf_police_new_index();
+	police->tcf_action = parm->action;
 #ifdef CONFIG_NET_ESTIMATOR
 	if (est)
-		gen_new_estimator(&p->bstats, &p->rate_est, p->stats_lock, est);
+		gen_new_estimator(&police->tcf_bstats, &police->tcf_rate_est,
+				  police->tcf_stats_lock, est);
 #endif
-	h = tcf_police_hash(p->index);
+	h = tcf_hash(police->tcf_index, POL_TAB_MASK);
 	write_lock_bh(&police_lock);
-	p->next = tcf_police_ht[h];
-	tcf_police_ht[h] = p;
+	police->tcf_next = tcf_police_ht[h];
+	tcf_police_ht[h] = &police->common;
 	write_unlock_bh(&police_lock);
-	return p;
+	return police;
 
 failure:
-	if (p->R_tab)
-		qdisc_put_rtab(p->R_tab);
-	kfree(p);
+	if (police->tcfp_R_tab)
+		qdisc_put_rtab(police->tcfp_R_tab);
+	kfree(police);
 	return NULL;
 }
 
-int tcf_police(struct sk_buff *skb, struct tcf_police *p)
+int tcf_police(struct sk_buff *skb, struct tcf_police *police)
 {
 	psched_time_t now;
 	long toks;
 	long ptoks = 0;
 
-	spin_lock(&p->lock);
+	spin_lock(&police->tcf_lock);
 
-	p->bstats.bytes += skb->len;
-	p->bstats.packets++;
+	police->tcf_bstats.bytes += skb->len;
+	police->tcf_bstats.packets++;
 
 #ifdef CONFIG_NET_ESTIMATOR
-	if (p->ewma_rate && p->rate_est.bps >= p->ewma_rate) {
-		p->qstats.overlimits++;
-		spin_unlock(&p->lock);
-		return p->action;
+	if (police->tcfp_ewma_rate &&
+	    police->tcf_rate_est.bps >= police->tcfp_ewma_rate) {
+		police->tcf_qstats.overlimits++;
+		spin_unlock(&police->tcf_lock);
+		return police->tcf_action;
 	}
 #endif
-
-	if (skb->len <= p->mtu) {
-		if (p->R_tab == NULL) {
-			spin_unlock(&p->lock);
-			return p->result;
+	if (skb->len <= police->tcfp_mtu) {
+		if (police->tcfp_R_tab == NULL) {
+			spin_unlock(&police->tcf_lock);
+			return police->tcfp_result;
 		}
 
 		PSCHED_GET_TIME(now);
-
-		toks = PSCHED_TDIFF_SAFE(now, p->t_c, p->burst);
-
-		if (p->P_tab) {
-			ptoks = toks + p->ptoks;
-			if (ptoks > (long)L2T_P(p, p->mtu))
-				ptoks = (long)L2T_P(p, p->mtu);
-			ptoks -= L2T_P(p, skb->len);
+		toks = PSCHED_TDIFF_SAFE(now, police->tcfp_t_c,
+					 police->tcfp_burst);
+		if (police->tcfp_P_tab) {
+			ptoks = toks + police->tcfp_ptoks;
+			if (ptoks > (long)L2T_P(police, police->tcfp_mtu))
+				ptoks = (long)L2T_P(police, police->tcfp_mtu);
+			ptoks -= L2T_P(police, skb->len);
 		}
-		toks += p->toks;
-		if (toks > (long)p->burst)
-			toks = p->burst;
-		toks -= L2T(p, skb->len);
-
+		toks += police->tcfp_toks;
+		if (toks > (long)police->tcfp_burst)
+			toks = police->tcfp_burst;
+		toks -= L2T(police, skb->len);
 		if ((toks|ptoks) >= 0) {
-			p->t_c = now;
-			p->toks = toks;
-			p->ptoks = ptoks;
-			spin_unlock(&p->lock);
-			return p->result;
+			police->tcfp_t_c = now;
+			police->tcfp_toks = toks;
+			police->tcfp_ptoks = ptoks;
+			spin_unlock(&police->tcf_lock);
+			return police->tcfp_result;
 		}
 	}
 
-	p->qstats.overlimits++;
-	spin_unlock(&p->lock);
-	return p->action;
+	police->tcf_qstats.overlimits++;
+	spin_unlock(&police->tcf_lock);
+	return police->tcf_action;
 }
 EXPORT_SYMBOL(tcf_police);
 
-int tcf_police_dump(struct sk_buff *skb, struct tcf_police *p)
+int tcf_police_dump(struct sk_buff *skb, struct tcf_police *police)
 {
-	unsigned char	 *b = skb->tail;
+	unsigned char *b = skb->tail;
 	struct tc_police opt;
 
-	opt.index = p->index;
-	opt.action = p->action;
-	opt.mtu = p->mtu;
-	opt.burst = p->burst;
-	if (p->R_tab)
-		opt.rate = p->R_tab->rate;
+	opt.index = police->tcf_index;
+	opt.action = police->tcf_action;
+	opt.mtu = police->tcfp_mtu;
+	opt.burst = police->tcfp_burst;
+	if (police->tcfp_R_tab)
+		opt.rate = police->tcfp_R_tab->rate;
 	else
 		memset(&opt.rate, 0, sizeof(opt.rate));
-	if (p->P_tab)
-		opt.peakrate = p->P_tab->rate;
+	if (police->tcfp_P_tab)
+		opt.peakrate = police->tcfp_P_tab->rate;
 	else
 		memset(&opt.peakrate, 0, sizeof(opt.peakrate));
 	RTA_PUT(skb, TCA_POLICE_TBF, sizeof(opt), &opt);
-	if (p->result)
-		RTA_PUT(skb, TCA_POLICE_RESULT, sizeof(int), &p->result);
+	if (police->tcfp_result)
+		RTA_PUT(skb, TCA_POLICE_RESULT, sizeof(int),
+			&police->tcfp_result);
 #ifdef CONFIG_NET_ESTIMATOR
-	if (p->ewma_rate)
-		RTA_PUT(skb, TCA_POLICE_AVRATE, 4, &p->ewma_rate);
+	if (police->tcfp_ewma_rate)
+		RTA_PUT(skb, TCA_POLICE_AVRATE, 4, &police->tcfp_ewma_rate);
 #endif
 	return skb->len;
 
@@ -574,19 +585,20 @@ rtattr_failure:
 	return -1;
 }
 
-int tcf_police_dump_stats(struct sk_buff *skb, struct tcf_police *p)
+int tcf_police_dump_stats(struct sk_buff *skb, struct tcf_police *police)
 {
 	struct gnet_dump d;
 	
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
-			TCA_XSTATS, p->stats_lock, &d) < 0)
+					 TCA_XSTATS, police->tcf_stats_lock,
+					 &d) < 0)
 		goto errout;
 	
-	if (gnet_stats_copy_basic(&d, &p->bstats) < 0 ||
+	if (gnet_stats_copy_basic(&d, &police->tcf_bstats) < 0 ||
 #ifdef CONFIG_NET_ESTIMATOR
-	    gnet_stats_copy_rate_est(&d, &p->rate_est) < 0 ||
+	    gnet_stats_copy_rate_est(&d, &police->tcf_rate_est) < 0 ||
 #endif
-	    gnet_stats_copy_queue(&d, &p->qstats) < 0)
+	    gnet_stats_copy_queue(&d, &police->tcf_qstats) < 0)
 		goto errout;
 
 	if (gnet_stats_finish_copy(&d) < 0)
diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c
index 17105c82537f..8c1ab8ad8fa6 100644
--- a/net/sched/act_simple.c
+++ b/net/sched/act_simple.c
@@ -20,54 +20,175 @@
 
 #define TCA_ACT_SIMP 22
 
-/* XXX: Hide all these common elements under some macro 
- * probably
-*/
 #include <linux/tc_act/tc_defact.h>
 #include <net/tc_act/tc_defact.h>
 
-/* use generic hash table with 8 buckets */
-#define MY_TAB_SIZE     8
-#define MY_TAB_MASK     (MY_TAB_SIZE - 1)
-static u32 idx_gen;
-static struct tcf_defact *tcf_simp_ht[MY_TAB_SIZE];
+#define SIMP_TAB_MASK     7
+static struct tcf_common *tcf_simp_ht[SIMP_TAB_MASK + 1];
+static u32 simp_idx_gen;
 static DEFINE_RWLOCK(simp_lock);
 
-/* override the defaults */
-#define tcf_st		tcf_defact
-#define tc_st		tc_defact
-#define tcf_t_lock	simp_lock
-#define tcf_ht		tcf_simp_ht
-
-#define CONFIG_NET_ACT_INIT 1
-#include <net/pkt_act.h>
-#include <net/act_generic.h>
+struct tcf_hashinfo simp_hash_info = {
+	.htab	=	tcf_simp_ht,
+	.hmask	=	SIMP_TAB_MASK,
+	.lock	=	&simp_lock,
+};
 
 static int tcf_simp(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res)
 {
-	struct tcf_defact *p = PRIV(a, defact);
+	struct tcf_defact *d = a->priv;
 
-	spin_lock(&p->lock);
-	p->tm.lastuse = jiffies;
-	p->bstats.bytes += skb->len;
-	p->bstats.packets++;
+	spin_lock(&d->tcf_lock);
+	d->tcf_tm.lastuse = jiffies;
+	d->tcf_bstats.bytes += skb->len;
+	d->tcf_bstats.packets++;
 
 	/* print policy string followed by _ then packet count 
 	 * Example if this was the 3rd packet and the string was "hello" 
 	 * then it would look like "hello_3" (without quotes) 
 	 **/
-	printk("simple: %s_%d\n", (char *)p->defdata, p->bstats.packets);
-	spin_unlock(&p->lock);
-	return p->action;
+	printk("simple: %s_%d\n",
+	       (char *)d->tcfd_defdata, d->tcf_bstats.packets);
+	spin_unlock(&d->tcf_lock);
+	return d->tcf_action;
+}
+
+static int tcf_simp_release(struct tcf_defact *d, int bind)
+{
+	int ret = 0;
+	if (d) {
+		if (bind)
+			d->tcf_bindcnt--;
+		d->tcf_refcnt--;
+		if (d->tcf_bindcnt <= 0 && d->tcf_refcnt <= 0) {
+			kfree(d->tcfd_defdata);
+			tcf_hash_destroy(&d->common, &simp_hash_info);
+			ret = 1;
+		}
+	}
+	return ret;
+}
+
+static int alloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata)
+{
+	d->tcfd_defdata = kmalloc(datalen, GFP_KERNEL);
+	if (unlikely(!d->tcfd_defdata))
+		return -ENOMEM;
+	d->tcfd_datalen = datalen;
+	memcpy(d->tcfd_defdata, defdata, datalen);
+	return 0;
+}
+
+static int realloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata)
+{
+	kfree(d->tcfd_defdata);
+	return alloc_defdata(d, datalen, defdata);
+}
+
+static int tcf_simp_init(struct rtattr *rta, struct rtattr *est,
+			 struct tc_action *a, int ovr, int bind)
+{
+	struct rtattr *tb[TCA_DEF_MAX];
+	struct tc_defact *parm;
+	struct tcf_defact *d;
+	struct tcf_common *pc;
+	void *defdata;
+	u32 datalen = 0;
+	int ret = 0;
+
+	if (rta == NULL || rtattr_parse_nested(tb, TCA_DEF_MAX, rta) < 0)
+		return -EINVAL;
+
+	if (tb[TCA_DEF_PARMS - 1] == NULL ||
+	    RTA_PAYLOAD(tb[TCA_DEF_PARMS - 1]) < sizeof(*parm))
+		return -EINVAL;
+
+	parm = RTA_DATA(tb[TCA_DEF_PARMS - 1]);
+	defdata = RTA_DATA(tb[TCA_DEF_DATA - 1]);
+	if (defdata == NULL)
+		return -EINVAL;
+
+	datalen = RTA_PAYLOAD(tb[TCA_DEF_DATA - 1]);
+	if (datalen <= 0)
+		return -EINVAL;
+
+	pc = tcf_hash_check(parm->index, a, bind, &simp_hash_info);
+	if (!pc) {
+		pc = tcf_hash_create(parm->index, est, a, sizeof(*d), bind,
+				     &simp_idx_gen, &simp_hash_info);
+		if (unlikely(!pc))
+			return -ENOMEM;
+
+		d = to_defact(pc);
+		ret = alloc_defdata(d, datalen, defdata);
+		if (ret < 0) {
+			kfree(pc);
+			return ret;
+		}
+		ret = ACT_P_CREATED;
+	} else {
+		d = to_defact(pc);
+		if (!ovr) {
+			tcf_simp_release(d, bind);
+			return -EEXIST;
+		}
+		realloc_defdata(d, datalen, defdata);
+	}
+
+	spin_lock_bh(&d->tcf_lock);
+	d->tcf_action = parm->action;
+	spin_unlock_bh(&d->tcf_lock);
+
+	if (ret == ACT_P_CREATED)
+		tcf_hash_insert(pc, &simp_hash_info);
+	return ret;
+}
+
+static inline int tcf_simp_cleanup(struct tc_action *a, int bind)
+{
+	struct tcf_defact *d = a->priv;
+
+	if (d)
+		return tcf_simp_release(d, bind);
+	return 0;
+}
+
+static inline int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a,
+				int bind, int ref)
+{
+	unsigned char *b = skb->tail;
+	struct tcf_defact *d = a->priv;
+	struct tc_defact opt;
+	struct tcf_t t;
+
+	opt.index = d->tcf_index;
+	opt.refcnt = d->tcf_refcnt - ref;
+	opt.bindcnt = d->tcf_bindcnt - bind;
+	opt.action = d->tcf_action;
+	RTA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt);
+	RTA_PUT(skb, TCA_DEF_DATA, d->tcfd_datalen, d->tcfd_defdata);
+	t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install);
+	t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse);
+	t.expires = jiffies_to_clock_t(d->tcf_tm.expires);
+	RTA_PUT(skb, TCA_DEF_TM, sizeof(t), &t);
+	return skb->len;
+
+rtattr_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
 }
 
 static struct tc_action_ops act_simp_ops = {
-	.kind = "simple",
-	.type = TCA_ACT_SIMP,
-	.capab = TCA_CAP_NONE,
-	.owner = THIS_MODULE,
-	.act = tcf_simp,
-	tca_use_default_ops
+	.kind		=	"simple",
+	.hinfo		=	&simp_hash_info,
+	.type		=	TCA_ACT_SIMP,
+	.capab		=	TCA_CAP_NONE,
+	.owner		=	THIS_MODULE,
+	.act		=	tcf_simp,
+	.dump		=	tcf_simp_dump,
+	.cleanup	=	tcf_simp_cleanup,
+	.init		=	tcf_simp_init,
+	.walk		=	tcf_generic_walker,
 };
 
 MODULE_AUTHOR("Jamal Hadi Salim(2005)");
-- 
cgit v1.2.3


From e0a1ad73d34fd6dfdb630479400511e9879069c0 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 22 Aug 2006 00:00:21 -0700
Subject: [IPv6] route: Simplify ip6_del_rt()

Provide a simple ip6_del_rt() for the majority of users and
an alternative for the exception via netlink. Avoids code
obfuscation.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |  5 +----
 net/ipv6/addrconf.c     |  6 +++---
 net/ipv6/ndisc.c        |  4 ++--
 net/ipv6/route.c        | 18 ++++++++++++------
 4 files changed, 18 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 01bfe404784f..a7e6086a2bd4 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -69,10 +69,7 @@ extern int			ip6_ins_rt(struct rt6_info *,
 					   struct nlmsghdr *,
 					   void *rtattr,
 					   struct netlink_skb_parms *req);
-extern int			ip6_del_rt(struct rt6_info *,
-					   struct nlmsghdr *,
-					   void *rtattr,
-					   struct netlink_skb_parms *req);
+extern int			ip6_del_rt(struct rt6_info *);
 
 extern int			ip6_rt_addr_add(struct in6_addr *addr,
 						struct net_device *dev,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index f1ede9004887..27f2e3309598 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -736,7 +736,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
 
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
 			if (onlink == 0) {
-				ip6_del_rt(rt, NULL, NULL, NULL);
+				ip6_del_rt(rt);
 				rt = NULL;
 			} else if (!(rt->rt6i_flags & RTF_EXPIRES)) {
 				rt->rt6i_expires = expires;
@@ -1662,7 +1662,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 		if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) {
 			if (rt->rt6i_flags&RTF_EXPIRES) {
 				if (valid_lft == 0) {
-					ip6_del_rt(rt, NULL, NULL, NULL);
+					ip6_del_rt(rt);
 					rt = NULL;
 				} else {
 					rt->rt6i_expires = jiffies + rt_expires;
@@ -3557,7 +3557,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 			addrconf_leave_anycast(ifp);
 		addrconf_leave_solict(ifp->idev, &ifp->addr);
 		dst_hold(&ifp->rt->u.dst);
-		if (ip6_del_rt(ifp->rt, NULL, NULL, NULL))
+		if (ip6_del_rt(ifp->rt))
 			dst_free(&ifp->rt->u.dst);
 		break;
 	}
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 5743e8bffefd..419d65163819 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -961,7 +961,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 			struct rt6_info *rt;
 			rt = rt6_get_dflt_router(saddr, dev);
 			if (rt)
-				ip6_del_rt(rt, NULL, NULL, NULL);
+				ip6_del_rt(rt);
 		}
 
 out:
@@ -1114,7 +1114,7 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 
 	if (rt && lifetime == 0) {
 		neigh_clone(neigh);
-		ip6_del_rt(rt, NULL, NULL, NULL);
+		ip6_del_rt(rt);
 		rt = NULL;
 	}
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 1aca787ead85..8d511de0db1b 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -457,7 +457,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 	rt = rt6_get_route_info(prefix, rinfo->prefix_len, gwaddr, dev->ifindex);
 
 	if (rt && !lifetime) {
-		ip6_del_rt(rt, NULL, NULL, NULL);
+		ip6_del_rt(rt);
 		rt = NULL;
 	}
 
@@ -813,7 +813,7 @@ static struct dst_entry *ip6_negative_advice(struct dst_entry *dst)
 
 	if (rt) {
 		if (rt->rt6i_flags & RTF_CACHE)
-			ip6_del_rt(rt, NULL, NULL, NULL);
+			ip6_del_rt(rt);
 		else
 			dst_release(dst);
 	}
@@ -1218,7 +1218,8 @@ out:
 	return err;
 }
 
-int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
+static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
+			void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
 	struct fib6_table *table;
@@ -1237,6 +1238,11 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct
 	return err;
 }
 
+int ip6_del_rt(struct rt6_info *rt)
+{
+	return __ip6_del_rt(rt, NULL, NULL, NULL);
+}
+
 static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 			 void *_rtattr, struct netlink_skb_parms *req,
 			 u32 table_id)
@@ -1271,7 +1277,7 @@ static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 			dst_hold(&rt->u.dst);
 			read_unlock_bh(&table->tb6_lock);
 
-			return ip6_del_rt(rt, nlh, _rtattr, req);
+			return __ip6_del_rt(rt, nlh, _rtattr, req);
 		}
 	}
 	read_unlock_bh(&table->tb6_lock);
@@ -1395,7 +1401,7 @@ restart:
 	call_netevent_notifiers(NETEVENT_REDIRECT, &netevent);
 
 	if (rt->rt6i_flags&RTF_CACHE) {
-		ip6_del_rt(rt, NULL, NULL, NULL);
+		ip6_del_rt(rt);
 		return;
 	}
 
@@ -1631,7 +1637,7 @@ restart:
 		if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) {
 			dst_hold(&rt->u.dst);
 			read_unlock_bh(&table->tb6_lock);
-			ip6_del_rt(rt, NULL, NULL, NULL);
+			ip6_del_rt(rt);
 			goto restart;
 		}
 	}
-- 
cgit v1.2.3


From 40e22e8f3d4d4f1ff68fb03683f007c53ee8b348 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 22 Aug 2006 00:00:45 -0700
Subject: [IPv6] route: Simplify ip6_ins_rt()

Provide a simple ip6_ins_rt() for the majority of users and
an alternative for the exception via netlink. Avoids code
obfuscation.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h |  5 +----
 net/ipv6/addrconf.c     |  2 +-
 net/ipv6/anycast.c      |  2 +-
 net/ipv6/route.c        | 19 ++++++++++++-------
 4 files changed, 15 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index a7e6086a2bd4..172c4761e2bf 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -65,10 +65,7 @@ extern int			ip6_route_add(struct in6_rtmsg *rtmsg,
 					      void *rtattr,
 					      struct netlink_skb_parms *req,
 					      u32 table_id);
-extern int			ip6_ins_rt(struct rt6_info *,
-					   struct nlmsghdr *,
-					   void *rtattr,
-					   struct netlink_skb_parms *req);
+extern int			ip6_ins_rt(struct rt6_info *);
 extern int			ip6_del_rt(struct rt6_info *);
 
 extern int			ip6_rt_addr_add(struct in6_addr *addr,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 27f2e3309598..aafba9ea9cb6 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3548,7 +3548,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 
 	switch (event) {
 	case RTM_NEWADDR:
-		ip6_ins_rt(ifp->rt, NULL, NULL, NULL);
+		ip6_ins_rt(ifp->rt);
 		if (ifp->idev->cnf.forwarding)
 			addrconf_join_anycast(ifp);
 		break;
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index f6881d7a0385..abbc35a13e08 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -335,7 +335,7 @@ int ipv6_dev_ac_inc(struct net_device *dev, struct in6_addr *addr)
 	write_unlock_bh(&idev->lock);
 
 	dst_hold(&rt->u.dst);
-	if (ip6_ins_rt(rt, NULL, NULL, NULL))
+	if (ip6_ins_rt(rt))
 		dst_release(&rt->u.dst);
 
 	addrconf_join_solict(dev, &aca->aca_addr);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8d511de0db1b..9ec348a72a95 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -546,8 +546,8 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
    be destroyed.
  */
 
-int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
-		void *_rtattr, struct netlink_skb_parms *req)
+static int __ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
+			void *_rtattr, struct netlink_skb_parms *req)
 {
 	int err;
 	struct fib6_table *table;
@@ -560,6 +560,11 @@ int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
 	return err;
 }
 
+int ip6_ins_rt(struct rt6_info *rt)
+{
+	return __ip6_ins_rt(rt, NULL, NULL, NULL);
+}
+
 static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *daddr,
 				      struct in6_addr *saddr)
 {
@@ -657,7 +662,7 @@ restart:
 
 	dst_hold(&rt->u.dst);
 	if (nrt) {
-		err = ip6_ins_rt(nrt, NULL, NULL, NULL);
+		err = ip6_ins_rt(nrt);
 		if (!err)
 			goto out2;
 	}
@@ -752,7 +757,7 @@ restart:
 
 	dst_hold(&rt->u.dst);
 	if (nrt) {
-		err = ip6_ins_rt(nrt, NULL, NULL, NULL);
+		err = ip6_ins_rt(nrt);
 		if (!err)
 			goto out2;
 	}
@@ -1206,7 +1211,7 @@ install_route:
 	rt->u.dst.dev = dev;
 	rt->rt6i_idev = idev;
 	rt->rt6i_table = table;
-	return ip6_ins_rt(rt, nlh, _rtattr, req);
+	return __ip6_ins_rt(rt, nlh, _rtattr, req);
 
 out:
 	if (dev)
@@ -1393,7 +1398,7 @@ restart:
 	nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
 	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&nrt->u.dst));
 
-	if (ip6_ins_rt(nrt, NULL, NULL, NULL))
+	if (ip6_ins_rt(nrt))
 		goto out;
 
 	netevent.old = &rt->u.dst;
@@ -1483,7 +1488,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		dst_set_expires(&nrt->u.dst, ip6_rt_mtu_expires);
 		nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES;
 
-		ip6_ins_rt(nrt, NULL, NULL, NULL);
+		ip6_ins_rt(nrt);
 	}
 out:
 	dst_release(&rt->u.dst);
-- 
cgit v1.2.3


From 86872cb57925c46a6499887d77afb880a892c0ec Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 22 Aug 2006 00:01:08 -0700
Subject: [IPv6] route: FIB6 configuration using struct fib6_config

Replaces the struct in6_rtmsg based interface orignating from
the ioctl interface with a struct fib6_config based on. Allows
changing the interface without breaking the ioctl interface
and avoids passing on tons of parameters.

The recently introduced struct nl_info is used to pass on
netlink authorship information for notifications.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h   |  38 ++++--
 include/net/ip6_route.h |   6 +-
 net/ipv6/addrconf.c     |  65 +++++-----
 net/ipv6/ip6_fib.c      |  19 ++-
 net/ipv6/route.c        | 331 ++++++++++++++++++++++++++++--------------------
 5 files changed, 259 insertions(+), 200 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 69c444209781..9610b887ffb5 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -16,14 +16,35 @@
 #ifdef __KERNEL__
 
 #include <linux/ipv6_route.h>
-
-#include <net/dst.h>
-#include <net/flow.h>
 #include <linux/rtnetlink.h>
 #include <linux/spinlock.h>
+#include <net/dst.h>
+#include <net/flow.h>
+#include <net/netlink.h>
 
 struct rt6_info;
 
+struct fib6_config
+{
+	u32		fc_table;
+	u32		fc_metric;
+	int		fc_dst_len;
+	int		fc_src_len;
+	int		fc_ifindex;
+	u32		fc_flags;
+	u32		fc_protocol;
+
+	struct in6_addr	fc_dst;
+	struct in6_addr	fc_src;
+	struct in6_addr	fc_gateway;
+
+	unsigned long	fc_expires;
+	struct nlattr	*fc_mx;
+	int		fc_mx_len;
+
+	struct nl_info	fc_nlinfo;
+};
+
 struct fib6_node
 {
 	struct fib6_node	*parent;
@@ -175,18 +196,13 @@ extern void			fib6_clean_all(int (*func)(struct rt6_info *, void *arg),
 
 extern int			fib6_add(struct fib6_node *root,
 					 struct rt6_info *rt,
-					 struct nlmsghdr *nlh,
-					 void *rtattr,
-					 struct netlink_skb_parms *req);
+					 struct nl_info *info);
 
 extern int			fib6_del(struct rt6_info *rt,
-					 struct nlmsghdr *nlh,
-					 void *rtattr,
-					 struct netlink_skb_parms *req);
+					 struct nl_info *info);
 
 extern void			inet6_rt_notify(int event, struct rt6_info *rt,
-						struct nlmsghdr *nlh,
-						struct netlink_skb_parms *req);
+						struct nl_info *info);
 
 extern void			fib6_run_gc(unsigned long dummy);
 
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 172c4761e2bf..3f170f667c7b 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -60,11 +60,7 @@ extern void			ip6_route_cleanup(void);
 
 extern int			ipv6_route_ioctl(unsigned int cmd, void __user *arg);
 
-extern int			ip6_route_add(struct in6_rtmsg *rtmsg,
-					      struct nlmsghdr *,
-					      void *rtattr,
-					      struct netlink_skb_parms *req,
-					      u32 table_id);
+extern int			ip6_route_add(struct fib6_config *cfg);
 extern int			ip6_ins_rt(struct rt6_info *);
 extern int			ip6_del_rt(struct rt6_info *);
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index aafba9ea9cb6..fc9cff3426c4 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1509,59 +1509,56 @@ static void
 addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
 		      unsigned long expires, u32 flags)
 {
-	struct in6_rtmsg rtmsg;
+	struct fib6_config cfg = {
+		.fc_table = RT6_TABLE_PREFIX,
+		.fc_metric = IP6_RT_PRIO_ADDRCONF,
+		.fc_ifindex = dev->ifindex,
+		.fc_expires = expires,
+		.fc_dst_len = plen,
+		.fc_flags = RTF_UP | flags,
+	};
 
-	memset(&rtmsg, 0, sizeof(rtmsg));
-	ipv6_addr_copy(&rtmsg.rtmsg_dst, pfx);
-	rtmsg.rtmsg_dst_len = plen;
-	rtmsg.rtmsg_metric = IP6_RT_PRIO_ADDRCONF;
-	rtmsg.rtmsg_ifindex = dev->ifindex;
-	rtmsg.rtmsg_info = expires;
-	rtmsg.rtmsg_flags = RTF_UP|flags;
-	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
+	ipv6_addr_copy(&cfg.fc_dst, pfx);
 
 	/* Prevent useless cloning on PtP SIT.
 	   This thing is done here expecting that the whole
 	   class of non-broadcast devices need not cloning.
 	 */
-	if (dev->type == ARPHRD_SIT && (dev->flags&IFF_POINTOPOINT))
-		rtmsg.rtmsg_flags |= RTF_NONEXTHOP;
+	if (dev->type == ARPHRD_SIT && (dev->flags & IFF_POINTOPOINT))
+		cfg.fc_flags |= RTF_NONEXTHOP;
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_PREFIX);
+	ip6_route_add(&cfg);
 }
 
 /* Create "default" multicast route to the interface */
 
 static void addrconf_add_mroute(struct net_device *dev)
 {
-	struct in6_rtmsg rtmsg;
-
-	memset(&rtmsg, 0, sizeof(rtmsg));
-	ipv6_addr_set(&rtmsg.rtmsg_dst,
-		      htonl(0xFF000000), 0, 0, 0);
-	rtmsg.rtmsg_dst_len = 8;
-	rtmsg.rtmsg_metric = IP6_RT_PRIO_ADDRCONF;
-	rtmsg.rtmsg_ifindex = dev->ifindex;
-	rtmsg.rtmsg_flags = RTF_UP;
-	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
-	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_LOCAL);
+	struct fib6_config cfg = {
+		.fc_table = RT6_TABLE_LOCAL,
+		.fc_metric = IP6_RT_PRIO_ADDRCONF,
+		.fc_ifindex = dev->ifindex,
+		.fc_dst_len = 8,
+		.fc_flags = RTF_UP,
+	};
+
+	ipv6_addr_set(&cfg.fc_dst, htonl(0xFF000000), 0, 0, 0);
+
+	ip6_route_add(&cfg);
 }
 
 static void sit_route_add(struct net_device *dev)
 {
-	struct in6_rtmsg rtmsg;
-
-	memset(&rtmsg, 0, sizeof(rtmsg));
-
-	rtmsg.rtmsg_type	= RTMSG_NEWROUTE;
-	rtmsg.rtmsg_metric	= IP6_RT_PRIO_ADDRCONF;
+	struct fib6_config cfg = {
+		.fc_table = RT6_TABLE_MAIN,
+		.fc_metric = IP6_RT_PRIO_ADDRCONF,
+		.fc_ifindex = dev->ifindex,
+		.fc_dst_len = 96,
+		.fc_flags = RTF_UP | RTF_NONEXTHOP,
+	};
 
 	/* prefix length - 96 bits "::d.d.d.d" */
-	rtmsg.rtmsg_dst_len	= 96;
-	rtmsg.rtmsg_flags	= RTF_UP|RTF_NONEXTHOP;
-	rtmsg.rtmsg_ifindex	= dev->ifindex;
-
-	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_MAIN);
+	ip6_route_add(&cfg);
 }
 
 static void addrconf_add_lroute(struct net_device *dev)
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index be36f4acda94..667b1b1ea25d 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -610,7 +610,7 @@ insert_above:
  */
 
 static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
-		struct nlmsghdr *nlh,  struct netlink_skb_parms *req)
+			    struct nl_info *info)
 {
 	struct rt6_info *iter = NULL;
 	struct rt6_info **ins;
@@ -665,7 +665,7 @@ out:
 	*ins = rt;
 	rt->rt6i_node = fn;
 	atomic_inc(&rt->rt6i_ref);
-	inet6_rt_notify(RTM_NEWROUTE, rt, nlh, req);
+	inet6_rt_notify(RTM_NEWROUTE, rt, info);
 	rt6_stats.fib_rt_entries++;
 
 	if ((fn->fn_flags & RTN_RTINFO) == 0) {
@@ -695,8 +695,7 @@ void fib6_force_start_gc(void)
  *	with source addr info in sub-trees
  */
 
-int fib6_add(struct fib6_node *root, struct rt6_info *rt, 
-		struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
+int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info)
 {
 	struct fib6_node *fn;
 	int err = -ENOMEM;
@@ -769,7 +768,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt,
 	}
 #endif
 
-	err = fib6_add_rt2node(fn, rt, nlh, req);
+	err = fib6_add_rt2node(fn, rt, info);
 
 	if (err == 0) {
 		fib6_start_gc(rt);
@@ -1076,7 +1075,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 }
 
 static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
-    struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
+			   struct nl_info *info)
 {
 	struct fib6_walker_t *w;
 	struct rt6_info *rt = *rtp;
@@ -1132,11 +1131,11 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 		if (atomic_read(&rt->rt6i_ref) != 1) BUG();
 	}
 
-	inet6_rt_notify(RTM_DELROUTE, rt, nlh, req);
+	inet6_rt_notify(RTM_DELROUTE, rt, info);
 	rt6_release(rt);
 }
 
-int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req)
+int fib6_del(struct rt6_info *rt, struct nl_info *info)
 {
 	struct fib6_node *fn = rt->rt6i_node;
 	struct rt6_info **rtp;
@@ -1161,7 +1160,7 @@ int fib6_del(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct ne
 
 	for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->u.next) {
 		if (*rtp == rt) {
-			fib6_del_route(fn, rtp, nlh, _rtattr, req);
+			fib6_del_route(fn, rtp, info);
 			return 0;
 		}
 	}
@@ -1290,7 +1289,7 @@ static int fib6_clean_node(struct fib6_walker_t *w)
 		res = c->func(rt, c->arg);
 		if (res < 0) {
 			w->leaf = rt;
-			res = fib6_del(rt, NULL, NULL, NULL);
+			res = fib6_del(rt, NULL);
 			if (res) {
 #if RT6_DEBUG >= 2
 				printk(KERN_DEBUG "fib6_clean_node: del failed: rt=%p@%p err=%d\n", rt, rt->rt6i_node, res);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 9ec348a72a95..7bcffa6ddba3 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -546,15 +546,14 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
    be destroyed.
  */
 
-static int __ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
-			void *_rtattr, struct netlink_skb_parms *req)
+static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info)
 {
 	int err;
 	struct fib6_table *table;
 
 	table = rt->rt6i_table;
 	write_lock_bh(&table->tb6_lock);
-	err = fib6_add(&table->tb6_root, rt, nlh, _rtattr, req);
+	err = fib6_add(&table->tb6_root, rt, info);
 	write_unlock_bh(&table->tb6_lock);
 
 	return err;
@@ -562,7 +561,7 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
 
 int ip6_ins_rt(struct rt6_info *rt)
 {
-	return __ip6_ins_rt(rt, NULL, NULL, NULL);
+	return __ip6_ins_rt(rt, NULL);
 }
 
 static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *daddr,
@@ -1014,30 +1013,24 @@ int ipv6_get_hoplimit(struct net_device *dev)
  *
  */
 
-int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, 
-		  void *_rtattr, struct netlink_skb_parms *req,
-		  u32 table_id)
+int ip6_route_add(struct fib6_config *cfg)
 {
 	int err;
-	struct rtmsg *r;
-	struct rtattr **rta;
 	struct rt6_info *rt = NULL;
 	struct net_device *dev = NULL;
 	struct inet6_dev *idev = NULL;
 	struct fib6_table *table;
 	int addr_type;
 
-	rta = (struct rtattr **) _rtattr;
-
-	if (rtmsg->rtmsg_dst_len > 128 || rtmsg->rtmsg_src_len > 128)
+	if (cfg->fc_dst_len > 128 || cfg->fc_src_len > 128)
 		return -EINVAL;
 #ifndef CONFIG_IPV6_SUBTREES
-	if (rtmsg->rtmsg_src_len)
+	if (cfg->fc_src_len)
 		return -EINVAL;
 #endif
-	if (rtmsg->rtmsg_ifindex) {
+	if (cfg->fc_ifindex) {
 		err = -ENODEV;
-		dev = dev_get_by_index(rtmsg->rtmsg_ifindex);
+		dev = dev_get_by_index(cfg->fc_ifindex);
 		if (!dev)
 			goto out;
 		idev = in6_dev_get(dev);
@@ -1045,10 +1038,10 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 			goto out;
 	}
 
-	if (rtmsg->rtmsg_metric == 0)
-		rtmsg->rtmsg_metric = IP6_RT_PRIO_USER;
+	if (cfg->fc_metric == 0)
+		cfg->fc_metric = IP6_RT_PRIO_USER;
 
-	table = fib6_new_table(table_id);
+	table = fib6_new_table(cfg->fc_table);
 	if (table == NULL) {
 		err = -ENOBUFS;
 		goto out;
@@ -1062,14 +1055,13 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 	}
 
 	rt->u.dst.obsolete = -1;
-	rt->rt6i_expires = jiffies + clock_t_to_jiffies(rtmsg->rtmsg_info);
-	if (nlh && (r = NLMSG_DATA(nlh))) {
-		rt->rt6i_protocol = r->rtm_protocol;
-	} else {
-		rt->rt6i_protocol = RTPROT_BOOT;
-	}
+	rt->rt6i_expires = jiffies + clock_t_to_jiffies(cfg->fc_expires);
+
+	if (cfg->fc_protocol == RTPROT_UNSPEC)
+		cfg->fc_protocol = RTPROT_BOOT;
+	rt->rt6i_protocol = cfg->fc_protocol;
 
-	addr_type = ipv6_addr_type(&rtmsg->rtmsg_dst);
+	addr_type = ipv6_addr_type(&cfg->fc_dst);
 
 	if (addr_type & IPV6_ADDR_MULTICAST)
 		rt->u.dst.input = ip6_mc_input;
@@ -1078,24 +1070,22 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 
 	rt->u.dst.output = ip6_output;
 
-	ipv6_addr_prefix(&rt->rt6i_dst.addr, 
-			 &rtmsg->rtmsg_dst, rtmsg->rtmsg_dst_len);
-	rt->rt6i_dst.plen = rtmsg->rtmsg_dst_len;
+	ipv6_addr_prefix(&rt->rt6i_dst.addr, &cfg->fc_dst, cfg->fc_dst_len);
+	rt->rt6i_dst.plen = cfg->fc_dst_len;
 	if (rt->rt6i_dst.plen == 128)
 	       rt->u.dst.flags = DST_HOST;
 
 #ifdef CONFIG_IPV6_SUBTREES
-	ipv6_addr_prefix(&rt->rt6i_src.addr, 
-			 &rtmsg->rtmsg_src, rtmsg->rtmsg_src_len);
-	rt->rt6i_src.plen = rtmsg->rtmsg_src_len;
+	ipv6_addr_prefix(&rt->rt6i_src.addr, &cfg->fc_src, cfg->fc_src_len);
+	rt->rt6i_src.plen = cfg->fc_src_len;
 #endif
 
-	rt->rt6i_metric = rtmsg->rtmsg_metric;
+	rt->rt6i_metric = cfg->fc_metric;
 
 	/* We cannot add true routes via loopback here,
 	   they would result in kernel looping; promote them to reject routes
 	 */
-	if ((rtmsg->rtmsg_flags&RTF_REJECT) ||
+	if ((cfg->fc_flags & RTF_REJECT) ||
 	    (dev && (dev->flags&IFF_LOOPBACK) && !(addr_type&IPV6_ADDR_LOOPBACK))) {
 		/* hold loopback dev/idev if we haven't done so. */
 		if (dev != &loopback_dev) {
@@ -1118,12 +1108,12 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 		goto install_route;
 	}
 
-	if (rtmsg->rtmsg_flags & RTF_GATEWAY) {
+	if (cfg->fc_flags & RTF_GATEWAY) {
 		struct in6_addr *gw_addr;
 		int gwa_type;
 
-		gw_addr = &rtmsg->rtmsg_gateway;
-		ipv6_addr_copy(&rt->rt6i_gateway, &rtmsg->rtmsg_gateway);
+		gw_addr = &cfg->fc_gateway;
+		ipv6_addr_copy(&rt->rt6i_gateway, gw_addr);
 		gwa_type = ipv6_addr_type(gw_addr);
 
 		if (gwa_type != (IPV6_ADDR_LINKLOCAL|IPV6_ADDR_UNICAST)) {
@@ -1140,7 +1130,7 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 			if (!(gwa_type&IPV6_ADDR_UNICAST))
 				goto out;
 
-			grt = rt6_lookup(gw_addr, NULL, rtmsg->rtmsg_ifindex, 1);
+			grt = rt6_lookup(gw_addr, NULL, cfg->fc_ifindex, 1);
 
 			err = -EHOSTUNREACH;
 			if (grt == NULL)
@@ -1172,7 +1162,7 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 	if (dev == NULL)
 		goto out;
 
-	if (rtmsg->rtmsg_flags & (RTF_GATEWAY|RTF_NONEXTHOP)) {
+	if (cfg->fc_flags & (RTF_GATEWAY | RTF_NONEXTHOP)) {
 		rt->rt6i_nexthop = __neigh_lookup_errno(&nd_tbl, &rt->rt6i_gateway, dev);
 		if (IS_ERR(rt->rt6i_nexthop)) {
 			err = PTR_ERR(rt->rt6i_nexthop);
@@ -1181,24 +1171,24 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
 		}
 	}
 
-	rt->rt6i_flags = rtmsg->rtmsg_flags;
+	rt->rt6i_flags = cfg->fc_flags;
 
 install_route:
-	if (rta && rta[RTA_METRICS-1]) {
-		int attrlen = RTA_PAYLOAD(rta[RTA_METRICS-1]);
-		struct rtattr *attr = RTA_DATA(rta[RTA_METRICS-1]);
-
-		while (RTA_OK(attr, attrlen)) {
-			unsigned flavor = attr->rta_type;
-			if (flavor) {
-				if (flavor > RTAX_MAX) {
+	if (cfg->fc_mx) {
+		struct nlattr *nla;
+		int remaining;
+
+		nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) {
+			int type = nla->nla_type;
+
+			if (type) {
+				if (type > RTAX_MAX) {
 					err = -EINVAL;
 					goto out;
 				}
-				rt->u.dst.metrics[flavor-1] =
-					*(u32 *)RTA_DATA(attr);
+
+				rt->u.dst.metrics[type - 1] = nla_get_u32(nla);
 			}
-			attr = RTA_NEXT(attr, attrlen);
 		}
 	}
 
@@ -1211,7 +1201,7 @@ install_route:
 	rt->u.dst.dev = dev;
 	rt->rt6i_idev = idev;
 	rt->rt6i_table = table;
-	return __ip6_ins_rt(rt, nlh, _rtattr, req);
+	return __ip6_ins_rt(rt, &cfg->fc_nlinfo);
 
 out:
 	if (dev)
@@ -1223,8 +1213,7 @@ out:
 	return err;
 }
 
-static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
-			void *_rtattr, struct netlink_skb_parms *req)
+static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 {
 	int err;
 	struct fib6_table *table;
@@ -1235,7 +1224,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
 	table = rt->rt6i_table;
 	write_lock_bh(&table->tb6_lock);
 
-	err = fib6_del(rt, nlh, _rtattr, req);
+	err = fib6_del(rt, info);
 	dst_release(&rt->u.dst);
 
 	write_unlock_bh(&table->tb6_lock);
@@ -1245,44 +1234,41 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh,
 
 int ip6_del_rt(struct rt6_info *rt)
 {
-	return __ip6_del_rt(rt, NULL, NULL, NULL);
+	return __ip6_del_rt(rt, NULL);
 }
 
-static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh,
-			 void *_rtattr, struct netlink_skb_parms *req,
-			 u32 table_id)
+static int ip6_route_del(struct fib6_config *cfg)
 {
 	struct fib6_table *table;
 	struct fib6_node *fn;
 	struct rt6_info *rt;
 	int err = -ESRCH;
 
-	table = fib6_get_table(table_id);
+	table = fib6_get_table(cfg->fc_table);
 	if (table == NULL)
 		return err;
 
 	read_lock_bh(&table->tb6_lock);
 
 	fn = fib6_locate(&table->tb6_root,
-			 &rtmsg->rtmsg_dst, rtmsg->rtmsg_dst_len,
-			 &rtmsg->rtmsg_src, rtmsg->rtmsg_src_len);
+			 &cfg->fc_dst, cfg->fc_dst_len,
+			 &cfg->fc_src, cfg->fc_src_len);
 	
 	if (fn) {
 		for (rt = fn->leaf; rt; rt = rt->u.next) {
-			if (rtmsg->rtmsg_ifindex &&
+			if (cfg->fc_ifindex &&
 			    (rt->rt6i_dev == NULL ||
-			     rt->rt6i_dev->ifindex != rtmsg->rtmsg_ifindex))
+			     rt->rt6i_dev->ifindex != cfg->fc_ifindex))
 				continue;
-			if (rtmsg->rtmsg_flags&RTF_GATEWAY &&
-			    !ipv6_addr_equal(&rtmsg->rtmsg_gateway, &rt->rt6i_gateway))
+			if (cfg->fc_flags & RTF_GATEWAY &&
+			    !ipv6_addr_equal(&cfg->fc_gateway, &rt->rt6i_gateway))
 				continue;
-			if (rtmsg->rtmsg_metric &&
-			    rtmsg->rtmsg_metric != rt->rt6i_metric)
+			if (cfg->fc_metric && cfg->fc_metric != rt->rt6i_metric)
 				continue;
 			dst_hold(&rt->u.dst);
 			read_unlock_bh(&table->tb6_lock);
 
-			return __ip6_del_rt(rt, nlh, _rtattr, req);
+			return __ip6_del_rt(rt, &cfg->fc_nlinfo);
 		}
 	}
 	read_unlock_bh(&table->tb6_lock);
@@ -1565,21 +1551,23 @@ static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixle
 					   struct in6_addr *gwaddr, int ifindex,
 					   unsigned pref)
 {
-	struct in6_rtmsg rtmsg;
+	struct fib6_config cfg = {
+		.fc_table	= RT6_TABLE_INFO,
+		.fc_metric	= 1024,
+		.fc_ifindex	= ifindex,
+		.fc_dst_len	= prefixlen,
+		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO |
+				  RTF_UP | RTF_PREF(pref),
+	};
+
+	ipv6_addr_copy(&cfg.fc_dst, prefix);
+	ipv6_addr_copy(&cfg.fc_gateway, gwaddr);
 
-	memset(&rtmsg, 0, sizeof(rtmsg));
-	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
-	ipv6_addr_copy(&rtmsg.rtmsg_dst, prefix);
-	rtmsg.rtmsg_dst_len = prefixlen;
-	ipv6_addr_copy(&rtmsg.rtmsg_gateway, gwaddr);
-	rtmsg.rtmsg_metric = 1024;
-	rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO | RTF_UP | RTF_PREF(pref);
 	/* We should treat it as a default route if prefix length is 0. */
 	if (!prefixlen)
-		rtmsg.rtmsg_flags |= RTF_DEFAULT;
-	rtmsg.rtmsg_ifindex = ifindex;
+		cfg.fc_flags |= RTF_DEFAULT;
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_INFO);
+	ip6_route_add(&cfg);
 
 	return rt6_get_route_info(prefix, prefixlen, gwaddr, ifindex);
 }
@@ -1611,18 +1599,18 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
 				     struct net_device *dev,
 				     unsigned int pref)
 {
-	struct in6_rtmsg rtmsg;
+	struct fib6_config cfg = {
+		.fc_table	= RT6_TABLE_DFLT,
+		.fc_metric	= 1024,
+		.fc_ifindex	= dev->ifindex,
+		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT |
+				  RTF_UP | RTF_EXPIRES | RTF_PREF(pref),
+	};
 
-	memset(&rtmsg, 0, sizeof(struct in6_rtmsg));
-	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
-	ipv6_addr_copy(&rtmsg.rtmsg_gateway, gwaddr);
-	rtmsg.rtmsg_metric = 1024;
-	rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP | RTF_EXPIRES |
-			    RTF_PREF(pref);
+	ipv6_addr_copy(&cfg.fc_gateway, gwaddr);
 
-	rtmsg.rtmsg_ifindex = dev->ifindex;
+	ip6_route_add(&cfg);
 
-	ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_DFLT);
 	return rt6_get_dflt_router(gwaddr, dev);
 }
 
@@ -1649,8 +1637,27 @@ restart:
 	read_unlock_bh(&table->tb6_lock);
 }
 
+static void rtmsg_to_fib6_config(struct in6_rtmsg *rtmsg,
+				 struct fib6_config *cfg)
+{
+	memset(cfg, 0, sizeof(*cfg));
+
+	cfg->fc_table = RT6_TABLE_MAIN;
+	cfg->fc_ifindex = rtmsg->rtmsg_ifindex;
+	cfg->fc_metric = rtmsg->rtmsg_metric;
+	cfg->fc_expires = rtmsg->rtmsg_info;
+	cfg->fc_dst_len = rtmsg->rtmsg_dst_len;
+	cfg->fc_src_len = rtmsg->rtmsg_src_len;
+	cfg->fc_flags = rtmsg->rtmsg_flags;
+
+	ipv6_addr_copy(&cfg->fc_dst, &rtmsg->rtmsg_dst);
+	ipv6_addr_copy(&cfg->fc_src, &rtmsg->rtmsg_src);
+	ipv6_addr_copy(&cfg->fc_gateway, &rtmsg->rtmsg_gateway);
+}
+
 int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
 {
+	struct fib6_config cfg;
 	struct in6_rtmsg rtmsg;
 	int err;
 
@@ -1663,16 +1670,16 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg)
 				     sizeof(struct in6_rtmsg));
 		if (err)
 			return -EFAULT;
-			
+
+		rtmsg_to_fib6_config(&rtmsg, &cfg);
+
 		rtnl_lock();
 		switch (cmd) {
 		case SIOCADDRT:
-			err = ip6_route_add(&rtmsg, NULL, NULL, NULL,
-					    RT6_TABLE_MAIN);
+			err = ip6_route_add(&cfg);
 			break;
 		case SIOCDELRT:
-			err = ip6_route_del(&rtmsg, NULL, NULL, NULL,
-					    RT6_TABLE_MAIN);
+			err = ip6_route_del(&cfg);
 			break;
 		default:
 			err = -EINVAL;
@@ -1823,66 +1830,104 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu)
 	fib6_clean_all(rt6_mtu_change_route, 0, &arg);
 }
 
-static int inet6_rtm_to_rtmsg(struct rtmsg *r, struct rtattr **rta,
-			      struct in6_rtmsg *rtmsg)
+static struct nla_policy rtm_ipv6_policy[RTA_MAX+1] __read_mostly = {
+	[RTA_GATEWAY]           = { .minlen = sizeof(struct in6_addr) },
+	[RTA_OIF]               = { .type = NLA_U32 },
+	[RTA_PRIORITY]          = { .type = NLA_U32 },
+	[RTA_METRICS]           = { .type = NLA_NESTED },
+};
+
+static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
+			      struct fib6_config *cfg)
 {
-	memset(rtmsg, 0, sizeof(*rtmsg));
+	struct rtmsg *rtm;
+	struct nlattr *tb[RTA_MAX+1];
+	int err;
+
+	err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy);
+	if (err < 0)
+		goto errout;
 
-	rtmsg->rtmsg_dst_len = r->rtm_dst_len;
-	rtmsg->rtmsg_src_len = r->rtm_src_len;
-	rtmsg->rtmsg_flags = RTF_UP;
-	if (r->rtm_type == RTN_UNREACHABLE)
-		rtmsg->rtmsg_flags |= RTF_REJECT;
+	err = -EINVAL;
+	rtm = nlmsg_data(nlh);
+	memset(cfg, 0, sizeof(*cfg));
 
-	if (rta[RTA_GATEWAY-1]) {
-		if (rta[RTA_GATEWAY-1]->rta_len != RTA_LENGTH(16))
-			return -EINVAL;
-		memcpy(&rtmsg->rtmsg_gateway, RTA_DATA(rta[RTA_GATEWAY-1]), 16);
-		rtmsg->rtmsg_flags |= RTF_GATEWAY;
-	}
-	if (rta[RTA_DST-1]) {
-		if (RTA_PAYLOAD(rta[RTA_DST-1]) < ((r->rtm_dst_len+7)>>3))
-			return -EINVAL;
-		memcpy(&rtmsg->rtmsg_dst, RTA_DATA(rta[RTA_DST-1]), ((r->rtm_dst_len+7)>>3));
+	cfg->fc_table = rtm->rtm_table;
+	cfg->fc_dst_len = rtm->rtm_dst_len;
+	cfg->fc_src_len = rtm->rtm_src_len;
+	cfg->fc_flags = RTF_UP;
+	cfg->fc_protocol = rtm->rtm_protocol;
+
+	if (rtm->rtm_type == RTN_UNREACHABLE)
+		cfg->fc_flags |= RTF_REJECT;
+
+	cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid;
+	cfg->fc_nlinfo.nlh = nlh;
+
+	if (tb[RTA_GATEWAY]) {
+		nla_memcpy(&cfg->fc_gateway, tb[RTA_GATEWAY], 16);
+		cfg->fc_flags |= RTF_GATEWAY;
 	}
-	if (rta[RTA_SRC-1]) {
-		if (RTA_PAYLOAD(rta[RTA_SRC-1]) < ((r->rtm_src_len+7)>>3))
-			return -EINVAL;
-		memcpy(&rtmsg->rtmsg_src, RTA_DATA(rta[RTA_SRC-1]), ((r->rtm_src_len+7)>>3));
+
+	if (tb[RTA_DST]) {
+		int plen = (rtm->rtm_dst_len + 7) >> 3;
+
+		if (nla_len(tb[RTA_DST]) < plen)
+			goto errout;
+
+		nla_memcpy(&cfg->fc_dst, tb[RTA_DST], plen);
 	}
-	if (rta[RTA_OIF-1]) {
-		if (rta[RTA_OIF-1]->rta_len != RTA_LENGTH(sizeof(int)))
-			return -EINVAL;
-		memcpy(&rtmsg->rtmsg_ifindex, RTA_DATA(rta[RTA_OIF-1]), sizeof(int));
+
+	if (tb[RTA_SRC]) {
+		int plen = (rtm->rtm_src_len + 7) >> 3;
+
+		if (nla_len(tb[RTA_SRC]) < plen)
+			goto errout;
+
+		nla_memcpy(&cfg->fc_src, tb[RTA_SRC], plen);
 	}
-	if (rta[RTA_PRIORITY-1]) {
-		if (rta[RTA_PRIORITY-1]->rta_len != RTA_LENGTH(4))
-			return -EINVAL;
-		memcpy(&rtmsg->rtmsg_metric, RTA_DATA(rta[RTA_PRIORITY-1]), 4);
+
+	if (tb[RTA_OIF])
+		cfg->fc_ifindex = nla_get_u32(tb[RTA_OIF]);
+
+	if (tb[RTA_PRIORITY])
+		cfg->fc_metric = nla_get_u32(tb[RTA_PRIORITY]);
+
+	if (tb[RTA_METRICS]) {
+		cfg->fc_mx = nla_data(tb[RTA_METRICS]);
+		cfg->fc_mx_len = nla_len(tb[RTA_METRICS]);
 	}
-	return 0;
+
+	if (tb[RTA_TABLE])
+		cfg->fc_table = nla_get_u32(tb[RTA_TABLE]);
+
+	err = 0;
+errout:
+	return err;
 }
 
 int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct rtmsg *r = NLMSG_DATA(nlh);
-	struct in6_rtmsg rtmsg;
+	struct fib6_config cfg;
+	int err;
 
-	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
-		return -EINVAL;
-	return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb),
-			     rtm_get_table(arg, r->rtm_table));
+	err = rtm_to_fib6_config(skb, nlh, &cfg);
+	if (err < 0)
+		return err;
+
+	return ip6_route_del(&cfg);
 }
 
 int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
 {
-	struct rtmsg *r = NLMSG_DATA(nlh);
-	struct in6_rtmsg rtmsg;
+	struct fib6_config cfg;
+	int err;
 
-	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
-		return -EINVAL;
-	return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb),
-			     rtm_get_table(arg, r->rtm_table));
+	err = rtm_to_fib6_config(skb, nlh, &cfg);
+	if (err < 0)
+		return err;
+
+	return ip6_route_add(&cfg);
 }
 
 static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
@@ -2063,15 +2108,21 @@ out_free:
 	goto out;	
 }
 
-void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh, 
-			struct netlink_skb_parms *req)
+void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
 {
 	struct sk_buff *skb;
-	u32 pid = req ? req->pid : 0;
-	u32 seq = nlh ? nlh->nlmsg_seq : 0;
+	u32 pid = 0, seq = 0;
+	struct nlmsghdr *nlh = NULL;
 	int payload = sizeof(struct rtmsg) + 256;
 	int err = -ENOBUFS;
 
+	if (info) {
+		pid = info->pid;
+		nlh = info->nlh;
+		if (nlh)
+			seq = nlh->nlmsg_seq;
+	}
+
 	skb = nlmsg_new(nlmsg_total_size(payload), gfp_any());
 	if (skb == NULL)
 		goto errout;
-- 
cgit v1.2.3


From ac0b04627269ff16c3c7ab854a65fe6780c6e3e5 Mon Sep 17 00:00:00 2001
From: Sridhar Samudrala <sri@us.ibm.com>
Date: Tue, 22 Aug 2006 00:15:33 -0700
Subject: [SCTP]: Extend /proc/net/sctp/snmp to provide more statistics.

This patch adds more statistics info under /proc/net/sctp/snmp
that should be useful for debugging. The additional events that
are counted now include timer expirations, retransmits, packet
and data chunk discards.

The Data chunk discards include all the cases where a data chunk
is discarded including high tsn, bad stream, dup tsn and the most
useful one(out of receive buffer/rwnd).

Also moved the SCTP MIB data structures from the generic include
directories to include/sctp/sctp.h.

Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/snmp.h    | 33 ---------------------------------
 include/net/sctp/sctp.h | 44 ++++++++++++++++++++++++++++++++++++++++++++
 include/net/snmp.h      |  6 ------
 net/sctp/input.c        |  8 ++++++--
 net/sctp/inqueue.c      |  4 ++--
 net/sctp/outqueue.c     |  6 +++++-
 net/sctp/proc.c         | 17 ++++++++++++++++-
 net/sctp/sm_statefuns.c | 15 +++++++++++++++
 8 files changed, 88 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/linux/snmp.h b/include/linux/snmp.h
index 30156556f78d..854aa6b543f1 100644
--- a/include/linux/snmp.h
+++ b/include/linux/snmp.h
@@ -160,39 +160,6 @@ enum
 	__UDP_MIB_MAX
 };
 
-/* sctp mib definitions */
-/*
- * draft-ietf-sigtran-sctp-mib-07.txt
- */
-enum
-{
-	SCTP_MIB_NUM = 0,
-	SCTP_MIB_CURRESTAB,			/* CurrEstab */
-	SCTP_MIB_ACTIVEESTABS,			/* ActiveEstabs */
-	SCTP_MIB_PASSIVEESTABS,			/* PassiveEstabs */
-	SCTP_MIB_ABORTEDS,			/* Aborteds */
-	SCTP_MIB_SHUTDOWNS,			/* Shutdowns */
-	SCTP_MIB_OUTOFBLUES,			/* OutOfBlues */
-	SCTP_MIB_CHECKSUMERRORS,		/* ChecksumErrors */
-	SCTP_MIB_OUTCTRLCHUNKS,			/* OutCtrlChunks */
-	SCTP_MIB_OUTORDERCHUNKS,		/* OutOrderChunks */
-	SCTP_MIB_OUTUNORDERCHUNKS,		/* OutUnorderChunks */
-	SCTP_MIB_INCTRLCHUNKS,			/* InCtrlChunks */
-	SCTP_MIB_INORDERCHUNKS,			/* InOrderChunks */
-	SCTP_MIB_INUNORDERCHUNKS,		/* InUnorderChunks */
-	SCTP_MIB_FRAGUSRMSGS,			/* FragUsrMsgs */
-	SCTP_MIB_REASMUSRMSGS,			/* ReasmUsrMsgs */
-	SCTP_MIB_OUTSCTPPACKS,			/* OutSCTPPacks */
-	SCTP_MIB_INSCTPPACKS,			/* InSCTPPacks */
-	SCTP_MIB_RTOALGORITHM,			/* RtoAlgorithm */
-	SCTP_MIB_RTOMIN,			/* RtoMin */
-	SCTP_MIB_RTOMAX,			/* RtoMax */
-	SCTP_MIB_RTOINITIAL,			/* RtoInitial */
-	SCTP_MIB_VALCOOKIELIFE,			/* ValCookieLife */
-	SCTP_MIB_MAXINITRETR,			/* MaxInitRetr */
-	__SCTP_MIB_MAX
-};
-
 /* linux mib definitions */
 enum
 {
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 1c1abce5f6b6..e274fd479990 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -216,6 +216,50 @@ DECLARE_SNMP_STAT(struct sctp_mib, sctp_statistics);
 
 #endif /* !TEST_FRAME */
 
+/* sctp mib definitions */
+enum
+{
+	SCTP_MIB_NUM = 0,
+	SCTP_MIB_CURRESTAB,			/* CurrEstab */
+	SCTP_MIB_ACTIVEESTABS,			/* ActiveEstabs */
+	SCTP_MIB_PASSIVEESTABS,			/* PassiveEstabs */
+	SCTP_MIB_ABORTEDS,			/* Aborteds */
+	SCTP_MIB_SHUTDOWNS,			/* Shutdowns */
+	SCTP_MIB_OUTOFBLUES,			/* OutOfBlues */
+	SCTP_MIB_CHECKSUMERRORS,		/* ChecksumErrors */
+	SCTP_MIB_OUTCTRLCHUNKS,			/* OutCtrlChunks */
+	SCTP_MIB_OUTORDERCHUNKS,		/* OutOrderChunks */
+	SCTP_MIB_OUTUNORDERCHUNKS,		/* OutUnorderChunks */
+	SCTP_MIB_INCTRLCHUNKS,			/* InCtrlChunks */
+	SCTP_MIB_INORDERCHUNKS,			/* InOrderChunks */
+	SCTP_MIB_INUNORDERCHUNKS,		/* InUnorderChunks */
+	SCTP_MIB_FRAGUSRMSGS,			/* FragUsrMsgs */
+	SCTP_MIB_REASMUSRMSGS,			/* ReasmUsrMsgs */
+	SCTP_MIB_OUTSCTPPACKS,			/* OutSCTPPacks */
+	SCTP_MIB_INSCTPPACKS,			/* InSCTPPacks */
+	SCTP_MIB_T1_INIT_EXPIREDS,
+	SCTP_MIB_T1_COOKIE_EXPIREDS,
+	SCTP_MIB_T2_SHUTDOWN_EXPIREDS,
+	SCTP_MIB_T3_RTX_EXPIREDS,
+	SCTP_MIB_T4_RTO_EXPIREDS,
+	SCTP_MIB_T5_SHUTDOWN_GUARD_EXPIREDS,
+	SCTP_MIB_DELAY_SACK_EXPIREDS,
+	SCTP_MIB_AUTOCLOSE_EXPIREDS,
+	SCTP_MIB_T3_RETRANSMITS,
+	SCTP_MIB_PMTUD_RETRANSMITS,
+	SCTP_MIB_FAST_RETRANSMITS,
+	SCTP_MIB_IN_PKT_SOFTIRQ,
+	SCTP_MIB_IN_PKT_BACKLOG,
+	SCTP_MIB_IN_PKT_DISCARDS,
+	SCTP_MIB_IN_DATA_CHUNK_DISCARDS,
+	__SCTP_MIB_MAX
+};
+
+#define SCTP_MIB_MAX    __SCTP_MIB_MAX
+struct sctp_mib {
+        unsigned long   mibs[SCTP_MIB_MAX];
+} __SNMP_MIB_ALIGN__;
+
 
 /* Print debugging messages.  */
 #if SCTP_DEBUG
diff --git a/include/net/snmp.h b/include/net/snmp.h
index a36bed8ea210..464970e39ec0 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -100,12 +100,6 @@ struct udp_mib {
 	unsigned long	mibs[UDP_MIB_MAX];
 } __SNMP_MIB_ALIGN__;
 
-/* SCTP */
-#define SCTP_MIB_MAX	__SCTP_MIB_MAX
-struct sctp_mib {
-	unsigned long	mibs[SCTP_MIB_MAX];
-} __SNMP_MIB_ALIGN__;
-
 /* Linux */
 #define LINUX_MIB_MAX	__LINUX_MIB_MAX
 struct linux_mib {
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 42b66e74bbb5..8a34d95602ce 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -255,10 +255,13 @@ int sctp_rcv(struct sk_buff *skb)
 	 */
 	sctp_bh_lock_sock(sk);
 
-	if (sock_owned_by_user(sk))
+	if (sock_owned_by_user(sk)) {
+		SCTP_INC_STATS_BH(SCTP_MIB_IN_PKT_BACKLOG);
 		sctp_add_backlog(sk, skb);
-	else
+	} else {
+		SCTP_INC_STATS_BH(SCTP_MIB_IN_PKT_SOFTIRQ);
 		sctp_inq_push(&chunk->rcvr->inqueue, chunk);
+	}
 
 	sctp_bh_unlock_sock(sk);
 
@@ -271,6 +274,7 @@ int sctp_rcv(struct sk_buff *skb)
 	return 0;
 
 discard_it:
+	SCTP_INC_STATS_BH(SCTP_MIB_IN_PKT_DISCARDS);
 	kfree_skb(skb);
 	return 0;
 
diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c
index cf0c767d43ae..cf6deed7e849 100644
--- a/net/sctp/inqueue.c
+++ b/net/sctp/inqueue.c
@@ -87,7 +87,7 @@ void sctp_inq_free(struct sctp_inq *queue)
 /* Put a new packet in an SCTP inqueue.
  * We assume that packet->sctp_hdr is set and in host byte order.
  */
-void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *packet)
+void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *chunk)
 {
 	/* Directly call the packet handling routine. */
 
@@ -96,7 +96,7 @@ void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *packet)
 	 * Eventually, we should clean up inqueue to not rely
 	 * on the BH related data structures.
 	 */
-	list_add_tail(&packet->list, &q->in_chunk_list);
+	list_add_tail(&chunk->list, &q->in_chunk_list);
 	q->immediate.func(q->immediate.data);
 }
 
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 30b710c54e64..37074a39ecbb 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -467,6 +467,7 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport,
 
 	switch(reason) {
 	case SCTP_RTXR_T3_RTX:
+		SCTP_INC_STATS(SCTP_MIB_T3_RETRANSMITS);
 		sctp_transport_lower_cwnd(transport, SCTP_LOWER_CWND_T3_RTX);
 		/* Update the retran path if the T3-rtx timer has expired for
 		 * the current retran path.
@@ -475,12 +476,15 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport,
 			sctp_assoc_update_retran_path(transport->asoc);
 		break;
 	case SCTP_RTXR_FAST_RTX:
+		SCTP_INC_STATS(SCTP_MIB_FAST_RETRANSMITS);
 		sctp_transport_lower_cwnd(transport, SCTP_LOWER_CWND_FAST_RTX);
 		fast_retransmit = 1;
 		break;
 	case SCTP_RTXR_PMTUD:
-	default:
+		SCTP_INC_STATS(SCTP_MIB_PMTUD_RETRANSMITS);
 		break;
+	default:
+		BUG();
 	}
 
 	sctp_retransmit_mark(q, transport, fast_retransmit);
diff --git a/net/sctp/proc.c b/net/sctp/proc.c
index 5b3b0e0ae7e5..a356d8d310a9 100644
--- a/net/sctp/proc.c
+++ b/net/sctp/proc.c
@@ -57,6 +57,21 @@ static struct snmp_mib sctp_snmp_list[] = {
 	SNMP_MIB_ITEM("SctpReasmUsrMsgs", SCTP_MIB_REASMUSRMSGS),
 	SNMP_MIB_ITEM("SctpOutSCTPPacks", SCTP_MIB_OUTSCTPPACKS),
 	SNMP_MIB_ITEM("SctpInSCTPPacks", SCTP_MIB_INSCTPPACKS),
+	SNMP_MIB_ITEM("SctpT1InitExpireds", SCTP_MIB_T1_INIT_EXPIREDS),
+	SNMP_MIB_ITEM("SctpT1CookieExpireds", SCTP_MIB_T1_COOKIE_EXPIREDS),
+	SNMP_MIB_ITEM("SctpT2ShutdownExpireds", SCTP_MIB_T2_SHUTDOWN_EXPIREDS),
+	SNMP_MIB_ITEM("SctpT3RtxExpireds", SCTP_MIB_T3_RTX_EXPIREDS),
+	SNMP_MIB_ITEM("SctpT4RtoExpireds", SCTP_MIB_T4_RTO_EXPIREDS),
+	SNMP_MIB_ITEM("SctpT5ShutdownGuardExpireds", SCTP_MIB_T5_SHUTDOWN_GUARD_EXPIREDS),
+	SNMP_MIB_ITEM("SctpDelaySackExpireds", SCTP_MIB_DELAY_SACK_EXPIREDS),
+	SNMP_MIB_ITEM("SctpAutocloseExpireds", SCTP_MIB_AUTOCLOSE_EXPIREDS),
+	SNMP_MIB_ITEM("SctpT3Retransmits", SCTP_MIB_T3_RETRANSMITS),
+	SNMP_MIB_ITEM("SctpPmtudRetransmits", SCTP_MIB_PMTUD_RETRANSMITS),
+	SNMP_MIB_ITEM("SctpFastRetransmits", SCTP_MIB_FAST_RETRANSMITS),
+	SNMP_MIB_ITEM("SctpInPktSoftirq", SCTP_MIB_IN_PKT_SOFTIRQ),
+	SNMP_MIB_ITEM("SctpInPktBacklog", SCTP_MIB_IN_PKT_BACKLOG),
+	SNMP_MIB_ITEM("SctpInPktDiscards", SCTP_MIB_IN_PKT_DISCARDS),
+	SNMP_MIB_ITEM("SctpInDataChunkDiscards", SCTP_MIB_IN_DATA_CHUNK_DISCARDS),
 	SNMP_MIB_SENTINEL
 };
 
@@ -328,8 +343,8 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v)
 			   "%8p %8p %-3d %-3d %-2d %-4d %4d %8d %8d %7d %5lu %-5d %5d ",
 			   assoc, sk, sctp_sk(sk)->type, sk->sk_state,
 			   assoc->state, hash, assoc->assoc_id,
-			   (sk->sk_rcvbuf - assoc->rwnd),
 			   assoc->sndbuf_used,
+			   (sk->sk_rcvbuf - assoc->rwnd),
 			   sock_i_uid(sk), sock_i_ino(sk),
 			   epb->bind_addr.port,
 			   assoc->peer.port);
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 5b5ae7958322..32f57f42af9e 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -2663,9 +2663,11 @@ sctp_disposition_t sctp_sf_eat_data_6_2(const struct sctp_endpoint *ep,
 		break;
 	case SCTP_IERROR_HIGH_TSN:
 	case SCTP_IERROR_BAD_STREAM:
+		SCTP_INC_STATS(SCTP_MIB_IN_DATA_CHUNK_DISCARDS);
 		goto discard_noforce;
 	case SCTP_IERROR_DUP_TSN:
 	case SCTP_IERROR_IGNORE_TSN:
+		SCTP_INC_STATS(SCTP_MIB_IN_DATA_CHUNK_DISCARDS);
 		goto discard_force;
 	case SCTP_IERROR_NO_DATA:
 		goto consume;
@@ -3652,6 +3654,7 @@ sctp_disposition_t sctp_sf_pdiscard(const struct sctp_endpoint *ep,
 				    void *arg,
 				    sctp_cmd_seq_t *commands)
 {
+	SCTP_INC_STATS(SCTP_MIB_IN_PKT_DISCARDS);
 	sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET, SCTP_NULL());
 
 	return SCTP_DISPOSITION_CONSUME;
@@ -4548,6 +4551,8 @@ sctp_disposition_t sctp_sf_do_6_3_3_rtx(const struct sctp_endpoint *ep,
 {
 	struct sctp_transport *transport = arg;
 
+	SCTP_INC_STATS(SCTP_MIB_T3_RTX_EXPIREDS);
+
 	if (asoc->overall_error_count >= asoc->max_retrans) {
 		sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR,
 				SCTP_ERROR(ETIMEDOUT));
@@ -4616,6 +4621,7 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep,
 				       void *arg,
 				       sctp_cmd_seq_t *commands)
 {
+	SCTP_INC_STATS(SCTP_MIB_DELAY_SACK_EXPIREDS);
 	sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_FORCE());
 	return SCTP_DISPOSITION_CONSUME;
 }
@@ -4650,6 +4656,7 @@ sctp_disposition_t sctp_sf_t1_init_timer_expire(const struct sctp_endpoint *ep,
 	int attempts = asoc->init_err_counter + 1;
 
 	SCTP_DEBUG_PRINTK("Timer T1 expired (INIT).\n");
+	SCTP_INC_STATS(SCTP_MIB_T1_INIT_EXPIREDS);
 
 	if (attempts <= asoc->max_init_attempts) {
 		bp = (struct sctp_bind_addr *) &asoc->base.bind_addr;
@@ -4709,6 +4716,7 @@ sctp_disposition_t sctp_sf_t1_cookie_timer_expire(const struct sctp_endpoint *ep
 	int attempts = asoc->init_err_counter + 1;
 
 	SCTP_DEBUG_PRINTK("Timer T1 expired (COOKIE-ECHO).\n");
+	SCTP_INC_STATS(SCTP_MIB_T1_COOKIE_EXPIREDS);
 
 	if (attempts <= asoc->max_init_attempts) {
 		repl = sctp_make_cookie_echo(asoc, NULL);
@@ -4753,6 +4761,8 @@ sctp_disposition_t sctp_sf_t2_timer_expire(const struct sctp_endpoint *ep,
 	struct sctp_chunk *reply = NULL;
 
 	SCTP_DEBUG_PRINTK("Timer T2 expired.\n");
+	SCTP_INC_STATS(SCTP_MIB_T2_SHUTDOWN_EXPIREDS);
+
 	if (asoc->overall_error_count >= asoc->max_retrans) {
 		sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR,
 				SCTP_ERROR(ETIMEDOUT));
@@ -4814,6 +4824,8 @@ sctp_disposition_t sctp_sf_t4_timer_expire(
 	struct sctp_chunk *chunk = asoc->addip_last_asconf;
 	struct sctp_transport *transport = chunk->transport;
 
+	SCTP_INC_STATS(SCTP_MIB_T4_RTO_EXPIREDS);
+
 	/* ADDIP 4.1 B1) Increment the error counters and perform path failure
 	 * detection on the appropriate destination address as defined in
 	 * RFC2960 [5] section 8.1 and 8.2.
@@ -4880,6 +4892,7 @@ sctp_disposition_t sctp_sf_t5_timer_expire(const struct sctp_endpoint *ep,
 	struct sctp_chunk *reply = NULL;
 
 	SCTP_DEBUG_PRINTK("Timer T5 expired.\n");
+	SCTP_INC_STATS(SCTP_MIB_T5_SHUTDOWN_GUARD_EXPIREDS);
 
 	reply = sctp_make_abort(asoc, NULL, 0);
 	if (!reply)
@@ -4910,6 +4923,8 @@ sctp_disposition_t sctp_sf_autoclose_timer_expire(
 {
 	int disposition;
 
+	SCTP_INC_STATS(SCTP_MIB_AUTOCLOSE_EXPIREDS);
+
 	/* From 9.2 Shutdown of an Association
 	 * Upon receipt of the SHUTDOWN primitive from its upper
 	 * layer, the endpoint enters SHUTDOWN-PENDING state and
-- 
cgit v1.2.3


From 9ba1627617d396135a4d679542a3623d5819e628 Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Tue, 22 Aug 2006 00:29:37 -0700
Subject: [NETFILTER]: x_tables: replace IPv4 dscp match by address family
 independent version

This replaces IPv4 dscp match by address family independent version.
This also
	- utilizes dsfield.h to get the DS field in IPv4/IPv6 header, and
	- checks for the DSCP value from user space.
	- fixes Kconfig help text.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_dscp.h       |  23 +++++++
 include/linux/netfilter_ipv4/ipt_dscp.h |  14 ++--
 net/ipv4/netfilter/Kconfig              |  11 ----
 net/ipv4/netfilter/Makefile             |   1 -
 net/ipv4/netfilter/ipt_dscp.c           |  54 ---------------
 net/netfilter/Kconfig                   |  11 ++++
 net/netfilter/Makefile                  |   1 +
 net/netfilter/xt_dscp.c                 | 113 ++++++++++++++++++++++++++++++++
 8 files changed, 154 insertions(+), 74 deletions(-)
 create mode 100644 include/linux/netfilter/xt_dscp.h
 delete mode 100644 net/ipv4/netfilter/ipt_dscp.c
 create mode 100644 net/netfilter/xt_dscp.c

(limited to 'include')

diff --git a/include/linux/netfilter/xt_dscp.h b/include/linux/netfilter/xt_dscp.h
new file mode 100644
index 000000000000..1da61e6acaf7
--- /dev/null
+++ b/include/linux/netfilter/xt_dscp.h
@@ -0,0 +1,23 @@
+/* x_tables module for matching the IPv4/IPv6 DSCP field
+ *
+ * (C) 2002 Harald Welte <laforge@gnumonks.org>
+ * This software is distributed under GNU GPL v2, 1991
+ *
+ * See RFC2474 for a description of the DSCP field within the IP Header.
+ *
+ * xt_dscp.h,v 1.3 2002/08/05 19:00:21 laforge Exp
+*/
+#ifndef _XT_DSCP_H
+#define _XT_DSCP_H
+
+#define XT_DSCP_MASK	0xfc	/* 11111100 */
+#define XT_DSCP_SHIFT	2
+#define XT_DSCP_MAX	0x3f	/* 00111111 */
+
+/* match info */
+struct xt_dscp_info {
+	u_int8_t dscp;
+	u_int8_t invert;
+};
+
+#endif /* _XT_DSCP_H */
diff --git a/include/linux/netfilter_ipv4/ipt_dscp.h b/include/linux/netfilter_ipv4/ipt_dscp.h
index 2fa6dfe92894..4b82ca912b0e 100644
--- a/include/linux/netfilter_ipv4/ipt_dscp.h
+++ b/include/linux/netfilter_ipv4/ipt_dscp.h
@@ -10,14 +10,12 @@
 #ifndef _IPT_DSCP_H
 #define _IPT_DSCP_H
 
-#define IPT_DSCP_MASK	0xfc	/* 11111100 */
-#define IPT_DSCP_SHIFT	2
-#define IPT_DSCP_MAX	0x3f	/* 00111111 */
+#include <linux/netfilter/xt_dscp.h>
 
-/* match info */
-struct ipt_dscp_info {
-	u_int8_t dscp;
-	u_int8_t invert;
-};
+#define IPT_DSCP_MASK	XT_DSCP_MASK
+#define IPT_DSCP_SHIFT	XT_DSCP_SHIFT
+#define IPT_DSCP_MAX	XT_DSCP_MAX
+
+#define ipt_dscp_info	xt_dscp_info
 
 #endif /* _IPT_DSCP_H */
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index ef0b5aac5838..d88d71d1ce0d 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -278,17 +278,6 @@ config IP_NF_MATCH_ECN
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-config IP_NF_MATCH_DSCP
-	tristate "DSCP match support"
-	depends on IP_NF_IPTABLES
-	help
-	  This option adds a `DSCP' match, which allows you to match against
-	  the IPv4 header DSCP field (DSCP codepoint).
-
-	  The DSCP codepoint can have any value between 0x0 and 0x4f.
-
-	  To compile it as a module, choose M here.  If unsure, say N.
-
 config IP_NF_MATCH_AH
 	tristate "AH match support"
 	depends on IP_NF_IPTABLES
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index 3ded4a3af59c..b946b0f3ea9d 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -59,7 +59,6 @@ obj-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o
 obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o
 obj-$(CONFIG_IP_NF_MATCH_RECENT) += ipt_recent.o
 obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o
-obj-$(CONFIG_IP_NF_MATCH_DSCP) += ipt_dscp.o
 obj-$(CONFIG_IP_NF_MATCH_AH) += ipt_ah.o
 obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o
 obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o
diff --git a/net/ipv4/netfilter/ipt_dscp.c b/net/ipv4/netfilter/ipt_dscp.c
deleted file mode 100644
index 47177591aeb6..000000000000
--- a/net/ipv4/netfilter/ipt_dscp.c
+++ /dev/null
@@ -1,54 +0,0 @@
-/* IP tables module for matching the value of the IPv4 DSCP field
- *
- * ipt_dscp.c,v 1.3 2002/08/05 19:00:21 laforge Exp
- *
- * (C) 2002 by Harald Welte <laforge@netfilter.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/module.h>
-#include <linux/skbuff.h>
-
-#include <linux/netfilter_ipv4/ipt_dscp.h>
-#include <linux/netfilter_ipv4/ip_tables.h>
-
-MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
-MODULE_DESCRIPTION("iptables DSCP matching module");
-MODULE_LICENSE("GPL");
-
-static int match(const struct sk_buff *skb,
-		 const struct net_device *in, const struct net_device *out,
-		 const struct xt_match *match, const void *matchinfo,
-		 int offset, unsigned int protoff, int *hotdrop)
-{
-	const struct ipt_dscp_info *info = matchinfo;
-	const struct iphdr *iph = skb->nh.iph;
-
-	u_int8_t sh_dscp = ((info->dscp << IPT_DSCP_SHIFT) & IPT_DSCP_MASK);
-
-	return ((iph->tos&IPT_DSCP_MASK) == sh_dscp) ^ info->invert;
-}
-
-static struct ipt_match dscp_match = {
-	.name		= "dscp",
-	.match		= match,
-	.matchsize	= sizeof(struct ipt_dscp_info),
-	.me		= THIS_MODULE,
-};
-
-static int __init ipt_dscp_init(void)
-{
-	return ipt_register_match(&dscp_match);
-}
-
-static void __exit ipt_dscp_fini(void)
-{
-	ipt_unregister_match(&dscp_match);
-
-}
-
-module_init(ipt_dscp_init);
-module_exit(ipt_dscp_fini);
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index a9894ddfd72a..f781405f5d65 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -263,6 +263,17 @@ config NETFILTER_XT_MATCH_DCCP
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/modules.txt>.  If unsure, say `N'.
 
+config NETFILTER_XT_MATCH_DSCP
+	tristate '"DSCP" match support'
+	depends on NETFILTER_XTABLES
+	help
+	  This option adds a `DSCP' match, which allows you to match against
+	  the IPv4/IPv6 header DSCP field (differentiated services codepoint).
+
+	  The DSCP field can have any value between 0x0 and 0x3f inclusive.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config NETFILTER_XT_MATCH_ESP
 	tristate '"ESP" match support'
 	depends on NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 6fa4b7580458..0b8a70c1df46 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -37,6 +37,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_CONNBYTES) += xt_connbytes.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_CONNMARK) += xt_connmark.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_CONNTRACK) += xt_conntrack.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_DCCP) += xt_dccp.o
+obj-$(CONFIG_NETFILTER_XT_MATCH_DSCP) += xt_dscp.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_ESP) += xt_esp.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_HELPER) += xt_helper.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_LENGTH) += xt_length.o
diff --git a/net/netfilter/xt_dscp.c b/net/netfilter/xt_dscp.c
new file mode 100644
index 000000000000..82e250d1f007
--- /dev/null
+++ b/net/netfilter/xt_dscp.c
@@ -0,0 +1,113 @@
+/* IP tables module for matching the value of the IPv4/IPv6 DSCP field
+ *
+ * xt_dscp.c,v 1.3 2002/08/05 19:00:21 laforge Exp
+ *
+ * (C) 2002 by Harald Welte <laforge@netfilter.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <net/dsfield.h>
+
+#include <linux/netfilter/xt_dscp.h>
+#include <linux/netfilter/x_tables.h>
+
+MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
+MODULE_DESCRIPTION("x_tables DSCP matching module");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("ipt_dscp");
+MODULE_ALIAS("ip6t_dscp");
+
+static int match(const struct sk_buff *skb,
+		 const struct net_device *in,
+		 const struct net_device *out,
+		 const struct xt_match *match,
+		 const void *matchinfo,
+		 int offset,
+		 unsigned int protoff,
+		 int *hotdrop)
+{
+	const struct xt_dscp_info *info = matchinfo;
+	u_int8_t dscp = ipv4_get_dsfield(skb->nh.iph) >> XT_DSCP_SHIFT;
+
+	return (dscp == info->dscp) ^ !!info->invert;
+}
+
+static int match6(const struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  const struct xt_match *match,
+		  const void *matchinfo,
+		  int offset,
+		  unsigned int protoff,
+		  int *hotdrop)
+{
+	const struct xt_dscp_info *info = matchinfo;
+	u_int8_t dscp = ipv6_get_dsfield(skb->nh.ipv6h) >> XT_DSCP_SHIFT;
+
+	return (dscp == info->dscp) ^ !!info->invert;
+}
+
+static int checkentry(const char *tablename,
+		      const void *info,
+		      const struct xt_match *match,
+		      void *matchinfo,
+		      unsigned int matchsize,
+		      unsigned int hook_mask)
+{
+	const u_int8_t dscp = ((struct xt_dscp_info *)matchinfo)->dscp;
+
+	if (dscp > XT_DSCP_MAX) {
+		printk(KERN_ERR "xt_dscp: dscp %x out of range\n", dscp);
+		return 0;
+	}
+
+	return 1;
+}
+
+static struct xt_match dscp_match = {
+	.name		= "dscp",
+	.match		= match,
+	.checkentry	= checkentry,
+	.matchsize	= sizeof(struct xt_dscp_info),
+	.family		= AF_INET,
+	.me		= THIS_MODULE,
+};
+
+static struct xt_match dscp6_match = {
+	.name		= "dscp",
+	.match		= match6,
+	.checkentry	= checkentry,
+	.matchsize	= sizeof(struct xt_dscp_info),
+	.family		= AF_INET6,
+	.me		= THIS_MODULE,
+};
+
+static int __init xt_dscp_match_init(void)
+{
+	int ret;
+	ret = xt_register_match(&dscp_match);
+	if (ret)
+		return ret;
+
+	ret = xt_register_match(&dscp6_match);
+	if (ret)
+		xt_unregister_match(&dscp_match);
+
+	return ret;
+}
+
+static void __exit xt_dscp_match_fini(void)
+{
+	xt_unregister_match(&dscp_match);
+	xt_unregister_match(&dscp6_match);
+}
+
+module_init(xt_dscp_match_init);
+module_exit(xt_dscp_match_fini);
-- 
cgit v1.2.3


From a468701db58a8b3e08e3f55fa6ac66db42014922 Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Tue, 22 Aug 2006 00:30:26 -0700
Subject: [NETFILTER]: x_tables: replace IPv4 DSCP target by address family
 independent version

This replaces IPv4 DSCP target by address family independent version.
This also
	- utilizes dsfield.h to get/mangle DS field in IPv4/IPv6 header
	- fixes Kconfig help text.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_DSCP.h       |  20 +++++
 include/linux/netfilter_ipv4/ipt_DSCP.h |   6 +-
 net/ipv4/netfilter/Kconfig              |  11 ---
 net/ipv4/netfilter/Makefile             |   1 -
 net/ipv4/netfilter/ipt_DSCP.c           |  96 -----------------------
 net/netfilter/Kconfig                   |  12 +++
 net/netfilter/Makefile                  |   1 +
 net/netfilter/xt_DSCP.c                 | 130 ++++++++++++++++++++++++++++++++
 8 files changed, 165 insertions(+), 112 deletions(-)
 create mode 100644 include/linux/netfilter/xt_DSCP.h
 delete mode 100644 net/ipv4/netfilter/ipt_DSCP.c
 create mode 100644 net/netfilter/xt_DSCP.c

(limited to 'include')

diff --git a/include/linux/netfilter/xt_DSCP.h b/include/linux/netfilter/xt_DSCP.h
new file mode 100644
index 000000000000..3c7c963997bd
--- /dev/null
+++ b/include/linux/netfilter/xt_DSCP.h
@@ -0,0 +1,20 @@
+/* x_tables module for setting the IPv4/IPv6 DSCP field
+ *
+ * (C) 2002 Harald Welte <laforge@gnumonks.org>
+ * based on ipt_FTOS.c (C) 2000 by Matthew G. Marsh <mgm@paktronix.com>
+ * This software is distributed under GNU GPL v2, 1991
+ *
+ * See RFC2474 for a description of the DSCP field within the IP Header.
+ *
+ * xt_DSCP.h,v 1.7 2002/03/14 12:03:13 laforge Exp
+*/
+#ifndef _XT_DSCP_TARGET_H
+#define _XT_DSCP_TARGET_H
+#include <linux/netfilter/xt_dscp.h>
+
+/* target info */
+struct xt_DSCP_info {
+	u_int8_t dscp;
+};
+
+#endif /* _XT_DSCP_TARGET_H */
diff --git a/include/linux/netfilter_ipv4/ipt_DSCP.h b/include/linux/netfilter_ipv4/ipt_DSCP.h
index b30f510b5bef..3491e524d5ea 100644
--- a/include/linux/netfilter_ipv4/ipt_DSCP.h
+++ b/include/linux/netfilter_ipv4/ipt_DSCP.h
@@ -11,10 +11,8 @@
 #ifndef _IPT_DSCP_TARGET_H
 #define _IPT_DSCP_TARGET_H
 #include <linux/netfilter_ipv4/ipt_dscp.h>
+#include <linux/netfilter/xt_DSCP.h>
 
-/* target info */
-struct ipt_DSCP_info {
-	u_int8_t dscp;
-};
+#define ipt_DSCP_info xt_DSCP_info
 
 #endif /* _IPT_DSCP_TARGET_H */
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index d88d71d1ce0d..a55b8ff70ded 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -557,17 +557,6 @@ config IP_NF_TARGET_ECN
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-config IP_NF_TARGET_DSCP
-	tristate "DSCP target support"
-	depends on IP_NF_MANGLE
-	help
-	  This option adds a `DSCP' match, which allows you to match against
-	  the IPv4 header DSCP field (DSCP codepoint).
-
-	  The DSCP codepoint can have any value between 0x0 and 0x4f.
-
-	  To compile it as a module, choose M here.  If unsure, say N.
-
 config IP_NF_TARGET_TTL
 	tristate  'TTL target support'
 	depends on IP_NF_MANGLE
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index b946b0f3ea9d..09aaed1a8063 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -67,7 +67,6 @@ obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o
 obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o
 obj-$(CONFIG_IP_NF_TARGET_TOS) += ipt_TOS.o
 obj-$(CONFIG_IP_NF_TARGET_ECN) += ipt_ECN.o
-obj-$(CONFIG_IP_NF_TARGET_DSCP) += ipt_DSCP.o
 obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o
 obj-$(CONFIG_IP_NF_TARGET_REDIRECT) += ipt_REDIRECT.o
 obj-$(CONFIG_IP_NF_TARGET_NETMAP) += ipt_NETMAP.o
diff --git a/net/ipv4/netfilter/ipt_DSCP.c b/net/ipv4/netfilter/ipt_DSCP.c
deleted file mode 100644
index c8e971288dfe..000000000000
--- a/net/ipv4/netfilter/ipt_DSCP.c
+++ /dev/null
@@ -1,96 +0,0 @@
-/* iptables module for setting the IPv4 DSCP field, Version 1.8
- *
- * (C) 2002 by Harald Welte <laforge@netfilter.org>
- * based on ipt_FTOS.c (C) 2000 by Matthew G. Marsh <mgm@paktronix.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as 
- * published by the Free Software Foundation.
- * 
- * See RFC2474 for a description of the DSCP field within the IP Header.
- *
- * ipt_DSCP.c,v 1.8 2002/08/06 18:41:57 laforge Exp
-*/
-
-#include <linux/module.h>
-#include <linux/skbuff.h>
-#include <linux/ip.h>
-#include <net/checksum.h>
-
-#include <linux/netfilter_ipv4/ip_tables.h>
-#include <linux/netfilter_ipv4/ipt_DSCP.h>
-
-MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
-MODULE_DESCRIPTION("iptables DSCP modification module");
-MODULE_LICENSE("GPL");
-
-static unsigned int
-target(struct sk_buff **pskb,
-       const struct net_device *in,
-       const struct net_device *out,
-       unsigned int hooknum,
-       const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
-{
-	const struct ipt_DSCP_info *dinfo = targinfo;
-	u_int8_t sh_dscp = ((dinfo->dscp << IPT_DSCP_SHIFT) & IPT_DSCP_MASK);
-
-
-	if (((*pskb)->nh.iph->tos & IPT_DSCP_MASK) != sh_dscp) {
-		u_int16_t diffs[2];
-
-		if (!skb_make_writable(pskb, sizeof(struct iphdr)))
-			return NF_DROP;
-
-		diffs[0] = htons((*pskb)->nh.iph->tos) ^ 0xFFFF;
-		(*pskb)->nh.iph->tos = ((*pskb)->nh.iph->tos & ~IPT_DSCP_MASK)
-			| sh_dscp;
-		diffs[1] = htons((*pskb)->nh.iph->tos);
-		(*pskb)->nh.iph->check
-			= csum_fold(csum_partial((char *)diffs,
-						 sizeof(diffs),
-						 (*pskb)->nh.iph->check
-						 ^ 0xFFFF));
-	}
-	return IPT_CONTINUE;
-}
-
-static int
-checkentry(const char *tablename,
-	   const void *e_void,
-	   const struct xt_target *target,
-           void *targinfo,
-           unsigned int targinfosize,
-           unsigned int hook_mask)
-{
-	const u_int8_t dscp = ((struct ipt_DSCP_info *)targinfo)->dscp;
-
-	if ((dscp > IPT_DSCP_MAX)) {
-		printk(KERN_WARNING "DSCP: dscp %x out of range\n", dscp);
-		return 0;
-	}
-	return 1;
-}
-
-static struct ipt_target ipt_dscp_reg = {
-	.name		= "DSCP",
-	.target		= target,
-	.targetsize	= sizeof(struct ipt_DSCP_info),
-	.table		= "mangle",
-	.checkentry	= checkentry,
-	.me		= THIS_MODULE,
-};
-
-static int __init ipt_dscp_init(void)
-{
-	return ipt_register_target(&ipt_dscp_reg);
-}
-
-static void __exit ipt_dscp_fini(void)
-{
-	ipt_unregister_target(&ipt_dscp_reg);
-}
-
-module_init(ipt_dscp_init);
-module_exit(ipt_dscp_fini);
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index f781405f5d65..0a28d2c5c44f 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -148,6 +148,18 @@ config NETFILTER_XT_TARGET_CONNMARK
 	  <file:Documentation/modules.txt>.  The module will be called
 	  ipt_CONNMARK.o.  If unsure, say `N'.
 
+config NETFILTER_XT_TARGET_DSCP
+	tristate '"DSCP" target support'
+	depends on NETFILTER_XTABLES
+	depends on IP_NF_MANGLE || IP6_NF_MANGLE
+	help
+	  This option adds a `DSCP' target, which allows you to manipulate
+	  the IPv4/IPv6 header DSCP field (differentiated services codepoint).
+
+	  The DSCP field can have any value between 0x0 and 0x3f inclusive.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config NETFILTER_XT_TARGET_MARK
 	tristate '"MARK" target support'
 	depends on NETFILTER_XTABLES
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 0b8a70c1df46..a74be492fd0a 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o
 # targets
 obj-$(CONFIG_NETFILTER_XT_TARGET_CLASSIFY) += xt_CLASSIFY.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_CONNMARK) += xt_CONNMARK.o
+obj-$(CONFIG_NETFILTER_XT_TARGET_DSCP) += xt_DSCP.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_MARK) += xt_MARK.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o
 obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o
diff --git a/net/netfilter/xt_DSCP.c b/net/netfilter/xt_DSCP.c
new file mode 100644
index 000000000000..79df8165cd79
--- /dev/null
+++ b/net/netfilter/xt_DSCP.c
@@ -0,0 +1,130 @@
+/* x_tables module for setting the IPv4/IPv6 DSCP field, Version 1.8
+ *
+ * (C) 2002 by Harald Welte <laforge@netfilter.org>
+ * based on ipt_FTOS.c (C) 2000 by Matthew G. Marsh <mgm@paktronix.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * See RFC2474 for a description of the DSCP field within the IP Header.
+ *
+ * xt_DSCP.c,v 1.8 2002/08/06 18:41:57 laforge Exp
+*/
+
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <net/dsfield.h>
+
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_DSCP.h>
+
+MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
+MODULE_DESCRIPTION("x_tables DSCP modification module");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("ipt_DSCP");
+MODULE_ALIAS("ip6t_DSCP");
+
+static unsigned int target(struct sk_buff **pskb,
+			   const struct net_device *in,
+			   const struct net_device *out,
+			   unsigned int hooknum,
+			   const struct xt_target *target,
+			   const void *targinfo,
+			   void *userinfo)
+{
+	const struct xt_DSCP_info *dinfo = targinfo;
+	u_int8_t dscp = ipv4_get_dsfield((*pskb)->nh.iph) >> XT_DSCP_SHIFT;
+
+	if (dscp != dinfo->dscp) {
+		if (!skb_make_writable(pskb, sizeof(struct iphdr)))
+			return NF_DROP;
+
+		ipv4_change_dsfield((*pskb)->nh.iph, (__u8)(~XT_DSCP_MASK),
+				    dinfo->dscp << XT_DSCP_SHIFT);
+
+	}
+	return XT_CONTINUE;
+}
+
+static unsigned int target6(struct sk_buff **pskb,
+			    const struct net_device *in,
+			    const struct net_device *out,
+			    unsigned int hooknum,
+			    const struct xt_target *target,
+			    const void *targinfo,
+			    void *userinfo)
+{
+	const struct xt_DSCP_info *dinfo = targinfo;
+	u_int8_t dscp = ipv6_get_dsfield((*pskb)->nh.ipv6h) >> XT_DSCP_SHIFT;
+
+	if (dscp != dinfo->dscp) {
+		if (!skb_make_writable(pskb, sizeof(struct ipv6hdr)))
+			return NF_DROP;
+
+		ipv6_change_dsfield((*pskb)->nh.ipv6h, (__u8)(~XT_DSCP_MASK),
+				    dinfo->dscp << XT_DSCP_SHIFT);
+	}
+	return XT_CONTINUE;
+}
+
+static int checkentry(const char *tablename,
+		      const void *e_void,
+		      const struct xt_target *target,
+		      void *targinfo,
+		      unsigned int targinfosize,
+		      unsigned int hook_mask)
+{
+	const u_int8_t dscp = ((struct xt_DSCP_info *)targinfo)->dscp;
+
+	if ((dscp > XT_DSCP_MAX)) {
+		printk(KERN_WARNING "DSCP: dscp %x out of range\n", dscp);
+		return 0;
+	}
+	return 1;
+}
+
+static struct xt_target xt_dscp_reg = {
+	.name		= "DSCP",
+	.target		= target,
+	.targetsize	= sizeof(struct xt_DSCP_info),
+	.table		= "mangle",
+	.checkentry	= checkentry,
+	.family		= AF_INET,
+	.me		= THIS_MODULE,
+};
+
+static struct xt_target xt_dscp6_reg = {
+	.name		= "DSCP",
+	.target		= target6,
+	.targetsize	= sizeof(struct xt_DSCP_info),
+	.table		= "mangle",
+	.checkentry	= checkentry,
+	.family		= AF_INET6,
+	.me		= THIS_MODULE,
+};
+
+static int __init xt_dscp_target_init(void)
+{
+	int ret;
+	ret = xt_register_target(&xt_dscp_reg);
+	if (ret)
+		return ret;
+
+	ret = xt_register_target(&xt_dscp6_reg);
+	if (ret)
+		xt_unregister_target(&xt_dscp_reg);
+
+	return ret;
+}
+
+static void __exit xt_dscp_target_fini(void)
+{
+	xt_unregister_target(&xt_dscp_reg);
+	xt_unregister_target(&xt_dscp6_reg);
+}
+
+module_init(xt_dscp_target_init);
+module_exit(xt_dscp_target_fini);
-- 
cgit v1.2.3


From 2521c12cf1a29f6c380b13ca32a38175f6beed08 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 22 Aug 2006 00:31:24 -0700
Subject: [NETFILTER]: conntrack: introduce connection mark event

This patch introduces the mark event. ctnetlink can use this to know if
the mark needs to be dumped.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nf_conntrack_common.h |  4 ++++
 net/netfilter/xt_CONNMARK.c                   | 16 ++++++++++++++--
 2 files changed, 18 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index d2e4bd7a7a14..9e0dae07861e 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -125,6 +125,10 @@ enum ip_conntrack_events
 	/* Counter highest bit has been set */
 	IPCT_COUNTER_FILLING_BIT = 11,
 	IPCT_COUNTER_FILLING = (1 << IPCT_COUNTER_FILLING_BIT),
+
+	/* Mark is set */
+	IPCT_MARK_BIT = 12,
+	IPCT_MARK = (1 << IPCT_MARK_BIT),
 };
 
 enum ip_conntrack_expect_events {
diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c
index 60c375d36f01..784482b74e58 100644
--- a/net/netfilter/xt_CONNMARK.c
+++ b/net/netfilter/xt_CONNMARK.c
@@ -52,13 +52,25 @@ target(struct sk_buff **pskb,
 	    switch(markinfo->mode) {
 	    case XT_CONNMARK_SET:
 		newmark = (*ctmark & ~markinfo->mask) | markinfo->mark;
-		if (newmark != *ctmark)
+		if (newmark != *ctmark) {
 		    *ctmark = newmark;
+#ifdef CONFIG_IP_NF_CONNTRACK_EVENTS
+		    ip_conntrack_event_cache(IPCT_MARK, *pskb);
+#else
+		    nf_conntrack_event_cache(IPCT_MARK, *pskb);
+#endif
+		}
 		break;
 	    case XT_CONNMARK_SAVE:
 		newmark = (*ctmark & ~markinfo->mask) | ((*pskb)->nfmark & markinfo->mask);
-		if (*ctmark != newmark)
+		if (*ctmark != newmark) {
 		    *ctmark = newmark;
+#ifdef CONFIG_IP_NF_CONNTRACK_EVENTS
+		    ip_conntrack_event_cache(IPCT_MARK, *pskb);
+#else
+		    nf_conntrack_event_cache(IPCT_MARK, *pskb);
+#endif
+		}
 		break;
 	    case XT_CONNMARK_RESTORE:
 		nfmark = (*pskb)->nfmark;
-- 
cgit v1.2.3


From 52d9c42ef2563d2c420eb23b96bf5a4cae9e167b Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Aug 2006 00:33:45 -0700
Subject: [NETFILTER]: x_tables: add helpers for mass match/target registration

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h |  5 ++++
 net/netfilter/x_tables.c           | 60 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 65 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 48cc32d83f77..9a9912430e3a 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -290,8 +290,13 @@ struct xt_table_info
 
 extern int xt_register_target(struct xt_target *target);
 extern void xt_unregister_target(struct xt_target *target);
+extern int xt_register_targets(struct xt_target *target, unsigned int n);
+extern void xt_unregister_targets(struct xt_target *target, unsigned int n);
+
 extern int xt_register_match(struct xt_match *target);
 extern void xt_unregister_match(struct xt_match *target);
+extern int xt_register_matches(struct xt_match *match, unsigned int n);
+extern void xt_unregister_matches(struct xt_match *match, unsigned int n);
 
 extern int xt_check_match(const struct xt_match *match, unsigned short family,
 			  unsigned int size, const char *table, unsigned int hook,
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index 174e8f970095..8037ba63d587 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -86,6 +86,36 @@ xt_unregister_target(struct xt_target *target)
 }
 EXPORT_SYMBOL(xt_unregister_target);
 
+int
+xt_register_targets(struct xt_target *target, unsigned int n)
+{
+	unsigned int i;
+	int err = 0;
+
+	for (i = 0; i < n; i++) {
+		err = xt_register_target(&target[i]);
+		if (err)
+			goto err;
+	}
+	return err;
+
+err:
+	if (i > 0)
+		xt_unregister_targets(target, i);
+	return err;
+}
+EXPORT_SYMBOL(xt_register_targets);
+
+void
+xt_unregister_targets(struct xt_target *target, unsigned int n)
+{
+	unsigned int i;
+
+	for (i = 0; i < n; i++)
+		xt_unregister_target(&target[i]);
+}
+EXPORT_SYMBOL(xt_unregister_targets);
+
 int
 xt_register_match(struct xt_match *match)
 {
@@ -113,6 +143,36 @@ xt_unregister_match(struct xt_match *match)
 }
 EXPORT_SYMBOL(xt_unregister_match);
 
+int
+xt_register_matches(struct xt_match *match, unsigned int n)
+{
+	unsigned int i;
+	int err = 0;
+
+	for (i = 0; i < n; i++) {
+		err = xt_register_match(&match[i]);
+		if (err)
+			goto err;
+	}
+	return err;
+
+err:
+	if (i > 0)
+		xt_unregister_matches(match, i);
+	return err;
+}
+EXPORT_SYMBOL(xt_register_matches);
+
+void
+xt_unregister_matches(struct xt_match *match, unsigned int n)
+{
+	unsigned int i;
+
+	for (i = 0; i < n; i++)
+		xt_unregister_match(&match[i]);
+}
+EXPORT_SYMBOL(xt_unregister_matches);
+
 
 /*
  * These are weird, but module loading must not be done with mutex
-- 
cgit v1.2.3


From fe1cb10873b44cf89082465823ee6d4d4ac63ad7 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Aug 2006 00:35:47 -0700
Subject: [NETFILTER]: x_tables: remove unused argument to target functions

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h        | 3 +--
 include/linux/netfilter_arp/arp_tables.h  | 3 +--
 include/linux/netfilter_ipv4/ip_tables.h  | 3 +--
 include/linux/netfilter_ipv6/ip6_tables.h | 3 +--
 net/ipv4/netfilter/arp_tables.c           | 9 +++------
 net/ipv4/netfilter/arpt_mangle.c          | 2 +-
 net/ipv4/netfilter/arptable_filter.c      | 2 +-
 net/ipv4/netfilter/ip_nat_rule.c          | 8 +++-----
 net/ipv4/netfilter/ip_tables.c            | 9 +++------
 net/ipv4/netfilter/ipt_CLUSTERIP.c        | 3 +--
 net/ipv4/netfilter/ipt_ECN.c              | 3 +--
 net/ipv4/netfilter/ipt_LOG.c              | 3 +--
 net/ipv4/netfilter/ipt_MASQUERADE.c       | 3 +--
 net/ipv4/netfilter/ipt_NETMAP.c           | 3 +--
 net/ipv4/netfilter/ipt_REDIRECT.c         | 3 +--
 net/ipv4/netfilter/ipt_REJECT.c           | 3 +--
 net/ipv4/netfilter/ipt_SAME.c             | 3 +--
 net/ipv4/netfilter/ipt_TCPMSS.c           | 3 +--
 net/ipv4/netfilter/ipt_TOS.c              | 3 +--
 net/ipv4/netfilter/ipt_TTL.c              | 2 +-
 net/ipv4/netfilter/ipt_ULOG.c             | 2 +-
 net/ipv4/netfilter/iptable_filter.c       | 4 ++--
 net/ipv4/netfilter/iptable_mangle.c       | 4 ++--
 net/ipv4/netfilter/iptable_raw.c          | 2 +-
 net/ipv6/netfilter/ip6_tables.c           | 9 +++------
 net/ipv6/netfilter/ip6t_HL.c              | 2 +-
 net/ipv6/netfilter/ip6t_LOG.c             | 3 +--
 net/ipv6/netfilter/ip6t_REJECT.c          | 3 +--
 net/ipv6/netfilter/ip6table_filter.c      | 4 ++--
 net/ipv6/netfilter/ip6table_mangle.c      | 4 ++--
 net/ipv6/netfilter/ip6table_raw.c         | 2 +-
 net/netfilter/xt_CLASSIFY.c               | 3 +--
 net/netfilter/xt_CONNMARK.c               | 3 +--
 net/netfilter/xt_CONNSECMARK.c            | 2 +-
 net/netfilter/xt_DSCP.c                   | 6 ++----
 net/netfilter/xt_MARK.c                   | 6 ++----
 net/netfilter/xt_NFQUEUE.c                | 3 +--
 net/netfilter/xt_NOTRACK.c                | 3 +--
 net/netfilter/xt_SECMARK.c                | 2 +-
 net/netfilter/xt_connbytes.c              | 2 +-
 net/sched/act_ipt.c                       | 2 +-
 41 files changed, 55 insertions(+), 90 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 9a9912430e3a..9cef0e91542b 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -211,8 +211,7 @@ struct xt_target
 			       const struct net_device *out,
 			       unsigned int hooknum,
 			       const struct xt_target *target,
-			       const void *targinfo,
-			       void *userdata);
+			       const void *targinfo);
 
 	/* Called when user tries to insert an entry of this type:
            hook_mask is a bitmask of hooks from which it can be
diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 62cc27daca4e..149e87c9ab13 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -248,8 +248,7 @@ extern unsigned int arpt_do_table(struct sk_buff **pskb,
 				  unsigned int hook,
 				  const struct net_device *in,
 				  const struct net_device *out,
-				  struct arpt_table *table,
-				  void *userdata);
+				  struct arpt_table *table);
 
 #define ARPT_ALIGN(s) (((s) + (__alignof__(struct arpt_entry)-1)) & ~(__alignof__(struct arpt_entry)-1))
 #endif /*__KERNEL__*/
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index c0dac16e1902..a536bbdef145 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -312,8 +312,7 @@ extern unsigned int ipt_do_table(struct sk_buff **pskb,
 				 unsigned int hook,
 				 const struct net_device *in,
 				 const struct net_device *out,
-				 struct ipt_table *table,
-				 void *userdata);
+				 struct ipt_table *table);
 
 #define IPT_ALIGN(s) XT_ALIGN(s)
 
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index d0d5d1ee4be3..d7a8e9c0dad0 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -300,8 +300,7 @@ extern unsigned int ip6t_do_table(struct sk_buff **pskb,
 				  unsigned int hook,
 				  const struct net_device *in,
 				  const struct net_device *out,
-				  struct ip6t_table *table,
-				  void *userdata);
+				  struct ip6t_table *table);
 
 /* Check for an extension */
 extern int ip6t_ext_hdr(u8 nexthdr);
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 8d1d7a6e72a5..c6bd270bf46a 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -208,8 +208,7 @@ static unsigned int arpt_error(struct sk_buff **pskb,
 			       const struct net_device *out,
 			       unsigned int hooknum,
 			       const struct xt_target *target,
-			       const void *targinfo,
-			       void *userinfo)
+			       const void *targinfo)
 {
 	if (net_ratelimit())
 		printk("arp_tables: error: '%s'\n", (char *)targinfo);
@@ -226,8 +225,7 @@ unsigned int arpt_do_table(struct sk_buff **pskb,
 			   unsigned int hook,
 			   const struct net_device *in,
 			   const struct net_device *out,
-			   struct arpt_table *table,
-			   void *userdata)
+			   struct arpt_table *table)
 {
 	static const char nulldevname[IFNAMSIZ];
 	unsigned int verdict = NF_DROP;
@@ -302,8 +300,7 @@ unsigned int arpt_do_table(struct sk_buff **pskb,
 								     in, out,
 								     hook,
 								     t->u.kernel.target,
-								     t->data,
-								     userdata);
+								     t->data);
 
 				/* Target might have changed stuff. */
 				arp = (*pskb)->nh.arph;
diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c
index a58325c1ceb9..05fb2421bb26 100644
--- a/net/ipv4/netfilter/arpt_mangle.c
+++ b/net/ipv4/netfilter/arpt_mangle.c
@@ -11,7 +11,7 @@ static unsigned int
 target(struct sk_buff **pskb,
        const struct net_device *in, const struct net_device *out,
        unsigned int hooknum, const struct xt_target *target,
-       const void *targinfo, void *userinfo)
+       const void *targinfo)
 {
 	const struct arpt_mangle *mangle = targinfo;
 	struct arphdr *arp;
diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c
index d7c472faa53b..7edea2a1696c 100644
--- a/net/ipv4/netfilter/arptable_filter.c
+++ b/net/ipv4/netfilter/arptable_filter.c
@@ -155,7 +155,7 @@ static unsigned int arpt_hook(unsigned int hook,
 			      const struct net_device *out,
 			      int (*okfn)(struct sk_buff *))
 {
-	return arpt_do_table(pskb, hook, in, out, &packet_filter, NULL);
+	return arpt_do_table(pskb, hook, in, out, &packet_filter);
 }
 
 static struct nf_hook_ops arpt_ops[] = {
diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c
index 1aba926c1cb0..1aa0e4f462a5 100644
--- a/net/ipv4/netfilter/ip_nat_rule.c
+++ b/net/ipv4/netfilter/ip_nat_rule.c
@@ -104,8 +104,7 @@ static unsigned int ipt_snat_target(struct sk_buff **pskb,
 				    const struct net_device *out,
 				    unsigned int hooknum,
 				    const struct ipt_target *target,
-				    const void *targinfo,
-				    void *userinfo)
+				    const void *targinfo)
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
@@ -147,8 +146,7 @@ static unsigned int ipt_dnat_target(struct sk_buff **pskb,
 				    const struct net_device *out,
 				    unsigned int hooknum,
 				    const struct ipt_target *target,
-				    const void *targinfo,
-				    void *userinfo)
+				    const void *targinfo)
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
@@ -255,7 +253,7 @@ int ip_nat_rule_find(struct sk_buff **pskb,
 {
 	int ret;
 
-	ret = ipt_do_table(pskb, hooknum, in, out, &nat_table, NULL);
+	ret = ipt_do_table(pskb, hooknum, in, out, &nat_table);
 
 	if (ret == NF_ACCEPT) {
 		if (!ip_nat_initialized(ct, HOOK2MANIP(hooknum)))
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 048514f15f2f..8ce5b6f76447 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -180,8 +180,7 @@ ipt_error(struct sk_buff **pskb,
 	  const struct net_device *out,
 	  unsigned int hooknum,
 	  const struct xt_target *target,
-	  const void *targinfo,
-	  void *userinfo)
+	  const void *targinfo)
 {
 	if (net_ratelimit())
 		printk("ip_tables: error: `%s'\n", (char *)targinfo);
@@ -217,8 +216,7 @@ ipt_do_table(struct sk_buff **pskb,
 	     unsigned int hook,
 	     const struct net_device *in,
 	     const struct net_device *out,
-	     struct ipt_table *table,
-	     void *userdata)
+	     struct ipt_table *table)
 {
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	u_int16_t offset;
@@ -308,8 +306,7 @@ ipt_do_table(struct sk_buff **pskb,
 								     in, out,
 								     hook,
 								     t->u.kernel.target,
-								     t->data,
-								     userdata);
+								     t->data);
 
 #ifdef CONFIG_NETFILTER_DEBUG
 				if (((struct ipt_entry *)table_base)->comefrom
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index d994c5f5744c..a08383cf9e7a 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -302,8 +302,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	const struct ipt_clusterip_tgt_info *cipinfo = targinfo;
 	enum ip_conntrack_info ctinfo;
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index 7e30e6d2b5da..1c3da4a48e5f 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -85,8 +85,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	const struct ipt_ECN_info *einfo = targinfo;
 
diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c
index b98f7b08b084..a8d356c6191f 100644
--- a/net/ipv4/netfilter/ipt_LOG.c
+++ b/net/ipv4/netfilter/ipt_LOG.c
@@ -416,8 +416,7 @@ ipt_log_target(struct sk_buff **pskb,
 	       const struct net_device *out,
 	       unsigned int hooknum,
 	       const struct xt_target *target,
-	       const void *targinfo,
-	       void *userinfo)
+	       const void *targinfo)
 {
 	const struct ipt_log_info *loginfo = targinfo;
 	struct nf_loginfo li;
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index ebd94f2abf0d..9659793c66c0 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -64,8 +64,7 @@ masquerade_target(struct sk_buff **pskb,
 		  const struct net_device *out,
 		  unsigned int hooknum,
 		  const struct xt_target *target,
-		  const void *targinfo,
-		  void *userinfo)
+		  const void *targinfo)
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c
index 736c4b5a86a7..fd5e74a19fb5 100644
--- a/net/ipv4/netfilter/ipt_NETMAP.c
+++ b/net/ipv4/netfilter/ipt_NETMAP.c
@@ -55,8 +55,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c
index f290463232de..839fe99f71d4 100644
--- a/net/ipv4/netfilter/ipt_REDIRECT.c
+++ b/net/ipv4/netfilter/ipt_REDIRECT.c
@@ -58,8 +58,7 @@ redirect_target(struct sk_buff **pskb,
 		const struct net_device *out,
 		unsigned int hooknum,
 		const struct xt_target *target,
-		const void *targinfo,
-		void *userinfo)
+		const void *targinfo)
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index 95c6662b663c..1dfd8e56be8b 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -228,8 +228,7 @@ static unsigned int reject(struct sk_buff **pskb,
 			   const struct net_device *out,
 			   unsigned int hooknum,
 			   const struct xt_target *target,
-			   const void *targinfo,
-			   void *userinfo)
+			   const void *targinfo)
 {
 	const struct ipt_reject_info *reject = targinfo;
 
diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c
index 7169b09b5a67..cf801749490f 100644
--- a/net/ipv4/netfilter/ipt_SAME.c
+++ b/net/ipv4/netfilter/ipt_SAME.c
@@ -133,8 +133,7 @@ same_target(struct sk_buff **pskb,
 		const struct net_device *out,
 		unsigned int hooknum,
 		const struct xt_target *target,
-		const void *targinfo,
-		void *userinfo)
+		const void *targinfo)
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c
index 0fce85e05507..6d668dcfc22a 100644
--- a/net/ipv4/netfilter/ipt_TCPMSS.c
+++ b/net/ipv4/netfilter/ipt_TCPMSS.c
@@ -41,8 +41,7 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 		  const struct net_device *out,
 		  unsigned int hooknum,
 		  const struct xt_target *target,
-		  const void *targinfo,
-		  void *userinfo)
+		  const void *targinfo)
 {
 	const struct ipt_tcpmss_info *tcpmssinfo = targinfo;
 	struct tcphdr *tcph;
diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c
index 52e9d705d48e..043df0137084 100644
--- a/net/ipv4/netfilter/ipt_TOS.c
+++ b/net/ipv4/netfilter/ipt_TOS.c
@@ -26,8 +26,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	const struct ipt_tos_target_info *tosinfo = targinfo;
 	struct iphdr *iph = (*pskb)->nh.iph;
diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c
index 2afb2a8aa8c5..164007107b5e 100644
--- a/net/ipv4/netfilter/ipt_TTL.c
+++ b/net/ipv4/netfilter/ipt_TTL.c
@@ -23,7 +23,7 @@ static unsigned int
 ipt_ttl_target(struct sk_buff **pskb,
 	       const struct net_device *in, const struct net_device *out,
 	       unsigned int hooknum, const struct xt_target *target,
-	       const void *targinfo, void *userinfo)
+	       const void *targinfo)
 {
 	struct iphdr *iph;
 	const struct ipt_TTL_info *info = targinfo;
diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c
index d46fd677fa11..4c5f0a117862 100644
--- a/net/ipv4/netfilter/ipt_ULOG.c
+++ b/net/ipv4/netfilter/ipt_ULOG.c
@@ -308,7 +308,7 @@ static unsigned int ipt_ulog_target(struct sk_buff **pskb,
 				    const struct net_device *out,
 				    unsigned int hooknum,
 				    const struct xt_target *target,
-				    const void *targinfo, void *userinfo)
+				    const void *targinfo)
 {
 	struct ipt_ulog_info *loginfo = (struct ipt_ulog_info *) targinfo;
 
diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c
index 7f417484bfbf..e2e7dd8d7903 100644
--- a/net/ipv4/netfilter/iptable_filter.c
+++ b/net/ipv4/netfilter/iptable_filter.c
@@ -90,7 +90,7 @@ ipt_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ipt_do_table(pskb, hook, in, out, &packet_filter, NULL);
+	return ipt_do_table(pskb, hook, in, out, &packet_filter);
 }
 
 static unsigned int
@@ -108,7 +108,7 @@ ipt_local_out_hook(unsigned int hook,
 		return NF_ACCEPT;
 	}
 
-	return ipt_do_table(pskb, hook, in, out, &packet_filter, NULL);
+	return ipt_do_table(pskb, hook, in, out, &packet_filter);
 }
 
 static struct nf_hook_ops ipt_ops[] = {
diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c
index 4e7998beda63..79336cb42527 100644
--- a/net/ipv4/netfilter/iptable_mangle.c
+++ b/net/ipv4/netfilter/iptable_mangle.c
@@ -119,7 +119,7 @@ ipt_route_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ipt_do_table(pskb, hook, in, out, &packet_mangler, NULL);
+	return ipt_do_table(pskb, hook, in, out, &packet_mangler);
 }
 
 static unsigned int
@@ -148,7 +148,7 @@ ipt_local_hook(unsigned int hook,
 	daddr = (*pskb)->nh.iph->daddr;
 	tos = (*pskb)->nh.iph->tos;
 
-	ret = ipt_do_table(pskb, hook, in, out, &packet_mangler, NULL);
+	ret = ipt_do_table(pskb, hook, in, out, &packet_mangler);
 	/* Reroute for ANY change. */
 	if (ret != NF_DROP && ret != NF_STOLEN && ret != NF_QUEUE
 	    && ((*pskb)->nh.iph->saddr != saddr
diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c
index 7912cce1e1b8..bcbeb4aeacd9 100644
--- a/net/ipv4/netfilter/iptable_raw.c
+++ b/net/ipv4/netfilter/iptable_raw.c
@@ -95,7 +95,7 @@ ipt_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ipt_do_table(pskb, hook, in, out, &packet_raw, NULL);
+	return ipt_do_table(pskb, hook, in, out, &packet_raw);
 }
 
 /* 'raw' is the very first table. */
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index c9d6b23cd3f7..38cd7ffda9a0 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -220,8 +220,7 @@ ip6t_error(struct sk_buff **pskb,
 	  const struct net_device *out,
 	  unsigned int hooknum,
 	  const struct xt_target *target,
-	  const void *targinfo,
-	  void *userinfo)
+	  const void *targinfo)
 {
 	if (net_ratelimit())
 		printk("ip6_tables: error: `%s'\n", (char *)targinfo);
@@ -258,8 +257,7 @@ ip6t_do_table(struct sk_buff **pskb,
 	      unsigned int hook,
 	      const struct net_device *in,
 	      const struct net_device *out,
-	      struct xt_table *table,
-	      void *userdata)
+	      struct xt_table *table)
 {
 	static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long))));
 	int offset = 0;
@@ -349,8 +347,7 @@ ip6t_do_table(struct sk_buff **pskb,
 								     in, out,
 								     hook,
 								     t->u.kernel.target,
-								     t->data,
-								     userdata);
+								     t->data);
 
 #ifdef CONFIG_NETFILTER_DEBUG
 				if (((struct ip6t_entry *)table_base)->comefrom
diff --git a/net/ipv6/netfilter/ip6t_HL.c b/net/ipv6/netfilter/ip6t_HL.c
index b8eff8ee69b1..c85d124f9a3d 100644
--- a/net/ipv6/netfilter/ip6t_HL.c
+++ b/net/ipv6/netfilter/ip6t_HL.c
@@ -22,7 +22,7 @@ static unsigned int ip6t_hl_target(struct sk_buff **pskb,
 				   const struct net_device *out,
 				   unsigned int hooknum,
 				   const struct xt_target *target,
-				   const void *targinfo, void *userinfo)
+				   const void *targinfo)
 {
 	struct ipv6hdr *ip6h;
 	const struct ip6t_HL_info *info = targinfo;
diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c
index 73c6300109d6..acb91733e1fd 100644
--- a/net/ipv6/netfilter/ip6t_LOG.c
+++ b/net/ipv6/netfilter/ip6t_LOG.c
@@ -427,8 +427,7 @@ ip6t_log_target(struct sk_buff **pskb,
 		const struct net_device *out,
 		unsigned int hooknum,
 		const struct xt_target *target,
-		const void *targinfo,
-		void *userinfo)
+		const void *targinfo)
 {
 	const struct ip6t_log_info *loginfo = targinfo;
 	struct nf_loginfo li;
diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index 7929ff402166..343acd3cbf5e 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -180,8 +180,7 @@ static unsigned int reject6_target(struct sk_buff **pskb,
 			   const struct net_device *out,
 			   unsigned int hooknum,
 			   const struct xt_target *target,
-			   const void *targinfo,
-			   void *userinfo)
+			   const void *targinfo)
 {
 	const struct ip6t_reject_info *reject = targinfo;
 
diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c
index 60976c0c58e8..2fc07c74decf 100644
--- a/net/ipv6/netfilter/ip6table_filter.c
+++ b/net/ipv6/netfilter/ip6table_filter.c
@@ -108,7 +108,7 @@ ip6t_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ip6t_do_table(pskb, hook, in, out, &packet_filter, NULL);
+	return ip6t_do_table(pskb, hook, in, out, &packet_filter);
 }
 
 static unsigned int
@@ -128,7 +128,7 @@ ip6t_local_out_hook(unsigned int hook,
 	}
 #endif
 
-	return ip6t_do_table(pskb, hook, in, out, &packet_filter, NULL);
+	return ip6t_do_table(pskb, hook, in, out, &packet_filter);
 }
 
 static struct nf_hook_ops ip6t_ops[] = {
diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c
index 03a13eab1dae..32db04fd8310 100644
--- a/net/ipv6/netfilter/ip6table_mangle.c
+++ b/net/ipv6/netfilter/ip6table_mangle.c
@@ -138,7 +138,7 @@ ip6t_route_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ip6t_do_table(pskb, hook, in, out, &packet_mangler, NULL);
+	return ip6t_do_table(pskb, hook, in, out, &packet_mangler);
 }
 
 static unsigned int
@@ -174,7 +174,7 @@ ip6t_local_hook(unsigned int hook,
 	/* flowlabel and prio (includes version, which shouldn't change either */
 	flowlabel = *((u_int32_t *) (*pskb)->nh.ipv6h);
 
-	ret = ip6t_do_table(pskb, hook, in, out, &packet_mangler, NULL);
+	ret = ip6t_do_table(pskb, hook, in, out, &packet_mangler);
 
 	if (ret != NF_DROP && ret != NF_STOLEN 
 		&& (memcmp(&(*pskb)->nh.ipv6h->saddr, &saddr, sizeof(saddr))
diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c
index 61a7c58e99f8..b4154da575c0 100644
--- a/net/ipv6/netfilter/ip6table_raw.c
+++ b/net/ipv6/netfilter/ip6table_raw.c
@@ -122,7 +122,7 @@ ip6t_hook(unsigned int hook,
 	 const struct net_device *out,
 	 int (*okfn)(struct sk_buff *))
 {
-	return ip6t_do_table(pskb, hook, in, out, &packet_raw, NULL);
+	return ip6t_do_table(pskb, hook, in, out, &packet_raw);
 }
 
 static struct nf_hook_ops ip6t_ops[] = { 
diff --git a/net/netfilter/xt_CLASSIFY.c b/net/netfilter/xt_CLASSIFY.c
index 1f92edd05933..50de965bb104 100644
--- a/net/netfilter/xt_CLASSIFY.c
+++ b/net/netfilter/xt_CLASSIFY.c
@@ -29,8 +29,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	const struct xt_classify_target_info *clinfo = targinfo;
 
diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c
index e577356b5c71..c2125f6ee128 100644
--- a/net/netfilter/xt_CONNMARK.c
+++ b/net/netfilter/xt_CONNMARK.c
@@ -38,8 +38,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	const struct xt_connmark_target_info *markinfo = targinfo;
 	u_int32_t diff;
diff --git a/net/netfilter/xt_CONNSECMARK.c b/net/netfilter/xt_CONNSECMARK.c
index 48f7fc3c85cd..4b9cc65bb82b 100644
--- a/net/netfilter/xt_CONNSECMARK.c
+++ b/net/netfilter/xt_CONNSECMARK.c
@@ -66,7 +66,7 @@ static void secmark_restore(struct sk_buff *skb)
 static unsigned int target(struct sk_buff **pskb, const struct net_device *in,
 			   const struct net_device *out, unsigned int hooknum,
 			   const struct xt_target *target,
-			   const void *targinfo, void *userinfo)
+			   const void *targinfo)
 {
 	struct sk_buff *skb = *pskb;
 	const struct xt_connsecmark_target_info *info = targinfo;
diff --git a/net/netfilter/xt_DSCP.c b/net/netfilter/xt_DSCP.c
index a1cd9723644f..9d23c9580d80 100644
--- a/net/netfilter/xt_DSCP.c
+++ b/net/netfilter/xt_DSCP.c
@@ -32,8 +32,7 @@ static unsigned int target(struct sk_buff **pskb,
 			   const struct net_device *out,
 			   unsigned int hooknum,
 			   const struct xt_target *target,
-			   const void *targinfo,
-			   void *userinfo)
+			   const void *targinfo)
 {
 	const struct xt_DSCP_info *dinfo = targinfo;
 	u_int8_t dscp = ipv4_get_dsfield((*pskb)->nh.iph) >> XT_DSCP_SHIFT;
@@ -54,8 +53,7 @@ static unsigned int target6(struct sk_buff **pskb,
 			    const struct net_device *out,
 			    unsigned int hooknum,
 			    const struct xt_target *target,
-			    const void *targinfo,
-			    void *userinfo)
+			    const void *targinfo)
 {
 	const struct xt_DSCP_info *dinfo = targinfo;
 	u_int8_t dscp = ipv6_get_dsfield((*pskb)->nh.ipv6h) >> XT_DSCP_SHIFT;
diff --git a/net/netfilter/xt_MARK.c b/net/netfilter/xt_MARK.c
index 0a6127219467..95a171c87994 100644
--- a/net/netfilter/xt_MARK.c
+++ b/net/netfilter/xt_MARK.c
@@ -27,8 +27,7 @@ target_v0(struct sk_buff **pskb,
 	  const struct net_device *out,
 	  unsigned int hooknum,
 	  const struct xt_target *target,
-	  const void *targinfo,
-	  void *userinfo)
+	  const void *targinfo)
 {
 	const struct xt_mark_target_info *markinfo = targinfo;
 
@@ -44,8 +43,7 @@ target_v1(struct sk_buff **pskb,
 	  const struct net_device *out,
 	  unsigned int hooknum,
 	  const struct xt_target *target,
-	  const void *targinfo,
-	  void *userinfo)
+	  const void *targinfo)
 {
 	const struct xt_mark_target_info_v1 *markinfo = targinfo;
 	int mark = 0;
diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c
index 7b982283abdb..db9b896e57c8 100644
--- a/net/netfilter/xt_NFQUEUE.c
+++ b/net/netfilter/xt_NFQUEUE.c
@@ -29,8 +29,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	const struct xt_NFQ_info *tinfo = targinfo;
 
diff --git a/net/netfilter/xt_NOTRACK.c b/net/netfilter/xt_NOTRACK.c
index cab881d4424c..6d00dcaed238 100644
--- a/net/netfilter/xt_NOTRACK.c
+++ b/net/netfilter/xt_NOTRACK.c
@@ -16,8 +16,7 @@ target(struct sk_buff **pskb,
        const struct net_device *out,
        unsigned int hooknum,
        const struct xt_target *target,
-       const void *targinfo,
-       void *userinfo)
+       const void *targinfo)
 {
 	/* Previously seen (loopback)? Ignore. */
 	if ((*pskb)->nfct != NULL)
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index 4300988786c9..8a04dcf2611e 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -31,7 +31,7 @@ static u8 mode;
 static unsigned int target(struct sk_buff **pskb, const struct net_device *in,
 			   const struct net_device *out, unsigned int hooknum,
 			   const struct xt_target *target,
-			   const void *targinfo, void *userinfo)
+			   const void *targinfo)
 {
 	u32 secmark = 0;
 	const struct xt_secmark_target_info *info = targinfo;
diff --git a/net/netfilter/xt_connbytes.c b/net/netfilter/xt_connbytes.c
index 2d49948d3c38..d725e8b84503 100644
--- a/net/netfilter/xt_connbytes.c
+++ b/net/netfilter/xt_connbytes.c
@@ -143,7 +143,7 @@ static int check(const char *tablename,
 	return 1;
 }
 
-static struct xt_match xt_connbytes_match = {
+static struct xt_match xt_connbytes_match[] = {
 	{
 		.name		= "connbytes",
 		.family		= AF_INET,
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index 224c078a398e..45a3143b8629 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -222,7 +222,7 @@ static int tcf_ipt(struct sk_buff *skb, struct tc_action *a,
 	ret = ipt->tcfi_t->u.kernel.target->target(&skb, skb->dev, NULL,
 						   ipt->tcfi_hook,
 						   ipt->tcfi_t->u.kernel.target,
-						   ipt->tcfi_t->data, NULL);
+						   ipt->tcfi_t->data);
 	switch (ret) {
 	case NF_ACCEPT:
 		result = TC_ACT_OK;
-- 
cgit v1.2.3


From efa741656e9ebf5fd6e0432b0d1b3c7f156392d3 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Aug 2006 00:36:37 -0700
Subject: [NETFILTER]: x_tables: remove unused size argument to check/destroy
 functions

The size is verified by x_tables and isn't needed by the modules anymore.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h   |  8 ++------
 net/ipv4/netfilter/arp_tables.c      |  5 +----
 net/ipv4/netfilter/arpt_mangle.c     |  2 +-
 net/ipv4/netfilter/ip_nat_rule.c     |  2 --
 net/ipv4/netfilter/ip_tables.c       | 14 +++-----------
 net/ipv4/netfilter/ipt_CLUSTERIP.c   |  4 +---
 net/ipv4/netfilter/ipt_ECN.c         |  1 -
 net/ipv4/netfilter/ipt_LOG.c         |  1 -
 net/ipv4/netfilter/ipt_MASQUERADE.c  |  1 -
 net/ipv4/netfilter/ipt_NETMAP.c      |  1 -
 net/ipv4/netfilter/ipt_REDIRECT.c    |  1 -
 net/ipv4/netfilter/ipt_REJECT.c      |  1 -
 net/ipv4/netfilter/ipt_SAME.c        |  4 +---
 net/ipv4/netfilter/ipt_TCPMSS.c      |  1 -
 net/ipv4/netfilter/ipt_TOS.c         |  1 -
 net/ipv4/netfilter/ipt_TTL.c         |  1 -
 net/ipv4/netfilter/ipt_ULOG.c        |  1 -
 net/ipv4/netfilter/ipt_ah.c          |  1 -
 net/ipv4/netfilter/ipt_ecn.c         |  3 +--
 net/ipv4/netfilter/ipt_hashlimit.c   |  4 +---
 net/ipv4/netfilter/ipt_owner.c       |  1 -
 net/ipv4/netfilter/ipt_recent.c      |  5 ++---
 net/ipv6/netfilter/ip6_tables.c      | 10 ++--------
 net/ipv6/netfilter/ip6t_HL.c         |  1 -
 net/ipv6/netfilter/ip6t_LOG.c        |  1 -
 net/ipv6/netfilter/ip6t_REJECT.c     |  1 -
 net/ipv6/netfilter/ip6t_ah.c         |  1 -
 net/ipv6/netfilter/ip6t_dst.c        |  1 -
 net/ipv6/netfilter/ip6t_frag.c       |  1 -
 net/ipv6/netfilter/ip6t_hbh.c        |  1 -
 net/ipv6/netfilter/ip6t_ipv6header.c |  1 -
 net/ipv6/netfilter/ip6t_owner.c      |  1 -
 net/ipv6/netfilter/ip6t_rt.c         |  1 -
 net/netfilter/xt_CONNMARK.c          |  1 -
 net/netfilter/xt_CONNSECMARK.c       |  2 +-
 net/netfilter/xt_DSCP.c              |  1 -
 net/netfilter/xt_MARK.c              |  2 --
 net/netfilter/xt_SECMARK.c           |  2 +-
 net/netfilter/xt_connbytes.c         |  1 -
 net/netfilter/xt_connmark.c          |  3 +--
 net/netfilter/xt_conntrack.c         |  3 +--
 net/netfilter/xt_dccp.c              |  1 -
 net/netfilter/xt_dscp.c              |  1 -
 net/netfilter/xt_esp.c               |  1 -
 net/netfilter/xt_helper.c            |  3 +--
 net/netfilter/xt_limit.c             |  1 -
 net/netfilter/xt_mark.c              |  1 -
 net/netfilter/xt_multiport.c         |  4 ----
 net/netfilter/xt_physdev.c           |  1 -
 net/netfilter/xt_policy.c            |  3 +--
 net/netfilter/xt_quota.c             |  2 +-
 net/netfilter/xt_sctp.c              |  1 -
 net/netfilter/xt_state.c             |  3 +--
 net/netfilter/xt_statistic.c         |  2 +-
 net/netfilter/xt_string.c            |  4 +---
 net/netfilter/xt_tcpudp.c            |  2 --
 net/sched/act_ipt.c                  |  4 +---
 57 files changed, 26 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 9cef0e91542b..9d97102a9347 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -174,12 +174,10 @@ struct xt_match
 			  const void *ip,
 			  const struct xt_match *match,
 			  void *matchinfo,
-			  unsigned int matchinfosize,
 			  unsigned int hook_mask);
 
 	/* Called when entry of this type deleted. */
-	void (*destroy)(const struct xt_match *match, void *matchinfo,
-			unsigned int matchinfosize);
+	void (*destroy)(const struct xt_match *match, void *matchinfo);
 
 	/* Called when userspace align differs from kernel space one */
 	int (*compat)(void *match, void **dstptr, int *size, int convert);
@@ -221,12 +219,10 @@ struct xt_target
 			  const void *entry,
 			  const struct xt_target *target,
 			  void *targinfo,
-			  unsigned int targinfosize,
 			  unsigned int hook_mask);
 
 	/* Called when entry of this type deleted. */
-	void (*destroy)(const struct xt_target *target, void *targinfo,
-			unsigned int targinfosize);
+	void (*destroy)(const struct xt_target *target, void *targinfo);
 
 	/* Called when userspace align differs from kernel space one */
 	int (*compat)(void *target, void **dstptr, int *size, int convert);
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index c6bd270bf46a..4f10b06413a1 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -491,8 +491,6 @@ static inline int check_entry(struct arpt_entry *e, const char *name, unsigned i
 		}
 	} else if (t->u.kernel.target->checkentry
 		   && !t->u.kernel.target->checkentry(name, e, target, t->data,
-						      t->u.target_size
-						      - sizeof(*t),
 						      e->comefrom)) {
 		duprintf("arp_tables: check failed for `%s'.\n",
 			 t->u.kernel.target->name);
@@ -559,8 +557,7 @@ static inline int cleanup_entry(struct arpt_entry *e, unsigned int *i)
 
 	t = arpt_get_target(e);
 	if (t->u.kernel.target->destroy)
-		t->u.kernel.target->destroy(t->u.kernel.target, t->data,
-					    t->u.target_size - sizeof(*t));
+		t->u.kernel.target->destroy(t->u.kernel.target, t->data);
 	module_put(t->u.kernel.target->me);
 	return 0;
 }
diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c
index 05fb2421bb26..d12b1df252a1 100644
--- a/net/ipv4/netfilter/arpt_mangle.c
+++ b/net/ipv4/netfilter/arpt_mangle.c
@@ -67,7 +67,7 @@ target(struct sk_buff **pskb,
 
 static int
 checkentry(const char *tablename, const void *e, const struct xt_target *target,
-           void *targinfo, unsigned int targinfosize, unsigned int hook_mask)
+           void *targinfo, unsigned int hook_mask)
 {
 	const struct arpt_mangle *mangle = targinfo;
 
diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c
index 1aa0e4f462a5..e59f5a8ecb6b 100644
--- a/net/ipv4/netfilter/ip_nat_rule.c
+++ b/net/ipv4/netfilter/ip_nat_rule.c
@@ -172,7 +172,6 @@ static int ipt_snat_checkentry(const char *tablename,
 			       const void *entry,
 			       const struct ipt_target *target,
 			       void *targinfo,
-			       unsigned int targinfosize,
 			       unsigned int hook_mask)
 {
 	struct ip_nat_multi_range_compat *mr = targinfo;
@@ -189,7 +188,6 @@ static int ipt_dnat_checkentry(const char *tablename,
 			       const void *entry,
 			       const struct ipt_target *target,
 			       void *targinfo,
-			       unsigned int targinfosize,
 			       unsigned int hook_mask)
 {
 	struct ip_nat_multi_range_compat *mr = targinfo;
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 8ce5b6f76447..a0f36806998c 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -464,8 +464,7 @@ cleanup_match(struct ipt_entry_match *m, unsigned int *i)
 		return 1;
 
 	if (m->u.kernel.match->destroy)
-		m->u.kernel.match->destroy(m->u.kernel.match, m->data,
-					   m->u.match_size - sizeof(*m));
+		m->u.kernel.match->destroy(m->u.kernel.match, m->data);
 	module_put(m->u.kernel.match->me);
 	return 0;
 }
@@ -518,7 +517,6 @@ check_match(struct ipt_entry_match *m,
 
 	if (m->u.kernel.match->checkentry
 	    && !m->u.kernel.match->checkentry(name, ip, match, m->data,
-					      m->u.match_size - sizeof(*m),
 					      hookmask)) {
 		duprintf("ip_tables: check failed for `%s'.\n",
 			 m->u.kernel.match->name);
@@ -579,8 +577,6 @@ check_entry(struct ipt_entry *e, const char *name, unsigned int size,
 		}
 	} else if (t->u.kernel.target->checkentry
 		   && !t->u.kernel.target->checkentry(name, e, target, t->data,
-						      t->u.target_size
-						      - sizeof(*t),
 						      e->comefrom)) {
 		duprintf("ip_tables: check failed for `%s'.\n",
 			 t->u.kernel.target->name);
@@ -652,8 +648,7 @@ cleanup_entry(struct ipt_entry *e, unsigned int *i)
 	IPT_MATCH_ITERATE(e, cleanup_match, NULL);
 	t = ipt_get_target(e);
 	if (t->u.kernel.target->destroy)
-		t->u.kernel.target->destroy(t->u.kernel.target, t->data,
-					    t->u.target_size - sizeof(*t));
+		t->u.kernel.target->destroy(t->u.kernel.target, t->data);
 	module_put(t->u.kernel.target->me);
 	return 0;
 }
@@ -1599,7 +1594,6 @@ static inline int compat_copy_match_from_user(struct ipt_entry_match *m,
 
 	if (m->u.kernel.match->checkentry
 	    && !m->u.kernel.match->checkentry(name, ip, match, dm->data,
-					      dm->u.match_size - sizeof(*dm),
 					      hookmask)) {
 		duprintf("ip_tables: check failed for `%s'.\n",
 			 m->u.kernel.match->name);
@@ -1658,8 +1652,7 @@ static int compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
 			goto out;
 	} else if (t->u.kernel.target->checkentry
 		   && !t->u.kernel.target->checkentry(name, de, target,
-				t->data, t->u.target_size - sizeof(*t),
-				de->comefrom)) {
+						      t->data, de->comefrom)) {
 		duprintf("ip_tables: compat: check failed for `%s'.\n",
 			 t->u.kernel.target->name);
 		goto out;
@@ -2182,7 +2175,6 @@ icmp_checkentry(const char *tablename,
 	   const void *info,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	const struct ipt_icmp *icmpinfo = matchinfo;
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index a08383cf9e7a..41589665fc5d 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -372,7 +372,6 @@ checkentry(const char *tablename,
 	   const void *e_void,
 	   const struct xt_target *target,
            void *targinfo,
-           unsigned int targinfosize,
            unsigned int hook_mask)
 {
 	struct ipt_clusterip_tgt_info *cipinfo = targinfo;
@@ -449,8 +448,7 @@ checkentry(const char *tablename,
 }
 
 /* drop reference count of cluster config when rule is deleted */
-static void destroy(const struct xt_target *target, void *targinfo,
-		    unsigned int targinfosize)
+static void destroy(const struct xt_target *target, void *targinfo)
 {
 	struct ipt_clusterip_tgt_info *cipinfo = targinfo;
 
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index 1c3da4a48e5f..23f9c7ebe7eb 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -106,7 +106,6 @@ checkentry(const char *tablename,
 	   const void *e_void,
 	   const struct xt_target *target,
            void *targinfo,
-           unsigned int targinfosize,
            unsigned int hook_mask)
 {
 	const struct ipt_ECN_info *einfo = (struct ipt_ECN_info *)targinfo;
diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c
index a8d356c6191f..7dc820df8bc5 100644
--- a/net/ipv4/netfilter/ipt_LOG.c
+++ b/net/ipv4/netfilter/ipt_LOG.c
@@ -439,7 +439,6 @@ static int ipt_log_checkentry(const char *tablename,
 			      const void *e,
 			      const struct xt_target *target,
 			      void *targinfo,
-			      unsigned int targinfosize,
 			      unsigned int hook_mask)
 {
 	const struct ipt_log_info *loginfo = targinfo;
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index 9659793c66c0..bc65168a3437 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -42,7 +42,6 @@ masquerade_check(const char *tablename,
 		 const void *e,
 		 const struct xt_target *target,
 		 void *targinfo,
-		 unsigned int targinfosize,
 		 unsigned int hook_mask)
 {
 	const struct ip_nat_multi_range_compat *mr = targinfo;
diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c
index fd5e74a19fb5..beb2914225ff 100644
--- a/net/ipv4/netfilter/ipt_NETMAP.c
+++ b/net/ipv4/netfilter/ipt_NETMAP.c
@@ -33,7 +33,6 @@ check(const char *tablename,
       const void *e,
       const struct xt_target *target,
       void *targinfo,
-      unsigned int targinfosize,
       unsigned int hook_mask)
 {
 	const struct ip_nat_multi_range_compat *mr = targinfo;
diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c
index 839fe99f71d4..f03d43671c6d 100644
--- a/net/ipv4/netfilter/ipt_REDIRECT.c
+++ b/net/ipv4/netfilter/ipt_REDIRECT.c
@@ -36,7 +36,6 @@ redirect_check(const char *tablename,
 	       const void *e,
 	       const struct xt_target *target,
 	       void *targinfo,
-	       unsigned int targinfosize,
 	       unsigned int hook_mask)
 {
 	const struct ip_nat_multi_range_compat *mr = targinfo;
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index 1dfd8e56be8b..b81821edd893 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -276,7 +276,6 @@ static int check(const char *tablename,
 		 const void *e_void,
 		 const struct xt_target *target,
 		 void *targinfo,
-		 unsigned int targinfosize,
 		 unsigned int hook_mask)
 {
  	const struct ipt_reject_info *rejinfo = targinfo;
diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c
index cf801749490f..efbcb1198832 100644
--- a/net/ipv4/netfilter/ipt_SAME.c
+++ b/net/ipv4/netfilter/ipt_SAME.c
@@ -52,7 +52,6 @@ same_check(const char *tablename,
 	      const void *e,
 	      const struct xt_target *target,
 	      void *targinfo,
-	      unsigned int targinfosize,
 	      unsigned int hook_mask)
 {
 	unsigned int count, countess, rangeip, index = 0;
@@ -116,8 +115,7 @@ same_check(const char *tablename,
 }
 
 static void 
-same_destroy(const struct xt_target *target, void *targinfo,
-		unsigned int targinfosize)
+same_destroy(const struct xt_target *target, void *targinfo)
 {
 	struct ipt_same_info *mr = targinfo;
 
diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c
index 6d668dcfc22a..ac8a35eeea3f 100644
--- a/net/ipv4/netfilter/ipt_TCPMSS.c
+++ b/net/ipv4/netfilter/ipt_TCPMSS.c
@@ -207,7 +207,6 @@ ipt_tcpmss_checkentry(const char *tablename,
 		      const void *e_void,
 		      const struct xt_target *target,
 		      void *targinfo,
-		      unsigned int targinfosize,
 		      unsigned int hook_mask)
 {
 	const struct ipt_tcpmss_info *tcpmssinfo = targinfo;
diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c
index 043df0137084..471a4c438b0a 100644
--- a/net/ipv4/netfilter/ipt_TOS.c
+++ b/net/ipv4/netfilter/ipt_TOS.c
@@ -49,7 +49,6 @@ checkentry(const char *tablename,
 	   const void *e_void,
 	   const struct xt_target *target,
            void *targinfo,
-           unsigned int targinfosize,
            unsigned int hook_mask)
 {
 	const u_int8_t tos = ((struct ipt_tos_target_info *)targinfo)->tos;
diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c
index 164007107b5e..214d9d9c428f 100644
--- a/net/ipv4/netfilter/ipt_TTL.c
+++ b/net/ipv4/netfilter/ipt_TTL.c
@@ -67,7 +67,6 @@ static int ipt_ttl_checkentry(const char *tablename,
 		const void *e,
 		const struct xt_target *target,
 		void *targinfo,
-		unsigned int targinfosize,
 		unsigned int hook_mask)
 {
 	struct ipt_TTL_info *info = targinfo;
diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c
index 4c5f0a117862..2b104ea54f48 100644
--- a/net/ipv4/netfilter/ipt_ULOG.c
+++ b/net/ipv4/netfilter/ipt_ULOG.c
@@ -346,7 +346,6 @@ static int ipt_ulog_checkentry(const char *tablename,
 			       const void *e,
 			       const struct xt_target *target,
 			       void *targinfo,
-			       unsigned int targinfosize,
 			       unsigned int hookmask)
 {
 	struct ipt_ulog_info *loginfo = (struct ipt_ulog_info *) targinfo;
diff --git a/net/ipv4/netfilter/ipt_ah.c b/net/ipv4/netfilter/ipt_ah.c
index 2927135873d7..1798f86bc534 100644
--- a/net/ipv4/netfilter/ipt_ah.c
+++ b/net/ipv4/netfilter/ipt_ah.c
@@ -74,7 +74,6 @@ checkentry(const char *tablename,
 	   const void *ip_void,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchinfosize,
 	   unsigned int hook_mask)
 {
 	const struct ipt_ah *ahinfo = matchinfo;
diff --git a/net/ipv4/netfilter/ipt_ecn.c b/net/ipv4/netfilter/ipt_ecn.c
index b28250414933..dafbdec0efc0 100644
--- a/net/ipv4/netfilter/ipt_ecn.c
+++ b/net/ipv4/netfilter/ipt_ecn.c
@@ -88,8 +88,7 @@ static int match(const struct sk_buff *skb,
 
 static int checkentry(const char *tablename, const void *ip_void,
 		      const struct xt_match *match,
-		      void *matchinfo, unsigned int matchsize,
-		      unsigned int hook_mask)
+		      void *matchinfo, unsigned int hook_mask)
 {
 	const struct ipt_ecn_info *info = matchinfo;
 	const struct ipt_ip *ip = ip_void;
diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c
index 3bd2368e1fc9..b5b74b07370c 100644
--- a/net/ipv4/netfilter/ipt_hashlimit.c
+++ b/net/ipv4/netfilter/ipt_hashlimit.c
@@ -478,7 +478,6 @@ hashlimit_checkentry(const char *tablename,
 		     const void *inf,
 		     const struct xt_match *match,
 		     void *matchinfo,
-		     unsigned int matchsize,
 		     unsigned int hook_mask)
 {
 	struct ipt_hashlimit_info *r = matchinfo;
@@ -529,8 +528,7 @@ hashlimit_checkentry(const char *tablename,
 }
 
 static void
-hashlimit_destroy(const struct xt_match *match, void *matchinfo,
-		  unsigned int matchsize)
+hashlimit_destroy(const struct xt_match *match, void *matchinfo)
 {
 	struct ipt_hashlimit_info *r = matchinfo;
 
diff --git a/net/ipv4/netfilter/ipt_owner.c b/net/ipv4/netfilter/ipt_owner.c
index 5ac6ac023b5e..78c336f12a9e 100644
--- a/net/ipv4/netfilter/ipt_owner.c
+++ b/net/ipv4/netfilter/ipt_owner.c
@@ -56,7 +56,6 @@ checkentry(const char *tablename,
            const void *ip,
 	   const struct xt_match *match,
            void *matchinfo,
-           unsigned int matchsize,
            unsigned int hook_mask)
 {
 	const struct ipt_owner_info *info = matchinfo;
diff --git a/net/ipv4/netfilter/ipt_recent.c b/net/ipv4/netfilter/ipt_recent.c
index 682c0946201e..32ae8d7ac506 100644
--- a/net/ipv4/netfilter/ipt_recent.c
+++ b/net/ipv4/netfilter/ipt_recent.c
@@ -238,7 +238,7 @@ out:
 static int
 ipt_recent_checkentry(const char *tablename, const void *ip,
 		      const struct xt_match *match, void *matchinfo,
-		      unsigned int matchsize, unsigned int hook_mask)
+		      unsigned int hook_mask)
 {
 	const struct ipt_recent_info *info = matchinfo;
 	struct recent_table *t;
@@ -294,8 +294,7 @@ out:
 }
 
 static void
-ipt_recent_destroy(const struct xt_match *match, void *matchinfo,
-		   unsigned int matchsize)
+ipt_recent_destroy(const struct xt_match *match, void *matchinfo)
 {
 	const struct ipt_recent_info *info = matchinfo;
 	struct recent_table *t;
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 38cd7ffda9a0..d1c315364ee7 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -504,8 +504,7 @@ cleanup_match(struct ip6t_entry_match *m, unsigned int *i)
 		return 1;
 
 	if (m->u.kernel.match->destroy)
-		m->u.kernel.match->destroy(m->u.kernel.match, m->data,
-					   m->u.match_size - sizeof(*m));
+		m->u.kernel.match->destroy(m->u.kernel.match, m->data);
 	module_put(m->u.kernel.match->me);
 	return 0;
 }
@@ -558,7 +557,6 @@ check_match(struct ip6t_entry_match *m,
 
 	if (m->u.kernel.match->checkentry
 	    && !m->u.kernel.match->checkentry(name, ipv6, match,  m->data,
-					      m->u.match_size - sizeof(*m),
 					      hookmask)) {
 		duprintf("ip_tables: check failed for `%s'.\n",
 			 m->u.kernel.match->name);
@@ -619,8 +617,6 @@ check_entry(struct ip6t_entry *e, const char *name, unsigned int size,
 		}
 	} else if (t->u.kernel.target->checkentry
 		   && !t->u.kernel.target->checkentry(name, e, target, t->data,
-						      t->u.target_size
-						      - sizeof(*t),
 						      e->comefrom)) {
 		duprintf("ip_tables: check failed for `%s'.\n",
 			 t->u.kernel.target->name);
@@ -692,8 +688,7 @@ cleanup_entry(struct ip6t_entry *e, unsigned int *i)
 	IP6T_MATCH_ITERATE(e, cleanup_match, NULL);
 	t = ip6t_get_target(e);
 	if (t->u.kernel.target->destroy)
-		t->u.kernel.target->destroy(t->u.kernel.target, t->data,
-					    t->u.target_size - sizeof(*t));
+		t->u.kernel.target->destroy(t->u.kernel.target, t->data);
 	module_put(t->u.kernel.target->me);
 	return 0;
 }
@@ -1349,7 +1344,6 @@ icmp6_checkentry(const char *tablename,
 	   const void *entry,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	const struct ip6t_icmp *icmpinfo = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_HL.c b/net/ipv6/netfilter/ip6t_HL.c
index c85d124f9a3d..e54ea92d107b 100644
--- a/net/ipv6/netfilter/ip6t_HL.c
+++ b/net/ipv6/netfilter/ip6t_HL.c
@@ -66,7 +66,6 @@ static int ip6t_hl_checkentry(const char *tablename,
 		const void *entry,
 		const struct xt_target *target,
 		void *targinfo,
-		unsigned int targinfosize,
 		unsigned int hook_mask)
 {
 	struct ip6t_HL_info *info = targinfo;
diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c
index acb91733e1fd..0cf537d30185 100644
--- a/net/ipv6/netfilter/ip6t_LOG.c
+++ b/net/ipv6/netfilter/ip6t_LOG.c
@@ -451,7 +451,6 @@ static int ip6t_log_checkentry(const char *tablename,
 			       const void *entry,
 			       const struct xt_target *target,
 			       void *targinfo,
-			       unsigned int targinfosize,
 			       unsigned int hook_mask)
 {
 	const struct ip6t_log_info *loginfo = targinfo;
diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index 343acd3cbf5e..311eae82feb3 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -223,7 +223,6 @@ static int check(const char *tablename,
 		 const void *entry,
 		 const struct xt_target *target,
 		 void *targinfo,
-		 unsigned int targinfosize,
 		 unsigned int hook_mask)
 {
  	const struct ip6t_reject_info *rejinfo = targinfo;
diff --git a/net/ipv6/netfilter/ip6t_ah.c b/net/ipv6/netfilter/ip6t_ah.c
index 2f7bb20c758b..ec1b1608156c 100644
--- a/net/ipv6/netfilter/ip6t_ah.c
+++ b/net/ipv6/netfilter/ip6t_ah.c
@@ -102,7 +102,6 @@ checkentry(const char *tablename,
           const void *entry,
 	  const struct xt_match *match,
           void *matchinfo,
-          unsigned int matchinfosize,
           unsigned int hook_mask)
 {
 	const struct ip6t_ah *ahinfo = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_dst.c b/net/ipv6/netfilter/ip6t_dst.c
index 9422413d0571..223c335467cc 100644
--- a/net/ipv6/netfilter/ip6t_dst.c
+++ b/net/ipv6/netfilter/ip6t_dst.c
@@ -182,7 +182,6 @@ checkentry(const char *tablename,
 	   const void *info,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchinfosize,
 	   unsigned int hook_mask)
 {
 	const struct ip6t_opts *optsinfo = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_frag.c b/net/ipv6/netfilter/ip6t_frag.c
index 06768c84bd31..78d9c8b9e28a 100644
--- a/net/ipv6/netfilter/ip6t_frag.c
+++ b/net/ipv6/netfilter/ip6t_frag.c
@@ -119,7 +119,6 @@ checkentry(const char *tablename,
 	   const void *ip,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchinfosize,
 	   unsigned int hook_mask)
 {
 	const struct ip6t_frag *fraginfo = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_hbh.c b/net/ipv6/netfilter/ip6t_hbh.c
index 374f1be85c0d..72defc816563 100644
--- a/net/ipv6/netfilter/ip6t_hbh.c
+++ b/net/ipv6/netfilter/ip6t_hbh.c
@@ -182,7 +182,6 @@ checkentry(const char *tablename,
 	   const void *entry,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchinfosize,
 	   unsigned int hook_mask)
 {
 	const struct ip6t_opts *optsinfo = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_ipv6header.c b/net/ipv6/netfilter/ip6t_ipv6header.c
index 9375eeb1369f..3093c398002f 100644
--- a/net/ipv6/netfilter/ip6t_ipv6header.c
+++ b/net/ipv6/netfilter/ip6t_ipv6header.c
@@ -128,7 +128,6 @@ ipv6header_checkentry(const char *tablename,
 		      const void *ip,
 		      const struct xt_match *match,
 		      void *matchinfo,
-		      unsigned int matchsize,
 		      unsigned int hook_mask)
 {
 	const struct ip6t_ipv6header_info *info = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_owner.c b/net/ipv6/netfilter/ip6t_owner.c
index 5d047990cd44..4eb9bbc4ebc3 100644
--- a/net/ipv6/netfilter/ip6t_owner.c
+++ b/net/ipv6/netfilter/ip6t_owner.c
@@ -57,7 +57,6 @@ checkentry(const char *tablename,
 	   const void *ip,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	const struct ip6t_owner_info *info = matchinfo;
diff --git a/net/ipv6/netfilter/ip6t_rt.c b/net/ipv6/netfilter/ip6t_rt.c
index fbb0184a41d8..bcb2e168a5bc 100644
--- a/net/ipv6/netfilter/ip6t_rt.c
+++ b/net/ipv6/netfilter/ip6t_rt.c
@@ -197,7 +197,6 @@ checkentry(const char *tablename,
 	   const void *entry,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchinfosize,
 	   unsigned int hook_mask)
 {
 	const struct ip6t_rt *rtinfo = matchinfo;
diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c
index c2125f6ee128..0e4249ddc17b 100644
--- a/net/netfilter/xt_CONNMARK.c
+++ b/net/netfilter/xt_CONNMARK.c
@@ -89,7 +89,6 @@ checkentry(const char *tablename,
 	   const void *entry,
 	   const struct xt_target *target,
 	   void *targinfo,
-	   unsigned int targinfosize,
 	   unsigned int hook_mask)
 {
 	struct xt_connmark_target_info *matchinfo = targinfo;
diff --git a/net/netfilter/xt_CONNSECMARK.c b/net/netfilter/xt_CONNSECMARK.c
index 4b9cc65bb82b..4b0e14bb1726 100644
--- a/net/netfilter/xt_CONNSECMARK.c
+++ b/net/netfilter/xt_CONNSECMARK.c
@@ -89,7 +89,7 @@ static unsigned int target(struct sk_buff **pskb, const struct net_device *in,
 
 static int checkentry(const char *tablename, const void *entry,
 		      const struct xt_target *target, void *targinfo,
-		      unsigned int targinfosize, unsigned int hook_mask)
+		      unsigned int hook_mask)
 {
 	struct xt_connsecmark_target_info *info = targinfo;
 
diff --git a/net/netfilter/xt_DSCP.c b/net/netfilter/xt_DSCP.c
index 9d23c9580d80..a7cc75aeb38d 100644
--- a/net/netfilter/xt_DSCP.c
+++ b/net/netfilter/xt_DSCP.c
@@ -72,7 +72,6 @@ static int checkentry(const char *tablename,
 		      const void *e_void,
 		      const struct xt_target *target,
 		      void *targinfo,
-		      unsigned int targinfosize,
 		      unsigned int hook_mask)
 {
 	const u_int8_t dscp = ((struct xt_DSCP_info *)targinfo)->dscp;
diff --git a/net/netfilter/xt_MARK.c b/net/netfilter/xt_MARK.c
index 95a171c87994..782f8d8c3edf 100644
--- a/net/netfilter/xt_MARK.c
+++ b/net/netfilter/xt_MARK.c
@@ -74,7 +74,6 @@ checkentry_v0(const char *tablename,
 	      const void *entry,
 	      const struct xt_target *target,
 	      void *targinfo,
-	      unsigned int targinfosize,
 	      unsigned int hook_mask)
 {
 	struct xt_mark_target_info *markinfo = targinfo;
@@ -91,7 +90,6 @@ checkentry_v1(const char *tablename,
 	      const void *entry,
 	      const struct xt_target *target,
 	      void *targinfo,
-	      unsigned int targinfosize,
 	      unsigned int hook_mask)
 {
 	struct xt_mark_target_info_v1 *markinfo = targinfo;
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index 8a04dcf2611e..451b67c4bb53 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -85,7 +85,7 @@ static int checkentry_selinux(struct xt_secmark_target_info *info)
 
 static int checkentry(const char *tablename, const void *entry,
 		      const struct xt_target *target, void *targinfo,
-		      unsigned int targinfosize, unsigned int hook_mask)
+		      unsigned int hook_mask)
 {
 	struct xt_secmark_target_info *info = targinfo;
 
diff --git a/net/netfilter/xt_connbytes.c b/net/netfilter/xt_connbytes.c
index d725e8b84503..dcc497ea8183 100644
--- a/net/netfilter/xt_connbytes.c
+++ b/net/netfilter/xt_connbytes.c
@@ -125,7 +125,6 @@ static int check(const char *tablename,
 		 const void *ip,
 		 const struct xt_match *match,
 		 void *matchinfo,
-		 unsigned int matchsize,
 		 unsigned int hook_mask)
 {
 	const struct xt_connbytes_info *sinfo = matchinfo;
diff --git a/net/netfilter/xt_connmark.c b/net/netfilter/xt_connmark.c
index a97b2d455b79..c9104d05a19c 100644
--- a/net/netfilter/xt_connmark.c
+++ b/net/netfilter/xt_connmark.c
@@ -55,7 +55,6 @@ checkentry(const char *tablename,
 	   const void *ip,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	struct xt_connmark_info *cm = matchinfo;
@@ -75,7 +74,7 @@ checkentry(const char *tablename,
 }
 
 static void
-destroy(const struct xt_match *match, void *matchinfo, unsigned int matchsize)
+destroy(const struct xt_match *match, void *matchinfo)
 {
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
 	nf_ct_l3proto_module_put(match->family);
diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c
index 1540885174ee..39c57e9f7563 100644
--- a/net/netfilter/xt_conntrack.c
+++ b/net/netfilter/xt_conntrack.c
@@ -208,7 +208,6 @@ checkentry(const char *tablename,
 	   const void *ip,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
@@ -222,7 +221,7 @@ checkentry(const char *tablename,
 }
 
 static void
-destroy(const struct xt_match *match, void *matchinfo, unsigned int matchsize)
+destroy(const struct xt_match *match, void *matchinfo)
 {
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
 	nf_ct_l3proto_module_put(match->family);
diff --git a/net/netfilter/xt_dccp.c b/net/netfilter/xt_dccp.c
index 5ca6f5288f46..3e6cf430e518 100644
--- a/net/netfilter/xt_dccp.c
+++ b/net/netfilter/xt_dccp.c
@@ -131,7 +131,6 @@ checkentry(const char *tablename,
 	   const void *inf,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	const struct xt_dccp_info *info = matchinfo;
diff --git a/net/netfilter/xt_dscp.c b/net/netfilter/xt_dscp.c
index d84075c30159..26c7f4ad102a 100644
--- a/net/netfilter/xt_dscp.c
+++ b/net/netfilter/xt_dscp.c
@@ -58,7 +58,6 @@ static int checkentry(const char *tablename,
 		      const void *info,
 		      const struct xt_match *match,
 		      void *matchinfo,
-		      unsigned int matchsize,
 		      unsigned int hook_mask)
 {
 	const u_int8_t dscp = ((struct xt_dscp_info *)matchinfo)->dscp;
diff --git a/net/netfilter/xt_esp.c b/net/netfilter/xt_esp.c
index 7b19bc9ea205..7c95f149d942 100644
--- a/net/netfilter/xt_esp.c
+++ b/net/netfilter/xt_esp.c
@@ -79,7 +79,6 @@ checkentry(const char *tablename,
 	   const void *ip_void,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchinfosize,
 	   unsigned int hook_mask)
 {
 	const struct xt_esp *espinfo = matchinfo;
diff --git a/net/netfilter/xt_helper.c b/net/netfilter/xt_helper.c
index db453a7a154e..5d7818b73e3a 100644
--- a/net/netfilter/xt_helper.c
+++ b/net/netfilter/xt_helper.c
@@ -139,7 +139,6 @@ static int check(const char *tablename,
 		 const void *inf,
 		 const struct xt_match *match,
 		 void *matchinfo,
-		 unsigned int matchsize,
 		 unsigned int hook_mask)
 {
 	struct xt_helper_info *info = matchinfo;
@@ -156,7 +155,7 @@ static int check(const char *tablename,
 }
 
 static void
-destroy(const struct xt_match *match, void *matchinfo, unsigned int matchsize)
+destroy(const struct xt_match *match, void *matchinfo)
 {
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
 	nf_ct_l3proto_module_put(match->family);
diff --git a/net/netfilter/xt_limit.c b/net/netfilter/xt_limit.c
index e8d5e7ac695a..b9c9ff3a06ea 100644
--- a/net/netfilter/xt_limit.c
+++ b/net/netfilter/xt_limit.c
@@ -110,7 +110,6 @@ ipt_limit_checkentry(const char *tablename,
 		     const void *inf,
 		     const struct xt_match *match,
 		     void *matchinfo,
-		     unsigned int matchsize,
 		     unsigned int hook_mask)
 {
 	struct xt_rateinfo *r = matchinfo;
diff --git a/net/netfilter/xt_mark.c b/net/netfilter/xt_mark.c
index 39f9b079f5d4..e8059cd17275 100644
--- a/net/netfilter/xt_mark.c
+++ b/net/netfilter/xt_mark.c
@@ -39,7 +39,6 @@ checkentry(const char *tablename,
            const void *entry,
 	   const struct xt_match *match,
            void *matchinfo,
-           unsigned int matchsize,
            unsigned int hook_mask)
 {
 	const struct xt_mark_info *minfo = matchinfo;
diff --git a/net/netfilter/xt_multiport.c b/net/netfilter/xt_multiport.c
index e74f9bb98b3c..d3aefd380930 100644
--- a/net/netfilter/xt_multiport.c
+++ b/net/netfilter/xt_multiport.c
@@ -176,7 +176,6 @@ checkentry(const char *tablename,
 	   const void *info,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	const struct ipt_ip *ip = info;
@@ -191,7 +190,6 @@ checkentry_v1(const char *tablename,
 	      const void *info,
 	      const struct xt_match *match,
 	      void *matchinfo,
-	      unsigned int matchsize,
 	      unsigned int hook_mask)
 {
 	const struct ipt_ip *ip = info;
@@ -206,7 +204,6 @@ checkentry6(const char *tablename,
 	    const void *info,
 	    const struct xt_match *match,
 	    void *matchinfo,
-	    unsigned int matchsize,
 	    unsigned int hook_mask)
 {
 	const struct ip6t_ip6 *ip = info;
@@ -221,7 +218,6 @@ checkentry6_v1(const char *tablename,
 	       const void *info,
 	       const struct xt_match *match,
 	       void *matchinfo,
-	       unsigned int matchsize,
 	       unsigned int hook_mask)
 {
 	const struct ip6t_ip6 *ip = info;
diff --git a/net/netfilter/xt_physdev.c b/net/netfilter/xt_physdev.c
index af3d70f96ecd..fd8f954cded5 100644
--- a/net/netfilter/xt_physdev.c
+++ b/net/netfilter/xt_physdev.c
@@ -106,7 +106,6 @@ checkentry(const char *tablename,
 		       const void *ip,
 		       const struct xt_match *match,
 		       void *matchinfo,
-		       unsigned int matchsize,
 		       unsigned int hook_mask)
 {
 	const struct xt_physdev_info *info = matchinfo;
diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c
index f5639c451112..e9d81378d653 100644
--- a/net/netfilter/xt_policy.c
+++ b/net/netfilter/xt_policy.c
@@ -135,8 +135,7 @@ static int match(const struct sk_buff *skb,
 
 static int checkentry(const char *tablename, const void *ip_void,
                       const struct xt_match *match,
-                      void *matchinfo, unsigned int matchsize,
-                      unsigned int hook_mask)
+                      void *matchinfo, unsigned int hook_mask)
 {
 	struct xt_policy_info *info = matchinfo;
 
diff --git a/net/netfilter/xt_quota.c b/net/netfilter/xt_quota.c
index cc44f87cb8e6..b75fa2c70e66 100644
--- a/net/netfilter/xt_quota.c
+++ b/net/netfilter/xt_quota.c
@@ -41,7 +41,7 @@ match(const struct sk_buff *skb,
 static int
 checkentry(const char *tablename, const void *entry,
 	   const struct xt_match *match, void *matchinfo,
-	   unsigned int matchsize, unsigned int hook_mask)
+	   unsigned int hook_mask)
 {
 	struct xt_quota_info *q = (struct xt_quota_info *)matchinfo;
 
diff --git a/net/netfilter/xt_sctp.c b/net/netfilter/xt_sctp.c
index 5628621170e6..7956acaaa24b 100644
--- a/net/netfilter/xt_sctp.c
+++ b/net/netfilter/xt_sctp.c
@@ -163,7 +163,6 @@ checkentry(const char *tablename,
 	   const void *inf,
 	   const struct xt_match *match,
 	   void *matchinfo,
-	   unsigned int matchsize,
 	   unsigned int hook_mask)
 {
 	const struct xt_sctp_info *info = matchinfo;
diff --git a/net/netfilter/xt_state.c b/net/netfilter/xt_state.c
index 5f9492e3b2b1..d9010b16a1f9 100644
--- a/net/netfilter/xt_state.c
+++ b/net/netfilter/xt_state.c
@@ -48,7 +48,6 @@ static int check(const char *tablename,
 		 const void *inf,
 		 const struct xt_match *match,
 		 void *matchinfo,
-		 unsigned int matchsize,
 		 unsigned int hook_mask)
 {
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
@@ -62,7 +61,7 @@ static int check(const char *tablename,
 }
 
 static void
-destroy(const struct xt_match *match, void *matchinfo, unsigned int matchsize)
+destroy(const struct xt_match *match, void *matchinfo)
 {
 #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
 	nf_ct_l3proto_module_put(match->family);
diff --git a/net/netfilter/xt_statistic.c b/net/netfilter/xt_statistic.c
index 5181630a87fc..091a9f89f5d5 100644
--- a/net/netfilter/xt_statistic.c
+++ b/net/netfilter/xt_statistic.c
@@ -55,7 +55,7 @@ match(const struct sk_buff *skb,
 static int
 checkentry(const char *tablename, const void *entry,
 	   const struct xt_match *match, void *matchinfo,
-	   unsigned int matchsize, unsigned int hook_mask)
+	   unsigned int hook_mask)
 {
 	struct xt_statistic_info *info = (struct xt_statistic_info *)matchinfo;
 
diff --git a/net/netfilter/xt_string.c b/net/netfilter/xt_string.c
index 1a1c1d17d85e..4453252400aa 100644
--- a/net/netfilter/xt_string.c
+++ b/net/netfilter/xt_string.c
@@ -46,7 +46,6 @@ static int checkentry(const char *tablename,
 		      const void *ip,
 		      const struct xt_match *match,
 		      void *matchinfo,
-		      unsigned int matchsize,
 		      unsigned int hook_mask)
 {
 	struct xt_string_info *conf = matchinfo;
@@ -69,8 +68,7 @@ static int checkentry(const char *tablename,
 	return 1;
 }
 
-static void destroy(const struct xt_match *match, void *matchinfo,
-		    unsigned int matchsize)
+static void destroy(const struct xt_match *match, void *matchinfo)
 {
 	textsearch_destroy(STRING_TEXT_PRIV(matchinfo)->config);
 }
diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c
index 54aab051af86..e76a68e0bc66 100644
--- a/net/netfilter/xt_tcpudp.c
+++ b/net/netfilter/xt_tcpudp.c
@@ -141,7 +141,6 @@ tcp_checkentry(const char *tablename,
 	       const void *info,
 	       const struct xt_match *match,
 	       void *matchinfo,
-	       unsigned int matchsize,
 	       unsigned int hook_mask)
 {
 	const struct xt_tcp *tcpinfo = matchinfo;
@@ -190,7 +189,6 @@ udp_checkentry(const char *tablename,
 	       const void *info,
 	       const struct xt_match *match,
 	       void *matchinfo,
-	       unsigned int matchsize,
 	       unsigned int hook_mask)
 {
 	const struct xt_tcp *udpinfo = matchinfo;
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index 45a3143b8629..d8c9310da6e5 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -69,7 +69,6 @@ static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int
 	if (t->u.kernel.target->checkentry
 	    && !t->u.kernel.target->checkentry(table, NULL,
 		    			       t->u.kernel.target, t->data,
-					       t->u.target_size - sizeof(*t),
 					       hook)) {
 		module_put(t->u.kernel.target->me);
 		ret = -EINVAL;
@@ -81,8 +80,7 @@ static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int
 static void ipt_destroy_target(struct ipt_entry_target *t)
 {
 	if (t->u.kernel.target->destroy)
-		t->u.kernel.target->destroy(t->u.kernel.target, t->data,
-		                            t->u.target_size - sizeof(*t));
+		t->u.kernel.target->destroy(t->u.kernel.target, t->data);
         module_put(t->u.kernel.target->me);
 }
 
-- 
cgit v1.2.3


From 53e26658282373b84ba85a0c9807cb762f7738a6 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Aug 2006 00:43:20 -0700
Subject: [NETFILTER]: nfnetlink: remove unnecessary packed attributes

Remove unnecessary packed attributes in nfnetlink structures. Unfortunately
in a few cases they have to stay to avoid changing structure sizes.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nfnetlink.h       | 4 ++--
 include/linux/netfilter/nfnetlink_log.h   | 6 +++---
 include/linux/netfilter/nfnetlink_queue.h | 8 ++++----
 3 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h
index 9f5b12cf489b..6d8e3e5a80e9 100644
--- a/include/linux/netfilter/nfnetlink.h
+++ b/include/linux/netfilter/nfnetlink.h
@@ -43,7 +43,7 @@ struct nfattr
 	u_int16_t nfa_len;
 	u_int16_t nfa_type;	/* we use 15 bits for the type, and the highest
 				 * bit to indicate whether the payload is nested */
-} __attribute__ ((packed));
+};
 
 /* FIXME: Apart from NFNL_NFA_NESTED shamelessly copy and pasted from
  * rtnetlink.h, it's time to put this in a generic file */
@@ -79,7 +79,7 @@ struct nfgenmsg {
 	u_int8_t  nfgen_family;		/* AF_xxx */
 	u_int8_t  version;		/* nfnetlink version */
 	u_int16_t res_id;		/* resource id */
-} __attribute__ ((packed));
+};
 
 #define NFNETLINK_V0	0
 
diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h
index a7497c7436df..87b92f8b988f 100644
--- a/include/linux/netfilter/nfnetlink_log.h
+++ b/include/linux/netfilter/nfnetlink_log.h
@@ -19,18 +19,18 @@ struct nfulnl_msg_packet_hdr {
 	u_int16_t	hw_protocol;	/* hw protocol (network order) */
 	u_int8_t	hook;		/* netfilter hook */
 	u_int8_t	_pad;
-} __attribute__ ((packed));
+};
 
 struct nfulnl_msg_packet_hw {
 	u_int16_t	hw_addrlen;
 	u_int16_t	_pad;
 	u_int8_t	hw_addr[8];
-} __attribute__ ((packed));
+};
 
 struct nfulnl_msg_packet_timestamp {
 	aligned_u64	sec;
 	aligned_u64	usec;
-} __attribute__ ((packed));
+};
 
 #define NFULNL_PREFIXLEN	30	/* just like old log target */
 
diff --git a/include/linux/netfilter/nfnetlink_queue.h b/include/linux/netfilter/nfnetlink_queue.h
index 9e774373244c..36af0360b56d 100644
--- a/include/linux/netfilter/nfnetlink_queue.h
+++ b/include/linux/netfilter/nfnetlink_queue.h
@@ -22,12 +22,12 @@ struct nfqnl_msg_packet_hw {
 	u_int16_t	hw_addrlen;
 	u_int16_t	_pad;
 	u_int8_t	hw_addr[8];
-} __attribute__ ((packed));
+};
 
 struct nfqnl_msg_packet_timestamp {
 	aligned_u64	sec;
 	aligned_u64	usec;
-} __attribute__ ((packed));
+};
 
 enum nfqnl_attr_type {
 	NFQA_UNSPEC,
@@ -49,7 +49,7 @@ enum nfqnl_attr_type {
 struct nfqnl_msg_verdict_hdr {
 	u_int32_t verdict;
 	u_int32_t id;
-} __attribute__ ((packed));
+};
 
 
 enum nfqnl_msg_config_cmds {
@@ -64,7 +64,7 @@ struct nfqnl_msg_config_cmd {
 	u_int8_t	command;	/* nfqnl_msg_config_cmds */
 	u_int8_t	_pad;
 	u_int16_t	pf;		/* AF_xxx for PF_[UN]BIND */
-} __attribute__ ((packed));
+};
 
 enum nfqnl_config_mode {
 	NFQNL_COPY_NONE,
-- 
cgit v1.2.3


From 91270cf81765152f6e77953440beb4d3b34a71b5 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 22 Aug 2006 00:43:38 -0700
Subject: [NETFILTER]: x_tables: add data member to struct xt_match

Shared match functions can use this to make runtime decisions basen on the
used match.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 9d97102a9347..03d1027fb0e8 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -185,6 +185,9 @@ struct xt_match
 	/* Set this to THIS_MODULE if you are a module, otherwise NULL */
 	struct module *me;
 
+	/* Free to use by each match */
+	unsigned long data;
+
 	char *table;
 	unsigned int matchsize;
 	unsigned int hooks;
-- 
cgit v1.2.3


From 3fd091e73b81f131e1567c4d4a1ec042940bf2f7 Mon Sep 17 00:00:00 2001
From: Vladislav Yasevich <vladislav.yasevich@hp.com>
Date: Tue, 22 Aug 2006 13:29:17 -0700
Subject: [SCTP]: Remove multiple levels of msecs to jiffies conversions.

The SCTP sysctl entries are displayed in milliseconds, but stored
internally in jiffies. This results in multiple levels of msecs to
jiffies conversion and as a result produces a truncation error. This
patch makes things consistent in that we store and display defaults
in milliseconds and only convert once for use by association.
This patch also adds some sane min/max values so that we don't go off
the deep end.

Signed-off-by: Vladislav Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/constants.h |  13 ++--
 include/net/sctp/structs.h   |  12 ++--
 net/sctp/protocol.c          |   2 +-
 net/sctp/socket.c            |  15 +++--
 net/sctp/sysctl.c            | 140 +++++++++++++++++++------------------------
 net/sctp/transport.c         |   2 +-
 6 files changed, 84 insertions(+), 100 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h
index 57166bfdf8eb..6c632e26f72d 100644
--- a/include/net/sctp/constants.h
+++ b/include/net/sctp/constants.h
@@ -264,10 +264,10 @@ enum { SCTP_MAX_DUP_TSNS = 16 };
 enum { SCTP_MAX_GABS = 16 };
 
 /* Heartbeat interval - 30 secs */
-#define SCTP_DEFAULT_TIMEOUT_HEARTBEAT	(30 * HZ)
+#define SCTP_DEFAULT_TIMEOUT_HEARTBEAT	(30*1000)
 
 /* Delayed sack timer - 200ms */
-#define SCTP_DEFAULT_TIMEOUT_SACK	((200 * HZ) / 1000)
+#define SCTP_DEFAULT_TIMEOUT_SACK	(200)
 
 /* RTO.Initial              - 3  seconds
  * RTO.Min                  - 1  second
@@ -275,9 +275,9 @@ enum { SCTP_MAX_GABS = 16 };
  * RTO.Alpha                - 1/8
  * RTO.Beta                 - 1/4
  */
-#define SCTP_RTO_INITIAL	(3 * HZ)
-#define SCTP_RTO_MIN		(1 * HZ)
-#define SCTP_RTO_MAX		(60 * HZ)
+#define SCTP_RTO_INITIAL	(3 * 1000)
+#define SCTP_RTO_MIN		(1 * 1000)
+#define SCTP_RTO_MAX		(60 * 1000)
 
 #define SCTP_RTO_ALPHA          3   /* 1/8 when converted to right shifts. */
 #define SCTP_RTO_BETA           2   /* 1/4 when converted to right shifts. */
@@ -290,8 +290,7 @@ enum { SCTP_MAX_GABS = 16 };
 #define SCTP_DEF_MAX_INIT 6
 #define SCTP_DEF_MAX_SEND 10
 
-#define SCTP_DEFAULT_COOKIE_LIFE_SEC	60 /* seconds */
-#define SCTP_DEFAULT_COOKIE_LIFE_USEC	0  /* microseconds */
+#define SCTP_DEFAULT_COOKIE_LIFE	(60 * 1000) /* 60 seconds */
 
 #define SCTP_DEFAULT_MINWINDOW	1500	/* default minimum rwnd size */
 #define SCTP_DEFAULT_MAXWINDOW	65535	/* default rwnd size */
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 0412e730c765..c6d93bb0dcd2 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -128,9 +128,9 @@ extern struct sctp_globals {
 	 * RTO.Alpha		    - 1/8  (3 when converted to right shifts.)
 	 * RTO.Beta		    - 1/4  (2 when converted to right shifts.)
 	 */
-	unsigned long rto_initial;
-	unsigned long rto_min;
-	unsigned long rto_max;
+	unsigned int rto_initial;
+	unsigned int rto_min;
+	unsigned int rto_max;
 
 	/* Note: rto_alpha and rto_beta are really defined as inverse
 	 * powers of two to facilitate integer operations.
@@ -145,13 +145,13 @@ extern struct sctp_globals {
 	int cookie_preserve_enable;
 
 	/* Valid.Cookie.Life	    - 60  seconds  */
-	unsigned long valid_cookie_life;
+	unsigned int valid_cookie_life;
 
 	/* Delayed SACK timeout  200ms default*/
-	unsigned long sack_timeout;
+	unsigned int sack_timeout;
 
 	/* HB.interval		    - 30 seconds  */
-	unsigned long hb_interval;
+	unsigned int hb_interval;
 
 	/* Association.Max.Retrans  - 10 attempts
 	 * Path.Max.Retrans	    - 5	 attempts (per destination address)
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index 1ab03a27a76e..5692ef5485d3 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -1049,7 +1049,7 @@ SCTP_STATIC __init int sctp_init(void)
 	sctp_rto_beta			= SCTP_RTO_BETA;
 
 	/* Valid.Cookie.Life        - 60  seconds */
-	sctp_valid_cookie_life		= 60 * HZ;
+	sctp_valid_cookie_life		= SCTP_DEFAULT_COOKIE_LIFE;
 
 	/* Whether Cookie Preservative is enabled(1) or not(0) */
 	sctp_cookie_preserve_enable 	= 1;
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 3b6e82cb372f..7c1dbb1d10df 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -3045,14 +3045,14 @@ SCTP_STATIC int sctp_init_sock(struct sock *sk)
 	sp->initmsg.sinit_num_ostreams   = sctp_max_outstreams;
 	sp->initmsg.sinit_max_instreams  = sctp_max_instreams;
 	sp->initmsg.sinit_max_attempts   = sctp_max_retrans_init;
-	sp->initmsg.sinit_max_init_timeo = jiffies_to_msecs(sctp_rto_max);
+	sp->initmsg.sinit_max_init_timeo = sctp_rto_max;
 
 	/* Initialize default RTO related parameters.  These parameters can
 	 * be modified for with the SCTP_RTOINFO socket option.
 	 */
-	sp->rtoinfo.srto_initial = jiffies_to_msecs(sctp_rto_initial);
-	sp->rtoinfo.srto_max     = jiffies_to_msecs(sctp_rto_max);
-	sp->rtoinfo.srto_min     = jiffies_to_msecs(sctp_rto_min);
+	sp->rtoinfo.srto_initial = sctp_rto_initial;
+	sp->rtoinfo.srto_max     = sctp_rto_max;
+	sp->rtoinfo.srto_min     = sctp_rto_min;
 
 	/* Initialize default association related parameters. These parameters
 	 * can be modified with the SCTP_ASSOCINFO socket option.
@@ -3061,8 +3061,7 @@ SCTP_STATIC int sctp_init_sock(struct sock *sk)
 	sp->assocparams.sasoc_number_peer_destinations = 0;
 	sp->assocparams.sasoc_peer_rwnd = 0;
 	sp->assocparams.sasoc_local_rwnd = 0;
-	sp->assocparams.sasoc_cookie_life = 
-		jiffies_to_msecs(sctp_valid_cookie_life);
+	sp->assocparams.sasoc_cookie_life = sctp_valid_cookie_life;
 
 	/* Initialize default event subscriptions. By default, all the
 	 * options are off. 
@@ -3072,10 +3071,10 @@ SCTP_STATIC int sctp_init_sock(struct sock *sk)
 	/* Default Peer Address Parameters.  These defaults can
 	 * be modified via SCTP_PEER_ADDR_PARAMS
 	 */
-	sp->hbinterval  = jiffies_to_msecs(sctp_hb_interval);
+	sp->hbinterval  = sctp_hb_interval;
 	sp->pathmaxrxt  = sctp_max_retrans_path;
 	sp->pathmtu     = 0; // allow default discovery
-	sp->sackdelay   = jiffies_to_msecs(sctp_sack_timeout);
+	sp->sackdelay   = sctp_sack_timeout;
 	sp->param_flags = SPP_HB_ENABLE |
 	                  SPP_PMTUD_ENABLE |
 	                  SPP_SACKDELAY_ENABLE;
diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c
index dc6f3ff32358..633cd178654b 100644
--- a/net/sctp/sysctl.c
+++ b/net/sctp/sysctl.c
@@ -45,9 +45,10 @@
 #include <net/sctp/sctp.h>
 #include <linux/sysctl.h>
 
-static ctl_handler sctp_sysctl_jiffies_ms;
-static long rto_timer_min = 1;
-static long rto_timer_max = 86400000; /* One day */
+static int zero = 0;
+static int one = 1;
+static int timer_max = 86400000; /* ms in one day */
+static int int_max = INT_MAX;
 static long sack_timer_min = 1;
 static long sack_timer_max = 500;
 
@@ -56,45 +57,45 @@ static ctl_table sctp_table[] = {
 		.ctl_name	= NET_SCTP_RTO_INITIAL,
 		.procname	= "rto_initial",
 		.data		= &sctp_rto_initial,
-		.maxlen		= sizeof(long),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
-		.strategy	= &sctp_sysctl_jiffies_ms,
-		.extra1         = &rto_timer_min,
-		.extra2         = &rto_timer_max
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1         = &one,
+		.extra2         = &timer_max
 	},
 	{
 		.ctl_name	= NET_SCTP_RTO_MIN,
 		.procname	= "rto_min",
 		.data		= &sctp_rto_min,
-		.maxlen		= sizeof(long),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
-		.strategy	= &sctp_sysctl_jiffies_ms,
-		.extra1         = &rto_timer_min,
-		.extra2         = &rto_timer_max
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1         = &one,
+		.extra2         = &timer_max
 	},
 	{
 		.ctl_name	= NET_SCTP_RTO_MAX,
 		.procname	= "rto_max",
 		.data		= &sctp_rto_max,
-		.maxlen		= sizeof(long),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
-		.strategy	= &sctp_sysctl_jiffies_ms,
-		.extra1         = &rto_timer_min,
-		.extra2         = &rto_timer_max
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1         = &one,
+		.extra2         = &timer_max
 	},
 	{
 		.ctl_name	= NET_SCTP_VALID_COOKIE_LIFE,
 		.procname	= "valid_cookie_life",
 		.data		= &sctp_valid_cookie_life,
-		.maxlen		= sizeof(long),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
-		.strategy	= &sctp_sysctl_jiffies_ms,
-		.extra1         = &rto_timer_min,
-		.extra2         = &rto_timer_max
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1         = &one,
+		.extra2         = &timer_max
 	},
 	{
 		.ctl_name	= NET_SCTP_MAX_BURST,
@@ -102,7 +103,10 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_max_burst,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+		.extra2		= &int_max
 	},
 	{
 		.ctl_name	= NET_SCTP_ASSOCIATION_MAX_RETRANS,
@@ -110,7 +114,10 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_max_retrans_association,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &one,
+		.extra2		= &int_max
 	},
 	{
 		.ctl_name	= NET_SCTP_SNDBUF_POLICY,
@@ -118,7 +125,8 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_sndbuf_policy,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_RCVBUF_POLICY,
@@ -126,7 +134,8 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_rcvbuf_policy,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_PATH_MAX_RETRANS,
@@ -134,7 +143,10 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_max_retrans_path,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &one,
+		.extra2		= &int_max
 	},
 	{
 		.ctl_name	= NET_SCTP_MAX_INIT_RETRANSMITS,
@@ -142,18 +154,21 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_max_retrans_init,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &one,
+		.extra2		= &int_max
 	},
 	{
 		.ctl_name	= NET_SCTP_HB_INTERVAL,
 		.procname	= "hb_interval",
 		.data		= &sctp_hb_interval,
-		.maxlen		= sizeof(long),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
-		.strategy	= &sctp_sysctl_jiffies_ms,
-		.extra1         = &rto_timer_min,
-		.extra2         = &rto_timer_max
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1         = &one,
+		.extra2         = &timer_max
 	},
 	{
 		.ctl_name	= NET_SCTP_PRESERVE_ENABLE,
@@ -161,23 +176,26 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_cookie_preserve_enable,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_RTO_ALPHA,
 		.procname	= "rto_alpha_exp_divisor",
 		.data		= &sctp_rto_alpha,
 		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.mode		= 0444,
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_RTO_BETA,
 		.procname	= "rto_beta_exp_divisor",
 		.data		= &sctp_rto_beta,
 		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.mode		= 0444,
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_ADDIP_ENABLE,
@@ -185,7 +203,8 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_addip_enable,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_PRSCTP_ENABLE,
@@ -193,7 +212,8 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_prsctp_enable,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec
+		.proc_handler	= &proc_dointvec,
+		.strategy	= &sysctl_intvec
 	},
 	{
 		.ctl_name	= NET_SCTP_SACK_TIMEOUT,
@@ -201,8 +221,8 @@ static ctl_table sctp_table[] = {
 		.data		= &sctp_sack_timeout,
 		.maxlen		= sizeof(long),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
-		.strategy	= &sctp_sysctl_jiffies_ms,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
 		.extra1         = &sack_timer_min,
 		.extra2         = &sack_timer_max,
 	},
@@ -242,37 +262,3 @@ void sctp_sysctl_unregister(void)
 {
 	unregister_sysctl_table(sctp_sysctl_header);
 }
-
-/* Strategy function to convert jiffies to milliseconds.  */
-static int sctp_sysctl_jiffies_ms(ctl_table *table, int __user *name, int nlen,
-		void __user *oldval, size_t __user *oldlenp,
-		void __user *newval, size_t newlen, void **context) {
-
-	if (oldval) {
-		size_t olen;
-
-		if (oldlenp) {
-			if (get_user(olen, oldlenp))
-				return -EFAULT;
-
-			if (olen != sizeof (int))
-				return -EINVAL;
-		}
-		if (put_user((*(int *)(table->data) * 1000) / HZ,
-			(int __user *)oldval) ||
-		    (oldlenp && put_user(sizeof (int), oldlenp)))
-			return -EFAULT;
-	}
-	if (newval && newlen) {
-		int new;
-
-		if (newlen != sizeof (int))
-			return -EINVAL;
-
-		if (get_user(new, (int __user *)newval))
-			return -EFAULT;
-
-		*(int *)(table->data) = (new * HZ) / 1000;
-	}
-	return 1;
-}
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index 2763aa93de1a..3e5936a5f671 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -75,7 +75,7 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer,
 	 * parameter 'RTO.Initial'.
 	 */
 	peer->rtt = 0;
-	peer->rto = sctp_rto_initial;
+	peer->rto = msecs_to_jiffies(sctp_rto_initial);
 	peer->rttvar = 0;
 	peer->srtt = 0;
 	peer->rto_pending = 0;
-- 
cgit v1.2.3


From 5e032e32ecc2e6cb0385dc115ca9bfe5e19a9539 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:12:24 -0700
Subject: [IPV6] NDISC: Take source address into account for redirects.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Ville Nuorvala <vnuorval@tcs.hut.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_route.h | 1 +
 net/ipv6/ndisc.c        | 3 ++-
 net/ipv6/route.c        | 5 +++--
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 3f170f667c7b..249ce4545ef0 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -110,6 +110,7 @@ extern int			rt6_route_rcv(struct net_device *dev,
 					      struct in6_addr *gwaddr);
 
 extern void			rt6_redirect(struct in6_addr *dest,
+					     struct in6_addr *src,
 					     struct in6_addr *saddr,
 					     struct neighbour *neigh,
 					     u8 *lladdr,
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 419d65163819..32f28dec399e 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1346,7 +1346,8 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
 
 	neigh = __neigh_lookup(&nd_tbl, target, skb->dev, 1);
 	if (neigh) {
-		rt6_redirect(dest, &skb->nh.ipv6h->saddr, neigh, lladdr, 
+		rt6_redirect(dest, &skb->nh.ipv6h->daddr,
+			     &skb->nh.ipv6h->saddr, neigh, lladdr,
 			     on_link);
 		neigh_release(neigh);
 	}
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 5d6e9083ca2c..a9b08a2422e0 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1279,7 +1279,8 @@ static int ip6_route_del(struct fib6_config *cfg)
 /*
  *	Handle redirects
  */
-void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr,
+void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
+		  struct in6_addr *saddr,
 		  struct neighbour *neigh, u8 *lladdr, int on_link)
 {
 	struct rt6_info *rt, *nrt = NULL;
@@ -1304,7 +1305,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr,
 	 */
 
 	read_lock_bh(&table->tb6_lock);
-	fn = fib6_lookup(&table->tb6_root, dest, NULL);
+	fn = fib6_lookup(&table->tb6_root, dest, src);
 restart:
 	for (rt = fn->leaf; rt; rt = rt->u.next) {
 		/*
-- 
cgit v1.2.3


From 8e1ef0a95b87e8b4292b2ba733e8cb854ea2d2fe Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Tue, 29 Aug 2006 17:15:09 -0700
Subject: [IPV6]: Cache source address as well in ipv6_pinfo{}.

Based on MIPL2 kernel patch.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Ville Nuorvala <vnuorval@tcs.hut.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h             | 3 +++
 include/net/ip6_route.h          | 9 ++++++---
 net/dccp/ipv6.c                  | 4 ++--
 net/ipv6/af_inet6.c              | 2 +-
 net/ipv6/datagram.c              | 7 ++++++-
 net/ipv6/inet6_connection_sock.c | 2 +-
 net/ipv6/ip6_output.c            | 3 +++
 net/ipv6/tcp_ipv6.c              | 4 ++--
 net/ipv6/udp.c                   | 7 ++++++-
 9 files changed, 30 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 297853c841b4..02d14a3ff2af 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -242,6 +242,9 @@ struct ipv6_pinfo {
 	struct in6_addr 	rcv_saddr;
 	struct in6_addr		daddr;
 	struct in6_addr		*daddr_cache;
+#ifdef CONFIG_IPV6_SUBTREES
+	struct in6_addr		*saddr_cache;
+#endif
 
 	__u32			flow_label;
 	__u32			frag_size;
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 249ce4545ef0..0d40f84df21b 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -144,21 +144,24 @@ extern rwlock_t rt6_lock;
  *	Store a destination cache entry in a socket
  */
 static inline void __ip6_dst_store(struct sock *sk, struct dst_entry *dst,
-				   struct in6_addr *daddr)
+				   struct in6_addr *daddr, struct in6_addr *saddr)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct rt6_info *rt = (struct rt6_info *) dst;
 
 	sk_setup_caps(sk, dst);
 	np->daddr_cache = daddr;
+#ifdef CONFIG_IPV6_SUBTREES
+	np->saddr_cache = saddr;
+#endif
 	np->dst_cookie = rt->rt6i_node ? rt->rt6i_node->fn_sernum : 0;
 }
 
 static inline void ip6_dst_store(struct sock *sk, struct dst_entry *dst,
-				 struct in6_addr *daddr)
+				 struct in6_addr *daddr, struct in6_addr *saddr)
 {
 	write_lock(&sk->sk_dst_lock);
-	__ip6_dst_store(sk, dst, daddr);
+	__ip6_dst_store(sk, dst, daddr, saddr);
 	write_unlock(&sk->sk_dst_lock);
 }
 
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 231bc7c7e749..f9c5e12d7038 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -231,7 +231,7 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	ipv6_addr_copy(&np->saddr, saddr);
 	inet->rcv_saddr = LOOPBACK4_IPV6;
 
-	__ip6_dst_store(sk, dst, NULL);
+	__ip6_dst_store(sk, dst, NULL, NULL);
 
 	icsk->icsk_ext_hdr_len = 0;
 	if (np->opt != NULL)
@@ -872,7 +872,7 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk,
 	 * comment in that function for the gory details. -acme
 	 */
 
-	__ip6_dst_store(newsk, dst, NULL);
+	__ip6_dst_store(newsk, dst, NULL, NULL);
 	newsk->sk_route_caps = dst->dev->features & ~(NETIF_F_IP_CSUM |
 						      NETIF_F_TSO);
 	newdp6 = (struct dccp6_sock *)newsk;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 2ff600cfe3a4..57ee5ddea96f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -659,7 +659,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 			return err;
 		}
 
-		__ip6_dst_store(sk, dst, NULL);
+		__ip6_dst_store(sk, dst, NULL, NULL);
 	}
 
 	return 0;
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index c73508e090a6..8561b9da6db6 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -193,7 +193,12 @@ ipv4_connected:
 
 	ip6_dst_store(sk, dst,
 		      ipv6_addr_equal(&fl.fl6_dst, &np->daddr) ?
-		      &np->daddr : NULL);
+		      &np->daddr : NULL,
+#ifdef CONFIG_IPV6_SUBTREES
+		      ipv6_addr_equal(&fl.fl6_src, &np->saddr) ?
+		      &np->saddr :
+#endif
+		      NULL);
 
 	sk->sk_state = TCP_ESTABLISHED;
 out:
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 7a51a258615d..827f41d1478b 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -186,7 +186,7 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok)
 			return err;
 		}
 
-		__ip6_dst_store(sk, dst, NULL);
+		__ip6_dst_store(sk, dst, NULL, NULL);
 	}
 
 	skb->dst = dst_clone(dst);
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 0a18cb6b1cbb..2a376b7d91b4 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -762,6 +762,9 @@ static struct dst_entry *ip6_sk_dst_check(struct sock *sk,
 	 * 2. oif also should be the same.
 	 */
 	if (ip6_rt_check(&rt->rt6i_dst, &fl->fl6_dst, np->daddr_cache) ||
+#ifdef CONFIG_IPV6_SUBTREES
+	    ip6_rt_check(&rt->rt6i_src, &fl->fl6_src, np->saddr_cache) ||
+#endif
 	    (fl->oif && fl->oif != dst->dev->ifindex)) {
 		dst_release(dst);
 		dst = NULL;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 7f1b660493b7..2b18918f3011 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -272,7 +272,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	inet->rcv_saddr = LOOPBACK4_IPV6;
 
 	sk->sk_gso_type = SKB_GSO_TCPV6;
-	__ip6_dst_store(sk, dst, NULL);
+	__ip6_dst_store(sk, dst, NULL, NULL);
 
 	icsk->icsk_ext_hdr_len = 0;
 	if (np->opt)
@@ -954,7 +954,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 	 */
 
 	newsk->sk_gso_type = SKB_GSO_TCPV6;
-	__ip6_dst_store(newsk, dst, NULL);
+	__ip6_dst_store(newsk, dst, NULL, NULL);
 
 	newtcp6sk = (struct tcp6_sock *)newsk;
 	inet_sk(newsk)->pinet6 = &newtcp6sk->inet6;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index eb9e1b39c8f8..b9cc55ccb000 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -847,7 +847,12 @@ do_append_data:
 		if (connected) {
 			ip6_dst_store(sk, dst,
 				      ipv6_addr_equal(&fl->fl6_dst, &np->daddr) ?
-				      &np->daddr : NULL);
+				      &np->daddr : NULL,
+#ifdef CONFIG_IPV6_SUBTREES
+				      ipv6_addr_equal(&fl->fl6_src, &np->saddr) ?
+				      &np->saddr :
+#endif
+				      NULL);
 		} else {
 			dst_release(dst);
 		}
-- 
cgit v1.2.3


From 7fc33165a74301b2c5c90b2f2a1f6907cbd5c6f1 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:22:24 -0700
Subject: [IPV6] ROUTE: Put SUBTREE() as FIB6_SUBTREE() into ip6_fib.h for
 future use.

Based on MIPL2 kernel patch.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: Ville Nuorvala <vnuorval@tcs.hut.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h |  5 +++++
 net/ipv6/ip6_fib.c    | 20 +++++++++-----------
 2 files changed, 14 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 9610b887ffb5..6a3f26a04509 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -60,6 +60,11 @@ struct fib6_node
 	__u32			fn_sernum;
 };
 
+#ifndef CONFIG_IPV6_SUBTREES
+#define FIB6_SUBTREE(fn)	NULL
+#else
+#define FIB6_SUBTREE(fn)	((fn)->subtree)
+#endif
 
 /*
  *	routing information
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index b706424e70b8..6536e33d8353 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -73,10 +73,8 @@ static DEFINE_RWLOCK(fib6_walker_lock);
 
 #ifdef CONFIG_IPV6_SUBTREES
 #define FWS_INIT FWS_S
-#define SUBTREE(fn) ((fn)->subtree)
 #else
 #define FWS_INIT FWS_L
-#define SUBTREE(fn) NULL
 #endif
 
 static void fib6_prune_clones(struct fib6_node *fn, struct rt6_info *rt);
@@ -854,7 +852,7 @@ static struct fib6_node * fib6_lookup_1(struct fib6_node *root,
 	}
 
 	while(fn) {
-		if (SUBTREE(fn) || fn->fn_flags & RTN_RTINFO) {
+		if (FIB6_SUBTREE(fn) || fn->fn_flags & RTN_RTINFO) {
 			struct rt6key *key;
 
 			key = (struct rt6key *) ((u8 *) fn->leaf +
@@ -985,7 +983,7 @@ static struct rt6_info * fib6_find_prefix(struct fib6_node *fn)
 		if(fn->right)
 			return fn->right->leaf;
 
-		fn = SUBTREE(fn);
+		fn = FIB6_SUBTREE(fn);
 	}
 	return NULL;
 }
@@ -1016,7 +1014,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 		if (fn->right) child = fn->right, children |= 1;
 		if (fn->left) child = fn->left, children |= 2;
 
-		if (children == 3 || SUBTREE(fn) 
+		if (children == 3 || FIB6_SUBTREE(fn)
 #ifdef CONFIG_IPV6_SUBTREES
 		    /* Subtree root (i.e. fn) may have one child */
 		    || (children && fn->fn_flags&RTN_ROOT)
@@ -1035,9 +1033,9 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 
 		pn = fn->parent;
 #ifdef CONFIG_IPV6_SUBTREES
-		if (SUBTREE(pn) == fn) {
+		if (FIB6_SUBTREE(pn) == fn) {
 			BUG_TRAP(fn->fn_flags&RTN_ROOT);
-			SUBTREE(pn) = NULL;
+			FIB6_SUBTREE(pn) = NULL;
 			nstate = FWS_L;
 		} else {
 			BUG_TRAP(!(fn->fn_flags&RTN_ROOT));
@@ -1085,7 +1083,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn)
 		read_unlock(&fib6_walker_lock);
 
 		node_free(fn);
-		if (pn->fn_flags&RTN_RTINFO || SUBTREE(pn))
+		if (pn->fn_flags&RTN_RTINFO || FIB6_SUBTREE(pn))
 			return pn;
 
 		rt6_release(pn->leaf);
@@ -1228,8 +1226,8 @@ static int fib6_walk_continue(struct fib6_walker_t *w)
 		switch (w->state) {
 #ifdef CONFIG_IPV6_SUBTREES
 		case FWS_S:
-			if (SUBTREE(fn)) {
-				w->node = SUBTREE(fn);
+			if (FIB6_SUBTREE(fn)) {
+				w->node = FIB6_SUBTREE(fn);
 				continue;
 			}
 			w->state = FWS_L;
@@ -1263,7 +1261,7 @@ static int fib6_walk_continue(struct fib6_walker_t *w)
 			pn = fn->parent;
 			w->node = pn;
 #ifdef CONFIG_IPV6_SUBTREES
-			if (SUBTREE(pn) == fn) {
+			if (FIB6_SUBTREE(pn) == fn) {
 				BUG_TRAP(fn->fn_flags&RTN_ROOT);
 				w->state = FWS_L;
 				continue;
-- 
cgit v1.2.3


From 77d16f450ae0452d7d4b009f78debb1294fb435c Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:25:05 -0700
Subject: [IPV6] ROUTE: Unify RT6_F_xxx and RT6_SELECT_F_xxx flags

Unify RT6_F_xxx and RT6_SELECT_F_xxx flags into
RT6_LOOKUP_F_xxx flags, and put them into ip6_route.h

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Acked-by: Ville Nuorvala <vnuorval@tcs.hut.fi
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip6_fib.h   |  3 ---
 include/net/ip6_route.h |  4 ++++
 net/ipv6/fib6_rules.c   |  2 +-
 net/ipv6/route.c        | 32 ++++++++++++--------------------
 4 files changed, 17 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 6a3f26a04509..e4438de3bd6b 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -173,9 +173,6 @@ struct fib6_table {
 #define RT6_TABLE_LOCAL		RT6_TABLE_MAIN
 #endif
 
-#define RT6_F_STRICT		1
-#define RT6_F_HAS_SADDR		2
-
 typedef struct rt6_info *(*pol_lookup_t)(struct fib6_table *,
 					 struct flowi *, int);
 
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 0d40f84df21b..297909570041 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -32,6 +32,10 @@ struct route_info {
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 
+#define RT6_LOOKUP_F_IFACE	0x1
+#define RT6_LOOKUP_F_REACHABLE	0x2
+#define RT6_LOOKUP_F_HAS_SADDR	0x4
+
 struct pol_chain {
 	int			type;
 	int			priority;
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 2c4fbc855e6c..7b4908cc52b3 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -117,7 +117,7 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 	if (!ipv6_prefix_equal(&fl->fl6_dst, &r->dst.addr, r->dst.plen))
 		return 0;
 
-	if ((flags & RT6_F_HAS_SADDR) &&
+	if ((flags & RT6_LOOKUP_F_HAS_SADDR) &&
 	    !ipv6_prefix_equal(&fl->fl6_src, &r->src.addr, r->src.plen))
 		return 0;
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fd6f2ec4fa09..20691285aee5 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -76,9 +76,6 @@
 
 #define CLONE_OFFLINK_ROUTE 0
 
-#define RT6_SELECT_F_IFACE	0x1
-#define RT6_SELECT_F_REACHABLE	0x2
-
 static int ip6_rt_max_size = 4096;
 static int ip6_rt_gc_min_interval = HZ / 2;
 static int ip6_rt_gc_timeout = 60*HZ;
@@ -340,7 +337,7 @@ static int rt6_score_route(struct rt6_info *rt, int oif,
 	int m, n;
 		
 	m = rt6_check_dev(rt, oif);
-	if (!m && (strict & RT6_SELECT_F_IFACE))
+	if (!m && (strict & RT6_LOOKUP_F_IFACE))
 		return -1;
 #ifdef CONFIG_IPV6_ROUTER_PREF
 	m |= IPV6_DECODE_PREF(IPV6_EXTRACT_PREF(rt->rt6i_flags)) << 2;
@@ -348,7 +345,7 @@ static int rt6_score_route(struct rt6_info *rt, int oif,
 	n = rt6_check_neigh(rt);
 	if (n > 1)
 		m |= 16;
-	else if (!n && strict & RT6_SELECT_F_REACHABLE)
+	else if (!n && strict & RT6_LOOKUP_F_REACHABLE)
 		return -1;
 	return m;
 }
@@ -388,7 +385,7 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif,
 	}
 
 	if (!match &&
-	    (strict & RT6_SELECT_F_REACHABLE) &&
+	    (strict & RT6_LOOKUP_F_REACHABLE) &&
 	    last && last != rt0) {
 		/* no entries matched; do round-robin */
 		static DEFINE_SPINLOCK(lock);
@@ -511,7 +508,7 @@ static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table,
 	fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src);
 restart:
 	rt = fn->leaf;
-	rt = rt6_device_match(rt, fl->oif, flags & RT6_F_STRICT);
+	rt = rt6_device_match(rt, fl->oif, flags);
 	BACKTRACK(&fl->fl6_src);
 	dst_hold(&rt->u.dst);
 out:
@@ -537,7 +534,7 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
 		},
 	};
 	struct dst_entry *dst;
-	int flags = strict ? RT6_F_STRICT : 0;
+	int flags = strict ? RT6_LOOKUP_F_IFACE : 0;
 
 	dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup);
 	if (dst->error == 0)
@@ -633,10 +630,9 @@ static struct rt6_info *ip6_pol_route_input(struct fib6_table *table,
 	int strict = 0;
 	int attempts = 3;
 	int err;
-	int reachable = RT6_SELECT_F_REACHABLE;
+	int reachable = RT6_LOOKUP_F_REACHABLE;
 
-	if (flags & RT6_F_STRICT)
-		strict = RT6_SELECT_F_IFACE;
+	strict |= flags & RT6_LOOKUP_F_IFACE;
 
 relookup:
 	read_lock_bh(&table->tb6_lock);
@@ -712,10 +708,7 @@ void ip6_route_input(struct sk_buff *skb)
 		},
 		.proto = iph->nexthdr,
 	};
-	int flags = 0;
-
-	if (rt6_need_strict(&iph->daddr))
-		flags |= RT6_F_STRICT;
+	int flags = rt6_need_strict(&iph->daddr) ? RT6_LOOKUP_F_IFACE : 0;
 
 	skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input);
 }
@@ -728,10 +721,9 @@ static struct rt6_info *ip6_pol_route_output(struct fib6_table *table,
 	int strict = 0;
 	int attempts = 3;
 	int err;
-	int reachable = RT6_SELECT_F_REACHABLE;
+	int reachable = RT6_LOOKUP_F_REACHABLE;
 
-	if (flags & RT6_F_STRICT)
-		strict = RT6_SELECT_F_IFACE;
+	strict |= flags & RT6_LOOKUP_F_IFACE;
 
 relookup:
 	read_lock_bh(&table->tb6_lock);
@@ -797,7 +789,7 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
 	int flags = 0;
 
 	if (rt6_need_strict(&fl->fl6_dst))
-		flags |= RT6_F_STRICT;
+		flags |= RT6_LOOKUP_F_IFACE;
 
 	return fib6_rule_lookup(fl, flags, ip6_pol_route_output);
 }
@@ -1362,7 +1354,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest,
 		},
 		.gateway = *gateway,
 	};
-	int flags = rt6_need_strict(dest) ? RT6_F_STRICT : 0;
+	int flags = rt6_need_strict(dest) ? RT6_LOOKUP_F_IFACE : 0;
 
 	return (struct rt6_info *)fib6_rule_lookup((struct flowi *)&rdfl, flags, __ip6_route_redirect);
 }
-- 
cgit v1.2.3


From 7e49e6de30efa716614e280d97963c570f3acf29 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Fri, 22 Sep 2006 15:05:15 -0700
Subject: [XFRM]: Add XFRM_MODE_xxx for future use.

Transformation mode is used as either IPsec transport or tunnel.
It is required to add two more items, route optimization and inbound trigger
for Mobile IPv6.
Based on MIPL2 kernel patch.

This patch was also written by: Ville Nuorvala <vnuorval@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h    |  6 ++++--
 include/net/xfrm.h      |  2 +-
 net/ipv4/ah4.c          |  2 +-
 net/ipv4/esp4.c         |  6 +++---
 net/ipv4/ipcomp.c       |  8 ++++----
 net/ipv4/xfrm4_input.c  |  2 +-
 net/ipv4/xfrm4_output.c |  4 ++--
 net/ipv4/xfrm4_policy.c |  2 +-
 net/ipv4/xfrm4_state.c  |  2 +-
 net/ipv4/xfrm4_tunnel.c |  2 +-
 net/ipv6/ah6.c          |  2 +-
 net/ipv6/esp6.c         |  4 ++--
 net/ipv6/ipcomp6.c      |  6 +++---
 net/ipv6/xfrm6_input.c  |  2 +-
 net/ipv6/xfrm6_output.c |  4 ++--
 net/ipv6/xfrm6_policy.c |  2 +-
 net/ipv6/xfrm6_state.c  |  2 +-
 net/ipv6/xfrm6_tunnel.c |  2 +-
 net/key/af_key.c        |  6 +++---
 net/xfrm/xfrm_policy.c  | 11 ++++++-----
 net/xfrm/xfrm_user.c    |  4 ++--
 21 files changed, 42 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 46a15c7a1a13..5154064b6d95 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -120,7 +120,9 @@ enum
 
 #define XFRM_MODE_TRANSPORT 0
 #define XFRM_MODE_TUNNEL 1
-#define XFRM_MODE_MAX 2
+#define XFRM_MODE_ROUTEOPTIMIZATION 2
+#define XFRM_MODE_IN_TRIGGER 3
+#define XFRM_MODE_MAX 4
 
 /* Netlink configuration messages.  */
 enum {
@@ -247,7 +249,7 @@ struct xfrm_usersa_info {
 	__u32				seq;
 	__u32				reqid;
 	__u16				family;
-	__u8				mode; /* 0=transport,1=tunnel */
+	__u8				mode;		/* XFRM_MODE_xxx */
 	__u8				replay_window;
 	__u8				flags;
 #define XFRM_STATE_NOECN	1
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 00bf86e6e82b..762795624b10 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -298,7 +298,7 @@ struct xfrm_tmpl
 
 	__u32			reqid;
 
-/* Mode: transport/tunnel */
+/* Mode: transport, tunnel etc. */
 	__u8			mode;
 
 /* Sharing mode: unique, this session only, this user only etc. */
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index 008e69d2e423..99542977e47e 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -265,7 +265,7 @@ static int ah_init_state(struct xfrm_state *x)
 		goto error;
 	
 	x->props.header_len = XFRM_ALIGN8(sizeof(struct ip_auth_hdr) + ahp->icv_trunc_len);
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct iphdr);
 	x->data = ahp;
 
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index b428489f6ccd..e87377e1d6b6 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -248,7 +248,7 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 		 *    as per draft-ietf-ipsec-udp-encaps-06,
 		 *    section 3.1.2
 		 */
-		if (!x->props.mode)
+		if (x->props.mode == XFRM_MODE_TRANSPORT)
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 
@@ -267,7 +267,7 @@ static u32 esp4_get_max_size(struct xfrm_state *x, int mtu)
 	struct esp_data *esp = x->data;
 	u32 blksize = ALIGN(crypto_blkcipher_blocksize(esp->conf.tfm), 4);
 
-	if (x->props.mode) {
+	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		mtu = ALIGN(mtu + 2, blksize);
 	} else {
 		/* The worst case. */
@@ -383,7 +383,7 @@ static int esp_init_state(struct xfrm_state *x)
 	if (crypto_blkcipher_setkey(tfm, esp->conf.key, esp->conf.key_len))
 		goto error;
 	x->props.header_len = sizeof(struct ip_esp_hdr) + esp->conf.ivlen;
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct iphdr);
 	if (x->encap) {
 		struct xfrm_encap_tmpl *encap = x->encap;
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index 5bb9c9f03fb6..17342430a843 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -176,7 +176,7 @@ static int ipcomp_output(struct xfrm_state *x, struct sk_buff *skb)
 	return 0;
 
 out_ok:
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		ip_send_check(iph);
 	return 0;
 }
@@ -216,7 +216,7 @@ static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x)
 	t->id.daddr.a4 = x->id.daddr.a4;
 	memcpy(&t->sel, &x->sel, sizeof(t->sel));
 	t->props.family = AF_INET;
-	t->props.mode = 1;
+	t->props.mode = XFRM_MODE_TUNNEL;
 	t->props.saddr.a4 = x->props.saddr.a4;
 	t->props.flags = x->props.flags;
 
@@ -416,7 +416,7 @@ static int ipcomp_init_state(struct xfrm_state *x)
 		goto out;
 
 	x->props.header_len = 0;
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct iphdr);
 
 	mutex_lock(&ipcomp_resource_mutex);
@@ -428,7 +428,7 @@ static int ipcomp_init_state(struct xfrm_state *x)
 		goto error;
 	mutex_unlock(&ipcomp_resource_mutex);
 
-	if (x->props.mode) {
+	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		err = ipcomp_tunnel_attach(x);
 		if (err)
 			goto error_tunnel;
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index 817ed84511a6..040e8475f295 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -106,7 +106,7 @@ int xfrm4_rcv_encap(struct sk_buff *skb, __u16 encap_type)
 		if (x->mode->input(x, skb))
 			goto drop;
 
-		if (x->props.mode) {
+		if (x->props.mode == XFRM_MODE_TUNNEL) {
 			decaps = 1;
 			break;
 		}
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 4a96a9e3ef3b..5fd115f0c547 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -54,7 +54,7 @@ static int xfrm4_output_one(struct sk_buff *skb)
 			goto error_nolock;
 	}
 
-	if (x->props.mode) {
+	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		err = xfrm4_tunnel_check_size(skb);
 		if (err)
 			goto error_nolock;
@@ -85,7 +85,7 @@ static int xfrm4_output_one(struct sk_buff *skb)
 		}
 		dst = skb->dst;
 		x = dst->xfrm;
-	} while (x && !x->props.mode);
+	} while (x && (x->props.mode != XFRM_MODE_TUNNEL));
 
 	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
 	err = 0;
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 8f50eae47d03..a5bed741de2c 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -96,7 +96,7 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 
 		dst1->next = dst_prev;
 		dst_prev = dst1;
-		if (xfrm[i]->props.mode) {
+		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
 			remote = xfrm[i]->id.daddr.a4;
 			local  = xfrm[i]->props.saddr.a4;
 			tunnel = 1;
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 81e1751c966e..97b0c7589711 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -42,7 +42,7 @@ __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->props.saddr = tmpl->saddr;
 	if (x->props.saddr.a4 == 0)
 		x->props.saddr.a4 = saddr->a4;
-	if (tmpl->mode && x->props.saddr.a4 == 0) {
+	if (tmpl->mode == XFRM_MODE_TUNNEL && x->props.saddr.a4 == 0) {
 		struct rtable *rt;
 	        struct flowi fl_tunnel = {
         	        .nl_u = {
diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c
index f8ceaa127c83..f110af5b1319 100644
--- a/net/ipv4/xfrm4_tunnel.c
+++ b/net/ipv4/xfrm4_tunnel.c
@@ -28,7 +28,7 @@ static int ipip_xfrm_rcv(struct xfrm_state *x, struct sk_buff *skb)
 
 static int ipip_init_state(struct xfrm_state *x)
 {
-	if (!x->props.mode)
+	if (x->props.mode != XFRM_MODE_TUNNEL)
 		return -EINVAL;
 
 	if (x->encap)
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 00ffa7bc6c9f..60954fc7eb36 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -398,7 +398,7 @@ static int ah6_init_state(struct xfrm_state *x)
 		goto error;
 	
 	x->props.header_len = XFRM_ALIGN8(sizeof(struct ipv6_auth_hdr) + ahp->icv_trunc_len);
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct ipv6hdr);
 	x->data = ahp;
 
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 2ebfd281e721..2b8e52e1d0ab 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -237,7 +237,7 @@ static u32 esp6_get_max_size(struct xfrm_state *x, int mtu)
 	struct esp_data *esp = x->data;
 	u32 blksize = ALIGN(crypto_blkcipher_blocksize(esp->conf.tfm), 4);
 
-	if (x->props.mode) {
+	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		mtu = ALIGN(mtu + 2, blksize);
 	} else {
 		/* The worst case. */
@@ -358,7 +358,7 @@ static int esp6_init_state(struct xfrm_state *x)
 	if (crypto_blkcipher_setkey(tfm, esp->conf.key, esp->conf.key_len))
 		goto error;
 	x->props.header_len = sizeof(struct ipv6_esp_hdr) + esp->conf.ivlen;
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct ipv6hdr);
 	x->data = esp;
 	return 0;
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index a81e9e9d93bd..19eba8d9f851 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -212,7 +212,7 @@ static struct xfrm_state *ipcomp6_tunnel_create(struct xfrm_state *x)
 	memcpy(t->id.daddr.a6, x->id.daddr.a6, sizeof(struct in6_addr));
 	memcpy(&t->sel, &x->sel, sizeof(t->sel));
 	t->props.family = AF_INET6;
-	t->props.mode = 1;
+	t->props.mode = XFRM_MODE_TUNNEL;
 	memcpy(t->props.saddr.a6, x->props.saddr.a6, sizeof(struct in6_addr));
 
 	if (xfrm_init_state(t))
@@ -417,7 +417,7 @@ static int ipcomp6_init_state(struct xfrm_state *x)
 		goto out;
 
 	x->props.header_len = 0;
-	if (x->props.mode)
+	if (x->props.mode == XFRM_MODE_TUNNEL)
 		x->props.header_len += sizeof(struct ipv6hdr);
 	
 	mutex_lock(&ipcomp6_resource_mutex);
@@ -429,7 +429,7 @@ static int ipcomp6_init_state(struct xfrm_state *x)
 		goto error;
 	mutex_unlock(&ipcomp6_resource_mutex);
 
-	if (x->props.mode) {
+	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		err = ipcomp6_tunnel_attach(x);
 		if (err)
 			goto error_tunnel;
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index 0405d74ff910..ee2f6b3908b6 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -72,7 +72,7 @@ int xfrm6_rcv_spi(struct sk_buff *skb, u32 spi)
 		if (x->mode->input(x, skb))
 			goto drop;
 
-		if (x->props.mode) { /* XXX */
+		if (x->props.mode == XFRM_MODE_TUNNEL) { /* XXX */
 			decaps = 1;
 			break;
 		}
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index 6d111743e508..26f18869f77b 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -47,7 +47,7 @@ static int xfrm6_output_one(struct sk_buff *skb)
 			goto error_nolock;
 	}
 
-	if (x->props.mode) {
+	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		err = xfrm6_tunnel_check_size(skb);
 		if (err)
 			goto error_nolock;
@@ -80,7 +80,7 @@ static int xfrm6_output_one(struct sk_buff *skb)
 		}
 		dst = skb->dst;
 		x = dst->xfrm;
-	} while (x && !x->props.mode);
+	} while (x && (x->props.mode != XFRM_MODE_TUNNEL));
 
 	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
 	err = 0;
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 73cd250aecbb..81355bb50328 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -114,7 +114,7 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 
 		dst1->next = dst_prev;
 		dst_prev = dst1;
-		if (xfrm[i]->props.mode) {
+		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
 			remote = (struct in6_addr*)&xfrm[i]->id.daddr;
 			local  = (struct in6_addr*)&xfrm[i]->props.saddr;
 			tunnel = 1;
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index b33296b3f6de..a1a1f5476442 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -42,7 +42,7 @@ __xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	memcpy(&x->props.saddr, &tmpl->saddr, sizeof(x->props.saddr));
 	if (ipv6_addr_any((struct in6_addr*)&x->props.saddr))
 		memcpy(&x->props.saddr, saddr, sizeof(x->props.saddr));
-	if (tmpl->mode && ipv6_addr_any((struct in6_addr*)&x->props.saddr)) {
+	if (tmpl->mode == XFRM_MODE_TUNNEL && ipv6_addr_any((struct in6_addr*)&x->props.saddr)) {
 		struct rt6_info *rt;
 		struct flowi fl_tunnel = {
 			.nl_u = {
diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
index c8f9369c2a87..59685ee8f700 100644
--- a/net/ipv6/xfrm6_tunnel.c
+++ b/net/ipv6/xfrm6_tunnel.c
@@ -307,7 +307,7 @@ static int xfrm6_tunnel_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 
 static int xfrm6_tunnel_init_state(struct xfrm_state *x)
 {
-	if (!x->props.mode)
+	if (x->props.mode != XFRM_MODE_TUNNEL)
 		return -EINVAL;
 
 	if (x->encap)
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 797c744a8438..19e047b0e678 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -1765,7 +1765,7 @@ parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq)
 	}
 
 	/* addresses present only in tunnel mode */
-	if (t->mode) {
+	if (t->mode == XFRM_MODE_TUNNEL) {
 		switch (xp->family) {
 		case AF_INET:
 			sin = (void*)(rq+1);
@@ -1997,7 +1997,7 @@ static void pfkey_xfrm_policy2msg(struct sk_buff *skb, struct xfrm_policy *xp, i
 		int req_size;
 
 		req_size = sizeof(struct sadb_x_ipsecrequest);
-		if (t->mode)
+		if (t->mode == XFRM_MODE_TUNNEL)
 			req_size += 2*socklen;
 		else
 			size -= 2*socklen;
@@ -2013,7 +2013,7 @@ static void pfkey_xfrm_policy2msg(struct sk_buff *skb, struct xfrm_policy *xp, i
 		if (t->optional)
 			rq->sadb_x_ipsecrequest_level = IPSEC_LEVEL_USE;
 		rq->sadb_x_ipsecrequest_reqid = t->reqid;
-		if (t->mode) {
+		if (t->mode == XFRM_MODE_TUNNEL) {
 			switch (xp->family) {
 			case AF_INET:
 				sin = (void*)(rq+1);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 32c963c90573..a0d58971391d 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -779,7 +779,7 @@ xfrm_tmpl_resolve(struct xfrm_policy *policy, struct flowi *fl,
 		xfrm_address_t *local  = saddr;
 		struct xfrm_tmpl *tmpl = &policy->xfrm_vec[i];
 
-		if (tmpl->mode) {
+		if (tmpl->mode == XFRM_MODE_TUNNEL) {
 			remote = &tmpl->id.daddr;
 			local = &tmpl->saddr;
 		}
@@ -1005,7 +1005,8 @@ xfrm_state_ok(struct xfrm_tmpl *tmpl, struct xfrm_state *x,
 		(x->props.reqid == tmpl->reqid || !tmpl->reqid) &&
 		x->props.mode == tmpl->mode &&
 		(tmpl->aalgos & (1<<x->props.aalgo)) &&
-		!(x->props.mode && xfrm_state_addr_cmp(tmpl, x, family));
+		!(x->props.mode != XFRM_MODE_TRANSPORT &&
+		  xfrm_state_addr_cmp(tmpl, x, family));
 }
 
 static inline int
@@ -1015,14 +1016,14 @@ xfrm_policy_ok(struct xfrm_tmpl *tmpl, struct sec_path *sp, int start,
 	int idx = start;
 
 	if (tmpl->optional) {
-		if (!tmpl->mode)
+		if (tmpl->mode == XFRM_MODE_TRANSPORT)
 			return start;
 	} else
 		start = -1;
 	for (; idx < sp->len; idx++) {
 		if (xfrm_state_ok(tmpl, sp->xvec[idx], family))
 			return ++idx;
-		if (sp->xvec[idx]->props.mode)
+		if (sp->xvec[idx]->props.mode != XFRM_MODE_TRANSPORT)
 			break;
 	}
 	return start;
@@ -1047,7 +1048,7 @@ EXPORT_SYMBOL(xfrm_decode_session);
 static inline int secpath_has_tunnel(struct sec_path *sp, int k)
 {
 	for (; k < sp->len; k++) {
-		if (sp->xvec[k]->props.mode)
+		if (sp->xvec[k]->props.mode != XFRM_MODE_TRANSPORT)
 			return 1;
 	}
 
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index f70e158874d2..0d580ac19771 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -174,8 +174,8 @@ static int verify_newsa_info(struct xfrm_usersa_info *p,
 
 	err = -EINVAL;
 	switch (p->mode) {
-	case 0:
-	case 1:
+	case XFRM_MODE_TRANSPORT:
+	case XFRM_MODE_TUNNEL:
 		break;
 
 	default:
-- 
cgit v1.2.3


From 5794708f11551b6d19b10673abf4b0202f66b44d Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Fri, 22 Sep 2006 15:06:24 -0700
Subject: [XFRM]: Introduce a helper to compare id protocol.

Put the helper to header for future use.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    | 6 ++++++
 net/xfrm/xfrm_state.c | 6 +++---
 2 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 762795624b10..5b364b0a6a28 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -9,6 +9,7 @@
 #include <linux/skbuff.h>
 #include <linux/socket.h>
 #include <linux/pfkeyv2.h>
+#include <linux/ipsec.h>
 #include <linux/in6.h>
 #include <linux/mutex.h>
 
@@ -835,6 +836,11 @@ static inline int xfrm_state_kern(struct xfrm_state *x)
 	return atomic_read(&x->tunnel_users);
 }
 
+static inline int xfrm_id_proto_match(u8 proto, u8 userproto)
+{
+	return (userproto == IPSEC_PROTO_ANY || proto == userproto);
+}
+
 /*
  * xfrm algorithm information
  */
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 1c796087ee78..34c038cbdf46 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -294,7 +294,7 @@ void xfrm_state_flush(u8 proto)
 restart:
 		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
 			if (!xfrm_state_kern(x) &&
-			    (proto == IPSEC_PROTO_ANY || x->id.proto == proto)) {
+			    xfrm_id_proto_match(x->id.proto, proto)) {
 				xfrm_state_hold(x);
 				spin_unlock_bh(&xfrm_state_lock);
 
@@ -772,7 +772,7 @@ int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
 	spin_lock_bh(&xfrm_state_lock);
 	for (i = 0; i < XFRM_DST_HSIZE; i++) {
 		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
-			if (proto == IPSEC_PROTO_ANY || x->id.proto == proto)
+			if (xfrm_id_proto_match(x->id.proto, proto))
 				count++;
 		}
 	}
@@ -783,7 +783,7 @@ int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
 
 	for (i = 0; i < XFRM_DST_HSIZE; i++) {
 		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
-			if (proto != IPSEC_PROTO_ANY && x->id.proto != proto)
+			if (!xfrm_id_proto_match(x->id.proto, proto))
 				continue;
 			err = func(x, --count, data);
 			if (err)
-- 
cgit v1.2.3


From dc00a525603650a1471c823a1e48c6505c2f9765 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:49:52 -0700
Subject: [XFRM] STATE: Allow non IPsec protocol.

It will be added two more transformation protocols (routing header
and destination options header) for Mobile IPv6.
xfrm_id_proto_match() can be handle zero as all, IPSEC_PROTO_ANY as
all IPsec and otherwise as exact one.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h   | 5 ++++-
 net/xfrm/xfrm_user.c | 2 +-
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 5b364b0a6a28..2a7d2132a1ae 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -838,7 +838,10 @@ static inline int xfrm_state_kern(struct xfrm_state *x)
 
 static inline int xfrm_id_proto_match(u8 proto, u8 userproto)
 {
-	return (userproto == IPSEC_PROTO_ANY || proto == userproto);
+	return (!userproto || proto == userproto ||
+		(userproto == IPSEC_PROTO_ANY && (proto == IPPROTO_AH ||
+						  proto == IPPROTO_ESP ||
+						  proto == IPPROTO_COMP)));
 }
 
 /*
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 0d580ac19771..41f3d51ffc33 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -542,7 +542,7 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb)
 	info.nlmsg_flags = NLM_F_MULTI;
 	info.this_idx = 0;
 	info.start_idx = cb->args[0];
-	(void) xfrm_state_walk(IPSEC_PROTO_ANY, dump_one_state, &info);
+	(void) xfrm_state_walk(0, dump_one_state, &info);
 	cb->args[0] = info.this_idx;
 
 	return skb->len;
-- 
cgit v1.2.3


From 622dc8281a80374873686514e46f852093d91106 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:52:01 -0700
Subject: [XFRM]: Expand XFRM_MAX_DEPTH for route optimization.

XFRM_MAX_DEPTH is a limit of transformation states to be applied to the same
flow. Two more extension headers are used by Mobile IPv6 transformation.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 2a7d2132a1ae..aa3be68041be 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -314,7 +314,7 @@ struct xfrm_tmpl
 	__u32			calgos;
 };
 
-#define XFRM_MAX_DEPTH		4
+#define XFRM_MAX_DEPTH		6
 
 struct xfrm_policy
 {
-- 
cgit v1.2.3


From 6c44e6b7ab500d7e3e3f406c83325671be51a752 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:53:57 -0700
Subject: [XFRM] STATE: Add source address list.

Support source address based searching.
Mobile IPv6 will use it.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 26 ++++++++++++++++++++++++++
 net/ipv4/xfrm4_state.c |  3 +++
 net/ipv6/xfrm6_state.c |  3 +++
 net/xfrm/xfrm_state.c  | 21 +++++++++++++++++++--
 4 files changed, 51 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index aa3be68041be..88145e3348d0 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -95,6 +95,7 @@ struct xfrm_state
 {
 	/* Note: bydst is re-used during gc */
 	struct list_head	bydst;
+	struct list_head	bysrc;
 	struct list_head	byspi;
 
 	atomic_t		refcnt;
@@ -236,6 +237,7 @@ extern int __xfrm_state_delete(struct xfrm_state *x);
 struct xfrm_state_afinfo {
 	unsigned short		family;
 	struct list_head	*state_bydst;
+	struct list_head	*state_bysrc;
 	struct list_head	*state_byspi;
 	int			(*init_flags)(struct xfrm_state *x);
 	void			(*init_tempsel)(struct xfrm_state *x, struct flowi *fl,
@@ -420,6 +422,30 @@ unsigned xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
 	return 0;
 }
 
+static __inline__
+unsigned __xfrm4_src_hash(xfrm_address_t *addr)
+{
+	return __xfrm4_dst_hash(addr);
+}
+
+static __inline__
+unsigned __xfrm6_src_hash(xfrm_address_t *addr)
+{
+	return __xfrm6_dst_hash(addr);
+}
+
+static __inline__
+unsigned xfrm_src_hash(xfrm_address_t *addr, unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		return __xfrm4_src_hash(addr);
+	case AF_INET6:
+		return __xfrm6_src_hash(addr);
+	}
+	return 0;
+}
+
 static __inline__
 unsigned __xfrm4_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
 {
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 97b0c7589711..c56b258fad73 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -122,6 +122,9 @@ __xfrm4_find_acq(u8 mode, u32 reqid, u8 proto,
 		add_timer(&x0->timer);
 		xfrm_state_hold(x0);
 		list_add_tail(&x0->bydst, xfrm4_state_afinfo.state_bydst+h);
+		h = __xfrm4_src_hash(saddr);
+		xfrm_state_hold(x0);
+		list_add_tail(&x0->bysrc, xfrm4_state_afinfo.state_bysrc+h);
 		wake_up(&km_waitq);
 	}
 	if (x0)
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index a1a1f5476442..2fb07850449f 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -126,6 +126,9 @@ __xfrm6_find_acq(u8 mode, u32 reqid, u8 proto,
 		add_timer(&x0->timer);
 		xfrm_state_hold(x0);
 		list_add_tail(&x0->bydst, xfrm6_state_afinfo.state_bydst+h);
+		h = __xfrm6_src_hash(saddr);
+		xfrm_state_hold(x0);
+		list_add_tail(&x0->bysrc, xfrm6_state_afinfo.state_bysrc+h);
 		wake_up(&km_waitq);
 	}
 	if (x0)
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 34c038cbdf46..2a9992894e69 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -45,6 +45,7 @@ static DEFINE_SPINLOCK(xfrm_state_lock);
  * Also, it can be used by ah/esp icmp error handler to find offending SA.
  */
 static struct list_head xfrm_state_bydst[XFRM_DST_HSIZE];
+static struct list_head xfrm_state_bysrc[XFRM_DST_HSIZE];
 static struct list_head xfrm_state_byspi[XFRM_DST_HSIZE];
 
 DECLARE_WAIT_QUEUE_HEAD(km_waitq);
@@ -200,6 +201,7 @@ struct xfrm_state *xfrm_state_alloc(void)
 		atomic_set(&x->refcnt, 1);
 		atomic_set(&x->tunnel_users, 0);
 		INIT_LIST_HEAD(&x->bydst);
+		INIT_LIST_HEAD(&x->bysrc);
 		INIT_LIST_HEAD(&x->byspi);
 		init_timer(&x->timer);
 		x->timer.function = xfrm_timer_handler;
@@ -240,6 +242,8 @@ int __xfrm_state_delete(struct xfrm_state *x)
 		spin_lock(&xfrm_state_lock);
 		list_del(&x->bydst);
 		__xfrm_state_put(x);
+		list_del(&x->bysrc);
+		__xfrm_state_put(x);
 		if (x->id.spi) {
 			list_del(&x->byspi);
 			__xfrm_state_put(x);
@@ -415,6 +419,8 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 			x->km.state = XFRM_STATE_ACQ;
 			list_add_tail(&x->bydst, xfrm_state_bydst+h);
 			xfrm_state_hold(x);
+			list_add_tail(&x->bysrc, xfrm_state_bysrc+h);
+			xfrm_state_hold(x);
 			if (x->id.spi) {
 				h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto, family);
 				list_add(&x->byspi, xfrm_state_byspi+h);
@@ -448,11 +454,19 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 	list_add(&x->bydst, xfrm_state_bydst+h);
 	xfrm_state_hold(x);
 
-	h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto, x->props.family);
+	h = xfrm_src_hash(&x->props.saddr, x->props.family);
 
-	list_add(&x->byspi, xfrm_state_byspi+h);
+	list_add(&x->bysrc, xfrm_state_bysrc+h);
 	xfrm_state_hold(x);
 
+	if (xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY)) {
+		h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto,
+				  x->props.family);
+
+		list_add(&x->byspi, xfrm_state_byspi+h);
+		xfrm_state_hold(x);
+	}
+
 	if (!mod_timer(&x->timer, jiffies + HZ))
 		xfrm_state_hold(x);
 
@@ -1075,6 +1089,7 @@ int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo)
 		err = -ENOBUFS;
 	else {
 		afinfo->state_bydst = xfrm_state_bydst;
+		afinfo->state_bysrc = xfrm_state_bysrc;
 		afinfo->state_byspi = xfrm_state_byspi;
 		xfrm_state_afinfo[afinfo->family] = afinfo;
 	}
@@ -1097,6 +1112,7 @@ int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo)
 		else {
 			xfrm_state_afinfo[afinfo->family] = NULL;
 			afinfo->state_byspi = NULL;
+			afinfo->state_bysrc = NULL;
 			afinfo->state_bydst = NULL;
 		}
 	}
@@ -1218,6 +1234,7 @@ void __init xfrm_state_init(void)
 
 	for (i=0; i<XFRM_DST_HSIZE; i++) {
 		INIT_LIST_HEAD(&xfrm_state_bydst[i]);
+		INIT_LIST_HEAD(&xfrm_state_bysrc[i]);
 		INIT_LIST_HEAD(&xfrm_state_byspi[i]);
 	}
 	INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task, NULL);
-- 
cgit v1.2.3


From eb2971b68a7d17a7d0fa2c7fc6fbc4bfe41cd694 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:56:04 -0700
Subject: [XFRM] STATE: Search by address using source address list.

This is a support to search transformation states by its addresses
by using source address list for Mobile IPv6 usage.
To use it from user-space, it is also added a message type for
source address as a xfrm state option.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h   |  1 +
 include/net/xfrm.h     |  2 ++
 net/ipv4/xfrm4_state.c |  9 ++++++++
 net/ipv6/xfrm6_state.c | 21 ++++++++++++++++++
 net/xfrm/xfrm_state.c  | 37 +++++++++++++++++++++++++++----
 net/xfrm/xfrm_user.c   | 59 +++++++++++++++++++++++++++++++++++++++++++++-----
 6 files changed, 119 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 5154064b6d95..66343d3d4b91 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -234,6 +234,7 @@ enum xfrm_attr_type_t {
 	XFRMA_REPLAY_VAL,
 	XFRMA_REPLAY_THRESH,
 	XFRMA_ETIMER_THRESH,
+	XFRMA_SRCADDR,		/* xfrm_address_t */
 	__XFRMA_MAX
 
 #define XFRMA_MAX (__XFRMA_MAX - 1)
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 88145e3348d0..d9c40e713184 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -244,6 +244,7 @@ struct xfrm_state_afinfo {
 						struct xfrm_tmpl *tmpl,
 						xfrm_address_t *daddr, xfrm_address_t *saddr);
 	struct xfrm_state	*(*state_lookup)(xfrm_address_t *daddr, u32 spi, u8 proto);
+	struct xfrm_state	*(*state_lookup_byaddr)(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto);
 	struct xfrm_state	*(*find_acq)(u8 mode, u32 reqid, u8 proto, 
 					     xfrm_address_t *daddr, xfrm_address_t *saddr, 
 					     int create);
@@ -937,6 +938,7 @@ extern void xfrm_state_insert(struct xfrm_state *x);
 extern int xfrm_state_add(struct xfrm_state *x);
 extern int xfrm_state_update(struct xfrm_state *x);
 extern struct xfrm_state *xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family);
+extern struct xfrm_state *xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family);
 extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq);
 extern int xfrm_state_delete(struct xfrm_state *x);
 extern void xfrm_state_flush(u8 proto);
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index c56b258fad73..616be131b4e3 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -80,6 +80,14 @@ __xfrm4_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto)
 	return NULL;
 }
 
+/* placeholder until ipv4's code is written */
+static struct xfrm_state *
+__xfrm4_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
+			    u8 proto)
+{
+	return NULL;
+}
+
 static struct xfrm_state *
 __xfrm4_find_acq(u8 mode, u32 reqid, u8 proto, 
 		 xfrm_address_t *daddr, xfrm_address_t *saddr, 
@@ -137,6 +145,7 @@ static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
 	.state_lookup		= __xfrm4_state_lookup,
+	.state_lookup_byaddr	= __xfrm4_state_lookup_byaddr,
 	.find_acq		= __xfrm4_find_acq,
 };
 
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 2fb07850449f..9c95b9d3e110 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -63,6 +63,26 @@ __xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->props.family = AF_INET6;
 }
 
+static struct xfrm_state *
+__xfrm6_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
+			    u8 proto)
+{
+	struct xfrm_state *x = NULL;
+	unsigned h;
+
+	h = __xfrm6_src_hash(saddr);
+	list_for_each_entry(x, xfrm6_state_afinfo.state_bysrc+h, bysrc) {
+		if (x->props.family == AF_INET6 &&
+		    ipv6_addr_equal((struct in6_addr *)daddr, (struct in6_addr *)x->id.daddr.a6) &&
+		    ipv6_addr_equal((struct in6_addr *)saddr, (struct in6_addr *)x->props.saddr.a6) &&
+		    proto == x->id.proto) {
+			xfrm_state_hold(x);
+			return x;
+		}
+	}
+	return NULL;
+}
+
 static struct xfrm_state *
 __xfrm6_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto)
 {
@@ -140,6 +160,7 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.state_lookup		= __xfrm6_state_lookup,
+	.state_lookup_byaddr	= __xfrm6_state_lookup_byaddr,
 	.find_acq		= __xfrm6_find_acq,
 };
 
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 2a9992894e69..11f480b12952 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -487,6 +487,16 @@ void xfrm_state_insert(struct xfrm_state *x)
 }
 EXPORT_SYMBOL(xfrm_state_insert);
 
+static inline struct xfrm_state *
+__xfrm_state_locate(struct xfrm_state_afinfo *afinfo, struct xfrm_state *x,
+		    int use_spi)
+{
+	if (use_spi)
+		return afinfo->state_lookup(&x->id.daddr, x->id.spi, x->id.proto);
+	else
+		return afinfo->state_lookup_byaddr(&x->id.daddr, &x->props.saddr, x->id.proto);
+}
+
 static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq);
 
 int xfrm_state_add(struct xfrm_state *x)
@@ -495,6 +505,7 @@ int xfrm_state_add(struct xfrm_state *x)
 	struct xfrm_state *x1;
 	int family;
 	int err;
+	int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY);
 
 	family = x->props.family;
 	afinfo = xfrm_state_get_afinfo(family);
@@ -503,7 +514,7 @@ int xfrm_state_add(struct xfrm_state *x)
 
 	spin_lock_bh(&xfrm_state_lock);
 
-	x1 = afinfo->state_lookup(&x->id.daddr, x->id.spi, x->id.proto);
+	x1 = __xfrm_state_locate(afinfo, x, use_spi);
 	if (x1) {
 		xfrm_state_put(x1);
 		x1 = NULL;
@@ -511,7 +522,7 @@ int xfrm_state_add(struct xfrm_state *x)
 		goto out;
 	}
 
-	if (x->km.seq) {
+	if (use_spi && x->km.seq) {
 		x1 = __xfrm_find_acq_byseq(x->km.seq);
 		if (x1 && xfrm_addr_cmp(&x1->id.daddr, &x->id.daddr, family)) {
 			xfrm_state_put(x1);
@@ -519,7 +530,7 @@ int xfrm_state_add(struct xfrm_state *x)
 		}
 	}
 
-	if (!x1)
+	if (use_spi && !x1)
 		x1 = afinfo->find_acq(
 			x->props.mode, x->props.reqid, x->id.proto,
 			&x->id.daddr, &x->props.saddr, 0);
@@ -548,13 +559,14 @@ int xfrm_state_update(struct xfrm_state *x)
 	struct xfrm_state_afinfo *afinfo;
 	struct xfrm_state *x1;
 	int err;
+	int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY);
 
 	afinfo = xfrm_state_get_afinfo(x->props.family);
 	if (unlikely(afinfo == NULL))
 		return -EAFNOSUPPORT;
 
 	spin_lock_bh(&xfrm_state_lock);
-	x1 = afinfo->state_lookup(&x->id.daddr, x->id.spi, x->id.proto);
+	x1 = __xfrm_state_locate(afinfo, x, use_spi);
 
 	err = -ESRCH;
 	if (!x1)
@@ -674,6 +686,23 @@ xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto,
 }
 EXPORT_SYMBOL(xfrm_state_lookup);
 
+struct xfrm_state *
+xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
+			 u8 proto, unsigned short family)
+{
+	struct xfrm_state *x;
+	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
+	if (!afinfo)
+		return NULL;
+
+	spin_lock_bh(&xfrm_state_lock);
+	x = afinfo->state_lookup_byaddr(daddr, saddr, proto);
+	spin_unlock_bh(&xfrm_state_lock);
+	xfrm_state_put_afinfo(afinfo);
+	return x;
+}
+EXPORT_SYMBOL(xfrm_state_lookup_byaddr);
+
 struct xfrm_state *
 xfrm_find_acq(u8 mode, u32 reqid, u8 proto, 
 	      xfrm_address_t *daddr, xfrm_address_t *saddr, 
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 41f3d51ffc33..b5f8ab71aa54 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -87,6 +87,22 @@ static int verify_encap_tmpl(struct rtattr **xfrma)
 	return 0;
 }
 
+static int verify_one_addr(struct rtattr **xfrma, enum xfrm_attr_type_t type,
+			   xfrm_address_t **addrp)
+{
+	struct rtattr *rt = xfrma[type - 1];
+
+	if (!rt)
+		return 0;
+
+	if ((rt->rta_len - sizeof(*rt)) < sizeof(**addrp))
+		return -EINVAL;
+
+	if (addrp)
+		*addrp = RTA_DATA(rt);
+
+	return 0;
+}
 
 static inline int verify_sec_ctx_len(struct rtattr **xfrma)
 {
@@ -418,16 +434,48 @@ out:
 	return err;
 }
 
+static struct xfrm_state *xfrm_user_state_lookup(struct xfrm_usersa_id *p,
+						 struct rtattr **xfrma,
+						 int *errp)
+{
+	struct xfrm_state *x = NULL;
+	int err;
+
+	if (xfrm_id_proto_match(p->proto, IPSEC_PROTO_ANY)) {
+		err = -ESRCH;
+		x = xfrm_state_lookup(&p->daddr, p->spi, p->proto, p->family);
+	} else {
+		xfrm_address_t *saddr = NULL;
+
+		err = verify_one_addr(xfrma, XFRMA_SRCADDR, &saddr);
+		if (err)
+			goto out;
+
+		if (!saddr) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		x = xfrm_state_lookup_byaddr(&p->daddr, saddr, p->proto,
+					     p->family);
+	}
+
+ out:
+	if (!x && errp)
+		*errp = err;
+	return x;
+}
+
 static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
 {
 	struct xfrm_state *x;
-	int err;
+	int err = -ESRCH;
 	struct km_event c;
 	struct xfrm_usersa_id *p = NLMSG_DATA(nlh);
 
-	x = xfrm_state_lookup(&p->daddr, p->spi, p->proto, p->family);
+	x = xfrm_user_state_lookup(p, (struct rtattr **)xfrma, &err);
 	if (x == NULL)
-		return -ESRCH;
+		return err;
 
 	if ((err = security_xfrm_state_delete(x)) != 0)
 		goto out;
@@ -578,10 +626,9 @@ static int xfrm_get_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
 	struct xfrm_usersa_id *p = NLMSG_DATA(nlh);
 	struct xfrm_state *x;
 	struct sk_buff *resp_skb;
-	int err;
+	int err = -ESRCH;
 
-	x = xfrm_state_lookup(&p->daddr, p->spi, p->proto, p->family);
-	err = -ESRCH;
+	x = xfrm_user_state_lookup(p, (struct rtattr **)xfrma, &err);
 	if (x == NULL)
 		goto out_noput;
 
-- 
cgit v1.2.3


From aee5adb4307c4c63a4dc5f3b49984d76f8a71b5b Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 17:57:28 -0700
Subject: [XFRM] STATE: Add a hook to find offset to be inserted header in
 outbound.

On current kernel, ip6_find_1stfragopt() is used by IPv6 IPsec to find
offset to be inserted header in outbound for transport mode. (BTW, no
usage may be needed for IPv4 case.)  Mobile IPv6 requires another
logic for routing header and destination options header
respectively. This patch is common platform for the offset and adopts
it to IPsec.

Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h              | 3 +++
 net/ipv6/ah6.c                  | 3 ++-
 net/ipv6/esp6.c                 | 3 ++-
 net/ipv6/ipcomp6.c              | 1 +
 net/ipv6/ipv6_syms.c            | 1 +
 net/ipv6/xfrm6_mode_transport.c | 2 +-
 net/ipv6/xfrm6_output.c         | 6 ++++++
 7 files changed, 16 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d9c40e713184..eed48f832ce1 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -265,6 +265,7 @@ struct xfrm_type
 	void			(*destructor)(struct xfrm_state *);
 	int			(*input)(struct xfrm_state *, struct sk_buff *skb);
 	int			(*output)(struct xfrm_state *, struct sk_buff *pskb);
+	int			(*hdr_offset)(struct xfrm_state *, struct sk_buff *, u8 **);
 	/* Estimate maximal size of result of transformation of a dgram */
 	u32			(*get_max_size)(struct xfrm_state *, int size);
 };
@@ -960,6 +961,8 @@ extern u32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr);
 extern void xfrm6_tunnel_free_spi(xfrm_address_t *saddr);
 extern u32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr);
 extern int xfrm6_output(struct sk_buff *skb);
+extern int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb,
+				 u8 **prevhdr);
 
 #ifdef CONFIG_XFRM
 extern int xfrm4_rcv_encap(struct sk_buff *skb, __u16 encap_type);
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 60954fc7eb36..6c0aa51319a5 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -435,7 +435,8 @@ static struct xfrm_type ah6_type =
 	.init_state	= ah6_init_state,
 	.destructor	= ah6_destroy,
 	.input		= ah6_input,
-	.output		= ah6_output
+	.output		= ah6_output,
+	.hdr_offset	= xfrm6_find_1stfragopt,
 };
 
 static struct inet6_protocol ah6_protocol = {
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 2b8e52e1d0ab..ae50b9511151 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -379,7 +379,8 @@ static struct xfrm_type esp6_type =
 	.destructor	= esp6_destroy,
 	.get_max_size	= esp6_get_max_size,
 	.input		= esp6_input,
-	.output		= esp6_output
+	.output		= esp6_output,
+	.hdr_offset	= xfrm6_find_1stfragopt,
 };
 
 static struct inet6_protocol esp6_protocol = {
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index 19eba8d9f851..ad9c6e824e62 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -461,6 +461,7 @@ static struct xfrm_type ipcomp6_type =
 	.destructor	= ipcomp6_destroy,
 	.input		= ipcomp6_input,
 	.output		= ipcomp6_output,
+	.hdr_offset	= xfrm6_find_1stfragopt,
 };
 
 static struct inet6_protocol ipcomp6_protocol = 
diff --git a/net/ipv6/ipv6_syms.c b/net/ipv6/ipv6_syms.c
index dd4d1ce77769..e1a741612888 100644
--- a/net/ipv6/ipv6_syms.c
+++ b/net/ipv6/ipv6_syms.c
@@ -31,6 +31,7 @@ EXPORT_SYMBOL(ipv6_chk_addr);
 EXPORT_SYMBOL(in6_dev_finish_destroy);
 #ifdef CONFIG_XFRM
 EXPORT_SYMBOL(xfrm6_rcv);
+EXPORT_SYMBOL(xfrm6_find_1stfragopt);
 #endif
 EXPORT_SYMBOL(rt6_lookup);
 EXPORT_SYMBOL(ipv6_push_nfrag_opts);
diff --git a/net/ipv6/xfrm6_mode_transport.c b/net/ipv6/xfrm6_mode_transport.c
index 711d713e36d8..a5dce216024d 100644
--- a/net/ipv6/xfrm6_mode_transport.c
+++ b/net/ipv6/xfrm6_mode_transport.c
@@ -35,7 +35,7 @@ static int xfrm6_transport_output(struct sk_buff *skb)
 	skb_push(skb, x->props.header_len);
 	iph = skb->nh.ipv6h;
 
-	hdr_len = ip6_find_1stfragopt(skb, &prevhdr);
+	hdr_len = x->type->hdr_offset(x, skb, &prevhdr);
 	skb->nh.raw = prevhdr - x->props.header_len;
 	skb->h.raw = skb->data + hdr_len;
 	memmove(skb->data, iph, hdr_len);
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index 26f18869f77b..b4628fbf8ff5 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -17,6 +17,12 @@
 #include <net/ipv6.h>
 #include <net/xfrm.h>
 
+int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb,
+			  u8 **prevhdr)
+{
+	return ip6_find_1stfragopt(skb, prevhdr);
+}
+
 static int xfrm6_tunnel_check_size(struct sk_buff *skb)
 {
 	int mtu, ret = 0;
-- 
cgit v1.2.3


From fbd9a5b47ee9c319ff0cae584391241ce78ffd6b Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 18:08:21 -0700
Subject: [XFRM] STATE: Common receive function for route optimization
 extension headers.

XFRM_STATE_WILDRECV flag is introduced; the last resort state is set
it and receives packet which is not route optimized but uses such
extension headers i.e. Mobile IPv6 signaling (binding update and
acknowledgement).  A node enabled Mobile IPv6 adds the state.

Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h   |   1 +
 include/net/xfrm.h     |   2 +
 net/ipv6/ipv6_syms.c   |   1 +
 net/ipv6/xfrm6_input.c | 108 +++++++++++++++++++++++++++++++++++++++++++++++++
 net/xfrm/xfrm_state.c  |   1 +
 5 files changed, 113 insertions(+)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 66343d3d4b91..a7c9e4cfb15b 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -256,6 +256,7 @@ struct xfrm_usersa_info {
 #define XFRM_STATE_NOECN	1
 #define XFRM_STATE_DECAP_DSCP	2
 #define XFRM_STATE_NOPMTUDISC	4
+#define XFRM_STATE_WILDRECV	8
 };
 
 struct xfrm_usersa_id {
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index eed48f832ce1..0d735a5aba61 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -955,6 +955,8 @@ extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler);
 extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler);
 extern int xfrm6_rcv_spi(struct sk_buff *skb, u32 spi);
 extern int xfrm6_rcv(struct sk_buff **pskb);
+extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
+			    xfrm_address_t *saddr, u8 proto);
 extern int xfrm6_tunnel_register(struct xfrm6_tunnel *handler);
 extern int xfrm6_tunnel_deregister(struct xfrm6_tunnel *handler);
 extern u32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr);
diff --git a/net/ipv6/ipv6_syms.c b/net/ipv6/ipv6_syms.c
index e1a741612888..7b7b90d9c3d0 100644
--- a/net/ipv6/ipv6_syms.c
+++ b/net/ipv6/ipv6_syms.c
@@ -31,6 +31,7 @@ EXPORT_SYMBOL(ipv6_chk_addr);
 EXPORT_SYMBOL(in6_dev_finish_destroy);
 #ifdef CONFIG_XFRM
 EXPORT_SYMBOL(xfrm6_rcv);
+EXPORT_SYMBOL(xfrm6_input_addr);
 EXPORT_SYMBOL(xfrm6_find_1stfragopt);
 #endif
 EXPORT_SYMBOL(rt6_lookup);
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index ee2f6b3908b6..a40a05789013 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -138,3 +138,111 @@ int xfrm6_rcv(struct sk_buff **pskb)
 {
 	return xfrm6_rcv_spi(*pskb, 0);
 }
+
+int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
+		     xfrm_address_t *saddr, u8 proto)
+{
+ 	struct xfrm_state *x = NULL;
+ 	int wildcard = 0;
+	struct in6_addr any;
+	xfrm_address_t *xany;
+	struct xfrm_state *xfrm_vec_one = NULL;
+ 	int nh = 0;
+	int i = 0;
+
+	ipv6_addr_set(&any, 0, 0, 0, 0);
+	xany = (xfrm_address_t *)&any;
+
+	for (i = 0; i < 3; i++) {
+		xfrm_address_t *dst, *src;
+		switch (i) {
+		case 0:
+			dst = daddr;
+			src = saddr;
+			break;
+		case 1:
+			/* lookup state with wild-card source address */
+			wildcard = 1;
+			dst = daddr;
+			src = xany;
+			break;
+		case 2:
+		default:
+ 			/* lookup state with wild-card addresses */
+			wildcard = 1; /* XXX */
+			dst = xany;
+			src = xany;
+			break;
+ 		}
+
+		x = xfrm_state_lookup_byaddr(dst, src, proto, AF_INET6);
+		if (!x)
+			continue;
+
+		spin_lock(&x->lock);
+
+		if (wildcard) {
+			if ((x->props.flags & XFRM_STATE_WILDRECV) == 0) {
+				spin_unlock(&x->lock);
+				xfrm_state_put(x);
+				x = NULL;
+				continue;
+			}
+		}
+
+		if (unlikely(x->km.state != XFRM_STATE_VALID)) {
+			spin_unlock(&x->lock);
+			xfrm_state_put(x);
+ 			x = NULL;
+ 			continue;
+		}
+		if (xfrm_state_check_expire(x)) {
+			spin_unlock(&x->lock);
+			xfrm_state_put(x);
+			x = NULL;
+			continue;
+		}
+
+		nh = x->type->input(x, skb);
+		if (nh <= 0) {
+			spin_unlock(&x->lock);
+			xfrm_state_put(x);
+			x = NULL;
+			continue;
+		}
+
+		x->curlft.bytes += skb->len;
+		x->curlft.packets++;
+
+		spin_unlock(&x->lock);
+
+		xfrm_vec_one = x;
+		break;
+	}
+
+	if (!xfrm_vec_one)
+		goto drop;
+
+	/* Allocate new secpath or COW existing one. */
+	if (!skb->sp || atomic_read(&skb->sp->refcnt) != 1) {
+		struct sec_path *sp;
+		sp = secpath_dup(skb->sp);
+		if (!sp)
+			goto drop;
+		if (skb->sp)
+			secpath_put(skb->sp);
+		skb->sp = sp;
+	}
+
+	if (1 + skb->sp->len > XFRM_MAX_DEPTH)
+		goto drop;
+
+	skb->sp->xvec[skb->sp->len] = xfrm_vec_one;
+	skb->sp->len ++;
+
+	return 1;
+drop:
+	if (xfrm_vec_one)
+		xfrm_state_put(xfrm_vec_one);
+	return -1;
+}
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 11f480b12952..f05371556cce 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -352,6 +352,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 	list_for_each_entry(x, xfrm_state_bydst+h, bydst) {
 		if (x->props.family == family &&
 		    x->props.reqid == tmpl->reqid &&
+		    !(x->props.flags & XFRM_STATE_WILDRECV) &&
 		    xfrm_state_addr_check(x, daddr, saddr, family) &&
 		    tmpl->mode == x->props.mode &&
 		    tmpl->id.proto == x->id.proto &&
-- 
cgit v1.2.3


From 99505a843673faeae962a8cde128c7c034ba6b5e Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 18:10:33 -0700
Subject: [XFRM] STATE: Add a hook to obtain local/remote outbound address.

Outbound transformation replaces both source and destination address with
state's end-point addresses at the same time when IPsec tunnel mode.
It is also required to change them for Mobile IPv6 route optimization, but we
should care about the following differences:
 - changing result is not end-point but care-of address
 - either source or destination is replaced for each state
This hook is a common platform to change outbound address.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  2 ++
 net/ipv6/xfrm6_policy.c | 20 ++++++++++++++++++--
 2 files changed, 20 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 0d735a5aba61..aa3ac994477b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -266,6 +266,8 @@ struct xfrm_type
 	int			(*input)(struct xfrm_state *, struct sk_buff *skb);
 	int			(*output)(struct xfrm_state *, struct sk_buff *pskb);
 	int			(*hdr_offset)(struct xfrm_state *, struct sk_buff *, u8 **);
+	xfrm_address_t		*(*local_addr)(struct xfrm_state *, xfrm_address_t *);
+	xfrm_address_t		*(*remote_addr)(struct xfrm_state *, xfrm_address_t *);
 	/* Estimate maximal size of result of transformation of a dgram */
 	u32			(*get_max_size)(struct xfrm_state *, int size);
 };
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 81355bb50328..9328fc88708a 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -59,6 +59,22 @@ __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 	return dst;
 }
 
+static inline struct in6_addr*
+__xfrm6_bundle_addr_remote(struct xfrm_state *x, struct in6_addr *addr)
+{
+	return (x->type->remote_addr) ?
+		(struct in6_addr*)x->type->remote_addr(x, (xfrm_address_t *)addr) :
+		(struct in6_addr*)&x->id.daddr;
+}
+
+static inline struct in6_addr*
+__xfrm6_bundle_addr_local(struct xfrm_state *x, struct in6_addr *addr)
+{
+	return (x->type->local_addr) ?
+		(struct in6_addr*)x->type->local_addr(x, (xfrm_address_t *)addr) :
+		(struct in6_addr*)&x->props.saddr;
+}
+
 /* Allocate chain of dst_entry's, attach known xfrm's, calculate
  * all the metrics... Shortly, bundle a bundle.
  */
@@ -115,8 +131,8 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		dst1->next = dst_prev;
 		dst_prev = dst1;
 		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
-			remote = (struct in6_addr*)&xfrm[i]->id.daddr;
-			local  = (struct in6_addr*)&xfrm[i]->props.saddr;
+			remote = __xfrm6_bundle_addr_remote(xfrm[i], remote);
+			local  = __xfrm6_bundle_addr_local(xfrm[i], local);
 			tunnel = 1;
 		}
 		header_len += xfrm[i]->props.header_len;
-- 
cgit v1.2.3


From 1b5c229987dc4d0c92a38fac0cde2aeec08cd775 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 18:11:50 -0700
Subject: [XFRM] STATE: Support non-fragment outbound transformation headers.

For originated outbound IPv6 packets which will fragment, ip6_append_data()
should know length of extension headers before sending them and
the length is carried by dst_entry.
IPv6 IPsec headers fragment then transformation was
designed to place all headers after fragment header.
OTOH Mobile IPv6 extension headers do not fragment then
it is a good idea to make dst_entry have non-fragment length to tell it
to ip6_append_data().

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/dst.h       |  1 +
 include/net/xfrm.h      |  2 ++
 net/ipv4/xfrm4_policy.c |  1 +
 net/ipv6/ip6_output.c   |  2 +-
 net/ipv6/xfrm6_policy.c | 24 ++++++++++++++++++++++--
 5 files changed, 27 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/dst.h b/include/net/dst.h
index 36d54fc248b0..a8d825f90305 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -54,6 +54,7 @@ struct dst_entry
 	unsigned long		expires;
 
 	unsigned short		header_len;	/* more space at head required */
+	unsigned short		nfheader_len;	/* more non-fragment space at head required */
 	unsigned short		trailer_len;	/* space to reserve at tail */
 
 	u32			metrics[RTAX_MAX];
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index aa3ac994477b..aa93cc1f6299 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -260,6 +260,8 @@ struct xfrm_type
 	char			*description;
 	struct module		*owner;
 	__u8			proto;
+	__u8			flags;
+#define XFRM_TYPE_NON_FRAGMENT	1
 
 	int			(*init_state)(struct xfrm_state *x);
 	void			(*destructor)(struct xfrm_state *);
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index a5bed741de2c..e517981ceadd 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -135,6 +135,7 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		dst_prev->flags	       |= DST_HOST;
 		dst_prev->lastuse	= jiffies;
 		dst_prev->header_len	= header_len;
+		dst_prev->nfheader_len	= 0;
 		dst_prev->trailer_len	= trailer_len;
 		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 2a376b7d91b4..258e3e45f5e0 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -971,7 +971,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 
 	hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);
 
-	fragheaderlen = sizeof(struct ipv6hdr) + (opt ? opt->opt_nflen : 0);
+	fragheaderlen = sizeof(struct ipv6hdr) + rt->u.dst.nfheader_len + (opt ? opt->opt_nflen : 0);
 	maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen - sizeof(struct frag_hdr);
 
 	if (mtu <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN) {
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 9328fc88708a..a3f68c8b737e 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -75,6 +75,24 @@ __xfrm6_bundle_addr_local(struct xfrm_state *x, struct in6_addr *addr)
 		(struct in6_addr*)&x->props.saddr;
 }
 
+static inline void
+__xfrm6_bundle_len_inc(int *len, int *nflen, struct xfrm_state *x)
+{
+	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
+		*nflen += x->props.header_len;
+	else
+		*len += x->props.header_len;
+}
+
+static inline void
+__xfrm6_bundle_len_dec(int *len, int *nflen, struct xfrm_state *x)
+{
+	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
+		*nflen -= x->props.header_len;
+	else
+		*len -= x->props.header_len;
+}
+
 /* Allocate chain of dst_entry's, attach known xfrm's, calculate
  * all the metrics... Shortly, bundle a bundle.
  */
@@ -99,6 +117,7 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 	int i;
 	int err = 0;
 	int header_len = 0;
+	int nfheader_len = 0;
 	int trailer_len = 0;
 
 	dst = dst_prev = NULL;
@@ -135,7 +154,7 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 			local  = __xfrm6_bundle_addr_local(xfrm[i], local);
 			tunnel = 1;
 		}
-		header_len += xfrm[i]->props.header_len;
+		__xfrm6_bundle_len_inc(&header_len, &nfheader_len, xfrm[i]);
 		trailer_len += xfrm[i]->props.trailer_len;
 
 		if (tunnel) {
@@ -170,6 +189,7 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		dst_prev->flags	       |= DST_HOST;
 		dst_prev->lastuse	= jiffies;
 		dst_prev->header_len	= header_len;
+		dst_prev->nfheader_len	= nfheader_len;
 		dst_prev->trailer_len	= trailer_len;
 		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
 
@@ -188,7 +208,7 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		x->u.rt6.rt6i_src      = rt0->rt6i_src;	
 		x->u.rt6.rt6i_idev     = rt0->rt6i_idev;
 		in6_dev_hold(rt0->rt6i_idev);
-		header_len -= x->u.dst.xfrm->props.header_len;
+		__xfrm6_bundle_len_dec(&header_len, &nfheader_len, x->u.dst.xfrm);
 		trailer_len -= x->u.dst.xfrm->props.trailer_len;
 	}
 
-- 
cgit v1.2.3


From 060f02a3bdd4d9ba8aa3c48e9b470672b1f3a585 Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Wed, 23 Aug 2006 18:18:55 -0700
Subject: [XFRM] STATE: Introduce care-of address.

Care-of address is carried by state as a transformation option like
IPsec encryption/authentication algorithm.

Based on MIPL2 kernel patch.

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/xfrm.h  |  1 +
 include/net/xfrm.h    |  3 +++
 net/xfrm/xfrm_state.c |  6 ++++++
 net/xfrm/xfrm_user.c  | 28 +++++++++++++++++++++++++++-
 4 files changed, 37 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index a7c9e4cfb15b..b53f799189af 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -235,6 +235,7 @@ enum xfrm_attr_type_t {
 	XFRMA_REPLAY_THRESH,
 	XFRMA_ETIMER_THRESH,
 	XFRMA_SRCADDR,		/* xfrm_address_t */
+	XFRMA_COADDR,		/* xfrm_address_t */
 	__XFRMA_MAX
 
 #define XFRMA_MAX (__XFRMA_MAX - 1)
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index aa93cc1f6299..872a2a4022b2 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -134,6 +134,9 @@ struct xfrm_state
 	/* Data for encapsulator */
 	struct xfrm_encap_tmpl	*encap;
 
+	/* Data for care-of address */
+	xfrm_address_t	*coaddr;
+
 	/* IPComp needs an IPIP tunnel for handling uncompressed packets */
 	struct xfrm_state	*tunnel;
 
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index f05371556cce..3da89c01ea71 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -78,6 +78,7 @@ static void xfrm_state_gc_destroy(struct xfrm_state *x)
 	kfree(x->ealg);
 	kfree(x->calg);
 	kfree(x->encap);
+	kfree(x->coaddr);
 	if (x->mode)
 		xfrm_put_mode(x->mode);
 	if (x->type) {
@@ -603,6 +604,11 @@ out:
 	if (likely(x1->km.state == XFRM_STATE_VALID)) {
 		if (x->encap && x1->encap)
 			memcpy(x1->encap, x->encap, sizeof(*x1->encap));
+		if (x->coaddr && x1->coaddr) {
+			memcpy(x1->coaddr, x->coaddr, sizeof(*x1->coaddr));
+		}
+		if (!use_spi && memcmp(&x1->sel, &x->sel, sizeof(x1->sel)))
+			memcpy(&x1->sel, &x->sel, sizeof(x1->sel));
 		memcpy(&x1->lft, &x->lft, sizeof(x1->lft));
 		x1->km.dying = 0;
 
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index b5f8ab71aa54..939808de9e20 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -187,11 +187,14 @@ static int verify_newsa_info(struct xfrm_usersa_info *p,
 		goto out;
 	if ((err = verify_sec_ctx_len(xfrma)))
 		goto out;
+	if ((err = verify_one_addr(xfrma, XFRMA_COADDR, NULL)))
+		goto out;
 
 	err = -EINVAL;
 	switch (p->mode) {
 	case XFRM_MODE_TRANSPORT:
 	case XFRM_MODE_TUNNEL:
+	case XFRM_MODE_ROUTEOPTIMIZATION:
 		break;
 
 	default:
@@ -276,6 +279,24 @@ static int attach_sec_ctx(struct xfrm_state *x, struct rtattr *u_arg)
 	return security_xfrm_state_alloc(x, uctx);
 }
 
+static int attach_one_addr(xfrm_address_t **addrpp, struct rtattr *u_arg)
+{
+	struct rtattr *rta = u_arg;
+	xfrm_address_t *p, *uaddrp;
+
+	if (!rta)
+		return 0;
+
+	uaddrp = RTA_DATA(rta);
+	p = kmalloc(sizeof(*p), GFP_KERNEL);
+	if (!p)
+		return -ENOMEM;
+
+	memcpy(p, uaddrp, sizeof(*p));
+	*addrpp = p;
+	return 0;
+}
+
 static void copy_from_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p)
 {
 	memcpy(&x->id, &p->id, sizeof(x->id));
@@ -365,7 +386,8 @@ static struct xfrm_state *xfrm_state_construct(struct xfrm_usersa_info *p,
 		goto error;
 	if ((err = attach_encap_tmpl(&x->encap, xfrma[XFRMA_ENCAP-1])))
 		goto error;
-
+	if ((err = attach_one_addr(&x->coaddr, xfrma[XFRMA_COADDR-1])))
+		goto error;
 	err = xfrm_init_state(x);
 	if (err)
 		goto error;
@@ -569,6 +591,10 @@ static int dump_one_state(struct xfrm_state *x, int count, void *ptr)
 		uctx->ctx_len = x->security->ctx_len;
 		memcpy(uctx + 1, x->security->ctx_str, x->security->ctx_len);
 	}
+
+	if (x->coaddr)
+		RTA_PUT(skb, XFRMA_COADDR, sizeof(*x->coaddr), x->coaddr);
+
 	nlh->nlmsg_len = skb->tail - b;
 out:
 	sp->this_idx++;
-- 
cgit v1.2.3


From 9afaca057980c02771f4657c455cc7592fcd7373 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 18:20:16 -0700
Subject: [XFRM] IPV6: Update outbound state timestamp for each sending.

With this patch transformation state is updated last used time
for each sending. Xtime is used for it like other state lifetime
expiration.
Mobile IPv6 enabled nodes will want to know traffic status of each
binding (e.g. judgement to request binding refresh by correspondent node,
or to keep home/care-of nonce alive by mobile node).
The last used timestamp is an important hint about it.
Based on MIPL2 kernel patch.

This patch was also written by: Henrik Petander <petander@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h    | 1 +
 include/net/xfrm.h      | 3 +++
 net/ipv6/xfrm6_output.c | 2 ++
 net/xfrm/xfrm_user.c    | 3 +++
 4 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index b53f799189af..1d8c1f22c12d 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -236,6 +236,7 @@ enum xfrm_attr_type_t {
 	XFRMA_ETIMER_THRESH,
 	XFRMA_SRCADDR,		/* xfrm_address_t */
 	XFRMA_COADDR,		/* xfrm_address_t */
+	XFRMA_LASTUSED,
 	__XFRMA_MAX
 
 #define XFRMA_MAX (__XFRMA_MAX - 1)
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 872a2a4022b2..248874ecf8df 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -167,6 +167,9 @@ struct xfrm_state
 	struct xfrm_lifetime_cur curlft;
 	struct timer_list	timer;
 
+	/* Last used time */
+	u64			lastused;
+
 	/* Reference to data common to all the instances of this
 	 * transformer. */
 	struct xfrm_type	*type;
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index b4628fbf8ff5..db58104e710b 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -75,6 +75,8 @@ static int xfrm6_output_one(struct sk_buff *skb)
 
 		x->curlft.bytes += skb->len;
 		x->curlft.packets++;
+		if (x->props.mode == XFRM_MODE_ROUTEOPTIMIZATION)
+			x->lastused = (u64)xtime.tv_sec;
 
 		spin_unlock_bh(&x->lock);
 
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 939808de9e20..f643063a1cbd 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -595,6 +595,9 @@ static int dump_one_state(struct xfrm_state *x, int count, void *ptr)
 	if (x->coaddr)
 		RTA_PUT(skb, XFRMA_COADDR, sizeof(*x->coaddr), x->coaddr);
 
+	if (x->lastused)
+		RTA_PUT(skb, XFRMA_LASTUSED, sizeof(x->lastused), &x->lastused);
+
 	nlh->nlmsg_len = skb->tail - b;
 out:
 	sp->this_idx++;
-- 
cgit v1.2.3


From e53820de0f81da1429048634cadc6ef5f50c2f8b Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 19:12:01 -0700
Subject: [XFRM] IPV6: Restrict bundle reusing

For outbound transformation, bundle is checked whether it is
suitable for current flow to be reused or not. In such IPv6 case
as below, transformation may apply incorrect bundle for the flow instead
of creating another bundle:

- The policy selector has destination prefix length < 128
  (Two or more addresses can be matched it)
- Its bundle holds dst entry of default route whose prefix length < 128
  (Previous traffic was used such route as next hop)
- The policy and the bundle were used a transport mode state and
  this time flow address is not matched the bundled state.

This issue is found by Mobile IPv6 usage to protect mobility signaling
by IPsec, but it is not a Mobile IPv6 specific.
This patch adds strict check to xfrm_bundle_ok() for each
state mode and address when prefix length is less than 128.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      | 19 ++++++++++++++++++-
 net/ipv4/xfrm4_policy.c |  2 +-
 net/ipv6/xfrm6_policy.c |  4 +++-
 net/xfrm/xfrm_policy.c  |  8 ++++++--
 4 files changed, 28 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 248874ecf8df..7f1630630dcf 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -869,6 +869,23 @@ xfrm_state_addr_check(struct xfrm_state *x,
 	return 0;
 }
 
+static __inline__ int
+xfrm_state_addr_flow_check(struct xfrm_state *x, struct flowi *fl,
+			   unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		return __xfrm4_state_addr_check(x,
+						(xfrm_address_t *)&fl->fl4_dst,
+						(xfrm_address_t *)&fl->fl4_src);
+	case AF_INET6:
+		return __xfrm6_state_addr_check(x,
+						(xfrm_address_t *)&fl->fl6_dst,
+						(xfrm_address_t *)&fl->fl6_src);
+	}
+	return 0;
+}
+
 static inline int xfrm_state_kern(struct xfrm_state *x)
 {
 	return atomic_read(&x->tunnel_users);
@@ -1014,7 +1031,7 @@ extern void xfrm_policy_flush(void);
 extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol);
 extern int xfrm_flush_bundles(void);
 extern void xfrm_flush_all_bundles(void);
-extern int xfrm_bundle_ok(struct xfrm_dst *xdst, struct flowi *fl, int family);
+extern int xfrm_bundle_ok(struct xfrm_dst *xdst, struct flowi *fl, int family, int strict);
 extern void xfrm_init_pmtu(struct dst_entry *dst);
 
 extern wait_queue_head_t km_waitq;
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index e517981ceadd..42d8ded0f96a 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -33,7 +33,7 @@ __xfrm4_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 		    xdst->u.rt.fl.fl4_dst == fl->fl4_dst &&
 	    	    xdst->u.rt.fl.fl4_src == fl->fl4_src &&
 	    	    xdst->u.rt.fl.fl4_tos == fl->fl4_tos &&
-		    xfrm_bundle_ok(xdst, fl, AF_INET)) {
+		    xfrm_bundle_ok(xdst, fl, AF_INET, 0)) {
 			dst_clone(dst);
 			break;
 		}
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index a3f68c8b737e..729b4748d6d3 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -50,7 +50,9 @@ __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 				 xdst->u.rt6.rt6i_src.plen);
 		if (ipv6_addr_equal(&xdst->u.rt6.rt6i_dst.addr, &fl_dst_prefix) &&
 		    ipv6_addr_equal(&xdst->u.rt6.rt6i_src.addr, &fl_src_prefix) &&
-		    xfrm_bundle_ok(xdst, fl, AF_INET6)) {
+		    xfrm_bundle_ok(xdst, fl, AF_INET6,
+				   (xdst->u.rt6.rt6i_dst.plen != 128 ||
+				    xdst->u.rt6.rt6i_src.plen != 128))) {
 			dst_clone(dst);
 			break;
 		}
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 56abb5c057d4..ad2a5cba1f5b 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1167,7 +1167,7 @@ static struct dst_entry *xfrm_dst_check(struct dst_entry *dst, u32 cookie)
 
 static int stale_bundle(struct dst_entry *dst)
 {
-	return !xfrm_bundle_ok((struct xfrm_dst *)dst, NULL, AF_UNSPEC);
+	return !xfrm_bundle_ok((struct xfrm_dst *)dst, NULL, AF_UNSPEC, 0);
 }
 
 void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev)
@@ -1282,7 +1282,7 @@ EXPORT_SYMBOL(xfrm_init_pmtu);
  * still valid.
  */
 
-int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family)
+int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family, int strict)
 {
 	struct dst_entry *dst = &first->u.dst;
 	struct xfrm_dst *last;
@@ -1304,6 +1304,10 @@ int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family)
 		if (dst->xfrm->km.state != XFRM_STATE_VALID)
 			return 0;
 
+		if (strict && fl && dst->xfrm->props.mode != XFRM_MODE_TUNNEL &&
+		    !xfrm_state_addr_flow_check(dst->xfrm, fl, family))
+			return 0;
+
 		mtu = dst_mtu(dst->child);
 		if (xdst->child_mtu_cached != mtu) {
 			last = xdst;
-- 
cgit v1.2.3


From 642ec62eee5bdc158e01029220c8a23c685778fb Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Wed, 23 Aug 2006 19:15:07 -0700
Subject: [IPV6] MIP6: Add routing header type 2 definition.

Add routing header type 2 definition for Mobile IPv6.
Based on MIPL2 kernel patch.

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/ipv6.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 02d14a3ff2af..d995662e94c4 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -29,6 +29,7 @@ struct in6_ifreq {
 
 #define IPV6_SRCRT_STRICT	0x01	/* this hop must be a neighbor	*/
 #define IPV6_SRCRT_TYPE_0	0	/* IPv6 type 0 Routing Header	*/
+#define IPV6_SRCRT_TYPE_2	2	/* IPv6 type 2 Routing Header	*/
 
 /*
  *	routing header
@@ -73,6 +74,18 @@ struct rt0_hdr {
 #define rt0_type		rt_hdr.type
 };
 
+/*
+ *	routing header type 2
+ */
+
+struct rt2_hdr {
+	struct ipv6_rt_hdr	rt_hdr;
+	__u32			reserved;
+	struct in6_addr		addr;
+
+#define rt2_type		rt_hdr.type
+};
+
 struct ipv6_auth_hdr {
 	__u8  nexthdr;
 	__u8  hdrlen;           /* This one is measured in 32 bit units! */
-- 
cgit v1.2.3


From 65d4ed92219b28875efb52de5700da8c3dfa83e1 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 19:16:22 -0700
Subject: [IPV6] MIP6: Add inbound interface of routing header type 2.

Add inbound interface of routing header type 2 for Mobile IPv6.
Based on MIPL2 kernel patch.

This patch was also written by: Ville Nuorvala <vnuorval@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h |  7 +++++
 net/ipv6/exthdrs.c     | 69 ++++++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 68 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 3d71251b3eca..5fc8627435eb 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -61,6 +61,13 @@ extern int			addrconf_set_dstaddr(void __user *arg);
 extern int			ipv6_chk_addr(struct in6_addr *addr,
 					      struct net_device *dev,
 					      int strict);
+/* XXX: this is a placeholder till addrconf supports */
+#ifdef CONFIG_IPV6_MIP6
+static inline int ipv6_chk_home_addr(struct in6_addr *addr)
+{
+	return 0;
+}
+#endif
 extern struct inet6_ifaddr *	ipv6_get_ifaddr(struct in6_addr *addr,
 						struct net_device *dev,
 						int strict);
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 05afa6b1912b..8d3a0e17314d 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -43,6 +43,9 @@
 #include <net/ndisc.h>
 #include <net/ip6_route.h>
 #include <net/addrconf.h>
+#ifdef CONFIG_IPV6_MIP6
+#include <net/xfrm.h>
+#endif
 
 #include <asm/uaccess.h>
 
@@ -219,7 +222,7 @@ static int ipv6_rthdr_rcv(struct sk_buff **skbp)
 {
 	struct sk_buff *skb = *skbp;
 	struct inet6_skb_parm *opt = IP6CB(skb);
-	struct in6_addr *addr;
+	struct in6_addr *addr = NULL;
 	struct in6_addr daddr;
 	int n, i;
 
@@ -244,6 +247,23 @@ static int ipv6_rthdr_rcv(struct sk_buff **skbp)
 
 looped_back:
 	if (hdr->segments_left == 0) {
+		switch (hdr->type) {
+#ifdef CONFIG_IPV6_MIP6
+		case IPV6_SRCRT_TYPE_2:
+			/* Silently discard type 2 header unless it was
+			 * processed by own
+			 */
+			if (!addr) {
+				IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS);
+				kfree_skb(skb);
+				return -1;
+			}
+			break;
+#endif
+		default:
+			break;
+		}
+
 		opt->lastopt = skb->h.raw - skb->nh.raw;
 		opt->srcrt = skb->h.raw - skb->nh.raw;
 		skb->h.raw += (hdr->hdrlen + 1) << 3;
@@ -253,17 +273,29 @@ looped_back:
 		return 1;
 	}
 
-	if (hdr->type != IPV6_SRCRT_TYPE_0) {
+	switch (hdr->type) {
+	case IPV6_SRCRT_TYPE_0:
+		if (hdr->hdrlen & 0x01) {
+			IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
+			icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->hdrlen) - skb->nh.raw);
+			return -1;
+		}
+		break;
+#ifdef CONFIG_IPV6_MIP6
+	case IPV6_SRCRT_TYPE_2:
+		/* Silently discard invalid RTH type 2 */
+		if (hdr->hdrlen != 2 || hdr->segments_left != 1) {
+			IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
+			kfree_skb(skb);
+			return -1;
+		}
+		break;
+#endif
+	default:
 		IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
 		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->type) - skb->nh.raw);
 		return -1;
 	}
-	
-	if (hdr->hdrlen & 0x01) {
-		IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
-		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->hdrlen) - skb->nh.raw);
-		return -1;
-	}
 
 	/*
 	 *	This is the routing header forwarding algorithm from
@@ -303,6 +335,27 @@ looped_back:
 	addr = rthdr->addr;
 	addr += i - 1;
 
+	switch (hdr->type) {
+#ifdef CONFIG_IPV6_MIP6
+	case IPV6_SRCRT_TYPE_2:
+		if (xfrm6_input_addr(skb, (xfrm_address_t *)addr,
+				     (xfrm_address_t *)&skb->nh.ipv6h->saddr,
+				     IPPROTO_ROUTING) < 0) {
+			IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS);
+			kfree_skb(skb);
+			return -1;
+		}
+		if (!ipv6_chk_home_addr(addr)) {
+			IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS);
+			kfree_skb(skb);
+			return -1;
+		}
+		break;
+#endif
+	default:
+		break;
+	}
+
 	if (ipv6_addr_is_multicast(addr)) {
 		IP6_INC_STATS_BH(IPSTATS_MIB_INADDRERRORS);
 		kfree_skb(skb);
-- 
cgit v1.2.3


From c61a404325093250b676f40ad8f4dd00f3bcab5f Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 19:18:35 -0700
Subject: [IPV6]: Find option offset by type.

This is a helper to search option offset from extension header which
can carry TLV option like destination options header.

Mobile IPv6 home address option will use it.

Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h |  2 ++
 net/ipv6/exthdrs.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index ece7e8a84ffd..c4ea12710576 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -506,6 +506,8 @@ extern int			ipv6_skip_exthdr(const struct sk_buff *, int start,
 
 extern int 			ipv6_ext_hdr(u8 nexthdr);
 
+extern int ipv6_find_tlv(struct sk_buff *skb, int offset, int type);
+
 extern struct ipv6_txoptions *	ipv6_invert_rthdr(struct sock *sk,
 						  struct ipv6_rt_hdr *hdr);
 
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 8d3a0e17314d..50ff49e518bc 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -49,6 +49,49 @@
 
 #include <asm/uaccess.h>
 
+int ipv6_find_tlv(struct sk_buff *skb, int offset, int type)
+{
+	int packet_len = skb->tail - skb->nh.raw;
+	struct ipv6_opt_hdr *hdr;
+	int len;
+
+	if (offset + 2 > packet_len)
+		goto bad;
+	hdr = (struct ipv6_opt_hdr*)(skb->nh.raw + offset);
+	len = ((hdr->hdrlen + 1) << 3);
+
+	if (offset + len > packet_len)
+		goto bad;
+
+	offset += 2;
+	len -= 2;
+
+	while (len > 0) {
+		int opttype = skb->nh.raw[offset];
+		int optlen;
+
+		if (opttype == type)
+			return offset;
+
+		switch (opttype) {
+		case IPV6_TLV_PAD0:
+			optlen = 1;
+			break;
+		default:
+			optlen = skb->nh.raw[offset + 1] + 2;
+			if (optlen > len)
+				goto bad;
+			break;
+		}
+		offset += optlen;
+		len -= optlen;
+	}
+	/* not_found */
+	return -1;
+ bad:
+	return -1;
+}
+
 /*
  *	Parsing tlv encoded headers.
  *
-- 
cgit v1.2.3


From a80ff03e05e4343d647780c116b02ec86078fd24 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 19:19:50 -0700
Subject: [IPV6]: Allow to replace skbuff by TLV parser.

In receiving Mobile IPv6 home address option which is a TLV carried by
destination options header, kernel will try to mangle source adderss
of packet. Think of cloned skbuff it is required to replace it by the
parser just like routing header case.

This is a framework to achieve that to allow TLV parser to replace
inbound skbuff pointer.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipv6.h   |  2 +-
 net/ipv6/exthdrs.c   | 29 +++++++++++++++++++----------
 net/ipv6/ip6_input.c |  2 +-
 3 files changed, 21 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index c4ea12710576..8e6ec6063f8c 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -229,7 +229,7 @@ extern int 			ip6_ra_control(struct sock *sk, int sel,
 					       void (*destructor)(struct sock *));
 
 
-extern int			ipv6_parse_hopopts(struct sk_buff *skb);
+extern int			ipv6_parse_hopopts(struct sk_buff **skbp);
 
 extern struct ipv6_txoptions *  ipv6_dup_options(struct sock *sk, struct ipv6_txoptions *opt);
 extern struct ipv6_txoptions *	ipv6_renew_options(struct sock *sk, struct ipv6_txoptions *opt,
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 50ff49e518bc..1cdd0f0b5d34 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -102,7 +102,7 @@ int ipv6_find_tlv(struct sk_buff *skb, int offset, int type)
 
 struct tlvtype_proc {
 	int	type;
-	int	(*func)(struct sk_buff *skb, int offset);
+	int	(*func)(struct sk_buff **skbp, int offset);
 };
 
 /*********************
@@ -111,8 +111,10 @@ struct tlvtype_proc {
 
 /* An unknown option is detected, decide what to do */
 
-static int ip6_tlvopt_unknown(struct sk_buff *skb, int optoff)
+static int ip6_tlvopt_unknown(struct sk_buff **skbp, int optoff)
 {
+	struct sk_buff *skb = *skbp;
+
 	switch ((skb->nh.raw[optoff] & 0xC0) >> 6) {
 	case 0: /* ignore */
 		return 1;
@@ -137,8 +139,9 @@ static int ip6_tlvopt_unknown(struct sk_buff *skb, int optoff)
 
 /* Parse tlv encoded option header (hop-by-hop or destination) */
 
-static int ip6_parse_tlv(struct tlvtype_proc *procs, struct sk_buff *skb)
+static int ip6_parse_tlv(struct tlvtype_proc *procs, struct sk_buff **skbp)
 {
+	struct sk_buff *skb = *skbp;
 	struct tlvtype_proc *curr;
 	int off = skb->h.raw - skb->nh.raw;
 	int len = ((skb->h.raw[1]+1)<<3);
@@ -168,13 +171,13 @@ static int ip6_parse_tlv(struct tlvtype_proc *procs, struct sk_buff *skb)
 					/* type specific length/alignment 
 					   checks will be performed in the 
 					   func(). */
-					if (curr->func(skb, off) == 0)
+					if (curr->func(skbp, off) == 0)
 						return 0;
 					break;
 				}
 			}
 			if (curr->type < 0) {
-				if (ip6_tlvopt_unknown(skb, off) == 0)
+				if (ip6_tlvopt_unknown(skbp, off) == 0)
 					return 0;
 			}
 			break;
@@ -213,7 +216,8 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp)
 	opt->lastopt = skb->h.raw - skb->nh.raw;
 	opt->dst1 = skb->h.raw - skb->nh.raw;
 
-	if (ip6_parse_tlv(tlvprocdestopt_lst, skb)) {
+	if (ip6_parse_tlv(tlvprocdestopt_lst, skbp)) {
+		skb = *skbp;
 		skb->h.raw += ((skb->h.raw[1]+1)<<3);
 		opt->nhoff = opt->dst1;
 		return 1;
@@ -517,8 +521,10 @@ EXPORT_SYMBOL_GPL(ipv6_invert_rthdr);
 
 /* Router Alert as of RFC 2711 */
 
-static int ipv6_hop_ra(struct sk_buff *skb, int optoff)
+static int ipv6_hop_ra(struct sk_buff **skbp, int optoff)
 {
+	struct sk_buff *skb = *skbp;
+
 	if (skb->nh.raw[optoff+1] == 2) {
 		IP6CB(skb)->ra = optoff;
 		return 1;
@@ -531,8 +537,9 @@ static int ipv6_hop_ra(struct sk_buff *skb, int optoff)
 
 /* Jumbo payload */
 
-static int ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
+static int ipv6_hop_jumbo(struct sk_buff **skbp, int optoff)
 {
+	struct sk_buff *skb = *skbp;
 	u32 pkt_len;
 
 	if (skb->nh.raw[optoff+1] != 4 || (optoff&3) != 2) {
@@ -581,8 +588,9 @@ static struct tlvtype_proc tlvprochopopt_lst[] = {
 	{ -1, }
 };
 
-int ipv6_parse_hopopts(struct sk_buff *skb)
+int ipv6_parse_hopopts(struct sk_buff **skbp)
 {
+	struct sk_buff *skb = *skbp;
 	struct inet6_skb_parm *opt = IP6CB(skb);
 
 	/*
@@ -598,7 +606,8 @@ int ipv6_parse_hopopts(struct sk_buff *skb)
 	}
 
 	opt->hop = sizeof(struct ipv6hdr);
-	if (ip6_parse_tlv(tlvprochopopt_lst, skb)) {
+	if (ip6_parse_tlv(tlvprochopopt_lst, skbp)) {
+		skb = *skbp;
 		skb->h.raw += (skb->h.raw[1]+1)<<3;
 		opt->nhoff = sizeof(struct ipv6hdr);
 		return 1;
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 25c2a9e03895..6b8e6d76a58b 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -111,7 +111,7 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 	}
 
 	if (hdr->nexthdr == NEXTHDR_HOP) {
-		if (ipv6_parse_hopopts(skb) < 0) {
+		if (ipv6_parse_hopopts(&skb) < 0) {
 			IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS);
 			return 0;
 		}
-- 
cgit v1.2.3


From 842426e719f86cd5709617208efae93ff1a1e2d8 Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Wed, 23 Aug 2006 19:21:34 -0700
Subject: [IPV6] MIP6: Add home address option definition.

Add home address option definition for Mobile IPv6.
Based on MIPL2 kernel patch.

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in6.h  |  1 +
 include/linux/ipv6.h | 10 ++++++++++
 2 files changed, 11 insertions(+)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index 304aaedea305..086ec2ac8c5f 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -142,6 +142,7 @@ struct in6_flowlabel_req
 #define IPV6_TLV_PADN		1
 #define IPV6_TLV_ROUTERALERT	5
 #define IPV6_TLV_JUMBO		194
+#define IPV6_TLV_HAO		201	/* home address option */
 
 /*
  *	IPV6 socket options
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index d995662e94c4..5bf4406e26d4 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -86,6 +86,16 @@ struct rt2_hdr {
 #define rt2_type		rt_hdr.type
 };
 
+/*
+ *	home address option in destination options header
+ */
+
+struct ipv6_destopt_hao {
+	__u8			type;
+	__u8			length;
+	struct in6_addr		addr;
+} __attribute__ ((__packed__));
+
 struct ipv6_auth_hdr {
 	__u8  nexthdr;
 	__u8  hdrlen;           /* This one is measured in 32 bit units! */
-- 
cgit v1.2.3


From a831f5bbc89a9978795504be9e1ff412043f8f77 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 19:24:48 -0700
Subject: [IPV6] MIP6: Add inbound interface of home address option.

Add inbound function of home address option by registering it to TLV
table for destination options header.

Based on MIPL2 kernel patch.

This patch was also written by: Ville Nuorvala <vnuorval@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h |  3 ++
 net/ipv6/exthdrs.c   | 84 +++++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 86 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 5bf4406e26d4..db3b2ba0f4f8 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -226,6 +226,9 @@ struct inet6_skb_parm {
 	__u16			dst0;
 	__u16			srcrt;
 	__u16			dst1;
+#ifdef CONFIG_IPV6_MIP6
+	__u16			dsthao;
+#endif
 	__u16			lastopt;
 	__u32			nhoff;
 	__u16			flags;
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 1cdd0f0b5d34..6a6466bb5f26 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -196,8 +196,80 @@ bad:
   Destination options header.
  *****************************/
 
+#ifdef CONFIG_IPV6_MIP6
+static int ipv6_dest_hao(struct sk_buff **skbp, int optoff)
+{
+	struct sk_buff *skb = *skbp;
+	struct ipv6_destopt_hao *hao;
+	struct inet6_skb_parm *opt = IP6CB(skb);
+	struct ipv6hdr *ipv6h = (struct ipv6hdr *)skb->nh.raw;
+	struct in6_addr tmp_addr;
+	int ret;
+
+	if (opt->dsthao) {
+		LIMIT_NETDEBUG(KERN_DEBUG "hao duplicated\n");
+		goto discard;
+	}
+	opt->dsthao = opt->dst1;
+	opt->dst1 = 0;
+
+	hao = (struct ipv6_destopt_hao *)(skb->nh.raw + optoff);
+
+	if (hao->length != 16) {
+		LIMIT_NETDEBUG(
+			KERN_DEBUG "hao invalid option length = %d\n", hao->length);
+		goto discard;
+	}
+
+	if (!(ipv6_addr_type(&hao->addr) & IPV6_ADDR_UNICAST)) {
+		LIMIT_NETDEBUG(
+			KERN_DEBUG "hao is not an unicast addr: " NIP6_FMT "\n", NIP6(hao->addr));
+		goto discard;
+	}
+
+	ret = xfrm6_input_addr(skb, (xfrm_address_t *)&ipv6h->daddr,
+			       (xfrm_address_t *)&hao->addr, IPPROTO_DSTOPTS);
+	if (unlikely(ret < 0))
+		goto discard;
+
+	if (skb_cloned(skb)) {
+		struct sk_buff *skb2 = skb_copy(skb, GFP_ATOMIC);
+		if (skb2 == NULL)
+			goto discard;
+
+		kfree_skb(skb);
+
+		/* update all variable using below by copied skbuff */
+		*skbp = skb = skb2;
+		hao = (struct ipv6_destopt_hao *)(skb2->nh.raw + optoff);
+		ipv6h = (struct ipv6hdr *)skb2->nh.raw;
+	}
+
+	if (skb->ip_summed == CHECKSUM_COMPLETE)
+		skb->ip_summed = CHECKSUM_NONE;
+
+	ipv6_addr_copy(&tmp_addr, &ipv6h->saddr);
+	ipv6_addr_copy(&ipv6h->saddr, &hao->addr);
+	ipv6_addr_copy(&hao->addr, &tmp_addr);
+
+	if (skb->tstamp.off_sec == 0)
+		__net_timestamp(skb);
+
+	return 1;
+
+ discard:
+	kfree_skb(skb);
+	return 0;
+}
+#endif
+
 static struct tlvtype_proc tlvprocdestopt_lst[] = {
-	/* No destination options are defined now */
+#ifdef CONFIG_IPV6_MIP6
+	{
+		.type	= IPV6_TLV_HAO,
+		.func	= ipv6_dest_hao,
+	},
+#endif
 	{-1,			NULL}
 };
 
@@ -205,6 +277,9 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp)
 {
 	struct sk_buff *skb = *skbp;
 	struct inet6_skb_parm *opt = IP6CB(skb);
+#ifdef CONFIG_IPV6_MIP6
+	__u16 dstbuf;
+#endif
 
 	if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+8) ||
 	    !pskb_may_pull(skb, (skb->h.raw-skb->data)+((skb->h.raw[1]+1)<<3))) {
@@ -215,11 +290,18 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp)
 
 	opt->lastopt = skb->h.raw - skb->nh.raw;
 	opt->dst1 = skb->h.raw - skb->nh.raw;
+#ifdef CONFIG_IPV6_MIP6
+	dstbuf = opt->dst1;
+#endif
 
 	if (ip6_parse_tlv(tlvprocdestopt_lst, skbp)) {
 		skb = *skbp;
 		skb->h.raw += ((skb->h.raw[1]+1)<<3);
+#ifdef CONFIG_IPV6_MIP6
+		opt->nhoff = dstbuf;
+#else
 		opt->nhoff = opt->dst1;
+#endif
 		return 1;
 	}
 
-- 
cgit v1.2.3


From 8dd7368dd97def967bbb3aec67b882e8dfd1a528 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Wed, 23 Aug 2006 19:25:55 -0700
Subject: [IPV6]: Put dsthao after flags in order to pack inet6_skb_parm
 better.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index db3b2ba0f4f8..1d6d3ccc9413 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -226,12 +226,12 @@ struct inet6_skb_parm {
 	__u16			dst0;
 	__u16			srcrt;
 	__u16			dst1;
-#ifdef CONFIG_IPV6_MIP6
-	__u16			dsthao;
-#endif
 	__u16			lastopt;
 	__u32			nhoff;
 	__u16			flags;
+#ifdef CONFIG_IPV6_MIP6
+	__u16			dsthao;
+#endif
 
 #define IP6SKB_XFRM_TRANSFORMED	1
 };
-- 
cgit v1.2.3


From 2c8d7ca0f76103855ad1f2a930e05683b64a00eb Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Wed, 23 Aug 2006 20:31:11 -0700
Subject: [IPV6] MIP6: Add routing header type 2 transformation.

Add routing header type 2 transformation for Mobile IPv6.
Based on MIPL2 kernel patch.

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mip6.h  |  31 +++++++++
 net/ipv6/Makefile   |   2 +
 net/ipv6/af_inet6.c |   9 +++
 net/ipv6/mip6.c     | 181 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 223 insertions(+)
 create mode 100644 include/net/mip6.h
 create mode 100644 net/ipv6/mip6.c

(limited to 'include')

diff --git a/include/net/mip6.h b/include/net/mip6.h
new file mode 100644
index 000000000000..644b8b673048
--- /dev/null
+++ b/include/net/mip6.h
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C)2003-2006 Helsinki University of Technology
+ * Copyright (C)2003-2006 USAGI/WIDE Project
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+/*
+ * Authors:
+ *	Noriaki TAKAMIYA @USAGI
+ *	Masahide NAKAMURA @USAGI
+ *	YOSHIFUJI Hideaki @USAGI
+ */
+#ifndef _NET_MIP6_H
+#define _NET_MIP6_H
+
+extern int mip6_init(void);
+extern void mip6_fini(void);
+
+#endif
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 87e912e31922..0213c6612b58 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -14,6 +14,8 @@ ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \
 	xfrm6_output.o
 ipv6-$(CONFIG_NETFILTER) += netfilter.o
 ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o
+ipv6-$(CONFIG_IPV6_MIP6) += mip6.o
+
 ipv6-objs += $(ipv6-y)
 
 obj-$(CONFIG_INET6_AH) += ah6.o
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 57ee5ddea96f..fc9c8a99bea6 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -59,6 +59,9 @@
 #ifdef CONFIG_IPV6_TUNNEL
 #include <net/ip6_tunnel.h>
 #endif
+#ifdef CONFIG_IPV6_MIP6
+#include <net/mip6.h>
+#endif
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -857,6 +860,9 @@ static int __init inet6_init(void)
 	ipv6_frag_init();
 	ipv6_nodata_init();
 	ipv6_destopt_init();
+#ifdef CONFIG_IPV6_MIP6
+	mip6_init();
+#endif
 
 	/* Init v6 transport protocols. */
 	udpv6_init();
@@ -919,6 +925,9 @@ static void __exit inet6_exit(void)
  	udp6_proc_exit();
  	tcp6_proc_exit();
  	raw6_proc_exit();
+#endif
+#ifdef CONFIG_IPV6_MIP6
+	mip6_fini();
 #endif
 	/* Cleanup code parts. */
 	sit_cleanup();
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
new file mode 100644
index 000000000000..63e548b6f81e
--- /dev/null
+++ b/net/ipv6/mip6.c
@@ -0,0 +1,181 @@
+/*
+ * Copyright (C)2003-2006 Helsinki University of Technology
+ * Copyright (C)2003-2006 USAGI/WIDE Project
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+/*
+ * Authors:
+ *	Noriaki TAKAMIYA @USAGI
+ *	Masahide NAKAMURA @USAGI
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <net/xfrm.h>
+#include <net/mip6.h>
+
+static xfrm_address_t *mip6_xfrm_addr(struct xfrm_state *x, xfrm_address_t *addr)
+{
+	return x->coaddr;
+}
+
+static int mip6_rthdr_input(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct rt2_hdr *rt2 = (struct rt2_hdr *)skb->data;
+
+	if (!ipv6_addr_equal(&rt2->addr, (struct in6_addr *)x->coaddr) &&
+	    !ipv6_addr_any((struct in6_addr *)x->coaddr))
+		return -ENOENT;
+
+	return rt2->rt_hdr.nexthdr;
+}
+
+/* Routing Header type 2 is inserted.
+ * IP Header's dst address is replaced with Routing Header's Home Address.
+ */
+static int mip6_rthdr_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct ipv6hdr *iph;
+	struct rt2_hdr *rt2;
+	u8 nexthdr;
+
+	iph = (struct ipv6hdr *)skb->data;
+	iph->payload_len = htons(skb->len - sizeof(*iph));
+
+	nexthdr = *skb->nh.raw;
+	*skb->nh.raw = IPPROTO_ROUTING;
+
+	rt2 = (struct rt2_hdr *)skb->h.raw;
+	rt2->rt_hdr.nexthdr = nexthdr;
+	rt2->rt_hdr.hdrlen = (x->props.header_len >> 3) - 1;
+	rt2->rt_hdr.type = IPV6_SRCRT_TYPE_2;
+	rt2->rt_hdr.segments_left = 1;
+	memset(&rt2->reserved, 0, sizeof(rt2->reserved));
+
+	BUG_TRAP(rt2->rt_hdr.hdrlen == 2);
+
+	memcpy(&rt2->addr, &iph->daddr, sizeof(rt2->addr));
+	memcpy(&iph->daddr, x->coaddr, sizeof(iph->daddr));
+
+	return 0;
+}
+
+static int mip6_rthdr_offset(struct xfrm_state *x, struct sk_buff *skb,
+			     u8 **nexthdr)
+{
+	u16 offset = sizeof(struct ipv6hdr);
+	struct ipv6_opt_hdr *exthdr = (struct ipv6_opt_hdr*)(skb->nh.ipv6h + 1);
+	unsigned int packet_len = skb->tail - skb->nh.raw;
+	int found_rhdr = 0;
+
+	*nexthdr = &skb->nh.ipv6h->nexthdr;
+
+	while (offset + 1 <= packet_len) {
+
+		switch (**nexthdr) {
+		case NEXTHDR_HOP:
+			break;
+		case NEXTHDR_ROUTING:
+			if (offset + 3 <= packet_len) {
+				struct ipv6_rt_hdr *rt;
+				rt = (struct ipv6_rt_hdr *)(skb->nh.raw + offset);
+				if (rt->type != 0)
+					return offset;
+			}
+			found_rhdr = 1;
+			break;
+		case NEXTHDR_DEST:
+			if (ipv6_find_tlv(skb, offset, IPV6_TLV_HAO) >= 0)
+				return offset;
+
+			if (found_rhdr)
+				return offset;
+
+			break;
+		default:
+			return offset;
+		}
+
+		offset += ipv6_optlen(exthdr);
+		*nexthdr = &exthdr->nexthdr;
+		exthdr = (struct ipv6_opt_hdr*)(skb->nh.raw + offset);
+	}
+
+	return offset;
+}
+
+static int mip6_rthdr_init_state(struct xfrm_state *x)
+{
+	if (x->id.spi) {
+		printk(KERN_INFO "%s: spi is not 0: %u\n", __FUNCTION__,
+		       x->id.spi);
+		return -EINVAL;
+	}
+	if (x->props.mode != XFRM_MODE_ROUTEOPTIMIZATION) {
+		printk(KERN_INFO "%s: state's mode is not %u: %u\n",
+		       __FUNCTION__, XFRM_MODE_ROUTEOPTIMIZATION, x->props.mode);
+		return -EINVAL;
+	}
+
+	x->props.header_len = sizeof(struct rt2_hdr);
+
+	return 0;
+}
+
+/*
+ * Do nothing about destroying since it has no specific operation for routing
+ * header type 2 unlike IPsec protocols.
+ */
+static void mip6_rthdr_destroy(struct xfrm_state *x)
+{
+}
+
+static struct xfrm_type mip6_rthdr_type =
+{
+	.description	= "MIP6RT",
+	.owner		= THIS_MODULE,
+	.proto	     	= IPPROTO_ROUTING,
+	.flags		= XFRM_TYPE_NON_FRAGMENT,
+	.init_state	= mip6_rthdr_init_state,
+	.destructor	= mip6_rthdr_destroy,
+	.input		= mip6_rthdr_input,
+	.output		= mip6_rthdr_output,
+	.hdr_offset	= mip6_rthdr_offset,
+	.remote_addr	= mip6_xfrm_addr,
+};
+
+int __init mip6_init(void)
+{
+	printk(KERN_INFO "Mobile IPv6\n");
+
+	if (xfrm_register_type(&mip6_rthdr_type, AF_INET6) < 0) {
+		printk(KERN_INFO "%s: can't add xfrm type(rthdr)\n", __FUNCTION__);
+		goto mip6_rthdr_xfrm_fail;
+	}
+	return 0;
+
+ mip6_rthdr_xfrm_fail:
+	return -EAGAIN;
+}
+
+void __exit mip6_fini(void)
+{
+	if (xfrm_unregister_type(&mip6_rthdr_type, AF_INET6) < 0)
+		printk(KERN_INFO "%s: can't remove xfrm type(rthdr)\n", __FUNCTION__);
+}
-- 
cgit v1.2.3


From 3d126890dd67beffec27c1b6f51c040fc8d0b526 Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Wed, 23 Aug 2006 20:32:34 -0700
Subject: [IPV6] MIP6: Add destination options header transformation.

Add destination options header transformation for Mobile IPv6.
Based on MIPL2 kernel patch.

This patch was also written by: Ville Nuorvala <vnuorval@tcs.hut.fi>

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mip6.h |   3 +
 net/ipv6/mip6.c    | 167 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 170 insertions(+)

(limited to 'include')

diff --git a/include/net/mip6.h b/include/net/mip6.h
index 644b8b673048..42b65bace122 100644
--- a/include/net/mip6.h
+++ b/include/net/mip6.h
@@ -25,6 +25,9 @@
 #ifndef _NET_MIP6_H
 #define _NET_MIP6_H
 
+#define MIP6_OPT_PAD_1	0
+#define MIP6_OPT_PAD_N	1
+
 extern int mip6_init(void);
 extern void mip6_fini(void);
 
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
index 63e548b6f81e..a8adf891fe0e 100644
--- a/net/ipv6/mip6.c
+++ b/net/ipv6/mip6.c
@@ -35,6 +35,165 @@ static xfrm_address_t *mip6_xfrm_addr(struct xfrm_state *x, xfrm_address_t *addr
 	return x->coaddr;
 }
 
+static inline unsigned int calc_padlen(unsigned int len, unsigned int n)
+{
+	return (n - len + 16) & 0x7;
+}
+
+static inline void *mip6_padn(__u8 *data, __u8 padlen)
+{
+	if (!data)
+		return NULL;
+	if (padlen == 1) {
+		data[0] = MIP6_OPT_PAD_1;
+	} else if (padlen > 1) {
+		data[0] = MIP6_OPT_PAD_N;
+		data[1] = padlen - 2;
+		if (padlen > 2)
+			memset(data+2, 0, data[1]);
+	}
+	return data + padlen;
+}
+
+static int mip6_destopt_input(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct ipv6hdr *iph = skb->nh.ipv6h;
+	struct ipv6_destopt_hdr *destopt = (struct ipv6_destopt_hdr *)skb->data;
+
+	if (!ipv6_addr_equal(&iph->saddr, (struct in6_addr *)x->coaddr) &&
+	    !ipv6_addr_any((struct in6_addr *)x->coaddr))
+		return -ENOENT;
+
+	return destopt->nexthdr;
+}
+
+/* Destination Option Header is inserted.
+ * IP Header's src address is replaced with Home Address Option in
+ * Destination Option Header.
+ */
+static int mip6_destopt_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct ipv6hdr *iph;
+	struct ipv6_destopt_hdr *dstopt;
+	struct ipv6_destopt_hao *hao;
+	u8 nexthdr;
+	int len;
+
+	iph = (struct ipv6hdr *)skb->data;
+	iph->payload_len = htons(skb->len - sizeof(*iph));
+
+	nexthdr = *skb->nh.raw;
+	*skb->nh.raw = IPPROTO_DSTOPTS;
+
+	dstopt = (struct ipv6_destopt_hdr *)skb->h.raw;
+	dstopt->nexthdr = nexthdr;
+
+	hao = mip6_padn((char *)(dstopt + 1),
+			calc_padlen(sizeof(*dstopt), 6));
+
+	hao->type = IPV6_TLV_HAO;
+	hao->length = sizeof(*hao) - 2;
+	BUG_TRAP(hao->length == 16);
+
+	len = ((char *)hao - (char *)dstopt) + sizeof(*hao);
+
+	memcpy(&hao->addr, &iph->saddr, sizeof(hao->addr));
+	memcpy(&iph->saddr, x->coaddr, sizeof(iph->saddr));
+
+	BUG_TRAP(len == x->props.header_len);
+	dstopt->hdrlen = (x->props.header_len >> 3) - 1;
+
+	return 0;
+}
+
+static int mip6_destopt_offset(struct xfrm_state *x, struct sk_buff *skb,
+			       u8 **nexthdr)
+{
+	u16 offset = sizeof(struct ipv6hdr);
+	struct ipv6_opt_hdr *exthdr = (struct ipv6_opt_hdr*)(skb->nh.ipv6h + 1);
+	unsigned int packet_len = skb->tail - skb->nh.raw;
+	int found_rhdr = 0;
+
+	*nexthdr = &skb->nh.ipv6h->nexthdr;
+
+	while (offset + 1 <= packet_len) {
+
+		switch (**nexthdr) {
+		case NEXTHDR_HOP:
+			break;
+		case NEXTHDR_ROUTING:
+			found_rhdr = 1;
+			break;
+		case NEXTHDR_DEST:
+			/*
+			 * HAO MUST NOT appear more than once.
+			 * XXX: It is better to try to find by the end of
+			 * XXX: packet if HAO exists.
+			 */
+			if (ipv6_find_tlv(skb, offset, IPV6_TLV_HAO) >= 0) {
+				LIMIT_NETDEBUG(KERN_WARNING "mip6: hao exists already, override\n");
+				return offset;
+			}
+
+			if (found_rhdr)
+				return offset;
+
+			break;
+		default:
+			return offset;
+		}
+
+		offset += ipv6_optlen(exthdr);
+		*nexthdr = &exthdr->nexthdr;
+		exthdr = (struct ipv6_opt_hdr*)(skb->nh.raw + offset);
+	}
+
+	return offset;
+}
+
+static int mip6_destopt_init_state(struct xfrm_state *x)
+{
+	if (x->id.spi) {
+		printk(KERN_INFO "%s: spi is not 0: %u\n", __FUNCTION__,
+		       x->id.spi);
+		return -EINVAL;
+	}
+	if (x->props.mode != XFRM_MODE_ROUTEOPTIMIZATION) {
+		printk(KERN_INFO "%s: state's mode is not %u: %u\n",
+		       __FUNCTION__, XFRM_MODE_ROUTEOPTIMIZATION, x->props.mode);
+		return -EINVAL;
+	}
+
+	x->props.header_len = sizeof(struct ipv6_destopt_hdr) +
+		calc_padlen(sizeof(struct ipv6_destopt_hdr), 6) +
+		sizeof(struct ipv6_destopt_hao);
+	BUG_TRAP(x->props.header_len == 24);
+
+	return 0;
+}
+
+/*
+ * Do nothing about destroying since it has no specific operation for
+ * destination options header unlike IPsec protocols.
+ */
+static void mip6_destopt_destroy(struct xfrm_state *x)
+{
+}
+
+static struct xfrm_type mip6_destopt_type =
+{
+	.description	= "MIP6DESTOPT",
+	.owner		= THIS_MODULE,
+	.proto	     	= IPPROTO_DSTOPTS,
+	.flags		= XFRM_TYPE_NON_FRAGMENT,
+	.init_state	= mip6_destopt_init_state,
+	.destructor	= mip6_destopt_destroy,
+	.input		= mip6_destopt_input,
+	.output		= mip6_destopt_output,
+	.hdr_offset	= mip6_destopt_offset,
+	.local_addr	= mip6_xfrm_addr,
+};
+
 static int mip6_rthdr_input(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct rt2_hdr *rt2 = (struct rt2_hdr *)skb->data;
@@ -164,6 +323,10 @@ int __init mip6_init(void)
 {
 	printk(KERN_INFO "Mobile IPv6\n");
 
+	if (xfrm_register_type(&mip6_destopt_type, AF_INET6) < 0) {
+		printk(KERN_INFO "%s: can't add xfrm type(destopt)\n", __FUNCTION__);
+		goto mip6_destopt_xfrm_fail;
+	}
 	if (xfrm_register_type(&mip6_rthdr_type, AF_INET6) < 0) {
 		printk(KERN_INFO "%s: can't add xfrm type(rthdr)\n", __FUNCTION__);
 		goto mip6_rthdr_xfrm_fail;
@@ -171,6 +334,8 @@ int __init mip6_init(void)
 	return 0;
 
  mip6_rthdr_xfrm_fail:
+	xfrm_unregister_type(&mip6_destopt_type, AF_INET6);
+ mip6_destopt_xfrm_fail:
 	return -EAGAIN;
 }
 
@@ -178,4 +343,6 @@ void __exit mip6_fini(void)
 {
 	if (xfrm_unregister_type(&mip6_rthdr_type, AF_INET6) < 0)
 		printk(KERN_INFO "%s: can't remove xfrm type(rthdr)\n", __FUNCTION__);
+	if (xfrm_unregister_type(&mip6_destopt_type, AF_INET6) < 0)
+		printk(KERN_INFO "%s: can't remove xfrm type(destopt)\n", __FUNCTION__);
 }
-- 
cgit v1.2.3


From 2b741653b6c824fe7520ee92b6795f11c5f24b24 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 20:34:26 -0700
Subject: [IPV6] MIP6: Add Mobility header definition.

Add Mobility header definition for Mobile IPv6.
Based on MIPL2 kernel patch.

This patch was also written by: Antti Tuominen <anttit@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in6.h |  1 +
 include/net/flow.h  |  9 +++++++++
 include/net/ipv6.h  |  1 +
 include/net/mip6.h  | 23 +++++++++++++++++++++++
 4 files changed, 34 insertions(+)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index 086ec2ac8c5f..d776829b443f 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -134,6 +134,7 @@ struct in6_flowlabel_req
 #define IPPROTO_ICMPV6		58	/* ICMPv6			*/
 #define IPPROTO_NONE		59	/* IPv6 no next header		*/
 #define IPPROTO_DSTOPTS		60	/* IPv6 destination options	*/
+#define IPPROTO_MH		135	/* IPv6 mobility header		*/
 
 /*
  *	IPv6 TLV options.
diff --git a/include/net/flow.h b/include/net/flow.h
index 21d988b2058a..e0522914316e 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -72,12 +72,21 @@ struct flowi {
 		} dnports;
 
 		__u32		spi;
+
+#ifdef CONFIG_IPV6_MIP6
+		struct {
+			__u8	type;
+		} mht;
+#endif
 	} uli_u;
 #define fl_ip_sport	uli_u.ports.sport
 #define fl_ip_dport	uli_u.ports.dport
 #define fl_icmp_type	uli_u.icmpt.type
 #define fl_icmp_code	uli_u.icmpt.code
 #define fl_ipsec_spi	uli_u.spi
+#ifdef CONFIG_IPV6_MIP6
+#define fl_mh_type	uli_u.mht.type
+#endif
 	__u32           secid;	/* used by xfrm; see secid.txt */
 } __attribute__((__aligned__(BITS_PER_LONG/8)));
 
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 8e6ec6063f8c..72bf47b2a4e0 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -40,6 +40,7 @@
 #define NEXTHDR_ICMP		58	/* ICMP for IPv6. */
 #define NEXTHDR_NONE		59	/* No next header */
 #define NEXTHDR_DEST		60	/* Destination options header. */
+#define NEXTHDR_MOBILITY	135	/* Mobility header. */
 
 #define NEXTHDR_MAX		255
 
diff --git a/include/net/mip6.h b/include/net/mip6.h
index 42b65bace122..fd43178faace 100644
--- a/include/net/mip6.h
+++ b/include/net/mip6.h
@@ -28,6 +28,29 @@
 #define MIP6_OPT_PAD_1	0
 #define MIP6_OPT_PAD_N	1
 
+/*
+ * Mobility Header
+ */
+struct ip6_mh {
+	__u8	ip6mh_proto;
+	__u8	ip6mh_hdrlen;
+	__u8	ip6mh_type;
+	__u8	ip6mh_reserved;
+	__u16	ip6mh_cksum;
+	/* Followed by type specific messages */
+	__u8	data[0];
+} __attribute__ ((__packed__));
+
+#define IP6_MH_TYPE_BRR		0   /* Binding Refresh Request */
+#define IP6_MH_TYPE_HOTI	1   /* HOTI Message   */
+#define IP6_MH_TYPE_COTI	2   /* COTI Message  */
+#define IP6_MH_TYPE_HOT		3   /* HOT Message   */
+#define IP6_MH_TYPE_COT		4   /* COT Message  */
+#define IP6_MH_TYPE_BU		5   /* Binding Update */
+#define IP6_MH_TYPE_BACK	6   /* Binding ACK */
+#define IP6_MH_TYPE_BERROR	7   /* Binding Error */
+#define IP6_MH_TYPE_MAX		IP6_MH_TYPE_BERROR
+
 extern int mip6_init(void);
 extern void mip6_fini(void);
 
-- 
cgit v1.2.3


From 7be96f7628469e56f91d51f13b03e9bcff113c7f Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 20:35:31 -0700
Subject: [IPV6] MIP6: Add receiving mobility header functions through raw
 socket.

Like ICMPv6, mobility header is handled through raw socket.
In inbound case, check only whether ICMPv6 error should be sent as a reply
or not by kernel.
Based on MIPL2 kernel patch.

This patch was also written by: Ville Nuorvala <vnuorval@tcs.hut.fi>
This patch was also written by: Antti Tuominen <anttit@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/mip6.h |  4 +++
 net/ipv6/mip6.c    | 83 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/raw.c     | 29 ++++++++++++++++++-
 3 files changed, 115 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/mip6.h b/include/net/mip6.h
index fd43178faace..68263c6d9996 100644
--- a/include/net/mip6.h
+++ b/include/net/mip6.h
@@ -25,6 +25,9 @@
 #ifndef _NET_MIP6_H
 #define _NET_MIP6_H
 
+#include <linux/skbuff.h>
+#include <net/sock.h>
+
 #define MIP6_OPT_PAD_1	0
 #define MIP6_OPT_PAD_N	1
 
@@ -53,5 +56,6 @@ struct ip6_mh {
 
 extern int mip6_init(void);
 extern void mip6_fini(void);
+extern int mip6_mh_filter(struct sock *sk, struct sk_buff *skb);
 
 #endif
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
index a8adf891fe0e..7b5f89321482 100644
--- a/net/ipv6/mip6.c
+++ b/net/ipv6/mip6.c
@@ -26,7 +26,10 @@
 #include <linux/module.h>
 #include <linux/skbuff.h>
 #include <linux/ipv6.h>
+#include <linux/icmpv6.h>
+#include <net/sock.h>
 #include <net/ipv6.h>
+#include <net/ip6_checksum.h>
 #include <net/xfrm.h>
 #include <net/mip6.h>
 
@@ -55,6 +58,86 @@ static inline void *mip6_padn(__u8 *data, __u8 padlen)
 	return data + padlen;
 }
 
+static inline void mip6_param_prob(struct sk_buff *skb, int code, int pos)
+{
+	icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos, skb->dev);
+}
+
+static int mip6_mh_len(int type)
+{
+	int len = 0;
+
+	switch (type) {
+	case IP6_MH_TYPE_BRR:
+		len = 0;
+		break;
+	case IP6_MH_TYPE_HOTI:
+	case IP6_MH_TYPE_COTI:
+	case IP6_MH_TYPE_BU:
+	case IP6_MH_TYPE_BACK:
+		len = 1;
+		break;
+	case IP6_MH_TYPE_HOT:
+	case IP6_MH_TYPE_COT:
+	case IP6_MH_TYPE_BERROR:
+		len = 2;
+		break;
+	}
+	return len;
+}
+
+int mip6_mh_filter(struct sock *sk, struct sk_buff *skb)
+{
+	struct ip6_mh *mh;
+	int mhlen;
+
+	if (!pskb_may_pull(skb, (skb->h.raw - skb->data) + 8) ||
+	    !pskb_may_pull(skb, (skb->h.raw - skb->data) + ((skb->h.raw[1] + 1) << 3)))
+		return -1;
+
+	mh = (struct ip6_mh *)skb->h.raw;
+
+	if (mh->ip6mh_hdrlen < mip6_mh_len(mh->ip6mh_type)) {
+		LIMIT_NETDEBUG(KERN_DEBUG "mip6: MH message too short: %d vs >=%d\n",
+			       mh->ip6mh_hdrlen, mip6_mh_len(mh->ip6mh_type));
+		mip6_param_prob(skb, 0, (&mh->ip6mh_hdrlen) - skb->nh.raw);
+		return -1;
+	}
+	mhlen = (mh->ip6mh_hdrlen + 1) << 3;
+
+	if (skb->ip_summed == CHECKSUM_COMPLETE) {
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		if (csum_ipv6_magic(&skb->nh.ipv6h->saddr,
+				    &skb->nh.ipv6h->daddr,
+				    mhlen, IPPROTO_MH,
+				    skb->csum)) {
+			LIMIT_NETDEBUG(KERN_DEBUG "mip6: MH hw checksum failed\n");
+			skb->ip_summed = CHECKSUM_NONE;
+		}
+	}
+	if (skb->ip_summed == CHECKSUM_NONE) {
+		if (csum_ipv6_magic(&skb->nh.ipv6h->saddr,
+				    &skb->nh.ipv6h->daddr,
+				    mhlen, IPPROTO_MH,
+				    skb_checksum(skb, 0, mhlen, 0))) {
+			LIMIT_NETDEBUG(KERN_DEBUG "mip6: MH checksum failed [%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x > %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x]\n",
+				       NIP6(skb->nh.ipv6h->saddr),
+				       NIP6(skb->nh.ipv6h->daddr));
+			return -1;
+		}
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+
+	if (mh->ip6mh_proto != IPPROTO_NONE) {
+		LIMIT_NETDEBUG(KERN_DEBUG "mip6: MH invalid payload proto = %d\n",
+			       mh->ip6mh_proto);
+		mip6_param_prob(skb, 0, (&mh->ip6mh_proto) - skb->nh.raw);
+		return -1;
+	}
+
+	return 0;
+}
+
 static int mip6_destopt_input(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct ipv6hdr *iph = skb->nh.ipv6h;
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index d4af1cb5e19f..ecca8aae3c4b 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -50,6 +50,9 @@
 #include <net/udp.h>
 #include <net/inet_common.h>
 #include <net/tcp_states.h>
+#ifdef CONFIG_IPV6_MIP6
+#include <net/mip6.h>
+#endif
 
 #include <net/rawv6.h>
 #include <net/xfrm.h>
@@ -169,8 +172,32 @@ int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
 	sk = __raw_v6_lookup(sk, nexthdr, daddr, saddr, IP6CB(skb)->iif);
 
 	while (sk) {
+		int filtered;
+
 		delivered = 1;
-		if (nexthdr != IPPROTO_ICMPV6 || !icmpv6_filter(sk, skb)) {
+		switch (nexthdr) {
+		case IPPROTO_ICMPV6:
+			filtered = icmpv6_filter(sk, skb);
+			break;
+#ifdef CONFIG_IPV6_MIP6
+		case IPPROTO_MH:
+			/* XXX: To validate MH only once for each packet,
+			 * this is placed here. It should be after checking
+			 * xfrm policy, however it doesn't. The checking xfrm
+			 * policy is placed in rawv6_rcv() because it is
+			 * required for each socket.
+			 */
+			filtered = mip6_mh_filter(sk, skb);
+			break;
+#endif
+		default:
+			filtered = 0;
+			break;
+		}
+
+		if (filtered < 0)
+			break;
+		if (filtered == 0) {
 			struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);
 
 			/* Not releasing hash table! */
-- 
cgit v1.2.3


From 2ce4272a699c731b9736d76126dc742353e381db Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 20:39:03 -0700
Subject: [IPV6] MIP6: Transformation support mobility header.

Transformation support mobility header.
Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      |  5 +++++
 net/ipv6/xfrm6_policy.c | 15 +++++++++++++++
 2 files changed, 20 insertions(+)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 7f1630630dcf..13488e7ba68c 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -546,6 +546,11 @@ u16 xfrm_flowi_sport(struct flowi *fl)
 	case IPPROTO_ICMPV6:
 		port = htons(fl->fl_icmp_type);
 		break;
+#ifdef CONFIG_IPV6_MIP6
+	case IPPROTO_MH:
+		port = htons(fl->fl_mh_type);
+		break;
+#endif
 	default:
 		port = 0;	/*XXX*/
 	}
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 729b4748d6d3..98c2fe449b3f 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -18,6 +18,9 @@
 #include <net/ip.h>
 #include <net/ipv6.h>
 #include <net/ip6_route.h>
+#ifdef CONFIG_IPV6_MIP6
+#include <net/mip6.h>
+#endif
 
 static struct dst_ops xfrm6_dst_ops;
 static struct xfrm_policy_afinfo xfrm6_policy_afinfo;
@@ -270,6 +273,18 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl)
 			fl->proto = nexthdr;
 			return;
 
+#ifdef CONFIG_IPV6_MIP6
+		case IPPROTO_MH:
+			if (pskb_may_pull(skb, skb->nh.raw + offset + 3 - skb->data)) {
+				struct ip6_mh *mh;
+				mh = (struct ip6_mh *)exthdr;
+
+				fl->fl_mh_type = mh->ip6mh_type;
+			}
+			fl->proto = nexthdr;
+			return;
+#endif
+
 		/* XXX Why are there these headers? */
 		case IPPROTO_AH:
 		case IPPROTO_ESP:
-- 
cgit v1.2.3


From df0ba92a99ca757039dfa84a929281ea3f7a50e8 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 20:41:00 -0700
Subject: [XFRM]: Trace which secpath state is reject factor.

For Mobile IPv6 usage, it is required to trace which secpath state is
reject factor in order to notify it to user space (to know the address
which cannot be used route optimized communication).

Based on MIPL2 kernel patch.

This patch was also written by: Henrik Petander <petander@tcs.hut.fi>

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  1 +
 net/xfrm/xfrm_policy.c | 55 +++++++++++++++++++++++++++++++++++++++++++-------
 2 files changed, 49 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 13488e7ba68c..9ebbdc1dd471 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -273,6 +273,7 @@ struct xfrm_type
 	void			(*destructor)(struct xfrm_state *);
 	int			(*input)(struct xfrm_state *, struct sk_buff *skb);
 	int			(*output)(struct xfrm_state *, struct sk_buff *pskb);
+	int			(*reject)(struct xfrm_state *, struct sk_buff *, struct flowi *);
 	int			(*hdr_offset)(struct xfrm_state *, struct sk_buff *, u8 **);
 	xfrm_address_t		*(*local_addr)(struct xfrm_state *, xfrm_address_t *);
 	xfrm_address_t		*(*remote_addr)(struct xfrm_state *, xfrm_address_t *);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index ad2a5cba1f5b..d125a2649037 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -988,6 +988,23 @@ error:
 }
 EXPORT_SYMBOL(xfrm_lookup);
 
+static inline int
+xfrm_secpath_reject(int idx, struct sk_buff *skb, struct flowi *fl)
+{
+	struct xfrm_state *x;
+	int err;
+
+	if (!skb->sp || idx < 0 || idx >= skb->sp->len)
+		return 0;
+	x = skb->sp->xvec[idx];
+	if (!x->type->reject)
+		return 0;
+	xfrm_state_hold(x);
+	err = x->type->reject(x, skb, fl);
+	xfrm_state_put(x);
+	return err;
+}
+
 /* When skb is transformed back to its "native" form, we have to
  * check policy restrictions. At the moment we make this in maximally
  * stupid way. Shame on me. :-) Of course, connected sockets must
@@ -1010,6 +1027,13 @@ xfrm_state_ok(struct xfrm_tmpl *tmpl, struct xfrm_state *x,
 		  xfrm_state_addr_cmp(tmpl, x, family));
 }
 
+/*
+ * 0 or more than 0 is returned when validation is succeeded (either bypass
+ * because of optional transport mode, or next index of the mathced secpath
+ * state with the template.
+ * -1 is returned when no matching template is found.
+ * Otherwise "-2 - errored_index" is returned.
+ */
 static inline int
 xfrm_policy_ok(struct xfrm_tmpl *tmpl, struct sec_path *sp, int start,
 	       unsigned short family)
@@ -1024,8 +1048,11 @@ xfrm_policy_ok(struct xfrm_tmpl *tmpl, struct sec_path *sp, int start,
 	for (; idx < sp->len; idx++) {
 		if (xfrm_state_ok(tmpl, sp->xvec[idx], family))
 			return ++idx;
-		if (sp->xvec[idx]->props.mode != XFRM_MODE_TRANSPORT)
+		if (sp->xvec[idx]->props.mode != XFRM_MODE_TRANSPORT) {
+			if (start == -1)
+				start = -2-idx;
 			break;
+		}
 	}
 	return start;
 }
@@ -1046,11 +1073,14 @@ xfrm_decode_session(struct sk_buff *skb, struct flowi *fl, unsigned short family
 }
 EXPORT_SYMBOL(xfrm_decode_session);
 
-static inline int secpath_has_nontransport(struct sec_path *sp, int k)
+static inline int secpath_has_nontransport(struct sec_path *sp, int k, int *idxp)
 {
 	for (; k < sp->len; k++) {
-		if (sp->xvec[k]->props.mode != XFRM_MODE_TRANSPORT)
+		if (sp->xvec[k]->props.mode != XFRM_MODE_TRANSPORT) {
+			if (idxp)
+				*idxp = k;
 			return 1;
+		}
 	}
 
 	return 0;
@@ -1062,6 +1092,8 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 	struct xfrm_policy *pol;
 	struct flowi fl;
 	u8 fl_dir = policy_to_flow_dir(dir);
+	int xerr_idx = -1;
+	int *xerr_idxp = &xerr_idx;
 
 	if (xfrm_decode_session(skb, &fl, family) < 0)
 		return 0;
@@ -1086,8 +1118,13 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		pol = flow_cache_lookup(&fl, family, fl_dir,
 					xfrm_policy_lookup);
 
-	if (!pol)
-		return !skb->sp || !secpath_has_nontransport(skb->sp, 0);
+	if (!pol) {
+		if (skb->sp && secpath_has_nontransport(skb->sp, 0, xerr_idxp)) {
+			xfrm_secpath_reject(xerr_idx, skb, &fl);
+			return 0;
+		}
+		return 1;
+	}
 
 	pol->curlft.use_time = (unsigned long)xtime.tv_sec;
 
@@ -1107,11 +1144,14 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		 */
 		for (i = pol->xfrm_nr-1, k = 0; i >= 0; i--) {
 			k = xfrm_policy_ok(pol->xfrm_vec+i, sp, k, family);
-			if (k < 0)
+			if (k < 0) {
+				if (k < -1 && xerr_idxp)
+					*xerr_idxp = -(2+k);
 				goto reject;
+			}
 		}
 
-		if (secpath_has_nontransport(sp, k))
+		if (secpath_has_nontransport(sp, k, xerr_idxp))
 			goto reject;
 
 		xfrm_pol_put(pol);
@@ -1119,6 +1159,7 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 	}
 
 reject:
+	xfrm_secpath_reject(xerr_idx, skb, &fl);
 	xfrm_pol_put(pol);
 	return 0;
 }
-- 
cgit v1.2.3


From 97a64b4577ae2bc5599dbd008a3cd9e25de9b9f5 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 20:44:06 -0700
Subject: [XFRM]: Introduce XFRM_MSG_REPORT.

XFRM_MSG_REPORT is a message as notification of state protocol and
selector from kernel to user-space.

Mobile IPv6 will use it when inbound reject is occurred at route
optimization to make user-space know a binding error requirement.

Based on MIPL2 kernel patch.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h  | 12 ++++++++++++
 include/net/xfrm.h    |  2 ++
 net/xfrm/xfrm_state.c | 19 +++++++++++++++++++
 net/xfrm/xfrm_user.c  | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 79 insertions(+)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 1d8c1f22c12d..4009f4445fa9 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -166,6 +166,10 @@ enum {
 #define XFRM_MSG_NEWAE XFRM_MSG_NEWAE
 	XFRM_MSG_GETAE,
 #define XFRM_MSG_GETAE XFRM_MSG_GETAE
+
+	XFRM_MSG_REPORT,
+#define XFRM_MSG_REPORT XFRM_MSG_REPORT
+
 	__XFRM_MSG_MAX
 };
 #define XFRM_MSG_MAX (__XFRM_MSG_MAX - 1)
@@ -325,12 +329,18 @@ struct xfrm_usersa_flush {
 	__u8				proto;
 };
 
+struct xfrm_user_report {
+	__u8				proto;
+	struct xfrm_selector		sel;
+};
+
 #ifndef __KERNEL__
 /* backwards compatibility for userspace */
 #define XFRMGRP_ACQUIRE		1
 #define XFRMGRP_EXPIRE		2
 #define XFRMGRP_SA		4
 #define XFRMGRP_POLICY		8
+#define XFRMGRP_REPORT		0x10
 #endif
 
 enum xfrm_nlgroups {
@@ -346,6 +356,8 @@ enum xfrm_nlgroups {
 #define XFRMNLGRP_POLICY	XFRMNLGRP_POLICY
 	XFRMNLGRP_AEVENTS,
 #define XFRMNLGRP_AEVENTS	XFRMNLGRP_AEVENTS
+	XFRMNLGRP_REPORT,
+#define XFRMNLGRP_REPORT	XFRMNLGRP_REPORT
 	__XFRMNLGRP_MAX
 };
 #define XFRMNLGRP_MAX	(__XFRMNLGRP_MAX - 1)
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 9ebbdc1dd471..0b223eed4c9b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -381,6 +381,7 @@ struct xfrm_mgr
 	struct xfrm_policy	*(*compile_policy)(struct sock *sk, int opt, u8 *data, int len, int *dir);
 	int			(*new_mapping)(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport);
 	int			(*notify_policy)(struct xfrm_policy *x, int dir, struct km_event *c);
+	int			(*report)(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr);
 };
 
 extern int xfrm_register_km(struct xfrm_mgr *km);
@@ -1043,6 +1044,7 @@ extern void xfrm_init_pmtu(struct dst_entry *dst);
 extern wait_queue_head_t km_waitq;
 extern int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport);
 extern void km_policy_expired(struct xfrm_policy *pol, int dir, int hard, u32 pid);
+extern int km_report(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr);
 
 extern void xfrm_input_init(void);
 extern int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, u32 *spi, u32 *seq);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 3da89c01ea71..a26ef6952c30 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1055,6 +1055,25 @@ void km_policy_expired(struct xfrm_policy *pol, int dir, int hard, u32 pid)
 }
 EXPORT_SYMBOL(km_policy_expired);
 
+int km_report(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr)
+{
+	int err = -EINVAL;
+	int ret;
+	struct xfrm_mgr *km;
+
+	read_lock(&xfrm_km_lock);
+	list_for_each_entry(km, &xfrm_km_list, list) {
+		if (km->report) {
+			ret = km->report(proto, sel, addr);
+			if (!ret)
+				err = ret;
+		}
+	}
+	read_unlock(&xfrm_km_lock);
+	return err;
+}
+EXPORT_SYMBOL(km_report);
+
 int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen)
 {
 	int err;
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 770bd2410749..7303b820bea4 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -1491,6 +1491,7 @@ static const int xfrm_msg_min[XFRM_NR_MSGTYPES] = {
 	[XFRM_MSG_FLUSHPOLICY - XFRM_MSG_BASE] = NLMSG_LENGTH(0),
 	[XFRM_MSG_NEWAE       - XFRM_MSG_BASE] = XMSGSIZE(xfrm_aevent_id),
 	[XFRM_MSG_GETAE       - XFRM_MSG_BASE] = XMSGSIZE(xfrm_aevent_id),
+	[XFRM_MSG_REPORT      - XFRM_MSG_BASE] = XMSGSIZE(xfrm_user_report),
 };
 
 #undef XMSGSIZE
@@ -2058,12 +2059,57 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_ev
 
 }
 
+static int build_report(struct sk_buff *skb, u8 proto,
+			struct xfrm_selector *sel, xfrm_address_t *addr)
+{
+	struct xfrm_user_report *ur;
+	struct nlmsghdr *nlh;
+	unsigned char *b = skb->tail;
+
+	nlh = NLMSG_PUT(skb, 0, 0, XFRM_MSG_REPORT, sizeof(*ur));
+	ur = NLMSG_DATA(nlh);
+	nlh->nlmsg_flags = 0;
+
+	ur->proto = proto;
+	memcpy(&ur->sel, sel, sizeof(ur->sel));
+
+	if (addr)
+		RTA_PUT(skb, XFRMA_COADDR, sizeof(*addr), addr);
+
+	nlh->nlmsg_len = skb->tail - b;
+	return skb->len;
+
+nlmsg_failure:
+rtattr_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
+}
+
+static int xfrm_send_report(u8 proto, struct xfrm_selector *sel,
+			    xfrm_address_t *addr)
+{
+	struct sk_buff *skb;
+	size_t len;
+
+	len = NLMSG_ALIGN(NLMSG_LENGTH(sizeof(struct xfrm_user_report)));
+	skb = alloc_skb(len, GFP_ATOMIC);
+	if (skb == NULL)
+		return -ENOMEM;
+
+	if (build_report(skb, proto, sel, addr) < 0)
+		BUG();
+
+	NETLINK_CB(skb).dst_group = XFRMNLGRP_REPORT;
+	return netlink_broadcast(xfrm_nl, skb, 0, XFRMNLGRP_REPORT, GFP_ATOMIC);
+}
+
 static struct xfrm_mgr netlink_mgr = {
 	.id		= "netlink",
 	.notify		= xfrm_send_state_notify,
 	.acquire	= xfrm_send_acquire,
 	.compile_policy	= xfrm_compile_policy,
 	.notify_policy	= xfrm_send_policy_notify,
+	.report		= xfrm_send_report,
 };
 
 static int __init xfrm_user_init(void)
-- 
cgit v1.2.3


From 4e81bb8336a0ac50289d4d4c7a55e559b994ee8f Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 22:43:30 -0700
Subject: [XFRM] POLICY: sub policy support.

Sub policy is introduced. Main and sub policy are applied the same flow.
(Policy that current kernel uses is named as main.)
It is required another transformation policy management to keep IPsec
and Mobile IPv6 lives separate.
Policy which lives shorter time in kernel should be a sub i.e. normally
main is for IPsec and sub is for Mobile IPv6.
(Such usage as two IPsec policies on different database can be used, too.)

Limitation or TODOs:
 - Sub policy is not supported for per socket one (it is always inserted as main).
 - Current kernel makes cached outbound with flowi to skip searching database.
   However this patch makes it disabled only when "two policies are used and
   the first matched one is bypass case" because neither flowi nor bundle
   information knows about transformation template size.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/xfrm.h   |   7 ++
 include/net/xfrm.h     |  45 +++++++--
 net/xfrm/xfrm_policy.c | 252 ++++++++++++++++++++++++++++++++++++++++++-------
 3 files changed, 260 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 4009f4445fa9..492fb9818747 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -102,6 +102,13 @@ struct xfrm_stats {
 	__u32	integrity_failed;
 };
 
+enum
+{
+	XFRM_POLICY_TYPE_MAIN	= 0,
+	XFRM_POLICY_TYPE_SUB	= 1,
+	XFRM_POLICY_TYPE_MAX	= 2
+};
+
 enum
 {
 	XFRM_POLICY_IN	= 0,
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 0b223eed4c9b..4655ca25f808 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -341,6 +341,7 @@ struct xfrm_policy
 	atomic_t		refcnt;
 	struct timer_list	timer;
 
+	u8			type;
 	u32			priority;
 	u32			index;
 	struct xfrm_selector	selector;
@@ -389,6 +390,19 @@ extern int xfrm_unregister_km(struct xfrm_mgr *km);
 
 
 extern struct xfrm_policy *xfrm_policy_list[XFRM_POLICY_MAX*2];
+#ifdef CONFIG_XFRM_SUB_POLICY
+extern struct xfrm_policy *xfrm_policy_list_sub[XFRM_POLICY_MAX*2];
+
+static inline int xfrm_policy_lists_empty(int dir)
+{
+	return (!xfrm_policy_list[dir] && !xfrm_policy_list_sub[dir]);
+}
+#else
+static inline int xfrm_policy_lists_empty(int dir)
+{
+	return (!xfrm_policy_list[dir]);
+}
+#endif
 
 static inline void xfrm_pol_hold(struct xfrm_policy *policy)
 {
@@ -404,6 +418,20 @@ static inline void xfrm_pol_put(struct xfrm_policy *policy)
 		__xfrm_policy_destroy(policy);
 }
 
+#ifdef CONFIG_XFRM_SUB_POLICY
+static inline void xfrm_pols_put(struct xfrm_policy **pols, int npols)
+{
+	int i;
+	for (i = npols - 1; i >= 0; --i)
+		xfrm_pol_put(pols[i]);
+}
+#else
+static inline void xfrm_pols_put(struct xfrm_policy **pols, int npols)
+{
+	xfrm_pol_put(pols[0]);
+}
+#endif
+
 #define XFRM_DST_HSIZE		1024
 
 static __inline__
@@ -737,8 +765,8 @@ static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *sk
 {
 	if (sk && sk->sk_policy[XFRM_POLICY_IN])
 		return __xfrm_policy_check(sk, dir, skb, family);
-		
-	return	(!xfrm_policy_list[dir] && !skb->sp) ||
+
+	return	(xfrm_policy_lists_empty(dir) && !skb->sp) ||
 		(skb->dst->flags & DST_NOPOLICY) ||
 		__xfrm_policy_check(sk, dir, skb, family);
 }
@@ -758,7 +786,7 @@ extern int __xfrm_route_forward(struct sk_buff *skb, unsigned short family);
 
 static inline int xfrm_route_forward(struct sk_buff *skb, unsigned short family)
 {
-	return	!xfrm_policy_list[XFRM_POLICY_OUT] ||
+	return	xfrm_policy_lists_empty(XFRM_POLICY_OUT) ||
 		(skb->dst->flags & DST_NOXFRM) ||
 		__xfrm_route_forward(skb, family);
 }
@@ -1023,18 +1051,19 @@ static inline int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl, unsig
 #endif
 
 struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp);
-extern int xfrm_policy_walk(int (*func)(struct xfrm_policy *, int, int, void*), void *);
+extern int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*), void *);
 int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl);
-struct xfrm_policy *xfrm_policy_bysel_ctx(int dir, struct xfrm_selector *sel,
+struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir,
+					  struct xfrm_selector *sel,
 					  struct xfrm_sec_ctx *ctx, int delete);
-struct xfrm_policy *xfrm_policy_byid(int dir, u32 id, int delete);
-void xfrm_policy_flush(void);
+struct xfrm_policy *xfrm_policy_byid(u8, int dir, u32 id, int delete);
+void xfrm_policy_flush(u8 type);
 u32 xfrm_get_acqseq(void);
 void xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi);
 struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto, 
 				  xfrm_address_t *daddr, xfrm_address_t *saddr, 
 				  int create, unsigned short family);
-extern void xfrm_policy_flush(void);
+extern void xfrm_policy_flush(u8 type);
 extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol);
 extern int xfrm_flush_bundles(void);
 extern void xfrm_flush_all_bundles(void);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index d125a2649037..96de6c76ed57 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -32,6 +32,24 @@ static DEFINE_RWLOCK(xfrm_policy_lock);
 
 struct xfrm_policy *xfrm_policy_list[XFRM_POLICY_MAX*2];
 EXPORT_SYMBOL(xfrm_policy_list);
+#ifdef CONFIG_XFRM_SUB_POLICY
+struct xfrm_policy *xfrm_policy_list_sub[XFRM_POLICY_MAX*2];
+EXPORT_SYMBOL(xfrm_policy_list_sub);
+
+#define XFRM_POLICY_LISTS(type) \
+	((type == XFRM_POLICY_TYPE_SUB) ? xfrm_policy_list_sub : \
+	 xfrm_policy_list)
+#define XFRM_POLICY_LISTHEAD(type, dir) \
+	((type == XFRM_POLICY_TYPE_SUB) ? xfrm_policy_list_sub[dir] : \
+	 xfrm_policy_list[dir])
+#define XFRM_POLICY_LISTHEADP(type, dir) \
+	((type == XFRM_POLICY_TYPE_SUB) ? &xfrm_policy_list_sub[dir] : \
+	 &xfrm_policy_list[dir])
+#else
+#define XFRM_POLICY_LISTS(type)              xfrm_policy_list
+#define XFRM_POLICY_LISTHEAD(type, dif)      xfrm_policy_list[dir]
+#define XFRM_POLICY_LISTHEADP(type, dif)     &xfrm_policy_list[dir]
+#endif
 
 static DEFINE_RWLOCK(xfrm_policy_afinfo_lock);
 static struct xfrm_policy_afinfo *xfrm_policy_afinfo[NPROTO];
@@ -397,7 +415,7 @@ static void xfrm_policy_kill(struct xfrm_policy *policy)
 
 /* Generate new index... KAME seems to generate them ordered by cost
  * of an absolute inpredictability of ordering of rules. This will not pass. */
-static u32 xfrm_gen_index(int dir)
+static u32 xfrm_gen_index(u8 type, int dir)
 {
 	u32 idx;
 	struct xfrm_policy *p;
@@ -408,7 +426,7 @@ static u32 xfrm_gen_index(int dir)
 		idx_generator += 8;
 		if (idx == 0)
 			idx = 8;
-		for (p = xfrm_policy_list[dir]; p; p = p->next) {
+		for (p = XFRM_POLICY_LISTHEAD(type, dir); p; p = p->next) {
 			if (p->index == idx)
 				break;
 		}
@@ -425,7 +443,7 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 	struct dst_entry *gc_list;
 
 	write_lock_bh(&xfrm_policy_lock);
-	for (p = &xfrm_policy_list[dir]; (pol=*p)!=NULL;) {
+	for (p = XFRM_POLICY_LISTHEADP(policy->type, dir); (pol=*p)!=NULL;) {
 		if (!delpol && memcmp(&policy->selector, &pol->selector, sizeof(pol->selector)) == 0 &&
 		    xfrm_sec_ctx_match(pol->security, policy->security)) {
 			if (excl) {
@@ -452,7 +470,7 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 	policy->next = *p;
 	*p = policy;
 	atomic_inc(&flow_cache_genid);
-	policy->index = delpol ? delpol->index : xfrm_gen_index(dir);
+	policy->index = delpol ? delpol->index : xfrm_gen_index(policy->type, dir);
 	policy->curlft.add_time = (unsigned long)xtime.tv_sec;
 	policy->curlft.use_time = 0;
 	if (!mod_timer(&policy->timer, jiffies + HZ))
@@ -493,13 +511,14 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 }
 EXPORT_SYMBOL(xfrm_policy_insert);
 
-struct xfrm_policy *xfrm_policy_bysel_ctx(int dir, struct xfrm_selector *sel,
+struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir,
+					  struct xfrm_selector *sel,
 					  struct xfrm_sec_ctx *ctx, int delete)
 {
 	struct xfrm_policy *pol, **p;
 
 	write_lock_bh(&xfrm_policy_lock);
-	for (p = &xfrm_policy_list[dir]; (pol=*p)!=NULL; p = &pol->next) {
+	for (p = XFRM_POLICY_LISTHEADP(type, dir); (pol=*p)!=NULL; p = &pol->next) {
 		if ((memcmp(sel, &pol->selector, sizeof(*sel)) == 0) &&
 		    (xfrm_sec_ctx_match(ctx, pol->security))) {
 			xfrm_pol_hold(pol);
@@ -518,12 +537,12 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(int dir, struct xfrm_selector *sel,
 }
 EXPORT_SYMBOL(xfrm_policy_bysel_ctx);
 
-struct xfrm_policy *xfrm_policy_byid(int dir, u32 id, int delete)
+struct xfrm_policy *xfrm_policy_byid(u8 type, int dir, u32 id, int delete)
 {
 	struct xfrm_policy *pol, **p;
 
 	write_lock_bh(&xfrm_policy_lock);
-	for (p = &xfrm_policy_list[dir]; (pol=*p)!=NULL; p = &pol->next) {
+	for (p = XFRM_POLICY_LISTHEADP(type, dir); (pol=*p)!=NULL; p = &pol->next) {
 		if (pol->index == id) {
 			xfrm_pol_hold(pol);
 			if (delete)
@@ -541,15 +560,16 @@ struct xfrm_policy *xfrm_policy_byid(int dir, u32 id, int delete)
 }
 EXPORT_SYMBOL(xfrm_policy_byid);
 
-void xfrm_policy_flush(void)
+void xfrm_policy_flush(u8 type)
 {
 	struct xfrm_policy *xp;
+	struct xfrm_policy **p_list = XFRM_POLICY_LISTS(type);
 	int dir;
 
 	write_lock_bh(&xfrm_policy_lock);
 	for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
-		while ((xp = xfrm_policy_list[dir]) != NULL) {
-			xfrm_policy_list[dir] = xp->next;
+		while ((xp = p_list[dir]) != NULL) {
+			p_list[dir] = xp->next;
 			write_unlock_bh(&xfrm_policy_lock);
 
 			xfrm_policy_kill(xp);
@@ -562,7 +582,7 @@ void xfrm_policy_flush(void)
 }
 EXPORT_SYMBOL(xfrm_policy_flush);
 
-int xfrm_policy_walk(int (*func)(struct xfrm_policy *, int, int, void*),
+int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*),
 		     void *data)
 {
 	struct xfrm_policy *xp;
@@ -572,7 +592,7 @@ int xfrm_policy_walk(int (*func)(struct xfrm_policy *, int, int, void*),
 
 	read_lock_bh(&xfrm_policy_lock);
 	for (dir = 0; dir < 2*XFRM_POLICY_MAX; dir++) {
-		for (xp = xfrm_policy_list[dir]; xp; xp = xp->next)
+		for (xp = XFRM_POLICY_LISTHEAD(type, dir); xp; xp = xp->next)
 			count++;
 	}
 
@@ -582,7 +602,7 @@ int xfrm_policy_walk(int (*func)(struct xfrm_policy *, int, int, void*),
 	}
 
 	for (dir = 0; dir < 2*XFRM_POLICY_MAX; dir++) {
-		for (xp = xfrm_policy_list[dir]; xp; xp = xp->next) {
+		for (xp = XFRM_POLICY_LISTHEAD(type, dir); xp; xp = xp->next) {
 			error = func(xp, dir%XFRM_POLICY_MAX, --count, data);
 			if (error)
 				goto out;
@@ -597,13 +617,13 @@ EXPORT_SYMBOL(xfrm_policy_walk);
 
 /* Find policy to apply to this flow. */
 
-static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir,
-			       void **objp, atomic_t **obj_refp)
+static struct xfrm_policy *xfrm_policy_lookup_bytype(u8 type, struct flowi *fl,
+						     u16 family, u8 dir)
 {
 	struct xfrm_policy *pol;
 
 	read_lock_bh(&xfrm_policy_lock);
-	for (pol = xfrm_policy_list[dir]; pol; pol = pol->next) {
+	for (pol = XFRM_POLICY_LISTHEAD(type, dir); pol; pol = pol->next) {
 		struct xfrm_selector *sel = &pol->selector;
 		int match;
 
@@ -620,6 +640,25 @@ static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir,
 		}
 	}
 	read_unlock_bh(&xfrm_policy_lock);
+
+	return pol;
+}
+
+static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir,
+			       void **objp, atomic_t **obj_refp)
+{
+	struct xfrm_policy *pol;
+
+#ifdef CONFIG_XFRM_SUB_POLICY
+	pol = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_SUB, fl, family, dir);
+	if (pol)
+		goto end;
+#endif
+	pol = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_MAIN, fl, family, dir);
+
+#ifdef CONFIG_XFRM_SUB_POLICY
+ end:
+#endif
 	if ((*objp = (void *) pol) != NULL)
 		*obj_refp = &pol->refcnt;
 }
@@ -665,8 +704,10 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc
 
 static void __xfrm_policy_link(struct xfrm_policy *pol, int dir)
 {
-	pol->next = xfrm_policy_list[dir];
-	xfrm_policy_list[dir] = pol;
+	struct xfrm_policy **p_list = XFRM_POLICY_LISTS(pol->type);
+
+	pol->next = p_list[dir];
+	p_list[dir] = pol;
 	xfrm_pol_hold(pol);
 }
 
@@ -675,7 +716,7 @@ static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol,
 {
 	struct xfrm_policy **polp;
 
-	for (polp = &xfrm_policy_list[dir];
+	for (polp = XFRM_POLICY_LISTHEADP(pol->type, dir);
 	     *polp != NULL; polp = &(*polp)->next) {
 		if (*polp == pol) {
 			*polp = pol->next;
@@ -704,12 +745,17 @@ int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol)
 {
 	struct xfrm_policy *old_pol;
 
+#ifdef CONFIG_XFRM_SUB_POLICY
+	if (pol && pol->type != XFRM_POLICY_TYPE_MAIN)
+		return -EINVAL;
+#endif
+
 	write_lock_bh(&xfrm_policy_lock);
 	old_pol = sk->sk_policy[dir];
 	sk->sk_policy[dir] = pol;
 	if (pol) {
 		pol->curlft.add_time = (unsigned long)xtime.tv_sec;
-		pol->index = xfrm_gen_index(XFRM_POLICY_MAX+dir);
+		pol->index = xfrm_gen_index(pol->type, XFRM_POLICY_MAX+dir);
 		__xfrm_policy_link(pol, XFRM_POLICY_MAX+dir);
 	}
 	if (old_pol)
@@ -738,6 +784,7 @@ static struct xfrm_policy *clone_policy(struct xfrm_policy *old, int dir)
 		newp->flags = old->flags;
 		newp->xfrm_nr = old->xfrm_nr;
 		newp->index = old->index;
+		newp->type = old->type;
 		memcpy(newp->xfrm_vec, old->xfrm_vec,
 		       newp->xfrm_nr*sizeof(struct xfrm_tmpl));
 		write_lock_bh(&xfrm_policy_lock);
@@ -764,9 +811,9 @@ int __xfrm_sk_clone_policy(struct sock *sk)
 /* Resolve list of templates for the flow, given policy. */
 
 static int
-xfrm_tmpl_resolve(struct xfrm_policy *policy, struct flowi *fl,
-		  struct xfrm_state **xfrm,
-		  unsigned short family)
+xfrm_tmpl_resolve_one(struct xfrm_policy *policy, struct flowi *fl,
+		      struct xfrm_state **xfrm,
+		      unsigned short family)
 {
 	int nx;
 	int i, error;
@@ -809,6 +856,38 @@ fail:
 	return error;
 }
 
+static int
+xfrm_tmpl_resolve(struct xfrm_policy **pols, int npols, struct flowi *fl,
+		  struct xfrm_state **xfrm,
+		  unsigned short family)
+{
+	int cnx = 0;
+	int error;
+	int ret;
+	int i;
+
+	for (i = 0; i < npols; i++) {
+		if (cnx + pols[i]->xfrm_nr >= XFRM_MAX_DEPTH) {
+			error = -ENOBUFS;
+			goto fail;
+		}
+		ret = xfrm_tmpl_resolve_one(pols[i], fl, &xfrm[cnx], family);
+		if (ret < 0) {
+			error = ret;
+			goto fail;
+		} else
+			cnx += ret;
+	}
+
+	return cnx;
+
+ fail:
+	for (cnx--; cnx>=0; cnx--)
+		xfrm_state_put(xfrm[cnx]);
+	return error;
+
+}
+
 /* Check that the bundle accepts the flow and its components are
  * still valid.
  */
@@ -855,6 +934,11 @@ int xfrm_lookup(struct dst_entry **dst_p, struct flowi *fl,
 		struct sock *sk, int flags)
 {
 	struct xfrm_policy *policy;
+	struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
+	int npols;
+	int pol_dead;
+	int xfrm_nr;
+	int pi;
 	struct xfrm_state *xfrm[XFRM_MAX_DEPTH];
 	struct dst_entry *dst, *dst_orig = *dst_p;
 	int nx = 0;
@@ -866,12 +950,18 @@ int xfrm_lookup(struct dst_entry **dst_p, struct flowi *fl,
 restart:
 	genid = atomic_read(&flow_cache_genid);
 	policy = NULL;
+	for (pi = 0; pi < ARRAY_SIZE(pols); pi++)
+		pols[pi] = NULL;
+	npols = 0;
+	pol_dead = 0;
+	xfrm_nr = 0;
+
 	if (sk && sk->sk_policy[1])
 		policy = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl);
 
 	if (!policy) {
 		/* To accelerate a bit...  */
-		if ((dst_orig->flags & DST_NOXFRM) || !xfrm_policy_list[XFRM_POLICY_OUT])
+		if ((dst_orig->flags & DST_NOXFRM) || xfrm_policy_lists_empty(XFRM_POLICY_OUT))
 			return 0;
 
 		policy = flow_cache_lookup(fl, dst_orig->ops->family,
@@ -883,6 +973,9 @@ restart:
 
 	family = dst_orig->ops->family;
 	policy->curlft.use_time = (unsigned long)xtime.tv_sec;
+	pols[0] = policy;
+	npols ++;
+	xfrm_nr += pols[0]->xfrm_nr;
 
 	switch (policy->action) {
 	case XFRM_POLICY_BLOCK:
@@ -891,11 +984,13 @@ restart:
 		goto error;
 
 	case XFRM_POLICY_ALLOW:
+#ifndef CONFIG_XFRM_SUB_POLICY
 		if (policy->xfrm_nr == 0) {
 			/* Flow passes not transformed. */
 			xfrm_pol_put(policy);
 			return 0;
 		}
+#endif
 
 		/* Try to find matching bundle.
 		 *
@@ -911,7 +1006,36 @@ restart:
 		if (dst)
 			break;
 
-		nx = xfrm_tmpl_resolve(policy, fl, xfrm, family);
+#ifdef CONFIG_XFRM_SUB_POLICY
+		if (pols[0]->type != XFRM_POLICY_TYPE_MAIN) {
+			pols[1] = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_MAIN,
+							    fl, family,
+							    XFRM_POLICY_OUT);
+			if (pols[1]) {
+				if (pols[1]->action == XFRM_POLICY_BLOCK) {
+					err = -EPERM;
+					goto error;
+				}
+				npols ++;
+				xfrm_nr += pols[1]->xfrm_nr;
+			}
+		}
+
+		/*
+		 * Because neither flowi nor bundle information knows about
+		 * transformation template size. On more than one policy usage
+		 * we can realize whether all of them is bypass or not after
+		 * they are searched. See above not-transformed bypass
+		 * is surrounded by non-sub policy configuration, too.
+		 */
+		if (xfrm_nr == 0) {
+			/* Flow passes not transformed. */
+			xfrm_pols_put(pols, npols);
+			return 0;
+		}
+
+#endif
+		nx = xfrm_tmpl_resolve(pols, npols, fl, xfrm, family);
 
 		if (unlikely(nx<0)) {
 			err = nx;
@@ -924,7 +1048,7 @@ restart:
 				set_current_state(TASK_RUNNING);
 				remove_wait_queue(&km_waitq, &wait);
 
-				nx = xfrm_tmpl_resolve(policy, fl, xfrm, family);
+				nx = xfrm_tmpl_resolve(pols, npols, fl, xfrm, family);
 
 				if (nx == -EAGAIN && signal_pending(current)) {
 					err = -ERESTART;
@@ -932,7 +1056,7 @@ restart:
 				}
 				if (nx == -EAGAIN ||
 				    genid != atomic_read(&flow_cache_genid)) {
-					xfrm_pol_put(policy);
+					xfrm_pols_put(pols, npols);
 					goto restart;
 				}
 				err = nx;
@@ -942,7 +1066,7 @@ restart:
 		}
 		if (nx == 0) {
 			/* Flow passes not transformed. */
-			xfrm_pol_put(policy);
+			xfrm_pols_put(pols, npols);
 			return 0;
 		}
 
@@ -956,8 +1080,14 @@ restart:
 			goto error;
 		}
 
+		for (pi = 0; pi < npols; pi++) {
+			read_lock_bh(&pols[pi]->lock);
+			pol_dead |= pols[pi]->dead;
+			read_unlock_bh(&pols[pi]->lock);
+		}
+
 		write_lock_bh(&policy->lock);
-		if (unlikely(policy->dead || stale_bundle(dst))) {
+		if (unlikely(pol_dead || stale_bundle(dst))) {
 			/* Wow! While we worked on resolving, this
 			 * policy has gone. Retry. It is not paranoia,
 			 * we just cannot enlist new bundle to dead object.
@@ -977,12 +1107,12 @@ restart:
 	}
 	*dst_p = dst;
 	dst_release(dst_orig);
-	xfrm_pol_put(policy);
+ 	xfrm_pols_put(pols, npols);
 	return 0;
 
 error:
 	dst_release(dst_orig);
-	xfrm_pol_put(policy);
+	xfrm_pols_put(pols, npols);
 	*dst_p = NULL;
 	return err;
 }
@@ -1090,6 +1220,10 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 			unsigned short family)
 {
 	struct xfrm_policy *pol;
+	struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
+	int npols = 0;
+	int xfrm_nr;
+	int pi;
 	struct flowi fl;
 	u8 fl_dir = policy_to_flow_dir(dir);
 	int xerr_idx = -1;
@@ -1128,22 +1262,50 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 
 	pol->curlft.use_time = (unsigned long)xtime.tv_sec;
 
+	pols[0] = pol;
+	npols ++;
+#ifdef CONFIG_XFRM_SUB_POLICY
+	if (pols[0]->type != XFRM_POLICY_TYPE_MAIN) {
+		pols[1] = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_MAIN,
+						    &fl, family,
+						    XFRM_POLICY_IN);
+		if (pols[1]) {
+			pols[1]->curlft.use_time = (unsigned long)xtime.tv_sec;
+			npols ++;
+		}
+	}
+#endif
+
 	if (pol->action == XFRM_POLICY_ALLOW) {
 		struct sec_path *sp;
 		static struct sec_path dummy;
+		struct xfrm_tmpl *tp[XFRM_MAX_DEPTH];
+		struct xfrm_tmpl **tpp = tp;
+		int ti = 0;
 		int i, k;
 
 		if ((sp = skb->sp) == NULL)
 			sp = &dummy;
 
+		for (pi = 0; pi < npols; pi++) {
+			if (pols[pi] != pol &&
+			    pols[pi]->action != XFRM_POLICY_ALLOW)
+				goto reject;
+			if (ti + pols[pi]->xfrm_nr >= XFRM_MAX_DEPTH)
+				goto reject_error;
+			for (i = 0; i < pols[pi]->xfrm_nr; i++)
+				tpp[ti++] = &pols[pi]->xfrm_vec[i];
+		}
+		xfrm_nr = ti;
+
 		/* For each tunnel xfrm, find the first matching tmpl.
 		 * For each tmpl before that, find corresponding xfrm.
 		 * Order is _important_. Later we will implement
 		 * some barriers, but at the moment barriers
 		 * are implied between each two transformations.
 		 */
-		for (i = pol->xfrm_nr-1, k = 0; i >= 0; i--) {
-			k = xfrm_policy_ok(pol->xfrm_vec+i, sp, k, family);
+		for (i = xfrm_nr-1, k = 0; i >= 0; i--) {
+			k = xfrm_policy_ok(tpp[i], sp, k, family);
 			if (k < 0) {
 				if (k < -1 && xerr_idxp)
 					*xerr_idxp = -(2+k);
@@ -1154,13 +1316,14 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		if (secpath_has_nontransport(sp, k, xerr_idxp))
 			goto reject;
 
-		xfrm_pol_put(pol);
+		xfrm_pols_put(pols, npols);
 		return 1;
 	}
 
 reject:
 	xfrm_secpath_reject(xerr_idx, skb, &fl);
-	xfrm_pol_put(pol);
+reject_error:
+	xfrm_pols_put(pols, npols);
 	return 0;
 }
 EXPORT_SYMBOL(__xfrm_policy_check);
@@ -1246,6 +1409,23 @@ static void xfrm_prune_bundles(int (*func)(struct dst_entry *))
 
 	read_lock_bh(&xfrm_policy_lock);
 	for (i=0; i<2*XFRM_POLICY_MAX; i++) {
+#ifdef CONFIG_XFRM_SUB_POLICY
+		for (pol = xfrm_policy_list_sub[i]; pol; pol = pol->next) {
+			write_lock(&pol->lock);
+			dstp = &pol->bundles;
+			while ((dst=*dstp) != NULL) {
+				if (func(dst)) {
+					*dstp = dst->next;
+					dst->next = gc_list;
+					gc_list = dst;
+				} else {
+					dstp = &dst->next;
+				}
+			}
+			write_unlock(&pol->lock);
+		}
+
+#endif
 		for (pol = xfrm_policy_list[i]; pol; pol = pol->next) {
 			write_lock(&pol->lock);
 			dstp = &pol->bundles;
-- 
cgit v1.2.3


From 41a49cc3c02ace59d4dddae91ea211c330970ee3 Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 22:48:31 -0700
Subject: [XFRM]: Add sorting interface for state and template.

Under two transformation policies it is required to merge them.
This is a platform to sort state for outbound and templates
for inbound respectively.
It will be used when Mobile IPv6 and IPsec are used at the same time.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 20 ++++++++++++++++++++
 net/xfrm/xfrm_policy.c | 16 ++++++++++++++--
 net/xfrm/xfrm_state.c  | 38 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 72 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 4655ca25f808..d341603e4ba8 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -254,6 +254,8 @@ struct xfrm_state_afinfo {
 	struct xfrm_state	*(*find_acq)(u8 mode, u32 reqid, u8 proto, 
 					     xfrm_address_t *daddr, xfrm_address_t *saddr, 
 					     int create);
+	int			(*tmpl_sort)(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n);
+	int			(*state_sort)(struct xfrm_state **dst, struct xfrm_state **src, int n);
 };
 
 extern int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo);
@@ -1002,6 +1004,24 @@ extern int xfrm_state_add(struct xfrm_state *x);
 extern int xfrm_state_update(struct xfrm_state *x);
 extern struct xfrm_state *xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family);
 extern struct xfrm_state *xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family);
+#ifdef CONFIG_XFRM_SUB_POLICY
+extern int xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src,
+			  int n, unsigned short family);
+extern int xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src,
+			   int n, unsigned short family);
+#else
+static inline int xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src,
+				 int n, unsigned short family)
+{
+	return -ENOSYS;
+}
+
+static inline int xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src,
+				  int n, unsigned short family)
+{
+	return -ENOSYS;
+}
+#endif
 extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq);
 extern int xfrm_state_delete(struct xfrm_state *x);
 extern void xfrm_state_flush(u8 proto);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 96de6c76ed57..1732159ffd01 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -861,6 +861,8 @@ xfrm_tmpl_resolve(struct xfrm_policy **pols, int npols, struct flowi *fl,
 		  struct xfrm_state **xfrm,
 		  unsigned short family)
 {
+	struct xfrm_state *tp[XFRM_MAX_DEPTH];
+	struct xfrm_state **tpp = (npols > 1) ? tp : xfrm;
 	int cnx = 0;
 	int error;
 	int ret;
@@ -871,7 +873,8 @@ xfrm_tmpl_resolve(struct xfrm_policy **pols, int npols, struct flowi *fl,
 			error = -ENOBUFS;
 			goto fail;
 		}
-		ret = xfrm_tmpl_resolve_one(pols[i], fl, &xfrm[cnx], family);
+
+		ret = xfrm_tmpl_resolve_one(pols[i], fl, &tpp[cnx], family);
 		if (ret < 0) {
 			error = ret;
 			goto fail;
@@ -879,11 +882,15 @@ xfrm_tmpl_resolve(struct xfrm_policy **pols, int npols, struct flowi *fl,
 			cnx += ret;
 	}
 
+	/* found states are sorted for outbound processing */
+	if (npols > 1)
+		xfrm_state_sort(xfrm, tpp, cnx, family);
+
 	return cnx;
 
  fail:
 	for (cnx--; cnx>=0; cnx--)
-		xfrm_state_put(xfrm[cnx]);
+		xfrm_state_put(tpp[cnx]);
 	return error;
 
 }
@@ -1280,6 +1287,7 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		struct sec_path *sp;
 		static struct sec_path dummy;
 		struct xfrm_tmpl *tp[XFRM_MAX_DEPTH];
+		struct xfrm_tmpl *stp[XFRM_MAX_DEPTH];
 		struct xfrm_tmpl **tpp = tp;
 		int ti = 0;
 		int i, k;
@@ -1297,6 +1305,10 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 				tpp[ti++] = &pols[pi]->xfrm_vec[i];
 		}
 		xfrm_nr = ti;
+		if (npols > 1) {
+			xfrm_tmpl_sort(stp, tpp, xfrm_nr, family);
+			tpp = stp;
+		}
 
 		/* For each tunnel xfrm, find the first matching tmpl.
 		 * For each tmpl before that, find corresponding xfrm.
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index a26ef6952c30..622e92a08d0b 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -728,6 +728,44 @@ xfrm_find_acq(u8 mode, u32 reqid, u8 proto,
 }
 EXPORT_SYMBOL(xfrm_find_acq);
 
+#ifdef CONFIG_XFRM_SUB_POLICY
+int
+xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n,
+	       unsigned short family)
+{
+	int err = 0;
+	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
+	if (!afinfo)
+		return -EAFNOSUPPORT;
+
+	spin_lock_bh(&xfrm_state_lock);
+	if (afinfo->tmpl_sort)
+		err = afinfo->tmpl_sort(dst, src, n);
+	spin_unlock_bh(&xfrm_state_lock);
+	xfrm_state_put_afinfo(afinfo);
+	return err;
+}
+EXPORT_SYMBOL(xfrm_tmpl_sort);
+
+int
+xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n,
+		unsigned short family)
+{
+	int err = 0;
+	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
+	if (!afinfo)
+		return -EAFNOSUPPORT;
+
+	spin_lock_bh(&xfrm_state_lock);
+	if (afinfo->state_sort)
+		err = afinfo->state_sort(dst, src, n);
+	spin_unlock_bh(&xfrm_state_lock);
+	xfrm_state_put_afinfo(afinfo);
+	return err;
+}
+EXPORT_SYMBOL(xfrm_state_sort);
+#endif
+
 /* Silly enough, but I'm lazy to build resolution list */
 
 static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq)
-- 
cgit v1.2.3


From f7b6983f0feeefcd2a594138adcffe640593d8de Mon Sep 17 00:00:00 2001
From: Masahide NAKAMURA <nakam@linux-ipv6.org>
Date: Wed, 23 Aug 2006 22:49:28 -0700
Subject: [XFRM] POLICY: Support netlink socket interface for sub policy.

Sub policy can be used through netlink socket.
PF_KEY uses main only and it is TODO to support sub.

Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h |   7 +++
 include/net/xfrm.h   |   1 +
 net/key/af_key.c     |  18 +++++--
 net/xfrm/xfrm_user.c | 134 ++++++++++++++++++++++++++++++++++++++++++++++-----
 4 files changed, 142 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 492fb9818747..14ecd19f4cdc 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -230,6 +230,12 @@ enum xfrm_ae_ftype_t {
 #define XFRM_AE_MAX (__XFRM_AE_MAX - 1)
 };
 
+struct xfrm_userpolicy_type {
+	__u8		type;
+	__u16		reserved1;
+	__u8		reserved2;
+};
+
 /* Netlink message attributes.  */
 enum xfrm_attr_type_t {
 	XFRMA_UNSPEC,
@@ -248,6 +254,7 @@ enum xfrm_attr_type_t {
 	XFRMA_SRCADDR,		/* xfrm_address_t */
 	XFRMA_COADDR,		/* xfrm_address_t */
 	XFRMA_LASTUSED,
+	XFRMA_POLICY_TYPE,	/* struct xfrm_userpolicy_type */
 	__XFRMA_MAX
 
 #define XFRMA_MAX (__XFRMA_MAX - 1)
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d341603e4ba8..c75b3287d8f8 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -203,6 +203,7 @@ struct km_event
 		u32 proto;
 		u32 byid;
 		u32 aevent;
+		u32 type;
 	} data;
 
 	u32	seq;
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 19e047b0e678..83b443ddc72f 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -1731,7 +1731,8 @@ static u32 gen_reqid(void)
 		++reqid;
 		if (reqid == 0)
 			reqid = IPSEC_MANUAL_REQID_MAX+1;
-		if (xfrm_policy_walk(check_reqid, (void*)&reqid) != -EEXIST)
+		if (xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, check_reqid,
+				     (void*)&reqid) != -EEXIST)
 			return reqid;
 	} while (reqid != start);
 	return 0;
@@ -2268,7 +2269,8 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg
 			return err;
 	}
 
-	xp = xfrm_policy_bysel_ctx(pol->sadb_x_policy_dir-1, &sel, tmp.security, 1);
+	xp = xfrm_policy_bysel_ctx(XFRM_POLICY_TYPE_MAIN, pol->sadb_x_policy_dir-1,
+				   &sel, tmp.security, 1);
 	security_xfrm_policy_free(&tmp);
 	if (xp == NULL)
 		return -ENOENT;
@@ -2330,7 +2332,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h
 	if (dir >= XFRM_POLICY_MAX)
 		return -EINVAL;
 
-	xp = xfrm_policy_byid(dir, pol->sadb_x_policy_id,
+	xp = xfrm_policy_byid(XFRM_POLICY_TYPE_MAIN, dir, pol->sadb_x_policy_id,
 			      hdr->sadb_msg_type == SADB_X_SPDDELETE2);
 	if (xp == NULL)
 		return -ENOENT;
@@ -2378,7 +2380,7 @@ static int pfkey_spddump(struct sock *sk, struct sk_buff *skb, struct sadb_msg *
 {
 	struct pfkey_dump_data data = { .skb = skb, .hdr = hdr, .sk = sk };
 
-	return xfrm_policy_walk(dump_sp, &data);
+	return xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, dump_sp, &data);
 }
 
 static int key_notify_policy_flush(struct km_event *c)
@@ -2405,7 +2407,8 @@ static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg
 {
 	struct km_event c;
 
-	xfrm_policy_flush();
+	xfrm_policy_flush(XFRM_POLICY_TYPE_MAIN);
+	c.data.type = XFRM_POLICY_TYPE_MAIN;
 	c.event = XFRM_MSG_FLUSHPOLICY;
 	c.pid = hdr->sadb_msg_pid;
 	c.seq = hdr->sadb_msg_seq;
@@ -2667,6 +2670,9 @@ static int pfkey_send_notify(struct xfrm_state *x, struct km_event *c)
 
 static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c)
 {
+	if (xp && xp->type != XFRM_POLICY_TYPE_MAIN)
+		return 0;
+
 	switch (c->event) {
 	case XFRM_MSG_POLEXPIRE:
 		return key_notify_policy_expire(xp, c);
@@ -2675,6 +2681,8 @@ static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_e
 	case XFRM_MSG_UPDPOLICY:
 		return key_notify_policy(xp, dir, c);
 	case XFRM_MSG_FLUSHPOLICY:
+		if (c->data.type != XFRM_POLICY_TYPE_MAIN)
+			break;
 		return key_notify_policy_flush(c);
 	default:
 		printk("pfkey: Unknown policy event %d\n", c->event);
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 7303b820bea4..c59a78d2923a 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -786,6 +786,22 @@ static int verify_policy_dir(__u8 dir)
 	return 0;
 }
 
+static int verify_policy_type(__u8 type)
+{
+	switch (type) {
+	case XFRM_POLICY_TYPE_MAIN:
+#ifdef CONFIG_XFRM_SUB_POLICY
+	case XFRM_POLICY_TYPE_SUB:
+#endif
+		break;
+
+	default:
+		return -EINVAL;
+	};
+
+	return 0;
+}
+
 static int verify_newpolicy_info(struct xfrm_userpolicy_info *p)
 {
 	switch (p->share) {
@@ -879,6 +895,29 @@ static int copy_from_user_tmpl(struct xfrm_policy *pol, struct rtattr **xfrma)
 	return 0;
 }
 
+static int copy_from_user_policy_type(u8 *tp, struct rtattr **xfrma)
+{
+	struct rtattr *rt = xfrma[XFRMA_POLICY_TYPE-1];
+	struct xfrm_userpolicy_type *upt;
+	__u8 type = XFRM_POLICY_TYPE_MAIN;
+	int err;
+
+	if (rt) {
+		if (rt->rta_len < sizeof(*upt))
+			return -EINVAL;
+
+		upt = RTA_DATA(rt);
+		type = upt->type;
+	}
+
+	err = verify_policy_type(type);
+	if (err)
+		return err;
+
+	*tp = type;
+	return 0;
+}
+
 static void copy_from_user_policy(struct xfrm_policy *xp, struct xfrm_userpolicy_info *p)
 {
 	xp->priority = p->priority;
@@ -917,16 +956,20 @@ static struct xfrm_policy *xfrm_policy_construct(struct xfrm_userpolicy_info *p,
 
 	copy_from_user_policy(xp, p);
 
+	err = copy_from_user_policy_type(&xp->type, xfrma);
+	if (err)
+		goto error;
+
 	if (!(err = copy_from_user_tmpl(xp, xfrma)))
 		err = copy_from_user_sec_ctx(xp, xfrma);
-
-	if (err) {
-		*errp = err;
-		kfree(xp);
-		xp = NULL;
-	}
+	if (err)
+		goto error;
 
 	return xp;
+ error:
+	*errp = err;
+	kfree(xp);
+	return NULL;
 }
 
 static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
@@ -1037,6 +1080,29 @@ static inline int copy_to_user_sec_ctx(struct xfrm_policy *xp, struct sk_buff *s
 	return 0;
 }
 
+#ifdef CONFIG_XFRM_SUB_POLICY
+static int copy_to_user_policy_type(struct xfrm_policy *xp, struct sk_buff *skb)
+{
+	struct xfrm_userpolicy_type upt;
+
+	memset(&upt, 0, sizeof(upt));
+	upt.type = xp->type;
+
+	RTA_PUT(skb, XFRMA_POLICY_TYPE, sizeof(upt), &upt);
+
+	return 0;
+
+rtattr_failure:
+	return -1;
+}
+
+#else
+static inline int copy_to_user_policy_type(struct xfrm_policy *xp, struct sk_buff *skb)
+{
+	return 0;
+}
+#endif
+
 static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr)
 {
 	struct xfrm_dump_info *sp = ptr;
@@ -1060,6 +1126,8 @@ static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr
 		goto nlmsg_failure;
 	if (copy_to_user_sec_ctx(xp, skb))
 		goto nlmsg_failure;
+	if (copy_to_user_policy_type(xp, skb) < 0)
+		goto nlmsg_failure;
 
 	nlh->nlmsg_len = skb->tail - b;
 out:
@@ -1081,7 +1149,10 @@ static int xfrm_dump_policy(struct sk_buff *skb, struct netlink_callback *cb)
 	info.nlmsg_flags = NLM_F_MULTI;
 	info.this_idx = 0;
 	info.start_idx = cb->args[0];
-	(void) xfrm_policy_walk(dump_one_policy, &info);
+	(void) xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, dump_one_policy, &info);
+#ifdef CONFIG_XFRM_SUB_POLICY
+	(void) xfrm_policy_walk(XFRM_POLICY_TYPE_SUB, dump_one_policy, &info);
+#endif
 	cb->args[0] = info.this_idx;
 
 	return skb->len;
@@ -1117,6 +1188,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr
 {
 	struct xfrm_policy *xp;
 	struct xfrm_userpolicy_id *p;
+	__u8 type = XFRM_POLICY_TYPE_MAIN;
 	int err;
 	struct km_event c;
 	int delete;
@@ -1124,12 +1196,16 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr
 	p = NLMSG_DATA(nlh);
 	delete = nlh->nlmsg_type == XFRM_MSG_DELPOLICY;
 
+	err = copy_from_user_policy_type(&type, (struct rtattr **)xfrma);
+	if (err)
+		return err;
+
 	err = verify_policy_dir(p->dir);
 	if (err)
 		return err;
 
 	if (p->index)
-		xp = xfrm_policy_byid(p->dir, p->index, delete);
+		xp = xfrm_policy_byid(type, p->dir, p->index, delete);
 	else {
 		struct rtattr **rtattrs = (struct rtattr **)xfrma;
 		struct rtattr *rt = rtattrs[XFRMA_SEC_CTX-1];
@@ -1146,7 +1222,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr
 			if ((err = security_xfrm_policy_alloc(&tmp, uctx)))
 				return err;
 		}
-		xp = xfrm_policy_bysel_ctx(p->dir, &p->sel, tmp.security, delete);
+		xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security, delete);
 		security_xfrm_policy_free(&tmp);
 	}
 	if (xp == NULL)
@@ -1329,9 +1405,16 @@ out:
 
 static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
 {
-struct km_event c;
+	struct km_event c;
+	__u8 type = XFRM_POLICY_TYPE_MAIN;
+	int err;
+
+	err = copy_from_user_policy_type(&type, (struct rtattr **)xfrma);
+	if (err)
+		return err;
 
-	xfrm_policy_flush();
+	xfrm_policy_flush(type);
+	c.data.type = type;
 	c.event = nlh->nlmsg_type;
 	c.seq = nlh->nlmsg_seq;
 	c.pid = nlh->nlmsg_pid;
@@ -1344,10 +1427,15 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, void *
 	struct xfrm_policy *xp;
 	struct xfrm_user_polexpire *up = NLMSG_DATA(nlh);
 	struct xfrm_userpolicy_info *p = &up->pol;
+	__u8 type = XFRM_POLICY_TYPE_MAIN;
 	int err = -ENOENT;
 
+	err = copy_from_user_policy_type(&type, (struct rtattr **)xfrma);
+	if (err)
+		return err;
+
 	if (p->index)
-		xp = xfrm_policy_byid(p->dir, p->index, 0);
+		xp = xfrm_policy_byid(type, p->dir, p->index, 0);
 	else {
 		struct rtattr **rtattrs = (struct rtattr **)xfrma;
 		struct rtattr *rt = rtattrs[XFRMA_SEC_CTX-1];
@@ -1364,7 +1452,7 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, void *
 			if ((err = security_xfrm_policy_alloc(&tmp, uctx)))
 				return err;
 		}
-		xp = xfrm_policy_bysel_ctx(p->dir, &p->sel, tmp.security, 0);
+		xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security, 0);
 		security_xfrm_policy_free(&tmp);
 	}
 
@@ -1818,6 +1906,8 @@ static int build_acquire(struct sk_buff *skb, struct xfrm_state *x,
 		goto nlmsg_failure;
 	if (copy_to_user_state_sec_ctx(x, skb))
 		goto nlmsg_failure;
+	if (copy_to_user_policy_type(xp, skb) < 0)
+		goto nlmsg_failure;
 
 	nlh->nlmsg_len = skb->tail - b;
 	return skb->len;
@@ -1898,6 +1988,7 @@ static struct xfrm_policy *xfrm_compile_policy(struct sock *sk, int opt,
 	}
 
 	copy_from_user_policy(xp, p);
+	xp->type = XFRM_POLICY_TYPE_MAIN;
 	copy_templates(xp, ut, nr);
 
 	if (!xp->security) {
@@ -1931,6 +2022,8 @@ static int build_polexpire(struct sk_buff *skb, struct xfrm_policy *xp,
 		goto nlmsg_failure;
 	if (copy_to_user_sec_ctx(xp, skb))
 		goto nlmsg_failure;
+	if (copy_to_user_policy_type(xp, skb) < 0)
+		goto nlmsg_failure;
 	upe->hard = !!hard;
 
 	nlh->nlmsg_len = skb->tail - b;
@@ -2002,6 +2095,8 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *
 	copy_to_user_policy(xp, p, dir);
 	if (copy_to_user_tmpl(xp, skb) < 0)
 		goto nlmsg_failure;
+	if (copy_to_user_policy_type(xp, skb) < 0)
+		goto nlmsg_failure;
 
 	nlh->nlmsg_len = skb->tail - b;
 
@@ -2019,6 +2114,9 @@ static int xfrm_notify_policy_flush(struct km_event *c)
 	struct nlmsghdr *nlh;
 	struct sk_buff *skb;
 	unsigned char *b;
+#ifdef CONFIG_XFRM_SUB_POLICY
+	struct xfrm_userpolicy_type upt;
+#endif
 	int len = NLMSG_LENGTH(0);
 
 	skb = alloc_skb(len, GFP_ATOMIC);
@@ -2028,6 +2126,13 @@ static int xfrm_notify_policy_flush(struct km_event *c)
 
 
 	nlh = NLMSG_PUT(skb, c->pid, c->seq, XFRM_MSG_FLUSHPOLICY, 0);
+	nlh->nlmsg_flags = 0;
+
+#ifdef CONFIG_XFRM_SUB_POLICY
+	memset(&upt, 0, sizeof(upt));
+	upt.type = c->data.type;
+	RTA_PUT(skb, XFRMA_POLICY_TYPE, sizeof(upt), &upt);
+#endif
 
 	nlh->nlmsg_len = skb->tail - b;
 
@@ -2035,6 +2140,9 @@ static int xfrm_notify_policy_flush(struct km_event *c)
 	return netlink_broadcast(xfrm_nl, skb, 0, XFRMNLGRP_POLICY, GFP_ATOMIC);
 
 nlmsg_failure:
+#ifdef CONFIG_XFRM_SUB_POLICY
+rtattr_failure:
+#endif
 	kfree_skb(skb);
 	return -1;
 }
-- 
cgit v1.2.3


From 2770834c9f44afd1bfa13914c7285470775af657 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 00:13:10 -0700
Subject: [XFRM]: Pull xfrm_state_bydst hash table knowledge out of afinfo.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  16 -------
 net/ipv4/xfrm4_state.c |  53 ------------------------
 net/ipv6/xfrm6_state.c |  56 -------------------------
 net/xfrm/xfrm_state.c  | 110 ++++++++++++++++++++++++++++++++++++++++++++-----
 4 files changed, 100 insertions(+), 135 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index c75b3287d8f8..cc83443f301e 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -243,7 +243,6 @@ extern int __xfrm_state_delete(struct xfrm_state *x);
 
 struct xfrm_state_afinfo {
 	unsigned short		family;
-	struct list_head	*state_bydst;
 	struct list_head	*state_bysrc;
 	struct list_head	*state_byspi;
 	int			(*init_flags)(struct xfrm_state *x);
@@ -252,9 +251,6 @@ struct xfrm_state_afinfo {
 						xfrm_address_t *daddr, xfrm_address_t *saddr);
 	struct xfrm_state	*(*state_lookup)(xfrm_address_t *daddr, u32 spi, u8 proto);
 	struct xfrm_state	*(*state_lookup_byaddr)(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto);
-	struct xfrm_state	*(*find_acq)(u8 mode, u32 reqid, u8 proto, 
-					     xfrm_address_t *daddr, xfrm_address_t *saddr, 
-					     int create);
 	int			(*tmpl_sort)(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n);
 	int			(*state_sort)(struct xfrm_state **dst, struct xfrm_state **src, int n);
 };
@@ -455,18 +451,6 @@ unsigned __xfrm6_dst_hash(xfrm_address_t *addr)
 	return h;
 }
 
-static __inline__
-unsigned xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
-{
-	switch (family) {
-	case AF_INET:
-		return __xfrm4_dst_hash(addr);
-	case AF_INET6:
-		return __xfrm6_dst_hash(addr);
-	}
-	return 0;
-}
-
 static __inline__
 unsigned __xfrm4_src_hash(xfrm_address_t *addr)
 {
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 616be131b4e3..9dc1afc17b6d 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -88,65 +88,12 @@ __xfrm4_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
 	return NULL;
 }
 
-static struct xfrm_state *
-__xfrm4_find_acq(u8 mode, u32 reqid, u8 proto, 
-		 xfrm_address_t *daddr, xfrm_address_t *saddr, 
-		 int create)
-{
-	struct xfrm_state *x, *x0;
-	unsigned h = __xfrm4_dst_hash(daddr);
-
-	x0 = NULL;
-
-	list_for_each_entry(x, xfrm4_state_afinfo.state_bydst+h, bydst) {
-		if (x->props.family == AF_INET &&
-		    daddr->a4 == x->id.daddr.a4 &&
-		    mode == x->props.mode &&
-		    proto == x->id.proto &&
-		    saddr->a4 == x->props.saddr.a4 &&
-		    reqid == x->props.reqid &&
-		    x->km.state == XFRM_STATE_ACQ &&
-		    !x->id.spi) {
-			    x0 = x;
-			    break;
-		    }
-	}
-	if (!x0 && create && (x0 = xfrm_state_alloc()) != NULL) {
-		x0->sel.daddr.a4 = daddr->a4;
-		x0->sel.saddr.a4 = saddr->a4;
-		x0->sel.prefixlen_d = 32;
-		x0->sel.prefixlen_s = 32;
-		x0->props.saddr.a4 = saddr->a4;
-		x0->km.state = XFRM_STATE_ACQ;
-		x0->id.daddr.a4 = daddr->a4;
-		x0->id.proto = proto;
-		x0->props.family = AF_INET;
-		x0->props.mode = mode;
-		x0->props.reqid = reqid;
-		x0->props.family = AF_INET;
-		x0->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES;
-		xfrm_state_hold(x0);
-		x0->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ;
-		add_timer(&x0->timer);
-		xfrm_state_hold(x0);
-		list_add_tail(&x0->bydst, xfrm4_state_afinfo.state_bydst+h);
-		h = __xfrm4_src_hash(saddr);
-		xfrm_state_hold(x0);
-		list_add_tail(&x0->bysrc, xfrm4_state_afinfo.state_bysrc+h);
-		wake_up(&km_waitq);
-	}
-	if (x0)
-		xfrm_state_hold(x0);
-	return x0;
-}
-
 static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
 	.state_lookup		= __xfrm4_state_lookup,
 	.state_lookup_byaddr	= __xfrm4_state_lookup_byaddr,
-	.find_acq		= __xfrm4_find_acq,
 };
 
 void __init xfrm4_state_init(void)
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 6269584e610e..40fcaab7e028 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -101,61 +101,6 @@ __xfrm6_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto)
 	return NULL;
 }
 
-static struct xfrm_state *
-__xfrm6_find_acq(u8 mode, u32 reqid, u8 proto, 
-		 xfrm_address_t *daddr, xfrm_address_t *saddr, 
-		 int create)
-{
-	struct xfrm_state *x, *x0;
-	unsigned h = __xfrm6_dst_hash(daddr);
-
-	x0 = NULL;
-
-	list_for_each_entry(x, xfrm6_state_afinfo.state_bydst+h, bydst) {
-		if (x->props.family == AF_INET6 &&
-		    ipv6_addr_equal((struct in6_addr *)daddr, (struct in6_addr *)x->id.daddr.a6) &&
-		    mode == x->props.mode &&
-		    proto == x->id.proto &&
-		    ipv6_addr_equal((struct in6_addr *)saddr, (struct in6_addr *)x->props.saddr.a6) &&
-		    reqid == x->props.reqid &&
-		    x->km.state == XFRM_STATE_ACQ &&
-		    !x->id.spi) {
-			    x0 = x;
-			    break;
-		    }
-	}
-	if (!x0 && create && (x0 = xfrm_state_alloc()) != NULL) {
-		ipv6_addr_copy((struct in6_addr *)x0->sel.daddr.a6,
-			       (struct in6_addr *)daddr);
-		ipv6_addr_copy((struct in6_addr *)x0->sel.saddr.a6,
-			       (struct in6_addr *)saddr);
-		x0->sel.prefixlen_d = 128;
-		x0->sel.prefixlen_s = 128;
-		ipv6_addr_copy((struct in6_addr *)x0->props.saddr.a6,
-			       (struct in6_addr *)saddr);
-		x0->km.state = XFRM_STATE_ACQ;
-		ipv6_addr_copy((struct in6_addr *)x0->id.daddr.a6,
-			       (struct in6_addr *)daddr);
-		x0->id.proto = proto;
-		x0->props.family = AF_INET6;
-		x0->props.mode = mode;
-		x0->props.reqid = reqid;
-		x0->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES;
-		xfrm_state_hold(x0);
-		x0->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ;
-		add_timer(&x0->timer);
-		xfrm_state_hold(x0);
-		list_add_tail(&x0->bydst, xfrm6_state_afinfo.state_bydst+h);
-		h = __xfrm6_src_hash(saddr);
-		xfrm_state_hold(x0);
-		list_add_tail(&x0->bysrc, xfrm6_state_afinfo.state_bysrc+h);
-		wake_up(&km_waitq);
-	}
-	if (x0)
-		xfrm_state_hold(x0);
-	return x0;
-}
-
 static int
 __xfrm6_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n)
 {
@@ -280,7 +225,6 @@ static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.init_tempsel		= __xfrm6_init_tempsel,
 	.state_lookup		= __xfrm6_state_lookup,
 	.state_lookup_byaddr	= __xfrm6_state_lookup_byaddr,
-	.find_acq		= __xfrm6_find_acq,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
 	.state_sort		= __xfrm6_state_sort,
 };
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 622e92a08d0b..80f5f9dc2b9e 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -48,6 +48,18 @@ static struct list_head xfrm_state_bydst[XFRM_DST_HSIZE];
 static struct list_head xfrm_state_bysrc[XFRM_DST_HSIZE];
 static struct list_head xfrm_state_byspi[XFRM_DST_HSIZE];
 
+static __inline__
+unsigned xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		return __xfrm4_dst_hash(addr);
+	case AF_INET6:
+		return __xfrm6_dst_hash(addr);
+	}
+	return 0;
+}
+
 DECLARE_WAIT_QUEUE_HEAD(km_waitq);
 EXPORT_SYMBOL(km_waitq);
 
@@ -489,6 +501,89 @@ void xfrm_state_insert(struct xfrm_state *x)
 }
 EXPORT_SYMBOL(xfrm_state_insert);
 
+/* xfrm_state_lock is held */
+static struct xfrm_state *__find_acq_core(unsigned short family, u8 mode, u32 reqid, u8 proto, xfrm_address_t *daddr, xfrm_address_t *saddr, int create)
+{
+	unsigned int h = xfrm_dst_hash(daddr, family);
+	struct xfrm_state *x;
+
+	list_for_each_entry(x, xfrm_state_bydst+h, bydst) {
+		if (x->props.reqid  != reqid ||
+		    x->props.mode   != mode ||
+		    x->props.family != family ||
+		    x->km.state     != XFRM_STATE_ACQ ||
+		    x->id.spi       != 0)
+			continue;
+
+		switch (family) {
+		case AF_INET:
+			if (x->id.daddr.a4    != daddr->a4 ||
+			    x->props.saddr.a4 != saddr->a4)
+				continue;
+			break;
+		case AF_INET6:
+			if (!ipv6_addr_equal((struct in6_addr *)x->id.daddr.a6,
+					     (struct in6_addr *)daddr) ||
+			    !ipv6_addr_equal((struct in6_addr *)
+					     x->props.saddr.a6,
+					     (struct in6_addr *)saddr))
+				continue;
+			break;
+		};
+
+		xfrm_state_hold(x);
+		return x;
+	}
+
+	if (!create)
+		return NULL;
+
+	x = xfrm_state_alloc();
+	if (likely(x)) {
+		switch (family) {
+		case AF_INET:
+			x->sel.daddr.a4 = daddr->a4;
+			x->sel.saddr.a4 = saddr->a4;
+			x->sel.prefixlen_d = 32;
+			x->sel.prefixlen_s = 32;
+			x->props.saddr.a4 = saddr->a4;
+			x->id.daddr.a4 = daddr->a4;
+			break;
+
+		case AF_INET6:
+			ipv6_addr_copy((struct in6_addr *)x->sel.daddr.a6,
+				       (struct in6_addr *)daddr);
+			ipv6_addr_copy((struct in6_addr *)x->sel.saddr.a6,
+				       (struct in6_addr *)saddr);
+			x->sel.prefixlen_d = 128;
+			x->sel.prefixlen_s = 128;
+			ipv6_addr_copy((struct in6_addr *)x->props.saddr.a6,
+				       (struct in6_addr *)saddr);
+			ipv6_addr_copy((struct in6_addr *)x->id.daddr.a6,
+				       (struct in6_addr *)daddr);
+			break;
+		};
+
+		x->km.state = XFRM_STATE_ACQ;
+		x->id.proto = proto;
+		x->props.family = family;
+		x->props.mode = mode;
+		x->props.reqid = reqid;
+		x->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES;
+		xfrm_state_hold(x);
+		x->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ;
+		add_timer(&x->timer);
+		xfrm_state_hold(x);
+		list_add_tail(&x->bydst, xfrm_state_bydst+h);
+		h = xfrm_src_hash(saddr, family);
+		xfrm_state_hold(x);
+		list_add_tail(&x->bysrc, xfrm_state_bysrc+h);
+		wake_up(&km_waitq);
+	}
+
+	return x;
+}
+
 static inline struct xfrm_state *
 __xfrm_state_locate(struct xfrm_state_afinfo *afinfo, struct xfrm_state *x,
 		    int use_spi)
@@ -533,9 +628,9 @@ int xfrm_state_add(struct xfrm_state *x)
 	}
 
 	if (use_spi && !x1)
-		x1 = afinfo->find_acq(
-			x->props.mode, x->props.reqid, x->id.proto,
-			&x->id.daddr, &x->props.saddr, 0);
+		x1 = __find_acq_core(family, x->props.mode, x->props.reqid,
+				     x->id.proto,
+				     &x->id.daddr, &x->props.saddr, 0);
 
 	__xfrm_state_insert(x);
 	err = 0;
@@ -716,14 +811,11 @@ xfrm_find_acq(u8 mode, u32 reqid, u8 proto,
 	      int create, unsigned short family)
 {
 	struct xfrm_state *x;
-	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
-	if (!afinfo)
-		return NULL;
 
 	spin_lock_bh(&xfrm_state_lock);
-	x = afinfo->find_acq(mode, reqid, proto, daddr, saddr, create);
+	x = __find_acq_core(family, mode, reqid, proto, daddr, saddr, create);
 	spin_unlock_bh(&xfrm_state_lock);
-	xfrm_state_put_afinfo(afinfo);
+
 	return x;
 }
 EXPORT_SYMBOL(xfrm_find_acq);
@@ -1181,7 +1273,6 @@ int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo)
 	if (unlikely(xfrm_state_afinfo[afinfo->family] != NULL))
 		err = -ENOBUFS;
 	else {
-		afinfo->state_bydst = xfrm_state_bydst;
 		afinfo->state_bysrc = xfrm_state_bysrc;
 		afinfo->state_byspi = xfrm_state_byspi;
 		xfrm_state_afinfo[afinfo->family] = afinfo;
@@ -1206,7 +1297,6 @@ int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo)
 			xfrm_state_afinfo[afinfo->family] = NULL;
 			afinfo->state_byspi = NULL;
 			afinfo->state_bysrc = NULL;
-			afinfo->state_bydst = NULL;
 		}
 	}
 	write_unlock_bh(&xfrm_state_afinfo_lock);
-- 
cgit v1.2.3


From edcd582152090bfb0ccb4ad444c151798a73eda8 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 00:42:45 -0700
Subject: [XFRM]: Pull xfrm_state_by{spi,src} hash table knowledge out of
 afinfo.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  78 ------------------
 net/ipv4/xfrm4_state.c |  28 -------
 net/ipv6/xfrm6_state.c |  40 ----------
 net/xfrm/xfrm_state.c  | 210 +++++++++++++++++++++++++++++++++++++------------
 4 files changed, 159 insertions(+), 197 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index cc83443f301e..dd3b84b9c04e 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -243,14 +243,10 @@ extern int __xfrm_state_delete(struct xfrm_state *x);
 
 struct xfrm_state_afinfo {
 	unsigned short		family;
-	struct list_head	*state_bysrc;
-	struct list_head	*state_byspi;
 	int			(*init_flags)(struct xfrm_state *x);
 	void			(*init_tempsel)(struct xfrm_state *x, struct flowi *fl,
 						struct xfrm_tmpl *tmpl,
 						xfrm_address_t *daddr, xfrm_address_t *saddr);
-	struct xfrm_state	*(*state_lookup)(xfrm_address_t *daddr, u32 spi, u8 proto);
-	struct xfrm_state	*(*state_lookup_byaddr)(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto);
 	int			(*tmpl_sort)(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n);
 	int			(*state_sort)(struct xfrm_state **dst, struct xfrm_state **src, int n);
 };
@@ -431,80 +427,6 @@ static inline void xfrm_pols_put(struct xfrm_policy **pols, int npols)
 }
 #endif
 
-#define XFRM_DST_HSIZE		1024
-
-static __inline__
-unsigned __xfrm4_dst_hash(xfrm_address_t *addr)
-{
-	unsigned h;
-	h = ntohl(addr->a4);
-	h = (h ^ (h>>16)) % XFRM_DST_HSIZE;
-	return h;
-}
-
-static __inline__
-unsigned __xfrm6_dst_hash(xfrm_address_t *addr)
-{
-	unsigned h;
-	h = ntohl(addr->a6[2]^addr->a6[3]);
-	h = (h ^ (h>>16)) % XFRM_DST_HSIZE;
-	return h;
-}
-
-static __inline__
-unsigned __xfrm4_src_hash(xfrm_address_t *addr)
-{
-	return __xfrm4_dst_hash(addr);
-}
-
-static __inline__
-unsigned __xfrm6_src_hash(xfrm_address_t *addr)
-{
-	return __xfrm6_dst_hash(addr);
-}
-
-static __inline__
-unsigned xfrm_src_hash(xfrm_address_t *addr, unsigned short family)
-{
-	switch (family) {
-	case AF_INET:
-		return __xfrm4_src_hash(addr);
-	case AF_INET6:
-		return __xfrm6_src_hash(addr);
-	}
-	return 0;
-}
-
-static __inline__
-unsigned __xfrm4_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
-{
-	unsigned h;
-	h = ntohl(addr->a4^spi^proto);
-	h = (h ^ (h>>10) ^ (h>>20)) % XFRM_DST_HSIZE;
-	return h;
-}
-
-static __inline__
-unsigned __xfrm6_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
-{
-	unsigned h;
-	h = ntohl(addr->a6[2]^addr->a6[3]^spi^proto);
-	h = (h ^ (h>>10) ^ (h>>20)) % XFRM_DST_HSIZE;
-	return h;
-}
-
-static __inline__
-unsigned xfrm_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto, unsigned short family)
-{
-	switch (family) {
-	case AF_INET:
-		return __xfrm4_spi_hash(addr, spi, proto);
-	case AF_INET6:
-		return __xfrm6_spi_hash(addr, spi, proto);
-	}
-	return 0;	/*XXX*/
-}
-
 extern void __xfrm_state_destroy(struct xfrm_state *);
 
 static inline void __xfrm_state_put(struct xfrm_state *x)
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 9dc1afc17b6d..6a2a4ab42772 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -62,38 +62,10 @@ __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->props.family = AF_INET;
 }
 
-static struct xfrm_state *
-__xfrm4_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto)
-{
-	unsigned h = __xfrm4_spi_hash(daddr, spi, proto);
-	struct xfrm_state *x;
-
-	list_for_each_entry(x, xfrm4_state_afinfo.state_byspi+h, byspi) {
-		if (x->props.family == AF_INET &&
-		    spi == x->id.spi &&
-		    daddr->a4 == x->id.daddr.a4 &&
-		    proto == x->id.proto) {
-			xfrm_state_hold(x);
-			return x;
-		}
-	}
-	return NULL;
-}
-
-/* placeholder until ipv4's code is written */
-static struct xfrm_state *
-__xfrm4_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
-			    u8 proto)
-{
-	return NULL;
-}
-
 static struct xfrm_state_afinfo xfrm4_state_afinfo = {
 	.family			= AF_INET,
 	.init_flags		= xfrm4_init_flags,
 	.init_tempsel		= __xfrm4_init_tempsel,
-	.state_lookup		= __xfrm4_state_lookup,
-	.state_lookup_byaddr	= __xfrm4_state_lookup_byaddr,
 };
 
 void __init xfrm4_state_init(void)
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 40fcaab7e028..d88cd92c864e 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -63,44 +63,6 @@ __xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->props.family = AF_INET6;
 }
 
-static struct xfrm_state *
-__xfrm6_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
-			    u8 proto)
-{
-	struct xfrm_state *x = NULL;
-	unsigned h;
-
-	h = __xfrm6_src_hash(saddr);
-	list_for_each_entry(x, xfrm6_state_afinfo.state_bysrc+h, bysrc) {
-		if (x->props.family == AF_INET6 &&
-		    ipv6_addr_equal((struct in6_addr *)daddr, (struct in6_addr *)x->id.daddr.a6) &&
-		    ipv6_addr_equal((struct in6_addr *)saddr, (struct in6_addr *)x->props.saddr.a6) &&
-		    proto == x->id.proto) {
-			xfrm_state_hold(x);
-			return x;
-		}
-	}
-	return NULL;
-}
-
-static struct xfrm_state *
-__xfrm6_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto)
-{
-	unsigned h = __xfrm6_spi_hash(daddr, spi, proto);
-	struct xfrm_state *x;
-
-	list_for_each_entry(x, xfrm6_state_afinfo.state_byspi+h, byspi) {
-		if (x->props.family == AF_INET6 &&
-		    spi == x->id.spi &&
-		    ipv6_addr_equal((struct in6_addr *)daddr, (struct in6_addr *)x->id.daddr.a6) &&
-		    proto == x->id.proto) {
-			xfrm_state_hold(x);
-			return x;
-		}
-	}
-	return NULL;
-}
-
 static int
 __xfrm6_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n)
 {
@@ -223,8 +185,6 @@ __xfrm6_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n)
 static struct xfrm_state_afinfo xfrm6_state_afinfo = {
 	.family			= AF_INET6,
 	.init_tempsel		= __xfrm6_init_tempsel,
-	.state_lookup		= __xfrm6_state_lookup,
-	.state_lookup_byaddr	= __xfrm6_state_lookup_byaddr,
 	.tmpl_sort		= __xfrm6_tmpl_sort,
 	.state_sort		= __xfrm6_state_sort,
 };
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 80f5f9dc2b9e..4a3832f81c37 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -38,6 +38,8 @@ EXPORT_SYMBOL(sysctl_xfrm_aevent_rseqth);
 
 static DEFINE_SPINLOCK(xfrm_state_lock);
 
+#define XFRM_DST_HSIZE		1024
+
 /* Hash table to find appropriate SA towards given target (endpoint
  * of tunnel or destination of transport mode) allowed by selector.
  *
@@ -48,6 +50,48 @@ static struct list_head xfrm_state_bydst[XFRM_DST_HSIZE];
 static struct list_head xfrm_state_bysrc[XFRM_DST_HSIZE];
 static struct list_head xfrm_state_byspi[XFRM_DST_HSIZE];
 
+static __inline__
+unsigned __xfrm4_dst_hash(xfrm_address_t *addr)
+{
+	unsigned h;
+	h = ntohl(addr->a4);
+	h = (h ^ (h>>16)) % XFRM_DST_HSIZE;
+	return h;
+}
+
+static __inline__
+unsigned __xfrm6_dst_hash(xfrm_address_t *addr)
+{
+	unsigned h;
+	h = ntohl(addr->a6[2]^addr->a6[3]);
+	h = (h ^ (h>>16)) % XFRM_DST_HSIZE;
+	return h;
+}
+
+static __inline__
+unsigned __xfrm4_src_hash(xfrm_address_t *addr)
+{
+	return __xfrm4_dst_hash(addr);
+}
+
+static __inline__
+unsigned __xfrm6_src_hash(xfrm_address_t *addr)
+{
+	return __xfrm6_dst_hash(addr);
+}
+
+static __inline__
+unsigned xfrm_src_hash(xfrm_address_t *addr, unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		return __xfrm4_src_hash(addr);
+	case AF_INET6:
+		return __xfrm6_src_hash(addr);
+	}
+	return 0;
+}
+
 static __inline__
 unsigned xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
 {
@@ -60,6 +104,36 @@ unsigned xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
 	return 0;
 }
 
+static __inline__
+unsigned __xfrm4_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
+{
+	unsigned h;
+	h = ntohl(addr->a4^spi^proto);
+	h = (h ^ (h>>10) ^ (h>>20)) % XFRM_DST_HSIZE;
+	return h;
+}
+
+static __inline__
+unsigned __xfrm6_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
+{
+	unsigned h;
+	h = ntohl(addr->a6[2]^addr->a6[3]^spi^proto);
+	h = (h ^ (h>>10) ^ (h>>20)) % XFRM_DST_HSIZE;
+	return h;
+}
+
+static __inline__
+unsigned xfrm_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto, unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		return __xfrm4_spi_hash(addr, spi, proto);
+	case AF_INET6:
+		return __xfrm6_spi_hash(addr, spi, proto);
+	}
+	return 0;	/*XXX*/
+}
+
 DECLARE_WAIT_QUEUE_HEAD(km_waitq);
 EXPORT_SYMBOL(km_waitq);
 
@@ -342,6 +416,83 @@ xfrm_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	return 0;
 }
 
+static struct xfrm_state *__xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family)
+{
+	unsigned int h = xfrm_spi_hash(daddr, spi, proto, family);
+	struct xfrm_state *x;
+
+	list_for_each_entry(x, xfrm_state_byspi+h, byspi) {
+		if (x->props.family != family ||
+		    x->id.spi       != spi ||
+		    x->id.proto     != proto)
+			continue;
+
+		switch (family) {
+		case AF_INET:
+			if (x->id.daddr.a4 != daddr->a4)
+				continue;
+			break;
+		case AF_INET6:
+			if (!ipv6_addr_equal((struct in6_addr *)daddr,
+					     (struct in6_addr *)
+					     x->id.daddr.a6))
+				continue;
+			break;
+		};
+
+		xfrm_state_hold(x);
+		return x;
+	}
+
+	return NULL;
+}
+
+static struct xfrm_state *__xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family)
+{
+	unsigned int h = xfrm_src_hash(saddr, family);
+	struct xfrm_state *x;
+
+	list_for_each_entry(x, xfrm_state_bysrc+h, bysrc) {
+		if (x->props.family != family ||
+		    x->id.proto     != proto)
+			continue;
+
+		switch (family) {
+		case AF_INET:
+			if (x->id.daddr.a4 != daddr->a4 ||
+			    x->props.saddr.a4 != saddr->a4)
+				continue;
+			break;
+		case AF_INET6:
+			if (!ipv6_addr_equal((struct in6_addr *)daddr,
+					     (struct in6_addr *)
+					     x->id.daddr.a6) ||
+			    !ipv6_addr_equal((struct in6_addr *)saddr,
+					     (struct in6_addr *)
+					     x->props.saddr.a6))
+				continue;
+			break;
+		};
+
+		xfrm_state_hold(x);
+		return x;
+	}
+
+	return NULL;
+}
+
+static inline struct xfrm_state *
+__xfrm_state_locate(struct xfrm_state *x, int use_spi, int family)
+{
+	if (use_spi)
+		return __xfrm_state_lookup(&x->id.daddr, x->id.spi,
+					   x->id.proto, family);
+	else
+		return __xfrm_state_lookup_byaddr(&x->id.daddr,
+						  &x->props.saddr,
+						  x->id.proto, family);
+}
+
 struct xfrm_state *
 xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, 
 		struct flowi *fl, struct xfrm_tmpl *tmpl,
@@ -353,14 +504,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 	int acquire_in_progress = 0;
 	int error = 0;
 	struct xfrm_state *best = NULL;
-	struct xfrm_state_afinfo *afinfo;
 	
-	afinfo = xfrm_state_get_afinfo(family);
-	if (afinfo == NULL) {
-		*err = -EAFNOSUPPORT;
-		return NULL;
-	}
-
 	spin_lock_bh(&xfrm_state_lock);
 	list_for_each_entry(x, xfrm_state_bydst+h, bydst) {
 		if (x->props.family == family &&
@@ -406,8 +550,8 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 	x = best;
 	if (!x && !error && !acquire_in_progress) {
 		if (tmpl->id.spi &&
-		    (x0 = afinfo->state_lookup(daddr, tmpl->id.spi,
-		                               tmpl->id.proto)) != NULL) {
+		    (x0 = __xfrm_state_lookup(daddr, tmpl->id.spi,
+					      tmpl->id.proto, family)) != NULL) {
 			xfrm_state_put(x0);
 			error = -EEXIST;
 			goto out;
@@ -457,7 +601,6 @@ out:
 	else
 		*err = acquire_in_progress ? -EAGAIN : error;
 	spin_unlock_bh(&xfrm_state_lock);
-	xfrm_state_put_afinfo(afinfo);
 	return x;
 }
 
@@ -584,34 +727,20 @@ static struct xfrm_state *__find_acq_core(unsigned short family, u8 mode, u32 re
 	return x;
 }
 
-static inline struct xfrm_state *
-__xfrm_state_locate(struct xfrm_state_afinfo *afinfo, struct xfrm_state *x,
-		    int use_spi)
-{
-	if (use_spi)
-		return afinfo->state_lookup(&x->id.daddr, x->id.spi, x->id.proto);
-	else
-		return afinfo->state_lookup_byaddr(&x->id.daddr, &x->props.saddr, x->id.proto);
-}
-
 static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq);
 
 int xfrm_state_add(struct xfrm_state *x)
 {
-	struct xfrm_state_afinfo *afinfo;
 	struct xfrm_state *x1;
 	int family;
 	int err;
 	int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY);
 
 	family = x->props.family;
-	afinfo = xfrm_state_get_afinfo(family);
-	if (unlikely(afinfo == NULL))
-		return -EAFNOSUPPORT;
 
 	spin_lock_bh(&xfrm_state_lock);
 
-	x1 = __xfrm_state_locate(afinfo, x, use_spi);
+	x1 = __xfrm_state_locate(x, use_spi, family);
 	if (x1) {
 		xfrm_state_put(x1);
 		x1 = NULL;
@@ -637,7 +766,6 @@ int xfrm_state_add(struct xfrm_state *x)
 
 out:
 	spin_unlock_bh(&xfrm_state_lock);
-	xfrm_state_put_afinfo(afinfo);
 
 	if (!err)
 		xfrm_flush_all_bundles();
@@ -653,17 +781,12 @@ EXPORT_SYMBOL(xfrm_state_add);
 
 int xfrm_state_update(struct xfrm_state *x)
 {
-	struct xfrm_state_afinfo *afinfo;
 	struct xfrm_state *x1;
 	int err;
 	int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY);
 
-	afinfo = xfrm_state_get_afinfo(x->props.family);
-	if (unlikely(afinfo == NULL))
-		return -EAFNOSUPPORT;
-
 	spin_lock_bh(&xfrm_state_lock);
-	x1 = __xfrm_state_locate(afinfo, x, use_spi);
+	x1 = __xfrm_state_locate(x, use_spi, x->props.family);
 
 	err = -ESRCH;
 	if (!x1)
@@ -683,7 +806,6 @@ int xfrm_state_update(struct xfrm_state *x)
 
 out:
 	spin_unlock_bh(&xfrm_state_lock);
-	xfrm_state_put_afinfo(afinfo);
 
 	if (err)
 		return err;
@@ -776,14 +898,10 @@ xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto,
 		  unsigned short family)
 {
 	struct xfrm_state *x;
-	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
-	if (!afinfo)
-		return NULL;
 
 	spin_lock_bh(&xfrm_state_lock);
-	x = afinfo->state_lookup(daddr, spi, proto);
+	x = __xfrm_state_lookup(daddr, spi, proto, family);
 	spin_unlock_bh(&xfrm_state_lock);
-	xfrm_state_put_afinfo(afinfo);
 	return x;
 }
 EXPORT_SYMBOL(xfrm_state_lookup);
@@ -793,14 +911,10 @@ xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr,
 			 u8 proto, unsigned short family)
 {
 	struct xfrm_state *x;
-	struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
-	if (!afinfo)
-		return NULL;
 
 	spin_lock_bh(&xfrm_state_lock);
-	x = afinfo->state_lookup_byaddr(daddr, saddr, proto);
+	x = __xfrm_state_lookup_byaddr(daddr, saddr, proto, family);
 	spin_unlock_bh(&xfrm_state_lock);
-	xfrm_state_put_afinfo(afinfo);
 	return x;
 }
 EXPORT_SYMBOL(xfrm_state_lookup_byaddr);
@@ -1272,11 +1386,8 @@ int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo)
 	write_lock_bh(&xfrm_state_afinfo_lock);
 	if (unlikely(xfrm_state_afinfo[afinfo->family] != NULL))
 		err = -ENOBUFS;
-	else {
-		afinfo->state_bysrc = xfrm_state_bysrc;
-		afinfo->state_byspi = xfrm_state_byspi;
+	else
 		xfrm_state_afinfo[afinfo->family] = afinfo;
-	}
 	write_unlock_bh(&xfrm_state_afinfo_lock);
 	return err;
 }
@@ -1293,11 +1404,8 @@ int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo)
 	if (likely(xfrm_state_afinfo[afinfo->family] != NULL)) {
 		if (unlikely(xfrm_state_afinfo[afinfo->family] != afinfo))
 			err = -EINVAL;
-		else {
+		else
 			xfrm_state_afinfo[afinfo->family] = NULL;
-			afinfo->state_byspi = NULL;
-			afinfo->state_bysrc = NULL;
-		}
 	}
 	write_unlock_bh(&xfrm_state_afinfo_lock);
 	return err;
-- 
cgit v1.2.3


From 8f126e37c0b250310a48a609bedf92a19a5559ec Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 02:45:07 -0700
Subject: [XFRM]: Convert xfrm_state hash linkage to hlists.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    |  6 ++--
 net/xfrm/xfrm_state.c | 92 ++++++++++++++++++++++++++++-----------------------
 2 files changed, 54 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index dd3b84b9c04e..3405e5d9d51c 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -94,9 +94,9 @@ extern struct mutex xfrm_cfg_mutex;
 struct xfrm_state
 {
 	/* Note: bydst is re-used during gc */
-	struct list_head	bydst;
-	struct list_head	bysrc;
-	struct list_head	byspi;
+	struct hlist_node	bydst;
+	struct hlist_node	bysrc;
+	struct hlist_node	byspi;
 
 	atomic_t		refcnt;
 	spinlock_t		lock;
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 4a3832f81c37..fe3c8c38d5e1 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -46,9 +46,9 @@ static DEFINE_SPINLOCK(xfrm_state_lock);
  * Main use is finding SA after policy selected tunnel or transport mode.
  * Also, it can be used by ah/esp icmp error handler to find offending SA.
  */
-static struct list_head xfrm_state_bydst[XFRM_DST_HSIZE];
-static struct list_head xfrm_state_bysrc[XFRM_DST_HSIZE];
-static struct list_head xfrm_state_byspi[XFRM_DST_HSIZE];
+static struct hlist_head xfrm_state_bydst[XFRM_DST_HSIZE];
+static struct hlist_head xfrm_state_bysrc[XFRM_DST_HSIZE];
+static struct hlist_head xfrm_state_byspi[XFRM_DST_HSIZE];
 
 static __inline__
 unsigned __xfrm4_dst_hash(xfrm_address_t *addr)
@@ -141,7 +141,7 @@ static DEFINE_RWLOCK(xfrm_state_afinfo_lock);
 static struct xfrm_state_afinfo *xfrm_state_afinfo[NPROTO];
 
 static struct work_struct xfrm_state_gc_work;
-static struct list_head xfrm_state_gc_list = LIST_HEAD_INIT(xfrm_state_gc_list);
+static HLIST_HEAD(xfrm_state_gc_list);
 static DEFINE_SPINLOCK(xfrm_state_gc_lock);
 
 static int xfrm_state_gc_flush_bundles;
@@ -178,8 +178,8 @@ static void xfrm_state_gc_destroy(struct xfrm_state *x)
 static void xfrm_state_gc_task(void *data)
 {
 	struct xfrm_state *x;
-	struct list_head *entry, *tmp;
-	struct list_head gc_list = LIST_HEAD_INIT(gc_list);
+	struct hlist_node *entry, *tmp;
+	struct hlist_head gc_list;
 
 	if (xfrm_state_gc_flush_bundles) {
 		xfrm_state_gc_flush_bundles = 0;
@@ -187,13 +187,13 @@ static void xfrm_state_gc_task(void *data)
 	}
 
 	spin_lock_bh(&xfrm_state_gc_lock);
-	list_splice_init(&xfrm_state_gc_list, &gc_list);
+	gc_list.first = xfrm_state_gc_list.first;
+	INIT_HLIST_HEAD(&xfrm_state_gc_list);
 	spin_unlock_bh(&xfrm_state_gc_lock);
 
-	list_for_each_safe(entry, tmp, &gc_list) {
-		x = list_entry(entry, struct xfrm_state, bydst);
+	hlist_for_each_entry_safe(x, entry, tmp, &gc_list, bydst)
 		xfrm_state_gc_destroy(x);
-	}
+
 	wake_up(&km_waitq);
 }
 
@@ -287,9 +287,9 @@ struct xfrm_state *xfrm_state_alloc(void)
 	if (x) {
 		atomic_set(&x->refcnt, 1);
 		atomic_set(&x->tunnel_users, 0);
-		INIT_LIST_HEAD(&x->bydst);
-		INIT_LIST_HEAD(&x->bysrc);
-		INIT_LIST_HEAD(&x->byspi);
+		INIT_HLIST_NODE(&x->bydst);
+		INIT_HLIST_NODE(&x->bysrc);
+		INIT_HLIST_NODE(&x->byspi);
 		init_timer(&x->timer);
 		x->timer.function = xfrm_timer_handler;
 		x->timer.data	  = (unsigned long)x;
@@ -314,7 +314,7 @@ void __xfrm_state_destroy(struct xfrm_state *x)
 	BUG_TRAP(x->km.state == XFRM_STATE_DEAD);
 
 	spin_lock_bh(&xfrm_state_gc_lock);
-	list_add(&x->bydst, &xfrm_state_gc_list);
+	hlist_add_head(&x->bydst, &xfrm_state_gc_list);
 	spin_unlock_bh(&xfrm_state_gc_lock);
 	schedule_work(&xfrm_state_gc_work);
 }
@@ -327,12 +327,12 @@ int __xfrm_state_delete(struct xfrm_state *x)
 	if (x->km.state != XFRM_STATE_DEAD) {
 		x->km.state = XFRM_STATE_DEAD;
 		spin_lock(&xfrm_state_lock);
-		list_del(&x->bydst);
+		hlist_del(&x->bydst);
 		__xfrm_state_put(x);
-		list_del(&x->bysrc);
+		hlist_del(&x->bysrc);
 		__xfrm_state_put(x);
 		if (x->id.spi) {
-			list_del(&x->byspi);
+			hlist_del(&x->byspi);
 			__xfrm_state_put(x);
 		}
 		spin_unlock(&xfrm_state_lock);
@@ -378,12 +378,13 @@ EXPORT_SYMBOL(xfrm_state_delete);
 void xfrm_state_flush(u8 proto)
 {
 	int i;
-	struct xfrm_state *x;
 
 	spin_lock_bh(&xfrm_state_lock);
 	for (i = 0; i < XFRM_DST_HSIZE; i++) {
+		struct hlist_node *entry;
+		struct xfrm_state *x;
 restart:
-		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
+		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
 			if (!xfrm_state_kern(x) &&
 			    xfrm_id_proto_match(x->id.proto, proto)) {
 				xfrm_state_hold(x);
@@ -420,8 +421,9 @@ static struct xfrm_state *__xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8
 {
 	unsigned int h = xfrm_spi_hash(daddr, spi, proto, family);
 	struct xfrm_state *x;
+	struct hlist_node *entry;
 
-	list_for_each_entry(x, xfrm_state_byspi+h, byspi) {
+	hlist_for_each_entry(x, entry, xfrm_state_byspi+h, byspi) {
 		if (x->props.family != family ||
 		    x->id.spi       != spi ||
 		    x->id.proto     != proto)
@@ -451,8 +453,9 @@ static struct xfrm_state *__xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm
 {
 	unsigned int h = xfrm_src_hash(saddr, family);
 	struct xfrm_state *x;
+	struct hlist_node *entry;
 
-	list_for_each_entry(x, xfrm_state_bysrc+h, bysrc) {
+	hlist_for_each_entry(x, entry, xfrm_state_bysrc+h, bysrc) {
 		if (x->props.family != family ||
 		    x->id.proto     != proto)
 			continue;
@@ -499,14 +502,15 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 		struct xfrm_policy *pol, int *err,
 		unsigned short family)
 {
-	unsigned h = xfrm_dst_hash(daddr, family);
+	unsigned int h = xfrm_dst_hash(daddr, family);
+	struct hlist_node *entry;
 	struct xfrm_state *x, *x0;
 	int acquire_in_progress = 0;
 	int error = 0;
 	struct xfrm_state *best = NULL;
 	
 	spin_lock_bh(&xfrm_state_lock);
-	list_for_each_entry(x, xfrm_state_bydst+h, bydst) {
+	hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) {
 		if (x->props.family == family &&
 		    x->props.reqid == tmpl->reqid &&
 		    !(x->props.flags & XFRM_STATE_WILDRECV) &&
@@ -575,13 +579,14 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
 
 		if (km_query(x, tmpl, pol) == 0) {
 			x->km.state = XFRM_STATE_ACQ;
-			list_add_tail(&x->bydst, xfrm_state_bydst+h);
+			hlist_add_head(&x->bydst, xfrm_state_bydst+h);
 			xfrm_state_hold(x);
-			list_add_tail(&x->bysrc, xfrm_state_bysrc+h);
+			h = xfrm_src_hash(saddr, family);
+			hlist_add_head(&x->bysrc, xfrm_state_bysrc+h);
 			xfrm_state_hold(x);
 			if (x->id.spi) {
 				h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto, family);
-				list_add(&x->byspi, xfrm_state_byspi+h);
+				hlist_add_head(&x->byspi, xfrm_state_byspi+h);
 				xfrm_state_hold(x);
 			}
 			x->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES;
@@ -608,19 +613,19 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 {
 	unsigned h = xfrm_dst_hash(&x->id.daddr, x->props.family);
 
-	list_add(&x->bydst, xfrm_state_bydst+h);
+	hlist_add_head(&x->bydst, xfrm_state_bydst+h);
 	xfrm_state_hold(x);
 
 	h = xfrm_src_hash(&x->props.saddr, x->props.family);
 
-	list_add(&x->bysrc, xfrm_state_bysrc+h);
+	hlist_add_head(&x->bysrc, xfrm_state_bysrc+h);
 	xfrm_state_hold(x);
 
 	if (xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY)) {
 		h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto,
 				  x->props.family);
 
-		list_add(&x->byspi, xfrm_state_byspi+h);
+		hlist_add_head(&x->byspi, xfrm_state_byspi+h);
 		xfrm_state_hold(x);
 	}
 
@@ -648,9 +653,10 @@ EXPORT_SYMBOL(xfrm_state_insert);
 static struct xfrm_state *__find_acq_core(unsigned short family, u8 mode, u32 reqid, u8 proto, xfrm_address_t *daddr, xfrm_address_t *saddr, int create)
 {
 	unsigned int h = xfrm_dst_hash(daddr, family);
+	struct hlist_node *entry;
 	struct xfrm_state *x;
 
-	list_for_each_entry(x, xfrm_state_bydst+h, bydst) {
+	hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) {
 		if (x->props.reqid  != reqid ||
 		    x->props.mode   != mode ||
 		    x->props.family != family ||
@@ -717,10 +723,10 @@ static struct xfrm_state *__find_acq_core(unsigned short family, u8 mode, u32 re
 		x->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ;
 		add_timer(&x->timer);
 		xfrm_state_hold(x);
-		list_add_tail(&x->bydst, xfrm_state_bydst+h);
+		hlist_add_head(&x->bydst, xfrm_state_bydst+h);
 		h = xfrm_src_hash(saddr, family);
 		xfrm_state_hold(x);
-		list_add_tail(&x->bysrc, xfrm_state_bysrc+h);
+		hlist_add_head(&x->bysrc, xfrm_state_bysrc+h);
 		wake_up(&km_waitq);
 	}
 
@@ -977,11 +983,14 @@ EXPORT_SYMBOL(xfrm_state_sort);
 static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq)
 {
 	int i;
-	struct xfrm_state *x;
 
 	for (i = 0; i < XFRM_DST_HSIZE; i++) {
-		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
-			if (x->km.seq == seq && x->km.state == XFRM_STATE_ACQ) {
+		struct hlist_node *entry;
+		struct xfrm_state *x;
+
+		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
+			if (x->km.seq == seq &&
+			    x->km.state == XFRM_STATE_ACQ) {
 				xfrm_state_hold(x);
 				return x;
 			}
@@ -1047,7 +1056,7 @@ xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi)
 	if (x->id.spi) {
 		spin_lock_bh(&xfrm_state_lock);
 		h = xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto, x->props.family);
-		list_add(&x->byspi, xfrm_state_byspi+h);
+		hlist_add_head(&x->byspi, xfrm_state_byspi+h);
 		xfrm_state_hold(x);
 		spin_unlock_bh(&xfrm_state_lock);
 		wake_up(&km_waitq);
@@ -1060,12 +1069,13 @@ int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
 {
 	int i;
 	struct xfrm_state *x;
+	struct hlist_node *entry;
 	int count = 0;
 	int err = 0;
 
 	spin_lock_bh(&xfrm_state_lock);
 	for (i = 0; i < XFRM_DST_HSIZE; i++) {
-		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
+		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
 			if (xfrm_id_proto_match(x->id.proto, proto))
 				count++;
 		}
@@ -1076,7 +1086,7 @@ int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
 	}
 
 	for (i = 0; i < XFRM_DST_HSIZE; i++) {
-		list_for_each_entry(x, xfrm_state_bydst+i, bydst) {
+		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
 			if (!xfrm_id_proto_match(x->id.proto, proto))
 				continue;
 			err = func(x, --count, data);
@@ -1524,9 +1534,9 @@ void __init xfrm_state_init(void)
 	int i;
 
 	for (i=0; i<XFRM_DST_HSIZE; i++) {
-		INIT_LIST_HEAD(&xfrm_state_bydst[i]);
-		INIT_LIST_HEAD(&xfrm_state_bysrc[i]);
-		INIT_LIST_HEAD(&xfrm_state_byspi[i]);
+		INIT_HLIST_HEAD(&xfrm_state_bydst[i]);
+		INIT_HLIST_HEAD(&xfrm_state_bysrc[i]);
+		INIT_HLIST_HEAD(&xfrm_state_byspi[i]);
 	}
 	INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task, NULL);
 }
-- 
cgit v1.2.3


From f034b5d4efdfe0fb9e2a1ce1d95fa7914f24de49 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 03:08:07 -0700
Subject: [XFRM]: Dynamic xfrm_state hash table sizing.

The grow algorithm is simple, we grow if:

1) we see a hash chain collision at insert, and
2) we haven't hit the hash size limit (currently 1*1024*1024 slots), and
3) the number of xfrm_state objects is > the current hash mask

All of this needs some tweaking.

Remove __initdata from "hashdist" so we can use it safely at run time.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/bootmem.h |   2 +-
 mm/page_alloc.c         |   2 +-
 net/xfrm/xfrm_state.c   | 247 ++++++++++++++++++++++++++++++++++++++----------
 3 files changed, 197 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 1021f508d82c..e319c649e4fd 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -114,7 +114,7 @@ extern void *__init alloc_large_system_hash(const char *tablename,
 #else
 #define HASHDIST_DEFAULT 0
 #endif
-extern int __initdata hashdist;		/* Distribute hashes across NUMA nodes? */
+extern int hashdist;		/* Distribute hashes across NUMA nodes? */
 
 
 #endif /* _LINUX_BOOTMEM_H */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 54a4f5375bba..3b5358a0561f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2363,7 +2363,7 @@ int percpu_pagelist_fraction_sysctl_handler(ctl_table *table, int write,
 	return 0;
 }
 
-__initdata int hashdist = HASHDIST_DEFAULT;
+int hashdist = HASHDIST_DEFAULT;
 
 #ifdef CONFIG_NUMA
 static int __init set_hashdist(char *str)
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index fe3c8c38d5e1..445263c54c94 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -18,6 +18,9 @@
 #include <linux/pfkeyv2.h>
 #include <linux/ipsec.h>
 #include <linux/module.h>
+#include <linux/bootmem.h>
+#include <linux/vmalloc.h>
+#include <linux/cache.h>
 #include <asm/uaccess.h>
 
 struct sock *xfrm_nl;
@@ -38,102 +41,230 @@ EXPORT_SYMBOL(sysctl_xfrm_aevent_rseqth);
 
 static DEFINE_SPINLOCK(xfrm_state_lock);
 
-#define XFRM_DST_HSIZE		1024
-
 /* Hash table to find appropriate SA towards given target (endpoint
  * of tunnel or destination of transport mode) allowed by selector.
  *
  * Main use is finding SA after policy selected tunnel or transport mode.
  * Also, it can be used by ah/esp icmp error handler to find offending SA.
  */
-static struct hlist_head xfrm_state_bydst[XFRM_DST_HSIZE];
-static struct hlist_head xfrm_state_bysrc[XFRM_DST_HSIZE];
-static struct hlist_head xfrm_state_byspi[XFRM_DST_HSIZE];
-
-static __inline__
-unsigned __xfrm4_dst_hash(xfrm_address_t *addr)
+static struct hlist_head *xfrm_state_bydst __read_mostly;
+static struct hlist_head *xfrm_state_bysrc __read_mostly;
+static struct hlist_head *xfrm_state_byspi __read_mostly;
+static unsigned int xfrm_state_hmask __read_mostly;
+static unsigned int xfrm_state_hashmax __read_mostly = 1 * 1024 * 1024;
+static unsigned int xfrm_state_num;
+
+static inline unsigned int __xfrm4_dst_hash(xfrm_address_t *addr, unsigned int hmask)
 {
-	unsigned h;
+	unsigned int h;
 	h = ntohl(addr->a4);
-	h = (h ^ (h>>16)) % XFRM_DST_HSIZE;
+	h = (h ^ (h>>16)) & hmask;
 	return h;
 }
 
-static __inline__
-unsigned __xfrm6_dst_hash(xfrm_address_t *addr)
+static inline unsigned int __xfrm6_dst_hash(xfrm_address_t *addr, unsigned int hmask)
 {
-	unsigned h;
+	unsigned int h;
 	h = ntohl(addr->a6[2]^addr->a6[3]);
-	h = (h ^ (h>>16)) % XFRM_DST_HSIZE;
+	h = (h ^ (h>>16)) & hmask;
 	return h;
 }
 
-static __inline__
-unsigned __xfrm4_src_hash(xfrm_address_t *addr)
+static inline unsigned int __xfrm4_src_hash(xfrm_address_t *addr, unsigned int hmask)
 {
-	return __xfrm4_dst_hash(addr);
+	return __xfrm4_dst_hash(addr, hmask);
 }
 
-static __inline__
-unsigned __xfrm6_src_hash(xfrm_address_t *addr)
+static inline unsigned int __xfrm6_src_hash(xfrm_address_t *addr, unsigned int hmask)
 {
-	return __xfrm6_dst_hash(addr);
+	return __xfrm6_dst_hash(addr, hmask);
 }
 
-static __inline__
-unsigned xfrm_src_hash(xfrm_address_t *addr, unsigned short family)
+static inline unsigned __xfrm_src_hash(xfrm_address_t *addr, unsigned short family,  unsigned int hmask)
 {
 	switch (family) {
 	case AF_INET:
-		return __xfrm4_src_hash(addr);
+		return __xfrm4_src_hash(addr, hmask);
 	case AF_INET6:
-		return __xfrm6_src_hash(addr);
+		return __xfrm6_src_hash(addr, hmask);
 	}
 	return 0;
 }
 
-static __inline__
-unsigned xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
+static inline unsigned xfrm_src_hash(xfrm_address_t *addr, unsigned short family)
+{
+	return __xfrm_src_hash(addr, family, xfrm_state_hmask);
+}
+
+static inline unsigned int __xfrm_dst_hash(xfrm_address_t *addr, unsigned short family, unsigned int hmask)
 {
 	switch (family) {
 	case AF_INET:
-		return __xfrm4_dst_hash(addr);
+		return __xfrm4_dst_hash(addr, hmask);
 	case AF_INET6:
-		return __xfrm6_dst_hash(addr);
+		return __xfrm6_dst_hash(addr, hmask);
 	}
 	return 0;
 }
 
-static __inline__
-unsigned __xfrm4_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
+static inline unsigned int xfrm_dst_hash(xfrm_address_t *addr, unsigned short family)
+{
+	return __xfrm_dst_hash(addr, family, xfrm_state_hmask);
+}
+
+static inline unsigned int __xfrm4_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto,
+					unsigned int hmask)
 {
-	unsigned h;
+	unsigned int h;
 	h = ntohl(addr->a4^spi^proto);
-	h = (h ^ (h>>10) ^ (h>>20)) % XFRM_DST_HSIZE;
+	h = (h ^ (h>>10) ^ (h>>20)) & hmask;
 	return h;
 }
 
-static __inline__
-unsigned __xfrm6_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto)
+static inline unsigned int __xfrm6_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto,
+					    unsigned int hmask)
 {
-	unsigned h;
+	unsigned int h;
 	h = ntohl(addr->a6[2]^addr->a6[3]^spi^proto);
-	h = (h ^ (h>>10) ^ (h>>20)) % XFRM_DST_HSIZE;
+	h = (h ^ (h>>10) ^ (h>>20)) & hmask;
 	return h;
 }
 
-static __inline__
-unsigned xfrm_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto, unsigned short family)
+static inline
+unsigned __xfrm_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto, unsigned short family,
+			 unsigned int hmask)
 {
 	switch (family) {
 	case AF_INET:
-		return __xfrm4_spi_hash(addr, spi, proto);
+		return __xfrm4_spi_hash(addr, spi, proto, hmask);
 	case AF_INET6:
-		return __xfrm6_spi_hash(addr, spi, proto);
+		return __xfrm6_spi_hash(addr, spi, proto, hmask);
 	}
 	return 0;	/*XXX*/
 }
 
+static inline unsigned int
+xfrm_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto, unsigned short family)
+{
+	return __xfrm_spi_hash(addr, spi, proto, family, xfrm_state_hmask);
+}
+
+static struct hlist_head *xfrm_state_hash_alloc(unsigned int sz)
+{
+	struct hlist_head *n;
+
+	if (sz <= PAGE_SIZE)
+		n = kmalloc(sz, GFP_KERNEL);
+	else if (hashdist)
+		n = __vmalloc(sz, GFP_KERNEL, PAGE_KERNEL);
+	else
+		n = (struct hlist_head *)
+			__get_free_pages(GFP_KERNEL, get_order(sz));
+
+	if (n)
+		memset(n, 0, sz);
+
+	return n;
+}
+
+static void xfrm_state_hash_free(struct hlist_head *n, unsigned int sz)
+{
+	if (sz <= PAGE_SIZE)
+		kfree(n);
+	else if (hashdist)
+		vfree(n);
+	else
+		free_pages((unsigned long)n, get_order(sz));
+}
+
+static void xfrm_hash_transfer(struct hlist_head *list,
+			       struct hlist_head *ndsttable,
+			       struct hlist_head *nsrctable,
+			       struct hlist_head *nspitable,
+			       unsigned int nhashmask)
+{
+	struct hlist_node *entry, *tmp;
+	struct xfrm_state *x;
+
+	hlist_for_each_entry_safe(x, entry, tmp, list, bydst) {
+		unsigned int h;
+
+		h = __xfrm_dst_hash(&x->id.daddr, x->props.family, nhashmask);
+		hlist_add_head(&x->bydst, ndsttable+h);
+
+		h = __xfrm_src_hash(&x->props.saddr, x->props.family,
+				    nhashmask);
+		hlist_add_head(&x->bysrc, nsrctable+h);
+
+		h = __xfrm_spi_hash(&x->id.daddr, x->id.spi, x->id.proto,
+				    x->props.family, nhashmask);
+		hlist_add_head(&x->byspi, nspitable+h);
+	}
+}
+
+static unsigned long xfrm_hash_new_size(void)
+{
+	return ((xfrm_state_hmask + 1) << 1) *
+		sizeof(struct hlist_head);
+}
+
+static DEFINE_MUTEX(hash_resize_mutex);
+
+static void xfrm_hash_resize(void *__unused)
+{
+	struct hlist_head *ndst, *nsrc, *nspi, *odst, *osrc, *ospi;
+	unsigned long nsize, osize;
+	unsigned int nhashmask, ohashmask;
+	int i;
+
+	mutex_lock(&hash_resize_mutex);
+
+	nsize = xfrm_hash_new_size();
+	ndst = xfrm_state_hash_alloc(nsize);
+	if (!ndst)
+		goto out_unlock;
+	nsrc = xfrm_state_hash_alloc(nsize);
+	if (!nsrc) {
+		xfrm_state_hash_free(ndst, nsize);
+		goto out_unlock;
+	}
+	nspi = xfrm_state_hash_alloc(nsize);
+	if (!nspi) {
+		xfrm_state_hash_free(ndst, nsize);
+		xfrm_state_hash_free(nsrc, nsize);
+		goto out_unlock;
+	}
+
+	spin_lock_bh(&xfrm_state_lock);
+
+	nhashmask = (nsize / sizeof(struct hlist_head)) - 1U;
+	for (i = xfrm_state_hmask; i >= 0; i--)
+		xfrm_hash_transfer(xfrm_state_bydst+i, ndst, nsrc, nspi,
+				   nhashmask);
+
+	odst = xfrm_state_bydst;
+	osrc = xfrm_state_bysrc;
+	ospi = xfrm_state_byspi;
+	ohashmask = xfrm_state_hmask;
+
+	xfrm_state_bydst = ndst;
+	xfrm_state_bysrc = nsrc;
+	xfrm_state_byspi = nspi;
+	xfrm_state_hmask = nhashmask;
+
+	spin_unlock_bh(&xfrm_state_lock);
+
+	osize = (ohashmask + 1) * sizeof(struct hlist_head);
+	xfrm_state_hash_free(odst, osize);
+	xfrm_state_hash_free(osrc, osize);
+	xfrm_state_hash_free(ospi, osize);
+
+out_unlock:
+	mutex_unlock(&hash_resize_mutex);
+}
+
+static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL);
+
 DECLARE_WAIT_QUEUE_HEAD(km_waitq);
 EXPORT_SYMBOL(km_waitq);
 
@@ -335,6 +466,7 @@ int __xfrm_state_delete(struct xfrm_state *x)
 			hlist_del(&x->byspi);
 			__xfrm_state_put(x);
 		}
+		xfrm_state_num--;
 		spin_unlock(&xfrm_state_lock);
 		if (del_timer(&x->timer))
 			__xfrm_state_put(x);
@@ -380,7 +512,7 @@ void xfrm_state_flush(u8 proto)
 	int i;
 
 	spin_lock_bh(&xfrm_state_lock);
-	for (i = 0; i < XFRM_DST_HSIZE; i++) {
+	for (i = 0; i < xfrm_state_hmask; i++) {
 		struct hlist_node *entry;
 		struct xfrm_state *x;
 restart:
@@ -611,7 +743,7 @@ out:
 
 static void __xfrm_state_insert(struct xfrm_state *x)
 {
-	unsigned h = xfrm_dst_hash(&x->id.daddr, x->props.family);
+	unsigned int h = xfrm_dst_hash(&x->id.daddr, x->props.family);
 
 	hlist_add_head(&x->bydst, xfrm_state_bydst+h);
 	xfrm_state_hold(x);
@@ -637,6 +769,13 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 		xfrm_state_hold(x);
 
 	wake_up(&km_waitq);
+
+	xfrm_state_num++;
+
+	if (x->bydst.next != NULL &&
+	    (xfrm_state_hmask + 1) < xfrm_state_hashmax &&
+	    xfrm_state_num > xfrm_state_hmask)
+		schedule_work(&xfrm_hash_work);
 }
 
 void xfrm_state_insert(struct xfrm_state *x)
@@ -984,7 +1123,7 @@ static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq)
 {
 	int i;
 
-	for (i = 0; i < XFRM_DST_HSIZE; i++) {
+	for (i = 0; i <= xfrm_state_hmask; i++) {
 		struct hlist_node *entry;
 		struct xfrm_state *x;
 
@@ -1026,7 +1165,7 @@ EXPORT_SYMBOL(xfrm_get_acqseq);
 void
 xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi)
 {
-	u32 h;
+	unsigned int h;
 	struct xfrm_state *x0;
 
 	if (x->id.spi)
@@ -1074,7 +1213,7 @@ int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
 	int err = 0;
 
 	spin_lock_bh(&xfrm_state_lock);
-	for (i = 0; i < XFRM_DST_HSIZE; i++) {
+	for (i = 0; i <= xfrm_state_hmask; i++) {
 		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
 			if (xfrm_id_proto_match(x->id.proto, proto))
 				count++;
@@ -1085,7 +1224,7 @@ int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*),
 		goto out;
 	}
 
-	for (i = 0; i < XFRM_DST_HSIZE; i++) {
+	for (i = 0; i <= xfrm_state_hmask; i++) {
 		hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) {
 			if (!xfrm_id_proto_match(x->id.proto, proto))
 				continue;
@@ -1531,13 +1670,17 @@ EXPORT_SYMBOL(xfrm_init_state);
  
 void __init xfrm_state_init(void)
 {
-	int i;
+	unsigned int sz;
+
+	sz = sizeof(struct hlist_head) * 8;
+
+	xfrm_state_bydst = xfrm_state_hash_alloc(sz);
+	xfrm_state_bysrc = xfrm_state_hash_alloc(sz);
+	xfrm_state_byspi = xfrm_state_hash_alloc(sz);
+	if (!xfrm_state_bydst || !xfrm_state_bysrc || !xfrm_state_byspi)
+		panic("XFRM: Cannot allocate bydst/bysrc/byspi hashes.");
+	xfrm_state_hmask = ((sz / sizeof(struct hlist_head)) - 1);
 
-	for (i=0; i<XFRM_DST_HSIZE; i++) {
-		INIT_HLIST_HEAD(&xfrm_state_bydst[i]);
-		INIT_HLIST_HEAD(&xfrm_state_bysrc[i]);
-		INIT_HLIST_HEAD(&xfrm_state_byspi[i]);
-	}
 	INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task, NULL);
 }
 
-- 
cgit v1.2.3


From 9d4a706d852411154d0c91b9ffb3bec68b94b25c Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 03:18:09 -0700
Subject: [XFRM]: Add generation count to xfrm_state and xfrm_dst.

Each xfrm_state inserted gets a new generation counter
value.  When a bundle is created, the xfrm_dst objects
get the current generation counter of the xfrm_state
they will attach to at dst->xfrm.

xfrm_bundle_ok() will return false if it sees an
xfrm_dst with a generation count different from the
generation count of the xfrm_state that dst points to.

This provides a facility by which to passively and
cheaply invalidate cached IPSEC routes during SA
database changes.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      | 3 +++
 net/ipv4/xfrm4_policy.c | 1 +
 net/ipv6/xfrm6_policy.c | 1 +
 net/xfrm/xfrm_policy.c  | 2 ++
 net/xfrm/xfrm_state.c   | 3 +++
 5 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 3405e5d9d51c..fd4a300b5baf 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -104,6 +104,8 @@ struct xfrm_state
 	struct xfrm_id		id;
 	struct xfrm_selector	sel;
 
+	u32			genid;
+
 	/* Key manger bits */
 	struct {
 		u8		state;
@@ -590,6 +592,7 @@ struct xfrm_dst
 		struct rt6_info		rt6;
 	} u;
 	struct dst_entry *route;
+	u32 genid;
 	u32 route_mtu_cached;
 	u32 child_mtu_cached;
 	u32 route_cookie;
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 42d8ded0f96a..479598566f1d 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -93,6 +93,7 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 
 		xdst = (struct xfrm_dst *)dst1;
 		xdst->route = &rt->u.dst;
+		xdst->genid = xfrm[i]->genid;
 
 		dst1->next = dst_prev;
 		dst_prev = dst1;
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 98c2fe449b3f..9391c4c94feb 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -149,6 +149,7 @@ __xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 
 		xdst = (struct xfrm_dst *)dst1;
 		xdst->route = &rt->u.dst;
+		xdst->genid = xfrm[i]->genid;
 		if (rt->rt6i_node)
 			xdst->route_cookie = rt->rt6i_node->fn_sernum;
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 1732159ffd01..7fc6944ee36f 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1536,6 +1536,8 @@ int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family, int str
 			return 0;
 		if (dst->xfrm->km.state != XFRM_STATE_VALID)
 			return 0;
+		if (xdst->genid != dst->xfrm->genid)
+			return 0;
 
 		if (strict && fl && dst->xfrm->props.mode != XFRM_MODE_TUNNEL &&
 		    !xfrm_state_addr_flow_check(dst->xfrm, fl, family))
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 445263c54c94..535d43c14720 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -53,6 +53,7 @@ static struct hlist_head *xfrm_state_byspi __read_mostly;
 static unsigned int xfrm_state_hmask __read_mostly;
 static unsigned int xfrm_state_hashmax __read_mostly = 1 * 1024 * 1024;
 static unsigned int xfrm_state_num;
+static unsigned int xfrm_state_genid;
 
 static inline unsigned int __xfrm4_dst_hash(xfrm_address_t *addr, unsigned int hmask)
 {
@@ -745,6 +746,8 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 {
 	unsigned int h = xfrm_dst_hash(&x->id.daddr, x->props.family);
 
+	x->genid = ++xfrm_state_genid;
+
 	hlist_add_head(&x->bydst, xfrm_state_bydst+h);
 	xfrm_state_hold(x);
 
-- 
cgit v1.2.3


From c7f5ea3a4d1ae6b3b426e113358fdc57494bc754 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 03:29:04 -0700
Subject: [XFRM]: Do not flush all bundles on SA insert.

Instead, simply set all potentially aliasing existing xfrm_state
objects to have the current generation counter value.

This will make routes get relooked up the next time an existing
route mentioning these aliased xfrm_state objects gets used,
via xfrm_dst_check().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  1 -
 net/xfrm/xfrm_policy.c | 10 ----------
 net/xfrm/xfrm_state.c  | 25 ++++++++++++++++++++-----
 3 files changed, 20 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index fd4a300b5baf..a620a43c9eeb 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -996,7 +996,6 @@ struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto,
 extern void xfrm_policy_flush(u8 type);
 extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol);
 extern int xfrm_flush_bundles(void);
-extern void xfrm_flush_all_bundles(void);
 extern int xfrm_bundle_ok(struct xfrm_dst *xdst, struct flowi *fl, int family, int strict);
 extern void xfrm_init_pmtu(struct dst_entry *dst);
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 7fc6944ee36f..cfa5c692f2e8 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1478,16 +1478,6 @@ int xfrm_flush_bundles(void)
 	return 0;
 }
 
-static int always_true(struct dst_entry *dst)
-{
-	return 1;
-}
-
-void xfrm_flush_all_bundles(void)
-{
-	xfrm_prune_bundles(always_true);
-}
-
 void xfrm_init_pmtu(struct dst_entry *dst)
 {
 	do {
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 98200397e098..77ef796c9d0d 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -761,13 +761,30 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 		schedule_work(&xfrm_hash_work);
 }
 
+/* xfrm_state_lock is held */
+static void __xfrm_state_bump_genids(struct xfrm_state *xnew)
+{
+	unsigned short family = xnew->props.family;
+	u32 reqid = xnew->props.reqid;
+	struct xfrm_state *x;
+	struct hlist_node *entry;
+	unsigned int h;
+
+	h = xfrm_dst_hash(&xnew->id.daddr, reqid, family);
+	hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) {
+		if (x->props.family	== family &&
+		    x->props.reqid	== reqid &&
+		    !xfrm_addr_cmp(&x->id.daddr, &xnew->id.daddr, family))
+			x->genid = xfrm_state_genid;
+	}
+}
+
 void xfrm_state_insert(struct xfrm_state *x)
 {
 	spin_lock_bh(&xfrm_state_lock);
+	__xfrm_state_bump_genids(x);
 	__xfrm_state_insert(x);
 	spin_unlock_bh(&xfrm_state_lock);
-
-	xfrm_flush_all_bundles();
 }
 EXPORT_SYMBOL(xfrm_state_insert);
 
@@ -889,15 +906,13 @@ int xfrm_state_add(struct xfrm_state *x)
 				     x->id.proto,
 				     &x->id.daddr, &x->props.saddr, 0);
 
+	__xfrm_state_bump_genids(x);
 	__xfrm_state_insert(x);
 	err = 0;
 
 out:
 	spin_unlock_bh(&xfrm_state_lock);
 
-	if (!err)
-		xfrm_flush_all_bundles();
-
 	if (x1) {
 		xfrm_state_delete(x1);
 		xfrm_state_put(x1);
-- 
cgit v1.2.3


From 1c0953997567b22e32fdf85d3b4bc0f2461fd161 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 03:30:28 -0700
Subject: [XFRM]: Purge dst references to deleted SAs passively.

Just let GC and other normal mechanisms take care of getting
rid of DST cache references to deleted xfrm_state objects
instead of walking all the policy bundles.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  1 -
 net/xfrm/xfrm_policy.c |  2 +-
 net/xfrm/xfrm_state.c  | 17 -----------------
 3 files changed, 1 insertion(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index a620a43c9eeb..c7870b6eae01 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -995,7 +995,6 @@ struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto,
 				  int create, unsigned short family);
 extern void xfrm_policy_flush(u8 type);
 extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol);
-extern int xfrm_flush_bundles(void);
 extern int xfrm_bundle_ok(struct xfrm_dst *xdst, struct flowi *fl, int family, int strict);
 extern void xfrm_init_pmtu(struct dst_entry *dst);
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index cfa5c692f2e8..1bcaae4adf3a 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1472,7 +1472,7 @@ static void __xfrm_garbage_collect(void)
 	xfrm_prune_bundles(unused_bundle);
 }
 
-int xfrm_flush_bundles(void)
+static int xfrm_flush_bundles(void)
 {
 	xfrm_prune_bundles(stale_bundle);
 	return 0;
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 77ef796c9d0d..9ff00b7d6ad3 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -256,8 +256,6 @@ static struct work_struct xfrm_state_gc_work;
 static HLIST_HEAD(xfrm_state_gc_list);
 static DEFINE_SPINLOCK(xfrm_state_gc_lock);
 
-static int xfrm_state_gc_flush_bundles;
-
 int __xfrm_state_delete(struct xfrm_state *x);
 
 static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned short family);
@@ -293,11 +291,6 @@ static void xfrm_state_gc_task(void *data)
 	struct hlist_node *entry, *tmp;
 	struct hlist_head gc_list;
 
-	if (xfrm_state_gc_flush_bundles) {
-		xfrm_state_gc_flush_bundles = 0;
-		xfrm_flush_bundles();
-	}
-
 	spin_lock_bh(&xfrm_state_gc_lock);
 	gc_list.first = xfrm_state_gc_list.first;
 	INIT_HLIST_HEAD(&xfrm_state_gc_list);
@@ -454,16 +447,6 @@ int __xfrm_state_delete(struct xfrm_state *x)
 		if (del_timer(&x->rtimer))
 			__xfrm_state_put(x);
 
-		/* The number two in this test is the reference
-		 * mentioned in the comment below plus the reference
-		 * our caller holds.  A larger value means that
-		 * there are DSTs attached to this xfrm_state.
-		 */
-		if (atomic_read(&x->refcnt) > 2) {
-			xfrm_state_gc_flush_bundles = 1;
-			schedule_work(&xfrm_state_gc_work);
-		}
-
 		/* All xfrm_state objects are created by xfrm_state_alloc.
 		 * The xfrm_state_alloc call gives a reference, and that
 		 * is what we are dropping here.
-- 
cgit v1.2.3


From 2518c7c2b3d7f0a6b302b4efe17c911f8dd4049f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 24 Aug 2006 04:45:07 -0700
Subject: [XFRM]: Hash policies when non-prefixed.

This idea is from Alexey Kuznetsov.

It is common for policies to be non-prefixed.  And for
that case we can optimize lookups, insert, etc. quite
a bit.

For each direction, we have a dynamically sized policy
hash table for non-prefixed policies.  We also have a
hash table on policy->index.

For prefixed policies, we have a list per-direction which
we will consult on lookups when a non-prefix hashtable
lookup fails.

This still isn't as efficient as I would like it.  There
are four immediate problems:

1) Lots of excessive refcounting, which can be fixed just
   like xfrm_state was
2) We do 2 hash probes on insert, one to look for dups and
   one to allocate a unique policy->index.  Althought I wonder
   how much this matters since xfrm_state inserts do up to
   3 hash probes and that seems to perform fine.
3) xfrm_policy_insert() is very complex because of the priority
   ordering and entry replacement logic.
4) Lots of counter bumping, in addition to policy refcounts,
   in the form of xfrm_policy_count[].  This is merely used
   to let code path(s) know that some IPSEC rules exist.  So
   this count is indexed per-direction, maybe that is overkill.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |  23 +-
 net/xfrm/xfrm_policy.c | 681 +++++++++++++++++++++++++++++++++++++++----------
 2 files changed, 546 insertions(+), 158 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index c7870b6eae01..0acabf2a0a8f 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -331,7 +331,8 @@ struct xfrm_tmpl
 struct xfrm_policy
 {
 	struct xfrm_policy	*next;
-	struct list_head	list;
+	struct hlist_node	bydst;
+	struct hlist_node	byidx;
 
 	/* This lock only affects elements except for entry. */
 	rwlock_t		lock;
@@ -385,21 +386,7 @@ struct xfrm_mgr
 extern int xfrm_register_km(struct xfrm_mgr *km);
 extern int xfrm_unregister_km(struct xfrm_mgr *km);
 
-
-extern struct xfrm_policy *xfrm_policy_list[XFRM_POLICY_MAX*2];
-#ifdef CONFIG_XFRM_SUB_POLICY
-extern struct xfrm_policy *xfrm_policy_list_sub[XFRM_POLICY_MAX*2];
-
-static inline int xfrm_policy_lists_empty(int dir)
-{
-	return (!xfrm_policy_list[dir] && !xfrm_policy_list_sub[dir]);
-}
-#else
-static inline int xfrm_policy_lists_empty(int dir)
-{
-	return (!xfrm_policy_list[dir]);
-}
-#endif
+extern unsigned int xfrm_policy_count[XFRM_POLICY_MAX*2];
 
 static inline void xfrm_pol_hold(struct xfrm_policy *policy)
 {
@@ -678,7 +665,7 @@ static inline int xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *sk
 	if (sk && sk->sk_policy[XFRM_POLICY_IN])
 		return __xfrm_policy_check(sk, dir, skb, family);
 
-	return	(xfrm_policy_lists_empty(dir) && !skb->sp) ||
+	return	(!xfrm_policy_count[dir] && !skb->sp) ||
 		(skb->dst->flags & DST_NOPOLICY) ||
 		__xfrm_policy_check(sk, dir, skb, family);
 }
@@ -698,7 +685,7 @@ extern int __xfrm_route_forward(struct sk_buff *skb, unsigned short family);
 
 static inline int xfrm_route_forward(struct sk_buff *skb, unsigned short family)
 {
-	return	xfrm_policy_lists_empty(XFRM_POLICY_OUT) ||
+	return	!xfrm_policy_count[XFRM_POLICY_OUT] ||
 		(skb->dst->flags & DST_NOXFRM) ||
 		__xfrm_route_forward(skb, family);
 }
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 1bcaae4adf3a..087a5443b051 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -22,6 +22,9 @@
 #include <linux/netdevice.h>
 #include <linux/netfilter.h>
 #include <linux/module.h>
+#include <linux/bootmem.h>
+#include <linux/vmalloc.h>
+#include <linux/cache.h>
 #include <net/xfrm.h>
 #include <net/ip.h>
 
@@ -30,26 +33,8 @@ EXPORT_SYMBOL(xfrm_cfg_mutex);
 
 static DEFINE_RWLOCK(xfrm_policy_lock);
 
-struct xfrm_policy *xfrm_policy_list[XFRM_POLICY_MAX*2];
-EXPORT_SYMBOL(xfrm_policy_list);
-#ifdef CONFIG_XFRM_SUB_POLICY
-struct xfrm_policy *xfrm_policy_list_sub[XFRM_POLICY_MAX*2];
-EXPORT_SYMBOL(xfrm_policy_list_sub);
-
-#define XFRM_POLICY_LISTS(type) \
-	((type == XFRM_POLICY_TYPE_SUB) ? xfrm_policy_list_sub : \
-	 xfrm_policy_list)
-#define XFRM_POLICY_LISTHEAD(type, dir) \
-	((type == XFRM_POLICY_TYPE_SUB) ? xfrm_policy_list_sub[dir] : \
-	 xfrm_policy_list[dir])
-#define XFRM_POLICY_LISTHEADP(type, dir) \
-	((type == XFRM_POLICY_TYPE_SUB) ? &xfrm_policy_list_sub[dir] : \
-	 &xfrm_policy_list[dir])
-#else
-#define XFRM_POLICY_LISTS(type)              xfrm_policy_list
-#define XFRM_POLICY_LISTHEAD(type, dif)      xfrm_policy_list[dir]
-#define XFRM_POLICY_LISTHEADP(type, dif)     &xfrm_policy_list[dir]
-#endif
+unsigned int xfrm_policy_count[XFRM_POLICY_MAX*2];
+EXPORT_SYMBOL(xfrm_policy_count);
 
 static DEFINE_RWLOCK(xfrm_policy_afinfo_lock);
 static struct xfrm_policy_afinfo *xfrm_policy_afinfo[NPROTO];
@@ -57,8 +42,7 @@ static struct xfrm_policy_afinfo *xfrm_policy_afinfo[NPROTO];
 static kmem_cache_t *xfrm_dst_cache __read_mostly;
 
 static struct work_struct xfrm_policy_gc_work;
-static struct list_head xfrm_policy_gc_list =
-	LIST_HEAD_INIT(xfrm_policy_gc_list);
+static HLIST_HEAD(xfrm_policy_gc_list);
 static DEFINE_SPINLOCK(xfrm_policy_gc_lock);
 
 static struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short family);
@@ -328,8 +312,10 @@ struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp)
 	policy = kzalloc(sizeof(struct xfrm_policy), gfp);
 
 	if (policy) {
-		atomic_set(&policy->refcnt, 1);
+		INIT_HLIST_NODE(&policy->bydst);
+		INIT_HLIST_NODE(&policy->byidx);
 		rwlock_init(&policy->lock);
+		atomic_set(&policy->refcnt, 1);
 		init_timer(&policy->timer);
 		policy->timer.data = (unsigned long)policy;
 		policy->timer.function = xfrm_policy_timer;
@@ -375,17 +361,16 @@ static void xfrm_policy_gc_kill(struct xfrm_policy *policy)
 static void xfrm_policy_gc_task(void *data)
 {
 	struct xfrm_policy *policy;
-	struct list_head *entry, *tmp;
-	struct list_head gc_list = LIST_HEAD_INIT(gc_list);
+	struct hlist_node *entry, *tmp;
+	struct hlist_head gc_list;
 
 	spin_lock_bh(&xfrm_policy_gc_lock);
-	list_splice_init(&xfrm_policy_gc_list, &gc_list);
+	gc_list.first = xfrm_policy_gc_list.first;
+	INIT_HLIST_HEAD(&xfrm_policy_gc_list);
 	spin_unlock_bh(&xfrm_policy_gc_lock);
 
-	list_for_each_safe(entry, tmp, &gc_list) {
-		policy = list_entry(entry, struct xfrm_policy, list);
+	hlist_for_each_entry_safe(policy, entry, tmp, &gc_list, bydst)
 		xfrm_policy_gc_kill(policy);
-	}
 }
 
 /* Rule must be locked. Release descentant resources, announce
@@ -407,70 +392,354 @@ static void xfrm_policy_kill(struct xfrm_policy *policy)
 	}
 
 	spin_lock(&xfrm_policy_gc_lock);
-	list_add(&policy->list, &xfrm_policy_gc_list);
+	hlist_add_head(&policy->bydst, &xfrm_policy_gc_list);
 	spin_unlock(&xfrm_policy_gc_lock);
 
 	schedule_work(&xfrm_policy_gc_work);
 }
 
+struct xfrm_policy_hash {
+	struct hlist_head	*table;
+	unsigned int		hmask;
+};
+
+static struct hlist_head xfrm_policy_inexact[XFRM_POLICY_MAX*2];
+static struct xfrm_policy_hash xfrm_policy_bydst[XFRM_POLICY_MAX*2] __read_mostly;
+static struct hlist_head *xfrm_policy_byidx __read_mostly;
+static unsigned int xfrm_idx_hmask __read_mostly;
+static unsigned int xfrm_policy_hashmax __read_mostly = 1 * 1024 * 1024;
+
+static inline unsigned int __idx_hash(u32 index, unsigned int hmask)
+{
+	return (index ^ (index >> 8)) & hmask;
+}
+
+static inline unsigned int idx_hash(u32 index)
+{
+	return __idx_hash(index, xfrm_idx_hmask);
+}
+
+static inline unsigned int __sel_hash(struct xfrm_selector *sel, unsigned short family, unsigned int hmask)
+{
+	xfrm_address_t *daddr = &sel->daddr;
+	xfrm_address_t *saddr = &sel->saddr;
+	unsigned int h = 0;
+
+	switch (family) {
+	case AF_INET:
+		if (sel->prefixlen_d != 32 ||
+		    sel->prefixlen_s != 32)
+			return hmask + 1;
+
+		h = ntohl(daddr->a4 ^ saddr->a4);
+		break;
+
+	case AF_INET6:
+		if (sel->prefixlen_d != 128 ||
+		    sel->prefixlen_s != 128)
+			return hmask + 1;
+
+		h = ntohl(daddr->a6[2] ^ daddr->a6[3] ^
+			  saddr->a6[2] ^ saddr->a6[3]);
+		break;
+	};
+	h ^= (h >> 16);
+	return h & hmask;
+}
+
+static inline unsigned int __addr_hash(xfrm_address_t *daddr, xfrm_address_t *saddr, unsigned short family, unsigned int hmask)
+{
+	unsigned int h = 0;
+
+	switch (family) {
+	case AF_INET:
+		h = ntohl(daddr->a4 ^ saddr->a4);
+		break;
+
+	case AF_INET6:
+		h = ntohl(daddr->a6[2] ^ daddr->a6[3] ^
+			  saddr->a6[2] ^ saddr->a6[3]);
+		break;
+	};
+	h ^= (h >> 16);
+	return h & hmask;
+}
+
+static struct hlist_head *policy_hash_bysel(struct xfrm_selector *sel, unsigned short family, int dir)
+{
+	unsigned int hmask = xfrm_policy_bydst[dir].hmask;
+	unsigned int hash = __sel_hash(sel, family, hmask);
+
+	return (hash == hmask + 1 ?
+		&xfrm_policy_inexact[dir] :
+		xfrm_policy_bydst[dir].table + hash);
+}
+
+static struct hlist_head *policy_hash_direct(xfrm_address_t *daddr, xfrm_address_t *saddr, unsigned short family, int dir)
+{
+	unsigned int hmask = xfrm_policy_bydst[dir].hmask;
+	unsigned int hash = __addr_hash(daddr, saddr, family, hmask);
+
+	return xfrm_policy_bydst[dir].table + hash;
+}
+
+static struct hlist_head *xfrm_policy_hash_alloc(unsigned int sz)
+{
+	struct hlist_head *n;
+
+	if (sz <= PAGE_SIZE)
+		n = kmalloc(sz, GFP_KERNEL);
+	else if (hashdist)
+		n = __vmalloc(sz, GFP_KERNEL, PAGE_KERNEL);
+	else
+		n = (struct hlist_head *)
+			__get_free_pages(GFP_KERNEL, get_order(sz));
+
+	if (n)
+		memset(n, 0, sz);
+
+	return n;
+}
+
+static void xfrm_policy_hash_free(struct hlist_head *n, unsigned int sz)
+{
+	if (sz <= PAGE_SIZE)
+		kfree(n);
+	else if (hashdist)
+		vfree(n);
+	else
+		free_pages((unsigned long)n, get_order(sz));
+}
+
+static void xfrm_dst_hash_transfer(struct hlist_head *list,
+				   struct hlist_head *ndsttable,
+				   unsigned int nhashmask)
+{
+	struct hlist_node *entry, *tmp;
+	struct xfrm_policy *pol;
+
+	hlist_for_each_entry_safe(pol, entry, tmp, list, bydst) {
+		unsigned int h;
+
+		h = __addr_hash(&pol->selector.daddr, &pol->selector.saddr,
+				pol->family, nhashmask);
+		hlist_add_head(&pol->bydst, ndsttable+h);
+	}
+}
+
+static void xfrm_idx_hash_transfer(struct hlist_head *list,
+				   struct hlist_head *nidxtable,
+				   unsigned int nhashmask)
+{
+	struct hlist_node *entry, *tmp;
+	struct xfrm_policy *pol;
+
+	hlist_for_each_entry_safe(pol, entry, tmp, list, byidx) {
+		unsigned int h;
+
+		h = __idx_hash(pol->index, nhashmask);
+		hlist_add_head(&pol->byidx, nidxtable+h);
+	}
+}
+
+static unsigned long xfrm_new_hash_mask(unsigned int old_hmask)
+{
+	return ((old_hmask + 1) << 1) - 1;
+}
+
+static void xfrm_bydst_resize(int dir)
+{
+	unsigned int hmask = xfrm_policy_bydst[dir].hmask;
+	unsigned int nhashmask = xfrm_new_hash_mask(hmask);
+	unsigned int nsize = (nhashmask + 1) * sizeof(struct hlist_head);
+	struct hlist_head *odst = xfrm_policy_bydst[dir].table;
+	struct hlist_head *ndst = xfrm_policy_hash_alloc(nsize);
+	int i;
+
+	if (!ndst)
+		return;
+
+	write_lock_bh(&xfrm_policy_lock);
+
+	for (i = hmask; i >= 0; i--)
+		xfrm_dst_hash_transfer(odst + i, ndst, nhashmask);
+
+	xfrm_policy_bydst[dir].table = ndst;
+	xfrm_policy_bydst[dir].hmask = nhashmask;
+
+	write_unlock_bh(&xfrm_policy_lock);
+
+	xfrm_policy_hash_free(odst, (hmask + 1) * sizeof(struct hlist_head));
+}
+
+static void xfrm_byidx_resize(int total)
+{
+	unsigned int hmask = xfrm_idx_hmask;
+	unsigned int nhashmask = xfrm_new_hash_mask(hmask);
+	unsigned int nsize = (nhashmask + 1) * sizeof(struct hlist_head);
+	struct hlist_head *oidx = xfrm_policy_byidx;
+	struct hlist_head *nidx = xfrm_policy_hash_alloc(nsize);
+	int i;
+
+	if (!nidx)
+		return;
+
+	write_lock_bh(&xfrm_policy_lock);
+
+	for (i = hmask; i >= 0; i--)
+		xfrm_idx_hash_transfer(oidx + i, nidx, nhashmask);
+
+	xfrm_policy_byidx = nidx;
+	xfrm_idx_hmask = nhashmask;
+
+	write_unlock_bh(&xfrm_policy_lock);
+
+	xfrm_policy_hash_free(oidx, (hmask + 1) * sizeof(struct hlist_head));
+}
+
+static inline int xfrm_bydst_should_resize(int dir, int *total)
+{
+	unsigned int cnt = xfrm_policy_count[dir];
+	unsigned int hmask = xfrm_policy_bydst[dir].hmask;
+
+	if (total)
+		*total += cnt;
+
+	if ((hmask + 1) < xfrm_policy_hashmax &&
+	    cnt > hmask)
+		return 1;
+
+	return 0;
+}
+
+static inline int xfrm_byidx_should_resize(int total)
+{
+	unsigned int hmask = xfrm_idx_hmask;
+
+	if ((hmask + 1) < xfrm_policy_hashmax &&
+	    total > hmask)
+		return 1;
+
+	return 0;
+}
+
+static DEFINE_MUTEX(hash_resize_mutex);
+
+static void xfrm_hash_resize(void *__unused)
+{
+	int dir, total;
+
+	mutex_lock(&hash_resize_mutex);
+
+	total = 0;
+	for (dir = 0; dir < XFRM_POLICY_MAX * 2; dir++) {
+		if (xfrm_bydst_should_resize(dir, &total))
+			xfrm_bydst_resize(dir);
+	}
+	if (xfrm_byidx_should_resize(total))
+		xfrm_byidx_resize(total);
+
+	mutex_unlock(&hash_resize_mutex);
+}
+
+static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL);
+
 /* Generate new index... KAME seems to generate them ordered by cost
  * of an absolute inpredictability of ordering of rules. This will not pass. */
 static u32 xfrm_gen_index(u8 type, int dir)
 {
-	u32 idx;
-	struct xfrm_policy *p;
 	static u32 idx_generator;
 
 	for (;;) {
+		struct hlist_node *entry;
+		struct hlist_head *list;
+		struct xfrm_policy *p;
+		u32 idx;
+		int found;
+
 		idx = (idx_generator | dir);
 		idx_generator += 8;
 		if (idx == 0)
 			idx = 8;
-		for (p = XFRM_POLICY_LISTHEAD(type, dir); p; p = p->next) {
-			if (p->index == idx)
+		list = xfrm_policy_byidx + idx_hash(idx);
+		found = 0;
+		hlist_for_each_entry(p, entry, list, byidx) {
+			if (p->index == idx) {
+				found = 1;
 				break;
+			}
 		}
-		if (!p)
+		if (!found)
 			return idx;
 	}
 }
 
+static inline int selector_cmp(struct xfrm_selector *s1, struct xfrm_selector *s2)
+{
+	u32 *p1 = (u32 *) s1;
+	u32 *p2 = (u32 *) s2;
+	int len = sizeof(struct xfrm_selector) / sizeof(u32);
+	int i;
+
+	for (i = 0; i < len; i++) {
+		if (p1[i] != p2[i])
+			return 1;
+	}
+
+	return 0;
+}
+
 int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 {
-	struct xfrm_policy *pol, **p;
-	struct xfrm_policy *delpol = NULL;
-	struct xfrm_policy **newpos = NULL;
+	struct xfrm_policy *pol;
+	struct xfrm_policy *delpol;
+	struct hlist_head *chain;
+	struct hlist_node *entry, *newpos, *last;
 	struct dst_entry *gc_list;
 
 	write_lock_bh(&xfrm_policy_lock);
-	for (p = XFRM_POLICY_LISTHEADP(policy->type, dir); (pol=*p)!=NULL;) {
-		if (!delpol && memcmp(&policy->selector, &pol->selector, sizeof(pol->selector)) == 0 &&
+	chain = policy_hash_bysel(&policy->selector, policy->family, dir);
+	delpol = NULL;
+	newpos = NULL;
+	last = NULL;
+	hlist_for_each_entry(pol, entry, chain, bydst) {
+		if (!delpol &&
+		    pol->type == policy->type &&
+		    !selector_cmp(&pol->selector, &policy->selector) &&
 		    xfrm_sec_ctx_match(pol->security, policy->security)) {
 			if (excl) {
 				write_unlock_bh(&xfrm_policy_lock);
 				return -EEXIST;
 			}
-			*p = pol->next;
 			delpol = pol;
 			if (policy->priority > pol->priority)
 				continue;
 		} else if (policy->priority >= pol->priority) {
-			p = &pol->next;
+			last = &pol->bydst;
 			continue;
 		}
 		if (!newpos)
-			newpos = p;
+			newpos = &pol->bydst;
 		if (delpol)
 			break;
-		p = &pol->next;
+		last = &pol->bydst;
 	}
+	if (!newpos)
+		newpos = last;
 	if (newpos)
-		p = newpos;
+		hlist_add_after(newpos, &policy->bydst);
+	else
+		hlist_add_head(&policy->bydst, chain);
 	xfrm_pol_hold(policy);
-	policy->next = *p;
-	*p = policy;
+	xfrm_policy_count[dir]++;
 	atomic_inc(&flow_cache_genid);
+	if (delpol) {
+		hlist_del(&delpol->bydst);
+		hlist_del(&delpol->byidx);
+		xfrm_policy_count[dir]--;
+	}
 	policy->index = delpol ? delpol->index : xfrm_gen_index(policy->type, dir);
+	hlist_add_head(&policy->byidx, xfrm_policy_byidx+idx_hash(policy->index));
 	policy->curlft.add_time = (unsigned long)xtime.tv_sec;
 	policy->curlft.use_time = 0;
 	if (!mod_timer(&policy->timer, jiffies + HZ))
@@ -479,10 +748,13 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 
 	if (delpol)
 		xfrm_policy_kill(delpol);
+	else if (xfrm_bydst_should_resize(dir, NULL))
+		schedule_work(&xfrm_hash_work);
 
 	read_lock_bh(&xfrm_policy_lock);
 	gc_list = NULL;
-	for (policy = policy->next; policy; policy = policy->next) {
+	entry = &policy->bydst;
+	hlist_for_each_entry_continue(policy, entry, bydst) {
 		struct dst_entry *dst;
 
 		write_lock(&policy->lock);
@@ -515,67 +787,112 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir,
 					  struct xfrm_selector *sel,
 					  struct xfrm_sec_ctx *ctx, int delete)
 {
-	struct xfrm_policy *pol, **p;
+	struct xfrm_policy *pol, *ret;
+	struct hlist_head *chain;
+	struct hlist_node *entry;
 
 	write_lock_bh(&xfrm_policy_lock);
-	for (p = XFRM_POLICY_LISTHEADP(type, dir); (pol=*p)!=NULL; p = &pol->next) {
-		if ((memcmp(sel, &pol->selector, sizeof(*sel)) == 0) &&
-		    (xfrm_sec_ctx_match(ctx, pol->security))) {
+	chain = policy_hash_bysel(sel, sel->family, dir);
+	ret = NULL;
+	hlist_for_each_entry(pol, entry, chain, bydst) {
+		if (pol->type == type &&
+		    !selector_cmp(sel, &pol->selector) &&
+		    xfrm_sec_ctx_match(ctx, pol->security)) {
 			xfrm_pol_hold(pol);
-			if (delete)
-				*p = pol->next;
+			if (delete) {
+				hlist_del(&pol->bydst);
+				hlist_del(&pol->byidx);
+				xfrm_policy_count[dir]--;
+			}
+			ret = pol;
 			break;
 		}
 	}
 	write_unlock_bh(&xfrm_policy_lock);
 
-	if (pol && delete) {
+	if (ret && delete) {
 		atomic_inc(&flow_cache_genid);
-		xfrm_policy_kill(pol);
+		xfrm_policy_kill(ret);
 	}
-	return pol;
+	return ret;
 }
 EXPORT_SYMBOL(xfrm_policy_bysel_ctx);
 
 struct xfrm_policy *xfrm_policy_byid(u8 type, int dir, u32 id, int delete)
 {
-	struct xfrm_policy *pol, **p;
+	struct xfrm_policy *pol, *ret;
+	struct hlist_head *chain;
+	struct hlist_node *entry;
 
 	write_lock_bh(&xfrm_policy_lock);
-	for (p = XFRM_POLICY_LISTHEADP(type, dir); (pol=*p)!=NULL; p = &pol->next) {
-		if (pol->index == id) {
+	chain = xfrm_policy_byidx + idx_hash(id);
+	ret = NULL;
+	hlist_for_each_entry(pol, entry, chain, byidx) {
+		if (pol->type == type && pol->index == id) {
 			xfrm_pol_hold(pol);
-			if (delete)
-				*p = pol->next;
+			if (delete) {
+				hlist_del(&pol->bydst);
+				hlist_del(&pol->byidx);
+				xfrm_policy_count[dir]--;
+			}
+			ret = pol;
 			break;
 		}
 	}
 	write_unlock_bh(&xfrm_policy_lock);
 
-	if (pol && delete) {
+	if (ret && delete) {
 		atomic_inc(&flow_cache_genid);
-		xfrm_policy_kill(pol);
+		xfrm_policy_kill(ret);
 	}
-	return pol;
+	return ret;
 }
 EXPORT_SYMBOL(xfrm_policy_byid);
 
 void xfrm_policy_flush(u8 type)
 {
-	struct xfrm_policy *xp;
-	struct xfrm_policy **p_list = XFRM_POLICY_LISTS(type);
 	int dir;
 
 	write_lock_bh(&xfrm_policy_lock);
 	for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
-		while ((xp = p_list[dir]) != NULL) {
-			p_list[dir] = xp->next;
+		struct xfrm_policy *pol;
+		struct hlist_node *entry;
+		int i;
+
+	again1:
+		hlist_for_each_entry(pol, entry,
+				     &xfrm_policy_inexact[dir], bydst) {
+			if (pol->type != type)
+				continue;
+			hlist_del(&pol->bydst);
+			hlist_del(&pol->byidx);
 			write_unlock_bh(&xfrm_policy_lock);
 
-			xfrm_policy_kill(xp);
+			xfrm_policy_kill(pol);
 
 			write_lock_bh(&xfrm_policy_lock);
+			goto again1;
+		}
+
+		for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) {
+	again2:
+			hlist_for_each_entry(pol, entry,
+					     xfrm_policy_bydst[dir].table + i,
+					     bydst) {
+				if (pol->type != type)
+					continue;
+				hlist_del(&pol->bydst);
+				hlist_del(&pol->byidx);
+				write_unlock_bh(&xfrm_policy_lock);
+
+				xfrm_policy_kill(pol);
+
+				write_lock_bh(&xfrm_policy_lock);
+				goto again2;
+			}
 		}
+
+		xfrm_policy_count[dir] = 0;
 	}
 	atomic_inc(&flow_cache_genid);
 	write_unlock_bh(&xfrm_policy_lock);
@@ -585,15 +902,27 @@ EXPORT_SYMBOL(xfrm_policy_flush);
 int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*),
 		     void *data)
 {
-	struct xfrm_policy *xp;
-	int dir;
-	int count = 0;
-	int error = 0;
+	struct xfrm_policy *pol;
+	struct hlist_node *entry;
+	int dir, count, error;
 
 	read_lock_bh(&xfrm_policy_lock);
+	count = 0;
 	for (dir = 0; dir < 2*XFRM_POLICY_MAX; dir++) {
-		for (xp = XFRM_POLICY_LISTHEAD(type, dir); xp; xp = xp->next)
-			count++;
+		struct hlist_head *table = xfrm_policy_bydst[dir].table;
+		int i;
+
+		hlist_for_each_entry(pol, entry,
+				     &xfrm_policy_inexact[dir], bydst) {
+			if (pol->type == type)
+				count++;
+		}
+		for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) {
+			hlist_for_each_entry(pol, entry, table + i, bydst) {
+				if (pol->type == type)
+					count++;
+			}
+		}
 	}
 
 	if (count == 0) {
@@ -602,13 +931,28 @@ int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*)
 	}
 
 	for (dir = 0; dir < 2*XFRM_POLICY_MAX; dir++) {
-		for (xp = XFRM_POLICY_LISTHEAD(type, dir); xp; xp = xp->next) {
-			error = func(xp, dir%XFRM_POLICY_MAX, --count, data);
+		struct hlist_head *table = xfrm_policy_bydst[dir].table;
+		int i;
+
+		hlist_for_each_entry(pol, entry,
+				     &xfrm_policy_inexact[dir], bydst) {
+			if (pol->type != type)
+				continue;
+			error = func(pol, dir % XFRM_POLICY_MAX, --count, data);
 			if (error)
 				goto out;
 		}
+		for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) {
+			hlist_for_each_entry(pol, entry, table + i, bydst) {
+				if (pol->type != type)
+					continue;
+				error = func(pol, dir % XFRM_POLICY_MAX, --count, data);
+				if (error)
+					goto out;
+			}
+		}
 	}
-
+	error = 0;
 out:
 	read_unlock_bh(&xfrm_policy_lock);
 	return error;
@@ -617,31 +961,61 @@ EXPORT_SYMBOL(xfrm_policy_walk);
 
 /* Find policy to apply to this flow. */
 
-static struct xfrm_policy *xfrm_policy_lookup_bytype(u8 type, struct flowi *fl,
-						     u16 family, u8 dir)
+static int xfrm_policy_match(struct xfrm_policy *pol, struct flowi *fl,
+			     u8 type, u16 family, int dir)
 {
-	struct xfrm_policy *pol;
+	struct xfrm_selector *sel = &pol->selector;
+	int match;
 
-	read_lock_bh(&xfrm_policy_lock);
-	for (pol = XFRM_POLICY_LISTHEAD(type, dir); pol; pol = pol->next) {
-		struct xfrm_selector *sel = &pol->selector;
-		int match;
+	if (pol->family != family ||
+	    pol->type != type)
+		return 0;
 
-		if (pol->family != family)
-			continue;
+	match = xfrm_selector_match(sel, fl, family);
+	if (match) {
+		if (!security_xfrm_policy_lookup(pol, fl->secid, dir))
+			return 1;
+	}
+
+	return 0;
+}
 
-		match = xfrm_selector_match(sel, fl, family);
+static struct xfrm_policy *xfrm_policy_lookup_bytype(u8 type, struct flowi *fl,
+						     u16 family, u8 dir)
+{
+	struct xfrm_policy *pol, *ret;
+	xfrm_address_t *daddr, *saddr;
+	struct hlist_node *entry;
+	struct hlist_head *chain;
 
-		if (match) {
- 			if (!security_xfrm_policy_lookup(pol, fl->secid, dir)) {
+	daddr = xfrm_flowi_daddr(fl, family);
+	saddr = xfrm_flowi_saddr(fl, family);
+	if (unlikely(!daddr || !saddr))
+		return NULL;
+
+	read_lock_bh(&xfrm_policy_lock);
+	chain = policy_hash_direct(daddr, saddr, family, dir);
+	ret = NULL;
+	hlist_for_each_entry(pol, entry, chain, bydst) {
+		if (xfrm_policy_match(pol, fl, type, family, dir)) {
+			xfrm_pol_hold(pol);
+			ret = pol;
+			break;
+		}
+	}
+	if (!ret) {
+		chain = &xfrm_policy_inexact[dir];
+		hlist_for_each_entry(pol, entry, chain, bydst) {
+			if (xfrm_policy_match(pol, fl, type, family, dir)) {
 				xfrm_pol_hold(pol);
+				ret = pol;
 				break;
 			}
 		}
 	}
 	read_unlock_bh(&xfrm_policy_lock);
 
-	return pol;
+	return ret;
 }
 
 static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir,
@@ -657,7 +1031,7 @@ static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir,
 	pol = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_MAIN, fl, family, dir);
 
 #ifdef CONFIG_XFRM_SUB_POLICY
- end:
+end:
 #endif
 	if ((*objp = (void *) pol) != NULL)
 		*obj_refp = &pol->refcnt;
@@ -704,26 +1078,29 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc
 
 static void __xfrm_policy_link(struct xfrm_policy *pol, int dir)
 {
-	struct xfrm_policy **p_list = XFRM_POLICY_LISTS(pol->type);
+	struct hlist_head *chain = policy_hash_bysel(&pol->selector,
+						     pol->family, dir);
 
-	pol->next = p_list[dir];
-	p_list[dir] = pol;
+	hlist_add_head(&pol->bydst, chain);
+	hlist_add_head(&pol->byidx, xfrm_policy_byidx+idx_hash(pol->index));
+	xfrm_policy_count[dir]++;
 	xfrm_pol_hold(pol);
+
+	if (xfrm_bydst_should_resize(dir, NULL))
+		schedule_work(&xfrm_hash_work);
 }
 
 static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol,
 						int dir)
 {
-	struct xfrm_policy **polp;
+	if (hlist_unhashed(&pol->bydst))
+		return NULL;
 
-	for (polp = XFRM_POLICY_LISTHEADP(pol->type, dir);
-	     *polp != NULL; polp = &(*polp)->next) {
-		if (*polp == pol) {
-			*polp = pol->next;
-			return pol;
-		}
-	}
-	return NULL;
+	hlist_del(&pol->bydst);
+	hlist_del(&pol->byidx);
+	xfrm_policy_count[dir]--;
+
+	return pol;
 }
 
 int xfrm_policy_delete(struct xfrm_policy *pol, int dir)
@@ -968,7 +1345,8 @@ restart:
 
 	if (!policy) {
 		/* To accelerate a bit...  */
-		if ((dst_orig->flags & DST_NOXFRM) || xfrm_policy_lists_empty(XFRM_POLICY_OUT))
+		if ((dst_orig->flags & DST_NOXFRM) ||
+		    !xfrm_policy_count[XFRM_POLICY_OUT])
 			return 0;
 
 		policy = flow_cache_lookup(fl, dst_orig->ops->family,
@@ -1413,50 +1791,50 @@ static struct dst_entry *xfrm_negative_advice(struct dst_entry *dst)
 	return dst;
 }
 
+static void prune_one_bundle(struct xfrm_policy *pol, int (*func)(struct dst_entry *), struct dst_entry **gc_list_p)
+{
+	struct dst_entry *dst, **dstp;
+
+	write_lock(&pol->lock);
+	dstp = &pol->bundles;
+	while ((dst=*dstp) != NULL) {
+		if (func(dst)) {
+			*dstp = dst->next;
+			dst->next = *gc_list_p;
+			*gc_list_p = dst;
+		} else {
+			dstp = &dst->next;
+		}
+	}
+	write_unlock(&pol->lock);
+}
+
 static void xfrm_prune_bundles(int (*func)(struct dst_entry *))
 {
-	int i;
-	struct xfrm_policy *pol;
-	struct dst_entry *dst, **dstp, *gc_list = NULL;
+	struct dst_entry *gc_list = NULL;
+	int dir;
 
 	read_lock_bh(&xfrm_policy_lock);
-	for (i=0; i<2*XFRM_POLICY_MAX; i++) {
-#ifdef CONFIG_XFRM_SUB_POLICY
-		for (pol = xfrm_policy_list_sub[i]; pol; pol = pol->next) {
-			write_lock(&pol->lock);
-			dstp = &pol->bundles;
-			while ((dst=*dstp) != NULL) {
-				if (func(dst)) {
-					*dstp = dst->next;
-					dst->next = gc_list;
-					gc_list = dst;
-				} else {
-					dstp = &dst->next;
-				}
-			}
-			write_unlock(&pol->lock);
-		}
+	for (dir = 0; dir < XFRM_POLICY_MAX * 2; dir++) {
+		struct xfrm_policy *pol;
+		struct hlist_node *entry;
+		struct hlist_head *table;
+		int i;
 
-#endif
-		for (pol = xfrm_policy_list[i]; pol; pol = pol->next) {
-			write_lock(&pol->lock);
-			dstp = &pol->bundles;
-			while ((dst=*dstp) != NULL) {
-				if (func(dst)) {
-					*dstp = dst->next;
-					dst->next = gc_list;
-					gc_list = dst;
-				} else {
-					dstp = &dst->next;
-				}
-			}
-			write_unlock(&pol->lock);
+		hlist_for_each_entry(pol, entry,
+				     &xfrm_policy_inexact[dir], bydst)
+			prune_one_bundle(pol, func, &gc_list);
+
+		table = xfrm_policy_bydst[dir].table;
+		for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) {
+			hlist_for_each_entry(pol, entry, table + i, bydst)
+				prune_one_bundle(pol, func, &gc_list);
 		}
 	}
 	read_unlock_bh(&xfrm_policy_lock);
 
 	while (gc_list) {
-		dst = gc_list;
+		struct dst_entry *dst = gc_list;
 		gc_list = dst->next;
 		dst_free(dst);
 	}
@@ -1680,6 +2058,9 @@ static struct notifier_block xfrm_dev_notifier = {
 
 static void __init xfrm_policy_init(void)
 {
+	unsigned int hmask, sz;
+	int dir;
+
 	xfrm_dst_cache = kmem_cache_create("xfrm_dst_cache",
 					   sizeof(struct xfrm_dst),
 					   0, SLAB_HWCACHE_ALIGN,
@@ -1687,6 +2068,26 @@ static void __init xfrm_policy_init(void)
 	if (!xfrm_dst_cache)
 		panic("XFRM: failed to allocate xfrm_dst_cache\n");
 
+	hmask = 8 - 1;
+	sz = (hmask+1) * sizeof(struct hlist_head);
+
+	xfrm_policy_byidx = xfrm_policy_hash_alloc(sz);
+	xfrm_idx_hmask = hmask;
+	if (!xfrm_policy_byidx)
+		panic("XFRM: failed to allocate byidx hash\n");
+
+	for (dir = 0; dir < XFRM_POLICY_MAX * 2; dir++) {
+		struct xfrm_policy_hash *htab;
+
+		INIT_HLIST_HEAD(&xfrm_policy_inexact[dir]);
+
+		htab = &xfrm_policy_bydst[dir];
+		htab->table = xfrm_policy_hash_alloc(sz);
+		htab->hmask = hmask;
+		if (!htab->table)
+			panic("XFRM: failed to allocate bydst hash\n");
+	}
+
 	INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task, NULL);
 	register_netdevice_notifier(&xfrm_dev_notifier);
 }
-- 
cgit v1.2.3


From e4bec827feda76d5e7417a2696a75424834d564f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 22 Sep 2006 15:17:35 -0700
Subject: [IPSEC] esp: Defer output IV initialization to first use.

First of all, if the xfrm_state only gets used for input
packets this entropy is a complete waste.

Secondly, it is often the case that a configuration loads
many rules (perhaps even dynamically) and they don't all
necessarily ever get used.

This get_random_bytes() call was showing up in the profiles
for xfrm_state inserts which is how I noticed this.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/esp.h | 5 +++--
 net/ipv4/esp4.c   | 9 +++++++--
 net/ipv6/esp6.c   | 9 +++++++--
 3 files changed, 17 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/esp.h b/include/net/esp.h
index 064366d66eea..713d039f4af7 100644
--- a/include/net/esp.h
+++ b/include/net/esp.h
@@ -15,13 +15,14 @@ struct esp_data
 	struct {
 		u8			*key;		/* Key */
 		int			key_len;	/* Key length */
-		u8			*ivec;		/* ivec buffer */
+		int			padlen;		/* 0..255 */
 		/* ivlen is offset from enc_data, where encrypted data start.
 		 * It is logically different of crypto_tfm_alg_ivsize(tfm).
 		 * We assume that it is either zero (no ivec), or
 		 * >= crypto_tfm_alg_ivsize(tfm). */
 		int			ivlen;
-		int			padlen;		/* 0..255 */
+		int			ivinitted;
+		u8			*ivec;		/* ivec buffer */
 		struct crypto_blkcipher	*tfm;		/* crypto handle */
 	} conf;
 
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index e87377e1d6b6..13b29360d102 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -95,8 +95,13 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 	esph->seq_no = htonl(++x->replay.oseq);
 	xfrm_aevent_doreplay(x);
 
-	if (esp->conf.ivlen)
+	if (esp->conf.ivlen) {
+		if (unlikely(!esp->conf.ivinitted)) {
+			get_random_bytes(esp->conf.ivec, esp->conf.ivlen);
+			esp->conf.ivinitted = 1;
+		}
 		crypto_blkcipher_set_iv(tfm, esp->conf.ivec, esp->conf.ivlen);
+	}
 
 	do {
 		struct scatterlist *sg = &esp->sgbuf[0];
@@ -378,7 +383,7 @@ static int esp_init_state(struct xfrm_state *x)
 		esp->conf.ivec = kmalloc(esp->conf.ivlen, GFP_KERNEL);
 		if (unlikely(esp->conf.ivec == NULL))
 			goto error;
-		get_random_bytes(esp->conf.ivec, esp->conf.ivlen);
+		esp->conf.ivinitted = 0;
 	}
 	if (crypto_blkcipher_setkey(tfm, esp->conf.key, esp->conf.key_len))
 		goto error;
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index ae50b9511151..e78680a9985b 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -99,8 +99,13 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 	esph->seq_no = htonl(++x->replay.oseq);
 	xfrm_aevent_doreplay(x);
 
-	if (esp->conf.ivlen)
+	if (esp->conf.ivlen) {
+		if (unlikely(!esp->conf.ivinitted)) {
+			get_random_bytes(esp->conf.ivec, esp->conf.ivlen);
+			esp->conf.ivinitted = 1;
+		}
 		crypto_blkcipher_set_iv(tfm, esp->conf.ivec, esp->conf.ivlen);
+	}
 
 	do {
 		struct scatterlist *sg = &esp->sgbuf[0];
@@ -353,7 +358,7 @@ static int esp6_init_state(struct xfrm_state *x)
 		esp->conf.ivec = kmalloc(esp->conf.ivlen, GFP_KERNEL);
 		if (unlikely(esp->conf.ivec == NULL))
 			goto error;
-		get_random_bytes(esp->conf.ivec, esp->conf.ivlen);
+		esp->conf.ivinitted = 0;
 	}
 	if (crypto_blkcipher_setkey(tfm, esp->conf.key, esp->conf.key_len))
 		goto error;
-- 
cgit v1.2.3


From 75bff8f023e02b045a8f68f36fa7da98dca124b8 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Mon, 21 Aug 2006 19:22:01 +0900
Subject: [IPV6] ROUTE: Routing by FWMARK.

Based on patch by Jean Lorchat <lorchat@sfc.wide.ad.jp>.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/fib_rules.h |  2 +-
 include/net/flow.h        |  2 ++
 net/ipv6/Kconfig          |  7 +++++++
 net/ipv6/fib6_rules.c     | 23 +++++++++++++++++++++++
 net/ipv6/route.c          |  1 +
 5 files changed, 34 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h
index 19a82b6c1c1f..2987549d6044 100644
--- a/include/linux/fib_rules.h
+++ b/include/linux/fib_rules.h
@@ -34,7 +34,7 @@ enum
 	FRA_UNUSED3,
 	FRA_UNUSED4,
 	FRA_UNUSED5,
-	FRA_FWMARK,	/* netfilter mark (IPv4) */
+	FRA_FWMARK,	/* netfilter mark (IPv4/IPv6) */
 	FRA_FLOW,	/* flow/class id */
 	FRA_UNUSED6,
 	FRA_UNUSED7,
diff --git a/include/net/flow.h b/include/net/flow.h
index e0522914316e..3ca210ec1379 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -26,6 +26,7 @@ struct flowi {
 		struct {
 			struct in6_addr		daddr;
 			struct in6_addr		saddr;
+			__u32			fwmark;
 			__u32			flowlabel;
 		} ip6_u;
 
@@ -42,6 +43,7 @@ struct flowi {
 #define fld_scope	nl_u.dn_u.scope
 #define fl6_dst		nl_u.ip6_u.daddr
 #define fl6_src		nl_u.ip6_u.saddr
+#define fl6_fwmark	nl_u.ip6_u.fwmark
 #define fl6_flowlabel	nl_u.ip6_u.flowlabel
 #define fl4_dst		nl_u.ip4_u.daddr
 #define fl4_src		nl_u.ip4_u.saddr
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 21e0cc808f44..a2d211da2aba 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -173,3 +173,10 @@ config IPV6_MULTIPLE_TABLES
 	---help---
 	  Support multiple routing tables.
 
+config IPV6_ROUTE_FWMARK
+	bool "IPv6: use netfilter MARK value as routing key"
+	depends on IPV6_MULTIPLE_TABLES && NETFILTER
+	---help---
+	  If you say Y here, you will be able to specify different routes for
+	  packets with different mark values (see iptables(8), MARK target).
+
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 91f6233d8efd..aebd9e2b85a8 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -26,6 +26,9 @@ struct fib6_rule
 	struct fib_rule		common;
 	struct rt6key		src;
 	struct rt6key		dst;
+#ifdef CONFIG_IPV6_ROUTE_FWMARK
+	u8			fwmark;
+#endif
 	u8			tclass;
 };
 
@@ -124,6 +127,11 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 	if (r->tclass && r->tclass != ((ntohl(fl->fl6_flowlabel) >> 20) & 0xff))
 		return 0;
 
+#ifdef CONFIG_IPV6_ROUTE_FWMARK
+	if (r->fwmark && (r->fwmark != fl->fl6_fwmark))
+		return 0;
+#endif
+
 	return 1;
 }
 
@@ -164,6 +172,11 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 		nla_memcpy(&rule6->dst.addr, tb[FRA_DST],
 			   sizeof(struct in6_addr));
 
+#ifdef CONFIG_IPV6_ROUTE_FWMARK
+	if (tb[FRA_FWMARK])
+		rule6->fwmark = nla_get_u32(tb[FRA_FWMARK]);
+#endif
+
 	rule6->src.plen = frh->src_len;
 	rule6->dst.plen = frh->dst_len;
 	rule6->tclass = frh->tos;
@@ -195,6 +208,11 @@ static int fib6_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
 	    nla_memcmp(tb[FRA_DST], &rule6->dst.addr, sizeof(struct in6_addr)))
 		return 0;
 
+#ifdef CONFIG_IPV6_ROUTE_FWMARK
+	if (tb[FRA_FWMARK] && (rule6->fwmark != nla_get_u32(tb[FRA_FWMARK])))
+		return 0;
+#endif
+
 	return 1;
 }
 
@@ -216,6 +234,11 @@ static int fib6_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
 		NLA_PUT(skb, FRA_SRC, sizeof(struct in6_addr),
 			&rule6->src.addr);
 
+#ifdef CONFIG_IPV6_ROUTE_FWMARK
+	if (rule6->fwmark)
+		NLA_PUT_U32(skb, FRA_FWMARK, rule6->fwmark);
+#endif
+
 	return 0;
 
 nla_put_failure:
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 20691285aee5..649350bd9299 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -703,6 +703,7 @@ void ip6_route_input(struct sk_buff *skb)
 			.ip6_u = {
 				.daddr = iph->daddr,
 				.saddr = iph->saddr,
+				.fwmark = skb->nfmark,
 				.flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK,
 			},
 		},
-- 
cgit v1.2.3


From 1aaec67f9335a17856dfacdd3e5cc6f4c18faeec Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sun, 25 Jun 2006 23:54:55 +0900
Subject: [NET]: Add common helper functions to convert IPv6/IPv4 address
 string to network address structure.

These helpers can be used in netfilter, cifs etc.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/inet.h |   2 +
 net/core/utils.c     | 215 +++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 217 insertions(+)

(limited to 'include')

diff --git a/include/linux/inet.h b/include/linux/inet.h
index 6c5587af118d..b7c6da7d6d32 100644
--- a/include/linux/inet.h
+++ b/include/linux/inet.h
@@ -46,5 +46,7 @@
 #include <linux/types.h>
 
 extern __be32 in_aton(const char *str);
+extern int in4_pton(const char *src, int srclen, u8 *dst, char delim, const char **end);
+extern int in6_pton(const char *src, int srclen, u8 *dst, char delim, const char **end);
 #endif
 #endif	/* _LINUX_INET_H */
diff --git a/net/core/utils.c b/net/core/utils.c
index e31c90e05594..5a06e8a72c17 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -4,6 +4,7 @@
  *	Authors:
  *	net_random Alan Cox
  *	net_ratelimit Andy Kleen
+ *	in{4,6}_pton YOSHIFUJI Hideaki, Copyright (C)2006 USAGI/WIDE Project
  *
  *	Created by Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
  *
@@ -191,3 +192,217 @@ __be32 in_aton(const char *str)
 }
 
 EXPORT_SYMBOL(in_aton);
+
+#define IN6PTON_XDIGIT		0x00010000
+#define IN6PTON_DIGIT		0x00020000
+#define IN6PTON_COLON_MASK	0x00700000
+#define IN6PTON_COLON_1		0x00100000	/* single : requested */
+#define IN6PTON_COLON_2		0x00200000	/* second : requested */
+#define IN6PTON_COLON_1_2	0x00400000	/* :: requested */
+#define IN6PTON_DOT		0x00800000	/* . */
+#define IN6PTON_DELIM		0x10000000
+#define IN6PTON_NULL		0x20000000	/* first/tail */
+#define IN6PTON_UNKNOWN		0x40000000
+
+static inline int digit2bin(char c, char delim)
+{
+	if (c == delim || c == '\0')
+		return IN6PTON_DELIM;
+	if (c == '.')
+		return IN6PTON_DOT;
+	if (c >= '0' && c <= '9')
+		return (IN6PTON_DIGIT | (c - '0'));
+	return IN6PTON_UNKNOWN;
+}
+
+static inline int xdigit2bin(char c, char delim)
+{
+	if (c == delim || c == '\0')
+		return IN6PTON_DELIM;
+	if (c == ':')
+		return IN6PTON_COLON_MASK;
+	if (c == '.')
+		return IN6PTON_DOT;
+	if (c >= '0' && c <= '9')
+		return (IN6PTON_XDIGIT | IN6PTON_DIGIT| (c - '0'));
+	if (c >= 'a' && c <= 'f')
+		return (IN6PTON_XDIGIT | (c - 'a' + 10));
+	if (c >= 'A' && c <= 'F')
+		return (IN6PTON_XDIGIT | (c - 'A' + 10));
+	return IN6PTON_UNKNOWN;
+}
+
+int in4_pton(const char *src, int srclen,
+	     u8 *dst,
+	     char delim, const char **end)
+{
+	const char *s;
+	u8 *d;
+	u8 dbuf[4];
+	int ret = 0;
+	int i;
+	int w = 0;
+
+	if (srclen < 0)
+		srclen = strlen(src);
+	s = src;
+	d = dbuf;
+	i = 0;
+	while(1) {
+		int c;
+		c = xdigit2bin(srclen > 0 ? *s : '\0', delim);
+		if (!(c & (IN6PTON_DIGIT | IN6PTON_DOT | IN6PTON_DELIM))) {
+			goto out;
+		}
+		if (c & (IN6PTON_DOT | IN6PTON_DELIM)) {
+			if (w == 0)
+				goto out;
+			*d++ = w & 0xff;
+			w = 0;
+			i++;
+			if (c & IN6PTON_DELIM) {
+				if (i != 4)
+					goto out;
+				break;
+			}
+			goto cont;
+		}
+		w = (w * 10) + c;
+		if ((w & 0xffff) > 255) {
+			goto out;
+		}
+cont:
+		if (i >= 4)
+			goto out;
+		s++;
+		srclen--;
+	}
+	ret = 1;
+	memcpy(dst, dbuf, sizeof(dbuf));
+out:
+	if (end)
+		*end = s;
+	return ret;
+}
+
+EXPORT_SYMBOL(in4_pton);
+
+int in6_pton(const char *src, int srclen,
+	     u8 *dst,
+	     char delim, const char **end)
+{
+	const char *s, *tok = NULL;
+	u8 *d, *dc = NULL;
+	u8 dbuf[16];
+	int ret = 0;
+	int i;
+	int state = IN6PTON_COLON_1_2 | IN6PTON_XDIGIT | IN6PTON_NULL;
+	int w = 0;
+
+	memset(dbuf, 0, sizeof(dbuf));
+
+	s = src;
+	d = dbuf;
+	if (srclen < 0)
+		srclen = strlen(src);
+
+	printf("srclen=%d\n", srclen);
+
+	while (1) {
+		int c;
+
+		c = xdigit2bin(srclen > 0 ? *s : '\0', delim);
+		if (!(c & state))
+			goto out;
+		if (c & (IN6PTON_DELIM | IN6PTON_COLON_MASK)) {
+			/* process one 16-bit word */
+			if (!(state & IN6PTON_NULL)) {
+				*d++ = (w >> 8) & 0xff;
+				*d++ = w & 0xff;
+			}
+			w = 0;
+			if (c & IN6PTON_DELIM) {
+				/* We've processed last word */
+				break;
+			}
+			/*
+			 * COLON_1 => XDIGIT
+			 * COLON_2 => XDIGIT|DELIM
+			 * COLON_1_2 => COLON_2
+			 */
+			switch (state & IN6PTON_COLON_MASK) {
+			case IN6PTON_COLON_2:
+				dc = d;
+				state = IN6PTON_XDIGIT | IN6PTON_DELIM;
+				if (dc - dbuf >= sizeof(dbuf))
+					state |= IN6PTON_NULL;
+				break;
+			case IN6PTON_COLON_1|IN6PTON_COLON_1_2:
+				state = IN6PTON_XDIGIT | IN6PTON_COLON_2;
+				break;
+			case IN6PTON_COLON_1:
+				state = IN6PTON_XDIGIT;
+				break;
+			case IN6PTON_COLON_1_2:
+				state = IN6PTON_COLON_2;
+				break;
+			default:
+				state = 0;
+			}
+			tok = s + 1;
+			goto cont;
+		}
+
+		if (c & IN6PTON_DOT) {
+			ret = in4_pton(tok ? tok : s, srclen + (int)(s - tok), d, delim, &s);
+			if (ret > 0) {
+				d += 4;
+				break;
+			}
+			goto out;
+		}
+
+		w = (w << 4) | (0xff & c);
+		state = IN6PTON_COLON_1 | IN6PTON_DELIM;
+		if (!(w & 0xf000)) {
+			state |= IN6PTON_XDIGIT;
+		}
+		if (!dc && d + 2 < dbuf + sizeof(dbuf)) {
+			state |= IN6PTON_COLON_1_2;
+			state &= ~IN6PTON_DELIM;
+		}
+		if (d + 2 >= dbuf + sizeof(dbuf)) {
+			state &= ~(IN6PTON_COLON_1|IN6PTON_COLON_1_2);
+		}
+cont:
+		if ((dc && d + 4 < dbuf + sizeof(dbuf)) ||
+		    d + 4 == dbuf + sizeof(dbuf)) {
+			state |= IN6PTON_DOT;
+		}
+		if (d >= dbuf + sizeof(dbuf)) {
+			state &= ~(IN6PTON_XDIGIT|IN6PTON_COLON_MASK);
+		}
+		s++;
+		srclen--;
+	}
+
+	i = 15; d--;
+
+	if (dc) {
+		while(d >= dc)
+			dst[i--] = *d--;
+		while(i >= dc - dbuf)
+			dst[i--] = 0;
+		while(i >= 0)
+			dst[i--] = *d--;
+	} else
+		memcpy(dst, dbuf, sizeof(dbuf));
+
+	ret = 1;
+out:
+	if (end)
+		*end = s;
+	return ret;
+}
+
+EXPORT_SYMBOL(in6_pton);
-- 
cgit v1.2.3


From bbfb39cbf63829d1db607aa90cbdca557a3a131d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 25 Aug 2006 16:10:14 -0700
Subject: [IPV4]: Add support for fwmark masks in routing rules

Add a FRA_FWMASK attributes for fwmark masks. For compatibility a mask of
0xFFFFFFFF is used when a mark value != 0 is sent without a mask.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/fib_rules.h |  3 ++-
 net/ipv4/fib_rules.c      | 21 +++++++++++++++++++--
 2 files changed, 21 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h
index 2987549d6044..4418c8d9d479 100644
--- a/include/linux/fib_rules.h
+++ b/include/linux/fib_rules.h
@@ -34,12 +34,13 @@ enum
 	FRA_UNUSED3,
 	FRA_UNUSED4,
 	FRA_UNUSED5,
-	FRA_FWMARK,	/* netfilter mark (IPv4/IPv6) */
+	FRA_FWMARK,	/* netfilter mark */
 	FRA_FLOW,	/* flow/class id */
 	FRA_UNUSED6,
 	FRA_UNUSED7,
 	FRA_UNUSED8,
 	FRA_TABLE,	/* Extended table id */
+	FRA_FWMASK,	/* mask for netfilter mark */
 	__FRA_MAX
 };
 
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index ce185ac6f260..280f424ca9c9 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -46,6 +46,7 @@ struct fib4_rule
 	u32			dstmask;
 #ifdef CONFIG_IP_ROUTE_FWMARK
 	u32			fwmark;
+	u32			fwmask;
 #endif
 #ifdef CONFIG_NET_CLS_ROUTE
 	u32			tclassid;
@@ -160,7 +161,7 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 		return 0;
 
 #ifdef CONFIG_IP_ROUTE_FWMARK
-	if (r->fwmark && (r->fwmark != fl->fl4_fwmark))
+	if ((r->fwmark ^ fl->fl4_fwmark) & r->fwmask)
 		return 0;
 #endif
 
@@ -183,6 +184,7 @@ static struct nla_policy fib4_rule_policy[FRA_MAX+1] __read_mostly = {
 	[FRA_SRC]	= { .type = NLA_U32 },
 	[FRA_DST]	= { .type = NLA_U32 },
 	[FRA_FWMARK]	= { .type = NLA_U32 },
+	[FRA_FWMASK]	= { .type = NLA_U32 },
 	[FRA_FLOW]	= { .type = NLA_U32 },
 	[FRA_TABLE]	= { .type = NLA_U32 },
 };
@@ -219,8 +221,17 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 		rule4->dst = nla_get_u32(tb[FRA_DST]);
 
 #ifdef CONFIG_IP_ROUTE_FWMARK
-	if (tb[FRA_FWMARK])
+	if (tb[FRA_FWMARK]) {
 		rule4->fwmark = nla_get_u32(tb[FRA_FWMARK]);
+		if (rule4->fwmark)
+			/* compatibility: if the mark value is non-zero all bits
+			 * are compared unless a mask is explicitly specified.
+			 */
+			rule4->fwmask = 0xFFFFFFFF;
+	}
+
+	if (tb[FRA_FWMASK])
+		rule4->fwmask = nla_get_u32(tb[FRA_FWMASK]);
 #endif
 
 #ifdef CONFIG_NET_CLS_ROUTE
@@ -256,6 +267,9 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
 #ifdef CONFIG_IP_ROUTE_FWMARK
 	if (tb[FRA_FWMARK] && (rule4->fwmark != nla_get_u32(tb[FRA_FWMARK])))
 		return 0;
+
+	if (tb[FRA_FWMASK] && (rule4->fwmask != nla_get_u32(tb[FRA_FWMASK])))
+		return 0;
 #endif
 
 #ifdef CONFIG_NET_CLS_ROUTE
@@ -285,6 +299,9 @@ static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
 #ifdef CONFIG_IP_ROUTE_FWMARK
 	if (rule4->fwmark)
 		NLA_PUT_U32(skb, FRA_FWMARK, rule4->fwmark);
+
+	if (rule4->fwmask || rule4->fwmark)
+		NLA_PUT_U32(skb, FRA_FWMASK, rule4->fwmask);
 #endif
 
 	if (rule4->dst_len)
-- 
cgit v1.2.3


From b4e9b520ca5d07a37ea59648e7f50f478e7487a3 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 25 Aug 2006 16:11:42 -0700
Subject: [NET_SCHED]: Add mask support to fwmark classifier

Support masking the nfmark value before the search. The mask value is
global for all filters contained in one instance. It can only be set
when a new instance is created, all filters must specify the same mask.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_cls.h |  1 +
 net/sched/cls_fw.c      | 25 ++++++++++++++++++++++++-
 2 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index bd2c5a2bbbf5..c3f01b3085a4 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -305,6 +305,7 @@ enum
 	TCA_FW_POLICE,
 	TCA_FW_INDEV, /*  used by CONFIG_NET_CLS_IND */
 	TCA_FW_ACT, /* used by CONFIG_NET_CLS_ACT */
+	TCA_FW_MASK,
 	__TCA_FW_MAX
 };
 
diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c
index e6973d9b686d..e54acc6bcccd 100644
--- a/net/sched/cls_fw.c
+++ b/net/sched/cls_fw.c
@@ -50,6 +50,7 @@
 struct fw_head
 {
 	struct fw_filter *ht[HTSIZE];
+	u32 mask;
 };
 
 struct fw_filter
@@ -101,7 +102,7 @@ static int fw_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	struct fw_filter *f;
 	int r;
 #ifdef CONFIG_NETFILTER
-	u32 id = skb->nfmark;
+	u32 id = skb->nfmark & head->mask;
 #else
 	u32 id = 0;
 #endif
@@ -209,7 +210,9 @@ static int
 fw_change_attrs(struct tcf_proto *tp, struct fw_filter *f,
 	struct rtattr **tb, struct rtattr **tca, unsigned long base)
 {
+	struct fw_head *head = (struct fw_head *)tp->root;
 	struct tcf_exts e;
+	u32 mask;
 	int err;
 
 	err = tcf_exts_validate(tp, tb, tca[TCA_RATE-1], &e, &fw_ext_map);
@@ -232,6 +235,15 @@ fw_change_attrs(struct tcf_proto *tp, struct fw_filter *f,
 	}
 #endif /* CONFIG_NET_CLS_IND */
 
+	if (tb[TCA_FW_MASK-1]) {
+		if (RTA_PAYLOAD(tb[TCA_FW_MASK-1]) != sizeof(u32))
+			goto errout;
+		mask = *(u32*)RTA_DATA(tb[TCA_FW_MASK-1]);
+		if (mask != head->mask)
+			goto errout;
+	} else if (head->mask != 0xFFFFFFFF)
+		goto errout;
+
 	tcf_exts_change(tp, &f->exts, &e);
 
 	return 0;
@@ -267,9 +279,17 @@ static int fw_change(struct tcf_proto *tp, unsigned long base,
 		return -EINVAL;
 
 	if (head == NULL) {
+		u32 mask = 0xFFFFFFFF;
+		if (tb[TCA_FW_MASK-1]) {
+			if (RTA_PAYLOAD(tb[TCA_FW_MASK-1]) != sizeof(u32))
+				return -EINVAL;
+			mask = *(u32*)RTA_DATA(tb[TCA_FW_MASK-1]);
+		}
+
 		head = kzalloc(sizeof(struct fw_head), GFP_KERNEL);
 		if (head == NULL)
 			return -ENOBUFS;
+		head->mask = mask;
 
 		tcf_tree_lock(tp);
 		tp->root = head;
@@ -330,6 +350,7 @@ static void fw_walk(struct tcf_proto *tp, struct tcf_walker *arg)
 static int fw_dump(struct tcf_proto *tp, unsigned long fh,
 		   struct sk_buff *skb, struct tcmsg *t)
 {
+	struct fw_head *head = (struct fw_head *)tp->root;
 	struct fw_filter *f = (struct fw_filter*)fh;
 	unsigned char	 *b = skb->tail;
 	struct rtattr *rta;
@@ -351,6 +372,8 @@ static int fw_dump(struct tcf_proto *tp, unsigned long fh,
 	if (strlen(f->indev))
 		RTA_PUT(skb, TCA_FW_INDEV, IFNAMSIZ, f->indev);
 #endif /* CONFIG_NET_CLS_IND */
+	if (head->mask != 0xFFFFFFFF)
+		RTA_PUT(skb, TCA_FW_MASK, 4, &head->mask);
 
 	if (tcf_exts_dump(skb, &f->exts, &fw_ext_map) < 0)
 		goto rtattr_failure;
-- 
cgit v1.2.3


From 97e5848dd39e7e76bd6077735ebb5473763ab9c5 Mon Sep 17 00:00:00 2001
From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Sat, 26 Aug 2006 19:16:45 -0700
Subject: [DCCP]: Introduce tx buffering

This adds transmit buffering to DCCP.

I have tested with CCID2/3 and with loss and rate limiting.

Signed off by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dccp.h |  2 ++
 net/dccp/dccp.h      |  2 +-
 net/dccp/output.c    | 90 +++++++++++++++++++++++++++++++++++++---------------
 net/dccp/proto.c     | 16 +++-------
 4 files changed, 73 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 676333b9fad0..2d7671c92c0b 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -438,6 +438,7 @@ struct dccp_ackvec;
  * @dccps_role - Role of this sock, one of %dccp_role
  * @dccps_ndp_count - number of Non Data Packets since last data packet
  * @dccps_hc_rx_ackvec - rx half connection ack vector
+ * @dccps_xmit_timer - timer for when CCID is not ready to send
  */
 struct dccp_sock {
 	/* inet_connection_sock has to be the first member of dccp_sock */
@@ -470,6 +471,7 @@ struct dccp_sock {
 	enum dccp_role			dccps_role:2;
 	__u8				dccps_hc_rx_insert_options:1;
 	__u8				dccps_hc_tx_insert_options:1;
+	struct timer_list		dccps_xmit_timer;
 };
  
 static inline struct dccp_sock *dccp_sk(const struct sock *sk)
diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h
index a5c5475724c0..0a21be437ed3 100644
--- a/net/dccp/dccp.h
+++ b/net/dccp/dccp.h
@@ -130,7 +130,7 @@ extern void dccp_send_delayed_ack(struct sock *sk);
 extern void dccp_send_sync(struct sock *sk, const u64 seq,
 			   const enum dccp_pkt_type pkt_type);
 
-extern int dccp_write_xmit(struct sock *sk, struct sk_buff *skb, long *timeo);
+extern void dccp_write_xmit(struct sock *sk, int block);
 extern void dccp_write_space(struct sock *sk);
 
 extern void dccp_init_xmit_timers(struct sock *sk);
diff --git a/net/dccp/output.c b/net/dccp/output.c
index 58669beee132..7102e3aed4ca 100644
--- a/net/dccp/output.c
+++ b/net/dccp/output.c
@@ -198,7 +198,7 @@ static int dccp_wait_for_ccid(struct sock *sk, struct sk_buff *skb,
 	while (1) {
 		prepare_to_wait(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE);
 
-		if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))
+		if (sk->sk_err)
 			goto do_error;
 		if (!*timeo)
 			goto do_nonblock;
@@ -234,37 +234,72 @@ do_interrupted:
 	goto out;
 }
 
-int dccp_write_xmit(struct sock *sk, struct sk_buff *skb, long *timeo)
+static void dccp_write_xmit_timer(unsigned long data) {
+	struct sock *sk = (struct sock *)data;
+	struct dccp_sock *dp = dccp_sk(sk);
+
+	bh_lock_sock(sk);
+	if (sock_owned_by_user(sk))
+		sk_reset_timer(sk, &dp->dccps_xmit_timer, jiffies+1);
+	else
+		dccp_write_xmit(sk, 0);
+	bh_unlock_sock(sk);
+	sock_put(sk);
+}
+
+void dccp_write_xmit(struct sock *sk, int block)
 {
-	const struct dccp_sock *dp = dccp_sk(sk);
-	int err = ccid_hc_tx_send_packet(dp->dccps_hc_tx_ccid, sk, skb,
+	struct dccp_sock *dp = dccp_sk(sk);
+	struct sk_buff *skb;
+	long timeo = 30000; 	/* If a packet is taking longer than 2 secs
+				   we have other issues */
+
+	while ((skb = skb_peek(&sk->sk_write_queue))) {
+		int err = ccid_hc_tx_send_packet(dp->dccps_hc_tx_ccid, sk, skb,
 					 skb->len);
 
-	if (err > 0)
-		err = dccp_wait_for_ccid(sk, skb, timeo);
+		if (err > 0) {
+			if (!block) {
+				sk_reset_timer(sk, &dp->dccps_xmit_timer,
+						msecs_to_jiffies(err)+jiffies);
+				break;
+			} else
+				err = dccp_wait_for_ccid(sk, skb, &timeo);
+			if (err) {
+				printk(KERN_CRIT "%s:err at dccp_wait_for_ccid"
+						 " %d\n", __FUNCTION__, err);
+				dump_stack();
+			}
+		}
 
-	if (err == 0) {
-		struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
-		const int len = skb->len;
+		skb_dequeue(&sk->sk_write_queue);
+		if (err == 0) {
+			struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
+			const int len = skb->len;
 
-		if (sk->sk_state == DCCP_PARTOPEN) {
-			/* See 8.1.5.  Handshake Completion */
-			inet_csk_schedule_ack(sk);
-			inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
+			if (sk->sk_state == DCCP_PARTOPEN) {
+				/* See 8.1.5.  Handshake Completion */
+				inet_csk_schedule_ack(sk);
+				inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
 						  inet_csk(sk)->icsk_rto,
 						  DCCP_RTO_MAX);
-			dcb->dccpd_type = DCCP_PKT_DATAACK;
-		} else if (dccp_ack_pending(sk))
-			dcb->dccpd_type = DCCP_PKT_DATAACK;
-		else
-			dcb->dccpd_type = DCCP_PKT_DATA;
-
-		err = dccp_transmit_skb(sk, skb);
-		ccid_hc_tx_packet_sent(dp->dccps_hc_tx_ccid, sk, 0, len);
-	} else
-		kfree_skb(skb);
-
-	return err;
+				dcb->dccpd_type = DCCP_PKT_DATAACK;
+			} else if (dccp_ack_pending(sk))
+				dcb->dccpd_type = DCCP_PKT_DATAACK;
+			else
+				dcb->dccpd_type = DCCP_PKT_DATA;
+
+			err = dccp_transmit_skb(sk, skb);
+			ccid_hc_tx_packet_sent(dp->dccps_hc_tx_ccid, sk, 0, len);
+			if (err) {
+				printk(KERN_CRIT "%s:err from "
+					         "ccid_hc_tx_packet_sent %d\n",
+					         __FUNCTION__, err);
+				dump_stack();
+			}
+		} else
+			kfree(skb);
+	}
 }
 
 int dccp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
@@ -426,6 +461,9 @@ static inline void dccp_connect_init(struct sock *sk)
 	dccp_set_seqno(&dp->dccps_awl, max48(dp->dccps_awl, dp->dccps_iss));
 
 	icsk->icsk_retransmits = 0;
+	init_timer(&dp->dccps_xmit_timer);
+	dp->dccps_xmit_timer.data = (unsigned long)sk;
+	dp->dccps_xmit_timer.function = dccp_write_xmit_timer;
 }
 
 int dccp_connect(struct sock *sk)
@@ -560,8 +598,10 @@ void dccp_send_close(struct sock *sk, const int active)
 					DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ;
 
 	if (active) {
+		dccp_write_xmit(sk, 1);
 		dccp_skb_entail(sk, skb);
 		dccp_transmit_skb(sk, skb_clone(skb, prio));
+		/* FIXME do we need a retransmit timer here? */
 	} else
 		dccp_transmit_skb(sk, skb);
 }
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 6f14bb5a28d4..962df0ea31aa 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -662,17 +662,8 @@ int dccp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	if (rc != 0)
 		goto out_discard;
 
-	rc = dccp_write_xmit(sk, skb, &timeo);
-	/*
-	 * XXX we don't use sk_write_queue, so just discard the packet.
-	 *     Current plan however is to _use_ sk_write_queue with
-	 *     an algorith similar to tcp_sendmsg, where the main difference
-	 *     is that in DCCP we have to respect packet boundaries, so
-	 *     no coalescing of skbs.
-	 *
-	 *     This bug was _quickly_ found & fixed by just looking at an OSTRA
-	 *     generated callgraph 8) -acme
-	 */
+	skb_queue_tail(&sk->sk_write_queue, skb);
+	dccp_write_xmit(sk,0);
 out_release:
 	release_sock(sk);
 	return rc ? : len;
@@ -846,6 +837,7 @@ static int dccp_close_state(struct sock *sk)
 
 void dccp_close(struct sock *sk, long timeout)
 {
+	struct dccp_sock *dp = dccp_sk(sk);
 	struct sk_buff *skb;
 	int state;
 
@@ -862,6 +854,8 @@ void dccp_close(struct sock *sk, long timeout)
 		goto adjudge_to_death;
 	}
 
+	sk_stop_timer(sk, &dp->dccps_xmit_timer);
+
 	/*
 	 * We need to flush the recv. buffs.  We do this only on the
 	 * descriptor close, not protocol-sourced closes, because the
-- 
cgit v1.2.3


From ff5dfe736dd9f6c74b206aa77c0465dfd503bdb9 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Sat, 26 Aug 2006 19:17:53 -0700
Subject: [NETLINK]: remove third bogus argument from NLA_PUT_FLAG

This patch removes the 'value' argument from NLA_PUT_FLAG which is
unused anyway. The documentation comment was already correct so it
doesn't need an update :)

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 47044da167c5..bcb27e3a312e 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -828,7 +828,7 @@ static inline int nla_put_msecs(struct sk_buff *skb, int attrtype,
 #define NLA_PUT_STRING(skb, attrtype, value) \
 	NLA_PUT(skb, attrtype, strlen(value) + 1, value)
 
-#define NLA_PUT_FLAG(skb, attrtype, value) \
+#define NLA_PUT_FLAG(skb, attrtype) \
 	NLA_PUT(skb, attrtype, 0, NULL)
 
 #define NLA_PUT_MSECS(skb, attrtype, jiffies) \
-- 
cgit v1.2.3


From 25030a7f9eeab2dcefff036469e0e2b4f956198f Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Sat, 26 Aug 2006 20:06:05 -0700
Subject: [UDP]: Unify UDPv4 and UDPv6 ->get_port()

This patch creates one common function which is called by
udp_v4_get_port() and udp_v6_get_port(). As a result,
  * duplicated code is removed
  * udp_port_rover and local port lookup can now be removed from udp.h
  * further savings follow since the same function will be used by UDP-Litev4
    and UDP-Litev6

In contrast to the patch sent in response to Yoshifujis comments
(fixed by this variant), the code below also removes the
EXPORT_SYMBOL(udp_port_rover), since udp_port_rover can now remain
local to net/ipv4/udp.c.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h | 18 ++---------
 net/ipv4/udp.c    | 96 ++++++++++++++++++++++++++++++++++---------------------
 net/ipv6/udp.c    | 76 ++-----------------------------------------
 3 files changed, 64 insertions(+), 126 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index 766fba1369ce..c490a0f662ac 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -30,25 +30,9 @@
 
 #define UDP_HTABLE_SIZE		128
 
-/* udp.c: This needs to be shared by v4 and v6 because the lookup
- *        and hashing code needs to work with different AF's yet
- *        the port space is shared.
- */
 extern struct hlist_head udp_hash[UDP_HTABLE_SIZE];
 extern rwlock_t udp_hash_lock;
 
-extern int udp_port_rover;
-
-static inline int udp_lport_inuse(u16 num)
-{
-	struct sock *sk;
-	struct hlist_node *node;
-
-	sk_for_each(sk, node, &udp_hash[num & (UDP_HTABLE_SIZE - 1)])
-		if (inet_sk(sk)->num == num)
-			return 1;
-	return 0;
-}
 
 /* Note: this must match 'valbool' in sock_setsockopt */
 #define UDP_CSUM_NOXMIT		1
@@ -63,6 +47,8 @@ extern struct proto udp_prot;
 
 struct sk_buff;
 
+extern int	udp_get_port(struct sock *sk, unsigned short snum,
+			     int (*saddr_cmp)(struct sock *, struct sock *));
 extern void	udp_err(struct sk_buff *, u32);
 
 extern int	udp_sendmsg(struct kiocb *iocb, struct sock *sk,
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 514c1e9ae810..7552b50bcd84 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -118,14 +118,34 @@ DEFINE_SNMP_STAT(struct udp_mib, udp_statistics) __read_mostly;
 struct hlist_head udp_hash[UDP_HTABLE_SIZE];
 DEFINE_RWLOCK(udp_hash_lock);
 
-/* Shared by v4/v6 udp. */
+/* Shared by v4/v6 udp_get_port */
 int udp_port_rover;
 
-static int udp_v4_get_port(struct sock *sk, unsigned short snum)
+static inline int udp_lport_inuse(u16 num)
 {
+	struct sock *sk;
 	struct hlist_node *node;
+
+	sk_for_each(sk, node, &udp_hash[num & (UDP_HTABLE_SIZE - 1)])
+		if (inet_sk(sk)->num == num)
+			return 1;
+	return 0;
+}
+
+/**
+ *  udp_get_port  -  common port lookup for IPv4 and IPv6
+ *
+ *  @sk:          socket struct in question
+ *  @snum:        port number to look up
+ *  @saddr_comp:  AF-dependent comparison of bound local IP addresses
+ */
+int udp_get_port(struct sock *sk, unsigned short snum,
+		 int (*saddr_cmp)(struct sock *sk1, struct sock *sk2))
+{
+	struct hlist_node *node;
+	struct hlist_head *head;
 	struct sock *sk2;
-	struct inet_sock *inet = inet_sk(sk);
+	int    error = 1;
 
 	write_lock_bh(&udp_hash_lock);
 	if (snum == 0) {
@@ -137,11 +157,10 @@ static int udp_v4_get_port(struct sock *sk, unsigned short snum)
 		best_size_so_far = 32767;
 		best = result = udp_port_rover;
 		for (i = 0; i < UDP_HTABLE_SIZE; i++, result++) {
-			struct hlist_head *list;
 			int size;
 
-			list = &udp_hash[result & (UDP_HTABLE_SIZE - 1)];
-			if (hlist_empty(list)) {
+			head = &udp_hash[result & (UDP_HTABLE_SIZE - 1)];
+			if (hlist_empty(head)) {
 				if (result > sysctl_local_port_range[1])
 					result = sysctl_local_port_range[0] +
 						((result - sysctl_local_port_range[0]) &
@@ -149,12 +168,11 @@ static int udp_v4_get_port(struct sock *sk, unsigned short snum)
 				goto gotit;
 			}
 			size = 0;
-			sk_for_each(sk2, node, list)
-				if (++size >= best_size_so_far)
-					goto next;
-			best_size_so_far = size;
-			best = result;
-		next:;
+			sk_for_each(sk2, node, head)
+				if (++size < best_size_so_far) {
+					best_size_so_far = size;
+					best = result;
+				}
 		}
 		result = best;
 		for(i = 0; i < (1 << 16) / UDP_HTABLE_SIZE; i++, result += UDP_HTABLE_SIZE) {
@@ -170,38 +188,44 @@ static int udp_v4_get_port(struct sock *sk, unsigned short snum)
 gotit:
 		udp_port_rover = snum = result;
 	} else {
-		sk_for_each(sk2, node,
-			    &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]) {
-			struct inet_sock *inet2 = inet_sk(sk2);
-
-			if (inet2->num == snum &&
-			    sk2 != sk &&
-			    !ipv6_only_sock(sk2) &&
-			    (!sk2->sk_bound_dev_if ||
-			     !sk->sk_bound_dev_if ||
-			     sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
-			    (!inet2->rcv_saddr ||
-			     !inet->rcv_saddr ||
-			     inet2->rcv_saddr == inet->rcv_saddr) &&
-			    (!sk2->sk_reuse || !sk->sk_reuse))
+		head = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)];
+
+		sk_for_each(sk2, node, head)
+			if (inet_sk(sk2)->num == snum                        &&
+			    sk2 != sk                                        &&
+			    (!sk2->sk_reuse        || !sk->sk_reuse)         &&
+			    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if
+			     || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
+			    (*saddr_cmp)(sk, sk2)                              )
 				goto fail;
-		}
 	}
-	inet->num = snum;
+	inet_sk(sk)->num = snum;
 	if (sk_unhashed(sk)) {
-		struct hlist_head *h = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)];
-
-		sk_add_node(sk, h);
+		head = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)];
+		sk_add_node(sk, head);
 		sock_prot_inc_use(sk->sk_prot);
 	}
-	write_unlock_bh(&udp_hash_lock);
-	return 0;
-
+	error = 0;
 fail:
 	write_unlock_bh(&udp_hash_lock);
-	return 1;
+	return error;
+}
+
+static inline int  ipv4_rcv_saddr_equal(struct sock *sk1, struct sock *sk2)
+{
+	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
+
+	return 	( !ipv6_only_sock(sk2)  &&
+		  (!inet1->rcv_saddr || !inet2->rcv_saddr ||
+		   inet1->rcv_saddr == inet2->rcv_saddr      ));
+}
+
+static inline int udp_v4_get_port(struct sock *sk, unsigned short snum)
+{
+	return udp_get_port(sk, snum, ipv4_rcv_saddr_equal);
 }
 
+
 static void udp_v4_hash(struct sock *sk)
 {
 	BUG();
@@ -1596,7 +1620,7 @@ EXPORT_SYMBOL(udp_disconnect);
 EXPORT_SYMBOL(udp_hash);
 EXPORT_SYMBOL(udp_hash_lock);
 EXPORT_SYMBOL(udp_ioctl);
-EXPORT_SYMBOL(udp_port_rover);
+EXPORT_SYMBOL(udp_get_port);
 EXPORT_SYMBOL(udp_prot);
 EXPORT_SYMBOL(udp_sendmsg);
 EXPORT_SYMBOL(udp_poll);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index b9cc55ccb000..9662561701d1 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -61,81 +61,9 @@
 
 DEFINE_SNMP_STAT(struct udp_mib, udp_stats_in6) __read_mostly;
 
-/* Grrr, addr_type already calculated by caller, but I don't want
- * to add some silly "cookie" argument to this method just for that.
- */
-static int udp_v6_get_port(struct sock *sk, unsigned short snum)
+static inline int udp_v6_get_port(struct sock *sk, unsigned short snum)
 {
-	struct sock *sk2;
-	struct hlist_node *node;
-
-	write_lock_bh(&udp_hash_lock);
-	if (snum == 0) {
-		int best_size_so_far, best, result, i;
-
-		if (udp_port_rover > sysctl_local_port_range[1] ||
-		    udp_port_rover < sysctl_local_port_range[0])
-			udp_port_rover = sysctl_local_port_range[0];
-		best_size_so_far = 32767;
-		best = result = udp_port_rover;
-		for (i = 0; i < UDP_HTABLE_SIZE; i++, result++) {
-			int size;
-			struct hlist_head *list;
-
-			list = &udp_hash[result & (UDP_HTABLE_SIZE - 1)];
-			if (hlist_empty(list)) {
-				if (result > sysctl_local_port_range[1])
-					result = sysctl_local_port_range[0] +
-						((result - sysctl_local_port_range[0]) &
-						 (UDP_HTABLE_SIZE - 1));
-				goto gotit;
-			}
-			size = 0;
-			sk_for_each(sk2, node, list)
-				if (++size >= best_size_so_far)
-					goto next;
-			best_size_so_far = size;
-			best = result;
-		next:;
-		}
-		result = best;
-		for(i = 0; i < (1 << 16) / UDP_HTABLE_SIZE; i++, result += UDP_HTABLE_SIZE) {
-			if (result > sysctl_local_port_range[1])
-				result = sysctl_local_port_range[0]
-					+ ((result - sysctl_local_port_range[0]) &
-					   (UDP_HTABLE_SIZE - 1));
-			if (!udp_lport_inuse(result))
-				break;
-		}
-		if (i >= (1 << 16) / UDP_HTABLE_SIZE)
-			goto fail;
-gotit:
-		udp_port_rover = snum = result;
-	} else {
-		sk_for_each(sk2, node,
-			    &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]) {
-			if (inet_sk(sk2)->num == snum &&
-			    sk2 != sk &&
-			    (!sk2->sk_bound_dev_if ||
-			     !sk->sk_bound_dev_if ||
-			     sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
-			    (!sk2->sk_reuse || !sk->sk_reuse) &&
-			    ipv6_rcv_saddr_equal(sk, sk2))
-				goto fail;
-		}
-	}
-
-	inet_sk(sk)->num = snum;
-	if (sk_unhashed(sk)) {
-		sk_add_node(sk, &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]);
-		sock_prot_inc_use(sk->sk_prot);
-	}
-	write_unlock_bh(&udp_hash_lock);
-	return 0;
-
-fail:
-	write_unlock_bh(&udp_hash_lock);
-	return 1;
+	return udp_get_port(sk, snum, ipv6_rcv_saddr_equal);
 }
 
 static void udp_v6_hash(struct sock *sk)
-- 
cgit v1.2.3


From e3b4eadbea77ecb3c3a74d1bc81b392f454c7f2e Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Sat, 26 Aug 2006 20:10:15 -0700
Subject: [UDP]: saddr_cmp function should take const socket pointers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This also kills a warning while building ipv6:

net/ipv6/udp.c: In function ‘udp_v6_get_port’:
net/ipv6/udp.c:66: warning: passing argument 3 of ‘udp_get_port’ from incompatible pointer type

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/udp.h | 2 +-
 net/ipv4/udp.c    | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/udp.h b/include/net/udp.h
index c490a0f662ac..db0c05f67546 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -48,7 +48,7 @@ extern struct proto udp_prot;
 struct sk_buff;
 
 extern int	udp_get_port(struct sock *sk, unsigned short snum,
-			     int (*saddr_cmp)(struct sock *, struct sock *));
+			     int (*saddr_cmp)(const struct sock *, const struct sock *));
 extern void	udp_err(struct sk_buff *, u32);
 
 extern int	udp_sendmsg(struct kiocb *iocb, struct sock *sk,
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index aa1823050b00..77e265d7bb8f 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -139,7 +139,7 @@ static inline int udp_lport_inuse(u16 num)
  *  @saddr_comp:  AF-dependent comparison of bound local IP addresses
  */
 int udp_get_port(struct sock *sk, unsigned short snum,
-		 int (*saddr_cmp)(struct sock *sk1, struct sock *sk2))
+		 int (*saddr_cmp)(const struct sock *sk1, const struct sock *sk2))
 {
 	struct hlist_node *node;
 	struct hlist_head *head;
@@ -210,7 +210,7 @@ fail:
 	return error;
 }
 
-static inline int  ipv4_rcv_saddr_equal(struct sock *sk1, struct sock *sk2)
+static inline int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
 {
 	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
 
-- 
cgit v1.2.3


From a5531a5d852008be40811496029012f4ad3093d1 Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Sat, 26 Aug 2006 20:11:47 -0700
Subject: [NETLINK]: Improve string attribute validation

Introduces a new attribute type NLA_NUL_STRING to support NUL
terminated strings. Attributes of this kind require to carry
a terminating NUL within the maximum specified in the policy.

The `old' NLA_STRING which is not required to be NUL terminated
is extended to provide means to specify a maximum length of the
string.

Aims at easing the pain with using nla_strlcpy() on temporary
buffers.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 15 +++++++++++----
 net/netlink/attr.c    | 49 +++++++++++++++++++++++++++++++++++++++----------
 2 files changed, 50 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index bcb27e3a312e..11dc2e7f679a 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -167,6 +167,7 @@ enum {
 	NLA_FLAG,
 	NLA_MSECS,
 	NLA_NESTED,
+	NLA_NUL_STRING,
 	__NLA_TYPE_MAX,
 };
 
@@ -175,21 +176,27 @@ enum {
 /**
  * struct nla_policy - attribute validation policy
  * @type: Type of attribute or NLA_UNSPEC
- * @minlen: Minimal length of payload required to be available
+ * @len: Type specific length of payload
  *
  * Policies are defined as arrays of this struct, the array must be
  * accessible by attribute type up to the highest identifier to be expected.
  *
+ * Meaning of `len' field:
+ *    NLA_STRING           Maximum length of string
+ *    NLA_NUL_STRING       Maximum length of string (excluding NUL)
+ *    NLA_FLAG             Unused
+ *    All other            Exact length of attribute payload
+ *
  * Example:
  * static struct nla_policy my_policy[ATTR_MAX+1] __read_mostly = {
  * 	[ATTR_FOO] = { .type = NLA_U16 },
- *	[ATTR_BAR] = { .type = NLA_STRING },
- *	[ATTR_BAZ] = { .minlen = sizeof(struct mystruct) },
+ *	[ATTR_BAR] = { .type = NLA_STRING, len = BARSIZ },
+ *	[ATTR_BAZ] = { .len = sizeof(struct mystruct) },
  * };
  */
 struct nla_policy {
 	u16		type;
-	u16		minlen;
+	u16		len;
 };
 
 /**
diff --git a/net/netlink/attr.c b/net/netlink/attr.c
index 136e529e5780..004139557e09 100644
--- a/net/netlink/attr.c
+++ b/net/netlink/attr.c
@@ -20,7 +20,6 @@ static u16 nla_attr_minlen[NLA_TYPE_MAX+1] __read_mostly = {
 	[NLA_U16]	= sizeof(u16),
 	[NLA_U32]	= sizeof(u32),
 	[NLA_U64]	= sizeof(u64),
-	[NLA_STRING]	= 1,
 	[NLA_NESTED]	= NLA_HDRLEN,
 };
 
@@ -28,7 +27,7 @@ static int validate_nla(struct nlattr *nla, int maxtype,
 			struct nla_policy *policy)
 {
 	struct nla_policy *pt;
-	int minlen = 0;
+	int minlen = 0, attrlen = nla_len(nla);
 
 	if (nla->nla_type <= 0 || nla->nla_type > maxtype)
 		return 0;
@@ -37,16 +36,46 @@ static int validate_nla(struct nlattr *nla, int maxtype,
 
 	BUG_ON(pt->type > NLA_TYPE_MAX);
 
-	if (pt->minlen)
-		minlen = pt->minlen;
-	else if (pt->type != NLA_UNSPEC)
-		minlen = nla_attr_minlen[pt->type];
+	switch (pt->type) {
+	case NLA_FLAG:
+		if (attrlen > 0)
+			return -ERANGE;
+		break;
 
-	if (pt->type == NLA_FLAG && nla_len(nla) > 0)
-		return -ERANGE;
+	case NLA_NUL_STRING:
+		if (pt->len)
+			minlen = min_t(int, attrlen, pt->len + 1);
+		else
+			minlen = attrlen;
 
-	if (nla_len(nla) < minlen)
-		return -ERANGE;
+		if (!minlen || memchr(nla_data(nla), '\0', minlen) == NULL)
+			return -EINVAL;
+		/* fall through */
+
+	case NLA_STRING:
+		if (attrlen < 1)
+			return -ERANGE;
+
+		if (pt->len) {
+			char *buf = nla_data(nla);
+
+			if (buf[attrlen - 1] == '\0')
+				attrlen--;
+
+			if (attrlen > pt->len)
+				return -ERANGE;
+		}
+		break;
+
+	default:
+		if (pt->len)
+			minlen = pt->len;
+		else if (pt->type != NLA_UNSPEC)
+			minlen = nla_attr_minlen[pt->type];
+
+		if (attrlen < minlen)
+			return -ERANGE;
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3


From def42ff4dd6f54ebcf78192579a8ff1f81d8e2e8 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Mon, 28 Aug 2006 23:57:56 -0700
Subject: [IPV4]: Make struct in_addr::s_addr __be32

There will be relatively small increase in sparse endian warnings, but
this (and sin_port) patch is a first step to make networking code
endian clean.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/in.h b/include/linux/in.h
index 94f557fa4636..9a9d5dd32e73 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -52,7 +52,7 @@ enum {
 
 /* Internet address. */
 struct in_addr {
-	__u32	s_addr;
+	__be32	s_addr;
 };
 
 #define IP_TOS		1
-- 
cgit v1.2.3


From cd360007a0eb8cbf17c006cca42aa884d33f96be Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Mon, 28 Aug 2006 23:58:32 -0700
Subject: [IPV4]: Make struct sockaddr_in::sin_port __be16

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/in.h b/include/linux/in.h
index 9a9d5dd32e73..bcaca8399aed 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -177,7 +177,7 @@ struct in_pktinfo
 #define __SOCK_SIZE__	16		/* sizeof(struct sockaddr)	*/
 struct sockaddr_in {
   sa_family_t		sin_family;	/* Address family		*/
-  unsigned short int	sin_port;	/* Port number			*/
+  __be16		sin_port;	/* Port number			*/
   struct in_addr	sin_addr;	/* Internet address		*/
 
   /* Pad to size of `struct sockaddr'. */
-- 
cgit v1.2.3


From 07317621d004e8e6967f2dac8562825267e56135 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Tue, 29 Aug 2006 17:48:17 -0700
Subject: [NETFILTER] bridge: code rearrangement for clarity

Cleanup and rearrangement for better style and clarity:
	Split the function nf_bridge_maybe_copy_header into two pieces
	Move copy portion out of line.
	Use Ethernet header size macros.
	Use header file to handle CONFIG_NETFILTER_BRIDGE differences

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_bridge.h | 26 +++++++-------------------
 net/bridge/br_forward.c          |  5 +----
 net/bridge/br_netfilter.c        | 27 +++++++++++++++++++++++++--
 3 files changed, 33 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h
index 427c67ff89e9..274fe4b33155 100644
--- a/include/linux/netfilter_bridge.h
+++ b/include/linux/netfilter_bridge.h
@@ -47,26 +47,12 @@ enum nf_br_hook_priorities {
 
 
 /* Only used in br_forward.c */
-static inline
-int nf_bridge_maybe_copy_header(struct sk_buff *skb)
+extern int nf_bridge_copy_header(struct sk_buff *skb);
+static inline int nf_bridge_maybe_copy_header(struct sk_buff *skb)
 {
-	int err;
-
-	if (skb->nf_bridge) {
-		if (skb->protocol == __constant_htons(ETH_P_8021Q)) {
-			err = skb_cow(skb, 18);
-			if (err)
-				return err;
-			memcpy(skb->data - 18, skb->nf_bridge->data, 18);
-			skb_push(skb, 4);
-		} else {
-			err = skb_cow(skb, 16);
-			if (err)
-				return err;
-			memcpy(skb->data - 16, skb->nf_bridge->data, 16);
-		}
-	}
-	return 0;
+	if (skb->nf_bridge)
+		return nf_bridge_copy_header(skb);
+  	return 0;
 }
 
 /* This is called by the IP fragmenting code and it ensures there is
@@ -90,6 +76,8 @@ struct bridge_skb_cb {
 };
 
 extern int brnf_deferred_hooks;
+#else
+#define nf_bridge_maybe_copy_header(skb)	(0)
 #endif /* CONFIG_BRIDGE_NETFILTER */
 
 #endif /* __KERNEL__ */
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 864fbbc7b24d..191b861e5e53 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -38,13 +38,10 @@ int br_dev_queue_push_xmit(struct sk_buff *skb)
 	if (packet_length(skb) > skb->dev->mtu && !skb_is_gso(skb))
 		kfree_skb(skb);
 	else {
-#ifdef CONFIG_BRIDGE_NETFILTER
 		/* ip_refrag calls ip_fragment, doesn't copy the MAC header. */
 		if (nf_bridge_maybe_copy_header(skb))
 			kfree_skb(skb);
-		else
-#endif
-		{
+		else {
 			skb_push(skb, ETH_HLEN);
 
 			dev_queue_xmit(skb);
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 05b3de888243..b498efcfe451 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -127,14 +127,37 @@ static inline struct nf_bridge_info *nf_bridge_alloc(struct sk_buff *skb)
 
 static inline void nf_bridge_save_header(struct sk_buff *skb)
 {
-        int header_size = 16;
+        int header_size = ETH_HLEN;
 
 	if (skb->protocol == htons(ETH_P_8021Q))
-		header_size = 18;
+		header_size += VLAN_HLEN;
 
 	memcpy(skb->nf_bridge->data, skb->data - header_size, header_size);
 }
 
+/*
+ * When forwarding bridge frames, we save a copy of the original
+ * header before processing.
+ */
+int nf_bridge_copy_header(struct sk_buff *skb)
+{
+	int err;
+        int header_size = ETH_HLEN;
+
+	if (skb->protocol == htons(ETH_P_8021Q))
+		header_size += VLAN_HLEN;
+
+	err = skb_cow(skb, header_size);
+	if (err)
+		return err;
+
+	memcpy(skb->data - header_size, skb->nf_bridge->data, header_size);
+
+	if (skb->protocol == htons(ETH_P_8021Q))
+		__skb_push(skb, VLAN_HLEN);
+	return 0;
+}
+
 /* PF_BRIDGE/PRE_ROUTING *********************************************/
 /* Undo the changes made for ip6tables PREROUTING and continue the
  * bridge PRE_ROUTING hook. */
-- 
cgit v1.2.3


From 9bcfcaf5e9cc887eb39236e43bdbe4b4b2572229 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Tue, 29 Aug 2006 17:48:57 -0700
Subject: [NETFILTER] bridge: simplify nf_bridge_pad

Do some simple optimization on the nf_bridge_pad() function
and don't use magic constants. Eliminate a double call and
the #ifdef'd code for CONFIG_BRIDGE_NETFILTER.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_bridge.h | 16 +++++-----------
 net/ipv4/ip_output.c             | 15 +++++++--------
 2 files changed, 12 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h
index 274fe4b33155..9a4dd11af86e 100644
--- a/include/linux/netfilter_bridge.h
+++ b/include/linux/netfilter_bridge.h
@@ -5,9 +5,8 @@
  */
 
 #include <linux/netfilter.h>
-#if defined(__KERNEL__) && defined(CONFIG_BRIDGE_NETFILTER)
 #include <linux/if_ether.h>
-#endif
+#include <linux/if_vlan.h>
 
 /* Bridge Hooks */
 /* After promisc drops, checksum checks. */
@@ -57,16 +56,10 @@ static inline int nf_bridge_maybe_copy_header(struct sk_buff *skb)
 
 /* This is called by the IP fragmenting code and it ensures there is
  * enough room for the encapsulating header (if there is one). */
-static inline
-int nf_bridge_pad(struct sk_buff *skb)
+static inline int nf_bridge_pad(const struct sk_buff *skb)
 {
-	if (skb->protocol == __constant_htons(ETH_P_IP))
-		return 0;
-	if (skb->nf_bridge) {
-		if (skb->protocol == __constant_htons(ETH_P_8021Q))
-			return 4;
-	}
-	return 0;
+ 	return (skb->nf_bridge && skb->protocol == htons(ETH_P_8021Q))
+		? VLAN_HLEN : 0;
 }
 
 struct bridge_skb_cb {
@@ -78,6 +71,7 @@ struct bridge_skb_cb {
 extern int brnf_deferred_hooks;
 #else
 #define nf_bridge_maybe_copy_header(skb)	(0)
+#define nf_bridge_pad(skb)			(0)
 #endif /* CONFIG_BRIDGE_NETFILTER */
 
 #endif /* __KERNEL__ */
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 81b2795a4c20..97aee76fb746 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -426,7 +426,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*))
 	int ptr;
 	struct net_device *dev;
 	struct sk_buff *skb2;
-	unsigned int mtu, hlen, left, len, ll_rs;
+	unsigned int mtu, hlen, left, len, ll_rs, pad;
 	int offset;
 	__be16 not_last_frag;
 	struct rtable *rt = (struct rtable*)skb->dst;
@@ -556,14 +556,13 @@ slow_path:
 	left = skb->len - hlen;		/* Space per frame */
 	ptr = raw + hlen;		/* Where to start from */
 
-#ifdef CONFIG_BRIDGE_NETFILTER
 	/* for bridged IP traffic encapsulated inside f.e. a vlan header,
-	 * we need to make room for the encapsulating header */
-	ll_rs = LL_RESERVED_SPACE_EXTRA(rt->u.dst.dev, nf_bridge_pad(skb));
-	mtu -= nf_bridge_pad(skb);
-#else
-	ll_rs = LL_RESERVED_SPACE(rt->u.dst.dev);
-#endif
+	 * we need to make room for the encapsulating header
+	 */
+	pad = nf_bridge_pad(skb);
+	ll_rs = LL_RESERVED_SPACE_EXTRA(rt->u.dst.dev, pad);
+	mtu -= pad;
+
 	/*
 	 *	Fragment the datagram.
 	 */
-- 
cgit v1.2.3


From 1b7f775209bbee6b993587bae69acb9fc12ceb17 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Aug 2006 17:54:17 -0700
Subject: [NetLabel]: remove unused function prototypes

Removed some older function prototypes for functions that no longer exist.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/cipso_ipv4.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include')

diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index c7175e725804..5aed72ab652b 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -200,15 +200,9 @@ static inline int cipso_v4_cache_add(const struct sk_buff *skb,
 
 #ifdef CONFIG_NETLABEL
 void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway);
-int cipso_v4_socket_setopt(struct socket *sock,
-			   unsigned char *opt,
-			   u32 opt_len);
 int cipso_v4_socket_setattr(const struct socket *sock,
 			    const struct cipso_v4_doi *doi_def,
 			    const struct netlbl_lsm_secattr *secattr);
-int cipso_v4_socket_getopt(const struct socket *sock,
-			   unsigned char **opt,
-			   u32 *opt_len);
 int cipso_v4_socket_getattr(const struct socket *sock,
 			    struct netlbl_lsm_secattr *secattr);
 int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
-- 
cgit v1.2.3


From 7a0e1d602288370801c353221c6a938eab925053 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Tue, 29 Aug 2006 17:56:04 -0700
Subject: [NetLabel]: add some missing #includes to various header files

Add some missing include files to the NetLabel related header files.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/cipso_ipv4.h                    | 2 ++
 include/net/netlabel.h                      | 1 +
 net/netlabel/netlabel_domainhash.h          | 4 ++++
 net/netlabel/netlabel_user.h                | 5 +++--
 security/selinux/include/selinux_netlabel.h | 9 +++++++++
 5 files changed, 19 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 5aed72ab652b..59406e0dc5b2 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -37,6 +37,8 @@
 #include <linux/types.h>
 #include <linux/rcupdate.h>
 #include <linux/list.h>
+#include <linux/net.h>
+#include <linux/skbuff.h>
 #include <net/netlabel.h>
 
 /* known doi values */
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index 7cae730832c7..fc2b72fc7e07 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -31,6 +31,7 @@
 #define _NETLABEL_H
 
 #include <linux/types.h>
+#include <linux/net.h>
 #include <linux/skbuff.h>
 #include <net/netlink.h>
 
diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h
index 9217863ce0d3..99a2287de246 100644
--- a/net/netlabel/netlabel_domainhash.h
+++ b/net/netlabel/netlabel_domainhash.h
@@ -32,6 +32,10 @@
 #ifndef _NETLABEL_DOMAINHASH_H
 #define _NETLABEL_DOMAINHASH_H
 
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+
 /* Domain hash table size */
 /* XXX - currently this number is an uneducated guess */
 #define NETLBL_DOMHSH_BITSIZE       7
diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h
index ccf237b3a128..385a6c7488c6 100644
--- a/net/netlabel/netlabel_user.h
+++ b/net/netlabel/netlabel_user.h
@@ -31,11 +31,12 @@
 #ifndef _NETLABEL_USER_H
 #define _NETLABEL_USER_H
 
+#include <linux/types.h>
 #include <linux/skbuff.h>
 #include <linux/capability.h>
-#include <linux/genetlink.h>
-#include <net/netlabel.h>
+#include <net/netlink.h>
 #include <net/genetlink.h>
+#include <net/netlabel.h>
 
 /* NetLabel NETLINK helper functions */
 
diff --git a/security/selinux/include/selinux_netlabel.h b/security/selinux/include/selinux_netlabel.h
index d69ec650cdbe..ecab4bddaaf4 100644
--- a/security/selinux/include/selinux_netlabel.h
+++ b/security/selinux/include/selinux_netlabel.h
@@ -27,6 +27,15 @@
 #ifndef _SELINUX_NETLABEL_H_
 #define _SELINUX_NETLABEL_H_
 
+#include <linux/types.h>
+#include <linux/fs.h>
+#include <linux/net.h>
+#include <linux/skbuff.h>
+#include <net/sock.h>
+
+#include "avc.h"
+#include "objsec.h"
+
 #ifdef CONFIG_NETLABEL
 void selinux_netlbl_cache_invalidate(void);
 int selinux_netlbl_socket_post_create(struct socket *sock,
-- 
cgit v1.2.3


From fda9ef5d679b07c9d9097aaf6ef7f069d794a8f9 Mon Sep 17 00:00:00 2001
From: Dmitry Mishin <dim@openvz.org>
Date: Thu, 31 Aug 2006 15:28:39 -0700
Subject: [NET]: Fix sk->sk_filter field access

Function sk_filter() is called from tcp_v{4,6}_rcv() functions with arg
needlock = 0, while socket is not locked at that moment. In order to avoid
this and similar issues in the future, use rcu for sk->sk_filter field read
protection.

Signed-off-by: Dmitry Mishin <dim@openvz.org>
Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
 include/linux/filter.h | 13 +++++++------
 include/net/sock.h     | 34 +++++++++++++++++-----------------
 net/core/filter.c      |  8 ++++----
 net/core/sock.c        | 22 +++++++++-------------
 net/dccp/ipv6.c        |  2 +-
 net/decnet/dn_nsp_in.c |  2 +-
 net/ipv4/tcp_ipv4.c    |  2 +-
 net/ipv6/tcp_ipv6.c    |  4 ++--
 net/packet/af_packet.c | 43 ++++++++++++++++++-------------------------
 net/sctp/input.c       |  2 +-
 10 files changed, 61 insertions(+), 71 deletions(-)

(limited to 'include')

diff --git a/include/linux/filter.h b/include/linux/filter.h
index c6cb8f095088..91b2e3b9251e 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -25,10 +25,10 @@
  
 struct sock_filter	/* Filter block */
 {
-        __u16	code;   /* Actual filter code */
-        __u8	jt;	/* Jump true */
-        __u8	jf;	/* Jump false */
-        __u32	k;      /* Generic multiuse field */
+	__u16	code;   /* Actual filter code */
+	__u8	jt;	/* Jump true */
+	__u8	jf;	/* Jump false */
+	__u32	k;      /* Generic multiuse field */
 };
 
 struct sock_fprog	/* Required for SO_ATTACH_FILTER. */
@@ -41,8 +41,9 @@ struct sock_fprog	/* Required for SO_ATTACH_FILTER. */
 struct sk_filter
 {
 	atomic_t		refcnt;
-        unsigned int         	len;	/* Number of filter blocks */
-        struct sock_filter     	insns[0];
+	unsigned int         	len;	/* Number of filter blocks */
+	struct rcu_head		rcu;
+	struct sock_filter     	insns[0];
 };
 
 static inline unsigned int sk_filter_len(struct sk_filter *fp)
diff --git a/include/net/sock.h b/include/net/sock.h
index 337ebec84c70..edd4d73ce7f5 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -862,30 +862,24 @@ extern void sock_init_data(struct socket *sock, struct sock *sk);
  *
  */
 
-static inline int sk_filter(struct sock *sk, struct sk_buff *skb, int needlock)
+static inline int sk_filter(struct sock *sk, struct sk_buff *skb)
 {
 	int err;
+	struct sk_filter *filter;
 	
 	err = security_sock_rcv_skb(sk, skb);
 	if (err)
 		return err;
 	
-	if (sk->sk_filter) {
-		struct sk_filter *filter;
-		
-		if (needlock)
-			bh_lock_sock(sk);
-		
-		filter = sk->sk_filter;
-		if (filter) {
-			unsigned int pkt_len = sk_run_filter(skb, filter->insns,
-							     filter->len);
-			err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM;
-		}
-
-		if (needlock)
-			bh_unlock_sock(sk);
+	rcu_read_lock_bh();
+	filter = sk->sk_filter;
+	if (filter) {
+		unsigned int pkt_len = sk_run_filter(skb, filter->insns,
+				filter->len);
+		err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM;
 	}
+ 	rcu_read_unlock_bh();
+
 	return err;
 }
 
@@ -897,6 +891,12 @@ static inline int sk_filter(struct sock *sk, struct sk_buff *skb, int needlock)
  *	Remove a filter from a socket and release its resources.
  */
  
+static inline void sk_filter_rcu_free(struct rcu_head *rcu)
+{
+	struct sk_filter *fp = container_of(rcu, struct sk_filter, rcu);
+	kfree(fp);
+}
+
 static inline void sk_filter_release(struct sock *sk, struct sk_filter *fp)
 {
 	unsigned int size = sk_filter_len(fp);
@@ -904,7 +904,7 @@ static inline void sk_filter_release(struct sock *sk, struct sk_filter *fp)
 	atomic_sub(size, &sk->sk_omem_alloc);
 
 	if (atomic_dec_and_test(&fp->refcnt))
-		kfree(fp);
+		call_rcu_bh(&fp->rcu, sk_filter_rcu_free);
 }
 
 static inline void sk_filter_charge(struct sock *sk, struct sk_filter *fp)
diff --git a/net/core/filter.c b/net/core/filter.c
index 5b4486a60cf6..6732782a5a40 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -422,10 +422,10 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk)
 	if (!err) {
 		struct sk_filter *old_fp;
 
-		spin_lock_bh(&sk->sk_lock.slock);
-		old_fp = sk->sk_filter;
-		sk->sk_filter = fp;
-		spin_unlock_bh(&sk->sk_lock.slock);
+		rcu_read_lock_bh();
+		old_fp = rcu_dereference(sk->sk_filter);
+		rcu_assign_pointer(sk->sk_filter, fp);
+		rcu_read_unlock_bh();
 		fp = old_fp;
 	}
 
diff --git a/net/core/sock.c b/net/core/sock.c
index cfaf09039b02..b77e155cbe6c 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -247,11 +247,7 @@ int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 		goto out;
 	}
 
-	/* It would be deadlock, if sock_queue_rcv_skb is used
-	   with socket lock! We assume that users of this
-	   function are lock free.
-	*/
-	err = sk_filter(sk, skb, 1);
+	err = sk_filter(sk, skb);
 	if (err)
 		goto out;
 
@@ -278,7 +274,7 @@ int sk_receive_skb(struct sock *sk, struct sk_buff *skb)
 {
 	int rc = NET_RX_SUCCESS;
 
-	if (sk_filter(sk, skb, 0))
+	if (sk_filter(sk, skb))
 		goto discard_and_relse;
 
 	skb->dev = NULL;
@@ -606,15 +602,15 @@ set_rcvbuf:
 			break;
 
 		case SO_DETACH_FILTER:
-			spin_lock_bh(&sk->sk_lock.slock);
-			filter = sk->sk_filter;
+			rcu_read_lock_bh();
+			filter = rcu_dereference(sk->sk_filter);
                         if (filter) {
-				sk->sk_filter = NULL;
-				spin_unlock_bh(&sk->sk_lock.slock);
+				rcu_assign_pointer(sk->sk_filter, NULL);
 				sk_filter_release(sk, filter);
+				rcu_read_unlock_bh();
 				break;
 			}
-			spin_unlock_bh(&sk->sk_lock.slock);
+			rcu_read_unlock_bh();
 			ret = -ENONET;
 			break;
 
@@ -884,10 +880,10 @@ void sk_free(struct sock *sk)
 	if (sk->sk_destruct)
 		sk->sk_destruct(sk);
 
-	filter = sk->sk_filter;
+	filter = rcu_dereference(sk->sk_filter);
 	if (filter) {
 		sk_filter_release(sk, filter);
-		sk->sk_filter = NULL;
+		rcu_assign_pointer(sk->sk_filter, NULL);
 	}
 
 	sock_disable_timestamp(sk);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index f9c5e12d7038..7a47399cf31f 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -970,7 +970,7 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb)
 	if (skb->protocol == htons(ETH_P_IP))
 		return dccp_v4_do_rcv(sk, skb);
 
-	if (sk_filter(sk, skb, 0))
+	if (sk_filter(sk, skb))
 		goto discard;
 
 	/*
diff --git a/net/decnet/dn_nsp_in.c b/net/decnet/dn_nsp_in.c
index 86f7f3b28e70..72ecc6e62ec4 100644
--- a/net/decnet/dn_nsp_in.c
+++ b/net/decnet/dn_nsp_in.c
@@ -586,7 +586,7 @@ static __inline__ int dn_queue_skb(struct sock *sk, struct sk_buff *skb, int sig
         	goto out;
         }
 
-	err = sk_filter(sk, skb, 0);
+	err = sk_filter(sk, skb);
 	if (err)
 		goto out;
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 23b46e36b147..39b179856082 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1104,7 +1104,7 @@ process:
 		goto discard_and_relse;
 	nf_reset(skb);
 
-	if (sk_filter(sk, skb, 0))
+	if (sk_filter(sk, skb))
 		goto discard_and_relse;
 
 	skb->dev = NULL;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 2b18918f3011..2546fc9f0a78 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1075,7 +1075,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb)
 	if (skb->protocol == htons(ETH_P_IP))
 		return tcp_v4_do_rcv(sk, skb);
 
-	if (sk_filter(sk, skb, 0))
+	if (sk_filter(sk, skb))
 		goto discard;
 
 	/*
@@ -1232,7 +1232,7 @@ process:
 	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
 		goto discard_and_relse;
 
-	if (sk_filter(sk, skb, 0))
+	if (sk_filter(sk, skb))
 		goto discard_and_relse;
 
 	skb->dev = NULL;
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 300215bdbf46..f4ccb90e6739 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -427,21 +427,24 @@ out_unlock:
 }
 #endif
 
-static inline unsigned run_filter(struct sk_buff *skb, struct sock *sk, unsigned res)
+static inline int run_filter(struct sk_buff *skb, struct sock *sk,
+							unsigned *snaplen)
 {
 	struct sk_filter *filter;
+	int err = 0;
 
-	bh_lock_sock(sk);
-	filter = sk->sk_filter;
-	/*
-	 * Our caller already checked that filter != NULL but we need to
-	 * verify that under bh_lock_sock() to be safe
-	 */
-	if (likely(filter != NULL))
-		res = sk_run_filter(skb, filter->insns, filter->len);
-	bh_unlock_sock(sk);
+	rcu_read_lock_bh();
+	filter = rcu_dereference(sk->sk_filter);
+	if (filter != NULL) {
+		err = sk_run_filter(skb, filter->insns, filter->len);
+		if (!err)
+			err = -EPERM;
+		else if (*snaplen > err)
+			*snaplen = err;
+	}
+	rcu_read_unlock_bh();
 
-	return res;
+	return err;
 }
 
 /*
@@ -491,13 +494,8 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet
 
 	snaplen = skb->len;
 
-	if (sk->sk_filter) {
-		unsigned res = run_filter(skb, sk, snaplen);
-		if (res == 0)
-			goto drop_n_restore;
-		if (snaplen > res)
-			snaplen = res;
-	}
+	if (run_filter(skb, sk, &snaplen) < 0)
+		goto drop_n_restore;
 
 	if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >=
 	    (unsigned)sk->sk_rcvbuf)
@@ -593,13 +591,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 
 	snaplen = skb->len;
 
-	if (sk->sk_filter) {
-		unsigned res = run_filter(skb, sk, snaplen);
-		if (res == 0)
-			goto drop_n_restore;
-		if (snaplen > res)
-			snaplen = res;
-	}
+	if (run_filter(skb, sk, &snaplen) < 0)
+		goto drop_n_restore;
 
 	if (sk->sk_type == SOCK_DGRAM) {
 		macoff = netoff = TPACKET_ALIGN(TPACKET_HDRLEN) + 16;
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 8a34d95602ce..03f65de75d88 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -228,7 +228,7 @@ int sctp_rcv(struct sk_buff *skb)
 		goto discard_release;
 	nf_reset(skb);
 
-	if (sk_filter(sk, skb, 1))
+	if (sk_filter(sk, skb))
                 goto discard_release;
 
 	/* Create an SCTP packet structure. */
-- 
cgit v1.2.3


From eb878e84575fbce21d2edb079eada78bfa27023d Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <hadi@cyberus.ca>
Date: Thu, 31 Aug 2006 17:42:59 -0700
Subject: [IPSEC]: output mode to take an xfrm state as input param

Expose IPSEC modes output path to take an xfrm state as input param.
This makes it consistent with the input mode processing (which already
takes the xfrm state as a param).

Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h              | 2 +-
 net/ipv4/xfrm4_mode_transport.c | 4 +---
 net/ipv4/xfrm4_mode_tunnel.c    | 3 +--
 net/ipv4/xfrm4_output.c         | 2 +-
 net/ipv6/xfrm6_mode_ro.c        | 3 +--
 net/ipv6/xfrm6_mode_transport.c | 3 +--
 net/ipv6/xfrm6_mode_tunnel.c    | 3 +--
 net/ipv6/xfrm6_output.c         | 2 +-
 8 files changed, 8 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 0acabf2a0a8f..4d6dc627df9b 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -285,7 +285,7 @@ extern void xfrm_put_type(struct xfrm_type *type);
 
 struct xfrm_mode {
 	int (*input)(struct xfrm_state *x, struct sk_buff *skb);
-	int (*output)(struct sk_buff *skb);
+	int (*output)(struct xfrm_state *x,struct sk_buff *skb);
 
 	struct module *owner;
 	unsigned int encap;
diff --git a/net/ipv4/xfrm4_mode_transport.c b/net/ipv4/xfrm4_mode_transport.c
index a9e6b3dd19c9..92676b7e4034 100644
--- a/net/ipv4/xfrm4_mode_transport.c
+++ b/net/ipv4/xfrm4_mode_transport.c
@@ -21,9 +21,8 @@
  * On exit, skb->h will be set to the start of the payload to be processed
  * by x->type->output and skb->nh will be set to the top IP header.
  */
-static int xfrm4_transport_output(struct sk_buff *skb)
+static int xfrm4_transport_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct xfrm_state *x;
 	struct iphdr *iph;
 	int ihl;
 
@@ -33,7 +32,6 @@ static int xfrm4_transport_output(struct sk_buff *skb)
 	ihl = iph->ihl * 4;
 	skb->h.raw += ihl;
 
-	x = skb->dst->xfrm;
 	skb->nh.raw = memmove(skb_push(skb, x->props.header_len), iph, ihl);
 	return 0;
 }
diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c
index 13cafbe56ce3..e23c21d31a53 100644
--- a/net/ipv4/xfrm4_mode_tunnel.c
+++ b/net/ipv4/xfrm4_mode_tunnel.c
@@ -33,10 +33,9 @@ static inline void ipip_ecn_decapsulate(struct sk_buff *skb)
  * On exit, skb->h will be set to the start of the payload to be processed
  * by x->type->output and skb->nh will be set to the top IP header.
  */
-static int xfrm4_tunnel_output(struct sk_buff *skb)
+static int xfrm4_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
-	struct xfrm_state *x = dst->xfrm;
 	struct iphdr *iph, *top_iph;
 	int flags;
 
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 5fd115f0c547..04403fb01a58 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -66,7 +66,7 @@ static int xfrm4_output_one(struct sk_buff *skb)
 		if (err)
 			goto error;
 
-		err = x->mode->output(skb);
+		err = x->mode->output(x, skb);
 		if (err)
 			goto error;
 
diff --git a/net/ipv6/xfrm6_mode_ro.c b/net/ipv6/xfrm6_mode_ro.c
index c11c335312f9..6031c16d46ca 100644
--- a/net/ipv6/xfrm6_mode_ro.c
+++ b/net/ipv6/xfrm6_mode_ro.c
@@ -43,9 +43,8 @@
  * its absence, that of the top IP header.  The value of skb->data will always
  * point to the top IP header.
  */
-static int xfrm6_ro_output(struct sk_buff *skb)
+static int xfrm6_ro_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct xfrm_state *x = skb->dst->xfrm;
 	struct ipv6hdr *iph;
 	u8 *prevhdr;
 	int hdr_len;
diff --git a/net/ipv6/xfrm6_mode_transport.c b/net/ipv6/xfrm6_mode_transport.c
index a5dce216024d..3a4b39b12bad 100644
--- a/net/ipv6/xfrm6_mode_transport.c
+++ b/net/ipv6/xfrm6_mode_transport.c
@@ -25,9 +25,8 @@
  * its absence, that of the top IP header.  The value of skb->data will always
  * point to the top IP header.
  */
-static int xfrm6_transport_output(struct sk_buff *skb)
+static int xfrm6_transport_output(struct xfrm_state *x, struct sk_buff *skb)
 {
-	struct xfrm_state *x = skb->dst->xfrm;
 	struct ipv6hdr *iph;
 	u8 *prevhdr;
 	int hdr_len;
diff --git a/net/ipv6/xfrm6_mode_tunnel.c b/net/ipv6/xfrm6_mode_tunnel.c
index 8af79be2edca..5e7d8a7d6414 100644
--- a/net/ipv6/xfrm6_mode_tunnel.c
+++ b/net/ipv6/xfrm6_mode_tunnel.c
@@ -37,10 +37,9 @@ static inline void ipip6_ecn_decapsulate(struct sk_buff *skb)
  * its absence, that of the top IP header.  The value of skb->data will always
  * point to the top IP header.
  */
-static int xfrm6_tunnel_output(struct sk_buff *skb)
+static int xfrm6_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb->dst;
-	struct xfrm_state *x = dst->xfrm;
 	struct ipv6hdr *iph, *top_iph;
 	int dsfield;
 
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index db58104e710b..c260ea104c52 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -65,7 +65,7 @@ static int xfrm6_output_one(struct sk_buff *skb)
 		if (err)
 			goto error;
 
-		err = x->mode->output(skb);
+		err = x->mode->output(x, skb);
 		if (err)
 			goto error;
 
-- 
cgit v1.2.3


From eb328111efde7bca782f340fe805756039ec6a0c Mon Sep 17 00:00:00 2001
From: Thomas Graf <tgraf@suug.ch>
Date: Mon, 18 Sep 2006 00:01:59 -0700
Subject: [GENL]: Provide more information to userspace about registered genl
 families

Additionaly exports the following information when providing
the list of registered generic netlink families:
  - protocol version
  - header size
  - maximum number of attributes
  - list of available operations including
      - id
      - flags
      - avaiability of policy and doit/dumpit function

libnl HEAD provides a utility to read this new information:

	0x0010 nlctrl version 1
	    hdrsize 0 maxattr 6
	      op GETFAMILY (0x03) [POLICY,DOIT,DUMPIT]
	0x0011 NLBL_MGMT version 1
	    hdrsize 0 maxattr 0
	      op unknown (0x02) [DOIT]
	      op unknown (0x03) [DOIT]
	      ....

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/genetlink.h | 18 ++++++++++++++++++
 include/net/genetlink.h   |  2 --
 net/netlink/genetlink.c   | 40 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 58 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h
index 84f12a41dc01..9049dc65ae51 100644
--- a/include/linux/genetlink.h
+++ b/include/linux/genetlink.h
@@ -16,6 +16,8 @@ struct genlmsghdr {
 
 #define GENL_HDRLEN	NLMSG_ALIGN(sizeof(struct genlmsghdr))
 
+#define GENL_ADMIN_PERM		0x01
+
 /*
  * List of reserved static generic netlink identifiers:
  */
@@ -43,9 +45,25 @@ enum {
 	CTRL_ATTR_UNSPEC,
 	CTRL_ATTR_FAMILY_ID,
 	CTRL_ATTR_FAMILY_NAME,
+	CTRL_ATTR_VERSION,
+	CTRL_ATTR_HDRSIZE,
+	CTRL_ATTR_MAXATTR,
+	CTRL_ATTR_OPS,
 	__CTRL_ATTR_MAX,
 };
 
 #define CTRL_ATTR_MAX (__CTRL_ATTR_MAX - 1)
 
+enum {
+	CTRL_ATTR_OP_UNSPEC,
+	CTRL_ATTR_OP_ID,
+	CTRL_ATTR_OP_FLAGS,
+	CTRL_ATTR_OP_POLICY,
+	CTRL_ATTR_OP_DOIT,
+	CTRL_ATTR_OP_DUMPIT,
+	__CTRL_ATTR_OP_MAX,
+};
+
+#define CTRL_ATTR_OP_MAX (__CTRL_ATTR_OP_MAX - 1)
+
 #endif	/* __LINUX_GENERIC_NETLINK_H */
diff --git a/include/net/genetlink.h b/include/net/genetlink.h
index 97d6d3aba9d2..4a38d85e4e25 100644
--- a/include/net/genetlink.h
+++ b/include/net/genetlink.h
@@ -27,8 +27,6 @@ struct genl_family
 	struct list_head	family_list;	/* private */
 };
 
-#define GENL_ADMIN_PERM		0x01
-
 /**
  * struct genl_info - receiving information
  * @snd_seq: sending sequence number
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index 3ac942cdb677..49bc2db7982b 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -387,7 +387,10 @@ static void genl_rcv(struct sock *sk, int len)
 static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq,
 			  u32 flags, struct sk_buff *skb, u8 cmd)
 {
+	struct nlattr *nla_ops;
+	struct genl_ops *ops;
 	void *hdr;
+	int idx = 1;
 
 	hdr = genlmsg_put(skb, pid, seq, GENL_ID_CTRL, 0, flags, cmd,
 			  family->version);
@@ -396,6 +399,37 @@ static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq,
 
 	NLA_PUT_STRING(skb, CTRL_ATTR_FAMILY_NAME, family->name);
 	NLA_PUT_U16(skb, CTRL_ATTR_FAMILY_ID, family->id);
+	NLA_PUT_U32(skb, CTRL_ATTR_VERSION, family->version);
+	NLA_PUT_U32(skb, CTRL_ATTR_HDRSIZE, family->hdrsize);
+	NLA_PUT_U32(skb, CTRL_ATTR_MAXATTR, family->maxattr);
+
+	nla_ops = nla_nest_start(skb, CTRL_ATTR_OPS);
+	if (nla_ops == NULL)
+		goto nla_put_failure;
+
+	list_for_each_entry(ops, &family->ops_list, ops_list) {
+		struct nlattr *nest;
+
+		nest = nla_nest_start(skb, idx++);
+		if (nest == NULL)
+			goto nla_put_failure;
+
+		NLA_PUT_U32(skb, CTRL_ATTR_OP_ID, ops->cmd);
+		NLA_PUT_U32(skb, CTRL_ATTR_OP_FLAGS, ops->flags);
+
+		if (ops->policy)
+			NLA_PUT_FLAG(skb, CTRL_ATTR_OP_POLICY);
+
+		if (ops->doit)
+			NLA_PUT_FLAG(skb, CTRL_ATTR_OP_DOIT);
+
+		if (ops->dumpit)
+			NLA_PUT_FLAG(skb, CTRL_ATTR_OP_DUMPIT);
+
+		nla_nest_end(skb, nest);
+	}
+
+	nla_nest_end(skb, nla_ops);
 
 	return genlmsg_end(skb, hdr);
 
@@ -411,6 +445,9 @@ static int ctrl_dumpfamily(struct sk_buff *skb, struct netlink_callback *cb)
 	int chains_to_skip = cb->args[0];
 	int fams_to_skip = cb->args[1];
 
+	if (chains_to_skip != 0)
+		genl_lock();
+
 	for (i = 0; i < GENL_FAM_TAB_SIZE; i++) {
 		if (i < chains_to_skip)
 			continue;
@@ -428,6 +465,9 @@ static int ctrl_dumpfamily(struct sk_buff *skb, struct netlink_callback *cb)
 	}
 
 errout:
+	if (chains_to_skip != 0)
+		genl_unlock();
+
 	cb->args[0] = i;
 	cb->args[1] = n;
 
-- 
cgit v1.2.3


From 161643660129dd7d98f0b12418c0a2710ffa7db6 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 18 Sep 2006 00:40:38 -0700
Subject: [SCTP]: Cleanups

This patch contains the following cleanups:
- make the following needlessly global function static:
  - socket.c: sctp_apply_peer_addr_params()
- add proper prototypes for the several global functions in
  include/net/sctp/sctp.h

Note that this fixes wrong prototypes for the following functions:
- sctp_snmp_proc_exit()
- sctp_eps_proc_exit()
- sctp_assocs_proc_exit()

The latter was spotted by the GNU C compiler and reported
by David Woodhouse.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/sctp.h | 13 +++++++++++++
 net/sctp/ipv6.c         |  1 -
 net/sctp/protocol.c     |  7 -------
 net/sctp/socket.c       | 14 +++++++-------
 4 files changed, 20 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index e274fd479990..ee68a3124076 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -128,6 +128,8 @@ extern int sctp_copy_local_addr_list(struct sctp_bind_addr *,
 				     int flags);
 extern struct sctp_pf *sctp_get_pf_specific(sa_family_t family);
 extern int sctp_register_pf(struct sctp_pf *, sa_family_t);
+int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev,
+                        void *ptr);
 
 /*
  * sctp/socket.c
@@ -177,6 +179,17 @@ void sctp_icmp_proto_unreachable(struct sock *sk,
 void sctp_backlog_migrate(struct sctp_association *assoc,
 			  struct sock *oldsk, struct sock *newsk);
 
+/*
+ * sctp/proc.c
+ */
+int sctp_snmp_proc_init(void);
+void sctp_snmp_proc_exit(void);
+int sctp_eps_proc_init(void);
+void sctp_eps_proc_exit(void);
+int sctp_assocs_proc_init(void);
+void sctp_assocs_proc_exit(void);
+
+
 /*
  *  Section:  Macros, externs, and inlines
  */
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index 99c0cefc04e0..fd87e3ceb56e 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -78,7 +78,6 @@
 
 #include <asm/uaccess.h>
 
-extern int sctp_inetaddr_event(struct notifier_block *, unsigned long, void *);
 static struct notifier_block sctp_inet6addr_notifier = {
 	.notifier_call = sctp_inetaddr_event,
 };
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index d9dd4c47bc29..fac7674438a4 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -82,13 +82,6 @@ static struct sctp_af *sctp_af_v6_specific;
 kmem_cache_t *sctp_chunk_cachep __read_mostly;
 kmem_cache_t *sctp_bucket_cachep __read_mostly;
 
-extern int sctp_snmp_proc_init(void);
-extern int sctp_snmp_proc_exit(void);
-extern int sctp_eps_proc_init(void);
-extern int sctp_eps_proc_exit(void);
-extern int sctp_assocs_proc_init(void);
-extern int sctp_assocs_proc_exit(void);
-
 /* Return the address of the control sock. */
 struct sock *sctp_get_ctl_sock(void)
 {
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 7c1dbb1d10df..79c3e072cf28 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -2081,13 +2081,13 @@ static int sctp_setsockopt_autoclose(struct sock *sk, char __user *optval,
  *                     SPP_SACKDELAY_ENABLE, setting both will have undefined
  *                     results.
  */
-int sctp_apply_peer_addr_params(struct sctp_paddrparams *params,
-				struct sctp_transport   *trans,
-				struct sctp_association *asoc,
-				struct sctp_sock        *sp,
-				int                      hb_change,
-				int                      pmtud_change,
-				int                      sackdelay_change)
+static int sctp_apply_peer_addr_params(struct sctp_paddrparams *params,
+				       struct sctp_transport   *trans,
+				       struct sctp_association *asoc,
+				       struct sctp_sock        *sp,
+				       int                      hb_change,
+				       int                      pmtud_change,
+				       int                      sackdelay_change)
 {
 	int error;
 
-- 
cgit v1.2.3


From 1ef9696c909060ccdae3ade245ca88692b49285b Mon Sep 17 00:00:00 2001
From: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Date: Tue, 19 Sep 2006 12:52:50 -0700
Subject: [TCP]: Send ACKs each 2nd received segment.

It does not affect either mss-sized connections (obviously) or
connections controlled by Nagle (because there is only one small
segment in flight).

The idea is to record the fact that a small segment arrives on a
connection, where one small segment has already been received and
still not-ACKed. In this case ACK is forced after tcp_recvmsg() drains
receive buffer.

In other words, it is a "soft" each-2nd-segment ACK, which is enough
to preserve ACK clock even when ABC is enabled.

Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_connection_sock.h | 3 ++-
 net/ipv4/tcp.c                     | 7 +++++--
 net/ipv4/tcp_input.c               | 2 ++
 3 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 9bf73fe50948..de4e83b6da4b 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -147,7 +147,8 @@ extern struct sock *inet_csk_clone(struct sock *sk,
 enum inet_csk_ack_state_t {
 	ICSK_ACK_SCHED	= 1,
 	ICSK_ACK_TIMER  = 2,
-	ICSK_ACK_PUSHED = 4
+	ICSK_ACK_PUSHED = 4,
+	ICSK_ACK_PUSHED2 = 8
 };
 
 extern void inet_csk_init_xmit_timers(struct sock *sk,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 29e3d606db78..66e9a729f6df 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -955,8 +955,11 @@ void tcp_cleanup_rbuf(struct sock *sk, int copied)
 		     * receive buffer and there was a small segment
 		     * in queue.
 		     */
-		    (copied > 0 && (icsk->icsk_ack.pending & ICSK_ACK_PUSHED) &&
-		     !icsk->icsk_ack.pingpong && !atomic_read(&sk->sk_rmem_alloc)))
+		    (copied > 0 &&
+		     ((icsk->icsk_ack.pending & ICSK_ACK_PUSHED2) ||
+		      ((icsk->icsk_ack.pending & ICSK_ACK_PUSHED) &&
+		       !icsk->icsk_ack.pingpong)) &&
+		      !atomic_read(&sk->sk_rmem_alloc)))
 			time_to_ack = 1;
 	}
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 511b738f118a..b3def0df14fb 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -156,6 +156,8 @@ static void tcp_measure_rcv_mss(struct sock *sk,
 				return;
 			}
 		}
+		if (icsk->icsk_ack.pending & ICSK_ACK_PUSHED)
+			icsk->icsk_ack.pending |= ICSK_ACK_PUSHED2;
 		icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
 	}
 }
-- 
cgit v1.2.3


From a1e59abf824969554b90facd44a4ab16e265afa4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 19 Sep 2006 12:57:34 -0700
Subject: [XFRM]: Fix wildcard as tunnel source

Hashing SAs by source address breaks templates with wildcards as tunnel
source since the source address used for hashing/lookup is still 0/0.
Move source address lookup to xfrm_tmpl_resolve_one() so we can use the
real address in the lookup.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      | 13 +++++++++++++
 net/ipv4/xfrm4_policy.c | 20 ++++++++++++++++++++
 net/ipv4/xfrm4_state.c  | 15 ---------------
 net/ipv6/xfrm6_policy.c | 21 +++++++++++++++++++++
 net/ipv6/xfrm6_state.c  | 16 ----------------
 net/xfrm/xfrm_policy.c  | 21 +++++++++++++++++++++
 6 files changed, 75 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 4d6dc627df9b..11e0b1d6bd47 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -222,6 +222,7 @@ struct xfrm_policy_afinfo {
 	struct dst_ops		*dst_ops;
 	void			(*garbage_collect)(void);
 	int			(*dst_lookup)(struct xfrm_dst **dst, struct flowi *fl);
+	int			(*get_saddr)(xfrm_address_t *saddr, xfrm_address_t *daddr);
 	struct dst_entry	*(*find_bundle)(struct flowi *fl, struct xfrm_policy *policy);
 	int			(*bundle_create)(struct xfrm_policy *policy, 
 						 struct xfrm_state **xfrm, 
@@ -630,6 +631,18 @@ secpath_reset(struct sk_buff *skb)
 #endif
 }
 
+static inline int
+xfrm_addr_any(xfrm_address_t *addr, unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		return addr->a4 == 0;
+	case AF_INET6:
+		return ipv6_addr_any((struct in6_addr *)&addr->a6);
+	}
+	return 0;
+}
+
 static inline int
 __xfrm4_state_addr_cmp(struct xfrm_tmpl *tmpl, struct xfrm_state *x)
 {
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 479598566f1d..eabcd27b1767 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -21,6 +21,25 @@ static int xfrm4_dst_lookup(struct xfrm_dst **dst, struct flowi *fl)
 	return __ip_route_output_key((struct rtable**)dst, fl);
 }
 
+static int xfrm4_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
+{
+	struct rtable *rt;
+	struct flowi fl_tunnel = {
+		.nl_u = {
+			.ip4_u = {
+				.daddr = daddr->a4,
+			},
+		},
+	};
+
+	if (!xfrm4_dst_lookup((struct xfrm_dst **)&rt, &fl_tunnel)) {
+		saddr->a4 = rt->rt_src;
+		dst_release(&rt->u.dst);
+		return 0;
+	}
+	return -EHOSTUNREACH;
+}
+
 static struct dst_entry *
 __xfrm4_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 {
@@ -298,6 +317,7 @@ static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
 	.family = 		AF_INET,
 	.dst_ops =		&xfrm4_dst_ops,
 	.dst_lookup =		xfrm4_dst_lookup,
+	.get_saddr =		xfrm4_get_saddr,
 	.find_bundle = 		__xfrm4_find_bundle,
 	.bundle_create =	__xfrm4_bundle_create,
 	.decode_session =	_decode_session4,
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index 6a2a4ab42772..fe2034494d08 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -42,21 +42,6 @@ __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->props.saddr = tmpl->saddr;
 	if (x->props.saddr.a4 == 0)
 		x->props.saddr.a4 = saddr->a4;
-	if (tmpl->mode == XFRM_MODE_TUNNEL && x->props.saddr.a4 == 0) {
-		struct rtable *rt;
-	        struct flowi fl_tunnel = {
-        	        .nl_u = {
-        			.ip4_u = {
-					.daddr = x->id.daddr.a4,
-				}
-			}
-		};
-		if (!xfrm_dst_lookup((struct xfrm_dst **)&rt,
-		                     &fl_tunnel, AF_INET)) {
-			x->props.saddr.a4 = rt->rt_src;
-			dst_release(&rt->u.dst);
-		}
-	}
 	x->props.mode = tmpl->mode;
 	x->props.reqid = tmpl->reqid;
 	x->props.family = AF_INET;
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 9391c4c94feb..6a252e2134d1 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -34,6 +34,26 @@ static int xfrm6_dst_lookup(struct xfrm_dst **dst, struct flowi *fl)
 	return err;
 }
 
+static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
+{
+	struct rt6_info *rt;
+	struct flowi fl_tunnel = {
+		.nl_u = {
+			.ip6_u = {
+				.daddr = *(struct in6_addr *)&daddr->a6,
+			},
+		},
+	};
+
+	if (!xfrm6_dst_lookup((struct xfrm_dst **)&rt, &fl_tunnel)) {
+		ipv6_get_saddr(&rt->u.dst, (struct in6_addr *)&daddr->a6,
+			       (struct in6_addr *)&saddr->a6);
+		dst_release(&rt->u.dst);
+		return 0;
+	}
+	return -EHOSTUNREACH;
+}
+
 static struct dst_entry *
 __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 {
@@ -362,6 +382,7 @@ static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
 	.family =		AF_INET6,
 	.dst_ops =		&xfrm6_dst_ops,
 	.dst_lookup =		xfrm6_dst_lookup,
+	.get_saddr = 		xfrm6_get_saddr,
 	.find_bundle =		__xfrm6_find_bundle,
 	.bundle_create =	__xfrm6_bundle_create,
 	.decode_session =	_decode_session6,
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index d88cd92c864e..711bfafb2472 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -42,22 +42,6 @@ __xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	memcpy(&x->props.saddr, &tmpl->saddr, sizeof(x->props.saddr));
 	if (ipv6_addr_any((struct in6_addr*)&x->props.saddr))
 		memcpy(&x->props.saddr, saddr, sizeof(x->props.saddr));
-	if (tmpl->mode == XFRM_MODE_TUNNEL && ipv6_addr_any((struct in6_addr*)&x->props.saddr)) {
-		struct rt6_info *rt;
-		struct flowi fl_tunnel = {
-			.nl_u = {
-				.ip6_u = {
-					.daddr = *(struct in6_addr *)daddr,
-				}
-			}
-		};
-		if (!xfrm_dst_lookup((struct xfrm_dst **)&rt,
-		                     &fl_tunnel, AF_INET6)) {
-			ipv6_get_saddr(&rt->u.dst, (struct in6_addr *)daddr,
-			               (struct in6_addr *)&x->props.saddr);
-			dst_release(&rt->u.dst);
-		}
-	}
 	x->props.mode = tmpl->mode;
 	x->props.reqid = tmpl->reqid;
 	x->props.family = AF_INET6;
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 537854fe47ca..b6e2e79d7261 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1107,6 +1107,20 @@ int __xfrm_sk_clone_policy(struct sock *sk)
 	return 0;
 }
 
+static int
+xfrm_get_saddr(xfrm_address_t *local, xfrm_address_t *remote,
+	       unsigned short family)
+{
+	int err;
+	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
+
+	if (unlikely(afinfo == NULL))
+		return -EINVAL;
+	err = afinfo->get_saddr(local, remote);
+	xfrm_policy_put_afinfo(afinfo);
+	return err;
+}
+
 /* Resolve list of templates for the flow, given policy. */
 
 static int
@@ -1118,6 +1132,7 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, struct flowi *fl,
 	int i, error;
 	xfrm_address_t *daddr = xfrm_flowi_daddr(fl, family);
 	xfrm_address_t *saddr = xfrm_flowi_saddr(fl, family);
+	xfrm_address_t tmp;
 
 	for (nx=0, i = 0; i < policy->xfrm_nr; i++) {
 		struct xfrm_state *x;
@@ -1128,6 +1143,12 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, struct flowi *fl,
 		if (tmpl->mode == XFRM_MODE_TUNNEL) {
 			remote = &tmpl->id.daddr;
 			local = &tmpl->saddr;
+			if (xfrm_addr_any(local, family)) {
+				error = xfrm_get_saddr(&tmp, remote, family);
+				if (error)
+					goto fail;
+				local = &tmp;
+			}
 		}
 
 		x = xfrm_state_find(remote, local, fl, tmpl, policy, &error, family);
-- 
cgit v1.2.3


From 1bf38a36b6a0e810dafae048fdbb999e587f0f2f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 11:57:09 -0700
Subject: [NETFILTER]: remove unused include file

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_logging.h | 33 ---------------------------------
 1 file changed, 33 deletions(-)
 delete mode 100644 include/linux/netfilter_logging.h

(limited to 'include')

diff --git a/include/linux/netfilter_logging.h b/include/linux/netfilter_logging.h
deleted file mode 100644
index 562bb6aad4e1..000000000000
--- a/include/linux/netfilter_logging.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/* Internal logging interface, which relies on the real 
-   LOG target modules */
-#ifndef __LINUX_NETFILTER_LOGGING_H
-#define __LINUX_NETFILTER_LOGGING_H
-
-#ifdef __KERNEL__
-#include <asm/atomic.h>
-
-struct nf_logging_t {
-	void (*nf_log_packet)(struct sk_buff **pskb,
-			      unsigned int hooknum,
-			      const struct net_device *in,
-			      const struct net_device *out,
-			      const char *prefix);
-	void (*nf_log)(char *pfh, size_t len,
-		       const char *prefix);
-};
-
-extern void nf_log_register(int pf, const struct nf_logging_t *logging);
-extern void nf_log_unregister(int pf, const struct nf_logging_t *logging);
-
-extern void nf_log_packet(int pf,
-			  struct sk_buff **pskb,
-			  unsigned int hooknum,
-			  const struct net_device *in,
-			  const struct net_device *out,
-			  const char *fmt, ...);
-extern void nf_log(int pf,
-		   char *pfh, size_t len,
-		   const char *fmt, ...);
-#endif /*__KERNEL__*/
-
-#endif /*__LINUX_NETFILTER_LOGGING_H*/
-- 
cgit v1.2.3


From df0933dcb027e156cb5253570ad694b81bd52b69 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 11:57:53 -0700
Subject: [NETFILTER]: kill listhelp.h

Kill listhelp.h and use the list.h functions instead.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h           |   4 -
 include/linux/netfilter_ipv4/listhelp.h      | 123 -----------------
 net/bridge/netfilter/ebtables.c              |  76 ++++++-----
 net/ipv4/netfilter/arp_tables.c              |   2 -
 net/ipv4/netfilter/ip_conntrack_core.c       | 189 ++++++++++++---------------
 net/ipv4/netfilter/ip_conntrack_proto_gre.c  |  24 ++--
 net/ipv4/netfilter/ip_conntrack_standalone.c |   1 -
 net/ipv4/netfilter/ip_nat_core.c             |   4 -
 net/ipv4/netfilter/ip_nat_helper.c           |   4 -
 net/ipv4/netfilter/ip_nat_rule.c             |   4 -
 net/ipv4/netfilter/ip_nat_standalone.c       |   4 -
 net/ipv6/netfilter/ip6_tables.c              |   3 -
 net/netfilter/nf_conntrack_core.c            | 185 ++++++++++++--------------
 net/netfilter/nf_conntrack_standalone.c      |   1 -
 net/netfilter/x_tables.c                     |  17 ++-
 15 files changed, 237 insertions(+), 404 deletions(-)
 delete mode 100644 include/linux/netfilter_ipv4/listhelp.h

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 03d1027fb0e8..c832295dbf61 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -138,10 +138,6 @@ struct xt_counters_info
 
 #include <linux/netdevice.h>
 
-#define ASSERT_READ_LOCK(x)
-#define ASSERT_WRITE_LOCK(x)
-#include <linux/netfilter_ipv4/listhelp.h>
-
 #ifdef CONFIG_COMPAT
 #define COMPAT_TO_USER		1
 #define COMPAT_FROM_USER	-1
diff --git a/include/linux/netfilter_ipv4/listhelp.h b/include/linux/netfilter_ipv4/listhelp.h
deleted file mode 100644
index 5d92cf044d91..000000000000
--- a/include/linux/netfilter_ipv4/listhelp.h
+++ /dev/null
@@ -1,123 +0,0 @@
-#ifndef _LISTHELP_H
-#define _LISTHELP_H
-#include <linux/list.h>
-
-/* Header to do more comprehensive job than linux/list.h; assume list
-   is first entry in structure. */
-
-/* Return pointer to first true entry, if any, or NULL.  A macro
-   required to allow inlining of cmpfn. */
-#define LIST_FIND(head, cmpfn, type, args...)		\
-({							\
-	const struct list_head *__i, *__j = NULL;	\
-							\
-	ASSERT_READ_LOCK(head);				\
-	list_for_each(__i, (head))			\
-		if (cmpfn((const type)__i , ## args)) {	\
-			__j = __i;			\
-			break;				\
-		}					\
-	(type)__j;					\
-})
-
-#define LIST_FIND_W(head, cmpfn, type, args...)		\
-({							\
-	const struct list_head *__i, *__j = NULL;	\
-							\
-	ASSERT_WRITE_LOCK(head);			\
-	list_for_each(__i, (head))			\
-		if (cmpfn((type)__i , ## args)) {	\
-			__j = __i;			\
-			break;				\
-		}					\
-	(type)__j;					\
-})
-
-/* Just like LIST_FIND but we search backwards */
-#define LIST_FIND_B(head, cmpfn, type, args...)		\
-({							\
-	const struct list_head *__i, *__j = NULL;	\
-							\
-	ASSERT_READ_LOCK(head);				\
-	list_for_each_prev(__i, (head))			\
-		if (cmpfn((const type)__i , ## args)) {	\
-			__j = __i;			\
-			break;				\
-		}					\
-	(type)__j;					\
-})
-
-static inline int
-__list_cmp_same(const void *p1, const void *p2) { return p1 == p2; }
-
-/* Is this entry in the list? */
-static inline int
-list_inlist(struct list_head *head, const void *entry)
-{
-	return LIST_FIND(head, __list_cmp_same, void *, entry) != NULL;
-}
-
-/* Delete from list. */
-#ifdef CONFIG_NETFILTER_DEBUG
-#define LIST_DELETE(head, oldentry)					\
-do {									\
-	ASSERT_WRITE_LOCK(head);					\
-	if (!list_inlist(head, oldentry))				\
-		printk("LIST_DELETE: %s:%u `%s'(%p) not in %s.\n",	\
-		       __FILE__, __LINE__, #oldentry, oldentry, #head);	\
-        else list_del((struct list_head *)oldentry);			\
-} while(0)
-#else
-#define LIST_DELETE(head, oldentry) list_del((struct list_head *)oldentry)
-#endif
-
-/* Append. */
-static inline void
-list_append(struct list_head *head, void *new)
-{
-	ASSERT_WRITE_LOCK(head);
-	list_add((new), (head)->prev);
-}
-
-/* Prepend. */
-static inline void
-list_prepend(struct list_head *head, void *new)
-{
-	ASSERT_WRITE_LOCK(head);
-	list_add(new, head);
-}
-
-/* Insert according to ordering function; insert before first true. */
-#define LIST_INSERT(head, new, cmpfn)				\
-do {								\
-	struct list_head *__i;					\
-	ASSERT_WRITE_LOCK(head);				\
-	list_for_each(__i, (head))				\
-		if ((new), (typeof (new))__i)			\
-			break;					\
-	list_add((struct list_head *)(new), __i->prev);		\
-} while(0)
-
-/* If the field after the list_head is a nul-terminated string, you
-   can use these functions. */
-static inline int __list_cmp_name(const void *i, const char *name)
-{
-	return strcmp(name, i+sizeof(struct list_head)) == 0;
-}
-
-/* Returns false if same name already in list, otherwise does insert. */
-static inline int
-list_named_insert(struct list_head *head, void *new)
-{
-	if (LIST_FIND(head, __list_cmp_name, void *,
-		      new + sizeof(struct list_head)))
-		return 0;
-	list_prepend(head, new);
-	return 1;
-}
-
-/* Find this named element in the list. */
-#define list_named_find(head, name)			\
-LIST_FIND(head, __list_cmp_name, void *, name)
-
-#endif /*_LISTHELP_H*/
diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c
index d06a5075b5f6..3df55b2bd91d 100644
--- a/net/bridge/netfilter/ebtables.c
+++ b/net/bridge/netfilter/ebtables.c
@@ -24,6 +24,7 @@
 #include <linux/vmalloc.h>
 #include <linux/netfilter_bridge/ebtables.h>
 #include <linux/spinlock.h>
+#include <linux/mutex.h>
 #include <asm/uaccess.h>
 #include <linux/smp.h>
 #include <linux/cpumask.h>
@@ -31,12 +32,6 @@
 /* needed for logical [in,out]-dev filtering */
 #include "../br_private.h"
 
-/* list_named_find */
-#define ASSERT_READ_LOCK(x)
-#define ASSERT_WRITE_LOCK(x)
-#include <linux/netfilter_ipv4/listhelp.h>
-#include <linux/mutex.h>
-
 #define BUGPRINT(format, args...) printk("kernel msg: ebtables bug: please "\
                                          "report to author: "format, ## args)
 /* #define BUGPRINT(format, args...) */
@@ -278,18 +273,22 @@ static inline void *
 find_inlist_lock_noload(struct list_head *head, const char *name, int *error,
    struct mutex *mutex)
 {
-	void *ret;
+	struct {
+		struct list_head list;
+		char name[EBT_FUNCTION_MAXNAMELEN];
+	} *e;
 
 	*error = mutex_lock_interruptible(mutex);
 	if (*error != 0)
 		return NULL;
 
-	ret = list_named_find(head, name);
-	if (!ret) {
-		*error = -ENOENT;
-		mutex_unlock(mutex);
+	list_for_each_entry(e, head, list) {
+		if (strcmp(e->name, name) == 0)
+			return e;
 	}
-	return ret;
+	*error = -ENOENT;
+	mutex_unlock(mutex);
+	return NULL;
 }
 
 #ifndef CONFIG_KMOD
@@ -1043,15 +1042,19 @@ free_newinfo:
 
 int ebt_register_target(struct ebt_target *target)
 {
+	struct ebt_target *t;
 	int ret;
 
 	ret = mutex_lock_interruptible(&ebt_mutex);
 	if (ret != 0)
 		return ret;
-	if (!list_named_insert(&ebt_targets, target)) {
-		mutex_unlock(&ebt_mutex);
-		return -EEXIST;
+	list_for_each_entry(t, &ebt_targets, list) {
+		if (strcmp(t->name, target->name) == 0) {
+			mutex_unlock(&ebt_mutex);
+			return -EEXIST;
+		}
 	}
+	list_add(&target->list, &ebt_targets);
 	mutex_unlock(&ebt_mutex);
 
 	return 0;
@@ -1060,21 +1063,25 @@ int ebt_register_target(struct ebt_target *target)
 void ebt_unregister_target(struct ebt_target *target)
 {
 	mutex_lock(&ebt_mutex);
-	LIST_DELETE(&ebt_targets, target);
+	list_del(&target->list);
 	mutex_unlock(&ebt_mutex);
 }
 
 int ebt_register_match(struct ebt_match *match)
 {
+	struct ebt_match *m;
 	int ret;
 
 	ret = mutex_lock_interruptible(&ebt_mutex);
 	if (ret != 0)
 		return ret;
-	if (!list_named_insert(&ebt_matches, match)) {
-		mutex_unlock(&ebt_mutex);
-		return -EEXIST;
+	list_for_each_entry(m, &ebt_matches, list) {
+		if (strcmp(m->name, match->name) == 0) {
+			mutex_unlock(&ebt_mutex);
+			return -EEXIST;
+		}
 	}
+	list_add(&match->list, &ebt_matches);
 	mutex_unlock(&ebt_mutex);
 
 	return 0;
@@ -1083,21 +1090,25 @@ int ebt_register_match(struct ebt_match *match)
 void ebt_unregister_match(struct ebt_match *match)
 {
 	mutex_lock(&ebt_mutex);
-	LIST_DELETE(&ebt_matches, match);
+	list_del(&match->list);
 	mutex_unlock(&ebt_mutex);
 }
 
 int ebt_register_watcher(struct ebt_watcher *watcher)
 {
+	struct ebt_watcher *w;
 	int ret;
 
 	ret = mutex_lock_interruptible(&ebt_mutex);
 	if (ret != 0)
 		return ret;
-	if (!list_named_insert(&ebt_watchers, watcher)) {
-		mutex_unlock(&ebt_mutex);
-		return -EEXIST;
+	list_for_each_entry(w, &ebt_watchers, list) {
+		if (strcmp(w->name, watcher->name) == 0) {
+			mutex_unlock(&ebt_mutex);
+			return -EEXIST;
+		}
 	}
+	list_add(&watcher->list, &ebt_watchers);
 	mutex_unlock(&ebt_mutex);
 
 	return 0;
@@ -1106,13 +1117,14 @@ int ebt_register_watcher(struct ebt_watcher *watcher)
 void ebt_unregister_watcher(struct ebt_watcher *watcher)
 {
 	mutex_lock(&ebt_mutex);
-	LIST_DELETE(&ebt_watchers, watcher);
+	list_del(&watcher->list);
 	mutex_unlock(&ebt_mutex);
 }
 
 int ebt_register_table(struct ebt_table *table)
 {
 	struct ebt_table_info *newinfo;
+	struct ebt_table *t;
 	int ret, i, countersize;
 
 	if (!table || !table->table ||!table->table->entries ||
@@ -1158,10 +1170,12 @@ int ebt_register_table(struct ebt_table *table)
 	if (ret != 0)
 		goto free_chainstack;
 
-	if (list_named_find(&ebt_tables, table->name)) {
-		ret = -EEXIST;
-		BUGPRINT("Table name already exists\n");
-		goto free_unlock;
+	list_for_each_entry(t, &ebt_tables, list) {
+		if (strcmp(t->name, table->name) == 0) {
+			ret = -EEXIST;
+			BUGPRINT("Table name already exists\n");
+			goto free_unlock;
+		}
 	}
 
 	/* Hold a reference count if the chains aren't empty */
@@ -1169,7 +1183,7 @@ int ebt_register_table(struct ebt_table *table)
 		ret = -ENOENT;
 		goto free_unlock;
 	}
-	list_prepend(&ebt_tables, table);
+	list_add(&table->list, &ebt_tables);
 	mutex_unlock(&ebt_mutex);
 	return 0;
 free_unlock:
@@ -1195,7 +1209,7 @@ void ebt_unregister_table(struct ebt_table *table)
 		return;
 	}
 	mutex_lock(&ebt_mutex);
-	LIST_DELETE(&ebt_tables, table);
+	list_del(&table->list);
 	mutex_unlock(&ebt_mutex);
 	vfree(table->private->entries);
 	if (table->private->chainstack) {
@@ -1465,7 +1479,7 @@ static int __init ebtables_init(void)
 	int ret;
 
 	mutex_lock(&ebt_mutex);
-	list_named_insert(&ebt_targets, &ebt_standard_target);
+	list_add(&ebt_standard_target.list, &ebt_targets);
 	mutex_unlock(&ebt_mutex);
 	if ((ret = nf_register_sockopt(&ebt_sockopts)) < 0)
 		return ret;
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 4f10b06413a1..aaeaa9ce0f28 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -56,8 +56,6 @@ do {								\
 #define ARP_NF_ASSERT(x)
 #endif
 
-#include <linux/netfilter_ipv4/listhelp.h>
-
 static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap,
 				      char *hdr_addr, int len)
 {
diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c
index 5da25ad50309..2568d480e9a9 100644
--- a/net/ipv4/netfilter/ip_conntrack_core.c
+++ b/net/ipv4/netfilter/ip_conntrack_core.c
@@ -47,7 +47,6 @@
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
 #include <linux/netfilter_ipv4/ip_conntrack_core.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #define IP_CONNTRACK_VERSION	"2.4"
 
@@ -294,15 +293,10 @@ void ip_ct_remove_expectations(struct ip_conntrack *ct)
 static void
 clean_from_lists(struct ip_conntrack *ct)
 {
-	unsigned int ho, hr;
-	
 	DEBUGP("clean_from_lists(%p)\n", ct);
 	ASSERT_WRITE_LOCK(&ip_conntrack_lock);
-
-	ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
-	hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
-	LIST_DELETE(&ip_conntrack_hash[ho], &ct->tuplehash[IP_CT_DIR_ORIGINAL]);
-	LIST_DELETE(&ip_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]);
+	list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
+	list_del(&ct->tuplehash[IP_CT_DIR_REPLY].list);
 
 	/* Destroy all pending expectations */
 	ip_ct_remove_expectations(ct);
@@ -367,16 +361,6 @@ static void death_by_timeout(unsigned long ul_conntrack)
 	ip_conntrack_put(ct);
 }
 
-static inline int
-conntrack_tuple_cmp(const struct ip_conntrack_tuple_hash *i,
-		    const struct ip_conntrack_tuple *tuple,
-		    const struct ip_conntrack *ignored_conntrack)
-{
-	ASSERT_READ_LOCK(&ip_conntrack_lock);
-	return tuplehash_to_ctrack(i) != ignored_conntrack
-		&& ip_ct_tuple_equal(tuple, &i->tuple);
-}
-
 struct ip_conntrack_tuple_hash *
 __ip_conntrack_find(const struct ip_conntrack_tuple *tuple,
 		    const struct ip_conntrack *ignored_conntrack)
@@ -386,7 +370,8 @@ __ip_conntrack_find(const struct ip_conntrack_tuple *tuple,
 
 	ASSERT_READ_LOCK(&ip_conntrack_lock);
 	list_for_each_entry(h, &ip_conntrack_hash[hash], list) {
-		if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) {
+		if (tuplehash_to_ctrack(h) != ignored_conntrack &&
+		    ip_ct_tuple_equal(tuple, &h->tuple)) {
 			CONNTRACK_STAT_INC(found);
 			return h;
 		}
@@ -417,10 +402,10 @@ static void __ip_conntrack_hash_insert(struct ip_conntrack *ct,
 					unsigned int repl_hash) 
 {
 	ct->id = ++ip_conntrack_next_id;
-	list_prepend(&ip_conntrack_hash[hash],
-		     &ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
-	list_prepend(&ip_conntrack_hash[repl_hash],
-		     &ct->tuplehash[IP_CT_DIR_REPLY].list);
+	list_add(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list,
+		 &ip_conntrack_hash[hash]);
+	list_add(&ct->tuplehash[IP_CT_DIR_REPLY].list,
+		 &ip_conntrack_hash[repl_hash]);
 }
 
 void ip_conntrack_hash_insert(struct ip_conntrack *ct)
@@ -440,6 +425,7 @@ int
 __ip_conntrack_confirm(struct sk_buff **pskb)
 {
 	unsigned int hash, repl_hash;
+	struct ip_conntrack_tuple_hash *h;
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
 
@@ -470,43 +456,43 @@ __ip_conntrack_confirm(struct sk_buff **pskb)
 	/* See if there's one in the list already, including reverse:
            NAT could have grabbed it without realizing, since we're
            not in the hash.  If there is, we lost race. */
-	if (!LIST_FIND(&ip_conntrack_hash[hash],
-		       conntrack_tuple_cmp,
-		       struct ip_conntrack_tuple_hash *,
-		       &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, NULL)
-	    && !LIST_FIND(&ip_conntrack_hash[repl_hash],
-			  conntrack_tuple_cmp,
-			  struct ip_conntrack_tuple_hash *,
-			  &ct->tuplehash[IP_CT_DIR_REPLY].tuple, NULL)) {
-		/* Remove from unconfirmed list */
-		list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
+	list_for_each_entry(h, &ip_conntrack_hash[hash], list)
+		if (ip_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple,
+				      &h->tuple))
+			goto out;
+	list_for_each_entry(h, &ip_conntrack_hash[repl_hash], list)
+		if (ip_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple,
+				      &h->tuple))
+			goto out;
 
-		__ip_conntrack_hash_insert(ct, hash, repl_hash);
-		/* Timer relative to confirmation time, not original
-		   setting time, otherwise we'd get timer wrap in
-		   weird delay cases. */
-		ct->timeout.expires += jiffies;
-		add_timer(&ct->timeout);
-		atomic_inc(&ct->ct_general.use);
-		set_bit(IPS_CONFIRMED_BIT, &ct->status);
-		CONNTRACK_STAT_INC(insert);
-		write_unlock_bh(&ip_conntrack_lock);
-		if (ct->helper)
-			ip_conntrack_event_cache(IPCT_HELPER, *pskb);
+	/* Remove from unconfirmed list */
+	list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
+
+	__ip_conntrack_hash_insert(ct, hash, repl_hash);
+	/* Timer relative to confirmation time, not original
+	   setting time, otherwise we'd get timer wrap in
+	   weird delay cases. */
+	ct->timeout.expires += jiffies;
+	add_timer(&ct->timeout);
+	atomic_inc(&ct->ct_general.use);
+	set_bit(IPS_CONFIRMED_BIT, &ct->status);
+	CONNTRACK_STAT_INC(insert);
+	write_unlock_bh(&ip_conntrack_lock);
+	if (ct->helper)
+		ip_conntrack_event_cache(IPCT_HELPER, *pskb);
 #ifdef CONFIG_IP_NF_NAT_NEEDED
-		if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) ||
-		    test_bit(IPS_DST_NAT_DONE_BIT, &ct->status))
-			ip_conntrack_event_cache(IPCT_NATINFO, *pskb);
+	if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) ||
+	    test_bit(IPS_DST_NAT_DONE_BIT, &ct->status))
+		ip_conntrack_event_cache(IPCT_NATINFO, *pskb);
 #endif
-		ip_conntrack_event_cache(master_ct(ct) ?
-					 IPCT_RELATED : IPCT_NEW, *pskb);
+	ip_conntrack_event_cache(master_ct(ct) ?
+				 IPCT_RELATED : IPCT_NEW, *pskb);
 
-		return NF_ACCEPT;
-	}
+	return NF_ACCEPT;
 
+out:
 	CONNTRACK_STAT_INC(insert_failed);
 	write_unlock_bh(&ip_conntrack_lock);
-
 	return NF_DROP;
 }
 
@@ -527,23 +513,21 @@ ip_conntrack_tuple_taken(const struct ip_conntrack_tuple *tuple,
 
 /* There's a small race here where we may free a just-assured
    connection.  Too bad: we're in trouble anyway. */
-static inline int unreplied(const struct ip_conntrack_tuple_hash *i)
-{
-	return !(test_bit(IPS_ASSURED_BIT, &tuplehash_to_ctrack(i)->status));
-}
-
 static int early_drop(struct list_head *chain)
 {
 	/* Traverse backwards: gives us oldest, which is roughly LRU */
 	struct ip_conntrack_tuple_hash *h;
-	struct ip_conntrack *ct = NULL;
+	struct ip_conntrack *ct = NULL, *tmp;
 	int dropped = 0;
 
 	read_lock_bh(&ip_conntrack_lock);
-	h = LIST_FIND_B(chain, unreplied, struct ip_conntrack_tuple_hash *);
-	if (h) {
-		ct = tuplehash_to_ctrack(h);
-		atomic_inc(&ct->ct_general.use);
+	list_for_each_entry_reverse(h, chain, list) {
+		tmp = tuplehash_to_ctrack(h);
+		if (!test_bit(IPS_ASSURED_BIT, &tmp->status)) {
+			ct = tmp;
+			atomic_inc(&ct->ct_general.use);
+			break;
+		}
 	}
 	read_unlock_bh(&ip_conntrack_lock);
 
@@ -559,18 +543,16 @@ static int early_drop(struct list_head *chain)
 	return dropped;
 }
 
-static inline int helper_cmp(const struct ip_conntrack_helper *i,
-			     const struct ip_conntrack_tuple *rtuple)
-{
-	return ip_ct_tuple_mask_cmp(rtuple, &i->tuple, &i->mask);
-}
-
 static struct ip_conntrack_helper *
 __ip_conntrack_helper_find( const struct ip_conntrack_tuple *tuple)
 {
-	return LIST_FIND(&helpers, helper_cmp,
-			 struct ip_conntrack_helper *,
-			 tuple);
+	struct ip_conntrack_helper *h;
+
+	list_for_each_entry(h, &helpers, list) {
+		if (ip_ct_tuple_mask_cmp(tuple, &h->tuple, &h->mask))
+			return h;
+	}
+	return NULL;
 }
 
 struct ip_conntrack_helper *
@@ -1062,7 +1044,7 @@ int ip_conntrack_helper_register(struct ip_conntrack_helper *me)
 {
 	BUG_ON(me->timeout == 0);
 	write_lock_bh(&ip_conntrack_lock);
-	list_prepend(&helpers, me);
+	list_add(&me->list, &helpers);
 	write_unlock_bh(&ip_conntrack_lock);
 
 	return 0;
@@ -1081,24 +1063,24 @@ __ip_conntrack_helper_find_byname(const char *name)
 	return NULL;
 }
 
-static inline int unhelp(struct ip_conntrack_tuple_hash *i,
-			 const struct ip_conntrack_helper *me)
+static inline void unhelp(struct ip_conntrack_tuple_hash *i,
+			  const struct ip_conntrack_helper *me)
 {
 	if (tuplehash_to_ctrack(i)->helper == me) {
  		ip_conntrack_event(IPCT_HELPER, tuplehash_to_ctrack(i));
 		tuplehash_to_ctrack(i)->helper = NULL;
 	}
-	return 0;
 }
 
 void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me)
 {
 	unsigned int i;
+	struct ip_conntrack_tuple_hash *h;
 	struct ip_conntrack_expect *exp, *tmp;
 
 	/* Need write lock here, to delete helper. */
 	write_lock_bh(&ip_conntrack_lock);
-	LIST_DELETE(&helpers, me);
+	list_del(&me->list);
 
 	/* Get rid of expectations */
 	list_for_each_entry_safe(exp, tmp, &ip_conntrack_expect_list, list) {
@@ -1108,10 +1090,12 @@ void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me)
 		}
 	}
 	/* Get rid of expecteds, set helpers to NULL. */
-	LIST_FIND_W(&unconfirmed, unhelp, struct ip_conntrack_tuple_hash*, me);
-	for (i = 0; i < ip_conntrack_htable_size; i++)
-		LIST_FIND_W(&ip_conntrack_hash[i], unhelp,
-			    struct ip_conntrack_tuple_hash *, me);
+	list_for_each_entry(h, &unconfirmed, list)
+		unhelp(h, me);
+	for (i = 0; i < ip_conntrack_htable_size; i++) {
+		list_for_each_entry(h, &ip_conntrack_hash[i], list)
+			unhelp(h, me);
+	}
 	write_unlock_bh(&ip_conntrack_lock);
 
 	/* Someone could be still looking at the helper in a bh. */
@@ -1237,46 +1221,43 @@ static void ip_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb)
 	nf_conntrack_get(nskb->nfct);
 }
 
-static inline int
-do_iter(const struct ip_conntrack_tuple_hash *i,
-	int (*iter)(struct ip_conntrack *i, void *data),
-	void *data)
-{
-	return iter(tuplehash_to_ctrack(i), data);
-}
-
 /* Bring out ya dead! */
-static struct ip_conntrack_tuple_hash *
+static struct ip_conntrack *
 get_next_corpse(int (*iter)(struct ip_conntrack *i, void *data),
 		void *data, unsigned int *bucket)
 {
-	struct ip_conntrack_tuple_hash *h = NULL;
+	struct ip_conntrack_tuple_hash *h;
+	struct ip_conntrack *ct;
 
 	write_lock_bh(&ip_conntrack_lock);
 	for (; *bucket < ip_conntrack_htable_size; (*bucket)++) {
-		h = LIST_FIND_W(&ip_conntrack_hash[*bucket], do_iter,
-				struct ip_conntrack_tuple_hash *, iter, data);
-		if (h)
-			break;
+		list_for_each_entry(h, &ip_conntrack_hash[*bucket], list) {
+			ct = tuplehash_to_ctrack(h);
+			if (iter(ct, data))
+				goto found;
+		}
+	}
+	list_for_each_entry(h, &unconfirmed, list) {
+		ct = tuplehash_to_ctrack(h);
+		if (iter(ct, data))
+			goto found;
 	}
-	if (!h)
-		h = LIST_FIND_W(&unconfirmed, do_iter,
-				struct ip_conntrack_tuple_hash *, iter, data);
-	if (h)
-		atomic_inc(&tuplehash_to_ctrack(h)->ct_general.use);
 	write_unlock_bh(&ip_conntrack_lock);
+	return NULL;
 
-	return h;
+found:
+	atomic_inc(&ct->ct_general.use);
+	write_unlock_bh(&ip_conntrack_lock);
+	return ct;
 }
 
 void
 ip_ct_iterate_cleanup(int (*iter)(struct ip_conntrack *i, void *), void *data)
 {
-	struct ip_conntrack_tuple_hash *h;
+	struct ip_conntrack *ct;
 	unsigned int bucket = 0;
 
-	while ((h = get_next_corpse(iter, data, &bucket)) != NULL) {
-		struct ip_conntrack *ct = tuplehash_to_ctrack(h);
+	while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) {
 		/* Time to push up daises... */
 		if (del_timer(&ct->timeout))
 			death_by_timeout((unsigned long)ct);
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_gre.c b/net/ipv4/netfilter/ip_conntrack_proto_gre.c
index 4ee016c427b4..92c6d8b178c9 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_gre.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_gre.c
@@ -37,7 +37,6 @@ static DEFINE_RWLOCK(ip_ct_gre_lock);
 #define ASSERT_READ_LOCK(x)
 #define ASSERT_WRITE_LOCK(x)
 
-#include <linux/netfilter_ipv4/listhelp.h>
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
 #include <linux/netfilter_ipv4/ip_conntrack_core.h>
@@ -82,10 +81,12 @@ static __be16 gre_keymap_lookup(struct ip_conntrack_tuple *t)
 	__be16 key = 0;
 
 	read_lock_bh(&ip_ct_gre_lock);
-	km = LIST_FIND(&gre_keymap_list, gre_key_cmpfn,
-			struct ip_ct_gre_keymap *, t);
-	if (km)
-		key = km->tuple.src.u.gre.key;
+	list_for_each_entry(km, &gre_keymap_list, list) {
+		if (gre_key_cmpfn(km, t)) {
+			key = km->tuple.src.u.gre.key;
+			break;
+		}
+	}
 	read_unlock_bh(&ip_ct_gre_lock);
 	
 	DEBUGP("lookup src key 0x%x up key for ", key);
@@ -99,7 +100,7 @@ int
 ip_ct_gre_keymap_add(struct ip_conntrack *ct,
 		     struct ip_conntrack_tuple *t, int reply)
 {
-	struct ip_ct_gre_keymap **exist_km, *km, *old;
+	struct ip_ct_gre_keymap **exist_km, *km;
 
 	if (!ct->helper || strcmp(ct->helper->name, "pptp")) {
 		DEBUGP("refusing to add GRE keymap to non-pptp session\n");
@@ -113,13 +114,10 @@ ip_ct_gre_keymap_add(struct ip_conntrack *ct,
 
 	if (*exist_km) {
 		/* check whether it's a retransmission */
-		old = LIST_FIND(&gre_keymap_list, gre_key_cmpfn,
-				struct ip_ct_gre_keymap *, t);
-		if (old == *exist_km) {
-			DEBUGP("retransmission\n");
-			return 0;
+		list_for_each_entry(km, &gre_keymap_list, list) {
+			if (gre_key_cmpfn(km, t) && km == *exist_km)
+				return 0;
 		}
-
 		DEBUGP("trying to override keymap_%s for ct %p\n", 
 			reply? "reply":"orig", ct);
 		return -EEXIST;
@@ -136,7 +134,7 @@ ip_ct_gre_keymap_add(struct ip_conntrack *ct,
 	DUMP_TUPLE_GRE(&km->tuple);
 
 	write_lock_bh(&ip_ct_gre_lock);
-	list_append(&gre_keymap_list, km);
+	list_add_tail(&km->list, &gre_keymap_list);
 	write_unlock_bh(&ip_ct_gre_lock);
 
 	return 0;
diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c
index 3f5d495b853b..02135756562e 100644
--- a/net/ipv4/netfilter/ip_conntrack_standalone.c
+++ b/net/ipv4/netfilter/ip_conntrack_standalone.c
@@ -35,7 +35,6 @@
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
 #include <linux/netfilter_ipv4/ip_conntrack_core.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
 #define DEBUGP printk
diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c
index 4c540d03d48e..71f3e09cbc84 100644
--- a/net/ipv4/netfilter/ip_nat_core.c
+++ b/net/ipv4/netfilter/ip_nat_core.c
@@ -22,9 +22,6 @@
 #include <linux/udp.h>
 #include <linux/jhash.h>
 
-#define ASSERT_READ_LOCK(x)
-#define ASSERT_WRITE_LOCK(x)
-
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_core.h>
 #include <linux/netfilter_ipv4/ip_conntrack_protocol.h>
@@ -33,7 +30,6 @@
 #include <linux/netfilter_ipv4/ip_nat_core.h>
 #include <linux/netfilter_ipv4/ip_nat_helper.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
 #define DEBUGP printk
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index 021c3daae3ed..7f6a75984f6c 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -27,16 +27,12 @@
 #include <net/tcp.h>
 #include <net/udp.h>
 
-#define ASSERT_READ_LOCK(x)
-#define ASSERT_WRITE_LOCK(x)
-
 #include <linux/netfilter_ipv4/ip_conntrack.h>
 #include <linux/netfilter_ipv4/ip_conntrack_helper.h>
 #include <linux/netfilter_ipv4/ip_nat.h>
 #include <linux/netfilter_ipv4/ip_nat_protocol.h>
 #include <linux/netfilter_ipv4/ip_nat_core.h>
 #include <linux/netfilter_ipv4/ip_nat_helper.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
 #define DEBUGP printk
diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c
index e59f5a8ecb6b..7b703839aa58 100644
--- a/net/ipv4/netfilter/ip_nat_rule.c
+++ b/net/ipv4/netfilter/ip_nat_rule.c
@@ -19,14 +19,10 @@
 #include <net/route.h>
 #include <linux/bitops.h>
 
-#define ASSERT_READ_LOCK(x)
-#define ASSERT_WRITE_LOCK(x)
-
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ip_nat.h>
 #include <linux/netfilter_ipv4/ip_nat_core.h>
 #include <linux/netfilter_ipv4/ip_nat_rule.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
 #define DEBUGP printk
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index f3b778355432..9c577db62047 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -30,9 +30,6 @@
 #include <net/checksum.h>
 #include <linux/spinlock.h>
 
-#define ASSERT_READ_LOCK(x)
-#define ASSERT_WRITE_LOCK(x)
-
 #include <linux/netfilter_ipv4/ip_nat.h>
 #include <linux/netfilter_ipv4/ip_nat_rule.h>
 #include <linux/netfilter_ipv4/ip_nat_protocol.h>
@@ -40,7 +37,6 @@
 #include <linux/netfilter_ipv4/ip_nat_helper.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ip_conntrack_core.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
 #define DEBUGP printk
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index d1c315364ee7..73d477ce216b 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -70,9 +70,6 @@ do {								\
 #define IP_NF_ASSERT(x)
 #endif
 
-
-#include <linux/netfilter_ipv4/listhelp.h>
-
 #if 0
 /* All the better to debug you with... */
 #define static
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 3b64dbee6620..927137b8b3b5 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -57,7 +57,6 @@
 #include <net/netfilter/nf_conntrack_protocol.h>
 #include <net/netfilter/nf_conntrack_helper.h>
 #include <net/netfilter/nf_conntrack_core.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #define NF_CONNTRACK_VERSION	"0.5.0"
 
@@ -539,15 +538,10 @@ void nf_ct_remove_expectations(struct nf_conn *ct)
 static void
 clean_from_lists(struct nf_conn *ct)
 {
-	unsigned int ho, hr;
-	
 	DEBUGP("clean_from_lists(%p)\n", ct);
 	ASSERT_WRITE_LOCK(&nf_conntrack_lock);
-
-	ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
-	hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
-	LIST_DELETE(&nf_conntrack_hash[ho], &ct->tuplehash[IP_CT_DIR_ORIGINAL]);
-	LIST_DELETE(&nf_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]);
+	list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
+	list_del(&ct->tuplehash[IP_CT_DIR_REPLY].list);
 
 	/* Destroy all pending expectations */
 	nf_ct_remove_expectations(ct);
@@ -617,16 +611,6 @@ static void death_by_timeout(unsigned long ul_conntrack)
 	nf_ct_put(ct);
 }
 
-static inline int
-conntrack_tuple_cmp(const struct nf_conntrack_tuple_hash *i,
-		    const struct nf_conntrack_tuple *tuple,
-		    const struct nf_conn *ignored_conntrack)
-{
-	ASSERT_READ_LOCK(&nf_conntrack_lock);
-	return nf_ct_tuplehash_to_ctrack(i) != ignored_conntrack
-		&& nf_ct_tuple_equal(tuple, &i->tuple);
-}
-
 struct nf_conntrack_tuple_hash *
 __nf_conntrack_find(const struct nf_conntrack_tuple *tuple,
 		    const struct nf_conn *ignored_conntrack)
@@ -636,7 +620,8 @@ __nf_conntrack_find(const struct nf_conntrack_tuple *tuple,
 
 	ASSERT_READ_LOCK(&nf_conntrack_lock);
 	list_for_each_entry(h, &nf_conntrack_hash[hash], list) {
-		if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) {
+		if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack &&
+		    nf_ct_tuple_equal(tuple, &h->tuple)) {
 			NF_CT_STAT_INC(found);
 			return h;
 		}
@@ -667,10 +652,10 @@ static void __nf_conntrack_hash_insert(struct nf_conn *ct,
 				       unsigned int repl_hash) 
 {
 	ct->id = ++nf_conntrack_next_id;
-	list_prepend(&nf_conntrack_hash[hash],
-		     &ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
-	list_prepend(&nf_conntrack_hash[repl_hash],
-		     &ct->tuplehash[IP_CT_DIR_REPLY].list);
+	list_add(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list,
+		 &nf_conntrack_hash[hash]);
+	list_add(&ct->tuplehash[IP_CT_DIR_REPLY].list,
+		 &nf_conntrack_hash[repl_hash]);
 }
 
 void nf_conntrack_hash_insert(struct nf_conn *ct)
@@ -690,7 +675,9 @@ int
 __nf_conntrack_confirm(struct sk_buff **pskb)
 {
 	unsigned int hash, repl_hash;
+	struct nf_conntrack_tuple_hash *h;
 	struct nf_conn *ct;
+	struct nf_conn_help *help;
 	enum ip_conntrack_info ctinfo;
 
 	ct = nf_ct_get(*pskb, &ctinfo);
@@ -720,41 +707,41 @@ __nf_conntrack_confirm(struct sk_buff **pskb)
 	/* See if there's one in the list already, including reverse:
 	   NAT could have grabbed it without realizing, since we're
 	   not in the hash.  If there is, we lost race. */
-	if (!LIST_FIND(&nf_conntrack_hash[hash],
-		       conntrack_tuple_cmp,
-		       struct nf_conntrack_tuple_hash *,
-		       &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, NULL)
-	    && !LIST_FIND(&nf_conntrack_hash[repl_hash],
-			  conntrack_tuple_cmp,
-			  struct nf_conntrack_tuple_hash *,
-			  &ct->tuplehash[IP_CT_DIR_REPLY].tuple, NULL)) {
-		struct nf_conn_help *help;
-		/* Remove from unconfirmed list */
-		list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
+	list_for_each_entry(h, &nf_conntrack_hash[hash], list)
+		if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple,
+				      &h->tuple))
+			goto out;
+	list_for_each_entry(h, &nf_conntrack_hash[repl_hash], list)
+		if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple,
+				      &h->tuple))
+			goto out;
 
-		__nf_conntrack_hash_insert(ct, hash, repl_hash);
-		/* Timer relative to confirmation time, not original
-		   setting time, otherwise we'd get timer wrap in
-		   weird delay cases. */
-		ct->timeout.expires += jiffies;
-		add_timer(&ct->timeout);
-		atomic_inc(&ct->ct_general.use);
-		set_bit(IPS_CONFIRMED_BIT, &ct->status);
-		NF_CT_STAT_INC(insert);
-		write_unlock_bh(&nf_conntrack_lock);
-		help = nfct_help(ct);
-		if (help && help->helper)
-			nf_conntrack_event_cache(IPCT_HELPER, *pskb);
+	/* Remove from unconfirmed list */
+	list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
+
+	__nf_conntrack_hash_insert(ct, hash, repl_hash);
+	/* Timer relative to confirmation time, not original
+	   setting time, otherwise we'd get timer wrap in
+	   weird delay cases. */
+	ct->timeout.expires += jiffies;
+	add_timer(&ct->timeout);
+	atomic_inc(&ct->ct_general.use);
+	set_bit(IPS_CONFIRMED_BIT, &ct->status);
+	NF_CT_STAT_INC(insert);
+	write_unlock_bh(&nf_conntrack_lock);
+	help = nfct_help(ct);
+	if (help && help->helper)
+		nf_conntrack_event_cache(IPCT_HELPER, *pskb);
 #ifdef CONFIG_NF_NAT_NEEDED
-		if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) ||
-		    test_bit(IPS_DST_NAT_DONE_BIT, &ct->status))
-			nf_conntrack_event_cache(IPCT_NATINFO, *pskb);
+	if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) ||
+	    test_bit(IPS_DST_NAT_DONE_BIT, &ct->status))
+		nf_conntrack_event_cache(IPCT_NATINFO, *pskb);
 #endif
-		nf_conntrack_event_cache(master_ct(ct) ?
-					 IPCT_RELATED : IPCT_NEW, *pskb);
-		return NF_ACCEPT;
-	}
+	nf_conntrack_event_cache(master_ct(ct) ?
+				 IPCT_RELATED : IPCT_NEW, *pskb);
+	return NF_ACCEPT;
 
+out:
 	NF_CT_STAT_INC(insert_failed);
 	write_unlock_bh(&nf_conntrack_lock);
 	return NF_DROP;
@@ -777,24 +764,21 @@ nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple,
 
 /* There's a small race here where we may free a just-assured
    connection.  Too bad: we're in trouble anyway. */
-static inline int unreplied(const struct nf_conntrack_tuple_hash *i)
-{
-	return !(test_bit(IPS_ASSURED_BIT,
-			  &nf_ct_tuplehash_to_ctrack(i)->status));
-}
-
 static int early_drop(struct list_head *chain)
 {
 	/* Traverse backwards: gives us oldest, which is roughly LRU */
 	struct nf_conntrack_tuple_hash *h;
-	struct nf_conn *ct = NULL;
+	struct nf_conn *ct = NULL, *tmp;
 	int dropped = 0;
 
 	read_lock_bh(&nf_conntrack_lock);
-	h = LIST_FIND_B(chain, unreplied, struct nf_conntrack_tuple_hash *);
-	if (h) {
-		ct = nf_ct_tuplehash_to_ctrack(h);
-		atomic_inc(&ct->ct_general.use);
+	list_for_each_entry_reverse(h, chain, list) {
+		tmp = nf_ct_tuplehash_to_ctrack(h);
+		if (!test_bit(IPS_ASSURED_BIT, &tmp->status)) {
+			ct = tmp;
+			atomic_inc(&ct->ct_general.use);
+			break;
+		}
 	}
 	read_unlock_bh(&nf_conntrack_lock);
 
@@ -810,18 +794,16 @@ static int early_drop(struct list_head *chain)
 	return dropped;
 }
 
-static inline int helper_cmp(const struct nf_conntrack_helper *i,
-			     const struct nf_conntrack_tuple *rtuple)
-{
-	return nf_ct_tuple_mask_cmp(rtuple, &i->tuple, &i->mask);
-}
-
 static struct nf_conntrack_helper *
 __nf_ct_helper_find(const struct nf_conntrack_tuple *tuple)
 {
-	return LIST_FIND(&helpers, helper_cmp,
-			 struct nf_conntrack_helper *,
-			 tuple);
+	struct nf_conntrack_helper *h;
+
+	list_for_each_entry(h, &helpers, list) {
+		if (nf_ct_tuple_mask_cmp(tuple, &h->tuple, &h->mask))
+			return h;
+	}
+	return NULL;
 }
 
 struct nf_conntrack_helper *
@@ -1323,7 +1305,7 @@ int nf_conntrack_helper_register(struct nf_conntrack_helper *me)
 		return ret;
 	}
 	write_lock_bh(&nf_conntrack_lock);
-	list_prepend(&helpers, me);
+	list_add(&me->list, &helpers);
 	write_unlock_bh(&nf_conntrack_lock);
 
 	return 0;
@@ -1342,8 +1324,8 @@ __nf_conntrack_helper_find_byname(const char *name)
 	return NULL;
 }
 
-static inline int unhelp(struct nf_conntrack_tuple_hash *i,
-			 const struct nf_conntrack_helper *me)
+static inline void unhelp(struct nf_conntrack_tuple_hash *i,
+			  const struct nf_conntrack_helper *me)
 {
 	struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(i);
 	struct nf_conn_help *help = nfct_help(ct);
@@ -1352,17 +1334,17 @@ static inline int unhelp(struct nf_conntrack_tuple_hash *i,
 		nf_conntrack_event(IPCT_HELPER, ct);
 		help->helper = NULL;
 	}
-	return 0;
 }
 
 void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me)
 {
 	unsigned int i;
+	struct nf_conntrack_tuple_hash *h;
 	struct nf_conntrack_expect *exp, *tmp;
 
 	/* Need write lock here, to delete helper. */
 	write_lock_bh(&nf_conntrack_lock);
-	LIST_DELETE(&helpers, me);
+	list_del(&me->list);
 
 	/* Get rid of expectations */
 	list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) {
@@ -1374,10 +1356,12 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me)
 	}
 
 	/* Get rid of expecteds, set helpers to NULL. */
-	LIST_FIND_W(&unconfirmed, unhelp, struct nf_conntrack_tuple_hash*, me);
-	for (i = 0; i < nf_conntrack_htable_size; i++)
-		LIST_FIND_W(&nf_conntrack_hash[i], unhelp,
-			    struct nf_conntrack_tuple_hash *, me);
+	list_for_each_entry(h, &unconfirmed, list)
+		unhelp(h, me);
+	for (i = 0; i < nf_conntrack_htable_size; i++) {
+		list_for_each_entry(h, &nf_conntrack_hash[i], list)
+			unhelp(h, me);
+	}
 	write_unlock_bh(&nf_conntrack_lock);
 
 	/* Someone could be still looking at the helper in a bh. */
@@ -1510,37 +1494,40 @@ do_iter(const struct nf_conntrack_tuple_hash *i,
 }
 
 /* Bring out ya dead! */
-static struct nf_conntrack_tuple_hash *
+static struct nf_conn *
 get_next_corpse(int (*iter)(struct nf_conn *i, void *data),
 		void *data, unsigned int *bucket)
 {
-	struct nf_conntrack_tuple_hash *h = NULL;
+	struct nf_conntrack_tuple_hash *h;
+	struct nf_conn *ct;
 
 	write_lock_bh(&nf_conntrack_lock);
 	for (; *bucket < nf_conntrack_htable_size; (*bucket)++) {
-		h = LIST_FIND_W(&nf_conntrack_hash[*bucket], do_iter,
-				struct nf_conntrack_tuple_hash *, iter, data);
-		if (h)
-			break;
+		list_for_each_entry(h, &nf_conntrack_hash[*bucket], list) {
+			ct = nf_ct_tuplehash_to_ctrack(h);
+			if (iter(ct, data))
+				goto found;
+		}
  	}
-	if (!h)
-		h = LIST_FIND_W(&unconfirmed, do_iter,
-				struct nf_conntrack_tuple_hash *, iter, data);
-	if (h)
-		atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use);
+	list_for_each_entry(h, &unconfirmed, list) {
+		ct = nf_ct_tuplehash_to_ctrack(h);
+		if (iter(ct, data))
+			goto found;
+	}
+	return NULL;
+found:
+	atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use);
 	write_unlock_bh(&nf_conntrack_lock);
-
-	return h;
+	return ct;
 }
 
 void
 nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data)
 {
-	struct nf_conntrack_tuple_hash *h;
+	struct nf_conn *ct;
 	unsigned int bucket = 0;
 
-	while ((h = get_next_corpse(iter, data, &bucket)) != NULL) {
-		struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
+	while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) {
 		/* Time to push up daises... */
 		if (del_timer(&ct->timeout))
 			death_by_timeout((unsigned long)ct);
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 9a1de0ca475b..5954f6773810 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -37,7 +37,6 @@
 #include <net/netfilter/nf_conntrack_protocol.h>
 #include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/nf_conntrack_helper.h>
-#include <linux/netfilter_ipv4/listhelp.h>
 
 #if 0
 #define DEBUGP printk
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index 8037ba63d587..be7baf4f6846 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -81,7 +81,7 @@ xt_unregister_target(struct xt_target *target)
 	int af = target->family;
 
 	mutex_lock(&xt[af].mutex);
-	LIST_DELETE(&xt[af].target, target);
+	list_del(&target->list);
 	mutex_unlock(&xt[af].mutex);
 }
 EXPORT_SYMBOL(xt_unregister_target);
@@ -138,7 +138,7 @@ xt_unregister_match(struct xt_match *match)
 	int af =  match->family;
 
 	mutex_lock(&xt[af].mutex);
-	LIST_DELETE(&xt[af].match, match);
+	list_del(&match->list);
 	mutex_unlock(&xt[af].mutex);
 }
 EXPORT_SYMBOL(xt_unregister_match);
@@ -575,15 +575,18 @@ int xt_register_table(struct xt_table *table,
 {
 	int ret;
 	struct xt_table_info *private;
+	struct xt_table *t;
 
 	ret = mutex_lock_interruptible(&xt[table->af].mutex);
 	if (ret != 0)
 		return ret;
 
 	/* Don't autoload: we'd eat our tail... */
-	if (list_named_find(&xt[table->af].tables, table->name)) {
-		ret = -EEXIST;
-		goto unlock;
+	list_for_each_entry(t, &xt[table->af].tables, list) {
+		if (strcmp(t->name, table->name) == 0) {
+			ret = -EEXIST;
+			goto unlock;
+		}
 	}
 
 	/* Simplifies replace_table code. */
@@ -598,7 +601,7 @@ int xt_register_table(struct xt_table *table,
 	/* save number of initial entries */
 	private->initial_entries = private->number;
 
-	list_prepend(&xt[table->af].tables, table);
+	list_add(&table->list, &xt[table->af].tables);
 
 	ret = 0;
  unlock:
@@ -613,7 +616,7 @@ void *xt_unregister_table(struct xt_table *table)
 
 	mutex_lock(&xt[table->af].mutex);
 	private = table->private;
-	LIST_DELETE(&xt[table->af].tables, table);
+	list_del(&table->list);
 	mutex_unlock(&xt[table->af].mutex);
 
 	return private;
-- 
cgit v1.2.3


From 9123de2c043996050bacf77031cad845f5976f5d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 11:59:42 -0700
Subject: [NETFILTER]: ip6table_mangle: reroute when nfmark changes in
 NF_IP6_LOCAL_OUT

Now that IPv6 supports policy routing we need to reroute in NF_IP6_LOCAL_OUT
when the mark value changes.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6.h       | 1 +
 include/net/ip6_route.h              | 2 --
 net/ipv6/netfilter/ip6table_mangle.c | 8 ++------
 3 files changed, 3 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index 52a7b9e76428..d97e268cdfe5 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -73,6 +73,7 @@ enum nf_ip6_hook_priorities {
 };
 
 #ifdef CONFIG_NETFILTER
+extern int ip6_route_me_harder(struct sk_buff *skb);
 extern unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
 				    unsigned int dataoff, u_int8_t protocol);
 
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 297909570041..6ca6b71dfe0f 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -57,8 +57,6 @@ extern void			ip6_route_input(struct sk_buff *skb);
 extern struct dst_entry *	ip6_route_output(struct sock *sk,
 						 struct flowi *fl);
 
-extern int			ip6_route_me_harder(struct sk_buff *skb);
-
 extern void			ip6_route_init(void);
 extern void			ip6_route_cleanup(void);
 
diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c
index 32db04fd8310..386ea260e767 100644
--- a/net/ipv6/netfilter/ip6table_mangle.c
+++ b/net/ipv6/netfilter/ip6table_mangle.c
@@ -180,12 +180,8 @@ ip6t_local_hook(unsigned int hook,
 		&& (memcmp(&(*pskb)->nh.ipv6h->saddr, &saddr, sizeof(saddr))
 		    || memcmp(&(*pskb)->nh.ipv6h->daddr, &daddr, sizeof(daddr))
 		    || (*pskb)->nfmark != nfmark
-		    || (*pskb)->nh.ipv6h->hop_limit != hop_limit)) {
-
-		/* something which could affect routing has changed */
-
-		DEBUGP("ip6table_mangle: we'd need to re-route a packet\n");
-	}
+		    || (*pskb)->nh.ipv6h->hop_limit != hop_limit))
+		return ip6_route_me_harder(*pskb) == 0 ? ret : NF_DROP;
 
 	return ret;
 }
-- 
cgit v1.2.3


From c1fe3ca5106d9568791433fa6c7f27e71ac69e1b Mon Sep 17 00:00:00 2001
From: George Hansper <georgeh@anstat.com.au>
Date: Wed, 20 Sep 2006 12:03:23 -0700
Subject: [NETFILTER]: TCP conntrack: improve dead connection detection

Don't count window updates as retransmissions.

Signed-off-by: George Hansper <georgeh@anstat.com.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/nf_conntrack_tcp.h  | 1 +
 net/ipv4/netfilter/ip_conntrack_proto_tcp.c | 4 +++-
 net/netfilter/nf_conntrack_proto_tcp.c      | 4 +++-
 3 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/nf_conntrack_tcp.h b/include/linux/netfilter/nf_conntrack_tcp.h
index b2feeffde384..6b01ba297727 100644
--- a/include/linux/netfilter/nf_conntrack_tcp.h
+++ b/include/linux/netfilter/nf_conntrack_tcp.h
@@ -49,6 +49,7 @@ struct ip_ct_tcp
 	u_int32_t	last_seq;	/* Last sequence number seen in dir */
 	u_int32_t	last_ack;	/* Last sequence number seen in opposite dir */
 	u_int32_t	last_end;	/* Last seq + len */
+	u_int16_t	last_win;	/* Last window advertisement seen in dir */
 };
 
 #endif /* __KERNEL__ */
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
index 75a7237eb8c1..03ae9a04cb37 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
@@ -731,13 +731,15 @@ static int tcp_in_window(struct ip_ct_tcp *state,
 			if (state->last_dir == dir
 			    && state->last_seq == seq
 			    && state->last_ack == ack
-			    && state->last_end == end)
+			    && state->last_end == end
+			    && state->last_win == win)
 				state->retrans++;
 			else {
 				state->last_dir = dir;
 				state->last_seq = seq;
 				state->last_ack = ack;
 				state->last_end = end;
+				state->last_win = win;
 				state->retrans = 0;
 			}
 		}
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 9fc0ee61f92a..238bbb5b72ef 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -688,13 +688,15 @@ static int tcp_in_window(struct ip_ct_tcp *state,
 			if (state->last_dir == dir
 			    && state->last_seq == seq
 			    && state->last_ack == ack
-			    && state->last_end == end)
+			    && state->last_end == end
+			    && state->last_win == win)
 				state->retrans++;
 			else {
 				state->last_dir = dir;
 				state->last_seq = seq;
 				state->last_ack = ack;
 				state->last_end = end;
+				state->last_win = win;
 				state->retrans = 0;
 			}
 		}
-- 
cgit v1.2.3


From 9fa492cdc160cd27ce1046cb36f47d3b2b1efa21 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 12:05:37 -0700
Subject: [NETFILTER]: x_tables: simplify compat API

Split the xt_compat_match/xt_compat_target into smaller type-safe functions
performing just one operation. Handle all alignment and size-related
conversions centrally in these function instead of requiring each module to
implement a full-blown conversion function. Replace ->compat callback by
->compat_from_user and ->compat_to_user callbacks, responsible for
converting just a single private structure.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h |  29 +++---
 net/ipv4/netfilter/ip_tables.c     | 115 +++++-----------------
 net/netfilter/x_tables.c           | 192 +++++++++++++++++++++----------------
 3 files changed, 151 insertions(+), 185 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index c832295dbf61..739a98eebe2c 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -138,12 +138,6 @@ struct xt_counters_info
 
 #include <linux/netdevice.h>
 
-#ifdef CONFIG_COMPAT
-#define COMPAT_TO_USER		1
-#define COMPAT_FROM_USER	-1
-#define COMPAT_CALC_SIZE	0
-#endif
-
 struct xt_match
 {
 	struct list_head list;
@@ -176,7 +170,8 @@ struct xt_match
 	void (*destroy)(const struct xt_match *match, void *matchinfo);
 
 	/* Called when userspace align differs from kernel space one */
-	int (*compat)(void *match, void **dstptr, int *size, int convert);
+	void (*compat_from_user)(void *dst, void *src);
+	int (*compat_to_user)(void __user *dst, void *src);
 
 	/* Set this to THIS_MODULE if you are a module, otherwise NULL */
 	struct module *me;
@@ -186,6 +181,7 @@ struct xt_match
 
 	char *table;
 	unsigned int matchsize;
+	unsigned int compatsize;
 	unsigned int hooks;
 	unsigned short proto;
 
@@ -224,13 +220,15 @@ struct xt_target
 	void (*destroy)(const struct xt_target *target, void *targinfo);
 
 	/* Called when userspace align differs from kernel space one */
-	int (*compat)(void *target, void **dstptr, int *size, int convert);
+	void (*compat_from_user)(void *dst, void *src);
+	int (*compat_to_user)(void __user *dst, void *src);
 
 	/* Set this to THIS_MODULE if you are a module, otherwise NULL */
 	struct module *me;
 
 	char *table;
 	unsigned int targetsize;
+	unsigned int compatsize;
 	unsigned int hooks;
 	unsigned short proto;
 
@@ -387,9 +385,18 @@ struct compat_xt_counters_info
 
 extern void xt_compat_lock(int af);
 extern void xt_compat_unlock(int af);
-extern int xt_compat_match(void *match, void **dstptr, int *size, int convert);
-extern int xt_compat_target(void *target, void **dstptr, int *size,
-		int convert);
+
+extern int xt_compat_match_offset(struct xt_match *match);
+extern void xt_compat_match_from_user(struct xt_entry_match *m,
+				      void **dstptr, int *size);
+extern int xt_compat_match_to_user(struct xt_entry_match *m,
+				   void * __user *dstptr, int *size);
+
+extern int xt_compat_target_offset(struct xt_target *target);
+extern void xt_compat_target_from_user(struct xt_entry_target *t,
+				       void **dstptr, int *size);
+extern int xt_compat_target_to_user(struct xt_entry_target *t,
+				    void * __user *dstptr, int *size);
 
 #endif /* CONFIG_COMPAT */
 #endif /* __KERNEL__ */
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 673581db986e..800067d69a9a 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -942,73 +942,28 @@ static short compat_calc_jump(u_int16_t offset)
 	return delta;
 }
 
-struct compat_ipt_standard_target
+static void compat_standard_from_user(void *dst, void *src)
 {
-	struct compat_xt_entry_target target;
-	compat_int_t verdict;
-};
-
-struct compat_ipt_standard
-{
-	struct compat_ipt_entry entry;
-	struct compat_ipt_standard_target target;
-};
+	int v = *(compat_int_t *)src;
 
-#define IPT_ST_LEN		XT_ALIGN(sizeof(struct ipt_standard_target))
-#define IPT_ST_COMPAT_LEN	COMPAT_XT_ALIGN(sizeof(struct compat_ipt_standard_target))
-#define IPT_ST_OFFSET		(IPT_ST_LEN - IPT_ST_COMPAT_LEN)
+	if (v > 0)
+		v += compat_calc_jump(v);
+	memcpy(dst, &v, sizeof(v));
+}
 
-static int compat_ipt_standard_fn(void *target,
-		void **dstptr, int *size, int convert)
+static int compat_standard_to_user(void __user *dst, void *src)
 {
-	struct compat_ipt_standard_target compat_st, *pcompat_st;
-	struct ipt_standard_target st, *pst;
-	int ret;
+	compat_int_t cv = *(int *)src;
 
-	ret = 0;
-	switch (convert) {
-		case COMPAT_TO_USER:
-			pst = target;
-			memcpy(&compat_st.target, &pst->target,
-				sizeof(compat_st.target));
-			compat_st.verdict = pst->verdict;
-			if (compat_st.verdict > 0)
-				compat_st.verdict -=
-					compat_calc_jump(compat_st.verdict);
-			compat_st.target.u.user.target_size = IPT_ST_COMPAT_LEN;
-			if (copy_to_user(*dstptr, &compat_st, IPT_ST_COMPAT_LEN))
-				ret = -EFAULT;
-			*size -= IPT_ST_OFFSET;
-			*dstptr += IPT_ST_COMPAT_LEN;
-			break;
-		case COMPAT_FROM_USER:
-			pcompat_st = target;
-			memcpy(&st.target, &pcompat_st->target, IPT_ST_COMPAT_LEN);
-			st.verdict = pcompat_st->verdict;
-			if (st.verdict > 0)
-				st.verdict += compat_calc_jump(st.verdict);
-			st.target.u.user.target_size = IPT_ST_LEN;
-			memcpy(*dstptr, &st, IPT_ST_LEN);
-			*size += IPT_ST_OFFSET;
-			*dstptr += IPT_ST_LEN;
-			break;
-		case COMPAT_CALC_SIZE:
-			*size += IPT_ST_OFFSET;
-			break;
-		default:
-			ret = -ENOPROTOOPT;
-			break;
-	}
-	return ret;
+	if (cv > 0)
+		cv -= compat_calc_jump(cv);
+	return copy_to_user(dst, &cv, sizeof(cv)) ? -EFAULT : 0;
 }
 
 static inline int
 compat_calc_match(struct ipt_entry_match *m, int * size)
 {
-	if (m->u.kernel.match->compat)
-		m->u.kernel.match->compat(m, NULL, size, COMPAT_CALC_SIZE);
-	else
-		xt_compat_match(m, NULL, size, COMPAT_CALC_SIZE);
+	*size += xt_compat_match_offset(m->u.kernel.match);
 	return 0;
 }
 
@@ -1023,10 +978,7 @@ static int compat_calc_entry(struct ipt_entry *e, struct xt_table_info *info,
 	entry_offset = (void *)e - base;
 	IPT_MATCH_ITERATE(e, compat_calc_match, &off);
 	t = ipt_get_target(e);
-	if (t->u.kernel.target->compat)
-		t->u.kernel.target->compat(t, NULL, &off, COMPAT_CALC_SIZE);
-	else
-		xt_compat_target(t, NULL, &off, COMPAT_CALC_SIZE);
+	off += xt_compat_target_offset(t->u.kernel.target);
 	newinfo->size -= off;
 	ret = compat_add_offset(entry_offset, off);
 	if (ret)
@@ -1412,17 +1364,13 @@ struct compat_ipt_replace {
 };
 
 static inline int compat_copy_match_to_user(struct ipt_entry_match *m,
-		void __user **dstptr, compat_uint_t *size)
+		void * __user *dstptr, compat_uint_t *size)
 {
-	if (m->u.kernel.match->compat)
-		return m->u.kernel.match->compat(m, dstptr, size,
-				COMPAT_TO_USER);
-	else
-		return xt_compat_match(m, dstptr, size, COMPAT_TO_USER);
+	return xt_compat_match_to_user(m, dstptr, size);
 }
 
 static int compat_copy_entry_to_user(struct ipt_entry *e,
-		void __user **dstptr, compat_uint_t *size)
+		void * __user *dstptr, compat_uint_t *size)
 {
 	struct ipt_entry_target __user *t;
 	struct compat_ipt_entry __user *ce;
@@ -1442,11 +1390,7 @@ static int compat_copy_entry_to_user(struct ipt_entry *e,
 	if (ret)
 		goto out;
 	t = ipt_get_target(e);
-	if (t->u.kernel.target->compat)
-		ret = t->u.kernel.target->compat(t, dstptr, size,
-				COMPAT_TO_USER);
-	else
-		ret = xt_compat_target(t, dstptr, size, COMPAT_TO_USER);
+	ret = xt_compat_target_to_user(t, dstptr, size);
 	if (ret)
 		goto out;
 	ret = -EFAULT;
@@ -1478,11 +1422,7 @@ compat_check_calc_match(struct ipt_entry_match *m,
 		return match ? PTR_ERR(match) : -ENOENT;
 	}
 	m->u.kernel.match = match;
-
-	if (m->u.kernel.match->compat)
-		m->u.kernel.match->compat(m, NULL, size, COMPAT_CALC_SIZE);
-	else
-		xt_compat_match(m, NULL, size, COMPAT_CALC_SIZE);
+	*size += xt_compat_match_offset(match);
 
 	(*i)++;
 	return 0;
@@ -1543,10 +1483,7 @@ check_compat_entry_size_and_hooks(struct ipt_entry *e,
 	}
 	t->u.kernel.target = target;
 
-	if (t->u.kernel.target->compat)
-		t->u.kernel.target->compat(t, NULL, &off, COMPAT_CALC_SIZE);
-	else
-		xt_compat_target(t, NULL, &off, COMPAT_CALC_SIZE);
+	off += xt_compat_target_offset(target);
 	*size += off;
 	ret = compat_add_offset(entry_offset, off);
 	if (ret)
@@ -1584,10 +1521,7 @@ static inline int compat_copy_match_from_user(struct ipt_entry_match *m,
 
 	dm = (struct ipt_entry_match *)*dstptr;
 	match = m->u.kernel.match;
-	if (match->compat)
-		match->compat(m, dstptr, size, COMPAT_FROM_USER);
-	else
-		xt_compat_match(m, dstptr, size, COMPAT_FROM_USER);
+	xt_compat_match_from_user(m, dstptr, size);
 
 	ret = xt_check_match(match, AF_INET, dm->u.match_size - sizeof(*dm),
 			     name, hookmask, ip->proto,
@@ -1635,10 +1569,7 @@ static int compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr,
 	de->target_offset = e->target_offset - (origsize - *size);
 	t = ipt_get_target(e);
 	target = t->u.kernel.target;
-	if (target->compat)
-		target->compat(t, dstptr, size, COMPAT_FROM_USER);
-	else
-		xt_compat_target(t, dstptr, size, COMPAT_FROM_USER);
+	xt_compat_target_from_user(t, dstptr, size);
 
 	de->next_offset = e->next_offset - (origsize - *size);
 	for (h = 0; h < NF_IP_NUMHOOKS; h++) {
@@ -2205,7 +2136,9 @@ static struct ipt_target ipt_standard_target = {
 	.targetsize	= sizeof(int),
 	.family		= AF_INET,
 #ifdef CONFIG_COMPAT
-	.compat		= &compat_ipt_standard_fn,
+	.compatsize	= sizeof(compat_int_t),
+	.compat_from_user = compat_standard_from_user,
+	.compat_to_user	= compat_standard_to_user,
 #endif
 };
 
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index be7baf4f6846..58522fc65d33 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -333,52 +333,65 @@ int xt_check_match(const struct xt_match *match, unsigned short family,
 EXPORT_SYMBOL_GPL(xt_check_match);
 
 #ifdef CONFIG_COMPAT
-int xt_compat_match(void *match, void **dstptr, int *size, int convert)
+int xt_compat_match_offset(struct xt_match *match)
 {
-	struct xt_match *m;
-	struct compat_xt_entry_match *pcompat_m;
-	struct xt_entry_match *pm;
-	u_int16_t msize;
-	int off, ret;
+	u_int16_t csize = match->compatsize ? : match->matchsize;
+	return XT_ALIGN(match->matchsize) - COMPAT_XT_ALIGN(csize);
+}
+EXPORT_SYMBOL_GPL(xt_compat_match_offset);
 
-	ret = 0;
-	m = ((struct xt_entry_match *)match)->u.kernel.match;
-	off = XT_ALIGN(m->matchsize) - COMPAT_XT_ALIGN(m->matchsize);
-	switch (convert) {
-		case COMPAT_TO_USER:
-			pm = (struct xt_entry_match *)match;
-			msize = pm->u.user.match_size;
-			if (copy_to_user(*dstptr, pm, msize)) {
-				ret = -EFAULT;
-				break;
-			}
-			msize -= off;
-			if (put_user(msize, (u_int16_t *)*dstptr))
-				ret = -EFAULT;
-			*size -= off;
-			*dstptr += msize;
-			break;
-		case COMPAT_FROM_USER:
-			pcompat_m = (struct compat_xt_entry_match *)match;
-			pm = (struct xt_entry_match *)*dstptr;
-			msize = pcompat_m->u.user.match_size;
-			memcpy(pm, pcompat_m, msize);
-			msize += off;
-			pm->u.user.match_size = msize;
-			*size += off;
-			*dstptr += msize;
-			break;
-		case COMPAT_CALC_SIZE:
-			*size += off;
-			break;
-		default:
-			ret = -ENOPROTOOPT;
-			break;
+void xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr,
+			       int *size)
+{
+	struct xt_match *match = m->u.kernel.match;
+	struct compat_xt_entry_match *cm = (struct compat_xt_entry_match *)m;
+	int pad, off = xt_compat_match_offset(match);
+	u_int16_t msize = cm->u.user.match_size;
+
+	m = *dstptr;
+	memcpy(m, cm, sizeof(*cm));
+	if (match->compat_from_user)
+		match->compat_from_user(m->data, cm->data);
+	else
+		memcpy(m->data, cm->data, msize - sizeof(*cm));
+	pad = XT_ALIGN(match->matchsize) - match->matchsize;
+	if (pad > 0)
+		memset(m->data + match->matchsize, 0, pad);
+
+	msize += off;
+	m->u.user.match_size = msize;
+
+	*size += off;
+	*dstptr += msize;
+}
+EXPORT_SYMBOL_GPL(xt_compat_match_from_user);
+
+int xt_compat_match_to_user(struct xt_entry_match *m, void __user **dstptr,
+			    int *size)
+{
+	struct xt_match *match = m->u.kernel.match;
+	struct compat_xt_entry_match __user *cm = *dstptr;
+	int off = xt_compat_match_offset(match);
+	u_int16_t msize = m->u.user.match_size - off;
+
+	if (copy_to_user(cm, m, sizeof(*cm)) ||
+	    put_user(msize, &cm->u.user.match_size))
+	    	return -EFAULT;
+
+	if (match->compat_to_user) {
+		if (match->compat_to_user((void __user *)cm->data, m->data))
+			return -EFAULT;
+	} else {
+		if (copy_to_user(cm->data, m->data, msize - sizeof(*cm)))
+			return -EFAULT;
 	}
-	return ret;
+
+	*size -= off;
+	*dstptr += msize;
+	return 0;
 }
-EXPORT_SYMBOL_GPL(xt_compat_match);
-#endif
+EXPORT_SYMBOL_GPL(xt_compat_match_to_user);
+#endif /* CONFIG_COMPAT */
 
 int xt_check_target(const struct xt_target *target, unsigned short family,
 		    unsigned int size, const char *table, unsigned int hook_mask,
@@ -410,51 +423,64 @@ int xt_check_target(const struct xt_target *target, unsigned short family,
 EXPORT_SYMBOL_GPL(xt_check_target);
 
 #ifdef CONFIG_COMPAT
-int xt_compat_target(void *target, void **dstptr, int *size, int convert)
+int xt_compat_target_offset(struct xt_target *target)
 {
-	struct xt_target *t;
-	struct compat_xt_entry_target *pcompat;
-	struct xt_entry_target *pt;
-	u_int16_t tsize;
-	int off, ret;
+	u_int16_t csize = target->compatsize ? : target->targetsize;
+	return XT_ALIGN(target->targetsize) - COMPAT_XT_ALIGN(csize);
+}
+EXPORT_SYMBOL_GPL(xt_compat_target_offset);
 
-	ret = 0;
-	t = ((struct xt_entry_target *)target)->u.kernel.target;
-	off = XT_ALIGN(t->targetsize) - COMPAT_XT_ALIGN(t->targetsize);
-	switch (convert) {
-		case COMPAT_TO_USER:
-			pt = (struct xt_entry_target *)target;
-			tsize = pt->u.user.target_size;
-			if (copy_to_user(*dstptr, pt, tsize)) {
-				ret = -EFAULT;
-				break;
-			}
-			tsize -= off;
-			if (put_user(tsize, (u_int16_t *)*dstptr))
-				ret = -EFAULT;
-			*size -= off;
-			*dstptr += tsize;
-			break;
-		case COMPAT_FROM_USER:
-			pcompat = (struct compat_xt_entry_target *)target;
-			pt = (struct xt_entry_target *)*dstptr;
-			tsize = pcompat->u.user.target_size;
-			memcpy(pt, pcompat, tsize);
-			tsize += off;
-			pt->u.user.target_size = tsize;
-			*size += off;
-			*dstptr += tsize;
-			break;
-		case COMPAT_CALC_SIZE:
-			*size += off;
-			break;
-		default:
-			ret = -ENOPROTOOPT;
-			break;
+void xt_compat_target_from_user(struct xt_entry_target *t, void **dstptr,
+			        int *size)
+{
+	struct xt_target *target = t->u.kernel.target;
+	struct compat_xt_entry_target *ct = (struct compat_xt_entry_target *)t;
+	int pad, off = xt_compat_target_offset(target);
+	u_int16_t tsize = ct->u.user.target_size;
+
+	t = *dstptr;
+	memcpy(t, ct, sizeof(*ct));
+	if (target->compat_from_user)
+		target->compat_from_user(t->data, ct->data);
+	else
+		memcpy(t->data, ct->data, tsize - sizeof(*ct));
+	pad = XT_ALIGN(target->targetsize) - target->targetsize;
+	if (pad > 0)
+		memset(t->data + target->targetsize, 0, pad);
+
+	tsize += off;
+	t->u.user.target_size = tsize;
+
+	*size += off;
+	*dstptr += tsize;
+}
+EXPORT_SYMBOL_GPL(xt_compat_target_from_user);
+
+int xt_compat_target_to_user(struct xt_entry_target *t, void __user **dstptr,
+			     int *size)
+{
+	struct xt_target *target = t->u.kernel.target;
+	struct compat_xt_entry_target __user *ct = *dstptr;
+	int off = xt_compat_target_offset(target);
+	u_int16_t tsize = t->u.user.target_size - off;
+
+	if (copy_to_user(ct, t, sizeof(*ct)) ||
+	    put_user(tsize, &ct->u.user.target_size))
+	    	return -EFAULT;
+
+	if (target->compat_to_user) {
+		if (target->compat_to_user((void __user *)ct->data, t->data))
+			return -EFAULT;
+	} else {
+		if (copy_to_user(ct->data, t->data, tsize - sizeof(*ct)))
+			return -EFAULT;
 	}
-	return ret;
+
+	*size -= off;
+	*dstptr += tsize;
+	return 0;
 }
-EXPORT_SYMBOL_GPL(xt_compat_target);
+EXPORT_SYMBOL_GPL(xt_compat_target_to_user);
 #endif
 
 struct xt_table_info *xt_alloc_table_info(unsigned int size)
-- 
cgit v1.2.3


From edd5a329cf69c112882e03c8ab55e985062a5d2a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 12:07:39 -0700
Subject: [NETFILTER]: PPTP conntrack: fix whitespace errors

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_pptp.h | 26 +++----
 net/ipv4/netfilter/ip_conntrack_helper_pptp.c    | 76 ++++++++++----------
 net/ipv4/netfilter/ip_conntrack_proto_gre.c      | 28 ++++----
 net/ipv4/netfilter/ip_nat_helper_pptp.c          | 92 ++++++++++++------------
 net/ipv4/netfilter/ip_nat_proto_gre.c            | 20 +++---
 5 files changed, 121 insertions(+), 121 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
index 816144c75de0..88f66d3c8765 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
@@ -285,19 +285,19 @@ struct PptpSetLinkInfo {
 };
 
 union pptp_ctrl_union {
-		struct PptpStartSessionRequest	sreq;
-		struct PptpStartSessionReply	srep;
-		struct PptpStopSessionRequest	streq;
-		struct PptpStopSessionReply	strep;
-                struct PptpOutCallRequest       ocreq;
-                struct PptpOutCallReply         ocack;
-                struct PptpInCallRequest        icreq;
-                struct PptpInCallReply          icack;
-                struct PptpInCallConnected      iccon;
-		struct PptpClearCallRequest	clrreq;
-                struct PptpCallDisconnectNotify disc;
-                struct PptpWanErrorNotify       wanerr;
-                struct PptpSetLinkInfo          setlink;
+	struct PptpStartSessionRequest	sreq;
+	struct PptpStartSessionReply	srep;
+	struct PptpStopSessionRequest	streq;
+	struct PptpStopSessionReply	strep;
+	struct PptpOutCallRequest	ocreq;
+	struct PptpOutCallReply		ocack;
+	struct PptpInCallRequest	icreq;
+	struct PptpInCallReply		icack;
+	struct PptpInCallConnected	iccon;
+	struct PptpClearCallRequest	clrreq;
+	struct PptpCallDisconnectNotify disc;
+	struct PptpWanErrorNotify	wanerr;
+	struct PptpSetLinkInfo		setlink;
 };
 
 extern int
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
index b020a33e65e9..6c94dd5d476c 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
@@ -20,11 +20,11 @@
  * 	 - We can only support one single call within each session
  *
  * TODO:
- *	 - testing of incoming PPTP calls 
+ *	 - testing of incoming PPTP calls
  *
- * Changes: 
+ * Changes:
  * 	2002-02-05 - Version 1.3
- * 	  - Call ip_conntrack_unexpect_related() from 
+ * 	  - Call ip_conntrack_unexpect_related() from
  * 	    pptp_destroy_siblings() to destroy expectations in case
  * 	    CALL_DISCONNECT_NOTIFY or tcp fin packet was seen
  * 	    (Philip Craig <philipc@snapgear.com>)
@@ -141,7 +141,7 @@ static void pptp_expectfn(struct ip_conntrack *ct,
 		invert_tuplepr(&inv_t, &exp->tuple);
 		DEBUGP("trying to unexpect other dir: ");
 		DUMP_TUPLE(&inv_t);
-	
+
 		exp_other = ip_conntrack_expect_find(&inv_t);
 		if (exp_other) {
 			/* delete other expectation.  */
@@ -194,7 +194,7 @@ static void pptp_destroy_siblings(struct ip_conntrack *ct)
 {
 	struct ip_conntrack_tuple t;
 
-	/* Since ct->sibling_list has literally rusted away in 2.6.11, 
+	/* Since ct->sibling_list has literally rusted away in 2.6.11,
 	 * we now need another way to find out about our sibling
 	 * contrack and expects... -HW */
 
@@ -264,7 +264,7 @@ exp_gre(struct ip_conntrack *master,
 	exp_orig->mask.dst.u.gre.key = htons(0xffff);
 	exp_orig->mask.dst.ip = 0xffffffff;
 	exp_orig->mask.dst.protonum = 0xff;
-		
+
 	exp_orig->master = master;
 	exp_orig->expectfn = pptp_expectfn;
 	exp_orig->flags = 0;
@@ -322,7 +322,7 @@ out_unexpect_orig:
 	goto out_put_both;
 }
 
-static inline int 
+static inline int
 pptp_inbound_pkt(struct sk_buff **pskb,
 		 struct tcphdr *tcph,
 		 unsigned int nexthdr_off,
@@ -336,7 +336,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 	struct ip_ct_pptp_master *info = &ct->help.ct_pptp_info;
 	u_int16_t msg;
 	__be16 *cid, *pcid;
-	u_int32_t seq;	
+	u_int32_t seq;
 
 	ctlh = skb_header_pointer(*pskb, nexthdr_off, sizeof(_ctlh), &_ctlh);
 	if (!ctlh) {
@@ -373,7 +373,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		}
 		if (pptpReq->srep.resultCode == PPTP_START_OK)
 			info->sstate = PPTP_SESSION_CONFIRMED;
-		else 
+		else
 			info->sstate = PPTP_SESSION_ERROR;
 		break;
 
@@ -420,22 +420,22 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		pcid = &pptpReq->ocack.peersCallID;
 
 		info->pac_call_id = ntohs(*cid);
-		
+
 		if (htons(info->pns_call_id) != *pcid) {
 			DEBUGP("%s for unknown callid %u\n",
 				pptp_msg_name[msg], ntohs(*pcid));
 			break;
 		}
 
-		DEBUGP("%s, CID=%X, PCID=%X\n", pptp_msg_name[msg], 
+		DEBUGP("%s, CID=%X, PCID=%X\n", pptp_msg_name[msg],
 			ntohs(*cid), ntohs(*pcid));
-		
+
 		info->cstate = PPTP_CALL_OUT_CONF;
 
 		seq = ntohl(tcph->seq) + sizeof(struct pptp_pkt_hdr)
 				       + sizeof(struct PptpControlHeader)
 				       + ((void *)pcid - (void *)pptpReq);
-			
+
 		if (exp_gre(ct, seq, *cid, *pcid) != 0)
 			printk("ip_conntrack_pptp: error during exp_gre\n");
 		break;
@@ -479,7 +479,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		cid = &info->pac_call_id;
 
 		if (info->pns_call_id != ntohs(*pcid)) {
-			DEBUGP("%s for unknown CallID %u\n", 
+			DEBUGP("%s for unknown CallID %u\n",
 				pptp_msg_name[msg], ntohs(*pcid));
 			break;
 		}
@@ -491,7 +491,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		seq = ntohl(tcph->seq) + sizeof(struct pptp_pkt_hdr)
 				       + sizeof(struct PptpControlHeader)
 				       + ((void *)pcid - (void *)pptpReq);
-			
+
 		if (exp_gre(ct, seq, *cid, *pcid) != 0)
 			printk("ip_conntrack_pptp: error during exp_gre\n");
 
@@ -554,7 +554,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		return NF_ACCEPT;
 	nexthdr_off += sizeof(_ctlh);
 	datalen -= sizeof(_ctlh);
-	
+
 	reqlen = datalen;
 	if (reqlen > sizeof(*pptpReq))
 		reqlen = sizeof(*pptpReq);
@@ -606,7 +606,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		/* client answers incoming call */
 		if (info->cstate != PPTP_CALL_IN_REQ
 		    && info->cstate != PPTP_CALL_IN_REP) {
-			DEBUGP("%s without incall_req\n", 
+			DEBUGP("%s without incall_req\n",
 				pptp_msg_name[msg]);
 			break;
 		}
@@ -616,7 +616,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		}
 		pcid = &pptpReq->icack.peersCallID;
 		if (info->pac_call_id != ntohs(*pcid)) {
-			DEBUGP("%s for unknown call %u\n", 
+			DEBUGP("%s for unknown call %u\n",
 				pptp_msg_name[msg], ntohs(*pcid));
 			break;
 		}
@@ -644,12 +644,12 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		/* I don't have to explain these ;) */
 		break;
 	default:
-		DEBUGP("invalid %s (TY=%d)\n", (msg <= PPTP_MSG_MAX)? 
+		DEBUGP("invalid %s (TY=%d)\n", (msg <= PPTP_MSG_MAX)?
 			pptp_msg_name[msg]:pptp_msg_name[0], msg);
 		/* unknown: no need to create GRE masq table entry */
 		break;
 	}
-	
+
 	if (ip_nat_pptp_hook_outbound)
 		return ip_nat_pptp_hook_outbound(pskb, ct, ctinfo, ctlh,
 						 pptpReq);
@@ -659,7 +659,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 
 
 /* track caller id inside control connection, call expect_related */
-static int 
+static int
 conntrack_pptp_help(struct sk_buff **pskb,
 		    struct ip_conntrack *ct, enum ip_conntrack_info ctinfo)
 
@@ -676,12 +676,12 @@ conntrack_pptp_help(struct sk_buff **pskb,
 	int ret;
 
 	/* don't do any tracking before tcp handshake complete */
-	if (ctinfo != IP_CT_ESTABLISHED 
+	if (ctinfo != IP_CT_ESTABLISHED
 	    && ctinfo != IP_CT_ESTABLISHED+IP_CT_IS_REPLY) {
 		DEBUGP("ctinfo = %u, skipping\n", ctinfo);
 		return NF_ACCEPT;
 	}
-	
+
 	nexthdr_off = (*pskb)->nh.iph->ihl*4;
 	tcph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_tcph), &_tcph);
 	BUG_ON(!tcph);
@@ -735,28 +735,28 @@ conntrack_pptp_help(struct sk_buff **pskb,
 }
 
 /* control protocol helper */
-static struct ip_conntrack_helper pptp = { 
+static struct ip_conntrack_helper pptp = {
 	.list = { NULL, NULL },
-	.name = "pptp", 
+	.name = "pptp",
 	.me = THIS_MODULE,
 	.max_expected = 2,
 	.timeout = 5 * 60,
-	.tuple = { .src = { .ip = 0, 
-		 	    .u = { .tcp = { .port =  
-				    __constant_htons(PPTP_CONTROL_PORT) } } 
-			  }, 
-		   .dst = { .ip = 0, 
+	.tuple = { .src = { .ip = 0,
+		 	    .u = { .tcp = { .port =
+				    __constant_htons(PPTP_CONTROL_PORT) } }
+			  },
+		   .dst = { .ip = 0,
 			    .u = { .all = 0 },
 			    .protonum = IPPROTO_TCP
-			  } 
+			  }
 		 },
-	.mask = { .src = { .ip = 0, 
-			   .u = { .tcp = { .port = __constant_htons(0xffff) } } 
-			 }, 
-		  .dst = { .ip = 0, 
+	.mask = { .src = { .ip = 0,
+			   .u = { .tcp = { .port = __constant_htons(0xffff) } }
+			 },
+		  .dst = { .ip = 0,
 			   .u = { .all = 0 },
-			   .protonum = 0xff 
-		 	 } 
+			   .protonum = 0xff
+		 	 }
 		},
 	.help = conntrack_pptp_help
 };
@@ -768,7 +768,7 @@ extern int __init ip_ct_proto_gre_init(void);
 static int __init ip_conntrack_helper_pptp_init(void)
 {
 	int retcode;
- 
+
 	retcode = ip_ct_proto_gre_init();
 	if (retcode < 0)
 		return retcode;
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_gre.c b/net/ipv4/netfilter/ip_conntrack_proto_gre.c
index 92c6d8b178c9..5fe026f467d3 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_gre.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_gre.c
@@ -1,15 +1,15 @@
 /*
- * ip_conntrack_proto_gre.c - Version 3.0 
+ * ip_conntrack_proto_gre.c - Version 3.0
  *
  * Connection tracking protocol helper module for GRE.
  *
  * GRE is a generic encapsulation protocol, which is generally not very
  * suited for NAT, as it has no protocol-specific part as port numbers.
  *
- * It has an optional key field, which may help us distinguishing two 
+ * It has an optional key field, which may help us distinguishing two
  * connections between the same two hosts.
  *
- * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 
+ * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784
  *
  * PPTP is built on top of a modified version of GRE, and has a mandatory
  * field called "CallID", which serves us for the same purpose as the key
@@ -61,7 +61,7 @@ MODULE_DESCRIPTION("netfilter connection tracking protocol helper for GRE");
 #define DEBUGP(x, args...)
 #define DUMP_TUPLE_GRE(x)
 #endif
-				
+
 /* GRE KEYMAP HANDLING FUNCTIONS */
 static LIST_HEAD(gre_keymap_list);
 
@@ -88,7 +88,7 @@ static __be16 gre_keymap_lookup(struct ip_conntrack_tuple *t)
 		}
 	}
 	read_unlock_bh(&ip_ct_gre_lock);
-	
+
 	DEBUGP("lookup src key 0x%x up key for ", key);
 	DUMP_TUPLE_GRE(t);
 
@@ -107,7 +107,7 @@ ip_ct_gre_keymap_add(struct ip_conntrack *ct,
 		return -1;
 	}
 
-	if (!reply) 
+	if (!reply)
 		exist_km = &ct->help.ct_pptp_info.keymap_orig;
 	else
 		exist_km = &ct->help.ct_pptp_info.keymap_reply;
@@ -118,7 +118,7 @@ ip_ct_gre_keymap_add(struct ip_conntrack *ct,
 			if (gre_key_cmpfn(km, t) && km == *exist_km)
 				return 0;
 		}
-		DEBUGP("trying to override keymap_%s for ct %p\n", 
+		DEBUGP("trying to override keymap_%s for ct %p\n",
 			reply? "reply":"orig", ct);
 		return -EEXIST;
 	}
@@ -152,7 +152,7 @@ void ip_ct_gre_keymap_destroy(struct ip_conntrack *ct)
 
 	write_lock_bh(&ip_ct_gre_lock);
 	if (ct->help.ct_pptp_info.keymap_orig) {
-		DEBUGP("removing %p from list\n", 
+		DEBUGP("removing %p from list\n",
 			ct->help.ct_pptp_info.keymap_orig);
 		list_del(&ct->help.ct_pptp_info.keymap_orig->list);
 		kfree(ct->help.ct_pptp_info.keymap_orig);
@@ -220,7 +220,7 @@ static int gre_pkt_to_tuple(const struct sk_buff *skb,
 static int gre_print_tuple(struct seq_file *s,
 			   const struct ip_conntrack_tuple *tuple)
 {
-	return seq_printf(s, "srckey=0x%x dstkey=0x%x ", 
+	return seq_printf(s, "srckey=0x%x dstkey=0x%x ",
 			  ntohs(tuple->src.u.gre.key),
 			  ntohs(tuple->dst.u.gre.key));
 }
@@ -250,14 +250,14 @@ static int gre_packet(struct ip_conntrack *ct,
 	} else
 		ip_ct_refresh_acct(ct, conntrackinfo, skb,
 				   ct->proto.gre.timeout);
-	
+
 	return NF_ACCEPT;
 }
 
 /* Called when a new connection for this protocol found. */
 static int gre_new(struct ip_conntrack *ct,
 		   const struct sk_buff *skb)
-{ 
+{
 	DEBUGP(": ");
 	DUMP_TUPLE_GRE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
 
@@ -283,9 +283,9 @@ static void gre_destroy(struct ip_conntrack *ct)
 }
 
 /* protocol helper struct */
-static struct ip_conntrack_protocol gre = { 
+static struct ip_conntrack_protocol gre = {
 	.proto		 = IPPROTO_GRE,
-	.name		 = "gre", 
+	.name		 = "gre",
 	.pkt_to_tuple	 = gre_pkt_to_tuple,
 	.invert_tuple	 = gre_invert_tuple,
 	.print_tuple	 = gre_print_tuple,
@@ -323,7 +323,7 @@ void ip_ct_proto_gre_fini(void)
 	}
 	write_unlock_bh(&ip_ct_gre_lock);
 
-	ip_conntrack_protocol_unregister(&gre); 
+	ip_conntrack_protocol_unregister(&gre);
 }
 
 EXPORT_SYMBOL(ip_ct_gre_keymap_add);
diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c
index 1d149964dc38..5dde1da1c300 100644
--- a/net/ipv4/netfilter/ip_nat_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c
@@ -32,7 +32,7 @@
  *     2005-06-10 - Version 3.0
  *       - kernel >= 2.6.11 version,
  *	   funded by Oxcoda NetBox Blue (http://www.netboxblue.com/)
- * 
+ *
  */
 
 #include <linux/module.h>
@@ -93,10 +93,10 @@ static void pptp_nat_expected(struct ip_conntrack *ct,
 		DEBUGP("we are PAC->PNS\n");
 		/* build tuple for PNS->PAC */
 		t.src.ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip;
-		t.src.u.gre.key = 
+		t.src.u.gre.key =
 			htons(master->nat.help.nat_pptp_info.pns_call_id);
 		t.dst.ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip;
-		t.dst.u.gre.key = 
+		t.dst.u.gre.key =
 			htons(master->nat.help.nat_pptp_info.pac_call_id);
 		t.dst.protonum = IPPROTO_GRE;
 	}
@@ -153,47 +153,47 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 	unsigned int cid_off;
 
 	new_callid = htons(ct_pptp_info->pns_call_id);
-	
+
 	switch (msg = ntohs(ctlh->messageType)) {
-		case PPTP_OUT_CALL_REQUEST:
-			cid_off = offsetof(union pptp_ctrl_union, ocreq.callID);
-			/* FIXME: ideally we would want to reserve a call ID
-			 * here.  current netfilter NAT core is not able to do
-			 * this :( For now we use TCP source port. This breaks
-			 * multiple calls within one control session */
-
-			/* save original call ID in nat_info */
-			nat_pptp_info->pns_call_id = ct_pptp_info->pns_call_id;
-
-			/* don't use tcph->source since we are at a DSTmanip
-			 * hook (e.g. PREROUTING) and pkt is not mangled yet */
-			new_callid = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.tcp.port;
-
-			/* save new call ID in ct info */
-			ct_pptp_info->pns_call_id = ntohs(new_callid);
-			break;
-		case PPTP_IN_CALL_REPLY:
-			cid_off = offsetof(union pptp_ctrl_union, icreq.callID);
-			break;
-		case PPTP_CALL_CLEAR_REQUEST:
-			cid_off = offsetof(union pptp_ctrl_union, clrreq.callID);
-			break;
-		default:
-			DEBUGP("unknown outbound packet 0x%04x:%s\n", msg,
-			      (msg <= PPTP_MSG_MAX)? 
-			      pptp_msg_name[msg]:pptp_msg_name[0]);
-			/* fall through */
-
-		case PPTP_SET_LINK_INFO:
-			/* only need to NAT in case PAC is behind NAT box */
-		case PPTP_START_SESSION_REQUEST:
-		case PPTP_START_SESSION_REPLY:
-		case PPTP_STOP_SESSION_REQUEST:
-		case PPTP_STOP_SESSION_REPLY:
-		case PPTP_ECHO_REQUEST:
-		case PPTP_ECHO_REPLY:
-			/* no need to alter packet */
-			return NF_ACCEPT;
+	case PPTP_OUT_CALL_REQUEST:
+		cid_off = offsetof(union pptp_ctrl_union, ocreq.callID);
+		/* FIXME: ideally we would want to reserve a call ID
+		 * here.  current netfilter NAT core is not able to do
+		 * this :( For now we use TCP source port. This breaks
+		 * multiple calls within one control session */
+
+		/* save original call ID in nat_info */
+		nat_pptp_info->pns_call_id = ct_pptp_info->pns_call_id;
+
+		/* don't use tcph->source since we are at a DSTmanip
+		 * hook (e.g. PREROUTING) and pkt is not mangled yet */
+		new_callid = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.tcp.port;
+
+		/* save new call ID in ct info */
+		ct_pptp_info->pns_call_id = ntohs(new_callid);
+		break;
+	case PPTP_IN_CALL_REPLY:
+		cid_off = offsetof(union pptp_ctrl_union, icreq.callID);
+		break;
+	case PPTP_CALL_CLEAR_REQUEST:
+		cid_off = offsetof(union pptp_ctrl_union, clrreq.callID);
+		break;
+	default:
+		DEBUGP("unknown outbound packet 0x%04x:%s\n", msg,
+		      (msg <= PPTP_MSG_MAX)?
+		      pptp_msg_name[msg]:pptp_msg_name[0]);
+		/* fall through */
+
+	case PPTP_SET_LINK_INFO:
+		/* only need to NAT in case PAC is behind NAT box */
+	case PPTP_START_SESSION_REQUEST:
+	case PPTP_START_SESSION_REPLY:
+	case PPTP_STOP_SESSION_REQUEST:
+	case PPTP_STOP_SESSION_REPLY:
+	case PPTP_ECHO_REQUEST:
+	case PPTP_ECHO_REPLY:
+		/* no need to alter packet */
+		return NF_ACCEPT;
 	}
 
 	/* only OUT_CALL_REQUEST, IN_CALL_REPLY, CALL_CLEAR_REQUEST pass
@@ -216,9 +216,9 @@ static int
 pptp_exp_gre(struct ip_conntrack_expect *expect_orig,
 	     struct ip_conntrack_expect *expect_reply)
 {
-	struct ip_ct_pptp_master *ct_pptp_info = 
+	struct ip_ct_pptp_master *ct_pptp_info =
 				&expect_orig->master->help.ct_pptp_info;
-	struct ip_nat_pptp *nat_pptp_info = 
+	struct ip_nat_pptp *nat_pptp_info =
 				&expect_orig->master->nat.help.nat_pptp_info;
 
 	struct ip_conntrack *ct = expect_orig->master;
@@ -324,7 +324,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		break;
 
 	default:
-		DEBUGP("unknown inbound packet %s\n", (msg <= PPTP_MSG_MAX)? 
+		DEBUGP("unknown inbound packet %s\n", (msg <= PPTP_MSG_MAX)?
 			pptp_msg_name[msg]:pptp_msg_name[0]);
 		/* fall through */
 
diff --git a/net/ipv4/netfilter/ip_nat_proto_gre.c b/net/ipv4/netfilter/ip_nat_proto_gre.c
index 70a65372225a..a5226691f02c 100644
--- a/net/ipv4/netfilter/ip_nat_proto_gre.c
+++ b/net/ipv4/netfilter/ip_nat_proto_gre.c
@@ -6,10 +6,10 @@
  * GRE is a generic encapsulation protocol, which is generally not very
  * suited for NAT, as it has no protocol-specific part as port numbers.
  *
- * It has an optional key field, which may help us distinguishing two 
+ * It has an optional key field, which may help us distinguishing two
  * connections between the same two hosts.
  *
- * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 
+ * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784
  *
  * PPTP is built on top of a modified version of GRE, and has a mandatory
  * field called "CallID", which serves us for the same purpose as the key
@@ -60,7 +60,7 @@ gre_in_range(const struct ip_conntrack_tuple *tuple,
 }
 
 /* generate unique tuple ... */
-static int 
+static int
 gre_unique_tuple(struct ip_conntrack_tuple *tuple,
 		 const struct ip_nat_range *range,
 		 enum ip_nat_manip_type maniptype,
@@ -84,7 +84,7 @@ gre_unique_tuple(struct ip_conntrack_tuple *tuple,
 		range_size = ntohs(range->max.gre.key) - min + 1;
 	}
 
-	DEBUGP("min = %u, range_size = %u\n", min, range_size); 
+	DEBUGP("min = %u, range_size = %u\n", min, range_size);
 
 	for (i = 0; i < range_size; i++, key++) {
 		*keyptr = htons(min + key % range_size);
@@ -117,7 +117,7 @@ gre_manip_pkt(struct sk_buff **pskb,
 	greh = (void *)(*pskb)->data + hdroff;
 	pgreh = (struct gre_hdr_pptp *) greh;
 
-	/* we only have destination manip of a packet, since 'source key' 
+	/* we only have destination manip of a packet, since 'source key'
 	 * is not present in the packet itself */
 	if (maniptype == IP_NAT_MANIP_DST) {
 		/* key manipulation is always dest */
@@ -129,7 +129,7 @@ gre_manip_pkt(struct sk_buff **pskb,
 			}
 			if (greh->csum) {
 				/* FIXME: Never tested this code... */
-				*(gre_csum(greh)) = 
+				*(gre_csum(greh)) =
 					nf_proto_csum_update(*pskb,
 							~*(gre_key(greh)),
 							tuple->dst.u.gre.key,
@@ -138,7 +138,7 @@ gre_manip_pkt(struct sk_buff **pskb,
 			*(gre_key(greh)) = tuple->dst.u.gre.key;
 			break;
 		case GRE_VERSION_PPTP:
-			DEBUGP("call_id -> 0x%04x\n", 
+			DEBUGP("call_id -> 0x%04x\n",
 				ntohs(tuple->dst.u.gre.key));
 			pgreh->call_id = tuple->dst.u.gre.key;
 			break;
@@ -152,8 +152,8 @@ gre_manip_pkt(struct sk_buff **pskb,
 }
 
 /* nat helper struct */
-static struct ip_nat_protocol gre = { 
-	.name		= "GRE", 
+static struct ip_nat_protocol gre = {
+	.name		= "GRE",
 	.protonum	= IPPROTO_GRE,
 	.manip_pkt	= gre_manip_pkt,
 	.in_range	= gre_in_range,
@@ -164,7 +164,7 @@ static struct ip_nat_protocol gre = {
 	.nfattr_to_range	= ip_nat_port_nfattr_to_range,
 #endif
 };
-				  
+
 int __init ip_nat_proto_gre_init(void)
 {
 	return ip_nat_protocol_register(&gre);
-- 
cgit v1.2.3


From 955b944293dd4c931ec866ebe19a6b2463b8f9a0 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 12:08:03 -0700
Subject: [NETFILTER]: PPTP conntrack: get rid of unnecessary byte order
 conversions

The conntrack structure contains the call ID in host byte order for no
reason, get rid of back and forth conversions.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_pptp.h   |  8 ++---
 .../linux/netfilter_ipv4/ip_conntrack_proto_gre.h  | 22 ++++++------
 include/linux/netfilter_ipv4/ip_nat_pptp.h         |  4 +--
 net/ipv4/netfilter/ip_conntrack_helper_pptp.c      | 22 ++++++------
 net/ipv4/netfilter/ip_nat_helper_pptp.c            | 42 +++++++++++-----------
 net/ipv4/netfilter/ip_nat_proto_gre.c              |  2 +-
 6 files changed, 50 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
index 88f66d3c8765..0d35623f9453 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
@@ -31,8 +31,8 @@ struct ip_ct_pptp_master {
 	/* everything below is going to be per-expectation in newnat,
 	 * since there could be more than one call within one session */
 	enum pptp_ctrlcall_state cstate;	/* call state */
-	u_int16_t pac_call_id;			/* call id of PAC, host byte order */
-	u_int16_t pns_call_id;			/* call id of PNS, host byte order */
+	__be16 pac_call_id;			/* call id of PAC, host byte order */
+	__be16 pns_call_id;			/* call id of PNS, host byte order */
 
 	/* in pre-2.6.11 this used to be per-expect. Now it is per-conntrack
 	 * and therefore imposes a fixed limit on the number of maps */
@@ -42,8 +42,8 @@ struct ip_ct_pptp_master {
 /* conntrack_expect private member */
 struct ip_ct_pptp_expect {
 	enum pptp_ctrlcall_state cstate; 	/* call state */
-	u_int16_t pac_call_id;			/* call id of PAC */
-	u_int16_t pns_call_id;			/* call id of PNS */
+	__be16 pac_call_id;			/* call id of PAC */
+	__be16 pns_call_id;			/* call id of PNS */
 };
 
 
diff --git a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h
index 8d090ef82f5f..1d853aa873eb 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h
@@ -49,18 +49,18 @@ struct gre_hdr {
 #else
 #error "Adjust your <asm/byteorder.h> defines"
 #endif
-	__u16	protocol;
+	__be16	protocol;
 };
 
 /* modified GRE header for PPTP */
 struct gre_hdr_pptp {
-	__u8  flags;		/* bitfield */
-	__u8  version;		/* should be GRE_VERSION_PPTP */
-	__u16 protocol;		/* should be GRE_PROTOCOL_PPTP */
-	__u16 payload_len;	/* size of ppp payload, not inc. gre header */
-	__u16 call_id;		/* peer's call_id for this session */
-	__u32 seq;		/* sequence number.  Present if S==1 */
-	__u32 ack;		/* seq number of highest packet recieved by */
+	__u8   flags;		/* bitfield */
+	__u8   version;		/* should be GRE_VERSION_PPTP */
+	__be16 protocol;	/* should be GRE_PROTOCOL_PPTP */
+	__be16 payload_len;	/* size of ppp payload, not inc. gre header */
+	__be16 call_id;		/* peer's call_id for this session */
+	__be32 seq;		/* sequence number.  Present if S==1 */
+	__be32 ack;		/* seq number of highest packet recieved by */
 				/*  sender in this session */
 };
 
@@ -92,13 +92,13 @@ void ip_ct_gre_keymap_destroy(struct ip_conntrack *ct);
 
 
 /* get pointer to gre key, if present */
-static inline u_int32_t *gre_key(struct gre_hdr *greh)
+static inline __be32 *gre_key(struct gre_hdr *greh)
 {
 	if (!greh->key)
 		return NULL;
 	if (greh->csum || greh->routing)
-		return (u_int32_t *) (greh+sizeof(*greh)+4);
-	return (u_int32_t *) (greh+sizeof(*greh));
+		return (__be32 *) (greh+sizeof(*greh)+4);
+	return (__be32 *) (greh+sizeof(*greh));
 }
 
 /* get pointer ot gre csum, if present */
diff --git a/include/linux/netfilter_ipv4/ip_nat_pptp.h b/include/linux/netfilter_ipv4/ip_nat_pptp.h
index eaf66c2e8f93..36668bf0f373 100644
--- a/include/linux/netfilter_ipv4/ip_nat_pptp.h
+++ b/include/linux/netfilter_ipv4/ip_nat_pptp.h
@@ -4,8 +4,8 @@
 
 /* conntrack private data */
 struct ip_nat_pptp {
-	u_int16_t pns_call_id;		/* NAT'ed PNS call id */
-	u_int16_t pac_call_id;		/* NAT'ed PAC call id */
+	__be16 pns_call_id;		/* NAT'ed PNS call id */
+	__be16 pac_call_id;		/* NAT'ed PAC call id */
 };
 
 #endif /* _NAT_PPTP_H */
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
index 6c94dd5d476c..57637ca2b82c 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
@@ -201,8 +201,8 @@ static void pptp_destroy_siblings(struct ip_conntrack *ct)
 	/* try original (pns->pac) tuple */
 	memcpy(&t, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, sizeof(t));
 	t.dst.protonum = IPPROTO_GRE;
-	t.src.u.gre.key = htons(ct->help.ct_pptp_info.pns_call_id);
-	t.dst.u.gre.key = htons(ct->help.ct_pptp_info.pac_call_id);
+	t.src.u.gre.key = ct->help.ct_pptp_info.pns_call_id;
+	t.dst.u.gre.key = ct->help.ct_pptp_info.pac_call_id;
 
 	if (!destroy_sibling_or_exp(&t))
 		DEBUGP("failed to timeout original pns->pac ct/exp\n");
@@ -210,8 +210,8 @@ static void pptp_destroy_siblings(struct ip_conntrack *ct)
 	/* try reply (pac->pns) tuple */
 	memcpy(&t, &ct->tuplehash[IP_CT_DIR_REPLY].tuple, sizeof(t));
 	t.dst.protonum = IPPROTO_GRE;
-	t.src.u.gre.key = htons(ct->help.ct_pptp_info.pac_call_id);
-	t.dst.u.gre.key = htons(ct->help.ct_pptp_info.pns_call_id);
+	t.src.u.gre.key = ct->help.ct_pptp_info.pac_call_id;
+	t.dst.u.gre.key = ct->help.ct_pptp_info.pns_call_id;
 
 	if (!destroy_sibling_or_exp(&t))
 		DEBUGP("failed to timeout reply pac->pns ct/exp\n");
@@ -419,9 +419,9 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		cid = &pptpReq->ocack.callID;
 		pcid = &pptpReq->ocack.peersCallID;
 
-		info->pac_call_id = ntohs(*cid);
+		info->pac_call_id = *cid;
 
-		if (htons(info->pns_call_id) != *pcid) {
+		if (info->pns_call_id != *pcid) {
 			DEBUGP("%s for unknown callid %u\n",
 				pptp_msg_name[msg], ntohs(*pcid));
 			break;
@@ -454,7 +454,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		pcid = &pptpReq->icack.peersCallID;
 		DEBUGP("%s, PCID=%X\n", pptp_msg_name[msg], ntohs(*pcid));
 		info->cstate = PPTP_CALL_IN_REQ;
-		info->pac_call_id = ntohs(*pcid);
+		info->pac_call_id = *pcid;
 		break;
 
 	case PPTP_IN_CALL_CONNECT:
@@ -478,7 +478,7 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		pcid = &pptpReq->iccon.peersCallID;
 		cid = &info->pac_call_id;
 
-		if (info->pns_call_id != ntohs(*pcid)) {
+		if (info->pns_call_id != *pcid) {
 			DEBUGP("%s for unknown CallID %u\n",
 				pptp_msg_name[msg], ntohs(*pcid));
 			break;
@@ -595,7 +595,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		/* track PNS call id */
 		cid = &pptpReq->ocreq.callID;
 		DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(*cid));
-		info->pns_call_id = ntohs(*cid);
+		info->pns_call_id = *cid;
 		break;
 	case PPTP_IN_CALL_REPLY:
 		if (reqlen < sizeof(_pptpReq.icack)) {
@@ -615,7 +615,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 			break;
 		}
 		pcid = &pptpReq->icack.peersCallID;
-		if (info->pac_call_id != ntohs(*pcid)) {
+		if (info->pac_call_id != *pcid) {
 			DEBUGP("%s for unknown call %u\n",
 				pptp_msg_name[msg], ntohs(*pcid));
 			break;
@@ -623,7 +623,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(*pcid));
 		/* part two of the three-way handshake */
 		info->cstate = PPTP_CALL_IN_REP;
-		info->pns_call_id = ntohs(pptpReq->icack.callID);
+		info->pns_call_id = pptpReq->icack.callID;
 		break;
 
 	case PPTP_CALL_CLEAR_REQUEST:
diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c
index 5dde1da1c300..6e8bd6b3431f 100644
--- a/net/ipv4/netfilter/ip_nat_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c
@@ -85,19 +85,17 @@ static void pptp_nat_expected(struct ip_conntrack *ct,
 		DEBUGP("we are PNS->PAC\n");
 		/* therefore, build tuple for PAC->PNS */
 		t.src.ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip;
-		t.src.u.gre.key = htons(master->help.ct_pptp_info.pac_call_id);
+		t.src.u.gre.key = master->help.ct_pptp_info.pac_call_id;
 		t.dst.ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip;
-		t.dst.u.gre.key = htons(master->help.ct_pptp_info.pns_call_id);
+		t.dst.u.gre.key = master->help.ct_pptp_info.pns_call_id;
 		t.dst.protonum = IPPROTO_GRE;
 	} else {
 		DEBUGP("we are PAC->PNS\n");
 		/* build tuple for PNS->PAC */
 		t.src.ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip;
-		t.src.u.gre.key =
-			htons(master->nat.help.nat_pptp_info.pns_call_id);
+		t.src.u.gre.key = master->nat.help.nat_pptp_info.pns_call_id;
 		t.dst.ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip;
-		t.dst.u.gre.key =
-			htons(master->nat.help.nat_pptp_info.pac_call_id);
+		t.dst.u.gre.key = master->nat.help.nat_pptp_info.pac_call_id;
 		t.dst.protonum = IPPROTO_GRE;
 	}
 
@@ -149,10 +147,11 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 {
 	struct ip_ct_pptp_master *ct_pptp_info = &ct->help.ct_pptp_info;
 	struct ip_nat_pptp *nat_pptp_info = &ct->nat.help.nat_pptp_info;
-	u_int16_t msg, new_callid;
+	u_int16_t msg;
+	__be16 new_callid;
 	unsigned int cid_off;
 
-	new_callid = htons(ct_pptp_info->pns_call_id);
+	new_callid = ct_pptp_info->pns_call_id;
 
 	switch (msg = ntohs(ctlh->messageType)) {
 	case PPTP_OUT_CALL_REQUEST:
@@ -170,7 +169,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 		new_callid = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.tcp.port;
 
 		/* save new call ID in ct info */
-		ct_pptp_info->pns_call_id = ntohs(new_callid);
+		ct_pptp_info->pns_call_id = new_callid;
 		break;
 	case PPTP_IN_CALL_REPLY:
 		cid_off = offsetof(union pptp_ctrl_union, icreq.callID);
@@ -235,14 +234,14 @@ pptp_exp_gre(struct ip_conntrack_expect *expect_orig,
 
 	/* alter expectation for PNS->PAC direction */
 	invert_tuplepr(&inv_t, &expect_orig->tuple);
-	expect_orig->saved_proto.gre.key = htons(ct_pptp_info->pns_call_id);
-	expect_orig->tuple.src.u.gre.key = htons(nat_pptp_info->pns_call_id);
-	expect_orig->tuple.dst.u.gre.key = htons(ct_pptp_info->pac_call_id);
+	expect_orig->saved_proto.gre.key = ct_pptp_info->pns_call_id;
+	expect_orig->tuple.src.u.gre.key = nat_pptp_info->pns_call_id;
+	expect_orig->tuple.dst.u.gre.key = ct_pptp_info->pac_call_id;
 	expect_orig->dir = IP_CT_DIR_ORIGINAL;
 	inv_t.src.ip = reply_t->src.ip;
 	inv_t.dst.ip = reply_t->dst.ip;
-	inv_t.src.u.gre.key = htons(nat_pptp_info->pac_call_id);
-	inv_t.dst.u.gre.key = htons(ct_pptp_info->pns_call_id);
+	inv_t.src.u.gre.key = nat_pptp_info->pac_call_id;
+	inv_t.dst.u.gre.key = ct_pptp_info->pns_call_id;
 
 	if (!ip_conntrack_expect_related(expect_orig)) {
 		DEBUGP("successfully registered expect\n");
@@ -253,14 +252,14 @@ pptp_exp_gre(struct ip_conntrack_expect *expect_orig,
 
 	/* alter expectation for PAC->PNS direction */
 	invert_tuplepr(&inv_t, &expect_reply->tuple);
-	expect_reply->saved_proto.gre.key = htons(nat_pptp_info->pns_call_id);
-	expect_reply->tuple.src.u.gre.key = htons(nat_pptp_info->pac_call_id);
-	expect_reply->tuple.dst.u.gre.key = htons(ct_pptp_info->pns_call_id);
+	expect_reply->saved_proto.gre.key = nat_pptp_info->pns_call_id;
+	expect_reply->tuple.src.u.gre.key = nat_pptp_info->pac_call_id;
+	expect_reply->tuple.dst.u.gre.key = ct_pptp_info->pns_call_id;
 	expect_reply->dir = IP_CT_DIR_REPLY;
 	inv_t.src.ip = orig_t->src.ip;
 	inv_t.dst.ip = orig_t->dst.ip;
-	inv_t.src.u.gre.key = htons(nat_pptp_info->pns_call_id);
-	inv_t.dst.u.gre.key = htons(ct_pptp_info->pac_call_id);
+	inv_t.src.u.gre.key = nat_pptp_info->pns_call_id;
+	inv_t.dst.u.gre.key = ct_pptp_info->pac_call_id;
 
 	if (!ip_conntrack_expect_related(expect_reply)) {
 		DEBUGP("successfully registered expect\n");
@@ -297,10 +296,11 @@ pptp_inbound_pkt(struct sk_buff **pskb,
 		 union pptp_ctrl_union *pptpReq)
 {
 	struct ip_nat_pptp *nat_pptp_info = &ct->nat.help.nat_pptp_info;
-	u_int16_t msg, new_cid = 0, new_pcid;
+	u_int16_t msg, new_cid = 0;
+	__be16 new_pcid;
 	unsigned int pcid_off, cid_off = 0;
 
-	new_pcid = htons(nat_pptp_info->pns_call_id);
+	new_pcid = nat_pptp_info->pns_call_id;
 
 	switch (msg = ntohs(ctlh->messageType)) {
 	case PPTP_OUT_CALL_REPLY:
diff --git a/net/ipv4/netfilter/ip_nat_proto_gre.c b/net/ipv4/netfilter/ip_nat_proto_gre.c
index a5226691f02c..bf91f9312b3c 100644
--- a/net/ipv4/netfilter/ip_nat_proto_gre.c
+++ b/net/ipv4/netfilter/ip_nat_proto_gre.c
@@ -67,7 +67,7 @@ gre_unique_tuple(struct ip_conntrack_tuple *tuple,
 		 const struct ip_conntrack *conntrack)
 {
 	static u_int16_t key;
-	u_int16_t *keyptr;
+	__be16 *keyptr;
 	unsigned int min, i, range_size;
 
 	if (maniptype == IP_NAT_MANIP_SRC)
-- 
cgit v1.2.3


From 6013c0a13e335674a783215e182c367406294392 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 12:08:56 -0700
Subject: [NETFILTER]: PPTP conntrack: fix header definitions

Fix a few header definitions to match RFC2637. Most importantly the
PptpOutCallRequest header included an invalid padding field and a
size check was disabled because of this.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_pptp.h | 9 +++++----
 net/ipv4/netfilter/ip_conntrack_helper_pptp.c    | 2 +-
 2 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
index 0d35623f9453..620bf06fabc2 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
@@ -107,8 +107,7 @@ struct PptpControlHeader {
 
 struct PptpStartSessionRequest {
 	__be16	protocolVersion;
-	__u8	reserved1;
-	__u8	reserved2;
+	__u16	reserved1;
 	__be32	framingCapability;
 	__be32	bearerCapability;
 	__be16	maxChannels;
@@ -143,6 +142,8 @@ struct PptpStartSessionReply {
 
 struct PptpStopSessionRequest {
 	__u8	reason;
+	__u8	reserved1;
+	__u16	reserved2;
 };
 
 /* PptpStopSessionResultCode */
@@ -152,6 +153,7 @@ struct PptpStopSessionRequest {
 struct PptpStopSessionReply {
 	__u8	resultCode;
 	__u8	generalErrorCode;
+	__u16	reserved1;
 };
 
 struct PptpEchoRequest {
@@ -188,9 +190,8 @@ struct PptpOutCallRequest {
 	__be32	framingType;
 	__be16	packetWindow;
 	__be16	packetProcDelay;
-	__u16	reserved1;
 	__be16	phoneNumberLength;
-	__u16	reserved2;
+	__u16	reserved1;
 	__u8	phoneNumber[64];
 	__u8	subAddress[64];
 };
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
index 0510ee50dc65..1a8da9015d8c 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
@@ -569,7 +569,7 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 	case PPTP_OUT_CALL_REQUEST:
 		if (reqlen < sizeof(_pptpReq.ocreq)) {
 			DEBUGP("%s: short packet\n", pptp_msg_name[msg]);
-			/* FIXME: break; */
+			break;
 		}
 
 		/* client initiating connection to server */
-- 
cgit v1.2.3


From cf9f81523ef3e95d9f222c896d266e4562999150 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 12:09:34 -0700
Subject: [NETFILTER]: PPTP conntrack: simplify expectation handling

Remove duplicated expectation handling in the NAT helper and simplify
the remains in the conntrack helper.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_pptp.h |  2 +-
 net/ipv4/netfilter/ip_conntrack_helper_pptp.c    | 92 ++++++++----------------
 net/ipv4/netfilter/ip_nat_helper_pptp.c          | 58 +--------------
 3 files changed, 35 insertions(+), 117 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
index 620bf06fabc2..2644b1faddd6 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_pptp.h
@@ -315,7 +315,7 @@ extern int
 			  struct PptpControlHeader *ctlh,
 			  union pptp_ctrl_union *pptpReq);
 
-extern int
+extern void
 (*ip_nat_pptp_hook_exp_gre)(struct ip_conntrack_expect *exp_orig,
 			    struct ip_conntrack_expect *exp_reply);
 
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
index 5f7af6ef3881..57eac6e3871a 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
@@ -80,7 +80,7 @@ int
 			  struct PptpControlHeader *ctlh,
 			  union pptp_ctrl_union *pptpReq);
 
-int
+void
 (*ip_nat_pptp_hook_exp_gre)(struct ip_conntrack_expect *expect_orig,
 			    struct ip_conntrack_expect *expect_reply);
 
@@ -219,93 +219,63 @@ static void pptp_destroy_siblings(struct ip_conntrack *ct)
 
 /* expect GRE connections (PNS->PAC and PAC->PNS direction) */
 static inline int
-exp_gre(struct ip_conntrack *master,
+exp_gre(struct ip_conntrack *ct,
 	__be16 callid,
 	__be16 peer_callid)
 {
-	struct ip_conntrack_tuple inv_tuple;
-	struct ip_conntrack_tuple exp_tuples[] = {
-		/* tuple in original direction, PNS->PAC */
-		{ .src = { .ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip,
-			   .u = { .gre = { .key = peer_callid } }
-			 },
-		  .dst = { .ip = master->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip,
-			   .u = { .gre = { .key = callid } },
-			   .protonum = IPPROTO_GRE
-			 },
-		 },
-		/* tuple in reply direction, PAC->PNS */
-		{ .src = { .ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip,
-			   .u = { .gre = { .key = callid } }
-			 },
-		  .dst = { .ip = master->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip,
-			   .u = { .gre = { .key = peer_callid } },
-			   .protonum = IPPROTO_GRE
-			 },
-		 }
-	};
 	struct ip_conntrack_expect *exp_orig, *exp_reply;
 	int ret = 1;
 
-	exp_orig = ip_conntrack_expect_alloc(master);
+	exp_orig = ip_conntrack_expect_alloc(ct);
 	if (exp_orig == NULL)
 		goto out;
 
-	exp_reply = ip_conntrack_expect_alloc(master);
+	exp_reply = ip_conntrack_expect_alloc(ct);
 	if (exp_reply == NULL)
 		goto out_put_orig;
 
-	memcpy(&exp_orig->tuple, &exp_tuples[0], sizeof(exp_orig->tuple));
+	/* original direction, PNS->PAC */
+	exp_orig->tuple.src.ip = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip;
+	exp_orig->tuple.src.u.gre.key = peer_callid;
+	exp_orig->tuple.dst.ip = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip;
+	exp_orig->tuple.dst.u.gre.key = callid;
+	exp_orig->tuple.dst.protonum = IPPROTO_GRE;
 
 	exp_orig->mask.src.ip = 0xffffffff;
 	exp_orig->mask.src.u.all = 0;
-	exp_orig->mask.dst.u.all = 0;
 	exp_orig->mask.dst.u.gre.key = htons(0xffff);
 	exp_orig->mask.dst.ip = 0xffffffff;
 	exp_orig->mask.dst.protonum = 0xff;
 
-	exp_orig->master = master;
+	exp_orig->master = ct;
 	exp_orig->expectfn = pptp_expectfn;
 	exp_orig->flags = 0;
 
 	/* both expectations are identical apart from tuple */
 	memcpy(exp_reply, exp_orig, sizeof(*exp_reply));
-	memcpy(&exp_reply->tuple, &exp_tuples[1], sizeof(exp_reply->tuple));
-
-	if (ip_nat_pptp_hook_exp_gre)
-		ret = ip_nat_pptp_hook_exp_gre(exp_orig, exp_reply);
-	else {
 
-		DEBUGP("calling expect_related PNS->PAC");
-		DUMP_TUPLE(&exp_orig->tuple);
+	/* reply direction, PAC->PNS */
+	exp_reply->tuple.src.ip = ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip;
+	exp_reply->tuple.src.u.gre.key = callid;
+	exp_reply->tuple.dst.ip = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip;
+	exp_reply->tuple.dst.u.gre.key = peer_callid;
+	exp_reply->tuple.dst.protonum = IPPROTO_GRE;
 
-		if (ip_conntrack_expect_related(exp_orig) != 0) {
-			DEBUGP("cannot expect_related()\n");
-			goto out_put_both;
-		}
-
-		DEBUGP("calling expect_related PAC->PNS");
-		DUMP_TUPLE(&exp_reply->tuple);
-
-		if (ip_conntrack_expect_related(exp_reply) != 0) {
-			DEBUGP("cannot expect_related()\n");
-			goto out_unexpect_orig;
-		}
-
-		/* Add GRE keymap entries */
-		if (ip_ct_gre_keymap_add(master, &exp_reply->tuple, 0) != 0) {
-			DEBUGP("cannot keymap_add() exp\n");
-			goto out_unexpect_both;
-		}
-
-		invert_tuplepr(&inv_tuple, &exp_reply->tuple);
-		if (ip_ct_gre_keymap_add(master, &inv_tuple, 1) != 0) {
-			ip_ct_gre_keymap_destroy(master);
-			DEBUGP("cannot keymap_add() exp_inv\n");
-			goto out_unexpect_both;
-		}
-		ret = 0;
+	if (ip_nat_pptp_hook_exp_gre)
+		ip_nat_pptp_hook_exp_gre(exp_orig, exp_reply);
+	if (ip_conntrack_expect_related(exp_orig) != 0)
+		goto out_put_both;
+	if (ip_conntrack_expect_related(exp_reply) != 0)
+		goto out_unexpect_orig;
+
+	/* Add GRE keymap entries */
+	if (ip_ct_gre_keymap_add(ct, &exp_orig->tuple, 0) != 0)
+		goto out_unexpect_both;
+	if (ip_ct_gre_keymap_add(ct, &exp_reply->tuple, 1) != 0) {
+		ip_ct_gre_keymap_destroy(ct);
+		goto out_unexpect_both;
 	}
+	ret = 0;
 
 out_put_both:
 	ip_conntrack_expect_put(exp_reply);
diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c
index 0f5e753b481d..84f6bd09fcd4 100644
--- a/net/ipv4/netfilter/ip_nat_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c
@@ -211,80 +211,28 @@ pptp_outbound_pkt(struct sk_buff **pskb,
 	return NF_ACCEPT;
 }
 
-static int
+static void
 pptp_exp_gre(struct ip_conntrack_expect *expect_orig,
 	     struct ip_conntrack_expect *expect_reply)
 {
-	struct ip_ct_pptp_master *ct_pptp_info =
-				&expect_orig->master->help.ct_pptp_info;
-	struct ip_nat_pptp *nat_pptp_info =
-				&expect_orig->master->nat.help.nat_pptp_info;
-
 	struct ip_conntrack *ct = expect_orig->master;
-
-	struct ip_conntrack_tuple inv_t;
-	struct ip_conntrack_tuple *orig_t, *reply_t;
+	struct ip_ct_pptp_master *ct_pptp_info = &ct->help.ct_pptp_info;
+	struct ip_nat_pptp *nat_pptp_info = &ct->nat.help.nat_pptp_info;
 
 	/* save original PAC call ID in nat_info */
 	nat_pptp_info->pac_call_id = ct_pptp_info->pac_call_id;
 
-	/* alter expectation */
-	orig_t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
-	reply_t = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-
 	/* alter expectation for PNS->PAC direction */
-	invert_tuplepr(&inv_t, &expect_orig->tuple);
 	expect_orig->saved_proto.gre.key = ct_pptp_info->pns_call_id;
 	expect_orig->tuple.src.u.gre.key = nat_pptp_info->pns_call_id;
 	expect_orig->tuple.dst.u.gre.key = ct_pptp_info->pac_call_id;
 	expect_orig->dir = IP_CT_DIR_ORIGINAL;
-	inv_t.src.ip = reply_t->src.ip;
-	inv_t.dst.ip = reply_t->dst.ip;
-	inv_t.src.u.gre.key = nat_pptp_info->pac_call_id;
-	inv_t.dst.u.gre.key = ct_pptp_info->pns_call_id;
-
-	if (!ip_conntrack_expect_related(expect_orig)) {
-		DEBUGP("successfully registered expect\n");
-	} else {
-		DEBUGP("can't expect_related(expect_orig)\n");
-		return 1;
-	}
 
 	/* alter expectation for PAC->PNS direction */
-	invert_tuplepr(&inv_t, &expect_reply->tuple);
 	expect_reply->saved_proto.gre.key = nat_pptp_info->pns_call_id;
 	expect_reply->tuple.src.u.gre.key = nat_pptp_info->pac_call_id;
 	expect_reply->tuple.dst.u.gre.key = ct_pptp_info->pns_call_id;
 	expect_reply->dir = IP_CT_DIR_REPLY;
-	inv_t.src.ip = orig_t->src.ip;
-	inv_t.dst.ip = orig_t->dst.ip;
-	inv_t.src.u.gre.key = nat_pptp_info->pns_call_id;
-	inv_t.dst.u.gre.key = ct_pptp_info->pac_call_id;
-
-	if (!ip_conntrack_expect_related(expect_reply)) {
-		DEBUGP("successfully registered expect\n");
-	} else {
-		DEBUGP("can't expect_related(expect_reply)\n");
-		ip_conntrack_unexpect_related(expect_orig);
-		return 1;
-	}
-
-	if (ip_ct_gre_keymap_add(ct, &expect_reply->tuple, 0) < 0) {
-		DEBUGP("can't register original keymap\n");
-		ip_conntrack_unexpect_related(expect_orig);
-		ip_conntrack_unexpect_related(expect_reply);
-		return 1;
-	}
-
-	if (ip_ct_gre_keymap_add(ct, &inv_t, 1) < 0) {
-		DEBUGP("can't register reply keymap\n");
-		ip_conntrack_unexpect_related(expect_orig);
-		ip_conntrack_unexpect_related(expect_reply);
-		ip_ct_gre_keymap_destroy(ct);
-		return 1;
-	}
-
-	return 0;
 }
 
 /* inbound packets == from PAC to PNS */
-- 
cgit v1.2.3


From 4c5de695cf7f71c85ad8cfff509f6475b8bd4d27 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 20 Sep 2006 12:11:30 -0700
Subject: [NETFILTER]: PPTP conntrack: fix another GRE keymap leak

When the master PPTP connection times out while still having unfullfilled
expectations (and a GRE keymap entry) associated with it, the keymap entry
is not destroyed.

Add a destroy callback to struct ip_conntrack_helper and use it to destroy
PPTP siblings when the master is destroyed.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_helper.h |  2 ++
 net/ipv4/netfilter/ip_conntrack_core.c             |  5 +++++
 net/ipv4/netfilter/ip_conntrack_helper_pptp.c      | 12 ++----------
 3 files changed, 9 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper.h b/include/linux/netfilter_ipv4/ip_conntrack_helper.h
index 8d69279ccfe4..77fe868d36ff 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_helper.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_helper.h
@@ -25,6 +25,8 @@ struct ip_conntrack_helper
 		    struct ip_conntrack *ct,
 		    enum ip_conntrack_info conntrackinfo);
 
+	void (*destroy)(struct ip_conntrack *ct);
+
 	int (*to_nfattr)(struct sk_buff *skb, const struct ip_conntrack *ct);
 };
 
diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c
index 2b6f24fc727e..c432b3163609 100644
--- a/net/ipv4/netfilter/ip_conntrack_core.c
+++ b/net/ipv4/netfilter/ip_conntrack_core.c
@@ -307,6 +307,7 @@ destroy_conntrack(struct nf_conntrack *nfct)
 {
 	struct ip_conntrack *ct = (struct ip_conntrack *)nfct;
 	struct ip_conntrack_protocol *proto;
+	struct ip_conntrack_helper *helper;
 
 	DEBUGP("destroy_conntrack(%p)\n", ct);
 	IP_NF_ASSERT(atomic_read(&nfct->use) == 0);
@@ -315,6 +316,10 @@ destroy_conntrack(struct nf_conntrack *nfct)
 	ip_conntrack_event(IPCT_DESTROY, ct);
 	set_bit(IPS_DYING_BIT, &ct->status);
 
+	helper = ct->helper;
+	if (helper && helper->destroy)
+		helper->destroy(ct);
+
 	/* To make sure we don't get any weird locking issues here:
 	 * destroy_conntrack() MUST NOT be called with a write lock
 	 * to ip_conntrack_lock!!! -HW */
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
index 98267b0d2a47..fb0aee691721 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
@@ -553,15 +553,6 @@ conntrack_pptp_help(struct sk_buff **pskb,
 	nexthdr_off += tcph->doff * 4;
  	datalen = tcplen - tcph->doff * 4;
 
-	if (tcph->fin || tcph->rst) {
-		DEBUGP("RST/FIN received, timeouting GRE\n");
-		/* can't do this after real newnat */
-		info->cstate = PPTP_CALL_NONE;
-
-		/* untrack this call id, unexpect GRE packets */
-		pptp_destroy_siblings(ct);
-	}
-
 	pptph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_pptph), &_pptph);
 	if (!pptph) {
 		DEBUGP("no full PPTP header, can't track\n");
@@ -640,7 +631,8 @@ static struct ip_conntrack_helper pptp = {
 			   .protonum = 0xff
 		 	 }
 		},
-	.help = conntrack_pptp_help
+	.help = conntrack_pptp_help,
+	.destroy = pptp_destroy_siblings,
 };
 
 extern void ip_ct_proto_gre_fini(void);
-- 
cgit v1.2.3


From 62dd93181aaa1d5a501a9cebcb254f44b8a48af7 Mon Sep 17 00:00:00 2001
From: Ville Nuorvala <vnuorval@tcs.hut.fi>
Date: Fri, 22 Sep 2006 14:43:19 -0700
Subject: [IPV6] NDISC: Set per-entry is_router flag in Proxy NA.

We have sent NA with router flag from the node-wide forwarding
configuration.  This is not appropriate for proxy NA, and it should be
set according to each proxy entry's configuration.

This is used by Mobile IPv6 home agent to support physical home link
in acting as a proxy router for mobile node which is not a router,
for example.

Based on MIPL2 kernel patch.

Signed-off-by: Ville Nuorvala <vnuorval@tcs.hut.fi>
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/net/neighbour.h |  1 +
 net/core/neighbour.c    | 11 ++++++++---
 net/ipv6/ndisc.c        | 14 +++++++++++---
 3 files changed, 20 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/neighbour.h b/include/net/neighbour.h
index bd187daffdb9..c8aacbd2e333 100644
--- a/include/net/neighbour.h
+++ b/include/net/neighbour.h
@@ -126,6 +126,7 @@ struct pneigh_entry
 {
 	struct pneigh_entry	*next;
 	struct net_device		*dev;
+	u8			flags;
 	u8			key[0];
 };
 
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index a45bd2124d6b..b6c69e1463e8 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -1544,9 +1544,14 @@ int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 		lladdr = tb[NDA_LLADDR] ? nla_data(tb[NDA_LLADDR]) : NULL;
 
 		if (ndm->ndm_flags & NTF_PROXY) {
-			err = 0;
-			if (pneigh_lookup(tbl, dst, dev, 1) == NULL)
-				err = -ENOBUFS;
+			struct pneigh_entry *pn;
+
+			err = -ENOBUFS;
+			pn = pneigh_lookup(tbl, dst, dev, 1);
+			if (pn) {
+				pn->flags = ndm->ndm_flags;
+				err = 0;
+			}
 			goto out_dev_put;
 		}
 
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 0e0d6ce69021..ddf038636f01 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -736,8 +736,10 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 	struct inet6_ifaddr *ifp;
 	struct inet6_dev *idev = NULL;
 	struct neighbour *neigh;
+	struct pneigh_entry *pneigh = NULL;
 	int dad = ipv6_addr_any(saddr);
 	int inc;
+	int is_router;
 
 	if (ipv6_addr_is_multicast(&msg->target)) {
 		ND_PRINTK2(KERN_WARNING 
@@ -822,7 +824,8 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 
 		if (ipv6_chk_acast_addr(dev, &msg->target) ||
 		    (idev->cnf.forwarding && 
-		     pneigh_lookup(&nd_tbl, &msg->target, dev, 0))) {
+		     (pneigh = pneigh_lookup(&nd_tbl,
+					     &msg->target, dev, 0)) != NULL)) {
 			if (!(NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED) &&
 			    skb->pkt_type != PACKET_HOST &&
 			    inc != 0 &&
@@ -843,12 +846,17 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 			goto out;
 	}
 
+	if (pneigh)
+		is_router = pneigh->flags & NTF_ROUTER;
+	else
+		is_router = idev->cnf.forwarding;
+
 	if (dad) {
 		struct in6_addr maddr;
 
 		ipv6_addr_all_nodes(&maddr);
 		ndisc_send_na(dev, NULL, &maddr, &msg->target,
-			      idev->cnf.forwarding, 0, (ifp != NULL), 1);
+			      is_router, 0, (ifp != NULL), 1);
 		goto out;
 	}
 
@@ -869,7 +877,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 			     NEIGH_UPDATE_F_OVERRIDE);
 	if (neigh || !dev->hard_header) {
 		ndisc_send_na(dev, neigh, saddr, &msg->target,
-			      idev->cnf.forwarding, 
+			      is_router,
 			      1, (ifp != NULL && inc), inc);
 		if (neigh)
 			neigh_release(neigh);
-- 
cgit v1.2.3


From fbea49e1e2404baa2d88ab47e2db89e49551b53b Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Fri, 22 Sep 2006 14:43:49 -0700
Subject: [IPV6] NDISC: Add proxy_ndp sysctl.

We do not always need proxy NDP functionality even we
enable forwarding.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ip-sysctl.txt |  3 +++
 include/linux/ipv6.h                   |  2 ++
 include/linux/sysctl.h                 |  1 +
 net/ipv6/addrconf.c                    | 11 +++++++++++
 net/ipv6/ip6_output.c                  |  4 +++-
 net/ipv6/ndisc.c                       |  8 +++++++-
 6 files changed, 27 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 307cd4ec8edd..935e298f674a 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -765,6 +765,9 @@ conf/all/forwarding - BOOLEAN
 
 	This referred to as global forwarding.
 
+proxy_ndp - BOOLEAN
+	Do proxy ndp.
+
 conf/interface/*:
 	Change special settings per interface.
 
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 1d6d3ccc9413..caca57df0d7d 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -176,6 +176,7 @@ struct ipv6_devconf {
 	__s32		accept_ra_rt_info_max_plen;
 #endif
 #endif
+	__s32		proxy_ndp;
 	void		*sysctl;
 };
 
@@ -203,6 +204,7 @@ enum {
 	DEVCONF_ACCEPT_RA_RTR_PREF,
 	DEVCONF_RTR_PROBE_INTERVAL,
 	DEVCONF_ACCEPT_RA_RT_INFO_MAX_PLEN,
+	DEVCONF_PROXY_NDP,
 	DEVCONF_MAX
 };
 
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index af61d9235409..736ed917a4f8 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -556,6 +556,7 @@ enum {
 	NET_IPV6_ACCEPT_RA_RTR_PREF=20,
 	NET_IPV6_RTR_PROBE_INTERVAL=21,
 	NET_IPV6_ACCEPT_RA_RT_INFO_MAX_PLEN=22,
+	NET_IPV6_PROXY_NDP=23,
 	__NET_IPV6_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 1e5a296d0a82..825a291d5aa5 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -175,6 +175,7 @@ struct ipv6_devconf ipv6_devconf __read_mostly = {
 	.accept_ra_rt_info_max_plen = 0,
 #endif
 #endif
+	.proxy_ndp		= 0,
 };
 
 static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
@@ -205,6 +206,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	.accept_ra_rt_info_max_plen = 0,
 #endif
 #endif
+	.proxy_ndp		= 0,
 };
 
 /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */
@@ -3337,6 +3339,7 @@ static void inline ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_ACCEPT_RA_RT_INFO_MAX_PLEN] = cnf->accept_ra_rt_info_max_plen;
 #endif
 #endif
+	array[DEVCONF_PROXY_NDP] = cnf->proxy_ndp;
 }
 
 /* Maximum length of ifinfomsg attributes */
@@ -3859,6 +3862,14 @@ static struct addrconf_sysctl_table
 		},
 #endif
 #endif
+		{
+			.ctl_name	=	NET_IPV6_PROXY_NDP,
+			.procname	=	"proxy_ndp",
+			.data		=	&ipv6_devconf.proxy_ndp,
+			.maxlen		=	sizeof(int),
+			.mode		=	0644,
+			.proc_handler	=	&proc_dointvec,
+		},
 		{
 			.ctl_name	=	0,	/* sentinel */
 		}
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index b2be749d2217..66716911962e 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -412,7 +412,9 @@ int ip6_forward(struct sk_buff *skb)
 		return -ETIMEDOUT;
 	}
 
-	if (pneigh_lookup(&nd_tbl, &hdr->daddr, skb->dev, 0)) {
+	/* XXX: idev->cnf.proxy_ndp? */
+	if (ipv6_devconf.proxy_ndp &&
+	    pneigh_lookup(&nd_tbl, &hdr->daddr, skb->dev, 0)) {
 		int proxied = ip6_forward_proxy_check(skb);
 		if (proxied > 0)
 			return ip6_input(skb);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index ddf038636f01..76517a5f6576 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -824,6 +824,7 @@ static void ndisc_recv_ns(struct sk_buff *skb)
 
 		if (ipv6_chk_acast_addr(dev, &msg->target) ||
 		    (idev->cnf.forwarding && 
+		     (ipv6_devconf.proxy_ndp || idev->cnf.proxy_ndp) &&
 		     (pneigh = pneigh_lookup(&nd_tbl,
 					     &msg->target, dev, 0)) != NULL)) {
 			if (!(NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED) &&
@@ -966,8 +967,13 @@ static void ndisc_recv_na(struct sk_buff *skb)
 		 * has already sent a NA to us.
 		 */
 		if (lladdr && !memcmp(lladdr, dev->dev_addr, dev->addr_len) &&
-		    pneigh_lookup(&nd_tbl, &msg->target, dev, 0))
+		    ipv6_devconf.forwarding && ipv6_devconf.proxy_ndp &&
+		    pneigh_lookup(&nd_tbl, &msg->target, dev, 0)) {
+			/* XXX: idev->cnf.prixy_ndp */
+			WARN_ON(skb->dst != NULL &&
+				((struct rt6_info *)skb->dst)->rt6i_idev);
 			goto out;
+		}
 
 		neigh_update(neigh, lladdr,
 			     msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE,
-- 
cgit v1.2.3


From 8814c4b533817df825485ff32ce6ac406c3a54d1 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Fri, 22 Sep 2006 14:44:24 -0700
Subject: [IPV6] ADDRCONF: Convert addrconf_lock to RCU.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/addrconf.h | 10 ++++------
 include/net/if_inet6.h |  1 +
 net/core/pktgen.c      |  4 ++--
 net/ipv6/addrconf.c    | 46 ++++++++++++++++++++++++----------------------
 net/ipv6/anycast.c     |  4 ++--
 net/ipv6/ipv6_syms.c   |  1 -
 net/sctp/ipv6.c        |  6 +++---
 7 files changed, 36 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 5fc8627435eb..aa2ed8f0a9dd 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -133,20 +133,18 @@ extern int unregister_inet6addr_notifier(struct notifier_block *nb);
 static inline struct inet6_dev *
 __in6_dev_get(struct net_device *dev)
 {
-	return (struct inet6_dev *)dev->ip6_ptr;
+	return rcu_dereference(dev->ip6_ptr);
 }
 
-extern rwlock_t addrconf_lock;
-
 static inline struct inet6_dev *
 in6_dev_get(struct net_device *dev)
 {
 	struct inet6_dev *idev = NULL;
-	read_lock(&addrconf_lock);
-	idev = dev->ip6_ptr;
+	rcu_read_lock();
+	idev = __in6_dev_get(dev);
 	if (idev)
 		atomic_inc(&idev->refcnt);
-	read_unlock(&addrconf_lock);
+	rcu_read_unlock();
 	return idev;
 }
 
diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h
index e459e1a0ae4a..34489c13c119 100644
--- a/include/net/if_inet6.h
+++ b/include/net/if_inet6.h
@@ -189,6 +189,7 @@ struct inet6_dev
 	struct ipv6_devconf	cnf;
 	struct ipv6_devstat	stats;
 	unsigned long		tstamp; /* ipv6InterfaceTable update timestamp */
+	struct rcu_head		rcu;
 };
 
 extern struct ipv6_devconf ipv6_devconf;
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 6a7320b39ed0..72145d4a2600 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -1786,7 +1786,7 @@ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev)
 			 * use ipv6_get_lladdr if/when it's get exported
 			 */
 
-			read_lock(&addrconf_lock);
+			rcu_read_lock();
 			if ((idev = __in6_dev_get(pkt_dev->odev)) != NULL) {
 				struct inet6_ifaddr *ifp;
 
@@ -1805,7 +1805,7 @@ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev)
 				}
 				read_unlock_bh(&idev->lock);
 			}
-			read_unlock(&addrconf_lock);
+			rcu_read_unlock();
 			if (err)
 				printk("pktgen: ERROR: IPv6 link address not availble.\n");
 		}
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 825a291d5aa5..c09ebb7bb98a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -119,9 +119,6 @@ static int ipv6_count_addresses(struct inet6_dev *idev);
 static struct inet6_ifaddr		*inet6_addr_lst[IN6_ADDR_HSIZE];
 static DEFINE_RWLOCK(addrconf_hash_lock);
 
-/* Protects inet6 devices */
-DEFINE_RWLOCK(addrconf_lock);
-
 static void addrconf_verify(unsigned long);
 
 static DEFINE_TIMER(addr_chk_timer, addrconf_verify, 0, 0);
@@ -318,6 +315,12 @@ static void addrconf_mod_timer(struct inet6_ifaddr *ifp,
 
 /* Nobody refers to this device, we may destroy it. */
 
+static void in6_dev_finish_destroy_rcu(struct rcu_head *head)
+{
+	struct inet6_dev *idev = container_of(head, struct inet6_dev, rcu);
+	kfree(idev);
+}
+
 void in6_dev_finish_destroy(struct inet6_dev *idev)
 {
 	struct net_device *dev = idev->dev;
@@ -332,7 +335,7 @@ void in6_dev_finish_destroy(struct inet6_dev *idev)
 		return;
 	}
 	snmp6_free_dev(idev);
-	kfree(idev);
+	call_rcu(&idev->rcu, in6_dev_finish_destroy_rcu);
 }
 
 static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
@@ -408,9 +411,8 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 	if (netif_carrier_ok(dev))
 		ndev->if_flags |= IF_READY;
 
-	write_lock_bh(&addrconf_lock);
-	dev->ip6_ptr = ndev;
-	write_unlock_bh(&addrconf_lock);
+	/* protected by rtnl_lock */
+	rcu_assign_pointer(dev->ip6_ptr, ndev);
 
 	ipv6_mc_init_dev(ndev);
 	ndev->tstamp = jiffies;
@@ -474,7 +476,7 @@ static void addrconf_forward_change(void)
 
 	read_lock(&dev_base_lock);
 	for (dev=dev_base; dev; dev=dev->next) {
-		read_lock(&addrconf_lock);
+		rcu_read_lock();
 		idev = __in6_dev_get(dev);
 		if (idev) {
 			int changed = (!idev->cnf.forwarding) ^ (!ipv6_devconf.forwarding);
@@ -482,7 +484,7 @@ static void addrconf_forward_change(void)
 			if (changed)
 				dev_forward_change(idev);
 		}
-		read_unlock(&addrconf_lock);
+		rcu_read_unlock();
 	}
 	read_unlock(&dev_base_lock);
 }
@@ -543,7 +545,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen,
 	int hash;
 	int err = 0;
 
-	read_lock_bh(&addrconf_lock);
+	rcu_read_lock_bh();
 	if (idev->dead) {
 		err = -ENODEV;			/*XXX*/
 		goto out2;
@@ -612,7 +614,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen,
 	in6_ifa_hold(ifa);
 	write_unlock(&idev->lock);
 out2:
-	read_unlock_bh(&addrconf_lock);
+	rcu_read_unlock_bh();
 
 	if (likely(err == 0))
 		atomic_notifier_call_chain(&inet6addr_chain, NETDEV_UP, ifa);
@@ -915,7 +917,7 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev,
 	memset(&hiscore, 0, sizeof(hiscore));
 
 	read_lock(&dev_base_lock);
-	read_lock(&addrconf_lock);
+	rcu_read_lock();
 
 	for (dev = dev_base; dev; dev=dev->next) {
 		struct inet6_dev *idev;
@@ -1127,7 +1129,7 @@ record_it:
 		}
 		read_unlock_bh(&idev->lock);
 	}
-	read_unlock(&addrconf_lock);
+	rcu_read_unlock();
 	read_unlock(&dev_base_lock);
 
 	if (!ifa_result)
@@ -1151,7 +1153,7 @@ int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr)
 	struct inet6_dev *idev;
 	int err = -EADDRNOTAVAIL;
 
-	read_lock(&addrconf_lock);
+	rcu_read_lock();
 	if ((idev = __in6_dev_get(dev)) != NULL) {
 		struct inet6_ifaddr *ifp;
 
@@ -1165,7 +1167,7 @@ int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr)
 		}
 		read_unlock_bh(&idev->lock);
 	}
-	read_unlock(&addrconf_lock);
+	rcu_read_unlock();
 	return err;
 }
 
@@ -1466,7 +1468,7 @@ static void ipv6_regen_rndid(unsigned long data)
 	struct inet6_dev *idev = (struct inet6_dev *) data;
 	unsigned long expires;
 
-	read_lock_bh(&addrconf_lock);
+	rcu_read_lock_bh();
 	write_lock_bh(&idev->lock);
 
 	if (idev->dead)
@@ -1490,7 +1492,7 @@ static void ipv6_regen_rndid(unsigned long data)
 
 out:
 	write_unlock_bh(&idev->lock);
-	read_unlock_bh(&addrconf_lock);
+	rcu_read_unlock_bh();
 	in6_dev_put(idev);
 }
 
@@ -2342,10 +2344,10 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 	           Do not dev_put!
 	 */
 	if (how == 1) {
-		write_lock_bh(&addrconf_lock);
-		dev->ip6_ptr = NULL;
 		idev->dead = 1;
-		write_unlock_bh(&addrconf_lock);
+
+		/* protected by rtnl_lock */
+		rcu_assign_pointer(dev->ip6_ptr, NULL);
 
 		/* Step 1.5: remove snmp6 entry */
 		snmp6_unregister_dev(idev);
@@ -3573,10 +3575,10 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 
 static void ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 {
-	read_lock_bh(&addrconf_lock);
+	rcu_read_lock_bh();
 	if (likely(ifp->idev->dead == 0))
 		__ipv6_ifa_notify(event, ifp);
-	read_unlock_bh(&addrconf_lock);
+	rcu_read_unlock_bh();
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index b80fc502ca03..a9604764e015 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -56,7 +56,7 @@ ip6_onlink(struct in6_addr *addr, struct net_device *dev)
 	int	onlink;
 
 	onlink = 0;
-	read_lock(&addrconf_lock);
+	rcu_read_lock();
 	idev = __in6_dev_get(dev);
 	if (idev) {
 		read_lock_bh(&idev->lock);
@@ -68,7 +68,7 @@ ip6_onlink(struct in6_addr *addr, struct net_device *dev)
 		}
 		read_unlock_bh(&idev->lock);
 	}
-	read_unlock(&addrconf_lock);
+	rcu_read_unlock();
 	return onlink;
 }
 
diff --git a/net/ipv6/ipv6_syms.c b/net/ipv6/ipv6_syms.c
index 7b7b90d9c3d0..0e8e0676a033 100644
--- a/net/ipv6/ipv6_syms.c
+++ b/net/ipv6/ipv6_syms.c
@@ -14,7 +14,6 @@ EXPORT_SYMBOL(ndisc_mc_map);
 EXPORT_SYMBOL(register_inet6addr_notifier);
 EXPORT_SYMBOL(unregister_inet6addr_notifier);
 EXPORT_SYMBOL(ip6_route_output);
-EXPORT_SYMBOL(addrconf_lock);
 EXPORT_SYMBOL(ipv6_setsockopt);
 EXPORT_SYMBOL(ipv6_getsockopt);
 EXPORT_SYMBOL(inet6_register_protosw);
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index fd87e3ceb56e..249e5033c1a8 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -321,9 +321,9 @@ static void sctp_v6_copy_addrlist(struct list_head *addrlist,
 	struct inet6_ifaddr *ifp;
 	struct sctp_sockaddr_entry *addr;
 
-	read_lock(&addrconf_lock);
+	rcu_read_lock();
 	if ((in6_dev = __in6_dev_get(dev)) == NULL) {
-		read_unlock(&addrconf_lock);
+		rcu_read_unlock();
 		return;
 	}
 
@@ -342,7 +342,7 @@ static void sctp_v6_copy_addrlist(struct list_head *addrlist,
 	}
 
 	read_unlock(&in6_dev->lock);
-	read_unlock(&addrconf_lock);
+	rcu_read_unlock();
 }
 
 /* Initialize a sockaddr_storage from in incoming skb. */
-- 
cgit v1.2.3


From 55ebaef1d5db9c1c76ba01a87fd986db5dee550d Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Fri, 22 Sep 2006 14:45:27 -0700
Subject: [IPV6] ADDRCONF: Allow non-DAD'able addresses.

IFA_F_NODAD flag, similar to IN6_IFF_NODAD in BSDs, is introduced
to skip DAD.

This flag should be set to Mobile IPv6 Home Address(es) on Mobile
Node because DAD would fail if we should perform DAD; our Home Agent
protects our Home Address(es).

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_addr.h |  1 +
 net/ipv6/addrconf.c     | 31 ++++++++++++++++---------------
 2 files changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_addr.h b/include/linux/if_addr.h
index e1590454db59..ca24b9de13fb 100644
--- a/include/linux/if_addr.h
+++ b/include/linux/if_addr.h
@@ -38,6 +38,7 @@ enum
 #define IFA_F_SECONDARY		0x01
 #define IFA_F_TEMPORARY		IFA_F_SECONDARY
 
+#define	IFA_F_NODAD		0x02
 #define IFA_F_DEPRECATED	0x20
 #define IFA_F_TENTATIVE		0x40
 #define IFA_F_PERMANENT		0x80
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index c09ebb7bb98a..adb583a26151 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1873,12 +1873,11 @@ err_exit:
  *	Manual configuration of address on an interface
  */
 static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen,
-			  __u32 prefered_lft, __u32 valid_lft)
+			  __u8 ifa_flags, __u32 prefered_lft, __u32 valid_lft)
 {
 	struct inet6_ifaddr *ifp;
 	struct inet6_dev *idev;
 	struct net_device *dev;
-	__u8 ifa_flags = 0;
 	int scope;
 
 	ASSERT_RTNL();
@@ -1971,7 +1970,7 @@ int addrconf_add_ifaddr(void __user *arg)
 
 	rtnl_lock();
 	err = inet6_addr_add(ireq.ifr6_ifindex, &ireq.ifr6_addr, ireq.ifr6_prefixlen,
-			     INFINITY_LIFE_TIME, INFINITY_LIFE_TIME);
+			     IFA_F_PERMANENT, INFINITY_LIFE_TIME, INFINITY_LIFE_TIME);
 	rtnl_unlock();
 	return err;
 }
@@ -2514,7 +2513,8 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags)
 	spin_lock_bh(&ifp->lock);
 
 	if (dev->flags&(IFF_NOARP|IFF_LOOPBACK) ||
-	    !(ifp->flags&IFA_F_TENTATIVE)) {
+	    !(ifp->flags&IFA_F_TENTATIVE) ||
+	    ifp->flags & IFA_F_NODAD) {
 		ifp->flags &= ~IFA_F_TENTATIVE;
 		spin_unlock_bh(&ifp->lock);
 		read_unlock_bh(&idev->lock);
@@ -2912,28 +2912,25 @@ inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	return inet6_addr_del(ifm->ifa_index, pfx, ifm->ifa_prefixlen);
 }
 
-static int inet6_addr_modify(struct inet6_ifaddr *ifp, u32 prefered_lft,
-			     u32 valid_lft)
+static int inet6_addr_modify(struct inet6_ifaddr *ifp, u8 ifa_flags,
+			     u32 prefered_lft, u32 valid_lft)
 {
-	int ifa_flags = 0;
-
 	if (!valid_lft || (prefered_lft > valid_lft))
 		return -EINVAL;
 
 	if (valid_lft == INFINITY_LIFE_TIME)
-		ifa_flags = IFA_F_PERMANENT;
+		ifa_flags |= IFA_F_PERMANENT;
 	else if (valid_lft >= 0x7FFFFFFF/HZ)
 		valid_lft = 0x7FFFFFFF/HZ;
 
 	if (prefered_lft == 0)
-		ifa_flags = IFA_F_DEPRECATED;
+		ifa_flags |= IFA_F_DEPRECATED;
 	else if ((prefered_lft >= 0x7FFFFFFF/HZ) &&
 		 (prefered_lft != INFINITY_LIFE_TIME))
 		prefered_lft = 0x7FFFFFFF/HZ;
 
 	spin_lock_bh(&ifp->lock);
-	ifp->flags = (ifp->flags & ~(IFA_F_DEPRECATED|IFA_F_PERMANENT)) | ifa_flags;
-
+	ifp->flags = (ifp->flags & ~(IFA_F_DEPRECATED | IFA_F_PERMANENT | IFA_F_NODAD)) | ifa_flags;
 	ifp->tstamp = jiffies;
 	ifp->valid_lft = valid_lft;
 	ifp->prefered_lft = prefered_lft;
@@ -2955,7 +2952,8 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	struct in6_addr *pfx;
 	struct inet6_ifaddr *ifa;
 	struct net_device *dev;
-	u32 valid_lft, preferred_lft;
+	u32 valid_lft = INFINITY_LIFE_TIME, preferred_lft = INFINITY_LIFE_TIME;
+	u8 ifa_flags;
 	int err;
 
 	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFA_MAX, ifa_ipv6_policy);
@@ -2982,6 +2980,9 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 	if (dev == NULL)
 		return -ENODEV;
 
+	/* We ignore other flags so far. */
+	ifa_flags = ifm->ifa_flags & IFA_F_NODAD;
+
 	ifa = ipv6_get_ifaddr(pfx, dev, 1);
 	if (ifa == NULL) {
 		/*
@@ -2989,14 +2990,14 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 		 * userspace alreay relies on not having to provide this.
 		 */
 		return inet6_addr_add(ifm->ifa_index, pfx, ifm->ifa_prefixlen,
-				      preferred_lft, valid_lft);
+				      ifa_flags, preferred_lft, valid_lft);
 	}
 
 	if (nlh->nlmsg_flags & NLM_F_EXCL ||
 	    !(nlh->nlmsg_flags & NLM_F_REPLACE))
 		err = -EEXIST;
 	else
-		err = inet6_addr_modify(ifa, preferred_lft, valid_lft);
+		err = inet6_addr_modify(ifa, ifa_flags, preferred_lft, valid_lft);
 
 	in6_ifa_put(ifa);
 
-- 
cgit v1.2.3


From 3b9f9a1c3903b64c38505f9fed3bb11e48dbc931 Mon Sep 17 00:00:00 2001
From: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Date: Fri, 22 Sep 2006 14:45:56 -0700
Subject: [IPV6] ADDRCONF: Mobile IPv6 Home Address support.

IFA_F_HOMEADDRESS is introduced for Mobile IPv6 Home Addresses on
Mobile Node.

The IFA_F_HOMEADDRESS flag should be set for Mobile IPv6 Home
Addresses for 2 purposes. 1) We need to check this on receipt of
Type 2 Routing Header (RFC3775 Secion 6.4), 2) We prefer Home
Address(es) in source address selection (RFC3484 Section 5 Rule 4).

Signed-off-by: Noriaki TAKAMIYA <takamiya@po.ntts.co.jp>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_addr.h |  1 +
 include/net/addrconf.h  |  6 +-----
 net/ipv6/addrconf.c     | 44 +++++++++++++++++++++++++++++++++++++++++---
 3 files changed, 43 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_addr.h b/include/linux/if_addr.h
index ca24b9de13fb..dbe8f6120a40 100644
--- a/include/linux/if_addr.h
+++ b/include/linux/if_addr.h
@@ -39,6 +39,7 @@ enum
 #define IFA_F_TEMPORARY		IFA_F_SECONDARY
 
 #define	IFA_F_NODAD		0x02
+#define	IFA_F_HOMEADDRESS	0x10
 #define IFA_F_DEPRECATED	0x20
 #define IFA_F_TENTATIVE		0x40
 #define IFA_F_PERMANENT		0x80
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index aa2ed8f0a9dd..44f1b673f916 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -61,12 +61,8 @@ extern int			addrconf_set_dstaddr(void __user *arg);
 extern int			ipv6_chk_addr(struct in6_addr *addr,
 					      struct net_device *dev,
 					      int strict);
-/* XXX: this is a placeholder till addrconf supports */
 #ifdef CONFIG_IPV6_MIP6
-static inline int ipv6_chk_home_addr(struct in6_addr *addr)
-{
-	return 0;
-}
+extern int			ipv6_chk_home_addr(struct in6_addr *addr);
 #endif
 extern struct inet6_ifaddr *	ipv6_get_ifaddr(struct in6_addr *addr,
 						struct net_device *dev,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index adb583a26151..c18676352397 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1038,9 +1038,27 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev,
 					continue;
 			}
 
-			/* Rule 4: Prefer home address -- not implemented yet */
+			/* Rule 4: Prefer home address */
+#ifdef CONFIG_IPV6_MIP6
+			if (hiscore.rule < 4) {
+				if (ifa_result->flags & IFA_F_HOMEADDRESS)
+					hiscore.attrs |= IPV6_SADDR_SCORE_HOA;
+				hiscore.rule++;
+			}
+			if (ifa->flags & IFA_F_HOMEADDRESS) {
+				score.attrs |= IPV6_SADDR_SCORE_HOA;
+				if (!(ifa_result->flags & IFA_F_HOMEADDRESS)) {
+					score.rule = 4;
+					goto record_it;
+				}
+			} else {
+				if (hiscore.attrs & IPV6_SADDR_SCORE_HOA)
+					continue;
+			}
+#else
 			if (hiscore.rule < 4)
 				hiscore.rule++;
+#endif
 
 			/* Rule 5: Prefer outgoing interface */
 			if (hiscore.rule < 5) {
@@ -2759,6 +2777,26 @@ void if6_proc_exit(void)
 }
 #endif	/* CONFIG_PROC_FS */
 
+#ifdef CONFIG_IPV6_MIP6
+/* Check if address is a home address configured on any interface. */
+int ipv6_chk_home_addr(struct in6_addr *addr)
+{
+	int ret = 0;
+	struct inet6_ifaddr * ifp;
+	u8 hash = ipv6_addr_hash(addr);
+	read_lock_bh(&addrconf_hash_lock);
+	for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) {
+		if (ipv6_addr_cmp(&ifp->addr, addr) == 0 &&
+		    (ifp->flags & IFA_F_HOMEADDRESS)) {
+			ret = 1;
+			break;
+		}
+	}
+	read_unlock_bh(&addrconf_hash_lock);
+	return ret;
+}
+#endif
+
 /*
  *	Periodic address status verification
  */
@@ -2930,7 +2968,7 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, u8 ifa_flags,
 		prefered_lft = 0x7FFFFFFF/HZ;
 
 	spin_lock_bh(&ifp->lock);
-	ifp->flags = (ifp->flags & ~(IFA_F_DEPRECATED | IFA_F_PERMANENT | IFA_F_NODAD)) | ifa_flags;
+	ifp->flags = (ifp->flags & ~(IFA_F_DEPRECATED | IFA_F_PERMANENT | IFA_F_NODAD | IFA_F_HOMEADDRESS)) | ifa_flags;
 	ifp->tstamp = jiffies;
 	ifp->valid_lft = valid_lft;
 	ifp->prefered_lft = prefered_lft;
@@ -2981,7 +3019,7 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 		return -ENODEV;
 
 	/* We ignore other flags so far. */
-	ifa_flags = ifm->ifa_flags & IFA_F_NODAD;
+	ifa_flags = ifm->ifa_flags & (IFA_F_NODAD | IFA_F_HOMEADDRESS);
 
 	ifa = ipv6_get_ifaddr(pfx, dev, 1);
 	if (ifa == NULL) {
-- 
cgit v1.2.3


From 64f817ba98095156149ba5991592d5d039f6da74 Mon Sep 17 00:00:00 2001
From: Ralph Campbell <ralph.campbell@qlogic.com>
Date: Fri, 22 Sep 2006 15:22:24 -0700
Subject: IB/uverbs: Allow resize CQ operation to return driver-specific data

Add a ib_uverbs_resize_cq_resp.driver_data field so that low-level
drivers can return data from a resize CQ operation to userspace.  Have
ib_uverbs_resize_cq() only copy the cqe field, to avoid having to bump
the userspace ABI.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/uverbs_cmd.c | 3 +--
 include/rdma/ib_user_verbs.h         | 2 ++
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 8b6df7cec0bf..deae43f31e79 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -879,11 +879,10 @@ ssize_t ib_uverbs_resize_cq(struct ib_uverbs_file *file,
 	if (ret)
 		goto out;
 
-	memset(&resp, 0, sizeof resp);
 	resp.cqe = cq->cqe;
 
 	if (copy_to_user((void __user *) (unsigned long) cmd.response,
-			 &resp, sizeof resp))
+			 &resp, sizeof resp.cqe))
 		ret = -EFAULT;
 
 out:
diff --git a/include/rdma/ib_user_verbs.h b/include/rdma/ib_user_verbs.h
index 7b5372010f4b..db1b814b62cc 100644
--- a/include/rdma/ib_user_verbs.h
+++ b/include/rdma/ib_user_verbs.h
@@ -275,6 +275,8 @@ struct ib_uverbs_resize_cq {
 
 struct ib_uverbs_resize_cq_resp {
 	__u32 cqe;
+	__u32 reserved;
+	__u64 driver_data[0];
 };
 
 struct ib_uverbs_poll_cq {
-- 
cgit v1.2.3


From 9bc57e2d19db4da81c1150120658cc3658a99ed4 Mon Sep 17 00:00:00 2001
From: Ralph Campbell <ralphc@pathscale.com>
Date: Fri, 11 Aug 2006 14:58:09 -0700
Subject: IB/uverbs: Pass userspace data to modify_srq and modify_qp methods

Pass a struct ib_udata to the low-level driver's ->modify_srq() and
->modify_qp() methods, so that it can get to the device-specific data
passed in by the userspace driver.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/uverbs_cmd.c      | 13 ++++++++++---
 drivers/infiniband/core/verbs.c           |  4 ++--
 drivers/infiniband/hw/ehca/ehca_iverbs.h  |  3 ++-
 drivers/infiniband/hw/ehca/ehca_qp.c      |  3 ++-
 drivers/infiniband/hw/ipath/ipath_qp.c    |  3 ++-
 drivers/infiniband/hw/ipath/ipath_srq.c   |  4 +++-
 drivers/infiniband/hw/ipath/ipath_verbs.h |  5 +++--
 drivers/infiniband/hw/mthca/mthca_dev.h   |  5 +++--
 drivers/infiniband/hw/mthca/mthca_qp.c    |  3 ++-
 drivers/infiniband/hw/mthca/mthca_srq.c   |  2 +-
 include/rdma/ib_verbs.h                   |  6 ++++--
 11 files changed, 34 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index deae43f31e79..3fcb5d189a23 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -841,7 +841,6 @@ ssize_t ib_uverbs_create_cq(struct ib_uverbs_file *file,
 err_copy:
 	idr_remove_uobj(&ib_uverbs_cq_idr, &obj->uobject);
 
-
 err_free:
 	ib_destroy_cq(cq);
 
@@ -1273,6 +1272,7 @@ ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file,
 			    int out_len)
 {
 	struct ib_uverbs_modify_qp cmd;
+	struct ib_udata            udata;
 	struct ib_qp              *qp;
 	struct ib_qp_attr         *attr;
 	int                        ret;
@@ -1280,6 +1280,9 @@ ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file,
 	if (copy_from_user(&cmd, buf, sizeof cmd))
 		return -EFAULT;
 
+	INIT_UDATA(&udata, buf + sizeof cmd, NULL, in_len - sizeof cmd,
+		   out_len);
+
 	attr = kmalloc(sizeof *attr, GFP_KERNEL);
 	if (!attr)
 		return -ENOMEM;
@@ -1336,7 +1339,7 @@ ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file,
 	attr->alt_ah_attr.ah_flags 	    = cmd.alt_dest.is_global ? IB_AH_GRH : 0;
 	attr->alt_ah_attr.port_num 	    = cmd.alt_dest.port_num;
 
-	ret = ib_modify_qp(qp, attr, cmd.attr_mask);
+	ret = qp->device->modify_qp(qp, attr, cmd.attr_mask, &udata);
 
 	put_qp_read(qp);
 
@@ -2054,6 +2057,7 @@ ssize_t ib_uverbs_modify_srq(struct ib_uverbs_file *file,
 			     int out_len)
 {
 	struct ib_uverbs_modify_srq cmd;
+	struct ib_udata             udata;
 	struct ib_srq              *srq;
 	struct ib_srq_attr          attr;
 	int                         ret;
@@ -2061,6 +2065,9 @@ ssize_t ib_uverbs_modify_srq(struct ib_uverbs_file *file,
 	if (copy_from_user(&cmd, buf, sizeof cmd))
 		return -EFAULT;
 
+	INIT_UDATA(&udata, buf + sizeof cmd, NULL, in_len - sizeof cmd,
+		   out_len);
+
 	srq = idr_read_srq(cmd.srq_handle, file->ucontext);
 	if (!srq)
 		return -EINVAL;
@@ -2068,7 +2075,7 @@ ssize_t ib_uverbs_modify_srq(struct ib_uverbs_file *file,
 	attr.max_wr    = cmd.max_wr;
 	attr.srq_limit = cmd.srq_limit;
 
-	ret = ib_modify_srq(srq, &attr, cmd.attr_mask);
+	ret = srq->device->modify_srq(srq, &attr, cmd.attr_mask, &udata);
 
 	put_srq_read(srq);
 
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 468999c38803..06f98e9e14f9 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -231,7 +231,7 @@ int ib_modify_srq(struct ib_srq *srq,
 		  struct ib_srq_attr *srq_attr,
 		  enum ib_srq_attr_mask srq_attr_mask)
 {
-	return srq->device->modify_srq(srq, srq_attr, srq_attr_mask);
+	return srq->device->modify_srq(srq, srq_attr, srq_attr_mask, NULL);
 }
 EXPORT_SYMBOL(ib_modify_srq);
 
@@ -547,7 +547,7 @@ int ib_modify_qp(struct ib_qp *qp,
 		 struct ib_qp_attr *qp_attr,
 		 int qp_attr_mask)
 {
-	return qp->device->modify_qp(qp, qp_attr, qp_attr_mask);
+	return qp->device->modify_qp(qp, qp_attr, qp_attr_mask, NULL);
 }
 EXPORT_SYMBOL(ib_modify_qp);
 
diff --git a/drivers/infiniband/hw/ehca/ehca_iverbs.h b/drivers/infiniband/hw/ehca/ehca_iverbs.h
index bbdc437f5167..319c39d47f3a 100644
--- a/drivers/infiniband/hw/ehca/ehca_iverbs.h
+++ b/drivers/infiniband/hw/ehca/ehca_iverbs.h
@@ -143,7 +143,8 @@ struct ib_qp *ehca_create_qp(struct ib_pd *pd,
 
 int ehca_destroy_qp(struct ib_qp *qp);
 
-int ehca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask);
+int ehca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
+		   struct ib_udata *udata);
 
 int ehca_query_qp(struct ib_qp *qp, struct ib_qp_attr *qp_attr,
 		  int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr);
diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c
index 4b27bedc6c24..4394123cdbd7 100644
--- a/drivers/infiniband/hw/ehca/ehca_qp.c
+++ b/drivers/infiniband/hw/ehca/ehca_qp.c
@@ -1230,7 +1230,8 @@ modify_qp_exit1:
 	return ret;
 }
 
-int ehca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
+int ehca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
+		   struct ib_udata *udata)
 {
 	struct ehca_qp *my_qp = container_of(ibqp, struct ehca_qp, ib_qp);
 	struct ehca_pd *my_pd = container_of(my_qp->ib_qp.pd, struct ehca_pd,
diff --git a/drivers/infiniband/hw/ipath/ipath_qp.c b/drivers/infiniband/hw/ipath/ipath_qp.c
index 83e557be591e..44c32d2db990 100644
--- a/drivers/infiniband/hw/ipath/ipath_qp.c
+++ b/drivers/infiniband/hw/ipath/ipath_qp.c
@@ -426,11 +426,12 @@ void ipath_error_qp(struct ipath_qp *qp)
  * @ibqp: the queue pair who's attributes we're modifying
  * @attr: the new attributes
  * @attr_mask: the mask of attributes to modify
+ * @udata: user data for ipathverbs.so
  *
  * Returns 0 on success, otherwise returns an errno.
  */
 int ipath_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
-		    int attr_mask)
+		    int attr_mask, struct ib_udata *udata)
 {
 	struct ipath_ibdev *dev = to_idev(ibqp->device);
 	struct ipath_qp *qp = to_iqp(ibqp);
diff --git a/drivers/infiniband/hw/ipath/ipath_srq.c b/drivers/infiniband/hw/ipath/ipath_srq.c
index f760434660bd..fa77da6667ed 100644
--- a/drivers/infiniband/hw/ipath/ipath_srq.c
+++ b/drivers/infiniband/hw/ipath/ipath_srq.c
@@ -188,9 +188,11 @@ bail:
  * @ibsrq: the SRQ to modify
  * @attr: the new attributes of the SRQ
  * @attr_mask: indicates which attributes to modify
+ * @udata: user data for ipathverbs.so
  */
 int ipath_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
-		     enum ib_srq_attr_mask attr_mask)
+		     enum ib_srq_attr_mask attr_mask,
+		     struct ib_udata *udata)
 {
 	struct ipath_srq *srq = to_isrq(ibsrq);
 	unsigned long flags;
diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.h b/drivers/infiniband/hw/ipath/ipath_verbs.h
index 2df684727dc1..698396778f00 100644
--- a/drivers/infiniband/hw/ipath/ipath_verbs.h
+++ b/drivers/infiniband/hw/ipath/ipath_verbs.h
@@ -579,7 +579,7 @@ struct ib_qp *ipath_create_qp(struct ib_pd *ibpd,
 int ipath_destroy_qp(struct ib_qp *ibqp);
 
 int ipath_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
-		    int attr_mask);
+		    int attr_mask, struct ib_udata *udata);
 
 int ipath_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		   int attr_mask, struct ib_qp_init_attr *init_attr);
@@ -638,7 +638,8 @@ struct ib_srq *ipath_create_srq(struct ib_pd *ibpd,
 				struct ib_udata *udata);
 
 int ipath_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
-		     enum ib_srq_attr_mask attr_mask);
+		     enum ib_srq_attr_mask attr_mask,
+		     struct ib_udata *udata);
 
 int ipath_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr);
 
diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h
index f8160b8de090..33bd0b8bfd13 100644
--- a/drivers/infiniband/hw/mthca/mthca_dev.h
+++ b/drivers/infiniband/hw/mthca/mthca_dev.h
@@ -506,7 +506,7 @@ int mthca_alloc_srq(struct mthca_dev *dev, struct mthca_pd *pd,
 		    struct ib_srq_attr *attr, struct mthca_srq *srq);
 void mthca_free_srq(struct mthca_dev *dev, struct mthca_srq *srq);
 int mthca_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
-		     enum ib_srq_attr_mask attr_mask);
+		     enum ib_srq_attr_mask attr_mask, struct ib_udata *udata);
 int mthca_query_srq(struct ib_srq *srq, struct ib_srq_attr *srq_attr);
 int mthca_max_srq_sge(struct mthca_dev *dev);
 void mthca_srq_event(struct mthca_dev *dev, u32 srqn,
@@ -521,7 +521,8 @@ void mthca_qp_event(struct mthca_dev *dev, u32 qpn,
 		    enum ib_event_type event_type);
 int mthca_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr_mask,
 		   struct ib_qp_init_attr *qp_init_attr);
-int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask);
+int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
+		    struct ib_udata *udata);
 int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 			  struct ib_send_wr **bad_wr);
 int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c
index 2e8f6f36e0a5..6d6ba4180a39 100644
--- a/drivers/infiniband/hw/mthca/mthca_qp.c
+++ b/drivers/infiniband/hw/mthca/mthca_qp.c
@@ -527,7 +527,8 @@ static int mthca_path_set(struct mthca_dev *dev, struct ib_ah_attr *ah,
 	return 0;
 }
 
-int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask)
+int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
+		    struct ib_udata *udata)
 {
 	struct mthca_dev *dev = to_mdev(ibqp->device);
 	struct mthca_qp *qp = to_mqp(ibqp);
diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c
index b60a9d79ae54..0f316c87bf64 100644
--- a/drivers/infiniband/hw/mthca/mthca_srq.c
+++ b/drivers/infiniband/hw/mthca/mthca_srq.c
@@ -358,7 +358,7 @@ void mthca_free_srq(struct mthca_dev *dev, struct mthca_srq *srq)
 }
 
 int mthca_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
-		     enum ib_srq_attr_mask attr_mask)
+		     enum ib_srq_attr_mask attr_mask, struct ib_udata *udata)
 {
 	struct mthca_dev *dev = to_mdev(ibsrq->device);
 	struct mthca_srq *srq = to_msrq(ibsrq);
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index ee1f3a355666..61eed3996117 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -888,7 +888,8 @@ struct ib_device {
 						 struct ib_udata *udata);
 	int                        (*modify_srq)(struct ib_srq *srq,
 						 struct ib_srq_attr *srq_attr,
-						 enum ib_srq_attr_mask srq_attr_mask);
+						 enum ib_srq_attr_mask srq_attr_mask,
+						 struct ib_udata *udata);
 	int                        (*query_srq)(struct ib_srq *srq,
 						struct ib_srq_attr *srq_attr);
 	int                        (*destroy_srq)(struct ib_srq *srq);
@@ -900,7 +901,8 @@ struct ib_device {
 						struct ib_udata *udata);
 	int                        (*modify_qp)(struct ib_qp *qp,
 						struct ib_qp_attr *qp_attr,
-						int qp_attr_mask);
+						int qp_attr_mask,
+						struct ib_udata *udata);
 	int                        (*query_qp)(struct ib_qp *qp,
 					       struct ib_qp_attr *qp_attr,
 					       int qp_attr_mask,
-- 
cgit v1.2.3


From 922a8e9fb2e0711212badce47a41137e2ca04cb3 Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Thu, 3 Aug 2006 16:02:40 -0500
Subject: RDMA: iWARP Connection Manager.

Add an iWARP Connection Manager (CM), which abstracts connection
management for iWARP devices (RNICs).  It is a logical instance of the
xx_cm where xx is the transport type (ib or iw).  The symbols exported
are used by the transport independent rdma_cm module, and are
available also for transport dependent ULPs.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/iwcm.c | 1019 ++++++++++++++++++++++++++++++++++++++++
 drivers/infiniband/core/iwcm.h |   62 +++
 include/rdma/iw_cm.h           |  258 ++++++++++
 3 files changed, 1339 insertions(+)
 create mode 100644 drivers/infiniband/core/iwcm.c
 create mode 100644 drivers/infiniband/core/iwcm.h
 create mode 100644 include/rdma/iw_cm.h

(limited to 'include')

diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c
new file mode 100644
index 000000000000..c3fb304a4e86
--- /dev/null
+++ b/drivers/infiniband/core/iwcm.c
@@ -0,0 +1,1019 @@
+/*
+ * Copyright (c) 2004, 2005 Intel Corporation.  All rights reserved.
+ * Copyright (c) 2004 Topspin Corporation.  All rights reserved.
+ * Copyright (c) 2004, 2005 Voltaire Corporation.  All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
+ * Copyright (c) 2005 Open Grid Computing, Inc. All rights reserved.
+ * Copyright (c) 2005 Network Appliance, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+#include <linux/dma-mapping.h>
+#include <linux/err.h>
+#include <linux/idr.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/rbtree.h>
+#include <linux/spinlock.h>
+#include <linux/workqueue.h>
+#include <linux/completion.h>
+
+#include <rdma/iw_cm.h>
+#include <rdma/ib_addr.h>
+
+#include "iwcm.h"
+
+MODULE_AUTHOR("Tom Tucker");
+MODULE_DESCRIPTION("iWARP CM");
+MODULE_LICENSE("Dual BSD/GPL");
+
+static struct workqueue_struct *iwcm_wq;
+struct iwcm_work {
+	struct work_struct work;
+	struct iwcm_id_private *cm_id;
+	struct list_head list;
+	struct iw_cm_event event;
+	struct list_head free_list;
+};
+
+/*
+ * The following services provide a mechanism for pre-allocating iwcm_work
+ * elements.  The design pre-allocates them  based on the cm_id type:
+ *	LISTENING IDS: 	Get enough elements preallocated to handle the
+ *			listen backlog.
+ *	ACTIVE IDS:	4: CONNECT_REPLY, ESTABLISHED, DISCONNECT, CLOSE
+ *	PASSIVE IDS:	3: ESTABLISHED, DISCONNECT, CLOSE
+ *
+ * Allocating them in connect and listen avoids having to deal
+ * with allocation failures on the event upcall from the provider (which
+ * is called in the interrupt context).
+ *
+ * One exception is when creating the cm_id for incoming connection requests.
+ * There are two cases:
+ * 1) in the event upcall, cm_event_handler(), for a listening cm_id.  If
+ *    the backlog is exceeded, then no more connection request events will
+ *    be processed.  cm_event_handler() returns -ENOMEM in this case.  Its up
+ *    to the provider to reject the connectino request.
+ * 2) in the connection request workqueue handler, cm_conn_req_handler().
+ *    If work elements cannot be allocated for the new connect request cm_id,
+ *    then IWCM will call the provider reject method.  This is ok since
+ *    cm_conn_req_handler() runs in the workqueue thread context.
+ */
+
+static struct iwcm_work *get_work(struct iwcm_id_private *cm_id_priv)
+{
+	struct iwcm_work *work;
+
+	if (list_empty(&cm_id_priv->work_free_list))
+		return NULL;
+	work = list_entry(cm_id_priv->work_free_list.next, struct iwcm_work,
+			  free_list);
+	list_del_init(&work->free_list);
+	return work;
+}
+
+static void put_work(struct iwcm_work *work)
+{
+	list_add(&work->free_list, &work->cm_id->work_free_list);
+}
+
+static void dealloc_work_entries(struct iwcm_id_private *cm_id_priv)
+{
+	struct list_head *e, *tmp;
+
+	list_for_each_safe(e, tmp, &cm_id_priv->work_free_list)
+		kfree(list_entry(e, struct iwcm_work, free_list));
+}
+
+static int alloc_work_entries(struct iwcm_id_private *cm_id_priv, int count)
+{
+	struct iwcm_work *work;
+
+	BUG_ON(!list_empty(&cm_id_priv->work_free_list));
+	while (count--) {
+		work = kmalloc(sizeof(struct iwcm_work), GFP_KERNEL);
+		if (!work) {
+			dealloc_work_entries(cm_id_priv);
+			return -ENOMEM;
+		}
+		work->cm_id = cm_id_priv;
+		INIT_LIST_HEAD(&work->list);
+		put_work(work);
+	}
+	return 0;
+}
+
+/*
+ * Save private data from incoming connection requests in the
+ * cm_id_priv so the low level driver doesn't have to.  Adjust
+ * the event ptr to point to the local copy.
+ */
+static int copy_private_data(struct iwcm_id_private *cm_id_priv,
+		       struct iw_cm_event *event)
+{
+	void *p;
+
+	p = kmalloc(event->private_data_len, GFP_ATOMIC);
+	if (!p)
+		return -ENOMEM;
+	memcpy(p, event->private_data, event->private_data_len);
+	event->private_data = p;
+	return 0;
+}
+
+/*
+ * Release a reference on cm_id. If the last reference is being removed
+ * and iw_destroy_cm_id is waiting, wake up the waiting thread.
+ */
+static int iwcm_deref_id(struct iwcm_id_private *cm_id_priv)
+{
+	int ret = 0;
+
+	BUG_ON(atomic_read(&cm_id_priv->refcount)==0);
+	if (atomic_dec_and_test(&cm_id_priv->refcount)) {
+		BUG_ON(!list_empty(&cm_id_priv->work_list));
+		if (waitqueue_active(&cm_id_priv->destroy_comp.wait)) {
+			BUG_ON(cm_id_priv->state != IW_CM_STATE_DESTROYING);
+			BUG_ON(test_bit(IWCM_F_CALLBACK_DESTROY,
+					&cm_id_priv->flags));
+			ret = 1;
+		}
+		complete(&cm_id_priv->destroy_comp);
+	}
+
+	return ret;
+}
+
+static void add_ref(struct iw_cm_id *cm_id)
+{
+	struct iwcm_id_private *cm_id_priv;
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	atomic_inc(&cm_id_priv->refcount);
+}
+
+static void rem_ref(struct iw_cm_id *cm_id)
+{
+	struct iwcm_id_private *cm_id_priv;
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	iwcm_deref_id(cm_id_priv);
+}
+
+static int cm_event_handler(struct iw_cm_id *cm_id, struct iw_cm_event *event);
+
+struct iw_cm_id *iw_create_cm_id(struct ib_device *device,
+				 iw_cm_handler cm_handler,
+				 void *context)
+{
+	struct iwcm_id_private *cm_id_priv;
+
+	cm_id_priv = kzalloc(sizeof(*cm_id_priv), GFP_KERNEL);
+	if (!cm_id_priv)
+		return ERR_PTR(-ENOMEM);
+
+	cm_id_priv->state = IW_CM_STATE_IDLE;
+	cm_id_priv->id.device = device;
+	cm_id_priv->id.cm_handler = cm_handler;
+	cm_id_priv->id.context = context;
+	cm_id_priv->id.event_handler = cm_event_handler;
+	cm_id_priv->id.add_ref = add_ref;
+	cm_id_priv->id.rem_ref = rem_ref;
+	spin_lock_init(&cm_id_priv->lock);
+	atomic_set(&cm_id_priv->refcount, 1);
+	init_waitqueue_head(&cm_id_priv->connect_wait);
+	init_completion(&cm_id_priv->destroy_comp);
+	INIT_LIST_HEAD(&cm_id_priv->work_list);
+	INIT_LIST_HEAD(&cm_id_priv->work_free_list);
+
+	return &cm_id_priv->id;
+}
+EXPORT_SYMBOL(iw_create_cm_id);
+
+
+static int iwcm_modify_qp_err(struct ib_qp *qp)
+{
+	struct ib_qp_attr qp_attr;
+
+	if (!qp)
+		return -EINVAL;
+
+	qp_attr.qp_state = IB_QPS_ERR;
+	return ib_modify_qp(qp, &qp_attr, IB_QP_STATE);
+}
+
+/*
+ * This is really the RDMAC CLOSING state. It is most similar to the
+ * IB SQD QP state.
+ */
+static int iwcm_modify_qp_sqd(struct ib_qp *qp)
+{
+	struct ib_qp_attr qp_attr;
+
+	BUG_ON(qp == NULL);
+	qp_attr.qp_state = IB_QPS_SQD;
+	return ib_modify_qp(qp, &qp_attr, IB_QP_STATE);
+}
+
+/*
+ * CM_ID <-- CLOSING
+ *
+ * Block if a passive or active connection is currenlty being processed. Then
+ * process the event as follows:
+ * - If we are ESTABLISHED, move to CLOSING and modify the QP state
+ *   based on the abrupt flag
+ * - If the connection is already in the CLOSING or IDLE state, the peer is
+ *   disconnecting concurrently with us and we've already seen the
+ *   DISCONNECT event -- ignore the request and return 0
+ * - Disconnect on a listening endpoint returns -EINVAL
+ */
+int iw_cm_disconnect(struct iw_cm_id *cm_id, int abrupt)
+{
+	struct iwcm_id_private *cm_id_priv;
+	unsigned long flags;
+	int ret = 0;
+	struct ib_qp *qp = NULL;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	/* Wait if we're currently in a connect or accept downcall */
+	wait_event(cm_id_priv->connect_wait,
+		   !test_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags));
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	switch (cm_id_priv->state) {
+	case IW_CM_STATE_ESTABLISHED:
+		cm_id_priv->state = IW_CM_STATE_CLOSING;
+
+		/* QP could be <nul> for user-mode client */
+		if (cm_id_priv->qp)
+			qp = cm_id_priv->qp;
+		else
+			ret = -EINVAL;
+		break;
+	case IW_CM_STATE_LISTEN:
+		ret = -EINVAL;
+		break;
+	case IW_CM_STATE_CLOSING:
+		/* remote peer closed first */
+	case IW_CM_STATE_IDLE:
+		/* accept or connect returned !0 */
+		break;
+	case IW_CM_STATE_CONN_RECV:
+		/*
+		 * App called disconnect before/without calling accept after
+		 * connect_request event delivered.
+		 */
+		break;
+	case IW_CM_STATE_CONN_SENT:
+		/* Can only get here if wait above fails */
+	default:
+		BUG();
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	if (qp) {
+		if (abrupt)
+			ret = iwcm_modify_qp_err(qp);
+		else
+			ret = iwcm_modify_qp_sqd(qp);
+
+		/*
+		 * If both sides are disconnecting the QP could
+		 * already be in ERR or SQD states
+		 */
+		ret = 0;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(iw_cm_disconnect);
+
+/*
+ * CM_ID <-- DESTROYING
+ *
+ * Clean up all resources associated with the connection and release
+ * the initial reference taken by iw_create_cm_id.
+ */
+static void destroy_cm_id(struct iw_cm_id *cm_id)
+{
+	struct iwcm_id_private *cm_id_priv;
+	unsigned long flags;
+	int ret;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	/*
+	 * Wait if we're currently in a connect or accept downcall. A
+	 * listening endpoint should never block here.
+	 */
+	wait_event(cm_id_priv->connect_wait,
+		   !test_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags));
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	switch (cm_id_priv->state) {
+	case IW_CM_STATE_LISTEN:
+		cm_id_priv->state = IW_CM_STATE_DESTROYING;
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		/* destroy the listening endpoint */
+		ret = cm_id->device->iwcm->destroy_listen(cm_id);
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+		break;
+	case IW_CM_STATE_ESTABLISHED:
+		cm_id_priv->state = IW_CM_STATE_DESTROYING;
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		/* Abrupt close of the connection */
+		(void)iwcm_modify_qp_err(cm_id_priv->qp);
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+		break;
+	case IW_CM_STATE_IDLE:
+	case IW_CM_STATE_CLOSING:
+		cm_id_priv->state = IW_CM_STATE_DESTROYING;
+		break;
+	case IW_CM_STATE_CONN_RECV:
+		/*
+		 * App called destroy before/without calling accept after
+		 * receiving connection request event notification.
+		 */
+		cm_id_priv->state = IW_CM_STATE_DESTROYING;
+		break;
+	case IW_CM_STATE_CONN_SENT:
+	case IW_CM_STATE_DESTROYING:
+	default:
+		BUG();
+		break;
+	}
+	if (cm_id_priv->qp) {
+		cm_id_priv->id.device->iwcm->rem_ref(cm_id_priv->qp);
+		cm_id_priv->qp = NULL;
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	(void)iwcm_deref_id(cm_id_priv);
+}
+
+/*
+ * This function is only called by the application thread and cannot
+ * be called by the event thread. The function will wait for all
+ * references to be released on the cm_id and then kfree the cm_id
+ * object.
+ */
+void iw_destroy_cm_id(struct iw_cm_id *cm_id)
+{
+	struct iwcm_id_private *cm_id_priv;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	BUG_ON(test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags));
+
+	destroy_cm_id(cm_id);
+
+	wait_for_completion(&cm_id_priv->destroy_comp);
+
+	dealloc_work_entries(cm_id_priv);
+
+	kfree(cm_id_priv);
+}
+EXPORT_SYMBOL(iw_destroy_cm_id);
+
+/*
+ * CM_ID <-- LISTEN
+ *
+ * Start listening for connect requests. Generates one CONNECT_REQUEST
+ * event for each inbound connect request.
+ */
+int iw_cm_listen(struct iw_cm_id *cm_id, int backlog)
+{
+	struct iwcm_id_private *cm_id_priv;
+	unsigned long flags;
+	int ret = 0;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+
+	ret = alloc_work_entries(cm_id_priv, backlog);
+	if (ret)
+		return ret;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	switch (cm_id_priv->state) {
+	case IW_CM_STATE_IDLE:
+		cm_id_priv->state = IW_CM_STATE_LISTEN;
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		ret = cm_id->device->iwcm->create_listen(cm_id, backlog);
+		if (ret)
+			cm_id_priv->state = IW_CM_STATE_IDLE;
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+		break;
+	default:
+		ret = -EINVAL;
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(iw_cm_listen);
+
+/*
+ * CM_ID <-- IDLE
+ *
+ * Rejects an inbound connection request. No events are generated.
+ */
+int iw_cm_reject(struct iw_cm_id *cm_id,
+		 const void *private_data,
+		 u8 private_data_len)
+{
+	struct iwcm_id_private *cm_id_priv;
+	unsigned long flags;
+	int ret;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	set_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	if (cm_id_priv->state != IW_CM_STATE_CONN_RECV) {
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+		wake_up_all(&cm_id_priv->connect_wait);
+		return -EINVAL;
+	}
+	cm_id_priv->state = IW_CM_STATE_IDLE;
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	ret = cm_id->device->iwcm->reject(cm_id, private_data,
+					  private_data_len);
+
+	clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+	wake_up_all(&cm_id_priv->connect_wait);
+
+	return ret;
+}
+EXPORT_SYMBOL(iw_cm_reject);
+
+/*
+ * CM_ID <-- ESTABLISHED
+ *
+ * Accepts an inbound connection request and generates an ESTABLISHED
+ * event. Callers of iw_cm_disconnect and iw_destroy_cm_id will block
+ * until the ESTABLISHED event is received from the provider.
+ */
+int iw_cm_accept(struct iw_cm_id *cm_id,
+		 struct iw_cm_conn_param *iw_param)
+{
+	struct iwcm_id_private *cm_id_priv;
+	struct ib_qp *qp;
+	unsigned long flags;
+	int ret;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	set_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	if (cm_id_priv->state != IW_CM_STATE_CONN_RECV) {
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+		wake_up_all(&cm_id_priv->connect_wait);
+		return -EINVAL;
+	}
+	/* Get the ib_qp given the QPN */
+	qp = cm_id->device->iwcm->get_qp(cm_id->device, iw_param->qpn);
+	if (!qp) {
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		return -EINVAL;
+	}
+	cm_id->device->iwcm->add_ref(qp);
+	cm_id_priv->qp = qp;
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	ret = cm_id->device->iwcm->accept(cm_id, iw_param);
+	if (ret) {
+		/* An error on accept precludes provider events */
+		BUG_ON(cm_id_priv->state != IW_CM_STATE_CONN_RECV);
+		cm_id_priv->state = IW_CM_STATE_IDLE;
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+		if (cm_id_priv->qp) {
+			cm_id->device->iwcm->rem_ref(qp);
+			cm_id_priv->qp = NULL;
+		}
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+		wake_up_all(&cm_id_priv->connect_wait);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(iw_cm_accept);
+
+/*
+ * Active Side: CM_ID <-- CONN_SENT
+ *
+ * If successful, results in the generation of a CONNECT_REPLY
+ * event. iw_cm_disconnect and iw_cm_destroy will block until the
+ * CONNECT_REPLY event is received from the provider.
+ */
+int iw_cm_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *iw_param)
+{
+	struct iwcm_id_private *cm_id_priv;
+	int ret = 0;
+	unsigned long flags;
+	struct ib_qp *qp;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+
+	ret = alloc_work_entries(cm_id_priv, 4);
+	if (ret)
+		return ret;
+
+	set_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+
+	if (cm_id_priv->state != IW_CM_STATE_IDLE) {
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+		wake_up_all(&cm_id_priv->connect_wait);
+		return -EINVAL;
+	}
+
+	/* Get the ib_qp given the QPN */
+	qp = cm_id->device->iwcm->get_qp(cm_id->device, iw_param->qpn);
+	if (!qp) {
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		return -EINVAL;
+	}
+	cm_id->device->iwcm->add_ref(qp);
+	cm_id_priv->qp = qp;
+	cm_id_priv->state = IW_CM_STATE_CONN_SENT;
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	ret = cm_id->device->iwcm->connect(cm_id, iw_param);
+	if (ret) {
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+		if (cm_id_priv->qp) {
+			cm_id->device->iwcm->rem_ref(qp);
+			cm_id_priv->qp = NULL;
+		}
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		BUG_ON(cm_id_priv->state != IW_CM_STATE_CONN_SENT);
+		cm_id_priv->state = IW_CM_STATE_IDLE;
+		clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+		wake_up_all(&cm_id_priv->connect_wait);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(iw_cm_connect);
+
+/*
+ * Passive Side: new CM_ID <-- CONN_RECV
+ *
+ * Handles an inbound connect request. The function creates a new
+ * iw_cm_id to represent the new connection and inherits the client
+ * callback function and other attributes from the listening parent.
+ *
+ * The work item contains a pointer to the listen_cm_id and the event. The
+ * listen_cm_id contains the client cm_handler, context and
+ * device. These are copied when the device is cloned. The event
+ * contains the new four tuple.
+ *
+ * An error on the child should not affect the parent, so this
+ * function does not return a value.
+ */
+static void cm_conn_req_handler(struct iwcm_id_private *listen_id_priv,
+				struct iw_cm_event *iw_event)
+{
+	unsigned long flags;
+	struct iw_cm_id *cm_id;
+	struct iwcm_id_private *cm_id_priv;
+	int ret;
+
+	/*
+	 * The provider should never generate a connection request
+	 * event with a bad status.
+	 */
+	BUG_ON(iw_event->status);
+
+	/*
+	 * We could be destroying the listening id. If so, ignore this
+	 * upcall.
+	 */
+	spin_lock_irqsave(&listen_id_priv->lock, flags);
+	if (listen_id_priv->state != IW_CM_STATE_LISTEN) {
+		spin_unlock_irqrestore(&listen_id_priv->lock, flags);
+		return;
+	}
+	spin_unlock_irqrestore(&listen_id_priv->lock, flags);
+
+	cm_id = iw_create_cm_id(listen_id_priv->id.device,
+				listen_id_priv->id.cm_handler,
+				listen_id_priv->id.context);
+	/* If the cm_id could not be created, ignore the request */
+	if (IS_ERR(cm_id))
+		return;
+
+	cm_id->provider_data = iw_event->provider_data;
+	cm_id->local_addr = iw_event->local_addr;
+	cm_id->remote_addr = iw_event->remote_addr;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	cm_id_priv->state = IW_CM_STATE_CONN_RECV;
+
+	ret = alloc_work_entries(cm_id_priv, 3);
+	if (ret) {
+		iw_cm_reject(cm_id, NULL, 0);
+		iw_destroy_cm_id(cm_id);
+		return;
+	}
+
+	/* Call the client CM handler */
+	ret = cm_id->cm_handler(cm_id, iw_event);
+	if (ret) {
+		set_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
+		destroy_cm_id(cm_id);
+		if (atomic_read(&cm_id_priv->refcount)==0)
+			kfree(cm_id);
+	}
+
+	if (iw_event->private_data_len)
+		kfree(iw_event->private_data);
+}
+
+/*
+ * Passive Side: CM_ID <-- ESTABLISHED
+ *
+ * The provider generated an ESTABLISHED event which means that
+ * the MPA negotion has completed successfully and we are now in MPA
+ * FPDU mode.
+ *
+ * This event can only be received in the CONN_RECV state. If the
+ * remote peer closed, the ESTABLISHED event would be received followed
+ * by the CLOSE event. If the app closes, it will block until we wake
+ * it up after processing this event.
+ */
+static int cm_conn_est_handler(struct iwcm_id_private *cm_id_priv,
+			       struct iw_cm_event *iw_event)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+
+	/*
+	 * We clear the CONNECT_WAIT bit here to allow the callback
+	 * function to call iw_cm_disconnect. Calling iw_destroy_cm_id
+	 * from a callback handler is not allowed.
+	 */
+	clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+	BUG_ON(cm_id_priv->state != IW_CM_STATE_CONN_RECV);
+	cm_id_priv->state = IW_CM_STATE_ESTABLISHED;
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+	ret = cm_id_priv->id.cm_handler(&cm_id_priv->id, iw_event);
+	wake_up_all(&cm_id_priv->connect_wait);
+
+	return ret;
+}
+
+/*
+ * Active Side: CM_ID <-- ESTABLISHED
+ *
+ * The app has called connect and is waiting for the established event to
+ * post it's requests to the server. This event will wake up anyone
+ * blocked in iw_cm_disconnect or iw_destroy_id.
+ */
+static int cm_conn_rep_handler(struct iwcm_id_private *cm_id_priv,
+			       struct iw_cm_event *iw_event)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	/*
+	 * Clear the connect wait bit so a callback function calling
+	 * iw_cm_disconnect will not wait and deadlock this thread
+	 */
+	clear_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags);
+	BUG_ON(cm_id_priv->state != IW_CM_STATE_CONN_SENT);
+	if (iw_event->status == IW_CM_EVENT_STATUS_ACCEPTED) {
+		cm_id_priv->id.local_addr = iw_event->local_addr;
+		cm_id_priv->id.remote_addr = iw_event->remote_addr;
+		cm_id_priv->state = IW_CM_STATE_ESTABLISHED;
+	} else {
+		/* REJECTED or RESET */
+		cm_id_priv->id.device->iwcm->rem_ref(cm_id_priv->qp);
+		cm_id_priv->qp = NULL;
+		cm_id_priv->state = IW_CM_STATE_IDLE;
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+	ret = cm_id_priv->id.cm_handler(&cm_id_priv->id, iw_event);
+
+	if (iw_event->private_data_len)
+		kfree(iw_event->private_data);
+
+	/* Wake up waiters on connect complete */
+	wake_up_all(&cm_id_priv->connect_wait);
+
+	return ret;
+}
+
+/*
+ * CM_ID <-- CLOSING
+ *
+ * If in the ESTABLISHED state, move to CLOSING.
+ */
+static void cm_disconnect_handler(struct iwcm_id_private *cm_id_priv,
+				  struct iw_cm_event *iw_event)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	if (cm_id_priv->state == IW_CM_STATE_ESTABLISHED)
+		cm_id_priv->state = IW_CM_STATE_CLOSING;
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+}
+
+/*
+ * CM_ID <-- IDLE
+ *
+ * If in the ESTBLISHED or CLOSING states, the QP will have have been
+ * moved by the provider to the ERR state. Disassociate the CM_ID from
+ * the QP,  move to IDLE, and remove the 'connected' reference.
+ *
+ * If in some other state, the cm_id was destroyed asynchronously.
+ * This is the last reference that will result in waking up
+ * the app thread blocked in iw_destroy_cm_id.
+ */
+static int cm_close_handler(struct iwcm_id_private *cm_id_priv,
+				  struct iw_cm_event *iw_event)
+{
+	unsigned long flags;
+	int ret = 0;
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+
+	if (cm_id_priv->qp) {
+		cm_id_priv->id.device->iwcm->rem_ref(cm_id_priv->qp);
+		cm_id_priv->qp = NULL;
+	}
+	switch (cm_id_priv->state) {
+	case IW_CM_STATE_ESTABLISHED:
+	case IW_CM_STATE_CLOSING:
+		cm_id_priv->state = IW_CM_STATE_IDLE;
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+		ret = cm_id_priv->id.cm_handler(&cm_id_priv->id, iw_event);
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+		break;
+	case IW_CM_STATE_DESTROYING:
+		break;
+	default:
+		BUG();
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+	return ret;
+}
+
+static int process_event(struct iwcm_id_private *cm_id_priv,
+			 struct iw_cm_event *iw_event)
+{
+	int ret = 0;
+
+	switch (iw_event->event) {
+	case IW_CM_EVENT_CONNECT_REQUEST:
+		cm_conn_req_handler(cm_id_priv, iw_event);
+		break;
+	case IW_CM_EVENT_CONNECT_REPLY:
+		ret = cm_conn_rep_handler(cm_id_priv, iw_event);
+		break;
+	case IW_CM_EVENT_ESTABLISHED:
+		ret = cm_conn_est_handler(cm_id_priv, iw_event);
+		break;
+	case IW_CM_EVENT_DISCONNECT:
+		cm_disconnect_handler(cm_id_priv, iw_event);
+		break;
+	case IW_CM_EVENT_CLOSE:
+		ret = cm_close_handler(cm_id_priv, iw_event);
+		break;
+	default:
+		BUG();
+	}
+
+	return ret;
+}
+
+/*
+ * Process events on the work_list for the cm_id. If the callback
+ * function requests that the cm_id be deleted, a flag is set in the
+ * cm_id flags to indicate that when the last reference is
+ * removed, the cm_id is to be destroyed. This is necessary to
+ * distinguish between an object that will be destroyed by the app
+ * thread asleep on the destroy_comp list vs. an object destroyed
+ * here synchronously when the last reference is removed.
+ */
+static void cm_work_handler(void *arg)
+{
+	struct iwcm_work *work = arg, lwork;
+	struct iwcm_id_private *cm_id_priv = work->cm_id;
+	unsigned long flags;
+	int empty;
+	int ret = 0;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	empty = list_empty(&cm_id_priv->work_list);
+	while (!empty) {
+		work = list_entry(cm_id_priv->work_list.next,
+				  struct iwcm_work, list);
+		list_del_init(&work->list);
+		empty = list_empty(&cm_id_priv->work_list);
+		lwork = *work;
+		put_work(work);
+		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+
+		ret = process_event(cm_id_priv, &work->event);
+		if (ret) {
+			set_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
+			destroy_cm_id(&cm_id_priv->id);
+		}
+		BUG_ON(atomic_read(&cm_id_priv->refcount)==0);
+		if (iwcm_deref_id(cm_id_priv))
+			return;
+
+		if (atomic_read(&cm_id_priv->refcount)==0 &&
+		    test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags)) {
+			dealloc_work_entries(cm_id_priv);
+			kfree(cm_id_priv);
+			return;
+		}
+		spin_lock_irqsave(&cm_id_priv->lock, flags);
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+}
+
+/*
+ * This function is called on interrupt context. Schedule events on
+ * the iwcm_wq thread to allow callback functions to downcall into
+ * the CM and/or block.  Events are queued to a per-CM_ID
+ * work_list. If this is the first event on the work_list, the work
+ * element is also queued on the iwcm_wq thread.
+ *
+ * Each event holds a reference on the cm_id. Until the last posted
+ * event has been delivered and processed, the cm_id cannot be
+ * deleted.
+ *
+ * Returns:
+ * 	      0	- the event was handled.
+ *	-ENOMEM	- the event was not handled due to lack of resources.
+ */
+static int cm_event_handler(struct iw_cm_id *cm_id,
+			     struct iw_cm_event *iw_event)
+{
+	struct iwcm_work *work;
+	struct iwcm_id_private *cm_id_priv;
+	unsigned long flags;
+	int ret = 0;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	work = get_work(cm_id_priv);
+	if (!work) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	INIT_WORK(&work->work, cm_work_handler, work);
+	work->cm_id = cm_id_priv;
+	work->event = *iw_event;
+
+	if ((work->event.event == IW_CM_EVENT_CONNECT_REQUEST ||
+	     work->event.event == IW_CM_EVENT_CONNECT_REPLY) &&
+	    work->event.private_data_len) {
+		ret = copy_private_data(cm_id_priv, &work->event);
+		if (ret) {
+			put_work(work);
+			goto out;
+		}
+	}
+
+	atomic_inc(&cm_id_priv->refcount);
+	if (list_empty(&cm_id_priv->work_list)) {
+		list_add_tail(&work->list, &cm_id_priv->work_list);
+		queue_work(iwcm_wq, &work->work);
+	} else
+		list_add_tail(&work->list, &cm_id_priv->work_list);
+out:
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+	return ret;
+}
+
+static int iwcm_init_qp_init_attr(struct iwcm_id_private *cm_id_priv,
+				  struct ib_qp_attr *qp_attr,
+				  int *qp_attr_mask)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	switch (cm_id_priv->state) {
+	case IW_CM_STATE_IDLE:
+	case IW_CM_STATE_CONN_SENT:
+	case IW_CM_STATE_CONN_RECV:
+	case IW_CM_STATE_ESTABLISHED:
+		*qp_attr_mask = IB_QP_STATE | IB_QP_ACCESS_FLAGS;
+		qp_attr->qp_access_flags = IB_ACCESS_LOCAL_WRITE |
+					   IB_ACCESS_REMOTE_WRITE|
+					   IB_ACCESS_REMOTE_READ;
+		ret = 0;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+	return ret;
+}
+
+static int iwcm_init_qp_rts_attr(struct iwcm_id_private *cm_id_priv,
+				  struct ib_qp_attr *qp_attr,
+				  int *qp_attr_mask)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&cm_id_priv->lock, flags);
+	switch (cm_id_priv->state) {
+	case IW_CM_STATE_IDLE:
+	case IW_CM_STATE_CONN_SENT:
+	case IW_CM_STATE_CONN_RECV:
+	case IW_CM_STATE_ESTABLISHED:
+		*qp_attr_mask = 0;
+		ret = 0;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
+	return ret;
+}
+
+int iw_cm_init_qp_attr(struct iw_cm_id *cm_id,
+		       struct ib_qp_attr *qp_attr,
+		       int *qp_attr_mask)
+{
+	struct iwcm_id_private *cm_id_priv;
+	int ret;
+
+	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+	switch (qp_attr->qp_state) {
+	case IB_QPS_INIT:
+	case IB_QPS_RTR:
+		ret = iwcm_init_qp_init_attr(cm_id_priv,
+					     qp_attr, qp_attr_mask);
+		break;
+	case IB_QPS_RTS:
+		ret = iwcm_init_qp_rts_attr(cm_id_priv,
+					    qp_attr, qp_attr_mask);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+EXPORT_SYMBOL(iw_cm_init_qp_attr);
+
+static int __init iw_cm_init(void)
+{
+	iwcm_wq = create_singlethread_workqueue("iw_cm_wq");
+	if (!iwcm_wq)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void __exit iw_cm_cleanup(void)
+{
+	destroy_workqueue(iwcm_wq);
+}
+
+module_init(iw_cm_init);
+module_exit(iw_cm_cleanup);
diff --git a/drivers/infiniband/core/iwcm.h b/drivers/infiniband/core/iwcm.h
new file mode 100644
index 000000000000..3f6cc82564c8
--- /dev/null
+++ b/drivers/infiniband/core/iwcm.h
@@ -0,0 +1,62 @@
+/*
+ * Copyright (c) 2005 Network Appliance, Inc. All rights reserved.
+ * Copyright (c) 2005 Open Grid Computing, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef IWCM_H
+#define IWCM_H
+
+enum iw_cm_state {
+	IW_CM_STATE_IDLE,             /* unbound, inactive */
+	IW_CM_STATE_LISTEN,           /* listen waiting for connect */
+	IW_CM_STATE_CONN_RECV,        /* inbound waiting for user accept */
+	IW_CM_STATE_CONN_SENT,        /* outbound waiting for peer accept */
+	IW_CM_STATE_ESTABLISHED,      /* established */
+	IW_CM_STATE_CLOSING,	      /* disconnect */
+	IW_CM_STATE_DESTROYING        /* object being deleted */
+};
+
+struct iwcm_id_private {
+	struct iw_cm_id	id;
+	enum iw_cm_state state;
+	unsigned long flags;
+	struct ib_qp *qp;
+	struct completion destroy_comp;
+	wait_queue_head_t connect_wait;
+	struct list_head work_list;
+	spinlock_t lock;
+	atomic_t refcount;
+	struct list_head work_free_list;
+};
+
+#define IWCM_F_CALLBACK_DESTROY   1
+#define IWCM_F_CONNECT_WAIT       2
+
+#endif /* IWCM_H */
diff --git a/include/rdma/iw_cm.h b/include/rdma/iw_cm.h
new file mode 100644
index 000000000000..aeefa9b740dc
--- /dev/null
+++ b/include/rdma/iw_cm.h
@@ -0,0 +1,258 @@
+/*
+ * Copyright (c) 2005 Network Appliance, Inc. All rights reserved.
+ * Copyright (c) 2005 Open Grid Computing, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef IW_CM_H
+#define IW_CM_H
+
+#include <linux/in.h>
+#include <rdma/ib_cm.h>
+
+struct iw_cm_id;
+
+enum iw_cm_event_type {
+	IW_CM_EVENT_CONNECT_REQUEST = 1, /* connect request received */
+	IW_CM_EVENT_CONNECT_REPLY,	 /* reply from active connect request */
+	IW_CM_EVENT_ESTABLISHED,	 /* passive side accept successful */
+	IW_CM_EVENT_DISCONNECT,		 /* orderly shutdown */
+	IW_CM_EVENT_CLOSE		 /* close complete */
+};
+
+enum iw_cm_event_status {
+	IW_CM_EVENT_STATUS_OK = 0,	 /* request successful */
+	IW_CM_EVENT_STATUS_ACCEPTED = 0, /* connect request accepted */
+	IW_CM_EVENT_STATUS_REJECTED,	 /* connect request rejected */
+	IW_CM_EVENT_STATUS_TIMEOUT,	 /* the operation timed out */
+	IW_CM_EVENT_STATUS_RESET,	 /* reset from remote peer */
+	IW_CM_EVENT_STATUS_EINVAL,	 /* asynchronous failure for bad parm */
+};
+
+struct iw_cm_event {
+	enum iw_cm_event_type event;
+	enum iw_cm_event_status status;
+	struct sockaddr_in local_addr;
+	struct sockaddr_in remote_addr;
+	void *private_data;
+	u8 private_data_len;
+	void* provider_data;
+};
+
+/**
+ * iw_cm_handler - Function to be called by the IW CM when delivering events
+ * to the client.
+ *
+ * @cm_id: The IW CM identifier associated with the event.
+ * @event: Pointer to the event structure.
+ */
+typedef int (*iw_cm_handler)(struct iw_cm_id *cm_id,
+			     struct iw_cm_event *event);
+
+/**
+ * iw_event_handler - Function called by the provider when delivering provider
+ * events to the IW CM.  Returns either 0 indicating the event was processed
+ * or -errno if the event could not be processed.
+ *
+ * @cm_id: The IW CM identifier associated with the event.
+ * @event: Pointer to the event structure.
+ */
+typedef int (*iw_event_handler)(struct iw_cm_id *cm_id,
+				 struct iw_cm_event *event);
+
+struct iw_cm_id {
+	iw_cm_handler		cm_handler;      /* client callback function */
+	void		        *context;	 /* client cb context */
+	struct ib_device	*device;
+	struct sockaddr_in      local_addr;
+	struct sockaddr_in	remote_addr;
+	void			*provider_data;	 /* provider private data */
+	iw_event_handler        event_handler;   /* cb for provider
+						    events */
+	/* Used by provider to add and remove refs on IW cm_id */
+	void (*add_ref)(struct iw_cm_id *);
+	void (*rem_ref)(struct iw_cm_id *);
+};
+
+struct iw_cm_conn_param {
+	const void *private_data;
+	u16 private_data_len;
+	u32 ord;
+	u32 ird;
+	u32 qpn;
+};
+
+struct iw_cm_verbs {
+	void		(*add_ref)(struct ib_qp *qp);
+
+	void		(*rem_ref)(struct ib_qp *qp);
+
+	struct ib_qp *	(*get_qp)(struct ib_device *device,
+				  int qpn);
+
+	int		(*connect)(struct iw_cm_id *cm_id,
+				   struct iw_cm_conn_param *conn_param);
+
+	int		(*accept)(struct iw_cm_id *cm_id,
+				  struct iw_cm_conn_param *conn_param);
+
+	int		(*reject)(struct iw_cm_id *cm_id,
+				  const void *pdata, u8 pdata_len);
+
+	int		(*create_listen)(struct iw_cm_id *cm_id,
+					 int backlog);
+
+	int		(*destroy_listen)(struct iw_cm_id *cm_id);
+};
+
+/**
+ * iw_create_cm_id - Create an IW CM identifier.
+ *
+ * @device: The IB device on which to create the IW CM identier.
+ * @event_handler: User callback invoked to report events associated with the
+ *   returned IW CM identifier.
+ * @context: User specified context associated with the id.
+ */
+struct iw_cm_id *iw_create_cm_id(struct ib_device *device,
+				 iw_cm_handler cm_handler, void *context);
+
+/**
+ * iw_destroy_cm_id - Destroy an IW CM identifier.
+ *
+ * @cm_id: The previously created IW CM identifier to destroy.
+ *
+ * The client can assume that no events will be delivered for the CM ID after
+ * this function returns.
+ */
+void iw_destroy_cm_id(struct iw_cm_id *cm_id);
+
+/**
+ * iw_cm_bind_qp - Unbind the specified IW CM identifier and QP
+ *
+ * @cm_id: The IW CM idenfier to unbind from the QP.
+ * @qp: The QP
+ *
+ * This is called by the provider when destroying the QP to ensure
+ * that any references held by the IWCM are released. It may also
+ * be called by the IWCM when destroying a CM_ID to that any
+ * references held by the provider are released.
+ */
+void iw_cm_unbind_qp(struct iw_cm_id *cm_id, struct ib_qp *qp);
+
+/**
+ * iw_cm_get_qp - Return the ib_qp associated with a QPN
+ *
+ * @ib_device: The IB device
+ * @qpn: The queue pair number
+ */
+struct ib_qp *iw_cm_get_qp(struct ib_device *device, int qpn);
+
+/**
+ * iw_cm_listen - Listen for incoming connection requests on the
+ * specified IW CM id.
+ *
+ * @cm_id: The IW CM identifier.
+ * @backlog: The maximum number of outstanding un-accepted inbound listen
+ *   requests to queue.
+ *
+ * The source address and port number are specified in the IW CM identifier
+ * structure.
+ */
+int iw_cm_listen(struct iw_cm_id *cm_id, int backlog);
+
+/**
+ * iw_cm_accept - Called to accept an incoming connect request.
+ *
+ * @cm_id: The IW CM identifier associated with the connection request.
+ * @iw_param: Pointer to a structure containing connection establishment
+ *   parameters.
+ *
+ * The specified cm_id will have been provided in the event data for a
+ * CONNECT_REQUEST event. Subsequent events related to this connection will be
+ * delivered to the specified IW CM identifier prior and may occur prior to
+ * the return of this function. If this function returns a non-zero value, the
+ * client can assume that no events will be delivered to the specified IW CM
+ * identifier.
+ */
+int iw_cm_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *iw_param);
+
+/**
+ * iw_cm_reject - Reject an incoming connection request.
+ *
+ * @cm_id: Connection identifier associated with the request.
+ * @private_daa: Pointer to data to deliver to the remote peer as part of the
+ *   reject message.
+ * @private_data_len: The number of bytes in the private_data parameter.
+ *
+ * The client can assume that no events will be delivered to the specified IW
+ * CM identifier following the return of this function. The private_data
+ * buffer is available for reuse when this function returns.
+ */
+int iw_cm_reject(struct iw_cm_id *cm_id, const void *private_data,
+		 u8 private_data_len);
+
+/**
+ * iw_cm_connect - Called to request a connection to a remote peer.
+ *
+ * @cm_id: The IW CM identifier for the connection.
+ * @iw_param: Pointer to a structure containing connection  establishment
+ *   parameters.
+ *
+ * Events may be delivered to the specified IW CM identifier prior to the
+ * return of this function. If this function returns a non-zero value, the
+ * client can assume that no events will be delivered to the specified IW CM
+ * identifier.
+ */
+int iw_cm_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *iw_param);
+
+/**
+ * iw_cm_disconnect - Close the specified connection.
+ *
+ * @cm_id: The IW CM identifier to close.
+ * @abrupt: If 0, the connection will be closed gracefully, otherwise, the
+ *   connection will be reset.
+ *
+ * The IW CM identifier is still active until the IW_CM_EVENT_CLOSE event is
+ * delivered.
+ */
+int iw_cm_disconnect(struct iw_cm_id *cm_id, int abrupt);
+
+/**
+ * iw_cm_init_qp_attr - Called to initialize the attributes of the QP
+ * associated with a IW CM identifier.
+ *
+ * @cm_id: The IW CM identifier associated with the QP
+ * @qp_attr: Pointer to the QP attributes structure.
+ * @qp_attr_mask: Pointer to a bit vector specifying which QP attributes are
+ *   valid.
+ */
+int iw_cm_init_qp_attr(struct iw_cm_id *cm_id, struct ib_qp_attr *qp_attr,
+		       int *qp_attr_mask);
+
+#endif /* IW_CM_H */
-- 
cgit v1.2.3


From 07ebafbaaa72aa6a35472879008f5a1d1d469a0c Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Thu, 3 Aug 2006 16:02:42 -0500
Subject: RDMA: iWARP Core Changes.

Modifications to the existing rdma header files, core files, drivers,
and ulp files to support iWARP, including:
 - Hook iWARP CM into the build system and use it in rdma_cm.
 - Convert enum ib_node_type to enum rdma_node_type, which includes
   the possibility of RDMA_NODE_RNIC, and update everything for this.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/Makefile             |   4 +-
 drivers/infiniband/core/addr.c               |  18 +-
 drivers/infiniband/core/cache.c              |   5 +-
 drivers/infiniband/core/cm.c                 |   3 +
 drivers/infiniband/core/cma.c                | 356 ++++++++++++++++++++++++---
 drivers/infiniband/core/device.c             |   4 +-
 drivers/infiniband/core/mad.c                |   7 +-
 drivers/infiniband/core/sa_query.c           |   5 +-
 drivers/infiniband/core/smi.c                |  16 +-
 drivers/infiniband/core/sysfs.c              |  11 +-
 drivers/infiniband/core/ucm.c                |   3 +-
 drivers/infiniband/core/user_mad.c           |   5 +-
 drivers/infiniband/core/verbs.c              |  17 ++
 drivers/infiniband/hw/ehca/ehca_main.c       |   2 +-
 drivers/infiniband/hw/ipath/ipath_verbs.c    |   2 +-
 drivers/infiniband/hw/mthca/mthca_provider.c |   2 +-
 drivers/infiniband/ulp/ipoib/ipoib_main.c    |   8 +-
 drivers/infiniband/ulp/srp/ib_srp.c          |   2 +-
 include/rdma/ib_addr.h                       |  17 +-
 include/rdma/ib_verbs.h                      |  25 +-
 20 files changed, 431 insertions(+), 81 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile
index 68e73ec2d1f8..163d991eb8c9 100644
--- a/drivers/infiniband/core/Makefile
+++ b/drivers/infiniband/core/Makefile
@@ -1,7 +1,7 @@
 infiniband-$(CONFIG_INFINIBAND_ADDR_TRANS)	:= ib_addr.o rdma_cm.o
 
 obj-$(CONFIG_INFINIBAND) +=		ib_core.o ib_mad.o ib_sa.o \
-					ib_cm.o $(infiniband-y)
+					ib_cm.o iw_cm.o $(infiniband-y)
 obj-$(CONFIG_INFINIBAND_USER_MAD) +=	ib_umad.o
 obj-$(CONFIG_INFINIBAND_USER_ACCESS) +=	ib_uverbs.o ib_ucm.o
 
@@ -14,6 +14,8 @@ ib_sa-y :=			sa_query.o
 
 ib_cm-y :=			cm.o
 
+iw_cm-y :=			iwcm.o
+
 rdma_cm-y :=			cma.o
 
 ib_addr-y :=			addr.o
diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index d8e54e002ce3..9cbf09e2052f 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -61,12 +61,15 @@ static LIST_HEAD(req_list);
 static DECLARE_WORK(work, process_req, NULL);
 static struct workqueue_struct *addr_wq;
 
-static int copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev,
-		     unsigned char *dst_dev_addr)
+int rdma_copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev,
+		     const unsigned char *dst_dev_addr)
 {
 	switch (dev->type) {
 	case ARPHRD_INFINIBAND:
-		dev_addr->dev_type = IB_NODE_CA;
+		dev_addr->dev_type = RDMA_NODE_IB_CA;
+		break;
+	case ARPHRD_ETHER:
+		dev_addr->dev_type = RDMA_NODE_RNIC;
 		break;
 	default:
 		return -EADDRNOTAVAIL;
@@ -78,6 +81,7 @@ static int copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev,
 		memcpy(dev_addr->dst_dev_addr, dst_dev_addr, MAX_ADDR_LEN);
 	return 0;
 }
+EXPORT_SYMBOL(rdma_copy_addr);
 
 int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr)
 {
@@ -89,7 +93,7 @@ int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr)
 	if (!dev)
 		return -EADDRNOTAVAIL;
 
-	ret = copy_addr(dev_addr, dev, NULL);
+	ret = rdma_copy_addr(dev_addr, dev, NULL);
 	dev_put(dev);
 	return ret;
 }
@@ -161,7 +165,7 @@ static int addr_resolve_remote(struct sockaddr_in *src_in,
 
 	/* If the device does ARP internally, return 'done' */
 	if (rt->idev->dev->flags & IFF_NOARP) {
-		copy_addr(addr, rt->idev->dev, NULL);
+		rdma_copy_addr(addr, rt->idev->dev, NULL);
 		goto put;
 	}
 
@@ -181,7 +185,7 @@ static int addr_resolve_remote(struct sockaddr_in *src_in,
 		src_in->sin_addr.s_addr = rt->rt_src;
 	}
 
-	ret = copy_addr(addr, neigh->dev, neigh->ha);
+	ret = rdma_copy_addr(addr, neigh->dev, neigh->ha);
 release:
 	neigh_release(neigh);
 put:
@@ -245,7 +249,7 @@ static int addr_resolve_local(struct sockaddr_in *src_in,
 	if (ZERONET(src_ip)) {
 		src_in->sin_family = dst_in->sin_family;
 		src_in->sin_addr.s_addr = dst_ip;
-		ret = copy_addr(addr, dev, dev->dev_addr);
+		ret = rdma_copy_addr(addr, dev, dev->dev_addr);
 	} else if (LOOPBACK(src_ip)) {
 		ret = rdma_translate_ip((struct sockaddr *)dst_in, addr);
 		if (!ret)
diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c
index 75313ade2e0d..20e9f64e67a6 100644
--- a/drivers/infiniband/core/cache.c
+++ b/drivers/infiniband/core/cache.c
@@ -62,12 +62,13 @@ struct ib_update_work {
 
 static inline int start_port(struct ib_device *device)
 {
-	return device->node_type == IB_NODE_SWITCH ? 0 : 1;
+	return (device->node_type == RDMA_NODE_IB_SWITCH) ? 0 : 1;
 }
 
 static inline int end_port(struct ib_device *device)
 {
-	return device->node_type == IB_NODE_SWITCH ? 0 : device->phys_port_cnt;
+	return (device->node_type == RDMA_NODE_IB_SWITCH) ?
+		0 : device->phys_port_cnt;
 }
 
 int ib_get_cached_gid(struct ib_device *device,
diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c
index 1c145fe92a54..e130d2e89515 100644
--- a/drivers/infiniband/core/cm.c
+++ b/drivers/infiniband/core/cm.c
@@ -3280,6 +3280,9 @@ static void cm_add_one(struct ib_device *device)
 	int ret;
 	u8 i;
 
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
 	cm_dev = kmalloc(sizeof(*cm_dev) + sizeof(*port) *
 			 device->phys_port_cnt, GFP_KERNEL);
 	if (!cm_dev)
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 9d58bb59cd45..e88a7c652ca0 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -35,6 +35,7 @@
 #include <linux/mutex.h>
 #include <linux/random.h>
 #include <linux/idr.h>
+#include <linux/inetdevice.h>
 
 #include <net/tcp.h>
 
@@ -43,6 +44,7 @@
 #include <rdma/ib_cache.h>
 #include <rdma/ib_cm.h>
 #include <rdma/ib_sa.h>
+#include <rdma/iw_cm.h>
 
 MODULE_AUTHOR("Sean Hefty");
 MODULE_DESCRIPTION("Generic RDMA CM Agent");
@@ -124,6 +126,7 @@ struct rdma_id_private {
 	int			query_id;
 	union {
 		struct ib_cm_id	*ib;
+		struct iw_cm_id	*iw;
 	} cm_id;
 
 	u32			seq_num;
@@ -259,14 +262,23 @@ static void cma_detach_from_dev(struct rdma_id_private *id_priv)
 	id_priv->cma_dev = NULL;
 }
 
-static int cma_acquire_ib_dev(struct rdma_id_private *id_priv)
+static int cma_acquire_dev(struct rdma_id_private *id_priv)
 {
+	enum rdma_node_type dev_type = id_priv->id.route.addr.dev_addr.dev_type;
 	struct cma_device *cma_dev;
 	union ib_gid gid;
 	int ret = -ENODEV;
 
-	ib_addr_get_sgid(&id_priv->id.route.addr.dev_addr, &gid),
-
+	switch (rdma_node_get_transport(dev_type)) {
+	case RDMA_TRANSPORT_IB:
+		ib_addr_get_sgid(&id_priv->id.route.addr.dev_addr, &gid);
+		break;
+	case RDMA_TRANSPORT_IWARP:
+		iw_addr_get_sgid(&id_priv->id.route.addr.dev_addr, &gid);
+		break;
+	default:
+		return -ENODEV;
+	}
 	mutex_lock(&lock);
 	list_for_each_entry(cma_dev, &dev_list, list) {
 		ret = ib_find_cached_gid(cma_dev->device, &gid,
@@ -280,16 +292,6 @@ static int cma_acquire_ib_dev(struct rdma_id_private *id_priv)
 	return ret;
 }
 
-static int cma_acquire_dev(struct rdma_id_private *id_priv)
-{
-	switch (id_priv->id.route.addr.dev_addr.dev_type) {
-	case IB_NODE_CA:
-		return cma_acquire_ib_dev(id_priv);
-	default:
-		return -ENODEV;
-	}
-}
-
 static void cma_deref_id(struct rdma_id_private *id_priv)
 {
 	if (atomic_dec_and_test(&id_priv->refcount))
@@ -347,6 +349,16 @@ static int cma_init_ib_qp(struct rdma_id_private *id_priv, struct ib_qp *qp)
 					  IB_QP_PKEY_INDEX | IB_QP_PORT);
 }
 
+static int cma_init_iw_qp(struct rdma_id_private *id_priv, struct ib_qp *qp)
+{
+	struct ib_qp_attr qp_attr;
+
+	qp_attr.qp_state = IB_QPS_INIT;
+	qp_attr.qp_access_flags = IB_ACCESS_LOCAL_WRITE;
+
+	return ib_modify_qp(qp, &qp_attr, IB_QP_STATE | IB_QP_ACCESS_FLAGS);
+}
+
 int rdma_create_qp(struct rdma_cm_id *id, struct ib_pd *pd,
 		   struct ib_qp_init_attr *qp_init_attr)
 {
@@ -362,10 +374,13 @@ int rdma_create_qp(struct rdma_cm_id *id, struct ib_pd *pd,
 	if (IS_ERR(qp))
 		return PTR_ERR(qp);
 
-	switch (id->device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id->device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		ret = cma_init_ib_qp(id_priv, qp);
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = cma_init_iw_qp(id_priv, qp);
+		break;
 	default:
 		ret = -ENOSYS;
 		break;
@@ -451,13 +466,17 @@ int rdma_init_qp_attr(struct rdma_cm_id *id, struct ib_qp_attr *qp_attr,
 	int ret;
 
 	id_priv = container_of(id, struct rdma_id_private, id);
-	switch (id_priv->id.device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id_priv->id.device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		ret = ib_cm_init_qp_attr(id_priv->cm_id.ib, qp_attr,
 					 qp_attr_mask);
 		if (qp_attr->qp_state == IB_QPS_RTR)
 			qp_attr->rq_psn = id_priv->seq_num;
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = iw_cm_init_qp_attr(id_priv->cm_id.iw, qp_attr,
+					qp_attr_mask);
+		break;
 	default:
 		ret = -ENOSYS;
 		break;
@@ -590,8 +609,8 @@ static int cma_notify_user(struct rdma_id_private *id_priv,
 
 static void cma_cancel_route(struct rdma_id_private *id_priv)
 {
-	switch (id_priv->id.device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id_priv->id.device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		if (id_priv->query)
 			ib_sa_cancel_query(id_priv->query_id, id_priv->query);
 		break;
@@ -611,11 +630,15 @@ static void cma_destroy_listen(struct rdma_id_private *id_priv)
 	cma_exch(id_priv, CMA_DESTROYING);
 
 	if (id_priv->cma_dev) {
-		switch (id_priv->id.device->node_type) {
-		case IB_NODE_CA:
+		switch (rdma_node_get_transport(id_priv->id.device->node_type)) {
+		case RDMA_TRANSPORT_IB:
 			if (id_priv->cm_id.ib && !IS_ERR(id_priv->cm_id.ib))
 				ib_destroy_cm_id(id_priv->cm_id.ib);
 			break;
+		case RDMA_TRANSPORT_IWARP:
+			if (id_priv->cm_id.iw && !IS_ERR(id_priv->cm_id.iw))
+				iw_destroy_cm_id(id_priv->cm_id.iw);
+			break;
 		default:
 			break;
 		}
@@ -690,11 +713,15 @@ void rdma_destroy_id(struct rdma_cm_id *id)
 	cma_cancel_operation(id_priv, state);
 
 	if (id_priv->cma_dev) {
-		switch (id->device->node_type) {
-		case IB_NODE_CA:
+		switch (rdma_node_get_transport(id->device->node_type)) {
+		case RDMA_TRANSPORT_IB:
 			if (id_priv->cm_id.ib && !IS_ERR(id_priv->cm_id.ib))
 				ib_destroy_cm_id(id_priv->cm_id.ib);
 			break;
+		case RDMA_TRANSPORT_IWARP:
+			if (id_priv->cm_id.iw && !IS_ERR(id_priv->cm_id.iw))
+				iw_destroy_cm_id(id_priv->cm_id.iw);
+			break;
 		default:
 			break;
 		}
@@ -869,7 +896,7 @@ static struct rdma_id_private *cma_new_id(struct rdma_cm_id *listen_id,
 	ib_addr_set_sgid(&rt->addr.dev_addr, &rt->path_rec[0].sgid);
 	ib_addr_set_dgid(&rt->addr.dev_addr, &rt->path_rec[0].dgid);
 	ib_addr_set_pkey(&rt->addr.dev_addr, be16_to_cpu(rt->path_rec[0].pkey));
-	rt->addr.dev_addr.dev_type = IB_NODE_CA;
+	rt->addr.dev_addr.dev_type = RDMA_NODE_IB_CA;
 
 	id_priv = container_of(id, struct rdma_id_private, id);
 	id_priv->state = CMA_CONNECT;
@@ -898,7 +925,7 @@ static int cma_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event)
 	}
 
 	atomic_inc(&conn_id->dev_remove);
-	ret = cma_acquire_ib_dev(conn_id);
+	ret = cma_acquire_dev(conn_id);
 	if (ret) {
 		ret = -ENODEV;
 		cma_release_remove(conn_id);
@@ -982,6 +1009,128 @@ static void cma_set_compare_data(enum rdma_port_space ps, struct sockaddr *addr,
 	}
 }
 
+static int cma_iw_handler(struct iw_cm_id *iw_id, struct iw_cm_event *iw_event)
+{
+	struct rdma_id_private *id_priv = iw_id->context;
+	enum rdma_cm_event_type event = 0;
+	struct sockaddr_in *sin;
+	int ret = 0;
+
+	atomic_inc(&id_priv->dev_remove);
+
+	switch (iw_event->event) {
+	case IW_CM_EVENT_CLOSE:
+		event = RDMA_CM_EVENT_DISCONNECTED;
+		break;
+	case IW_CM_EVENT_CONNECT_REPLY:
+		sin = (struct sockaddr_in *) &id_priv->id.route.addr.src_addr;
+		*sin = iw_event->local_addr;
+		sin = (struct sockaddr_in *) &id_priv->id.route.addr.dst_addr;
+		*sin = iw_event->remote_addr;
+		if (iw_event->status)
+			event = RDMA_CM_EVENT_REJECTED;
+		else
+			event = RDMA_CM_EVENT_ESTABLISHED;
+		break;
+	case IW_CM_EVENT_ESTABLISHED:
+		event = RDMA_CM_EVENT_ESTABLISHED;
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	ret = cma_notify_user(id_priv, event, iw_event->status,
+			      iw_event->private_data,
+			      iw_event->private_data_len);
+	if (ret) {
+		/* Destroy the CM ID by returning a non-zero value. */
+		id_priv->cm_id.iw = NULL;
+		cma_exch(id_priv, CMA_DESTROYING);
+		cma_release_remove(id_priv);
+		rdma_destroy_id(&id_priv->id);
+		return ret;
+	}
+
+	cma_release_remove(id_priv);
+	return ret;
+}
+
+static int iw_conn_req_handler(struct iw_cm_id *cm_id,
+			       struct iw_cm_event *iw_event)
+{
+	struct rdma_cm_id *new_cm_id;
+	struct rdma_id_private *listen_id, *conn_id;
+	struct sockaddr_in *sin;
+	struct net_device *dev = NULL;
+	int ret;
+
+	listen_id = cm_id->context;
+	atomic_inc(&listen_id->dev_remove);
+	if (!cma_comp(listen_id, CMA_LISTEN)) {
+		ret = -ECONNABORTED;
+		goto out;
+	}
+
+	/* Create a new RDMA id for the new IW CM ID */
+	new_cm_id = rdma_create_id(listen_id->id.event_handler,
+				   listen_id->id.context,
+				   RDMA_PS_TCP);
+	if (!new_cm_id) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	conn_id = container_of(new_cm_id, struct rdma_id_private, id);
+	atomic_inc(&conn_id->dev_remove);
+	conn_id->state = CMA_CONNECT;
+
+	dev = ip_dev_find(iw_event->local_addr.sin_addr.s_addr);
+	if (!dev) {
+		ret = -EADDRNOTAVAIL;
+		cma_release_remove(conn_id);
+		rdma_destroy_id(new_cm_id);
+		goto out;
+	}
+	ret = rdma_copy_addr(&conn_id->id.route.addr.dev_addr, dev, NULL);
+	if (ret) {
+		cma_release_remove(conn_id);
+		rdma_destroy_id(new_cm_id);
+		goto out;
+	}
+
+	ret = cma_acquire_dev(conn_id);
+	if (ret) {
+		cma_release_remove(conn_id);
+		rdma_destroy_id(new_cm_id);
+		goto out;
+	}
+
+	conn_id->cm_id.iw = cm_id;
+	cm_id->context = conn_id;
+	cm_id->cm_handler = cma_iw_handler;
+
+	sin = (struct sockaddr_in *) &new_cm_id->route.addr.src_addr;
+	*sin = iw_event->local_addr;
+	sin = (struct sockaddr_in *) &new_cm_id->route.addr.dst_addr;
+	*sin = iw_event->remote_addr;
+
+	ret = cma_notify_user(conn_id, RDMA_CM_EVENT_CONNECT_REQUEST, 0,
+			      iw_event->private_data,
+			      iw_event->private_data_len);
+	if (ret) {
+		/* User wants to destroy the CM ID */
+		conn_id->cm_id.iw = NULL;
+		cma_exch(conn_id, CMA_DESTROYING);
+		cma_release_remove(conn_id);
+		rdma_destroy_id(&conn_id->id);
+	}
+
+out:
+	if (dev)
+		dev_put(dev);
+	cma_release_remove(listen_id);
+	return ret;
+}
+
 static int cma_ib_listen(struct rdma_id_private *id_priv)
 {
 	struct ib_cm_compare_data compare_data;
@@ -1011,6 +1160,30 @@ static int cma_ib_listen(struct rdma_id_private *id_priv)
 	return ret;
 }
 
+static int cma_iw_listen(struct rdma_id_private *id_priv, int backlog)
+{
+	int ret;
+	struct sockaddr_in *sin;
+
+	id_priv->cm_id.iw = iw_create_cm_id(id_priv->id.device,
+					    iw_conn_req_handler,
+					    id_priv);
+	if (IS_ERR(id_priv->cm_id.iw))
+		return PTR_ERR(id_priv->cm_id.iw);
+
+	sin = (struct sockaddr_in *) &id_priv->id.route.addr.src_addr;
+	id_priv->cm_id.iw->local_addr = *sin;
+
+	ret = iw_cm_listen(id_priv->cm_id.iw, backlog);
+
+	if (ret) {
+		iw_destroy_cm_id(id_priv->cm_id.iw);
+		id_priv->cm_id.iw = NULL;
+	}
+
+	return ret;
+}
+
 static int cma_listen_handler(struct rdma_cm_id *id,
 			      struct rdma_cm_event *event)
 {
@@ -1087,12 +1260,17 @@ int rdma_listen(struct rdma_cm_id *id, int backlog)
 
 	id_priv->backlog = backlog;
 	if (id->device) {
-		switch (id->device->node_type) {
-		case IB_NODE_CA:
+		switch (rdma_node_get_transport(id->device->node_type)) {
+		case RDMA_TRANSPORT_IB:
 			ret = cma_ib_listen(id_priv);
 			if (ret)
 				goto err;
 			break;
+		case RDMA_TRANSPORT_IWARP:
+			ret = cma_iw_listen(id_priv, backlog);
+			if (ret)
+				goto err;
+			break;
 		default:
 			ret = -ENOSYS;
 			goto err;
@@ -1231,6 +1409,23 @@ err:
 }
 EXPORT_SYMBOL(rdma_set_ib_paths);
 
+static int cma_resolve_iw_route(struct rdma_id_private *id_priv, int timeout_ms)
+{
+	struct cma_work *work;
+
+	work = kzalloc(sizeof *work, GFP_KERNEL);
+	if (!work)
+		return -ENOMEM;
+
+	work->id = id_priv;
+	INIT_WORK(&work->work, cma_work_handler, work);
+	work->old_state = CMA_ROUTE_QUERY;
+	work->new_state = CMA_ROUTE_RESOLVED;
+	work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED;
+	queue_work(cma_wq, &work->work);
+	return 0;
+}
+
 int rdma_resolve_route(struct rdma_cm_id *id, int timeout_ms)
 {
 	struct rdma_id_private *id_priv;
@@ -1241,10 +1436,13 @@ int rdma_resolve_route(struct rdma_cm_id *id, int timeout_ms)
 		return -EINVAL;
 
 	atomic_inc(&id_priv->refcount);
-	switch (id->device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id->device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		ret = cma_resolve_ib_route(id_priv, timeout_ms);
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = cma_resolve_iw_route(id_priv, timeout_ms);
+		break;
 	default:
 		ret = -ENOSYS;
 		break;
@@ -1649,6 +1847,47 @@ out:
 	return ret;
 }
 
+static int cma_connect_iw(struct rdma_id_private *id_priv,
+			  struct rdma_conn_param *conn_param)
+{
+	struct iw_cm_id *cm_id;
+	struct sockaddr_in* sin;
+	int ret;
+	struct iw_cm_conn_param iw_param;
+
+	cm_id = iw_create_cm_id(id_priv->id.device, cma_iw_handler, id_priv);
+	if (IS_ERR(cm_id)) {
+		ret = PTR_ERR(cm_id);
+		goto out;
+	}
+
+	id_priv->cm_id.iw = cm_id;
+
+	sin = (struct sockaddr_in*) &id_priv->id.route.addr.src_addr;
+	cm_id->local_addr = *sin;
+
+	sin = (struct sockaddr_in*) &id_priv->id.route.addr.dst_addr;
+	cm_id->remote_addr = *sin;
+
+	ret = cma_modify_qp_rtr(&id_priv->id);
+	if (ret) {
+		iw_destroy_cm_id(cm_id);
+		return ret;
+	}
+
+	iw_param.ord = conn_param->initiator_depth;
+	iw_param.ird = conn_param->responder_resources;
+	iw_param.private_data = conn_param->private_data;
+	iw_param.private_data_len = conn_param->private_data_len;
+	if (id_priv->id.qp)
+		iw_param.qpn = id_priv->qp_num;
+	else
+		iw_param.qpn = conn_param->qp_num;
+	ret = iw_cm_connect(cm_id, &iw_param);
+out:
+	return ret;
+}
+
 int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param)
 {
 	struct rdma_id_private *id_priv;
@@ -1664,10 +1903,13 @@ int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param)
 		id_priv->srq = conn_param->srq;
 	}
 
-	switch (id->device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id->device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		ret = cma_connect_ib(id_priv, conn_param);
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = cma_connect_iw(id_priv, conn_param);
+		break;
 	default:
 		ret = -ENOSYS;
 		break;
@@ -1708,6 +1950,28 @@ static int cma_accept_ib(struct rdma_id_private *id_priv,
 	return ib_send_cm_rep(id_priv->cm_id.ib, &rep);
 }
 
+static int cma_accept_iw(struct rdma_id_private *id_priv,
+		  struct rdma_conn_param *conn_param)
+{
+	struct iw_cm_conn_param iw_param;
+	int ret;
+
+	ret = cma_modify_qp_rtr(&id_priv->id);
+	if (ret)
+		return ret;
+
+	iw_param.ord = conn_param->initiator_depth;
+	iw_param.ird = conn_param->responder_resources;
+	iw_param.private_data = conn_param->private_data;
+	iw_param.private_data_len = conn_param->private_data_len;
+	if (id_priv->id.qp) {
+		iw_param.qpn = id_priv->qp_num;
+	} else
+		iw_param.qpn = conn_param->qp_num;
+
+	return iw_cm_accept(id_priv->cm_id.iw, &iw_param);
+}
+
 int rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param)
 {
 	struct rdma_id_private *id_priv;
@@ -1723,13 +1987,16 @@ int rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param)
 		id_priv->srq = conn_param->srq;
 	}
 
-	switch (id->device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id->device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		if (conn_param)
 			ret = cma_accept_ib(id_priv, conn_param);
 		else
 			ret = cma_rep_recv(id_priv);
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = cma_accept_iw(id_priv, conn_param);
+		break;
 	default:
 		ret = -ENOSYS;
 		break;
@@ -1756,12 +2023,16 @@ int rdma_reject(struct rdma_cm_id *id, const void *private_data,
 	if (!cma_comp(id_priv, CMA_CONNECT))
 		return -EINVAL;
 
-	switch (id->device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id->device->node_type)) {
+	case RDMA_TRANSPORT_IB:
 		ret = ib_send_cm_rej(id_priv->cm_id.ib,
 				     IB_CM_REJ_CONSUMER_DEFINED, NULL, 0,
 				     private_data, private_data_len);
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = iw_cm_reject(id_priv->cm_id.iw,
+				   private_data, private_data_len);
+		break;
 	default:
 		ret = -ENOSYS;
 		break;
@@ -1780,17 +2051,20 @@ int rdma_disconnect(struct rdma_cm_id *id)
 	    !cma_comp(id_priv, CMA_DISCONNECT))
 		return -EINVAL;
 
-	ret = cma_modify_qp_err(id);
-	if (ret)
-		goto out;
-
-	switch (id->device->node_type) {
-	case IB_NODE_CA:
+	switch (rdma_node_get_transport(id->device->node_type)) {
+	case RDMA_TRANSPORT_IB:
+		ret = cma_modify_qp_err(id);
+		if (ret)
+			goto out;
 		/* Initiate or respond to a disconnect. */
 		if (ib_send_cm_dreq(id_priv->cm_id.ib, NULL, 0))
 			ib_send_cm_drep(id_priv->cm_id.ib, NULL, 0);
 		break;
+	case RDMA_TRANSPORT_IWARP:
+		ret = iw_cm_disconnect(id_priv->cm_id.iw, 0);
+		break;
 	default:
+		ret = -EINVAL;
 		break;
 	}
 out:
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index b2f3cb91d9bc..d978fbe97535 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -505,7 +505,7 @@ int ib_query_port(struct ib_device *device,
 		  u8 port_num,
 		  struct ib_port_attr *port_attr)
 {
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		if (port_num)
 			return -EINVAL;
 	} else if (port_num < 1 || port_num > device->phys_port_cnt)
@@ -580,7 +580,7 @@ int ib_modify_port(struct ib_device *device,
 		   u8 port_num, int port_modify_mask,
 		   struct ib_port_modify *port_modify)
 {
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		if (port_num)
 			return -EINVAL;
 	} else if (port_num < 1 || port_num > device->phys_port_cnt)
diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index 32d3028b274b..082f03c158f0 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -2876,7 +2876,10 @@ static void ib_mad_init_device(struct ib_device *device)
 {
 	int start, end, i;
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		start = 0;
 		end   = 0;
 	} else {
@@ -2923,7 +2926,7 @@ static void ib_mad_remove_device(struct ib_device *device)
 {
 	int i, num_ports, cur_port;
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		num_ports = 1;
 		cur_port = 0;
 	} else {
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index df762ba4868f..ca8760a7d88c 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -919,7 +919,10 @@ static void ib_sa_add_one(struct ib_device *device)
 	struct ib_sa_device *sa_dev;
 	int s, e, i;
 
-	if (device->node_type == IB_NODE_SWITCH)
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
+	if (device->node_type == RDMA_NODE_IB_SWITCH)
 		s = e = 0;
 	else {
 		s = 1;
diff --git a/drivers/infiniband/core/smi.c b/drivers/infiniband/core/smi.c
index 35852e794e26..54b81e17ad50 100644
--- a/drivers/infiniband/core/smi.c
+++ b/drivers/infiniband/core/smi.c
@@ -64,7 +64,7 @@ int smi_handle_dr_smp_send(struct ib_smp *smp,
 
 		/* C14-9:2 */
 		if (hop_ptr && hop_ptr < hop_cnt) {
-			if (node_type != IB_NODE_SWITCH)
+			if (node_type != RDMA_NODE_IB_SWITCH)
 				return 0;
 
 			/* smp->return_path set when received */
@@ -77,7 +77,7 @@ int smi_handle_dr_smp_send(struct ib_smp *smp,
 		if (hop_ptr == hop_cnt) {
 			/* smp->return_path set when received */
 			smp->hop_ptr++;
-			return (node_type == IB_NODE_SWITCH ||
+			return (node_type == RDMA_NODE_IB_SWITCH ||
 				smp->dr_dlid == IB_LID_PERMISSIVE);
 		}
 
@@ -95,7 +95,7 @@ int smi_handle_dr_smp_send(struct ib_smp *smp,
 
 		/* C14-13:2 */
 		if (2 <= hop_ptr && hop_ptr <= hop_cnt) {
-			if (node_type != IB_NODE_SWITCH)
+			if (node_type != RDMA_NODE_IB_SWITCH)
 				return 0;
 
 			smp->hop_ptr--;
@@ -107,7 +107,7 @@ int smi_handle_dr_smp_send(struct ib_smp *smp,
 		if (hop_ptr == 1) {
 			smp->hop_ptr--;
 			/* C14-13:3 -- SMPs destined for SM shouldn't be here */
-			return (node_type == IB_NODE_SWITCH ||
+			return (node_type == RDMA_NODE_IB_SWITCH ||
 				smp->dr_slid == IB_LID_PERMISSIVE);
 		}
 
@@ -142,7 +142,7 @@ int smi_handle_dr_smp_recv(struct ib_smp *smp,
 
 		/* C14-9:2 -- intermediate hop */
 		if (hop_ptr && hop_ptr < hop_cnt) {
-			if (node_type != IB_NODE_SWITCH)
+			if (node_type != RDMA_NODE_IB_SWITCH)
 				return 0;
 
 			smp->return_path[hop_ptr] = port_num;
@@ -156,7 +156,7 @@ int smi_handle_dr_smp_recv(struct ib_smp *smp,
 				smp->return_path[hop_ptr] = port_num;
 			/* smp->hop_ptr updated when sending */
 
-			return (node_type == IB_NODE_SWITCH ||
+			return (node_type == RDMA_NODE_IB_SWITCH ||
 				smp->dr_dlid == IB_LID_PERMISSIVE);
 		}
 
@@ -175,7 +175,7 @@ int smi_handle_dr_smp_recv(struct ib_smp *smp,
 
 		/* C14-13:2 */
 		if (2 <= hop_ptr && hop_ptr <= hop_cnt) {
-			if (node_type != IB_NODE_SWITCH)
+			if (node_type != RDMA_NODE_IB_SWITCH)
 				return 0;
 
 			/* smp->hop_ptr updated when sending */
@@ -190,7 +190,7 @@ int smi_handle_dr_smp_recv(struct ib_smp *smp,
 				return 1;
 			}
 			/* smp->hop_ptr updated when sending */
-			return (node_type == IB_NODE_SWITCH);
+			return (node_type == RDMA_NODE_IB_SWITCH);
 		}
 
 		/* C14-13:4 -- hop_ptr = 0 -> give to SM */
diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index fb6660564a30..709323c14c5d 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -589,10 +589,11 @@ static ssize_t show_node_type(struct class_device *cdev, char *buf)
 		return -ENODEV;
 
 	switch (dev->node_type) {
-	case IB_NODE_CA:     return sprintf(buf, "%d: CA\n", dev->node_type);
-	case IB_NODE_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type);
-	case IB_NODE_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type);
-	default:             return sprintf(buf, "%d: <unknown>\n", dev->node_type);
+	case RDMA_NODE_IB_CA:	  return sprintf(buf, "%d: CA\n", dev->node_type);
+	case RDMA_NODE_RNIC:	  return sprintf(buf, "%d: RNIC\n", dev->node_type);
+	case RDMA_NODE_IB_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type);
+	case RDMA_NODE_IB_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type);
+	default:		  return sprintf(buf, "%d: <unknown>\n", dev->node_type);
 	}
 }
 
@@ -708,7 +709,7 @@ int ib_device_register_sysfs(struct ib_device *device)
 	if (ret)
 		goto err_put;
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		ret = add_port(device, 0);
 		if (ret)
 			goto err_put;
diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c
index e74c964af7fa..ad4f4d5c2924 100644
--- a/drivers/infiniband/core/ucm.c
+++ b/drivers/infiniband/core/ucm.c
@@ -1247,7 +1247,8 @@ static void ib_ucm_add_one(struct ib_device *device)
 {
 	struct ib_ucm_device *ucm_dev;
 
-	if (!device->alloc_ucontext)
+	if (!device->alloc_ucontext ||
+	    rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
 		return;
 
 	ucm_dev = kzalloc(sizeof *ucm_dev, GFP_KERNEL);
diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c
index 8a455aec758f..807fbd6b8414 100644
--- a/drivers/infiniband/core/user_mad.c
+++ b/drivers/infiniband/core/user_mad.c
@@ -1032,7 +1032,10 @@ static void ib_umad_add_one(struct ib_device *device)
 	struct ib_umad_device *umad_dev;
 	int s, e, i;
 
-	if (device->node_type == IB_NODE_SWITCH)
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
+	if (device->node_type == RDMA_NODE_IB_SWITCH)
 		s = e = 0;
 	else {
 		s = 1;
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 06f98e9e14f9..8b5dd3649bbf 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -79,6 +79,23 @@ enum ib_rate mult_to_ib_rate(int mult)
 }
 EXPORT_SYMBOL(mult_to_ib_rate);
 
+enum rdma_transport_type
+rdma_node_get_transport(enum rdma_node_type node_type)
+{
+	switch (node_type) {
+	case RDMA_NODE_IB_CA:
+	case RDMA_NODE_IB_SWITCH:
+	case RDMA_NODE_IB_ROUTER:
+		return RDMA_TRANSPORT_IB;
+	case RDMA_NODE_RNIC:
+		return RDMA_TRANSPORT_IWARP;
+	default:
+		BUG();
+		return 0;
+	}
+}
+EXPORT_SYMBOL(rdma_node_get_transport);
+
 /* Protection domains */
 
 struct ib_pd *ib_alloc_pd(struct ib_device *device)
diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c
index 2a99f2d13cdb..2380994418a5 100644
--- a/drivers/infiniband/hw/ehca/ehca_main.c
+++ b/drivers/infiniband/hw/ehca/ehca_main.c
@@ -269,7 +269,7 @@ int ehca_register_device(struct ehca_shca *shca)
 		(1ull << IB_USER_VERBS_CMD_ATTACH_MCAST)	|
 		(1ull << IB_USER_VERBS_CMD_DETACH_MCAST);
 
-	shca->ib_device.node_type           = IB_NODE_CA;
+	shca->ib_device.node_type           = RDMA_NODE_IB_CA;
 	shca->ib_device.phys_port_cnt       = shca->num_ports;
 	shca->ib_device.dma_device          = &shca->ibmebus_dev->ofdev.dev;
 	shca->ib_device.query_device        = ehca_query_device;
diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.c b/drivers/infiniband/hw/ipath/ipath_verbs.c
index fbda7739715f..b8381c5e72bd 100644
--- a/drivers/infiniband/hw/ipath/ipath_verbs.c
+++ b/drivers/infiniband/hw/ipath/ipath_verbs.c
@@ -1538,7 +1538,7 @@ int ipath_register_ib_device(struct ipath_devdata *dd)
 		(1ull << IB_USER_VERBS_CMD_QUERY_SRQ)		|
 		(1ull << IB_USER_VERBS_CMD_DESTROY_SRQ)		|
 		(1ull << IB_USER_VERBS_CMD_POST_SRQ_RECV);
-	dev->node_type = IB_NODE_CA;
+	dev->node_type = RDMA_NODE_IB_CA;
 	dev->phys_port_cnt = 1;
 	dev->dma_device = &dd->pcidev->dev;
 	dev->class_dev.dev = dev->dma_device;
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c
index 265b1d1c4a62..981fe2eebdfa 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.c
+++ b/drivers/infiniband/hw/mthca/mthca_provider.c
@@ -1288,7 +1288,7 @@ int mthca_register_device(struct mthca_dev *dev)
 		(1ull << IB_USER_VERBS_CMD_DESTROY_QP)		|
 		(1ull << IB_USER_VERBS_CMD_ATTACH_MCAST)	|
 		(1ull << IB_USER_VERBS_CMD_DETACH_MCAST);
-	dev->ib_dev.node_type            = IB_NODE_CA;
+	dev->ib_dev.node_type            = RDMA_NODE_IB_CA;
 	dev->ib_dev.phys_port_cnt        = dev->limits.num_ports;
 	dev->ib_dev.dma_device           = &dev->pdev->dev;
 	dev->ib_dev.class_dev.dev        = &dev->pdev->dev;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 36d76987a481..e9a7659eb1d7 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -1111,13 +1111,16 @@ static void ipoib_add_one(struct ib_device *device)
 	struct ipoib_dev_priv *priv;
 	int s, e, p;
 
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
 	dev_list = kmalloc(sizeof *dev_list, GFP_KERNEL);
 	if (!dev_list)
 		return;
 
 	INIT_LIST_HEAD(dev_list);
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		s = 0;
 		e = 0;
 	} else {
@@ -1141,6 +1144,9 @@ static void ipoib_remove_one(struct ib_device *device)
 	struct ipoib_dev_priv *priv, *tmp;
 	struct list_head *dev_list;
 
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
 	dev_list = ib_get_client_data(device, &ipoib_client);
 
 	list_for_each_entry_safe(priv, tmp, dev_list, list) {
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index 61c13d1e0506..feb1fcd0f2fb 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -1899,7 +1899,7 @@ static void srp_add_one(struct ib_device *device)
 	if (IS_ERR(srp_dev->fmr_pool))
 		srp_dev->fmr_pool = NULL;
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		s = 0;
 		e = 0;
 	} else {
diff --git a/include/rdma/ib_addr.h b/include/rdma/ib_addr.h
index 0ff67398928d..81b62307621d 100644
--- a/include/rdma/ib_addr.h
+++ b/include/rdma/ib_addr.h
@@ -40,7 +40,7 @@ struct rdma_dev_addr {
 	unsigned char src_dev_addr[MAX_ADDR_LEN];
 	unsigned char dst_dev_addr[MAX_ADDR_LEN];
 	unsigned char broadcast[MAX_ADDR_LEN];
-	enum ib_node_type dev_type;
+	enum rdma_node_type dev_type;
 };
 
 /**
@@ -72,6 +72,9 @@ int rdma_resolve_ip(struct sockaddr *src_addr, struct sockaddr *dst_addr,
 
 void rdma_addr_cancel(struct rdma_dev_addr *addr);
 
+int rdma_copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev,
+	      const unsigned char *dst_dev_addr);
+
 static inline int ip_addr_size(struct sockaddr *addr)
 {
 	return addr->sa_family == AF_INET6 ?
@@ -113,4 +116,16 @@ static inline void ib_addr_set_dgid(struct rdma_dev_addr *dev_addr,
 	memcpy(dev_addr->dst_dev_addr + 4, gid, sizeof *gid);
 }
 
+static inline void iw_addr_get_sgid(struct rdma_dev_addr *dev_addr,
+				    union ib_gid *gid)
+{
+	memcpy(gid, dev_addr->src_dev_addr, sizeof *gid);
+}
+
+static inline void iw_addr_get_dgid(struct rdma_dev_addr *dev_addr,
+				    union ib_gid *gid)
+{
+	memcpy(gid, dev_addr->dst_dev_addr, sizeof *gid);
+}
+
 #endif /* IB_ADDR_H */
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 61eed3996117..8eacc3510993 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -56,12 +56,22 @@ union ib_gid {
 	} global;
 };
 
-enum ib_node_type {
-	IB_NODE_CA 	= 1,
-	IB_NODE_SWITCH,
-	IB_NODE_ROUTER
+enum rdma_node_type {
+	/* IB values map to NodeInfo:NodeType. */
+	RDMA_NODE_IB_CA 	= 1,
+	RDMA_NODE_IB_SWITCH,
+	RDMA_NODE_IB_ROUTER,
+	RDMA_NODE_RNIC
 };
 
+enum rdma_transport_type {
+	RDMA_TRANSPORT_IB,
+	RDMA_TRANSPORT_IWARP
+};
+
+enum rdma_transport_type
+rdma_node_get_transport(enum rdma_node_type node_type) __attribute_const__;
+
 enum ib_device_cap_flags {
 	IB_DEVICE_RESIZE_MAX_WR		= 1,
 	IB_DEVICE_BAD_PKEY_CNTR		= (1<<1),
@@ -78,6 +88,9 @@ enum ib_device_cap_flags {
 	IB_DEVICE_RC_RNR_NAK_GEN	= (1<<12),
 	IB_DEVICE_SRQ_RESIZE		= (1<<13),
 	IB_DEVICE_N_NOTIFY_CQ		= (1<<14),
+	IB_DEVICE_ZERO_STAG		= (1<<15),
+	IB_DEVICE_SEND_W_INV		= (1<<16),
+	IB_DEVICE_MEM_WINDOW		= (1<<17)
 };
 
 enum ib_atomic_cap {
@@ -835,6 +848,8 @@ struct ib_cache {
 	u8                     *lmc_cache;
 };
 
+struct iw_cm_verbs;
+
 struct ib_device {
 	struct device                *dma_device;
 
@@ -851,6 +866,8 @@ struct ib_device {
 
 	u32                           flags;
 
+	struct iw_cm_verbs	     *iwcm;
+
 	int		           (*query_device)(struct ib_device *device,
 						   struct ib_device_attr *device_attr);
 	int		           (*query_port)(struct ib_device *device,
-- 
cgit v1.2.3


From c1a0b23bf477c2e1068905f4e2b5c3cee139e853 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Mon, 21 Aug 2006 16:40:12 -0700
Subject: IB/sa: Require SA registration

Require users to register with SA module, to prevent the sa_query
module text from going away while an SA query callback is still
running.  Update all in-tree users for the new interface.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/cma.c                  |  7 ++-
 drivers/infiniband/core/sa_query.c             | 60 ++++++++++++++++++++++----
 drivers/infiniband/ulp/ipoib/ipoib.h           |  2 +
 drivers/infiniband/ulp/ipoib/ipoib_main.c      | 12 ++++--
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 12 +++---
 drivers/infiniband/ulp/srp/ib_srp.c            |  9 +++-
 include/rdma/ib_sa.h                           | 41 +++++++++++++++---
 7 files changed, 116 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 488fa1d642a7..1178bd434d1b 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -62,6 +62,7 @@ static struct ib_client cma_client = {
 	.remove = cma_remove_one
 };
 
+static struct ib_sa_client sa_client;
 static LIST_HEAD(dev_list);
 static LIST_HEAD(listen_any_list);
 static DEFINE_MUTEX(lock);
@@ -1323,7 +1324,7 @@ static int cma_query_ib_route(struct rdma_id_private *id_priv, int timeout_ms,
 	path_rec.pkey = cpu_to_be16(ib_addr_get_pkey(addr));
 	path_rec.numb_path = 1;
 
-	id_priv->query_id = ib_sa_path_rec_get(id_priv->id.device,
+	id_priv->query_id = ib_sa_path_rec_get(&sa_client, id_priv->id.device,
 				id_priv->id.port_num, &path_rec,
 				IB_SA_PATH_REC_DGID | IB_SA_PATH_REC_SGID |
 				IB_SA_PATH_REC_PKEY | IB_SA_PATH_REC_NUMB_PATH,
@@ -2199,12 +2200,15 @@ static int cma_init(void)
 	if (!cma_wq)
 		return -ENOMEM;
 
+	ib_sa_register_client(&sa_client);
+
 	ret = ib_register_client(&cma_client);
 	if (ret)
 		goto err;
 	return 0;
 
 err:
+	ib_sa_unregister_client(&sa_client);
 	destroy_workqueue(cma_wq);
 	return ret;
 }
@@ -2212,6 +2216,7 @@ err:
 static void cma_cleanup(void)
 {
 	ib_unregister_client(&cma_client);
+	ib_sa_unregister_client(&sa_client);
 	destroy_workqueue(cma_wq);
 	idr_destroy(&sdp_ps);
 	idr_destroy(&tcp_ps);
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index ca8760a7d88c..1706d3c7e95e 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -1,6 +1,7 @@
 /*
  * Copyright (c) 2004 Topspin Communications.  All rights reserved.
  * Copyright (c) 2005 Voltaire, Inc.  All rights reserved.
+ * Copyright (c) 2006 Intel Corporation.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -75,6 +76,7 @@ struct ib_sa_device {
 struct ib_sa_query {
 	void (*callback)(struct ib_sa_query *, int, struct ib_sa_mad *);
 	void (*release)(struct ib_sa_query *);
+	struct ib_sa_client    *client;
 	struct ib_sa_port      *port;
 	struct ib_mad_send_buf *mad_buf;
 	struct ib_sa_sm_ah     *sm_ah;
@@ -415,6 +417,31 @@ static void ib_sa_event(struct ib_event_handler *handler, struct ib_event *event
 	}
 }
 
+void ib_sa_register_client(struct ib_sa_client *client)
+{
+	atomic_set(&client->users, 1);
+	init_completion(&client->comp);
+}
+EXPORT_SYMBOL(ib_sa_register_client);
+
+static inline void ib_sa_client_get(struct ib_sa_client *client)
+{
+	atomic_inc(&client->users);
+}
+
+static inline void ib_sa_client_put(struct ib_sa_client *client)
+{
+	if (atomic_dec_and_test(&client->users))
+		complete(&client->comp);
+}
+
+void ib_sa_unregister_client(struct ib_sa_client *client)
+{
+	ib_sa_client_put(client);
+	wait_for_completion(&client->comp);
+}
+EXPORT_SYMBOL(ib_sa_unregister_client);
+
 /**
  * ib_sa_cancel_query - try to cancel an SA query
  * @id:ID of query to cancel
@@ -557,6 +584,7 @@ static void ib_sa_path_rec_release(struct ib_sa_query *sa_query)
 
 /**
  * ib_sa_path_rec_get - Start a Path get query
+ * @client:SA client
  * @device:device to send query on
  * @port_num: port number to send query on
  * @rec:Path Record to send in query
@@ -579,7 +607,8 @@ static void ib_sa_path_rec_release(struct ib_sa_query *sa_query)
  * error code.  Otherwise it is a query ID that can be used to cancel
  * the query.
  */
-int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
+int ib_sa_path_rec_get(struct ib_sa_client *client,
+		       struct ib_device *device, u8 port_num,
 		       struct ib_sa_path_rec *rec,
 		       ib_sa_comp_mask comp_mask,
 		       int timeout_ms, gfp_t gfp_mask,
@@ -614,8 +643,10 @@ int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
 		goto err1;
 	}
 
-	query->callback = callback;
-	query->context  = context;
+	ib_sa_client_get(client);
+	query->sa_query.client = client;
+	query->callback        = callback;
+	query->context         = context;
 
 	mad = query->sa_query.mad_buf->mad;
 	init_mad(mad, agent);
@@ -639,6 +670,7 @@ int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
 
 err2:
 	*sa_query = NULL;
+	ib_sa_client_put(query->sa_query.client);
 	ib_free_send_mad(query->sa_query.mad_buf);
 
 err1:
@@ -671,6 +703,7 @@ static void ib_sa_service_rec_release(struct ib_sa_query *sa_query)
 
 /**
  * ib_sa_service_rec_query - Start Service Record operation
+ * @client:SA client
  * @device:device to send request on
  * @port_num: port number to send request on
  * @method:SA method - should be get, set, or delete
@@ -695,7 +728,8 @@ static void ib_sa_service_rec_release(struct ib_sa_query *sa_query)
  * error code.  Otherwise it is a request ID that can be used to cancel
  * the query.
  */
-int ib_sa_service_rec_query(struct ib_device *device, u8 port_num, u8 method,
+int ib_sa_service_rec_query(struct ib_sa_client *client,
+			    struct ib_device *device, u8 port_num, u8 method,
 			    struct ib_sa_service_rec *rec,
 			    ib_sa_comp_mask comp_mask,
 			    int timeout_ms, gfp_t gfp_mask,
@@ -735,8 +769,10 @@ int ib_sa_service_rec_query(struct ib_device *device, u8 port_num, u8 method,
 		goto err1;
 	}
 
-	query->callback = callback;
-	query->context  = context;
+	ib_sa_client_get(client);
+	query->sa_query.client = client;
+	query->callback        = callback;
+	query->context         = context;
 
 	mad = query->sa_query.mad_buf->mad;
 	init_mad(mad, agent);
@@ -761,6 +797,7 @@ int ib_sa_service_rec_query(struct ib_device *device, u8 port_num, u8 method,
 
 err2:
 	*sa_query = NULL;
+	ib_sa_client_put(query->sa_query.client);
 	ib_free_send_mad(query->sa_query.mad_buf);
 
 err1:
@@ -791,7 +828,8 @@ static void ib_sa_mcmember_rec_release(struct ib_sa_query *sa_query)
 	kfree(container_of(sa_query, struct ib_sa_mcmember_query, sa_query));
 }
 
-int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
+int ib_sa_mcmember_rec_query(struct ib_sa_client *client,
+			     struct ib_device *device, u8 port_num,
 			     u8 method,
 			     struct ib_sa_mcmember_rec *rec,
 			     ib_sa_comp_mask comp_mask,
@@ -827,8 +865,10 @@ int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
 		goto err1;
 	}
 
-	query->callback = callback;
-	query->context  = context;
+	ib_sa_client_get(client);
+	query->sa_query.client = client;
+	query->callback        = callback;
+	query->context         = context;
 
 	mad = query->sa_query.mad_buf->mad;
 	init_mad(mad, agent);
@@ -853,6 +893,7 @@ int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
 
 err2:
 	*sa_query = NULL;
+	ib_sa_client_put(query->sa_query.client);
 	ib_free_send_mad(query->sa_query.mad_buf);
 
 err1:
@@ -889,6 +930,7 @@ static void send_handler(struct ib_mad_agent *agent,
 
 	ib_free_send_mad(mad_send_wc->send_buf);
 	kref_put(&query->sm_ah->ref, free_sm_ah);
+	ib_sa_client_put(query->client);
 	query->release(query);
 }
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index 474aa214ab57..0b8a79d53a00 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -336,6 +336,8 @@ static inline void ipoib_unregister_debugfs(void) { }
 extern int ipoib_sendq_size;
 extern int ipoib_recvq_size;
 
+extern struct ib_sa_client ipoib_sa_client;
+
 #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG
 extern int ipoib_debug_level;
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index e9a7659eb1d7..ae3a4982cddb 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -82,6 +82,8 @@ static const u8 ipv4_bcast_addr[] = {
 
 struct workqueue_struct *ipoib_workqueue;
 
+struct ib_sa_client ipoib_sa_client;
+
 static void ipoib_add_one(struct ib_device *device);
 static void ipoib_remove_one(struct ib_device *device);
 
@@ -463,7 +465,7 @@ static int path_rec_start(struct net_device *dev,
 	init_completion(&path->done);
 
 	path->query_id =
-		ib_sa_path_rec_get(priv->ca, priv->port,
+		ib_sa_path_rec_get(&ipoib_sa_client, priv->ca, priv->port,
 				   &path->pathrec,
 				   IB_SA_PATH_REC_DGID		|
 				   IB_SA_PATH_REC_SGID		|
@@ -1191,13 +1193,16 @@ static int __init ipoib_init_module(void)
 		goto err_fs;
 	}
 
+	ib_sa_register_client(&ipoib_sa_client);
+
 	ret = ib_register_client(&ipoib_client);
 	if (ret)
-		goto err_wq;
+		goto err_sa;
 
 	return 0;
 
-err_wq:
+err_sa:
+	ib_sa_unregister_client(&ipoib_sa_client);
 	destroy_workqueue(ipoib_workqueue);
 
 err_fs:
@@ -1209,6 +1214,7 @@ err_fs:
 static void __exit ipoib_cleanup_module(void)
 {
 	ib_unregister_client(&ipoib_client);
+	ib_sa_unregister_client(&ipoib_sa_client);
 	ipoib_unregister_debugfs();
 	destroy_workqueue(ipoib_workqueue);
 }
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 60b09f5cb347..fb3e4875a46d 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -361,7 +361,7 @@ static int ipoib_mcast_sendonly_join(struct ipoib_mcast *mcast)
 
 	init_completion(&mcast->done);
 
-	ret = ib_sa_mcmember_rec_set(priv->ca, priv->port, &rec,
+	ret = ib_sa_mcmember_rec_set(&ipoib_sa_client, priv->ca, priv->port, &rec,
 				     IB_SA_MCMEMBER_REC_MGID		|
 				     IB_SA_MCMEMBER_REC_PORT_GID	|
 				     IB_SA_MCMEMBER_REC_PKEY		|
@@ -485,9 +485,9 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast,
 
 	init_completion(&mcast->done);
 
-	ret = ib_sa_mcmember_rec_set(priv->ca, priv->port, &rec, comp_mask,
-				     mcast->backoff * 1000, GFP_ATOMIC,
-				     ipoib_mcast_join_complete,
+	ret = ib_sa_mcmember_rec_set(&ipoib_sa_client, priv->ca, priv->port,
+				     &rec, comp_mask, mcast->backoff * 1000,
+				     GFP_ATOMIC, ipoib_mcast_join_complete,
 				     mcast, &mcast->query);
 
 	if (ret < 0) {
@@ -528,7 +528,7 @@ void ipoib_mcast_join_task(void *dev_ptr)
 			priv->local_rate = attr.active_speed *
 				ib_width_enum_to_int(attr.active_width);
 		} else
-			ipoib_warn(priv, "ib_query_port failed\n");
+		ipoib_warn(priv, "ib_query_port failed\n");
 	}
 
 	if (!priv->broadcast) {
@@ -681,7 +681,7 @@ static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast)
 	 * Just make one shot at leaving and don't wait for a reply;
 	 * if we fail, too bad.
 	 */
-	ret = ib_sa_mcmember_rec_delete(priv->ca, priv->port, &rec,
+	ret = ib_sa_mcmember_rec_delete(&ipoib_sa_client, priv->ca, priv->port, &rec,
 					IB_SA_MCMEMBER_REC_MGID		|
 					IB_SA_MCMEMBER_REC_PORT_GID	|
 					IB_SA_MCMEMBER_REC_PKEY		|
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index feb1fcd0f2fb..44b9e5be6687 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -96,6 +96,8 @@ static struct ib_client srp_client = {
 	.remove = srp_remove_one
 };
 
+static struct ib_sa_client srp_sa_client;
+
 static inline struct srp_target_port *host_to_target(struct Scsi_Host *host)
 {
 	return (struct srp_target_port *) host->hostdata;
@@ -267,7 +269,8 @@ static int srp_lookup_path(struct srp_target_port *target)
 
 	init_completion(&target->done);
 
-	target->path_query_id = ib_sa_path_rec_get(target->srp_host->dev->dev,
+	target->path_query_id = ib_sa_path_rec_get(&srp_sa_client,
+						   target->srp_host->dev->dev,
 						   target->srp_host->port,
 						   &target->path,
 						   IB_SA_PATH_REC_DGID		|
@@ -1998,9 +2001,12 @@ static int __init srp_init_module(void)
 		return ret;
 	}
 
+	ib_sa_register_client(&srp_sa_client);
+
 	ret = ib_register_client(&srp_client);
 	if (ret) {
 		printk(KERN_ERR PFX "couldn't register IB client\n");
+		ib_sa_unregister_client(&srp_sa_client);
 		class_unregister(&srp_class);
 		return ret;
 	}
@@ -2011,6 +2017,7 @@ static int __init srp_init_module(void)
 static void __exit srp_cleanup_module(void)
 {
 	ib_unregister_client(&srp_client);
+	ib_sa_unregister_client(&srp_sa_client);
 	class_unregister(&srp_class);
 }
 
diff --git a/include/rdma/ib_sa.h b/include/rdma/ib_sa.h
index c99e4420fd7e..58bb5f716fe3 100644
--- a/include/rdma/ib_sa.h
+++ b/include/rdma/ib_sa.h
@@ -1,6 +1,7 @@
 /*
  * Copyright (c) 2004 Topspin Communications.  All rights reserved.
  * Copyright (c) 2005 Voltaire, Inc.  All rights reserved.
+ * Copyright (c) 2006 Intel Corporation.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -36,8 +37,11 @@
 #ifndef IB_SA_H
 #define IB_SA_H
 
+#include <linux/completion.h>
 #include <linux/compiler.h>
 
+#include <asm/atomic.h>
+
 #include <rdma/ib_verbs.h>
 #include <rdma/ib_mad.h>
 
@@ -250,11 +254,28 @@ struct ib_sa_service_rec {
 	u64		data64[2];
 };
 
+struct ib_sa_client {
+	atomic_t users;
+	struct completion comp;
+};
+
+/**
+ * ib_sa_register_client - Register an SA client.
+ */
+void ib_sa_register_client(struct ib_sa_client *client);
+
+/**
+ * ib_sa_unregister_client - Deregister an SA client.
+ * @client: Client object to deregister.
+ */
+void ib_sa_unregister_client(struct ib_sa_client *client);
+
 struct ib_sa_query;
 
 void ib_sa_cancel_query(int id, struct ib_sa_query *query);
 
-int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
+int ib_sa_path_rec_get(struct ib_sa_client *client,
+		       struct ib_device *device, u8 port_num,
 		       struct ib_sa_path_rec *rec,
 		       ib_sa_comp_mask comp_mask,
 		       int timeout_ms, gfp_t gfp_mask,
@@ -264,7 +285,8 @@ int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
 		       void *context,
 		       struct ib_sa_query **query);
 
-int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
+int ib_sa_mcmember_rec_query(struct ib_sa_client *client,
+			     struct ib_device *device, u8 port_num,
 			     u8 method,
 			     struct ib_sa_mcmember_rec *rec,
 			     ib_sa_comp_mask comp_mask,
@@ -275,7 +297,8 @@ int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
 			     void *context,
 			     struct ib_sa_query **query);
 
-int ib_sa_service_rec_query(struct ib_device *device, u8 port_num,
+int ib_sa_service_rec_query(struct ib_sa_client *client,
+			 struct ib_device *device, u8 port_num,
 			 u8 method,
 			 struct ib_sa_service_rec *rec,
 			 ib_sa_comp_mask comp_mask,
@@ -288,6 +311,7 @@ int ib_sa_service_rec_query(struct ib_device *device, u8 port_num,
 
 /**
  * ib_sa_mcmember_rec_set - Start an MCMember set query
+ * @client:SA client
  * @device:device to send query on
  * @port_num: port number to send query on
  * @rec:MCMember Record to send in query
@@ -311,7 +335,8 @@ int ib_sa_service_rec_query(struct ib_device *device, u8 port_num,
  * cancel the query.
  */
 static inline int
-ib_sa_mcmember_rec_set(struct ib_device *device, u8 port_num,
+ib_sa_mcmember_rec_set(struct ib_sa_client *client,
+		       struct ib_device *device, u8 port_num,
 		       struct ib_sa_mcmember_rec *rec,
 		       ib_sa_comp_mask comp_mask,
 		       int timeout_ms, gfp_t gfp_mask,
@@ -321,7 +346,7 @@ ib_sa_mcmember_rec_set(struct ib_device *device, u8 port_num,
 		       void *context,
 		       struct ib_sa_query **query)
 {
-	return ib_sa_mcmember_rec_query(device, port_num,
+	return ib_sa_mcmember_rec_query(client, device, port_num,
 					IB_MGMT_METHOD_SET,
 					rec, comp_mask,
 					timeout_ms, gfp_mask, callback,
@@ -330,6 +355,7 @@ ib_sa_mcmember_rec_set(struct ib_device *device, u8 port_num,
 
 /**
  * ib_sa_mcmember_rec_delete - Start an MCMember delete query
+ * @client:SA client
  * @device:device to send query on
  * @port_num: port number to send query on
  * @rec:MCMember Record to send in query
@@ -353,7 +379,8 @@ ib_sa_mcmember_rec_set(struct ib_device *device, u8 port_num,
  * cancel the query.
  */
 static inline int
-ib_sa_mcmember_rec_delete(struct ib_device *device, u8 port_num,
+ib_sa_mcmember_rec_delete(struct ib_sa_client *client,
+			  struct ib_device *device, u8 port_num,
 			  struct ib_sa_mcmember_rec *rec,
 			  ib_sa_comp_mask comp_mask,
 			  int timeout_ms, gfp_t gfp_mask,
@@ -363,7 +390,7 @@ ib_sa_mcmember_rec_delete(struct ib_device *device, u8 port_num,
 			  void *context,
 			  struct ib_sa_query **query)
 {
-	return ib_sa_mcmember_rec_query(device, port_num,
+	return ib_sa_mcmember_rec_query(client, device, port_num,
 					IB_SA_METHOD_DELETE,
 					rec, comp_mask,
 					timeout_ms, gfp_mask, callback,
-- 
cgit v1.2.3


From 07eeec0627e93a1a753c4df004a97a4d0a7b9ceb Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Tue, 12 Sep 2006 09:03:33 -0700
Subject: RDMA/cma: Document rdma_destroy_id() function

Clarify that rdma_destroy_id cancels outstanding asynchronous operations on the
Associated id.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 include/rdma/rdma_cm.h | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'include')

diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h
index 402c63d7226b..1566be568ab6 100644
--- a/include/rdma/rdma_cm.h
+++ b/include/rdma/rdma_cm.h
@@ -117,6 +117,14 @@ struct rdma_cm_id {
 struct rdma_cm_id *rdma_create_id(rdma_cm_event_handler event_handler,
 				  void *context, enum rdma_port_space ps);
 
+/**
+  * rdma_destroy_id - Destroys an RDMA identifier.
+  *
+  * @id: RDMA identifier.
+  *
+  * Note: calling this function has the effect of canceling in-flight
+  * asynchronous operations associated with the id.
+  */
 void rdma_destroy_id(struct rdma_cm_id *id);
 
 /**
-- 
cgit v1.2.3


From 951f7fc1372da3d826b1d975b3cc5e3db92af5d0 Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Fri, 22 Sep 2006 15:22:54 -0700
Subject: RDMA/cma: Document rdma_accept() error handling

Document the reject sending and modifying QP to error done in rdma_accept().

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 include/rdma/rdma_cm.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h
index 1566be568ab6..deb5a0a4cee5 100644
--- a/include/rdma/rdma_cm.h
+++ b/include/rdma/rdma_cm.h
@@ -245,6 +245,10 @@ int rdma_listen(struct rdma_cm_id *id, int backlog);
  * Typically, this routine is only called by the listener to accept a connection
  * request.  It must also be called on the active side of a connection if the
  * user is performing their own QP transitions.
+ *
+ * In the case of error, a reject message is sent to the remote side and the
+ * state of the qp associated with the id is modified to error, such that any
+ * previously posted receive buffers would be flushed.
  */
 int rdma_accept(struct rdma_cm_id *id, struct rdma_conn_param *conn_param);
 
-- 
cgit v1.2.3


From aec79fcc3ea3b536a2788b4e22b7ebabbb176485 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Mon, 18 Sep 2006 22:17:08 +0300
Subject: IB/sa: fix ib_sa_selector names

Relevant SA queries are actually "greater than" / "less than", not
"greater than or equal" / "less than or equal" as the names imply.
(See IB spec 1.2 Vol 1, 15.2.5.16 PATHRECORD/Table 205 PathRecord)

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 include/rdma/ib_sa.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/rdma/ib_sa.h b/include/rdma/ib_sa.h
index 58bb5f716fe3..97715b0c20b6 100644
--- a/include/rdma/ib_sa.h
+++ b/include/rdma/ib_sa.h
@@ -83,8 +83,8 @@ enum {
 };
 
 enum ib_sa_selector {
-	IB_SA_GTE  = 0,
-	IB_SA_LTE  = 1,
+	IB_SA_GT   = 0,
+	IB_SA_LT   = 1,
 	IB_SA_EQ   = 2,
 	/*
 	 * The meaning of "best" depends on the attribute: for
-- 
cgit v1.2.3


From 1d22e05df8183b36b3cc0760344774040abc74d5 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Fri, 22 Sep 2006 02:28:13 +0200
Subject: [PATCH] Cirrus Logic ep93xx ethernet driver

The Cirrus Logic ep93xx is an ARM SoC that includes an ethernet MAC
-- this patch adds a driver for that ethernet MAC.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/arm/Kconfig                   |   7 +
 drivers/net/arm/Makefile                  |   1 +
 drivers/net/arm/ep93xx_eth.c              | 944 ++++++++++++++++++++++++++++++
 include/asm-arm/arch-ep93xx/ep93xx-regs.h |   1 +
 include/asm-arm/arch-ep93xx/platform.h    |   6 +
 5 files changed, 959 insertions(+)
 create mode 100644 drivers/net/arm/ep93xx_eth.c

(limited to 'include')

diff --git a/drivers/net/arm/Kconfig b/drivers/net/arm/Kconfig
index 77fe20dbea35..678e4f48d36b 100644
--- a/drivers/net/arm/Kconfig
+++ b/drivers/net/arm/Kconfig
@@ -39,3 +39,10 @@ config ARM_AT91_ETHER
 	help
 	  If you wish to compile a kernel for the AT91RM9200 and enable
 	  ethernet support, then you should always answer Y to this.
+
+config EP93XX_ETH
+	tristate "EP93xx Ethernet support"
+	depends on NET_ETHERNET && ARM && ARCH_EP93XX
+	help
+	  This is a driver for the ethernet hardware included in EP93xx CPUs.
+	  Say Y if you are building a kernel for EP93xx based devices.
diff --git a/drivers/net/arm/Makefile b/drivers/net/arm/Makefile
index 42c95b79c261..a4c868278e11 100644
--- a/drivers/net/arm/Makefile
+++ b/drivers/net/arm/Makefile
@@ -8,3 +8,4 @@ obj-$(CONFIG_ARM_ETHERH)	+= etherh.o
 obj-$(CONFIG_ARM_ETHER3)	+= ether3.o
 obj-$(CONFIG_ARM_ETHER1)	+= ether1.o
 obj-$(CONFIG_ARM_AT91_ETHER)	+= at91_ether.o
+obj-$(CONFIG_EP93XX_ETH)	+= ep93xx_eth.o
diff --git a/drivers/net/arm/ep93xx_eth.c b/drivers/net/arm/ep93xx_eth.c
new file mode 100644
index 000000000000..cef00744a9dc
--- /dev/null
+++ b/drivers/net/arm/ep93xx_eth.c
@@ -0,0 +1,944 @@
+/*
+ * EP93xx ethernet network device driver
+ * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
+ * Dedicated to Marija Kulikova.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/config.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/mii.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/init.h>
+#include <linux/moduleparam.h>
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+#include <asm/arch/ep93xx-regs.h>
+#include <asm/arch/platform.h>
+#include <asm/io.h>
+
+#define DRV_MODULE_NAME		"ep93xx-eth"
+#define DRV_MODULE_VERSION	"0.1"
+
+#define RX_QUEUE_ENTRIES	64
+#define TX_QUEUE_ENTRIES	8
+
+#define MAX_PKT_SIZE		2044
+#define PKT_BUF_SIZE		2048
+
+#define REG_RXCTL		0x0000
+#define  REG_RXCTL_DEFAULT	0x00073800
+#define REG_TXCTL		0x0004
+#define  REG_TXCTL_ENABLE	0x00000001
+#define REG_MIICMD		0x0010
+#define  REG_MIICMD_READ	0x00008000
+#define  REG_MIICMD_WRITE	0x00004000
+#define REG_MIIDATA		0x0014
+#define REG_MIISTS		0x0018
+#define  REG_MIISTS_BUSY	0x00000001
+#define REG_SELFCTL		0x0020
+#define  REG_SELFCTL_RESET	0x00000001
+#define REG_INTEN		0x0024
+#define  REG_INTEN_TX		0x00000008
+#define  REG_INTEN_RX		0x00000007
+#define REG_INTSTSP		0x0028
+#define  REG_INTSTS_TX		0x00000008
+#define  REG_INTSTS_RX		0x00000004
+#define REG_INTSTSC		0x002c
+#define REG_AFP			0x004c
+#define REG_INDAD0		0x0050
+#define REG_INDAD1		0x0051
+#define REG_INDAD2		0x0052
+#define REG_INDAD3		0x0053
+#define REG_INDAD4		0x0054
+#define REG_INDAD5		0x0055
+#define REG_GIINTMSK		0x0064
+#define  REG_GIINTMSK_ENABLE	0x00008000
+#define REG_BMCTL		0x0080
+#define  REG_BMCTL_ENABLE_TX	0x00000100
+#define  REG_BMCTL_ENABLE_RX	0x00000001
+#define REG_BMSTS		0x0084
+#define  REG_BMSTS_RX_ACTIVE	0x00000008
+#define REG_RXDQBADD		0x0090
+#define REG_RXDQBLEN		0x0094
+#define REG_RXDCURADD		0x0098
+#define REG_RXDENQ		0x009c
+#define REG_RXSTSQBADD		0x00a0
+#define REG_RXSTSQBLEN		0x00a4
+#define REG_RXSTSQCURADD	0x00a8
+#define REG_RXSTSENQ		0x00ac
+#define REG_TXDQBADD		0x00b0
+#define REG_TXDQBLEN		0x00b4
+#define REG_TXDQCURADD		0x00b8
+#define REG_TXDENQ		0x00bc
+#define REG_TXSTSQBADD		0x00c0
+#define REG_TXSTSQBLEN		0x00c4
+#define REG_TXSTSQCURADD	0x00c8
+#define REG_MAXFRMLEN		0x00e8
+
+struct ep93xx_rdesc
+{
+	u32	buf_addr;
+	u32	rdesc1;
+};
+
+#define RDESC1_NSOF		0x80000000
+#define RDESC1_BUFFER_INDEX	0x7fff0000
+#define RDESC1_BUFFER_LENGTH	0x0000ffff
+
+struct ep93xx_rstat
+{
+	u32	rstat0;
+	u32	rstat1;
+};
+
+#define RSTAT0_RFP		0x80000000
+#define RSTAT0_RWE		0x40000000
+#define RSTAT0_EOF		0x20000000
+#define RSTAT0_EOB		0x10000000
+#define RSTAT0_AM		0x00c00000
+#define RSTAT0_RX_ERR		0x00200000
+#define RSTAT0_OE		0x00100000
+#define RSTAT0_FE		0x00080000
+#define RSTAT0_RUNT		0x00040000
+#define RSTAT0_EDATA		0x00020000
+#define RSTAT0_CRCE		0x00010000
+#define RSTAT0_CRCI		0x00008000
+#define RSTAT0_HTI		0x00003f00
+#define RSTAT1_RFP		0x80000000
+#define RSTAT1_BUFFER_INDEX	0x7fff0000
+#define RSTAT1_FRAME_LENGTH	0x0000ffff
+
+struct ep93xx_tdesc
+{
+	u32	buf_addr;
+	u32	tdesc1;
+};
+
+#define TDESC1_EOF		0x80000000
+#define TDESC1_BUFFER_INDEX	0x7fff0000
+#define TDESC1_BUFFER_ABORT	0x00008000
+#define TDESC1_BUFFER_LENGTH	0x00000fff
+
+struct ep93xx_tstat
+{
+	u32	tstat0;
+};
+
+#define TSTAT0_TXFP		0x80000000
+#define TSTAT0_TXWE		0x40000000
+#define TSTAT0_FA		0x20000000
+#define TSTAT0_LCRS		0x10000000
+#define TSTAT0_OW		0x04000000
+#define TSTAT0_TXU		0x02000000
+#define TSTAT0_ECOLL		0x01000000
+#define TSTAT0_NCOLL		0x001f0000
+#define TSTAT0_BUFFER_INDEX	0x00007fff
+
+struct ep93xx_descs
+{
+	struct ep93xx_rdesc	rdesc[RX_QUEUE_ENTRIES];
+	struct ep93xx_tdesc	tdesc[TX_QUEUE_ENTRIES];
+	struct ep93xx_rstat	rstat[RX_QUEUE_ENTRIES];
+	struct ep93xx_tstat	tstat[TX_QUEUE_ENTRIES];
+};
+
+struct ep93xx_priv
+{
+	struct resource		*res;
+	void			*base_addr;
+	int			irq;
+
+	struct ep93xx_descs	*descs;
+	dma_addr_t		descs_dma_addr;
+
+	void			*rx_buf[RX_QUEUE_ENTRIES];
+	void			*tx_buf[TX_QUEUE_ENTRIES];
+
+	spinlock_t		rx_lock;
+	unsigned int		rx_pointer;
+	unsigned int		tx_clean_pointer;
+	unsigned int		tx_pointer;
+	spinlock_t		tx_pending_lock;
+	unsigned int		tx_pending;
+
+	struct net_device_stats	stats;
+
+	struct mii_if_info	mii;
+	u8			mdc_divisor;
+};
+
+#define rdb(ep, off)		__raw_readb((ep)->base_addr + (off))
+#define rdw(ep, off)		__raw_readw((ep)->base_addr + (off))
+#define rdl(ep, off)		__raw_readl((ep)->base_addr + (off))
+#define wrb(ep, off, val)	__raw_writeb((val), (ep)->base_addr + (off))
+#define wrw(ep, off, val)	__raw_writew((val), (ep)->base_addr + (off))
+#define wrl(ep, off, val)	__raw_writel((val), (ep)->base_addr + (off))
+
+static int ep93xx_mdio_read(struct net_device *dev, int phy_id, int reg);
+
+static struct net_device_stats *ep93xx_get_stats(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	return &(ep->stats);
+}
+
+static int ep93xx_rx(struct net_device *dev, int *budget)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int tail_offset;
+	int rx_done;
+	int processed;
+
+	tail_offset = rdl(ep, REG_RXSTSQCURADD) - ep->descs_dma_addr;
+
+	rx_done = 0;
+	processed = 0;
+	while (*budget > 0) {
+		int entry;
+		struct ep93xx_rstat *rstat;
+		u32 rstat0;
+		u32 rstat1;
+		int length;
+		struct sk_buff *skb;
+
+		entry = ep->rx_pointer;
+		rstat = ep->descs->rstat + entry;
+		if ((void *)rstat - (void *)ep->descs == tail_offset) {
+			rx_done = 1;
+			break;
+		}
+
+		rstat0 = rstat->rstat0;
+		rstat1 = rstat->rstat1;
+		rstat->rstat0 = 0;
+		rstat->rstat1 = 0;
+
+		if (!(rstat0 & RSTAT0_RFP))
+			printk(KERN_CRIT "ep93xx_rx: buffer not done "
+					 " %.8x %.8x\n", rstat0, rstat1);
+		if (!(rstat0 & RSTAT0_EOF))
+			printk(KERN_CRIT "ep93xx_rx: not end-of-frame "
+					 " %.8x %.8x\n", rstat0, rstat1);
+		if (!(rstat0 & RSTAT0_EOB))
+			printk(KERN_CRIT "ep93xx_rx: not end-of-buffer "
+					 " %.8x %.8x\n", rstat0, rstat1);
+		if (!(rstat1 & RSTAT1_RFP))
+			printk(KERN_CRIT "ep93xx_rx: buffer1 not done "
+					 " %.8x %.8x\n", rstat0, rstat1);
+		if ((rstat1 & RSTAT1_BUFFER_INDEX) >> 16 != entry)
+			printk(KERN_CRIT "ep93xx_rx: entry mismatch "
+					 " %.8x %.8x\n", rstat0, rstat1);
+
+		if (!(rstat0 & RSTAT0_RWE)) {
+			printk(KERN_NOTICE "ep93xx_rx: receive error "
+					 " %.8x %.8x\n", rstat0, rstat1);
+
+			ep->stats.rx_errors++;
+			if (rstat0 & RSTAT0_OE)
+				ep->stats.rx_fifo_errors++;
+			if (rstat0 & RSTAT0_FE)
+				ep->stats.rx_frame_errors++;
+			if (rstat0 & (RSTAT0_RUNT | RSTAT0_EDATA))
+				ep->stats.rx_length_errors++;
+			if (rstat0 & RSTAT0_CRCE)
+				ep->stats.rx_crc_errors++;
+			goto err;
+		}
+
+		length = rstat1 & RSTAT1_FRAME_LENGTH;
+		if (length > MAX_PKT_SIZE) {
+			printk(KERN_NOTICE "ep93xx_rx: invalid length "
+					 " %.8x %.8x\n", rstat0, rstat1);
+			goto err;
+		}
+
+		/* Strip FCS.  */
+		if (rstat0 & RSTAT0_CRCI)
+			length -= 4;
+
+		skb = dev_alloc_skb(length + 2);
+		if (likely(skb != NULL)) {
+			skb->dev = dev;
+			skb_reserve(skb, 2);
+			dma_sync_single(NULL, ep->descs->rdesc[entry].buf_addr,
+						length, DMA_FROM_DEVICE);
+			eth_copy_and_sum(skb, ep->rx_buf[entry], length, 0);
+			skb_put(skb, length);
+			skb->protocol = eth_type_trans(skb, dev);
+
+			dev->last_rx = jiffies;
+
+			netif_receive_skb(skb);
+
+			ep->stats.rx_packets++;
+			ep->stats.rx_bytes += length;
+		} else {
+			ep->stats.rx_dropped++;
+		}
+
+err:
+		ep->rx_pointer = (entry + 1) & (RX_QUEUE_ENTRIES - 1);
+		processed++;
+		dev->quota--;
+		(*budget)--;
+	}
+
+	if (processed) {
+		wrw(ep, REG_RXDENQ, processed);
+		wrw(ep, REG_RXSTSENQ, processed);
+	}
+
+	return !rx_done;
+}
+
+static int ep93xx_have_more_rx(struct ep93xx_priv *ep)
+{
+	struct ep93xx_rstat *rstat;
+	int tail_offset;
+
+	rstat = ep->descs->rstat + ep->rx_pointer;
+	tail_offset = rdl(ep, REG_RXSTSQCURADD) - ep->descs_dma_addr;
+
+	return !((void *)rstat - (void *)ep->descs == tail_offset);
+}
+
+static int ep93xx_poll(struct net_device *dev, int *budget)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+
+	/*
+	 * @@@ Have to stop polling if device is downed while we
+	 * are polling.
+	 */
+
+poll_some_more:
+	if (ep93xx_rx(dev, budget))
+		return 1;
+
+	netif_rx_complete(dev);
+
+	spin_lock_irq(&ep->rx_lock);
+	wrl(ep, REG_INTEN, REG_INTEN_TX | REG_INTEN_RX);
+	if (ep93xx_have_more_rx(ep)) {
+		wrl(ep, REG_INTEN, REG_INTEN_TX);
+		wrl(ep, REG_INTSTSP, REG_INTSTS_RX);
+		spin_unlock_irq(&ep->rx_lock);
+
+		if (netif_rx_reschedule(dev, 0))
+			goto poll_some_more;
+
+		return 0;
+	}
+	spin_unlock_irq(&ep->rx_lock);
+
+	return 0;
+}
+
+static int ep93xx_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int entry;
+
+	if (unlikely(skb->len) > MAX_PKT_SIZE) {
+		ep->stats.tx_dropped++;
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+
+	entry = ep->tx_pointer;
+	ep->tx_pointer = (ep->tx_pointer + 1) & (TX_QUEUE_ENTRIES - 1);
+
+	ep->descs->tdesc[entry].tdesc1 =
+		TDESC1_EOF | (entry << 16) | (skb->len & 0xfff);
+	skb_copy_and_csum_dev(skb, ep->tx_buf[entry]);
+	dma_sync_single(NULL, ep->descs->tdesc[entry].buf_addr,
+				skb->len, DMA_TO_DEVICE);
+	dev_kfree_skb(skb);
+
+	dev->trans_start = jiffies;
+
+	spin_lock_irq(&ep->tx_pending_lock);
+	ep->tx_pending++;
+	if (ep->tx_pending == TX_QUEUE_ENTRIES)
+		netif_stop_queue(dev);
+	spin_unlock_irq(&ep->tx_pending_lock);
+
+	wrl(ep, REG_TXDENQ, 1);
+
+	return NETDEV_TX_OK;
+}
+
+static void ep93xx_tx_complete(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int tail_offset;
+	int wake;
+
+	tail_offset = rdl(ep, REG_TXSTSQCURADD) - ep->descs_dma_addr;
+	wake = 0;
+
+	spin_lock(&ep->tx_pending_lock);
+	while (1) {
+		int entry;
+		struct ep93xx_tstat *tstat;
+		u32 tstat0;
+
+		entry = ep->tx_clean_pointer;
+		tstat = ep->descs->tstat + entry;
+		if ((void *)tstat - (void *)ep->descs == tail_offset)
+			break;
+
+		tstat0 = tstat->tstat0;
+		tstat->tstat0 = 0;
+
+		if (!(tstat0 & TSTAT0_TXFP))
+			printk(KERN_CRIT "ep93xx_tx_complete: buffer not done "
+					 " %.8x\n", tstat0);
+		if (tstat0 & TSTAT0_FA)
+			printk(KERN_CRIT "ep93xx_tx_complete: frame aborted "
+					 " %.8x\n", tstat0);
+		if ((tstat0 & TSTAT0_BUFFER_INDEX) != entry)
+			printk(KERN_CRIT "ep93xx_tx_complete: entry mismatch "
+					 " %.8x\n", tstat0);
+
+		if (tstat0 & TSTAT0_TXWE) {
+			int length = ep->descs->tdesc[entry].tdesc1 & 0xfff;
+
+			ep->stats.tx_packets++;
+			ep->stats.tx_bytes += length;
+		} else {
+			ep->stats.tx_errors++;
+		}
+
+		if (tstat0 & TSTAT0_OW)
+			ep->stats.tx_window_errors++;
+		if (tstat0 & TSTAT0_TXU)
+			ep->stats.tx_fifo_errors++;
+		ep->stats.collisions += (tstat0 >> 16) & 0x1f;
+
+		ep->tx_clean_pointer = (entry + 1) & (TX_QUEUE_ENTRIES - 1);
+		if (ep->tx_pending == TX_QUEUE_ENTRIES)
+			wake = 1;
+		ep->tx_pending--;
+	}
+	spin_unlock(&ep->tx_pending_lock);
+
+	if (wake)
+		netif_wake_queue(dev);
+}
+
+static irqreturn_t ep93xx_irq(int irq, void *dev_id, struct pt_regs *regs)
+{
+	struct net_device *dev = dev_id;
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	u32 status;
+
+	status = rdl(ep, REG_INTSTSC);
+	if (status == 0)
+		return IRQ_NONE;
+
+	if (status & REG_INTSTS_RX) {
+		spin_lock(&ep->rx_lock);
+		if (likely(__netif_rx_schedule_prep(dev))) {
+			wrl(ep, REG_INTEN, REG_INTEN_TX);
+			__netif_rx_schedule(dev);
+		}
+		spin_unlock(&ep->rx_lock);
+	}
+
+	if (status & REG_INTSTS_TX)
+		ep93xx_tx_complete(dev);
+
+	return IRQ_HANDLED;
+}
+
+static void ep93xx_free_buffers(struct ep93xx_priv *ep)
+{
+	int i;
+
+	for (i = 0; i < RX_QUEUE_ENTRIES; i += 2) {
+		dma_addr_t d;
+
+		d = ep->descs->rdesc[i].buf_addr;
+		if (d)
+			dma_unmap_single(NULL, d, PAGE_SIZE, DMA_FROM_DEVICE);
+
+		if (ep->rx_buf[i] != NULL)
+			free_page((unsigned long)ep->rx_buf[i]);
+	}
+
+	for (i = 0; i < TX_QUEUE_ENTRIES; i += 2) {
+		dma_addr_t d;
+
+		d = ep->descs->tdesc[i].buf_addr;
+		if (d)
+			dma_unmap_single(NULL, d, PAGE_SIZE, DMA_TO_DEVICE);
+
+		if (ep->tx_buf[i] != NULL)
+			free_page((unsigned long)ep->tx_buf[i]);
+	}
+
+	dma_free_coherent(NULL, sizeof(struct ep93xx_descs), ep->descs,
+							ep->descs_dma_addr);
+}
+
+/*
+ * The hardware enforces a sub-2K maximum packet size, so we put
+ * two buffers on every hardware page.
+ */
+static int ep93xx_alloc_buffers(struct ep93xx_priv *ep)
+{
+	int i;
+
+	ep->descs = dma_alloc_coherent(NULL, sizeof(struct ep93xx_descs),
+				&ep->descs_dma_addr, GFP_KERNEL | GFP_DMA);
+	if (ep->descs == NULL)
+		return 1;
+
+	for (i = 0; i < RX_QUEUE_ENTRIES; i += 2) {
+		void *page;
+		dma_addr_t d;
+
+		page = (void *)__get_free_page(GFP_KERNEL | GFP_DMA);
+		if (page == NULL)
+			goto err;
+
+		d = dma_map_single(NULL, page, PAGE_SIZE, DMA_FROM_DEVICE);
+		if (dma_mapping_error(d)) {
+			free_page((unsigned long)page);
+			goto err;
+		}
+
+		ep->rx_buf[i] = page;
+		ep->descs->rdesc[i].buf_addr = d;
+		ep->descs->rdesc[i].rdesc1 = (i << 16) | PKT_BUF_SIZE;
+
+		ep->rx_buf[i + 1] = page + PKT_BUF_SIZE;
+		ep->descs->rdesc[i + 1].buf_addr = d + PKT_BUF_SIZE;
+		ep->descs->rdesc[i + 1].rdesc1 = ((i + 1) << 16) | PKT_BUF_SIZE;
+	}
+
+	for (i = 0; i < TX_QUEUE_ENTRIES; i += 2) {
+		void *page;
+		dma_addr_t d;
+
+		page = (void *)__get_free_page(GFP_KERNEL | GFP_DMA);
+		if (page == NULL)
+			goto err;
+
+		d = dma_map_single(NULL, page, PAGE_SIZE, DMA_TO_DEVICE);
+		if (dma_mapping_error(d)) {
+			free_page((unsigned long)page);
+			goto err;
+		}
+
+		ep->tx_buf[i] = page;
+		ep->descs->tdesc[i].buf_addr = d;
+
+		ep->tx_buf[i + 1] = page + PKT_BUF_SIZE;
+		ep->descs->tdesc[i + 1].buf_addr = d + PKT_BUF_SIZE;
+	}
+
+	return 0;
+
+err:
+	ep93xx_free_buffers(ep);
+	return 1;
+}
+
+static int ep93xx_start_hw(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	unsigned long addr;
+	int i;
+
+	wrl(ep, REG_SELFCTL, REG_SELFCTL_RESET);
+	for (i = 0; i < 10; i++) {
+		if ((rdl(ep, REG_SELFCTL) & REG_SELFCTL_RESET) == 0)
+			break;
+		msleep(1);
+	}
+
+	if (i == 10) {
+		printk(KERN_CRIT DRV_MODULE_NAME ": hw failed to reset\n");
+		return 1;
+	}
+
+	wrl(ep, REG_SELFCTL, ((ep->mdc_divisor - 1) << 9));
+
+	/* Does the PHY support preamble suppress?  */
+	if ((ep93xx_mdio_read(dev, ep->mii.phy_id, MII_BMSR) & 0x0040) != 0)
+		wrl(ep, REG_SELFCTL, ((ep->mdc_divisor - 1) << 9) | (1 << 8));
+
+	/* Receive descriptor ring.  */
+	addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, rdesc);
+	wrl(ep, REG_RXDQBADD, addr);
+	wrl(ep, REG_RXDCURADD, addr);
+	wrw(ep, REG_RXDQBLEN, RX_QUEUE_ENTRIES * sizeof(struct ep93xx_rdesc));
+
+	/* Receive status ring.  */
+	addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, rstat);
+	wrl(ep, REG_RXSTSQBADD, addr);
+	wrl(ep, REG_RXSTSQCURADD, addr);
+	wrw(ep, REG_RXSTSQBLEN, RX_QUEUE_ENTRIES * sizeof(struct ep93xx_rstat));
+
+	/* Transmit descriptor ring.  */
+	addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, tdesc);
+	wrl(ep, REG_TXDQBADD, addr);
+	wrl(ep, REG_TXDQCURADD, addr);
+	wrw(ep, REG_TXDQBLEN, TX_QUEUE_ENTRIES * sizeof(struct ep93xx_tdesc));
+
+	/* Transmit status ring.  */
+	addr = ep->descs_dma_addr + offsetof(struct ep93xx_descs, tstat);
+	wrl(ep, REG_TXSTSQBADD, addr);
+	wrl(ep, REG_TXSTSQCURADD, addr);
+	wrw(ep, REG_TXSTSQBLEN, TX_QUEUE_ENTRIES * sizeof(struct ep93xx_tstat));
+
+	wrl(ep, REG_BMCTL, REG_BMCTL_ENABLE_TX | REG_BMCTL_ENABLE_RX);
+	wrl(ep, REG_INTEN, REG_INTEN_TX | REG_INTEN_RX);
+	wrl(ep, REG_GIINTMSK, 0);
+
+	for (i = 0; i < 10; i++) {
+		if ((rdl(ep, REG_BMSTS) & REG_BMSTS_RX_ACTIVE) != 0)
+			break;
+		msleep(1);
+	}
+
+	if (i == 10) {
+		printk(KERN_CRIT DRV_MODULE_NAME ": hw failed to start\n");
+		return 1;
+	}
+
+	wrl(ep, REG_RXDENQ, RX_QUEUE_ENTRIES);
+	wrl(ep, REG_RXSTSENQ, RX_QUEUE_ENTRIES);
+
+	wrb(ep, REG_INDAD0, dev->dev_addr[0]);
+	wrb(ep, REG_INDAD1, dev->dev_addr[1]);
+	wrb(ep, REG_INDAD2, dev->dev_addr[2]);
+	wrb(ep, REG_INDAD3, dev->dev_addr[3]);
+	wrb(ep, REG_INDAD4, dev->dev_addr[4]);
+	wrb(ep, REG_INDAD5, dev->dev_addr[5]);
+	wrl(ep, REG_AFP, 0);
+
+	wrl(ep, REG_MAXFRMLEN, (MAX_PKT_SIZE << 16) | MAX_PKT_SIZE);
+
+	wrl(ep, REG_RXCTL, REG_RXCTL_DEFAULT);
+	wrl(ep, REG_TXCTL, REG_TXCTL_ENABLE);
+
+	return 0;
+}
+
+static void ep93xx_stop_hw(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int i;
+
+	wrl(ep, REG_SELFCTL, REG_SELFCTL_RESET);
+	for (i = 0; i < 10; i++) {
+		if ((rdl(ep, REG_SELFCTL) & REG_SELFCTL_RESET) == 0)
+			break;
+		msleep(1);
+	}
+
+	if (i == 10)
+		printk(KERN_CRIT DRV_MODULE_NAME ": hw failed to reset\n");
+}
+
+static int ep93xx_open(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int err;
+
+	if (ep93xx_alloc_buffers(ep))
+		return -ENOMEM;
+
+	if (is_zero_ether_addr(dev->dev_addr)) {
+		random_ether_addr(dev->dev_addr);
+		printk(KERN_INFO "%s: generated random MAC address "
+			"%.2x:%.2x:%.2x:%.2x:%.2x:%.2x.\n", dev->name,
+			dev->dev_addr[0], dev->dev_addr[1],
+			dev->dev_addr[2], dev->dev_addr[3],
+			dev->dev_addr[4], dev->dev_addr[5]);
+	}
+
+	if (ep93xx_start_hw(dev)) {
+		ep93xx_free_buffers(ep);
+		return -EIO;
+	}
+
+	spin_lock_init(&ep->rx_lock);
+	ep->rx_pointer = 0;
+	ep->tx_clean_pointer = 0;
+	ep->tx_pointer = 0;
+	spin_lock_init(&ep->tx_pending_lock);
+	ep->tx_pending = 0;
+
+	err = request_irq(ep->irq, ep93xx_irq, IRQF_SHARED, dev->name, dev);
+	if (err) {
+		ep93xx_stop_hw(dev);
+		ep93xx_free_buffers(ep);
+		return err;
+	}
+
+	wrl(ep, REG_GIINTMSK, REG_GIINTMSK_ENABLE);
+
+	netif_start_queue(dev);
+
+	return 0;
+}
+
+static int ep93xx_close(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+
+	netif_stop_queue(dev);
+
+	wrl(ep, REG_GIINTMSK, 0);
+	free_irq(ep->irq, dev);
+	ep93xx_stop_hw(dev);
+	ep93xx_free_buffers(ep);
+
+	return 0;
+}
+
+static int ep93xx_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	struct mii_ioctl_data *data = if_mii(ifr);
+
+	return generic_mii_ioctl(&ep->mii, data, cmd, NULL);
+}
+
+static int ep93xx_mdio_read(struct net_device *dev, int phy_id, int reg)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int data;
+	int i;
+
+	wrl(ep, REG_MIICMD, REG_MIICMD_READ | (phy_id << 5) | reg);
+
+	for (i = 0; i < 10; i++) {
+		if ((rdl(ep, REG_MIISTS) & REG_MIISTS_BUSY) == 0)
+			break;
+		msleep(1);
+	}
+
+	if (i == 10) {
+		printk(KERN_INFO DRV_MODULE_NAME ": mdio read timed out\n");
+		data = 0xffff;
+	} else {
+		data = rdl(ep, REG_MIIDATA);
+	}
+
+	return data;
+}
+
+static void ep93xx_mdio_write(struct net_device *dev, int phy_id, int reg, int data)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	int i;
+
+	wrl(ep, REG_MIIDATA, data);
+	wrl(ep, REG_MIICMD, REG_MIICMD_WRITE | (phy_id << 5) | reg);
+
+	for (i = 0; i < 10; i++) {
+		if ((rdl(ep, REG_MIISTS) & REG_MIISTS_BUSY) == 0)
+			break;
+		msleep(1);
+	}
+
+	if (i == 10)
+		printk(KERN_INFO DRV_MODULE_NAME ": mdio write timed out\n");
+}
+
+static void ep93xx_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
+{
+	strcpy(info->driver, DRV_MODULE_NAME);
+	strcpy(info->version, DRV_MODULE_VERSION);
+}
+
+static int ep93xx_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	return mii_ethtool_gset(&ep->mii, cmd);
+}
+
+static int ep93xx_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	return mii_ethtool_sset(&ep->mii, cmd);
+}
+
+static int ep93xx_nway_reset(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	return mii_nway_restart(&ep->mii);
+}
+
+static u32 ep93xx_get_link(struct net_device *dev)
+{
+	struct ep93xx_priv *ep = netdev_priv(dev);
+	return mii_link_ok(&ep->mii);
+}
+
+static struct ethtool_ops ep93xx_ethtool_ops = {
+	.get_drvinfo		= ep93xx_get_drvinfo,
+	.get_settings		= ep93xx_get_settings,
+	.set_settings		= ep93xx_set_settings,
+	.nway_reset		= ep93xx_nway_reset,
+	.get_link		= ep93xx_get_link,
+};
+
+struct net_device *ep93xx_dev_alloc(struct ep93xx_eth_data *data)
+{
+	struct net_device *dev;
+	struct ep93xx_priv *ep;
+
+	dev = alloc_etherdev(sizeof(struct ep93xx_priv));
+	if (dev == NULL)
+		return NULL;
+	ep = netdev_priv(dev);
+
+	memcpy(dev->dev_addr, data->dev_addr, ETH_ALEN);
+
+	dev->get_stats = ep93xx_get_stats;
+	dev->ethtool_ops = &ep93xx_ethtool_ops;
+	dev->poll = ep93xx_poll;
+	dev->hard_start_xmit = ep93xx_xmit;
+	dev->open = ep93xx_open;
+	dev->stop = ep93xx_close;
+	dev->do_ioctl = ep93xx_ioctl;
+
+	dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
+	dev->weight = 64;
+
+	return dev;
+}
+
+
+static int ep93xx_eth_remove(struct platform_device *pdev)
+{
+	struct net_device *dev;
+	struct ep93xx_priv *ep;
+
+	dev = platform_get_drvdata(pdev);
+	if (dev == NULL)
+		return 0;
+	platform_set_drvdata(pdev, NULL);
+
+	ep = netdev_priv(dev);
+
+	/* @@@ Force down.  */
+	unregister_netdev(dev);
+	ep93xx_free_buffers(ep);
+
+	if (ep->base_addr != NULL)
+		iounmap(ep->base_addr);
+
+	if (ep->res != NULL) {
+		release_resource(ep->res);
+		kfree(ep->res);
+	}
+
+	free_netdev(dev);
+
+	return 0;
+}
+
+static int ep93xx_eth_probe(struct platform_device *pdev)
+{
+	struct ep93xx_eth_data *data;
+	struct net_device *dev;
+	struct ep93xx_priv *ep;
+	int err;
+
+	data = pdev->dev.platform_data;
+	if (pdev == NULL)
+		return -ENODEV;
+
+	dev = ep93xx_dev_alloc(data);
+	if (dev == NULL) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+	ep = netdev_priv(dev);
+
+	platform_set_drvdata(pdev, dev);
+
+	ep->res = request_mem_region(pdev->resource[0].start,
+			pdev->resource[0].end - pdev->resource[0].start + 1,
+			pdev->dev.bus_id);
+	if (ep->res == NULL) {
+		dev_err(&pdev->dev, "Could not reserve memory region\n");
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	ep->base_addr = ioremap(pdev->resource[0].start,
+			pdev->resource[0].end - pdev->resource[0].start);
+	if (ep->base_addr == NULL) {
+		dev_err(&pdev->dev, "Failed to ioremap ethernet registers\n");
+		err = -EIO;
+		goto err_out;
+	}
+	ep->irq = pdev->resource[1].start;
+
+	ep->mii.phy_id = data->phy_id;
+	ep->mii.phy_id_mask = 0x1f;
+	ep->mii.reg_num_mask = 0x1f;
+	ep->mii.dev = dev;
+	ep->mii.mdio_read = ep93xx_mdio_read;
+	ep->mii.mdio_write = ep93xx_mdio_write;
+	ep->mdc_divisor = 40;	/* Max HCLK 100 MHz, min MDIO clk 2.5 MHz.  */
+
+	err = register_netdev(dev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to register netdev\n");
+		goto err_out;
+	}
+
+	printk(KERN_INFO "%s: ep93xx on-chip ethernet, IRQ %d, "
+			 "%.2x:%.2x:%.2x:%.2x:%.2x:%.2x.\n", dev->name,
+			ep->irq, data->dev_addr[0], data->dev_addr[1],
+			data->dev_addr[2], data->dev_addr[3],
+			data->dev_addr[4], data->dev_addr[5]);
+
+	return 0;
+
+err_out:
+	ep93xx_eth_remove(pdev);
+	return err;
+}
+
+
+static struct platform_driver ep93xx_eth_driver = {
+	.probe		= ep93xx_eth_probe,
+	.remove		= ep93xx_eth_remove,
+	.driver		= {
+		.name	= "ep93xx-eth",
+	},
+};
+
+static int __init ep93xx_eth_init_module(void)
+{
+	printk(KERN_INFO DRV_MODULE_NAME " version " DRV_MODULE_VERSION " loading\n");
+	return platform_driver_register(&ep93xx_eth_driver);
+}
+
+static void __exit ep93xx_eth_cleanup_module(void)
+{
+	platform_driver_unregister(&ep93xx_eth_driver);
+}
+
+module_init(ep93xx_eth_init_module);
+module_exit(ep93xx_eth_cleanup_module);
+MODULE_LICENSE("GPL");
diff --git a/include/asm-arm/arch-ep93xx/ep93xx-regs.h b/include/asm-arm/arch-ep93xx/ep93xx-regs.h
index 8c322975f96e..593f562f85c3 100644
--- a/include/asm-arm/arch-ep93xx/ep93xx-regs.h
+++ b/include/asm-arm/arch-ep93xx/ep93xx-regs.h
@@ -27,6 +27,7 @@
 #define EP93XX_DMA_BASE			(EP93XX_AHB_VIRT_BASE + 0x00000000)
 
 #define EP93XX_ETHERNET_BASE		(EP93XX_AHB_VIRT_BASE + 0x00010000)
+#define EP93XX_ETHERNET_PHYS_BASE	(EP93XX_AHB_PHYS_BASE + 0x00010000)
 
 #define EP93XX_USB_BASE			(EP93XX_AHB_VIRT_BASE + 0x00020000)
 #define EP93XX_USB_PHYS_BASE		(EP93XX_AHB_PHYS_BASE + 0x00020000)
diff --git a/include/asm-arm/arch-ep93xx/platform.h b/include/asm-arm/arch-ep93xx/platform.h
index d7a34ce20293..b4a8deb8bdef 100644
--- a/include/asm-arm/arch-ep93xx/platform.h
+++ b/include/asm-arm/arch-ep93xx/platform.h
@@ -11,5 +11,11 @@ void ep93xx_init_devices(void);
 void ep93xx_clock_init(void);
 extern struct sys_timer ep93xx_timer;
 
+struct ep93xx_eth_data
+{
+	unsigned char	dev_addr[6];
+	unsigned char	phy_id;
+};
+
 
 #endif
-- 
cgit v1.2.3


From a83fbf635992442edf6aa3252e4008d4a08edf12 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 23 Sep 2006 00:10:18 +0100
Subject: [PATCH] fix missing ifdefs in syscall classes hookup for generic
 targets

several targets have no ....at() family and m32r calls its only chown variant
chown32(), with __NR_chown being undefined.  creat(2) is also absent in some
targets.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/audit_change_attr.h | 4 ++++
 include/asm-generic/audit_dir_write.h   | 4 ++++
 lib/audit.c                             | 2 ++
 3 files changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/asm-generic/audit_change_attr.h b/include/asm-generic/audit_change_attr.h
index cb05bf69745a..50764550a60c 100644
--- a/include/asm-generic/audit_change_attr.h
+++ b/include/asm-generic/audit_change_attr.h
@@ -1,16 +1,20 @@
 __NR_chmod,
 __NR_fchmod,
+#ifdef __NR_chown
 __NR_chown,
 __NR_fchown,
 __NR_lchown,
+#endif
 __NR_setxattr,
 __NR_lsetxattr,
 __NR_fsetxattr,
 __NR_removexattr,
 __NR_lremovexattr,
 __NR_fremovexattr,
+#ifdef __NR_fchownat
 __NR_fchownat,
 __NR_fchmodat,
+#endif
 #ifdef __NR_chown32
 __NR_chown32,
 __NR_fchown32,
diff --git a/include/asm-generic/audit_dir_write.h b/include/asm-generic/audit_dir_write.h
index 161a7a58fbab..6621bd82cbe8 100644
--- a/include/asm-generic/audit_dir_write.h
+++ b/include/asm-generic/audit_dir_write.h
@@ -1,14 +1,18 @@
 __NR_rename,
 __NR_mkdir,
 __NR_rmdir,
+#ifdef __NR_creat
 __NR_creat,
+#endif
 __NR_link,
 __NR_unlink,
 __NR_symlink,
 __NR_mknod,
+#ifdef __NR_mkdirat
 __NR_mkdirat,
 __NR_mknodat,
 __NR_unlinkat,
 __NR_renameat,
 __NR_linkat,
 __NR_symlinkat,
+#endif
diff --git a/lib/audit.c b/lib/audit.c
index 8c21625ef938..3b1289fadf06 100644
--- a/lib/audit.c
+++ b/lib/audit.c
@@ -28,8 +28,10 @@ int audit_classify_syscall(int abi, unsigned syscall)
 	switch(syscall) {
 	case __NR_open:
 		return 2;
+#ifdef __NR_openat
 	case __NR_openat:
 		return 3;
+#endif
 #ifdef __NR_socketcall
 	case __NR_socketcall:
 		return 4;
-- 
cgit v1.2.3


From 1c3c07e9f6cc50dab2aeb8051325e317d4f6c70e Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 25 Jul 2006 11:28:18 -0400
Subject: NFS: Add a new ACCESS rpc call cache to the linux nfs client

The current access cache only allows one entry at a time to be cached for each
inode. Add a per-inode red-black tree in order to allow more than one to
be cached at a time.

Should significantly cut down the time spent in path traversal for shared
directories such as ${PATH}, /usr/share, etc.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 133 ++++++++++++++++++++++++++++++++++++++++++-------
 fs/nfs/inode.c         |  13 ++---
 include/linux/nfs_fs.h |   5 +-
 3 files changed, 124 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index e7ffb4deb3e5..094afded2b11 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1638,35 +1638,134 @@ out:
 	return error;
 }
 
-int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res)
+static void nfs_access_free_entry(struct nfs_access_entry *entry)
+{
+	put_rpccred(entry->cred);
+	kfree(entry);
+}
+
+static void __nfs_access_zap_cache(struct inode *inode)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
-	struct nfs_access_entry *cache = &nfsi->cache_access;
+	struct rb_root *root_node = &nfsi->access_cache;
+	struct rb_node *n, *dispose = NULL;
+	struct nfs_access_entry *entry;
+
+	/* Unhook entries from the cache */
+	while ((n = rb_first(root_node)) != NULL) {
+		entry = rb_entry(n, struct nfs_access_entry, rb_node);
+		rb_erase(n, root_node);
+		n->rb_left = dispose;
+		dispose = n;
+	}
+	nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS;
+	spin_unlock(&inode->i_lock);
 
-	if (cache->cred != cred
-			|| time_after(jiffies, cache->jiffies + NFS_ATTRTIMEO(inode))
-			|| (nfsi->cache_validity & NFS_INO_INVALID_ACCESS))
-		return -ENOENT;
-	memcpy(res, cache, sizeof(*res));
-	return 0;
+	/* Now kill them all! */
+	while (dispose != NULL) {
+		n = dispose;
+		dispose = n->rb_left;
+		nfs_access_free_entry(rb_entry(n, struct nfs_access_entry, rb_node));
+	}
 }
 
-void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set)
+void nfs_access_zap_cache(struct inode *inode)
 {
-	struct nfs_inode *nfsi = NFS_I(inode);
-	struct nfs_access_entry *cache = &nfsi->cache_access;
+	spin_lock(&inode->i_lock);
+	/* This will release the spinlock */
+	__nfs_access_zap_cache(inode);
+}
 
-	if (cache->cred != set->cred) {
-		if (cache->cred)
-			put_rpccred(cache->cred);
-		cache->cred = get_rpccred(set->cred);
+static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, struct rpc_cred *cred)
+{
+	struct rb_node *n = NFS_I(inode)->access_cache.rb_node;
+	struct nfs_access_entry *entry;
+
+	while (n != NULL) {
+		entry = rb_entry(n, struct nfs_access_entry, rb_node);
+
+		if (cred < entry->cred)
+			n = n->rb_left;
+		else if (cred > entry->cred)
+			n = n->rb_right;
+		else
+			return entry;
 	}
-	/* FIXME: replace current access_cache BKL reliance with inode->i_lock */
+	return NULL;
+}
+
+int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res)
+{
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct nfs_access_entry *cache;
+	int err = -ENOENT;
+
 	spin_lock(&inode->i_lock);
-	nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS;
+	if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS)
+		goto out_zap;
+	cache = nfs_access_search_rbtree(inode, cred);
+	if (cache == NULL)
+		goto out;
+	if (time_after(jiffies, cache->jiffies + NFS_ATTRTIMEO(inode)))
+		goto out_stale;
+	res->jiffies = cache->jiffies;
+	res->cred = cache->cred;
+	res->mask = cache->mask;
+	err = 0;
+out:
+	spin_unlock(&inode->i_lock);
+	return err;
+out_stale:
+	rb_erase(&cache->rb_node, &nfsi->access_cache);
+	spin_unlock(&inode->i_lock);
+	nfs_access_free_entry(cache);
+	return -ENOENT;
+out_zap:
+	/* This will release the spinlock */
+	__nfs_access_zap_cache(inode);
+	return -ENOENT;
+}
+
+static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *set)
+{
+	struct rb_root *root_node = &NFS_I(inode)->access_cache;
+	struct rb_node **p = &root_node->rb_node;
+	struct rb_node *parent = NULL;
+	struct nfs_access_entry *entry;
+
+	spin_lock(&inode->i_lock);
+	while (*p != NULL) {
+		parent = *p;
+		entry = rb_entry(parent, struct nfs_access_entry, rb_node);
+
+		if (set->cred < entry->cred)
+			p = &parent->rb_left;
+		else if (set->cred > entry->cred)
+			p = &parent->rb_right;
+		else
+			goto found;
+	}
+	rb_link_node(&set->rb_node, parent, p);
+	rb_insert_color(&set->rb_node, root_node);
 	spin_unlock(&inode->i_lock);
+	return;
+found:
+	rb_replace_node(parent, &set->rb_node, root_node);
+	spin_unlock(&inode->i_lock);
+	nfs_access_free_entry(entry);
+}
+
+void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set)
+{
+	struct nfs_access_entry *cache = kmalloc(sizeof(*cache), GFP_KERNEL);
+	if (cache == NULL)
+		return;
+	RB_CLEAR_NODE(&cache->rb_node);
 	cache->jiffies = set->jiffies;
+	cache->cred = get_rpccred(set->cred);
 	cache->mask = set->mask;
+
+	nfs_access_add_rbtree(inode, cache);
 }
 
 static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index d349fb2245da..b94ab060bb1e 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -76,19 +76,14 @@ int nfs_write_inode(struct inode *inode, int sync)
 
 void nfs_clear_inode(struct inode *inode)
 {
-	struct nfs_inode *nfsi = NFS_I(inode);
-	struct rpc_cred *cred;
-
 	/*
 	 * The following should never happen...
 	 */
 	BUG_ON(nfs_have_writebacks(inode));
-	BUG_ON (!list_empty(&nfsi->open_files));
+	BUG_ON(!list_empty(&NFS_I(inode)->open_files));
+	BUG_ON(atomic_read(&NFS_I(inode)->data_updates) != 0);
 	nfs_zap_acl_cache(inode);
-	cred = nfsi->cache_access.cred;
-	if (cred)
-		put_rpccred(cred);
-	BUG_ON(atomic_read(&nfsi->data_updates) != 0);
+	nfs_access_zap_cache(inode);
 }
 
 /**
@@ -290,7 +285,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 		nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
 		nfsi->attrtimeo_timestamp = jiffies;
 		memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
-		nfsi->cache_access.cred = NULL;
+		nfsi->access_cache = RB_ROOT;
 
 		unlock_new_inode(inode);
 	} else
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 6c2066caeaab..cc013ed2e52e 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -42,6 +42,7 @@
 #include <linux/in.h>
 #include <linux/mm.h>
 #include <linux/pagemap.h>
+#include <linux/rbtree.h>
 #include <linux/rwsem.h>
 #include <linux/wait.h>
 
@@ -69,6 +70,7 @@
  * NFSv3/v4 Access mode cache entry
  */
 struct nfs_access_entry {
+	struct rb_node		rb_node;
 	unsigned long		jiffies;
 	struct rpc_cred *	cred;
 	int			mask;
@@ -145,7 +147,7 @@ struct nfs_inode {
 	 */
 	atomic_t		data_updates;
 
-	struct nfs_access_entry	cache_access;
+	struct rb_root		access_cache;
 #ifdef CONFIG_NFS_V3_ACL
 	struct posix_acl	*acl_access;
 	struct posix_acl	*acl_default;
@@ -297,6 +299,7 @@ extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 extern int nfs_permission(struct inode *, int, struct nameidata *);
 extern int nfs_access_get_cached(struct inode *, struct rpc_cred *, struct nfs_access_entry *);
 extern void nfs_access_add_cache(struct inode *, struct nfs_access_entry *);
+extern void nfs_access_zap_cache(struct inode *inode);
 extern int nfs_open(struct inode *, struct file *);
 extern int nfs_release(struct inode *, struct file *);
 extern int nfs_attribute_timeout(struct inode *inode);
-- 
cgit v1.2.3


From cfcea3e8c66c2dcde98d5c2693d4bff50b5cac97 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 25 Jul 2006 11:28:18 -0400
Subject: NFS: Add a global LRU list for the ACCESS cache

...in order to allow the addition of a memory shrinker.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c           | 35 ++++++++++++++++++++++++++++++++++-
 fs/nfs/inode.c         |  2 ++
 include/linux/nfs_fs.h |  4 ++++
 3 files changed, 40 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 094afded2b11..bf4f5ffda703 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1638,10 +1638,17 @@ out:
 	return error;
 }
 
+static DEFINE_SPINLOCK(nfs_access_lru_lock);
+static LIST_HEAD(nfs_access_lru_list);
+static atomic_long_t nfs_access_nr_entries;
+
 static void nfs_access_free_entry(struct nfs_access_entry *entry)
 {
 	put_rpccred(entry->cred);
 	kfree(entry);
+	smp_mb__before_atomic_dec();
+	atomic_long_dec(&nfs_access_nr_entries);
+	smp_mb__after_atomic_dec();
 }
 
 static void __nfs_access_zap_cache(struct inode *inode)
@@ -1655,6 +1662,7 @@ static void __nfs_access_zap_cache(struct inode *inode)
 	while ((n = rb_first(root_node)) != NULL) {
 		entry = rb_entry(n, struct nfs_access_entry, rb_node);
 		rb_erase(n, root_node);
+		list_del(&entry->lru);
 		n->rb_left = dispose;
 		dispose = n;
 	}
@@ -1671,6 +1679,13 @@ static void __nfs_access_zap_cache(struct inode *inode)
 
 void nfs_access_zap_cache(struct inode *inode)
 {
+	/* Remove from global LRU init */
+	if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) {
+		spin_lock(&nfs_access_lru_lock);
+		list_del_init(&NFS_I(inode)->access_cache_inode_lru);
+		spin_unlock(&nfs_access_lru_lock);
+	}
+
 	spin_lock(&inode->i_lock);
 	/* This will release the spinlock */
 	__nfs_access_zap_cache(inode);
@@ -1711,12 +1726,14 @@ int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, struct nfs
 	res->jiffies = cache->jiffies;
 	res->cred = cache->cred;
 	res->mask = cache->mask;
+	list_move_tail(&cache->lru, &nfsi->access_cache_entry_lru);
 	err = 0;
 out:
 	spin_unlock(&inode->i_lock);
 	return err;
 out_stale:
 	rb_erase(&cache->rb_node, &nfsi->access_cache);
+	list_del(&cache->lru);
 	spin_unlock(&inode->i_lock);
 	nfs_access_free_entry(cache);
 	return -ENOENT;
@@ -1728,7 +1745,8 @@ out_zap:
 
 static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *set)
 {
-	struct rb_root *root_node = &NFS_I(inode)->access_cache;
+	struct nfs_inode *nfsi = NFS_I(inode);
+	struct rb_root *root_node = &nfsi->access_cache;
 	struct rb_node **p = &root_node->rb_node;
 	struct rb_node *parent = NULL;
 	struct nfs_access_entry *entry;
@@ -1747,10 +1765,13 @@ static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *
 	}
 	rb_link_node(&set->rb_node, parent, p);
 	rb_insert_color(&set->rb_node, root_node);
+	list_add_tail(&set->lru, &nfsi->access_cache_entry_lru);
 	spin_unlock(&inode->i_lock);
 	return;
 found:
 	rb_replace_node(parent, &set->rb_node, root_node);
+	list_add_tail(&set->lru, &nfsi->access_cache_entry_lru);
+	list_del(&entry->lru);
 	spin_unlock(&inode->i_lock);
 	nfs_access_free_entry(entry);
 }
@@ -1766,6 +1787,18 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set)
 	cache->mask = set->mask;
 
 	nfs_access_add_rbtree(inode, cache);
+
+	/* Update accounting */
+	smp_mb__before_atomic_inc();
+	atomic_long_inc(&nfs_access_nr_entries);
+	smp_mb__after_atomic_inc();
+
+	/* Add inode to global LRU list */
+	if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) {
+		spin_lock(&nfs_access_lru_lock);
+		list_add_tail(&NFS_I(inode)->access_cache_inode_lru, &nfs_access_lru_list);
+		spin_unlock(&nfs_access_lru_lock);
+	}
 }
 
 static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask)
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index b94ab060bb1e..6ed018c9aad2 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1104,6 +1104,8 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 		INIT_LIST_HEAD(&nfsi->dirty);
 		INIT_LIST_HEAD(&nfsi->commit);
 		INIT_LIST_HEAD(&nfsi->open_files);
+		INIT_LIST_HEAD(&nfsi->access_cache_entry_lru);
+		INIT_LIST_HEAD(&nfsi->access_cache_inode_lru);
 		INIT_RADIX_TREE(&nfsi->nfs_page_tree, GFP_ATOMIC);
 		atomic_set(&nfsi->data_updates, 0);
 		nfsi->ndirty = 0;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index cc013ed2e52e..a36e01cd6321 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -71,6 +71,7 @@
  */
 struct nfs_access_entry {
 	struct rb_node		rb_node;
+	struct list_head	lru;
 	unsigned long		jiffies;
 	struct rpc_cred *	cred;
 	int			mask;
@@ -148,6 +149,8 @@ struct nfs_inode {
 	atomic_t		data_updates;
 
 	struct rb_root		access_cache;
+	struct list_head	access_cache_entry_lru;
+	struct list_head	access_cache_inode_lru;
 #ifdef CONFIG_NFS_V3_ACL
 	struct posix_acl	*acl_access;
 	struct posix_acl	*acl_default;
@@ -201,6 +204,7 @@ struct nfs_inode {
 #define NFS_INO_REVALIDATING	(0)		/* revalidating attrs */
 #define NFS_INO_ADVISE_RDPLUS	(1)		/* advise readdirplus */
 #define NFS_INO_STALE		(2)		/* possible stale inode */
+#define NFS_INO_ACL_LRU_SET	(3)		/* Inode is on the LRU list */
 
 static inline struct nfs_inode *NFS_I(struct inode *inode)
 {
-- 
cgit v1.2.3


From 770bfad846ab6628444428467b11fa6773ae9ea1 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:07 -0400
Subject: NFS: Add dentry materialisation op

The attached patch adds a new directory cache management function that prepares
a disconnected anonymous function to be connected into the dentry tree. The
anonymous dentry is transferred the name and parentage from another dentry.

The following changes were made in [try #2]:

 (*) d_materialise_dentry() now switches the parentage of the two nodes around
     correctly when one or other of them is self-referential.

The following changes were made in [try #7]:

 (*) d_instantiate_unique() has had the interior part split out as function
     __d_instantiate_unique(). Callers of this latter function must be holding
     the appropriate locks.

 (*) _d_rehash() has been added as a wrapper around __d_rehash() to call it
     with the most obvious hash list (the one from the name). d_rehash() now
     calls _d_rehash().

 (*) d_materialise_dentry() is now __d_materialise_dentry() and is static.

 (*) d_materialise_unique() added to perform the combination of d_find_alias(),
     d_materialise_dentry() and d_add_unique() that the NFS client was doing
     twice, all within a single dcache_lock critical section. This reduces the
     number of times two different spinlocks were being accessed.

The following further changes were made:

 (*) Add the dentries onto their parents d_subdirs lists.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/dcache.c            | 164 ++++++++++++++++++++++++++++++++++++++++++++-----
 include/linux/dcache.h |   1 +
 2 files changed, 151 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/fs/dcache.c b/fs/dcache.c
index 1b4a3a34ec57..17b392a2049e 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -828,17 +828,19 @@ void d_instantiate(struct dentry *entry, struct inode * inode)
  * (or otherwise set) by the caller to indicate that it is now
  * in use by the dcache.
  */
-struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
+static struct dentry *__d_instantiate_unique(struct dentry *entry,
+					     struct inode *inode)
 {
 	struct dentry *alias;
 	int len = entry->d_name.len;
 	const char *name = entry->d_name.name;
 	unsigned int hash = entry->d_name.hash;
 
-	BUG_ON(!list_empty(&entry->d_alias));
-	spin_lock(&dcache_lock);
-	if (!inode)
-		goto do_negative;
+	if (!inode) {
+		entry->d_inode = NULL;
+		return NULL;
+	}
+
 	list_for_each_entry(alias, &inode->i_dentry, d_alias) {
 		struct qstr *qstr = &alias->d_name;
 
@@ -851,19 +853,35 @@ struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
 		if (memcmp(qstr->name, name, len))
 			continue;
 		dget_locked(alias);
-		spin_unlock(&dcache_lock);
-		BUG_ON(!d_unhashed(alias));
-		iput(inode);
 		return alias;
 	}
+
 	list_add(&entry->d_alias, &inode->i_dentry);
-do_negative:
 	entry->d_inode = inode;
 	fsnotify_d_instantiate(entry, inode);
-	spin_unlock(&dcache_lock);
-	security_d_instantiate(entry, inode);
 	return NULL;
 }
+
+struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
+{
+	struct dentry *result;
+
+	BUG_ON(!list_empty(&entry->d_alias));
+
+	spin_lock(&dcache_lock);
+	result = __d_instantiate_unique(entry, inode);
+	spin_unlock(&dcache_lock);
+
+	if (!result) {
+		security_d_instantiate(entry, inode);
+		return NULL;
+	}
+
+	BUG_ON(!d_unhashed(result));
+	iput(inode);
+	return result;
+}
+
 EXPORT_SYMBOL(d_instantiate_unique);
 
 /**
@@ -1235,6 +1253,11 @@ static void __d_rehash(struct dentry * entry, struct hlist_head *list)
  	hlist_add_head_rcu(&entry->d_hash, list);
 }
 
+static void _d_rehash(struct dentry * entry)
+{
+	__d_rehash(entry, d_hash(entry->d_parent, entry->d_name.hash));
+}
+
 /**
  * d_rehash	- add an entry back to the hash
  * @entry: dentry to add to the hash
@@ -1244,11 +1267,9 @@ static void __d_rehash(struct dentry * entry, struct hlist_head *list)
  
 void d_rehash(struct dentry * entry)
 {
-	struct hlist_head *list = d_hash(entry->d_parent, entry->d_name.hash);
-
 	spin_lock(&dcache_lock);
 	spin_lock(&entry->d_lock);
-	__d_rehash(entry, list);
+	_d_rehash(entry);
 	spin_unlock(&entry->d_lock);
 	spin_unlock(&dcache_lock);
 }
@@ -1386,6 +1407,120 @@ already_unhashed:
 	spin_unlock(&dcache_lock);
 }
 
+/*
+ * Prepare an anonymous dentry for life in the superblock's dentry tree as a
+ * named dentry in place of the dentry to be replaced.
+ */
+static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon)
+{
+	struct dentry *dparent, *aparent;
+
+	switch_names(dentry, anon);
+	do_switch(dentry->d_name.len, anon->d_name.len);
+	do_switch(dentry->d_name.hash, anon->d_name.hash);
+
+	dparent = dentry->d_parent;
+	aparent = anon->d_parent;
+
+	dentry->d_parent = (aparent == anon) ? dentry : aparent;
+	list_del(&dentry->d_u.d_child);
+	if (!IS_ROOT(dentry))
+		list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs);
+	else
+		INIT_LIST_HEAD(&dentry->d_u.d_child);
+
+	anon->d_parent = (dparent == dentry) ? anon : dparent;
+	list_del(&anon->d_u.d_child);
+	if (!IS_ROOT(anon))
+		list_add(&anon->d_u.d_child, &anon->d_parent->d_subdirs);
+	else
+		INIT_LIST_HEAD(&anon->d_u.d_child);
+
+	anon->d_flags &= ~DCACHE_DISCONNECTED;
+}
+
+/**
+ * d_materialise_unique - introduce an inode into the tree
+ * @dentry: candidate dentry
+ * @inode: inode to bind to the dentry, to which aliases may be attached
+ *
+ * Introduces an dentry into the tree, substituting an extant disconnected
+ * root directory alias in its place if there is one
+ */
+struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
+{
+	struct dentry *alias, *actual;
+
+	BUG_ON(!d_unhashed(dentry));
+
+	spin_lock(&dcache_lock);
+
+	if (!inode) {
+		actual = dentry;
+		dentry->d_inode = NULL;
+		goto found_lock;
+	}
+
+	/* See if a disconnected directory already exists as an anonymous root
+	 * that we should splice into the tree instead */
+	if (S_ISDIR(inode->i_mode) && (alias = __d_find_alias(inode, 1))) {
+		spin_lock(&alias->d_lock);
+
+		/* Is this a mountpoint that we could splice into our tree? */
+		if (IS_ROOT(alias))
+			goto connect_mountpoint;
+
+		if (alias->d_name.len == dentry->d_name.len &&
+		    alias->d_parent == dentry->d_parent &&
+		    memcmp(alias->d_name.name,
+			   dentry->d_name.name,
+			   dentry->d_name.len) == 0)
+			goto replace_with_alias;
+
+		spin_unlock(&alias->d_lock);
+
+		/* Doh! Seem to be aliasing directories for some reason... */
+		dput(alias);
+	}
+
+	/* Add a unique reference */
+	actual = __d_instantiate_unique(dentry, inode);
+	if (!actual)
+		actual = dentry;
+	else if (unlikely(!d_unhashed(actual)))
+		goto shouldnt_be_hashed;
+
+found_lock:
+	spin_lock(&actual->d_lock);
+found:
+	_d_rehash(actual);
+	spin_unlock(&actual->d_lock);
+	spin_unlock(&dcache_lock);
+
+	if (actual == dentry) {
+		security_d_instantiate(dentry, inode);
+		return NULL;
+	}
+
+	iput(inode);
+	return actual;
+
+	/* Convert the anonymous/root alias into an ordinary dentry */
+connect_mountpoint:
+	__d_materialise_dentry(dentry, alias);
+
+	/* Replace the candidate dentry with the alias in the tree */
+replace_with_alias:
+	__d_drop(alias);
+	actual = alias;
+	goto found;
+
+shouldnt_be_hashed:
+	spin_unlock(&dcache_lock);
+	BUG();
+	goto shouldnt_be_hashed;
+}
+
 /**
  * d_path - return the path of a dentry
  * @dentry: dentry to report
@@ -1784,6 +1919,7 @@ EXPORT_SYMBOL(d_instantiate);
 EXPORT_SYMBOL(d_invalidate);
 EXPORT_SYMBOL(d_lookup);
 EXPORT_SYMBOL(d_move);
+EXPORT_SYMBOL_GPL(d_materialise_unique);
 EXPORT_SYMBOL(d_path);
 EXPORT_SYMBOL(d_prune_aliases);
 EXPORT_SYMBOL(d_rehash);
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 471781ffeab1..44605be59409 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -221,6 +221,7 @@ static inline int dname_external(struct dentry *dentry)
  */
 extern void d_instantiate(struct dentry *, struct inode *);
 extern struct dentry * d_instantiate_unique(struct dentry *, struct inode *);
+extern struct dentry * d_materialise_unique(struct dentry *, struct inode *);
 extern void d_delete(struct dentry *);
 
 /* allocate/de-allocate */
-- 
cgit v1.2.3


From adfa6f980bd46974e6b32b22dd0c45e3f52063f4 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:08 -0400
Subject: NFS: Rename struct nfs4_client to struct nfs_client

Rename struct nfs4_client to struct nfs_client so that it can become the basis
for a general client record for NFS2 and NFS3 in addition to NFS4.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/callback.c         |  2 +-
 fs/nfs/callback_proc.c    |  4 ++--
 fs/nfs/delegation.c       | 24 +++++++++++------------
 fs/nfs/delegation.h       | 10 +++++-----
 fs/nfs/idmap.c            | 12 ++++++------
 fs/nfs/nfs4_fs.h          | 30 ++++++++++++++--------------
 fs/nfs/nfs4proc.c         | 32 +++++++++++++++---------------
 fs/nfs/nfs4renewd.c       |  8 ++++----
 fs/nfs/nfs4state.c        | 50 +++++++++++++++++++++++------------------------
 fs/nfs/nfs4xdr.c          | 18 ++++++++---------
 fs/nfs/super.c            |  4 ++--
 include/linux/nfs_fs_sb.h |  2 +-
 include/linux/nfs_idmap.h | 14 ++++++-------
 13 files changed, 105 insertions(+), 105 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index b1f7dc415392..1b596b6d9dc2 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -167,7 +167,7 @@ void nfs_callback_down(void)
 static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 {
 	struct in_addr *addr = &rqstp->rq_addr.sin_addr;
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 
 	/* Don't talk to strangers */
 	clp = nfs4_find_client(addr);
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index 7719483ecdfc..55d6e2ec157f 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -15,7 +15,7 @@
  
 unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	struct nfs_delegation *delegation;
 	struct nfs_inode *nfsi;
 	struct inode *inode;
@@ -56,7 +56,7 @@ out:
 
 unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	struct inode *inode;
 	unsigned res;
 	
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 9540a316c05e..5a1105c258bd 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -114,7 +114,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st
  */
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res)
 {
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int status = 0;
@@ -176,7 +176,7 @@ static void nfs_msync_inode(struct inode *inode)
  */
 int __nfs_inode_return_delegation(struct inode *inode)
 {
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
@@ -208,7 +208,7 @@ int __nfs_inode_return_delegation(struct inode *inode)
  */
 void nfs_return_all_delegations(struct super_block *sb)
 {
-	struct nfs4_client *clp = NFS_SB(sb)->nfs4_state;
+	struct nfs_client *clp = NFS_SB(sb)->nfs4_state;
 	struct nfs_delegation *delegation;
 	struct inode *inode;
 
@@ -232,7 +232,7 @@ restart:
 
 int nfs_do_expire_all_delegations(void *ptr)
 {
-	struct nfs4_client *clp = ptr;
+	struct nfs_client *clp = ptr;
 	struct nfs_delegation *delegation;
 	struct inode *inode;
 
@@ -258,7 +258,7 @@ out:
 	module_put_and_exit(0);
 }
 
-void nfs_expire_all_delegations(struct nfs4_client *clp)
+void nfs_expire_all_delegations(struct nfs_client *clp)
 {
 	struct task_struct *task;
 
@@ -276,7 +276,7 @@ void nfs_expire_all_delegations(struct nfs4_client *clp)
 /*
  * Return all delegations following an NFS4ERR_CB_PATH_DOWN error.
  */
-void nfs_handle_cb_pathdown(struct nfs4_client *clp)
+void nfs_handle_cb_pathdown(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation;
 	struct inode *inode;
@@ -299,7 +299,7 @@ restart:
 
 struct recall_threadargs {
 	struct inode *inode;
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	const nfs4_stateid *stateid;
 
 	struct completion started;
@@ -310,7 +310,7 @@ static int recall_thread(void *data)
 {
 	struct recall_threadargs *args = (struct recall_threadargs *)data;
 	struct inode *inode = igrab(args->inode);
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 
@@ -371,7 +371,7 @@ out_module_put:
 /*
  * Retrieve the inode associated with a delegation
  */
-struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle)
+struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle)
 {
 	struct nfs_delegation *delegation;
 	struct inode *res = NULL;
@@ -389,7 +389,7 @@ struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nf
 /*
  * Mark all delegations as needing to be reclaimed
  */
-void nfs_delegation_mark_reclaim(struct nfs4_client *clp)
+void nfs_delegation_mark_reclaim(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation;
 	spin_lock(&clp->cl_lock);
@@ -401,7 +401,7 @@ void nfs_delegation_mark_reclaim(struct nfs4_client *clp)
 /*
  * Reap all unclaimed delegations after reboot recovery is done
  */
-void nfs_delegation_reap_unclaimed(struct nfs4_client *clp)
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation, *n;
 	LIST_HEAD(head);
@@ -423,7 +423,7 @@ void nfs_delegation_reap_unclaimed(struct nfs4_client *clp)
 
 int nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode)
 {
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h
index 3858694652fa..2cfd4b24c7fe 100644
--- a/fs/nfs/delegation.h
+++ b/fs/nfs/delegation.h
@@ -29,13 +29,13 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st
 int __nfs_inode_return_delegation(struct inode *inode);
 int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
 
-struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle);
+struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle);
 void nfs_return_all_delegations(struct super_block *sb);
-void nfs_expire_all_delegations(struct nfs4_client *clp);
-void nfs_handle_cb_pathdown(struct nfs4_client *clp);
+void nfs_expire_all_delegations(struct nfs_client *clp);
+void nfs_handle_cb_pathdown(struct nfs_client *clp);
 
-void nfs_delegation_mark_reclaim(struct nfs4_client *clp);
-void nfs_delegation_reap_unclaimed(struct nfs4_client *clp);
+void nfs_delegation_mark_reclaim(struct nfs_client *clp);
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp);
 
 /* NFSv4 delegation-related procedures */
 int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid);
diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c
index 873deb96a6cd..d05148ec9414 100644
--- a/fs/nfs/idmap.c
+++ b/fs/nfs/idmap.c
@@ -109,7 +109,7 @@ static struct rpc_pipe_ops idmap_upcall_ops = {
 };
 
 void
-nfs_idmap_new(struct nfs4_client *clp)
+nfs_idmap_new(struct nfs_client *clp)
 {
 	struct idmap *idmap;
 
@@ -138,7 +138,7 @@ nfs_idmap_new(struct nfs4_client *clp)
 }
 
 void
-nfs_idmap_delete(struct nfs4_client *clp)
+nfs_idmap_delete(struct nfs_client *clp)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
@@ -491,27 +491,27 @@ static unsigned int fnvhash32(const void *buf, size_t buflen)
 	return (hash);
 }
 
-int nfs_map_name_to_uid(struct nfs4_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_name_to_uid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
 	return nfs_idmap_id(idmap, &idmap->idmap_user_hash, name, namelen, uid);
 }
 
-int nfs_map_group_to_gid(struct nfs4_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_group_to_gid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
 	return nfs_idmap_id(idmap, &idmap->idmap_group_hash, name, namelen, uid);
 }
 
-int nfs_map_uid_to_name(struct nfs4_client *clp, __u32 uid, char *buf)
+int nfs_map_uid_to_name(struct nfs_client *clp, __u32 uid, char *buf)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
 	return nfs_idmap_name(idmap, &idmap->idmap_user_hash, uid, buf);
 }
-int nfs_map_gid_to_group(struct nfs4_client *clp, __u32 uid, char *buf)
+int nfs_map_gid_to_group(struct nfs_client *clp, __u32 uid, char *buf)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 9a102860df37..4e334cb48498 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -43,9 +43,9 @@ enum nfs4_client_state {
 };
 
 /*
- * The nfs4_client identifies our client state to the server.
+ * The nfs_client identifies our client state to the server.
  */
-struct nfs4_client {
+struct nfs_client {
 	struct list_head	cl_servers;	/* Global list of servers */
 	struct in_addr		cl_addr;	/* Server identifier */
 	u64			cl_clientid;	/* constant */
@@ -127,7 +127,7 @@ static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status
 struct nfs4_state_owner {
 	spinlock_t	     so_lock;
 	struct list_head     so_list;	 /* per-clientid list of state_owners */
-	struct nfs4_client   *so_client;
+	struct nfs_client    *so_client;
 	u32                  so_id;      /* 32-bit identifier, unique */
 	atomic_t	     so_count;
 
@@ -210,10 +210,10 @@ extern ssize_t nfs4_listxattr(struct dentry *, char *, size_t);
 
 /* nfs4proc.c */
 extern int nfs4_map_errors(int err);
-extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short, struct rpc_cred *);
-extern int nfs4_proc_setclientid_confirm(struct nfs4_client *, struct rpc_cred *);
-extern int nfs4_proc_async_renew(struct nfs4_client *, struct rpc_cred *);
-extern int nfs4_proc_renew(struct nfs4_client *, struct rpc_cred *);
+extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *);
+extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
@@ -231,19 +231,19 @@ extern const u32 nfs4_fsinfo_bitmap[2];
 extern const u32 nfs4_fs_locations_bitmap[2];
 
 /* nfs4renewd.c */
-extern void nfs4_schedule_state_renewal(struct nfs4_client *);
+extern void nfs4_schedule_state_renewal(struct nfs_client *);
 extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
-extern void nfs4_kill_renewd(struct nfs4_client *);
+extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(void *);
 
 /* nfs4state.c */
 extern void init_nfsv4_state(struct nfs_server *);
 extern void destroy_nfsv4_state(struct nfs_server *);
-extern struct nfs4_client *nfs4_get_client(struct in_addr *);
-extern void nfs4_put_client(struct nfs4_client *clp);
-extern struct nfs4_client *nfs4_find_client(struct in_addr *);
-struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp);
-extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *);
+extern struct nfs_client *nfs4_get_client(struct in_addr *);
+extern void nfs4_put_client(struct nfs_client *clp);
+extern struct nfs_client *nfs4_find_client(struct in_addr *);
+struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp);
+extern u32 nfs4_alloc_lockowner_id(struct nfs_client *);
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
 extern void nfs4_put_state_owner(struct nfs4_state_owner *);
@@ -252,7 +252,7 @@ extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state
 extern void nfs4_put_open_state(struct nfs4_state *);
 extern void nfs4_close_state(struct nfs4_state *, mode_t);
 extern void nfs4_state_set_mode_locked(struct nfs4_state *, mode_t);
-extern void nfs4_schedule_state_recovery(struct nfs4_client *);
+extern void nfs4_schedule_state_recovery(struct nfs_client *);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b14145b7b87f..168f3ffb059f 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -64,7 +64,7 @@ static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinf
 static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *);
 static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry);
 static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception);
-static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp);
+static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp);
 
 /* Prevent leaks of NFSv4 errors into userland */
 int nfs4_map_errors(int err)
@@ -195,7 +195,7 @@ static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry,
 
 static void renew_lease(const struct nfs_server *server, unsigned long timestamp)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	spin_lock(&clp->cl_lock);
 	if (time_before(clp->cl_last_renewal,timestamp))
 		clp->cl_last_renewal = timestamp;
@@ -792,7 +792,7 @@ out:
 
 int nfs4_recover_expired_lease(struct nfs_server *server)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 
 	if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		nfs4_schedule_state_recovery(clp);
@@ -867,7 +867,7 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred
 {
 	struct nfs_delegation *delegation;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs4_state_owner *sp = NULL;
 	struct nfs4_state *state = NULL;
@@ -953,7 +953,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st
 	struct nfs4_state_owner  *sp;
 	struct nfs4_state     *state = NULL;
 	struct nfs_server       *server = NFS_SERVER(dir);
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs4_opendata *opendata;
 	int                     status;
 
@@ -2521,7 +2521,7 @@ static void nfs4_proc_commit_setup(struct nfs_write_data *data, int how)
  */
 static void nfs4_renew_done(struct rpc_task *task, void *data)
 {
-	struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp;
+	struct nfs_client *clp = (struct nfs_client *)task->tk_msg.rpc_argp;
 	unsigned long timestamp = (unsigned long)data;
 
 	if (task->tk_status < 0) {
@@ -2543,7 +2543,7 @@ static const struct rpc_call_ops nfs4_renew_ops = {
 	.rpc_call_done = nfs4_renew_done,
 };
 
-int nfs4_proc_async_renew(struct nfs4_client *clp, struct rpc_cred *cred)
+int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_RENEW],
@@ -2555,7 +2555,7 @@ int nfs4_proc_async_renew(struct nfs4_client *clp, struct rpc_cred *cred)
 			&nfs4_renew_ops, (void *)jiffies);
 }
 
-int nfs4_proc_renew(struct nfs4_client *clp, struct rpc_cred *cred)
+int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_RENEW],
@@ -2791,7 +2791,7 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 static int
 nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 
 	if (!clp || task->tk_status >= 0)
 		return 0;
@@ -2828,7 +2828,7 @@ static int nfs4_wait_bit_interruptible(void *word)
 	return 0;
 }
 
-static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp)
+static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp)
 {
 	sigset_t oldset;
 	int res;
@@ -2871,7 +2871,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
  */
 int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	int ret = errorcode;
 
 	exception->retry = 0;
@@ -2898,7 +2898,7 @@ int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct
 	return nfs4_map_errors(ret);
 }
 
-int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
+int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
 {
 	nfs4_verifier sc_verifier;
 	struct nfs4_setclientid setclientid = {
@@ -2945,7 +2945,7 @@ int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short p
 	return status;
 }
 
-static int _nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cred *cred)
+static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct nfs_fsinfo fsinfo;
 	struct rpc_message msg = {
@@ -2969,7 +2969,7 @@ static int _nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cr
 	return status;
 }
 
-int nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cred *cred)
+int nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	long timeout;
 	int err;
@@ -3106,7 +3106,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 {
 	struct inode *inode = state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs_lockt_args arg = {
 		.fh = NFS_FH(inode),
 		.fl = request,
@@ -3513,7 +3513,7 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request
 
 static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request)
 {
-	struct nfs4_client *clp = state->owner->so_client;
+	struct nfs_client *clp = state->owner->so_client;
 	unsigned char fl_flags = request->fl_flags;
 	int status;
 
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 5d764d8e6d8a..208764069f61 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -61,7 +61,7 @@
 void
 nfs4_renew_state(void *data)
 {
-	struct nfs4_client *clp = (struct nfs4_client *)data;
+	struct nfs_client *clp = (struct nfs_client *)data;
 	struct rpc_cred *cred;
 	long lease, timeout;
 	unsigned long last, now;
@@ -108,7 +108,7 @@ out:
 
 /* Must be called with clp->cl_sem locked for writes */
 void
-nfs4_schedule_state_renewal(struct nfs4_client *clp)
+nfs4_schedule_state_renewal(struct nfs_client *clp)
 {
 	long timeout;
 
@@ -127,7 +127,7 @@ nfs4_schedule_state_renewal(struct nfs4_client *clp)
 void
 nfs4_renewd_prepare_shutdown(struct nfs_server *server)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 
 	if (!clp)
 		return;
@@ -140,7 +140,7 @@ nfs4_renewd_prepare_shutdown(struct nfs_server *server)
 
 /* Must be called with clp->cl_sem locked for writes */
 void
-nfs4_kill_renewd(struct nfs4_client *clp)
+nfs4_kill_renewd(struct nfs_client *clp)
 {
 	down_read(&clp->cl_sem);
 	if (!list_empty(&clp->cl_superblocks)) {
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 090a36b07a22..c0b6439f1f71 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -83,10 +83,10 @@ destroy_nfsv4_state(struct nfs_server *server)
  * Since these are allocated/deallocated very rarely, we don't
  * bother putting them in a slab cache...
  */
-static struct nfs4_client *
+static struct nfs_client *
 nfs4_alloc_client(struct in_addr *addr)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 
 	if (nfs_callback_up() < 0)
 		return NULL;
@@ -111,7 +111,7 @@ nfs4_alloc_client(struct in_addr *addr)
 }
 
 static void
-nfs4_free_client(struct nfs4_client *clp)
+nfs4_free_client(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
 
@@ -130,9 +130,9 @@ nfs4_free_client(struct nfs4_client *clp)
 	nfs_callback_down();
 }
 
-static struct nfs4_client *__nfs4_find_client(struct in_addr *addr)
+static struct nfs_client *__nfs4_find_client(struct in_addr *addr)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	list_for_each_entry(clp, &nfs4_clientid_list, cl_servers) {
 		if (memcmp(&clp->cl_addr, addr, sizeof(clp->cl_addr)) == 0) {
 			atomic_inc(&clp->cl_count);
@@ -142,19 +142,19 @@ static struct nfs4_client *__nfs4_find_client(struct in_addr *addr)
 	return NULL;
 }
 
-struct nfs4_client *nfs4_find_client(struct in_addr *addr)
+struct nfs_client *nfs4_find_client(struct in_addr *addr)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	spin_lock(&state_spinlock);
 	clp = __nfs4_find_client(addr);
 	spin_unlock(&state_spinlock);
 	return clp;
 }
 
-struct nfs4_client *
+struct nfs_client *
 nfs4_get_client(struct in_addr *addr)
 {
-	struct nfs4_client *clp, *new = NULL;
+	struct nfs_client *clp, *new = NULL;
 
 	spin_lock(&state_spinlock);
 	for (;;) {
@@ -180,7 +180,7 @@ nfs4_get_client(struct in_addr *addr)
 }
 
 void
-nfs4_put_client(struct nfs4_client *clp)
+nfs4_put_client(struct nfs_client *clp)
 {
 	if (!atomic_dec_and_lock(&clp->cl_count, &state_spinlock))
 		return;
@@ -192,7 +192,7 @@ nfs4_put_client(struct nfs4_client *clp)
 	nfs4_free_client(clp);
 }
 
-static int nfs4_init_client(struct nfs4_client *clp, struct rpc_cred *cred)
+static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	int status = nfs4_proc_setclientid(clp, NFS4_CALLBACK,
 			nfs_callback_tcpport, cred);
@@ -204,13 +204,13 @@ static int nfs4_init_client(struct nfs4_client *clp, struct rpc_cred *cred)
 }
 
 u32
-nfs4_alloc_lockowner_id(struct nfs4_client *clp)
+nfs4_alloc_lockowner_id(struct nfs_client *clp)
 {
 	return clp->cl_lockowner_id ++;
 }
 
 static struct nfs4_state_owner *
-nfs4_client_grab_unused(struct nfs4_client *clp, struct rpc_cred *cred)
+nfs4_client_grab_unused(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct nfs4_state_owner *sp = NULL;
 
@@ -224,7 +224,7 @@ nfs4_client_grab_unused(struct nfs4_client *clp, struct rpc_cred *cred)
 	return sp;
 }
 
-struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp)
+struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
 	struct rpc_cred *cred = NULL;
@@ -238,7 +238,7 @@ struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp)
 	return cred;
 }
 
-struct rpc_cred *nfs4_get_setclientid_cred(struct nfs4_client *clp)
+struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
 
@@ -251,7 +251,7 @@ struct rpc_cred *nfs4_get_setclientid_cred(struct nfs4_client *clp)
 }
 
 static struct nfs4_state_owner *
-nfs4_find_state_owner(struct nfs4_client *clp, struct rpc_cred *cred)
+nfs4_find_state_owner(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct nfs4_state_owner *sp, *res = NULL;
 
@@ -294,7 +294,7 @@ nfs4_alloc_state_owner(void)
 void
 nfs4_drop_state_owner(struct nfs4_state_owner *sp)
 {
-	struct nfs4_client *clp = sp->so_client;
+	struct nfs_client *clp = sp->so_client;
 	spin_lock(&clp->cl_lock);
 	list_del_init(&sp->so_list);
 	spin_unlock(&clp->cl_lock);
@@ -306,7 +306,7 @@ nfs4_drop_state_owner(struct nfs4_state_owner *sp)
  */
 struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct rpc_cred *cred)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs4_state_owner *sp, *new;
 
 	get_rpccred(cred);
@@ -337,7 +337,7 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct
  */
 void nfs4_put_state_owner(struct nfs4_state_owner *sp)
 {
-	struct nfs4_client *clp = sp->so_client;
+	struct nfs_client *clp = sp->so_client;
 	struct rpc_cred *cred = sp->so_cred;
 
 	if (!atomic_dec_and_lock(&sp->so_count, &clp->cl_lock))
@@ -540,7 +540,7 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
 static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
 {
 	struct nfs4_lock_state *lsp;
-	struct nfs4_client *clp = state->owner->so_client;
+	struct nfs_client *clp = state->owner->so_client;
 
 	lsp = kzalloc(sizeof(*lsp), GFP_KERNEL);
 	if (lsp == NULL)
@@ -752,7 +752,7 @@ out:
 
 static int reclaimer(void *);
 
-static inline void nfs4_clear_recover_bit(struct nfs4_client *clp)
+static inline void nfs4_clear_recover_bit(struct nfs_client *clp)
 {
 	smp_mb__before_clear_bit();
 	clear_bit(NFS4CLNT_STATE_RECOVER, &clp->cl_state);
@@ -764,7 +764,7 @@ static inline void nfs4_clear_recover_bit(struct nfs4_client *clp)
 /*
  * State recovery routine
  */
-static void nfs4_recover_state(struct nfs4_client *clp)
+static void nfs4_recover_state(struct nfs_client *clp)
 {
 	struct task_struct *task;
 
@@ -782,7 +782,7 @@ static void nfs4_recover_state(struct nfs4_client *clp)
 /*
  * Schedule a state recovery attempt
  */
-void nfs4_schedule_state_recovery(struct nfs4_client *clp)
+void nfs4_schedule_state_recovery(struct nfs_client *clp)
 {
 	if (!clp)
 		return;
@@ -879,7 +879,7 @@ out_err:
 	return status;
 }
 
-static void nfs4_state_mark_reclaim(struct nfs4_client *clp)
+static void nfs4_state_mark_reclaim(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
 	struct nfs4_state *state;
@@ -903,7 +903,7 @@ static void nfs4_state_mark_reclaim(struct nfs4_client *clp)
 
 static int reclaimer(void *ptr)
 {
-	struct nfs4_client *clp = ptr;
+	struct nfs_client *clp = ptr;
 	struct nfs4_state_owner *sp;
 	struct nfs4_state_recovery_ops *ops;
 	struct rpc_cred *cred;
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1dee6ef7e5a9..04748ab9ed55 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1160,7 +1160,7 @@ static int encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, con
 	return 0;
 }
 
-static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client_stateid)
+static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid)
 {
 	uint32_t *p;
 
@@ -1246,7 +1246,7 @@ static int encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclien
 	return 0;
 }
 
-static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4_client *client_state)
+static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state)
 {
         uint32_t *p;
 
@@ -1945,7 +1945,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, uint32_t *p, const str
 /*
  * a RENEW request
  */
-static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs4_client *clp)
+static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -1975,7 +1975,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, uint32_t *p, struct nf
 /*
  * a SETCLIENTID_CONFIRM request
  */
-static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs4_client *clp)
+static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -2132,7 +2132,7 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected)
 }
 
 /* Dummy routine */
-static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs4_client *clp)
+static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp)
 {
 	uint32_t *p;
 	unsigned int strlen;
@@ -2636,7 +2636,7 @@ static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t
 	return 0;
 }
 
-static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_client *clp, int32_t *uid)
+static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *uid)
 {
 	uint32_t len, *p;
 
@@ -2660,7 +2660,7 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf
 	return 0;
 }
 
-static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_client *clp, int32_t *gid)
+static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *gid)
 {
 	uint32_t len, *p;
 
@@ -3565,7 +3565,7 @@ static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res)
 	return 0;
 }
 
-static int decode_setclientid(struct xdr_stream *xdr, struct nfs4_client *clp)
+static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp)
 {
 	uint32_t *p;
 	uint32_t opnum;
@@ -4335,7 +4335,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, uint32_t *p, void *dummy)
  * a SETCLIENTID request
  */
 static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p,
-		struct nfs4_client *clp)
+		struct nfs_client *clp)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 63497345806b..d03ede5b1aca 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1099,7 +1099,7 @@ static int nfs_clone_nfs_sb(struct file_system_type *fs_type,
 static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 	struct rpc_timeout *timeparms, int proto, rpc_authflavor_t flavor)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	struct rpc_xprt *xprt = NULL;
 	struct rpc_clnt *clnt = NULL;
 	int err = -EIO;
@@ -1416,7 +1416,7 @@ static inline char *nfs4_dup_path(const struct dentry *dentry)
 static struct super_block *nfs4_clone_sb(struct nfs_server *server, struct nfs_clone_mount *data)
 {
 	const struct dentry *dentry = data->dentry;
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct super_block *sb;
 
 	server->fsid = data->fattr->fsid;
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 6b4a13c79474..4db90df2aed0 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -43,7 +43,7 @@ struct nfs_server {
 	 */
 	char			ip_addr[16];
 	char *			mnt_path;
-	struct nfs4_client *	nfs4_state;	/* all NFSv4 state starts here */
+	struct nfs_client *	nfs4_state;	/* all NFSv4 state starts here */
 	struct list_head	nfs4_siblings;	/* List of other nfs_server structs
 						 * that share the same clientid
 						 */
diff --git a/include/linux/nfs_idmap.h b/include/linux/nfs_idmap.h
index 102e56094296..678fe68982ef 100644
--- a/include/linux/nfs_idmap.h
+++ b/include/linux/nfs_idmap.h
@@ -62,15 +62,15 @@ struct idmap_msg {
 #ifdef __KERNEL__
 
 /* Forward declaration to make this header independent of others */
-struct nfs4_client;
+struct nfs_client;
 
-void nfs_idmap_new(struct nfs4_client *);
-void nfs_idmap_delete(struct nfs4_client *);
+void nfs_idmap_new(struct nfs_client *);
+void nfs_idmap_delete(struct nfs_client *);
 
-int nfs_map_name_to_uid(struct nfs4_client *, const char *, size_t, __u32 *);
-int nfs_map_group_to_gid(struct nfs4_client *, const char *, size_t, __u32 *);
-int nfs_map_uid_to_name(struct nfs4_client *, __u32, char *);
-int nfs_map_gid_to_group(struct nfs4_client *, __u32, char *);
+int nfs_map_name_to_uid(struct nfs_client *, const char *, size_t, __u32 *);
+int nfs_map_group_to_gid(struct nfs_client *, const char *, size_t, __u32 *);
+int nfs_map_uid_to_name(struct nfs_client *, __u32, char *);
+int nfs_map_gid_to_group(struct nfs_client *, __u32, char *);
 
 extern unsigned int nfs_idmap_cache_timeout;
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From 7539bbab8062aadc1db95a22b377146843cfa88f Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:09 -0400
Subject: NFS: Rename nfs_server::nfs4_state

Rename nfs_server::nfs4_state to nfs_client as it will be used to represent the
client state for NFS2 and NFS3 also.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/delegation.c       | 12 ++++++------
 fs/nfs/nfs4proc.c         | 26 +++++++++++++-------------
 fs/nfs/nfs4renewd.c       |  2 +-
 fs/nfs/nfs4state.c        | 10 +++++-----
 fs/nfs/nfs4xdr.c          | 10 +++++-----
 fs/nfs/super.c            |  6 +++---
 include/linux/nfs_fs_sb.h |  2 +-
 7 files changed, 34 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 5a1105c258bd..cfe239736ac0 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -52,7 +52,7 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_
 			case -NFS4ERR_EXPIRED:
 				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
 			case -NFS4ERR_STALE_CLIENTID:
-				nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs4_state);
+				nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs_client);
 				goto out_err;
 		}
 	}
@@ -114,7 +114,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st
  */
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res)
 {
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int status = 0;
@@ -176,7 +176,7 @@ static void nfs_msync_inode(struct inode *inode)
  */
 int __nfs_inode_return_delegation(struct inode *inode)
 {
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
@@ -208,7 +208,7 @@ int __nfs_inode_return_delegation(struct inode *inode)
  */
 void nfs_return_all_delegations(struct super_block *sb)
 {
-	struct nfs_client *clp = NFS_SB(sb)->nfs4_state;
+	struct nfs_client *clp = NFS_SB(sb)->nfs_client;
 	struct nfs_delegation *delegation;
 	struct inode *inode;
 
@@ -310,7 +310,7 @@ static int recall_thread(void *data)
 {
 	struct recall_threadargs *args = (struct recall_threadargs *)data;
 	struct inode *inode = igrab(args->inode);
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 
@@ -423,7 +423,7 @@ void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
 
 int nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode)
 {
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 168f3ffb059f..b46597fc81e1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -195,7 +195,7 @@ static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry,
 
 static void renew_lease(const struct nfs_server *server, unsigned long timestamp)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	spin_lock(&clp->cl_lock);
 	if (time_before(clp->cl_last_renewal,timestamp))
 		clp->cl_last_renewal = timestamp;
@@ -252,7 +252,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry,
 	atomic_inc(&sp->so_count);
 	p->o_arg.fh = NFS_FH(dir);
 	p->o_arg.open_flags = flags,
-	p->o_arg.clientid = server->nfs4_state->cl_clientid;
+	p->o_arg.clientid = server->nfs_client->cl_clientid;
 	p->o_arg.id = sp->so_id;
 	p->o_arg.name = &dentry->d_name;
 	p->o_arg.server = server;
@@ -550,7 +550,7 @@ int nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state)
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
 				/* Don't recall a delegation if it was lost */
-				nfs4_schedule_state_recovery(server->nfs4_state);
+				nfs4_schedule_state_recovery(server->nfs_client);
 				return err;
 		}
 		err = nfs4_handle_exception(server, err, &exception);
@@ -792,7 +792,7 @@ out:
 
 int nfs4_recover_expired_lease(struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 
 	if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		nfs4_schedule_state_recovery(clp);
@@ -867,7 +867,7 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred
 {
 	struct nfs_delegation *delegation;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs4_state_owner *sp = NULL;
 	struct nfs4_state *state = NULL;
@@ -953,7 +953,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st
 	struct nfs4_state_owner  *sp;
 	struct nfs4_state     *state = NULL;
 	struct nfs_server       *server = NFS_SERVER(dir);
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs4_opendata *opendata;
 	int                     status;
 
@@ -1133,7 +1133,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 			break;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
-			nfs4_schedule_state_recovery(server->nfs4_state);
+			nfs4_schedule_state_recovery(server->nfs_client);
 			break;
 		default:
 			if (nfs4_async_handle_error(task, server) == -EAGAIN) {
@@ -2791,7 +2791,7 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 static int
 nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 
 	if (!clp || task->tk_status >= 0)
 		return 0;
@@ -2871,7 +2871,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
  */
 int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	int ret = errorcode;
 
 	exception->retry = 0;
@@ -3077,7 +3077,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4
 		switch (err) {
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
-				nfs4_schedule_state_recovery(server->nfs4_state);
+				nfs4_schedule_state_recovery(server->nfs_client);
 			case 0:
 				return 0;
 		}
@@ -3106,7 +3106,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 {
 	struct inode *inode = state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs_lockt_args arg = {
 		.fh = NFS_FH(inode),
 		.fl = request,
@@ -3231,7 +3231,7 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			break;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
-			nfs4_schedule_state_recovery(calldata->server->nfs4_state);
+			nfs4_schedule_state_recovery(calldata->server->nfs_client);
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server) == -EAGAIN) {
@@ -3343,7 +3343,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	if (p->arg.lock_seqid == NULL)
 		goto out_free;
 	p->arg.lock_stateid = &lsp->ls_stateid;
-	p->arg.lock_owner.clientid = server->nfs4_state->cl_clientid;
+	p->arg.lock_owner.clientid = server->nfs_client->cl_clientid;
 	p->arg.lock_owner.id = lsp->ls_id;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 208764069f61..ff947ecb8b81 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -127,7 +127,7 @@ nfs4_schedule_state_renewal(struct nfs_client *clp)
 void
 nfs4_renewd_prepare_shutdown(struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 
 	if (!clp)
 		return;
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index c0b6439f1f71..fa51a7d4c022 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -61,7 +61,7 @@ static LIST_HEAD(nfs4_clientid_list);
 void
 init_nfsv4_state(struct nfs_server *server)
 {
-	server->nfs4_state = NULL;
+	server->nfs_client = NULL;
 	INIT_LIST_HEAD(&server->nfs4_siblings);
 }
 
@@ -70,9 +70,9 @@ destroy_nfsv4_state(struct nfs_server *server)
 {
 	kfree(server->mnt_path);
 	server->mnt_path = NULL;
-	if (server->nfs4_state) {
-		nfs4_put_client(server->nfs4_state);
-		server->nfs4_state = NULL;
+	if (server->nfs_client) {
+		nfs4_put_client(server->nfs_client);
+		server->nfs_client = NULL;
 	}
 }
 
@@ -306,7 +306,7 @@ nfs4_drop_state_owner(struct nfs4_state_owner *sp)
  */
 struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct rpc_cred *cred)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs4_state_owner *sp, *new;
 
 	get_rpccred(cred);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 04748ab9ed55..99926067eca4 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -529,7 +529,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s
 	if (iap->ia_valid & ATTR_MODE)
 		len += 4;
 	if (iap->ia_valid & ATTR_UID) {
-		owner_namelen = nfs_map_uid_to_name(server->nfs4_state, iap->ia_uid, owner_name);
+		owner_namelen = nfs_map_uid_to_name(server->nfs_client, iap->ia_uid, owner_name);
 		if (owner_namelen < 0) {
 			printk(KERN_WARNING "nfs: couldn't resolve uid %d to string\n",
 			       iap->ia_uid);
@@ -541,7 +541,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s
 		len += 4 + (XDR_QUADLEN(owner_namelen) << 2);
 	}
 	if (iap->ia_valid & ATTR_GID) {
-		owner_grouplen = nfs_map_gid_to_group(server->nfs4_state, iap->ia_gid, owner_group);
+		owner_grouplen = nfs_map_gid_to_group(server->nfs_client, iap->ia_gid, owner_group);
 		if (owner_grouplen < 0) {
 			printk(KERN_WARNING "nfs4: couldn't resolve gid %d to string\n",
 			       iap->ia_gid);
@@ -3051,9 +3051,9 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons
 	fattr->mode |= fmode;
 	if ((status = decode_attr_nlink(xdr, bitmap, &fattr->nlink)) != 0)
 		goto xdr_error;
-	if ((status = decode_attr_owner(xdr, bitmap, server->nfs4_state, &fattr->uid)) != 0)
+	if ((status = decode_attr_owner(xdr, bitmap, server->nfs_client, &fattr->uid)) != 0)
 		goto xdr_error;
-	if ((status = decode_attr_group(xdr, bitmap, server->nfs4_state, &fattr->gid)) != 0)
+	if ((status = decode_attr_group(xdr, bitmap, server->nfs_client, &fattr->gid)) != 0)
 		goto xdr_error;
 	if ((status = decode_attr_rdev(xdr, bitmap, &fattr->rdev)) != 0)
 		goto xdr_error;
@@ -3254,7 +3254,7 @@ static int decode_delegation(struct xdr_stream *xdr, struct nfs_openres *res)
 			if (decode_space_limit(xdr, &res->maxsize) < 0)
 				return -EIO;
 	}
-	return decode_ace(xdr, NULL, res->server->nfs4_state);
+	return decode_ace(xdr, NULL, res->server->nfs_client);
 }
 
 static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res)
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index d03ede5b1aca..ab4c78ee840c 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1141,7 +1141,7 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
 	clnt = rpc_clone_client(clp->cl_rpcclient);
 	if (!IS_ERR(clnt))
-		server->nfs4_state = clp;
+		server->nfs_client = clp;
 	up_write(&clp->cl_sem);
 	clp = NULL;
 
@@ -1151,7 +1151,7 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 		return clnt;
 	}
 
-	if (server->nfs4_state->cl_idmap == NULL) {
+	if (server->nfs_client->cl_idmap == NULL) {
 		dprintk("%s: failed to create idmapper.\n", __FUNCTION__);
 		return ERR_PTR(-ENOMEM);
 	}
@@ -1416,7 +1416,7 @@ static inline char *nfs4_dup_path(const struct dentry *dentry)
 static struct super_block *nfs4_clone_sb(struct nfs_server *server, struct nfs_clone_mount *data)
 {
 	const struct dentry *dentry = data->dentry;
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct super_block *sb;
 
 	server->fsid = data->fattr->fsid;
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 4db90df2aed0..fc20d6b934fb 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -43,7 +43,7 @@ struct nfs_server {
 	 */
 	char			ip_addr[16];
 	char *			mnt_path;
-	struct nfs_client *	nfs4_state;	/* all NFSv4 state starts here */
+	struct nfs_client *	nfs_client;	/* all NFSv4 state starts here */
 	struct list_head	nfs4_siblings;	/* List of other nfs_server structs
 						 * that share the same clientid
 						 */
-- 
cgit v1.2.3


From b7162792b5c0e0f6e91b8997f8e6bbc76ec5420a Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:09 -0400
Subject: NFS: Return an error when starting the idmapping pipe

Return an error when starting the idmapping pipe so that we can detect it
failing.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/idmap.c            | 12 ++++++++----
 fs/nfs/super.c            |  3 ++-
 include/linux/nfs_idmap.h |  2 +-
 3 files changed, 11 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c
index d05148ec9414..231c20ffc0ff 100644
--- a/fs/nfs/idmap.c
+++ b/fs/nfs/idmap.c
@@ -108,15 +108,17 @@ static struct rpc_pipe_ops idmap_upcall_ops = {
         .destroy_msg    = idmap_pipe_destroy_msg,
 };
 
-void
+int
 nfs_idmap_new(struct nfs_client *clp)
 {
 	struct idmap *idmap;
+	int error;
 
 	if (clp->cl_idmap != NULL)
-		return;
+		return 0;
+
         if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL)
-                return;
+                return -ENOMEM;
 
 	snprintf(idmap->idmap_path, sizeof(idmap->idmap_path),
 	    "%s/idmap", clp->cl_rpcclient->cl_pathname);
@@ -124,8 +126,9 @@ nfs_idmap_new(struct nfs_client *clp)
         idmap->idmap_dentry = rpc_mkpipe(idmap->idmap_path,
 	    idmap, &idmap_upcall_ops, 0);
         if (IS_ERR(idmap->idmap_dentry)) {
+		error = PTR_ERR(idmap->idmap_dentry);
 		kfree(idmap);
-		return;
+		return error;
 	}
 
         mutex_init(&idmap->idmap_lock);
@@ -135,6 +138,7 @@ nfs_idmap_new(struct nfs_client *clp)
 	idmap->idmap_group_hash.h_type = IDMAP_TYPE_GROUP;
 
 	clp->cl_idmap = idmap;
+	return 0;
 }
 
 void
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index ab4c78ee840c..3ee85c4e65d8 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1136,7 +1136,8 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 		clnt->cl_softrtry = 1;
 		clp->cl_rpcclient = clnt;
 		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
-		nfs_idmap_new(clp);
+		if (nfs_idmap_new(clp) < 0)
+			goto out_fail;
 	}
 	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
 	clnt = rpc_clone_client(clp->cl_rpcclient);
diff --git a/include/linux/nfs_idmap.h b/include/linux/nfs_idmap.h
index 678fe68982ef..15a9f3b7289a 100644
--- a/include/linux/nfs_idmap.h
+++ b/include/linux/nfs_idmap.h
@@ -64,7 +64,7 @@ struct idmap_msg {
 /* Forward declaration to make this header independent of others */
 struct nfs_client;
 
-void nfs_idmap_new(struct nfs_client *);
+int nfs_idmap_new(struct nfs_client *);
 void nfs_idmap_delete(struct nfs_client *);
 
 int nfs_map_name_to_uid(struct nfs_client *, const char *, size_t, __u32 *);
-- 
cgit v1.2.3


From 2b3de4411b3ccaeb00018c99d1bbe7203554cf7f Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:09 -0400
Subject: NFS: Add a lookupfh NFS RPC op

Add a lookup filehandle NFS RPC op so that a file handle can be looked up
without requiring dentries and inodes and other VFS stuff when doing an NFS4
pathwalk during mounting.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs_xdr.h |  3 +++
 2 files changed, 50 insertions(+)

(limited to 'include')

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b46597fc81e1..de2006f754ef 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1583,6 +1583,52 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 	return status;
 }
 
+static int _nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
+		struct qstr *name, struct nfs_fh *fhandle,
+		struct nfs_fattr *fattr)
+{
+	int		       status;
+	struct nfs4_lookup_arg args = {
+		.bitmask = server->attr_bitmask,
+		.dir_fh = dirfh,
+		.name = name,
+	};
+	struct nfs4_lookup_res res = {
+		.server = server,
+		.fattr = fattr,
+		.fh = fhandle,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+
+	nfs_fattr_init(fattr);
+
+	dprintk("NFS call  lookupfh %s\n", name->name);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply lookupfh: %d\n", status);
+	if (status == -NFS4ERR_MOVED)
+		status = -EREMOTE;
+	return status;
+}
+
+static int nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
+			      struct qstr *name, struct nfs_fh *fhandle,
+			      struct nfs_fattr *fattr)
+{
+	struct nfs4_exception exception = { };
+	int err;
+	do {
+		err = nfs4_handle_exception(server,
+				_nfs4_proc_lookupfh(server, dirfh, name,
+						    fhandle, fattr),
+				&exception);
+	} while (exception.retry);
+	return err;
+}
+
 static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name,
 		struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
@@ -3723,6 +3769,7 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
 	.getroot	= nfs4_proc_get_root,
 	.getattr	= nfs4_proc_getattr,
 	.setattr	= nfs4_proc_setattr,
+	.lookupfh	= nfs4_proc_lookupfh,
 	.lookup		= nfs4_proc_lookup,
 	.access		= nfs4_proc_access,
 	.readlink	= nfs4_proc_readlink,
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 41e5a19199e9..26879771831d 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -770,6 +770,9 @@ struct nfs_rpc_ops {
 
 	int	(*getroot) (struct nfs_server *, struct nfs_fh *,
 			    struct nfs_fsinfo *);
+	int	(*lookupfh)(struct nfs_server *, struct nfs_fh *,
+			    struct qstr *, struct nfs_fh *,
+			    struct nfs_fattr *);
 	int	(*getattr) (struct nfs_server *, struct nfs_fh *,
 			    struct nfs_fattr *);
 	int	(*setattr) (struct dentry *, struct nfs_fattr *,
-- 
cgit v1.2.3


From e9326dcab413848e70ab746c7c5363da13e5f801 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:10 -0400
Subject: NFS: Add a server capabilities NFS RPC op

Add a set_capabilities NFS RPC op so that the server capabilities can be set.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 1 +
 include/linux/nfs_xdr.h | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index de2006f754ef..850f0851023a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3790,6 +3790,7 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
 	.statfs		= nfs4_proc_statfs,
 	.fsinfo		= nfs4_proc_fsinfo,
 	.pathconf	= nfs4_proc_pathconf,
+	.set_capabilities = nfs4_server_capabilities,
 	.decode_dirent	= nfs4_decode_dirent,
 	.read_setup	= nfs4_proc_read_setup,
 	.read_done	= nfs4_read_done,
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 26879771831d..dd9ae6761f71 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -809,6 +809,7 @@ struct nfs_rpc_ops {
 			    struct nfs_fsinfo *);
 	int	(*pathconf) (struct nfs_server *, struct nfs_fh *,
 			     struct nfs_pathconf *);
+	int	(*set_capabilities)(struct nfs_server *, struct nfs_fh *);
 	u32 *	(*decode_dirent)(u32 *, struct nfs_entry *, int plus);
 	void	(*read_setup)   (struct nfs_read_data *);
 	int	(*read_done)  (struct rpc_task *, struct nfs_read_data *);
-- 
cgit v1.2.3


From 24c8dbbb5f777187d660393599641ab3307b4b97 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:10 -0400
Subject: NFS: Generalise the nfs_client structure

Generalise the nfs_client structure by:

 (1) Moving nfs_client to a more general place (nfs_fs_sb.h).

 (2) Renaming its maintenance routines to be non-NFS4 specific.

 (3) Move those maintenance routines to a new non-NFS4 specific file (client.c)
     and move the declarations to internal.h.

 (4) Make nfs_find/get_client() take a full sockaddr_in to include the port
     number (will be required for NFS2/3).

 (5) Make nfs_find/get_client() take the NFS protocol version (again will be
     required to differentiate NFS2, 3 & 4 client records).

Also:

 (6) Make nfs_client construction proceed akin to inodes, marking them as under
     construction and providing a function to indicate completion.

 (7) Make nfs_get_client() wait interruptibly if it finds a client that it can
     share, but that client is currently being constructed.

 (8) Make nfs4_create_client() use (6) and (7) instead of locking cl_sem.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/Makefile           |   6 +-
 fs/nfs/callback.c         |   9 +-
 fs/nfs/callback_proc.c    |   9 +-
 fs/nfs/client.c           | 312 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/delegation.c       |   9 +-
 fs/nfs/internal.h         |   6 +
 fs/nfs/nfs4_fs.h          |  52 --------
 fs/nfs/nfs4proc.c         |   2 +-
 fs/nfs/nfs4state.c        | 128 +------------------
 fs/nfs/super.c            |  53 ++++----
 include/linux/nfs_fs.h    |   1 +
 include/linux/nfs_fs_sb.h |  60 +++++++++
 12 files changed, 425 insertions(+), 222 deletions(-)
 create mode 100644 fs/nfs/client.c

(limited to 'include')

diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile
index 0b572a0c1967..3b993a6f8163 100644
--- a/fs/nfs/Makefile
+++ b/fs/nfs/Makefile
@@ -4,9 +4,9 @@
 
 obj-$(CONFIG_NFS_FS) += nfs.o
 
-nfs-y 			:= dir.o file.o inode.o super.o nfs2xdr.o pagelist.o \
-			   proc.o read.o symlink.o unlink.o write.o \
-			   namespace.o
+nfs-y 			:= client.o dir.o file.o inode.o super.o nfs2xdr.o \
+			   pagelist.o proc.o read.o symlink.o unlink.o \
+			   write.o namespace.o
 nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o mount_clnt.o      
 nfs-$(CONFIG_NFS_V3)	+= nfs3proc.o nfs3xdr.o
 nfs-$(CONFIG_NFS_V3_ACL)	+= nfs3acl.o
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 1b596b6d9dc2..a3ee11364db0 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -19,6 +19,7 @@
 
 #include "nfs4_fs.h"
 #include "callback.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
 
@@ -166,15 +167,15 @@ void nfs_callback_down(void)
 
 static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 {
-	struct in_addr *addr = &rqstp->rq_addr.sin_addr;
+	struct sockaddr_in *addr = &rqstp->rq_addr;
 	struct nfs_client *clp;
 
 	/* Don't talk to strangers */
-	clp = nfs4_find_client(addr);
+	clp = nfs_find_client(addr, 4);
 	if (clp == NULL)
 		return SVC_DROP;
-	dprintk("%s: %u.%u.%u.%u NFSv4 callback!\n", __FUNCTION__, NIPQUAD(addr));
-	nfs4_put_client(clp);
+	dprintk("%s: %u.%u.%u.%u NFSv4 callback!\n", __FUNCTION__, NIPQUAD(addr->sin_addr));
+	nfs_put_client(clp);
 	switch (rqstp->rq_authop->flavour) {
 		case RPC_AUTH_NULL:
 			if (rqstp->rq_proc != CB_NULL)
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index 55d6e2ec157f..97cf8f71451f 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -10,6 +10,7 @@
 #include "nfs4_fs.h"
 #include "callback.h"
 #include "delegation.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
  
@@ -22,7 +23,7 @@ unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres
 	
 	res->bitmap[0] = res->bitmap[1] = 0;
 	res->status = htonl(NFS4ERR_BADHANDLE);
-	clp = nfs4_find_client(&args->addr->sin_addr);
+	clp = nfs_find_client(args->addr, 4);
 	if (clp == NULL)
 		goto out;
 	inode = nfs_delegation_find_inode(clp, &args->fh);
@@ -48,7 +49,7 @@ out_iput:
 	up_read(&nfsi->rwsem);
 	iput(inode);
 out_putclient:
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 out:
 	dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res->status));
 	return res->status;
@@ -61,7 +62,7 @@ unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 	unsigned res;
 	
 	res = htonl(NFS4ERR_BADHANDLE);
-	clp = nfs4_find_client(&args->addr->sin_addr);
+	clp = nfs_find_client(args->addr, 4);
 	if (clp == NULL)
 		goto out;
 	inode = nfs_delegation_find_inode(clp, &args->fh);
@@ -80,7 +81,7 @@ unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 	}
 	iput(inode);
 out_putclient:
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 out:
 	dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res));
 	return res;
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
new file mode 100644
index 000000000000..cb5e92463bdb
--- /dev/null
+++ b/fs/nfs/client.c
@@ -0,0 +1,312 @@
+/* client.c: NFS client sharing and management code
+ *
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <linux/time.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/stat.h>
+#include <linux/errno.h>
+#include <linux/unistd.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/stats.h>
+#include <linux/sunrpc/metrics.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/nfs4_mount.h>
+#include <linux/lockd/bind.h>
+#include <linux/smp_lock.h>
+#include <linux/seq_file.h>
+#include <linux/mount.h>
+#include <linux/nfs_idmap.h>
+#include <linux/vfs.h>
+#include <linux/inet.h>
+#include <linux/nfs_xdr.h>
+
+#include <asm/system.h>
+
+#include "nfs4_fs.h"
+#include "callback.h"
+#include "delegation.h"
+#include "iostat.h"
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_CLIENT
+
+static DEFINE_SPINLOCK(nfs_client_lock);
+static LIST_HEAD(nfs_client_list);
+static DECLARE_WAIT_QUEUE_HEAD(nfs_client_active_wq);
+
+/*
+ * Allocate a shared client record
+ *
+ * Since these are allocated/deallocated very rarely, we don't
+ * bother putting them in a slab cache...
+ */
+static struct nfs_client *nfs_alloc_client(const char *hostname,
+					   const struct sockaddr_in *addr,
+					   int nfsversion)
+{
+	struct nfs_client *clp;
+	int error;
+
+	if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL)
+		goto error_0;
+
+	error = rpciod_up();
+	if (error < 0) {
+		dprintk("%s: couldn't start rpciod! Error = %d\n",
+				__FUNCTION__, error);
+		__set_bit(NFS_CS_RPCIOD, &clp->cl_res_state);
+		goto error_1;
+	}
+
+	if (nfsversion == 4) {
+		if (nfs_callback_up() < 0)
+			goto error_2;
+		__set_bit(NFS_CS_CALLBACK, &clp->cl_res_state);
+	}
+
+	atomic_set(&clp->cl_count, 1);
+	clp->cl_cons_state = NFS_CS_INITING;
+
+	clp->cl_nfsversion = nfsversion;
+	memcpy(&clp->cl_addr, addr, sizeof(clp->cl_addr));
+
+	if (hostname) {
+		clp->cl_hostname = kstrdup(hostname, GFP_KERNEL);
+		if (!clp->cl_hostname)
+			goto error_3;
+	}
+
+	INIT_LIST_HEAD(&clp->cl_superblocks);
+	clp->cl_rpcclient = ERR_PTR(-EINVAL);
+
+#ifdef CONFIG_NFS_V4
+	init_rwsem(&clp->cl_sem);
+	INIT_LIST_HEAD(&clp->cl_delegations);
+	INIT_LIST_HEAD(&clp->cl_state_owners);
+	INIT_LIST_HEAD(&clp->cl_unused);
+	spin_lock_init(&clp->cl_lock);
+	INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp);
+	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client");
+	clp->cl_boot_time = CURRENT_TIME;
+	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
+#endif
+
+	return clp;
+
+error_3:
+	nfs_callback_down();
+	__clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state);
+error_2:
+	rpciod_down();
+	__clear_bit(NFS_CS_RPCIOD, &clp->cl_res_state);
+error_1:
+	kfree(clp);
+error_0:
+	return NULL;
+}
+
+/*
+ * Destroy a shared client record
+ */
+static void nfs_free_client(struct nfs_client *clp)
+{
+	dprintk("--> nfs_free_client(%d)\n", clp->cl_nfsversion);
+
+#ifdef CONFIG_NFS_V4
+	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state)) {
+		while (!list_empty(&clp->cl_unused)) {
+			struct nfs4_state_owner *sp;
+
+			sp = list_entry(clp->cl_unused.next,
+					struct nfs4_state_owner,
+					so_list);
+			list_del(&sp->so_list);
+			kfree(sp);
+		}
+		BUG_ON(!list_empty(&clp->cl_state_owners));
+		nfs_idmap_delete(clp);
+	}
+#endif
+
+	/* -EIO all pending I/O */
+	if (!IS_ERR(clp->cl_rpcclient))
+		rpc_shutdown_client(clp->cl_rpcclient);
+
+	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
+		nfs_callback_down();
+
+	if (__test_and_clear_bit(NFS_CS_RPCIOD, &clp->cl_res_state))
+	rpciod_down();
+
+	kfree(clp->cl_hostname);
+	kfree(clp);
+
+	dprintk("<-- nfs_free_client()\n");
+}
+
+/*
+ * Release a reference to a shared client record
+ */
+void nfs_put_client(struct nfs_client *clp)
+{
+	dprintk("--> nfs_put_client({%d})\n", atomic_read(&clp->cl_count));
+
+	if (atomic_dec_and_lock(&clp->cl_count, &nfs_client_lock)) {
+		list_del(&clp->cl_share_link);
+		spin_unlock(&nfs_client_lock);
+
+		BUG_ON(!list_empty(&clp->cl_superblocks));
+
+		nfs_free_client(clp);
+	}
+}
+
+/*
+ * Find a client by address
+ * - caller must hold nfs_client_lock
+ */
+static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion)
+{
+	struct nfs_client *clp;
+
+	list_for_each_entry(clp, &nfs_client_list, cl_share_link) {
+		/* Different NFS versions cannot share the same nfs_client */
+		if (clp->cl_nfsversion != nfsversion)
+			continue;
+
+		if (memcmp(&clp->cl_addr.sin_addr, &addr->sin_addr,
+			   sizeof(clp->cl_addr.sin_addr)) != 0)
+			continue;
+
+		if (clp->cl_addr.sin_port == addr->sin_port)
+			goto found;
+	}
+
+	return NULL;
+
+found:
+	atomic_inc(&clp->cl_count);
+	return clp;
+}
+
+/*
+ * Find a client by IP address and protocol version
+ * - returns NULL if no such client
+ */
+struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversion)
+{
+	struct nfs_client *clp;
+
+	spin_lock(&nfs_client_lock);
+	clp = __nfs_find_client(addr, nfsversion);
+	spin_unlock(&nfs_client_lock);
+
+	BUG_ON(clp->cl_cons_state == 0);
+
+	return clp;
+}
+
+/*
+ * Look up a client by IP address and protocol version
+ * - creates a new record if one doesn't yet exist
+ */
+struct nfs_client *nfs_get_client(const char *hostname,
+				  const struct sockaddr_in *addr,
+				  int nfsversion)
+{
+	struct nfs_client *clp, *new = NULL;
+	int error;
+
+	dprintk("--> nfs_get_client(%s,"NIPQUAD_FMT":%d,%d)\n",
+		hostname ?: "", NIPQUAD(addr->sin_addr),
+		addr->sin_port, nfsversion);
+
+	/* see if the client already exists */
+	do {
+		spin_lock(&nfs_client_lock);
+
+		clp = __nfs_find_client(addr, nfsversion);
+		if (clp)
+			goto found_client;
+		if (new)
+			goto install_client;
+
+		spin_unlock(&nfs_client_lock);
+
+		new = nfs_alloc_client(hostname, addr, nfsversion);
+	} while (new);
+
+	return ERR_PTR(-ENOMEM);
+
+	/* install a new client and return with it unready */
+install_client:
+	clp = new;
+	list_add(&clp->cl_share_link, &nfs_client_list);
+	spin_unlock(&nfs_client_lock);
+	dprintk("--> nfs_get_client() = %p [new]\n", clp);
+	return clp;
+
+	/* found an existing client
+	 * - make sure it's ready before returning
+	 */
+found_client:
+	spin_unlock(&nfs_client_lock);
+
+	if (new)
+		nfs_free_client(new);
+
+	if (clp->cl_cons_state == NFS_CS_INITING) {
+		DECLARE_WAITQUEUE(myself, current);
+
+		add_wait_queue(&nfs_client_active_wq, &myself);
+
+		for (;;) {
+			set_current_state(TASK_INTERRUPTIBLE);
+			if (signal_pending(current) ||
+			    clp->cl_cons_state > NFS_CS_READY)
+				break;
+			schedule();
+		}
+
+		remove_wait_queue(&nfs_client_active_wq, &myself);
+
+		if (signal_pending(current)) {
+			nfs_put_client(clp);
+			return ERR_PTR(-ERESTARTSYS);
+		}
+	}
+
+	if (clp->cl_cons_state < NFS_CS_READY) {
+		error = clp->cl_cons_state;
+		nfs_put_client(clp);
+		return ERR_PTR(error);
+	}
+
+	dprintk("--> nfs_get_client() = %p [share]\n", clp);
+	return clp;
+}
+
+/*
+ * Mark a server as ready or failed
+ */
+void nfs_mark_client_ready(struct nfs_client *clp, int state)
+{
+	clp->cl_cons_state = state;
+	wake_up_all(&nfs_client_active_wq);
+}
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index cfe239736ac0..57133678db16 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -18,6 +18,7 @@
 
 #include "nfs4_fs.h"
 #include "delegation.h"
+#include "internal.h"
 
 static struct nfs_delegation *nfs_alloc_delegation(void)
 {
@@ -145,7 +146,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct
 					sizeof(delegation->stateid)) != 0 ||
 				delegation->type != nfsi->delegation->type) {
 			printk("%s: server %u.%u.%u.%u, handed out a duplicate delegation!\n",
-					__FUNCTION__, NIPQUAD(clp->cl_addr));
+					__FUNCTION__, NIPQUAD(clp->cl_addr.sin_addr));
 			status = -EIO;
 		}
 	}
@@ -254,7 +255,7 @@ restart:
 	}
 out:
 	spin_unlock(&clp->cl_lock);
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 	module_put_and_exit(0);
 }
 
@@ -266,10 +267,10 @@ void nfs_expire_all_delegations(struct nfs_client *clp)
 	atomic_inc(&clp->cl_count);
 	task = kthread_run(nfs_do_expire_all_delegations, clp,
 			"%u.%u.%u.%u-delegreturn",
-			NIPQUAD(clp->cl_addr));
+			NIPQUAD(clp->cl_addr.sin_addr));
 	if (!IS_ERR(task))
 		return;
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 	module_put(THIS_MODULE);
 }
 
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 4802157963f8..ac370d5d4494 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -15,6 +15,12 @@ struct nfs_clone_mount {
 	rpc_authflavor_t authflavor;
 };
 
+/* client.c */
+extern void nfs_put_client(struct nfs_client *);
+extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int);
+extern struct nfs_client *nfs_get_client(const char *, const struct sockaddr_in *, int);
+extern void nfs_mark_client_ready(struct nfs_client *, int);
+
 /* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
 extern struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry);
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 4e334cb48498..e7879245361e 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -42,55 +42,6 @@ enum nfs4_client_state {
 	NFS4CLNT_LEASE_EXPIRED,
 };
 
-/*
- * The nfs_client identifies our client state to the server.
- */
-struct nfs_client {
-	struct list_head	cl_servers;	/* Global list of servers */
-	struct in_addr		cl_addr;	/* Server identifier */
-	u64			cl_clientid;	/* constant */
-	nfs4_verifier		cl_confirm;
-	unsigned long		cl_state;
-
-	u32			cl_lockowner_id;
-
-	/*
-	 * The following rwsem ensures exclusive access to the server
-	 * while we recover the state following a lease expiration.
-	 */
-	struct rw_semaphore	cl_sem;
-
-	struct list_head	cl_delegations;
-	struct list_head	cl_state_owners;
-	struct list_head	cl_unused;
-	int			cl_nunused;
-	spinlock_t		cl_lock;
-	atomic_t		cl_count;
-
-	struct rpc_clnt *	cl_rpcclient;
-
-	struct list_head	cl_superblocks;	/* List of nfs_server structs */
-
-	unsigned long		cl_lease_time;
-	unsigned long		cl_last_renewal;
-	struct work_struct	cl_renewd;
-	struct work_struct	cl_recoverd;
-
-	struct rpc_wait_queue	cl_rpcwaitq;
-
-	/* used for the setclientid verifier */
-	struct timespec		cl_boot_time;
-
-	/* idmapper */
-	struct idmap *		cl_idmap;
-
-	/* Our own IP address, as a null-terminated string.
-	 * This is used to generate the clientid, and the callback address.
-	 */
-	char			cl_ipaddr[16];
-	unsigned char		cl_id_uniquifier;
-};
-
 /*
  * struct rpc_sequence ensures that RPC calls are sent in the exact
  * order that they appear on the list.
@@ -239,9 +190,6 @@ extern void nfs4_renew_state(void *);
 /* nfs4state.c */
 extern void init_nfsv4_state(struct nfs_server *);
 extern void destroy_nfsv4_state(struct nfs_server *);
-extern struct nfs_client *nfs4_get_client(struct in_addr *);
-extern void nfs4_put_client(struct nfs_client *clp);
-extern struct nfs_client *nfs4_find_client(struct in_addr *);
 struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp);
 extern u32 nfs4_alloc_lockowner_id(struct nfs_client *);
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 850f0851023a..803c31b88bb5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2968,7 +2968,7 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po
 	for(;;) {
 		setclientid.sc_name_len = scnprintf(setclientid.sc_name,
 				sizeof(setclientid.sc_name), "%s/%u.%u.%u.%u %s %u",
-				clp->cl_ipaddr, NIPQUAD(clp->cl_addr.s_addr),
+				clp->cl_ipaddr, NIPQUAD(clp->cl_addr.sin_addr),
 				cred->cr_ops->cr_name,
 				clp->cl_id_uniquifier);
 		setclientid.sc_netid_len = scnprintf(setclientid.sc_netid,
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index fa51a7d4c022..058811e39555 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -50,12 +50,12 @@
 #include "nfs4_fs.h"
 #include "callback.h"
 #include "delegation.h"
+#include "internal.h"
 
 #define OPENOWNER_POOL_SIZE	8
 
 const nfs4_stateid zero_stateid;
 
-static DEFINE_SPINLOCK(state_spinlock);
 static LIST_HEAD(nfs4_clientid_list);
 
 void
@@ -71,127 +71,11 @@ destroy_nfsv4_state(struct nfs_server *server)
 	kfree(server->mnt_path);
 	server->mnt_path = NULL;
 	if (server->nfs_client) {
-		nfs4_put_client(server->nfs_client);
+		nfs_put_client(server->nfs_client);
 		server->nfs_client = NULL;
 	}
 }
 
-/*
- * nfs4_get_client(): returns an empty client structure
- * nfs4_put_client(): drops reference to client structure
- *
- * Since these are allocated/deallocated very rarely, we don't
- * bother putting them in a slab cache...
- */
-static struct nfs_client *
-nfs4_alloc_client(struct in_addr *addr)
-{
-	struct nfs_client *clp;
-
-	if (nfs_callback_up() < 0)
-		return NULL;
-	if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL) {
-		nfs_callback_down();
-		return NULL;
-	}
-	memcpy(&clp->cl_addr, addr, sizeof(clp->cl_addr));
-	init_rwsem(&clp->cl_sem);
-	INIT_LIST_HEAD(&clp->cl_delegations);
-	INIT_LIST_HEAD(&clp->cl_state_owners);
-	INIT_LIST_HEAD(&clp->cl_unused);
-	spin_lock_init(&clp->cl_lock);
-	atomic_set(&clp->cl_count, 1);
-	INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp);
-	INIT_LIST_HEAD(&clp->cl_superblocks);
-	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS4 client");
-	clp->cl_rpcclient = ERR_PTR(-EINVAL);
-	clp->cl_boot_time = CURRENT_TIME;
-	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
-	return clp;
-}
-
-static void
-nfs4_free_client(struct nfs_client *clp)
-{
-	struct nfs4_state_owner *sp;
-
-	while (!list_empty(&clp->cl_unused)) {
-		sp = list_entry(clp->cl_unused.next,
-				struct nfs4_state_owner,
-				so_list);
-		list_del(&sp->so_list);
-		kfree(sp);
-	}
-	BUG_ON(!list_empty(&clp->cl_state_owners));
-	nfs_idmap_delete(clp);
-	if (!IS_ERR(clp->cl_rpcclient))
-		rpc_shutdown_client(clp->cl_rpcclient);
-	kfree(clp);
-	nfs_callback_down();
-}
-
-static struct nfs_client *__nfs4_find_client(struct in_addr *addr)
-{
-	struct nfs_client *clp;
-	list_for_each_entry(clp, &nfs4_clientid_list, cl_servers) {
-		if (memcmp(&clp->cl_addr, addr, sizeof(clp->cl_addr)) == 0) {
-			atomic_inc(&clp->cl_count);
-			return clp;
-		}
-	}
-	return NULL;
-}
-
-struct nfs_client *nfs4_find_client(struct in_addr *addr)
-{
-	struct nfs_client *clp;
-	spin_lock(&state_spinlock);
-	clp = __nfs4_find_client(addr);
-	spin_unlock(&state_spinlock);
-	return clp;
-}
-
-struct nfs_client *
-nfs4_get_client(struct in_addr *addr)
-{
-	struct nfs_client *clp, *new = NULL;
-
-	spin_lock(&state_spinlock);
-	for (;;) {
-		clp = __nfs4_find_client(addr);
-		if (clp != NULL)
-			break;
-		clp = new;
-		if (clp != NULL) {
-			list_add(&clp->cl_servers, &nfs4_clientid_list);
-			new = NULL;
-			break;
-		}
-		spin_unlock(&state_spinlock);
-		new = nfs4_alloc_client(addr);
-		spin_lock(&state_spinlock);
-		if (new == NULL)
-			break;
-	}
-	spin_unlock(&state_spinlock);
-	if (new)
-		nfs4_free_client(new);
-	return clp;
-}
-
-void
-nfs4_put_client(struct nfs_client *clp)
-{
-	if (!atomic_dec_and_lock(&clp->cl_count, &state_spinlock))
-		return;
-	list_del(&clp->cl_servers);
-	spin_unlock(&state_spinlock);
-	BUG_ON(!list_empty(&clp->cl_superblocks));
-	rpc_wake_up(&clp->cl_rpcwaitq);
-	nfs4_kill_renewd(clp);
-	nfs4_free_client(clp);
-}
-
 static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	int status = nfs4_proc_setclientid(clp, NFS4_CALLBACK,
@@ -771,11 +655,11 @@ static void nfs4_recover_state(struct nfs_client *clp)
 	__module_get(THIS_MODULE);
 	atomic_inc(&clp->cl_count);
 	task = kthread_run(reclaimer, clp, "%u.%u.%u.%u-reclaim",
-			NIPQUAD(clp->cl_addr));
+			NIPQUAD(clp->cl_addr.sin_addr));
 	if (!IS_ERR(task))
 		return;
 	nfs4_clear_recover_bit(clp);
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 	module_put(THIS_MODULE);
 }
 
@@ -970,12 +854,12 @@ out:
 	if (status == -NFS4ERR_CB_PATH_DOWN)
 		nfs_handle_cb_pathdown(clp);
 	nfs4_clear_recover_bit(clp);
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 	module_put_and_exit(0);
 	return 0;
 out_error:
 	printk(KERN_WARNING "Error: state recovery failed on NFSv4 server %u.%u.%u.%u with error %d\n",
-				NIPQUAD(clp->cl_addr.s_addr), -status);
+				NIPQUAD(clp->cl_addr.sin_addr), -status);
 	set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 	goto out;
 }
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 3ee85c4e65d8..f97d7d9c5c32 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1104,47 +1104,46 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 	struct rpc_clnt *clnt = NULL;
 	int err = -EIO;
 
-	clp = nfs4_get_client(&server->addr.sin_addr);
+	clp = nfs_get_client(server->hostname, &server->addr, 4);
 	if (!clp) {
 		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
 		return ERR_PTR(err);
 	}
 
 	/* Now create transport and client */
-	down_write(&clp->cl_sem);
-	if (IS_ERR(clp->cl_rpcclient)) {
+	if (clp->cl_cons_state == NFS_CS_INITING) {
 		xprt = xprt_create_proto(proto, &server->addr, timeparms);
 		if (IS_ERR(xprt)) {
-			up_write(&clp->cl_sem);
 			err = PTR_ERR(xprt);
 			dprintk("%s: cannot create RPC transport. Error = %d\n",
 					__FUNCTION__, err);
-			goto out_fail;
+			goto client_init_error;
 		}
 		/* Bind to a reserved port! */
 		xprt->resvport = 1;
 		clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
 				server->rpc_ops->version, flavor);
 		if (IS_ERR(clnt)) {
-			up_write(&clp->cl_sem);
 			err = PTR_ERR(clnt);
 			dprintk("%s: cannot create RPC client. Error = %d\n",
 					__FUNCTION__, err);
-			goto out_fail;
+			goto client_init_error;
 		}
 		clnt->cl_intr     = 1;
 		clnt->cl_softrtry = 1;
 		clp->cl_rpcclient = clnt;
 		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
-		if (nfs_idmap_new(clp) < 0)
-			goto out_fail;
+		err = nfs_idmap_new(clp);
+		if (err < 0) {
+			dprintk("%s: failed to create idmapper.\n",
+				__FUNCTION__);
+			goto client_init_error;
+		}
+		__set_bit(NFS_CS_IDMAP, &clp->cl_res_state);
+		nfs_mark_client_ready(clp, 0);
 	}
-	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
+
 	clnt = rpc_clone_client(clp->cl_rpcclient);
-	if (!IS_ERR(clnt))
-		server->nfs_client = clp;
-	up_write(&clp->cl_sem);
-	clp = NULL;
 
 	if (IS_ERR(clnt)) {
 		dprintk("%s: cannot create RPC client. Error = %d\n",
@@ -1152,11 +1151,6 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 		return clnt;
 	}
 
-	if (server->nfs_client->cl_idmap == NULL) {
-		dprintk("%s: failed to create idmapper.\n", __FUNCTION__);
-		return ERR_PTR(-ENOMEM);
-	}
-
 	if (clnt->cl_auth->au_flavor != flavor) {
 		struct rpc_auth *auth;
 
@@ -1166,11 +1160,16 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 			return (struct rpc_clnt *)auth;
 		}
 	}
+
+	server->nfs_client = clp;
+	down_write(&clp->cl_sem);
+	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
+	up_write(&clp->cl_sem);
 	return clnt;
 
- out_fail:
-	if (clp)
-		nfs4_put_client(clp);
+client_init_error:
+	nfs_mark_client_ready(clp, err);
+	nfs_put_client(clp);
 	return ERR_PTR(err);
 }
 
@@ -1329,14 +1328,6 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 		goto out_free;
 	}
 
-	/* Fire up rpciod if not yet running */
-	error = rpciod_up();
-	if (error < 0) {
-		dprintk("%s: couldn't start rpciod! Error = %d\n",
-				__FUNCTION__, error);
-		goto out_free;
-	}
-
 	s = sget(fs_type, nfs4_compare_super, nfs_set_super, server);
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
@@ -1383,8 +1374,6 @@ static void nfs4_kill_super(struct super_block *sb)
 
 	destroy_nfsv4_state(server);
 
-	rpciod_down();
-
 	nfs_free_iostats(server->io_stats);
 	kfree(server->hostname);
 	kfree(server);
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index a36e01cd6321..70e1dc9162e2 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -586,6 +586,7 @@ extern void * nfs_root_data(void);
 #define NFSDBG_FILE		0x0040
 #define NFSDBG_ROOT		0x0080
 #define NFSDBG_CALLBACK		0x0100
+#define NFSDBG_CLIENT		0x0200
 #define NFSDBG_ALL		0xFFFF
 
 #ifdef __KERNEL__
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index fc20d6b934fb..a727657e0ad3 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -6,6 +6,66 @@
 
 struct nfs_iostats;
 
+/*
+ * The nfs_client identifies our client state to the server.
+ */
+struct nfs_client {
+	atomic_t		cl_count;
+	int			cl_cons_state;	/* current construction state (-ve: init error) */
+#define NFS_CS_READY		0		/* ready to be used */
+#define NFS_CS_INITING		1		/* busy initialising */
+	int			cl_nfsversion;	/* NFS protocol version */
+	unsigned long		cl_res_state;	/* NFS resources state */
+#define NFS_CS_RPCIOD		0		/* - rpciod started */
+#define NFS_CS_CALLBACK		1		/* - callback started */
+#define NFS_CS_IDMAP		2		/* - idmap started */
+	struct sockaddr_in	cl_addr;	/* server identifier */
+	char *			cl_hostname;	/* hostname of server */
+	struct list_head	cl_share_link;	/* link in global client list */
+	struct list_head	cl_superblocks;	/* List of nfs_server structs */
+
+	struct rpc_clnt *	cl_rpcclient;
+
+#ifdef CONFIG_NFS_V4
+	u64			cl_clientid;	/* constant */
+	nfs4_verifier		cl_confirm;
+	unsigned long		cl_state;
+
+	u32			cl_lockowner_id;
+
+	/*
+	 * The following rwsem ensures exclusive access to the server
+	 * while we recover the state following a lease expiration.
+	 */
+	struct rw_semaphore	cl_sem;
+
+	struct list_head	cl_delegations;
+	struct list_head	cl_state_owners;
+	struct list_head	cl_unused;
+	int			cl_nunused;
+	spinlock_t		cl_lock;
+
+	unsigned long		cl_lease_time;
+	unsigned long		cl_last_renewal;
+	struct work_struct	cl_renewd;
+	struct work_struct	cl_recoverd;
+
+	struct rpc_wait_queue	cl_rpcwaitq;
+
+	/* used for the setclientid verifier */
+	struct timespec		cl_boot_time;
+
+	/* idmapper */
+	struct idmap *		cl_idmap;
+
+	/* Our own IP address, as a null-terminated string.
+	 * This is used to generate the clientid, and the callback address.
+	 */
+	char			cl_ipaddr[16];
+	unsigned char		cl_id_uniquifier;
+#endif
+};
+
 /*
  * NFS client parameters stored in the superblock.
  */
-- 
cgit v1.2.3


From 509de8111656a7d89b4a1a5f430f4460ce510f0f Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:11 -0400
Subject: NFS: Add extra const qualifiers

Add some extra const qualifiers into NFS.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/namespace.c        |  3 ++-
 fs/nfs/nfs3proc.c         |  2 +-
 fs/nfs/nfs4namespace.c    |  8 ++++----
 fs/nfs/nfs4proc.c         |  2 +-
 fs/nfs/proc.c             |  2 +-
 fs/nfs/super.c            | 10 +++++-----
 include/linux/nfs_fs_sb.h |  2 +-
 include/linux/nfs_xdr.h   |  6 +++---
 8 files changed, 18 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 86b3169c8cac..85d9ed1dcf42 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -172,7 +172,8 @@ void nfs_release_automount_timer(void)
 /*
  * Clone a mountpoint of the appropriate type
  */
-static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server, char *devname,
+static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
+					   const char *devname,
 					   struct nfs_clone_mount *mountdata)
 {
 #ifdef CONFIG_NFS_V4
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 7143b1f82cea..3e5371241cea 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -886,7 +886,7 @@ nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
 	return nlmclnt_proc(filp->f_dentry->d_inode, cmd, fl);
 }
 
-struct nfs_rpc_ops	nfs_v3_clientops = {
+const struct nfs_rpc_ops nfs_v3_clientops = {
 	.version	= 3,			/* protocol version */
 	.dentry_ops	= &nfs_dentry_operations,
 	.dir_inode_ops	= &nfs3_dir_inode_operations,
diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c
index ea38d27b74e6..faed9bcba50f 100644
--- a/fs/nfs/nfs4namespace.c
+++ b/fs/nfs/nfs4namespace.c
@@ -23,7 +23,7 @@
 /*
  * Check if fs_root is valid
  */
-static inline char *nfs4_pathname_string(struct nfs4_pathname *pathname,
+static inline char *nfs4_pathname_string(const struct nfs4_pathname *pathname,
 					 char *buffer, ssize_t buflen)
 {
 	char *end = buffer + buflen;
@@ -34,7 +34,7 @@ static inline char *nfs4_pathname_string(struct nfs4_pathname *pathname,
 
 	n = pathname->ncomponents;
 	while (--n >= 0) {
-		struct nfs4_string *component = &pathname->components[n];
+		const struct nfs4_string *component = &pathname->components[n];
 		buflen -= component->len + 1;
 		if (buflen < 0)
 			goto Elong;
@@ -60,7 +60,7 @@ Elong:
  */
 static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
 					    const struct dentry *dentry,
-					    struct nfs4_fs_locations *locations)
+					    const struct nfs4_fs_locations *locations)
 {
 	struct vfsmount *mnt = ERR_PTR(-ENOENT);
 	struct nfs_clone_mount mountdata = {
@@ -108,7 +108,7 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
 
 	loc = 0;
 	while (loc < locations->nlocations && IS_ERR(mnt)) {
-		struct nfs4_fs_location *location = &locations->locations[loc];
+		const struct nfs4_fs_location *location = &locations->locations[loc];
 		char *mnt_path;
 
 		if (location == NULL || location->nservers <= 0 ||
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 803c31b88bb5..061be713b206 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3761,7 +3761,7 @@ static struct inode_operations nfs4_file_inode_operations = {
 	.listxattr	= nfs4_listxattr,
 };
 
-struct nfs_rpc_ops	nfs_v4_clientops = {
+const struct nfs_rpc_ops nfs_v4_clientops = {
 	.version	= 4,			/* protocol version */
 	.dentry_ops	= &nfs4_dentry_operations,
 	.dir_inode_ops	= &nfs4_dir_inode_operations,
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index b3899ea3229e..77676903e0f5 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -671,7 +671,7 @@ nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
 }
 
 
-struct nfs_rpc_ops	nfs_v2_clientops = {
+const struct nfs_rpc_ops nfs_v2_clientops = {
 	.version	= 2,		       /* protocol version */
 	.dentry_ops	= &nfs_dentry_operations,
 	.dir_inode_ops	= &nfs_dir_inode_operations,
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index a41d516ed595..c97f30967955 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -329,10 +329,10 @@ static const char *nfs_pseudoflavour_to_name(rpc_authflavor_t flavour)
  */
 static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, int showdefaults)
 {
-	static struct proc_nfs_info {
+	static const struct proc_nfs_info {
 		int flag;
-		char *str;
-		char *nostr;
+		const char *str;
+		const char *nostr;
 	} nfs_info[] = {
 		{ NFS_MOUNT_SOFT, ",soft", ",hard" },
 		{ NFS_MOUNT_INTR, ",intr", "" },
@@ -342,9 +342,9 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 		{ NFS_MOUNT_NOACL, ",noacl", "" },
 		{ 0, NULL, NULL }
 	};
-	struct proc_nfs_info *nfs_infop;
+	const struct proc_nfs_info *nfs_infop;
 	char buf[12];
-	char *proto;
+	const char *proto;
 
 	seq_printf(m, ",vers=%d", nfss->rpc_ops->version);
 	seq_printf(m, ",rsize=%d", nfss->rsize);
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index a727657e0ad3..95f32d5f6e9c 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -73,7 +73,7 @@ struct nfs_server {
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_sys;	/* 2nd handle for FSINFO */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
-	struct nfs_rpc_ops *	rpc_ops;	/* NFS protocol vector */
+	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
 	int			flags;		/* various flags */
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index dd9ae6761f71..2426b11b6cce 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -833,9 +833,9 @@ struct nfs_rpc_ops {
 /*
  * Function vectors etc. for the NFS client
  */
-extern struct nfs_rpc_ops	nfs_v2_clientops;
-extern struct nfs_rpc_ops	nfs_v3_clientops;
-extern struct nfs_rpc_ops	nfs_v4_clientops;
+extern const struct nfs_rpc_ops	nfs_v2_clientops;
+extern const struct nfs_rpc_ops	nfs_v3_clientops;
+extern const struct nfs_rpc_ops	nfs_v4_clientops;
 extern struct rpc_version	nfs_version2;
 extern struct rpc_version	nfs_version3;
 extern struct rpc_version	nfs_version4;
-- 
cgit v1.2.3


From 27951bd26031f6c27d38df9e94623bbe208a2464 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:11 -0400
Subject: NFS: Maintain a common server record for NFS2/3 as well as for NFS4

Maintain a common server record for NFS2/3 as well as for NFS4 so that common
stuff can be moved there from struct nfs_server.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/super.c            | 21 ++++++++++++++++++++-
 include/linux/nfs_fs_sb.h |  2 +-
 2 files changed, 21 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index c97f30967955..d1b4a5b36e33 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -658,11 +658,19 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, unsigned
 static struct rpc_clnt *
 nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 {
+	struct nfs_client	*clp;
 	struct rpc_timeout	timeparms;
 	struct rpc_xprt		*xprt = NULL;
 	struct rpc_clnt		*clnt = NULL;
 	int			proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
 
+	clp = nfs_get_client(server->hostname, &server->addr,
+			     server->rpc_ops->version);
+	if (!clp) {
+		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
+		return ERR_PTR(PTR_ERR(clp));
+	}
+
 	nfs_init_timeout_values(&timeparms, proto, data->timeo, data->retrans);
 
 	server->retrans_timeo = timeparms.to_initval;
@@ -673,6 +681,8 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 	if (IS_ERR(xprt)) {
 		dprintk("%s: cannot create RPC transport. Error = %ld\n",
 				__FUNCTION__, PTR_ERR(xprt));
+		nfs_mark_client_ready(clp, PTR_ERR(xprt));
+		nfs_put_client(clp);
 		return (struct rpc_clnt *)xprt;
 	}
 	clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
@@ -686,9 +696,13 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 	clnt->cl_intr     = 1;
 	clnt->cl_softrtry = 1;
 
+	nfs_mark_client_ready(clp, 0);
+	server->nfs_client = clp;
 	return clnt;
 
 out_fail:
+	nfs_mark_client_ready(clp, PTR_ERR(xprt));
+	nfs_put_client(clp);
 	return clnt;
 }
 
@@ -764,6 +778,7 @@ static int nfs_clone_generic_sb(struct nfs_clone_mount *data,
 	if (server == NULL)
 		goto out_err;
 	memcpy(server, parent, sizeof(*server));
+	atomic_inc(&server->nfs_client->cl_count);
 	hostname = (data->hostname != NULL) ? data->hostname : parent->hostname;
 	len = strlen(hostname) + 1;
 	server->hostname = kmalloc(len, GFP_KERNEL);
@@ -796,6 +811,7 @@ out_deactivate:
 out_rpciod_down:
 	rpciod_down();
 	kfree(server->hostname);
+	nfs_put_client(server->nfs_client);
 	kfree(server);
 	return simple_set_mnt(mnt, sb);
 kill_rpciod:
@@ -803,6 +819,7 @@ kill_rpciod:
 free_hostname:
 	kfree(server->hostname);
 free_server:
+	nfs_put_client(server->nfs_client);
 	kfree(server);
 out_err:
 	return error;
@@ -1071,6 +1088,7 @@ static void nfs_kill_super(struct super_block *s)
 
 	nfs_free_iostats(server->io_stats);
 	kfree(server->hostname);
+	nfs_put_client(server->nfs_client);
 	kfree(server);
 	nfs_release_automount_timer();
 }
@@ -1421,7 +1439,6 @@ static struct super_block *nfs4_clone_sb(struct nfs_server *server, struct nfs_c
 	nfs4_server_capabilities(server, &server->fh);
 
 	down_write(&clp->cl_sem);
-	atomic_inc(&clp->cl_count);
 	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
 	up_write(&clp->cl_sem);
 	return sb;
@@ -1476,6 +1493,8 @@ static struct nfs_server *nfs4_referral_server(struct super_block *sb, struct nf
 	retrans = 1;
 	nfs_init_timeout_values(&timeparms, proto, timeo, retrans);
 
+	nfs_put_client(server->nfs_client);
+	server->nfs_client = NULL;
 	server->client = nfs4_create_client(server, &timeparms, proto, data->authflavor);
 	if (IS_ERR((err = server->client)))
 		goto out_err;
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 95f32d5f6e9c..e7d7662f51fd 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -70,6 +70,7 @@ struct nfs_client {
  * NFS client parameters stored in the superblock.
  */
 struct nfs_server {
+	struct nfs_client *	nfs_client;	/* shared client and NFS4 state */
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_sys;	/* 2nd handle for FSINFO */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
@@ -103,7 +104,6 @@ struct nfs_server {
 	 */
 	char			ip_addr[16];
 	char *			mnt_path;
-	struct nfs_client *	nfs_client;	/* all NFSv4 state starts here */
 	struct list_head	nfs4_siblings;	/* List of other nfs_server structs
 						 * that share the same clientid
 						 */
-- 
cgit v1.2.3


From 8fa5c000d7f986ef9cdc6d95f9f7fcee20e0a7d6 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:12 -0400
Subject: NFS: Move rpc_ops from nfs_server to nfs_client

Move the rpc_ops from the nfs_server struct to the nfs_client struct as they're
common to all server records of a particular NFS protocol version.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c              |  2 +-
 fs/nfs/inode.c            |  4 ++--
 fs/nfs/namespace.c        |  6 +++--
 fs/nfs/nfs4proc.c         |  2 +-
 fs/nfs/super.c            | 59 ++++++++++++++++++++++++++---------------------
 include/linux/nfs_fs.h    |  2 +-
 include/linux/nfs_fs_sb.h |  2 +-
 7 files changed, 43 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 067d144d141b..19362712452f 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1147,7 +1147,7 @@ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle,
 	}
 	if (!(fattr->valid & NFS_ATTR_FATTR)) {
 		struct nfs_server *server = NFS_SB(dentry->d_sb);
-		error = server->rpc_ops->getattr(server, fhandle, fattr);
+		error = server->nfs_client->rpc_ops->getattr(server, fhandle, fattr);
 		if (error < 0)
 			goto out_err;
 	}
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 6ed018c9aad2..771c3b833757 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -237,13 +237,13 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 		/* Why so? Because we want revalidate for devices/FIFOs, and
 		 * that's precisely what we have in nfs_file_inode_operations.
 		 */
-		inode->i_op = NFS_SB(sb)->rpc_ops->file_inode_ops;
+		inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->file_inode_ops;
 		if (S_ISREG(inode->i_mode)) {
 			inode->i_fop = &nfs_file_operations;
 			inode->i_data.a_ops = &nfs_file_aops;
 			inode->i_data.backing_dev_info = &NFS_SB(sb)->backing_dev_info;
 		} else if (S_ISDIR(inode->i_mode)) {
-			inode->i_op = NFS_SB(sb)->rpc_ops->dir_inode_ops;
+			inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops;
 			inode->i_fop = &nfs_dir_operations;
 			if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS)
 			    && fattr->size <= NFS_LIMIT_READDIRPLUS)
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 85d9ed1dcf42..d8b8d56266cb 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -104,7 +104,9 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 		goto out_follow;
 	/* Look it up again */
 	parent = dget_parent(nd->dentry);
-	err = server->rpc_ops->lookup(parent->d_inode, &nd->dentry->d_name, &fh, &fattr);
+	err = server->nfs_client->rpc_ops->lookup(parent->d_inode,
+						  &nd->dentry->d_name,
+						  &fh, &fattr);
 	dput(parent);
 	if (err != 0)
 		goto out_err;
@@ -178,7 +180,7 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
 {
 #ifdef CONFIG_NFS_V4
 	struct vfsmount *mnt = NULL;
-	switch (server->rpc_ops->version) {
+	switch (server->nfs_client->cl_nfsversion) {
 		case 2:
 		case 3:
 			mnt = vfs_kern_mount(&clone_nfs_fs_type, 0, devname, mountdata);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b731b1945270..1573eeb07ce1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -758,7 +758,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 	}
 	nfs_confirm_seqid(&data->owner->so_seqid, 0);
 	if (!(o_res->f_attr->valid & NFS_ATTR_FATTR))
-		return server->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr);
+		return server->nfs_client->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr);
 	return 0;
 }
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index d1b4a5b36e33..e1e5eab0259b 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -252,7 +252,7 @@ static int nfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 
 	lock_kernel();
 
-	error = server->rpc_ops->statfs(server, fh, &res);
+	error = server->nfs_client->rpc_ops->statfs(server, fh, &res);
 	buf->f_type = NFS_SUPER_MAGIC;
 	if (error < 0)
 		goto out_err;
@@ -343,10 +343,11 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 		{ 0, NULL, NULL }
 	};
 	const struct proc_nfs_info *nfs_infop;
+	struct nfs_client *clp = nfss->nfs_client;
 	char buf[12];
 	const char *proto;
 
-	seq_printf(m, ",vers=%d", nfss->rpc_ops->version);
+	seq_printf(m, ",vers=%d", clp->rpc_ops->version);
 	seq_printf(m, ",rsize=%d", nfss->rsize);
 	seq_printf(m, ",wsize=%d", nfss->wsize);
 	if (nfss->acregmin != 3*HZ || showdefaults)
@@ -427,7 +428,7 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt)
 	seq_printf(m, ",namelen=%d", nfss->namelen);
 
 #ifdef CONFIG_NFS_V4
-	if (nfss->rpc_ops->version == 4) {
+	if (nfss->nfs_client->cl_nfsversion == 4) {
 		seq_printf(m, "\n\tnfsv4:\t");
 		seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]);
 		seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]);
@@ -503,7 +504,7 @@ nfs_get_root(struct super_block *sb, struct nfs_fh *rootfh, struct nfs_fsinfo *f
 	struct nfs_server	*server = NFS_SB(sb);
 	int			error;
 
-	error = server->rpc_ops->getroot(server, rootfh, fsinfo);
+	error = server->nfs_client->rpc_ops->getroot(server, rootfh, fsinfo);
 	if (error < 0) {
 		dprintk("nfs_get_root: getattr error = %d\n", -error);
 		return ERR_PTR(error);
@@ -553,14 +554,14 @@ nfs_sb_init(struct super_block *sb, rpc_authflavor_t authflavor)
 		no_root_error = -ENOMEM;
 		goto out_no_root;
 	}
-	sb->s_root->d_op = server->rpc_ops->dentry_ops;
+	sb->s_root->d_op = server->nfs_client->rpc_ops->dentry_ops;
 
 	/* mount time stamp, in seconds */
 	server->mount_time = jiffies;
 
 	/* Get some general file system info */
 	if (server->namelen == 0 &&
-	    server->rpc_ops->pathconf(server, &server->fh, &pathinfo) >= 0)
+	    server->nfs_client->rpc_ops->pathconf(server, &server->fh, &pathinfo) >= 0)
 		server->namelen = pathinfo.max_namelen;
 	/* Work out a lot of parameters */
 	if (server->rsize == 0)
@@ -663,9 +664,14 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 	struct rpc_xprt		*xprt = NULL;
 	struct rpc_clnt		*clnt = NULL;
 	int			proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
+	int			nfsversion = 2;
 
-	clp = nfs_get_client(server->hostname, &server->addr,
-			     server->rpc_ops->version);
+#ifdef CONFIG_NFS_V3
+	if (server->flags & NFS_MOUNT_VER3)
+		nfsversion = 3;
+#endif
+
+	clp = nfs_get_client(server->hostname, &server->addr, nfsversion);
 	if (!clp) {
 		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
 		return ERR_PTR(PTR_ERR(clp));
@@ -676,6 +682,19 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 	server->retrans_timeo = timeparms.to_initval;
 	server->retrans_count = timeparms.to_retries;
 
+	/* Check NFS protocol revision and initialize RPC op vector
+	 * and file handle pool. */
+#ifdef CONFIG_NFS_V3
+	if (nfsversion == 3) {
+		clp->rpc_ops = &nfs_v3_clientops;
+		server->caps |= NFS_CAP_READDIRPLUS;
+	} else {
+		clp->rpc_ops = &nfs_v2_clientops;
+	}
+#else
+	clp->rpc_ops = &nfs_v2_clientops;
+#endif
+
 	/* create transport and client */
 	xprt = xprt_create_proto(proto, &server->addr, &timeparms);
 	if (IS_ERR(xprt)) {
@@ -686,7 +705,7 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 		return (struct rpc_clnt *)xprt;
 	}
 	clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
-				 server->rpc_ops->version, data->pseudoflavor);
+				 clp->cl_nfsversion, data->pseudoflavor);
 	if (IS_ERR(clnt)) {
 		dprintk("%s: cannot create RPC client. Error = %ld\n",
 				__FUNCTION__, PTR_ERR(xprt));
@@ -750,7 +769,7 @@ static struct nfs_server *nfs_clone_server(struct super_block *sb, struct nfs_cl
 	fsinfo.fattr = data->fattr;
 	if (NFS_PROTO(root_inode)->fsinfo(server, data->fh, &fsinfo) == 0)
 		nfs_super_set_maxbytes(sb, fsinfo.maxfilesize);
-	sb->s_root->d_op = server->rpc_ops->dentry_ops;
+	sb->s_root->d_op = server->nfs_client->rpc_ops->dentry_ops;
 	sb->s_flags |= MS_ACTIVE;
 	return server;
 out_put_root:
@@ -865,19 +884,6 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
 		return -ENOMEM;
 	strcpy(server->hostname, data->hostname);
 
-	/* Check NFS protocol revision and initialize RPC op vector
-	 * and file handle pool. */
-#ifdef CONFIG_NFS_V3
-	if (server->flags & NFS_MOUNT_VER3) {
-		server->rpc_ops = &nfs_v3_clientops;
-		server->caps |= NFS_CAP_READDIRPLUS;
-	} else {
-		server->rpc_ops = &nfs_v2_clientops;
-	}
-#else
-	server->rpc_ops = &nfs_v2_clientops;
-#endif
-
 	/* Fill in pseudoflavor for mount version < 5 */
 	if (!(data->flags & NFS_MOUNT_SECFLAVOUR))
 		data->pseudoflavor = RPC_AUTH_UNIX;
@@ -888,6 +894,7 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
 	server->client = nfs_create_client(server, data);
 	if (IS_ERR(server->client))
 		return PTR_ERR(server->client);
+
 	/* RFC 2623, sec 2.3.2 */
 	if (authflavor != RPC_AUTH_UNIX) {
 		struct rpc_auth *auth;
@@ -1129,6 +1136,8 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 
 	/* Now create transport and client */
 	if (clp->cl_cons_state == NFS_CS_INITING) {
+		clp->rpc_ops = &nfs_v4_clientops;
+
 		xprt = xprt_create_proto(proto, &server->addr, timeparms);
 		if (IS_ERR(xprt)) {
 			err = PTR_ERR(xprt);
@@ -1139,7 +1148,7 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 		/* Bind to a reserved port! */
 		xprt->resvport = 1;
 		clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
-				server->rpc_ops->version, flavor);
+				clp->cl_nfsversion, flavor);
 		if (IS_ERR(clnt)) {
 			err = PTR_ERR(clnt);
 			dprintk("%s: cannot create RPC client. Error = %d\n",
@@ -1215,8 +1224,6 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data,
 	server->acdirmin = data->acdirmin*HZ;
 	server->acdirmax = data->acdirmax*HZ;
 
-	server->rpc_ops = &nfs_v4_clientops;
-
 	nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans);
 
 	server->retrans_timeo = timeparms.to_initval;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 70e1dc9162e2..51e9bd90dedc 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -215,7 +215,7 @@ static inline struct nfs_inode *NFS_I(struct inode *inode)
 #define NFS_FH(inode)			(&NFS_I(inode)->fh)
 #define NFS_SERVER(inode)		(NFS_SB(inode->i_sb))
 #define NFS_CLIENT(inode)		(NFS_SERVER(inode)->client)
-#define NFS_PROTO(inode)		(NFS_SERVER(inode)->rpc_ops)
+#define NFS_PROTO(inode)		(NFS_SERVER(inode)->nfs_client->rpc_ops)
 #define NFS_ADDR(inode)			(RPC_PEERADDR(NFS_CLIENT(inode)))
 #define NFS_COOKIEVERF(inode)		(NFS_I(inode)->cookieverf)
 #define NFS_READTIME(inode)		(NFS_I(inode)->read_cache_jiffies)
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index e7d7662f51fd..aae7c117597a 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -25,6 +25,7 @@ struct nfs_client {
 	struct list_head	cl_superblocks;	/* List of nfs_server structs */
 
 	struct rpc_clnt *	cl_rpcclient;
+	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
 
 #ifdef CONFIG_NFS_V4
 	u64			cl_clientid;	/* constant */
@@ -74,7 +75,6 @@ struct nfs_server {
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_sys;	/* 2nd handle for FSINFO */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
-	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
 	int			flags;		/* various flags */
-- 
cgit v1.2.3


From 5006a76cca8f86c6975c16fcf67e83b8b0eee2b6 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:12 -0400
Subject: NFS: Eliminate client_sys in favour of cl_rpcclient

Eliminate nfs_server::client_sys in favour of nfs_client::cl_rpcclient as we
only really need one per server that we're talking to since it doesn't have any
security on it.

The retransmission management variables are also moved to the common struct as
they're required to set up the cl_rpcclient connection.

The NFS2/3 client and client_acl connections are thenceforth derived by cloning
the cl_rpcclient connection and post-applying the authorisation flavour.

The code for setting up the initial common connection has been moved to
client.c as nfs_create_rpc_client().  All the NFS program definition tables are
also moved there as that's where they're now required rather than super.c.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 119 +++++++++++++++++++++++++
 fs/nfs/internal.h         |   2 +
 fs/nfs/nfs3proc.c         |   6 +-
 fs/nfs/proc.c             |   4 +-
 fs/nfs/super.c            | 222 +++++++++++-----------------------------------
 include/linux/nfs_fs_sb.h |   5 +-
 6 files changed, 179 insertions(+), 179 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index cb5e92463bdb..c08cab935ad5 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -50,6 +50,48 @@ static DEFINE_SPINLOCK(nfs_client_lock);
 static LIST_HEAD(nfs_client_list);
 static DECLARE_WAIT_QUEUE_HEAD(nfs_client_active_wq);
 
+/*
+ * RPC cruft for NFS
+ */
+static struct rpc_version *nfs_version[5] = {
+	[2]			= &nfs_version2,
+#ifdef CONFIG_NFS_V3
+	[3]			= &nfs_version3,
+#endif
+#ifdef CONFIG_NFS_V4
+	[4]			= &nfs_version4,
+#endif
+};
+
+struct rpc_program nfs_program = {
+	.name			= "nfs",
+	.number			= NFS_PROGRAM,
+	.nrvers			= ARRAY_SIZE(nfs_version),
+	.version		= nfs_version,
+	.stats			= &nfs_rpcstat,
+	.pipe_dir_name		= "/nfs",
+};
+
+struct rpc_stat nfs_rpcstat = {
+	.program		= &nfs_program
+};
+
+
+#ifdef CONFIG_NFS_V3_ACL
+static struct rpc_stat		nfsacl_rpcstat = { &nfsacl_program };
+static struct rpc_version *	nfsacl_version[] = {
+	[3]			= &nfsacl_version3,
+};
+
+struct rpc_program		nfsacl_program = {
+	.name			= "nfsacl",
+	.number			= NFS_ACL_PROGRAM,
+	.nrvers			= ARRAY_SIZE(nfsacl_version),
+	.version		= nfsacl_version,
+	.stats			= &nfsacl_rpcstat,
+};
+#endif  /* CONFIG_NFS_V3_ACL */
+
 /*
  * Allocate a shared client record
  *
@@ -310,3 +352,80 @@ void nfs_mark_client_ready(struct nfs_client *clp, int state)
 	clp->cl_cons_state = state;
 	wake_up_all(&nfs_client_active_wq);
 }
+
+/*
+ * Initialise the timeout values for a connection
+ */
+static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
+				    unsigned int timeo, unsigned int retrans)
+{
+	to->to_initval = timeo * HZ / 10;
+	to->to_retries = retrans;
+	if (!to->to_retries)
+		to->to_retries = 2;
+
+	switch (proto) {
+	case IPPROTO_TCP:
+		if (!to->to_initval)
+			to->to_initval = 60 * HZ;
+		if (to->to_initval > NFS_MAX_TCP_TIMEOUT)
+			to->to_initval = NFS_MAX_TCP_TIMEOUT;
+		to->to_increment = to->to_initval;
+		to->to_maxval = to->to_initval + (to->to_increment * to->to_retries);
+		to->to_exponential = 0;
+		break;
+	case IPPROTO_UDP:
+	default:
+		if (!to->to_initval)
+			to->to_initval = 11 * HZ / 10;
+		if (to->to_initval > NFS_MAX_UDP_TIMEOUT)
+			to->to_initval = NFS_MAX_UDP_TIMEOUT;
+		to->to_maxval = NFS_MAX_UDP_TIMEOUT;
+		to->to_exponential = 1;
+		break;
+	}
+}
+
+/*
+ * Create an RPC client handle
+ */
+int nfs_create_rpc_client(struct nfs_client *clp, int proto,
+			  unsigned int timeo,
+			  unsigned int retrans,
+			  rpc_authflavor_t flavor)
+{
+	struct rpc_timeout	timeparms;
+	struct rpc_xprt		*xprt = NULL;
+	struct rpc_clnt		*clnt = NULL;
+
+	if (!IS_ERR(clp->cl_rpcclient))
+		return 0;
+
+	nfs_init_timeout_values(&timeparms, proto, timeo, retrans);
+	clp->retrans_timeo = timeparms.to_initval;
+	clp->retrans_count = timeparms.to_retries;
+
+	/* create transport and client */
+	xprt = xprt_create_proto(proto, &clp->cl_addr, &timeparms);
+	if (IS_ERR(xprt)) {
+		dprintk("%s: cannot create RPC transport. Error = %ld\n",
+				__FUNCTION__, PTR_ERR(xprt));
+		return PTR_ERR(xprt);
+	}
+
+	/* Bind to a reserved port! */
+	xprt->resvport = 1;
+	/* Create the client RPC handle */
+	clnt = rpc_create_client(xprt, clp->cl_hostname, &nfs_program,
+				 clp->rpc_ops->version, RPC_AUTH_UNIX);
+	if (IS_ERR(clnt)) {
+		dprintk("%s: cannot create RPC client. Error = %ld\n",
+				__FUNCTION__, PTR_ERR(clnt));
+		return PTR_ERR(clnt);
+	}
+
+	clnt->cl_intr     = 1;
+	clnt->cl_softrtry = 1;
+	clp->cl_rpcclient = clnt;
+	return 0;
+}
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index ac370d5d4494..2f3aa52fbefc 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -20,6 +20,8 @@ extern void nfs_put_client(struct nfs_client *);
 extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int);
 extern struct nfs_client *nfs_get_client(const char *, const struct sockaddr_in *, int);
 extern void nfs_mark_client_ready(struct nfs_client *, int);
+extern int nfs_create_rpc_client(struct nfs_client *, int, unsigned int,
+				 unsigned int, rpc_authflavor_t);
 
 /* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 3e5371241cea..0622af0122be 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -90,8 +90,8 @@ nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
 	int	status;
 
 	status = do_proc_get_root(server->client, fhandle, info);
-	if (status && server->client_sys != server->client)
-		status = do_proc_get_root(server->client_sys, fhandle, info);
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = do_proc_get_root(server->nfs_client->cl_rpcclient, fhandle, info);
 	return status;
 }
 
@@ -785,7 +785,7 @@ nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
 
 	dprintk("NFS call  fsinfo\n");
 	nfs_fattr_init(info->fattr);
-	status = rpc_call_sync(server->client_sys, &msg, 0);
+	status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
 	dprintk("NFS reply fsinfo: %d\n", status);
 	return status;
 }
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 77676903e0f5..5a8b9407ee9a 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -66,14 +66,14 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
 
 	dprintk("%s: call getattr\n", __FUNCTION__);
 	nfs_fattr_init(fattr);
-	status = rpc_call_sync(server->client_sys, &msg, 0);
+	status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
 	dprintk("%s: reply getattr: %d\n", __FUNCTION__, status);
 	if (status)
 		return status;
 	dprintk("%s: call statfs\n", __FUNCTION__);
 	msg.rpc_proc = &nfs_procedures[NFSPROC_STATFS];
 	msg.rpc_resp = &fsinfo;
-	status = rpc_call_sync(server->client_sys, &msg, 0);
+	status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
 	dprintk("%s: reply statfs: %d\n", __FUNCTION__, status);
 	if (status)
 		return status;
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index e1e5eab0259b..85583414a3ca 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -60,52 +60,6 @@
  */
 #define NFS_MAX_READAHEAD	(RPC_DEF_SLOT_TABLE - 1)
 
-/*
- * RPC cruft for NFS
- */
-static struct rpc_version * nfs_version[] = {
-	NULL,
-	NULL,
-	&nfs_version2,
-#if defined(CONFIG_NFS_V3)
-	&nfs_version3,
-#elif defined(CONFIG_NFS_V4)
-	NULL,
-#endif
-#if defined(CONFIG_NFS_V4)
-	&nfs_version4,
-#endif
-};
-
-static struct rpc_program nfs_program = {
-	.name			= "nfs",
-	.number			= NFS_PROGRAM,
-	.nrvers			= ARRAY_SIZE(nfs_version),
-	.version		= nfs_version,
-	.stats			= &nfs_rpcstat,
-	.pipe_dir_name		= "/nfs",
-};
-
-struct rpc_stat nfs_rpcstat = {
-	.program		= &nfs_program
-};
-
-
-#ifdef CONFIG_NFS_V3_ACL
-static struct rpc_stat		nfsacl_rpcstat = { &nfsacl_program };
-static struct rpc_version *	nfsacl_version[] = {
-	[3]			= &nfsacl_version3,
-};
-
-struct rpc_program		nfsacl_program = {
-	.name =			"nfsacl",
-	.number =		NFS_ACL_PROGRAM,
-	.nrvers =		ARRAY_SIZE(nfsacl_version),
-	.version =		nfsacl_version,
-	.stats =		&nfsacl_rpcstat,
-};
-#endif  /* CONFIG_NFS_V3_ACL */
-
 static void nfs_umount_begin(struct vfsmount *, int);
 static int  nfs_statfs(struct dentry *, struct kstatfs *);
 static int  nfs_show_options(struct seq_file *, struct vfsmount *);
@@ -376,8 +330,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 			proto = buf;
 	}
 	seq_printf(m, ",proto=%s", proto);
-	seq_printf(m, ",timeo=%lu", 10U * nfss->retrans_timeo / HZ);
-	seq_printf(m, ",retrans=%u", nfss->retrans_count);
+	seq_printf(m, ",timeo=%lu", 10U * clp->retrans_timeo / HZ);
+	seq_printf(m, ",retrans=%u", clp->retrans_count);
 	seq_printf(m, ",sec=%s", nfs_pseudoflavour_to_name(nfss->client->cl_auth->au_flavor));
 }
 
@@ -621,38 +575,6 @@ out_no_root:
 	return no_root_error;
 }
 
-/*
- * Initialise the timeout values for a connection
- */
-static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, unsigned int timeo, unsigned int retrans)
-{
-	to->to_initval = timeo * HZ / 10;
-	to->to_retries = retrans;
-	if (!to->to_retries)
-		to->to_retries = 2;
-
-	switch (proto) {
-	case IPPROTO_TCP:
-		if (!to->to_initval)
-			to->to_initval = 60 * HZ;
-		if (to->to_initval > NFS_MAX_TCP_TIMEOUT)
-			to->to_initval = NFS_MAX_TCP_TIMEOUT;
-		to->to_increment = to->to_initval;
-		to->to_maxval = to->to_initval + (to->to_increment * to->to_retries);
-		to->to_exponential = 0;
-		break;
-	case IPPROTO_UDP:
-	default:
-		if (!to->to_initval)
-			to->to_initval = 11 * HZ / 10;
-		if (to->to_initval > NFS_MAX_UDP_TIMEOUT)
-			to->to_initval = NFS_MAX_UDP_TIMEOUT;
-		to->to_maxval = NFS_MAX_UDP_TIMEOUT;
-		to->to_exponential = 1;
-		break;
-	}
-}
-
 /*
  * Create an RPC client handle.
  */
@@ -660,11 +582,10 @@ static struct rpc_clnt *
 nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 {
 	struct nfs_client	*clp;
-	struct rpc_timeout	timeparms;
-	struct rpc_xprt		*xprt = NULL;
-	struct rpc_clnt		*clnt = NULL;
+	struct rpc_clnt		*clnt;
 	int			proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
 	int			nfsversion = 2;
+	int			err;
 
 #ifdef CONFIG_NFS_V3
 	if (server->flags & NFS_MOUNT_VER3)
@@ -677,52 +598,54 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
 		return ERR_PTR(PTR_ERR(clp));
 	}
 
-	nfs_init_timeout_values(&timeparms, proto, data->timeo, data->retrans);
-
-	server->retrans_timeo = timeparms.to_initval;
-	server->retrans_count = timeparms.to_retries;
-
-	/* Check NFS protocol revision and initialize RPC op vector
-	 * and file handle pool. */
+	if (clp->cl_cons_state == NFS_CS_INITING) {
+		/* Check NFS protocol revision and initialize RPC op
+		 * vector and file handle pool. */
 #ifdef CONFIG_NFS_V3
-	if (nfsversion == 3) {
-		clp->rpc_ops = &nfs_v3_clientops;
-		server->caps |= NFS_CAP_READDIRPLUS;
-	} else {
-		clp->rpc_ops = &nfs_v2_clientops;
-	}
+		if (nfsversion == 3) {
+			clp->rpc_ops = &nfs_v3_clientops;
+			server->caps |= NFS_CAP_READDIRPLUS;
+		} else {
+			clp->rpc_ops = &nfs_v2_clientops;
+		}
 #else
-	clp->rpc_ops = &nfs_v2_clientops;
+		clp->rpc_ops = &nfs_v2_clientops;
 #endif
 
-	/* create transport and client */
-	xprt = xprt_create_proto(proto, &server->addr, &timeparms);
-	if (IS_ERR(xprt)) {
-		dprintk("%s: cannot create RPC transport. Error = %ld\n",
-				__FUNCTION__, PTR_ERR(xprt));
-		nfs_mark_client_ready(clp, PTR_ERR(xprt));
-		nfs_put_client(clp);
-		return (struct rpc_clnt *)xprt;
+		/* create transport and client */
+		err = nfs_create_rpc_client(clp, proto, data->timeo,
+					    data->retrans, RPC_AUTH_UNIX);
+		if (err < 0)
+			goto client_init_error;
+
+		nfs_mark_client_ready(clp, 0);
 	}
-	clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
-				 clp->cl_nfsversion, data->pseudoflavor);
+
+	/* create an nfs_server-specific client */
+	clnt = rpc_clone_client(clp->cl_rpcclient);
 	if (IS_ERR(clnt)) {
-		dprintk("%s: cannot create RPC client. Error = %ld\n",
-				__FUNCTION__, PTR_ERR(xprt));
-		goto out_fail;
+		dprintk("%s: couldn't create rpc_client!\n", __FUNCTION__);
+		nfs_put_client(clp);
+		return ERR_PTR(PTR_ERR(clnt));
 	}
 
-	clnt->cl_intr     = 1;
-	clnt->cl_softrtry = 1;
+	if (data->pseudoflavor != clp->cl_rpcclient->cl_auth->au_flavor) {
+		struct rpc_auth *auth;
+
+		auth = rpcauth_create(data->pseudoflavor, server->client);
+		if (IS_ERR(auth)) {
+			dprintk("%s: couldn't create credcache!\n", __FUNCTION__);
+			return ERR_PTR(PTR_ERR(auth));
+		}
+	}
 
-	nfs_mark_client_ready(clp, 0);
 	server->nfs_client = clp;
 	return clnt;
 
-out_fail:
-	nfs_mark_client_ready(clp, PTR_ERR(xprt));
+client_init_error:
+	nfs_mark_client_ready(clp, err);
 	nfs_put_client(clp);
-	return clnt;
+	return ERR_PTR(err);
 }
 
 /*
@@ -741,7 +664,7 @@ static struct nfs_server *nfs_clone_server(struct super_block *sb, struct nfs_cl
 	sb->s_blocksize_bits = data->sb->s_blocksize_bits;
 	sb->s_maxbytes = data->sb->s_maxbytes;
 
-	server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
+	server->client_acl = ERR_PTR(-EINVAL);
 	server->io_stats = nfs_alloc_iostats();
 	if (server->io_stats == NULL)
 		goto out;
@@ -750,11 +673,6 @@ static struct nfs_server *nfs_clone_server(struct super_block *sb, struct nfs_cl
 	if (IS_ERR((err = server->client)))
 		goto out;
 
-	if (!IS_ERR(parent->client_sys)) {
-		server->client_sys = rpc_clone_client(parent->client_sys);
-		if (IS_ERR((err = server->client_sys)))
-			goto out;
-	}
 	if (!IS_ERR(parent->client_acl)) {
 		server->client_acl = rpc_clone_client(parent->client_acl);
 		if (IS_ERR((err = server->client_acl)))
@@ -813,7 +731,7 @@ static int nfs_clone_generic_sb(struct nfs_clone_mount *data,
 		error = PTR_ERR(sb);
 		goto kill_rpciod;
 	}
-		
+
 	if (sb->s_root)
 		goto out_rpciod_down;
 
@@ -896,19 +814,6 @@ nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
 		return PTR_ERR(server->client);
 
 	/* RFC 2623, sec 2.3.2 */
-	if (authflavor != RPC_AUTH_UNIX) {
-		struct rpc_auth *auth;
-
-		server->client_sys = rpc_clone_client(server->client);
-		if (IS_ERR(server->client_sys))
-			return PTR_ERR(server->client_sys);
-		auth = rpcauth_create(RPC_AUTH_UNIX, server->client_sys);
-		if (IS_ERR(auth))
-			return PTR_ERR(auth);
-	} else {
-		atomic_inc(&server->client->cl_count);
-		server->client_sys = server->client;
-	}
 	if (server->flags & NFS_MOUNT_VER3) {
 #ifdef CONFIG_NFS_V3_ACL
 		if (!(server->flags & NFS_MOUNT_NOACL)) {
@@ -1012,7 +917,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 		goto out_err_noserver;
 	/* Zero out the NFS state stuff */
 	init_nfsv4_state(server);
-	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
+	server->client = server->client_acl = ERR_PTR(-EINVAL);
 
 	root = &server->fh;
 	if (data->flags & NFS_MOUNT_VER3)
@@ -1083,8 +988,6 @@ static void nfs_kill_super(struct super_block *s)
 
 	if (!IS_ERR(server->client))
 		rpc_shutdown_client(server->client);
-	if (!IS_ERR(server->client_sys))
-		rpc_shutdown_client(server->client_sys);
 	if (!IS_ERR(server->client_acl))
 		rpc_shutdown_client(server->client_acl);
 
@@ -1121,10 +1024,9 @@ static int nfs_clone_nfs_sb(struct file_system_type *fs_type,
 
 #ifdef CONFIG_NFS_V4
 static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
-	struct rpc_timeout *timeparms, int proto, rpc_authflavor_t flavor)
+	int timeo, int retrans, int proto, rpc_authflavor_t flavor)
 {
 	struct nfs_client *clp;
-	struct rpc_xprt *xprt = NULL;
 	struct rpc_clnt *clnt = NULL;
 	int err = -EIO;
 
@@ -1138,26 +1040,10 @@ static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 	if (clp->cl_cons_state == NFS_CS_INITING) {
 		clp->rpc_ops = &nfs_v4_clientops;
 
-		xprt = xprt_create_proto(proto, &server->addr, timeparms);
-		if (IS_ERR(xprt)) {
-			err = PTR_ERR(xprt);
-			dprintk("%s: cannot create RPC transport. Error = %d\n",
-					__FUNCTION__, err);
+		err = nfs_create_rpc_client(clp, proto, timeo, retrans, flavor);
+		if (err < 0)
 			goto client_init_error;
-		}
-		/* Bind to a reserved port! */
-		xprt->resvport = 1;
-		clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
-				clp->cl_nfsversion, flavor);
-		if (IS_ERR(clnt)) {
-			err = PTR_ERR(clnt);
-			dprintk("%s: cannot create RPC client. Error = %d\n",
-					__FUNCTION__, err);
-			goto client_init_error;
-		}
-		clnt->cl_intr     = 1;
-		clnt->cl_softrtry = 1;
-		clp->cl_rpcclient = clnt;
+
 		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
 		err = nfs_idmap_new(clp);
 		if (err < 0) {
@@ -1205,7 +1091,6 @@ client_init_error:
 static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, int silent)
 {
 	struct nfs_server *server;
-	struct rpc_timeout timeparms;
 	rpc_authflavor_t authflavour;
 	int err = -EIO;
 
@@ -1224,11 +1109,6 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data,
 	server->acdirmin = data->acdirmin*HZ;
 	server->acdirmax = data->acdirmax*HZ;
 
-	nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans);
-
-	server->retrans_timeo = timeparms.to_initval;
-	server->retrans_count = timeparms.to_retries;
-
 	/* Now create transport and client */
 	authflavour = RPC_AUTH_UNIX;
 	if (data->auth_flavourlen != 0) {
@@ -1244,7 +1124,8 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data,
 		}
 	}
 
-	server->client = nfs4_create_client(server, &timeparms, data->proto, authflavour);
+	server->client = nfs4_create_client(server, data->timeo, data->retrans,
+					    data->proto, authflavour);
 	if (IS_ERR(server->client)) {
 		err = PTR_ERR(server->client);
 			dprintk("%s: cannot create RPC client. Error = %d\n",
@@ -1318,7 +1199,7 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 		return -ENOMEM;
 	/* Zero out the NFS state stuff */
 	init_nfsv4_state(server);
-	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
+	server->client = server->client_acl = ERR_PTR(-EINVAL);
 
 	p = nfs_copy_user_string(NULL, &data->hostname, 256);
 	if (IS_ERR(p))
@@ -1489,7 +1370,6 @@ err:
 static struct nfs_server *nfs4_referral_server(struct super_block *sb, struct nfs_clone_mount *data)
 {
 	struct nfs_server *server = NFS_SB(sb);
-	struct rpc_timeout timeparms;
 	int proto, timeo, retrans;
 	void *err;
 
@@ -1498,11 +1378,11 @@ static struct nfs_server *nfs4_referral_server(struct super_block *sb, struct nf
 	   set the timeouts and retries to low values */
 	timeo = 2;
 	retrans = 1;
-	nfs_init_timeout_values(&timeparms, proto, timeo, retrans);
 
 	nfs_put_client(server->nfs_client);
 	server->nfs_client = NULL;
-	server->client = nfs4_create_client(server, &timeparms, proto, data->authflavor);
+	server->client = nfs4_create_client(server, timeo, retrans, proto,
+					    data->authflavor);
 	if (IS_ERR((err = server->client)))
 		goto out_err;
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index aae7c117597a..d404ceca9168 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -26,6 +26,8 @@ struct nfs_client {
 
 	struct rpc_clnt *	cl_rpcclient;
 	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
+	unsigned long		retrans_timeo;	/* retransmit timeout */
+	unsigned int		retrans_count;	/* number of retransmit tries */
 
 #ifdef CONFIG_NFS_V4
 	u64			cl_clientid;	/* constant */
@@ -73,7 +75,6 @@ struct nfs_client {
 struct nfs_server {
 	struct nfs_client *	nfs_client;	/* shared client and NFS4 state */
 	struct rpc_clnt *	client;		/* RPC client handle */
-	struct rpc_clnt *	client_sys;	/* 2nd handle for FSINFO */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
@@ -90,8 +91,6 @@ struct nfs_server {
 	unsigned int		acregmax;
 	unsigned int		acdirmin;
 	unsigned int		acdirmax;
-	unsigned long		retrans_timeo;	/* retransmit timeout */
-	unsigned int		retrans_count;	/* number of retransmit tries */
 	unsigned int		namelen;
 	char *			hostname;	/* remote hostname */
 	struct nfs_fh		fh;
-- 
cgit v1.2.3


From 54ceac4515986030c2502960be620198dd8fe25b Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 22 Aug 2006 20:06:13 -0400
Subject: NFS: Share NFS superblocks per-protocol per-server per-FSID

The attached patch makes NFS share superblocks between mounts from the same
server and FSID over the same protocol.

It does this by creating each superblock with a false root and returning the
real root dentry in the vfsmount presented by get_sb(). The root dentry set
starts off as an anonymous dentry if we don't already have the dentry for its
inode, otherwise it simply returns the dentry we already have.

We may thus end up with several trees of dentries in the superblock, and if at
some later point one of anonymous tree roots is discovered by normal filesystem
activity to be located in another tree within the superblock, the anonymous
root is named and materialises attached to the second tree at the appropriate
point.

Why do it this way? Why not pass an extra argument to the mount() syscall to
indicate the subpath and then pathwalk from the server root to the desired
directory? You can't guarantee this will work for two reasons:

 (1) The root and intervening nodes may not be accessible to the client.

     With NFS2 and NFS3, for instance, mountd is called on the server to get
     the filehandle for the tip of a path. mountd won't give us handles for
     anything we don't have permission to access, and so we can't set up NFS
     inodes for such nodes, and so can't easily set up dentries (we'd have to
     have ghost inodes or something).

     With this patch we don't actually create dentries until we get handles
     from the server that we can use to set up their inodes, and we don't
     actually bind them into the tree until we know for sure where they go.

 (2) Inaccessible symbolic links.

     If we're asked to mount two exports from the server, eg:

	mount warthog:/warthog/aaa/xxx /mmm
	mount warthog:/warthog/bbb/yyy /nnn

     We may not be able to access anything nearer the root than xxx and yyy,
     but we may find out later that /mmm/www/yyy, say, is actually the same
     directory as the one mounted on /nnn. What we might then find out, for
     example, is that /warthog/bbb was actually a symbolic link to
     /warthog/aaa/xxx/www, but we can't actually determine that by talking to
     the server until /warthog is made available by NFS.

     This would lead to having constructed an errneous dentry tree which we
     can't easily fix. We can end up with a dentry marked as a directory when
     it should actually be a symlink, or we could end up with an apparently
     hardlinked directory.

     With this patch we need not make assumptions about the type of a dentry
     for which we can't retrieve information, nor need we assume we know its
     place in the grand scheme of things until we actually see that place.

This patch reduces the possibility of aliasing in the inode and page caches for
inodes that may be accessed by more than one NFS export. It also reduces the
number of superblocks required for NFS where there are many NFS exports being
used from a server (home directory server + autofs for example).

This in turn makes it simpler to do local caching of network filesystems, as it
can then be guaranteed that there won't be links from multiple inodes in
separate superblocks to the same cache file.

Obviously, cache aliasing between different levels of NFS protocol could still
be a problem, but at least that gives us another key to use when indexing the
cache.

This patch makes the following changes:

 (1) The server record construction/destruction has been abstracted out into
     its own set of functions to make things easier to get right.  These have
     been moved into fs/nfs/client.c.

     All the code in fs/nfs/client.c has to do with the management of
     connections to servers, and doesn't touch superblocks in any way; the
     remaining code in fs/nfs/super.c has to do with VFS superblock management.

 (2) The sequence of events undertaken by NFS mount is now reordered:

     (a) A volume representation (struct nfs_server) is allocated.

     (b) A server representation (struct nfs_client) is acquired.  This may be
     	 allocated or shared, and is keyed on server address, port and NFS
     	 version.

     (c) If allocated, the client representation is initialised.  The state
     	 member variable of nfs_client is used to prevent a race during
     	 initialisation from two mounts.

     (d) For NFS4 a simple pathwalk is performed, walking from FH to FH to find
     	 the root filehandle for the mount (fs/nfs/getroot.c).  For NFS2/3 we
     	 are given the root FH in advance.

     (e) The volume FSID is probed for on the root FH.

     (f) The volume representation is initialised from the FSINFO record
     	 retrieved on the root FH.

     (g) sget() is called to acquire a superblock.  This may be allocated or
     	 shared, keyed on client pointer and FSID.

     (h) If allocated, the superblock is initialised.

     (i) If the superblock is shared, then the new nfs_server record is
     	 discarded.

     (j) The root dentry for this mount is looked up from the root FH.

     (k) The root dentry for this mount is assigned to the vfsmount.

 (3) nfs_readdir_lookup() creates dentries for each of the entries readdir()
     returns; this function now attaches disconnected trees from alternate
     roots that happen to be discovered attached to a directory being read (in
     the same way nfs_lookup() is made to do for lookup ops).

     The new d_materialise_unique() function is now used to do this, thus
     permitting the whole thing to be done under one set of locks, and thus
     avoiding any race between mount and lookup operations on the same
     directory.

 (4) The client management code uses a new debug facility: NFSDBG_CLIENT which
     is set by echoing 1024 to /proc/net/sunrpc/nfs_debug.

 (5) Clone mounts are now called xdev mounts.

 (6) Use the dentry passed to the statfs() op as the handle for retrieving fs
     statistics rather than the root dentry of the superblock (which is now a
     dummy).

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/Makefile           |    2 +-
 fs/nfs/client.c           |  735 ++++++++++++++++++++++++++-
 fs/nfs/dir.c              |   16 +-
 fs/nfs/getroot.c          |  306 ++++++++++++
 fs/nfs/idmap.c            |    3 +-
 fs/nfs/inode.c            |    2 +-
 fs/nfs/internal.h         |   82 +--
 fs/nfs/namespace.c        |   25 +-
 fs/nfs/nfs3proc.c         |    2 +-
 fs/nfs/nfs4_fs.h          |    6 -
 fs/nfs/nfs4namespace.c    |  110 ++++-
 fs/nfs/nfs4proc.c         |   59 +--
 fs/nfs/nfs4renewd.c       |   13 -
 fs/nfs/nfs4state.c        |   18 -
 fs/nfs/read.c             |    2 +-
 fs/nfs/super.c            | 1207 +++++++++++++++++----------------------------
 fs/nfs/write.c            |    2 +-
 include/linux/nfs_fs_sb.h |   21 +-
 18 files changed, 1655 insertions(+), 956 deletions(-)
 create mode 100644 fs/nfs/getroot.c

(limited to 'include')

diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile
index 3b993a6f8163..f4580b44eef4 100644
--- a/fs/nfs/Makefile
+++ b/fs/nfs/Makefile
@@ -4,7 +4,7 @@
 
 obj-$(CONFIG_NFS_FS) += nfs.o
 
-nfs-y 			:= client.o dir.o file.o inode.o super.o nfs2xdr.o \
+nfs-y 			:= client.o dir.o file.o getroot.o inode.o super.o nfs2xdr.o \
 			   pagelist.o proc.o read.o symlink.o unlink.o \
 			   write.o namespace.o
 nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o mount_clnt.o      
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index c08cab935ad5..dafba608c0a0 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -48,6 +48,7 @@
 
 static DEFINE_SPINLOCK(nfs_client_lock);
 static LIST_HEAD(nfs_client_list);
+static LIST_HEAD(nfs_volume_list);
 static DECLARE_WAIT_QUEUE_HEAD(nfs_client_active_wq);
 
 /*
@@ -268,9 +269,9 @@ struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversio
  * Look up a client by IP address and protocol version
  * - creates a new record if one doesn't yet exist
  */
-struct nfs_client *nfs_get_client(const char *hostname,
-				  const struct sockaddr_in *addr,
-				  int nfsversion)
+static struct nfs_client *nfs_get_client(const char *hostname,
+					 const struct sockaddr_in *addr,
+					 int nfsversion)
 {
 	struct nfs_client *clp, *new = NULL;
 	int error;
@@ -340,6 +341,8 @@ found_client:
 		return ERR_PTR(error);
 	}
 
+	BUG_ON(clp->cl_cons_state != NFS_CS_READY);
+
 	dprintk("--> nfs_get_client() = %p [share]\n", clp);
 	return clp;
 }
@@ -347,7 +350,7 @@ found_client:
 /*
  * Mark a server as ready or failed
  */
-void nfs_mark_client_ready(struct nfs_client *clp, int state)
+static void nfs_mark_client_ready(struct nfs_client *clp, int state)
 {
 	clp->cl_cons_state = state;
 	wake_up_all(&nfs_client_active_wq);
@@ -389,10 +392,10 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 /*
  * Create an RPC client handle
  */
-int nfs_create_rpc_client(struct nfs_client *clp, int proto,
-			  unsigned int timeo,
-			  unsigned int retrans,
-			  rpc_authflavor_t flavor)
+static int nfs_create_rpc_client(struct nfs_client *clp, int proto,
+						unsigned int timeo,
+						unsigned int retrans,
+						rpc_authflavor_t flavor)
 {
 	struct rpc_timeout	timeparms;
 	struct rpc_xprt		*xprt = NULL;
@@ -429,3 +432,719 @@ int nfs_create_rpc_client(struct nfs_client *clp, int proto,
 	clp->cl_rpcclient = clnt;
 	return 0;
 }
+
+/*
+ * Version 2 or 3 client destruction
+ */
+static void nfs_destroy_server(struct nfs_server *server)
+{
+	if (!IS_ERR(server->client_acl))
+		rpc_shutdown_client(server->client_acl);
+
+	if (!(server->flags & NFS_MOUNT_NONLM))
+		lockd_down();	/* release rpc.lockd */
+}
+
+/*
+ * Version 2 or 3 lockd setup
+ */
+static int nfs_start_lockd(struct nfs_server *server)
+{
+	int error = 0;
+
+	if (server->nfs_client->cl_nfsversion > 3)
+		goto out;
+	if (server->flags & NFS_MOUNT_NONLM)
+		goto out;
+	error = lockd_up();
+	if (error < 0)
+		server->flags |= NFS_MOUNT_NONLM;
+	else
+		server->destroy = nfs_destroy_server;
+out:
+	return error;
+}
+
+/*
+ * Initialise an NFSv3 ACL client connection
+ */
+#ifdef CONFIG_NFS_V3_ACL
+static void nfs_init_server_aclclient(struct nfs_server *server)
+{
+	if (server->nfs_client->cl_nfsversion != 3)
+		goto out_noacl;
+	if (server->flags & NFS_MOUNT_NOACL)
+		goto out_noacl;
+
+	server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3);
+	if (IS_ERR(server->client_acl))
+		goto out_noacl;
+
+	/* No errors! Assume that Sun nfsacls are supported */
+	server->caps |= NFS_CAP_ACLS;
+	return;
+
+out_noacl:
+	server->caps &= ~NFS_CAP_ACLS;
+}
+#else
+static inline void nfs_init_server_aclclient(struct nfs_server *server)
+{
+	server->flags &= ~NFS_MOUNT_NOACL;
+	server->caps &= ~NFS_CAP_ACLS;
+}
+#endif
+
+/*
+ * Create a general RPC client
+ */
+static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t pseudoflavour)
+{
+	struct nfs_client *clp = server->nfs_client;
+
+	server->client = rpc_clone_client(clp->cl_rpcclient);
+	if (IS_ERR(server->client)) {
+		dprintk("%s: couldn't create rpc_client!\n", __FUNCTION__);
+		return PTR_ERR(server->client);
+	}
+
+	if (pseudoflavour != clp->cl_rpcclient->cl_auth->au_flavor) {
+		struct rpc_auth *auth;
+
+		auth = rpcauth_create(pseudoflavour, server->client);
+		if (IS_ERR(auth)) {
+			dprintk("%s: couldn't create credcache!\n", __FUNCTION__);
+			return PTR_ERR(auth);
+		}
+	}
+	server->client->cl_softrtry = 0;
+	if (server->flags & NFS_MOUNT_SOFT)
+		server->client->cl_softrtry = 1;
+
+	server->client->cl_intr = 0;
+	if (server->flags & NFS4_MOUNT_INTR)
+		server->client->cl_intr = 1;
+
+	return 0;
+}
+
+/*
+ * Initialise an NFS2 or NFS3 client
+ */
+static int nfs_init_client(struct nfs_client *clp, const struct nfs_mount_data *data)
+{
+	int proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
+	int error;
+
+	if (clp->cl_cons_state == NFS_CS_READY) {
+		/* the client is already initialised */
+		dprintk("<-- nfs_init_client() = 0 [already %p]\n", clp);
+		return 0;
+	}
+
+	/* Check NFS protocol revision and initialize RPC op vector */
+	clp->rpc_ops = &nfs_v2_clientops;
+#ifdef CONFIG_NFS_V3
+	if (clp->cl_nfsversion == 3)
+		clp->rpc_ops = &nfs_v3_clientops;
+#endif
+	/*
+	 * Create a client RPC handle for doing FSSTAT with UNIX auth only
+	 * - RFC 2623, sec 2.3.2
+	 */
+	error = nfs_create_rpc_client(clp, proto, data->timeo, data->retrans,
+			RPC_AUTH_UNIX);
+	if (error < 0)
+		goto error;
+	nfs_mark_client_ready(clp, NFS_CS_READY);
+	return 0;
+
+error:
+	nfs_mark_client_ready(clp, error);
+	dprintk("<-- nfs_init_client() = xerror %d\n", error);
+	return error;
+}
+
+/*
+ * Create a version 2 or 3 client
+ */
+static int nfs_init_server(struct nfs_server *server, const struct nfs_mount_data *data)
+{
+	struct nfs_client *clp;
+	int error, nfsvers = 2;
+
+	dprintk("--> nfs_init_server()\n");
+
+#ifdef CONFIG_NFS_V3
+	if (data->flags & NFS_MOUNT_VER3)
+		nfsvers = 3;
+#endif
+
+	/* Allocate or find a client reference we can use */
+	clp = nfs_get_client(data->hostname, &data->addr, nfsvers);
+	if (IS_ERR(clp)) {
+		dprintk("<-- nfs_init_server() = error %ld\n", PTR_ERR(clp));
+		return PTR_ERR(clp);
+	}
+
+	error = nfs_init_client(clp, data);
+	if (error < 0)
+		goto error;
+
+	server->nfs_client = clp;
+
+	/* Initialise the client representation from the mount data */
+	server->flags = data->flags & NFS_MOUNT_FLAGMASK;
+
+	if (data->rsize)
+		server->rsize = nfs_block_size(data->rsize, NULL);
+	if (data->wsize)
+		server->wsize = nfs_block_size(data->wsize, NULL);
+
+	server->acregmin = data->acregmin * HZ;
+	server->acregmax = data->acregmax * HZ;
+	server->acdirmin = data->acdirmin * HZ;
+	server->acdirmax = data->acdirmax * HZ;
+
+	/* Start lockd here, before we might error out */
+	error = nfs_start_lockd(server);
+	if (error < 0)
+		goto error;
+
+	error = nfs_init_server_rpcclient(server, data->pseudoflavor);
+	if (error < 0)
+		goto error;
+
+	server->namelen  = data->namlen;
+	/* Create a client RPC handle for the NFSv3 ACL management interface */
+	nfs_init_server_aclclient(server);
+	if (clp->cl_nfsversion == 3) {
+		if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN)
+			server->namelen = NFS3_MAXNAMLEN;
+		server->caps |= NFS_CAP_READDIRPLUS;
+	} else {
+		if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN)
+			server->namelen = NFS2_MAXNAMLEN;
+	}
+
+	dprintk("<-- nfs_init_server() = 0 [new %p]\n", clp);
+	return 0;
+
+error:
+	server->nfs_client = NULL;
+	nfs_put_client(clp);
+	dprintk("<-- nfs_init_server() = xerror %d\n", error);
+	return error;
+}
+
+/*
+ * Load up the server record from information gained in an fsinfo record
+ */
+static void nfs_server_set_fsinfo(struct nfs_server *server, struct nfs_fsinfo *fsinfo)
+{
+	unsigned long max_rpc_payload;
+
+	/* Work out a lot of parameters */
+	if (server->rsize == 0)
+		server->rsize = nfs_block_size(fsinfo->rtpref, NULL);
+	if (server->wsize == 0)
+		server->wsize = nfs_block_size(fsinfo->wtpref, NULL);
+
+	if (fsinfo->rtmax >= 512 && server->rsize > fsinfo->rtmax)
+		server->rsize = nfs_block_size(fsinfo->rtmax, NULL);
+	if (fsinfo->wtmax >= 512 && server->wsize > fsinfo->wtmax)
+		server->wsize = nfs_block_size(fsinfo->wtmax, NULL);
+
+	max_rpc_payload = nfs_block_size(rpc_max_payload(server->client), NULL);
+	if (server->rsize > max_rpc_payload)
+		server->rsize = max_rpc_payload;
+	if (server->rsize > NFS_MAX_FILE_IO_SIZE)
+		server->rsize = NFS_MAX_FILE_IO_SIZE;
+	server->rpages = (server->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	server->backing_dev_info.ra_pages = server->rpages * NFS_MAX_READAHEAD;
+
+	if (server->wsize > max_rpc_payload)
+		server->wsize = max_rpc_payload;
+	if (server->wsize > NFS_MAX_FILE_IO_SIZE)
+		server->wsize = NFS_MAX_FILE_IO_SIZE;
+	server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	server->wtmult = nfs_block_bits(fsinfo->wtmult, NULL);
+
+	server->dtsize = nfs_block_size(fsinfo->dtpref, NULL);
+	if (server->dtsize > PAGE_CACHE_SIZE)
+		server->dtsize = PAGE_CACHE_SIZE;
+	if (server->dtsize > server->rsize)
+		server->dtsize = server->rsize;
+
+	if (server->flags & NFS_MOUNT_NOAC) {
+		server->acregmin = server->acregmax = 0;
+		server->acdirmin = server->acdirmax = 0;
+	}
+
+	server->maxfilesize = fsinfo->maxfilesize;
+
+	/* We're airborne Set socket buffersize */
+	rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100);
+}
+
+/*
+ * Probe filesystem information, including the FSID on v2/v3
+ */
+static int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs_fattr *fattr)
+{
+	struct nfs_fsinfo fsinfo;
+	struct nfs_client *clp = server->nfs_client;
+	int error;
+
+	dprintk("--> nfs_probe_fsinfo()\n");
+
+	if (clp->rpc_ops->set_capabilities != NULL) {
+		error = clp->rpc_ops->set_capabilities(server, mntfh);
+		if (error < 0)
+			goto out_error;
+	}
+
+	fsinfo.fattr = fattr;
+	nfs_fattr_init(fattr);
+	error = clp->rpc_ops->fsinfo(server, mntfh, &fsinfo);
+	if (error < 0)
+		goto out_error;
+
+	nfs_server_set_fsinfo(server, &fsinfo);
+
+	/* Get some general file system info */
+	if (server->namelen == 0) {
+		struct nfs_pathconf pathinfo;
+
+		pathinfo.fattr = fattr;
+		nfs_fattr_init(fattr);
+
+		if (clp->rpc_ops->pathconf(server, mntfh, &pathinfo) >= 0)
+			server->namelen = pathinfo.max_namelen;
+	}
+
+	dprintk("<-- nfs_probe_fsinfo() = 0\n");
+	return 0;
+
+out_error:
+	dprintk("nfs_probe_fsinfo: error = %d\n", -error);
+	return error;
+}
+
+/*
+ * Copy useful information when duplicating a server record
+ */
+static void nfs_server_copy_userdata(struct nfs_server *target, struct nfs_server *source)
+{
+	target->flags = source->flags;
+	target->acregmin = source->acregmin;
+	target->acregmax = source->acregmax;
+	target->acdirmin = source->acdirmin;
+	target->acdirmax = source->acdirmax;
+	target->caps = source->caps;
+}
+
+/*
+ * Allocate and initialise a server record
+ */
+static struct nfs_server *nfs_alloc_server(void)
+{
+	struct nfs_server *server;
+
+	server = kzalloc(sizeof(struct nfs_server), GFP_KERNEL);
+	if (!server)
+		return NULL;
+
+	server->client = server->client_acl = ERR_PTR(-EINVAL);
+
+	/* Zero out the NFS state stuff */
+	INIT_LIST_HEAD(&server->client_link);
+	INIT_LIST_HEAD(&server->master_link);
+
+	server->io_stats = nfs_alloc_iostats();
+	if (!server->io_stats) {
+		kfree(server);
+		return NULL;
+	}
+
+	return server;
+}
+
+/*
+ * Free up a server record
+ */
+void nfs_free_server(struct nfs_server *server)
+{
+	dprintk("--> nfs_free_server()\n");
+
+	spin_lock(&nfs_client_lock);
+	list_del(&server->client_link);
+	list_del(&server->master_link);
+	spin_unlock(&nfs_client_lock);
+
+	if (server->destroy != NULL)
+		server->destroy(server);
+	if (!IS_ERR(server->client))
+		rpc_shutdown_client(server->client);
+
+	nfs_put_client(server->nfs_client);
+
+	nfs_free_iostats(server->io_stats);
+	kfree(server);
+	nfs_release_automount_timer();
+	dprintk("<-- nfs_free_server()\n");
+}
+
+/*
+ * Create a version 2 or 3 volume record
+ * - keyed on server and FSID
+ */
+struct nfs_server *nfs_create_server(const struct nfs_mount_data *data,
+				     struct nfs_fh *mntfh)
+{
+	struct nfs_server *server;
+	struct nfs_fattr fattr;
+	int error;
+
+	server = nfs_alloc_server();
+	if (!server)
+		return ERR_PTR(-ENOMEM);
+
+	/* Get a client representation */
+	error = nfs_init_server(server, data);
+	if (error < 0)
+		goto error;
+
+	BUG_ON(!server->nfs_client);
+	BUG_ON(!server->nfs_client->rpc_ops);
+	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
+
+	/* Probe the root fh to retrieve its FSID */
+	error = nfs_probe_fsinfo(server, mntfh, &fattr);
+	if (error < 0)
+		goto error;
+	if (!(fattr.valid & NFS_ATTR_FATTR)) {
+		error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr);
+		if (error < 0) {
+			dprintk("nfs_create_server: getattr error = %d\n", -error);
+			goto error;
+		}
+	}
+	memcpy(&server->fsid, &fattr.fsid, sizeof(server->fsid));
+
+	dprintk("Server FSID: %llx:%llx\n", server->fsid.major, server->fsid.minor);
+
+	BUG_ON(!server->nfs_client);
+	BUG_ON(!server->nfs_client->rpc_ops);
+	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
+
+	spin_lock(&nfs_client_lock);
+	list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
+	list_add_tail(&server->master_link, &nfs_volume_list);
+	spin_unlock(&nfs_client_lock);
+
+	server->mount_time = jiffies;
+	return server;
+
+error:
+	nfs_free_server(server);
+	return ERR_PTR(error);
+}
+
+#ifdef CONFIG_NFS_V4
+/*
+ * Initialise an NFS4 client record
+ */
+static int nfs4_init_client(struct nfs_client *clp,
+		int proto, int timeo, int retrans,
+		rpc_authflavor_t authflavour)
+{
+	int error;
+
+	if (clp->cl_cons_state == NFS_CS_READY) {
+		/* the client is initialised already */
+		dprintk("<-- nfs4_init_client() = 0 [already %p]\n", clp);
+		return 0;
+	}
+
+	/* Check NFS protocol revision and initialize RPC op vector */
+	clp->rpc_ops = &nfs_v4_clientops;
+
+	error = nfs_create_rpc_client(clp, proto, timeo, retrans, authflavour);
+	if (error < 0)
+		goto error;
+
+	error = nfs_idmap_new(clp);
+	if (error < 0) {
+		dprintk("%s: failed to create idmapper. Error = %d\n",
+			__FUNCTION__, error);
+		__set_bit(NFS_CS_IDMAP, &clp->cl_res_state);
+		goto error;
+	}
+
+	nfs_mark_client_ready(clp, NFS_CS_READY);
+	return 0;
+
+error:
+	nfs_mark_client_ready(clp, error);
+	dprintk("<-- nfs4_init_client() = xerror %d\n", error);
+	return error;
+}
+
+/*
+ * Set up an NFS4 client
+ */
+static int nfs4_set_client(struct nfs_server *server,
+		const char *hostname, const struct sockaddr_in *addr,
+		rpc_authflavor_t authflavour,
+		int proto, int timeo, int retrans)
+{
+	struct nfs_client *clp;
+	int error;
+
+	dprintk("--> nfs4_set_client()\n");
+
+	/* Allocate or find a client reference we can use */
+	clp = nfs_get_client(hostname, addr, 4);
+	if (IS_ERR(clp)) {
+		error = PTR_ERR(clp);
+		goto error;
+	}
+	error = nfs4_init_client(clp, proto, timeo, retrans, authflavour);
+	if (error < 0)
+		goto error_put;
+
+	server->nfs_client = clp;
+	dprintk("<-- nfs4_set_client() = 0 [new %p]\n", clp);
+	return 0;
+
+error_put:
+	nfs_put_client(clp);
+error:
+	dprintk("<-- nfs4_set_client() = xerror %d\n", error);
+	return error;
+}
+
+/*
+ * Create a version 4 volume record
+ */
+static int nfs4_init_server(struct nfs_server *server,
+		const struct nfs4_mount_data *data, rpc_authflavor_t authflavour)
+{
+	int error;
+
+	dprintk("--> nfs4_init_server()\n");
+
+	/* Initialise the client representation from the mount data */
+	server->flags = data->flags & NFS_MOUNT_FLAGMASK;
+	server->caps |= NFS_CAP_ATOMIC_OPEN;
+
+	if (data->rsize)
+		server->rsize = nfs_block_size(data->rsize, NULL);
+	if (data->wsize)
+		server->wsize = nfs_block_size(data->wsize, NULL);
+
+	server->acregmin = data->acregmin * HZ;
+	server->acregmax = data->acregmax * HZ;
+	server->acdirmin = data->acdirmin * HZ;
+	server->acdirmax = data->acdirmax * HZ;
+
+	error = nfs_init_server_rpcclient(server, authflavour);
+
+	/* Done */
+	dprintk("<-- nfs4_init_server() = %d\n", error);
+	return error;
+}
+
+/*
+ * Create a version 4 volume record
+ * - keyed on server and FSID
+ */
+struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data,
+				      const char *hostname,
+				      const struct sockaddr_in *addr,
+				      const char *mntpath,
+				      const char *ip_addr,
+				      rpc_authflavor_t authflavour,
+				      struct nfs_fh *mntfh)
+{
+	struct nfs_fattr fattr;
+	struct nfs_server *server;
+	int error;
+
+	dprintk("--> nfs4_create_server()\n");
+
+	server = nfs_alloc_server();
+	if (!server)
+		return ERR_PTR(-ENOMEM);
+
+	/* Get a client record */
+	error = nfs4_set_client(server, hostname, addr, authflavour,
+			data->proto, data->timeo, data->retrans);
+	if (error < 0)
+		goto error;
+
+	/* set up the general RPC client */
+	error = nfs4_init_server(server, data, authflavour);
+	if (error < 0)
+		goto error;
+
+	BUG_ON(!server->nfs_client);
+	BUG_ON(!server->nfs_client->rpc_ops);
+	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
+
+	/* Probe the root fh to retrieve its FSID */
+	error = nfs4_path_walk(server, mntfh, mntpath);
+	if (error < 0)
+		goto error;
+
+	dprintk("Server FSID: %llx:%llx\n", server->fsid.major, server->fsid.minor);
+	dprintk("Mount FH: %d\n", mntfh->size);
+
+	error = nfs_probe_fsinfo(server, mntfh, &fattr);
+	if (error < 0)
+		goto error;
+
+	BUG_ON(!server->nfs_client);
+	BUG_ON(!server->nfs_client->rpc_ops);
+	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
+
+	spin_lock(&nfs_client_lock);
+	list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
+	list_add_tail(&server->master_link, &nfs_volume_list);
+	spin_unlock(&nfs_client_lock);
+
+	server->mount_time = jiffies;
+	dprintk("<-- nfs4_create_server() = %p\n", server);
+	return server;
+
+error:
+	nfs_free_server(server);
+	dprintk("<-- nfs4_create_server() = error %d\n", error);
+	return ERR_PTR(error);
+}
+
+/*
+ * Create an NFS4 referral server record
+ */
+struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
+					       struct nfs_fh *fh)
+{
+	struct nfs_client *parent_client;
+	struct nfs_server *server, *parent_server;
+	struct nfs_fattr fattr;
+	int error;
+
+	dprintk("--> nfs4_create_referral_server()\n");
+
+	server = nfs_alloc_server();
+	if (!server)
+		return ERR_PTR(-ENOMEM);
+
+	parent_server = NFS_SB(data->sb);
+	parent_client = parent_server->nfs_client;
+
+	/* Get a client representation.
+	 * Note: NFSv4 always uses TCP, */
+	error = nfs4_set_client(server, data->hostname, data->addr,
+			data->authflavor,
+			parent_server->client->cl_xprt->prot,
+			parent_client->retrans_timeo,
+			parent_client->retrans_count);
+
+	/* Initialise the client representation from the parent server */
+	nfs_server_copy_userdata(server, parent_server);
+	server->caps |= NFS_CAP_ATOMIC_OPEN;
+
+	error = nfs_init_server_rpcclient(server, data->authflavor);
+	if (error < 0)
+		goto error;
+
+	BUG_ON(!server->nfs_client);
+	BUG_ON(!server->nfs_client->rpc_ops);
+	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
+
+	/* probe the filesystem info for this server filesystem */
+	error = nfs_probe_fsinfo(server, fh, &fattr);
+	if (error < 0)
+		goto error;
+
+	dprintk("Referral FSID: %llx:%llx\n",
+		server->fsid.major, server->fsid.minor);
+
+	spin_lock(&nfs_client_lock);
+	list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
+	list_add_tail(&server->master_link, &nfs_volume_list);
+	spin_unlock(&nfs_client_lock);
+
+	server->mount_time = jiffies;
+
+	dprintk("<-- nfs_create_referral_server() = %p\n", server);
+	return server;
+
+error:
+	nfs_free_server(server);
+	dprintk("<-- nfs4_create_referral_server() = error %d\n", error);
+	return ERR_PTR(error);
+}
+
+#endif /* CONFIG_NFS_V4 */
+
+/*
+ * Clone an NFS2, NFS3 or NFS4 server record
+ */
+struct nfs_server *nfs_clone_server(struct nfs_server *source,
+				    struct nfs_fh *fh,
+				    struct nfs_fattr *fattr)
+{
+	struct nfs_server *server;
+	struct nfs_fattr fattr_fsinfo;
+	int error;
+
+	dprintk("--> nfs_clone_server(,%llx:%llx,)\n",
+		fattr->fsid.major, fattr->fsid.minor);
+
+	server = nfs_alloc_server();
+	if (!server)
+		return ERR_PTR(-ENOMEM);
+
+	/* Copy data from the source */
+	server->nfs_client = source->nfs_client;
+	atomic_inc(&server->nfs_client->cl_count);
+	nfs_server_copy_userdata(server, source);
+
+	server->fsid = fattr->fsid;
+
+	error = nfs_init_server_rpcclient(server, source->client->cl_auth->au_flavor);
+	if (error < 0)
+		goto out_free_server;
+	if (!IS_ERR(source->client_acl))
+		nfs_init_server_aclclient(server);
+
+	/* probe the filesystem info for this server filesystem */
+	error = nfs_probe_fsinfo(server, fh, &fattr_fsinfo);
+	if (error < 0)
+		goto out_free_server;
+
+	dprintk("Cloned FSID: %llx:%llx\n",
+		server->fsid.major, server->fsid.minor);
+
+	error = nfs_start_lockd(server);
+	if (error < 0)
+		goto out_free_server;
+
+	spin_lock(&nfs_client_lock);
+	list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks);
+	list_add_tail(&server->master_link, &nfs_volume_list);
+	spin_unlock(&nfs_client_lock);
+
+	server->mount_time = jiffies;
+
+	dprintk("<-- nfs_clone_server() = %p\n", server);
+	return server;
+
+out_free_server:
+	nfs_free_server(server);
+	dprintk("<-- nfs_clone_server() = error %d\n", error);
+	return ERR_PTR(error);
+}
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 19362712452f..9b496ef4abea 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -31,6 +31,7 @@
 #include <linux/pagemap.h>
 #include <linux/smp_lock.h>
 #include <linux/namei.h>
+#include <linux/mount.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -870,14 +871,14 @@ int nfs_is_exclusive_create(struct inode *dir, struct nameidata *nd)
 	return (nd->intent.open.flags & O_EXCL) != 0;
 }
 
-static inline int nfs_reval_fsid(struct inode *dir,
-		struct nfs_fh *fh, struct nfs_fattr *fattr)
+static inline int nfs_reval_fsid(struct vfsmount *mnt, struct inode *dir,
+				 struct nfs_fh *fh, struct nfs_fattr *fattr)
 {
 	struct nfs_server *server = NFS_SERVER(dir);
 
 	if (!nfs_fsid_equal(&server->fsid, &fattr->fsid))
 		/* Revalidate fsid on root dir */
-		return __nfs_revalidate_inode(server, dir->i_sb->s_root->d_inode);
+		return __nfs_revalidate_inode(server, mnt->mnt_root->d_inode);
 	return 0;
 }
 
@@ -913,7 +914,7 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
 		res = ERR_PTR(error);
 		goto out_unlock;
 	}
-	error = nfs_reval_fsid(dir, &fhandle, &fattr);
+	error = nfs_reval_fsid(nd->mnt, dir, &fhandle, &fattr);
 	if (error < 0) {
 		res = ERR_PTR(error);
 		goto out_unlock;
@@ -922,8 +923,9 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
 	res = (struct dentry *)inode;
 	if (IS_ERR(res))
 		goto out_unlock;
+
 no_entry:
-	res = d_add_unique(dentry, inode);
+	res = d_materialise_unique(dentry, inode);
 	if (res != NULL)
 		dentry = res;
 	nfs_renew_times(dentry);
@@ -1117,11 +1119,13 @@ static struct dentry *nfs_readdir_lookup(nfs_readdir_descriptor_t *desc)
 		dput(dentry);
 		return NULL;
 	}
-	alias = d_add_unique(dentry, inode);
+
+	alias = d_materialise_unique(dentry, inode);
 	if (alias != NULL) {
 		dput(dentry);
 		dentry = alias;
 	}
+
 	nfs_renew_times(dentry);
 	nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
 	return dentry;
diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c
new file mode 100644
index 000000000000..977e59088eeb
--- /dev/null
+++ b/fs/nfs/getroot.c
@@ -0,0 +1,306 @@
+/* getroot.c: get the root dentry for an NFS mount
+ *
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <linux/time.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/stat.h>
+#include <linux/errno.h>
+#include <linux/unistd.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/stats.h>
+#include <linux/nfs_fs.h>
+#include <linux/nfs_mount.h>
+#include <linux/nfs4_mount.h>
+#include <linux/lockd/bind.h>
+#include <linux/smp_lock.h>
+#include <linux/seq_file.h>
+#include <linux/mount.h>
+#include <linux/nfs_idmap.h>
+#include <linux/vfs.h>
+#include <linux/namei.h>
+#include <linux/namespace.h>
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+
+#include "nfs4_fs.h"
+#include "delegation.h"
+#include "internal.h"
+
+#define NFSDBG_FACILITY		NFSDBG_CLIENT
+#define NFS_PARANOIA 1
+
+/*
+ * get an NFS2/NFS3 root dentry from the root filehandle
+ */
+struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh)
+{
+	struct nfs_server *server = NFS_SB(sb);
+	struct nfs_fsinfo fsinfo;
+	struct nfs_fattr fattr;
+	struct dentry *mntroot;
+	struct inode *inode;
+	int error;
+
+	/* create a dummy root dentry with dummy inode for this superblock */
+	if (!sb->s_root) {
+		struct nfs_fh dummyfh;
+		struct dentry *root;
+		struct inode *iroot;
+
+		memset(&dummyfh, 0, sizeof(dummyfh));
+		memset(&fattr, 0, sizeof(fattr));
+		nfs_fattr_init(&fattr);
+		fattr.valid = NFS_ATTR_FATTR;
+		fattr.type = NFDIR;
+		fattr.mode = S_IFDIR | S_IRUSR | S_IWUSR;
+		fattr.nlink = 2;
+
+		iroot = nfs_fhget(sb, &dummyfh, &fattr);
+		if (IS_ERR(iroot))
+			return ERR_PTR(PTR_ERR(iroot));
+
+		root = d_alloc_root(iroot);
+		if (!root) {
+			iput(iroot);
+			return ERR_PTR(-ENOMEM);
+		}
+
+		sb->s_root = root;
+	}
+
+	/* get the actual root for this mount */
+	fsinfo.fattr = &fattr;
+
+	error = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
+	if (error < 0) {
+		dprintk("nfs_get_root: getattr error = %d\n", -error);
+		return ERR_PTR(error);
+	}
+
+	inode = nfs_fhget(sb, mntfh, fsinfo.fattr);
+	if (IS_ERR(inode)) {
+		dprintk("nfs_get_root: get root inode failed\n");
+		return ERR_PTR(PTR_ERR(inode));
+	}
+
+	/* root dentries normally start off anonymous and get spliced in later
+	 * if the dentry tree reaches them; however if the dentry already
+	 * exists, we'll pick it up at this point and use it as the root
+	 */
+	mntroot = d_alloc_anon(inode);
+	if (!mntroot) {
+		iput(inode);
+		dprintk("nfs_get_root: get root dentry failed\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	if (!mntroot->d_op)
+		mntroot->d_op = server->nfs_client->rpc_ops->dentry_ops;
+
+	return mntroot;
+}
+
+#ifdef CONFIG_NFS_V4
+
+/*
+ * Do a simple pathwalk from the root FH of the server to the nominated target
+ * of the mountpoint
+ * - give error on symlinks
+ * - give error on ".." occurring in the path
+ * - follow traversals
+ */
+int nfs4_path_walk(struct nfs_server *server,
+		   struct nfs_fh *mntfh,
+		   const char *path)
+{
+	struct nfs_fsinfo fsinfo;
+	struct nfs_fattr fattr;
+	struct nfs_fh lastfh;
+	struct qstr name;
+	int ret;
+	//int referral_count = 0;
+
+	dprintk("--> nfs4_path_walk(,,%s)\n", path);
+
+	fsinfo.fattr = &fattr;
+	nfs_fattr_init(&fattr);
+
+	if (*path++ != '/') {
+		dprintk("nfs4_get_root: Path does not begin with a slash\n");
+		return -EINVAL;
+	}
+
+	/* Start by getting the root filehandle from the server */
+	ret = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
+	if (ret < 0) {
+		dprintk("nfs4_get_root: getroot error = %d\n", -ret);
+		return ret;
+	}
+
+	if (fattr.type != NFDIR) {
+		printk(KERN_ERR "nfs4_get_root:"
+		       " getroot encountered non-directory\n");
+		return -ENOTDIR;
+	}
+
+	if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) {
+		printk(KERN_ERR "nfs4_get_root:"
+		       " getroot obtained referral\n");
+		return -EREMOTE;
+	}
+
+next_component:
+	dprintk("Next: %s\n", path);
+
+	/* extract the next bit of the path */
+	if (!*path)
+		goto path_walk_complete;
+
+	name.name = path;
+	while (*path && *path != '/')
+		path++;
+	name.len = path - (const char *) name.name;
+
+eat_dot_dir:
+	while (*path == '/')
+		path++;
+
+	if (path[0] == '.' && (path[1] == '/' || !path[1])) {
+		path += 2;
+		goto eat_dot_dir;
+	}
+
+	if (path[0] == '.' && path[1] == '.' && (path[2] == '/' || !path[2])
+	    ) {
+		printk(KERN_ERR "nfs4_get_root:"
+		       " Mount path contains reference to \"..\"\n");
+		return -EINVAL;
+	}
+
+	/* lookup the next FH in the sequence */
+	memcpy(&lastfh, mntfh, sizeof(lastfh));
+
+	dprintk("LookupFH: %*.*s [%s]\n", name.len, name.len, name.name, path);
+
+	ret = server->nfs_client->rpc_ops->lookupfh(server, &lastfh, &name,
+						    mntfh, &fattr);
+	if (ret < 0) {
+		dprintk("nfs4_get_root: getroot error = %d\n", -ret);
+		return ret;
+	}
+
+	if (fattr.type != NFDIR) {
+		printk(KERN_ERR "nfs4_get_root:"
+		       " lookupfh encountered non-directory\n");
+		return -ENOTDIR;
+	}
+
+	if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) {
+		printk(KERN_ERR "nfs4_get_root:"
+		       " lookupfh obtained referral\n");
+		return -EREMOTE;
+	}
+
+	goto next_component;
+
+path_walk_complete:
+	memcpy(&server->fsid, &fattr.fsid, sizeof(server->fsid));
+	dprintk("<-- nfs4_path_walk() = 0\n");
+	return 0;
+}
+
+/*
+ * get an NFS4 root dentry from the root filehandle
+ */
+struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh)
+{
+	struct nfs_server *server = NFS_SB(sb);
+	struct nfs_fattr fattr;
+	struct dentry *mntroot;
+	struct inode *inode;
+	int error;
+
+	dprintk("--> nfs4_get_root()\n");
+
+	/* create a dummy root dentry with dummy inode for this superblock */
+	if (!sb->s_root) {
+		struct nfs_fh dummyfh;
+		struct dentry *root;
+		struct inode *iroot;
+
+		memset(&dummyfh, 0, sizeof(dummyfh));
+		memset(&fattr, 0, sizeof(fattr));
+		nfs_fattr_init(&fattr);
+		fattr.valid = NFS_ATTR_FATTR;
+		fattr.type = NFDIR;
+		fattr.mode = S_IFDIR | S_IRUSR | S_IWUSR;
+		fattr.nlink = 2;
+
+		iroot = nfs_fhget(sb, &dummyfh, &fattr);
+		if (IS_ERR(iroot))
+			return ERR_PTR(PTR_ERR(iroot));
+
+		root = d_alloc_root(iroot);
+		if (!root) {
+			iput(iroot);
+			return ERR_PTR(-ENOMEM);
+		}
+
+		sb->s_root = root;
+	}
+
+	/* get the info about the server and filesystem */
+	error = nfs4_server_capabilities(server, mntfh);
+	if (error < 0) {
+		dprintk("nfs_get_root: getcaps error = %d\n",
+			-error);
+		return ERR_PTR(error);
+	}
+
+	/* get the actual root for this mount */
+	error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr);
+	if (error < 0) {
+		dprintk("nfs_get_root: getattr error = %d\n", -error);
+		return ERR_PTR(error);
+	}
+
+	inode = nfs_fhget(sb, mntfh, &fattr);
+	if (IS_ERR(inode)) {
+		dprintk("nfs_get_root: get root inode failed\n");
+		return ERR_PTR(PTR_ERR(inode));
+	}
+
+	/* root dentries normally start off anonymous and get spliced in later
+	 * if the dentry tree reaches them; however if the dentry already
+	 * exists, we'll pick it up at this point and use it as the root
+	 */
+	mntroot = d_alloc_anon(inode);
+	if (!mntroot) {
+		iput(inode);
+		dprintk("nfs_get_root: get root dentry failed\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	if (!mntroot->d_op)
+		mntroot->d_op = server->nfs_client->rpc_ops->dentry_ops;
+
+	dprintk("<-- nfs4_get_root()\n");
+	return mntroot;
+}
+
+#endif /* CONFIG_NFS_V4 */
diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c
index 231c20ffc0ff..f96dfac7dc9a 100644
--- a/fs/nfs/idmap.c
+++ b/fs/nfs/idmap.c
@@ -114,8 +114,7 @@ nfs_idmap_new(struct nfs_client *clp)
 	struct idmap *idmap;
 	int error;
 
-	if (clp->cl_idmap != NULL)
-		return 0;
+	BUG_ON(clp->cl_idmap != NULL);
 
         if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL)
                 return -ENOMEM;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 771c3b833757..a547c58a83e6 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1020,7 +1020,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
  out_fileid:
 	printk(KERN_ERR "NFS: server %s error: fileid changed\n"
 		"fsid %s: expected fileid 0x%Lx, got 0x%Lx\n",
-		NFS_SERVER(inode)->hostname, inode->i_sb->s_id,
+		NFS_SERVER(inode)->nfs_client->cl_hostname, inode->i_sb->s_id,
 		(long long)nfsi->fileid, (long long)fattr->fileid);
 	goto out_err;
 }
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 2f3aa52fbefc..e73ba4f1052a 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -4,6 +4,18 @@
 
 #include <linux/mount.h>
 
+struct nfs_string;
+struct nfs_mount_data;
+struct nfs4_mount_data;
+
+/* Maximum number of readahead requests
+ * FIXME: this should really be a sysctl so that users may tune it to suit
+ *        their needs. People that do NFS over a slow network, might for
+ *        instance want to reduce it to something closer to 1 for improved
+ *        interactive response.
+ */
+#define NFS_MAX_READAHEAD	(RPC_DEF_SLOT_TABLE - 1)
+
 struct nfs_clone_mount {
 	const struct super_block *sb;
 	const struct dentry *dentry;
@@ -16,12 +28,25 @@ struct nfs_clone_mount {
 };
 
 /* client.c */
+extern struct rpc_program nfs_program;
+
 extern void nfs_put_client(struct nfs_client *);
 extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int);
-extern struct nfs_client *nfs_get_client(const char *, const struct sockaddr_in *, int);
-extern void nfs_mark_client_ready(struct nfs_client *, int);
-extern int nfs_create_rpc_client(struct nfs_client *, int, unsigned int,
-				 unsigned int, rpc_authflavor_t);
+extern struct nfs_server *nfs_create_server(const struct nfs_mount_data *,
+					    struct nfs_fh *);
+extern struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *,
+					     const char *,
+					     const struct sockaddr_in *,
+					     const char *,
+					     const char *,
+					     rpc_authflavor_t,
+					     struct nfs_fh *);
+extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *,
+						      struct nfs_fh *);
+extern void nfs_free_server(struct nfs_server *server);
+extern struct nfs_server *nfs_clone_server(struct nfs_server *,
+					   struct nfs_fh *,
+					   struct nfs_fattr *);
 
 /* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
@@ -89,10 +114,10 @@ extern void nfs4_clear_inode(struct inode *);
 #endif
 
 /* super.c */
-extern struct file_system_type nfs_referral_nfs4_fs_type;
-extern struct file_system_type clone_nfs_fs_type;
+extern struct file_system_type nfs_xdev_fs_type;
 #ifdef CONFIG_NFS_V4
-extern struct file_system_type clone_nfs4_fs_type;
+extern struct file_system_type nfs4_xdev_fs_type;
+extern struct file_system_type nfs4_referral_fs_type;
 #endif
 
 extern struct rpc_stat nfs_rpcstat;
@@ -101,28 +126,30 @@ extern int __init register_nfs_fs(void);
 extern void __exit unregister_nfs_fs(void);
 
 /* namespace.c */
-extern char *nfs_path(const char *base, const struct dentry *dentry,
+extern char *nfs_path(const char *base,
+		      const struct dentry *droot,
+		      const struct dentry *dentry,
 		      char *buffer, ssize_t buflen);
 
-/*
- * Determine the mount path as a string
- */
+/* getroot.c */
+extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *);
 #ifdef CONFIG_NFS_V4
-static inline char *
-nfs4_path(const struct dentry *dentry, char *buffer, ssize_t buflen)
-{
-	return nfs_path(NFS_SB(dentry->d_sb)->mnt_path, dentry, buffer, buflen);
-}
+extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *);
+
+extern int nfs4_path_walk(struct nfs_server *server,
+			  struct nfs_fh *mntfh,
+			  const char *path);
 #endif
 
 /*
  * Determine the device name as a string
  */
 static inline char *nfs_devname(const struct vfsmount *mnt_parent,
-			 const struct dentry *dentry,
-			 char *buffer, ssize_t buflen)
+				const struct dentry *dentry,
+				char *buffer, ssize_t buflen)
 {
-	return nfs_path(mnt_parent->mnt_devname, dentry, buffer, buflen);
+	return nfs_path(mnt_parent->mnt_devname, mnt_parent->mnt_root,
+			dentry, buffer, buflen);
 }
 
 /*
@@ -178,20 +205,3 @@ void nfs_super_set_maxbytes(struct super_block *sb, __u64 maxfilesize)
 	if (sb->s_maxbytes > MAX_LFS_FILESIZE || sb->s_maxbytes <= 0)
 		sb->s_maxbytes = MAX_LFS_FILESIZE;
 }
-
-/*
- * Check if the string represents a "valid" IPv4 address
- */
-static inline int valid_ipaddr4(const char *buf)
-{
-	int rc, count, in[4];
-
-	rc = sscanf(buf, "%d.%d.%d.%d", &in[0], &in[1], &in[2], &in[3]);
-	if (rc != 4)
-		return -EINVAL;
-	for (count = 0; count < 4; count++) {
-		if (in[count] > 255)
-			return -EINVAL;
-	}
-	return 0;
-}
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index d8b8d56266cb..77b00684894d 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -2,6 +2,7 @@
  * linux/fs/nfs/namespace.c
  *
  * Copyright (C) 2005 Trond Myklebust <Trond.Myklebust@netapp.com>
+ * - Modified by David Howells <dhowells@redhat.com>
  *
  * NFS namespace
  */
@@ -28,6 +29,7 @@ int nfs_mountpoint_expiry_timeout = 500 * HZ;
 /*
  * nfs_path - reconstruct the path given an arbitrary dentry
  * @base - arbitrary string to prepend to the path
+ * @droot - pointer to root dentry for mountpoint
  * @dentry - pointer to dentry
  * @buffer - result buffer
  * @buflen - length of buffer
@@ -38,7 +40,9 @@ int nfs_mountpoint_expiry_timeout = 500 * HZ;
  * This is mainly for use in figuring out the path on the
  * server side when automounting on top of an existing partition.
  */
-char *nfs_path(const char *base, const struct dentry *dentry,
+char *nfs_path(const char *base,
+	       const struct dentry *droot,
+	       const struct dentry *dentry,
 	       char *buffer, ssize_t buflen)
 {
 	char *end = buffer+buflen;
@@ -47,7 +51,7 @@ char *nfs_path(const char *base, const struct dentry *dentry,
 	*--end = '\0';
 	buflen--;
 	spin_lock(&dcache_lock);
-	while (!IS_ROOT(dentry)) {
+	while (!IS_ROOT(dentry) && dentry != droot) {
 		namelen = dentry->d_name.len;
 		buflen -= namelen + 1;
 		if (buflen < 0)
@@ -96,12 +100,13 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	struct nfs_fattr fattr;
 	int err;
 
+	dprintk("--> nfs_follow_mountpoint()\n");
+
 	BUG_ON(IS_ROOT(dentry));
 	dprintk("%s: enter\n", __FUNCTION__);
 	dput(nd->dentry);
 	nd->dentry = dget(dentry);
-	if (d_mountpoint(nd->dentry))
-		goto out_follow;
+
 	/* Look it up again */
 	parent = dget_parent(nd->dentry);
 	err = server->nfs_client->rpc_ops->lookup(parent->d_inode,
@@ -134,6 +139,8 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
 out:
 	dprintk("%s: done, returned %d\n", __FUNCTION__, err);
+
+	dprintk("<-- nfs_follow_mountpoint() = %d\n", err);
 	return ERR_PTR(err);
 out_err:
 	path_release(nd);
@@ -183,14 +190,14 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
 	switch (server->nfs_client->cl_nfsversion) {
 		case 2:
 		case 3:
-			mnt = vfs_kern_mount(&clone_nfs_fs_type, 0, devname, mountdata);
+			mnt = vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata);
 			break;
 		case 4:
-			mnt = vfs_kern_mount(&clone_nfs4_fs_type, 0, devname, mountdata);
+			mnt = vfs_kern_mount(&nfs4_xdev_fs_type, 0, devname, mountdata);
 	}
 	return mnt;
 #else
-	return vfs_kern_mount(&clone_nfs_fs_type, 0, devname, mountdata);
+	return vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata);
 #endif
 }
 
@@ -216,6 +223,8 @@ struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
 	char *page = (char *) __get_free_page(GFP_USER);
 	char *devname;
 
+	dprintk("--> nfs_do_submount()\n");
+
 	dprintk("%s: submounting on %s/%s\n", __FUNCTION__,
 			dentry->d_parent->d_name.name,
 			dentry->d_name.name);
@@ -230,5 +239,7 @@ free_page:
 	free_page((unsigned long)page);
 out:
 	dprintk("%s: done\n", __FUNCTION__);
+
+	dprintk("<-- nfs_do_submount() = %p\n", mnt);
 	return mnt;
 }
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 0622af0122be..9e8258ece6fd 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -81,7 +81,7 @@ do_proc_get_root(struct rpc_clnt *client, struct nfs_fh *fhandle,
 }
 
 /*
- * Bare-bones access to getattr: this is for nfs_read_super.
+ * Bare-bones access to getattr: this is for nfs_get_root/nfs_get_sb
  */
 static int
 nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index e7879245361e..61095fe4b5ca 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -188,8 +188,6 @@ extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(void *);
 
 /* nfs4state.c */
-extern void init_nfsv4_state(struct nfs_server *);
-extern void destroy_nfsv4_state(struct nfs_server *);
 struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp);
 extern u32 nfs4_alloc_lockowner_id(struct nfs_client *);
 
@@ -224,10 +222,6 @@ extern struct svc_version nfs4_callback_version1;
 
 #else
 
-#define init_nfsv4_state(server)  do { } while (0)
-#define destroy_nfsv4_state(server)       do { } while (0)
-#define nfs4_put_state_owner(inode, owner) do { } while (0)
-#define nfs4_put_open_state(state) do { } while (0)
 #define nfs4_close_state(a, b) do { } while (0)
 
 #endif /* CONFIG_NFS_V4 */
diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c
index faed9bcba50f..24e47f3bbd17 100644
--- a/fs/nfs/nfs4namespace.c
+++ b/fs/nfs/nfs4namespace.c
@@ -2,6 +2,7 @@
  * linux/fs/nfs/nfs4namespace.c
  *
  * Copyright (C) 2005 Trond Myklebust <Trond.Myklebust@netapp.com>
+ * - Modified by David Howells <dhowells@redhat.com>
  *
  * NFSv4 namespace
  */
@@ -47,6 +48,68 @@ Elong:
 	return ERR_PTR(-ENAMETOOLONG);
 }
 
+/*
+ * Determine the mount path as a string
+ */
+static char *nfs4_path(const struct vfsmount *mnt_parent,
+		       const struct dentry *dentry,
+		       char *buffer, ssize_t buflen)
+{
+	const char *srvpath;
+
+	srvpath = strchr(mnt_parent->mnt_devname, ':');
+	if (srvpath)
+		srvpath++;
+	else
+		srvpath = mnt_parent->mnt_devname;
+
+	return nfs_path(srvpath, mnt_parent->mnt_root, dentry, buffer, buflen);
+}
+
+/*
+ * Check that fs_locations::fs_root [RFC3530 6.3] is a prefix for what we
+ * believe to be the server path to this dentry
+ */
+static int nfs4_validate_fspath(const struct vfsmount *mnt_parent,
+				const struct dentry *dentry,
+				const struct nfs4_fs_locations *locations,
+				char *page, char *page2)
+{
+	const char *path, *fs_path;
+
+	path = nfs4_path(mnt_parent, dentry, page, PAGE_SIZE);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	fs_path = nfs4_pathname_string(&locations->fs_path, page2, PAGE_SIZE);
+	if (IS_ERR(fs_path))
+		return PTR_ERR(fs_path);
+
+	if (strncmp(path, fs_path, strlen(fs_path)) != 0) {
+		dprintk("%s: path %s does not begin with fsroot %s\n",
+			__FUNCTION__, path, fs_path);
+		return -ENOENT;
+	}
+
+	return 0;
+}
+
+/*
+ * Check if the string represents a "valid" IPv4 address
+ */
+static inline int valid_ipaddr4(const char *buf)
+{
+	int rc, count, in[4];
+
+	rc = sscanf(buf, "%d.%d.%d.%d", &in[0], &in[1], &in[2], &in[3]);
+	if (rc != 4)
+		return -EINVAL;
+	for (count = 0; count < 4; count++) {
+		if (in[count] > 255)
+			return -EINVAL;
+	}
+	return 0;
+}
 
 /**
  * nfs_follow_referral - set up mountpoint when hitting a referral on moved error
@@ -68,10 +131,9 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
 		.dentry = dentry,
 		.authflavor = NFS_SB(mnt_parent->mnt_sb)->client->cl_auth->au_flavor,
 	};
-	char *page, *page2;
-	char *path, *fs_path;
+	char *page = NULL, *page2 = NULL;
 	char *devname;
-	int loc, s;
+	int loc, s, error;
 
 	if (locations == NULL || locations->nlocations <= 0)
 		goto out;
@@ -79,31 +141,25 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
 	dprintk("%s: referral at %s/%s\n", __FUNCTION__,
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
-	/* Ensure fs path is a prefix of current dentry path */
 	page = (char *) __get_free_page(GFP_USER);
-	if (page == NULL)
+	if (!page)
 		goto out;
+
 	page2 = (char *) __get_free_page(GFP_USER);
-	if (page2 == NULL)
+	if (!page2)
 		goto out;
 
-	path = nfs4_path(dentry, page, PAGE_SIZE);
-	if (IS_ERR(path))
-		goto out_free;
-
-	fs_path = nfs4_pathname_string(&locations->fs_path, page2, PAGE_SIZE);
-	if (IS_ERR(fs_path))
-		goto out_free;
-
-	if (strncmp(path, fs_path, strlen(fs_path)) != 0) {
-		dprintk("%s: path %s does not begin with fsroot %s\n", __FUNCTION__, path, fs_path);
-		goto out_free;
+	/* Ensure fs path is a prefix of current dentry path */
+	error = nfs4_validate_fspath(mnt_parent, dentry, locations, page, page2);
+	if (error < 0) {
+		mnt = ERR_PTR(error);
+		goto out;
 	}
 
 	devname = nfs_devname(mnt_parent, dentry, page, PAGE_SIZE);
 	if (IS_ERR(devname)) {
 		mnt = (struct vfsmount *)devname;
-		goto out_free;
+		goto out;
 	}
 
 	loc = 0;
@@ -140,7 +196,7 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
 			addr.sin_port = htons(NFS_PORT);
 			mountdata.addr = &addr;
 
-			mnt = vfs_kern_mount(&nfs_referral_nfs4_fs_type, 0, devname, &mountdata);
+			mnt = vfs_kern_mount(&nfs4_referral_fs_type, 0, devname, &mountdata);
 			if (!IS_ERR(mnt)) {
 				break;
 			}
@@ -149,10 +205,9 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
 		loc++;
 	}
 
-out_free:
-	free_page((unsigned long)page);
-	free_page((unsigned long)page2);
 out:
+	free_page((unsigned long) page);
+	free_page((unsigned long) page2);
 	dprintk("%s: done\n", __FUNCTION__);
 	return mnt;
 }
@@ -165,7 +220,7 @@ out:
  */
 struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry)
 {
-	struct vfsmount *mnt = ERR_PTR(-ENOENT);
+	struct vfsmount *mnt = ERR_PTR(-ENOMEM);
 	struct dentry *parent;
 	struct nfs4_fs_locations *fs_locations = NULL;
 	struct page *page;
@@ -183,11 +238,16 @@ struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentr
 		goto out_free;
 
 	/* Get locations */
+	mnt = ERR_PTR(-ENOENT);
+
 	parent = dget_parent(dentry);
-	dprintk("%s: getting locations for %s/%s\n", __FUNCTION__, parent->d_name.name, dentry->d_name.name);
+	dprintk("%s: getting locations for %s/%s\n",
+		__FUNCTION__, parent->d_name.name, dentry->d_name.name);
+
 	err = nfs4_proc_fs_locations(parent->d_inode, dentry, fs_locations, page);
 	dput(parent);
-	if (err != 0 || fs_locations->nlocations <= 0 ||
+	if (err != 0 ||
+	    fs_locations->nlocations <= 0 ||
 	    fs_locations->fs_path.ncomponents <= 0)
 		goto out_free;
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 1573eeb07ce1..a825547e8214 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1393,70 +1393,19 @@ static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 	return err;
 }
 
+/*
+ * get the file handle for the "/" directory on the server
+ */
 static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
-		struct nfs_fsinfo *info)
+			      struct nfs_fsinfo *info)
 {
-	struct nfs_fattr *	fattr = info->fattr;
-	unsigned char *		p;
-	struct qstr		q;
-	struct nfs4_lookup_arg args = {
-		.dir_fh = fhandle,
-		.name = &q,
-		.bitmask = nfs4_fattr_bitmap,
-	};
-	struct nfs4_lookup_res res = {
-		.server = server,
-		.fattr = fattr,
-		.fh = fhandle,
-	};
-	struct rpc_message msg = {
-		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP],
-		.rpc_argp = &args,
-		.rpc_resp = &res,
-	};
 	int status;
 
-	/*
-	 * Now we do a separate LOOKUP for each component of the mount path.
-	 * The LOOKUPs are done separately so that we can conveniently
-	 * catch an ERR_WRONGSEC if it occurs along the way...
-	 */
 	status = nfs4_lookup_root(server, fhandle, info);
-	if (status)
-		goto out;
-
-	p = server->mnt_path;
-	for (;;) {
-		struct nfs4_exception exception = { };
-
-		while (*p == '/')
-			p++;
-		if (!*p)
-			break;
-		q.name = p;
-		while (*p && (*p != '/'))
-			p++;
-		q.len = p - q.name;
-
-		do {
-			nfs_fattr_init(fattr);
-			status = nfs4_handle_exception(server,
-					rpc_call_sync(server->client, &msg, 0),
-					&exception);
-		} while (exception.retry);
-		if (status == 0)
-			continue;
-		if (status == -ENOENT) {
-			printk(KERN_NOTICE "NFS: mount path %s does not exist!\n", server->mnt_path);
-			printk(KERN_NOTICE "NFS: suggestion: try mounting '/' instead.\n");
-		}
-		break;
-	}
 	if (status == 0)
 		status = nfs4_server_capabilities(server, fhandle);
 	if (status == 0)
 		status = nfs4_do_fsinfo(server, fhandle, info);
-out:
 	return nfs4_map_errors(status);
 }
 
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index ff947ecb8b81..f2c893690ac4 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -127,26 +127,13 @@ nfs4_schedule_state_renewal(struct nfs_client *clp)
 void
 nfs4_renewd_prepare_shutdown(struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs_client;
-
-	if (!clp)
-		return;
 	flush_scheduled_work();
-	down_write(&clp->cl_sem);
-	if (!list_empty(&server->nfs4_siblings))
-		list_del_init(&server->nfs4_siblings);
-	up_write(&clp->cl_sem);
 }
 
-/* Must be called with clp->cl_sem locked for writes */
 void
 nfs4_kill_renewd(struct nfs_client *clp)
 {
 	down_read(&clp->cl_sem);
-	if (!list_empty(&clp->cl_superblocks)) {
-		up_read(&clp->cl_sem);
-		return;
-	}
 	cancel_delayed_work(&clp->cl_renewd);
 	up_read(&clp->cl_sem);
 	flush_scheduled_work();
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 058811e39555..5fffbdfa971f 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -58,24 +58,6 @@ const nfs4_stateid zero_stateid;
 
 static LIST_HEAD(nfs4_clientid_list);
 
-void
-init_nfsv4_state(struct nfs_server *server)
-{
-	server->nfs_client = NULL;
-	INIT_LIST_HEAD(&server->nfs4_siblings);
-}
-
-void
-destroy_nfsv4_state(struct nfs_server *server)
-{
-	kfree(server->mnt_path);
-	server->mnt_path = NULL;
-	if (server->nfs_client) {
-		nfs_put_client(server->nfs_client);
-		server->nfs_client = NULL;
-	}
-}
-
 static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	int status = nfs4_proc_setclientid(clp, NFS4_CALLBACK,
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index f0aff824a291..dae33c1e8a77 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -171,7 +171,7 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode,
 		rdata->args.offset = page_offset(page) + rdata->args.pgbase;
 
 		dprintk("NFS: nfs_proc_read(%s, (%s/%Ld), %Lu, %u)\n",
-			NFS_SERVER(inode)->hostname,
+			NFS_SERVER(inode)->nfs_client->cl_hostname,
 			inode->i_sb->s_id,
 			(long long)NFS_FILEID(inode),
 			(unsigned long long)rdata->args.pgbase,
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 5842d510d732..867b5dcd3a40 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -13,6 +13,11 @@
  *
  *  Split from inode.c by David Howells <dhowells@redhat.com>
  *
+ * - superblocks are indexed on server only - all inodes, dentries, etc. associated with a
+ *   particular server are held in the same superblock
+ * - NFS superblocks can have several effective roots to the dentry tree
+ * - directory type roots are spliced into the tree when a path from one root reaches the root
+ *   of another (see nfs_lookup())
  */
 
 #include <linux/config.h>
@@ -52,20 +57,12 @@
 
 #define NFSDBG_FACILITY		NFSDBG_VFS
 
-/* Maximum number of readahead requests
- * FIXME: this should really be a sysctl so that users may tune it to suit
- *        their needs. People that do NFS over a slow network, might for
- *        instance want to reduce it to something closer to 1 for improved
- *        interactive response.
- */
-#define NFS_MAX_READAHEAD	(RPC_DEF_SLOT_TABLE - 1)
-
 static void nfs_umount_begin(struct vfsmount *, int);
 static int  nfs_statfs(struct dentry *, struct kstatfs *);
 static int  nfs_show_options(struct seq_file *, struct vfsmount *);
 static int  nfs_show_stats(struct seq_file *, struct vfsmount *);
 static int nfs_get_sb(struct file_system_type *, int, const char *, void *, struct vfsmount *);
-static int nfs_clone_nfs_sb(struct file_system_type *fs_type,
+static int nfs_xdev_get_sb(struct file_system_type *fs_type,
 		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
 static void nfs_kill_super(struct super_block *);
 
@@ -77,10 +74,10 @@ static struct file_system_type nfs_fs_type = {
 	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
-struct file_system_type clone_nfs_fs_type = {
+struct file_system_type nfs_xdev_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "nfs",
-	.get_sb		= nfs_clone_nfs_sb,
+	.get_sb		= nfs_xdev_get_sb,
 	.kill_sb	= nfs_kill_super,
 	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
@@ -99,10 +96,10 @@ static struct super_operations nfs_sops = {
 #ifdef CONFIG_NFS_V4
 static int nfs4_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
-static int nfs_clone_nfs4_sb(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
-static int nfs_referral_nfs4_sb(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
+static int nfs4_xdev_get_sb(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
+static int nfs4_referral_get_sb(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
 static void nfs4_kill_super(struct super_block *sb);
 
 static struct file_system_type nfs4_fs_type = {
@@ -113,18 +110,18 @@ static struct file_system_type nfs4_fs_type = {
 	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
-struct file_system_type clone_nfs4_fs_type = {
+struct file_system_type nfs4_xdev_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "nfs4",
-	.get_sb		= nfs_clone_nfs4_sb,
+	.get_sb		= nfs4_xdev_get_sb,
 	.kill_sb	= nfs4_kill_super,
 	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
-struct file_system_type nfs_referral_nfs4_fs_type = {
+struct file_system_type nfs4_referral_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "nfs4",
-	.get_sb		= nfs_referral_nfs4_sb,
+	.get_sb		= nfs4_referral_get_sb,
 	.kill_sb	= nfs4_kill_super,
 	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
@@ -345,7 +342,7 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
 	nfs_show_mount_options(m, nfss, 0);
 
 	seq_puts(m, ",addr=");
-	seq_escape(m, nfss->hostname, " \t\n\\");
+	seq_escape(m, nfss->nfs_client->cl_hostname, " \t\n\\");
 
 	return 0;
 }
@@ -429,714 +426,351 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt)
 
 /*
  * Begin unmount by attempting to remove all automounted mountpoints we added
- * in response to traversals
+ * in response to xdev traversals and referrals
  */
 static void nfs_umount_begin(struct vfsmount *vfsmnt, int flags)
 {
-	struct nfs_server *server;
-	struct rpc_clnt	*rpc;
-
 	shrink_submounts(vfsmnt, &nfs_automount_list);
-	if (!(flags & MNT_FORCE))
-		return;
-	/* -EIO all pending I/O */
-	server = NFS_SB(vfsmnt->mnt_sb);
-	rpc = server->client;
-	if (!IS_ERR(rpc))
-		rpc_killall_tasks(rpc);
-	rpc = server->client_acl;
-	if (!IS_ERR(rpc))
-		rpc_killall_tasks(rpc);
 }
 
 /*
- * Obtain the root inode of the file system.
+ * Validate the NFS2/NFS3 mount data
+ * - fills in the mount root filehandle
  */
-static struct inode *
-nfs_get_root(struct super_block *sb, struct nfs_fh *rootfh, struct nfs_fsinfo *fsinfo)
+static int nfs_validate_mount_data(struct nfs_mount_data *data,
+				   struct nfs_fh *mntfh)
 {
-	struct nfs_server	*server = NFS_SB(sb);
-	int			error;
-
-	error = server->nfs_client->rpc_ops->getroot(server, rootfh, fsinfo);
-	if (error < 0) {
-		dprintk("nfs_get_root: getattr error = %d\n", -error);
-		return ERR_PTR(error);
+	if (data == NULL) {
+		dprintk("%s: missing data argument\n", __FUNCTION__);
+		return -EINVAL;
 	}
 
-	server->fsid = fsinfo->fattr->fsid;
-	return nfs_fhget(sb, rootfh, fsinfo->fattr);
-}
-
-/*
- * Do NFS version-independent mount processing, and sanity checking
- */
-static int
-nfs_sb_init(struct super_block *sb, rpc_authflavor_t authflavor)
-{
-	struct nfs_server	*server;
-	struct inode		*root_inode;
-	struct nfs_fattr	fattr;
-	struct nfs_fsinfo	fsinfo = {
-					.fattr = &fattr,
-				};
-	struct nfs_pathconf pathinfo = {
-			.fattr = &fattr,
-	};
-	int no_root_error = 0;
-	unsigned long max_rpc_payload;
-
-	/* We probably want something more informative here */
-	snprintf(sb->s_id, sizeof(sb->s_id), "%x:%x", MAJOR(sb->s_dev), MINOR(sb->s_dev));
-
-	server = NFS_SB(sb);
-
-	sb->s_magic      = NFS_SUPER_MAGIC;
-
-	server->io_stats = nfs_alloc_iostats();
-	if (server->io_stats == NULL)
-		return -ENOMEM;
+	if (data->version <= 0 || data->version > NFS_MOUNT_VERSION) {
+		dprintk("%s: bad mount version\n", __FUNCTION__);
+		return -EINVAL;
+	}
 
-	root_inode = nfs_get_root(sb, &server->fh, &fsinfo);
-	/* Did getting the root inode fail? */
-	if (IS_ERR(root_inode)) {
-		no_root_error = PTR_ERR(root_inode);
-		goto out_no_root;
+	switch (data->version) {
+		case 1:
+			data->namlen = 0;
+		case 2:
+			data->bsize  = 0;
+		case 3:
+			if (data->flags & NFS_MOUNT_VER3) {
+				dprintk("%s: mount structure version %d does not support NFSv3\n",
+						__FUNCTION__,
+						data->version);
+				return -EINVAL;
+			}
+			data->root.size = NFS2_FHSIZE;
+			memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
+		case 4:
+			if (data->flags & NFS_MOUNT_SECFLAVOUR) {
+				dprintk("%s: mount structure version %d does not support strong security\n",
+						__FUNCTION__,
+						data->version);
+				return -EINVAL;
+			}
+			/* Fill in pseudoflavor for mount version < 5 */
+			data->pseudoflavor = RPC_AUTH_UNIX;
+		case 5:
+			memset(data->context, 0, sizeof(data->context));
 	}
-	sb->s_root = d_alloc_root(root_inode);
-	if (!sb->s_root) {
-		no_root_error = -ENOMEM;
-		goto out_no_root;
+
+#ifndef CONFIG_NFS_V3
+	/* If NFSv3 is not compiled in, return -EPROTONOSUPPORT */
+	if (data->flags & NFS_MOUNT_VER3) {
+		dprintk("%s: NFSv3 not compiled into kernel\n", __FUNCTION__);
+		return -EPROTONOSUPPORT;
 	}
-	sb->s_root->d_op = server->nfs_client->rpc_ops->dentry_ops;
-
-	/* mount time stamp, in seconds */
-	server->mount_time = jiffies;
-
-	/* Get some general file system info */
-	if (server->namelen == 0 &&
-	    server->nfs_client->rpc_ops->pathconf(server, &server->fh, &pathinfo) >= 0)
-		server->namelen = pathinfo.max_namelen;
-	/* Work out a lot of parameters */
-	if (server->rsize == 0)
-		server->rsize = nfs_block_size(fsinfo.rtpref, NULL);
-	if (server->wsize == 0)
-		server->wsize = nfs_block_size(fsinfo.wtpref, NULL);
-
-	if (fsinfo.rtmax >= 512 && server->rsize > fsinfo.rtmax)
-		server->rsize = nfs_block_size(fsinfo.rtmax, NULL);
-	if (fsinfo.wtmax >= 512 && server->wsize > fsinfo.wtmax)
-		server->wsize = nfs_block_size(fsinfo.wtmax, NULL);
-
-	max_rpc_payload = nfs_block_size(rpc_max_payload(server->client), NULL);
-	if (server->rsize > max_rpc_payload)
-		server->rsize = max_rpc_payload;
-	if (server->rsize > NFS_MAX_FILE_IO_SIZE)
-		server->rsize = NFS_MAX_FILE_IO_SIZE;
-	server->rpages = (server->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
-
-	if (server->wsize > max_rpc_payload)
-		server->wsize = max_rpc_payload;
-	if (server->wsize > NFS_MAX_FILE_IO_SIZE)
-		server->wsize = NFS_MAX_FILE_IO_SIZE;
-	server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+#endif /* CONFIG_NFS_V3 */
 
-	if (sb->s_blocksize == 0)
-		sb->s_blocksize = nfs_block_bits(server->wsize,
-							 &sb->s_blocksize_bits);
-	server->wtmult = nfs_block_bits(fsinfo.wtmult, NULL);
-
-	server->dtsize = nfs_block_size(fsinfo.dtpref, NULL);
-	if (server->dtsize > PAGE_CACHE_SIZE)
-		server->dtsize = PAGE_CACHE_SIZE;
-	if (server->dtsize > server->rsize)
-		server->dtsize = server->rsize;
-
-	if (server->flags & NFS_MOUNT_NOAC) {
-		server->acregmin = server->acregmax = 0;
-		server->acdirmin = server->acdirmax = 0;
-		sb->s_flags |= MS_SYNCHRONOUS;
+	/* We now require that the mount process passes the remote address */
+	if (data->addr.sin_addr.s_addr == INADDR_ANY) {
+		dprintk("%s: mount program didn't pass remote address!\n",
+			__FUNCTION__);
+		return -EINVAL;
 	}
-	server->backing_dev_info.ra_pages = server->rpages * NFS_MAX_READAHEAD;
 
-	nfs_super_set_maxbytes(sb, fsinfo.maxfilesize);
+	/* Prepare the root filehandle */
+	if (data->flags & NFS_MOUNT_VER3)
+		mntfh->size = data->root.size;
+	else
+		mntfh->size = NFS2_FHSIZE;
 
-	server->client->cl_intr = (server->flags & NFS_MOUNT_INTR) ? 1 : 0;
-	server->client->cl_softrtry = (server->flags & NFS_MOUNT_SOFT) ? 1 : 0;
+	if (mntfh->size > sizeof(mntfh->data)) {
+		dprintk("%s: invalid root filehandle\n", __FUNCTION__);
+		return -EINVAL;
+	}
+
+	memcpy(mntfh->data, data->root.data, mntfh->size);
+	if (mntfh->size < sizeof(mntfh->data))
+		memset(mntfh->data + mntfh->size, 0,
+		       sizeof(mntfh->data) - mntfh->size);
 
-	/* We're airborne Set socket buffersize */
-	rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100);
 	return 0;
-	/* Yargs. It didn't work out. */
-out_no_root:
-	dprintk("nfs_sb_init: get root inode failed: errno %d\n", -no_root_error);
-	if (!IS_ERR(root_inode))
-		iput(root_inode);
-	return no_root_error;
 }
 
 /*
- * Create an RPC client handle.
+ * Initialise the common bits of the superblock
  */
-static struct rpc_clnt *
-nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
+static inline void nfs_initialise_sb(struct super_block *sb)
 {
-	struct nfs_client	*clp;
-	struct rpc_clnt		*clnt;
-	int			proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
-	int			nfsversion = 2;
-	int			err;
-
-#ifdef CONFIG_NFS_V3
-	if (server->flags & NFS_MOUNT_VER3)
-		nfsversion = 3;
-#endif
-
-	clp = nfs_get_client(server->hostname, &server->addr, nfsversion);
-	if (!clp) {
-		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
-		return ERR_PTR(PTR_ERR(clp));
-	}
-
-	if (clp->cl_cons_state == NFS_CS_INITING) {
-		/* Check NFS protocol revision and initialize RPC op
-		 * vector and file handle pool. */
-#ifdef CONFIG_NFS_V3
-		if (nfsversion == 3) {
-			clp->rpc_ops = &nfs_v3_clientops;
-			server->caps |= NFS_CAP_READDIRPLUS;
-		} else {
-			clp->rpc_ops = &nfs_v2_clientops;
-		}
-#else
-		clp->rpc_ops = &nfs_v2_clientops;
-#endif
-
-		/* create transport and client */
-		err = nfs_create_rpc_client(clp, proto, data->timeo,
-					    data->retrans, RPC_AUTH_UNIX);
-		if (err < 0)
-			goto client_init_error;
-
-		nfs_mark_client_ready(clp, 0);
-	}
+	struct nfs_server *server = NFS_SB(sb);
 
-	/* create an nfs_server-specific client */
-	clnt = rpc_clone_client(clp->cl_rpcclient);
-	if (IS_ERR(clnt)) {
-		dprintk("%s: couldn't create rpc_client!\n", __FUNCTION__);
-		nfs_put_client(clp);
-		return ERR_PTR(PTR_ERR(clnt));
-	}
+	sb->s_magic = NFS_SUPER_MAGIC;
 
-	if (data->pseudoflavor != clp->cl_rpcclient->cl_auth->au_flavor) {
-		struct rpc_auth *auth;
+	/* We probably want something more informative here */
+	snprintf(sb->s_id, sizeof(sb->s_id),
+		 "%x:%x", MAJOR(sb->s_dev), MINOR(sb->s_dev));
 
-		auth = rpcauth_create(data->pseudoflavor, server->client);
-		if (IS_ERR(auth)) {
-			dprintk("%s: couldn't create credcache!\n", __FUNCTION__);
-			return ERR_PTR(PTR_ERR(auth));
-		}
-	}
+	if (sb->s_blocksize == 0)
+		sb->s_blocksize = nfs_block_bits(server->wsize,
+						 &sb->s_blocksize_bits);
 
-	server->nfs_client = clp;
-	return clnt;
+	if (server->flags & NFS_MOUNT_NOAC)
+		sb->s_flags |= MS_SYNCHRONOUS;
 
-client_init_error:
-	nfs_mark_client_ready(clp, err);
-	nfs_put_client(clp);
-	return ERR_PTR(err);
+	nfs_super_set_maxbytes(sb, server->maxfilesize);
 }
 
 /*
- * Clone a server record
+ * Finish setting up an NFS2/3 superblock
  */
-static struct nfs_server *nfs_clone_server(struct super_block *sb, struct nfs_clone_mount *data)
+static void nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data)
 {
 	struct nfs_server *server = NFS_SB(sb);
-	struct nfs_server *parent = NFS_SB(data->sb);
-	struct inode *root_inode;
-	struct nfs_fsinfo fsinfo;
-	void *err = ERR_PTR(-ENOMEM);
-
-	sb->s_op = data->sb->s_op;
-	sb->s_blocksize = data->sb->s_blocksize;
-	sb->s_blocksize_bits = data->sb->s_blocksize_bits;
-	sb->s_maxbytes = data->sb->s_maxbytes;
-
-	server->client_acl = ERR_PTR(-EINVAL);
-	server->io_stats = nfs_alloc_iostats();
-	if (server->io_stats == NULL)
-		goto out;
-
-	server->client = rpc_clone_client(parent->client);
-	if (IS_ERR((err = server->client)))
-		goto out;
-
-	if (!IS_ERR(parent->client_acl)) {
-		server->client_acl = rpc_clone_client(parent->client_acl);
-		if (IS_ERR((err = server->client_acl)))
-			goto out;
-	}
-	root_inode = nfs_fhget(sb, data->fh, data->fattr);
-	if (!root_inode)
-		goto out;
-	sb->s_root = d_alloc_root(root_inode);
-	if (!sb->s_root)
-		goto out_put_root;
-	fsinfo.fattr = data->fattr;
-	if (NFS_PROTO(root_inode)->fsinfo(server, data->fh, &fsinfo) == 0)
-		nfs_super_set_maxbytes(sb, fsinfo.maxfilesize);
-	sb->s_root->d_op = server->nfs_client->rpc_ops->dentry_ops;
-	sb->s_flags |= MS_ACTIVE;
-	return server;
-out_put_root:
-	iput(root_inode);
-out:
-	return err;
-}
-
-/*
- * Copy an existing superblock and attach revised data
- */
-static int nfs_clone_generic_sb(struct nfs_clone_mount *data,
-		struct super_block *(*fill_sb)(struct nfs_server *, struct nfs_clone_mount *),
-		struct nfs_server *(*fill_server)(struct super_block *, struct nfs_clone_mount *),
-		struct vfsmount *mnt)
-{
-	struct nfs_server *server;
-	struct nfs_server *parent = NFS_SB(data->sb);
-	struct super_block *sb = ERR_PTR(-EINVAL);
-	char *hostname;
-	int error = -ENOMEM;
-	int len;
-
-	server = kmalloc(sizeof(struct nfs_server), GFP_KERNEL);
-	if (server == NULL)
-		goto out_err;
-	memcpy(server, parent, sizeof(*server));
-	atomic_inc(&server->nfs_client->cl_count);
-	hostname = (data->hostname != NULL) ? data->hostname : parent->hostname;
-	len = strlen(hostname) + 1;
-	server->hostname = kmalloc(len, GFP_KERNEL);
-	if (server->hostname == NULL)
-		goto free_server;
-	memcpy(server->hostname, hostname, len);
-
-	sb = fill_sb(server, data);
-	if (IS_ERR(sb)) {
-		error = PTR_ERR(sb);
-		goto free_hostname;
-	}
 
-	if (sb->s_root)
-		goto out_share;
+	sb->s_blocksize_bits = 0;
+	sb->s_blocksize = 0;
+	if (data->bsize)
+		sb->s_blocksize = nfs_block_size(data->bsize, &sb->s_blocksize_bits);
 
-	server = fill_server(sb, data);
-	if (IS_ERR(server)) {
-		error = PTR_ERR(server);
-		goto out_deactivate;
+	if (server->flags & NFS_MOUNT_VER3) {
+		/* The VFS shouldn't apply the umask to mode bits. We will do
+		 * so ourselves when necessary.
+		 */
+		sb->s_flags |= MS_POSIXACL;
+		sb->s_time_gran = 1;
 	}
-	return simple_set_mnt(mnt, sb);
-out_deactivate:
-	up_write(&sb->s_umount);
-	deactivate_super(sb);
-	return error;
-out_share:
-	kfree(server->hostname);
-	nfs_put_client(server->nfs_client);
-	kfree(server);
-	return simple_set_mnt(mnt, sb);
-free_hostname:
-	kfree(server->hostname);
-free_server:
-	nfs_put_client(server->nfs_client);
-	kfree(server);
-out_err:
-	return error;
+
+	sb->s_op = &nfs_sops;
+ 	nfs_initialise_sb(sb);
 }
 
 /*
- * Set up an NFS2/3 superblock
- *
- * The way this works is that the mount process passes a structure
- * in the data argument which contains the server's IP address
- * and the root file handle obtained from the server's mount
- * daemon. We stash these away in the private superblock fields.
+ * Finish setting up a cloned NFS2/3 superblock
  */
-static int
-nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
+static void nfs_clone_super(struct super_block *sb,
+			    const struct super_block *old_sb)
 {
-	struct nfs_server	*server;
-	rpc_authflavor_t	authflavor;
+	struct nfs_server *server = NFS_SB(sb);
+
+	sb->s_blocksize_bits = old_sb->s_blocksize_bits;
+	sb->s_blocksize = old_sb->s_blocksize;
+	sb->s_maxbytes = old_sb->s_maxbytes;
 
-	server           = NFS_SB(sb);
-	sb->s_blocksize_bits = 0;
-	sb->s_blocksize = 0;
-	if (data->bsize)
-		sb->s_blocksize = nfs_block_size(data->bsize, &sb->s_blocksize_bits);
-	if (data->rsize)
-		server->rsize = nfs_block_size(data->rsize, NULL);
-	if (data->wsize)
-		server->wsize = nfs_block_size(data->wsize, NULL);
-	server->flags    = data->flags & NFS_MOUNT_FLAGMASK;
-
-	server->acregmin = data->acregmin*HZ;
-	server->acregmax = data->acregmax*HZ;
-	server->acdirmin = data->acdirmin*HZ;
-	server->acdirmax = data->acdirmax*HZ;
-
-	/* Start lockd here, before we might error out */
-	if (!(server->flags & NFS_MOUNT_NONLM))
-		lockd_up();
-
-	server->namelen  = data->namlen;
-	server->hostname = kmalloc(strlen(data->hostname) + 1, GFP_KERNEL);
-	if (!server->hostname)
-		return -ENOMEM;
-	strcpy(server->hostname, data->hostname);
-
-	/* Fill in pseudoflavor for mount version < 5 */
-	if (!(data->flags & NFS_MOUNT_SECFLAVOUR))
-		data->pseudoflavor = RPC_AUTH_UNIX;
-	authflavor = data->pseudoflavor;	/* save for sb_init() */
-	/* XXX maybe we want to add a server->pseudoflavor field */
-
-	/* Create RPC client handles */
-	server->client = nfs_create_client(server, data);
-	if (IS_ERR(server->client))
-		return PTR_ERR(server->client);
-
-	/* RFC 2623, sec 2.3.2 */
 	if (server->flags & NFS_MOUNT_VER3) {
-#ifdef CONFIG_NFS_V3_ACL
-		if (!(server->flags & NFS_MOUNT_NOACL)) {
-			server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3);
-			/* No errors! Assume that Sun nfsacls are supported */
-			if (!IS_ERR(server->client_acl))
-				server->caps |= NFS_CAP_ACLS;
-		}
-#else
-		server->flags &= ~NFS_MOUNT_NOACL;
-#endif /* CONFIG_NFS_V3_ACL */
-		/*
-		 * The VFS shouldn't apply the umask to mode bits. We will
-		 * do so ourselves when necessary.
+		/* The VFS shouldn't apply the umask to mode bits. We will do
+		 * so ourselves when necessary.
 		 */
 		sb->s_flags |= MS_POSIXACL;
-		if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN)
-			server->namelen = NFS3_MAXNAMLEN;
 		sb->s_time_gran = 1;
-	} else {
-		if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN)
-			server->namelen = NFS2_MAXNAMLEN;
 	}
 
-	sb->s_op = &nfs_sops;
-	return nfs_sb_init(sb, authflavor);
+	sb->s_op = old_sb->s_op;
+ 	nfs_initialise_sb(sb);
 }
 
-static int nfs_set_super(struct super_block *s, void *data)
+static int nfs_set_super(struct super_block *s, void *_server)
 {
-	s->s_fs_info = data;
-	return set_anon_super(s, data);
+	struct nfs_server *server = _server;
+	int ret;
+
+	s->s_fs_info = server;
+	ret = set_anon_super(s, server);
+	if (ret == 0)
+		server->s_dev = s->s_dev;
+	return ret;
 }
 
 static int nfs_compare_super(struct super_block *sb, void *data)
 {
-	struct nfs_server *server = data;
-	struct nfs_server *old = NFS_SB(sb);
+	struct nfs_server *server = data, *old = NFS_SB(sb);
 
-	if (old->addr.sin_addr.s_addr != server->addr.sin_addr.s_addr)
+	if (old->nfs_client != server->nfs_client)
 		return 0;
-	if (old->addr.sin_port != server->addr.sin_port)
+	if (memcmp(&old->fsid, &server->fsid, sizeof(old->fsid)) != 0)
 		return 0;
-	return !nfs_compare_fh(&old->fh, &server->fh);
+	return 1;
 }
 
 static int nfs_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
 {
-	int error;
 	struct nfs_server *server = NULL;
 	struct super_block *s;
-	struct nfs_fh *root;
+	struct nfs_fh mntfh;
 	struct nfs_mount_data *data = raw_data;
+	struct dentry *mntroot;
+	int error;
 
-	error = -EINVAL;
-	if (data == NULL) {
-		dprintk("%s: missing data argument\n", __FUNCTION__);
-		goto out_err_noserver;
-	}
-	if (data->version <= 0 || data->version > NFS_MOUNT_VERSION) {
-		dprintk("%s: bad mount version\n", __FUNCTION__);
-		goto out_err_noserver;
-	}
-	switch (data->version) {
-		case 1:
-			data->namlen = 0;
-		case 2:
-			data->bsize  = 0;
-		case 3:
-			if (data->flags & NFS_MOUNT_VER3) {
-				dprintk("%s: mount structure version %d does not support NFSv3\n",
-						__FUNCTION__,
-						data->version);
-				goto out_err_noserver;
-			}
-			data->root.size = NFS2_FHSIZE;
-			memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
-		case 4:
-			if (data->flags & NFS_MOUNT_SECFLAVOUR) {
-				dprintk("%s: mount structure version %d does not support strong security\n",
-						__FUNCTION__,
-						data->version);
-				goto out_err_noserver;
-			}
-		case 5:
-			memset(data->context, 0, sizeof(data->context));
-	}
-#ifndef CONFIG_NFS_V3
-	/* If NFSv3 is not compiled in, return -EPROTONOSUPPORT */
-	error = -EPROTONOSUPPORT;
-	if (data->flags & NFS_MOUNT_VER3) {
-		dprintk("%s: NFSv3 not compiled into kernel\n", __FUNCTION__);
-		goto out_err_noserver;
-	}
-#endif /* CONFIG_NFS_V3 */
+	/* Validate the mount data */
+	error = nfs_validate_mount_data(data, &mntfh);
+	if (error < 0)
+		return error;
 
-	error = -ENOMEM;
-	server = kzalloc(sizeof(struct nfs_server), GFP_KERNEL);
-	if (!server)
+	/* Get a volume representation */
+	server = nfs_create_server(data, &mntfh);
+	if (IS_ERR(server)) {
+		error = PTR_ERR(server);
 		goto out_err_noserver;
-	/* Zero out the NFS state stuff */
-	init_nfsv4_state(server);
-	server->client = server->client_acl = ERR_PTR(-EINVAL);
-
-	root = &server->fh;
-	if (data->flags & NFS_MOUNT_VER3)
-		root->size = data->root.size;
-	else
-		root->size = NFS2_FHSIZE;
-	error = -EINVAL;
-	if (root->size > sizeof(root->data)) {
-		dprintk("%s: invalid root filehandle\n", __FUNCTION__);
-		goto out_err;
-	}
-	memcpy(root->data, data->root.data, root->size);
-
-	/* We now require that the mount process passes the remote address */
-	memcpy(&server->addr, &data->addr, sizeof(server->addr));
-	if (server->addr.sin_addr.s_addr == INADDR_ANY) {
-		dprintk("%s: mount program didn't pass remote address!\n",
-				__FUNCTION__);
-		goto out_err;
 	}
 
+	/* Get a superblock - note that we may end up sharing one that already exists */
 	s = sget(fs_type, nfs_compare_super, nfs_set_super, server);
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
-		goto out_err;
+		goto out_err_nosb;
 	}
 
-	if (s->s_root)
-		goto out_share;
+	if (s->s_fs_info != server) {
+		nfs_free_server(server);
+		server = NULL;
+	}
 
-	s->s_flags = flags;
+	if (!s->s_root) {
+		/* initial superblock/root creation */
+		s->s_flags = flags;
+		nfs_fill_super(s, data);
+	}
 
-	error = nfs_fill_super(s, data, flags & MS_SILENT ? 1 : 0);
-	if (error) {
-		up_write(&s->s_umount);
-		deactivate_super(s);
-		return error;
+	mntroot = nfs_get_root(s, &mntfh);
+	if (IS_ERR(mntroot)) {
+		error = PTR_ERR(mntroot);
+		goto error_splat_super;
 	}
-	s->s_flags |= MS_ACTIVE;
-	return simple_set_mnt(mnt, s);
 
-out_share:
-	kfree(server);
-	return simple_set_mnt(mnt, s);
+	s->s_flags |= MS_ACTIVE;
+	mnt->mnt_sb = s;
+	mnt->mnt_root = mntroot;
+	return 0;
 
-out_err:
-	kfree(server);
+out_err_nosb:
+	nfs_free_server(server);
 out_err_noserver:
 	return error;
+
+error_splat_super:
+	up_write(&s->s_umount);
+	deactivate_super(s);
+	return error;
 }
 
+/*
+ * Destroy an NFS2/3 superblock
+ */
 static void nfs_kill_super(struct super_block *s)
 {
 	struct nfs_server *server = NFS_SB(s);
 
 	kill_anon_super(s);
-
-	if (!IS_ERR(server->client))
-		rpc_shutdown_client(server->client);
-	if (!IS_ERR(server->client_acl))
-		rpc_shutdown_client(server->client_acl);
-
-	if (!(server->flags & NFS_MOUNT_NONLM))
-		lockd_down();	/* release rpc.lockd */
-
-	nfs_free_iostats(server->io_stats);
-	kfree(server->hostname);
-	nfs_put_client(server->nfs_client);
-	kfree(server);
-	nfs_release_automount_timer();
+	nfs_free_server(server);
 }
 
-static struct super_block *nfs_clone_sb(struct nfs_server *server, struct nfs_clone_mount *data)
-{
-	struct super_block *sb;
-
-	server->fsid = data->fattr->fsid;
-	nfs_copy_fh(&server->fh, data->fh);
-	sb = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server);
-	if (!IS_ERR(sb) && sb->s_root == NULL && !(server->flags & NFS_MOUNT_NONLM))
-		lockd_up();
-	return sb;
-}
-
-static int nfs_clone_nfs_sb(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
+/*
+ * Clone an NFS2/3 server record on xdev traversal (FSID-change)
+ */
+static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags,
+			   const char *dev_name, void *raw_data,
+			   struct vfsmount *mnt)
 {
 	struct nfs_clone_mount *data = raw_data;
-	return nfs_clone_generic_sb(data, nfs_clone_sb, nfs_clone_server, mnt);
-}
+	struct super_block *s;
+	struct nfs_server *server;
+	struct dentry *mntroot;
+	int error;
 
-#ifdef CONFIG_NFS_V4
-static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
-	int timeo, int retrans, int proto, rpc_authflavor_t flavor)
-{
-	struct nfs_client *clp;
-	struct rpc_clnt *clnt = NULL;
-	int err = -EIO;
-
-	clp = nfs_get_client(server->hostname, &server->addr, 4);
-	if (!clp) {
-		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
-		return ERR_PTR(err);
-	}
+	dprintk("--> nfs_xdev_get_sb()\n");
 
-	/* Now create transport and client */
-	if (clp->cl_cons_state == NFS_CS_INITING) {
-		clp->rpc_ops = &nfs_v4_clientops;
+	/* create a new volume representation */
+	server = nfs_clone_server(NFS_SB(data->sb), data->fh, data->fattr);
+	if (IS_ERR(server)) {
+		error = PTR_ERR(server);
+		goto out_err_noserver;
+	}
 
-		err = nfs_create_rpc_client(clp, proto, timeo, retrans, flavor);
-		if (err < 0)
-			goto client_init_error;
+	/* Get a superblock - note that we may end up sharing one that already exists */
+	s = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server);
+	if (IS_ERR(s)) {
+		error = PTR_ERR(s);
+		goto out_err_nosb;
+	}
 
-		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
-		err = nfs_idmap_new(clp);
-		if (err < 0) {
-			dprintk("%s: failed to create idmapper.\n",
-				__FUNCTION__);
-			goto client_init_error;
-		}
-		__set_bit(NFS_CS_IDMAP, &clp->cl_res_state);
-		nfs_mark_client_ready(clp, 0);
+	if (s->s_fs_info != server) {
+		nfs_free_server(server);
+		server = NULL;
 	}
 
-	clnt = rpc_clone_client(clp->cl_rpcclient);
+	if (!s->s_root) {
+		/* initial superblock/root creation */
+		s->s_flags = flags;
+		nfs_clone_super(s, data->sb);
+	}
 
-	if (IS_ERR(clnt)) {
-		dprintk("%s: cannot create RPC client. Error = %d\n",
-				__FUNCTION__, err);
-		return clnt;
+	mntroot = nfs_get_root(s, data->fh);
+	if (IS_ERR(mntroot)) {
+		error = PTR_ERR(mntroot);
+		goto error_splat_super;
 	}
 
-	if (clnt->cl_auth->au_flavor != flavor) {
-		struct rpc_auth *auth;
+	s->s_flags |= MS_ACTIVE;
+	mnt->mnt_sb = s;
+	mnt->mnt_root = mntroot;
 
-		auth = rpcauth_create(flavor, clnt);
-		if (IS_ERR(auth)) {
-			dprintk("%s: couldn't create credcache!\n", __FUNCTION__);
-			return (struct rpc_clnt *)auth;
-		}
-	}
+	dprintk("<-- nfs_xdev_get_sb() = 0\n");
+	return 0;
 
-	server->nfs_client = clp;
-	down_write(&clp->cl_sem);
-	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
-	up_write(&clp->cl_sem);
-	return clnt;
+out_err_nosb:
+	nfs_free_server(server);
+out_err_noserver:
+	dprintk("<-- nfs_xdev_get_sb() = %d [error]\n", error);
+	return error;
 
-client_init_error:
-	nfs_mark_client_ready(clp, err);
-	nfs_put_client(clp);
-	return ERR_PTR(err);
+error_splat_super:
+	up_write(&s->s_umount);
+	deactivate_super(s);
+	dprintk("<-- nfs_xdev_get_sb() = %d [splat]\n", error);
+	return error;
 }
 
+#ifdef CONFIG_NFS_V4
+
 /*
- * Set up an NFS4 superblock
+ * Finish setting up a cloned NFS4 superblock
  */
-static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, int silent)
+static void nfs4_clone_super(struct super_block *sb,
+			    const struct super_block *old_sb)
 {
-	struct nfs_server *server;
-	rpc_authflavor_t authflavour;
-	int err = -EIO;
-
-	sb->s_blocksize_bits = 0;
-	sb->s_blocksize = 0;
-	server = NFS_SB(sb);
-	if (data->rsize != 0)
-		server->rsize = nfs_block_size(data->rsize, NULL);
-	if (data->wsize != 0)
-		server->wsize = nfs_block_size(data->wsize, NULL);
-	server->flags = data->flags & NFS_MOUNT_FLAGMASK;
-	server->caps = NFS_CAP_ATOMIC_OPEN;
-
-	server->acregmin = data->acregmin*HZ;
-	server->acregmax = data->acregmax*HZ;
-	server->acdirmin = data->acdirmin*HZ;
-	server->acdirmax = data->acdirmax*HZ;
-
-	/* Now create transport and client */
-	authflavour = RPC_AUTH_UNIX;
-	if (data->auth_flavourlen != 0) {
-		if (data->auth_flavourlen != 1) {
-			dprintk("%s: Invalid number of RPC auth flavours %d.\n",
-					__FUNCTION__, data->auth_flavourlen);
-			err = -EINVAL;
-			goto out_fail;
-		}
-		if (copy_from_user(&authflavour, data->auth_flavours, sizeof(authflavour))) {
-			err = -EFAULT;
-			goto out_fail;
-		}
-	}
-
-	server->client = nfs4_create_client(server, data->timeo, data->retrans,
-					    data->proto, authflavour);
-	if (IS_ERR(server->client)) {
-		err = PTR_ERR(server->client);
-			dprintk("%s: cannot create RPC client. Error = %d\n",
-					__FUNCTION__, err);
-			goto out_fail;
-	}
-
+	sb->s_blocksize_bits = old_sb->s_blocksize_bits;
+	sb->s_blocksize = old_sb->s_blocksize;
+	sb->s_maxbytes = old_sb->s_maxbytes;
 	sb->s_time_gran = 1;
-
-	sb->s_op = &nfs4_sops;
-	err = nfs_sb_init(sb, authflavour);
-
- out_fail:
-	return err;
+	sb->s_op = old_sb->s_op;
+ 	nfs_initialise_sb(sb);
 }
 
-static int nfs4_compare_super(struct super_block *sb, void *data)
+/*
+ * Set up an NFS4 superblock
+ */
+static void nfs4_fill_super(struct super_block *sb)
 {
-	struct nfs_server *server = data;
-	struct nfs_server *old = NFS_SB(sb);
-
-	if (strcmp(server->hostname, old->hostname) != 0)
-		return 0;
-	if (strcmp(server->mnt_path, old->mnt_path) != 0)
-		return 0;
-	return 1;
+	sb->s_time_gran = 1;
+	sb->s_op = &nfs4_sops;
+	nfs_initialise_sb(sb);
 }
 
-static void *
-nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen)
+static void *nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen)
 {
 	void *p = NULL;
 
@@ -1157,14 +791,22 @@ nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen)
 	return dst;
 }
 
+/*
+ * Get the superblock for an NFS4 mountpoint
+ */
 static int nfs4_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
 {
-	int error;
-	struct nfs_server *server;
-	struct super_block *s;
 	struct nfs4_mount_data *data = raw_data;
+	struct super_block *s;
+	struct nfs_server *server;
+	struct sockaddr_in addr;
+	rpc_authflavor_t authflavour;
+	struct nfs_fh mntfh;
+	struct dentry *mntroot;
+	char *mntpath = NULL, *hostname = NULL, ip_addr[16];
 	void *p;
+	int error;
 
 	if (data == NULL) {
 		dprintk("%s: missing data argument\n", __FUNCTION__);
@@ -1175,75 +817,107 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 		return -EINVAL;
 	}
 
-	server = kzalloc(sizeof(struct nfs_server), GFP_KERNEL);
-	if (!server)
-		return -ENOMEM;
-	/* Zero out the NFS state stuff */
-	init_nfsv4_state(server);
-	server->client = server->client_acl = ERR_PTR(-EINVAL);
+	/* We now require that the mount process passes the remote address */
+	if (data->host_addrlen != sizeof(addr))
+		return -EINVAL;
+
+	if (copy_from_user(&addr, data->host_addr, sizeof(addr)))
+		return -EFAULT;
+
+	if (addr.sin_family != AF_INET ||
+	    addr.sin_addr.s_addr == INADDR_ANY
+	    ) {
+		dprintk("%s: mount program didn't pass remote IP address!\n",
+				__FUNCTION__);
+		return -EINVAL;
+	}
+
+	/* Grab the authentication type */
+	authflavour = RPC_AUTH_UNIX;
+	if (data->auth_flavourlen != 0) {
+		if (data->auth_flavourlen != 1) {
+			dprintk("%s: Invalid number of RPC auth flavours %d.\n",
+					__FUNCTION__, data->auth_flavourlen);
+			error = -EINVAL;
+			goto out_err_noserver;
+		}
+
+		if (copy_from_user(&authflavour, data->auth_flavours,
+				   sizeof(authflavour))) {
+			error = -EFAULT;
+			goto out_err_noserver;
+		}
+	}
 
 	p = nfs_copy_user_string(NULL, &data->hostname, 256);
 	if (IS_ERR(p))
 		goto out_err;
-	server->hostname = p;
+	hostname = p;
 
 	p = nfs_copy_user_string(NULL, &data->mnt_path, 1024);
 	if (IS_ERR(p))
 		goto out_err;
-	server->mnt_path = p;
+	mntpath = p;
 
-	p = nfs_copy_user_string(server->ip_addr, &data->client_addr,
-			sizeof(server->ip_addr) - 1);
+	dprintk("MNTPATH: %s\n", mntpath);
+
+	p = nfs_copy_user_string(ip_addr, &data->client_addr,
+				 sizeof(ip_addr) - 1);
 	if (IS_ERR(p))
 		goto out_err;
 
-	/* We now require that the mount process passes the remote address */
-	if (data->host_addrlen != sizeof(server->addr)) {
-		error = -EINVAL;
-		goto out_free;
-	}
-	if (copy_from_user(&server->addr, data->host_addr, sizeof(server->addr))) {
-		error = -EFAULT;
-		goto out_free;
-	}
-	if (server->addr.sin_family != AF_INET ||
-	    server->addr.sin_addr.s_addr == INADDR_ANY) {
-		dprintk("%s: mount program didn't pass remote IP address!\n",
-				__FUNCTION__);
-		error = -EINVAL;
-		goto out_free;
+	/* Get a volume representation */
+	server = nfs4_create_server(data, hostname, &addr, mntpath, ip_addr,
+				    authflavour, &mntfh);
+	if (IS_ERR(server)) {
+		error = PTR_ERR(server);
+		goto out_err_noserver;
 	}
 
-	s = sget(fs_type, nfs4_compare_super, nfs_set_super, server);
+	/* Get a superblock - note that we may end up sharing one that already exists */
+	s = sget(fs_type, nfs_compare_super, nfs_set_super, server);
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
 		goto out_free;
 	}
 
-	if (s->s_root) {
-		kfree(server->mnt_path);
-		kfree(server->hostname);
-		kfree(server);
-		return simple_set_mnt(mnt, s);
-	}
+	if (!s->s_root) {
+		/* initial superblock/root creation */
+		s->s_flags = flags;
 
-	s->s_flags = flags;
+		nfs4_fill_super(s);
+	} else {
+		nfs_free_server(server);
+	}
 
-	error = nfs4_fill_super(s, data, flags & MS_SILENT ? 1 : 0);
-	if (error) {
-		up_write(&s->s_umount);
-		deactivate_super(s);
-		return error;
+	mntroot = nfs4_get_root(s, &mntfh);
+	if (IS_ERR(mntroot)) {
+		error = PTR_ERR(mntroot);
+		goto error_splat_super;
 	}
+
 	s->s_flags |= MS_ACTIVE;
-	return simple_set_mnt(mnt, s);
+	mnt->mnt_sb = s;
+	mnt->mnt_root = mntroot;
+	kfree(mntpath);
+	kfree(hostname);
+	return 0;
+
 out_err:
 	error = PTR_ERR(p);
+	goto out_err_noserver;
+
 out_free:
-	kfree(server->mnt_path);
-	kfree(server->hostname);
-	kfree(server);
+	nfs_free_server(server);
+out_err_noserver:
+	kfree(mntpath);
+	kfree(hostname);
 	return error;
+
+error_splat_super:
+	up_write(&s->s_umount);
+	deactivate_super(s);
+	goto out_err_noserver;
 }
 
 static void nfs4_kill_super(struct super_block *sb)
@@ -1254,133 +928,140 @@ static void nfs4_kill_super(struct super_block *sb)
 	kill_anon_super(sb);
 
 	nfs4_renewd_prepare_shutdown(server);
-
-	if (server->client != NULL && !IS_ERR(server->client))
-		rpc_shutdown_client(server->client);
-
-	destroy_nfsv4_state(server);
-
-	nfs_free_iostats(server->io_stats);
-	kfree(server->hostname);
-	kfree(server);
-	nfs_release_automount_timer();
+	nfs_free_server(server);
 }
 
 /*
- * Constructs the SERVER-side path
+ * Clone an NFS4 server record on xdev traversal (FSID-change)
  */
-static inline char *nfs4_dup_path(const struct dentry *dentry)
+static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags,
+			    const char *dev_name, void *raw_data,
+			    struct vfsmount *mnt)
 {
-	char *page = (char *) __get_free_page(GFP_USER);
-	char *path;
+	struct nfs_clone_mount *data = raw_data;
+	struct super_block *s;
+	struct nfs_server *server;
+	struct dentry *mntroot;
+	int error;
 
-	path = nfs4_path(dentry, page, PAGE_SIZE);
-	if (!IS_ERR(path)) {
-		int len = PAGE_SIZE + page - path;
-		char *tmp = path;
+	dprintk("--> nfs4_xdev_get_sb()\n");
 
-		path = kmalloc(len, GFP_KERNEL);
-		if (path)
-			memcpy(path, tmp, len);
-		else
-			path = ERR_PTR(-ENOMEM);
+	/* create a new volume representation */
+	server = nfs_clone_server(NFS_SB(data->sb), data->fh, data->fattr);
+	if (IS_ERR(server)) {
+		error = PTR_ERR(server);
+		goto out_err_noserver;
 	}
-	free_page((unsigned long)page);
-	return path;
-}
 
-static struct super_block *nfs4_clone_sb(struct nfs_server *server, struct nfs_clone_mount *data)
-{
-	const struct dentry *dentry = data->dentry;
-	struct nfs_client *clp = server->nfs_client;
-	struct super_block *sb;
-
-	server->fsid = data->fattr->fsid;
-	nfs_copy_fh(&server->fh, data->fh);
-	server->mnt_path = nfs4_dup_path(dentry);
-	if (IS_ERR(server->mnt_path)) {
-		sb = (struct super_block *)server->mnt_path;
-		goto err;
+	/* Get a superblock - note that we may end up sharing one that already exists */
+	s = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server);
+	if (IS_ERR(s)) {
+		error = PTR_ERR(s);
+		goto out_err_nosb;
 	}
-	sb = sget(&nfs4_fs_type, nfs4_compare_super, nfs_set_super, server);
-	if (IS_ERR(sb) || sb->s_root)
-		goto free_path;
-	nfs4_server_capabilities(server, &server->fh);
-
-	down_write(&clp->cl_sem);
-	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
-	up_write(&clp->cl_sem);
-	return sb;
-free_path:
-	kfree(server->mnt_path);
-err:
-	server->mnt_path = NULL;
-	return sb;
-}
 
-static int nfs_clone_nfs4_sb(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
-{
-	struct nfs_clone_mount *data = raw_data;
-	return nfs_clone_generic_sb(data, nfs4_clone_sb, nfs_clone_server, mnt);
-}
+	if (s->s_fs_info != server) {
+		nfs_free_server(server);
+		server = NULL;
+	}
 
-static struct super_block *nfs4_referral_sb(struct nfs_server *server, struct nfs_clone_mount *data)
-{
-	struct super_block *sb = ERR_PTR(-ENOMEM);
-	int len;
-
-	len = strlen(data->mnt_path) + 1;
-	server->mnt_path = kmalloc(len, GFP_KERNEL);
-	if (server->mnt_path == NULL)
-		goto err;
-	memcpy(server->mnt_path, data->mnt_path, len);
-	memcpy(&server->addr, data->addr, sizeof(struct sockaddr_in));
-
-	sb = sget(&nfs4_fs_type, nfs4_compare_super, nfs_set_super, server);
-	if (IS_ERR(sb) || sb->s_root)
-		goto free_path;
-	return sb;
-free_path:
-	kfree(server->mnt_path);
-err:
-	server->mnt_path = NULL;
-	return sb;
-}
+	if (!s->s_root) {
+		/* initial superblock/root creation */
+		s->s_flags = flags;
+		nfs4_clone_super(s, data->sb);
+	}
 
-static struct nfs_server *nfs4_referral_server(struct super_block *sb, struct nfs_clone_mount *data)
-{
-	struct nfs_server *server = NFS_SB(sb);
-	int proto, timeo, retrans;
-	void *err;
-
-	proto = IPPROTO_TCP;
-	/* Since we are following a referral and there may be alternatives,
-	   set the timeouts and retries to low values */
-	timeo = 2;
-	retrans = 1;
-
-	nfs_put_client(server->nfs_client);
-	server->nfs_client = NULL;
-	server->client = nfs4_create_client(server, timeo, retrans, proto,
-					    data->authflavor);
-	if (IS_ERR((err = server->client)))
-		goto out_err;
+	mntroot = nfs4_get_root(s, data->fh);
+	if (IS_ERR(mntroot)) {
+		error = PTR_ERR(mntroot);
+		goto error_splat_super;
+	}
 
-	sb->s_time_gran = 1;
-	sb->s_op = &nfs4_sops;
-	err = ERR_PTR(nfs_sb_init(sb, data->authflavor));
-	if (!IS_ERR(err))
-		return server;
-out_err:
-	return (struct nfs_server *)err;
+	s->s_flags |= MS_ACTIVE;
+	mnt->mnt_sb = s;
+	mnt->mnt_root = mntroot;
+
+	dprintk("<-- nfs4_xdev_get_sb() = 0\n");
+	return 0;
+
+out_err_nosb:
+	nfs_free_server(server);
+out_err_noserver:
+	dprintk("<-- nfs4_xdev_get_sb() = %d [error]\n", error);
+	return error;
+
+error_splat_super:
+	up_write(&s->s_umount);
+	deactivate_super(s);
+	dprintk("<-- nfs4_xdev_get_sb() = %d [splat]\n", error);
+	return error;
 }
 
-static int nfs_referral_nfs4_sb(struct file_system_type *fs_type,
-		int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
+/*
+ * Create an NFS4 server record on referral traversal
+ */
+static int nfs4_referral_get_sb(struct file_system_type *fs_type, int flags,
+				const char *dev_name, void *raw_data,
+				struct vfsmount *mnt)
 {
 	struct nfs_clone_mount *data = raw_data;
-	return nfs_clone_generic_sb(data, nfs4_referral_sb, nfs4_referral_server, mnt);
+	struct super_block *s;
+	struct nfs_server *server;
+	struct dentry *mntroot;
+	struct nfs_fh mntfh;
+	int error;
+
+	dprintk("--> nfs4_referral_get_sb()\n");
+
+	/* create a new volume representation */
+	server = nfs4_create_referral_server(data, &mntfh);
+	if (IS_ERR(server)) {
+		error = PTR_ERR(server);
+		goto out_err_noserver;
+	}
+
+	/* Get a superblock - note that we may end up sharing one that already exists */
+	s = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server);
+	if (IS_ERR(s)) {
+		error = PTR_ERR(s);
+		goto out_err_nosb;
+	}
+
+	if (s->s_fs_info != server) {
+		nfs_free_server(server);
+		server = NULL;
+	}
+
+	if (!s->s_root) {
+		/* initial superblock/root creation */
+		s->s_flags = flags;
+		nfs4_fill_super(s);
+	}
+
+	mntroot = nfs4_get_root(s, data->fh);
+	if (IS_ERR(mntroot)) {
+		error = PTR_ERR(mntroot);
+		goto error_splat_super;
+	}
+
+	s->s_flags |= MS_ACTIVE;
+	mnt->mnt_sb = s;
+	mnt->mnt_root = mntroot;
+
+	dprintk("<-- nfs4_referral_get_sb() = 0\n");
+	return 0;
+
+out_err_nosb:
+	nfs_free_server(server);
+out_err_noserver:
+	dprintk("<-- nfs4_referral_get_sb() = %d [error]\n", error);
+	return error;
+
+error_splat_super:
+	up_write(&s->s_umount);
+	deactivate_super(s);
+	dprintk("<-- nfs4_referral_get_sb() = %d [splat]\n", error);
+	return error;
 }
 
-#endif
+#endif /* CONFIG_NFS_V4 */
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 7084ac9a6455..453d44666ea5 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1273,7 +1273,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		if (time_before(complain, jiffies)) {
 			dprintk("NFS: faulty NFS server %s:"
 				" (committed = %d) != (stable = %d)\n",
-				NFS_SERVER(data->inode)->hostname,
+				NFS_SERVER(data->inode)->nfs_client->cl_hostname,
 				resp->verf->committed, argp->stable);
 			complain = jiffies + 300 * HZ;
 		}
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index d404ceca9168..6d0be0efd1b5 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -51,7 +51,6 @@ struct nfs_client {
 	unsigned long		cl_lease_time;
 	unsigned long		cl_last_renewal;
 	struct work_struct	cl_renewd;
-	struct work_struct	cl_recoverd;
 
 	struct rpc_wait_queue	cl_rpcwaitq;
 
@@ -74,6 +73,10 @@ struct nfs_client {
  */
 struct nfs_server {
 	struct nfs_client *	nfs_client;	/* shared client and NFS4 state */
+	struct list_head	client_link;	/* List of other nfs_server structs
+						 * that share the same client
+						 */
+	struct list_head	master_link;	/* link in master servers list */
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
@@ -92,20 +95,13 @@ struct nfs_server {
 	unsigned int		acdirmin;
 	unsigned int		acdirmax;
 	unsigned int		namelen;
-	char *			hostname;	/* remote hostname */
-	struct nfs_fh		fh;
-	struct sockaddr_in	addr;
+
 	struct nfs_fsid		fsid;
+	__u64			maxfilesize;	/* maximum file size */
 	unsigned long		mount_time;	/* when this fs was mounted */
+	dev_t			s_dev;		/* superblock dev numbers */
+
 #ifdef CONFIG_NFS_V4
-	/* Our own IP address, as a null-terminated string.
-	 * This is used to generate the clientid, and the callback address.
-	 */
-	char			ip_addr[16];
-	char *			mnt_path;
-	struct list_head	nfs4_siblings;	/* List of other nfs_server structs
-						 * that share the same clientid
-						 */
 	u32			attr_bitmask[2];/* V4 bitmask representing the set
 						   of attributes supported on this
 						   filesystem */
@@ -113,6 +109,7 @@ struct nfs_server {
 						   that are supported on this
 						   filesystem */
 #endif
+	void (*destroy)(struct nfs_server *);
 };
 
 /* Server capabilities */
-- 
cgit v1.2.3


From ec739ef03dc926d05051c8c5838971445504470a Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:15 -0400
Subject: SUNRPC: Create a helper to tell whether a transport is bound

Hide the contents and format of xprt->addr by eliminating direct uses
of the xprt->addr.sin_port field.  This change is required to support
alternate RPC host address formats (eg IPv6).

Test-plan:
Destructive testing (unplugging the network temporarily).  Repeated runs of
Connectathon locking suite with UDP and TCP.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h | 16 ++++++++++++++++
 net/sunrpc/clnt.c           | 10 +++++-----
 net/sunrpc/pmap_clnt.c      |  5 ++++-
 net/sunrpc/xprt.c           |  2 +-
 net/sunrpc/xprtsock.c       | 14 ++++++++++----
 5 files changed, 36 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 3a0cca255b76..a71106723d71 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -269,6 +269,7 @@ int			xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to);
 #define XPRT_CONNECTED		(1)
 #define XPRT_CONNECTING		(2)
 #define XPRT_CLOSE_WAIT		(3)
+#define XPRT_BOUND		(4)
 
 static inline void xprt_set_connected(struct rpc_xprt *xprt)
 {
@@ -312,6 +313,21 @@ static inline int xprt_test_and_set_connecting(struct rpc_xprt *xprt)
 	return test_and_set_bit(XPRT_CONNECTING, &xprt->state);
 }
 
+static inline void xprt_set_bound(struct rpc_xprt *xprt)
+{
+	test_and_set_bit(XPRT_BOUND, &xprt->state);
+}
+
+static inline int xprt_bound(struct rpc_xprt *xprt)
+{
+	return test_bit(XPRT_BOUND, &xprt->state);
+}
+
+static inline void xprt_clear_bound(struct rpc_xprt *xprt)
+{
+	clear_bit(XPRT_BOUND, &xprt->state);
+}
+
 #endif /* __KERNEL__*/
 
 #endif /* _LINUX_SUNRPC_XPRT_H */
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 3e19d321067a..0b8d03d08561 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -148,7 +148,6 @@ rpc_new_client(struct rpc_xprt *xprt, char *servname,
 	clnt->cl_maxproc  = version->nrprocs;
 	clnt->cl_protname = program->name;
 	clnt->cl_pmap	  = &clnt->cl_pmap_default;
-	clnt->cl_port     = xprt->addr.sin_port;
 	clnt->cl_prog     = program->number;
 	clnt->cl_vers     = version->number;
 	clnt->cl_prot     = xprt->prot;
@@ -156,7 +155,7 @@ rpc_new_client(struct rpc_xprt *xprt, char *servname,
 	clnt->cl_metrics  = rpc_alloc_iostats(clnt);
 	rpc_init_wait_queue(&clnt->cl_pmap_default.pm_bindwait, "bindwait");
 
-	if (!clnt->cl_port)
+	if (!xprt_bound(clnt->cl_xprt))
 		clnt->cl_autobind = 1;
 
 	clnt->cl_rtt = &clnt->cl_rtt_default;
@@ -570,7 +569,7 @@ EXPORT_SYMBOL(rpc_max_payload);
 void rpc_force_rebind(struct rpc_clnt *clnt)
 {
 	if (clnt->cl_autobind)
-		clnt->cl_port = 0;
+		xprt_clear_bound(clnt->cl_xprt);
 }
 EXPORT_SYMBOL(rpc_force_rebind);
 
@@ -782,14 +781,15 @@ static void
 call_bind(struct rpc_task *task)
 {
 	struct rpc_clnt	*clnt = task->tk_client;
+	struct rpc_xprt *xprt = task->tk_xprt;
 
 	dprintk("RPC: %4d call_bind (status %d)\n",
 				task->tk_pid, task->tk_status);
 
 	task->tk_action = call_connect;
-	if (!clnt->cl_port) {
+	if (!xprt_bound(xprt)) {
 		task->tk_action = call_bind_status;
-		task->tk_timeout = task->tk_xprt->bind_timeout;
+		task->tk_timeout = xprt->bind_timeout;
 		rpc_getport(task, clnt);
 	}
 }
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index 623180f224c9..209ffdfee10b 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -142,15 +142,17 @@ pmap_getport_done(struct rpc_task *task)
 	dprintk("RPC: %4d pmap_getport_done(status %d, port %d)\n",
 			task->tk_pid, task->tk_status, clnt->cl_port);
 
-	xprt->ops->set_port(xprt, 0);
 	if (task->tk_status < 0) {
 		/* Make the calling task exit with an error */
+		xprt->ops->set_port(xprt, 0);
 		task->tk_action = rpc_exit_task;
 	} else if (clnt->cl_port == 0) {
 		/* Program not registered */
+		xprt->ops->set_port(xprt, 0);
 		rpc_exit(task, -EACCES);
 	} else {
 		xprt->ops->set_port(xprt, clnt->cl_port);
+		xprt_set_bound(xprt);
 		clnt->cl_port = htons(clnt->cl_port);
 	}
 	spin_lock(&pmap_lock);
@@ -218,6 +220,7 @@ pmap_create(char *hostname, struct sockaddr_in *srvaddr, int proto, int privileg
 	if (IS_ERR(xprt))
 		return (struct rpc_clnt *)xprt;
 	xprt->ops->set_port(xprt, RPC_PMAP_PORT);
+	xprt_set_bound(xprt);
 	if (!privileged)
 		xprt->resvport = 0;
 
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index e8c2bc4977f3..e239ef985ef7 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -534,7 +534,7 @@ void xprt_connect(struct rpc_task *task)
 	dprintk("RPC: %4d xprt_connect xprt %p %s connected\n", task->tk_pid,
 			xprt, (xprt_connected(xprt) ? "is" : "is not"));
 
-	if (!xprt->addr.sin_port) {
+	if (!xprt_bound(xprt)) {
 		task->tk_status = -EIO;
 		return;
 	}
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 441bd53f5eca..123ac1e5ba15 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1016,7 +1016,7 @@ static void xs_udp_connect_worker(void *args)
 	struct socket *sock = xprt->sock;
 	int err, status = -EIO;
 
-	if (xprt->shutdown || xprt->addr.sin_port == 0)
+	if (xprt->shutdown || !xprt_bound(xprt))
 		goto out;
 
 	dprintk("RPC:      xs_udp_connect_worker for xprt %p\n", xprt);
@@ -1099,7 +1099,7 @@ static void xs_tcp_connect_worker(void *args)
 	struct socket *sock = xprt->sock;
 	int err, status = -EIO;
 
-	if (xprt->shutdown || xprt->addr.sin_port == 0)
+	if (xprt->shutdown || !xprt_bound(xprt))
 		goto out;
 
 	dprintk("RPC:      xs_tcp_connect_worker for xprt %p\n", xprt);
@@ -1307,8 +1307,11 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 	if (xprt->slot == NULL)
 		return -ENOMEM;
 
-	xprt->prot = IPPROTO_UDP;
+	if (ntohs(xprt->addr.sin_port) != 0)
+		xprt_set_bound(xprt);
 	xprt->port = xs_get_random_port();
+
+	xprt->prot = IPPROTO_UDP;
 	xprt->tsh_size = 0;
 	xprt->resvport = capable(CAP_NET_BIND_SERVICE) ? 1 : 0;
 	/* XXX: header size can vary due to auth type, IPv6, etc. */
@@ -1348,8 +1351,11 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 	if (xprt->slot == NULL)
 		return -ENOMEM;
 
-	xprt->prot = IPPROTO_TCP;
+	if (ntohs(xprt->addr.sin_port) != 0)
+		xprt_set_bound(xprt);
 	xprt->port = xs_get_random_port();
+
+	xprt->prot = IPPROTO_TCP;
 	xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32);
 	xprt->resvport = capable(CAP_NET_BIND_SERVICE) ? 1 : 0;
 	xprt->max_payload = RPC_MAX_FRAGMENT_SIZE;
-- 
cgit v1.2.3


From 4a68179d38874c37be2802442a71b847f5d1a2a9 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:15 -0400
Subject: SUNRPC: Make RPC portmapper use per-transport storage

Move connection and bind state that was maintained in the rpc_clnt
structure to the rpc_xprt structure.  This will allow the creation of
a clean API for plugging in different types of bind mechanisms.

This brings improvements such as the elimination of a single spin lock to
control serialization for all in-kernel RPC binding.  A set of per-xprt
bitops is used to serialize tasks during RPC binding, just like it now
works for making RPC transport connections.

Test-plan:
Destructive testing (unplugging the network temporarily).  Connectathon
with UDP and TCP.  NFSv2/3 and NFSv4 mounting should be carefully checked.
Probably need to rig a server where certain services aren't running, or
that returns an error for some typical operation.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  23 +------
 include/linux/sunrpc/xprt.h |  14 ++++
 net/sunrpc/clnt.c           |   8 +--
 net/sunrpc/pmap_clnt.c      | 158 +++++++++++++++++++++++++++++---------------
 net/sunrpc/xprt.c           |   1 +
 5 files changed, 123 insertions(+), 81 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 8fe9f35eba31..00e9dbaec9c5 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -18,18 +18,6 @@
 #include <linux/sunrpc/timer.h>
 #include <asm/signal.h>
 
-/*
- * This defines an RPC port mapping
- */
-struct rpc_portmap {
-	__u32			pm_prog;
-	__u32			pm_vers;
-	__u32			pm_prot;
-	__u16			pm_port;
-	unsigned char		pm_binding : 1;	/* doing a getport() */
-	struct rpc_wait_queue	pm_bindwait;	/* waiting on getport() */
-};
-
 struct rpc_inode;
 
 /*
@@ -40,7 +28,9 @@ struct rpc_clnt {
 	atomic_t		cl_users;	/* number of references */
 	struct rpc_xprt *	cl_xprt;	/* transport */
 	struct rpc_procinfo *	cl_procinfo;	/* procedure info */
-	u32			cl_maxproc;	/* max procedure number */
+	u32			cl_prog,	/* RPC program number */
+				cl_vers,	/* RPC version number */
+				cl_maxproc;	/* max procedure number */
 
 	char *			cl_server;	/* server machine name */
 	char *			cl_protname;	/* protocol name */
@@ -55,7 +45,6 @@ struct rpc_clnt {
 				cl_dead     : 1;/* abandoned */
 
 	struct rpc_rtt *	cl_rtt;		/* RTO estimator data */
-	struct rpc_portmap *	cl_pmap;	/* port mapping */
 
 	int			cl_nodelen;	/* nodename length */
 	char 			cl_nodename[UNX_MAXNODENAME];
@@ -64,14 +53,8 @@ struct rpc_clnt {
 	struct dentry *		cl_dentry;	/* inode */
 	struct rpc_clnt *	cl_parent;	/* Points to parent of clones */
 	struct rpc_rtt		cl_rtt_default;
-	struct rpc_portmap	cl_pmap_default;
 	char			cl_inline_name[32];
 };
-#define cl_timeout		cl_xprt->timeout
-#define cl_prog			cl_pmap->pm_prog
-#define cl_vers			cl_pmap->pm_vers
-#define cl_port			cl_pmap->pm_port
-#define cl_prot			cl_pmap->pm_prot
 
 /*
  * General RPC program info
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index a71106723d71..4ce82616873d 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -138,6 +138,7 @@ struct rpc_xprt {
 	unsigned int		tsh_size;	/* size of transport specific
 						   header */
 
+	struct rpc_wait_queue	binding;	/* requests waiting on rpcbind */
 	struct rpc_wait_queue	sending;	/* requests waiting to send */
 	struct rpc_wait_queue	resend;		/* requests waiting to resend */
 	struct rpc_wait_queue	pending;	/* requests in flight */
@@ -270,6 +271,7 @@ int			xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to);
 #define XPRT_CONNECTING		(2)
 #define XPRT_CLOSE_WAIT		(3)
 #define XPRT_BOUND		(4)
+#define XPRT_BINDING		(5)
 
 static inline void xprt_set_connected(struct rpc_xprt *xprt)
 {
@@ -328,6 +330,18 @@ static inline void xprt_clear_bound(struct rpc_xprt *xprt)
 	clear_bit(XPRT_BOUND, &xprt->state);
 }
 
+static inline void xprt_clear_binding(struct rpc_xprt *xprt)
+{
+	smp_mb__before_clear_bit();
+	clear_bit(XPRT_BINDING, &xprt->state);
+	smp_mb__after_clear_bit();
+}
+
+static inline int xprt_test_and_set_binding(struct rpc_xprt *xprt)
+{
+	return test_and_set_bit(XPRT_BINDING, &xprt->state);
+}
+
 #endif /* __KERNEL__*/
 
 #endif /* _LINUX_SUNRPC_XPRT_H */
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 0b8d03d08561..cee504162a3f 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -147,13 +147,10 @@ rpc_new_client(struct rpc_xprt *xprt, char *servname,
 	clnt->cl_procinfo = version->procs;
 	clnt->cl_maxproc  = version->nrprocs;
 	clnt->cl_protname = program->name;
-	clnt->cl_pmap	  = &clnt->cl_pmap_default;
 	clnt->cl_prog     = program->number;
 	clnt->cl_vers     = version->number;
-	clnt->cl_prot     = xprt->prot;
 	clnt->cl_stats    = program->stats;
 	clnt->cl_metrics  = rpc_alloc_iostats(clnt);
-	rpc_init_wait_queue(&clnt->cl_pmap_default.pm_bindwait, "bindwait");
 
 	if (!xprt_bound(clnt->cl_xprt))
 		clnt->cl_autobind = 1;
@@ -243,8 +240,6 @@ rpc_clone_client(struct rpc_clnt *clnt)
 	atomic_set(&new->cl_users, 0);
 	new->cl_parent = clnt;
 	atomic_inc(&clnt->cl_count);
-	/* Duplicate portmapper */
-	rpc_init_wait_queue(&new->cl_pmap_default.pm_bindwait, "bindwait");
 	/* Turn off autobind on clones */
 	new->cl_autobind = 0;
 	new->cl_oneshot = 0;
@@ -254,8 +249,7 @@ rpc_clone_client(struct rpc_clnt *clnt)
 	rpc_init_rtt(&new->cl_rtt_default, clnt->cl_xprt->timeout.to_initval);
 	if (new->cl_auth)
 		atomic_inc(&new->cl_auth->au_count);
-	new->cl_pmap		= &new->cl_pmap_default;
-	new->cl_metrics         = rpc_alloc_iostats(clnt);
+	new->cl_metrics = rpc_alloc_iostats(clnt);
 	return new;
 out_no_clnt:
 	printk(KERN_INFO "RPC: out of memory in %s\n", __FUNCTION__);
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index 209ffdfee10b..59d542436ca9 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -24,11 +24,57 @@
 #define PMAP_UNSET		2
 #define PMAP_GETPORT		3
 
+struct portmap_args {
+	u32			pm_prog;
+	u32			pm_vers;
+	u32			pm_prot;
+	unsigned short		pm_port;
+	struct rpc_task *	pm_task;
+};
+
 static struct rpc_procinfo	pmap_procedures[];
 static struct rpc_clnt *	pmap_create(char *, struct sockaddr_in *, int, int);
-static void			pmap_getport_done(struct rpc_task *);
+static void			pmap_getport_done(struct rpc_task *, void *);
 static struct rpc_program	pmap_program;
-static DEFINE_SPINLOCK(pmap_lock);
+
+static void pmap_getport_prepare(struct rpc_task *task, void *calldata)
+{
+	struct portmap_args *map = calldata;
+	struct rpc_message msg = {
+		.rpc_proc	= &pmap_procedures[PMAP_GETPORT],
+		.rpc_argp	= map,
+		.rpc_resp	= &map->pm_port,
+	};
+
+	rpc_call_setup(task, &msg, 0);
+}
+
+static inline struct portmap_args *pmap_map_alloc(void)
+{
+	return kmalloc(sizeof(struct portmap_args), GFP_NOFS);
+}
+
+static inline void pmap_map_free(struct portmap_args *map)
+{
+	kfree(map);
+}
+
+static void pmap_map_release(void *data)
+{
+	pmap_map_free(data);
+}
+
+static const struct rpc_call_ops pmap_getport_ops = {
+	.rpc_call_prepare	= pmap_getport_prepare,
+	.rpc_call_done		= pmap_getport_done,
+	.rpc_release		= pmap_map_release,
+};
+
+static inline void pmap_wake_portmap_waiters(struct rpc_xprt *xprt)
+{
+	xprt_clear_binding(xprt);
+	rpc_wake_up(&xprt->binding);
+}
 
 /*
  * Obtain the port for a given RPC service on a given host. This one can
@@ -37,67 +83,71 @@ static DEFINE_SPINLOCK(pmap_lock);
 void
 rpc_getport(struct rpc_task *task, struct rpc_clnt *clnt)
 {
-	struct rpc_portmap *map = clnt->cl_pmap;
-	struct sockaddr_in *sap = &clnt->cl_xprt->addr;
-	struct rpc_message msg = {
-		.rpc_proc	= &pmap_procedures[PMAP_GETPORT],
-		.rpc_argp	= map,
-		.rpc_resp	= &clnt->cl_port,
-		.rpc_cred	= NULL
-	};
+	struct rpc_xprt *xprt = task->tk_xprt;
+	struct sockaddr_in *sap = &xprt->addr;
+	struct portmap_args *map;
 	struct rpc_clnt	*pmap_clnt;
-	struct rpc_task	*child;
+	struct rpc_task *child;
 
-	dprintk("RPC: %4d rpc_getport(%s, %d, %d, %d)\n",
+	dprintk("RPC: %4d rpc_getport(%s, %u, %u, %d)\n",
 			task->tk_pid, clnt->cl_server,
-			map->pm_prog, map->pm_vers, map->pm_prot);
+			clnt->cl_prog, clnt->cl_vers, xprt->prot);
 
 	/* Autobind on cloned rpc clients is discouraged */
 	BUG_ON(clnt->cl_parent != clnt);
 
-	spin_lock(&pmap_lock);
-	if (map->pm_binding) {
-		rpc_sleep_on(&map->pm_bindwait, task, NULL, NULL);
-		spin_unlock(&pmap_lock);
+	if (xprt_test_and_set_binding(xprt)) {
+		task->tk_status = -EACCES;	/* tell caller to check again */
+		rpc_sleep_on(&xprt->binding, task, NULL, NULL);
 		return;
 	}
-	map->pm_binding = 1;
-	spin_unlock(&pmap_lock);
+
+	/* Someone else may have bound if we slept */
+	if (xprt_bound(xprt)) {
+		task->tk_status = 0;
+		goto bailout_nofree;
+	}
+
+	map = pmap_map_alloc();
+	if (!map) {
+		task->tk_status = -ENOMEM;
+		goto bailout_nofree;
+	}
+	map->pm_prog = clnt->cl_prog;
+	map->pm_vers = clnt->cl_vers;
+	map->pm_prot = xprt->prot;
+	map->pm_port = 0;
+	map->pm_task = task;
 
 	pmap_clnt = pmap_create(clnt->cl_server, sap, map->pm_prot, 0);
 	if (IS_ERR(pmap_clnt)) {
 		task->tk_status = PTR_ERR(pmap_clnt);
 		goto bailout;
 	}
-	task->tk_status = 0;
 
-	/*
-	 * Note: rpc_new_child will release client after a failure.
-	 */
-	if (!(child = rpc_new_child(pmap_clnt, task)))
+	child = rpc_run_task(pmap_clnt, RPC_TASK_ASYNC, &pmap_getport_ops, map);
+	if (IS_ERR(child)) {
+		task->tk_status = -EIO;
 		goto bailout;
+	}
+	rpc_release_task(child);
 
-	/* Setup the call info struct */
-	rpc_call_setup(child, &msg, 0);
+	rpc_sleep_on(&xprt->binding, task, NULL, NULL);
 
-	/* ... and run the child task */
 	task->tk_xprt->stat.bind_count++;
-	rpc_run_child(task, child, pmap_getport_done);
 	return;
 
 bailout:
-	spin_lock(&pmap_lock);
-	map->pm_binding = 0;
-	rpc_wake_up(&map->pm_bindwait);
-	spin_unlock(&pmap_lock);
-	rpc_exit(task, -EIO);
+	pmap_map_free(map);
+bailout_nofree:
+	pmap_wake_portmap_waiters(xprt);
 }
 
 #ifdef CONFIG_ROOT_NFS
 int
 rpc_getport_external(struct sockaddr_in *sin, __u32 prog, __u32 vers, int prot)
 {
-	struct rpc_portmap map = {
+	struct portmap_args map = {
 		.pm_prog	= prog,
 		.pm_vers	= vers,
 		.pm_prot	= prot,
@@ -133,32 +183,32 @@ rpc_getport_external(struct sockaddr_in *sin, __u32 prog, __u32 vers, int prot)
 #endif
 
 static void
-pmap_getport_done(struct rpc_task *task)
+pmap_getport_done(struct rpc_task *child, void *data)
 {
-	struct rpc_clnt	*clnt = task->tk_client;
+	struct portmap_args *map = data;
+	struct rpc_task *task = map->pm_task;
 	struct rpc_xprt *xprt = task->tk_xprt;
-	struct rpc_portmap *map = clnt->cl_pmap;
-
-	dprintk("RPC: %4d pmap_getport_done(status %d, port %d)\n",
-			task->tk_pid, task->tk_status, clnt->cl_port);
+	int status = child->tk_status;
 
-	if (task->tk_status < 0) {
-		/* Make the calling task exit with an error */
+	if (status < 0) {
+		/* Portmapper not available */
 		xprt->ops->set_port(xprt, 0);
-		task->tk_action = rpc_exit_task;
-	} else if (clnt->cl_port == 0) {
-		/* Program not registered */
+		task->tk_status = status;
+	} else if (map->pm_port == 0) {
+		/* Requested RPC service wasn't registered */
 		xprt->ops->set_port(xprt, 0);
-		rpc_exit(task, -EACCES);
+		task->tk_status = -EACCES;
 	} else {
-		xprt->ops->set_port(xprt, clnt->cl_port);
+		/* Succeeded */
+		xprt->ops->set_port(xprt, map->pm_port);
 		xprt_set_bound(xprt);
-		clnt->cl_port = htons(clnt->cl_port);
+		task->tk_status = 0;
 	}
-	spin_lock(&pmap_lock);
-	map->pm_binding = 0;
-	rpc_wake_up(&map->pm_bindwait);
-	spin_unlock(&pmap_lock);
+
+	dprintk("RPC: %4d pmap_getport_done(status %d, port %u)\n",
+			child->tk_pid, child->tk_status, map->pm_port);
+
+	pmap_wake_portmap_waiters(xprt);
 }
 
 /*
@@ -172,7 +222,7 @@ rpc_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay)
 		.sin_family	= AF_INET,
 		.sin_addr.s_addr = htonl(INADDR_LOOPBACK),
 	};
-	struct rpc_portmap	map = {
+	struct portmap_args	map = {
 		.pm_prog	= prog,
 		.pm_vers	= vers,
 		.pm_prot	= prot,
@@ -239,7 +289,7 @@ pmap_create(char *hostname, struct sockaddr_in *srvaddr, int proto, int privileg
  * XDR encode/decode functions for PMAP
  */
 static int
-xdr_encode_mapping(struct rpc_rqst *req, u32 *p, struct rpc_portmap *map)
+xdr_encode_mapping(struct rpc_rqst *req, u32 *p, struct portmap_args *map)
 {
 	dprintk("RPC: xdr_encode_mapping(%d, %d, %d, %d)\n",
 		map->pm_prog, map->pm_vers, map->pm_prot, map->pm_port);
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index e239ef985ef7..b45abd0743cb 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -928,6 +928,7 @@ static struct rpc_xprt *xprt_setup(int proto, struct sockaddr_in *ap, struct rpc
 	xprt->last_used = jiffies;
 	xprt->cwnd = RPC_INITCWND;
 
+	rpc_init_wait_queue(&xprt->binding, "xprt_binding");
 	rpc_init_wait_queue(&xprt->pending, "xprt_pending");
 	rpc_init_wait_queue(&xprt->sending, "xprt_sending");
 	rpc_init_wait_queue(&xprt->resend, "xprt_resend");
-- 
cgit v1.2.3


From 5b1eacbcd78930d976eb50a93f1779d311b553d1 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:16 -0400
Subject: SUNRPC: Support for RPC child tasks no longer needed

The previous patches removed the last user of RPC child tasks, so we can
remove support for child tasks from net/sunrpc/sched.c now.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/sched.h |  5 ---
 net/sunrpc/sched.c           | 82 --------------------------------------------
 2 files changed, 87 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 82a91bb22362..f399c138f79d 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -127,7 +127,6 @@ struct rpc_call_ops {
  */
 #define RPC_TASK_ASYNC		0x0001		/* is an async task */
 #define RPC_TASK_SWAPPER	0x0002		/* is swapping in/out */
-#define RPC_TASK_CHILD		0x0008		/* is child of other task */
 #define RPC_CALL_MAJORSEEN	0x0020		/* major timeout seen */
 #define RPC_TASK_ROOTCREDS	0x0040		/* force root creds */
 #define RPC_TASK_DYNAMIC	0x0080		/* task was kmalloc'ed */
@@ -136,7 +135,6 @@ struct rpc_call_ops {
 #define RPC_TASK_NOINTR		0x0400		/* uninterruptible task */
 
 #define RPC_IS_ASYNC(t)		((t)->tk_flags & RPC_TASK_ASYNC)
-#define RPC_IS_CHILD(t)		((t)->tk_flags & RPC_TASK_CHILD)
 #define RPC_IS_SWAPPER(t)	((t)->tk_flags & RPC_TASK_SWAPPER)
 #define RPC_DO_ROOTOVERRIDE(t)	((t)->tk_flags & RPC_TASK_ROOTCREDS)
 #define RPC_ASSASSINATED(t)	((t)->tk_flags & RPC_TASK_KILLED)
@@ -253,7 +251,6 @@ struct rpc_task *rpc_new_task(struct rpc_clnt *, int flags,
 				const struct rpc_call_ops *ops, void *data);
 struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags,
 				const struct rpc_call_ops *ops, void *data);
-struct rpc_task *rpc_new_child(struct rpc_clnt *, struct rpc_task *parent);
 void		rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt,
 				int flags, const struct rpc_call_ops *ops,
 				void *data);
@@ -261,8 +258,6 @@ void		rpc_release_task(struct rpc_task *);
 void		rpc_exit_task(struct rpc_task *);
 void		rpc_killall_tasks(struct rpc_clnt *);
 int		rpc_execute(struct rpc_task *);
-void		rpc_run_child(struct rpc_task *parent, struct rpc_task *child,
-					rpc_action action);
 void		rpc_init_priority_wait_queue(struct rpc_wait_queue *, const char *);
 void		rpc_init_wait_queue(struct rpc_wait_queue *, const char *);
 void		rpc_sleep_on(struct rpc_wait_queue *, struct rpc_task *,
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 5c3eee768504..015ffe423a2f 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -44,12 +44,6 @@ static void			__rpc_default_timer(struct rpc_task *task);
 static void			rpciod_killall(void);
 static void			rpc_async_schedule(void *);
 
-/*
- * RPC tasks that create another task (e.g. for contacting the portmapper)
- * will wait on this queue for their child's completion
- */
-static RPC_WAITQ(childq, "childq");
-
 /*
  * RPC tasks sit here while waiting for conditions to improve.
  */
@@ -323,16 +317,6 @@ static void rpc_make_runnable(struct rpc_task *task)
 		wake_up_bit(&task->tk_runstate, RPC_TASK_QUEUED);
 }
 
-/*
- * Place a newly initialized task on the workqueue.
- */
-static inline void
-rpc_schedule_run(struct rpc_task *task)
-{
-	rpc_set_active(task);
-	rpc_make_runnable(task);
-}
-
 /*
  * Prepare for sleeping on a wait queue.
  * By always appending tasks to the list we ensure FIFO behavior.
@@ -933,72 +917,6 @@ struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags,
 }
 EXPORT_SYMBOL(rpc_run_task);
 
-/**
- * rpc_find_parent - find the parent of a child task.
- * @child: child task
- * @parent: parent task
- *
- * Checks that the parent task is still sleeping on the
- * queue 'childq'. If so returns a pointer to the parent.
- * Upon failure returns NULL.
- *
- * Caller must hold childq.lock
- */
-static inline struct rpc_task *rpc_find_parent(struct rpc_task *child, struct rpc_task *parent)
-{
-	struct rpc_task	*task;
-	struct list_head *le;
-
-	task_for_each(task, le, &childq.tasks[0])
-		if (task == parent)
-			return parent;
-
-	return NULL;
-}
-
-static void rpc_child_exit(struct rpc_task *child, void *calldata)
-{
-	struct rpc_task	*parent;
-
-	spin_lock_bh(&childq.lock);
-	if ((parent = rpc_find_parent(child, calldata)) != NULL) {
-		parent->tk_status = child->tk_status;
-		__rpc_wake_up_task(parent);
-	}
-	spin_unlock_bh(&childq.lock);
-}
-
-static const struct rpc_call_ops rpc_child_ops = {
-	.rpc_call_done = rpc_child_exit,
-};
-
-/*
- * Note: rpc_new_task releases the client after a failure.
- */
-struct rpc_task *
-rpc_new_child(struct rpc_clnt *clnt, struct rpc_task *parent)
-{
-	struct rpc_task	*task;
-
-	task = rpc_new_task(clnt, RPC_TASK_ASYNC | RPC_TASK_CHILD, &rpc_child_ops, parent);
-	if (!task)
-		goto fail;
-	return task;
-
-fail:
-	parent->tk_status = -ENOMEM;
-	return NULL;
-}
-
-void rpc_run_child(struct rpc_task *task, struct rpc_task *child, rpc_action func)
-{
-	spin_lock_bh(&childq.lock);
-	/* N.B. Is it possible for the child to have already finished? */
-	__rpc_sleep_on(&childq, task, func, NULL);
-	rpc_schedule_run(child);
-	spin_unlock_bh(&childq.lock);
-}
-
 /*
  * Kill all tasks for the given client.
  * XXX: kill their descendants as well?
-- 
cgit v1.2.3


From bbf7c1dd2ae2b4040b41b1065ee9b1b6905b1605 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:16 -0400
Subject: SUNRPC: Introduce transport switch callout for pluggable rpcbind

Introduce a clean transport switch API for plugging in different types of
rpcbind mechanisms.  For instance, rpcbind can cleanly replace the
existing portmapper client, or a transport can choose to implement RPC
binding any way it likes.

Test plan:
Destructive testing (unplugging the network temporarily).  Connectathon
with UDP and TCP.  NFSv2/3 and NFSv4 mounting should be carefully checked.
Probably need to rig a server where certain services aren't running, or
that returns an error for some typical operation.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h | 2 +-
 include/linux/sunrpc/xprt.h | 1 +
 net/sunrpc/clnt.c           | 3 +--
 net/sunrpc/pmap_clnt.c      | 4 ++--
 net/sunrpc/xprtsock.c       | 2 ++
 5 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 00e9dbaec9c5..2e68ac0aa022 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -106,7 +106,7 @@ struct rpc_clnt *rpc_clone_client(struct rpc_clnt *);
 int		rpc_shutdown_client(struct rpc_clnt *);
 int		rpc_destroy_client(struct rpc_clnt *);
 void		rpc_release_client(struct rpc_clnt *);
-void		rpc_getport(struct rpc_task *, struct rpc_clnt *);
+void		rpc_getport(struct rpc_task *);
 int		rpc_register(u32, u32, int, unsigned short, int *);
 
 void		rpc_call_setup(struct rpc_task *, struct rpc_message *, int);
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 4ce82616873d..84122559fa17 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -105,6 +105,7 @@ struct rpc_xprt_ops {
 	void		(*set_buffer_size)(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize);
 	int		(*reserve_xprt)(struct rpc_task *task);
 	void		(*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task);
+	void		(*rpcbind)(struct rpc_task *task);
 	void		(*set_port)(struct rpc_xprt *xprt, unsigned short port);
 	void		(*connect)(struct rpc_task *task);
 	void *		(*buf_alloc)(struct rpc_task *task, size_t size);
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index cee504162a3f..d003c2f5688f 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -774,7 +774,6 @@ call_encode(struct rpc_task *task)
 static void
 call_bind(struct rpc_task *task)
 {
-	struct rpc_clnt	*clnt = task->tk_client;
 	struct rpc_xprt *xprt = task->tk_xprt;
 
 	dprintk("RPC: %4d call_bind (status %d)\n",
@@ -784,7 +783,7 @@ call_bind(struct rpc_task *task)
 	if (!xprt_bound(xprt)) {
 		task->tk_action = call_bind_status;
 		task->tk_timeout = xprt->bind_timeout;
-		rpc_getport(task, clnt);
+		xprt->ops->rpcbind(task);
 	}
 }
 
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index 0efcbf1302a2..f7b279a63baa 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -81,13 +81,13 @@ static inline void pmap_wake_portmap_waiters(struct rpc_xprt *xprt)
 /**
  * rpc_getport - obtain the port for a given RPC service on a given host
  * @task: task that is waiting for portmapper request
- * @clnt: controlling rpc_clnt
  *
  * This one can be called for an ongoing RPC request, and can be used in
  * an async (rpciod) context.
  */
-void rpc_getport(struct rpc_task *task, struct rpc_clnt *clnt)
+void rpc_getport(struct rpc_task *task)
 {
+	struct rpc_clnt *clnt = task->tk_client;
 	struct rpc_xprt *xprt = task->tk_xprt;
 	struct sockaddr_in *sap = &xprt->addr;
 	struct portmap_args *map;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 123ac1e5ba15..4c98b89a5b48 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1262,6 +1262,7 @@ static struct rpc_xprt_ops xs_udp_ops = {
 	.set_buffer_size	= xs_udp_set_buffer_size,
 	.reserve_xprt		= xprt_reserve_xprt_cong,
 	.release_xprt		= xprt_release_xprt_cong,
+	.rpcbind		= rpc_getport,
 	.set_port		= xs_set_port,
 	.connect		= xs_connect,
 	.buf_alloc		= rpc_malloc,
@@ -1278,6 +1279,7 @@ static struct rpc_xprt_ops xs_udp_ops = {
 static struct rpc_xprt_ops xs_tcp_ops = {
 	.reserve_xprt		= xprt_reserve_xprt,
 	.release_xprt		= xs_tcp_release_xprt,
+	.rpcbind		= rpc_getport,
 	.set_port		= xs_set_port,
 	.connect		= xs_connect,
 	.buf_alloc		= rpc_malloc,
-- 
cgit v1.2.3


From ed39440a2573abc926f230267000f21fa5a87822 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:17 -0400
Subject: SUNRPC: create API for getting remote peer address

Provide an API for retrieving the remote peer address without allowing
direct access to the rpc_xprt struct.

Test-plan:
Compile kernel with CONFIG_NFS enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  1 +
 net/sunrpc/clnt.c           | 21 +++++++++++++++++++++
 2 files changed, 22 insertions(+)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 2e68ac0aa022..65196b03f0ab 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -123,6 +123,7 @@ void		rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int);
 size_t		rpc_max_payload(struct rpc_clnt *);
 void		rpc_force_rebind(struct rpc_clnt *);
 int		rpc_ping(struct rpc_clnt *clnt, int flags);
+size_t		rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t);
 
 /*
  * Helper function for NFSroot support
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index d003c2f5688f..94768cf5fd5b 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -533,6 +533,27 @@ rpc_call_setup(struct rpc_task *task, struct rpc_message *msg, int flags)
 		task->tk_action = rpc_exit_task;
 }
 
+/**
+ * rpc_peeraddr - extract remote peer address from clnt's xprt
+ * @clnt: RPC client structure
+ * @buf: target buffer
+ * @size: length of target buffer
+ *
+ * Returns the number of bytes that are actually in the stored address.
+ */
+size_t rpc_peeraddr(struct rpc_clnt *clnt, struct sockaddr *buf, size_t bufsize)
+{
+	size_t bytes;
+	struct rpc_xprt *xprt = clnt->cl_xprt;
+
+	bytes = sizeof(xprt->addr);
+	if (bytes > bufsize)
+		bytes = bufsize;
+	memcpy(buf, &clnt->cl_xprt->addr, bytes);
+	return sizeof(xprt->addr);
+}
+EXPORT_SYMBOL(rpc_peeraddr);
+
 void
 rpc_setbufsize(struct rpc_clnt *clnt, unsigned int sndsize, unsigned int rcvsize)
 {
-- 
cgit v1.2.3


From 39d7bbcb5ba5e9d8d658b70903dd7939400e57db Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:18 -0400
Subject: SUNRPC: remove extraneous header inclusions

include/linux/sunrpc/clnt.h already includes include/linux/sunrpc/xprt.h.
We can remove xprt.h from source files that already include clnt.h.
Likewise include/linux/sunrpc/timer.h.

Test plan:
Compile kernel with CONFIG_NFS enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c     | 1 -
 include/linux/nfs_xdr.h | 1 -
 net/sunrpc/pmap_clnt.c  | 1 -
 net/sunrpc/sched.c      | 1 -
 net/sunrpc/timer.c      | 2 --
 5 files changed, 6 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 445abb4d4214..41274874b9a5 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -14,7 +14,6 @@
 #include <linux/net.h>
 #include <linux/in.h>
 #include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/xprt.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/nfs_fs.h>
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 2426b11b6cce..0f33e621892f 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -1,7 +1,6 @@
 #ifndef _LINUX_NFS_XDR_H
 #define _LINUX_NFS_XDR_H
 
-#include <linux/sunrpc/xprt.h>
 #include <linux/nfsacl.h>
 
 /*
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index 3eee8e907275..523f0e825dea 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -15,7 +15,6 @@
 #include <linux/uio.h>
 #include <linux/in.h>
 #include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/xprt.h>
 #include <linux/sunrpc/sched.h>
 
 #ifdef RPC_DEBUG
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 015ffe423a2f..ecf366351bf7 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -21,7 +21,6 @@
 #include <linux/mutex.h>
 
 #include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/xprt.h>
 
 #ifdef RPC_DEBUG
 #define RPCDBG_FACILITY		RPCDBG_SCHED
diff --git a/net/sunrpc/timer.c b/net/sunrpc/timer.c
index bcbdf6430d5c..8142fdb8a930 100644
--- a/net/sunrpc/timer.c
+++ b/net/sunrpc/timer.c
@@ -19,8 +19,6 @@
 #include <linux/unistd.h>
 
 #include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/xprt.h>
-#include <linux/sunrpc/timer.h>
 
 #define RPC_RTO_MAX (60*HZ)
 #define RPC_RTO_INIT (HZ/5)
-- 
cgit v1.2.3


From edb267a688fcee5335d596752f117a30c7152e44 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:18 -0400
Subject: SUNRPC: add xprt switch API for printing formatted remote peer
 addresses

Add a new method to the transport switch API to provide a way to convert
the opaque contents of xprt->addr to a human-readable string.

Test plan:
Compile kernel with CONFIG_NFS enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h | 11 +++++++
 net/sunrpc/xprtsock.c       | 79 ++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 82 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 84122559fa17..8372ab8fc9b5 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -51,6 +51,14 @@ struct rpc_timeout {
 	unsigned char		to_exponential;
 };
 
+enum rpc_display_format_t {
+	RPC_DISPLAY_ADDR = 0,
+	RPC_DISPLAY_PORT,
+	RPC_DISPLAY_PROTO,
+	RPC_DISPLAY_ALL,
+	RPC_DISPLAY_MAX,
+};
+
 struct rpc_task;
 struct rpc_xprt;
 struct seq_file;
@@ -103,6 +111,7 @@ struct rpc_rqst {
 
 struct rpc_xprt_ops {
 	void		(*set_buffer_size)(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize);
+	char *		(*print_addr)(struct rpc_xprt *xprt, enum rpc_display_format_t format);
 	int		(*reserve_xprt)(struct rpc_task *task);
 	void		(*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task);
 	void		(*rpcbind)(struct rpc_task *task);
@@ -207,6 +216,8 @@ struct rpc_xprt {
 	void			(*old_data_ready)(struct sock *, int);
 	void			(*old_state_change)(struct sock *);
 	void			(*old_write_space)(struct sock *);
+
+	char *			address_strings[RPC_DISPLAY_MAX];
 };
 
 #define XPRT_LAST_FRAG		(1 << 0)
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 4c98b89a5b48..cb8e6c34e12f 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -125,6 +125,47 @@ static inline void xs_pktdump(char *msg, u32 *packet, unsigned int count)
 }
 #endif
 
+static void xs_format_peer_addresses(struct rpc_xprt *xprt)
+{
+	struct sockaddr_in *addr = (struct sockaddr_in *) &xprt->addr;
+	char *buf;
+
+	buf = kzalloc(20, GFP_KERNEL);
+	if (buf) {
+		snprintf(buf, 20, "%u.%u.%u.%u",
+				NIPQUAD(addr->sin_addr.s_addr));
+	}
+	xprt->address_strings[RPC_DISPLAY_ADDR] = buf;
+
+	buf = kzalloc(8, GFP_KERNEL);
+	if (buf) {
+		snprintf(buf, 8, "%u",
+				ntohs(addr->sin_port));
+	}
+	xprt->address_strings[RPC_DISPLAY_PORT] = buf;
+
+	if (xprt->prot == IPPROTO_UDP)
+		xprt->address_strings[RPC_DISPLAY_PROTO] = "udp";
+	else
+		xprt->address_strings[RPC_DISPLAY_PROTO] = "tcp";
+
+	buf = kzalloc(48, GFP_KERNEL);
+	if (buf) {
+		snprintf(buf, 48, "addr=%u.%u.%u.%u port=%u proto=%s",
+			NIPQUAD(addr->sin_addr.s_addr),
+			ntohs(addr->sin_port),
+			xprt->prot == IPPROTO_UDP ? "udp" : "tcp");
+	}
+	xprt->address_strings[RPC_DISPLAY_ALL] = buf;
+}
+
+static void xs_free_peer_addresses(struct rpc_xprt *xprt)
+{
+	kfree(xprt->address_strings[RPC_DISPLAY_ADDR]);
+	kfree(xprt->address_strings[RPC_DISPLAY_PORT]);
+	kfree(xprt->address_strings[RPC_DISPLAY_ALL]);
+}
+
 #define XS_SENDMSG_FLAGS	(MSG_DONTWAIT | MSG_NOSIGNAL)
 
 static inline int xs_send_head(struct socket *sock, struct sockaddr *addr, int addrlen, struct xdr_buf *xdr, unsigned int base, unsigned int len)
@@ -490,6 +531,7 @@ static void xs_destroy(struct rpc_xprt *xprt)
 
 	xprt_disconnect(xprt);
 	xs_close(xprt);
+	xs_free_peer_addresses(xprt);
 	kfree(xprt->slot);
 }
 
@@ -964,6 +1006,19 @@ static unsigned short xs_get_random_port(void)
 	return rand + xprt_min_resvport;
 }
 
+/**
+ * xs_print_peer_address - format an IPv4 address for printing
+ * @xprt: generic transport
+ * @format: flags field indicating which parts of the address to render
+ */
+static char *xs_print_peer_address(struct rpc_xprt *xprt, enum rpc_display_format_t format)
+{
+	if (xprt->address_strings[format] != NULL)
+		return xprt->address_strings[format];
+	else
+		return "unprintable";
+}
+
 /**
  * xs_set_port - reset the port number in the remote endpoint address
  * @xprt: generic transport
@@ -1019,8 +1074,6 @@ static void xs_udp_connect_worker(void *args)
 	if (xprt->shutdown || !xprt_bound(xprt))
 		goto out;
 
-	dprintk("RPC:      xs_udp_connect_worker for xprt %p\n", xprt);
-
 	/* Start by resetting any existing state */
 	xs_close(xprt);
 
@@ -1034,6 +1087,9 @@ static void xs_udp_connect_worker(void *args)
 		goto out;
 	}
 
+	dprintk("RPC:      worker connecting xprt %p to address: %s\n",
+			xprt, xs_print_peer_address(xprt, RPC_DISPLAY_ALL));
+
 	if (!xprt->inet) {
 		struct sock *sk = sock->sk;
 
@@ -1102,8 +1158,6 @@ static void xs_tcp_connect_worker(void *args)
 	if (xprt->shutdown || !xprt_bound(xprt))
 		goto out;
 
-	dprintk("RPC:      xs_tcp_connect_worker for xprt %p\n", xprt);
-
 	if (!xprt->sock) {
 		/* start from scratch */
 		if ((err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) {
@@ -1119,6 +1173,9 @@ static void xs_tcp_connect_worker(void *args)
 		/* "close" the socket, preserving the local port */
 		xs_tcp_reuse_connection(xprt);
 
+	dprintk("RPC:      worker connecting xprt %p to address: %s\n",
+			xprt, xs_print_peer_address(xprt, RPC_DISPLAY_ALL));
+
 	if (!xprt->inet) {
 		struct sock *sk = sock->sk;
 
@@ -1260,6 +1317,7 @@ static void xs_tcp_print_stats(struct rpc_xprt *xprt, struct seq_file *seq)
 
 static struct rpc_xprt_ops xs_udp_ops = {
 	.set_buffer_size	= xs_udp_set_buffer_size,
+	.print_addr		= xs_print_peer_address,
 	.reserve_xprt		= xprt_reserve_xprt_cong,
 	.release_xprt		= xprt_release_xprt_cong,
 	.rpcbind		= rpc_getport,
@@ -1277,6 +1335,7 @@ static struct rpc_xprt_ops xs_udp_ops = {
 };
 
 static struct rpc_xprt_ops xs_tcp_ops = {
+	.print_addr		= xs_print_peer_address,
 	.reserve_xprt		= xprt_reserve_xprt,
 	.release_xprt		= xs_tcp_release_xprt,
 	.rpcbind		= rpc_getport,
@@ -1301,8 +1360,6 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 {
 	size_t slot_table_size;
 
-	dprintk("RPC:      setting up udp-ipv4 transport...\n");
-
 	xprt->max_reqs = xprt_udp_slot_table_entries;
 	slot_table_size = xprt->max_reqs * sizeof(xprt->slot[0]);
 	xprt->slot = kzalloc(slot_table_size, GFP_KERNEL);
@@ -1332,6 +1389,10 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 	else
 		xprt_set_timeout(&xprt->timeout, 5, 5 * HZ);
 
+	xs_format_peer_addresses(xprt);
+	dprintk("RPC:      set up transport to address %s\n",
+			xs_print_peer_address(xprt, RPC_DISPLAY_ALL));
+
 	return 0;
 }
 
@@ -1345,8 +1406,6 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 {
 	size_t slot_table_size;
 
-	dprintk("RPC:      setting up tcp-ipv4 transport...\n");
-
 	xprt->max_reqs = xprt_tcp_slot_table_entries;
 	slot_table_size = xprt->max_reqs * sizeof(xprt->slot[0]);
 	xprt->slot = kzalloc(slot_table_size, GFP_KERNEL);
@@ -1375,5 +1434,9 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 	else
 		xprt_set_timeout(&xprt->timeout, 2, 60 * HZ);
 
+	xs_format_peer_addresses(xprt);
+	dprintk("RPC:      set up transport to address %s\n",
+			xs_print_peer_address(xprt, RPC_DISPLAY_ALL));
+
 	return 0;
 }
-- 
cgit v1.2.3


From f425eba437f0051bde979ea2eef8bc875a77cd00 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:18 -0400
Subject: SUNRPC: Create API for displaying remote peer address

Provide an API for formatting the remote peer address for printing without
exposing its internal structure.  The address could be dynamic, so we
support a function call to get the address rather than reading it straight
out of a structure.

Test-plan:
Destructive testing (unplugging the network temporarily).  Probably need
to rig a server where certain services aren't running, or that returns an
error for some typical operation.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  1 +
 net/sunrpc/clnt.c           | 13 +++++++++++++
 2 files changed, 14 insertions(+)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 65196b03f0ab..b7d47f018353 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -124,6 +124,7 @@ size_t		rpc_max_payload(struct rpc_clnt *);
 void		rpc_force_rebind(struct rpc_clnt *);
 int		rpc_ping(struct rpc_clnt *clnt, int flags);
 size_t		rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t);
+char *		rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t);
 
 /*
  * Helper function for NFSroot support
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 94768cf5fd5b..e5b19e348d88 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -554,6 +554,19 @@ size_t rpc_peeraddr(struct rpc_clnt *clnt, struct sockaddr *buf, size_t bufsize)
 }
 EXPORT_SYMBOL(rpc_peeraddr);
 
+/**
+ * rpc_peeraddr2str - return remote peer address in printable format
+ * @clnt: RPC client structure
+ * @format: address format
+ *
+ */
+char *rpc_peeraddr2str(struct rpc_clnt *clnt, enum rpc_display_format_t format)
+{
+	struct rpc_xprt *xprt = clnt->cl_xprt;
+	return xprt->ops->print_addr(xprt, format);
+}
+EXPORT_SYMBOL(rpc_peeraddr2str);
+
 void
 rpc_setbufsize(struct rpc_clnt *clnt, unsigned int sndsize, unsigned int rcvsize)
 {
-- 
cgit v1.2.3


From c4efcb1d3e0bc76aeb9ca6301d19a5079893c6c9 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:19 -0400
Subject: SUNRPC: Use "sockaddr_storage" for storing RPC client's remote peer
 address

IPv6 addresses are big (128 bytes).  Now that no RPC client consumers treat
the addr field in rpc_xprt structs as an opaque, and access it only via the
API calls, we can safely widen the field in the rpc_xprt struct to
accomodate larger addresses.

Test plan:
Compile kernel with CONFIG_NFS enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h |  3 ++-
 net/sunrpc/clnt.c           |  2 +-
 net/sunrpc/xprt.c           |  3 ++-
 net/sunrpc/xprtsock.c       | 15 ++++++++++-----
 4 files changed, 15 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 8372ab8fc9b5..fc05cfbd5805 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -134,7 +134,8 @@ struct rpc_xprt {
 	struct sock *		inet;		/* INET layer */
 
 	struct rpc_timeout	timeout;	/* timeout parms */
-	struct sockaddr_in	addr;		/* server address */
+	struct sockaddr_storage	addr;		/* server address */
+	size_t			addrlen;	/* size of server address */
 	int			prot;		/* IP protocol */
 
 	unsigned long		cong;		/* current congestion */
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index e5b19e348d88..ff1e90fd81ab 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -550,7 +550,7 @@ size_t rpc_peeraddr(struct rpc_clnt *clnt, struct sockaddr *buf, size_t bufsize)
 	if (bytes > bufsize)
 		bytes = bufsize;
 	memcpy(buf, &clnt->cl_xprt->addr, bytes);
-	return sizeof(xprt->addr);
+	return xprt->addrlen;
 }
 EXPORT_SYMBOL(rpc_peeraddr);
 
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index b45abd0743cb..4987517cc74b 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -896,7 +896,8 @@ static struct rpc_xprt *xprt_setup(int proto, struct sockaddr_in *ap, struct rpc
 	if ((xprt = kzalloc(sizeof(struct rpc_xprt), GFP_KERNEL)) == NULL)
 		return ERR_PTR(-ENOMEM);
 
-	xprt->addr = *ap;
+	memcpy(&xprt->addr, ap, sizeof(*ap));
+	xprt->addrlen = sizeof(*ap);
 
 	switch (proto) {
 	case IPPROTO_UDP:
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index cb8e6c34e12f..17179aa4c207 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -341,7 +341,7 @@ static int xs_udp_send_request(struct rpc_task *task)
 
 	req->rq_xtime = jiffies;
 	status = xs_sendpages(xprt->sock, (struct sockaddr *) &xprt->addr,
-				sizeof(xprt->addr), xdr, req->rq_bytes_sent);
+				xprt->addrlen, xdr, req->rq_bytes_sent);
 
 	dprintk("RPC:      xs_udp_send_request(%u) = %d\n",
 			xdr->len - req->rq_bytes_sent, status);
@@ -1027,8 +1027,11 @@ static char *xs_print_peer_address(struct rpc_xprt *xprt, enum rpc_display_forma
  */
 static void xs_set_port(struct rpc_xprt *xprt, unsigned short port)
 {
+	struct sockaddr_in *sap = (struct sockaddr_in *) &xprt->addr;
+
 	dprintk("RPC:      setting port for xprt %p to %u\n", xprt, port);
-	xprt->addr.sin_port = htons(port);
+
+	sap->sin_port = htons(port);
 }
 
 static int xs_bindresvport(struct rpc_xprt *xprt, struct socket *sock)
@@ -1209,7 +1212,7 @@ static void xs_tcp_connect_worker(void *args)
 	xprt->stat.connect_count++;
 	xprt->stat.connect_start = jiffies;
 	status = sock->ops->connect(sock, (struct sockaddr *) &xprt->addr,
-			sizeof(xprt->addr), O_NONBLOCK);
+			xprt->addrlen, O_NONBLOCK);
 	dprintk("RPC: %p  connect status %d connected %d sock state %d\n",
 			xprt, -status, xprt_connected(xprt), sock->sk->sk_state);
 	if (status < 0) {
@@ -1359,6 +1362,7 @@ static struct rpc_xprt_ops xs_tcp_ops = {
 int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 {
 	size_t slot_table_size;
+	struct sockaddr_in *addr = (struct sockaddr_in *) &xprt->addr;
 
 	xprt->max_reqs = xprt_udp_slot_table_entries;
 	slot_table_size = xprt->max_reqs * sizeof(xprt->slot[0]);
@@ -1366,7 +1370,7 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 	if (xprt->slot == NULL)
 		return -ENOMEM;
 
-	if (ntohs(xprt->addr.sin_port) != 0)
+	if (ntohs(addr->sin_port != 0))
 		xprt_set_bound(xprt);
 	xprt->port = xs_get_random_port();
 
@@ -1405,6 +1409,7 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 {
 	size_t slot_table_size;
+	struct sockaddr_in *addr = (struct sockaddr_in *) &xprt->addr;
 
 	xprt->max_reqs = xprt_tcp_slot_table_entries;
 	slot_table_size = xprt->max_reqs * sizeof(xprt->slot[0]);
@@ -1412,7 +1417,7 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 	if (xprt->slot == NULL)
 		return -ENOMEM;
 
-	if (ntohs(xprt->addr.sin_port) != 0)
+	if (ntohs(addr->sin_port) != 0)
 		xprt_set_bound(xprt);
 	xprt->port = xs_get_random_port();
 
-- 
cgit v1.2.3


From 6ca948238724c945bd353f51d54ae7d285f3889f Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:19 -0400
Subject: SUNRPC: Clean-up after previous patches.

Remove some unused macros related to accessing an RPC peer address

Test plan:
Compile kernel with CONFIG_NFS option enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/host.c             | 1 -
 include/linux/nfs_fs.h      | 1 -
 include/linux/sunrpc/clnt.h | 3 ---
 3 files changed, 5 deletions(-)

(limited to 'include')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 38b0e8a1aec0..a516a01561b8 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -26,7 +26,6 @@
 #define NLM_HOST_REBIND		(60 * HZ)
 #define NLM_HOST_EXPIRE		((nrhosts > NLM_HOST_MAX)? 300 * HZ : 120 * HZ)
 #define NLM_HOST_COLLECT	((nrhosts > NLM_HOST_MAX)? 120 * HZ :  60 * HZ)
-#define NLM_HOST_ADDR(sv)	(&(sv)->s_nlmclnt->cl_xprt->addr)
 
 static struct nlm_host *	nlm_hosts[NLM_HOST_NRHASH];
 static unsigned long		next_gc;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 51e9bd90dedc..3b5b04193fee 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -216,7 +216,6 @@ static inline struct nfs_inode *NFS_I(struct inode *inode)
 #define NFS_SERVER(inode)		(NFS_SB(inode->i_sb))
 #define NFS_CLIENT(inode)		(NFS_SERVER(inode)->client)
 #define NFS_PROTO(inode)		(NFS_SERVER(inode)->nfs_client->rpc_ops)
-#define NFS_ADDR(inode)			(RPC_PEERADDR(NFS_CLIENT(inode)))
 #define NFS_COOKIEVERF(inode)		(NFS_I(inode)->cookieverf)
 #define NFS_READTIME(inode)		(NFS_I(inode)->read_cache_jiffies)
 #define NFS_CHANGE_ATTR(inode)		(NFS_I(inode)->change_attr)
diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index b7d47f018353..a26d69583c7a 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -89,9 +89,6 @@ struct rpc_procinfo {
 	char *			p_name;		/* name of procedure */
 };
 
-#define RPC_CONGESTED(clnt)	(RPCXPRT_CONGESTED((clnt)->cl_xprt))
-#define RPC_PEERADDR(clnt)	(&(clnt)->cl_xprt->addr)
-
 #ifdef __KERNEL__
 
 struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
-- 
cgit v1.2.3


From c2866763b4029411d166040306691773c12d4caf Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:20 -0400
Subject: SUNRPC: use sockaddr + size when creating remote transport endpoints

Prepare for more generic transport endpoint handling needed by transports
that might use different forms of addressing, such as IPv6.

Introduce a single function call to replace the two-call
xprt_create_proto/rpc_create_client API.  Define a new rpc_create_args
structure that allows callers to pass in remote endpoint addresses of
varying length.

Test-plan:
Compile kernel with CONFIG_NFS enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h | 22 +++++++++++++
 include/linux/sunrpc/xprt.h |  1 +
 net/sunrpc/clnt.c           | 61 ++++++++++++++++++++++++++++++++++++
 net/sunrpc/xprt.c           | 75 +++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 159 insertions(+)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index a26d69583c7a..7817ba82f1b2 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -97,6 +97,28 @@ struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
 struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname,
 				struct rpc_program *info,
 				u32 version, rpc_authflavor_t authflavor);
+
+struct rpc_create_args {
+	int			protocol;
+	struct sockaddr		*address;
+	size_t			addrsize;
+	struct rpc_timeout	*timeout;
+	char			*servername;
+	struct rpc_program	*program;
+	u32			version;
+	rpc_authflavor_t	authflavor;
+	unsigned long		flags;
+};
+
+/* Values for "flags" field */
+#define RPC_CLNT_CREATE_HARDRTRY	(1UL << 0)
+#define RPC_CLNT_CREATE_INTR		(1UL << 1)
+#define RPC_CLNT_CREATE_AUTOBIND	(1UL << 2)
+#define RPC_CLNT_CREATE_ONESHOT		(1UL << 3)
+#define RPC_CLNT_CREATE_NONPRIVPORT	(1UL << 4)
+#define RPC_CLNT_CREATE_NOPING		(1UL << 5)
+
+struct rpc_clnt *rpc_create(struct rpc_create_args *args);
 struct rpc_clnt	*rpc_bind_new_program(struct rpc_clnt *,
 				struct rpc_program *, int);
 struct rpc_clnt *rpc_clone_client(struct rpc_clnt *);
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index fc05cfbd5805..bc80fcfdd892 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -237,6 +237,7 @@ void			xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long
 /*
  * Generic internal transport functions
  */
+struct rpc_xprt *	xprt_create_transport(int proto, struct sockaddr *addr, size_t size, struct rpc_timeout *toparms);
 void			xprt_connect(struct rpc_task *task);
 void			xprt_reserve(struct rpc_task *task);
 int			xprt_reserve_xprt(struct rpc_task *task);
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index ff1e90fd81ab..dbb93bdf6cc9 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -192,6 +192,67 @@ out_no_xprt:
 	return ERR_PTR(err);
 }
 
+/*
+ * rpc_create - create an RPC client and transport with one call
+ * @args: rpc_clnt create argument structure
+ *
+ * Creates and initializes an RPC transport and an RPC client.
+ *
+ * It can ping the server in order to determine if it is up, and to see if
+ * it supports this program and version.  RPC_CLNT_CREATE_NOPING disables
+ * this behavior so asynchronous tasks can also use rpc_create.
+ */
+struct rpc_clnt *rpc_create(struct rpc_create_args *args)
+{
+	struct rpc_xprt *xprt;
+	struct rpc_clnt *clnt;
+
+	xprt = xprt_create_transport(args->protocol, args->address,
+					args->addrsize, args->timeout);
+	if (IS_ERR(xprt))
+		return (struct rpc_clnt *)xprt;
+
+	/*
+	 * By default, kernel RPC client connects from a reserved port.
+	 * CAP_NET_BIND_SERVICE will not be set for unprivileged requesters,
+	 * but it is always enabled for rpciod, which handles the connect
+	 * operation.
+	 */
+	xprt->resvport = 1;
+	if (args->flags & RPC_CLNT_CREATE_NONPRIVPORT)
+		xprt->resvport = 0;
+
+	dprintk("RPC:       creating %s client for %s (xprt %p)\n",
+		args->program->name, args->servername, xprt);
+
+	clnt = rpc_new_client(xprt, args->servername, args->program,
+				args->version, args->authflavor);
+	if (IS_ERR(clnt))
+		return clnt;
+
+	if (!(args->flags & RPC_CLNT_CREATE_NOPING)) {
+		int err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR);
+		if (err != 0) {
+			rpc_shutdown_client(clnt);
+			return ERR_PTR(err);
+		}
+	}
+
+	clnt->cl_softrtry = 1;
+	if (args->flags & RPC_CLNT_CREATE_HARDRTRY)
+		clnt->cl_softrtry = 0;
+
+	if (args->flags & RPC_CLNT_CREATE_INTR)
+		clnt->cl_intr = 1;
+	if (args->flags & RPC_CLNT_CREATE_AUTOBIND)
+		clnt->cl_autobind = 1;
+	if (args->flags & RPC_CLNT_CREATE_ONESHOT)
+		clnt->cl_oneshot = 1;
+
+	return clnt;
+}
+EXPORT_SYMBOL(rpc_create);
+
 /**
  * Create an RPC client
  * @xprt - pointer to xprt struct
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 4987517cc74b..17f56cfe2412 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -887,6 +887,81 @@ void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long i
 	to->to_exponential = 0;
 }
 
+/**
+ * xprt_create_transport - create an RPC transport
+ * @proto: requested transport protocol
+ * @ap: remote peer address
+ * @size: length of address
+ * @to: timeout parameters
+ *
+ */
+struct rpc_xprt *xprt_create_transport(int proto, struct sockaddr *ap, size_t size, struct rpc_timeout *to)
+{
+	int result;
+	struct rpc_xprt	*xprt;
+	struct rpc_rqst	*req;
+
+	if ((xprt = kzalloc(sizeof(struct rpc_xprt), GFP_KERNEL)) == NULL) {
+		dprintk("RPC:      xprt_create_transport: no memory\n");
+		return ERR_PTR(-ENOMEM);
+	}
+	if (size <= sizeof(xprt->addr)) {
+		memcpy(&xprt->addr, ap, size);
+		xprt->addrlen = size;
+	} else {
+		kfree(xprt);
+		dprintk("RPC:      xprt_create_transport: address too large\n");
+		return ERR_PTR(-EBADF);
+	}
+
+	switch (proto) {
+	case IPPROTO_UDP:
+		result = xs_setup_udp(xprt, to);
+		break;
+	case IPPROTO_TCP:
+		result = xs_setup_tcp(xprt, to);
+		break;
+	default:
+		printk(KERN_ERR "RPC: unrecognized transport protocol: %d\n",
+				proto);
+		return ERR_PTR(-EIO);
+	}
+	if (result) {
+		kfree(xprt);
+		dprintk("RPC:      xprt_create_transport: failed, %d\n", result);
+		return ERR_PTR(result);
+	}
+
+	spin_lock_init(&xprt->transport_lock);
+	spin_lock_init(&xprt->reserve_lock);
+
+	INIT_LIST_HEAD(&xprt->free);
+	INIT_LIST_HEAD(&xprt->recv);
+	INIT_WORK(&xprt->task_cleanup, xprt_autoclose, xprt);
+	init_timer(&xprt->timer);
+	xprt->timer.function = xprt_init_autodisconnect;
+	xprt->timer.data = (unsigned long) xprt;
+	xprt->last_used = jiffies;
+	xprt->cwnd = RPC_INITCWND;
+
+	rpc_init_wait_queue(&xprt->binding, "xprt_binding");
+	rpc_init_wait_queue(&xprt->pending, "xprt_pending");
+	rpc_init_wait_queue(&xprt->sending, "xprt_sending");
+	rpc_init_wait_queue(&xprt->resend, "xprt_resend");
+	rpc_init_priority_wait_queue(&xprt->backlog, "xprt_backlog");
+
+	/* initialize free list */
+	for (req = &xprt->slot[xprt->max_reqs-1]; req >= &xprt->slot[0]; req--)
+		list_add(&req->rq_list, &xprt->free);
+
+	xprt_init_xid(xprt);
+
+	dprintk("RPC:      created transport %p with %u slots\n", xprt,
+			xprt->max_reqs);
+
+	return xprt;
+}
+
 static struct rpc_xprt *xprt_setup(int proto, struct sockaddr_in *ap, struct rpc_timeout *to)
 {
 	int result;
-- 
cgit v1.2.3


From ff9aa5e56df60cc8565a93cc868fe25ae3f20e49 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:21 -0400
Subject: SUNRPC: Eliminate xprt_create_proto and rpc_create_client

The two function call API for creating a new RPC client is now obsolete.
Remove it.

Also, remove an unnecessary check to see whether the caller is capable of
using privileged network services.  The kernel RPC client always uses a
privileged ephemeral port by default; callers are responsible for checking
the authority of users to make use of any RPC service, or for specifying
that a nonprivileged port is acceptable.

Test plan:
Repeated runs of Connectathon locking suite.  Check network trace to ensure
correctness of NLM requests and replies.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |  7 ----
 include/linux/sunrpc/xprt.h |  1 -
 net/sunrpc/clnt.c           | 42 +-----------------------
 net/sunrpc/sunrpc_syms.c    |  3 --
 net/sunrpc/xprt.c           | 79 ---------------------------------------------
 net/sunrpc/xprtsock.c       |  2 --
 6 files changed, 1 insertion(+), 133 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 7817ba82f1b2..f6d1d646ce05 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -91,13 +91,6 @@ struct rpc_procinfo {
 
 #ifdef __KERNEL__
 
-struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
-				struct rpc_program *info,
-				u32 version, rpc_authflavor_t authflavor);
-struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname,
-				struct rpc_program *info,
-				u32 version, rpc_authflavor_t authflavor);
-
 struct rpc_create_args {
 	int			protocol;
 	struct sockaddr		*address;
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index bc80fcfdd892..de4efea7c856 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -231,7 +231,6 @@ struct rpc_xprt {
 /*
  * Transport operations used by ULPs
  */
-struct rpc_xprt *	xprt_create_proto(int proto, struct sockaddr_in *addr, struct rpc_timeout *to);
 void			xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long incr);
 
 /*
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index dbb93bdf6cc9..428704dd5b3e 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -97,17 +97,7 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
 	}
 }
 
-/*
- * Create an RPC client
- * FIXME: This should also take a flags argument (as in task->tk_flags).
- * It's called (among others) from pmap_create_client, which may in
- * turn be called by an async task. In this case, rpciod should not be
- * made to sleep too long.
- */
-struct rpc_clnt *
-rpc_new_client(struct rpc_xprt *xprt, char *servname,
-		  struct rpc_program *program, u32 vers,
-		  rpc_authflavor_t flavor)
+static struct rpc_clnt * rpc_new_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *program, u32 vers, rpc_authflavor_t flavor)
 {
 	struct rpc_version	*version;
 	struct rpc_clnt		*clnt = NULL;
@@ -253,36 +243,6 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 }
 EXPORT_SYMBOL(rpc_create);
 
-/**
- * Create an RPC client
- * @xprt - pointer to xprt struct
- * @servname - name of server
- * @info - rpc_program
- * @version - rpc_program version
- * @authflavor - rpc_auth flavour to use
- *
- * Creates an RPC client structure, then pings the server in order to
- * determine if it is up, and if it supports this program and version.
- *
- * This function should never be called by asynchronous tasks such as
- * the portmapper.
- */
-struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname,
-		struct rpc_program *info, u32 version, rpc_authflavor_t authflavor)
-{
-	struct rpc_clnt *clnt;
-	int err;
-	
-	clnt = rpc_new_client(xprt, servname, info, version, authflavor);
-	if (IS_ERR(clnt))
-		return clnt;
-	err = rpc_ping(clnt, RPC_TASK_SOFT|RPC_TASK_NOINTR);
-	if (err == 0)
-		return clnt;
-	rpc_shutdown_client(clnt);
-	return ERR_PTR(err);
-}
-
 /*
  * This function clones the RPC client structure. It allows us to share the
  * same transport while varying parameters such as the authentication
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index f38f939ce95f..26c0531d7e25 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -36,8 +36,6 @@ EXPORT_SYMBOL(rpc_wake_up_status);
 EXPORT_SYMBOL(rpc_release_task);
 
 /* RPC client functions */
-EXPORT_SYMBOL(rpc_create_client);
-EXPORT_SYMBOL(rpc_new_client);
 EXPORT_SYMBOL(rpc_clone_client);
 EXPORT_SYMBOL(rpc_bind_new_program);
 EXPORT_SYMBOL(rpc_destroy_client);
@@ -57,7 +55,6 @@ EXPORT_SYMBOL(rpc_queue_upcall);
 EXPORT_SYMBOL(rpc_mkpipe);
 
 /* Client transport */
-EXPORT_SYMBOL(xprt_create_proto);
 EXPORT_SYMBOL(xprt_set_timeout);
 
 /* Client credential cache */
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 17f56cfe2412..e4f64fb58ff2 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -962,85 +962,6 @@ struct rpc_xprt *xprt_create_transport(int proto, struct sockaddr *ap, size_t si
 	return xprt;
 }
 
-static struct rpc_xprt *xprt_setup(int proto, struct sockaddr_in *ap, struct rpc_timeout *to)
-{
-	int result;
-	struct rpc_xprt	*xprt;
-	struct rpc_rqst	*req;
-
-	if ((xprt = kzalloc(sizeof(struct rpc_xprt), GFP_KERNEL)) == NULL)
-		return ERR_PTR(-ENOMEM);
-
-	memcpy(&xprt->addr, ap, sizeof(*ap));
-	xprt->addrlen = sizeof(*ap);
-
-	switch (proto) {
-	case IPPROTO_UDP:
-		result = xs_setup_udp(xprt, to);
-		break;
-	case IPPROTO_TCP:
-		result = xs_setup_tcp(xprt, to);
-		break;
-	default:
-		printk(KERN_ERR "RPC: unrecognized transport protocol: %d\n",
-				proto);
-		result = -EIO;
-		break;
-	}
-	if (result) {
-		kfree(xprt);
-		return ERR_PTR(result);
-	}
-
-	spin_lock_init(&xprt->transport_lock);
-	spin_lock_init(&xprt->reserve_lock);
-
-	INIT_LIST_HEAD(&xprt->free);
-	INIT_LIST_HEAD(&xprt->recv);
-	INIT_WORK(&xprt->task_cleanup, xprt_autoclose, xprt);
-	init_timer(&xprt->timer);
-	xprt->timer.function = xprt_init_autodisconnect;
-	xprt->timer.data = (unsigned long) xprt;
-	xprt->last_used = jiffies;
-	xprt->cwnd = RPC_INITCWND;
-
-	rpc_init_wait_queue(&xprt->binding, "xprt_binding");
-	rpc_init_wait_queue(&xprt->pending, "xprt_pending");
-	rpc_init_wait_queue(&xprt->sending, "xprt_sending");
-	rpc_init_wait_queue(&xprt->resend, "xprt_resend");
-	rpc_init_priority_wait_queue(&xprt->backlog, "xprt_backlog");
-
-	/* initialize free list */
-	for (req = &xprt->slot[xprt->max_reqs-1]; req >= &xprt->slot[0]; req--)
-		list_add(&req->rq_list, &xprt->free);
-
-	xprt_init_xid(xprt);
-
-	dprintk("RPC:      created transport %p with %u slots\n", xprt,
-			xprt->max_reqs);
-	
-	return xprt;
-}
-
-/**
- * xprt_create_proto - create an RPC client transport
- * @proto: requested transport protocol
- * @sap: remote peer's address
- * @to: timeout parameters for new transport
- *
- */
-struct rpc_xprt *xprt_create_proto(int proto, struct sockaddr_in *sap, struct rpc_timeout *to)
-{
-	struct rpc_xprt	*xprt;
-
-	xprt = xprt_setup(proto, sap, to);
-	if (IS_ERR(xprt))
-		dprintk("RPC:      xprt_create_proto failed\n");
-	else
-		dprintk("RPC:      xprt_create_proto created xprt %p\n", xprt);
-	return xprt;
-}
-
 /**
  * xprt_destroy - destroy an RPC transport, killing off all requests.
  * @xprt: transport to destroy
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 17179aa4c207..0b84fab68d7e 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1376,7 +1376,6 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 
 	xprt->prot = IPPROTO_UDP;
 	xprt->tsh_size = 0;
-	xprt->resvport = capable(CAP_NET_BIND_SERVICE) ? 1 : 0;
 	/* XXX: header size can vary due to auth type, IPv6, etc. */
 	xprt->max_payload = (1U << 16) - (MAX_HEADER << 3);
 
@@ -1423,7 +1422,6 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to)
 
 	xprt->prot = IPPROTO_TCP;
 	xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32);
-	xprt->resvport = capable(CAP_NET_BIND_SERVICE) ? 1 : 0;
 	xprt->max_payload = RPC_MAX_FRAGMENT_SIZE;
 
 	INIT_WORK(&xprt->connect_worker, xs_tcp_connect_worker, xprt);
-- 
cgit v1.2.3


From 4f390c152bc87165da4b1f5b7d870b46fb106d4e Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:22 -0400
Subject: NFS: Fix double d_drop in nfs_instantiate() error path

If the LOOKUP or GETATTR in nfs_instantiate fail, nfs_instantiate will do a
d_drop before returning.  But some callers already do a d_drop in the case
of an error return.  Make certain we do only one d_drop in all error paths.

This issue was introduced because over time, the symlink proc API diverged
slightly from the create/mkdir/mknod proc API.  To prevent other coding
mistakes of this type, change the symlink proc API to be more like
create/mkdir/mknod and move the nfs_instantiate call into the symlink proc
routines so it is used in exactly the same way for create, mkdir, mknod,
and symlink.

Test plan:
Connectathon, all versions of NFS.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c            | 16 ++++------------
 fs/nfs/nfs3proc.c       | 26 ++++++++++++++++----------
 fs/nfs/nfs4proc.c       | 31 ++++++++++++++++---------------
 fs/nfs/proc.c           | 29 +++++++++++++++++++++--------
 include/linux/nfs_xdr.h |  5 ++---
 5 files changed, 59 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 084e8cb41c84..affd3ae52e55 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1147,23 +1147,20 @@ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle,
 		struct inode *dir = dentry->d_parent->d_inode;
 		error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr);
 		if (error)
-			goto out_err;
+			return error;
 	}
 	if (!(fattr->valid & NFS_ATTR_FATTR)) {
 		struct nfs_server *server = NFS_SB(dentry->d_sb);
 		error = server->nfs_client->rpc_ops->getattr(server, fhandle, fattr);
 		if (error < 0)
-			goto out_err;
+			return error;
 	}
 	inode = nfs_fhget(dentry->d_sb, fhandle, fattr);
 	error = PTR_ERR(inode);
 	if (IS_ERR(inode))
-		goto out_err;
+		return error;
 	d_instantiate(dentry, inode);
 	return 0;
-out_err:
-	d_drop(dentry);
-	return error;
 }
 
 /*
@@ -1448,8 +1445,6 @@ static int
 nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 {
 	struct iattr attr;
-	struct nfs_fattr sym_attr;
-	struct nfs_fh sym_fh;
 	struct qstr qsymname;
 	int error;
 
@@ -1473,12 +1468,9 @@ dentry->d_parent->d_name.name, dentry->d_name.name);
 
 	lock_kernel();
 	nfs_begin_data_update(dir);
-	error = NFS_PROTO(dir)->symlink(dir, &dentry->d_name, &qsymname,
-					  &attr, &sym_fh, &sym_attr);
+	error = NFS_PROTO(dir)->symlink(dir, dentry, &qsymname, &attr);
 	nfs_end_data_update(dir);
 	if (!error)
-		error = nfs_instantiate(dentry, &sym_fh, &sym_attr);
-	else
 		d_drop(dentry);
 	unlock_kernel();
 	return error;
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 9e8258ece6fd..d85ac427c326 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -544,23 +544,23 @@ nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 }
 
 static int
-nfs3_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path,
-		  struct iattr *sattr, struct nfs_fh *fhandle,
-		  struct nfs_fattr *fattr)
+nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
+		  struct iattr *sattr)
 {
-	struct nfs_fattr	dir_attr;
+	struct nfs_fh fhandle;
+	struct nfs_fattr fattr, dir_attr;
 	struct nfs3_symlinkargs	arg = {
 		.fromfh		= NFS_FH(dir),
-		.fromname	= name->name,
-		.fromlen	= name->len,
+		.fromname	= dentry->d_name.name,
+		.fromlen	= dentry->d_name.len,
 		.topath		= path->name,
 		.tolen		= path->len,
 		.sattr		= sattr
 	};
 	struct nfs3_diropres	res = {
 		.dir_attr	= &dir_attr,
-		.fh		= fhandle,
-		.fattr		= fattr
+		.fh		= &fhandle,
+		.fattr		= &fattr
 	};
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs3_procedures[NFS3PROC_SYMLINK],
@@ -571,11 +571,17 @@ nfs3_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path,
 
 	if (path->len > NFS3_MAXPATHLEN)
 		return -ENAMETOOLONG;
-	dprintk("NFS call  symlink %s -> %s\n", name->name, path->name);
+
+	dprintk("NFS call  symlink %s -> %s\n", dentry->d_name.name,
+			path->name);
 	nfs_fattr_init(&dir_attr);
-	nfs_fattr_init(fattr);
+	nfs_fattr_init(&fattr);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_post_op_update_inode(dir, &dir_attr);
+	if (status != 0)
+		goto out;
+	status = nfs_instantiate(dentry, &fhandle, &fattr);
+out:
 	dprintk("NFS reply symlink: %d\n", status);
 	return status;
 }
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a825547e8214..2d18eac6bee5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2084,24 +2084,24 @@ static int nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *n
 	return err;
 }
 
-static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name,
-		struct qstr *path, struct iattr *sattr, struct nfs_fh *fhandle,
-		struct nfs_fattr *fattr)
+static int _nfs4_proc_symlink(struct inode *dir, struct dentry *dentry,
+		struct qstr *path, struct iattr *sattr)
 {
 	struct nfs_server *server = NFS_SERVER(dir);
-	struct nfs_fattr dir_fattr;
+	struct nfs_fh fhandle;
+	struct nfs_fattr fattr, dir_fattr;
 	struct nfs4_create_arg arg = {
 		.dir_fh = NFS_FH(dir),
 		.server = server,
-		.name = name,
+		.name = &dentry->d_name,
 		.attrs = sattr,
 		.ftype = NF4LNK,
 		.bitmask = server->attr_bitmask,
 	};
 	struct nfs4_create_res res = {
 		.server = server,
-		.fh = fhandle,
-		.fattr = fattr,
+		.fh = &fhandle,
+		.fattr = &fattr,
 		.dir_fattr = &dir_fattr,
 	};
 	struct rpc_message msg = {
@@ -2113,27 +2113,28 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name,
 
 	if (path->len > NFS4_MAXPATHLEN)
 		return -ENAMETOOLONG;
+
 	arg.u.symlink = path;
-	nfs_fattr_init(fattr);
+	nfs_fattr_init(&fattr);
 	nfs_fattr_init(&dir_fattr);
 	
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
-	if (!status)
+	if (!status) {
 		update_changeattr(dir, &res.dir_cinfo);
-	nfs_post_op_update_inode(dir, res.dir_fattr);
+		nfs_post_op_update_inode(dir, res.dir_fattr);
+		status = nfs_instantiate(dentry, &fhandle, &fattr);
+	}
 	return status;
 }
 
-static int nfs4_proc_symlink(struct inode *dir, struct qstr *name,
-		struct qstr *path, struct iattr *sattr, struct nfs_fh *fhandle,
-		struct nfs_fattr *fattr)
+static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry,
+		struct qstr *path, struct iattr *sattr)
 {
 	struct nfs4_exception exception = { };
 	int err;
 	do {
 		err = nfs4_handle_exception(NFS_SERVER(dir),
-				_nfs4_proc_symlink(dir, name, path, sattr,
-					fhandle, fattr),
+				_nfs4_proc_symlink(dir, dentry, path, sattr),
 				&exception);
 	} while (exception.retry);
 	return err;
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 5a8b9407ee9a..0b507bf0f330 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -425,14 +425,15 @@ nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 }
 
 static int
-nfs_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path,
-		 struct iattr *sattr, struct nfs_fh *fhandle,
-		 struct nfs_fattr *fattr)
+nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
+		 struct iattr *sattr)
 {
+	struct nfs_fh fhandle;
+	struct nfs_fattr fattr;
 	struct nfs_symlinkargs	arg = {
 		.fromfh		= NFS_FH(dir),
-		.fromname	= name->name,
-		.fromlen	= name->len,
+		.fromname	= dentry->d_name.name,
+		.fromlen	= dentry->d_name.len,
 		.topath		= path->name,
 		.tolen		= path->len,
 		.sattr		= sattr
@@ -445,11 +446,23 @@ nfs_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path,
 
 	if (path->len > NFS2_MAXPATHLEN)
 		return -ENAMETOOLONG;
-	dprintk("NFS call  symlink %s -> %s\n", name->name, path->name);
-	nfs_fattr_init(fattr);
-	fhandle->size = 0;
+
+	dprintk("NFS call  symlink %s -> %s\n", dentry->d_name.name,
+			path->name);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_mark_for_revalidate(dir);
+
+	/*
+	 * V2 SYMLINK requests don't return any attributes.  Setting the
+	 * filehandle size to zero indicates to nfs_instantiate that it
+	 * should fill in the data with a LOOKUP call on the wire.
+	 */
+	if (status == 0) {
+		nfs_fattr_init(&fattr);
+		fhandle.size = 0;
+		status = nfs_instantiate(dentry, &fhandle, &fattr);
+	}
+
 	dprintk("NFS reply symlink: %d\n", status);
 	return status;
 }
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 0f33e621892f..ddf5d75e97a2 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -793,9 +793,8 @@ struct nfs_rpc_ops {
 	int	(*rename)  (struct inode *, struct qstr *,
 			    struct inode *, struct qstr *);
 	int	(*link)    (struct inode *, struct inode *, struct qstr *);
-	int	(*symlink) (struct inode *, struct qstr *, struct qstr *,
-			    struct iattr *, struct nfs_fh *,
-			    struct nfs_fattr *);
+	int	(*symlink) (struct inode *, struct dentry *, struct qstr *,
+			    struct iattr *);
 	int	(*mkdir)   (struct inode *, struct dentry *, struct iattr *);
 	int	(*rmdir)   (struct inode *, struct qstr *);
 	int	(*readdir) (struct dentry *, struct rpc_cred *,
-- 
cgit v1.2.3


From 94a6d75320b3681e6e728b70e18bd186cb55e682 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Tue, 22 Aug 2006 20:06:23 -0400
Subject: NFS: Use cached page as buffer for NFS symlink requests

Now that we have a copy of the symlink path in the page cache, we can pass
a struct page down to the XDR routines instead of a string buffer.

Test plan:
Connectathon, all NFS versions.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/dir.c            |  8 +-------
 fs/nfs/nfs2xdr.c        | 21 ++++++++++++++++++---
 fs/nfs/nfs3proc.c       | 14 +++++++-------
 fs/nfs/nfs3xdr.c        |  7 +++++--
 fs/nfs/nfs4proc.c       | 12 +++++++-----
 fs/nfs/nfs4xdr.c        |  8 ++++----
 fs/nfs/proc.c           | 14 +++++++-------
 include/linux/nfs_xdr.h | 17 ++++++++++-------
 8 files changed, 59 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index b483e5d206cb..51328ae640dd 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1464,10 +1464,6 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym
 	char *kaddr;
 	struct iattr attr;
 	unsigned int pathlen = strlen(symname);
-	struct qstr qsymname = {
-		.name	= symname,
-		.len	= pathlen,
-	};
 	int error;
 
 	dfprintk(VFS, "NFS: symlink(%s/%ld, %s, %s)\n", dir->i_sb->s_id,
@@ -1493,10 +1489,8 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym
 		memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen);
 	kunmap_atomic(kaddr, KM_USER0);
 
-	/* XXX: eventually this will pass in {page, pathlen},
-	 *	instead of qsymname; need XDR changes for that */
 	nfs_begin_data_update(dir);
-	error = NFS_PROTO(dir)->symlink(dir, dentry, &qsymname, &attr);
+	error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr);
 	nfs_end_data_update(dir);
 	if (error != 0) {
 		dfprintk(VFS, "NFS: symlink(%s/%ld, %s, %s) error %d\n",
diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c
index 67391eef6b93..b49501fc0a79 100644
--- a/fs/nfs/nfs2xdr.c
+++ b/fs/nfs/nfs2xdr.c
@@ -51,7 +51,7 @@
 #define NFS_createargs_sz	(NFS_diropargs_sz+NFS_sattr_sz)
 #define NFS_renameargs_sz	(NFS_diropargs_sz+NFS_diropargs_sz)
 #define NFS_linkargs_sz		(NFS_fhandle_sz+NFS_diropargs_sz)
-#define NFS_symlinkargs_sz	(NFS_diropargs_sz+NFS_path_sz+NFS_sattr_sz)
+#define NFS_symlinkargs_sz	(NFS_diropargs_sz+1+NFS_sattr_sz)
 #define NFS_readdirargs_sz	(NFS_fhandle_sz+2)
 
 #define NFS_attrstat_sz		(1+NFS_fattr_sz)
@@ -351,11 +351,26 @@ nfs_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs_linkargs *args)
 static int
 nfs_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_symlinkargs *args)
 {
+	struct xdr_buf *sndbuf = &req->rq_snd_buf;
+	size_t pad;
+
 	p = xdr_encode_fhandle(p, args->fromfh);
 	p = xdr_encode_array(p, args->fromname, args->fromlen);
-	p = xdr_encode_array(p, args->topath, args->tolen);
+	*p++ = htonl(args->pathlen);
+	sndbuf->len = xdr_adjust_iovec(sndbuf->head, p);
+
+	xdr_encode_pages(sndbuf, args->pages, 0, args->pathlen);
+
+	/*
+	 * xdr_encode_pages may have added a few bytes to ensure the
+	 * pathname ends on a 4-byte boundary.  Start encoding the
+	 * attributes after the pad bytes.
+	 */
+	pad = sndbuf->tail->iov_len;
+	if (pad > 0)
+		p++;
 	p = xdr_encode_sattr(p, args->sattr);
-	req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
+	sndbuf->len += xdr_adjust_iovec(sndbuf->tail, p) - pad;
 	return 0;
 }
 
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index d85ac427c326..f8688eaa0001 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -544,8 +544,8 @@ nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 }
 
 static int
-nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
-		  struct iattr *sattr)
+nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
+		  unsigned int len, struct iattr *sattr)
 {
 	struct nfs_fh fhandle;
 	struct nfs_fattr fattr, dir_attr;
@@ -553,8 +553,8 @@ nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
 		.fromfh		= NFS_FH(dir),
 		.fromname	= dentry->d_name.name,
 		.fromlen	= dentry->d_name.len,
-		.topath		= path->name,
-		.tolen		= path->len,
+		.pages		= &page,
+		.pathlen	= len,
 		.sattr		= sattr
 	};
 	struct nfs3_diropres	res = {
@@ -569,11 +569,11 @@ nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
 	};
 	int			status;
 
-	if (path->len > NFS3_MAXPATHLEN)
+	if (len > NFS3_MAXPATHLEN)
 		return -ENAMETOOLONG;
 
-	dprintk("NFS call  symlink %s -> %s\n", dentry->d_name.name,
-			path->name);
+	dprintk("NFS call  symlink %s\n", dentry->d_name.name);
+
 	nfs_fattr_init(&dir_attr);
 	nfs_fattr_init(&fattr);
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c
index 0250269e9753..16556fa4effb 100644
--- a/fs/nfs/nfs3xdr.c
+++ b/fs/nfs/nfs3xdr.c
@@ -56,7 +56,7 @@
 #define NFS3_writeargs_sz	(NFS3_fh_sz+5)
 #define NFS3_createargs_sz	(NFS3_diropargs_sz+NFS3_sattr_sz)
 #define NFS3_mkdirargs_sz	(NFS3_diropargs_sz+NFS3_sattr_sz)
-#define NFS3_symlinkargs_sz	(NFS3_diropargs_sz+NFS3_path_sz+NFS3_sattr_sz)
+#define NFS3_symlinkargs_sz	(NFS3_diropargs_sz+1+NFS3_sattr_sz)
 #define NFS3_mknodargs_sz	(NFS3_diropargs_sz+2+NFS3_sattr_sz)
 #define NFS3_renameargs_sz	(NFS3_diropargs_sz+NFS3_diropargs_sz)
 #define NFS3_linkargs_sz		(NFS3_fh_sz+NFS3_diropargs_sz)
@@ -398,8 +398,11 @@ nfs3_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_symlinkargs *args
 	p = xdr_encode_fhandle(p, args->fromfh);
 	p = xdr_encode_array(p, args->fromname, args->fromlen);
 	p = xdr_encode_sattr(p, args->sattr);
-	p = xdr_encode_array(p, args->topath, args->tolen);
+	*p++ = htonl(args->pathlen);
 	req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
+
+	/* Copy the page */
+	xdr_encode_pages(&req->rq_snd_buf, args->pages, 0, args->pathlen);
 	return 0;
 }
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 2d18eac6bee5..7f60beb40df3 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2085,7 +2085,7 @@ static int nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *n
 }
 
 static int _nfs4_proc_symlink(struct inode *dir, struct dentry *dentry,
-		struct qstr *path, struct iattr *sattr)
+		struct page *page, unsigned int len, struct iattr *sattr)
 {
 	struct nfs_server *server = NFS_SERVER(dir);
 	struct nfs_fh fhandle;
@@ -2111,10 +2111,11 @@ static int _nfs4_proc_symlink(struct inode *dir, struct dentry *dentry,
 	};
 	int			status;
 
-	if (path->len > NFS4_MAXPATHLEN)
+	if (len > NFS4_MAXPATHLEN)
 		return -ENAMETOOLONG;
 
-	arg.u.symlink = path;
+	arg.u.symlink.pages = &page;
+	arg.u.symlink.len = len;
 	nfs_fattr_init(&fattr);
 	nfs_fattr_init(&dir_fattr);
 	
@@ -2128,13 +2129,14 @@ static int _nfs4_proc_symlink(struct inode *dir, struct dentry *dentry,
 }
 
 static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry,
-		struct qstr *path, struct iattr *sattr)
+		struct page *page, unsigned int len, struct iattr *sattr)
 {
 	struct nfs4_exception exception = { };
 	int err;
 	do {
 		err = nfs4_handle_exception(NFS_SERVER(dir),
-				_nfs4_proc_symlink(dir, dentry, path, sattr),
+				_nfs4_proc_symlink(dir, dentry, page,
+							len, sattr),
 				&exception);
 	} while (exception.retry);
 	return err;
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 99926067eca4..3dd413f52da1 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -128,7 +128,7 @@ static int nfs4_stat_to_errno(int);
 #define decode_link_maxsz	(op_decode_hdr_maxsz + 5)
 #define encode_symlink_maxsz	(op_encode_hdr_maxsz + \
 				1 + nfs4_name_maxsz + \
-				nfs4_path_maxsz + \
+				1 + \
 				nfs4_fattr_maxsz)
 #define decode_symlink_maxsz	(op_decode_hdr_maxsz + 8)
 #define encode_create_maxsz	(op_encode_hdr_maxsz + \
@@ -673,9 +673,9 @@ static int encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *c
 
 	switch (create->ftype) {
 	case NF4LNK:
-		RESERVE_SPACE(4 + create->u.symlink->len);
-		WRITE32(create->u.symlink->len);
-		WRITEMEM(create->u.symlink->name, create->u.symlink->len);
+		RESERVE_SPACE(4);
+		WRITE32(create->u.symlink.len);
+		xdr_write_pages(xdr, create->u.symlink.pages, 0, create->u.symlink.len);
 		break;
 
 	case NF4BLK: case NF4CHR:
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 0b507bf0f330..630e50647bbb 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -425,8 +425,8 @@ nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
 }
 
 static int
-nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
-		 struct iattr *sattr)
+nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page,
+		 unsigned int len, struct iattr *sattr)
 {
 	struct nfs_fh fhandle;
 	struct nfs_fattr fattr;
@@ -434,8 +434,8 @@ nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
 		.fromfh		= NFS_FH(dir),
 		.fromname	= dentry->d_name.name,
 		.fromlen	= dentry->d_name.len,
-		.topath		= path->name,
-		.tolen		= path->len,
+		.pages		= &page,
+		.pathlen	= len,
 		.sattr		= sattr
 	};
 	struct rpc_message msg = {
@@ -444,11 +444,11 @@ nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct qstr *path,
 	};
 	int			status;
 
-	if (path->len > NFS2_MAXPATHLEN)
+	if (len > NFS2_MAXPATHLEN)
 		return -ENAMETOOLONG;
 
-	dprintk("NFS call  symlink %s -> %s\n", dentry->d_name.name,
-			path->name);
+	dprintk("NFS call  symlink %s\n", dentry->d_name.name);
+
 	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
 	nfs_mark_for_revalidate(dir);
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index ddf5d75e97a2..dc5397d9d23c 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -358,8 +358,8 @@ struct nfs_symlinkargs {
 	struct nfs_fh *		fromfh;
 	const char *		fromname;
 	unsigned int		fromlen;
-	const char *		topath;
-	unsigned int		tolen;
+	struct page **		pages;
+	unsigned int		pathlen;
 	struct iattr *		sattr;
 };
 
@@ -434,8 +434,8 @@ struct nfs3_symlinkargs {
 	struct nfs_fh *		fromfh;
 	const char *		fromname;
 	unsigned int		fromlen;
-	const char *		topath;
-	unsigned int		tolen;
+	struct page **		pages;
+	unsigned int		pathlen;
 	struct iattr *		sattr;
 };
 
@@ -533,7 +533,10 @@ struct nfs4_accessres {
 struct nfs4_create_arg {
 	u32				ftype;
 	union {
-		struct qstr *		symlink;    /* NF4LNK */
+		struct {
+			struct page **	pages;
+			unsigned int	len;
+		} symlink;   /* NF4LNK */
 		struct {
 			u32		specdata1;
 			u32		specdata2;
@@ -793,8 +796,8 @@ struct nfs_rpc_ops {
 	int	(*rename)  (struct inode *, struct qstr *,
 			    struct inode *, struct qstr *);
 	int	(*link)    (struct inode *, struct inode *, struct qstr *);
-	int	(*symlink) (struct inode *, struct dentry *, struct qstr *,
-			    struct iattr *);
+	int	(*symlink) (struct inode *, struct dentry *, struct page *,
+			    unsigned int, struct iattr *);
 	int	(*mkdir)   (struct inode *, struct dentry *, struct iattr *);
 	int	(*rmdir)   (struct inode *, struct qstr *);
 	int	(*readdir) (struct dentry *, struct rpc_cred *,
-- 
cgit v1.2.3


From 275a082fe9308e710324e26ccb5363c53d8fd45f Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 22 Aug 2006 20:06:24 -0400
Subject: Add a real API for dealing with blk_congestion_wait()

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 block/ll_rw_blk.c         | 12 ++++++++++++
 fs/nfs/write.c            |  1 +
 include/linux/blkdev.h    |  1 +
 include/linux/writeback.h |  1 +
 mm/page-writeback.c       |  9 +++++++++
 5 files changed, 24 insertions(+)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index ddd9253f9d55..dcbd6ff1fa33 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -2734,6 +2734,18 @@ long blk_congestion_wait(int rw, long timeout)
 
 EXPORT_SYMBOL(blk_congestion_wait);
 
+/**
+ * blk_congestion_end - wake up sleepers on a congestion queue
+ * @rw: READ or WRITE
+ */
+void blk_congestion_end(int rw)
+{
+	wait_queue_head_t *wqh = &congestion_wqh[rw];
+
+	if (waitqueue_active(wqh))
+		wake_up(wqh);
+}
+
 /*
  * Has to be called with the request spinlock acquired
  */
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 453d44666ea5..38ba5c09af08 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -396,6 +396,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 out:
 	clear_bit(BDI_write_congested, &bdi->state);
 	wake_up_all(&nfs_write_congestion);
+	writeback_congestion_end();
 	return err;
 }
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index aafe82788b4e..96c9040c00a8 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -746,6 +746,7 @@ extern void blk_queue_free_tags(request_queue_t *);
 extern int blk_queue_resize_tags(request_queue_t *, int);
 extern void blk_queue_invalidate_tags(request_queue_t *);
 extern long blk_congestion_wait(int rw, long timeout);
+extern void blk_congestion_end(int rw);
 
 extern void blk_rq_bio_prep(request_queue_t *, struct request *, struct bio *);
 extern int blkdev_issue_flush(struct block_device *, sector_t *);
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 9e38b566d0e7..0422036af4eb 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -85,6 +85,7 @@ int wakeup_pdflush(long nr_pages);
 void laptop_io_completion(void);
 void laptop_sync_completion(void);
 void throttle_vm_writeout(void);
+void writeback_congestion_end(void);
 
 /* These are exported to sysctl. */
 extern int dirty_background_ratio;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index e630188ccc40..77a0bc4e261a 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -802,6 +802,15 @@ int test_set_page_writeback(struct page *page)
 }
 EXPORT_SYMBOL(test_set_page_writeback);
 
+/*
+ * Wakes up tasks that are being throttled due to writeback congestion
+ */
+void writeback_congestion_end(void)
+{
+	blk_congestion_end(WRITE);
+}
+EXPORT_SYMBOL(writeback_congestion_end);
+
 /*
  * Return true if any of the pages in the mapping are marged with the
  * passed tag.
-- 
cgit v1.2.3


From 5dd3177ae5012c1e2ad7a9ffdbd0e0d0de2f60e4 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 24 Aug 2006 01:03:05 -0400
Subject: NFSv4: Fix a use-after-free issue with the nfs server.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 36 +++++++++++++++++++++---------------
 fs/nfs/nfs4renewd.c       |  1 +
 fs/nfs/super.c            |  8 +++++---
 include/linux/nfs_fs_sb.h |  1 +
 4 files changed, 28 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 12941a8a6d75..f1ff2aec2ca5 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -164,6 +164,26 @@ error_0:
 	return NULL;
 }
 
+static void nfs4_shutdown_client(struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4
+	if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state))
+		nfs4_kill_renewd(clp);
+	while (!list_empty(&clp->cl_unused)) {
+		struct nfs4_state_owner *sp;
+
+		sp = list_entry(clp->cl_unused.next,
+				struct nfs4_state_owner,
+				so_list);
+		list_del(&sp->so_list);
+		kfree(sp);
+	}
+	BUG_ON(!list_empty(&clp->cl_state_owners));
+	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state))
+		nfs_idmap_delete(clp);
+#endif
+}
+
 /*
  * Destroy a shared client record
  */
@@ -171,21 +191,7 @@ static void nfs_free_client(struct nfs_client *clp)
 {
 	dprintk("--> nfs_free_client(%d)\n", clp->cl_nfsversion);
 
-#ifdef CONFIG_NFS_V4
-	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state)) {
-		while (!list_empty(&clp->cl_unused)) {
-			struct nfs4_state_owner *sp;
-
-			sp = list_entry(clp->cl_unused.next,
-					struct nfs4_state_owner,
-					so_list);
-			list_del(&sp->so_list);
-			kfree(sp);
-		}
-		BUG_ON(!list_empty(&clp->cl_state_owners));
-		nfs_idmap_delete(clp);
-	}
-#endif
+	nfs4_shutdown_client(clp);
 
 	/* -EIO all pending I/O */
 	if (!IS_ERR(clp->cl_rpcclient))
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index f2c893690ac4..7b6df1852e75 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -121,6 +121,7 @@ nfs4_schedule_state_renewal(struct nfs_client *clp)
 			__FUNCTION__, (timeout + HZ - 1) / HZ);
 	cancel_delayed_work(&clp->cl_renewd);
 	schedule_delayed_work(&clp->cl_renewd, timeout);
+	set_bit(NFS_CS_RENEWD, &clp->cl_res_state);
 	spin_unlock(&clp->cl_lock);
 }
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 97cfb143e09f..665949d27798 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -883,13 +883,15 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 		goto out_free;
 	}
 
+	if (s->s_fs_info != server) {
+		nfs_free_server(server);
+		server = NULL;
+	}
+
 	if (!s->s_root) {
 		/* initial superblock/root creation */
 		s->s_flags = flags;
-
 		nfs4_fill_super(s);
-	} else {
-		nfs_free_server(server);
 	}
 
 	mntroot = nfs4_get_root(s, &mntfh);
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 6d0be0efd1b5..7ccfc7ef0a83 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -19,6 +19,7 @@ struct nfs_client {
 #define NFS_CS_RPCIOD		0		/* - rpciod started */
 #define NFS_CS_CALLBACK		1		/* - callback started */
 #define NFS_CS_IDMAP		2		/* - idmap started */
+#define NFS_CS_RENEWD		3		/* - renewd started */
 	struct sockaddr_in	cl_addr;	/* server identifier */
 	char *			cl_hostname;	/* hostname of server */
 	struct list_head	cl_share_link;	/* link in global client list */
-- 
cgit v1.2.3


From 158998b6fe36f6acef087f574c96d44713499cc9 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Thu, 24 Aug 2006 01:03:17 -0400
Subject: SUNRPC: Make rpc_mkpipe() take the parent dentry as an argument

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/idmap.c                     |  6 +-----
 include/linux/sunrpc/rpc_pipe_fs.h |  2 +-
 net/sunrpc/auth_gss/auth_gss.c     |  7 ++-----
 net/sunrpc/rpc_pipe.c              | 38 +++++++++++++++++++++++---------------
 4 files changed, 27 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c
index f96dfac7dc9a..82ad7110a1c0 100644
--- a/fs/nfs/idmap.c
+++ b/fs/nfs/idmap.c
@@ -84,7 +84,6 @@ struct idmap_hashtable {
 };
 
 struct idmap {
-	char                  idmap_path[48];
 	struct dentry        *idmap_dentry;
 	wait_queue_head_t     idmap_wq;
 	struct idmap_msg      idmap_im;
@@ -119,10 +118,7 @@ nfs_idmap_new(struct nfs_client *clp)
         if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL)
                 return -ENOMEM;
 
-	snprintf(idmap->idmap_path, sizeof(idmap->idmap_path),
-	    "%s/idmap", clp->cl_rpcclient->cl_pathname);
-
-        idmap->idmap_dentry = rpc_mkpipe(idmap->idmap_path,
+        idmap->idmap_dentry = rpc_mkpipe(clp->cl_rpcclient->cl_dentry, "idmap",
 	    idmap, &idmap_upcall_ops, 0);
         if (IS_ERR(idmap->idmap_dentry)) {
 		error = PTR_ERR(idmap->idmap_dentry);
diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h
index a481472c9484..a2eb9b4a9de3 100644
--- a/include/linux/sunrpc/rpc_pipe_fs.h
+++ b/include/linux/sunrpc/rpc_pipe_fs.h
@@ -43,7 +43,7 @@ extern int rpc_queue_upcall(struct inode *, struct rpc_pipe_msg *);
 
 extern struct dentry *rpc_mkdir(char *, struct rpc_clnt *);
 extern int rpc_rmdir(struct dentry *);
-extern struct dentry *rpc_mkpipe(char *, void *, struct rpc_pipe_ops *, int flags);
+extern struct dentry *rpc_mkpipe(struct dentry *, const char *, void *, struct rpc_pipe_ops *, int flags);
 extern int rpc_unlink(struct dentry *);
 extern struct vfsmount *rpc_get_mount(void);
 extern void rpc_put_mount(void);
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index ef1cf5b476c8..6eed3e166ba3 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -88,7 +88,6 @@ struct gss_auth {
 	struct list_head upcalls;
 	struct rpc_clnt *client;
 	struct dentry *dentry;
-	char path[48];
 	spinlock_t lock;
 };
 
@@ -690,10 +689,8 @@ gss_create(struct rpc_clnt *clnt, rpc_authflavor_t flavor)
 	if (err)
 		goto err_put_mech;
 
-	snprintf(gss_auth->path, sizeof(gss_auth->path), "%s/%s",
-			clnt->cl_pathname,
-			gss_auth->mech->gm_name);
-	gss_auth->dentry = rpc_mkpipe(gss_auth->path, clnt, &gss_upcall_ops, RPC_PIPE_WAIT_FOR_OPEN);
+	gss_auth->dentry = rpc_mkpipe(clnt->cl_dentry, gss_auth->mech->gm_name,
+			clnt, &gss_upcall_ops, RPC_PIPE_WAIT_FOR_OPEN);
 	if (IS_ERR(gss_auth->dentry)) {
 		err = PTR_ERR(gss_auth->dentry);
 		goto err_put_mech;
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index c21dc07f2a8c..11ec12a09d70 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -621,17 +621,13 @@ __rpc_rmdir(struct inode *dir, struct dentry *dentry)
 }
 
 static struct dentry *
-rpc_lookup_negative(char *path, struct nameidata *nd)
+rpc_lookup_create(struct dentry *parent, const char *name, int len)
 {
+	struct inode *dir = parent->d_inode;
 	struct dentry *dentry;
-	struct inode *dir;
-	int error;
 
-	if ((error = rpc_lookup_parent(path, nd)) != 0)
-		return ERR_PTR(error);
-	dir = nd->dentry->d_inode;
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
-	dentry = lookup_one_len(nd->last.name, nd->dentry, nd->last.len);
+	dentry = lookup_one_len(name, parent, len);
 	if (IS_ERR(dentry))
 		goto out_err;
 	if (dentry->d_inode) {
@@ -642,7 +638,20 @@ rpc_lookup_negative(char *path, struct nameidata *nd)
 	return dentry;
 out_err:
 	mutex_unlock(&dir->i_mutex);
-	rpc_release_path(nd);
+	return dentry;
+}
+
+static struct dentry *
+rpc_lookup_negative(char *path, struct nameidata *nd)
+{
+	struct dentry *dentry;
+	int error;
+
+	if ((error = rpc_lookup_parent(path, nd)) != 0)
+		return ERR_PTR(error);
+	dentry = rpc_lookup_create(nd->dentry, nd->last.name, nd->last.len);
+	if (IS_ERR(dentry))
+		rpc_release_path(nd);
 	return dentry;
 }
 
@@ -701,17 +710,16 @@ rpc_rmdir(struct dentry *dentry)
 }
 
 struct dentry *
-rpc_mkpipe(char *path, void *private, struct rpc_pipe_ops *ops, int flags)
+rpc_mkpipe(struct dentry *parent, const char *name, void *private, struct rpc_pipe_ops *ops, int flags)
 {
-	struct nameidata nd;
 	struct dentry *dentry;
 	struct inode *dir, *inode;
 	struct rpc_inode *rpci;
 
-	dentry = rpc_lookup_negative(path, &nd);
+	dentry = rpc_lookup_create(parent, name, strlen(name));
 	if (IS_ERR(dentry))
 		return dentry;
-	dir = nd.dentry->d_inode;
+	dir = parent->d_inode;
 	inode = rpc_get_inode(dir->i_sb, S_IFSOCK | S_IRUSR | S_IWUSR);
 	if (!inode)
 		goto err_dput;
@@ -726,13 +734,13 @@ rpc_mkpipe(char *path, void *private, struct rpc_pipe_ops *ops, int flags)
 	dget(dentry);
 out:
 	mutex_unlock(&dir->i_mutex);
-	rpc_release_path(&nd);
 	return dentry;
 err_dput:
 	dput(dentry);
 	dentry = ERR_PTR(-ENOMEM);
-	printk(KERN_WARNING "%s: %s() failed to create pipe %s (errno = %d)\n",
-			__FILE__, __FUNCTION__, path, -ENOMEM);
+	printk(KERN_WARNING "%s: %s() failed to create pipe %s/%s (errno = %d)\n",
+			__FILE__, __FUNCTION__, parent->d_name.name, name,
+			-ENOMEM);
 	goto out;
 }
 
-- 
cgit v1.2.3


From 6b6ca86b77b62b798cf9ca2599036420abce7796 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 5 Sep 2006 12:55:57 -0400
Subject: SUNRPC: Add refcounting to the struct rpc_xprt

In a subsequent patch, this will allow the portmapper to take a reference
to the rpc_xprt for which it is updating the port number, fixing an Oops.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/xprt.h |  5 ++++-
 net/sunrpc/clnt.c           |  8 +++-----
 net/sunrpc/xprt.c           | 28 +++++++++++++++++++++++++---
 3 files changed, 32 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index de4efea7c856..bdeba8538c71 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -12,6 +12,7 @@
 #include <linux/uio.h>
 #include <linux/socket.h>
 #include <linux/in.h>
+#include <linux/kref.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/xdr.h>
 
@@ -129,6 +130,7 @@ struct rpc_xprt_ops {
 };
 
 struct rpc_xprt {
+	struct kref		kref;		/* Reference count */
 	struct rpc_xprt_ops *	ops;		/* transport methods */
 	struct socket *		sock;		/* BSD socket layer */
 	struct sock *		inet;		/* INET layer */
@@ -248,7 +250,8 @@ int			xprt_adjust_timeout(struct rpc_rqst *req);
 void			xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task);
 void			xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task);
 void			xprt_release(struct rpc_task *task);
-int			xprt_destroy(struct rpc_xprt *xprt);
+struct rpc_xprt *	xprt_get(struct rpc_xprt *xprt);
+void			xprt_put(struct rpc_xprt *xprt);
 
 static inline u32 *xprt_skip_transport_header(struct rpc_xprt *xprt, u32 *p)
 {
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index ceadb728f0da..084a0ad5c64e 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -177,7 +177,7 @@ out_no_path:
 		kfree(clnt->cl_server);
 	kfree(clnt);
 out_err:
-	xprt_destroy(xprt);
+	xprt_put(xprt);
 out_no_xprt:
 	return ERR_PTR(err);
 }
@@ -261,6 +261,7 @@ rpc_clone_client(struct rpc_clnt *clnt)
 	atomic_set(&new->cl_users, 0);
 	new->cl_parent = clnt;
 	atomic_inc(&clnt->cl_count);
+	new->cl_xprt = xprt_get(clnt->cl_xprt);
 	/* Turn off autobind on clones */
 	new->cl_autobind = 0;
 	new->cl_oneshot = 0;
@@ -337,15 +338,12 @@ rpc_destroy_client(struct rpc_clnt *clnt)
 		rpc_rmdir(clnt->cl_dentry);
 		rpc_put_mount();
 	}
-	if (clnt->cl_xprt) {
-		xprt_destroy(clnt->cl_xprt);
-		clnt->cl_xprt = NULL;
-	}
 	if (clnt->cl_server != clnt->cl_inline_name)
 		kfree(clnt->cl_server);
 out_free:
 	rpc_free_iostats(clnt->cl_metrics);
 	clnt->cl_metrics = NULL;
+	xprt_put(clnt->cl_xprt);
 	kfree(clnt);
 	return 0;
 }
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index a85f82baefc1..1f786f68729d 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -926,6 +926,7 @@ struct rpc_xprt *xprt_create_transport(int proto, struct sockaddr *ap, size_t si
 		return ERR_PTR(result);
 	}
 
+	kref_init(&xprt->kref);
 	spin_lock_init(&xprt->transport_lock);
 	spin_lock_init(&xprt->reserve_lock);
 
@@ -958,16 +959,37 @@ struct rpc_xprt *xprt_create_transport(int proto, struct sockaddr *ap, size_t si
 
 /**
  * xprt_destroy - destroy an RPC transport, killing off all requests.
- * @xprt: transport to destroy
+ * @kref: kref for the transport to destroy
  *
  */
-int xprt_destroy(struct rpc_xprt *xprt)
+static void xprt_destroy(struct kref *kref)
 {
+	struct rpc_xprt *xprt = container_of(kref, struct rpc_xprt, kref);
+
 	dprintk("RPC:      destroying transport %p\n", xprt);
 	xprt->shutdown = 1;
 	del_timer_sync(&xprt->timer);
 	xprt->ops->destroy(xprt);
 	kfree(xprt);
+}
 
-	return 0;
+/**
+ * xprt_put - release a reference to an RPC transport.
+ * @xprt: pointer to the transport
+ *
+ */
+void xprt_put(struct rpc_xprt *xprt)
+{
+	kref_put(&xprt->kref, xprt_destroy);
+}
+
+/**
+ * xprt_get - return a reference to an RPC transport.
+ * @xprt: pointer to the transport
+ *
+ */
+struct rpc_xprt *xprt_get(struct rpc_xprt *xprt)
+{
+	kref_get(&xprt->kref);
+	return xprt;
 }
-- 
cgit v1.2.3


From 42750b04c5baa7c5ffdf0a8be2b9b320efdf069f Mon Sep 17 00:00:00 2001
From: Jaroslav Kysela <perex@suse.cz>
Date: Thu, 1 Jun 2006 18:34:01 +0200
Subject: [ALSA] Control API - TLV implementation for additional information
 like dB scale

This patch implements a TLV mechanism to transfer an additional information
like dB scale to the user space. The types might be extended in future.
Acked-by: Takashi Iwai <tiwai@suse.de>

Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/asound.h          |  9 ++++++++-
 include/sound/control.h         |  2 ++
 include/sound/tlv.h             | 43 +++++++++++++++++++++++++++++++++++++++++
 sound/core/control.c            | 41 +++++++++++++++++++++++++++++++++++++--
 sound/pci/ca0106/ca0106_mixer.c |  4 ++++
 5 files changed, 96 insertions(+), 3 deletions(-)
 create mode 100644 include/sound/tlv.h

(limited to 'include')

diff --git a/include/sound/asound.h b/include/sound/asound.h
index 41885f48ad91..76a20406bd18 100644
--- a/include/sound/asound.h
+++ b/include/sound/asound.h
@@ -688,7 +688,7 @@ struct snd_timer_tread {
  *                                                                          *
  ****************************************************************************/
 
-#define SNDRV_CTL_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 3)
+#define SNDRV_CTL_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 4)
 
 struct snd_ctl_card_info {
 	int card;			/* card number */
@@ -818,6 +818,12 @@ struct snd_ctl_elem_value {
         unsigned char reserved[128-sizeof(struct timespec)];
 };
 
+struct snd_ctl_tlv {
+        unsigned int numid;	/* control element numeric identification */
+        unsigned int length;	/* in bytes aligned to 4 */
+        unsigned int tlv[0];	/* first TLV */
+};
+
 enum {
 	SNDRV_CTL_IOCTL_PVERSION = _IOR('U', 0x00, int),
 	SNDRV_CTL_IOCTL_CARD_INFO = _IOR('U', 0x01, struct snd_ctl_card_info),
@@ -831,6 +837,7 @@ enum {
 	SNDRV_CTL_IOCTL_ELEM_ADD = _IOWR('U', 0x17, struct snd_ctl_elem_info),
 	SNDRV_CTL_IOCTL_ELEM_REPLACE = _IOWR('U', 0x18, struct snd_ctl_elem_info),
 	SNDRV_CTL_IOCTL_ELEM_REMOVE = _IOWR('U', 0x19, struct snd_ctl_elem_id),
+	SNDRV_CTL_IOCTL_TLV_READ = _IOWR('U', 0x1a, struct snd_ctl_tlv),
 	SNDRV_CTL_IOCTL_HWDEP_NEXT_DEVICE = _IOWR('U', 0x20, int),
 	SNDRV_CTL_IOCTL_HWDEP_INFO = _IOR('U', 0x21, struct snd_hwdep_info),
 	SNDRV_CTL_IOCTL_PCM_NEXT_DEVICE = _IOR('U', 0x30, int),
diff --git a/include/sound/control.h b/include/sound/control.h
index 2489b1eb0110..a93a58d0e688 100644
--- a/include/sound/control.h
+++ b/include/sound/control.h
@@ -42,6 +42,7 @@ struct snd_kcontrol_new {
 	snd_kcontrol_info_t *info;
 	snd_kcontrol_get_t *get;
 	snd_kcontrol_put_t *put;
+	unsigned int *tlv;
 	unsigned long private_value;
 };
 
@@ -58,6 +59,7 @@ struct snd_kcontrol {
 	snd_kcontrol_info_t *info;
 	snd_kcontrol_get_t *get;
 	snd_kcontrol_put_t *put;
+	unsigned int *tlv;
 	unsigned long private_value;
 	void *private_data;
 	void (*private_free)(struct snd_kcontrol *kcontrol);
diff --git a/include/sound/tlv.h b/include/sound/tlv.h
new file mode 100644
index 000000000000..b826e1df1da6
--- /dev/null
+++ b/include/sound/tlv.h
@@ -0,0 +1,43 @@
+#ifndef __SOUND_TLV_H
+#define __SOUND_TLV_H
+
+/*
+ *  Advanced Linux Sound Architecture - ALSA - Driver
+ *  Copyright (c) 2006 by Jaroslav Kysela <perex@suse.cz>
+ *
+ *
+ *   This program is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; either version 2 of the License, or
+ *   (at your option) any later version.
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ *
+ *   You should have received a copy of the GNU General Public License
+ *   along with this program; if not, write to the Free Software
+ *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ *
+ */
+
+/*
+ * TLV structure is right behind the struct snd_ctl_tlv:
+ *   unsigned int type  	- see SNDRV_CTL_TLVT_*
+ *   unsigned int length
+ *   .... data aligned to sizeof(unsigned int), use
+ *        block_length = (length + (sizeof(unsigned int) - 1)) &
+ *                       ~(sizeof(unsigned int) - 1)) ....
+ */
+
+#define SNDRV_CTL_TLVT_CONTAINER 0	/* one level down - group of TLVs */
+#define SNDRV_CTL_TLVT_DB_SCALE	1       /* dB scale */
+
+#define DECLARE_TLV_DB_SCALE(name, min, step, mute) \
+unsigned int name[] = { \
+        SNDRV_CTL_TLVT_DB_SCALE, 2 * sizeof(unsigned int), \
+        (min), ((step) & 0xffff) | ((mute) ? 0x10000 : 0) \
+}
+
+#endif /* __SOUND_TLV_H */
diff --git a/sound/core/control.c b/sound/core/control.c
index bb397eaa7187..e9c8854d2f7b 100644
--- a/sound/core/control.c
+++ b/sound/core/control.c
@@ -241,6 +241,7 @@ struct snd_kcontrol *snd_ctl_new1(const struct snd_kcontrol_new *ncontrol,
 	kctl.info = ncontrol->info;
 	kctl.get = ncontrol->get;
 	kctl.put = ncontrol->put;
+	kctl.tlv = ncontrol->tlv;
 	kctl.private_value = ncontrol->private_value;
 	kctl.private_data = private_data;
 	return snd_ctl_new(&kctl, access);
@@ -1067,6 +1068,40 @@ static int snd_ctl_subscribe_events(struct snd_ctl_file *file, int __user *ptr)
 	return 0;
 }
 
+static int snd_ctl_tlv_read(struct snd_card *card,
+                            struct snd_ctl_tlv __user *_tlv)
+{
+	struct snd_ctl_tlv tlv;
+	struct snd_kcontrol *kctl;
+	unsigned int len;
+	int err = 0;
+
+	if (copy_from_user(&tlv, _tlv, sizeof(tlv)))
+		return -EFAULT;
+        if (tlv.length < sizeof(unsigned int) * 3)
+                return -EINVAL;
+        down_read(&card->controls_rwsem);
+        kctl = snd_ctl_find_numid(card, tlv.numid);
+        if (kctl == NULL) {
+                err = -ENOENT;
+                goto __kctl_end;
+        }
+        if (kctl->tlv == NULL) {
+                err = -ENXIO;
+                goto __kctl_end;
+        }
+        len = kctl->tlv[1] + 2 * sizeof(unsigned int);
+        if (tlv.length < len) {
+                err = -ENOMEM;
+                goto __kctl_end;
+        }
+        if (copy_to_user(_tlv->tlv, kctl->tlv, len))
+        	err = -EFAULT;
+      __kctl_end:
+        up_read(&card->controls_rwsem);
+        return err;
+}
+
 static long snd_ctl_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct snd_ctl_file *ctl;
@@ -1086,11 +1121,11 @@ static long snd_ctl_ioctl(struct file *file, unsigned int cmd, unsigned long arg
 	case SNDRV_CTL_IOCTL_CARD_INFO:
 		return snd_ctl_card_info(card, ctl, cmd, argp);
 	case SNDRV_CTL_IOCTL_ELEM_LIST:
-		return snd_ctl_elem_list(ctl->card, argp);
+		return snd_ctl_elem_list(card, argp);
 	case SNDRV_CTL_IOCTL_ELEM_INFO:
 		return snd_ctl_elem_info_user(ctl, argp);
 	case SNDRV_CTL_IOCTL_ELEM_READ:
-		return snd_ctl_elem_read_user(ctl->card, argp);
+		return snd_ctl_elem_read_user(card, argp);
 	case SNDRV_CTL_IOCTL_ELEM_WRITE:
 		return snd_ctl_elem_write_user(ctl, argp);
 	case SNDRV_CTL_IOCTL_ELEM_LOCK:
@@ -1105,6 +1140,8 @@ static long snd_ctl_ioctl(struct file *file, unsigned int cmd, unsigned long arg
 		return snd_ctl_elem_remove(ctl, argp);
 	case SNDRV_CTL_IOCTL_SUBSCRIBE_EVENTS:
 		return snd_ctl_subscribe_events(ctl, ip);
+        case SNDRV_CTL_IOCTL_TLV_READ:
+                return snd_ctl_tlv_read(card, argp);
 	case SNDRV_CTL_IOCTL_POWER:
 		return -ENOPROTOOPT;
 	case SNDRV_CTL_IOCTL_POWER_STATE:
diff --git a/sound/pci/ca0106/ca0106_mixer.c b/sound/pci/ca0106/ca0106_mixer.c
index 146eed70dce6..35309b3ed8c0 100644
--- a/sound/pci/ca0106/ca0106_mixer.c
+++ b/sound/pci/ca0106/ca0106_mixer.c
@@ -70,9 +70,12 @@
 #include <sound/pcm.h>
 #include <sound/ac97_codec.h>
 #include <sound/info.h>
+#include <sound/tlv.h>
 
 #include "ca0106.h"
 
+static DECLARE_TLV_DB_SCALE(snd_ca0106_db_scale, -5150, 75, 1);
+
 static int snd_ca0106_shared_spdif_info(struct snd_kcontrol *kcontrol,
 					struct snd_ctl_elem_info *uinfo)
 {
@@ -472,6 +475,7 @@ static int snd_ca0106_i2c_volume_put(struct snd_kcontrol *kcontrol,
 	.info =	 snd_ca0106_volume_info,			\
 	.get =   snd_ca0106_volume_get,				\
 	.put =   snd_ca0106_volume_put,				\
+	.tlv =	 snd_ca0106_db_scale,				\
 	.private_value = ((chid) << 8) | (reg)			\
 }
 
-- 
cgit v1.2.3


From 746d4a02e68499fc6c1f8d0c43d2271853ade181 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 23 Jun 2006 14:37:59 +0200
Subject: [ALSA] Fix disconnection of proc interface

- Add the linked list to each proc entry to enable a single-shot
  disconnection (unregister)
- Deprecate snd_info_unregister(), use snd_info_free_entry()
- Removed NULL checks of snd_info_free_entry()

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/info.h                |   7 +--
 sound/core/hwdep.c                  |   2 +-
 sound/core/info.c                   | 102 ++++++++++++++++++------------------
 sound/core/info_oss.c               |   6 +--
 sound/core/init.c                   |   7 +--
 sound/core/oss/mixer_oss.c          |   6 +--
 sound/core/oss/pcm_oss.c            |   8 ++-
 sound/core/pcm.c                    |  53 +++++++------------
 sound/core/pcm_memory.c             |   2 +-
 sound/core/rawmidi.c                |   2 +-
 sound/core/seq/oss/seq_oss.c        |   3 +-
 sound/core/seq/seq_device.c         |   2 +-
 sound/core/seq/seq_info.c           |   6 +--
 sound/core/sound.c                  |   3 +-
 sound/core/sound_oss.c              |   3 +-
 sound/core/timer.c                  |   2 +-
 sound/drivers/opl4/opl4_proc.c      |   3 +-
 sound/pci/ac97/ac97_proc.c          |  18 +++----
 sound/pci/cs46xx/dsp_spos.c         |  52 +++++++-----------
 sound/pci/cs46xx/dsp_spos_scb_lib.c |   2 +-
 sound/synth/emux/emux_proc.c        |   6 +--
 21 files changed, 123 insertions(+), 172 deletions(-)

(limited to 'include')

diff --git a/include/sound/info.h b/include/sound/info.h
index 74f6996769c7..97ffc4fb9969 100644
--- a/include/sound/info.h
+++ b/include/sound/info.h
@@ -71,7 +71,6 @@ struct snd_info_entry {
 	mode_t mode;
 	long size;
 	unsigned short content;
-	unsigned short disconnected: 1;
 	union {
 		struct snd_info_entry_text text;
 		struct snd_info_entry_ops *ops;
@@ -83,6 +82,8 @@ struct snd_info_entry {
 	void (*private_free)(struct snd_info_entry *entry);
 	struct proc_dir_entry *p;
 	struct mutex access;
+	struct list_head children;
+	struct list_head list;
 };
 
 #if defined(CONFIG_SND_OSSEMUL) && defined(CONFIG_PROC_FS)
@@ -122,8 +123,8 @@ int snd_info_restore_text(struct snd_info_entry * entry);
 int snd_info_card_create(struct snd_card * card);
 int snd_info_card_register(struct snd_card * card);
 int snd_info_card_free(struct snd_card * card);
+void snd_info_card_disconnect(struct snd_card * card);
 int snd_info_register(struct snd_info_entry * entry);
-int snd_info_unregister(struct snd_info_entry * entry);
 
 /* for card drivers */
 int snd_card_proc_new(struct snd_card *card, const char *name, struct snd_info_entry **entryp);
@@ -156,8 +157,8 @@ static inline void snd_info_free_entry(struct snd_info_entry * entry) { ; }
 static inline int snd_info_card_create(struct snd_card * card) { return 0; }
 static inline int snd_info_card_register(struct snd_card * card) { return 0; }
 static inline int snd_info_card_free(struct snd_card * card) { return 0; }
+static inline void snd_info_card_disconnect(struct snd_card * card) { }
 static inline int snd_info_register(struct snd_info_entry * entry) { return 0; }
-static inline int snd_info_unregister(struct snd_info_entry * entry) { return 0; }
 
 static inline int snd_card_proc_new(struct snd_card *card, const char *name,
 				    struct snd_info_entry **entryp) { return -EINVAL; }
diff --git a/sound/core/hwdep.c b/sound/core/hwdep.c
index 8bd0dcc93eba..cbd8a63282b6 100644
--- a/sound/core/hwdep.c
+++ b/sound/core/hwdep.c
@@ -497,7 +497,7 @@ static void __init snd_hwdep_proc_init(void)
 
 static void __exit snd_hwdep_proc_done(void)
 {
-	snd_info_unregister(snd_hwdep_proc_entry);
+	snd_info_free_entry(snd_hwdep_proc_entry);
 }
 #else /* !CONFIG_PROC_FS */
 #define snd_hwdep_proc_init()
diff --git a/sound/core/info.c b/sound/core/info.c
index 340332c6d973..9663b6be9c3a 100644
--- a/sound/core/info.c
+++ b/sound/core/info.c
@@ -78,6 +78,7 @@ struct snd_info_private_data {
 
 static int snd_info_version_init(void);
 static int snd_info_version_done(void);
+static void snd_info_disconnect(struct snd_info_entry *entry);
 
 
 /* resize the proc r/w buffer */
@@ -304,7 +305,7 @@ static int snd_info_entry_open(struct inode *inode, struct file *file)
 	mutex_lock(&info_mutex);
 	p = PDE(inode);
 	entry = p == NULL ? NULL : (struct snd_info_entry *)p->data;
-	if (entry == NULL || entry->disconnected) {
+	if (entry == NULL || ! entry->p) {
 		mutex_unlock(&info_mutex);
 		return -ENODEV;
 	}
@@ -586,10 +587,10 @@ int __exit snd_info_done(void)
 	snd_info_version_done();
 	if (snd_proc_root) {
 #if defined(CONFIG_SND_SEQUENCER) || defined(CONFIG_SND_SEQUENCER_MODULE)
-		snd_info_unregister(snd_seq_root);
+		snd_info_free_entry(snd_seq_root);
 #endif
 #ifdef CONFIG_SND_OSSEMUL
-		snd_info_unregister(snd_oss_root);
+		snd_info_free_entry(snd_oss_root);
 #endif
 		snd_remove_proc_entry(&proc_root, snd_proc_root);
 	}
@@ -648,17 +649,28 @@ int snd_info_card_register(struct snd_card *card)
  * de-register the card proc file
  * called from init.c
  */
-int snd_info_card_free(struct snd_card *card)
+void snd_info_card_disconnect(struct snd_card *card)
 {
-	snd_assert(card != NULL, return -ENXIO);
+	snd_assert(card != NULL, return);
+	mutex_lock(&info_mutex);
 	if (card->proc_root_link) {
 		snd_remove_proc_entry(snd_proc_root, card->proc_root_link);
 		card->proc_root_link = NULL;
 	}
-	if (card->proc_root) {
-		snd_info_unregister(card->proc_root);
-		card->proc_root = NULL;
-	}
+	if (card->proc_root)
+		snd_info_disconnect(card->proc_root);
+	mutex_unlock(&info_mutex);
+}
+
+/*
+ * release the card proc file resources
+ * called from init.c
+ */
+int snd_info_card_free(struct snd_card *card)
+{
+	snd_assert(card != NULL, return -ENXIO);
+	snd_info_free_entry(card->proc_root);
+	card->proc_root = NULL;
 	return 0;
 }
 
@@ -767,6 +779,8 @@ static struct snd_info_entry *snd_info_create_entry(const char *name)
 	entry->mode = S_IFREG | S_IRUGO;
 	entry->content = SNDRV_INFO_CONTENT_TEXT;
 	mutex_init(&entry->access);
+	INIT_LIST_HEAD(&entry->children);
+	INIT_LIST_HEAD(&entry->list);
 	return entry;
 }
 
@@ -819,30 +833,35 @@ struct snd_info_entry *snd_info_create_card_entry(struct snd_card *card,
 
 EXPORT_SYMBOL(snd_info_create_card_entry);
 
-static int snd_info_dev_free_entry(struct snd_device *device)
+static void snd_info_disconnect(struct snd_info_entry *entry)
 {
-	struct snd_info_entry *entry = device->device_data;
-	snd_info_free_entry(entry);
-	return 0;
-}
+	struct list_head *p, *n;
+	struct proc_dir_entry *root;
 
-static int snd_info_dev_register_entry(struct snd_device *device)
-{
-	struct snd_info_entry *entry = device->device_data;
-	return snd_info_register(entry);
+	list_for_each_safe(p, n, &entry->children) {
+		snd_info_disconnect(list_entry(p, struct snd_info_entry, list));
+	}
+
+	if (! entry->p)
+		return;
+	list_del_init(&entry->list);
+	root = entry->parent == NULL ? snd_proc_root : entry->parent->p;
+	snd_assert(root, return);
+	snd_remove_proc_entry(root, entry->p);
+	entry->p = NULL;
 }
 
-static int snd_info_dev_disconnect_entry(struct snd_device *device)
+static int snd_info_dev_free_entry(struct snd_device *device)
 {
 	struct snd_info_entry *entry = device->device_data;
-	entry->disconnected = 1;
+	snd_info_free_entry(entry);
 	return 0;
 }
 
-static int snd_info_dev_unregister_entry(struct snd_device *device)
+static int snd_info_dev_register_entry(struct snd_device *device)
 {
 	struct snd_info_entry *entry = device->device_data;
-	return snd_info_unregister(entry);
+	return snd_info_register(entry);
 }
 
 /**
@@ -871,8 +890,7 @@ int snd_card_proc_new(struct snd_card *card, const char *name,
 	static struct snd_device_ops ops = {
 		.dev_free = snd_info_dev_free_entry,
 		.dev_register =	snd_info_dev_register_entry,
-		.dev_disconnect = snd_info_dev_disconnect_entry,
-		.dev_unregister = snd_info_dev_unregister_entry
+		/* disconnect is done via snd_info_card_disconnect() */
 	};
 	struct snd_info_entry *entry;
 	int err;
@@ -901,6 +919,11 @@ void snd_info_free_entry(struct snd_info_entry * entry)
 {
 	if (entry == NULL)
 		return;
+	if (entry->p) {
+		mutex_lock(&info_mutex);
+		snd_info_disconnect(entry);
+		mutex_unlock(&info_mutex);
+	}
 	kfree(entry->name);
 	if (entry->private_free)
 		entry->private_free(entry);
@@ -935,38 +958,14 @@ int snd_info_register(struct snd_info_entry * entry)
 	p->size = entry->size;
 	p->data = entry;
 	entry->p = p;
+	if (entry->parent)
+		list_add_tail(&entry->list, &entry->parent->children);
 	mutex_unlock(&info_mutex);
 	return 0;
 }
 
 EXPORT_SYMBOL(snd_info_register);
 
-/**
- * snd_info_unregister - de-register the info entry
- * @entry: the info entry
- *
- * De-registers the info entry and releases the instance.
- *
- * Returns zero if successful, or a negative error code on failure.
- */
-int snd_info_unregister(struct snd_info_entry * entry)
-{
-	struct proc_dir_entry *root;
-
-	if (! entry)
-		return 0;
-	snd_assert(entry->p != NULL, return -ENXIO);
-	root = entry->parent == NULL ? snd_proc_root : entry->parent->p;
-	snd_assert(root, return -ENXIO);
-	mutex_lock(&info_mutex);
-	snd_remove_proc_entry(root, entry->p);
-	mutex_unlock(&info_mutex);
-	snd_info_free_entry(entry);
-	return 0;
-}
-
-EXPORT_SYMBOL(snd_info_unregister);
-
 /*
 
  */
@@ -999,8 +998,7 @@ static int __init snd_info_version_init(void)
 
 static int __exit snd_info_version_done(void)
 {
-	if (snd_info_version_entry)
-		snd_info_unregister(snd_info_version_entry);
+	snd_info_free_entry(snd_info_version_entry);
 	return 0;
 }
 
diff --git a/sound/core/info_oss.c b/sound/core/info_oss.c
index bb2c40d0ab66..3ebc34919c76 100644
--- a/sound/core/info_oss.c
+++ b/sound/core/info_oss.c
@@ -131,10 +131,8 @@ int snd_info_minor_register(void)
 
 int snd_info_minor_unregister(void)
 {
-	if (snd_sndstat_proc_entry) {
-		snd_info_unregister(snd_sndstat_proc_entry);
-		snd_sndstat_proc_entry = NULL;
-	}
+	snd_info_free_entry(snd_sndstat_proc_entry);
+	snd_sndstat_proc_entry = NULL;
 	return 0;
 }
 
diff --git a/sound/core/init.c b/sound/core/init.c
index 4d9258884e44..1ecb029ff4c9 100644
--- a/sound/core/init.c
+++ b/sound/core/init.c
@@ -310,6 +310,7 @@ int snd_card_disconnect(struct snd_card *card)
 	if (err < 0)
 		snd_printk(KERN_ERR "not all devices for card %i can be disconnected\n", card->number);
 
+	snd_info_card_disconnect(card);
 	return 0;	
 }
 
@@ -360,7 +361,7 @@ int snd_card_free(struct snd_card *card)
 	}
 	if (card->private_free)
 		card->private_free(card);
-	snd_info_unregister(card->proc_id);
+	snd_info_free_entry(card->proc_id);
 	if (snd_info_card_free(card) < 0) {
 		snd_printk(KERN_WARNING "unable to free card info\n");
 		/* Not fatal error */
@@ -625,9 +626,9 @@ int __init snd_card_info_init(void)
 
 int __exit snd_card_info_done(void)
 {
-	snd_info_unregister(snd_card_info_entry);
+	snd_info_free_entry(snd_card_info_entry);
 #ifdef MODULE
-	snd_info_unregister(snd_card_module_info_entry);
+	snd_info_free_entry(snd_card_module_info_entry);
 #endif
 	return 0;
 }
diff --git a/sound/core/oss/mixer_oss.c b/sound/core/oss/mixer_oss.c
index 75a9505c7445..00c95def95aa 100644
--- a/sound/core/oss/mixer_oss.c
+++ b/sound/core/oss/mixer_oss.c
@@ -1193,10 +1193,8 @@ static void snd_mixer_oss_proc_init(struct snd_mixer_oss *mixer)
 
 static void snd_mixer_oss_proc_done(struct snd_mixer_oss *mixer)
 {
-	if (mixer->proc_entry) {
-		snd_info_unregister(mixer->proc_entry);
-		mixer->proc_entry = NULL;
-	}
+	snd_info_free_entry(mixer->proc_entry);
+	mixer->proc_entry = NULL;
 }
 #else /* !CONFIG_PROC_FS */
 #define snd_mixer_oss_proc_init(mix)
diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c
index 472fce0ee0e8..a92b93e5ebd5 100644
--- a/sound/core/oss/pcm_oss.c
+++ b/sound/core/oss/pcm_oss.c
@@ -2846,11 +2846,9 @@ static void snd_pcm_oss_proc_done(struct snd_pcm *pcm)
 	int stream;
 	for (stream = 0; stream < 2; ++stream) {
 		struct snd_pcm_str *pstr = &pcm->streams[stream];
-		if (pstr->oss.proc_entry) {
-			snd_info_unregister(pstr->oss.proc_entry);
-			pstr->oss.proc_entry = NULL;
-			snd_pcm_oss_proc_free_setup_list(pstr);
-		}
+		snd_info_free_entry(pstr->oss.proc_entry);
+		pstr->oss.proc_entry = NULL;
+		snd_pcm_oss_proc_free_setup_list(pstr);
 	}
 }
 #else /* !CONFIG_SND_VERBOSE_PROCFS */
diff --git a/sound/core/pcm.c b/sound/core/pcm.c
index 7581edd7b9ff..b8602471f7e5 100644
--- a/sound/core/pcm.c
+++ b/sound/core/pcm.c
@@ -494,19 +494,13 @@ static int snd_pcm_stream_proc_init(struct snd_pcm_str *pstr)
 static int snd_pcm_stream_proc_done(struct snd_pcm_str *pstr)
 {
 #ifdef CONFIG_SND_PCM_XRUN_DEBUG
-	if (pstr->proc_xrun_debug_entry) {
-		snd_info_unregister(pstr->proc_xrun_debug_entry);
-		pstr->proc_xrun_debug_entry = NULL;
-	}
+	snd_info_free_entry(pstr->proc_xrun_debug_entry);
+	pstr->proc_xrun_debug_entry = NULL;
 #endif
-	if (pstr->proc_info_entry) {
-		snd_info_unregister(pstr->proc_info_entry);
-		pstr->proc_info_entry = NULL;
-	}
-	if (pstr->proc_root) {
-		snd_info_unregister(pstr->proc_root);
-		pstr->proc_root = NULL;
-	}
+	snd_info_free_entry(pstr->proc_info_entry);
+	pstr->proc_info_entry = NULL;
+	snd_info_free_entry(pstr->proc_root);
+	pstr->proc_root = NULL;
 	return 0;
 }
 
@@ -570,29 +564,19 @@ static int snd_pcm_substream_proc_init(struct snd_pcm_substream *substream)
 
 	return 0;
 }
-		
+
 static int snd_pcm_substream_proc_done(struct snd_pcm_substream *substream)
 {
-	if (substream->proc_info_entry) {
-		snd_info_unregister(substream->proc_info_entry);
-		substream->proc_info_entry = NULL;
-	}
-	if (substream->proc_hw_params_entry) {
-		snd_info_unregister(substream->proc_hw_params_entry);
-		substream->proc_hw_params_entry = NULL;
-	}
-	if (substream->proc_sw_params_entry) {
-		snd_info_unregister(substream->proc_sw_params_entry);
-		substream->proc_sw_params_entry = NULL;
-	}
-	if (substream->proc_status_entry) {
-		snd_info_unregister(substream->proc_status_entry);
-		substream->proc_status_entry = NULL;
-	}
-	if (substream->proc_root) {
-		snd_info_unregister(substream->proc_root);
-		substream->proc_root = NULL;
-	}
+	snd_info_free_entry(substream->proc_info_entry);
+	substream->proc_info_entry = NULL;
+	snd_info_free_entry(substream->proc_hw_params_entry);
+	substream->proc_hw_params_entry = NULL;
+	snd_info_free_entry(substream->proc_sw_params_entry);
+	substream->proc_sw_params_entry = NULL;
+	snd_info_free_entry(substream->proc_status_entry);
+	substream->proc_status_entry = NULL;
+	snd_info_free_entry(substream->proc_root);
+	substream->proc_root = NULL;
 	return 0;
 }
 #else /* !CONFIG_SND_VERBOSE_PROCFS */
@@ -1090,8 +1074,7 @@ static void snd_pcm_proc_init(void)
 
 static void snd_pcm_proc_done(void)
 {
-	if (snd_pcm_proc_entry)
-		snd_info_unregister(snd_pcm_proc_entry);
+	snd_info_free_entry(snd_pcm_proc_entry);
 }
 
 #else /* !CONFIG_PROC_FS */
diff --git a/sound/core/pcm_memory.c b/sound/core/pcm_memory.c
index 067d2056db9a..be030cb4d373 100644
--- a/sound/core/pcm_memory.c
+++ b/sound/core/pcm_memory.c
@@ -101,7 +101,7 @@ int snd_pcm_lib_preallocate_free(struct snd_pcm_substream *substream)
 {
 	snd_pcm_lib_preallocate_dma_free(substream);
 #ifdef CONFIG_SND_VERBOSE_PROCFS
-	snd_info_unregister(substream->proc_prealloc_entry);
+	snd_info_free_entry(substream->proc_prealloc_entry);
 	substream->proc_prealloc_entry = NULL;
 #endif
 	return 0;
diff --git a/sound/core/rawmidi.c b/sound/core/rawmidi.c
index 8c15c66eb4aa..51577c22f8ce 100644
--- a/sound/core/rawmidi.c
+++ b/sound/core/rawmidi.c
@@ -1599,7 +1599,7 @@ static int snd_rawmidi_dev_unregister(struct snd_device *device)
 	mutex_lock(&register_mutex);
 	list_del(&rmidi->list);
 	if (rmidi->proc_entry) {
-		snd_info_unregister(rmidi->proc_entry);
+		snd_info_free_entry(rmidi->proc_entry);
 		rmidi->proc_entry = NULL;
 	}
 #ifdef CONFIG_SND_OSSEMUL
diff --git a/sound/core/seq/oss/seq_oss.c b/sound/core/seq/oss/seq_oss.c
index e7234135641c..92858cf8b6eb 100644
--- a/sound/core/seq/oss/seq_oss.c
+++ b/sound/core/seq/oss/seq_oss.c
@@ -303,8 +303,7 @@ register_proc(void)
 static void
 unregister_proc(void)
 {
-	if (info_entry)
-		snd_info_unregister(info_entry);
+	snd_info_free_entry(info_entry);
 	info_entry = NULL;
 }
 #endif /* CONFIG_PROC_FS */
diff --git a/sound/core/seq/seq_device.c b/sound/core/seq/seq_device.c
index 102ff548ce69..b85954e956d4 100644
--- a/sound/core/seq/seq_device.c
+++ b/sound/core/seq/seq_device.c
@@ -573,7 +573,7 @@ static void __exit alsa_seq_device_exit(void)
 {
 	remove_drivers();
 #ifdef CONFIG_PROC_FS
-	snd_info_unregister(info_entry);
+	snd_info_free_entry(info_entry);
 #endif
 	if (num_ops)
 		snd_printk(KERN_ERR "drivers not released (%d)\n", num_ops);
diff --git a/sound/core/seq/seq_info.c b/sound/core/seq/seq_info.c
index 142e9e6882c9..8a7fe5cca1c9 100644
--- a/sound/core/seq/seq_info.c
+++ b/sound/core/seq/seq_info.c
@@ -64,9 +64,9 @@ int __init snd_seq_info_init(void)
 
 int __exit snd_seq_info_done(void)
 {
-	snd_info_unregister(queues_entry);
-	snd_info_unregister(clients_entry);
-	snd_info_unregister(timer_entry);
+	snd_info_free_entry(queues_entry);
+	snd_info_free_entry(clients_entry);
+	snd_info_free_entry(timer_entry);
 	return 0;
 }
 #endif
diff --git a/sound/core/sound.c b/sound/core/sound.c
index 7edd1fc58b17..b4430db3fa4c 100644
--- a/sound/core/sound.c
+++ b/sound/core/sound.c
@@ -387,8 +387,7 @@ int __init snd_minor_info_init(void)
 
 int __exit snd_minor_info_done(void)
 {
-	if (snd_minor_info_entry)
-		snd_info_unregister(snd_minor_info_entry);
+	snd_info_free_entry(snd_minor_info_entry);
 	return 0;
 }
 #endif /* CONFIG_PROC_FS */
diff --git a/sound/core/sound_oss.c b/sound/core/sound_oss.c
index 74f0fe5a1ba0..b2fc40aa520b 100644
--- a/sound/core/sound_oss.c
+++ b/sound/core/sound_oss.c
@@ -270,8 +270,7 @@ int __init snd_minor_info_oss_init(void)
 
 int __exit snd_minor_info_oss_done(void)
 {
-	if (snd_minor_info_oss_entry)
-		snd_info_unregister(snd_minor_info_oss_entry);
+	snd_info_free_entry(snd_minor_info_oss_entry);
 	return 0;
 }
 #endif /* CONFIG_PROC_FS */
diff --git a/sound/core/timer.c b/sound/core/timer.c
index 0a984e881c10..52ecbe1e9abb 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -1126,7 +1126,7 @@ static void __init snd_timer_proc_init(void)
 
 static void __exit snd_timer_proc_done(void)
 {
-	snd_info_unregister(snd_timer_proc_entry);
+	snd_info_free_entry(snd_timer_proc_entry);
 }
 #else /* !CONFIG_PROC_FS */
 #define snd_timer_proc_init()
diff --git a/sound/drivers/opl4/opl4_proc.c b/sound/drivers/opl4/opl4_proc.c
index e552ec34166f..11dd811771a4 100644
--- a/sound/drivers/opl4/opl4_proc.c
+++ b/sound/drivers/opl4/opl4_proc.c
@@ -159,8 +159,7 @@ int snd_opl4_create_proc(struct snd_opl4 *opl4)
 
 void snd_opl4_free_proc(struct snd_opl4 *opl4)
 {
-	if (opl4->proc_entry)
-		snd_info_unregister(opl4->proc_entry);
+	snd_info_free_entry(opl4->proc_entry);
 }
 
 #endif /* CONFIG_PROC_FS */
diff --git a/sound/pci/ac97/ac97_proc.c b/sound/pci/ac97/ac97_proc.c
index 2118df50b9d6..a3fdd7da911c 100644
--- a/sound/pci/ac97/ac97_proc.c
+++ b/sound/pci/ac97/ac97_proc.c
@@ -457,14 +457,10 @@ void snd_ac97_proc_init(struct snd_ac97 * ac97)
 
 void snd_ac97_proc_done(struct snd_ac97 * ac97)
 {
-	if (ac97->proc_regs) {
-		snd_info_unregister(ac97->proc_regs);
-		ac97->proc_regs = NULL;
-	}
-	if (ac97->proc) {
-		snd_info_unregister(ac97->proc);
-		ac97->proc = NULL;
-	}
+	snd_info_free_entry(ac97->proc_regs);
+	ac97->proc_regs = NULL;
+	snd_info_free_entry(ac97->proc);
+	ac97->proc = NULL;
 }
 
 void snd_ac97_bus_proc_init(struct snd_ac97_bus * bus)
@@ -485,8 +481,6 @@ void snd_ac97_bus_proc_init(struct snd_ac97_bus * bus)
 
 void snd_ac97_bus_proc_done(struct snd_ac97_bus * bus)
 {
-	if (bus->proc) {
-		snd_info_unregister(bus->proc);
-		bus->proc = NULL;
-	}
+	snd_info_free_entry(bus->proc);
+	bus->proc = NULL;
 }
diff --git a/sound/pci/cs46xx/dsp_spos.c b/sound/pci/cs46xx/dsp_spos.c
index 5c9711c0265c..89c402770a1d 100644
--- a/sound/pci/cs46xx/dsp_spos.c
+++ b/sound/pci/cs46xx/dsp_spos.c
@@ -868,35 +868,23 @@ int cs46xx_dsp_proc_done (struct snd_cs46xx *chip)
 	struct dsp_spos_instance * ins = chip->dsp_spos_instance;
 	int i;
 
-	if (ins->proc_sym_info_entry) {
-		snd_info_unregister(ins->proc_sym_info_entry);
-		ins->proc_sym_info_entry = NULL;
-	}
-  
-	if (ins->proc_modules_info_entry) {
-		snd_info_unregister(ins->proc_modules_info_entry);
-		ins->proc_modules_info_entry = NULL;
-	}
- 
-	if (ins->proc_parameter_dump_info_entry) {
-		snd_info_unregister(ins->proc_parameter_dump_info_entry);
-		ins->proc_parameter_dump_info_entry = NULL;
-	}
-  
-	if (ins->proc_sample_dump_info_entry) {
-		snd_info_unregister(ins->proc_sample_dump_info_entry);
-		ins->proc_sample_dump_info_entry = NULL;
-	}
-  
-	if (ins->proc_scb_info_entry) {
-		snd_info_unregister(ins->proc_scb_info_entry);
-		ins->proc_scb_info_entry = NULL;
-	}
-  
-	if (ins->proc_task_info_entry) {
-		snd_info_unregister(ins->proc_task_info_entry);
-		ins->proc_task_info_entry = NULL;
-	}
+	snd_info_free_entry(ins->proc_sym_info_entry);
+	ins->proc_sym_info_entry = NULL;
+
+	snd_info_free_entry(ins->proc_modules_info_entry);
+	ins->proc_modules_info_entry = NULL;
+
+	snd_info_free_entry(ins->proc_parameter_dump_info_entry);
+	ins->proc_parameter_dump_info_entry = NULL;
+
+	snd_info_free_entry(ins->proc_sample_dump_info_entry);
+	ins->proc_sample_dump_info_entry = NULL;
+
+	snd_info_free_entry(ins->proc_scb_info_entry);
+	ins->proc_scb_info_entry = NULL;
+
+	snd_info_free_entry(ins->proc_task_info_entry);
+	ins->proc_task_info_entry = NULL;
 
 	mutex_lock(&chip->spos_mutex);
 	for (i = 0; i < ins->nscb; ++i) {
@@ -905,10 +893,8 @@ int cs46xx_dsp_proc_done (struct snd_cs46xx *chip)
 	}
 	mutex_unlock(&chip->spos_mutex);
 
-	if (ins->proc_dsp_dir) {
-		snd_info_unregister (ins->proc_dsp_dir);
-		ins->proc_dsp_dir = NULL;
-	}
+	snd_info_free_entry(ins->proc_dsp_dir);
+	ins->proc_dsp_dir = NULL;
 
 	return 0;
 }
diff --git a/sound/pci/cs46xx/dsp_spos_scb_lib.c b/sound/pci/cs46xx/dsp_spos_scb_lib.c
index 232b337852ff..343f51d5311b 100644
--- a/sound/pci/cs46xx/dsp_spos_scb_lib.c
+++ b/sound/pci/cs46xx/dsp_spos_scb_lib.c
@@ -233,7 +233,7 @@ void cs46xx_dsp_proc_free_scb_desc (struct dsp_scb_descriptor * scb)
 
 		snd_printdd("cs46xx_dsp_proc_free_scb_desc: freeing %s\n",scb->scb_name);
 
-		snd_info_unregister(scb->proc_info);
+		snd_info_free_entry(scb->proc_info);
 		scb->proc_info = NULL;
 
 		snd_assert (scb_info != NULL, return);
diff --git a/sound/synth/emux/emux_proc.c b/sound/synth/emux/emux_proc.c
index 58b9601f3ad0..59144ec026e4 100644
--- a/sound/synth/emux/emux_proc.c
+++ b/sound/synth/emux/emux_proc.c
@@ -128,10 +128,8 @@ void snd_emux_proc_init(struct snd_emux *emu, struct snd_card *card, int device)
 
 void snd_emux_proc_free(struct snd_emux *emu)
 {
-	if (emu->proc) {
-		snd_info_unregister(emu->proc);
-		emu->proc = NULL;
-	}
+	snd_info_free_entry(emu->proc);
+	emu->proc = NULL;
 }
 
 #endif /* CONFIG_PROC_FS */
-- 
cgit v1.2.3


From c461482c8072bb073e6146db320d3da85cdc89ad Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 23 Jun 2006 14:38:23 +0200
Subject: [ALSA] Unregister device files at disconnection

Orignally proposed by Sam Revitch <sam.revitch@gmail.com>.
Unregister device files at disconnection to avoid the futher accesses.
Also, the dev_unregister callback is removed and replaced with the
combination of disconnect + free.
A new function snd_card_free_when_closed() is introduced, which is
used in USB disconnect callback.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/core.h        |  3 +-
 include/sound/timer.h       |  1 -
 sound/core/control.c        | 27 ++++++------------
 sound/core/device.c         | 20 ++++++-------
 sound/core/hwdep.c          | 10 +++----
 sound/core/init.c           | 69 +++++++++++++++++++++++++++++++++++----------
 sound/core/oss/mixer_oss.c  | 16 +++++------
 sound/core/oss/pcm_oss.c    | 16 +++++------
 sound/core/pcm.c            | 48 ++++++++++++-------------------
 sound/core/rawmidi.c        | 35 ++++++-----------------
 sound/core/rtctimer.c       |  2 +-
 sound/core/seq/seq_device.c | 11 --------
 sound/core/timer.c          | 52 +++++++++++++++-------------------
 sound/pci/ac97/ac97_codec.c |  8 +++---
 sound/usb/usbaudio.c        |  2 +-
 15 files changed, 148 insertions(+), 172 deletions(-)

(limited to 'include')

diff --git a/include/sound/core.h b/include/sound/core.h
index bab3ff457e40..cf4001cf6248 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -71,7 +71,6 @@ struct snd_device_ops {
 	int (*dev_free)(struct snd_device *dev);
 	int (*dev_register)(struct snd_device *dev);
 	int (*dev_disconnect)(struct snd_device *dev);
-	int (*dev_unregister)(struct snd_device *dev);
 };
 
 struct snd_device {
@@ -131,6 +130,7 @@ struct snd_card {
 								state */
 	spinlock_t files_lock;		/* lock the files for this card */
 	int shutdown;			/* this card is going down */
+	int free_on_last_close;		/* free in context of file_release */
 	wait_queue_head_t shutdown_sleep;
 	struct work_struct free_workq;	/* for free in workqueue */
 	struct device *dev;
@@ -244,6 +244,7 @@ struct snd_card *snd_card_new(int idx, const char *id,
 			 struct module *module, int extra_size);
 int snd_card_disconnect(struct snd_card *card);
 int snd_card_free(struct snd_card *card);
+int snd_card_free_when_closed(struct snd_card *card);
 int snd_card_free_in_thread(struct snd_card *card);
 int snd_card_register(struct snd_card *card);
 int snd_card_info_init(void);
diff --git a/include/sound/timer.h b/include/sound/timer.h
index 5ece2bf541dc..d42c083db1da 100644
--- a/include/sound/timer.h
+++ b/include/sound/timer.h
@@ -129,7 +129,6 @@ void snd_timer_notify(struct snd_timer *timer, int event, struct timespec *tstam
 int snd_timer_global_new(char *id, int device, struct snd_timer **rtimer);
 int snd_timer_global_free(struct snd_timer *timer);
 int snd_timer_global_register(struct snd_timer *timer);
-int snd_timer_global_unregister(struct snd_timer *timer);
 
 int snd_timer_open(struct snd_timer_instance **ti, char *owner, struct snd_timer_id *tid, unsigned int slave_id);
 int snd_timer_close(struct snd_timer_instance *timeri);
diff --git a/sound/core/control.c b/sound/core/control.c
index e9c8854d2f7b..f0c7272a2d48 100644
--- a/sound/core/control.c
+++ b/sound/core/control.c
@@ -1375,6 +1375,11 @@ static int snd_ctl_dev_disconnect(struct snd_device *device)
 	struct snd_card *card = device->device_data;
 	struct list_head *flist;
 	struct snd_ctl_file *ctl;
+	int err, cardnum;
+
+	snd_assert(card != NULL, return -ENXIO);
+	cardnum = card->number;
+	snd_assert(cardnum >= 0 && cardnum < SNDRV_CARDS, return -ENXIO);
 
 	down_read(&card->controls_rwsem);
 	list_for_each(flist, &card->ctl_files) {
@@ -1383,6 +1388,10 @@ static int snd_ctl_dev_disconnect(struct snd_device *device)
 		kill_fasync(&ctl->fasync, SIGIO, POLL_ERR);
 	}
 	up_read(&card->controls_rwsem);
+
+	if ((err = snd_unregister_device(SNDRV_DEVICE_TYPE_CONTROL,
+					 card, -1)) < 0)
+		return err;
 	return 0;
 }
 
@@ -1403,23 +1412,6 @@ static int snd_ctl_dev_free(struct snd_device *device)
 	return 0;
 }
 
-/*
- * de-registration of the control device
- */
-static int snd_ctl_dev_unregister(struct snd_device *device)
-{
-	struct snd_card *card = device->device_data;
-	int err, cardnum;
-
-	snd_assert(card != NULL, return -ENXIO);
-	cardnum = card->number;
-	snd_assert(cardnum >= 0 && cardnum < SNDRV_CARDS, return -ENXIO);
-	if ((err = snd_unregister_device(SNDRV_DEVICE_TYPE_CONTROL,
-					 card, -1)) < 0)
-		return err;
-	return snd_ctl_dev_free(device);
-}
-
 /*
  * create control core:
  * called from init.c
@@ -1430,7 +1422,6 @@ int snd_ctl_create(struct snd_card *card)
 		.dev_free = snd_ctl_dev_free,
 		.dev_register =	snd_ctl_dev_register,
 		.dev_disconnect = snd_ctl_dev_disconnect,
-		.dev_unregister = snd_ctl_dev_unregister
 	};
 
 	snd_assert(card != NULL, return -ENXIO);
diff --git a/sound/core/device.c b/sound/core/device.c
index 6ce4da4a1081..ccb25816ac9e 100644
--- a/sound/core/device.c
+++ b/sound/core/device.c
@@ -71,7 +71,7 @@ EXPORT_SYMBOL(snd_device_new);
  * @device_data: the data pointer to release
  *
  * Removes the device from the list on the card and invokes the
- * callback, dev_unregister or dev_free, corresponding to the state.
+ * callbacks, dev_disconnect and dev_free, corresponding to the state.
  * Then release the device.
  *
  * Returns zero if successful, or a negative error code on failure or if the
@@ -90,16 +90,14 @@ int snd_device_free(struct snd_card *card, void *device_data)
 			continue;
 		/* unlink */
 		list_del(&dev->list);
-		if ((dev->state == SNDRV_DEV_REGISTERED ||
-		     dev->state == SNDRV_DEV_DISCONNECTED) &&
-		    dev->ops->dev_unregister) {
-			if (dev->ops->dev_unregister(dev))
-				snd_printk(KERN_ERR "device unregister failure\n");
-		} else {
-			if (dev->ops->dev_free) {
-				if (dev->ops->dev_free(dev))
-					snd_printk(KERN_ERR "device free failure\n");
-			}
+		if (dev->state == SNDRV_DEV_REGISTERED &&
+		    dev->ops->dev_disconnect)
+			if (dev->ops->dev_disconnect(dev))
+				snd_printk(KERN_ERR
+					   "device disconnect failure\n");
+		if (dev->ops->dev_free) {
+			if (dev->ops->dev_free(dev))
+				snd_printk(KERN_ERR "device free failure\n");
 		}
 		kfree(dev);
 		return 0;
diff --git a/sound/core/hwdep.c b/sound/core/hwdep.c
index cbd8a63282b6..9aa9d94891f0 100644
--- a/sound/core/hwdep.c
+++ b/sound/core/hwdep.c
@@ -42,7 +42,7 @@ static DEFINE_MUTEX(register_mutex);
 static int snd_hwdep_free(struct snd_hwdep *hwdep);
 static int snd_hwdep_dev_free(struct snd_device *device);
 static int snd_hwdep_dev_register(struct snd_device *device);
-static int snd_hwdep_dev_unregister(struct snd_device *device);
+static int snd_hwdep_dev_disconnect(struct snd_device *device);
 
 
 static struct snd_hwdep *snd_hwdep_search(struct snd_card *card, int device)
@@ -353,7 +353,7 @@ int snd_hwdep_new(struct snd_card *card, char *id, int device,
 	static struct snd_device_ops ops = {
 		.dev_free = snd_hwdep_dev_free,
 		.dev_register = snd_hwdep_dev_register,
-		.dev_unregister = snd_hwdep_dev_unregister
+		.dev_disconnect = snd_hwdep_dev_disconnect,
 	};
 
 	snd_assert(rhwdep != NULL, return -EINVAL);
@@ -439,7 +439,7 @@ static int snd_hwdep_dev_register(struct snd_device *device)
 	return 0;
 }
 
-static int snd_hwdep_dev_unregister(struct snd_device *device)
+static int snd_hwdep_dev_disconnect(struct snd_device *device)
 {
 	struct snd_hwdep *hwdep = device->device_data;
 
@@ -454,9 +454,9 @@ static int snd_hwdep_dev_unregister(struct snd_device *device)
 		snd_unregister_oss_device(hwdep->oss_type, hwdep->card, hwdep->device);
 #endif
 	snd_unregister_device(SNDRV_DEVICE_TYPE_HWDEP, hwdep->card, hwdep->device);
-	list_del(&hwdep->list);
+	list_del_init(&hwdep->list);
 	mutex_unlock(&register_mutex);
-	return snd_hwdep_free(hwdep);
+	return 0;
 }
 
 #ifdef CONFIG_PROC_FS
diff --git a/sound/core/init.c b/sound/core/init.c
index 1ecb029ff4c9..5850d99d21e3 100644
--- a/sound/core/init.c
+++ b/sound/core/init.c
@@ -327,22 +327,10 @@ EXPORT_SYMBOL(snd_card_disconnect);
  *  Returns zero. Frees all associated devices and frees the control
  *  interface associated to given soundcard.
  */
-int snd_card_free(struct snd_card *card)
+static int snd_card_do_free(struct snd_card *card)
 {
 	struct snd_shutdown_f_ops *s_f_ops;
 
-	if (card == NULL)
-		return -EINVAL;
-	mutex_lock(&snd_card_mutex);
-	snd_cards[card->number] = NULL;
-	mutex_unlock(&snd_card_mutex);
-
-#ifdef CONFIG_PM
-	wake_up(&card->power_sleep);
-#endif
-	/* wait, until all devices are ready for the free operation */
-	wait_event(card->shutdown_sleep, card->files == NULL);
-
 #if defined(CONFIG_SND_MIXER_OSS) || defined(CONFIG_SND_MIXER_OSS_MODULE)
 	if (snd_mixer_oss_notify_callback)
 		snd_mixer_oss_notify_callback(card, SND_MIXER_OSS_NOTIFY_FREE);
@@ -371,10 +359,55 @@ int snd_card_free(struct snd_card *card)
 		card->s_f_ops = s_f_ops->next;
 		kfree(s_f_ops);
 	}
+	kfree(card);
+	return 0;
+}
+
+static int snd_card_free_prepare(struct snd_card *card)
+{
+	if (card == NULL)
+		return -EINVAL;
+	(void) snd_card_disconnect(card);
 	mutex_lock(&snd_card_mutex);
+	snd_cards[card->number] = NULL;
 	snd_cards_lock &= ~(1 << card->number);
 	mutex_unlock(&snd_card_mutex);
-	kfree(card);
+#ifdef CONFIG_PM
+	wake_up(&card->power_sleep);
+#endif
+	return 0;
+}
+
+int snd_card_free_when_closed(struct snd_card *card)
+{
+	int free_now = 0;
+	int ret = snd_card_free_prepare(card);
+	if (ret)
+		return ret;
+
+	spin_lock(&card->files_lock);
+	if (card->files == NULL)
+		free_now = 1;
+	else
+		card->free_on_last_close = 1;
+	spin_unlock(&card->files_lock);
+
+	if (free_now)
+		snd_card_do_free(card);
+	return 0;
+}
+
+EXPORT_SYMBOL(snd_card_free_when_closed);
+
+int snd_card_free(struct snd_card *card)
+{
+	int ret = snd_card_free_prepare(card);
+	if (ret)
+		return ret;
+
+	/* wait, until all devices are ready for the free operation */
+	wait_event(card->shutdown_sleep, card->files == NULL);
+	snd_card_do_free(card);
 	return 0;
 }
 
@@ -718,6 +751,7 @@ EXPORT_SYMBOL(snd_card_file_add);
 int snd_card_file_remove(struct snd_card *card, struct file *file)
 {
 	struct snd_monitor_file *mfile, *pfile = NULL;
+	int last_close = 0;
 
 	spin_lock(&card->files_lock);
 	mfile = card->files;
@@ -732,9 +766,14 @@ int snd_card_file_remove(struct snd_card *card, struct file *file)
 		pfile = mfile;
 		mfile = mfile->next;
 	}
-	spin_unlock(&card->files_lock);
 	if (card->files == NULL)
+		last_close = 1;
+	spin_unlock(&card->files_lock);
+	if (last_close) {
 		wake_up(&card->shutdown_sleep);
+		if (card->free_on_last_close)
+			snd_card_do_free(card);
+	}
 	if (!mfile) {
 		snd_printk(KERN_ERR "ALSA card file remove problem (%p)\n", file);
 		return -ENOENT;
diff --git a/sound/core/oss/mixer_oss.c b/sound/core/oss/mixer_oss.c
index 00c95def95aa..f4c67042e3ac 100644
--- a/sound/core/oss/mixer_oss.c
+++ b/sound/core/oss/mixer_oss.c
@@ -1310,21 +1310,19 @@ static int snd_mixer_oss_notify_handler(struct snd_card *card, int cmd)
 		card->mixer_oss = mixer;
 		snd_mixer_oss_build(mixer);
 		snd_mixer_oss_proc_init(mixer);
-	} else if (cmd == SND_MIXER_OSS_NOTIFY_DISCONNECT) {
-		mixer = card->mixer_oss;
-		if (mixer == NULL || !mixer->oss_dev_alloc)
-			return 0;
-		snd_unregister_oss_device(SNDRV_OSS_DEVICE_TYPE_MIXER, mixer->card, 0);
-		mixer->oss_dev_alloc = 0;
-	} else {		/* free */
+	} else {
 		mixer = card->mixer_oss;
 		if (mixer == NULL)
 			return 0;
+		if (mixer->oss_dev_alloc) {
 #ifdef SNDRV_OSS_INFO_DEV_MIXERS
-		snd_oss_info_unregister(SNDRV_OSS_INFO_DEV_MIXERS, mixer->card->number);
+			snd_oss_info_unregister(SNDRV_OSS_INFO_DEV_MIXERS, mixer->card->number);
 #endif
-		if (mixer->oss_dev_alloc)
 			snd_unregister_oss_device(SNDRV_OSS_DEVICE_TYPE_MIXER, mixer->card, 0);
+			mixer->oss_dev_alloc = 0;
+		}
+		if (cmd == SND_MIXER_OSS_NOTIFY_DISCONNECT)
+			return 0;
 		snd_mixer_oss_proc_done(mixer);
 		return snd_mixer_oss_free1(mixer);
 	}
diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c
index a92b93e5ebd5..505b23ec4058 100644
--- a/sound/core/oss/pcm_oss.c
+++ b/sound/core/oss/pcm_oss.c
@@ -2929,25 +2929,23 @@ static int snd_pcm_oss_disconnect_minor(struct snd_pcm *pcm)
 			snd_unregister_oss_device(SNDRV_OSS_DEVICE_TYPE_PCM,
 						  pcm->card, 1);
 		}
-	}
-	return 0;
-}
-
-static int snd_pcm_oss_unregister_minor(struct snd_pcm *pcm)
-{
-	snd_pcm_oss_disconnect_minor(pcm);
-	if (pcm->oss.reg) {
 		if (dsp_map[pcm->card->number] == (int)pcm->device) {
 #ifdef SNDRV_OSS_INFO_DEV_AUDIO
 			snd_oss_info_unregister(SNDRV_OSS_INFO_DEV_AUDIO, pcm->card->number);
 #endif
 		}
 		pcm->oss.reg = 0;
-		snd_pcm_oss_proc_done(pcm);
 	}
 	return 0;
 }
 
+static int snd_pcm_oss_unregister_minor(struct snd_pcm *pcm)
+{
+	snd_pcm_oss_disconnect_minor(pcm);
+	snd_pcm_oss_proc_done(pcm);
+	return 0;
+}
+
 static struct snd_pcm_notify snd_pcm_oss_notify =
 {
 	.n_register =	snd_pcm_oss_register_minor,
diff --git a/sound/core/pcm.c b/sound/core/pcm.c
index b8602471f7e5..f52178abf120 100644
--- a/sound/core/pcm.c
+++ b/sound/core/pcm.c
@@ -42,7 +42,6 @@ static int snd_pcm_free(struct snd_pcm *pcm);
 static int snd_pcm_dev_free(struct snd_device *device);
 static int snd_pcm_dev_register(struct snd_device *device);
 static int snd_pcm_dev_disconnect(struct snd_device *device);
-static int snd_pcm_dev_unregister(struct snd_device *device);
 
 static struct snd_pcm *snd_pcm_search(struct snd_card *card, int device)
 {
@@ -680,7 +679,6 @@ int snd_pcm_new(struct snd_card *card, char *id, int device,
 		.dev_free = snd_pcm_dev_free,
 		.dev_register =	snd_pcm_dev_register,
 		.dev_disconnect = snd_pcm_dev_disconnect,
-		.dev_unregister = snd_pcm_dev_unregister
 	};
 
 	snd_assert(rpcm != NULL, return -EINVAL);
@@ -724,6 +722,7 @@ static void snd_pcm_free_stream(struct snd_pcm_str * pstr)
 	substream = pstr->substream;
 	while (substream) {
 		substream_next = substream->next;
+		snd_pcm_timer_done(substream);
 		snd_pcm_substream_proc_done(substream);
 		kfree(substream);
 		substream = substream_next;
@@ -740,7 +739,12 @@ static void snd_pcm_free_stream(struct snd_pcm_str * pstr)
 
 static int snd_pcm_free(struct snd_pcm *pcm)
 {
+	struct snd_pcm_notify *notify;
+
 	snd_assert(pcm != NULL, return -ENXIO);
+	list_for_each_entry(notify, &snd_pcm_notify_list, list) {
+		notify->n_unregister(pcm);
+	}
 	if (pcm->private_free)
 		pcm->private_free(pcm);
 	snd_pcm_lib_preallocate_free_for_all(pcm);
@@ -955,35 +959,22 @@ static int snd_pcm_dev_register(struct snd_device *device)
 static int snd_pcm_dev_disconnect(struct snd_device *device)
 {
 	struct snd_pcm *pcm = device->device_data;
-	struct list_head *list;
+	struct snd_pcm_notify *notify;
 	struct snd_pcm_substream *substream;
-	int cidx;
+	int cidx, devtype;
 
 	mutex_lock(&register_mutex);
+	if (list_empty(&pcm->list))
+		goto unlock;
+
 	list_del_init(&pcm->list);
 	for (cidx = 0; cidx < 2; cidx++)
 		for (substream = pcm->streams[cidx].substream; substream; substream = substream->next)
 			if (substream->runtime)
 				substream->runtime->status->state = SNDRV_PCM_STATE_DISCONNECTED;
-	list_for_each(list, &snd_pcm_notify_list) {
-		struct snd_pcm_notify *notify;
-		notify = list_entry(list, struct snd_pcm_notify, list);
+	list_for_each_entry(notify, &snd_pcm_notify_list, list) {
 		notify->n_disconnect(pcm);
 	}
-	mutex_unlock(&register_mutex);
-	return 0;
-}
-
-static int snd_pcm_dev_unregister(struct snd_device *device)
-{
-	int cidx, devtype;
-	struct snd_pcm_substream *substream;
-	struct list_head *list;
-	struct snd_pcm *pcm = device->device_data;
-
-	snd_assert(pcm != NULL, return -ENXIO);
-	mutex_lock(&register_mutex);
-	list_del(&pcm->list);
 	for (cidx = 0; cidx < 2; cidx++) {
 		devtype = -1;
 		switch (cidx) {
@@ -995,23 +986,20 @@ static int snd_pcm_dev_unregister(struct snd_device *device)
 			break;
 		}
 		snd_unregister_device(devtype, pcm->card, pcm->device);
-		for (substream = pcm->streams[cidx].substream; substream; substream = substream->next)
-			snd_pcm_timer_done(substream);
-	}
-	list_for_each(list, &snd_pcm_notify_list) {
-		struct snd_pcm_notify *notify;
-		notify = list_entry(list, struct snd_pcm_notify, list);
-		notify->n_unregister(pcm);
 	}
+ unlock:
 	mutex_unlock(&register_mutex);
-	return snd_pcm_free(pcm);
+	return 0;
 }
 
 int snd_pcm_notify(struct snd_pcm_notify *notify, int nfree)
 {
 	struct list_head *p;
 
-	snd_assert(notify != NULL && notify->n_register != NULL && notify->n_unregister != NULL, return -EINVAL);
+	snd_assert(notify != NULL &&
+		   notify->n_register != NULL &&
+		   notify->n_unregister != NULL &&
+		   notify->n_disconnect, return -EINVAL);
 	mutex_lock(&register_mutex);
 	if (nfree) {
 		list_del(&notify->list);
diff --git a/sound/core/rawmidi.c b/sound/core/rawmidi.c
index 51577c22f8ce..8a2bdfae63e3 100644
--- a/sound/core/rawmidi.c
+++ b/sound/core/rawmidi.c
@@ -55,7 +55,6 @@ static int snd_rawmidi_free(struct snd_rawmidi *rawmidi);
 static int snd_rawmidi_dev_free(struct snd_device *device);
 static int snd_rawmidi_dev_register(struct snd_device *device);
 static int snd_rawmidi_dev_disconnect(struct snd_device *device);
-static int snd_rawmidi_dev_unregister(struct snd_device *device);
 
 static LIST_HEAD(snd_rawmidi_devices);
 static DEFINE_MUTEX(register_mutex);
@@ -1426,7 +1425,6 @@ int snd_rawmidi_new(struct snd_card *card, char *id, int device,
 		.dev_free = snd_rawmidi_dev_free,
 		.dev_register = snd_rawmidi_dev_register,
 		.dev_disconnect = snd_rawmidi_dev_disconnect,
-		.dev_unregister = snd_rawmidi_dev_unregister
 	};
 
 	snd_assert(rrawmidi != NULL, return -EINVAL);
@@ -1479,6 +1477,14 @@ static void snd_rawmidi_free_substreams(struct snd_rawmidi_str *stream)
 static int snd_rawmidi_free(struct snd_rawmidi *rmidi)
 {
 	snd_assert(rmidi != NULL, return -ENXIO);	
+
+	snd_info_free_entry(rmidi->proc_entry);
+	rmidi->proc_entry = NULL;
+	mutex_lock(&register_mutex);
+	if (rmidi->ops && rmidi->ops->dev_unregister)
+		rmidi->ops->dev_unregister(rmidi);
+	mutex_unlock(&register_mutex);
+
 	snd_rawmidi_free_substreams(&rmidi->streams[SNDRV_RAWMIDI_STREAM_INPUT]);
 	snd_rawmidi_free_substreams(&rmidi->streams[SNDRV_RAWMIDI_STREAM_OUTPUT]);
 	if (rmidi->private_free)
@@ -1587,21 +1593,6 @@ static int snd_rawmidi_dev_disconnect(struct snd_device *device)
 
 	mutex_lock(&register_mutex);
 	list_del_init(&rmidi->list);
-	mutex_unlock(&register_mutex);
-	return 0;
-}
-
-static int snd_rawmidi_dev_unregister(struct snd_device *device)
-{
-	struct snd_rawmidi *rmidi = device->device_data;
-
-	snd_assert(rmidi != NULL, return -ENXIO);
-	mutex_lock(&register_mutex);
-	list_del(&rmidi->list);
-	if (rmidi->proc_entry) {
-		snd_info_free_entry(rmidi->proc_entry);
-		rmidi->proc_entry = NULL;
-	}
 #ifdef CONFIG_SND_OSSEMUL
 	if (rmidi->ossreg) {
 		if ((int)rmidi->device == midi_map[rmidi->card->number]) {
@@ -1615,17 +1606,9 @@ static int snd_rawmidi_dev_unregister(struct snd_device *device)
 		rmidi->ossreg = 0;
 	}
 #endif /* CONFIG_SND_OSSEMUL */
-	if (rmidi->ops && rmidi->ops->dev_unregister)
-		rmidi->ops->dev_unregister(rmidi);
 	snd_unregister_device(SNDRV_DEVICE_TYPE_RAWMIDI, rmidi->card, rmidi->device);
 	mutex_unlock(&register_mutex);
-#if defined(CONFIG_SND_SEQUENCER) || (defined(MODULE) && defined(CONFIG_SND_SEQUENCER_MODULE))
-	if (rmidi->seq_dev) {
-		snd_device_free(rmidi->card, rmidi->seq_dev);
-		rmidi->seq_dev = NULL;
-	}
-#endif
-	return snd_rawmidi_free(rmidi);
+	return 0;
 }
 
 /**
diff --git a/sound/core/rtctimer.c b/sound/core/rtctimer.c
index 84704ccb1829..412dd62b654e 100644
--- a/sound/core/rtctimer.c
+++ b/sound/core/rtctimer.c
@@ -156,7 +156,7 @@ static int __init rtctimer_init(void)
 static void __exit rtctimer_exit(void)
 {
 	if (rtctimer) {
-		snd_timer_global_unregister(rtctimer);
+		snd_timer_global_free(rtctimer);
 		rtctimer = NULL;
 	}
 }
diff --git a/sound/core/seq/seq_device.c b/sound/core/seq/seq_device.c
index b85954e956d4..b79d011813c0 100644
--- a/sound/core/seq/seq_device.c
+++ b/sound/core/seq/seq_device.c
@@ -90,7 +90,6 @@ static int snd_seq_device_free(struct snd_seq_device *dev);
 static int snd_seq_device_dev_free(struct snd_device *device);
 static int snd_seq_device_dev_register(struct snd_device *device);
 static int snd_seq_device_dev_disconnect(struct snd_device *device);
-static int snd_seq_device_dev_unregister(struct snd_device *device);
 
 static int init_device(struct snd_seq_device *dev, struct ops_list *ops);
 static int free_device(struct snd_seq_device *dev, struct ops_list *ops);
@@ -189,7 +188,6 @@ int snd_seq_device_new(struct snd_card *card, int device, char *id, int argsize,
 		.dev_free = snd_seq_device_dev_free,
 		.dev_register = snd_seq_device_dev_register,
 		.dev_disconnect = snd_seq_device_dev_disconnect,
-		.dev_unregister = snd_seq_device_dev_unregister
 	};
 
 	if (result)
@@ -308,15 +306,6 @@ static int snd_seq_device_dev_disconnect(struct snd_device *device)
 	return 0;
 }
 
-/*
- * unregister the existing device
- */
-static int snd_seq_device_dev_unregister(struct snd_device *device)
-{
-	struct snd_seq_device *dev = device->device_data;
-	return snd_seq_device_free(dev);
-}
-
 /*
  * register device driver
  * id = driver id
diff --git a/sound/core/timer.c b/sound/core/timer.c
index 52ecbe1e9abb..7e5e562fe356 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -88,7 +88,7 @@ static DEFINE_MUTEX(register_mutex);
 static int snd_timer_free(struct snd_timer *timer);
 static int snd_timer_dev_free(struct snd_device *device);
 static int snd_timer_dev_register(struct snd_device *device);
-static int snd_timer_dev_unregister(struct snd_device *device);
+static int snd_timer_dev_disconnect(struct snd_device *device);
 
 static void snd_timer_reschedule(struct snd_timer * timer, unsigned long ticks_left);
 
@@ -773,7 +773,7 @@ int snd_timer_new(struct snd_card *card, char *id, struct snd_timer_id *tid,
 	static struct snd_device_ops ops = {
 		.dev_free = snd_timer_dev_free,
 		.dev_register = snd_timer_dev_register,
-		.dev_unregister = snd_timer_dev_unregister
+		.dev_disconnect = snd_timer_dev_disconnect,
 	};
 
 	snd_assert(tid != NULL, return -EINVAL);
@@ -813,6 +813,21 @@ int snd_timer_new(struct snd_card *card, char *id, struct snd_timer_id *tid,
 static int snd_timer_free(struct snd_timer *timer)
 {
 	snd_assert(timer != NULL, return -ENXIO);
+
+	mutex_lock(&register_mutex);
+	if (! list_empty(&timer->open_list_head)) {
+		struct list_head *p, *n;
+		struct snd_timer_instance *ti;
+		snd_printk(KERN_WARNING "timer %p is busy?\n", timer);
+		list_for_each_safe(p, n, &timer->open_list_head) {
+			list_del_init(p);
+			ti = list_entry(p, struct snd_timer_instance, open_list);
+			ti->timer = NULL;
+		}
+	}
+	list_del(&timer->device_list);
+	mutex_unlock(&register_mutex);
+
 	if (timer->private_free)
 		timer->private_free(timer);
 	kfree(timer);
@@ -867,30 +882,13 @@ static int snd_timer_dev_register(struct snd_device *dev)
 	return 0;
 }
 
-static int snd_timer_unregister(struct snd_timer *timer)
+static int snd_timer_dev_disconnect(struct snd_device *device)
 {
-	struct list_head *p, *n;
-	struct snd_timer_instance *ti;
-
-	snd_assert(timer != NULL, return -ENXIO);
+	struct snd_timer *timer = device->device_data;
 	mutex_lock(&register_mutex);
-	if (! list_empty(&timer->open_list_head)) {
-		snd_printk(KERN_WARNING "timer 0x%lx is busy?\n", (long)timer);
-		list_for_each_safe(p, n, &timer->open_list_head) {
-			list_del_init(p);
-			ti = list_entry(p, struct snd_timer_instance, open_list);
-			ti->timer = NULL;
-		}
-	}
-	list_del(&timer->device_list);
+	list_del_init(&timer->device_list);
 	mutex_unlock(&register_mutex);
-	return snd_timer_free(timer);
-}
-
-static int snd_timer_dev_unregister(struct snd_device *device)
-{
-	struct snd_timer *timer = device->device_data;
-	return snd_timer_unregister(timer);
+	return 0;
 }
 
 void snd_timer_notify(struct snd_timer *timer, int event, struct timespec *tstamp)
@@ -955,11 +953,6 @@ int snd_timer_global_register(struct snd_timer *timer)
 	return snd_timer_dev_register(&dev);
 }
 
-int snd_timer_global_unregister(struct snd_timer *timer)
-{
-	return snd_timer_unregister(timer);
-}
-
 /*
  *  System timer
  */
@@ -1982,7 +1975,7 @@ static void __exit alsa_timer_exit(void)
 	/* unregister the system timer */
 	list_for_each_safe(p, n, &snd_timer_list) {
 		struct snd_timer *timer = list_entry(p, struct snd_timer, device_list);
-		snd_timer_unregister(timer);
+		snd_timer_free(timer);
 	}
 	snd_timer_proc_done();
 #ifdef SNDRV_OSS_INFO_DEV_TIMERS
@@ -2005,5 +1998,4 @@ EXPORT_SYMBOL(snd_timer_notify);
 EXPORT_SYMBOL(snd_timer_global_new);
 EXPORT_SYMBOL(snd_timer_global_free);
 EXPORT_SYMBOL(snd_timer_global_register);
-EXPORT_SYMBOL(snd_timer_global_unregister);
 EXPORT_SYMBOL(snd_timer_interrupt);
diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c
index 51e83d7a839a..b35280ca2465 100644
--- a/sound/pci/ac97/ac97_codec.c
+++ b/sound/pci/ac97/ac97_codec.c
@@ -1817,13 +1817,13 @@ static int snd_ac97_dev_register(struct snd_device *device)
 	return 0;
 }
 
-/* unregister ac97 codec */
-static int snd_ac97_dev_unregister(struct snd_device *device)
+/* disconnect ac97 codec */
+static int snd_ac97_dev_disconnect(struct snd_device *device)
 {
 	struct snd_ac97 *ac97 = device->device_data;
 	if (ac97->dev.bus)
 		device_unregister(&ac97->dev);
-	return snd_ac97_free(ac97);
+	return 0;
 }
 
 /* build_ops to do nothing */
@@ -1860,7 +1860,7 @@ int snd_ac97_mixer(struct snd_ac97_bus *bus, struct snd_ac97_template *template,
 	static struct snd_device_ops ops = {
 		.dev_free =	snd_ac97_dev_free,
 		.dev_register =	snd_ac97_dev_register,
-		.dev_unregister =	snd_ac97_dev_unregister,
+		.dev_disconnect =	snd_ac97_dev_disconnect,
 	};
 
 	snd_assert(rac97 != NULL, return -EINVAL);
diff --git a/sound/usb/usbaudio.c b/sound/usb/usbaudio.c
index 1b7f499c549d..314431385913 100644
--- a/sound/usb/usbaudio.c
+++ b/sound/usb/usbaudio.c
@@ -3499,7 +3499,7 @@ static void snd_usb_audio_disconnect(struct usb_device *dev, void *ptr)
 		}
 		usb_chip[chip->index] = NULL;
 		mutex_unlock(&register_mutex);
-		snd_card_free(card);
+		snd_card_free_when_closed(card);
 	} else {
 		mutex_unlock(&register_mutex);
 	}
-- 
cgit v1.2.3


From 2b29b13c5794f648cd5e839796496704d787f5a6 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 23 Jun 2006 14:38:26 +0200
Subject: [ALSA] Deprecate snd_card_free_in_thread()

Deprecated snd_card_free_in_thread(), replaced with
snd_card_free_when_closed().

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 .../sound/alsa/DocBook/writing-an-alsa-driver.tmpl |  5 +-
 include/sound/core.h                               |  3 --
 sound/core/init.c                                  | 56 ++--------------------
 sound/drivers/mpu401/mpu401.c                      |  2 +-
 sound/pcmcia/pdaudiocf/pdaudiocf.c                 |  2 +-
 sound/pcmcia/vx/vxpocket.c                         |  4 +-
 6 files changed, 9 insertions(+), 63 deletions(-)

(limited to 'include')

diff --git a/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl b/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl
index b8dc51ca776c..4807ef79a94d 100644
--- a/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl
+++ b/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl
@@ -1054,9 +1054,8 @@
 
       <para>
       For a device which allows hotplugging, you can use
-      <function>snd_card_free_in_thread</function>.  This one will
-      postpone the destruction and wait in a kernel-thread until all
-      devices are closed.
+      <function>snd_card_free_when_closed</function>.  This one will
+      postpone the destruction until all devices are closed.
       </para>
 
     </section>
diff --git a/include/sound/core.h b/include/sound/core.h
index cf4001cf6248..1359c532b68e 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -25,7 +25,6 @@
 #include <linux/sched.h>		/* wake_up() */
 #include <linux/mutex.h>		/* struct mutex */
 #include <linux/rwsem.h>		/* struct rw_semaphore */
-#include <linux/workqueue.h>		/* struct workqueue_struct */
 #include <linux/pm.h>			/* pm_message_t */
 
 /* forward declarations */
@@ -132,7 +131,6 @@ struct snd_card {
 	int shutdown;			/* this card is going down */
 	int free_on_last_close;		/* free in context of file_release */
 	wait_queue_head_t shutdown_sleep;
-	struct work_struct free_workq;	/* for free in workqueue */
 	struct device *dev;
 
 #ifdef CONFIG_PM
@@ -245,7 +243,6 @@ struct snd_card *snd_card_new(int idx, const char *id,
 int snd_card_disconnect(struct snd_card *card);
 int snd_card_free(struct snd_card *card);
 int snd_card_free_when_closed(struct snd_card *card);
-int snd_card_free_in_thread(struct snd_card *card);
 int snd_card_register(struct snd_card *card);
 int snd_card_info_init(void);
 int snd_card_info_done(void);
diff --git a/sound/core/init.c b/sound/core/init.c
index 5850d99d21e3..d7607a25acdf 100644
--- a/sound/core/init.c
+++ b/sound/core/init.c
@@ -81,8 +81,6 @@ static inline int init_info_for_card(struct snd_card *card)
 #define init_info_for_card(card)
 #endif
 
-static void snd_card_free_thread(void * __card);
-
 /**
  *  snd_card_new - create and initialize a soundcard structure
  *  @idx: card index (address) [0 ... (SNDRV_CARDS-1)]
@@ -145,7 +143,6 @@ struct snd_card *snd_card_new(int idx, const char *xid,
 	INIT_LIST_HEAD(&card->ctl_files);
 	spin_lock_init(&card->files_lock);
 	init_waitqueue_head(&card->shutdown_sleep);
-	INIT_WORK(&card->free_workq, snd_card_free_thread, card);
 #ifdef CONFIG_PM
 	mutex_init(&card->power_lock);
 	init_waitqueue_head(&card->power_sleep);
@@ -413,53 +410,6 @@ int snd_card_free(struct snd_card *card)
 
 EXPORT_SYMBOL(snd_card_free);
 
-static void snd_card_free_thread(void * __card)
-{
-	struct snd_card *card = __card;
-	struct module * module = card->module;
-
-	if (!try_module_get(module)) {
-		snd_printk(KERN_ERR "unable to lock toplevel module for card %i in free thread\n", card->number);
-		module = NULL;
-	}
-
-	snd_card_free(card);
-
-	module_put(module);
-}
-
-/**
- *  snd_card_free_in_thread - call snd_card_free() in thread
- *  @card: soundcard structure
- *
- *  This function schedules the call of snd_card_free() function in a
- *  work queue.  When all devices are released (non-busy), the work
- *  is woken up and calls snd_card_free().
- *
- *  When a card can be disconnected at any time by hotplug service,
- *  this function should be used in disconnect (or detach) callback
- *  instead of calling snd_card_free() directly.
- *  
- *  Returns - zero otherwise a negative error code if the start of thread failed.
- */
-int snd_card_free_in_thread(struct snd_card *card)
-{
-	if (card->files == NULL) {
-		snd_card_free(card);
-		return 0;
-	}
-
-	if (schedule_work(&card->free_workq))
-		return 0;
-
-	snd_printk(KERN_ERR "schedule_work() failed in snd_card_free_in_thread for card %i\n", card->number);
-	/* try to free the structure immediately */
-	snd_card_free(card);
-	return -EFAULT;
-}
-
-EXPORT_SYMBOL(snd_card_free_in_thread);
-
 static void choose_default_id(struct snd_card *card)
 {
 	int i, len, idx_flag = 0, loops = SNDRV_CARDS;
@@ -742,9 +692,9 @@ EXPORT_SYMBOL(snd_card_file_add);
  *
  *  This function removes the file formerly added to the card via
  *  snd_card_file_add() function.
- *  If all files are removed and the release of the card is
- *  scheduled, it will wake up the the thread to call snd_card_free()
- *  (see snd_card_free_in_thread() function).
+ *  If all files are removed and snd_card_free_when_closed() was
+ *  called beforehand, it processes the pending release of
+ *  resources.
  *
  *  Returns zero or a negative error code.
  */
diff --git a/sound/drivers/mpu401/mpu401.c b/sound/drivers/mpu401/mpu401.c
index 17cc105b26fc..2de181ad0b05 100644
--- a/sound/drivers/mpu401/mpu401.c
+++ b/sound/drivers/mpu401/mpu401.c
@@ -211,7 +211,7 @@ static void __devexit snd_mpu401_pnp_remove(struct pnp_dev *dev)
 	struct snd_card *card = (struct snd_card *) pnp_get_drvdata(dev);
 
 	snd_card_disconnect(card);
-	snd_card_free_in_thread(card);
+	snd_card_free_when_closed(card);
 }
 
 static struct pnp_driver snd_mpu401_pnp_driver = {
diff --git a/sound/pcmcia/pdaudiocf/pdaudiocf.c b/sound/pcmcia/pdaudiocf/pdaudiocf.c
index 1c09e5f49da8..fd3590fcaedb 100644
--- a/sound/pcmcia/pdaudiocf/pdaudiocf.c
+++ b/sound/pcmcia/pdaudiocf/pdaudiocf.c
@@ -206,7 +206,7 @@ static void snd_pdacf_detach(struct pcmcia_device *link)
 		snd_pdacf_powerdown(chip);
 	chip->chip_status |= PDAUDIOCF_STAT_IS_STALE; /* to be sure */
 	snd_card_disconnect(chip->card);
-	snd_card_free_in_thread(chip->card);
+	snd_card_free_when_closed(chip->card);
 }
 
 /*
diff --git a/sound/pcmcia/vx/vxpocket.c b/sound/pcmcia/vx/vxpocket.c
index cafe6640cc1a..76c85cffb40e 100644
--- a/sound/pcmcia/vx/vxpocket.c
+++ b/sound/pcmcia/vx/vxpocket.c
@@ -65,7 +65,7 @@ static void vxpocket_release(struct pcmcia_device *link)
 }
 
 /*
- * destructor, called from snd_card_free_in_thread()
+ * destructor, called from snd_card_free_when_closed()
  */
 static int snd_vxpocket_dev_free(struct snd_device *device)
 {
@@ -363,7 +363,7 @@ static void vxpocket_detach(struct pcmcia_device *link)
 	chip->chip_status |= VX_STAT_IS_STALE; /* to be sure */
 	snd_card_disconnect(chip->card);
 	vxpocket_release(link);
-	snd_card_free_in_thread(chip->card);
+	snd_card_free_when_closed(chip->card);
 }
 
 /*
-- 
cgit v1.2.3


From 6dbe662874ba08585eaf732d126762c25ac8e3f7 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Tue, 27 Jun 2006 18:28:53 +0200
Subject: [ALSA] Add experimental support of aggressive AC97 power-saving mode

Added CONFIG_SND_AC97_POWER_SAVE kernel config to enable the support
of aggressive AC97 power-saving mode.  In this mode, the AC97
powerdown register bits are dynamically controlled at each open/close
of PCM streams.
The mode is activated via power_save option for snd-ac97-codec
driver.  As default it's off.  It can be turned on/off on the fly
via sysfs, too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/ac97_codec.h  |  32 ++++++
 sound/drivers/Kconfig       |  13 +++
 sound/pci/ac97/ac97_codec.c | 264 +++++++++++++++++++++++++++++++++++++-------
 sound/pci/ac97/ac97_pcm.c   |  18 +++
 sound/pci/intel8x0.c        |  14 ++-
 sound/pci/via82xx.c         |  13 ++-
 6 files changed, 315 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/include/sound/ac97_codec.h b/include/sound/ac97_codec.h
index 758f8bf133c7..4c43521cc493 100644
--- a/include/sound/ac97_codec.h
+++ b/include/sound/ac97_codec.h
@@ -27,6 +27,7 @@
 
 #include <linux/bitops.h>
 #include <linux/device.h>
+#include <linux/workqueue.h>
 #include "pcm.h"
 #include "control.h"
 #include "info.h"
@@ -140,6 +141,20 @@
 #define AC97_GP_DRSS_1011	0x0000	/* LR(C) 10+11(+12) */
 #define AC97_GP_DRSS_78		0x0400	/* LR 7+8 */
 
+/* powerdown bits */
+#define AC97_PD_ADC_STATUS	0x0001	/* ADC status (RO) */
+#define AC97_PD_DAC_STATUS	0x0002	/* DAC status (RO) */
+#define AC97_PD_MIXER_STATUS	0x0004	/* Analog mixer status (RO) */
+#define AC97_PD_VREF_STATUS	0x0008	/* Vref status (RO) */
+#define AC97_PD_PR0		0x0100	/* Power down PCM ADCs and input MUX */
+#define AC97_PD_PR1		0x0200	/* Power down PCM front DAC */
+#define AC97_PD_PR2		0x0400	/* Power down Mixer (Vref still on) */
+#define AC97_PD_PR3		0x0800	/* Power down Mixer (Vref off) */
+#define AC97_PD_PR4		0x1000	/* Power down AC-Link */
+#define AC97_PD_PR5		0x2000	/* Disable internal clock usage */
+#define AC97_PD_PR6		0x4000	/* Headphone amplifier */
+#define AC97_PD_EAPD		0x8000	/* External Amplifer Power Down (EAPD) */
+
 /* extended audio ID bit defines */
 #define AC97_EI_VRA		0x0001	/* Variable bit rate supported */
 #define AC97_EI_DRA		0x0002	/* Double rate supported */
@@ -359,6 +374,7 @@
 #define AC97_SCAP_INV_EAPD	(1<<7)	/* inverted EAPD */
 #define AC97_SCAP_DETECT_BY_VENDOR (1<<8) /* use vendor registers for read tests */
 #define AC97_SCAP_NO_SPDIF	(1<<9)	/* don't build SPDIF controls */
+#define AC97_SCAP_EAPD_LED	(1<<10)	/* EAPD as mute LED */
 
 /* ac97->flags */
 #define AC97_HAS_PC_BEEP	(1<<0)	/* force PC Speaker usage */
@@ -491,6 +507,12 @@ struct snd_ac97 {
 	/* jack-sharing info */
 	unsigned char indep_surround;
 	unsigned char channel_mode;
+
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	unsigned int power_up;	/* power states */
+	struct workqueue_struct *power_workq;
+	struct work_struct power_work;
+#endif
 	struct device dev;
 };
 
@@ -532,6 +554,15 @@ unsigned short snd_ac97_read(struct snd_ac97 *ac97, unsigned short reg);
 void snd_ac97_write_cache(struct snd_ac97 *ac97, unsigned short reg, unsigned short value);
 int snd_ac97_update(struct snd_ac97 *ac97, unsigned short reg, unsigned short value);
 int snd_ac97_update_bits(struct snd_ac97 *ac97, unsigned short reg, unsigned short mask, unsigned short value);
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+int snd_ac97_update_power(struct snd_ac97 *ac97, int reg, int powerup);
+#else
+static inline int snd_ac97_update_power(struct snd_ac97 *ac97, int reg,
+					int powerup)
+{
+	return 0;
+}
+#endif
 #ifdef CONFIG_PM
 void snd_ac97_suspend(struct snd_ac97 *ac97);
 void snd_ac97_resume(struct snd_ac97 *ac97);
@@ -583,6 +614,7 @@ struct ac97_pcm {
 		     copy_flag: 1,	   /* lowlevel driver must fill all entries */
 		     spdif: 1;		   /* spdif pcm */
 	unsigned short aslots;		   /* active slots */
+	unsigned short cur_dbl;		   /* current double-rate state */
 	unsigned int rates;		   /* available rates */
 	struct {
 		unsigned short slots;	   /* driver input: requested AC97 slot numbers */
diff --git a/sound/drivers/Kconfig b/sound/drivers/Kconfig
index 395c4ef52ac9..897dc2dfd7dd 100644
--- a/sound/drivers/Kconfig
+++ b/sound/drivers/Kconfig
@@ -100,4 +100,17 @@ config SND_MPU401
 	  To compile this driver as a module, choose M here: the module
 	  will be called snd-mpu401.
 
+config SND_AC97_POWER_SAVE
+	bool "AC97 Power-Saving Mode"
+	depends on SND_AC97_CODEC && EXPERIMENTAL
+	default n
+	help
+	  Say Y here to enable the aggressive power-saving support of
+	  AC97 codecs.  In this mode, the power-mode is dynamically
+	  controlled at each open/close.
+
+	  The mode is activated by passing power_save=1 option to
+	  snd-ac97-codec driver.  You can toggle it dynamically over
+	  sysfs, too.
+
 endmenu
diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c
index b35280ca2465..f82c636e99a9 100644
--- a/sound/pci/ac97/ac97_codec.c
+++ b/sound/pci/ac97/ac97_codec.c
@@ -47,6 +47,11 @@ static int enable_loopback;
 module_param(enable_loopback, bool, 0444);
 MODULE_PARM_DESC(enable_loopback, "Enable AC97 ADC/DAC Loopback Control");
 
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+static int power_save;
+module_param(power_save, bool, 0644);
+MODULE_PARM_DESC(power_save, "Enable AC97 power-saving control");
+#endif
 /*
 
  */
@@ -187,6 +192,8 @@ static const struct ac97_codec_id snd_ac97_codec_ids[] = {
 };
 
 
+static void update_power_regs(struct snd_ac97 *ac97);
+
 /*
  *  I/O routines
  */
@@ -554,6 +561,18 @@ int snd_ac97_put_volsw(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value
 	}
 	err = snd_ac97_update_bits(ac97, reg, val_mask, val);
 	snd_ac97_page_restore(ac97, page_save);
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	/* check analog mixer power-down */
+	if ((val_mask & 0x8000) &&
+	    (kcontrol->private_value & (1<<30))) {
+		if (val & 0x8000)
+			ac97->power_up &= ~(1 << (reg>>1));
+		else
+			ac97->power_up |= 1 << (reg>>1);
+		if (power_save)
+			update_power_regs(ac97);
+	}
+#endif
 	return err;
 }
 
@@ -962,6 +981,10 @@ static int snd_ac97_bus_dev_free(struct snd_device *device)
 static int snd_ac97_free(struct snd_ac97 *ac97)
 {
 	if (ac97) {
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+		if (ac97->power_workq)
+			destroy_workqueue(ac97->power_workq);
+#endif
 		snd_ac97_proc_done(ac97);
 		if (ac97->bus)
 			ac97->bus->codec[ac97->num] = NULL;
@@ -1117,7 +1140,9 @@ struct snd_kcontrol *snd_ac97_cnew(const struct snd_kcontrol_new *_template, str
 /*
  * create mute switch(es) for normal stereo controls
  */
-static int snd_ac97_cmute_new_stereo(struct snd_card *card, char *name, int reg, int check_stereo, struct snd_ac97 *ac97)
+static int snd_ac97_cmute_new_stereo(struct snd_card *card, char *name, int reg,
+				     int check_stereo, int check_amix,
+				     struct snd_ac97 *ac97)
 {
 	struct snd_kcontrol *kctl;
 	int err;
@@ -1137,10 +1162,14 @@ static int snd_ac97_cmute_new_stereo(struct snd_card *card, char *name, int reg,
 	}
 	if (mute_mask == 0x8080) {
 		struct snd_kcontrol_new tmp = AC97_DOUBLE(name, reg, 15, 7, 1, 1);
+		if (check_amix)
+			tmp.private_value |= (1 << 30);
 		tmp.index = ac97->num;
 		kctl = snd_ctl_new1(&tmp, ac97);
 	} else {
 		struct snd_kcontrol_new tmp = AC97_SINGLE(name, reg, 15, 1, 1);
+		if (check_amix)
+			tmp.private_value |= (1 << 30);
 		tmp.index = ac97->num;
 		kctl = snd_ctl_new1(&tmp, ac97);
 	}
@@ -1186,7 +1215,9 @@ static int snd_ac97_cvol_new(struct snd_card *card, char *name, int reg, unsigne
 /*
  * create a mute-switch and a volume for normal stereo/mono controls
  */
-static int snd_ac97_cmix_new_stereo(struct snd_card *card, const char *pfx, int reg, int check_stereo, struct snd_ac97 *ac97)
+static int snd_ac97_cmix_new_stereo(struct snd_card *card, const char *pfx,
+				    int reg, int check_stereo, int check_amix,
+				    struct snd_ac97 *ac97)
 {
 	int err;
 	char name[44];
@@ -1197,7 +1228,9 @@ static int snd_ac97_cmix_new_stereo(struct snd_card *card, const char *pfx, int
 
 	if (snd_ac97_try_bit(ac97, reg, 15)) {
 		sprintf(name, "%s Switch", pfx);
-		if ((err = snd_ac97_cmute_new_stereo(card, name, reg, check_stereo, ac97)) < 0)
+		if ((err = snd_ac97_cmute_new_stereo(card, name, reg,
+						     check_stereo, check_amix,
+						     ac97)) < 0)
 			return err;
 	}
 	check_volume_resolution(ac97, reg, &lo_max, &hi_max);
@@ -1209,8 +1242,10 @@ static int snd_ac97_cmix_new_stereo(struct snd_card *card, const char *pfx, int
 	return 0;
 }
 
-#define snd_ac97_cmix_new(card, pfx, reg, ac97)	snd_ac97_cmix_new_stereo(card, pfx, reg, 0, ac97)
-#define snd_ac97_cmute_new(card, name, reg, ac97)	snd_ac97_cmute_new_stereo(card, name, reg, 0, ac97)
+#define snd_ac97_cmix_new(card, pfx, reg, acheck, ac97) \
+	snd_ac97_cmix_new_stereo(card, pfx, reg, 0, acheck, ac97)
+#define snd_ac97_cmute_new(card, name, reg, acheck, ac97) \
+	snd_ac97_cmute_new_stereo(card, name, reg, 0, acheck, ac97)
 
 static unsigned int snd_ac97_determine_spdif_rates(struct snd_ac97 *ac97);
 
@@ -1226,9 +1261,11 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	/* AD claims to remove this control from AD1887, although spec v2.2 does not allow this */
 	if (snd_ac97_try_volume_mix(ac97, AC97_MASTER)) {
 		if (ac97->flags & AC97_HAS_NO_MASTER_VOL)
-			err = snd_ac97_cmute_new(card, "Master Playback Switch", AC97_MASTER, ac97);
+			err = snd_ac97_cmute_new(card, "Master Playback Switch",
+						 AC97_MASTER, 0, ac97);
 		else
-			err = snd_ac97_cmix_new(card, "Master Playback", AC97_MASTER, ac97);
+			err = snd_ac97_cmix_new(card, "Master Playback",
+						AC97_MASTER, 0, ac97);
 		if (err < 0)
 			return err;
 	}
@@ -1265,19 +1302,23 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	if ((snd_ac97_try_volume_mix(ac97, AC97_SURROUND_MASTER)) 
 		&& !(ac97->flags & AC97_AD_MULTI)) {
 		/* Surround Master (0x38) is with stereo mutes */
-		if ((err = snd_ac97_cmix_new_stereo(card, "Surround Playback", AC97_SURROUND_MASTER, 1, ac97)) < 0)
+		if ((err = snd_ac97_cmix_new_stereo(card, "Surround Playback",
+						    AC97_SURROUND_MASTER, 1, 0,
+						    ac97)) < 0)
 			return err;
 	}
 
 	/* build headphone controls */
 	if (snd_ac97_try_volume_mix(ac97, AC97_HEADPHONE)) {
-		if ((err = snd_ac97_cmix_new(card, "Headphone Playback", AC97_HEADPHONE, ac97)) < 0)
+		if ((err = snd_ac97_cmix_new(card, "Headphone Playback",
+					     AC97_HEADPHONE, 0, ac97)) < 0)
 			return err;
 	}
 	
 	/* build master mono controls */
 	if (snd_ac97_try_volume_mix(ac97, AC97_MASTER_MONO)) {
-		if ((err = snd_ac97_cmix_new(card, "Master Mono Playback", AC97_MASTER_MONO, ac97)) < 0)
+		if ((err = snd_ac97_cmix_new(card, "Master Mono Playback",
+					     AC97_MASTER_MONO, 0, ac97)) < 0)
 			return err;
 	}
 	
@@ -1310,7 +1351,8 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	/* build Phone controls */
 	if (!(ac97->flags & AC97_HAS_NO_PHONE)) {
 		if (snd_ac97_try_volume_mix(ac97, AC97_PHONE)) {
-			if ((err = snd_ac97_cmix_new(card, "Phone Playback", AC97_PHONE, ac97)) < 0)
+			if ((err = snd_ac97_cmix_new(card, "Phone Playback",
+						     AC97_PHONE, 1, ac97)) < 0)
 				return err;
 		}
 	}
@@ -1318,7 +1360,8 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	/* build MIC controls */
 	if (!(ac97->flags & AC97_HAS_NO_MIC)) {
 		if (snd_ac97_try_volume_mix(ac97, AC97_MIC)) {
-			if ((err = snd_ac97_cmix_new(card, "Mic Playback", AC97_MIC, ac97)) < 0)
+			if ((err = snd_ac97_cmix_new(card, "Mic Playback",
+						     AC97_MIC, 1, ac97)) < 0)
 				return err;
 			if ((err = snd_ctl_add(card, snd_ac97_cnew(&snd_ac97_controls_mic_boost, ac97))) < 0)
 				return err;
@@ -1327,14 +1370,16 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 
 	/* build Line controls */
 	if (snd_ac97_try_volume_mix(ac97, AC97_LINE)) {
-		if ((err = snd_ac97_cmix_new(card, "Line Playback", AC97_LINE, ac97)) < 0)
+		if ((err = snd_ac97_cmix_new(card, "Line Playback",
+					     AC97_LINE, 1, ac97)) < 0)
 			return err;
 	}
 	
 	/* build CD controls */
 	if (!(ac97->flags & AC97_HAS_NO_CD)) {
 		if (snd_ac97_try_volume_mix(ac97, AC97_CD)) {
-			if ((err = snd_ac97_cmix_new(card, "CD Playback", AC97_CD, ac97)) < 0)
+			if ((err = snd_ac97_cmix_new(card, "CD Playback",
+						     AC97_CD, 1, ac97)) < 0)
 				return err;
 		}
 	}
@@ -1342,7 +1387,8 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	/* build Video controls */
 	if (!(ac97->flags & AC97_HAS_NO_VIDEO)) {
 		if (snd_ac97_try_volume_mix(ac97, AC97_VIDEO)) {
-			if ((err = snd_ac97_cmix_new(card, "Video Playback", AC97_VIDEO, ac97)) < 0)
+			if ((err = snd_ac97_cmix_new(card, "Video Playback",
+						     AC97_VIDEO, 1, ac97)) < 0)
 				return err;
 		}
 	}
@@ -1350,7 +1396,8 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	/* build Aux controls */
 	if (!(ac97->flags & AC97_HAS_NO_AUX)) {
 		if (snd_ac97_try_volume_mix(ac97, AC97_AUX)) {
-			if ((err = snd_ac97_cmix_new(card, "Aux Playback", AC97_AUX, ac97)) < 0)
+			if ((err = snd_ac97_cmix_new(card, "Aux Playback",
+						     AC97_AUX, 1, ac97)) < 0)
 				return err;
 		}
 	}
@@ -1385,9 +1432,12 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 	} else {
 		if (!(ac97->flags & AC97_HAS_NO_STD_PCM)) {
 			if (ac97->flags & AC97_HAS_NO_PCM_VOL)
-				err = snd_ac97_cmute_new(card, "PCM Playback Switch", AC97_PCM, ac97);
+				err = snd_ac97_cmute_new(card,
+							 "PCM Playback Switch",
+							 AC97_PCM, 0, ac97);
 			else
-				err = snd_ac97_cmix_new(card, "PCM Playback", AC97_PCM, ac97);
+				err = snd_ac97_cmix_new(card, "PCM Playback",
+							AC97_PCM, 0, ac97);
 			if (err < 0)
 				return err;
 		}
@@ -1398,7 +1448,9 @@ static int snd_ac97_mixer_build(struct snd_ac97 * ac97)
 		if ((err = snd_ctl_add(card, snd_ac97_cnew(&snd_ac97_control_capture_src, ac97))) < 0)
 			return err;
 		if (snd_ac97_try_bit(ac97, AC97_REC_GAIN, 15)) {
-			if ((err = snd_ac97_cmute_new(card, "Capture Switch", AC97_REC_GAIN, ac97)) < 0)
+			err = snd_ac97_cmute_new(card, "Capture Switch",
+						 AC97_REC_GAIN, 0, ac97);
+			if (err < 0)
 				return err;
 		}
 		if ((err = snd_ctl_add(card, snd_ac97_cnew(&snd_ac97_control_capture_vol, ac97))) < 0)
@@ -1829,6 +1881,13 @@ static int snd_ac97_dev_disconnect(struct snd_device *device)
 /* build_ops to do nothing */
 static struct snd_ac97_build_ops null_build_ops;
 
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+static void do_update_power(void *data)
+{
+	update_power_regs(data);
+}
+#endif
+
 /**
  * snd_ac97_mixer - create an Codec97 component
  * @bus: the AC97 bus which codec is attached to
@@ -1883,6 +1942,10 @@ int snd_ac97_mixer(struct snd_ac97_bus *bus, struct snd_ac97_template *template,
 	bus->codec[ac97->num] = ac97;
 	mutex_init(&ac97->reg_mutex);
 	mutex_init(&ac97->page_mutex);
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	ac97->power_workq = create_workqueue("ac97");
+	INIT_WORK(&ac97->power_work, do_update_power, ac97);
+#endif
 
 #ifdef CONFIG_PCI
 	if (ac97->pci) {
@@ -2117,15 +2180,8 @@ int snd_ac97_mixer(struct snd_ac97_bus *bus, struct snd_ac97_template *template,
 			return -ENOMEM;
 		}
 	}
-	/* make sure the proper powerdown bits are cleared */
-	if (ac97->scaps && ac97_is_audio(ac97)) {
-		reg = snd_ac97_read(ac97, AC97_EXTENDED_STATUS);
-		if (ac97->scaps & AC97_SCAP_SURROUND_DAC) 
-			reg &= ~AC97_EA_PRJ;
-		if (ac97->scaps & AC97_SCAP_CENTER_LFE_DAC) 
-			reg &= ~(AC97_EA_PRI | AC97_EA_PRK);
-		snd_ac97_write_cache(ac97, AC97_EXTENDED_STATUS, reg);
-	}
+	if (ac97_is_audio(ac97))
+		update_power_regs(ac97);
 	snd_ac97_proc_init(ac97);
 	if ((err = snd_device_new(card, SNDRV_DEV_CODEC, ac97, &ops)) < 0) {
 		snd_ac97_free(ac97);
@@ -2153,22 +2209,155 @@ static void snd_ac97_powerdown(struct snd_ac97 *ac97)
 		snd_ac97_write(ac97, AC97_HEADPHONE, 0x9f9f);
 	}
 
-	power = ac97->regs[AC97_POWERDOWN] | 0x8000;	/* EAPD */
-	power |= 0x4000;	/* Headphone amplifier powerdown */
-	power |= 0x0300;	/* ADC & DAC powerdown */
+	/* surround, CLFE, mic powerdown */
+	power = ac97->regs[AC97_EXTENDED_STATUS];
+	if (ac97->scaps & AC97_SCAP_SURROUND_DAC)
+		power |= AC97_EA_PRJ;
+	if (ac97->scaps & AC97_SCAP_CENTER_LFE_DAC)
+		power |= AC97_EA_PRI | AC97_EA_PRK;
+	power |= AC97_EA_PRL;
+	snd_ac97_write(ac97, AC97_EXTENDED_STATUS, power);
+
+	/* powerdown external amplifier */
+	if (ac97->scaps & AC97_SCAP_INV_EAPD)
+		power = ac97->regs[AC97_POWERDOWN] & ~AC97_PD_EAPD;
+	else if (! (ac97->scaps & AC97_SCAP_EAPD_LED))
+		power = ac97->regs[AC97_POWERDOWN] | AC97_PD_EAPD;
+	power |= AC97_PD_PR6;	/* Headphone amplifier powerdown */
+	power |= AC97_PD_PR0 | AC97_PD_PR1;	/* ADC & DAC powerdown */
 	snd_ac97_write(ac97, AC97_POWERDOWN, power);
 	udelay(100);
-	power |= 0x0400;	/* Analog Mixer powerdown (Vref on) */
-	snd_ac97_write(ac97, AC97_POWERDOWN, power);
-	udelay(100);
-#if 0
-	/* FIXME: this causes click noises on some boards at resume */
-	power |= 0x3800;	/* AC-link powerdown, internal Clk disable */
+	power |= AC97_PD_PR2 | AC97_PD_PR3;	/* Analog Mixer powerdown */
 	snd_ac97_write(ac97, AC97_POWERDOWN, power);
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	if (power_save) {
+		udelay(100);
+		/* AC-link powerdown, internal Clk disable */
+		/* FIXME: this may cause click noises on some boards */
+		power |= AC97_PD_PR4 | AC97_PD_PR5;
+		snd_ac97_write(ac97, AC97_POWERDOWN, power);
+	}
 #endif
 }
 
 
+struct ac97_power_reg {
+	unsigned short reg;
+	unsigned short power_reg;
+	unsigned short mask;
+};
+
+enum { PWIDX_ADC, PWIDX_FRONT, PWIDX_CLFE, PWIDX_SURR, PWIDX_MIC, PWIDX_SIZE };
+
+static struct ac97_power_reg power_regs[PWIDX_SIZE] = {
+	[PWIDX_ADC] = { AC97_PCM_LR_ADC_RATE, AC97_POWERDOWN, AC97_PD_PR0},
+	[PWIDX_FRONT] = { AC97_PCM_FRONT_DAC_RATE, AC97_POWERDOWN, AC97_PD_PR1},
+	[PWIDX_CLFE] = { AC97_PCM_LFE_DAC_RATE, AC97_EXTENDED_STATUS,
+			 AC97_EA_PRI | AC97_EA_PRK},
+	[PWIDX_SURR] = { AC97_PCM_SURR_DAC_RATE, AC97_EXTENDED_STATUS,
+			 AC97_EA_PRJ},
+	[PWIDX_MIC] = { AC97_PCM_MIC_ADC_RATE, AC97_EXTENDED_STATUS,
+			AC97_EA_PRL},
+};
+
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+/**
+ * snd_ac97_update_power - update the powerdown register
+ * @ac97: the codec instance
+ * @reg: the rate register, e.g. AC97_PCM_FRONT_DAC_RATE
+ * @powerup: non-zero when power up the part
+ *
+ * Update the AC97 powerdown register bits of the given part.
+ */
+int snd_ac97_update_power(struct snd_ac97 *ac97, int reg, int powerup)
+{
+	int i;
+
+	if (! ac97)
+		return 0;
+
+	if (reg) {
+		/* SPDIF requires DAC power, too */
+		if (reg == AC97_SPDIF)
+			reg = AC97_PCM_FRONT_DAC_RATE;
+		for (i = 0; i < PWIDX_SIZE; i++) {
+			if (power_regs[i].reg == reg) {
+				if (powerup)
+					ac97->power_up |= (1 << i);
+				else
+					ac97->power_up &= ~(1 << i);
+				break;
+			}
+		}
+	}
+
+	if (! power_save)
+		return 0;
+
+	if (! powerup && ac97->power_workq)
+		/* adjust power-down bits after two seconds delay
+		 * (for avoiding loud click noises for many (OSS) apps
+		 *  that open/close frequently)
+		 */
+		queue_delayed_work(ac97->power_workq, &ac97->power_work, HZ*2);
+	else
+		update_power_regs(ac97);
+
+	return 0;
+}
+
+EXPORT_SYMBOL(snd_ac97_update_power);
+#endif /* CONFIG_SND_AC97_POWER_SAVE */
+
+static void update_power_regs(struct snd_ac97 *ac97)
+{
+	unsigned int power_up, bits;
+	int i;
+
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	if (power_save)
+		power_up = ac97->power_up;
+	else {
+#endif
+		power_up = (1 << PWIDX_FRONT) | (1 << PWIDX_ADC);
+		power_up |= (1 << PWIDX_MIC);
+		if (ac97->scaps & AC97_SCAP_SURROUND_DAC)
+			power_up |= (1 << PWIDX_SURR);
+		if (ac97->scaps & AC97_SCAP_CENTER_LFE_DAC)
+			power_up |= (1 << PWIDX_CLFE);
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	}
+#endif
+	if (power_up) {
+		if (ac97->regs[AC97_POWERDOWN] & AC97_PD_PR2) {
+			/* needs power-up analog mix and vref */
+			snd_ac97_update_bits(ac97, AC97_POWERDOWN,
+					     AC97_PD_PR3, 0);
+			msleep(1);
+			snd_ac97_update_bits(ac97, AC97_POWERDOWN,
+					     AC97_PD_PR2, 0);
+		}
+	}
+	for (i = 0; i < PWIDX_SIZE; i++) {
+		if (power_up & (1 << i))
+			bits = 0;
+		else
+			bits = power_regs[i].mask;
+		snd_ac97_update_bits(ac97, power_regs[i].power_reg,
+				     power_regs[i].mask, bits);
+	}
+	if (! power_up) {
+		if (! (ac97->regs[AC97_POWERDOWN] & AC97_PD_PR2)) {
+			/* power down analog mix and vref */
+			snd_ac97_update_bits(ac97, AC97_POWERDOWN,
+					     AC97_PD_PR2, AC97_PD_PR2);
+			snd_ac97_update_bits(ac97, AC97_POWERDOWN,
+					     AC97_PD_PR3, AC97_PD_PR3);
+		}
+	}
+}
+
+
 #ifdef CONFIG_PM
 /**
  * snd_ac97_suspend - General suspend function for AC97 codec
@@ -2484,6 +2673,7 @@ static int tune_mute_led(struct snd_ac97 *ac97)
 	msw->put = master_mute_sw_put;
 	snd_ac97_remove_ctl(ac97, "External Amplifier", NULL);
 	snd_ac97_update_bits(ac97, AC97_POWERDOWN, 0x8000, 0x8000); /* mute LED on */
+	ac97->scaps |= AC97_SCAP_EAPD_LED;
 	return 0;
 }
 
diff --git a/sound/pci/ac97/ac97_pcm.c b/sound/pci/ac97/ac97_pcm.c
index f684aa2c0067..3758d07182f8 100644
--- a/sound/pci/ac97/ac97_pcm.c
+++ b/sound/pci/ac97/ac97_pcm.c
@@ -269,6 +269,7 @@ int snd_ac97_set_rate(struct snd_ac97 *ac97, int reg, unsigned int rate)
 			return -EINVAL;
 	}
 
+	snd_ac97_update_power(ac97, reg, 1);
 	switch (reg) {
 	case AC97_PCM_MIC_ADC_RATE:
 		if ((ac97->regs[AC97_EXTENDED_STATUS] & AC97_EA_VRM) == 0)	/* MIC VRA */
@@ -606,6 +607,7 @@ int snd_ac97_pcm_open(struct ac97_pcm *pcm, unsigned int rate,
 			goto error;
 		}
 	}
+	pcm->cur_dbl = r;
 	spin_unlock_irq(&pcm->bus->bus_lock);
 	for (i = 3; i < 12; i++) {
 		if (!(slots & (1 << i)))
@@ -651,6 +653,21 @@ int snd_ac97_pcm_close(struct ac97_pcm *pcm)
 	unsigned short slots = pcm->aslots;
 	int i, cidx;
 
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	int r = pcm->cur_dbl;
+	for (i = 3; i < 12; i++) {
+		if (!(slots & (1 << i)))
+			continue;
+		for (cidx = 0; cidx < 4; cidx++) {
+			if (pcm->r[r].rslots[cidx] & (1 << i)) {
+				int reg = get_slot_reg(pcm, cidx, i, r);
+				snd_ac97_update_power(pcm->r[r].codec[cidx],
+						      reg, 0);
+			}
+		}
+	}
+#endif
+
 	bus = pcm->bus;
 	spin_lock_irq(&pcm->bus->bus_lock);
 	for (i = 3; i < 12; i++) {
@@ -660,6 +677,7 @@ int snd_ac97_pcm_close(struct ac97_pcm *pcm)
 			bus->used_slots[pcm->stream][cidx] &= ~(1 << i);
 	}
 	pcm->aslots = 0;
+	pcm->cur_dbl = 0;
 	spin_unlock_irq(&pcm->bus->bus_lock);
 	return 0;
 }
diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c
index 6874263f1681..72dbaedcbdf5 100644
--- a/sound/pci/intel8x0.c
+++ b/sound/pci/intel8x0.c
@@ -2251,6 +2251,16 @@ static int snd_intel8x0_ich_chip_init(struct intel8x0 *chip, int probing)
 	/* ACLink on, 2 channels */
 	cnt = igetdword(chip, ICHREG(GLOB_CNT));
 	cnt &= ~(ICH_ACLINK | ICH_PCM_246_MASK);
+#ifdef CONFIG_SND_AC97_POWER_SAVE
+	/* do cold reset - the full ac97 powerdown may leave the controller
+	 * in a warm state but actually it cannot communicate with the codec.
+	 */
+	iputdword(chip, ICHREG(GLOB_CNT), cnt & ~ICH_AC97COLD);
+	cnt = igetdword(chip, ICHREG(GLOB_CNT));
+	udelay(10);
+	iputdword(chip, ICHREG(GLOB_CNT), cnt | ICH_AC97COLD);
+	msleep(1);
+#else
 	/* finish cold or do warm reset */
 	cnt |= (cnt & ICH_AC97COLD) == 0 ? ICH_AC97COLD : ICH_AC97WARM;
 	iputdword(chip, ICHREG(GLOB_CNT), cnt);
@@ -2265,6 +2275,7 @@ static int snd_intel8x0_ich_chip_init(struct intel8x0 *chip, int probing)
 	return -EIO;
 
       __ok:
+#endif
 	if (probing) {
 		/* wait for any codec ready status.
 		 * Once it becomes ready it should remain ready
@@ -2485,7 +2496,7 @@ static int intel8x0_resume(struct pci_dev *pci)
 		    card->shortname, chip);
 	chip->irq = pci->irq;
 	synchronize_irq(chip->irq);
-	snd_intel8x0_chip_init(chip, 1);
+	snd_intel8x0_chip_init(chip, 0);
 
 	/* re-initialize mixer stuff */
 	if (chip->device_type == DEVICE_INTEL_ICH4) {
@@ -2615,6 +2626,7 @@ static void __devinit intel8x0_measure_ac97_clock(struct intel8x0 *chip)
 		/* not 48000Hz, tuning the clock.. */
 		chip->ac97_bus->clock = (chip->ac97_bus->clock * 48000) / pos;
 	printk(KERN_INFO "intel8x0: clocking to %d\n", chip->ac97_bus->clock);
+	snd_ac97_update_power(chip->ac97[0], AC97_PCM_FRONT_DAC_RATE, 0);
 }
 
 #ifdef CONFIG_PROC_FS
diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c
index 08da9234efb3..2c23a665c3e3 100644
--- a/sound/pci/via82xx.c
+++ b/sound/pci/via82xx.c
@@ -1277,7 +1277,18 @@ static int snd_via82xx_pcm_close(struct snd_pcm_substream *substream)
 	if (! ratep->used)
 		ratep->rate = 0;
 	spin_unlock_irq(&ratep->lock);
-
+	if (! ratep->rate) {
+		if (! viadev->direction) {
+			snd_ac97_update_power(chip->ac97,
+					      AC97_PCM_FRONT_DAC_RATE, 0);
+			snd_ac97_update_power(chip->ac97,
+					      AC97_PCM_SURR_DAC_RATE, 0);
+			snd_ac97_update_power(chip->ac97,
+					      AC97_PCM_LFE_DAC_RATE, 0);
+		} else
+			snd_ac97_update_power(chip->ac97,
+					      AC97_PCM_LR_ADC_RATE, 0);
+	}
 	viadev->substream = NULL;
 	return 0;
 }
-- 
cgit v1.2.3


From 8aa9b586e42099817163aba01d925c2660c4dbbe Mon Sep 17 00:00:00 2001
From: Jaroslav Kysela <perex@suse.cz>
Date: Wed, 5 Jul 2006 17:34:51 +0200
Subject: [ALSA] Control API - more robust TLV implementation

- added callback option
- added READ/WRITE/COMMAND flags to access member
- added WRITE/COMMAND ioctls
- added SNDRV_CTL_EVENT_MASK_TLV for TLV change notifications
- added TLV support to ELEM_ADD ioctl

Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/asound.h  |  10 +++-
 include/sound/control.h |  16 +++++-
 sound/core/control.c    | 139 +++++++++++++++++++++++++++++++++++++-----------
 3 files changed, 132 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/sound/asound.h b/include/sound/asound.h
index 76a20406bd18..c1621c650a9a 100644
--- a/include/sound/asound.h
+++ b/include/sound/asound.h
@@ -727,10 +727,15 @@ typedef int __bitwise snd_ctl_elem_iface_t;
 #define SNDRV_CTL_ELEM_ACCESS_WRITE		(1<<1)
 #define SNDRV_CTL_ELEM_ACCESS_READWRITE		(SNDRV_CTL_ELEM_ACCESS_READ|SNDRV_CTL_ELEM_ACCESS_WRITE)
 #define SNDRV_CTL_ELEM_ACCESS_VOLATILE		(1<<2)	/* control value may be changed without a notification */
-#define SNDRV_CTL_ELEM_ACCESS_TIMESTAMP		(1<<2)	/* when was control changed */
+#define SNDRV_CTL_ELEM_ACCESS_TIMESTAMP		(1<<3)	/* when was control changed */
+#define SNDRV_CTL_ELEM_ACCESS_TLV_READ		(1<<4)	/* TLV read is possible */
+#define SNDRV_CTL_ELEM_ACCESS_TLV_WRITE		(1<<5)	/* TLV write is possible */
+#define SNDRV_CTL_ELEM_ACCESS_TLV_READWRITE	(SNDRV_CTL_ELEM_ACCESS_TLV_READ|SNDRV_CTL_ELEM_ACCESS_TLV_WRITE)
+#define SNDRV_CTL_ELEM_ACCESS_TLV_COMMAND	(1<<6)	/* TLV command is possible */
 #define SNDRV_CTL_ELEM_ACCESS_INACTIVE		(1<<8)	/* control does actually nothing, but may be updated */
 #define SNDRV_CTL_ELEM_ACCESS_LOCK		(1<<9)	/* write lock */
 #define SNDRV_CTL_ELEM_ACCESS_OWNER		(1<<10)	/* write lock owner */
+#define SNDRV_CTL_ELEM_ACCESS_TLV_CALLBACK	(1<<28)	/* kernel use a TLV callback */ 
 #define SNDRV_CTL_ELEM_ACCESS_USER		(1<<29) /* user space element */
 #define SNDRV_CTL_ELEM_ACCESS_DINDIRECT		(1<<30)	/* indirect access for matrix dimensions in the info structure */
 #define SNDRV_CTL_ELEM_ACCESS_INDIRECT		(1<<31)	/* indirect access for element value in the value structure */
@@ -838,6 +843,8 @@ enum {
 	SNDRV_CTL_IOCTL_ELEM_REPLACE = _IOWR('U', 0x18, struct snd_ctl_elem_info),
 	SNDRV_CTL_IOCTL_ELEM_REMOVE = _IOWR('U', 0x19, struct snd_ctl_elem_id),
 	SNDRV_CTL_IOCTL_TLV_READ = _IOWR('U', 0x1a, struct snd_ctl_tlv),
+	SNDRV_CTL_IOCTL_TLV_WRITE = _IOWR('U', 0x1b, struct snd_ctl_tlv),
+	SNDRV_CTL_IOCTL_TLV_COMMAND = _IOWR('U', 0x1c, struct snd_ctl_tlv),
 	SNDRV_CTL_IOCTL_HWDEP_NEXT_DEVICE = _IOWR('U', 0x20, int),
 	SNDRV_CTL_IOCTL_HWDEP_INFO = _IOR('U', 0x21, struct snd_hwdep_info),
 	SNDRV_CTL_IOCTL_PCM_NEXT_DEVICE = _IOR('U', 0x30, int),
@@ -862,6 +869,7 @@ enum sndrv_ctl_event_type {
 #define SNDRV_CTL_EVENT_MASK_VALUE	(1<<0)	/* element value was changed */
 #define SNDRV_CTL_EVENT_MASK_INFO	(1<<1)	/* element info was changed */
 #define SNDRV_CTL_EVENT_MASK_ADD	(1<<2)	/* element was added */
+#define SNDRV_CTL_EVENT_MASK_TLV	(1<<3)	/* element TLV tree was changed */
 #define SNDRV_CTL_EVENT_MASK_REMOVE	(~0U)	/* element was removed */
 
 struct snd_ctl_event {
diff --git a/include/sound/control.h b/include/sound/control.h
index a93a58d0e688..e3905c5a0950 100644
--- a/include/sound/control.h
+++ b/include/sound/control.h
@@ -30,6 +30,11 @@ struct snd_kcontrol;
 typedef int (snd_kcontrol_info_t) (struct snd_kcontrol * kcontrol, struct snd_ctl_elem_info * uinfo);
 typedef int (snd_kcontrol_get_t) (struct snd_kcontrol * kcontrol, struct snd_ctl_elem_value * ucontrol);
 typedef int (snd_kcontrol_put_t) (struct snd_kcontrol * kcontrol, struct snd_ctl_elem_value * ucontrol);
+typedef int (snd_kcontrol_tlv_rw_t)(struct snd_kcontrol *kcontrol,
+				    int op_flag, /* 0=read,1=write,-1=command */
+				    unsigned int size,
+				    unsigned int __user *tlv);
+
 
 struct snd_kcontrol_new {
 	snd_ctl_elem_iface_t iface;	/* interface identifier */
@@ -42,7 +47,10 @@ struct snd_kcontrol_new {
 	snd_kcontrol_info_t *info;
 	snd_kcontrol_get_t *get;
 	snd_kcontrol_put_t *put;
-	unsigned int *tlv;
+	union {
+		snd_kcontrol_tlv_rw_t *c;
+		unsigned int *p;
+	} tlv;
 	unsigned long private_value;
 };
 
@@ -59,7 +67,11 @@ struct snd_kcontrol {
 	snd_kcontrol_info_t *info;
 	snd_kcontrol_get_t *get;
 	snd_kcontrol_put_t *put;
-	unsigned int *tlv;
+	snd_kcontrol_tlv_rw_t *tlv_rw;
+	union {
+		snd_kcontrol_tlv_rw_t *c;
+		unsigned int *p;
+	} tlv;
 	unsigned long private_value;
 	void *private_data;
 	void (*private_free)(struct snd_kcontrol *kcontrol);
diff --git a/sound/core/control.c b/sound/core/control.c
index f0c7272a2d48..31ad58154c06 100644
--- a/sound/core/control.c
+++ b/sound/core/control.c
@@ -236,12 +236,16 @@ struct snd_kcontrol *snd_ctl_new1(const struct snd_kcontrol_new *ncontrol,
 	kctl.id.index = ncontrol->index;
 	kctl.count = ncontrol->count ? ncontrol->count : 1;
 	access = ncontrol->access == 0 ? SNDRV_CTL_ELEM_ACCESS_READWRITE :
-		 (ncontrol->access & (SNDRV_CTL_ELEM_ACCESS_READWRITE|SNDRV_CTL_ELEM_ACCESS_INACTIVE|
-		 		      SNDRV_CTL_ELEM_ACCESS_DINDIRECT|SNDRV_CTL_ELEM_ACCESS_INDIRECT));
+		 (ncontrol->access & (SNDRV_CTL_ELEM_ACCESS_READWRITE|
+				      SNDRV_CTL_ELEM_ACCESS_INACTIVE|
+		 		      SNDRV_CTL_ELEM_ACCESS_DINDIRECT|
+		 		      SNDRV_CTL_ELEM_ACCESS_INDIRECT|
+		 		      SNDRV_CTL_ELEM_ACCESS_TLV_READWRITE|
+		 		      SNDRV_CTL_ELEM_ACCESS_TLV_CALLBACK));
 	kctl.info = ncontrol->info;
 	kctl.get = ncontrol->get;
 	kctl.put = ncontrol->put;
-	kctl.tlv = ncontrol->tlv;
+	kctl.tlv.p = ncontrol->tlv.p;
 	kctl.private_value = ncontrol->private_value;
 	kctl.private_data = private_data;
 	return snd_ctl_new(&kctl, access);
@@ -883,6 +887,8 @@ struct user_element {
 	struct snd_ctl_elem_info info;
 	void *elem_data;		/* element data */
 	unsigned long elem_data_size;	/* size of element data in bytes */
+	void *tlv_data;			/* TLV data */
+	unsigned long tlv_data_size;	/* TLV data size */
 	void *priv_data;		/* private data (like strings for enumerated type) */
 	unsigned long priv_data_size;	/* size of private data in bytes */
 };
@@ -917,9 +923,46 @@ static int snd_ctl_elem_user_put(struct snd_kcontrol *kcontrol,
 	return change;
 }
 
+static int snd_ctl_elem_user_tlv(struct snd_kcontrol *kcontrol,
+				 int op_flag,
+				 unsigned int size,
+				 unsigned int __user *tlv)
+{
+	struct user_element *ue = kcontrol->private_data;
+	int change = 0;
+	void *new_data;
+
+	if (op_flag > 0) {
+		if (size > 1024 * 128)	/* sane value */
+			return -EINVAL;
+		new_data = kmalloc(size, GFP_KERNEL);
+		if (new_data == NULL)
+			return -ENOMEM;
+		if (copy_from_user(new_data, tlv, size)) {
+			kfree(new_data);
+			return -EFAULT;
+		}
+		change = ue->tlv_data_size != size;
+		if (!change)
+			change = memcmp(ue->tlv_data, new_data, size);
+		kfree(ue->tlv_data);
+		ue->tlv_data = new_data;
+		ue->tlv_data_size = size;
+	} else {
+		if (size < ue->tlv_data_size)
+			return -ENOSPC;
+		if (copy_to_user(tlv, ue->tlv_data, ue->tlv_data_size))
+			return -EFAULT;
+	}
+	return change;
+}
+
 static void snd_ctl_elem_user_free(struct snd_kcontrol *kcontrol)
 {
-	kfree(kcontrol->private_data);
+	struct user_element *ue = kcontrol->private_data;
+	if (ue->tlv_data)
+		kfree(ue->tlv_data);
+	kfree(ue);
 }
 
 static int snd_ctl_elem_add(struct snd_ctl_file *file,
@@ -938,7 +981,8 @@ static int snd_ctl_elem_add(struct snd_ctl_file *file,
 		return -EINVAL;
 	access = info->access == 0 ? SNDRV_CTL_ELEM_ACCESS_READWRITE :
 		(info->access & (SNDRV_CTL_ELEM_ACCESS_READWRITE|
-				 SNDRV_CTL_ELEM_ACCESS_INACTIVE));
+				 SNDRV_CTL_ELEM_ACCESS_INACTIVE|
+				 SNDRV_CTL_ELEM_ACCESS_TLV_READWRITE));
 	info->id.numid = 0;
 	memset(&kctl, 0, sizeof(kctl));
 	down_write(&card->controls_rwsem);
@@ -964,6 +1008,10 @@ static int snd_ctl_elem_add(struct snd_ctl_file *file,
 		kctl.get = snd_ctl_elem_user_get;
 	if (access & SNDRV_CTL_ELEM_ACCESS_WRITE)
 		kctl.put = snd_ctl_elem_user_put;
+	if (access & SNDRV_CTL_ELEM_ACCESS_TLV_READWRITE) {
+		kctl.tlv.c = snd_ctl_elem_user_tlv;
+		access |= SNDRV_CTL_ELEM_ACCESS_TLV_CALLBACK;
+	}
 	switch (info->type) {
 	case SNDRV_CTL_ELEM_TYPE_BOOLEAN:
 		private_size = sizeof(char);
@@ -1068,38 +1116,65 @@ static int snd_ctl_subscribe_events(struct snd_ctl_file *file, int __user *ptr)
 	return 0;
 }
 
-static int snd_ctl_tlv_read(struct snd_card *card,
-                            struct snd_ctl_tlv __user *_tlv)
+static int snd_ctl_tlv_ioctl(struct snd_ctl_file *file,
+                             struct snd_ctl_tlv __user *_tlv,
+                             int op_flag)
 {
+	struct snd_card *card = file->card;
 	struct snd_ctl_tlv tlv;
 	struct snd_kcontrol *kctl;
+	struct snd_kcontrol_volatile *vd;
 	unsigned int len;
 	int err = 0;
 
 	if (copy_from_user(&tlv, _tlv, sizeof(tlv)))
 		return -EFAULT;
-        if (tlv.length < sizeof(unsigned int) * 3)
-                return -EINVAL;
-        down_read(&card->controls_rwsem);
-        kctl = snd_ctl_find_numid(card, tlv.numid);
-        if (kctl == NULL) {
-                err = -ENOENT;
-                goto __kctl_end;
-        }
-        if (kctl->tlv == NULL) {
-                err = -ENXIO;
-                goto __kctl_end;
-        }
-        len = kctl->tlv[1] + 2 * sizeof(unsigned int);
-        if (tlv.length < len) {
-                err = -ENOMEM;
-                goto __kctl_end;
-        }
-        if (copy_to_user(_tlv->tlv, kctl->tlv, len))
-        	err = -EFAULT;
+	if (tlv.length < sizeof(unsigned int) * 3)
+		return -EINVAL;
+	down_read(&card->controls_rwsem);
+	kctl = snd_ctl_find_numid(card, tlv.numid);
+	if (kctl == NULL) {
+		err = -ENOENT;
+		goto __kctl_end;
+	}
+	if (kctl->tlv.p == NULL) {
+		err = -ENXIO;
+		goto __kctl_end;
+	}
+	vd = &kctl->vd[tlv.numid - kctl->id.numid];
+	if ((op_flag == 0 && (vd->access & SNDRV_CTL_ELEM_ACCESS_TLV_READ) == 0) ||
+	    (op_flag > 0 && (vd->access & SNDRV_CTL_ELEM_ACCESS_TLV_WRITE) == 0) ||
+	    (op_flag < 0 && (vd->access & SNDRV_CTL_ELEM_ACCESS_TLV_COMMAND) == 0)) {
+	    	err = -ENXIO;
+	    	goto __kctl_end;
+	}
+	if (vd->access & SNDRV_CTL_ELEM_ACCESS_TLV_CALLBACK) {
+		if (file && vd->owner != NULL && vd->owner != file) {
+			err = -EPERM;
+			goto __kctl_end;
+		}
+		err = kctl->tlv.c(kctl, op_flag, tlv.length, _tlv->tlv); 
+		if (err > 0) {
+			up_read(&card->controls_rwsem);
+			snd_ctl_notify(card, SNDRV_CTL_EVENT_MASK_TLV, &kctl->id);
+			return 0;
+		}
+	} else {
+		if (op_flag) {
+			err = -ENXIO;
+			goto __kctl_end;
+		}
+		len = kctl->tlv.p[1] + 2 * sizeof(unsigned int);
+		if (tlv.length < len) {
+			err = -ENOMEM;
+			goto __kctl_end;
+		}
+		if (copy_to_user(_tlv->tlv, kctl->tlv.p, len))
+			err = -EFAULT;
+	}
       __kctl_end:
-        up_read(&card->controls_rwsem);
-        return err;
+	up_read(&card->controls_rwsem);
+	return err;
 }
 
 static long snd_ctl_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
@@ -1140,8 +1215,12 @@ static long snd_ctl_ioctl(struct file *file, unsigned int cmd, unsigned long arg
 		return snd_ctl_elem_remove(ctl, argp);
 	case SNDRV_CTL_IOCTL_SUBSCRIBE_EVENTS:
 		return snd_ctl_subscribe_events(ctl, ip);
-        case SNDRV_CTL_IOCTL_TLV_READ:
-                return snd_ctl_tlv_read(card, argp);
+	case SNDRV_CTL_IOCTL_TLV_READ:
+		return snd_ctl_tlv_ioctl(ctl, argp, 0);
+	case SNDRV_CTL_IOCTL_TLV_WRITE:
+		return snd_ctl_tlv_ioctl(ctl, argp, 1);
+	case SNDRV_CTL_IOCTL_TLV_COMMAND:
+		return snd_ctl_tlv_ioctl(ctl, argp, -1);
 	case SNDRV_CTL_IOCTL_POWER:
 		return -ENOPROTOOPT;
 	case SNDRV_CTL_IOCTL_POWER_STATE:
-- 
cgit v1.2.3


From 6a65d793b0a82c7e190d9fd92a479401b6a127ca Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 14 Jul 2006 14:39:34 +0200
Subject: [ALSA] Remove unused tlv_rw field from struct snd_kcontrol

Remove unused tlv_rw field from struct snd_kcontrol.  The callback is
set in tlv.c field, instead.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/control.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/control.h b/include/sound/control.h
index e3905c5a0950..1de148b0fd94 100644
--- a/include/sound/control.h
+++ b/include/sound/control.h
@@ -67,7 +67,6 @@ struct snd_kcontrol {
 	snd_kcontrol_info_t *info;
 	snd_kcontrol_get_t *get;
 	snd_kcontrol_put_t *put;
-	snd_kcontrol_tlv_rw_t *tlv_rw;
 	union {
 		snd_kcontrol_tlv_rw_t *c;
 		unsigned int *p;
-- 
cgit v1.2.3


From 31508f83f591dc8764427b6321c89f8f9e84bad2 Mon Sep 17 00:00:00 2001
From: James Courtier-Dutton <James@superbug.co.uk>
Date: Sat, 22 Jul 2006 17:02:10 +0100
Subject: [ALSA] snd-emu10k1: Implement dB gain infomation.

Signed-off-by: James Courtier-Dutton <James@superbug.co.uk>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/emu10k1.h   |  4 ++++
 sound/pci/emu10k1/emufx.c | 11 +++++++++++
 sound/pci/emu10k1/p16v.c  |  5 +++++
 3 files changed, 20 insertions(+)

(limited to 'include')

diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h
index 884bbf54cd36..892e310c504d 100644
--- a/include/sound/emu10k1.h
+++ b/include/sound/emu10k1.h
@@ -1524,6 +1524,10 @@ struct snd_emu10k1_fx8010_control_gpr {
 	unsigned int value[32];		/* initial values */
 	unsigned int min;		/* minimum range */
 	unsigned int max;		/* maximum range */
+	union {
+		snd_kcontrol_tlv_rw_t *c;
+		unsigned int *p;
+	} tlv;
 	unsigned int translation;	/* translation type (EMU10K1_GPR_TRANSLATION*) */
 };
 
diff --git a/sound/pci/emu10k1/emufx.c b/sound/pci/emu10k1/emufx.c
index dfba00230d4d..00fc904c251d 100644
--- a/sound/pci/emu10k1/emufx.c
+++ b/sound/pci/emu10k1/emufx.c
@@ -35,6 +35,7 @@
 #include <linux/mutex.h>
 
 #include <sound/core.h>
+#include <sound/tlv.h>
 #include <sound/emu10k1.h>
 
 #if 0		/* for testing purposes - digital out -> capture */
@@ -290,6 +291,9 @@ static const u32 db_table[101] = {
 	0x7fffffff,
 };
 
+/* EMU10k1/EMU10k2 DSP control db gain */
+static DECLARE_TLV_DB_SCALE(snd_emu10k1_db_scale1, -4000, 40, 1);
+
 static const u32 onoff_table[2] = {
 	0x00000000, 0x00000001
 };
@@ -755,6 +759,11 @@ static int snd_emu10k1_add_controls(struct snd_emu10k1 *emu,
 		knew.device = gctl->id.device;
 		knew.subdevice = gctl->id.subdevice;
 		knew.info = snd_emu10k1_gpr_ctl_info;
+		if (gctl->tlv.p) {
+			knew.tlv.p = gctl->tlv.p;
+			knew.access = SNDRV_CTL_ELEM_ACCESS_READWRITE |
+				SNDRV_CTL_ELEM_ACCESS_TLV_READ;
+		} 
 		knew.get = snd_emu10k1_gpr_ctl_get;
 		knew.put = snd_emu10k1_gpr_ctl_put;
 		memset(nctl, 0, sizeof(*nctl));
@@ -1013,6 +1022,7 @@ snd_emu10k1_init_mono_control(struct snd_emu10k1_fx8010_control_gpr *ctl,
 	ctl->gpr[0] = gpr + 0; ctl->value[0] = defval;
 	ctl->min = 0;
 	ctl->max = 100;
+	ctl->tlv.p = snd_emu10k1_db_scale1;
 	ctl->translation = EMU10K1_GPR_TRANSLATION_TABLE100;	
 }
 
@@ -1027,6 +1037,7 @@ snd_emu10k1_init_stereo_control(struct snd_emu10k1_fx8010_control_gpr *ctl,
 	ctl->gpr[1] = gpr + 1; ctl->value[1] = defval;
 	ctl->min = 0;
 	ctl->max = 100;
+	ctl->tlv.p = snd_emu10k1_db_scale1;
 	ctl->translation = EMU10K1_GPR_TRANSLATION_TABLE100;
 }
 
diff --git a/sound/pci/emu10k1/p16v.c b/sound/pci/emu10k1/p16v.c
index 9905651935fb..1e44714b8623 100644
--- a/sound/pci/emu10k1/p16v.c
+++ b/sound/pci/emu10k1/p16v.c
@@ -100,6 +100,7 @@
 #include <sound/pcm.h>
 #include <sound/ac97_codec.h>
 #include <sound/info.h>
+#include <sound/tlv.h>
 #include <sound/emu10k1.h>
 #include "p16v.h"
 
@@ -784,12 +785,16 @@ static int snd_p16v_capture_channel_put(struct snd_kcontrol *kcontrol,
 	}
         return change;
 }
+static DECLARE_TLV_DB_SCALE(snd_p16v_db_scale1, -5175, 25, 1);
 
 #define P16V_VOL(xname,xreg,xhl) { \
 	.iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \
+        .access = SNDRV_CTL_ELEM_ACCESS_READWRITE |             \
+                  SNDRV_CTL_ELEM_ACCESS_TLV_READ,               \
 	.info = snd_p16v_volume_info, \
 	.get = snd_p16v_volume_get, \
 	.put = snd_p16v_volume_put, \
+	.tlv.p = snd_p16v_db_scale1, \
 	.private_value = ((xreg) | ((xhl) << 8)) \
 }
 
-- 
cgit v1.2.3


From 548a648b98318e4b843b636dd2c7f42377e19a00 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 31 Jul 2006 16:51:51 +0200
Subject: [ALSA] Fix control/status mmap with shared PCM substream

The flag to avoid 32bit-incompatible mmap for control/status records
should be outside the pcm substream instance since a substream can be
shared among multiple opens.  Now it's flagged in pcm_file list that
is directly assigned to file->private_data.
Also, removed snd_pcm_add_file() and remove_file() functions and
substream.files field that are not really used in the code.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/pcm.h     |  4 +---
 sound/core/pcm_compat.c |  2 +-
 sound/core/pcm_native.c | 49 ++++++++++---------------------------------------
 3 files changed, 12 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/include/sound/pcm.h b/include/sound/pcm.h
index f84d84993a31..60d40b34efc0 100644
--- a/include/sound/pcm.h
+++ b/include/sound/pcm.h
@@ -190,7 +190,7 @@ struct snd_pcm_ops {
 
 struct snd_pcm_file {
 	struct snd_pcm_substream *substream;
-	struct snd_pcm_file *next;
+	int no_compat_mmap;
 };
 
 struct snd_pcm_hw_rule;
@@ -384,7 +384,6 @@ struct snd_pcm_substream {
 	struct snd_info_entry *proc_prealloc_entry;
 #endif
 	/* misc flags */
-	unsigned int no_mmap_ctrl: 1;
 	unsigned int hw_opened: 1;
 };
 
@@ -402,7 +401,6 @@ struct snd_pcm_str {
 	/* -- OSS things -- */
 	struct snd_pcm_oss_stream oss;
 #endif
-	struct snd_pcm_file *files;
 #ifdef CONFIG_SND_VERBOSE_PROCFS
 	struct snd_info_entry *proc_root;
 	struct snd_info_entry *proc_info_entry;
diff --git a/sound/core/pcm_compat.c b/sound/core/pcm_compat.c
index 2b8aab6fd6cd..2b539799d23b 100644
--- a/sound/core/pcm_compat.c
+++ b/sound/core/pcm_compat.c
@@ -478,7 +478,7 @@ static long snd_pcm_ioctl_compat(struct file *file, unsigned int cmd, unsigned l
 	 * mmap of PCM status/control records because of the size
 	 * incompatibility.
 	 */
-	substream->no_mmap_ctrl = 1;
+	pcm_file->no_compat_mmap = 1;
 
 	switch (cmd) {
 	case SNDRV_PCM_IOCTL_PVERSION:
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index 439f047929e1..0224c70414f5 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -1992,35 +1992,9 @@ int snd_pcm_hw_constraints_complete(struct snd_pcm_substream *substream)
 	return 0;
 }
 
-static void snd_pcm_add_file(struct snd_pcm_str *str,
-			     struct snd_pcm_file *pcm_file)
-{
-	pcm_file->next = str->files;
-	str->files = pcm_file;
-}
-
-static void snd_pcm_remove_file(struct snd_pcm_str *str,
-				struct snd_pcm_file *pcm_file)
-{
-	struct snd_pcm_file * pcm_file1;
-	if (str->files == pcm_file) {
-		str->files = pcm_file->next;
-	} else {
-		pcm_file1 = str->files;
-		while (pcm_file1 && pcm_file1->next != pcm_file)
-			pcm_file1 = pcm_file1->next;
-		if (pcm_file1 != NULL)
-			pcm_file1->next = pcm_file->next;
-	}
-}
-
 static void pcm_release_private(struct snd_pcm_substream *substream)
 {
-	struct snd_pcm_file *pcm_file = substream->file;
-
 	snd_pcm_unlink(substream);
-	snd_pcm_remove_file(substream->pstr, pcm_file);
-	kfree(pcm_file);
 }
 
 void snd_pcm_release_substream(struct snd_pcm_substream *substream)
@@ -2060,7 +2034,6 @@ int snd_pcm_open_substream(struct snd_pcm *pcm, int stream,
 		return 0;
 	}
 
-	substream->no_mmap_ctrl = 0;
 	err = snd_pcm_hw_constraints_init(substream);
 	if (err < 0) {
 		snd_printd("snd_pcm_hw_constraints_init failed\n");
@@ -2105,19 +2078,16 @@ static int snd_pcm_open_file(struct file *file,
 	if (err < 0)
 		return err;
 
-	if (substream->ref_count > 1)
-		pcm_file = substream->file;
-	else {
-		pcm_file = kzalloc(sizeof(*pcm_file), GFP_KERNEL);
-		if (pcm_file == NULL) {
-			snd_pcm_release_substream(substream);
-			return -ENOMEM;
-		}
+	pcm_file = kzalloc(sizeof(*pcm_file), GFP_KERNEL);
+	if (pcm_file == NULL) {
+		snd_pcm_release_substream(substream);
+		return -ENOMEM;
+	}
+	pcm_file->substream = substream;
+	if (substream->ref_count == 1) {
 		str = substream->pstr;
 		substream->file = pcm_file;
 		substream->pcm_release = pcm_release_private;
-		pcm_file->substream = substream;
-		snd_pcm_add_file(str, pcm_file);
 	}
 	file->private_data = pcm_file;
 	*rpcm_file = pcm_file;
@@ -2209,6 +2179,7 @@ static int snd_pcm_release(struct inode *inode, struct file *file)
 	fasync_helper(-1, file, 0, &substream->runtime->fasync);
 	mutex_lock(&pcm->open_mutex);
 	snd_pcm_release_substream(substream);
+	kfree(pcm_file);
 	mutex_unlock(&pcm->open_mutex);
 	wake_up(&pcm->open_wait);
 	module_put(pcm->card->module);
@@ -3270,11 +3241,11 @@ static int snd_pcm_mmap(struct file *file, struct vm_area_struct *area)
 	offset = area->vm_pgoff << PAGE_SHIFT;
 	switch (offset) {
 	case SNDRV_PCM_MMAP_OFFSET_STATUS:
-		if (substream->no_mmap_ctrl)
+		if (pcm_file->no_compat_mmap)
 			return -ENXIO;
 		return snd_pcm_mmap_status(substream, file, area);
 	case SNDRV_PCM_MMAP_OFFSET_CONTROL:
-		if (substream->no_mmap_ctrl)
+		if (pcm_file->no_compat_mmap)
 			return -ENXIO;
 		return snd_pcm_mmap_control(substream, file, area);
 	default:
-- 
cgit v1.2.3


From 683fe1537e660c322c8af953773921e814791193 Mon Sep 17 00:00:00 2001
From: Jochen Voss <voss@seehuhn.de>
Date: Tue, 8 Aug 2006 21:12:44 +0200
Subject: [ALSA] Revolution 5.1 - add AK5365 ADC support

Add support for the AK5365 ADC.

Signed-off-by: Jochen Voss <voss@seehuhn.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/ak4xxx-adda.h   |  3 ++-
 sound/i2c/other/ak4xxx-adda.c | 25 +++++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/ak4xxx-adda.h b/include/sound/ak4xxx-adda.h
index 3d9888492026..65ddfa3cac1f 100644
--- a/include/sound/ak4xxx-adda.h
+++ b/include/sound/ak4xxx-adda.h
@@ -53,7 +53,8 @@ struct snd_akm4xxx {
 	unsigned int idx_offset;		/* control index offset */
 	enum {
 		SND_AK4524, SND_AK4528, SND_AK4529,
-		SND_AK4355, SND_AK4358, SND_AK4381
+		SND_AK4355, SND_AK4358, SND_AK4381,
+		SND_AK5365
 	} type;
 	unsigned int *num_stereo;	/* array of combined counts
 					 * for the mixer
diff --git a/sound/i2c/other/ak4xxx-adda.c b/sound/i2c/other/ak4xxx-adda.c
index dc7cc2001b74..7d562f084207 100644
--- a/sound/i2c/other/ak4xxx-adda.c
+++ b/sound/i2c/other/ak4xxx-adda.c
@@ -598,6 +598,31 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak)
 		if (err < 0)
 			goto __error;
 	}
+
+	if (ak->type == SND_AK5365) {
+		memset(ctl, 0, sizeof(*ctl));
+		if (ak->channel_names == NULL)
+			strcpy(ctl->id.name, "Capture Volume");
+		else
+			strcpy(ctl->id.name, ak->channel_names[0]);
+		ctl->id.index = ak->idx_offset * 2;
+		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
+		ctl->count = 1;
+		ctl->info = snd_akm4xxx_stereo_volume_info;
+		ctl->get = snd_akm4xxx_stereo_volume_get;
+		ctl->put = snd_akm4xxx_stereo_volume_put;
+		/* Registers 4 & 5 (see AK5365 data sheet, pages 34 and 35):
+		 * valid values are from 0x00 (mute) to 0x98 (+12dB).  */
+		ctl->private_value =
+			AK_COMPOSE(0, 4, 0, 0x98);
+		ctl->private_data = ak;
+		err = snd_ctl_add(ak->card,
+				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
+					      SNDRV_CTL_ELEM_ACCESS_WRITE));
+		if (err < 0)
+			goto __error;
+	}
+
 	if (ak->type == SND_AK4355 || ak->type == SND_AK4358)
 		num_emphs = 1;
 	else
-- 
cgit v1.2.3


From eac06a10d2b814dfacc36a8fff35ef07bf4eec8e Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Tue, 22 Aug 2006 13:16:25 +0200
Subject: [ALSA] Add dB scale information to ad1848 driver

Added the dB scale information to ad1848 driver.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/ad1848.h        | 22 +++++++++++++------
 sound/isa/ad1848/ad1848_lib.c | 49 ++++++++++++++++++++++++++++++-------------
 2 files changed, 50 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/sound/ad1848.h b/include/sound/ad1848.h
index 57af1fe7b309..c8de6f83338f 100644
--- a/include/sound/ad1848.h
+++ b/include/sound/ad1848.h
@@ -179,14 +179,13 @@ enum { AD1848_MIX_SINGLE, AD1848_MIX_DOUBLE, AD1848_MIX_CAPTURE };
 #define AD1848_MIXVAL_DOUBLE(left_reg, right_reg, shift_left, shift_right, mask, invert) \
 	((left_reg) | ((right_reg) << 8) | ((shift_left) << 16) | ((shift_right) << 19) | ((mask) << 24) | ((invert) << 22))
 
-int snd_ad1848_add_ctl(struct snd_ad1848 *chip, const char *name, int index, int type, unsigned long value);
-
 /* for ease of use */
 struct ad1848_mix_elem {
 	const char *name;
 	int index;
 	int type;
 	unsigned long private_value;
+	unsigned int *tlv;
 };
 
 #define AD1848_SINGLE(xname, xindex, reg, shift, mask, invert) \
@@ -195,15 +194,26 @@ struct ad1848_mix_elem {
   .type = AD1848_MIX_SINGLE, \
   .private_value = AD1848_MIXVAL_SINGLE(reg, shift, mask, invert) }
 
+#define AD1848_SINGLE_TLV(xname, xindex, reg, shift, mask, invert, xtlv) \
+{ .name = xname, \
+  .index = xindex, \
+  .type = AD1848_MIX_SINGLE, \
+  .private_value = AD1848_MIXVAL_SINGLE(reg, shift, mask, invert), \
+  .tlv = xtlv }
+
 #define AD1848_DOUBLE(xname, xindex, left_reg, right_reg, shift_left, shift_right, mask, invert) \
 { .name = xname, \
   .index = xindex, \
   .type = AD1848_MIX_DOUBLE, \
   .private_value = AD1848_MIXVAL_DOUBLE(left_reg, right_reg, shift_left, shift_right, mask, invert) }
 
-static inline int snd_ad1848_add_ctl_elem(struct snd_ad1848 *chip, const struct ad1848_mix_elem *c)
-{
-	return snd_ad1848_add_ctl(chip, c->name, c->index, c->type, c->private_value);
-}
+#define AD1848_DOUBLE_TLV(xname, xindex, left_reg, right_reg, shift_left, shift_right, mask, invert, xtlv) \
+{ .name = xname, \
+  .index = xindex, \
+  .type = AD1848_MIX_DOUBLE, \
+  .private_value = AD1848_MIXVAL_DOUBLE(left_reg, right_reg, shift_left, shift_right, mask, invert), \
+  .tlv = xtlv }
+
+int snd_ad1848_add_ctl_elem(struct snd_ad1848 *chip, const struct ad1848_mix_elem *c);
 
 #endif /* __SOUND_AD1848_H */
diff --git a/sound/isa/ad1848/ad1848_lib.c b/sound/isa/ad1848/ad1848_lib.c
index e711f87d5fd1..a6fbd5d1d62f 100644
--- a/sound/isa/ad1848/ad1848_lib.c
+++ b/sound/isa/ad1848/ad1848_lib.c
@@ -29,6 +29,7 @@
 #include <sound/core.h>
 #include <sound/ad1848.h>
 #include <sound/control.h>
+#include <sound/tlv.h>
 #include <sound/pcm_params.h>
 
 #include <asm/io.h>
@@ -118,6 +119,8 @@ void snd_ad1848_out(struct snd_ad1848 *chip,
 #endif
 }
 
+EXPORT_SYMBOL(snd_ad1848_out);
+
 static void snd_ad1848_dout(struct snd_ad1848 *chip,
 			    unsigned char reg, unsigned char value)
 {
@@ -941,6 +944,8 @@ int snd_ad1848_create(struct snd_card *card,
 	return 0;
 }
 
+EXPORT_SYMBOL(snd_ad1848_create);
+
 static struct snd_pcm_ops snd_ad1848_playback_ops = {
 	.open =		snd_ad1848_playback_open,
 	.close =	snd_ad1848_playback_close,
@@ -988,12 +993,16 @@ int snd_ad1848_pcm(struct snd_ad1848 *chip, int device, struct snd_pcm **rpcm)
 	return 0;
 }
 
+EXPORT_SYMBOL(snd_ad1848_pcm);
+
 const struct snd_pcm_ops *snd_ad1848_get_pcm_ops(int direction)
 {
 	return direction == SNDRV_PCM_STREAM_PLAYBACK ?
 		&snd_ad1848_playback_ops : &snd_ad1848_capture_ops;
 }
 
+EXPORT_SYMBOL(snd_ad1848_get_pcm_ops);
+
 /*
  *  MIXER part
  */
@@ -1171,7 +1180,8 @@ static int snd_ad1848_put_double(struct snd_kcontrol *kcontrol, struct snd_ctl_e
 
 /*
  */
-int snd_ad1848_add_ctl(struct snd_ad1848 *chip, const char *name, int index, int type, unsigned long value)
+int snd_ad1848_add_ctl_elem(struct snd_ad1848 *chip,
+			    const struct ad1848_mix_elem *c)
 {
 	static struct snd_kcontrol_new newctls[] = {
 		[AD1848_MIX_SINGLE] = {
@@ -1196,32 +1206,46 @@ int snd_ad1848_add_ctl(struct snd_ad1848 *chip, const char *name, int index, int
 	struct snd_kcontrol *ctl;
 	int err;
 
-	ctl = snd_ctl_new1(&newctls[type], chip);
+	ctl = snd_ctl_new1(&newctls[c->type], chip);
 	if (! ctl)
 		return -ENOMEM;
-	strlcpy(ctl->id.name, name, sizeof(ctl->id.name));
-	ctl->id.index = index;
-	ctl->private_value = value;
+	strlcpy(ctl->id.name, c->name, sizeof(ctl->id.name));
+	ctl->id.index = c->index;
+	ctl->private_value = c->private_value;
+	if (c->tlv) {
+		ctl->vd[0].access |= SNDRV_CTL_ELEM_ACCESS_TLV_READ;
+		ctl->tlv.p = c->tlv;
+	}
 	if ((err = snd_ctl_add(chip->card, ctl)) < 0)
 		return err;
 	return 0;
 }
 
+EXPORT_SYMBOL(snd_ad1848_add_ctl_elem);
+
+static DECLARE_TLV_DB_SCALE(db_scale_6bit, -9450, 150, 0);
+static DECLARE_TLV_DB_SCALE(db_scale_5bit_12db_max, -3450, 150, 0);
+static DECLARE_TLV_DB_SCALE(db_scale_rec_gain, 0, 150, 0);
 
 static struct ad1848_mix_elem snd_ad1848_controls[] = {
 AD1848_DOUBLE("PCM Playback Switch", 0, AD1848_LEFT_OUTPUT, AD1848_RIGHT_OUTPUT, 7, 7, 1, 1),
-AD1848_DOUBLE("PCM Playback Volume", 0, AD1848_LEFT_OUTPUT, AD1848_RIGHT_OUTPUT, 0, 0, 63, 1),
+AD1848_DOUBLE_TLV("PCM Playback Volume", 0, AD1848_LEFT_OUTPUT, AD1848_RIGHT_OUTPUT, 0, 0, 63, 1,
+		  db_scale_6bit),
 AD1848_DOUBLE("Aux Playback Switch", 0, AD1848_AUX1_LEFT_INPUT, AD1848_AUX1_RIGHT_INPUT, 7, 7, 1, 1),
-AD1848_DOUBLE("Aux Playback Volume", 0, AD1848_AUX1_LEFT_INPUT, AD1848_AUX1_RIGHT_INPUT, 0, 0, 31, 1),
+AD1848_DOUBLE_TLV("Aux Playback Volume", 0, AD1848_AUX1_LEFT_INPUT, AD1848_AUX1_RIGHT_INPUT, 0, 0, 31, 1,
+		  db_scale_5bit_12db_max),
 AD1848_DOUBLE("Aux Playback Switch", 1, AD1848_AUX2_LEFT_INPUT, AD1848_AUX2_RIGHT_INPUT, 7, 7, 1, 1),
-AD1848_DOUBLE("Aux Playback Volume", 1, AD1848_AUX2_LEFT_INPUT, AD1848_AUX2_RIGHT_INPUT, 0, 0, 31, 1),
-AD1848_DOUBLE("Capture Volume", 0, AD1848_LEFT_INPUT, AD1848_RIGHT_INPUT, 0, 0, 15, 0),
+AD1848_DOUBLE_TLV("Aux Playback Volume", 1, AD1848_AUX2_LEFT_INPUT, AD1848_AUX2_RIGHT_INPUT, 0, 0, 31, 1,
+		  db_scale_5bit_12db_max),
+AD1848_DOUBLE_TLV("Capture Volume", 0, AD1848_LEFT_INPUT, AD1848_RIGHT_INPUT, 0, 0, 15, 0,
+		  db_scale_rec_gain),
 {
 	.name = "Capture Source",
 	.type = AD1848_MIX_CAPTURE,
 },
 AD1848_SINGLE("Loopback Capture Switch", 0, AD1848_LOOPBACK, 0, 1, 0),
-AD1848_SINGLE("Loopback Capture Volume", 0, AD1848_LOOPBACK, 1, 63, 0)
+AD1848_SINGLE_TLV("Loopback Capture Volume", 0, AD1848_LOOPBACK, 1, 63, 0,
+		  db_scale_6bit),
 };
                                         
 int snd_ad1848_mixer(struct snd_ad1848 *chip)
@@ -1245,12 +1269,7 @@ int snd_ad1848_mixer(struct snd_ad1848 *chip)
 	return 0;
 }
 
-EXPORT_SYMBOL(snd_ad1848_out);
-EXPORT_SYMBOL(snd_ad1848_create);
-EXPORT_SYMBOL(snd_ad1848_pcm);
-EXPORT_SYMBOL(snd_ad1848_get_pcm_ops);
 EXPORT_SYMBOL(snd_ad1848_mixer);
-EXPORT_SYMBOL(snd_ad1848_add_ctl);
 
 /*
  *  INIT part
-- 
cgit v1.2.3


From 1186ed8c7dc9c0185e783beddf241509cc224f1a Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 23 Aug 2006 19:53:28 +0200
Subject: [ALSA] Add dB scale information to vxpocket and vx222 drivers

Added the dB scale information to vxpocket and vx222 drivers.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/vx_core.h     |  1 +
 sound/drivers/vx/vx_mixer.c | 17 +++++++++++++++--
 sound/pci/vx222/vx222.c     |  7 +++++++
 sound/pci/vx222/vx222_ops.c |  9 +++++++++
 sound/pcmcia/vx/vxpocket.c  |  5 +++++
 5 files changed, 37 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/sound/vx_core.h b/include/sound/vx_core.h
index 9821a6194caa..dbca14170615 100644
--- a/include/sound/vx_core.h
+++ b/include/sound/vx_core.h
@@ -128,6 +128,7 @@ struct snd_vx_hardware {
 	unsigned int num_ins;
 	unsigned int num_outs;
 	unsigned int output_level_max;
+	unsigned int *output_level_db_scale;
 };
 
 /* hwdep id string */
diff --git a/sound/drivers/vx/vx_mixer.c b/sound/drivers/vx/vx_mixer.c
index c1d7fcdd1973..1613ed844ac6 100644
--- a/sound/drivers/vx/vx_mixer.c
+++ b/sound/drivers/vx/vx_mixer.c
@@ -23,6 +23,7 @@
 #include <sound/driver.h>
 #include <sound/core.h>
 #include <sound/control.h>
+#include <sound/tlv.h>
 #include <sound/vx_core.h>
 #include "vx_cmd.h"
 
@@ -455,10 +456,13 @@ static int vx_output_level_put(struct snd_kcontrol *kcontrol, struct snd_ctl_ele
 
 static struct snd_kcontrol_new vx_control_output_level = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
+	.access =	(SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			 SNDRV_CTL_ELEM_ACCESS_TLV_READ),
 	.name =		"Master Playback Volume",
 	.info =		vx_output_level_info,
 	.get =		vx_output_level_get,
 	.put =		vx_output_level_put,
+	/* tlv will be filled later */
 };
 
 /*
@@ -712,12 +716,17 @@ static int vx_monitor_sw_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_
 	return 0;
 }
 
+static DECLARE_TLV_DB_SCALE(db_scale_audio_gain, -10975, 25, 0);
+
 static struct snd_kcontrol_new vx_control_audio_gain = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
+	.access =	(SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			 SNDRV_CTL_ELEM_ACCESS_TLV_READ),
 	/* name will be filled later */
 	.info =         vx_audio_gain_info,
 	.get =          vx_audio_gain_get,
-	.put =          vx_audio_gain_put
+	.put =          vx_audio_gain_put,
+	.tlv = { .p = db_scale_audio_gain },
 };
 static struct snd_kcontrol_new vx_control_output_switch = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
@@ -729,9 +738,12 @@ static struct snd_kcontrol_new vx_control_output_switch = {
 static struct snd_kcontrol_new vx_control_monitor_gain = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
 	.name =         "Monitoring Volume",
+	.access =	(SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			 SNDRV_CTL_ELEM_ACCESS_TLV_READ),
 	.info =         vx_audio_gain_info,	/* shared */
 	.get =          vx_audio_monitor_get,
-	.put =          vx_audio_monitor_put
+	.put =          vx_audio_monitor_put,
+	.tlv = { .p = db_scale_audio_gain },
 };
 static struct snd_kcontrol_new vx_control_monitor_switch = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
@@ -918,6 +930,7 @@ int snd_vx_mixer_new(struct vx_core *chip)
 	for (i = 0; i < chip->hw->num_outs; i++) {
 		temp = vx_control_output_level;
 		temp.index = i;
+		temp.tlv.p = chip->hw->output_level_db_scale;
 		if ((err = snd_ctl_add(card, snd_ctl_new1(&temp, chip))) < 0)
 			return err;
 	}
diff --git a/sound/pci/vx222/vx222.c b/sound/pci/vx222/vx222.c
index 9c03c6b4e490..e7cd8acab59a 100644
--- a/sound/pci/vx222/vx222.c
+++ b/sound/pci/vx222/vx222.c
@@ -26,6 +26,7 @@
 #include <linux/moduleparam.h>
 #include <sound/core.h>
 #include <sound/initval.h>
+#include <sound/tlv.h>
 #include "vx222.h"
 
 #define CARD_NAME "VX222"
@@ -72,6 +73,9 @@ MODULE_DEVICE_TABLE(pci, snd_vx222_ids);
 /*
  */
 
+static DECLARE_TLV_DB_SCALE(db_scale_old_vol, -11350, 50, 0);
+static DECLARE_TLV_DB_SCALE(db_scale_akm, -7350, 50, 0);
+
 static struct snd_vx_hardware vx222_old_hw = {
 
 	.name = "VX222/Old",
@@ -81,6 +85,7 @@ static struct snd_vx_hardware vx222_old_hw = {
 	.num_ins = 1,
 	.num_outs = 1,
 	.output_level_max = VX_ANALOG_OUT_LEVEL_MAX,
+	.output_level_db_scale = db_scale_old_vol,
 };
 
 static struct snd_vx_hardware vx222_v2_hw = {
@@ -92,6 +97,7 @@ static struct snd_vx_hardware vx222_v2_hw = {
 	.num_ins = 1,
 	.num_outs = 1,
 	.output_level_max = VX2_AKM_LEVEL_MAX,
+	.output_level_db_scale = db_scale_akm,
 };
 
 static struct snd_vx_hardware vx222_mic_hw = {
@@ -103,6 +109,7 @@ static struct snd_vx_hardware vx222_mic_hw = {
 	.num_ins = 1,
 	.num_outs = 1,
 	.output_level_max = VX2_AKM_LEVEL_MAX,
+	.output_level_db_scale = db_scale_akm,
 };
 
 
diff --git a/sound/pci/vx222/vx222_ops.c b/sound/pci/vx222/vx222_ops.c
index 9b6d345b83a6..5e51950e05f9 100644
--- a/sound/pci/vx222/vx222_ops.c
+++ b/sound/pci/vx222/vx222_ops.c
@@ -28,6 +28,7 @@
 
 #include <sound/core.h>
 #include <sound/control.h>
+#include <sound/tlv.h>
 #include <asm/io.h>
 #include "vx222.h"
 
@@ -845,6 +846,8 @@ static void vx2_set_input_level(struct snd_vx222 *chip)
 
 #define MIC_LEVEL_MAX	0xff
 
+static DECLARE_TLV_DB_SCALE(db_scale_mic, -6450, 50, 0);
+
 /*
  * controls API for input levels
  */
@@ -922,18 +925,24 @@ static int vx_mic_level_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_v
 
 static struct snd_kcontrol_new vx_control_input_level = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
+	.access =	(SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			 SNDRV_CTL_ELEM_ACCESS_TLV_READ),
 	.name =		"Capture Volume",
 	.info =		vx_input_level_info,
 	.get =		vx_input_level_get,
 	.put =		vx_input_level_put,
+	.tlv = { .p = db_scale_mic },
 };
 
 static struct snd_kcontrol_new vx_control_mic_level = {
 	.iface =	SNDRV_CTL_ELEM_IFACE_MIXER,
+	.access =	(SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			 SNDRV_CTL_ELEM_ACCESS_TLV_READ),
 	.name =		"Mic Capture Volume",
 	.info =		vx_mic_level_info,
 	.get =		vx_mic_level_get,
 	.put =		vx_mic_level_put,
+	.tlv = { .p = db_scale_mic },
 };
 
 /*
diff --git a/sound/pcmcia/vx/vxpocket.c b/sound/pcmcia/vx/vxpocket.c
index 76c85cffb40e..3089fcca800e 100644
--- a/sound/pcmcia/vx/vxpocket.c
+++ b/sound/pcmcia/vx/vxpocket.c
@@ -27,6 +27,7 @@
 #include <pcmcia/ciscode.h>
 #include <pcmcia/cisreg.h>
 #include <sound/initval.h>
+#include <sound/tlv.h>
 
 /*
  */
@@ -90,6 +91,8 @@ static int snd_vxpocket_dev_free(struct snd_device *device)
  * Only output levels can be modified
  */
 
+static DECLARE_TLV_DB_SCALE(db_scale_old_vol, -11350, 50, 0);
+
 static struct snd_vx_hardware vxpocket_hw = {
 	.name = "VXPocket",
 	.type = VX_TYPE_VXPOCKET,
@@ -99,6 +102,7 @@ static struct snd_vx_hardware vxpocket_hw = {
 	.num_ins = 1,
 	.num_outs = 1,
 	.output_level_max = VX_ANALOG_OUT_LEVEL_MAX,
+	.output_level_db_scale = db_scale_old_vol,
 };	
 
 /* VX-pocket 440
@@ -120,6 +124,7 @@ static struct snd_vx_hardware vxp440_hw = {
 	.num_ins = 2,
 	.num_outs = 2,
 	.output_level_max = VX_ANALOG_OUT_LEVEL_MAX,
+	.output_level_db_scale = db_scale_old_vol,
 };	
 
 
-- 
cgit v1.2.3


From 063a40d9111ce7558f2fdfa4f85acfc47eb27353 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 28 Aug 2006 13:20:13 +0200
Subject: [ALSA] Add the definition of linear volume TLV

Added the definition of linear volume TLV type.
Some DSP chips and codecs (e.g. AK codec) use linear volume control.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/tlv.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'include')

diff --git a/include/sound/tlv.h b/include/sound/tlv.h
index b826e1df1da6..7905841643df 100644
--- a/include/sound/tlv.h
+++ b/include/sound/tlv.h
@@ -33,6 +33,7 @@
 
 #define SNDRV_CTL_TLVT_CONTAINER 0	/* one level down - group of TLVs */
 #define SNDRV_CTL_TLVT_DB_SCALE	1       /* dB scale */
+#define SNDRV_CTL_TLVT_DB_LINEAR 2	/* linear volume */
 
 #define DECLARE_TLV_DB_SCALE(name, min, step, mute) \
 unsigned int name[] = { \
@@ -40,4 +41,13 @@ unsigned int name[] = { \
         (min), ((step) & 0xffff) | ((mute) ? 0x10000 : 0) \
 }
 
+/* linear volume between min_dB and max_dB (.01dB unit) */
+#define DECLARE_TLV_DB_LINEAR(name, min_dB, max_dB)	\
+unsigned int name[] = { \
+        SNDRV_CTL_TLVT_DB_LINEAR, 2 * sizeof(unsigned int), \
+        (min_dB), (max_dB)				\
+}
+
+#define TLV_DB_GAIN_MUTE	-9999999
+
 #endif /* __SOUND_TLV_H */
-- 
cgit v1.2.3


From 723b2b0d36fa7cea81a962af2d40d88520d5a5f1 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 30 Aug 2006 16:49:54 +0200
Subject: [ALSA] Clean up and add TLV support to AK4xxx i2c driver

- Clean up the code in AK4xxx-ADDA i2c code.
- Fix capture gain controls for AK5365
- Changed the static table for DAC/ADC mixer labels to use
  structs
- Implemented TLV entries for each AK codec
  The volumes in AK4524, AK4528 and AK5365 are corrected with
  a table to be suitable for dB conversion.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/ak4xxx-adda.h   |  40 +++-
 sound/i2c/other/ak4xxx-adda.c | 523 ++++++++++++++++++++++++------------------
 sound/pci/ice1712/revo.c      |  47 ++--
 3 files changed, 354 insertions(+), 256 deletions(-)

(limited to 'include')

diff --git a/include/sound/ak4xxx-adda.h b/include/sound/ak4xxx-adda.h
index 65ddfa3cac1f..026e4072a9a1 100644
--- a/include/sound/ak4xxx-adda.h
+++ b/include/sound/ak4xxx-adda.h
@@ -39,14 +39,26 @@ struct snd_ak4xxx_ops {
 
 #define AK4XXX_IMAGE_SIZE	(AK4XXX_MAX_CHIPS * 16)	/* 64 bytes */
 
+/* DAC label and channels */
+struct snd_akm4xxx_dac_channel {
+	char *name;		/* mixer volume name */
+	unsigned int num_channels;
+};
+
+/* ADC labels and channels */
+struct snd_akm4xxx_adc_channel {
+	char *name;		/* capture gain volume label */
+	char *gain_name;	/* IPGA */
+	char *switch_name;	/* capture switch */
+	unsigned int num_channels;
+};
+
 struct snd_akm4xxx {
 	struct snd_card *card;
 	unsigned int num_adcs;			/* AK4524 or AK4528 ADCs */
 	unsigned int num_dacs;			/* AK4524 or AK4528 DACs */
 	unsigned char images[AK4XXX_IMAGE_SIZE]; /* saved register image */
-	unsigned char ipga_gain[AK4XXX_MAX_CHIPS][2]; /* saved register image
-						       * for IPGA (AK4528)
-						       */
+	unsigned char volumes[AK4XXX_IMAGE_SIZE]; /* saved volume values */
 	unsigned long private_value[AK4XXX_MAX_CHIPS];	/* helper for driver */
 	void *private_data[AK4XXX_MAX_CHIPS];		/* helper for driver */
 	/* template should fill the following fields */
@@ -56,10 +68,11 @@ struct snd_akm4xxx {
 		SND_AK4355, SND_AK4358, SND_AK4381,
 		SND_AK5365
 	} type;
-	unsigned int *num_stereo;	/* array of combined counts
-					 * for the mixer
-					 */
-	char **channel_names;		/* array of mixer channel names */
+
+	/* (array) information of combined codecs */
+	struct snd_akm4xxx_dac_channel *dac_info;
+	struct snd_akm4xxx_adc_channel *adc_info;
+
 	struct snd_ak4xxx_ops ops;
 };
 
@@ -73,9 +86,18 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak);
 	(ak)->images[(chip) * 16 + (reg)]
 #define snd_akm4xxx_set(ak,chip,reg,val) \
 	((ak)->images[(chip) * 16 + (reg)] = (val))
+#define snd_akm4xxx_get_vol(ak,chip,reg) \
+	(ak)->volumes[(chip) * 16 + (reg)]
+#define snd_akm4xxx_set_vol(ak,chip,reg,val) \
+	((ak)->volumes[(chip) * 16 + (reg)] = (val))
+
+/* Warning: IPGA is tricky - we assume the addr + 4 is unused
+ *   so far, it's OK for all AK codecs with IPGA:
+ *   AK4524, AK4528 and EK5365
+ */
 #define snd_akm4xxx_get_ipga(ak,chip,reg) \
-	(ak)->ipga_gain[chip][(reg)-4]
+	snd_akm4xxx_get_vol(ak, chip, (reg) + 4)
 #define snd_akm4xxx_set_ipga(ak,chip,reg,val) \
-	((ak)->ipga_gain[chip][(reg)-4] = (val))
+	snd_akm4xxx_set_vol(ak, chip, (reg) + 4, val)
 
 #endif /* __SOUND_AK4XXX_ADDA_H */
diff --git a/sound/i2c/other/ak4xxx-adda.c b/sound/i2c/other/ak4xxx-adda.c
index 89fc3cbc2356..c34cb4684607 100644
--- a/sound/i2c/other/ak4xxx-adda.c
+++ b/sound/i2c/other/ak4xxx-adda.c
@@ -28,12 +28,14 @@
 #include <linux/init.h>
 #include <sound/core.h>
 #include <sound/control.h>
+#include <sound/tlv.h>
 #include <sound/ak4xxx-adda.h>
 
 MODULE_AUTHOR("Jaroslav Kysela <perex@suse.cz>, Takashi Iwai <tiwai@suse.de>");
 MODULE_DESCRIPTION("Routines for control of AK452x / AK43xx  AD/DA converters");
 MODULE_LICENSE("GPL");
 
+/* write the given register and save the data to the cache */
 void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg,
 		       unsigned char val)
 {
@@ -41,15 +43,10 @@ void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg,
 	ak->ops.write(ak, chip, reg, val);
 
 	/* save the data */
-	if (ak->type == SND_AK4524 || ak->type == SND_AK4528) {
-		if ((reg != 0x04 && reg != 0x05) || (val & 0x80) == 0)
-			snd_akm4xxx_set(ak, chip, reg, val);
-		else
-			snd_akm4xxx_set_ipga(ak, chip, reg, val);
-	} else {
-		/* AK4529, or else */
+	/* don't overwrite with IPGA data */
+	if ((ak->type != SND_AK4524 && ak->type != SND_AK5365) ||
+	    (reg != 0x04 && reg != 0x05) || (val & 0x80) == 0)
 		snd_akm4xxx_set(ak, chip, reg, val);
-	}
 	ak->ops.unlock(ak, chip);
 }
 
@@ -78,7 +75,7 @@ static void ak4524_reset(struct snd_akm4xxx *ak, int state)
 		/* IPGA */
 		for (reg = 0x04; reg < 0x06; reg++)
 			snd_akm4xxx_write(ak, chip, reg,
-					  snd_akm4xxx_get_ipga(ak, chip, reg));
+					  snd_akm4xxx_get_ipga(ak, chip, reg) | 0x80);
 	}
 }
 
@@ -144,6 +141,42 @@ void snd_akm4xxx_reset(struct snd_akm4xxx *ak, int state)
 
 EXPORT_SYMBOL(snd_akm4xxx_reset);
 
+
+/*
+ * Volume conversion table for non-linear volumes
+ * from -63.5dB (mute) to 0dB step 0.5dB
+ *
+ * Used for AK4524 input/ouput attenuation, AK4528, and
+ * AK5365 input attenuation
+ */
+static unsigned char vol_cvt_datt[128] = {
+	0x00, 0x01, 0x01, 0x02, 0x02, 0x03, 0x03, 0x04,
+	0x04, 0x04, 0x04, 0x05, 0x05, 0x05, 0x06, 0x06,
+	0x06, 0x07, 0x07, 0x08, 0x08, 0x08, 0x09, 0x0a,
+	0x0a, 0x0b, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, 0x0f,
+	0x10, 0x10, 0x11, 0x12, 0x12, 0x13, 0x13, 0x14,
+	0x15, 0x16, 0x17, 0x17, 0x18, 0x19, 0x1a, 0x1c,
+	0x1d, 0x1e, 0x1f, 0x20, 0x21, 0x22, 0x23, 0x23,
+	0x24, 0x25, 0x26, 0x28, 0x29, 0x2a, 0x2b, 0x2d,
+	0x2e, 0x30, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35,
+	0x37, 0x38, 0x39, 0x3b, 0x3c, 0x3e, 0x3f, 0x40,
+	0x41, 0x42, 0x43, 0x44, 0x46, 0x47, 0x48, 0x4a,
+	0x4b, 0x4d, 0x4e, 0x50, 0x51, 0x52, 0x53, 0x54,
+	0x55, 0x56, 0x58, 0x59, 0x5b, 0x5c, 0x5e, 0x5f,
+	0x60, 0x61, 0x62, 0x64, 0x65, 0x66, 0x67, 0x69,
+	0x6a, 0x6c, 0x6d, 0x6f, 0x70, 0x71, 0x72, 0x73,
+	0x75, 0x76, 0x77, 0x79, 0x7a, 0x7c, 0x7d, 0x7f,
+};
+
+/*
+ * dB tables
+ */
+static DECLARE_TLV_DB_SCALE(db_scale_vol_datt, -6350, 50, 1);
+static DECLARE_TLV_DB_SCALE(db_scale_8bit, -12750, 50, 1);
+static DECLARE_TLV_DB_SCALE(db_scale_7bit, -6350, 50, 1);
+static DECLARE_TLV_DB_LINEAR(db_scale_linear, TLV_DB_GAIN_MUTE, 0);
+static DECLARE_TLV_DB_SCALE(db_scale_ipga, 0, 50, 0);
+
 /*
  * initialize all the ak4xxx chips
  */
@@ -240,6 +273,9 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak)
 	int chip, num_chips;
 	unsigned char *ptr, reg, data, *inits;
 
+	memset(ak->images, 0, sizeof(ak->images));
+	memset(ak->volumes, 0, sizeof(ak->volumes));
+
 	switch (ak->type) {
 	case SND_AK4524:
 		inits = inits_ak4524;
@@ -265,6 +301,9 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak)
 		inits = inits_ak4381;
 		num_chips = ak->num_dacs / 2;
 		break;
+	case SND_AK5365:
+		/* FIXME: any init sequence? */
+		return;
 	default:
 		snd_BUG();
 		return;
@@ -282,16 +321,21 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak)
 
 EXPORT_SYMBOL(snd_akm4xxx_init);
 
+/*
+ * Mixer callbacks
+ */
+#define AK_VOL_CVT 			(1<<21)	/* need dB conversion */
+#define AK_NEEDSMSB 			(1<<22)	/* need MSB update bit */
+#define AK_INVERT 			(1<<23)	/* data is inverted */
 #define AK_GET_CHIP(val)		(((val) >> 8) & 0xff)
 #define AK_GET_ADDR(val)		((val) & 0xff)
-#define AK_GET_SHIFT(val)		(((val) >> 16) & 0x3f)
+#define AK_GET_SHIFT(val)		(((val) >> 16) & 0x1f)
+#define AK_GET_VOL_CVT(val)		(((val) >> 21) & 1)
 #define AK_GET_NEEDSMSB(val)		(((val) >> 22) & 1)
 #define AK_GET_INVERT(val)		(((val) >> 23) & 1)
 #define AK_GET_MASK(val)		(((val) >> 24) & 0xff)
 #define AK_COMPOSE(chip,addr,shift,mask) \
 	(((chip) << 8) | (addr) | ((shift) << 16) | ((mask) << 24))
-#define AK_NEEDSMSB 			(1<<22)
-#define AK_INVERT 			(1<<23)
 
 static int snd_akm4xxx_volume_info(struct snd_kcontrol *kcontrol,
 				   struct snd_ctl_elem_info *uinfo)
@@ -311,37 +355,37 @@ static int snd_akm4xxx_volume_get(struct snd_kcontrol *kcontrol,
 	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
 	int chip = AK_GET_CHIP(kcontrol->private_value);
 	int addr = AK_GET_ADDR(kcontrol->private_value);
-	int needsmsb = AK_GET_NEEDSMSB(kcontrol->private_value);
-	int invert = AK_GET_INVERT(kcontrol->private_value);
-	unsigned int mask = AK_GET_MASK(kcontrol->private_value);
-	unsigned char val = snd_akm4xxx_get(ak, chip, addr);
 
-	if (needsmsb)
-		val &= 0x7f;
-	ucontrol->value.integer.value[0] = invert ? mask - val : val;
+	ucontrol->value.integer.value[0] = snd_akm4xxx_get_vol(ak, chip, addr);
 	return 0;
 }
 
-static int snd_akm4xxx_volume_put(struct snd_kcontrol *kcontrol,
-				  struct snd_ctl_elem_value *ucontrol)
+static int put_ak_reg(struct snd_kcontrol *kcontrol, int addr,
+		      unsigned char nval)
 {
 	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
-	int chip = AK_GET_CHIP(kcontrol->private_value);
-	int addr = AK_GET_ADDR(kcontrol->private_value);
-	int needsmsb = AK_GET_NEEDSMSB(kcontrol->private_value);
-	int invert = AK_GET_INVERT(kcontrol->private_value);
 	unsigned int mask = AK_GET_MASK(kcontrol->private_value);
-	unsigned char nval = ucontrol->value.integer.value[0] % (mask+1);
-	int change;
+	int chip = AK_GET_CHIP(kcontrol->private_value);
 
-	if (invert)
+	if (snd_akm4xxx_get_vol(ak, chip, addr) == nval)
+		return 0;
+
+	snd_akm4xxx_set_vol(ak, chip, addr, nval);
+	if (AK_GET_VOL_CVT(kcontrol->private_value))
+		nval = vol_cvt_datt[nval];
+	if (AK_GET_INVERT(kcontrol->private_value))
 		nval = mask - nval;
-	if (needsmsb)
+	if (AK_GET_NEEDSMSB(kcontrol->private_value))
 		nval |= 0x80;
-	change = snd_akm4xxx_get(ak, chip, addr) != nval;
-	if (change)
-		snd_akm4xxx_write(ak, chip, addr, nval);
-	return change;
+	snd_akm4xxx_write(ak, chip, addr, nval);
+	return 1;
+}
+
+static int snd_akm4xxx_volume_put(struct snd_kcontrol *kcontrol,
+				  struct snd_ctl_elem_value *ucontrol)
+{
+	return put_ak_reg(kcontrol, AK_GET_ADDR(kcontrol->private_value),
+			  ucontrol->value.integer.value[0]);
 }
 
 static int snd_akm4xxx_stereo_volume_info(struct snd_kcontrol *kcontrol,
@@ -362,66 +406,25 @@ static int snd_akm4xxx_stereo_volume_get(struct snd_kcontrol *kcontrol,
 	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
 	int chip = AK_GET_CHIP(kcontrol->private_value);
 	int addr = AK_GET_ADDR(kcontrol->private_value);
-	int needsmsb = AK_GET_NEEDSMSB(kcontrol->private_value);
-	int invert = AK_GET_INVERT(kcontrol->private_value);
-	unsigned int mask = AK_GET_MASK(kcontrol->private_value);
-	unsigned char val;
-
-	val = snd_akm4xxx_get(ak, chip, addr);
-	if (needsmsb)
-		val &= 0x7f;
-	ucontrol->value.integer.value[0] = invert ? mask - val : val;
-
-	val = snd_akm4xxx_get(ak, chip, addr+1);
-	if (needsmsb)
-		val &= 0x7f;
-	ucontrol->value.integer.value[1] = invert ? mask - val : val;
 
+	ucontrol->value.integer.value[0] = snd_akm4xxx_get_vol(ak, chip, addr);
+	ucontrol->value.integer.value[1] = snd_akm4xxx_get_vol(ak, chip, addr+1);
 	return 0;
 }
 
 static int snd_akm4xxx_stereo_volume_put(struct snd_kcontrol *kcontrol,
 					 struct snd_ctl_elem_value *ucontrol)
 {
-	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
-	int chip = AK_GET_CHIP(kcontrol->private_value);
 	int addr = AK_GET_ADDR(kcontrol->private_value);
-	int needsmsb = AK_GET_NEEDSMSB(kcontrol->private_value);
-	int invert = AK_GET_INVERT(kcontrol->private_value);
-	unsigned int mask = AK_GET_MASK(kcontrol->private_value);
-	unsigned char nval = ucontrol->value.integer.value[0] % (mask+1);
-	int change0, change1;
-
-	if (invert)
-		nval = mask - nval;
-	if (needsmsb)
-		nval |= 0x80;
-	change0 = snd_akm4xxx_get(ak, chip, addr) != nval;
-	if (change0)
-		snd_akm4xxx_write(ak, chip, addr, nval);
-
-	nval = ucontrol->value.integer.value[1] % (mask+1);
-	if (invert)
-		nval = mask - nval;
-	if (needsmsb)
-		nval |= 0x80;
-	change1 = snd_akm4xxx_get(ak, chip, addr+1) != nval;
-	if (change1)
-		snd_akm4xxx_write(ak, chip, addr+1, nval);
-
+	int change;
 
-	return change0 || change1;
+	change = put_ak_reg(kcontrol, addr, ucontrol->value.integer.value[0]);
+	change |= put_ak_reg(kcontrol, addr + 1,
+			     ucontrol->value.integer.value[1]);
+	return change;
 }
 
-static int snd_akm4xxx_ipga_gain_info(struct snd_kcontrol *kcontrol,
-				      struct snd_ctl_elem_info *uinfo)
-{
-	uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
-	uinfo->count = 1;
-	uinfo->value.integer.min = 0;
-	uinfo->value.integer.max = 36;
-	return 0;
-}
+#define snd_akm4xxx_ipga_gain_info	snd_akm4xxx_volume_info
 
 static int snd_akm4xxx_ipga_gain_get(struct snd_kcontrol *kcontrol,
 				     struct snd_ctl_elem_value *ucontrol)
@@ -429,21 +432,57 @@ static int snd_akm4xxx_ipga_gain_get(struct snd_kcontrol *kcontrol,
 	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
 	int chip = AK_GET_CHIP(kcontrol->private_value);
 	int addr = AK_GET_ADDR(kcontrol->private_value);
+
 	ucontrol->value.integer.value[0] =
-		snd_akm4xxx_get_ipga(ak, chip, addr) & 0x7f;
+		snd_akm4xxx_get_ipga(ak, chip, addr);
 	return 0;
 }
 
+static int put_ak_ipga(struct snd_kcontrol *kcontrol, int addr,
+		       unsigned char nval)
+{
+	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
+	int chip = AK_GET_CHIP(kcontrol->private_value);
+
+	if (snd_akm4xxx_get_ipga(ak, chip, addr) == nval)
+		return 0;
+	snd_akm4xxx_set_ipga(ak, chip, addr, nval);
+	snd_akm4xxx_write(ak, chip, addr, nval | 0x80); /* need MSB */
+	return 1;
+}
+
 static int snd_akm4xxx_ipga_gain_put(struct snd_kcontrol *kcontrol,
 				     struct snd_ctl_elem_value *ucontrol)
+{
+	return put_ak_ipga(kcontrol, AK_GET_ADDR(kcontrol->private_value),
+			   ucontrol->value.integer.value[0]);
+}
+
+#define snd_akm4xxx_stereo_gain_info	snd_akm4xxx_stereo_volume_info
+
+static int snd_akm4xxx_stereo_gain_get(struct snd_kcontrol *kcontrol,
+				       struct snd_ctl_elem_value *ucontrol)
 {
 	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
 	int chip = AK_GET_CHIP(kcontrol->private_value);
 	int addr = AK_GET_ADDR(kcontrol->private_value);
-	unsigned char nval = (ucontrol->value.integer.value[0] % 37) | 0x80;
-	int change = snd_akm4xxx_get_ipga(ak, chip, addr) != nval;
-	if (change)
-		snd_akm4xxx_write(ak, chip, addr, nval);
+
+	ucontrol->value.integer.value[0] =
+		snd_akm4xxx_get_ipga(ak, chip, addr);
+	ucontrol->value.integer.value[1] =
+		snd_akm4xxx_get_ipga(ak, chip, addr + 1);
+	return 0;
+}
+
+static int snd_akm4xxx_stereo_gain_put(struct snd_kcontrol *kcontrol,
+				       struct snd_ctl_elem_value *ucontrol)
+{
+	int addr = AK_GET_ADDR(kcontrol->private_value);
+	int change;
+
+	change = put_ak_ipga(kcontrol, addr, ucontrol->value.integer.value[0]);
+	change |= put_ak_ipga(kcontrol, addr + 1,
+			      ucontrol->value.integer.value[1]);
 	return change;
 }
 
@@ -548,221 +587,247 @@ static int ak4xxx_switch_put(struct snd_kcontrol *kcontrol,
  * build AK4xxx controls
  */
 
-int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak)
+static int build_dac_controls(struct snd_akm4xxx *ak)
 {
-	unsigned int idx, num_emphs;
-	struct snd_kcontrol *ctl;
-	int err;
-	int mixer_ch = 0;
-	int num_stereo;
-
-	ctl = kmalloc(sizeof(*ctl), GFP_KERNEL);
-	if (! ctl)
-		return -ENOMEM;
+	int idx, err, mixer_ch, num_stereo;
+	struct snd_kcontrol_new knew;
 
+	mixer_ch = 0;
 	for (idx = 0; idx < ak->num_dacs; ) {
-		memset(ctl, 0, sizeof(*ctl));
-		if (ak->channel_names == NULL) {
-			strcpy(ctl->id.name, "DAC Volume");
+		memset(&knew, 0, sizeof(knew));
+		if (! ak->dac_info || ! ak->dac_info[mixer_ch].name) {
+			knew.name = "DAC Volume";
+			knew.index = mixer_ch + ak->idx_offset * 2;
 			num_stereo = 1;
-			ctl->id.index = mixer_ch + ak->idx_offset * 2;
 		} else {
-			strcpy(ctl->id.name, ak->channel_names[mixer_ch]);
-			num_stereo = ak->num_stereo[mixer_ch];
-			ctl->id.index = 0;
+			knew.name = ak->dac_info[mixer_ch].name;
+			num_stereo = ak->dac_info[mixer_ch].num_channels;
 		}
-		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
-		ctl->count = 1;
+		knew.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
+		knew.count = 1;
+		knew.access = SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			SNDRV_CTL_ELEM_ACCESS_TLV_READ;
 		if (num_stereo == 2) {
-			ctl->info = snd_akm4xxx_stereo_volume_info;
-			ctl->get = snd_akm4xxx_stereo_volume_get;
-			ctl->put = snd_akm4xxx_stereo_volume_put;
+			knew.info = snd_akm4xxx_stereo_volume_info;
+			knew.get = snd_akm4xxx_stereo_volume_get;
+			knew.put = snd_akm4xxx_stereo_volume_put;
 		} else {
-			ctl->info = snd_akm4xxx_volume_info;
-			ctl->get = snd_akm4xxx_volume_get;
-			ctl->put = snd_akm4xxx_volume_put;
+			knew.info = snd_akm4xxx_volume_info;
+			knew.get = snd_akm4xxx_volume_get;
+			knew.put = snd_akm4xxx_volume_put;
 		}
 		switch (ak->type) {
 		case SND_AK4524:
 			/* register 6 & 7 */
-			ctl->private_value =
-				AK_COMPOSE(idx/2, (idx%2) + 6, 0, 127);
+			knew.private_value =
+				AK_COMPOSE(idx/2, (idx%2) + 6, 0, 127) |
+				AK_VOL_CVT;
+			knew.tlv.p = db_scale_vol_datt;
 			break;
 		case SND_AK4528:
 			/* register 4 & 5 */
-			ctl->private_value =
-				AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127);
+			knew.private_value =
+				AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127) |
+				AK_VOL_CVT;
+			knew.tlv.p = db_scale_vol_datt;
 			break;
 		case SND_AK4529: {
 			/* registers 2-7 and b,c */
 			int val = idx < 6 ? idx + 2 : (idx - 6) + 0xb;
-			ctl->private_value =
+			knew.private_value =
 				AK_COMPOSE(0, val, 0, 255) | AK_INVERT;
+			knew.tlv.p = db_scale_8bit;
 			break;
 		}
 		case SND_AK4355:
 			/* register 4-9, chip #0 only */
-			ctl->private_value = AK_COMPOSE(0, idx + 4, 0, 255);
+			knew.private_value = AK_COMPOSE(0, idx + 4, 0, 255);
+			knew.tlv.p = db_scale_8bit;
 			break;
 		case SND_AK4358: {
 			/* register 4-9 and 11-12, chip #0 only */
 			int  addr = idx < 6 ? idx + 4 : idx + 5;
-			ctl->private_value =
+			knew.private_value =
 				AK_COMPOSE(0, addr, 0, 127) | AK_NEEDSMSB;
+			knew.tlv.p = db_scale_7bit;
 			break;
 		}
 		case SND_AK4381:
 			/* register 3 & 4 */
-			ctl->private_value =
+			knew.private_value =
 				AK_COMPOSE(idx/2, (idx%2) + 3, 0, 255);
+			knew.tlv.p = db_scale_linear;
 			break;
 		default:
-			err = -EINVAL;
-			goto __error;
+			return -EINVAL;
 		}
 
-		ctl->private_data = ak;
-		err = snd_ctl_add(ak->card,
-				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
-					      SNDRV_CTL_ELEM_ACCESS_WRITE));
+		err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
 		if (err < 0)
-			goto __error;
+			return err;
 
 		idx += num_stereo;
 		mixer_ch++;
 	}
-	for (idx = 0; idx < ak->num_adcs && ak->type == SND_AK4524; ++idx) {
-		memset(ctl, 0, sizeof(*ctl));
-		strcpy(ctl->id.name, "ADC Volume");
-		ctl->id.index = idx + ak->idx_offset * 2;
-		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
-		ctl->count = 1;
-		ctl->info = snd_akm4xxx_volume_info;
-		ctl->get = snd_akm4xxx_volume_get;
-		ctl->put = snd_akm4xxx_volume_put;
-		/* register 4 & 5 */
-		ctl->private_value =
-			AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127);
-		ctl->private_data = ak;
-		err = snd_ctl_add(ak->card,
-				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
-					      SNDRV_CTL_ELEM_ACCESS_WRITE));
-		if (err < 0)
-			goto __error;
-
-		memset(ctl, 0, sizeof(*ctl));
-		strcpy(ctl->id.name, "IPGA Analog Capture Volume");
-		ctl->id.index = idx + ak->idx_offset * 2;
-		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
-		ctl->count = 1;
-		ctl->info = snd_akm4xxx_ipga_gain_info;
-		ctl->get = snd_akm4xxx_ipga_gain_get;
-		ctl->put = snd_akm4xxx_ipga_gain_put;
+	return 0;
+}
+
+static int build_adc_controls(struct snd_akm4xxx *ak)
+{
+	int idx, err, mixer_ch, num_stereo;
+	struct snd_kcontrol_new knew;
+
+	mixer_ch = 0;
+	for (idx = 0; idx < ak->num_adcs;) {
+		memset(&knew, 0, sizeof(knew));
+		if (! ak->adc_info || ! ak->adc_info[mixer_ch].name) {
+			knew.name = "ADC Volume";
+			knew.index = mixer_ch + ak->idx_offset * 2;
+			num_stereo = 1;
+		} else {
+			knew.name = ak->adc_info[mixer_ch].name;
+			num_stereo = ak->adc_info[mixer_ch].num_channels;
+		}
+		knew.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
+		knew.count = 1;
+		knew.access = SNDRV_CTL_ELEM_ACCESS_READWRITE |
+			SNDRV_CTL_ELEM_ACCESS_TLV_READ;
+		if (num_stereo == 2) {
+			knew.info = snd_akm4xxx_stereo_volume_info;
+			knew.get = snd_akm4xxx_stereo_volume_get;
+			knew.put = snd_akm4xxx_stereo_volume_put;
+		} else {
+			knew.info = snd_akm4xxx_volume_info;
+			knew.get = snd_akm4xxx_volume_get;
+			knew.put = snd_akm4xxx_volume_put;
+		}
 		/* register 4 & 5 */
-		ctl->private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0, 0);
-		ctl->private_data = ak;
-		err = snd_ctl_add(ak->card,
-				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
-					      SNDRV_CTL_ELEM_ACCESS_WRITE));
+		knew.private_value =
+			AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127) |
+			AK_VOL_CVT;
+		knew.tlv.p = db_scale_vol_datt;
+		err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
 		if (err < 0)
-			goto __error;
-	}
+			return err;
 
-	if (ak->type == SND_AK5365) {
-		memset(ctl, 0, sizeof(*ctl));
-		if (ak->channel_names == NULL)
-			strcpy(ctl->id.name, "Capture Volume");
+		if (! ak->adc_info || ! ak->adc_info[mixer_ch].gain_name)
+			knew.name = "IPGA Analog Capture Volume";
 		else
-			strcpy(ctl->id.name, ak->channel_names[0]);
-		ctl->id.index = ak->idx_offset * 2;
-		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
-		ctl->count = 1;
-		ctl->info = snd_akm4xxx_stereo_volume_info;
-		ctl->get = snd_akm4xxx_stereo_volume_get;
-		ctl->put = snd_akm4xxx_stereo_volume_put;
-		/* Registers 4 & 5 (see AK5365 data sheet, pages 34 and 35):
-		 * valid values are from 0x00 (mute) to 0x98 (+12dB).  */
-		ctl->private_value =
-			AK_COMPOSE(0, 4, 0, 0x98);
-		ctl->private_data = ak;
-		err = snd_ctl_add(ak->card,
-				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
-					      SNDRV_CTL_ELEM_ACCESS_WRITE));
+			knew.name = ak->adc_info[mixer_ch].gain_name;
+		if (num_stereo == 2) {
+			knew.info = snd_akm4xxx_stereo_gain_info;
+			knew.get = snd_akm4xxx_stereo_gain_get;
+			knew.put = snd_akm4xxx_stereo_gain_put;
+		} else {
+			knew.info = snd_akm4xxx_ipga_gain_info;
+			knew.get = snd_akm4xxx_ipga_gain_get;
+			knew.put = snd_akm4xxx_ipga_gain_put;
+		}
+		/* register 4 & 5 */
+		if (ak->type == SND_AK4524)
+			knew.private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0,
+							24);
+		else /* AK5365 */
+			knew.private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0,
+							36);
+		knew.tlv.p = db_scale_ipga;
+		err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
 		if (err < 0)
-			goto __error;
+			return err;
+
+		if (ak->type == SND_AK5365 && (idx % 2) == 0) {
+			if (! ak->adc_info || 
+			    ! ak->adc_info[mixer_ch].switch_name)
+				knew.name = "Capture Switch";
+			else
+				knew.name = ak->adc_info[mixer_ch].switch_name;
+			knew.info = ak4xxx_switch_info;
+			knew.get = ak4xxx_switch_get;
+			knew.put = ak4xxx_switch_put;
+			knew.access = 0;
+			/* register 2, bit 0 (SMUTE): 0 = normal operation,
+			   1 = mute */
+			knew.private_value =
+				AK_COMPOSE(idx/2, 2, 0, 0) | AK_INVERT;
+			err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
+			if (err < 0)
+				return err;
+		}
 
-		memset(ctl, 0, sizeof(*ctl));
-		if (ak->channel_names == NULL)
-			strcpy(ctl->id.name, "Capture Switch");
-		else
-			strcpy(ctl->id.name, ak->channel_names[1]);
-		ctl->id.index = ak->idx_offset * 2;
-		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
-		ctl->count = 1;
-		ctl->info = ak4xxx_switch_info;
-		ctl->get = ak4xxx_switch_get;
-		ctl->put = ak4xxx_switch_put;
-		/* register 2, bit 0 (SMUTE): 0 = normal operation, 1 = mute */
-		ctl->private_value =
-			AK_COMPOSE(0, 2, 0, 0) | AK_INVERT;
-		ctl->private_data = ak;
-		err = snd_ctl_add(ak->card,
-				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
-					      SNDRV_CTL_ELEM_ACCESS_WRITE));
-		if (err < 0)
-			goto __error;
+		idx += num_stereo;
+		mixer_ch++;
 	}
+	return 0;
+}
+
+static int build_deemphasis(struct snd_akm4xxx *ak, int num_emphs)
+{
+	int idx, err;
+	struct snd_kcontrol_new knew;
 
-	if (ak->type == SND_AK4355 || ak->type == SND_AK4358)
-		num_emphs = 1;
-	else
-		num_emphs = ak->num_dacs / 2;
 	for (idx = 0; idx < num_emphs; idx++) {
-		memset(ctl, 0, sizeof(*ctl));
-		strcpy(ctl->id.name, "Deemphasis");
-		ctl->id.index = idx + ak->idx_offset;
-		ctl->id.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
-		ctl->count = 1;
-		ctl->info = snd_akm4xxx_deemphasis_info;
-		ctl->get = snd_akm4xxx_deemphasis_get;
-		ctl->put = snd_akm4xxx_deemphasis_put;
+		memset(&knew, 0, sizeof(knew));
+		knew.name = "Deemphasis";
+		knew.index = idx + ak->idx_offset;
+		knew.iface = SNDRV_CTL_ELEM_IFACE_MIXER;
+		knew.count = 1;
+		knew.info = snd_akm4xxx_deemphasis_info;
+		knew.get = snd_akm4xxx_deemphasis_get;
+		knew.put = snd_akm4xxx_deemphasis_put;
 		switch (ak->type) {
 		case SND_AK4524:
 		case SND_AK4528:
 			/* register 3 */
-			ctl->private_value = AK_COMPOSE(idx, 3, 0, 0);
+			knew.private_value = AK_COMPOSE(idx, 3, 0, 0);
 			break;
 		case SND_AK4529: {
 			int shift = idx == 3 ? 6 : (2 - idx) * 2;
 			/* register 8 with shift */
-			ctl->private_value = AK_COMPOSE(0, 8, shift, 0);
+			knew.private_value = AK_COMPOSE(0, 8, shift, 0);
 			break;
 		}
 		case SND_AK4355:
 		case SND_AK4358:
-			ctl->private_value = AK_COMPOSE(idx, 3, 0, 0);
+			knew.private_value = AK_COMPOSE(idx, 3, 0, 0);
 			break;
 		case SND_AK4381:
-			ctl->private_value = AK_COMPOSE(idx, 1, 1, 0);
+			knew.private_value = AK_COMPOSE(idx, 1, 1, 0);
 			break;
 		default:
-			err = -EINVAL;
-			goto __error;
+			return -EINVAL;
 		}
-		ctl->private_data = ak;
-		err = snd_ctl_add(ak->card,
-				  snd_ctl_new(ctl, SNDRV_CTL_ELEM_ACCESS_READ|
-					      SNDRV_CTL_ELEM_ACCESS_WRITE));
+		err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
 		if (err < 0)
-			goto __error;
+			return err;
 	}
-	err = 0;
-
- __error:
-	kfree(ctl);
-	return err;
+	return 0;
 }
 
+int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak)
+{
+	int err, num_emphs;
+
+	err = build_dac_controls(ak);
+	if (err < 0)
+		return err;
+
+	if (ak->type == SND_AK4524 || ak->type == SND_AK5365) {
+		err = build_adc_controls(ak);
+		if (err < 0)
+			return err;
+	}
+
+	if (ak->type == SND_AK4355 || ak->type == SND_AK4358)
+		num_emphs = 1;
+	else
+		num_emphs = ak->num_dacs / 2;
+	err = build_deemphasis(ak, num_emphs);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+	
 EXPORT_SYMBOL(snd_akm4xxx_build_controls);
 
 static int __init alsa_akm4xxx_module_init(void)
diff --git a/sound/pci/ice1712/revo.c b/sound/pci/ice1712/revo.c
index 1134a57f9e65..c9eefa9bbfff 100644
--- a/sound/pci/ice1712/revo.c
+++ b/sound/pci/ice1712/revo.c
@@ -87,19 +87,34 @@ static void revo_set_rate_val(struct snd_akm4xxx *ak, unsigned int rate)
  * initialize the chips on M-Audio Revolution cards
  */
 
-static unsigned int revo71_num_stereo_front[] = {2};
-static char *revo71_channel_names_front[] = {"PCM Playback Volume"};
+#define AK_DAC(xname,xch) { .name = xname, .num_channels = xch }
 
-static unsigned int revo71_num_stereo_surround[] = {1, 1, 2, 2};
-static char *revo71_channel_names_surround[] = {"PCM Center Playback Volume", "PCM LFE Playback Volume",
-						"PCM Side Playback Volume", "PCM Rear Playback Volume"};
+static struct snd_akm4xxx_dac_channel revo71_front[] = {
+	AK_DAC("PCM Playback Volume", 2)
+};
+
+static struct snd_akm4xxx_dac_channel revo71_surround[] = {
+	AK_DAC("PCM Center Playback Volume", 1),
+	AK_DAC("PCM LFE Playback Volume", 1),
+	AK_DAC("PCM Side Playback Volume", 2),
+	AK_DAC("PCM Rear Playback Volume", 2),
+};
 
-static unsigned int revo51_num_stereo[] = {2, 1, 1, 2};
-static char *revo51_channel_names[] = {"PCM Playback Volume", "PCM Center Playback Volume",
-					"PCM LFE Playback Volume", "PCM Rear Playback Volume"};
+static struct snd_akm4xxx_dac_channel revo51_dac[] = {
+	AK_DAC("PCM Playback Volume", 2),
+	AK_DAC("PCM Center Playback Volume", 1),
+	AK_DAC("PCM LFE Playback Volume", 1),
+	AK_DAC("PCM Rear Playback Volume", 2),
+};
 
-static unsigned int revo51_adc_num_stereo[] = {2};
-static char *revo51_adc_channel_names[] = {"PCM Capture Volume","PCM Capture Switch"};
+static struct snd_akm4xxx_adc_channel revo51_adc[] = {
+	{
+		.name = "PCM Capture Volume",
+		.gain_name = "PCM Capture Gain Volume",
+		.switch_name = "PCM Capture Switch",
+		.num_channels = 2
+	},
+};
 
 static struct snd_akm4xxx akm_revo_front __devinitdata = {
 	.type = SND_AK4381,
@@ -107,8 +122,7 @@ static struct snd_akm4xxx akm_revo_front __devinitdata = {
 	.ops = {
 		.set_rate_val = revo_set_rate_val
 	},
-	.num_stereo = revo71_num_stereo_front,
-	.channel_names = revo71_channel_names_front
+	.dac_info = revo71_front,
 };
 
 static struct snd_ak4xxx_private akm_revo_front_priv __devinitdata = {
@@ -130,8 +144,7 @@ static struct snd_akm4xxx akm_revo_surround __devinitdata = {
 	.ops = {
 		.set_rate_val = revo_set_rate_val
 	},
-	.num_stereo = revo71_num_stereo_surround,
-	.channel_names = revo71_channel_names_surround
+	.dac_info = revo71_surround,
 };
 
 static struct snd_ak4xxx_private akm_revo_surround_priv __devinitdata = {
@@ -152,8 +165,7 @@ static struct snd_akm4xxx akm_revo51 __devinitdata = {
 	.ops = {
 		.set_rate_val = revo_set_rate_val
 	},
-	.num_stereo = revo51_num_stereo,
-	.channel_names = revo51_channel_names
+	.dac_info = revo51_dac,
 };
 
 static struct snd_ak4xxx_private akm_revo51_priv __devinitdata = {
@@ -171,8 +183,7 @@ static struct snd_ak4xxx_private akm_revo51_priv __devinitdata = {
 static struct snd_akm4xxx akm_revo51_adc __devinitdata = {
 	.type = SND_AK5365,
 	.num_adcs = 2,
-	.num_stereo = revo51_adc_num_stereo,
-	.channel_names = revo51_adc_channel_names
+	.adc_info = revo51_adc,
 };
 
 static struct snd_ak4xxx_private akm_revo51_adc_priv __devinitdata = {
-- 
cgit v1.2.3


From 55a29af5ed5d914f017e6a7c613a4d7cc34f82d9 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 6 Sep 2006 12:15:34 +0200
Subject: [ALSA] Add definition of TLV dB range compound

Added the definition of TLV dB range compound.  It contains one or
more dB-range or linear-volume TLV entries with min/max ranges.
Used for volume controls with non-linear curves.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/tlv.h | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/sound/tlv.h b/include/sound/tlv.h
index 7905841643df..d93a96b91875 100644
--- a/include/sound/tlv.h
+++ b/include/sound/tlv.h
@@ -34,19 +34,26 @@
 #define SNDRV_CTL_TLVT_CONTAINER 0	/* one level down - group of TLVs */
 #define SNDRV_CTL_TLVT_DB_SCALE	1       /* dB scale */
 #define SNDRV_CTL_TLVT_DB_LINEAR 2	/* linear volume */
+#define SNDRV_CTL_TLVT_DB_RANGE 3	/* dB range container */
 
+#define TLV_DB_SCALE_ITEM(min, step, mute)			\
+	SNDRV_CTL_TLVT_DB_SCALE, 2 * sizeof(unsigned int),	\
+	(min), ((step) & 0xffff) | ((mute) ? 0x10000 : 0)
 #define DECLARE_TLV_DB_SCALE(name, min, step, mute) \
-unsigned int name[] = { \
-        SNDRV_CTL_TLVT_DB_SCALE, 2 * sizeof(unsigned int), \
-        (min), ((step) & 0xffff) | ((mute) ? 0x10000 : 0) \
-}
+	unsigned int name[] = { TLV_DB_SCALE_ITEM(min, step, mute) }
 
 /* linear volume between min_dB and max_dB (.01dB unit) */
+#define TLV_DB_LINEAR_ITEM(min_dB, max_dB)		    \
+	SNDRV_CTL_TLVT_DB_LINEAR, 2 * sizeof(unsigned int), \
+	(min_dB), (max_dB)
 #define DECLARE_TLV_DB_LINEAR(name, min_dB, max_dB)	\
-unsigned int name[] = { \
-        SNDRV_CTL_TLVT_DB_LINEAR, 2 * sizeof(unsigned int), \
-        (min_dB), (max_dB)				\
-}
+	unsigned int name[] = { TLV_DB_LINEAR_ITEM(min_dB, max_dB) }
+
+/* dB range container */
+/* Each item is: <min> <max> <TLV> */
+/* The below assumes that each item TLV is 4 words like DB_SCALE or LINEAR */
+#define TLV_DB_RANGE_HEAD(num)			\
+	SNDRV_CTL_TLVT_DB_RANGE, 6 * (num) * sizeof(unsigned int)
 
 #define TLV_DB_GAIN_MUTE	-9999999
 
-- 
cgit v1.2.3


From 9d19f48cfe2570562c2c6226780a7ca627b0f1f1 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 6 Sep 2006 14:27:46 +0200
Subject: [ALSA] Add pcm_class attribute to PCM sysfs entry

This patch adds a new attribute, pcm_class, to each PCM sysfs entry.
It's useful to detect what kind of PCM stream is, for example, HAL
can check whether it's a modem or not.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/core.h |  4 ++++
 sound/core/pcm.c     | 24 ++++++++++++++++++++++
 sound/core/sound.c   | 56 +++++++++++++++++++++++++++++++++++++++++-----------
 3 files changed, 72 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/sound/core.h b/include/sound/core.h
index 1359c532b68e..b056ea925ecf 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -26,6 +26,7 @@
 #include <linux/mutex.h>		/* struct mutex */
 #include <linux/rwsem.h>		/* struct rw_semaphore */
 #include <linux/pm.h>			/* pm_message_t */
+#include <linux/device.h>
 
 /* forward declarations */
 #ifdef CONFIG_PCI
@@ -186,6 +187,7 @@ struct snd_minor {
 	int device;			/* device number */
 	const struct file_operations *f_ops;	/* file operations */
 	void *private_data;		/* private data for f_ops->open */
+	struct class_device *class_dev;	/* class device for sysfs */
 };
 
 /* sound.c */
@@ -200,6 +202,8 @@ int snd_register_device(int type, struct snd_card *card, int dev,
 			const char *name);
 int snd_unregister_device(int type, struct snd_card *card, int dev);
 void *snd_lookup_minor_data(unsigned int minor, int type);
+int snd_add_device_sysfs_file(int type, struct snd_card *card, int dev,
+			      const struct class_device_attribute *attr);
 
 #ifdef CONFIG_SND_OSSEMUL
 int snd_register_oss_device(int type, struct snd_card *card, int dev,
diff --git a/sound/core/pcm.c b/sound/core/pcm.c
index ed3b09469560..bf8f412988b8 100644
--- a/sound/core/pcm.c
+++ b/sound/core/pcm.c
@@ -907,6 +907,28 @@ void snd_pcm_detach_substream(struct snd_pcm_substream *substream)
 	substream->pstr->substream_opened--;
 }
 
+static ssize_t show_pcm_class(struct class_device *class_device, char *buf)
+{
+	struct snd_pcm *pcm;
+	const char *str;
+	static const char *strs[SNDRV_PCM_CLASS_LAST + 1] = {
+		[SNDRV_PCM_CLASS_GENERIC] = "generic",
+		[SNDRV_PCM_CLASS_MULTI] = "multi",
+		[SNDRV_PCM_CLASS_MODEM] = "modem",
+		[SNDRV_PCM_CLASS_DIGITIZER] = "digitizer",
+	};
+
+	if (! (pcm = class_get_devdata(class_device)) ||
+	    pcm->dev_class > SNDRV_PCM_CLASS_LAST)
+		str = "none";
+	else
+		str = strs[pcm->dev_class];
+        return snprintf(buf, PAGE_SIZE, "%s\n", str);
+}
+
+static struct class_device_attribute pcm_attrs =
+	__ATTR(pcm_class, S_IRUGO, show_pcm_class, NULL);
+
 static int snd_pcm_dev_register(struct snd_device *device)
 {
 	int cidx, err;
@@ -945,6 +967,8 @@ static int snd_pcm_dev_register(struct snd_device *device)
 			mutex_unlock(&register_mutex);
 			return err;
 		}
+		snd_add_device_sysfs_file(devtype, pcm->card, pcm->device,
+					  &pcm_attrs);
 		for (substream = pcm->streams[cidx].substream; substream; substream = substream->next)
 			snd_pcm_timer_init(substream);
 	}
diff --git a/sound/core/sound.c b/sound/core/sound.c
index b4430db3fa4c..efa476c5210a 100644
--- a/sound/core/sound.c
+++ b/sound/core/sound.c
@@ -268,7 +268,11 @@ int snd_register_device(int type, struct snd_card *card, int dev,
 	snd_minors[minor] = preg;
 	if (card)
 		device = card->dev;
-	class_device_create(sound_class, NULL, MKDEV(major, minor), device, "%s", name);
+	preg->class_dev = class_device_create(sound_class, NULL,
+					      MKDEV(major, minor),
+					      device, "%s", name);
+	if (preg->class_dev)
+		class_set_devdata(preg->class_dev, private_data);
 
 	mutex_unlock(&sound_mutex);
 	return 0;
@@ -276,6 +280,24 @@ int snd_register_device(int type, struct snd_card *card, int dev,
 
 EXPORT_SYMBOL(snd_register_device);
 
+/* find the matching minor record
+ * return the index of snd_minor, or -1 if not found
+ */
+static int find_snd_minor(int type, struct snd_card *card, int dev)
+{
+	int cardnum, minor;
+	struct snd_minor *mptr;
+
+	cardnum = card ? card->number : -1;
+	for (minor = 0; minor < ARRAY_SIZE(snd_minors); ++minor)
+		if ((mptr = snd_minors[minor]) != NULL &&
+		    mptr->type == type &&
+		    mptr->card == cardnum &&
+		    mptr->device == dev)
+			return minor;
+	return -1;
+}
+
 /**
  * snd_unregister_device - unregister the device on the given card
  * @type: the device type, SNDRV_DEVICE_TYPE_XXX
@@ -289,32 +311,42 @@ EXPORT_SYMBOL(snd_register_device);
  */
 int snd_unregister_device(int type, struct snd_card *card, int dev)
 {
-	int cardnum, minor;
-	struct snd_minor *mptr;
+	int minor;
 
-	cardnum = card ? card->number : -1;
 	mutex_lock(&sound_mutex);
-	for (minor = 0; minor < ARRAY_SIZE(snd_minors); ++minor)
-		if ((mptr = snd_minors[minor]) != NULL &&
-		    mptr->type == type &&
-		    mptr->card == cardnum &&
-		    mptr->device == dev)
-			break;
-	if (minor == ARRAY_SIZE(snd_minors)) {
+	minor = find_snd_minor(type, card, dev);
+	if (minor < 0) {
 		mutex_unlock(&sound_mutex);
 		return -EINVAL;
 	}
 
 	class_device_destroy(sound_class, MKDEV(major, minor));
 
+	kfree(snd_minors[minor]);
 	snd_minors[minor] = NULL;
 	mutex_unlock(&sound_mutex);
-	kfree(mptr);
 	return 0;
 }
 
 EXPORT_SYMBOL(snd_unregister_device);
 
+int snd_add_device_sysfs_file(int type, struct snd_card *card, int dev,
+			      const struct class_device_attribute *attr)
+{
+	int minor, ret = -EINVAL;
+	struct class_device *cdev;
+
+	mutex_lock(&sound_mutex);
+	minor = find_snd_minor(type, card, dev);
+	if (minor >= 0 && (cdev = snd_minors[minor]->class_dev) != NULL)
+		ret = class_device_create_file(cdev, attr);
+	mutex_unlock(&sound_mutex);
+	return ret;
+
+}
+
+EXPORT_SYMBOL(snd_add_device_sysfs_file);
+
 #ifdef CONFIG_PROC_FS
 /*
  *  INFO PART
-- 
cgit v1.2.3


From 854b66e44260320c21ebe4b8a18e189f2e45b5be Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 8 Sep 2006 12:27:38 +0200
Subject: [ALSA] ak4xxx - Remove bogus IPGA controls

Remove IPGA volume controls and merge the IPGA range to ADC volume
controls.  These two volumes are not really independent but connected
simply in different ranges 0-0x7f and 0x80-max.  It doesn't make sense
to provide two controls.
Since both 0x7f and 0x80 specify 0dB, a hack is needed for IPGA range
to skip 0x80 (increment one) for such controls.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
---
 include/sound/ak4xxx-adda.h   |  10 ----
 sound/i2c/other/ak4xxx-adda.c | 128 ++++++------------------------------------
 sound/pci/ice1712/revo.c      |   1 -
 3 files changed, 18 insertions(+), 121 deletions(-)

(limited to 'include')

diff --git a/include/sound/ak4xxx-adda.h b/include/sound/ak4xxx-adda.h
index 026e4072a9a1..d0deca669b92 100644
--- a/include/sound/ak4xxx-adda.h
+++ b/include/sound/ak4xxx-adda.h
@@ -48,7 +48,6 @@ struct snd_akm4xxx_dac_channel {
 /* ADC labels and channels */
 struct snd_akm4xxx_adc_channel {
 	char *name;		/* capture gain volume label */
-	char *gain_name;	/* IPGA */
 	char *switch_name;	/* capture switch */
 	unsigned int num_channels;
 };
@@ -91,13 +90,4 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak);
 #define snd_akm4xxx_set_vol(ak,chip,reg,val) \
 	((ak)->volumes[(chip) * 16 + (reg)] = (val))
 
-/* Warning: IPGA is tricky - we assume the addr + 4 is unused
- *   so far, it's OK for all AK codecs with IPGA:
- *   AK4524, AK4528 and EK5365
- */
-#define snd_akm4xxx_get_ipga(ak,chip,reg) \
-	snd_akm4xxx_get_vol(ak, chip, (reg) + 4)
-#define snd_akm4xxx_set_ipga(ak,chip,reg,val) \
-	snd_akm4xxx_set_vol(ak, chip, (reg) + 4, val)
-
 #endif /* __SOUND_AK4XXX_ADDA_H */
diff --git a/sound/i2c/other/ak4xxx-adda.c b/sound/i2c/other/ak4xxx-adda.c
index c34cb4684607..5da49e2eb350 100644
--- a/sound/i2c/other/ak4xxx-adda.c
+++ b/sound/i2c/other/ak4xxx-adda.c
@@ -43,10 +43,7 @@ void snd_akm4xxx_write(struct snd_akm4xxx *ak, int chip, unsigned char reg,
 	ak->ops.write(ak, chip, reg, val);
 
 	/* save the data */
-	/* don't overwrite with IPGA data */
-	if ((ak->type != SND_AK4524 && ak->type != SND_AK5365) ||
-	    (reg != 0x04 && reg != 0x05) || (val & 0x80) == 0)
-		snd_akm4xxx_set(ak, chip, reg, val);
+	snd_akm4xxx_set(ak, chip, reg, val);
 	ak->ops.unlock(ak, chip);
 }
 
@@ -70,12 +67,6 @@ static void ak4524_reset(struct snd_akm4xxx *ak, int state)
 		for (reg = 0x04; reg < maxreg; reg++)
 			snd_akm4xxx_write(ak, chip, reg,
 					  snd_akm4xxx_get(ak, chip, reg));
-		if (ak->type == SND_AK4528)
-			continue;
-		/* IPGA */
-		for (reg = 0x04; reg < 0x06; reg++)
-			snd_akm4xxx_write(ak, chip, reg,
-					  snd_akm4xxx_get_ipga(ak, chip, reg) | 0x80);
 	}
 }
 
@@ -175,7 +166,6 @@ static DECLARE_TLV_DB_SCALE(db_scale_vol_datt, -6350, 50, 1);
 static DECLARE_TLV_DB_SCALE(db_scale_8bit, -12750, 50, 1);
 static DECLARE_TLV_DB_SCALE(db_scale_7bit, -6350, 50, 1);
 static DECLARE_TLV_DB_LINEAR(db_scale_linear, TLV_DB_GAIN_MUTE, 0);
-static DECLARE_TLV_DB_SCALE(db_scale_ipga, 0, 50, 0);
 
 /*
  * initialize all the ak4xxx chips
@@ -190,8 +180,6 @@ void snd_akm4xxx_init(struct snd_akm4xxx *ak)
 		0x01, 0x03, /* 1: ADC/DAC enable */
 		0x04, 0x00, /* 4: ADC left muted */
 		0x05, 0x00, /* 5: ADC right muted */
-		0x04, 0x80, /* 4: ADC IPGA gain 0dB */
-		0x05, 0x80, /* 5: ADC IPGA gain 0dB */
 		0x06, 0x00, /* 6: DAC left muted */
 		0x07, 0x00, /* 7: DAC right muted */
 		0xff, 0xff
@@ -324,13 +312,15 @@ EXPORT_SYMBOL(snd_akm4xxx_init);
 /*
  * Mixer callbacks
  */
+#define AK_IPGA 			(1<<20)	/* including IPGA */
 #define AK_VOL_CVT 			(1<<21)	/* need dB conversion */
 #define AK_NEEDSMSB 			(1<<22)	/* need MSB update bit */
 #define AK_INVERT 			(1<<23)	/* data is inverted */
 #define AK_GET_CHIP(val)		(((val) >> 8) & 0xff)
 #define AK_GET_ADDR(val)		((val) & 0xff)
-#define AK_GET_SHIFT(val)		(((val) >> 16) & 0x1f)
+#define AK_GET_SHIFT(val)		(((val) >> 16) & 0x0f)
 #define AK_GET_VOL_CVT(val)		(((val) >> 21) & 1)
+#define AK_GET_IPGA(val)		(((val) >> 20) & 1)
 #define AK_GET_NEEDSMSB(val)		(((val) >> 22) & 1)
 #define AK_GET_INVERT(val)		(((val) >> 23) & 1)
 #define AK_GET_MASK(val)		(((val) >> 24) & 0xff)
@@ -371,8 +361,10 @@ static int put_ak_reg(struct snd_kcontrol *kcontrol, int addr,
 		return 0;
 
 	snd_akm4xxx_set_vol(ak, chip, addr, nval);
-	if (AK_GET_VOL_CVT(kcontrol->private_value))
+	if (AK_GET_VOL_CVT(kcontrol->private_value) && nval < 128)
 		nval = vol_cvt_datt[nval];
+	if (AK_GET_IPGA(kcontrol->private_value) && nval >= 128)
+		nval++; /* need to correct + 1 since both 127 and 128 are 0dB */
 	if (AK_GET_INVERT(kcontrol->private_value))
 		nval = mask - nval;
 	if (AK_GET_NEEDSMSB(kcontrol->private_value))
@@ -424,68 +416,6 @@ static int snd_akm4xxx_stereo_volume_put(struct snd_kcontrol *kcontrol,
 	return change;
 }
 
-#define snd_akm4xxx_ipga_gain_info	snd_akm4xxx_volume_info
-
-static int snd_akm4xxx_ipga_gain_get(struct snd_kcontrol *kcontrol,
-				     struct snd_ctl_elem_value *ucontrol)
-{
-	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
-	int chip = AK_GET_CHIP(kcontrol->private_value);
-	int addr = AK_GET_ADDR(kcontrol->private_value);
-
-	ucontrol->value.integer.value[0] =
-		snd_akm4xxx_get_ipga(ak, chip, addr);
-	return 0;
-}
-
-static int put_ak_ipga(struct snd_kcontrol *kcontrol, int addr,
-		       unsigned char nval)
-{
-	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
-	int chip = AK_GET_CHIP(kcontrol->private_value);
-
-	if (snd_akm4xxx_get_ipga(ak, chip, addr) == nval)
-		return 0;
-	snd_akm4xxx_set_ipga(ak, chip, addr, nval);
-	snd_akm4xxx_write(ak, chip, addr, nval | 0x80); /* need MSB */
-	return 1;
-}
-
-static int snd_akm4xxx_ipga_gain_put(struct snd_kcontrol *kcontrol,
-				     struct snd_ctl_elem_value *ucontrol)
-{
-	return put_ak_ipga(kcontrol, AK_GET_ADDR(kcontrol->private_value),
-			   ucontrol->value.integer.value[0]);
-}
-
-#define snd_akm4xxx_stereo_gain_info	snd_akm4xxx_stereo_volume_info
-
-static int snd_akm4xxx_stereo_gain_get(struct snd_kcontrol *kcontrol,
-				       struct snd_ctl_elem_value *ucontrol)
-{
-	struct snd_akm4xxx *ak = snd_kcontrol_chip(kcontrol);
-	int chip = AK_GET_CHIP(kcontrol->private_value);
-	int addr = AK_GET_ADDR(kcontrol->private_value);
-
-	ucontrol->value.integer.value[0] =
-		snd_akm4xxx_get_ipga(ak, chip, addr);
-	ucontrol->value.integer.value[1] =
-		snd_akm4xxx_get_ipga(ak, chip, addr + 1);
-	return 0;
-}
-
-static int snd_akm4xxx_stereo_gain_put(struct snd_kcontrol *kcontrol,
-				       struct snd_ctl_elem_value *ucontrol)
-{
-	int addr = AK_GET_ADDR(kcontrol->private_value);
-	int change;
-
-	change = put_ak_ipga(kcontrol, addr, ucontrol->value.integer.value[0]);
-	change |= put_ak_ipga(kcontrol, addr + 1,
-			      ucontrol->value.integer.value[1]);
-	return change;
-}
-
 static int snd_akm4xxx_deemphasis_info(struct snd_kcontrol *kcontrol,
 				       struct snd_ctl_elem_info *uinfo)
 {
@@ -702,35 +632,15 @@ static int build_adc_controls(struct snd_akm4xxx *ak)
 			knew.put = snd_akm4xxx_volume_put;
 		}
 		/* register 4 & 5 */
-		knew.private_value =
-			AK_COMPOSE(idx/2, (idx%2) + 4, 0, 127) |
-			AK_VOL_CVT;
-		knew.tlv.p = db_scale_vol_datt;
-		err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
-		if (err < 0)
-			return err;
-
-		if (! ak->adc_info || ! ak->adc_info[mixer_ch].gain_name)
-			knew.name = "IPGA Analog Capture Volume";
+		if (ak->type == SND_AK5365)
+			knew.private_value =
+				AK_COMPOSE(idx/2, (idx%2) + 4, 0, 151) |
+				AK_VOL_CVT | AK_IPGA;
 		else
-			knew.name = ak->adc_info[mixer_ch].gain_name;
-		if (num_stereo == 2) {
-			knew.info = snd_akm4xxx_stereo_gain_info;
-			knew.get = snd_akm4xxx_stereo_gain_get;
-			knew.put = snd_akm4xxx_stereo_gain_put;
-		} else {
-			knew.info = snd_akm4xxx_ipga_gain_info;
-			knew.get = snd_akm4xxx_ipga_gain_get;
-			knew.put = snd_akm4xxx_ipga_gain_put;
-		}
-		/* register 4 & 5 */
-		if (ak->type == SND_AK4524)
-			knew.private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0,
-							24);
-		else /* AK5365 */
-			knew.private_value = AK_COMPOSE(idx/2, (idx%2) + 4, 0,
-							36);
-		knew.tlv.p = db_scale_ipga;
+			knew.private_value =
+				AK_COMPOSE(idx/2, (idx%2) + 4, 0, 163) |
+				AK_VOL_CVT | AK_IPGA;
+		knew.tlv.p = db_scale_vol_datt;
 		err = snd_ctl_add(ak->card, snd_ctl_new1(&knew, ak));
 		if (err < 0)
 			return err;
@@ -811,11 +721,9 @@ int snd_akm4xxx_build_controls(struct snd_akm4xxx *ak)
 	if (err < 0)
 		return err;
 
-	if (ak->type == SND_AK4524 || ak->type == SND_AK5365) {
-		err = build_adc_controls(ak);
-		if (err < 0)
-			return err;
-	}
+	err = build_adc_controls(ak);
+	if (err < 0)
+		return err;
 
 	if (ak->type == SND_AK4355 || ak->type == SND_AK4358)
 		num_emphs = 1;
diff --git a/sound/pci/ice1712/revo.c b/sound/pci/ice1712/revo.c
index c9eefa9bbfff..bf98ea34feb0 100644
--- a/sound/pci/ice1712/revo.c
+++ b/sound/pci/ice1712/revo.c
@@ -110,7 +110,6 @@ static struct snd_akm4xxx_dac_channel revo51_dac[] = {
 static struct snd_akm4xxx_adc_channel revo51_adc[] = {
 	{
 		.name = "PCM Capture Volume",
-		.gain_name = "PCM Capture Gain Volume",
 		.switch_name = "PCM Capture Switch",
 		.num_channels = 2
 	},
-- 
cgit v1.2.3


From 4c8bd7eeee4c8f157fb61fb64b57500990b42e0e Mon Sep 17 00:00:00 2001
From: David Miller <davem@davemloft.net>
Date: Fri, 22 Sep 2006 22:31:36 -0700
Subject: [KERNEL] Do not truncate to 'int' in ALIGN() macro.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kernel.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 851aa1bcfc1a..2b2ae4fdce8b 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -31,7 +31,7 @@ extern const char linux_banner[];
 #define STACK_MAGIC	0xdeadbeef
 
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
-#define ALIGN(x,a) (((x)+(a)-1)&~((a)-1))
+#define ALIGN(x,a) (((x)+(a)-1UL)&~((a)-1UL))
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
 #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
 
-- 
cgit v1.2.3


From 2efc80cb8ddc341d81de996920e3b2ad8a12b1f7 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 23 Sep 2006 16:45:55 +0100
Subject: [PATCH] #elif that should've been #elif defined

 #elif CONFIG_44x
in ibm4xx.h should've been
 #elif defined(CONFIG_44x)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-ppc/ibm4xx.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-ppc/ibm4xx.h b/include/asm-ppc/ibm4xx.h
index cf62b69cb69a..499c14691c71 100644
--- a/include/asm-ppc/ibm4xx.h
+++ b/include/asm-ppc/ibm4xx.h
@@ -86,7 +86,7 @@ void ppc4xx_init(unsigned long r3, unsigned long r4, unsigned long r5,
 #define PCI_DRAM_OFFSET	0
 #endif
 
-#elif CONFIG_44x
+#elif defined(CONFIG_44x)
 
 #if defined(CONFIG_BAMBOO)
 #include <platforms/4xx/bamboo.h>
-- 
cgit v1.2.3


From dfdc58ba354adb80d67c99f7be84f95a8e02e466 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Wed, 20 Sep 2006 12:00:18 -0400
Subject: [SCSI] SPI transport class: misc DV fixes

Key more of the domain validation settings off the inquiry data from
the disk (in particular, don't try IU or DT unless the disk claims to
support them.

Also add a new dv_in_progress flag to prevent recursive DV.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_spi.c | 30 +++++++++++++++++++++++-------
 include/scsi/scsi_transport_spi.h |  3 ++-
 2 files changed, 25 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c
index 29a9a53cdd1a..9f070f0d0f2b 100644
--- a/drivers/scsi/scsi_transport_spi.c
+++ b/drivers/scsi/scsi_transport_spi.c
@@ -47,6 +47,7 @@
 
 /* Private data accessors (keep these out of the header file) */
 #define spi_dv_pending(x) (((struct spi_transport_attrs *)&(x)->starget_data)->dv_pending)
+#define spi_dv_in_progress(x) (((struct spi_transport_attrs *)&(x)->starget_data)->dv_in_progress)
 #define spi_dv_mutex(x) (((struct spi_transport_attrs *)&(x)->starget_data)->dv_mutex)
 
 struct spi_internal {
@@ -240,6 +241,7 @@ static int spi_setup_transport_attrs(struct transport_container *tc,
 	spi_pcomp_en(starget) = 0;
 	spi_hold_mcs(starget) = 0;
 	spi_dv_pending(starget) = 0;
+	spi_dv_in_progress(starget) = 0;
 	spi_initial_dv(starget) = 0;
 	mutex_init(&spi_dv_mutex(starget));
 
@@ -830,28 +832,37 @@ spi_dv_device_internal(struct scsi_device *sdev, u8 *buffer)
 	DV_SET(period, spi_min_period(starget));
 	/* try QAS requests; this should be harmless to set if the
 	 * target supports it */
-	if (scsi_device_qas(sdev))
+	if (scsi_device_qas(sdev)) {
 		DV_SET(qas, 1);
-	/* Also try IU transfers */
-	if (scsi_device_ius(sdev))
+	} else {
+		DV_SET(qas, 0);
+	}
+
+	if (scsi_device_ius(sdev) && spi_min_period(starget) < 9) {
+		/* This u320 (or u640). Set IU transfers */
 		DV_SET(iu, 1);
-	if (spi_min_period(starget) < 9) {
-		/* This u320 (or u640). Ignore the coupled parameters
-		 * like DT and IU, but set the optional ones */
+		/* Then set the optional parameters */
 		DV_SET(rd_strm, 1);
 		DV_SET(wr_flow, 1);
 		DV_SET(rti, 1);
 		if (spi_min_period(starget) == 8)
 			DV_SET(pcomp_en, 1);
+	} else {
+		DV_SET(iu, 0);
 	}
+
 	/* now that we've done all this, actually check the bus
 	 * signal type (if known).  Some devices are stupid on
 	 * a SE bus and still claim they can try LVD only settings */
 	if (i->f->get_signalling)
 		i->f->get_signalling(shost);
 	if (spi_signalling(shost) == SPI_SIGNAL_SE ||
-	    spi_signalling(shost) == SPI_SIGNAL_HVD)
+	    spi_signalling(shost) == SPI_SIGNAL_HVD ||
+	    !scsi_device_dt(sdev)) {
 		DV_SET(dt, 0);
+	} else {
+		DV_SET(dt, 1);
+	}
 	/* Do the read only INQUIRY tests */
 	spi_dv_retrain(sdev, buffer, buffer + sdev->inquiry_len,
 		       spi_dv_device_compare_inquiry);
@@ -907,6 +918,10 @@ spi_dv_device(struct scsi_device *sdev)
 	if (unlikely(scsi_device_get(sdev)))
 		return;
 
+	if (unlikely(spi_dv_in_progress(starget)))
+		return;
+	spi_dv_in_progress(starget) = 1;
+
 	buffer = kzalloc(len, GFP_KERNEL);
 
 	if (unlikely(!buffer))
@@ -938,6 +953,7 @@ spi_dv_device(struct scsi_device *sdev)
  out_free:
 	kfree(buffer);
  out_put:
+	spi_dv_in_progress(starget) = 0;
 	scsi_device_put(sdev);
 }
 EXPORT_SYMBOL(spi_dv_device);
diff --git a/include/scsi/scsi_transport_spi.h b/include/scsi/scsi_transport_spi.h
index 302680c0c0de..da180f738477 100644
--- a/include/scsi/scsi_transport_spi.h
+++ b/include/scsi/scsi_transport_spi.h
@@ -53,7 +53,8 @@ struct spi_transport_attrs {
 	unsigned int support_ius; /* support Information Units */
 	unsigned int support_qas; /* supports quick arbitration and selection */
 	/* Private Fields */
-	unsigned int dv_pending:1; /* Internal flag */
+	unsigned int dv_pending:1; /* Internal flag: DV Requested */
+	unsigned int dv_in_progress:1;	/* Internal: DV started */
 	struct mutex dv_mutex; /* semaphore to serialise dv */
 };
 
-- 
cgit v1.2.3


From 42431acbac43eb47c774c29d370f5c59136805bf Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 24 Sep 2006 10:44:09 +0100
Subject: [MMC] MMC_CAP_BYTEBLOCK flag for non-log2 block sizes capable hosts

Some MMC hosts can only handle log2 block sizes.  Unfortunately,
the MMC password support needs to be able to send non-log2 block
sizes.  Provide a capability so that the MMC password support can
decide whether it should use this support or not.

The unfortunate side effect of this host limitation is that any
MMC card protected by a password which is not a log2 block size
can not be accessed on a host which only allows a log2 block size.

This change just adds the flag.  The MMC password support code
needs updating to use it (if and when it is finally submitted.)

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mmc/at91_mci.c   | 1 +
 drivers/mmc/imxmmc.c     | 2 +-
 drivers/mmc/omap.c       | 7 ++++---
 drivers/mmc/sdhci.c      | 2 +-
 drivers/mmc/wbsd.c       | 2 +-
 include/linux/mmc/host.h | 1 +
 6 files changed, 9 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/drivers/mmc/at91_mci.c b/drivers/mmc/at91_mci.c
index 6b7638b84290..d34b7d9d92ed 100644
--- a/drivers/mmc/at91_mci.c
+++ b/drivers/mmc/at91_mci.c
@@ -822,6 +822,7 @@ static int at91_mci_probe(struct platform_device *pdev)
 	mmc->f_min = 375000;
 	mmc->f_max = 25000000;
 	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
+	mmc->caps = MMC_CAP_BYTEBLOCK;
 
 	host = mmc_priv(mmc);
 	host->mmc = mmc;
diff --git a/drivers/mmc/imxmmc.c b/drivers/mmc/imxmmc.c
index fb6565b98f32..1b79dd271aae 100644
--- a/drivers/mmc/imxmmc.c
+++ b/drivers/mmc/imxmmc.c
@@ -956,7 +956,7 @@ static int imxmci_probe(struct platform_device *pdev)
 	mmc->f_min = 150000;
 	mmc->f_max = CLK_RATE/2;
 	mmc->ocr_avail = MMC_VDD_32_33;
-	mmc->caps |= MMC_CAP_4_BIT_DATA;
+	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_BYTEBLOCK;
 
 	/* MMC core transfer sizes tunable parameters */
 	mmc->max_hw_segs = 64;
diff --git a/drivers/mmc/omap.c b/drivers/mmc/omap.c
index ddf06b32c159..52c9e52e6b78 100644
--- a/drivers/mmc/omap.c
+++ b/drivers/mmc/omap.c
@@ -1034,13 +1034,14 @@ static int __init mmc_omap_probe(struct platform_device *pdev)
 	host->irq = pdev->resource[1].start;
 	host->base = (void __iomem*)IO_ADDRESS(r->start);
 
-	if (minfo->wire4)
-		 mmc->caps |= MMC_CAP_4_BIT_DATA;
-
 	mmc->ops = &mmc_omap_ops;
 	mmc->f_min = 400000;
 	mmc->f_max = 24000000;
 	mmc->ocr_avail = MMC_VDD_32_33|MMC_VDD_33_34;
+	mmc->caps = MMC_CAP_BYTEBLOCK;
+
+	if (minfo->wire4)
+		 mmc->caps |= MMC_CAP_4_BIT_DATA;
 
 	/* Use scatterlist DMA to reduce per-transfer costs.
 	 * NOTE max_seg_size assumption that small blocks aren't
diff --git a/drivers/mmc/sdhci.c b/drivers/mmc/sdhci.c
index dea4edd1c434..fdfc3838dd79 100644
--- a/drivers/mmc/sdhci.c
+++ b/drivers/mmc/sdhci.c
@@ -1262,7 +1262,7 @@ static int __devinit sdhci_probe_slot(struct pci_dev *pdev, int slot)
 	mmc->ops = &sdhci_ops;
 	mmc->f_min = host->max_clk / 256;
 	mmc->f_max = host->max_clk;
-	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
+	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE | MMC_CAP_BYTEBLOCK;
 
 	mmc->ocr_avail = 0;
 	if (caps & SDHCI_CAN_VDD_330)
diff --git a/drivers/mmc/wbsd.c b/drivers/mmc/wbsd.c
index 4a6617d9a49f..6435a6822ad3 100644
--- a/drivers/mmc/wbsd.c
+++ b/drivers/mmc/wbsd.c
@@ -1323,7 +1323,7 @@ static int __devinit wbsd_alloc_mmc(struct device *dev)
 	mmc->f_min = 375000;
 	mmc->f_max = 24000000;
 	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
+	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE | MMC_CAP_BYTEBLOCK;
 
 	spin_lock_init(&host->lock);
 
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index b282ec9bba08..587264a58d56 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -86,6 +86,7 @@ struct mmc_host {
 
 #define MMC_CAP_4_BIT_DATA	(1 << 0)	/* Can the host do 4 bit transfers */
 #define MMC_CAP_MULTIWRITE	(1 << 1)	/* Can accurately report bytes sent to card on error */
+#define MMC_CAP_BYTEBLOCK	(1 << 2)	/* Can do non-log2 block sizes */
 
 	/* host specific block data */
 	unsigned int		max_seg_size;	/* see blk_queue_max_segment_size */
-- 
cgit v1.2.3


From e18fa700c9a31360bc8f193aa543b7ef7b39a06b Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Sun, 24 Sep 2006 11:13:19 -0400
Subject: Move several *_SUPER_MAGIC symbols to include/linux/magic.h.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 fs/affs/affs.h               |  1 -
 fs/affs/super.c              |  1 +
 fs/autofs/autofs_i.h         |  2 --
 fs/autofs/inode.c            |  1 +
 fs/autofs4/autofs_i.h        |  2 --
 fs/autofs4/inode.c           |  1 +
 fs/hpfs/hpfs_fn.h            |  1 -
 fs/hpfs/super.c              |  1 +
 fs/openpromfs/inode.c        |  2 +-
 include/linux/Kbuild         |  4 +---
 include/linux/adfs_fs.h      |  2 +-
 include/linux/affs_fs.h      |  7 -------
 include/linux/coda_psdev.h   |  4 ++--
 include/linux/efs_fs_sb.h    |  3 +--
 include/linux/ext2_fs.h      |  6 +-----
 include/linux/ext3_fs.h      |  6 +-----
 include/linux/hpfs_fs.h      |  8 --------
 include/linux/iso_fs.h       |  6 +++---
 include/linux/jffs2.h        |  4 ++--
 include/linux/magic.h        | 37 +++++++++++++++++++++++++++++++++++++
 include/linux/minix_fs.h     |  6 ++----
 include/linux/msdos_fs.h     |  4 ++--
 include/linux/ncp_fs.h       |  5 +----
 include/linux/nfs_fs.h       |  7 ++-----
 include/linux/openprom_fs.h  | 10 ----------
 include/linux/proc_fs.h      |  3 +--
 include/linux/qnx4_fs.h      |  2 +-
 include/linux/reiserfs_fs.h  | 10 ++--------
 include/linux/smb.h          |  3 +--
 include/linux/usbdevice_fs.h |  3 +--
 30 files changed, 67 insertions(+), 85 deletions(-)
 delete mode 100644 include/linux/affs_fs.h
 delete mode 100644 include/linux/hpfs_fs.h
 create mode 100644 include/linux/magic.h
 delete mode 100644 include/linux/openprom_fs.h

(limited to 'include')

diff --git a/fs/affs/affs.h b/fs/affs/affs.h
index 0ddd4cc0d1a0..1dc8438ef389 100644
--- a/fs/affs/affs.h
+++ b/fs/affs/affs.h
@@ -1,7 +1,6 @@
 #include <linux/types.h>
 #include <linux/fs.h>
 #include <linux/buffer_head.h>
-#include <linux/affs_fs.h>
 #include <linux/amigaffs.h>
 
 /* AmigaOS allows file names with up to 30 characters length.
diff --git a/fs/affs/super.c b/fs/affs/super.c
index 5200f4938df0..17352011ab67 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -14,6 +14,7 @@
 #include <linux/init.h>
 #include <linux/statfs.h>
 #include <linux/parser.h>
+#include <linux/magic.h>
 #include "affs.h"
 
 extern struct timezone sys_tz;
diff --git a/fs/autofs/autofs_i.h b/fs/autofs/autofs_i.h
index a62327f1bdff..c7700d9b3f96 100644
--- a/fs/autofs/autofs_i.h
+++ b/fs/autofs/autofs_i.h
@@ -37,8 +37,6 @@
 #define DPRINTK(D) ((void)0)
 #endif
 
-#define AUTOFS_SUPER_MAGIC 0x0187
-
 /*
  * If the daemon returns a negative response (AUTOFS_IOC_FAIL) then the
  * kernel will keep the negative response cached for up to the time given
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index 65e5ed42190e..af2efbbb5d76 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -16,6 +16,7 @@
 #include <linux/file.h>
 #include <linux/parser.h>
 #include <linux/bitops.h>
+#include <linux/magic.h>
 #include "autofs_i.h"
 #include <linux/module.h>
 
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index d6603d02304c..480ab178cba5 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -40,8 +40,6 @@
 #define DPRINTK(fmt,args...) do {} while(0)
 #endif
 
-#define AUTOFS_SUPER_MAGIC 0x0187
-
 /* Unified info structure.  This is pointed to by both the dentry and
    inode structures.  Each file in the filesystem has an instance of this
    structure.  It holds a reference to the dentry, so dentries are never
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index fde78b110ddd..11a6a9ae51b7 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -19,6 +19,7 @@
 #include <linux/parser.h>
 #include <linux/bitops.h>
 #include <linux/smp_lock.h>
+#include <linux/magic.h>
 #include "autofs_i.h"
 #include <linux/module.h>
 
diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h
index f687d54ed442..32ab51e42b96 100644
--- a/fs/hpfs/hpfs_fn.h
+++ b/fs/hpfs/hpfs_fn.h
@@ -12,7 +12,6 @@
 #include <linux/mutex.h>
 #include <linux/pagemap.h>
 #include <linux/buffer_head.h>
-#include <linux/hpfs_fs.h>
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
 
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index f798480a363f..8fe51c343786 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -11,6 +11,7 @@
 #include <linux/parser.h>
 #include <linux/init.h>
 #include <linux/statfs.h>
+#include <linux/magic.h>
 
 /* Mark the filesystem dirty, so that chkdsk checks it when os/2 booted */
 
diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c
index 93a56bd4a2b7..592a6402e851 100644
--- a/fs/openpromfs/inode.c
+++ b/fs/openpromfs/inode.c
@@ -8,10 +8,10 @@
 #include <linux/types.h>
 #include <linux/string.h>
 #include <linux/fs.h>
-#include <linux/openprom_fs.h>
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/seq_file.h>
+#include <linux/magic.h>
 
 #include <asm/openprom.h>
 #include <asm/oplib.h>
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 7d076d97b2f7..67383605f2e5 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -12,7 +12,6 @@ header-y += netfilter_bridge/
 header-y += netfilter_ipv4/
 header-y += netfilter_ipv6/
 
-header-y += affs_fs.h
 header-y += affs_hardblocks.h
 header-y += aio_abi.h
 header-y += a.out.h
@@ -67,7 +66,6 @@ header-y += genetlink.h
 header-y += gen_stats.h
 header-y += gigaset_dev.h
 header-y += hdsmart.h
-header-y += hpfs_fs.h
 header-y += hysdn_if.h
 header-y += i2c-dev.h
 header-y += i8k.h
@@ -103,6 +101,7 @@ header-y += ixjuser.h
 header-y += jffs2.h
 header-y += keyctl.h
 header-y += limits.h
+header-y += magic.h
 header-y += major.h
 header-y += matroxfb.h
 header-y += meye.h
@@ -116,7 +115,6 @@ header-y += netrom.h
 header-y += nfs2.h
 header-y += nfs4_mount.h
 header-y += nfs_mount.h
-header-y += openprom_fs.h
 header-y += param.h
 header-y += pci_ids.h
 header-y += pci_regs.h
diff --git a/include/linux/adfs_fs.h b/include/linux/adfs_fs.h
index 4a5d50c2bdbf..ef788c2085a1 100644
--- a/include/linux/adfs_fs.h
+++ b/include/linux/adfs_fs.h
@@ -2,6 +2,7 @@
 #define _ADFS_FS_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
 
 /*
  * Disc Record at disc address 0xc00
@@ -38,7 +39,6 @@ struct adfs_discrecord {
 #define ADFS_DR_OFFSET		(0x1c0)
 #define ADFS_DR_SIZE		 60
 #define ADFS_DR_SIZE_BITS	(ADFS_DR_SIZE << 3)
-#define ADFS_SUPER_MAGIC	 0xadf5
 
 #ifdef __KERNEL__
 #include <linux/adfs_fs_i.h>
diff --git a/include/linux/affs_fs.h b/include/linux/affs_fs.h
deleted file mode 100644
index c57b5ee87d55..000000000000
--- a/include/linux/affs_fs.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef _AFFS_FS_H
-#define _AFFS_FS_H
-/*
- * The affs filesystem constants/structures
- */
-#define AFFS_SUPER_MAGIC 0xadff
-#endif
diff --git a/include/linux/coda_psdev.h b/include/linux/coda_psdev.h
index 98f6c52c152b..b541bb3d1f4b 100644
--- a/include/linux/coda_psdev.h
+++ b/include/linux/coda_psdev.h
@@ -1,11 +1,11 @@
 #ifndef __CODA_PSDEV_H
 #define __CODA_PSDEV_H
 
+#include <linux/magic.h>
+
 #define CODA_PSDEV_MAJOR 67
 #define MAX_CODADEVS  5	   /* how many do we allow */
 
-#define CODA_SUPER_MAGIC	0x73757245
-
 struct kstatfs;
 
 struct coda_sb_info
diff --git a/include/linux/efs_fs_sb.h b/include/linux/efs_fs_sb.h
index c76088baef28..ff1945e37790 100644
--- a/include/linux/efs_fs_sb.h
+++ b/include/linux/efs_fs_sb.h
@@ -9,8 +9,7 @@
 #ifndef __EFS_FS_SB_H__
 #define __EFS_FS_SB_H__
 
-/* statfs() magic number for EFS */
-#define EFS_SUPER_MAGIC	0x414A53
+#include <linux/magic.h>
 
 /* EFS superblock magic numbers */
 #define EFS_MAGIC	0x072959
diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index facf34e98954..33a1aa107329 100644
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -17,6 +17,7 @@
 #define _LINUX_EXT2_FS_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
 
 /*
  * The second extended filesystem constants/structures
@@ -63,11 +64,6 @@
 /* First non-reserved inode for old ext2 filesystems */
 #define EXT2_GOOD_OLD_FIRST_INO	11
 
-/*
- * The second extended file system magic number
- */
-#define EXT2_SUPER_MAGIC	0xEF53
-
 #ifdef __KERNEL__
 #include <linux/ext2_fs_sb.h>
 static inline struct ext2_sb_info *EXT2_SB(struct super_block *sb)
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 9f9cce7bd86d..0eed918b3816 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -17,6 +17,7 @@
 #define _LINUX_EXT3_FS_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
 
 /*
  * The second extended filesystem constants/structures
@@ -66,11 +67,6 @@
 /* First non-reserved inode for old ext3 filesystems */
 #define EXT3_GOOD_OLD_FIRST_INO	11
 
-/*
- * The second extended file system magic number
- */
-#define EXT3_SUPER_MAGIC	0xEF53
-
 /*
  * Maximal count of links to a file
  */
diff --git a/include/linux/hpfs_fs.h b/include/linux/hpfs_fs.h
deleted file mode 100644
index a5028dd94d31..000000000000
--- a/include/linux/hpfs_fs.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef _LINUX_HPFS_FS_H
-#define _LINUX_HPFS_FS_H
-
-/* HPFS magic number (word 0 of block 16) */
-
-#define HPFS_SUPER_MAGIC 0xf995e849
-
-#endif
diff --git a/include/linux/iso_fs.h b/include/linux/iso_fs.h
index 47967878bfef..4688ac4284e2 100644
--- a/include/linux/iso_fs.h
+++ b/include/linux/iso_fs.h
@@ -2,6 +2,8 @@
 #define _ISOFS_FS_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
+
 /*
  * The isofs filesystem constants/structures
  */
@@ -160,6 +162,4 @@ struct iso_directory_record {
 #define ISOFS_BUFFER_SIZE(INODE) ((INODE)->i_sb->s_blocksize)
 #define ISOFS_BUFFER_BITS(INODE) ((INODE)->i_sb->s_blocksize_bits)
 
-#define ISOFS_SUPER_MAGIC 0x9660
-
-#endif
+#endif /* _ISOFS_FS_H */
diff --git a/include/linux/jffs2.h b/include/linux/jffs2.h
index c9c760700bc3..840631fa5ff1 100644
--- a/include/linux/jffs2.h
+++ b/include/linux/jffs2.h
@@ -15,12 +15,12 @@
 #ifndef __LINUX_JFFS2_H__
 #define __LINUX_JFFS2_H__
 
+#include <linux/magic.h>
+
 /* You must include something which defines the C99 uintXX_t types. 
    We don't do it from here because this file is used in too many
    different environments. */
 
-#define JFFS2_SUPER_MAGIC 0x72b6
-
 /* Values we may expect to find in the 'magic' field */
 #define JFFS2_OLD_MAGIC_BITMASK 0x1984
 #define JFFS2_MAGIC_BITMASK 0x1985
diff --git a/include/linux/magic.h b/include/linux/magic.h
new file mode 100644
index 000000000000..22036dd2ba36
--- /dev/null
+++ b/include/linux/magic.h
@@ -0,0 +1,37 @@
+#ifndef __LINUX_MAGIC_H__
+#define __LINUX_MAGIC_H__
+
+#define ADFS_SUPER_MAGIC	0xadf5
+#define AFFS_SUPER_MAGIC	0xadff
+#define AUTOFS_SUPER_MAGIC	0x0187
+#define CODA_SUPER_MAGIC	0x73757245
+#define EFS_SUPER_MAGIC		0x414A53
+#define EXT2_SUPER_MAGIC	0xEF53
+#define EXT3_SUPER_MAGIC	0xEF53
+#define HPFS_SUPER_MAGIC	0xf995e849
+#define ISOFS_SUPER_MAGIC	0x9660
+#define JFFS2_SUPER_MAGIC	0x72b6
+
+#define MINIX_SUPER_MAGIC	0x137F		/* original minix fs */
+#define MINIX_SUPER_MAGIC2	0x138F		/* minix fs, 30 char names */
+#define MINIX2_SUPER_MAGIC	0x2468		/* minix V2 fs */
+#define MINIX2_SUPER_MAGIC2	0x2478		/* minix V2 fs, 30 char names */
+
+#define MSDOS_SUPER_MAGIC	0x4d44		/* MD */
+#define NCP_SUPER_MAGIC		0x564c		/* Guess, what 0x564c is :-) */
+#define NFS_SUPER_MAGIC		0x6969
+#define OPENPROM_SUPER_MAGIC	0x9fa1
+#define PROC_SUPER_MAGIC	0x9fa0
+#define QNX4_SUPER_MAGIC	0x002f		/* qnx4 fs detection */
+
+#define REISERFS_SUPER_MAGIC	0x52654973	/* used by gcc */
+					/* used by file system utilities that
+	                                   look at the superblock, etc.  */
+#define REISERFS_SUPER_MAGIC_STRING	"ReIsErFs"
+#define REISER2FS_SUPER_MAGIC_STRING	"ReIsEr2Fs"
+#define REISER2FS_JR_SUPER_MAGIC_STRING	"ReIsEr3Fs"
+
+#define SMB_SUPER_MAGIC		0x517B
+#define USBDEVICE_SUPER_MAGIC	0x9fa2
+
+#endif /* __LINUX_MAGIC_H__ */
diff --git a/include/linux/minix_fs.h b/include/linux/minix_fs.h
index 1ecc3cc8cef5..916e8f72c63d 100644
--- a/include/linux/minix_fs.h
+++ b/include/linux/minix_fs.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_MINIX_FS_H
 #define _LINUX_MINIX_FS_H
 
+#include <linux/magic.h>
+
 /*
  * The minix filesystem constants/structures
  */
@@ -19,10 +21,6 @@
 
 #define MINIX_I_MAP_SLOTS	8
 #define MINIX_Z_MAP_SLOTS	64
-#define MINIX_SUPER_MAGIC	0x137F		/* original minix fs */
-#define MINIX_SUPER_MAGIC2	0x138F		/* minix fs, 30 char names */
-#define MINIX2_SUPER_MAGIC	0x2468		/* minix V2 fs */
-#define MINIX2_SUPER_MAGIC2	0x2478		/* minix V2 fs, 30 char names */
 #define MINIX_VALID_FS		0x0001		/* Clean fs. */
 #define MINIX_ERROR_FS		0x0002		/* fs has errors. */
 
diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index d9035c73e5d1..bae62d62dc3e 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_MSDOS_FS_H
 #define _LINUX_MSDOS_FS_H
 
+#include <linux/magic.h>
+
 /*
  * The MS-DOS filesystem constants/structures
  */
@@ -18,8 +20,6 @@
 #define CT_LE_L(v)	cpu_to_le32(v)
 
 
-#define MSDOS_SUPER_MAGIC 0x4d44 /* MD */
-
 #define MSDOS_ROOT_INO	1	/* == MINIX_ROOT_INO */
 #define MSDOS_DIR_BITS	5	/* log2(sizeof(struct msdos_dir_entry)) */
 
diff --git a/include/linux/ncp_fs.h b/include/linux/ncp_fs.h
index b208f0cd556b..02e352be717e 100644
--- a/include/linux/ncp_fs.h
+++ b/include/linux/ncp_fs.h
@@ -11,6 +11,7 @@
 #include <linux/fs.h>
 #include <linux/in.h>
 #include <linux/types.h>
+#include <linux/magic.h>
 
 #include <linux/ipx.h>
 #include <linux/ncp_no.h>
@@ -185,10 +186,6 @@ struct ncp_entry_info {
 	__u8			file_handle[6];
 };
 
-/* Guess, what 0x564c is :-) */
-#define NCP_SUPER_MAGIC  0x564c
-
-
 static inline struct ncp_server *NCP_SBP(struct super_block *sb)
 {
 	return sb->s_fs_info;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 3b5b04193fee..36f5bcf513b0 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -9,6 +9,8 @@
 #ifndef _LINUX_NFS_FS_H
 #define _LINUX_NFS_FS_H
 
+#include <linux/magic.h>
+
 /*
  * Enable debugging support for nfs client.
  * Requires RPC_DEBUG.
@@ -21,11 +23,6 @@
 #define NFS_MAX_UDP_TIMEOUT	(60*HZ)
 #define NFS_MAX_TCP_TIMEOUT	(600*HZ)
 
-/*
- * superblock magic number for NFS
- */
-#define NFS_SUPER_MAGIC			0x6969
-
 /*
  * When flushing a cluster of dirty pages, there can be different
  * strategies:
diff --git a/include/linux/openprom_fs.h b/include/linux/openprom_fs.h
deleted file mode 100644
index a837aab8217e..000000000000
--- a/include/linux/openprom_fs.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef _LINUX_OPENPROM_FS_H
-#define _LINUX_OPENPROM_FS_H
-
-/*
- * The openprom filesystem constants/structures
- */
-
-#define OPENPROM_SUPER_MAGIC 0x9fa1
-
-#endif /* _LINUX_OPENPROM_FS_H */
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index 17e75783e3a5..3435ca38dd14 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -4,6 +4,7 @@
 #include <linux/slab.h>
 #include <linux/fs.h>
 #include <linux/spinlock.h>
+#include <linux/magic.h>
 #include <asm/atomic.h>
 
 /*
@@ -24,8 +25,6 @@ enum {
 	PROC_ROOT_INO = 1,
 };
 
-#define PROC_SUPER_MAGIC 0x9fa0
-
 /*
  * This is not completely implemented yet. The idea is to
  * create an in-memory tree (like the actual /proc filesystem
diff --git a/include/linux/qnx4_fs.h b/include/linux/qnx4_fs.h
index 27f49c85d5d6..0c7ac444fd35 100644
--- a/include/linux/qnx4_fs.h
+++ b/include/linux/qnx4_fs.h
@@ -11,6 +11,7 @@
 #define _LINUX_QNX4_FS_H
 
 #include <linux/qnxtypes.h>
+#include <linux/magic.h>
 
 #define QNX4_ROOT_INO 1
 
@@ -25,7 +26,6 @@
 
 #define QNX4_I_MAP_SLOTS	8
 #define QNX4_Z_MAP_SLOTS	64
-#define QNX4_SUPER_MAGIC	0x002f	/* qnx4 fs detection */
 #define QNX4_VALID_FS		0x0001	/* Clean fs. */
 #define QNX4_ERROR_FS		0x0002	/* fs has errors. */
 #define QNX4_BLOCK_SIZE         0x200	/* blocksize of 512 bytes */
diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h
index daa2d83cefe8..28493ffaafe7 100644
--- a/include/linux/reiserfs_fs.h
+++ b/include/linux/reiserfs_fs.h
@@ -12,6 +12,8 @@
 #define _LINUX_REISER_FS_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
+
 #ifdef __KERNEL__
 #include <linux/slab.h>
 #include <linux/interrupt.h>
@@ -227,14 +229,6 @@ struct reiserfs_super_block {
          ((!is_reiserfs_jr(SB_DISK_SUPER_BLOCK(s)) ? \
          SB_ONDISK_JOURNAL_SIZE(s) + 1 : SB_ONDISK_RESERVED_FOR_JOURNAL(s)))
 
-				/* used by gcc */
-#define REISERFS_SUPER_MAGIC 0x52654973
-				/* used by file system utilities that
-				   look at the superblock, etc. */
-#define REISERFS_SUPER_MAGIC_STRING "ReIsErFs"
-#define REISER2FS_SUPER_MAGIC_STRING "ReIsEr2Fs"
-#define REISER2FS_JR_SUPER_MAGIC_STRING "ReIsEr3Fs"
-
 int is_reiserfs_3_5(struct reiserfs_super_block *rs);
 int is_reiserfs_3_6(struct reiserfs_super_block *rs);
 int is_reiserfs_jr(struct reiserfs_super_block *rs);
diff --git a/include/linux/smb.h b/include/linux/smb.h
index b0162208c963..6df3b1501559 100644
--- a/include/linux/smb.h
+++ b/include/linux/smb.h
@@ -10,6 +10,7 @@
 #define _LINUX_SMB_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
 
 enum smb_protocol { 
 	SMB_PROTOCOL_NONE, 
@@ -101,8 +102,6 @@ enum smb_conn_state {
 	CONN_RETRYING		/* Currently trying to reconnect */
 };
 
-#define SMB_SUPER_MAGIC               0x517B
-
 #define SMB_HEADER_LEN   37     /* includes everything up to, but not
                                  * including smb_bcc */
 
diff --git a/include/linux/usbdevice_fs.h b/include/linux/usbdevice_fs.h
index 7b7aadb69092..617d8a1c59ae 100644
--- a/include/linux/usbdevice_fs.h
+++ b/include/linux/usbdevice_fs.h
@@ -32,11 +32,10 @@
 #define _LINUX_USBDEVICE_FS_H
 
 #include <linux/types.h>
+#include <linux/magic.h>
 
 /* --------------------------------------------------------------------- */
 
-#define USBDEVICE_SUPER_MAGIC 0x9fa2
-
 /* usbdevfs ioctl codes */
 
 struct usbdevfs_ctrltransfer {
-- 
cgit v1.2.3


From 00e4d116a7ef94eb910be037912b0b2fc09f608b Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Fri, 22 Sep 2006 09:33:58 +0100
Subject: [DCCP]: Allow default/fallback service code.

This has been discussed on dccp@vger and removes the necessity for applications
to supply service codes in each and every case.

If an application does not want to provide a service code, that's fine, it will
be given 0. Otherwise, service codes can be set via socket options as before.

This patch has been tested using various client/server configurations
(including listening on multiple service codes).

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
---
 Documentation/networking/dccp.txt |  8 +++++---
 include/linux/dccp.h              |  6 +-----
 net/dccp/ipv4.c                   |  3 ---
 net/dccp/proto.c                  | 11 +----------
 4 files changed, 7 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt
index c45daabd3bfe..74563b38ffd9 100644
--- a/Documentation/networking/dccp.txt
+++ b/Documentation/networking/dccp.txt
@@ -1,7 +1,6 @@
 DCCP protocol
 ============
 
-Last updated: 10 November 2005
 
 Contents
 ========
@@ -42,8 +41,11 @@ Socket options
 DCCP_SOCKOPT_PACKET_SIZE is used for CCID3 to set default packet size for
 calculations.
 
-DCCP_SOCKOPT_SERVICE sets the service. This is compulsory as per the
-specification. If you don't set it you will get EPROTO.
+DCCP_SOCKOPT_SERVICE sets the service. The specification mandates use of
+service codes (RFC 4340, sec. 8.1.2); if this socket option is not set,
+the socket will fall back to 0 (which means that no meaningful service code
+is present). Connecting sockets set at most one service option; for
+listening sockets, multiple service codes can be specified.
 
 Notes
 =====
diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 2d7671c92c0b..29f9291e45c0 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -404,6 +404,7 @@ struct dccp_service_list {
 };
 
 #define DCCP_SERVICE_INVALID_VALUE htonl((__u32)-1)
+#define DCCP_SERVICE_CODE_IS_ABSENT 		 0
 
 static inline int dccp_list_has_service(const struct dccp_service_list *sl,
 					const __be32 service)
@@ -484,11 +485,6 @@ static inline struct dccp_minisock *dccp_msk(const struct sock *sk)
 	return (struct dccp_minisock *)&dccp_sk(sk)->dccps_minisock;
 }
 
-static inline int dccp_service_not_initialized(const struct sock *sk)
-{
-	return dccp_sk(sk)->dccps_service == DCCP_SERVICE_INVALID_VALUE;
-}
-
 static inline const char *dccp_role(const struct sock *sk)
 {
 	switch (dccp_sk(sk)->dccps_role) {
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 9a1a76a7dc41..66be29b6f508 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -56,9 +56,6 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 
 	dp->dccps_role = DCCP_ROLE_CLIENT;
 
-	if (dccp_service_not_initialized(sk))
-		return -EPROTO;
-
 	if (addr_len < sizeof(struct sockaddr_in))
 		return -EINVAL;
 
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 962df0ea31aa..72cbdcfc2c65 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -217,7 +217,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized)
 	icsk->icsk_sync_mss	= dccp_sync_mss;
 	dp->dccps_mss_cache	= 536;
 	dp->dccps_role		= DCCP_ROLE_UNDEFINED;
-	dp->dccps_service	= DCCP_SERVICE_INVALID_VALUE;
+	dp->dccps_service	= DCCP_SERVICE_CODE_IS_ABSENT;
 	dp->dccps_l_ack_ratio	= dp->dccps_r_ack_ratio = 1;
 
 	return 0;
@@ -267,12 +267,6 @@ static inline int dccp_listen_start(struct sock *sk)
 	struct dccp_sock *dp = dccp_sk(sk);
 
 	dp->dccps_role = DCCP_ROLE_LISTEN;
-	/*
-	 * Apps need to use setsockopt(DCCP_SOCKOPT_SERVICE)
-	 * before calling listen()
-	 */
-	if (dccp_service_not_initialized(sk))
-		return -EPROTO;
 	return inet_csk_listen_start(sk, TCP_SYNQ_HSIZE);
 }
 
@@ -540,9 +534,6 @@ static int dccp_getsockopt_service(struct sock *sk, int len,
 	int err = -ENOENT, slen = 0, total_len = sizeof(u32);
 
 	lock_sock(sk);
-	if (dccp_service_not_initialized(sk))
-		goto out;
-
 	if ((sl = dp->dccps_service_list) != NULL) {
 		slen = sl->dccpsl_nr * sizeof(u32);
 		total_len += slen;
-- 
cgit v1.2.3


From 349457ccf2592c14bdf13b6706170ae2e94931b1 Mon Sep 17 00:00:00 2001
From: Mark Fasheh <mark.fasheh@oracle.com>
Date: Fri, 8 Sep 2006 14:22:21 -0700
Subject: [PATCH] Allow file systems to manually d_move() inside of ->rename()

Some file systems want to manually d_move() the dentries involved in a
rename.  We can do this by making use of the FS_ODD_RENAME flag if we just
have nfs_rename() unconditionally do the d_move().  While there, we rename
the flag to be more descriptive.

OCFS2 uses this to protect that part of the rename operation with a cluster
lock.

Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 fs/namei.c         |  6 +++---
 fs/nfs/dir.c       |  3 +--
 fs/nfs/super.c     | 10 +++++-----
 include/linux/fs.h |  7 ++++---
 4 files changed, 13 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index 432d6bc6fab0..6b591c01b09f 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2370,7 +2370,8 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry,
 		dput(new_dentry);
 	}
 	if (!error)
-		d_move(old_dentry,new_dentry);
+		if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE))
+			d_move(old_dentry,new_dentry);
 	return error;
 }
 
@@ -2393,8 +2394,7 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry,
 	else
 		error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry);
 	if (!error) {
-		/* The following d_move() should become unconditional */
-		if (!(old_dir->i_sb->s_type->fs_flags & FS_ODD_RENAME))
+		if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE))
 			d_move(old_dentry, new_dentry);
 	}
 	if (target)
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 3419c2da9ba9..7432f1a43f3d 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1669,8 +1669,7 @@ out:
 	if (rehash)
 		d_rehash(rehash);
 	if (!error) {
-		if (!S_ISDIR(old_inode->i_mode))
-			d_move(old_dentry, new_dentry);
+		d_move(old_dentry, new_dentry);
 		nfs_renew_times(new_dentry);
 		nfs_set_verifier(new_dentry, nfs_save_change_attribute(new_dir));
 	}
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index b99113b0f65f..e8d40030cab4 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -71,7 +71,7 @@ static struct file_system_type nfs_fs_type = {
 	.name		= "nfs",
 	.get_sb		= nfs_get_sb,
 	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
 struct file_system_type nfs_xdev_fs_type = {
@@ -79,7 +79,7 @@ struct file_system_type nfs_xdev_fs_type = {
 	.name		= "nfs",
 	.get_sb		= nfs_xdev_get_sb,
 	.kill_sb	= nfs_kill_super,
-	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
 static struct super_operations nfs_sops = {
@@ -107,7 +107,7 @@ static struct file_system_type nfs4_fs_type = {
 	.name		= "nfs4",
 	.get_sb		= nfs4_get_sb,
 	.kill_sb	= nfs4_kill_super,
-	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
 struct file_system_type nfs4_xdev_fs_type = {
@@ -115,7 +115,7 @@ struct file_system_type nfs4_xdev_fs_type = {
 	.name		= "nfs4",
 	.get_sb		= nfs4_xdev_get_sb,
 	.kill_sb	= nfs4_kill_super,
-	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
 struct file_system_type nfs4_referral_fs_type = {
@@ -123,7 +123,7 @@ struct file_system_type nfs4_referral_fs_type = {
 	.name		= "nfs4",
 	.get_sb		= nfs4_referral_get_sb,
 	.kill_sb	= nfs4_kill_super,
-	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
 static struct super_operations nfs4_sops = {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 555bc195c420..1d3e601ece73 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -92,9 +92,10 @@ extern int dir_notify_enable;
 #define FS_REQUIRES_DEV 1 
 #define FS_BINARY_MOUNTDATA 2
 #define FS_REVAL_DOT	16384	/* Check the paths ".", ".." for staleness */
-#define FS_ODD_RENAME	32768	/* Temporary stuff; will go away as soon
-				  * as nfs_rename() will be cleaned up
-				  */
+#define FS_RENAME_DOES_D_MOVE	32768	/* FS will handle d_move()
+					 * during rename() internally.
+					 */
+
 /*
  * These are the fs-independent mount-flags: up to 32 flags are supported
  */
-- 
cgit v1.2.3


From b83eff641ed39bd631535b9a8971e161b056f541 Mon Sep 17 00:00:00 2001
From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Fri, 22 Sep 2006 14:25:36 +1200
Subject: [DCCP]: Introduce constants for CCID numbers

Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
---
 include/linux/dccp.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 29f9291e45c0..d6f4ec467a4b 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -169,6 +169,12 @@ enum {
 	DCCPO_MAX_CCID_SPECIFIC = 255,
 };
 
+/* DCCP CCIDS */
+enum {
+	DCCPC_CCID2 = 2,
+	DCCPC_CCID3 = 3,
+};
+
 /* DCCP features */
 enum {
 	DCCPF_RESERVED = 0,
@@ -320,7 +326,7 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb)
 /* initial values for each feature */
 #define DCCPF_INITIAL_SEQUENCE_WINDOW		100
 #define DCCPF_INITIAL_ACK_RATIO			2
-#define DCCPF_INITIAL_CCID			2
+#define DCCPF_INITIAL_CCID			DCCPC_CCID2
 #define DCCPF_INITIAL_SEND_ACK_VECTOR		1
 /* FIXME: for now we're default to 1 but it should really be 0 */
 #define DCCPF_INITIAL_SEND_NDP_COUNT		1
-- 
cgit v1.2.3


From 9e72cbf353e259bd30ab472d72d7bdb9be23fb12 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Sun, 24 Sep 2006 22:06:48 +0100
Subject: Remove dead netfilter_logging.h from include/linux/Kbuild

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/linux/Kbuild | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 19a3664f1dfe..1df2ac30a4d2 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -266,7 +266,6 @@ unifdef-y += netfilter_decnet.h
 unifdef-y += netfilter.h
 unifdef-y += netfilter_ipv4.h
 unifdef-y += netfilter_ipv6.h
-unifdef-y += netfilter_logging.h
 unifdef-y += net.h
 unifdef-y += netlink.h
 unifdef-y += nfs3.h
-- 
cgit v1.2.3


From b4daf69722c49670d355d66439abda5ab5d4c5db Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Sun, 24 Sep 2006 22:07:25 +0100
Subject: [S390] Unexport <asm/z90crypt.h>, export <asm/zcrypt.h> in its place.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/asm-s390/Kbuild | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-s390/Kbuild b/include/asm-s390/Kbuild
index 088969d55e72..e92b429d2be1 100644
--- a/include/asm-s390/Kbuild
+++ b/include/asm-s390/Kbuild
@@ -6,7 +6,7 @@ header-y += qeth.h
 header-y += tape390.h
 header-y += ucontext.h
 header-y += vtoc.h
-header-y += z90crypt.h
+header-y += zcrypt.h
 
 unifdef-y += cmb.h
 unifdef-y += debug.h
-- 
cgit v1.2.3


From 638b093255e12e066ae16b8750f37383603f7bd8 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sun, 24 Sep 2006 23:39:25 +0100
Subject: [PATCH] restore libata build on frv

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/libata-portmap.h | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 include/asm-frv/libata-portmap.h

(limited to 'include')

diff --git a/include/asm-frv/libata-portmap.h b/include/asm-frv/libata-portmap.h
new file mode 100644
index 000000000000..75484ef0c743
--- /dev/null
+++ b/include/asm-frv/libata-portmap.h
@@ -0,0 +1 @@
+#include <asm-generic/libata-portmap.h>
-- 
cgit v1.2.3


From 1db2ea398ffbfd9ea46d509ff0e4a85bb4b8c0ea Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sun, 24 Sep 2006 23:41:42 +0100
Subject: [PATCH] netlabel gfp annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/net/netlabel.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index fc2b72fc7e07..dd5780b36919 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -108,7 +108,7 @@
  */
 static inline struct sk_buff *netlbl_netlink_alloc_skb(size_t head,
 						       size_t body,
-						       int gfp_flags)
+						       gfp_t gfp_flags)
 {
 	struct sk_buff *skb;
 
-- 
cgit v1.2.3


From 3e597c6045502dd0fa98a61aa95ba178f8a2cc03 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sun, 24 Sep 2006 23:42:20 +0100
Subject: [PATCH] fix iptables __user misannotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/netfilter/x_tables.h | 4 ++--
 net/ipv4/netfilter/ip_tables.c     | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 739a98eebe2c..04319a76103a 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -390,13 +390,13 @@ extern int xt_compat_match_offset(struct xt_match *match);
 extern void xt_compat_match_from_user(struct xt_entry_match *m,
 				      void **dstptr, int *size);
 extern int xt_compat_match_to_user(struct xt_entry_match *m,
-				   void * __user *dstptr, int *size);
+				   void __user **dstptr, int *size);
 
 extern int xt_compat_target_offset(struct xt_target *target);
 extern void xt_compat_target_from_user(struct xt_entry_target *t,
 				       void **dstptr, int *size);
 extern int xt_compat_target_to_user(struct xt_entry_target *t,
-				    void * __user *dstptr, int *size);
+				    void __user **dstptr, int *size);
 
 #endif /* CONFIG_COMPAT */
 #endif /* __KERNEL__ */
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 800067d69a9a..78a44b01c035 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1364,15 +1364,15 @@ struct compat_ipt_replace {
 };
 
 static inline int compat_copy_match_to_user(struct ipt_entry_match *m,
-		void * __user *dstptr, compat_uint_t *size)
+		void __user **dstptr, compat_uint_t *size)
 {
 	return xt_compat_match_to_user(m, dstptr, size);
 }
 
 static int compat_copy_entry_to_user(struct ipt_entry *e,
-		void * __user *dstptr, compat_uint_t *size)
+		void __user **dstptr, compat_uint_t *size)
 {
-	struct ipt_entry_target __user *t;
+	struct ipt_entry_target *t;
 	struct compat_ipt_entry __user *ce;
 	u_int16_t target_offset, next_offset;
 	compat_uint_t origsize;
-- 
cgit v1.2.3


From 0d6c7ae22d4fadbc83677ea1a6144eea8911e319 Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Sun, 24 Sep 2006 19:28:47 -0700
Subject: [NETFILTER]: Add dscp,DSCP headers to header-y

This patch adds xt_dscp.h and xt_DSCP.h to the kernel headers which are
exported via 'make headers_install'. These are necessary for userspace
to add rules using dscp match and DSCP target.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/Kbuild | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index 9a285cecf249..312bd2ffee33 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -10,6 +10,8 @@ header-y += xt_connmark.h
 header-y += xt_CONNMARK.h
 header-y += xt_conntrack.h
 header-y += xt_dccp.h
+header-y += xt_dscp.h
+header-y += xt_DSCP.h
 header-y += xt_esp.h
 header-y += xt_helper.h
 header-y += xt_length.h
-- 
cgit v1.2.3


From 3cc27547d6ee2d50ecdd11e9127bc3cd1947e8dd Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Mon, 25 Sep 2006 02:55:40 +0100
Subject: [PATCH] SCSI gfp_t annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/scsi/aic94xx/aic94xx.h      |  2 +-
 drivers/scsi/aic94xx/aic94xx_hwi.c  |  8 ++++----
 drivers/scsi/aic94xx/aic94xx_hwi.h  |  6 +++---
 drivers/scsi/aic94xx/aic94xx_task.c | 10 +++++-----
 drivers/scsi/libsas/sas_scsi_host.c |  2 +-
 include/scsi/libsas.h               |  4 ++--
 6 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/drivers/scsi/aic94xx/aic94xx.h b/drivers/scsi/aic94xx/aic94xx.h
index 1bd5b4ecf3d5..71a031df7a34 100644
--- a/drivers/scsi/aic94xx/aic94xx.h
+++ b/drivers/scsi/aic94xx/aic94xx.h
@@ -94,7 +94,7 @@ void asd_dev_gone(struct domain_device *dev);
 
 void asd_invalidate_edb(struct asd_ascb *ascb, int edb_id);
 
-int  asd_execute_task(struct sas_task *, int num, unsigned long gfp_flags);
+int  asd_execute_task(struct sas_task *, int num, gfp_t gfp_flags);
 
 /* ---------- TMFs ---------- */
 int  asd_abort_task(struct sas_task *);
diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c
index a24201351108..1d8c5e5f442e 100644
--- a/drivers/scsi/aic94xx/aic94xx_hwi.c
+++ b/drivers/scsi/aic94xx/aic94xx_hwi.c
@@ -267,7 +267,7 @@ static int asd_init_dl(struct asd_ha_struct *asd_ha)
 
 /* ---------- EDB and ESCB init ---------- */
 
-static int asd_alloc_edbs(struct asd_ha_struct *asd_ha, unsigned int gfp_flags)
+static int asd_alloc_edbs(struct asd_ha_struct *asd_ha, gfp_t gfp_flags)
 {
 	struct asd_seq_data *seq = &asd_ha->seq;
 	int i;
@@ -298,7 +298,7 @@ Err_unroll:
 }
 
 static int asd_alloc_escbs(struct asd_ha_struct *asd_ha,
-			   unsigned int gfp_flags)
+			   gfp_t gfp_flags)
 {
 	struct asd_seq_data *seq = &asd_ha->seq;
 	struct asd_ascb *escb;
@@ -1028,7 +1028,7 @@ irqreturn_t asd_hw_isr(int irq, void *dev_id, struct pt_regs *regs)
 /* ---------- SCB handling ---------- */
 
 static inline struct asd_ascb *asd_ascb_alloc(struct asd_ha_struct *asd_ha,
-					      unsigned int gfp_flags)
+					      gfp_t gfp_flags)
 {
 	extern kmem_cache_t *asd_ascb_cache;
 	struct asd_seq_data *seq = &asd_ha->seq;
@@ -1086,7 +1086,7 @@ undo:
  */
 struct asd_ascb *asd_ascb_alloc_list(struct asd_ha_struct
 				     *asd_ha, int *num,
-				     unsigned int gfp_flags)
+				     gfp_t gfp_flags)
 {
 	struct asd_ascb *first = NULL;
 
diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.h b/drivers/scsi/aic94xx/aic94xx_hwi.h
index c7d505388fed..8498144aa5e1 100644
--- a/drivers/scsi/aic94xx/aic94xx_hwi.h
+++ b/drivers/scsi/aic94xx/aic94xx_hwi.h
@@ -242,7 +242,7 @@ struct asd_ha_struct {
 
 /* ---------- DMA allocs ---------- */
 
-static inline struct asd_dma_tok *asd_dmatok_alloc(unsigned int flags)
+static inline struct asd_dma_tok *asd_dmatok_alloc(gfp_t flags)
 {
 	return kmem_cache_alloc(asd_dma_token_cache, flags);
 }
@@ -254,7 +254,7 @@ static inline void asd_dmatok_free(struct asd_dma_tok *token)
 
 static inline struct asd_dma_tok *asd_alloc_coherent(struct asd_ha_struct *
 						     asd_ha, size_t size,
-						     unsigned int flags)
+						     gfp_t flags)
 {
 	struct asd_dma_tok *token = asd_dmatok_alloc(flags);
 	if (token) {
@@ -376,7 +376,7 @@ irqreturn_t asd_hw_isr(int irq, void *dev_id, struct pt_regs *regs);
 
 struct asd_ascb *asd_ascb_alloc_list(struct asd_ha_struct
 				     *asd_ha, int *num,
-				     unsigned int gfp_mask);
+				     gfp_t gfp_mask);
 
 int  asd_post_ascb_list(struct asd_ha_struct *asd_ha, struct asd_ascb *ascb,
 			int num);
diff --git a/drivers/scsi/aic94xx/aic94xx_task.c b/drivers/scsi/aic94xx/aic94xx_task.c
index 285e70dae933..d202ed5a6709 100644
--- a/drivers/scsi/aic94xx/aic94xx_task.c
+++ b/drivers/scsi/aic94xx/aic94xx_task.c
@@ -53,7 +53,7 @@ static const u8 data_dir_flags[] = {
 
 static inline int asd_map_scatterlist(struct sas_task *task,
 				      struct sg_el *sg_arr,
-				      unsigned long gfp_flags)
+				      gfp_t gfp_flags)
 {
 	struct asd_ascb *ascb = task->lldd_task;
 	struct asd_ha_struct *asd_ha = ascb->ha;
@@ -368,7 +368,7 @@ Again:
 /* ---------- ATA ---------- */
 
 static int asd_build_ata_ascb(struct asd_ascb *ascb, struct sas_task *task,
-			      unsigned long gfp_flags)
+			      gfp_t gfp_flags)
 {
 	struct domain_device *dev = task->dev;
 	struct scb *scb;
@@ -437,7 +437,7 @@ static void asd_unbuild_ata_ascb(struct asd_ascb *a)
 /* ---------- SMP ---------- */
 
 static int asd_build_smp_ascb(struct asd_ascb *ascb, struct sas_task *task,
-			      unsigned long gfp_flags)
+			      gfp_t gfp_flags)
 {
 	struct asd_ha_struct *asd_ha = ascb->ha;
 	struct domain_device *dev = task->dev;
@@ -487,7 +487,7 @@ static void asd_unbuild_smp_ascb(struct asd_ascb *a)
 /* ---------- SSP ---------- */
 
 static int asd_build_ssp_ascb(struct asd_ascb *ascb, struct sas_task *task,
-			      unsigned long gfp_flags)
+			      gfp_t gfp_flags)
 {
 	struct domain_device *dev = task->dev;
 	struct scb *scb;
@@ -550,7 +550,7 @@ static inline int asd_can_queue(struct asd_ha_struct *asd_ha, int num)
 }
 
 int asd_execute_task(struct sas_task *task, const int num,
-		     unsigned long gfp_flags)
+		     gfp_t gfp_flags)
 {
 	int res = 0;
 	LIST_HEAD(alist);
diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 43e0e4e36934..7f9e89bcac7e 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -134,7 +134,7 @@ static enum task_attribute sas_scsi_get_task_attr(struct scsi_cmnd *cmd)
 
 static struct sas_task *sas_create_task(struct scsi_cmnd *cmd,
 					       struct domain_device *dev,
-					       unsigned long gfp_flags)
+					       gfp_t gfp_flags)
 {
 	struct sas_task *task = sas_alloc_task(gfp_flags);
 	struct scsi_lun lun;
diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h
index 8e39982fc3db..9582e8401669 100644
--- a/include/scsi/libsas.h
+++ b/include/scsi/libsas.h
@@ -534,7 +534,7 @@ struct sas_task {
 #define SAS_TASK_STATE_DONE     2
 #define SAS_TASK_STATE_ABORTED  4
 
-static inline struct sas_task *sas_alloc_task(unsigned long flags)
+static inline struct sas_task *sas_alloc_task(gfp_t flags)
 {
 	extern kmem_cache_t *sas_task_cache;
 	struct sas_task *task = kmem_cache_alloc(sas_task_cache, flags);
@@ -570,7 +570,7 @@ struct sas_domain_function_template {
 	void (*lldd_dev_gone)(struct domain_device *);
 
 	int (*lldd_execute_task)(struct sas_task *, int num,
-				 unsigned long gfp_flags);
+				 gfp_t gfp_flags);
 
 	/* Task Management Functions. Must be called from process context. */
 	int (*lldd_abort_task)(struct sas_task *);
-- 
cgit v1.2.3


From fd88edd20fb0e8e2729aa8ce565316242189ceea Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 15 Sep 2006 23:34:34 +0100
Subject: [ARM] 3795/1: S3C24XX: add base AC97 registers

Add base definitions for the AC97 register
definitions.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/map.h       |  5 +++++
 include/asm-arm/arch-s3c2410/regs-ac97.h | 23 +++++++++++++++++++++++
 2 files changed, 28 insertions(+)
 create mode 100644 include/asm-arm/arch-s3c2410/regs-ac97.h

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/map.h b/include/asm-arm/arch-s3c2410/map.h
index 27ba0ac3fdd5..7895042d176b 100644
--- a/include/asm-arm/arch-s3c2410/map.h
+++ b/include/asm-arm/arch-s3c2410/map.h
@@ -160,6 +160,11 @@
 #define S3C2440_PA_CAMIF   (0x4F000000)
 #define S3C2440_SZ_CAMIF   SZ_1M
 
+/* AC97 */
+
+#define S3C2440_PA_AC97	   (0x5B000000)
+#define S3C2440_SZ_AC97	   SZ_1M
+
 /* ISA style IO, for each machine to sort out mappings for, if it
  * implements it. We reserve two 16M regions for ISA.
  */
diff --git a/include/asm-arm/arch-s3c2410/regs-ac97.h b/include/asm-arm/arch-s3c2410/regs-ac97.h
new file mode 100644
index 000000000000..bdd6a4f93d7f
--- /dev/null
+++ b/include/asm-arm/arch-s3c2410/regs-ac97.h
@@ -0,0 +1,23 @@
+/* linux/include/asm-arm/arch-s3c2410/regs-ac97.h
+ *
+ * Copyright (c) 2006 Simtec Electronics <linux@simtec.co.uk>
+ *		http://www.simtec.co.uk/products/SWLINUX/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * S3C2440 AC97 Controller
+*/
+
+#ifndef __ASM_ARCH_REGS_AC97_H
+#define __ASM_ARCH_REGS_AC97_H __FILE__
+
+#define S3C_AC97_GLBCTRL	(0x00)
+#define S3C_AC97_GLBSTAT	(0x04)
+#define S3C_AC97_CODEC_CMD	(0x08)
+#define S3C_AC97_PCM_ADDR	(0x10)
+#define S3C_AC97_PCM_DATA	(0x18)
+#define S3C_AC97_MIC_DATA	(0x1C)
+
+#endif /* __ASM_ARCH_REGS_AC97_H */
-- 
cgit v1.2.3


From 505788cccbb96cd496b646594c8a5fcdc26bc2d9 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 15 Sep 2006 23:42:24 +0100
Subject: [ARM] 3796/1: S3C24XX: Add per-cpu DMA channel mapper

Allow each CPU type in the S3C24XX range to
select the DMA channel mapping it supports.

We change the DMA registration to use an
virtual channel number that the DMA system
will allocate to a hardware channel at
request time.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/Makefile     |   6 +
 arch/arm/mach-s3c2410/dma.c        | 241 ++++++++++++++++++++++++++-----------
 arch/arm/mach-s3c2410/dma.h        |  45 +++++++
 include/asm-arm/arch-s3c2410/dma.h |  34 ++++++
 4 files changed, 257 insertions(+), 69 deletions(-)
 create mode 100644 arch/arm/mach-s3c2410/dma.h

(limited to 'include')

diff --git a/arch/arm/mach-s3c2410/Makefile b/arch/arm/mach-s3c2410/Makefile
index 0eadec916214..e4c32f6ffb71 100644
--- a/arch/arm/mach-s3c2410/Makefile
+++ b/arch/arm/mach-s3c2410/Makefile
@@ -9,6 +9,8 @@ obj-y			:= cpu.o irq.o time.o gpio.o clock.o devs.o
 obj-m			:=
 obj-n			:=
 obj-			:=
+obj-dma-y		:=
+obj-dma-n		:=
 
 # DMA
 obj-$(CONFIG_S3C2410_DMA)	+= dma.o
@@ -57,6 +59,10 @@ obj-$(CONFIG_CPU_S3C2442)	+= s3c2442-clock.o
 
 obj-$(CONFIG_BAST_PC104_IRQ)	+= bast-irq.o
 
+# merge in dma objects
+
+obj-y				+= $(obj-dma-y)
+
 # machine specific support
 
 obj-$(CONFIG_MACH_ANUBIS)	+= mach-anubis.o
diff --git a/arch/arm/mach-s3c2410/dma.c b/arch/arm/mach-s3c2410/dma.c
index cc92a7b2db88..d264bbbd8bef 100644
--- a/arch/arm/mach-s3c2410/dma.c
+++ b/arch/arm/mach-s3c2410/dma.c
@@ -1,35 +1,16 @@
-/* linux/arch/arm/mach-bast/dma.c
+/* linux/arch/arm/mach-s3c2410/dma.c
  *
- * (c) 2003-2005 Simtec Electronics
+ * (c) 2003-2005,2006 Simtec Electronics
  *	Ben Dooks <ben@simtec.co.uk>
  *
  * S3C2410 DMA core
  *
- * http://www.simtec.co.uk/products/EB2410ITX/
+ * http://armlinux.simtec.co.uk/
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
- * Changelog:
- *  27-Feb-2005 BJD  Added kmem cache for dma descriptors
- *  18-Nov-2004 BJD  Removed error for loading onto stopped channel
- *  10-Nov-2004 BJD  Ensure all external symbols exported for modules
- *  10-Nov-2004 BJD  Use sys_device and sysdev_class for power management
- *  08-Aug-2004 BJD  Apply rmk's suggestions
- *  21-Jul-2004 BJD  Ported to linux 2.6
- *  12-Jul-2004 BJD  Finished re-write and change of API
- *  06-Jul-2004 BJD  Rewrote dma code to try and cope with various problems
- *  23-May-2003 BJD  Created file
- *  19-Aug-2003 BJD  Cleanup, header fix, added URL
- *
- * This file is based on the Sangwook Lee/Samsung patches, re-written due
- * to various ommisions from the code (such as flexible dma configuration)
- * for use with the BAST system board.
- *
- * The re-write is pretty much complete, and should be good enough for any
- * possible DMA function
- */
+*/
 
 
 #ifdef CONFIG_S3C2410_DMA_DEBUG
@@ -55,10 +36,14 @@
 #include <asm/mach/dma.h>
 #include <asm/arch/map.h>
 
+#include "dma.h"
+
 /* io map for dma */
 static void __iomem *dma_base;
 static kmem_cache_t *dma_kmem;
 
+struct s3c24xx_dma_selection dma_sel;
+
 /* dma channel state information */
 struct s3c2410_dma_chan s3c2410_chans[S3C2410_DMA_CHANNELS];
 
@@ -79,7 +64,6 @@ dma_wrreg(struct s3c2410_dma_chan *chan, int reg, unsigned long val)
 	pr_debug("writing %08x to register %08x\n",(unsigned int)val,reg);
 	writel(val, dma_regaddr(chan, reg));
 }
-
 #endif
 
 #define dma_rdreg(chan, reg) readl((chan)->regs + (reg))
@@ -151,12 +135,20 @@ dmadbg_showregs(const char *fname, int line, struct s3c2410_dma_chan *chan)
 #define dbg_showchan(chan) do { } while(0)
 #endif /* CONFIG_S3C2410_DMA_DEBUG */
 
-#define check_channel(chan) \
-  do { if ((chan) >= S3C2410_DMA_CHANNELS) { \
-    printk(KERN_ERR "%s: invalid channel %d\n", __FUNCTION__, (chan)); \
-    return -EINVAL; \
-  } } while(0)
+static struct s3c2410_dma_chan *dma_chan_map[DMACH_MAX];
 
+/* lookup_dma_channel
+ *
+ * change the dma channel number given into a real dma channel id
+*/
+
+static struct s3c2410_dma_chan *lookup_dma_channel(unsigned int channel)
+{
+	if (channel & DMACH_LOW_LEVEL)
+		return &s3c2410_chans[channel & ~DMACH_LOW_LEVEL];
+	else
+		return dma_chan_map[channel];
+}
 
 /* s3c2410_dma_stats_timeout
  *
@@ -321,8 +313,10 @@ static inline void
 s3c2410_dma_buffdone(struct s3c2410_dma_chan *chan, struct s3c2410_dma_buf *buf,
 		     enum s3c2410_dma_buffresult result)
 {
+#if 0
 	pr_debug("callback_fn=%p, buf=%p, id=%p, size=%d, result=%d\n",
 		 chan->callback_fn, buf, buf->id, buf->size, result);
+#endif
 
 	if (chan->callback_fn != NULL) {
 		(chan->callback_fn)(chan, buf->id, buf->size, result);
@@ -439,7 +433,6 @@ s3c2410_dma_canload(struct s3c2410_dma_chan *chan)
 	return 0;
 }
 
-
 /* s3c2410_dma_enqueue
  *
  * queue an given buffer for dma transfer.
@@ -460,11 +453,12 @@ s3c2410_dma_canload(struct s3c2410_dma_chan *chan)
 int s3c2410_dma_enqueue(unsigned int channel, void *id,
 			dma_addr_t data, int size)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 	struct s3c2410_dma_buf *buf;
 	unsigned long flags;
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	pr_debug("%s: id=%p, data=%08x, size=%d\n",
 		 __FUNCTION__, id, (unsigned int)data, size);
@@ -562,8 +556,10 @@ s3c2410_dma_freebuf(struct s3c2410_dma_buf *buf)
 static inline void
 s3c2410_dma_lastxfer(struct s3c2410_dma_chan *chan)
 {
+#if 0
 	pr_debug("dma%d: s3c2410_dma_lastxfer: load_state %d\n",
 		 chan->number, chan->load_state);
+#endif
 
 	switch (chan->load_state) {
 	case S3C2410_DMALOAD_NONE:
@@ -718,7 +714,8 @@ s3c2410_dma_irq(int irq, void *devpw, struct pt_regs *regs)
 		if (chan->load_state == S3C2410_DMALOAD_NONE) {
 			pr_debug("dma%d: end of transfer, stopping channel (%ld)\n",
 				 chan->number, jiffies);
-			s3c2410_dma_ctrl(chan->number, S3C2410_DMAOP_STOP);
+			s3c2410_dma_ctrl(chan->number | DMACH_LOW_LEVEL,
+					 S3C2410_DMAOP_STOP);
 		}
 	}
 
@@ -726,37 +723,34 @@ s3c2410_dma_irq(int irq, void *devpw, struct pt_regs *regs)
 	return IRQ_HANDLED;
 }
 
+static struct s3c2410_dma_chan *s3c2410_dma_map_channel(int channel);
+
 /* s3c2410_request_dma
  *
  * get control of an dma channel
 */
 
-int s3c2410_dma_request(unsigned int channel, struct s3c2410_dma_client *client,
+int s3c2410_dma_request(unsigned int channel,
+			struct s3c2410_dma_client *client,
 			void *dev)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan;
 	unsigned long flags;
 	int err;
 
 	pr_debug("dma%d: s3c2410_request_dma: client=%s, dev=%p\n",
 		 channel, client->name, dev);
 
-	check_channel(channel);
-
 	local_irq_save(flags);
 
-	dbg_showchan(chan);
-
-	if (chan->in_use) {
-		if (client != chan->client) {
-			printk(KERN_ERR "dma%d: already in use\n", channel);
-			local_irq_restore(flags);
-			return -EBUSY;
-		} else {
-			printk(KERN_ERR "dma%d: client already has channel\n", channel);
-		}
+	chan = s3c2410_dma_map_channel(channel);
+	if (chan == NULL) {
+		local_irq_restore(flags);
+		return -EBUSY;
 	}
 
+	dbg_showchan(chan);
+
 	chan->client = client;
 	chan->in_use = 1;
 
@@ -809,14 +803,14 @@ EXPORT_SYMBOL(s3c2410_dma_request);
 
 int s3c2410_dma_free(dmach_t channel, struct s3c2410_dma_client *client)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 	unsigned long flags;
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	local_irq_save(flags);
 
-
 	if (chan->client != client) {
 		printk(KERN_WARNING "dma%d: possible free from different client (channel %p, passed %p)\n",
 		       channel, chan->client, client);
@@ -837,8 +831,12 @@ int s3c2410_dma_free(dmach_t channel, struct s3c2410_dma_client *client)
 
 	if (chan->irq_claimed)
 		free_irq(chan->irq, (void *)chan);
+
 	chan->irq_claimed = 0;
 
+	if (!(channel & DMACH_LOW_LEVEL))
+		dma_chan_map[channel] = NULL;
+
 	local_irq_restore(flags);
 
 	return 0;
@@ -848,8 +846,8 @@ EXPORT_SYMBOL(s3c2410_dma_free);
 
 static int s3c2410_dma_dostop(struct s3c2410_dma_chan *chan)
 {
-	unsigned long tmp;
 	unsigned long flags;
+	unsigned long tmp;
 
 	pr_debug("%s:\n", __FUNCTION__);
 
@@ -997,9 +995,10 @@ s3c2410_dma_started(struct s3c2410_dma_chan *chan)
 int
 s3c2410_dma_ctrl(dmach_t channel, enum s3c2410_chan_op op)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	switch (op) {
 	case S3C2410_DMAOP_START:
@@ -1046,12 +1045,19 @@ int s3c2410_dma_config(dmach_t channel,
 		       int xferunit,
 		       int dcon)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
 	pr_debug("%s: chan=%d, xfer_unit=%d, dcon=%08x\n",
 		 __FUNCTION__, channel, xferunit, dcon);
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
+
+	printk("Initial dcon is %08x\n", dcon);
+
+	dcon |= chan->dcon & dma_sel.dcon_mask;
+
+	printk("New dcon is %08x\n", dcon);
 
 	switch (xferunit) {
 	case 1:
@@ -1086,9 +1092,10 @@ EXPORT_SYMBOL(s3c2410_dma_config);
 
 int s3c2410_dma_setflags(dmach_t channel, unsigned int flags)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	pr_debug("%s: chan=%p, flags=%08x\n", __FUNCTION__, chan, flags);
 
@@ -1106,9 +1113,10 @@ EXPORT_SYMBOL(s3c2410_dma_setflags);
 
 int s3c2410_dma_set_opfn(dmach_t channel, s3c2410_dma_opfn_t rtn)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	pr_debug("%s: chan=%p, op rtn=%p\n", __FUNCTION__, chan, rtn);
 
@@ -1121,9 +1129,10 @@ EXPORT_SYMBOL(s3c2410_dma_set_opfn);
 
 int s3c2410_dma_set_buffdone_fn(dmach_t channel, s3c2410_dma_cbfn_t rtn)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	pr_debug("%s: chan=%p, callback rtn=%p\n", __FUNCTION__, chan, rtn);
 
@@ -1153,9 +1162,10 @@ int s3c2410_dma_devconfig(int channel,
 			  int hwcfg,
 			  unsigned long devaddr)
 {
-	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
-	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	pr_debug("%s: source=%d, hwcfg=%08x, devaddr=%08lx\n",
 		 __FUNCTION__, (int)source, hwcfg, devaddr);
@@ -1200,9 +1210,10 @@ EXPORT_SYMBOL(s3c2410_dma_devconfig);
 
 int s3c2410_dma_getposition(dmach_t channel, dma_addr_t *src, dma_addr_t *dst)
 {
- 	struct s3c2410_dma_chan *chan = &s3c2410_chans[channel];
+ 	struct s3c2410_dma_chan *chan = lookup_dma_channel(channel);
 
- 	check_channel(channel);
+	if (chan == NULL)
+		return -EINVAL;
 
 	if (src != NULL)
  		*src = dma_rdreg(chan, S3C2410_DMA_DCSRC);
@@ -1252,7 +1263,7 @@ static int s3c2410_dma_resume(struct sys_device *dev)
 #define s3c2410_dma_resume  NULL
 #endif /* CONFIG_PM */
 
-static struct sysdev_class dma_sysclass = {
+struct sysdev_class dma_sysclass = {
 	set_kset_name("s3c24xx-dma"),
 	.suspend	= s3c2410_dma_suspend,
 	.resume		= s3c2410_dma_resume,
@@ -1265,7 +1276,6 @@ static void s3c2410_dma_cache_ctor(void *p, kmem_cache_t *c, unsigned long f)
 	memset(p, 0, sizeof(struct s3c2410_dma_buf));
 }
 
-
 /* initialisation code */
 
 static int __init s3c2410_init_dma(void)
@@ -1274,7 +1284,7 @@ static int __init s3c2410_init_dma(void)
 	int channel;
 	int ret;
 
-	printk("S3C2410 DMA Driver, (c) 2003-2004 Simtec Electronics\n");
+	printk("S3C24XX DMA Driver, (c) 2003-2004,2006 Simtec Electronics\n");
 
 	dma_base = ioremap(S3C24XX_PA_DMA, 0x200);
 	if (dma_base == NULL) {
@@ -1282,6 +1292,8 @@ static int __init s3c2410_init_dma(void)
 		return -ENOMEM;
 	}
 
+	printk("Registering sysclass\n");
+
 	ret = sysdev_class_register(&dma_sysclass);
 	if (ret != 0) {
 		printk(KERN_ERR "dma sysclass registration failed\n");
@@ -1335,4 +1347,95 @@ static int __init s3c2410_init_dma(void)
 	return ret;
 }
 
-__initcall(s3c2410_init_dma);
+core_initcall(s3c2410_init_dma);
+
+static inline int is_channel_valid(unsigned int channel)
+{
+	return (channel & DMA_CH_VALID);
+}
+
+/* s3c2410_dma_map_channel()
+ *
+ * turn the virtual channel number into a real, and un-used hardware
+ * channel.
+ *
+ * currently this code uses first-free channel from the specified harware
+ * map, not taking into account anything that the board setup code may
+ * have to say about the likely peripheral set to be in use.
+*/
+
+struct s3c2410_dma_chan *s3c2410_dma_map_channel(int channel)
+{
+	struct s3c24xx_dma_map *ch_map;
+	struct s3c2410_dma_chan *dmach;
+	int ch;
+
+	if (dma_sel.map == NULL || channel > dma_sel.map_size)
+		return NULL;
+
+	ch_map = dma_sel.map + channel;
+
+	for (ch = 0; ch < S3C2410_DMA_CHANNELS; ch++) {
+		if (!is_channel_valid(ch_map->channels[ch]))
+			continue;
+
+		if (s3c2410_chans[ch].in_use == 0) {
+			printk("mapped channel %d to %d\n", channel, ch);
+			break;
+		}
+	}
+
+	if (ch >= S3C2410_DMA_CHANNELS)
+		return NULL;
+
+	/* update our channel mapping */
+
+	dmach = &s3c2410_chans[ch];
+	dma_chan_map[channel] = dmach;
+
+	/* select the channel */
+
+	(dma_sel.select)(dmach, ch_map);
+
+	return dmach;
+}
+
+static void s3c24xx_dma_show_ch(struct s3c24xx_dma_map *map, int ch)
+{
+	/* show the channel configuration */
+
+	printk("%2d: %20s, channels %c%c%c%c\n", ch, map->name,
+	       (is_channel_valid(map->channels[0]) ? '0' : '-'),
+	       (is_channel_valid(map->channels[1]) ? '1' : '-'),
+	       (is_channel_valid(map->channels[2]) ? '2' : '-'),
+	       (is_channel_valid(map->channels[3]) ? '3' : '-'));
+}
+
+static int s3c24xx_dma_check_entry(struct s3c24xx_dma_map *map, int ch)
+{
+	if (1)
+		s3c24xx_dma_show_ch(map, ch);
+
+	return 0;
+}
+
+int __init s3c24xx_dma_init_map(struct s3c24xx_dma_selection *sel)
+{
+	struct s3c24xx_dma_map *nmap;
+	size_t map_sz = sizeof(*nmap) * sel->map_size;
+	int ptr;
+
+	nmap = kmalloc(map_sz, GFP_KERNEL);
+	if (nmap == NULL)
+		return -ENOMEM;
+
+	memcpy(nmap, sel->map, map_sz);
+	memcpy(&dma_sel, sel, sizeof(*sel));
+
+	dma_sel.map = nmap;
+
+	for (ptr = 0; ptr < sel->map_size; ptr++)
+		s3c24xx_dma_check_entry(nmap+ptr, ptr);
+
+	return 0;
+}
diff --git a/arch/arm/mach-s3c2410/dma.h b/arch/arm/mach-s3c2410/dma.h
new file mode 100644
index 000000000000..0ebfe0aab80b
--- /dev/null
+++ b/arch/arm/mach-s3c2410/dma.h
@@ -0,0 +1,45 @@
+/* arch/arm/mach-s3c2410/dma.h
+ *
+ * Copyright (C) 2006 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * Samsung S3C24XX DMA support
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+extern struct sysdev_class dma_sysclass;
+extern struct s3c2410_dma_chan s3c2410_chans[S3C2410_DMA_CHANNELS];
+
+#define DMA_CH_VALID		(1<<31)
+
+struct s3c24xx_dma_addr {
+	unsigned long		from;
+	unsigned long		to;
+};
+
+/* struct s3c24xx_dma_map
+ *
+ * this holds the mapping information for the channel selected
+ * to be connected to the specified device
+*/
+
+struct s3c24xx_dma_map {
+	const char		*name;
+	struct s3c24xx_dma_addr  hw_addr;
+
+	unsigned long		 channels[S3C2410_DMA_CHANNELS];
+};
+
+struct s3c24xx_dma_selection {
+	struct s3c24xx_dma_map	*map;
+	unsigned long		 map_size;
+	unsigned long		 dcon_mask;
+
+	void	(*select)(struct s3c2410_dma_chan *chan,
+			  struct s3c24xx_dma_map *map);
+};
+
+extern int s3c24xx_dma_init_map(struct s3c24xx_dma_selection *sel);
diff --git a/include/asm-arm/arch-s3c2410/dma.h b/include/asm-arm/arch-s3c2410/dma.h
index 3661e465b0a5..166fc89d62d7 100644
--- a/include/asm-arm/arch-s3c2410/dma.h
+++ b/include/asm-arm/arch-s3c2410/dma.h
@@ -23,6 +23,36 @@
 #define MAX_DMA_ADDRESS		0x40000000
 #define MAX_DMA_TRANSFER_SIZE   0x100000 /* Data Unit is half word  */
 
+/* We use `virtual` dma channels to hide the fact we have only a limited
+ * number of DMA channels, and not of all of them (dependant on the device)
+ * can be attached to any DMA source. We therefore let the DMA core handle
+ * the allocation of hardware channels to clients.
+*/
+
+enum dma_ch {
+	DMACH_XD0,
+	DMACH_XD1,
+	DMACH_SDI,
+	DMACH_SPI0,
+	DMACH_SPI1,
+	DMACH_UART0,
+	DMACH_UART1,
+	DMACH_UART2,
+	DMACH_TIMER,
+	DMACH_I2S_IN,
+	DMACH_I2S_OUT,
+	DMACH_PCM_IN,
+	DMACH_PCM_OUT,
+	DMACH_MIC_IN,
+	DMACH_USB_EP1,
+	DMACH_USB_EP2,
+	DMACH_USB_EP3,
+	DMACH_USB_EP4,
+	DMACH_MAX,		/* the end entry */
+};
+
+#define DMACH_LOW_LEVEL	(1<<28)	/* use this to specifiy hardware ch no */
+
 /* we have 4 dma channels */
 #define S3C2410_DMA_CHANNELS        (4)
 
@@ -149,6 +179,8 @@ struct s3c2410_dma_stats {
 	unsigned long		timeout_failed;
 };
 
+struct s3c2410_dma_map;
+
 /* struct s3c2410_dma_chan
  *
  * full state information for each DMA channel
@@ -174,6 +206,8 @@ struct s3c2410_dma_chan {
 	unsigned long		 load_timeout;
 	unsigned int		 flags;		/* channel flags */
 
+	struct s3c24xx_dma_map	*map;		/* channel hw maps */
+
 	/* channel's hardware position and configuration */
 	void __iomem		*regs;		/* channels registers */
 	void __iomem		*addr_reg;	/* data address register */
-- 
cgit v1.2.3


From 3e9fc8e5de0fb00226325cf34eb08411eb72ec6d Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 16 Sep 2006 00:11:32 +0100
Subject: [ARM] 3804/1: S3C2442: LCD register update

Add LCD register definitions for the S3C2442.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/regs-lcd.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/regs-lcd.h b/include/asm-arm/arch-s3c2410/regs-lcd.h
index b306d6e3135d..d8f1adfd17f0 100644
--- a/include/asm-arm/arch-s3c2410/regs-lcd.h
+++ b/include/asm-arm/arch-s3c2410/regs-lcd.h
@@ -113,6 +113,13 @@
 #define	S3C2410_LCDINT_FRSYNC	(1<<1)
 #define S3C2410_LCDINT_FICNT	(1<<0)
 
+/* s3c2442 extra stn registers */
+
+#define S3C2442_REDLUT		S3C2410_LCDREG(0x20)
+#define S3C2442_GREENLUT	S3C2410_LCDREG(0x24)
+#define S3C2442_BLUELUT		S3C2410_LCDREG(0x28)
+#define S3C2442_DITHMODE	S3C2410_LCDREG(0x20)
+
 #define S3C2410_LPCSEL	   S3C2410_LCDREG(0x60)
 
 #define S3C2410_TFTPAL(x)  S3C2410_LCDREG((0x400 + (x)*4))
-- 
cgit v1.2.3


From 34148c6990d2f0107b53fe4ddf29b1ba30e613d3 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 16 Sep 2006 00:12:53 +0100
Subject: [ARM] 3805/1: S3C2412: LCD register update

Add LCD register definitions for the S3C2412.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/regs-lcd.h | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/regs-lcd.h b/include/asm-arm/arch-s3c2410/regs-lcd.h
index d8f1adfd17f0..6d7881c8cfc8 100644
--- a/include/asm-arm/arch-s3c2410/regs-lcd.h
+++ b/include/asm-arm/arch-s3c2410/regs-lcd.h
@@ -63,6 +63,8 @@
 #define S3C2410_LCDCON3_GET_HBPD(x) ( ((x) >> 19) & 0x7F)
 #define S3C2410_LCDCON3_GET_HFPD(x) ( ((x) >>  0) & 0xFF)
 
+/* LDCCON4 changes for STN mode on the S3C2412 */
+
 #define S3C2410_LCDCON4_MVAL(x)	    ((x) << 8)
 #define S3C2410_LCDCON4_HSPW(x)	    ((x) << 0)
 #define S3C2410_LCDCON4_WLH(x)	    ((x) << 0)
@@ -124,6 +126,27 @@
 
 #define S3C2410_TFTPAL(x)  S3C2410_LCDREG((0x400 + (x)*4))
 
+/* S3C2412 registers */
+
+#define S3C2412_TPAL		S3C2410_LCDREG(0x20)
+
+#define S3C2412_LCDINTPND	S3C2410_LCDREG(0x24)
+#define S3C2412_LCDSRCPND	S3C2410_LCDREG(0x28)
+#define S3C2412_LCDINTMSK	S3C2410_LCDREG(0x2C)
+
+#define S3C2412_TCONSEL		S3C2410_LCDREG(0x30)
+
+#define S3C2412_LCDCON6		S3C2410_LCDREG(0x34)
+#define S3C2412_LCDCON7		S3C2410_LCDREG(0x38)
+#define S3C2412_LCDCON8		S3C2410_LCDREG(0x3C)
+#define S3C2412_LCDCON9		S3C2410_LCDREG(0x40)
+
+#define S3C2412_REDLUT(x)	S3C2410_LCDREG(0x44 + ((x)*4))
+#define S3C2412_GREENLUT(x)	S3C2410_LCDREG(0x60 + ((x)*4))
+#define S3C2412_BLUELUT(x)	S3C2410_LCDREG(0x98 + ((x)*4))
+
+#define S3C2412_FRCPAT(x)	S3C2410_LCDREG(0xB4 + ((x)*4))
+
 #endif /* ___ASM_ARCH_REGS_LCD_H */
 
 
-- 
cgit v1.2.3


From 8dd5c845bbc26c3517398abc3e5477b4b42e7176 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sat, 16 Sep 2006 10:47:18 +0100
Subject: [ARM] 3810/1: switch atomic helpers over to
 raw_local_irq_{save,restore}

Now that we have raw_* variants of local_irq_$FOO(), switch the atomic
helpers over to use those raw_* variants.  This is necessary when using
lockdep on pre-ARMv6 hardware, as lockdep uses atomic_t counters in the
trace_hardirqs_off() path.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/atomic.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/atomic.h b/include/asm-arm/atomic.h
index 4b0ce3e7de9a..ea88aa6bfc78 100644
--- a/include/asm-arm/atomic.h
+++ b/include/asm-arm/atomic.h
@@ -128,10 +128,10 @@ static inline int atomic_add_return(int i, atomic_t *v)
 	unsigned long flags;
 	int val;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	val = v->counter;
 	v->counter = val += i;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 
 	return val;
 }
@@ -141,10 +141,10 @@ static inline int atomic_sub_return(int i, atomic_t *v)
 	unsigned long flags;
 	int val;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	val = v->counter;
 	v->counter = val -= i;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 
 	return val;
 }
@@ -154,11 +154,11 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
 	int ret;
 	unsigned long flags;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	ret = v->counter;
 	if (likely(ret == old))
 		v->counter = new;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 
 	return ret;
 }
@@ -167,9 +167,9 @@ static inline void atomic_clear_mask(unsigned long mask, unsigned long *addr)
 {
 	unsigned long flags;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	*addr &= ~mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 }
 
 #endif /* __LINUX_ARM_ARCH__ */
-- 
cgit v1.2.3


From 51635ad282ead58b9d164f07e1fb62a9456b427c Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Sat, 16 Sep 2006 10:50:22 +0100
Subject: [ARM] 3813/1: prevent >= 4G /dev/mem mmap()

Prevent userland from mapping in physical address regions >= 4G by
checking for that in valid_mmap_phys_addr_range().

Unfortunately, we cannot override valid_mmap_phys_addr_range() without
also overriding valid_phys_addr_range(), so copy drivers/char/mem.c's
version of valid_phys_addr_range() over to arch/arm/mm/mmap.c as well.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/mmap.c   | 22 ++++++++++++++++++++++
 include/asm-arm/io.h |  4 ++++
 2 files changed, 26 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
index 29e54807c5bc..b0b5f4694070 100644
--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -114,3 +114,25 @@ full_search:
 	}
 }
 
+
+/*
+ * You really shouldn't be using read() or write() on /dev/mem.  This
+ * might go away in the future.
+ */
+int valid_phys_addr_range(unsigned long addr, size_t size)
+{
+	if (addr + size > __pa(high_memory))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * We don't use supersection mappings for mmap() on /dev/mem, which
+ * means that we can't map the memory area above the 4G barrier into
+ * userspace.
+ */
+int valid_mmap_phys_addr_range(unsigned long pfn, size_t size)
+{
+	return !(pfn + (size >> PAGE_SHIFT) > 0x00100000);
+}
diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h
index bf7b9dea30f1..8076a85c3675 100644
--- a/include/asm-arm/io.h
+++ b/include/asm-arm/io.h
@@ -280,6 +280,10 @@ extern void pci_iounmap(struct pci_dev *dev, void __iomem *addr);
 #define BIOVEC_MERGEABLE(vec1, vec2)	\
 	((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2)))
 
+#define ARCH_HAS_VALID_PHYS_ADDR_RANGE
+extern int valid_phys_addr_range(unsigned long addr, size_t size);
+extern int valid_mmap_phys_addr_range(unsigned long pfn, size_t size);
+
 /*
  * Convert a physical pointer to a virtual kernel pointer for /dev/mem
  * access
-- 
cgit v1.2.3


From 34348012d6b43eca5e241fe97381420d5758866c Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 18 Sep 2006 23:52:03 +0100
Subject: [ARM] 3800/2: S3C2412: DMA channel mappings

DMA channel mappings for the S3C2312

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/Makefile      |   1 +
 arch/arm/mach-s3c2410/s3c2412-dma.c | 160 ++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-s3c2410/dma.h  |  32 ++++++++
 3 files changed, 193 insertions(+)
 create mode 100644 arch/arm/mach-s3c2410/s3c2412-dma.c

(limited to 'include')

diff --git a/arch/arm/mach-s3c2410/Makefile b/arch/arm/mach-s3c2410/Makefile
index 23f0909915db..0d87a9de0034 100644
--- a/arch/arm/mach-s3c2410/Makefile
+++ b/arch/arm/mach-s3c2410/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_PM_SIMTEC)		+= pm-simtec.o
 obj-$(CONFIG_CPU_S3C2412)	+= s3c2412.o
 obj-$(CONFIG_CPU_S3C2412)	+= s3c2412-irq.o
 obj-$(CONFIG_CPU_S3C2412)	+= s3c2412-clock.o
+obj-dma-$(CONFIG_CPU_S3C2412)	+= s3c2412-dma.o
 
 #
 # S3C244X support
diff --git a/arch/arm/mach-s3c2410/s3c2412-dma.c b/arch/arm/mach-s3c2410/s3c2412-dma.c
new file mode 100644
index 000000000000..171f3706d36d
--- /dev/null
+++ b/arch/arm/mach-s3c2410/s3c2412-dma.c
@@ -0,0 +1,160 @@
+/* linux/arch/arm/mach-s3c2410/s3c2412-dma.c
+ *
+ * (c) 2006 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * S3C2412 DMA selection
+ *
+ * http://armlinux.simtec.co.uk/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sysdev.h>
+
+#include <asm/dma.h>
+#include <asm/arch/dma.h>
+#include <asm/io.h>
+
+#include "dma.h"
+#include "cpu.h"
+
+#include <asm/arch/regs-serial.h>
+#include <asm/arch/regs-gpio.h>
+#include <asm/arch/regs-ac97.h>
+#include <asm/arch/regs-mem.h>
+#include <asm/arch/regs-lcd.h>
+#include <asm/arch/regs-sdi.h>
+#include <asm/arch/regs-iis.h>
+#include <asm/arch/regs-spi.h>
+
+#define MAP(x) { (x)| DMA_CH_VALID, (x)| DMA_CH_VALID, (x)| DMA_CH_VALID, (x)| DMA_CH_VALID }
+
+static struct s3c24xx_dma_map __initdata s3c2412_dma_mappings[] = {
+	[DMACH_XD0] = {
+		.name		= "xdreq0",
+		.channels	= MAP(S3C2412_DMAREQSEL_XDREQ0),
+	},
+	[DMACH_XD1] = {
+		.name		= "xdreq1",
+		.channels	= MAP(S3C2412_DMAREQSEL_XDREQ1),
+	},
+	[DMACH_SDI] = {
+		.name		= "sdi",
+		.channels	= MAP(S3C2412_DMAREQSEL_SDI),
+		.hw_addr.to	= S3C2410_PA_IIS + S3C2410_IISFIFO,
+		.hw_addr.from	= S3C2410_PA_IIS + S3C2410_IISFIFO,
+	},
+	[DMACH_SPI0] = {
+		.name		= "spi0",
+		.channels	= MAP(S3C2412_DMAREQSEL_SPI0TX),
+		.hw_addr.to	= S3C2410_PA_SPI + S3C2410_SPTDAT,
+		.hw_addr.from	= S3C2410_PA_SPI + S3C2410_SPRDAT,
+	},
+	[DMACH_SPI1] = {
+		.name		= "spi1",
+		.channels	= MAP(S3C2412_DMAREQSEL_SPI1TX),
+		.hw_addr.to	= S3C2410_PA_SPI + 0x20 + S3C2410_SPTDAT,
+		.hw_addr.from	= S3C2410_PA_SPI + 0x20 + S3C2410_SPRDAT,
+	},
+	[DMACH_UART0] = {
+		.name		= "uart0",
+		.channels	= MAP(S3C2412_DMAREQSEL_UART0_0),
+		.hw_addr.to	= S3C2410_PA_UART0 + S3C2410_UTXH,
+		.hw_addr.from	= S3C2410_PA_UART0 + S3C2410_URXH,
+	},
+	[DMACH_UART1] = {
+		.name		= "uart1",
+		.channels	= MAP(S3C2412_DMAREQSEL_UART1_0),
+		.hw_addr.to	= S3C2410_PA_UART1 + S3C2410_UTXH,
+		.hw_addr.from	= S3C2410_PA_UART1 + S3C2410_URXH,
+	},
+      	[DMACH_UART2] = {
+		.name		= "uart2",
+		.channels	= MAP(S3C2412_DMAREQSEL_UART2_0),
+		.hw_addr.to	= S3C2410_PA_UART2 + S3C2410_UTXH,
+		.hw_addr.from	= S3C2410_PA_UART2 + S3C2410_URXH,
+	},
+	[DMACH_UART0_SRC2] = {
+		.name		= "uart0",
+		.channels	= MAP(S3C2412_DMAREQSEL_UART0_1),
+		.hw_addr.to	= S3C2410_PA_UART0 + S3C2410_UTXH,
+		.hw_addr.from	= S3C2410_PA_UART0 + S3C2410_URXH,
+	},
+	[DMACH_UART1_SRC2] = {
+		.name		= "uart1",
+		.channels	= MAP(S3C2412_DMAREQSEL_UART1_1),
+		.hw_addr.to	= S3C2410_PA_UART1 + S3C2410_UTXH,
+		.hw_addr.from	= S3C2410_PA_UART1 + S3C2410_URXH,
+	},
+      	[DMACH_UART2_SRC2] = {
+		.name		= "uart2",
+		.channels	= MAP(S3C2412_DMAREQSEL_UART2_1),
+		.hw_addr.to	= S3C2410_PA_UART2 + S3C2410_UTXH,
+		.hw_addr.from	= S3C2410_PA_UART2 + S3C2410_URXH,
+	},
+	[DMACH_TIMER] = {
+		.name		= "timer",
+		.channels	= MAP(S3C2412_DMAREQSEL_TIMER),
+	},
+	[DMACH_I2S_IN] = {
+		.name		= "i2s-sdi",
+		.channels	= MAP(S3C2412_DMAREQSEL_I2SRX),
+		.hw_addr.from	= S3C2410_PA_IIS + S3C2410_IISFIFO,
+	},
+	[DMACH_I2S_OUT] = {
+		.name		= "i2s-sdo",
+		.channels	= MAP(S3C2412_DMAREQSEL_I2STX),
+		.hw_addr.to	= S3C2410_PA_IIS + S3C2410_IISFIFO,
+	},
+	[DMACH_USB_EP1] = {
+		.name		= "usb-ep1",
+		.channels	= MAP(S3C2412_DMAREQSEL_USBEP1),
+	},
+	[DMACH_USB_EP2] = {
+		.name		= "usb-ep2",
+		.channels	= MAP(S3C2412_DMAREQSEL_USBEP2),
+	},
+	[DMACH_USB_EP3] = {
+		.name		= "usb-ep3",
+		.channels	= MAP(S3C2412_DMAREQSEL_USBEP3),
+	},
+	[DMACH_USB_EP4] = {
+		.name		= "usb-ep4",
+		.channels	= MAP(S3C2412_DMAREQSEL_USBEP4),
+	},
+};
+
+static void s3c2412_dma_select(struct s3c2410_dma_chan *chan,
+			       struct s3c24xx_dma_map *map)
+{
+	writel(chan->regs + S3C2412_DMA_DMAREQSEL,
+	       map->channels[0] | S3C2412_DMAREQSEL_HW);
+}
+
+static struct s3c24xx_dma_selection __initdata s3c2412_dma_sel = {
+	.select		= s3c2412_dma_select,
+	.dcon_mask	= 0,
+	.map		= s3c2412_dma_mappings,
+	.map_size	= ARRAY_SIZE(s3c2412_dma_mappings),
+};
+
+static int s3c2412_dma_add(struct sys_device *sysdev)
+{
+	return s3c24xx_dma_init_map(&s3c2412_dma_sel);
+}
+
+static struct sysdev_driver s3c2412_dma_driver = {
+	.add	= s3c2412_dma_add,
+};
+
+static int __init s3c2412_dma_init(void)
+{
+	return sysdev_driver_register(&s3c2412_sysclass, &s3c2412_dma_driver);
+}
+
+arch_initcall(s3c2412_dma_init);
diff --git a/include/asm-arm/arch-s3c2410/dma.h b/include/asm-arm/arch-s3c2410/dma.h
index 166fc89d62d7..7ac224836971 100644
--- a/include/asm-arm/arch-s3c2410/dma.h
+++ b/include/asm-arm/arch-s3c2410/dma.h
@@ -48,6 +48,9 @@ enum dma_ch {
 	DMACH_USB_EP2,
 	DMACH_USB_EP3,
 	DMACH_USB_EP4,
+	DMACH_UART0_SRC2,	/* s3c2412 second uart sources */
+	DMACH_UART1_SRC2,
+	DMACH_UART2_SRC2,
 	DMACH_MAX,		/* the end entry */
 };
 
@@ -317,6 +320,7 @@ extern int s3c2410_dma_set_buffdone_fn(dmach_t, s3c2410_dma_cbfn_t rtn);
 #define S3C2410_DMA_DCSRC       (0x18)
 #define S3C2410_DMA_DCDST       (0x1C)
 #define S3C2410_DMA_DMASKTRIG   (0x20)
+#define S3C2412_DMA_DMAREQSEL	(0x24)
 
 #define S3C2410_DISRCC_INC	(1<<0)
 #define S3C2410_DISRCC_APB	(1<<1)
@@ -383,4 +387,32 @@ extern int s3c2410_dma_set_buffdone_fn(dmach_t, s3c2410_dma_cbfn_t rtn);
 #define S3C2440_DCON_CH3_PCMOUT	(6<<24)
 #endif
 
+#ifdef CONFIG_CPU_S3C2412
+
+#define S3C2412_DMAREQSEL_SRC(x)	((x)<<1)
+
+#define S3C2412_DMAREQSEL_HW		(1)
+
+#define S3C2412_DMAREQSEL_SPI0TX	S3C2412_DMAREQSEL_SRC(0)
+#define S3C2412_DMAREQSEL_SPI0RX	S3C2412_DMAREQSEL_SRC(1)
+#define S3C2412_DMAREQSEL_SPI1TX	S3C2412_DMAREQSEL_SRC(2)
+#define S3C2412_DMAREQSEL_SPI1RX	S3C2412_DMAREQSEL_SRC(3)
+#define S3C2412_DMAREQSEL_I2STX		S3C2412_DMAREQSEL_SRC(4)
+#define S3C2412_DMAREQSEL_I2SRX		S3C2412_DMAREQSEL_SRC(5)
+#define S3C2412_DMAREQSEL_TIMER		S3C2412_DMAREQSEL_SRC(9)
+#define S3C2412_DMAREQSEL_SDI		S3C2412_DMAREQSEL_SRC(10)
+#define S3C2412_DMAREQSEL_USBEP1	S3C2412_DMAREQSEL_SRC(13)
+#define S3C2412_DMAREQSEL_USBEP2	S3C2412_DMAREQSEL_SRC(14)
+#define S3C2412_DMAREQSEL_USBEP3	S3C2412_DMAREQSEL_SRC(15)
+#define S3C2412_DMAREQSEL_USBEP4	S3C2412_DMAREQSEL_SRC(16)
+#define S3C2412_DMAREQSEL_XDREQ0	S3C2412_DMAREQSEL_SRC(17)
+#define S3C2412_DMAREQSEL_XDREQ1	S3C2412_DMAREQSEL_SRC(18)
+#define S3C2412_DMAREQSEL_UART0_0	S3C2412_DMAREQSEL_SRC(19)
+#define S3C2412_DMAREQSEL_UART0_1	S3C2412_DMAREQSEL_SRC(20)
+#define S3C2412_DMAREQSEL_UART1_0	S3C2412_DMAREQSEL_SRC(21)
+#define S3C2412_DMAREQSEL_UART1_1	S3C2412_DMAREQSEL_SRC(22)
+#define S3C2412_DMAREQSEL_UART2_0	S3C2412_DMAREQSEL_SRC(23)
+#define S3C2412_DMAREQSEL_UART2_1	S3C2412_DMAREQSEL_SRC(24)
+
+#endif
 #endif /* __ASM_ARCH_DMA_H */
-- 
cgit v1.2.3


From 98954df6917cb8f7e65f4f0f79ed641112fcf6b6 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:02:25 +0100
Subject: [ARM] 3816/1: iop3xx: rename config symbols

Rename CONFIG_ARCH_IOP321 to CONFIG_ARCH_IOP32X and
CONFIG_ARCH_IOP331 to CONFIG_ARCH_IOP33X.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/boot/compressed/head.S           |  2 +-
 arch/arm/configs/ep80219_defconfig        |  4 ++--
 arch/arm/configs/iq31244_defconfig        |  4 ++--
 arch/arm/configs/iq80321_defconfig        |  4 ++--
 arch/arm/configs/iq80331_defconfig        |  2 +-
 arch/arm/configs/iq80332_defconfig        |  2 +-
 arch/arm/mach-iop3xx/Kconfig              | 16 ++++++++--------
 arch/arm/mach-iop3xx/Makefile             |  4 ++--
 arch/arm/mach-iop3xx/Makefile.boot        |  2 +-
 arch/arm/mach-iop3xx/iop331-setup.c       |  2 +-
 arch/arm/oprofile/op_model_xscale.c       |  4 ++--
 drivers/i2c/busses/i2c-iop3xx.c           |  4 ++--
 include/asm-arm/arch-iop3xx/debug-macro.S |  2 +-
 include/asm-arm/arch-iop3xx/entry-macro.S |  4 ++--
 include/asm-arm/arch-iop3xx/iop321.h      |  2 +-
 include/asm-arm/arch-iop3xx/iop331.h      |  4 ++--
 include/asm-arm/arch-iop3xx/irqs.h        |  4 ++--
 include/asm-arm/arch-iop3xx/memory.h      |  6 +++---
 include/asm-arm/arch-iop3xx/system.h      |  4 ++--
 include/asm-arm/arch-iop3xx/uncompress.h  |  4 ++--
 20 files changed, 40 insertions(+), 40 deletions(-)

(limited to 'include')

diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
index 14a9ff9c68df..d7ecc7ece82d 100644
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -42,7 +42,7 @@
 		add	\rb, \rb, #0x00010000	@ Ser1
 #endif
 		.endm
-#elif defined(CONFIG_ARCH_IOP331)
+#elif defined(CONFIG_ARCH_IOP33X)
 		.macro loadsp, rb
                 mov   	\rb, #0xff000000
                 orr     \rb, \rb, #0x00ff0000
diff --git a/arch/arm/configs/ep80219_defconfig b/arch/arm/configs/ep80219_defconfig
index 3c73b707c2f3..199f1567345d 100644
--- a/arch/arm/configs/ep80219_defconfig
+++ b/arch/arm/configs/ep80219_defconfig
@@ -93,8 +93,8 @@ CONFIG_ARCH_IQ31244=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
 CONFIG_ARCH_EP80219=y
-CONFIG_ARCH_IOP321=y
-# CONFIG_ARCH_IOP331 is not set
+CONFIG_ARCH_IOP32X=y
+# CONFIG_ARCH_IOP33X is not set
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq31244_defconfig b/arch/arm/configs/iq31244_defconfig
index 32467160a6df..27f884417b90 100644
--- a/arch/arm/configs/iq31244_defconfig
+++ b/arch/arm/configs/iq31244_defconfig
@@ -94,8 +94,8 @@ CONFIG_ARCH_IQ31244=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP321=y
-# CONFIG_ARCH_IOP331 is not set
+CONFIG_ARCH_IOP32X=y
+# CONFIG_ARCH_IOP33X is not set
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq80321_defconfig b/arch/arm/configs/iq80321_defconfig
index b000da753c41..33f21a24cf4e 100644
--- a/arch/arm/configs/iq80321_defconfig
+++ b/arch/arm/configs/iq80321_defconfig
@@ -93,8 +93,8 @@ CONFIG_ARCH_IQ80321=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP321=y
-# CONFIG_ARCH_IOP331 is not set
+CONFIG_ARCH_IOP32X=y
+# CONFIG_ARCH_IOP33X is not set
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq80331_defconfig b/arch/arm/configs/iq80331_defconfig
index 46c79e1efe07..67a8bd7dba29 100644
--- a/arch/arm/configs/iq80331_defconfig
+++ b/arch/arm/configs/iq80331_defconfig
@@ -93,7 +93,7 @@ CONFIG_ARCH_IOP3XX=y
 CONFIG_ARCH_IQ80331=y
 # CONFIG_MACH_IQ80332 is not set
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP331=y
+CONFIG_ARCH_IOP33X=y
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq80332_defconfig b/arch/arm/configs/iq80332_defconfig
index 11959b705d82..c8dea45cb017 100644
--- a/arch/arm/configs/iq80332_defconfig
+++ b/arch/arm/configs/iq80332_defconfig
@@ -93,7 +93,7 @@ CONFIG_ARCH_IOP3XX=y
 # CONFIG_ARCH_IQ80331 is not set
 CONFIG_MACH_IQ80332=y
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP331=y
+CONFIG_ARCH_IOP33X=y
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/mach-iop3xx/Kconfig b/arch/arm/mach-iop3xx/Kconfig
index 4422f2388607..887c3753727f 100644
--- a/arch/arm/mach-iop3xx/Kconfig
+++ b/arch/arm/mach-iop3xx/Kconfig
@@ -6,48 +6,48 @@ comment "IOP3xx Platform Types"
 
 config ARCH_IQ80321
 	bool "Enable support for IQ80321"
-	select ARCH_IOP321
+	select ARCH_IOP32X
 	help
 	  Say Y here if you want to run your kernel on the Intel IQ80321
 	  evaluation kit for the IOP321 chipset.
 
 config ARCH_IQ31244
 	bool "Enable support for IQ31244"
-	select ARCH_IOP321
+	select ARCH_IOP32X
 	help
 	  Say Y here if you want to run your kernel on the Intel IQ31244
 	  evaluation kit for the IOP321 chipset.
 
 config ARCH_IQ80331
 	bool "Enable support for IQ80331"
-	select ARCH_IOP331
+	select ARCH_IOP33X
 	help
 	  Say Y here if you want to run your kernel on the Intel IQ80331
 	  evaluation kit for the IOP331 chipset.
 
 config MACH_IQ80332
 	bool "Enable support for IQ80332"
-	select ARCH_IOP331
+	select ARCH_IOP33X
 	help
 	  Say Y here if you want to run your kernel on the Intel IQ80332
 	  evaluation kit for the IOP332 chipset.
 
 config ARCH_EP80219
 	bool "Enable support for EP80219"
-	select ARCH_IOP321
+	select ARCH_IOP32X
 	select ARCH_IQ31244
 	help
 	  Say Y here if you want to run your kernel on the Intel EP80219
 	  evaluation kit for the Intel 80219 chipset (a IOP321 variant).
 
 # Which IOP variant are we running?
-config ARCH_IOP321
+config ARCH_IOP32X
 	bool
 	help
 	  The IQ80321 uses the IOP321 variant.
 	  The IQ31244 and EP80219 uses the IOP321 variant.
 
-config ARCH_IOP331
+config ARCH_IOP33X
 	bool
 	default ARCH_IQ80331
 	help
@@ -57,7 +57,7 @@ comment "IOP3xx Chipset Features"
 
 config IOP331_STEPD
 	bool "Chip stepping D of the IOP80331 processor or IOP80333"
-	depends on (ARCH_IOP331)
+	depends on ARCH_IOP33X
 	help
 	  Say Y here if you have StepD of the IOP80331 or IOP8033
 	  based platforms.
diff --git a/arch/arm/mach-iop3xx/Makefile b/arch/arm/mach-iop3xx/Makefile
index b17eb1f46102..ad3051524480 100644
--- a/arch/arm/mach-iop3xx/Makefile
+++ b/arch/arm/mach-iop3xx/Makefile
@@ -10,9 +10,9 @@ obj-m			:=
 obj-n			:=
 obj-			:=
 
-obj-$(CONFIG_ARCH_IOP321)  += iop321-setup.o iop321-irq.o iop321-pci.o iop321-time.o
+obj-$(CONFIG_ARCH_IOP32X)  += iop321-setup.o iop321-irq.o iop321-pci.o iop321-time.o
 
-obj-$(CONFIG_ARCH_IOP331)  += iop331-setup.o iop331-irq.o iop331-pci.o iop331-time.o
+obj-$(CONFIG_ARCH_IOP33X)  += iop331-setup.o iop331-irq.o iop331-pci.o iop331-time.o
 
 obj-$(CONFIG_ARCH_IQ80321) += iq80321-mm.o iq80321-pci.o
 
diff --git a/arch/arm/mach-iop3xx/Makefile.boot b/arch/arm/mach-iop3xx/Makefile.boot
index 6387aa20461b..e120c233b706 100644
--- a/arch/arm/mach-iop3xx/Makefile.boot
+++ b/arch/arm/mach-iop3xx/Makefile.boot
@@ -1,7 +1,7 @@
    zreladdr-y	:= 0xa0008000
 params_phys-y	:= 0xa0000100
 initrd_phys-y	:= 0xa0800000
-ifeq ($(CONFIG_ARCH_IOP331),y)
+ifeq ($(CONFIG_ARCH_IOP33X),y)
    zreladdr-y	:= 0x00008000
 params_phys-y	:= 0x00000100
 initrd_phys-y	:= 0x00800000
diff --git a/arch/arm/mach-iop3xx/iop331-setup.c b/arch/arm/mach-iop3xx/iop331-setup.c
index 3cc98d892ad4..b2f24931d143 100644
--- a/arch/arm/mach-iop3xx/iop331-setup.c
+++ b/arch/arm/mach-iop3xx/iop331-setup.c
@@ -177,7 +177,7 @@ void __init iop331_map_io(void)
 	iotable_init(iop331_std_desc, ARRAY_SIZE(iop331_std_desc));
 }
 
-#ifdef CONFIG_ARCH_IOP331
+#ifdef CONFIG_ARCH_IOP33X
 extern void iop331_init_irq(void);
 extern struct sys_timer iop331_timer;
 #endif
diff --git a/arch/arm/oprofile/op_model_xscale.c b/arch/arm/oprofile/op_model_xscale.c
index 34fdc733743b..7e0cc5b15b15 100644
--- a/arch/arm/oprofile/op_model_xscale.c
+++ b/arch/arm/oprofile/op_model_xscale.c
@@ -36,10 +36,10 @@
 #ifdef CONFIG_ARCH_IOP310
 #define XSCALE_PMU_IRQ  IRQ_XS80200_PMU
 #endif
-#ifdef CONFIG_ARCH_IOP321
+#ifdef CONFIG_ARCH_IOP32X
 #define XSCALE_PMU_IRQ  IRQ_IOP321_CORE_PMU
 #endif
-#ifdef CONFIG_ARCH_IOP331
+#ifdef CONFIG_ARCH_IOP33X
 #define XSCALE_PMU_IRQ  IRQ_IOP331_CORE_PMU
 #endif
 #ifdef CONFIG_ARCH_PXA
diff --git a/drivers/i2c/busses/i2c-iop3xx.c b/drivers/i2c/busses/i2c-iop3xx.c
index 48c56939c861..035417f5a003 100644
--- a/drivers/i2c/busses/i2c-iop3xx.c
+++ b/drivers/i2c/busses/i2c-iop3xx.c
@@ -82,9 +82,9 @@ iop3xx_i2c_enable(struct i2c_algo_iop3xx_data *iop3xx_adap)
 
 	/* 
 	 * Every time unit enable is asserted, GPOD needs to be cleared
-	 * on IOP321 to avoid data corruption on the bus.
+	 * on IOP32X to avoid data corruption on the bus.
 	 */
-#ifdef CONFIG_ARCH_IOP321
+#ifdef CONFIG_ARCH_IOP32X
 #define IOP321_GPOD_I2C0    0x00c0  /* clear these bits to enable ch0 */
 #define IOP321_GPOD_I2C1    0x0030  /* clear these bits to enable ch1 */
 
diff --git a/include/asm-arm/arch-iop3xx/debug-macro.S b/include/asm-arm/arch-iop3xx/debug-macro.S
index ce007e531994..dcc6856d14ff 100644
--- a/include/asm-arm/arch-iop3xx/debug-macro.S
+++ b/include/asm-arm/arch-iop3xx/debug-macro.S
@@ -15,7 +15,7 @@
 		mov	\rx, #0xfe000000	@ physical
 #if defined(CONFIG_ARCH_IQ80321) || defined(CONFIG_ARCH_IQ31244)
 		orr	\rx, \rx, #0x00800000	@ location of the UART
-#elif defined(CONFIG_ARCH_IOP331)
+#elif defined(CONFIG_ARCH_IOP33X)
 		mrc	p15, 0, \rx, c1, c0
 		tst	\rx, #1			@ MMU enabled?
 		moveq	\rx, #0x000fe000	@ Physical Base
diff --git a/include/asm-arm/arch-iop3xx/entry-macro.S b/include/asm-arm/arch-iop3xx/entry-macro.S
index 926668c098a5..f3db54637ad3 100644
--- a/include/asm-arm/arch-iop3xx/entry-macro.S
+++ b/include/asm-arm/arch-iop3xx/entry-macro.S
@@ -9,7 +9,7 @@
  */
 #include <asm/arch/irqs.h>
 
-#if defined(CONFIG_ARCH_IOP321)
+#if defined(CONFIG_ARCH_IOP32X)
 		.macro  disable_fiq
 		.endm
 
@@ -28,7 +28,7 @@
 1001:
 		.endm
 
-#elif defined(CONFIG_ARCH_IOP331)
+#elif defined(CONFIG_ARCH_IOP33X)
 		.macro  disable_fiq
 		.endm
 
diff --git a/include/asm-arm/arch-iop3xx/iop321.h b/include/asm-arm/arch-iop3xx/iop321.h
index f8df778a356f..d198d72a50a4 100644
--- a/include/asm-arm/arch-iop3xx/iop321.h
+++ b/include/asm-arm/arch-iop3xx/iop321.h
@@ -21,7 +21,7 @@
  * IOP3xx variants but behave slightly differently on each.
  */
 #ifndef __ASSEMBLY__
-#ifdef	CONFIG_ARCH_IOP321
+#ifdef	CONFIG_ARCH_IOP32X
 #define iop_is_321() (((processor_id & 0xfffff5e0) == 0x69052420))
 #else
 #define	iop_is_321()	0
diff --git a/include/asm-arm/arch-iop3xx/iop331.h b/include/asm-arm/arch-iop3xx/iop331.h
index fbf0cc11bdd9..4d7bcc62cb3e 100644
--- a/include/asm-arm/arch-iop3xx/iop331.h
+++ b/include/asm-arm/arch-iop3xx/iop331.h
@@ -20,7 +20,7 @@
  * IOP3xx variants but behave slightly differently on each.
  */
 #ifndef __ASSEMBLY__
-#ifdef	CONFIG_ARCH_IOP331
+#ifdef	CONFIG_ARCH_IOP33X
 /*#define	iop_is_331()	((processor_id & 0xffffffb0) == 0x69054090) */
 #define	iop_is_331()	((processor_id & 0xffffff30) == 0x69054010)
 #else
@@ -257,7 +257,7 @@
 #define IOP331_TU_TISR    (volatile u32 *)IOP331_REG_ADDR(0x000007E8)
 #define IOP331_TU_WDTCR   (volatile u32 *)IOP331_REG_ADDR(0x000007EC)
 
-#if defined(CONFIG_ARCH_IOP331)
+#if defined(CONFIG_ARCH_IOP33X)
 #define	IOP331_TICK_RATE	266000000	/* 266 MHz IB clock */
 #endif
 
diff --git a/include/asm-arm/arch-iop3xx/irqs.h b/include/asm-arm/arch-iop3xx/irqs.h
index b2c03f4c269c..4f7c7aa87b4a 100644
--- a/include/asm-arm/arch-iop3xx/irqs.h
+++ b/include/asm-arm/arch-iop3xx/irqs.h
@@ -12,10 +12,10 @@
 /*
  * Chipset-specific bits
  */
-#ifdef CONFIG_ARCH_IOP321
+#ifdef CONFIG_ARCH_IOP32X
 #include "iop321-irqs.h"
 #endif
 
-#ifdef CONFIG_ARCH_IOP331
+#ifdef CONFIG_ARCH_IOP33X
 #include "iop331-irqs.h"
 #endif
diff --git a/include/asm-arm/arch-iop3xx/memory.h b/include/asm-arm/arch-iop3xx/memory.h
index e43ebd984745..25666184e8fc 100644
--- a/include/asm-arm/arch-iop3xx/memory.h
+++ b/include/asm-arm/arch-iop3xx/memory.h
@@ -10,7 +10,7 @@
 /*
  * Physical DRAM offset.
  */
-#ifndef CONFIG_ARCH_IOP331
+#ifndef CONFIG_ARCH_IOP33X
 #define PHYS_OFFSET	UL(0xa0000000)
 #else
 #define PHYS_OFFSET	UL(0x00000000)
@@ -23,12 +23,12 @@
  * bus_to_virt: Used to convert an address for DMA operations
  *		to an address that the kernel can use.
  */
-#if defined(CONFIG_ARCH_IOP321)
+#if defined(CONFIG_ARCH_IOP32X)
 
 #define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP321_IATVR2)) | ((*IOP321_IABAR2) & 0xfffffff0))
 #define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP321_IALR2)) | ( *IOP321_IATVR2)))
 
-#elif defined(CONFIG_ARCH_IOP331)
+#elif defined(CONFIG_ARCH_IOP33X)
 
 #define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP331_IATVR2)) | ((*IOP331_IABAR2) & 0xfffffff0))
 #define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP331_IALR2)) | ( *IOP331_IATVR2)))
diff --git a/include/asm-arm/arch-iop3xx/system.h b/include/asm-arm/arch-iop3xx/system.h
index af6ae8cd36c9..a16cbb77a7f6 100644
--- a/include/asm-arm/arch-iop3xx/system.h
+++ b/include/asm-arm/arch-iop3xx/system.h
@@ -16,11 +16,11 @@ static inline void arch_idle(void)
 
 static inline void arch_reset(char mode)
 {
-#ifdef CONFIG_ARCH_IOP321
+#ifdef CONFIG_ARCH_IOP32X
     *IOP321_PCSR = 0x30;
 #endif
 
-#ifdef CONFIG_ARCH_IOP331
+#ifdef CONFIG_ARCH_IOP33X
     *IOP331_PCSR = 0x30;
 #endif
 
diff --git a/include/asm-arm/arch-iop3xx/uncompress.h b/include/asm-arm/arch-iop3xx/uncompress.h
index fbdd5af644fe..066c16bc1250 100644
--- a/include/asm-arm/arch-iop3xx/uncompress.h
+++ b/include/asm-arm/arch-iop3xx/uncompress.h
@@ -6,9 +6,9 @@
 #include <linux/serial_reg.h>
 #include <asm/hardware.h>
 
-#ifdef CONFIG_ARCH_IOP321
+#ifdef CONFIG_ARCH_IOP32X
 #define UTYPE unsigned char *
-#elif defined(CONFIG_ARCH_IOP331)
+#elif defined(CONFIG_ARCH_IOP33X)
 #define UTYPE u32 *
 #else
 #error "Missing IOP3xx arch type def"
-- 
cgit v1.2.3


From 3f7e5815f4b774270e6506962de37af85aa9c830 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:10:26 +0100
Subject: [ARM] 3817/1: iop3xx: split the iop3xx mach into iop32x and iop33x

Split the iop3xx mach type into iop32x and iop33x -- split the config
symbols, and move the code in the mach-iop3xx directory to the mach-iop32x
and mach-iop33x directories.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                          |  20 +-
 arch/arm/Makefile                         |   3 +-
 arch/arm/configs/ep80219_defconfig        |   5 +-
 arch/arm/configs/iq31244_defconfig        |   5 +-
 arch/arm/configs/iq80321_defconfig        |   5 +-
 arch/arm/configs/iq80331_defconfig        |   4 +-
 arch/arm/configs/iq80332_defconfig        |   4 +-
 arch/arm/mach-iop32x/Kconfig              |  28 +++
 arch/arm/mach-iop32x/Makefile             |  11 +
 arch/arm/mach-iop32x/Makefile.boot        |   3 +
 arch/arm/mach-iop32x/common.c             |  72 ++++++
 arch/arm/mach-iop32x/iq31244-mm.c         |  45 ++++
 arch/arm/mach-iop32x/iq31244-pci.c        | 129 +++++++++++
 arch/arm/mach-iop32x/iq80321-mm.c         |  45 ++++
 arch/arm/mach-iop32x/iq80321-pci.c        | 123 ++++++++++
 arch/arm/mach-iop32x/irq.c                |  97 ++++++++
 arch/arm/mach-iop32x/pci.c                | 220 ++++++++++++++++++
 arch/arm/mach-iop32x/setup.c              | 173 ++++++++++++++
 arch/arm/mach-iop32x/time.c               | 108 +++++++++
 arch/arm/mach-iop33x/Kconfig              |  27 +++
 arch/arm/mach-iop33x/Makefile             |  11 +
 arch/arm/mach-iop33x/Makefile.boot        |   3 +
 arch/arm/mach-iop33x/common.c             |  22 ++
 arch/arm/mach-iop33x/iq80331-mm.c         |  35 +++
 arch/arm/mach-iop33x/iq80331-pci.c        | 119 ++++++++++
 arch/arm/mach-iop33x/iq80332-mm.c         |  35 +++
 arch/arm/mach-iop33x/iq80332-pci.c        | 125 ++++++++++
 arch/arm/mach-iop33x/irq.c                | 129 +++++++++++
 arch/arm/mach-iop33x/pci.c                | 222 ++++++++++++++++++
 arch/arm/mach-iop33x/setup.c              | 221 ++++++++++++++++++
 arch/arm/mach-iop33x/time.c               | 106 +++++++++
 arch/arm/mach-iop3xx/Kconfig              |  66 ------
 arch/arm/mach-iop3xx/Makefile             |  23 --
 arch/arm/mach-iop3xx/Makefile.boot        |   9 -
 arch/arm/mach-iop3xx/common.c             |  72 ------
 arch/arm/mach-iop3xx/iop321-irq.c         |  97 --------
 arch/arm/mach-iop3xx/iop321-pci.c         | 220 ------------------
 arch/arm/mach-iop3xx/iop321-setup.c       | 173 --------------
 arch/arm/mach-iop3xx/iop321-time.c        | 108 ---------
 arch/arm/mach-iop3xx/iop331-irq.c         | 129 -----------
 arch/arm/mach-iop3xx/iop331-pci.c         | 222 ------------------
 arch/arm/mach-iop3xx/iop331-setup.c       | 221 ------------------
 arch/arm/mach-iop3xx/iop331-time.c        | 106 ---------
 arch/arm/mach-iop3xx/iq31244-mm.c         |  45 ----
 arch/arm/mach-iop3xx/iq31244-pci.c        | 129 -----------
 arch/arm/mach-iop3xx/iq80321-mm.c         |  45 ----
 arch/arm/mach-iop3xx/iq80321-pci.c        | 123 ----------
 arch/arm/mach-iop3xx/iq80331-mm.c         |  35 ---
 arch/arm/mach-iop3xx/iq80331-pci.c        | 119 ----------
 arch/arm/mach-iop3xx/iq80332-mm.c         |  35 ---
 arch/arm/mach-iop3xx/iq80332-pci.c        | 125 ----------
 arch/arm/mm/Kconfig                       |   2 +-
 drivers/i2c/busses/Kconfig                |   2 +-
 include/asm-arm/arch-iop32x/debug-macro.S |  20 ++
 include/asm-arm/arch-iop32x/dma.h         |   9 +
 include/asm-arm/arch-iop32x/entry-macro.S |  28 +++
 include/asm-arm/arch-iop32x/hardware.h    |  54 +++++
 include/asm-arm/arch-iop32x/io.h          |  21 ++
 include/asm-arm/arch-iop32x/iop321.h      | 341 ++++++++++++++++++++++++++++
 include/asm-arm/arch-iop32x/iq31244.h     |  24 ++
 include/asm-arm/arch-iop32x/iq80321.h     |  24 ++
 include/asm-arm/arch-iop32x/irqs.h        |  98 ++++++++
 include/asm-arm/arch-iop32x/memory.h      |  27 +++
 include/asm-arm/arch-iop32x/system.h      |  29 +++
 include/asm-arm/arch-iop32x/timex.h       |   8 +
 include/asm-arm/arch-iop32x/uncompress.h  |  38 ++++
 include/asm-arm/arch-iop32x/vmalloc.h     |  16 ++
 include/asm-arm/arch-iop33x/debug-macro.S |  24 ++
 include/asm-arm/arch-iop33x/dma.h         |   9 +
 include/asm-arm/arch-iop33x/entry-macro.S |  34 +++
 include/asm-arm/arch-iop33x/hardware.h    |  54 +++++
 include/asm-arm/arch-iop33x/io.h          |  21 ++
 include/asm-arm/arch-iop33x/iop331.h      | 358 +++++++++++++++++++++++++++++
 include/asm-arm/arch-iop33x/iq80331.h     |  23 ++
 include/asm-arm/arch-iop33x/iq80332.h     |  23 ++
 include/asm-arm/arch-iop33x/irqs.h        | 130 +++++++++++
 include/asm-arm/arch-iop33x/memory.h      |  26 +++
 include/asm-arm/arch-iop33x/system.h      |  29 +++
 include/asm-arm/arch-iop33x/timex.h       |   8 +
 include/asm-arm/arch-iop33x/uncompress.h  |  36 +++
 include/asm-arm/arch-iop33x/vmalloc.h     |  16 ++
 include/asm-arm/arch-iop3xx/debug-macro.S |  35 ---
 include/asm-arm/arch-iop3xx/dma.h         |   9 -
 include/asm-arm/arch-iop3xx/entry-macro.S |  57 -----
 include/asm-arm/arch-iop3xx/hardware.h    |  57 -----
 include/asm-arm/arch-iop3xx/io.h          |  21 --
 include/asm-arm/arch-iop3xx/iop321-irqs.h | 100 --------
 include/asm-arm/arch-iop3xx/iop321.h      | 345 ----------------------------
 include/asm-arm/arch-iop3xx/iop331-irqs.h | 132 -----------
 include/asm-arm/arch-iop3xx/iop331.h      | 363 ------------------------------
 include/asm-arm/arch-iop3xx/iq31244.h     |  24 --
 include/asm-arm/arch-iop3xx/iq80321.h     |  24 --
 include/asm-arm/arch-iop3xx/iq80331.h     |  23 --
 include/asm-arm/arch-iop3xx/iq80332.h     |  23 --
 include/asm-arm/arch-iop3xx/irqs.h        |  21 --
 include/asm-arm/arch-iop3xx/memory.h      |  38 ----
 include/asm-arm/arch-iop3xx/system.h      |  35 ---
 include/asm-arm/arch-iop3xx/timex.h       |  20 --
 include/asm-arm/arch-iop3xx/uncompress.h  |  48 ----
 include/asm-arm/arch-iop3xx/vmalloc.h     |  16 --
 100 files changed, 3666 insertions(+), 3514 deletions(-)
 create mode 100644 arch/arm/mach-iop32x/Kconfig
 create mode 100644 arch/arm/mach-iop32x/Makefile
 create mode 100644 arch/arm/mach-iop32x/Makefile.boot
 create mode 100644 arch/arm/mach-iop32x/common.c
 create mode 100644 arch/arm/mach-iop32x/iq31244-mm.c
 create mode 100644 arch/arm/mach-iop32x/iq31244-pci.c
 create mode 100644 arch/arm/mach-iop32x/iq80321-mm.c
 create mode 100644 arch/arm/mach-iop32x/iq80321-pci.c
 create mode 100644 arch/arm/mach-iop32x/irq.c
 create mode 100644 arch/arm/mach-iop32x/pci.c
 create mode 100644 arch/arm/mach-iop32x/setup.c
 create mode 100644 arch/arm/mach-iop32x/time.c
 create mode 100644 arch/arm/mach-iop33x/Kconfig
 create mode 100644 arch/arm/mach-iop33x/Makefile
 create mode 100644 arch/arm/mach-iop33x/Makefile.boot
 create mode 100644 arch/arm/mach-iop33x/common.c
 create mode 100644 arch/arm/mach-iop33x/iq80331-mm.c
 create mode 100644 arch/arm/mach-iop33x/iq80331-pci.c
 create mode 100644 arch/arm/mach-iop33x/iq80332-mm.c
 create mode 100644 arch/arm/mach-iop33x/iq80332-pci.c
 create mode 100644 arch/arm/mach-iop33x/irq.c
 create mode 100644 arch/arm/mach-iop33x/pci.c
 create mode 100644 arch/arm/mach-iop33x/setup.c
 create mode 100644 arch/arm/mach-iop33x/time.c
 delete mode 100644 arch/arm/mach-iop3xx/Kconfig
 delete mode 100644 arch/arm/mach-iop3xx/Makefile
 delete mode 100644 arch/arm/mach-iop3xx/Makefile.boot
 delete mode 100644 arch/arm/mach-iop3xx/common.c
 delete mode 100644 arch/arm/mach-iop3xx/iop321-irq.c
 delete mode 100644 arch/arm/mach-iop3xx/iop321-pci.c
 delete mode 100644 arch/arm/mach-iop3xx/iop321-setup.c
 delete mode 100644 arch/arm/mach-iop3xx/iop321-time.c
 delete mode 100644 arch/arm/mach-iop3xx/iop331-irq.c
 delete mode 100644 arch/arm/mach-iop3xx/iop331-pci.c
 delete mode 100644 arch/arm/mach-iop3xx/iop331-setup.c
 delete mode 100644 arch/arm/mach-iop3xx/iop331-time.c
 delete mode 100644 arch/arm/mach-iop3xx/iq31244-mm.c
 delete mode 100644 arch/arm/mach-iop3xx/iq31244-pci.c
 delete mode 100644 arch/arm/mach-iop3xx/iq80321-mm.c
 delete mode 100644 arch/arm/mach-iop3xx/iq80321-pci.c
 delete mode 100644 arch/arm/mach-iop3xx/iq80331-mm.c
 delete mode 100644 arch/arm/mach-iop3xx/iq80331-pci.c
 delete mode 100644 arch/arm/mach-iop3xx/iq80332-mm.c
 delete mode 100644 arch/arm/mach-iop3xx/iq80332-pci.c
 create mode 100644 include/asm-arm/arch-iop32x/debug-macro.S
 create mode 100644 include/asm-arm/arch-iop32x/dma.h
 create mode 100644 include/asm-arm/arch-iop32x/entry-macro.S
 create mode 100644 include/asm-arm/arch-iop32x/hardware.h
 create mode 100644 include/asm-arm/arch-iop32x/io.h
 create mode 100644 include/asm-arm/arch-iop32x/iop321.h
 create mode 100644 include/asm-arm/arch-iop32x/iq31244.h
 create mode 100644 include/asm-arm/arch-iop32x/iq80321.h
 create mode 100644 include/asm-arm/arch-iop32x/irqs.h
 create mode 100644 include/asm-arm/arch-iop32x/memory.h
 create mode 100644 include/asm-arm/arch-iop32x/system.h
 create mode 100644 include/asm-arm/arch-iop32x/timex.h
 create mode 100644 include/asm-arm/arch-iop32x/uncompress.h
 create mode 100644 include/asm-arm/arch-iop32x/vmalloc.h
 create mode 100644 include/asm-arm/arch-iop33x/debug-macro.S
 create mode 100644 include/asm-arm/arch-iop33x/dma.h
 create mode 100644 include/asm-arm/arch-iop33x/entry-macro.S
 create mode 100644 include/asm-arm/arch-iop33x/hardware.h
 create mode 100644 include/asm-arm/arch-iop33x/io.h
 create mode 100644 include/asm-arm/arch-iop33x/iop331.h
 create mode 100644 include/asm-arm/arch-iop33x/iq80331.h
 create mode 100644 include/asm-arm/arch-iop33x/iq80332.h
 create mode 100644 include/asm-arm/arch-iop33x/irqs.h
 create mode 100644 include/asm-arm/arch-iop33x/memory.h
 create mode 100644 include/asm-arm/arch-iop33x/system.h
 create mode 100644 include/asm-arm/arch-iop33x/timex.h
 create mode 100644 include/asm-arm/arch-iop33x/uncompress.h
 create mode 100644 include/asm-arm/arch-iop33x/vmalloc.h
 delete mode 100644 include/asm-arm/arch-iop3xx/debug-macro.S
 delete mode 100644 include/asm-arm/arch-iop3xx/dma.h
 delete mode 100644 include/asm-arm/arch-iop3xx/entry-macro.S
 delete mode 100644 include/asm-arm/arch-iop3xx/hardware.h
 delete mode 100644 include/asm-arm/arch-iop3xx/io.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iop321-irqs.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iop321.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iop331-irqs.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iop331.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iq31244.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iq80321.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iq80331.h
 delete mode 100644 include/asm-arm/arch-iop3xx/iq80332.h
 delete mode 100644 include/asm-arm/arch-iop3xx/irqs.h
 delete mode 100644 include/asm-arm/arch-iop3xx/memory.h
 delete mode 100644 include/asm-arm/arch-iop3xx/system.h
 delete mode 100644 include/asm-arm/arch-iop3xx/timex.h
 delete mode 100644 include/asm-arm/arch-iop3xx/uncompress.h
 delete mode 100644 include/asm-arm/arch-iop3xx/vmalloc.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index d9376f048ed7..0fb17bbb86bc 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -202,12 +202,20 @@ config ARCH_IMX
 	help
 	  Support for Motorola's i.MX family of processors (MX1, MXL).
 
-config ARCH_IOP3XX
-	bool "IOP3xx-based"
+config ARCH_IOP32X
+	bool "IOP32x-based"
 	depends on MMU
 	select PCI
 	help
-	  Support for Intel's IOP3XX (XScale) family of processors.
+	  Support for Intel's 80219 and IOP32X (XScale) family of
+	  processors.
+
+config ARCH_IOP33X
+	bool "IOP33x-based"
+	depends on MMU
+	select PCI
+	help
+	  Support for Intel's IOP33X (XScale) family of processors.
 
 config ARCH_IXP4XX
 	bool "IXP4xx-based"
@@ -312,7 +320,9 @@ source "arch/arm/mach-footbridge/Kconfig"
 
 source "arch/arm/mach-integrator/Kconfig"
 
-source "arch/arm/mach-iop3xx/Kconfig"
+source "arch/arm/mach-iop32x/Kconfig"
+
+source "arch/arm/mach-iop33x/Kconfig"
 
 source "arch/arm/mach-ixp4xx/Kconfig"
 
@@ -848,7 +858,7 @@ source "drivers/block/Kconfig"
 
 source "drivers/acorn/block/Kconfig"
 
-if PCMCIA || ARCH_CLPS7500 || ARCH_IOP3XX || ARCH_IXP4XX \
+if PCMCIA || ARCH_CLPS7500 || ARCH_IOP32X || ARCH_IOP33X || ARCH_IXP4XX \
 	|| ARCH_L7200 || ARCH_LH7A40X || ARCH_PXA || ARCH_RPC \
 	|| ARCH_S3C2410 || ARCH_SA1100 || ARCH_SHARK || FOOTBRIDGE \
 	|| ARCH_IXP23XX
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 92873cdee31f..ac786bf0f546 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -101,7 +101,8 @@ endif
  machine-$(CONFIG_ARCH_INTEGRATOR) := integrator
  textofs-$(CONFIG_ARCH_CLPS711X)   := 0x00028000
  machine-$(CONFIG_ARCH_CLPS711X)   := clps711x
- machine-$(CONFIG_ARCH_IOP3XX)	   := iop3xx
+ machine-$(CONFIG_ARCH_IOP32X)	   := iop32x
+ machine-$(CONFIG_ARCH_IOP33X)	   := iop33x
  machine-$(CONFIG_ARCH_IXP4XX)	   := ixp4xx
  machine-$(CONFIG_ARCH_IXP2000)    := ixp2000
  machine-$(CONFIG_ARCH_IXP23XX)    := ixp23xx
diff --git a/arch/arm/configs/ep80219_defconfig b/arch/arm/configs/ep80219_defconfig
index 199f1567345d..f91cf8ff359d 100644
--- a/arch/arm/configs/ep80219_defconfig
+++ b/arch/arm/configs/ep80219_defconfig
@@ -66,7 +66,8 @@ CONFIG_KMOD=y
 # CONFIG_ARCH_EBSA110 is not set
 # CONFIG_ARCH_FOOTBRIDGE is not set
 # CONFIG_ARCH_INTEGRATOR is not set
-CONFIG_ARCH_IOP3XX=y
+CONFIG_ARCH_IOP32X=y
+# CONFIG_ARCH_IOP33X is not set
 # CONFIG_ARCH_IXP4XX is not set
 # CONFIG_ARCH_IXP2000 is not set
 # CONFIG_ARCH_L7200 is not set
@@ -93,8 +94,6 @@ CONFIG_ARCH_IQ31244=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
 CONFIG_ARCH_EP80219=y
-CONFIG_ARCH_IOP32X=y
-# CONFIG_ARCH_IOP33X is not set
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq31244_defconfig b/arch/arm/configs/iq31244_defconfig
index 27f884417b90..ce1b1f20b763 100644
--- a/arch/arm/configs/iq31244_defconfig
+++ b/arch/arm/configs/iq31244_defconfig
@@ -67,7 +67,8 @@ CONFIG_KMOD=y
 # CONFIG_ARCH_EBSA110 is not set
 # CONFIG_ARCH_FOOTBRIDGE is not set
 # CONFIG_ARCH_INTEGRATOR is not set
-CONFIG_ARCH_IOP3XX=y
+CONFIG_ARCH_IOP32X=y
+# CONFIG_ARCH_IOP33X is not set
 # CONFIG_ARCH_IXP4XX is not set
 # CONFIG_ARCH_IXP2000 is not set
 # CONFIG_ARCH_L7200 is not set
@@ -94,8 +95,6 @@ CONFIG_ARCH_IQ31244=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP32X=y
-# CONFIG_ARCH_IOP33X is not set
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq80321_defconfig b/arch/arm/configs/iq80321_defconfig
index 33f21a24cf4e..f00b0d2159dd 100644
--- a/arch/arm/configs/iq80321_defconfig
+++ b/arch/arm/configs/iq80321_defconfig
@@ -66,7 +66,8 @@ CONFIG_KMOD=y
 # CONFIG_ARCH_EBSA110 is not set
 # CONFIG_ARCH_FOOTBRIDGE is not set
 # CONFIG_ARCH_INTEGRATOR is not set
-CONFIG_ARCH_IOP3XX=y
+CONFIG_ARCH_IOP32X=y
+# CONFIG_ARCH_IOP33X is not set
 # CONFIG_ARCH_IXP4XX is not set
 # CONFIG_ARCH_IXP2000 is not set
 # CONFIG_ARCH_L7200 is not set
@@ -93,8 +94,6 @@ CONFIG_ARCH_IQ80321=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP32X=y
-# CONFIG_ARCH_IOP33X is not set
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq80331_defconfig b/arch/arm/configs/iq80331_defconfig
index 67a8bd7dba29..af3a87e469d5 100644
--- a/arch/arm/configs/iq80331_defconfig
+++ b/arch/arm/configs/iq80331_defconfig
@@ -66,7 +66,8 @@ CONFIG_KMOD=y
 # CONFIG_ARCH_EBSA110 is not set
 # CONFIG_ARCH_FOOTBRIDGE is not set
 # CONFIG_ARCH_INTEGRATOR is not set
-CONFIG_ARCH_IOP3XX=y
+# CONFIG_ARCH_IOP32X is not set
+CONFIG_ARCH_IOP33X=y
 # CONFIG_ARCH_IXP4XX is not set
 # CONFIG_ARCH_IXP2000 is not set
 # CONFIG_ARCH_L7200 is not set
@@ -93,7 +94,6 @@ CONFIG_ARCH_IOP3XX=y
 CONFIG_ARCH_IQ80331=y
 # CONFIG_MACH_IQ80332 is not set
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP33X=y
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/configs/iq80332_defconfig b/arch/arm/configs/iq80332_defconfig
index c8dea45cb017..931c78755a30 100644
--- a/arch/arm/configs/iq80332_defconfig
+++ b/arch/arm/configs/iq80332_defconfig
@@ -66,7 +66,8 @@ CONFIG_KMOD=y
 # CONFIG_ARCH_EBSA110 is not set
 # CONFIG_ARCH_FOOTBRIDGE is not set
 # CONFIG_ARCH_INTEGRATOR is not set
-CONFIG_ARCH_IOP3XX=y
+# CONFIG_ARCH_IOP32X is not set
+CONFIG_ARCH_IOP33X=y
 # CONFIG_ARCH_IXP4XX is not set
 # CONFIG_ARCH_IXP2000 is not set
 # CONFIG_ARCH_L7200 is not set
@@ -93,7 +94,6 @@ CONFIG_ARCH_IOP3XX=y
 # CONFIG_ARCH_IQ80331 is not set
 CONFIG_MACH_IQ80332=y
 # CONFIG_ARCH_EP80219 is not set
-CONFIG_ARCH_IOP33X=y
 
 #
 # IOP3xx Chipset Features
diff --git a/arch/arm/mach-iop32x/Kconfig b/arch/arm/mach-iop32x/Kconfig
new file mode 100644
index 000000000000..21e3e8c8c2ad
--- /dev/null
+++ b/arch/arm/mach-iop32x/Kconfig
@@ -0,0 +1,28 @@
+if ARCH_IOP32X
+
+menu "IOP32x Implementation Options"
+
+comment "IOP32x Platform Types"
+
+config ARCH_IQ80321
+	bool "Enable support for IQ80321"
+	help
+	  Say Y here if you want to run your kernel on the Intel IQ80321
+	  evaluation kit for the IOP321 chipset.
+
+config ARCH_IQ31244
+	bool "Enable support for IQ31244"
+	help
+	  Say Y here if you want to run your kernel on the Intel IQ31244
+	  evaluation kit for the IOP321 chipset.
+
+config ARCH_EP80219
+	bool "Enable support for EP80219"
+	select ARCH_IQ31244
+	help
+	  Say Y here if you want to run your kernel on the Intel EP80219
+	  evaluation kit for the Intel 80219 chipset (a IOP321 variant).
+
+endmenu
+
+endif
diff --git a/arch/arm/mach-iop32x/Makefile b/arch/arm/mach-iop32x/Makefile
new file mode 100644
index 000000000000..94144f7efa6c
--- /dev/null
+++ b/arch/arm/mach-iop32x/Makefile
@@ -0,0 +1,11 @@
+#
+# Makefile for the linux kernel.
+#
+
+obj-y			:= common.o setup.o irq.o pci.o time.o
+obj-m			:=
+obj-n			:=
+obj-			:=
+
+obj-$(CONFIG_ARCH_IQ80321) += iq80321-mm.o iq80321-pci.o
+obj-$(CONFIG_ARCH_IQ31244) += iq31244-mm.o iq31244-pci.o
diff --git a/arch/arm/mach-iop32x/Makefile.boot b/arch/arm/mach-iop32x/Makefile.boot
new file mode 100644
index 000000000000..47000dccd61f
--- /dev/null
+++ b/arch/arm/mach-iop32x/Makefile.boot
@@ -0,0 +1,3 @@
+   zreladdr-y	:= 0xa0008000
+params_phys-y	:= 0xa0000100
+initrd_phys-y	:= 0xa0800000
diff --git a/arch/arm/mach-iop32x/common.c b/arch/arm/mach-iop32x/common.c
new file mode 100644
index 000000000000..8044af6a54e1
--- /dev/null
+++ b/arch/arm/mach-iop32x/common.c
@@ -0,0 +1,72 @@
+/*
+ * arch/arm/mach-iop32x/common.c
+ *
+ * Common routines shared across all IOP3xx implementations
+ *
+ * Author: Deepak Saxena <dsaxena@mvista.com>
+ *
+ * Copyright 2003 (c) MontaVista, Software, Inc.
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/delay.h>
+#include <asm/hardware.h>
+
+/*
+ * Shared variables
+ */
+unsigned long iop3xx_pcibios_min_io = 0;
+unsigned long iop3xx_pcibios_min_mem = 0;
+
+#ifdef CONFIG_ARCH_EP80219
+#include <linux/kernel.h>
+/*
+ * Default power-off for EP80219
+ */
+
+static inline void ep80219_send_to_pic(__u8 c) {
+}
+
+void ep80219_power_off(void)
+{
+	/*
+     * This function will send a SHUTDOWN_COMPLETE message to the PIC controller
+     * over I2C.  We are not using the i2c subsystem since we are going to power
+     * off and it may be removed
+     */
+
+	/* Send the Address byte w/ the start condition */
+	*IOP321_IDBR1 = 0x60;
+	*IOP321_ICR1 = 0xE9;
+    mdelay(1);
+
+	/* Send the START_MSG byte w/ no start or stop condition */
+	*IOP321_IDBR1 = 0x0F;
+	*IOP321_ICR1 = 0xE8;
+    mdelay(1);
+
+	/* Send the SHUTDOWN_COMPLETE Message ID byte w/ no start or stop condition */
+	*IOP321_IDBR1 = 0x03;
+	*IOP321_ICR1 = 0xE8;
+    mdelay(1);
+
+	/* Send an ignored byte w/ stop condition */
+	*IOP321_IDBR1 = 0x00;
+	*IOP321_ICR1 = 0xEA;
+
+	while (1) ;
+}
+
+#include <linux/init.h>
+#include <linux/pm.h>
+
+static int __init ep80219_init(void)
+{
+	pm_power_off = ep80219_power_off;
+	return 0;
+}
+arch_initcall(ep80219_init);
+#endif
diff --git a/arch/arm/mach-iop32x/iq31244-mm.c b/arch/arm/mach-iop32x/iq31244-mm.c
new file mode 100644
index 000000000000..ffc6692c95ad
--- /dev/null
+++ b/arch/arm/mach-iop32x/iq31244-mm.c
@@ -0,0 +1,45 @@
+/*
+ * linux/arch/arm/mach-iop32x/iq31244-mm.c
+ *
+ * Low level memory initialization for iq80321 platform
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+
+#include <asm/mach/map.h>
+
+
+/*
+ * IQ80321 specific IO mappings
+ *
+ * We use RedBoot's setup for the onboard devices.
+ */
+static struct map_desc iq31244_io_desc[] __initdata = {
+	{	/* on-board devices */
+		.virtual	= IQ31244_UART,
+		.pfn		= __phys_to_pfn(IQ31244_UART),
+		.length		= 0x00100000,
+		.type		= MT_DEVICE
+	}
+};
+
+void __init iq31244_map_io(void)
+{
+	iop321_map_io();
+
+	iotable_init(iq31244_io_desc, ARRAY_SIZE(iq31244_io_desc));
+}
diff --git a/arch/arm/mach-iop32x/iq31244-pci.c b/arch/arm/mach-iop32x/iq31244-pci.c
new file mode 100644
index 000000000000..e447c8103ecd
--- /dev/null
+++ b/arch/arm/mach-iop32x/iq31244-pci.c
@@ -0,0 +1,129 @@
+/*
+ * arch/arm/mach-iop32x/iq31244-pci.c
+ *
+ * PCI support for the Intel IQ31244 reference board
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+
+#include <asm/hardware.h>
+#include <asm/irq.h>
+#include <asm/mach/pci.h>
+#include <asm/mach-types.h>
+
+/*
+ * The following macro is used to lookup irqs in a standard table
+ * format for those systems that do not already have PCI
+ * interrupts properly routed.  We assume 1 <= pin <= 4
+ */
+#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
+({ int _ctl_ = -1;				\
+   unsigned int _idsel = idsel - minid;		\
+   if (_idsel <= maxid)				\
+      _ctl_ = pci_irq_table[_idsel][pin-1];	\
+   _ctl_; })
+
+#define INTA	IRQ_IQ31244_INTA
+#define INTB	IRQ_IQ31244_INTB
+#define INTC	IRQ_IQ31244_INTC
+#define INTD	IRQ_IQ31244_INTD
+
+#define INTE	IRQ_IQ31244_I82546
+
+static inline int __init
+iq31244_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
+{
+	static int pci_irq_table[][4] = {
+		/*
+		 * PCI IDSEL/INTPIN->INTLINE
+		 * A       B       C       D
+		 */
+#ifdef CONFIG_ARCH_EP80219
+		{INTB, INTB, INTB, INTB}, /* CFlash */
+		{INTE, INTE, INTE, INTE}, /* 82551 Pro 100 */
+		{INTD, INTD, INTD, INTD}, /* PCI-X Slot */
+		{INTC, INTC, INTC, INTC}, /* SATA   */
+#else
+		{INTB, INTB, INTB, INTB}, /* CFlash */
+		{INTC, INTC, INTC, INTC}, /* SATA   */
+		{INTD, INTD, INTD, INTD}, /* PCI-X Slot */
+		{INTE, INTE, INTE, INTE}, /* 82546 GigE */
+#endif // CONFIG_ARCH_EP80219
+	};
+
+	BUG_ON(pin < 1 || pin > 4);
+
+	return PCI_IRQ_TABLE_LOOKUP(0, 7);
+}
+
+static int iq31244_setup(int nr, struct pci_sys_data *sys)
+{
+	struct resource *res;
+
+	if(nr != 0)
+		return 0;
+
+	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
+	if (!res)
+		panic("PCI: unable to alloc resources");
+
+	res[0].start = IOP321_PCI_LOWER_IO_VA;
+	res[0].end   = IOP321_PCI_UPPER_IO_VA;
+	res[0].name  = "IQ31244 PCI I/O Space";
+	res[0].flags = IORESOURCE_IO;
+
+	res[1].start = IOP321_PCI_LOWER_MEM_PA;
+	res[1].end   = IOP321_PCI_UPPER_MEM_PA;
+	res[1].name  = "IQ31244 PCI Memory Space";
+	res[1].flags = IORESOURCE_MEM;
+
+	request_resource(&ioport_resource, &res[0]);
+	request_resource(&iomem_resource, &res[1]);
+
+	sys->mem_offset = IOP321_PCI_MEM_OFFSET;
+	sys->io_offset  = IOP321_PCI_IO_OFFSET;
+
+	sys->resource[0] = &res[0];
+	sys->resource[1] = &res[1];
+	sys->resource[2] = NULL;
+
+	return 1;
+}
+
+static void iq31244_preinit(void)
+{
+	iop321_init();
+}
+
+static struct hw_pci iq31244_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iq31244_setup,
+	.scan		= iop321_scan_bus,
+	.preinit	= iq31244_preinit,
+	.map_irq	= iq31244_map_irq
+};
+
+static int __init iq31244_pci_init(void)
+{
+	if (machine_is_iq31244())
+		pci_common_init(&iq31244_pci);
+	return 0;
+}
+
+subsys_initcall(iq31244_pci_init);
+
+
+
+
diff --git a/arch/arm/mach-iop32x/iq80321-mm.c b/arch/arm/mach-iop32x/iq80321-mm.c
new file mode 100644
index 000000000000..5f702a7ff622
--- /dev/null
+++ b/arch/arm/mach-iop32x/iq80321-mm.c
@@ -0,0 +1,45 @@
+/*
+ * linux/arch/arm/mach-iop32x/iq80321-mm.c
+ *
+ * Low level memory initialization for iq80321 platform
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+
+#include <asm/mach/map.h>
+
+
+/*
+ * IQ80321 specific IO mappings
+ *
+ * We use RedBoot's setup for the onboard devices.
+ */
+static struct map_desc iq80321_io_desc[] __initdata = {
+ 	{	/* on-board devices */
+		.virtual	= IQ80321_UART,
+		.pfn		= __phys_to_pfn(IQ80321_UART),
+		.length		= 0x00100000,
+		.type		= MT_DEVICE
+	}
+};
+
+void __init iq80321_map_io(void)
+{
+	iop321_map_io();
+
+	iotable_init(iq80321_io_desc, ARRAY_SIZE(iq80321_io_desc));
+}
diff --git a/arch/arm/mach-iop32x/iq80321-pci.c b/arch/arm/mach-iop32x/iq80321-pci.c
new file mode 100644
index 000000000000..8767950e131f
--- /dev/null
+++ b/arch/arm/mach-iop32x/iq80321-pci.c
@@ -0,0 +1,123 @@
+/*
+ * arch/arm/mach-iop32x/iq80321-pci.c
+ *
+ * PCI support for the Intel IQ80321 reference board
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+
+#include <asm/hardware.h>
+#include <asm/irq.h>
+#include <asm/mach/pci.h>
+#include <asm/mach-types.h>
+
+/*
+ * The following macro is used to lookup irqs in a standard table
+ * format for those systems that do not already have PCI
+ * interrupts properly routed.  We assume 1 <= pin <= 4
+ */
+#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
+({ int _ctl_ = -1;				\
+   unsigned int _idsel = idsel - minid;		\
+   if (_idsel <= maxid)				\
+      _ctl_ = pci_irq_table[_idsel][pin-1];	\
+   _ctl_; })
+
+#define INTA	IRQ_IQ80321_INTA
+#define INTB	IRQ_IQ80321_INTB
+#define INTC	IRQ_IQ80321_INTC
+#define INTD	IRQ_IQ80321_INTD
+
+#define INTE	IRQ_IQ80321_I82544
+
+static inline int __init
+iq80321_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
+{
+	static int pci_irq_table[][4] = {
+		/*
+		 * PCI IDSEL/INTPIN->INTLINE
+		 * A       B       C       D
+		 */
+		{INTE, INTE, INTE, INTE}, /* Gig-E */
+		{-1, -1, -1, -1}, 	  /* Unused */
+		{INTC, INTD, INTA, INTB}, /* PCI-X Slot */
+		{-1, -1, -1, -1},
+	};
+
+	BUG_ON(pin < 1 || pin > 4);
+
+//	return PCI_IRQ_TABLE_LOOKUP(4, 7);
+	return pci_irq_table[idsel%4][pin-1];
+}
+
+static int iq80321_setup(int nr, struct pci_sys_data *sys)
+{
+	struct resource *res;
+
+	if(nr != 0)
+		return 0;
+
+	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
+	if (!res)
+		panic("PCI: unable to alloc resources");
+
+	res[0].start = IOP321_PCI_LOWER_IO_VA;
+	res[0].end   = IOP321_PCI_UPPER_IO_VA;
+	res[0].name  = "IQ80321 PCI I/O Space";
+	res[0].flags = IORESOURCE_IO;
+
+	res[1].start = IOP321_PCI_LOWER_MEM_PA;
+	res[1].end   = IOP321_PCI_UPPER_MEM_PA;
+	res[1].name  = "IQ80321 PCI Memory Space";
+	res[1].flags = IORESOURCE_MEM;
+
+	request_resource(&ioport_resource, &res[0]);
+	request_resource(&iomem_resource, &res[1]);
+
+	sys->mem_offset = IOP321_PCI_MEM_OFFSET;
+	sys->io_offset  = IOP321_PCI_IO_OFFSET;
+
+	sys->resource[0] = &res[0];
+	sys->resource[1] = &res[1];
+	sys->resource[2] = NULL;
+
+	return 1;
+}
+
+static void iq80321_preinit(void)
+{
+	iop321_init();
+}
+
+static struct hw_pci iq80321_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iq80321_setup,
+	.scan		= iop321_scan_bus,
+	.preinit	= iq80321_preinit,
+	.map_irq	= iq80321_map_irq
+};
+
+static int __init iq80321_pci_init(void)
+{
+	if (machine_is_iq80321())
+		pci_common_init(&iq80321_pci);
+	return 0;
+}
+
+subsys_initcall(iq80321_pci_init);
+
+
+
+
diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
new file mode 100644
index 000000000000..c5f633d2d632
--- /dev/null
+++ b/arch/arm/mach-iop32x/irq.c
@@ -0,0 +1,97 @@
+/*
+ * linux/arch/arm/mach-iop32x/irq.c
+ *
+ * Generic IOP32X IRQ handling functionality
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Added IOP3XX chipset and IQ80321 board masking code.
+ *
+ */
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+
+#include <asm/mach/irq.h>
+#include <asm/irq.h>
+#include <asm/hardware.h>
+
+#include <asm/mach-types.h>
+
+static u32 iop321_mask /* = 0 */;
+
+static inline void intctl_write(u32 val)
+{
+	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
+}
+
+static inline void intstr_write(u32 val)
+{
+	asm volatile("mcr p6,0,%0,c4,c0,0"::"r" (val));
+}
+
+static void
+iop321_irq_mask (unsigned int irq)
+{
+
+	iop321_mask &= ~(1 << (irq - IOP321_IRQ_OFS));
+
+	intctl_write(iop321_mask);
+}
+
+static void
+iop321_irq_unmask (unsigned int irq)
+{
+	iop321_mask |= (1 << (irq - IOP321_IRQ_OFS));
+
+	intctl_write(iop321_mask);
+}
+
+struct irq_chip ext_chip = {
+	.name	= "IOP",
+	.ack    = iop321_irq_mask,
+	.mask   = iop321_irq_mask,
+	.unmask = iop321_irq_unmask,
+};
+
+void __init iop321_init_irq(void)
+{
+	unsigned int i, tmp;
+
+	/* Enable access to coprocessor 6 for dealing with IRQs.
+	 * From RMK:
+	 * Basically, the Intel documentation here is poor.  It appears that
+	 * you need to set the bit to be able to access the coprocessor from
+	 * SVC mode.  Whether that allows access from user space or not is
+	 * unclear.
+	 */
+	asm volatile (
+		"mrc p15, 0, %0, c15, c1, 0\n\t"
+		"orr %0, %0, %1\n\t"
+		"mcr p15, 0, %0, c15, c1, 0\n\t"
+		/* The action is delayed, so we have to do this: */
+		"mrc p15, 0, %0, c15, c1, 0\n\t"
+		"mov %0, %0\n\t"
+		"sub pc, pc, #4"
+		: "=r" (tmp) : "i" (1 << 6) );
+
+	intctl_write(0);		// disable all interrupts
+	intstr_write(0);		// treat all as IRQ
+	if(machine_is_iq80321() ||
+	   machine_is_iq31244()) 	// all interrupts are inputs to chip
+		*IOP321_PCIIRSR = 0x0f;
+
+	for(i = IOP321_IRQ_OFS; i < NR_IRQS; i++)
+	{
+		set_irq_chip(i, &ext_chip);
+		set_irq_handler(i, do_level_IRQ);
+		set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
+
+	}
+}
+
diff --git a/arch/arm/mach-iop32x/pci.c b/arch/arm/mach-iop32x/pci.c
new file mode 100644
index 000000000000..1a7c683673f2
--- /dev/null
+++ b/arch/arm/mach-iop32x/pci.c
@@ -0,0 +1,220 @@
+/*
+ * arch/arm/mach-iop32x/pci.c
+ *
+ * PCI support for the Intel IOP321 chipset
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/hardware.h>
+#include <asm/mach/pci.h>
+
+#include <asm/arch/iop321.h>
+
+// #define DEBUG
+
+#ifdef DEBUG
+#define  DBG(x...) printk(x)
+#else
+#define  DBG(x...) do { } while (0)
+#endif
+
+/*
+ * This routine builds either a type0 or type1 configuration command.  If the
+ * bus is on the 80321 then a type0 made, else a type1 is created.
+ */
+static u32 iop321_cfg_address(struct pci_bus *bus, int devfn, int where)
+{
+	struct pci_sys_data *sys = bus->sysdata;
+	u32 addr;
+
+	if (sys->busnr == bus->number)
+		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
+	else
+		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
+
+	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
+
+	return addr;
+}
+
+/*
+ * This routine checks the status of the last configuration cycle.  If an error
+ * was detected it returns a 1, else it returns a 0.  The errors being checked
+ * are parity, master abort, target abort (master and target).  These types of
+ * errors occure during a config cycle where there is no device, like during
+ * the discovery stage.
+ */
+static int iop321_pci_status(void)
+{
+	unsigned int status;
+	int ret = 0;
+
+	/*
+	 * Check the status registers.
+	 */
+	status = *IOP321_ATUSR;
+	if (status & 0xf900)
+	{
+		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
+		*IOP321_ATUSR = status & 0xf900;
+		ret = 1;
+	}
+	status = *IOP321_ATUISR;
+	if (status & 0x679f)
+	{
+		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
+		*IOP321_ATUISR = status & 0x679f;
+		ret = 1;
+	}
+	return ret;
+}
+
+/*
+ * Simply write the address register and read the configuration
+ * data.  Note that the 4 nop's ensure that we are able to handle
+ * a delayed abort (in theory.)
+ */
+static inline u32 iop321_read(unsigned long addr)
+{
+	u32 val;
+
+	__asm__ __volatile__(
+		"str	%1, [%2]\n\t"
+		"ldr	%0, [%3]\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		: "=r" (val)
+		: "r" (addr), "r" (IOP321_OCCAR), "r" (IOP321_OCCDR));
+
+	return val;
+}
+
+/*
+ * The read routines must check the error status of the last configuration
+ * cycle.  If there was an error, the routine returns all hex f's.
+ */
+static int
+iop321_read_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 *value)
+{
+	unsigned long addr = iop321_cfg_address(bus, devfn, where);
+	u32 val = iop321_read(addr) >> ((where & 3) * 8);
+
+	if( iop321_pci_status() )
+		val = 0xffffffff;
+
+	*value = val;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+iop321_write_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 value)
+{
+	unsigned long addr = iop321_cfg_address(bus, devfn, where);
+	u32 val;
+
+	if (size != 4) {
+		val = iop321_read(addr);
+		if (!iop321_pci_status() == 0)
+			return PCIBIOS_SUCCESSFUL;
+
+		where = (where & 3) * 8;
+
+		if (size == 1)
+			val &= ~(0xff << where);
+		else
+			val &= ~(0xffff << where);
+
+		*IOP321_OCCDR = val | value << where;
+	} else {
+		asm volatile(
+			"str	%1, [%2]\n\t"
+			"str	%0, [%3]\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			:
+			: "r" (value), "r" (addr),
+			  "r" (IOP321_OCCAR), "r" (IOP321_OCCDR));
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops iop321_ops = {
+	.read	= iop321_read_config,
+	.write	= iop321_write_config,
+};
+
+/*
+ * When a PCI device does not exist during config cycles, the 80200 gets a
+ * bus error instead of returning 0xffffffff. This handler simply returns.
+ */
+int
+iop321_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
+{
+	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
+		addr, fsr, regs->ARM_pc, regs->ARM_lr);
+
+	/*
+	 * If it was an imprecise abort, then we need to correct the
+	 * return address to be _after_ the instruction.
+	 */
+	if (fsr & (1 << 10))
+		regs->ARM_pc += 4;
+
+	return 0;
+}
+
+/*
+ * Scan an IOP321 PCI bus.  sys->bus defines which bus we scan.
+ */
+struct pci_bus *iop321_scan_bus(int nr, struct pci_sys_data *sys)
+{
+	return pci_scan_bus(sys->busnr, &iop321_ops, sys);
+}
+
+void iop321_init(void)
+{
+	DBG("PCI:  Intel 80321 PCI init code.\n");
+	DBG("ATU: IOP321_ATUCMD=0x%04x\n", *IOP321_ATUCMD);
+	DBG("ATU: IOP321_OMWTVR0=0x%04x, IOP321_OIOWTVR=0x%04x\n",
+			*IOP321_OMWTVR0,
+			*IOP321_OIOWTVR);
+	DBG("ATU: IOP321_ATUCR=0x%08x\n", *IOP321_ATUCR);
+	DBG("ATU: IOP321_IABAR0=0x%08x IOP321_IALR0=0x%08x IOP321_IATVR0=%08x\n",
+			*IOP321_IABAR0, *IOP321_IALR0, *IOP321_IATVR0);
+	DBG("ATU: IOP321_OMWTVR0=0x%08x\n", *IOP321_OMWTVR0);
+	DBG("ATU: IOP321_IABAR1=0x%08x IOP321_IALR1=0x%08x\n",
+			*IOP321_IABAR1, *IOP321_IALR1);
+	DBG("ATU: IOP321_ERBAR=0x%08x IOP321_ERLR=0x%08x IOP321_ERTVR=%08x\n",
+			*IOP321_ERBAR, *IOP321_ERLR, *IOP321_ERTVR);
+	DBG("ATU: IOP321_IABAR2=0x%08x IOP321_IALR2=0x%08x IOP321_IATVR2=%08x\n",
+			*IOP321_IABAR2, *IOP321_IALR2, *IOP321_IATVR2);
+	DBG("ATU: IOP321_IABAR3=0x%08x IOP321_IALR3=0x%08x IOP321_IATVR3=%08x\n",
+			*IOP321_IABAR3, *IOP321_IALR3, *IOP321_IATVR3);
+
+	hook_fault_code(16+6, iop321_pci_abort, SIGBUS, "imprecise external abort");
+}
+
diff --git a/arch/arm/mach-iop32x/setup.c b/arch/arm/mach-iop32x/setup.c
new file mode 100644
index 000000000000..729678799271
--- /dev/null
+++ b/arch/arm/mach-iop32x/setup.c
@@ -0,0 +1,173 @@
+/*
+ * linux/arch/arm/mach-iop32x/setup.c
+ *
+ * Author: Nicolas Pitre <nico@cam.org>
+ * Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+#include <linux/serial.h>
+#include <linux/tty.h>
+#include <linux/serial_core.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/mach/map.h>
+#include <asm/setup.h>
+#include <asm/system.h>
+#include <asm/memory.h>
+#include <asm/hardware.h>
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+
+#define IOP321_UART_XTAL 1843200
+
+/*
+ * Standard IO mapping for all IOP321 based systems
+ */
+static struct map_desc iop321_std_desc[] __initdata = {
+	 {	/* mem mapped registers */
+		.virtual	= IOP321_VIRT_MEM_BASE,
+		.pfn		= __phys_to_pfn(IOP321_PHYS_MEM_BASE),
+		.length		= 0x00002000,
+		.type		= MT_DEVICE
+	 }, {	/* PCI IO space */
+		.virtual	= IOP321_PCI_LOWER_IO_VA,
+		.pfn		= __phys_to_pfn(IOP321_PCI_LOWER_IO_PA),
+		.length		= IOP321_PCI_IO_WINDOW_SIZE,
+		.type		= MT_DEVICE
+	 }
+};
+
+#ifdef CONFIG_ARCH_IQ80321
+#define UARTBASE IQ80321_UART
+#define IRQ_UART IRQ_IQ80321_UART
+#endif
+
+#ifdef CONFIG_ARCH_IQ31244
+#define UARTBASE IQ31244_UART
+#define IRQ_UART IRQ_IQ31244_UART
+#endif
+
+static struct uart_port iop321_serial_ports[] = {
+	{
+		.membase	= (char*)(UARTBASE),
+		.mapbase	= (UARTBASE),
+		.irq		= IRQ_UART,
+		.flags		= UPF_SKIP_TEST,
+		.iotype		= UPIO_MEM,
+		.regshift	= 0,
+		.uartclk	= IOP321_UART_XTAL,
+		.line		= 0,
+		.type		= PORT_16550A,
+		.fifosize	= 16
+	}
+};
+
+static struct resource iop32x_i2c_0_resources[] = {
+	[0] = {
+		.start = 0xfffff680,
+		.end = 0xfffff698,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_IOP321_I2C_0,
+		.end = IRQ_IOP321_I2C_0,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource iop32x_i2c_1_resources[] = {
+	[0] = {
+		.start = 0xfffff6a0,
+		.end = 0xfffff6b8,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_IOP321_I2C_1,
+		.end = IRQ_IOP321_I2C_1,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct platform_device iop32x_i2c_0_controller = {
+	.name = "IOP3xx-I2C",
+	.id = 0,
+	.num_resources = 2,
+	.resource = iop32x_i2c_0_resources
+};
+
+static struct platform_device iop32x_i2c_1_controller = {
+	.name = "IOP3xx-I2C",
+	.id = 1,
+	.num_resources = 2,
+	.resource = iop32x_i2c_1_resources
+};
+
+static struct platform_device *iop32x_devices[] __initdata = {
+	&iop32x_i2c_0_controller,
+	&iop32x_i2c_1_controller
+};
+
+void __init iop32x_init(void)
+{
+	if(iop_is_321())
+	{
+		platform_add_devices(iop32x_devices,
+				ARRAY_SIZE(iop32x_devices));
+	}
+}
+
+void __init iop321_map_io(void)
+{
+	iotable_init(iop321_std_desc, ARRAY_SIZE(iop321_std_desc));
+	early_serial_setup(&iop321_serial_ports[0]);
+}
+
+#ifdef CONFIG_ARCH_IQ80321
+extern void iq80321_map_io(void);
+extern struct sys_timer iop321_timer;
+extern void iop321_init_time(void);
+#endif
+
+#ifdef CONFIG_ARCH_IQ31244
+extern void iq31244_map_io(void);
+extern struct sys_timer iop321_timer;
+extern void iop321_init_time(void);
+#endif
+
+#if defined(CONFIG_ARCH_IQ80321)
+MACHINE_START(IQ80321, "Intel IQ80321")
+	/* Maintainer: Intel Corporation */
+	.phys_io	= IQ80321_UART,
+	.io_pg_offst	= ((IQ80321_UART) >> 18) & 0xfffc,
+	.map_io		= iq80321_map_io,
+	.init_irq	= iop321_init_irq,
+	.timer		= &iop321_timer,
+	.boot_params	= 0xa0000100,
+	.init_machine	= iop32x_init,
+MACHINE_END
+#elif defined(CONFIG_ARCH_IQ31244)
+MACHINE_START(IQ31244, "Intel IQ31244")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= IQ31244_UART,
+	.io_pg_offst	= ((IQ31244_UART) >> 18) & 0xfffc,
+	.map_io		= iq31244_map_io,
+	.init_irq	= iop321_init_irq,
+	.timer		= &iop321_timer,
+	.boot_params	= 0xa0000100,
+	.init_machine	= iop32x_init,
+MACHINE_END
+#else
+#error No machine descriptor defined for this IOP3XX implementation
+#endif
diff --git a/arch/arm/mach-iop32x/time.c b/arch/arm/mach-iop32x/time.c
new file mode 100644
index 000000000000..3cabbbca0a02
--- /dev/null
+++ b/arch/arm/mach-iop32x/time.c
@@ -0,0 +1,108 @@
+/*
+ * arch/arm/mach-iop32x/time.c
+ *
+ * Timer code for IOP321 based systems
+ *
+ * Author: Deepak Saxena <dsaxena@mvista.com>
+ *
+ * Copyright 2002-2003 MontaVista Software Inc.
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/time.h>
+#include <linux/init.h>
+#include <linux/timex.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+#include <asm/mach/irq.h>
+#include <asm/mach/time.h>
+
+#define IOP321_TIME_SYNC 0
+
+static inline unsigned long get_elapsed(void)
+{
+	return LATCH - *IOP321_TU_TCR0;
+}
+
+static unsigned long iop321_gettimeoffset(void)
+{
+	unsigned long elapsed, usec;
+	u32 tisr1, tisr2;
+
+	/*
+	 * If an interrupt was pending before we read the timer,
+	 * we've already wrapped.  Factor this into the time.
+	 * If an interrupt was pending after we read the timer,
+	 * it may have wrapped between checking the interrupt
+	 * status and reading the timer.  Re-read the timer to
+	 * be sure its value is after the wrap.
+	 */
+
+	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr1));
+	elapsed = get_elapsed();
+	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr2));
+
+	if(tisr1 & 1)
+		elapsed += LATCH;
+	else if (tisr2 & 1)
+		elapsed = LATCH + get_elapsed();
+
+	/*
+	 * Now convert them to usec.
+	 */
+	usec = (unsigned long)(elapsed / (CLOCK_TICK_RATE/1000000));
+
+	return usec;
+}
+
+static irqreturn_t
+iop321_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	u32 tisr;
+
+	write_seqlock(&xtime_lock);
+
+	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr));
+	tisr |= 1;
+	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (tisr));
+
+	timer_tick(regs);
+
+	write_sequnlock(&xtime_lock);
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction iop321_timer_irq = {
+	.name		= "IOP321 Timer Tick",
+	.handler	= iop321_timer_interrupt,
+	.flags		= IRQF_DISABLED | IRQF_TIMER,
+};
+
+static void __init iop321_timer_init(void)
+{
+	u32 timer_ctl;
+
+	setup_irq(IRQ_IOP321_TIMER0, &iop321_timer_irq);
+
+	timer_ctl = IOP321_TMR_EN | IOP321_TMR_PRIVILEGED | IOP321_TMR_RELOAD |
+			IOP321_TMR_RATIO_1_1;
+
+	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (LATCH));
+
+	asm volatile("mcr p6, 0, %0, c0, c1, 0"	: : "r" (timer_ctl));
+}
+
+struct sys_timer iop321_timer = {
+	.init		= &iop321_timer_init,
+	.offset		= iop321_gettimeoffset,
+};
diff --git a/arch/arm/mach-iop33x/Kconfig b/arch/arm/mach-iop33x/Kconfig
new file mode 100644
index 000000000000..410df546e954
--- /dev/null
+++ b/arch/arm/mach-iop33x/Kconfig
@@ -0,0 +1,27 @@
+if ARCH_IOP33X
+
+menu "IOP33x Implementation Options"
+
+comment "IOP33x Platform Types"
+
+config ARCH_IQ80331
+	bool "Enable support for IQ80331"
+	help
+	  Say Y here if you want to run your kernel on the Intel IQ80331
+	  evaluation kit for the IOP331 chipset.
+
+config MACH_IQ80332
+	bool "Enable support for IQ80332"
+	help
+	  Say Y here if you want to run your kernel on the Intel IQ80332
+	  evaluation kit for the IOP332 chipset.
+
+config IOP331_STEPD
+	bool "Chip stepping D of the IOP80331 processor or IOP80333"
+	help
+	  Say Y here if you have StepD of the IOP80331 or IOP8033
+	  based platforms.
+
+endmenu
+
+endif
diff --git a/arch/arm/mach-iop33x/Makefile b/arch/arm/mach-iop33x/Makefile
new file mode 100644
index 000000000000..0f1a8fffe187
--- /dev/null
+++ b/arch/arm/mach-iop33x/Makefile
@@ -0,0 +1,11 @@
+#
+# Makefile for the linux kernel.
+#
+
+obj-y			:= common.o setup.o irq.o pci.o time.o
+obj-m			:=
+obj-n			:=
+obj-			:=
+
+obj-$(CONFIG_ARCH_IQ80331) += iq80331-mm.o iq80331-pci.o
+obj-$(CONFIG_MACH_IQ80332) += iq80332-mm.o iq80332-pci.o
diff --git a/arch/arm/mach-iop33x/Makefile.boot b/arch/arm/mach-iop33x/Makefile.boot
new file mode 100644
index 000000000000..67039c3e0c48
--- /dev/null
+++ b/arch/arm/mach-iop33x/Makefile.boot
@@ -0,0 +1,3 @@
+   zreladdr-y	:= 0x00008000
+params_phys-y	:= 0x00000100
+initrd_phys-y	:= 0x00800000
diff --git a/arch/arm/mach-iop33x/common.c b/arch/arm/mach-iop33x/common.c
new file mode 100644
index 000000000000..dd75f2b8e74f
--- /dev/null
+++ b/arch/arm/mach-iop33x/common.c
@@ -0,0 +1,22 @@
+/*
+ * arch/arm/mach-iop33x/common.c
+ *
+ * Common routines shared across all IOP3xx implementations
+ *
+ * Author: Deepak Saxena <dsaxena@mvista.com>
+ *
+ * Copyright 2003 (c) MontaVista, Software, Inc.
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/delay.h>
+#include <asm/hardware.h>
+
+/*
+ * Shared variables
+ */
+unsigned long iop3xx_pcibios_min_io = 0;
+unsigned long iop3xx_pcibios_min_mem = 0;
diff --git a/arch/arm/mach-iop33x/iq80331-mm.c b/arch/arm/mach-iop33x/iq80331-mm.c
new file mode 100644
index 000000000000..eb59445a0fca
--- /dev/null
+++ b/arch/arm/mach-iop33x/iq80331-mm.c
@@ -0,0 +1,35 @@
+/*
+ * linux/arch/arm/mach-iop33x/iq80331-mm.c
+ *
+ * Low level memory initialization for iq80331 platform
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2003 Intel Corp.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+
+#include <asm/mach/map.h>
+
+
+/*
+ * IQ80331 specific IO mappings
+ *
+ * We use RedBoot's setup for the onboard devices.
+ */
+
+void __init iq80331_map_io(void)
+{
+	iop331_map_io();
+}
diff --git a/arch/arm/mach-iop33x/iq80331-pci.c b/arch/arm/mach-iop33x/iq80331-pci.c
new file mode 100644
index 000000000000..71adb12a9421
--- /dev/null
+++ b/arch/arm/mach-iop33x/iq80331-pci.c
@@ -0,0 +1,119 @@
+/*
+ * arch/arm/mach-iop33x/iq80331-pci.c
+ *
+ * PCI support for the Intel IQ80331 reference board
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2003, 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+
+#include <asm/hardware.h>
+#include <asm/irq.h>
+#include <asm/mach/pci.h>
+#include <asm/mach-types.h>
+
+/*
+ * The following macro is used to lookup irqs in a standard table
+ * format for those systems that do not already have PCI
+ * interrupts properly routed.  We assume 1 <= pin <= 4
+ */
+#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
+({ int _ctl_ = -1;				\
+   unsigned int _idsel = idsel - minid;		\
+   if (_idsel <= maxid)				\
+      _ctl_ = pci_irq_table[_idsel][pin-1];	\
+   _ctl_; })
+
+#define INTA	IRQ_IQ80331_INTA
+#define INTB	IRQ_IQ80331_INTB
+#define INTC	IRQ_IQ80331_INTC
+#define INTD	IRQ_IQ80331_INTD
+
+//#define INTE	IRQ_IQ80331_I82544
+
+static inline int __init
+iq80331_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
+{
+	static int pci_irq_table[][4] = {
+		/*
+		 * PCI IDSEL/INTPIN->INTLINE
+		 * A       B       C       D
+		 */
+		{INTB, INTC, INTD, INTA}, /* PCI-X Slot */
+		{INTC, INTC, INTC, INTC}, /* GigE  */
+	};
+
+	BUG_ON(pin < 1 || pin > 4);
+
+	return PCI_IRQ_TABLE_LOOKUP(1, 7);
+}
+
+static int iq80331_setup(int nr, struct pci_sys_data *sys)
+{
+	struct resource *res;
+
+	if(nr != 0)
+		return 0;
+
+	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
+	if (!res)
+		panic("PCI: unable to alloc resources");
+
+	res[0].start = IOP331_PCI_LOWER_IO_VA;
+	res[0].end   = IOP331_PCI_UPPER_IO_VA;
+	res[0].name  = "IQ80331 PCI I/O Space";
+	res[0].flags = IORESOURCE_IO;
+
+	res[1].start = IOP331_PCI_LOWER_MEM_PA;
+	res[1].end   = IOP331_PCI_UPPER_MEM_PA;
+	res[1].name  = "IQ80331 PCI Memory Space";
+	res[1].flags = IORESOURCE_MEM;
+
+	request_resource(&ioport_resource, &res[0]);
+	request_resource(&iomem_resource, &res[1]);
+
+	sys->mem_offset = IOP331_PCI_MEM_OFFSET;
+	sys->io_offset  = IOP331_PCI_IO_OFFSET;
+
+	sys->resource[0] = &res[0];
+	sys->resource[1] = &res[1];
+	sys->resource[2] = NULL;
+
+	return 1;
+}
+
+static void iq80331_preinit(void)
+{
+	iop331_init();
+}
+
+static struct hw_pci iq80331_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iq80331_setup,
+	.scan		= iop331_scan_bus,
+	.preinit	= iq80331_preinit,
+	.map_irq	= iq80331_map_irq
+};
+
+static int __init iq80331_pci_init(void)
+{
+	if (machine_is_iq80331())
+		pci_common_init(&iq80331_pci);
+	return 0;
+}
+
+subsys_initcall(iq80331_pci_init);
+
+
+
+
diff --git a/arch/arm/mach-iop33x/iq80332-mm.c b/arch/arm/mach-iop33x/iq80332-mm.c
new file mode 100644
index 000000000000..f7bc73113ea8
--- /dev/null
+++ b/arch/arm/mach-iop33x/iq80332-mm.c
@@ -0,0 +1,35 @@
+/*
+ * linux/arch/arm/mach-iop33x/iq80332-mm.c
+ *
+ * Low level memory initialization for iq80332 platform
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+
+#include <asm/mach/map.h>
+
+
+/*
+ * IQ80332 specific IO mappings
+ *
+ * We use RedBoot's setup for the onboard devices.
+ */
+
+void __init iq80332_map_io(void)
+{
+	iop331_map_io();
+}
diff --git a/arch/arm/mach-iop33x/iq80332-pci.c b/arch/arm/mach-iop33x/iq80332-pci.c
new file mode 100644
index 000000000000..f3010f85b1a2
--- /dev/null
+++ b/arch/arm/mach-iop33x/iq80332-pci.c
@@ -0,0 +1,125 @@
+/*
+ * arch/arm/mach-iop33x/iq80332-pci.c
+ *
+ * PCI support for the Intel IQ80332 reference board
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+
+#include <asm/hardware.h>
+#include <asm/irq.h>
+#include <asm/mach/pci.h>
+#include <asm/mach-types.h>
+
+/*
+ * The following macro is used to lookup irqs in a standard table
+ * format for those systems that do not already have PCI
+ * interrupts properly routed.  We assume 1 <= pin <= 4
+ */
+#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
+({ int _ctl_ = -1;				\
+   unsigned int _idsel = idsel - minid;		\
+   if (_idsel <= maxid)				\
+      _ctl_ = pci_irq_table[_idsel][pin-1];	\
+   _ctl_; })
+
+#define INTA	IRQ_IQ80332_INTA
+#define INTB	IRQ_IQ80332_INTB
+#define INTC	IRQ_IQ80332_INTC
+#define INTD	IRQ_IQ80332_INTD
+
+//#define INTE	IRQ_IQ80332_I82544
+
+static inline int __init
+iq80332_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
+{
+	static int pci_irq_table[][8] = {
+		/*
+		 * PCI IDSEL/INTPIN->INTLINE
+		 * A       B       C       D
+		 */
+		{-1,   -1,   -1,   -1},
+		{-1,   -1,   -1,   -1},
+		{-1,   -1,   -1,   -1},
+		{INTA, INTB, INTC, INTD}, /* PCI-X Slot */
+		{-1,   -1,   -1,   -1},
+		{INTC, INTC, INTC, INTC}, /* GigE  */
+		{-1,   -1,   -1,   -1},
+		{-1,   -1,   -1,   -1},
+	};
+
+	BUG_ON(pin < 1 || pin > 4);
+
+	return PCI_IRQ_TABLE_LOOKUP(1, 7);
+}
+
+static int iq80332_setup(int nr, struct pci_sys_data *sys)
+{
+	struct resource *res;
+
+	if(nr != 0)
+		return 0;
+
+	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
+	if (!res)
+		panic("PCI: unable to alloc resources");
+
+	res[0].start = IOP331_PCI_LOWER_IO_VA;
+	res[0].end   = IOP331_PCI_UPPER_IO_VA;
+	res[0].name  = "IQ80332 PCI I/O Space";
+	res[0].flags = IORESOURCE_IO;
+
+	res[1].start = IOP331_PCI_LOWER_MEM_PA;
+	res[1].end   = IOP331_PCI_UPPER_MEM_PA;
+	res[1].name  = "IQ80332 PCI Memory Space";
+	res[1].flags = IORESOURCE_MEM;
+
+	request_resource(&ioport_resource, &res[0]);
+	request_resource(&iomem_resource, &res[1]);
+
+	sys->mem_offset = IOP331_PCI_MEM_OFFSET;
+	sys->io_offset  = IOP331_PCI_IO_OFFSET;
+
+	sys->resource[0] = &res[0];
+	sys->resource[1] = &res[1];
+	sys->resource[2] = NULL;
+
+	return 1;
+}
+
+static void iq80332_preinit(void)
+{
+	iop331_init();
+}
+
+static struct hw_pci iq80332_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iq80332_setup,
+	.scan		= iop331_scan_bus,
+	.preinit	= iq80332_preinit,
+	.map_irq	= iq80332_map_irq
+};
+
+static int __init iq80332_pci_init(void)
+{
+	if (machine_is_iq80332())
+		pci_common_init(&iq80332_pci);
+	return 0;
+}
+
+subsys_initcall(iq80332_pci_init);
+
+
+
+
diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
new file mode 100644
index 000000000000..029fa78876a5
--- /dev/null
+++ b/arch/arm/mach-iop33x/irq.c
@@ -0,0 +1,129 @@
+/*
+ * linux/arch/arm/mach-iop33x/irq.c
+ *
+ * Generic IOP331 IRQ handling functionality
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2003 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *
+ */
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+
+#include <asm/mach/irq.h>
+#include <asm/irq.h>
+#include <asm/hardware.h>
+
+#include <asm/mach-types.h>
+
+static u32 iop331_mask0 = 0;
+static u32 iop331_mask1 = 0;
+
+static inline void intctl_write0(u32 val)
+{
+    // INTCTL0
+	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
+}
+
+static inline void intctl_write1(u32 val)
+{
+    // INTCTL1
+    asm volatile("mcr p6,0,%0,c1,c0,0"::"r" (val));
+}
+
+static inline void intstr_write0(u32 val)
+{
+    // INTSTR0
+	asm volatile("mcr p6,0,%0,c2,c0,0"::"r" (val));
+}
+
+static inline void intstr_write1(u32 val)
+{
+    // INTSTR1
+	asm volatile("mcr p6,0,%0,c3,c0,0"::"r" (val));
+}
+
+static void
+iop331_irq_mask1 (unsigned int irq)
+{
+        iop331_mask0 &= ~(1 << (irq - IOP331_IRQ_OFS));
+        intctl_write0(iop331_mask0);
+}
+
+static void
+iop331_irq_mask2 (unsigned int irq)
+{
+        iop331_mask1 &= ~(1 << (irq - IOP331_IRQ_OFS - 32));
+        intctl_write1(iop331_mask1);
+}
+
+static void
+iop331_irq_unmask1(unsigned int irq)
+{
+        iop331_mask0 |= (1 << (irq - IOP331_IRQ_OFS));
+        intctl_write0(iop331_mask0);
+}
+
+static void
+iop331_irq_unmask2(unsigned int irq)
+{
+        iop331_mask1 |= (1 << (irq - IOP331_IRQ_OFS - 32));
+        intctl_write1(iop331_mask1);
+}
+
+struct irq_chip iop331_irqchip1 = {
+	.name	= "IOP-1",
+	.ack    = iop331_irq_mask1,
+	.mask   = iop331_irq_mask1,
+	.unmask = iop331_irq_unmask1,
+};
+
+struct irq_chip iop331_irqchip2 = {
+	.name	= "IOP-2",
+	.ack    = iop331_irq_mask2,
+	.mask   = iop331_irq_mask2,
+	.unmask = iop331_irq_unmask2,
+};
+
+void __init iop331_init_irq(void)
+{
+	unsigned int i, tmp;
+
+	/* Enable access to coprocessor 6 for dealing with IRQs.
+	 * From RMK:
+	 * Basically, the Intel documentation here is poor.  It appears that
+	 * you need to set the bit to be able to access the coprocessor from
+	 * SVC mode.  Whether that allows access from user space or not is
+	 * unclear.
+	 */
+	asm volatile (
+		"mrc p15, 0, %0, c15, c1, 0\n\t"
+		"orr %0, %0, %1\n\t"
+		"mcr p15, 0, %0, c15, c1, 0\n\t"
+		/* The action is delayed, so we have to do this: */
+		"mrc p15, 0, %0, c15, c1, 0\n\t"
+		"mov %0, %0\n\t"
+		"sub pc, pc, #4"
+		: "=r" (tmp) : "i" (1 << 6) );
+
+	intctl_write0(0);		// disable all interrupts
+    	intctl_write1(0);
+	intstr_write0(0);		// treat all as IRQ
+    	intstr_write1(0);
+	if(machine_is_iq80331()) 	// all interrupts are inputs to chip
+		*IOP331_PCIIRSR = 0x0f;
+
+	for(i = IOP331_IRQ_OFS; i < NR_IRQS; i++)
+	{
+		set_irq_chip(i, (i < 32) ? &iop331_irqchip1 : &iop331_irqchip2);
+		set_irq_handler(i, do_level_IRQ);
+		set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
+	}
+}
+
diff --git a/arch/arm/mach-iop33x/pci.c b/arch/arm/mach-iop33x/pci.c
new file mode 100644
index 000000000000..bf6aba91b47d
--- /dev/null
+++ b/arch/arm/mach-iop33x/pci.c
@@ -0,0 +1,222 @@
+/*
+ * arch/arm/mach-iop33x/pci.c
+ *
+ * PCI support for the Intel IOP331 chipset
+ *
+ * Author: Dave Jiang (dave.jiang@intel.com)
+ * Copyright (C) 2003, 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/hardware.h>
+#include <asm/mach/pci.h>
+
+#include <asm/arch/iop331.h>
+
+#undef DEBUG
+#undef DEBUG1
+
+#ifdef DEBUG
+#define  DBG(x...) printk(x)
+#else
+#define  DBG(x...) do { } while (0)
+#endif
+
+#ifdef DEBUG1
+#define  DBG1(x...) printk(x)
+#else
+#define  DBG1(x...) do { } while (0)
+#endif
+
+/*
+ * This routine builds either a type0 or type1 configuration command.  If the
+ * bus is on the 80331 then a type0 made, else a type1 is created.
+ */
+static u32 iop331_cfg_address(struct pci_bus *bus, int devfn, int where)
+{
+	struct pci_sys_data *sys = bus->sysdata;
+	u32 addr;
+
+	if (sys->busnr == bus->number)
+		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
+	else
+		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
+
+	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
+
+	return addr;
+}
+
+/*
+ * This routine checks the status of the last configuration cycle.  If an error
+ * was detected it returns a 1, else it returns a 0.  The errors being checked
+ * are parity, master abort, target abort (master and target).  These types of
+ * errors occure during a config cycle where there is no device, like during
+ * the discovery stage.
+ */
+static int iop331_pci_status(void)
+{
+	unsigned int status;
+	int ret = 0;
+
+	/*
+	 * Check the status registers.
+	 */
+	status = *IOP331_ATUSR;
+	if (status & 0xf900)
+	{
+		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
+		*IOP331_ATUSR = status & 0xf900;
+		ret = 1;
+	}
+	status = *IOP331_ATUISR;
+	if (status & 0x679f)
+	{
+		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
+		*IOP331_ATUISR = status & 0x679f;
+		ret = 1;
+	}
+	return ret;
+}
+
+/*
+ * Simply write the address register and read the configuration
+ * data.  Note that the 4 nop's ensure that we are able to handle
+ * a delayed abort (in theory.)
+ */
+static inline u32 iop331_read(unsigned long addr)
+{
+	u32 val;
+
+	__asm__ __volatile__(
+		"str	%1, [%2]\n\t"
+		"ldr	%0, [%3]\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		: "=r" (val)
+		: "r" (addr), "r" (IOP331_OCCAR), "r" (IOP331_OCCDR));
+
+	return val;
+}
+
+/*
+ * The read routines must check the error status of the last configuration
+ * cycle.  If there was an error, the routine returns all hex f's.
+ */
+static int
+iop331_read_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 *value)
+{
+	unsigned long addr = iop331_cfg_address(bus, devfn, where);
+	u32 val = iop331_read(addr) >> ((where & 3) * 8);
+
+	if( iop331_pci_status() )
+		val = 0xffffffff;
+
+	*value = val;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+iop331_write_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 value)
+{
+	unsigned long addr = iop331_cfg_address(bus, devfn, where);
+	u32 val;
+
+	if (size != 4) {
+		val = iop331_read(addr);
+		if (!iop331_pci_status() == 0)
+			return PCIBIOS_SUCCESSFUL;
+
+		where = (where & 3) * 8;
+
+		if (size == 1)
+			val &= ~(0xff << where);
+		else
+			val &= ~(0xffff << where);
+
+		*IOP331_OCCDR = val | value << where;
+	} else {
+		asm volatile(
+			"str	%1, [%2]\n\t"
+			"str	%0, [%3]\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			:
+			: "r" (value), "r" (addr),
+			  "r" (IOP331_OCCAR), "r" (IOP331_OCCDR));
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops iop331_ops = {
+	.read	= iop331_read_config,
+	.write	= iop331_write_config,
+};
+
+/*
+ * When a PCI device does not exist during config cycles, the XScale gets a
+ * bus error instead of returning 0xffffffff. This handler simply returns.
+ */
+int
+iop331_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
+{
+	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
+		addr, fsr, regs->ARM_pc, regs->ARM_lr);
+
+	/*
+	 * If it was an imprecise abort, then we need to correct the
+	 * return address to be _after_ the instruction.
+	 */
+	if (fsr & (1 << 10))
+		regs->ARM_pc += 4;
+
+	return 0;
+}
+
+/*
+ * Scan an IOP331 PCI bus.  sys->bus defines which bus we scan.
+ */
+struct pci_bus *iop331_scan_bus(int nr, struct pci_sys_data *sys)
+{
+	return pci_scan_bus(sys->busnr, &iop331_ops, sys);
+}
+
+void iop331_init(void)
+{
+	DBG1("PCI:  Intel 80331 PCI init code.\n");
+	DBG1("\tATU: IOP331_ATUCMD=0x%04x\n", *IOP331_ATUCMD);
+	DBG1("\tATU: IOP331_OMWTVR0=0x%04x, IOP331_OIOWTVR=0x%04x\n",
+			*IOP331_OMWTVR0,
+			*IOP331_OIOWTVR);
+	DBG1("\tATU: IOP331_OMWTVR1=0x%04x\n", *IOP331_OMWTVR1);
+	DBG1("\tATU: IOP331_ATUCR=0x%08x\n", *IOP331_ATUCR);
+	DBG1("\tATU: IOP331_IABAR0=0x%08x IOP331_IALR0=0x%08x IOP331_IATVR0=%08x\n", *IOP331_IABAR0, *IOP331_IALR0, *IOP331_IATVR0);
+	DBG1("\tATU: IOP31_IABAR1=0x%08x IOP331_IALR1=0x%08x\n", *IOP331_IABAR1, *IOP331_IALR1);
+	DBG1("\tATU: IOP331_ERBAR=0x%08x IOP331_ERLR=0x%08x IOP331_ERTVR=%08x\n", *IOP331_ERBAR, *IOP331_ERLR, *IOP331_ERTVR);
+	DBG1("\tATU: IOP331_IABAR2=0x%08x IOP331_IALR2=0x%08x IOP331_IATVR2=%08x\n", *IOP331_IABAR2, *IOP331_IALR2, *IOP331_IATVR2);
+	DBG1("\tATU: IOP331_IABAR3=0x%08x IOP331_IALR3=0x%08x IOP331_IATVR3=%08x\n", *IOP331_IABAR3, *IOP331_IALR3, *IOP331_IATVR3);
+
+	hook_fault_code(16+6, iop331_pci_abort, SIGBUS, "imprecise external abort");
+}
+
diff --git a/arch/arm/mach-iop33x/setup.c b/arch/arm/mach-iop33x/setup.c
new file mode 100644
index 000000000000..c14c7da5ec4c
--- /dev/null
+++ b/arch/arm/mach-iop33x/setup.c
@@ -0,0 +1,221 @@
+/*
+ * linux/arch/arm/mach-iop33x/setup.c
+ *
+ * Author: Dave Jiang (dave.jiang@intel.com)
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+#include <linux/serial.h>
+#include <linux/tty.h>
+#include <linux/serial_8250.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/mach/map.h>
+#include <asm/setup.h>
+#include <asm/system.h>
+#include <asm/memory.h>
+#include <asm/hardware.h>
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+
+#define IOP331_UART_XTAL 33334000
+
+/*
+ * Standard IO mapping for all IOP331 based systems
+ */
+static struct map_desc iop331_std_desc[] __initdata = {
+	{	/* mem mapped registers */
+		.virtual	= IOP331_VIRT_MEM_BASE,
+		.pfn		= __phys_to_pfn(IOP331_PHYS_MEM_BASE),
+		.length		= 0x00002000,
+		.type		= MT_DEVICE
+	}, {	/* PCI IO space */
+		.virtual	= IOP331_PCI_LOWER_IO_VA,
+		.pfn		= __phys_to_pfn(IOP331_PCI_LOWER_IO_PA),
+		.length		= IOP331_PCI_IO_WINDOW_SIZE,
+		.type		= MT_DEVICE
+	}
+};
+
+static struct resource iop33x_uart0_resources[] = {
+	[0] = {
+		.start = IOP331_UART0_PHYS,
+		.end = IOP331_UART0_PHYS + 0x3f,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_IOP331_UART0,
+		.end = IRQ_IOP331_UART0,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource iop33x_uart1_resources[] = {
+	[0] = {
+		.start = IOP331_UART1_PHYS,
+		.end = IOP331_UART1_PHYS + 0x3f,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_IOP331_UART1,
+		.end = IRQ_IOP331_UART1,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct plat_serial8250_port iop33x_uart0_data[] = {
+	{
+       .membase     = (char*)(IOP331_UART0_VIRT),
+       .mapbase     = (IOP331_UART0_PHYS),
+       .irq         = IRQ_IOP331_UART0,
+       .uartclk     = IOP331_UART_XTAL,
+       .regshift    = 2,
+       .iotype      = UPIO_MEM,
+       .flags       = UPF_SKIP_TEST,
+	},
+	{  },
+};
+
+static struct plat_serial8250_port iop33x_uart1_data[] = {
+	{
+       .membase     = (char*)(IOP331_UART1_VIRT),
+       .mapbase     = (IOP331_UART1_PHYS),
+       .irq         = IRQ_IOP331_UART1,
+       .uartclk     = IOP331_UART_XTAL,
+       .regshift    = 2,
+       .iotype      = UPIO_MEM,
+       .flags       = UPF_SKIP_TEST,
+	},
+	{  },
+};
+
+static struct platform_device iop33x_uart0 = {
+       .name = "serial8250",
+       .id = PLAT8250_DEV_PLATFORM,
+       .dev.platform_data = iop33x_uart0_data,
+       .num_resources = 2,
+       .resource = iop33x_uart0_resources,
+};
+
+static struct platform_device iop33x_uart1 = {
+       .name = "serial8250",
+       .id = PLAT8250_DEV_PLATFORM1,
+       .dev.platform_data = iop33x_uart1_data,
+       .num_resources = 2,
+       .resource = iop33x_uart1_resources,
+};
+
+static struct resource iop33x_i2c_0_resources[] = {
+	[0] = {
+		.start = 0xfffff680,
+		.end = 0xfffff698,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_IOP331_I2C_0,
+		.end = IRQ_IOP331_I2C_0,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource iop33x_i2c_1_resources[] = {
+	[0] = {
+		.start = 0xfffff6a0,
+		.end = 0xfffff6b8,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_IOP331_I2C_1,
+		.end = IRQ_IOP331_I2C_1,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct platform_device iop33x_i2c_0_controller = {
+	.name = "IOP3xx-I2C",
+	.id = 0,
+	.num_resources = 2,
+	.resource = iop33x_i2c_0_resources
+};
+
+static struct platform_device iop33x_i2c_1_controller = {
+	.name = "IOP3xx-I2C",
+	.id = 1,
+	.num_resources = 2,
+	.resource = iop33x_i2c_1_resources
+};
+
+static struct platform_device *iop33x_devices[] __initdata = {
+	&iop33x_uart0,
+	&iop33x_uart1,
+	&iop33x_i2c_0_controller,
+	&iop33x_i2c_1_controller
+};
+
+void __init iop33x_init(void)
+{
+	if(iop_is_331())
+	{
+		platform_add_devices(iop33x_devices,
+				ARRAY_SIZE(iop33x_devices));
+	}
+}
+
+void __init iop331_map_io(void)
+{
+	iotable_init(iop331_std_desc, ARRAY_SIZE(iop331_std_desc));
+}
+
+#ifdef CONFIG_ARCH_IOP33X
+extern void iop331_init_irq(void);
+extern struct sys_timer iop331_timer;
+#endif
+
+#ifdef CONFIG_ARCH_IQ80331
+extern void iq80331_map_io(void);
+#endif
+
+#ifdef CONFIG_MACH_IQ80332
+extern void iq80332_map_io(void);
+#endif
+
+#if defined(CONFIG_ARCH_IQ80331)
+MACHINE_START(IQ80331, "Intel IQ80331")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= 0xfefff000,
+	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
+	.map_io		= iq80331_map_io,
+	.init_irq	= iop331_init_irq,
+	.timer		= &iop331_timer,
+	.boot_params	= 0x0100,
+	.init_machine	= iop33x_init,
+MACHINE_END
+
+#elif defined(CONFIG_MACH_IQ80332)
+MACHINE_START(IQ80332, "Intel IQ80332")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= 0xfefff000,
+	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
+	.map_io		= iq80332_map_io,
+	.init_irq	= iop331_init_irq,
+	.timer		= &iop331_timer,
+	.boot_params	= 0x0100,
+	.init_machine	= iop33x_init,
+MACHINE_END
+
+#else
+#error No machine descriptor defined for this IOP3XX implementation
+#endif
+
+
diff --git a/arch/arm/mach-iop33x/time.c b/arch/arm/mach-iop33x/time.c
new file mode 100644
index 000000000000..d839cd0d926e
--- /dev/null
+++ b/arch/arm/mach-iop33x/time.c
@@ -0,0 +1,106 @@
+/*
+ * arch/arm/mach-iop33x/time.c
+ *
+ * Timer code for IOP331 based systems
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ *
+ * Copyright 2003 Intel Corp.
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/time.h>
+#include <linux/init.h>
+#include <linux/timex.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+#include <asm/mach/irq.h>
+#include <asm/mach/time.h>
+
+static inline unsigned long get_elapsed(void)
+{
+	return LATCH - *IOP331_TU_TCR0;
+}
+
+static unsigned long iop331_gettimeoffset(void)
+{
+	unsigned long elapsed, usec;
+	u32 tisr1, tisr2;
+
+	/*
+	 * If an interrupt was pending before we read the timer,
+	 * we've already wrapped.  Factor this into the time.
+	 * If an interrupt was pending after we read the timer,
+	 * it may have wrapped between checking the interrupt
+	 * status and reading the timer.  Re-read the timer to
+	 * be sure its value is after the wrap.
+	 */
+
+	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr1));
+	elapsed = get_elapsed();
+	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr2));
+
+	if(tisr1 & 1)
+		elapsed += LATCH;
+	else if (tisr2 & 1)
+		elapsed = LATCH + get_elapsed();
+
+	/*
+	 * Now convert them to usec.
+	 */
+	usec = (unsigned long)(elapsed / (CLOCK_TICK_RATE/1000000));
+
+	return usec;
+}
+
+static irqreturn_t
+iop331_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	u32 tisr;
+
+	write_seqlock(&xtime_lock);
+
+	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr));
+	tisr |= 1;
+	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (tisr));
+
+	timer_tick(regs);
+
+	write_sequnlock(&xtime_lock);
+	return IRQ_HANDLED;
+}
+
+static struct irqaction iop331_timer_irq = {
+	.name		= "IOP331 Timer Tick",
+	.handler	= iop331_timer_interrupt,
+	.flags		= IRQF_DISABLED | IRQF_TIMER,
+};
+
+static void __init iop331_timer_init(void)
+{
+	u32 timer_ctl;
+
+	setup_irq(IRQ_IOP331_TIMER0, &iop331_timer_irq);
+
+	timer_ctl = IOP331_TMR_EN | IOP331_TMR_PRIVILEGED | IOP331_TMR_RELOAD |
+			IOP331_TMR_RATIO_1_1;
+
+	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (LATCH));
+
+	asm volatile("mcr p6, 0, %0, c0, c1, 0"	: : "r" (timer_ctl));
+
+}
+
+struct sys_timer iop331_timer = {
+	.init		= iop331_timer_init,
+	.offset		= iop331_gettimeoffset,
+};
diff --git a/arch/arm/mach-iop3xx/Kconfig b/arch/arm/mach-iop3xx/Kconfig
deleted file mode 100644
index 887c3753727f..000000000000
--- a/arch/arm/mach-iop3xx/Kconfig
+++ /dev/null
@@ -1,66 +0,0 @@
-if ARCH_IOP3XX
-
-menu "IOP3xx Implementation Options"
-
-comment "IOP3xx Platform Types"
-
-config ARCH_IQ80321
-	bool "Enable support for IQ80321"
-	select ARCH_IOP32X
-	help
-	  Say Y here if you want to run your kernel on the Intel IQ80321
-	  evaluation kit for the IOP321 chipset.
-
-config ARCH_IQ31244
-	bool "Enable support for IQ31244"
-	select ARCH_IOP32X
-	help
-	  Say Y here if you want to run your kernel on the Intel IQ31244
-	  evaluation kit for the IOP321 chipset.
-
-config ARCH_IQ80331
-	bool "Enable support for IQ80331"
-	select ARCH_IOP33X
-	help
-	  Say Y here if you want to run your kernel on the Intel IQ80331
-	  evaluation kit for the IOP331 chipset.
-
-config MACH_IQ80332
-	bool "Enable support for IQ80332"
-	select ARCH_IOP33X
-	help
-	  Say Y here if you want to run your kernel on the Intel IQ80332
-	  evaluation kit for the IOP332 chipset.
-
-config ARCH_EP80219
-	bool "Enable support for EP80219"
-	select ARCH_IOP32X
-	select ARCH_IQ31244
-	help
-	  Say Y here if you want to run your kernel on the Intel EP80219
-	  evaluation kit for the Intel 80219 chipset (a IOP321 variant).
-
-# Which IOP variant are we running?
-config ARCH_IOP32X
-	bool
-	help
-	  The IQ80321 uses the IOP321 variant.
-	  The IQ31244 and EP80219 uses the IOP321 variant.
-
-config ARCH_IOP33X
-	bool
-	default ARCH_IQ80331
-	help
-	  The IQ80331, IQ80332, and IQ80333 uses the IOP331 variant.
-
-comment "IOP3xx Chipset Features"
-
-config IOP331_STEPD
-	bool "Chip stepping D of the IOP80331 processor or IOP80333"
-	depends on ARCH_IOP33X
-	help
-	  Say Y here if you have StepD of the IOP80331 or IOP8033
-	  based platforms.
-
-endmenu
-endif
diff --git a/arch/arm/mach-iop3xx/Makefile b/arch/arm/mach-iop3xx/Makefile
deleted file mode 100644
index ad3051524480..000000000000
--- a/arch/arm/mach-iop3xx/Makefile
+++ /dev/null
@@ -1,23 +0,0 @@
-#
-# Makefile for the linux kernel.
-#
-
-# Object file lists.
-
-obj-y			:= common.o
-
-obj-m			:=
-obj-n			:=
-obj-			:=
-
-obj-$(CONFIG_ARCH_IOP32X)  += iop321-setup.o iop321-irq.o iop321-pci.o iop321-time.o
-
-obj-$(CONFIG_ARCH_IOP33X)  += iop331-setup.o iop331-irq.o iop331-pci.o iop331-time.o
-
-obj-$(CONFIG_ARCH_IQ80321) += iq80321-mm.o iq80321-pci.o
-
-obj-$(CONFIG_ARCH_IQ31244) += iq31244-mm.o iq31244-pci.o
-
-obj-$(CONFIG_ARCH_IQ80331) += iq80331-mm.o iq80331-pci.o
-
-obj-$(CONFIG_MACH_IQ80332) += iq80332-mm.o iq80332-pci.o
diff --git a/arch/arm/mach-iop3xx/Makefile.boot b/arch/arm/mach-iop3xx/Makefile.boot
deleted file mode 100644
index e120c233b706..000000000000
--- a/arch/arm/mach-iop3xx/Makefile.boot
+++ /dev/null
@@ -1,9 +0,0 @@
-   zreladdr-y	:= 0xa0008000
-params_phys-y	:= 0xa0000100
-initrd_phys-y	:= 0xa0800000
-ifeq ($(CONFIG_ARCH_IOP33X),y)
-   zreladdr-y	:= 0x00008000
-params_phys-y	:= 0x00000100
-initrd_phys-y	:= 0x00800000
-endif
-
diff --git a/arch/arm/mach-iop3xx/common.c b/arch/arm/mach-iop3xx/common.c
deleted file mode 100644
index d7f50e57e753..000000000000
--- a/arch/arm/mach-iop3xx/common.c
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/common.c
- *
- * Common routines shared across all IOP3xx implementations
- *
- * Author: Deepak Saxena <dsaxena@mvista.com>
- *
- * Copyright 2003 (c) MontaVista, Software, Inc.
- *
- * This file is licensed under  the terms of the GNU General Public
- * License version 2. This program is licensed "as is" without any
- * warranty of any kind, whether express or implied.
- */
-
-#include <linux/delay.h>
-#include <asm/hardware.h>
-
-/*
- * Shared variables
- */
-unsigned long iop3xx_pcibios_min_io = 0;
-unsigned long iop3xx_pcibios_min_mem = 0;
-
-#ifdef CONFIG_ARCH_EP80219
-#include <linux/kernel.h>
-/*
- * Default power-off for EP80219
- */
-
-static inline void ep80219_send_to_pic(__u8 c) {
-}
-
-void ep80219_power_off(void)
-{
-	/*
-     * This function will send a SHUTDOWN_COMPLETE message to the PIC controller
-     * over I2C.  We are not using the i2c subsystem since we are going to power
-     * off and it may be removed
-     */
-
-	/* Send the Address byte w/ the start condition */
-	*IOP321_IDBR1 = 0x60;
-	*IOP321_ICR1 = 0xE9;
-    mdelay(1);
-
-	/* Send the START_MSG byte w/ no start or stop condition */
-	*IOP321_IDBR1 = 0x0F;
-	*IOP321_ICR1 = 0xE8;
-    mdelay(1);
-
-	/* Send the SHUTDOWN_COMPLETE Message ID byte w/ no start or stop condition */
-	*IOP321_IDBR1 = 0x03;
-	*IOP321_ICR1 = 0xE8;
-    mdelay(1);
-
-	/* Send an ignored byte w/ stop condition */
-	*IOP321_IDBR1 = 0x00;
-	*IOP321_ICR1 = 0xEA;
-
-	while (1) ;
-}
-
-#include <linux/init.h>
-#include <linux/pm.h>
-
-static int __init ep80219_init(void)
-{
-	pm_power_off = ep80219_power_off;
-	return 0;
-}
-arch_initcall(ep80219_init);
-#endif
diff --git a/arch/arm/mach-iop3xx/iop321-irq.c b/arch/arm/mach-iop3xx/iop321-irq.c
deleted file mode 100644
index 88ac333472c8..000000000000
--- a/arch/arm/mach-iop3xx/iop321-irq.c
+++ /dev/null
@@ -1,97 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/iop321-irq.c
- *
- * Generic IOP321 IRQ handling functionality
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * Added IOP3XX chipset and IQ80321 board masking code.
- *
- */
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/list.h>
-
-#include <asm/mach/irq.h>
-#include <asm/irq.h>
-#include <asm/hardware.h>
-
-#include <asm/mach-types.h>
-
-static u32 iop321_mask /* = 0 */;
-
-static inline void intctl_write(u32 val)
-{
-	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
-}
-
-static inline void intstr_write(u32 val)
-{
-	asm volatile("mcr p6,0,%0,c4,c0,0"::"r" (val));
-}
-
-static void
-iop321_irq_mask (unsigned int irq)
-{
-
-	iop321_mask &= ~(1 << (irq - IOP321_IRQ_OFS));
-
-	intctl_write(iop321_mask);
-}
-
-static void
-iop321_irq_unmask (unsigned int irq)
-{
-	iop321_mask |= (1 << (irq - IOP321_IRQ_OFS));
-
-	intctl_write(iop321_mask);
-}
-
-struct irq_chip ext_chip = {
-	.name	= "IOP",
-	.ack    = iop321_irq_mask,
-	.mask   = iop321_irq_mask,
-	.unmask = iop321_irq_unmask,
-};
-
-void __init iop321_init_irq(void)
-{
-	unsigned int i, tmp;
-
-	/* Enable access to coprocessor 6 for dealing with IRQs.
-	 * From RMK:
-	 * Basically, the Intel documentation here is poor.  It appears that
-	 * you need to set the bit to be able to access the coprocessor from
-	 * SVC mode.  Whether that allows access from user space or not is
-	 * unclear.
-	 */
-	asm volatile (
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"orr %0, %0, %1\n\t"
-		"mcr p15, 0, %0, c15, c1, 0\n\t"
-		/* The action is delayed, so we have to do this: */
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"mov %0, %0\n\t"
-		"sub pc, pc, #4"
-		: "=r" (tmp) : "i" (1 << 6) );
-
-	intctl_write(0);		// disable all interrupts
-	intstr_write(0);		// treat all as IRQ
-	if(machine_is_iq80321() ||
-	   machine_is_iq31244()) 	// all interrupts are inputs to chip
-		*IOP321_PCIIRSR = 0x0f;
-
-	for(i = IOP321_IRQ_OFS; i < NR_IOP321_IRQS; i++)
-	{
-		set_irq_chip(i, &ext_chip);
-		set_irq_handler(i, do_level_IRQ);
-		set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
-
-	}
-}
-
diff --git a/arch/arm/mach-iop3xx/iop321-pci.c b/arch/arm/mach-iop3xx/iop321-pci.c
deleted file mode 100644
index 8ba6a0e23134..000000000000
--- a/arch/arm/mach-iop3xx/iop321-pci.c
+++ /dev/null
@@ -1,220 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iop321-pci.c
- *
- * PCI support for the Intel IOP321 chipset
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/ioport.h>
-
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/system.h>
-#include <asm/hardware.h>
-#include <asm/mach/pci.h>
-
-#include <asm/arch/iop321.h>
-
-// #define DEBUG
-
-#ifdef DEBUG
-#define  DBG(x...) printk(x)
-#else
-#define  DBG(x...) do { } while (0)
-#endif
-
-/*
- * This routine builds either a type0 or type1 configuration command.  If the
- * bus is on the 80321 then a type0 made, else a type1 is created.
- */
-static u32 iop321_cfg_address(struct pci_bus *bus, int devfn, int where)
-{
-	struct pci_sys_data *sys = bus->sysdata;
-	u32 addr;
-
-	if (sys->busnr == bus->number)
-		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
-	else
-		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
-
-	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
-
-	return addr;
-}
-
-/*
- * This routine checks the status of the last configuration cycle.  If an error
- * was detected it returns a 1, else it returns a 0.  The errors being checked
- * are parity, master abort, target abort (master and target).  These types of
- * errors occure during a config cycle where there is no device, like during
- * the discovery stage.
- */
-static int iop321_pci_status(void)
-{
-	unsigned int status;
-	int ret = 0;
-
-	/*
-	 * Check the status registers.
-	 */
-	status = *IOP321_ATUSR;
-	if (status & 0xf900)
-	{
-		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
-		*IOP321_ATUSR = status & 0xf900;
-		ret = 1;
-	}
-	status = *IOP321_ATUISR;
-	if (status & 0x679f)
-	{
-		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
-		*IOP321_ATUISR = status & 0x679f;
-		ret = 1;
-	}
-	return ret;
-}
-
-/*
- * Simply write the address register and read the configuration
- * data.  Note that the 4 nop's ensure that we are able to handle
- * a delayed abort (in theory.)
- */
-static inline u32 iop321_read(unsigned long addr)
-{
-	u32 val;
-
-	__asm__ __volatile__(
-		"str	%1, [%2]\n\t"
-		"ldr	%0, [%3]\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		: "=r" (val)
-		: "r" (addr), "r" (IOP321_OCCAR), "r" (IOP321_OCCDR));
-
-	return val;
-}
-
-/*
- * The read routines must check the error status of the last configuration
- * cycle.  If there was an error, the routine returns all hex f's.
- */
-static int
-iop321_read_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 *value)
-{
-	unsigned long addr = iop321_cfg_address(bus, devfn, where);
-	u32 val = iop321_read(addr) >> ((where & 3) * 8);
-
-	if( iop321_pci_status() )
-		val = 0xffffffff;
-
-	*value = val;
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int
-iop321_write_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 value)
-{
-	unsigned long addr = iop321_cfg_address(bus, devfn, where);
-	u32 val;
-
-	if (size != 4) {
-		val = iop321_read(addr);
-		if (!iop321_pci_status() == 0)
-			return PCIBIOS_SUCCESSFUL;
-
-		where = (where & 3) * 8;
-
-		if (size == 1)
-			val &= ~(0xff << where);
-		else
-			val &= ~(0xffff << where);
-
-		*IOP321_OCCDR = val | value << where;
-	} else {
-		asm volatile(
-			"str	%1, [%2]\n\t"
-			"str	%0, [%3]\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			:
-			: "r" (value), "r" (addr),
-			  "r" (IOP321_OCCAR), "r" (IOP321_OCCDR));
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops iop321_ops = {
-	.read	= iop321_read_config,
-	.write	= iop321_write_config,
-};
-
-/*
- * When a PCI device does not exist during config cycles, the 80200 gets a
- * bus error instead of returning 0xffffffff. This handler simply returns.
- */
-int
-iop321_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
-{
-	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
-		addr, fsr, regs->ARM_pc, regs->ARM_lr);
-
-	/*
-	 * If it was an imprecise abort, then we need to correct the
-	 * return address to be _after_ the instruction.
-	 */
-	if (fsr & (1 << 10))
-		regs->ARM_pc += 4;
-
-	return 0;
-}
-
-/*
- * Scan an IOP321 PCI bus.  sys->bus defines which bus we scan.
- */
-struct pci_bus *iop321_scan_bus(int nr, struct pci_sys_data *sys)
-{
-	return pci_scan_bus(sys->busnr, &iop321_ops, sys);
-}
-
-void iop321_init(void)
-{
-	DBG("PCI:  Intel 80321 PCI init code.\n");
-	DBG("ATU: IOP321_ATUCMD=0x%04x\n", *IOP321_ATUCMD);
-	DBG("ATU: IOP321_OMWTVR0=0x%04x, IOP321_OIOWTVR=0x%04x\n",
-			*IOP321_OMWTVR0,
-			*IOP321_OIOWTVR);
-	DBG("ATU: IOP321_ATUCR=0x%08x\n", *IOP321_ATUCR);
-	DBG("ATU: IOP321_IABAR0=0x%08x IOP321_IALR0=0x%08x IOP321_IATVR0=%08x\n",
-			*IOP321_IABAR0, *IOP321_IALR0, *IOP321_IATVR0);
-	DBG("ATU: IOP321_OMWTVR0=0x%08x\n", *IOP321_OMWTVR0);
-	DBG("ATU: IOP321_IABAR1=0x%08x IOP321_IALR1=0x%08x\n",
-			*IOP321_IABAR1, *IOP321_IALR1);
-	DBG("ATU: IOP321_ERBAR=0x%08x IOP321_ERLR=0x%08x IOP321_ERTVR=%08x\n",
-			*IOP321_ERBAR, *IOP321_ERLR, *IOP321_ERTVR);
-	DBG("ATU: IOP321_IABAR2=0x%08x IOP321_IALR2=0x%08x IOP321_IATVR2=%08x\n",
-			*IOP321_IABAR2, *IOP321_IALR2, *IOP321_IATVR2);
-	DBG("ATU: IOP321_IABAR3=0x%08x IOP321_IALR3=0x%08x IOP321_IATVR3=%08x\n",
-			*IOP321_IABAR3, *IOP321_IALR3, *IOP321_IATVR3);
-
-	hook_fault_code(16+6, iop321_pci_abort, SIGBUS, "imprecise external abort");
-}
-
diff --git a/arch/arm/mach-iop3xx/iop321-setup.c b/arch/arm/mach-iop3xx/iop321-setup.c
deleted file mode 100644
index b6d096903c4a..000000000000
--- a/arch/arm/mach-iop3xx/iop321-setup.c
+++ /dev/null
@@ -1,173 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/iop321-setup.c
- *
- * Author: Nicolas Pitre <nico@cam.org>
- * Copyright (C) 2001 MontaVista Software, Inc.
- * Copyright (C) 2004 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/major.h>
-#include <linux/fs.h>
-#include <linux/platform_device.h>
-#include <linux/serial.h>
-#include <linux/tty.h>
-#include <linux/serial_core.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/mach/map.h>
-#include <asm/setup.h>
-#include <asm/system.h>
-#include <asm/memory.h>
-#include <asm/hardware.h>
-#include <asm/mach-types.h>
-#include <asm/mach/arch.h>
-
-#define IOP321_UART_XTAL 1843200
-
-/*
- * Standard IO mapping for all IOP321 based systems
- */
-static struct map_desc iop321_std_desc[] __initdata = {
-	 {	/* mem mapped registers */
-		.virtual	= IOP321_VIRT_MEM_BASE,
-		.pfn		= __phys_to_pfn(IOP321_PHYS_MEM_BASE),
-		.length		= 0x00002000,
-		.type		= MT_DEVICE
-	 }, {	/* PCI IO space */
-		.virtual	= IOP321_PCI_LOWER_IO_VA,
-		.pfn		= __phys_to_pfn(IOP321_PCI_LOWER_IO_PA),
-		.length		= IOP321_PCI_IO_WINDOW_SIZE,
-		.type		= MT_DEVICE
-	 }
-};
-
-#ifdef CONFIG_ARCH_IQ80321
-#define UARTBASE IQ80321_UART
-#define IRQ_UART IRQ_IQ80321_UART
-#endif
-
-#ifdef CONFIG_ARCH_IQ31244
-#define UARTBASE IQ31244_UART
-#define IRQ_UART IRQ_IQ31244_UART
-#endif
-
-static struct uart_port iop321_serial_ports[] = {
-	{
-		.membase	= (char*)(UARTBASE),
-		.mapbase	= (UARTBASE),
-		.irq		= IRQ_UART,
-		.flags		= UPF_SKIP_TEST,
-		.iotype		= UPIO_MEM,
-		.regshift	= 0,
-		.uartclk	= IOP321_UART_XTAL,
-		.line		= 0,
-		.type		= PORT_16550A,
-		.fifosize	= 16
-	}
-};
-
-static struct resource iop32x_i2c_0_resources[] = {
-	[0] = {
-		.start = 0xfffff680,
-		.end = 0xfffff698,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP321_I2C_0,
-		.end = IRQ_IOP321_I2C_0,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct resource iop32x_i2c_1_resources[] = {
-	[0] = {
-		.start = 0xfffff6a0,
-		.end = 0xfffff6b8,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP321_I2C_1,
-		.end = IRQ_IOP321_I2C_1,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct platform_device iop32x_i2c_0_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 0,
-	.num_resources = 2,
-	.resource = iop32x_i2c_0_resources
-};
-
-static struct platform_device iop32x_i2c_1_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 1,
-	.num_resources = 2,
-	.resource = iop32x_i2c_1_resources
-};
-
-static struct platform_device *iop32x_devices[] __initdata = {
-	&iop32x_i2c_0_controller,
-	&iop32x_i2c_1_controller
-};
-
-void __init iop32x_init(void)
-{
-	if(iop_is_321())
-	{
-		platform_add_devices(iop32x_devices,
-				ARRAY_SIZE(iop32x_devices));
-	}
-}
-
-void __init iop321_map_io(void)
-{
-	iotable_init(iop321_std_desc, ARRAY_SIZE(iop321_std_desc));
-	early_serial_setup(&iop321_serial_ports[0]);
-}
-
-#ifdef CONFIG_ARCH_IQ80321
-extern void iq80321_map_io(void);
-extern struct sys_timer iop321_timer;
-extern void iop321_init_time(void);
-#endif
-
-#ifdef CONFIG_ARCH_IQ31244
-extern void iq31244_map_io(void);
-extern struct sys_timer iop321_timer;
-extern void iop321_init_time(void);
-#endif
-
-#if defined(CONFIG_ARCH_IQ80321)
-MACHINE_START(IQ80321, "Intel IQ80321")
-	/* Maintainer: Intel Corporation */
-	.phys_io	= IQ80321_UART,
-	.io_pg_offst	= ((IQ80321_UART) >> 18) & 0xfffc,
-	.map_io		= iq80321_map_io,
-	.init_irq	= iop321_init_irq,
-	.timer		= &iop321_timer,
-	.boot_params	= 0xa0000100,
-	.init_machine	= iop32x_init,
-MACHINE_END
-#elif defined(CONFIG_ARCH_IQ31244)
-MACHINE_START(IQ31244, "Intel IQ31244")
-	/* Maintainer: Intel Corp. */
-	.phys_io	= IQ31244_UART,
-	.io_pg_offst	= ((IQ31244_UART) >> 18) & 0xfffc,
-	.map_io		= iq31244_map_io,
-	.init_irq	= iop321_init_irq,
-	.timer		= &iop321_timer,
-	.boot_params	= 0xa0000100,
-	.init_machine	= iop32x_init,
-MACHINE_END
-#else
-#error No machine descriptor defined for this IOP3XX implementation
-#endif
diff --git a/arch/arm/mach-iop3xx/iop321-time.c b/arch/arm/mach-iop3xx/iop321-time.c
deleted file mode 100644
index 04b1a6f7ebae..000000000000
--- a/arch/arm/mach-iop3xx/iop321-time.c
+++ /dev/null
@@ -1,108 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iop321-time.c
- *
- * Timer code for IOP321 based systems
- *
- * Author: Deepak Saxena <dsaxena@mvista.com>
- *
- * Copyright 2002-2003 MontaVista Software Inc.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/interrupt.h>
-#include <linux/time.h>
-#include <linux/init.h>
-#include <linux/timex.h>
-
-#include <asm/hardware.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
-#include <asm/mach/irq.h>
-#include <asm/mach/time.h>
-
-#define IOP321_TIME_SYNC 0
-
-static inline unsigned long get_elapsed(void)
-{
-	return LATCH - *IOP321_TU_TCR0;
-}
-
-static unsigned long iop321_gettimeoffset(void)
-{
-	unsigned long elapsed, usec;
-	u32 tisr1, tisr2;
-
-	/*
-	 * If an interrupt was pending before we read the timer,
-	 * we've already wrapped.  Factor this into the time.
-	 * If an interrupt was pending after we read the timer,
-	 * it may have wrapped between checking the interrupt
-	 * status and reading the timer.  Re-read the timer to
-	 * be sure its value is after the wrap.
-	 */
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr1));
-	elapsed = get_elapsed();
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr2));
-
-	if(tisr1 & 1)
-		elapsed += LATCH;
-	else if (tisr2 & 1)
-		elapsed = LATCH + get_elapsed();
-
-	/*
-	 * Now convert them to usec.
-	 */
-	usec = (unsigned long)(elapsed / (CLOCK_TICK_RATE/1000000));
-
-	return usec;
-}
-
-static irqreturn_t
-iop321_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
-{
-	u32 tisr;
-
-	write_seqlock(&xtime_lock);
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr));
-	tisr |= 1;
-	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (tisr));
-
-	timer_tick(regs);
-
-	write_sequnlock(&xtime_lock);
-
-	return IRQ_HANDLED;
-}
-
-static struct irqaction iop321_timer_irq = {
-	.name		= "IOP321 Timer Tick",
-	.handler	= iop321_timer_interrupt,
-	.flags		= IRQF_DISABLED | IRQF_TIMER,
-};
-
-static void __init iop321_timer_init(void)
-{
-	u32 timer_ctl;
-
-	setup_irq(IRQ_IOP321_TIMER0, &iop321_timer_irq);
-
-	timer_ctl = IOP321_TMR_EN | IOP321_TMR_PRIVILEGED | IOP321_TMR_RELOAD |
-			IOP321_TMR_RATIO_1_1;
-
-	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (LATCH));
-
-	asm volatile("mcr p6, 0, %0, c0, c1, 0"	: : "r" (timer_ctl));
-}
-
-struct sys_timer iop321_timer = {
-	.init		= &iop321_timer_init,
-	.offset		= iop321_gettimeoffset,
-};
diff --git a/arch/arm/mach-iop3xx/iop331-irq.c b/arch/arm/mach-iop3xx/iop331-irq.c
deleted file mode 100644
index cab11722ced2..000000000000
--- a/arch/arm/mach-iop3xx/iop331-irq.c
+++ /dev/null
@@ -1,129 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/iop331-irq.c
- *
- * Generic IOP331 IRQ handling functionality
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2003 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- *
- */
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/list.h>
-
-#include <asm/mach/irq.h>
-#include <asm/irq.h>
-#include <asm/hardware.h>
-
-#include <asm/mach-types.h>
-
-static u32 iop331_mask0 = 0;
-static u32 iop331_mask1 = 0;
-
-static inline void intctl_write0(u32 val)
-{
-    // INTCTL0
-	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
-}
-
-static inline void intctl_write1(u32 val)
-{
-    // INTCTL1
-    asm volatile("mcr p6,0,%0,c1,c0,0"::"r" (val));
-}
-
-static inline void intstr_write0(u32 val)
-{
-    // INTSTR0
-	asm volatile("mcr p6,0,%0,c2,c0,0"::"r" (val));
-}
-
-static inline void intstr_write1(u32 val)
-{
-    // INTSTR1
-	asm volatile("mcr p6,0,%0,c3,c0,0"::"r" (val));
-}
-
-static void
-iop331_irq_mask1 (unsigned int irq)
-{
-        iop331_mask0 &= ~(1 << (irq - IOP331_IRQ_OFS));
-        intctl_write0(iop331_mask0);
-}
-
-static void
-iop331_irq_mask2 (unsigned int irq)
-{
-        iop331_mask1 &= ~(1 << (irq - IOP331_IRQ_OFS - 32));
-        intctl_write1(iop331_mask1);
-}
-
-static void
-iop331_irq_unmask1(unsigned int irq)
-{
-        iop331_mask0 |= (1 << (irq - IOP331_IRQ_OFS));
-        intctl_write0(iop331_mask0);
-}
-
-static void
-iop331_irq_unmask2(unsigned int irq)
-{
-        iop331_mask1 |= (1 << (irq - IOP331_IRQ_OFS - 32));
-        intctl_write1(iop331_mask1);
-}
-
-struct irq_chip iop331_irqchip1 = {
-	.name	= "IOP-1",
-	.ack    = iop331_irq_mask1,
-	.mask   = iop331_irq_mask1,
-	.unmask = iop331_irq_unmask1,
-};
-
-struct irq_chip iop331_irqchip2 = {
-	.name	= "IOP-2",
-	.ack    = iop331_irq_mask2,
-	.mask   = iop331_irq_mask2,
-	.unmask = iop331_irq_unmask2,
-};
-
-void __init iop331_init_irq(void)
-{
-	unsigned int i, tmp;
-
-	/* Enable access to coprocessor 6 for dealing with IRQs.
-	 * From RMK:
-	 * Basically, the Intel documentation here is poor.  It appears that
-	 * you need to set the bit to be able to access the coprocessor from
-	 * SVC mode.  Whether that allows access from user space or not is
-	 * unclear.
-	 */
-	asm volatile (
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"orr %0, %0, %1\n\t"
-		"mcr p15, 0, %0, c15, c1, 0\n\t"
-		/* The action is delayed, so we have to do this: */
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"mov %0, %0\n\t"
-		"sub pc, pc, #4"
-		: "=r" (tmp) : "i" (1 << 6) );
-
-	intctl_write0(0);		// disable all interrupts
-    	intctl_write1(0);
-	intstr_write0(0);		// treat all as IRQ
-    	intstr_write1(0);
-	if(machine_is_iq80331()) 	// all interrupts are inputs to chip
-		*IOP331_PCIIRSR = 0x0f;
-
-	for(i = IOP331_IRQ_OFS; i < NR_IOP331_IRQS; i++)
-	{
-		set_irq_chip(i, (i < 32) ? &iop331_irqchip1 : &iop331_irqchip2);
-		set_irq_handler(i, do_level_IRQ);
-		set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
-	}
-}
-
diff --git a/arch/arm/mach-iop3xx/iop331-pci.c b/arch/arm/mach-iop3xx/iop331-pci.c
deleted file mode 100644
index 44dd213b48a3..000000000000
--- a/arch/arm/mach-iop3xx/iop331-pci.c
+++ /dev/null
@@ -1,222 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iop331-pci.c
- *
- * PCI support for the Intel IOP331 chipset
- *
- * Author: Dave Jiang (dave.jiang@intel.com)
- * Copyright (C) 2003, 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/ioport.h>
-
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/system.h>
-#include <asm/hardware.h>
-#include <asm/mach/pci.h>
-
-#include <asm/arch/iop331.h>
-
-#undef DEBUG
-#undef DEBUG1
-
-#ifdef DEBUG
-#define  DBG(x...) printk(x)
-#else
-#define  DBG(x...) do { } while (0)
-#endif
-
-#ifdef DEBUG1
-#define  DBG1(x...) printk(x)
-#else
-#define  DBG1(x...) do { } while (0)
-#endif
-
-/*
- * This routine builds either a type0 or type1 configuration command.  If the
- * bus is on the 80331 then a type0 made, else a type1 is created.
- */
-static u32 iop331_cfg_address(struct pci_bus *bus, int devfn, int where)
-{
-	struct pci_sys_data *sys = bus->sysdata;
-	u32 addr;
-
-	if (sys->busnr == bus->number)
-		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
-	else
-		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
-
-	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
-
-	return addr;
-}
-
-/*
- * This routine checks the status of the last configuration cycle.  If an error
- * was detected it returns a 1, else it returns a 0.  The errors being checked
- * are parity, master abort, target abort (master and target).  These types of
- * errors occure during a config cycle where there is no device, like during
- * the discovery stage.
- */
-static int iop331_pci_status(void)
-{
-	unsigned int status;
-	int ret = 0;
-
-	/*
-	 * Check the status registers.
-	 */
-	status = *IOP331_ATUSR;
-	if (status & 0xf900)
-	{
-		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
-		*IOP331_ATUSR = status & 0xf900;
-		ret = 1;
-	}
-	status = *IOP331_ATUISR;
-	if (status & 0x679f)
-	{
-		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
-		*IOP331_ATUISR = status & 0x679f;
-		ret = 1;
-	}
-	return ret;
-}
-
-/*
- * Simply write the address register and read the configuration
- * data.  Note that the 4 nop's ensure that we are able to handle
- * a delayed abort (in theory.)
- */
-static inline u32 iop331_read(unsigned long addr)
-{
-	u32 val;
-
-	__asm__ __volatile__(
-		"str	%1, [%2]\n\t"
-		"ldr	%0, [%3]\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		: "=r" (val)
-		: "r" (addr), "r" (IOP331_OCCAR), "r" (IOP331_OCCDR));
-
-	return val;
-}
-
-/*
- * The read routines must check the error status of the last configuration
- * cycle.  If there was an error, the routine returns all hex f's.
- */
-static int
-iop331_read_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 *value)
-{
-	unsigned long addr = iop331_cfg_address(bus, devfn, where);
-	u32 val = iop331_read(addr) >> ((where & 3) * 8);
-
-	if( iop331_pci_status() )
-		val = 0xffffffff;
-
-	*value = val;
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int
-iop331_write_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 value)
-{
-	unsigned long addr = iop331_cfg_address(bus, devfn, where);
-	u32 val;
-
-	if (size != 4) {
-		val = iop331_read(addr);
-		if (!iop331_pci_status() == 0)
-			return PCIBIOS_SUCCESSFUL;
-
-		where = (where & 3) * 8;
-
-		if (size == 1)
-			val &= ~(0xff << where);
-		else
-			val &= ~(0xffff << where);
-
-		*IOP331_OCCDR = val | value << where;
-	} else {
-		asm volatile(
-			"str	%1, [%2]\n\t"
-			"str	%0, [%3]\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			:
-			: "r" (value), "r" (addr),
-			  "r" (IOP331_OCCAR), "r" (IOP331_OCCDR));
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops iop331_ops = {
-	.read	= iop331_read_config,
-	.write	= iop331_write_config,
-};
-
-/*
- * When a PCI device does not exist during config cycles, the XScale gets a
- * bus error instead of returning 0xffffffff. This handler simply returns.
- */
-int
-iop331_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
-{
-	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
-		addr, fsr, regs->ARM_pc, regs->ARM_lr);
-
-	/*
-	 * If it was an imprecise abort, then we need to correct the
-	 * return address to be _after_ the instruction.
-	 */
-	if (fsr & (1 << 10))
-		regs->ARM_pc += 4;
-
-	return 0;
-}
-
-/*
- * Scan an IOP331 PCI bus.  sys->bus defines which bus we scan.
- */
-struct pci_bus *iop331_scan_bus(int nr, struct pci_sys_data *sys)
-{
-	return pci_scan_bus(sys->busnr, &iop331_ops, sys);
-}
-
-void iop331_init(void)
-{
-	DBG1("PCI:  Intel 80331 PCI init code.\n");
-	DBG1("\tATU: IOP331_ATUCMD=0x%04x\n", *IOP331_ATUCMD);
-	DBG1("\tATU: IOP331_OMWTVR0=0x%04x, IOP331_OIOWTVR=0x%04x\n",
-			*IOP331_OMWTVR0,
-			*IOP331_OIOWTVR);
-	DBG1("\tATU: IOP331_OMWTVR1=0x%04x\n", *IOP331_OMWTVR1);
-	DBG1("\tATU: IOP331_ATUCR=0x%08x\n", *IOP331_ATUCR);
-	DBG1("\tATU: IOP331_IABAR0=0x%08x IOP331_IALR0=0x%08x IOP331_IATVR0=%08x\n", *IOP331_IABAR0, *IOP331_IALR0, *IOP331_IATVR0);
-	DBG1("\tATU: IOP31_IABAR1=0x%08x IOP331_IALR1=0x%08x\n", *IOP331_IABAR1, *IOP331_IALR1);
-	DBG1("\tATU: IOP331_ERBAR=0x%08x IOP331_ERLR=0x%08x IOP331_ERTVR=%08x\n", *IOP331_ERBAR, *IOP331_ERLR, *IOP331_ERTVR);
-	DBG1("\tATU: IOP331_IABAR2=0x%08x IOP331_IALR2=0x%08x IOP331_IATVR2=%08x\n", *IOP331_IABAR2, *IOP331_IALR2, *IOP331_IATVR2);
-	DBG1("\tATU: IOP331_IABAR3=0x%08x IOP331_IALR3=0x%08x IOP331_IATVR3=%08x\n", *IOP331_IABAR3, *IOP331_IALR3, *IOP331_IATVR3);
-
-	hook_fault_code(16+6, iop331_pci_abort, SIGBUS, "imprecise external abort");
-}
-
diff --git a/arch/arm/mach-iop3xx/iop331-setup.c b/arch/arm/mach-iop3xx/iop331-setup.c
deleted file mode 100644
index b2f24931d143..000000000000
--- a/arch/arm/mach-iop3xx/iop331-setup.c
+++ /dev/null
@@ -1,221 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/iop331-setup.c
- *
- * Author: Dave Jiang (dave.jiang@intel.com)
- * Copyright (C) 2004 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/major.h>
-#include <linux/fs.h>
-#include <linux/platform_device.h>
-#include <linux/serial.h>
-#include <linux/tty.h>
-#include <linux/serial_8250.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/mach/map.h>
-#include <asm/setup.h>
-#include <asm/system.h>
-#include <asm/memory.h>
-#include <asm/hardware.h>
-#include <asm/mach-types.h>
-#include <asm/mach/arch.h>
-
-#define IOP331_UART_XTAL 33334000
-
-/*
- * Standard IO mapping for all IOP331 based systems
- */
-static struct map_desc iop331_std_desc[] __initdata = {
-	{	/* mem mapped registers */
-		.virtual	= IOP331_VIRT_MEM_BASE,
-		.pfn		= __phys_to_pfn(IOP331_PHYS_MEM_BASE),
-		.length		= 0x00002000,
-		.type		= MT_DEVICE
-	}, {	/* PCI IO space */
-		.virtual	= IOP331_PCI_LOWER_IO_VA,
-		.pfn		= __phys_to_pfn(IOP331_PCI_LOWER_IO_PA),
-		.length		= IOP331_PCI_IO_WINDOW_SIZE,
-		.type		= MT_DEVICE
-	}
-};
-
-static struct resource iop33x_uart0_resources[] = {
-	[0] = {
-		.start = IOP331_UART0_PHYS,
-		.end = IOP331_UART0_PHYS + 0x3f,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_UART0,
-		.end = IRQ_IOP331_UART0,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct resource iop33x_uart1_resources[] = {
-	[0] = {
-		.start = IOP331_UART1_PHYS,
-		.end = IOP331_UART1_PHYS + 0x3f,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_UART1,
-		.end = IRQ_IOP331_UART1,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct plat_serial8250_port iop33x_uart0_data[] = {
-	{
-       .membase     = (char*)(IOP331_UART0_VIRT),
-       .mapbase     = (IOP331_UART0_PHYS),
-       .irq         = IRQ_IOP331_UART0,
-       .uartclk     = IOP331_UART_XTAL,
-       .regshift    = 2,
-       .iotype      = UPIO_MEM,
-       .flags       = UPF_SKIP_TEST,
-	},
-	{  },
-};
-
-static struct plat_serial8250_port iop33x_uart1_data[] = {
-	{
-       .membase     = (char*)(IOP331_UART1_VIRT),
-       .mapbase     = (IOP331_UART1_PHYS),
-       .irq         = IRQ_IOP331_UART1,
-       .uartclk     = IOP331_UART_XTAL,
-       .regshift    = 2,
-       .iotype      = UPIO_MEM,
-       .flags       = UPF_SKIP_TEST,
-	},
-	{  },
-};
-
-static struct platform_device iop33x_uart0 = {
-       .name = "serial8250",
-       .id = PLAT8250_DEV_PLATFORM,
-       .dev.platform_data = iop33x_uart0_data,
-       .num_resources = 2,
-       .resource = iop33x_uart0_resources,
-};
-
-static struct platform_device iop33x_uart1 = {
-       .name = "serial8250",
-       .id = PLAT8250_DEV_PLATFORM1,
-       .dev.platform_data = iop33x_uart1_data,
-       .num_resources = 2,
-       .resource = iop33x_uart1_resources,
-};
-
-static struct resource iop33x_i2c_0_resources[] = {
-	[0] = {
-		.start = 0xfffff680,
-		.end = 0xfffff698,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_I2C_0,
-		.end = IRQ_IOP331_I2C_0,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct resource iop33x_i2c_1_resources[] = {
-	[0] = {
-		.start = 0xfffff6a0,
-		.end = 0xfffff6b8,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_I2C_1,
-		.end = IRQ_IOP331_I2C_1,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct platform_device iop33x_i2c_0_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 0,
-	.num_resources = 2,
-	.resource = iop33x_i2c_0_resources
-};
-
-static struct platform_device iop33x_i2c_1_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 1,
-	.num_resources = 2,
-	.resource = iop33x_i2c_1_resources
-};
-
-static struct platform_device *iop33x_devices[] __initdata = {
-	&iop33x_uart0,
-	&iop33x_uart1,
-	&iop33x_i2c_0_controller,
-	&iop33x_i2c_1_controller
-};
-
-void __init iop33x_init(void)
-{
-	if(iop_is_331())
-	{
-		platform_add_devices(iop33x_devices,
-				ARRAY_SIZE(iop33x_devices));
-	}
-}
-
-void __init iop331_map_io(void)
-{
-	iotable_init(iop331_std_desc, ARRAY_SIZE(iop331_std_desc));
-}
-
-#ifdef CONFIG_ARCH_IOP33X
-extern void iop331_init_irq(void);
-extern struct sys_timer iop331_timer;
-#endif
-
-#ifdef CONFIG_ARCH_IQ80331
-extern void iq80331_map_io(void);
-#endif
-
-#ifdef CONFIG_MACH_IQ80332
-extern void iq80332_map_io(void);
-#endif
-
-#if defined(CONFIG_ARCH_IQ80331)
-MACHINE_START(IQ80331, "Intel IQ80331")
-	/* Maintainer: Intel Corp. */
-	.phys_io	= 0xfefff000,
-	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
-	.map_io		= iq80331_map_io,
-	.init_irq	= iop331_init_irq,
-	.timer		= &iop331_timer,
-	.boot_params	= 0x0100,
-	.init_machine	= iop33x_init,
-MACHINE_END
-
-#elif defined(CONFIG_MACH_IQ80332)
-MACHINE_START(IQ80332, "Intel IQ80332")
-	/* Maintainer: Intel Corp. */
-	.phys_io	= 0xfefff000,
-	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
-	.map_io		= iq80332_map_io,
-	.init_irq	= iop331_init_irq,
-	.timer		= &iop331_timer,
-	.boot_params	= 0x0100,
-	.init_machine	= iop33x_init,
-MACHINE_END
-
-#else
-#error No machine descriptor defined for this IOP3XX implementation
-#endif
-
-
diff --git a/arch/arm/mach-iop3xx/iop331-time.c b/arch/arm/mach-iop3xx/iop331-time.c
deleted file mode 100644
index 0c09e74c5740..000000000000
--- a/arch/arm/mach-iop3xx/iop331-time.c
+++ /dev/null
@@ -1,106 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iop331-time.c
- *
- * Timer code for IOP331 based systems
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- *
- * Copyright 2003 Intel Corp.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/interrupt.h>
-#include <linux/time.h>
-#include <linux/init.h>
-#include <linux/timex.h>
-
-#include <asm/hardware.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
-#include <asm/mach/irq.h>
-#include <asm/mach/time.h>
-
-static inline unsigned long get_elapsed(void)
-{
-	return LATCH - *IOP331_TU_TCR0;
-}
-
-static unsigned long iop331_gettimeoffset(void)
-{
-	unsigned long elapsed, usec;
-	u32 tisr1, tisr2;
-
-	/*
-	 * If an interrupt was pending before we read the timer,
-	 * we've already wrapped.  Factor this into the time.
-	 * If an interrupt was pending after we read the timer,
-	 * it may have wrapped between checking the interrupt
-	 * status and reading the timer.  Re-read the timer to
-	 * be sure its value is after the wrap.
-	 */
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr1));
-	elapsed = get_elapsed();
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr2));
-
-	if(tisr1 & 1)
-		elapsed += LATCH;
-	else if (tisr2 & 1)
-		elapsed = LATCH + get_elapsed();
-
-	/*
-	 * Now convert them to usec.
-	 */
-	usec = (unsigned long)(elapsed / (CLOCK_TICK_RATE/1000000));
-
-	return usec;
-}
-
-static irqreturn_t
-iop331_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
-{
-	u32 tisr;
-
-	write_seqlock(&xtime_lock);
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr));
-	tisr |= 1;
-	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (tisr));
-
-	timer_tick(regs);
-
-	write_sequnlock(&xtime_lock);
-	return IRQ_HANDLED;
-}
-
-static struct irqaction iop331_timer_irq = {
-	.name		= "IOP331 Timer Tick",
-	.handler	= iop331_timer_interrupt,
-	.flags		= IRQF_DISABLED | IRQF_TIMER,
-};
-
-static void __init iop331_timer_init(void)
-{
-	u32 timer_ctl;
-
-	setup_irq(IRQ_IOP331_TIMER0, &iop331_timer_irq);
-
-	timer_ctl = IOP331_TMR_EN | IOP331_TMR_PRIVILEGED | IOP331_TMR_RELOAD |
-			IOP331_TMR_RATIO_1_1;
-
-	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (LATCH));
-
-	asm volatile("mcr p6, 0, %0, c0, c1, 0"	: : "r" (timer_ctl));
-
-}
-
-struct sys_timer iop331_timer = {
-	.init		= iop331_timer_init,
-	.offset		= iop331_gettimeoffset,
-};
diff --git a/arch/arm/mach-iop3xx/iq31244-mm.c b/arch/arm/mach-iop3xx/iq31244-mm.c
deleted file mode 100644
index e874b54eefe3..000000000000
--- a/arch/arm/mach-iop3xx/iq31244-mm.c
+++ /dev/null
@@ -1,45 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/mm.c
- *
- * Low level memory initialization for iq80321 platform
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-
-
-/*
- * IQ80321 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-static struct map_desc iq31244_io_desc[] __initdata = {
-	{	/* on-board devices */
-		.virtual	= IQ31244_UART,
-		.pfn		= __phys_to_pfn(IQ31244_UART),
-		.length		= 0x00100000,
-		.type		= MT_DEVICE
-	}
-};
-
-void __init iq31244_map_io(void)
-{
-	iop321_map_io();
-
-	iotable_init(iq31244_io_desc, ARRAY_SIZE(iq31244_io_desc));
-}
diff --git a/arch/arm/mach-iop3xx/iq31244-pci.c b/arch/arm/mach-iop3xx/iq31244-pci.c
deleted file mode 100644
index f3c6413fa5bd..000000000000
--- a/arch/arm/mach-iop3xx/iq31244-pci.c
+++ /dev/null
@@ -1,129 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iq80321-pci.c
- *
- * PCI support for the Intel IQ80321 reference board
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ31244_INTA
-#define INTB	IRQ_IQ31244_INTB
-#define INTC	IRQ_IQ31244_INTC
-#define INTD	IRQ_IQ31244_INTD
-
-#define INTE	IRQ_IQ31244_I82546
-
-static inline int __init
-iq31244_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][4] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-#ifdef CONFIG_ARCH_EP80219
-		{INTB, INTB, INTB, INTB}, /* CFlash */
-		{INTE, INTE, INTE, INTE}, /* 82551 Pro 100 */
-		{INTD, INTD, INTD, INTD}, /* PCI-X Slot */
-		{INTC, INTC, INTC, INTC}, /* SATA   */
-#else
-		{INTB, INTB, INTB, INTB}, /* CFlash */
-		{INTC, INTC, INTC, INTC}, /* SATA   */
-		{INTD, INTD, INTD, INTD}, /* PCI-X Slot */
-		{INTE, INTE, INTE, INTE}, /* 82546 GigE */
-#endif // CONFIG_ARCH_EP80219
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-	return PCI_IRQ_TABLE_LOOKUP(0, 7);
-}
-
-static int iq31244_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP321_PCI_LOWER_IO_VA;
-	res[0].end   = IOP321_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ31244 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP321_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP321_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ31244 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP321_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP321_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq31244_preinit(void)
-{
-	iop321_init();
-}
-
-static struct hw_pci iq31244_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iq31244_setup,
-	.scan		= iop321_scan_bus,
-	.preinit	= iq31244_preinit,
-	.map_irq	= iq31244_map_irq
-};
-
-static int __init iq31244_pci_init(void)
-{
-	if (machine_is_iq31244())
-		pci_common_init(&iq31244_pci);
-	return 0;
-}
-
-subsys_initcall(iq31244_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop3xx/iq80321-mm.c b/arch/arm/mach-iop3xx/iq80321-mm.c
deleted file mode 100644
index d9cac5e1fc3d..000000000000
--- a/arch/arm/mach-iop3xx/iq80321-mm.c
+++ /dev/null
@@ -1,45 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/mm.c
- *
- * Low level memory initialization for iq80321 platform
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-
-
-/*
- * IQ80321 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-static struct map_desc iq80321_io_desc[] __initdata = {
- 	{	/* on-board devices */
-		.virtual	= IQ80321_UART,
-		.pfn		= __phys_to_pfn(IQ80321_UART),
-		.length		= 0x00100000,
-		.type		= MT_DEVICE
-	}
-};
-
-void __init iq80321_map_io(void)
-{
-	iop321_map_io();
-
-	iotable_init(iq80321_io_desc, ARRAY_SIZE(iq80321_io_desc));
-}
diff --git a/arch/arm/mach-iop3xx/iq80321-pci.c b/arch/arm/mach-iop3xx/iq80321-pci.c
deleted file mode 100644
index d9758d3f6e7f..000000000000
--- a/arch/arm/mach-iop3xx/iq80321-pci.c
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iq80321-pci.c
- *
- * PCI support for the Intel IQ80321 reference board
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ80321_INTA
-#define INTB	IRQ_IQ80321_INTB
-#define INTC	IRQ_IQ80321_INTC
-#define INTD	IRQ_IQ80321_INTD
-
-#define INTE	IRQ_IQ80321_I82544
-
-static inline int __init
-iq80321_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][4] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-		{INTE, INTE, INTE, INTE}, /* Gig-E */
-		{-1, -1, -1, -1}, 	  /* Unused */
-		{INTC, INTD, INTA, INTB}, /* PCI-X Slot */
-		{-1, -1, -1, -1},
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-//	return PCI_IRQ_TABLE_LOOKUP(4, 7);
-	return pci_irq_table[idsel%4][pin-1];
-}
-
-static int iq80321_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP321_PCI_LOWER_IO_VA;
-	res[0].end   = IOP321_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ80321 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP321_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP321_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ80321 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP321_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP321_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq80321_preinit(void)
-{
-	iop321_init();
-}
-
-static struct hw_pci iq80321_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iq80321_setup,
-	.scan		= iop321_scan_bus,
-	.preinit	= iq80321_preinit,
-	.map_irq	= iq80321_map_irq
-};
-
-static int __init iq80321_pci_init(void)
-{
-	if (machine_is_iq80321())
-		pci_common_init(&iq80321_pci);
-	return 0;
-}
-
-subsys_initcall(iq80321_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop3xx/iq80331-mm.c b/arch/arm/mach-iop3xx/iq80331-mm.c
deleted file mode 100644
index 129eb49b0670..000000000000
--- a/arch/arm/mach-iop3xx/iq80331-mm.c
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/mm.c
- *
- * Low level memory initialization for iq80331 platform
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2003 Intel Corp.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-
-
-/*
- * IQ80331 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-
-void __init iq80331_map_io(void)
-{
-	iop331_map_io();
-}
diff --git a/arch/arm/mach-iop3xx/iq80331-pci.c b/arch/arm/mach-iop3xx/iq80331-pci.c
deleted file mode 100644
index 40d861002492..000000000000
--- a/arch/arm/mach-iop3xx/iq80331-pci.c
+++ /dev/null
@@ -1,119 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iq80331-pci.c
- *
- * PCI support for the Intel IQ80331 reference board
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2003, 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ80331_INTA
-#define INTB	IRQ_IQ80331_INTB
-#define INTC	IRQ_IQ80331_INTC
-#define INTD	IRQ_IQ80331_INTD
-
-//#define INTE	IRQ_IQ80331_I82544
-
-static inline int __init
-iq80331_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][4] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-		{INTB, INTC, INTD, INTA}, /* PCI-X Slot */
-		{INTC, INTC, INTC, INTC}, /* GigE  */
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-	return PCI_IRQ_TABLE_LOOKUP(1, 7);
-}
-
-static int iq80331_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP331_PCI_LOWER_IO_VA;
-	res[0].end   = IOP331_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ80331 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP331_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP331_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ80331 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP331_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP331_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq80331_preinit(void)
-{
-	iop331_init();
-}
-
-static struct hw_pci iq80331_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iq80331_setup,
-	.scan		= iop331_scan_bus,
-	.preinit	= iq80331_preinit,
-	.map_irq	= iq80331_map_irq
-};
-
-static int __init iq80331_pci_init(void)
-{
-	if (machine_is_iq80331())
-		pci_common_init(&iq80331_pci);
-	return 0;
-}
-
-subsys_initcall(iq80331_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop3xx/iq80332-mm.c b/arch/arm/mach-iop3xx/iq80332-mm.c
deleted file mode 100644
index 2feaf7591f53..000000000000
--- a/arch/arm/mach-iop3xx/iq80332-mm.c
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * linux/arch/arm/mach-iop3xx/mm.c
- *
- * Low level memory initialization for iq80332 platform
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-
-
-/*
- * IQ80332 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-
-void __init iq80332_map_io(void)
-{
-	iop331_map_io();
-}
diff --git a/arch/arm/mach-iop3xx/iq80332-pci.c b/arch/arm/mach-iop3xx/iq80332-pci.c
deleted file mode 100644
index afc0676318e4..000000000000
--- a/arch/arm/mach-iop3xx/iq80332-pci.c
+++ /dev/null
@@ -1,125 +0,0 @@
-/*
- * arch/arm/mach-iop3xx/iq80332-pci.c
- *
- * PCI support for the Intel IQ80332 reference board
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ80332_INTA
-#define INTB	IRQ_IQ80332_INTB
-#define INTC	IRQ_IQ80332_INTC
-#define INTD	IRQ_IQ80332_INTD
-
-//#define INTE	IRQ_IQ80332_I82544
-
-static inline int __init
-iq80332_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][8] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-		{-1,   -1,   -1,   -1},
-		{-1,   -1,   -1,   -1},
-		{-1,   -1,   -1,   -1},
-		{INTA, INTB, INTC, INTD}, /* PCI-X Slot */
-		{-1,   -1,   -1,   -1},
-		{INTC, INTC, INTC, INTC}, /* GigE  */
-		{-1,   -1,   -1,   -1},
-		{-1,   -1,   -1,   -1},
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-	return PCI_IRQ_TABLE_LOOKUP(1, 7);
-}
-
-static int iq80332_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP331_PCI_LOWER_IO_VA;
-	res[0].end   = IOP331_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ80332 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP331_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP331_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ80332 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP331_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP331_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq80332_preinit(void)
-{
-	iop331_init();
-}
-
-static struct hw_pci iq80332_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iq80332_setup,
-	.scan		= iop331_scan_bus,
-	.preinit	= iq80332_preinit,
-	.map_irq	= iq80332_map_irq
-};
-
-static int __init iq80332_pci_init(void)
-{
-	if (machine_is_iq80332())
-		pci_common_init(&iq80332_pci);
-	return 0;
-}
-
-subsys_initcall(iq80332_pci_init);
-
-
-
-
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index b4f220dd5eb8..b59c74100a84 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -232,7 +232,7 @@ config CPU_SA1100
 # XScale
 config CPU_XSCALE
 	bool
-	depends on ARCH_IOP3XX || ARCH_PXA || ARCH_IXP4XX || ARCH_IXP2000
+	depends on ARCH_IOP32X || ARCH_IOP33X || ARCH_PXA || ARCH_IXP4XX || ARCH_IXP2000
 	default y
 	select CPU_32v5
 	select CPU_ABRT_EV5T
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 884320e70403..a843d6b10a0c 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -196,7 +196,7 @@ config I2C_IBM_IIC
 
 config I2C_IOP3XX
 	tristate "Intel IOP3xx and IXP4xx on-chip I2C interface"
-	depends on (ARCH_IOP3XX || ARCH_IXP4XX) && I2C
+	depends on (ARCH_IOP32X || ARCH_IOP33X || ARCH_IXP4XX) && I2C
 	help
 	  Say Y here if you want to use the IIC bus controller on
 	  the Intel IOP3xx I/O Processors or IXP4xx Network Processors.
diff --git a/include/asm-arm/arch-iop32x/debug-macro.S b/include/asm-arm/arch-iop32x/debug-macro.S
new file mode 100644
index 000000000000..75ab2e0d8c67
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/debug-macro.S
@@ -0,0 +1,20 @@
+/* linux/include/asm-arm/arch-iop32x/debug-macro.S
+ *
+ * Debugging macro include header
+ *
+ *  Copyright (C) 1994-1999 Russell King
+ *  Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+*/
+
+		.macro	addruart,rx
+		mov	\rx, #0xfe000000	@ physical
+		orr	\rx, \rx, #0x00800000	@ location of the UART
+		.endm
+
+#define UART_SHIFT	0
+#include <asm/hardware/debug-8250.S>
diff --git a/include/asm-arm/arch-iop32x/dma.h b/include/asm-arm/arch-iop32x/dma.h
new file mode 100644
index 000000000000..5be36676e58f
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/dma.h
@@ -0,0 +1,9 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/dma.h
+ *
+ *  Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
diff --git a/include/asm-arm/arch-iop32x/entry-macro.S b/include/asm-arm/arch-iop32x/entry-macro.S
new file mode 100644
index 000000000000..52d9435c6a34
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/entry-macro.S
@@ -0,0 +1,28 @@
+/*
+ * include/asm-arm/arch-iop32x/entry-macro.S
+ *
+ * Low-level IRQ helper macros for IOP32x-based platforms
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+#include <asm/arch/irqs.h>
+
+		.macro  disable_fiq
+		.endm
+
+		/*
+		 * Note: only deal with normal interrupts, not FIQ
+		 */
+		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
+		mov     \irqnr, #0
+		mrc     p6, 0, \irqstat, c8, c0, 0      @ Read IINTSRC
+		cmp     \irqstat, #0
+		beq     1001f
+		clz     \irqnr, \irqstat
+		mov     \base, #31
+		subs    \irqnr,\base,\irqnr
+		add     \irqnr,\irqnr,#IRQ_IOP321_DMA0_EOT
+1001:
+		.endm
diff --git a/include/asm-arm/arch-iop32x/hardware.h b/include/asm-arm/arch-iop32x/hardware.h
new file mode 100644
index 000000000000..8fb10134a107
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/hardware.h
@@ -0,0 +1,54 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/hardware.h
+ */
+#ifndef __ASM_ARCH_HARDWARE_H
+#define __ASM_ARCH_HARDWARE_H
+
+#include <asm/types.h>
+
+/*
+ * Note about PCI IO space mappings
+ *
+ * To make IO space accesses efficient, we store virtual addresses in
+ * the IO resources.
+ *
+ * The PCI IO space is located at virtual 0xfe000000 from physical
+ * 0x90000000.  The PCI BARs must be programmed with physical addresses,
+ * but when we read them, we convert them to virtual addresses.  See
+ * arch/arm/mach-iop3xx/iop3xx-pci.c
+ */
+
+#define pcibios_assign_all_busses() 1
+
+
+/*
+ * The min PCI I/O and MEM space are dependent on what specific
+ * chipset/platform we are running on, so instead of hardcoding with
+ * #ifdefs, we just fill these in the platform level PCI init code.
+ */
+#ifndef __ASSEMBLY__
+extern unsigned long iop3xx_pcibios_min_io;
+extern unsigned long iop3xx_pcibios_min_mem;
+
+extern unsigned int processor_id;
+#endif
+
+/*
+ * We just set these to zero since they are really bogus anyways
+ */
+#define PCIBIOS_MIN_IO      (iop3xx_pcibios_min_io)
+#define PCIBIOS_MIN_MEM     (iop3xx_pcibios_min_mem)
+
+/*
+ * Generic chipset bits
+ *
+ */
+#include "iop321.h"
+
+/*
+ * Board specific bits
+ */
+#include "iq80321.h"
+#include "iq31244.h"
+
+#endif  /* _ASM_ARCH_HARDWARE_H */
diff --git a/include/asm-arm/arch-iop32x/io.h b/include/asm-arm/arch-iop32x/io.h
new file mode 100644
index 000000000000..36d05ada12c4
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/io.h
@@ -0,0 +1,21 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/io.h
+ *
+ *  Copyright (C) 2001  MontaVista Software, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARM_ARCH_IO_H
+#define __ASM_ARM_ARCH_IO_H
+
+#include <asm/hardware.h>
+
+#define IO_SPACE_LIMIT 0xffffffff
+
+#define __io(p)			((void __iomem *)(p))
+#define __mem_pci(a)		(a)
+
+#endif
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
new file mode 100644
index 000000000000..7ba93faf8da4
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -0,0 +1,341 @@
+/*
+ * linux/include/asm/arch-iop32x/iop321.h
+ *
+ * Intel IOP321 Chip definitions
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _IOP321_HW_H_
+#define _IOP321_HW_H_
+
+
+/*
+ * This is needed for mixed drivers that need to work on all
+ * IOP3xx variants but behave slightly differently on each.
+ */
+#ifndef __ASSEMBLY__
+#define iop_is_321()		1
+#endif
+
+/*
+ * IOP321 I/O and Mem space regions for PCI autoconfiguration
+ */
+#define IOP321_PCI_IO_WINDOW_SIZE   0x00010000
+#define IOP321_PCI_LOWER_IO_PA      0x90000000
+#define IOP321_PCI_LOWER_IO_VA      0xfe000000
+#define IOP321_PCI_LOWER_IO_BA      (*IOP321_OIOWTVR)
+#define IOP321_PCI_UPPER_IO_PA      (IOP321_PCI_LOWER_IO_PA + IOP321_PCI_IO_WINDOW_SIZE - 1)
+#define IOP321_PCI_UPPER_IO_VA      (IOP321_PCI_LOWER_IO_VA + IOP321_PCI_IO_WINDOW_SIZE - 1)
+#define IOP321_PCI_UPPER_IO_BA      (IOP321_PCI_LOWER_IO_BA + IOP321_PCI_IO_WINDOW_SIZE - 1)
+#define IOP321_PCI_IO_OFFSET        (IOP321_PCI_LOWER_IO_VA - IOP321_PCI_LOWER_IO_BA)
+
+/* #define IOP321_PCI_MEM_WINDOW_SIZE  (~*IOP321_IALR1 + 1) */
+#define IOP321_PCI_MEM_WINDOW_SIZE  0x04000000 /* 64M outbound window */
+#define IOP321_PCI_LOWER_MEM_PA     0x80000000
+#define IOP321_PCI_LOWER_MEM_BA     (*IOP321_OMWTVR0)
+#define IOP321_PCI_UPPER_MEM_PA     (IOP321_PCI_LOWER_MEM_PA + IOP321_PCI_MEM_WINDOW_SIZE - 1)
+#define IOP321_PCI_UPPER_MEM_BA     (IOP321_PCI_LOWER_MEM_BA + IOP321_PCI_MEM_WINDOW_SIZE - 1)
+#define IOP321_PCI_MEM_OFFSET       (IOP321_PCI_LOWER_MEM_PA - IOP321_PCI_LOWER_MEM_BA)
+
+
+/*
+ * IOP321 chipset registers
+ */
+#define IOP321_VIRT_MEM_BASE 0xfeffe000  /* chip virtual mem address*/
+#define IOP321_PHYS_MEM_BASE 0xffffe000  /* chip physical memory address */
+#define IOP321_REG_ADDR(reg) (IOP321_VIRT_MEM_BASE | (reg))
+
+/* Reserved 0x00000000 through 0x000000FF */
+
+/* Address Translation Unit 0x00000100 through 0x000001FF */
+#define IOP321_ATUVID     (volatile u16 *)IOP321_REG_ADDR(0x00000100)
+#define IOP321_ATUDID     (volatile u16 *)IOP321_REG_ADDR(0x00000102)
+#define IOP321_ATUCMD     (volatile u16 *)IOP321_REG_ADDR(0x00000104)
+#define IOP321_ATUSR      (volatile u16 *)IOP321_REG_ADDR(0x00000106)
+#define IOP321_ATURID     (volatile u8  *)IOP321_REG_ADDR(0x00000108)
+#define IOP321_ATUCCR     (volatile u32 *)IOP321_REG_ADDR(0x00000109)
+#define IOP321_ATUCLSR    (volatile u8  *)IOP321_REG_ADDR(0x0000010C)
+#define IOP321_ATULT      (volatile u8  *)IOP321_REG_ADDR(0x0000010D)
+#define IOP321_ATUHTR     (volatile u8  *)IOP321_REG_ADDR(0x0000010E)
+#define IOP321_ATUBIST    (volatile u8  *)IOP321_REG_ADDR(0x0000010F)
+#define IOP321_IABAR0     (volatile u32 *)IOP321_REG_ADDR(0x00000110)
+#define IOP321_IAUBAR0    (volatile u32 *)IOP321_REG_ADDR(0x00000114)
+#define IOP321_IABAR1     (volatile u32 *)IOP321_REG_ADDR(0x00000118)
+#define IOP321_IAUBAR1    (volatile u32 *)IOP321_REG_ADDR(0x0000011C)
+#define IOP321_IABAR2     (volatile u32 *)IOP321_REG_ADDR(0x00000120)
+#define IOP321_IAUBAR2    (volatile u32 *)IOP321_REG_ADDR(0x00000124)
+#define IOP321_ASVIR      (volatile u16 *)IOP321_REG_ADDR(0x0000012C)
+#define IOP321_ASIR       (volatile u16 *)IOP321_REG_ADDR(0x0000012E)
+#define IOP321_ERBAR      (volatile u32 *)IOP321_REG_ADDR(0x00000130)
+/* Reserved 0x00000134 through 0x0000013B */
+#define IOP321_ATUILR     (volatile u8  *)IOP321_REG_ADDR(0x0000013C)
+#define IOP321_ATUIPR     (volatile u8  *)IOP321_REG_ADDR(0x0000013D)
+#define IOP321_ATUMGNT    (volatile u8  *)IOP321_REG_ADDR(0x0000013E)
+#define IOP321_ATUMLAT    (volatile u8  *)IOP321_REG_ADDR(0x0000013F)
+#define IOP321_IALR0      (volatile u32 *)IOP321_REG_ADDR(0x00000140)
+#define IOP321_IATVR0     (volatile u32 *)IOP321_REG_ADDR(0x00000144)
+#define IOP321_ERLR       (volatile u32 *)IOP321_REG_ADDR(0x00000148)
+#define IOP321_ERTVR      (volatile u32 *)IOP321_REG_ADDR(0x0000014C)
+#define IOP321_IALR1      (volatile u32 *)IOP321_REG_ADDR(0x00000150)
+#define IOP321_IALR2      (volatile u32 *)IOP321_REG_ADDR(0x00000154)
+#define IOP321_IATVR2     (volatile u32 *)IOP321_REG_ADDR(0x00000158)
+#define IOP321_OIOWTVR    (volatile u32 *)IOP321_REG_ADDR(0x0000015C)
+#define IOP321_OMWTVR0    (volatile u32 *)IOP321_REG_ADDR(0x00000160)
+#define IOP321_OUMWTVR0   (volatile u32 *)IOP321_REG_ADDR(0x00000164)
+#define IOP321_OMWTVR1    (volatile u32 *)IOP321_REG_ADDR(0x00000168)
+#define IOP321_OUMWTVR1   (volatile u32 *)IOP321_REG_ADDR(0x0000016C)
+/* Reserved 0x00000170 through 0x00000177*/
+#define IOP321_OUDWTVR    (volatile u32 *)IOP321_REG_ADDR(0x00000178)
+/* Reserved 0x0000017C through 0x0000017F*/
+#define IOP321_ATUCR      (volatile u32 *)IOP321_REG_ADDR(0x00000180)
+#define IOP321_PCSR       (volatile u32 *)IOP321_REG_ADDR(0x00000184)
+#define IOP321_ATUISR     (volatile u32 *)IOP321_REG_ADDR(0x00000188)
+#define IOP321_ATUIMR     (volatile u32 *)IOP321_REG_ADDR(0x0000018C)
+#define IOP321_IABAR3     (volatile u32 *)IOP321_REG_ADDR(0x00000190)
+#define IOP321_IAUBAR3    (volatile u32 *)IOP321_REG_ADDR(0x00000194)
+#define IOP321_IALR3      (volatile u32 *)IOP321_REG_ADDR(0x00000198)
+#define IOP321_IATVR3     (volatile u32 *)IOP321_REG_ADDR(0x0000019C)
+/* Reserved 0x000001A0 through 0x000001A3*/
+#define IOP321_OCCAR      (volatile u32 *)IOP321_REG_ADDR(0x000001A4)
+/* Reserved 0x000001A8 through 0x000001AB*/
+#define IOP321_OCCDR      (volatile u32 *)IOP321_REG_ADDR(0x000001AC)
+/* Reserved 0x000001B0 through 0x000001BB*/
+#define IOP321_PDSCR      (volatile u32 *)IOP321_REG_ADDR(0x000001BC)
+#define IOP321_PMCAPID    (volatile u8  *)IOP321_REG_ADDR(0x000001C0)
+#define IOP321_PMNEXT     (volatile u8  *)IOP321_REG_ADDR(0x000001C1)
+#define IOP321_APMCR      (volatile u16 *)IOP321_REG_ADDR(0x000001C2)
+#define IOP321_APMCSR     (volatile u16 *)IOP321_REG_ADDR(0x000001C4)
+/* Reserved 0x000001C6 through 0x000001DF */
+#define IOP321_PCIXCAPID  (volatile u8  *)IOP321_REG_ADDR(0x000001E0)
+#define IOP321_PCIXNEXT   (volatile u8  *)IOP321_REG_ADDR(0x000001E1)
+#define IOP321_PCIXCMD    (volatile u16 *)IOP321_REG_ADDR(0x000001E2)
+#define IOP321_PCIXSR     (volatile u32 *)IOP321_REG_ADDR(0x000001E4)
+#define IOP321_PCIIRSR    (volatile u32 *)IOP321_REG_ADDR(0x000001EC)
+
+/* Messaging Unit 0x00000300 through 0x000003FF */
+
+/* Reserved 0x00000300 through 0x0000030c */
+#define IOP321_IMR0       (volatile u32 *)IOP321_REG_ADDR(0x00000310)
+#define IOP321_IMR1       (volatile u32 *)IOP321_REG_ADDR(0x00000314)
+#define IOP321_OMR0       (volatile u32 *)IOP321_REG_ADDR(0x00000318)
+#define IOP321_OMR1       (volatile u32 *)IOP321_REG_ADDR(0x0000031C)
+#define IOP321_IDR        (volatile u32 *)IOP321_REG_ADDR(0x00000320)
+#define IOP321_IISR       (volatile u32 *)IOP321_REG_ADDR(0x00000324)
+#define IOP321_IIMR       (volatile u32 *)IOP321_REG_ADDR(0x00000328)
+#define IOP321_ODR        (volatile u32 *)IOP321_REG_ADDR(0x0000032C)
+#define IOP321_OISR       (volatile u32 *)IOP321_REG_ADDR(0x00000330)
+#define IOP321_OIMR       (volatile u32 *)IOP321_REG_ADDR(0x00000334)
+/* Reserved 0x00000338 through 0x0000034F */
+#define IOP321_MUCR       (volatile u32 *)IOP321_REG_ADDR(0x00000350)
+#define IOP321_QBAR       (volatile u32 *)IOP321_REG_ADDR(0x00000354)
+/* Reserved 0x00000358 through 0x0000035C */
+#define IOP321_IFHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000360)
+#define IOP321_IFTPR      (volatile u32 *)IOP321_REG_ADDR(0x00000364)
+#define IOP321_IPHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000368)
+#define IOP321_IPTPR      (volatile u32 *)IOP321_REG_ADDR(0x0000036C)
+#define IOP321_OFHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000370)
+#define IOP321_OFTPR      (volatile u32 *)IOP321_REG_ADDR(0x00000374)
+#define IOP321_OPHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000378)
+#define IOP321_OPTPR      (volatile u32 *)IOP321_REG_ADDR(0x0000037C)
+#define IOP321_IAR        (volatile u32 *)IOP321_REG_ADDR(0x00000380)
+
+#define IOP321_IIxR_MASK	0x7f /* masks all */
+#define IOP321_IIxR_IRI		0x40 /* RC Index Register Interrupt */
+#define IOP321_IIxR_OFQF	0x20 /* RC Output Free Q Full (ERROR) */
+#define IOP321_IIxR_ipq		0x10 /* RC Inbound Post Q (post) */
+#define IOP321_IIxR_ERRDI	0x08 /* RO Error Doorbell Interrupt */
+#define IOP321_IIxR_IDI		0x04 /* RO Inbound Doorbell Interrupt */
+#define IOP321_IIxR_IM1		0x02 /* RC Inbound Message 1 Interrupt */
+#define IOP321_IIxR_IM0		0x01 /* RC Inbound Message 0 Interrupt */
+
+/* Reserved 0x00000384 through 0x000003FF */
+
+/* DMA Controller 0x00000400 through 0x000004FF */
+#define IOP321_DMA0_CCR   (volatile u32 *)IOP321_REG_ADDR(0x00000400)
+#define IOP321_DMA0_CSR   (volatile u32 *)IOP321_REG_ADDR(0x00000404)
+#define IOP321_DMA0_DAR   (volatile u32 *)IOP321_REG_ADDR(0x0000040C)
+#define IOP321_DMA0_NDAR  (volatile u32 *)IOP321_REG_ADDR(0x00000410)
+#define IOP321_DMA0_PADR  (volatile u32 *)IOP321_REG_ADDR(0x00000414)
+#define IOP321_DMA0_PUADR (volatile u32 *)IOP321_REG_ADDR(0x00000418)
+#define IOP321_DMA0_LADR  (volatile u32 *)IOP321_REG_ADDR(0X0000041C)
+#define IOP321_DMA0_BCR   (volatile u32 *)IOP321_REG_ADDR(0x00000420)
+#define IOP321_DMA0_DCR   (volatile u32 *)IOP321_REG_ADDR(0x00000424)
+/* Reserved 0x00000428 through 0x0000043C */
+#define IOP321_DMA1_CCR   (volatile u32 *)IOP321_REG_ADDR(0x00000440)
+#define IOP321_DMA1_CSR   (volatile u32 *)IOP321_REG_ADDR(0x00000444)
+#define IOP321_DMA1_DAR   (volatile u32 *)IOP321_REG_ADDR(0x0000044C)
+#define IOP321_DMA1_NDAR  (volatile u32 *)IOP321_REG_ADDR(0x00000450)
+#define IOP321_DMA1_PADR  (volatile u32 *)IOP321_REG_ADDR(0x00000454)
+#define IOP321_DMA1_PUADR (volatile u32 *)IOP321_REG_ADDR(0x00000458)
+#define IOP321_DMA1_LADR  (volatile u32 *)IOP321_REG_ADDR(0x0000045C)
+#define IOP321_DMA1_BCR   (volatile u32 *)IOP321_REG_ADDR(0x00000460)
+#define IOP321_DMA1_DCR   (volatile u32 *)IOP321_REG_ADDR(0x00000464)
+/* Reserved 0x00000468 through 0x000004FF */
+
+/* Memory controller 0x00000500 through 0x0005FF */
+
+/* Peripheral bus interface unit 0x00000680 through 0x0006FF */
+#define IOP321_PBCR       (volatile u32 *)IOP321_REG_ADDR(0x00000680)
+#define IOP321_PBISR      (volatile u32 *)IOP321_REG_ADDR(0x00000684)
+#define IOP321_PBBAR0     (volatile u32 *)IOP321_REG_ADDR(0x00000688)
+#define IOP321_PBLR0      (volatile u32 *)IOP321_REG_ADDR(0x0000068C)
+#define IOP321_PBBAR1     (volatile u32 *)IOP321_REG_ADDR(0x00000690)
+#define IOP321_PBLR1      (volatile u32 *)IOP321_REG_ADDR(0x00000694)
+#define IOP321_PBBAR2     (volatile u32 *)IOP321_REG_ADDR(0x00000698)
+#define IOP321_PBLR2      (volatile u32 *)IOP321_REG_ADDR(0x0000069C)
+#define IOP321_PBBAR3     (volatile u32 *)IOP321_REG_ADDR(0x000006A0)
+#define IOP321_PBLR3      (volatile u32 *)IOP321_REG_ADDR(0x000006A4)
+#define IOP321_PBBAR4     (volatile u32 *)IOP321_REG_ADDR(0x000006A8)
+#define IOP321_PBLR4      (volatile u32 *)IOP321_REG_ADDR(0x000006AC)
+#define IOP321_PBBAR5     (volatile u32 *)IOP321_REG_ADDR(0x000006B0)
+#define IOP321_PBLR5      (volatile u32 *)IOP321_REG_ADDR(0x000006B4)
+#define IOP321_PBDSCR     (volatile u32 *)IOP321_REG_ADDR(0x000006B8)
+/* Reserved 0x000006BC */
+#define IOP321_PMBR0      (volatile u32 *)IOP321_REG_ADDR(0x000006C0)
+/* Reserved 0x000006C4 through 0x000006DC */
+#define IOP321_PMBR1      (volatile u32 *)IOP321_REG_ADDR(0x000006E0)
+#define IOP321_PMBR2      (volatile u32 *)IOP321_REG_ADDR(0x000006E4)
+
+#define IOP321_PBCR_EN    0x1
+
+#define IOP321_PBISR_BOOR_ERR 0x1
+
+/* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
+#define IOP321_GTMR	(volatile u32 *)IOP321_REG_ADDR(0x00000700)
+#define IOP321_ESR	(volatile u32 *)IOP321_REG_ADDR(0x00000704)
+#define IOP321_EMISR	(volatile u32 *)IOP321_REG_ADDR(0x00000708)
+/* reserved 0x00000070c */
+#define IOP321_GTSR	(volatile u32 *)IOP321_REG_ADDR(0x00000710)
+/* PERC0 DOESN'T EXIST - index from 1! */
+#define IOP321_PERCR0	(volatile u32 *)IOP321_REG_ADDR(0x00000710)
+
+#define IOP321_GTMR_NGCE	0x04 /* (Not) Global Counter Enable */
+
+/* Internal arbitration unit 0x00000780 through 0x0007BF */
+#define IOP321_IACR	(volatile u32 *)IOP321_REG_ADDR(0x00000780)
+#define IOP321_MTTR1	(volatile u32 *)IOP321_REG_ADDR(0x00000784)
+#define IOP321_MTTR2	(volatile u32 *)IOP321_REG_ADDR(0x00000788)
+
+/* General Purpose I/O Registers */
+#define IOP321_GPOE       (volatile u32 *)IOP321_REG_ADDR(0x000007C4)
+#define IOP321_GPID       (volatile u32 *)IOP321_REG_ADDR(0x000007C8)
+#define IOP321_GPOD       (volatile u32 *)IOP321_REG_ADDR(0x000007CC)
+
+/* Interrupt Controller */
+#define IOP321_INTCTL     (volatile u32 *)IOP321_REG_ADDR(0x000007D0)
+#define IOP321_INTSTR     (volatile u32 *)IOP321_REG_ADDR(0x000007D4)
+#define IOP321_IINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007D8)
+#define IOP321_FINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007DC)
+
+/* Timers */
+
+#define IOP321_TU_TMR0		(volatile u32 *)IOP321_REG_ADDR(0x000007E0)
+#define IOP321_TU_TMR1		(volatile u32 *)IOP321_REG_ADDR(0x000007E4)
+
+#ifdef CONFIG_ARCH_IQ80321
+#define	IOP321_TICK_RATE	200000000	/* 200 MHz clock */
+#elif defined(CONFIG_ARCH_IQ31244)
+#define IOP321_TICK_RATE	198000000	/* 33.000 MHz crystal */
+#endif
+
+#ifdef CONFIG_ARCH_EP80219
+#undef IOP321_TICK_RATE
+#define IOP321_TICK_RATE 200000000 /* 33.333333 Mhz crystal */
+#endif
+
+#define IOP321_TMR_TC		0x01
+#define	IOP321_TMR_EN		0x02
+#define IOP321_TMR_RELOAD	0x04
+#define	IOP321_TMR_PRIVILEGED	0x09
+
+#define	IOP321_TMR_RATIO_1_1	0x00
+#define	IOP321_TMR_RATIO_4_1	0x10
+#define	IOP321_TMR_RATIO_8_1	0x20
+#define	IOP321_TMR_RATIO_16_1	0x30
+
+#define IOP321_TU_TCR0    (volatile u32 *)IOP321_REG_ADDR(0x000007E8)
+#define IOP321_TU_TCR1    (volatile u32 *)IOP321_REG_ADDR(0x000007EC)
+#define IOP321_TU_TRR0    (volatile u32 *)IOP321_REG_ADDR(0x000007F0)
+#define IOP321_TU_TRR1    (volatile u32 *)IOP321_REG_ADDR(0x000007F4)
+#define IOP321_TU_TISR    (volatile u32 *)IOP321_REG_ADDR(0x000007F8)
+#define IOP321_TU_WDTCR   (volatile u32 *)IOP321_REG_ADDR(0x000007FC)
+
+/* Application accelerator unit 0x00000800 - 0x000008FF */
+#define IOP321_AAU_ACR     (volatile u32 *)IOP321_REG_ADDR(0x00000800)
+#define IOP321_AAU_ASR     (volatile u32 *)IOP321_REG_ADDR(0x00000804)
+#define IOP321_AAU_ADAR    (volatile u32 *)IOP321_REG_ADDR(0x00000808)
+#define IOP321_AAU_ANDAR   (volatile u32 *)IOP321_REG_ADDR(0x0000080C)
+#define IOP321_AAU_SAR1    (volatile u32 *)IOP321_REG_ADDR(0x00000810)
+#define IOP321_AAU_SAR2    (volatile u32 *)IOP321_REG_ADDR(0x00000814)
+#define IOP321_AAU_SAR3    (volatile u32 *)IOP321_REG_ADDR(0x00000818)
+#define IOP321_AAU_SAR4    (volatile u32 *)IOP321_REG_ADDR(0x0000081C)
+#define IOP321_AAU_SAR5    (volatile u32 *)IOP321_REG_ADDR(0x0000082C)
+#define IOP321_AAU_SAR6    (volatile u32 *)IOP321_REG_ADDR(0x00000830)
+#define IOP321_AAU_SAR7    (volatile u32 *)IOP321_REG_ADDR(0x00000834)
+#define IOP321_AAU_SAR8    (volatile u32 *)IOP321_REG_ADDR(0x00000838)
+#define IOP321_AAU_SAR9    (volatile u32 *)IOP321_REG_ADDR(0x00000840)
+#define IOP321_AAU_SAR10   (volatile u32 *)IOP321_REG_ADDR(0x00000844)
+#define IOP321_AAU_SAR11   (volatile u32 *)IOP321_REG_ADDR(0x00000848)
+#define IOP321_AAU_SAR12   (volatile u32 *)IOP321_REG_ADDR(0x0000084C)
+#define IOP321_AAU_SAR13   (volatile u32 *)IOP321_REG_ADDR(0x00000850)
+#define IOP321_AAU_SAR14   (volatile u32 *)IOP321_REG_ADDR(0x00000854)
+#define IOP321_AAU_SAR15   (volatile u32 *)IOP321_REG_ADDR(0x00000858)
+#define IOP321_AAU_SAR16   (volatile u32 *)IOP321_REG_ADDR(0x0000085C)
+#define IOP321_AAU_SAR17   (volatile u32 *)IOP321_REG_ADDR(0x00000864)
+#define IOP321_AAU_SAR18   (volatile u32 *)IOP321_REG_ADDR(0x00000868)
+#define IOP321_AAU_SAR19   (volatile u32 *)IOP321_REG_ADDR(0x0000086C)
+#define IOP321_AAU_SAR20   (volatile u32 *)IOP321_REG_ADDR(0x00000870)
+#define IOP321_AAU_SAR21   (volatile u32 *)IOP321_REG_ADDR(0x00000874)
+#define IOP321_AAU_SAR22   (volatile u32 *)IOP321_REG_ADDR(0x00000878)
+#define IOP321_AAU_SAR23   (volatile u32 *)IOP321_REG_ADDR(0x0000087C)
+#define IOP321_AAU_SAR24   (volatile u32 *)IOP321_REG_ADDR(0x00000880)
+#define IOP321_AAU_SAR25   (volatile u32 *)IOP321_REG_ADDR(0x00000888)
+#define IOP321_AAU_SAR26   (volatile u32 *)IOP321_REG_ADDR(0x0000088C)
+#define IOP321_AAU_SAR27   (volatile u32 *)IOP321_REG_ADDR(0x00000890)
+#define IOP321_AAU_SAR28   (volatile u32 *)IOP321_REG_ADDR(0x00000894)
+#define IOP321_AAU_SAR29   (volatile u32 *)IOP321_REG_ADDR(0x00000898)
+#define IOP321_AAU_SAR30   (volatile u32 *)IOP321_REG_ADDR(0x0000089C)
+#define IOP321_AAU_SAR31   (volatile u32 *)IOP321_REG_ADDR(0x000008A0)
+#define IOP321_AAU_SAR32   (volatile u32 *)IOP321_REG_ADDR(0x000008A4)
+#define IOP321_AAU_DAR     (volatile u32 *)IOP321_REG_ADDR(0x00000820)
+#define IOP321_AAU_ABCR    (volatile u32 *)IOP321_REG_ADDR(0x00000824)
+#define IOP321_AAU_ADCR    (volatile u32 *)IOP321_REG_ADDR(0x00000828)
+#define IOP321_AAU_EDCR0   (volatile u32 *)IOP321_REG_ADDR(0x0000083c)
+#define IOP321_AAU_EDCR1   (volatile u32 *)IOP321_REG_ADDR(0x00000860)
+#define IOP321_AAU_EDCR2   (volatile u32 *)IOP321_REG_ADDR(0x00000884)
+
+
+/* SSP serial port unit 0x00001600 - 0x0000167F */
+/* I2C bus interface unit 0x00001680 - 0x000016FF */
+#define IOP321_ICR0       (volatile u32 *)IOP321_REG_ADDR(0x00001680)
+#define IOP321_ISR0       (volatile u32 *)IOP321_REG_ADDR(0x00001684)
+#define IOP321_ISAR0      (volatile u32 *)IOP321_REG_ADDR(0x00001688)
+#define IOP321_IDBR0      (volatile u32 *)IOP321_REG_ADDR(0x0000168C)
+/* Reserved 0x00001690 */
+#define IOP321_IBMR0      (volatile u32 *)IOP321_REG_ADDR(0x00001694)
+/* Reserved 0x00001698 */
+/* Reserved 0x0000169C */
+#define IOP321_ICR1       (volatile u32 *)IOP321_REG_ADDR(0x000016A0)
+#define IOP321_ISR1       (volatile u32 *)IOP321_REG_ADDR(0x000016A4)
+#define IOP321_ISAR1      (volatile u32 *)IOP321_REG_ADDR(0x000016A8)
+#define IOP321_IDBR1      (volatile u32 *)IOP321_REG_ADDR(0x000016AC)
+#define IOP321_IBMR1      (volatile u32 *)IOP321_REG_ADDR(0x000016B4)
+/* Reserved 0x000016B8 through 0x000016FC */
+
+/* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
+
+
+#ifndef __ASSEMBLY__
+extern void iop321_map_io(void);
+extern void iop321_init_irq(void);
+extern void iop321_time_init(void);
+#endif
+
+#endif // _IOP321_HW_H_
diff --git a/include/asm-arm/arch-iop32x/iq31244.h b/include/asm-arm/arch-iop32x/iq31244.h
new file mode 100644
index 000000000000..f490063d2156
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/iq31244.h
@@ -0,0 +1,24 @@
+/*
+ * linux/include/asm/arch-iop32x/iq31244.h
+ *
+ * Intel IQ31244 evaluation board registers
+ */
+
+#ifndef _IQ31244_H_
+#define _IQ31244_H_
+
+#define	IQ31244_FLASHBASE	0xf0000000	/* Flash */
+#define	IQ31244_FLASHSIZE	0x00800000
+#define	IQ31244_FLASHWIDTH	2
+
+#define IQ31244_UART		0xfe800000	/* UART #1 */
+#define IQ31244_7SEG_1		0xfe840000	/* 7-Segment MSB */
+#define IQ31244_7SEG_0		0xfe850000	/* 7-Segment LSB (WO) */
+#define IQ31244_ROTARY_SW	0xfe8d0000	/* Rotary Switch */
+#define IQ31244_BATT_STAT	0xfe8f0000	/* Battery Status */
+
+#ifndef __ASSEMBLY__
+extern void iq31244_map_io(void);
+#endif
+
+#endif	// _IQ31244_H_
diff --git a/include/asm-arm/arch-iop32x/iq80321.h b/include/asm-arm/arch-iop32x/iq80321.h
new file mode 100644
index 000000000000..7015a605ab64
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/iq80321.h
@@ -0,0 +1,24 @@
+/*
+ * linux/include/asm/arch-iop32x/iq80321.h
+ *
+ * Intel IQ80321 evaluation board registers
+ */
+
+#ifndef _IQ80321_H_
+#define _IQ80321_H_
+
+#define	IQ80321_FLASHBASE	0xf0000000	/* Flash */
+#define	IQ80321_FLASHSIZE	0x00800000
+#define	IQ80321_FLASHWIDTH	1
+
+#define IQ80321_UART		0xfe800000	/* UART #1 */
+#define IQ80321_7SEG_1		0xfe840000	/* 7-Segment MSB */
+#define IQ80321_7SEG_0		0xfe850000	/* 7-Segment LSB (WO) */
+#define IQ80321_ROTARY_SW	0xfe8d0000	/* Rotary Switch */
+#define IQ80321_BATT_STAT	0xfe8f0000	/* Battery Status */
+
+#ifndef __ASSEMBLY__
+extern void iq80321_map_io(void);
+#endif
+
+#endif	// _IQ80321_H_
diff --git a/include/asm-arm/arch-iop32x/irqs.h b/include/asm-arm/arch-iop32x/irqs.h
new file mode 100644
index 000000000000..4b0c82711f96
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/irqs.h
@@ -0,0 +1,98 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/irqs.h
+ *
+ * Author:	Rory Bolt <rorybolt@pacbell.net>
+ * Copyright:	(C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#ifndef _IRQS_H_
+#define _IRQS_H_
+
+/*
+ * IOP80321 chipset interrupts
+ */
+#define IOP321_IRQ_OFS		0
+#define IOP321_IRQ(x)		(IOP321_IRQ_OFS + (x))
+
+/*
+ * On IRQ or FIQ register
+ */
+#define IRQ_IOP321_DMA0_EOT	IOP321_IRQ(0)
+#define IRQ_IOP321_DMA0_EOC	IOP321_IRQ(1)
+#define IRQ_IOP321_DMA1_EOT	IOP321_IRQ(2)
+#define IRQ_IOP321_DMA1_EOC	IOP321_IRQ(3)
+#define IRQ_IOP321_RSVD_4	IOP321_IRQ(4)
+#define IRQ_IOP321_RSVD_5	IOP321_IRQ(5)
+#define IRQ_IOP321_AA_EOT	IOP321_IRQ(6)
+#define IRQ_IOP321_AA_EOC	IOP321_IRQ(7)
+#define IRQ_IOP321_CORE_PMON	IOP321_IRQ(8)
+#define IRQ_IOP321_TIMER0	IOP321_IRQ(9)
+#define IRQ_IOP321_TIMER1	IOP321_IRQ(10)
+#define IRQ_IOP321_I2C_0	IOP321_IRQ(11)
+#define IRQ_IOP321_I2C_1	IOP321_IRQ(12)
+#define IRQ_IOP321_MESSAGING	IOP321_IRQ(13)
+#define IRQ_IOP321_ATU_BIST	IOP321_IRQ(14)
+#define IRQ_IOP321_PERFMON	IOP321_IRQ(15)
+#define IRQ_IOP321_CORE_PMU	IOP321_IRQ(16)
+#define IRQ_IOP321_BIU_ERR	IOP321_IRQ(17)
+#define IRQ_IOP321_ATU_ERR	IOP321_IRQ(18)
+#define IRQ_IOP321_MCU_ERR	IOP321_IRQ(19)
+#define IRQ_IOP321_DMA0_ERR	IOP321_IRQ(20)
+#define IRQ_IOP321_DMA1_ERR	IOP321_IRQ(21)
+#define IRQ_IOP321_RSVD_22	IOP321_IRQ(22)
+#define IRQ_IOP321_AA_ERR	IOP321_IRQ(23)
+#define IRQ_IOP321_MSG_ERR	IOP321_IRQ(24)
+#define IRQ_IOP321_SSP		IOP321_IRQ(25)
+#define IRQ_IOP321_RSVD_26	IOP321_IRQ(26)
+#define IRQ_IOP321_XINT0	IOP321_IRQ(27)
+#define IRQ_IOP321_XINT1	IOP321_IRQ(28)
+#define IRQ_IOP321_XINT2	IOP321_IRQ(29)
+#define IRQ_IOP321_XINT3	IOP321_IRQ(30)
+#define IRQ_IOP321_HPI		IOP321_IRQ(31)
+
+#define NR_IRQS			(IOP321_IRQ(31) + 1)
+
+
+/*
+ * Interrupts available on the IQ80321 board
+ */
+
+/*
+ * On board devices
+ */
+#define	IRQ_IQ80321_I82544	IRQ_IOP321_XINT0
+#define IRQ_IQ80321_UART	IRQ_IOP321_XINT1
+
+/*
+ * PCI interrupts
+ */
+#define	IRQ_IQ80321_INTA	IRQ_IOP321_XINT0
+#define	IRQ_IQ80321_INTB	IRQ_IOP321_XINT1
+#define	IRQ_IQ80321_INTC	IRQ_IOP321_XINT2
+#define	IRQ_IQ80321_INTD	IRQ_IOP321_XINT3
+
+/*
+ * Interrupts on the IQ31244 board
+ */
+
+/*
+ * On board devices
+ */
+#define IRQ_IQ31244_UART	IRQ_IOP321_XINT1
+#define	IRQ_IQ31244_I82546	IRQ_IOP321_XINT0
+#define IRQ_IQ31244_SATA	IRQ_IOP321_XINT2
+#define	IRQ_IQ31244_PCIX_SLOT	IRQ_IOP321_XINT3
+
+/*
+ * PCI interrupts
+ */
+#define	IRQ_IQ31244_INTA	IRQ_IOP321_XINT0
+#define	IRQ_IQ31244_INTB	IRQ_IOP321_XINT1
+#define	IRQ_IQ31244_INTC	IRQ_IOP321_XINT2
+#define	IRQ_IQ31244_INTD	IRQ_IOP321_XINT3
+
+#endif // _IRQ_H_
diff --git a/include/asm-arm/arch-iop32x/memory.h b/include/asm-arm/arch-iop32x/memory.h
new file mode 100644
index 000000000000..b4073f15b405
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/memory.h
@@ -0,0 +1,27 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/memory.h
+ */
+
+#ifndef __ASM_ARCH_MEMORY_H
+#define __ASM_ARCH_MEMORY_H
+
+#include <asm/hardware.h>
+
+/*
+ * Physical DRAM offset.
+ */
+#define PHYS_OFFSET	UL(0xa0000000)
+
+/*
+ * Virtual view <-> PCI DMA view memory address translations
+ * virt_to_bus: Used to translate the virtual address to an
+ *		address suitable to be passed to set_dma_addr
+ * bus_to_virt: Used to convert an address for DMA operations
+ *		to an address that the kernel can use.
+ */
+
+#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP321_IATVR2)) | ((*IOP321_IABAR2) & 0xfffffff0))
+#define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP321_IALR2)) | ( *IOP321_IATVR2)))
+
+
+#endif
diff --git a/include/asm-arm/arch-iop32x/system.h b/include/asm-arm/arch-iop32x/system.h
new file mode 100644
index 000000000000..d4c8d691e1b0
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/system.h
@@ -0,0 +1,29 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/system.h
+ *
+ *  Copyright (C) 2001 MontaVista Software, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+static inline void arch_idle(void)
+{
+	cpu_do_idle();
+}
+
+
+static inline void arch_reset(char mode)
+{
+    *IOP321_PCSR = 0x30;
+
+	if ( 1 && mode == 's') {
+		/* Jump into ROM at address 0 */
+		cpu_reset(0);
+	} else {
+		/* No on-chip reset capability */
+		cpu_reset(0);
+	}
+}
+
diff --git a/include/asm-arm/arch-iop32x/timex.h b/include/asm-arm/arch-iop32x/timex.h
new file mode 100644
index 000000000000..08badde2e820
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/timex.h
@@ -0,0 +1,8 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/timex.h
+ *
+ * IOP3xx architecture timex specifications
+ */
+#include <asm/hardware.h>
+
+#define CLOCK_TICK_RATE IOP321_TICK_RATE
diff --git a/include/asm-arm/arch-iop32x/uncompress.h b/include/asm-arm/arch-iop32x/uncompress.h
new file mode 100644
index 000000000000..4a85f20c796f
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/uncompress.h
@@ -0,0 +1,38 @@
+/*
+ *  linux/include/asm-arm/arch-iop32x/uncompress.h
+ */
+#include <asm/types.h>
+#include <asm/mach-types.h>
+#include <linux/serial_reg.h>
+#include <asm/hardware.h>
+
+static volatile u8 *uart_base;
+
+#define TX_DONE (UART_LSR_TEMT|UART_LSR_THRE)
+
+static inline void putc(char c)
+{
+	while ((uart_base[UART_LSR] & TX_DONE) != TX_DONE)
+		barrier();
+	*uart_base = c;
+}
+
+static inline void flush(void)
+{
+}
+
+static __inline__ void __arch_decomp_setup(unsigned long arch_id)
+{
+	if (machine_is_iq80321())
+		uart_base = (volatile u8 *)IQ80321_UART;
+	else if (machine_is_iq31244())
+		uart_base = (volatile u8 *)IQ31244_UART;
+	else
+		uart_base = (volatile u8 *)0xfe800000;
+}
+
+/*
+ * nothing to do
+ */
+#define arch_decomp_setup()	__arch_decomp_setup(arch_id)
+#define arch_decomp_wdog()
diff --git a/include/asm-arm/arch-iop32x/vmalloc.h b/include/asm-arm/arch-iop32x/vmalloc.h
new file mode 100644
index 000000000000..8492e1708a63
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/vmalloc.h
@@ -0,0 +1,16 @@
+/*
+ * linux/include/asm-arm/arch-iop32x/vmalloc.h
+ */
+
+/*
+ * Just any arbitrary offset to the start of the vmalloc VM area: the
+ * current 8MB value just means that there will be a 8MB "hole" after the
+ * physical memory until the kernel virtual memory starts.  That means that
+ * any out-of-bounds memory accesses will hopefully be caught.
+ * The vmalloc() routines leaves a hole of 4kB between each vmalloced
+ * area for the same reason. ;)
+ */
+//#define VMALLOC_END       (0xe8000000)
+/* increase usable physical RAM to ~992M per RMK */
+#define VMALLOC_END       (0xfe000000)
+
diff --git a/include/asm-arm/arch-iop33x/debug-macro.S b/include/asm-arm/arch-iop33x/debug-macro.S
new file mode 100644
index 000000000000..b647edff475d
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/debug-macro.S
@@ -0,0 +1,24 @@
+/* linux/include/asm-arm/arch-iop33x/debug-macro.S
+ *
+ * Debugging macro include header
+ *
+ *  Copyright (C) 1994-1999 Russell King
+ *  Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+*/
+
+		.macro	addruart,rx
+		mrc	p15, 0, \rx, c1, c0
+		tst	\rx, #1			@ mmu enabled?
+		moveq	\rx, #0xff000000	@ physical
+		movne	\rx, #0xfe000000	@ virtual
+		orr	\rx, \rx, #0x00ff0000
+		orr	\rx, \rx, #0x0000f700
+		.endm
+
+#define UART_SHIFT	2
+#include <asm/hardware/debug-8250.S>
diff --git a/include/asm-arm/arch-iop33x/dma.h b/include/asm-arm/arch-iop33x/dma.h
new file mode 100644
index 000000000000..d577ca59f4b0
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/dma.h
@@ -0,0 +1,9 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/dma.h
+ *
+ *  Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
diff --git a/include/asm-arm/arch-iop33x/entry-macro.S b/include/asm-arm/arch-iop33x/entry-macro.S
new file mode 100644
index 000000000000..980ec9b1ac83
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/entry-macro.S
@@ -0,0 +1,34 @@
+/*
+ * include/asm-arm/arch-iop33x/entry-macro.S
+ *
+ * Low-level IRQ helper macros for IOP33x-based platforms
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+#include <asm/arch/irqs.h>
+
+		.macro  disable_fiq
+		.endm
+
+		/*
+		 * Note: only deal with normal interrupts, not FIQ
+		 */
+		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
+		mov     \irqnr, #0
+		mrc     p6, 0, \irqstat, c4, c0, 0      @ Read IINTSRC0
+		cmp     \irqstat, #0
+		bne     1002f
+		mrc     p6, 0, \irqstat, c5, c0, 0      @ Read IINTSRC1
+		cmp	\irqstat, #0
+		beq	1001f
+		clz	\irqnr, \irqstat
+		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
+		add	\irqnr,\irqnr,#IRQ_IOP331_XINT8
+		b	1001f
+1002:		clz     \irqnr, \irqstat
+		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
+		add     \irqnr,\irqnr,#IRQ_IOP331_DMA0_EOT
+1001:
+		.endm
diff --git a/include/asm-arm/arch-iop33x/hardware.h b/include/asm-arm/arch-iop33x/hardware.h
new file mode 100644
index 000000000000..4a457084c5c6
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/hardware.h
@@ -0,0 +1,54 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/hardware.h
+ */
+#ifndef __ASM_ARCH_HARDWARE_H
+#define __ASM_ARCH_HARDWARE_H
+
+#include <asm/types.h>
+
+/*
+ * Note about PCI IO space mappings
+ *
+ * To make IO space accesses efficient, we store virtual addresses in
+ * the IO resources.
+ *
+ * The PCI IO space is located at virtual 0xfe000000 from physical
+ * 0x90000000.  The PCI BARs must be programmed with physical addresses,
+ * but when we read them, we convert them to virtual addresses.  See
+ * arch/arm/mach-iop33x/pci.c
+ */
+
+#define pcibios_assign_all_busses() 1
+
+
+/*
+ * The min PCI I/O and MEM space are dependent on what specific
+ * chipset/platform we are running on, so instead of hardcoding with
+ * #ifdefs, we just fill these in the platform level PCI init code.
+ */
+#ifndef __ASSEMBLY__
+extern unsigned long iop3xx_pcibios_min_io;
+extern unsigned long iop3xx_pcibios_min_mem;
+
+extern unsigned int processor_id;
+#endif
+
+/*
+ * We just set these to zero since they are really bogus anyways
+ */
+#define PCIBIOS_MIN_IO      (iop3xx_pcibios_min_io)
+#define PCIBIOS_MIN_MEM     (iop3xx_pcibios_min_mem)
+
+/*
+ * Generic chipset bits
+ *
+ */
+#include "iop331.h"
+
+/*
+ * Board specific bits
+ */
+#include "iq80331.h"
+#include "iq80332.h"
+
+#endif  /* _ASM_ARCH_HARDWARE_H */
diff --git a/include/asm-arm/arch-iop33x/io.h b/include/asm-arm/arch-iop33x/io.h
new file mode 100644
index 000000000000..a9949d5d4953
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/io.h
@@ -0,0 +1,21 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/io.h
+ *
+ *  Copyright (C) 2001  MontaVista Software, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARM_ARCH_IO_H
+#define __ASM_ARM_ARCH_IO_H
+
+#include <asm/hardware.h>
+
+#define IO_SPACE_LIMIT 0xffffffff
+
+#define __io(p)			((void __iomem *)(p))
+#define __mem_pci(a)		(a)
+
+#endif
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
new file mode 100644
index 000000000000..780b707edb1e
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -0,0 +1,358 @@
+/*
+ * linux/include/asm/arch-iop33x/iop331.h
+ *
+ * Intel IOP331 Chip definitions
+ *
+ * Author: Dave Jiang (dave.jiang@intel.com)
+ * Copyright (C) 2003, 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _IOP331_HW_H_
+#define _IOP331_HW_H_
+
+
+/*
+ * This is needed for mixed drivers that need to work on all
+ * IOP3xx variants but behave slightly differently on each.
+ */
+#ifndef __ASSEMBLY__
+#define	iop_is_331()		1
+#endif
+
+/*
+ * IOP331 I/O and Mem space regions for PCI autoconfiguration
+ */
+#define IOP331_PCI_IO_WINDOW_SIZE   0x00010000
+#define IOP331_PCI_LOWER_IO_PA      0x90000000
+#define IOP331_PCI_LOWER_IO_VA      0xfe000000
+#define IOP331_PCI_LOWER_IO_BA      (*IOP331_OIOWTVR)
+#define IOP331_PCI_UPPER_IO_PA      (IOP331_PCI_LOWER_IO_PA + IOP331_PCI_IO_WINDOW_SIZE - 1)
+#define IOP331_PCI_UPPER_IO_VA      (IOP331_PCI_LOWER_IO_VA + IOP331_PCI_IO_WINDOW_SIZE - 1)
+#define IOP331_PCI_UPPER_IO_BA      (IOP331_PCI_LOWER_IO_BA + IOP331_PCI_IO_WINDOW_SIZE - 1)
+#define IOP331_PCI_IO_OFFSET        (IOP331_PCI_LOWER_IO_VA - IOP331_PCI_LOWER_IO_BA)
+
+/* this can be 128M if OMWTVR1 is set */
+#define IOP331_PCI_MEM_WINDOW_SIZE	0x04000000 /* 64M outbound window */
+/* #define IOP331_PCI_MEM_WINDOW_SIZE  (~*IOP331_IALR1 + 1) */
+#define IOP331_PCI_LOWER_MEM_PA     0x80000000
+#define IOP331_PCI_LOWER_MEM_BA     (*IOP331_OMWTVR0)
+#define IOP331_PCI_UPPER_MEM_PA     (IOP331_PCI_LOWER_MEM_PA + IOP331_PCI_MEM_WINDOW_SIZE - 1)
+#define IOP331_PCI_UPPER_MEM_BA     (IOP331_PCI_LOWER_MEM_BA + IOP331_PCI_MEM_WINDOW_SIZE - 1)
+#define IOP331_PCI_MEM_OFFSET       (IOP331_PCI_LOWER_MEM_PA - IOP331_PCI_LOWER_MEM_BA)
+
+/*
+ * IOP331 chipset registers
+ */
+#define IOP331_VIRT_MEM_BASE  0xfeffe000  /* chip virtual mem address*/
+#define IOP331_PHYS_MEM_BASE  0xffffe000  /* chip physical memory address */
+#define IOP331_REG_ADDR(reg) (IOP331_VIRT_MEM_BASE | (reg))
+
+/* Reserved 0x00000000 through 0x000000FF */
+
+/* Address Translation Unit 0x00000100 through 0x000001FF */
+#define IOP331_ATUVID     (volatile u16 *)IOP331_REG_ADDR(0x00000100)
+#define IOP331_ATUDID     (volatile u16 *)IOP331_REG_ADDR(0x00000102)
+#define IOP331_ATUCMD     (volatile u16 *)IOP331_REG_ADDR(0x00000104)
+#define IOP331_ATUSR      (volatile u16 *)IOP331_REG_ADDR(0x00000106)
+#define IOP331_ATURID     (volatile u8  *)IOP331_REG_ADDR(0x00000108)
+#define IOP331_ATUCCR     (volatile u32 *)IOP331_REG_ADDR(0x00000109)
+#define IOP331_ATUCLSR    (volatile u8  *)IOP331_REG_ADDR(0x0000010C)
+#define IOP331_ATULT      (volatile u8  *)IOP331_REG_ADDR(0x0000010D)
+#define IOP331_ATUHTR     (volatile u8  *)IOP331_REG_ADDR(0x0000010E)
+#define IOP331_ATUBIST    (volatile u8  *)IOP331_REG_ADDR(0x0000010F)
+#define IOP331_IABAR0     (volatile u32 *)IOP331_REG_ADDR(0x00000110)
+#define IOP331_IAUBAR0    (volatile u32 *)IOP331_REG_ADDR(0x00000114)
+#define IOP331_IABAR1     (volatile u32 *)IOP331_REG_ADDR(0x00000118)
+#define IOP331_IAUBAR1    (volatile u32 *)IOP331_REG_ADDR(0x0000011C)
+#define IOP331_IABAR2     (volatile u32 *)IOP331_REG_ADDR(0x00000120)
+#define IOP331_IAUBAR2    (volatile u32 *)IOP331_REG_ADDR(0x00000124)
+#define IOP331_ASVIR      (volatile u16 *)IOP331_REG_ADDR(0x0000012C)
+#define IOP331_ASIR       (volatile u16 *)IOP331_REG_ADDR(0x0000012E)
+#define IOP331_ERBAR      (volatile u32 *)IOP331_REG_ADDR(0x00000130)
+#define IOP331_ATU_CAPPTR (volatile u32 *)IOP331_REG_ADDR(0x00000134)
+/* Reserved 0x00000138 through 0x0000013B */
+#define IOP331_ATUILR     (volatile u8  *)IOP331_REG_ADDR(0x0000013C)
+#define IOP331_ATUIPR     (volatile u8  *)IOP331_REG_ADDR(0x0000013D)
+#define IOP331_ATUMGNT    (volatile u8  *)IOP331_REG_ADDR(0x0000013E)
+#define IOP331_ATUMLAT    (volatile u8  *)IOP331_REG_ADDR(0x0000013F)
+#define IOP331_IALR0      (volatile u32 *)IOP331_REG_ADDR(0x00000140)
+#define IOP331_IATVR0     (volatile u32 *)IOP331_REG_ADDR(0x00000144)
+#define IOP331_ERLR       (volatile u32 *)IOP331_REG_ADDR(0x00000148)
+#define IOP331_ERTVR      (volatile u32 *)IOP331_REG_ADDR(0x0000014C)
+#define IOP331_IALR1      (volatile u32 *)IOP331_REG_ADDR(0x00000150)
+#define IOP331_IALR2      (volatile u32 *)IOP331_REG_ADDR(0x00000154)
+#define IOP331_IATVR2     (volatile u32 *)IOP331_REG_ADDR(0x00000158)
+#define IOP331_OIOWTVR    (volatile u32 *)IOP331_REG_ADDR(0x0000015C)
+#define IOP331_OMWTVR0    (volatile u32 *)IOP331_REG_ADDR(0x00000160)
+#define IOP331_OUMWTVR0   (volatile u32 *)IOP331_REG_ADDR(0x00000164)
+#define IOP331_OMWTVR1    (volatile u32 *)IOP331_REG_ADDR(0x00000168)
+#define IOP331_OUMWTVR1   (volatile u32 *)IOP331_REG_ADDR(0x0000016C)
+/* Reserved 0x00000170 through 0x00000177*/
+#define IOP331_OUDWTVR    (volatile u32 *)IOP331_REG_ADDR(0x00000178)
+/* Reserved 0x0000017C through 0x0000017F*/
+#define IOP331_ATUCR      (volatile u32 *)IOP331_REG_ADDR(0x00000180)
+#define IOP331_PCSR       (volatile u32 *)IOP331_REG_ADDR(0x00000184)
+#define IOP331_ATUISR     (volatile u32 *)IOP331_REG_ADDR(0x00000188)
+#define IOP331_ATUIMR     (volatile u32 *)IOP331_REG_ADDR(0x0000018C)
+#define IOP331_IABAR3     (volatile u32 *)IOP331_REG_ADDR(0x00000190)
+#define IOP331_IAUBAR3    (volatile u32 *)IOP331_REG_ADDR(0x00000194)
+#define IOP331_IALR3      (volatile u32 *)IOP331_REG_ADDR(0x00000198)
+#define IOP331_IATVR3     (volatile u32 *)IOP331_REG_ADDR(0x0000019C)
+/* Reserved 0x000001A0 through 0x000001A3*/
+#define IOP331_OCCAR      (volatile u32 *)IOP331_REG_ADDR(0x000001A4)
+/* Reserved 0x000001A8 through 0x000001AB*/
+#define IOP331_OCCDR      (volatile u32 *)IOP331_REG_ADDR(0x000001AC)
+/* Reserved 0x000001B0 through 0x000001BB*/
+#define IOP331_VPDCAPID   (volatile u8 *)IOP331_REG_ADDR(0x000001B8)
+#define IOP331_VPDNXTP    (volatile u8 *)IOP331_REG_ADDR(0x000001B9)
+#define IOP331_VPDAR	  (volatile u16 *)IOP331_REG_ADDR(0x000001BA)
+#define IOP331_VPDDR      (volatile u32 *)IOP331_REG_ADDR(0x000001BC)
+#define IOP331_PMCAPID    (volatile u8 *)IOP331_REG_ADDR(0x000001C0)
+#define IOP331_PMNEXT     (volatile u8 *)IOP331_REG_ADDR(0x000001C1)
+#define IOP331_APMCR      (volatile u16 *)IOP331_REG_ADDR(0x000001C2)
+#define IOP331_APMCSR     (volatile u16 *)IOP331_REG_ADDR(0x000001C4)
+/* Reserved 0x000001C6 through 0x000001CF */
+#define IOP331_MSICAPID   (volatile u8 *)IOP331_REG_ADDR(0x000001D0)
+#define IOP331_MSINXTP	  (volatile u8 *)IOP331_REG_ADDR(0x000001D1)
+#define IOP331_MSIMCR     (volatile u16 *)IOP331_REG_ADDR(0x000001D2)
+#define IOP331_MSIMAR     (volatile u32 *)IOP331_REG_ADDR(0x000001D4)
+#define IOP331_MSIMUAR	  (volatile u32 *)IOP331_REG_ADDR(0x000001D8)
+#define IOP331_MSIMDR	  (volatile u32 *)IOP331_REG_ADDR(0x000001DC)
+#define IOP331_PCIXCAPID  (volatile u8 *)IOP331_REG_ADDR(0x000001E0)
+#define IOP331_PCIXNEXT   (volatile u8 *)IOP331_REG_ADDR(0x000001E1)
+#define IOP331_PCIXCMD    (volatile u16 *)IOP331_REG_ADDR(0x000001E2)
+#define IOP331_PCIXSR     (volatile u32 *)IOP331_REG_ADDR(0x000001E4)
+#define IOP331_PCIIRSR    (volatile u32 *)IOP331_REG_ADDR(0x000001EC)
+
+/* Messaging Unit 0x00000300 through 0x000003FF */
+
+/* Reserved 0x00000300 through 0x0000030c */
+#define IOP331_IMR0       (volatile u32 *)IOP331_REG_ADDR(0x00000310)
+#define IOP331_IMR1       (volatile u32 *)IOP331_REG_ADDR(0x00000314)
+#define IOP331_OMR0       (volatile u32 *)IOP331_REG_ADDR(0x00000318)
+#define IOP331_OMR1       (volatile u32 *)IOP331_REG_ADDR(0x0000031C)
+#define IOP331_IDR        (volatile u32 *)IOP331_REG_ADDR(0x00000320)
+#define IOP331_IISR       (volatile u32 *)IOP331_REG_ADDR(0x00000324)
+#define IOP331_IIMR       (volatile u32 *)IOP331_REG_ADDR(0x00000328)
+#define IOP331_ODR        (volatile u32 *)IOP331_REG_ADDR(0x0000032C)
+#define IOP331_OISR       (volatile u32 *)IOP331_REG_ADDR(0x00000330)
+#define IOP331_OIMR       (volatile u32 *)IOP331_REG_ADDR(0x00000334)
+/* Reserved 0x00000338 through 0x0000034F */
+#define IOP331_MUCR       (volatile u32 *)IOP331_REG_ADDR(0x00000350)
+#define IOP331_QBAR       (volatile u32 *)IOP331_REG_ADDR(0x00000354)
+/* Reserved 0x00000358 through 0x0000035C */
+#define IOP331_IFHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000360)
+#define IOP331_IFTPR      (volatile u32 *)IOP331_REG_ADDR(0x00000364)
+#define IOP331_IPHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000368)
+#define IOP331_IPTPR      (volatile u32 *)IOP331_REG_ADDR(0x0000036C)
+#define IOP331_OFHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000370)
+#define IOP331_OFTPR      (volatile u32 *)IOP331_REG_ADDR(0x00000374)
+#define IOP331_OPHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000378)
+#define IOP331_OPTPR      (volatile u32 *)IOP331_REG_ADDR(0x0000037C)
+#define IOP331_IAR        (volatile u32 *)IOP331_REG_ADDR(0x00000380)
+/* Reserved 0x00000384 through 0x000003FF */
+
+/* DMA Controller 0x00000400 through 0x000004FF */
+#define IOP331_DMA0_CCR   (volatile u32 *)IOP331_REG_ADDR(0x00000400)
+#define IOP331_DMA0_CSR   (volatile u32 *)IOP331_REG_ADDR(0x00000404)
+#define IOP331_DMA0_DAR   (volatile u32 *)IOP331_REG_ADDR(0x0000040C)
+#define IOP331_DMA0_NDAR  (volatile u32 *)IOP331_REG_ADDR(0x00000410)
+#define IOP331_DMA0_PADR  (volatile u32 *)IOP331_REG_ADDR(0x00000414)
+#define IOP331_DMA0_PUADR (volatile u32 *)IOP331_REG_ADDR(0x00000418)
+#define IOP331_DMA0_LADR  (volatile u32 *)IOP331_REG_ADDR(0X0000041C)
+#define IOP331_DMA0_BCR   (volatile u32 *)IOP331_REG_ADDR(0x00000420)
+#define IOP331_DMA0_DCR   (volatile u32 *)IOP331_REG_ADDR(0x00000424)
+/* Reserved 0x00000428 through 0x0000043C */
+#define IOP331_DMA1_CCR   (volatile u32 *)IOP331_REG_ADDR(0x00000440)
+#define IOP331_DMA1_CSR   (volatile u32 *)IOP331_REG_ADDR(0x00000444)
+#define IOP331_DMA1_DAR   (volatile u32 *)IOP331_REG_ADDR(0x0000044C)
+#define IOP331_DMA1_NDAR  (volatile u32 *)IOP331_REG_ADDR(0x00000450)
+#define IOP331_DMA1_PADR  (volatile u32 *)IOP331_REG_ADDR(0x00000454)
+#define IOP331_DMA1_PUADR (volatile u32 *)IOP331_REG_ADDR(0x00000458)
+#define IOP331_DMA1_LADR  (volatile u32 *)IOP331_REG_ADDR(0x0000045C)
+#define IOP331_DMA1_BCR   (volatile u32 *)IOP331_REG_ADDR(0x00000460)
+#define IOP331_DMA1_DCR   (volatile u32 *)IOP331_REG_ADDR(0x00000464)
+/* Reserved 0x00000468 through 0x000004FF */
+
+/* Memory controller 0x00000500 through 0x0005FF */
+
+/* Peripheral bus interface unit 0x00000680 through 0x0006FF */
+#define IOP331_PBCR       (volatile u32 *)IOP331_REG_ADDR(0x00000680)
+#define IOP331_PBISR      (volatile u32 *)IOP331_REG_ADDR(0x00000684)
+#define IOP331_PBBAR0     (volatile u32 *)IOP331_REG_ADDR(0x00000688)
+#define IOP331_PBLR0      (volatile u32 *)IOP331_REG_ADDR(0x0000068C)
+#define IOP331_PBBAR1     (volatile u32 *)IOP331_REG_ADDR(0x00000690)
+#define IOP331_PBLR1      (volatile u32 *)IOP331_REG_ADDR(0x00000694)
+#define IOP331_PBBAR2     (volatile u32 *)IOP331_REG_ADDR(0x00000698)
+#define IOP331_PBLR2      (volatile u32 *)IOP331_REG_ADDR(0x0000069C)
+#define IOP331_PBBAR3     (volatile u32 *)IOP331_REG_ADDR(0x000006A0)
+#define IOP331_PBLR3      (volatile u32 *)IOP331_REG_ADDR(0x000006A4)
+#define IOP331_PBBAR4     (volatile u32 *)IOP331_REG_ADDR(0x000006A8)
+#define IOP331_PBLR4      (volatile u32 *)IOP331_REG_ADDR(0x000006AC)
+#define IOP331_PBBAR5     (volatile u32 *)IOP331_REG_ADDR(0x000006B0)
+#define IOP331_PBLR5      (volatile u32 *)IOP331_REG_ADDR(0x000006B4)
+#define IOP331_PBDSCR     (volatile u32 *)IOP331_REG_ADDR(0x000006B8)
+/* Reserved 0x000006BC */
+#define IOP331_PMBR0      (volatile u32 *)IOP331_REG_ADDR(0x000006C0)
+/* Reserved 0x000006C4 through 0x000006DC */
+#define IOP331_PMBR1      (volatile u32 *)IOP331_REG_ADDR(0x000006E0)
+#define IOP331_PMBR2      (volatile u32 *)IOP331_REG_ADDR(0x000006E4)
+
+#define IOP331_PBCR_EN    0x1
+
+#define IOP331_PBISR_BOOR_ERR 0x1
+
+
+
+/* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
+/* Internal arbitration unit 0x00000780 through 0x0007BF */
+
+/* Interrupt Controller */
+#define IOP331_INTCTL0    (volatile u32 *)IOP331_REG_ADDR(0x00000790)
+#define IOP331_INTCTL1    (volatile u32 *)IOP331_REG_ADDR(0x00000794)
+#define IOP331_INTSTR0    (volatile u32 *)IOP331_REG_ADDR(0x00000798)
+#define IOP331_INTSTR1    (volatile u32 *)IOP331_REG_ADDR(0x0000079C)
+#define IOP331_IINTSRC0   (volatile u32 *)IOP331_REG_ADDR(0x000007A0)
+#define IOP331_IINTSRC1   (volatile u32 *)IOP331_REG_ADDR(0x000007A4)
+#define IOP331_FINTSRC0   (volatile u32 *)IOP331_REG_ADDR(0x000007A8)
+#define IOP331_FINTSRC1   (volatile u32 *)IOP331_REG_ADDR(0x000007AC)
+#define IOP331_IPR0       (volatile u32 *)IOP331_REG_ADDR(0x000007B0)
+#define IOP331_IPR1       (volatile u32 *)IOP331_REG_ADDR(0x000007B4)
+#define IOP331_IPR2       (volatile u32 *)IOP331_REG_ADDR(0x000007B8)
+#define IOP331_IPR3       (volatile u32 *)IOP331_REG_ADDR(0x000007BC)
+#define IOP331_INTBASE    (volatile u32 *)IOP331_REG_ADDR(0x000007C0)
+#define IOP331_INTSIZE    (volatile u32 *)IOP331_REG_ADDR(0x000007C4)
+#define IOP331_IINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007C8)
+#define IOP331_FINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007CC)
+
+
+/* Timers */
+
+#define IOP331_TU_TMR0		(volatile u32 *)IOP331_REG_ADDR(0x000007D0)
+#define IOP331_TU_TMR1		(volatile u32 *)IOP331_REG_ADDR(0x000007D4)
+
+#define IOP331_TMR_TC		0x01
+#define	IOP331_TMR_EN		0x02
+#define IOP331_TMR_RELOAD	0x04
+#define	IOP331_TMR_PRIVILEGED	0x09
+
+#define	IOP331_TMR_RATIO_1_1	0x00
+#define	IOP331_TMR_RATIO_4_1	0x10
+#define	IOP331_TMR_RATIO_8_1	0x20
+#define	IOP331_TMR_RATIO_16_1	0x30
+
+#define IOP331_TU_TCR0    (volatile u32 *)IOP331_REG_ADDR(0x000007D8)
+#define IOP331_TU_TCR1    (volatile u32 *)IOP331_REG_ADDR(0x000007DC)
+#define IOP331_TU_TRR0    (volatile u32 *)IOP331_REG_ADDR(0x000007E0)
+#define IOP331_TU_TRR1    (volatile u32 *)IOP331_REG_ADDR(0x000007E4)
+#define IOP331_TU_TISR    (volatile u32 *)IOP331_REG_ADDR(0x000007E8)
+#define IOP331_TU_WDTCR   (volatile u32 *)IOP331_REG_ADDR(0x000007EC)
+
+#if defined(CONFIG_ARCH_IOP33X)
+#define	IOP331_TICK_RATE	266000000	/* 266 MHz IB clock */
+#endif
+
+#if defined(CONFIG_IOP331_STEPD) || defined(CONFIG_ARCH_IQ80333)
+#undef IOP331_TICK_RATE
+#define IOP331_TICK_RATE	333000000	/* 333 Mhz IB clock */
+#endif
+
+/* Application accelerator unit 0x00000800 - 0x000008FF */
+#define IOP331_AAU_ACR     (volatile u32 *)IOP331_REG_ADDR(0x00000800)
+#define IOP331_AAU_ASR     (volatile u32 *)IOP331_REG_ADDR(0x00000804)
+#define IOP331_AAU_ADAR    (volatile u32 *)IOP331_REG_ADDR(0x00000808)
+#define IOP331_AAU_ANDAR   (volatile u32 *)IOP331_REG_ADDR(0x0000080C)
+#define IOP331_AAU_SAR1    (volatile u32 *)IOP331_REG_ADDR(0x00000810)
+#define IOP331_AAU_SAR2    (volatile u32 *)IOP331_REG_ADDR(0x00000814)
+#define IOP331_AAU_SAR3    (volatile u32 *)IOP331_REG_ADDR(0x00000818)
+#define IOP331_AAU_SAR4    (volatile u32 *)IOP331_REG_ADDR(0x0000081C)
+#define IOP331_AAU_SAR5    (volatile u32 *)IOP331_REG_ADDR(0x0000082C)
+#define IOP331_AAU_SAR6    (volatile u32 *)IOP331_REG_ADDR(0x00000830)
+#define IOP331_AAU_SAR7    (volatile u32 *)IOP331_REG_ADDR(0x00000834)
+#define IOP331_AAU_SAR8    (volatile u32 *)IOP331_REG_ADDR(0x00000838)
+#define IOP331_AAU_SAR9    (volatile u32 *)IOP331_REG_ADDR(0x00000840)
+#define IOP331_AAU_SAR10   (volatile u32 *)IOP331_REG_ADDR(0x00000844)
+#define IOP331_AAU_SAR11   (volatile u32 *)IOP331_REG_ADDR(0x00000848)
+#define IOP331_AAU_SAR12   (volatile u32 *)IOP331_REG_ADDR(0x0000084C)
+#define IOP331_AAU_SAR13   (volatile u32 *)IOP331_REG_ADDR(0x00000850)
+#define IOP331_AAU_SAR14   (volatile u32 *)IOP331_REG_ADDR(0x00000854)
+#define IOP331_AAU_SAR15   (volatile u32 *)IOP331_REG_ADDR(0x00000858)
+#define IOP331_AAU_SAR16   (volatile u32 *)IOP331_REG_ADDR(0x0000085C)
+#define IOP331_AAU_SAR17   (volatile u32 *)IOP331_REG_ADDR(0x00000864)
+#define IOP331_AAU_SAR18   (volatile u32 *)IOP331_REG_ADDR(0x00000868)
+#define IOP331_AAU_SAR19   (volatile u32 *)IOP331_REG_ADDR(0x0000086C)
+#define IOP331_AAU_SAR20   (volatile u32 *)IOP331_REG_ADDR(0x00000870)
+#define IOP331_AAU_SAR21   (volatile u32 *)IOP331_REG_ADDR(0x00000874)
+#define IOP331_AAU_SAR22   (volatile u32 *)IOP331_REG_ADDR(0x00000878)
+#define IOP331_AAU_SAR23   (volatile u32 *)IOP331_REG_ADDR(0x0000087C)
+#define IOP331_AAU_SAR24   (volatile u32 *)IOP331_REG_ADDR(0x00000880)
+#define IOP331_AAU_SAR25   (volatile u32 *)IOP331_REG_ADDR(0x00000888)
+#define IOP331_AAU_SAR26   (volatile u32 *)IOP331_REG_ADDR(0x0000088C)
+#define IOP331_AAU_SAR27   (volatile u32 *)IOP331_REG_ADDR(0x00000890)
+#define IOP331_AAU_SAR28   (volatile u32 *)IOP331_REG_ADDR(0x00000894)
+#define IOP331_AAU_SAR29   (volatile u32 *)IOP331_REG_ADDR(0x00000898)
+#define IOP331_AAU_SAR30   (volatile u32 *)IOP331_REG_ADDR(0x0000089C)
+#define IOP331_AAU_SAR31   (volatile u32 *)IOP331_REG_ADDR(0x000008A0)
+#define IOP331_AAU_SAR32   (volatile u32 *)IOP331_REG_ADDR(0x000008A4)
+#define IOP331_AAU_DAR     (volatile u32 *)IOP331_REG_ADDR(0x00000820)
+#define IOP331_AAU_ABCR    (volatile u32 *)IOP331_REG_ADDR(0x00000824)
+#define IOP331_AAU_ADCR    (volatile u32 *)IOP331_REG_ADDR(0x00000828)
+#define IOP331_AAU_EDCR0   (volatile u32 *)IOP331_REG_ADDR(0x0000083c)
+#define IOP331_AAU_EDCR1   (volatile u32 *)IOP331_REG_ADDR(0x00000860)
+#define IOP331_AAU_EDCR2   (volatile u32 *)IOP331_REG_ADDR(0x00000884)
+
+
+#define IOP331_SPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C0)
+#define IOP331_PPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C8)
+/* SSP serial port unit 0x00001600 - 0x0000167F */
+
+/* I2C bus interface unit 0x00001680 - 0x000016FF */
+/* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
+
+#define IOP331_ICR0       (volatile u32 *)IOP331_REG_ADDR(0x00001680)
+#define IOP331_ISR0       (volatile u32 *)IOP331_REG_ADDR(0x00001684)
+#define IOP331_ISAR0      (volatile u32 *)IOP331_REG_ADDR(0x00001688)
+#define IOP331_IDBR0      (volatile u32 *)IOP331_REG_ADDR(0x0000168C)
+/* Reserved 0x00001690 */
+#define IOP331_IBMR0      (volatile u32 *)IOP331_REG_ADDR(0x00001694)
+/* Reserved 0x00001698 */
+/* Reserved 0x0000169C */
+#define IOP331_ICR1       (volatile u32 *)IOP331_REG_ADDR(0x000016A0)
+#define IOP331_ISR1       (volatile u32 *)IOP331_REG_ADDR(0x000016A4)
+#define IOP331_ISAR1      (volatile u32 *)IOP331_REG_ADDR(0x000016A8)
+#define IOP331_IDBR1      (volatile u32 *)IOP331_REG_ADDR(0x000016AC)
+#define IOP331_IBMR1      (volatile u32 *)IOP331_REG_ADDR(0x000016B4)
+/* Reserved 0x000016B8 through 0x000016FF */
+
+/* 0x00001700 through 0x0000172C  UART 0 */
+
+/* Reserved 0x00001730 through 0x0000173F */
+
+/* 0x00001740 through 0x0000176C UART 1 */
+
+#define IOP331_UART0_PHYS  (IOP331_PHYS_MEM_BASE | 0x00001700)	/* UART #1 physical */
+#define IOP331_UART1_PHYS  (IOP331_PHYS_MEM_BASE | 0x00001740)	/* UART #2 physical */
+#define IOP331_UART0_VIRT  (IOP331_VIRT_MEM_BASE | 0x00001700) /* UART #1 virtual addr */
+#define IOP331_UART1_VIRT  (IOP331_VIRT_MEM_BASE | 0x00001740) /* UART #2 virtual addr */
+
+/* Reserved 0x00001770 through 0x0000177F */
+
+/* General Purpose I/O Registers */
+#define IOP331_GPOE       (volatile u32 *)IOP331_REG_ADDR(0x00001780)
+#define IOP331_GPID       (volatile u32 *)IOP331_REG_ADDR(0x00001784)
+#define IOP331_GPOD       (volatile u32 *)IOP331_REG_ADDR(0x00001788)
+
+/* Reserved 0x0000178c through 0x000019ff */
+
+
+#ifndef __ASSEMBLY__
+extern void iop331_map_io(void);
+extern void iop331_init_irq(void);
+extern void iop331_time_init(void);
+#endif
+
+#endif // _IOP331_HW_H_
diff --git a/include/asm-arm/arch-iop33x/iq80331.h b/include/asm-arm/arch-iop33x/iq80331.h
new file mode 100644
index 000000000000..bda7ab6d55cf
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/iq80331.h
@@ -0,0 +1,23 @@
+/*
+ * linux/include/asm/arch-iop33x/iq80331.h
+ *
+ * Intel IQ80331 evaluation board registers
+ */
+
+#ifndef _IQ80331_H_
+#define _IQ80331_H_
+
+#define	IQ80331_FLASHBASE	0xc0000000	/* Flash */
+#define	IQ80331_FLASHSIZE	0x00800000
+#define	IQ80331_FLASHWIDTH	1
+
+#define IQ80331_7SEG_1		0xce840000	/* 7-Segment MSB */
+#define IQ80331_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
+#define IQ80331_ROTARY_SW	0xce8d0000	/* Rotary Switch */
+#define IQ80331_BATT_STAT	0xce8f0000	/* Battery Status */
+
+#ifndef __ASSEMBLY__
+extern void iq80331_map_io(void);
+#endif
+
+#endif	// _IQ80331_H_
diff --git a/include/asm-arm/arch-iop33x/iq80332.h b/include/asm-arm/arch-iop33x/iq80332.h
new file mode 100644
index 000000000000..f728e04378ab
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/iq80332.h
@@ -0,0 +1,23 @@
+/*
+ * linux/include/asm/arch-iop33x/iq80332.h
+ *
+ * Intel IQ80332 evaluation board registers
+ */
+
+#ifndef _IQ80332_H_
+#define _IQ80332_H_
+
+#define	IQ80332_FLASHBASE	0xc0000000	/* Flash */
+#define	IQ80332_FLASHSIZE	0x00800000
+#define	IQ80332_FLASHWIDTH	1
+
+#define IQ80332_7SEG_1		0xce840000	/* 7-Segment MSB */
+#define IQ80332_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
+#define IQ80332_ROTARY_SW	0xce8d0000	/* Rotary Switch */
+#define IQ80332_BATT_STAT	0xce8f0000	/* Battery Status */
+
+#ifndef __ASSEMBLY__
+extern void iq80332_map_io(void);
+#endif
+
+#endif	// _IQ80332_H_
diff --git a/include/asm-arm/arch-iop33x/irqs.h b/include/asm-arm/arch-iop33x/irqs.h
new file mode 100644
index 000000000000..45856a12815a
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/irqs.h
@@ -0,0 +1,130 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/irqs.h
+ *
+ * Author:	Dave Jiang (dave.jiang@intel.com)
+ * Copyright:	(C) 2003 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#ifndef _IRQS_H_
+#define _IRQS_H_
+
+/*
+ * IOP80331 chipset interrupts
+ */
+#define IOP331_IRQ_OFS		0
+#define IOP331_IRQ(x)		(IOP331_IRQ_OFS + (x))
+
+/*
+ * On IRQ or FIQ register
+ */
+#define IRQ_IOP331_DMA0_EOT	IOP331_IRQ(0)
+#define IRQ_IOP331_DMA0_EOC	IOP331_IRQ(1)
+#define IRQ_IOP331_DMA1_EOT	IOP331_IRQ(2)
+#define IRQ_IOP331_DMA1_EOC	IOP331_IRQ(3)
+#define IRQ_IOP331_RSVD_4	IOP331_IRQ(4)
+#define IRQ_IOP331_RSVD_5	IOP331_IRQ(5)
+#define IRQ_IOP331_AA_EOT	IOP331_IRQ(6)
+#define IRQ_IOP331_AA_EOC	IOP331_IRQ(7)
+#define IRQ_IOP331_TIMER0	IOP331_IRQ(8)
+#define IRQ_IOP331_TIMER1	IOP331_IRQ(9)
+#define IRQ_IOP331_I2C_0	IOP331_IRQ(10)
+#define IRQ_IOP331_I2C_1	IOP331_IRQ(11)
+#define IRQ_IOP331_MSG	    IOP331_IRQ(12)
+#define IRQ_IOP331_MSGIBQ   IOP331_IRQ(13)
+#define IRQ_IOP331_ATU_BIST	IOP331_IRQ(14)
+#define IRQ_IOP331_PERFMON	IOP331_IRQ(15)
+#define IRQ_IOP331_CORE_PMU	IOP331_IRQ(16)
+#define IRQ_IOP331_RSVD_17	IOP331_IRQ(17)
+#define IRQ_IOP331_RSVD_18	IOP331_IRQ(18)
+#define IRQ_IOP331_RSVD_19	IOP331_IRQ(19)
+#define IRQ_IOP331_RSVD_20	IOP331_IRQ(20)
+#define IRQ_IOP331_RSVD_21	IOP331_IRQ(21)
+#define IRQ_IOP331_RSVD_22	IOP331_IRQ(22)
+#define IRQ_IOP331_RSVD_23	IOP331_IRQ(23)
+#define IRQ_IOP331_XINT0	IOP331_IRQ(24)
+#define IRQ_IOP331_XINT1	IOP331_IRQ(25)
+#define IRQ_IOP331_XINT2	IOP331_IRQ(26)
+#define IRQ_IOP331_XINT3	IOP331_IRQ(27)
+#define IRQ_IOP331_RSVD_28  IOP331_IRQ(28)
+#define IRQ_IOP331_RSVD_29  IOP331_IRQ(29)
+#define IRQ_IOP331_RSVD_30  IOP331_IRQ(30)
+#define IRQ_IOP331_RSVD_31  IOP331_IRQ(31)
+#define IRQ_IOP331_XINT8    IOP331_IRQ(32)  // 0
+#define IRQ_IOP331_XINT9    IOP331_IRQ(33)  // 1
+#define IRQ_IOP331_XINT10   IOP331_IRQ(34)  // 2
+#define IRQ_IOP331_XINT11   IOP331_IRQ(35)  // 3
+#define IRQ_IOP331_XINT12   IOP331_IRQ(36)  // 4
+#define IRQ_IOP331_XINT13   IOP331_IRQ(37)  // 5
+#define IRQ_IOP331_XINT14   IOP331_IRQ(38)  // 6
+#define IRQ_IOP331_XINT15   IOP331_IRQ(39)  // 7
+#define IRQ_IOP331_RSVD_40  IOP331_IRQ(40)  // 8
+#define IRQ_IOP331_RSVD_41  IOP331_IRQ(41)  // 9
+#define IRQ_IOP331_RSVD_42  IOP331_IRQ(42)  // 10
+#define IRQ_IOP331_RSVD_43  IOP331_IRQ(43)  // 11
+#define IRQ_IOP331_RSVD_44  IOP331_IRQ(44)  // 12
+#define IRQ_IOP331_RSVD_45  IOP331_IRQ(45)  // 13
+#define IRQ_IOP331_RSVD_46  IOP331_IRQ(46)  // 14
+#define IRQ_IOP331_RSVD_47  IOP331_IRQ(47)  // 15
+#define IRQ_IOP331_RSVD_48  IOP331_IRQ(48)  // 16
+#define IRQ_IOP331_RSVD_49  IOP331_IRQ(49)  // 17
+#define IRQ_IOP331_RSVD_50  IOP331_IRQ(50)  // 18
+#define IRQ_IOP331_UART0    IOP331_IRQ(51)  // 19
+#define IRQ_IOP331_UART1    IOP331_IRQ(52)  // 20
+#define IRQ_IOP331_PBIE     IOP331_IRQ(53)  // 21
+#define IRQ_IOP331_ATU_CRW  IOP331_IRQ(54)  // 22
+#define IRQ_IOP331_ATU_ERR	IOP331_IRQ(55)  // 23
+#define IRQ_IOP331_MCU_ERR	IOP331_IRQ(56)  // 24
+#define IRQ_IOP331_DMA0_ERR	IOP331_IRQ(57)  // 25
+#define IRQ_IOP331_DMA1_ERR	IOP331_IRQ(58)  // 26
+#define IRQ_IOP331_RSVD_59  IOP331_IRQ(59)  // 27
+#define IRQ_IOP331_AA_ERR	IOP331_IRQ(60)  // 28
+#define IRQ_IOP331_RSVD_61  IOP331_IRQ(61)  // 29
+#define IRQ_IOP331_MSG_ERR	IOP331_IRQ(62)  // 30
+#define IRQ_IOP331_HPI		IOP331_IRQ(63)  // 31
+
+#define NR_IRQS			(IOP331_IRQ(63) + 1)
+
+
+/*
+ * Interrupts available on the IQ80331 board
+ */
+
+/*
+ * On board devices
+ */
+#define	IRQ_IQ80331_I82544	IRQ_IOP331_XINT0
+#define IRQ_IQ80331_UART0	IRQ_IOP331_UART0
+#define IRQ_IQ80331_UART1	IRQ_IOP331_UART1
+
+/*
+ * PCI interrupts
+ */
+#define	IRQ_IQ80331_INTA	IRQ_IOP331_XINT0
+#define	IRQ_IQ80331_INTB	IRQ_IOP331_XINT1
+#define	IRQ_IQ80331_INTC	IRQ_IOP331_XINT2
+#define	IRQ_IQ80331_INTD	IRQ_IOP331_XINT3
+
+/*
+ * Interrupts available on the IQ80332 board
+ */
+
+/*
+ * On board devices
+ */
+#define	IRQ_IQ80332_I82544	IRQ_IOP331_XINT0
+#define IRQ_IQ80332_UART0	IRQ_IOP331_UART0
+#define IRQ_IQ80332_UART1	IRQ_IOP331_UART1
+
+/*
+ * PCI interrupts
+ */
+#define	IRQ_IQ80332_INTA	IRQ_IOP331_XINT0
+#define	IRQ_IQ80332_INTB	IRQ_IOP331_XINT1
+#define	IRQ_IQ80332_INTC	IRQ_IOP331_XINT2
+#define	IRQ_IQ80332_INTD	IRQ_IOP331_XINT3
+
+#endif // _IRQ_H_
diff --git a/include/asm-arm/arch-iop33x/memory.h b/include/asm-arm/arch-iop33x/memory.h
new file mode 100644
index 000000000000..5e47164934ce
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/memory.h
@@ -0,0 +1,26 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/memory.h
+ */
+
+#ifndef __ASM_ARCH_MEMORY_H
+#define __ASM_ARCH_MEMORY_H
+
+#include <asm/hardware.h>
+
+/*
+ * Physical DRAM offset.
+ */
+#define PHYS_OFFSET	UL(0x00000000)
+
+/*
+ * Virtual view <-> PCI DMA view memory address translations
+ * virt_to_bus: Used to translate the virtual address to an
+ *		address suitable to be passed to set_dma_addr
+ * bus_to_virt: Used to convert an address for DMA operations
+ *		to an address that the kernel can use.
+ */
+#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP331_IATVR2)) | ((*IOP331_IABAR2) & 0xfffffff0))
+#define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP331_IALR2)) | ( *IOP331_IATVR2)))
+
+
+#endif
diff --git a/include/asm-arm/arch-iop33x/system.h b/include/asm-arm/arch-iop33x/system.h
new file mode 100644
index 000000000000..43cc787ea629
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/system.h
@@ -0,0 +1,29 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/system.h
+ *
+ *  Copyright (C) 2001 MontaVista Software, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+static inline void arch_idle(void)
+{
+	cpu_do_idle();
+}
+
+
+static inline void arch_reset(char mode)
+{
+    *IOP331_PCSR = 0x30;
+
+	if ( 1 && mode == 's') {
+		/* Jump into ROM at address 0 */
+		cpu_reset(0);
+	} else {
+		/* No on-chip reset capability */
+		cpu_reset(0);
+	}
+}
+
diff --git a/include/asm-arm/arch-iop33x/timex.h b/include/asm-arm/arch-iop33x/timex.h
new file mode 100644
index 000000000000..cc8085fa2a1e
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/timex.h
@@ -0,0 +1,8 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/timex.h
+ *
+ * IOP3xx architecture timex specifications
+ */
+#include <asm/hardware.h>
+
+#define CLOCK_TICK_RATE IOP331_TICK_RATE
diff --git a/include/asm-arm/arch-iop33x/uncompress.h b/include/asm-arm/arch-iop33x/uncompress.h
new file mode 100644
index 000000000000..62904ae3b038
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/uncompress.h
@@ -0,0 +1,36 @@
+/*
+ *  linux/include/asm-arm/arch-iop33x/uncompress.h
+ */
+#include <asm/types.h>
+#include <asm/mach-types.h>
+#include <linux/serial_reg.h>
+#include <asm/hardware.h>
+
+static volatile u32 *uart_base;
+
+#define TX_DONE (UART_LSR_TEMT|UART_LSR_THRE)
+
+static inline void putc(char c)
+{
+	while ((uart_base[UART_LSR] & TX_DONE) != TX_DONE)
+		barrier();
+	*uart_base = c;
+}
+
+static inline void flush(void)
+{
+}
+
+static __inline__ void __arch_decomp_setup(unsigned long arch_id)
+{
+	if (machine_is_iq80331() || machine_is_iq80332())
+		uart_base = (volatile u32 *)IOP331_UART0_PHYS;
+	else
+		uart_base = (volatile u32 *)0xfe800000;
+}
+
+/*
+ * nothing to do
+ */
+#define arch_decomp_setup()	__arch_decomp_setup(arch_id)
+#define arch_decomp_wdog()
diff --git a/include/asm-arm/arch-iop33x/vmalloc.h b/include/asm-arm/arch-iop33x/vmalloc.h
new file mode 100644
index 000000000000..b5092027449e
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/vmalloc.h
@@ -0,0 +1,16 @@
+/*
+ * linux/include/asm-arm/arch-iop33x/vmalloc.h
+ */
+
+/*
+ * Just any arbitrary offset to the start of the vmalloc VM area: the
+ * current 8MB value just means that there will be a 8MB "hole" after the
+ * physical memory until the kernel virtual memory starts.  That means that
+ * any out-of-bounds memory accesses will hopefully be caught.
+ * The vmalloc() routines leaves a hole of 4kB between each vmalloced
+ * area for the same reason. ;)
+ */
+//#define VMALLOC_END       (0xe8000000)
+/* increase usable physical RAM to ~992M per RMK */
+#define VMALLOC_END       (0xfe000000)
+
diff --git a/include/asm-arm/arch-iop3xx/debug-macro.S b/include/asm-arm/arch-iop3xx/debug-macro.S
deleted file mode 100644
index dcc6856d14ff..000000000000
--- a/include/asm-arm/arch-iop3xx/debug-macro.S
+++ /dev/null
@@ -1,35 +0,0 @@
-/* linux/include/asm-arm/arch-iop3xx/debug-macro.S
- *
- * Debugging macro include header
- *
- *  Copyright (C) 1994-1999 Russell King
- *  Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
-*/
-
-		.macro	addruart,rx
-		mov	\rx, #0xfe000000	@ physical
-#if defined(CONFIG_ARCH_IQ80321) || defined(CONFIG_ARCH_IQ31244)
-		orr	\rx, \rx, #0x00800000	@ location of the UART
-#elif defined(CONFIG_ARCH_IOP33X)
-		mrc	p15, 0, \rx, c1, c0
-		tst	\rx, #1			@ MMU enabled?
-		moveq	\rx, #0x000fe000	@ Physical Base
-		movne	\rx, #0
-		orr	\rx, \rx, #0xfe000000
-		orr	\rx, \rx, #0x00f00000	@ Virtual Base
-		orr	\rx, \rx, #0x00001700   @ location of the UART
-#else
-#error Unknown IOP3XX implementation
-#endif
-		.endm
-
-#if !defined(CONFIG_ARCH_IQ80321) || !defined(CONFIG_ARCH_IQ31244) || !defined(CONFIG_ARCH_IQ80331)
-#define FLOW_CONTROL
-#endif
-#define UART_SHIFT	0
-#include <asm/hardware/debug-8250.S>
diff --git a/include/asm-arm/arch-iop3xx/dma.h b/include/asm-arm/arch-iop3xx/dma.h
deleted file mode 100644
index 1e808db8af2a..000000000000
--- a/include/asm-arm/arch-iop3xx/dma.h
+++ /dev/null
@@ -1,9 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/dma.h
- *
- *  Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
diff --git a/include/asm-arm/arch-iop3xx/entry-macro.S b/include/asm-arm/arch-iop3xx/entry-macro.S
deleted file mode 100644
index f3db54637ad3..000000000000
--- a/include/asm-arm/arch-iop3xx/entry-macro.S
+++ /dev/null
@@ -1,57 +0,0 @@
-/*
- * include/asm-arm/arch-iop3xx/entry-macro.S
- *
- * Low-level IRQ helper macros for IOP3xx-based platforms
- *
- * This file is licensed under  the terms of the GNU General Public
- * License version 2. This program is licensed "as is" without any
- * warranty of any kind, whether express or implied.
- */
-#include <asm/arch/irqs.h>
-
-#if defined(CONFIG_ARCH_IOP32X)
-		.macro  disable_fiq
-		.endm
-
-		/*
-		 * Note: only deal with normal interrupts, not FIQ
-		 */
-		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
-		mov     \irqnr, #0
-		mrc     p6, 0, \irqstat, c8, c0, 0      @ Read IINTSRC
-		cmp     \irqstat, #0
-		beq     1001f
-		clz     \irqnr, \irqstat
-		mov     \base, #31
-		subs    \irqnr,\base,\irqnr
-		add     \irqnr,\irqnr,#IRQ_IOP321_DMA0_EOT
-1001:
-		.endm
-
-#elif defined(CONFIG_ARCH_IOP33X)
-		.macro  disable_fiq
-		.endm
-
-		/*
-		 * Note: only deal with normal interrupts, not FIQ
-		 */
-		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
-		mov     \irqnr, #0
-		mrc     p6, 0, \irqstat, c4, c0, 0      @ Read IINTSRC0
-		cmp     \irqstat, #0
-		bne     1002f
-		mrc     p6, 0, \irqstat, c5, c0, 0      @ Read IINTSRC1
-		cmp	\irqstat, #0
-		beq	1001f
-		clz	\irqnr, \irqstat
-		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
-		add	\irqnr,\irqnr,#IRQ_IOP331_XINT8
-		b	1001f
-1002:	clz     \irqnr, \irqstat
-		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
-		add     \irqnr,\irqnr,#IRQ_IOP331_DMA0_EOT
-1001:
-		.endm
-
-#endif
-
diff --git a/include/asm-arm/arch-iop3xx/hardware.h b/include/asm-arm/arch-iop3xx/hardware.h
deleted file mode 100644
index 3b138171d086..000000000000
--- a/include/asm-arm/arch-iop3xx/hardware.h
+++ /dev/null
@@ -1,57 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/hardware.h
- */
-#ifndef __ASM_ARCH_HARDWARE_H
-#define __ASM_ARCH_HARDWARE_H
-
-#include <asm/types.h>
-
-/*
- * Note about PCI IO space mappings
- *
- * To make IO space accesses efficient, we store virtual addresses in
- * the IO resources.
- *
- * The PCI IO space is located at virtual 0xfe000000 from physical
- * 0x90000000.  The PCI BARs must be programmed with physical addresses,
- * but when we read them, we convert them to virtual addresses.  See
- * arch/arm/mach-iop3xx/iop3xx-pci.c
- */
-
-#define pcibios_assign_all_busses() 1
-
-
-/*
- * The min PCI I/O and MEM space are dependent on what specific
- * chipset/platform we are running on, so instead of hardcoding with
- * #ifdefs, we just fill these in the platform level PCI init code.
- */
-#ifndef __ASSEMBLY__
-extern unsigned long iop3xx_pcibios_min_io;
-extern unsigned long iop3xx_pcibios_min_mem;
-
-extern unsigned int processor_id;
-#endif
-
-/*
- * We just set these to zero since they are really bogus anyways
- */
-#define PCIBIOS_MIN_IO      (iop3xx_pcibios_min_io)
-#define PCIBIOS_MIN_MEM     (iop3xx_pcibios_min_mem)
-
-/*
- * Generic chipset bits
- *
- */
-#include "iop321.h"
-#include "iop331.h"
-
-/*
- * Board specific bits
- */
-#include "iq80321.h"
-#include "iq31244.h"
-#include "iq80331.h"
-#include "iq80332.h"
-
-#endif  /* _ASM_ARCH_HARDWARE_H */
diff --git a/include/asm-arm/arch-iop3xx/io.h b/include/asm-arm/arch-iop3xx/io.h
deleted file mode 100644
index 36adbdf5055a..000000000000
--- a/include/asm-arm/arch-iop3xx/io.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/io.h
- *
- *  Copyright (C) 2001  MontaVista Software, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef __ASM_ARM_ARCH_IO_H
-#define __ASM_ARM_ARCH_IO_H
-
-#include <asm/hardware.h>
-
-#define IO_SPACE_LIMIT 0xffffffff
-
-#define __io(p)			((void __iomem *)(p))
-#define __mem_pci(a)		(a)
-
-#endif
diff --git a/include/asm-arm/arch-iop3xx/iop321-irqs.h b/include/asm-arm/arch-iop3xx/iop321-irqs.h
deleted file mode 100644
index 2fcc1654cb9d..000000000000
--- a/include/asm-arm/arch-iop3xx/iop321-irqs.h
+++ /dev/null
@@ -1,100 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/irqs.h
- *
- * Author:	Rory Bolt <rorybolt@pacbell.net>
- * Copyright:	(C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#ifndef _IOP321_IRQS_H_
-#define _IOP321_IRQS_H_
-
-/*
- * IOP80321 chipset interrupts
- */
-#define IOP321_IRQ_OFS		0
-#define IOP321_IRQ(x)		(IOP321_IRQ_OFS + (x))
-
-/*
- * On IRQ or FIQ register
- */
-#define IRQ_IOP321_DMA0_EOT	IOP321_IRQ(0)
-#define IRQ_IOP321_DMA0_EOC	IOP321_IRQ(1)
-#define IRQ_IOP321_DMA1_EOT	IOP321_IRQ(2)
-#define IRQ_IOP321_DMA1_EOC	IOP321_IRQ(3)
-#define IRQ_IOP321_RSVD_4	IOP321_IRQ(4)
-#define IRQ_IOP321_RSVD_5	IOP321_IRQ(5)
-#define IRQ_IOP321_AA_EOT	IOP321_IRQ(6)
-#define IRQ_IOP321_AA_EOC	IOP321_IRQ(7)
-#define IRQ_IOP321_CORE_PMON	IOP321_IRQ(8)
-#define IRQ_IOP321_TIMER0	IOP321_IRQ(9)
-#define IRQ_IOP321_TIMER1	IOP321_IRQ(10)
-#define IRQ_IOP321_I2C_0	IOP321_IRQ(11)
-#define IRQ_IOP321_I2C_1	IOP321_IRQ(12)
-#define IRQ_IOP321_MESSAGING	IOP321_IRQ(13)
-#define IRQ_IOP321_ATU_BIST	IOP321_IRQ(14)
-#define IRQ_IOP321_PERFMON	IOP321_IRQ(15)
-#define IRQ_IOP321_CORE_PMU	IOP321_IRQ(16)
-#define IRQ_IOP321_BIU_ERR	IOP321_IRQ(17)
-#define IRQ_IOP321_ATU_ERR	IOP321_IRQ(18)
-#define IRQ_IOP321_MCU_ERR	IOP321_IRQ(19)
-#define IRQ_IOP321_DMA0_ERR	IOP321_IRQ(20)
-#define IRQ_IOP321_DMA1_ERR	IOP321_IRQ(21)
-#define IRQ_IOP321_RSVD_22	IOP321_IRQ(22)
-#define IRQ_IOP321_AA_ERR	IOP321_IRQ(23)
-#define IRQ_IOP321_MSG_ERR	IOP321_IRQ(24)
-#define IRQ_IOP321_SSP		IOP321_IRQ(25)
-#define IRQ_IOP321_RSVD_26	IOP321_IRQ(26)
-#define IRQ_IOP321_XINT0	IOP321_IRQ(27)
-#define IRQ_IOP321_XINT1	IOP321_IRQ(28)
-#define IRQ_IOP321_XINT2	IOP321_IRQ(29)
-#define IRQ_IOP321_XINT3	IOP321_IRQ(30)
-#define IRQ_IOP321_HPI		IOP321_IRQ(31)
-
-#define NR_IOP321_IRQS		(IOP321_IRQ(31) + 1)
-
-#define NR_IRQS			NR_IOP321_IRQS
-
-
-/*
- * Interrupts available on the IQ80321 board
- */
-
-/*
- * On board devices
- */
-#define	IRQ_IQ80321_I82544	IRQ_IOP321_XINT0
-#define IRQ_IQ80321_UART	IRQ_IOP321_XINT1
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ80321_INTA	IRQ_IOP321_XINT0
-#define	IRQ_IQ80321_INTB	IRQ_IOP321_XINT1
-#define	IRQ_IQ80321_INTC	IRQ_IOP321_XINT2
-#define	IRQ_IQ80321_INTD	IRQ_IOP321_XINT3
-
-/*
- * Interrupts on the IQ31244 board
- */
-
-/*
- * On board devices
- */
-#define IRQ_IQ31244_UART	IRQ_IOP321_XINT1
-#define	IRQ_IQ31244_I82546	IRQ_IOP321_XINT0
-#define IRQ_IQ31244_SATA	IRQ_IOP321_XINT2
-#define	IRQ_IQ31244_PCIX_SLOT	IRQ_IOP321_XINT3
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ31244_INTA	IRQ_IOP321_XINT0
-#define	IRQ_IQ31244_INTB	IRQ_IOP321_XINT1
-#define	IRQ_IQ31244_INTC	IRQ_IOP321_XINT2
-#define	IRQ_IQ31244_INTD	IRQ_IOP321_XINT3
-
-#endif // _IOP321_IRQ_H_
diff --git a/include/asm-arm/arch-iop3xx/iop321.h b/include/asm-arm/arch-iop3xx/iop321.h
deleted file mode 100644
index d198d72a50a4..000000000000
--- a/include/asm-arm/arch-iop3xx/iop321.h
+++ /dev/null
@@ -1,345 +0,0 @@
-/*
- * linux/include/asm/arch-iop3xx/iop321.h
- *
- * Intel IOP321 Chip definitions
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef _IOP321_HW_H_
-#define _IOP321_HW_H_
-
-
-/*
- * This is needed for mixed drivers that need to work on all
- * IOP3xx variants but behave slightly differently on each.
- */
-#ifndef __ASSEMBLY__
-#ifdef	CONFIG_ARCH_IOP32X
-#define iop_is_321() (((processor_id & 0xfffff5e0) == 0x69052420))
-#else
-#define	iop_is_321()	0
-#endif
-#endif
-
-/*
- * IOP321 I/O and Mem space regions for PCI autoconfiguration
- */
-#define IOP321_PCI_IO_WINDOW_SIZE   0x00010000
-#define IOP321_PCI_LOWER_IO_PA      0x90000000
-#define IOP321_PCI_LOWER_IO_VA      0xfe000000
-#define IOP321_PCI_LOWER_IO_BA      (*IOP321_OIOWTVR)
-#define IOP321_PCI_UPPER_IO_PA      (IOP321_PCI_LOWER_IO_PA + IOP321_PCI_IO_WINDOW_SIZE - 1)
-#define IOP321_PCI_UPPER_IO_VA      (IOP321_PCI_LOWER_IO_VA + IOP321_PCI_IO_WINDOW_SIZE - 1)
-#define IOP321_PCI_UPPER_IO_BA      (IOP321_PCI_LOWER_IO_BA + IOP321_PCI_IO_WINDOW_SIZE - 1)
-#define IOP321_PCI_IO_OFFSET        (IOP321_PCI_LOWER_IO_VA - IOP321_PCI_LOWER_IO_BA)
-
-/* #define IOP321_PCI_MEM_WINDOW_SIZE  (~*IOP321_IALR1 + 1) */
-#define IOP321_PCI_MEM_WINDOW_SIZE  0x04000000 /* 64M outbound window */
-#define IOP321_PCI_LOWER_MEM_PA     0x80000000
-#define IOP321_PCI_LOWER_MEM_BA     (*IOP321_OMWTVR0)
-#define IOP321_PCI_UPPER_MEM_PA     (IOP321_PCI_LOWER_MEM_PA + IOP321_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP321_PCI_UPPER_MEM_BA     (IOP321_PCI_LOWER_MEM_BA + IOP321_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP321_PCI_MEM_OFFSET       (IOP321_PCI_LOWER_MEM_PA - IOP321_PCI_LOWER_MEM_BA)
-
-
-/*
- * IOP321 chipset registers
- */
-#define IOP321_VIRT_MEM_BASE 0xfeffe000  /* chip virtual mem address*/
-#define IOP321_PHYS_MEM_BASE 0xffffe000  /* chip physical memory address */
-#define IOP321_REG_ADDR(reg) (IOP321_VIRT_MEM_BASE | (reg))
-
-/* Reserved 0x00000000 through 0x000000FF */
-
-/* Address Translation Unit 0x00000100 through 0x000001FF */
-#define IOP321_ATUVID     (volatile u16 *)IOP321_REG_ADDR(0x00000100)
-#define IOP321_ATUDID     (volatile u16 *)IOP321_REG_ADDR(0x00000102)
-#define IOP321_ATUCMD     (volatile u16 *)IOP321_REG_ADDR(0x00000104)
-#define IOP321_ATUSR      (volatile u16 *)IOP321_REG_ADDR(0x00000106)
-#define IOP321_ATURID     (volatile u8  *)IOP321_REG_ADDR(0x00000108)
-#define IOP321_ATUCCR     (volatile u32 *)IOP321_REG_ADDR(0x00000109)
-#define IOP321_ATUCLSR    (volatile u8  *)IOP321_REG_ADDR(0x0000010C)
-#define IOP321_ATULT      (volatile u8  *)IOP321_REG_ADDR(0x0000010D)
-#define IOP321_ATUHTR     (volatile u8  *)IOP321_REG_ADDR(0x0000010E)
-#define IOP321_ATUBIST    (volatile u8  *)IOP321_REG_ADDR(0x0000010F)
-#define IOP321_IABAR0     (volatile u32 *)IOP321_REG_ADDR(0x00000110)
-#define IOP321_IAUBAR0    (volatile u32 *)IOP321_REG_ADDR(0x00000114)
-#define IOP321_IABAR1     (volatile u32 *)IOP321_REG_ADDR(0x00000118)
-#define IOP321_IAUBAR1    (volatile u32 *)IOP321_REG_ADDR(0x0000011C)
-#define IOP321_IABAR2     (volatile u32 *)IOP321_REG_ADDR(0x00000120)
-#define IOP321_IAUBAR2    (volatile u32 *)IOP321_REG_ADDR(0x00000124)
-#define IOP321_ASVIR      (volatile u16 *)IOP321_REG_ADDR(0x0000012C)
-#define IOP321_ASIR       (volatile u16 *)IOP321_REG_ADDR(0x0000012E)
-#define IOP321_ERBAR      (volatile u32 *)IOP321_REG_ADDR(0x00000130)
-/* Reserved 0x00000134 through 0x0000013B */
-#define IOP321_ATUILR     (volatile u8  *)IOP321_REG_ADDR(0x0000013C)
-#define IOP321_ATUIPR     (volatile u8  *)IOP321_REG_ADDR(0x0000013D)
-#define IOP321_ATUMGNT    (volatile u8  *)IOP321_REG_ADDR(0x0000013E)
-#define IOP321_ATUMLAT    (volatile u8  *)IOP321_REG_ADDR(0x0000013F)
-#define IOP321_IALR0      (volatile u32 *)IOP321_REG_ADDR(0x00000140)
-#define IOP321_IATVR0     (volatile u32 *)IOP321_REG_ADDR(0x00000144)
-#define IOP321_ERLR       (volatile u32 *)IOP321_REG_ADDR(0x00000148)
-#define IOP321_ERTVR      (volatile u32 *)IOP321_REG_ADDR(0x0000014C)
-#define IOP321_IALR1      (volatile u32 *)IOP321_REG_ADDR(0x00000150)
-#define IOP321_IALR2      (volatile u32 *)IOP321_REG_ADDR(0x00000154)
-#define IOP321_IATVR2     (volatile u32 *)IOP321_REG_ADDR(0x00000158)
-#define IOP321_OIOWTVR    (volatile u32 *)IOP321_REG_ADDR(0x0000015C)
-#define IOP321_OMWTVR0    (volatile u32 *)IOP321_REG_ADDR(0x00000160)
-#define IOP321_OUMWTVR0   (volatile u32 *)IOP321_REG_ADDR(0x00000164)
-#define IOP321_OMWTVR1    (volatile u32 *)IOP321_REG_ADDR(0x00000168)
-#define IOP321_OUMWTVR1   (volatile u32 *)IOP321_REG_ADDR(0x0000016C)
-/* Reserved 0x00000170 through 0x00000177*/
-#define IOP321_OUDWTVR    (volatile u32 *)IOP321_REG_ADDR(0x00000178)
-/* Reserved 0x0000017C through 0x0000017F*/
-#define IOP321_ATUCR      (volatile u32 *)IOP321_REG_ADDR(0x00000180)
-#define IOP321_PCSR       (volatile u32 *)IOP321_REG_ADDR(0x00000184)
-#define IOP321_ATUISR     (volatile u32 *)IOP321_REG_ADDR(0x00000188)
-#define IOP321_ATUIMR     (volatile u32 *)IOP321_REG_ADDR(0x0000018C)
-#define IOP321_IABAR3     (volatile u32 *)IOP321_REG_ADDR(0x00000190)
-#define IOP321_IAUBAR3    (volatile u32 *)IOP321_REG_ADDR(0x00000194)
-#define IOP321_IALR3      (volatile u32 *)IOP321_REG_ADDR(0x00000198)
-#define IOP321_IATVR3     (volatile u32 *)IOP321_REG_ADDR(0x0000019C)
-/* Reserved 0x000001A0 through 0x000001A3*/
-#define IOP321_OCCAR      (volatile u32 *)IOP321_REG_ADDR(0x000001A4)
-/* Reserved 0x000001A8 through 0x000001AB*/
-#define IOP321_OCCDR      (volatile u32 *)IOP321_REG_ADDR(0x000001AC)
-/* Reserved 0x000001B0 through 0x000001BB*/
-#define IOP321_PDSCR      (volatile u32 *)IOP321_REG_ADDR(0x000001BC)
-#define IOP321_PMCAPID    (volatile u8  *)IOP321_REG_ADDR(0x000001C0)
-#define IOP321_PMNEXT     (volatile u8  *)IOP321_REG_ADDR(0x000001C1)
-#define IOP321_APMCR      (volatile u16 *)IOP321_REG_ADDR(0x000001C2)
-#define IOP321_APMCSR     (volatile u16 *)IOP321_REG_ADDR(0x000001C4)
-/* Reserved 0x000001C6 through 0x000001DF */
-#define IOP321_PCIXCAPID  (volatile u8  *)IOP321_REG_ADDR(0x000001E0)
-#define IOP321_PCIXNEXT   (volatile u8  *)IOP321_REG_ADDR(0x000001E1)
-#define IOP321_PCIXCMD    (volatile u16 *)IOP321_REG_ADDR(0x000001E2)
-#define IOP321_PCIXSR     (volatile u32 *)IOP321_REG_ADDR(0x000001E4)
-#define IOP321_PCIIRSR    (volatile u32 *)IOP321_REG_ADDR(0x000001EC)
-
-/* Messaging Unit 0x00000300 through 0x000003FF */
-
-/* Reserved 0x00000300 through 0x0000030c */
-#define IOP321_IMR0       (volatile u32 *)IOP321_REG_ADDR(0x00000310)
-#define IOP321_IMR1       (volatile u32 *)IOP321_REG_ADDR(0x00000314)
-#define IOP321_OMR0       (volatile u32 *)IOP321_REG_ADDR(0x00000318)
-#define IOP321_OMR1       (volatile u32 *)IOP321_REG_ADDR(0x0000031C)
-#define IOP321_IDR        (volatile u32 *)IOP321_REG_ADDR(0x00000320)
-#define IOP321_IISR       (volatile u32 *)IOP321_REG_ADDR(0x00000324)
-#define IOP321_IIMR       (volatile u32 *)IOP321_REG_ADDR(0x00000328)
-#define IOP321_ODR        (volatile u32 *)IOP321_REG_ADDR(0x0000032C)
-#define IOP321_OISR       (volatile u32 *)IOP321_REG_ADDR(0x00000330)
-#define IOP321_OIMR       (volatile u32 *)IOP321_REG_ADDR(0x00000334)
-/* Reserved 0x00000338 through 0x0000034F */
-#define IOP321_MUCR       (volatile u32 *)IOP321_REG_ADDR(0x00000350)
-#define IOP321_QBAR       (volatile u32 *)IOP321_REG_ADDR(0x00000354)
-/* Reserved 0x00000358 through 0x0000035C */
-#define IOP321_IFHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000360)
-#define IOP321_IFTPR      (volatile u32 *)IOP321_REG_ADDR(0x00000364)
-#define IOP321_IPHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000368)
-#define IOP321_IPTPR      (volatile u32 *)IOP321_REG_ADDR(0x0000036C)
-#define IOP321_OFHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000370)
-#define IOP321_OFTPR      (volatile u32 *)IOP321_REG_ADDR(0x00000374)
-#define IOP321_OPHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000378)
-#define IOP321_OPTPR      (volatile u32 *)IOP321_REG_ADDR(0x0000037C)
-#define IOP321_IAR        (volatile u32 *)IOP321_REG_ADDR(0x00000380)
-
-#define IOP321_IIxR_MASK	0x7f /* masks all */
-#define IOP321_IIxR_IRI		0x40 /* RC Index Register Interrupt */
-#define IOP321_IIxR_OFQF	0x20 /* RC Output Free Q Full (ERROR) */
-#define IOP321_IIxR_ipq		0x10 /* RC Inbound Post Q (post) */
-#define IOP321_IIxR_ERRDI	0x08 /* RO Error Doorbell Interrupt */
-#define IOP321_IIxR_IDI		0x04 /* RO Inbound Doorbell Interrupt */
-#define IOP321_IIxR_IM1		0x02 /* RC Inbound Message 1 Interrupt */
-#define IOP321_IIxR_IM0		0x01 /* RC Inbound Message 0 Interrupt */
-
-/* Reserved 0x00000384 through 0x000003FF */
-
-/* DMA Controller 0x00000400 through 0x000004FF */
-#define IOP321_DMA0_CCR   (volatile u32 *)IOP321_REG_ADDR(0x00000400)
-#define IOP321_DMA0_CSR   (volatile u32 *)IOP321_REG_ADDR(0x00000404)
-#define IOP321_DMA0_DAR   (volatile u32 *)IOP321_REG_ADDR(0x0000040C)
-#define IOP321_DMA0_NDAR  (volatile u32 *)IOP321_REG_ADDR(0x00000410)
-#define IOP321_DMA0_PADR  (volatile u32 *)IOP321_REG_ADDR(0x00000414)
-#define IOP321_DMA0_PUADR (volatile u32 *)IOP321_REG_ADDR(0x00000418)
-#define IOP321_DMA0_LADR  (volatile u32 *)IOP321_REG_ADDR(0X0000041C)
-#define IOP321_DMA0_BCR   (volatile u32 *)IOP321_REG_ADDR(0x00000420)
-#define IOP321_DMA0_DCR   (volatile u32 *)IOP321_REG_ADDR(0x00000424)
-/* Reserved 0x00000428 through 0x0000043C */
-#define IOP321_DMA1_CCR   (volatile u32 *)IOP321_REG_ADDR(0x00000440)
-#define IOP321_DMA1_CSR   (volatile u32 *)IOP321_REG_ADDR(0x00000444)
-#define IOP321_DMA1_DAR   (volatile u32 *)IOP321_REG_ADDR(0x0000044C)
-#define IOP321_DMA1_NDAR  (volatile u32 *)IOP321_REG_ADDR(0x00000450)
-#define IOP321_DMA1_PADR  (volatile u32 *)IOP321_REG_ADDR(0x00000454)
-#define IOP321_DMA1_PUADR (volatile u32 *)IOP321_REG_ADDR(0x00000458)
-#define IOP321_DMA1_LADR  (volatile u32 *)IOP321_REG_ADDR(0x0000045C)
-#define IOP321_DMA1_BCR   (volatile u32 *)IOP321_REG_ADDR(0x00000460)
-#define IOP321_DMA1_DCR   (volatile u32 *)IOP321_REG_ADDR(0x00000464)
-/* Reserved 0x00000468 through 0x000004FF */
-
-/* Memory controller 0x00000500 through 0x0005FF */
-
-/* Peripheral bus interface unit 0x00000680 through 0x0006FF */
-#define IOP321_PBCR       (volatile u32 *)IOP321_REG_ADDR(0x00000680)
-#define IOP321_PBISR      (volatile u32 *)IOP321_REG_ADDR(0x00000684)
-#define IOP321_PBBAR0     (volatile u32 *)IOP321_REG_ADDR(0x00000688)
-#define IOP321_PBLR0      (volatile u32 *)IOP321_REG_ADDR(0x0000068C)
-#define IOP321_PBBAR1     (volatile u32 *)IOP321_REG_ADDR(0x00000690)
-#define IOP321_PBLR1      (volatile u32 *)IOP321_REG_ADDR(0x00000694)
-#define IOP321_PBBAR2     (volatile u32 *)IOP321_REG_ADDR(0x00000698)
-#define IOP321_PBLR2      (volatile u32 *)IOP321_REG_ADDR(0x0000069C)
-#define IOP321_PBBAR3     (volatile u32 *)IOP321_REG_ADDR(0x000006A0)
-#define IOP321_PBLR3      (volatile u32 *)IOP321_REG_ADDR(0x000006A4)
-#define IOP321_PBBAR4     (volatile u32 *)IOP321_REG_ADDR(0x000006A8)
-#define IOP321_PBLR4      (volatile u32 *)IOP321_REG_ADDR(0x000006AC)
-#define IOP321_PBBAR5     (volatile u32 *)IOP321_REG_ADDR(0x000006B0)
-#define IOP321_PBLR5      (volatile u32 *)IOP321_REG_ADDR(0x000006B4)
-#define IOP321_PBDSCR     (volatile u32 *)IOP321_REG_ADDR(0x000006B8)
-/* Reserved 0x000006BC */
-#define IOP321_PMBR0      (volatile u32 *)IOP321_REG_ADDR(0x000006C0)
-/* Reserved 0x000006C4 through 0x000006DC */
-#define IOP321_PMBR1      (volatile u32 *)IOP321_REG_ADDR(0x000006E0)
-#define IOP321_PMBR2      (volatile u32 *)IOP321_REG_ADDR(0x000006E4)
-
-#define IOP321_PBCR_EN    0x1
-
-#define IOP321_PBISR_BOOR_ERR 0x1
-
-/* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
-#define IOP321_GTMR	(volatile u32 *)IOP321_REG_ADDR(0x00000700)
-#define IOP321_ESR	(volatile u32 *)IOP321_REG_ADDR(0x00000704)
-#define IOP321_EMISR	(volatile u32 *)IOP321_REG_ADDR(0x00000708)
-/* reserved 0x00000070c */
-#define IOP321_GTSR	(volatile u32 *)IOP321_REG_ADDR(0x00000710)
-/* PERC0 DOESN'T EXIST - index from 1! */
-#define IOP321_PERCR0	(volatile u32 *)IOP321_REG_ADDR(0x00000710)
-
-#define IOP321_GTMR_NGCE	0x04 /* (Not) Global Counter Enable */
-
-/* Internal arbitration unit 0x00000780 through 0x0007BF */
-#define IOP321_IACR	(volatile u32 *)IOP321_REG_ADDR(0x00000780)
-#define IOP321_MTTR1	(volatile u32 *)IOP321_REG_ADDR(0x00000784)
-#define IOP321_MTTR2	(volatile u32 *)IOP321_REG_ADDR(0x00000788)
-
-/* General Purpose I/O Registers */
-#define IOP321_GPOE       (volatile u32 *)IOP321_REG_ADDR(0x000007C4)
-#define IOP321_GPID       (volatile u32 *)IOP321_REG_ADDR(0x000007C8)
-#define IOP321_GPOD       (volatile u32 *)IOP321_REG_ADDR(0x000007CC)
-
-/* Interrupt Controller */
-#define IOP321_INTCTL     (volatile u32 *)IOP321_REG_ADDR(0x000007D0)
-#define IOP321_INTSTR     (volatile u32 *)IOP321_REG_ADDR(0x000007D4)
-#define IOP321_IINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007D8)
-#define IOP321_FINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007DC)
-
-/* Timers */
-
-#define IOP321_TU_TMR0		(volatile u32 *)IOP321_REG_ADDR(0x000007E0)
-#define IOP321_TU_TMR1		(volatile u32 *)IOP321_REG_ADDR(0x000007E4)
-
-#ifdef CONFIG_ARCH_IQ80321
-#define	IOP321_TICK_RATE	200000000	/* 200 MHz clock */
-#elif defined(CONFIG_ARCH_IQ31244)
-#define IOP321_TICK_RATE	198000000	/* 33.000 MHz crystal */
-#endif
-
-#ifdef CONFIG_ARCH_EP80219
-#undef IOP321_TICK_RATE
-#define IOP321_TICK_RATE 200000000 /* 33.333333 Mhz crystal */
-#endif
-
-#define IOP321_TMR_TC		0x01
-#define	IOP321_TMR_EN		0x02
-#define IOP321_TMR_RELOAD	0x04
-#define	IOP321_TMR_PRIVILEGED	0x09
-
-#define	IOP321_TMR_RATIO_1_1	0x00
-#define	IOP321_TMR_RATIO_4_1	0x10
-#define	IOP321_TMR_RATIO_8_1	0x20
-#define	IOP321_TMR_RATIO_16_1	0x30
-
-#define IOP321_TU_TCR0    (volatile u32 *)IOP321_REG_ADDR(0x000007E8)
-#define IOP321_TU_TCR1    (volatile u32 *)IOP321_REG_ADDR(0x000007EC)
-#define IOP321_TU_TRR0    (volatile u32 *)IOP321_REG_ADDR(0x000007F0)
-#define IOP321_TU_TRR1    (volatile u32 *)IOP321_REG_ADDR(0x000007F4)
-#define IOP321_TU_TISR    (volatile u32 *)IOP321_REG_ADDR(0x000007F8)
-#define IOP321_TU_WDTCR   (volatile u32 *)IOP321_REG_ADDR(0x000007FC)
-
-/* Application accelerator unit 0x00000800 - 0x000008FF */
-#define IOP321_AAU_ACR     (volatile u32 *)IOP321_REG_ADDR(0x00000800)
-#define IOP321_AAU_ASR     (volatile u32 *)IOP321_REG_ADDR(0x00000804)
-#define IOP321_AAU_ADAR    (volatile u32 *)IOP321_REG_ADDR(0x00000808)
-#define IOP321_AAU_ANDAR   (volatile u32 *)IOP321_REG_ADDR(0x0000080C)
-#define IOP321_AAU_SAR1    (volatile u32 *)IOP321_REG_ADDR(0x00000810)
-#define IOP321_AAU_SAR2    (volatile u32 *)IOP321_REG_ADDR(0x00000814)
-#define IOP321_AAU_SAR3    (volatile u32 *)IOP321_REG_ADDR(0x00000818)
-#define IOP321_AAU_SAR4    (volatile u32 *)IOP321_REG_ADDR(0x0000081C)
-#define IOP321_AAU_SAR5    (volatile u32 *)IOP321_REG_ADDR(0x0000082C)
-#define IOP321_AAU_SAR6    (volatile u32 *)IOP321_REG_ADDR(0x00000830)
-#define IOP321_AAU_SAR7    (volatile u32 *)IOP321_REG_ADDR(0x00000834)
-#define IOP321_AAU_SAR8    (volatile u32 *)IOP321_REG_ADDR(0x00000838)
-#define IOP321_AAU_SAR9    (volatile u32 *)IOP321_REG_ADDR(0x00000840)
-#define IOP321_AAU_SAR10   (volatile u32 *)IOP321_REG_ADDR(0x00000844)
-#define IOP321_AAU_SAR11   (volatile u32 *)IOP321_REG_ADDR(0x00000848)
-#define IOP321_AAU_SAR12   (volatile u32 *)IOP321_REG_ADDR(0x0000084C)
-#define IOP321_AAU_SAR13   (volatile u32 *)IOP321_REG_ADDR(0x00000850)
-#define IOP321_AAU_SAR14   (volatile u32 *)IOP321_REG_ADDR(0x00000854)
-#define IOP321_AAU_SAR15   (volatile u32 *)IOP321_REG_ADDR(0x00000858)
-#define IOP321_AAU_SAR16   (volatile u32 *)IOP321_REG_ADDR(0x0000085C)
-#define IOP321_AAU_SAR17   (volatile u32 *)IOP321_REG_ADDR(0x00000864)
-#define IOP321_AAU_SAR18   (volatile u32 *)IOP321_REG_ADDR(0x00000868)
-#define IOP321_AAU_SAR19   (volatile u32 *)IOP321_REG_ADDR(0x0000086C)
-#define IOP321_AAU_SAR20   (volatile u32 *)IOP321_REG_ADDR(0x00000870)
-#define IOP321_AAU_SAR21   (volatile u32 *)IOP321_REG_ADDR(0x00000874)
-#define IOP321_AAU_SAR22   (volatile u32 *)IOP321_REG_ADDR(0x00000878)
-#define IOP321_AAU_SAR23   (volatile u32 *)IOP321_REG_ADDR(0x0000087C)
-#define IOP321_AAU_SAR24   (volatile u32 *)IOP321_REG_ADDR(0x00000880)
-#define IOP321_AAU_SAR25   (volatile u32 *)IOP321_REG_ADDR(0x00000888)
-#define IOP321_AAU_SAR26   (volatile u32 *)IOP321_REG_ADDR(0x0000088C)
-#define IOP321_AAU_SAR27   (volatile u32 *)IOP321_REG_ADDR(0x00000890)
-#define IOP321_AAU_SAR28   (volatile u32 *)IOP321_REG_ADDR(0x00000894)
-#define IOP321_AAU_SAR29   (volatile u32 *)IOP321_REG_ADDR(0x00000898)
-#define IOP321_AAU_SAR30   (volatile u32 *)IOP321_REG_ADDR(0x0000089C)
-#define IOP321_AAU_SAR31   (volatile u32 *)IOP321_REG_ADDR(0x000008A0)
-#define IOP321_AAU_SAR32   (volatile u32 *)IOP321_REG_ADDR(0x000008A4)
-#define IOP321_AAU_DAR     (volatile u32 *)IOP321_REG_ADDR(0x00000820)
-#define IOP321_AAU_ABCR    (volatile u32 *)IOP321_REG_ADDR(0x00000824)
-#define IOP321_AAU_ADCR    (volatile u32 *)IOP321_REG_ADDR(0x00000828)
-#define IOP321_AAU_EDCR0   (volatile u32 *)IOP321_REG_ADDR(0x0000083c)
-#define IOP321_AAU_EDCR1   (volatile u32 *)IOP321_REG_ADDR(0x00000860)
-#define IOP321_AAU_EDCR2   (volatile u32 *)IOP321_REG_ADDR(0x00000884)
-
-
-/* SSP serial port unit 0x00001600 - 0x0000167F */
-/* I2C bus interface unit 0x00001680 - 0x000016FF */
-#define IOP321_ICR0       (volatile u32 *)IOP321_REG_ADDR(0x00001680)
-#define IOP321_ISR0       (volatile u32 *)IOP321_REG_ADDR(0x00001684)
-#define IOP321_ISAR0      (volatile u32 *)IOP321_REG_ADDR(0x00001688)
-#define IOP321_IDBR0      (volatile u32 *)IOP321_REG_ADDR(0x0000168C)
-/* Reserved 0x00001690 */
-#define IOP321_IBMR0      (volatile u32 *)IOP321_REG_ADDR(0x00001694)
-/* Reserved 0x00001698 */
-/* Reserved 0x0000169C */
-#define IOP321_ICR1       (volatile u32 *)IOP321_REG_ADDR(0x000016A0)
-#define IOP321_ISR1       (volatile u32 *)IOP321_REG_ADDR(0x000016A4)
-#define IOP321_ISAR1      (volatile u32 *)IOP321_REG_ADDR(0x000016A8)
-#define IOP321_IDBR1      (volatile u32 *)IOP321_REG_ADDR(0x000016AC)
-#define IOP321_IBMR1      (volatile u32 *)IOP321_REG_ADDR(0x000016B4)
-/* Reserved 0x000016B8 through 0x000016FC */
-
-/* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
-
-
-#ifndef __ASSEMBLY__
-extern void iop321_map_io(void);
-extern void iop321_init_irq(void);
-extern void iop321_time_init(void);
-#endif
-
-#endif // _IOP321_HW_H_
diff --git a/include/asm-arm/arch-iop3xx/iop331-irqs.h b/include/asm-arm/arch-iop3xx/iop331-irqs.h
deleted file mode 100644
index 7135ad7e335e..000000000000
--- a/include/asm-arm/arch-iop3xx/iop331-irqs.h
+++ /dev/null
@@ -1,132 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/irqs.h
- *
- * Author:	Dave Jiang (dave.jiang@intel.com)
- * Copyright:	(C) 2003 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#ifndef _IOP331_IRQS_H_
-#define _IOP331_IRQS_H_
-
-/*
- * IOP80331 chipset interrupts
- */
-#define IOP331_IRQ_OFS		0
-#define IOP331_IRQ(x)		(IOP331_IRQ_OFS + (x))
-
-/*
- * On IRQ or FIQ register
- */
-#define IRQ_IOP331_DMA0_EOT	IOP331_IRQ(0)
-#define IRQ_IOP331_DMA0_EOC	IOP331_IRQ(1)
-#define IRQ_IOP331_DMA1_EOT	IOP331_IRQ(2)
-#define IRQ_IOP331_DMA1_EOC	IOP331_IRQ(3)
-#define IRQ_IOP331_RSVD_4	IOP331_IRQ(4)
-#define IRQ_IOP331_RSVD_5	IOP331_IRQ(5)
-#define IRQ_IOP331_AA_EOT	IOP331_IRQ(6)
-#define IRQ_IOP331_AA_EOC	IOP331_IRQ(7)
-#define IRQ_IOP331_TIMER0	IOP331_IRQ(8)
-#define IRQ_IOP331_TIMER1	IOP331_IRQ(9)
-#define IRQ_IOP331_I2C_0	IOP331_IRQ(10)
-#define IRQ_IOP331_I2C_1	IOP331_IRQ(11)
-#define IRQ_IOP331_MSG	    IOP331_IRQ(12)
-#define IRQ_IOP331_MSGIBQ   IOP331_IRQ(13)
-#define IRQ_IOP331_ATU_BIST	IOP331_IRQ(14)
-#define IRQ_IOP331_PERFMON	IOP331_IRQ(15)
-#define IRQ_IOP331_CORE_PMU	IOP331_IRQ(16)
-#define IRQ_IOP331_RSVD_17	IOP331_IRQ(17)
-#define IRQ_IOP331_RSVD_18	IOP331_IRQ(18)
-#define IRQ_IOP331_RSVD_19	IOP331_IRQ(19)
-#define IRQ_IOP331_RSVD_20	IOP331_IRQ(20)
-#define IRQ_IOP331_RSVD_21	IOP331_IRQ(21)
-#define IRQ_IOP331_RSVD_22	IOP331_IRQ(22)
-#define IRQ_IOP331_RSVD_23	IOP331_IRQ(23)
-#define IRQ_IOP331_XINT0	IOP331_IRQ(24)
-#define IRQ_IOP331_XINT1	IOP331_IRQ(25)
-#define IRQ_IOP331_XINT2	IOP331_IRQ(26)
-#define IRQ_IOP331_XINT3	IOP331_IRQ(27)
-#define IRQ_IOP331_RSVD_28  IOP331_IRQ(28)
-#define IRQ_IOP331_RSVD_29  IOP331_IRQ(29)
-#define IRQ_IOP331_RSVD_30  IOP331_IRQ(30)
-#define IRQ_IOP331_RSVD_31  IOP331_IRQ(31)
-#define IRQ_IOP331_XINT8    IOP331_IRQ(32)  // 0
-#define IRQ_IOP331_XINT9    IOP331_IRQ(33)  // 1
-#define IRQ_IOP331_XINT10   IOP331_IRQ(34)  // 2
-#define IRQ_IOP331_XINT11   IOP331_IRQ(35)  // 3
-#define IRQ_IOP331_XINT12   IOP331_IRQ(36)  // 4
-#define IRQ_IOP331_XINT13   IOP331_IRQ(37)  // 5
-#define IRQ_IOP331_XINT14   IOP331_IRQ(38)  // 6
-#define IRQ_IOP331_XINT15   IOP331_IRQ(39)  // 7
-#define IRQ_IOP331_RSVD_40  IOP331_IRQ(40)  // 8
-#define IRQ_IOP331_RSVD_41  IOP331_IRQ(41)  // 9
-#define IRQ_IOP331_RSVD_42  IOP331_IRQ(42)  // 10
-#define IRQ_IOP331_RSVD_43  IOP331_IRQ(43)  // 11
-#define IRQ_IOP331_RSVD_44  IOP331_IRQ(44)  // 12
-#define IRQ_IOP331_RSVD_45  IOP331_IRQ(45)  // 13
-#define IRQ_IOP331_RSVD_46  IOP331_IRQ(46)  // 14
-#define IRQ_IOP331_RSVD_47  IOP331_IRQ(47)  // 15
-#define IRQ_IOP331_RSVD_48  IOP331_IRQ(48)  // 16
-#define IRQ_IOP331_RSVD_49  IOP331_IRQ(49)  // 17
-#define IRQ_IOP331_RSVD_50  IOP331_IRQ(50)  // 18
-#define IRQ_IOP331_UART0    IOP331_IRQ(51)  // 19
-#define IRQ_IOP331_UART1    IOP331_IRQ(52)  // 20
-#define IRQ_IOP331_PBIE     IOP331_IRQ(53)  // 21
-#define IRQ_IOP331_ATU_CRW  IOP331_IRQ(54)  // 22
-#define IRQ_IOP331_ATU_ERR	IOP331_IRQ(55)  // 23
-#define IRQ_IOP331_MCU_ERR	IOP331_IRQ(56)  // 24
-#define IRQ_IOP331_DMA0_ERR	IOP331_IRQ(57)  // 25
-#define IRQ_IOP331_DMA1_ERR	IOP331_IRQ(58)  // 26
-#define IRQ_IOP331_RSVD_59  IOP331_IRQ(59)  // 27
-#define IRQ_IOP331_AA_ERR	IOP331_IRQ(60)  // 28
-#define IRQ_IOP331_RSVD_61  IOP331_IRQ(61)  // 29
-#define IRQ_IOP331_MSG_ERR	IOP331_IRQ(62)  // 30
-#define IRQ_IOP331_HPI		IOP331_IRQ(63)  // 31
-
-#define NR_IOP331_IRQS		(IOP331_IRQ(63) + 1)
-
-#define NR_IRQS			NR_IOP331_IRQS
-
-
-/*
- * Interrupts available on the IQ80331 board
- */
-
-/*
- * On board devices
- */
-#define	IRQ_IQ80331_I82544	IRQ_IOP331_XINT0
-#define IRQ_IQ80331_UART0	IRQ_IOP331_UART0
-#define IRQ_IQ80331_UART1	IRQ_IOP331_UART1
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ80331_INTA	IRQ_IOP331_XINT0
-#define	IRQ_IQ80331_INTB	IRQ_IOP331_XINT1
-#define	IRQ_IQ80331_INTC	IRQ_IOP331_XINT2
-#define	IRQ_IQ80331_INTD	IRQ_IOP331_XINT3
-
-/*
- * Interrupts available on the IQ80332 board
- */
-
-/*
- * On board devices
- */
-#define	IRQ_IQ80332_I82544	IRQ_IOP331_XINT0
-#define IRQ_IQ80332_UART0	IRQ_IOP331_UART0
-#define IRQ_IQ80332_UART1	IRQ_IOP331_UART1
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ80332_INTA	IRQ_IOP331_XINT0
-#define	IRQ_IQ80332_INTB	IRQ_IOP331_XINT1
-#define	IRQ_IQ80332_INTC	IRQ_IOP331_XINT2
-#define	IRQ_IQ80332_INTD	IRQ_IOP331_XINT3
-
-#endif // _IOP331_IRQ_H_
diff --git a/include/asm-arm/arch-iop3xx/iop331.h b/include/asm-arm/arch-iop3xx/iop331.h
deleted file mode 100644
index 4d7bcc62cb3e..000000000000
--- a/include/asm-arm/arch-iop3xx/iop331.h
+++ /dev/null
@@ -1,363 +0,0 @@
-/*
- * linux/include/asm/arch-iop3xx/iop331.h
- *
- * Intel IOP331 Chip definitions
- *
- * Author: Dave Jiang (dave.jiang@intel.com)
- * Copyright (C) 2003, 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef _IOP331_HW_H_
-#define _IOP331_HW_H_
-
-
-/*
- * This is needed for mixed drivers that need to work on all
- * IOP3xx variants but behave slightly differently on each.
- */
-#ifndef __ASSEMBLY__
-#ifdef	CONFIG_ARCH_IOP33X
-/*#define	iop_is_331()	((processor_id & 0xffffffb0) == 0x69054090) */
-#define	iop_is_331()	((processor_id & 0xffffff30) == 0x69054010)
-#else
-#define	iop_is_331()	0
-#endif
-#endif
-
-/*
- * IOP331 I/O and Mem space regions for PCI autoconfiguration
- */
-#define IOP331_PCI_IO_WINDOW_SIZE   0x00010000
-#define IOP331_PCI_LOWER_IO_PA      0x90000000
-#define IOP331_PCI_LOWER_IO_VA      0xfe000000
-#define IOP331_PCI_LOWER_IO_BA      (*IOP331_OIOWTVR)
-#define IOP331_PCI_UPPER_IO_PA      (IOP331_PCI_LOWER_IO_PA + IOP331_PCI_IO_WINDOW_SIZE - 1)
-#define IOP331_PCI_UPPER_IO_VA      (IOP331_PCI_LOWER_IO_VA + IOP331_PCI_IO_WINDOW_SIZE - 1)
-#define IOP331_PCI_UPPER_IO_BA      (IOP331_PCI_LOWER_IO_BA + IOP331_PCI_IO_WINDOW_SIZE - 1)
-#define IOP331_PCI_IO_OFFSET        (IOP331_PCI_LOWER_IO_VA - IOP331_PCI_LOWER_IO_BA)
-
-/* this can be 128M if OMWTVR1 is set */
-#define IOP331_PCI_MEM_WINDOW_SIZE	0x04000000 /* 64M outbound window */
-/* #define IOP331_PCI_MEM_WINDOW_SIZE  (~*IOP331_IALR1 + 1) */
-#define IOP331_PCI_LOWER_MEM_PA     0x80000000
-#define IOP331_PCI_LOWER_MEM_BA     (*IOP331_OMWTVR0)
-#define IOP331_PCI_UPPER_MEM_PA     (IOP331_PCI_LOWER_MEM_PA + IOP331_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP331_PCI_UPPER_MEM_BA     (IOP331_PCI_LOWER_MEM_BA + IOP331_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP331_PCI_MEM_OFFSET       (IOP331_PCI_LOWER_MEM_PA - IOP331_PCI_LOWER_MEM_BA)
-
-/*
- * IOP331 chipset registers
- */
-#define IOP331_VIRT_MEM_BASE  0xfeffe000  /* chip virtual mem address*/
-#define IOP331_PHYS_MEM_BASE  0xffffe000  /* chip physical memory address */
-#define IOP331_REG_ADDR(reg) (IOP331_VIRT_MEM_BASE | (reg))
-
-/* Reserved 0x00000000 through 0x000000FF */
-
-/* Address Translation Unit 0x00000100 through 0x000001FF */
-#define IOP331_ATUVID     (volatile u16 *)IOP331_REG_ADDR(0x00000100)
-#define IOP331_ATUDID     (volatile u16 *)IOP331_REG_ADDR(0x00000102)
-#define IOP331_ATUCMD     (volatile u16 *)IOP331_REG_ADDR(0x00000104)
-#define IOP331_ATUSR      (volatile u16 *)IOP331_REG_ADDR(0x00000106)
-#define IOP331_ATURID     (volatile u8  *)IOP331_REG_ADDR(0x00000108)
-#define IOP331_ATUCCR     (volatile u32 *)IOP331_REG_ADDR(0x00000109)
-#define IOP331_ATUCLSR    (volatile u8  *)IOP331_REG_ADDR(0x0000010C)
-#define IOP331_ATULT      (volatile u8  *)IOP331_REG_ADDR(0x0000010D)
-#define IOP331_ATUHTR     (volatile u8  *)IOP331_REG_ADDR(0x0000010E)
-#define IOP331_ATUBIST    (volatile u8  *)IOP331_REG_ADDR(0x0000010F)
-#define IOP331_IABAR0     (volatile u32 *)IOP331_REG_ADDR(0x00000110)
-#define IOP331_IAUBAR0    (volatile u32 *)IOP331_REG_ADDR(0x00000114)
-#define IOP331_IABAR1     (volatile u32 *)IOP331_REG_ADDR(0x00000118)
-#define IOP331_IAUBAR1    (volatile u32 *)IOP331_REG_ADDR(0x0000011C)
-#define IOP331_IABAR2     (volatile u32 *)IOP331_REG_ADDR(0x00000120)
-#define IOP331_IAUBAR2    (volatile u32 *)IOP331_REG_ADDR(0x00000124)
-#define IOP331_ASVIR      (volatile u16 *)IOP331_REG_ADDR(0x0000012C)
-#define IOP331_ASIR       (volatile u16 *)IOP331_REG_ADDR(0x0000012E)
-#define IOP331_ERBAR      (volatile u32 *)IOP331_REG_ADDR(0x00000130)
-#define IOP331_ATU_CAPPTR (volatile u32 *)IOP331_REG_ADDR(0x00000134)
-/* Reserved 0x00000138 through 0x0000013B */
-#define IOP331_ATUILR     (volatile u8  *)IOP331_REG_ADDR(0x0000013C)
-#define IOP331_ATUIPR     (volatile u8  *)IOP331_REG_ADDR(0x0000013D)
-#define IOP331_ATUMGNT    (volatile u8  *)IOP331_REG_ADDR(0x0000013E)
-#define IOP331_ATUMLAT    (volatile u8  *)IOP331_REG_ADDR(0x0000013F)
-#define IOP331_IALR0      (volatile u32 *)IOP331_REG_ADDR(0x00000140)
-#define IOP331_IATVR0     (volatile u32 *)IOP331_REG_ADDR(0x00000144)
-#define IOP331_ERLR       (volatile u32 *)IOP331_REG_ADDR(0x00000148)
-#define IOP331_ERTVR      (volatile u32 *)IOP331_REG_ADDR(0x0000014C)
-#define IOP331_IALR1      (volatile u32 *)IOP331_REG_ADDR(0x00000150)
-#define IOP331_IALR2      (volatile u32 *)IOP331_REG_ADDR(0x00000154)
-#define IOP331_IATVR2     (volatile u32 *)IOP331_REG_ADDR(0x00000158)
-#define IOP331_OIOWTVR    (volatile u32 *)IOP331_REG_ADDR(0x0000015C)
-#define IOP331_OMWTVR0    (volatile u32 *)IOP331_REG_ADDR(0x00000160)
-#define IOP331_OUMWTVR0   (volatile u32 *)IOP331_REG_ADDR(0x00000164)
-#define IOP331_OMWTVR1    (volatile u32 *)IOP331_REG_ADDR(0x00000168)
-#define IOP331_OUMWTVR1   (volatile u32 *)IOP331_REG_ADDR(0x0000016C)
-/* Reserved 0x00000170 through 0x00000177*/
-#define IOP331_OUDWTVR    (volatile u32 *)IOP331_REG_ADDR(0x00000178)
-/* Reserved 0x0000017C through 0x0000017F*/
-#define IOP331_ATUCR      (volatile u32 *)IOP331_REG_ADDR(0x00000180)
-#define IOP331_PCSR       (volatile u32 *)IOP331_REG_ADDR(0x00000184)
-#define IOP331_ATUISR     (volatile u32 *)IOP331_REG_ADDR(0x00000188)
-#define IOP331_ATUIMR     (volatile u32 *)IOP331_REG_ADDR(0x0000018C)
-#define IOP331_IABAR3     (volatile u32 *)IOP331_REG_ADDR(0x00000190)
-#define IOP331_IAUBAR3    (volatile u32 *)IOP331_REG_ADDR(0x00000194)
-#define IOP331_IALR3      (volatile u32 *)IOP331_REG_ADDR(0x00000198)
-#define IOP331_IATVR3     (volatile u32 *)IOP331_REG_ADDR(0x0000019C)
-/* Reserved 0x000001A0 through 0x000001A3*/
-#define IOP331_OCCAR      (volatile u32 *)IOP331_REG_ADDR(0x000001A4)
-/* Reserved 0x000001A8 through 0x000001AB*/
-#define IOP331_OCCDR      (volatile u32 *)IOP331_REG_ADDR(0x000001AC)
-/* Reserved 0x000001B0 through 0x000001BB*/
-#define IOP331_VPDCAPID   (volatile u8 *)IOP331_REG_ADDR(0x000001B8)
-#define IOP331_VPDNXTP    (volatile u8 *)IOP331_REG_ADDR(0x000001B9)
-#define IOP331_VPDAR	  (volatile u16 *)IOP331_REG_ADDR(0x000001BA)
-#define IOP331_VPDDR      (volatile u32 *)IOP331_REG_ADDR(0x000001BC)
-#define IOP331_PMCAPID    (volatile u8 *)IOP331_REG_ADDR(0x000001C0)
-#define IOP331_PMNEXT     (volatile u8 *)IOP331_REG_ADDR(0x000001C1)
-#define IOP331_APMCR      (volatile u16 *)IOP331_REG_ADDR(0x000001C2)
-#define IOP331_APMCSR     (volatile u16 *)IOP331_REG_ADDR(0x000001C4)
-/* Reserved 0x000001C6 through 0x000001CF */
-#define IOP331_MSICAPID   (volatile u8 *)IOP331_REG_ADDR(0x000001D0)
-#define IOP331_MSINXTP	  (volatile u8 *)IOP331_REG_ADDR(0x000001D1)
-#define IOP331_MSIMCR     (volatile u16 *)IOP331_REG_ADDR(0x000001D2)
-#define IOP331_MSIMAR     (volatile u32 *)IOP331_REG_ADDR(0x000001D4)
-#define IOP331_MSIMUAR	  (volatile u32 *)IOP331_REG_ADDR(0x000001D8)
-#define IOP331_MSIMDR	  (volatile u32 *)IOP331_REG_ADDR(0x000001DC)
-#define IOP331_PCIXCAPID  (volatile u8 *)IOP331_REG_ADDR(0x000001E0)
-#define IOP331_PCIXNEXT   (volatile u8 *)IOP331_REG_ADDR(0x000001E1)
-#define IOP331_PCIXCMD    (volatile u16 *)IOP331_REG_ADDR(0x000001E2)
-#define IOP331_PCIXSR     (volatile u32 *)IOP331_REG_ADDR(0x000001E4)
-#define IOP331_PCIIRSR    (volatile u32 *)IOP331_REG_ADDR(0x000001EC)
-
-/* Messaging Unit 0x00000300 through 0x000003FF */
-
-/* Reserved 0x00000300 through 0x0000030c */
-#define IOP331_IMR0       (volatile u32 *)IOP331_REG_ADDR(0x00000310)
-#define IOP331_IMR1       (volatile u32 *)IOP331_REG_ADDR(0x00000314)
-#define IOP331_OMR0       (volatile u32 *)IOP331_REG_ADDR(0x00000318)
-#define IOP331_OMR1       (volatile u32 *)IOP331_REG_ADDR(0x0000031C)
-#define IOP331_IDR        (volatile u32 *)IOP331_REG_ADDR(0x00000320)
-#define IOP331_IISR       (volatile u32 *)IOP331_REG_ADDR(0x00000324)
-#define IOP331_IIMR       (volatile u32 *)IOP331_REG_ADDR(0x00000328)
-#define IOP331_ODR        (volatile u32 *)IOP331_REG_ADDR(0x0000032C)
-#define IOP331_OISR       (volatile u32 *)IOP331_REG_ADDR(0x00000330)
-#define IOP331_OIMR       (volatile u32 *)IOP331_REG_ADDR(0x00000334)
-/* Reserved 0x00000338 through 0x0000034F */
-#define IOP331_MUCR       (volatile u32 *)IOP331_REG_ADDR(0x00000350)
-#define IOP331_QBAR       (volatile u32 *)IOP331_REG_ADDR(0x00000354)
-/* Reserved 0x00000358 through 0x0000035C */
-#define IOP331_IFHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000360)
-#define IOP331_IFTPR      (volatile u32 *)IOP331_REG_ADDR(0x00000364)
-#define IOP331_IPHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000368)
-#define IOP331_IPTPR      (volatile u32 *)IOP331_REG_ADDR(0x0000036C)
-#define IOP331_OFHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000370)
-#define IOP331_OFTPR      (volatile u32 *)IOP331_REG_ADDR(0x00000374)
-#define IOP331_OPHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000378)
-#define IOP331_OPTPR      (volatile u32 *)IOP331_REG_ADDR(0x0000037C)
-#define IOP331_IAR        (volatile u32 *)IOP331_REG_ADDR(0x00000380)
-/* Reserved 0x00000384 through 0x000003FF */
-
-/* DMA Controller 0x00000400 through 0x000004FF */
-#define IOP331_DMA0_CCR   (volatile u32 *)IOP331_REG_ADDR(0x00000400)
-#define IOP331_DMA0_CSR   (volatile u32 *)IOP331_REG_ADDR(0x00000404)
-#define IOP331_DMA0_DAR   (volatile u32 *)IOP331_REG_ADDR(0x0000040C)
-#define IOP331_DMA0_NDAR  (volatile u32 *)IOP331_REG_ADDR(0x00000410)
-#define IOP331_DMA0_PADR  (volatile u32 *)IOP331_REG_ADDR(0x00000414)
-#define IOP331_DMA0_PUADR (volatile u32 *)IOP331_REG_ADDR(0x00000418)
-#define IOP331_DMA0_LADR  (volatile u32 *)IOP331_REG_ADDR(0X0000041C)
-#define IOP331_DMA0_BCR   (volatile u32 *)IOP331_REG_ADDR(0x00000420)
-#define IOP331_DMA0_DCR   (volatile u32 *)IOP331_REG_ADDR(0x00000424)
-/* Reserved 0x00000428 through 0x0000043C */
-#define IOP331_DMA1_CCR   (volatile u32 *)IOP331_REG_ADDR(0x00000440)
-#define IOP331_DMA1_CSR   (volatile u32 *)IOP331_REG_ADDR(0x00000444)
-#define IOP331_DMA1_DAR   (volatile u32 *)IOP331_REG_ADDR(0x0000044C)
-#define IOP331_DMA1_NDAR  (volatile u32 *)IOP331_REG_ADDR(0x00000450)
-#define IOP331_DMA1_PADR  (volatile u32 *)IOP331_REG_ADDR(0x00000454)
-#define IOP331_DMA1_PUADR (volatile u32 *)IOP331_REG_ADDR(0x00000458)
-#define IOP331_DMA1_LADR  (volatile u32 *)IOP331_REG_ADDR(0x0000045C)
-#define IOP331_DMA1_BCR   (volatile u32 *)IOP331_REG_ADDR(0x00000460)
-#define IOP331_DMA1_DCR   (volatile u32 *)IOP331_REG_ADDR(0x00000464)
-/* Reserved 0x00000468 through 0x000004FF */
-
-/* Memory controller 0x00000500 through 0x0005FF */
-
-/* Peripheral bus interface unit 0x00000680 through 0x0006FF */
-#define IOP331_PBCR       (volatile u32 *)IOP331_REG_ADDR(0x00000680)
-#define IOP331_PBISR      (volatile u32 *)IOP331_REG_ADDR(0x00000684)
-#define IOP331_PBBAR0     (volatile u32 *)IOP331_REG_ADDR(0x00000688)
-#define IOP331_PBLR0      (volatile u32 *)IOP331_REG_ADDR(0x0000068C)
-#define IOP331_PBBAR1     (volatile u32 *)IOP331_REG_ADDR(0x00000690)
-#define IOP331_PBLR1      (volatile u32 *)IOP331_REG_ADDR(0x00000694)
-#define IOP331_PBBAR2     (volatile u32 *)IOP331_REG_ADDR(0x00000698)
-#define IOP331_PBLR2      (volatile u32 *)IOP331_REG_ADDR(0x0000069C)
-#define IOP331_PBBAR3     (volatile u32 *)IOP331_REG_ADDR(0x000006A0)
-#define IOP331_PBLR3      (volatile u32 *)IOP331_REG_ADDR(0x000006A4)
-#define IOP331_PBBAR4     (volatile u32 *)IOP331_REG_ADDR(0x000006A8)
-#define IOP331_PBLR4      (volatile u32 *)IOP331_REG_ADDR(0x000006AC)
-#define IOP331_PBBAR5     (volatile u32 *)IOP331_REG_ADDR(0x000006B0)
-#define IOP331_PBLR5      (volatile u32 *)IOP331_REG_ADDR(0x000006B4)
-#define IOP331_PBDSCR     (volatile u32 *)IOP331_REG_ADDR(0x000006B8)
-/* Reserved 0x000006BC */
-#define IOP331_PMBR0      (volatile u32 *)IOP331_REG_ADDR(0x000006C0)
-/* Reserved 0x000006C4 through 0x000006DC */
-#define IOP331_PMBR1      (volatile u32 *)IOP331_REG_ADDR(0x000006E0)
-#define IOP331_PMBR2      (volatile u32 *)IOP331_REG_ADDR(0x000006E4)
-
-#define IOP331_PBCR_EN    0x1
-
-#define IOP331_PBISR_BOOR_ERR 0x1
-
-
-
-/* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
-/* Internal arbitration unit 0x00000780 through 0x0007BF */
-
-/* Interrupt Controller */
-#define IOP331_INTCTL0    (volatile u32 *)IOP331_REG_ADDR(0x00000790)
-#define IOP331_INTCTL1    (volatile u32 *)IOP331_REG_ADDR(0x00000794)
-#define IOP331_INTSTR0    (volatile u32 *)IOP331_REG_ADDR(0x00000798)
-#define IOP331_INTSTR1    (volatile u32 *)IOP331_REG_ADDR(0x0000079C)
-#define IOP331_IINTSRC0   (volatile u32 *)IOP331_REG_ADDR(0x000007A0)
-#define IOP331_IINTSRC1   (volatile u32 *)IOP331_REG_ADDR(0x000007A4)
-#define IOP331_FINTSRC0   (volatile u32 *)IOP331_REG_ADDR(0x000007A8)
-#define IOP331_FINTSRC1   (volatile u32 *)IOP331_REG_ADDR(0x000007AC)
-#define IOP331_IPR0       (volatile u32 *)IOP331_REG_ADDR(0x000007B0)
-#define IOP331_IPR1       (volatile u32 *)IOP331_REG_ADDR(0x000007B4)
-#define IOP331_IPR2       (volatile u32 *)IOP331_REG_ADDR(0x000007B8)
-#define IOP331_IPR3       (volatile u32 *)IOP331_REG_ADDR(0x000007BC)
-#define IOP331_INTBASE    (volatile u32 *)IOP331_REG_ADDR(0x000007C0)
-#define IOP331_INTSIZE    (volatile u32 *)IOP331_REG_ADDR(0x000007C4)
-#define IOP331_IINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007C8)
-#define IOP331_FINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007CC)
-
-
-/* Timers */
-
-#define IOP331_TU_TMR0		(volatile u32 *)IOP331_REG_ADDR(0x000007D0)
-#define IOP331_TU_TMR1		(volatile u32 *)IOP331_REG_ADDR(0x000007D4)
-
-#define IOP331_TMR_TC		0x01
-#define	IOP331_TMR_EN		0x02
-#define IOP331_TMR_RELOAD	0x04
-#define	IOP331_TMR_PRIVILEGED	0x09
-
-#define	IOP331_TMR_RATIO_1_1	0x00
-#define	IOP331_TMR_RATIO_4_1	0x10
-#define	IOP331_TMR_RATIO_8_1	0x20
-#define	IOP331_TMR_RATIO_16_1	0x30
-
-#define IOP331_TU_TCR0    (volatile u32 *)IOP331_REG_ADDR(0x000007D8)
-#define IOP331_TU_TCR1    (volatile u32 *)IOP331_REG_ADDR(0x000007DC)
-#define IOP331_TU_TRR0    (volatile u32 *)IOP331_REG_ADDR(0x000007E0)
-#define IOP331_TU_TRR1    (volatile u32 *)IOP331_REG_ADDR(0x000007E4)
-#define IOP331_TU_TISR    (volatile u32 *)IOP331_REG_ADDR(0x000007E8)
-#define IOP331_TU_WDTCR   (volatile u32 *)IOP331_REG_ADDR(0x000007EC)
-
-#if defined(CONFIG_ARCH_IOP33X)
-#define	IOP331_TICK_RATE	266000000	/* 266 MHz IB clock */
-#endif
-
-#if defined(CONFIG_IOP331_STEPD) || defined(CONFIG_ARCH_IQ80333)
-#undef IOP331_TICK_RATE
-#define IOP331_TICK_RATE	333000000	/* 333 Mhz IB clock */
-#endif
-
-/* Application accelerator unit 0x00000800 - 0x000008FF */
-#define IOP331_AAU_ACR     (volatile u32 *)IOP331_REG_ADDR(0x00000800)
-#define IOP331_AAU_ASR     (volatile u32 *)IOP331_REG_ADDR(0x00000804)
-#define IOP331_AAU_ADAR    (volatile u32 *)IOP331_REG_ADDR(0x00000808)
-#define IOP331_AAU_ANDAR   (volatile u32 *)IOP331_REG_ADDR(0x0000080C)
-#define IOP331_AAU_SAR1    (volatile u32 *)IOP331_REG_ADDR(0x00000810)
-#define IOP331_AAU_SAR2    (volatile u32 *)IOP331_REG_ADDR(0x00000814)
-#define IOP331_AAU_SAR3    (volatile u32 *)IOP331_REG_ADDR(0x00000818)
-#define IOP331_AAU_SAR4    (volatile u32 *)IOP331_REG_ADDR(0x0000081C)
-#define IOP331_AAU_SAR5    (volatile u32 *)IOP331_REG_ADDR(0x0000082C)
-#define IOP331_AAU_SAR6    (volatile u32 *)IOP331_REG_ADDR(0x00000830)
-#define IOP331_AAU_SAR7    (volatile u32 *)IOP331_REG_ADDR(0x00000834)
-#define IOP331_AAU_SAR8    (volatile u32 *)IOP331_REG_ADDR(0x00000838)
-#define IOP331_AAU_SAR9    (volatile u32 *)IOP331_REG_ADDR(0x00000840)
-#define IOP331_AAU_SAR10   (volatile u32 *)IOP331_REG_ADDR(0x00000844)
-#define IOP331_AAU_SAR11   (volatile u32 *)IOP331_REG_ADDR(0x00000848)
-#define IOP331_AAU_SAR12   (volatile u32 *)IOP331_REG_ADDR(0x0000084C)
-#define IOP331_AAU_SAR13   (volatile u32 *)IOP331_REG_ADDR(0x00000850)
-#define IOP331_AAU_SAR14   (volatile u32 *)IOP331_REG_ADDR(0x00000854)
-#define IOP331_AAU_SAR15   (volatile u32 *)IOP331_REG_ADDR(0x00000858)
-#define IOP331_AAU_SAR16   (volatile u32 *)IOP331_REG_ADDR(0x0000085C)
-#define IOP331_AAU_SAR17   (volatile u32 *)IOP331_REG_ADDR(0x00000864)
-#define IOP331_AAU_SAR18   (volatile u32 *)IOP331_REG_ADDR(0x00000868)
-#define IOP331_AAU_SAR19   (volatile u32 *)IOP331_REG_ADDR(0x0000086C)
-#define IOP331_AAU_SAR20   (volatile u32 *)IOP331_REG_ADDR(0x00000870)
-#define IOP331_AAU_SAR21   (volatile u32 *)IOP331_REG_ADDR(0x00000874)
-#define IOP331_AAU_SAR22   (volatile u32 *)IOP331_REG_ADDR(0x00000878)
-#define IOP331_AAU_SAR23   (volatile u32 *)IOP331_REG_ADDR(0x0000087C)
-#define IOP331_AAU_SAR24   (volatile u32 *)IOP331_REG_ADDR(0x00000880)
-#define IOP331_AAU_SAR25   (volatile u32 *)IOP331_REG_ADDR(0x00000888)
-#define IOP331_AAU_SAR26   (volatile u32 *)IOP331_REG_ADDR(0x0000088C)
-#define IOP331_AAU_SAR27   (volatile u32 *)IOP331_REG_ADDR(0x00000890)
-#define IOP331_AAU_SAR28   (volatile u32 *)IOP331_REG_ADDR(0x00000894)
-#define IOP331_AAU_SAR29   (volatile u32 *)IOP331_REG_ADDR(0x00000898)
-#define IOP331_AAU_SAR30   (volatile u32 *)IOP331_REG_ADDR(0x0000089C)
-#define IOP331_AAU_SAR31   (volatile u32 *)IOP331_REG_ADDR(0x000008A0)
-#define IOP331_AAU_SAR32   (volatile u32 *)IOP331_REG_ADDR(0x000008A4)
-#define IOP331_AAU_DAR     (volatile u32 *)IOP331_REG_ADDR(0x00000820)
-#define IOP331_AAU_ABCR    (volatile u32 *)IOP331_REG_ADDR(0x00000824)
-#define IOP331_AAU_ADCR    (volatile u32 *)IOP331_REG_ADDR(0x00000828)
-#define IOP331_AAU_EDCR0   (volatile u32 *)IOP331_REG_ADDR(0x0000083c)
-#define IOP331_AAU_EDCR1   (volatile u32 *)IOP331_REG_ADDR(0x00000860)
-#define IOP331_AAU_EDCR2   (volatile u32 *)IOP331_REG_ADDR(0x00000884)
-
-
-#define IOP331_SPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C0)
-#define IOP331_PPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C8)
-/* SSP serial port unit 0x00001600 - 0x0000167F */
-
-/* I2C bus interface unit 0x00001680 - 0x000016FF */
-/* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
-
-#define IOP331_ICR0       (volatile u32 *)IOP331_REG_ADDR(0x00001680)
-#define IOP331_ISR0       (volatile u32 *)IOP331_REG_ADDR(0x00001684)
-#define IOP331_ISAR0      (volatile u32 *)IOP331_REG_ADDR(0x00001688)
-#define IOP331_IDBR0      (volatile u32 *)IOP331_REG_ADDR(0x0000168C)
-/* Reserved 0x00001690 */
-#define IOP331_IBMR0      (volatile u32 *)IOP331_REG_ADDR(0x00001694)
-/* Reserved 0x00001698 */
-/* Reserved 0x0000169C */
-#define IOP331_ICR1       (volatile u32 *)IOP331_REG_ADDR(0x000016A0)
-#define IOP331_ISR1       (volatile u32 *)IOP331_REG_ADDR(0x000016A4)
-#define IOP331_ISAR1      (volatile u32 *)IOP331_REG_ADDR(0x000016A8)
-#define IOP331_IDBR1      (volatile u32 *)IOP331_REG_ADDR(0x000016AC)
-#define IOP331_IBMR1      (volatile u32 *)IOP331_REG_ADDR(0x000016B4)
-/* Reserved 0x000016B8 through 0x000016FF */
-
-/* 0x00001700 through 0x0000172C  UART 0 */
-
-/* Reserved 0x00001730 through 0x0000173F */
-
-/* 0x00001740 through 0x0000176C UART 1 */
-
-#define IOP331_UART0_PHYS  (IOP331_PHYS_MEM_BASE | 0x00001700)	/* UART #1 physical */
-#define IOP331_UART1_PHYS  (IOP331_PHYS_MEM_BASE | 0x00001740)	/* UART #2 physical */
-#define IOP331_UART0_VIRT  (IOP331_VIRT_MEM_BASE | 0x00001700) /* UART #1 virtual addr */
-#define IOP331_UART1_VIRT  (IOP331_VIRT_MEM_BASE | 0x00001740) /* UART #2 virtual addr */
-
-/* Reserved 0x00001770 through 0x0000177F */
-
-/* General Purpose I/O Registers */
-#define IOP331_GPOE       (volatile u32 *)IOP331_REG_ADDR(0x00001780)
-#define IOP331_GPID       (volatile u32 *)IOP331_REG_ADDR(0x00001784)
-#define IOP331_GPOD       (volatile u32 *)IOP331_REG_ADDR(0x00001788)
-
-/* Reserved 0x0000178c through 0x000019ff */
-
-
-#ifndef __ASSEMBLY__
-extern void iop331_map_io(void);
-extern void iop331_init_irq(void);
-extern void iop331_time_init(void);
-#endif
-
-#endif // _IOP331_HW_H_
diff --git a/include/asm-arm/arch-iop3xx/iq31244.h b/include/asm-arm/arch-iop3xx/iq31244.h
deleted file mode 100644
index 4177cfa8100f..000000000000
--- a/include/asm-arm/arch-iop3xx/iq31244.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/*
- * linux/include/asm/arch-iop3xx/iq31244.h
- *
- * Intel IQ31244 evaluation board registers
- */
-
-#ifndef _IQ31244_H_
-#define _IQ31244_H_
-
-#define	IQ31244_FLASHBASE	0xf0000000	/* Flash */
-#define	IQ31244_FLASHSIZE	0x00800000
-#define	IQ31244_FLASHWIDTH	2
-
-#define IQ31244_UART		0xfe800000	/* UART #1 */
-#define IQ31244_7SEG_1		0xfe840000	/* 7-Segment MSB */
-#define IQ31244_7SEG_0		0xfe850000	/* 7-Segment LSB (WO) */
-#define IQ31244_ROTARY_SW	0xfe8d0000	/* Rotary Switch */
-#define IQ31244_BATT_STAT	0xfe8f0000	/* Battery Status */
-
-#ifndef __ASSEMBLY__
-extern void iq31244_map_io(void);
-#endif
-
-#endif	// _IQ31244_H_
diff --git a/include/asm-arm/arch-iop3xx/iq80321.h b/include/asm-arm/arch-iop3xx/iq80321.h
deleted file mode 100644
index cb8725979ffa..000000000000
--- a/include/asm-arm/arch-iop3xx/iq80321.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/*
- * linux/include/asm/arch-iop3xx/iq80321.h
- *
- * Intel IQ80321 evaluation board registers
- */
-
-#ifndef _IQ80321_H_
-#define _IQ80321_H_
-
-#define	IQ80321_FLASHBASE	0xf0000000	/* Flash */
-#define	IQ80321_FLASHSIZE	0x00800000
-#define	IQ80321_FLASHWIDTH	1
-
-#define IQ80321_UART		0xfe800000	/* UART #1 */
-#define IQ80321_7SEG_1		0xfe840000	/* 7-Segment MSB */
-#define IQ80321_7SEG_0		0xfe850000	/* 7-Segment LSB (WO) */
-#define IQ80321_ROTARY_SW	0xfe8d0000	/* Rotary Switch */
-#define IQ80321_BATT_STAT	0xfe8f0000	/* Battery Status */
-
-#ifndef __ASSEMBLY__
-extern void iq80321_map_io(void);
-#endif
-
-#endif	// _IQ80321_H_
diff --git a/include/asm-arm/arch-iop3xx/iq80331.h b/include/asm-arm/arch-iop3xx/iq80331.h
deleted file mode 100644
index 0668e78d483e..000000000000
--- a/include/asm-arm/arch-iop3xx/iq80331.h
+++ /dev/null
@@ -1,23 +0,0 @@
-/*
- * linux/include/asm/arch-iop3xx/iq80331.h
- *
- * Intel IQ80331 evaluation board registers
- */
-
-#ifndef _IQ80331_H_
-#define _IQ80331_H_
-
-#define	IQ80331_FLASHBASE	0xc0000000	/* Flash */
-#define	IQ80331_FLASHSIZE	0x00800000
-#define	IQ80331_FLASHWIDTH	1
-
-#define IQ80331_7SEG_1		0xce840000	/* 7-Segment MSB */
-#define IQ80331_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
-#define IQ80331_ROTARY_SW	0xce8d0000	/* Rotary Switch */
-#define IQ80331_BATT_STAT	0xce8f0000	/* Battery Status */
-
-#ifndef __ASSEMBLY__
-extern void iq80331_map_io(void);
-#endif
-
-#endif	// _IQ80331_H_
diff --git a/include/asm-arm/arch-iop3xx/iq80332.h b/include/asm-arm/arch-iop3xx/iq80332.h
deleted file mode 100644
index e5fff1775d1a..000000000000
--- a/include/asm-arm/arch-iop3xx/iq80332.h
+++ /dev/null
@@ -1,23 +0,0 @@
-/*
- * linux/include/asm/arch-iop3xx/iq80332.h
- *
- * Intel IQ80332 evaluation board registers
- */
-
-#ifndef _IQ80332_H_
-#define _IQ80332_H_
-
-#define	IQ80332_FLASHBASE	0xc0000000	/* Flash */
-#define	IQ80332_FLASHSIZE	0x00800000
-#define	IQ80332_FLASHWIDTH	1
-
-#define IQ80332_7SEG_1		0xce840000	/* 7-Segment MSB */
-#define IQ80332_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
-#define IQ80332_ROTARY_SW	0xce8d0000	/* Rotary Switch */
-#define IQ80332_BATT_STAT	0xce8f0000	/* Battery Status */
-
-#ifndef __ASSEMBLY__
-extern void iq80332_map_io(void);
-#endif
-
-#endif	// _IQ80332_H_
diff --git a/include/asm-arm/arch-iop3xx/irqs.h b/include/asm-arm/arch-iop3xx/irqs.h
deleted file mode 100644
index 4f7c7aa87b4a..000000000000
--- a/include/asm-arm/arch-iop3xx/irqs.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/irqs.h
- *
- * Copyright:	(C) 2001-2003 MontaVista Software Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-
-/*
- * Chipset-specific bits
- */
-#ifdef CONFIG_ARCH_IOP32X
-#include "iop321-irqs.h"
-#endif
-
-#ifdef CONFIG_ARCH_IOP33X
-#include "iop331-irqs.h"
-#endif
diff --git a/include/asm-arm/arch-iop3xx/memory.h b/include/asm-arm/arch-iop3xx/memory.h
deleted file mode 100644
index 25666184e8fc..000000000000
--- a/include/asm-arm/arch-iop3xx/memory.h
+++ /dev/null
@@ -1,38 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/memory.h
- */
-
-#ifndef __ASM_ARCH_MEMORY_H
-#define __ASM_ARCH_MEMORY_H
-
-#include <asm/hardware.h>
-
-/*
- * Physical DRAM offset.
- */
-#ifndef CONFIG_ARCH_IOP33X
-#define PHYS_OFFSET	UL(0xa0000000)
-#else
-#define PHYS_OFFSET	UL(0x00000000)
-#endif
-
-/*
- * Virtual view <-> PCI DMA view memory address translations
- * virt_to_bus: Used to translate the virtual address to an
- *		address suitable to be passed to set_dma_addr
- * bus_to_virt: Used to convert an address for DMA operations
- *		to an address that the kernel can use.
- */
-#if defined(CONFIG_ARCH_IOP32X)
-
-#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP321_IATVR2)) | ((*IOP321_IABAR2) & 0xfffffff0))
-#define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP321_IALR2)) | ( *IOP321_IATVR2)))
-
-#elif defined(CONFIG_ARCH_IOP33X)
-
-#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP331_IATVR2)) | ((*IOP331_IABAR2) & 0xfffffff0))
-#define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP331_IALR2)) | ( *IOP331_IATVR2)))
-
-#endif
-
-#endif
diff --git a/include/asm-arm/arch-iop3xx/system.h b/include/asm-arm/arch-iop3xx/system.h
deleted file mode 100644
index a16cbb77a7f6..000000000000
--- a/include/asm-arm/arch-iop3xx/system.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/system.h
- *
- *  Copyright (C) 2001 MontaVista Software, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-static inline void arch_idle(void)
-{
-	cpu_do_idle();
-}
-
-
-static inline void arch_reset(char mode)
-{
-#ifdef CONFIG_ARCH_IOP32X
-    *IOP321_PCSR = 0x30;
-#endif
-
-#ifdef CONFIG_ARCH_IOP33X
-    *IOP331_PCSR = 0x30;
-#endif
-
-	if ( 1 && mode == 's') {
-		/* Jump into ROM at address 0 */
-		cpu_reset(0);
-	} else {
-		/* No on-chip reset capability */
-		cpu_reset(0);
-	}
-}
-
diff --git a/include/asm-arm/arch-iop3xx/timex.h b/include/asm-arm/arch-iop3xx/timex.h
deleted file mode 100644
index 14ca8d0f7b29..000000000000
--- a/include/asm-arm/arch-iop3xx/timex.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/timex.h
- *
- * IOP3xx architecture timex specifications
- */
-#include <asm/hardware.h>
-
-#if defined(CONFIG_ARCH_IQ80321) || defined(CONFIG_ARCH_IQ31244)
-
-#define CLOCK_TICK_RATE IOP321_TICK_RATE
-
-#elif defined(CONFIG_ARCH_IQ80331) || defined(CONFIG_MACH_IQ80332)
-
-#define CLOCK_TICK_RATE IOP331_TICK_RATE
-
-#else
-
-#error "No IOP3xx timex information for this architecture"
-
-#endif
diff --git a/include/asm-arm/arch-iop3xx/uncompress.h b/include/asm-arm/arch-iop3xx/uncompress.h
deleted file mode 100644
index 066c16bc1250..000000000000
--- a/include/asm-arm/arch-iop3xx/uncompress.h
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- *  linux/include/asm-arm/arch-iop3xx/uncompress.h
- */
-#include <asm/types.h>
-#include <asm/mach-types.h>
-#include <linux/serial_reg.h>
-#include <asm/hardware.h>
-
-#ifdef CONFIG_ARCH_IOP32X
-#define UTYPE unsigned char *
-#elif defined(CONFIG_ARCH_IOP33X)
-#define UTYPE u32 *
-#else
-#error "Missing IOP3xx arch type def"
-#endif
-
-static volatile UTYPE uart_base;
-
-#define TX_DONE (UART_LSR_TEMT|UART_LSR_THRE)
-
-static inline void putc(char c)
-{
-	while ((uart_base[UART_LSR] & TX_DONE) != TX_DONE)
-		barrier();
-	*uart_base = c;
-}
-
-static inline void flush(void)
-{
-}
-
-static __inline__ void __arch_decomp_setup(unsigned long arch_id)
-{
-        if(machine_is_iq80321())
-			uart_base = (volatile UTYPE)IQ80321_UART;
-		else if(machine_is_iq31244())
-			uart_base = (volatile UTYPE)IQ31244_UART;
-		else if(machine_is_iq80331() || machine_is_iq80332())
-			uart_base = (volatile UTYPE)IOP331_UART0_PHYS;
-		else
-			uart_base = (volatile UTYPE)0xfe800000;
-}
-
-/*
- * nothing to do
- */
-#define arch_decomp_setup()	__arch_decomp_setup(arch_id)
-#define arch_decomp_wdog()
diff --git a/include/asm-arm/arch-iop3xx/vmalloc.h b/include/asm-arm/arch-iop3xx/vmalloc.h
deleted file mode 100644
index 0f2f6847f93c..000000000000
--- a/include/asm-arm/arch-iop3xx/vmalloc.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- * linux/include/asm-arm/arch-iop3xx/vmalloc.h
- */
-
-/*
- * Just any arbitrary offset to the start of the vmalloc VM area: the
- * current 8MB value just means that there will be a 8MB "hole" after the
- * physical memory until the kernel virtual memory starts.  That means that
- * any out-of-bounds memory accesses will hopefully be caught.
- * The vmalloc() routines leaves a hole of 4kB between each vmalloced
- * area for the same reason. ;)
- */
-//#define VMALLOC_END       (0xe8000000)
-/* increase usable physical RAM to ~992M per RMK */
-#define VMALLOC_END       (0xfe000000)
-
-- 
cgit v1.2.3


From 7ae1f7ec525c32db441836ab0ab010b85cb819a2 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:12:53 +0100
Subject: [ARM] 3818/1: iop3xx: introduce arch/arm/plat-iop for shared
 iop32x/iop33x code

Introduce the arch/arm/plat-iop directory, for code shared between the
iop32x and iop33x, and move the common memory map setup bits there.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                     |  5 +++++
 arch/arm/Makefile                    |  1 +
 arch/arm/mach-iop32x/iq31244-mm.c    |  3 ++-
 arch/arm/mach-iop32x/iq80321-mm.c    |  3 ++-
 arch/arm/mach-iop32x/setup.c         | 22 ------------------
 arch/arm/mach-iop33x/Makefile        |  4 ++--
 arch/arm/mach-iop33x/iq80331-mm.c    | 35 -----------------------------
 arch/arm/mach-iop33x/iq80332-mm.c    | 35 -----------------------------
 arch/arm/mach-iop33x/setup.c         | 35 +++--------------------------
 arch/arm/plat-iop/Makefile           |  8 +++++++
 arch/arm/plat-iop/setup.c            | 38 +++++++++++++++++++++++++++++++
 include/asm-arm/arch-iop32x/iop321.h |  1 -
 include/asm-arm/arch-iop33x/iop331.h |  1 -
 include/asm-arm/hardware/iop3xx.h    | 43 ++++++++++++++++++++++++++++++++++++
 14 files changed, 104 insertions(+), 130 deletions(-)
 delete mode 100644 arch/arm/mach-iop33x/iq80331-mm.c
 delete mode 100644 arch/arm/mach-iop33x/iq80332-mm.c
 create mode 100644 arch/arm/plat-iop/Makefile
 create mode 100644 arch/arm/plat-iop/setup.c
 create mode 100644 include/asm-arm/hardware/iop3xx.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 0fb17bbb86bc..665ce173e664 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -205,6 +205,7 @@ config ARCH_IMX
 config ARCH_IOP32X
 	bool "IOP32x-based"
 	depends on MMU
+	select PLAT_IOP
 	select PCI
 	help
 	  Support for Intel's 80219 and IOP32X (XScale) family of
@@ -213,6 +214,7 @@ config ARCH_IOP32X
 config ARCH_IOP33X
 	bool "IOP33x-based"
 	depends on MMU
+	select PLAT_IOP
 	select PCI
 	help
 	  Support for Intel's IOP33X (XScale) family of processors.
@@ -362,6 +364,9 @@ source "arch/arm/mach-netx/Kconfig"
 config ARCH_ACORN
 	bool
 
+config PLAT_IOP
+	bool
+
 source arch/arm/mm/Kconfig
 
 #  bool 'Use XScale PMU as timer source' CONFIG_XSCALE_PMU_TIMER
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index ac786bf0f546..80cee786500c 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -158,6 +158,7 @@ core-$(CONFIG_FPE_FASTFPE)	+= $(FASTFPE_OBJ)
 core-$(CONFIG_VFP)		+= arch/arm/vfp/
 
 # If we have a common platform directory, then include it in the build.
+core-$(CONFIG_PLAT_IOP)		+= arch/arm/plat-iop/
 core-$(CONFIG_ARCH_OMAP)	+= arch/arm/plat-omap/
 
 drivers-$(CONFIG_OPROFILE)      += arch/arm/oprofile/
diff --git a/arch/arm/mach-iop32x/iq31244-mm.c b/arch/arm/mach-iop32x/iq31244-mm.c
index ffc6692c95ad..fba22d5d908f 100644
--- a/arch/arm/mach-iop32x/iq31244-mm.c
+++ b/arch/arm/mach-iop32x/iq31244-mm.c
@@ -21,6 +21,7 @@
 #include <asm/page.h>
 
 #include <asm/mach/map.h>
+#include <asm/hardware/iop3xx.h>
 
 
 /*
@@ -39,7 +40,7 @@ static struct map_desc iq31244_io_desc[] __initdata = {
 
 void __init iq31244_map_io(void)
 {
-	iop321_map_io();
+	iop3xx_map_io();
 
 	iotable_init(iq31244_io_desc, ARRAY_SIZE(iq31244_io_desc));
 }
diff --git a/arch/arm/mach-iop32x/iq80321-mm.c b/arch/arm/mach-iop32x/iq80321-mm.c
index 5f702a7ff622..b6a3079ad29e 100644
--- a/arch/arm/mach-iop32x/iq80321-mm.c
+++ b/arch/arm/mach-iop32x/iq80321-mm.c
@@ -21,6 +21,7 @@
 #include <asm/page.h>
 
 #include <asm/mach/map.h>
+#include <asm/hardware/iop3xx.h>
 
 
 /*
@@ -39,7 +40,7 @@ static struct map_desc iq80321_io_desc[] __initdata = {
 
 void __init iq80321_map_io(void)
 {
-	iop321_map_io();
+	iop3xx_map_io();
 
 	iotable_init(iq80321_io_desc, ARRAY_SIZE(iq80321_io_desc));
 }
diff --git a/arch/arm/mach-iop32x/setup.c b/arch/arm/mach-iop32x/setup.c
index 729678799271..48d806474f40 100644
--- a/arch/arm/mach-iop32x/setup.c
+++ b/arch/arm/mach-iop32x/setup.c
@@ -32,23 +32,6 @@
 
 #define IOP321_UART_XTAL 1843200
 
-/*
- * Standard IO mapping for all IOP321 based systems
- */
-static struct map_desc iop321_std_desc[] __initdata = {
-	 {	/* mem mapped registers */
-		.virtual	= IOP321_VIRT_MEM_BASE,
-		.pfn		= __phys_to_pfn(IOP321_PHYS_MEM_BASE),
-		.length		= 0x00002000,
-		.type		= MT_DEVICE
-	 }, {	/* PCI IO space */
-		.virtual	= IOP321_PCI_LOWER_IO_VA,
-		.pfn		= __phys_to_pfn(IOP321_PCI_LOWER_IO_PA),
-		.length		= IOP321_PCI_IO_WINDOW_SIZE,
-		.type		= MT_DEVICE
-	 }
-};
-
 #ifdef CONFIG_ARCH_IQ80321
 #define UARTBASE IQ80321_UART
 #define IRQ_UART IRQ_IQ80321_UART
@@ -126,11 +109,6 @@ void __init iop32x_init(void)
 		platform_add_devices(iop32x_devices,
 				ARRAY_SIZE(iop32x_devices));
 	}
-}
-
-void __init iop321_map_io(void)
-{
-	iotable_init(iop321_std_desc, ARRAY_SIZE(iop321_std_desc));
 	early_serial_setup(&iop321_serial_ports[0]);
 }
 
diff --git a/arch/arm/mach-iop33x/Makefile b/arch/arm/mach-iop33x/Makefile
index 0f1a8fffe187..7865923b0a60 100644
--- a/arch/arm/mach-iop33x/Makefile
+++ b/arch/arm/mach-iop33x/Makefile
@@ -7,5 +7,5 @@ obj-m			:=
 obj-n			:=
 obj-			:=
 
-obj-$(CONFIG_ARCH_IQ80331) += iq80331-mm.o iq80331-pci.o
-obj-$(CONFIG_MACH_IQ80332) += iq80332-mm.o iq80332-pci.o
+obj-$(CONFIG_ARCH_IQ80331) += iq80331-pci.o
+obj-$(CONFIG_MACH_IQ80332) += iq80332-pci.o
diff --git a/arch/arm/mach-iop33x/iq80331-mm.c b/arch/arm/mach-iop33x/iq80331-mm.c
deleted file mode 100644
index eb59445a0fca..000000000000
--- a/arch/arm/mach-iop33x/iq80331-mm.c
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * linux/arch/arm/mach-iop33x/iq80331-mm.c
- *
- * Low level memory initialization for iq80331 platform
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2003 Intel Corp.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-
-
-/*
- * IQ80331 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-
-void __init iq80331_map_io(void)
-{
-	iop331_map_io();
-}
diff --git a/arch/arm/mach-iop33x/iq80332-mm.c b/arch/arm/mach-iop33x/iq80332-mm.c
deleted file mode 100644
index f7bc73113ea8..000000000000
--- a/arch/arm/mach-iop33x/iq80332-mm.c
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * linux/arch/arm/mach-iop33x/iq80332-mm.c
- *
- * Low level memory initialization for iq80332 platform
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-
-
-/*
- * IQ80332 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-
-void __init iq80332_map_io(void)
-{
-	iop331_map_io();
-}
diff --git a/arch/arm/mach-iop33x/setup.c b/arch/arm/mach-iop33x/setup.c
index c14c7da5ec4c..67cb21cb0ad9 100644
--- a/arch/arm/mach-iop33x/setup.c
+++ b/arch/arm/mach-iop33x/setup.c
@@ -28,26 +28,10 @@
 #include <asm/hardware.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
+#include <asm/hardware/iop3xx.h>
 
 #define IOP331_UART_XTAL 33334000
 
-/*
- * Standard IO mapping for all IOP331 based systems
- */
-static struct map_desc iop331_std_desc[] __initdata = {
-	{	/* mem mapped registers */
-		.virtual	= IOP331_VIRT_MEM_BASE,
-		.pfn		= __phys_to_pfn(IOP331_PHYS_MEM_BASE),
-		.length		= 0x00002000,
-		.type		= MT_DEVICE
-	}, {	/* PCI IO space */
-		.virtual	= IOP331_PCI_LOWER_IO_VA,
-		.pfn		= __phys_to_pfn(IOP331_PCI_LOWER_IO_PA),
-		.length		= IOP331_PCI_IO_WINDOW_SIZE,
-		.type		= MT_DEVICE
-	}
-};
-
 static struct resource iop33x_uart0_resources[] = {
 	[0] = {
 		.start = IOP331_UART0_PHYS,
@@ -172,30 +156,17 @@ void __init iop33x_init(void)
 	}
 }
 
-void __init iop331_map_io(void)
-{
-	iotable_init(iop331_std_desc, ARRAY_SIZE(iop331_std_desc));
-}
-
 #ifdef CONFIG_ARCH_IOP33X
 extern void iop331_init_irq(void);
 extern struct sys_timer iop331_timer;
 #endif
 
-#ifdef CONFIG_ARCH_IQ80331
-extern void iq80331_map_io(void);
-#endif
-
-#ifdef CONFIG_MACH_IQ80332
-extern void iq80332_map_io(void);
-#endif
-
 #if defined(CONFIG_ARCH_IQ80331)
 MACHINE_START(IQ80331, "Intel IQ80331")
 	/* Maintainer: Intel Corp. */
 	.phys_io	= 0xfefff000,
 	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
-	.map_io		= iq80331_map_io,
+	.map_io		= iop3xx_map_io,
 	.init_irq	= iop331_init_irq,
 	.timer		= &iop331_timer,
 	.boot_params	= 0x0100,
@@ -207,7 +178,7 @@ MACHINE_START(IQ80332, "Intel IQ80332")
 	/* Maintainer: Intel Corp. */
 	.phys_io	= 0xfefff000,
 	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
-	.map_io		= iq80332_map_io,
+	.map_io		= iop3xx_map_io,
 	.init_irq	= iop331_init_irq,
 	.timer		= &iop331_timer,
 	.boot_params	= 0x0100,
diff --git a/arch/arm/plat-iop/Makefile b/arch/arm/plat-iop/Makefile
new file mode 100644
index 000000000000..b299103537c5
--- /dev/null
+++ b/arch/arm/plat-iop/Makefile
@@ -0,0 +1,8 @@
+#
+# Makefile for the linux kernel.
+#
+
+obj-y			:= setup.o
+obj-m			:=
+obj-n			:=
+obj-			:=
diff --git a/arch/arm/plat-iop/setup.c b/arch/arm/plat-iop/setup.c
new file mode 100644
index 000000000000..4689db638e95
--- /dev/null
+++ b/arch/arm/plat-iop/setup.c
@@ -0,0 +1,38 @@
+/*
+ * arch/arm/plat-iop/setup.c
+ *
+ * Author: Nicolas Pitre <nico@cam.org>
+ * Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <asm/mach/map.h>
+#include <asm/hardware/iop3xx.h>
+
+/*
+ * Standard IO mapping for all IOP3xx based systems
+ */
+static struct map_desc iop3xx_std_desc[] __initdata = {
+	 {	/* mem mapped registers */
+		.virtual	= IOP3XX_PERIPHERAL_VIRT_BASE,
+		.pfn		= __phys_to_pfn(IOP3XX_PERIPHERAL_PHYS_BASE),
+		.length		= IOP3XX_PERIPHERAL_SIZE,
+		.type		= MT_DEVICE,
+	 }, {	/* PCI IO space */
+		.virtual	= IOP3XX_PCI_LOWER_IO_VA,
+		.pfn		= __phys_to_pfn(IOP3XX_PCI_LOWER_IO_PA),
+		.length		= IOP3XX_PCI_IO_WINDOW_SIZE,
+		.type		= MT_DEVICE,
+	 },
+};
+
+void __init iop3xx_map_io(void)
+{
+	iotable_init(iop3xx_std_desc, ARRAY_SIZE(iop3xx_std_desc));
+}
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index 7ba93faf8da4..307272b07809 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -333,7 +333,6 @@
 
 
 #ifndef __ASSEMBLY__
-extern void iop321_map_io(void);
 extern void iop321_init_irq(void);
 extern void iop321_time_init(void);
 #endif
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index 780b707edb1e..21430f877ea7 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -350,7 +350,6 @@
 
 
 #ifndef __ASSEMBLY__
-extern void iop331_map_io(void);
 extern void iop331_init_irq(void);
 extern void iop331_time_init(void);
 #endif
diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
new file mode 100644
index 000000000000..c17cc19cdfab
--- /dev/null
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -0,0 +1,43 @@
+/*
+ * include/asm-arm/hardware/iop3xx.h
+ *
+ * Intel IOP32X and IOP33X register definitions
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __IOP3XX_H
+#define __IOP3XX_H
+
+/*
+ * IOP3XX processor registers
+ */
+#define IOP3XX_PERIPHERAL_PHYS_BASE	0xffffe000
+#define IOP3XX_PERIPHERAL_VIRT_BASE	0xfeffe000
+#define IOP3XX_PERIPHERAL_SIZE		0x00002000
+#define IOP3XX_REG_ADDR(reg)		(IOP3XX_PERIPHERAL_VIRT_BASE + (reg))
+
+
+/*
+ * IOP3XX I/O and Mem space regions for PCI autoconfiguration
+ */
+#define IOP3XX_PCI_MEM_WINDOW_SIZE	0x04000000
+#define IOP3XX_PCI_LOWER_MEM_PA		0x80000000
+
+#define IOP3XX_PCI_IO_WINDOW_SIZE	0x00010000
+#define IOP3XX_PCI_LOWER_IO_PA		0x90000000
+#define IOP3XX_PCI_LOWER_IO_VA		0xfe000000
+
+
+#ifndef __ASSEMBLY__
+void iop3xx_map_io(void);
+#endif
+
+
+#endif
-- 
cgit v1.2.3


From e25d64f1242e8586f6e20c26fd876a4d956a6c45 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:15:21 +0100
Subject: [ARM] 3819/1: iop3xx: factor out shared i2c code

Move the i2c bits shared between iop32x and iop33x to plat-iop/i2c.c
and include/asm-arm/hardware/iop3xx.h.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop32x/common.c        | 17 ++++----
 arch/arm/mach-iop32x/setup.c         | 53 ++---------------------
 arch/arm/mach-iop33x/setup.c         | 44 +-------------------
 arch/arm/plat-iop/Makefile           |  2 +-
 arch/arm/plat-iop/i2c.c              | 81 ++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-iop32x/iop321.h | 14 -------
 include/asm-arm/arch-iop33x/iop331.h | 16 -------
 include/asm-arm/hardware/iop3xx.h    | 15 +++++++
 8 files changed, 111 insertions(+), 131 deletions(-)
 create mode 100644 arch/arm/plat-iop/i2c.c

(limited to 'include')

diff --git a/arch/arm/mach-iop32x/common.c b/arch/arm/mach-iop32x/common.c
index 8044af6a54e1..17e7d650fecb 100644
--- a/arch/arm/mach-iop32x/common.c
+++ b/arch/arm/mach-iop32x/common.c
@@ -14,6 +14,7 @@
 
 #include <linux/delay.h>
 #include <asm/hardware.h>
+#include <asm/hardware/iop3xx.h>
 
 /*
  * Shared variables
@@ -39,23 +40,23 @@ void ep80219_power_off(void)
      */
 
 	/* Send the Address byte w/ the start condition */
-	*IOP321_IDBR1 = 0x60;
-	*IOP321_ICR1 = 0xE9;
+	*IOP3XX_IDBR1 = 0x60;
+	*IOP3XX_ICR1 = 0xE9;
     mdelay(1);
 
 	/* Send the START_MSG byte w/ no start or stop condition */
-	*IOP321_IDBR1 = 0x0F;
-	*IOP321_ICR1 = 0xE8;
+	*IOP3XX_IDBR1 = 0x0F;
+	*IOP3XX_ICR1 = 0xE8;
     mdelay(1);
 
 	/* Send the SHUTDOWN_COMPLETE Message ID byte w/ no start or stop condition */
-	*IOP321_IDBR1 = 0x03;
-	*IOP321_ICR1 = 0xE8;
+	*IOP3XX_IDBR1 = 0x03;
+	*IOP3XX_ICR1 = 0xE8;
     mdelay(1);
 
 	/* Send an ignored byte w/ stop condition */
-	*IOP321_IDBR1 = 0x00;
-	*IOP321_ICR1 = 0xEA;
+	*IOP3XX_IDBR1 = 0x00;
+	*IOP3XX_ICR1 = 0xEA;
 
 	while (1) ;
 }
diff --git a/arch/arm/mach-iop32x/setup.c b/arch/arm/mach-iop32x/setup.c
index 48d806474f40..1a03b0b41baa 100644
--- a/arch/arm/mach-iop32x/setup.c
+++ b/arch/arm/mach-iop32x/setup.c
@@ -29,6 +29,7 @@
 #include <asm/hardware.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
+#include <asm/hardware/iop3xx.h>
 
 #define IOP321_UART_XTAL 1843200
 
@@ -57,58 +58,10 @@ static struct uart_port iop321_serial_ports[] = {
 	}
 };
 
-static struct resource iop32x_i2c_0_resources[] = {
-	[0] = {
-		.start = 0xfffff680,
-		.end = 0xfffff698,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP321_I2C_0,
-		.end = IRQ_IOP321_I2C_0,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct resource iop32x_i2c_1_resources[] = {
-	[0] = {
-		.start = 0xfffff6a0,
-		.end = 0xfffff6b8,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP321_I2C_1,
-		.end = IRQ_IOP321_I2C_1,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct platform_device iop32x_i2c_0_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 0,
-	.num_resources = 2,
-	.resource = iop32x_i2c_0_resources
-};
-
-static struct platform_device iop32x_i2c_1_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 1,
-	.num_resources = 2,
-	.resource = iop32x_i2c_1_resources
-};
-
-static struct platform_device *iop32x_devices[] __initdata = {
-	&iop32x_i2c_0_controller,
-	&iop32x_i2c_1_controller
-};
-
 void __init iop32x_init(void)
 {
-	if(iop_is_321())
-	{
-		platform_add_devices(iop32x_devices,
-				ARRAY_SIZE(iop32x_devices));
-	}
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
 	early_serial_setup(&iop321_serial_ports[0]);
 }
 
diff --git a/arch/arm/mach-iop33x/setup.c b/arch/arm/mach-iop33x/setup.c
index 67cb21cb0ad9..ad737d6582b5 100644
--- a/arch/arm/mach-iop33x/setup.c
+++ b/arch/arm/mach-iop33x/setup.c
@@ -100,51 +100,9 @@ static struct platform_device iop33x_uart1 = {
        .resource = iop33x_uart1_resources,
 };
 
-static struct resource iop33x_i2c_0_resources[] = {
-	[0] = {
-		.start = 0xfffff680,
-		.end = 0xfffff698,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_I2C_0,
-		.end = IRQ_IOP331_I2C_0,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct resource iop33x_i2c_1_resources[] = {
-	[0] = {
-		.start = 0xfffff6a0,
-		.end = 0xfffff6b8,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_I2C_1,
-		.end = IRQ_IOP331_I2C_1,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct platform_device iop33x_i2c_0_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 0,
-	.num_resources = 2,
-	.resource = iop33x_i2c_0_resources
-};
-
-static struct platform_device iop33x_i2c_1_controller = {
-	.name = "IOP3xx-I2C",
-	.id = 1,
-	.num_resources = 2,
-	.resource = iop33x_i2c_1_resources
-};
-
 static struct platform_device *iop33x_devices[] __initdata = {
 	&iop33x_uart0,
 	&iop33x_uart1,
-	&iop33x_i2c_0_controller,
-	&iop33x_i2c_1_controller
 };
 
 void __init iop33x_init(void)
@@ -154,6 +112,8 @@ void __init iop33x_init(void)
 		platform_add_devices(iop33x_devices,
 				ARRAY_SIZE(iop33x_devices));
 	}
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
 }
 
 #ifdef CONFIG_ARCH_IOP33X
diff --git a/arch/arm/plat-iop/Makefile b/arch/arm/plat-iop/Makefile
index b299103537c5..be6ab7c71ca3 100644
--- a/arch/arm/plat-iop/Makefile
+++ b/arch/arm/plat-iop/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= setup.o
+obj-y			:= i2c.o setup.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/plat-iop/i2c.c b/arch/arm/plat-iop/i2c.c
new file mode 100644
index 000000000000..7ae149c2e982
--- /dev/null
+++ b/arch/arm/plat-iop/i2c.c
@@ -0,0 +1,81 @@
+/*
+ * arch/arm/plat-iop/i2c.c
+ *
+ * Author: Nicolas Pitre <nico@cam.org>
+ * Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+#include <linux/serial.h>
+#include <linux/tty.h>
+#include <linux/serial_core.h>
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/mach/map.h>
+#include <asm/setup.h>
+#include <asm/system.h>
+#include <asm/memory.h>
+#include <asm/hardware.h>
+#include <asm/hardware/iop3xx.h>
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+
+#ifdef CONFIG_ARCH_IOP32X
+#define IRQ_IOP3XX_I2C_0	IRQ_IOP321_I2C_0
+#define IRQ_IOP3XX_I2C_1	IRQ_IOP321_I2C_1
+#endif
+#ifdef CONFIG_ARCH_IOP33X
+#define IRQ_IOP3XX_I2C_0	IRQ_IOP331_I2C_0
+#define IRQ_IOP3XX_I2C_1	IRQ_IOP331_I2C_1
+#endif
+
+static struct resource iop3xx_i2c0_resources[] = {
+	[0] = {
+		.start	= 0xfffff680,
+		.end	= 0xfffff697,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_IOP3XX_I2C_0,
+		.end	= IRQ_IOP3XX_I2C_0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+struct platform_device iop3xx_i2c0_device = {
+	.name		= "IOP3xx-I2C",
+	.id		= 0,
+	.num_resources	= 2,
+	.resource	= iop3xx_i2c0_resources,
+};
+
+
+static struct resource iop3xx_i2c1_resources[] = {
+	[0] = {
+		.start	= 0xfffff6a0,
+		.end	= 0xfffff6b7,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_IOP3XX_I2C_1,
+		.end	= IRQ_IOP3XX_I2C_1,
+		.flags	= IORESOURCE_IRQ,
+	}
+};
+
+struct platform_device iop3xx_i2c1_device = {
+	.name		= "IOP3xx-I2C",
+	.id		= 1,
+	.num_resources	= 2,
+	.resource	= iop3xx_i2c1_resources,
+};
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index 307272b07809..1a82dd96bf50 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -314,20 +314,6 @@
 
 /* SSP serial port unit 0x00001600 - 0x0000167F */
 /* I2C bus interface unit 0x00001680 - 0x000016FF */
-#define IOP321_ICR0       (volatile u32 *)IOP321_REG_ADDR(0x00001680)
-#define IOP321_ISR0       (volatile u32 *)IOP321_REG_ADDR(0x00001684)
-#define IOP321_ISAR0      (volatile u32 *)IOP321_REG_ADDR(0x00001688)
-#define IOP321_IDBR0      (volatile u32 *)IOP321_REG_ADDR(0x0000168C)
-/* Reserved 0x00001690 */
-#define IOP321_IBMR0      (volatile u32 *)IOP321_REG_ADDR(0x00001694)
-/* Reserved 0x00001698 */
-/* Reserved 0x0000169C */
-#define IOP321_ICR1       (volatile u32 *)IOP321_REG_ADDR(0x000016A0)
-#define IOP321_ISR1       (volatile u32 *)IOP321_REG_ADDR(0x000016A4)
-#define IOP321_ISAR1      (volatile u32 *)IOP321_REG_ADDR(0x000016A8)
-#define IOP321_IDBR1      (volatile u32 *)IOP321_REG_ADDR(0x000016AC)
-#define IOP321_IBMR1      (volatile u32 *)IOP321_REG_ADDR(0x000016B4)
-/* Reserved 0x000016B8 through 0x000016FC */
 
 /* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
 
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index 21430f877ea7..a7f47122c5e1 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -311,22 +311,6 @@
 /* SSP serial port unit 0x00001600 - 0x0000167F */
 
 /* I2C bus interface unit 0x00001680 - 0x000016FF */
-/* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
-
-#define IOP331_ICR0       (volatile u32 *)IOP331_REG_ADDR(0x00001680)
-#define IOP331_ISR0       (volatile u32 *)IOP331_REG_ADDR(0x00001684)
-#define IOP331_ISAR0      (volatile u32 *)IOP331_REG_ADDR(0x00001688)
-#define IOP331_IDBR0      (volatile u32 *)IOP331_REG_ADDR(0x0000168C)
-/* Reserved 0x00001690 */
-#define IOP331_IBMR0      (volatile u32 *)IOP331_REG_ADDR(0x00001694)
-/* Reserved 0x00001698 */
-/* Reserved 0x0000169C */
-#define IOP331_ICR1       (volatile u32 *)IOP331_REG_ADDR(0x000016A0)
-#define IOP331_ISR1       (volatile u32 *)IOP331_REG_ADDR(0x000016A4)
-#define IOP331_ISAR1      (volatile u32 *)IOP331_REG_ADDR(0x000016A8)
-#define IOP331_IDBR1      (volatile u32 *)IOP331_REG_ADDR(0x000016AC)
-#define IOP331_IBMR1      (volatile u32 *)IOP331_REG_ADDR(0x000016B4)
-/* Reserved 0x000016B8 through 0x000016FF */
 
 /* 0x00001700 through 0x0000172C  UART 0 */
 
diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
index c17cc19cdfab..ea7d05970001 100644
--- a/include/asm-arm/hardware/iop3xx.h
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -23,6 +23,18 @@
 #define IOP3XX_PERIPHERAL_SIZE		0x00002000
 #define IOP3XX_REG_ADDR(reg)		(IOP3XX_PERIPHERAL_VIRT_BASE + (reg))
 
+/* I2C bus interface unit  */
+#define IOP3XX_ICR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1680)
+#define IOP3XX_ISR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1684)
+#define IOP3XX_ISAR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1688)
+#define IOP3XX_IDBR0		(volatile u32 *)IOP3XX_REG_ADDR(0x168c)
+#define IOP3XX_IBMR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1694)
+#define IOP3XX_ICR1		(volatile u32 *)IOP3XX_REG_ADDR(0x16a0)
+#define IOP3XX_ISR1		(volatile u32 *)IOP3XX_REG_ADDR(0x16a4)
+#define IOP3XX_ISAR1		(volatile u32 *)IOP3XX_REG_ADDR(0x16a8)
+#define IOP3XX_IDBR1		(volatile u32 *)IOP3XX_REG_ADDR(0x16ac)
+#define IOP3XX_IBMR1		(volatile u32 *)IOP3XX_REG_ADDR(0x16b4)
+
 
 /*
  * IOP3XX I/O and Mem space regions for PCI autoconfiguration
@@ -37,6 +49,9 @@
 
 #ifndef __ASSEMBLY__
 void iop3xx_map_io(void);
+
+extern struct platform_device iop3xx_i2c0_device;
+extern struct platform_device iop3xx_i2c1_device;
 #endif
 
 
-- 
cgit v1.2.3


From 0cb015f9dea8a40d82d170be1a4f39ff909890bf Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:16:23 +0100
Subject: [ARM] 3820/1: iop3xx: factor out shared pci code

Merge the iop32x PCI code and iop33x PCI code into plat-iop/pci.c.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/plat-iop/Makefile        |   2 +-
 arch/arm/plat-iop/pci.c           | 247 ++++++++++++++++++++++++++++++++++++++
 include/asm-arm/hardware/iop3xx.h |  60 +++++++++
 include/asm-arm/mach/pci.h        |   4 +
 4 files changed, 312 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/plat-iop/pci.c

(limited to 'include')

diff --git a/arch/arm/plat-iop/Makefile b/arch/arm/plat-iop/Makefile
index be6ab7c71ca3..efde7a513fb7 100644
--- a/arch/arm/plat-iop/Makefile
+++ b/arch/arm/plat-iop/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= i2c.o setup.o
+obj-y			:= i2c.o pci.o setup.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/plat-iop/pci.c b/arch/arm/plat-iop/pci.c
new file mode 100644
index 000000000000..e647812654f2
--- /dev/null
+++ b/arch/arm/plat-iop/pci.c
@@ -0,0 +1,247 @@
+/*
+ * arch/arm/plat-iop/pci.c
+ *
+ * PCI support for the Intel IOP32X and IOP33X processors
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/hardware.h>
+#include <asm/mach/pci.h>
+#include <asm/hardware/iop3xx.h>
+
+// #define DEBUG
+
+#ifdef DEBUG
+#define  DBG(x...) printk(x)
+#else
+#define  DBG(x...) do { } while (0)
+#endif
+
+/*
+ * This routine builds either a type0 or type1 configuration command.  If the
+ * bus is on the 803xx then a type0 made, else a type1 is created.
+ */
+static u32 iop3xx_cfg_address(struct pci_bus *bus, int devfn, int where)
+{
+	struct pci_sys_data *sys = bus->sysdata;
+	u32 addr;
+
+	if (sys->busnr == bus->number)
+		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
+	else
+		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
+
+	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
+
+	return addr;
+}
+
+/*
+ * This routine checks the status of the last configuration cycle.  If an error
+ * was detected it returns a 1, else it returns a 0.  The errors being checked
+ * are parity, master abort, target abort (master and target).  These types of
+ * errors occure during a config cycle where there is no device, like during
+ * the discovery stage.
+ */
+static int iop3xx_pci_status(void)
+{
+	unsigned int status;
+	int ret = 0;
+
+	/*
+	 * Check the status registers.
+	 */
+	status = *IOP3XX_ATUSR;
+	if (status & 0xf900) {
+		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
+		*IOP3XX_ATUSR = status & 0xf900;
+		ret = 1;
+	}
+
+	status = *IOP3XX_ATUISR;
+	if (status & 0x679f) {
+		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
+		*IOP3XX_ATUISR = status & 0x679f;
+		ret = 1;
+	}
+
+	return ret;
+}
+
+/*
+ * Simply write the address register and read the configuration
+ * data.  Note that the 4 nop's ensure that we are able to handle
+ * a delayed abort (in theory.)
+ */
+static inline u32 iop3xx_read(unsigned long addr)
+{
+	u32 val;
+
+	__asm__ __volatile__(
+		"str	%1, [%2]\n\t"
+		"ldr	%0, [%3]\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		: "=r" (val)
+		: "r" (addr), "r" (IOP3XX_OCCAR), "r" (IOP3XX_OCCDR));
+
+	return val;
+}
+
+/*
+ * The read routines must check the error status of the last configuration
+ * cycle.  If there was an error, the routine returns all hex f's.
+ */
+static int
+iop3xx_read_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 *value)
+{
+	unsigned long addr = iop3xx_cfg_address(bus, devfn, where);
+	u32 val = iop3xx_read(addr) >> ((where & 3) * 8);
+
+	if (iop3xx_pci_status())
+		val = 0xffffffff;
+
+	*value = val;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+iop3xx_write_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 value)
+{
+	unsigned long addr = iop3xx_cfg_address(bus, devfn, where);
+	u32 val;
+
+	if (size != 4) {
+		val = iop3xx_read(addr);
+		if (iop3xx_pci_status())
+			return PCIBIOS_SUCCESSFUL;
+
+		where = (where & 3) * 8;
+
+		if (size == 1)
+			val &= ~(0xff << where);
+		else
+			val &= ~(0xffff << where);
+
+		*IOP3XX_OCCDR = val | value << where;
+	} else {
+		asm volatile(
+			"str	%1, [%2]\n\t"
+			"str	%0, [%3]\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			:
+			: "r" (value), "r" (addr),
+			  "r" (IOP3XX_OCCAR), "r" (IOP3XX_OCCDR));
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops iop3xx_ops = {
+	.read	= iop3xx_read_config,
+	.write	= iop3xx_write_config,
+};
+
+/*
+ * When a PCI device does not exist during config cycles, the 80200 gets a
+ * bus error instead of returning 0xffffffff. This handler simply returns.
+ */
+static int
+iop3xx_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
+{
+	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
+		addr, fsr, regs->ARM_pc, regs->ARM_lr);
+
+	/*
+	 * If it was an imprecise abort, then we need to correct the
+	 * return address to be _after_ the instruction.
+	 */
+	if (fsr & (1 << 10))
+		regs->ARM_pc += 4;
+
+	return 0;
+}
+
+int iop3xx_pci_setup(int nr, struct pci_sys_data *sys)
+{
+	struct resource *res;
+
+	if (nr != 0)
+		return 0;
+
+	res = kzalloc(2 * sizeof(struct resource), GFP_KERNEL);
+	if (!res)
+		panic("PCI: unable to alloc resources");
+
+	res[0].start = IOP3XX_PCI_LOWER_IO_VA;
+	res[0].end   = IOP3XX_PCI_LOWER_IO_VA + IOP3XX_PCI_IO_WINDOW_SIZE - 1;
+	res[0].name  = "IOP3XX PCI I/O Space";
+	res[0].flags = IORESOURCE_IO;
+	request_resource(&ioport_resource, &res[0]);
+
+	res[1].start = IOP3XX_PCI_LOWER_MEM_PA;
+	res[1].end   = IOP3XX_PCI_LOWER_MEM_PA + IOP3XX_PCI_MEM_WINDOW_SIZE - 1;
+	res[1].name  = "IOP3XX PCI Memory Space";
+	res[1].flags = IORESOURCE_MEM;
+	request_resource(&iomem_resource, &res[1]);
+
+	sys->mem_offset = IOP3XX_PCI_LOWER_MEM_PA - IOP3XX_PCI_LOWER_MEM_BA;
+	sys->io_offset  = IOP3XX_PCI_LOWER_IO_VA - IOP3XX_PCI_LOWER_IO_BA;
+
+	sys->resource[0] = &res[0];
+	sys->resource[1] = &res[1];
+	sys->resource[2] = NULL;
+
+	return 1;
+}
+
+struct pci_bus *iop3xx_pci_scan_bus(int nr, struct pci_sys_data *sys)
+{
+	return pci_scan_bus(sys->busnr, &iop3xx_ops, sys);
+}
+
+void iop3xx_pci_preinit(void)
+{
+	DBG("PCI:  Intel 803xx PCI init code.\n");
+	DBG("ATU: IOP3XX_ATUCMD=0x%04x\n", *IOP3XX_ATUCMD);
+	DBG("ATU: IOP3XX_OMWTVR0=0x%04x, IOP3XX_OIOWTVR=0x%04x\n",
+			*IOP3XX_OMWTVR0,
+			*IOP3XX_OIOWTVR);
+	DBG("ATU: IOP3XX_ATUCR=0x%08x\n", *IOP3XX_ATUCR);
+	DBG("ATU: IOP3XX_IABAR0=0x%08x IOP3XX_IALR0=0x%08x IOP3XX_IATVR0=%08x\n",
+			*IOP3XX_IABAR0, *IOP3XX_IALR0, *IOP3XX_IATVR0);
+	DBG("ATU: IOP3XX_OMWTVR0=0x%08x\n", *IOP3XX_OMWTVR0);
+	DBG("ATU: IOP3XX_IABAR1=0x%08x IOP3XX_IALR1=0x%08x\n",
+			*IOP3XX_IABAR1, *IOP3XX_IALR1);
+	DBG("ATU: IOP3XX_ERBAR=0x%08x IOP3XX_ERLR=0x%08x IOP3XX_ERTVR=%08x\n",
+			*IOP3XX_ERBAR, *IOP3XX_ERLR, *IOP3XX_ERTVR);
+	DBG("ATU: IOP3XX_IABAR2=0x%08x IOP3XX_IALR2=0x%08x IOP3XX_IATVR2=%08x\n",
+			*IOP3XX_IABAR2, *IOP3XX_IALR2, *IOP3XX_IATVR2);
+	DBG("ATU: IOP3XX_IABAR3=0x%08x IOP3XX_IALR3=0x%08x IOP3XX_IATVR3=%08x\n",
+			*IOP3XX_IABAR3, *IOP3XX_IALR3, *IOP3XX_IATVR3);
+
+	hook_fault_code(16+6, iop3xx_pci_abort, SIGBUS, "imprecise external abort");
+}
diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
index ea7d05970001..d488ced2e12d 100644
--- a/include/asm-arm/hardware/iop3xx.h
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -23,6 +23,64 @@
 #define IOP3XX_PERIPHERAL_SIZE		0x00002000
 #define IOP3XX_REG_ADDR(reg)		(IOP3XX_PERIPHERAL_VIRT_BASE + (reg))
 
+/* Address Translation Unit  */
+#define IOP3XX_ATUVID		(volatile u16 *)IOP3XX_REG_ADDR(0x0100)
+#define IOP3XX_ATUDID		(volatile u16 *)IOP3XX_REG_ADDR(0x0102)
+#define IOP3XX_ATUCMD		(volatile u16 *)IOP3XX_REG_ADDR(0x0104)
+#define IOP3XX_ATUSR		(volatile u16 *)IOP3XX_REG_ADDR(0x0106)
+#define IOP3XX_ATURID		(volatile u8  *)IOP3XX_REG_ADDR(0x0108)
+#define IOP3XX_ATUCCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0109)
+#define IOP3XX_ATUCLSR		(volatile u8  *)IOP3XX_REG_ADDR(0x010c)
+#define IOP3XX_ATULT		(volatile u8  *)IOP3XX_REG_ADDR(0x010d)
+#define IOP3XX_ATUHTR		(volatile u8  *)IOP3XX_REG_ADDR(0x010e)
+#define IOP3XX_ATUBIST		(volatile u8  *)IOP3XX_REG_ADDR(0x010f)
+#define IOP3XX_IABAR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0110)
+#define IOP3XX_IAUBAR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0114)
+#define IOP3XX_IABAR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0118)
+#define IOP3XX_IAUBAR1		(volatile u32 *)IOP3XX_REG_ADDR(0x011c)
+#define IOP3XX_IABAR2		(volatile u32 *)IOP3XX_REG_ADDR(0x0120)
+#define IOP3XX_IAUBAR2		(volatile u32 *)IOP3XX_REG_ADDR(0x0124)
+#define IOP3XX_ASVIR		(volatile u16 *)IOP3XX_REG_ADDR(0x012c)
+#define IOP3XX_ASIR		(volatile u16 *)IOP3XX_REG_ADDR(0x012e)
+#define IOP3XX_ERBAR		(volatile u32 *)IOP3XX_REG_ADDR(0x0130)
+#define IOP3XX_ATUILR		(volatile u8  *)IOP3XX_REG_ADDR(0x013c)
+#define IOP3XX_ATUIPR		(volatile u8  *)IOP3XX_REG_ADDR(0x013d)
+#define IOP3XX_ATUMGNT		(volatile u8  *)IOP3XX_REG_ADDR(0x013e)
+#define IOP3XX_ATUMLAT		(volatile u8  *)IOP3XX_REG_ADDR(0x013f)
+#define IOP3XX_IALR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0140)
+#define IOP3XX_IATVR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0144)
+#define IOP3XX_ERLR		(volatile u32 *)IOP3XX_REG_ADDR(0x0148)
+#define IOP3XX_ERTVR		(volatile u32 *)IOP3XX_REG_ADDR(0x014c)
+#define IOP3XX_IALR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0150)
+#define IOP3XX_IALR2		(volatile u32 *)IOP3XX_REG_ADDR(0x0154)
+#define IOP3XX_IATVR2		(volatile u32 *)IOP3XX_REG_ADDR(0x0158)
+#define IOP3XX_OIOWTVR		(volatile u32 *)IOP3XX_REG_ADDR(0x015c)
+#define IOP3XX_OMWTVR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0160)
+#define IOP3XX_OUMWTVR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0164)
+#define IOP3XX_OMWTVR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0168)
+#define IOP3XX_OUMWTVR1		(volatile u32 *)IOP3XX_REG_ADDR(0x016c)
+#define IOP3XX_OUDWTVR		(volatile u32 *)IOP3XX_REG_ADDR(0x0178)
+#define IOP3XX_ATUCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0180)
+#define IOP3XX_PCSR		(volatile u32 *)IOP3XX_REG_ADDR(0x0184)
+#define IOP3XX_ATUISR		(volatile u32 *)IOP3XX_REG_ADDR(0x0188)
+#define IOP3XX_ATUIMR		(volatile u32 *)IOP3XX_REG_ADDR(0x018c)
+#define IOP3XX_IABAR3		(volatile u32 *)IOP3XX_REG_ADDR(0x0190)
+#define IOP3XX_IAUBAR3		(volatile u32 *)IOP3XX_REG_ADDR(0x0194)
+#define IOP3XX_IALR3		(volatile u32 *)IOP3XX_REG_ADDR(0x0198)
+#define IOP3XX_IATVR3		(volatile u32 *)IOP3XX_REG_ADDR(0x019c)
+#define IOP3XX_OCCAR		(volatile u32 *)IOP3XX_REG_ADDR(0x01a4)
+#define IOP3XX_OCCDR		(volatile u32 *)IOP3XX_REG_ADDR(0x01ac)
+#define IOP3XX_PDSCR		(volatile u32 *)IOP3XX_REG_ADDR(0x01bc)
+#define IOP3XX_PMCAPID		(volatile u8  *)IOP3XX_REG_ADDR(0x01c0)
+#define IOP3XX_PMNEXT		(volatile u8  *)IOP3XX_REG_ADDR(0x01c1)
+#define IOP3XX_APMCR		(volatile u16 *)IOP3XX_REG_ADDR(0x01c2)
+#define IOP3XX_APMCSR		(volatile u16 *)IOP3XX_REG_ADDR(0x01c4)
+#define IOP3XX_PCIXCAPID	(volatile u8  *)IOP3XX_REG_ADDR(0x01e0)
+#define IOP3XX_PCIXNEXT		(volatile u8  *)IOP3XX_REG_ADDR(0x01e1)
+#define IOP3XX_PCIXCMD		(volatile u16 *)IOP3XX_REG_ADDR(0x01e2)
+#define IOP3XX_PCIXSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01e4)
+#define IOP3XX_PCIIRSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01ec)
+
 /* I2C bus interface unit  */
 #define IOP3XX_ICR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1680)
 #define IOP3XX_ISR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1684)
@@ -41,10 +99,12 @@
  */
 #define IOP3XX_PCI_MEM_WINDOW_SIZE	0x04000000
 #define IOP3XX_PCI_LOWER_MEM_PA		0x80000000
+#define IOP3XX_PCI_LOWER_MEM_BA		(*IOP3XX_OMWTVR0)
 
 #define IOP3XX_PCI_IO_WINDOW_SIZE	0x00010000
 #define IOP3XX_PCI_LOWER_IO_PA		0x90000000
 #define IOP3XX_PCI_LOWER_IO_VA		0xfe000000
+#define IOP3XX_PCI_LOWER_IO_BA		(*IOP3XX_OIOWTVR)
 
 
 #ifndef __ASSEMBLY__
diff --git a/include/asm-arm/mach/pci.h b/include/asm-arm/mach/pci.h
index 923e0ca66200..cb41defad4a1 100644
--- a/include/asm-arm/mach/pci.h
+++ b/include/asm-arm/mach/pci.h
@@ -52,6 +52,10 @@ void pci_common_init(struct hw_pci *);
 /*
  * PCI controllers
  */
+extern int iop3xx_pci_setup(int nr, struct pci_sys_data *);
+extern struct pci_bus *iop3xx_pci_scan_bus(int nr, struct pci_sys_data *);
+extern void iop3xx_pci_preinit(void);
+
 extern int iop321_setup(int nr, struct pci_sys_data *);
 extern struct pci_bus *iop321_scan_bus(int nr, struct pci_sys_data *);
 extern void iop321_init(void);
-- 
cgit v1.2.3


From 7e9740b11529a0a69789fbe92d324f293e6266f6 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:17:36 +0100
Subject: [ARM] 3821/1: iop3xx: switch iop32x/iop33x over to shared pci code

Switch the iop32x and iop33x code over to the common PCI implementation,
and remove the (nearly identical) iop32x and iop33x PCI implementations.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop32x/Makefile          |   2 +-
 arch/arm/mach-iop32x/common.c          |   6 -
 arch/arm/mach-iop32x/iq31244-pci.c     |  45 +------
 arch/arm/mach-iop32x/iq80321-pci.c     |  45 +------
 arch/arm/mach-iop32x/irq.c             |   2 +-
 arch/arm/mach-iop32x/pci.c             | 220 --------------------------------
 arch/arm/mach-iop33x/Makefile          |   2 +-
 arch/arm/mach-iop33x/common.c          |  22 ----
 arch/arm/mach-iop33x/iq80331-pci.c     |  45 +------
 arch/arm/mach-iop33x/iq80332-pci.c     |  45 +------
 arch/arm/mach-iop33x/irq.c             |   2 +-
 arch/arm/mach-iop33x/pci.c             | 222 ---------------------------------
 include/asm-arm/arch-iop32x/hardware.h |  20 +--
 include/asm-arm/arch-iop32x/iop321.h   |  86 +------------
 include/asm-arm/arch-iop32x/memory.h   |   4 +-
 include/asm-arm/arch-iop32x/system.h   |   2 +-
 include/asm-arm/arch-iop33x/hardware.h |  20 +--
 include/asm-arm/arch-iop33x/iop331.h   |  96 +-------------
 include/asm-arm/arch-iop33x/memory.h   |   4 +-
 include/asm-arm/arch-iop33x/system.h   |   2 +-
 include/asm-arm/mach/pci.h             |   8 --
 21 files changed, 30 insertions(+), 870 deletions(-)
 delete mode 100644 arch/arm/mach-iop32x/pci.c
 delete mode 100644 arch/arm/mach-iop33x/common.c
 delete mode 100644 arch/arm/mach-iop33x/pci.c

(limited to 'include')

diff --git a/arch/arm/mach-iop32x/Makefile b/arch/arm/mach-iop32x/Makefile
index 94144f7efa6c..7266224ab6d7 100644
--- a/arch/arm/mach-iop32x/Makefile
+++ b/arch/arm/mach-iop32x/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= common.o setup.o irq.o pci.o time.o
+obj-y			:= common.o setup.o irq.o time.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/mach-iop32x/common.c b/arch/arm/mach-iop32x/common.c
index 17e7d650fecb..9a17a081327d 100644
--- a/arch/arm/mach-iop32x/common.c
+++ b/arch/arm/mach-iop32x/common.c
@@ -16,12 +16,6 @@
 #include <asm/hardware.h>
 #include <asm/hardware/iop3xx.h>
 
-/*
- * Shared variables
- */
-unsigned long iop3xx_pcibios_min_io = 0;
-unsigned long iop3xx_pcibios_min_mem = 0;
-
 #ifdef CONFIG_ARCH_EP80219
 #include <linux/kernel.h>
 /*
diff --git a/arch/arm/mach-iop32x/iq31244-pci.c b/arch/arm/mach-iop32x/iq31244-pci.c
index e447c8103ecd..605b79553747 100644
--- a/arch/arm/mach-iop32x/iq31244-pci.c
+++ b/arch/arm/mach-iop32x/iq31244-pci.c
@@ -67,51 +67,12 @@ iq31244_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
 	return PCI_IRQ_TABLE_LOOKUP(0, 7);
 }
 
-static int iq31244_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP321_PCI_LOWER_IO_VA;
-	res[0].end   = IOP321_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ31244 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP321_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP321_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ31244 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP321_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP321_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq31244_preinit(void)
-{
-	iop321_init();
-}
-
 static struct hw_pci iq31244_pci __initdata = {
 	.swizzle	= pci_std_swizzle,
 	.nr_controllers = 1,
-	.setup		= iq31244_setup,
-	.scan		= iop321_scan_bus,
-	.preinit	= iq31244_preinit,
+	.setup		= iop3xx_pci_setup,
+	.scan		= iop3xx_pci_scan_bus,
+	.preinit	= iop3xx_pci_preinit,
 	.map_irq	= iq31244_map_irq
 };
 
diff --git a/arch/arm/mach-iop32x/iq80321-pci.c b/arch/arm/mach-iop32x/iq80321-pci.c
index 8767950e131f..cedc37b968b7 100644
--- a/arch/arm/mach-iop32x/iq80321-pci.c
+++ b/arch/arm/mach-iop32x/iq80321-pci.c
@@ -61,51 +61,12 @@ iq80321_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
 	return pci_irq_table[idsel%4][pin-1];
 }
 
-static int iq80321_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP321_PCI_LOWER_IO_VA;
-	res[0].end   = IOP321_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ80321 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP321_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP321_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ80321 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP321_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP321_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq80321_preinit(void)
-{
-	iop321_init();
-}
-
 static struct hw_pci iq80321_pci __initdata = {
 	.swizzle	= pci_std_swizzle,
 	.nr_controllers = 1,
-	.setup		= iq80321_setup,
-	.scan		= iop321_scan_bus,
-	.preinit	= iq80321_preinit,
+	.setup		= iop3xx_pci_setup,
+	.scan		= iop3xx_pci_scan_bus,
+	.preinit	= iop3xx_pci_preinit,
 	.map_irq	= iq80321_map_irq
 };
 
diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index c5f633d2d632..76f2d561dbc1 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -84,7 +84,7 @@ void __init iop321_init_irq(void)
 	intstr_write(0);		// treat all as IRQ
 	if(machine_is_iq80321() ||
 	   machine_is_iq31244()) 	// all interrupts are inputs to chip
-		*IOP321_PCIIRSR = 0x0f;
+		*IOP3XX_PCIIRSR = 0x0f;
 
 	for(i = IOP321_IRQ_OFS; i < NR_IRQS; i++)
 	{
diff --git a/arch/arm/mach-iop32x/pci.c b/arch/arm/mach-iop32x/pci.c
deleted file mode 100644
index 1a7c683673f2..000000000000
--- a/arch/arm/mach-iop32x/pci.c
+++ /dev/null
@@ -1,220 +0,0 @@
-/*
- * arch/arm/mach-iop32x/pci.c
- *
- * PCI support for the Intel IOP321 chipset
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/ioport.h>
-
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/system.h>
-#include <asm/hardware.h>
-#include <asm/mach/pci.h>
-
-#include <asm/arch/iop321.h>
-
-// #define DEBUG
-
-#ifdef DEBUG
-#define  DBG(x...) printk(x)
-#else
-#define  DBG(x...) do { } while (0)
-#endif
-
-/*
- * This routine builds either a type0 or type1 configuration command.  If the
- * bus is on the 80321 then a type0 made, else a type1 is created.
- */
-static u32 iop321_cfg_address(struct pci_bus *bus, int devfn, int where)
-{
-	struct pci_sys_data *sys = bus->sysdata;
-	u32 addr;
-
-	if (sys->busnr == bus->number)
-		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
-	else
-		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
-
-	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
-
-	return addr;
-}
-
-/*
- * This routine checks the status of the last configuration cycle.  If an error
- * was detected it returns a 1, else it returns a 0.  The errors being checked
- * are parity, master abort, target abort (master and target).  These types of
- * errors occure during a config cycle where there is no device, like during
- * the discovery stage.
- */
-static int iop321_pci_status(void)
-{
-	unsigned int status;
-	int ret = 0;
-
-	/*
-	 * Check the status registers.
-	 */
-	status = *IOP321_ATUSR;
-	if (status & 0xf900)
-	{
-		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
-		*IOP321_ATUSR = status & 0xf900;
-		ret = 1;
-	}
-	status = *IOP321_ATUISR;
-	if (status & 0x679f)
-	{
-		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
-		*IOP321_ATUISR = status & 0x679f;
-		ret = 1;
-	}
-	return ret;
-}
-
-/*
- * Simply write the address register and read the configuration
- * data.  Note that the 4 nop's ensure that we are able to handle
- * a delayed abort (in theory.)
- */
-static inline u32 iop321_read(unsigned long addr)
-{
-	u32 val;
-
-	__asm__ __volatile__(
-		"str	%1, [%2]\n\t"
-		"ldr	%0, [%3]\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		: "=r" (val)
-		: "r" (addr), "r" (IOP321_OCCAR), "r" (IOP321_OCCDR));
-
-	return val;
-}
-
-/*
- * The read routines must check the error status of the last configuration
- * cycle.  If there was an error, the routine returns all hex f's.
- */
-static int
-iop321_read_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 *value)
-{
-	unsigned long addr = iop321_cfg_address(bus, devfn, where);
-	u32 val = iop321_read(addr) >> ((where & 3) * 8);
-
-	if( iop321_pci_status() )
-		val = 0xffffffff;
-
-	*value = val;
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int
-iop321_write_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 value)
-{
-	unsigned long addr = iop321_cfg_address(bus, devfn, where);
-	u32 val;
-
-	if (size != 4) {
-		val = iop321_read(addr);
-		if (!iop321_pci_status() == 0)
-			return PCIBIOS_SUCCESSFUL;
-
-		where = (where & 3) * 8;
-
-		if (size == 1)
-			val &= ~(0xff << where);
-		else
-			val &= ~(0xffff << where);
-
-		*IOP321_OCCDR = val | value << where;
-	} else {
-		asm volatile(
-			"str	%1, [%2]\n\t"
-			"str	%0, [%3]\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			:
-			: "r" (value), "r" (addr),
-			  "r" (IOP321_OCCAR), "r" (IOP321_OCCDR));
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops iop321_ops = {
-	.read	= iop321_read_config,
-	.write	= iop321_write_config,
-};
-
-/*
- * When a PCI device does not exist during config cycles, the 80200 gets a
- * bus error instead of returning 0xffffffff. This handler simply returns.
- */
-int
-iop321_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
-{
-	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
-		addr, fsr, regs->ARM_pc, regs->ARM_lr);
-
-	/*
-	 * If it was an imprecise abort, then we need to correct the
-	 * return address to be _after_ the instruction.
-	 */
-	if (fsr & (1 << 10))
-		regs->ARM_pc += 4;
-
-	return 0;
-}
-
-/*
- * Scan an IOP321 PCI bus.  sys->bus defines which bus we scan.
- */
-struct pci_bus *iop321_scan_bus(int nr, struct pci_sys_data *sys)
-{
-	return pci_scan_bus(sys->busnr, &iop321_ops, sys);
-}
-
-void iop321_init(void)
-{
-	DBG("PCI:  Intel 80321 PCI init code.\n");
-	DBG("ATU: IOP321_ATUCMD=0x%04x\n", *IOP321_ATUCMD);
-	DBG("ATU: IOP321_OMWTVR0=0x%04x, IOP321_OIOWTVR=0x%04x\n",
-			*IOP321_OMWTVR0,
-			*IOP321_OIOWTVR);
-	DBG("ATU: IOP321_ATUCR=0x%08x\n", *IOP321_ATUCR);
-	DBG("ATU: IOP321_IABAR0=0x%08x IOP321_IALR0=0x%08x IOP321_IATVR0=%08x\n",
-			*IOP321_IABAR0, *IOP321_IALR0, *IOP321_IATVR0);
-	DBG("ATU: IOP321_OMWTVR0=0x%08x\n", *IOP321_OMWTVR0);
-	DBG("ATU: IOP321_IABAR1=0x%08x IOP321_IALR1=0x%08x\n",
-			*IOP321_IABAR1, *IOP321_IALR1);
-	DBG("ATU: IOP321_ERBAR=0x%08x IOP321_ERLR=0x%08x IOP321_ERTVR=%08x\n",
-			*IOP321_ERBAR, *IOP321_ERLR, *IOP321_ERTVR);
-	DBG("ATU: IOP321_IABAR2=0x%08x IOP321_IALR2=0x%08x IOP321_IATVR2=%08x\n",
-			*IOP321_IABAR2, *IOP321_IALR2, *IOP321_IATVR2);
-	DBG("ATU: IOP321_IABAR3=0x%08x IOP321_IALR3=0x%08x IOP321_IATVR3=%08x\n",
-			*IOP321_IABAR3, *IOP321_IALR3, *IOP321_IATVR3);
-
-	hook_fault_code(16+6, iop321_pci_abort, SIGBUS, "imprecise external abort");
-}
-
diff --git a/arch/arm/mach-iop33x/Makefile b/arch/arm/mach-iop33x/Makefile
index 7865923b0a60..6c74f13b1139 100644
--- a/arch/arm/mach-iop33x/Makefile
+++ b/arch/arm/mach-iop33x/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= common.o setup.o irq.o pci.o time.o
+obj-y			:= setup.o irq.o time.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/mach-iop33x/common.c b/arch/arm/mach-iop33x/common.c
deleted file mode 100644
index dd75f2b8e74f..000000000000
--- a/arch/arm/mach-iop33x/common.c
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- * arch/arm/mach-iop33x/common.c
- *
- * Common routines shared across all IOP3xx implementations
- *
- * Author: Deepak Saxena <dsaxena@mvista.com>
- *
- * Copyright 2003 (c) MontaVista, Software, Inc.
- *
- * This file is licensed under  the terms of the GNU General Public
- * License version 2. This program is licensed "as is" without any
- * warranty of any kind, whether express or implied.
- */
-
-#include <linux/delay.h>
-#include <asm/hardware.h>
-
-/*
- * Shared variables
- */
-unsigned long iop3xx_pcibios_min_io = 0;
-unsigned long iop3xx_pcibios_min_mem = 0;
diff --git a/arch/arm/mach-iop33x/iq80331-pci.c b/arch/arm/mach-iop33x/iq80331-pci.c
index 71adb12a9421..8b0bed5e2f91 100644
--- a/arch/arm/mach-iop33x/iq80331-pci.c
+++ b/arch/arm/mach-iop33x/iq80331-pci.c
@@ -57,51 +57,12 @@ iq80331_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
 	return PCI_IRQ_TABLE_LOOKUP(1, 7);
 }
 
-static int iq80331_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP331_PCI_LOWER_IO_VA;
-	res[0].end   = IOP331_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ80331 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP331_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP331_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ80331 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP331_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP331_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq80331_preinit(void)
-{
-	iop331_init();
-}
-
 static struct hw_pci iq80331_pci __initdata = {
 	.swizzle	= pci_std_swizzle,
 	.nr_controllers = 1,
-	.setup		= iq80331_setup,
-	.scan		= iop331_scan_bus,
-	.preinit	= iq80331_preinit,
+	.setup		= iop3xx_pci_setup,
+	.scan		= iop3xx_pci_scan_bus,
+	.preinit	= iop3xx_pci_preinit,
 	.map_irq	= iq80331_map_irq
 };
 
diff --git a/arch/arm/mach-iop33x/iq80332-pci.c b/arch/arm/mach-iop33x/iq80332-pci.c
index f3010f85b1a2..0de8aa748dd8 100644
--- a/arch/arm/mach-iop33x/iq80332-pci.c
+++ b/arch/arm/mach-iop33x/iq80332-pci.c
@@ -63,51 +63,12 @@ iq80332_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
 	return PCI_IRQ_TABLE_LOOKUP(1, 7);
 }
 
-static int iq80332_setup(int nr, struct pci_sys_data *sys)
-{
-	struct resource *res;
-
-	if(nr != 0)
-		return 0;
-
-	res = kzalloc(sizeof(struct resource) * 2, GFP_KERNEL);
-	if (!res)
-		panic("PCI: unable to alloc resources");
-
-	res[0].start = IOP331_PCI_LOWER_IO_VA;
-	res[0].end   = IOP331_PCI_UPPER_IO_VA;
-	res[0].name  = "IQ80332 PCI I/O Space";
-	res[0].flags = IORESOURCE_IO;
-
-	res[1].start = IOP331_PCI_LOWER_MEM_PA;
-	res[1].end   = IOP331_PCI_UPPER_MEM_PA;
-	res[1].name  = "IQ80332 PCI Memory Space";
-	res[1].flags = IORESOURCE_MEM;
-
-	request_resource(&ioport_resource, &res[0]);
-	request_resource(&iomem_resource, &res[1]);
-
-	sys->mem_offset = IOP331_PCI_MEM_OFFSET;
-	sys->io_offset  = IOP331_PCI_IO_OFFSET;
-
-	sys->resource[0] = &res[0];
-	sys->resource[1] = &res[1];
-	sys->resource[2] = NULL;
-
-	return 1;
-}
-
-static void iq80332_preinit(void)
-{
-	iop331_init();
-}
-
 static struct hw_pci iq80332_pci __initdata = {
 	.swizzle	= pci_std_swizzle,
 	.nr_controllers = 1,
-	.setup		= iq80332_setup,
-	.scan		= iop331_scan_bus,
-	.preinit	= iq80332_preinit,
+	.setup		= iop3xx_pci_setup,
+	.scan		= iop3xx_pci_scan_bus,
+	.preinit	= iop3xx_pci_preinit,
 	.map_irq	= iq80332_map_irq
 };
 
diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
index 029fa78876a5..bcffc33a5be8 100644
--- a/arch/arm/mach-iop33x/irq.c
+++ b/arch/arm/mach-iop33x/irq.c
@@ -117,7 +117,7 @@ void __init iop331_init_irq(void)
 	intstr_write0(0);		// treat all as IRQ
     	intstr_write1(0);
 	if(machine_is_iq80331()) 	// all interrupts are inputs to chip
-		*IOP331_PCIIRSR = 0x0f;
+		*IOP3XX_PCIIRSR = 0x0f;
 
 	for(i = IOP331_IRQ_OFS; i < NR_IRQS; i++)
 	{
diff --git a/arch/arm/mach-iop33x/pci.c b/arch/arm/mach-iop33x/pci.c
deleted file mode 100644
index bf6aba91b47d..000000000000
--- a/arch/arm/mach-iop33x/pci.c
+++ /dev/null
@@ -1,222 +0,0 @@
-/*
- * arch/arm/mach-iop33x/pci.c
- *
- * PCI support for the Intel IOP331 chipset
- *
- * Author: Dave Jiang (dave.jiang@intel.com)
- * Copyright (C) 2003, 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/ioport.h>
-
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/system.h>
-#include <asm/hardware.h>
-#include <asm/mach/pci.h>
-
-#include <asm/arch/iop331.h>
-
-#undef DEBUG
-#undef DEBUG1
-
-#ifdef DEBUG
-#define  DBG(x...) printk(x)
-#else
-#define  DBG(x...) do { } while (0)
-#endif
-
-#ifdef DEBUG1
-#define  DBG1(x...) printk(x)
-#else
-#define  DBG1(x...) do { } while (0)
-#endif
-
-/*
- * This routine builds either a type0 or type1 configuration command.  If the
- * bus is on the 80331 then a type0 made, else a type1 is created.
- */
-static u32 iop331_cfg_address(struct pci_bus *bus, int devfn, int where)
-{
-	struct pci_sys_data *sys = bus->sysdata;
-	u32 addr;
-
-	if (sys->busnr == bus->number)
-		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
-	else
-		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
-
-	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
-
-	return addr;
-}
-
-/*
- * This routine checks the status of the last configuration cycle.  If an error
- * was detected it returns a 1, else it returns a 0.  The errors being checked
- * are parity, master abort, target abort (master and target).  These types of
- * errors occure during a config cycle where there is no device, like during
- * the discovery stage.
- */
-static int iop331_pci_status(void)
-{
-	unsigned int status;
-	int ret = 0;
-
-	/*
-	 * Check the status registers.
-	 */
-	status = *IOP331_ATUSR;
-	if (status & 0xf900)
-	{
-		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
-		*IOP331_ATUSR = status & 0xf900;
-		ret = 1;
-	}
-	status = *IOP331_ATUISR;
-	if (status & 0x679f)
-	{
-		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
-		*IOP331_ATUISR = status & 0x679f;
-		ret = 1;
-	}
-	return ret;
-}
-
-/*
- * Simply write the address register and read the configuration
- * data.  Note that the 4 nop's ensure that we are able to handle
- * a delayed abort (in theory.)
- */
-static inline u32 iop331_read(unsigned long addr)
-{
-	u32 val;
-
-	__asm__ __volatile__(
-		"str	%1, [%2]\n\t"
-		"ldr	%0, [%3]\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		: "=r" (val)
-		: "r" (addr), "r" (IOP331_OCCAR), "r" (IOP331_OCCDR));
-
-	return val;
-}
-
-/*
- * The read routines must check the error status of the last configuration
- * cycle.  If there was an error, the routine returns all hex f's.
- */
-static int
-iop331_read_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 *value)
-{
-	unsigned long addr = iop331_cfg_address(bus, devfn, where);
-	u32 val = iop331_read(addr) >> ((where & 3) * 8);
-
-	if( iop331_pci_status() )
-		val = 0xffffffff;
-
-	*value = val;
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int
-iop331_write_config(struct pci_bus *bus, unsigned int devfn, int where,
-		int size, u32 value)
-{
-	unsigned long addr = iop331_cfg_address(bus, devfn, where);
-	u32 val;
-
-	if (size != 4) {
-		val = iop331_read(addr);
-		if (!iop331_pci_status() == 0)
-			return PCIBIOS_SUCCESSFUL;
-
-		where = (where & 3) * 8;
-
-		if (size == 1)
-			val &= ~(0xff << where);
-		else
-			val &= ~(0xffff << where);
-
-		*IOP331_OCCDR = val | value << where;
-	} else {
-		asm volatile(
-			"str	%1, [%2]\n\t"
-			"str	%0, [%3]\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			"nop\n\t"
-			:
-			: "r" (value), "r" (addr),
-			  "r" (IOP331_OCCAR), "r" (IOP331_OCCDR));
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops iop331_ops = {
-	.read	= iop331_read_config,
-	.write	= iop331_write_config,
-};
-
-/*
- * When a PCI device does not exist during config cycles, the XScale gets a
- * bus error instead of returning 0xffffffff. This handler simply returns.
- */
-int
-iop331_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
-{
-	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
-		addr, fsr, regs->ARM_pc, regs->ARM_lr);
-
-	/*
-	 * If it was an imprecise abort, then we need to correct the
-	 * return address to be _after_ the instruction.
-	 */
-	if (fsr & (1 << 10))
-		regs->ARM_pc += 4;
-
-	return 0;
-}
-
-/*
- * Scan an IOP331 PCI bus.  sys->bus defines which bus we scan.
- */
-struct pci_bus *iop331_scan_bus(int nr, struct pci_sys_data *sys)
-{
-	return pci_scan_bus(sys->busnr, &iop331_ops, sys);
-}
-
-void iop331_init(void)
-{
-	DBG1("PCI:  Intel 80331 PCI init code.\n");
-	DBG1("\tATU: IOP331_ATUCMD=0x%04x\n", *IOP331_ATUCMD);
-	DBG1("\tATU: IOP331_OMWTVR0=0x%04x, IOP331_OIOWTVR=0x%04x\n",
-			*IOP331_OMWTVR0,
-			*IOP331_OIOWTVR);
-	DBG1("\tATU: IOP331_OMWTVR1=0x%04x\n", *IOP331_OMWTVR1);
-	DBG1("\tATU: IOP331_ATUCR=0x%08x\n", *IOP331_ATUCR);
-	DBG1("\tATU: IOP331_IABAR0=0x%08x IOP331_IALR0=0x%08x IOP331_IATVR0=%08x\n", *IOP331_IABAR0, *IOP331_IALR0, *IOP331_IATVR0);
-	DBG1("\tATU: IOP31_IABAR1=0x%08x IOP331_IALR1=0x%08x\n", *IOP331_IABAR1, *IOP331_IALR1);
-	DBG1("\tATU: IOP331_ERBAR=0x%08x IOP331_ERLR=0x%08x IOP331_ERTVR=%08x\n", *IOP331_ERBAR, *IOP331_ERLR, *IOP331_ERTVR);
-	DBG1("\tATU: IOP331_IABAR2=0x%08x IOP331_IALR2=0x%08x IOP331_IATVR2=%08x\n", *IOP331_IABAR2, *IOP331_IALR2, *IOP331_IATVR2);
-	DBG1("\tATU: IOP331_IABAR3=0x%08x IOP331_IALR3=0x%08x IOP331_IATVR3=%08x\n", *IOP331_IABAR3, *IOP331_IALR3, *IOP331_IATVR3);
-
-	hook_fault_code(16+6, iop331_pci_abort, SIGBUS, "imprecise external abort");
-}
-
diff --git a/include/asm-arm/arch-iop32x/hardware.h b/include/asm-arm/arch-iop32x/hardware.h
index 8fb10134a107..16d0630ab252 100644
--- a/include/asm-arm/arch-iop32x/hardware.h
+++ b/include/asm-arm/arch-iop32x/hardware.h
@@ -19,26 +19,10 @@
  */
 
 #define pcibios_assign_all_busses() 1
+#define PCIBIOS_MIN_IO		0x00000000
+#define PCIBIOS_MIN_MEM		0x00000000
 
 
-/*
- * The min PCI I/O and MEM space are dependent on what specific
- * chipset/platform we are running on, so instead of hardcoding with
- * #ifdefs, we just fill these in the platform level PCI init code.
- */
-#ifndef __ASSEMBLY__
-extern unsigned long iop3xx_pcibios_min_io;
-extern unsigned long iop3xx_pcibios_min_mem;
-
-extern unsigned int processor_id;
-#endif
-
-/*
- * We just set these to zero since they are really bogus anyways
- */
-#define PCIBIOS_MIN_IO      (iop3xx_pcibios_min_io)
-#define PCIBIOS_MIN_MEM     (iop3xx_pcibios_min_mem)
-
 /*
  * Generic chipset bits
  *
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index 1a82dd96bf50..e3c85a05e73a 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -24,27 +24,6 @@
 #define iop_is_321()		1
 #endif
 
-/*
- * IOP321 I/O and Mem space regions for PCI autoconfiguration
- */
-#define IOP321_PCI_IO_WINDOW_SIZE   0x00010000
-#define IOP321_PCI_LOWER_IO_PA      0x90000000
-#define IOP321_PCI_LOWER_IO_VA      0xfe000000
-#define IOP321_PCI_LOWER_IO_BA      (*IOP321_OIOWTVR)
-#define IOP321_PCI_UPPER_IO_PA      (IOP321_PCI_LOWER_IO_PA + IOP321_PCI_IO_WINDOW_SIZE - 1)
-#define IOP321_PCI_UPPER_IO_VA      (IOP321_PCI_LOWER_IO_VA + IOP321_PCI_IO_WINDOW_SIZE - 1)
-#define IOP321_PCI_UPPER_IO_BA      (IOP321_PCI_LOWER_IO_BA + IOP321_PCI_IO_WINDOW_SIZE - 1)
-#define IOP321_PCI_IO_OFFSET        (IOP321_PCI_LOWER_IO_VA - IOP321_PCI_LOWER_IO_BA)
-
-/* #define IOP321_PCI_MEM_WINDOW_SIZE  (~*IOP321_IALR1 + 1) */
-#define IOP321_PCI_MEM_WINDOW_SIZE  0x04000000 /* 64M outbound window */
-#define IOP321_PCI_LOWER_MEM_PA     0x80000000
-#define IOP321_PCI_LOWER_MEM_BA     (*IOP321_OMWTVR0)
-#define IOP321_PCI_UPPER_MEM_PA     (IOP321_PCI_LOWER_MEM_PA + IOP321_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP321_PCI_UPPER_MEM_BA     (IOP321_PCI_LOWER_MEM_BA + IOP321_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP321_PCI_MEM_OFFSET       (IOP321_PCI_LOWER_MEM_PA - IOP321_PCI_LOWER_MEM_BA)
-
-
 /*
  * IOP321 chipset registers
  */
@@ -55,69 +34,6 @@
 /* Reserved 0x00000000 through 0x000000FF */
 
 /* Address Translation Unit 0x00000100 through 0x000001FF */
-#define IOP321_ATUVID     (volatile u16 *)IOP321_REG_ADDR(0x00000100)
-#define IOP321_ATUDID     (volatile u16 *)IOP321_REG_ADDR(0x00000102)
-#define IOP321_ATUCMD     (volatile u16 *)IOP321_REG_ADDR(0x00000104)
-#define IOP321_ATUSR      (volatile u16 *)IOP321_REG_ADDR(0x00000106)
-#define IOP321_ATURID     (volatile u8  *)IOP321_REG_ADDR(0x00000108)
-#define IOP321_ATUCCR     (volatile u32 *)IOP321_REG_ADDR(0x00000109)
-#define IOP321_ATUCLSR    (volatile u8  *)IOP321_REG_ADDR(0x0000010C)
-#define IOP321_ATULT      (volatile u8  *)IOP321_REG_ADDR(0x0000010D)
-#define IOP321_ATUHTR     (volatile u8  *)IOP321_REG_ADDR(0x0000010E)
-#define IOP321_ATUBIST    (volatile u8  *)IOP321_REG_ADDR(0x0000010F)
-#define IOP321_IABAR0     (volatile u32 *)IOP321_REG_ADDR(0x00000110)
-#define IOP321_IAUBAR0    (volatile u32 *)IOP321_REG_ADDR(0x00000114)
-#define IOP321_IABAR1     (volatile u32 *)IOP321_REG_ADDR(0x00000118)
-#define IOP321_IAUBAR1    (volatile u32 *)IOP321_REG_ADDR(0x0000011C)
-#define IOP321_IABAR2     (volatile u32 *)IOP321_REG_ADDR(0x00000120)
-#define IOP321_IAUBAR2    (volatile u32 *)IOP321_REG_ADDR(0x00000124)
-#define IOP321_ASVIR      (volatile u16 *)IOP321_REG_ADDR(0x0000012C)
-#define IOP321_ASIR       (volatile u16 *)IOP321_REG_ADDR(0x0000012E)
-#define IOP321_ERBAR      (volatile u32 *)IOP321_REG_ADDR(0x00000130)
-/* Reserved 0x00000134 through 0x0000013B */
-#define IOP321_ATUILR     (volatile u8  *)IOP321_REG_ADDR(0x0000013C)
-#define IOP321_ATUIPR     (volatile u8  *)IOP321_REG_ADDR(0x0000013D)
-#define IOP321_ATUMGNT    (volatile u8  *)IOP321_REG_ADDR(0x0000013E)
-#define IOP321_ATUMLAT    (volatile u8  *)IOP321_REG_ADDR(0x0000013F)
-#define IOP321_IALR0      (volatile u32 *)IOP321_REG_ADDR(0x00000140)
-#define IOP321_IATVR0     (volatile u32 *)IOP321_REG_ADDR(0x00000144)
-#define IOP321_ERLR       (volatile u32 *)IOP321_REG_ADDR(0x00000148)
-#define IOP321_ERTVR      (volatile u32 *)IOP321_REG_ADDR(0x0000014C)
-#define IOP321_IALR1      (volatile u32 *)IOP321_REG_ADDR(0x00000150)
-#define IOP321_IALR2      (volatile u32 *)IOP321_REG_ADDR(0x00000154)
-#define IOP321_IATVR2     (volatile u32 *)IOP321_REG_ADDR(0x00000158)
-#define IOP321_OIOWTVR    (volatile u32 *)IOP321_REG_ADDR(0x0000015C)
-#define IOP321_OMWTVR0    (volatile u32 *)IOP321_REG_ADDR(0x00000160)
-#define IOP321_OUMWTVR0   (volatile u32 *)IOP321_REG_ADDR(0x00000164)
-#define IOP321_OMWTVR1    (volatile u32 *)IOP321_REG_ADDR(0x00000168)
-#define IOP321_OUMWTVR1   (volatile u32 *)IOP321_REG_ADDR(0x0000016C)
-/* Reserved 0x00000170 through 0x00000177*/
-#define IOP321_OUDWTVR    (volatile u32 *)IOP321_REG_ADDR(0x00000178)
-/* Reserved 0x0000017C through 0x0000017F*/
-#define IOP321_ATUCR      (volatile u32 *)IOP321_REG_ADDR(0x00000180)
-#define IOP321_PCSR       (volatile u32 *)IOP321_REG_ADDR(0x00000184)
-#define IOP321_ATUISR     (volatile u32 *)IOP321_REG_ADDR(0x00000188)
-#define IOP321_ATUIMR     (volatile u32 *)IOP321_REG_ADDR(0x0000018C)
-#define IOP321_IABAR3     (volatile u32 *)IOP321_REG_ADDR(0x00000190)
-#define IOP321_IAUBAR3    (volatile u32 *)IOP321_REG_ADDR(0x00000194)
-#define IOP321_IALR3      (volatile u32 *)IOP321_REG_ADDR(0x00000198)
-#define IOP321_IATVR3     (volatile u32 *)IOP321_REG_ADDR(0x0000019C)
-/* Reserved 0x000001A0 through 0x000001A3*/
-#define IOP321_OCCAR      (volatile u32 *)IOP321_REG_ADDR(0x000001A4)
-/* Reserved 0x000001A8 through 0x000001AB*/
-#define IOP321_OCCDR      (volatile u32 *)IOP321_REG_ADDR(0x000001AC)
-/* Reserved 0x000001B0 through 0x000001BB*/
-#define IOP321_PDSCR      (volatile u32 *)IOP321_REG_ADDR(0x000001BC)
-#define IOP321_PMCAPID    (volatile u8  *)IOP321_REG_ADDR(0x000001C0)
-#define IOP321_PMNEXT     (volatile u8  *)IOP321_REG_ADDR(0x000001C1)
-#define IOP321_APMCR      (volatile u16 *)IOP321_REG_ADDR(0x000001C2)
-#define IOP321_APMCSR     (volatile u16 *)IOP321_REG_ADDR(0x000001C4)
-/* Reserved 0x000001C6 through 0x000001DF */
-#define IOP321_PCIXCAPID  (volatile u8  *)IOP321_REG_ADDR(0x000001E0)
-#define IOP321_PCIXNEXT   (volatile u8  *)IOP321_REG_ADDR(0x000001E1)
-#define IOP321_PCIXCMD    (volatile u16 *)IOP321_REG_ADDR(0x000001E2)
-#define IOP321_PCIXSR     (volatile u32 *)IOP321_REG_ADDR(0x000001E4)
-#define IOP321_PCIIRSR    (volatile u32 *)IOP321_REG_ADDR(0x000001EC)
 
 /* Messaging Unit 0x00000300 through 0x000003FF */
 
@@ -317,6 +233,8 @@
 
 /* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
 
+#include <asm/hardware/iop3xx.h>
+
 
 #ifndef __ASSEMBLY__
 extern void iop321_init_irq(void);
diff --git a/include/asm-arm/arch-iop32x/memory.h b/include/asm-arm/arch-iop32x/memory.h
index b4073f15b405..4c64d9e7229b 100644
--- a/include/asm-arm/arch-iop32x/memory.h
+++ b/include/asm-arm/arch-iop32x/memory.h
@@ -20,8 +20,8 @@
  *		to an address that the kernel can use.
  */
 
-#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP321_IATVR2)) | ((*IOP321_IABAR2) & 0xfffffff0))
-#define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP321_IALR2)) | ( *IOP321_IATVR2)))
+#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP3XX_IATVR2)) | ((*IOP3XX_IABAR2) & 0xfffffff0))
+#define __bus_to_virt(x)	(__phys_to_virt(((x) & ~(*IOP3XX_IALR2)) | ( *IOP3XX_IATVR2)))
 
 
 #endif
diff --git a/include/asm-arm/arch-iop32x/system.h b/include/asm-arm/arch-iop32x/system.h
index d4c8d691e1b0..1ac207a0d52e 100644
--- a/include/asm-arm/arch-iop32x/system.h
+++ b/include/asm-arm/arch-iop32x/system.h
@@ -16,7 +16,7 @@ static inline void arch_idle(void)
 
 static inline void arch_reset(char mode)
 {
-    *IOP321_PCSR = 0x30;
+    *IOP3XX_PCSR = 0x30;
 
 	if ( 1 && mode == 's') {
 		/* Jump into ROM at address 0 */
diff --git a/include/asm-arm/arch-iop33x/hardware.h b/include/asm-arm/arch-iop33x/hardware.h
index 4a457084c5c6..5e3cb32af020 100644
--- a/include/asm-arm/arch-iop33x/hardware.h
+++ b/include/asm-arm/arch-iop33x/hardware.h
@@ -19,26 +19,10 @@
  */
 
 #define pcibios_assign_all_busses() 1
+#define PCIBIOS_MIN_IO		0x00000000
+#define PCIBIOS_MIN_MEM		0x00000000
 
 
-/*
- * The min PCI I/O and MEM space are dependent on what specific
- * chipset/platform we are running on, so instead of hardcoding with
- * #ifdefs, we just fill these in the platform level PCI init code.
- */
-#ifndef __ASSEMBLY__
-extern unsigned long iop3xx_pcibios_min_io;
-extern unsigned long iop3xx_pcibios_min_mem;
-
-extern unsigned int processor_id;
-#endif
-
-/*
- * We just set these to zero since they are really bogus anyways
- */
-#define PCIBIOS_MIN_IO      (iop3xx_pcibios_min_io)
-#define PCIBIOS_MIN_MEM     (iop3xx_pcibios_min_mem)
-
 /*
  * Generic chipset bits
  *
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index a7f47122c5e1..e85e1a2e1a86 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -23,27 +23,6 @@
 #define	iop_is_331()		1
 #endif
 
-/*
- * IOP331 I/O and Mem space regions for PCI autoconfiguration
- */
-#define IOP331_PCI_IO_WINDOW_SIZE   0x00010000
-#define IOP331_PCI_LOWER_IO_PA      0x90000000
-#define IOP331_PCI_LOWER_IO_VA      0xfe000000
-#define IOP331_PCI_LOWER_IO_BA      (*IOP331_OIOWTVR)
-#define IOP331_PCI_UPPER_IO_PA      (IOP331_PCI_LOWER_IO_PA + IOP331_PCI_IO_WINDOW_SIZE - 1)
-#define IOP331_PCI_UPPER_IO_VA      (IOP331_PCI_LOWER_IO_VA + IOP331_PCI_IO_WINDOW_SIZE - 1)
-#define IOP331_PCI_UPPER_IO_BA      (IOP331_PCI_LOWER_IO_BA + IOP331_PCI_IO_WINDOW_SIZE - 1)
-#define IOP331_PCI_IO_OFFSET        (IOP331_PCI_LOWER_IO_VA - IOP331_PCI_LOWER_IO_BA)
-
-/* this can be 128M if OMWTVR1 is set */
-#define IOP331_PCI_MEM_WINDOW_SIZE	0x04000000 /* 64M outbound window */
-/* #define IOP331_PCI_MEM_WINDOW_SIZE  (~*IOP331_IALR1 + 1) */
-#define IOP331_PCI_LOWER_MEM_PA     0x80000000
-#define IOP331_PCI_LOWER_MEM_BA     (*IOP331_OMWTVR0)
-#define IOP331_PCI_UPPER_MEM_PA     (IOP331_PCI_LOWER_MEM_PA + IOP331_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP331_PCI_UPPER_MEM_BA     (IOP331_PCI_LOWER_MEM_BA + IOP331_PCI_MEM_WINDOW_SIZE - 1)
-#define IOP331_PCI_MEM_OFFSET       (IOP331_PCI_LOWER_MEM_PA - IOP331_PCI_LOWER_MEM_BA)
-
 /*
  * IOP331 chipset registers
  */
@@ -54,79 +33,6 @@
 /* Reserved 0x00000000 through 0x000000FF */
 
 /* Address Translation Unit 0x00000100 through 0x000001FF */
-#define IOP331_ATUVID     (volatile u16 *)IOP331_REG_ADDR(0x00000100)
-#define IOP331_ATUDID     (volatile u16 *)IOP331_REG_ADDR(0x00000102)
-#define IOP331_ATUCMD     (volatile u16 *)IOP331_REG_ADDR(0x00000104)
-#define IOP331_ATUSR      (volatile u16 *)IOP331_REG_ADDR(0x00000106)
-#define IOP331_ATURID     (volatile u8  *)IOP331_REG_ADDR(0x00000108)
-#define IOP331_ATUCCR     (volatile u32 *)IOP331_REG_ADDR(0x00000109)
-#define IOP331_ATUCLSR    (volatile u8  *)IOP331_REG_ADDR(0x0000010C)
-#define IOP331_ATULT      (volatile u8  *)IOP331_REG_ADDR(0x0000010D)
-#define IOP331_ATUHTR     (volatile u8  *)IOP331_REG_ADDR(0x0000010E)
-#define IOP331_ATUBIST    (volatile u8  *)IOP331_REG_ADDR(0x0000010F)
-#define IOP331_IABAR0     (volatile u32 *)IOP331_REG_ADDR(0x00000110)
-#define IOP331_IAUBAR0    (volatile u32 *)IOP331_REG_ADDR(0x00000114)
-#define IOP331_IABAR1     (volatile u32 *)IOP331_REG_ADDR(0x00000118)
-#define IOP331_IAUBAR1    (volatile u32 *)IOP331_REG_ADDR(0x0000011C)
-#define IOP331_IABAR2     (volatile u32 *)IOP331_REG_ADDR(0x00000120)
-#define IOP331_IAUBAR2    (volatile u32 *)IOP331_REG_ADDR(0x00000124)
-#define IOP331_ASVIR      (volatile u16 *)IOP331_REG_ADDR(0x0000012C)
-#define IOP331_ASIR       (volatile u16 *)IOP331_REG_ADDR(0x0000012E)
-#define IOP331_ERBAR      (volatile u32 *)IOP331_REG_ADDR(0x00000130)
-#define IOP331_ATU_CAPPTR (volatile u32 *)IOP331_REG_ADDR(0x00000134)
-/* Reserved 0x00000138 through 0x0000013B */
-#define IOP331_ATUILR     (volatile u8  *)IOP331_REG_ADDR(0x0000013C)
-#define IOP331_ATUIPR     (volatile u8  *)IOP331_REG_ADDR(0x0000013D)
-#define IOP331_ATUMGNT    (volatile u8  *)IOP331_REG_ADDR(0x0000013E)
-#define IOP331_ATUMLAT    (volatile u8  *)IOP331_REG_ADDR(0x0000013F)
-#define IOP331_IALR0      (volatile u32 *)IOP331_REG_ADDR(0x00000140)
-#define IOP331_IATVR0     (volatile u32 *)IOP331_REG_ADDR(0x00000144)
-#define IOP331_ERLR       (volatile u32 *)IOP331_REG_ADDR(0x00000148)
-#define IOP331_ERTVR      (volatile u32 *)IOP331_REG_ADDR(0x0000014C)
-#define IOP331_IALR1      (volatile u32 *)IOP331_REG_ADDR(0x00000150)
-#define IOP331_IALR2      (volatile u32 *)IOP331_REG_ADDR(0x00000154)
-#define IOP331_IATVR2     (volatile u32 *)IOP331_REG_ADDR(0x00000158)
-#define IOP331_OIOWTVR    (volatile u32 *)IOP331_REG_ADDR(0x0000015C)
-#define IOP331_OMWTVR0    (volatile u32 *)IOP331_REG_ADDR(0x00000160)
-#define IOP331_OUMWTVR0   (volatile u32 *)IOP331_REG_ADDR(0x00000164)
-#define IOP331_OMWTVR1    (volatile u32 *)IOP331_REG_ADDR(0x00000168)
-#define IOP331_OUMWTVR1   (volatile u32 *)IOP331_REG_ADDR(0x0000016C)
-/* Reserved 0x00000170 through 0x00000177*/
-#define IOP331_OUDWTVR    (volatile u32 *)IOP331_REG_ADDR(0x00000178)
-/* Reserved 0x0000017C through 0x0000017F*/
-#define IOP331_ATUCR      (volatile u32 *)IOP331_REG_ADDR(0x00000180)
-#define IOP331_PCSR       (volatile u32 *)IOP331_REG_ADDR(0x00000184)
-#define IOP331_ATUISR     (volatile u32 *)IOP331_REG_ADDR(0x00000188)
-#define IOP331_ATUIMR     (volatile u32 *)IOP331_REG_ADDR(0x0000018C)
-#define IOP331_IABAR3     (volatile u32 *)IOP331_REG_ADDR(0x00000190)
-#define IOP331_IAUBAR3    (volatile u32 *)IOP331_REG_ADDR(0x00000194)
-#define IOP331_IALR3      (volatile u32 *)IOP331_REG_ADDR(0x00000198)
-#define IOP331_IATVR3     (volatile u32 *)IOP331_REG_ADDR(0x0000019C)
-/* Reserved 0x000001A0 through 0x000001A3*/
-#define IOP331_OCCAR      (volatile u32 *)IOP331_REG_ADDR(0x000001A4)
-/* Reserved 0x000001A8 through 0x000001AB*/
-#define IOP331_OCCDR      (volatile u32 *)IOP331_REG_ADDR(0x000001AC)
-/* Reserved 0x000001B0 through 0x000001BB*/
-#define IOP331_VPDCAPID   (volatile u8 *)IOP331_REG_ADDR(0x000001B8)
-#define IOP331_VPDNXTP    (volatile u8 *)IOP331_REG_ADDR(0x000001B9)
-#define IOP331_VPDAR	  (volatile u16 *)IOP331_REG_ADDR(0x000001BA)
-#define IOP331_VPDDR      (volatile u32 *)IOP331_REG_ADDR(0x000001BC)
-#define IOP331_PMCAPID    (volatile u8 *)IOP331_REG_ADDR(0x000001C0)
-#define IOP331_PMNEXT     (volatile u8 *)IOP331_REG_ADDR(0x000001C1)
-#define IOP331_APMCR      (volatile u16 *)IOP331_REG_ADDR(0x000001C2)
-#define IOP331_APMCSR     (volatile u16 *)IOP331_REG_ADDR(0x000001C4)
-/* Reserved 0x000001C6 through 0x000001CF */
-#define IOP331_MSICAPID   (volatile u8 *)IOP331_REG_ADDR(0x000001D0)
-#define IOP331_MSINXTP	  (volatile u8 *)IOP331_REG_ADDR(0x000001D1)
-#define IOP331_MSIMCR     (volatile u16 *)IOP331_REG_ADDR(0x000001D2)
-#define IOP331_MSIMAR     (volatile u32 *)IOP331_REG_ADDR(0x000001D4)
-#define IOP331_MSIMUAR	  (volatile u32 *)IOP331_REG_ADDR(0x000001D8)
-#define IOP331_MSIMDR	  (volatile u32 *)IOP331_REG_ADDR(0x000001DC)
-#define IOP331_PCIXCAPID  (volatile u8 *)IOP331_REG_ADDR(0x000001E0)
-#define IOP331_PCIXNEXT   (volatile u8 *)IOP331_REG_ADDR(0x000001E1)
-#define IOP331_PCIXCMD    (volatile u16 *)IOP331_REG_ADDR(0x000001E2)
-#define IOP331_PCIXSR     (volatile u32 *)IOP331_REG_ADDR(0x000001E4)
-#define IOP331_PCIIRSR    (volatile u32 *)IOP331_REG_ADDR(0x000001EC)
 
 /* Messaging Unit 0x00000300 through 0x000003FF */
 
@@ -332,6 +238,8 @@
 
 /* Reserved 0x0000178c through 0x000019ff */
 
+#include <asm/hardware/iop3xx.h>
+
 
 #ifndef __ASSEMBLY__
 extern void iop331_init_irq(void);
diff --git a/include/asm-arm/arch-iop33x/memory.h b/include/asm-arm/arch-iop33x/memory.h
index 5e47164934ce..de208d2cca4e 100644
--- a/include/asm-arm/arch-iop33x/memory.h
+++ b/include/asm-arm/arch-iop33x/memory.h
@@ -19,8 +19,8 @@
  * bus_to_virt: Used to convert an address for DMA operations
  *		to an address that the kernel can use.
  */
-#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP331_IATVR2)) | ((*IOP331_IABAR2) & 0xfffffff0))
-#define __bus_to_virt(x)    (__phys_to_virt(((x) & ~(*IOP331_IALR2)) | ( *IOP331_IATVR2)))
+#define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP3XX_IATVR2)) | ((*IOP3XX_IABAR2) & 0xfffffff0))
+#define __bus_to_virt(x)	(__phys_to_virt(((x) & ~(*IOP3XX_IALR2)) | ( *IOP3XX_IATVR2)))
 
 
 #endif
diff --git a/include/asm-arm/arch-iop33x/system.h b/include/asm-arm/arch-iop33x/system.h
index 43cc787ea629..8270ad9f86c8 100644
--- a/include/asm-arm/arch-iop33x/system.h
+++ b/include/asm-arm/arch-iop33x/system.h
@@ -16,7 +16,7 @@ static inline void arch_idle(void)
 
 static inline void arch_reset(char mode)
 {
-    *IOP331_PCSR = 0x30;
+    *IOP3XX_PCSR = 0x30;
 
 	if ( 1 && mode == 's') {
 		/* Jump into ROM at address 0 */
diff --git a/include/asm-arm/mach/pci.h b/include/asm-arm/mach/pci.h
index cb41defad4a1..24621c49a0c7 100644
--- a/include/asm-arm/mach/pci.h
+++ b/include/asm-arm/mach/pci.h
@@ -56,14 +56,6 @@ extern int iop3xx_pci_setup(int nr, struct pci_sys_data *);
 extern struct pci_bus *iop3xx_pci_scan_bus(int nr, struct pci_sys_data *);
 extern void iop3xx_pci_preinit(void);
 
-extern int iop321_setup(int nr, struct pci_sys_data *);
-extern struct pci_bus *iop321_scan_bus(int nr, struct pci_sys_data *);
-extern void iop321_init(void);
-
-extern int iop331_setup(int nr, struct pci_sys_data *);
-extern struct pci_bus *iop331_scan_bus(int nr, struct pci_sys_data *);
-extern void iop331_init(void);
-
 extern int dc21285_setup(int nr, struct pci_sys_data *);
 extern struct pci_bus *dc21285_scan_bus(int nr, struct pci_sys_data *);
 extern void dc21285_preinit(void);
-- 
cgit v1.2.3


From 48388b2a56ae5e0f1c422e84d536f31729469b17 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:18:16 +0100
Subject: [ARM] 3822/1: iop3xx: rewrite time handling

Merge and rewrite the iop32x/iop33x time code to do lost jiffy
tracking properly, and put the result in plat-iop/time.c.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/plat-iop/Makefile           |  2 +-
 arch/arm/plat-iop/time.c             | 94 ++++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-iop32x/iop321.h |  6 +++
 include/asm-arm/arch-iop33x/iop331.h |  6 +++
 include/asm-arm/hardware/iop3xx.h    | 20 ++++++++
 5 files changed, 127 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/plat-iop/time.c

(limited to 'include')

diff --git a/arch/arm/plat-iop/Makefile b/arch/arm/plat-iop/Makefile
index efde7a513fb7..d20cdec3a944 100644
--- a/arch/arm/plat-iop/Makefile
+++ b/arch/arm/plat-iop/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= i2c.o pci.o setup.o
+obj-y			:= i2c.o pci.o setup.o time.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/plat-iop/time.c b/arch/arm/plat-iop/time.c
new file mode 100644
index 000000000000..5730a0d7ed67
--- /dev/null
+++ b/arch/arm/plat-iop/time.c
@@ -0,0 +1,94 @@
+/*
+ * arch/arm/plat-iop/time.c
+ *
+ * Timer code for IOP32x and IOP33x based systems
+ *
+ * Author: Deepak Saxena <dsaxena@mvista.com>
+ *
+ * Copyright 2002-2003 MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/time.h>
+#include <linux/init.h>
+#include <linux/timex.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+#include <asm/mach/irq.h>
+#include <asm/mach/time.h>
+
+#ifdef CONFIG_ARCH_IOP32X
+#define IRQ_IOP3XX_TIMER0	IRQ_IOP321_TIMER0
+#else
+#ifdef CONFIG_ARCH_IOP33X
+#define IRQ_IOP3XX_TIMER0	IRQ_IOP331_TIMER0
+#endif
+#endif
+
+static unsigned long ticks_per_jiffy;
+static unsigned long ticks_per_usec;
+static unsigned long next_jiffy_time;
+
+unsigned long iop3xx_gettimeoffset(void)
+{
+	unsigned long offset;
+
+	offset = next_jiffy_time - *IOP3XX_TU_TCR1;
+
+	return offset / ticks_per_usec;
+}
+
+static irqreturn_t
+iop3xx_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	write_seqlock(&xtime_lock);
+
+	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (1));
+
+	while ((signed long)(next_jiffy_time - *IOP3XX_TU_TCR1)
+							>= ticks_per_jiffy) {
+		timer_tick(regs);
+		next_jiffy_time -= ticks_per_jiffy;
+	}
+
+	write_sequnlock(&xtime_lock);
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction iop3xx_timer_irq = {
+	.name		= "IOP3XX Timer Tick",
+	.handler	= iop3xx_timer_interrupt,
+	.flags		= IRQF_DISABLED | IRQF_TIMER,
+};
+
+void __init iop3xx_init_time(unsigned long tick_rate)
+{
+	u32 timer_ctl;
+
+	ticks_per_jiffy = (tick_rate + HZ/2) / HZ;
+	ticks_per_usec = tick_rate / 1000000;
+	next_jiffy_time = 0xffffffff;
+
+	timer_ctl = IOP3XX_TMR_EN | IOP3XX_TMR_PRIVILEGED |
+			IOP3XX_TMR_RELOAD | IOP3XX_TMR_RATIO_1_1;
+
+	/*
+	 * We use timer 0 for our timer interrupt, and timer 1 as
+	 * monotonic counter for tracking missed jiffies.
+	 */
+	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (ticks_per_jiffy - 1));
+	asm volatile("mcr p6, 0, %0, c0, c1, 0" : : "r" (timer_ctl));
+	asm volatile("mcr p6, 0, %0, c5, c1, 0" : : "r" (0xffffffff));
+	asm volatile("mcr p6, 0, %0, c1, c1, 0" : : "r" (timer_ctl));
+
+	setup_irq(IRQ_IOP3XX_TIMER0, &iop3xx_timer_irq);
+}
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index e3c85a05e73a..bd96b8d55a76 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -233,6 +233,12 @@
 
 /* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
 
+/*
+ * Peripherals that are shared between the iop32x and iop33x but
+ * located at different addresses.
+ */
+#define IOP3XX_TIMER_REG(reg)  (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07e0 + (reg))
+
 #include <asm/hardware/iop3xx.h>
 
 
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index e85e1a2e1a86..b301ef8f7f32 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -238,6 +238,12 @@
 
 /* Reserved 0x0000178c through 0x000019ff */
 
+/*
+ * Peripherals that are shared between the iop32x and iop33x but
+ * located at different addresses.
+ */
+#define IOP3XX_TIMER_REG(reg)  (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07d0 + (reg))
+
 #include <asm/hardware/iop3xx.h>
 
 
diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
index d488ced2e12d..b21ea41b149e 100644
--- a/include/asm-arm/hardware/iop3xx.h
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -81,6 +81,24 @@
 #define IOP3XX_PCIXSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01e4)
 #define IOP3XX_PCIIRSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01ec)
 
+/* Timers  */
+#define IOP3XX_TU_TMR0		(volatile u32 *)IOP3XX_TIMER_REG(0x0000)
+#define IOP3XX_TU_TMR1		(volatile u32 *)IOP3XX_TIMER_REG(0x0004)
+#define IOP3XX_TU_TCR0		(volatile u32 *)IOP3XX_TIMER_REG(0x0008)
+#define IOP3XX_TU_TCR1		(volatile u32 *)IOP3XX_TIMER_REG(0x000c)
+#define IOP3XX_TU_TRR0		(volatile u32 *)IOP3XX_TIMER_REG(0x0010)
+#define IOP3XX_TU_TRR1		(volatile u32 *)IOP3XX_TIMER_REG(0x0014)
+#define IOP3XX_TU_TISR		(volatile u32 *)IOP3XX_TIMER_REG(0x0018)
+#define IOP3XX_TU_WDTCR		(volatile u32 *)IOP3XX_TIMER_REG(0x001c)
+#define IOP3XX_TMR_TC		0x01
+#define IOP3XX_TMR_EN		0x02
+#define IOP3XX_TMR_RELOAD	0x04
+#define IOP3XX_TMR_PRIVILEGED	0x09
+#define IOP3XX_TMR_RATIO_1_1	0x00
+#define IOP3XX_TMR_RATIO_4_1	0x10
+#define IOP3XX_TMR_RATIO_8_1	0x20
+#define IOP3XX_TMR_RATIO_16_1	0x30
+
 /* I2C bus interface unit  */
 #define IOP3XX_ICR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1680)
 #define IOP3XX_ISR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1684)
@@ -109,6 +127,8 @@
 
 #ifndef __ASSEMBLY__
 void iop3xx_map_io(void);
+void iop3xx_init_time(unsigned long);
+unsigned long iop3xx_gettimeoffset(void);
 
 extern struct platform_device iop3xx_i2c0_device;
 extern struct platform_device iop3xx_i2c1_device;
-- 
cgit v1.2.3


From 863753a81e4f863015be34900dc2ba3637622f34 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:19:02 +0100
Subject: [ARM] 3823/1: iop3xx: switch iop32x/iop33x over to shared time code

Switch the iop32x and iop33x code over to the common time implementation,
and remove the (nearly identical) iop32x and iop33x time implementations.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop32x/Makefile        |   2 +-
 arch/arm/mach-iop32x/setup.c         |  15 +++--
 arch/arm/mach-iop32x/time.c          | 108 -----------------------------------
 arch/arm/mach-iop33x/Makefile        |   2 +-
 arch/arm/mach-iop33x/setup.c         |  12 +++-
 arch/arm/mach-iop33x/time.c          | 106 ----------------------------------
 include/asm-arm/arch-iop32x/iop321.h |  21 -------
 include/asm-arm/arch-iop32x/timex.h  |   2 +-
 include/asm-arm/arch-iop33x/iop331.h |  21 -------
 include/asm-arm/arch-iop33x/timex.h  |   2 +-
 10 files changed, 26 insertions(+), 265 deletions(-)
 delete mode 100644 arch/arm/mach-iop32x/time.c
 delete mode 100644 arch/arm/mach-iop33x/time.c

(limited to 'include')

diff --git a/arch/arm/mach-iop32x/Makefile b/arch/arm/mach-iop32x/Makefile
index 7266224ab6d7..ef561db20c9b 100644
--- a/arch/arm/mach-iop32x/Makefile
+++ b/arch/arm/mach-iop32x/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= common.o setup.o irq.o time.o
+obj-y			:= common.o setup.o irq.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/mach-iop32x/setup.c b/arch/arm/mach-iop32x/setup.c
index 1a03b0b41baa..68de247a4cca 100644
--- a/arch/arm/mach-iop32x/setup.c
+++ b/arch/arm/mach-iop32x/setup.c
@@ -29,6 +29,7 @@
 #include <asm/hardware.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
+#include <asm/mach/time.h>
 #include <asm/hardware/iop3xx.h>
 
 #define IOP321_UART_XTAL 1843200
@@ -67,16 +68,22 @@ void __init iop32x_init(void)
 
 #ifdef CONFIG_ARCH_IQ80321
 extern void iq80321_map_io(void);
-extern struct sys_timer iop321_timer;
-extern void iop321_init_time(void);
 #endif
 
 #ifdef CONFIG_ARCH_IQ31244
 extern void iq31244_map_io(void);
-extern struct sys_timer iop321_timer;
-extern void iop321_init_time(void);
 #endif
 
+static void __init iop3xx_timer_init(void)
+{
+	iop3xx_init_time(IOP321_TICK_RATE);
+}
+
+struct sys_timer iop321_timer = {
+	.init		= iop3xx_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
 #if defined(CONFIG_ARCH_IQ80321)
 MACHINE_START(IQ80321, "Intel IQ80321")
 	/* Maintainer: Intel Corporation */
diff --git a/arch/arm/mach-iop32x/time.c b/arch/arm/mach-iop32x/time.c
deleted file mode 100644
index 3cabbbca0a02..000000000000
--- a/arch/arm/mach-iop32x/time.c
+++ /dev/null
@@ -1,108 +0,0 @@
-/*
- * arch/arm/mach-iop32x/time.c
- *
- * Timer code for IOP321 based systems
- *
- * Author: Deepak Saxena <dsaxena@mvista.com>
- *
- * Copyright 2002-2003 MontaVista Software Inc.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/interrupt.h>
-#include <linux/time.h>
-#include <linux/init.h>
-#include <linux/timex.h>
-
-#include <asm/hardware.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
-#include <asm/mach/irq.h>
-#include <asm/mach/time.h>
-
-#define IOP321_TIME_SYNC 0
-
-static inline unsigned long get_elapsed(void)
-{
-	return LATCH - *IOP321_TU_TCR0;
-}
-
-static unsigned long iop321_gettimeoffset(void)
-{
-	unsigned long elapsed, usec;
-	u32 tisr1, tisr2;
-
-	/*
-	 * If an interrupt was pending before we read the timer,
-	 * we've already wrapped.  Factor this into the time.
-	 * If an interrupt was pending after we read the timer,
-	 * it may have wrapped between checking the interrupt
-	 * status and reading the timer.  Re-read the timer to
-	 * be sure its value is after the wrap.
-	 */
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr1));
-	elapsed = get_elapsed();
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr2));
-
-	if(tisr1 & 1)
-		elapsed += LATCH;
-	else if (tisr2 & 1)
-		elapsed = LATCH + get_elapsed();
-
-	/*
-	 * Now convert them to usec.
-	 */
-	usec = (unsigned long)(elapsed / (CLOCK_TICK_RATE/1000000));
-
-	return usec;
-}
-
-static irqreturn_t
-iop321_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
-{
-	u32 tisr;
-
-	write_seqlock(&xtime_lock);
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr));
-	tisr |= 1;
-	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (tisr));
-
-	timer_tick(regs);
-
-	write_sequnlock(&xtime_lock);
-
-	return IRQ_HANDLED;
-}
-
-static struct irqaction iop321_timer_irq = {
-	.name		= "IOP321 Timer Tick",
-	.handler	= iop321_timer_interrupt,
-	.flags		= IRQF_DISABLED | IRQF_TIMER,
-};
-
-static void __init iop321_timer_init(void)
-{
-	u32 timer_ctl;
-
-	setup_irq(IRQ_IOP321_TIMER0, &iop321_timer_irq);
-
-	timer_ctl = IOP321_TMR_EN | IOP321_TMR_PRIVILEGED | IOP321_TMR_RELOAD |
-			IOP321_TMR_RATIO_1_1;
-
-	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (LATCH));
-
-	asm volatile("mcr p6, 0, %0, c0, c1, 0"	: : "r" (timer_ctl));
-}
-
-struct sys_timer iop321_timer = {
-	.init		= &iop321_timer_init,
-	.offset		= iop321_gettimeoffset,
-};
diff --git a/arch/arm/mach-iop33x/Makefile b/arch/arm/mach-iop33x/Makefile
index 6c74f13b1139..f825cee57d9f 100644
--- a/arch/arm/mach-iop33x/Makefile
+++ b/arch/arm/mach-iop33x/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= setup.o irq.o time.o
+obj-y			:= setup.o irq.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/mach-iop33x/setup.c b/arch/arm/mach-iop33x/setup.c
index ad737d6582b5..7cf5015436f3 100644
--- a/arch/arm/mach-iop33x/setup.c
+++ b/arch/arm/mach-iop33x/setup.c
@@ -28,6 +28,7 @@
 #include <asm/hardware.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
+#include <asm/mach/time.h>
 #include <asm/hardware/iop3xx.h>
 
 #define IOP331_UART_XTAL 33334000
@@ -118,9 +119,18 @@ void __init iop33x_init(void)
 
 #ifdef CONFIG_ARCH_IOP33X
 extern void iop331_init_irq(void);
-extern struct sys_timer iop331_timer;
 #endif
 
+static void __init iop3xx_timer_init(void)
+{
+	iop3xx_init_time(IOP331_TICK_RATE);
+}
+
+struct sys_timer iop331_timer = {
+	.init		= iop3xx_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
 #if defined(CONFIG_ARCH_IQ80331)
 MACHINE_START(IQ80331, "Intel IQ80331")
 	/* Maintainer: Intel Corp. */
diff --git a/arch/arm/mach-iop33x/time.c b/arch/arm/mach-iop33x/time.c
deleted file mode 100644
index d839cd0d926e..000000000000
--- a/arch/arm/mach-iop33x/time.c
+++ /dev/null
@@ -1,106 +0,0 @@
-/*
- * arch/arm/mach-iop33x/time.c
- *
- * Timer code for IOP331 based systems
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- *
- * Copyright 2003 Intel Corp.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/interrupt.h>
-#include <linux/time.h>
-#include <linux/init.h>
-#include <linux/timex.h>
-
-#include <asm/hardware.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
-#include <asm/mach/irq.h>
-#include <asm/mach/time.h>
-
-static inline unsigned long get_elapsed(void)
-{
-	return LATCH - *IOP331_TU_TCR0;
-}
-
-static unsigned long iop331_gettimeoffset(void)
-{
-	unsigned long elapsed, usec;
-	u32 tisr1, tisr2;
-
-	/*
-	 * If an interrupt was pending before we read the timer,
-	 * we've already wrapped.  Factor this into the time.
-	 * If an interrupt was pending after we read the timer,
-	 * it may have wrapped between checking the interrupt
-	 * status and reading the timer.  Re-read the timer to
-	 * be sure its value is after the wrap.
-	 */
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr1));
-	elapsed = get_elapsed();
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr2));
-
-	if(tisr1 & 1)
-		elapsed += LATCH;
-	else if (tisr2 & 1)
-		elapsed = LATCH + get_elapsed();
-
-	/*
-	 * Now convert them to usec.
-	 */
-	usec = (unsigned long)(elapsed / (CLOCK_TICK_RATE/1000000));
-
-	return usec;
-}
-
-static irqreturn_t
-iop331_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
-{
-	u32 tisr;
-
-	write_seqlock(&xtime_lock);
-
-	asm volatile("mrc p6, 0, %0, c6, c1, 0" : "=r" (tisr));
-	tisr |= 1;
-	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (tisr));
-
-	timer_tick(regs);
-
-	write_sequnlock(&xtime_lock);
-	return IRQ_HANDLED;
-}
-
-static struct irqaction iop331_timer_irq = {
-	.name		= "IOP331 Timer Tick",
-	.handler	= iop331_timer_interrupt,
-	.flags		= IRQF_DISABLED | IRQF_TIMER,
-};
-
-static void __init iop331_timer_init(void)
-{
-	u32 timer_ctl;
-
-	setup_irq(IRQ_IOP331_TIMER0, &iop331_timer_irq);
-
-	timer_ctl = IOP331_TMR_EN | IOP331_TMR_PRIVILEGED | IOP331_TMR_RELOAD |
-			IOP331_TMR_RATIO_1_1;
-
-	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (LATCH));
-
-	asm volatile("mcr p6, 0, %0, c0, c1, 0"	: : "r" (timer_ctl));
-
-}
-
-struct sys_timer iop331_timer = {
-	.init		= iop331_timer_init,
-	.offset		= iop331_gettimeoffset,
-};
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index bd96b8d55a76..34fe07f0a44e 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -151,10 +151,6 @@
 #define IOP321_FINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007DC)
 
 /* Timers */
-
-#define IOP321_TU_TMR0		(volatile u32 *)IOP321_REG_ADDR(0x000007E0)
-#define IOP321_TU_TMR1		(volatile u32 *)IOP321_REG_ADDR(0x000007E4)
-
 #ifdef CONFIG_ARCH_IQ80321
 #define	IOP321_TICK_RATE	200000000	/* 200 MHz clock */
 #elif defined(CONFIG_ARCH_IQ31244)
@@ -166,23 +162,6 @@
 #define IOP321_TICK_RATE 200000000 /* 33.333333 Mhz crystal */
 #endif
 
-#define IOP321_TMR_TC		0x01
-#define	IOP321_TMR_EN		0x02
-#define IOP321_TMR_RELOAD	0x04
-#define	IOP321_TMR_PRIVILEGED	0x09
-
-#define	IOP321_TMR_RATIO_1_1	0x00
-#define	IOP321_TMR_RATIO_4_1	0x10
-#define	IOP321_TMR_RATIO_8_1	0x20
-#define	IOP321_TMR_RATIO_16_1	0x30
-
-#define IOP321_TU_TCR0    (volatile u32 *)IOP321_REG_ADDR(0x000007E8)
-#define IOP321_TU_TCR1    (volatile u32 *)IOP321_REG_ADDR(0x000007EC)
-#define IOP321_TU_TRR0    (volatile u32 *)IOP321_REG_ADDR(0x000007F0)
-#define IOP321_TU_TRR1    (volatile u32 *)IOP321_REG_ADDR(0x000007F4)
-#define IOP321_TU_TISR    (volatile u32 *)IOP321_REG_ADDR(0x000007F8)
-#define IOP321_TU_WDTCR   (volatile u32 *)IOP321_REG_ADDR(0x000007FC)
-
 /* Application accelerator unit 0x00000800 - 0x000008FF */
 #define IOP321_AAU_ACR     (volatile u32 *)IOP321_REG_ADDR(0x00000800)
 #define IOP321_AAU_ASR     (volatile u32 *)IOP321_REG_ADDR(0x00000804)
diff --git a/include/asm-arm/arch-iop32x/timex.h b/include/asm-arm/arch-iop32x/timex.h
index 08badde2e820..328f37282c3e 100644
--- a/include/asm-arm/arch-iop32x/timex.h
+++ b/include/asm-arm/arch-iop32x/timex.h
@@ -5,4 +5,4 @@
  */
 #include <asm/hardware.h>
 
-#define CLOCK_TICK_RATE IOP321_TICK_RATE
+#define CLOCK_TICK_RATE		(100 * HZ)
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index b301ef8f7f32..4ebcd7197c86 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -137,27 +137,6 @@
 
 
 /* Timers */
-
-#define IOP331_TU_TMR0		(volatile u32 *)IOP331_REG_ADDR(0x000007D0)
-#define IOP331_TU_TMR1		(volatile u32 *)IOP331_REG_ADDR(0x000007D4)
-
-#define IOP331_TMR_TC		0x01
-#define	IOP331_TMR_EN		0x02
-#define IOP331_TMR_RELOAD	0x04
-#define	IOP331_TMR_PRIVILEGED	0x09
-
-#define	IOP331_TMR_RATIO_1_1	0x00
-#define	IOP331_TMR_RATIO_4_1	0x10
-#define	IOP331_TMR_RATIO_8_1	0x20
-#define	IOP331_TMR_RATIO_16_1	0x30
-
-#define IOP331_TU_TCR0    (volatile u32 *)IOP331_REG_ADDR(0x000007D8)
-#define IOP331_TU_TCR1    (volatile u32 *)IOP331_REG_ADDR(0x000007DC)
-#define IOP331_TU_TRR0    (volatile u32 *)IOP331_REG_ADDR(0x000007E0)
-#define IOP331_TU_TRR1    (volatile u32 *)IOP331_REG_ADDR(0x000007E4)
-#define IOP331_TU_TISR    (volatile u32 *)IOP331_REG_ADDR(0x000007E8)
-#define IOP331_TU_WDTCR   (volatile u32 *)IOP331_REG_ADDR(0x000007EC)
-
 #if defined(CONFIG_ARCH_IOP33X)
 #define	IOP331_TICK_RATE	266000000	/* 266 MHz IB clock */
 #endif
diff --git a/include/asm-arm/arch-iop33x/timex.h b/include/asm-arm/arch-iop33x/timex.h
index cc8085fa2a1e..8994322a09f4 100644
--- a/include/asm-arm/arch-iop33x/timex.h
+++ b/include/asm-arm/arch-iop33x/timex.h
@@ -5,4 +5,4 @@
  */
 #include <asm/hardware.h>
 
-#define CLOCK_TICK_RATE IOP331_TICK_RATE
+#define CLOCK_TICK_RATE		(100 * HZ)
-- 
cgit v1.2.3


From 0b29de4a6ac0936f56b974a3c19bd9c24ac5b5d7 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:20:55 +0100
Subject: [ARM] 3824/1: iop3xx: add cp6 enable/disable macros

Add macros to enable and disable access to CP6.  On the iop3xx, enabling
CP6 access unfortunately also enables access to that coprocessor from
unprivileged code, so we need these macros to enable and disable access
to the coprocessor whenever we need to access it.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/hardware/iop3xx.h | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
index b21ea41b149e..98b7cbc405e6 100644
--- a/include/asm-arm/hardware/iop3xx.h
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -132,6 +132,34 @@ unsigned long iop3xx_gettimeoffset(void);
 
 extern struct platform_device iop3xx_i2c0_device;
 extern struct platform_device iop3xx_i2c1_device;
+
+extern inline void iop3xx_cp6_enable(void)
+{
+	u32 temp;
+
+	asm volatile (
+		"mrc	p15, 0, %0, c15, c1, 0\n\t"
+		"orr	%0, %0, #(1 << 6)\n\t"
+		"mcr	p15, 0, %0, c15, c1, 0\n\t"
+		"mrc	p15, 0, %0, c15, c1, 0\n\t"
+		"mov	%0, %0\n\t"
+		"sub	pc, pc, #4\n\t"
+		: "=r" (temp) );
+}
+
+extern inline void iop3xx_cp6_disable(void)
+{
+	u32 temp;
+
+	asm volatile (
+		"mrc	p15, 0, %0, c15, c1, 0\n\t"
+		"bic	%0, %0, #(1 << 6)\n\t"
+		"mcr	p15, 0, %0, c15, c1, 0\n\t"
+		"mrc	p15, 0, %0, c15, c1, 0\n\t"
+		"mov	%0, %0\n\t"
+		"sub	pc, pc, #4\n\t"
+		: "=r" (temp) );
+}
 #endif
 
 
-- 
cgit v1.2.3


From 38ce73ebd74a9a1738b73619557f2397c59ba628 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:21:38 +0100
Subject: [ARM] 3825/1: iop3xx: use cp6 enable/disable macros

Add CP6 enable/disable sequences to the timekeeping code and the IRQ
code.  As a result, we can't depend on CP6 access being enabled when
we enter get_irqnr_and_base anymore, so switch the latter over to
using memory-mapped accesses for now.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop32x/irq.c                | 23 +++++------------------
 arch/arm/mach-iop33x/irq.c                | 27 +++++++++------------------
 arch/arm/plat-iop/time.c                  |  4 ++++
 include/asm-arm/arch-iop32x/entry-macro.S |  3 ++-
 include/asm-arm/arch-iop33x/entry-macro.S |  5 +++--
 5 files changed, 23 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index 76f2d561dbc1..cdd6943ce768 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -27,12 +27,16 @@ static u32 iop321_mask /* = 0 */;
 
 static inline void intctl_write(u32 val)
 {
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
+	iop3xx_cp6_disable();
 }
 
 static inline void intstr_write(u32 val)
 {
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6,0,%0,c4,c0,0"::"r" (val));
+	iop3xx_cp6_disable();
 }
 
 static void
@@ -61,24 +65,7 @@ struct irq_chip ext_chip = {
 
 void __init iop321_init_irq(void)
 {
-	unsigned int i, tmp;
-
-	/* Enable access to coprocessor 6 for dealing with IRQs.
-	 * From RMK:
-	 * Basically, the Intel documentation here is poor.  It appears that
-	 * you need to set the bit to be able to access the coprocessor from
-	 * SVC mode.  Whether that allows access from user space or not is
-	 * unclear.
-	 */
-	asm volatile (
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"orr %0, %0, %1\n\t"
-		"mcr p15, 0, %0, c15, c1, 0\n\t"
-		/* The action is delayed, so we have to do this: */
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"mov %0, %0\n\t"
-		"sub pc, pc, #4"
-		: "=r" (tmp) : "i" (1 << 6) );
+	unsigned int i;
 
 	intctl_write(0);		// disable all interrupts
 	intstr_write(0);		// treat all as IRQ
diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
index bcffc33a5be8..d667439c8573 100644
--- a/arch/arm/mach-iop33x/irq.c
+++ b/arch/arm/mach-iop33x/irq.c
@@ -28,25 +28,33 @@ static u32 iop331_mask1 = 0;
 static inline void intctl_write0(u32 val)
 {
     // INTCTL0
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
+	iop3xx_cp6_disable();
 }
 
 static inline void intctl_write1(u32 val)
 {
     // INTCTL1
+	iop3xx_cp6_enable();
     asm volatile("mcr p6,0,%0,c1,c0,0"::"r" (val));
+	iop3xx_cp6_disable();
 }
 
 static inline void intstr_write0(u32 val)
 {
     // INTSTR0
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6,0,%0,c2,c0,0"::"r" (val));
+	iop3xx_cp6_disable();
 }
 
 static inline void intstr_write1(u32 val)
 {
     // INTSTR1
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6,0,%0,c3,c0,0"::"r" (val));
+	iop3xx_cp6_disable();
 }
 
 static void
@@ -93,24 +101,7 @@ struct irq_chip iop331_irqchip2 = {
 
 void __init iop331_init_irq(void)
 {
-	unsigned int i, tmp;
-
-	/* Enable access to coprocessor 6 for dealing with IRQs.
-	 * From RMK:
-	 * Basically, the Intel documentation here is poor.  It appears that
-	 * you need to set the bit to be able to access the coprocessor from
-	 * SVC mode.  Whether that allows access from user space or not is
-	 * unclear.
-	 */
-	asm volatile (
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"orr %0, %0, %1\n\t"
-		"mcr p15, 0, %0, c15, c1, 0\n\t"
-		/* The action is delayed, so we have to do this: */
-		"mrc p15, 0, %0, c15, c1, 0\n\t"
-		"mov %0, %0\n\t"
-		"sub pc, pc, #4"
-		: "=r" (tmp) : "i" (1 << 6) );
+	unsigned int i;
 
 	intctl_write0(0);		// disable all interrupts
     	intctl_write1(0);
diff --git a/arch/arm/plat-iop/time.c b/arch/arm/plat-iop/time.c
index 5730a0d7ed67..bed20f3669f4 100644
--- a/arch/arm/plat-iop/time.c
+++ b/arch/arm/plat-iop/time.c
@@ -51,7 +51,9 @@ iop3xx_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
 	write_seqlock(&xtime_lock);
 
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6, 0, %0, c6, c1, 0" : : "r" (1));
+	iop3xx_cp6_disable();
 
 	while ((signed long)(next_jiffy_time - *IOP3XX_TU_TCR1)
 							>= ticks_per_jiffy) {
@@ -85,10 +87,12 @@ void __init iop3xx_init_time(unsigned long tick_rate)
 	 * We use timer 0 for our timer interrupt, and timer 1 as
 	 * monotonic counter for tracking missed jiffies.
 	 */
+	iop3xx_cp6_enable();
 	asm volatile("mcr p6, 0, %0, c4, c1, 0" : : "r" (ticks_per_jiffy - 1));
 	asm volatile("mcr p6, 0, %0, c0, c1, 0" : : "r" (timer_ctl));
 	asm volatile("mcr p6, 0, %0, c5, c1, 0" : : "r" (0xffffffff));
 	asm volatile("mcr p6, 0, %0, c1, c1, 0" : : "r" (timer_ctl));
+	iop3xx_cp6_disable();
 
 	setup_irq(IRQ_IOP3XX_TIMER0, &iop3xx_timer_irq);
 }
diff --git a/include/asm-arm/arch-iop32x/entry-macro.S b/include/asm-arm/arch-iop32x/entry-macro.S
index 52d9435c6a34..00038c17317a 100644
--- a/include/asm-arm/arch-iop32x/entry-macro.S
+++ b/include/asm-arm/arch-iop32x/entry-macro.S
@@ -17,7 +17,8 @@
 		 */
 		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
 		mov     \irqnr, #0
-		mrc     p6, 0, \irqstat, c8, c0, 0      @ Read IINTSRC
+		ldr	\base, =IOP3XX_REG_ADDR(0x07D8)
+		ldr	\irqstat, [\base]		@ Read IINTSRC
 		cmp     \irqstat, #0
 		beq     1001f
 		clz     \irqnr, \irqstat
diff --git a/include/asm-arm/arch-iop33x/entry-macro.S b/include/asm-arm/arch-iop33x/entry-macro.S
index 980ec9b1ac83..57f6ea0069e4 100644
--- a/include/asm-arm/arch-iop33x/entry-macro.S
+++ b/include/asm-arm/arch-iop33x/entry-macro.S
@@ -17,10 +17,11 @@
 		 */
 		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
 		mov     \irqnr, #0
-		mrc     p6, 0, \irqstat, c4, c0, 0      @ Read IINTSRC0
+		ldr	\base, =IOP3XX_REG_ADDR(0x7A0)
+		ldr	\irqstat, [\base]		@ Read IINTSRC0
 		cmp     \irqstat, #0
 		bne     1002f
-		mrc     p6, 0, \irqstat, c5, c0, 0      @ Read IINTSRC1
+		ldr	\irqstat, [\base, #4]		@ Read IINTSRC1
 		cmp	\irqstat, #0
 		beq	1001f
 		clz	\irqnr, \irqstat
-- 
cgit v1.2.3


From 610300e8f4f833904096ca1233ffd9dbd73fb11f Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:22:24 +0100
Subject: [ARM] 3826/1: iop3xx: remove IOP3??_IRQ_OFS irq offset

Get rid of the unused IOP3??_IRQ_OFS irq offset define, start IRQ
numbering from zero.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop32x/irq.c                |   6 +-
 arch/arm/mach-iop33x/irq.c                |  10 +--
 include/asm-arm/arch-iop32x/entry-macro.S |   1 -
 include/asm-arm/arch-iop32x/irqs.h        |  68 ++++++++----------
 include/asm-arm/arch-iop33x/entry-macro.S |   1 -
 include/asm-arm/arch-iop33x/irqs.h        | 110 +++++++++++-------------------
 6 files changed, 76 insertions(+), 120 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index cdd6943ce768..ff049e02f5f7 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -43,7 +43,7 @@ static void
 iop321_irq_mask (unsigned int irq)
 {
 
-	iop321_mask &= ~(1 << (irq - IOP321_IRQ_OFS));
+	iop321_mask &= ~(1 << irq);
 
 	intctl_write(iop321_mask);
 }
@@ -51,7 +51,7 @@ iop321_irq_mask (unsigned int irq)
 static void
 iop321_irq_unmask (unsigned int irq)
 {
-	iop321_mask |= (1 << (irq - IOP321_IRQ_OFS));
+	iop321_mask |= (1 << irq);
 
 	intctl_write(iop321_mask);
 }
@@ -73,7 +73,7 @@ void __init iop321_init_irq(void)
 	   machine_is_iq31244()) 	// all interrupts are inputs to chip
 		*IOP3XX_PCIIRSR = 0x0f;
 
-	for(i = IOP321_IRQ_OFS; i < NR_IRQS; i++)
+	for(i = 0; i < NR_IRQS; i++)
 	{
 		set_irq_chip(i, &ext_chip);
 		set_irq_handler(i, do_level_IRQ);
diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
index d667439c8573..675ed3989973 100644
--- a/arch/arm/mach-iop33x/irq.c
+++ b/arch/arm/mach-iop33x/irq.c
@@ -60,28 +60,28 @@ static inline void intstr_write1(u32 val)
 static void
 iop331_irq_mask1 (unsigned int irq)
 {
-        iop331_mask0 &= ~(1 << (irq - IOP331_IRQ_OFS));
+        iop331_mask0 &= ~(1 << irq);
         intctl_write0(iop331_mask0);
 }
 
 static void
 iop331_irq_mask2 (unsigned int irq)
 {
-        iop331_mask1 &= ~(1 << (irq - IOP331_IRQ_OFS - 32));
+        iop331_mask1 &= ~(1 << (irq - 32));
         intctl_write1(iop331_mask1);
 }
 
 static void
 iop331_irq_unmask1(unsigned int irq)
 {
-        iop331_mask0 |= (1 << (irq - IOP331_IRQ_OFS));
+        iop331_mask0 |= (1 << irq);
         intctl_write0(iop331_mask0);
 }
 
 static void
 iop331_irq_unmask2(unsigned int irq)
 {
-        iop331_mask1 |= (1 << (irq - IOP331_IRQ_OFS - 32));
+        iop331_mask1 |= (1 << (irq - 32));
         intctl_write1(iop331_mask1);
 }
 
@@ -110,7 +110,7 @@ void __init iop331_init_irq(void)
 	if(machine_is_iq80331()) 	// all interrupts are inputs to chip
 		*IOP3XX_PCIIRSR = 0x0f;
 
-	for(i = IOP331_IRQ_OFS; i < NR_IRQS; i++)
+	for(i = 0; i < NR_IRQS; i++)
 	{
 		set_irq_chip(i, (i < 32) ? &iop331_irqchip1 : &iop331_irqchip2);
 		set_irq_handler(i, do_level_IRQ);
diff --git a/include/asm-arm/arch-iop32x/entry-macro.S b/include/asm-arm/arch-iop32x/entry-macro.S
index 00038c17317a..c5ec1e23cbea 100644
--- a/include/asm-arm/arch-iop32x/entry-macro.S
+++ b/include/asm-arm/arch-iop32x/entry-macro.S
@@ -24,6 +24,5 @@
 		clz     \irqnr, \irqstat
 		mov     \base, #31
 		subs    \irqnr,\base,\irqnr
-		add     \irqnr,\irqnr,#IRQ_IOP321_DMA0_EOT
 1001:
 		.endm
diff --git a/include/asm-arm/arch-iop32x/irqs.h b/include/asm-arm/arch-iop32x/irqs.h
index 4b0c82711f96..9fefcf3372b1 100644
--- a/include/asm-arm/arch-iop32x/irqs.h
+++ b/include/asm-arm/arch-iop32x/irqs.h
@@ -15,46 +15,36 @@
 /*
  * IOP80321 chipset interrupts
  */
-#define IOP321_IRQ_OFS		0
-#define IOP321_IRQ(x)		(IOP321_IRQ_OFS + (x))
+#define IRQ_IOP321_DMA0_EOT	0
+#define IRQ_IOP321_DMA0_EOC	1
+#define IRQ_IOP321_DMA1_EOT	2
+#define IRQ_IOP321_DMA1_EOC	3
+#define IRQ_IOP321_AA_EOT	6
+#define IRQ_IOP321_AA_EOC	7
+#define IRQ_IOP321_CORE_PMON	8
+#define IRQ_IOP321_TIMER0	9
+#define IRQ_IOP321_TIMER1	10
+#define IRQ_IOP321_I2C_0	11
+#define IRQ_IOP321_I2C_1	12
+#define IRQ_IOP321_MESSAGING	13
+#define IRQ_IOP321_ATU_BIST	14
+#define IRQ_IOP321_PERFMON	15
+#define IRQ_IOP321_CORE_PMU	16
+#define IRQ_IOP321_BIU_ERR	17
+#define IRQ_IOP321_ATU_ERR	18
+#define IRQ_IOP321_MCU_ERR	19
+#define IRQ_IOP321_DMA0_ERR	20
+#define IRQ_IOP321_DMA1_ERR	21
+#define IRQ_IOP321_AA_ERR	23
+#define IRQ_IOP321_MSG_ERR	24
+#define IRQ_IOP321_SSP		25
+#define IRQ_IOP321_XINT0	27
+#define IRQ_IOP321_XINT1	28
+#define IRQ_IOP321_XINT2	29
+#define IRQ_IOP321_XINT3	30
+#define IRQ_IOP321_HPI		31
 
-/*
- * On IRQ or FIQ register
- */
-#define IRQ_IOP321_DMA0_EOT	IOP321_IRQ(0)
-#define IRQ_IOP321_DMA0_EOC	IOP321_IRQ(1)
-#define IRQ_IOP321_DMA1_EOT	IOP321_IRQ(2)
-#define IRQ_IOP321_DMA1_EOC	IOP321_IRQ(3)
-#define IRQ_IOP321_RSVD_4	IOP321_IRQ(4)
-#define IRQ_IOP321_RSVD_5	IOP321_IRQ(5)
-#define IRQ_IOP321_AA_EOT	IOP321_IRQ(6)
-#define IRQ_IOP321_AA_EOC	IOP321_IRQ(7)
-#define IRQ_IOP321_CORE_PMON	IOP321_IRQ(8)
-#define IRQ_IOP321_TIMER0	IOP321_IRQ(9)
-#define IRQ_IOP321_TIMER1	IOP321_IRQ(10)
-#define IRQ_IOP321_I2C_0	IOP321_IRQ(11)
-#define IRQ_IOP321_I2C_1	IOP321_IRQ(12)
-#define IRQ_IOP321_MESSAGING	IOP321_IRQ(13)
-#define IRQ_IOP321_ATU_BIST	IOP321_IRQ(14)
-#define IRQ_IOP321_PERFMON	IOP321_IRQ(15)
-#define IRQ_IOP321_CORE_PMU	IOP321_IRQ(16)
-#define IRQ_IOP321_BIU_ERR	IOP321_IRQ(17)
-#define IRQ_IOP321_ATU_ERR	IOP321_IRQ(18)
-#define IRQ_IOP321_MCU_ERR	IOP321_IRQ(19)
-#define IRQ_IOP321_DMA0_ERR	IOP321_IRQ(20)
-#define IRQ_IOP321_DMA1_ERR	IOP321_IRQ(21)
-#define IRQ_IOP321_RSVD_22	IOP321_IRQ(22)
-#define IRQ_IOP321_AA_ERR	IOP321_IRQ(23)
-#define IRQ_IOP321_MSG_ERR	IOP321_IRQ(24)
-#define IRQ_IOP321_SSP		IOP321_IRQ(25)
-#define IRQ_IOP321_RSVD_26	IOP321_IRQ(26)
-#define IRQ_IOP321_XINT0	IOP321_IRQ(27)
-#define IRQ_IOP321_XINT1	IOP321_IRQ(28)
-#define IRQ_IOP321_XINT2	IOP321_IRQ(29)
-#define IRQ_IOP321_XINT3	IOP321_IRQ(30)
-#define IRQ_IOP321_HPI		IOP321_IRQ(31)
-
-#define NR_IRQS			(IOP321_IRQ(31) + 1)
+#define NR_IRQS			32
 
 
 /*
diff --git a/include/asm-arm/arch-iop33x/entry-macro.S b/include/asm-arm/arch-iop33x/entry-macro.S
index 57f6ea0069e4..425aa7aafa0e 100644
--- a/include/asm-arm/arch-iop33x/entry-macro.S
+++ b/include/asm-arm/arch-iop33x/entry-macro.S
@@ -30,6 +30,5 @@
 		b	1001f
 1002:		clz     \irqnr, \irqstat
 		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
-		add     \irqnr,\irqnr,#IRQ_IOP331_DMA0_EOT
 1001:
 		.endm
diff --git a/include/asm-arm/arch-iop33x/irqs.h b/include/asm-arm/arch-iop33x/irqs.h
index 45856a12815a..2e3ade3b5ff9 100644
--- a/include/asm-arm/arch-iop33x/irqs.h
+++ b/include/asm-arm/arch-iop33x/irqs.h
@@ -15,78 +15,46 @@
 /*
  * IOP80331 chipset interrupts
  */
-#define IOP331_IRQ_OFS		0
-#define IOP331_IRQ(x)		(IOP331_IRQ_OFS + (x))
+#define IRQ_IOP331_DMA0_EOT	0
+#define IRQ_IOP331_DMA0_EOC	1
+#define IRQ_IOP331_DMA1_EOT	2
+#define IRQ_IOP331_DMA1_EOC	3
+#define IRQ_IOP331_AA_EOT	6
+#define IRQ_IOP331_AA_EOC	7
+#define IRQ_IOP331_TIMER0	8
+#define IRQ_IOP331_TIMER1	9
+#define IRQ_IOP331_I2C_0	10
+#define IRQ_IOP331_I2C_1	11
+#define IRQ_IOP331_MSG		12
+#define IRQ_IOP331_MSGIBQ	13
+#define IRQ_IOP331_ATU_BIST	14
+#define IRQ_IOP331_PERFMON	15
+#define IRQ_IOP331_CORE_PMU	16
+#define IRQ_IOP331_XINT0	24
+#define IRQ_IOP331_XINT1	25
+#define IRQ_IOP331_XINT2	26
+#define IRQ_IOP331_XINT3	27
+#define IRQ_IOP331_XINT8	32
+#define IRQ_IOP331_XINT9	33
+#define IRQ_IOP331_XINT10	34
+#define IRQ_IOP331_XINT11	35
+#define IRQ_IOP331_XINT12	36
+#define IRQ_IOP331_XINT13	37
+#define IRQ_IOP331_XINT14	38
+#define IRQ_IOP331_XINT15	39
+#define IRQ_IOP331_UART0	51
+#define IRQ_IOP331_UART1	52
+#define IRQ_IOP331_PBIE		53
+#define IRQ_IOP331_ATU_CRW	54
+#define IRQ_IOP331_ATU_ERR	55
+#define IRQ_IOP331_MCU_ERR	56
+#define IRQ_IOP331_DMA0_ERR	57
+#define IRQ_IOP331_DMA1_ERR	58
+#define IRQ_IOP331_AA_ERR	60
+#define IRQ_IOP331_MSG_ERR	62
+#define IRQ_IOP331_HPI		63
 
-/*
- * On IRQ or FIQ register
- */
-#define IRQ_IOP331_DMA0_EOT	IOP331_IRQ(0)
-#define IRQ_IOP331_DMA0_EOC	IOP331_IRQ(1)
-#define IRQ_IOP331_DMA1_EOT	IOP331_IRQ(2)
-#define IRQ_IOP331_DMA1_EOC	IOP331_IRQ(3)
-#define IRQ_IOP331_RSVD_4	IOP331_IRQ(4)
-#define IRQ_IOP331_RSVD_5	IOP331_IRQ(5)
-#define IRQ_IOP331_AA_EOT	IOP331_IRQ(6)
-#define IRQ_IOP331_AA_EOC	IOP331_IRQ(7)
-#define IRQ_IOP331_TIMER0	IOP331_IRQ(8)
-#define IRQ_IOP331_TIMER1	IOP331_IRQ(9)
-#define IRQ_IOP331_I2C_0	IOP331_IRQ(10)
-#define IRQ_IOP331_I2C_1	IOP331_IRQ(11)
-#define IRQ_IOP331_MSG	    IOP331_IRQ(12)
-#define IRQ_IOP331_MSGIBQ   IOP331_IRQ(13)
-#define IRQ_IOP331_ATU_BIST	IOP331_IRQ(14)
-#define IRQ_IOP331_PERFMON	IOP331_IRQ(15)
-#define IRQ_IOP331_CORE_PMU	IOP331_IRQ(16)
-#define IRQ_IOP331_RSVD_17	IOP331_IRQ(17)
-#define IRQ_IOP331_RSVD_18	IOP331_IRQ(18)
-#define IRQ_IOP331_RSVD_19	IOP331_IRQ(19)
-#define IRQ_IOP331_RSVD_20	IOP331_IRQ(20)
-#define IRQ_IOP331_RSVD_21	IOP331_IRQ(21)
-#define IRQ_IOP331_RSVD_22	IOP331_IRQ(22)
-#define IRQ_IOP331_RSVD_23	IOP331_IRQ(23)
-#define IRQ_IOP331_XINT0	IOP331_IRQ(24)
-#define IRQ_IOP331_XINT1	IOP331_IRQ(25)
-#define IRQ_IOP331_XINT2	IOP331_IRQ(26)
-#define IRQ_IOP331_XINT3	IOP331_IRQ(27)
-#define IRQ_IOP331_RSVD_28  IOP331_IRQ(28)
-#define IRQ_IOP331_RSVD_29  IOP331_IRQ(29)
-#define IRQ_IOP331_RSVD_30  IOP331_IRQ(30)
-#define IRQ_IOP331_RSVD_31  IOP331_IRQ(31)
-#define IRQ_IOP331_XINT8    IOP331_IRQ(32)  // 0
-#define IRQ_IOP331_XINT9    IOP331_IRQ(33)  // 1
-#define IRQ_IOP331_XINT10   IOP331_IRQ(34)  // 2
-#define IRQ_IOP331_XINT11   IOP331_IRQ(35)  // 3
-#define IRQ_IOP331_XINT12   IOP331_IRQ(36)  // 4
-#define IRQ_IOP331_XINT13   IOP331_IRQ(37)  // 5
-#define IRQ_IOP331_XINT14   IOP331_IRQ(38)  // 6
-#define IRQ_IOP331_XINT15   IOP331_IRQ(39)  // 7
-#define IRQ_IOP331_RSVD_40  IOP331_IRQ(40)  // 8
-#define IRQ_IOP331_RSVD_41  IOP331_IRQ(41)  // 9
-#define IRQ_IOP331_RSVD_42  IOP331_IRQ(42)  // 10
-#define IRQ_IOP331_RSVD_43  IOP331_IRQ(43)  // 11
-#define IRQ_IOP331_RSVD_44  IOP331_IRQ(44)  // 12
-#define IRQ_IOP331_RSVD_45  IOP331_IRQ(45)  // 13
-#define IRQ_IOP331_RSVD_46  IOP331_IRQ(46)  // 14
-#define IRQ_IOP331_RSVD_47  IOP331_IRQ(47)  // 15
-#define IRQ_IOP331_RSVD_48  IOP331_IRQ(48)  // 16
-#define IRQ_IOP331_RSVD_49  IOP331_IRQ(49)  // 17
-#define IRQ_IOP331_RSVD_50  IOP331_IRQ(50)  // 18
-#define IRQ_IOP331_UART0    IOP331_IRQ(51)  // 19
-#define IRQ_IOP331_UART1    IOP331_IRQ(52)  // 20
-#define IRQ_IOP331_PBIE     IOP331_IRQ(53)  // 21
-#define IRQ_IOP331_ATU_CRW  IOP331_IRQ(54)  // 22
-#define IRQ_IOP331_ATU_ERR	IOP331_IRQ(55)  // 23
-#define IRQ_IOP331_MCU_ERR	IOP331_IRQ(56)  // 24
-#define IRQ_IOP331_DMA0_ERR	IOP331_IRQ(57)  // 25
-#define IRQ_IOP331_DMA1_ERR	IOP331_IRQ(58)  // 26
-#define IRQ_IOP331_RSVD_59  IOP331_IRQ(59)  // 27
-#define IRQ_IOP331_AA_ERR	IOP331_IRQ(60)  // 28
-#define IRQ_IOP331_RSVD_61  IOP331_IRQ(61)  // 29
-#define IRQ_IOP331_MSG_ERR	IOP331_IRQ(62)  // 30
-#define IRQ_IOP331_HPI		IOP331_IRQ(63)  // 31
-
-#define NR_IRQS			(IOP331_IRQ(63) + 1)
+#define NR_IRQS			64
 
 
 /*
-- 
cgit v1.2.3


From 72edd84a6b2db1a21d1ed07929cae560e276a0a6 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:23:07 +0100
Subject: [ARM] 3827/1: iop3xx: add common gpio module

Implement the gpio_line_{config,get,set} API for iop3xx.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/plat-iop/Makefile           |  2 +-
 arch/arm/plat-iop/gpio.c             | 48 ++++++++++++++++++++++++++++++++++++
 drivers/i2c/busses/i2c-iop3xx.c      | 16 ++++++------
 include/asm-arm/arch-iop32x/iop321.h |  1 +
 include/asm-arm/arch-iop33x/iop331.h |  1 +
 include/asm-arm/hardware/iop3xx.h    | 21 ++++++++++++++++
 6 files changed, 81 insertions(+), 8 deletions(-)
 create mode 100644 arch/arm/plat-iop/gpio.c

(limited to 'include')

diff --git a/arch/arm/plat-iop/Makefile b/arch/arm/plat-iop/Makefile
index d20cdec3a944..23da00b11517 100644
--- a/arch/arm/plat-iop/Makefile
+++ b/arch/arm/plat-iop/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= i2c.o pci.o setup.o time.o
+obj-y			:= gpio.o i2c.o pci.o setup.o time.o
 obj-m			:=
 obj-n			:=
 obj-			:=
diff --git a/arch/arm/plat-iop/gpio.c b/arch/arm/plat-iop/gpio.c
new file mode 100644
index 000000000000..eda436083417
--- /dev/null
+++ b/arch/arm/plat-iop/gpio.c
@@ -0,0 +1,48 @@
+/*
+ * arch/arm/plat-iop/gpio.c
+ * GPIO handling for Intel IOP3xx processors.
+ *
+ * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or (at
+ * your option) any later version.
+ */
+
+#include <linux/device.h>
+#include <asm/hardware/iop3xx.h>
+
+void gpio_line_config(int line, int direction)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	if (direction == GPIO_IN) {
+		*IOP3XX_GPOE |= 1 << line;
+	} else if (direction == GPIO_OUT) {
+		*IOP3XX_GPOE &= ~(1 << line);
+	}
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(gpio_line_config);
+
+int gpio_line_get(int line)
+{
+	return !!(*IOP3XX_GPID & (1 << line));
+}
+EXPORT_SYMBOL(gpio_line_get);
+
+void gpio_line_set(int line, int value)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	if (value == GPIO_LOW) {
+		*IOP3XX_GPOD &= ~(1 << line);
+	} else if (value == GPIO_HIGH) {
+		*IOP3XX_GPOD |= 1 << line;
+	}
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(gpio_line_set);
diff --git a/drivers/i2c/busses/i2c-iop3xx.c b/drivers/i2c/busses/i2c-iop3xx.c
index 035417f5a003..a2d0cc3a7785 100644
--- a/drivers/i2c/busses/i2c-iop3xx.c
+++ b/drivers/i2c/busses/i2c-iop3xx.c
@@ -82,14 +82,16 @@ iop3xx_i2c_enable(struct i2c_algo_iop3xx_data *iop3xx_adap)
 
 	/* 
 	 * Every time unit enable is asserted, GPOD needs to be cleared
-	 * on IOP32X to avoid data corruption on the bus.
+	 * on IOP3XX to avoid data corruption on the bus.
 	 */
-#ifdef CONFIG_ARCH_IOP32X
-#define IOP321_GPOD_I2C0    0x00c0  /* clear these bits to enable ch0 */
-#define IOP321_GPOD_I2C1    0x0030  /* clear these bits to enable ch1 */
-
-	*IOP321_GPOD &= (iop3xx_adap->id == 0) ? ~IOP321_GPOD_I2C0 : 
-		~IOP321_GPOD_I2C1;
+#ifdef CONFIG_PLAT_IOP
+	if (iop3xx_adap->id == 0) {
+		gpio_line_set(IOP3XX_GPIO_LINE(7), GPIO_LOW);
+		gpio_line_set(IOP3XX_GPIO_LINE(6), GPIO_LOW);
+	} else {
+		gpio_line_set(IOP3XX_GPIO_LINE(5), GPIO_LOW);
+		gpio_line_set(IOP3XX_GPIO_LINE(4), GPIO_LOW);
+	}
 #endif
 	/* NB SR bits not same position as CR IE bits :-( */
 	iop3xx_adap->SR_enabled = 
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index 34fe07f0a44e..1e57e0094767 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -216,6 +216,7 @@
  * Peripherals that are shared between the iop32x and iop33x but
  * located at different addresses.
  */
+#define IOP3XX_GPIO_REG(reg)   (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07c0 + (reg))
 #define IOP3XX_TIMER_REG(reg)  (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07e0 + (reg))
 
 #include <asm/hardware/iop3xx.h>
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index 4ebcd7197c86..d12a95aa967a 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -221,6 +221,7 @@
  * Peripherals that are shared between the iop32x and iop33x but
  * located at different addresses.
  */
+#define IOP3XX_GPIO_REG(reg)   (IOP3XX_PERIPHERAL_VIRT_BASE + 0x1780 + (reg))
 #define IOP3XX_TIMER_REG(reg)  (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07d0 + (reg))
 
 #include <asm/hardware/iop3xx.h>
diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
index 98b7cbc405e6..f3c61d041fca 100644
--- a/include/asm-arm/hardware/iop3xx.h
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -15,6 +15,22 @@
 #ifndef __IOP3XX_H
 #define __IOP3XX_H
 
+/*
+ * IOP3XX GPIO handling
+ */
+#define GPIO_IN			0
+#define GPIO_OUT		1
+#define GPIO_LOW		0
+#define GPIO_HIGH		1
+#define IOP3XX_GPIO_LINE(x)	(x)
+
+#ifndef __ASSEMBLY__
+extern void gpio_line_config(int line, int direction);
+extern int  gpio_line_get(int line);
+extern void gpio_line_set(int line, int value);
+#endif
+
+
 /*
  * IOP3XX processor registers
  */
@@ -81,6 +97,11 @@
 #define IOP3XX_PCIXSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01e4)
 #define IOP3XX_PCIIRSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01ec)
 
+/* General Purpose I/O  */
+#define IOP3XX_GPOE		(volatile u32 *)IOP3XX_GPIO_REG(0x0004)
+#define IOP3XX_GPID		(volatile u32 *)IOP3XX_GPIO_REG(0x0008)
+#define IOP3XX_GPOD		(volatile u32 *)IOP3XX_GPIO_REG(0x000c)
+
 /* Timers  */
 #define IOP3XX_TU_TMR0		(volatile u32 *)IOP3XX_TIMER_REG(0x0000)
 #define IOP3XX_TU_TMR1		(volatile u32 *)IOP3XX_TIMER_REG(0x0004)
-- 
cgit v1.2.3


From 7412b10f7967ef4210ed6f793004d23642dc5140 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:24:10 +0100
Subject: [ARM] 3829/1: iop3xx: optimise irq entry macros

Squeeze three instructions out of the iop32x irq demuxer, and nine
out of the iop33x irq demuxer by using the hardware vector generator.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop33x/irq.c                | 16 ++++++++++++++++
 include/asm-arm/arch-iop32x/entry-macro.S | 10 +++-------
 include/asm-arm/arch-iop33x/entry-macro.S | 22 +++++-----------------
 3 files changed, 24 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
index 675ed3989973..3c720551ac12 100644
--- a/arch/arm/mach-iop33x/irq.c
+++ b/arch/arm/mach-iop33x/irq.c
@@ -57,6 +57,20 @@ static inline void intstr_write1(u32 val)
 	iop3xx_cp6_disable();
 }
 
+static inline void intbase_write(u32 val)
+{
+	iop3xx_cp6_enable();
+	asm volatile("mcr p6, 0, %0, c12, c0, 0" : : "r" (val));
+	iop3xx_cp6_disable();
+}
+
+static inline void intsize_write(u32 val)
+{
+	iop3xx_cp6_enable();
+	asm volatile("mcr p6, 0, %0, c13, c0, 0" : : "r" (val));
+	iop3xx_cp6_disable();
+}
+
 static void
 iop331_irq_mask1 (unsigned int irq)
 {
@@ -107,6 +121,8 @@ void __init iop331_init_irq(void)
     	intctl_write1(0);
 	intstr_write0(0);		// treat all as IRQ
     	intstr_write1(0);
+	intbase_write(0);
+	intsize_write(1);
 	if(machine_is_iq80331()) 	// all interrupts are inputs to chip
 		*IOP3XX_PCIIRSR = 0x0f;
 
diff --git a/include/asm-arm/arch-iop32x/entry-macro.S b/include/asm-arm/arch-iop32x/entry-macro.S
index c5ec1e23cbea..3497fef0b890 100644
--- a/include/asm-arm/arch-iop32x/entry-macro.S
+++ b/include/asm-arm/arch-iop32x/entry-macro.S
@@ -16,13 +16,9 @@
 		 * Note: only deal with normal interrupts, not FIQ
 		 */
 		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
-		mov     \irqnr, #0
 		ldr	\base, =IOP3XX_REG_ADDR(0x07D8)
 		ldr	\irqstat, [\base]		@ Read IINTSRC
-		cmp     \irqstat, #0
-		beq     1001f
-		clz     \irqnr, \irqstat
-		mov     \base, #31
-		subs    \irqnr,\base,\irqnr
-1001:
+		cmp	\irqstat, #0
+		clzne	\irqnr, \irqstat
+		rsbne	\irqnr, \irqnr, #31
 		.endm
diff --git a/include/asm-arm/arch-iop33x/entry-macro.S b/include/asm-arm/arch-iop33x/entry-macro.S
index 425aa7aafa0e..4750e98e9b4a 100644
--- a/include/asm-arm/arch-iop33x/entry-macro.S
+++ b/include/asm-arm/arch-iop33x/entry-macro.S
@@ -12,23 +12,11 @@
 		.macro  disable_fiq
 		.endm
 
-		/*
-		 * Note: only deal with normal interrupts, not FIQ
-		 */
 		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
-		mov     \irqnr, #0
-		ldr	\base, =IOP3XX_REG_ADDR(0x7A0)
-		ldr	\irqstat, [\base]		@ Read IINTSRC0
-		cmp     \irqstat, #0
-		bne     1002f
-		ldr	\irqstat, [\base, #4]		@ Read IINTSRC1
+		ldr	\base, =IOP3XX_REG_ADDR(0x07C8)
+		ldr	\irqstat, [\base]		@ Read IINTVEC
 		cmp	\irqstat, #0
-		beq	1001f
-		clz	\irqnr, \irqstat
-		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
-		add	\irqnr,\irqnr,#IRQ_IOP331_XINT8
-		b	1001f
-1002:		clz     \irqnr, \irqstat
-		rsbs    \irqnr,\irqnr,#31   @ recommend by RMK
-1001:
+		ldreq	\irqstat, [\base]		@ erratum 63 workaround
+		adds	\irqnr, \irqstat, #1
+		movne	\irqnr, \irqstat, lsr #2
 		.endm
-- 
cgit v1.2.3


From c680b77efe4542830bb170e1cc40db1c47c569bc Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:24:52 +0100
Subject: [ARM] 3830/1: iop3xx: board support file cleanup

Revamp the iop3xx board support: move the support code for each iop
board type into its own file, start using platform serial and platform
physmap flash devices, switch to a per-board time tick rate, and get
rid of the ARCH_EP80219 and STEPD config options by doing the relevant
checks at run time.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/ep80219_defconfig     |   7 +-
 arch/arm/configs/iq31244_defconfig     |   7 +-
 arch/arm/configs/iq80321_defconfig     |   5 +-
 arch/arm/configs/iq80331_defconfig     |  10 +-
 arch/arm/configs/iq80332_defconfig     |  10 +-
 arch/arm/mach-iop32x/Kconfig           |   7 -
 arch/arm/mach-iop32x/Makefile          |   6 +-
 arch/arm/mach-iop32x/common.c          |  67 --------
 arch/arm/mach-iop32x/iq31244-mm.c      |  46 ------
 arch/arm/mach-iop32x/iq31244-pci.c     |  90 ----------
 arch/arm/mach-iop32x/iq31244.c         | 293 +++++++++++++++++++++++++++++++++
 arch/arm/mach-iop32x/iq80321-mm.c      |  46 ------
 arch/arm/mach-iop32x/iq80321-pci.c     |  84 ----------
 arch/arm/mach-iop32x/iq80321.c         | 193 ++++++++++++++++++++++
 arch/arm/mach-iop32x/setup.c           | 111 -------------
 arch/arm/mach-iop33x/Kconfig           |   6 -
 arch/arm/mach-iop33x/Makefile          |   6 +-
 arch/arm/mach-iop33x/iq80331-pci.c     |  80 ---------
 arch/arm/mach-iop33x/iq80331.c         | 148 +++++++++++++++++
 arch/arm/mach-iop33x/iq80332-pci.c     |  86 ----------
 arch/arm/mach-iop33x/iq80332.c         | 148 +++++++++++++++++
 arch/arm/mach-iop33x/setup.c           | 162 ------------------
 arch/arm/mach-iop33x/uart.c            | 106 ++++++++++++
 include/asm-arm/arch-iop32x/iop321.h   |  12 --
 include/asm-arm/arch-iop32x/iq31244.h  |   7 -
 include/asm-arm/arch-iop32x/iq80321.h  |   7 -
 include/asm-arm/arch-iop32x/irqs.h     |  38 -----
 include/asm-arm/arch-iop33x/hardware.h |   5 +
 include/asm-arm/arch-iop33x/iop331.h   |  10 --
 include/asm-arm/arch-iop33x/iq80331.h  |   7 -
 include/asm-arm/arch-iop33x/iq80332.h  |   7 -
 include/asm-arm/arch-iop33x/irqs.h     |  38 -----
 32 files changed, 911 insertions(+), 944 deletions(-)
 delete mode 100644 arch/arm/mach-iop32x/common.c
 delete mode 100644 arch/arm/mach-iop32x/iq31244-mm.c
 delete mode 100644 arch/arm/mach-iop32x/iq31244-pci.c
 create mode 100644 arch/arm/mach-iop32x/iq31244.c
 delete mode 100644 arch/arm/mach-iop32x/iq80321-mm.c
 delete mode 100644 arch/arm/mach-iop32x/iq80321-pci.c
 create mode 100644 arch/arm/mach-iop32x/iq80321.c
 delete mode 100644 arch/arm/mach-iop32x/setup.c
 delete mode 100644 arch/arm/mach-iop33x/iq80331-pci.c
 create mode 100644 arch/arm/mach-iop33x/iq80331.c
 delete mode 100644 arch/arm/mach-iop33x/iq80332-pci.c
 create mode 100644 arch/arm/mach-iop33x/iq80332.c
 delete mode 100644 arch/arm/mach-iop33x/setup.c
 create mode 100644 arch/arm/mach-iop33x/uart.c

(limited to 'include')

diff --git a/arch/arm/configs/ep80219_defconfig b/arch/arm/configs/ep80219_defconfig
index f91cf8ff359d..c10f3a8afb59 100644
--- a/arch/arm/configs/ep80219_defconfig
+++ b/arch/arm/configs/ep80219_defconfig
@@ -93,7 +93,6 @@ CONFIG_ARCH_IOP32X=y
 CONFIG_ARCH_IQ31244=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
-CONFIG_ARCH_EP80219=y
 
 #
 # IOP3xx Chipset Features
@@ -231,9 +230,9 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xf0000000
-CONFIG_MTD_PHYSMAP_LEN=0x00800000
-CONFIG_MTD_PHYSMAP_BANKWIDTH=2
+CONFIG_MTD_PHYSMAP_START=0x0
+CONFIG_MTD_PHYSMAP_LEN=0x0
+CONFIG_MTD_PHYSMAP_BANKWIDTH=1
 # CONFIG_MTD_ARM_INTEGRATOR is not set
 # CONFIG_MTD_EDB7312 is not set
 
diff --git a/arch/arm/configs/iq31244_defconfig b/arch/arm/configs/iq31244_defconfig
index ce1b1f20b763..0955577243de 100644
--- a/arch/arm/configs/iq31244_defconfig
+++ b/arch/arm/configs/iq31244_defconfig
@@ -94,7 +94,6 @@ CONFIG_ARCH_IOP32X=y
 CONFIG_ARCH_IQ31244=y
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
-# CONFIG_ARCH_EP80219 is not set
 
 #
 # IOP3xx Chipset Features
@@ -232,9 +231,9 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xf0000000
-CONFIG_MTD_PHYSMAP_LEN=0x00800000
-CONFIG_MTD_PHYSMAP_BANKWIDTH=2
+CONFIG_MTD_PHYSMAP_START=0x0
+CONFIG_MTD_PHYSMAP_LEN=0x0
+CONFIG_MTD_PHYSMAP_BANKWIDTH=1
 # CONFIG_MTD_ARM_INTEGRATOR is not set
 # CONFIG_MTD_EDB7312 is not set
 
diff --git a/arch/arm/configs/iq80321_defconfig b/arch/arm/configs/iq80321_defconfig
index f00b0d2159dd..96be4c058546 100644
--- a/arch/arm/configs/iq80321_defconfig
+++ b/arch/arm/configs/iq80321_defconfig
@@ -93,7 +93,6 @@ CONFIG_ARCH_IQ80321=y
 # CONFIG_ARCH_IQ31244 is not set
 # CONFIG_ARCH_IQ80331 is not set
 # CONFIG_MACH_IQ80332 is not set
-# CONFIG_ARCH_EP80219 is not set
 
 #
 # IOP3xx Chipset Features
@@ -231,8 +230,8 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xf0000000
-CONFIG_MTD_PHYSMAP_LEN=0x00800000
+CONFIG_MTD_PHYSMAP_START=0x0
+CONFIG_MTD_PHYSMAP_LEN=0x0
 CONFIG_MTD_PHYSMAP_BANKWIDTH=1
 # CONFIG_MTD_ARM_INTEGRATOR is not set
 # CONFIG_MTD_EDB7312 is not set
diff --git a/arch/arm/configs/iq80331_defconfig b/arch/arm/configs/iq80331_defconfig
index af3a87e469d5..874872ab6bdd 100644
--- a/arch/arm/configs/iq80331_defconfig
+++ b/arch/arm/configs/iq80331_defconfig
@@ -93,12 +93,6 @@ CONFIG_ARCH_IOP33X=y
 # CONFIG_ARCH_IQ31244 is not set
 CONFIG_ARCH_IQ80331=y
 # CONFIG_MACH_IQ80332 is not set
-# CONFIG_ARCH_EP80219 is not set
-
-#
-# IOP3xx Chipset Features
-#
-CONFIG_IOP331_STEPD=y
 
 #
 # Processor Type
@@ -236,8 +230,8 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xc0000000
-CONFIG_MTD_PHYSMAP_LEN=0x00800000
+CONFIG_MTD_PHYSMAP_START=0x0
+CONFIG_MTD_PHYSMAP_LEN=0x0
 CONFIG_MTD_PHYSMAP_BANKWIDTH=1
 # CONFIG_MTD_ARM_INTEGRATOR is not set
 # CONFIG_MTD_EDB7312 is not set
diff --git a/arch/arm/configs/iq80332_defconfig b/arch/arm/configs/iq80332_defconfig
index 931c78755a30..a79895316535 100644
--- a/arch/arm/configs/iq80332_defconfig
+++ b/arch/arm/configs/iq80332_defconfig
@@ -93,12 +93,6 @@ CONFIG_ARCH_IOP33X=y
 # CONFIG_ARCH_IQ31244 is not set
 # CONFIG_ARCH_IQ80331 is not set
 CONFIG_MACH_IQ80332=y
-# CONFIG_ARCH_EP80219 is not set
-
-#
-# IOP3xx Chipset Features
-#
-# CONFIG_IOP331_STEPD is not set
 
 #
 # Processor Type
@@ -236,8 +230,8 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xc0000000
-CONFIG_MTD_PHYSMAP_LEN=0x00800000
+CONFIG_MTD_PHYSMAP_START=0x0
+CONFIG_MTD_PHYSMAP_LEN=0x0
 CONFIG_MTD_PHYSMAP_BANKWIDTH=1
 # CONFIG_MTD_ARM_INTEGRATOR is not set
 # CONFIG_MTD_EDB7312 is not set
diff --git a/arch/arm/mach-iop32x/Kconfig b/arch/arm/mach-iop32x/Kconfig
index 21e3e8c8c2ad..ff8a77a8866e 100644
--- a/arch/arm/mach-iop32x/Kconfig
+++ b/arch/arm/mach-iop32x/Kconfig
@@ -16,13 +16,6 @@ config ARCH_IQ31244
 	  Say Y here if you want to run your kernel on the Intel IQ31244
 	  evaluation kit for the IOP321 chipset.
 
-config ARCH_EP80219
-	bool "Enable support for EP80219"
-	select ARCH_IQ31244
-	help
-	  Say Y here if you want to run your kernel on the Intel EP80219
-	  evaluation kit for the Intel 80219 chipset (a IOP321 variant).
-
 endmenu
 
 endif
diff --git a/arch/arm/mach-iop32x/Makefile b/arch/arm/mach-iop32x/Makefile
index ef561db20c9b..af1747ae392d 100644
--- a/arch/arm/mach-iop32x/Makefile
+++ b/arch/arm/mach-iop32x/Makefile
@@ -2,10 +2,10 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= common.o setup.o irq.o
+obj-y			:= irq.o
 obj-m			:=
 obj-n			:=
 obj-			:=
 
-obj-$(CONFIG_ARCH_IQ80321) += iq80321-mm.o iq80321-pci.o
-obj-$(CONFIG_ARCH_IQ31244) += iq31244-mm.o iq31244-pci.o
+obj-$(CONFIG_ARCH_IQ80321) += iq80321.o
+obj-$(CONFIG_ARCH_IQ31244) += iq31244.o
diff --git a/arch/arm/mach-iop32x/common.c b/arch/arm/mach-iop32x/common.c
deleted file mode 100644
index 9a17a081327d..000000000000
--- a/arch/arm/mach-iop32x/common.c
+++ /dev/null
@@ -1,67 +0,0 @@
-/*
- * arch/arm/mach-iop32x/common.c
- *
- * Common routines shared across all IOP3xx implementations
- *
- * Author: Deepak Saxena <dsaxena@mvista.com>
- *
- * Copyright 2003 (c) MontaVista, Software, Inc.
- *
- * This file is licensed under  the terms of the GNU General Public
- * License version 2. This program is licensed "as is" without any
- * warranty of any kind, whether express or implied.
- */
-
-#include <linux/delay.h>
-#include <asm/hardware.h>
-#include <asm/hardware/iop3xx.h>
-
-#ifdef CONFIG_ARCH_EP80219
-#include <linux/kernel.h>
-/*
- * Default power-off for EP80219
- */
-
-static inline void ep80219_send_to_pic(__u8 c) {
-}
-
-void ep80219_power_off(void)
-{
-	/*
-     * This function will send a SHUTDOWN_COMPLETE message to the PIC controller
-     * over I2C.  We are not using the i2c subsystem since we are going to power
-     * off and it may be removed
-     */
-
-	/* Send the Address byte w/ the start condition */
-	*IOP3XX_IDBR1 = 0x60;
-	*IOP3XX_ICR1 = 0xE9;
-    mdelay(1);
-
-	/* Send the START_MSG byte w/ no start or stop condition */
-	*IOP3XX_IDBR1 = 0x0F;
-	*IOP3XX_ICR1 = 0xE8;
-    mdelay(1);
-
-	/* Send the SHUTDOWN_COMPLETE Message ID byte w/ no start or stop condition */
-	*IOP3XX_IDBR1 = 0x03;
-	*IOP3XX_ICR1 = 0xE8;
-    mdelay(1);
-
-	/* Send an ignored byte w/ stop condition */
-	*IOP3XX_IDBR1 = 0x00;
-	*IOP3XX_ICR1 = 0xEA;
-
-	while (1) ;
-}
-
-#include <linux/init.h>
-#include <linux/pm.h>
-
-static int __init ep80219_init(void)
-{
-	pm_power_off = ep80219_power_off;
-	return 0;
-}
-arch_initcall(ep80219_init);
-#endif
diff --git a/arch/arm/mach-iop32x/iq31244-mm.c b/arch/arm/mach-iop32x/iq31244-mm.c
deleted file mode 100644
index fba22d5d908f..000000000000
--- a/arch/arm/mach-iop32x/iq31244-mm.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- * linux/arch/arm/mach-iop32x/iq31244-mm.c
- *
- * Low level memory initialization for iq80321 platform
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-#include <asm/hardware/iop3xx.h>
-
-
-/*
- * IQ80321 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-static struct map_desc iq31244_io_desc[] __initdata = {
-	{	/* on-board devices */
-		.virtual	= IQ31244_UART,
-		.pfn		= __phys_to_pfn(IQ31244_UART),
-		.length		= 0x00100000,
-		.type		= MT_DEVICE
-	}
-};
-
-void __init iq31244_map_io(void)
-{
-	iop3xx_map_io();
-
-	iotable_init(iq31244_io_desc, ARRAY_SIZE(iq31244_io_desc));
-}
diff --git a/arch/arm/mach-iop32x/iq31244-pci.c b/arch/arm/mach-iop32x/iq31244-pci.c
deleted file mode 100644
index 605b79553747..000000000000
--- a/arch/arm/mach-iop32x/iq31244-pci.c
+++ /dev/null
@@ -1,90 +0,0 @@
-/*
- * arch/arm/mach-iop32x/iq31244-pci.c
- *
- * PCI support for the Intel IQ31244 reference board
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ31244_INTA
-#define INTB	IRQ_IQ31244_INTB
-#define INTC	IRQ_IQ31244_INTC
-#define INTD	IRQ_IQ31244_INTD
-
-#define INTE	IRQ_IQ31244_I82546
-
-static inline int __init
-iq31244_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][4] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-#ifdef CONFIG_ARCH_EP80219
-		{INTB, INTB, INTB, INTB}, /* CFlash */
-		{INTE, INTE, INTE, INTE}, /* 82551 Pro 100 */
-		{INTD, INTD, INTD, INTD}, /* PCI-X Slot */
-		{INTC, INTC, INTC, INTC}, /* SATA   */
-#else
-		{INTB, INTB, INTB, INTB}, /* CFlash */
-		{INTC, INTC, INTC, INTC}, /* SATA   */
-		{INTD, INTD, INTD, INTD}, /* PCI-X Slot */
-		{INTE, INTE, INTE, INTE}, /* 82546 GigE */
-#endif // CONFIG_ARCH_EP80219
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-	return PCI_IRQ_TABLE_LOOKUP(0, 7);
-}
-
-static struct hw_pci iq31244_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iop3xx_pci_setup,
-	.scan		= iop3xx_pci_scan_bus,
-	.preinit	= iop3xx_pci_preinit,
-	.map_irq	= iq31244_map_irq
-};
-
-static int __init iq31244_pci_init(void)
-{
-	if (machine_is_iq31244())
-		pci_common_init(&iq31244_pci);
-	return 0;
-}
-
-subsys_initcall(iq31244_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop32x/iq31244.c b/arch/arm/mach-iop32x/iq31244.c
new file mode 100644
index 000000000000..88b77d32b0ac
--- /dev/null
+++ b/arch/arm/mach-iop32x/iq31244.c
@@ -0,0 +1,293 @@
+/*
+ * arch/arm/mach-iop32x/iq31244.c
+ *
+ * Board support code for the Intel EP80219 and IQ31244 platforms.
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright 2003 (c) MontaVista, Software, Inc.
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/pm.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/serial_core.h>
+#include <linux/serial_8250.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/pci.h>
+#include <asm/mach/time.h>
+#include <asm/mach-types.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+
+/*
+ * The EP80219 and IQ31244 use the same machine ID.  To find out
+ * which of the two we're running on, we look at the processor ID.
+ */
+static int is_80219(void)
+{
+	extern int processor_id;
+	return !!((processor_id & 0xffffffe0) == 0x69052e20);
+}
+
+
+/*
+ * EP80219/IQ31244 timer tick configuration.
+ */
+static void __init iq31244_timer_init(void)
+{
+	if (is_80219()) {
+		/* 33.333 MHz crystal.  */
+		iop3xx_init_time(200000000);
+	} else {
+		/* 33.000 MHz crystal.  */
+		iop3xx_init_time(198000000);
+	}
+}
+
+static struct sys_timer iq31244_timer = {
+	.init		= iq31244_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
+
+/*
+ * IQ31244 I/O.
+ */
+static struct map_desc iq31244_io_desc[] __initdata = {
+	{	/* on-board devices */
+		.virtual	= IQ31244_UART,
+		.pfn		= __phys_to_pfn(IQ31244_UART),
+		.length		= 0x00100000,
+		.type		= MT_DEVICE,
+	},
+};
+
+void __init iq31244_map_io(void)
+{
+	iop3xx_map_io();
+	iotable_init(iq31244_io_desc, ARRAY_SIZE(iq31244_io_desc));
+}
+
+
+/*
+ * EP80219/IQ31244 PCI.
+ */
+static inline int __init
+ep80219_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+
+	if (slot == 0) {
+		/* CFlash */
+		irq = IRQ_IOP321_XINT1;
+	} else if (slot == 1) {
+		/* 82551 Pro 100 */
+		irq = IRQ_IOP321_XINT0;
+	} else if (slot == 2) {
+		/* PCI-X Slot */
+		irq = IRQ_IOP321_XINT3;
+	} else if (slot == 3) {
+		/* SATA */
+		irq = IRQ_IOP321_XINT2;
+	} else {
+		printk(KERN_ERR "ep80219_pci_map_irq() called for unknown "
+			"device PCI:%d:%d:%d\n", dev->bus->number,
+			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+		irq = -1;
+	}
+
+	return irq;
+}
+
+static struct hw_pci ep80219_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= ep80219_pci_map_irq,
+};
+
+static inline int __init
+iq31244_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+
+	if (slot == 0) {
+		/* CFlash */
+		irq = IRQ_IOP321_XINT1;
+	} else if (slot == 1) {
+		/* SATA */
+		irq = IRQ_IOP321_XINT2;
+	} else if (slot == 2) {
+		/* PCI-X Slot */
+		irq = IRQ_IOP321_XINT3;
+	} else if (slot == 3) {
+		/* 82546 GigE */
+		irq = IRQ_IOP321_XINT0;
+	} else {
+		printk(KERN_ERR "iq31244_pci_map_irq() called for unknown "
+			"device PCI:%d:%d:%d\n", dev->bus->number,
+			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+		irq = -1;
+	}
+
+	return irq;
+}
+
+static struct hw_pci iq31244_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= iq31244_pci_map_irq,
+};
+
+static int __init iq31244_pci_init(void)
+{
+	if (machine_is_iq31244()) {
+		if (is_80219()) {
+			pci_common_init(&ep80219_pci);
+		} else {
+			pci_common_init(&iq31244_pci);
+		}
+	}
+
+	return 0;
+}
+
+subsys_initcall(iq31244_pci_init);
+
+
+/*
+ * IQ31244 machine initialisation.
+ */
+static struct physmap_flash_data iq31244_flash_data = {
+	.width		= 2,
+};
+
+static struct resource iq31244_flash_resource = {
+	.start		= 0xf0000000,
+	.end		= 0xf07fffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device iq31244_flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &iq31244_flash_data,
+	},
+	.num_resources	= 1,
+	.resource	= &iq31244_flash_resource,
+};
+
+static struct plat_serial8250_port iq31244_serial_port[] = {
+	{
+		.mapbase	= IQ31244_UART,
+		.membase	= (char *)IQ31244_UART,
+		.irq		= IRQ_IOP321_XINT1,
+		.flags		= UPF_SKIP_TEST,
+		.iotype		= UPIO_MEM,
+		.regshift	= 0,
+		.uartclk	= 1843200,
+	},
+	{ },
+};
+
+static struct resource iq31244_uart_resource = {
+	.start		= IQ31244_UART,
+	.end		= IQ31244_UART + 7,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device iq31244_serial_device = {
+	.name		= "serial8250",
+	.id		= PLAT8250_DEV_PLATFORM,
+	.dev		= {
+		.platform_data		= iq31244_serial_port,
+	},
+	.num_resources	= 1,
+	.resource	= &iq31244_uart_resource,
+};
+
+/*
+ * This function will send a SHUTDOWN_COMPLETE message to the PIC
+ * controller over I2C.  We are not using the i2c subsystem since
+ * we are going to power off and it may be removed
+ */
+void ep80219_power_off(void)
+{
+	/*
+	 * Send the Address byte w/ the start condition
+	 */
+	*IOP3XX_IDBR1 = 0x60;
+	*IOP3XX_ICR1 = 0xE9;
+	mdelay(1);
+
+	/*
+	 * Send the START_MSG byte w/ no start or stop condition
+	 */
+	*IOP3XX_IDBR1 = 0x0F;
+	*IOP3XX_ICR1 = 0xE8;
+	mdelay(1);
+
+	/*
+	 * Send the SHUTDOWN_COMPLETE Message ID byte w/ no start or
+	 * stop condition
+	 */
+	*IOP3XX_IDBR1 = 0x03;
+	*IOP3XX_ICR1 = 0xE8;
+	mdelay(1);
+
+	/*
+	 * Send an ignored byte w/ stop condition
+	 */
+	*IOP3XX_IDBR1 = 0x00;
+	*IOP3XX_ICR1 = 0xEA;
+
+	while (1)
+		;
+}
+
+static void __init iq31244_init_machine(void)
+{
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
+	platform_device_register(&iq31244_flash_device);
+	platform_device_register(&iq31244_serial_device);
+
+	if (is_80219())
+		pm_power_off = ep80219_power_off;
+}
+
+MACHINE_START(IQ31244, "Intel IQ31244")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= IQ31244_UART,
+	.io_pg_offst	= ((IQ31244_UART) >> 18) & 0xfffc,
+	.boot_params	= 0xa0000100,
+	.map_io		= iq31244_map_io,
+	.init_irq	= iop321_init_irq,
+	.timer		= &iq31244_timer,
+	.init_machine	= iq31244_init_machine,
+MACHINE_END
diff --git a/arch/arm/mach-iop32x/iq80321-mm.c b/arch/arm/mach-iop32x/iq80321-mm.c
deleted file mode 100644
index b6a3079ad29e..000000000000
--- a/arch/arm/mach-iop32x/iq80321-mm.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- * linux/arch/arm/mach-iop32x/iq80321-mm.c
- *
- * Low level memory initialization for iq80321 platform
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <linux/mm.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-
-#include <asm/mach/map.h>
-#include <asm/hardware/iop3xx.h>
-
-
-/*
- * IQ80321 specific IO mappings
- *
- * We use RedBoot's setup for the onboard devices.
- */
-static struct map_desc iq80321_io_desc[] __initdata = {
- 	{	/* on-board devices */
-		.virtual	= IQ80321_UART,
-		.pfn		= __phys_to_pfn(IQ80321_UART),
-		.length		= 0x00100000,
-		.type		= MT_DEVICE
-	}
-};
-
-void __init iq80321_map_io(void)
-{
-	iop3xx_map_io();
-
-	iotable_init(iq80321_io_desc, ARRAY_SIZE(iq80321_io_desc));
-}
diff --git a/arch/arm/mach-iop32x/iq80321-pci.c b/arch/arm/mach-iop32x/iq80321-pci.c
deleted file mode 100644
index cedc37b968b7..000000000000
--- a/arch/arm/mach-iop32x/iq80321-pci.c
+++ /dev/null
@@ -1,84 +0,0 @@
-/*
- * arch/arm/mach-iop32x/iq80321-pci.c
- *
- * PCI support for the Intel IQ80321 reference board
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ80321_INTA
-#define INTB	IRQ_IQ80321_INTB
-#define INTC	IRQ_IQ80321_INTC
-#define INTD	IRQ_IQ80321_INTD
-
-#define INTE	IRQ_IQ80321_I82544
-
-static inline int __init
-iq80321_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][4] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-		{INTE, INTE, INTE, INTE}, /* Gig-E */
-		{-1, -1, -1, -1}, 	  /* Unused */
-		{INTC, INTD, INTA, INTB}, /* PCI-X Slot */
-		{-1, -1, -1, -1},
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-//	return PCI_IRQ_TABLE_LOOKUP(4, 7);
-	return pci_irq_table[idsel%4][pin-1];
-}
-
-static struct hw_pci iq80321_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iop3xx_pci_setup,
-	.scan		= iop3xx_pci_scan_bus,
-	.preinit	= iop3xx_pci_preinit,
-	.map_irq	= iq80321_map_irq
-};
-
-static int __init iq80321_pci_init(void)
-{
-	if (machine_is_iq80321())
-		pci_common_init(&iq80321_pci);
-	return 0;
-}
-
-subsys_initcall(iq80321_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop32x/iq80321.c b/arch/arm/mach-iop32x/iq80321.c
new file mode 100644
index 000000000000..3c9b86271759
--- /dev/null
+++ b/arch/arm/mach-iop32x/iq80321.c
@@ -0,0 +1,193 @@
+/*
+ * arch/arm/mach-iop32x/iq80321.c
+ *
+ * Board support code for the Intel IQ80321 platform.
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/serial_core.h>
+#include <linux/serial_8250.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/pci.h>
+#include <asm/mach/time.h>
+#include <asm/mach-types.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+/*
+ * IQ80321 timer tick configuration.
+ */
+static void __init iq80321_timer_init(void)
+{
+	/* 33.333 MHz crystal.  */
+	iop3xx_init_time(200000000);
+}
+
+static struct sys_timer iq80321_timer = {
+	.init		= iq80321_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
+
+/*
+ * IQ80321 I/O.
+ */
+static struct map_desc iq80321_io_desc[] __initdata = {
+ 	{	/* on-board devices */
+		.virtual	= IQ80321_UART,
+		.pfn		= __phys_to_pfn(IQ80321_UART),
+		.length		= 0x00100000,
+		.type		= MT_DEVICE,
+	},
+};
+
+void __init iq80321_map_io(void)
+{
+	iop3xx_map_io();
+	iotable_init(iq80321_io_desc, ARRAY_SIZE(iq80321_io_desc));
+}
+
+
+/*
+ * IQ80321 PCI.
+ */
+static inline int __init
+iq80321_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+
+	if ((slot == 2 || slot == 6) && pin == 1) {
+		/* PCI-X Slot INTA */
+		irq = IRQ_IOP321_XINT2;
+	} else if ((slot == 2 || slot == 6) && pin == 2) {
+		/* PCI-X Slot INTA */
+		irq = IRQ_IOP321_XINT3;
+	} else if ((slot == 2 || slot == 6) && pin == 3) {
+		/* PCI-X Slot INTA */
+		irq = IRQ_IOP321_XINT0;
+	} else if ((slot == 2 || slot == 6) && pin == 4) {
+		/* PCI-X Slot INTA */
+		irq = IRQ_IOP321_XINT1;
+	} else if (slot == 4 || slot == 8) {
+		/* Gig-E */
+		irq = IRQ_IOP321_XINT0;
+	} else {
+		printk(KERN_ERR "iq80321_pci_map_irq() called for unknown "
+			"device PCI:%d:%d:%d\n", dev->bus->number,
+			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+		irq = -1;
+	}
+
+	return irq;
+}
+
+static struct hw_pci iq80321_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= iq80321_pci_map_irq,
+};
+
+static int __init iq80321_pci_init(void)
+{
+	if (machine_is_iq80321())
+		pci_common_init(&iq80321_pci);
+
+	return 0;
+}
+
+subsys_initcall(iq80321_pci_init);
+
+
+/*
+ * IQ80321 machine initialisation.
+ */
+static struct physmap_flash_data iq80321_flash_data = {
+	.width		= 1,
+};
+
+static struct resource iq80321_flash_resource = {
+	.start		= 0xf0000000,
+	.end		= 0xf07fffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device iq80321_flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &iq80321_flash_data,
+	},
+	.num_resources	= 1,
+	.resource	= &iq80321_flash_resource,
+};
+
+static struct plat_serial8250_port iq80321_serial_port[] = {
+	{
+		.mapbase	= IQ80321_UART,
+		.membase	= (char *)IQ80321_UART,
+		.irq		= IRQ_IOP321_XINT1,
+		.flags		= UPF_SKIP_TEST,
+		.iotype		= UPIO_MEM,
+		.regshift	= 0,
+		.uartclk	= 1843200,
+	},
+	{ },
+};
+
+static struct resource iq80321_uart_resource = {
+	.start		= IQ80321_UART,
+	.end		= IQ80321_UART + 7,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device iq80321_serial_device = {
+	.name		= "serial8250",
+	.id		= PLAT8250_DEV_PLATFORM,
+	.dev		= {
+		.platform_data		= iq80321_serial_port,
+	},
+	.num_resources	= 1,
+	.resource	= &iq80321_uart_resource,
+};
+
+static void __init iq80321_init_machine(void)
+{
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
+	platform_device_register(&iq80321_flash_device);
+	platform_device_register(&iq80321_serial_device);
+}
+
+MACHINE_START(IQ80321, "Intel IQ80321")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= IQ80321_UART,
+	.io_pg_offst	= ((IQ80321_UART) >> 18) & 0xfffc,
+	.boot_params	= 0xa0000100,
+	.map_io		= iq80321_map_io,
+	.init_irq	= iop321_init_irq,
+	.timer		= &iq80321_timer,
+	.init_machine	= iq80321_init_machine,
+MACHINE_END
diff --git a/arch/arm/mach-iop32x/setup.c b/arch/arm/mach-iop32x/setup.c
deleted file mode 100644
index 68de247a4cca..000000000000
--- a/arch/arm/mach-iop32x/setup.c
+++ /dev/null
@@ -1,111 +0,0 @@
-/*
- * linux/arch/arm/mach-iop32x/setup.c
- *
- * Author: Nicolas Pitre <nico@cam.org>
- * Copyright (C) 2001 MontaVista Software, Inc.
- * Copyright (C) 2004 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/major.h>
-#include <linux/fs.h>
-#include <linux/platform_device.h>
-#include <linux/serial.h>
-#include <linux/tty.h>
-#include <linux/serial_core.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/mach/map.h>
-#include <asm/setup.h>
-#include <asm/system.h>
-#include <asm/memory.h>
-#include <asm/hardware.h>
-#include <asm/mach-types.h>
-#include <asm/mach/arch.h>
-#include <asm/mach/time.h>
-#include <asm/hardware/iop3xx.h>
-
-#define IOP321_UART_XTAL 1843200
-
-#ifdef CONFIG_ARCH_IQ80321
-#define UARTBASE IQ80321_UART
-#define IRQ_UART IRQ_IQ80321_UART
-#endif
-
-#ifdef CONFIG_ARCH_IQ31244
-#define UARTBASE IQ31244_UART
-#define IRQ_UART IRQ_IQ31244_UART
-#endif
-
-static struct uart_port iop321_serial_ports[] = {
-	{
-		.membase	= (char*)(UARTBASE),
-		.mapbase	= (UARTBASE),
-		.irq		= IRQ_UART,
-		.flags		= UPF_SKIP_TEST,
-		.iotype		= UPIO_MEM,
-		.regshift	= 0,
-		.uartclk	= IOP321_UART_XTAL,
-		.line		= 0,
-		.type		= PORT_16550A,
-		.fifosize	= 16
-	}
-};
-
-void __init iop32x_init(void)
-{
-	platform_device_register(&iop3xx_i2c0_device);
-	platform_device_register(&iop3xx_i2c1_device);
-	early_serial_setup(&iop321_serial_ports[0]);
-}
-
-#ifdef CONFIG_ARCH_IQ80321
-extern void iq80321_map_io(void);
-#endif
-
-#ifdef CONFIG_ARCH_IQ31244
-extern void iq31244_map_io(void);
-#endif
-
-static void __init iop3xx_timer_init(void)
-{
-	iop3xx_init_time(IOP321_TICK_RATE);
-}
-
-struct sys_timer iop321_timer = {
-	.init		= iop3xx_timer_init,
-	.offset		= iop3xx_gettimeoffset,
-};
-
-#if defined(CONFIG_ARCH_IQ80321)
-MACHINE_START(IQ80321, "Intel IQ80321")
-	/* Maintainer: Intel Corporation */
-	.phys_io	= IQ80321_UART,
-	.io_pg_offst	= ((IQ80321_UART) >> 18) & 0xfffc,
-	.map_io		= iq80321_map_io,
-	.init_irq	= iop321_init_irq,
-	.timer		= &iop321_timer,
-	.boot_params	= 0xa0000100,
-	.init_machine	= iop32x_init,
-MACHINE_END
-#elif defined(CONFIG_ARCH_IQ31244)
-MACHINE_START(IQ31244, "Intel IQ31244")
-	/* Maintainer: Intel Corp. */
-	.phys_io	= IQ31244_UART,
-	.io_pg_offst	= ((IQ31244_UART) >> 18) & 0xfffc,
-	.map_io		= iq31244_map_io,
-	.init_irq	= iop321_init_irq,
-	.timer		= &iop321_timer,
-	.boot_params	= 0xa0000100,
-	.init_machine	= iop32x_init,
-MACHINE_END
-#else
-#error No machine descriptor defined for this IOP3XX implementation
-#endif
diff --git a/arch/arm/mach-iop33x/Kconfig b/arch/arm/mach-iop33x/Kconfig
index 410df546e954..9aa016bb18f9 100644
--- a/arch/arm/mach-iop33x/Kconfig
+++ b/arch/arm/mach-iop33x/Kconfig
@@ -16,12 +16,6 @@ config MACH_IQ80332
 	  Say Y here if you want to run your kernel on the Intel IQ80332
 	  evaluation kit for the IOP332 chipset.
 
-config IOP331_STEPD
-	bool "Chip stepping D of the IOP80331 processor or IOP80333"
-	help
-	  Say Y here if you have StepD of the IOP80331 or IOP8033
-	  based platforms.
-
 endmenu
 
 endif
diff --git a/arch/arm/mach-iop33x/Makefile b/arch/arm/mach-iop33x/Makefile
index f825cee57d9f..90081d8c9d16 100644
--- a/arch/arm/mach-iop33x/Makefile
+++ b/arch/arm/mach-iop33x/Makefile
@@ -2,10 +2,10 @@
 # Makefile for the linux kernel.
 #
 
-obj-y			:= setup.o irq.o
+obj-y			:= irq.o uart.o
 obj-m			:=
 obj-n			:=
 obj-			:=
 
-obj-$(CONFIG_ARCH_IQ80331) += iq80331-pci.o
-obj-$(CONFIG_MACH_IQ80332) += iq80332-pci.o
+obj-$(CONFIG_ARCH_IQ80331) += iq80331.o
+obj-$(CONFIG_MACH_IQ80332) += iq80332.o
diff --git a/arch/arm/mach-iop33x/iq80331-pci.c b/arch/arm/mach-iop33x/iq80331-pci.c
deleted file mode 100644
index 8b0bed5e2f91..000000000000
--- a/arch/arm/mach-iop33x/iq80331-pci.c
+++ /dev/null
@@ -1,80 +0,0 @@
-/*
- * arch/arm/mach-iop33x/iq80331-pci.c
- *
- * PCI support for the Intel IQ80331 reference board
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2003, 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ80331_INTA
-#define INTB	IRQ_IQ80331_INTB
-#define INTC	IRQ_IQ80331_INTC
-#define INTD	IRQ_IQ80331_INTD
-
-//#define INTE	IRQ_IQ80331_I82544
-
-static inline int __init
-iq80331_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][4] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-		{INTB, INTC, INTD, INTA}, /* PCI-X Slot */
-		{INTC, INTC, INTC, INTC}, /* GigE  */
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-	return PCI_IRQ_TABLE_LOOKUP(1, 7);
-}
-
-static struct hw_pci iq80331_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iop3xx_pci_setup,
-	.scan		= iop3xx_pci_scan_bus,
-	.preinit	= iop3xx_pci_preinit,
-	.map_irq	= iq80331_map_irq
-};
-
-static int __init iq80331_pci_init(void)
-{
-	if (machine_is_iq80331())
-		pci_common_init(&iq80331_pci);
-	return 0;
-}
-
-subsys_initcall(iq80331_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop33x/iq80331.c b/arch/arm/mach-iop33x/iq80331.c
new file mode 100644
index 000000000000..6b8475da3df6
--- /dev/null
+++ b/arch/arm/mach-iop33x/iq80331.c
@@ -0,0 +1,148 @@
+/*
+ * arch/arm/mach-iop33x/iq80331.c
+ *
+ * Board support code for the Intel IQ80331 platform.
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2003 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/serial_core.h>
+#include <linux/serial_8250.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/pci.h>
+#include <asm/mach/time.h>
+#include <asm/mach-types.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+/*
+ * IQ80331 timer tick configuration.
+ */
+static void __init iq80331_timer_init(void)
+{
+	/* D-Step parts run at a higher internal bus frequency */
+	if (*IOP3XX_ATURID >= 0xa)
+		iop3xx_init_time(333000000);
+	else
+		iop3xx_init_time(266000000);
+}
+
+static struct sys_timer iq80331_timer = {
+	.init		= iq80331_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
+
+/*
+ * IQ80331 PCI.
+ */
+static inline int __init
+iq80331_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+
+	if (slot == 1 && pin == 1) {
+		/* PCI-X Slot INTA */
+		irq = IRQ_IOP331_XINT1;
+	} else if (slot == 1 && pin == 2) {
+		/* PCI-X Slot INTB */
+		irq = IRQ_IOP331_XINT2;
+	} else if (slot == 1 && pin == 3) {
+		/* PCI-X Slot INTC */
+		irq = IRQ_IOP331_XINT3;
+	} else if (slot == 1 && pin == 4) {
+		/* PCI-X Slot INTD */
+		irq = IRQ_IOP331_XINT0;
+	} else if (slot == 2) {
+		/* GigE */
+		irq = IRQ_IOP331_XINT2;
+	} else {
+		printk(KERN_ERR "iq80331_pci_map_irq() called for unknown "
+			"device PCI:%d:%d:%d\n", dev->bus->number,
+			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+		irq = -1;
+	}
+
+	return irq;
+}
+
+static struct hw_pci iq80331_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= iq80331_pci_map_irq,
+};
+
+static int __init iq80331_pci_init(void)
+{
+	if (machine_is_iq80331())
+		pci_common_init(&iq80331_pci);
+
+	return 0;
+}
+
+subsys_initcall(iq80331_pci_init);
+
+
+/*
+ * IQ80331 machine initialisation.
+ */
+static struct physmap_flash_data iq80331_flash_data = {
+	.width		= 1,
+};
+
+static struct resource iq80331_flash_resource = {
+	.start		= 0xc0000000,
+	.end		= 0xc07fffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device iq80331_flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &iq80331_flash_data,
+	},
+	.num_resources	= 1,
+	.resource	= &iq80331_flash_resource,
+};
+
+static void __init iq80331_init_machine(void)
+{
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
+	platform_device_register(&iop33x_uart0_device);
+	platform_device_register(&iop33x_uart1_device);
+	platform_device_register(&iq80331_flash_device);
+}
+
+MACHINE_START(IQ80331, "Intel IQ80331")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= 0xfefff000,
+	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc,
+	.boot_params	= 0x00000100,
+	.map_io		= iop3xx_map_io,
+	.init_irq	= iop331_init_irq,
+	.timer		= &iq80331_timer,
+	.init_machine	= iq80331_init_machine,
+MACHINE_END
diff --git a/arch/arm/mach-iop33x/iq80332-pci.c b/arch/arm/mach-iop33x/iq80332-pci.c
deleted file mode 100644
index 0de8aa748dd8..000000000000
--- a/arch/arm/mach-iop33x/iq80332-pci.c
+++ /dev/null
@@ -1,86 +0,0 @@
-/*
- * arch/arm/mach-iop33x/iq80332-pci.c
- *
- * PCI support for the Intel IQ80332 reference board
- *
- * Author: Dave Jiang <dave.jiang@intel.com>
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/kernel.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-
-#include <asm/hardware.h>
-#include <asm/irq.h>
-#include <asm/mach/pci.h>
-#include <asm/mach-types.h>
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those systems that do not already have PCI
- * interrupts properly routed.  We assume 1 <= pin <= 4
- */
-#define PCI_IRQ_TABLE_LOOKUP(minid,maxid)	\
-({ int _ctl_ = -1;				\
-   unsigned int _idsel = idsel - minid;		\
-   if (_idsel <= maxid)				\
-      _ctl_ = pci_irq_table[_idsel][pin-1];	\
-   _ctl_; })
-
-#define INTA	IRQ_IQ80332_INTA
-#define INTB	IRQ_IQ80332_INTB
-#define INTC	IRQ_IQ80332_INTC
-#define INTD	IRQ_IQ80332_INTD
-
-//#define INTE	IRQ_IQ80332_I82544
-
-static inline int __init
-iq80332_map_irq(struct pci_dev *dev, u8 idsel, u8 pin)
-{
-	static int pci_irq_table[][8] = {
-		/*
-		 * PCI IDSEL/INTPIN->INTLINE
-		 * A       B       C       D
-		 */
-		{-1,   -1,   -1,   -1},
-		{-1,   -1,   -1,   -1},
-		{-1,   -1,   -1,   -1},
-		{INTA, INTB, INTC, INTD}, /* PCI-X Slot */
-		{-1,   -1,   -1,   -1},
-		{INTC, INTC, INTC, INTC}, /* GigE  */
-		{-1,   -1,   -1,   -1},
-		{-1,   -1,   -1,   -1},
-	};
-
-	BUG_ON(pin < 1 || pin > 4);
-
-	return PCI_IRQ_TABLE_LOOKUP(1, 7);
-}
-
-static struct hw_pci iq80332_pci __initdata = {
-	.swizzle	= pci_std_swizzle,
-	.nr_controllers = 1,
-	.setup		= iop3xx_pci_setup,
-	.scan		= iop3xx_pci_scan_bus,
-	.preinit	= iop3xx_pci_preinit,
-	.map_irq	= iq80332_map_irq
-};
-
-static int __init iq80332_pci_init(void)
-{
-	if (machine_is_iq80332())
-		pci_common_init(&iq80332_pci);
-	return 0;
-}
-
-subsys_initcall(iq80332_pci_init);
-
-
-
-
diff --git a/arch/arm/mach-iop33x/iq80332.c b/arch/arm/mach-iop33x/iq80332.c
new file mode 100644
index 000000000000..150f3fd5de0b
--- /dev/null
+++ b/arch/arm/mach-iop33x/iq80332.c
@@ -0,0 +1,148 @@
+/*
+ * arch/arm/mach-iop33x/iq80332.c
+ *
+ * Board support code for the Intel IQ80332 platform.
+ *
+ * Author: Dave Jiang <dave.jiang@intel.com>
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/serial_core.h>
+#include <linux/serial_8250.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/pci.h>
+#include <asm/mach/time.h>
+#include <asm/mach-types.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+/*
+ * IQ80332 timer tick configuration.
+ */
+static void __init iq80332_timer_init(void)
+{
+	/* D-Step parts and the iop333 run at a higher internal bus frequency */
+	if (*IOP3XX_ATURID >= 0xa || *IOP3XX_ATUDID == 0x374)
+		iop3xx_init_time(333000000);
+	else
+		iop3xx_init_time(266000000);
+}
+
+static struct sys_timer iq80332_timer = {
+	.init		= iq80332_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
+
+/*
+ * IQ80332 PCI.
+ */
+static inline int __init
+iq80332_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+
+	if (slot == 4 && pin == 1) {
+		/* PCI-X Slot INTA */
+		irq = IRQ_IOP331_XINT0;
+	} else if (slot == 4 && pin == 2) {
+		/* PCI-X Slot INTB */
+		irq = IRQ_IOP331_XINT1;
+	} else if (slot == 4 && pin == 3) {
+		/* PCI-X Slot INTC */
+		irq = IRQ_IOP331_XINT2;
+	} else if (slot == 4 && pin == 4) {
+		/* PCI-X Slot INTD */
+		irq = IRQ_IOP331_XINT3;
+	} else if (slot == 6) {
+		/* GigE */
+		irq = IRQ_IOP331_XINT2;
+	} else {
+		printk(KERN_ERR "iq80332_pci_map_irq() called for unknown "
+			"device PCI:%d:%d:%d\n", dev->bus->number,
+			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+		irq = -1;
+	}
+
+	return irq;
+}
+
+static struct hw_pci iq80332_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= iq80332_pci_map_irq,
+};
+
+static int __init iq80332_pci_init(void)
+{
+	if (machine_is_iq80332())
+		pci_common_init(&iq80332_pci);
+
+	return 0;
+}
+
+subsys_initcall(iq80332_pci_init);
+
+
+/*
+ * IQ80332 machine initialisation.
+ */
+static struct physmap_flash_data iq80332_flash_data = {
+	.width		= 1,
+};
+
+static struct resource iq80332_flash_resource = {
+	.start		= 0xc0000000,
+	.end		= 0xc07fffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device iq80332_flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &iq80332_flash_data,
+	},
+	.num_resources	= 1,
+	.resource	= &iq80332_flash_resource,
+};
+
+static void __init iq80332_init_machine(void)
+{
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
+	platform_device_register(&iop33x_uart0_device);
+	platform_device_register(&iop33x_uart1_device);
+	platform_device_register(&iq80332_flash_device);
+}
+
+MACHINE_START(IQ80332, "Intel IQ80332")
+	/* Maintainer: Intel Corp. */
+	.phys_io	= 0xfefff000,
+	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc,
+	.boot_params	= 0x00000100,
+	.map_io		= iop3xx_map_io,
+	.init_irq	= iop331_init_irq,
+	.timer		= &iq80332_timer,
+	.init_machine	= iq80332_init_machine,
+MACHINE_END
diff --git a/arch/arm/mach-iop33x/setup.c b/arch/arm/mach-iop33x/setup.c
deleted file mode 100644
index 7cf5015436f3..000000000000
--- a/arch/arm/mach-iop33x/setup.c
+++ /dev/null
@@ -1,162 +0,0 @@
-/*
- * linux/arch/arm/mach-iop33x/setup.c
- *
- * Author: Dave Jiang (dave.jiang@intel.com)
- * Copyright (C) 2004 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- */
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <linux/major.h>
-#include <linux/fs.h>
-#include <linux/platform_device.h>
-#include <linux/serial.h>
-#include <linux/tty.h>
-#include <linux/serial_8250.h>
-
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/mach/map.h>
-#include <asm/setup.h>
-#include <asm/system.h>
-#include <asm/memory.h>
-#include <asm/hardware.h>
-#include <asm/mach-types.h>
-#include <asm/mach/arch.h>
-#include <asm/mach/time.h>
-#include <asm/hardware/iop3xx.h>
-
-#define IOP331_UART_XTAL 33334000
-
-static struct resource iop33x_uart0_resources[] = {
-	[0] = {
-		.start = IOP331_UART0_PHYS,
-		.end = IOP331_UART0_PHYS + 0x3f,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_UART0,
-		.end = IRQ_IOP331_UART0,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct resource iop33x_uart1_resources[] = {
-	[0] = {
-		.start = IOP331_UART1_PHYS,
-		.end = IOP331_UART1_PHYS + 0x3f,
-		.flags = IORESOURCE_MEM,
-	},
-	[1] = {
-		.start = IRQ_IOP331_UART1,
-		.end = IRQ_IOP331_UART1,
-		.flags = IORESOURCE_IRQ
-	}
-};
-
-static struct plat_serial8250_port iop33x_uart0_data[] = {
-	{
-       .membase     = (char*)(IOP331_UART0_VIRT),
-       .mapbase     = (IOP331_UART0_PHYS),
-       .irq         = IRQ_IOP331_UART0,
-       .uartclk     = IOP331_UART_XTAL,
-       .regshift    = 2,
-       .iotype      = UPIO_MEM,
-       .flags       = UPF_SKIP_TEST,
-	},
-	{  },
-};
-
-static struct plat_serial8250_port iop33x_uart1_data[] = {
-	{
-       .membase     = (char*)(IOP331_UART1_VIRT),
-       .mapbase     = (IOP331_UART1_PHYS),
-       .irq         = IRQ_IOP331_UART1,
-       .uartclk     = IOP331_UART_XTAL,
-       .regshift    = 2,
-       .iotype      = UPIO_MEM,
-       .flags       = UPF_SKIP_TEST,
-	},
-	{  },
-};
-
-static struct platform_device iop33x_uart0 = {
-       .name = "serial8250",
-       .id = PLAT8250_DEV_PLATFORM,
-       .dev.platform_data = iop33x_uart0_data,
-       .num_resources = 2,
-       .resource = iop33x_uart0_resources,
-};
-
-static struct platform_device iop33x_uart1 = {
-       .name = "serial8250",
-       .id = PLAT8250_DEV_PLATFORM1,
-       .dev.platform_data = iop33x_uart1_data,
-       .num_resources = 2,
-       .resource = iop33x_uart1_resources,
-};
-
-static struct platform_device *iop33x_devices[] __initdata = {
-	&iop33x_uart0,
-	&iop33x_uart1,
-};
-
-void __init iop33x_init(void)
-{
-	if(iop_is_331())
-	{
-		platform_add_devices(iop33x_devices,
-				ARRAY_SIZE(iop33x_devices));
-	}
-	platform_device_register(&iop3xx_i2c0_device);
-	platform_device_register(&iop3xx_i2c1_device);
-}
-
-#ifdef CONFIG_ARCH_IOP33X
-extern void iop331_init_irq(void);
-#endif
-
-static void __init iop3xx_timer_init(void)
-{
-	iop3xx_init_time(IOP331_TICK_RATE);
-}
-
-struct sys_timer iop331_timer = {
-	.init		= iop3xx_timer_init,
-	.offset		= iop3xx_gettimeoffset,
-};
-
-#if defined(CONFIG_ARCH_IQ80331)
-MACHINE_START(IQ80331, "Intel IQ80331")
-	/* Maintainer: Intel Corp. */
-	.phys_io	= 0xfefff000,
-	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
-	.map_io		= iop3xx_map_io,
-	.init_irq	= iop331_init_irq,
-	.timer		= &iop331_timer,
-	.boot_params	= 0x0100,
-	.init_machine	= iop33x_init,
-MACHINE_END
-
-#elif defined(CONFIG_MACH_IQ80332)
-MACHINE_START(IQ80332, "Intel IQ80332")
-	/* Maintainer: Intel Corp. */
-	.phys_io	= 0xfefff000,
-	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc, // virtual, physical
-	.map_io		= iop3xx_map_io,
-	.init_irq	= iop331_init_irq,
-	.timer		= &iop331_timer,
-	.boot_params	= 0x0100,
-	.init_machine	= iop33x_init,
-MACHINE_END
-
-#else
-#error No machine descriptor defined for this IOP3XX implementation
-#endif
-
-
diff --git a/arch/arm/mach-iop33x/uart.c b/arch/arm/mach-iop33x/uart.c
new file mode 100644
index 000000000000..d221d4abaa87
--- /dev/null
+++ b/arch/arm/mach-iop33x/uart.c
@@ -0,0 +1,106 @@
+/*
+ * linux/arch/arm/mach-iop33x/uart.c
+ *
+ * Author: Dave Jiang (dave.jiang@intel.com)
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+#include <linux/serial.h>
+#include <linux/tty.h>
+#include <linux/serial_8250.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/mach/map.h>
+#include <asm/setup.h>
+#include <asm/system.h>
+#include <asm/memory.h>
+#include <asm/hardware.h>
+#include <asm/hardware/iop3xx.h>
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+
+#define IOP331_UART_XTAL 33334000
+
+static struct plat_serial8250_port iop33x_uart0_data[] = {
+	{
+		.membase	= (char *)IOP331_UART0_VIRT,
+		.mapbase	= IOP331_UART0_PHYS,
+		.irq		= IRQ_IOP331_UART0,
+		.uartclk	= IOP331_UART_XTAL,
+		.regshift	= 2,
+		.iotype		= UPIO_MEM,
+		.flags		= UPF_SKIP_TEST,
+	},
+	{ },
+};
+
+static struct resource iop33x_uart0_resources[] = {
+	[0] = {
+		.start	= IOP331_UART0_PHYS,
+		.end	= IOP331_UART0_PHYS + 0x3f,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_IOP331_UART0,
+		.end	= IRQ_IOP331_UART0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+struct platform_device iop33x_uart0_device = {
+	.name		= "serial8250",
+	.id		= PLAT8250_DEV_PLATFORM,
+	.dev		= {
+		.platform_data		= iop33x_uart0_data,
+	},
+	.num_resources	= 2,
+	.resource	= iop33x_uart0_resources,
+};
+
+
+static struct resource iop33x_uart1_resources[] = {
+	[0] = {
+		.start	= IOP331_UART1_PHYS,
+		.end	= IOP331_UART1_PHYS + 0x3f,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_IOP331_UART1,
+		.end	= IRQ_IOP331_UART1,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct plat_serial8250_port iop33x_uart1_data[] = {
+	{
+		.membase	= (char *)IOP331_UART1_VIRT,
+		.mapbase	= IOP331_UART1_PHYS,
+		.irq		= IRQ_IOP331_UART1,
+		.uartclk	= IOP331_UART_XTAL,
+		.regshift	= 2,
+		.iotype		= UPIO_MEM,
+		.flags		= UPF_SKIP_TEST,
+	},
+	{ },
+};
+
+struct platform_device iop33x_uart1_device = {
+	.name		= "serial8250",
+	.id		= PLAT8250_DEV_PLATFORM1,
+	.dev		= {
+		.platform_data		= iop33x_uart1_data,
+	},
+	.num_resources	= 2,
+	.resource	= iop33x_uart1_resources,
+};
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index 1e57e0094767..8042946327ed 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -150,18 +150,6 @@
 #define IOP321_IINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007D8)
 #define IOP321_FINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007DC)
 
-/* Timers */
-#ifdef CONFIG_ARCH_IQ80321
-#define	IOP321_TICK_RATE	200000000	/* 200 MHz clock */
-#elif defined(CONFIG_ARCH_IQ31244)
-#define IOP321_TICK_RATE	198000000	/* 33.000 MHz crystal */
-#endif
-
-#ifdef CONFIG_ARCH_EP80219
-#undef IOP321_TICK_RATE
-#define IOP321_TICK_RATE 200000000 /* 33.333333 Mhz crystal */
-#endif
-
 /* Application accelerator unit 0x00000800 - 0x000008FF */
 #define IOP321_AAU_ACR     (volatile u32 *)IOP321_REG_ADDR(0x00000800)
 #define IOP321_AAU_ASR     (volatile u32 *)IOP321_REG_ADDR(0x00000804)
diff --git a/include/asm-arm/arch-iop32x/iq31244.h b/include/asm-arm/arch-iop32x/iq31244.h
index f490063d2156..cf2d2343398d 100644
--- a/include/asm-arm/arch-iop32x/iq31244.h
+++ b/include/asm-arm/arch-iop32x/iq31244.h
@@ -7,18 +7,11 @@
 #ifndef _IQ31244_H_
 #define _IQ31244_H_
 
-#define	IQ31244_FLASHBASE	0xf0000000	/* Flash */
-#define	IQ31244_FLASHSIZE	0x00800000
-#define	IQ31244_FLASHWIDTH	2
-
 #define IQ31244_UART		0xfe800000	/* UART #1 */
 #define IQ31244_7SEG_1		0xfe840000	/* 7-Segment MSB */
 #define IQ31244_7SEG_0		0xfe850000	/* 7-Segment LSB (WO) */
 #define IQ31244_ROTARY_SW	0xfe8d0000	/* Rotary Switch */
 #define IQ31244_BATT_STAT	0xfe8f0000	/* Battery Status */
 
-#ifndef __ASSEMBLY__
-extern void iq31244_map_io(void);
-#endif
 
 #endif	// _IQ31244_H_
diff --git a/include/asm-arm/arch-iop32x/iq80321.h b/include/asm-arm/arch-iop32x/iq80321.h
index 7015a605ab64..55d70f49b7fd 100644
--- a/include/asm-arm/arch-iop32x/iq80321.h
+++ b/include/asm-arm/arch-iop32x/iq80321.h
@@ -7,18 +7,11 @@
 #ifndef _IQ80321_H_
 #define _IQ80321_H_
 
-#define	IQ80321_FLASHBASE	0xf0000000	/* Flash */
-#define	IQ80321_FLASHSIZE	0x00800000
-#define	IQ80321_FLASHWIDTH	1
-
 #define IQ80321_UART		0xfe800000	/* UART #1 */
 #define IQ80321_7SEG_1		0xfe840000	/* 7-Segment MSB */
 #define IQ80321_7SEG_0		0xfe850000	/* 7-Segment LSB (WO) */
 #define IQ80321_ROTARY_SW	0xfe8d0000	/* Rotary Switch */
 #define IQ80321_BATT_STAT	0xfe8f0000	/* Battery Status */
 
-#ifndef __ASSEMBLY__
-extern void iq80321_map_io(void);
-#endif
 
 #endif	// _IQ80321_H_
diff --git a/include/asm-arm/arch-iop32x/irqs.h b/include/asm-arm/arch-iop32x/irqs.h
index 9fefcf3372b1..a48327ced92e 100644
--- a/include/asm-arm/arch-iop32x/irqs.h
+++ b/include/asm-arm/arch-iop32x/irqs.h
@@ -47,42 +47,4 @@
 #define NR_IRQS			32
 
 
-/*
- * Interrupts available on the IQ80321 board
- */
-
-/*
- * On board devices
- */
-#define	IRQ_IQ80321_I82544	IRQ_IOP321_XINT0
-#define IRQ_IQ80321_UART	IRQ_IOP321_XINT1
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ80321_INTA	IRQ_IOP321_XINT0
-#define	IRQ_IQ80321_INTB	IRQ_IOP321_XINT1
-#define	IRQ_IQ80321_INTC	IRQ_IOP321_XINT2
-#define	IRQ_IQ80321_INTD	IRQ_IOP321_XINT3
-
-/*
- * Interrupts on the IQ31244 board
- */
-
-/*
- * On board devices
- */
-#define IRQ_IQ31244_UART	IRQ_IOP321_XINT1
-#define	IRQ_IQ31244_I82546	IRQ_IOP321_XINT0
-#define IRQ_IQ31244_SATA	IRQ_IOP321_XINT2
-#define	IRQ_IQ31244_PCIX_SLOT	IRQ_IOP321_XINT3
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ31244_INTA	IRQ_IOP321_XINT0
-#define	IRQ_IQ31244_INTB	IRQ_IOP321_XINT1
-#define	IRQ_IQ31244_INTC	IRQ_IOP321_XINT2
-#define	IRQ_IQ31244_INTD	IRQ_IOP321_XINT3
-
 #endif // _IRQ_H_
diff --git a/include/asm-arm/arch-iop33x/hardware.h b/include/asm-arm/arch-iop33x/hardware.h
index 5e3cb32af020..3ebfdc6fea99 100644
--- a/include/asm-arm/arch-iop33x/hardware.h
+++ b/include/asm-arm/arch-iop33x/hardware.h
@@ -22,6 +22,11 @@
 #define PCIBIOS_MIN_IO		0x00000000
 #define PCIBIOS_MIN_MEM		0x00000000
 
+#ifndef __ASSEMBLY__
+extern struct platform_device iop33x_uart0_device;
+extern struct platform_device iop33x_uart1_device;
+#endif
+
 
 /*
  * Generic chipset bits
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index d12a95aa967a..a21872abd877 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -136,16 +136,6 @@
 #define IOP331_FINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007CC)
 
 
-/* Timers */
-#if defined(CONFIG_ARCH_IOP33X)
-#define	IOP331_TICK_RATE	266000000	/* 266 MHz IB clock */
-#endif
-
-#if defined(CONFIG_IOP331_STEPD) || defined(CONFIG_ARCH_IQ80333)
-#undef IOP331_TICK_RATE
-#define IOP331_TICK_RATE	333000000	/* 333 Mhz IB clock */
-#endif
-
 /* Application accelerator unit 0x00000800 - 0x000008FF */
 #define IOP331_AAU_ACR     (volatile u32 *)IOP331_REG_ADDR(0x00000800)
 #define IOP331_AAU_ASR     (volatile u32 *)IOP331_REG_ADDR(0x00000804)
diff --git a/include/asm-arm/arch-iop33x/iq80331.h b/include/asm-arm/arch-iop33x/iq80331.h
index bda7ab6d55cf..186762bf8944 100644
--- a/include/asm-arm/arch-iop33x/iq80331.h
+++ b/include/asm-arm/arch-iop33x/iq80331.h
@@ -7,17 +7,10 @@
 #ifndef _IQ80331_H_
 #define _IQ80331_H_
 
-#define	IQ80331_FLASHBASE	0xc0000000	/* Flash */
-#define	IQ80331_FLASHSIZE	0x00800000
-#define	IQ80331_FLASHWIDTH	1
-
 #define IQ80331_7SEG_1		0xce840000	/* 7-Segment MSB */
 #define IQ80331_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
 #define IQ80331_ROTARY_SW	0xce8d0000	/* Rotary Switch */
 #define IQ80331_BATT_STAT	0xce8f0000	/* Battery Status */
 
-#ifndef __ASSEMBLY__
-extern void iq80331_map_io(void);
-#endif
 
 #endif	// _IQ80331_H_
diff --git a/include/asm-arm/arch-iop33x/iq80332.h b/include/asm-arm/arch-iop33x/iq80332.h
index f728e04378ab..2a5d4ee01df9 100644
--- a/include/asm-arm/arch-iop33x/iq80332.h
+++ b/include/asm-arm/arch-iop33x/iq80332.h
@@ -7,17 +7,10 @@
 #ifndef _IQ80332_H_
 #define _IQ80332_H_
 
-#define	IQ80332_FLASHBASE	0xc0000000	/* Flash */
-#define	IQ80332_FLASHSIZE	0x00800000
-#define	IQ80332_FLASHWIDTH	1
-
 #define IQ80332_7SEG_1		0xce840000	/* 7-Segment MSB */
 #define IQ80332_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
 #define IQ80332_ROTARY_SW	0xce8d0000	/* Rotary Switch */
 #define IQ80332_BATT_STAT	0xce8f0000	/* Battery Status */
 
-#ifndef __ASSEMBLY__
-extern void iq80332_map_io(void);
-#endif
 
 #endif	// _IQ80332_H_
diff --git a/include/asm-arm/arch-iop33x/irqs.h b/include/asm-arm/arch-iop33x/irqs.h
index 2e3ade3b5ff9..a875404a07fc 100644
--- a/include/asm-arm/arch-iop33x/irqs.h
+++ b/include/asm-arm/arch-iop33x/irqs.h
@@ -57,42 +57,4 @@
 #define NR_IRQS			64
 
 
-/*
- * Interrupts available on the IQ80331 board
- */
-
-/*
- * On board devices
- */
-#define	IRQ_IQ80331_I82544	IRQ_IOP331_XINT0
-#define IRQ_IQ80331_UART0	IRQ_IOP331_UART0
-#define IRQ_IQ80331_UART1	IRQ_IOP331_UART1
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ80331_INTA	IRQ_IOP331_XINT0
-#define	IRQ_IQ80331_INTB	IRQ_IOP331_XINT1
-#define	IRQ_IQ80331_INTC	IRQ_IOP331_XINT2
-#define	IRQ_IQ80331_INTD	IRQ_IOP331_XINT3
-
-/*
- * Interrupts available on the IQ80332 board
- */
-
-/*
- * On board devices
- */
-#define	IRQ_IQ80332_I82544	IRQ_IOP331_XINT0
-#define IRQ_IQ80332_UART0	IRQ_IOP331_UART0
-#define IRQ_IQ80332_UART1	IRQ_IOP331_UART1
-
-/*
- * PCI interrupts
- */
-#define	IRQ_IQ80332_INTA	IRQ_IOP331_XINT0
-#define	IRQ_IQ80332_INTB	IRQ_IOP331_XINT1
-#define	IRQ_IQ80332_INTC	IRQ_IOP331_XINT2
-#define	IRQ_IQ80332_INTD	IRQ_IOP331_XINT3
-
 #endif // _IRQ_H_
-- 
cgit v1.2.3


From 475549faa161f4e002225f2ef75fdd2a6d83d151 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:25:33 +0100
Subject: [ARM] 3831/1: iop3xx: factor out common register defines

Factor out the register defines for a number of other peripherals
common to the iop32x and iop33x.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-iop32x/iop321.h | 132 -----------------------------------
 include/asm-arm/arch-iop33x/iop331.h | 115 ------------------------------
 include/asm-arm/hardware/iop3xx.h    | 114 ++++++++++++++++++++++++++++++
 3 files changed, 114 insertions(+), 247 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
index 8042946327ed..1757222a4cad 100644
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ b/include/asm-arm/arch-iop32x/iop321.h
@@ -37,102 +37,13 @@
 
 /* Messaging Unit 0x00000300 through 0x000003FF */
 
-/* Reserved 0x00000300 through 0x0000030c */
-#define IOP321_IMR0       (volatile u32 *)IOP321_REG_ADDR(0x00000310)
-#define IOP321_IMR1       (volatile u32 *)IOP321_REG_ADDR(0x00000314)
-#define IOP321_OMR0       (volatile u32 *)IOP321_REG_ADDR(0x00000318)
-#define IOP321_OMR1       (volatile u32 *)IOP321_REG_ADDR(0x0000031C)
-#define IOP321_IDR        (volatile u32 *)IOP321_REG_ADDR(0x00000320)
-#define IOP321_IISR       (volatile u32 *)IOP321_REG_ADDR(0x00000324)
-#define IOP321_IIMR       (volatile u32 *)IOP321_REG_ADDR(0x00000328)
-#define IOP321_ODR        (volatile u32 *)IOP321_REG_ADDR(0x0000032C)
-#define IOP321_OISR       (volatile u32 *)IOP321_REG_ADDR(0x00000330)
-#define IOP321_OIMR       (volatile u32 *)IOP321_REG_ADDR(0x00000334)
-/* Reserved 0x00000338 through 0x0000034F */
-#define IOP321_MUCR       (volatile u32 *)IOP321_REG_ADDR(0x00000350)
-#define IOP321_QBAR       (volatile u32 *)IOP321_REG_ADDR(0x00000354)
-/* Reserved 0x00000358 through 0x0000035C */
-#define IOP321_IFHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000360)
-#define IOP321_IFTPR      (volatile u32 *)IOP321_REG_ADDR(0x00000364)
-#define IOP321_IPHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000368)
-#define IOP321_IPTPR      (volatile u32 *)IOP321_REG_ADDR(0x0000036C)
-#define IOP321_OFHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000370)
-#define IOP321_OFTPR      (volatile u32 *)IOP321_REG_ADDR(0x00000374)
-#define IOP321_OPHPR      (volatile u32 *)IOP321_REG_ADDR(0x00000378)
-#define IOP321_OPTPR      (volatile u32 *)IOP321_REG_ADDR(0x0000037C)
-#define IOP321_IAR        (volatile u32 *)IOP321_REG_ADDR(0x00000380)
-
-#define IOP321_IIxR_MASK	0x7f /* masks all */
-#define IOP321_IIxR_IRI		0x40 /* RC Index Register Interrupt */
-#define IOP321_IIxR_OFQF	0x20 /* RC Output Free Q Full (ERROR) */
-#define IOP321_IIxR_ipq		0x10 /* RC Inbound Post Q (post) */
-#define IOP321_IIxR_ERRDI	0x08 /* RO Error Doorbell Interrupt */
-#define IOP321_IIxR_IDI		0x04 /* RO Inbound Doorbell Interrupt */
-#define IOP321_IIxR_IM1		0x02 /* RC Inbound Message 1 Interrupt */
-#define IOP321_IIxR_IM0		0x01 /* RC Inbound Message 0 Interrupt */
-
-/* Reserved 0x00000384 through 0x000003FF */
-
 /* DMA Controller 0x00000400 through 0x000004FF */
-#define IOP321_DMA0_CCR   (volatile u32 *)IOP321_REG_ADDR(0x00000400)
-#define IOP321_DMA0_CSR   (volatile u32 *)IOP321_REG_ADDR(0x00000404)
-#define IOP321_DMA0_DAR   (volatile u32 *)IOP321_REG_ADDR(0x0000040C)
-#define IOP321_DMA0_NDAR  (volatile u32 *)IOP321_REG_ADDR(0x00000410)
-#define IOP321_DMA0_PADR  (volatile u32 *)IOP321_REG_ADDR(0x00000414)
-#define IOP321_DMA0_PUADR (volatile u32 *)IOP321_REG_ADDR(0x00000418)
-#define IOP321_DMA0_LADR  (volatile u32 *)IOP321_REG_ADDR(0X0000041C)
-#define IOP321_DMA0_BCR   (volatile u32 *)IOP321_REG_ADDR(0x00000420)
-#define IOP321_DMA0_DCR   (volatile u32 *)IOP321_REG_ADDR(0x00000424)
-/* Reserved 0x00000428 through 0x0000043C */
-#define IOP321_DMA1_CCR   (volatile u32 *)IOP321_REG_ADDR(0x00000440)
-#define IOP321_DMA1_CSR   (volatile u32 *)IOP321_REG_ADDR(0x00000444)
-#define IOP321_DMA1_DAR   (volatile u32 *)IOP321_REG_ADDR(0x0000044C)
-#define IOP321_DMA1_NDAR  (volatile u32 *)IOP321_REG_ADDR(0x00000450)
-#define IOP321_DMA1_PADR  (volatile u32 *)IOP321_REG_ADDR(0x00000454)
-#define IOP321_DMA1_PUADR (volatile u32 *)IOP321_REG_ADDR(0x00000458)
-#define IOP321_DMA1_LADR  (volatile u32 *)IOP321_REG_ADDR(0x0000045C)
-#define IOP321_DMA1_BCR   (volatile u32 *)IOP321_REG_ADDR(0x00000460)
-#define IOP321_DMA1_DCR   (volatile u32 *)IOP321_REG_ADDR(0x00000464)
-/* Reserved 0x00000468 through 0x000004FF */
 
 /* Memory controller 0x00000500 through 0x0005FF */
 
 /* Peripheral bus interface unit 0x00000680 through 0x0006FF */
-#define IOP321_PBCR       (volatile u32 *)IOP321_REG_ADDR(0x00000680)
-#define IOP321_PBISR      (volatile u32 *)IOP321_REG_ADDR(0x00000684)
-#define IOP321_PBBAR0     (volatile u32 *)IOP321_REG_ADDR(0x00000688)
-#define IOP321_PBLR0      (volatile u32 *)IOP321_REG_ADDR(0x0000068C)
-#define IOP321_PBBAR1     (volatile u32 *)IOP321_REG_ADDR(0x00000690)
-#define IOP321_PBLR1      (volatile u32 *)IOP321_REG_ADDR(0x00000694)
-#define IOP321_PBBAR2     (volatile u32 *)IOP321_REG_ADDR(0x00000698)
-#define IOP321_PBLR2      (volatile u32 *)IOP321_REG_ADDR(0x0000069C)
-#define IOP321_PBBAR3     (volatile u32 *)IOP321_REG_ADDR(0x000006A0)
-#define IOP321_PBLR3      (volatile u32 *)IOP321_REG_ADDR(0x000006A4)
-#define IOP321_PBBAR4     (volatile u32 *)IOP321_REG_ADDR(0x000006A8)
-#define IOP321_PBLR4      (volatile u32 *)IOP321_REG_ADDR(0x000006AC)
-#define IOP321_PBBAR5     (volatile u32 *)IOP321_REG_ADDR(0x000006B0)
-#define IOP321_PBLR5      (volatile u32 *)IOP321_REG_ADDR(0x000006B4)
-#define IOP321_PBDSCR     (volatile u32 *)IOP321_REG_ADDR(0x000006B8)
-/* Reserved 0x000006BC */
-#define IOP321_PMBR0      (volatile u32 *)IOP321_REG_ADDR(0x000006C0)
-/* Reserved 0x000006C4 through 0x000006DC */
-#define IOP321_PMBR1      (volatile u32 *)IOP321_REG_ADDR(0x000006E0)
-#define IOP321_PMBR2      (volatile u32 *)IOP321_REG_ADDR(0x000006E4)
-
-#define IOP321_PBCR_EN    0x1
-
-#define IOP321_PBISR_BOOR_ERR 0x1
 
 /* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
-#define IOP321_GTMR	(volatile u32 *)IOP321_REG_ADDR(0x00000700)
-#define IOP321_ESR	(volatile u32 *)IOP321_REG_ADDR(0x00000704)
-#define IOP321_EMISR	(volatile u32 *)IOP321_REG_ADDR(0x00000708)
-/* reserved 0x00000070c */
-#define IOP321_GTSR	(volatile u32 *)IOP321_REG_ADDR(0x00000710)
-/* PERC0 DOESN'T EXIST - index from 1! */
-#define IOP321_PERCR0	(volatile u32 *)IOP321_REG_ADDR(0x00000710)
-
-#define IOP321_GTMR_NGCE	0x04 /* (Not) Global Counter Enable */
 
 /* Internal arbitration unit 0x00000780 through 0x0007BF */
 #define IOP321_IACR	(volatile u32 *)IOP321_REG_ADDR(0x00000780)
@@ -151,49 +62,6 @@
 #define IOP321_FINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007DC)
 
 /* Application accelerator unit 0x00000800 - 0x000008FF */
-#define IOP321_AAU_ACR     (volatile u32 *)IOP321_REG_ADDR(0x00000800)
-#define IOP321_AAU_ASR     (volatile u32 *)IOP321_REG_ADDR(0x00000804)
-#define IOP321_AAU_ADAR    (volatile u32 *)IOP321_REG_ADDR(0x00000808)
-#define IOP321_AAU_ANDAR   (volatile u32 *)IOP321_REG_ADDR(0x0000080C)
-#define IOP321_AAU_SAR1    (volatile u32 *)IOP321_REG_ADDR(0x00000810)
-#define IOP321_AAU_SAR2    (volatile u32 *)IOP321_REG_ADDR(0x00000814)
-#define IOP321_AAU_SAR3    (volatile u32 *)IOP321_REG_ADDR(0x00000818)
-#define IOP321_AAU_SAR4    (volatile u32 *)IOP321_REG_ADDR(0x0000081C)
-#define IOP321_AAU_SAR5    (volatile u32 *)IOP321_REG_ADDR(0x0000082C)
-#define IOP321_AAU_SAR6    (volatile u32 *)IOP321_REG_ADDR(0x00000830)
-#define IOP321_AAU_SAR7    (volatile u32 *)IOP321_REG_ADDR(0x00000834)
-#define IOP321_AAU_SAR8    (volatile u32 *)IOP321_REG_ADDR(0x00000838)
-#define IOP321_AAU_SAR9    (volatile u32 *)IOP321_REG_ADDR(0x00000840)
-#define IOP321_AAU_SAR10   (volatile u32 *)IOP321_REG_ADDR(0x00000844)
-#define IOP321_AAU_SAR11   (volatile u32 *)IOP321_REG_ADDR(0x00000848)
-#define IOP321_AAU_SAR12   (volatile u32 *)IOP321_REG_ADDR(0x0000084C)
-#define IOP321_AAU_SAR13   (volatile u32 *)IOP321_REG_ADDR(0x00000850)
-#define IOP321_AAU_SAR14   (volatile u32 *)IOP321_REG_ADDR(0x00000854)
-#define IOP321_AAU_SAR15   (volatile u32 *)IOP321_REG_ADDR(0x00000858)
-#define IOP321_AAU_SAR16   (volatile u32 *)IOP321_REG_ADDR(0x0000085C)
-#define IOP321_AAU_SAR17   (volatile u32 *)IOP321_REG_ADDR(0x00000864)
-#define IOP321_AAU_SAR18   (volatile u32 *)IOP321_REG_ADDR(0x00000868)
-#define IOP321_AAU_SAR19   (volatile u32 *)IOP321_REG_ADDR(0x0000086C)
-#define IOP321_AAU_SAR20   (volatile u32 *)IOP321_REG_ADDR(0x00000870)
-#define IOP321_AAU_SAR21   (volatile u32 *)IOP321_REG_ADDR(0x00000874)
-#define IOP321_AAU_SAR22   (volatile u32 *)IOP321_REG_ADDR(0x00000878)
-#define IOP321_AAU_SAR23   (volatile u32 *)IOP321_REG_ADDR(0x0000087C)
-#define IOP321_AAU_SAR24   (volatile u32 *)IOP321_REG_ADDR(0x00000880)
-#define IOP321_AAU_SAR25   (volatile u32 *)IOP321_REG_ADDR(0x00000888)
-#define IOP321_AAU_SAR26   (volatile u32 *)IOP321_REG_ADDR(0x0000088C)
-#define IOP321_AAU_SAR27   (volatile u32 *)IOP321_REG_ADDR(0x00000890)
-#define IOP321_AAU_SAR28   (volatile u32 *)IOP321_REG_ADDR(0x00000894)
-#define IOP321_AAU_SAR29   (volatile u32 *)IOP321_REG_ADDR(0x00000898)
-#define IOP321_AAU_SAR30   (volatile u32 *)IOP321_REG_ADDR(0x0000089C)
-#define IOP321_AAU_SAR31   (volatile u32 *)IOP321_REG_ADDR(0x000008A0)
-#define IOP321_AAU_SAR32   (volatile u32 *)IOP321_REG_ADDR(0x000008A4)
-#define IOP321_AAU_DAR     (volatile u32 *)IOP321_REG_ADDR(0x00000820)
-#define IOP321_AAU_ABCR    (volatile u32 *)IOP321_REG_ADDR(0x00000824)
-#define IOP321_AAU_ADCR    (volatile u32 *)IOP321_REG_ADDR(0x00000828)
-#define IOP321_AAU_EDCR0   (volatile u32 *)IOP321_REG_ADDR(0x0000083c)
-#define IOP321_AAU_EDCR1   (volatile u32 *)IOP321_REG_ADDR(0x00000860)
-#define IOP321_AAU_EDCR2   (volatile u32 *)IOP321_REG_ADDR(0x00000884)
-
 
 /* SSP serial port unit 0x00001600 - 0x0000167F */
 /* I2C bus interface unit 0x00001680 - 0x000016FF */
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
index a21872abd877..8c7ec583615f 100644
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ b/include/asm-arm/arch-iop33x/iop331.h
@@ -36,83 +36,11 @@
 
 /* Messaging Unit 0x00000300 through 0x000003FF */
 
-/* Reserved 0x00000300 through 0x0000030c */
-#define IOP331_IMR0       (volatile u32 *)IOP331_REG_ADDR(0x00000310)
-#define IOP331_IMR1       (volatile u32 *)IOP331_REG_ADDR(0x00000314)
-#define IOP331_OMR0       (volatile u32 *)IOP331_REG_ADDR(0x00000318)
-#define IOP331_OMR1       (volatile u32 *)IOP331_REG_ADDR(0x0000031C)
-#define IOP331_IDR        (volatile u32 *)IOP331_REG_ADDR(0x00000320)
-#define IOP331_IISR       (volatile u32 *)IOP331_REG_ADDR(0x00000324)
-#define IOP331_IIMR       (volatile u32 *)IOP331_REG_ADDR(0x00000328)
-#define IOP331_ODR        (volatile u32 *)IOP331_REG_ADDR(0x0000032C)
-#define IOP331_OISR       (volatile u32 *)IOP331_REG_ADDR(0x00000330)
-#define IOP331_OIMR       (volatile u32 *)IOP331_REG_ADDR(0x00000334)
-/* Reserved 0x00000338 through 0x0000034F */
-#define IOP331_MUCR       (volatile u32 *)IOP331_REG_ADDR(0x00000350)
-#define IOP331_QBAR       (volatile u32 *)IOP331_REG_ADDR(0x00000354)
-/* Reserved 0x00000358 through 0x0000035C */
-#define IOP331_IFHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000360)
-#define IOP331_IFTPR      (volatile u32 *)IOP331_REG_ADDR(0x00000364)
-#define IOP331_IPHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000368)
-#define IOP331_IPTPR      (volatile u32 *)IOP331_REG_ADDR(0x0000036C)
-#define IOP331_OFHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000370)
-#define IOP331_OFTPR      (volatile u32 *)IOP331_REG_ADDR(0x00000374)
-#define IOP331_OPHPR      (volatile u32 *)IOP331_REG_ADDR(0x00000378)
-#define IOP331_OPTPR      (volatile u32 *)IOP331_REG_ADDR(0x0000037C)
-#define IOP331_IAR        (volatile u32 *)IOP331_REG_ADDR(0x00000380)
-/* Reserved 0x00000384 through 0x000003FF */
-
 /* DMA Controller 0x00000400 through 0x000004FF */
-#define IOP331_DMA0_CCR   (volatile u32 *)IOP331_REG_ADDR(0x00000400)
-#define IOP331_DMA0_CSR   (volatile u32 *)IOP331_REG_ADDR(0x00000404)
-#define IOP331_DMA0_DAR   (volatile u32 *)IOP331_REG_ADDR(0x0000040C)
-#define IOP331_DMA0_NDAR  (volatile u32 *)IOP331_REG_ADDR(0x00000410)
-#define IOP331_DMA0_PADR  (volatile u32 *)IOP331_REG_ADDR(0x00000414)
-#define IOP331_DMA0_PUADR (volatile u32 *)IOP331_REG_ADDR(0x00000418)
-#define IOP331_DMA0_LADR  (volatile u32 *)IOP331_REG_ADDR(0X0000041C)
-#define IOP331_DMA0_BCR   (volatile u32 *)IOP331_REG_ADDR(0x00000420)
-#define IOP331_DMA0_DCR   (volatile u32 *)IOP331_REG_ADDR(0x00000424)
-/* Reserved 0x00000428 through 0x0000043C */
-#define IOP331_DMA1_CCR   (volatile u32 *)IOP331_REG_ADDR(0x00000440)
-#define IOP331_DMA1_CSR   (volatile u32 *)IOP331_REG_ADDR(0x00000444)
-#define IOP331_DMA1_DAR   (volatile u32 *)IOP331_REG_ADDR(0x0000044C)
-#define IOP331_DMA1_NDAR  (volatile u32 *)IOP331_REG_ADDR(0x00000450)
-#define IOP331_DMA1_PADR  (volatile u32 *)IOP331_REG_ADDR(0x00000454)
-#define IOP331_DMA1_PUADR (volatile u32 *)IOP331_REG_ADDR(0x00000458)
-#define IOP331_DMA1_LADR  (volatile u32 *)IOP331_REG_ADDR(0x0000045C)
-#define IOP331_DMA1_BCR   (volatile u32 *)IOP331_REG_ADDR(0x00000460)
-#define IOP331_DMA1_DCR   (volatile u32 *)IOP331_REG_ADDR(0x00000464)
-/* Reserved 0x00000468 through 0x000004FF */
 
 /* Memory controller 0x00000500 through 0x0005FF */
 
 /* Peripheral bus interface unit 0x00000680 through 0x0006FF */
-#define IOP331_PBCR       (volatile u32 *)IOP331_REG_ADDR(0x00000680)
-#define IOP331_PBISR      (volatile u32 *)IOP331_REG_ADDR(0x00000684)
-#define IOP331_PBBAR0     (volatile u32 *)IOP331_REG_ADDR(0x00000688)
-#define IOP331_PBLR0      (volatile u32 *)IOP331_REG_ADDR(0x0000068C)
-#define IOP331_PBBAR1     (volatile u32 *)IOP331_REG_ADDR(0x00000690)
-#define IOP331_PBLR1      (volatile u32 *)IOP331_REG_ADDR(0x00000694)
-#define IOP331_PBBAR2     (volatile u32 *)IOP331_REG_ADDR(0x00000698)
-#define IOP331_PBLR2      (volatile u32 *)IOP331_REG_ADDR(0x0000069C)
-#define IOP331_PBBAR3     (volatile u32 *)IOP331_REG_ADDR(0x000006A0)
-#define IOP331_PBLR3      (volatile u32 *)IOP331_REG_ADDR(0x000006A4)
-#define IOP331_PBBAR4     (volatile u32 *)IOP331_REG_ADDR(0x000006A8)
-#define IOP331_PBLR4      (volatile u32 *)IOP331_REG_ADDR(0x000006AC)
-#define IOP331_PBBAR5     (volatile u32 *)IOP331_REG_ADDR(0x000006B0)
-#define IOP331_PBLR5      (volatile u32 *)IOP331_REG_ADDR(0x000006B4)
-#define IOP331_PBDSCR     (volatile u32 *)IOP331_REG_ADDR(0x000006B8)
-/* Reserved 0x000006BC */
-#define IOP331_PMBR0      (volatile u32 *)IOP331_REG_ADDR(0x000006C0)
-/* Reserved 0x000006C4 through 0x000006DC */
-#define IOP331_PMBR1      (volatile u32 *)IOP331_REG_ADDR(0x000006E0)
-#define IOP331_PMBR2      (volatile u32 *)IOP331_REG_ADDR(0x000006E4)
-
-#define IOP331_PBCR_EN    0x1
-
-#define IOP331_PBISR_BOOR_ERR 0x1
-
-
 
 /* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
 /* Internal arbitration unit 0x00000780 through 0x0007BF */
@@ -137,49 +65,6 @@
 
 
 /* Application accelerator unit 0x00000800 - 0x000008FF */
-#define IOP331_AAU_ACR     (volatile u32 *)IOP331_REG_ADDR(0x00000800)
-#define IOP331_AAU_ASR     (volatile u32 *)IOP331_REG_ADDR(0x00000804)
-#define IOP331_AAU_ADAR    (volatile u32 *)IOP331_REG_ADDR(0x00000808)
-#define IOP331_AAU_ANDAR   (volatile u32 *)IOP331_REG_ADDR(0x0000080C)
-#define IOP331_AAU_SAR1    (volatile u32 *)IOP331_REG_ADDR(0x00000810)
-#define IOP331_AAU_SAR2    (volatile u32 *)IOP331_REG_ADDR(0x00000814)
-#define IOP331_AAU_SAR3    (volatile u32 *)IOP331_REG_ADDR(0x00000818)
-#define IOP331_AAU_SAR4    (volatile u32 *)IOP331_REG_ADDR(0x0000081C)
-#define IOP331_AAU_SAR5    (volatile u32 *)IOP331_REG_ADDR(0x0000082C)
-#define IOP331_AAU_SAR6    (volatile u32 *)IOP331_REG_ADDR(0x00000830)
-#define IOP331_AAU_SAR7    (volatile u32 *)IOP331_REG_ADDR(0x00000834)
-#define IOP331_AAU_SAR8    (volatile u32 *)IOP331_REG_ADDR(0x00000838)
-#define IOP331_AAU_SAR9    (volatile u32 *)IOP331_REG_ADDR(0x00000840)
-#define IOP331_AAU_SAR10   (volatile u32 *)IOP331_REG_ADDR(0x00000844)
-#define IOP331_AAU_SAR11   (volatile u32 *)IOP331_REG_ADDR(0x00000848)
-#define IOP331_AAU_SAR12   (volatile u32 *)IOP331_REG_ADDR(0x0000084C)
-#define IOP331_AAU_SAR13   (volatile u32 *)IOP331_REG_ADDR(0x00000850)
-#define IOP331_AAU_SAR14   (volatile u32 *)IOP331_REG_ADDR(0x00000854)
-#define IOP331_AAU_SAR15   (volatile u32 *)IOP331_REG_ADDR(0x00000858)
-#define IOP331_AAU_SAR16   (volatile u32 *)IOP331_REG_ADDR(0x0000085C)
-#define IOP331_AAU_SAR17   (volatile u32 *)IOP331_REG_ADDR(0x00000864)
-#define IOP331_AAU_SAR18   (volatile u32 *)IOP331_REG_ADDR(0x00000868)
-#define IOP331_AAU_SAR19   (volatile u32 *)IOP331_REG_ADDR(0x0000086C)
-#define IOP331_AAU_SAR20   (volatile u32 *)IOP331_REG_ADDR(0x00000870)
-#define IOP331_AAU_SAR21   (volatile u32 *)IOP331_REG_ADDR(0x00000874)
-#define IOP331_AAU_SAR22   (volatile u32 *)IOP331_REG_ADDR(0x00000878)
-#define IOP331_AAU_SAR23   (volatile u32 *)IOP331_REG_ADDR(0x0000087C)
-#define IOP331_AAU_SAR24   (volatile u32 *)IOP331_REG_ADDR(0x00000880)
-#define IOP331_AAU_SAR25   (volatile u32 *)IOP331_REG_ADDR(0x00000888)
-#define IOP331_AAU_SAR26   (volatile u32 *)IOP331_REG_ADDR(0x0000088C)
-#define IOP331_AAU_SAR27   (volatile u32 *)IOP331_REG_ADDR(0x00000890)
-#define IOP331_AAU_SAR28   (volatile u32 *)IOP331_REG_ADDR(0x00000894)
-#define IOP331_AAU_SAR29   (volatile u32 *)IOP331_REG_ADDR(0x00000898)
-#define IOP331_AAU_SAR30   (volatile u32 *)IOP331_REG_ADDR(0x0000089C)
-#define IOP331_AAU_SAR31   (volatile u32 *)IOP331_REG_ADDR(0x000008A0)
-#define IOP331_AAU_SAR32   (volatile u32 *)IOP331_REG_ADDR(0x000008A4)
-#define IOP331_AAU_DAR     (volatile u32 *)IOP331_REG_ADDR(0x00000820)
-#define IOP331_AAU_ABCR    (volatile u32 *)IOP331_REG_ADDR(0x00000824)
-#define IOP331_AAU_ADCR    (volatile u32 *)IOP331_REG_ADDR(0x00000828)
-#define IOP331_AAU_EDCR0   (volatile u32 *)IOP331_REG_ADDR(0x0000083c)
-#define IOP331_AAU_EDCR1   (volatile u32 *)IOP331_REG_ADDR(0x00000860)
-#define IOP331_AAU_EDCR2   (volatile u32 *)IOP331_REG_ADDR(0x00000884)
-
 
 #define IOP331_SPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C0)
 #define IOP331_PPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C8)
diff --git a/include/asm-arm/hardware/iop3xx.h b/include/asm-arm/hardware/iop3xx.h
index f3c61d041fca..1018a7486ab7 100644
--- a/include/asm-arm/hardware/iop3xx.h
+++ b/include/asm-arm/hardware/iop3xx.h
@@ -97,6 +97,76 @@ extern void gpio_line_set(int line, int value);
 #define IOP3XX_PCIXSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01e4)
 #define IOP3XX_PCIIRSR		(volatile u32 *)IOP3XX_REG_ADDR(0x01ec)
 
+/* Messaging Unit  */
+#define IOP3XX_IMR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0310)
+#define IOP3XX_IMR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0314)
+#define IOP3XX_OMR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0318)
+#define IOP3XX_OMR1		(volatile u32 *)IOP3XX_REG_ADDR(0x031c)
+#define IOP3XX_IDR		(volatile u32 *)IOP3XX_REG_ADDR(0x0320)
+#define IOP3XX_IISR		(volatile u32 *)IOP3XX_REG_ADDR(0x0324)
+#define IOP3XX_IIMR		(volatile u32 *)IOP3XX_REG_ADDR(0x0328)
+#define IOP3XX_ODR		(volatile u32 *)IOP3XX_REG_ADDR(0x032c)
+#define IOP3XX_OISR		(volatile u32 *)IOP3XX_REG_ADDR(0x0330)
+#define IOP3XX_OIMR		(volatile u32 *)IOP3XX_REG_ADDR(0x0334)
+#define IOP3XX_MUCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0350)
+#define IOP3XX_QBAR		(volatile u32 *)IOP3XX_REG_ADDR(0x0354)
+#define IOP3XX_IFHPR		(volatile u32 *)IOP3XX_REG_ADDR(0x0360)
+#define IOP3XX_IFTPR		(volatile u32 *)IOP3XX_REG_ADDR(0x0364)
+#define IOP3XX_IPHPR		(volatile u32 *)IOP3XX_REG_ADDR(0x0368)
+#define IOP3XX_IPTPR		(volatile u32 *)IOP3XX_REG_ADDR(0x036c)
+#define IOP3XX_OFHPR		(volatile u32 *)IOP3XX_REG_ADDR(0x0370)
+#define IOP3XX_OFTPR		(volatile u32 *)IOP3XX_REG_ADDR(0x0374)
+#define IOP3XX_OPHPR		(volatile u32 *)IOP3XX_REG_ADDR(0x0378)
+#define IOP3XX_OPTPR		(volatile u32 *)IOP3XX_REG_ADDR(0x037c)
+#define IOP3XX_IAR		(volatile u32 *)IOP3XX_REG_ADDR(0x0380)
+
+/* DMA Controller  */
+#define IOP3XX_DMA0_CCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0400)
+#define IOP3XX_DMA0_CSR		(volatile u32 *)IOP3XX_REG_ADDR(0x0404)
+#define IOP3XX_DMA0_DAR		(volatile u32 *)IOP3XX_REG_ADDR(0x040c)
+#define IOP3XX_DMA0_NDAR	(volatile u32 *)IOP3XX_REG_ADDR(0x0410)
+#define IOP3XX_DMA0_PADR	(volatile u32 *)IOP3XX_REG_ADDR(0x0414)
+#define IOP3XX_DMA0_PUADR	(volatile u32 *)IOP3XX_REG_ADDR(0x0418)
+#define IOP3XX_DMA0_LADR	(volatile u32 *)IOP3XX_REG_ADDR(0x041c)
+#define IOP3XX_DMA0_BCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0420)
+#define IOP3XX_DMA0_DCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0424)
+#define IOP3XX_DMA1_CCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0440)
+#define IOP3XX_DMA1_CSR		(volatile u32 *)IOP3XX_REG_ADDR(0x0444)
+#define IOP3XX_DMA1_DAR		(volatile u32 *)IOP3XX_REG_ADDR(0x044c)
+#define IOP3XX_DMA1_NDAR	(volatile u32 *)IOP3XX_REG_ADDR(0x0450)
+#define IOP3XX_DMA1_PADR	(volatile u32 *)IOP3XX_REG_ADDR(0x0454)
+#define IOP3XX_DMA1_PUADR	(volatile u32 *)IOP3XX_REG_ADDR(0x0458)
+#define IOP3XX_DMA1_LADR	(volatile u32 *)IOP3XX_REG_ADDR(0x045c)
+#define IOP3XX_DMA1_BCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0460)
+#define IOP3XX_DMA1_DCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0464)
+
+/* Peripheral bus interface  */
+#define IOP3XX_PBCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0680)
+#define IOP3XX_PBISR		(volatile u32 *)IOP3XX_REG_ADDR(0x0684)
+#define IOP3XX_PBBAR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0688)
+#define IOP3XX_PBLR0		(volatile u32 *)IOP3XX_REG_ADDR(0x068c)
+#define IOP3XX_PBBAR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0690)
+#define IOP3XX_PBLR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0694)
+#define IOP3XX_PBBAR2		(volatile u32 *)IOP3XX_REG_ADDR(0x0698)
+#define IOP3XX_PBLR2		(volatile u32 *)IOP3XX_REG_ADDR(0x069c)
+#define IOP3XX_PBBAR3		(volatile u32 *)IOP3XX_REG_ADDR(0x06a0)
+#define IOP3XX_PBLR3		(volatile u32 *)IOP3XX_REG_ADDR(0x06a4)
+#define IOP3XX_PBBAR4		(volatile u32 *)IOP3XX_REG_ADDR(0x06a8)
+#define IOP3XX_PBLR4		(volatile u32 *)IOP3XX_REG_ADDR(0x06ac)
+#define IOP3XX_PBBAR5		(volatile u32 *)IOP3XX_REG_ADDR(0x06b0)
+#define IOP3XX_PBLR5		(volatile u32 *)IOP3XX_REG_ADDR(0x06b4)
+#define IOP3XX_PMBR0		(volatile u32 *)IOP3XX_REG_ADDR(0x06c0)
+#define IOP3XX_PMBR1		(volatile u32 *)IOP3XX_REG_ADDR(0x06e0)
+#define IOP3XX_PMBR2		(volatile u32 *)IOP3XX_REG_ADDR(0x06e4)
+
+/* Peripheral performance monitoring unit  */
+#define IOP3XX_GTMR		(volatile u32 *)IOP3XX_REG_ADDR(0x0700)
+#define IOP3XX_ESR		(volatile u32 *)IOP3XX_REG_ADDR(0x0704)
+#define IOP3XX_EMISR		(volatile u32 *)IOP3XX_REG_ADDR(0x0708)
+#define IOP3XX_GTSR		(volatile u32 *)IOP3XX_REG_ADDR(0x0710)
+/* PERCR0 DOESN'T EXIST - index from 1! */
+#define IOP3XX_PERCR0		(volatile u32 *)IOP3XX_REG_ADDR(0x0710)
+
 /* General Purpose I/O  */
 #define IOP3XX_GPOE		(volatile u32 *)IOP3XX_GPIO_REG(0x0004)
 #define IOP3XX_GPID		(volatile u32 *)IOP3XX_GPIO_REG(0x0008)
@@ -120,6 +190,50 @@ extern void gpio_line_set(int line, int value);
 #define IOP3XX_TMR_RATIO_8_1	0x20
 #define IOP3XX_TMR_RATIO_16_1	0x30
 
+/* Application accelerator unit  */
+#define IOP3XX_AAU_ACR		(volatile u32 *)IOP3XX_REG_ADDR(0x0800)
+#define IOP3XX_AAU_ASR		(volatile u32 *)IOP3XX_REG_ADDR(0x0804)
+#define IOP3XX_AAU_ADAR		(volatile u32 *)IOP3XX_REG_ADDR(0x0808)
+#define IOP3XX_AAU_ANDAR	(volatile u32 *)IOP3XX_REG_ADDR(0x080c)
+#define IOP3XX_AAU_SAR1		(volatile u32 *)IOP3XX_REG_ADDR(0x0810)
+#define IOP3XX_AAU_SAR2		(volatile u32 *)IOP3XX_REG_ADDR(0x0814)
+#define IOP3XX_AAU_SAR3		(volatile u32 *)IOP3XX_REG_ADDR(0x0818)
+#define IOP3XX_AAU_SAR4		(volatile u32 *)IOP3XX_REG_ADDR(0x081c)
+#define IOP3XX_AAU_DAR		(volatile u32 *)IOP3XX_REG_ADDR(0x0820)
+#define IOP3XX_AAU_ABCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0824)
+#define IOP3XX_AAU_ADCR		(volatile u32 *)IOP3XX_REG_ADDR(0x0828)
+#define IOP3XX_AAU_SAR5		(volatile u32 *)IOP3XX_REG_ADDR(0x082c)
+#define IOP3XX_AAU_SAR6		(volatile u32 *)IOP3XX_REG_ADDR(0x0830)
+#define IOP3XX_AAU_SAR7		(volatile u32 *)IOP3XX_REG_ADDR(0x0834)
+#define IOP3XX_AAU_SAR8		(volatile u32 *)IOP3XX_REG_ADDR(0x0838)
+#define IOP3XX_AAU_EDCR0	(volatile u32 *)IOP3XX_REG_ADDR(0x083c)
+#define IOP3XX_AAU_SAR9		(volatile u32 *)IOP3XX_REG_ADDR(0x0840)
+#define IOP3XX_AAU_SAR10	(volatile u32 *)IOP3XX_REG_ADDR(0x0844)
+#define IOP3XX_AAU_SAR11	(volatile u32 *)IOP3XX_REG_ADDR(0x0848)
+#define IOP3XX_AAU_SAR12	(volatile u32 *)IOP3XX_REG_ADDR(0x084c)
+#define IOP3XX_AAU_SAR13	(volatile u32 *)IOP3XX_REG_ADDR(0x0850)
+#define IOP3XX_AAU_SAR14	(volatile u32 *)IOP3XX_REG_ADDR(0x0854)
+#define IOP3XX_AAU_SAR15	(volatile u32 *)IOP3XX_REG_ADDR(0x0858)
+#define IOP3XX_AAU_SAR16	(volatile u32 *)IOP3XX_REG_ADDR(0x085c)
+#define IOP3XX_AAU_EDCR1	(volatile u32 *)IOP3XX_REG_ADDR(0x0860)
+#define IOP3XX_AAU_SAR17	(volatile u32 *)IOP3XX_REG_ADDR(0x0864)
+#define IOP3XX_AAU_SAR18	(volatile u32 *)IOP3XX_REG_ADDR(0x0868)
+#define IOP3XX_AAU_SAR19	(volatile u32 *)IOP3XX_REG_ADDR(0x086c)
+#define IOP3XX_AAU_SAR20	(volatile u32 *)IOP3XX_REG_ADDR(0x0870)
+#define IOP3XX_AAU_SAR21	(volatile u32 *)IOP3XX_REG_ADDR(0x0874)
+#define IOP3XX_AAU_SAR22	(volatile u32 *)IOP3XX_REG_ADDR(0x0878)
+#define IOP3XX_AAU_SAR23	(volatile u32 *)IOP3XX_REG_ADDR(0x087c)
+#define IOP3XX_AAU_SAR24	(volatile u32 *)IOP3XX_REG_ADDR(0x0880)
+#define IOP3XX_AAU_EDCR2	(volatile u32 *)IOP3XX_REG_ADDR(0x0884)
+#define IOP3XX_AAU_SAR25	(volatile u32 *)IOP3XX_REG_ADDR(0x0888)
+#define IOP3XX_AAU_SAR26	(volatile u32 *)IOP3XX_REG_ADDR(0x088c)
+#define IOP3XX_AAU_SAR27	(volatile u32 *)IOP3XX_REG_ADDR(0x0890)
+#define IOP3XX_AAU_SAR28	(volatile u32 *)IOP3XX_REG_ADDR(0x0894)
+#define IOP3XX_AAU_SAR29	(volatile u32 *)IOP3XX_REG_ADDR(0x0898)
+#define IOP3XX_AAU_SAR30	(volatile u32 *)IOP3XX_REG_ADDR(0x089c)
+#define IOP3XX_AAU_SAR31	(volatile u32 *)IOP3XX_REG_ADDR(0x08a0)
+#define IOP3XX_AAU_SAR32	(volatile u32 *)IOP3XX_REG_ADDR(0x08a4)
+
 /* I2C bus interface unit  */
 #define IOP3XX_ICR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1680)
 #define IOP3XX_ISR0		(volatile u32 *)IOP3XX_REG_ADDR(0x1684)
-- 
cgit v1.2.3


From c852ac80440db9b0a47f48578e9c6303078abbc1 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 18 Sep 2006 23:26:25 +0100
Subject: [ARM] 3832/1: iop3xx: coding style cleanup

Since the iop32x code isn't iop321-specific, and the iop33x code isn't
iop331-specfic, do a s/iop321/iop32x/ and s/iop331/iop33x/, and tidy up
the code to conform to the coding style guidelines somewhat better.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop32x/Kconfig              |   9 +--
 arch/arm/mach-iop32x/iq31244.c            |  22 +++---
 arch/arm/mach-iop32x/iq80321.c            |  14 ++--
 arch/arm/mach-iop32x/irq.c                |  54 ++++++---------
 arch/arm/mach-iop33x/iq80331.c            |  12 ++--
 arch/arm/mach-iop33x/iq80332.c            |  12 ++--
 arch/arm/mach-iop33x/irq.c                |  95 ++++++++++++--------------
 arch/arm/mach-iop33x/uart.c               |  37 +++++-----
 arch/arm/oprofile/op_model_xscale.c       |   6 +-
 arch/arm/plat-iop/i2c.c                   |   8 +--
 arch/arm/plat-iop/time.c                  |   4 +-
 include/asm-arm/arch-iop32x/debug-macro.S |  14 ++--
 include/asm-arm/arch-iop32x/dma.h         |   4 +-
 include/asm-arm/arch-iop32x/entry-macro.S |  11 ++-
 include/asm-arm/arch-iop32x/hardware.h    |  24 ++++---
 include/asm-arm/arch-iop32x/io.h          |  11 +--
 include/asm-arm/arch-iop32x/iop321.h      |  86 -----------------------
 include/asm-arm/arch-iop32x/iop32x.h      |  28 ++++++++
 include/asm-arm/arch-iop32x/iq31244.h     |   8 +--
 include/asm-arm/arch-iop32x/iq80321.h     |   8 +--
 include/asm-arm/arch-iop32x/irqs.h        |  66 +++++++++---------
 include/asm-arm/arch-iop32x/memory.h      |   7 +-
 include/asm-arm/arch-iop32x/system.h      |  17 ++---
 include/asm-arm/arch-iop32x/timex.h       |   5 +-
 include/asm-arm/arch-iop32x/uncompress.h  |   7 +-
 include/asm-arm/arch-iop32x/vmalloc.h     |  15 +---
 include/asm-arm/arch-iop33x/debug-macro.S |  12 ++--
 include/asm-arm/arch-iop33x/dma.h         |   4 +-
 include/asm-arm/arch-iop33x/entry-macro.S |   8 +--
 include/asm-arm/arch-iop33x/hardware.h    |  19 +++---
 include/asm-arm/arch-iop33x/io.h          |  12 ++--
 include/asm-arm/arch-iop33x/iop331.h      | 110 ------------------------------
 include/asm-arm/arch-iop33x/iop33x.h      |  33 +++++++++
 include/asm-arm/arch-iop33x/iq80331.h     |   8 +--
 include/asm-arm/arch-iop33x/iq80332.h     |   8 +--
 include/asm-arm/arch-iop33x/irqs.h        |  86 +++++++++++------------
 include/asm-arm/arch-iop33x/memory.h      |   6 +-
 include/asm-arm/arch-iop33x/system.h      |  17 ++---
 include/asm-arm/arch-iop33x/timex.h       |   3 +-
 include/asm-arm/arch-iop33x/uncompress.h  |   9 +--
 include/asm-arm/arch-iop33x/vmalloc.h     |  15 +---
 41 files changed, 376 insertions(+), 558 deletions(-)
 delete mode 100644 include/asm-arm/arch-iop32x/iop321.h
 create mode 100644 include/asm-arm/arch-iop32x/iop32x.h
 delete mode 100644 include/asm-arm/arch-iop33x/iop331.h
 create mode 100644 include/asm-arm/arch-iop33x/iop33x.h

(limited to 'include')

diff --git a/arch/arm/mach-iop32x/Kconfig b/arch/arm/mach-iop32x/Kconfig
index ff8a77a8866e..d7abfaa525c3 100644
--- a/arch/arm/mach-iop32x/Kconfig
+++ b/arch/arm/mach-iop32x/Kconfig
@@ -8,13 +8,14 @@ config ARCH_IQ80321
 	bool "Enable support for IQ80321"
 	help
 	  Say Y here if you want to run your kernel on the Intel IQ80321
-	  evaluation kit for the IOP321 chipset.
+	  evaluation kit for the IOP321 processor.
 
 config ARCH_IQ31244
-	bool "Enable support for IQ31244"
+	bool "Enable support for EP80219/IQ31244"
 	help
-	  Say Y here if you want to run your kernel on the Intel IQ31244
-	  evaluation kit for the IOP321 chipset.
+	  Say Y here if you want to run your kernel on the Intel EP80219
+	  evaluation kit for the Intel 80219 processor (a IOP321 variant)
+	  or the IQ31244 evaluation kit for the IOP321 processor.
 
 endmenu
 
diff --git a/arch/arm/mach-iop32x/iq31244.c b/arch/arm/mach-iop32x/iq31244.c
index 88b77d32b0ac..be4aedfa0de6 100644
--- a/arch/arm/mach-iop32x/iq31244.c
+++ b/arch/arm/mach-iop32x/iq31244.c
@@ -98,16 +98,16 @@ ep80219_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 
 	if (slot == 0) {
 		/* CFlash */
-		irq = IRQ_IOP321_XINT1;
+		irq = IRQ_IOP32X_XINT1;
 	} else if (slot == 1) {
 		/* 82551 Pro 100 */
-		irq = IRQ_IOP321_XINT0;
+		irq = IRQ_IOP32X_XINT0;
 	} else if (slot == 2) {
 		/* PCI-X Slot */
-		irq = IRQ_IOP321_XINT3;
+		irq = IRQ_IOP32X_XINT3;
 	} else if (slot == 3) {
 		/* SATA */
-		irq = IRQ_IOP321_XINT2;
+		irq = IRQ_IOP32X_XINT2;
 	} else {
 		printk(KERN_ERR "ep80219_pci_map_irq() called for unknown "
 			"device PCI:%d:%d:%d\n", dev->bus->number,
@@ -134,18 +134,18 @@ iq31244_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 
 	if (slot == 0) {
 		/* CFlash */
-		irq = IRQ_IOP321_XINT1;
+		irq = IRQ_IOP32X_XINT1;
 	} else if (slot == 1) {
 		/* SATA */
-		irq = IRQ_IOP321_XINT2;
+		irq = IRQ_IOP32X_XINT2;
 	} else if (slot == 2) {
 		/* PCI-X Slot */
-		irq = IRQ_IOP321_XINT3;
+		irq = IRQ_IOP32X_XINT3;
 	} else if (slot == 3) {
 		/* 82546 GigE */
-		irq = IRQ_IOP321_XINT0;
+		irq = IRQ_IOP32X_XINT0;
 	} else {
-		printk(KERN_ERR "iq31244_pci_map_irq() called for unknown "
+		printk(KERN_ERR "iq31244_pci_map_irq called for unknown "
 			"device PCI:%d:%d:%d\n", dev->bus->number,
 			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
 		irq = -1;
@@ -206,7 +206,7 @@ static struct plat_serial8250_port iq31244_serial_port[] = {
 	{
 		.mapbase	= IQ31244_UART,
 		.membase	= (char *)IQ31244_UART,
-		.irq		= IRQ_IOP321_XINT1,
+		.irq		= IRQ_IOP32X_XINT1,
 		.flags		= UPF_SKIP_TEST,
 		.iotype		= UPIO_MEM,
 		.regshift	= 0,
@@ -287,7 +287,7 @@ MACHINE_START(IQ31244, "Intel IQ31244")
 	.io_pg_offst	= ((IQ31244_UART) >> 18) & 0xfffc,
 	.boot_params	= 0xa0000100,
 	.map_io		= iq31244_map_io,
-	.init_irq	= iop321_init_irq,
+	.init_irq	= iop32x_init_irq,
 	.timer		= &iq31244_timer,
 	.init_machine	= iq31244_init_machine,
 MACHINE_END
diff --git a/arch/arm/mach-iop32x/iq80321.c b/arch/arm/mach-iop32x/iq80321.c
index 3c9b86271759..1f37b5501888 100644
--- a/arch/arm/mach-iop32x/iq80321.c
+++ b/arch/arm/mach-iop32x/iq80321.c
@@ -78,19 +78,19 @@ iq80321_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 
 	if ((slot == 2 || slot == 6) && pin == 1) {
 		/* PCI-X Slot INTA */
-		irq = IRQ_IOP321_XINT2;
+		irq = IRQ_IOP32X_XINT2;
 	} else if ((slot == 2 || slot == 6) && pin == 2) {
 		/* PCI-X Slot INTA */
-		irq = IRQ_IOP321_XINT3;
+		irq = IRQ_IOP32X_XINT3;
 	} else if ((slot == 2 || slot == 6) && pin == 3) {
 		/* PCI-X Slot INTA */
-		irq = IRQ_IOP321_XINT0;
+		irq = IRQ_IOP32X_XINT0;
 	} else if ((slot == 2 || slot == 6) && pin == 4) {
 		/* PCI-X Slot INTA */
-		irq = IRQ_IOP321_XINT1;
+		irq = IRQ_IOP32X_XINT1;
 	} else if (slot == 4 || slot == 8) {
 		/* Gig-E */
-		irq = IRQ_IOP321_XINT0;
+		irq = IRQ_IOP32X_XINT0;
 	} else {
 		printk(KERN_ERR "iq80321_pci_map_irq() called for unknown "
 			"device PCI:%d:%d:%d\n", dev->bus->number,
@@ -148,7 +148,7 @@ static struct plat_serial8250_port iq80321_serial_port[] = {
 	{
 		.mapbase	= IQ80321_UART,
 		.membase	= (char *)IQ80321_UART,
-		.irq		= IRQ_IOP321_XINT1,
+		.irq		= IRQ_IOP32X_XINT1,
 		.flags		= UPF_SKIP_TEST,
 		.iotype		= UPIO_MEM,
 		.regshift	= 0,
@@ -187,7 +187,7 @@ MACHINE_START(IQ80321, "Intel IQ80321")
 	.io_pg_offst	= ((IQ80321_UART) >> 18) & 0xfffc,
 	.boot_params	= 0xa0000100,
 	.map_io		= iq80321_map_io,
-	.init_irq	= iop321_init_irq,
+	.init_irq	= iop32x_init_irq,
 	.timer		= &iq80321_timer,
 	.init_machine	= iq80321_init_machine,
 MACHINE_END
diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index ff049e02f5f7..21294be5a369 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -1,5 +1,5 @@
 /*
- * linux/arch/arm/mach-iop32x/irq.c
+ * arch/arm/mach-iop32x/irq.c
  *
  * Generic IOP32X IRQ handling functionality
  *
@@ -9,76 +9,66 @@
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
- * Added IOP3XX chipset and IQ80321 board masking code.
- *
  */
+
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/list.h>
-
 #include <asm/mach/irq.h>
 #include <asm/irq.h>
 #include <asm/hardware.h>
-
 #include <asm/mach-types.h>
 
-static u32 iop321_mask /* = 0 */;
+static u32 iop32x_mask;
 
 static inline void intctl_write(u32 val)
 {
 	iop3xx_cp6_enable();
-	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
+	asm volatile("mcr p6, 0, %0, c0, c0, 0" : : "r" (val));
 	iop3xx_cp6_disable();
 }
 
 static inline void intstr_write(u32 val)
 {
 	iop3xx_cp6_enable();
-	asm volatile("mcr p6,0,%0,c4,c0,0"::"r" (val));
+	asm volatile("mcr p6, 0, %0, c4, c0, 0" : : "r" (val));
 	iop3xx_cp6_disable();
 }
 
 static void
-iop321_irq_mask (unsigned int irq)
+iop32x_irq_mask(unsigned int irq)
 {
-
-	iop321_mask &= ~(1 << irq);
-
-	intctl_write(iop321_mask);
+	iop32x_mask &= ~(1 << irq);
+	intctl_write(iop32x_mask);
 }
 
 static void
-iop321_irq_unmask (unsigned int irq)
+iop32x_irq_unmask(unsigned int irq)
 {
-	iop321_mask |= (1 << irq);
-
-	intctl_write(iop321_mask);
+	iop32x_mask |= 1 << irq;
+	intctl_write(iop32x_mask);
 }
 
 struct irq_chip ext_chip = {
-	.name	= "IOP",
-	.ack    = iop321_irq_mask,
-	.mask   = iop321_irq_mask,
-	.unmask = iop321_irq_unmask,
+	.name	= "IOP32x",
+	.ack	= iop32x_irq_mask,
+	.mask	= iop32x_irq_mask,
+	.unmask	= iop32x_irq_unmask,
 };
 
-void __init iop321_init_irq(void)
+void __init iop32x_init_irq(void)
 {
-	unsigned int i;
+	int i;
 
-	intctl_write(0);		// disable all interrupts
-	intstr_write(0);		// treat all as IRQ
-	if(machine_is_iq80321() ||
-	   machine_is_iq31244()) 	// all interrupts are inputs to chip
+	intctl_write(0);
+	intstr_write(0);
+	if (machine_is_iq80321() ||
+	    machine_is_iq31244())
 		*IOP3XX_PCIIRSR = 0x0f;
 
-	for(i = 0; i < NR_IRQS; i++)
-	{
+	for (i = 0; i < NR_IRQS; i++) {
 		set_irq_chip(i, &ext_chip);
 		set_irq_handler(i, do_level_IRQ);
 		set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
-
 	}
 }
-
diff --git a/arch/arm/mach-iop33x/iq80331.c b/arch/arm/mach-iop33x/iq80331.c
index 6b8475da3df6..97a7b7488264 100644
--- a/arch/arm/mach-iop33x/iq80331.c
+++ b/arch/arm/mach-iop33x/iq80331.c
@@ -61,19 +61,19 @@ iq80331_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 
 	if (slot == 1 && pin == 1) {
 		/* PCI-X Slot INTA */
-		irq = IRQ_IOP331_XINT1;
+		irq = IRQ_IOP33X_XINT1;
 	} else if (slot == 1 && pin == 2) {
 		/* PCI-X Slot INTB */
-		irq = IRQ_IOP331_XINT2;
+		irq = IRQ_IOP33X_XINT2;
 	} else if (slot == 1 && pin == 3) {
 		/* PCI-X Slot INTC */
-		irq = IRQ_IOP331_XINT3;
+		irq = IRQ_IOP33X_XINT3;
 	} else if (slot == 1 && pin == 4) {
 		/* PCI-X Slot INTD */
-		irq = IRQ_IOP331_XINT0;
+		irq = IRQ_IOP33X_XINT0;
 	} else if (slot == 2) {
 		/* GigE */
-		irq = IRQ_IOP331_XINT2;
+		irq = IRQ_IOP33X_XINT2;
 	} else {
 		printk(KERN_ERR "iq80331_pci_map_irq() called for unknown "
 			"device PCI:%d:%d:%d\n", dev->bus->number,
@@ -142,7 +142,7 @@ MACHINE_START(IQ80331, "Intel IQ80331")
 	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc,
 	.boot_params	= 0x00000100,
 	.map_io		= iop3xx_map_io,
-	.init_irq	= iop331_init_irq,
+	.init_irq	= iop33x_init_irq,
 	.timer		= &iq80331_timer,
 	.init_machine	= iq80331_init_machine,
 MACHINE_END
diff --git a/arch/arm/mach-iop33x/iq80332.c b/arch/arm/mach-iop33x/iq80332.c
index 150f3fd5de0b..9887bfc1c078 100644
--- a/arch/arm/mach-iop33x/iq80332.c
+++ b/arch/arm/mach-iop33x/iq80332.c
@@ -61,19 +61,19 @@ iq80332_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 
 	if (slot == 4 && pin == 1) {
 		/* PCI-X Slot INTA */
-		irq = IRQ_IOP331_XINT0;
+		irq = IRQ_IOP33X_XINT0;
 	} else if (slot == 4 && pin == 2) {
 		/* PCI-X Slot INTB */
-		irq = IRQ_IOP331_XINT1;
+		irq = IRQ_IOP33X_XINT1;
 	} else if (slot == 4 && pin == 3) {
 		/* PCI-X Slot INTC */
-		irq = IRQ_IOP331_XINT2;
+		irq = IRQ_IOP33X_XINT2;
 	} else if (slot == 4 && pin == 4) {
 		/* PCI-X Slot INTD */
-		irq = IRQ_IOP331_XINT3;
+		irq = IRQ_IOP33X_XINT3;
 	} else if (slot == 6) {
 		/* GigE */
-		irq = IRQ_IOP331_XINT2;
+		irq = IRQ_IOP33X_XINT2;
 	} else {
 		printk(KERN_ERR "iq80332_pci_map_irq() called for unknown "
 			"device PCI:%d:%d:%d\n", dev->bus->number,
@@ -142,7 +142,7 @@ MACHINE_START(IQ80332, "Intel IQ80332")
 	.io_pg_offst	= ((0xfffff000) >> 18) & 0xfffc,
 	.boot_params	= 0x00000100,
 	.map_io		= iop3xx_map_io,
-	.init_irq	= iop331_init_irq,
+	.init_irq	= iop33x_init_irq,
 	.timer		= &iq80332_timer,
 	.init_machine	= iq80332_init_machine,
 MACHINE_END
diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
index 3c720551ac12..63304b3d0d76 100644
--- a/arch/arm/mach-iop33x/irq.c
+++ b/arch/arm/mach-iop33x/irq.c
@@ -1,5 +1,5 @@
 /*
- * linux/arch/arm/mach-iop33x/irq.c
+ * arch/arm/mach-iop33x/irq.c
  *
  * Generic IOP331 IRQ handling functionality
  *
@@ -9,51 +9,44 @@
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
- *
  */
+
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/list.h>
-
 #include <asm/mach/irq.h>
 #include <asm/irq.h>
 #include <asm/hardware.h>
-
 #include <asm/mach-types.h>
 
-static u32 iop331_mask0 = 0;
-static u32 iop331_mask1 = 0;
+static u32 iop33x_mask0;
+static u32 iop33x_mask1;
 
-static inline void intctl_write0(u32 val)
+static inline void intctl0_write(u32 val)
 {
-    // INTCTL0
 	iop3xx_cp6_enable();
-	asm volatile("mcr p6,0,%0,c0,c0,0"::"r" (val));
+	asm volatile("mcr p6, 0, %0, c0, c0, 0" : : "r" (val));
 	iop3xx_cp6_disable();
 }
 
-static inline void intctl_write1(u32 val)
+static inline void intctl1_write(u32 val)
 {
-    // INTCTL1
 	iop3xx_cp6_enable();
-    asm volatile("mcr p6,0,%0,c1,c0,0"::"r" (val));
+	asm volatile("mcr p6, 0, %0, c1, c0, 0" : : "r" (val));
 	iop3xx_cp6_disable();
 }
 
-static inline void intstr_write0(u32 val)
+static inline void intstr0_write(u32 val)
 {
-    // INTSTR0
 	iop3xx_cp6_enable();
-	asm volatile("mcr p6,0,%0,c2,c0,0"::"r" (val));
+	asm volatile("mcr p6, 0, %0, c2, c0, 0" : : "r" (val));
 	iop3xx_cp6_disable();
 }
 
-static inline void intstr_write1(u32 val)
+static inline void intstr1_write(u32 val)
 {
-    // INTSTR1
 	iop3xx_cp6_enable();
-	asm volatile("mcr p6,0,%0,c3,c0,0"::"r" (val));
+	asm volatile("mcr p6, 0, %0, c3, c0, 0" : : "r" (val));
 	iop3xx_cp6_disable();
 }
 
@@ -72,65 +65,63 @@ static inline void intsize_write(u32 val)
 }
 
 static void
-iop331_irq_mask1 (unsigned int irq)
+iop33x_irq_mask1 (unsigned int irq)
 {
-        iop331_mask0 &= ~(1 << irq);
-        intctl_write0(iop331_mask0);
+	iop33x_mask0 &= ~(1 << irq);
+	intctl0_write(iop33x_mask0);
 }
 
 static void
-iop331_irq_mask2 (unsigned int irq)
+iop33x_irq_mask2 (unsigned int irq)
 {
-        iop331_mask1 &= ~(1 << (irq - 32));
-        intctl_write1(iop331_mask1);
+	iop33x_mask1 &= ~(1 << (irq - 32));
+	intctl1_write(iop33x_mask1);
 }
 
 static void
-iop331_irq_unmask1(unsigned int irq)
+iop33x_irq_unmask1(unsigned int irq)
 {
-        iop331_mask0 |= (1 << irq);
-        intctl_write0(iop331_mask0);
+	iop33x_mask0 |= 1 << irq;
+	intctl0_write(iop33x_mask0);
 }
 
 static void
-iop331_irq_unmask2(unsigned int irq)
+iop33x_irq_unmask2(unsigned int irq)
 {
-        iop331_mask1 |= (1 << (irq - 32));
-        intctl_write1(iop331_mask1);
+	iop33x_mask1 |= (1 << (irq - 32));
+	intctl1_write(iop33x_mask1);
 }
 
-struct irq_chip iop331_irqchip1 = {
-	.name	= "IOP-1",
-	.ack    = iop331_irq_mask1,
-	.mask   = iop331_irq_mask1,
-	.unmask = iop331_irq_unmask1,
+struct irq_chip iop33x_irqchip1 = {
+	.name	= "IOP33x-1",
+	.ack	= iop33x_irq_mask1,
+	.mask	= iop33x_irq_mask1,
+	.unmask	= iop33x_irq_unmask1,
 };
 
-struct irq_chip iop331_irqchip2 = {
-	.name	= "IOP-2",
-	.ack    = iop331_irq_mask2,
-	.mask   = iop331_irq_mask2,
-	.unmask = iop331_irq_unmask2,
+struct irq_chip iop33x_irqchip2 = {
+	.name	= "IOP33x-2",
+	.ack	= iop33x_irq_mask2,
+	.mask	= iop33x_irq_mask2,
+	.unmask	= iop33x_irq_unmask2,
 };
 
-void __init iop331_init_irq(void)
+void __init iop33x_init_irq(void)
 {
-	unsigned int i;
+	int i;
 
-	intctl_write0(0);		// disable all interrupts
-    	intctl_write1(0);
-	intstr_write0(0);		// treat all as IRQ
-    	intstr_write1(0);
+	intctl0_write(0);
+	intctl1_write(0);
+	intstr0_write(0);
+	intstr1_write(0);
 	intbase_write(0);
 	intsize_write(1);
-	if(machine_is_iq80331()) 	// all interrupts are inputs to chip
+	if (machine_is_iq80331())
 		*IOP3XX_PCIIRSR = 0x0f;
 
-	for(i = 0; i < NR_IRQS; i++)
-	{
-		set_irq_chip(i, (i < 32) ? &iop331_irqchip1 : &iop331_irqchip2);
+	for (i = 0; i < NR_IRQS; i++) {
+		set_irq_chip(i, (i < 32) ? &iop33x_irqchip1 : &iop33x_irqchip2);
 		set_irq_handler(i, do_level_IRQ);
 		set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
 	}
 }
-
diff --git a/arch/arm/mach-iop33x/uart.c b/arch/arm/mach-iop33x/uart.c
index d221d4abaa87..ac297cd0276c 100644
--- a/arch/arm/mach-iop33x/uart.c
+++ b/arch/arm/mach-iop33x/uart.c
@@ -1,5 +1,5 @@
 /*
- * linux/arch/arm/mach-iop33x/uart.c
+ * arch/arm/mach-iop33x/uart.c
  *
  * Author: Dave Jiang (dave.jiang@intel.com)
  * Copyright (C) 2004 Intel Corporation.
@@ -17,7 +17,6 @@
 #include <linux/serial.h>
 #include <linux/tty.h>
 #include <linux/serial_8250.h>
-
 #include <asm/io.h>
 #include <asm/pgtable.h>
 #include <asm/page.h>
@@ -30,14 +29,14 @@
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
 
-#define IOP331_UART_XTAL 33334000
+#define IOP33X_UART_XTAL 33334000
 
 static struct plat_serial8250_port iop33x_uart0_data[] = {
 	{
-		.membase	= (char *)IOP331_UART0_VIRT,
-		.mapbase	= IOP331_UART0_PHYS,
-		.irq		= IRQ_IOP331_UART0,
-		.uartclk	= IOP331_UART_XTAL,
+		.membase	= (char *)IOP33X_UART0_VIRT,
+		.mapbase	= IOP33X_UART0_PHYS,
+		.irq		= IRQ_IOP33X_UART0,
+		.uartclk	= IOP33X_UART_XTAL,
 		.regshift	= 2,
 		.iotype		= UPIO_MEM,
 		.flags		= UPF_SKIP_TEST,
@@ -47,13 +46,13 @@ static struct plat_serial8250_port iop33x_uart0_data[] = {
 
 static struct resource iop33x_uart0_resources[] = {
 	[0] = {
-		.start	= IOP331_UART0_PHYS,
-		.end	= IOP331_UART0_PHYS + 0x3f,
+		.start	= IOP33X_UART0_PHYS,
+		.end	= IOP33X_UART0_PHYS + 0x3f,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= IRQ_IOP331_UART0,
-		.end	= IRQ_IOP331_UART0,
+		.start	= IRQ_IOP33X_UART0,
+		.end	= IRQ_IOP33X_UART0,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -71,23 +70,23 @@ struct platform_device iop33x_uart0_device = {
 
 static struct resource iop33x_uart1_resources[] = {
 	[0] = {
-		.start	= IOP331_UART1_PHYS,
-		.end	= IOP331_UART1_PHYS + 0x3f,
+		.start	= IOP33X_UART1_PHYS,
+		.end	= IOP33X_UART1_PHYS + 0x3f,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= IRQ_IOP331_UART1,
-		.end	= IRQ_IOP331_UART1,
+		.start	= IRQ_IOP33X_UART1,
+		.end	= IRQ_IOP33X_UART1,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
 
 static struct plat_serial8250_port iop33x_uart1_data[] = {
 	{
-		.membase	= (char *)IOP331_UART1_VIRT,
-		.mapbase	= IOP331_UART1_PHYS,
-		.irq		= IRQ_IOP331_UART1,
-		.uartclk	= IOP331_UART_XTAL,
+		.membase	= (char *)IOP33X_UART1_VIRT,
+		.mapbase	= IOP33X_UART1_PHYS,
+		.irq		= IRQ_IOP33X_UART1,
+		.uartclk	= IOP33X_UART_XTAL,
 		.regshift	= 2,
 		.iotype		= UPIO_MEM,
 		.flags		= UPF_SKIP_TEST,
diff --git a/arch/arm/oprofile/op_model_xscale.c b/arch/arm/oprofile/op_model_xscale.c
index 7e0cc5b15b15..6576143f2559 100644
--- a/arch/arm/oprofile/op_model_xscale.c
+++ b/arch/arm/oprofile/op_model_xscale.c
@@ -37,10 +37,10 @@
 #define XSCALE_PMU_IRQ  IRQ_XS80200_PMU
 #endif
 #ifdef CONFIG_ARCH_IOP32X
-#define XSCALE_PMU_IRQ  IRQ_IOP321_CORE_PMU
+#define XSCALE_PMU_IRQ  IRQ_IOP32X_CORE_PMU
 #endif
 #ifdef CONFIG_ARCH_IOP33X
-#define XSCALE_PMU_IRQ  IRQ_IOP331_CORE_PMU
+#define XSCALE_PMU_IRQ  IRQ_IOP33X_CORE_PMU
 #endif
 #ifdef CONFIG_ARCH_PXA
 #define XSCALE_PMU_IRQ  IRQ_PMU
@@ -88,7 +88,7 @@ static struct pmu_counter results[MAX_COUNTERS];
 /*
  * There are two versions of the PMU in current XScale processors
  * with differing register layouts and number of performance counters.
- * e.g. IOP321 is xsc1 whilst IOP331 is xsc2.
+ * e.g. IOP32x is xsc1 whilst IOP33x is xsc2.
  * We detect which register layout to use in xscale_detect_pmu()
  */
 enum { PMU_XSC1, PMU_XSC2 };
diff --git a/arch/arm/plat-iop/i2c.c b/arch/arm/plat-iop/i2c.c
index 7ae149c2e982..e99909bdba71 100644
--- a/arch/arm/plat-iop/i2c.c
+++ b/arch/arm/plat-iop/i2c.c
@@ -31,12 +31,12 @@
 #include <asm/mach/arch.h>
 
 #ifdef CONFIG_ARCH_IOP32X
-#define IRQ_IOP3XX_I2C_0	IRQ_IOP321_I2C_0
-#define IRQ_IOP3XX_I2C_1	IRQ_IOP321_I2C_1
+#define IRQ_IOP3XX_I2C_0	IRQ_IOP32X_I2C_0
+#define IRQ_IOP3XX_I2C_1	IRQ_IOP32X_I2C_1
 #endif
 #ifdef CONFIG_ARCH_IOP33X
-#define IRQ_IOP3XX_I2C_0	IRQ_IOP331_I2C_0
-#define IRQ_IOP3XX_I2C_1	IRQ_IOP331_I2C_1
+#define IRQ_IOP3XX_I2C_0	IRQ_IOP33X_I2C_0
+#define IRQ_IOP3XX_I2C_1	IRQ_IOP33X_I2C_1
 #endif
 
 static struct resource iop3xx_i2c0_resources[] = {
diff --git a/arch/arm/plat-iop/time.c b/arch/arm/plat-iop/time.c
index bed20f3669f4..06282dffbdc6 100644
--- a/arch/arm/plat-iop/time.c
+++ b/arch/arm/plat-iop/time.c
@@ -26,10 +26,10 @@
 #include <asm/mach/time.h>
 
 #ifdef CONFIG_ARCH_IOP32X
-#define IRQ_IOP3XX_TIMER0	IRQ_IOP321_TIMER0
+#define IRQ_IOP3XX_TIMER0	IRQ_IOP32X_TIMER0
 #else
 #ifdef CONFIG_ARCH_IOP33X
-#define IRQ_IOP3XX_TIMER0	IRQ_IOP331_TIMER0
+#define IRQ_IOP3XX_TIMER0	IRQ_IOP33X_TIMER0
 #endif
 #endif
 
diff --git a/include/asm-arm/arch-iop32x/debug-macro.S b/include/asm-arm/arch-iop32x/debug-macro.S
index 75ab2e0d8c67..9022b6849e23 100644
--- a/include/asm-arm/arch-iop32x/debug-macro.S
+++ b/include/asm-arm/arch-iop32x/debug-macro.S
@@ -1,18 +1,18 @@
-/* linux/include/asm-arm/arch-iop32x/debug-macro.S
+/*
+ * include/asm-arm/arch-iop32x/debug-macro.S
  *
  * Debugging macro include header
  *
- *  Copyright (C) 1994-1999 Russell King
- *  Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
+ * Copyright (C) 1994-1999 Russell King
+ * Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
-*/
+ */
 
-		.macro	addruart,rx
-		mov	\rx, #0xfe000000	@ physical
+		.macro	addruart, rx
+		mov	\rx, #0xfe000000	@ physical as well as virtual
 		orr	\rx, \rx, #0x00800000	@ location of the UART
 		.endm
 
diff --git a/include/asm-arm/arch-iop32x/dma.h b/include/asm-arm/arch-iop32x/dma.h
index 5be36676e58f..e977a9ef3160 100644
--- a/include/asm-arm/arch-iop32x/dma.h
+++ b/include/asm-arm/arch-iop32x/dma.h
@@ -1,7 +1,7 @@
 /*
- * linux/include/asm-arm/arch-iop32x/dma.h
+ * include/asm-arm/arch-iop32x/dma.h
  *
- *  Copyright (C) 2004 Intel Corp.
+ * Copyright (C) 2004 Intel Corp.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
diff --git a/include/asm-arm/arch-iop32x/entry-macro.S b/include/asm-arm/arch-iop32x/entry-macro.S
index 3497fef0b890..1500cbbd2295 100644
--- a/include/asm-arm/arch-iop32x/entry-macro.S
+++ b/include/asm-arm/arch-iop32x/entry-macro.S
@@ -3,19 +3,16 @@
  *
  * Low-level IRQ helper macros for IOP32x-based platforms
  *
- * This file is licensed under  the terms of the GNU General Public
+ * This file is licensed under the terms of the GNU General Public
  * License version 2. This program is licensed "as is" without any
  * warranty of any kind, whether express or implied.
  */
-#include <asm/arch/irqs.h>
+#include <asm/arch/iop32x.h>
 
-		.macro  disable_fiq
+		.macro	disable_fiq
 		.endm
 
-		/*
-		 * Note: only deal with normal interrupts, not FIQ
-		 */
-		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
+		.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
 		ldr	\base, =IOP3XX_REG_ADDR(0x07D8)
 		ldr	\irqstat, [\base]		@ Read IINTSRC
 		cmp	\irqstat, #0
diff --git a/include/asm-arm/arch-iop32x/hardware.h b/include/asm-arm/arch-iop32x/hardware.h
index 16d0630ab252..6a3001f2f7e0 100644
--- a/include/asm-arm/arch-iop32x/hardware.h
+++ b/include/asm-arm/arch-iop32x/hardware.h
@@ -1,8 +1,9 @@
 /*
- * linux/include/asm-arm/arch-iop32x/hardware.h
+ * include/asm-arm/arch-iop32x/hardware.h
  */
-#ifndef __ASM_ARCH_HARDWARE_H
-#define __ASM_ARCH_HARDWARE_H
+
+#ifndef __HARDWARE_H
+#define __HARDWARE_H
 
 #include <asm/types.h>
 
@@ -13,21 +14,23 @@
  * the IO resources.
  *
  * The PCI IO space is located at virtual 0xfe000000 from physical
- * 0x90000000.  The PCI BARs must be programmed with physical addresses,
- * but when we read them, we convert them to virtual addresses.  See
- * arch/arm/mach-iop3xx/iop3xx-pci.c
+ * 0x90000000. The PCI BARs must be programmed with physical addresses,
+ * but when we read them, we convert them to virtual addresses. See
+ * arch/arm/plat-iop/pci.c.
  */
-
 #define pcibios_assign_all_busses() 1
 #define PCIBIOS_MIN_IO		0x00000000
 #define PCIBIOS_MIN_MEM		0x00000000
 
+#ifndef __ASSEMBLY__
+void iop32x_init_irq(void);
+#endif
+
 
 /*
  * Generic chipset bits
- *
  */
-#include "iop321.h"
+#include "iop32x.h"
 
 /*
  * Board specific bits
@@ -35,4 +38,5 @@
 #include "iq80321.h"
 #include "iq31244.h"
 
-#endif  /* _ASM_ARCH_HARDWARE_H */
+
+#endif
diff --git a/include/asm-arm/arch-iop32x/io.h b/include/asm-arm/arch-iop32x/io.h
index 36d05ada12c4..12d9ee02cde3 100644
--- a/include/asm-arm/arch-iop32x/io.h
+++ b/include/asm-arm/arch-iop32x/io.h
@@ -1,21 +1,22 @@
 /*
- * linux/include/asm-arm/arch-iop32x/io.h
+ * include/asm-arm/arch-iop32x/io.h
  *
- *  Copyright (C) 2001  MontaVista Software, Inc.
+ * Copyright (C) 2001 MontaVista Software, Inc.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
 
-#ifndef __ASM_ARM_ARCH_IO_H
-#define __ASM_ARM_ARCH_IO_H
+#ifndef __IO_H
+#define __IO_H
 
 #include <asm/hardware.h>
 
-#define IO_SPACE_LIMIT 0xffffffff
+#define IO_SPACE_LIMIT		0xffffffff
 
 #define __io(p)			((void __iomem *)(p))
 #define __mem_pci(a)		(a)
 
+
 #endif
diff --git a/include/asm-arm/arch-iop32x/iop321.h b/include/asm-arm/arch-iop32x/iop321.h
deleted file mode 100644
index 1757222a4cad..000000000000
--- a/include/asm-arm/arch-iop32x/iop321.h
+++ /dev/null
@@ -1,86 +0,0 @@
-/*
- * linux/include/asm/arch-iop32x/iop321.h
- *
- * Intel IOP321 Chip definitions
- *
- * Author: Rory Bolt <rorybolt@pacbell.net>
- * Copyright (C) 2002 Rory Bolt
- * Copyright (C) 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef _IOP321_HW_H_
-#define _IOP321_HW_H_
-
-
-/*
- * This is needed for mixed drivers that need to work on all
- * IOP3xx variants but behave slightly differently on each.
- */
-#ifndef __ASSEMBLY__
-#define iop_is_321()		1
-#endif
-
-/*
- * IOP321 chipset registers
- */
-#define IOP321_VIRT_MEM_BASE 0xfeffe000  /* chip virtual mem address*/
-#define IOP321_PHYS_MEM_BASE 0xffffe000  /* chip physical memory address */
-#define IOP321_REG_ADDR(reg) (IOP321_VIRT_MEM_BASE | (reg))
-
-/* Reserved 0x00000000 through 0x000000FF */
-
-/* Address Translation Unit 0x00000100 through 0x000001FF */
-
-/* Messaging Unit 0x00000300 through 0x000003FF */
-
-/* DMA Controller 0x00000400 through 0x000004FF */
-
-/* Memory controller 0x00000500 through 0x0005FF */
-
-/* Peripheral bus interface unit 0x00000680 through 0x0006FF */
-
-/* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
-
-/* Internal arbitration unit 0x00000780 through 0x0007BF */
-#define IOP321_IACR	(volatile u32 *)IOP321_REG_ADDR(0x00000780)
-#define IOP321_MTTR1	(volatile u32 *)IOP321_REG_ADDR(0x00000784)
-#define IOP321_MTTR2	(volatile u32 *)IOP321_REG_ADDR(0x00000788)
-
-/* General Purpose I/O Registers */
-#define IOP321_GPOE       (volatile u32 *)IOP321_REG_ADDR(0x000007C4)
-#define IOP321_GPID       (volatile u32 *)IOP321_REG_ADDR(0x000007C8)
-#define IOP321_GPOD       (volatile u32 *)IOP321_REG_ADDR(0x000007CC)
-
-/* Interrupt Controller */
-#define IOP321_INTCTL     (volatile u32 *)IOP321_REG_ADDR(0x000007D0)
-#define IOP321_INTSTR     (volatile u32 *)IOP321_REG_ADDR(0x000007D4)
-#define IOP321_IINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007D8)
-#define IOP321_FINTSRC    (volatile u32 *)IOP321_REG_ADDR(0x000007DC)
-
-/* Application accelerator unit 0x00000800 - 0x000008FF */
-
-/* SSP serial port unit 0x00001600 - 0x0000167F */
-/* I2C bus interface unit 0x00001680 - 0x000016FF */
-
-/* for I2C bit defs see drivers/i2c/i2c-iop3xx.h */
-
-/*
- * Peripherals that are shared between the iop32x and iop33x but
- * located at different addresses.
- */
-#define IOP3XX_GPIO_REG(reg)   (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07c0 + (reg))
-#define IOP3XX_TIMER_REG(reg)  (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07e0 + (reg))
-
-#include <asm/hardware/iop3xx.h>
-
-
-#ifndef __ASSEMBLY__
-extern void iop321_init_irq(void);
-extern void iop321_time_init(void);
-#endif
-
-#endif // _IOP321_HW_H_
diff --git a/include/asm-arm/arch-iop32x/iop32x.h b/include/asm-arm/arch-iop32x/iop32x.h
new file mode 100644
index 000000000000..4bbd85f3ed2a
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/iop32x.h
@@ -0,0 +1,28 @@
+/*
+ * include/asm-arm/arch-iop32x/iop32x.h
+ *
+ * Intel IOP32X Chip definitions
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __IOP32X_H
+#define __IOP32X_H
+
+/*
+ * Peripherals that are shared between the iop32x and iop33x but
+ * located at different addresses.
+ */
+#define IOP3XX_GPIO_REG(reg)	(IOP3XX_PERIPHERAL_VIRT_BASE + 0x07c0 + (reg))
+#define IOP3XX_TIMER_REG(reg)	(IOP3XX_PERIPHERAL_VIRT_BASE + 0x07e0 + (reg))
+
+#include <asm/hardware/iop3xx.h>
+
+
+#endif
diff --git a/include/asm-arm/arch-iop32x/iq31244.h b/include/asm-arm/arch-iop32x/iq31244.h
index cf2d2343398d..fff4eafa1f6b 100644
--- a/include/asm-arm/arch-iop32x/iq31244.h
+++ b/include/asm-arm/arch-iop32x/iq31244.h
@@ -1,11 +1,11 @@
 /*
- * linux/include/asm/arch-iop32x/iq31244.h
+ * include/asm-arm/arch-iop32x/iq31244.h
  *
  * Intel IQ31244 evaluation board registers
  */
 
-#ifndef _IQ31244_H_
-#define _IQ31244_H_
+#ifndef __IQ31244_H
+#define __IQ31244_H
 
 #define IQ31244_UART		0xfe800000	/* UART #1 */
 #define IQ31244_7SEG_1		0xfe840000	/* 7-Segment MSB */
@@ -14,4 +14,4 @@
 #define IQ31244_BATT_STAT	0xfe8f0000	/* Battery Status */
 
 
-#endif	// _IQ31244_H_
+#endif
diff --git a/include/asm-arm/arch-iop32x/iq80321.h b/include/asm-arm/arch-iop32x/iq80321.h
index 55d70f49b7fd..eb69db9b9a06 100644
--- a/include/asm-arm/arch-iop32x/iq80321.h
+++ b/include/asm-arm/arch-iop32x/iq80321.h
@@ -1,11 +1,11 @@
 /*
- * linux/include/asm/arch-iop32x/iq80321.h
+ * include/asm-arm/arch-iop32x/iq80321.h
  *
  * Intel IQ80321 evaluation board registers
  */
 
-#ifndef _IQ80321_H_
-#define _IQ80321_H_
+#ifndef __IQ80321_H
+#define __IQ80321_H
 
 #define IQ80321_UART		0xfe800000	/* UART #1 */
 #define IQ80321_7SEG_1		0xfe840000	/* 7-Segment MSB */
@@ -14,4 +14,4 @@
 #define IQ80321_BATT_STAT	0xfe8f0000	/* Battery Status */
 
 
-#endif	// _IQ80321_H_
+#endif
diff --git a/include/asm-arm/arch-iop32x/irqs.h b/include/asm-arm/arch-iop32x/irqs.h
index a48327ced92e..bbaef873afce 100644
--- a/include/asm-arm/arch-iop32x/irqs.h
+++ b/include/asm-arm/arch-iop32x/irqs.h
@@ -1,5 +1,5 @@
 /*
- * linux/include/asm-arm/arch-iop32x/irqs.h
+ * include/asm-arm/arch-iop32x/irqs.h
  *
  * Author:	Rory Bolt <rorybolt@pacbell.net>
  * Copyright:	(C) 2002 Rory Bolt
@@ -7,44 +7,44 @@
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
  */
-#ifndef _IRQS_H_
-#define _IRQS_H_
+
+#ifndef __IRQS_H
+#define __IRQS_H
 
 /*
  * IOP80321 chipset interrupts
  */
-#define IRQ_IOP321_DMA0_EOT	0
-#define IRQ_IOP321_DMA0_EOC	1
-#define IRQ_IOP321_DMA1_EOT	2
-#define IRQ_IOP321_DMA1_EOC	3
-#define IRQ_IOP321_AA_EOT	6
-#define IRQ_IOP321_AA_EOC	7
-#define IRQ_IOP321_CORE_PMON	8
-#define IRQ_IOP321_TIMER0	9
-#define IRQ_IOP321_TIMER1	10
-#define IRQ_IOP321_I2C_0	11
-#define IRQ_IOP321_I2C_1	12
-#define IRQ_IOP321_MESSAGING	13
-#define IRQ_IOP321_ATU_BIST	14
-#define IRQ_IOP321_PERFMON	15
-#define IRQ_IOP321_CORE_PMU	16
-#define IRQ_IOP321_BIU_ERR	17
-#define IRQ_IOP321_ATU_ERR	18
-#define IRQ_IOP321_MCU_ERR	19
-#define IRQ_IOP321_DMA0_ERR	20
-#define IRQ_IOP321_DMA1_ERR	21
-#define IRQ_IOP321_AA_ERR	23
-#define IRQ_IOP321_MSG_ERR	24
-#define IRQ_IOP321_SSP		25
-#define IRQ_IOP321_XINT0	27
-#define IRQ_IOP321_XINT1	28
-#define IRQ_IOP321_XINT2	29
-#define IRQ_IOP321_XINT3	30
-#define IRQ_IOP321_HPI		31
+#define IRQ_IOP32X_DMA0_EOT	0
+#define IRQ_IOP32X_DMA0_EOC	1
+#define IRQ_IOP32X_DMA1_EOT	2
+#define IRQ_IOP32X_DMA1_EOC	3
+#define IRQ_IOP32X_AA_EOT	6
+#define IRQ_IOP32X_AA_EOC	7
+#define IRQ_IOP32X_CORE_PMON	8
+#define IRQ_IOP32X_TIMER0	9
+#define IRQ_IOP32X_TIMER1	10
+#define IRQ_IOP32X_I2C_0	11
+#define IRQ_IOP32X_I2C_1	12
+#define IRQ_IOP32X_MESSAGING	13
+#define IRQ_IOP32X_ATU_BIST	14
+#define IRQ_IOP32X_PERFMON	15
+#define IRQ_IOP32X_CORE_PMU	16
+#define IRQ_IOP32X_BIU_ERR	17
+#define IRQ_IOP32X_ATU_ERR	18
+#define IRQ_IOP32X_MCU_ERR	19
+#define IRQ_IOP32X_DMA0_ERR	20
+#define IRQ_IOP32X_DMA1_ERR	21
+#define IRQ_IOP32X_AA_ERR	23
+#define IRQ_IOP32X_MSG_ERR	24
+#define IRQ_IOP32X_SSP		25
+#define IRQ_IOP32X_XINT0	27
+#define IRQ_IOP32X_XINT1	28
+#define IRQ_IOP32X_XINT2	29
+#define IRQ_IOP32X_XINT3	30
+#define IRQ_IOP32X_HPI		31
 
 #define NR_IRQS			32
 
 
-#endif // _IRQ_H_
+#endif
diff --git a/include/asm-arm/arch-iop32x/memory.h b/include/asm-arm/arch-iop32x/memory.h
index 4c64d9e7229b..764cd3f0d416 100644
--- a/include/asm-arm/arch-iop32x/memory.h
+++ b/include/asm-arm/arch-iop32x/memory.h
@@ -1,9 +1,9 @@
 /*
- * linux/include/asm-arm/arch-iop32x/memory.h
+ * include/asm-arm/arch-iop32x/memory.h
  */
 
-#ifndef __ASM_ARCH_MEMORY_H
-#define __ASM_ARCH_MEMORY_H
+#ifndef __MEMORY_H
+#define __MEMORY_H
 
 #include <asm/hardware.h>
 
@@ -19,7 +19,6 @@
  * bus_to_virt: Used to convert an address for DMA operations
  *		to an address that the kernel can use.
  */
-
 #define __virt_to_bus(x)	(((__virt_to_phys(x)) & ~(*IOP3XX_IATVR2)) | ((*IOP3XX_IABAR2) & 0xfffffff0))
 #define __bus_to_virt(x)	(__phys_to_virt(((x) & ~(*IOP3XX_IALR2)) | ( *IOP3XX_IATVR2)))
 
diff --git a/include/asm-arm/arch-iop32x/system.h b/include/asm-arm/arch-iop32x/system.h
index 1ac207a0d52e..c65ede3e627a 100644
--- a/include/asm-arm/arch-iop32x/system.h
+++ b/include/asm-arm/arch-iop32x/system.h
@@ -1,7 +1,7 @@
 /*
- * linux/include/asm-arm/arch-iop32x/system.h
+ * include/asm-arm/arch-iop32x/system.h
  *
- *  Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2001 MontaVista Software, Inc.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -13,17 +13,10 @@ static inline void arch_idle(void)
 	cpu_do_idle();
 }
 
-
 static inline void arch_reset(char mode)
 {
-    *IOP3XX_PCSR = 0x30;
+	*IOP3XX_PCSR = 0x30;
 
-	if ( 1 && mode == 's') {
-		/* Jump into ROM at address 0 */
-		cpu_reset(0);
-	} else {
-		/* No on-chip reset capability */
-		cpu_reset(0);
-	}
+	/* Jump into ROM at address 0 */
+	cpu_reset(0);
 }
-
diff --git a/include/asm-arm/arch-iop32x/timex.h b/include/asm-arm/arch-iop32x/timex.h
index 328f37282c3e..9934b087311b 100644
--- a/include/asm-arm/arch-iop32x/timex.h
+++ b/include/asm-arm/arch-iop32x/timex.h
@@ -1,8 +1,9 @@
 /*
- * linux/include/asm-arm/arch-iop32x/timex.h
+ * include/asm-arm/arch-iop32x/timex.h
  *
- * IOP3xx architecture timex specifications
+ * IOP32x architecture timex specifications
  */
+
 #include <asm/hardware.h>
 
 #define CLOCK_TICK_RATE		(100 * HZ)
diff --git a/include/asm-arm/arch-iop32x/uncompress.h b/include/asm-arm/arch-iop32x/uncompress.h
index 4a85f20c796f..e64f52bf2bce 100644
--- a/include/asm-arm/arch-iop32x/uncompress.h
+++ b/include/asm-arm/arch-iop32x/uncompress.h
@@ -1,6 +1,7 @@
 /*
- *  linux/include/asm-arm/arch-iop32x/uncompress.h
+ * include/asm-arm/arch-iop32x/uncompress.h
  */
+
 #include <asm/types.h>
 #include <asm/mach-types.h>
 #include <linux/serial_reg.h>
@@ -8,13 +9,13 @@
 
 static volatile u8 *uart_base;
 
-#define TX_DONE (UART_LSR_TEMT|UART_LSR_THRE)
+#define TX_DONE		(UART_LSR_TEMT | UART_LSR_THRE)
 
 static inline void putc(char c)
 {
 	while ((uart_base[UART_LSR] & TX_DONE) != TX_DONE)
 		barrier();
-	*uart_base = c;
+	uart_base[UART_TX] = c;
 }
 
 static inline void flush(void)
diff --git a/include/asm-arm/arch-iop32x/vmalloc.h b/include/asm-arm/arch-iop32x/vmalloc.h
index 8492e1708a63..0a70baa19517 100644
--- a/include/asm-arm/arch-iop32x/vmalloc.h
+++ b/include/asm-arm/arch-iop32x/vmalloc.h
@@ -1,16 +1,5 @@
 /*
- * linux/include/asm-arm/arch-iop32x/vmalloc.h
+ * include/asm-arm/arch-iop32x/vmalloc.h
  */
 
-/*
- * Just any arbitrary offset to the start of the vmalloc VM area: the
- * current 8MB value just means that there will be a 8MB "hole" after the
- * physical memory until the kernel virtual memory starts.  That means that
- * any out-of-bounds memory accesses will hopefully be caught.
- * The vmalloc() routines leaves a hole of 4kB between each vmalloced
- * area for the same reason. ;)
- */
-//#define VMALLOC_END       (0xe8000000)
-/* increase usable physical RAM to ~992M per RMK */
-#define VMALLOC_END       (0xfe000000)
-
+#define VMALLOC_END	0xfe000000
diff --git a/include/asm-arm/arch-iop33x/debug-macro.S b/include/asm-arm/arch-iop33x/debug-macro.S
index b647edff475d..9e7132ebe6a7 100644
--- a/include/asm-arm/arch-iop33x/debug-macro.S
+++ b/include/asm-arm/arch-iop33x/debug-macro.S
@@ -1,17 +1,17 @@
-/* linux/include/asm-arm/arch-iop33x/debug-macro.S
+/*
+ * include/asm-arm/arch-iop33x/debug-macro.S
  *
  * Debugging macro include header
  *
- *  Copyright (C) 1994-1999 Russell King
- *  Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
+ * Copyright (C) 1994-1999 Russell King
+ * Moved from linux/arch/arm/kernel/debug.S by Ben Dooks
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
-*/
+ */
 
-		.macro	addruart,rx
+		.macro	addruart, rx
 		mrc	p15, 0, \rx, c1, c0
 		tst	\rx, #1			@ mmu enabled?
 		moveq	\rx, #0xff000000	@ physical
diff --git a/include/asm-arm/arch-iop33x/dma.h b/include/asm-arm/arch-iop33x/dma.h
index d577ca59f4b0..b7775fdc5ad3 100644
--- a/include/asm-arm/arch-iop33x/dma.h
+++ b/include/asm-arm/arch-iop33x/dma.h
@@ -1,7 +1,7 @@
 /*
- * linux/include/asm-arm/arch-iop33x/dma.h
+ * include/asm-arm/arch-iop33x/dma.h
  *
- *  Copyright (C) 2004 Intel Corp.
+ * Copyright (C) 2004 Intel Corp.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
diff --git a/include/asm-arm/arch-iop33x/entry-macro.S b/include/asm-arm/arch-iop33x/entry-macro.S
index 4750e98e9b4a..92b791702e34 100644
--- a/include/asm-arm/arch-iop33x/entry-macro.S
+++ b/include/asm-arm/arch-iop33x/entry-macro.S
@@ -3,16 +3,16 @@
  *
  * Low-level IRQ helper macros for IOP33x-based platforms
  *
- * This file is licensed under  the terms of the GNU General Public
+ * This file is licensed under the terms of the GNU General Public
  * License version 2. This program is licensed "as is" without any
  * warranty of any kind, whether express or implied.
  */
-#include <asm/arch/irqs.h>
+#include <asm/arch/iop33x.h>
 
-		.macro  disable_fiq
+		.macro	disable_fiq
 		.endm
 
-		.macro  get_irqnr_and_base, irqnr, irqstat, base, tmp
+		.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
 		ldr	\base, =IOP3XX_REG_ADDR(0x07C8)
 		ldr	\irqstat, [\base]		@ Read IINTVEC
 		cmp	\irqstat, #0
diff --git a/include/asm-arm/arch-iop33x/hardware.h b/include/asm-arm/arch-iop33x/hardware.h
index 3ebfdc6fea99..0659cf94d040 100644
--- a/include/asm-arm/arch-iop33x/hardware.h
+++ b/include/asm-arm/arch-iop33x/hardware.h
@@ -1,8 +1,9 @@
 /*
- * linux/include/asm-arm/arch-iop33x/hardware.h
+ * include/asm-arm/arch-iop33x/hardware.h
  */
-#ifndef __ASM_ARCH_HARDWARE_H
-#define __ASM_ARCH_HARDWARE_H
+
+#ifndef __HARDWARE_H
+#define __HARDWARE_H
 
 #include <asm/types.h>
 
@@ -15,14 +16,15 @@
  * The PCI IO space is located at virtual 0xfe000000 from physical
  * 0x90000000.  The PCI BARs must be programmed with physical addresses,
  * but when we read them, we convert them to virtual addresses.  See
- * arch/arm/mach-iop33x/pci.c
+ * arch/arm/mach-iop3xx/iop3xx-pci.c
  */
-
-#define pcibios_assign_all_busses() 1
+#define pcibios_assign_all_busses()	1
 #define PCIBIOS_MIN_IO		0x00000000
 #define PCIBIOS_MIN_MEM		0x00000000
 
 #ifndef __ASSEMBLY__
+void iop33x_init_irq(void);
+
 extern struct platform_device iop33x_uart0_device;
 extern struct platform_device iop33x_uart1_device;
 #endif
@@ -32,7 +34,7 @@ extern struct platform_device iop33x_uart1_device;
  * Generic chipset bits
  *
  */
-#include "iop331.h"
+#include "iop33x.h"
 
 /*
  * Board specific bits
@@ -40,4 +42,5 @@ extern struct platform_device iop33x_uart1_device;
 #include "iq80331.h"
 #include "iq80332.h"
 
-#endif  /* _ASM_ARCH_HARDWARE_H */
+
+#endif
diff --git a/include/asm-arm/arch-iop33x/io.h b/include/asm-arm/arch-iop33x/io.h
index a9949d5d4953..c017402bab96 100644
--- a/include/asm-arm/arch-iop33x/io.h
+++ b/include/asm-arm/arch-iop33x/io.h
@@ -1,21 +1,21 @@
 /*
- * linux/include/asm-arm/arch-iop33x/io.h
+ * include/asm-arm/arch-iop33x/io.h
  *
- *  Copyright (C) 2001  MontaVista Software, Inc.
+ * Copyright (C) 2001  MontaVista Software, Inc.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
 
-#ifndef __ASM_ARM_ARCH_IO_H
-#define __ASM_ARM_ARCH_IO_H
+#ifndef __IO_H
+#define __IO_H
 
 #include <asm/hardware.h>
 
-#define IO_SPACE_LIMIT 0xffffffff
-
+#define IO_SPACE_LIMIT		0xffffffff
 #define __io(p)			((void __iomem *)(p))
 #define __mem_pci(a)		(a)
 
+
 #endif
diff --git a/include/asm-arm/arch-iop33x/iop331.h b/include/asm-arm/arch-iop33x/iop331.h
deleted file mode 100644
index 8c7ec583615f..000000000000
--- a/include/asm-arm/arch-iop33x/iop331.h
+++ /dev/null
@@ -1,110 +0,0 @@
-/*
- * linux/include/asm/arch-iop33x/iop331.h
- *
- * Intel IOP331 Chip definitions
- *
- * Author: Dave Jiang (dave.jiang@intel.com)
- * Copyright (C) 2003, 2004 Intel Corp.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef _IOP331_HW_H_
-#define _IOP331_HW_H_
-
-
-/*
- * This is needed for mixed drivers that need to work on all
- * IOP3xx variants but behave slightly differently on each.
- */
-#ifndef __ASSEMBLY__
-#define	iop_is_331()		1
-#endif
-
-/*
- * IOP331 chipset registers
- */
-#define IOP331_VIRT_MEM_BASE  0xfeffe000  /* chip virtual mem address*/
-#define IOP331_PHYS_MEM_BASE  0xffffe000  /* chip physical memory address */
-#define IOP331_REG_ADDR(reg) (IOP331_VIRT_MEM_BASE | (reg))
-
-/* Reserved 0x00000000 through 0x000000FF */
-
-/* Address Translation Unit 0x00000100 through 0x000001FF */
-
-/* Messaging Unit 0x00000300 through 0x000003FF */
-
-/* DMA Controller 0x00000400 through 0x000004FF */
-
-/* Memory controller 0x00000500 through 0x0005FF */
-
-/* Peripheral bus interface unit 0x00000680 through 0x0006FF */
-
-/* Peripheral performance monitoring unit 0x00000700 through 0x00077F */
-/* Internal arbitration unit 0x00000780 through 0x0007BF */
-
-/* Interrupt Controller */
-#define IOP331_INTCTL0    (volatile u32 *)IOP331_REG_ADDR(0x00000790)
-#define IOP331_INTCTL1    (volatile u32 *)IOP331_REG_ADDR(0x00000794)
-#define IOP331_INTSTR0    (volatile u32 *)IOP331_REG_ADDR(0x00000798)
-#define IOP331_INTSTR1    (volatile u32 *)IOP331_REG_ADDR(0x0000079C)
-#define IOP331_IINTSRC0   (volatile u32 *)IOP331_REG_ADDR(0x000007A0)
-#define IOP331_IINTSRC1   (volatile u32 *)IOP331_REG_ADDR(0x000007A4)
-#define IOP331_FINTSRC0   (volatile u32 *)IOP331_REG_ADDR(0x000007A8)
-#define IOP331_FINTSRC1   (volatile u32 *)IOP331_REG_ADDR(0x000007AC)
-#define IOP331_IPR0       (volatile u32 *)IOP331_REG_ADDR(0x000007B0)
-#define IOP331_IPR1       (volatile u32 *)IOP331_REG_ADDR(0x000007B4)
-#define IOP331_IPR2       (volatile u32 *)IOP331_REG_ADDR(0x000007B8)
-#define IOP331_IPR3       (volatile u32 *)IOP331_REG_ADDR(0x000007BC)
-#define IOP331_INTBASE    (volatile u32 *)IOP331_REG_ADDR(0x000007C0)
-#define IOP331_INTSIZE    (volatile u32 *)IOP331_REG_ADDR(0x000007C4)
-#define IOP331_IINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007C8)
-#define IOP331_FINTVEC    (volatile u32 *)IOP331_REG_ADDR(0x000007CC)
-
-
-/* Application accelerator unit 0x00000800 - 0x000008FF */
-
-#define IOP331_SPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C0)
-#define IOP331_PPDSCR	  (volatile u32 *)IOP331_REG_ADDR(0x000015C8)
-/* SSP serial port unit 0x00001600 - 0x0000167F */
-
-/* I2C bus interface unit 0x00001680 - 0x000016FF */
-
-/* 0x00001700 through 0x0000172C  UART 0 */
-
-/* Reserved 0x00001730 through 0x0000173F */
-
-/* 0x00001740 through 0x0000176C UART 1 */
-
-#define IOP331_UART0_PHYS  (IOP331_PHYS_MEM_BASE | 0x00001700)	/* UART #1 physical */
-#define IOP331_UART1_PHYS  (IOP331_PHYS_MEM_BASE | 0x00001740)	/* UART #2 physical */
-#define IOP331_UART0_VIRT  (IOP331_VIRT_MEM_BASE | 0x00001700) /* UART #1 virtual addr */
-#define IOP331_UART1_VIRT  (IOP331_VIRT_MEM_BASE | 0x00001740) /* UART #2 virtual addr */
-
-/* Reserved 0x00001770 through 0x0000177F */
-
-/* General Purpose I/O Registers */
-#define IOP331_GPOE       (volatile u32 *)IOP331_REG_ADDR(0x00001780)
-#define IOP331_GPID       (volatile u32 *)IOP331_REG_ADDR(0x00001784)
-#define IOP331_GPOD       (volatile u32 *)IOP331_REG_ADDR(0x00001788)
-
-/* Reserved 0x0000178c through 0x000019ff */
-
-/*
- * Peripherals that are shared between the iop32x and iop33x but
- * located at different addresses.
- */
-#define IOP3XX_GPIO_REG(reg)   (IOP3XX_PERIPHERAL_VIRT_BASE + 0x1780 + (reg))
-#define IOP3XX_TIMER_REG(reg)  (IOP3XX_PERIPHERAL_VIRT_BASE + 0x07d0 + (reg))
-
-#include <asm/hardware/iop3xx.h>
-
-
-#ifndef __ASSEMBLY__
-extern void iop331_init_irq(void);
-extern void iop331_time_init(void);
-#endif
-
-#endif // _IOP331_HW_H_
diff --git a/include/asm-arm/arch-iop33x/iop33x.h b/include/asm-arm/arch-iop33x/iop33x.h
new file mode 100644
index 000000000000..7ac6e93db5ff
--- /dev/null
+++ b/include/asm-arm/arch-iop33x/iop33x.h
@@ -0,0 +1,33 @@
+/*
+ * include/asm-arm/arch-iop33x/iop33x.h
+ *
+ * Intel IOP33X Chip definitions
+ *
+ * Author: Dave Jiang (dave.jiang@intel.com)
+ * Copyright (C) 2003, 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __IOP33X_H
+#define __IOP33X_H
+
+/*
+ * Peripherals that are shared between the iop32x and iop33x but
+ * located at different addresses.
+ */
+#define IOP3XX_GPIO_REG(reg)	(IOP3XX_PERIPHERAL_VIRT_BASE + 0x1780 + (reg))
+#define IOP3XX_TIMER_REG(reg)	(IOP3XX_PERIPHERAL_VIRT_BASE + 0x07d0 + (reg))
+
+#include <asm/hardware/iop3xx.h>
+
+/* UARTs  */
+#define IOP33X_UART0_PHYS	(IOP3XX_PERIPHERAL_PHYS_BASE + 0x1700)
+#define IOP33X_UART0_VIRT	(IOP3XX_PERIPHERAL_VIRT_BASE + 0x1700)
+#define IOP33X_UART1_PHYS	(IOP3XX_PERIPHERAL_PHYS_BASE + 0x1740)
+#define IOP33X_UART1_VIRT	(IOP3XX_PERIPHERAL_VIRT_BASE + 0x1740)
+
+
+#endif
diff --git a/include/asm-arm/arch-iop33x/iq80331.h b/include/asm-arm/arch-iop33x/iq80331.h
index 186762bf8944..79b9302017ea 100644
--- a/include/asm-arm/arch-iop33x/iq80331.h
+++ b/include/asm-arm/arch-iop33x/iq80331.h
@@ -1,11 +1,11 @@
 /*
- * linux/include/asm/arch-iop33x/iq80331.h
+ * include/asm-arm/arch-iop33x/iq80331.h
  *
  * Intel IQ80331 evaluation board registers
  */
 
-#ifndef _IQ80331_H_
-#define _IQ80331_H_
+#ifndef __IQ80331_H
+#define __IQ80331_H
 
 #define IQ80331_7SEG_1		0xce840000	/* 7-Segment MSB */
 #define IQ80331_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
@@ -13,4 +13,4 @@
 #define IQ80331_BATT_STAT	0xce8f0000	/* Battery Status */
 
 
-#endif	// _IQ80331_H_
+#endif
diff --git a/include/asm-arm/arch-iop33x/iq80332.h b/include/asm-arm/arch-iop33x/iq80332.h
index 2a5d4ee01df9..053165629492 100644
--- a/include/asm-arm/arch-iop33x/iq80332.h
+++ b/include/asm-arm/arch-iop33x/iq80332.h
@@ -1,11 +1,11 @@
 /*
- * linux/include/asm/arch-iop33x/iq80332.h
+ * include/asm-arm/arch-iop33x/iq80332.h
  *
  * Intel IQ80332 evaluation board registers
  */
 
-#ifndef _IQ80332_H_
-#define _IQ80332_H_
+#ifndef __IQ80332_H
+#define __IQ80332_H
 
 #define IQ80332_7SEG_1		0xce840000	/* 7-Segment MSB */
 #define IQ80332_7SEG_0		0xce850000	/* 7-Segment LSB (WO) */
@@ -13,4 +13,4 @@
 #define IQ80332_BATT_STAT	0xce8f0000	/* Battery Status */
 
 
-#endif	// _IQ80332_H_
+#endif
diff --git a/include/asm-arm/arch-iop33x/irqs.h b/include/asm-arm/arch-iop33x/irqs.h
index a875404a07fc..d045f8403396 100644
--- a/include/asm-arm/arch-iop33x/irqs.h
+++ b/include/asm-arm/arch-iop33x/irqs.h
@@ -1,5 +1,5 @@
 /*
- * linux/include/asm-arm/arch-iop33x/irqs.h
+ * include/asm-arm/arch-iop33x/irqs.h
  *
  * Author:	Dave Jiang (dave.jiang@intel.com)
  * Copyright:	(C) 2003 Intel Corp.
@@ -7,54 +7,54 @@
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
  */
-#ifndef _IRQS_H_
-#define _IRQS_H_
+
+#ifndef __IRQS_H
+#define __IRQS_H
 
 /*
  * IOP80331 chipset interrupts
  */
-#define IRQ_IOP331_DMA0_EOT	0
-#define IRQ_IOP331_DMA0_EOC	1
-#define IRQ_IOP331_DMA1_EOT	2
-#define IRQ_IOP331_DMA1_EOC	3
-#define IRQ_IOP331_AA_EOT	6
-#define IRQ_IOP331_AA_EOC	7
-#define IRQ_IOP331_TIMER0	8
-#define IRQ_IOP331_TIMER1	9
-#define IRQ_IOP331_I2C_0	10
-#define IRQ_IOP331_I2C_1	11
-#define IRQ_IOP331_MSG		12
-#define IRQ_IOP331_MSGIBQ	13
-#define IRQ_IOP331_ATU_BIST	14
-#define IRQ_IOP331_PERFMON	15
-#define IRQ_IOP331_CORE_PMU	16
-#define IRQ_IOP331_XINT0	24
-#define IRQ_IOP331_XINT1	25
-#define IRQ_IOP331_XINT2	26
-#define IRQ_IOP331_XINT3	27
-#define IRQ_IOP331_XINT8	32
-#define IRQ_IOP331_XINT9	33
-#define IRQ_IOP331_XINT10	34
-#define IRQ_IOP331_XINT11	35
-#define IRQ_IOP331_XINT12	36
-#define IRQ_IOP331_XINT13	37
-#define IRQ_IOP331_XINT14	38
-#define IRQ_IOP331_XINT15	39
-#define IRQ_IOP331_UART0	51
-#define IRQ_IOP331_UART1	52
-#define IRQ_IOP331_PBIE		53
-#define IRQ_IOP331_ATU_CRW	54
-#define IRQ_IOP331_ATU_ERR	55
-#define IRQ_IOP331_MCU_ERR	56
-#define IRQ_IOP331_DMA0_ERR	57
-#define IRQ_IOP331_DMA1_ERR	58
-#define IRQ_IOP331_AA_ERR	60
-#define IRQ_IOP331_MSG_ERR	62
-#define IRQ_IOP331_HPI		63
+#define IRQ_IOP33X_DMA0_EOT	0
+#define IRQ_IOP33X_DMA0_EOC	1
+#define IRQ_IOP33X_DMA1_EOT	2
+#define IRQ_IOP33X_DMA1_EOC	3
+#define IRQ_IOP33X_AA_EOT	6
+#define IRQ_IOP33X_AA_EOC	7
+#define IRQ_IOP33X_TIMER0	8
+#define IRQ_IOP33X_TIMER1	9
+#define IRQ_IOP33X_I2C_0	10
+#define IRQ_IOP33X_I2C_1	11
+#define IRQ_IOP33X_MSG		12
+#define IRQ_IOP33X_MSGIBQ	13
+#define IRQ_IOP33X_ATU_BIST	14
+#define IRQ_IOP33X_PERFMON	15
+#define IRQ_IOP33X_CORE_PMU	16
+#define IRQ_IOP33X_XINT0	24
+#define IRQ_IOP33X_XINT1	25
+#define IRQ_IOP33X_XINT2	26
+#define IRQ_IOP33X_XINT3	27
+#define IRQ_IOP33X_XINT8	32
+#define IRQ_IOP33X_XINT9	33
+#define IRQ_IOP33X_XINT10	34
+#define IRQ_IOP33X_XINT11	35
+#define IRQ_IOP33X_XINT12	36
+#define IRQ_IOP33X_XINT13	37
+#define IRQ_IOP33X_XINT14	38
+#define IRQ_IOP33X_XINT15	39
+#define IRQ_IOP33X_UART0	51
+#define IRQ_IOP33X_UART1	52
+#define IRQ_IOP33X_PBIE		53
+#define IRQ_IOP33X_ATU_CRW	54
+#define IRQ_IOP33X_ATU_ERR	55
+#define IRQ_IOP33X_MCU_ERR	56
+#define IRQ_IOP33X_DMA0_ERR	57
+#define IRQ_IOP33X_DMA1_ERR	58
+#define IRQ_IOP33X_AA_ERR	60
+#define IRQ_IOP33X_MSG_ERR	62
+#define IRQ_IOP33X_HPI		63
 
 #define NR_IRQS			64
 
 
-#endif // _IRQ_H_
+#endif
diff --git a/include/asm-arm/arch-iop33x/memory.h b/include/asm-arm/arch-iop33x/memory.h
index de208d2cca4e..0d39139b241e 100644
--- a/include/asm-arm/arch-iop33x/memory.h
+++ b/include/asm-arm/arch-iop33x/memory.h
@@ -1,9 +1,9 @@
 /*
- * linux/include/asm-arm/arch-iop33x/memory.h
+ * include/asm-arm/arch-iop33x/memory.h
  */
 
-#ifndef __ASM_ARCH_MEMORY_H
-#define __ASM_ARCH_MEMORY_H
+#ifndef __MEMORY_H
+#define __MEMORY_H
 
 #include <asm/hardware.h>
 
diff --git a/include/asm-arm/arch-iop33x/system.h b/include/asm-arm/arch-iop33x/system.h
index 8270ad9f86c8..00dd07ece262 100644
--- a/include/asm-arm/arch-iop33x/system.h
+++ b/include/asm-arm/arch-iop33x/system.h
@@ -1,7 +1,7 @@
 /*
- * linux/include/asm-arm/arch-iop33x/system.h
+ * include/asm-arm/arch-iop33x/system.h
  *
- *  Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2001 MontaVista Software, Inc.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -13,17 +13,10 @@ static inline void arch_idle(void)
 	cpu_do_idle();
 }
 
-
 static inline void arch_reset(char mode)
 {
-    *IOP3XX_PCSR = 0x30;
+	*IOP3XX_PCSR = 0x30;
 
-	if ( 1 && mode == 's') {
-		/* Jump into ROM at address 0 */
-		cpu_reset(0);
-	} else {
-		/* No on-chip reset capability */
-		cpu_reset(0);
-	}
+	/* Jump into ROM at address 0 */
+	cpu_reset(0);
 }
-
diff --git a/include/asm-arm/arch-iop33x/timex.h b/include/asm-arm/arch-iop33x/timex.h
index 8994322a09f4..fe3e1e369ff9 100644
--- a/include/asm-arm/arch-iop33x/timex.h
+++ b/include/asm-arm/arch-iop33x/timex.h
@@ -1,8 +1,9 @@
 /*
- * linux/include/asm-arm/arch-iop33x/timex.h
+ * include/asm-arm/arch-iop33x/timex.h
  *
  * IOP3xx architecture timex specifications
  */
+
 #include <asm/hardware.h>
 
 #define CLOCK_TICK_RATE		(100 * HZ)
diff --git a/include/asm-arm/arch-iop33x/uncompress.h b/include/asm-arm/arch-iop33x/uncompress.h
index 62904ae3b038..e17fbc05877b 100644
--- a/include/asm-arm/arch-iop33x/uncompress.h
+++ b/include/asm-arm/arch-iop33x/uncompress.h
@@ -1,6 +1,7 @@
 /*
- *  linux/include/asm-arm/arch-iop33x/uncompress.h
+ * include/asm-arm/arch-iop33x/uncompress.h
  */
+
 #include <asm/types.h>
 #include <asm/mach-types.h>
 #include <linux/serial_reg.h>
@@ -8,13 +9,13 @@
 
 static volatile u32 *uart_base;
 
-#define TX_DONE (UART_LSR_TEMT|UART_LSR_THRE)
+#define TX_DONE		(UART_LSR_TEMT | UART_LSR_THRE)
 
 static inline void putc(char c)
 {
 	while ((uart_base[UART_LSR] & TX_DONE) != TX_DONE)
 		barrier();
-	*uart_base = c;
+	uart_base[UART_TX] = c;
 }
 
 static inline void flush(void)
@@ -24,7 +25,7 @@ static inline void flush(void)
 static __inline__ void __arch_decomp_setup(unsigned long arch_id)
 {
 	if (machine_is_iq80331() || machine_is_iq80332())
-		uart_base = (volatile u32 *)IOP331_UART0_PHYS;
+		uart_base = (volatile u32 *)IOP33X_UART0_PHYS;
 	else
 		uart_base = (volatile u32 *)0xfe800000;
 }
diff --git a/include/asm-arm/arch-iop33x/vmalloc.h b/include/asm-arm/arch-iop33x/vmalloc.h
index b5092027449e..66f545a7f4fc 100644
--- a/include/asm-arm/arch-iop33x/vmalloc.h
+++ b/include/asm-arm/arch-iop33x/vmalloc.h
@@ -1,16 +1,5 @@
 /*
- * linux/include/asm-arm/arch-iop33x/vmalloc.h
+ * include/asm-arm/arch-iop33x/vmalloc.h
  */
 
-/*
- * Just any arbitrary offset to the start of the vmalloc VM area: the
- * current 8MB value just means that there will be a 8MB "hole" after the
- * physical memory until the kernel virtual memory starts.  That means that
- * any out-of-bounds memory accesses will hopefully be caught.
- * The vmalloc() routines leaves a hole of 4kB between each vmalloced
- * area for the same reason. ;)
- */
-//#define VMALLOC_END       (0xe8000000)
-/* increase usable physical RAM to ~992M per RMK */
-#define VMALLOC_END       (0xfe000000)
-
+#define VMALLOC_END	0xfe000000
-- 
cgit v1.2.3


From 3b7a86c2f01dafa797908fdcf386f51eb0d01f29 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 20 Sep 2006 21:57:06 +0100
Subject: [ARM] 3846/1: S3C24XX: Fix osiris memory map

The memory mapping for the Osiris machine
are all off by one bit, and the base address
has been fixed for writing (bit25 is being
checked by the write, but not on read)

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-s3c2410/osiris-map.h | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/arch-s3c2410/osiris-map.h b/include/asm-arm/arch-s3c2410/osiris-map.h
index e2d406218ae5..a14164dfa525 100644
--- a/include/asm-arm/arch-s3c2410/osiris-map.h
+++ b/include/asm-arm/arch-s3c2410/osiris-map.h
@@ -18,22 +18,22 @@
 
 /* start peripherals off after the S3C2410 */
 
-#define OSIRIS_IOADDR(x)	(S3C2410_ADDR((x) + 0x05000000))
+#define OSIRIS_IOADDR(x)	(S3C2410_ADDR((x) + 0x04000000))
 
-#define OSIRIS_PA_CPLD		(S3C2410_CS1 | (3<<25))
+#define OSIRIS_PA_CPLD		(S3C2410_CS1 | (1<<26))
 
 /* we put the CPLD registers next, to get them out of the way */
 
-#define OSIRIS_VA_CTRL1	    OSIRIS_IOADDR(0x00000000)	 /* 0x01300000 */
+#define OSIRIS_VA_CTRL1	    OSIRIS_IOADDR(0x00000000)
 #define OSIRIS_PA_CTRL1	    (OSIRIS_PA_CPLD)
 
-#define OSIRIS_VA_CTRL2	    OSIRIS_IOADDR(0x00100000)	 /* 0x01400000 */
-#define OSIRIS_PA_CTRL2	    (OSIRIS_PA_CPLD + (1<<24))
+#define OSIRIS_VA_CTRL2	    OSIRIS_IOADDR(0x00100000)
+#define OSIRIS_PA_CTRL2	    (OSIRIS_PA_CPLD + (1<<23))
 
-#define OSIRIS_VA_CTRL3	    OSIRIS_IOADDR(0x00200000)	 /* 0x01500000 */
-#define OSIRIS_PA_CTRL3	    (OSIRIS_PA_CPLD + (2<<24))
+#define OSIRIS_VA_CTRL3	    OSIRIS_IOADDR(0x00200000)
+#define OSIRIS_PA_CTRL3	    (OSIRIS_PA_CPLD + (2<<23))
 
-#define OSIRIS_VA_CTRL4	    OSIRIS_IOADDR(0x00300000)	 /* 0x01600000 */
-#define OSIRIS_PA_CTRL4	    (OSIRIS_PA_CPLD + (3<<24))
+#define OSIRIS_VA_CTRL4	    OSIRIS_IOADDR(0x00300000)
+#define OSIRIS_PA_CTRL4	    (OSIRIS_PA_CPLD + (3<<23))
 
 #endif /* __ASM_ARCH_OSIRISMAP_H */
-- 
cgit v1.2.3


From 17b602b1c1a38f3f0a4461bb1f571346e751b36b Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Thu, 21 Sep 2006 02:24:38 +0100
Subject: [ARM] 3849/1: fix get_unaligned() for gcc >= 4.1

gcc 4.1's __typeof__ propagates 'const', which breaks get_unaligned().
Rewrite get_unaligned() not to use __typeof__.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/unaligned.h | 62 ++++++++++++++++++---------------------------
 1 file changed, 25 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/unaligned.h b/include/asm-arm/unaligned.h
index 1b39c2f322c9..795b9e5b9e6a 100644
--- a/include/asm-arm/unaligned.h
+++ b/include/asm-arm/unaligned.h
@@ -3,7 +3,7 @@
 
 #include <asm/types.h>
 
-extern int __bug_unaligned_x(void *ptr);
+extern int __bug_unaligned_x(const void *ptr);
 
 /*
  * What is the most efficient way of loading/storing an unaligned value?
@@ -51,44 +51,32 @@ extern int __bug_unaligned_x(void *ptr);
 #define __get_unaligned_4_be(__p)					\
 	(__p[0] << 24 | __p[1] << 16 | __p[2] << 8 | __p[3])
 
-#define __get_unaligned_le(ptr)					\
-	({							\
-		__typeof__(*(ptr)) __v;				\
-		__u8 *__p = (__u8 *)(ptr);			\
-		switch (sizeof(*(ptr))) {			\
-		case 1:	__v = *(ptr);			break;	\
-		case 2: __v = __get_unaligned_2_le(__p);	break;	\
-		case 4: __v = __get_unaligned_4_le(__p);	break;	\
-		case 8: {					\
-				unsigned int __v1, __v2;	\
-				__v2 = __get_unaligned_4_le((__p+4)); \
-				__v1 = __get_unaligned_4_le(__p);	\
-				__v = ((unsigned long long)__v2 << 32 | __v1);	\
-			}					\
-			break;					\
-		default: __v = __bug_unaligned_x(__p);	break;	\
-		}						\
-		__v;						\
+#define __get_unaligned_8_le(__p)					\
+	((unsigned long long)__get_unaligned_4_le((__p+4)) << 32 |	\
+		__get_unaligned_4_le(__p))
+
+#define __get_unaligned_8_be(__p)					\
+	((unsigned long long)__get_unaligned_4_be(__p) << 32 |		\
+		__get_unaligned_4_be((__p+4)))
+
+#define __get_unaligned_le(ptr)						\
+	({								\
+		const __u8 *__p = (const __u8 *)(ptr);			\
+		__builtin_choose_expr(sizeof(*(ptr)) == 1, *__p,	\
+		  __builtin_choose_expr(sizeof(*(ptr)) == 2, __get_unaligned_2_le(__p),	\
+		  __builtin_choose_expr(sizeof(*(ptr)) == 4, __get_unaligned_4_le(__p),	\
+		  __builtin_choose_expr(sizeof(*(ptr)) == 8, __get_unaligned_8_le(__p),	\
+		    (void)__bug_unaligned_x(__p)))));			\
 	})
 
-#define __get_unaligned_be(ptr)					\
-	({							\
-		__typeof__(*(ptr)) __v;				\
-		__u8 *__p = (__u8 *)(ptr);			\
-		switch (sizeof(*(ptr))) {			\
-		case 1:	__v = *(ptr);			break;	\
-		case 2: __v = __get_unaligned_2_be(__p);	break;	\
-		case 4: __v = __get_unaligned_4_be(__p);	break;	\
-		case 8: {					\
-				unsigned int __v1, __v2;	\
-				__v2 = __get_unaligned_4_be(__p); \
-				__v1 = __get_unaligned_4_be((__p+4));	\
-				__v = ((unsigned long long)__v2 << 32 | __v1);	\
-			}					\
-			break;					\
-		default: __v = __bug_unaligned_x(__p);	break;	\
-		}						\
-		__v;						\
+#define __get_unaligned_be(ptr)						\
+	({								\
+		const __u8 *__p = (const __u8 *)(ptr);			\
+		__builtin_choose_expr(sizeof(*(ptr)) == 1, *__p,	\
+		  __builtin_choose_expr(sizeof(*(ptr)) == 2, __get_unaligned_2_be(__p),	\
+		  __builtin_choose_expr(sizeof(*(ptr)) == 4, __get_unaligned_4_be(__p),	\
+		  __builtin_choose_expr(sizeof(*(ptr)) == 8, __get_unaligned_8_be(__p),	\
+		    (void)__bug_unaligned_x(__p)))));			\
 	})
 
 
-- 
cgit v1.2.3


From e60d07b6cd38a7afb85f2cf51aebcb3359b63819 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Thu, 21 Sep 2006 02:42:12 +0100
Subject: [ARM] 3850/1: iop3xx: add thecus n2100 support

Add support for the Thecus n2100 (80219-based.)

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/iop32x_defconfig      | 119 +++++++++++++++-
 arch/arm/mach-iop32x/Kconfig           |   6 +
 arch/arm/mach-iop32x/Makefile          |   1 +
 arch/arm/mach-iop32x/irq.c             |   3 +-
 arch/arm/mach-iop32x/n2100.c           | 251 +++++++++++++++++++++++++++++++++
 include/asm-arm/arch-iop32x/hardware.h |   1 +
 include/asm-arm/arch-iop32x/n2100.h    |  19 +++
 include/asm-arm/arch-iop32x/system.h   |  11 ++
 8 files changed, 408 insertions(+), 3 deletions(-)
 create mode 100644 arch/arm/mach-iop32x/n2100.c
 create mode 100644 include/asm-arm/arch-iop32x/n2100.h

(limited to 'include')

diff --git a/arch/arm/configs/iop32x_defconfig b/arch/arm/configs/iop32x_defconfig
index af7ead33c6dd..af3b40c0508a 100644
--- a/arch/arm/configs/iop32x_defconfig
+++ b/arch/arm/configs/iop32x_defconfig
@@ -128,6 +128,7 @@ CONFIG_ARCH_IOP32X=y
 #
 CONFIG_ARCH_IQ80321=y
 CONFIG_ARCH_IQ31244=y
+CONFIG_MACH_N2100=y
 CONFIG_PLAT_IOP=y
 
 #
@@ -601,7 +602,8 @@ CONFIG_E1000_NAPI=y
 # CONFIG_NS83820 is not set
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
-# CONFIG_R8169 is not set
+CONFIG_R8169=y
+# CONFIG_R8169_NAPI is not set
 # CONFIG_SIS190 is not set
 # CONFIG_SKGE is not set
 # CONFIG_SKY2 is not set
@@ -897,12 +899,125 @@ CONFIG_DUMMY_CONSOLE=y
 CONFIG_USB_ARCH_HAS_HCD=y
 CONFIG_USB_ARCH_HAS_OHCI=y
 CONFIG_USB_ARCH_HAS_EHCI=y
-# CONFIG_USB is not set
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+# CONFIG_USB_DEVICEFS is not set
+# CONFIG_USB_BANDWIDTH is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+CONFIG_USB_EHCI_HCD=y
+CONFIG_USB_EHCI_SPLIT_ISO=y
+CONFIG_USB_EHCI_ROOT_HUB_TT=y
+CONFIG_USB_EHCI_TT_NEWSCHED=y
+# CONFIG_USB_ISP116X_HCD is not set
+# CONFIG_USB_OHCI_HCD is not set
+CONFIG_USB_UHCI_HCD=y
+# CONFIG_USB_SL811_HCD is not set
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_ACM is not set
+# CONFIG_USB_PRINTER is not set
 
 #
 # NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
 #
 
+#
+# may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=y
+# CONFIG_USB_STORAGE_DEBUG is not set
+# CONFIG_USB_STORAGE_DATAFAB is not set
+# CONFIG_USB_STORAGE_FREECOM is not set
+# CONFIG_USB_STORAGE_DPCM is not set
+# CONFIG_USB_STORAGE_USBAT is not set
+# CONFIG_USB_STORAGE_SDDR09 is not set
+# CONFIG_USB_STORAGE_SDDR55 is not set
+# CONFIG_USB_STORAGE_JUMPSHOT is not set
+# CONFIG_USB_STORAGE_ALAUDA is not set
+# CONFIG_USB_LIBUSUAL is not set
+
+#
+# USB Input Devices
+#
+# CONFIG_USB_HID is not set
+
+#
+# USB HID Boot Protocol drivers
+#
+# CONFIG_USB_KBD is not set
+# CONFIG_USB_MOUSE is not set
+# CONFIG_USB_AIPTEK is not set
+# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
+# CONFIG_USB_KBTAB is not set
+# CONFIG_USB_POWERMATE is not set
+# CONFIG_USB_TOUCHSCREEN is not set
+# CONFIG_USB_YEALINK is not set
+# CONFIG_USB_XPAD is not set
+# CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_ATI_REMOTE2 is not set
+# CONFIG_USB_KEYSPAN_REMOTE is not set
+# CONFIG_USB_APPLETOUCH is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+# CONFIG_USB_SERIAL is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGETKIT is not set
+# CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_APPLEDISPLAY is not set
+# CONFIG_USB_SISUSBVGA is not set
+# CONFIG_USB_LD is not set
+
+#
+# USB DSL modem support
+#
+
 #
 # USB Gadget Support
 #
diff --git a/arch/arm/mach-iop32x/Kconfig b/arch/arm/mach-iop32x/Kconfig
index d7abfaa525c3..0a397a4a6ac9 100644
--- a/arch/arm/mach-iop32x/Kconfig
+++ b/arch/arm/mach-iop32x/Kconfig
@@ -17,6 +17,12 @@ config ARCH_IQ31244
 	  evaluation kit for the Intel 80219 processor (a IOP321 variant)
 	  or the IQ31244 evaluation kit for the IOP321 processor.
 
+config MACH_N2100
+	bool "Enable support for the Thecus n2100"
+	help
+	  Say Y here if you want to run your kernel on the Thecus n2100
+	  NAS appliance.
+
 endmenu
 
 endif
diff --git a/arch/arm/mach-iop32x/Makefile b/arch/arm/mach-iop32x/Makefile
index af1747ae392d..9c62470cd4f5 100644
--- a/arch/arm/mach-iop32x/Makefile
+++ b/arch/arm/mach-iop32x/Makefile
@@ -9,3 +9,4 @@ obj-			:=
 
 obj-$(CONFIG_ARCH_IQ80321) += iq80321.o
 obj-$(CONFIG_ARCH_IQ31244) += iq31244.o
+obj-$(CONFIG_MACH_N2100) += n2100.o
diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index 21294be5a369..17fb94e6e29f 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -63,7 +63,8 @@ void __init iop32x_init_irq(void)
 	intctl_write(0);
 	intstr_write(0);
 	if (machine_is_iq80321() ||
-	    machine_is_iq31244())
+	    machine_is_iq31244() ||
+	    machine_is_n2100())
 		*IOP3XX_PCIIRSR = 0x0f;
 
 	for (i = 0; i < NR_IRQS; i++) {
diff --git a/arch/arm/mach-iop32x/n2100.c b/arch/arm/mach-iop32x/n2100.c
new file mode 100644
index 000000000000..a2c94a47b2b2
--- /dev/null
+++ b/arch/arm/mach-iop32x/n2100.c
@@ -0,0 +1,251 @@
+/*
+ * arch/arm/mach-iop32x/n2100.c
+ *
+ * Board support code for the Thecus N2100 platform.
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ * Copyright 2003 (c) MontaVista, Software, Inc.
+ * Copyright (C) 2004 Intel Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/pm.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/serial_core.h>
+#include <linux/serial_8250.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <linux/reboot.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/pci.h>
+#include <asm/mach/time.h>
+#include <asm/mach-types.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+
+/*
+ * N2100 timer tick configuration.
+ */
+static void __init n2100_timer_init(void)
+{
+	/* 33.000 MHz crystal.  */
+	iop3xx_init_time(198000000);
+}
+
+static struct sys_timer n2100_timer = {
+	.init		= n2100_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
+
+/*
+ * N2100 I/O.
+ */
+static struct map_desc n2100_io_desc[] __initdata = {
+	{	/* on-board devices */
+		.virtual	= N2100_UART,
+		.pfn		= __phys_to_pfn(N2100_UART),
+		.length		= 0x00100000,
+		.type		= MT_DEVICE
+	},
+};
+
+void __init n2100_map_io(void)
+{
+	iop3xx_map_io();
+	iotable_init(n2100_io_desc, ARRAY_SIZE(n2100_io_desc));
+}
+
+
+/*
+ * N2100 PCI.
+ */
+static inline int __init
+n2100_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq;
+
+	if (PCI_SLOT(dev->devfn) == 1) {
+		/* RTL8110SB #1 */
+		irq = IRQ_IOP32X_XINT0;
+	} else if (PCI_SLOT(dev->devfn) == 2) {
+		/* RTL8110SB #2 */
+		irq = IRQ_IOP32X_XINT1;
+	} else if (PCI_SLOT(dev->devfn) == 3) {
+		/* Sil3512 */
+		irq = IRQ_IOP32X_XINT2;
+	} else if (PCI_SLOT(dev->devfn) == 4 && pin == 1) {
+		/* VT6212 INTA */
+		irq = IRQ_IOP32X_XINT1;
+	} else if (PCI_SLOT(dev->devfn) == 4 && pin == 2) {
+		/* VT6212 INTB */
+		irq = IRQ_IOP32X_XINT0;
+	} else if (PCI_SLOT(dev->devfn) == 4 && pin == 3) {
+		/* VT6212 INTC */
+		irq = IRQ_IOP32X_XINT2;
+	} else if (PCI_SLOT(dev->devfn) == 5) {
+		/* Mini-PCI slot */
+		irq = IRQ_IOP32X_XINT3;
+	} else {
+		printk(KERN_ERR "n2100_pci_map_irq() called for unknown "
+			"device PCI:%d:%d:%d\n", dev->bus->number,
+			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+		irq = -1;
+	}
+
+	return irq;
+}
+
+static struct hw_pci n2100_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= n2100_pci_map_irq,
+};
+
+static int __init n2100_pci_init(void)
+{
+	if (machine_is_n2100())
+		pci_common_init(&n2100_pci);
+
+	return 0;
+}
+
+subsys_initcall(n2100_pci_init);
+
+
+/*
+ * N2100 machine initialisation.
+ */
+static struct physmap_flash_data n2100_flash_data = {
+	.width		= 2,
+};
+
+static struct resource n2100_flash_resource = {
+	.start		= 0xf0000000,
+	.end		= 0xf0ffffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device n2100_flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &n2100_flash_data,
+	},
+	.num_resources	= 1,
+	.resource	= &n2100_flash_resource,
+};
+
+
+static struct plat_serial8250_port n2100_serial_port[] = {
+	{
+		.mapbase	= N2100_UART,
+		.membase	= (char *)N2100_UART,
+		.irq		= 0,
+		.flags		= UPF_SKIP_TEST,
+		.iotype		= UPIO_MEM,
+		.regshift	= 0,
+		.uartclk	= 1843200,
+	},
+	{ },
+};
+
+static struct resource n2100_uart_resource = {
+	.start		= N2100_UART,
+	.end		= N2100_UART + 7,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device n2100_serial_device = {
+	.name		= "serial8250",
+	.id		= PLAT8250_DEV_PLATFORM,
+	.dev		= {
+		.platform_data		= n2100_serial_port,
+	},
+	.num_resources	= 1,
+	.resource	= &n2100_uart_resource,
+};
+
+
+/*
+ * Pull PCA9532 GPIO #8 low to power off the machine.
+ */
+static void n2100_power_off(void)
+{
+	local_irq_disable();
+
+	/* Start condition, I2C address of PCA9532, write transaction.  */
+	*IOP3XX_IDBR0 = 0xc0;
+	*IOP3XX_ICR0 = 0xe9;
+	mdelay(1);
+
+	/* Write address 0x08.  */
+	*IOP3XX_IDBR0 = 0x08;
+	*IOP3XX_ICR0 = 0xe8;
+	mdelay(1);
+
+	/* Write data 0x01, stop condition.  */
+	*IOP3XX_IDBR0 = 0x01;
+	*IOP3XX_ICR0 = 0xea;
+
+	while (1)
+		;
+}
+
+
+static struct timer_list power_button_poll_timer;
+
+static void power_button_poll(unsigned long dummy)
+{
+	if (gpio_line_get(N2100_POWER_BUTTON) == 0) {
+		ctrl_alt_del();
+		return;
+	}
+
+	power_button_poll_timer.expires = jiffies + (HZ / 10);
+	add_timer(&power_button_poll_timer);
+}
+
+
+static void __init n2100_init_machine(void)
+{
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&n2100_flash_device);
+	platform_device_register(&n2100_serial_device);
+
+	pm_power_off = n2100_power_off;
+
+	init_timer(&power_button_poll_timer);
+	power_button_poll_timer.function = power_button_poll;
+	power_button_poll_timer.expires = jiffies + (HZ / 10);
+	add_timer(&power_button_poll_timer);
+}
+
+MACHINE_START(N2100, "Thecus N2100")
+	/* Maintainer: Lennert Buytenhek <buytenh@wantstofly.org> */
+	.phys_io	= N2100_UART,
+	.io_pg_offst	= ((N2100_UART) >> 18) & 0xfffc,
+	.boot_params	= 0xa0000100,
+	.map_io		= n2100_map_io,
+	.init_irq	= iop32x_init_irq,
+	.timer		= &n2100_timer,
+	.init_machine	= n2100_init_machine,
+MACHINE_END
diff --git a/include/asm-arm/arch-iop32x/hardware.h b/include/asm-arm/arch-iop32x/hardware.h
index 6a3001f2f7e0..c6f58b91ddbc 100644
--- a/include/asm-arm/arch-iop32x/hardware.h
+++ b/include/asm-arm/arch-iop32x/hardware.h
@@ -37,6 +37,7 @@ void iop32x_init_irq(void);
  */
 #include "iq80321.h"
 #include "iq31244.h"
+#include "n2100.h"
 
 
 #endif
diff --git a/include/asm-arm/arch-iop32x/n2100.h b/include/asm-arm/arch-iop32x/n2100.h
new file mode 100644
index 000000000000..fed31a648425
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/n2100.h
@@ -0,0 +1,19 @@
+/*
+ * include/asm/arch-iop32x/n2100.h
+ *
+ * Thecus N2100 board registers
+ */
+
+#ifndef __N2100_H
+#define __N2100_H
+
+#define N2100_UART		0xfe800000	/* UART */
+
+#define N2100_COPY_BUTTON	IOP3XX_GPIO_LINE(0)
+#define N2100_PCA9532_RESET	IOP3XX_GPIO_LINE(2)
+#define N2100_RESET_BUTTON	IOP3XX_GPIO_LINE(3)
+#define N2100_HARDWARE_RESET	IOP3XX_GPIO_LINE(4)
+#define N2100_POWER_BUTTON	IOP3XX_GPIO_LINE(5)
+
+
+#endif
diff --git a/include/asm-arm/arch-iop32x/system.h b/include/asm-arm/arch-iop32x/system.h
index c65ede3e627a..17b7eb7e9c0d 100644
--- a/include/asm-arm/arch-iop32x/system.h
+++ b/include/asm-arm/arch-iop32x/system.h
@@ -8,6 +8,8 @@
  * published by the Free Software Foundation.
  */
 
+#include <asm/mach-types.h>
+
 static inline void arch_idle(void)
 {
 	cpu_do_idle();
@@ -15,6 +17,15 @@ static inline void arch_idle(void)
 
 static inline void arch_reset(char mode)
 {
+	local_irq_disable();
+
+	if (machine_is_n2100()) {
+		gpio_line_set(N2100_HARDWARE_RESET, GPIO_LOW);
+		gpio_line_config(N2100_HARDWARE_RESET, GPIO_OUT);
+		while (1)
+			;
+	}
+
 	*IOP3XX_PCSR = 0x30;
 
 	/* Jump into ROM at address 0 */
-- 
cgit v1.2.3


From 0c92e830bd39f3e6cf7b151dffecafbdc623496c Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Thu, 21 Sep 2006 02:46:03 +0100
Subject: [ARM] 3851/1: iop3xx: add io-data glantank support

Add support for the IO-Data GLAN Tank, from Martin Michlmayr.

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/configs/iop32x_defconfig      |   1 +
 arch/arm/mach-iop32x/Kconfig           |   7 ++
 arch/arm/mach-iop32x/Makefile          |   1 +
 arch/arm/mach-iop32x/glantank.c        | 195 +++++++++++++++++++++++++++++++++
 arch/arm/mach-iop32x/irq.c             |   3 +-
 include/asm-arm/arch-iop32x/glantank.h |  13 +++
 include/asm-arm/arch-iop32x/hardware.h |   1 +
 7 files changed, 220 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mach-iop32x/glantank.c
 create mode 100644 include/asm-arm/arch-iop32x/glantank.h

(limited to 'include')

diff --git a/arch/arm/configs/iop32x_defconfig b/arch/arm/configs/iop32x_defconfig
index af3b40c0508a..0d67f66e78c2 100644
--- a/arch/arm/configs/iop32x_defconfig
+++ b/arch/arm/configs/iop32x_defconfig
@@ -126,6 +126,7 @@ CONFIG_ARCH_IOP32X=y
 #
 # IOP32x Platform Types
 #
+CONFIG_MACH_GLANTANK=y
 CONFIG_ARCH_IQ80321=y
 CONFIG_ARCH_IQ31244=y
 CONFIG_MACH_N2100=y
diff --git a/arch/arm/mach-iop32x/Kconfig b/arch/arm/mach-iop32x/Kconfig
index 0a397a4a6ac9..c072d94070da 100644
--- a/arch/arm/mach-iop32x/Kconfig
+++ b/arch/arm/mach-iop32x/Kconfig
@@ -4,6 +4,13 @@ menu "IOP32x Implementation Options"
 
 comment "IOP32x Platform Types"
 
+config MACH_GLANTANK
+	bool "Enable support for the IO-Data GLAN Tank"
+	help
+	  Say Y here if you want to run your kernel on the GLAN Tank
+	  NAS appliance or machines from IO-Data's HDL-Gxxx, HDL-GWxxx
+	  and HDL-GZxxx series.
+
 config ARCH_IQ80321
 	bool "Enable support for IQ80321"
 	help
diff --git a/arch/arm/mach-iop32x/Makefile b/arch/arm/mach-iop32x/Makefile
index 9c62470cd4f5..7b05b37e1f94 100644
--- a/arch/arm/mach-iop32x/Makefile
+++ b/arch/arm/mach-iop32x/Makefile
@@ -7,6 +7,7 @@ obj-m			:=
 obj-n			:=
 obj-			:=
 
+obj-$(CONFIG_MACH_GLANTANK) += glantank.o
 obj-$(CONFIG_ARCH_IQ80321) += iq80321.o
 obj-$(CONFIG_ARCH_IQ31244) += iq31244.o
 obj-$(CONFIG_MACH_N2100) += n2100.o
diff --git a/arch/arm/mach-iop32x/glantank.c b/arch/arm/mach-iop32x/glantank.c
new file mode 100644
index 000000000000..b9b765057dbe
--- /dev/null
+++ b/arch/arm/mach-iop32x/glantank.c
@@ -0,0 +1,195 @@
+/*
+ * arch/arm/mach-iop32x/glantank.c
+ *
+ * Board support code for the GLAN Tank.
+ *
+ * Copyright (C) 2006 Martin Michlmayr <tbm@cyrius.com>
+ * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/serial_core.h>
+#include <linux/serial_8250.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/pci.h>
+#include <asm/mach/time.h>
+#include <asm/mach-types.h>
+#include <asm/page.h>
+
+/*
+ * GLAN Tank timer tick configuration.
+ */
+static void __init glantank_timer_init(void)
+{
+	/* 33.333 MHz crystal.  */
+	iop3xx_init_time(200000000);
+}
+
+static struct sys_timer glantank_timer = {
+	.init		= glantank_timer_init,
+	.offset		= iop3xx_gettimeoffset,
+};
+
+
+/*
+ * GLAN Tank I/O.
+ */
+static struct map_desc glantank_io_desc[] __initdata = {
+	{	/* on-board devices */
+		.virtual	= GLANTANK_UART,
+		.pfn		= __phys_to_pfn(GLANTANK_UART),
+		.length		= 0x00100000,
+		.type		= MT_DEVICE
+	},
+};
+
+void __init glantank_map_io(void)
+{
+	iop3xx_map_io();
+	iotable_init(glantank_io_desc, ARRAY_SIZE(glantank_io_desc));
+}
+
+
+/*
+ * GLAN Tank PCI.
+ */
+#define INTA	IRQ_IOP32X_XINT0
+#define INTB	IRQ_IOP32X_XINT1
+#define INTC	IRQ_IOP32X_XINT2
+#define INTD	IRQ_IOP32X_XINT3
+
+static inline int __init
+glantank_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	static int pci_irq_table[][4] = {
+		/*
+		 * PCI IDSEL/INTPIN->INTLINE
+		 * A       B       C       D
+		 */
+		{INTD, INTD, INTD, INTD}, /* UART (8250) */
+		{INTA, INTA, INTA, INTA}, /* Ethernet (E1000) */
+		{INTB, INTB, INTB, INTB}, /* IDE (AEC6280R) */
+		{INTC, INTC, INTC, INTC}, /* USB (NEC) */
+	};
+
+	BUG_ON(pin < 1 || pin > 4);
+
+	return pci_irq_table[slot % 4][pin - 1];
+}
+
+static struct hw_pci glantank_pci __initdata = {
+	.swizzle	= pci_std_swizzle,
+	.nr_controllers = 1,
+	.setup		= iop3xx_pci_setup,
+	.preinit	= iop3xx_pci_preinit,
+	.scan		= iop3xx_pci_scan_bus,
+	.map_irq	= glantank_pci_map_irq,
+};
+
+static int __init glantank_pci_init(void)
+{
+	if (machine_is_glantank())
+		pci_common_init(&glantank_pci);
+
+	return 0;
+}
+
+subsys_initcall(glantank_pci_init);
+
+
+/*
+ * GLAN Tank machine initialization.
+ */
+static struct physmap_flash_data glantank_flash_data = {
+	.width		= 1,
+};
+
+static struct resource glantank_flash_resource = {
+	.start		= 0xf0000000,
+	.end		= 0xf007ffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device glantank_flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &glantank_flash_data,
+	},
+	.num_resources	= 1,
+	.resource	= &glantank_flash_resource,
+};
+
+static struct plat_serial8250_port glantank_serial_port[] = {
+	{
+		.mapbase	= GLANTANK_UART,
+		.membase	= (char *)GLANTANK_UART,
+		.irq		= IRQ_IOP32X_XINT3,
+		.flags		= UPF_SKIP_TEST,
+		.iotype		= UPIO_MEM,
+		.regshift	= 0,
+		.uartclk	= 1843200,
+	},
+	{ },
+};
+
+static struct resource glantank_uart_resource = {
+	.start		= GLANTANK_UART,
+	.end		= GLANTANK_UART + 7,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device glantank_serial_device = {
+	.name		= "serial8250",
+	.id		= PLAT8250_DEV_PLATFORM,
+	.dev		= {
+		.platform_data		= glantank_serial_port,
+	},
+	.num_resources	= 1,
+	.resource	= &glantank_uart_resource,
+};
+
+static void glantank_power_off(void)
+{
+	__raw_writeb(0x01, 0xfe8d0004);
+
+	while (1)
+		;
+}
+
+static void __init glantank_init_machine(void)
+{
+	platform_device_register(&iop3xx_i2c0_device);
+	platform_device_register(&iop3xx_i2c1_device);
+	platform_device_register(&glantank_flash_device);
+	platform_device_register(&glantank_serial_device);
+
+	pm_power_off = glantank_power_off;
+}
+
+MACHINE_START(GLANTANK, "GLAN Tank")
+	/* Maintainer: Lennert Buytenhek <buytenh@wantstofly.org> */
+	.phys_io	= GLANTANK_UART,
+	.io_pg_offst	= ((GLANTANK_UART) >> 18) & 0xfffc,
+	.boot_params	= 0xa0000100,
+	.map_io		= glantank_map_io,
+	.init_irq	= iop32x_init_irq,
+	.timer		= &glantank_timer,
+	.init_machine	= glantank_init_machine,
+MACHINE_END
diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index 17fb94e6e29f..69d6302f40cf 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -62,7 +62,8 @@ void __init iop32x_init_irq(void)
 
 	intctl_write(0);
 	intstr_write(0);
-	if (machine_is_iq80321() ||
+	if (machine_is_glantank() ||
+	    machine_is_iq80321() ||
 	    machine_is_iq31244() ||
 	    machine_is_n2100())
 		*IOP3XX_PCIIRSR = 0x0f;
diff --git a/include/asm-arm/arch-iop32x/glantank.h b/include/asm-arm/arch-iop32x/glantank.h
new file mode 100644
index 000000000000..3b065618dd00
--- /dev/null
+++ b/include/asm-arm/arch-iop32x/glantank.h
@@ -0,0 +1,13 @@
+/*
+ * include/asm/arch-iop32x/glantank.h
+ *
+ * IO-Data GLAN Tank board registers
+ */
+
+#ifndef __GLANTANK_H
+#define __GLANTANK_H
+
+#define GLANTANK_UART		0xfe800000	/* UART */
+
+
+#endif
diff --git a/include/asm-arm/arch-iop32x/hardware.h b/include/asm-arm/arch-iop32x/hardware.h
index c6f58b91ddbc..6556ed5eee31 100644
--- a/include/asm-arm/arch-iop32x/hardware.h
+++ b/include/asm-arm/arch-iop32x/hardware.h
@@ -35,6 +35,7 @@ void iop32x_init_irq(void);
 /*
  * Board specific bits
  */
+#include "glantank.h"
 #include "iq80321.h"
 #include "iq31244.h"
 #include "n2100.h"
-- 
cgit v1.2.3


From e7cc2c59cc83558fc26f17eee3c8f901119f0a7c Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Thu, 21 Sep 2006 03:35:20 +0100
Subject: [ARM] 3852/1: convert atomic bitops and __xchg over to
 raw_local_irq_{save,restore}

Thomas Gleixner noticed that bitops.h should also use the raw_* irq
disable/enable variants, and __xchg needs them as well.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/bitops.h | 24 ++++++++++++------------
 include/asm-arm/system.h |  8 ++++----
 2 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/bitops.h b/include/asm-arm/bitops.h
index 0ac54b1a8bad..b41831b6432f 100644
--- a/include/asm-arm/bitops.h
+++ b/include/asm-arm/bitops.h
@@ -37,9 +37,9 @@ static inline void ____atomic_set_bit(unsigned int bit, volatile unsigned long *
 
 	p += bit >> 5;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	*p |= mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 }
 
 static inline void ____atomic_clear_bit(unsigned int bit, volatile unsigned long *p)
@@ -49,9 +49,9 @@ static inline void ____atomic_clear_bit(unsigned int bit, volatile unsigned long
 
 	p += bit >> 5;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	*p &= ~mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 }
 
 static inline void ____atomic_change_bit(unsigned int bit, volatile unsigned long *p)
@@ -61,9 +61,9 @@ static inline void ____atomic_change_bit(unsigned int bit, volatile unsigned lon
 
 	p += bit >> 5;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	*p ^= mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 }
 
 static inline int
@@ -75,10 +75,10 @@ ____atomic_test_and_set_bit(unsigned int bit, volatile unsigned long *p)
 
 	p += bit >> 5;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	res = *p;
 	*p = res | mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 
 	return res & mask;
 }
@@ -92,10 +92,10 @@ ____atomic_test_and_clear_bit(unsigned int bit, volatile unsigned long *p)
 
 	p += bit >> 5;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	res = *p;
 	*p = res & ~mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 
 	return res & mask;
 }
@@ -109,10 +109,10 @@ ____atomic_test_and_change_bit(unsigned int bit, volatile unsigned long *p)
 
 	p += bit >> 5;
 
-	local_irq_save(flags);
+	raw_local_irq_save(flags);
 	res = *p;
 	*p = res ^ mask;
-	local_irq_restore(flags);
+	raw_local_irq_restore(flags);
 
 	return res & mask;
 }
diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index 174ff52661b0..c19c5b009f71 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -282,17 +282,17 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size
 #error SMP is not supported on this platform
 #endif
 	case 1:
-		local_irq_save(flags);
+		raw_local_irq_save(flags);
 		ret = *(volatile unsigned char *)ptr;
 		*(volatile unsigned char *)ptr = x;
-		local_irq_restore(flags);
+		raw_local_irq_restore(flags);
 		break;
 
 	case 4:
-		local_irq_save(flags);
+		raw_local_irq_save(flags);
 		ret = *(volatile unsigned long *)ptr;
 		*(volatile unsigned long *)ptr = x;
-		local_irq_restore(flags);
+		raw_local_irq_restore(flags);
 		break;
 #else
 	case 1:
-- 
cgit v1.2.3


From 93bda4c0214441b0bb03b61c2bf1d6727896a750 Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <samuel.ortiz@solidboot.com>
Date: Mon, 25 Sep 2006 12:41:22 +0300
Subject: ARM: OMAP: Added OMAP24xx camera IRQ definition

Signed-off-by: Samuel Ortiz <samuel.ortiz@solidboot.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 include/asm-arm/arch-omap/irqs.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-omap/irqs.h b/include/asm-arm/arch-omap/irqs.h
index 2542495d8a43..636f21cae3ee 100644
--- a/include/asm-arm/arch-omap/irqs.h
+++ b/include/asm-arm/arch-omap/irqs.h
@@ -237,6 +237,7 @@
 #define INT_24XX_SDMA_IRQ1	13
 #define INT_24XX_SDMA_IRQ2	14
 #define INT_24XX_SDMA_IRQ3	15
+#define INT_24XX_CAM_IRQ	24
 #define INT_24XX_DSS_IRQ	25
 #define INT_24XX_GPIO_BANK1	29
 #define INT_24XX_GPIO_BANK2	30
-- 
cgit v1.2.3


From abc45e1d69542281fb2b40968e5d112f51976623 Mon Sep 17 00:00:00 2001
From: Kyungmin Park <kyungmin.park@samsung.com>
Date: Mon, 25 Sep 2006 12:41:25 +0300
Subject: ARM: OMAP: Apollon MMC support

Apollon board MMC supports on OMAP2

TODO:
        We have to check MMC on H4

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/board-apollon.c |  7 +++++--
 arch/arm/mach-omap2/mux.c           | 14 ++++++++++++++
 arch/arm/plat-omap/devices.c        | 20 +++++++++++++++++---
 include/asm-arm/arch-omap/irqs.h    |  1 +
 include/asm-arm/arch-omap/mux.h     | 14 ++++++++++++++
 5 files changed, 51 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-omap2/board-apollon.c b/arch/arm/mach-omap2/board-apollon.c
index 7993b7bae2bd..2db6b732b084 100644
--- a/arch/arm/mach-omap2/board-apollon.c
+++ b/arch/arm/mach-omap2/board-apollon.c
@@ -166,8 +166,8 @@ static struct omap_uart_config apollon_uart_config __initdata = {
 
 static struct omap_mmc_config apollon_mmc_config __initdata = {
 	.mmc [0] = {
-		.enabled 	= 0,
-		.wire4		= 0,
+		.enabled 	= 1,
+		.wire4		= 1,
 		.wp_pin		= -1,
 		.power_pin	= -1,
 		.switch_pin	= -1,
@@ -257,6 +257,9 @@ static void __init omap_apollon_init(void)
 	/* REVISIT: where's the correct place */
 	omap_cfg_reg(W19_24XX_SYS_NIRQ);
 
+	/* Use Interal loop-back in MMC/SDIO Module Input Clock selection */
+	CONTROL_DEVCONF |= (1 << 24);
+
 	/*
  	 * Make sure the serial ports are muxed on at this point.
 	 * You have to mux them off in device drivers later on
diff --git a/arch/arm/mach-omap2/mux.c b/arch/arm/mach-omap2/mux.c
index 60ef084faffd..f538d0fdb13c 100644
--- a/arch/arm/mach-omap2/mux.c
+++ b/arch/arm/mach-omap2/mux.c
@@ -104,6 +104,20 @@ MUX_CFG_24XX("P20_24XX_TSC_IRQ",	0x108,	0,	0,	0,	1)
 MUX_CFG_24XX("K15_24XX_UART3_TX",	0x118,	0,	0,	0,	1)
 MUX_CFG_24XX("K14_24XX_UART3_RX",	0x119,	0,	0,	0,	1)
 
+/* MMC/SDIO */
+MUX_CFG_24XX("G19_24XX_MMC_CLKO",	0x0f3,	0,	0,	0,	1)
+MUX_CFG_24XX("H18_24XX_MMC_CMD",	0x0f4,	0,	0,	0,	1)
+MUX_CFG_24XX("F20_24XX_MMC_DAT0",	0x0f5,	0,	0,	0,	1)
+MUX_CFG_24XX("H14_24XX_MMC_DAT1",	0x0f6,	0,	0,	0,	1)
+MUX_CFG_24XX("E19_24XX_MMC_DAT2",	0x0f7,	0,	0,	0,	1)
+MUX_CFG_24XX("D19_24XX_MMC_DAT3",	0x0f8,	0,	0,	0,	1)
+MUX_CFG_24XX("F19_24XX_MMC_DAT_DIR0",	0x0f9,	0,	0,	0,	1)
+MUX_CFG_24XX("E20_24XX_MMC_DAT_DIR1",	0x0fa,	0,	0,	0,	1)
+MUX_CFG_24XX("F18_24XX_MMC_DAT_DIR2",	0x0fb,	0,	0,	0,	1)
+MUX_CFG_24XX("E18_24XX_MMC_DAT_DIR3",	0x0fc,	0,	0,	0,	1)
+MUX_CFG_24XX("G18_24XX_MMC_CMD_DIR",	0x0fd,	0,	0,	0,	1)
+MUX_CFG_24XX("H15_24XX_MMC_CLKI",	0x0fe,	0,	0,	0,	1)
+
 /* Keypad GPIO*/
 MUX_CFG_24XX("T19_24XX_KBR0",		0x106,	3,	1,	1,	1)
 MUX_CFG_24XX("R19_24XX_KBR1",		0x107,	3,	1,	1,	1)
diff --git a/arch/arm/plat-omap/devices.c b/arch/arm/plat-omap/devices.c
index 1812f237d12f..dbc3f44e07a6 100644
--- a/arch/arm/plat-omap/devices.c
+++ b/arch/arm/plat-omap/devices.c
@@ -148,7 +148,7 @@ static inline void omap_init_kp(void) {}
 
 #ifdef CONFIG_ARCH_OMAP24XX
 #define	OMAP_MMC1_BASE		0x4809c000
-#define OMAP_MMC1_INT		83
+#define OMAP_MMC1_INT		INT_24XX_MMC_IRQ
 #else
 #define	OMAP_MMC1_BASE		0xfffb7800
 #define OMAP_MMC1_INT		INT_MMC
@@ -225,7 +225,14 @@ static void __init omap_init_mmc(void)
 	/* block 1 is always available and has just one pinout option */
 	mmc = &mmc_conf->mmc[0];
 	if (mmc->enabled) {
-		if (!cpu_is_omap24xx()) {
+		if (cpu_is_omap24xx()) {
+			omap_cfg_reg(H18_24XX_MMC_CMD);
+			omap_cfg_reg(H15_24XX_MMC_CLKI);
+			omap_cfg_reg(G19_24XX_MMC_CLKO);
+			omap_cfg_reg(F20_24XX_MMC_DAT0);
+			omap_cfg_reg(F19_24XX_MMC_DAT_DIR0);
+			omap_cfg_reg(G18_24XX_MMC_CMD_DIR);
+		} else {
 			omap_cfg_reg(MMC_CMD);
 			omap_cfg_reg(MMC_CLK);
 			omap_cfg_reg(MMC_DAT0);
@@ -236,7 +243,14 @@ static void __init omap_init_mmc(void)
 			}
 		}
 		if (mmc->wire4) {
-			if (!cpu_is_omap24xx()) {
+			if (cpu_is_omap24xx()) {
+				omap_cfg_reg(H14_24XX_MMC_DAT1);
+				omap_cfg_reg(E19_24XX_MMC_DAT2);
+				omap_cfg_reg(D19_24XX_MMC_DAT3);
+				omap_cfg_reg(E20_24XX_MMC_DAT_DIR1);
+				omap_cfg_reg(F18_24XX_MMC_DAT_DIR2);
+				omap_cfg_reg(E18_24XX_MMC_DAT_DIR3);
+			} else {
 				omap_cfg_reg(MMC_DAT1);
 				/* NOTE:  DAT2 can be on W10 (here) or M15 */
 				if (!mmc->nomux)
diff --git a/include/asm-arm/arch-omap/irqs.h b/include/asm-arm/arch-omap/irqs.h
index 636f21cae3ee..c5bb05a69b81 100644
--- a/include/asm-arm/arch-omap/irqs.h
+++ b/include/asm-arm/arch-omap/irqs.h
@@ -262,6 +262,7 @@
 #define INT_24XX_UART1_IRQ	72
 #define INT_24XX_UART2_IRQ	73
 #define INT_24XX_UART3_IRQ	74
+#define INT_24XX_MMC_IRQ	83
 
 /* Max. 128 level 2 IRQs (OMAP1610), 192 GPIOs (OMAP730) and
  * 16 MPUIO lines */
diff --git a/include/asm-arm/arch-omap/mux.h b/include/asm-arm/arch-omap/mux.h
index 679869c5e68f..a6df97c11839 100644
--- a/include/asm-arm/arch-omap/mux.h
+++ b/include/asm-arm/arch-omap/mux.h
@@ -461,6 +461,20 @@ enum omap24xx_index {
 	K15_24XX_UART3_TX,
 	K14_24XX_UART3_RX,
 
+	/* MMC/SDIO */
+	G19_24XX_MMC_CLKO,
+	H18_24XX_MMC_CMD,
+	F20_24XX_MMC_DAT0,
+	H14_24XX_MMC_DAT1,
+	E19_24XX_MMC_DAT2,
+	D19_24XX_MMC_DAT3,
+	F19_24XX_MMC_DAT_DIR0,
+	E20_24XX_MMC_DAT_DIR1,
+	F18_24XX_MMC_DAT_DIR2,
+	E18_24XX_MMC_DAT_DIR3,
+	G18_24XX_MMC_CMD_DIR,
+	H15_24XX_MMC_CLKI,
+
 	/* Keypad GPIO*/
 	T19_24XX_KBR0,
 	R19_24XX_KBR1,
-- 
cgit v1.2.3


From 75a1d10e2f110380adaa9b993fd417537e2f85ba Mon Sep 17 00:00:00 2001
From: Mark Howell <mhowell@northlink.com>
Date: Mon, 25 Sep 2006 12:41:29 +0300
Subject: ARM: OMAP: mux: add config for 16xx SPI pins

This patch adds pin mux info for the SPI master/slave interface on
OMAP16xx. Data from OMAP 1611/1612 TRM and errata. Works for me on my
1611/H2 with current git kernel.

Signed-off-by: Mark Howell <mhowell@northlink.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap1/mux.c       | 11 +++++++++++
 include/asm-arm/arch-omap/mux.h | 11 +++++++++++
 2 files changed, 22 insertions(+)

(limited to 'include')

diff --git a/arch/arm/mach-omap1/mux.c b/arch/arm/mach-omap1/mux.c
index fa74ef7af15f..5432335bc493 100644
--- a/arch/arm/mach-omap1/mux.c
+++ b/arch/arm/mach-omap1/mux.c
@@ -199,6 +199,17 @@ MUX_CFG("N14_1610_UWIRE_CS0",	 8,    9,    1,	  1,  21,   0,	  1,	 1,  1)
 MUX_CFG("P15_1610_UWIRE_CS3",	 8,   12,    1,	  1,  22,   0,	  1,	 1,  1)
 MUX_CFG("N15_1610_UWIRE_CS1",	 7,   18,    2,	  1,  14,   0,	 NA,	 0,  1)
 
+/* OMAP-1610 SPI */
+MUX_CFG("U19_1610_SPIF_SCK",	 7,    21,   6,	  1,  15,   0,	  1,	 1,  1)
+MUX_CFG("U18_1610_SPIF_DIN",	 8,    0,    6,	  1,  18,   1,	  1,	 0,  1)
+MUX_CFG("P20_1610_SPIF_DIN",	 6,    27,   4,   1,   7,   1,    1,     0,  1)
+MUX_CFG("W21_1610_SPIF_DOUT",	 8,    3,    6,	  1,  19,   0,	  1,	 0,  1)
+MUX_CFG("R18_1610_SPIF_DOUT",	 7,    9,    3,	  1,  11,   0,	  1,	 0,  1)
+MUX_CFG("N14_1610_SPIF_CS0",	 8,    9,    6,	  1,  21,   0,	  1,	 1,  1)
+MUX_CFG("N15_1610_SPIF_CS1",	 7,    18,   6,	  1,  14,   0,	  1,	 1,  1)
+MUX_CFG("T19_1610_SPIF_CS2",	 7,    15,   4,	  1,  13,   0,	  1,	 1,  1)
+MUX_CFG("P15_1610_SPIF_CS3",	 8,    12,   3,	  1,  22,   0,	  1,	 1,  1)
+
 /* OMAP-1610 Flash */
 MUX_CFG("L3_1610_FLASH_CS2B_OE",10,    6,    1,	 NA,   0,   0,	 NA,	 0,  1)
 MUX_CFG("M8_1610_FLASH_CS2B_WE",10,    3,    1,	 NA,   0,   0,	 NA,	 0,  1)
diff --git a/include/asm-arm/arch-omap/mux.h b/include/asm-arm/arch-omap/mux.h
index a6df97c11839..828cc5c114e1 100644
--- a/include/asm-arm/arch-omap/mux.h
+++ b/include/asm-arm/arch-omap/mux.h
@@ -320,6 +320,17 @@ enum omap1xxx_index {
 	P15_1610_UWIRE_CS3,
 	N15_1610_UWIRE_CS1,
 
+	/* OMAP-1610 SPI */
+	U19_1610_SPIF_SCK,
+	U18_1610_SPIF_DIN,
+	P20_1610_SPIF_DIN,
+	W21_1610_SPIF_DOUT,
+	R18_1610_SPIF_DOUT,
+	N14_1610_SPIF_CS0,
+	N15_1610_SPIF_CS1,
+	T19_1610_SPIF_CS2,
+	P15_1610_SPIF_CS3,
+
 	/* OMAP-1610 Flash */
 	L3_1610_FLASH_CS2B_OE,
 	M8_1610_FLASH_CS2B_WE,
-- 
cgit v1.2.3


From 2eaff915744b6c8db7770fa9e841fd1c0105fb0b Mon Sep 17 00:00:00 2001
From: Jonathan McDowell <noodles@earth.li>
Date: Mon, 25 Sep 2006 12:41:31 +0300
Subject: ARM: OMAP: Add some extra #defines for Amstrad Delta

This patch adds some further #defines regarding GPIOs and latch bits for
the Amstrad Delta; the drivers that use them will be submitted at a
later date but there's no reason not to have the information already
there and available for use.

Signed-off-by: Jonathan McDowell <noodles@earth.li>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 include/asm-arm/arch-omap/board-ams-delta.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include')

diff --git a/include/asm-arm/arch-omap/board-ams-delta.h b/include/asm-arm/arch-omap/board-ams-delta.h
index 0070f6d3b75c..9aee15d97145 100644
--- a/include/asm-arm/arch-omap/board-ams-delta.h
+++ b/include/asm-arm/arch-omap/board-ams-delta.h
@@ -50,9 +50,20 @@
 #define AMS_DELTA_LATCH2_NAND_NWE	0x0020
 #define AMS_DELTA_LATCH2_NAND_ALE	0x0040
 #define AMS_DELTA_LATCH2_NAND_CLE	0x0080
+#define AMD_DELTA_LATCH2_KEYBRD_PWR	0x0100
+#define AMD_DELTA_LATCH2_KEYBRD_DATA	0x0200
+#define AMD_DELTA_LATCH2_SCARD_RSTIN	0x0400
+#define AMD_DELTA_LATCH2_SCARD_CMDVCC	0x0800
 #define AMS_DELTA_LATCH2_MODEM_NRESET	0x1000
 #define AMS_DELTA_LATCH2_MODEM_CODEC	0x2000
 
+#define AMS_DELTA_GPIO_PIN_KEYBRD_DATA	0
+#define AMS_DELTA_GPIO_PIN_KEYBRD_CLK	1
+#define AMS_DELTA_GPIO_PIN_MODEM_IRQ	2
+#define AMS_DELTA_GPIO_PIN_HOOK_SWITCH	4
+#define AMS_DELTA_GPIO_PIN_SCARD_NOFF	6
+#define AMS_DELTA_GPIO_PIN_SCARD_IO	7
+#define AMS_DELTA_GPIO_PIN_CONFIG	11
 #define AMS_DELTA_GPIO_PIN_NAND_RB	12
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.3


From f37e4580c409e290f6e482007c3573cdb4470bf9 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@solidboot.com>
Date: Mon, 25 Sep 2006 12:41:33 +0300
Subject: ARM: OMAP2: Dynamic allocator for GPMC memory space

Add support for assigning memory regions dynamically to peripherals
attached to GPMC interface. Platform specific code should now call
gpmc_cs_request to get a free GPMC memory region instead of using
a fixed address.

Make the H4 and Apollon platform initialization use the new API.

Signed-off-by: Imre Deak <imre.deak@solidboot.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/gpmc.c       | 180 ++++++++++++++++++++++++++++++++++++++-
 include/asm-arm/arch-omap/gpmc.h |   4 +-
 2 files changed, 180 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-omap2/gpmc.c b/arch/arm/mach-omap2/gpmc.c
index c7a48f921fef..f4f04d87df32 100644
--- a/arch/arm/mach-omap2/gpmc.c
+++ b/arch/arm/mach-omap2/gpmc.c
@@ -13,6 +13,8 @@
 #include <linux/init.h>
 #include <linux/err.h>
 #include <linux/clk.h>
+#include <linux/ioport.h>
+#include <linux/spinlock.h>
 
 #include <asm/io.h>
 #include <asm/arch/gpmc.h>
@@ -41,6 +43,19 @@
 #define GPMC_CS0		0x60
 #define GPMC_CS_SIZE		0x30
 
+#define GPMC_CS_NUM		8
+#define GPMC_MEM_START		0x00000000
+#define GPMC_MEM_END		0x3FFFFFFF
+#define BOOT_ROM_SPACE		0x100000	/* 1MB */
+
+#define GPMC_CHUNK_SHIFT	24		/* 16 MB */
+#define GPMC_SECTION_SHIFT	28		/* 128 MB */
+
+static struct resource	gpmc_mem_root;
+static struct resource	gpmc_cs_mem[GPMC_CS_NUM];
+static spinlock_t	gpmc_mem_lock = SPIN_LOCK_UNLOCKED;
+static unsigned		gpmc_cs_map;
+
 static void __iomem *gpmc_base =
 	(void __iomem *) IO_ADDRESS(GPMC_BASE);
 static void __iomem *gpmc_cs_base =
@@ -187,9 +202,168 @@ int gpmc_cs_set_timings(int cs, const struct gpmc_timings *t)
 	return 0;
 }
 
-unsigned long gpmc_cs_get_base_addr(int cs)
+static void gpmc_cs_enable_mem(int cs, u32 base, u32 size)
+{
+	u32 l;
+	u32 mask;
+
+	mask = (1 << GPMC_SECTION_SHIFT) - size;
+	l = gpmc_cs_read_reg(cs, GPMC_CS_CONFIG7);
+	l &= ~0x3f;
+	l = (base >> GPMC_CHUNK_SHIFT) & 0x3f;
+	l &= ~(0x0f << 8);
+	l |= ((mask >> GPMC_CHUNK_SHIFT) & 0x0f) << 8;
+	l |= 1 << 6;		/* CSVALID */
+	gpmc_cs_write_reg(cs, GPMC_CS_CONFIG7, l);
+}
+
+static void gpmc_cs_disable_mem(int cs)
+{
+	u32 l;
+
+	l = gpmc_cs_read_reg(cs, GPMC_CS_CONFIG7);
+	l &= ~(1 << 6);		/* CSVALID */
+	gpmc_cs_write_reg(cs, GPMC_CS_CONFIG7, l);
+}
+
+static void gpmc_cs_get_memconf(int cs, u32 *base, u32 *size)
+{
+	u32 l;
+	u32 mask;
+
+	l = gpmc_cs_read_reg(cs, GPMC_CS_CONFIG7);
+	*base = (l & 0x3f) << GPMC_CHUNK_SHIFT;
+	mask = (l >> 8) & 0x0f;
+	*size = (1 << GPMC_SECTION_SHIFT) - (mask << GPMC_CHUNK_SHIFT);
+}
+
+static int gpmc_cs_mem_enabled(int cs)
+{
+	u32 l;
+
+	l = gpmc_cs_read_reg(cs, GPMC_CS_CONFIG7);
+	return l & (1 << 6);
+}
+
+static void gpmc_cs_set_reserved(int cs, int reserved)
 {
-	return (gpmc_cs_read_reg(cs, GPMC_CS_CONFIG7) & 0x1f) << 24;
+	gpmc_cs_map &= ~(1 << cs);
+	gpmc_cs_map |= (reserved ? 1 : 0) << cs;
+}
+
+static int gpmc_cs_reserved(int cs)
+{
+	return gpmc_cs_map & (1 << cs);
+}
+
+static unsigned long gpmc_mem_align(unsigned long size)
+{
+	int order;
+
+	size = (size - 1) >> (GPMC_CHUNK_SHIFT - 1);
+	order = GPMC_CHUNK_SHIFT - 1;
+	do {
+		size >>= 1;
+		order++;
+	} while (size);
+	size = 1 << order;
+	return size;
+}
+
+static int gpmc_cs_insert_mem(int cs, unsigned long base, unsigned long size)
+{
+	struct resource	*res = &gpmc_cs_mem[cs];
+	int r;
+
+	size = gpmc_mem_align(size);
+	spin_lock(&gpmc_mem_lock);
+	res->start = base;
+	res->end = base + size - 1;
+	r = request_resource(&gpmc_mem_root, res);
+	spin_unlock(&gpmc_mem_lock);
+
+	return r;
+}
+
+int gpmc_cs_request(int cs, unsigned long size, unsigned long *base)
+{
+	struct resource *res = &gpmc_cs_mem[cs];
+	int r = -1;
+
+	if (cs > GPMC_CS_NUM)
+		return -ENODEV;
+
+	size = gpmc_mem_align(size);
+	if (size > (1 << GPMC_SECTION_SHIFT))
+		return -ENOMEM;
+
+	spin_lock(&gpmc_mem_lock);
+	if (gpmc_cs_reserved(cs)) {
+		r = -EBUSY;
+		goto out;
+	}
+	if (gpmc_cs_mem_enabled(cs))
+		r = adjust_resource(res, res->start & ~(size - 1), size);
+	if (r < 0)
+		r = allocate_resource(&gpmc_mem_root, res, size, 0, ~0,
+				      size, NULL, NULL);
+	if (r < 0)
+		goto out;
+
+	gpmc_cs_enable_mem(cs, res->start, res->end - res->start + 1);
+	*base = res->start;
+	gpmc_cs_set_reserved(cs, 1);
+out:
+	spin_unlock(&gpmc_mem_lock);
+	return r;
+}
+
+void gpmc_cs_free(int cs)
+{
+	spin_lock(&gpmc_mem_lock);
+	if (cs >= GPMC_CS_NUM || !gpmc_cs_reserved(cs)) {
+		printk(KERN_ERR "Trying to free non-reserved GPMC CS%d\n", cs);
+		BUG();
+		spin_unlock(&gpmc_mem_lock);
+		return;
+	}
+	gpmc_cs_disable_mem(cs);
+	release_resource(&gpmc_cs_mem[cs]);
+	gpmc_cs_set_reserved(cs, 0);
+	spin_unlock(&gpmc_mem_lock);
+}
+
+void __init gpmc_mem_init(void)
+{
+	int cs;
+	unsigned long boot_rom_space = 0;
+
+	if (cpu_is_omap242x()) {
+		u32 l;
+		l = omap_readl(OMAP242X_CONTROL_STATUS);
+		/* In case of internal boot the 1st MB is redirected to the
+		 * boot ROM memory space.
+		 */
+		if (l & (1 << 3))
+			boot_rom_space = BOOT_ROM_SPACE;
+	} else
+		/* We assume internal boot if the mode can't be
+		 * determined.
+		 */
+		boot_rom_space = BOOT_ROM_SPACE;
+	gpmc_mem_root.start = GPMC_MEM_START + boot_rom_space;
+	gpmc_mem_root.end = GPMC_MEM_END;
+
+	/* Reserve all regions that has been set up by bootloader */
+	for (cs = 0; cs < GPMC_CS_NUM; cs++) {
+		u32 base, size;
+
+		if (!gpmc_cs_mem_enabled(cs))
+			continue;
+		gpmc_cs_get_memconf(cs, &base, &size);
+		if (gpmc_cs_insert_mem(cs, base, size) < 0)
+			BUG();
+	}
 }
 
 void __init gpmc_init(void)
@@ -206,4 +380,6 @@ void __init gpmc_init(void)
 	l &= 0x03 << 3;
 	l |= (0x02 << 3) | (1 << 0);
 	gpmc_write_reg(GPMC_SYSCONFIG, l);
+
+	gpmc_mem_init();
 }
diff --git a/include/asm-arm/arch-omap/gpmc.h b/include/asm-arm/arch-omap/gpmc.h
index 1a0a5207822d..7c03ef6c14c4 100644
--- a/include/asm-arm/arch-omap/gpmc.h
+++ b/include/asm-arm/arch-omap/gpmc.h
@@ -85,7 +85,7 @@ extern void gpmc_cs_write_reg(int cs, int idx, u32 val);
 extern u32 gpmc_cs_read_reg(int cs, int idx);
 extern int gpmc_cs_calc_divider(int cs, unsigned int sync_clk);
 extern int gpmc_cs_set_timings(int cs, const struct gpmc_timings *t);
-extern unsigned long gpmc_cs_get_base_addr(int cs);
-
+extern int gpmc_cs_request(int cs, unsigned long size, unsigned long *base);
+extern void gpmc_cs_free(int cs);
 
 #endif
-- 
cgit v1.2.3


From 123e9a5573098dbb10194c18d6d575620d0e94f3 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Mon, 25 Sep 2006 12:41:34 +0300
Subject: ARM: OMAP: DMA source and destination addresses are unsigned

Also export some omap24xx specific DMA functions.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/dma.c        | 2 ++
 include/asm-arm/arch-omap/dma.h | 4 ++--
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/arm/plat-omap/dma.c b/arch/arm/plat-omap/dma.c
index 9948d7a26339..0889769cb225 100644
--- a/arch/arm/plat-omap/dma.c
+++ b/arch/arm/plat-omap/dma.c
@@ -1422,11 +1422,13 @@ EXPORT_SYMBOL(omap_request_dma);
 EXPORT_SYMBOL(omap_free_dma);
 EXPORT_SYMBOL(omap_start_dma);
 EXPORT_SYMBOL(omap_stop_dma);
+EXPORT_SYMBOL(omap_set_dma_callback);
 EXPORT_SYMBOL(omap_enable_dma_irq);
 EXPORT_SYMBOL(omap_disable_dma_irq);
 
 EXPORT_SYMBOL(omap_set_dma_transfer_params);
 EXPORT_SYMBOL(omap_set_dma_color_mode);
+EXPORT_SYMBOL(omap_set_dma_write_mode);
 
 EXPORT_SYMBOL(omap_set_dma_src_params);
 EXPORT_SYMBOL(omap_set_dma_src_index);
diff --git a/include/asm-arm/arch-omap/dma.h b/include/asm-arm/arch-omap/dma.h
index 1b1b02307e77..33cd48d9a853 100644
--- a/include/asm-arm/arch-omap/dma.h
+++ b/include/asm-arm/arch-omap/dma.h
@@ -338,13 +338,13 @@ struct omap_dma_channel_params {
 
 	int src_port;		/* Only on OMAP1 REVISIT: Is this needed? */
 	int src_amode;		/* constant , post increment, indexed , double indexed */
-	int src_start;		/* source address : physical */
+	unsigned long src_start;	/* source address : physical */
 	int src_ei;		/* source element index */
 	int src_fi;		/* source frame index */
 
 	int dst_port;		/* Only on OMAP1 REVISIT: Is this needed? */
 	int dst_amode;		/* constant , post increment, indexed , double indexed */
-	int dst_start;		/* source address : physical */
+	unsigned long dst_start;	/* source address : physical */
 	int dst_ei;		/* source element index */
 	int dst_fi;		/* source frame index */
 
-- 
cgit v1.2.3


From 12583a70ac6b6641905e37fdd61a7f711fb4ce2b Mon Sep 17 00:00:00 2001
From: Timo Teras <timo.teras@solidboot.com>
Date: Mon, 25 Sep 2006 12:41:42 +0300
Subject: ARM: OMAP: Add enable/disable functions for dmtimer

Add enable/disable functions which effectively control the GPT iclk and fclk.

Signed-off-by: Timo Teras <timo.teras@solidboot.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/dmtimer.c        | 73 ++++++++++++++++---------------------
 include/asm-arm/arch-omap/dmtimer.h |  2 +
 2 files changed, 33 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/arch/arm/plat-omap/dmtimer.c b/arch/arm/plat-omap/dmtimer.c
index fb2a23e7c172..8d6197497a74 100644
--- a/arch/arm/plat-omap/dmtimer.c
+++ b/arch/arm/plat-omap/dmtimer.c
@@ -75,6 +75,7 @@ struct omap_dm_timer {
 #endif
 	void __iomem *io_base;
 	unsigned reserved:1;
+	unsigned enabled:1;
 };
 
 #ifdef CONFIG_ARCH_OMAP1
@@ -164,7 +165,7 @@ static void omap_dm_timer_reset(struct omap_dm_timer *timer)
 		omap_dm_timer_write_reg(timer, OMAP_TIMER_IF_CTRL_REG, 0x06);
 		omap_dm_timer_wait_for_reset(timer);
 	}
-	omap_dm_timer_set_source(timer, OMAP_TIMER_SRC_SYS_CLK);
+	omap_dm_timer_set_source(timer, OMAP_TIMER_SRC_32_KHZ);
 
 	/* Set to smart-idle mode */
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_OCP_CFG_REG);
@@ -174,15 +175,8 @@ static void omap_dm_timer_reset(struct omap_dm_timer *timer)
 
 static void omap_dm_timer_prepare(struct omap_dm_timer *timer)
 {
-	omap_dm_clk_enable(timer->fclk);
-	omap_dm_clk_enable(timer->iclk);
-
+	omap_dm_timer_enable(timer);
 	omap_dm_timer_reset(timer);
-
-	/* Leave iclk enabled for GPT1 as it is needed for the
-	 * system timer to work properly. */
-	if (timer != &dm_timers[0])
-		omap_dm_clk_disable(timer->iclk);
 }
 
 struct omap_dm_timer *omap_dm_timer_request(void)
@@ -233,18 +227,36 @@ struct omap_dm_timer *omap_dm_timer_request_specific(int id)
 
 void omap_dm_timer_free(struct omap_dm_timer *timer)
 {
-	omap_dm_clk_enable(timer->iclk);
+	omap_dm_timer_enable(timer);
 	omap_dm_timer_reset(timer);
-	omap_dm_clk_disable(timer->iclk);
-
-	if (timer == &dm_timers[0])
-		omap_dm_clk_disable(timer->iclk);
-	omap_dm_clk_disable(timer->fclk);
+	omap_dm_timer_disable(timer);
 
 	WARN_ON(!timer->reserved);
 	timer->reserved = 0;
 }
 
+void omap_dm_timer_enable(struct omap_dm_timer *timer)
+{
+	if (timer->enabled)
+		return;
+
+	omap_dm_clk_enable(timer->fclk);
+	omap_dm_clk_enable(timer->iclk);
+
+	timer->enabled = 1;
+}
+
+void omap_dm_timer_disable(struct omap_dm_timer *timer)
+{
+	if (!timer->enabled)
+		return;
+
+	omap_dm_clk_disable(timer->iclk);
+	omap_dm_clk_disable(timer->fclk);
+
+	timer->enabled = 0;
+}
+
 int omap_dm_timer_get_irq(struct omap_dm_timer *timer)
 {
 	return timer->irq;
@@ -301,35 +313,29 @@ __u32 omap_dm_timer_modify_idlect_mask(__u32 inputmask)
 
 void omap_dm_timer_trigger(struct omap_dm_timer *timer)
 {
-	omap_dm_clk_enable(timer->iclk);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_TRIGGER_REG, 0);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 void omap_dm_timer_start(struct omap_dm_timer *timer)
 {
 	u32 l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
 	if (!(l & OMAP_TIMER_CTRL_ST)) {
 		l |= OMAP_TIMER_CTRL_ST;
 		omap_dm_timer_write_reg(timer, OMAP_TIMER_CTRL_REG, l);
 	}
-	omap_dm_clk_disable(timer->iclk);
 }
 
 void omap_dm_timer_stop(struct omap_dm_timer *timer)
 {
 	u32 l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
 	if (l & OMAP_TIMER_CTRL_ST) {
 		l &= ~0x1;
 		omap_dm_timer_write_reg(timer, OMAP_TIMER_CTRL_REG, l);
 	}
-	omap_dm_clk_disable(timer->iclk);
 }
 
 #ifdef CONFIG_ARCH_OMAP1
@@ -367,7 +373,6 @@ void omap_dm_timer_set_load(struct omap_dm_timer *timer, int autoreload,
 {
 	u32 l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
 	if (autoreload)
 		l |= OMAP_TIMER_CTRL_AR;
@@ -376,7 +381,6 @@ void omap_dm_timer_set_load(struct omap_dm_timer *timer, int autoreload,
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_CTRL_REG, l);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_LOAD_REG, load);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_TRIGGER_REG, 0);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 void omap_dm_timer_set_match(struct omap_dm_timer *timer, int enable,
@@ -384,7 +388,6 @@ void omap_dm_timer_set_match(struct omap_dm_timer *timer, int enable,
 {
 	u32 l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
 	if (enable)
 		l |= OMAP_TIMER_CTRL_CE;
@@ -392,7 +395,6 @@ void omap_dm_timer_set_match(struct omap_dm_timer *timer, int enable,
 		l &= ~OMAP_TIMER_CTRL_CE;
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_CTRL_REG, l);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_MATCH_REG, match);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 
@@ -401,7 +403,6 @@ void omap_dm_timer_set_pwm(struct omap_dm_timer *timer, int def_on,
 {
 	u32 l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
 	l &= ~(OMAP_TIMER_CTRL_GPOCFG | OMAP_TIMER_CTRL_SCPWM |
 	       OMAP_TIMER_CTRL_PT | (0x03 << 10));
@@ -411,14 +412,12 @@ void omap_dm_timer_set_pwm(struct omap_dm_timer *timer, int def_on,
 		l |= OMAP_TIMER_CTRL_PT;
 	l |= trigger << 10;
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_CTRL_REG, l);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 void omap_dm_timer_set_prescaler(struct omap_dm_timer *timer, int prescaler)
 {
 	u32 l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
 	l &= ~(OMAP_TIMER_CTRL_PRE | (0x07 << 2));
 	if (prescaler >= 0x00 && prescaler <= 0x07) {
@@ -426,51 +425,40 @@ void omap_dm_timer_set_prescaler(struct omap_dm_timer *timer, int prescaler)
 		l |= prescaler << 2;
 	}
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_CTRL_REG, l);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 void omap_dm_timer_set_int_enable(struct omap_dm_timer *timer,
 				  unsigned int value)
 {
-	omap_dm_clk_enable(timer->iclk);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_INT_EN_REG, value);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 unsigned int omap_dm_timer_read_status(struct omap_dm_timer *timer)
 {
 	unsigned int l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_STAT_REG);
-	omap_dm_clk_disable(timer->iclk);
 
 	return l;
 }
 
 void omap_dm_timer_write_status(struct omap_dm_timer *timer, unsigned int value)
 {
-	omap_dm_clk_enable(timer->iclk);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_STAT_REG, value);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 unsigned int omap_dm_timer_read_counter(struct omap_dm_timer *timer)
 {
 	unsigned int l;
 
-	omap_dm_clk_enable(timer->iclk);
 	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_COUNTER_REG);
-	omap_dm_clk_disable(timer->iclk);
 
 	return l;
 }
 
 void omap_dm_timer_write_counter(struct omap_dm_timer *timer, unsigned int value)
 {
-	omap_dm_clk_enable(timer->iclk);
 	omap_dm_timer_write_reg(timer, OMAP_TIMER_COUNTER_REG, value);
-	omap_dm_clk_disable(timer->iclk);
 }
 
 int omap_dm_timers_active(void)
@@ -481,13 +469,14 @@ int omap_dm_timers_active(void)
 		struct omap_dm_timer *timer;
 
 		timer = &dm_timers[i];
-		omap_dm_clk_enable(timer->iclk);
+
+		if (!timer->enabled)
+			continue;
+
 		if (omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG) &
 		    OMAP_TIMER_CTRL_ST) {
-			omap_dm_clk_disable(timer->iclk);
 			return 1;
 		}
-		omap_dm_clk_disable(timer->iclk);
 	}
 	return 0;
 }
diff --git a/include/asm-arm/arch-omap/dmtimer.h b/include/asm-arm/arch-omap/dmtimer.h
index 7a289ff07404..b5f3a71b899d 100644
--- a/include/asm-arm/arch-omap/dmtimer.h
+++ b/include/asm-arm/arch-omap/dmtimer.h
@@ -52,6 +52,8 @@ int omap_dm_timer_init(void);
 struct omap_dm_timer *omap_dm_timer_request(void);
 struct omap_dm_timer *omap_dm_timer_request_specific(int timer_id);
 void omap_dm_timer_free(struct omap_dm_timer *timer);
+void omap_dm_timer_enable(struct omap_dm_timer *timer);
+void omap_dm_timer_disable(struct omap_dm_timer *timer);
 
 int omap_dm_timer_get_irq(struct omap_dm_timer *timer);
 
-- 
cgit v1.2.3


From 709eb3e5ccb304dca011c41456da5ffd246d7271 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Mon, 25 Sep 2006 12:45:45 +0300
Subject: ARM: OMAP: Sync DMA with linux-omap tree

This patch syncs OMAP DMA code with linux-omap tree.
Mostly allow changing DMA callback function and set
OMAP2 specific transfer mode.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/dma.c        | 85 +++++++++++++++++++++++++++++++----------
 include/asm-arm/arch-omap/dma.h | 12 +++++-
 2 files changed, 75 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/arm/plat-omap/dma.c b/arch/arm/plat-omap/dma.c
index 0889769cb225..1bbb431843ce 100644
--- a/arch/arm/plat-omap/dma.c
+++ b/arch/arm/plat-omap/dma.c
@@ -119,32 +119,41 @@ static void clear_lch_regs(int lch)
 		omap_writew(0, lch_base + i);
 }
 
-void omap_set_dma_priority(int dst_port, int priority)
+void omap_set_dma_priority(int lch, int dst_port, int priority)
 {
 	unsigned long reg;
 	u32 l;
 
-	switch (dst_port) {
-	case OMAP_DMA_PORT_OCP_T1:	/* FFFECC00 */
-		reg = OMAP_TC_OCPT1_PRIOR;
-		break;
-	case OMAP_DMA_PORT_OCP_T2:	/* FFFECCD0 */
-		reg = OMAP_TC_OCPT2_PRIOR;
-		break;
-	case OMAP_DMA_PORT_EMIFF:	/* FFFECC08 */
-		reg = OMAP_TC_EMIFF_PRIOR;
-		break;
-	case OMAP_DMA_PORT_EMIFS:	/* FFFECC04 */
-		reg = OMAP_TC_EMIFS_PRIOR;
-		break;
-	default:
-		BUG();
-		return;
+	if (cpu_class_is_omap1()) {
+		switch (dst_port) {
+		case OMAP_DMA_PORT_OCP_T1:	/* FFFECC00 */
+			reg = OMAP_TC_OCPT1_PRIOR;
+			break;
+		case OMAP_DMA_PORT_OCP_T2:	/* FFFECCD0 */
+			reg = OMAP_TC_OCPT2_PRIOR;
+			break;
+		case OMAP_DMA_PORT_EMIFF:	/* FFFECC08 */
+			reg = OMAP_TC_EMIFF_PRIOR;
+			break;
+		case OMAP_DMA_PORT_EMIFS:	/* FFFECC04 */
+			reg = OMAP_TC_EMIFS_PRIOR;
+			break;
+		default:
+			BUG();
+			return;
+		}
+		l = omap_readl(reg);
+		l &= ~(0xf << 8);
+		l |= (priority & 0xf) << 8;
+		omap_writel(l, reg);
+	}
+
+	if (cpu_is_omap24xx()) {
+		if (priority)
+			OMAP_DMA_CCR_REG(lch) |= (1 << 6);
+		else
+			OMAP_DMA_CCR_REG(lch) &= ~(1 << 6);
 	}
-	l = omap_readl(reg);
-	l &= ~(0xf << 8);
-	l |= (priority & 0xf) << 8;
-	omap_writel(l, reg);
 }
 
 void omap_set_dma_transfer_params(int lch, int data_type, int elem_count,
@@ -234,6 +243,14 @@ void omap_set_dma_color_mode(int lch, enum omap_dma_color_mode mode, u32 color)
 	OMAP1_DMA_LCH_CTRL_REG(lch) = w;
 }
 
+void omap_set_dma_write_mode(int lch, enum omap_dma_write_mode mode)
+{
+	if (cpu_is_omap24xx()) {
+		OMAP_DMA_CSDP_REG(lch) &= ~(0x3 << 16);
+		OMAP_DMA_CSDP_REG(lch) |= (mode << 16);
+	}
+}
+
 /* Note that src_port is only for omap1 */
 void omap_set_dma_src_params(int lch, int src_port, int src_amode,
 			     unsigned long src_start,
@@ -697,6 +714,32 @@ void omap_stop_dma(int lch)
 	dma_chan[lch].flags &= ~OMAP_DMA_ACTIVE;
 }
 
+/*
+ * Allows changing the DMA callback function or data. This may be needed if
+ * the driver shares a single DMA channel for multiple dma triggers.
+ */
+int omap_set_dma_callback(int lch,
+			  void (* callback)(int lch, u16 ch_status, void *data),
+			  void *data)
+{
+	unsigned long flags;
+
+	if (lch < 0)
+		return -ENODEV;
+
+	spin_lock_irqsave(&dma_chan_lock, flags);
+	if (dma_chan[lch].dev_id == -1) {
+		printk(KERN_ERR "DMA callback for not set for free channel\n");
+		spin_unlock_irqrestore(&dma_chan_lock, flags);
+		return -EINVAL;
+	}
+	dma_chan[lch].callback = callback;
+	dma_chan[lch].data = data;
+	spin_unlock_irqrestore(&dma_chan_lock, flags);
+
+	return 0;
+}
+
 /*
  * Returns current physical source address for the given DMA channel.
  * If the channel is running the caller must disable interrupts prior calling
diff --git a/include/asm-arm/arch-omap/dma.h b/include/asm-arm/arch-omap/dma.h
index 33cd48d9a853..d591d0585bba 100644
--- a/include/asm-arm/arch-omap/dma.h
+++ b/include/asm-arm/arch-omap/dma.h
@@ -331,6 +331,12 @@ enum omap_dma_color_mode {
 	OMAP_DMA_TRANSPARENT_COPY
 };
 
+enum omap_dma_write_mode {
+	OMAP_DMA_WRITE_NON_POSTED = 0,
+	OMAP_DMA_WRITE_POSTED,
+	OMAP_DMA_WRITE_LAST_NON_POSTED
+};
+
 struct omap_dma_channel_params {
 	int data_type;		/* data type 8,16,32 */
 	int elem_count;		/* number of elements in a frame */
@@ -356,7 +362,7 @@ struct omap_dma_channel_params {
 };
 
 
-extern void omap_set_dma_priority(int dst_port, int priority);
+extern void omap_set_dma_priority(int lch, int dst_port, int priority);
 extern int omap_request_dma(int dev_id, const char *dev_name,
 			    void (* callback)(int lch, u16 ch_status, void *data),
 			    void *data, int *dma_ch);
@@ -371,6 +377,7 @@ extern void omap_set_dma_transfer_params(int lch, int data_type,
 					 int dma_trigger, int src_or_dst_synch);
 extern void omap_set_dma_color_mode(int lch, enum omap_dma_color_mode mode,
 				    u32 color);
+extern void omap_set_dma_write_mode(int lch, enum omap_dma_write_mode mode);
 
 extern void omap_set_dma_src_params(int lch, int src_port, int src_amode,
 				    unsigned long src_start,
@@ -394,6 +401,9 @@ extern void omap_set_dma_params(int lch,
 extern void omap_dma_link_lch (int lch_head, int lch_queue);
 extern void omap_dma_unlink_lch (int lch_head, int lch_queue);
 
+extern int omap_set_dma_callback(int lch,
+			void (* callback)(int lch, u16 ch_status, void *data),
+			void *data);
 extern dma_addr_t omap_get_dma_src_pos(int lch);
 extern dma_addr_t omap_get_dma_dst_pos(int lch);
 extern int omap_get_dma_src_addr_counter(int lch);
-- 
cgit v1.2.3


From 90afd5cb2ac0977c38e83b6b21493da911b242b3 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Mon, 25 Sep 2006 13:27:20 +0300
Subject: ARM: OMAP: Sync clocks with linux-omap tree

Mostly clean up CONFIG_OMAP_RESET_CLOCKS. Also includes a
patch from Imre Deak to make McSPI clocks use id.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap1/clock.c       | 107 +++++++++++++++++---------------------
 arch/arm/mach-omap1/clock.h       |  14 +++++
 arch/arm/mach-omap2/clock.c       |  39 ++++++--------
 arch/arm/mach-omap2/clock.h       |  18 ++++---
 arch/arm/plat-omap/clock.c        |  25 +++++++++
 include/asm-arm/arch-omap/clock.h |   1 +
 6 files changed, 117 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-omap1/clock.c b/arch/arm/mach-omap1/clock.c
index f1958e882e86..638490e62d5f 100644
--- a/arch/arm/mach-omap1/clock.c
+++ b/arch/arm/mach-omap1/clock.c
@@ -20,6 +20,7 @@
 #include <linux/clk.h>
 
 #include <asm/io.h>
+#include <asm/mach-types.h>
 
 #include <asm/arch/cpu.h>
 #include <asm/arch/usb.h>
@@ -586,77 +587,53 @@ static int omap1_clk_set_rate(struct clk *clk, unsigned long rate)
  *-------------------------------------------------------------------------*/
 
 #ifdef CONFIG_OMAP_RESET_CLOCKS
-/*
- * Resets some clocks that may be left on from bootloader,
- * but leaves serial clocks on. See also omap_late_clk_reset().
- */
-static inline void omap1_early_clk_reset(void)
-{
-	//omap_writel(0x3 << 29, MOD_CONF_CTRL_0);
-}
 
-static int __init omap1_late_clk_reset(void)
+static void __init omap1_clk_disable_unused(struct clk *clk)
 {
-	/* Turn off all unused clocks */
-	struct clk *p;
 	__u32 regval32;
 
-	/* USB_REQ_EN will be disabled later if necessary (usb_dc_ck) */
-	regval32 = omap_readw(SOFT_REQ_REG) & (1 << 4);
-	omap_writew(regval32, SOFT_REQ_REG);
-	omap_writew(0, SOFT_REQ_REG2);
-
-	list_for_each_entry(p, &clocks, node) {
-		if (p->usecount > 0 || (p->flags & ALWAYS_ENABLED) ||
-			p->enable_reg == 0)
-			continue;
-
-		/* Clocks in the DSP domain need api_ck. Just assume bootloader
-		 * has not enabled any DSP clocks */
-		if ((u32)p->enable_reg == DSP_IDLECT2) {
-			printk(KERN_INFO "Skipping reset check for DSP domain "
-			       "clock \"%s\"\n", p->name);
-			continue;
-		}
+	/* Clocks in the DSP domain need api_ck. Just assume bootloader
+	 * has not enabled any DSP clocks */
+	if ((u32)clk->enable_reg == DSP_IDLECT2) {
+		printk(KERN_INFO "Skipping reset check for DSP domain "
+		       "clock \"%s\"\n", clk->name);
+		return;
+	}
 
-		/* Is the clock already disabled? */
-		if (p->flags & ENABLE_REG_32BIT) {
-			if (p->flags & VIRTUAL_IO_ADDRESS)
-				regval32 = __raw_readl(p->enable_reg);
-			else
-				regval32 = omap_readl(p->enable_reg);
-		} else {
-			if (p->flags & VIRTUAL_IO_ADDRESS)
-				regval32 = __raw_readw(p->enable_reg);
+	/* Is the clock already disabled? */
+	if (clk->flags & ENABLE_REG_32BIT) {
+		if (clk->flags & VIRTUAL_IO_ADDRESS)
+			regval32 = __raw_readl(clk->enable_reg);
 			else
-				regval32 = omap_readw(p->enable_reg);
-		}
-
-		if ((regval32 & (1 << p->enable_bit)) == 0)
-			continue;
+				regval32 = omap_readl(clk->enable_reg);
+	} else {
+		if (clk->flags & VIRTUAL_IO_ADDRESS)
+			regval32 = __raw_readw(clk->enable_reg);
+		else
+			regval32 = omap_readw(clk->enable_reg);
+	}
 
-		/* FIXME: This clock seems to be necessary but no-one
-		 * has asked for its activation. */
-		if (p == &tc2_ck         // FIX: pm.c (SRAM), CCP, Camera
-		    || p == &ck_dpll1out.clk // FIX: SoSSI, SSR
-		    || p == &arm_gpio_ck // FIX: GPIO code for 1510
-		    ) {
-			printk(KERN_INFO "FIXME: Clock \"%s\" seems unused\n",
-			       p->name);
-			continue;
-		}
+	if ((regval32 & (1 << clk->enable_bit)) == 0)
+		return;
 
-		printk(KERN_INFO "Disabling unused clock \"%s\"... ", p->name);
-		p->disable(p);
-		printk(" done\n");
+	/* FIXME: This clock seems to be necessary but no-one
+	 * has asked for its activation. */
+	if (clk == &tc2_ck		// FIX: pm.c (SRAM), CCP, Camera
+	    || clk == &ck_dpll1out.clk	// FIX: SoSSI, SSR
+	    || clk == &arm_gpio_ck	// FIX: GPIO code for 1510
+		) {
+		printk(KERN_INFO "FIXME: Clock \"%s\" seems unused\n",
+		       clk->name);
+		return;
 	}
 
-	return 0;
+	printk(KERN_INFO "Disabling unused clock \"%s\"... ", clk->name);
+	clk->disable(clk);
+	printk(" done\n");
 }
-late_initcall(omap1_late_clk_reset);
 
 #else
-#define omap1_early_clk_reset()	{}
+#define omap1_clk_disable_unused	NULL
 #endif
 
 static struct clk_functions omap1_clk_functions = {
@@ -664,6 +641,7 @@ static struct clk_functions omap1_clk_functions = {
 	.clk_disable		= omap1_clk_disable,
 	.clk_round_rate		= omap1_clk_round_rate,
 	.clk_set_rate		= omap1_clk_set_rate,
+	.clk_disable_unused	= omap1_clk_disable_unused,
 };
 
 int __init omap1_clk_init(void)
@@ -671,8 +649,13 @@ int __init omap1_clk_init(void)
 	struct clk ** clkp;
 	const struct omap_clock_config *info;
 	int crystal_type = 0; /* Default 12 MHz */
+	u32 reg;
+
+	/* USB_REQ_EN will be disabled later if necessary (usb_dc_ck) */
+	reg = omap_readw(SOFT_REQ_REG) & (1 << 4);
+	omap_writew(reg, SOFT_REQ_REG);
+	omap_writew(0, SOFT_REQ_REG2);
 
-	omap1_early_clk_reset();
 	clk_init(&omap1_clk_functions);
 
 	/* By default all idlect1 clocks are allowed to idle */
@@ -772,6 +755,12 @@ int __init omap1_clk_init(void)
 	omap_writew(omap_readw(OMAP730_PCC_UPLD_CTRL) & ~0x1, OMAP730_PCC_UPLD_CTRL);
 #endif
 
+	/* Amstrad Delta wants BCLK high when inactive */
+	if (machine_is_ams_delta())
+		omap_writel(omap_readl(ULPD_CLOCK_CTRL) |
+				(1 << SDW_MCLK_INV_BIT),
+				ULPD_CLOCK_CTRL);
+
 	/* Turn off DSP and ARM_TIMXO. Make sure ARM_INTHCK is not divided */
 	/* (on 730, bit 13 must not be cleared) */
 	if (cpu_is_omap730())
diff --git a/arch/arm/mach-omap1/clock.h b/arch/arm/mach-omap1/clock.h
index b7c68819c4e7..f7df00205c4a 100644
--- a/arch/arm/mach-omap1/clock.h
+++ b/arch/arm/mach-omap1/clock.h
@@ -89,6 +89,7 @@ struct arm_idlect1_clk {
 #define EN_DSPTIMCK	5
 
 /* Various register defines for clock controls scattered around OMAP chip */
+#define SDW_MCLK_INV_BIT	2	/* In ULPD_CLKC_CTRL */
 #define USB_MCLK_EN_BIT		4	/* In ULPD_CLKC_CTRL */
 #define USB_HOST_HHC_UHOST_EN	9	/* In MOD_CONF_CTRL_0 */
 #define SWD_ULPD_PLL_CLK_REQ	1	/* In SWD_CLK_DIV_CTRL_SEL */
@@ -741,6 +742,18 @@ static struct clk i2c_fck = {
 	.disable	= &omap1_clk_disable_generic,
 };
 
+static struct clk i2c_ick = {
+	.name		= "i2c_ick",
+	.id		= 1,
+	.flags		= CLOCK_IN_OMAP16XX |
+			  VIRTUAL_CLOCK | CLOCK_NO_IDLE_PARENT |
+			  ALWAYS_ENABLED,
+	.parent		= &armper_ck.clk,
+	.recalc		= &followparent_recalc,
+	.enable		= &omap1_clk_enable_generic,
+	.disable	= &omap1_clk_disable_generic,
+};
+
 static struct clk * onchip_clks[] = {
 	/* non-ULPD clocks */
 	&ck_ref,
@@ -790,6 +803,7 @@ static struct clk * onchip_clks[] = {
 	/* Virtual clocks */
 	&virtual_ck_mpu,
 	&i2c_fck,
+	&i2c_ick,
 };
 
 #endif
diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c
index 26ac49ecb78d..0de201c3d50b 100644
--- a/arch/arm/mach-omap2/clock.c
+++ b/arch/arm/mach-omap2/clock.c
@@ -1025,12 +1025,29 @@ static int omap2_select_table_rate(struct clk * clk, unsigned long rate)
  * Omap2 clock reset and init functions
  *-------------------------------------------------------------------------*/
 
+#ifdef CONFIG_OMAP_RESET_CLOCKS
+static void __init omap2_clk_disable_unused(struct clk *clk)
+{
+	u32 regval32;
+
+	regval32 = __raw_readl(clk->enable_reg);
+	if ((regval32 & (1 << clk->enable_bit)) == 0)
+		return;
+
+	printk(KERN_INFO "Disabling unused clock \"%s\"\n", clk->name);
+	_omap2_clk_disable(clk);
+}
+#else
+#define omap2_clk_disable_unused	NULL
+#endif
+
 static struct clk_functions omap2_clk_functions = {
 	.clk_enable		= omap2_clk_enable,
 	.clk_disable		= omap2_clk_disable,
 	.clk_round_rate		= omap2_clk_round_rate,
 	.clk_set_rate		= omap2_clk_set_rate,
 	.clk_set_parent		= omap2_clk_set_parent,
+	.clk_disable_unused	= omap2_clk_disable_unused,
 };
 
 static void __init omap2_get_crystal_rate(struct clk *osc, struct clk *sys)
@@ -1070,28 +1087,6 @@ void omap2_clk_prepare_for_reboot(void)
 	clk_set_rate(vclk, rate);
 }
 
-#ifdef CONFIG_OMAP_RESET_CLOCKS
-static void __init omap2_disable_unused_clocks(void)
-{
-	struct clk *ck;
-	u32 regval32;
-
-	list_for_each_entry(ck, &clocks, node) {
-		if (ck->usecount > 0 || (ck->flags & ALWAYS_ENABLED) ||
-			ck->enable_reg == 0)
-			continue;
-
-		regval32 = __raw_readl(ck->enable_reg);
-		if ((regval32 & (1 << ck->enable_bit)) == 0)
-			continue;
-
-		printk(KERN_INFO "Disabling unused clock \"%s\"\n", ck->name);
-		_omap2_clk_disable(ck);
-	}
-}
-late_initcall(omap2_disable_unused_clocks);
-#endif
-
 /*
  * Switch the MPU rate if specified on cmdline.
  * We cannot do this early until cmdline is parsed.
diff --git a/arch/arm/mach-omap2/clock.h b/arch/arm/mach-omap2/clock.h
index 223152d9663f..8816f5a33a28 100644
--- a/arch/arm/mach-omap2/clock.h
+++ b/arch/arm/mach-omap2/clock.h
@@ -1368,7 +1368,8 @@ static struct clk mcbsp5_fck = {
 };
 
 static struct clk mcspi1_ick = {
-	.name		= "mcspi1_ick",
+	.name		= "mcspi_ick",
+	.id		= 1,
 	.parent		= &l4_ck,
 	.flags		= CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X,
 	.enable_reg	= (void __iomem *)&CM_ICLKEN1_CORE,
@@ -1377,7 +1378,8 @@ static struct clk mcspi1_ick = {
 };
 
 static struct clk mcspi1_fck = {
-	.name		= "mcspi1_fck",
+	.name		= "mcspi_fck",
+	.id		= 1,
 	.parent		= &func_48m_ck,
 	.flags		= CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X,
 	.enable_reg	= (void __iomem *)&CM_FCLKEN1_CORE,
@@ -1386,7 +1388,8 @@ static struct clk mcspi1_fck = {
 };
 
 static struct clk mcspi2_ick = {
-	.name		= "mcspi2_ick",
+	.name		= "mcspi_ick",
+	.id		= 2,
 	.parent		= &l4_ck,
 	.flags		= CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X,
 	.enable_reg	= (void __iomem *)&CM_ICLKEN1_CORE,
@@ -1395,7 +1398,8 @@ static struct clk mcspi2_ick = {
 };
 
 static struct clk mcspi2_fck = {
-	.name		= "mcspi2_fck",
+	.name		= "mcspi_fck",
+	.id		= 2,
 	.parent		= &func_48m_ck,
 	.flags		= CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X,
 	.enable_reg	= (void __iomem *)&CM_FCLKEN1_CORE,
@@ -1404,7 +1408,8 @@ static struct clk mcspi2_fck = {
 };
 
 static struct clk mcspi3_ick = {
-	.name		= "mcspi3_ick",
+	.name		= "mcspi_ick",
+	.id		= 3,
 	.parent		= &l4_ck,
 	.flags		= CLOCK_IN_OMAP243X,
 	.enable_reg	= (void __iomem *)&CM_ICLKEN2_CORE,
@@ -1413,7 +1418,8 @@ static struct clk mcspi3_ick = {
 };
 
 static struct clk mcspi3_fck = {
-	.name		= "mcspi3_fck",
+	.name		= "mcspi_fck",
+	.id		= 3,
 	.parent		= &func_48m_ck,
 	.flags		= CLOCK_IN_OMAP243X,
 	.enable_reg	= (void __iomem *)&CM_FCLKEN2_CORE,
diff --git a/arch/arm/plat-omap/clock.c b/arch/arm/plat-omap/clock.c
index 4abf2714f2d9..f1179ad4be1b 100644
--- a/arch/arm/plat-omap/clock.c
+++ b/arch/arm/plat-omap/clock.c
@@ -323,6 +323,31 @@ EXPORT_SYMBOL(clk_allow_idle);
 
 /*-------------------------------------------------------------------------*/
 
+#ifdef CONFIG_OMAP_RESET_CLOCKS
+/*
+ * Disable any unused clocks left on by the bootloader
+ */
+static int __init clk_disable_unused(void)
+{
+	struct clk *ck;
+	unsigned long flags;
+
+	list_for_each_entry(ck, &clocks, node) {
+		if (ck->usecount > 0 || (ck->flags & ALWAYS_ENABLED) ||
+			ck->enable_reg == 0)
+			continue;
+
+		spin_lock_irqsave(&clockfw_lock, flags);
+		if (arch_clock->clk_disable_unused)
+			arch_clock->clk_disable_unused(ck);
+		spin_unlock_irqrestore(&clockfw_lock, flags);
+	}
+
+	return 0;
+}
+late_initcall(clk_disable_unused);
+#endif
+
 int __init clk_init(struct clk_functions * custom_clocks)
 {
 	if (!custom_clocks) {
diff --git a/include/asm-arm/arch-omap/clock.h b/include/asm-arm/arch-omap/clock.h
index f83003f5287b..fa6881049903 100644
--- a/include/asm-arm/arch-omap/clock.h
+++ b/include/asm-arm/arch-omap/clock.h
@@ -45,6 +45,7 @@ struct clk_functions {
 	struct clk *	(*clk_get_parent)(struct clk *clk);
 	void		(*clk_allow_idle)(struct clk *clk);
 	void		(*clk_deny_idle)(struct clk *clk);
+	void		(*clk_disable_unused)(struct clk *clk);
 };
 
 extern unsigned int mpurate;
-- 
cgit v1.2.3


From a6d967a485c67ec8a1276261f39d81ace6a3e308 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 25 Sep 2006 15:33:09 -0400
Subject: [libata] No need for all those arch libata-portmap.h headers

They all contain the same thing.  Instead, have a single generic one in
include/asm-generic, and permit an arch to override as needed.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/asm-alpha/libata-portmap.h   | 1 -
 include/asm-frv/libata-portmap.h     | 1 -
 include/asm-i386/libata-portmap.h    | 1 -
 include/asm-ia64/libata-portmap.h    | 1 -
 include/asm-powerpc/libata-portmap.h | 1 -
 include/asm-sparc/libata-portmap.h   | 1 -
 include/asm-sparc64/libata-portmap.h | 1 -
 include/asm-x86_64/libata-portmap.h  | 1 -
 include/linux/libata.h               | 8 ++++++++
 9 files changed, 8 insertions(+), 8 deletions(-)
 delete mode 100644 include/asm-alpha/libata-portmap.h
 delete mode 100644 include/asm-frv/libata-portmap.h
 delete mode 100644 include/asm-i386/libata-portmap.h
 delete mode 100644 include/asm-ia64/libata-portmap.h
 delete mode 100644 include/asm-powerpc/libata-portmap.h
 delete mode 100644 include/asm-sparc/libata-portmap.h
 delete mode 100644 include/asm-sparc64/libata-portmap.h
 delete mode 100644 include/asm-x86_64/libata-portmap.h

(limited to 'include')

diff --git a/include/asm-alpha/libata-portmap.h b/include/asm-alpha/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-alpha/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-frv/libata-portmap.h b/include/asm-frv/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-frv/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-i386/libata-portmap.h b/include/asm-i386/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-i386/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-ia64/libata-portmap.h b/include/asm-ia64/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-ia64/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-powerpc/libata-portmap.h b/include/asm-powerpc/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-powerpc/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-sparc/libata-portmap.h b/include/asm-sparc/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-sparc/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-sparc64/libata-portmap.h b/include/asm-sparc64/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-sparc64/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-x86_64/libata-portmap.h b/include/asm-x86_64/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-x86_64/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 1ef3d3901b47..d6a3d4b345fc 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -36,7 +36,15 @@
 #include <linux/workqueue.h>
 #include <scsi/scsi_host.h>
 
+/*
+ * Define if arch has non-standard setup.  This is a _PCI_ standard
+ * not a legacy or ISA standard.
+ */
+#ifdef CONFIG_ATA_NONSTANDARD
 #include <asm/libata-portmap.h>
+#else
+#include <asm-generic/libata-portmap.h>
+#endif
 
 /*
  * compile-time options: to be removed as soon as all the drivers are
-- 
cgit v1.2.3


From baef186519c69b11cf7e48c26e75feb1e6173baa Mon Sep 17 00:00:00 2001
From: "John W. Linville" <linville@tuxdriver.com>
Date: Fri, 8 Sep 2006 16:04:05 -0400
Subject: [PATCH] WE-21 support (core API)

This is version 21 of the Wireless Extensions. Changelog :
	o finishes migrating the ESSID API (remove the +1)
	o netdev->get_wireless_stats is no more
	o long/short retry

This is a redacted version of a patch originally submitted by Jean
Tourrilhes.  I removed most of the additions, in order to minimize
future support requirements for nl80211 (or other WE successor).

CC: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/netdevice.h |  1 -
 include/linux/wireless.h  | 24 ++++++++++++++---
 net/core/net-sysfs.c      |  5 +---
 net/core/wireless.c       | 67 +++++++++++++++++++++++++++++------------------
 4 files changed, 62 insertions(+), 35 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 43289127b458..ac4f50213bc3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -334,7 +334,6 @@ struct net_device
 
 
 	struct net_device_stats* (*get_stats)(struct net_device *dev);
-	struct iw_statistics*	(*get_wireless_stats)(struct net_device *dev);
 
 	/* List of functions to handle Wireless Extensions (instead of ioctl).
 	 * See <net/iw_handler.h> for details. Jean II */
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 13588564b42b..a50a0130fd9e 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -1,7 +1,7 @@
 /*
  * This file define a set of standard wireless extensions
  *
- * Version :	20	17.2.06
+ * Version :	21	14.3.06
  *
  * Authors :	Jean Tourrilhes - HPL - <jt@hpl.hp.com>
  * Copyright (c) 1997-2006 Jean Tourrilhes, All Rights Reserved.
@@ -69,9 +69,14 @@
 
 /***************************** INCLUDES *****************************/
 
+/* This header is used in user-space, therefore need to be sanitised
+ * for that purpose. Those includes are usually not compatible with glibc.
+ * To know which includes to use in user-space, check iwlib.h. */
+#ifdef __KERNEL__
 #include <linux/types.h>		/* for "caddr_t" et al		*/
 #include <linux/socket.h>		/* for "struct sockaddr" et al	*/
 #include <linux/if.h>			/* for IFNAMSIZ and co... */
+#endif	/* __KERNEL__ */
 
 /***************************** VERSION *****************************/
 /*
@@ -80,7 +85,7 @@
  * (there is some stuff that will be added in the future...)
  * I just plan to increment with each new version.
  */
-#define WIRELESS_EXT	20
+#define WIRELESS_EXT	21
 
 /*
  * Changes :
@@ -208,6 +213,14 @@
  * V19 to V20
  * ----------
  *	- RtNetlink requests support (SET/GET)
+ *
+ * V20 to V21
+ * ----------
+ *	- Remove (struct net_device *)->get_wireless_stats()
+ *	- Change length in ESSID and NICK to strlen() instead of strlen()+1
+ *	- Add IW_RETRY_SHORT/IW_RETRY_LONG retry modifiers
+ *	- Power/Retry relative values no longer * 100000
+ *	- Add explicit flag to tell stats are in 802.11k RCPI : IW_QUAL_RCPI
  */
 
 /**************************** CONSTANTS ****************************/
@@ -448,6 +461,7 @@
 #define IW_QUAL_QUAL_INVALID	0x10	/* Driver doesn't provide value */
 #define IW_QUAL_LEVEL_INVALID	0x20
 #define IW_QUAL_NOISE_INVALID	0x40
+#define IW_QUAL_RCPI		0x80	/* Level + Noise are 802.11k RCPI */
 #define IW_QUAL_ALL_INVALID	0x70
 
 /* Frequency flags */
@@ -500,10 +514,12 @@
 #define IW_RETRY_TYPE		0xF000	/* Type of parameter */
 #define IW_RETRY_LIMIT		0x1000	/* Maximum number of retries*/
 #define IW_RETRY_LIFETIME	0x2000	/* Maximum duration of retries in us */
-#define IW_RETRY_MODIFIER	0x000F	/* Modify a parameter */
+#define IW_RETRY_MODIFIER	0x00FF	/* Modify a parameter */
 #define IW_RETRY_MIN		0x0001	/* Value is a minimum  */
 #define IW_RETRY_MAX		0x0002	/* Value is a maximum */
 #define IW_RETRY_RELATIVE	0x0004	/* Value is not in seconds/ms/us */
+#define IW_RETRY_SHORT		0x0010	/* Value is for short packets  */
+#define IW_RETRY_LONG		0x0020	/* Value is for long packets */
 
 /* Scanning request flags */
 #define IW_SCAN_DEFAULT		0x0000	/* Default scan of the driver */
@@ -1017,7 +1033,7 @@ struct	iw_range
 	/* Note : this frequency list doesn't need to fit channel numbers,
 	 * because each entry contain its channel index */
 
-	__u32		enc_capa; /* IW_ENC_CAPA_* bit field */
+	__u32		enc_capa;	/* IW_ENC_CAPA_* bit field */
 };
 
 /*
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 13472762b18b..f47f319bb7dc 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -344,8 +344,6 @@ static ssize_t wireless_show(struct class_device *cd, char *buf,
 		if(dev->wireless_handlers &&
 		   dev->wireless_handlers->get_wireless_stats)
 			iw = dev->wireless_handlers->get_wireless_stats(dev);
-		else if (dev->get_wireless_stats)
-			iw = dev->get_wireless_stats(dev);
 		if (iw != NULL)
 			ret = (*format)(iw, buf);
 	}
@@ -465,8 +463,7 @@ int netdev_register_sysfs(struct net_device *net)
 		*groups++ = &netstat_group;
 
 #ifdef WIRELESS_EXT
-	if (net->get_wireless_stats
-	    || (net->wireless_handlers && net->wireless_handlers->get_wireless_stats))
+	if (net->wireless_handlers && net->wireless_handlers->get_wireless_stats)
 		*groups++ = &wireless_group;
 #endif
 
diff --git a/net/core/wireless.c b/net/core/wireless.c
index 3168fca312f7..ffff0da46c6e 100644
--- a/net/core/wireless.c
+++ b/net/core/wireless.c
@@ -68,6 +68,14 @@
  *
  * v8 - 17.02.06 - Jean II
  *	o RtNetlink requests support (SET/GET)
+ *
+ * v8b - 03.08.06 - Herbert Xu
+ *	o Fix Wireless Event locking issues.
+ *
+ * v9 - 14.3.06 - Jean II
+ *	o Change length in ESSID and NICK to strlen() instead of strlen()+1
+ *	o Make standard_ioctl_num and standard_event_num unsigned
+ *	o Remove (struct net_device *)->get_wireless_stats()
  */
 
 /***************************** INCLUDES *****************************/
@@ -234,24 +242,24 @@ static const struct iw_ioctl_description standard_ioctl[] = {
 	[SIOCSIWESSID	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 		.flags		= IW_DESCR_FLAG_EVENT,
 	},
 	[SIOCGIWESSID	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 		.flags		= IW_DESCR_FLAG_DUMP,
 	},
 	[SIOCSIWNICKN	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 	},
 	[SIOCGIWNICKN	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 	},
 	[SIOCSIWRATE	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_PARAM,
@@ -338,8 +346,8 @@ static const struct iw_ioctl_description standard_ioctl[] = {
 		.max_tokens	= sizeof(struct iw_pmksa),
 	},
 };
-static const int standard_ioctl_num = (sizeof(standard_ioctl) /
-				       sizeof(struct iw_ioctl_description));
+static const unsigned standard_ioctl_num = (sizeof(standard_ioctl) /
+					    sizeof(struct iw_ioctl_description));
 
 /*
  * Meta-data about all the additional standard Wireless Extension events
@@ -389,8 +397,8 @@ static const struct iw_ioctl_description standard_event[] = {
 		.max_tokens	= sizeof(struct iw_pmkid_cand),
 	},
 };
-static const int standard_event_num = (sizeof(standard_event) /
-				       sizeof(struct iw_ioctl_description));
+static const unsigned standard_event_num = (sizeof(standard_event) /
+					    sizeof(struct iw_ioctl_description));
 
 /* Size (in bytes) of the various private data types */
 static const char iw_priv_type_size[] = {
@@ -465,17 +473,6 @@ static inline struct iw_statistics *get_wireless_stats(struct net_device *dev)
 	   (dev->wireless_handlers->get_wireless_stats != NULL))
 		return dev->wireless_handlers->get_wireless_stats(dev);
 
-	/* Old location, field to be removed in next WE */
-	if(dev->get_wireless_stats) {
-		static int printed_message;
-
-		if (!printed_message++)
-			printk(KERN_DEBUG "%s (WE) : Driver using old /proc/net/wireless support, please fix driver !\n",
-				dev->name);
-
-		return dev->get_wireless_stats(dev);
-	}
-
 	/* Not found */
 	return (struct iw_statistics *) NULL;
 }
@@ -1843,8 +1840,33 @@ int wireless_rtnetlink_set(struct net_device *	dev,
  */
 
 #ifdef WE_EVENT_RTNETLINK
+/* ---------------------------------------------------------------- */
+/*
+ * Locking...
+ * ----------
+ *
+ * Thanks to Herbert Xu <herbert@gondor.apana.org.au> for fixing
+ * the locking issue in here and implementing this code !
+ *
+ * The issue : wireless_send_event() is often called in interrupt context,
+ * while the Netlink layer can never be called in interrupt context.
+ * The fully formed RtNetlink events are queued, and then a tasklet is run
+ * to feed those to Netlink.
+ * The skb_queue is interrupt safe, and its lock is not held while calling
+ * Netlink, so there is no possibility of dealock.
+ * Jean II
+ */
+
 static struct sk_buff_head wireless_nlevent_queue;
 
+static int __init wireless_nlevent_init(void)
+{
+	skb_queue_head_init(&wireless_nlevent_queue);
+	return 0;
+}
+
+subsys_initcall(wireless_nlevent_init);
+
 static void wireless_nlevent_process(unsigned long data)
 {
 	struct sk_buff *skb;
@@ -1921,13 +1943,6 @@ static inline void rtmsg_iwinfo(struct net_device *	dev,
 	tasklet_schedule(&wireless_nlevent_tasklet);
 }
 
-static int __init wireless_nlevent_init(void)
-{
-	skb_queue_head_init(&wireless_nlevent_queue);
-	return 0;
-}
-
-subsys_initcall(wireless_nlevent_init);
 #endif	/* WE_EVENT_RTNETLINK */
 
 /* ---------------------------------------------------------------- */
-- 
cgit v1.2.3


From 14a72f53fb1bb5d5c2bdd8cf172219519664729a Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:52:01 -0700
Subject: [NetLabel]: correct improper handling of non-NetLabel peer contexts

Fix a problem where NetLabel would always set the value of
sk_security_struct->peer_sid in selinux_netlbl_sock_graft() to the context of
the socket, causing problems when users would query the context of the
connection.  This patch fixes this so that the value in
sk_security_struct->peer_sid is only set when the connection is NetLabel based,
otherwise the value is untouched.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/cipso_ipv4.h       |  7 ++++++
 include/net/netlabel.h         |  8 +++++++
 net/ipv4/cipso_ipv4.c          | 48 +++++++++++++++++++++++++++++-------------
 net/netlabel/netlabel_kapi.c   | 23 ++++++++++++++++++++
 security/selinux/ss/services.c | 12 ++++++++++-
 5 files changed, 82 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 59406e0dc5b2..6718452a5cd0 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -205,6 +205,7 @@ void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway);
 int cipso_v4_socket_setattr(const struct socket *sock,
 			    const struct cipso_v4_doi *doi_def,
 			    const struct netlbl_lsm_secattr *secattr);
+int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr);
 int cipso_v4_socket_getattr(const struct socket *sock,
 			    struct netlbl_lsm_secattr *secattr);
 int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
@@ -225,6 +226,12 @@ static inline int cipso_v4_socket_setattr(const struct socket *sock,
 	return -ENOSYS;
 }
 
+static inline int cipso_v4_sock_getattr(struct sock *sk,
+					struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
 static inline int cipso_v4_socket_getattr(const struct socket *sock,
 					  struct netlbl_lsm_secattr *secattr)
 {
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index dd5780b36919..bf7b564e3540 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -238,6 +238,8 @@ static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr,
 #ifdef CONFIG_NETLABEL
 int netlbl_socket_setattr(const struct socket *sock,
 			  const struct netlbl_lsm_secattr *secattr);
+int netlbl_sock_getattr(struct sock *sk,
+			struct netlbl_lsm_secattr *secattr);
 int netlbl_socket_getattr(const struct socket *sock,
 			  struct netlbl_lsm_secattr *secattr);
 int netlbl_skbuff_getattr(const struct sk_buff *skb,
@@ -250,6 +252,12 @@ static inline int netlbl_socket_setattr(const struct socket *sock,
 	return -ENOSYS;
 }
 
+static inline int netlbl_sock_getattr(struct sock *sk,
+				      struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
 static inline int netlbl_socket_getattr(const struct socket *sock,
 					struct netlbl_lsm_secattr *secattr)
 {
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 80a2a0911b49..a3bae2ca8acc 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -1486,43 +1486,40 @@ socket_setattr_failure:
 }
 
 /**
- * cipso_v4_socket_getattr - Get the security attributes from a socket
- * @sock: the socket
+ * cipso_v4_sock_getattr - Get the security attributes from a sock
+ * @sk: the sock
  * @secattr: the security attributes
  *
  * Description:
- * Query @sock to see if there is a CIPSO option attached to the socket and if
- * there is return the CIPSO security attributes in @secattr.  Returns zero on
- * success and negative values on failure.
+ * Query @sk to see if there is a CIPSO option attached to the sock and if
+ * there is return the CIPSO security attributes in @secattr.  This function
+ * requires that @sk be locked, or privately held, but it does not do any
+ * locking itself.  Returns zero on success and negative values on failure.
  *
  */
-int cipso_v4_socket_getattr(const struct socket *sock,
-			    struct netlbl_lsm_secattr *secattr)
+int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
 {
 	int ret_val = -ENOMSG;
-	struct sock *sk;
 	struct inet_sock *sk_inet;
 	unsigned char *cipso_ptr;
 	u32 doi;
 	struct cipso_v4_doi *doi_def;
 
-	sk = sock->sk;
-	lock_sock(sk);
 	sk_inet = inet_sk(sk);
 	if (sk_inet->opt == NULL || sk_inet->opt->cipso == 0)
-		goto socket_getattr_return;
+		return -ENOMSG;
 	cipso_ptr = sk_inet->opt->__data + sk_inet->opt->cipso -
 		sizeof(struct iphdr);
 	ret_val = cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr);
 	if (ret_val == 0)
-		goto socket_getattr_return;
+		return ret_val;
 
 	doi = ntohl(*(u32 *)&cipso_ptr[2]);
 	rcu_read_lock();
 	doi_def = cipso_v4_doi_getdef(doi);
 	if (doi_def == NULL) {
 		rcu_read_unlock();
-		goto socket_getattr_return;
+		return -ENOMSG;
 	}
 	switch (cipso_ptr[6]) {
 	case CIPSO_V4_TAG_RBITMAP:
@@ -1533,8 +1530,29 @@ int cipso_v4_socket_getattr(const struct socket *sock,
 	}
 	rcu_read_unlock();
 
-socket_getattr_return:
-	release_sock(sk);
+	return ret_val;
+}
+
+/**
+ * cipso_v4_socket_getattr - Get the security attributes from a socket
+ * @sock: the socket
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Query @sock to see if there is a CIPSO option attached to the socket and if
+ * there is return the CIPSO security attributes in @secattr.  Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_socket_getattr(const struct socket *sock,
+			    struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+
+	lock_sock(sock->sk);
+	ret_val = cipso_v4_sock_getattr(sock->sk, secattr);
+	release_sock(sock->sk);
+
 	return ret_val;
 }
 
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index 0fd8aaafe23f..54fb7de3c2b1 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -84,6 +84,29 @@ socket_setattr_return:
 	return ret_val;
 }
 
+/**
+ * netlbl_sock_getattr - Determine the security attributes of a sock
+ * @sk: the sock
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Examines the given sock to see any NetLabel style labeling has been
+ * applied to the sock, if so it parses the socket label and returns the
+ * security attributes in @secattr.  Returns zero on success, negative values
+ * on failure.
+ *
+ */
+int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+
+	ret_val = cipso_v4_sock_getattr(sk, secattr);
+	if (ret_val == 0)
+		return 0;
+
+	return netlbl_unlabel_getattr(secattr);
+}
+
 /**
  * netlbl_socket_getattr - Determine the security attributes of a socket
  * @sock: the socket
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 7eb69a602d8f..d67f7e658529 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2502,14 +2502,24 @@ void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock)
 {
 	struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
 	struct sk_security_struct *sksec = sk->sk_security;
+	struct netlbl_lsm_secattr secattr;
+	u32 nlbl_peer_sid;
 
 	sksec->sclass = isec->sclass;
 
 	if (sk->sk_family != PF_INET)
 		return;
 
+	netlbl_secattr_init(&secattr);
+	if (netlbl_sock_getattr(sk, &secattr) == 0 &&
+	    selinux_netlbl_secattr_to_sid(NULL,
+					  &secattr,
+					  sksec->sid,
+					  &nlbl_peer_sid) == 0)
+		sksec->peer_sid = nlbl_peer_sid;
+	netlbl_secattr_destroy(&secattr, 0);
+
 	sksec->nlbl_state = NLBL_REQUIRE;
-	sksec->peer_sid = sksec->sid;
 
 	/* Try to set the NetLabel on the socket to save time later, if we fail
 	 * here we will pick up the pieces in later calls to
-- 
cgit v1.2.3


From 22acb19a91d2b551ea37647747972e5286284b22 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:53:37 -0700
Subject: [NETLINK]: add nla_for_each_nested() to the interface list

At the top of include/net/netlink.h is a list of Netlink interfaces, however,
the nla_for_each_nested() macro was not listed.  This patch adds this interface
to the list at the top of the header file.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 11dc2e7f679a..2897a73f81eb 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -151,6 +151,7 @@
  *   nla_parse()			parse and validate stream of attrs
  *   nla_parse_nested()			parse nested attribuets
  *   nla_for_each_attr()		loop over all attributes
+ *   nla_for_each_nested()		loop over the nested attributes
  *=========================================================================
  */
 
-- 
cgit v1.2.3


From 4fe5d5c07ab615a52fd1b0ceba5aeed7c612821a Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:54:03 -0700
Subject: [Netlink]: add nla_validate_nested()

Add a new function, nla_validate_nested(), to validate nested Netlink
attributes.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 2897a73f81eb..4ab68a7a636a 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -146,6 +146,7 @@
  *   nla_ok(nla, remaining)		does nla fit into remaining bytes?
  *   nla_next(nla, remaining)		get next netlink attribute
  *   nla_validate()			validate a stream of attributes
+ *   nla_validate_nested()		validate a stream of nested attributes
  *   nla_find()				find attribute in stream of attributes
  *   nla_find_nested()			find attribute in nested attributes
  *   nla_parse()			parse and validate stream of attrs
@@ -950,6 +951,24 @@ static inline int nla_nest_cancel(struct sk_buff *skb, struct nlattr *start)
 	return nlmsg_trim(skb, start);
 }
 
+/**
+ * nla_validate_nested - Validate a stream of nested attributes
+ * @start: container attribute
+ * @maxtype: maximum attribute type to be expected
+ * @policy: validation policy
+ *
+ * Validates all attributes in the nested attribute stream against the
+ * specified policy. Attributes with a type exceeding maxtype will be
+ * ignored. See documenation of struct nla_policy for more details.
+ *
+ * Returns 0 on success or a negative error code.
+ */
+static inline int nla_validate_nested(struct nlattr *start, int maxtype,
+				      struct nla_policy *policy)
+{
+	return nla_validate(nla_data(start), nla_len(start), maxtype, policy);
+}
+
 /**
  * nla_for_each_attr - iterate over a stream of attributes
  * @pos: loop counter, set to current attribute
-- 
cgit v1.2.3


From fcd48280643e92ec6cb29a04e9079dd7b6b5bfef Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:56:09 -0700
Subject: [NetLabel]: rework the Netlink attribute handling (part 1)

At the suggestion of Thomas Graf, rewrite NetLabel's use of Netlink attributes
to better follow the common Netlink attribute usage.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/cipso_ipv4.h           |  16 ++-
 include/net/netlabel.h             |  49 +--------
 net/ipv4/cipso_ipv4.c              | 203 +++++--------------------------------
 net/netlabel/netlabel_domainhash.c | 183 +++++++--------------------------
 net/netlabel/netlabel_domainhash.h |   6 +-
 net/netlabel/netlabel_user.c       |  82 ---------------
 net/netlabel/netlabel_user.h       | 141 +-------------------------
 7 files changed, 76 insertions(+), 604 deletions(-)

(limited to 'include')

diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 6718452a5cd0..2d72496c2029 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -130,8 +130,9 @@ extern int cipso_v4_rbm_strictvalid;
 int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
 int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head));
 struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi);
-struct sk_buff *cipso_v4_doi_dump_all(size_t headroom);
-struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom);
+int cipso_v4_doi_walk(u32 *skip_cnt,
+		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
+	             void *cb_arg);
 int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain);
 int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
 			       const char *domain);
@@ -152,14 +153,11 @@ static inline struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
 	return NULL;
 }
 
-static inline struct sk_buff *cipso_v4_doi_dump_all(size_t headroom)
+static inline int cipso_v4_doi_walk(u32 *skip_cnt,
+		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
+		     void *cb_arg)
 {
-	return NULL;
-}
-
-static inline struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom)
-{
-	return NULL;
+	return 0;
 }
 
 static inline int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def,
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index bf7b564e3540..6692430063fd 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -57,9 +57,8 @@
  * The payload is dependent on the subsystem specified in the
  * 'nlmsghdr->nlmsg_type' and should be defined below, supporting functions
  * should be defined in the corresponding net/netlabel/netlabel_<subsys>.h|c
- * file.  All of the fields in the NetLabel payload are NETLINK attributes, the
- * length of each field is the length of the NETLINK attribute payload, see
- * include/net/netlink.h for more information on NETLINK attributes.
+ * file.  All of the fields in the NetLabel payload are NETLINK attributes, see
+ * the include/net/netlink.h file for more information on NETLINK attributes.
  *
  */
 
@@ -82,50 +81,6 @@
 #define NETLBL_NLTYPE_UNLABELED         5
 #define NETLBL_NLTYPE_UNLABELED_NAME    "NLBL_UNLBL"
 
-/* NetLabel return codes */
-#define NETLBL_E_OK                     0
-
-/*
- * Helper functions
- */
-
-#define NETLBL_LEN_U8                   nla_total_size(sizeof(u8))
-#define NETLBL_LEN_U16                  nla_total_size(sizeof(u16))
-#define NETLBL_LEN_U32                  nla_total_size(sizeof(u32))
-
-/**
- * netlbl_netlink_alloc_skb - Allocate a NETLINK message buffer
- * @head: the amount of headroom in bytes
- * @body: the desired size (minus headroom) in bytes
- * @gfp_flags: the alloc flags to pass to alloc_skb()
- *
- * Description:
- * Allocate a NETLINK message buffer based on the sizes given in @head and
- * @body.  If @head is greater than zero skb_reserve() is called to reserve
- * @head bytes at the start of the buffer.  Returns a valid sk_buff pointer on
- * success, NULL on failure.
- *
- */
-static inline struct sk_buff *netlbl_netlink_alloc_skb(size_t head,
-						       size_t body,
-						       gfp_t gfp_flags)
-{
-	struct sk_buff *skb;
-
-	skb = alloc_skb(NLMSG_ALIGN(head + body), gfp_flags);
-	if (skb == NULL)
-		return NULL;
-	if (head > 0) {
-		skb_reserve(skb, head);
-		if (skb_tailroom(skb) < body) {
-			kfree_skb(skb);
-			return NULL;
-		}
-	}
-
-	return skb;
-}
-
 /*
  * NetLabel - Kernel API for accessing the network packet label mappings.
  *
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 87e71563335d..e6ce0b3ba62a 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -530,197 +530,42 @@ struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
 }
 
 /**
- * cipso_v4_doi_dump_all - Dump all the CIPSO DOI definitions into a sk_buff
- * @headroom: the amount of headroom to allocate for the sk_buff
+ * cipso_v4_doi_walk - Iterate through the DOI definitions
+ * @skip_cnt: skip past this number of DOI definitions, updated
+ * @callback: callback for each DOI definition
+ * @cb_arg: argument for the callback function
  *
  * Description:
- * Dump a list of all the configured DOI values into a sk_buff.  The returned
- * sk_buff has room at the front of the sk_buff for @headroom bytes.  See
- * net/netlabel/netlabel_cipso_v4.h for the LISTALL message format.  This
- * function may fail if another process is changing the DOI list at the same
- * time.  Returns a pointer to a sk_buff on success, NULL on error.
+ * Iterate over the DOI definition list, skipping the first @skip_cnt entries.
+ * For each entry call @callback, if @callback returns a negative value stop
+ * 'walking' through the list and return.  Updates the value in @skip_cnt upon
+ * return.  Returns zero on success, negative values on failure.
  *
  */
-struct sk_buff *cipso_v4_doi_dump_all(size_t headroom)
+int cipso_v4_doi_walk(u32 *skip_cnt,
+		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
+		     void *cb_arg)
 {
-	struct sk_buff *skb = NULL;
-	struct cipso_v4_doi *iter;
+	int ret_val = -ENOENT;
 	u32 doi_cnt = 0;
-	ssize_t buf_len;
-
-	buf_len = NETLBL_LEN_U32;
-	rcu_read_lock();
-	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
-		if (iter->valid) {
-			doi_cnt += 1;
-			buf_len += 2 * NETLBL_LEN_U32;
-		}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto doi_dump_all_failure;
-
-	if (nla_put_u32(skb, NLA_U32, doi_cnt) != 0)
-		goto doi_dump_all_failure;
-	buf_len -= NETLBL_LEN_U32;
-	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
-		if (iter->valid) {
-			if (buf_len < 2 * NETLBL_LEN_U32)
-				goto doi_dump_all_failure;
-			if (nla_put_u32(skb, NLA_U32, iter->doi) != 0)
-				goto doi_dump_all_failure;
-			if (nla_put_u32(skb, NLA_U32, iter->type) != 0)
-				goto doi_dump_all_failure;
-			buf_len -= 2 * NETLBL_LEN_U32;
-		}
-	rcu_read_unlock();
-
-	return skb;
-
-doi_dump_all_failure:
-	rcu_read_unlock();
-	kfree(skb);
-	return NULL;
-}
-
-/**
- * cipso_v4_doi_dump - Dump a CIPSO DOI definition into a sk_buff
- * @doi: the DOI value
- * @headroom: the amount of headroom to allocate for the sk_buff
- *
- * Description:
- * Lookup the DOI definition matching @doi and dump it's contents into a
- * sk_buff.  The returned sk_buff has room at the front of the sk_buff for
- * @headroom bytes.  See net/netlabel/netlabel_cipso_v4.h for the LIST message
- * format.  This function may fail if another process is changing the DOI list
- * at the same time.  Returns a pointer to a sk_buff on success, NULL on error.
- *
- */
-struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom)
-{
-	struct sk_buff *skb = NULL;
-	struct cipso_v4_doi *iter;
-	u32 tag_cnt = 0;
-	u32 lvl_cnt = 0;
-	u32 cat_cnt = 0;
-	ssize_t buf_len;
-	ssize_t tmp;
+	struct cipso_v4_doi *iter_doi;
 
 	rcu_read_lock();
-	iter = cipso_v4_doi_getdef(doi);
-	if (iter == NULL)
-		goto doi_dump_failure;
-	buf_len = NETLBL_LEN_U32;
-	switch (iter->type) {
-	case CIPSO_V4_MAP_PASS:
-		buf_len += NETLBL_LEN_U32;
-		while(tag_cnt < CIPSO_V4_TAG_MAXCNT &&
-		      iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
-			tag_cnt += 1;
-			buf_len += NETLBL_LEN_U8;
-		}
-		break;
-	case CIPSO_V4_MAP_STD:
-		buf_len += 3 * NETLBL_LEN_U32;
-		while (tag_cnt < CIPSO_V4_TAG_MAXCNT &&
-		       iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
-			tag_cnt += 1;
-			buf_len += NETLBL_LEN_U8;
-		}
-		for (tmp = 0; tmp < iter->map.std->lvl.local_size; tmp++)
-			if (iter->map.std->lvl.local[tmp] !=
-			    CIPSO_V4_INV_LVL) {
-				lvl_cnt += 1;
-				buf_len += NETLBL_LEN_U32 + NETLBL_LEN_U8;
+	list_for_each_entry_rcu(iter_doi, &cipso_v4_doi_list, list)
+		if (iter_doi->valid) {
+			if (doi_cnt++ < *skip_cnt)
+				continue;
+			ret_val = callback(iter_doi, cb_arg);
+			if (ret_val < 0) {
+				doi_cnt--;
+				goto doi_walk_return;
 			}
-		for (tmp = 0; tmp < iter->map.std->cat.local_size; tmp++)
-			if (iter->map.std->cat.local[tmp] !=
-			    CIPSO_V4_INV_CAT) {
-				cat_cnt += 1;
-				buf_len += NETLBL_LEN_U32 + NETLBL_LEN_U16;
-			}
-		break;
-	}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto doi_dump_failure;
-
-	if (nla_put_u32(skb, NLA_U32, iter->type) != 0)
-		goto doi_dump_failure;
-	buf_len -= NETLBL_LEN_U32;
-	if (iter != cipso_v4_doi_getdef(doi))
-		goto doi_dump_failure;
-	switch (iter->type) {
-	case CIPSO_V4_MAP_PASS:
-		if (nla_put_u32(skb, NLA_U32, tag_cnt) != 0)
-			goto doi_dump_failure;
-		buf_len -= NETLBL_LEN_U32;
-		for (tmp = 0;
-		     tmp < CIPSO_V4_TAG_MAXCNT &&
-			     iter->tags[tmp] != CIPSO_V4_TAG_INVALID;
-		     tmp++) {
-			if (buf_len < NETLBL_LEN_U8)
-				goto doi_dump_failure;
-			if (nla_put_u8(skb, NLA_U8, iter->tags[tmp]) != 0)
-				goto doi_dump_failure;
-			buf_len -= NETLBL_LEN_U8;
 		}
-		break;
-	case CIPSO_V4_MAP_STD:
-		if (nla_put_u32(skb, NLA_U32, tag_cnt) != 0)
-			goto doi_dump_failure;
-		if (nla_put_u32(skb, NLA_U32, lvl_cnt) != 0)
-			goto doi_dump_failure;
-		if (nla_put_u32(skb, NLA_U32, cat_cnt) != 0)
-			goto doi_dump_failure;
-		buf_len -= 3 * NETLBL_LEN_U32;
-		for (tmp = 0;
-		     tmp < CIPSO_V4_TAG_MAXCNT &&
-			     iter->tags[tmp] != CIPSO_V4_TAG_INVALID;
-		     tmp++) {
-			if (buf_len < NETLBL_LEN_U8)
-				goto doi_dump_failure;
-			if (nla_put_u8(skb, NLA_U8, iter->tags[tmp]) != 0)
-				goto doi_dump_failure;
-			buf_len -= NETLBL_LEN_U8;
-		}
-		for (tmp = 0; tmp < iter->map.std->lvl.local_size; tmp++)
-			if (iter->map.std->lvl.local[tmp] !=
-			    CIPSO_V4_INV_LVL) {
-				if (buf_len < NETLBL_LEN_U32 + NETLBL_LEN_U8)
-					goto doi_dump_failure;
-				if (nla_put_u32(skb, NLA_U32, tmp) != 0)
-					goto doi_dump_failure;
-				if (nla_put_u8(skb,
-					   NLA_U8,
-					   iter->map.std->lvl.local[tmp]) != 0)
-					goto doi_dump_failure;
-				buf_len -= NETLBL_LEN_U32 + NETLBL_LEN_U8;
-			}
-		for (tmp = 0; tmp < iter->map.std->cat.local_size; tmp++)
-			if (iter->map.std->cat.local[tmp] !=
-			    CIPSO_V4_INV_CAT) {
-				if (buf_len < NETLBL_LEN_U32 + NETLBL_LEN_U16)
-					goto doi_dump_failure;
-				if (nla_put_u32(skb, NLA_U32, tmp) != 0)
-					goto doi_dump_failure;
-				if (nla_put_u16(skb,
-					   NLA_U16,
-					   iter->map.std->cat.local[tmp]) != 0)
-					goto doi_dump_failure;
-				buf_len -= NETLBL_LEN_U32 + NETLBL_LEN_U16;
-			}
-		break;
-	}
-	rcu_read_unlock();
-
-	return skb;
 
-doi_dump_failure:
+doi_walk_return:
 	rcu_read_unlock();
-	kfree(skb);
-	return NULL;
+	*skip_cnt = doi_cnt;
+	return ret_val;
 }
 
 /**
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index 0489a1378101..f56d7a8ac7b7 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -354,160 +354,51 @@ struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain)
 }
 
 /**
- * netlbl_domhsh_dump - Dump the domain hash table into a sk_buff
+ * netlbl_domhsh_walk - Iterate through the domain mapping hash table
+ * @skip_bkt: the number of buckets to skip at the start
+ * @skip_chain: the number of entries to skip in the first iterated bucket
+ * @callback: callback for each entry
+ * @cb_arg: argument for the callback function
  *
  * Description:
- * Dump the domain hash table into a buffer suitable for returning to an
- * application in response to a NetLabel management DOMAIN message.  This
- * function may fail if another process is growing the hash table at the same
- * time.  The returned sk_buff has room at the front of the sk_buff for
- * @headroom bytes.  See netlabel.h for the DOMAIN message format.  Returns a
- * pointer to a sk_buff on success, NULL on error.
+ * Interate over the domain mapping hash table, skipping the first @skip_bkt
+ * buckets and @skip_chain entries.  For each entry in the table call
+ * @callback, if @callback returns a negative value stop 'walking' through the
+ * table and return.  Updates the values in @skip_bkt and @skip_chain on
+ * return.  Returns zero on succcess, negative values on failure.
  *
  */
-struct sk_buff *netlbl_domhsh_dump(size_t headroom)
+int netlbl_domhsh_walk(u32 *skip_bkt,
+		     u32 *skip_chain,
+		     int (*callback) (struct netlbl_dom_map *entry, void *arg),
+		     void *cb_arg)
 {
-	struct sk_buff *skb = NULL;
-	ssize_t buf_len;
-	u32 bkt_iter;
-	u32 dom_cnt = 0;
-	struct netlbl_domhsh_tbl *hsh_tbl;
-	struct netlbl_dom_map *list_iter;
-	ssize_t tmp_len;
+	int ret_val = -ENOENT;
+	u32 iter_bkt;
+	struct netlbl_dom_map *iter_entry;
+	u32 chain_cnt = 0;
 
-	buf_len = NETLBL_LEN_U32;
 	rcu_read_lock();
-	hsh_tbl = rcu_dereference(netlbl_domhsh);
-	for (bkt_iter = 0; bkt_iter < hsh_tbl->size; bkt_iter++)
-		list_for_each_entry_rcu(list_iter,
-					&hsh_tbl->tbl[bkt_iter], list) {
-			buf_len += NETLBL_LEN_U32 +
-				nla_total_size(strlen(list_iter->domain) + 1);
-			switch (list_iter->type) {
-			case NETLBL_NLTYPE_UNLABELED:
-				break;
-			case NETLBL_NLTYPE_CIPSOV4:
-				buf_len += 2 * NETLBL_LEN_U32;
-				break;
-			}
-			dom_cnt++;
-		}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto dump_failure;
-
-	if (nla_put_u32(skb, NLA_U32, dom_cnt) != 0)
-		goto dump_failure;
-	buf_len -= NETLBL_LEN_U32;
-	hsh_tbl = rcu_dereference(netlbl_domhsh);
-	for (bkt_iter = 0; bkt_iter < hsh_tbl->size; bkt_iter++)
-		list_for_each_entry_rcu(list_iter,
-					&hsh_tbl->tbl[bkt_iter], list) {
-			tmp_len = nla_total_size(strlen(list_iter->domain) +
-						 1);
-			if (buf_len < NETLBL_LEN_U32 + tmp_len)
-				goto dump_failure;
-			if (nla_put_string(skb,
-					   NLA_STRING,
-					   list_iter->domain) != 0)
-				goto dump_failure;
-			if (nla_put_u32(skb, NLA_U32, list_iter->type) != 0)
-				goto dump_failure;
-			buf_len -= NETLBL_LEN_U32 + tmp_len;
-			switch (list_iter->type) {
-			case NETLBL_NLTYPE_UNLABELED:
-				break;
-			case NETLBL_NLTYPE_CIPSOV4:
-				if (buf_len < 2 * NETLBL_LEN_U32)
-					goto dump_failure;
-				if (nla_put_u32(skb,
-				       NLA_U32,
-				       list_iter->type_def.cipsov4->type) != 0)
-					goto dump_failure;
-				if (nla_put_u32(skb,
-				       NLA_U32,
-				       list_iter->type_def.cipsov4->doi) != 0)
-					goto dump_failure;
-				buf_len -= 2 * NETLBL_LEN_U32;
-				break;
+	for (iter_bkt = *skip_bkt;
+	     iter_bkt < rcu_dereference(netlbl_domhsh)->size;
+	     iter_bkt++, chain_cnt = 0) {
+		list_for_each_entry_rcu(iter_entry,
+					&netlbl_domhsh->tbl[iter_bkt],
+					list)
+			if (iter_entry->valid) {
+				if (chain_cnt++ < *skip_chain)
+					continue;
+				ret_val = callback(iter_entry, cb_arg);
+				if (ret_val < 0) {
+					chain_cnt--;
+					goto walk_return;
+				}
 			}
-		}
-	rcu_read_unlock();
-
-	return skb;
-
-dump_failure:
-	rcu_read_unlock();
-	kfree_skb(skb);
-	return NULL;
-}
-
-/**
- * netlbl_domhsh_dump_default - Dump the default domain mapping into a sk_buff
- *
- * Description:
- * Dump the default domain mapping into a buffer suitable for returning to an
- * application in response to a NetLabel management DEFDOMAIN message.  This
- * function may fail if another process is changing the default domain mapping
- * at the same time.  The returned sk_buff has room at the front of the
- * skb_buff for @headroom bytes.  See netlabel.h for the DEFDOMAIN message
- * format.  Returns a pointer to a sk_buff on success, NULL on error.
- *
- */
-struct sk_buff *netlbl_domhsh_dump_default(size_t headroom)
-{
-	struct sk_buff *skb;
-	ssize_t buf_len;
-	struct netlbl_dom_map *entry;
-
-	buf_len = NETLBL_LEN_U32;
-	rcu_read_lock();
-	entry = rcu_dereference(netlbl_domhsh_def);
-	if (entry != NULL)
-		switch (entry->type) {
-		case NETLBL_NLTYPE_UNLABELED:
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			buf_len += 2 * NETLBL_LEN_U32;
-			break;
-		}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto dump_default_failure;
-
-	if (entry != rcu_dereference(netlbl_domhsh_def))
-		goto dump_default_failure;
-	if (entry != NULL) {
-		if (nla_put_u32(skb, NLA_U32, entry->type) != 0)
-			goto dump_default_failure;
-		buf_len -= NETLBL_LEN_U32;
-		switch (entry->type) {
-		case NETLBL_NLTYPE_UNLABELED:
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			if (buf_len < 2 * NETLBL_LEN_U32)
-				goto dump_default_failure;
-			if (nla_put_u32(skb,
-					NLA_U32,
-					entry->type_def.cipsov4->type) != 0)
-				goto dump_default_failure;
-			if (nla_put_u32(skb,
-					NLA_U32,
-					entry->type_def.cipsov4->doi) != 0)
-				goto dump_default_failure;
-			buf_len -= 2 * NETLBL_LEN_U32;
-			break;
-		}
-	} else
-		nla_put_u32(skb, NLA_U32, NETLBL_NLTYPE_NONE);
-	rcu_read_unlock();
-
-	return skb;
+	}
 
-dump_default_failure:
+walk_return:
 	rcu_read_unlock();
-	kfree_skb(skb);
-	return NULL;
+	*skip_bkt = iter_bkt;
+	*skip_chain = chain_cnt;
+	return ret_val;
 }
diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h
index 99a2287de246..02af72a7877c 100644
--- a/net/netlabel/netlabel_domainhash.h
+++ b/net/netlabel/netlabel_domainhash.h
@@ -61,7 +61,9 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry);
 int netlbl_domhsh_add_default(struct netlbl_dom_map *entry);
 int netlbl_domhsh_remove_default(void);
 struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain);
-struct sk_buff *netlbl_domhsh_dump(size_t headroom);
-struct sk_buff *netlbl_domhsh_dump_default(size_t headroom);
+int netlbl_domhsh_walk(u32 *skip_bkt,
+		     u32 *skip_chain,
+		     int (*callback) (struct netlbl_dom_map *entry, void *arg),
+		     void *cb_arg);
 
 #endif
diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c
index 73cbe66e42ff..eeb7d768d2bb 100644
--- a/net/netlabel/netlabel_user.c
+++ b/net/netlabel/netlabel_user.c
@@ -74,85 +74,3 @@ int netlbl_netlink_init(void)
 
 	return 0;
 }
-
-/*
- * NetLabel Common Protocol Functions
- */
-
-/**
- * netlbl_netlink_send_ack - Send an ACK message
- * @info: the generic NETLINK information
- * @genl_family: the generic NETLINK family ID value
- * @ack_cmd: the generic NETLINK family ACK command value
- * @ret_code: return code to use
- *
- * Description:
- * This function sends an ACK message to the sender of the NETLINK message
- * specified by @info.
- *
- */
-void netlbl_netlink_send_ack(const struct genl_info *info,
-			     u32 genl_family,
-			     u8 ack_cmd,
-			     u32 ret_code)
-{
-	size_t data_size;
-	struct sk_buff *skb;
-
-	data_size = GENL_HDRLEN + 2 * NETLBL_LEN_U32;
-	skb = netlbl_netlink_alloc_skb(0, data_size, GFP_KERNEL);
-	if (skb == NULL)
-		return;
-
-	if (netlbl_netlink_hdr_put(skb,
-				   info->snd_pid,
-				   0,
-				   genl_family,
-				   ack_cmd) == NULL)
-		goto send_ack_failure;
-
-	if (nla_put_u32(skb, NLA_U32, info->snd_seq) != 0)
-		goto send_ack_failure;
-	if (nla_put_u32(skb, NLA_U32, ret_code) != 0)
-		goto send_ack_failure;
-
-	netlbl_netlink_snd(skb, info->snd_pid);
-	return;
-
-send_ack_failure:
-	kfree_skb(skb);
-}
-
-/*
- * NETLINK I/O Functions
- */
-
-/**
- * netlbl_netlink_snd - Send a NetLabel message
- * @skb: NetLabel message
- * @pid: destination PID
- *
- * Description:
- * Sends a unicast NetLabel message over the NETLINK socket.
- *
- */
-int netlbl_netlink_snd(struct sk_buff *skb, u32 pid)
-{
-	return genlmsg_unicast(skb, pid);
-}
-
-/**
- * netlbl_netlink_snd - Send a NetLabel message
- * @skb: NetLabel message
- * @pid: sending PID
- * @group: multicast group id
- *
- * Description:
- * Sends a multicast NetLabel message over the NETLINK socket to all members
- * of @group except @pid.
- *
- */
-int netlbl_netlink_snd_multicast(struct sk_buff *skb, u32 pid, u32 group)
-{
-	return genlmsg_multicast(skb, pid, group, GFP_KERNEL);
-}
diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h
index 385a6c7488c6..3f9386b917df 100644
--- a/net/netlabel/netlabel_user.h
+++ b/net/netlabel/netlabel_user.h
@@ -40,72 +40,6 @@
 
 /* NetLabel NETLINK helper functions */
 
-/**
- * netlbl_netlink_cap_check - Check the NETLINK msg capabilities
- * @skb: the NETLINK buffer
- * @req_cap: the required capability
- *
- * Description:
- * Check the NETLINK buffer's capabilities against the required capabilities.
- * Returns zero on success, negative values on failure.
- *
- */
-static inline int netlbl_netlink_cap_check(const struct sk_buff *skb,
-					   kernel_cap_t req_cap)
-{
-	if (cap_raised(NETLINK_CB(skb).eff_cap, req_cap))
-		return 0;
-	return -EPERM;
-}
-
-/**
- * netlbl_getinc_u8 - Read a u8 value from a nlattr stream and move on
- * @nla: the attribute
- * @rem_len: remaining length
- *
- * Description:
- * Return a u8 value pointed to by @nla and advance it to the next attribute.
- *
- */
-static inline u8 netlbl_getinc_u8(struct nlattr **nla, int *rem_len)
-{
-	u8 val = nla_get_u8(*nla);
-	*nla = nla_next(*nla, rem_len);
-	return val;
-}
-
-/**
- * netlbl_getinc_u16 - Read a u16 value from a nlattr stream and move on
- * @nla: the attribute
- * @rem_len: remaining length
- *
- * Description:
- * Return a u16 value pointed to by @nla and advance it to the next attribute.
- *
- */
-static inline u16 netlbl_getinc_u16(struct nlattr **nla, int *rem_len)
-{
-	u16 val = nla_get_u16(*nla);
-	*nla = nla_next(*nla, rem_len);
-	return val;
-}
-
-/**
- * netlbl_getinc_u32 - Read a u32 value from a nlattr stream and move on
- * @nla: the attribute
- * @rem_len: remaining length
- *
- * Description:
- * Return a u32 value pointed to by @nla and advance it to the next attribute.
- *
- */
-static inline u32 netlbl_getinc_u32(struct nlattr **nla, int *rem_len)
-{
-	u32 val = nla_get_u32(*nla);
-	*nla = nla_next(*nla, rem_len);
-	return val;
-}
-
 /**
  * netlbl_netlink_hdr_put - Write the NETLINK buffers into a sk_buff
  * @skb: the packet
@@ -124,6 +58,7 @@ static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb,
 					   u32 pid,
 					   u32 seq,
 					   int type,
+					   int flags,
 					   u8 cmd)
 {
 	return genlmsg_put(skb,
@@ -131,85 +66,13 @@ static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb,
 			   seq,
 			   type,
 			   0,
-			   0,
+			   flags,
 			   cmd,
 			   NETLBL_PROTO_VERSION);
 }
 
-/**
- * netlbl_netlink_hdr_push - Write the NETLINK buffers into a sk_buff
- * @skb: the packet
- * @pid: the PID of the receipient
- * @seq: the sequence number
- * @type: the generic NETLINK message family type
- * @cmd: command
- *
- * Description:
- * Write both a NETLINK nlmsghdr structure and a Generic NETLINK genlmsghdr
- * struct to the packet.
- *
- */
-static inline void netlbl_netlink_hdr_push(struct sk_buff *skb,
-					   u32 pid,
-					   u32 seq,
-					   int type,
-					   u8 cmd)
-
-{
-	struct nlmsghdr *nlh;
-	struct genlmsghdr *hdr;
-
-	nlh = (struct nlmsghdr *)skb_push(skb, NLMSG_SPACE(GENL_HDRLEN));
-	nlh->nlmsg_type = type;
-	nlh->nlmsg_len = skb->len;
-	nlh->nlmsg_flags = 0;
-	nlh->nlmsg_pid = pid;
-	nlh->nlmsg_seq = seq;
-
-	hdr = nlmsg_data(nlh);
-	hdr->cmd = cmd;
-	hdr->version = NETLBL_PROTO_VERSION;
-	hdr->reserved = 0;
-}
-
-/**
- * netlbl_netlink_payload_len - Return the length of the payload
- * @skb: the NETLINK buffer
- *
- * Description:
- * This function returns the length of the NetLabel payload.
- *
- */
-static inline u32 netlbl_netlink_payload_len(const struct sk_buff *skb)
-{
-	return nlmsg_len((struct nlmsghdr *)skb->data) - GENL_HDRLEN;
-}
-
-/**
- * netlbl_netlink_payload_data - Returns a pointer to the start of the payload
- * @skb: the NETLINK buffer
- *
- * Description:
- * This function returns a pointer to the start of the NetLabel payload.
- *
- */
-static inline void *netlbl_netlink_payload_data(const struct sk_buff *skb)
-{
-  return (unsigned char *)nlmsg_data((struct nlmsghdr *)skb->data) +
-	  GENL_HDRLEN;
-}
-
-/* NetLabel common protocol functions */
-
-void netlbl_netlink_send_ack(const struct genl_info *info,
-			     u32 genl_family,
-			     u8 ack_cmd,
-			     u32 ret_code);
-
 /* NetLabel NETLINK I/O functions */
 
 int netlbl_netlink_init(void);
-int netlbl_netlink_snd(struct sk_buff *skb, u32 pid);
-int netlbl_netlink_snd_multicast(struct sk_buff *skb, u32 pid, u32 group);
 
 #endif
-- 
cgit v1.2.3


From 0b680e753724d31a9c45f059d1aad29df54584a1 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:54:10 -0700
Subject: [PATCH] bonding: Add priv_flag to avoid event mishandling

Add priv_flag to specifically identify bonding-involved devices.  Needed
because IFF_MASTER is an unreliable identifier (vlan interfaces above bonding
will inherit IFF_MASTER).  Misidentification of devices would cause
notifier events for other devices to be erroneously processed by bonding,
causing various havoc.

Bug discovered by Martin Papik <martin.papik@ipsec.info>; this patch is
modified from his original.

Signed-off-by: Martin Papik <martin.papik@ipsec.info>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_main.c | 7 ++++++-
 include/linux/if.h              | 1 +
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d2f460b0dbab..9e5a533a1622 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1371,6 +1371,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	}
 
 	new_slave->dev = slave_dev;
+	slave_dev->priv_flags |= IFF_BONDING;
 
 	if ((bond->params.mode == BOND_MODE_TLB) ||
 	    (bond->params.mode == BOND_MODE_ALB)) {
@@ -1784,7 +1785,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 	dev_set_mac_address(slave_dev, &addr);
 
 	slave_dev->priv_flags &= ~(IFF_MASTER_8023AD | IFF_MASTER_ALB |
-				   IFF_SLAVE_INACTIVE);
+				   IFF_SLAVE_INACTIVE | IFF_BONDING);
 
 	kfree(slave);
 
@@ -3216,6 +3217,9 @@ static int bond_netdev_event(struct notifier_block *this, unsigned long event, v
 		(event_dev ? event_dev->name : "None"),
 		event);
 
+	if (!(event_dev->priv_flags & IFF_BONDING))
+		return NOTIFY_DONE;
+
 	if (event_dev->flags & IFF_MASTER) {
 		dprintk("IFF_MASTER\n");
 		return bond_master_netdev_event(event, event_dev);
@@ -4185,6 +4189,7 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params)
 	/* Initialize the device options */
 	bond_dev->tx_queue_len = 0;
 	bond_dev->flags |= IFF_MASTER|IFF_MULTICAST;
+	bond_dev->priv_flags |= IFF_BONDING;
 
 	/* At first, we block adding VLANs. That's the only way to
 	 * prevent problems that occur when adding VLANs over an
diff --git a/include/linux/if.h b/include/linux/if.h
index cd080d765324..a023ec1274fe 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -59,6 +59,7 @@
 #define IFF_SLAVE_INACTIVE	0x4	/* bonding slave not the curr. active */
 #define IFF_MASTER_8023AD	0x8	/* bonding master, 802.3ad. 	*/
 #define IFF_MASTER_ALB	0x10		/* bonding master, balance-alb.	*/
+#define IFF_BONDING	0x20		/* bonding master or slave	*/
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
-- 
cgit v1.2.3


From f5b2b966f032f22d3a289045a5afd4afa09f09c6 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:54:53 -0700
Subject: [PATCH] bonding: Validate probe replies in ARP monitor

	Add logic to check ARP request / reply packets used for ARP
monitor link integrity checking.

	The current method simply examines the slave device to see if it
has sent and received traffic; this can be fooled by extraneous traffic.
For example, if multiple hosts running bonding are behind a common
switch, the probe traffic from the multiple instances of bonding will
update the tx/rx times on each other's slave devices.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 Documentation/networking/bonding.txt |  59 ++++++++++++
 drivers/net/bonding/bond_main.c      | 182 +++++++++++++++++++++++++++++++++--
 drivers/net/bonding/bond_sysfs.c     |  54 +++++++++++
 drivers/net/bonding/bonding.h        |  32 +++++-
 include/linux/if.h                   |   1 +
 include/linux/netdevice.h            |   7 +-
 6 files changed, 325 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index afac780445cd..dc942eaf490f 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -192,6 +192,17 @@ or, for backwards compatibility, the option value.  E.g.,
 arp_interval
 
 	Specifies the ARP link monitoring frequency in milliseconds.
+
+	The ARP monitor works by periodically checking the slave
+	devices to determine whether they have sent or received
+	traffic recently (the precise criteria depends upon the
+	bonding mode, and the state of the slave).  Regular traffic is
+	generated via ARP probes issued for the addresses specified by
+	the arp_ip_target option.
+
+	This behavior can be modified by the arp_validate option,
+	below.
+
 	If ARP monitoring is used in an etherchannel compatible mode
 	(modes 0 and 2), the switch should be configured in a mode
 	that evenly distributes packets across all links. If the
@@ -213,6 +224,54 @@ arp_ip_target
 	maximum number of targets that can be specified is 16.  The
 	default value is no IP addresses.
 
+arp_validate
+
+	Specifies whether or not ARP probes and replies should be
+	validated in the active-backup mode.  This causes the ARP
+	monitor to examine the incoming ARP requests and replies, and
+	only consider a slave to be up if it is receiving the
+	appropriate ARP traffic.
+
+	Possible values are:
+
+	none or 0
+
+		No validation is performed.  This is the default.
+
+	active or 1
+
+		Validation is performed only for the active slave.
+
+	backup or 2
+
+		Validation is performed only for backup slaves.
+
+	all or 3
+
+		Validation is performed for all slaves.
+
+	For the active slave, the validation checks ARP replies to
+	confirm that they were generated by an arp_ip_target.  Since
+	backup slaves do not typically receive these replies, the
+	validation performed for backup slaves is on the ARP request
+	sent out via the active slave.  It is possible that some
+	switch or network configurations may result in situations
+	wherein the backup slaves do not receive the ARP requests; in
+	such a situation, validation of backup slaves must be
+	disabled.
+
+	This option is useful in network configurations in which
+	multiple bonding hosts are concurrently issuing ARPs to one or
+	more targets beyond a common switch.  Should the link between
+	the switch and target fail (but not the switch itself), the
+	probe traffic generated by the multiple bonding instances will
+	fool the standard ARP monitor into considering the links as
+	still up.  Use of the arp_validate option can resolve this, as
+	the ARP monitor will only consider ARP requests and replies
+	associated with its own instance of bonding.
+
+	This option was added in bonding version 3.1.0.
+
 downdelay
 
 	Specifies the time, in milliseconds, to wait before disabling
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index bafe62f7c9b7..fd521b05db83 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -96,6 +96,7 @@ static char *lacp_rate	= NULL;
 static char *xmit_hash_policy = NULL;
 static int arp_interval = BOND_LINK_ARP_INTERV;
 static char *arp_ip_target[BOND_MAX_ARP_TARGETS] = { NULL, };
+static char *arp_validate = NULL;
 struct bond_params bonding_defaults;
 
 module_param(max_bonds, int, 0);
@@ -127,6 +128,8 @@ module_param(arp_interval, int, 0);
 MODULE_PARM_DESC(arp_interval, "arp interval in milliseconds");
 module_param_array(arp_ip_target, charp, NULL, 0);
 MODULE_PARM_DESC(arp_ip_target, "arp targets in n.n.n.n form");
+module_param(arp_validate, charp, 0);
+MODULE_PARM_DESC(arp_validate, "validate src/dst of ARP probes: none (default), active, backup or all");
 
 /*----------------------------- Global variables ----------------------------*/
 
@@ -170,6 +173,14 @@ struct bond_parm_tbl xmit_hashtype_tbl[] = {
 {	NULL,			-1},
 };
 
+struct bond_parm_tbl arp_validate_tbl[] = {
+{	"none",			BOND_ARP_VALIDATE_NONE},
+{	"active",		BOND_ARP_VALIDATE_ACTIVE},
+{	"backup",		BOND_ARP_VALIDATE_BACKUP},
+{	"all",			BOND_ARP_VALIDATE_ALL},
+{	NULL,			-1},
+};
+
 /*-------------------------- Forward declarations ---------------------------*/
 
 static void bond_send_gratuitous_arp(struct bonding *bond);
@@ -1424,6 +1435,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 
 	bond_compute_features(bond);
 
+	new_slave->last_arp_rx = jiffies;
+
 	if (bond->params.miimon && !bond->params.use_carrier) {
 		link_reporting = bond_check_dev_link(bond, slave_dev, 1);
 
@@ -1785,7 +1798,8 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 	dev_set_mac_address(slave_dev, &addr);
 
 	slave_dev->priv_flags &= ~(IFF_MASTER_8023AD | IFF_MASTER_ALB |
-				   IFF_SLAVE_INACTIVE | IFF_BONDING);
+				   IFF_SLAVE_INACTIVE | IFF_BONDING |
+				   IFF_SLAVE_NEEDARP);
 
 	kfree(slave);
 
@@ -2298,6 +2312,25 @@ static int bond_has_ip(struct bonding *bond)
 	return 0;
 }
 
+static int bond_has_this_ip(struct bonding *bond, u32 ip)
+{
+	struct vlan_entry *vlan, *vlan_next;
+
+	if (ip == bond->master_ip)
+		return 1;
+
+	if (list_empty(&bond->vlan_list))
+		return 0;
+
+	list_for_each_entry_safe(vlan, vlan_next, &bond->vlan_list,
+				 vlan_list) {
+		if (ip == vlan->vlan_ip)
+			return 1;
+	}
+
+	return 0;
+}
+
 /*
  * We go to the (large) trouble of VLAN tagging ARP frames because
  * switches in VLAN mode (especially if ports are configured as
@@ -2436,6 +2469,93 @@ static void bond_send_gratuitous_arp(struct bonding *bond)
 	}
 }
 
+static void bond_validate_arp(struct bonding *bond, struct slave *slave, u32 sip, u32 tip)
+{
+	int i;
+	u32 *targets = bond->params.arp_targets;
+
+	targets = bond->params.arp_targets;
+	for (i = 0; (i < BOND_MAX_ARP_TARGETS) && targets[i]; i++) {
+		dprintk("bva: sip %u.%u.%u.%u tip %u.%u.%u.%u t[%d] "
+			"%u.%u.%u.%u bhti(tip) %d\n",
+		       NIPQUAD(sip), NIPQUAD(tip), i, NIPQUAD(targets[i]),
+		       bond_has_this_ip(bond, tip));
+		if (sip == targets[i]) {
+			if (bond_has_this_ip(bond, tip))
+				slave->last_arp_rx = jiffies;
+			return;
+		}
+	}
+}
+
+static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev)
+{
+	struct arphdr *arp;
+	struct slave *slave;
+	struct bonding *bond;
+	unsigned char *arp_ptr;
+	u32 sip, tip;
+
+	if (!(dev->priv_flags & IFF_BONDING) || !(dev->flags & IFF_MASTER))
+		goto out;
+
+	bond = dev->priv;
+	read_lock(&bond->lock);
+
+	dprintk("bond_arp_rcv: bond %s skb->dev %s orig_dev %s\n",
+		bond->dev->name, skb->dev ? skb->dev->name : "NULL",
+		orig_dev ? orig_dev->name : "NULL");
+
+	slave = bond_get_slave_by_dev(bond, orig_dev);
+	if (!slave || !slave_do_arp_validate(bond, slave))
+		goto out_unlock;
+
+	/* ARP header, plus 2 device addresses, plus 2 IP addresses.  */
+	if (!pskb_may_pull(skb, (sizeof(struct arphdr) +
+				 (2 * dev->addr_len) +
+				 (2 * sizeof(u32)))))
+		goto out_unlock;
+
+	arp = skb->nh.arph;
+	if (arp->ar_hln != dev->addr_len ||
+	    skb->pkt_type == PACKET_OTHERHOST ||
+	    skb->pkt_type == PACKET_LOOPBACK ||
+	    arp->ar_hrd != htons(ARPHRD_ETHER) ||
+	    arp->ar_pro != htons(ETH_P_IP) ||
+	    arp->ar_pln != 4)
+		goto out_unlock;
+
+	arp_ptr = (unsigned char *)(arp + 1);
+	arp_ptr += dev->addr_len;
+	memcpy(&sip, arp_ptr, 4);
+	arp_ptr += 4 + dev->addr_len;
+	memcpy(&tip, arp_ptr, 4);
+
+	dprintk("bond_arp_rcv: %s %s/%d av %d sv %d sip %u.%u.%u.%u"
+		" tip %u.%u.%u.%u\n", bond->dev->name, slave->dev->name,
+		slave->state, bond->params.arp_validate,
+		slave_do_arp_validate(bond, slave), NIPQUAD(sip), NIPQUAD(tip));
+
+	/*
+	 * Backup slaves won't see the ARP reply, but do come through
+	 * here for each ARP probe (so we swap the sip/tip to validate
+	 * the probe).  In a "redundant switch, common router" type of
+	 * configuration, the ARP probe will (hopefully) travel from
+	 * the active, through one switch, the router, then the other
+	 * switch before reaching the backup.
+	 */
+	if (slave->state == BOND_STATE_ACTIVE)
+		bond_validate_arp(bond, slave, sip, tip);
+	else
+		bond_validate_arp(bond, slave, tip, sip);
+
+out_unlock:
+	read_unlock(&bond->lock);
+out:
+	dev_kfree_skb(skb);
+	return NET_RX_SUCCESS;
+}
+
 /*
  * this function is called regularly to monitor each slave's link
  * ensuring that traffic is being sent and received when arp monitoring
@@ -2600,7 +2720,8 @@ void bond_activebackup_arp_mon(struct net_device *bond_dev)
 	 */
 	bond_for_each_slave(bond, slave, i) {
 		if (slave->link != BOND_LINK_UP) {
-			if ((jiffies - slave->dev->last_rx) <= delta_in_ticks) {
+			if ((jiffies - slave_last_rx(bond, slave)) <=
+			     delta_in_ticks) {
 
 				slave->link = BOND_LINK_UP;
 
@@ -2645,7 +2766,7 @@ void bond_activebackup_arp_mon(struct net_device *bond_dev)
 
 			if ((slave != bond->curr_active_slave) &&
 			    (!bond->current_arp_slave) &&
-			    (((jiffies - slave->dev->last_rx) >= 3*delta_in_ticks) &&
+			    (((jiffies - slave_last_rx(bond, slave)) >= 3*delta_in_ticks) &&
 			     bond_has_ip(bond))) {
 				/* a backup slave has gone down; three times
 				 * the delta allows the current slave to be
@@ -2692,7 +2813,7 @@ void bond_activebackup_arp_mon(struct net_device *bond_dev)
 		 * if it is up and needs to take over as the curr_active_slave
 		 */
 		if ((((jiffies - slave->dev->trans_start) >= (2*delta_in_ticks)) ||
-	    (((jiffies - slave->dev->last_rx) >= (2*delta_in_ticks)) &&
+	    (((jiffies - slave_last_rx(bond, slave)) >= (2*delta_in_ticks)) &&
 	     bond_has_ip(bond))) &&
 		    ((jiffies - slave->jiffies) >= 2*delta_in_ticks)) {
 
@@ -3315,6 +3436,21 @@ static void bond_unregister_lacpdu(struct bonding *bond)
 	dev_remove_pack(&(BOND_AD_INFO(bond).ad_pkt_type));
 }
 
+void bond_register_arp(struct bonding *bond)
+{
+	struct packet_type *pt = &bond->arp_mon_pt;
+
+	pt->type = htons(ETH_P_ARP);
+	pt->dev = NULL; /*bond->dev;XXX*/
+	pt->func = bond_arp_rcv;
+	dev_add_pack(pt);
+}
+
+void bond_unregister_arp(struct bonding *bond)
+{
+	dev_remove_pack(&bond->arp_mon_pt);
+}
+
 /*---------------------------- Hashing Policies -----------------------------*/
 
 /*
@@ -3401,6 +3537,9 @@ static int bond_open(struct net_device *bond_dev)
 		} else {
 			arp_timer->function = (void *)&bond_loadbalance_arp_mon;
 		}
+		if (bond->params.arp_validate)
+			bond_register_arp(bond);
+
 		add_timer(arp_timer);
 	}
 
@@ -3428,6 +3567,9 @@ static int bond_close(struct net_device *bond_dev)
 		bond_unregister_lacpdu(bond);
 	}
 
+	if (bond->params.arp_validate)
+		bond_unregister_arp(bond);
+
 	write_lock_bh(&bond->lock);
 
 
@@ -4281,6 +4423,8 @@ int bond_parse_parm(char *mode_arg, struct bond_parm_tbl *tbl)
 
 static int bond_check_params(struct bond_params *params)
 {
+	int arp_validate_value;
+
 	/*
 	 * Convert string parameters.
 	 */
@@ -4484,6 +4628,29 @@ static int bond_check_params(struct bond_params *params)
 		arp_interval = 0;
 	}
 
+	if (arp_validate) {
+		if (bond_mode != BOND_MODE_ACTIVEBACKUP) {
+			printk(KERN_ERR DRV_NAME
+	       ": arp_validate only supported in active-backup mode\n");
+			return -EINVAL;
+		}
+		if (!arp_interval) {
+			printk(KERN_ERR DRV_NAME
+			       ": arp_validate requires arp_interval\n");
+			return -EINVAL;
+		}
+
+		arp_validate_value = bond_parse_parm(arp_validate,
+						     arp_validate_tbl);
+		if (arp_validate_value == -1) {
+			printk(KERN_ERR DRV_NAME
+			       ": Error: invalid arp_validate \"%s\"\n",
+			       arp_validate == NULL ? "NULL" : arp_validate);
+			return -EINVAL;
+		}
+	} else
+		arp_validate_value = 0;
+
 	if (miimon) {
 		printk(KERN_INFO DRV_NAME
 		       ": MII link monitoring set to %d ms\n",
@@ -4492,8 +4659,10 @@ static int bond_check_params(struct bond_params *params)
 		int i;
 
 		printk(KERN_INFO DRV_NAME
-		       ": ARP monitoring set to %d ms with %d target(s):",
-		       arp_interval, arp_ip_count);
+		       ": ARP monitoring set to %d ms, validate %s, with %d target(s):",
+		       arp_interval,
+		       arp_validate_tbl[arp_validate_value].modename,
+		       arp_ip_count);
 
 		for (i = 0; i < arp_ip_count; i++)
 			printk (" %s", arp_ip_target[i]);
@@ -4527,6 +4696,7 @@ static int bond_check_params(struct bond_params *params)
 	params->xmit_policy = xmit_hashtype;
 	params->miimon = miimon;
 	params->arp_interval = arp_interval;
+	params->arp_validate = arp_validate_value;
 	params->updelay = updelay;
 	params->downdelay = downdelay;
 	params->use_carrier = use_carrier;
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index 15b6a29bb4d4..ced9ed8f995a 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -51,6 +51,7 @@ extern struct bond_params bonding_defaults;
 extern struct bond_parm_tbl bond_mode_tbl[];
 extern struct bond_parm_tbl bond_lacp_tbl[];
 extern struct bond_parm_tbl xmit_hashtype_tbl[];
+extern struct bond_parm_tbl arp_validate_tbl[];
 
 static int expected_refcount = -1;
 static struct class *netdev_class;
@@ -502,6 +503,53 @@ out:
 }
 static CLASS_DEVICE_ATTR(xmit_hash_policy, S_IRUGO | S_IWUSR, bonding_show_xmit_hash, bonding_store_xmit_hash);
 
+/*
+ * Show and set arp_validate.
+ */
+static ssize_t bonding_show_arp_validate(struct class_device *cd, char *buf)
+{
+	struct bonding *bond = to_bond(cd);
+
+	return sprintf(buf, "%s %d\n",
+		       arp_validate_tbl[bond->params.arp_validate].modename,
+		       bond->params.arp_validate) + 1;
+}
+
+static ssize_t bonding_store_arp_validate(struct class_device *cd, const char *buf, size_t count)
+{
+	int new_value;
+	struct bonding *bond = to_bond(cd);
+
+	new_value = bond_parse_parm((char *)buf, arp_validate_tbl);
+	if (new_value < 0) {
+		printk(KERN_ERR DRV_NAME
+		       ": %s: Ignoring invalid arp_validate value %s\n",
+		       bond->dev->name, buf);
+		return -EINVAL;
+	}
+	if (new_value && (bond->params.mode != BOND_MODE_ACTIVEBACKUP)) {
+		printk(KERN_ERR DRV_NAME
+		       ": %s: arp_validate only supported in active-backup mode.\n",
+		       bond->dev->name);
+		return -EINVAL;
+	}
+	printk(KERN_INFO DRV_NAME ": %s: setting arp_validate to %s (%d).\n",
+	       bond->dev->name, arp_validate_tbl[new_value].modename,
+	       new_value);
+
+	if (!bond->params.arp_validate && new_value) {
+		bond_register_arp(bond);
+	} else if (bond->params.arp_validate && !new_value) {
+		bond_unregister_arp(bond);
+	}
+
+	bond->params.arp_validate = new_value;
+
+	return count;
+}
+
+static CLASS_DEVICE_ATTR(arp_validate, S_IRUGO | S_IWUSR, bonding_show_arp_validate, bonding_store_arp_validate);
+
 /*
  * Show and set the arp timer interval.  There are two tricky bits
  * here.  First, if ARP monitoring is activated, then we must disable
@@ -914,6 +962,11 @@ static ssize_t bonding_store_miimon(struct class_device *cd, const char *buf, si
 			       "ARP monitoring. Disabling ARP monitoring...\n",
 			       bond->dev->name);
 			bond->params.arp_interval = 0;
+			if (bond->params.arp_validate) {
+				bond_unregister_arp(bond);
+				bond->params.arp_validate =
+					BOND_ARP_VALIDATE_NONE;
+			}
 			/* Kill ARP timer, else it brings bond's link down */
 			if (bond->mii_timer.function) {
 				printk(KERN_INFO DRV_NAME
@@ -1273,6 +1326,7 @@ static CLASS_DEVICE_ATTR(ad_partner_mac, S_IRUGO, bonding_show_ad_partner_mac, N
 static struct attribute *per_bond_attrs[] = {
 	&class_device_attr_slaves.attr,
 	&class_device_attr_mode.attr,
+	&class_device_attr_arp_validate.attr,
 	&class_device_attr_arp_interval.attr,
 	&class_device_attr_arp_ip_target.attr,
 	&class_device_attr_downdelay.attr,
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 17caafe58247..db16fee40a5f 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -22,8 +22,8 @@
 #include "bond_3ad.h"
 #include "bond_alb.h"
 
-#define DRV_VERSION	"3.0.3"
-#define DRV_RELDATE	"March 23, 2006"
+#define DRV_VERSION	"3.1.0-test"
+#define DRV_RELDATE	"September 9, 2006"
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"
 
@@ -126,6 +126,7 @@ struct bond_params {
 	int xmit_policy;
 	int miimon;
 	int arp_interval;
+	int arp_validate;
 	int use_carrier;
 	int updelay;
 	int downdelay;
@@ -151,6 +152,7 @@ struct slave {
 	struct slave *prev;
 	int    delay;
 	u32    jiffies;
+	u32    last_arp_rx;
 	s8     link;    /* one of BOND_LINK_XXXX */
 	s8     state;   /* one of BOND_STATE_XXXX */
 	u32    original_flags;
@@ -198,6 +200,7 @@ struct bonding {
 	struct   bond_params params;
 	struct   list_head vlan_list;
 	struct   vlan_group *vlgrp;
+	struct   packet_type arp_mon_pt;
 };
 
 /**
@@ -228,6 +231,25 @@ static inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
 	return (struct bonding *)slave->dev->master->priv;
 }
 
+#define BOND_ARP_VALIDATE_NONE		0
+#define BOND_ARP_VALIDATE_ACTIVE	(1 << BOND_STATE_ACTIVE)
+#define BOND_ARP_VALIDATE_BACKUP	(1 << BOND_STATE_BACKUP)
+#define BOND_ARP_VALIDATE_ALL		(BOND_ARP_VALIDATE_ACTIVE | \
+					 BOND_ARP_VALIDATE_BACKUP)
+
+extern inline int slave_do_arp_validate(struct bonding *bond, struct slave *slave)
+{
+	return bond->params.arp_validate & (1 << slave->state);
+}
+
+extern inline u32 slave_last_rx(struct bonding *bond, struct slave *slave)
+{
+	if (slave_do_arp_validate(bond, slave))
+		return slave->last_arp_rx;
+
+	return slave->dev->last_rx;
+}
+
 static inline void bond_set_slave_inactive_flags(struct slave *slave)
 {
 	struct bonding *bond = slave->dev->master->priv;
@@ -235,12 +257,14 @@ static inline void bond_set_slave_inactive_flags(struct slave *slave)
 	    bond->params.mode != BOND_MODE_ALB)
 		slave->state = BOND_STATE_BACKUP;
 	slave->dev->priv_flags |= IFF_SLAVE_INACTIVE;
+	if (slave_do_arp_validate(bond, slave))
+		slave->dev->priv_flags |= IFF_SLAVE_NEEDARP;
 }
 
 static inline void bond_set_slave_active_flags(struct slave *slave)
 {
 	slave->state = BOND_STATE_ACTIVE;
-	slave->dev->priv_flags &= ~IFF_SLAVE_INACTIVE;
+	slave->dev->priv_flags &= ~(IFF_SLAVE_INACTIVE | IFF_SLAVE_NEEDARP);
 }
 
 static inline void bond_set_master_3ad_flags(struct bonding *bond)
@@ -284,6 +308,8 @@ int bond_parse_parm(char *mode_arg, struct bond_parm_tbl *tbl);
 const char *bond_mode_name(int mode);
 void bond_select_active_slave(struct bonding *bond);
 void bond_change_active_slave(struct bonding *bond, struct slave *new_active);
+void bond_register_arp(struct bonding *);
+void bond_unregister_arp(struct bonding *);
 
 #endif /* _LINUX_BONDING_H */
 
diff --git a/include/linux/if.h b/include/linux/if.h
index a023ec1274fe..8018c2e22c0c 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -60,6 +60,7 @@
 #define IFF_MASTER_8023AD	0x8	/* bonding master, 802.3ad. 	*/
 #define IFF_MASTER_ALB	0x10		/* bonding master, balance-alb.	*/
 #define IFF_BONDING	0x20		/* bonding master or slave	*/
+#define IFF_SLAVE_NEEDARP 0x40		/* need ARPs for validation	*/
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 43289127b458..afd80eff2725 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1016,7 +1016,8 @@ static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
 }
 
 /* On bonding slaves other than the currently active slave, suppress
- * duplicates except for 802.3ad ETH_P_SLOW and alb non-mcast/bcast.
+ * duplicates except for 802.3ad ETH_P_SLOW, alb non-mcast/bcast, and
+ * ARP on active-backup slaves with arp_validate enabled.
  */
 static inline int skb_bond_should_drop(struct sk_buff *skb)
 {
@@ -1025,6 +1026,10 @@ static inline int skb_bond_should_drop(struct sk_buff *skb)
 
 	if (master &&
 	    (dev->priv_flags & IFF_SLAVE_INACTIVE)) {
+		if ((dev->priv_flags & IFF_SLAVE_NEEDARP) &&
+		    skb->protocol == __constant_htons(ETH_P_ARP))
+			return 0;
+
 		if (master->priv_flags & IFF_MASTER_ALB) {
 			if (skb->pkt_type != PACKET_BROADCAST &&
 			    skb->pkt_type != PACKET_MULTICAST)
-- 
cgit v1.2.3


From 3212fe1594e577463bc8601d28aa008f520c3377 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Mon, 25 Sep 2006 16:25:31 -0700
Subject: [PATCH] cpu to node relationship fixup: map cpu to node

Assume that a cpu is *physically* offlined at boot time...

Because smpboot.c::smp_boot_cpu_map() canoot find cpu's sapicid,
numa.c::build_cpu_to_node_map() cannot build cpu<->node map for
offlined cpu.

For such cpus, cpu_to_node map should be fixed at cpu-hot-add.
This mapping should be done before cpu onlining.

This patch also handles cpu hotremove case.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/numa.c     | 34 +++++++++++++++++++++++++++++++---
 arch/ia64/kernel/topology.c |  6 +++++-
 include/asm-ia64/numa.h     |  4 ++++
 3 files changed, 40 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c
index 1cc360c83e7a..20340631179f 100644
--- a/arch/ia64/kernel/numa.c
+++ b/arch/ia64/kernel/numa.c
@@ -29,6 +29,36 @@ EXPORT_SYMBOL(cpu_to_node_map);
 
 cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned;
 
+void __cpuinit map_cpu_to_node(int cpu, int nid)
+{
+	int oldnid;
+	if (nid < 0) { /* just initialize by zero */
+		cpu_to_node_map[cpu] = 0;
+		return;
+	}
+	/* sanity check first */
+	oldnid = cpu_to_node_map[cpu];
+	if (cpu_isset(cpu, node_to_cpu_mask[oldnid])) {
+		return; /* nothing to do */
+	}
+	/* we don't have cpu-driven node hot add yet...
+	   In usual case, node is created from SRAT at boot time. */
+	if (!node_online(nid))
+		nid = first_online_node;
+	cpu_to_node_map[cpu] = nid;
+	cpu_set(cpu, node_to_cpu_mask[nid]);
+	return;
+}
+
+void __cpuinit unmap_cpu_from_node(int cpu, int nid)
+{
+	WARN_ON(!cpu_isset(cpu, node_to_cpu_mask[nid]));
+	WARN_ON(cpu_to_node_map[cpu] != nid);
+	cpu_to_node_map[cpu] = 0;
+	cpu_clear(cpu, node_to_cpu_mask[nid]);
+}
+
+
 /**
  * build_cpu_to_node_map - setup cpu to node and node to cpumask arrays
  *
@@ -49,8 +79,6 @@ void __init build_cpu_to_node_map(void)
 				node = node_cpuid[i].nid;
 				break;
 			}
-		cpu_to_node_map[cpu] = (node >= 0) ? node : 0;
-		if (node >= 0)
-			cpu_set(cpu, node_to_cpu_mask[node]);
+		map_cpu_to_node(cpu, node);
 	}
 }
diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index f648c610b10c..05bdf7affb43 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -36,6 +36,9 @@ int arch_register_cpu(int num)
 	 */
 	if (!can_cpei_retarget() && is_cpu_cpei_target(num))
 		sysfs_cpus[num].cpu.no_control = 1;
+#ifdef CONFIG_NUMA
+	map_cpu_to_node(num, node_cpuid[num].nid);
+#endif
 #endif
 
 	return register_cpu(&sysfs_cpus[num].cpu, num);
@@ -45,7 +48,8 @@ int arch_register_cpu(int num)
 
 void arch_unregister_cpu(int num)
 {
-	return unregister_cpu(&sysfs_cpus[num].cpu);
+	unregister_cpu(&sysfs_cpus[num].cpu);
+	unmap_cpu_from_node(num, cpu_to_node(num));
 }
 EXPORT_SYMBOL(arch_register_cpu);
 EXPORT_SYMBOL(arch_unregister_cpu);
diff --git a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h
index e5a8260593a5..e0a1d173e42d 100644
--- a/include/asm-ia64/numa.h
+++ b/include/asm-ia64/numa.h
@@ -64,6 +64,10 @@ extern int paddr_to_nid(unsigned long paddr);
 
 #define local_nodeid (cpu_to_node_map[smp_processor_id()])
 
+extern void map_cpu_to_node(int cpu, int nid);
+extern void unmap_cpu_from_node(int cpu, int nid);
+
+
 #else /* !CONFIG_NUMA */
 
 #define paddr_to_nid(addr)	0
-- 
cgit v1.2.3


From ddd5d35a8f7c1924049e8b6877b3177c1787e6a3 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor@insightbb.com>
Date: Thu, 10 Aug 2006 01:18:18 -0400
Subject: class_device_create(): make fmt argument 'const char *'

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 3 ++-
 include/linux/device.h | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 46336f1b93b6..75057aaab809 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -679,7 +679,8 @@ int class_device_register(struct class_device *class_dev)
 struct class_device *class_device_create(struct class *cls,
 					 struct class_device *parent,
 					 dev_t devt,
-					 struct device *device, char *fmt, ...)
+					 struct device *device,
+					 const char *fmt, ...)
 {
 	va_list args;
 	struct class_device *class_dev = NULL;
diff --git a/include/linux/device.h b/include/linux/device.h
index 1e5f30da98bc..1fec28546525 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -277,7 +277,7 @@ extern struct class_device *class_device_create(struct class *cls,
 						struct class_device *parent,
 						dev_t devt,
 						struct device *device,
-						char *fmt, ...)
+						const char *fmt, ...)
 					__attribute__((format(printf,5,6)));
 extern void class_device_destroy(struct class *cls, dev_t devt);
 
-- 
cgit v1.2.3


From 5cbe5f8a5897470698222ac9924429056e57d84c Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 10 Aug 2006 01:19:19 -0400
Subject: device_create(): make fmt argument 'const char *'

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 2 +-
 include/linux/device.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 04d089fd4f76..5d4b7e04471e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -583,7 +583,7 @@ static void device_create_release(struct device *dev)
  * been created with a call to class_create().
  */
 struct device *device_create(struct class *class, struct device *parent,
-			     dev_t devt, char *fmt, ...)
+			     dev_t devt, const char *fmt, ...)
 {
 	va_list args;
 	struct device *dev = NULL;
diff --git a/include/linux/device.h b/include/linux/device.h
index 1fec28546525..8d92013f9ce1 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -384,7 +384,7 @@ extern void device_reprobe(struct device *dev);
  * Easy functions for dynamically creating devices on the fly
  */
 extern struct device *device_create(struct class *cls, struct device *parent,
-				    dev_t devt, char *fmt, ...)
+				    dev_t devt, const char *fmt, ...)
 				    __attribute__((format(printf,4,5)));
 extern void device_destroy(struct class *cls, dev_t devt);
 
-- 
cgit v1.2.3


From ab7d7371acc68fa9130b079a9ba879191202035f Mon Sep 17 00:00:00 2001
From: Miguel Ojeda Sandonis <maxextreme@gmail.com>
Date: Wed, 13 Sep 2006 15:34:05 +0200
Subject: Driver core: add const to class_create

Adds const to class_create second parameter, because:

struct class {
	const char * name;

	/*...*/
}

Signed-off-by: Miguel Ojeda Sandonis <maxextreme@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 2 +-
 include/linux/device.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 75057aaab809..e078bc21d52e 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -197,7 +197,7 @@ static int class_device_create_uevent(struct class_device *class_dev,
  * Note, the pointer created here is to be destroyed when finished by
  * making a call to class_destroy().
  */
-struct class *class_create(struct module *owner, char *name)
+struct class *class_create(struct module *owner, const char *name)
 {
 	struct class *cls;
 	int retval;
diff --git a/include/linux/device.h b/include/linux/device.h
index 8d92013f9ce1..8a648cd94fa9 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -271,7 +271,7 @@ struct class_interface {
 extern int class_interface_register(struct class_interface *);
 extern void class_interface_unregister(struct class_interface *);
 
-extern struct class *class_create(struct module *owner, char *name);
+extern struct class *class_create(struct module *owner, const char *name);
 extern void class_destroy(struct class *cls);
 extern struct class_device *class_device_create(struct class *cls,
 						struct class_device *parent,
-- 
cgit v1.2.3


From 7c8265f51073bc8632a99de78d5fd19117ed78b7 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@osdl.org>
Date: Sat, 24 Jun 2006 14:50:29 -0700
Subject: Suspend infrastructure cleanup and extension

Allow devices to participate in the suspend process more intimately,
in particular, allow the final phase (with interrupts disabled) to
also be open to normal devices, not just system devices.

Also, allow classes to participate in device suspend.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/resume.c  |  28 ++++++++--
 drivers/base/power/suspend.c | 122 ++++++++++++++++++++++++++++++++-----------
 include/linux/device.h       |  11 +++-
 include/linux/pm.h           |   1 +
 kernel/power/main.c          |   4 ++
 5 files changed, 130 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c
index 826093ef4c7e..48e3d49d7b65 100644
--- a/drivers/base/power/resume.c
+++ b/drivers/base/power/resume.c
@@ -38,13 +38,35 @@ int resume_device(struct device * dev)
 		dev_dbg(dev,"resuming\n");
 		error = dev->bus->resume(dev);
 	}
+	if (dev->class && dev->class->resume) {
+		dev_dbg(dev,"class resume\n");
+		error = dev->class->resume(dev);
+	}
 	up(&dev->sem);
 	TRACE_RESUME(error);
 	return error;
 }
 
 
+static int resume_device_early(struct device * dev)
+{
+	int error = 0;
 
+	TRACE_DEVICE(dev);
+	TRACE_RESUME(0);
+	if (dev->bus && dev->bus->resume_early) {
+		dev_dbg(dev,"EARLY resume\n");
+		error = dev->bus->resume_early(dev);
+	}
+	TRACE_RESUME(error);
+	return error;
+}
+
+/*
+ * Resume the devices that have either not gone through
+ * the late suspend, or that did go through it but also
+ * went through the early resume
+ */
 void dpm_resume(void)
 {
 	down(&dpm_list_sem);
@@ -99,10 +121,8 @@ void dpm_power_up(void)
 		struct list_head * entry = dpm_off_irq.next;
 		struct device * dev = to_device(entry);
 
-		get_device(dev);
-		list_move_tail(entry, &dpm_active);
-		resume_device(dev);
-		put_device(dev);
+		list_move_tail(entry, &dpm_off);
+		resume_device_early(dev);
 	}
 }
 
diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 69509e02f703..10e8032aee1a 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -65,7 +65,19 @@ int suspend_device(struct device * dev, pm_message_t state)
 
 	dev->power.prev_state = dev->power.power_state;
 
-	if (dev->bus && dev->bus->suspend && !dev->power.power_state.event) {
+	if (dev->class && dev->class->suspend && !dev->power.power_state.event) {
+		dev_dbg(dev, "class %s%s\n",
+			suspend_verb(state.event),
+			((state.event == PM_EVENT_SUSPEND)
+					&& device_may_wakeup(dev))
+				? ", may wakeup"
+				: ""
+			);
+		error = dev->class->suspend(dev, state);
+		suspend_report_result(dev->class->suspend, error);
+	}
+
+	if (!error && dev->bus && dev->bus->suspend && !dev->power.power_state.event) {
 		dev_dbg(dev, "%s%s\n",
 			suspend_verb(state.event),
 			((state.event == PM_EVENT_SUSPEND)
@@ -81,15 +93,74 @@ int suspend_device(struct device * dev, pm_message_t state)
 }
 
 
+/*
+ * This is called with interrupts off, only a single CPU
+ * running. We can't do down() on a semaphore (and we don't
+ * need the protection)
+ */
+static int suspend_device_late(struct device *dev, pm_message_t state)
+{
+	int error = 0;
+
+	if (dev->power.power_state.event) {
+		dev_dbg(dev, "PM: suspend_late %d-->%d\n",
+			dev->power.power_state.event, state.event);
+	}
+
+	if (dev->bus && dev->bus->suspend_late && !dev->power.power_state.event) {
+		dev_dbg(dev, "LATE %s%s\n",
+			suspend_verb(state.event),
+			((state.event == PM_EVENT_SUSPEND)
+					&& device_may_wakeup(dev))
+				? ", may wakeup"
+				: ""
+			);
+		error = dev->bus->suspend_late(dev, state);
+		suspend_report_result(dev->bus->suspend_late, error);
+	}
+	return error;
+}
+
+/**
+ *	device_prepare_suspend - save state and prepare to suspend
+ *
+ *	NOTE! Devices cannot detach at this point - not only do we
+ *	hold the device list semaphores over the whole prepare, but
+ *	the whole point is to do non-invasive preparatory work, not
+ *	the actual suspend.
+ */
+int device_prepare_suspend(pm_message_t state)
+{
+	int error = 0;
+	struct device * dev;
+
+	down(&dpm_sem);
+	down(&dpm_list_sem);
+	list_for_each_entry_reverse(dev, &dpm_active, power.entry) {
+		if (!dev->bus || !dev->bus->suspend_prepare)
+			continue;
+		error = dev->bus->suspend_prepare(dev, state);
+		if (error)
+			break;
+	}
+	up(&dpm_list_sem);
+	up(&dpm_sem);
+	return error;
+}
+
 /**
  *	device_suspend - Save state and stop all devices in system.
  *	@state:		Power state to put each device in.
  *
  *	Walk the dpm_active list, call ->suspend() for each device, and move
- *	it to dpm_off.
- *	Check the return value for each. If it returns 0, then we move the
- *	the device to the dpm_off list. If it returns -EAGAIN, we move it to
- *	the dpm_off_irq list. If we get a different error, try and back out.
+ *	it to the dpm_off list.
+ *
+ *	(For historical reasons, if it returns -EAGAIN, that used to mean
+ *	that the device would be called again with interrupts disabled.
+ *	These days, we use the "suspend_late()" callback for that, so we
+ *	print a warning and consider it an error).
+ *
+ *	If we get a different error, try and back out.
  *
  *	If we hit a failure with any of the devices, call device_resume()
  *	above to bring the suspended devices back to life.
@@ -115,39 +186,27 @@ int device_suspend(pm_message_t state)
 
 		/* Check if the device got removed */
 		if (!list_empty(&dev->power.entry)) {
-			/* Move it to the dpm_off or dpm_off_irq list */
+			/* Move it to the dpm_off list */
 			if (!error)
 				list_move(&dev->power.entry, &dpm_off);
-			else if (error == -EAGAIN) {
-				list_move(&dev->power.entry, &dpm_off_irq);
-				error = 0;
-			}
 		}
 		if (error)
 			printk(KERN_ERR "Could not suspend device %s: "
-				"error %d\n", kobject_name(&dev->kobj), error);
+				"error %d%s\n",
+				kobject_name(&dev->kobj), error,
+				error == -EAGAIN ? " (please convert to suspend_late)" : "");
 		put_device(dev);
 	}
 	up(&dpm_list_sem);
-	if (error) {
-		/* we failed... before resuming, bring back devices from
-		 * dpm_off_irq list back to main dpm_off list, we do want
-		 * to call resume() on them, in case they partially suspended
-		 * despite returning -EAGAIN
-		 */
-		while (!list_empty(&dpm_off_irq)) {
-			struct list_head * entry = dpm_off_irq.next;
-			list_move(entry, &dpm_off);
-		}
+	if (error)
 		dpm_resume();
-	}
+
 	up(&dpm_sem);
 	return error;
 }
 
 EXPORT_SYMBOL_GPL(device_suspend);
 
-
 /**
  *	device_power_down - Shut down special devices.
  *	@state:		Power state to enter.
@@ -162,14 +221,17 @@ int device_power_down(pm_message_t state)
 	int error = 0;
 	struct device * dev;
 
-	list_for_each_entry_reverse(dev, &dpm_off_irq, power.entry) {
-		if ((error = suspend_device(dev, state)))
-			break;
+	while (!list_empty(&dpm_off)) {
+		struct list_head * entry = dpm_off.prev;
+
+		dev = to_device(entry);
+		error = suspend_device_late(dev, state);
+		if (error)
+			goto Error;
+		list_move(&dev->power.entry, &dpm_off_irq);
 	}
-	if (error)
-		goto Error;
-	if ((error = sysdev_suspend(state)))
-		goto Error;
+
+	error = sysdev_suspend(state);
  Done:
 	return error;
  Error:
diff --git a/include/linux/device.h b/include/linux/device.h
index 8a648cd94fa9..b40be6fca6fa 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -51,8 +51,12 @@ struct bus_type {
 	int		(*probe)(struct device * dev);
 	int		(*remove)(struct device * dev);
 	void		(*shutdown)(struct device * dev);
-	int		(*suspend)(struct device * dev, pm_message_t state);
-	int		(*resume)(struct device * dev);
+
+	int (*suspend_prepare)(struct device * dev, pm_message_t state);
+	int (*suspend)(struct device * dev, pm_message_t state);
+	int (*suspend_late)(struct device * dev, pm_message_t state);
+	int (*resume_early)(struct device * dev);
+	int (*resume)(struct device * dev);
 };
 
 extern int bus_register(struct bus_type * bus);
@@ -154,6 +158,9 @@ struct class {
 
 	void	(*release)(struct class_device *dev);
 	void	(*class_release)(struct class *class);
+
+	int	(*suspend)(struct device *, pm_message_t state);
+	int	(*resume)(struct device *);
 };
 
 extern int class_register(struct class *);
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 658c1b93d5bb..096fb6f754cf 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -190,6 +190,7 @@ extern void device_resume(void);
 extern suspend_disk_method_t pm_disk_mode;
 
 extern int device_suspend(pm_message_t state);
+extern int device_prepare_suspend(pm_message_t state);
 
 #define device_set_wakeup_enable(dev,val) \
 	((dev)->power.should_wakeup = !!(val))
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6d295c776794..0c3ed6ac938e 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -57,6 +57,10 @@ static int suspend_prepare(suspend_state_t state)
 	if (!pm_ops || !pm_ops->enter)
 		return -EPERM;
 
+	error = device_prepare_suspend(PMSG_SUSPEND);
+	if (error)
+		return error;
+
 	pm_prepare_console();
 
 	disable_nonboot_cpus();
-- 
cgit v1.2.3


From cbd69dbbf1adfce6e048f15afc8629901ca9dae5 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@osdl.org>
Date: Sat, 24 Jun 2006 14:50:29 -0700
Subject: Suspend changes for PCI core

Changes the PCI core to use the new suspend infrastructure changes.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci-driver.c | 41 ++++++++++++++++++++++++++++++++++++++++-
 include/linux/pci.h      |  3 +++
 2 files changed, 43 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 474e9cd0e9e4..9e7d6ceb3805 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -264,6 +264,19 @@ static int pci_device_remove(struct device * dev)
 	return 0;
 }
 
+static int pci_device_suspend_prepare(struct device * dev, pm_message_t state)
+{
+	struct pci_dev * pci_dev = to_pci_dev(dev);
+	struct pci_driver * drv = pci_dev->driver;
+	int i = 0;
+
+	if (drv && drv->suspend_prepare) {
+		i = drv->suspend_prepare(pci_dev, state);
+		suspend_report_result(drv->suspend_prepare, i);
+	}
+	return i;
+}
+
 static int pci_device_suspend(struct device * dev, pm_message_t state)
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
@@ -279,6 +292,18 @@ static int pci_device_suspend(struct device * dev, pm_message_t state)
 	return i;
 }
 
+static int pci_device_suspend_late(struct device * dev, pm_message_t state)
+{
+	struct pci_dev * pci_dev = to_pci_dev(dev);
+	struct pci_driver * drv = pci_dev->driver;
+	int i = 0;
+
+	if (drv && drv->suspend_late) {
+		i = drv->suspend_late(pci_dev, state);
+		suspend_report_result(drv->suspend_late, i);
+	}
+	return i;
+}
 
 /*
  * Default resume method for devices that have no driver provided resume,
@@ -313,6 +338,17 @@ static int pci_device_resume(struct device * dev)
 	return error;
 }
 
+static int pci_device_resume_early(struct device * dev)
+{
+	int error = 0;
+	struct pci_dev * pci_dev = to_pci_dev(dev);
+	struct pci_driver * drv = pci_dev->driver;
+
+	if (drv && drv->resume_early)
+		error = drv->resume_early(pci_dev);
+	return error;
+}
+
 static void pci_device_shutdown(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
@@ -508,9 +544,12 @@ struct bus_type pci_bus_type = {
 	.uevent		= pci_uevent,
 	.probe		= pci_device_probe,
 	.remove		= pci_device_remove,
+	.suspend_prepare= pci_device_suspend_prepare,
 	.suspend	= pci_device_suspend,
-	.shutdown	= pci_device_shutdown,
+	.suspend_late	= pci_device_suspend_late,
+	.resume_early	= pci_device_resume_early,
 	.resume		= pci_device_resume,
+	.shutdown	= pci_device_shutdown,
 	.dev_attrs	= pci_dev_attrs,
 };
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 8565b81d7fbc..4b2e629467c7 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -345,7 +345,10 @@ struct pci_driver {
 	const struct pci_device_id *id_table;	/* must be non-NULL for probe to be called */
 	int  (*probe)  (struct pci_dev *dev, const struct pci_device_id *id);	/* New device inserted */
 	void (*remove) (struct pci_dev *dev);	/* Device removed (NULL if not a hot-plug capable driver) */
+	int  (*suspend_prepare) (struct pci_dev *dev, pm_message_t state);
 	int  (*suspend) (struct pci_dev *dev, pm_message_t state);	/* Device suspended */
+	int  (*suspend_late) (struct pci_dev *dev, pm_message_t state);
+	int  (*resume_early) (struct pci_dev *dev);
 	int  (*resume) (struct pci_dev *dev);	                /* Device woken up */
 	int  (*enable_wake) (struct pci_dev *dev, pci_power_t state, int enable);   /* Enable wake event */
 	void (*shutdown) (struct pci_dev *dev);
-- 
cgit v1.2.3


From 82bb67f2c1f9ef438c56ac24e7dca027fe7289b5 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:04 -0700
Subject: PM: define PM_EVENT_PRETHAW

This adds a new pm_message_t event type to use when preparing to restore a
swsusp snapshot.  Devices that have been initialized by Linux after resume
(rather than left in power-up-reset state) may need to be reset; this new
event type give drivers the chance to do that.

The drivers that will care about this are those which understand more hardware
states than just "on" and "reset", relying on hardware state during resume()
methods to be either the state left by the preceding suspend(), or a
power-lost reset.  The best current example of this class of drivers are USB
host controller drivers, which currently do not work through swsusp when
they're statically linked.

When the swsusp freeze/thaw mechanism kicks in, a troublesome third state
could exist: one state set up by a different kernel instance, before a
snapshot image is resumed.  This mechanism lets drivers prevent that state.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/pm.h | 62 +++++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 47 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/pm.h b/include/linux/pm.h
index 096fb6f754cf..6b27e07aef19 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -142,29 +142,61 @@ typedef struct pm_message {
 } pm_message_t;
 
 /*
- * There are 4 important states driver can be in:
- * ON     -- driver is working
- * FREEZE -- stop operations and apply whatever policy is applicable to a
- *           suspended driver of that class, freeze queues for block like IDE
- *           does, drop packets for ethernet, etc... stop DMA engine too etc...
- *           so a consistent image can be saved; but do not power any hardware
- *           down.
- * SUSPEND - like FREEZE, but hardware is doing as much powersaving as
- *           possible. Roughly pci D3.
+ * Several driver power state transitions are externally visible, affecting
+ * the state of pending I/O queues and (for drivers that touch hardware)
+ * interrupts, wakeups, DMA, and other hardware state.  There may also be
+ * internal transitions to various low power modes, which are transparent
+ * to the rest of the driver stack (such as a driver that's ON gating off
+ * clocks which are not in active use).
  *
- * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3
- * (SUSPEND).  We'll need to fix the drivers. So yes, putting 3 to all different
- * defines is intentional, and will go away as soon as drivers are fixed.  Also
- * note that typedef is neccessary, we'll probably want to switch to
- *   typedef struct pm_message_t { int event; int flags; } pm_message_t
- * or something similar soon.
+ * One transition is triggered by resume(), after a suspend() call; the
+ * message is implicit:
+ *
+ * ON		Driver starts working again, responding to hardware events
+ * 		and software requests.  The hardware may have gone through
+ * 		a power-off reset, or it may have maintained state from the
+ * 		previous suspend() which the driver will rely on while
+ * 		resuming.  On most platforms, there are no restrictions on
+ * 		availability of resources like clocks during resume().
+ *
+ * Other transitions are triggered by messages sent using suspend().  All
+ * these transitions quiesce the driver, so that I/O queues are inactive.
+ * That commonly entails turning off IRQs and DMA; there may be rules
+ * about how to quiesce that are specific to the bus or the device's type.
+ * (For example, network drivers mark the link state.)  Other details may
+ * differ according to the message:
+ *
+ * SUSPEND	Quiesce, enter a low power device state appropriate for
+ * 		the upcoming system state (such as PCI_D3hot), and enable
+ * 		wakeup events as appropriate.
+ *
+ * FREEZE	Quiesce operations so that a consistent image can be saved;
+ * 		but do NOT otherwise enter a low power device state, and do
+ * 		NOT emit system wakeup events.
+ *
+ * PRETHAW	Quiesce as if for FREEZE; additionally, prepare for restoring
+ * 		the system from a snapshot taken after an earlier FREEZE.
+ * 		Some drivers will need to reset their hardware state instead
+ * 		of preserving it, to ensure that it's never mistaken for the
+ * 		state which that earlier snapshot had set up.
+ *
+ * A minimally power-aware driver treats all messages as SUSPEND, fully
+ * reinitializes its device during resume() -- whether or not it was reset
+ * during the suspend/resume cycle -- and can't issue wakeup events.
+ *
+ * More power-aware drivers may also use low power states at runtime as
+ * well as during system sleep states like PM_SUSPEND_STANDBY.  They may
+ * be able to use wakeup events to exit from runtime low-power states,
+ * or from system low-power states such as standby or suspend-to-RAM.
  */
 
 #define PM_EVENT_ON 0
 #define PM_EVENT_FREEZE 1
 #define PM_EVENT_SUSPEND 2
+#define PM_EVENT_PRETHAW 3
 
 #define PMSG_FREEZE	((struct pm_message){ .event = PM_EVENT_FREEZE, })
+#define PMSG_PRETHAW	((struct pm_message){ .event = PM_EVENT_PRETHAW, })
 #define PMSG_SUSPEND	((struct pm_message){ .event = PM_EVENT_SUSPEND, })
 #define PMSG_ON		((struct pm_message){ .event = PM_EVENT_ON, })
 
-- 
cgit v1.2.3


From 1d3a82af45428c5e8deaa119cdeb79611ae46371 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 30 Aug 2006 14:09:47 -0700
Subject: PM: no suspend_prepare() phase

Remove the new suspend_prepare() phase.  It doesn't seem very usable,
has never been tested, doesn't address fault cleanup, and would need
a sibling resume_complete(); plus there are no real use cases.  It
could be restored later if those issues get resolved.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/suspend.c | 27 ---------------------------
 drivers/pci/pci-driver.c     | 14 --------------
 include/linux/device.h       |  1 -
 include/linux/pci.h          |  1 -
 kernel/power/main.c          |  4 ----
 5 files changed, 47 deletions(-)

(limited to 'include')

diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index e86db83746ac..6453c25103d2 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -117,33 +117,6 @@ static int suspend_device_late(struct device *dev, pm_message_t state)
 	return error;
 }
 
-/**
- *	device_prepare_suspend - save state and prepare to suspend
- *
- *	NOTE! Devices cannot detach at this point - not only do we
- *	hold the device list semaphores over the whole prepare, but
- *	the whole point is to do non-invasive preparatory work, not
- *	the actual suspend.
- */
-int device_prepare_suspend(pm_message_t state)
-{
-	int error = 0;
-	struct device * dev;
-
-	down(&dpm_sem);
-	down(&dpm_list_sem);
-	list_for_each_entry_reverse(dev, &dpm_active, power.entry) {
-		if (!dev->bus || !dev->bus->suspend_prepare)
-			continue;
-		error = dev->bus->suspend_prepare(dev, state);
-		if (error)
-			break;
-	}
-	up(&dpm_list_sem);
-	up(&dpm_sem);
-	return error;
-}
-
 /**
  *	device_suspend - Save state and stop all devices in system.
  *	@state:		Power state to put each device in.
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 9e7d6ceb3805..8948ac9ab681 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -264,19 +264,6 @@ static int pci_device_remove(struct device * dev)
 	return 0;
 }
 
-static int pci_device_suspend_prepare(struct device * dev, pm_message_t state)
-{
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
-	int i = 0;
-
-	if (drv && drv->suspend_prepare) {
-		i = drv->suspend_prepare(pci_dev, state);
-		suspend_report_result(drv->suspend_prepare, i);
-	}
-	return i;
-}
-
 static int pci_device_suspend(struct device * dev, pm_message_t state)
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
@@ -544,7 +531,6 @@ struct bus_type pci_bus_type = {
 	.uevent		= pci_uevent,
 	.probe		= pci_device_probe,
 	.remove		= pci_device_remove,
-	.suspend_prepare= pci_device_suspend_prepare,
 	.suspend	= pci_device_suspend,
 	.suspend_late	= pci_device_suspend_late,
 	.resume_early	= pci_device_resume_early,
diff --git a/include/linux/device.h b/include/linux/device.h
index b40be6fca6fa..b3646462d6dc 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -52,7 +52,6 @@ struct bus_type {
 	int		(*remove)(struct device * dev);
 	void		(*shutdown)(struct device * dev);
 
-	int (*suspend_prepare)(struct device * dev, pm_message_t state);
 	int (*suspend)(struct device * dev, pm_message_t state);
 	int (*suspend_late)(struct device * dev, pm_message_t state);
 	int (*resume_early)(struct device * dev);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 4b2e629467c7..9514bbfe96e2 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -345,7 +345,6 @@ struct pci_driver {
 	const struct pci_device_id *id_table;	/* must be non-NULL for probe to be called */
 	int  (*probe)  (struct pci_dev *dev, const struct pci_device_id *id);	/* New device inserted */
 	void (*remove) (struct pci_dev *dev);	/* Device removed (NULL if not a hot-plug capable driver) */
-	int  (*suspend_prepare) (struct pci_dev *dev, pm_message_t state);
 	int  (*suspend) (struct pci_dev *dev, pm_message_t state);	/* Device suspended */
 	int  (*suspend_late) (struct pci_dev *dev, pm_message_t state);
 	int  (*resume_early) (struct pci_dev *dev);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 0c3ed6ac938e..6d295c776794 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -57,10 +57,6 @@ static int suspend_prepare(suspend_state_t state)
 	if (!pm_ops || !pm_ops->enter)
 		return -EPERM;
 
-	error = device_prepare_suspend(PMSG_SUSPEND);
-	if (error)
-		return error;
-
 	pm_prepare_console();
 
 	disable_nonboot_cpus();
-- 
cgit v1.2.3


From 386415d88b1ae50304f9c61aa3e0db082fa90428 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 3 Sep 2006 13:16:45 -0700
Subject: PM: platform_bus and late_suspend/early_resume

Teach platform_bus about the new suspend_late/resume_early PM calls,
issued with IRQs off.  Do we really need sysdev and friends any more,
or can janitors start switching its users over to platform_device so
we can do a minor code-ectomy?

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/platform.c         | 30 ++++++++++++++++++++++++++++--
 include/linux/platform_device.h |  2 ++
 2 files changed, 30 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index 2b8755db76c6..940ce41f1887 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -505,12 +505,36 @@ static int platform_match(struct device * dev, struct device_driver * drv)
 	return (strncmp(pdev->name, drv->name, BUS_ID_SIZE) == 0);
 }
 
-static int platform_suspend(struct device * dev, pm_message_t state)
+static int platform_suspend(struct device *dev, pm_message_t mesg)
 {
 	int ret = 0;
 
 	if (dev->driver && dev->driver->suspend)
-		ret = dev->driver->suspend(dev, state);
+		ret = dev->driver->suspend(dev, mesg);
+
+	return ret;
+}
+
+static int platform_suspend_late(struct device *dev, pm_message_t mesg)
+{
+	struct platform_driver *drv = to_platform_driver(dev->driver);
+	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
+	int ret = 0;
+
+	if (dev->driver && drv->suspend_late)
+		ret = drv->suspend_late(pdev, mesg);
+
+	return ret;
+}
+
+static int platform_resume_early(struct device *dev)
+{
+	struct platform_driver *drv = to_platform_driver(dev->driver);
+	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
+	int ret = 0;
+
+	if (dev->driver && drv->resume_early)
+		ret = drv->resume_early(pdev);
 
 	return ret;
 }
@@ -531,6 +555,8 @@ struct bus_type platform_bus_type = {
 	.match		= platform_match,
 	.uevent		= platform_uevent,
 	.suspend	= platform_suspend,
+	.suspend_late	= platform_suspend_late,
+	.resume_early	= platform_resume_early,
 	.resume		= platform_resume,
 };
 EXPORT_SYMBOL_GPL(platform_bus_type);
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index 782090c68932..29cd6dee13db 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -49,6 +49,8 @@ struct platform_driver {
 	int (*remove)(struct platform_device *);
 	void (*shutdown)(struct platform_device *);
 	int (*suspend)(struct platform_device *, pm_message_t state);
+	int (*suspend_late)(struct platform_device *, pm_message_t state);
+	int (*resume_early)(struct platform_device *);
 	int (*resume)(struct platform_device *);
 	struct device_driver driver;
 };
-- 
cgit v1.2.3


From de0ff00d723fd821d372496e2c084805644aa5e1 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 27 Jun 2006 00:06:09 -0700
Subject: Driver core: add groups support to struct device

This is needed for the network class devices in order to be able to
convert over to use struct device.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 34 ++++++++++++++++++++++++++++++++++
 include/linux/device.h |  1 +
 2 files changed, 35 insertions(+)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 5d4b7e04471e..641c0c42bb48 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -197,6 +197,35 @@ static ssize_t store_uevent(struct device *dev, struct device_attribute *attr,
 	return count;
 }
 
+static int device_add_groups(struct device *dev)
+{
+	int i;
+	int error = 0;
+
+	if (dev->groups) {
+		for (i = 0; dev->groups[i]; i++) {
+			error = sysfs_create_group(&dev->kobj, dev->groups[i]);
+			if (error) {
+				while (--i >= 0)
+					sysfs_remove_group(&dev->kobj, dev->groups[i]);
+				goto out;
+			}
+		}
+	}
+out:
+	return error;
+}
+
+static void device_remove_groups(struct device *dev)
+{
+	int i;
+	if (dev->groups) {
+		for (i = 0; dev->groups[i]; i++) {
+			sysfs_remove_group(&dev->kobj, dev->groups[i]);
+		}
+	}
+}
+
 static ssize_t show_dev(struct device *dev, struct device_attribute *attr,
 			char *buf)
 {
@@ -350,6 +379,8 @@ int device_add(struct device *dev)
 		sysfs_create_link(&dev->parent->kobj, &dev->kobj, class_name);
 	}
 
+	if ((error = device_add_groups(dev)))
+		goto GroupError;
 	if ((error = device_pm_add(dev)))
 		goto PMError;
 	if ((error = bus_add_device(dev)))
@@ -376,6 +407,8 @@ int device_add(struct device *dev)
  BusError:
 	device_pm_remove(dev);
  PMError:
+	device_remove_groups(dev);
+ GroupError:
 	if (dev->devt_attr) {
 		device_remove_file(dev, dev->devt_attr);
 		kfree(dev->devt_attr);
@@ -470,6 +503,7 @@ void device_del(struct device * dev)
 		up(&dev->class->sem);
 	}
 	device_remove_file(dev, &dev->uevent_attr);
+	device_remove_groups(dev);
 
 	/* Notify the platform of the removal, in case they
 	 * need to do anything...
diff --git a/include/linux/device.h b/include/linux/device.h
index b3646462d6dc..994d3ebd53f4 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -344,6 +344,7 @@ struct device {
 	struct list_head	node;
 	struct class		*class;		/* optional*/
 	dev_t			devt;		/* dev_t, creates the sysfs "dev" */
+	struct attribute_group	**groups;	/* optional groups */
 
 	void	(*release)(struct device * dev);
 };
-- 
cgit v1.2.3


From 2620efef7029bb040430f50f0fc148f2d5e002ad Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 28 Jun 2006 16:19:58 -0700
Subject: Driver core: add ability for classes to handle devices properly

This adds two new callbacks to the class structure:
	int	(*dev_uevent)(struct device *dev, char **envp, int num_envp,
			char *buffer, int buffer_size);
	void	(*dev_release)(struct device *dev);

And one pointer:
	struct device_attribute		* dev_attrs;

which all corrispond with the same thing as the "normal" class devices
do, yet this is for when a struct device is bound to a class.

Someday soon, struct class_device will go away, and then the other
fields in this structure can be removed too.  But this is necessary in
order to get the transition to work properly.

Tested out on a network core patch that converted it to use struct
device instead of struct class_device.


Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/device.h |  4 ++++
 2 files changed, 57 insertions(+)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 5c91d0d81e78..f1228f25efe0 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -94,6 +94,8 @@ static void device_release(struct kobject * kobj)
 
 	if (dev->release)
 		dev->release(dev);
+	else if (dev->class && dev->class->dev_release)
+		dev->class->dev_release(dev);
 	else {
 		printk(KERN_ERR "Device '%s' does not have a release() function, "
 			"it is broken and must be fixed.\n",
@@ -183,6 +185,15 @@ static int dev_uevent(struct kset *kset, struct kobject *kobj, char **envp,
 		}
 	}
 
+	if (dev->class && dev->class->dev_uevent) {
+		/* have the class specific function add its stuff */
+		retval = dev->class->dev_uevent(dev, envp, num_envp, buffer, buffer_size);
+			if (retval) {
+				pr_debug("%s - dev_uevent() returned %d\n",
+					 __FUNCTION__, retval);
+		}
+	}
+
 	return retval;
 }
 
@@ -228,6 +239,43 @@ static void device_remove_groups(struct device *dev)
 	}
 }
 
+static int device_add_attrs(struct device *dev)
+{
+	struct class *class = dev->class;
+	int error = 0;
+	int i;
+
+	if (!class)
+		return 0;
+
+	if (class->dev_attrs) {
+		for (i = 0; attr_name(class->dev_attrs[i]); i++) {
+			error = device_create_file(dev, &class->dev_attrs[i]);
+			if (error)
+				break;
+		}
+	}
+	if (error)
+		while (--i >= 0)
+			device_remove_file(dev, &class->dev_attrs[i]);
+	return error;
+}
+
+static void device_remove_attrs(struct device *dev)
+{
+	struct class *class = dev->class;
+	int i;
+
+	if (!class)
+		return;
+
+	if (class->dev_attrs) {
+		for (i = 0; attr_name(class->dev_attrs[i]); i++)
+			device_remove_file(dev, &class->dev_attrs[i]);
+	}
+}
+
+
 static ssize_t show_dev(struct device *dev, struct device_attribute *attr,
 			char *buf)
 {
@@ -382,6 +430,8 @@ int device_add(struct device *dev)
 		}
 	}
 
+	if ((error = device_add_attrs(dev)))
+		goto AttrsError;
 	if ((error = device_add_groups(dev)))
 		goto GroupError;
 	if ((error = device_pm_add(dev)))
@@ -412,6 +462,8 @@ int device_add(struct device *dev)
  PMError:
 	device_remove_groups(dev);
  GroupError:
+ 	device_remove_attrs(dev);
+ AttrsError:
 	if (dev->devt_attr) {
 		device_remove_file(dev, dev->devt_attr);
 		kfree(dev->devt_attr);
@@ -509,6 +561,7 @@ void device_del(struct device * dev)
 	}
 	device_remove_file(dev, &dev->uevent_attr);
 	device_remove_groups(dev);
+	device_remove_attrs(dev);
 
 	/* Notify the platform of the removal, in case they
 	 * need to do anything...
diff --git a/include/linux/device.h b/include/linux/device.h
index 994d3ebd53f4..3122bd25ce42 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -151,12 +151,16 @@ struct class {
 
 	struct class_attribute		* class_attrs;
 	struct class_device_attribute	* class_dev_attrs;
+	struct device_attribute		* dev_attrs;
 
 	int	(*uevent)(struct class_device *dev, char **envp,
 			   int num_envp, char *buffer, int buffer_size);
+	int	(*dev_uevent)(struct device *dev, char **envp, int num_envp,
+				char *buffer, int buffer_size);
 
 	void	(*release)(struct class_device *dev);
 	void	(*class_release)(struct class *class);
+	void	(*dev_release)(struct device *dev);
 
 	int	(*suspend)(struct device *, pm_message_t state);
 	int	(*resume)(struct device *);
-- 
cgit v1.2.3


From a2de48cace5d0993da6cfa28b276ae724dc3569b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 3 Jul 2006 14:31:12 -0700
Subject: Driver core: add device_rename function

The network layer needs this to convert to using struct device instead
of a struct class_device.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/device.h |  1 +
 2 files changed, 56 insertions(+)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index f1228f25efe0..0db47561331e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -736,3 +736,58 @@ void device_destroy(struct class *class, dev_t devt)
 		device_unregister(dev);
 }
 EXPORT_SYMBOL_GPL(device_destroy);
+
+/**
+ * device_rename - renames a device
+ * @dev: the pointer to the struct device to be renamed
+ * @new_name: the new name of the device
+ */
+int device_rename(struct device *dev, char *new_name)
+{
+	char *old_class_name = NULL;
+	char *new_class_name = NULL;
+	char *old_symlink_name = NULL;
+	int error;
+
+	dev = get_device(dev);
+	if (!dev)
+		return -EINVAL;
+
+	pr_debug("DEVICE: renaming '%s' to '%s'\n", dev->bus_id, new_name);
+
+	if ((dev->class) && (dev->parent))
+		old_class_name = make_class_name(dev->class->name, &dev->kobj);
+
+	if (dev->class) {
+		old_symlink_name = kmalloc(BUS_ID_SIZE, GFP_KERNEL);
+		if (!old_symlink_name)
+			return -ENOMEM;
+		strlcpy(old_symlink_name, dev->bus_id, BUS_ID_SIZE);
+	}
+
+	strlcpy(dev->bus_id, new_name, BUS_ID_SIZE);
+
+	error = kobject_rename(&dev->kobj, new_name);
+
+	if (old_class_name) {
+		new_class_name = make_class_name(dev->class->name, &dev->kobj);
+		if (new_class_name) {
+			sysfs_create_link(&dev->parent->kobj, &dev->kobj,
+					  new_class_name);
+			sysfs_remove_link(&dev->parent->kobj, old_class_name);
+		}
+	}
+	if (dev->class) {
+		sysfs_remove_link(&dev->class->subsys.kset.kobj,
+				  old_symlink_name);
+		sysfs_create_link(&dev->class->subsys.kset.kobj, &dev->kobj,
+				  dev->bus_id);
+	}
+	put_device(dev);
+
+	kfree(old_class_name);
+	kfree(new_class_name);
+	kfree(old_symlink_name);
+
+	return error;
+}
diff --git a/include/linux/device.h b/include/linux/device.h
index 3122bd25ce42..3400e09bf458 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -380,6 +380,7 @@ extern int device_add(struct device * dev);
 extern void device_del(struct device * dev);
 extern int device_for_each_child(struct device *, void *,
 		     int (*fn)(struct device *, void *));
+extern int device_rename(struct device *dev, char *new_name);
 
 /*
  * Manual binding of a device to driver. See drivers/base/bus.c
-- 
cgit v1.2.3


From c205ef4880273d2de4ee5388d4e52227ff688cc4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 7 Aug 2006 22:19:37 -0700
Subject: Driver core: create devices/virtual/ tree

This change creates a devices/virtual/CLASS_NAME tree for struct devices
that belong to a class, yet do not have a "real" struct device for a
parent.  It automatically creates the directories on the fly as needed.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 17 +++++++++++++++++
 drivers/base/core.c    |  7 +++++++
 include/linux/device.h |  5 ++++-
 3 files changed, 28 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index e078bc21d52e..cbdf47c0c60d 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -19,6 +19,8 @@
 #include <linux/slab.h>
 #include "base.h"
 
+extern struct subsystem devices_subsys;
+
 #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr)
 #define to_class(obj) container_of(obj, struct class, subsys.kset.kobj)
 
@@ -878,7 +880,22 @@ void class_interface_unregister(struct class_interface *class_intf)
 	class_put(parent);
 }
 
+int virtual_device_parent(struct device *dev)
+{
+	if (!dev->class)
+		return -ENODEV;
+
+	if (!dev->class->virtual_dir) {
+		static struct kobject *virtual_dir = NULL;
 
+		if (!virtual_dir)
+			virtual_dir = kobject_add_dir(&devices_subsys.kset.kobj, "virtual");
+		dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name);
+	}
+
+	dev->kobj.parent = dev->class->virtual_dir;
+	return 0;
+}
 
 int __init classes_init(void)
 {
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 0db47561331e..e21a65fc043e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -378,6 +378,13 @@ int device_add(struct device *dev)
 	if (!dev || !strlen(dev->bus_id))
 		goto Error;
 
+	/* if this is a class device, and has no parent, create one */
+	if ((dev->class) && (dev->parent == NULL)) {
+		error = virtual_device_parent(dev);
+		if (error)
+			goto Error;
+	}
+
 	parent = get_device(dev->parent);
 
 	pr_debug("DEV: registering device: ID = '%s'\n", dev->bus_id);
diff --git a/include/linux/device.h b/include/linux/device.h
index 3400e09bf458..bbb0d6b5d232 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -149,6 +149,8 @@ struct class {
 	struct list_head	interfaces;
 	struct semaphore	sem;	/* locks both the children and interfaces lists */
 
+	struct kobject		*virtual_dir;
+
 	struct class_attribute		* class_attrs;
 	struct class_device_attribute	* class_dev_attrs;
 	struct device_attribute		* dev_attrs;
@@ -291,7 +293,6 @@ extern struct class_device *class_device_create(struct class *cls,
 					__attribute__((format(printf,5,6)));
 extern void class_device_destroy(struct class *cls, dev_t devt);
 
-
 /* interface for exporting device attributes */
 struct device_attribute {
 	struct attribute	attr;
@@ -400,6 +401,8 @@ extern struct device *device_create(struct class *cls, struct device *parent,
 				    __attribute__((format(printf,4,5)));
 extern void device_destroy(struct class *cls, dev_t devt);
 
+extern int virtual_device_parent(struct device *dev);
+
 /*
  * Platform "fixup" functions - allow the platform to have their say
  * about devices and actions that the general device layer doesn't
-- 
cgit v1.2.3


From c47ed219ba81632595e9f02e27318151fec16c9e Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 13 Sep 2006 15:34:05 +0200
Subject: Class: add support for class interfaces for devices

When moving class_device usage over to device, we need to handle
class_interfaces properly with devices.  This patch adds that support.


Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 10 ++++++++++
 drivers/base/core.c    | 14 +++++++++++++-
 include/linux/device.h |  2 ++
 3 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index cbdf47c0c60d..b06b0e2b9c62 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -842,6 +842,7 @@ int class_interface_register(struct class_interface *class_intf)
 {
 	struct class *parent;
 	struct class_device *class_dev;
+	struct device *dev;
 
 	if (!class_intf || !class_intf->class)
 		return -ENODEV;
@@ -856,6 +857,10 @@ int class_interface_register(struct class_interface *class_intf)
 		list_for_each_entry(class_dev, &parent->children, node)
 			class_intf->add(class_dev, class_intf);
 	}
+	if (class_intf->add_dev) {
+		list_for_each_entry(dev, &parent->devices, node)
+			class_intf->add_dev(dev, class_intf);
+	}
 	up(&parent->sem);
 
 	return 0;
@@ -865,6 +870,7 @@ void class_interface_unregister(struct class_interface *class_intf)
 {
 	struct class * parent = class_intf->class;
 	struct class_device *class_dev;
+	struct device *dev;
 
 	if (!parent)
 		return;
@@ -875,6 +881,10 @@ void class_interface_unregister(struct class_interface *class_intf)
 		list_for_each_entry(class_dev, &parent->children, node)
 			class_intf->remove(class_dev, class_intf);
 	}
+	if (class_intf->remove_dev) {
+		list_for_each_entry(dev, &parent->devices, node)
+			class_intf->remove_dev(dev, class_intf);
+	}
 	up(&parent->sem);
 
 	class_put(parent);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index e21a65fc043e..1d3d3582fcca 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -372,6 +372,7 @@ int device_add(struct device *dev)
 {
 	struct device *parent = NULL;
 	char *class_name = NULL;
+	struct class_interface *class_intf;
 	int error = -EINVAL;
 
 	dev = get_device(dev);
@@ -451,9 +452,14 @@ int device_add(struct device *dev)
 		klist_add_tail(&dev->knode_parent, &parent->klist_children);
 
 	if (dev->class) {
-		/* tie the class to the device */
 		down(&dev->class->sem);
+		/* tie the class to the device */
 		list_add_tail(&dev->node, &dev->class->devices);
+
+		/* notify any interfaces that the device is here */
+		list_for_each_entry(class_intf, &dev->class->interfaces, node)
+			if (class_intf->add_dev)
+				class_intf->add_dev(dev, class_intf);
 		up(&dev->class->sem);
 	}
 
@@ -548,6 +554,7 @@ void device_del(struct device * dev)
 {
 	struct device * parent = dev->parent;
 	char *class_name = NULL;
+	struct class_interface *class_intf;
 
 	if (parent)
 		klist_del(&dev->knode_parent);
@@ -563,6 +570,11 @@ void device_del(struct device * dev)
 		}
 		kfree(class_name);
 		down(&dev->class->sem);
+		/* notify any interfaces that the device is now gone */
+		list_for_each_entry(class_intf, &dev->class->interfaces, node)
+			if (class_intf->remove_dev)
+				class_intf->remove_dev(dev, class_intf);
+		/* remove the device from the class list */
 		list_del_init(&dev->node);
 		up(&dev->class->sem);
 	}
diff --git a/include/linux/device.h b/include/linux/device.h
index bbb0d6b5d232..e0fae0e76fa9 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -278,6 +278,8 @@ struct class_interface {
 
 	int (*add)	(struct class_device *, struct class_interface *);
 	void (*remove)	(struct class_device *, struct class_interface *);
+	int (*add_dev)		(struct device *, struct class_interface *);
+	void (*remove_dev)	(struct device *, struct class_interface *);
 };
 
 extern int class_interface_register(struct class_interface *);
-- 
cgit v1.2.3


From 2589f1887b0bf9f08ec3d7f3c5705ccb7c628076 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 19 Sep 2006 09:39:19 -0700
Subject: Driver core: add ability for devices to create and remove bin files

Makes it easier for devices to create and remove binary attribute files
so they don't have to call directly into sysfs.  This is needed to help
with the conversion from struct class_device to struct device.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 26 ++++++++++++++++++++++++++
 include/linux/device.h |  4 ++++
 2 files changed, 30 insertions(+)

(limited to 'include')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 1d3d3582fcca..bc9f35c81691 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -319,6 +319,32 @@ void device_remove_file(struct device * dev, struct device_attribute * attr)
 	}
 }
 
+/**
+ * device_create_bin_file - create sysfs binary attribute file for device.
+ * @dev: device.
+ * @attr: device binary attribute descriptor.
+ */
+int device_create_bin_file(struct device *dev, struct bin_attribute *attr)
+{
+	int error = -EINVAL;
+	if (dev)
+		error = sysfs_create_bin_file(&dev->kobj, attr);
+	return error;
+}
+EXPORT_SYMBOL_GPL(device_create_bin_file);
+
+/**
+ * device_remove_bin_file - remove sysfs binary attribute file
+ * @dev: device.
+ * @attr: device binary attribute descriptor.
+ */
+void device_remove_bin_file(struct device *dev, struct bin_attribute *attr)
+{
+	if (dev)
+		sysfs_remove_bin_file(&dev->kobj, attr);
+}
+EXPORT_SYMBOL_GPL(device_remove_bin_file);
+
 static void klist_children_get(struct klist_node *n)
 {
 	struct device *dev = container_of(n, struct device, knode_parent);
diff --git a/include/linux/device.h b/include/linux/device.h
index e0fae0e76fa9..7d447d7271c5 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -309,6 +309,10 @@ struct device_attribute dev_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
 extern int device_create_file(struct device *device, struct device_attribute * entry);
 extern void device_remove_file(struct device * dev, struct device_attribute * attr);
+extern int __must_check device_create_bin_file(struct device *dev,
+					       struct bin_attribute *attr);
+extern void device_remove_bin_file(struct device *dev,
+				   struct bin_attribute *attr);
 struct device {
 	struct klist		klist_children;
 	struct klist_node	knode_parent;		/* node in sibling list */
-- 
cgit v1.2.3


From 995982ca79d9262869513948ec7c540f32035491 Mon Sep 17 00:00:00 2001
From: "Randy.Dunlap" <rdunlap@xenotime.net>
Date: Mon, 10 Jul 2006 23:05:25 -0700
Subject: sysfs_remove_bin_file: no return value, dump_stack on error

Make sysfs_remove_bin_file() void.  If it detects an error,
printk the file name and call dump_stack().

sysfs_hash_and_remove() now returns an error code indicating
its success or failure so that sysfs_remove_bin_file() can
know success/failure.

Convert the only driver that checked the return value of
sysfs_remove_bin_file().

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_ibm.c |  4 +---
 fs/sysfs/bin.c                    | 13 ++++++++-----
 fs/sysfs/inode.c                  | 11 ++++++++---
 fs/sysfs/sysfs.h                  |  2 +-
 include/linux/sysfs.h             |  2 +-
 5 files changed, 19 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c
index 317457dd4014..d0a07d9ab30c 100644
--- a/drivers/pci/hotplug/acpiphp_ibm.c
+++ b/drivers/pci/hotplug/acpiphp_ibm.c
@@ -487,9 +487,7 @@ static void __exit ibm_acpiphp_exit(void)
 	if (ACPI_FAILURE(status))
 		err("%s: Notification handler removal failed\n", __FUNCTION__);
 	/* remove the /sys entries */
-	if (sysfs_remove_bin_file(sysdir, &ibm_apci_table_attr))
-		err("%s: removal of sysfs file apci_table failed\n",
-				__FUNCTION__);
+	sysfs_remove_bin_file(sysdir, &ibm_apci_table_attr);
 }
 
 module_init(ibm_acpiphp_init);
diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c
index c16a93c353c0..98022e41cda1 100644
--- a/fs/sysfs/bin.c
+++ b/fs/sysfs/bin.c
@@ -10,6 +10,7 @@
 
 #include <linux/errno.h>
 #include <linux/fs.h>
+#include <linux/kernel.h>
 #include <linux/kobject.h>
 #include <linux/module.h>
 #include <linux/slab.h>
@@ -176,7 +177,6 @@ const struct file_operations bin_fops = {
  *	sysfs_create_bin_file - create binary file for object.
  *	@kobj:	object.
  *	@attr:	attribute descriptor.
- *
  */
 
 int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
@@ -191,13 +191,16 @@ int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
  *	sysfs_remove_bin_file - remove binary file for object.
  *	@kobj:	object.
  *	@attr:	attribute descriptor.
- *
  */
 
-int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr)
+void sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr)
 {
-	sysfs_hash_and_remove(kobj->dentry,attr->attr.name);
-	return 0;
+	if (sysfs_hash_and_remove(kobj->dentry, attr->attr.name) < 0) {
+		printk(KERN_ERR "%s: "
+			"bad dentry or inode or no such file: \"%s\"\n",
+			__FUNCTION__, attr->attr.name);
+		dump_stack();
+	}
 }
 
 EXPORT_SYMBOL_GPL(sysfs_create_bin_file);
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 9889e54e1f13..fd7cd5f843d2 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -12,6 +12,7 @@
 #include <linux/namei.h>
 #include <linux/backing-dev.h>
 #include <linux/capability.h>
+#include <linux/errno.h>
 #include "sysfs.h"
 
 extern struct super_block * sysfs_sb;
@@ -234,17 +235,18 @@ void sysfs_drop_dentry(struct sysfs_dirent * sd, struct dentry * parent)
 	}
 }
 
-void sysfs_hash_and_remove(struct dentry * dir, const char * name)
+int sysfs_hash_and_remove(struct dentry * dir, const char * name)
 {
 	struct sysfs_dirent * sd;
 	struct sysfs_dirent * parent_sd;
+	int found = 0;
 
 	if (!dir)
-		return;
+		return -ENOENT;
 
 	if (dir->d_inode == NULL)
 		/* no inode means this hasn't been made visible yet */
-		return;
+		return -ENOENT;
 
 	parent_sd = dir->d_fsdata;
 	mutex_lock(&dir->d_inode->i_mutex);
@@ -255,8 +257,11 @@ void sysfs_hash_and_remove(struct dentry * dir, const char * name)
 			list_del_init(&sd->s_sibling);
 			sysfs_drop_dentry(sd, dir);
 			sysfs_put(sd);
+			found = 1;
 			break;
 		}
 	}
 	mutex_unlock(&dir->d_inode->i_mutex);
+
+	return found ? 0 : -ENOENT;
 }
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index 3651ffb5ec09..6f3d6bd52887 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -10,7 +10,7 @@ extern int sysfs_make_dirent(struct sysfs_dirent *, struct dentry *, void *,
 				umode_t, int);
 
 extern int sysfs_add_file(struct dentry *, const struct attribute *, int);
-extern void sysfs_hash_and_remove(struct dentry * dir, const char * name);
+extern int sysfs_hash_and_remove(struct dentry * dir, const char * name);
 extern struct sysfs_dirent *sysfs_find(struct sysfs_dirent *dir, const char * name);
 
 extern int sysfs_create_subdir(struct kobject *, const char *, struct dentry **);
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 1ea5d3cda6ae..95f6db54d8d2 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -114,7 +114,7 @@ extern void
 sysfs_remove_link(struct kobject *, const char * name);
 
 int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
-int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr);
+void sysfs_remove_bin_file(struct kobject *kobj, struct bin_attribute *attr);
 
 int sysfs_create_group(struct kobject *, const struct attribute_group *);
 void sysfs_remove_group(struct kobject *, const struct attribute_group *);
-- 
cgit v1.2.3


From 4a7fb6363f2d1a6c09a10253937f672f3c7929e1 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:17 -0700
Subject: add __must_check to device management code

We're getting a lot of crashes in the sysfs/kobject/device/bus/class code and
they're very hard to diagnose.

I'm suspecting that in some cases this is because drivers aren't checking
return values and aren't handling errors correctly.  So the code blithely
blunders on and crashes later in very obscure ways.

There's just no reason to ignore errors which can and do occur.  So the patch
sprinkles __must_check all over these APIs.

Causes 1,513 new warnings.  Heh.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/device.h  | 52 ++++++++++++++++++++++++++-----------------------
 include/linux/kobject.h | 16 ++++++++-------
 include/linux/pci.h     | 34 +++++++++++++++++---------------
 include/linux/sysfs.h   | 19 ++++++++++--------
 4 files changed, 66 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/include/linux/device.h b/include/linux/device.h
index 7d447d7271c5..ad4db72f5733 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -15,6 +15,7 @@
 #include <linux/kobject.h>
 #include <linux/klist.h>
 #include <linux/list.h>
+#include <linux/compiler.h>
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pm.h>
@@ -58,7 +59,7 @@ struct bus_type {
 	int (*resume)(struct device * dev);
 };
 
-extern int bus_register(struct bus_type * bus);
+extern int __must_check bus_register(struct bus_type * bus);
 extern void bus_unregister(struct bus_type * bus);
 
 extern void bus_rescan_devices(struct bus_type * bus);
@@ -70,9 +71,9 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data,
 struct device * bus_find_device(struct bus_type *bus, struct device *start,
 				void *data, int (*match)(struct device *, void *));
 
-int bus_for_each_drv(struct bus_type * bus, struct device_driver * start, 
-		     void * data, int (*fn)(struct device_driver *, void *));
-
+int __must_check bus_for_each_drv(struct bus_type *bus,
+		struct device_driver *start, void *data,
+		int (*fn)(struct device_driver *, void *));
 
 /* driverfs interface for exporting bus attributes */
 
@@ -85,7 +86,8 @@ struct bus_attribute {
 #define BUS_ATTR(_name,_mode,_show,_store)	\
 struct bus_attribute bus_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
-extern int bus_create_file(struct bus_type *, struct bus_attribute *);
+extern int __must_check bus_create_file(struct bus_type *,
+					struct bus_attribute *);
 extern void bus_remove_file(struct bus_type *, struct bus_attribute *);
 
 struct device_driver {
@@ -107,14 +109,13 @@ struct device_driver {
 };
 
 
-extern int driver_register(struct device_driver * drv);
+extern int __must_check driver_register(struct device_driver * drv);
 extern void driver_unregister(struct device_driver * drv);
 
 extern struct device_driver * get_driver(struct device_driver * drv);
 extern void put_driver(struct device_driver * drv);
 extern struct device_driver *driver_find(const char *name, struct bus_type *bus);
 
-
 /* driverfs interface for exporting driver attributes */
 
 struct driver_attribute {
@@ -126,16 +127,17 @@ struct driver_attribute {
 #define DRIVER_ATTR(_name,_mode,_show,_store)	\
 struct driver_attribute driver_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
-extern int driver_create_file(struct device_driver *, struct driver_attribute *);
+extern int __must_check driver_create_file(struct device_driver *,
+					struct driver_attribute *);
 extern void driver_remove_file(struct device_driver *, struct driver_attribute *);
 
-extern int driver_for_each_device(struct device_driver * drv, struct device * start,
-				  void * data, int (*fn)(struct device *, void *));
+extern int __must_check driver_for_each_device(struct device_driver * drv,
+		struct device *start, void *data,
+		int (*fn)(struct device *, void *));
 struct device * driver_find_device(struct device_driver *drv,
 				   struct device *start, void *data,
 				   int (*match)(struct device *, void *));
 
-
 /*
  * device classes
  */
@@ -168,7 +170,7 @@ struct class {
 	int	(*resume)(struct device *);
 };
 
-extern int class_register(struct class *);
+extern int __must_check class_register(struct class *);
 extern void class_unregister(struct class *);
 
 
@@ -181,7 +183,8 @@ struct class_attribute {
 #define CLASS_ATTR(_name,_mode,_show,_store)			\
 struct class_attribute class_attr_##_name = __ATTR(_name,_mode,_show,_store) 
 
-extern int class_create_file(struct class *, const struct class_attribute *);
+extern int __must_check class_create_file(struct class *,
+					const struct class_attribute *);
 extern void class_remove_file(struct class *, const struct class_attribute *);
 
 struct class_device_attribute {
@@ -194,7 +197,7 @@ struct class_device_attribute {
 struct class_device_attribute class_device_attr_##_name = 	\
 	__ATTR(_name,_mode,_show,_store)
 
-extern int class_device_create_file(struct class_device *,
+extern int __must_check class_device_create_file(struct class_device *,
 				    const struct class_device_attribute *);
 
 /**
@@ -254,10 +257,10 @@ class_set_devdata (struct class_device *dev, void *data)
 }
 
 
-extern int class_device_register(struct class_device *);
+extern int __must_check class_device_register(struct class_device *);
 extern void class_device_unregister(struct class_device *);
 extern void class_device_initialize(struct class_device *);
-extern int class_device_add(struct class_device *);
+extern int __must_check class_device_add(struct class_device *);
 extern void class_device_del(struct class_device *);
 
 extern int class_device_rename(struct class_device *, char *);
@@ -267,7 +270,7 @@ extern void class_device_put(struct class_device *);
 
 extern void class_device_remove_file(struct class_device *, 
 				     const struct class_device_attribute *);
-extern int class_device_create_bin_file(struct class_device *,
+extern int __must_check class_device_create_bin_file(struct class_device *,
 					struct bin_attribute *);
 extern void class_device_remove_bin_file(struct class_device *,
 					 struct bin_attribute *);
@@ -282,7 +285,7 @@ struct class_interface {
 	void (*remove_dev)	(struct device *, struct class_interface *);
 };
 
-extern int class_interface_register(struct class_interface *);
+extern int __must_check class_interface_register(struct class_interface *);
 extern void class_interface_unregister(struct class_interface *);
 
 extern struct class *class_create(struct module *owner, const char *name);
@@ -307,7 +310,8 @@ struct device_attribute {
 #define DEVICE_ATTR(_name,_mode,_show,_store) \
 struct device_attribute dev_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
-extern int device_create_file(struct device *device, struct device_attribute * entry);
+extern int __must_check device_create_file(struct device *device,
+					struct device_attribute * entry);
 extern void device_remove_file(struct device * dev, struct device_attribute * attr);
 extern int __must_check device_create_bin_file(struct device *dev,
 					       struct bin_attribute *attr);
@@ -380,12 +384,12 @@ static inline int device_is_registered(struct device *dev)
 /*
  * High level routines for use by the bus drivers
  */
-extern int device_register(struct device * dev);
+extern int __must_check device_register(struct device * dev);
 extern void device_unregister(struct device * dev);
 extern void device_initialize(struct device * dev);
-extern int device_add(struct device * dev);
+extern int __must_check device_add(struct device * dev);
 extern void device_del(struct device * dev);
-extern int device_for_each_child(struct device *, void *,
+extern int __must_check device_for_each_child(struct device *, void *,
 		     int (*fn)(struct device *, void *));
 extern int device_rename(struct device *dev, char *new_name);
 
@@ -395,7 +399,7 @@ extern int device_rename(struct device *dev, char *new_name);
  */
 extern void device_bind_driver(struct device * dev);
 extern void device_release_driver(struct device * dev);
-extern int  device_attach(struct device * dev);
+extern int  __must_check device_attach(struct device * dev);
 extern void driver_attach(struct device_driver * drv);
 extern void device_reprobe(struct device *dev);
 
@@ -433,7 +437,7 @@ extern void device_shutdown(void);
 
 
 /* drivers/base/firmware.c */
-extern int firmware_register(struct subsystem *);
+extern int __must_check firmware_register(struct subsystem *);
 extern void firmware_unregister(struct subsystem *);
 
 /* debugging and troubleshooting/diagnostic helpers. */
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 2d229327959e..bcd9cd173c2c 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -20,6 +20,7 @@
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/sysfs.h>
+#include <linux/compiler.h>
 #include <linux/spinlock.h>
 #include <linux/rwsem.h>
 #include <linux/kref.h>
@@ -71,12 +72,12 @@ static inline const char * kobject_name(const struct kobject * kobj)
 extern void kobject_init(struct kobject *);
 extern void kobject_cleanup(struct kobject *);
 
-extern int kobject_add(struct kobject *);
+extern int __must_check kobject_add(struct kobject *);
 extern void kobject_del(struct kobject *);
 
-extern int kobject_rename(struct kobject *, const char *new_name);
+extern int __must_check kobject_rename(struct kobject *, const char *new_name);
 
-extern int kobject_register(struct kobject *);
+extern int __must_check kobject_register(struct kobject *);
 extern void kobject_unregister(struct kobject *);
 
 extern struct kobject * kobject_get(struct kobject *);
@@ -128,8 +129,8 @@ struct kset {
 
 
 extern void kset_init(struct kset * k);
-extern int kset_add(struct kset * k);
-extern int kset_register(struct kset * k);
+extern int __must_check kset_add(struct kset * k);
+extern int __must_check kset_register(struct kset * k);
 extern void kset_unregister(struct kset * k);
 
 static inline struct kset * to_kset(struct kobject * kobj)
@@ -239,7 +240,7 @@ extern struct subsystem hypervisor_subsys;
 	(obj)->subsys.kset.kobj.kset = &(_subsys).kset
 
 extern void subsystem_init(struct subsystem *);
-extern int subsystem_register(struct subsystem *);
+extern int __must_check subsystem_register(struct subsystem *);
 extern void subsystem_unregister(struct subsystem *);
 
 static inline struct subsystem * subsys_get(struct subsystem * s)
@@ -258,7 +259,8 @@ struct subsys_attribute {
 	ssize_t (*store)(struct subsystem *, const char *, size_t); 
 };
 
-extern int subsys_create_file(struct subsystem * , struct subsys_attribute *);
+extern int __must_check subsys_create_file(struct subsystem * ,
+					struct subsys_attribute *);
 
 #if defined(CONFIG_HOTPLUG)
 void kobject_uevent(struct kobject *kobj, enum kobject_action action);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 9514bbfe96e2..3ec72551ac31 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -49,6 +49,7 @@
 #include <linux/types.h>
 #include <linux/ioport.h>
 #include <linux/list.h>
+#include <linux/compiler.h>
 #include <linux/errno.h>
 #include <linux/device.h>
 
@@ -403,7 +404,7 @@ extern struct list_head pci_root_buses;	/* list of all known PCI buses */
 extern struct list_head pci_devices;	/* list of all devices */
 
 void pcibios_fixup_bus(struct pci_bus *);
-int pcibios_enable_device(struct pci_dev *, int mask);
+int __must_check pcibios_enable_device(struct pci_dev *, int mask);
 char *pcibios_setup (char *str);
 
 /* Used only when drivers/pci/setup.c is used */
@@ -490,19 +491,19 @@ static inline int pci_write_config_dword(struct pci_dev *dev, int where, u32 val
 	return pci_bus_write_config_dword (dev->bus, dev->devfn, where, val);
 }
 
-int pci_enable_device(struct pci_dev *dev);
-int pci_enable_device_bars(struct pci_dev *dev, int mask);
+int __must_check pci_enable_device(struct pci_dev *dev);
+int __must_check pci_enable_device_bars(struct pci_dev *dev, int mask);
 void pci_disable_device(struct pci_dev *dev);
 void pci_set_master(struct pci_dev *dev);
 #define HAVE_PCI_SET_MWI
-int pci_set_mwi(struct pci_dev *dev);
+int __must_check pci_set_mwi(struct pci_dev *dev);
 void pci_clear_mwi(struct pci_dev *dev);
 void pci_intx(struct pci_dev *dev, int enable);
 int pci_set_dma_mask(struct pci_dev *dev, u64 mask);
 int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask);
 void pci_update_resource(struct pci_dev *dev, struct resource *res, int resno);
-int pci_assign_resource(struct pci_dev *dev, int i);
-int pci_assign_resource_fixed(struct pci_dev *dev, int i);
+int __must_check pci_assign_resource(struct pci_dev *dev, int i);
+int __must_check pci_assign_resource_fixed(struct pci_dev *dev, int i);
 void pci_restore_bars(struct pci_dev *dev);
 
 /* ROM control related routines */
@@ -528,23 +529,24 @@ void pdev_sort_resources(struct pci_dev *, struct resource_list *);
 void pci_fixup_irqs(u8 (*)(struct pci_dev *, u8 *),
 		    int (*)(struct pci_dev *, u8, u8));
 #define HAVE_PCI_REQ_REGIONS	2
-int pci_request_regions(struct pci_dev *, const char *);
+int __must_check pci_request_regions(struct pci_dev *, const char *);
 void pci_release_regions(struct pci_dev *);
-int pci_request_region(struct pci_dev *, int, const char *);
+int __must_check pci_request_region(struct pci_dev *, int, const char *);
 void pci_release_region(struct pci_dev *, int);
 
 /* drivers/pci/bus.c */
-int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
-			   resource_size_t size, resource_size_t align,
-			   resource_size_t min, unsigned int type_mask,
-			   void (*alignf)(void *, struct resource *,
-					  resource_size_t, resource_size_t),
-			   void *alignf_data);
+int __must_check pci_bus_alloc_resource(struct pci_bus *bus,
+			struct resource *res, resource_size_t size,
+			resource_size_t align, resource_size_t min,
+			unsigned int type_mask,
+			void (*alignf)(void *, struct resource *,
+				resource_size_t, resource_size_t),
+			void *alignf_data);
 void pci_enable_bridges(struct pci_bus *bus);
 
 /* Proper probing supporting hot-pluggable devices */
-int __pci_register_driver(struct pci_driver *, struct module *);
-static inline int pci_register_driver(struct pci_driver *driver)
+int __must_check __pci_register_driver(struct pci_driver *, struct module *);
+static inline int __must_check pci_register_driver(struct pci_driver *driver)
 {
 	return __pci_register_driver(driver, THIS_MODULE);
 }
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 95f6db54d8d2..02ad790921fd 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -10,6 +10,7 @@
 #ifndef _SYSFS_H_
 #define _SYSFS_H_
 
+#include <linux/compiler.h>
 #include <asm/atomic.h>
 
 struct kobject;
@@ -86,37 +87,39 @@ struct sysfs_dirent {
 
 #ifdef CONFIG_SYSFS
 
-extern int
+extern int __must_check
 sysfs_create_dir(struct kobject *);
 
 extern void
 sysfs_remove_dir(struct kobject *);
 
-extern int
+extern int __must_check
 sysfs_rename_dir(struct kobject *, const char *new_name);
 
-extern int
+extern int __must_check
 sysfs_create_file(struct kobject *, const struct attribute *);
 
-extern int
+extern int __must_check
 sysfs_update_file(struct kobject *, const struct attribute *);
 
-extern int
+extern int __must_check
 sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode);
 
 extern void
 sysfs_remove_file(struct kobject *, const struct attribute *);
 
-extern int
+extern int __must_check
 sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name);
 
 extern void
 sysfs_remove_link(struct kobject *, const char * name);
 
-int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
+int __must_check sysfs_create_bin_file(struct kobject *kobj,
+					struct bin_attribute *attr);
 void sysfs_remove_bin_file(struct kobject *kobj, struct bin_attribute *attr);
 
-int sysfs_create_group(struct kobject *, const struct attribute_group *);
+int __must_check sysfs_create_group(struct kobject *,
+					const struct attribute_group *);
 void sysfs_remove_group(struct kobject *, const struct attribute_group *);
 void sysfs_notify(struct kobject * k, char *dir, char *attr);
 
-- 
cgit v1.2.3


From cebc04ba9aeb3a646cc746300421fc0e5aa4f253 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:18 -0700
Subject: add CONFIG_ENABLE_MUST_CHECK

Those 1500 warnings can be a bit of a pain.  Add a config option to shut them
up.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/compiler.h | 5 +++++
 lib/Kconfig.debug        | 7 +++++++
 2 files changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 9b4f11094937..060b96112ec6 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -99,6 +99,11 @@ extern void __chk_io_ptr(void __iomem *);
 #define __must_check
 #endif
 
+#ifndef CONFIG_ENABLE_MUST_CHECK
+#undef __must_check
+#define __must_check
+#endif
+
 /*
  * Allow us to avoid 'defined but not used' warnings on functions and data,
  * as well as force them to be emitted to the assembly file.
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 554ee688a9f8..5c114c3f03c5 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -8,6 +8,13 @@ config PRINTK_TIME
 	  operations.  This is useful for identifying long delays
 	  in kernel startup.
 
+config ENABLE_MUST_CHECK
+	bool "Enable __must_check logic"
+	default y
+	help
+	  Enable the __must_check logic in the kernel build.  Disable this to
+	  suppress the "warning: ignoring return value of 'foo', declared with
+	  attribute warn_unused_result" messages.
 
 config MAGIC_SYSRQ
 	bool "Magic SysRq key"
-- 
cgit v1.2.3


From 8a6914ab7c95d471c23b42268aa8e1f55b3d2fdb Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:19 -0700
Subject: v4l-dev2: handle __must_check

We get hundreds of these:

include/media/v4l2-dev.h:348: warning: ignoring return value of 'class_device_create_file', declared with attribute warn_unused_result

Handle it, and propagate the __must_check back a level.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/media/v4l2-dev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/media/v4l2-dev.h b/include/media/v4l2-dev.h
index 810462f8a374..bb495b7f4680 100644
--- a/include/media/v4l2-dev.h
+++ b/include/media/v4l2-dev.h
@@ -341,7 +341,7 @@ extern int video_usercopy(struct inode *inode, struct file *file,
 extern struct video_device* video_devdata(struct file*);
 
 #define to_video_device(cd) container_of(cd, struct video_device, class_dev)
-static inline int
+static inline int __must_check
 video_device_create_file(struct video_device *vfd,
 			 struct class_device_attribute *attr)
 {
-- 
cgit v1.2.3


From f86db396ff455ed586751d21816a1ebd431264e5 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:20 -0700
Subject: drivers/base: check errors

Add lots of return-value checking.

<pcornelia.huck@de.ibm.com>: fix bus_rescan_devices()]
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |   2 +-
 drivers/base/bus.c     | 107 +++++++++++++++++++++++++++++++++----------------
 drivers/base/dd.c      |  37 ++++++++++++-----
 include/linux/device.h |   8 ++--
 4 files changed, 104 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index c3b8dc98b8a7..d26644a59537 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -16,7 +16,7 @@ extern int cpu_dev_init(void);
 extern int attribute_container_init(void);
 
 extern int bus_add_device(struct device * dev);
-extern void bus_attach_device(struct device * dev);
+extern int bus_attach_device(struct device * dev);
 extern void bus_remove_device(struct device * dev);
 extern struct bus_type *get_bus(struct bus_type * bus);
 extern void put_bus(struct bus_type * bus);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 4d22a1d10a1c..aa685a20b649 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -371,12 +371,20 @@ int bus_add_device(struct device * dev)
 	if (bus) {
 		pr_debug("bus %s: add device %s\n", bus->name, dev->bus_id);
 		error = device_add_attrs(bus, dev);
-		if (!error) {
-			sysfs_create_link(&bus->devices.kobj, &dev->kobj, dev->bus_id);
-			sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "subsystem");
-			sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "bus");
-		}
+		if (error)
+			goto out;
+		error = sysfs_create_link(&bus->devices.kobj,
+						&dev->kobj, dev->bus_id);
+		if (error)
+			goto out;
+		error = sysfs_create_link(&dev->kobj,
+				&dev->bus->subsys.kset.kobj, "subsystem");
+		if (error)
+			goto out;
+		error = sysfs_create_link(&dev->kobj,
+				&dev->bus->subsys.kset.kobj, "bus");
 	}
+out:
 	return error;
 }
 
@@ -386,14 +394,19 @@ int bus_add_device(struct device * dev)
  *
  *	- Try to attach to driver.
  */
-void bus_attach_device(struct device * dev)
+int bus_attach_device(struct device * dev)
 {
-	struct bus_type * bus = dev->bus;
+	struct bus_type *bus = dev->bus;
+	int ret = 0;
 
 	if (bus) {
-		device_attach(dev);
-		klist_add_tail(&dev->knode_bus, &bus->klist_devices);
+		ret = device_attach(dev);
+		if (ret >= 0) {
+			klist_add_tail(&dev->knode_bus, &bus->klist_devices);
+			ret = 0;
+		}
 	}
+	return ret;
 }
 
 /**
@@ -455,10 +468,17 @@ static void driver_remove_attrs(struct bus_type * bus, struct device_driver * dr
  * Thanks to drivers making their tables __devinit, we can't allow manual
  * bind and unbind from userspace unless CONFIG_HOTPLUG is enabled.
  */
-static void add_bind_files(struct device_driver *drv)
+static int __must_check add_bind_files(struct device_driver *drv)
 {
-	driver_create_file(drv, &driver_attr_unbind);
-	driver_create_file(drv, &driver_attr_bind);
+	int ret;
+
+	ret = driver_create_file(drv, &driver_attr_unbind);
+	if (ret == 0) {
+		ret = driver_create_file(drv, &driver_attr_bind);
+		if (ret)
+			driver_remove_file(drv, &driver_attr_unbind);
+	}
+	return ret;
 }
 
 static void remove_bind_files(struct device_driver *drv)
@@ -476,7 +496,7 @@ static inline void remove_bind_files(struct device_driver *drv) {}
  *	@drv:	driver.
  *
  */
-int bus_add_driver(struct device_driver * drv)
+int bus_add_driver(struct device_driver *drv)
 {
 	struct bus_type * bus = get_bus(drv->bus);
 	int error = 0;
@@ -484,27 +504,39 @@ int bus_add_driver(struct device_driver * drv)
 	if (bus) {
 		pr_debug("bus %s: add driver %s\n", bus->name, drv->name);
 		error = kobject_set_name(&drv->kobj, "%s", drv->name);
-		if (error) {
-			put_bus(bus);
-			return error;
-		}
+		if (error)
+			goto out_put_bus;
 		drv->kobj.kset = &bus->drivers;
-		if ((error = kobject_register(&drv->kobj))) {
-			put_bus(bus);
-			return error;
-		}
+		if ((error = kobject_register(&drv->kobj)))
+			goto out_put_bus;
 
-		driver_attach(drv);
+		error = driver_attach(drv);
+		if (error)
+			goto out_unregister;
 		klist_add_tail(&drv->knode_bus, &bus->klist_drivers);
 		module_add_driver(drv->owner, drv);
 
-		driver_add_attrs(bus, drv);
-		add_bind_files(drv);
+		error = driver_add_attrs(bus, drv);
+		if (error) {
+			/* How the hell do we get out of this pickle? Give up */
+			printk(KERN_ERR "%s: driver_add_attrs(%s) failed\n",
+				__FUNCTION__, drv->name);
+		}
+		error = add_bind_files(drv);
+		if (error) {
+			/* Ditto */
+			printk(KERN_ERR "%s: add_bind_files(%s) failed\n",
+				__FUNCTION__, drv->name);
+		}
 	}
 	return error;
+out_unregister:
+	kobject_unregister(&drv->kobj);
+out_put_bus:
+	put_bus(bus);
+	return error;
 }
 
-
 /**
  *	bus_remove_driver - delete driver from bus's knowledge.
  *	@drv:	driver.
@@ -530,16 +562,21 @@ void bus_remove_driver(struct device_driver * drv)
 
 
 /* Helper for bus_rescan_devices's iter */
-static int bus_rescan_devices_helper(struct device *dev, void *data)
+static int __must_check bus_rescan_devices_helper(struct device *dev,
+						void *data)
 {
+	int ret = 0;
+
 	if (!dev->driver) {
 		if (dev->parent)	/* Needed for USB */
 			down(&dev->parent->sem);
-		device_attach(dev);
+		ret = device_attach(dev);
 		if (dev->parent)
 			up(&dev->parent->sem);
+		if (ret > 0)
+			ret = 0;
 	}
-	return 0;
+	return ret < 0 ? ret : 0;
 }
 
 /**
@@ -550,9 +587,9 @@ static int bus_rescan_devices_helper(struct device *dev, void *data)
  * attached and rescan it against existing drivers to see if it matches
  * any by calling device_attach() for the unbound devices.
  */
-void bus_rescan_devices(struct bus_type * bus)
+int bus_rescan_devices(struct bus_type * bus)
 {
-	bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper);
+	return bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper);
 }
 
 /**
@@ -564,7 +601,7 @@ void bus_rescan_devices(struct bus_type * bus)
  * to use if probing criteria changed during a devices lifetime and
  * driver attachment should change accordingly.
  */
-void device_reprobe(struct device *dev)
+int device_reprobe(struct device *dev)
 {
 	if (dev->driver) {
 		if (dev->parent)        /* Needed for USB */
@@ -573,14 +610,14 @@ void device_reprobe(struct device *dev)
 		if (dev->parent)
 			up(&dev->parent->sem);
 	}
-
-	bus_rescan_devices_helper(dev, NULL);
+	return bus_rescan_devices_helper(dev, NULL);
 }
 EXPORT_SYMBOL_GPL(device_reprobe);
 
-struct bus_type * get_bus(struct bus_type * bus)
+struct bus_type *get_bus(struct bus_type *bus)
 {
-	return bus ? container_of(subsys_get(&bus->subsys), struct bus_type, subsys) : NULL;
+	return bus ? container_of(subsys_get(&bus->subsys),
+				struct bus_type, subsys) : NULL;
 }
 
 void put_bus(struct bus_type * bus)
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 889c71111239..9f6f11ca0ab6 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -38,17 +38,29 @@
  *
  *	This function must be called with @dev->sem held.
  */
-void device_bind_driver(struct device * dev)
+int device_bind_driver(struct device *dev)
 {
-	if (klist_node_attached(&dev->knode_driver))
-		return;
+	int ret;
+
+	if (klist_node_attached(&dev->knode_driver)) {
+		printk(KERN_WARNING "%s: device %s already bound\n",
+			__FUNCTION__, kobject_name(&dev->kobj));
+		return 0;
+	}
 
 	pr_debug("bound device '%s' to driver '%s'\n",
 		 dev->bus_id, dev->driver->name);
 	klist_add_tail(&dev->knode_driver, &dev->driver->klist_devices);
-	sysfs_create_link(&dev->driver->kobj, &dev->kobj,
+	ret = sysfs_create_link(&dev->driver->kobj, &dev->kobj,
 			  kobject_name(&dev->kobj));
-	sysfs_create_link(&dev->kobj, &dev->driver->kobj, "driver");
+	if (ret == 0) {
+		ret = sysfs_create_link(&dev->kobj, &dev->driver->kobj,
+					"driver");
+		if (ret)
+			sysfs_remove_link(&dev->driver->kobj,
+					kobject_name(&dev->kobj));
+	}
+	return ret;
 }
 
 /**
@@ -91,7 +103,11 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 			goto ProbeFailed;
 		}
 	}
-	device_bind_driver(dev);
+	if (device_bind_driver(dev)) {
+		printk(KERN_ERR "%s: device_bind_driver(%s) failed\n",
+			__FUNCTION__, dev->bus_id);
+		/* How does undo a ->probe?  We're screwed. */
+	}
 	ret = 1;
 	pr_debug("%s: Bound Device %s to Driver %s\n",
 		 drv->bus->name, dev->bus_id, drv->name);
@@ -139,8 +155,9 @@ int device_attach(struct device * dev)
 
 	down(&dev->sem);
 	if (dev->driver) {
-		device_bind_driver(dev);
-		ret = 1;
+		ret = device_bind_driver(dev);
+		if (ret == 0)
+			ret = 1;
 	} else
 		ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach);
 	up(&dev->sem);
@@ -182,9 +199,9 @@ static int __driver_attach(struct device * dev, void * data)
  *	returns 0 and the @dev->driver is set, we've found a
  *	compatible pair.
  */
-void driver_attach(struct device_driver * drv)
+int driver_attach(struct device_driver * drv)
 {
-	bus_for_each_dev(drv->bus, NULL, drv, __driver_attach);
+	return bus_for_each_dev(drv->bus, NULL, drv, __driver_attach);
 }
 
 /**
diff --git a/include/linux/device.h b/include/linux/device.h
index ad4db72f5733..b3da9a870cfc 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -62,7 +62,7 @@ struct bus_type {
 extern int __must_check bus_register(struct bus_type * bus);
 extern void bus_unregister(struct bus_type * bus);
 
-extern void bus_rescan_devices(struct bus_type * bus);
+extern int __must_check bus_rescan_devices(struct bus_type * bus);
 
 /* iterator helpers for buses */
 
@@ -397,11 +397,11 @@ extern int device_rename(struct device *dev, char *new_name);
  * Manual binding of a device to driver. See drivers/base/bus.c
  * for information on use.
  */
-extern void device_bind_driver(struct device * dev);
+extern int __must_check device_bind_driver(struct device *dev);
 extern void device_release_driver(struct device * dev);
 extern int  __must_check device_attach(struct device * dev);
-extern void driver_attach(struct device_driver * drv);
-extern void device_reprobe(struct device *dev);
+extern int __must_check driver_attach(struct device_driver *drv);
+extern int __must_check device_reprobe(struct device *dev);
 
 /*
  * Easy functions for dynamically creating devices on the fly
-- 
cgit v1.2.3


From f20a9ead0d005fbeeae3fc21a96f9bf197ac1c1c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:23 -0700
Subject: sysfs: add proper sysfs_init() prototype

Don't be crufty.  Mark it __must_check too.

Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/namespace.c        | 10 +---------
 include/linux/sysfs.h |  7 +++++++
 2 files changed, 8 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/fs/namespace.c b/fs/namespace.c
index fa7ed6a9fc2d..36d180858136 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -17,6 +17,7 @@
 #include <linux/acct.h>
 #include <linux/capability.h>
 #include <linux/module.h>
+#include <linux/sysfs.h>
 #include <linux/seq_file.h>
 #include <linux/namespace.h>
 #include <linux/namei.h>
@@ -28,15 +29,6 @@
 
 extern int __init init_rootfs(void);
 
-#ifdef CONFIG_SYSFS
-extern int __init sysfs_init(void);
-#else
-static inline int sysfs_init(void)
-{
-	return 0;
-}
-#endif
-
 /* spinlock for vfsmount related operations, inplace of dcache_lock */
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock);
 
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 02ad790921fd..6d5c43d31dec 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -123,6 +123,8 @@ int __must_check sysfs_create_group(struct kobject *,
 void sysfs_remove_group(struct kobject *, const struct attribute_group *);
 void sysfs_notify(struct kobject * k, char *dir, char *attr);
 
+extern int __must_check sysfs_init(void);
+
 #else /* CONFIG_SYSFS */
 
 static inline int sysfs_create_dir(struct kobject * k)
@@ -194,6 +196,11 @@ static inline void sysfs_notify(struct kobject * k, char *dir, char *attr)
 {
 }
 
+static inline int __must_check sysfs_init(void)
+{
+	return 0;
+}
+
 #endif /* CONFIG_SYSFS */
 
 #endif /* _SYSFS_H_ */
-- 
cgit v1.2.3


From d779249ed4cb3b50690de6de8448829d65a1cd08 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 18 Jul 2006 10:59:59 -0700
Subject: Driver Core: add ability for drivers to do a threaded probe

This adds the infrastructure for drivers to do a threaded probe, and
waits at init time for all currently outstanding probes to complete.

A new kernel thread will be created when the probe() function for the
driver is called, if the multithread_probe bit is set in the driver
saying it can support this kind of operation.

I have tested this with USB and PCI, and it works, and shaves off a lot
of time in the boot process, but there are issues with finding root boot
disks, and some USB drivers assume that this can never happen, so it is
currently not enabled for any bus type.  Individual drivers can enable
this right now if they wish, and bus authors can selectivly turn it on
as well, once they determine that their subsystem will work properly
with it.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/dd.c      | 108 ++++++++++++++++++++++++++++++++++++-------------
 include/linux/device.h |   3 ++
 init/do_mounts.c       |   5 +++
 3 files changed, 89 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 9f6f11ca0ab6..319a73be4180 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -17,6 +17,7 @@
 
 #include <linux/device.h>
 #include <linux/module.h>
+#include <linux/kthread.h>
 
 #include "base.h"
 #include "power/power.h"
@@ -63,44 +64,35 @@ int device_bind_driver(struct device *dev)
 	return ret;
 }
 
-/**
- *	driver_probe_device - attempt to bind device & driver.
- *	@drv:	driver.
- *	@dev:	device.
- *
- *	First, we call the bus's match function, if one present, which
- *	should compare the device IDs the driver supports with the
- *	device IDs of the device. Note we don't do this ourselves
- *	because we don't know the format of the ID structures, nor what
- *	is to be considered a match and what is not.
- *
- *	This function returns 1 if a match is found, an error if one
- *	occurs (that is not -ENODEV or -ENXIO), and 0 otherwise.
- *
- *	This function must be called with @dev->sem held.  When called
- *	for a USB interface, @dev->parent->sem must be held as well.
- */
-int driver_probe_device(struct device_driver * drv, struct device * dev)
+struct stupid_thread_structure {
+	struct device_driver *drv;
+	struct device *dev;
+};
+
+static atomic_t probe_count = ATOMIC_INIT(0);
+static int really_probe(void *void_data)
 {
+	struct stupid_thread_structure *data = void_data;
+	struct device_driver *drv = data->drv;
+	struct device *dev = data->dev;
 	int ret = 0;
 
-	if (drv->bus->match && !drv->bus->match(dev, drv))
-		goto Done;
+	atomic_inc(&probe_count);
+	pr_debug("%s: Probing driver %s with device %s\n",
+		 drv->bus->name, drv->name, dev->bus_id);
 
-	pr_debug("%s: Matched Device %s with Driver %s\n",
-		 drv->bus->name, dev->bus_id, drv->name);
 	dev->driver = drv;
 	if (dev->bus->probe) {
 		ret = dev->bus->probe(dev);
 		if (ret) {
 			dev->driver = NULL;
-			goto ProbeFailed;
+			goto probe_failed;
 		}
 	} else if (drv->probe) {
 		ret = drv->probe(dev);
 		if (ret) {
 			dev->driver = NULL;
-			goto ProbeFailed;
+			goto probe_failed;
 		}
 	}
 	if (device_bind_driver(dev)) {
@@ -111,9 +103,9 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 	ret = 1;
 	pr_debug("%s: Bound Device %s to Driver %s\n",
 		 drv->bus->name, dev->bus_id, drv->name);
-	goto Done;
+	goto done;
 
- ProbeFailed:
+probe_failed:
 	if (ret == -ENODEV || ret == -ENXIO) {
 		/* Driver matched, but didn't support device
 		 * or device not found.
@@ -126,7 +118,69 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 		       "%s: probe of %s failed with error %d\n",
 		       drv->name, dev->bus_id, ret);
 	}
- Done:
+done:
+	kfree(data);
+	atomic_dec(&probe_count);
+	return ret;
+}
+
+/**
+ * driver_probe_done
+ * Determine if the probe sequence is finished or not.
+ *
+ * Should somehow figure out how to use a semaphore, not an atomic variable...
+ */
+int driver_probe_done(void)
+{
+	pr_debug("%s: probe_count = %d\n", __FUNCTION__,
+		 atomic_read(&probe_count));
+	if (atomic_read(&probe_count))
+		return -EBUSY;
+	return 0;
+}
+
+/**
+ * driver_probe_device - attempt to bind device & driver together
+ * @drv: driver to bind a device to
+ * @dev: device to try to bind to the driver
+ *
+ * First, we call the bus's match function, if one present, which should
+ * compare the device IDs the driver supports with the device IDs of the
+ * device. Note we don't do this ourselves because we don't know the
+ * format of the ID structures, nor what is to be considered a match and
+ * what is not.
+ *
+ * This function returns 1 if a match is found, an error if one occurs
+ * (that is not -ENODEV or -ENXIO), and 0 otherwise.
+ *
+ * This function must be called with @dev->sem held.  When called for a
+ * USB interface, @dev->parent->sem must be held as well.
+ */
+int driver_probe_device(struct device_driver * drv, struct device * dev)
+{
+	struct stupid_thread_structure *data;
+	struct task_struct *probe_task;
+	int ret = 0;
+
+	if (drv->bus->match && !drv->bus->match(dev, drv))
+		goto done;
+
+	pr_debug("%s: Matched Device %s with Driver %s\n",
+		 drv->bus->name, dev->bus_id, drv->name);
+
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	data->drv = drv;
+	data->dev = dev;
+
+	if (drv->multithread_probe) {
+		probe_task = kthread_run(really_probe, data,
+					 "probe-%s", dev->bus_id);
+		if (IS_ERR(probe_task))
+			ret = PTR_ERR(probe_task);
+	} else
+		ret = really_probe(data);
+
+done:
 	return ret;
 }
 
diff --git a/include/linux/device.h b/include/linux/device.h
index b3da9a870cfc..74246efba931 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -106,6 +106,8 @@ struct device_driver {
 	void	(*shutdown)	(struct device * dev);
 	int	(*suspend)	(struct device * dev, pm_message_t state);
 	int	(*resume)	(struct device * dev);
+
+	unsigned int multithread_probe:1;
 };
 
 
@@ -115,6 +117,7 @@ extern void driver_unregister(struct device_driver * drv);
 extern struct device_driver * get_driver(struct device_driver * drv);
 extern void put_driver(struct device_driver * drv);
 extern struct device_driver *driver_find(const char *name, struct bus_type *bus);
+extern int driver_probe_done(void);
 
 /* driverfs interface for exporting driver attributes */
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 94aeec7aa917..b290aadb1d3f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -8,6 +8,7 @@
 #include <linux/security.h>
 #include <linux/delay.h>
 #include <linux/mount.h>
+#include <linux/device.h>
 
 #include <linux/nfs_fs.h>
 #include <linux/nfs_fs_sb.h>
@@ -403,6 +404,10 @@ void __init prepare_namespace(void)
 		ssleep(root_delay);
 	}
 
+	/* wait for the known devices to complete their probing */
+	while (driver_probe_done() != 0)
+		msleep(100);
+
 	md_run_setup();
 
 	if (saved_root_name[0]) {
-- 
cgit v1.2.3


From f2eaae197f4590c4d96f31b09b0ee9067421a95c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 18 Sep 2006 16:22:34 -0400
Subject: Driver core: Fix potential deadlock in driver core

There is a potential deadlock in the driver core.  It boils down to
the fact that bus_remove_device() calls klist_remove() instead of
klist_del(), thereby waiting until the reference count of the
klist_node in the bus's klist of devices drops to 0.  The refcount
can't reach 0 so long as a modprobe process is trying to bind a new
driver to the device being removed, by calling __driver_attach().  The
problem is that __driver_attach() tries to acquire the device's
parent's semaphore, but the caller of bus_remove_device() is quite
likely to own that semaphore already.

It isn't sufficient just to replace klist_remove() with klist_del().
Doing so runs the risk that the device would remain on the bus's klist
of devices for some time, and so could be bound to another driver even
after it was unregistered.  What's needed is a new way to distinguish
whether or not a device is registered, based on a criterion other than
whether its klist_node is linked into the bus's klist of devices.  That
way driver binding can fail when the device is unregistered, even if
it is still linked into the klist.

This patch (as782) implements the solution, by adding a new bitflag to
indiate when a struct device is registered, by testing the flag before
allowing a driver to bind a device, and by changing the definition of
the device_is_registered() inline.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 8 ++++++--
 drivers/base/dd.c      | 2 ++
 include/linux/device.h | 3 ++-
 3 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index aa685a20b649..636af538a2b5 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -392,6 +392,7 @@ out:
  *	bus_attach_device - add device to bus
  *	@dev:	device tried to attach to a driver
  *
+ *	- Add device to bus's list of devices.
  *	- Try to attach to driver.
  */
 int bus_attach_device(struct device * dev)
@@ -400,11 +401,13 @@ int bus_attach_device(struct device * dev)
 	int ret = 0;
 
 	if (bus) {
+		dev->is_registered = 1;
 		ret = device_attach(dev);
 		if (ret >= 0) {
 			klist_add_tail(&dev->knode_bus, &bus->klist_devices);
 			ret = 0;
-		}
+		} else
+			dev->is_registered = 0;
 	}
 	return ret;
 }
@@ -425,7 +428,8 @@ void bus_remove_device(struct device * dev)
 		sysfs_remove_link(&dev->kobj, "bus");
 		sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id);
 		device_remove_attrs(dev->bus, dev);
-		klist_remove(&dev->knode_bus);
+		dev->is_registered = 0;
+		klist_del(&dev->knode_bus);
 		pr_debug("bus %s: remove device %s\n", dev->bus->name, dev->bus_id);
 		device_release_driver(dev);
 		put_bus(dev->bus);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 319a73be4180..b5f43c3e44fa 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -162,6 +162,8 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 	struct task_struct *probe_task;
 	int ret = 0;
 
+	if (!device_is_registered(dev))
+		return -ENODEV;
 	if (drv->bus->match && !drv->bus->match(dev, drv))
 		goto done;
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 74246efba931..662e6a10144e 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -329,6 +329,7 @@ struct device {
 
 	struct kobject kobj;
 	char	bus_id[BUS_ID_SIZE];	/* position on parent bus */
+	unsigned		is_registered:1;
 	struct device_attribute uevent_attr;
 	struct device_attribute *devt_attr;
 
@@ -381,7 +382,7 @@ dev_set_drvdata (struct device *dev, void *data)
 
 static inline int device_is_registered(struct device *dev)
 {
-	return klist_node_attached(&dev->knode_bus);
+	return dev->is_registered;
 }
 
 /*
-- 
cgit v1.2.3


From 874c4fe389d1358f82c96dc9b5092fc5c7690604 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] i386: Allow to use GENERICARCH for UP kernels

There are some machines around (large xSeries or Unisys ES7000) that
need physical IO-APIC destination mode to access all of their IO
devices. This currently doesn't work in UP kernels as used in
distribution installers.

This patch allows to compile even UP kernels as GENERICARCH which
allows to use physical or clustered APIC mode.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/Kconfig                        |  7 ++--
 arch/i386/kernel/io_apic.c               |  1 +
 arch/i386/kernel/mpparse.c               |  1 +
 arch/i386/mach-generic/bigsmp.c          |  1 +
 arch/i386/mach-generic/es7000.c          |  1 +
 arch/i386/mach-generic/probe.c           |  2 +
 arch/i386/mach-generic/summit.c          |  1 +
 include/asm-i386/genapic.h               | 69 +++++++++++++++++++-------------
 include/asm-i386/mach-es7000/mach_apic.h |  4 ++
 include/asm-i386/mach-summit/mach_apic.h | 11 ++++-
 include/asm-i386/smp.h                   | 19 +++++----
 11 files changed, 76 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index b2751eadbc56..c134a545b37e 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -166,7 +166,6 @@ config X86_VISWS
 
 config X86_GENERICARCH
        bool "Generic architecture (Summit, bigsmp, ES7000, default)"
-       depends on SMP
        help
           This option compiles in the Summit, bigsmp, ES7000, default subarchitectures.
 	  It is intended for a generic binary kernel.
@@ -263,7 +262,7 @@ source "kernel/Kconfig.preempt"
 
 config X86_UP_APIC
 	bool "Local APIC support on uniprocessors"
-	depends on !SMP && !(X86_VISWS || X86_VOYAGER)
+	depends on !SMP && !(X86_VISWS || X86_VOYAGER || X86_GENERICARCH)
 	help
 	  A local APIC (Advanced Programmable Interrupt Controller) is an
 	  integrated interrupt controller in the CPU. If you have a single-CPU
@@ -288,12 +287,12 @@ config X86_UP_IOAPIC
 
 config X86_LOCAL_APIC
 	bool
-	depends on X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER)
+	depends on X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER) || X86_GENERICARCH
 	default y
 
 config X86_IO_APIC
 	bool
-	depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER))
+	depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER)) || X86_GENERICARCH
 	default y
 
 config X86_VISWS_APIC
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 4fb32c551fe0..4eacd52eeb74 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -40,6 +40,7 @@
 #include <asm/nmi.h>
 
 #include <mach_apic.h>
+#include <mach_apicdef.h>
 
 #include "io_ports.h"
 
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index a70b5fa0ef06..827569688562 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -30,6 +30,7 @@
 #include <asm/io_apic.h>
 
 #include <mach_apic.h>
+#include <mach_apicdef.h>
 #include <mach_mpparse.h>
 #include <bios_ebda.h>
 
diff --git a/arch/i386/mach-generic/bigsmp.c b/arch/i386/mach-generic/bigsmp.c
index ef7a6e6fcb9f..33d9f93557ba 100644
--- a/arch/i386/mach-generic/bigsmp.c
+++ b/arch/i386/mach-generic/bigsmp.c
@@ -5,6 +5,7 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
diff --git a/arch/i386/mach-generic/es7000.c b/arch/i386/mach-generic/es7000.c
index 845cdd0b3593..aa144d82334d 100644
--- a/arch/i386/mach-generic/es7000.c
+++ b/arch/i386/mach-generic/es7000.c
@@ -4,6 +4,7 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
diff --git a/arch/i386/mach-generic/probe.c b/arch/i386/mach-generic/probe.c
index bcd1bcfaa723..793d1b473251 100644
--- a/arch/i386/mach-generic/probe.c
+++ b/arch/i386/mach-generic/probe.c
@@ -119,7 +119,9 @@ int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 0;	
 }
 
+#ifdef CONFIG_SMP
 int hard_smp_processor_id(void)
 {
 	return genapic->get_apic_id(*(unsigned long *)(APIC_BASE+APIC_ID));
 }
+#endif
diff --git a/arch/i386/mach-generic/summit.c b/arch/i386/mach-generic/summit.c
index b73501ddd653..f7e5d66648dc 100644
--- a/arch/i386/mach-generic/summit.c
+++ b/arch/i386/mach-generic/summit.c
@@ -4,6 +4,7 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
diff --git a/include/asm-i386/genapic.h b/include/asm-i386/genapic.h
index b3783a32abee..8ffbb0f07457 100644
--- a/include/asm-i386/genapic.h
+++ b/include/asm-i386/genapic.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_GENAPIC_H
 #define _ASM_GENAPIC_H 1
 
+#include <asm/mpspec.h>
+
 /*
  * Generic APIC driver interface.
  *
@@ -63,14 +65,25 @@ struct genapic {
 	unsigned (*get_apic_id)(unsigned long x);
 	unsigned long apic_id_mask;
 	unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask);
-	
+
+#ifdef CONFIG_SMP
 	/* ipi */
 	void (*send_IPI_mask)(cpumask_t mask, int vector);
 	void (*send_IPI_allbutself)(int vector);
 	void (*send_IPI_all)(int vector);
+#endif
 }; 
 
-#define APICFUNC(x) .x = x
+#define APICFUNC(x) .x = x,
+
+/* More functions could be probably marked IPIFUNC and save some space
+   in UP GENERICARCH kernels, but I don't have the nerve right now
+   to untangle this mess. -AK  */
+#ifdef CONFIG_SMP
+#define IPIFUNC(x) APICFUNC(x)
+#else
+#define IPIFUNC(x)
+#endif
 
 #define APIC_INIT(aname, aprobe) { \
 	.name = aname, \
@@ -80,33 +93,33 @@ struct genapic {
 	.no_balance_irq = NO_BALANCE_IRQ, \
 	.ESR_DISABLE = esr_disable, \
 	.apic_destination_logical = APIC_DEST_LOGICAL, \
-	APICFUNC(apic_id_registered), \
-	APICFUNC(target_cpus), \
-	APICFUNC(check_apicid_used), \
-	APICFUNC(check_apicid_present), \
-	APICFUNC(init_apic_ldr), \
-	APICFUNC(ioapic_phys_id_map), \
-	APICFUNC(clustered_apic_check), \
-	APICFUNC(multi_timer_check), \
-	APICFUNC(apicid_to_node), \
-	APICFUNC(cpu_to_logical_apicid), \
-	APICFUNC(cpu_present_to_apicid), \
-	APICFUNC(apicid_to_cpu_present), \
-	APICFUNC(mpc_apic_id), \
-	APICFUNC(setup_portio_remap), \
-	APICFUNC(check_phys_apicid_present), \
-	APICFUNC(mpc_oem_bus_info), \
-	APICFUNC(mpc_oem_pci_bus), \
-	APICFUNC(mps_oem_check), \
-	APICFUNC(get_apic_id), \
+	APICFUNC(apic_id_registered) \
+	APICFUNC(target_cpus) \
+	APICFUNC(check_apicid_used) \
+	APICFUNC(check_apicid_present) \
+	APICFUNC(init_apic_ldr) \
+	APICFUNC(ioapic_phys_id_map) \
+	APICFUNC(clustered_apic_check) \
+	APICFUNC(multi_timer_check) \
+	APICFUNC(apicid_to_node) \
+	APICFUNC(cpu_to_logical_apicid) \
+	APICFUNC(cpu_present_to_apicid) \
+	APICFUNC(apicid_to_cpu_present) \
+	APICFUNC(mpc_apic_id) \
+	APICFUNC(setup_portio_remap) \
+	APICFUNC(check_phys_apicid_present) \
+	APICFUNC(mpc_oem_bus_info) \
+	APICFUNC(mpc_oem_pci_bus) \
+	APICFUNC(mps_oem_check) \
+	APICFUNC(get_apic_id) \
 	.apic_id_mask = APIC_ID_MASK, \
-	APICFUNC(cpu_mask_to_apicid), \
-	APICFUNC(acpi_madt_oem_check), \
-	APICFUNC(send_IPI_mask), \
-	APICFUNC(send_IPI_allbutself), \
-	APICFUNC(send_IPI_all), \
-	APICFUNC(enable_apic_mode), \
-	APICFUNC(phys_pkg_id), \
+	APICFUNC(cpu_mask_to_apicid) \
+	APICFUNC(acpi_madt_oem_check) \
+	IPIFUNC(send_IPI_mask) \
+	IPIFUNC(send_IPI_allbutself) \
+	IPIFUNC(send_IPI_all) \
+	APICFUNC(enable_apic_mode) \
+	APICFUNC(phys_pkg_id) \
 	}
 
 extern struct genapic *genapic;
diff --git a/include/asm-i386/mach-es7000/mach_apic.h b/include/asm-i386/mach-es7000/mach_apic.h
index b5f3f0d0b2bc..26333685a7fb 100644
--- a/include/asm-i386/mach-es7000/mach_apic.h
+++ b/include/asm-i386/mach-es7000/mach_apic.h
@@ -123,9 +123,13 @@ extern u8 cpu_2_logical_apicid[];
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
 {
+#ifdef CONFIG_SMP
        if (cpu >= NR_CPUS)
 	       return BAD_APICID;
        return (int)cpu_2_logical_apicid[cpu];
+#else
+	return logical_smp_processor_id();
+#endif
 }
 
 static inline int mpc_apic_id(struct mpc_config_processor *m, struct mpc_config_translation *unused)
diff --git a/include/asm-i386/mach-summit/mach_apic.h b/include/asm-i386/mach-summit/mach_apic.h
index 9fd073286289..a81b05961595 100644
--- a/include/asm-i386/mach-summit/mach_apic.h
+++ b/include/asm-i386/mach-summit/mach_apic.h
@@ -46,10 +46,12 @@ extern u8 cpu_2_logical_apicid[];
 static inline void init_apic_ldr(void)
 {
 	unsigned long val, id;
-	int i, count;
-	u8 lid;
+	int count = 0;
 	u8 my_id = (u8)hard_smp_processor_id();
 	u8 my_cluster = (u8)apicid_cluster(my_id);
+#ifdef CONFIG_SMP
+	u8 lid;
+	int i;
 
 	/* Create logical APIC IDs by counting CPUs already in cluster. */
 	for (count = 0, i = NR_CPUS; --i >= 0; ) {
@@ -57,6 +59,7 @@ static inline void init_apic_ldr(void)
 		if (lid != BAD_APICID && apicid_cluster(lid) == my_cluster)
 			++count;
 	}
+#endif
 	/* We only have a 4 wide bitmap in cluster mode.  If a deranged
 	 * BIOS puts 5 CPUs in one APIC cluster, we're hosed. */
 	BUG_ON(count >= XAPIC_DEST_CPUS_SHIFT);
@@ -91,9 +94,13 @@ static inline int apicid_to_node(int logical_apicid)
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
 {
+#ifdef CONFIG_SMP
        if (cpu >= NR_CPUS)
 	       return BAD_APICID;
 	return (int)cpu_2_logical_apicid[cpu];
+#else
+	return logical_smp_processor_id();
+#endif
 }
 
 static inline int cpu_present_to_apicid(int mps_cpu)
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 142d10e34ade..f87826039a51 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -80,17 +80,11 @@ static inline int hard_smp_processor_id(void)
 	return GET_APIC_ID(*(unsigned long *)(APIC_BASE+APIC_ID));
 }
 #endif
-
-static __inline int logical_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
-}
-
 #endif
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
+
 #endif /* !__ASSEMBLY__ */
 
 #else /* CONFIG_SMP */
@@ -100,4 +94,15 @@ extern void __cpu_die(unsigned int cpu);
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 
 #endif
+
+#ifndef __ASSEMBLY__
+#ifdef CONFIG_X86_LOCAL_APIC
+static __inline int logical_smp_processor_id(void)
+{
+	/* we don't want to mark this access volatile - bad code generation */
+	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
+}
+#endif
+#endif
+
 #endif
-- 
cgit v1.2.3


From b07f8915cda3fcd73b8b68075ba1e6cd0673365d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] x86: Temporarily revert parts of the Core 2 nmi nmi watchdog
 support

This makes merging easier.  They are readded a few patches later.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c                  | 65 +-------------------------
 arch/x86_64/kernel/nmi.c                | 81 ++-------------------------------
 include/asm-i386/intel_arch_perfmon.h   | 19 --------
 include/asm-x86_64/intel_arch_perfmon.h | 19 --------
 4 files changed, 6 insertions(+), 178 deletions(-)
 delete mode 100644 include/asm-i386/intel_arch_perfmon.h
 delete mode 100644 include/asm-x86_64/intel_arch_perfmon.h

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index acb351478e42..1282d70ff971 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -24,7 +24,6 @@
 
 #include <asm/smp.h>
 #include <asm/nmi.h>
-#include <asm/intel_arch_perfmon.h>
 
 #include "mach_traps.h"
 
@@ -96,9 +95,6 @@ int nmi_active;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
-#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
-#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
-
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -211,8 +207,6 @@ static int __init setup_nmi_watchdog(char *str)
 
 __setup("nmi_watchdog=", setup_nmi_watchdog);
 
-static void disable_intel_arch_watchdog(void);
-
 static void disable_lapic_nmi_watchdog(void)
 {
 	if (nmi_active <= 0)
@@ -222,10 +216,6 @@ static void disable_lapic_nmi_watchdog(void)
 		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
 		break;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			disable_intel_arch_watchdog();
-			break;
-		}
 		switch (boot_cpu_data.x86) {
 		case 6:
 			if (boot_cpu_data.x86_model > 0xd)
@@ -454,53 +444,6 @@ static int setup_p4_watchdog(void)
 	return 1;
 }
 
-static void disable_intel_arch_watchdog(void)
-{
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if (!(ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, 0, 0);
-}
-
-static int setup_intel_arch_watchdog(void)
-{
-	unsigned int evntsel;
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if ((ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		return 0;
-
-	nmi_perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
-
-	clear_msr_range(MSR_ARCH_PERFMON_EVENTSEL0, 2);
-	clear_msr_range(MSR_ARCH_PERFMON_PERFCTR0, 2);
-
-	evntsel = ARCH_PERFMON_EVENTSEL_INT
-		| ARCH_PERFMON_EVENTSEL_OS
-		| ARCH_PERFMON_EVENTSEL_USR
-		| ARCH_PERFMON_NMI_EVENT_SEL
-		| ARCH_PERFMON_NMI_EVENT_UMASK;
-
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	write_watchdog_counter("INTEL_ARCH_PERFCTR0");
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	return 1;
-}
-
 void setup_apic_nmi_watchdog (void)
 {
 	switch (boot_cpu_data.x86_vendor) {
@@ -510,11 +453,6 @@ void setup_apic_nmi_watchdog (void)
 		setup_k7_watchdog();
 		break;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			if (!setup_intel_arch_watchdog())
-				return;
-			break;
-		}
 		switch (boot_cpu_data.x86) {
 		case 6:
 			if (boot_cpu_data.x86_model > 0xd)
@@ -619,8 +557,7 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 			apic_write(APIC_LVTPC, APIC_DM_NMI);
 		}
-		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0 ||
-		         nmi_perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
+		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0) {
 			/* Only P6 based Pentium M need to re-unmask
 			 * the apic vector but it doesn't hurt
 			 * other P6 variant */
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5baa0c726e97..42c05d6907b9 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -26,7 +26,6 @@
 #include <asm/proto.h>
 #include <asm/kdebug.h>
 #include <asm/mce.h>
-#include <asm/intel_arch_perfmon.h>
 
 /*
  * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
@@ -66,9 +65,6 @@ static unsigned int nmi_p4_cccr_val;
 #define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
 #define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
 
-#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
-#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
-
 #define MSR_P4_MISC_ENABLE	0x1A0
 #define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
 #define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
@@ -100,10 +96,7 @@ static __cpuinit inline int nmi_known_cpu(void)
 	case X86_VENDOR_AMD:
 		return boot_cpu_data.x86 == 15;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
-			return 1;
-		else
-			return (boot_cpu_data.x86 == 15);
+		return boot_cpu_data.x86 == 15;
 	}
 	return 0;
 }
@@ -209,8 +202,6 @@ int __init setup_nmi_watchdog(char *str)
 
 __setup("nmi_watchdog=", setup_nmi_watchdog);
 
-static void disable_intel_arch_watchdog(void);
-
 static void disable_lapic_nmi_watchdog(void)
 {
 	if (nmi_active <= 0)
@@ -223,8 +214,6 @@ static void disable_lapic_nmi_watchdog(void)
 		if (boot_cpu_data.x86 == 15) {
 			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
 			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
-		} else if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			disable_intel_arch_watchdog();
 		}
 		break;
 	}
@@ -377,53 +366,6 @@ static void setup_k7_watchdog(void)
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
 }
 
-static void disable_intel_arch_watchdog(void)
-{
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if (!(ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, 0, 0);
-}
-
-static int setup_intel_arch_watchdog(void)
-{
-	unsigned int evntsel;
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if ((ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		return 0;
-
-	nmi_perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
-
-	clear_msr_range(MSR_ARCH_PERFMON_EVENTSEL0, 2);
-	clear_msr_range(MSR_ARCH_PERFMON_PERFCTR0, 2);
-
-	evntsel = ARCH_PERFMON_EVENTSEL_INT
-		| ARCH_PERFMON_EVENTSEL_OS
-		| ARCH_PERFMON_EVENTSEL_USR
-		| ARCH_PERFMON_NMI_EVENT_SEL
-		| ARCH_PERFMON_NMI_EVENT_UMASK;
-
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	wrmsrl(MSR_ARCH_PERFMON_PERFCTR0, -((u64)cpu_khz * 1000 / nmi_hz));
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	return 1;
-}
-
 
 static int setup_p4_watchdog(void)
 {
@@ -477,16 +419,10 @@ void setup_apic_nmi_watchdog(void)
 		setup_k7_watchdog();
 		break;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			if (!setup_intel_arch_watchdog())
-				return;
-		} else if (boot_cpu_data.x86 == 15) {
-			if (!setup_p4_watchdog())
-				return;
-		} else {
+		if (boot_cpu_data.x86 != 15)
+			return;
+		if (!setup_p4_watchdog())
 			return;
-		}
-
 		break;
 
 	default:
@@ -571,14 +507,7 @@ void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
  			 */
  			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
  			apic_write(APIC_LVTPC, APIC_DM_NMI);
- 		} else if (nmi_perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
-			/*
-			 * For Intel based architectural perfmon
-			 * - LVTPC is masked on interrupt and must be
-			 *   unmasked by the LVTPC handler.
-			 */
-			apic_write(APIC_LVTPC, APIC_DM_NMI);
-		}
+ 		}
 		wrmsrl(nmi_perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 	}
 }
diff --git a/include/asm-i386/intel_arch_perfmon.h b/include/asm-i386/intel_arch_perfmon.h
deleted file mode 100644
index 134ea9cc5283..000000000000
--- a/include/asm-i386/intel_arch_perfmon.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef X86_INTEL_ARCH_PERFMON_H
-#define X86_INTEL_ARCH_PERFMON_H 1
-
-#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
-#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
-
-#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
-#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
-
-#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
-#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
-#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
-#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
-
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT (1 << 0)
-
-#endif	/* X86_INTEL_ARCH_PERFMON_H */
diff --git a/include/asm-x86_64/intel_arch_perfmon.h b/include/asm-x86_64/intel_arch_perfmon.h
deleted file mode 100644
index 59c396431569..000000000000
--- a/include/asm-x86_64/intel_arch_perfmon.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef X86_64_INTEL_ARCH_PERFMON_H
-#define X86_64_INTEL_ARCH_PERFMON_H 1
-
-#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
-#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
-
-#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
-#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
-
-#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
-#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
-#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
-#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
-
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT (1 << 0)
-
-#endif	/* X86_64_INTEL_ARCH_PERFMON_H */
-- 
cgit v1.2.3


From 828f0afda123a96ff4e8078f057a302f4b4232ae Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] x86: Add performance counter reservation framework for UP
 kernels

Adds basic infrastructure to allow subsystems to reserve performance
counters on the x86 chips.  Only UP kernels are supported in this patch to
make reviewing easier.  The SMP portion makes a lot more changes.

Think of this as a locking mechanism where each bit represents a different
counter.  In addition, each subsystem should also reserve an appropriate
event selection register that will correspond to the performance counter it
will be using (this is mainly neccessary for the Pentium 4 chips as they
break the 1:1 relationship to performance counters).

This will help prevent subsystems like oprofile from interfering with the
nmi watchdog.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c   | 188 +++++++++++++++++++++++++++++++++++++++--------
 arch/x86_64/kernel/nmi.c | 178 ++++++++++++++++++++++++++++++++++----------
 include/asm-i386/nmi.h   |   7 ++
 include/asm-x86_64/nmi.h |   8 +-
 4 files changed, 308 insertions(+), 73 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 1282d70ff971..5d58dfeacd59 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -34,6 +34,20 @@ static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
 static unsigned int nmi_p4_cccr_val;
 extern void show_registers(struct pt_regs *regs);
 
+/* perfctr_nmi_owner tracks the ownership of the perfctr registers:
+ * evtsel_nmi_owner tracks the ownership of the event selection
+ * - different performance counters/ event selection may be reserved for
+ *   different subsystems this reservation system just tries to coordinate
+ *   things a little
+ */
+static DEFINE_PER_CPU(unsigned long, perfctr_nmi_owner);
+static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]);
+
+/* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's
+ * offset from MSR_P4_BSU_ESCR0.  It will be the max for all platforms (for now)
+ */
+#define NMI_MAX_COUNTER_BITS 66
+
 /*
  * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
  * - it may be reserved by some other driver, or not
@@ -95,6 +109,105 @@ int nmi_active;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the performance counter register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_PERFCTR0);
+	case X86_VENDOR_INTEL:
+		switch (boot_cpu_data.x86) {
+		case 6:
+			return (msr - MSR_P6_PERFCTR0);
+		case 15:
+			return (msr - MSR_P4_BPU_PERFCTR0);
+		}
+	}
+	return 0;
+}
+
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the event selection register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_EVNTSEL0);
+	case X86_VENDOR_INTEL:
+		switch (boot_cpu_data.x86) {
+		case 6:
+			return (msr - MSR_P6_EVNTSEL0);
+		case 15:
+			return (msr - MSR_P4_BSU_ESCR0);
+		}
+	}
+	return 0;
+}
+
+/* checks for a bit availability (hack for oprofile) */
+int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
+{
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+/* checks the an msr for availability */
+int avail_to_resrv_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+int reserve_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(perfctr_nmi_owner)))
+		return 1;
+	return 0;
+}
+
+void release_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(perfctr_nmi_owner));
+}
+
+int reserve_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]))
+		return 1;
+	return 0;
+}
+
+void release_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]);
+}
+
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -344,14 +457,6 @@ late_initcall(init_lapic_nmi_sysfs);
  * Original code written by Keith Owens.
  */
 
-static void clear_msr_range(unsigned int base, unsigned int n)
-{
-	unsigned int i;
-
-	for(i = 0; i < n; ++i)
-		wrmsr(base+i, 0, 0);
-}
-
 static void write_watchdog_counter(const char *descr)
 {
 	u64 count = (u64)cpu_khz * 1000;
@@ -362,14 +467,19 @@ static void write_watchdog_counter(const char *descr)
 	wrmsrl(nmi_perfctr_msr, 0 - count);
 }
 
-static void setup_k7_watchdog(void)
+static int setup_k7_watchdog(void)
 {
 	unsigned int evntsel;
 
 	nmi_perfctr_msr = MSR_K7_PERFCTR0;
 
-	clear_msr_range(MSR_K7_EVNTSEL0, 4);
-	clear_msr_range(MSR_K7_PERFCTR0, 4);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+		goto fail1;
+
+	wrmsrl(MSR_K7_PERFCTR0, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
@@ -381,16 +491,24 @@ static void setup_k7_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
-static void setup_p6_watchdog(void)
+static int setup_p6_watchdog(void)
 {
 	unsigned int evntsel;
 
 	nmi_perfctr_msr = MSR_P6_PERFCTR0;
 
-	clear_msr_range(MSR_P6_EVNTSEL0, 2);
-	clear_msr_range(MSR_P6_PERFCTR0, 2);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_P6_EVNTSEL0))
+		goto fail1;
 
 	evntsel = P6_EVNTSEL_INT
 		| P6_EVNTSEL_OS
@@ -402,6 +520,11 @@ static void setup_p6_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= P6_EVNTSEL0_ENABLE;
 	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
+	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 static int setup_p4_watchdog(void)
@@ -419,22 +542,11 @@ static int setup_p4_watchdog(void)
 		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
 #endif
 
-	if (!(misc_enable & MSR_P4_MISC_ENABLE_PEBS_UNAVAIL))
-		clear_msr_range(0x3F1, 2);
-	/* MSR 0x3F0 seems to have a default value of 0xFC00, but current
-	   docs doesn't fully define it, so leave it alone for now. */
-	if (boot_cpu_data.x86_model >= 0x3) {
-		/* MSR_P4_IQ_ESCR0/1 (0x3ba/0x3bb) removed */
-		clear_msr_range(0x3A0, 26);
-		clear_msr_range(0x3BC, 3);
-	} else {
-		clear_msr_range(0x3A0, 31);
-	}
-	clear_msr_range(0x3C0, 6);
-	clear_msr_range(0x3C8, 6);
-	clear_msr_range(0x3E0, 2);
-	clear_msr_range(MSR_P4_CCCR0, 18);
-	clear_msr_range(MSR_P4_PERFCTR0, 18);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+		goto fail1;
 
 	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
 	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
@@ -442,6 +554,10 @@ static int setup_p4_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 void setup_apic_nmi_watchdog (void)
@@ -450,7 +566,8 @@ void setup_apic_nmi_watchdog (void)
 	case X86_VENDOR_AMD:
 		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
 			return;
-		setup_k7_watchdog();
+		if (!setup_k7_watchdog())
+			return;
 		break;
 	case X86_VENDOR_INTEL:
 		switch (boot_cpu_data.x86) {
@@ -458,7 +575,8 @@ void setup_apic_nmi_watchdog (void)
 			if (boot_cpu_data.x86_model > 0xd)
 				return;
 
-			setup_p6_watchdog();
+			if(!setup_p6_watchdog())
+				return;
 			break;
 		case 15:
 			if (boot_cpu_data.x86_model > 0x4)
@@ -612,6 +730,12 @@ int proc_unknown_nmi_panic(ctl_table *table, int write, struct file *file,
 
 EXPORT_SYMBOL(nmi_active);
 EXPORT_SYMBOL(nmi_watchdog);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
+EXPORT_SYMBOL(reserve_perfctr_nmi);
+EXPORT_SYMBOL(release_perfctr_nmi);
+EXPORT_SYMBOL(reserve_evntsel_nmi);
+EXPORT_SYMBOL(release_evntsel_nmi);
 EXPORT_SYMBOL(reserve_lapic_nmi);
 EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 42c05d6907b9..b7a7c9973849 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -27,6 +27,20 @@
 #include <asm/kdebug.h>
 #include <asm/mce.h>
 
+/* perfctr_nmi_owner tracks the ownership of the perfctr registers:
+ * evtsel_nmi_owner tracks the ownership of the event selection
+ * - different performance counters/ event selection may be reserved for
+ *   different subsystems this reservation system just tries to coordinate
+ *   things a little
+ */
+static DEFINE_PER_CPU(unsigned, perfctr_nmi_owner);
+static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]);
+
+/* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's
+ * offset from MSR_P4_BSU_ESCR0.  It will be the max for all platforms (for now)
+ */
+#define NMI_MAX_COUNTER_BITS 66
+
 /*
  * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
  * - it may be reserved by some other driver, or not
@@ -90,6 +104,95 @@ static unsigned int nmi_p4_cccr_val;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the performance counter register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_PERFCTR0);
+	case X86_VENDOR_INTEL:
+		return (msr - MSR_P4_BPU_PERFCTR0);
+	}
+	return 0;
+}
+
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the event selection register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_EVNTSEL0);
+	case X86_VENDOR_INTEL:
+		return (msr - MSR_P4_BSU_ESCR0);
+	}
+	return 0;
+}
+
+/* checks for a bit availability (hack for oprofile) */
+int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
+{
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+/* checks the an msr for availability */
+int avail_to_resrv_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+int reserve_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(perfctr_nmi_owner)))
+		return 1;
+	return 0;
+}
+
+void release_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(perfctr_nmi_owner));
+}
+
+int reserve_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(evntsel_nmi_owner)))
+		return 1;
+	return 0;
+}
+
+void release_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner));
+}
+
 static __cpuinit inline int nmi_known_cpu(void)
 {
 	switch (boot_cpu_data.x86_vendor) {
@@ -325,34 +428,22 @@ late_initcall(init_lapic_nmi_sysfs);
 
 #endif	/* CONFIG_PM */
 
-/*
- * Activate the NMI watchdog via the local APIC.
- * Original code written by Keith Owens.
- */
-
-static void clear_msr_range(unsigned int base, unsigned int n)
+static int setup_k7_watchdog(void)
 {
-	unsigned int i;
-
-	for(i = 0; i < n; ++i)
-		wrmsr(base+i, 0, 0);
-}
-
-static void setup_k7_watchdog(void)
-{
-	int i;
 	unsigned int evntsel;
 
 	nmi_perfctr_msr = MSR_K7_PERFCTR0;
 
-	for(i = 0; i < 4; ++i) {
-		/* Simulator may not support it */
-		if (checking_wrmsrl(MSR_K7_EVNTSEL0+i, 0UL)) {
-			nmi_perfctr_msr = 0;
-			return;
-		}
-		wrmsrl(MSR_K7_PERFCTR0+i, 0UL);
-	}
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+		goto fail1;
+
+	/* Simulator may not support it */
+	if (checking_wrmsrl(MSR_K7_EVNTSEL0, 0UL))
+		goto fail2;
+	wrmsrl(MSR_K7_PERFCTR0, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
@@ -364,6 +455,13 @@ static void setup_k7_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	return 1;
+fail2:
+	release_evntsel_nmi(MSR_K7_EVNTSEL0);
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 
@@ -382,22 +480,11 @@ static int setup_p4_watchdog(void)
 		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
 #endif
 
-	if (!(misc_enable & MSR_P4_MISC_ENABLE_PEBS_UNAVAIL))
-		clear_msr_range(0x3F1, 2);
-	/* MSR 0x3F0 seems to have a default value of 0xFC00, but current
-	   docs doesn't fully define it, so leave it alone for now. */
-	if (boot_cpu_data.x86_model >= 0x3) {
-		/* MSR_P4_IQ_ESCR0/1 (0x3ba/0x3bb) removed */
-		clear_msr_range(0x3A0, 26);
-		clear_msr_range(0x3BC, 3);
-	} else {
-		clear_msr_range(0x3A0, 31);
-	}
-	clear_msr_range(0x3C0, 6);
-	clear_msr_range(0x3C8, 6);
-	clear_msr_range(0x3E0, 2);
-	clear_msr_range(MSR_P4_CCCR0, 18);
-	clear_msr_range(MSR_P4_PERFCTR0, 18);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+		goto fail1;
 
 	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
 	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
@@ -406,6 +493,10 @@ static int setup_p4_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 void setup_apic_nmi_watchdog(void)
@@ -416,7 +507,8 @@ void setup_apic_nmi_watchdog(void)
 			return;
 		if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
 			return;
-		setup_k7_watchdog();
+		if (!setup_k7_watchdog())
+			return;
 		break;
 	case X86_VENDOR_INTEL:
 		if (boot_cpu_data.x86 != 15)
@@ -588,6 +680,12 @@ int proc_unknown_nmi_panic(struct ctl_table *table, int write, struct file *file
 
 EXPORT_SYMBOL(nmi_active);
 EXPORT_SYMBOL(nmi_watchdog);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
+EXPORT_SYMBOL(reserve_perfctr_nmi);
+EXPORT_SYMBOL(release_perfctr_nmi);
+EXPORT_SYMBOL(reserve_evntsel_nmi);
+EXPORT_SYMBOL(release_evntsel_nmi);
 EXPORT_SYMBOL(reserve_lapic_nmi);
 EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 67d994799999..27fc9e6f630e 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -25,6 +25,13 @@ void set_nmi_callback(nmi_callback_t callback);
  */
 void unset_nmi_callback(void);
 
+extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
+extern int avail_to_resrv_perfctr_nmi(unsigned int);
+extern int reserve_perfctr_nmi(unsigned int);
+extern void release_perfctr_nmi(unsigned int);
+extern int reserve_evntsel_nmi(unsigned int);
+extern void release_evntsel_nmi(unsigned int);
+
 extern void setup_apic_nmi_watchdog (void);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index efb45c894d76..62a784cb8f0c 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -56,7 +56,13 @@ extern int panic_on_timeout;
 extern int unknown_nmi_panic;
 
 extern int check_nmi_watchdog(void);
- 
+extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
+extern int avail_to_resrv_perfctr_nmi(unsigned int);
+extern int reserve_perfctr_nmi(unsigned int);
+extern void release_perfctr_nmi(unsigned int);
+extern int reserve_evntsel_nmi(unsigned int);
+extern void release_evntsel_nmi(unsigned int);
+
 extern void setup_apic_nmi_watchdog (void);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
-- 
cgit v1.2.3


From f2802e7f571c05f9a901b1f5bd144aa730ccc88e Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] Add SMP support on x86_64 to reservation framework

This patch includes the changes to make the nmi watchdog on x86_64 SMP
aware.  A bunch of code was moved around to make it simpler to read.  In
addition, it is now possible to determine if a particular NMI was the result
of the watchdog or not.  This feature allows the kernel to filter out
unknown NMIs easier.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c |   3 +-
 arch/x86_64/kernel/nmi.c  | 426 +++++++++++++++++++++++++++++++---------------
 include/asm-x86_64/nmi.h  |   3 +-
 3 files changed, 290 insertions(+), 142 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 2b8cef037a65..692e9b579743 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -479,8 +479,7 @@ void __cpuinit setup_local_APIC (void)
 	}
 
 	nmi_watchdog_default();
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		setup_apic_nmi_watchdog();
+	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
 }
 
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index b7a7c9973849..d42374a952d7 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -56,53 +56,29 @@ static unsigned int lapic_nmi_owner;
 #define LAPIC_NMI_RESERVED	(1<<1)
 
 /* nmi_active:
- * +1: the lapic NMI watchdog is active, but can be disabled
- *  0: the lapic NMI watchdog has not been set up, and cannot
+ * >0: the lapic NMI watchdog is active, but can be disabled
+ * <0: the lapic NMI watchdog has not been set up, and cannot
  *     be enabled
- * -1: the lapic NMI watchdog is disabled, but can be enabled
+ *  0: the lapic NMI watchdog is disabled, but can be enabled
  */
-int nmi_active;		/* oprofile uses this */
+atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
 int panic_on_timeout;
 
 unsigned int nmi_watchdog = NMI_DEFAULT;
 static unsigned int nmi_hz = HZ;
-static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
-static unsigned int nmi_p4_cccr_val;
 
-/* Note that these events don't tick when the CPU idles. This means
-   the frequency varies with CPU load. */
-
-#define K7_EVNTSEL_ENABLE	(1 << 22)
-#define K7_EVNTSEL_INT		(1 << 20)
-#define K7_EVNTSEL_OS		(1 << 17)
-#define K7_EVNTSEL_USR		(1 << 16)
-#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
-#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+struct nmi_watchdog_ctlblk {
+	int enabled;
+	u64 check_bit;
+	unsigned int cccr_msr;
+	unsigned int perfctr_msr;  /* the MSR to reset in NMI handler */
+	unsigned int evntsel_msr;  /* the MSR to select the events to handle */
+};
+static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
-#define MSR_P4_MISC_ENABLE	0x1A0
-#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
-#define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
-#define MSR_P4_PERFCTR0		0x300
-#define MSR_P4_CCCR0		0x360
-#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
-#define P4_ESCR_OS		(1<<3)
-#define P4_ESCR_USR		(1<<2)
-#define P4_CCCR_OVF_PMI0	(1<<26)
-#define P4_CCCR_OVF_PMI1	(1<<27)
-#define P4_CCCR_THRESHOLD(N)	((N)<<20)
-#define P4_CCCR_COMPLEMENT	(1<<19)
-#define P4_CCCR_COMPARE		(1<<18)
-#define P4_CCCR_REQUIRED	(3<<16)
-#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
-#define P4_CCCR_ENABLE		(1<<12)
-/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
-   CRU_ESCR0 (with any non-null event selector) through a complemented
-   max threshold. [IA32-Vol3, Section 14.9.9] */
-#define MSR_P4_IQ_COUNTER0	0x30C
-#define P4_NMI_CRU_ESCR0	(P4_ESCR_EVENT_SELECT(0x3F)|P4_ESCR_OS|P4_ESCR_USR)
-#define P4_NMI_IQ_CCCR0	\
-	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
-	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
+/* local prototypes */
+static void stop_apic_nmi_watchdog(void *unused);
+static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
 
 /* converts an msr to an appropriate reservation bit */
 static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
@@ -241,6 +217,12 @@ int __init check_nmi_watchdog (void)
 	int *counts;
 	int cpu;
 
+	if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT))
+		return 0;
+
+	if (!atomic_read(&nmi_active))
+		return 0;
+
 	counts = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
 	if (!counts)
 		return -1;
@@ -258,19 +240,22 @@ int __init check_nmi_watchdog (void)
 	mdelay((10*1000)/nmi_hz); // wait 10 ticks
 
 	for_each_online_cpu(cpu) {
+		if (!per_cpu(nmi_watchdog_ctlblk, cpu).enabled)
+			continue;
 		if (cpu_pda(cpu)->__nmi_count - counts[cpu] <= 5) {
-			endflag = 1;
 			printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
 			       cpu,
 			       counts[cpu],
 			       cpu_pda(cpu)->__nmi_count);
-			nmi_active = 0;
-			lapic_nmi_owner &= ~LAPIC_NMI_WATCHDOG;
-			nmi_perfctr_msr = 0;
-			kfree(counts);
-			return -1;
+			per_cpu(nmi_watchdog_ctlblk, cpu).enabled = 0;
+			atomic_dec(&nmi_active);
 		}
 	}
+	if (!atomic_read(&nmi_active)) {
+		kfree(counts);
+		atomic_set(&nmi_active, -1);
+		return -1;
+	}
 	endflag = 1;
 	printk("OK.\n");
 
@@ -297,8 +282,11 @@ int __init setup_nmi_watchdog(char *str)
 
 	get_option(&str, &nmi);
 
-	if (nmi >= NMI_INVALID)
+	if ((nmi >= NMI_INVALID) || (nmi < NMI_NONE))
 		return 0;
+
+	if ((nmi == NMI_LOCAL_APIC) && (nmi_known_cpu() == 0))
+		return 0;  /* no lapic support */
 	nmi_watchdog = nmi;
 	return 1;
 }
@@ -307,31 +295,30 @@ __setup("nmi_watchdog=", setup_nmi_watchdog);
 
 static void disable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active <= 0)
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
-		break;
-	case X86_VENDOR_INTEL:
-		if (boot_cpu_data.x86 == 15) {
-			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
-			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
-		}
-		break;
-	}
-	nmi_active = -1;
-	/* tell do_nmi() and others that we're not active any more */
-	nmi_watchdog = 0;
+
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 static void enable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_LOCAL_APIC;
-		touch_nmi_watchdog();
-		setup_apic_nmi_watchdog();
-	}
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	/* are we already enabled */
+	if (atomic_read(&nmi_active) != 0)
+		return;
+
+	/* are we lapic aware */
+	if (nmi_known_cpu() <= 0)
+		return;
+
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	touch_nmi_watchdog();
 }
 
 int reserve_lapic_nmi(void)
@@ -363,21 +350,24 @@ void release_lapic_nmi(void)
 
 void disable_timer_nmi_watchdog(void)
 {
-	if ((nmi_watchdog != NMI_IO_APIC) || (nmi_active <= 0))
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
 
 	disable_irq(0);
-	unset_nmi_callback();
-	nmi_active = -1;
-	nmi_watchdog = NMI_NONE;
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 void enable_timer_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_IO_APIC;
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) == 0) {
 		touch_nmi_watchdog();
-		nmi_active = 1;
+		on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
 		enable_irq(0);
 	}
 }
@@ -388,7 +378,7 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
-	nmi_pm_active = nmi_active;
+	nmi_pm_active = atomic_read(&nmi_active);
 	disable_lapic_nmi_watchdog();
 	return 0;
 }
@@ -396,7 +386,7 @@ static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 static int lapic_nmi_resume(struct sys_device *dev)
 {
 	if (nmi_pm_active > 0)
-	enable_lapic_nmi_watchdog();
+		enable_lapic_nmi_watchdog();
 	return 0;
 }
 
@@ -415,7 +405,13 @@ static int __init init_lapic_nmi_sysfs(void)
 {
 	int error;
 
-	if (nmi_active == 0 || nmi_watchdog != NMI_LOCAL_APIC)
+	/* should really be a BUG_ON but b/c this is an
+	 * init call, it just doesn't work.  -dcz
+	 */
+	if (nmi_watchdog != NMI_LOCAL_APIC)
+		return 0;
+
+	if ( atomic_read(&nmi_active) < 0 )
 		return 0;
 
 	error = sysdev_class_register(&nmi_sysclass);
@@ -428,100 +424,232 @@ late_initcall(init_lapic_nmi_sysfs);
 
 #endif	/* CONFIG_PM */
 
+/*
+ * Activate the NMI watchdog via the local APIC.
+ * Original code written by Keith Owens.
+ */
+
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define K7_EVNTSEL_ENABLE	(1 << 22)
+#define K7_EVNTSEL_INT		(1 << 20)
+#define K7_EVNTSEL_OS		(1 << 17)
+#define K7_EVNTSEL_USR		(1 << 16)
+#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
+#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+
 static int setup_k7_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr;
 	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	nmi_perfctr_msr = MSR_K7_PERFCTR0;
-
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	perfctr_msr = MSR_K7_PERFCTR0;
+	evntsel_msr = MSR_K7_EVNTSEL0;
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
 	/* Simulator may not support it */
-	if (checking_wrmsrl(MSR_K7_EVNTSEL0, 0UL))
+	if (checking_wrmsrl(evntsel_msr, 0UL))
 		goto fail2;
-	wrmsrl(MSR_K7_PERFCTR0, 0UL);
+	wrmsrl(perfctr_msr, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
 		| K7_EVNTSEL_USR
 		| K7_NMI_EVENT;
 
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
-	wrmsrl(MSR_K7_PERFCTR0, -((u64)cpu_khz * 1000 / nmi_hz));
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsrl(perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL<<63;
 	return 1;
 fail2:
-	release_evntsel_nmi(MSR_K7_EVNTSEL0);
+	release_evntsel_nmi(evntsel_msr);
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
+static void stop_k7_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
+#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
+#define P4_ESCR_OS		(1<<3)
+#define P4_ESCR_USR		(1<<2)
+#define P4_CCCR_OVF_PMI0	(1<<26)
+#define P4_CCCR_OVF_PMI1	(1<<27)
+#define P4_CCCR_THRESHOLD(N)	((N)<<20)
+#define P4_CCCR_COMPLEMENT	(1<<19)
+#define P4_CCCR_COMPARE		(1<<18)
+#define P4_CCCR_REQUIRED	(3<<16)
+#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
+#define P4_CCCR_ENABLE		(1<<12)
+#define P4_CCCR_OVF 		(1<<31)
+/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
+   CRU_ESCR0 (with any non-null event selector) through a complemented
+   max threshold. [IA32-Vol3, Section 14.9.9] */
 
 static int setup_p4_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr, cccr_msr;
+	unsigned int evntsel, cccr_val;
 	unsigned int misc_enable, dummy;
+	unsigned int ht_num;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	rdmsr(MSR_P4_MISC_ENABLE, misc_enable, dummy);
+	rdmsr(MSR_IA32_MISC_ENABLE, misc_enable, dummy);
 	if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL))
 		return 0;
 
-	nmi_perfctr_msr = MSR_P4_IQ_COUNTER0;
-	nmi_p4_cccr_val = P4_NMI_IQ_CCCR0;
 #ifdef CONFIG_SMP
-	if (smp_num_siblings == 2)
-		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
+	/* detect which hyperthread we are on */
+	if (smp_num_siblings == 2) {
+		unsigned int ebx, apicid;
+
+        	ebx = cpuid_ebx(1);
+	        apicid = (ebx >> 24) & 0xff;
+        	ht_num = apicid & 1;
+	} else
 #endif
+		ht_num = 0;
+
+	/* performance counters are shared resources
+	 * assign each hyperthread its own set
+	 * (re-use the ESCR0 register, seems safe
+	 * and keeps the cccr_val the same)
+	 */
+	if (!ht_num) {
+		/* logical cpu 0 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR0;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR0;
+		cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4);
+	} else {
+		/* logical cpu 1 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR1;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR1;
+		cccr_val = P4_CCCR_OVF_PMI1 | P4_CCCR_ESCR_SELECT(4);
+	}
 
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
-	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
-	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
-	Dprintk("setting P4_IQ_COUNTER0 to 0x%08lx\n", -(cpu_khz * 1000UL / nmi_hz));
-	wrmsrl(MSR_P4_IQ_COUNTER0, -((u64)cpu_khz * 1000 / nmi_hz));
+	evntsel = P4_ESCR_EVENT_SELECT(0x3F)
+	 	| P4_ESCR_OS
+		| P4_ESCR_USR;
+
+	cccr_val |= P4_CCCR_THRESHOLD(15)
+		 | P4_CCCR_COMPLEMENT
+		 | P4_CCCR_COMPARE
+		 | P4_CCCR_REQUIRED;
+
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsr(cccr_msr, cccr_val, 0);
+	wrmsrl(perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
+	cccr_val |= P4_CCCR_ENABLE;
+	wrmsr(cccr_msr, cccr_val, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = cccr_msr;
+	wd->check_bit = 1ULL<<39;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
-void setup_apic_nmi_watchdog(void)
+static void stop_p4_watchdog(void)
 {
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		if (boot_cpu_data.x86 != 15)
-			return;
-		if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
-			return;
-		if (!setup_k7_watchdog())
-			return;
-		break;
-	case X86_VENDOR_INTEL:
-		if (boot_cpu_data.x86 != 15)
-			return;
-		if (!setup_p4_watchdog())
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->cccr_msr, 0, 0);
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+void setup_apic_nmi_watchdog(void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
+				return;
+			if (!setup_k7_watchdog())
+				return;
+			break;
+		case X86_VENDOR_INTEL:
+			if (!setup_p4_watchdog())
+				return;
+			break;
+		default:
 			return;
-		break;
+		}
+	}
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	atomic_inc(&nmi_active);
+}
 
-	default:
-		return;
+static void stop_apic_nmi_watchdog(void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
+				return;
+			stop_k7_watchdog();
+			break;
+		case X86_VENDOR_INTEL:
+			stop_p4_watchdog();
+			break;
+		default:
+			return;
+		}
 	}
-	lapic_nmi_owner = LAPIC_NMI_WATCHDOG;
-	nmi_active = 1;
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	atomic_dec(&nmi_active);
 }
 
 /*
@@ -558,50 +686,70 @@ void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	u64 dummy;
+
+	/* check for other users first */
+	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
+			== NOTIFY_STOP) {
+		touched = 1;
+	}
 
 	sum = read_pda(apic_timer_irqs);
 	if (__get_cpu_var(nmi_touch)) {
 		__get_cpu_var(nmi_touch) = 0;
 		touched = 1;
 	}
+
 #ifdef CONFIG_X86_MCE
 	/* Could check oops_in_progress here too, but it's safer
 	   not too */
 	if (atomic_read(&mce_entry) > 0)
 		touched = 1;
 #endif
+	/* if the apic timer isn't firing, this cpu isn't doing much */
 	if (!touched && __get_cpu_var(last_irq_sum) == sum) {
 		/*
 		 * Ayiee, looks like this CPU is stuck ...
 		 * wait a few IRQs (5 seconds) before doing the oops ...
 		 */
 		local_inc(&__get_cpu_var(alert_counter));
-		if (local_read(&__get_cpu_var(alert_counter)) == 5*nmi_hz) {
-			if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
-							== NOTIFY_STOP) {
-				local_set(&__get_cpu_var(alert_counter), 0);
-				return;
-			}
+		if (local_read(&__get_cpu_var(alert_counter)) == 5*nmi_hz)
 			die_nmi("NMI Watchdog detected LOCKUP on CPU %d\n", regs);
-		}
 	} else {
 		__get_cpu_var(last_irq_sum) = sum;
 		local_set(&__get_cpu_var(alert_counter), 0);
 	}
-	if (nmi_perfctr_msr) {
- 		if (nmi_perfctr_msr == MSR_P4_IQ_COUNTER0) {
- 			/*
- 			 * P4 quirks:
- 			 * - An overflown perfctr will assert its interrupt
- 			 *   until the OVF flag in its CCCR is cleared.
- 			 * - LVTPC is masked on interrupt and must be
- 			 *   unmasked by the LVTPC handler.
- 			 */
- 			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
- 			apic_write(APIC_LVTPC, APIC_DM_NMI);
- 		}
-		wrmsrl(nmi_perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
+
+	/* see if the nmi watchdog went off */
+	if (wd->enabled) {
+		if (nmi_watchdog == NMI_LOCAL_APIC) {
+			rdmsrl(wd->perfctr_msr, dummy);
+			if (dummy & wd->check_bit){
+				/* this wasn't a watchdog timer interrupt */
+				goto done;
+			}
+
+			/* only Intel uses the cccr msr */
+	 		if (wd->cccr_msr != 0) {
+	 			/*
+	 			 * P4 quirks:
+	 			 * - An overflown perfctr will assert its interrupt
+	 			 *   until the OVF flag in its CCCR is cleared.
+	 			 * - LVTPC is masked on interrupt and must be
+	 			 *   unmasked by the LVTPC handler.
+	 			 */
+				rdmsrl(wd->cccr_msr, dummy);
+				dummy &= ~P4_CCCR_OVF;
+	 			wrmsrl(wd->cccr_msr, dummy);
+	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
+	 		}
+			/* start the cycle over again */
+			wrmsrl(wd->perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
+		}
 	}
+done:
+	return;
 }
 
 static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 62a784cb8f0c..5918136fd853 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -63,7 +63,7 @@ extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
-extern void setup_apic_nmi_watchdog (void);
+extern void setup_apic_nmi_watchdog (void *);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
@@ -73,6 +73,7 @@ extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 extern void nmi_watchdog_default(void);
 extern int setup_nmi_watchdog(char *);
 
+extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
 #define NMI_DEFAULT	-1
 #define NMI_NONE	0
-- 
cgit v1.2.3


From b7471c6da94d30d3deadc55986cc38d1ff57f9ca Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] i386: Add SMP support on i386 to reservation framework

This patch includes the changes to make the nmi watchdog on i386 SMP aware.
A bunch of code was moved around to make it simpler to read.  In addition,
it is now possible to determine if a particular NMI was the result of the
watchdog or not.  This feature allows the kernel to filter out unknown NMIs
easier.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/apic.c            |   3 +-
 arch/i386/kernel/nmi.c             | 537 ++++++++++++++++++++++++-------------
 arch/i386/kernel/traps.c           |   2 +-
 arch/i386/oprofile/nmi_timer_int.c |   4 +-
 include/asm-i386/nmi.h             |   5 +-
 5 files changed, 360 insertions(+), 191 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 8c844d07862f..1a34fc57800b 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -586,8 +586,7 @@ void __devinit setup_local_APIC(void)
 			printk("No ESR for 82489DX.\n");
 	}
 
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		setup_apic_nmi_watchdog();
+	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
 }
 
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 5d58dfeacd59..d88004343034 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -24,16 +24,10 @@
 
 #include <asm/smp.h>
 #include <asm/nmi.h>
+#include <asm/kdebug.h>
 
 #include "mach_traps.h"
 
-unsigned int nmi_watchdog = NMI_NONE;
-extern int unknown_nmi_panic;
-static unsigned int nmi_hz = HZ;
-static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
-static unsigned int nmi_p4_cccr_val;
-extern void show_registers(struct pt_regs *regs);
-
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
  * - different performance counters/ event selection may be reserved for
@@ -63,51 +57,31 @@ static unsigned int lapic_nmi_owner;
 #define LAPIC_NMI_RESERVED	(1<<1)
 
 /* nmi_active:
- * +1: the lapic NMI watchdog is active, but can be disabled
- *  0: the lapic NMI watchdog has not been set up, and cannot
+ * >0: the lapic NMI watchdog is active, but can be disabled
+ * <0: the lapic NMI watchdog has not been set up, and cannot
  *     be enabled
- * -1: the lapic NMI watchdog is disabled, but can be enabled
+ *  0: the lapic NMI watchdog is disabled, but can be enabled
  */
-int nmi_active;
+atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
 
-#define K7_EVNTSEL_ENABLE	(1 << 22)
-#define K7_EVNTSEL_INT		(1 << 20)
-#define K7_EVNTSEL_OS		(1 << 17)
-#define K7_EVNTSEL_USR		(1 << 16)
-#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
-#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+unsigned int nmi_watchdog = NMI_DEFAULT;
+static unsigned int nmi_hz = HZ;
 
-#define P6_EVNTSEL0_ENABLE	(1 << 22)
-#define P6_EVNTSEL_INT		(1 << 20)
-#define P6_EVNTSEL_OS		(1 << 17)
-#define P6_EVNTSEL_USR		(1 << 16)
-#define P6_EVENT_CPU_CLOCKS_NOT_HALTED	0x79
-#define P6_NMI_EVENT		P6_EVENT_CPU_CLOCKS_NOT_HALTED
+struct nmi_watchdog_ctlblk {
+	int enabled;
+	u64 check_bit;
+	unsigned int cccr_msr;
+	unsigned int perfctr_msr;  /* the MSR to reset in NMI handler */
+	unsigned int evntsel_msr;  /* the MSR to select the events to handle */
+};
+static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
-#define MSR_P4_MISC_ENABLE	0x1A0
-#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
-#define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
-#define MSR_P4_PERFCTR0		0x300
-#define MSR_P4_CCCR0		0x360
-#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
-#define P4_ESCR_OS		(1<<3)
-#define P4_ESCR_USR		(1<<2)
-#define P4_CCCR_OVF_PMI0	(1<<26)
-#define P4_CCCR_OVF_PMI1	(1<<27)
-#define P4_CCCR_THRESHOLD(N)	((N)<<20)
-#define P4_CCCR_COMPLEMENT	(1<<19)
-#define P4_CCCR_COMPARE		(1<<18)
-#define P4_CCCR_REQUIRED	(3<<16)
-#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
-#define P4_CCCR_ENABLE		(1<<12)
-/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
-   CRU_ESCR0 (with any non-null event selector) through a complemented
-   max threshold. [IA32-Vol3, Section 14.9.9] */
-#define MSR_P4_IQ_COUNTER0	0x30C
-#define P4_NMI_CRU_ESCR0	(P4_ESCR_EVENT_SELECT(0x3F)|P4_ESCR_OS|P4_ESCR_USR)
-#define P4_NMI_IQ_CCCR0	\
-	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
-	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
+/* local prototypes */
+static void stop_apic_nmi_watchdog(void *unused);
+static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
+
+extern void show_registers(struct pt_regs *regs);
+extern int unknown_nmi_panic;
 
 /* converts an msr to an appropriate reservation bit */
 static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
@@ -208,6 +182,17 @@ void release_evntsel_nmi(unsigned int msr)
 	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]);
 }
 
+static __cpuinit inline int nmi_known_cpu(void)
+{
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+	case X86_VENDOR_INTEL:
+		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+	}
+	return 0;
+}
+
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -234,7 +219,10 @@ static int __init check_nmi_watchdog(void)
 	unsigned int *prev_nmi_count;
 	int cpu;
 
-	if (nmi_watchdog == NMI_NONE)
+	if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT))
+		return 0;
+
+	if (!atomic_read(&nmi_active))
 		return 0;
 
 	prev_nmi_count = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
@@ -258,18 +246,22 @@ static int __init check_nmi_watchdog(void)
 		if (!cpu_isset(cpu, cpu_callin_map))
 			continue;
 #endif
+		if (!per_cpu(nmi_watchdog_ctlblk, cpu).enabled)
+			continue;
 		if (nmi_count(cpu) - prev_nmi_count[cpu] <= 5) {
-			endflag = 1;
 			printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
 				cpu,
 				prev_nmi_count[cpu],
 				nmi_count(cpu));
-			nmi_active = 0;
-			lapic_nmi_owner &= ~LAPIC_NMI_WATCHDOG;
-			kfree(prev_nmi_count);
-			return -1;
+			per_cpu(nmi_watchdog_ctlblk, cpu).enabled = 0;
+			atomic_dec(&nmi_active);
 		}
 	}
+	if (!atomic_read(&nmi_active)) {
+		kfree(prev_nmi_count);
+		atomic_set(&nmi_active, -1);
+		return -1;
+	}
 	endflag = 1;
 	printk("OK.\n");
 
@@ -290,31 +282,16 @@ static int __init setup_nmi_watchdog(char *str)
 
 	get_option(&str, &nmi);
 
-	if (nmi >= NMI_INVALID)
+	if ((nmi >= NMI_INVALID) || (nmi < NMI_NONE))
 		return 0;
-	if (nmi == NMI_NONE)
-		nmi_watchdog = nmi;
 	/*
 	 * If any other x86 CPU has a local APIC, then
 	 * please test the NMI stuff there and send me the
 	 * missing bits. Right now Intel P6/P4 and AMD K7 only.
 	 */
-	if ((nmi == NMI_LOCAL_APIC) &&
-			(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
-			(boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15))
-		nmi_watchdog = nmi;
-	if ((nmi == NMI_LOCAL_APIC) &&
-			(boot_cpu_data.x86_vendor == X86_VENDOR_AMD) &&
-	  		(boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15))
-		nmi_watchdog = nmi;
-	/*
-	 * We can enable the IO-APIC watchdog
-	 * unconditionally.
-	 */
-	if (nmi == NMI_IO_APIC) {
-		nmi_active = 1;
-		nmi_watchdog = nmi;
-	}
+	if ((nmi == NMI_LOCAL_APIC) && (nmi_known_cpu() == 0))
+		return 0;  /* no lapic support */
+	nmi_watchdog = nmi;
 	return 1;
 }
 
@@ -322,41 +299,30 @@ __setup("nmi_watchdog=", setup_nmi_watchdog);
 
 static void disable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active <= 0)
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
-		break;
-	case X86_VENDOR_INTEL:
-		switch (boot_cpu_data.x86) {
-		case 6:
-			if (boot_cpu_data.x86_model > 0xd)
-				break;
 
-			wrmsr(MSR_P6_EVNTSEL0, 0, 0);
-			break;
-		case 15:
-			if (boot_cpu_data.x86_model > 0x4)
-				break;
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
 
-			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
-			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
-			break;
-		}
-		break;
-	}
-	nmi_active = -1;
-	/* tell do_nmi() and others that we're not active any more */
-	nmi_watchdog = 0;
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 static void enable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_LOCAL_APIC;
-		setup_apic_nmi_watchdog();
-	}
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	/* are we already enabled */
+	if (atomic_read(&nmi_active) != 0)
+		return;
+
+	/* are we lapic aware */
+	if (nmi_known_cpu() <= 0)
+		return;
+
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	touch_nmi_watchdog();
 }
 
 int reserve_lapic_nmi(void)
@@ -388,20 +354,25 @@ void release_lapic_nmi(void)
 
 void disable_timer_nmi_watchdog(void)
 {
-	if ((nmi_watchdog != NMI_IO_APIC) || (nmi_active <= 0))
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
 
-	unset_nmi_callback();
-	nmi_active = -1;
-	nmi_watchdog = NMI_NONE;
+	disable_irq(0);
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 void enable_timer_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_IO_APIC;
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) == 0) {
 		touch_nmi_watchdog();
-		nmi_active = 1;
+		on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+		enable_irq(0);
 	}
 }
 
@@ -411,7 +382,7 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
-	nmi_pm_active = nmi_active;
+	nmi_pm_active = atomic_read(&nmi_active);
 	disable_lapic_nmi_watchdog();
 	return 0;
 }
@@ -439,7 +410,13 @@ static int __init init_lapic_nmi_sysfs(void)
 {
 	int error;
 
-	if (nmi_active == 0 || nmi_watchdog != NMI_LOCAL_APIC)
+	/* should really be a BUG_ON but b/c this is an
+	 * init call, it just doesn't work.  -dcz
+	 */
+	if (nmi_watchdog != NMI_LOCAL_APIC)
+		return 0;
+
+	if ( atomic_read(&nmi_active) < 0 )
 		return 0;
 
 	error = sysdev_class_register(&nmi_sysclass);
@@ -457,143 +434,312 @@ late_initcall(init_lapic_nmi_sysfs);
  * Original code written by Keith Owens.
  */
 
-static void write_watchdog_counter(const char *descr)
+static void write_watchdog_counter(unsigned int perfctr_msr, const char *descr)
 {
 	u64 count = (u64)cpu_khz * 1000;
 
 	do_div(count, nmi_hz);
 	if(descr)
 		Dprintk("setting %s to -0x%08Lx\n", descr, count);
-	wrmsrl(nmi_perfctr_msr, 0 - count);
+	wrmsrl(perfctr_msr, 0 - count);
 }
 
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define K7_EVNTSEL_ENABLE	(1 << 22)
+#define K7_EVNTSEL_INT		(1 << 20)
+#define K7_EVNTSEL_OS		(1 << 17)
+#define K7_EVNTSEL_USR		(1 << 16)
+#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
+#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+
 static int setup_k7_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr;
 	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	nmi_perfctr_msr = MSR_K7_PERFCTR0;
-
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	perfctr_msr = MSR_K7_PERFCTR0;
+	evntsel_msr = MSR_K7_EVNTSEL0;
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
-	wrmsrl(MSR_K7_PERFCTR0, 0UL);
+	wrmsrl(perfctr_msr, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
 		| K7_EVNTSEL_USR
 		| K7_NMI_EVENT;
 
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
-	write_watchdog_counter("K7_PERFCTR0");
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "K7_PERFCTR0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL<<63;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
+static void stop_k7_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+#define P6_EVNTSEL0_ENABLE	(1 << 22)
+#define P6_EVNTSEL_INT		(1 << 20)
+#define P6_EVNTSEL_OS		(1 << 17)
+#define P6_EVNTSEL_USR		(1 << 16)
+#define P6_EVENT_CPU_CLOCKS_NOT_HALTED	0x79
+#define P6_NMI_EVENT		P6_EVENT_CPU_CLOCKS_NOT_HALTED
+
 static int setup_p6_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr;
 	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	nmi_perfctr_msr = MSR_P6_PERFCTR0;
-
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	perfctr_msr = MSR_P6_PERFCTR0;
+	evntsel_msr = MSR_P6_EVNTSEL0;
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_P6_EVNTSEL0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
+	wrmsrl(perfctr_msr, 0UL);
+
 	evntsel = P6_EVNTSEL_INT
 		| P6_EVNTSEL_OS
 		| P6_EVNTSEL_USR
 		| P6_NMI_EVENT;
 
-	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
-	write_watchdog_counter("P6_PERFCTR0");
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "P6_PERFCTR0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= P6_EVNTSEL0_ENABLE;
-	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL<<39;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
+static void stop_p6_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
+#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
+#define P4_ESCR_OS		(1<<3)
+#define P4_ESCR_USR		(1<<2)
+#define P4_CCCR_OVF_PMI0	(1<<26)
+#define P4_CCCR_OVF_PMI1	(1<<27)
+#define P4_CCCR_THRESHOLD(N)	((N)<<20)
+#define P4_CCCR_COMPLEMENT	(1<<19)
+#define P4_CCCR_COMPARE		(1<<18)
+#define P4_CCCR_REQUIRED	(3<<16)
+#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
+#define P4_CCCR_ENABLE		(1<<12)
+#define P4_CCCR_OVF 		(1<<31)
+/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
+   CRU_ESCR0 (with any non-null event selector) through a complemented
+   max threshold. [IA32-Vol3, Section 14.9.9] */
+
 static int setup_p4_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr, cccr_msr;
+	unsigned int evntsel, cccr_val;
 	unsigned int misc_enable, dummy;
+	unsigned int ht_num;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	rdmsr(MSR_P4_MISC_ENABLE, misc_enable, dummy);
+	rdmsr(MSR_IA32_MISC_ENABLE, misc_enable, dummy);
 	if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL))
 		return 0;
 
-	nmi_perfctr_msr = MSR_P4_IQ_COUNTER0;
-	nmi_p4_cccr_val = P4_NMI_IQ_CCCR0;
 #ifdef CONFIG_SMP
-	if (smp_num_siblings == 2)
-		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
+	/* detect which hyperthread we are on */
+	if (smp_num_siblings == 2) {
+		unsigned int ebx, apicid;
+
+        	ebx = cpuid_ebx(1);
+	        apicid = (ebx >> 24) & 0xff;
+        	ht_num = apicid & 1;
+	} else
 #endif
+		ht_num = 0;
+
+	/* performance counters are shared resources
+	 * assign each hyperthread its own set
+	 * (re-use the ESCR0 register, seems safe
+	 * and keeps the cccr_val the same)
+	 */
+	if (!ht_num) {
+		/* logical cpu 0 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR0;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR0;
+		cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4);
+	} else {
+		/* logical cpu 1 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR1;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR1;
+		cccr_val = P4_CCCR_OVF_PMI1 | P4_CCCR_ESCR_SELECT(4);
+	}
 
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
-	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
-	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
-	write_watchdog_counter("P4_IQ_COUNTER0");
+	evntsel = P4_ESCR_EVENT_SELECT(0x3F)
+	 	| P4_ESCR_OS
+		| P4_ESCR_USR;
+
+	cccr_val |= P4_CCCR_THRESHOLD(15)
+		 | P4_CCCR_COMPLEMENT
+		 | P4_CCCR_COMPARE
+		 | P4_CCCR_REQUIRED;
+
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsr(cccr_msr, cccr_val, 0);
+	write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
+	cccr_val |= P4_CCCR_ENABLE;
+	wrmsr(cccr_msr, cccr_val, 0);
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = cccr_msr;
+	wd->check_bit = 1ULL<<39;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
-void setup_apic_nmi_watchdog (void)
+static void stop_p4_watchdog(void)
 {
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
-			return;
-		if (!setup_k7_watchdog())
-			return;
-		break;
-	case X86_VENDOR_INTEL:
-		switch (boot_cpu_data.x86) {
-		case 6:
-			if (boot_cpu_data.x86_model > 0xd)
-				return;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->cccr_msr, 0, 0);
+	wrmsr(wd->evntsel_msr, 0, 0);
 
-			if(!setup_p6_watchdog())
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+void setup_apic_nmi_watchdog (void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
 				return;
-			break;
-		case 15:
-			if (boot_cpu_data.x86_model > 0x4)
+			if (!setup_k7_watchdog())
 				return;
+			break;
+		case X86_VENDOR_INTEL:
+			switch (boot_cpu_data.x86) {
+			case 6:
+				if (boot_cpu_data.x86_model > 0xd)
+					return;
+
+				if (!setup_p6_watchdog())
+					return;
+				break;
+			case 15:
+				if (boot_cpu_data.x86_model > 0x4)
+					return;
 
-			if (!setup_p4_watchdog())
+				if (!setup_p4_watchdog())
+					return;
+				break;
+			default:
 				return;
+			}
+			break;
+		default:
+			return;
+		}
+	}
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	atomic_inc(&nmi_active);
+}
+
+static void stop_apic_nmi_watchdog(void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			stop_k7_watchdog();
+			break;
+		case X86_VENDOR_INTEL:
+			switch (boot_cpu_data.x86) {
+			case 6:
+				if (boot_cpu_data.x86_model > 0xd)
+					break;
+				stop_p6_watchdog();
+				break;
+			case 15:
+				if (boot_cpu_data.x86_model > 0x4)
+					break;
+				stop_p4_watchdog();
+				break;
+			}
 			break;
 		default:
 			return;
 		}
-		break;
-	default:
-		return;
 	}
-	lapic_nmi_owner = LAPIC_NMI_WATCHDOG;
-	nmi_active = 1;
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	atomic_dec(&nmi_active);
 }
 
 /*
@@ -635,7 +781,7 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-void nmi_watchdog_tick (struct pt_regs * regs)
+void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 {
 
 	/*
@@ -644,11 +790,21 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 	 * smp_processor_id().
 	 */
 	unsigned int sum;
+	int touched = 0;
 	int cpu = smp_processor_id();
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	u64 dummy;
+
+	/* check for other users first */
+	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
+			== NOTIFY_STOP) {
+		touched = 1;
+	}
 
 	sum = per_cpu(irq_stat, cpu).apic_timer_irqs;
 
-	if (last_irq_sums[cpu] == sum) {
+	/* if the apic timer isn't firing, this cpu isn't doing much */
+	if (!touched && last_irq_sums[cpu] == sum) {
 		/*
 		 * Ayiee, looks like this CPU is stuck ...
 		 * wait a few IRQs (5 seconds) before doing the oops ...
@@ -663,26 +819,41 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 		last_irq_sums[cpu] = sum;
 		alert_counter[cpu] = 0;
 	}
-	if (nmi_perfctr_msr) {
-		if (nmi_perfctr_msr == MSR_P4_IQ_COUNTER0) {
-			/*
-			 * P4 quirks:
-			 * - An overflown perfctr will assert its interrupt
-			 *   until the OVF flag in its CCCR is cleared.
-			 * - LVTPC is masked on interrupt and must be
-			 *   unmasked by the LVTPC handler.
-			 */
-			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
-			apic_write(APIC_LVTPC, APIC_DM_NMI);
-		}
-		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0) {
-			/* Only P6 based Pentium M need to re-unmask
-			 * the apic vector but it doesn't hurt
-			 * other P6 variant */
-			apic_write(APIC_LVTPC, APIC_DM_NMI);
+	/* see if the nmi watchdog went off */
+	if (wd->enabled) {
+		if (nmi_watchdog == NMI_LOCAL_APIC) {
+			rdmsrl(wd->perfctr_msr, dummy);
+			if (dummy & wd->check_bit){
+				/* this wasn't a watchdog timer interrupt */
+				goto done;
+			}
+
+			/* only Intel P4 uses the cccr msr */
+	 		if (wd->cccr_msr != 0) {
+	 			/*
+	 			 * P4 quirks:
+	 			 * - An overflown perfctr will assert its interrupt
+	 			 *   until the OVF flag in its CCCR is cleared.
+	 			 * - LVTPC is masked on interrupt and must be
+	 			 *   unmasked by the LVTPC handler.
+	 			 */
+				rdmsrl(wd->cccr_msr, dummy);
+				dummy &= ~P4_CCCR_OVF;
+	 			wrmsrl(wd->cccr_msr, dummy);
+	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
+	 		}
+			else if (wd->perfctr_msr == MSR_P6_PERFCTR0) {
+				/* Only P6 based Pentium M need to re-unmask
+				 * the apic vector but it doesn't hurt
+				 * other P6 variant */
+				apic_write(APIC_LVTPC, APIC_DM_NMI);
+			}
+			/* start the cycle over again */
+			write_watchdog_counter(wd->perfctr_msr, NULL);
 		}
-		write_watchdog_counter(NULL);
 	}
+done:
+	return;
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 7e9edafffd8a..3a07b2677e2a 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -724,7 +724,7 @@ static void default_do_nmi(struct pt_regs * regs)
 		 * so it must be the NMI watchdog.
 		 */
 		if (nmi_watchdog) {
-			nmi_watchdog_tick(regs);
+			nmi_watchdog_tick(regs, reason);
 			return;
 		}
 #endif
diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c
index 930a1127bb30..a33a73bb502d 100644
--- a/arch/i386/oprofile/nmi_timer_int.c
+++ b/arch/i386/oprofile/nmi_timer_int.c
@@ -42,9 +42,7 @@ static void timer_stop(void)
 
 int __init op_nmi_timer_init(struct oprofile_operations * ops)
 {
-	extern int nmi_active;
-
-	if (nmi_active <= 0)
+	if (atomic_read(&nmi_active) <= 0)
 		return -ENODEV;
 
 	ops->start = timer_start;
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 27fc9e6f630e..4cda6801ecb8 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -32,13 +32,14 @@ extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
-extern void setup_apic_nmi_watchdog (void);
+extern void setup_apic_nmi_watchdog (void *);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
-extern void nmi_watchdog_tick (struct pt_regs * regs);
+extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
+extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
 #define NMI_DEFAULT     -1
 #define NMI_NONE	0
-- 
cgit v1.2.3


From 3adbbcce9a49b900d4cc118cdccfdefa78bf1afb Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] x86: Cleanup NMI interrupt path

This patch cleans up the NMI interrupt path.  Instead of being gated by if
the 'nmi callback' is set, the interrupt handler now calls everyone who is
registered on the die_chain and additionally checks the nmi watchdog,
reseting it if enabled.  This allows more subsystems to hook into the NMI if
they need to (without being block by set_nmi_callback).

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c     | 16 +++++++++++++---
 arch/i386/kernel/traps.c   | 24 +++++++++++-------------
 arch/x86_64/kernel/nmi.c   | 26 +++++++++++++++++++-------
 arch/x86_64/kernel/traps.c |  8 ++++----
 include/asm-i386/nmi.h     |  2 +-
 include/asm-x86_64/nmi.h   | 10 +++++++++-
 6 files changed, 57 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index d88004343034..bd96ea4f2942 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -781,7 +781,7 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
+int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 {
 
 	/*
@@ -794,10 +794,12 @@ void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 	int cpu = smp_processor_id();
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 	u64 dummy;
+	int rc=0;
 
 	/* check for other users first */
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
 			== NOTIFY_STOP) {
+		rc = 1;
 		touched = 1;
 	}
 
@@ -850,10 +852,18 @@ void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 			}
 			/* start the cycle over again */
 			write_watchdog_counter(wd->perfctr_msr, NULL);
-		}
+			rc = 1;
+		} else if (nmi_watchdog == NMI_IO_APIC) {
+			/* don't know how to accurately check for this.
+			 * just assume it was a watchdog timer interrupt
+			 * This matches the old behaviour.
+			 */
+			rc = 1;
+		} else
+			printk(KERN_WARNING "Unknown enabled NMI hardware?!\n");
 	}
 done:
-	return;
+	return rc;
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 3a07b2677e2a..282f0bd40dfd 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -706,6 +706,13 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	do_exit(SIGSEGV);
 }
 
+static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
+{
+	return 0;
+}
+
+static nmi_callback_t nmi_callback = dummy_nmi_callback;
+
 static void default_do_nmi(struct pt_regs * regs)
 {
 	unsigned char reason = 0;
@@ -723,12 +730,11 @@ static void default_do_nmi(struct pt_regs * regs)
 		 * Ok, so this is none of the documented NMI sources,
 		 * so it must be the NMI watchdog.
 		 */
-		if (nmi_watchdog) {
-			nmi_watchdog_tick(regs, reason);
+		if (nmi_watchdog_tick(regs, reason))
 			return;
-		}
 #endif
-		unknown_nmi_error(reason, regs);
+		if (!rcu_dereference(nmi_callback)(regs, smp_processor_id()))
+			unknown_nmi_error(reason, regs);
 		return;
 	}
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
@@ -744,13 +750,6 @@ static void default_do_nmi(struct pt_regs * regs)
 	reassert_nmi();
 }
 
-static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
-{
-	return 0;
-}
- 
-static nmi_callback_t nmi_callback = dummy_nmi_callback;
- 
 fastcall void do_nmi(struct pt_regs * regs, long error_code)
 {
 	int cpu;
@@ -761,8 +760,7 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
 
 	++nmi_count(cpu);
 
-	if (!rcu_dereference(nmi_callback)(regs, cpu))
-		default_do_nmi(regs);
+	default_do_nmi(regs);
 
 	nmi_exit();
 }
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index d42374a952d7..f6b881b23a70 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -682,16 +682,18 @@ void touch_nmi_watchdog (void)
  	touch_softlockup_watchdog();
 }
 
-void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
+int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 	u64 dummy;
+	int rc=0;
 
 	/* check for other users first */
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
 			== NOTIFY_STOP) {
+		rc = 1;
 		touched = 1;
 	}
 
@@ -746,10 +748,18 @@ void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 	 		}
 			/* start the cycle over again */
 			wrmsrl(wd->perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
-		}
+			rc = 1;
+		} else 	if (nmi_watchdog == NMI_IO_APIC) {
+			/* don't know how to accurately check for this.
+			 * just assume it was a watchdog timer interrupt
+			 * This matches the old behaviour.
+			 */
+			rc = 1;
+		} else
+			printk(KERN_WARNING "Unknown enabled NMI hardware?!\n");
 	}
 done:
-	return;
+	return rc;
 }
 
 static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
@@ -761,15 +771,17 @@ static nmi_callback_t nmi_callback = dummy_nmi_callback;
  
 asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 {
-	int cpu = safe_smp_processor_id();
-
 	nmi_enter();
 	add_pda(__nmi_count,1);
-	if (!rcu_dereference(nmi_callback)(regs, cpu))
-		default_do_nmi(regs);
+	default_do_nmi(regs);
 	nmi_exit();
 }
 
+int do_nmi_callback(struct pt_regs * regs, int cpu)
+{
+	return rcu_dereference(nmi_callback)(regs, cpu);
+}
+
 void set_nmi_callback(nmi_callback_t callback)
 {
 	vmalloc_sync_all();
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index b1249774d1e8..42bc070fdf11 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -781,12 +781,12 @@ asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 		 * Ok, so this is none of the documented NMI sources,
 		 * so it must be the NMI watchdog.
 		 */
-		if (nmi_watchdog > 0) {
-			nmi_watchdog_tick(regs,reason);
+		if (nmi_watchdog_tick(regs,reason))
 			return;
-		}
+		if (!do_nmi_callback(regs,cpu))
 #endif
-		unknown_nmi_error(reason, regs);
+			unknown_nmi_error(reason, regs);
+
 		return;
 	}
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 4cda6801ecb8..da0e0b4e9139 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -37,7 +37,7 @@ extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
-extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
+extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
 extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 5918136fd853..8f02a2a416e6 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -26,6 +26,14 @@ void set_nmi_callback(nmi_callback_t callback);
  */
 void unset_nmi_callback(void);
 
+/**
+ * do_nmi_callback
+ *
+ * Check to see if a callback exists and execute it.  Return 1
+ * if the handler exists and was handled successfully.
+ */
+int do_nmi_callback(struct pt_regs *regs, int cpu);
+
 #ifdef CONFIG_PM
  
 /** Replace the PM callback routine for NMI. */
@@ -68,7 +76,7 @@ extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
-extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
+extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
 extern void nmi_watchdog_default(void);
 extern int setup_nmi_watchdog(char *);
-- 
cgit v1.2.3


From 1d001df19d5323e642ba8ac821c713675ebccd82 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] Add TIF_RESTORE_SIGMASK

We need TIF_RESTORE_SIGMASK in order to support ppoll() and pselect()
system calls. This patch originally came from Andi, and was based
heavily on David Howells' implementation of same on i386. I fixed a typo
which was causing do_signal() to use the wrong signal mask.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32_signal.c   | 28 ++++++--------
 arch/x86_64/kernel/signal.c      | 82 +++++++++++++++++-----------------------
 include/asm-x86_64/signal.h      |  4 --
 include/asm-x86_64/thread_info.h |  2 +
 include/asm-x86_64/unistd.h      |  1 +
 5 files changed, 49 insertions(+), 68 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c
index 25e5ca22204c..549de439fb2d 100644
--- a/arch/x86_64/ia32/ia32_signal.c
+++ b/arch/x86_64/ia32/ia32_signal.c
@@ -113,25 +113,19 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
 }
 
 asmlinkage long
-sys32_sigsuspend(int history0, int history1, old_sigset_t mask,
-		 struct pt_regs *regs)
+sys32_sigsuspend(int history0, int history1, old_sigset_t mask)
 {
-	sigset_t saveset;
-
 	mask &= _BLOCKABLE;
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	siginitset(&current->blocked, mask);
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs->rax = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(regs, &saveset))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 asmlinkage long
@@ -508,11 +502,11 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return 1;
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
-	return 0;
+	return -EFAULT;
 }
 
 int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
@@ -595,7 +589,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	regs->ss = __USER32_DS; 
 
 	set_fs(USER_DS);
-    regs->eflags &= ~TF_MASK;
+	regs->eflags &= ~TF_MASK;
     if (test_thread_flag(TIF_SINGLESTEP))
         ptrace_notify(SIGTRAP);
 
@@ -604,9 +598,9 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return 1;
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
-	return 0;
+	return -EFAULT;
 }
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 28161170fb0a..7f58bc9a056d 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -37,37 +37,6 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 int ia32_setup_frame(int sig, struct k_sigaction *ka,
             sigset_t *set, struct pt_regs * regs); 
 
-asmlinkage long
-sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize, struct pt_regs *regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-#ifdef DEBUG_SIG
-	printk("rt_sigsuspend savset(%lx) newset(%lx) regs(%p) rip(%lx)\n",
-		saveset, newset, regs, regs->rip);
-#endif 
-	regs->rax = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(regs, &saveset))
-			return -EINTR;
-	}
-}
-
 asmlinkage long
 sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 		struct pt_regs *regs)
@@ -341,11 +310,11 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return 1;
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
-	return 0;
+	return -EFAULT;
 }
 
 /*
@@ -408,7 +377,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 #endif
 	ret = setup_rt_frame(sig, ka, info, oldset, regs);
 
-	if (ret) {
+	if (ret == 0) {
 		spin_lock_irq(&current->sighand->siglock);
 		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
 		if (!(ka->sa.sa_flags & SA_NODEFER))
@@ -425,11 +394,12 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
  * want to handle. Thus you cannot kill init even with a SIGKILL even by
  * mistake.
  */
-int do_signal(struct pt_regs *regs, sigset_t *oldset)
+static void do_signal(struct pt_regs *regs)
 {
 	struct k_sigaction ka;
 	siginfo_t info;
 	int signr;
+	sigset_t *oldset;
 
 	/*
 	 * We want the common case to go fast, which
@@ -438,9 +408,11 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
@@ -454,30 +426,46 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 			set_debugreg(current->thread.debugreg7, 7);
 
 		/* Whee!  Actually deliver the signal.  */
-		return handle_signal(signr, &info, &ka, oldset, regs);
+		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
+		return;
 	}
 
 	/* Did we come from a system call? */
 	if ((long)regs->orig_rax >= 0) {
 		/* Restart the system call - no handlers present */
 		long res = regs->rax;
-		if (res == -ERESTARTNOHAND ||
-		    res == -ERESTARTSYS ||
-		    res == -ERESTARTNOINTR) {
+		switch (res) {
+		case -ERESTARTNOHAND:
+		case -ERESTARTSYS:
+		case -ERESTARTNOINTR:
 			regs->rax = regs->orig_rax;
 			regs->rip -= 2;
-		}
-		if (regs->rax == (unsigned long)-ERESTART_RESTARTBLOCK) {
+			break;
+		case -ERESTART_RESTARTBLOCK:
 			regs->rax = test_thread_flag(TIF_IA32) ?
 					__NR_ia32_restart_syscall :
 					__NR_restart_syscall;
 			regs->rip -= 2;
+			break;
 		}
 	}
-	return 0;
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	   back. */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
 }
 
-void do_notify_resume(struct pt_regs *regs, sigset_t *oldset, __u32 thread_info_flags)
+void
+do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 {
 #ifdef DEBUG_SIG
 	printk("do_notify_resume flags:%x rip:%lx rsp:%lx caller:%lx pending:%lx\n",
@@ -491,8 +479,8 @@ void do_notify_resume(struct pt_regs *regs, sigset_t *oldset, __u32 thread_info_
 	}
 
 	/* deal with pending signal delivery */
-	if (thread_info_flags & _TIF_SIGPENDING)
-		do_signal(regs,oldset);
+	if (thread_info_flags & (_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK))
+		do_signal(regs);
 }
 
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
diff --git a/include/asm-x86_64/signal.h b/include/asm-x86_64/signal.h
index 3ede2a61973a..4581f978b299 100644
--- a/include/asm-x86_64/signal.h
+++ b/include/asm-x86_64/signal.h
@@ -24,10 +24,6 @@ typedef struct {
 } sigset_t;
 
 
-struct pt_regs; 
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
-
-
 #else
 /* Here we must cater to libcs that poke about in kernel headers.  */
 
diff --git a/include/asm-x86_64/thread_info.h b/include/asm-x86_64/thread_info.h
index 2029b00351f3..790c512a4369 100644
--- a/include/asm-x86_64/thread_info.h
+++ b/include/asm-x86_64/thread_info.h
@@ -114,6 +114,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_IRET		5	/* force IRET */
 #define TIF_SYSCALL_AUDIT	7	/* syscall auditing active */
 #define TIF_SECCOMP		8	/* secure computing */
+#define TIF_RESTORE_SIGMASK	9	/* restore signal mask in do_signal */
 /* 16 free */
 #define TIF_IA32		17	/* 32bit process */ 
 #define TIF_FORK		18	/* ret_from_fork */
@@ -128,6 +129,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_IRET		(1<<TIF_IRET)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
+#define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
 #define _TIF_IA32		(1<<TIF_IA32)
 #define _TIF_FORK		(1<<TIF_FORK)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index 80fd48e84bbb..f266de294003 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -658,6 +658,7 @@ do { \
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
 #define __ARCH_WANT_SYS_TIME
 #define __ARCH_WANT_COMPAT_SYS_TIME
 
-- 
cgit v1.2.3


From 957dc87c1bd849440f0eef27e2ade67387001e13 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] Add ppoll/pselect syscalls

Needed TIF_RESTORE_SIGMASK first

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32entry.S | 4 ++--
 include/asm-x86_64/unistd.h  | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 5d4a7d125ed0..30ed0f6f4a2b 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -703,8 +703,8 @@ ia32_sys_call_table:
 	.quad sys_readlinkat		/* 305 */
 	.quad sys_fchmodat
 	.quad sys_faccessat
-	.quad quiet_ni_syscall		/* pselect6 for now */
-	.quad quiet_ni_syscall		/* ppoll for now */
+	.quad compat_sys_pselect6
+	.quad compat_sys_ppoll
 	.quad sys_unshare		/* 310 */
 	.quad compat_sys_set_robust_list
 	.quad compat_sys_get_robust_list
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index f266de294003..eeb98c168e98 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -600,9 +600,9 @@ __SYSCALL(__NR_fchmodat, sys_fchmodat)
 #define __NR_faccessat		269
 __SYSCALL(__NR_faccessat, sys_faccessat)
 #define __NR_pselect6		270
-__SYSCALL(__NR_pselect6, sys_ni_syscall)	/* for now */
+__SYSCALL(__NR_pselect6, sys_pselect6)
 #define __NR_ppoll		271
-__SYSCALL(__NR_ppoll,	sys_ni_syscall)		/* for now */
+__SYSCALL(__NR_ppoll,	sys_ppoll)
 #define __NR_unshare		272
 __SYSCALL(__NR_unshare,	sys_unshare)
 #define __NR_set_robust_list	273
-- 
cgit v1.2.3


From 2fbe7b25c8edaf2d10e6c1a4cc9f8afe714c4764 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386/x86-64: Remove un/set_nmi_callback and
 reserve/release_lapic_nmi functions

Removes the un/set_nmi_callback and reserve/release_lapic_nmi functions as
they are no longer needed.  The various subsystems are modified to register
with the die_notifier instead.

Also includes compile fixes by Andrew Morton.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/crash.c           |  20 +++++++-
 arch/i386/kernel/nmi.c             |  85 ++++---------------------------
 arch/i386/kernel/traps.c           |  23 +--------
 arch/i386/oprofile/nmi_int.c       |  47 ++++++++++-------
 arch/i386/oprofile/nmi_timer_int.c |  33 +++++++++---
 arch/x86_64/kernel/crash.c         |  20 +++++++-
 arch/x86_64/kernel/nmi.c           | 102 +++----------------------------------
 include/asm-i386/nmi.h             |  21 ++------
 include/asm-x86_64/nmi.h           |  21 --------
 kernel/sysctl.c                    |   4 +-
 10 files changed, 116 insertions(+), 260 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 5b96f038367f..736c76d6b31d 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -22,6 +22,8 @@
 #include <asm/nmi.h>
 #include <asm/hw_irq.h>
 #include <asm/apic.h>
+#include <asm/kdebug.h>
+
 #include <mach_ipi.h>
 
 
@@ -93,9 +95,18 @@ static void crash_save_self(struct pt_regs *regs)
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 static atomic_t waiting_for_crash_ipi;
 
-static int crash_nmi_callback(struct pt_regs *regs, int cpu)
+static int crash_nmi_callback(struct notifier_block *self,
+			unsigned long val, void *data)
 {
+	struct pt_regs *regs;
 	struct pt_regs fixed_regs;
+	int cpu;
+
+	if (val != DIE_NMI)
+		return NOTIFY_OK;
+
+	regs = ((struct die_args *)data)->regs;
+	cpu = raw_smp_processor_id();
 
 	/* Don't do anything if this handler is invoked on crashing cpu.
 	 * Otherwise, system will completely hang. Crashing cpu can get
@@ -125,13 +136,18 @@ static void smp_send_nmi_allbutself(void)
 	send_IPI_allbutself(NMI_VECTOR);
 }
 
+static struct notifier_block crash_nmi_nb = {
+	.notifier_call = crash_nmi_callback,
+};
+
 static void nmi_shootdown_cpus(void)
 {
 	unsigned long msecs;
 
 	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
 	/* Would it be better to replace the trap vector here? */
-	set_nmi_callback(crash_nmi_callback);
+	if (register_die_notifier(&crash_nmi_nb))
+		return;		/* return what? */
 	/* Ensure the new callback function is set before sending
 	 * out the NMI
 	 */
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index bd96ea4f2942..acd3fdea2a21 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -42,20 +42,6 @@ static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]);
  */
 #define NMI_MAX_COUNTER_BITS 66
 
-/*
- * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
- * - it may be reserved by some other driver, or not
- * - when not reserved by some other driver, it may be used for
- *   the NMI watchdog, or not
- *
- * This is maintained separately from nmi_active because the NMI
- * watchdog may also be driven from the I/O APIC timer.
- */
-static DEFINE_SPINLOCK(lapic_nmi_owner_lock);
-static unsigned int lapic_nmi_owner;
-#define LAPIC_NMI_WATCHDOG	(1<<0)
-#define LAPIC_NMI_RESERVED	(1<<1)
-
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
  * <0: the lapic NMI watchdog has not been set up, and cannot
@@ -325,33 +311,6 @@ static void enable_lapic_nmi_watchdog(void)
 	touch_nmi_watchdog();
 }
 
-int reserve_lapic_nmi(void)
-{
-	unsigned int old_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	old_owner = lapic_nmi_owner;
-	lapic_nmi_owner |= LAPIC_NMI_RESERVED;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (old_owner & LAPIC_NMI_RESERVED)
-		return -EBUSY;
-	if (old_owner & LAPIC_NMI_WATCHDOG)
-		disable_lapic_nmi_watchdog();
-	return 0;
-}
-
-void release_lapic_nmi(void)
-{
-	unsigned int new_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED;
-	lapic_nmi_owner = new_owner;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (new_owner & LAPIC_NMI_WATCHDOG)
-		enable_lapic_nmi_watchdog();
-}
-
 void disable_timer_nmi_watchdog(void)
 {
 	BUG_ON(nmi_watchdog != NMI_IO_APIC);
@@ -866,6 +825,15 @@ done:
 	return rc;
 }
 
+int do_nmi_callback(struct pt_regs * regs, int cpu)
+{
+#ifdef CONFIG_SYSCTL
+	if (unknown_nmi_panic)
+		return unknown_nmi_panic_callback(regs, cpu);
+#endif
+	return 0;
+}
+
 #ifdef CONFIG_SYSCTL
 
 static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
@@ -873,37 +841,8 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	unsigned char reason = get_nmi_reason();
 	char buf[64];
 
-	if (!(reason & 0xc0)) {
-		sprintf(buf, "NMI received for unknown reason %02x\n", reason);
-		die_nmi(regs, buf);
-	}
-	return 0;
-}
-
-/*
- * proc handler for /proc/sys/kernel/unknown_nmi_panic
- */
-int proc_unknown_nmi_panic(ctl_table *table, int write, struct file *file,
-			void __user *buffer, size_t *length, loff_t *ppos)
-{
-	int old_state;
-
-	old_state = unknown_nmi_panic;
-	proc_dointvec(table, write, file, buffer, length, ppos);
-	if (!!old_state == !!unknown_nmi_panic)
-		return 0;
-
-	if (unknown_nmi_panic) {
-		if (reserve_lapic_nmi() < 0) {
-			unknown_nmi_panic = 0;
-			return -EBUSY;
-		} else {
-			set_nmi_callback(unknown_nmi_panic_callback);
-		}
-	} else {
-		release_lapic_nmi();
-		unset_nmi_callback();
-	}
+	sprintf(buf, "NMI received for unknown reason %02x\n", reason);
+	die_nmi(regs, buf);
 	return 0;
 }
 
@@ -917,7 +856,5 @@ EXPORT_SYMBOL(reserve_perfctr_nmi);
 EXPORT_SYMBOL(release_perfctr_nmi);
 EXPORT_SYMBOL(reserve_evntsel_nmi);
 EXPORT_SYMBOL(release_evntsel_nmi);
-EXPORT_SYMBOL(reserve_lapic_nmi);
-EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
 EXPORT_SYMBOL(enable_timer_nmi_watchdog);
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 282f0bd40dfd..7db664d0b25c 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -706,13 +706,6 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	do_exit(SIGSEGV);
 }
 
-static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
-{
-	return 0;
-}
-
-static nmi_callback_t nmi_callback = dummy_nmi_callback;
-
 static void default_do_nmi(struct pt_regs * regs)
 {
 	unsigned char reason = 0;
@@ -732,9 +725,10 @@ static void default_do_nmi(struct pt_regs * regs)
 		 */
 		if (nmi_watchdog_tick(regs, reason))
 			return;
+		if (!do_nmi_callback(regs, smp_processor_id()))
 #endif
-		if (!rcu_dereference(nmi_callback)(regs, smp_processor_id()))
 			unknown_nmi_error(reason, regs);
+
 		return;
 	}
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
@@ -765,19 +759,6 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
 	nmi_exit();
 }
 
-void set_nmi_callback(nmi_callback_t callback)
-{
-	vmalloc_sync_all();
-	rcu_assign_pointer(nmi_callback, callback);
-}
-EXPORT_SYMBOL_GPL(set_nmi_callback);
-
-void unset_nmi_callback(void)
-{
-	nmi_callback = dummy_nmi_callback;
-}
-EXPORT_SYMBOL_GPL(unset_nmi_callback);
-
 #ifdef CONFIG_KPROBES
 fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code)
 {
diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index 8710ca081b1e..b3610188bcf0 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -17,14 +17,15 @@
 #include <asm/nmi.h>
 #include <asm/msr.h>
 #include <asm/apic.h>
+#include <asm/kdebug.h>
  
 #include "op_counter.h"
 #include "op_x86_model.h"
- 
+
 static struct op_x86_model_spec const * model;
 static struct op_msrs cpu_msrs[NR_CPUS];
 static unsigned long saved_lvtpc[NR_CPUS];
- 
+
 static int nmi_start(void);
 static void nmi_stop(void);
 
@@ -82,13 +83,24 @@ static void exit_driverfs(void)
 #define exit_driverfs() do { } while (0)
 #endif /* CONFIG_PM */
 
-
-static int nmi_callback(struct pt_regs * regs, int cpu)
+int profile_exceptions_notify(struct notifier_block *self,
+				unsigned long val, void *data)
 {
-	return model->check_ctrs(regs, &cpu_msrs[cpu]);
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+	int cpu = smp_processor_id();
+
+	switch(val) {
+	case DIE_NMI:
+		if (model->check_ctrs(args->regs, &cpu_msrs[cpu]))
+			ret = NOTIFY_STOP;
+		break;
+	default:
+		break;
+	}
+	return ret;
 }
- 
- 
+
 static void nmi_cpu_save_registers(struct op_msrs * msrs)
 {
 	unsigned int const nr_ctrs = model->num_counters;
@@ -174,27 +186,29 @@ static void nmi_cpu_setup(void * dummy)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 }
 
+static struct notifier_block profile_exceptions_nb = {
+	.notifier_call = profile_exceptions_notify,
+	.next = NULL,
+	.priority = 0
+};
 
 static int nmi_setup(void)
 {
+	int err=0;
+
 	if (!allocate_msrs())
 		return -ENOMEM;
 
-	/* We walk a thin line between law and rape here.
-	 * We need to be careful to install our NMI handler
-	 * without actually triggering any NMIs as this will
-	 * break the core code horrifically.
-	 */
-	if (reserve_lapic_nmi() < 0) {
+	if ((err = register_die_notifier(&profile_exceptions_nb))){
 		free_msrs();
-		return -EBUSY;
+		return err;
 	}
+
 	/* We need to serialize save and setup for HT because the subset
 	 * of msrs are distinct for save and setup operations
 	 */
 	on_each_cpu(nmi_save_registers, NULL, 0, 1);
 	on_each_cpu(nmi_cpu_setup, NULL, 0, 1);
-	set_nmi_callback(nmi_callback);
 	nmi_enabled = 1;
 	return 0;
 }
@@ -250,8 +264,7 @@ static void nmi_shutdown(void)
 {
 	nmi_enabled = 0;
 	on_each_cpu(nmi_cpu_shutdown, NULL, 0, 1);
-	unset_nmi_callback();
-	release_lapic_nmi();
+	unregister_die_notifier(&profile_exceptions_nb);
 	free_msrs();
 }
 
diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c
index a33a73bb502d..93ca48f804ac 100644
--- a/arch/i386/oprofile/nmi_timer_int.c
+++ b/arch/i386/oprofile/nmi_timer_int.c
@@ -17,32 +17,49 @@
 #include <asm/nmi.h>
 #include <asm/apic.h>
 #include <asm/ptrace.h>
+#include <asm/kdebug.h>
  
-static int nmi_timer_callback(struct pt_regs * regs, int cpu)
+int profile_timer_exceptions_notify(struct notifier_block *self,
+				unsigned long val, void *data)
 {
-	oprofile_add_sample(regs, 0);
-	return 1;
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	switch(val) {
+	case DIE_NMI:
+		oprofile_add_sample(args->regs, 0);
+		ret = NOTIFY_STOP;
+		break;
+	default:
+		break;
+	}
+	return ret;
 }
 
+static struct notifier_block profile_timer_exceptions_nb = {
+	.notifier_call = profile_timer_exceptions_notify,
+	.next = NULL,
+	.priority = 0
+};
+
 static int timer_start(void)
 {
-	disable_timer_nmi_watchdog();
-	set_nmi_callback(nmi_timer_callback);
+	if (register_die_notifier(&profile_timer_exceptions_nb))
+		return 1;
 	return 0;
 }
 
 
 static void timer_stop(void)
 {
-	enable_timer_nmi_watchdog();
-	unset_nmi_callback();
+	unregister_die_notifier(&profile_timer_exceptions_nb);
 	synchronize_sched();  /* Allow already-started NMIs to complete. */
 }
 
 
 int __init op_nmi_timer_init(struct oprofile_operations * ops)
 {
-	if (atomic_read(&nmi_active) <= 0)
+	if ((nmi_watchdog != NMI_IO_APIC) || (atomic_read(&nmi_active) <= 0))
 		return -ENODEV;
 
 	ops->start = timer_start;
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index d8d5750d6106..44c8af65325e 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -23,6 +23,7 @@
 #include <asm/nmi.h>
 #include <asm/hw_irq.h>
 #include <asm/mach_apic.h>
+#include <asm/kdebug.h>
 
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
@@ -95,8 +96,18 @@ static void crash_save_self(struct pt_regs *regs)
 #ifdef CONFIG_SMP
 static atomic_t waiting_for_crash_ipi;
 
-static int crash_nmi_callback(struct pt_regs *regs, int cpu)
+static int crash_nmi_callback(struct notifier_block *self,
+				unsigned long val, void *data)
 {
+	struct pt_regs *regs;
+	int cpu;
+
+	if (val != DIE_NMI)
+		return NOTIFY_OK;
+
+	regs = ((struct die_args *)data)->regs;
+	cpu = raw_smp_processor_id();
+
 	/*
 	 * Don't do anything if this handler is invoked on crashing cpu.
 	 * Otherwise, system will completely hang. Crashing cpu can get
@@ -127,12 +138,17 @@ static void smp_send_nmi_allbutself(void)
  * cpu hotplug shouldn't matter.
  */
 
+static struct notifier_block crash_nmi_nb = {
+	.notifier_call = crash_nmi_callback,
+};
+
 static void nmi_shootdown_cpus(void)
 {
 	unsigned long msecs;
 
 	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
-	set_nmi_callback(crash_nmi_callback);
+	if (register_die_notifier(&crash_nmi_nb))
+		return;         /* return what? */
 
 	/*
 	 * Ensure the new callback function is set before sending
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index f6b881b23a70..9d175dcf3a2d 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -41,20 +41,6 @@ static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]);
  */
 #define NMI_MAX_COUNTER_BITS 66
 
-/*
- * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
- * - it may be reserved by some other driver, or not
- * - when not reserved by some other driver, it may be used for
- *   the NMI watchdog, or not
- *
- * This is maintained separately from nmi_active because the NMI
- * watchdog may also be driven from the I/O APIC timer.
- */
-static DEFINE_SPINLOCK(lapic_nmi_owner_lock);
-static unsigned int lapic_nmi_owner;
-#define LAPIC_NMI_WATCHDOG	(1<<0)
-#define LAPIC_NMI_RESERVED	(1<<1)
-
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
  * <0: the lapic NMI watchdog has not been set up, and cannot
@@ -321,33 +307,6 @@ static void enable_lapic_nmi_watchdog(void)
 	touch_nmi_watchdog();
 }
 
-int reserve_lapic_nmi(void)
-{
-	unsigned int old_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	old_owner = lapic_nmi_owner;
-	lapic_nmi_owner |= LAPIC_NMI_RESERVED;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (old_owner & LAPIC_NMI_RESERVED)
-		return -EBUSY;
-	if (old_owner & LAPIC_NMI_WATCHDOG)
-		disable_lapic_nmi_watchdog();
-	return 0;
-}
-
-void release_lapic_nmi(void)
-{
-	unsigned int new_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED;
-	lapic_nmi_owner = new_owner;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (new_owner & LAPIC_NMI_WATCHDOG)
-		enable_lapic_nmi_watchdog();
-}
-
 void disable_timer_nmi_watchdog(void)
 {
 	BUG_ON(nmi_watchdog != NMI_IO_APIC);
@@ -762,13 +721,6 @@ done:
 	return rc;
 }
 
-static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
-{
-	return 0;
-}
- 
-static nmi_callback_t nmi_callback = dummy_nmi_callback;
- 
 asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 {
 	nmi_enter();
@@ -779,21 +731,12 @@ asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 
 int do_nmi_callback(struct pt_regs * regs, int cpu)
 {
-	return rcu_dereference(nmi_callback)(regs, cpu);
-}
-
-void set_nmi_callback(nmi_callback_t callback)
-{
-	vmalloc_sync_all();
-	rcu_assign_pointer(nmi_callback, callback);
-}
-EXPORT_SYMBOL_GPL(set_nmi_callback);
-
-void unset_nmi_callback(void)
-{
-	nmi_callback = dummy_nmi_callback;
+#ifdef CONFIG_SYSCTL
+	if (unknown_nmi_panic)
+		return unknown_nmi_panic_callback(regs, cpu);
+#endif
+	return 0;
 }
-EXPORT_SYMBOL_GPL(unset_nmi_callback);
 
 #ifdef CONFIG_SYSCTL
 
@@ -802,37 +745,8 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	unsigned char reason = get_nmi_reason();
 	char buf[64];
 
-	if (!(reason & 0xc0)) {
-		sprintf(buf, "NMI received for unknown reason %02x\n", reason);
-		die_nmi(buf,regs);
-	}
-	return 0;
-}
-
-/*
- * proc handler for /proc/sys/kernel/unknown_nmi_panic
- */
-int proc_unknown_nmi_panic(struct ctl_table *table, int write, struct file *file,
-			void __user *buffer, size_t *length, loff_t *ppos)
-{
-	int old_state;
-
-	old_state = unknown_nmi_panic;
-	proc_dointvec(table, write, file, buffer, length, ppos);
-	if (!!old_state == !!unknown_nmi_panic)
-		return 0;
-
-	if (unknown_nmi_panic) {
-		if (reserve_lapic_nmi() < 0) {
-			unknown_nmi_panic = 0;
-			return -EBUSY;
-		} else {
-			set_nmi_callback(unknown_nmi_panic_callback);
-		}
-	} else {
-		release_lapic_nmi();
-		unset_nmi_callback();
-	}
+	sprintf(buf, "NMI received for unknown reason %02x\n", reason);
+	die_nmi(buf,regs);
 	return 0;
 }
 
@@ -846,8 +760,6 @@ EXPORT_SYMBOL(reserve_perfctr_nmi);
 EXPORT_SYMBOL(release_perfctr_nmi);
 EXPORT_SYMBOL(reserve_evntsel_nmi);
 EXPORT_SYMBOL(release_evntsel_nmi);
-EXPORT_SYMBOL(reserve_lapic_nmi);
-EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
 EXPORT_SYMBOL(enable_timer_nmi_watchdog);
 EXPORT_SYMBOL(touch_nmi_watchdog);
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index da0e0b4e9139..34d6bf063b6e 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -6,24 +6,13 @@
 
 #include <linux/pm.h>
 
-struct pt_regs;
-
-typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu);
-
-/**
- * set_nmi_callback
- *
- * Set a handler for an NMI. Only one handler may be
- * set. Return 1 if the NMI was handled.
- */
-void set_nmi_callback(nmi_callback_t callback);
-
 /**
- * unset_nmi_callback
+ * do_nmi_callback
  *
- * Remove the handler previously set.
+ * Check to see if a callback exists and execute it.  Return 1
+ * if the handler exists and was handled successfully.
  */
-void unset_nmi_callback(void);
+int do_nmi_callback(struct pt_regs *regs, int cpu);
 
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
 extern int avail_to_resrv_perfctr_nmi(unsigned int);
@@ -33,8 +22,6 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
-extern int reserve_lapic_nmi(void);
-extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 8f02a2a416e6..8818c39d34e0 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -7,25 +7,6 @@
 #include <linux/pm.h>
 #include <asm/io.h>
  
-struct pt_regs;
-
-typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu);
-
-/**
- * set_nmi_callback
- *
- * Set a handler for an NMI. Only one handler may be
- * set. Return 1 if the NMI was handled.
- */
-void set_nmi_callback(nmi_callback_t callback);
-
-/**
- * unset_nmi_callback
- *
- * Remove the handler previously set.
- */
-void unset_nmi_callback(void);
-
 /**
  * do_nmi_callback
  *
@@ -72,8 +53,6 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
-extern int reserve_lapic_nmi(void);
-extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 362a0cc37138..83f168361624 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -76,8 +76,6 @@ extern int compat_log;
 
 #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86)
 int unknown_nmi_panic;
-extern int proc_unknown_nmi_panic(ctl_table *, int, struct file *,
-				  void __user *, size_t *, loff_t *);
 #endif
 
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
@@ -628,7 +626,7 @@ static ctl_table kern_table[] = {
 		.data           = &unknown_nmi_panic,
 		.maxlen         = sizeof (int),
 		.mode           = 0644,
-		.proc_handler   = &proc_unknown_nmi_panic,
+		.proc_handler   = &proc_dointvec,
 	},
 #endif
 #if defined(CONFIG_X86)
-- 
cgit v1.2.3


From 407984f1af259b31957c7c05075a454a751bb801 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: Add abilty to enable/disable nmi watchdog with sysctl

Adds a new /proc/sys/kernel/nmi call that will enable/disable the nmi
watchdog.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c   | 52 ++++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/nmi.c | 48 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-i386/nmi.h   |  1 +
 include/asm-x86_64/nmi.h |  1 +
 include/linux/sysctl.h   |  1 +
 kernel/sysctl.c          | 11 ++++++++++
 6 files changed, 114 insertions(+)

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index acd3fdea2a21..28065d0b71a9 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -846,6 +846,58 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	return 0;
 }
 
+/*
+ * proc handler for /proc/sys/kernel/nmi_watchdog
+ */
+int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
+			void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int old_state;
+
+	nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0;
+	old_state = nmi_watchdog_enabled;
+	proc_dointvec(table, write, file, buffer, length, ppos);
+	if (!!old_state == !!nmi_watchdog_enabled)
+		return 0;
+
+	if (atomic_read(&nmi_active) < 0) {
+		printk(KERN_WARNING "NMI watchdog is permanently disabled\n");
+		return -EINVAL;
+	}
+
+	if (nmi_watchdog == NMI_DEFAULT) {
+		if (nmi_known_cpu() > 0)
+			nmi_watchdog = NMI_LOCAL_APIC;
+		else
+			nmi_watchdog = NMI_IO_APIC;
+	}
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+	{
+		if (nmi_watchdog_enabled)
+			enable_lapic_nmi_watchdog();
+		else
+			disable_lapic_nmi_watchdog();
+	} else if (nmi_watchdog == NMI_IO_APIC) {
+		/* FIXME
+		 * for some reason these functions don't work
+		 */
+		printk("Can not enable/disable NMI on IO APIC\n");
+		return -EINVAL;
+#if 0
+		if (nmi_watchdog_enabled)
+			enable_timer_nmi_watchdog();
+		else
+			disable_timer_nmi_watchdog();
+#endif
+	} else {
+		printk( KERN_WARNING
+			"NMI watchdog doesn't know what hardware to touch\n");
+		return -EIO;
+	}
+	return 0;
+}
+
 #endif
 
 EXPORT_SYMBOL(nmi_active);
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 9d175dcf3a2d..3a17411a9a19 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -750,6 +750,54 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	return 0;
 }
 
+/*
+ * proc handler for /proc/sys/kernel/nmi
+ */
+int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
+			void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int old_state;
+
+	nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0;
+	old_state = nmi_watchdog_enabled;
+	proc_dointvec(table, write, file, buffer, length, ppos);
+	if (!!old_state == !!nmi_watchdog_enabled)
+		return 0;
+
+	if (atomic_read(&nmi_active) < 0) {
+		printk( KERN_WARNING "NMI watchdog is permanently disabled\n");
+		return -EINVAL;
+	}
+
+	/* if nmi_watchdog is not set yet, then set it */
+	nmi_watchdog_default();
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+	{
+		if (nmi_watchdog_enabled)
+			enable_lapic_nmi_watchdog();
+		else
+			disable_lapic_nmi_watchdog();
+	} else if (nmi_watchdog == NMI_IO_APIC) {
+		/* FIXME
+		 * for some reason these functions don't work
+		 */
+		printk("Can not enable/disable NMI on IO APIC\n");
+		return -EIO;
+#if 0
+		if (nmi_watchdog_enabled)
+			enable_timer_nmi_watchdog();
+		else
+			disable_timer_nmi_watchdog();
+#endif
+	} else {
+		printk(KERN_WARNING
+			"NMI watchdog doesn't know what hardware to touch\n");
+		return -EIO;
+	}
+	return 0;
+}
+
 #endif
 
 EXPORT_SYMBOL(nmi_active);
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 34d6bf063b6e..13b5d8311bf7 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -14,6 +14,7 @@
  */
 int do_nmi_callback(struct pt_regs *regs, int cpu);
 
+extern int nmi_watchdog_enabled;
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
 extern int avail_to_resrv_perfctr_nmi(unsigned int);
 extern int reserve_perfctr_nmi(unsigned int);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 8818c39d34e0..2c23b0df87d2 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -43,6 +43,7 @@ extern void die_nmi(char *str, struct pt_regs *regs);
 
 extern int panic_on_timeout;
 extern int unknown_nmi_panic;
+extern int nmi_watchdog_enabled;
 
 extern int check_nmi_watchdog(void);
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 736ed917a4f8..ecb79ba52ae1 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -150,6 +150,7 @@ enum
 	KERN_IA64_UNALIGNED=72, /* int: ia64 unaligned userland trap enable */
 	KERN_COMPAT_LOG=73,	/* int: print compat layer  messages */
 	KERN_MAX_LOCK_DEPTH=74,
+	KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */
 };
 
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 83f168361624..040de6bd74dd 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -76,6 +76,9 @@ extern int compat_log;
 
 #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86)
 int unknown_nmi_panic;
+int nmi_watchdog_enabled;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
 #endif
 
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
@@ -628,6 +631,14 @@ static ctl_table kern_table[] = {
 		.mode           = 0644,
 		.proc_handler   = &proc_dointvec,
 	},
+	{
+		.ctl_name       = KERN_NMI_WATCHDOG,
+		.procname       = "nmi_watchdog",
+		.data           = &nmi_watchdog_enabled,
+		.maxlen         = sizeof (int),
+		.mode           = 0644,
+		.proc_handler   = &proc_nmi_enabled,
+	},
 #endif
 #if defined(CONFIG_X86)
 	{
-- 
cgit v1.2.3


From 8da5adda91df3d2fcc5300e68da491694c9af019 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: Allow users to force a panic on NMI

To quote Alan Cox:

The default Linux behaviour on an NMI of either memory or unknown is to
continue operation. For many environments such as scientific computing
it is preferable that the box is taken out and the error dealt with than
an uncorrected parity/ECC error get propogated.

A small number of systems do generate NMI's for bizarre random reasons
such as power management so the default is unchanged. In other respects
the new proc/sys entry works like the existing panic controls already in
that directory.

This is separate to the edac support - EDAC allows supported chipsets to
handle ECC errors well, this change allows unsupported cases to at least
panic rather than cause problems further down the line.

Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c   | 6 ++++++
 arch/x86_64/kernel/traps.c | 6 ++++++
 include/linux/kernel.h     | 1 +
 include/linux/sysctl.h     | 1 +
 kernel/panic.c             | 1 +
 kernel/sysctl.c            | 8 ++++++++
 6 files changed, 23 insertions(+)

(limited to 'include')

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 7db664d0b25c..2f6cb8276480 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -635,6 +635,8 @@ static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
 			"to continue\n");
 	printk(KERN_EMERG "You probably have a hardware problem with your RAM "
 			"chips\n");
+	if (panic_on_unrecovered_nmi)
+                panic("NMI: Not continuing");
 
 	/* Clear and disable the memory parity error line. */
 	clear_mem_error(reason);
@@ -670,6 +672,10 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 		reason, smp_processor_id());
 	printk("Dazed and confused, but trying to continue\n");
 	printk("Do you have a strange power saving mode enabled?\n");
+
+	if (panic_on_unrecovered_nmi)
+                panic("NMI: Not continuing");
+
 }
 
 static DEFINE_SPINLOCK(nmi_print_lock);
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 42bc070fdf11..b18829db2a6a 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -732,6 +732,8 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
 	printk("You probably have a hardware problem with your RAM chips\n");
+	if (panic_on_unrecovered_nmi)
+               panic("NMI: Not continuing");
 
 	/* Clear and disable the memory parity error line. */
 	reason = (reason & 0xf) | 4;
@@ -757,6 +759,10 @@ unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {	printk("Uhhuh. NMI received for unknown reason %02x.\n", reason);
 	printk("Dazed and confused, but trying to continue\n");
 	printk("Do you have a strange power saving mode enabled?\n");
+
+	if (panic_on_unrecovered_nmi)
+                panic("NMI: Not continuing");
+
 }
 
 /* Runs on IST stack. This code must keep interrupts off all the time.
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b2ae4fdce8b..1ff9609300b4 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -186,6 +186,7 @@ extern void bust_spinlocks(int yes);
 extern int oops_in_progress;		/* If set, an oops, panic(), BUG() or die() is in progress */
 extern int panic_timeout;
 extern int panic_on_oops;
+extern int panic_on_unrecovered_nmi;
 extern int tainted;
 extern const char *print_tainted(void);
 extern void add_taint(unsigned);
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index ecb79ba52ae1..432778446ad2 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -151,6 +151,7 @@ enum
 	KERN_COMPAT_LOG=73,	/* int: print compat layer  messages */
 	KERN_MAX_LOCK_DEPTH=74,
 	KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */
+	KERN_PANIC_ON_NMI=76, /* int: whether we will panic on an unrecovered */
 };
 
 
diff --git a/kernel/panic.c b/kernel/panic.c
index 8010b9b17aca..d2db3e2209e0 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -21,6 +21,7 @@
 #include <linux/debug_locks.h>
 
 int panic_on_oops;
+int panic_on_unrecovered_nmi;
 int tainted;
 static int pause_on_oops;
 static int pause_on_oops_flag;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 040de6bd74dd..220e20564124 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -641,6 +641,14 @@ static ctl_table kern_table[] = {
 	},
 #endif
 #if defined(CONFIG_X86)
+	{
+		.ctl_name	= KERN_PANIC_ON_NMI,
+		.procname	= "panic_on_unrecovered_nmi",
+		.data		= &panic_on_unrecovered_nmi,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 	{
 		.ctl_name	= KERN_BOOTLOADER_TYPE,
 		.procname	= "bootloader_type",
-- 
cgit v1.2.3


From 4038f901cf102a40715b900984ed7540a9fa637f Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386/x86-64: Fix NMI watchdog suspend/resume

Making NMI suspend/resume work with SMP. We use CPU hotplug to offline
APs in SMP suspend/resume. Only BSP executes sysdev's .suspend/.resume
method. APs should follow CPU hotplug code path.

And:

+From: Don Zickus <dzickus@redhat.com>

Makes the start/stop paths of nmi watchdog more robust to handle the
suspend/resume cases more gracefully.

AK: I merged the two patches together

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/i386/kernel/nmi.c       | 33 ++++++++++++++++++++++++++-------
 arch/i386/kernel/smpboot.c   |  3 ++-
 arch/x86_64/kernel/nmi.c     | 33 ++++++++++++++++++++++++++-------
 arch/x86_64/kernel/smpboot.c |  2 ++
 include/asm-i386/nmi.h       |  1 +
 include/asm-x86_64/nmi.h     |  1 +
 6 files changed, 58 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 6241e4448cab..8e4ed930ce6b 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -63,7 +63,6 @@ struct nmi_watchdog_ctlblk {
 static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
 /* local prototypes */
-static void stop_apic_nmi_watchdog(void *unused);
 static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
 
 extern void show_registers(struct pt_regs *regs);
@@ -341,15 +340,20 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
+	/* only CPU0 goes here, other CPUs should be offline */
 	nmi_pm_active = atomic_read(&nmi_active);
-	disable_lapic_nmi_watchdog();
+	stop_apic_nmi_watchdog(NULL);
+	BUG_ON(atomic_read(&nmi_active) != 0);
 	return 0;
 }
 
 static int lapic_nmi_resume(struct sys_device *dev)
 {
-	if (nmi_pm_active > 0)
-		enable_lapic_nmi_watchdog();
+	/* only CPU0 goes here, other CPUs should be offline */
+	if (nmi_pm_active > 0) {
+		setup_apic_nmi_watchdog(NULL);
+		touch_nmi_watchdog();
+	}
 	return 0;
 }
 
@@ -626,11 +630,21 @@ static void stop_p4_watchdog(void)
 
 void setup_apic_nmi_watchdog (void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 1)
+		return;
+
+	/* cheap hack to support suspend/resume */
+	/* if cpu0 is not active neither should the other cpus */
+	if ((smp_processor_id() != 0) && (atomic_read(&nmi_active) <= 0))
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -663,17 +677,22 @@ void setup_apic_nmi_watchdog (void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	wd->enabled = 1;
 	atomic_inc(&nmi_active);
 }
 
-static void stop_apic_nmi_watchdog(void *unused)
+void stop_apic_nmi_watchdog(void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 0)
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -697,7 +716,7 @@ static void stop_apic_nmi_watchdog(void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	wd->enabled = 0;
 	atomic_dec(&nmi_active);
 }
 
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index efe07990e7fc..9367af76ce37 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -1376,7 +1376,8 @@ int __cpu_disable(void)
 	 */
 	if (cpu == 0)
 		return -EBUSY;
-
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		stop_apic_nmi_watchdog(NULL);
 	clear_local_APIC();
 	/* Allow any queued timer interrupts to get serviced */
 	local_irq_enable();
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index dd57410dad51..5a35975e5763 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -63,7 +63,6 @@ struct nmi_watchdog_ctlblk {
 static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
 /* local prototypes */
-static void stop_apic_nmi_watchdog(void *unused);
 static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
 
 /* converts an msr to an appropriate reservation bit */
@@ -337,15 +336,20 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
+	/* only CPU0 goes here, other CPUs should be offline */
 	nmi_pm_active = atomic_read(&nmi_active);
-	disable_lapic_nmi_watchdog();
+	stop_apic_nmi_watchdog(NULL);
+	BUG_ON(atomic_read(&nmi_active) != 0);
 	return 0;
 }
 
 static int lapic_nmi_resume(struct sys_device *dev)
 {
-	if (nmi_pm_active > 0)
-		enable_lapic_nmi_watchdog();
+	/* only CPU0 goes here, other CPUs should be offline */
+	if (nmi_pm_active > 0) {
+		setup_apic_nmi_watchdog(NULL);
+		touch_nmi_watchdog();
+	}
 	return 0;
 }
 
@@ -561,11 +565,21 @@ static void stop_p4_watchdog(void)
 
 void setup_apic_nmi_watchdog(void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 1)
+		return;
+
+	/* cheap hack to support suspend/resume */
+	/* if cpu0 is not active neither should the other cpus */
+	if ((smp_processor_id() != 0) && (atomic_read(&nmi_active) <= 0))
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -582,17 +596,22 @@ void setup_apic_nmi_watchdog(void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	wd->enabled = 1;
 	atomic_inc(&nmi_active);
 }
 
-static void stop_apic_nmi_watchdog(void *unused)
+void stop_apic_nmi_watchdog(void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 0)
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -607,7 +626,7 @@ static void stop_apic_nmi_watchdog(void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	wd->enabled = 0;
 	atomic_dec(&nmi_active);
 }
 
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 975380207b46..15555879ce95 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1233,6 +1233,8 @@ int __cpu_disable(void)
 	if (cpu == 0)
 		return -EBUSY;
 
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		stop_apic_nmi_watchdog(NULL);
 	clear_local_APIC();
 
 	/*
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 13b5d8311bf7..303bcd4592bb 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -23,6 +23,7 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
+extern void stop_apic_nmi_watchdog (void *);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 2c23b0df87d2..578596494275 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -54,6 +54,7 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
+extern void stop_apic_nmi_watchdog (void *);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
-- 
cgit v1.2.3


From fac58550e80c307bf17cfa0dd544fca4eff120a5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] Fix up panic messages for different NMI panics

When a unknown NMI happened the panic would claim a NMI watchdog timeout.
Also it would check the variable set by nmi_watchdog=panic and panic then.

Fix up the panic message to be generic
Unconditionally panic on unknown NMI when panic on unknown nmi is enabled.

Noticed by Jan Beulich

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/nmi.c   | 5 +++--
 arch/x86_64/kernel/traps.c | 7 +++----
 include/asm-x86_64/nmi.h   | 2 +-
 3 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5a35975e5763..1b76d1574529 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -695,7 +695,8 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 		 */
 		local_inc(&__get_cpu_var(alert_counter));
 		if (local_read(&__get_cpu_var(alert_counter)) == 5*nmi_hz)
-			die_nmi("NMI Watchdog detected LOCKUP on CPU %d\n", regs);
+			die_nmi("NMI Watchdog detected LOCKUP on CPU %d\n", regs,
+				panic_on_timeout);
 	} else {
 		__get_cpu_var(last_irq_sum) = sum;
 		local_set(&__get_cpu_var(alert_counter), 0);
@@ -765,7 +766,7 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	char buf[64];
 
 	sprintf(buf, "NMI received for unknown reason %02x\n", reason);
-	die_nmi(buf,regs);
+	die_nmi(buf, regs, 1);	/* Always panic here */
 	return 0;
 }
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index dae10df60926..96f62a033242 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -570,7 +570,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 	do_exit(SIGSEGV); 
 }
 
-void __kprobes die_nmi(char *str, struct pt_regs *regs)
+void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
 {
 	unsigned long flags = oops_begin();
 
@@ -582,9 +582,8 @@ void __kprobes die_nmi(char *str, struct pt_regs *regs)
 	show_registers(regs);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
-	if (panic_on_timeout || panic_on_oops)
-		panic("nmi watchdog");
-	printk("console shuts up ...\n");
+	if (do_panic || panic_on_oops)
+		panic("Non maskable interrupt");
 	oops_end(flags);
 	nmi_exit();
 	local_irq_enable();
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 578596494275..cbf2669bca71 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -37,7 +37,7 @@ static inline void unset_nmi_pm_callback(struct pm_dev * dev)
 #endif /* CONFIG_PM */
  
 extern void default_do_nmi(struct pt_regs *);
-extern void die_nmi(char *str, struct pt_regs *regs);
+extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 
 #define get_nmi_reason() inb(0x61)
 
-- 
cgit v1.2.3


From 248dcb2ffffe8f3e4a369556a68988788c208111 Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: i386/x86-64 Add nmi watchdog support for new Intel CPUs

AK: This redoes the changes I temporarily reverted.

Intel now has support for Architectural Performance Monitoring Counters
( Refer to IA-32 Intel Architecture Software Developer's Manual
http://www.intel.com/design/pentium4/manuals/253669.htm ). This
feature is present starting from Intel Core Duo and Intel Core Solo processors.

What this means is, the performance monitoring counters and some performance
monitoring events are now defined in an architectural way (using cpuid).
And there will be no need to check for family/model etc for these architectural
events.

Below is the patch to use this performance counters in nmi watchdog driver.
Patch handles both i386 and x86-64 kernels.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c                  | 126 +++++++++++++++++++++++++++++--
 arch/x86_64/kernel/nmi.c                | 130 ++++++++++++++++++++++++++++++--
 include/asm-i386/intel_arch_perfmon.h   |  31 ++++++++
 include/asm-x86_64/intel_arch_perfmon.h |  31 ++++++++
 4 files changed, 308 insertions(+), 10 deletions(-)
 create mode 100644 include/asm-i386/intel_arch_perfmon.h
 create mode 100644 include/asm-x86_64/intel_arch_perfmon.h

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 6e5085d5d2f6..7b9a053effa3 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -26,6 +26,7 @@
 #include <asm/smp.h>
 #include <asm/nmi.h>
 #include <asm/kdebug.h>
+#include <asm/intel_arch_perfmon.h>
 
 #include "mach_traps.h"
 
@@ -77,6 +78,9 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_PERFCTR0);
 	case X86_VENDOR_INTEL:
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+
 		switch (boot_cpu_data.x86) {
 		case 6:
 			return (msr - MSR_P6_PERFCTR0);
@@ -95,6 +99,9 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_EVNTSEL0);
 	case X86_VENDOR_INTEL:
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+
 		switch (boot_cpu_data.x86) {
 		case 6:
 			return (msr - MSR_P6_EVNTSEL0);
@@ -174,7 +181,10 @@ static __cpuinit inline int nmi_known_cpu(void)
 	case X86_VENDOR_AMD:
 		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
 	case X86_VENDOR_INTEL:
-		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return 1;
+		else
+			return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
 	}
 	return 0;
 }
@@ -261,8 +271,24 @@ static int __init check_nmi_watchdog(void)
 
 	/* now that we know it works we can reduce NMI frequency to
 	   something more reasonable; makes a difference in some configs */
-	if (nmi_watchdog == NMI_LOCAL_APIC)
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 		nmi_hz = 1;
+		/*
+		 * On Intel CPUs with ARCH_PERFMON only 32 bits in the counter
+		 * are writable, with higher bits sign extending from bit 31.
+		 * So, we can only program the counter with 31 bit values and
+		 * 32nd bit should be 1, for 33.. to be 1.
+		 * Find the appropriate nmi_hz
+		 */
+	 	if (wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0 &&
+			((u64)cpu_khz * 1000) > 0x7fffffffULL) {
+			u64 count = (u64)cpu_khz * 1000;
+			do_div(count, 0x7fffffffUL);
+			nmi_hz = count + 1;
+		}
+	}
 
 	kfree(prev_nmi_count);
 	return 0;
@@ -637,6 +663,85 @@ static void stop_p4_watchdog(void)
 	release_perfctr_nmi(wd->perfctr_msr);
 }
 
+#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
+#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
+
+static int setup_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		goto fail;
+
+	perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
+	evntsel_msr = MSR_ARCH_PERFMON_EVENTSEL0;
+
+	if (!reserve_perfctr_nmi(perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(evntsel_msr))
+		goto fail1;
+
+	wrmsrl(perfctr_msr, 0UL);
+
+	evntsel = ARCH_PERFMON_EVENTSEL_INT
+		| ARCH_PERFMON_EVENTSEL_OS
+		| ARCH_PERFMON_EVENTSEL_USR
+		| ARCH_PERFMON_NMI_EVENT_SEL
+		| ARCH_PERFMON_NMI_EVENT_UMASK;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "INTEL_ARCH_PERFCTR0");
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL << (eax.split.bit_width - 1);
+	return 1;
+fail1:
+	release_perfctr_nmi(perfctr_msr);
+fail:
+	return 0;
+}
+
+static void stop_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		return;
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
 void setup_apic_nmi_watchdog (void *unused)
 {
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
@@ -663,6 +768,11 @@ void setup_apic_nmi_watchdog (void *unused)
 				return;
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				if (!setup_intel_arch_watchdog())
+					return;
+				break;
+			}
 			switch (boot_cpu_data.x86) {
 			case 6:
 				if (boot_cpu_data.x86_model > 0xd)
@@ -708,6 +818,10 @@ void stop_apic_nmi_watchdog(void *unused)
 			stop_k7_watchdog();
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				stop_intel_arch_watchdog();
+				break;
+			}
 			switch (boot_cpu_data.x86) {
 			case 6:
 				if (boot_cpu_data.x86_model > 0xd)
@@ -831,10 +945,12 @@ int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 	 			wrmsrl(wd->cccr_msr, dummy);
 	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
 	 		}
-			else if (wd->perfctr_msr == MSR_P6_PERFCTR0) {
-				/* Only P6 based Pentium M need to re-unmask
+			else if (wd->perfctr_msr == MSR_P6_PERFCTR0 ||
+				 wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
+				/* P6 based Pentium M need to re-unmask
 				 * the apic vector but it doesn't hurt
-				 * other P6 variant */
+				 * other P6 variant.
+				 * ArchPerfom/Core Duo also needs this */
 				apic_write(APIC_LVTPC, APIC_DM_NMI);
 			}
 			/* start the cycle over again */
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 1b76d1574529..4d6fb047952e 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -26,6 +26,7 @@
 #include <asm/proto.h>
 #include <asm/kdebug.h>
 #include <asm/mce.h>
+#include <asm/intel_arch_perfmon.h>
 
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
@@ -73,7 +74,10 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_PERFCTR0);
 	case X86_VENDOR_INTEL:
-		return (msr - MSR_P4_BPU_PERFCTR0);
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+		else
+			return (msr - MSR_P4_BPU_PERFCTR0);
 	}
 	return 0;
 }
@@ -86,7 +90,10 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_EVNTSEL0);
 	case X86_VENDOR_INTEL:
-		return (msr - MSR_P4_BSU_ESCR0);
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+		else
+			return (msr - MSR_P4_BSU_ESCR0);
 	}
 	return 0;
 }
@@ -160,7 +167,10 @@ static __cpuinit inline int nmi_known_cpu(void)
 	case X86_VENDOR_AMD:
 		return boot_cpu_data.x86 == 15;
 	case X86_VENDOR_INTEL:
-		return boot_cpu_data.x86 == 15;
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return 1;
+		else
+			return (boot_cpu_data.x86 == 15);
 	}
 	return 0;
 }
@@ -246,8 +256,22 @@ int __init check_nmi_watchdog (void)
 
 	/* now that we know it works we can reduce NMI frequency to
 	   something more reasonable; makes a difference in some configs */
-	if (nmi_watchdog == NMI_LOCAL_APIC)
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 		nmi_hz = 1;
+		/*
+		 * On Intel CPUs with ARCH_PERFMON only 32 bits in the counter
+		 * are writable, with higher bits sign extending from bit 31.
+		 * So, we can only program the counter with 31 bit values and
+		 * 32nd bit should be 1, for 33.. to be 1.
+		 * Find the appropriate nmi_hz
+		 */
+	 	if (wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0 &&
+			((u64)cpu_khz * 1000) > 0x7fffffffULL) {
+			nmi_hz = ((u64)cpu_khz * 1000) / 0x7fffffffUL + 1;
+		}
+	}
 
 	kfree(counts);
 	return 0;
@@ -563,6 +587,87 @@ static void stop_p4_watchdog(void)
 	release_perfctr_nmi(wd->perfctr_msr);
 }
 
+#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
+#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
+
+static int setup_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		goto fail;
+
+	perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
+	evntsel_msr = MSR_ARCH_PERFMON_EVENTSEL0;
+
+	if (!reserve_perfctr_nmi(perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(evntsel_msr))
+		goto fail1;
+
+	wrmsrl(perfctr_msr, 0UL);
+
+	evntsel = ARCH_PERFMON_EVENTSEL_INT
+		| ARCH_PERFMON_EVENTSEL_OS
+		| ARCH_PERFMON_EVENTSEL_USR
+		| ARCH_PERFMON_NMI_EVENT_SEL
+		| ARCH_PERFMON_NMI_EVENT_UMASK;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsrl(perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL << (eax.split.bit_width - 1);
+	return 1;
+fail1:
+	release_perfctr_nmi(perfctr_msr);
+fail:
+	return 0;
+}
+
+static void stop_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		return;
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
 void setup_apic_nmi_watchdog(void *unused)
 {
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
@@ -589,6 +694,11 @@ void setup_apic_nmi_watchdog(void *unused)
 				return;
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				if (!setup_intel_arch_watchdog())
+					return;
+				break;
+			}
 			if (!setup_p4_watchdog())
 				return;
 			break;
@@ -620,6 +730,10 @@ void stop_apic_nmi_watchdog(void *unused)
 			stop_k7_watchdog();
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				stop_intel_arch_watchdog();
+				break;
+			}
 			stop_p4_watchdog();
 			break;
 		default:
@@ -724,7 +838,13 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 				dummy &= ~P4_CCCR_OVF;
 	 			wrmsrl(wd->cccr_msr, dummy);
 	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
-	 		}
+	 		} else if (wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
+				/*
+				 * ArchPerfom/Core Duo needs to re-unmask
+				 * the apic vector
+				 */
+				apic_write(APIC_LVTPC, APIC_DM_NMI);
+			}
 			/* start the cycle over again */
 			wrmsrl(wd->perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 			rc = 1;
diff --git a/include/asm-i386/intel_arch_perfmon.h b/include/asm-i386/intel_arch_perfmon.h
new file mode 100644
index 000000000000..b52cd60a075b
--- /dev/null
+++ b/include/asm-i386/intel_arch_perfmon.h
@@ -0,0 +1,31 @@
+#ifndef X86_INTEL_ARCH_PERFMON_H
+#define X86_INTEL_ARCH_PERFMON_H 1
+
+#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
+#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
+
+#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
+#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
+
+#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
+#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
+#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
+#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
+
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX (0)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT \
+				(1 << (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX))
+
+union cpuid10_eax {
+	struct {
+		unsigned int version_id:8;
+		unsigned int num_counters:8;
+		unsigned int bit_width:8;
+		unsigned int mask_length:8;
+	} split;
+	unsigned int full;
+};
+
+#endif	/* X86_INTEL_ARCH_PERFMON_H */
diff --git a/include/asm-x86_64/intel_arch_perfmon.h b/include/asm-x86_64/intel_arch_perfmon.h
new file mode 100644
index 000000000000..8633331420ec
--- /dev/null
+++ b/include/asm-x86_64/intel_arch_perfmon.h
@@ -0,0 +1,31 @@
+#ifndef X86_64_INTEL_ARCH_PERFMON_H
+#define X86_64_INTEL_ARCH_PERFMON_H 1
+
+#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
+#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
+
+#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
+#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
+
+#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
+#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
+#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
+#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
+
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX (0)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT \
+				(1 << (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX))
+
+union cpuid10_eax {
+	struct {
+		unsigned int version_id:8;
+		unsigned int num_counters:8;
+		unsigned int bit_width:8;
+		unsigned int mask_length:8;
+	} split;
+	unsigned int full;
+};
+
+#endif	/* X86_64_INTEL_ARCH_PERFMON_H */
-- 
cgit v1.2.3


From 81af4449af9c9b686a4eeeb00112614621655704 Mon Sep 17 00:00:00 2001
From: Vojtech Pavlik <vojtech@suse.cz>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Add macros for rdtscp

This patch adds macros for reading tsc via the RDTSCP instruction, as well
as writing the auxilliary MSR read by RDTSCP to msr.h

[AK: changed rdtscp definition for old binutils]

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/msr.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include')

diff --git a/include/asm-x86_64/msr.h b/include/asm-x86_64/msr.h
index 10f8b51cec8b..37e194169fac 100644
--- a/include/asm-x86_64/msr.h
+++ b/include/asm-x86_64/msr.h
@@ -66,14 +66,25 @@
 #define rdtscl(low) \
      __asm__ __volatile__ ("rdtsc" : "=a" (low) : : "edx")
 
+#define rdtscp(low,high,aux) \
+     asm volatile (".byte 0x0f,0x01,0xf9" : "=a" (low), "=d" (high), "=c" (aux))
+
 #define rdtscll(val) do { \
      unsigned int __a,__d; \
      asm volatile("rdtsc" : "=a" (__a), "=d" (__d)); \
      (val) = ((unsigned long)__a) | (((unsigned long)__d)<<32); \
 } while(0)
 
+#define rdtscpll(val, aux) do { \
+     unsigned long __a, __d; \
+     asm volatile (".byte 0x0f,0x01,0xf9" : "=a" (__a), "=d" (__d), "=c" (aux)); \
+     (val) = (__d << 32) | __a; \
+} while (0)
+
 #define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
 
+#define write_rdtscp_aux(val) wrmsr(0xc0000103, val, 0)
+
 #define rdpmc(counter,low,high) \
      __asm__ __volatile__("rdpmc" \
 			  : "=a" (low), "=d" (high) \
-- 
cgit v1.2.3


From a670fad0adb1cc6202a607d250f10bd380593905 Mon Sep 17 00:00:00 2001
From: Vojtech Pavlik <vojtech@suse.cz>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Add initalization of the RDTSCP auxilliary values

This patch adds initalization of the RDTSCP auxilliary values to CPU numbers
to time.c. If RDTSCP is available, the MSRs are written with the respective
values. It can be later used to initalize per-cpu timekeeping variables.

AK: Some cleanups. Move externs into headers and fix CPU hotplug.

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/smpboot.c |  2 ++
 arch/x86_64/kernel/time.c    | 47 +++++++++++++++++++++++++++++++++-----------
 include/asm-x86_64/proto.h   |  1 +
 3 files changed, 38 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 15555879ce95..582896f7d420 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1181,6 +1181,8 @@ void __init smp_cpus_done(unsigned int max_cpus)
 #endif
 
 	check_nmi_watchdog();
+
+	time_init_gtod();
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 7a9b18224182..97b9e46d1992 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -24,6 +24,8 @@
 #include <linux/device.h>
 #include <linux/sysdev.h>
 #include <linux/bcd.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
 #include <linux/kallsyms.h>
 #include <linux/acpi.h>
 #ifdef CONFIG_ACPI
@@ -49,7 +51,7 @@ static void cpufreq_delayed_get(void);
 extern void i8254_timer_resume(void);
 extern int using_apic_timer;
 
-static char *time_init_gtod(void);
+static char *timename = NULL;
 
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
@@ -893,11 +895,21 @@ static struct irqaction irq0 = {
 	timer_interrupt, IRQF_DISABLED, CPU_MASK_NONE, "timer", NULL, NULL
 };
 
-void __init time_init(void)
+static int __cpuinit
+time_cpu_notifier(struct notifier_block *nb, unsigned long action, void *hcpu)
 {
-	char *timename;
-	char *gtod;
+	unsigned cpu = (unsigned long) hcpu;
+	if (action == CPU_ONLINE &&
+		cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) {
+		unsigned p;
+		p = smp_processor_id() | (cpu_to_node(smp_processor_id())<<12);
+		write_rdtscp_aux(p);
+	}
+	return NOTIFY_DONE;
+}
 
+void __init time_init(void)
+{
 	if (nohpet)
 		vxtime.hpet_address = 0;
 
@@ -931,18 +943,19 @@ void __init time_init(void)
 	}
 
 	vxtime.mode = VXTIME_TSC;
-	gtod = time_init_gtod();
-
-	printk(KERN_INFO "time.c: Using %ld.%06ld MHz WALL %s GTOD %s timer.\n",
-	       vxtime_hz / 1000000, vxtime_hz % 1000000, timename, gtod);
-	printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
-		cpu_khz / 1000, cpu_khz % 1000);
 	vxtime.quot = (USEC_PER_SEC << US_SCALE) / vxtime_hz;
 	vxtime.tsc_quot = (USEC_PER_MSEC << US_SCALE) / cpu_khz;
 	vxtime.last_tsc = get_cycles_sync();
 	setup_irq(0, &irq0);
 
 	set_cyc2ns_scale(cpu_khz);
+
+	hotcpu_notifier(time_cpu_notifier, 0);
+	time_cpu_notifier(NULL, CPU_ONLINE, (void *)(long)smp_processor_id());
+
+#ifndef CONFIG_SMP
+	time_init_gtod();
+#endif
 }
 
 /*
@@ -973,12 +986,13 @@ __cpuinit int unsynchronized_tsc(void)
 /*
  * Decide what mode gettimeofday should use.
  */
-__init static char *time_init_gtod(void)
+void time_init_gtod(void)
 {
 	char *timetype;
 
 	if (unsynchronized_tsc())
 		notsc = 1;
+
 	if (vxtime.hpet_address && notsc) {
 		timetype = hpet_use_timer ? "HPET" : "PIT/HPET";
 		if (hpet_use_timer)
@@ -1001,7 +1015,16 @@ __init static char *time_init_gtod(void)
 		timetype = hpet_use_timer ? "HPET/TSC" : "PIT/TSC";
 		vxtime.mode = VXTIME_TSC;
 	}
-	return timetype;
+
+	printk(KERN_INFO "time.c: Using %ld.%06ld MHz WALL %s GTOD %s timer.\n",
+	       vxtime_hz / 1000000, vxtime_hz % 1000000, timename, timetype);
+	printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
+		cpu_khz / 1000, cpu_khz % 1000);
+	vxtime.quot = (USEC_PER_SEC << US_SCALE) / vxtime_hz;
+	vxtime.tsc_quot = (USEC_PER_MSEC << US_SCALE) / cpu_khz;
+	vxtime.last_tsc = get_cycles_sync();
+
+	set_cyc2ns_scale(cpu_khz);
 }
 
 __setup("report_lost_ticks", time_setup);
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 038fe1f47e6f..3b1c60247902 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -51,6 +51,7 @@ extern unsigned long long monotonic_base;
 extern int sysctl_vsyscall;
 extern int nohpet;
 extern unsigned long vxtime_hz;
+extern void time_init_gtod(void);
 
 extern int numa_setup(char *opt);
 
-- 
cgit v1.2.3


From c08c820508233b424deab3302bc404bbecc6493a Mon Sep 17 00:00:00 2001
From: Vojtech Pavlik <vojtech@suse.cz>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Add the vgetcpu vsyscall

This patch adds a vgetcpu vsyscall, which depending on the CPU RDTSCP
capability uses either the RDTSCP or CPUID to obtain a CPU and node
numbers and pass them to the program.

AK: Lots of changes over Vojtech's original code:
Better prototype for vgetcpu()
It's better to pass the cpu / node numbers as separate arguments
to avoid mistakes when going from SMP to NUMA.
Also add a fast time stamp based cache using a user supplied
argument to speed things more up.
Use fast method from Chuck Ebbert to retrieve node/cpu from
GDT limit instead of CPUID
Made sure RDTSCP init is always executed after node is known.
Drop printk

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/head.S        |  2 +-
 arch/x86_64/kernel/time.c        | 13 ++++---
 arch/x86_64/kernel/vmlinux.lds.S |  3 ++
 arch/x86_64/kernel/vsyscall.c    | 84 +++++++++++++++++++++++++++++++++++++++-
 include/asm-x86_64/segment.h     |  5 ++-
 include/asm-x86_64/smp.h         | 12 ++++--
 include/asm-x86_64/vsyscall.h    |  9 +++++
 include/linux/getcpu.h           | 16 ++++++++
 8 files changed, 130 insertions(+), 14 deletions(-)
 create mode 100644 include/linux/getcpu.h

(limited to 'include')

diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S
index c9739ca81d06..505ec4a57506 100644
--- a/arch/x86_64/kernel/head.S
+++ b/arch/x86_64/kernel/head.S
@@ -371,7 +371,7 @@ ENTRY(cpu_gdt_table)
 	.quad	0,0			/* TSS */
 	.quad	0,0			/* LDT */
 	.quad   0,0,0			/* three TLS descriptors */ 
-	.quad	0			/* unused */
+	.quad	0x0000f40000000000	/* node/CPU stored in limit */
 gdt_end:	
 	/* asm/segment.h:GDT_ENTRIES must match this */	
 	/* This should be a multiple of the cache line size */
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 97b9e46d1992..560ed944dc0e 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -899,12 +899,8 @@ static int __cpuinit
 time_cpu_notifier(struct notifier_block *nb, unsigned long action, void *hcpu)
 {
 	unsigned cpu = (unsigned long) hcpu;
-	if (action == CPU_ONLINE &&
-		cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) {
-		unsigned p;
-		p = smp_processor_id() | (cpu_to_node(smp_processor_id())<<12);
-		write_rdtscp_aux(p);
-	}
+	if (action == CPU_ONLINE)
+		vsyscall_set_cpu(cpu);
 	return NOTIFY_DONE;
 }
 
@@ -993,6 +989,11 @@ void time_init_gtod(void)
 	if (unsynchronized_tsc())
 		notsc = 1;
 
+ 	if (cpu_has(&boot_cpu_data, X86_FEATURE_RDTSCP))
+		vgetcpu_mode = VGETCPU_RDTSCP;
+	else
+		vgetcpu_mode = VGETCPU_LSL;
+
 	if (vxtime.hpet_address && notsc) {
 		timetype = hpet_use_timer ? "HPET" : "PIT/HPET";
 		if (hpet_use_timer)
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 7c4de31471d4..8d5a5149bb3a 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -99,6 +99,9 @@ SECTIONS
   .vxtime : AT(VLOAD(.vxtime)) { *(.vxtime) }
   vxtime = VVIRT(.vxtime);
 
+  .vgetcpu_mode : AT(VLOAD(.vgetcpu_mode)) { *(.vgetcpu_mode) }
+  vgetcpu_mode = VVIRT(.vgetcpu_mode);
+
   .wall_jiffies : AT(VLOAD(.wall_jiffies)) { *(.wall_jiffies) }
   wall_jiffies = VVIRT(.wall_jiffies);
 
diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index f603037df162..902783bc4d53 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -26,6 +26,7 @@
 #include <linux/seqlock.h>
 #include <linux/jiffies.h>
 #include <linux/sysctl.h>
+#include <linux/getcpu.h>
 
 #include <asm/vsyscall.h>
 #include <asm/pgtable.h>
@@ -33,11 +34,15 @@
 #include <asm/fixmap.h>
 #include <asm/errno.h>
 #include <asm/io.h>
+#include <asm/segment.h>
+#include <asm/desc.h>
+#include <asm/topology.h>
 
 #define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
 
 int __sysctl_vsyscall __section_sysctl_vsyscall = 1;
 seqlock_t __xtime_lock __section_xtime_lock = SEQLOCK_UNLOCKED;
+int __vgetcpu_mode __section_vgetcpu_mode;
 
 #include <asm/unistd.h>
 
@@ -127,9 +132,46 @@ time_t __vsyscall(1) vtime(time_t *t)
 	return __xtime.tv_sec;
 }
 
-long __vsyscall(2) venosys_0(void)
+/* Fast way to get current CPU and node.
+   This helps to do per node and per CPU caches in user space.
+   The result is not guaranteed without CPU affinity, but usually
+   works out because the scheduler tries to keep a thread on the same
+   CPU.
+
+   tcache must point to a two element sized long array.
+   All arguments can be NULL. */
+long __vsyscall(2)
+vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 {
-	return -ENOSYS;
+	unsigned int dummy, p;
+	unsigned long j = 0;
+
+	/* Fast cache - only recompute value once per jiffies and avoid
+	   relatively costly rdtscp/cpuid otherwise.
+	   This works because the scheduler usually keeps the process
+	   on the same CPU and this syscall doesn't guarantee its
+	   results anyways.
+	   We do this here because otherwise user space would do it on
+	   its own in a likely inferior way (no access to jiffies).
+	   If you don't like it pass NULL. */
+	if (tcache && tcache->t0 == (j = __jiffies)) {
+		p = tcache->t1;
+	} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
+		/* Load per CPU data from RDTSCP */
+		rdtscp(dummy, dummy, p);
+	} else {
+		/* Load per CPU data from GDT */
+		asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
+	}
+	if (tcache) {
+		tcache->t0 = j;
+		tcache->t1 = p;
+	}
+	if (cpu)
+		*cpu = p & 0xfff;
+	if (node)
+		*node = p >> 12;
+	return 0;
 }
 
 long __vsyscall(3) venosys_1(void)
@@ -200,6 +242,43 @@ static ctl_table kernel_root_table2[] = {
 
 #endif
 
+static void __cpuinit write_rdtscp_cb(void *info)
+{
+	write_rdtscp_aux((unsigned long)info);
+}
+
+void __cpuinit vsyscall_set_cpu(int cpu)
+{
+	unsigned long *d;
+	unsigned long node = 0;
+#ifdef CONFIG_NUMA
+	node = cpu_to_node[cpu];
+#endif
+	if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) {
+		void *info = (void *)((node << 12) | cpu);
+		/* Can happen on preemptive kernel */
+		if (get_cpu() == cpu)
+			write_rdtscp_cb(info);
+#ifdef CONFIG_SMP
+		else {
+			/* the notifier is unfortunately not executed on the
+			   target CPU */
+			smp_call_function_single(cpu,write_rdtscp_cb,info,0,1);
+		}
+#endif
+		put_cpu();
+	}
+
+	/* Store cpu number in limit so that it can be loaded quickly
+	   in user space in vgetcpu.
+	   12 bits for the CPU and 8 bits for the node. */
+	d = (unsigned long *)(cpu_gdt(cpu) + GDT_ENTRY_PER_CPU);
+	*d = 0x0f40000000000ULL;
+	*d |= cpu;
+	*d |= (node & 0xf) << 12;
+	*d |= (node >> 4) << 48;
+}
+
 static void __init map_vsyscall(void)
 {
 	extern char __vsyscall_0;
@@ -214,6 +293,7 @@ static int __init vsyscall_init(void)
 			VSYSCALL_ADDR(__NR_vgettimeofday)));
 	BUG_ON((unsigned long) &vtime != VSYSCALL_ADDR(__NR_vtime));
 	BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
+	BUG_ON((unsigned long) &vgetcpu != VSYSCALL_ADDR(__NR_vgetcpu));
 	map_vsyscall();
 #ifdef CONFIG_SYSCTL
 	register_sysctl_table(kernel_root_table2, 0);
diff --git a/include/asm-x86_64/segment.h b/include/asm-x86_64/segment.h
index d4bed33fb32c..334ddcdd8f92 100644
--- a/include/asm-x86_64/segment.h
+++ b/include/asm-x86_64/segment.h
@@ -20,15 +20,16 @@
 #define __USER_CS     0x33   /* 6*8+3 */ 
 #define __USER32_DS	__USER_DS 
 
-#define GDT_ENTRY_TLS 1
 #define GDT_ENTRY_TSS 8	/* needs two entries */
 #define GDT_ENTRY_LDT 10 /* needs two entries */
 #define GDT_ENTRY_TLS_MIN 12
 #define GDT_ENTRY_TLS_MAX 14
-/* 15 free */
 
 #define GDT_ENTRY_TLS_ENTRIES 3
 
+#define GDT_ENTRY_PER_CPU 15	/* Abused to load per CPU data from limit */
+#define __PER_CPU_SEG	(GDT_ENTRY_PER_CPU * 8 + 3)
+
 /* TLS indexes for 64bit - hardcoded in arch_prctl */
 #define FS_TLS 0	
 #define GS_TLS 1	
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6805e1feb300..d61547fd833b 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -133,13 +133,19 @@ static __inline int logical_smp_processor_id(void)
 	/* we don't want to mark this access volatile - bad code generation */
 	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
 }
-#endif
 
 #ifdef CONFIG_SMP
 #define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
 #else
 #define cpu_physical_id(cpu)		boot_cpu_id
-#endif
-
+static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
+				void *info, int retry, int wait)
+{
+	/* Disable interrupts here? */
+	func(info);
+	return 0;
+}
+#endif /* !CONFIG_SMP */
+#endif /* !__ASSEMBLY */
 #endif
 
diff --git a/include/asm-x86_64/vsyscall.h b/include/asm-x86_64/vsyscall.h
index 146b24402a5f..2281e9399b96 100644
--- a/include/asm-x86_64/vsyscall.h
+++ b/include/asm-x86_64/vsyscall.h
@@ -4,6 +4,7 @@
 enum vsyscall_num {
 	__NR_vgettimeofday,
 	__NR_vtime,
+	__NR_vgetcpu,
 };
 
 #define VSYSCALL_START (-10UL << 20)
@@ -15,6 +16,7 @@ enum vsyscall_num {
 #include <linux/seqlock.h>
 
 #define __section_vxtime __attribute__ ((unused, __section__ (".vxtime"), aligned(16)))
+#define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16)))
 #define __section_wall_jiffies __attribute__ ((unused, __section__ (".wall_jiffies"), aligned(16)))
 #define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16)))
 #define __section_sys_tz __attribute__ ((unused, __section__ (".sys_tz"), aligned(16)))
@@ -26,6 +28,9 @@ enum vsyscall_num {
 #define VXTIME_HPET	2
 #define VXTIME_PMTMR	3
 
+#define VGETCPU_RDTSCP	1
+#define VGETCPU_LSL	2
+
 struct vxtime_data {
 	long hpet_address;	/* HPET base address */
 	int last;
@@ -40,6 +45,7 @@ struct vxtime_data {
 
 /* vsyscall space (readonly) */
 extern struct vxtime_data __vxtime;
+extern int __vgetcpu_mode;
 extern struct timespec __xtime;
 extern volatile unsigned long __jiffies;
 extern unsigned long __wall_jiffies;
@@ -48,6 +54,7 @@ extern seqlock_t __xtime_lock;
 
 /* kernel space (writeable) */
 extern struct vxtime_data vxtime;
+extern int vgetcpu_mode;
 extern unsigned long wall_jiffies;
 extern struct timezone sys_tz;
 extern int sysctl_vsyscall;
@@ -55,6 +62,8 @@ extern seqlock_t xtime_lock;
 
 extern int sysctl_vsyscall;
 
+extern void vsyscall_set_cpu(int cpu);
+
 #define ARCH_HAVE_XTIME_LOCK 1
 
 #endif /* __KERNEL__ */
diff --git a/include/linux/getcpu.h b/include/linux/getcpu.h
new file mode 100644
index 000000000000..031ed3780e45
--- /dev/null
+++ b/include/linux/getcpu.h
@@ -0,0 +1,16 @@
+#ifndef _LINUX_GETCPU_H
+#define _LINUX_GETCPU_H 1
+
+/* Cache for getcpu() to speed it up. Results might be upto a jiffie
+   out of date, but will be faster.
+   User programs should not refer to the contents of this structure.
+   It is only a cache for vgetcpu(). It might change in future kernels.
+   The user program must store this information per thread (__thread)
+   If you want 100% accurate information pass NULL instead. */
+struct getcpu_cache {
+	unsigned long t0;
+	unsigned long t1;
+	unsigned long res[4];
+};
+
+#endif
-- 
cgit v1.2.3


From 3cfc348bf90ffaa777c188652aa297f04eb94de8 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] x86: Add portable getcpu call

For NUMA optimization and some other algorithms it is useful to have a fast
to get the current CPU and node numbers in user space.

x86-64 added a fast way to do this in a vsyscall. This adds a generic
syscall for other architectures to make it a generic portable facility.

I expect some of them will also implement it as a faster vsyscall.

The cache is an optimization for the x86-64 vsyscall optimization. Since
what the syscall returns is an approximation anyways and user space
often wants very fast results it can be cached for some time.  The norma
methods to get this information in user space are relatively slow

The vsyscall is in a better position to manage the cache because it has direct
access to a fast time stamp (jiffies). For the generic syscall optimization
it doesn't help much, but enforce a valid argument to keep programs
portable

I only added an i386 syscall entry for now. Other architectures can follow
as needed.

AK: Also added some cleanups from Andrew Morton

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/syscall_table.S |  1 +
 arch/x86_64/ia32/ia32entry.S     |  1 +
 include/asm-i386/unistd.h        |  3 ++-
 include/linux/syscalls.h         |  2 ++
 kernel/sys.c                     | 31 +++++++++++++++++++++++++++++++
 5 files changed, 37 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index dd63d4775398..7e639f78b0b9 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -317,3 +317,4 @@ ENTRY(sys_call_table)
 	.long sys_tee			/* 315 */
 	.long sys_vmsplice
 	.long sys_move_pages
+	.long sys_getcpu
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 30ed0f6f4a2b..32fd32bea07c 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -713,4 +713,5 @@ ia32_sys_call_table:
 	.quad sys_tee
 	.quad compat_sys_vmsplice
 	.quad compat_sys_move_pages
+	.quad sys_getcpu
 ia32_syscall_end:		
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index fc1c8ddae149..565d0897b205 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -323,10 +323,11 @@
 #define __NR_tee		315
 #define __NR_vmsplice		316
 #define __NR_move_pages		317
+#define __NR_getcpu		318
 
 #ifdef __KERNEL__
 
-#define NR_syscalls 318
+#define NR_syscalls 319
 
 /*
  * user-visible error numbers are in the range -1 - -128: see
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 008f04c56737..3f0f716225ec 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -53,6 +53,7 @@ struct mq_attr;
 struct compat_stat;
 struct compat_timeval;
 struct robust_list_head;
+struct getcpu_cache;
 
 #include <linux/types.h>
 #include <linux/aio_abi.h>
@@ -596,5 +597,6 @@ asmlinkage long sys_get_robust_list(int pid,
 				    size_t __user *len_ptr);
 asmlinkage long sys_set_robust_list(struct robust_list_head __user *head,
 				    size_t len);
+asmlinkage long sys_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *cache);
 
 #endif
diff --git a/kernel/sys.c b/kernel/sys.c
index e236f98f7ec5..3f894775488d 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -28,6 +28,7 @@
 #include <linux/tty.h>
 #include <linux/signal.h>
 #include <linux/cn_proc.h>
+#include <linux/getcpu.h>
 
 #include <linux/compat.h>
 #include <linux/syscalls.h>
@@ -2062,3 +2063,33 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,
 	}
 	return error;
 }
+
+asmlinkage long sys_getcpu(unsigned __user *cpup, unsigned __user *nodep,
+	   		   struct getcpu_cache __user *cache)
+{
+	int err = 0;
+	int cpu = raw_smp_processor_id();
+	if (cpup)
+		err |= put_user(cpu, cpup);
+	if (nodep)
+		err |= put_user(cpu_to_node(cpu), nodep);
+	if (cache) {
+		/*
+		 * The cache is not needed for this implementation,
+		 * but make sure user programs pass something
+		 * valid. vsyscall implementations can instead make
+		 * good use of the cache. Only use t0 and t1 because
+		 * these are available in both 32bit and 64bit ABI (no
+		 * need for a compat_getcpu). 32bit has enough
+		 * padding
+		 */
+		unsigned long t0, t1;
+		get_user(t0, &cache->t0);
+		get_user(t1, &cache->t1);
+		t0++;
+		t1++;
+		put_user(t0, &cache->t0);
+		put_user(t1, &cache->t1);
+	}
+	return err ? -EFAULT : 0;
+}
-- 
cgit v1.2.3


From 2f766d16062d0147edff91be15de4a950667ca42 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Clean up asm/smp.h includes

No need to include it from entry.S
Drop all the #ifdef __ASSEMBLY__

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S |  2 --
 include/asm-x86_64/smp.h   | 12 ------------
 2 files changed, 14 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index aa8d8939abc1..7eb1678e098d 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -27,10 +27,8 @@
  * - schedule it carefully for the final hardware.
  */
 
-#define ASSEMBLY 1
 #include <linux/linkage.h>
 #include <asm/segment.h>
-#include <asm/smp.h>
 #include <asm/cache.h>
 #include <asm/errno.h>
 #include <asm/dwarf2.h>
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index d61547fd833b..612d208961a2 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -4,15 +4,12 @@
 /*
  * We need the APIC definitions automatically as part of 'smp.h'
  */
-#ifndef __ASSEMBLY__
 #include <linux/threads.h>
 #include <linux/cpumask.h>
 #include <linux/bitops.h>
 extern int disable_apic;
-#endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
-#ifndef __ASSEMBLY__
 #include <asm/fixmap.h>
 #include <asm/mpspec.h>
 #ifdef CONFIG_X86_IO_APIC
@@ -21,10 +18,8 @@ extern int disable_apic;
 #include <asm/apic.h>
 #include <asm/thread_info.h>
 #endif
-#endif
 
 #ifdef CONFIG_SMP
-#ifndef ASSEMBLY
 
 #include <asm/pda.h>
 
@@ -83,13 +78,10 @@ extern void prefill_possible_map(void);
 extern unsigned num_processors;
 extern unsigned disabled_cpus;
 
-#endif /* !ASSEMBLY */
-
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 
 #endif
 
-#ifndef ASSEMBLY
 /*
  * Some lowlevel functions might want to know about
  * the real APIC ID <-> CPU # mapping.
@@ -111,8 +103,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 		return BAD_APICID;
 }
 
-#endif /* !ASSEMBLY */
-
 #ifndef CONFIG_SMP
 #define stack_smp_processor_id() 0
 #define safe_smp_processor_id() 0
@@ -127,7 +117,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 })
 #endif
 
-#ifndef __ASSEMBLY__
 static __inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
@@ -146,6 +135,5 @@ static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
 	return 0;
 }
 #endif /* !CONFIG_SMP */
-#endif /* !__ASSEMBLY */
 #endif
 
-- 
cgit v1.2.3


From d3a4f48d4866b8623ca9adde8ce4e5fde979c132 Mon Sep 17 00:00:00 2001
From: Stephane Eranian <eranian@hpl.hp.com>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] x86-64 TIF flags for debug regs and io bitmap in ctxsw

Hello,

Following my discussion with Andi. Here is a patch that introduces
two new TIF flags to simplify the context switch code in __switch_to().
The idea is to minimize the number of cache lines accessed in the common
case, i.e., when neither the debug registers nor the I/O bitmap are used.

This patch covers the x86-64 modifications. A patch for i386 follows.

Changelog:
	- add TIF_DEBUG to track when debug registers are active
	- add TIF_IO_BITMAP to track when I/O bitmap is used
	- modify __switch_to() to use the new TIF flags

<signed-off-by>: eranian@hpl.hp.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ptrace32.c      |  4 +++
 arch/x86_64/kernel/ioport.c      |  1 +
 arch/x86_64/kernel/process.c     | 73 +++++++++++++++++++++++-----------------
 arch/x86_64/kernel/ptrace.c      |  8 +++--
 include/asm-x86_64/thread_info.h |  7 ++++
 5 files changed, 60 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/ia32/ptrace32.c b/arch/x86_64/ia32/ptrace32.c
index 659c0722f6b8..72bf92a9d375 100644
--- a/arch/x86_64/ia32/ptrace32.c
+++ b/arch/x86_64/ia32/ptrace32.c
@@ -117,6 +117,10 @@ static int putreg32(struct task_struct *child, unsigned regno, u32 val)
 			if ((0x5454 >> ((val >> (16 + 4*i)) & 0xf)) & 1)
 			       return -EIO;
 		child->thread.debugreg7 = val; 
+		if (val)
+			set_tsk_thread_flag(child, TIF_DEBUG);
+		else
+			clear_tsk_thread_flag(child, TIF_DEBUG);
 		break; 
 		    
 	default:
diff --git a/arch/x86_64/kernel/ioport.c b/arch/x86_64/kernel/ioport.c
index b81614970ecc..fe063d3cfe42 100644
--- a/arch/x86_64/kernel/ioport.c
+++ b/arch/x86_64/kernel/ioport.c
@@ -56,6 +56,7 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
 
 		memset(bitmap, 0xff, IO_BITMAP_BYTES);
 		t->io_bitmap_ptr = bitmap;
+		set_thread_flag(TIF_IO_BITMAP);
 	}
 
 	/*
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index bb6745d13b8f..6e0527635b4c 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -350,6 +350,7 @@ void exit_thread(void)
 
 		kfree(t->io_bitmap_ptr);
 		t->io_bitmap_ptr = NULL;
+		clear_thread_flag(TIF_IO_BITMAP);
 		/*
 		 * Careful, clear this in the TSS too:
 		 */
@@ -369,6 +370,7 @@ void flush_thread(void)
 		if (t->flags & _TIF_IA32)
 			current_thread_info()->status |= TS_COMPAT;
 	}
+	t->flags &= ~_TIF_DEBUG;
 
 	tsk->thread.debugreg0 = 0;
 	tsk->thread.debugreg1 = 0;
@@ -461,7 +463,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 	asm("mov %%es,%0" : "=m" (p->thread.es));
 	asm("mov %%ds,%0" : "=m" (p->thread.ds));
 
-	if (unlikely(me->thread.io_bitmap_ptr != NULL)) { 
+	if (unlikely(test_tsk_thread_flag(me, TIF_IO_BITMAP))) {
 		p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
 		if (!p->thread.io_bitmap_ptr) {
 			p->thread.io_bitmap_max = 0;
@@ -469,6 +471,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 		}
 		memcpy(p->thread.io_bitmap_ptr, me->thread.io_bitmap_ptr,
 				IO_BITMAP_BYTES);
+		set_tsk_thread_flag(p, TIF_IO_BITMAP);
 	} 
 
 	/*
@@ -498,6 +501,40 @@ out:
  */
 #define loaddebug(thread,r) set_debugreg(thread->debugreg ## r, r)
 
+static inline void __switch_to_xtra(struct task_struct *prev_p,
+			     	    struct task_struct *next_p,
+			     	    struct tss_struct *tss)
+{
+	struct thread_struct *prev, *next;
+
+	prev = &prev_p->thread,
+	next = &next_p->thread;
+
+	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
+		loaddebug(next, 0);
+		loaddebug(next, 1);
+		loaddebug(next, 2);
+		loaddebug(next, 3);
+		/* no 4 and 5 */
+		loaddebug(next, 6);
+		loaddebug(next, 7);
+	}
+
+	if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
+		/*
+		 * Copy the relevant range of the IO bitmap.
+		 * Normally this is 128 bytes or less:
+		 */
+		memcpy(tss->io_bitmap, next->io_bitmap_ptr,
+		       max(prev->io_bitmap_max, next->io_bitmap_max));
+	} else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) {
+		/*
+		 * Clear any possible leftover bits:
+		 */
+		memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
+	}
+}
+
 /*
  *	switch_to(x,y) should switch tasks from x to y.
  *
@@ -586,37 +623,11 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 		  task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
 
 	/*
-	 * Now maybe reload the debug registers
+	 * Now maybe reload the debug registers and handle I/O bitmaps
 	 */
-	if (unlikely(next->debugreg7)) {
-		loaddebug(next, 0);
-		loaddebug(next, 1);
-		loaddebug(next, 2);
-		loaddebug(next, 3);
-		/* no 4 and 5 */
-		loaddebug(next, 6);
-		loaddebug(next, 7);
-	}
-
-
-	/* 
-	 * Handle the IO bitmap 
-	 */ 
-	if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr)) {
-		if (next->io_bitmap_ptr)
-			/*
-			 * Copy the relevant range of the IO bitmap.
-			 * Normally this is 128 bytes or less:
- 			 */
-			memcpy(tss->io_bitmap, next->io_bitmap_ptr,
-				max(prev->io_bitmap_max, next->io_bitmap_max));
-		else {
-			/*
-			 * Clear any possible leftover bits:
-			 */
-			memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
-		}
-	}
+	if (unlikely((task_thread_info(next_p)->flags & _TIF_WORK_CTXSW))
+	    || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))
+		__switch_to_xtra(prev_p, next_p, tss);
 
 	return prev_p;
 }
diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c
index 2d50024c9f30..d35ec1bc696a 100644
--- a/arch/x86_64/kernel/ptrace.c
+++ b/arch/x86_64/kernel/ptrace.c
@@ -420,9 +420,13 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 				if ((0x5554 >> ((data >> (16 + 4*i)) & 0xf)) & 1)
 					break;
 			if (i == 4) {
-				child->thread.debugreg7 = data;
+			  child->thread.debugreg7 = data;
+			  if (data)
+			  	set_tsk_thread_flag(child, TIF_DEBUG);
+			  else
+			  	clear_tsk_thread_flag(child, TIF_DEBUG);
 			  ret = 0;
-		  }
+		  	}
 		  break;
 		}
 		break;
diff --git a/include/asm-x86_64/thread_info.h b/include/asm-x86_64/thread_info.h
index 790c512a4369..787a08114b48 100644
--- a/include/asm-x86_64/thread_info.h
+++ b/include/asm-x86_64/thread_info.h
@@ -120,6 +120,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_FORK		18	/* ret_from_fork */
 #define TIF_ABI_PENDING		19
 #define TIF_MEMDIE		20
+#define TIF_DEBUG		21	/* uses debug registers */
+#define TIF_IO_BITMAP		22	/* uses I/O bitmap */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
@@ -133,6 +135,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_IA32		(1<<TIF_IA32)
 #define _TIF_FORK		(1<<TIF_FORK)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
+#define _TIF_DEBUG		(1<<TIF_DEBUG)
+#define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
@@ -140,6 +144,9 @@ static inline struct thread_info *stack_thread_info(void)
 /* work to do on any return to user space */
 #define _TIF_ALLWORK_MASK (0x0000FFFF & ~_TIF_SECCOMP)
 
+/* flags to check in __switch_to() */
+#define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP)
+
 #define PREEMPT_ACTIVE     0x10000000
 
 /*
-- 
cgit v1.2.3


From 0cb91a2293648507886563ccb91979cfc94d6a4b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] i386: Account spinlocks to the caller during profiling for
 !FP kernels

This ports the algorithm from x86-64 (with improvements) to i386.
Previously this only worked for frame pointer enabled kernels.
But spinlocks have a very simple stack frame that can be manually
analyzed. Do this.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/time.c   | 23 +++++++++++++++++++----
 include/asm-i386/ptrace.h |  4 ----
 kernel/spinlock.c         |  5 +++++
 3 files changed, 24 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index edd00f6cee37..5af802ef00b2 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -130,18 +130,33 @@ static int set_rtc_mmss(unsigned long nowtime)
 
 int timer_ack;
 
-#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
 unsigned long profile_pc(struct pt_regs *regs)
 {
 	unsigned long pc = instruction_pointer(regs);
 
-	if (!user_mode_vm(regs) && in_lock_functions(pc))
+#ifdef CONFIG_SMP
+	if (!user_mode_vm(regs) && in_lock_functions(pc)) {
+#ifdef CONFIG_FRAME_POINTER
 		return *(unsigned long *)(regs->ebp + 4);
-
+#else
+		unsigned long *sp;
+		if ((regs->xcs & 3) == 0)
+			sp = (unsigned long *)&regs->esp;
+		else
+			sp = (unsigned long *)regs->esp;
+		/* Return address is either directly at stack pointer
+		   or above a saved eflags. Eflags has bits 22-31 zero,
+		   kernel addresses don't. */
+ 		if (sp[0] >> 22)
+			return sp[0];
+		if (sp[1] >> 22)
+			return sp[1];
+#endif
+	}
+#endif
 	return pc;
 }
 EXPORT_SYMBOL(profile_pc);
-#endif
 
 /*
  * This is the same as the above, except we _also_ save the current
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index f324c53b6f9a..30a442ec2059 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -80,11 +80,7 @@ static inline int user_mode_vm(struct pt_regs *regs)
 	return ((regs->xcs & 3) | (regs->eflags & VM_MASK)) != 0;
 }
 #define instruction_pointer(regs) ((regs)->eip)
-#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
 extern unsigned long profile_pc(struct pt_regs *regs);
-#else
-#define profile_pc(regs) instruction_pointer(regs)
-#endif
 #endif /* __KERNEL__ */
 
 #endif
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index fb524b009eef..9644a41e0bef 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -7,6 +7,11 @@
  *
  * This file contains the spinlock/rwlock implementations for the
  * SMP and the DEBUG_SPINLOCK cases. (UP-nondebug inlines them)
+ *
+ * Note that some architectures have special knowledge about the
+ * stack frames of these functions in their profile_pc. If you
+ * change anything significant here that could change the stack
+ * frame contact the architecture maintainers.
  */
 
 #include <linux/linkage.h>
-- 
cgit v1.2.3


From b4062b16094038334d9bbadac0397a3fc9e981b0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Support patchable lock prefix for pure assembly files

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/alternative-asm.i | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 include/asm-x86_64/alternative-asm.i

(limited to 'include')

diff --git a/include/asm-x86_64/alternative-asm.i b/include/asm-x86_64/alternative-asm.i
new file mode 100644
index 000000000000..e4041f4fa4dc
--- /dev/null
+++ b/include/asm-x86_64/alternative-asm.i
@@ -0,0 +1,14 @@
+#include <linux/config.h>
+
+#ifdef CONFIG_SMP
+	.macro LOCK_PREFIX
+1:	lock
+	.section .smp_locks,"a"
+	.align 8
+	.quad 1b
+	.previous
+	.endm
+#else
+	.macro LOCK_PREFIX
+	.endm
+#endif
-- 
cgit v1.2.3


From 9a0b26e6bc4ae1979d9bcc6194e57a71b2b5cac6 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Clean up read write lock assembly

- Move the slow path fallbacks to their own assembly files
This makes them much easier to read and is needed for the next change.
- Add CFI annotations for unwinding (XXX need review)
- Remove constant case which can never happen with out of line spinlocks
- Use patchable LOCK prefixes
- Don't use lock sections anymore for inline code because they can't
be expressed by the unwinder (this adds one taken jump to the lock
fast path)

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/lib/Makefile      |  2 +-
 arch/x86_64/lib/rwlock.S      | 38 ++++++++++++++++++++++++
 arch/x86_64/lib/thunk.S       | 30 -------------------
 include/asm-x86_64/rwlock.h   | 68 +++++++------------------------------------
 include/asm-x86_64/spinlock.h | 11 ++-----
 5 files changed, 51 insertions(+), 98 deletions(-)
 create mode 100644 arch/x86_64/lib/rwlock.S

(limited to 'include')

diff --git a/arch/x86_64/lib/Makefile b/arch/x86_64/lib/Makefile
index ccef6ae747a3..b78d4170fce2 100644
--- a/arch/x86_64/lib/Makefile
+++ b/arch/x86_64/lib/Makefile
@@ -9,4 +9,4 @@ obj-y := io.o iomap_copy.o
 lib-y := csum-partial.o csum-copy.o csum-wrappers.o delay.o \
 	usercopy.o getuser.o putuser.o  \
 	thunk.o clear_page.o copy_page.o bitstr.o bitops.o
-lib-y += memcpy.o memmove.o memset.o copy_user.o
+lib-y += memcpy.o memmove.o memset.o copy_user.o rwlock.o
diff --git a/arch/x86_64/lib/rwlock.S b/arch/x86_64/lib/rwlock.S
new file mode 100644
index 000000000000..0cde1f807314
--- /dev/null
+++ b/arch/x86_64/lib/rwlock.S
@@ -0,0 +1,38 @@
+/* Slow paths of read/write spinlocks. */
+
+#include <linux/linkage.h>
+#include <asm/rwlock.h>
+#include <asm/alternative-asm.i>
+#include <asm/dwarf2.h>
+
+/* rdi:	pointer to rwlock_t */
+ENTRY(__write_lock_failed)
+	CFI_STARTPROC
+	LOCK_PREFIX
+	addl $RW_LOCK_BIAS,(%rdi)
+1:	rep
+	nop
+	cmpl $RW_LOCK_BIAS,(%rdi)
+	jne 1b
+	LOCK_PREFIX
+	subl $RW_LOCK_BIAS,(%rdi)
+	jnz  __write_lock_failed
+	ret
+	CFI_ENDPROC
+END(__write_lock_failed)
+
+/* rdi:	pointer to rwlock_t */
+ENTRY(__read_lock_failed)
+	CFI_STARTPROC
+	LOCK_PREFIX
+	incl (%rdi)
+1:	rep
+	nop
+	cmpl $1,(%rdi)
+	js 1b
+	LOCK_PREFIX
+	decl (%rdi)
+	js __read_lock_failed
+	ret
+	CFI_ENDPROC
+END(__read_lock_failed)
diff --git a/arch/x86_64/lib/thunk.S b/arch/x86_64/lib/thunk.S
index 332ea5dff916..6cff27c775ae 100644
--- a/arch/x86_64/lib/thunk.S
+++ b/arch/x86_64/lib/thunk.S
@@ -67,33 +67,3 @@ restore_norax:
 	RESTORE_ARGS 1
 	ret
 	CFI_ENDPROC
-
-#ifdef CONFIG_SMP
-/* Support for read/write spinlocks. */
-	.text
-/* rax:	pointer to rwlock_t */	
-ENTRY(__write_lock_failed)
-	lock
-	addl $RW_LOCK_BIAS,(%rax)
-1:	rep
-	nop
-	cmpl $RW_LOCK_BIAS,(%rax)
-	jne 1b
-	lock 
-	subl $RW_LOCK_BIAS,(%rax)
-	jnz  __write_lock_failed
-	ret
-
-/* rax:	pointer to rwlock_t */	
-ENTRY(__read_lock_failed)
-	lock
-	incl (%rax)
-1:	rep
-	nop
-	cmpl $1,(%rax)
-	js 1b
-	lock
-	decl (%rax)
-	js __read_lock_failed
-	ret
-#endif
diff --git a/include/asm-x86_64/rwlock.h b/include/asm-x86_64/rwlock.h
index dea0e9459264..28a080d23119 100644
--- a/include/asm-x86_64/rwlock.h
+++ b/include/asm-x86_64/rwlock.h
@@ -18,69 +18,21 @@
 #ifndef _ASM_X86_64_RWLOCK_H
 #define _ASM_X86_64_RWLOCK_H
 
-#include <linux/stringify.h>
-
 #define RW_LOCK_BIAS		 0x01000000
-#define RW_LOCK_BIAS_STR	"0x01000000"
+#define RW_LOCK_BIAS_STR	 "0x01000000"
 
-#define __build_read_lock_ptr(rw, helper)   \
+#define __build_read_lock(rw)   \
 	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t" \
-		     "js 2f\n" \
-		     "1:\n" \
-		    LOCK_SECTION_START("") \
-		     "2:\tcall " helper "\n\t" \
-		     "jmp 1b\n" \
-		    LOCK_SECTION_END \
-		     ::"a" (rw) : "memory")
-
-#define __build_read_lock_const(rw, helper)   \
-	asm volatile(LOCK_PREFIX "subl $1,%0\n\t" \
-		     "js 2f\n" \
+		     "jns 1f\n" \
+		     "call __read_lock_failed\n" \
 		     "1:\n" \
-		    LOCK_SECTION_START("") \
-		     "2:\tpushq %%rax\n\t" \
-		     "leaq %0,%%rax\n\t" \
-		     "call " helper "\n\t" \
-		     "popq %%rax\n\t" \
-		     "jmp 1b\n" \
-		    LOCK_SECTION_END \
-		     :"=m" (*((volatile int *)rw))::"memory")
-
-#define __build_read_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_read_lock_const(rw, helper); \
-						else \
-							__build_read_lock_ptr(rw, helper); \
-					} while (0)
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
 
-#define __build_write_lock_ptr(rw, helper) \
-	asm volatile(LOCK_PREFIX "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
-		     "jnz 2f\n" \
+#define __build_write_lock(rw) \
+	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t" \
+		     "jz 1f\n" \
+		     "\tcall __write_lock_failed\n\t" \
 		     "1:\n" \
-		     LOCK_SECTION_START("") \
-		     "2:\tcall " helper "\n\t" \
-		     "jmp 1b\n" \
-		     LOCK_SECTION_END \
-		     ::"a" (rw) : "memory")
-
-#define __build_write_lock_const(rw, helper) \
-	asm volatile(LOCK_PREFIX "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
-		     "jnz 2f\n" \
-		     "1:\n" \
-		    LOCK_SECTION_START("") \
-		     "2:\tpushq %%rax\n\t" \
-		     "leaq %0,%%rax\n\t" \
-		     "call " helper "\n\t" \
-		     "popq %%rax\n\t" \
-		     "jmp 1b\n" \
-		    LOCK_SECTION_END \
-		     :"=m" (*((volatile long *)rw))::"memory")
-
-#define __build_write_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_write_lock_const(rw, helper); \
-						else \
-							__build_write_lock_ptr(rw, helper); \
-					} while (0)
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
 
 #endif
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index 248a79f0eaff..a8e3d89f591f 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -79,13 +79,6 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
  *
  * On x86, we implement read-write locks as a 32-bit counter
  * with the high bit (sign) being the "contended" bit.
- *
- * The inline assembly is non-obvious. Think about it.
- *
- * Changed to use the same technique as rw semaphores.  See
- * semaphore.h for details.  -ben
- *
- * the helpers are in arch/i386/kernel/semaphore.c
  */
 
 #define __raw_read_can_lock(x)		((int)(x)->lock > 0)
@@ -93,12 +86,12 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	__build_read_lock(rw, "__read_lock_failed");
+	__build_read_lock(rw);
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	__build_write_lock(rw, "__write_lock_failed");
+	__build_write_lock(rw);
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.3


From 1a015b5644ec6df0a2c4cbeff1f8a3d24ba0478e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: Remove const case for rwlocks

rwlocks are now out of line, so it near never triggers.  Also it was
incompatible with the new dwarf2 unwinder because it had unannotiatable
push/pops.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/rwlock.h | 38 ++------------------------------------
 1 file changed, 2 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h
index 87c069ccba08..f40ccbd8cb7f 100644
--- a/include/asm-i386/rwlock.h
+++ b/include/asm-i386/rwlock.h
@@ -20,52 +20,18 @@
 #define RW_LOCK_BIAS		 0x01000000
 #define RW_LOCK_BIAS_STR	"0x01000000"
 
-#define __build_read_lock_ptr(rw, helper)   \
+#define __build_read_lock(rw, helper)   \
 	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" \
 			"jns 1f\n" \
 			"call " helper "\n\t" \
 			"1:\n" \
 			::"a" (rw) : "memory")
 
-#define __build_read_lock_const(rw, helper)   \
-	asm volatile(LOCK_PREFIX " subl $1,%0\n\t" \
-			"jns 1f\n" \
-			"pushl %%eax\n\t" \
-			"leal %0,%%eax\n\t" \
-			"call " helper "\n\t" \
-			"popl %%eax\n\t" \
-			"1:\n" \
-			:"+m" (*(volatile int *)rw) : : "memory")
-
-#define __build_read_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_read_lock_const(rw, helper); \
-						else \
-							__build_read_lock_ptr(rw, helper); \
-					} while (0)
-
-#define __build_write_lock_ptr(rw, helper) \
+#define __build_write_lock(rw, helper) \
 	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
 			"jz 1f\n" \
 			"call " helper "\n\t" \
 			"1:\n" \
 			::"a" (rw) : "memory")
 
-#define __build_write_lock_const(rw, helper) \
-	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
-			"jz 1f\n" \
-			"pushl %%eax\n\t" \
-			"leal %0,%%eax\n\t" \
-			"call " helper "\n\t" \
-			"popl %%eax\n\t" \
-			"1:\n" \
-			:"+m" (*(volatile int *)rw) : : "memory")
-
-#define __build_write_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_write_lock_const(rw, helper); \
-						else \
-							__build_write_lock_ptr(rw, helper); \
-					} while (0)
-
 #endif
-- 
cgit v1.2.3


From b06babac45e1546dfb504f1f25eb0495632bfc41 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Add proper alignment to ENTRY

Previously it didn't align. Use the same one as the C compiler
in blended mode, which is good for K8 and Core2 and doesn't hurt
on P4.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S   | 3 +--
 include/asm-x86_64/linkage.h | 2 +-
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 7eb1678e098d..2dc5c01f754d 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -617,8 +617,7 @@ retint_signal:
 #ifdef CONFIG_PREEMPT
 	/* Returning to kernel space. Check if we need preemption */
 	/* rcx:	 threadinfo. interrupts off. */
-	.p2align
-retint_kernel:	
+ENTRY(retint_kernel)
 	cmpl $0,threadinfo_preempt_count(%rcx)
 	jnz  retint_restore_args
 	bt  $TIF_NEED_RESCHED,threadinfo_flags(%rcx)
diff --git a/include/asm-x86_64/linkage.h b/include/asm-x86_64/linkage.h
index 291c2d01c44f..b5f39d0189ce 100644
--- a/include/asm-x86_64/linkage.h
+++ b/include/asm-x86_64/linkage.h
@@ -1,6 +1,6 @@
 #ifndef __ASM_LINKAGE_H
 #define __ASM_LINKAGE_H
 
-/* Nothing to see here... */
+#define __ALIGN .p2align 4,,15
 
 #endif
-- 
cgit v1.2.3


From 07c9819b31eda7954feddc83f2fae035f31c11e1 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: add alternative-asm.h to allow LOCK_PREFIX replacement
 in .S files

LOCK_PREFIX is replaced by nops on UP systems, so it has to be a special
macro.  Previously this was only possible from C. Allow it for pure
assembly files too. Similar to earlier x86-64 patch.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/alternative-asm.i | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 include/asm-i386/alternative-asm.i

(limited to 'include')

diff --git a/include/asm-i386/alternative-asm.i b/include/asm-i386/alternative-asm.i
new file mode 100644
index 000000000000..6c47e3b9484b
--- /dev/null
+++ b/include/asm-i386/alternative-asm.i
@@ -0,0 +1,14 @@
+#include <linux/config.h>
+
+#ifdef CONFIG_SMP
+	.macro LOCK_PREFIX
+1:	lock
+	.section .smp_locks,"a"
+	.align 4
+	.long 1b
+	.previous
+	.endm
+#else
+	.macro LOCK_PREFIX
+	.endm
+#endif
-- 
cgit v1.2.3


From ecaf45ee5ce60afe7cc46e91d82c1b0cbda09387 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: Redo semaphore and rwlock assembly helpers

- Move them to a pure assembly file. Previously they were in
a C file that only consisted of inline assembly. Doing it in pure
assembler is much nicer.
- Add a frame.i include with FRAME/ENDFRAME macros to easily
add frame pointers to assembly functions
- Add dwarf2 annotation to them so that the new dwarf2 unwinder
doesn't get stuck on them
- Random cleanups

Includes feedback from Jan Beulich and a UML build fix from Andrew
Morton.

Cc: jbeulich@novell.com
Cc: jdike@addtoit.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/Makefile        |   2 +-
 arch/i386/kernel/semaphore.c     | 134 ----------------------------------
 arch/i386/lib/Makefile           |   2 +-
 arch/i386/lib/semaphore.S        | 154 +++++++++++++++++++++++++++++++++++++++
 arch/um/sys-i386/Makefile        |   2 +-
 include/asm-i386/frame.i         |  24 ++++++
 include/asm-um/alternative-asm.i |   6 ++
 include/asm-um/frame.i           |   6 ++
 8 files changed, 193 insertions(+), 137 deletions(-)
 delete mode 100644 arch/i386/kernel/semaphore.c
 create mode 100644 arch/i386/lib/semaphore.S
 create mode 100644 include/asm-i386/frame.i
 create mode 100644 include/asm-um/alternative-asm.i
 create mode 100644 include/asm-um/frame.i

(limited to 'include')

diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 5427a842e841..dab497472deb 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -4,7 +4,7 @@
 
 extra-y := head.o init_task.o vmlinux.lds
 
-obj-y	:= process.o semaphore.o signal.o entry.o traps.o irq.o \
+obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \
 		pci-dma.o i386_ksyms.o i387.o bootflag.o \
 		quirks.o i8237.o topology.o alternative.o i8253.o tsc.o
diff --git a/arch/i386/kernel/semaphore.c b/arch/i386/kernel/semaphore.c
deleted file mode 100644
index 98352c374c76..000000000000
--- a/arch/i386/kernel/semaphore.c
+++ /dev/null
@@ -1,134 +0,0 @@
-/*
- * i386 semaphore implementation.
- *
- * (C) Copyright 1999 Linus Torvalds
- *
- * Portions Copyright 1999 Red Hat, Inc.
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org>
- */
-#include <asm/semaphore.h>
-
-/*
- * The semaphore operations have a special calling sequence that
- * allow us to do a simpler in-line version of them. These routines
- * need to convert that sequence back into the C sequence when
- * there is contention on the semaphore.
- *
- * %eax contains the semaphore pointer on entry. Save the C-clobbered
- * registers (%eax, %edx and %ecx) except %eax whish is either a return
- * value or just clobbered..
- */
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __down_failed\n"
-"__down_failed:\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"pushl %ebp\n\t"
-	"movl  %esp,%ebp\n\t"
-#endif
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __down\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"movl %ebp,%esp\n\t"
-	"popl %ebp\n\t"
-#endif
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __down_failed_interruptible\n"
-"__down_failed_interruptible:\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"pushl %ebp\n\t"
-	"movl  %esp,%ebp\n\t"
-#endif
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __down_interruptible\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"movl %ebp,%esp\n\t"
-	"popl %ebp\n\t"
-#endif
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __down_failed_trylock\n"
-"__down_failed_trylock:\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"pushl %ebp\n\t"
-	"movl  %esp,%ebp\n\t"
-#endif
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __down_trylock\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"movl %ebp,%esp\n\t"
-	"popl %ebp\n\t"
-#endif
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __up_wakeup\n"
-"__up_wakeup:\n\t"
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __up\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-	"ret"
-);
-
-/*
- * rw spinlock fallbacks
- */
-#if defined(CONFIG_SMP)
-asm(
-".section .sched.text\n"
-".align	4\n"
-".globl	__write_lock_failed\n"
-"__write_lock_failed:\n\t"
-	LOCK_PREFIX "addl	$" RW_LOCK_BIAS_STR ",(%eax)\n"
-"1:	rep; nop\n\t"
-	"cmpl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
-	"jne	1b\n\t"
-	LOCK_PREFIX "subl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
-	"jnz	__write_lock_failed\n\t"
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align	4\n"
-".globl	__read_lock_failed\n"
-"__read_lock_failed:\n\t"
-	LOCK_PREFIX "incl	(%eax)\n"
-"1:	rep; nop\n\t"
-	"cmpl	$1,(%eax)\n\t"
-	"js	1b\n\t"
-	LOCK_PREFIX "decl	(%eax)\n\t"
-	"js	__read_lock_failed\n\t"
-	"ret"
-);
-#endif
diff --git a/arch/i386/lib/Makefile b/arch/i386/lib/Makefile
index 914933e9ec3d..d86a548b8d54 100644
--- a/arch/i386/lib/Makefile
+++ b/arch/i386/lib/Makefile
@@ -4,6 +4,6 @@
 
 
 lib-y = checksum.o delay.o usercopy.o getuser.o putuser.o memcpy.o strstr.o \
-	bitops.o
+	bitops.o semaphore.o
 
 lib-$(CONFIG_X86_USE_3DNOW) += mmx.o
diff --git a/arch/i386/lib/semaphore.S b/arch/i386/lib/semaphore.S
new file mode 100644
index 000000000000..e16ff1696390
--- /dev/null
+++ b/arch/i386/lib/semaphore.S
@@ -0,0 +1,154 @@
+/*
+ * i386 semaphore implementation.
+ *
+ * (C) Copyright 1999 Linus Torvalds
+ *
+ * Portions Copyright 1999 Red Hat, Inc.
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org>
+ */
+
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <asm/rwlock.h>
+#include <asm/alternative-asm.i>
+#include <asm/frame.i>
+#include <asm/dwarf2.h>
+
+/*
+ * The semaphore operations have a special calling sequence that
+ * allow us to do a simpler in-line version of them. These routines
+ * need to convert that sequence back into the C sequence when
+ * there is contention on the semaphore.
+ *
+ * %eax contains the semaphore pointer on entry. Save the C-clobbered
+ * registers (%eax, %edx and %ecx) except %eax whish is either a return
+ * value or just clobbered..
+ */
+	.section .sched.text
+ENTRY(__down_failed)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __down
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__down_failed)
+
+ENTRY(__down_failed_interruptible)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __down_interruptible
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__down_failed_interruptible)
+
+ENTRY(__down_failed_trylock)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __down_trylock
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__down_failed_trylock)
+
+ENTRY(__up_wakeup)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __up
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__up_wakeup)
+
+/*
+ * rw spinlock fallbacks
+ */
+#ifdef CONFIG_SMP
+ENTRY(__write_lock_failed)
+	CFI_STARTPROC simple
+	FRAME
+2: 	LOCK_PREFIX
+	addl	$ RW_LOCK_BIAS,(%eax)
+1:	rep; nop
+	cmpl	$ RW_LOCK_BIAS,(%eax)
+	jne	1b
+	LOCK_PREFIX
+	subl	$ RW_LOCK_BIAS,(%eax)
+	jnz	2b
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__write_lock_failed)
+
+ENTRY(__read_lock_failed)
+	CFI_STARTPROC
+	FRAME
+2: 	LOCK_PREFIX
+	incl	(%eax)
+1:	rep; nop
+	cmpl	$1,(%eax)
+	js	1b
+	LOCK_PREFIX
+	decl	(%eax)
+	js	2b
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__read_lock_failed)
+
+#endif
diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile
index 374d61a19439..6289779ca67c 100644
--- a/arch/um/sys-i386/Makefile
+++ b/arch/um/sys-i386/Makefile
@@ -4,7 +4,7 @@ obj-y = bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \
 
 obj-$(CONFIG_MODE_SKAS) += stub.o stub_segv.o
 
-subarch-obj-y = lib/bitops.o kernel/semaphore.o
+subarch-obj-y = lib/bitops.o lib/semaphore.o
 subarch-obj-$(CONFIG_HIGHMEM) += mm/highmem.o
 subarch-obj-$(CONFIG_MODULES) += kernel/module.o
 
diff --git a/include/asm-i386/frame.i b/include/asm-i386/frame.i
new file mode 100644
index 000000000000..4d68ddce18b6
--- /dev/null
+++ b/include/asm-i386/frame.i
@@ -0,0 +1,24 @@
+#include <linux/config.h>
+#include <asm/dwarf2.h>
+
+/* The annotation hides the frame from the unwinder and makes it look
+   like a ordinary ebp save/restore. This avoids some special cases for
+   frame pointer later */
+#ifdef CONFIG_FRAME_POINTER
+	.macro FRAME
+	pushl %ebp
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ebp,0
+	movl %esp,%ebp
+	.endm
+	.macro ENDFRAME
+	popl %ebp
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ebp
+	.endm
+#else
+	.macro FRAME
+	.endm
+	.macro ENDFRAME
+	.endm
+#endif
diff --git a/include/asm-um/alternative-asm.i b/include/asm-um/alternative-asm.i
new file mode 100644
index 000000000000..cae9faca132f
--- /dev/null
+++ b/include/asm-um/alternative-asm.i
@@ -0,0 +1,6 @@
+#ifndef __UM_ALTERNATIVE_ASM_I
+#define __UM_ALTERNATIVE_ASM_I
+
+#include "asm/arch/alternative-asm.i"
+
+#endif
diff --git a/include/asm-um/frame.i b/include/asm-um/frame.i
new file mode 100644
index 000000000000..09d5dca5d928
--- /dev/null
+++ b/include/asm-um/frame.i
@@ -0,0 +1,6 @@
+#ifndef __UM_FRAME_I
+#define __UM_FRAME_I
+
+#include "asm/arch/frame.i"
+
+#endif
-- 
cgit v1.2.3


From 7f11d8a5efd625ffa41cde1d8472f85e885478ec Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove all ifdefs for local/io apic

IO-APIC or local APIC can only be disabled at runtime anyways and
Kconfig has forced these options on for a long time now.

The Kconfigs are kept only now for the benefit of the shared acpi
boot.c code.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/Makefile  | 4 ++--
 arch/x86_64/kernel/apic.c    | 6 +-----
 arch/x86_64/kernel/crash.c   | 2 --
 arch/x86_64/kernel/entry.S   | 2 --
 arch/x86_64/kernel/head64.c  | 2 --
 arch/x86_64/kernel/i8259.c   | 9 ---------
 arch/x86_64/kernel/irq.c     | 6 ------
 arch/x86_64/kernel/mpparse.c | 9 ---------
 arch/x86_64/kernel/setup.c   | 6 ------
 arch/x86_64/kernel/smpboot.c | 5 -----
 arch/x86_64/kernel/time.c    | 8 --------
 arch/x86_64/kernel/traps.c   | 2 --
 include/asm-x86_64/apic.h    | 4 ----
 include/asm-x86_64/fixmap.h  | 4 ----
 include/asm-x86_64/io_apic.h | 6 ------
 include/asm-x86_64/irq.h     | 2 --
 include/asm-x86_64/mpspec.h  | 2 --
 include/asm-x86_64/smp.h     | 4 ----
 18 files changed, 3 insertions(+), 80 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index b5aaeafc1cd3..0ef39553c0f6 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -20,8 +20,8 @@ obj-$(CONFIG_X86_MSR)		+= msr.o
 obj-$(CONFIG_MICROCODE)		+= microcode.o
 obj-$(CONFIG_X86_CPUID)		+= cpuid.o
 obj-$(CONFIG_SMP)		+= smp.o smpboot.o trampoline.o
-obj-$(CONFIG_X86_LOCAL_APIC)	+= apic.o  nmi.o
-obj-$(CONFIG_X86_IO_APIC)	+= io_apic.o mpparse.o \
+obj-y				+= apic.o  nmi.o
+obj-y				+= io_apic.o mpparse.o \
 		genapic.o genapic_cluster.o genapic_flat.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump.o
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 63d9b037afc6..c6ca100a7dea 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -661,7 +661,6 @@ void __init init_apic_mappings(void)
 	 */
 	boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
 
-#ifdef CONFIG_X86_IO_APIC
 	{
 		unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
 		int i;
@@ -679,7 +678,6 @@ void __init init_apic_mappings(void)
 			idx++;
 		}
 	}
-#endif
 }
 
 /*
@@ -1129,12 +1127,10 @@ int __init APIC_init_uniprocessor (void)
 
 	setup_local_APIC();
 
-#ifdef CONFIG_X86_IO_APIC
 	if (smp_found_config && !skip_ioapic_setup && nr_ioapics)
-			setup_IO_APIC();
+		setup_IO_APIC();
 	else
 		nr_ioapics = 0;
-#endif
 	setup_boot_APIC_clock();
 	check_nmi_watchdog();
 	return 0;
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index 7d7a9e75f70c..3525f884af82 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -194,9 +194,7 @@ void machine_crash_shutdown(struct pt_regs *regs)
 	if(cpu_has_apic)
 		 disable_local_APIC();
 
-#if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
-#endif
 
 	crash_save_self(regs);
 }
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 2dc5c01f754d..a0cf36ba7c56 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -676,7 +676,6 @@ ENTRY(call_function_interrupt)
 END(call_function_interrupt)
 #endif
 
-#ifdef CONFIG_X86_LOCAL_APIC	
 ENTRY(apic_timer_interrupt)
 	apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
 END(apic_timer_interrupt)
@@ -688,7 +687,6 @@ END(error_interrupt)
 ENTRY(spurious_interrupt)
 	apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt
 END(spurious_interrupt)
-#endif
 				
 /*
  * Exception entry points.
diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index 36647ce6aecb..bacbd75c63ad 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -111,10 +111,8 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	if (s != NULL)
 		numa_setup(s+5);
 #endif
-#ifdef CONFIG_X86_IO_APIC
 	if (strstr(saved_command_line, "disableapic"))
 		disable_apic = 1;
-#endif
 	/* You need early console to see that */
 	if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
 		panic("Kernel too big for kernel mapping\n");
diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c
index 0434b1f8e3dd..d2f5acde5749 100644
--- a/arch/x86_64/kernel/i8259.c
+++ b/arch/x86_64/kernel/i8259.c
@@ -55,7 +55,6 @@
  */
 BUILD_16_IRQS(0x0)
 
-#ifdef CONFIG_X86_LOCAL_APIC
 /*
  * The IO-APIC gives us many more interrupt sources. Most of these 
  * are unused but an SMP system is supposed to have enough memory ...
@@ -75,8 +74,6 @@ BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd)
 	BUILD_15_IRQS(0xe)
 #endif
 
-#endif
-
 #undef BUILD_16_IRQS
 #undef BUILD_15_IRQS
 #undef BI
@@ -100,7 +97,6 @@ BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd)
 void (*interrupt[NR_IRQS])(void) = {
 	IRQLIST_16(0x0),
 
-#ifdef CONFIG_X86_IO_APIC
 			 IRQLIST_16(0x1), IRQLIST_16(0x2), IRQLIST_16(0x3),
 	IRQLIST_16(0x4), IRQLIST_16(0x5), IRQLIST_16(0x6), IRQLIST_16(0x7),
 	IRQLIST_16(0x8), IRQLIST_16(0x9), IRQLIST_16(0xa), IRQLIST_16(0xb),
@@ -110,7 +106,6 @@ void (*interrupt[NR_IRQS])(void) = {
 	, IRQLIST_15(0xe)
 #endif
 
-#endif
 };
 
 #undef IRQ
@@ -453,9 +448,7 @@ void __init init_ISA_irqs (void)
 {
 	int i;
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	init_bsp_APIC();
-#endif
 	init_8259A(0);
 
 	for (i = 0; i < NR_IRQS; i++) {
@@ -581,14 +574,12 @@ void __init init_IRQ(void)
 	set_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt);
 	set_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt);
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	/* self generated IPI for local APIC timer */
 	set_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);
 
 	/* IPI vectors for APIC spurious and error interrupts */
 	set_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	set_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
-#endif
 
 	/*
 	 * Set the clock to HZ Hz, we already have a valid
diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c
index 5221a53e90c1..bfbcf92cfa4a 100644
--- a/arch/x86_64/kernel/irq.c
+++ b/arch/x86_64/kernel/irq.c
@@ -20,11 +20,9 @@
 #include <asm/idle.h>
 
 atomic_t irq_err_count;
-#ifdef CONFIG_X86_IO_APIC
 #ifdef APIC_MISMATCH_DEBUG
 atomic_t irq_mis_count;
 #endif
-#endif
 
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 /*
@@ -92,17 +90,13 @@ skip:
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
 		seq_putc(p, '\n');
-#ifdef CONFIG_X86_LOCAL_APIC
 		seq_printf(p, "LOC: ");
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
 		seq_putc(p, '\n');
-#endif
 		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
-#ifdef CONFIG_X86_IO_APIC
 #ifdef APIC_MISMATCH_DEBUG
 		seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count));
-#endif
 #endif
 	}
 	return 0;
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index a1ab4197f8a1..c73dd1f41e65 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -74,14 +74,10 @@ physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
 /* ACPI MADT entry parsing functions */
 #ifdef CONFIG_ACPI
 extern struct acpi_boot_flags acpi_boot;
-#ifdef CONFIG_X86_LOCAL_APIC
 extern int acpi_parse_lapic (acpi_table_entry_header *header);
 extern int acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header);
 extern int acpi_parse_lapic_nmi (acpi_table_entry_header *header);
-#endif /*CONFIG_X86_LOCAL_APIC*/
-#ifdef CONFIG_X86_IO_APIC
 extern int acpi_parse_ioapic (acpi_table_entry_header *header);
-#endif /*CONFIG_X86_IO_APIC*/
 #endif /*CONFIG_ACPI*/
 
 u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
@@ -661,9 +657,7 @@ void __init find_intel_smp (void)
  */
 void __init find_smp_config (void)
 {
-#ifdef CONFIG_X86_LOCAL_APIC
 	find_intel_smp();
-#endif
 }
 
 
@@ -717,8 +711,6 @@ void __cpuinit mp_register_lapic (
 	MP_processor_info(&processor);
 }
 
-#ifdef CONFIG_X86_IO_APIC
-
 #define MP_ISA_BUS		0
 #define MP_MAX_IOAPIC_PIN	127
 
@@ -997,5 +989,4 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity)
 	return gsi;
 }
 
-#endif /*CONFIG_X86_IO_APIC*/
 #endif /*CONFIG_ACPI*/
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 34afad704824..fbe9f7faa2db 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -340,10 +340,8 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 		else if (fullarg(from, "acpi=strict")) {
 			acpi_strict = 1;
 		}
-#ifdef CONFIG_X86_IO_APIC
 		else if (fullarg(from, "acpi_skip_timer_override"))
 			acpi_skip_timer_override = 1;
-#endif
 #endif
 
 		if (fullarg(from, "disable_timer_pin_1"))
@@ -625,12 +623,10 @@ void __init setup_arch(char **cmdline_p)
         */
        acpi_reserve_bootmem();
 #endif
-#ifdef CONFIG_X86_LOCAL_APIC
 	/*
 	 * Find and reserve possible boot-time SMP configuration:
 	 */
 	find_smp_config();
-#endif
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (LOADER_TYPE && INITRD_START) {
 		if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
@@ -674,14 +670,12 @@ void __init setup_arch(char **cmdline_p)
 
 	init_cpu_to_node();
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	/*
 	 * get boot-time SMP configuration:
 	 */
 	if (smp_found_config)
 		get_smp_config();
 	init_apic_mappings();
-#endif
 
 	/*
 	 * Request address space for all standard RAM and ROM resources
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 582896f7d420..d29571e24953 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1175,13 +1175,8 @@ int __cpuinit __cpu_up(unsigned int cpu)
 void __init smp_cpus_done(unsigned int max_cpus)
 {
 	smp_cleanup_boot();
-
-#ifdef CONFIG_X86_IO_APIC
 	setup_ioapic_dest();
-#endif
-
 	check_nmi_watchdog();
-
 	time_init_gtod();
 }
 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index ea00915d393a..d66c7f750e75 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -41,9 +41,7 @@
 #include <asm/sections.h>
 #include <linux/cpufreq.h>
 #include <linux/hpet.h>
-#ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/apic.h>
-#endif
 
 #ifdef CONFIG_CPU_FREQ
 static void cpufreq_delayed_get(void);
@@ -438,12 +436,8 @@ void main_timer_handler(struct pt_regs *regs)
  * have to call the local interrupt handler.
  */
 
-#ifndef CONFIG_X86_LOCAL_APIC
-	profile_tick(CPU_PROFILING, regs);
-#else
 	if (!using_apic_timer)
 		smp_local_timer_interrupt(regs);
-#endif
 
 /*
  * If we have an externally synchronized Linux clock, then update CMOS clock
@@ -467,10 +461,8 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	if (apic_runs_main_timer > 1)
 		return IRQ_HANDLED;
 	main_timer_handler(regs);
-#ifdef CONFIG_X86_LOCAL_APIC
 	if (using_apic_timer)
 		smp_send_timer_broadcast_ipi();
-#endif
 	return IRQ_HANDLED;
 }
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 96f62a033242..56d7ff0c894c 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -788,7 +788,6 @@ asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 		if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
 								== NOTIFY_STOP)
 			return;
-#ifdef CONFIG_X86_LOCAL_APIC
 		/*
 		 * Ok, so this is none of the documented NMI sources,
 		 * so it must be the NMI watchdog.
@@ -796,7 +795,6 @@ asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 		if (nmi_watchdog_tick(regs,reason))
 			return;
 		if (!do_nmi_callback(regs,cpu))
-#endif
 			unknown_nmi_error(reason, regs);
 
 		return;
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index 9c96a0a8d1bd..8ed0f4d67b8d 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -29,8 +29,6 @@ extern int apic_runs_main_timer;
 			printk(s, ##a);    \
 	} while (0)
 
-#ifdef CONFIG_X86_LOCAL_APIC
-
 struct pt_regs;
 
 /*
@@ -104,8 +102,6 @@ void switch_ipi_to_APIC_timer(void *cpumask);
 
 #define ARCH_APICTIMER_STOPS_ON_C3	1
 
-#endif /* CONFIG_X86_LOCAL_APIC */
-
 extern unsigned boot_cpu_id;
 
 #endif /* __ASM_APIC_H */
diff --git a/include/asm-x86_64/fixmap.h b/include/asm-x86_64/fixmap.h
index 0b4ffbd1a125..1b620db5b9e3 100644
--- a/include/asm-x86_64/fixmap.h
+++ b/include/asm-x86_64/fixmap.h
@@ -37,13 +37,9 @@ enum fixed_addresses {
 	VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
 	VSYSCALL_HPET,
 	FIX_HPET_BASE,
-#ifdef CONFIG_X86_LOCAL_APIC
 	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
-#endif
-#ifdef CONFIG_X86_IO_APIC
 	FIX_IO_APIC_BASE_0,
 	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
-#endif
 	__end_of_fixed_addresses
 };
 
diff --git a/include/asm-x86_64/io_apic.h b/include/asm-x86_64/io_apic.h
index fb7a0909a174..5d1b5c68e36e 100644
--- a/include/asm-x86_64/io_apic.h
+++ b/include/asm-x86_64/io_apic.h
@@ -10,8 +10,6 @@
  * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar
  */
 
-#ifdef CONFIG_X86_IO_APIC
-
 #ifdef CONFIG_PCI_MSI
 static inline int use_pci_vector(void)	{return 1;}
 static inline void disable_edge_ioapic_vector(unsigned int vector) { }
@@ -209,10 +207,6 @@ extern int timer_uses_ioapic_pin_0;
 
 extern int sis_apic_bug; /* dummy */ 
 
-#else  /* !CONFIG_X86_IO_APIC */
-#define io_apic_assign_pci_irqs 0
-#endif
-
 extern int assign_irq_vector(int irq);
 
 void enable_NMI_through_LVT0 (void * dummy);
diff --git a/include/asm-x86_64/irq.h b/include/asm-x86_64/irq.h
index 9db5a1b4f7b1..43469d8ab71a 100644
--- a/include/asm-x86_64/irq.h
+++ b/include/asm-x86_64/irq.h
@@ -44,9 +44,7 @@ static __inline__ int irq_canonicalize(int irq)
 	return ((irq == 2) ? 9 : irq);
 }
 
-#ifdef CONFIG_X86_LOCAL_APIC
 #define ARCH_HAS_NMI_WATCHDOG		/* See include/linux/nmi.h */
-#endif
 
 #ifdef CONFIG_HOTPLUG_CPU
 #include <linux/cpumask.h>
diff --git a/include/asm-x86_64/mpspec.h b/include/asm-x86_64/mpspec.h
index 14fc3ddd9031..3c7aae9fd157 100644
--- a/include/asm-x86_64/mpspec.h
+++ b/include/asm-x86_64/mpspec.h
@@ -184,12 +184,10 @@ extern int pic_mode;
 extern void mp_register_lapic (u8 id, u8 enabled);
 extern void mp_register_lapic_address (u64 address);
 
-#ifdef CONFIG_X86_IO_APIC
 extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
 extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger, u32 gsi);
 extern void mp_config_acpi_legacy_irqs (void);
 extern int mp_register_gsi (u32 gsi, int triggering, int polarity);
-#endif /*CONFIG_X86_IO_APIC*/
 #endif
 
 extern int using_apic_timer;
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 612d208961a2..6353fa41bebe 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -9,15 +9,11 @@
 #include <linux/bitops.h>
 extern int disable_apic;
 
-#ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/fixmap.h>
 #include <asm/mpspec.h>
-#ifdef CONFIG_X86_IO_APIC
 #include <asm/io_apic.h>
-#endif
 #include <asm/apic.h>
 #include <asm/thread_info.h>
-#endif
 
 #ifdef CONFIG_SMP
 
-- 
cgit v1.2.3


From b1c78c0fcc29097567e1afc39701012e6d89adb7 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Clean up and minor fixes to TLB flush

- Convert CR* accesses to dedicated inline functions and rewrite
the rest as C inlines
- Don't do a double flush for global flushes (pointed out by Zach Amsden)
This was a bug workaround for old CPUs that don't do 64bit and is obsolete.
- Add a proper memory clobber to invlpg
- Remove an unused extern

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pgtable.h  |  2 --
 include/asm-x86_64/tlbflush.h | 70 +++++++++++++++++++++----------------------
 2 files changed, 35 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index a31ab4e68a9b..0c1e2422400a 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -25,8 +25,6 @@ extern int nonx_setup(char *str);
 extern void paging_init(void);
 extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 
-extern unsigned long pgkern_mask;
-
 /*
  * ZERO_PAGE is a global shared page that is always zero: used
  * for zero-mapped memory areas etc..
diff --git a/include/asm-x86_64/tlbflush.h b/include/asm-x86_64/tlbflush.h
index d16d5b60f419..983bd296c81a 100644
--- a/include/asm-x86_64/tlbflush.h
+++ b/include/asm-x86_64/tlbflush.h
@@ -4,44 +4,44 @@
 #include <linux/mm.h>
 #include <asm/processor.h>
 
-#define __flush_tlb()							\
-	do {								\
-		unsigned long tmpreg;					\
-									\
-		__asm__ __volatile__(					\
-			"movq %%cr3, %0;  # flush TLB \n"		\
-			"movq %0, %%cr3;              \n"		\
-			: "=r" (tmpreg)					\
-			:: "memory");					\
-	} while (0)
+static inline unsigned long get_cr3(void)
+{
+	unsigned long cr3;
+	asm volatile("mov %%cr3,%0" : "=r" (cr3));
+	return cr3;
+}
 
-/*
- * Global pages have to be flushed a bit differently. Not a real
- * performance problem because this does not happen often.
- */
-#define __flush_tlb_global()						\
-	do {								\
-		unsigned long tmpreg, cr4, cr4_orig;			\
-									\
-		__asm__ __volatile__(					\
-			"movq %%cr4, %2;  # turn off PGE     \n"	\
-			"movq %2, %1;                        \n"	\
-			"andq %3, %1;                        \n"	\
-			"movq %1, %%cr4;                     \n"	\
-			"movq %%cr3, %0;  # flush TLB        \n"	\
-			"movq %0, %%cr3;                     \n"	\
-			"movq %2, %%cr4;  # turn PGE back on \n"	\
-			: "=&r" (tmpreg), "=&r" (cr4), "=&r" (cr4_orig)	\
-			: "i" (~X86_CR4_PGE)				\
-			: "memory");					\
-	} while (0)
-
-extern unsigned long pgkern_mask;
-
-#define __flush_tlb_all() __flush_tlb_global()
+static inline void set_cr3(unsigned long cr3)
+{
+	asm volatile("mov %0,%%cr3" :: "r" (cr3) : "memory");
+}
+
+static inline void __flush_tlb(void)
+{
+	set_cr3(get_cr3());
+}
+
+static inline unsigned long get_cr4(void)
+{
+	unsigned long cr4;
+	asm volatile("mov %%cr4,%0" : "=r" (cr4));
+	return cr4;
+}
+
+static inline void set_cr4(unsigned long cr4)
+{
+	asm volatile("mov %0,%%cr4" :: "r" (cr4) : "memory");
+}
+
+static inline void __flush_tlb_all(void)
+{
+	unsigned long cr4 = get_cr4();
+	set_cr4(cr4 & ~X86_CR4_PGE);	/* clear PGE */
+	set_cr4(cr4);			/* write old PGE again and flush TLBs */
+}
 
 #define __flush_tlb_one(addr) \
-	__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
+	__asm__ __volatile__("invlpg (%0)" :: "r" (addr) : "memory")
 
 
 /*
-- 
cgit v1.2.3


From 107878bb14b1fb6bddc646f4d5e72e8beaa2f6a2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: Minor fixes & cleanup to tlb flush

(based on x86-64 changes)
- Add a proper memory clobber to invlpg
- Remove an unused extern

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/tlbflush.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/tlbflush.h b/include/asm-i386/tlbflush.h
index d57ca5c540b6..360648b0f2b3 100644
--- a/include/asm-i386/tlbflush.h
+++ b/include/asm-i386/tlbflush.h
@@ -36,8 +36,6 @@
 			: "memory");					\
 	} while (0)
 
-extern unsigned long pgkern_mask;
-
 # define __flush_tlb_all()						\
 	do {								\
 		if (cpu_has_pge)					\
@@ -49,7 +47,7 @@ extern unsigned long pgkern_mask;
 #define cpu_has_invlpg	(boot_cpu_data.x86 > 3)
 
 #define __flush_tlb_single(addr) \
-	__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
+	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory")
 
 #ifdef CONFIG_X86_INVLPG
 # define __flush_tlb_one(addr) __flush_tlb_single(addr)
-- 
cgit v1.2.3


From a8fcf1a24a16e1c735c795d99773b5dcefb71518 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Remove obsolete PIC mode

PIC mode is an outdated way to drive the APICs that was used on
some early MP boards. It is not supported in the ACPI model.

It is unlikely to be ever configured by any x86-64 system

Remove it thus.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c    | 92 ++++++++++++++------------------------------
 arch/x86_64/kernel/mpparse.c |  8 ----
 arch/x86_64/kernel/smpboot.c |  1 -
 include/asm-x86_64/mpspec.h  |  1 -
 include/asm-x86_64/smp.h     |  1 -
 5 files changed, 29 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 78c43b1a3064..20a76f209629 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -136,72 +136,40 @@ void clear_local_APIC(void)
 	apic_read(APIC_ESR);
 }
 
-void __init connect_bsp_APIC(void)
-{
-	if (pic_mode) {
-		/*
-		 * Do not trust the local APIC being empty at bootup.
-		 */
-		clear_local_APIC();
-		/*
-		 * PIC mode, enable APIC mode in the IMCR, i.e.
-		 * connect BSP's local APIC to INT and NMI lines.
-		 */
-		apic_printk(APIC_VERBOSE, "leaving PIC mode, enabling APIC mode.\n");
-		outb(0x70, 0x22);
-		outb(0x01, 0x23);
-	}
-}
-
 void disconnect_bsp_APIC(int virt_wire_setup)
 {
-	if (pic_mode) {
-		/*
-		 * Put the board back into PIC mode (has an effect
-		 * only on certain older boards).  Note that APIC
-		 * interrupts, including IPIs, won't work beyond
-		 * this point!  The only exception are INIT IPIs.
-		 */
-		apic_printk(APIC_QUIET, "disabling APIC mode, entering PIC mode.\n");
-		outb(0x70, 0x22);
-		outb(0x00, 0x23);
-	}
-	else {
-		/* Go back to Virtual Wire compatibility mode */
-		unsigned long value;
-
-		/* For the spurious interrupt use vector F, and enable it */
-		value = apic_read(APIC_SPIV);
-		value &= ~APIC_VECTOR_MASK;
-		value |= APIC_SPIV_APIC_ENABLED;
-		value |= 0xf;
-		apic_write(APIC_SPIV, value);
-
-		if (!virt_wire_setup) {
-			/* For LVT0 make it edge triggered, active high, external and enabled */
-			value = apic_read(APIC_LVT0);
-			value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
-				APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
-				APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED );
-			value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-			value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
-			apic_write(APIC_LVT0, value);
-		}
-		else {
-			/* Disable LVT0 */
-			apic_write(APIC_LVT0, APIC_LVT_MASKED);
-		}
+	/* Go back to Virtual Wire compatibility mode */
+	unsigned long value;
 
-		/* For LVT1 make it edge triggered, active high, nmi and enabled */
-		value = apic_read(APIC_LVT1);
-		value &= ~(
-			APIC_MODE_MASK | APIC_SEND_PENDING |
+	/* For the spurious interrupt use vector F, and enable it */
+	value = apic_read(APIC_SPIV);
+	value &= ~APIC_VECTOR_MASK;
+	value |= APIC_SPIV_APIC_ENABLED;
+	value |= 0xf;
+	apic_write(APIC_SPIV, value);
+
+	if (!virt_wire_setup) {
+		/* For LVT0 make it edge triggered, active high, external and enabled */
+		value = apic_read(APIC_LVT0);
+		value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
 			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
-			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED );
 		value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
-		apic_write(APIC_LVT1, value);
+		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
+		apic_write(APIC_LVT0, value);
+	} else {
+		/* Disable LVT0 */
+		apic_write(APIC_LVT0, APIC_LVT_MASKED);
 	}
+
+	/* For LVT1 make it edge triggered, active high, nmi and enabled */
+	value = apic_read(APIC_LVT1);
+	value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
+			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+	value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+	value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
+	apic_write(APIC_LVT1, value);
 }
 
 void disable_local_APIC(void)
@@ -418,7 +386,7 @@ void __cpuinit setup_local_APIC (void)
 	 * TODO: set up through-local-APIC from through-I/O-APIC? --macro
 	 */
 	value = apic_read(APIC_LVT0) & APIC_LVT_MASKED;
-	if (!smp_processor_id() && (pic_mode || !value)) {
+	if (!smp_processor_id() && !value) {
 		value = APIC_DM_EXTINT;
 		apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", smp_processor_id());
 	} else {
@@ -1096,8 +1064,6 @@ int __init APIC_init_uniprocessor (void)
 
 	verify_local_APIC();
 
-	connect_bsp_APIC();
-
 	phys_cpu_present_map = physid_mask_of_physid(boot_cpu_id);
 	apic_write(APIC_ID, SET_APIC_ID(boot_cpu_id));
 
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 90e99cf27aa5..022624503552 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -56,7 +56,6 @@ struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 int mp_irq_entries;
 
 int nr_ioapics;
-int pic_mode;
 unsigned long mp_lapic_addr = 0;
 
 
@@ -514,13 +513,6 @@ void __init get_smp_config (void)
  		printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
 
 	printk("Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
-	if (mpf->mpf_feature2 & (1<<7)) {
-		printk(KERN_INFO "    IMCR and PIC compatibility mode.\n");
-		pic_mode = 1;
-	} else {
-		printk(KERN_INFO "    Virtual Wire compatibility mode.\n");
-		pic_mode = 0;
-	}
 
 	/*
 	 * Now see if we need to read further.
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index d29571e24953..ae7d6d84e352 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1090,7 +1090,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	/*
 	 * Switch from PIC to APIC mode.
 	 */
-	connect_bsp_APIC();
 	setup_local_APIC();
 
 	if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_id) {
diff --git a/include/asm-x86_64/mpspec.h b/include/asm-x86_64/mpspec.h
index 3c7aae9fd157..1dc83436cc91 100644
--- a/include/asm-x86_64/mpspec.h
+++ b/include/asm-x86_64/mpspec.h
@@ -178,7 +178,6 @@ extern int mp_irq_entries;
 extern struct mpc_config_intsrc mp_irqs [MAX_IRQ_SOURCES];
 extern int mpc_default_type;
 extern unsigned long mp_lapic_addr;
-extern int pic_mode;
 
 #ifdef CONFIG_ACPI
 extern void mp_register_lapic (u8 id, u8 enabled);
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6353fa41bebe..498fbc1fc179 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -33,7 +33,6 @@ extern cpumask_t cpu_initialized;
  
 extern void smp_alloc_memory(void);
 extern volatile unsigned long smp_invalidate_needed;
-extern int pic_mode;
 extern void lock_ipi_call_lock(void);
 extern void unlock_ipi_call_lock(void);
 extern int smp_num_siblings;
-- 
cgit v1.2.3


From dfa4698c50bf85b7927214b0e4a3dc4bc3b3c4a9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Move early chipset quirks out to new file

They did not really belong into io_apic.c. Move them into a new file
and clean it up a bit.

Also remove outdated ATI quirk that was obsolete,

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/Makefile       |   2 +-
 arch/x86_64/kernel/early-quirks.c | 118 ++++++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/io_apic.c      | 101 --------------------------------
 arch/x86_64/kernel/setup.c        |   2 +-
 include/asm-x86_64/proto.h        |   2 +-
 5 files changed, 121 insertions(+), 104 deletions(-)
 create mode 100644 arch/x86_64/kernel/early-quirks.c

(limited to 'include')

diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 0ef39553c0f6..000e67e8f028 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -8,7 +8,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_x86_64.o \
 		x8664_ksyms.o i387.o syscall.o vsyscall.o \
 		setup64.o bootflag.o e820.o reboot.o quirks.o i8237.o \
-		pci-dma.o pci-nommu.o alternative.o
+		pci-dma.o pci-nommu.o alternative.o early-quirks.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-$(CONFIG_X86_MCE)         += mce.o
diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c
new file mode 100644
index 000000000000..d637cff1c4b1
--- /dev/null
+++ b/arch/x86_64/kernel/early-quirks.c
@@ -0,0 +1,118 @@
+/* Various workarounds for chipset bugs.
+   This code runs very early and can't use the regular PCI subsystem
+   The entries are keyed to PCI bridges which usually identify chipsets
+   uniquely.
+   This is only for whole classes of chipsets with specific problems which
+   need early invasive action (e.g. before the timers are initialized).
+   Most PCI device specific workarounds can be done later and should be
+   in standard PCI quirks
+   Mainboard specific bugs should be handled by DMI entries.
+   CPU specific bugs in setup.c */
+
+#include <linux/pci.h>
+#include <linux/acpi.h>
+#include <linux/pci_ids.h>
+#include <asm/pci-direct.h>
+#include <asm/proto.h>
+#include <asm/dma.h>
+
+static void via_bugs(void)
+{
+#ifdef CONFIG_IOMMU
+	if ((end_pfn > MAX_DMA32_PFN ||  force_iommu) &&
+	    !iommu_aperture_allowed) {
+		printk(KERN_INFO
+  "Looks like a VIA chipset. Disabling IOMMU. Override with iommu=allowed\n");
+		iommu_aperture_disabled = 1;
+	}
+#endif
+}
+
+#ifdef CONFIG_ACPI
+
+static int nvidia_hpet_detected __initdata;
+
+static int __init nvidia_hpet_check(unsigned long phys, unsigned long size)
+{
+	nvidia_hpet_detected = 1;
+	return 0;
+}
+#endif
+
+static void nvidia_bugs(void)
+{
+#ifdef CONFIG_ACPI
+	/*
+	 * All timer overrides on Nvidia are
+	 * wrong unless HPET is enabled.
+	 */
+	nvidia_hpet_detected = 0;
+	acpi_table_parse(ACPI_HPET, nvidia_hpet_check);
+	if (nvidia_hpet_detected == 0) {
+		acpi_skip_timer_override = 1;
+		printk(KERN_INFO "Nvidia board "
+		       "detected. Ignoring ACPI "
+		       "timer override.\n");
+	}
+#endif
+	/* RED-PEN skip them on mptables too? */
+
+}
+
+static void ati_bugs(void)
+{
+#if 1 /* for testing */
+	printk("ATI board detected\n");
+#endif
+	/* No bugs right now */
+}
+
+struct chipset {
+	u16 vendor;
+	void (*f)(void);
+};
+
+static struct chipset early_qrk[] = {
+	{ PCI_VENDOR_ID_NVIDIA, nvidia_bugs },
+	{ PCI_VENDOR_ID_VIA, via_bugs },
+	{ PCI_VENDOR_ID_ATI, ati_bugs },
+	{}
+};
+
+void __init early_quirks(void)
+{
+	int num, slot, func;
+	/* Poor man's PCI discovery */
+	for (num = 0; num < 32; num++) {
+		for (slot = 0; slot < 32; slot++) {
+			for (func = 0; func < 8; func++) {
+				u32 class;
+				u32 vendor;
+				u8 type;
+				int i;
+				class = read_pci_config(num,slot,func,
+							PCI_CLASS_REVISION);
+				if (class == 0xffffffff)
+					break;
+
+		       		if ((class >> 16) != PCI_CLASS_BRIDGE_PCI)
+					continue;
+
+				vendor = read_pci_config(num, slot, func,
+							 PCI_VENDOR_ID);
+				vendor &= 0xffff;
+
+				for (i = 0; early_qrk[i].f; i++)
+					if (early_qrk[i].vendor == vendor) {
+						early_qrk[i].f();
+						return;
+					}
+
+				type = read_pci_config_byte(num, slot, func,
+							    PCI_HEADER_TYPE);
+				if (!(type & 0x80))
+					break;
+			}
+		}
+	}
+}
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index a02ed17d05dc..8bbff64c20d7 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -280,107 +280,6 @@ static int __init setup_enable_8254_timer(char *s)
 __setup("disable_8254_timer", setup_disable_8254_timer);
 __setup("enable_8254_timer", setup_enable_8254_timer);
 
-#include <asm/pci-direct.h>
-#include <linux/pci_ids.h>
-#include <linux/pci.h>
-
-
-#ifdef CONFIG_ACPI
-
-static int nvidia_hpet_detected __initdata;
-
-static int __init nvidia_hpet_check(unsigned long phys, unsigned long size)
-{
-	nvidia_hpet_detected = 1;
-	return 0;
-}
-#endif
-
-/* Temporary Hack. Nvidia and VIA boards currently only work with IO-APIC
-   off. Check for an Nvidia or VIA PCI bridge and turn it off.
-   Use pci direct infrastructure because this runs before the PCI subsystem. 
-
-   Can be overwritten with "apic"
-
-   And another hack to disable the IOMMU on VIA chipsets.
-
-   ... and others. Really should move this somewhere else.
-
-   Kludge-O-Rama. */
-void __init check_ioapic(void) 
-{ 
-	int num,slot,func; 
-	/* Poor man's PCI discovery */
-	for (num = 0; num < 32; num++) { 
-		for (slot = 0; slot < 32; slot++) { 
-			for (func = 0; func < 8; func++) { 
-				u32 class;
-				u32 vendor;
-				u8 type;
-				class = read_pci_config(num,slot,func,
-							PCI_CLASS_REVISION);
-				if (class == 0xffffffff)
-					break; 
-
-		       		if ((class >> 16) != PCI_CLASS_BRIDGE_PCI)
-					continue; 
-
-				vendor = read_pci_config(num, slot, func, 
-							 PCI_VENDOR_ID);
-				vendor &= 0xffff;
-				switch (vendor) { 
-				case PCI_VENDOR_ID_VIA:
-#ifdef CONFIG_IOMMU
-					if ((end_pfn > MAX_DMA32_PFN ||
-					     force_iommu) &&
-					    !iommu_aperture_allowed) {
-						printk(KERN_INFO
-    "Looks like a VIA chipset. Disabling IOMMU. Override with \"iommu=allowed\"\n");
-						iommu_aperture_disabled = 1;
-					}
-#endif
-					return;
-				case PCI_VENDOR_ID_NVIDIA:
-#ifdef CONFIG_ACPI
-					/*
-					 * All timer overrides on Nvidia are
-					 * wrong unless HPET is enabled.
-					 */
-					nvidia_hpet_detected = 0;
-					acpi_table_parse(ACPI_HPET,
-							nvidia_hpet_check);
-					if (nvidia_hpet_detected == 0) {
-						acpi_skip_timer_override = 1;
-						printk(KERN_INFO "Nvidia board "
-						    "detected. Ignoring ACPI "
-						    "timer override.\n");
-					}
-#endif
-					/* RED-PEN skip them on mptables too? */
-					return;
-
-				/* This should be actually default, but
-				   for 2.6.16 let's do it for ATI only where
-				   it's really needed. */
-				case PCI_VENDOR_ID_ATI:
-					if (timer_over_8254 == 1) {	
-						timer_over_8254 = 0;	
-					printk(KERN_INFO
-		"ATI board detected. Disabling timer routing over 8254.\n");
-					}	
-					return;
-				} 
-
-
-				/* No multi-function device? */
-				type = read_pci_config_byte(num,slot,func,
-							    PCI_HEADER_TYPE);
-				if (!(type & 0x80))
-					break;
-			} 
-		}
-	}
-} 
 
 /*
  * Find the IRQ entry number of a certain pin.
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index fbe9f7faa2db..aab4bd66aa0d 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -653,7 +653,7 @@ void __init setup_arch(char **cmdline_p)
 
 	paging_init();
 
-	check_ioapic();
+	early_quirks();
 
 	/*
 	 * set this early, so we dont allocate cpu0
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 3b1c60247902..58fec91318e4 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -92,7 +92,7 @@ extern void syscall32_cpu_init(void);
 
 extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end);
 
-extern void check_ioapic(void);
+extern void early_quirks(void);
 extern void check_efer(void);
 
 extern int unhandled_signal(struct task_struct *tsk, int sig);
-- 
cgit v1.2.3


From 55f05ffaa788e039df2f1ebe0d7bfbcb6f39d0b4 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Replace mp bus array with bitmap for bus not pci

Since we only support PCI and ISA legacy busses now there is no need to
have an full array with checking.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 92 +++++++++++---------------------------------
 arch/x86_64/kernel/mpparse.c |  9 ++---
 include/asm-x86_64/mpspec.h  |  8 +---
 3 files changed, 27 insertions(+), 82 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 8bbff64c20d7..a1412042b918 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -308,7 +308,7 @@ static int __init find_isa_irq_pin(int irq, int type)
 	for (i = 0; i < mp_irq_entries; i++) {
 		int lbus = mp_irqs[i].mpc_srcbus;
 
-		if (mp_bus_id_to_type[lbus] == MP_BUS_ISA &&
+		if (test_bit(lbus, mp_bus_not_pci) &&
 		    (mp_irqs[i].mpc_irqtype == type) &&
 		    (mp_irqs[i].mpc_srcbusirq == irq))
 
@@ -324,7 +324,7 @@ static int __init find_isa_irq_apic(int irq, int type)
 	for (i = 0; i < mp_irq_entries; i++) {
 		int lbus = mp_irqs[i].mpc_srcbus;
 
-		if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA) &&
+		if (test_bit(lbus, mp_bus_not_pci) &&
 		    (mp_irqs[i].mpc_irqtype == type) &&
 		    (mp_irqs[i].mpc_srcbusirq == irq))
 			break;
@@ -364,7 +364,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
 			    mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
 				break;
 
-		if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) &&
+		if (!test_bit(lbus, mp_bus_not_pci) &&
 		    !mp_irqs[i].mpc_irqtype &&
 		    (bus == lbus) &&
 		    (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
@@ -410,28 +410,11 @@ static int __init MPBIOS_polarity(int idx)
 	switch (mp_irqs[idx].mpc_irqflag & 3)
 	{
 		case 0: /* conforms, ie. bus-type dependent polarity */
-		{
-			switch (mp_bus_id_to_type[bus])
-			{
-				case MP_BUS_ISA: /* ISA pin */
-				{
-					polarity = default_ISA_polarity(idx);
-					break;
-				}
-				case MP_BUS_PCI: /* PCI pin */
-				{
-					polarity = default_PCI_polarity(idx);
-					break;
-				}
-				default:
-				{
-					printk(KERN_WARNING "broken BIOS!!\n");
-					polarity = 1;
-					break;
-				}
-			}
+			if (test_bit(bus, mp_bus_not_pci))
+				polarity = default_ISA_polarity(idx);
+			else
+				polarity = default_PCI_polarity(idx);
 			break;
-		}
 		case 1: /* high active */
 		{
 			polarity = 0;
@@ -469,28 +452,11 @@ static int MPBIOS_trigger(int idx)
 	switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
 	{
 		case 0: /* conforms, ie. bus-type dependent */
-		{
-			switch (mp_bus_id_to_type[bus])
-			{
-				case MP_BUS_ISA: /* ISA pin */
-				{
-					trigger = default_ISA_trigger(idx);
-					break;
-				}
-				case MP_BUS_PCI: /* PCI pin */
-				{
-					trigger = default_PCI_trigger(idx);
-					break;
-				}
-				default:
-				{
-					printk(KERN_WARNING "broken BIOS!!\n");
-					trigger = 1;
-					break;
-				}
-			}
+			if (test_bit(bus, mp_bus_not_pci))
+				trigger = default_ISA_trigger(idx);
+			else
+				trigger = default_PCI_trigger(idx);
 			break;
-		}
 		case 1: /* edge */
 		{
 			trigger = 0;
@@ -596,31 +562,17 @@ static int pin_2_irq(int idx, int apic, int pin)
 	if (mp_irqs[idx].mpc_dstirq != pin)
 		printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
 
-	switch (mp_bus_id_to_type[bus])
-	{
-		case MP_BUS_ISA: /* ISA pin */
-		{
-			irq = mp_irqs[idx].mpc_srcbusirq;
-			break;
-		}
-		case MP_BUS_PCI: /* PCI pin */
-		{
-			/*
-			 * PCI IRQs are mapped in order
-			 */
-			i = irq = 0;
-			while (i < apic)
-				irq += nr_ioapic_registers[i++];
-			irq += pin;
-			irq = gsi_irq_sharing(irq);
-			break;
-		}
-		default:
-		{
-			printk(KERN_ERR "unknown bus type %d.\n",bus); 
-			irq = 0;
-			break;
-		}
+	if (test_bit(bus, mp_bus_not_pci)) {
+		irq = mp_irqs[idx].mpc_srcbusirq;
+	} else {
+		/*
+		 * PCI IRQs are mapped in order
+		 */
+		i = irq = 0;
+		while (i < apic)
+			irq += nr_ioapic_registers[i++];
+		irq += pin;
+		irq = gsi_irq_sharing(irq);
 	}
 	BUG_ON(irq >= NR_IRQS);
 	return irq;
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index c12acc3b5552..a8d38c0bb449 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -42,7 +42,7 @@ int acpi_found_madt;
  * MP-table.
  */
 unsigned char apic_version [MAX_APICS];
-unsigned char mp_bus_id_to_type [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
+DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 
 static int mp_current_pci_id = 0;
@@ -173,9 +173,9 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	Dprintk("Bus #%d is %s\n", m->mpc_busid, str);
 
 	if (strncmp(str, "ISA", 3) == 0) {
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
+		set_bit(m->mpc_busid, mp_bus_not_pci);
 	} else if (strncmp(str, "PCI", 3) == 0) {
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
+		clear_bit(m->mpc_busid, mp_bus_not_pci);
 		mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
 		mp_current_pci_id++;
 	} else {
@@ -808,8 +808,7 @@ void __init mp_config_acpi_legacy_irqs (void)
 	/* 
 	 * Fabricate the legacy ISA bus (bus #31).
 	 */
-	mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
-	Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
+	set_bit(MP_ISA_BUS, mp_bus_not_pci);
 
 	/* 
 	 * Locate the IOAPIC that manages the ISA IRQs (0-15). 
diff --git a/include/asm-x86_64/mpspec.h b/include/asm-x86_64/mpspec.h
index 1dc83436cc91..017fddb61dc5 100644
--- a/include/asm-x86_64/mpspec.h
+++ b/include/asm-x86_64/mpspec.h
@@ -159,13 +159,7 @@ struct mpc_config_lintsrc
 #define MAX_MP_BUSSES 256
 /* Each PCI slot may be a combo card with its own bus.  4 IRQ pins per slot. */
 #define MAX_IRQ_SOURCES (MAX_MP_BUSSES * 4)
-enum mp_bustype {
-	MP_BUS_ISA = 1,
-	MP_BUS_EISA,
-	MP_BUS_PCI,
-	MP_BUS_MCA
-};
-extern unsigned char mp_bus_id_to_type [MAX_MP_BUSSES];
+extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 extern int mp_bus_id_to_pci_bus [MAX_MP_BUSSES];
 
 extern unsigned int boot_cpu_physical_apicid;
-- 
cgit v1.2.3


From e2414910f212c52d9d7c64c99a22863488ac5b48 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] x86: Detect CFI support in the assembler at runtime

... instead of using a CONFIG option. The config option still controls
if the resulting executable actually has unwind information.

This is useful to prevent compilation errors when users select
CONFIG_STACK_UNWIND on old binutils and also allows to use
CFI in the future for non kernel debugging applications.

Cc: jbeulich@novell.com
Cc: sam@ravnborg.org

Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/kbuild/makefiles.txt | 5 +++++
 arch/i386/Makefile                 | 4 ++++
 arch/x86_64/Makefile               | 3 +++
 include/asm-x86_64/dwarf2.h        | 2 +-
 scripts/Kbuild.include             | 7 +++++++
 5 files changed, 20 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/Documentation/kbuild/makefiles.txt b/Documentation/kbuild/makefiles.txt
index b7d6abb501a6..e2cbd59cf2d0 100644
--- a/Documentation/kbuild/makefiles.txt
+++ b/Documentation/kbuild/makefiles.txt
@@ -421,6 +421,11 @@ more details, with real examples.
 	The second argument is optional, and if supplied will be used
 	if first argument is not supported.
 
+    as-instr
+	as-instr checks if the assembler reports a specific instruction
+	and then outputs either option1 or option2
+	C escapes are supported in the test instruction
+
     cc-option
 	cc-option is used to check if $(CC) supports a given option, and not
 	supported to use an optional second option.
diff --git a/arch/i386/Makefile b/arch/i386/Makefile
index 3e4adb1e2244..508cdbeb3a09 100644
--- a/arch/i386/Makefile
+++ b/arch/i386/Makefile
@@ -46,6 +46,10 @@ cflags-y += -ffreestanding
 # a lot more stack due to the lack of sharing of stacklots:
 CFLAGS				+= $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-option,-fno-unit-at-a-time); fi ;)
 
+# do binutils support CFI?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+
 CFLAGS += $(cflags-y)
 
 # Default subarch .c files
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 431bb4bc36cd..d6472ddf5f6e 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -54,6 +54,9 @@ endif
 cflags-y += $(call cc-option,-funit-at-a-time)
 # prevent gcc from generating any FP code by mistake
 cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+# do binutils support CFI?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
 CFLAGS += $(cflags-y)
 CFLAGS_KERNEL += $(cflags-kernel-y)
diff --git a/include/asm-x86_64/dwarf2.h b/include/asm-x86_64/dwarf2.h
index 0744db777676..2b9368365fad 100644
--- a/include/asm-x86_64/dwarf2.h
+++ b/include/asm-x86_64/dwarf2.h
@@ -13,7 +13,7 @@
    away for older version. 
  */
 
-#ifdef CONFIG_UNWIND_INFO
+#ifdef CONFIG_AS_CFI
 
 #define CFI_STARTPROC .cfi_startproc
 #define CFI_ENDPROC .cfi_endproc
diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include
index 3d523899fdc0..7adef12a0c26 100644
--- a/scripts/Kbuild.include
+++ b/scripts/Kbuild.include
@@ -63,6 +63,13 @@ as-option = $(shell if $(CC) $(CFLAGS) $(1) -Wa,-Z -c -o /dev/null \
 	     -xassembler /dev/null > /dev/null 2>&1; then echo "$(1)"; \
 	     else echo "$(2)"; fi ;)
 
+# as-instr
+# Usage: cflags-y += $(call as-instr, instr, option1, option2)
+
+as-instr = $(shell if echo -e "$(1)" | $(AS) -Z -o astest$$$$.out \
+		   2>&1 >/dev/null ; then echo "$(2)"; else echo "$(3)"; fi; \
+	           rm -f astest$$$$.out)
+
 # cc-option
 # Usage: cflags-y += $(call cc-option, -march=winchip-c6, -march=i586)
 
-- 
cgit v1.2.3


From 3b94355c47e2b025a7ececa0a14180e8fce6b0f1 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] remove int_delivery_dest

The genapic field and the accessor macro weren't used anywhere.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/genapic_cluster.c | 1 -
 arch/x86_64/kernel/genapic_flat.c    | 2 --
 include/asm-x86_64/genapic.h         | 1 -
 include/asm-x86_64/mach_apic.h       | 1 -
 4 files changed, 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/genapic_cluster.c b/arch/x86_64/kernel/genapic_cluster.c
index 3020917546de..cdb90e671b88 100644
--- a/arch/x86_64/kernel/genapic_cluster.c
+++ b/arch/x86_64/kernel/genapic_cluster.c
@@ -118,7 +118,6 @@ struct genapic apic_cluster = {
 	.name = "clustered",
 	.int_delivery_mode = dest_Fixed,
 	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
-	.int_delivery_dest = APIC_DEST_PHYSICAL | APIC_DM_FIXED,
 	.target_cpus = cluster_target_cpus,
 	.apic_id_registered = cluster_apic_id_registered,
 	.init_apic_ldr = cluster_init_apic_ldr,
diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c
index eb86d374813a..2ceb8b872b57 100644
--- a/arch/x86_64/kernel/genapic_flat.c
+++ b/arch/x86_64/kernel/genapic_flat.c
@@ -121,7 +121,6 @@ struct genapic apic_flat =  {
 	.name = "flat",
 	.int_delivery_mode = dest_LowestPrio,
 	.int_dest_mode = (APIC_DEST_LOGICAL != 0),
-	.int_delivery_dest = APIC_DEST_LOGICAL | APIC_DM_LOWEST,
 	.target_cpus = flat_target_cpus,
 	.apic_id_registered = flat_apic_id_registered,
 	.init_apic_ldr = flat_init_apic_ldr,
@@ -180,7 +179,6 @@ struct genapic apic_physflat =  {
 	.name = "physical flat",
 	.int_delivery_mode = dest_Fixed,
 	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
-	.int_delivery_dest = APIC_DEST_PHYSICAL | APIC_DM_FIXED,
 	.target_cpus = physflat_target_cpus,
 	.apic_id_registered = flat_apic_id_registered,
 	.init_apic_ldr = flat_init_apic_ldr,/*not needed, but shouldn't hurt*/
diff --git a/include/asm-x86_64/genapic.h b/include/asm-x86_64/genapic.h
index 50b38e7c58e4..81e714665344 100644
--- a/include/asm-x86_64/genapic.h
+++ b/include/asm-x86_64/genapic.h
@@ -16,7 +16,6 @@ struct genapic {
 	char *name;
 	u32 int_delivery_mode;
 	u32 int_dest_mode;
-	u32 int_delivery_dest;	/* for quick IPIs */
 	int (*apic_id_registered)(void);
 	cpumask_t (*target_cpus)(void);
 	void (*init_apic_ldr)(void);
diff --git a/include/asm-x86_64/mach_apic.h b/include/asm-x86_64/mach_apic.h
index 0acea44c9377..d33422450c00 100644
--- a/include/asm-x86_64/mach_apic.h
+++ b/include/asm-x86_64/mach_apic.h
@@ -16,7 +16,6 @@
 
 #define INT_DELIVERY_MODE (genapic->int_delivery_mode)
 #define INT_DEST_MODE (genapic->int_dest_mode)
-#define INT_DELIVERY_DEST (genapic->int_delivery_dest)
 #define TARGET_CPUS	  (genapic->target_cpus())
 #define apic_id_registered (genapic->apic_id_registered)
 #define init_apic_ldr (genapic->init_apic_ldr)
-- 
cgit v1.2.3


From 5f4a7a93886ce1a4327f6028cc05d423f39eebf0 Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: rearrange 'struct iommu_table' members

Rearrange struct members loosely based on size for improved alignment
and to save a few bytes.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/calgary.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/calgary.h b/include/asm-x86_64/calgary.h
index 4e3919524240..0a03bda94d03 100644
--- a/include/asm-x86_64/calgary.h
+++ b/include/asm-x86_64/calgary.h
@@ -34,12 +34,12 @@ struct iommu_table {
 	unsigned long  it_base;      /* mapped address of tce table */
 	unsigned long  it_hint;      /* Hint for next alloc */
 	unsigned long *it_map;       /* A simple allocation bitmap for now */
+	void __iomem  *bbar;         /* Bridge BAR */
+	u64	       tar_val;      /* Table Address Register */
+	struct timer_list watchdog_timer;
 	spinlock_t     it_lock;      /* Protects it_map */
 	unsigned int   it_size;      /* Size of iommu table in entries */
 	unsigned char  it_busno;     /* Bus number this table belongs to */
-	void __iomem  *bbar;
-	u64	       tar_val;
-	struct timer_list watchdog_timer;
 };
 
 #define TCE_TABLE_SIZE_UNSPECIFIED	~0
-- 
cgit v1.2.3


From f38db651d5da5e10235fd7dd31095969fb7ef6fb Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: consolidate per bus data structures

Move the tce_table_kva array, disabled bitmap and bus_to_phb array
into a new per bus 'struct calgary_bus_info'. Also slightly reorganize
build_tce_table and tce_table_setparms to avoid exporting bus_info to
tce.c.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 53 +++++++++++++++++++++-------------------
 arch/x86_64/kernel/tce.c         | 10 --------
 include/asm-x86_64/tce.h         |  1 -
 3 files changed, 28 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 146924ba5df5..96f6a866afad 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -111,17 +111,17 @@ static const unsigned long phb_offsets[] = {
 	0xB000 /* PHB3 */
 };
 
-static char bus_to_phb[MAX_PHB_BUS_NUM];
-void* tce_table_kva[MAX_PHB_BUS_NUM];
 unsigned int specified_table_size = TCE_TABLE_SIZE_UNSPECIFIED;
 static int translate_empty_slots __read_mostly = 0;
 static int calgary_detected __read_mostly = 0;
 
-/*
- * the bitmap of PHBs the user requested that we disable
- * translation on.
- */
-static DECLARE_BITMAP(translation_disabled, MAX_PHB_BUS_NUM);
+struct calgary_bus_info {
+	void *tce_space;
+	int translation_disabled;
+	signed char phbid;
+};
+
+static struct calgary_bus_info bus_info[MAX_PHB_BUS_NUM] = { { NULL, 0, 0 }, };
 
 static void tce_cache_blast(struct iommu_table *tbl);
 
@@ -149,7 +149,7 @@ static inline unsigned int num_dma_pages(unsigned long dma, unsigned int dmalen)
 
 static inline int translate_phb(struct pci_dev* dev)
 {
-	int disabled = test_bit(dev->bus->number, translation_disabled);
+	int disabled = bus_info[dev->bus->number].translation_disabled;
 	return !disabled;
 }
 
@@ -454,7 +454,7 @@ static struct dma_mapping_ops calgary_dma_ops = {
 
 static inline int busno_to_phbid(unsigned char num)
 {
-	return bus_to_phb[num];
+	return bus_info[num].phbid;
 }
 
 static inline unsigned long split_queue_offset(unsigned char num)
@@ -631,6 +631,10 @@ static int __init calgary_setup_tar(struct pci_dev *dev, void __iomem *bbar)
 	if (ret)
 		return ret;
 
+	tbl = dev->sysdata;
+	tbl->it_base = (unsigned long)bus_info[dev->bus->number].tce_space;
+	tce_free(tbl, 0, tbl->it_size);
+
 	calgary_reserve_regions(dev);
 
 	/* set TARs for each PHB */
@@ -824,7 +828,7 @@ static int __init calgary_init(void)
 			calgary_init_one_nontraslated(dev);
 			continue;
 		}
-		if (!tce_table_kva[dev->bus->number] && !translate_empty_slots) {
+		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots) {
 			pci_dev_put(dev);
 			continue;
 		}
@@ -844,7 +848,7 @@ error:
 			pci_dev_put(dev);
 			continue;
 		}
-		if (!tce_table_kva[dev->bus->number] && !translate_empty_slots)
+		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots)
 			continue;
 		calgary_disable_translation(dev);
 		calgary_free_tar(dev);
@@ -894,9 +898,8 @@ void __init detect_calgary(void)
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		int dev;
-
-		tce_table_kva[bus] = NULL;
-		bus_to_phb[bus] = -1;
+		struct calgary_bus_info *info = &bus_info[bus];
+		info->phbid = -1;
 
 		if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY)
 			continue;
@@ -907,12 +910,9 @@ void __init detect_calgary(void)
 		 */
 		phb = (phb + 1) % PHBS_PER_CALGARY;
 
-		if (test_bit(bus, translation_disabled)) {
-			printk(KERN_INFO "Calgary: translation is disabled for "
-			       "PHB 0x%x\n", bus);
-			/* skip this phb, don't allocate a tbl for it */
+		if (info->translation_disabled)
 			continue;
-		}
+
 		/*
 		 * Scan the slots of the PCI bus to see if there is a device present.
 		 * The parent bus will be the zero-ith device, so start at 1.
@@ -923,8 +923,8 @@ void __init detect_calgary(void)
 				tbl = alloc_tce_table();
 				if (!tbl)
 					goto cleanup;
-				tce_table_kva[bus] = tbl;
-				bus_to_phb[bus] = phb;
+				info->tce_space = tbl;
+				info->phbid = phb;
 				calgary_found = 1;
 				break;
 			}
@@ -940,9 +940,12 @@ void __init detect_calgary(void)
 	return;
 
 cleanup:
-	for (--bus; bus >= 0; --bus)
-		if (tce_table_kva[bus])
-			free_tce_table(tce_table_kva[bus]);
+	for (--bus; bus >= 0; --bus) {
+		struct calgary_bus_info *info = &bus_info[bus];
+
+		if (info->tce_space)
+			free_tce_table(info->tce_space);
+	}
 }
 
 int __init calgary_iommu_init(void)
@@ -1016,7 +1019,7 @@ static int __init calgary_parse_options(char *p)
 			if (bridge < MAX_PHB_BUS_NUM) {
 				printk(KERN_INFO "Calgary: disabling "
 				       "translation for PHB 0x%x\n", bridge);
-				set_bit(bridge, translation_disabled);
+				bus_info[bridge].translation_disabled = 1;
 			}
 		}
 
diff --git a/arch/x86_64/kernel/tce.c b/arch/x86_64/kernel/tce.c
index 0905ce6b9b32..cbabfdf78e06 100644
--- a/arch/x86_64/kernel/tce.c
+++ b/arch/x86_64/kernel/tce.c
@@ -106,14 +106,6 @@ static int tce_table_setparms(struct pci_dev *dev, struct iommu_table *tbl)
 	/* set the tce table size - measured in entries */
 	tbl->it_size = table_size_to_number_of_entries(specified_table_size);
 
-	tbl->it_base = (unsigned long)tce_table_kva[dev->bus->number];
-	if (!tbl->it_base) {
-		printk(KERN_ERR "Calgary: iommu_table_setparms: "
-		       "no table allocated?!\n");
-		ret = -ENOMEM;
-		goto done;
-	}
-
 	/*
 	 * number of bytes needed for the bitmap size in number of
 	 * entries; we need one bit per entry
@@ -162,8 +154,6 @@ int build_tce_table(struct pci_dev *dev, void __iomem *bbar)
 	if (ret)
 		goto free_tbl;
 
-	tce_free(tbl, 0, tbl->it_size);
-
 	tbl->bbar = bbar;
 
 	/*
diff --git a/include/asm-x86_64/tce.h b/include/asm-x86_64/tce.h
index 53e9a68b3336..dbb047febc5e 100644
--- a/include/asm-x86_64/tce.h
+++ b/include/asm-x86_64/tce.h
@@ -24,7 +24,6 @@
 #ifndef _ASM_X86_64_TCE_H
 #define _ASM_X86_64_TCE_H
 
-extern void* tce_table_kva[];
 extern unsigned int specified_table_size;
 struct iommu_table;
 
-- 
cgit v1.2.3


From 01215ad8d83e18321d99e9b5750a6f21cac243a2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] i386: Remove lock section support in mutex.h

Lock sections don't work the new dwarf2 unwinder
This generates slightly smaller code. It adds one more taken
jump to the fast path.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/mutex.h | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/mutex.h b/include/asm-i386/mutex.h
index 05a538531229..7a17d9e58ad6 100644
--- a/include/asm-i386/mutex.h
+++ b/include/asm-i386/mutex.h
@@ -30,14 +30,10 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   decl (%%eax)	\n"			\
-			"   js 2f		\n"			\
+			"   jns 1f		\n"			\
+			"   call "#fail_fn"	\n"			\
 			"1:			\n"			\
 									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
-									\
 		:"=a" (dummy)						\
 		: "a" (count)						\
 		: "memory", "ecx", "edx");				\
@@ -86,14 +82,10 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   incl (%%eax)	\n"			\
-			"   jle 2f		\n"			\
+			"   jg	1f		\n"			\
+			"   call "#fail_fn"	\n"			\
 			"1:			\n"			\
 									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
-									\
 		:"=a" (dummy)						\
 		: "a" (count)						\
 		: "memory", "ecx", "edx");				\
-- 
cgit v1.2.3


From add659bf8aa92f8b3f01a8c0220557c959507fb1 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] i386: Remove lock section support in rwsem.h

Lock sections don't work the new dwarf2 unwinder
This generates slightly smaller code. It adds one more taken
jump to the fast path.

Also move the trampolines into semaphore.S and add proper CFI
annotations.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/lib/semaphore.S | 63 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-i386/rwsem.h  | 62 +++++++++-------------------------------------
 2 files changed, 75 insertions(+), 50 deletions(-)

(limited to 'include')

diff --git a/arch/i386/lib/semaphore.S b/arch/i386/lib/semaphore.S
index e16ff1696390..01f80b5c45d2 100644
--- a/arch/i386/lib/semaphore.S
+++ b/arch/i386/lib/semaphore.S
@@ -152,3 +152,66 @@ ENTRY(__read_lock_failed)
 	END(__read_lock_failed)
 
 #endif
+
+/* Fix up special calling conventions */
+ENTRY(call_rwsem_down_read_failed)
+	CFI_STARTPROC
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	push %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	call rwsem_down_read_failed
+	pop %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	ret
+	CFI_ENDPROC
+	END(call_rwsem_down_read_failed)
+
+ENTRY(call_rwsem_down_write_failed)
+	CFI_STARTPROC
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	calll rwsem_down_write_failed
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	ret
+	CFI_ENDPROC
+	END(call_rwsem_down_write_failed)
+
+ENTRY(call_rwsem_wake)
+	CFI_STARTPROC
+	decw %dx    /* do nothing if still outstanding active readers */
+	jnz 1f
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call rwsem_wake
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+1:	ret
+	CFI_ENDPROC
+	END(call_rwsem_wake)
+
+/* Fix up special calling conventions */
+ENTRY(call_rwsem_downgrade_wake)
+	CFI_STARTPROC
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	push %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	call rwsem_downgrade_wake
+	pop %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	ret
+	CFI_ENDPROC
+	END(call_rwsem_downgrade_wake)
+
diff --git a/include/asm-i386/rwsem.h b/include/asm-i386/rwsem.h
index 43113f5608eb..bc598d6388e3 100644
--- a/include/asm-i386/rwsem.h
+++ b/include/asm-i386/rwsem.h
@@ -99,17 +99,9 @@ static inline void __down_read(struct rw_semaphore *sem)
 	__asm__ __volatile__(
 		"# beginning down_read\n\t"
 LOCK_PREFIX	"  incl      (%%eax)\n\t" /* adds 0x00000001, returns the old value */
-		"  js        2f\n\t" /* jump if we weren't granted the lock */
+		"  jns        1f\n"
+		"  call call_rwsem_down_read_failed\n"
 		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  pushl     %%ecx\n\t"
-		"  pushl     %%edx\n\t"
-		"  call      rwsem_down_read_failed\n\t"
-		"  popl      %%edx\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
 		"# ending down_read\n\t"
 		: "+m" (sem->count)
 		: "a" (sem)
@@ -151,15 +143,9 @@ static inline void __down_write_nested(struct rw_semaphore *sem, int subclass)
 		"# beginning down_write\n\t"
 LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtract 0x0000ffff, returns the old value */
 		"  testl     %%edx,%%edx\n\t" /* was the count 0 before? */
-		"  jnz       2f\n\t" /* jump if we weren't granted the lock */
-		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  pushl     %%ecx\n\t"
-		"  call      rwsem_down_write_failed\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
+		"  jz        1f\n"
+		"  call call_rwsem_down_write_failed\n"
+		"1:\n"
 		"# ending down_write"
 		: "+m" (sem->count), "=d" (tmp)
 		: "a" (sem), "1" (tmp)
@@ -193,17 +179,9 @@ static inline void __up_read(struct rw_semaphore *sem)
 	__asm__ __volatile__(
 		"# beginning __up_read\n\t"
 LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtracts 1, returns the old value */
-		"  js        2f\n\t" /* jump if the lock is being waited upon */
-		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  decw      %%dx\n\t" /* do nothing if still outstanding active readers */
-		"  jnz       1b\n\t"
-		"  pushl     %%ecx\n\t"
-		"  call      rwsem_wake\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
+		"  jns        1f\n\t"
+		"  call call_rwsem_wake\n"
+		"1:\n"
 		"# ending __up_read\n"
 		: "+m" (sem->count), "=d" (tmp)
 		: "a" (sem), "1" (tmp)
@@ -219,17 +197,9 @@ static inline void __up_write(struct rw_semaphore *sem)
 		"# beginning __up_write\n\t"
 		"  movl      %2,%%edx\n\t"
 LOCK_PREFIX	"  xaddl     %%edx,(%%eax)\n\t" /* tries to transition 0xffff0001 -> 0x00000000 */
-		"  jnz       2f\n\t" /* jump if the lock is being waited upon */
+		"  jz       1f\n"
+		"  call call_rwsem_wake\n"
 		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  decw      %%dx\n\t" /* did the active count reduce to 0? */
-		"  jnz       1b\n\t" /* jump back if not */
-		"  pushl     %%ecx\n\t"
-		"  call      rwsem_wake\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
 		"# ending __up_write\n"
 		: "+m" (sem->count)
 		: "a" (sem), "i" (-RWSEM_ACTIVE_WRITE_BIAS)
@@ -244,17 +214,9 @@ static inline void __downgrade_write(struct rw_semaphore *sem)
 	__asm__ __volatile__(
 		"# beginning __downgrade_write\n\t"
 LOCK_PREFIX	"  addl      %2,(%%eax)\n\t" /* transitions 0xZZZZ0001 -> 0xYYYY0001 */
-		"  js        2f\n\t" /* jump if the lock is being waited upon */
+		"  jns       1f\n\t"
+		"  call call_rwsem_downgrade_wake\n"
 		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  pushl     %%ecx\n\t"
-		"  pushl     %%edx\n\t"
-		"  call      rwsem_downgrade_wake\n\t"
-		"  popl      %%edx\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
 		"# ending __downgrade_write\n"
 		: "+m" (sem->count)
 		: "a" (sem), "i" (-RWSEM_WAITING_BIAS)
-- 
cgit v1.2.3


From 7ca2b49b06a6d26e89e3535653889f1d7892b085 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] i386: Remove lock section support in semaphore.h

Lock sections don't work the new dwarf2 unwinder
This generates slightly smaller code. It adds one more taken
jump to the fast path.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/semaphore.h | 49 +++++++++++++++-----------------------------
 1 file changed, 17 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/semaphore.h b/include/asm-i386/semaphore.h
index d51e800acf29..e63b6a68f04c 100644
--- a/include/asm-i386/semaphore.h
+++ b/include/asm-i386/semaphore.h
@@ -100,13 +100,10 @@ static inline void down(struct semaphore * sem)
 	__asm__ __volatile__(
 		"# atomic down operation\n\t"
 		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
-		"js 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %0,%%eax\n\t"
-		"call __down_failed\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n"
+		"\tlea %0,%%eax\n\t"
+		"call __down_failed\n"
+		"2:"
 		:"+m" (sem->count)
 		:
 		:"memory","ax");
@@ -123,15 +120,12 @@ static inline int down_interruptible(struct semaphore * sem)
 	might_sleep();
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %1,%%eax\n\t"
-		"call __down_failed_interruptible\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n\t"
+		"lea %1,%%eax\n\t"
+		"call __down_failed_interruptible\n"
+		"2:"
 		:"=a" (result), "+m" (sem->count)
 		:
 		:"memory");
@@ -148,15 +142,12 @@ static inline int down_trylock(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %1,%%eax\n\t"
+		"jns 2f\n\t"
+		"lea %1,%%eax\n\t"
 		"call __down_failed_trylock\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"2:\n"
 		:"=a" (result), "+m" (sem->count)
 		:
 		:"memory");
@@ -166,22 +157,16 @@ static inline int down_trylock(struct semaphore * sem)
 /*
  * Note! This is subtle. We jump to wake people up only if
  * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
  */
 static inline void up(struct semaphore * sem)
 {
 	__asm__ __volatile__(
 		"# atomic up operation\n\t"
 		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
-		"jle 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %0,%%eax\n\t"
-		"call __up_wakeup\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
-		".subsection 0\n"
+		"jg 1f\n\t"
+		"lea %0,%%eax\n\t"
+		"call __up_wakeup\n"
+		"1:"
 		:"+m" (sem->count)
 		:
 		:"memory","ax");
-- 
cgit v1.2.3


From 8578bdf835d6d6ece6718858d351456367b8a5cf Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Don't use lock section for mutexes and semaphores

Lock sections cannot be handled by the dwarf2 unwinder.

Disadvantage is a taken branch in the hot path.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/mutex.h     | 20 ++++++--------------
 include/asm-x86_64/semaphore.h | 40 ++++++++++++++--------------------------
 include/asm-x86_64/spinlock.h  |  9 ++++-----
 3 files changed, 24 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/mutex.h b/include/asm-x86_64/mutex.h
index 06fab6de2a88..16396b1de3e4 100644
--- a/include/asm-x86_64/mutex.h
+++ b/include/asm-x86_64/mutex.h
@@ -25,13 +25,9 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   decl (%%rdi)	\n"			\
-			"   js 2f		\n"			\
-			"1:			\n"			\
-									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
+			"   jns 1f		\n"			\
+			"   call "#fail_fn"	\n"			\
+			"1:"						\
 									\
 		:"=D" (dummy)						\
 		: "D" (v)						\
@@ -75,13 +71,9 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   incl (%%rdi)	\n"			\
-			"   jle 2f		\n"			\
-			"1:			\n"			\
-									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
+			"   jg 1f		\n"			\
+			"   call "#fail_fn"	\n"			\
+			"1:			  "			\
 									\
 		:"=D" (dummy)						\
 		: "D" (v)						\
diff --git a/include/asm-x86_64/semaphore.h b/include/asm-x86_64/semaphore.h
index 064df08b9a0f..107bd90429e8 100644
--- a/include/asm-x86_64/semaphore.h
+++ b/include/asm-x86_64/semaphore.h
@@ -107,12 +107,9 @@ static inline void down(struct semaphore * sem)
 	__asm__ __volatile__(
 		"# atomic down operation\n\t"
 		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
-		"js 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __down_failed\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 1f\n\t"
+		"call __down_failed\n"
+		"1:"
 		:"=m" (sem->count)
 		:"D" (sem)
 		:"memory");
@@ -130,14 +127,11 @@ static inline int down_interruptible(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __down_failed_interruptible\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n\t"
+		"call __down_failed_interruptible\n"
+		"2:\n"
 		:"=a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory");
@@ -154,14 +148,11 @@ static inline int down_trylock(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __down_failed_trylock\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n\t"
+		"call __down_failed_trylock\n\t"
+		"2:\n"
 		:"=a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory","cc");
@@ -179,12 +170,9 @@ static inline void up(struct semaphore * sem)
 	__asm__ __volatile__(
 		"# atomic up operation\n\t"
 		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
-		"jle 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __up_wakeup\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jg 1f\n\t"
+		"call __up_wakeup\n"
+		"1:"
 		:"=m" (sem->count)
 		:"D" (sem)
 		:"memory");
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index a8e3d89f591f..1d0733b38a4b 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -22,14 +22,13 @@
 #define __raw_spin_lock_string \
 	"\n1:\t" \
 	LOCK_PREFIX " ; decl %0\n\t" \
-	"js 2f\n" \
-	LOCK_SECTION_START("") \
-	"2:\t" \
+	"jns 2f\n" \
+	"3:\n" \
 	"rep;nop\n\t" \
 	"cmpl $0,%0\n\t" \
-	"jle 2b\n\t" \
+	"jle 3b\n\t" \
 	"jmp 1b\n" \
-	LOCK_SECTION_END
+	"2:\t" \
 
 #define __raw_spin_lock_string_up \
 	"\n\tdecl %0"
-- 
cgit v1.2.3


From 8b059d2373c16b6d32787a49daf8ccf72dc61b71 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Clean up spin/rwlocks

- Inline spinlock strings into their inline functions
- Convert macros to typesafe inlines
- Replace some leftover __asm__ __volatile__s with asm volatile

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/rwlock.h   | 14 +--------
 include/asm-x86_64/spinlock.h | 71 ++++++++++++++++++++++++-------------------
 2 files changed, 41 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/rwlock.h b/include/asm-x86_64/rwlock.h
index 28a080d23119..72aeebed920b 100644
--- a/include/asm-x86_64/rwlock.h
+++ b/include/asm-x86_64/rwlock.h
@@ -21,18 +21,6 @@
 #define RW_LOCK_BIAS		 0x01000000
 #define RW_LOCK_BIAS_STR	 "0x01000000"
 
-#define __build_read_lock(rw)   \
-	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t" \
-		     "jns 1f\n" \
-		     "call __read_lock_failed\n" \
-		     "1:\n" \
-		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
-
-#define __build_write_lock(rw) \
-	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t" \
-		     "jz 1f\n" \
-		     "\tcall __write_lock_failed\n\t" \
-		     "1:\n" \
-		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
+/* Actual code is in asm/spinlock.h or in arch/x86_64/lib/rwlock.S */
 
 #endif
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index 1d0733b38a4b..be7a9e629fb2 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -16,30 +16,23 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-#define __raw_spin_is_locked(x) \
-		(*(volatile signed int *)(&(x)->slock) <= 0)
-
-#define __raw_spin_lock_string \
-	"\n1:\t" \
-	LOCK_PREFIX " ; decl %0\n\t" \
-	"jns 2f\n" \
-	"3:\n" \
-	"rep;nop\n\t" \
-	"cmpl $0,%0\n\t" \
-	"jle 3b\n\t" \
-	"jmp 1b\n" \
-	"2:\t" \
-
-#define __raw_spin_lock_string_up \
-	"\n\tdecl %0"
-
-#define __raw_spin_unlock_string \
-	"movl $1,%0" \
-		:"=m" (lock->slock) : : "memory"
+static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
+{
+	return *(volatile signed int *)(&(lock)->slock) <= 0;
+}
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	asm volatile(__raw_spin_lock_string : "=m" (lock->slock) : : "memory");
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; decl %0\n\t"
+		"jns 2f\n"
+		"3:\n"
+		"rep;nop\n\t"
+		"cmpl $0,%0\n\t"
+		"jle 3b\n\t"
+		"jmp 1b\n"
+		"2:\t" : "=m" (lock->slock) : : "memory");
 }
 
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
@@ -48,7 +41,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
 	int oldval;
 
-	__asm__ __volatile__(
+	asm volatile(
 		"xchgl %0,%1"
 		:"=q" (oldval), "=m" (lock->slock)
 		:"0" (0) : "memory");
@@ -58,13 +51,14 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		__raw_spin_unlock_string
-	);
+	asm volatile("movl $1,%0" :"=m" (lock->slock) :: "memory");
 }
 
-#define __raw_spin_unlock_wait(lock) \
-	do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
+static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
+{
+	while (__raw_spin_is_locked(lock))
+		cpu_relax();
+}
 
 /*
  * Read-write spinlocks, allowing multiple readers
@@ -80,17 +74,32 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
  * with the high bit (sign) being the "contended" bit.
  */
 
-#define __raw_read_can_lock(x)		((int)(x)->lock > 0)
-#define __raw_write_can_lock(x)		((x)->lock == RW_LOCK_BIAS)
+static inline int __raw_read_can_lock(raw_rwlock_t *lock)
+{
+	return (int)(lock)->lock > 0;
+}
+
+static inline int __raw_write_can_lock(raw_rwlock_t *lock)
+{
+	return (lock)->lock == RW_LOCK_BIAS;
+}
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	__build_read_lock(rw);
+	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t"
+		     "jns 1f\n"
+		     "call __read_lock_failed\n"
+		     "1:\n"
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	__build_write_lock(rw);
+	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t"
+		     "jz 1f\n"
+		     "\tcall __write_lock_failed\n\t"
+		     "1:\n"
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.3


From fb2e28485679418e459583605f9b19807a72ceca Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] i386: Clean up spin/rwlocks

- Inline spinlock strings into their inline functions
- Convert macros to typesafe inlines
- Replace some leftover __asm__ __volatile__s with asm volatile

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/rwlock.h   |  14 +----
 include/asm-i386/spinlock.h | 131 ++++++++++++++++++++++----------------------
 2 files changed, 68 insertions(+), 77 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h
index f40ccbd8cb7f..c3e5db32fa48 100644
--- a/include/asm-i386/rwlock.h
+++ b/include/asm-i386/rwlock.h
@@ -20,18 +20,6 @@
 #define RW_LOCK_BIAS		 0x01000000
 #define RW_LOCK_BIAS_STR	"0x01000000"
 
-#define __build_read_lock(rw, helper)   \
-	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" \
-			"jns 1f\n" \
-			"call " helper "\n\t" \
-			"1:\n" \
-			::"a" (rw) : "memory")
-
-#define __build_write_lock(rw, helper) \
-	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
-			"jz 1f\n" \
-			"call " helper "\n\t" \
-			"1:\n" \
-			::"a" (rw) : "memory")
+/* Code is in asm-i386/spinlock.h */
 
 #endif
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index d1020363c41a..324329313af8 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -4,6 +4,7 @@
 #include <asm/atomic.h>
 #include <asm/rwlock.h>
 #include <asm/page.h>
+#include <asm/processor.h>
 #include <linux/compiler.h>
 
 /*
@@ -17,67 +18,64 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-#define __raw_spin_is_locked(x) \
-		(*(volatile signed char *)(&(x)->slock) <= 0)
-
-#define __raw_spin_lock_string \
-	"\n1:\t" \
-	LOCK_PREFIX " ; decb %0\n\t" \
-	"jns 3f\n" \
-	"2:\t" \
-	"rep;nop\n\t" \
-	"cmpb $0,%0\n\t" \
-	"jle 2b\n\t" \
-	"jmp 1b\n" \
-	"3:\n\t"
-
-/*
- * NOTE: there's an irqs-on section here, which normally would have to be
- * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use
- * __raw_spin_lock_string_flags().
- */
-#define __raw_spin_lock_string_flags \
-	"\n1:\t" \
-	LOCK_PREFIX " ; decb %0\n\t" \
-	"jns 5f\n" \
-	"2:\t" \
-	"testl $0x200, %1\n\t" \
-	"jz 4f\n\t" \
-	"sti\n" \
-	"3:\t" \
-	"rep;nop\n\t" \
-	"cmpb $0, %0\n\t" \
-	"jle 3b\n\t" \
-	"cli\n\t" \
-	"jmp 1b\n" \
-	"4:\t" \
-	"rep;nop\n\t" \
-	"cmpb $0, %0\n\t" \
-	"jg 1b\n\t" \
-	"jmp 4b\n" \
-	"5:\n\t"
+static inline int __raw_spin_is_locked(raw_spinlock_t *x)
+{
+	return *(volatile signed char *)(&(x)->slock) <= 0;
+}
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	asm(__raw_spin_lock_string : "+m" (lock->slock) : : "memory");
+	asm volatile("\n1:\t"
+		     LOCK_PREFIX " ; decb %0\n\t"
+		     "jns 3f\n"
+		     "2:\t"
+		     "rep;nop\n\t"
+		     "cmpb $0,%0\n\t"
+		     "jle 2b\n\t"
+		     "jmp 1b\n"
+		     "3:\n\t"
+		     : "+m" (lock->slock) : : "memory");
 }
 
 /*
  * It is easier for the lock validator if interrupts are not re-enabled
  * in the middle of a lock-acquire. This is a performance feature anyway
  * so we turn it off:
+ *
+ * NOTE: there's an irqs-on section here, which normally would have to be
+ * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
  */
 #ifndef CONFIG_PROVE_LOCKING
 static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
 {
-	asm(__raw_spin_lock_string_flags : "+m" (lock->slock) : "r" (flags) : "memory");
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; decb %0\n\t"
+		"jns 5f\n"
+		"2:\t"
+		"testl $0x200, %1\n\t"
+		"jz 4f\n\t"
+		"sti\n"
+		"3:\t"
+		"rep;nop\n\t"
+		"cmpb $0, %0\n\t"
+		"jle 3b\n\t"
+		"cli\n\t"
+		"jmp 1b\n"
+		"4:\t"
+		"rep;nop\n\t"
+		"cmpb $0, %0\n\t"
+		"jg 1b\n\t"
+		"jmp 4b\n"
+		"5:\n\t"
+		: "+m" (lock->slock) : "r" (flags) : "memory");
 }
 #endif
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
 	char oldval;
-	__asm__ __volatile__(
+	asm volatile(
 		"xchgb %b0,%1"
 		:"=q" (oldval), "+m" (lock->slock)
 		:"0" (0) : "memory");
@@ -93,38 +91,29 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 #if !defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE)
 
-#define __raw_spin_unlock_string \
-	"movb $1,%0" \
-		:"+m" (lock->slock) : : "memory"
-
-
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		__raw_spin_unlock_string
-	);
+	asm volatile("movb $1,%0" : "+m" (lock->slock) :: "memory");
 }
 
 #else
 
-#define __raw_spin_unlock_string \
-	"xchgb %b0, %1" \
-		:"=q" (oldval), "+m" (lock->slock) \
-		:"0" (oldval) : "memory"
-
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
 	char oldval = 1;
 
-	__asm__ __volatile__(
-		__raw_spin_unlock_string
-	);
+	asm volatile("xchgb %b0, %1"
+		     : "=q" (oldval), "+m" (lock->slock)
+		     : "0" (oldval) : "memory");
 }
 
 #endif
 
-#define __raw_spin_unlock_wait(lock) \
-	do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
+static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
+{
+	while (__raw_spin_is_locked(lock))
+		cpu_relax();
+}
 
 /*
  * Read-write spinlocks, allowing multiple readers
@@ -151,22 +140,36 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
  * read_can_lock - would read_trylock() succeed?
  * @lock: the rwlock in question.
  */
-#define __raw_read_can_lock(x)		((int)(x)->lock > 0)
+static inline int __raw_read_can_lock(raw_rwlock_t *x)
+{
+	return (int)(x)->lock > 0;
+}
 
 /**
  * write_can_lock - would write_trylock() succeed?
  * @lock: the rwlock in question.
  */
-#define __raw_write_can_lock(x)		((x)->lock == RW_LOCK_BIAS)
+static inline int __raw_write_can_lock(raw_rwlock_t *x)
+{
+	return (x)->lock == RW_LOCK_BIAS;
+}
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	__build_read_lock(rw, "__read_lock_failed");
+	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t"
+		     "jns 1f\n"
+		     "call __read_lock_failed\n\t"
+		     "1:\n"
+		     ::"a" (rw) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	__build_write_lock(rw, "__write_lock_failed");
+	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t"
+		     "jz 1f\n"
+		     "call __write_lock_failed\n\t"
+		     "1:\n"
+		     ::"a" (rw) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.3


From 1a3f239ddf9208f2e52d36fef1c1c4518cbbbabe Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] i386: Replace i386 open-coded cmdline parsing with

This patch replaces the open-coded early commandline parsing
throughout the i386 boot code with the generic mechanism (already used
by ppc, powerpc, ia64 and s390).  The code was inconsistent with
whether it deletes the option from the cmdline or not, meaning some of
these will get passed through the environment into init.

This transformation is mainly mechanical, but there are some notable
parts:

1) Grammar: s/linux never set's it up/linux never sets it up/

2) Remove hacked-in earlyprintk= option scanning.  When someone
   actually implements CONFIG_EARLY_PRINTK, then they can use
   early_param().
[AK: actually it is implemented, but I'm adding the early_param it in the next
x86-64 patch]

3) Move declaration of generic_apic_probe() from setup.c into asm/apic.h

4) Various parameters now moved into their appropriate files (thanks Andi).

5) All parse functions which examine arg need to check for NULL,
   except one where it has subtle humor value.

AK: readded acpi_sci handling which was completely dropped
AK: moved some more variables into acpi/boot.c

Cc: len.brown@intel.com

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c     |  76 ++++++++-
 arch/i386/kernel/apic.c          |  15 ++
 arch/i386/kernel/io_apic.c       |  24 ++-
 arch/i386/kernel/machine_kexec.c |  23 +++
 arch/i386/kernel/setup.c         | 350 +++++++++++++--------------------------
 arch/i386/kernel/smpboot.c       |  13 ++
 arch/i386/mach-generic/probe.c   |  58 ++++---
 arch/i386/mm/init.c              |  18 +-
 include/asm-i386/acpi.h          |  14 --
 include/asm-i386/apic.h          |   4 +-
 include/asm-i386/io_apic.h       |  11 ++
 include/asm-i386/pgtable.h       |   2 -
 12 files changed, 319 insertions(+), 289 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index ee003bc0e8b1..87e2ab50b694 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -36,6 +36,8 @@
 #include <asm/io.h>
 #include <asm/mpspec.h>
 
+int __initdata acpi_force = 0;
+
 #ifdef	CONFIG_X86_64
 
 extern void __init clustered_apic_check(void);
@@ -860,8 +862,6 @@ static void __init acpi_process_madt(void)
 	return;
 }
 
-extern int acpi_force;
-
 #ifdef __i386__
 
 static int __init disable_acpi_irq(struct dmi_system_id *d)
@@ -1163,3 +1163,75 @@ int __init acpi_boot_init(void)
 
 	return 0;
 }
+
+static int __init parse_acpi(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
+
+	/* "acpi=off" disables both ACPI table parsing and interpreter */
+	if (strcmp(arg, "off") == 0) {
+		disable_acpi();
+	}
+	/* acpi=force to over-ride black-list */
+	else if (strcmp(arg, "force") == 0) {
+		acpi_force = 1;
+		acpi_ht = 1;
+		acpi_disabled = 0;
+	}
+	/* acpi=strict disables out-of-spec workarounds */
+	else if (strcmp(arg, "strict") == 0) {
+		acpi_strict = 1;
+	}
+	/* Limit ACPI just to boot-time to enable HT */
+	else if (strcmp(arg, "ht") == 0) {
+		if (!acpi_force)
+			disable_acpi();
+		acpi_ht = 1;
+	}
+	/* "acpi=noirq" disables ACPI interrupt routing */
+	else if (strcmp(arg, "noirq") == 0) {
+		acpi_noirq_set();
+	} else {
+		/* Core will printk when we return error. */
+		return -EINVAL;
+	}
+	return 0;
+}
+early_param("acpi", parse_acpi);
+
+/* FIXME: Using pci= for an ACPI parameter is a travesty. */
+static int __init parse_pci(char *arg)
+{
+	if (arg && strcmp(arg, "noacpi") == 0)
+		acpi_disable_pci();
+	return 0;
+}
+early_param("pci", parse_pci);
+
+#ifdef CONFIG_X86_IO_APIC
+static int __init parse_acpi_skip_timer_override(char *arg)
+{
+	acpi_skip_timer_override = 1;
+	return 0;
+}
+early_param("acpi_skip_timer_override", parse_acpi_skip_timer_override);
+#endif /* CONFIG_X86_IO_APIC */
+
+static int __init setup_acpi_sci(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "edge"))
+		acpi_sci_flags.trigger = 1;
+	else if (!strcmp(s, "level"))
+		acpi_sci_flags.trigger = 3;
+	else if (!strcmp(s, "high"))
+		acpi_sci_flags.polarity = 1;
+	else if (!strcmp(s, "low"))
+		acpi_sci_flags.polarity = 3;
+	else
+		return -EINVAL;
+	return 0;
+}
+early_param("acpi_sci", setup_acpi_sci);
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 1a34fc57800b..ce75e71b694f 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -1372,3 +1372,18 @@ int __init APIC_init_uniprocessor (void)
 
 	return 0;
 }
+
+static int __init parse_lapic(char *arg)
+{
+	lapic_enable();
+	return 0;
+}
+early_param("lapic", parse_lapic);
+
+static int __init parse_nolapic(char *arg)
+{
+	lapic_disable();
+	return 0;
+}
+early_param("nolapic", parse_nolapic);
+
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index b713f27d7e86..fd0df75cfbda 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -66,7 +66,7 @@ int sis_apic_bug = -1;
  */
 int nr_ioapic_registers[MAX_IO_APICS];
 
-int disable_timer_pin_1 __initdata;
+static int disable_timer_pin_1 __initdata;
 
 /*
  * Rough estimation of how many shared IRQs there are, can
@@ -2691,3 +2691,25 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a
 }
 
 #endif /* CONFIG_ACPI */
+
+static int __init parse_disable_timer_pin_1(char *arg)
+{
+	disable_timer_pin_1 = 1;
+	return 0;
+}
+early_param("disable_timer_pin_1", parse_disable_timer_pin_1);
+
+static int __init parse_enable_timer_pin_1(char *arg)
+{
+	disable_timer_pin_1 = -1;
+	return 0;
+}
+early_param("enable_timer_pin_1", parse_enable_timer_pin_1);
+
+static int __init parse_noapic(char *arg)
+{
+	/* disable IO-APIC */
+	disable_ioapic_setup();
+	return 0;
+}
+early_param("noapic", parse_noapic);
diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c
index 6b1ae6ba76f0..66c3dc99a655 100644
--- a/arch/i386/kernel/machine_kexec.c
+++ b/arch/i386/kernel/machine_kexec.c
@@ -9,6 +9,7 @@
 #include <linux/mm.h>
 #include <linux/kexec.h>
 #include <linux/delay.h>
+#include <linux/init.h>
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
@@ -209,3 +210,25 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	rnk = (relocate_new_kernel_t) reboot_code_buffer;
 	(*rnk)(page_list, reboot_code_buffer, image->start, cpu_has_pae);
 }
+
+/* crashkernel=size@addr specifies the location to reserve for
+ * a crash kernel.  By reserving this memory we guarantee
+ * that linux never sets it up as a DMA target.
+ * Useful for holding code to do something appropriate
+ * after a kernel panic.
+ */
+static int __init parse_crashkernel(char *arg)
+{
+	unsigned long size, base;
+	size = memparse(arg, &arg);
+	if (*arg == '@') {
+		base = memparse(arg+1, &arg);
+		/* FIXME: Do I want a sanity check
+		 * to validate the memory range?
+		 */
+		crashk_res.start = base;
+		crashk_res.end   = base + size - 1;
+	}
+	return 0;
+}
+early_param("crashkernel", parse_crashkernel);
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 71a540362b78..c6e31ed386f5 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -96,11 +96,6 @@ unsigned long mmu_cr4_features;
 #endif
 EXPORT_SYMBOL(acpi_disabled);
 
-#ifdef	CONFIG_ACPI
-int __initdata acpi_force = 0;
-extern acpi_interrupt_flags	acpi_sci_flags;
-#endif
-
 /* for MCA, but anyone else can use it if they want */
 unsigned int machine_id;
 #ifdef CONFIG_MCA
@@ -148,7 +143,6 @@ EXPORT_SYMBOL(ist_info);
 struct e820map e820;
 
 extern void early_cpu_init(void);
-extern void generic_apic_probe(char *);
 extern int root_mountflags;
 
 unsigned long saved_videomode;
@@ -700,238 +694,132 @@ static inline void copy_edd(void)
 }
 #endif
 
-static void __init parse_cmdline_early (char ** cmdline_p)
-{
-	char c = ' ', *to = command_line, *from = saved_command_line;
-	int len = 0;
-	int userdef = 0;
+static int __initdata user_defined_memmap = 0;
 
-	/* Save unparsed command line copy for /proc/cmdline */
-	saved_command_line[COMMAND_LINE_SIZE-1] = '\0';
+/*
+ * "mem=nopentium" disables the 4MB page tables.
+ * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
+ * to <mem>, overriding the bios size.
+ * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
+ * <start> to <start>+<mem>, overriding the bios size.
+ *
+ * HPA tells me bootloaders need to parse mem=, so no new
+ * option should be mem=  [also see Documentation/i386/boot.txt]
+ */
+static int __init parse_mem(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-	for (;;) {
-		if (c != ' ')
-			goto next_char;
-		/*
-		 * "mem=nopentium" disables the 4MB page tables.
-		 * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
-		 * to <mem>, overriding the bios size.
-		 * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
-		 * <start> to <start>+<mem>, overriding the bios size.
-		 *
-		 * HPA tells me bootloaders need to parse mem=, so no new
-		 * option should be mem=  [also see Documentation/i386/boot.txt]
+	if (strcmp(arg, "nopentium") == 0) {
+		clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+		disable_pse = 1;
+	} else {
+		/* If the user specifies memory size, we
+		 * limit the BIOS-provided memory map to
+		 * that size. exactmap can be used to specify
+		 * the exact map. mem=number can be used to
+		 * trim the existing memory map.
 		 */
-		if (!memcmp(from, "mem=", 4)) {
-			if (to != command_line)
-				to--;
-			if (!memcmp(from+4, "nopentium", 9)) {
-				from += 9+4;
-				clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
-				disable_pse = 1;
-			} else {
-				/* If the user specifies memory size, we
-				 * limit the BIOS-provided memory map to
-				 * that size. exactmap can be used to specify
-				 * the exact map. mem=number can be used to
-				 * trim the existing memory map.
-				 */
-				unsigned long long mem_size;
+		unsigned long long mem_size;
  
-				mem_size = memparse(from+4, &from);
-				limit_regions(mem_size);
-				userdef=1;
-			}
-		}
-
-		else if (!memcmp(from, "memmap=", 7)) {
-			if (to != command_line)
-				to--;
-			if (!memcmp(from+7, "exactmap", 8)) {
-#ifdef CONFIG_CRASH_DUMP
-				/* If we are doing a crash dump, we
-				 * still need to know the real mem
-				 * size before original memory map is
-				 * reset.
-				 */
-				find_max_pfn();
-				saved_max_pfn = max_pfn;
-#endif
-				from += 8+7;
-				e820.nr_map = 0;
-				userdef = 1;
-			} else {
-				/* If the user specifies memory size, we
-				 * limit the BIOS-provided memory map to
-				 * that size. exactmap can be used to specify
-				 * the exact map. mem=number can be used to
-				 * trim the existing memory map.
-				 */
-				unsigned long long start_at, mem_size;
- 
-				mem_size = memparse(from+7, &from);
-				if (*from == '@') {
-					start_at = memparse(from+1, &from);
-					add_memory_region(start_at, mem_size, E820_RAM);
-				} else if (*from == '#') {
-					start_at = memparse(from+1, &from);
-					add_memory_region(start_at, mem_size, E820_ACPI);
-				} else if (*from == '$') {
-					start_at = memparse(from+1, &from);
-					add_memory_region(start_at, mem_size, E820_RESERVED);
-				} else {
-					limit_regions(mem_size);
-					userdef=1;
-				}
-			}
-		}
-
-		else if (!memcmp(from, "noexec=", 7))
-			noexec_setup(from + 7);
+		mem_size = memparse(arg, &arg);
+		limit_regions(mem_size);
+		user_defined_memmap = 1;
+	}
+	return 0;
+}
+early_param("mem", parse_mem);
 
+static int __init parse_memmap(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-#ifdef  CONFIG_X86_SMP
-		/*
-		 * If the BIOS enumerates physical processors before logical,
-		 * maxcpus=N at enumeration-time can be used to disable HT.
+	if (strcmp(arg, "exactmap") == 0) {
+#ifdef CONFIG_CRASH_DUMP
+		/* If we are doing a crash dump, we
+		 * still need to know the real mem
+		 * size before original memory map is
+		 * reset.
 		 */
-		else if (!memcmp(from, "maxcpus=", 8)) {
-			extern unsigned int maxcpus;
-
-			maxcpus = simple_strtoul(from + 8, NULL, 0);
-		}
+		find_max_pfn();
+		saved_max_pfn = max_pfn;
 #endif
-
-#ifdef CONFIG_ACPI
-		/* "acpi=off" disables both ACPI table parsing and interpreter */
-		else if (!memcmp(from, "acpi=off", 8)) {
-			disable_acpi();
-		}
-
-		/* acpi=force to over-ride black-list */
-		else if (!memcmp(from, "acpi=force", 10)) {
-			acpi_force = 1;
-			acpi_ht = 1;
-			acpi_disabled = 0;
-		}
-
-		/* acpi=strict disables out-of-spec workarounds */
-		else if (!memcmp(from, "acpi=strict", 11)) {
-			acpi_strict = 1;
-		}
-
-		/* Limit ACPI just to boot-time to enable HT */
-		else if (!memcmp(from, "acpi=ht", 7)) {
-			if (!acpi_force)
-				disable_acpi();
-			acpi_ht = 1;
-		}
-		
-		/* "pci=noacpi" disable ACPI IRQ routing and PCI scan */
-		else if (!memcmp(from, "pci=noacpi", 10)) {
-			acpi_disable_pci();
-		}
-		/* "acpi=noirq" disables ACPI interrupt routing */
-		else if (!memcmp(from, "acpi=noirq", 10)) {
-			acpi_noirq_set();
+		e820.nr_map = 0;
+		user_defined_memmap = 1;
+	} else {
+		/* If the user specifies memory size, we
+		 * limit the BIOS-provided memory map to
+		 * that size. exactmap can be used to specify
+		 * the exact map. mem=number can be used to
+		 * trim the existing memory map.
+		 */
+		unsigned long long start_at, mem_size;
+
+		mem_size = memparse(arg, &arg);
+		if (*arg == '@') {
+			start_at = memparse(arg+1, &arg);
+			add_memory_region(start_at, mem_size, E820_RAM);
+		} else if (*arg == '#') {
+			start_at = memparse(arg+1, &arg);
+			add_memory_region(start_at, mem_size, E820_ACPI);
+		} else if (*arg == '$') {
+			start_at = memparse(arg+1, &arg);
+			add_memory_region(start_at, mem_size, E820_RESERVED);
+		} else {
+			limit_regions(mem_size);
+			user_defined_memmap = 1;
 		}
+	}
+	return 0;
+}
+early_param("memmap", parse_memmap);
 
-		else if (!memcmp(from, "acpi_sci=edge", 13))
-			acpi_sci_flags.trigger =  1;
-
-		else if (!memcmp(from, "acpi_sci=level", 14))
-			acpi_sci_flags.trigger = 3;
-
-		else if (!memcmp(from, "acpi_sci=high", 13))
-			acpi_sci_flags.polarity = 1;
-
-		else if (!memcmp(from, "acpi_sci=low", 12))
-			acpi_sci_flags.polarity = 3;
-
-#ifdef CONFIG_X86_IO_APIC
-		else if (!memcmp(from, "acpi_skip_timer_override", 24))
-			acpi_skip_timer_override = 1;
-
-		if (!memcmp(from, "disable_timer_pin_1", 19))
-			disable_timer_pin_1 = 1;
-		if (!memcmp(from, "enable_timer_pin_1", 18))
-			disable_timer_pin_1 = -1;
+#ifdef CONFIG_PROC_VMCORE
+/* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel.
+ */
+static int __init parse_elfcorehdr(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-		/* disable IO-APIC */
-		else if (!memcmp(from, "noapic", 6))
-			disable_ioapic_setup();
-#endif /* CONFIG_X86_IO_APIC */
-#endif /* CONFIG_ACPI */
+	elfcorehdr_addr = memparse(arg, &arg);
+	return 0;
+}
+early_param("elfcorehdr", parse_elfcorehdr);
+#endif /* CONFIG_PROC_VMCORE */
 
-#ifdef CONFIG_X86_LOCAL_APIC
-		/* enable local APIC */
-		else if (!memcmp(from, "lapic", 5))
-			lapic_enable();
+/*
+ * highmem=size forces highmem to be exactly 'size' bytes.
+ * This works even on boxes that have no highmem otherwise.
+ * This also works to reduce highmem size on bigger boxes.
+ */
+static int __init parse_highmem(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-		/* disable local APIC */
-		else if (!memcmp(from, "nolapic", 6))
-			lapic_disable();
-#endif /* CONFIG_X86_LOCAL_APIC */
+	highmem_pages = memparse(arg, &arg) >> PAGE_SHIFT;
+	return 0;
+}
+early_param("highmem", parse_highmem);
 
-#ifdef CONFIG_KEXEC
-		/* crashkernel=size@addr specifies the location to reserve for
-		 * a crash kernel.  By reserving this memory we guarantee
-		 * that linux never set's it up as a DMA target.
-		 * Useful for holding code to do something appropriate
-		 * after a kernel panic.
-		 */
-		else if (!memcmp(from, "crashkernel=", 12)) {
-			unsigned long size, base;
-			size = memparse(from+12, &from);
-			if (*from == '@') {
-				base = memparse(from+1, &from);
-				/* FIXME: Do I want a sanity check
-				 * to validate the memory range?
-				 */
-				crashk_res.start = base;
-				crashk_res.end   = base + size - 1;
-			}
-		}
-#endif
-#ifdef CONFIG_PROC_VMCORE
-		/* elfcorehdr= specifies the location of elf core header
-		 * stored by the crashed kernel.
-		 */
-		else if (!memcmp(from, "elfcorehdr=", 11))
-			elfcorehdr_addr = memparse(from+11, &from);
-#endif
+/*
+ * vmalloc=size forces the vmalloc area to be exactly 'size'
+ * bytes. This can be used to increase (or decrease) the
+ * vmalloc area - the default is 128m.
+ */
+static int __init parse_vmalloc(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-		/*
-		 * highmem=size forces highmem to be exactly 'size' bytes.
-		 * This works even on boxes that have no highmem otherwise.
-		 * This also works to reduce highmem size on bigger boxes.
-		 */
-		else if (!memcmp(from, "highmem=", 8))
-			highmem_pages = memparse(from+8, &from) >> PAGE_SHIFT;
-	
-		/*
-		 * vmalloc=size forces the vmalloc area to be exactly 'size'
-		 * bytes. This can be used to increase (or decrease) the
-		 * vmalloc area - the default is 128m.
-		 */
-		else if (!memcmp(from, "vmalloc=", 8))
-			__VMALLOC_RESERVE = memparse(from+8, &from);
-
-	next_char:
-		c = *(from++);
-		if (!c)
-			break;
-		if (COMMAND_LINE_SIZE <= ++len)
-			break;
-		*(to++) = c;
-	}
-	*to = '\0';
-	*cmdline_p = command_line;
-	if (userdef) {
-		printk(KERN_INFO "user-defined physical RAM map:\n");
-		print_memory_map("user");
-	}
+	__VMALLOC_RESERVE = memparse(arg, &arg);
+	return 0;
 }
+early_param("vmalloc", parse_vmalloc);
 
 /*
  * Callback for efi_memory_walk.
@@ -1507,17 +1395,15 @@ void __init setup_arch(char **cmdline_p)
 	data_resource.start = virt_to_phys(_etext);
 	data_resource.end = virt_to_phys(_edata)-1;
 
-	parse_cmdline_early(cmdline_p);
+	parse_early_param();
 
-#ifdef CONFIG_EARLY_PRINTK
-	{
-		char *s = strstr(*cmdline_p, "earlyprintk=");
-		if (s) {
-			setup_early_printk(strchr(s, '=') + 1);
-			printk("early console enabled\n");
-		}
+	if (user_defined_memmap) {
+		printk(KERN_INFO "user-defined physical RAM map:\n");
+		print_memory_map("user");
 	}
-#endif
+
+	strlcpy(command_line, saved_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
 
 	max_low_pfn = setup_memory();
 
@@ -1546,7 +1432,7 @@ void __init setup_arch(char **cmdline_p)
 	dmi_scan_machine();
 
 #ifdef CONFIG_X86_GENERICARCH
-	generic_apic_probe(*cmdline_p);
+	generic_apic_probe();
 #endif	
 	if (efi_enabled)
 		efi_map_memmap();
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 9367af76ce37..517eb3874550 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -1491,3 +1491,16 @@ void __init smp_intr_init(void)
 	/* IPI for generic function call */
 	set_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 }
+
+/*
+ * If the BIOS enumerates physical processors before logical,
+ * maxcpus=N at enumeration-time can be used to disable HT.
+ */
+static int __init parse_maxcpus(char *arg)
+{
+	extern unsigned int maxcpus;
+
+	maxcpus = simple_strtoul(arg, NULL, 0);
+	return 0;
+}
+early_param("maxcpus", parse_maxcpus);
diff --git a/arch/i386/mach-generic/probe.c b/arch/i386/mach-generic/probe.c
index 793d1b473251..94b1fd9cbe3c 100644
--- a/arch/i386/mach-generic/probe.c
+++ b/arch/i386/mach-generic/probe.c
@@ -9,6 +9,7 @@
 #include <linux/kernel.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
+#include <linux/errno.h>
 #include <asm/fixmap.h>
 #include <asm/mpspec.h>
 #include <asm/apicdef.h>
@@ -29,7 +30,24 @@ struct genapic *apic_probe[] __initdata = {
 	NULL,
 };
 
-static int cmdline_apic;
+static int cmdline_apic __initdata;
+static int __init parse_apic(char *arg)
+{
+	int i;
+
+	if (!arg)
+		return -EINVAL;
+
+	for (i = 0; apic_probe[i]; i++) {
+		if (!strcmp(apic_probe[i]->name, arg)) {
+			genapic = apic_probe[i];
+			cmdline_apic = 1;
+			return 0;
+		}
+	}
+	return -ENOENT;
+}
+early_param("apic", parse_apic);
 
 void __init generic_bigsmp_probe(void)
 {
@@ -48,40 +66,20 @@ void __init generic_bigsmp_probe(void)
 		}
 }
 
-void __init generic_apic_probe(char *command_line) 
+void __init generic_apic_probe(void)
 { 
-	char *s;
-	int i;
-	int changed = 0;
-
-	s = strstr(command_line, "apic=");
-	if (s && (s == command_line || isspace(s[-1]))) { 
-		char *p = strchr(s, ' '), old; 
-		if (!p)
-			p = strchr(s, '\0'); 
-		old = *p; 
-		*p = 0; 
-		for (i = 0; !changed && apic_probe[i]; i++) {
-			if (!strcmp(apic_probe[i]->name, s+5)) { 
-				changed = 1;
+	if (!cmdline_apic) {
+		int i;
+		for (i = 0; apic_probe[i]; i++) {
+			if (apic_probe[i]->probe()) {
 				genapic = apic_probe[i];
+				break;
 			}
 		}
-		if (!changed)
-			printk(KERN_ERR "Unknown genapic `%s' specified.\n", s);
-		*p = old;
-		cmdline_apic = changed;
-	} 
-	for (i = 0; !changed && apic_probe[i]; i++) { 
-		if (apic_probe[i]->probe()) {
-			changed = 1;
-			genapic = apic_probe[i]; 
-		} 
+		/* Not visible without early console */
+		if (!apic_probe[i])
+			panic("Didn't find an APIC driver");
 	}
-	/* Not visible without early console */ 
-	if (!changed) 
-		panic("Didn't find an APIC driver"); 
-
 	printk(KERN_INFO "Using APIC driver %s\n", genapic->name);
 } 
 
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index d0c2fdf6a4d7..951386606d09 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -435,16 +435,22 @@ u64 __supported_pte_mask __read_mostly = ~_PAGE_NX;
  * on      Enable
  * off     Disable
  */
-void __init noexec_setup(const char *str)
+static int __init noexec_setup(char *str)
 {
-	if (!strncmp(str, "on",2) && cpu_has_nx) {
-		__supported_pte_mask |= _PAGE_NX;
-		disable_nx = 0;
-	} else if (!strncmp(str,"off",3)) {
+	if (!str || !strcmp(str, "on")) {
+		if (cpu_has_nx) {
+			__supported_pte_mask |= _PAGE_NX;
+			disable_nx = 0;
+		}
+	} else if (!strcmp(str,"off")) {
 		disable_nx = 1;
 		__supported_pte_mask &= ~_PAGE_NX;
-	}
+	} else
+		return -EINVAL;
+
+	return 0;
 }
+early_param("noexec", noexec_setup);
 
 int nx_enabled = 0;
 #ifdef CONFIG_X86_PAE
diff --git a/include/asm-i386/acpi.h b/include/asm-i386/acpi.h
index 20f523954218..6016632d032f 100644
--- a/include/asm-i386/acpi.h
+++ b/include/asm-i386/acpi.h
@@ -131,21 +131,7 @@ static inline void disable_acpi(void)
 extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
 
 #ifdef CONFIG_X86_IO_APIC
-extern int skip_ioapic_setup;
 extern int acpi_skip_timer_override;
-
-static inline void disable_ioapic_setup(void)
-{
-	skip_ioapic_setup = 1;
-}
-
-static inline int ioapic_setup_disabled(void)
-{
-	return skip_ioapic_setup;
-}
-
-#else
-static inline void disable_ioapic_setup(void) { }
 #endif
 
 static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h
index 2c1e371cebb6..8c5331ceca7b 100644
--- a/include/asm-i386/apic.h
+++ b/include/asm-i386/apic.h
@@ -42,6 +42,8 @@ static inline void lapic_enable(void)
 	} while (0)
 
 
+extern void generic_apic_probe(void);
+
 #ifdef CONFIG_X86_LOCAL_APIC
 
 /*
@@ -117,8 +119,6 @@ extern void enable_APIC_timer(void);
 
 extern void enable_NMI_through_LVT0 (void * dummy);
 
-extern int disable_timer_pin_1;
-
 void smp_send_timer_broadcast_ipi(struct pt_regs *regs);
 void switch_APIC_timer_to_ipi(void *cpumask);
 void switch_ipi_to_APIC_timer(void *cpumask);
diff --git a/include/asm-i386/io_apic.h b/include/asm-i386/io_apic.h
index 5092e819b8a2..5d309275a1dc 100644
--- a/include/asm-i386/io_apic.h
+++ b/include/asm-i386/io_apic.h
@@ -188,6 +188,16 @@ static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned
 /* 1 if "noapic" boot option passed */
 extern int skip_ioapic_setup;
 
+static inline void disable_ioapic_setup(void)
+{
+	skip_ioapic_setup = 1;
+}
+
+static inline int ioapic_setup_disabled(void)
+{
+	return skip_ioapic_setup;
+}
+
 /*
  * If we use the IO-APIC for IRQ routing, disable automatic
  * assignment of PCI IRQ's.
@@ -206,6 +216,7 @@ extern int (*ioapic_renumber_irq)(int ioapic, int irq);
 
 #else  /* !CONFIG_X86_IO_APIC */
 #define io_apic_assign_pci_irqs 0
+static inline void disable_ioapic_setup(void) { }
 #endif
 
 extern int assign_irq_vector(int irq);
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 09697fec3d2b..64140f2f1b95 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -391,8 +391,6 @@ extern pte_t *lookup_address(unsigned long address);
  static inline int set_kernel_exec(unsigned long vaddr, int enable) { return 0;}
 #endif
 
-extern void noexec_setup(const char *str);
-
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
 	((pte_t *)kmap_atomic(pmd_page(*(dir)),KM_PTE0) + pte_index(address))
-- 
cgit v1.2.3


From 2c8c0e6b8d7700a990da8d24eff767f9ca223b96 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Convert x86-64 to early param

Instead of hackish manual parsing

Requires earlier i386 patchkit, but also fixes i386 early_printk again.

I removed some obsolete really early parameters which didn't do anything useful.
Also made a few parameters that needed it early (mostly oops printing setup)

Also removed one panic check that wasn't visible without
early console anyways (the early console is now initialized after that
panic)

This cleans up a lot of code.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c          |  34 ++++---
 arch/x86_64/kernel/e820.c          |  53 ++++++++--
 arch/x86_64/kernel/early_printk.c  |  20 ++--
 arch/x86_64/kernel/head64.c        |  15 ---
 arch/x86_64/kernel/io_apic.c       |  15 ++-
 arch/x86_64/kernel/machine_kexec.c |  28 ++++++
 arch/x86_64/kernel/pci-dma.c       |   7 +-
 arch/x86_64/kernel/setup.c         | 197 ++++---------------------------------
 arch/x86_64/kernel/setup64.c       |   8 +-
 arch/x86_64/kernel/smpboot.c       |   6 +-
 arch/x86_64/kernel/traps.c         |  24 +++--
 arch/x86_64/mm/numa.c              |   9 +-
 include/asm-x86_64/apic.h          |   4 +-
 include/asm-x86_64/e820.h          |   3 +-
 include/asm-x86_64/pgtable.h       |   1 -
 include/asm-x86_64/proto.h         |  10 --
 16 files changed, 159 insertions(+), 275 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 496cd1bd2ae5..660602c51f51 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -36,6 +36,7 @@
 #include <asm/idle.h>
 #include <asm/proto.h>
 #include <asm/timex.h>
+#include <asm/apic.h>
 
 int apic_verbosity;
 int apic_runs_main_timer;
@@ -546,18 +547,24 @@ static void apic_pm_activate(void) { }
 
 static int __init apic_set_verbosity(char *str)
 {
+	if (str == NULL)  {
+		skip_ioapic_setup = 0;
+		ioapic_force = 1;
+		return 0;
+	}
 	if (strcmp("debug", str) == 0)
 		apic_verbosity = APIC_DEBUG;
 	else if (strcmp("verbose", str) == 0)
 		apic_verbosity = APIC_VERBOSE;
-	else
+	else {
 		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
-				" use apic=verbose or apic=debug", str);
+				" use apic=verbose or apic=debug\n", str);
+		return -EINVAL;
+	}
 
-	return 1;
+	return 0;
 }
-
-__setup("apic=", apic_set_verbosity);
+early_param("apic", apic_set_verbosity);
 
 /*
  * Detect and enable local APICs on non-SMP boards.
@@ -1078,14 +1085,17 @@ int __init APIC_init_uniprocessor (void)
 static __init int setup_disableapic(char *str) 
 { 
 	disable_apic = 1;
-	return 1;
-} 
+	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+	return 0;
+}
+early_param("disableapic", setup_disableapic);
 
+/* same as disableapic, for compatibility */
 static __init int setup_nolapic(char *str) 
 { 
-	disable_apic = 1;
-	return 1;
+	return setup_disableapic(str);
 } 
+early_param("nolapic", setup_nolapic);
 
 static __init int setup_noapictimer(char *str) 
 { 
@@ -1118,11 +1128,5 @@ static __init int setup_apicpmtimer(char *s)
 }
 __setup("apicpmtimer", setup_apicpmtimer);
 
-/* dummy parsing: see setup.c */
-
-__setup("disableapic", setup_disableapic); 
-__setup("nolapic", setup_nolapic);  /* same as disableapic, for compatibility */
-
 __setup("noapictimer", setup_noapictimer); 
 
-/* no "lapic" flag - we only use the lapic when the BIOS tells us so. */
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 1362aad4f3aa..a4739dc82937 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -596,31 +596,64 @@ void __init setup_memory_region(void)
 	e820_print_map(who);
 }
 
-void __init parse_memopt(char *p, char **from) 
-{ 
-	end_user_pfn = memparse(p, from);
+static int __init parse_memopt(char *p)
+{
+	if (!p)
+		return -EINVAL;
+	end_user_pfn = memparse(p, &p);
 	end_user_pfn >>= PAGE_SHIFT;	
+	return 0;
 } 
+early_param("mem", parse_memopt);
+
+static int userdef __initdata;
 
-void __init parse_memmapopt(char *p, char **from)
+static int __init parse_memmap_opt(char *p)
 {
+	char *oldp;
 	unsigned long long start_at, mem_size;
 
-	mem_size = memparse(p, from);
-	p = *from;
+	if (!strcmp(p, "exactmap")) {
+#ifdef CONFIG_CRASH_DUMP
+		/* If we are doing a crash dump, we
+		 * still need to know the real mem
+		 * size before original memory map is
+		 * reset.
+		 */
+		saved_max_pfn = e820_end_of_ram();
+#endif
+		end_pfn_map = 0;
+		e820.nr_map = 0;
+		userdef = 1;
+		return 0;
+	}
+
+	oldp = p;
+	mem_size = memparse(p, &p);
+	if (p == oldp)
+		return -EINVAL;
 	if (*p == '@') {
-		start_at = memparse(p+1, from);
+		start_at = memparse(p+1, &p);
 		add_memory_region(start_at, mem_size, E820_RAM);
 	} else if (*p == '#') {
-		start_at = memparse(p+1, from);
+		start_at = memparse(p+1, &p);
 		add_memory_region(start_at, mem_size, E820_ACPI);
 	} else if (*p == '$') {
-		start_at = memparse(p+1, from);
+		start_at = memparse(p+1, &p);
 		add_memory_region(start_at, mem_size, E820_RESERVED);
 	} else {
 		end_user_pfn = (mem_size >> PAGE_SHIFT);
 	}
-	p = *from;
+	return *p == '\0' ? 0 : -EINVAL;
+}
+early_param("memmap", parse_memmap_opt);
+
+void finish_e820_parsing(void)
+{
+	if (userdef) {
+		printk(KERN_INFO "user-defined physical RAM map:\n");
+		e820_print_map("user");
+	}
 }
 
 unsigned long pci_mem_start = 0xaeedbabe;
diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index 140051e07fa6..e22ecd54870d 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -215,20 +215,16 @@ void early_printk(const char *fmt, ...)
 
 static int __initdata keep_early;
 
-int __init setup_early_printk(char *opt)
+static int __init setup_early_printk(char *buf)
 {
-	char *space;
-	char buf[256];
+	if (!buf)
+		return 0;
 
 	if (early_console_initialized)
-		return 1;
-
-	strlcpy(buf,opt,sizeof(buf));
-	space = strchr(buf, ' ');
-	if (space)
-		*space = 0;
+		return 0;
+	early_console_initialized = 1;
 
-	if (strstr(buf,"keep"))
+	if (!strcmp(buf,"keep"))
 		keep_early = 1;
 
 	if (!strncmp(buf, "serial", 6)) {
@@ -248,11 +244,12 @@ int __init setup_early_printk(char *opt)
  		early_console = &simnow_console;
  		keep_early = 1;
 	}
-	early_console_initialized = 1;
 	register_console(early_console);
 	return 0;
 }
 
+early_param("earlyprintk", setup_early_printk);
+
 void __init disable_early_printk(void)
 {
 	if (!early_console_initialized || !early_console)
@@ -266,4 +263,3 @@ void __init disable_early_printk(void)
 	}
 }
 
-__setup("earlyprintk=", setup_early_printk);
diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index f2461fde9f8f..6716cbfc34ac 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -58,7 +58,6 @@ static void __init copy_bootdata(char *real_mode_data)
 
 void __init x86_64_start_kernel(char * real_mode_data)
 {
-	char *s;
 	int i;
 
 	for (i = 0; i < 256; i++)
@@ -85,19 +84,5 @@ void __init x86_64_start_kernel(char * real_mode_data)
 #ifdef CONFIG_SMP
 	cpu_set(0, cpu_online_map);
 #endif
-	s = strstr(saved_command_line, "earlyprintk=");
-	if (s != NULL)
-		setup_early_printk(strchr(s, '=') + 1);
-#ifdef CONFIG_NUMA
-	s = strstr(saved_command_line, "numa=");
-	if (s != NULL)
-		numa_setup(s+5);
-#endif
-	if (strstr(saved_command_line, "disableapic"))
-		disable_apic = 1;
-	/* You need early console to see that */
-	if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
-		panic("Kernel too big for kernel mapping\n");
-
 	start_kernel();
 }
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index a1412042b918..afac3dbb3729 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -48,7 +48,7 @@ int sis_apic_bug; /* not actually supported, dummy for compile */
 
 static int no_timer_check;
 
-int disable_timer_pin_1 __initdata;
+static int disable_timer_pin_1 __initdata;
 
 int timer_over_8254 __initdata = 0;
 
@@ -253,18 +253,17 @@ int ioapic_force;
 static int __init disable_ioapic_setup(char *str)
 {
 	skip_ioapic_setup = 1;
-	return 1;
+	return 0;
 }
+early_param("noapic", disable_ioapic_setup);
 
-static int __init enable_ioapic_setup(char *str)
+/* Actually the next is obsolete, but keep it for paranoid reasons -AK */
+static int __init disable_timer_pin_setup(char *arg)
 {
-	ioapic_force = 1;
-	skip_ioapic_setup = 0;
+	disable_timer_pin_1 = 1;
 	return 1;
 }
-
-__setup("noapic", disable_ioapic_setup);
-__setup("apic", enable_ioapic_setup);
+__setup("disable_timer_pin_1", disable_timer_pin_setup);
 
 static int __init setup_disable_8254_timer(char *s)
 {
diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index 106076b370fc..2e94c072d84a 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -226,3 +226,31 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	rnk = (relocate_new_kernel_t) control_code_buffer;
 	(*rnk)(page_list, control_code_buffer, image->start, start_pgtable);
 }
+
+/* crashkernel=size@addr specifies the location to reserve for
+ * a crash kernel.  By reserving this memory we guarantee
+ * that linux never set's it up as a DMA target.
+ * Useful for holding code to do something appropriate
+ * after a kernel panic.
+ */
+static int __init setup_crashkernel(char *arg)
+{
+	unsigned long size, base;
+	char *p;
+	if (!arg)
+		return -EINVAL;
+	size = memparse(arg, &p);
+	if (arg == p)
+		return -EINVAL;
+	if (*p == '@') {
+		base = memparse(p+1, &p);
+		/* FIXME: Do I want a sanity check to validate the
+		 * memory range?  Yes you do, but it's too early for
+		 * e820 -AK */
+		crashk_res.start = base;
+		crashk_res.end   = base + size - 1;
+	}
+	return 0;
+}
+early_param("crashkernel", setup_crashkernel);
+
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 9c44f4f2433d..4dcb671bd19f 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/arch/x86_64/kernel/pci-dma.c
@@ -236,6 +236,9 @@ __init int iommu_setup(char *p)
 {
     iommu_merge = 1;
 
+	if (!p)
+		return -EINVAL;
+
     while (*p) {
 	    if (!strncmp(p,"off",3))
 		    no_iommu = 1;
@@ -278,9 +281,9 @@ __init int iommu_setup(char *p)
 	    if (*p == ',')
 		    ++p;
     }
-    return 1;
+    return 0;
 }
-__setup("iommu=", iommu_setup);
+early_param("iommu", iommu_setup);
 
 void __init pci_iommu_alloc(void)
 {
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 73f1cdd140fe..f55540a6084e 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -76,11 +76,6 @@ unsigned long mmu_cr4_features;
 
 int acpi_disabled;
 EXPORT_SYMBOL(acpi_disabled);
-#ifdef	CONFIG_ACPI
-extern int __initdata acpi_ht;
-extern acpi_interrupt_flags	acpi_sci_flags;
-int __initdata acpi_force = 0;
-#endif
 
 int acpi_numa __initdata;
 
@@ -276,183 +271,22 @@ static void __init probe_roms(void)
 	}
 }
 
-/* Check for full argument with no trailing characters */
-static int fullarg(char *p, char *arg)
+#ifdef CONFIG_PROC_VMCORE
+/* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel. This option will be passed
+ * by kexec loader to the capture kernel.
+ */
+static int __init setup_elfcorehdr(char *arg)
 {
-	int l = strlen(arg);
-	return !memcmp(p, arg, l) && (p[l] == 0 || isspace(p[l]));
+	char *end;
+	if (!arg)
+		return -EINVAL;
+	elfcorehdr_addr = memparse(arg, &end);
+	return end > arg ? 0 : -EINVAL;
 }
-
-static __init void parse_cmdline_early (char ** cmdline_p)
-{
-	char c = ' ', *to = command_line, *from = COMMAND_LINE;
-	int len = 0;
-	int userdef = 0;
-
-	for (;;) {
-		if (c != ' ') 
-			goto next_char; 
-
-#ifdef  CONFIG_SMP
-		/*
-		 * If the BIOS enumerates physical processors before logical,
-		 * maxcpus=N at enumeration-time can be used to disable HT.
-		 */
-		else if (!memcmp(from, "maxcpus=", 8)) {
-			extern unsigned int maxcpus;
-
-			maxcpus = simple_strtoul(from + 8, NULL, 0);
-		}
-#endif
-#ifdef CONFIG_ACPI
-		/* "acpi=off" disables both ACPI table parsing and interpreter init */
-		if (fullarg(from,"acpi=off"))
-			disable_acpi();
-
-		if (fullarg(from, "acpi=force")) { 
-			/* add later when we do DMI horrors: */
-			acpi_force = 1;
-			acpi_disabled = 0;
-		}
-
-		/* acpi=ht just means: do ACPI MADT parsing 
-		   at bootup, but don't enable the full ACPI interpreter */
-		if (fullarg(from, "acpi=ht")) { 
-			if (!acpi_force)
-				disable_acpi();
-			acpi_ht = 1; 
-		}
-                else if (fullarg(from, "pci=noacpi")) 
-			acpi_disable_pci();
-		else if (fullarg(from, "acpi=noirq"))
-			acpi_noirq_set();
-
-		else if (fullarg(from, "acpi_sci=edge"))
-			acpi_sci_flags.trigger =  1;
-		else if (fullarg(from, "acpi_sci=level"))
-			acpi_sci_flags.trigger = 3;
-		else if (fullarg(from, "acpi_sci=high"))
-			acpi_sci_flags.polarity = 1;
-		else if (fullarg(from, "acpi_sci=low"))
-			acpi_sci_flags.polarity = 3;
-
-		/* acpi=strict disables out-of-spec workarounds */
-		else if (fullarg(from, "acpi=strict")) {
-			acpi_strict = 1;
-		}
-		else if (fullarg(from, "acpi_skip_timer_override"))
-			acpi_skip_timer_override = 1;
+early_param("elfcorehdr", setup_elfcorehdr);
 #endif
 
-		if (fullarg(from, "disable_timer_pin_1"))
-			disable_timer_pin_1 = 1;
-		if (fullarg(from, "enable_timer_pin_1"))
-			disable_timer_pin_1 = -1;
-
-		if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
-			clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
-			disable_apic = 1;
-		}
-
-		if (fullarg(from, "noapic"))
-			skip_ioapic_setup = 1;
-
-		if (fullarg(from,"apic")) {
-			skip_ioapic_setup = 0;
-			ioapic_force = 1;
-		}
-			
-		if (!memcmp(from, "mem=", 4))
-			parse_memopt(from+4, &from); 
-
-		if (!memcmp(from, "memmap=", 7)) {
-			/* exactmap option is for used defined memory */
-			if (!memcmp(from+7, "exactmap", 8)) {
-#ifdef CONFIG_CRASH_DUMP
-				/* If we are doing a crash dump, we
-				 * still need to know the real mem
-				 * size before original memory map is
-				 * reset.
-				 */
-				saved_max_pfn = e820_end_of_ram();
-#endif
-				from += 8+7;
-				end_pfn_map = 0;
-				e820.nr_map = 0;
-				userdef = 1;
-			}
-			else {
-				parse_memmapopt(from+7, &from);
-				userdef = 1;
-			}
-		}
-
-#ifdef CONFIG_NUMA
-		if (!memcmp(from, "numa=", 5))
-			numa_setup(from+5); 
-#endif
-
-		if (!memcmp(from,"iommu=",6)) { 
-			iommu_setup(from+6); 
-		}
-
-		if (fullarg(from,"oops=panic"))
-			panic_on_oops = 1;
-
-		if (!memcmp(from, "noexec=", 7))
-			nonx_setup(from + 7);
-
-#ifdef CONFIG_KEXEC
-		/* crashkernel=size@addr specifies the location to reserve for
-		 * a crash kernel.  By reserving this memory we guarantee
-		 * that linux never set's it up as a DMA target.
-		 * Useful for holding code to do something appropriate
-		 * after a kernel panic.
-		 */
-		else if (!memcmp(from, "crashkernel=", 12)) {
-			unsigned long size, base;
-			size = memparse(from+12, &from);
-			if (*from == '@') {
-				base = memparse(from+1, &from);
-				/* FIXME: Do I want a sanity check
-				 * to validate the memory range?
-				 */
-				crashk_res.start = base;
-				crashk_res.end   = base + size - 1;
-			}
-		}
-#endif
-
-#ifdef CONFIG_PROC_VMCORE
-		/* elfcorehdr= specifies the location of elf core header
-		 * stored by the crashed kernel. This option will be passed
-		 * by kexec loader to the capture kernel.
-		 */
-		else if(!memcmp(from, "elfcorehdr=", 11))
-			elfcorehdr_addr = memparse(from+11, &from);
-#endif
-
-#ifdef CONFIG_HOTPLUG_CPU
-		else if (!memcmp(from, "additional_cpus=", 16))
-			setup_additional_cpus(from+16);
-#endif
-
-	next_char:
-		c = *(from++);
-		if (!c)
-			break;
-		if (COMMAND_LINE_SIZE <= ++len)
-			break;
-		*(to++) = c;
-	}
-	if (userdef) {
-		printk(KERN_INFO "user-defined physical RAM map:\n");
-		e820_print_map("user");
-	}
-	*to = '\0';
-	*cmdline_p = command_line;
-}
-
 #ifndef CONFIG_NUMA
 static void __init
 contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
@@ -547,7 +381,12 @@ void __init setup_arch(char **cmdline_p)
 
 	early_identify_cpu(&boot_cpu_data);
 
-	parse_cmdline_early(cmdline_p);
+	strlcpy(command_line, saved_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
+
+	parse_early_param();
+
+	finish_e820_parsing();
 
 	/*
 	 * partially used pages are not usable - thus
diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 417de564456e..646caa0a20bb 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -46,8 +46,10 @@ Control non executable mappings for 64bit processes.
 on	Enable(default)
 off	Disable
 */ 
-int __init nonx_setup(char *str)
+static int __init nonx_setup(char *str)
 {
+	if (!str)
+		return -EINVAL;
 	if (!strncmp(str, "on", 2)) {
                 __supported_pte_mask |= _PAGE_NX; 
  		do_not_nx = 0; 
@@ -55,9 +57,9 @@ int __init nonx_setup(char *str)
 		do_not_nx = 1;
 		__supported_pte_mask &= ~_PAGE_NX;
         }
-	return 1;
+	return 0;
 } 
-__setup("noexec=", nonx_setup);	/* parsed early actually */
+early_param("noexec", nonx_setup);
 
 int force_personality32 = 0; 
 
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index ae7d6d84e352..0b9f28b05e7d 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1270,11 +1270,11 @@ void __cpu_die(unsigned int cpu)
  	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
 }
 
-__init int setup_additional_cpus(char *s)
+static __init int setup_additional_cpus(char *s)
 {
-	return get_option(&s, &additional_cpus);
+	return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL;
 }
-__setup("additional_cpus=", setup_additional_cpus);
+early_param("additional_cpus", setup_additional_cpus);
 
 #else /* ... !CONFIG_HOTPLUG_CPU */
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 56d7ff0c894c..9ec2b1d5893d 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -1119,24 +1119,30 @@ void __init trap_init(void)
 }
 
 
-/* Actual parsing is done early in setup.c. */
-static int __init oops_dummy(char *s)
+static int __init oops_setup(char *s)
 { 
-	panic_on_oops = 1;
-	return 1;
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "panic"))
+		panic_on_oops = 1;
+	return 0;
 } 
-__setup("oops=", oops_dummy); 
+early_param("oops", oops_setup);
 
 static int __init kstack_setup(char *s)
 {
+	if (!s)
+		return -EINVAL;
 	kstack_depth_to_print = simple_strtoul(s,NULL,0);
-	return 1;
+	return 0;
 }
-__setup("kstack=", kstack_setup);
+early_param("kstack", kstack_setup);
 
 #ifdef CONFIG_STACK_UNWIND
 static int __init call_trace_setup(char *s)
 {
+	if (!s)
+		return -EINVAL;
 	if (strcmp(s, "old") == 0)
 		call_trace = -1;
 	else if (strcmp(s, "both") == 0)
@@ -1145,7 +1151,7 @@ static int __init call_trace_setup(char *s)
 		call_trace = 1;
 	else if (strcmp(s, "new") == 0)
 		call_trace = 2;
-	return 1;
+	return 0;
 }
-__setup("call_trace=", call_trace_setup);
+early_param("call_trace", call_trace_setup);
 #endif
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index b2fac14baac0..d64d6d93d04b 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -348,9 +348,10 @@ void __init paging_init(void)
 	}
 } 
 
-/* [numa=off] */
-__init int numa_setup(char *opt) 
+static __init int numa_setup(char *opt)
 { 
+	if (!opt)
+		return -EINVAL;
 	if (!strncmp(opt,"off",3))
 		numa_off = 1;
 #ifdef CONFIG_NUMA_EMU
@@ -366,9 +367,11 @@ __init int numa_setup(char *opt)
 	if (!strncmp(opt,"hotadd=", 7))
 		hotadd_percent = simple_strtoul(opt+7, NULL, 10);
 #endif
-	return 1;
+	return 0;
 } 
 
+early_param("numa", numa_setup);
+
 /*
  * Setup early cpu_to_node.
  *
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index 8ed0f4d67b8d..29ee735278f2 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -17,6 +17,7 @@
 
 extern int apic_verbosity;
 extern int apic_runs_main_timer;
+extern int ioapic_force;
 
 /*
  * Define the default level of output to be very little
@@ -93,9 +94,6 @@ extern void setup_APIC_extened_lvt(unsigned char lvt_off, unsigned char vector,
 #define K8_APIC_EXT_INT_MSG_EXT 0x7
 #define K8_APIC_EXT_LVT_ENTRY_THRESHOLD    0
 
-extern int disable_timer_pin_1;
-
-
 void smp_send_timer_broadcast_ipi(void);
 void switch_APIC_timer_to_ipi(void *cpumask);
 void switch_ipi_to_APIC_timer(void *cpumask);
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index 670a3388e70a..dba012151856 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -55,8 +55,7 @@ extern void e820_setup_gap(void);
 extern unsigned long e820_hole_size(unsigned long start_pfn,
 				    unsigned long end_pfn);
 
-extern void __init parse_memopt(char *p, char **end);
-extern void __init parse_memmapopt(char *p, char **end);
+extern void finish_e820_parsing(void);
 
 extern struct e820map e820;
 
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index 0c1e2422400a..d79e7441b513 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -21,7 +21,6 @@ extern unsigned long __supported_pte_mask;
 
 #define swapper_pg_dir init_level4_pgt
 
-extern int nonx_setup(char *str);
 extern void paging_init(void);
 extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 58fec91318e4..151826f8ce69 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -53,9 +53,6 @@ extern int nohpet;
 extern unsigned long vxtime_hz;
 extern void time_init_gtod(void);
 
-extern int numa_setup(char *opt);
-
-extern int setup_early_printk(char *); 
 extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 
 extern void early_identify_cpu(struct cpuinfo_x86 *c);
@@ -104,13 +101,7 @@ extern void select_idle_routine(const struct cpuinfo_x86 *c);
 extern unsigned long table_start, table_end;
 
 extern int exception_trace;
-extern int using_apic_timer;
-extern int disable_apic;
 extern unsigned cpu_khz;
-extern int ioapic_force;
-extern int skip_ioapic_setup;
-extern int acpi_ht;
-extern int acpi_disabled;
 
 extern void no_iommu_init(void);
 extern int force_iommu, no_iommu;
@@ -132,7 +123,6 @@ extern int fix_aperture;
 
 extern int reboot_force;
 extern int notsc_setup(char *);
-extern int setup_additional_cpus(char *);
 
 extern void smp_local_timer_interrupt(struct pt_regs * regs);
 
-- 
cgit v1.2.3


From 91cd444e56ebe0c2acd9576a045d77490b26f607 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] x86: Remove unneeded externs in acpi/boot.c

And move one into proto.h

Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 3 ---
 include/asm-x86_64/proto.h   | 2 ++
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index f3761b98c8dd..c809a3f4c0bf 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -47,9 +47,6 @@ EXPORT_SYMBOL(acpi_disabled);
 
 #ifdef	CONFIG_X86_64
 
-extern void __init clustered_apic_check(void);
-
-extern int gsi_irq_sharing(int gsi);
 #include <asm/proto.h>
 
 static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 151826f8ce69..b73d0c76613c 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -124,6 +124,8 @@ extern int fix_aperture;
 extern int reboot_force;
 extern int notsc_setup(char *);
 
+extern int gsi_irq_sharing(int gsi);
+
 extern void smp_local_timer_interrupt(struct pt_regs * regs);
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
-- 
cgit v1.2.3


From 5a1b3999d6cb7ab87f1f3b1700bc91839fd6fa29 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] x86: Some preparationary cleanup for stack trace

- Remove unused all_contexts parameter
No caller used it
- Move skip argument into the structure (needed for
followon patches)

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/stacktrace.c   | 11 +++--------
 arch/s390/kernel/stacktrace.c   | 17 ++++++++---------
 arch/x86_64/kernel/stacktrace.c | 14 +++++---------
 include/linux/stacktrace.h      |  7 ++++---
 kernel/lockdep.c                |  5 ++++-
 5 files changed, 24 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/stacktrace.c b/arch/i386/kernel/stacktrace.c
index e62a037ab399..ae3c32a87add 100644
--- a/arch/i386/kernel/stacktrace.c
+++ b/arch/i386/kernel/stacktrace.c
@@ -61,12 +61,8 @@ save_context_stack(struct stack_trace *trace, unsigned int skip,
 
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
- * If all_contexts is set, all contexts (hardirq, softirq and process)
- * are saved. If not set then only the current context is saved.
  */
-void save_stack_trace(struct stack_trace *trace,
-		      struct task_struct *task, int all_contexts,
-		      unsigned int skip)
+void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
 	unsigned long ebp;
 	unsigned long *stack = &ebp;
@@ -85,10 +81,9 @@ void save_stack_trace(struct stack_trace *trace,
 		struct thread_info *context = (struct thread_info *)
 				((unsigned long)stack & (~(THREAD_SIZE - 1)));
 
-		ebp = save_context_stack(trace, skip, context, stack, ebp);
+		ebp = save_context_stack(trace, trace->skip, context, stack, ebp);
 		stack = (unsigned long *)context->previous_esp;
-		if (!all_contexts || !stack ||
-				trace->nr_entries >= trace->max_entries)
+		if (!stack || trace->nr_entries >= trace->max_entries)
 			break;
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 		if (trace->nr_entries >= trace->max_entries)
diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index de83f38288d0..d9428a0fc8fb 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -59,9 +59,7 @@ static inline unsigned long save_context_stack(struct stack_trace *trace,
 	}
 }
 
-void save_stack_trace(struct stack_trace *trace,
-		      struct task_struct *task, int all_contexts,
-		      unsigned int skip)
+void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
 	register unsigned long sp asm ("15");
 	unsigned long orig_sp;
@@ -69,22 +67,23 @@ void save_stack_trace(struct stack_trace *trace,
 	sp &= PSW_ADDR_INSN;
 	orig_sp = sp;
 
-	sp = save_context_stack(trace, &skip, sp,
+	sp = save_context_stack(trace, &trace->skip, sp,
 				S390_lowcore.panic_stack - PAGE_SIZE,
 				S390_lowcore.panic_stack);
-	if ((sp != orig_sp) && !all_contexts)
+	if ((sp != orig_sp) && !trace->all_contexts)
 		return;
-	sp = save_context_stack(trace, &skip, sp,
+	sp = save_context_stack(trace, &trace->skip, sp,
 				S390_lowcore.async_stack - ASYNC_SIZE,
 				S390_lowcore.async_stack);
-	if ((sp != orig_sp) && !all_contexts)
+	if ((sp != orig_sp) && !trace->all_contexts)
 		return;
 	if (task)
-		save_context_stack(trace, &skip, sp,
+		save_context_stack(trace, &trace->skip, sp,
 				   (unsigned long) task_stack_page(task),
 				   (unsigned long) task_stack_page(task) + THREAD_SIZE);
 	else
-		save_context_stack(trace, &skip, sp, S390_lowcore.thread_info,
+		save_context_stack(trace, &trace->skip, sp,
+				   S390_lowcore.thread_info,
 				   S390_lowcore.thread_info + THREAD_SIZE);
 	return;
 }
diff --git a/arch/x86_64/kernel/stacktrace.c b/arch/x86_64/kernel/stacktrace.c
index 32cf55eb9af8..1c022af8fe1e 100644
--- a/arch/x86_64/kernel/stacktrace.c
+++ b/arch/x86_64/kernel/stacktrace.c
@@ -109,9 +109,10 @@ out_restore:
  * Save stack-backtrace addresses into a stack_trace buffer:
  */
 static inline unsigned long
-save_context_stack(struct stack_trace *trace, unsigned int skip,
+save_context_stack(struct stack_trace *trace,
 		   unsigned long stack, unsigned long stack_end)
 {
+	int skip = trace->skip;
 	unsigned long addr;
 
 #ifdef CONFIG_FRAME_POINTER
@@ -159,12 +160,8 @@ save_context_stack(struct stack_trace *trace, unsigned int skip,
 
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
- * If all_contexts is set, all contexts (hardirq, softirq and process)
- * are saved. If not set then only the current context is saved.
  */
-void save_stack_trace(struct stack_trace *trace,
-		      struct task_struct *task, int all_contexts,
-		      unsigned int skip)
+void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
 	unsigned long stack = (unsigned long)&stack;
 	int i, nr_stacks = 0, stacks_done[MAX_STACKS];
@@ -207,9 +204,8 @@ void save_stack_trace(struct stack_trace *trace,
 				return;
 		stacks_done[nr_stacks] = stack_end;
 
-		stack = save_context_stack(trace, skip, stack, stack_end);
-		if (!all_contexts || !stack ||
-				trace->nr_entries >= trace->max_entries)
+		stack = save_context_stack(trace, stack, stack_end);
+		if (!stack || trace->nr_entries >= trace->max_entries)
 			return;
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 		if (trace->nr_entries >= trace->max_entries)
diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h
index 9cc81e572224..50e2b01e517c 100644
--- a/include/linux/stacktrace.h
+++ b/include/linux/stacktrace.h
@@ -5,15 +5,16 @@
 struct stack_trace {
 	unsigned int nr_entries, max_entries;
 	unsigned long *entries;
+	int skip;	/* input argument: How many entries to skip */
+	int all_contexts; /* input argument: if true do than one stack */
 };
 
 extern void save_stack_trace(struct stack_trace *trace,
-			     struct task_struct *task, int all_contexts,
-			     unsigned int skip);
+			     struct task_struct *task);
 
 extern void print_stack_trace(struct stack_trace *trace, int spaces);
 #else
-# define save_stack_trace(trace, task, all, skip)	do { } while (0)
+# define save_stack_trace(trace, task)			do { } while (0)
 # define print_stack_trace(trace)			do { } while (0)
 #endif
 
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 9bad17884513..900b4cb1a024 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -224,7 +224,10 @@ static int save_trace(struct stack_trace *trace)
 	trace->max_entries = MAX_STACK_TRACE_ENTRIES - nr_stack_trace_entries;
 	trace->entries = stack_trace + nr_stack_trace_entries;
 
-	save_stack_trace(trace, NULL, 0, 3);
+	trace->skip = 3;
+	trace->all_contexts = 0;
+
+	save_stack_trace(trace, NULL);
 
 	trace->max_entries = trace->nr_entries;
 
-- 
cgit v1.2.3


From b7f5e3c7742d5332b78b831131f43fc3630e6322 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Don't access the APIC in safe_smp_processor_id when it is not
 mapped yet

Lockdep can call the dwarf2 unwinder early, and the dwarf2 code
uses safe_smp_processor_id which tries to access the local APIC page.
But that doesn't work before the APIC code has set up its fixmap.

Check for this case and always return boot cpu then.

Cc: jbeulich@novell.com
Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c | 2 ++
 arch/x86_64/kernel/smp.c  | 2 +-
 include/asm-x86_64/apic.h | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index b95d1e378b96..6472e321cad7 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -38,6 +38,7 @@
 #include <asm/timex.h>
 #include <asm/apic.h>
 
+int apic_mapped;
 int apic_verbosity;
 int apic_runs_main_timer;
 int apic_calibrate_pmtmr __initdata;
@@ -600,6 +601,7 @@ void __init init_apic_mappings(void)
 		apic_phys = mp_lapic_addr;
 
 	set_fixmap_nocache(FIX_APIC_BASE, apic_phys);
+	apic_mapped = 1;
 	apic_printk(APIC_VERBOSE,"mapped APIC to %16lx (%16lx)\n", APIC_BASE, apic_phys);
 
 	/*
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 06af6ca60129..2df988bbf312 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -526,7 +526,7 @@ int safe_smp_processor_id(void)
 {
 	unsigned apicid, i;
 
-	if (disable_apic)
+	if (disable_apic || !apic_mapped)
 		return 0;
 
 	apicid = hard_smp_processor_id();
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index 29ee735278f2..9e66d32330c9 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -18,6 +18,7 @@
 extern int apic_verbosity;
 extern int apic_runs_main_timer;
 extern int ioapic_force;
+extern int apic_mapped;
 
 /*
  * Define the default level of output to be very little
-- 
cgit v1.2.3


From c0b766f13d8e1189ce4d00e54700c9d96b543b9a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Merge stacktrace and show_trace

This unifies the standard backtracer and the new stacktrace
in memory backtracer. The standard one is converted to use callbacks
and then reimplement stacktrace using new callbacks.

The main advantage is that stacktrace can now use the new dwarf2 unwinder
and avoid false positives in many cases.

I kept it simple to make sure the standard backtracer stays reliable.

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/stacktrace.c | 214 +++++-----------------------------------
 arch/x86_64/kernel/traps.c      |  99 ++++++++++++++-----
 include/asm-x86_64/stacktrace.h |  18 ++++
 3 files changed, 120 insertions(+), 211 deletions(-)
 create mode 100644 include/asm-x86_64/stacktrace.h

(limited to 'include')

diff --git a/arch/x86_64/kernel/stacktrace.c b/arch/x86_64/kernel/stacktrace.c
index 1c022af8fe1e..6026b31d037e 100644
--- a/arch/x86_64/kernel/stacktrace.c
+++ b/arch/x86_64/kernel/stacktrace.c
@@ -7,211 +7,49 @@
  */
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
+#include <linux/module.h>
+#include <asm/stacktrace.h>
 
-#include <asm/smp.h>
-
-static inline int
-in_range(unsigned long start, unsigned long addr, unsigned long end)
+static void save_stack_warning(void *data, char *msg)
 {
-	return addr >= start && addr <= end;
 }
 
-static unsigned long
-get_stack_end(struct task_struct *task, unsigned long stack)
+static void
+save_stack_warning_symbol(void *data, char *msg, unsigned long symbol)
 {
-	unsigned long stack_start, stack_end, flags;
-	int i, cpu;
-
-	/*
-	 * The most common case is that we are in the task stack:
-	 */
-	stack_start = (unsigned long)task->thread_info;
-	stack_end = stack_start + THREAD_SIZE;
-
-	if (in_range(stack_start, stack, stack_end))
-		return stack_end;
-
-	/*
-	 * We are in an interrupt if irqstackptr is set:
-	 */
-	raw_local_irq_save(flags);
-	cpu = safe_smp_processor_id();
-	stack_end = (unsigned long)cpu_pda(cpu)->irqstackptr;
-
-	if (stack_end) {
-		stack_start = stack_end & ~(IRQSTACKSIZE-1);
-		if (in_range(stack_start, stack, stack_end))
-			goto out_restore;
-		/*
-		 * We get here if we are in an IRQ context but we
-		 * are also in an exception stack.
-		 */
-	}
-
-	/*
-	 * Iterate over all exception stacks, and figure out whether
-	 * 'stack' is in one of them:
-	 */
-	for (i = 0; i < N_EXCEPTION_STACKS; i++) {
-		/*
-		 * set 'end' to the end of the exception stack.
-		 */
-		stack_end = per_cpu(init_tss, cpu).ist[i];
-		stack_start = stack_end - EXCEPTION_STKSZ;
-
-		/*
-		 * Is 'stack' above this exception frame's end?
-		 * If yes then skip to the next frame.
-		 */
-		if (stack >= stack_end)
-			continue;
-		/*
-		 * Is 'stack' above this exception frame's start address?
-		 * If yes then we found the right frame.
-		 */
-		if (stack >= stack_start)
-			goto out_restore;
-
-		/*
-		 * If this is a debug stack, and if it has a larger size than
-		 * the usual exception stacks, then 'stack' might still
-		 * be within the lower portion of the debug stack:
-		 */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		if (i == DEBUG_STACK - 1 && stack >= stack_end - DEBUG_STKSZ) {
-			/*
-			 * Black magic. A large debug stack is composed of
-			 * multiple exception stack entries, which we
-			 * iterate through now. Dont look:
-			 */
-			do {
-				stack_end -= EXCEPTION_STKSZ;
-				stack_start -= EXCEPTION_STKSZ;
-			} while (stack < stack_start);
-
-			goto out_restore;
-		}
-#endif
-	}
-	/*
-	 * Ok, 'stack' is not pointing to any of the system stacks.
-	 */
-	stack_end = 0;
-
-out_restore:
-	raw_local_irq_restore(flags);
-
-	return stack_end;
 }
 
-
-/*
- * Save stack-backtrace addresses into a stack_trace buffer:
- */
-static inline unsigned long
-save_context_stack(struct stack_trace *trace,
-		   unsigned long stack, unsigned long stack_end)
+static int save_stack_stack(void *data, char *name)
 {
-	int skip = trace->skip;
-	unsigned long addr;
-
-#ifdef CONFIG_FRAME_POINTER
-	unsigned long prev_stack = 0;
+	struct stack_trace *trace = (struct stack_trace *)data;
+	return trace->all_contexts ? 0 : -1;
+}
 
-	while (in_range(prev_stack, stack, stack_end)) {
-		pr_debug("stack:          %p\n", (void *)stack);
-		addr = (unsigned long)(((unsigned long *)stack)[1]);
-		pr_debug("addr:           %p\n", (void *)addr);
-		if (!skip)
-			trace->entries[trace->nr_entries++] = addr-1;
-		else
-			skip--;
-		if (trace->nr_entries >= trace->max_entries)
-			break;
-		if (!addr)
-			return 0;
-		/*
-		 * Stack frames must go forwards (otherwise a loop could
-		 * happen if the stackframe is corrupted), so we move
-		 * prev_stack forwards:
-		 */
-		prev_stack = stack;
-		stack = (unsigned long)(((unsigned long *)stack)[0]);
-	}
-	pr_debug("invalid:        %p\n", (void *)stack);
-#else
-	while (stack < stack_end) {
-		addr = ((unsigned long *)stack)[0];
-		stack += sizeof(long);
-		if (__kernel_text_address(addr)) {
-			if (!skip)
-				trace->entries[trace->nr_entries++] = addr-1;
-			else
-				skip--;
-			if (trace->nr_entries >= trace->max_entries)
-				break;
-		}
+static void save_stack_address(void *data, unsigned long addr)
+{
+	struct stack_trace *trace = (struct stack_trace *)data;
+	if (trace->skip > 0) {
+		trace->skip--;
+		return;
 	}
-#endif
-	return stack;
+	if (trace->nr_entries < trace->max_entries - 1)
+		trace->entries[trace->nr_entries++] = addr;
 }
 
-#define MAX_STACKS 10
+static struct stacktrace_ops save_stack_ops = {
+	.warning = save_stack_warning,
+	.warning_symbol = save_stack_warning_symbol,
+	.stack = save_stack_stack,
+	.address = save_stack_address,
+};
 
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
  */
 void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
-	unsigned long stack = (unsigned long)&stack;
-	int i, nr_stacks = 0, stacks_done[MAX_STACKS];
-
-	WARN_ON(trace->nr_entries || !trace->max_entries);
-
-	if (!task)
-		task = current;
-
-	pr_debug("task: %p, ti: %p\n", task, task->thread_info);
-
-	if (!task || task == current) {
-		/* Grab rbp right from our regs: */
-		asm ("mov %%rbp, %0" : "=r" (stack));
-		pr_debug("rbp:            %p\n", (void *)stack);
-	} else {
-		/* rbp is the last reg pushed by switch_to(): */
-		stack = task->thread.rsp;
-		pr_debug("other task rsp: %p\n", (void *)stack);
-		stack = (unsigned long)(((unsigned long *)stack)[0]);
-		pr_debug("other task rbp: %p\n", (void *)stack);
-	}
-
-	while (1) {
-		unsigned long stack_end = get_stack_end(task, stack);
-
-		pr_debug("stack:          %p\n", (void *)stack);
-		pr_debug("stack end:      %p\n", (void *)stack_end);
-
-		/*
-		 * Invalid stack addres?
-		 */
-		if (!stack_end)
-			return;
-		/*
-		 * Were we in this stack already? (recursion)
-		 */
-		for (i = 0; i < nr_stacks; i++)
-			if (stacks_done[i] == stack_end)
-				return;
-		stacks_done[nr_stacks] = stack_end;
-
-		stack = save_context_stack(trace, stack, stack_end);
-		if (!stack || trace->nr_entries >= trace->max_entries)
-			return;
-		trace->entries[trace->nr_entries++] = ULONG_MAX;
-		if (trace->nr_entries >= trace->max_entries)
-			return;
-		if (++nr_stacks >= MAX_STACKS)
-			return;
-	}
+	dump_trace(task, NULL, NULL, &save_stack_ops, trace);
+	trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
+EXPORT_SYMBOL(save_stack_trace);
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 9ec2b1d5893d..4ac18b02eada 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -45,6 +45,7 @@
 #include <asm/pda.h>
 #include <asm/proto.h>
 #include <asm/nmi.h>
+#include <asm/stacktrace.h>
 
 asmlinkage void divide_error(void);
 asmlinkage void debug(void);
@@ -142,7 +143,7 @@ void printk_address(unsigned long address)
 #endif
 
 static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
-					unsigned *usedp, const char **idp)
+					unsigned *usedp, char **idp)
 {
 	static char ids[][8] = {
 		[DEBUG_STACK - 1] = "#DB",
@@ -234,13 +235,19 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 	return NULL;
 }
 
-static int show_trace_unwind(struct unwind_frame_info *info, void *context)
+struct ops_and_data {
+	struct stacktrace_ops *ops;
+	void *data;
+};
+
+static int dump_trace_unwind(struct unwind_frame_info *info, void *context)
 {
+	struct ops_and_data *oad = (struct ops_and_data *)context;
 	int n = 0;
 
 	while (unwind(info) == 0 && UNW_PC(info)) {
 		n++;
-		printk_address(UNW_PC(info));
+		oad->ops->address(oad->data, UNW_PC(info));
 		if (arch_unw_user_mode(info))
 			break;
 	}
@@ -254,45 +261,51 @@ static int show_trace_unwind(struct unwind_frame_info *info, void *context)
  * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
  */
 
-void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack)
+void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack,
+		struct stacktrace_ops *ops, void *data)
 {
 	const unsigned cpu = safe_smp_processor_id();
 	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
 	unsigned used = 0;
 
-	printk("\nCall Trace:\n");
-
 	if (!tsk)
 		tsk = current;
 
 	if (call_trace >= 0) {
 		int unw_ret = 0;
 		struct unwind_frame_info info;
+		struct ops_and_data oad = { .ops = ops, .data = data };
 
 		if (regs) {
 			if (unwind_init_frame_info(&info, tsk, regs) == 0)
-				unw_ret = show_trace_unwind(&info, NULL);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		} else if (tsk == current)
-			unw_ret = unwind_init_running(&info, show_trace_unwind, NULL);
+			unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad);
 		else {
 			if (unwind_init_blocked(&info, tsk) == 0)
-				unw_ret = show_trace_unwind(&info, NULL);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		}
 		if (unw_ret > 0) {
 			if (call_trace == 1 && !arch_unw_user_mode(&info)) {
-				print_symbol("DWARF2 unwinder stuck at %s\n",
+				ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n",
 					     UNW_PC(&info));
 				if ((long)UNW_SP(&info) < 0) {
-					printk("Leftover inexact backtrace:\n");
+					ops->warning(data, "Leftover inexact backtrace:\n");
 					stack = (unsigned long *)UNW_SP(&info);
 				} else
-					printk("Full inexact backtrace again:\n");
+					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
 				return;
 			else
-				printk("Full inexact backtrace again:\n");
+				ops->warning(data, "Full inexact backtrace again:\n");
 		} else
-			printk("Inexact backtrace:\n");
+			ops->warning(data, "Inexact backtrace:\n");
+	}
+	if (!stack) {
+		unsigned long dummy;
+		stack = &dummy;
+		if (tsk && tsk != current)
+			stack = (unsigned long *)tsk->thread.rsp;
 	}
 
 	/*
@@ -312,7 +325,7 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 			 * down the cause of the crash will be able to figure \
 			 * out the call path that was taken. \
 			 */ \
-			printk_address(addr); \
+			ops->address(data, addr);   \
 		} \
 	} while (0)
 
@@ -321,16 +334,17 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 	 * current stack address. If the stacks consist of nested
 	 * exceptions
 	 */
-	for ( ; ; ) {
-		const char *id;
+	for (;;) {
+		char *id;
 		unsigned long *estack_end;
 		estack_end = in_exception_stack(cpu, (unsigned long)stack,
 						&used, &id);
 
 		if (estack_end) {
-			printk(" <%s>", id);
+			if (ops->stack(data, id) < 0)
+				break;
 			HANDLE_STACK (stack < estack_end);
-			printk(" <EOE>");
+			ops->stack(data, "<EOE>");
 			/*
 			 * We link to the next stack via the
 			 * second-to-last pointer (index -2 to end) in the
@@ -345,7 +359,8 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 				(IRQSTACKSIZE - 64) / sizeof(*irqstack);
 
 			if (stack >= irqstack && stack < irqstack_end) {
-				printk(" <IRQ>");
+				if (ops->stack(data, "IRQ") < 0)
+					break;
 				HANDLE_STACK (stack < irqstack_end);
 				/*
 				 * We link to the next stack (which would be
@@ -354,7 +369,7 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 				 */
 				stack = (unsigned long *) (irqstack_end[-1]);
 				irqstack_end = NULL;
-				printk(" <EOI>");
+				ops->stack(data, "EOI");
 				continue;
 			}
 		}
@@ -362,15 +377,53 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 	}
 
 	/*
-	 * This prints the process stack:
+	 * This handles the process stack:
 	 */
 	HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0);
 #undef HANDLE_STACK
+}
+EXPORT_SYMBOL(dump_trace);
 
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s\n", msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	printk(" <%s> ", name);
+	return 0;
+}
+
+static void print_trace_address(void *data, unsigned long addr)
+{
+	printk_address(addr);
+}
+
+static struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+void
+show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack)
+{
+	printk("\nCall Trace:\n");
+	dump_trace(tsk, regs, stack, &print_trace_ops, NULL);
 	printk("\n");
 }
 
-static void _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long * rsp)
+static void
+_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
 {
 	unsigned long *stack;
 	int i;
diff --git a/include/asm-x86_64/stacktrace.h b/include/asm-x86_64/stacktrace.h
new file mode 100644
index 000000000000..5eb9799bef76
--- /dev/null
+++ b/include/asm-x86_64/stacktrace.h
@@ -0,0 +1,18 @@
+#ifndef _ASM_STACKTRACE_H
+#define _ASM_STACKTRACE_H 1
+
+/* Generic stack tracer with callbacks */
+
+struct stacktrace_ops {
+	void (*warning)(void *data, char *msg);
+	/* msg must contain %s for the symbol */
+	void (*warning_symbol)(void *data, char *msg, unsigned long symbol);
+	void (*address)(void *data, unsigned long address);
+	/* On negative return stop dumping */
+	int (*stack)(void *data, char *name);
+};
+
+void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack,
+		struct stacktrace_ops *ops, void *data);
+
+#endif
-- 
cgit v1.2.3


From 2b14a78cd07a52001b8c3865ed615d8b9b905b78 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: Do stacktracer conversion too

Following x86-64 patches. Reuses code from them in fact.

Convert the standard backtracer to do all output using
callbacks.   Use the x86-64 stack tracer implementation
that uses these callbacks to implement the stacktrace interface.

This allows to use the new dwarf2 unwinder for stacktrace
and get better backtraces.

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/Makefile     |   1 +
 arch/i386/kernel/stacktrace.c |  93 ------------------------------------
 arch/i386/kernel/traps.c      | 108 +++++++++++++++++++++++++++++++-----------
 include/asm-i386/stacktrace.h |   1 +
 4 files changed, 82 insertions(+), 121 deletions(-)
 delete mode 100644 arch/i386/kernel/stacktrace.c
 create mode 100644 include/asm-i386/stacktrace.h

(limited to 'include')

diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index dab497472deb..1a884b6e6e5c 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -81,4 +81,5 @@ $(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
 	$(call if_changed,syscall)
 
 k8-y                      += ../../x86_64/kernel/k8.o
+stacktrace-y		  += ../../x86_64/kernel/stacktrace.o
 
diff --git a/arch/i386/kernel/stacktrace.c b/arch/i386/kernel/stacktrace.c
deleted file mode 100644
index ae3c32a87add..000000000000
--- a/arch/i386/kernel/stacktrace.c
+++ /dev/null
@@ -1,93 +0,0 @@
-/*
- * arch/i386/kernel/stacktrace.c
- *
- * Stack trace management functions
- *
- *  Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
- */
-#include <linux/sched.h>
-#include <linux/stacktrace.h>
-
-static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
-{
-	return	p > (void *)tinfo &&
-		p < (void *)tinfo + THREAD_SIZE - 3;
-}
-
-/*
- * Save stack-backtrace addresses into a stack_trace buffer:
- */
-static inline unsigned long
-save_context_stack(struct stack_trace *trace, unsigned int skip,
-		   struct thread_info *tinfo, unsigned long *stack,
-		   unsigned long ebp)
-{
-	unsigned long addr;
-
-#ifdef CONFIG_FRAME_POINTER
-	while (valid_stack_ptr(tinfo, (void *)ebp)) {
-		addr = *(unsigned long *)(ebp + 4);
-		if (!skip)
-			trace->entries[trace->nr_entries++] = addr;
-		else
-			skip--;
-		if (trace->nr_entries >= trace->max_entries)
-			break;
-		/*
-		 * break out of recursive entries (such as
-		 * end_of_stack_stop_unwind_function):
-	 	 */
-		if (ebp == *(unsigned long *)ebp)
-			break;
-
-		ebp = *(unsigned long *)ebp;
-	}
-#else
-	while (valid_stack_ptr(tinfo, stack)) {
-		addr = *stack++;
-		if (__kernel_text_address(addr)) {
-			if (!skip)
-				trace->entries[trace->nr_entries++] = addr;
-			else
-				skip--;
-			if (trace->nr_entries >= trace->max_entries)
-				break;
-		}
-	}
-#endif
-
-	return ebp;
-}
-
-/*
- * Save stack-backtrace addresses into a stack_trace buffer.
- */
-void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
-{
-	unsigned long ebp;
-	unsigned long *stack = &ebp;
-
-	WARN_ON(trace->nr_entries || !trace->max_entries);
-
-	if (!task || task == current) {
-		/* Grab ebp right from our regs: */
-		asm ("movl %%ebp, %0" : "=r" (ebp));
-	} else {
-		/* ebp is the last reg pushed by switch_to(): */
-		ebp = *(unsigned long *) task->thread.esp;
-	}
-
-	while (1) {
-		struct thread_info *context = (struct thread_info *)
-				((unsigned long)stack & (~(THREAD_SIZE - 1)));
-
-		ebp = save_context_stack(trace, trace->skip, context, stack, ebp);
-		stack = (unsigned long *)context->previous_esp;
-		if (!stack || trace->nr_entries >= trace->max_entries)
-			break;
-		trace->entries[trace->nr_entries++] = ULONG_MAX;
-		if (trace->nr_entries >= trace->max_entries)
-			break;
-	}
-}
-
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 3c85c89f68d8..4ced4285163b 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -51,6 +51,7 @@
 #include <asm/smp.h>
 #include <asm/arch_hooks.h>
 #include <asm/kdebug.h>
+#include <asm/stacktrace.h>
 
 #include <linux/module.h>
 
@@ -118,26 +119,16 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
 		p < (void *)tinfo + THREAD_SIZE - 3;
 }
 
-/*
- * Print one address/symbol entries per line.
- */
-static inline void print_addr_and_symbol(unsigned long addr, char *log_lvl)
-{
-	printk(" [<%08lx>] ", addr);
-
-	print_symbol("%s\n", addr);
-}
-
 static inline unsigned long print_context_stack(struct thread_info *tinfo,
 				unsigned long *stack, unsigned long ebp,
-				char *log_lvl)
+				struct stacktrace_ops *ops, void *data)
 {
 	unsigned long addr;
 
 #ifdef	CONFIG_FRAME_POINTER
 	while (valid_stack_ptr(tinfo, (void *)ebp)) {
 		addr = *(unsigned long *)(ebp + 4);
-		print_addr_and_symbol(addr, log_lvl);
+		ops->address(data, addr);
 		/*
 		 * break out of recursive entries (such as
 		 * end_of_stack_stop_unwind_function):
@@ -150,28 +141,35 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
 	while (valid_stack_ptr(tinfo, stack)) {
 		addr = *stack++;
 		if (__kernel_text_address(addr))
-			print_addr_and_symbol(addr, log_lvl);
+			ops->address(data, addr);
 	}
 #endif
 	return ebp;
 }
 
+struct ops_and_data {
+	struct stacktrace_ops *ops;
+	void *data;
+};
+
 static asmlinkage int
-show_trace_unwind(struct unwind_frame_info *info, void *log_lvl)
+dump_trace_unwind(struct unwind_frame_info *info, void *data)
 {
+	struct ops_and_data *oad = (struct ops_and_data *)data;
 	int n = 0;
 
 	while (unwind(info) == 0 && UNW_PC(info)) {
 		n++;
-		print_addr_and_symbol(UNW_PC(info), log_lvl);
+		oad->ops->address(oad->data, UNW_PC(info));
 		if (arch_unw_user_mode(info))
 			break;
 	}
 	return n;
 }
 
-static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-			       unsigned long *stack, char *log_lvl)
+void dump_trace(struct task_struct *task, struct pt_regs *regs,
+	        unsigned long *stack,
+		struct stacktrace_ops *ops, void *data)
 {
 	unsigned long ebp;
 
@@ -181,31 +179,37 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	if (call_trace >= 0) {
 		int unw_ret = 0;
 		struct unwind_frame_info info;
+		struct ops_and_data oad = { .ops = ops, .data = data };
 
 		if (regs) {
 			if (unwind_init_frame_info(&info, task, regs) == 0)
-				unw_ret = show_trace_unwind(&info, log_lvl);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		} else if (task == current)
-			unw_ret = unwind_init_running(&info, show_trace_unwind, log_lvl);
+			unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad);
 		else {
 			if (unwind_init_blocked(&info, task) == 0)
-				unw_ret = show_trace_unwind(&info, log_lvl);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		}
 		if (unw_ret > 0) {
 			if (call_trace == 1 && !arch_unw_user_mode(&info)) {
-				print_symbol("DWARF2 unwinder stuck at %s\n",
+				ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n",
 					     UNW_PC(&info));
 				if (UNW_SP(&info) >= PAGE_OFFSET) {
-					printk("Leftover inexact backtrace:\n");
+					ops->warning(data, "Leftover inexact backtrace:\n");
 					stack = (void *)UNW_SP(&info);
 				} else
-					printk("Full inexact backtrace again:\n");
+					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
 				return;
 			else
-				printk("Full inexact backtrace again:\n");
+				ops->warning(data, "Full inexact backtrace again:\n");
 		} else
-			printk("Inexact backtrace:\n");
+			ops->warning(data, "Inexact backtrace:\n");
+	} else 	if (!stack) {
+		unsigned long dummy;
+		stack = &dummy;
+		if (task && task != current)
+			stack = (unsigned long *)task->thread.esp;
 	}
 
 	if (task == current) {
@@ -220,15 +224,63 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 		struct thread_info *context;
 		context = (struct thread_info *)
 			((unsigned long)stack & (~(THREAD_SIZE - 1)));
-		ebp = print_context_stack(context, stack, ebp, log_lvl);
+		ebp = print_context_stack(context, stack, ebp, ops, data);
+		/* Should be after the line below, but somewhere
+		   in early boot context comes out corrupted and we
+		   can't reference it -AK */
+		if (ops->stack(data, "IRQ") < 0)
+			break;
 		stack = (unsigned long*)context->previous_esp;
 		if (!stack)
 			break;
-		printk("%s =======================\n", log_lvl);
 	}
 }
+EXPORT_SYMBOL(dump_trace);
+
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	printk(data);
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s%s\n", (char *)data, msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	return 0;
+}
+
+/*
+ * Print one address/symbol entries per line.
+ */
+static void print_trace_address(void *data, unsigned long addr)
+{
+	printk("%s [<%08lx>] ", (char *)data, addr);
+	print_symbol("%s\n", addr);
+}
+
+static struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+static void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		   unsigned long * stack, char *log_lvl)
+{
+	dump_trace(task, regs, stack, &print_trace_ops, log_lvl);
+	printk("%s =======================\n", log_lvl);
+}
 
-void show_trace(struct task_struct *task, struct pt_regs *regs, unsigned long * stack)
+void show_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long * stack)
 {
 	show_trace_log_lvl(task, regs, stack, "");
 }
diff --git a/include/asm-i386/stacktrace.h b/include/asm-i386/stacktrace.h
new file mode 100644
index 000000000000..7d1f6a5cbfca
--- /dev/null
+++ b/include/asm-i386/stacktrace.h
@@ -0,0 +1 @@
+#include <asm-x86_64/stacktrace.h>
-- 
cgit v1.2.3


From a32cf3975bed3b84491f8ffeb24abe8c45d86ab0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: Get ebp from unwinder state when continuing fallback
 backtrace

Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c  | 22 ++++++++++++++--------
 include/asm-i386/unwind.h |  3 +++
 2 files changed, 17 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 86fa7e47f301..bdf949c30c7c 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -171,7 +171,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	        unsigned long *stack,
 		struct stacktrace_ops *ops, void *data)
 {
-	unsigned long ebp;
+	unsigned long ebp = 0;
 
 	if (!task)
 		task = current;
@@ -199,6 +199,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 					stack = (void *)UNW_SP(&info);
 					if (!stack)
 						return;
+					ebp = UNW_FP(&info);
 				} else
 					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
@@ -207,20 +208,25 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 				ops->warning(data, "Full inexact backtrace again:\n");
 		} else
 			ops->warning(data, "Inexact backtrace:\n");
-	} else 	if (!stack) {
+	}
+	if (!stack) {
 		unsigned long dummy;
 		stack = &dummy;
 		if (task && task != current)
 			stack = (unsigned long *)task->thread.esp;
 	}
 
-	if (task == current) {
-		/* Grab ebp right from our regs */
-		asm ("movl %%ebp, %0" : "=r" (ebp) : );
-	} else {
-		/* ebp is the last reg pushed by switch_to */
-		ebp = *(unsigned long *) task->thread.esp;
+#ifdef CONFIG_FRAME_POINTER
+	if (!ebp) {
+		if (task == current) {
+			/* Grab ebp right from our regs */
+			asm ("movl %%ebp, %0" : "=r" (ebp) : );
+		} else {
+			/* ebp is the last reg pushed by switch_to */
+			ebp = *(unsigned long *) task->thread.esp;
+		}
 	}
+#endif
 
 	while (1) {
 		struct thread_info *context;
diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h
index 4c1a0b968569..f0ac399bae3c 100644
--- a/include/asm-i386/unwind.h
+++ b/include/asm-i386/unwind.h
@@ -28,6 +28,8 @@ struct unwind_frame_info
 #define FRAME_LINK_OFFSET    0
 #define STACK_BOTTOM(tsk)    STACK_LIMIT((tsk)->thread.esp0)
 #define STACK_TOP(tsk)       ((tsk)->thread.esp0)
+#else
+#define UNW_FP(frame) ((void)(frame), 0)
 #endif
 #define STACK_LIMIT(ptr)     (((ptr) - 1) & ~(THREAD_SIZE - 1))
 
@@ -88,6 +90,7 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info)
 
 #define UNW_PC(frame) ((void)(frame), 0)
 #define UNW_SP(frame) ((void)(frame), 0)
+#define UNW_FP(frame) ((void)(frame), 0)
 
 static inline int arch_unw_user_mode(const void *info)
 {
-- 
cgit v1.2.3


From d28c4393a7bf558538e9def269c1caeab6ec056f Mon Sep 17 00:00:00 2001
From: "Prasanna S.P" <prasanna@in.ibm.com>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] x86: error_code is not safe for kprobes

This patch moves the entry.S:error_entry to .kprobes.text section,
since code marked unsafe for kprobes jumps directly to entry.S::error_entry,
that must be marked unsafe as well.
This patch also moves all the ".previous.text" asm directives to ".previous"
for kprobes section.

AK: Following a similar i386 patch from Chuck Ebbert
AK: Also merged Jeremy's fix in.

+From: Jeremy Fitzhardinge <jeremy@goop.org>

KPROBE_ENTRY does a .section .kprobes.text, and expects its users to
do a .previous at the end of the function.

Unfortunately, if any code within the function switches sections, for
example .fixup, then the .previous ends up putting all subsequent code
into .fixup.  Worse, any subsequent .fixup code gets intermingled with
the code its supposed to be fixing (which is also in .fixup).  It's
surprising this didn't cause more havok.

The fix is to use .pushsection/.popsection, so this stuff nests
properly.  A further cleanup would be to get rid of all
.section/.previous pairs, since they're inherently fragile.

+From: Chuck Ebbert <76306.1226@compuserve.com>

Because code marked unsafe for kprobes jumps directly to
entry.S::error_code, that must be marked unsafe as well.
The easiest way to do that is to move the page fault entry
point to just before error_code and let it inherit the same
section.

Also moved all the ".previous" asm directives for kprobes
sections to column 1 and removed ".text" from them.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S   | 25 +++++++++++++------------
 arch/x86_64/kernel/entry.S | 19 +++++++------------
 include/linux/linkage.h    |  6 +++++-
 3 files changed, 25 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 87f9f60b803b..ba22ec8fab54 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -591,11 +591,9 @@ ENTRY(name)				\
 /* The include is where all of the SMP etc. interrupts come from */
 #include "entry_arch.h"
 
-ENTRY(divide_error)
-	RING0_INT_FRAME
-	pushl $0			# no error code
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_divide_error
+KPROBE_ENTRY(page_fault)
+	RING0_EC_FRAME
+	pushl $do_page_fault
 	CFI_ADJUST_CFA_OFFSET 4
 	ALIGN
 error_code:
@@ -645,6 +643,7 @@ error_code:
 	call *%edi
 	jmp ret_from_exception
 	CFI_ENDPROC
+KPROBE_END(page_fault)
 
 ENTRY(coprocessor_error)
 	RING0_INT_FRAME
@@ -720,7 +719,8 @@ debug_stack_correct:
 	call do_debug
 	jmp ret_from_exception
 	CFI_ENDPROC
-	.previous .text
+KPROBE_END(debug)
+
 /*
  * NMI is doubly nasty. It can happen _while_ we're handling
  * a debug fault, and the debug fault hasn't yet been able to
@@ -816,7 +816,7 @@ KPROBE_ENTRY(int3)
 	call do_int3
 	jmp ret_from_exception
 	CFI_ENDPROC
-	.previous .text
+KPROBE_END(int3)
 
 ENTRY(overflow)
 	RING0_INT_FRAME
@@ -881,7 +881,7 @@ KPROBE_ENTRY(general_protection)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
-	.previous .text
+KPROBE_END(general_protection)
 
 ENTRY(alignment_check)
 	RING0_EC_FRAME
@@ -890,13 +890,14 @@ ENTRY(alignment_check)
 	jmp error_code
 	CFI_ENDPROC
 
-KPROBE_ENTRY(page_fault)
-	RING0_EC_FRAME
-	pushl $do_page_fault
+ENTRY(divide_error)
+	RING0_INT_FRAME
+	pushl $0			# no error code
+	CFI_ADJUST_CFA_OFFSET 4
+	pushl $do_divide_error
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
-	.previous .text
 
 #ifdef CONFIG_X86_MCE
 ENTRY(machine_check)
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 2092f565aa87..780f9b26169f 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -819,7 +819,7 @@ paranoid_schedule\trace:
  * Exception entry point. This expects an error code/orig_rax on the stack
  * and the exception handler in %rax.	
  */ 		  				
-ENTRY(error_entry)
+KPROBE_ENTRY(error_entry)
 	_frame RDI
 	/* rdi slot contains rax, oldrax contains error code */
 	cld	
@@ -903,7 +903,7 @@ error_kernelspace:
 	cmpq $gs_change,RIP(%rsp)
         je   error_swapgs
 	jmp  error_sti
-END(error_entry)
+KPROBE_END(error_entry)
 	
        /* Reload gs selector with exception handling */
        /* edi:  new selector */ 
@@ -1025,8 +1025,7 @@ ENDPROC(execve)
 
 KPROBE_ENTRY(page_fault)
 	errorentry do_page_fault
-END(page_fault)
-	.previous .text
+KPROBE_END(page_fault)
 
 ENTRY(coprocessor_error)
 	zeroentry do_coprocessor_error
@@ -1047,8 +1046,7 @@ KPROBE_ENTRY(debug)
 	CFI_ADJUST_CFA_OFFSET 8		
 	paranoidentry do_debug, DEBUG_STACK
 	paranoidexit
-END(debug)
-	.previous .text
+KPROBE_END(debug)
 
 	/* runs on exception stack */	
 KPROBE_ENTRY(nmi)
@@ -1062,8 +1060,7 @@ KPROBE_ENTRY(nmi)
 	jmp paranoid_exit1
  	CFI_ENDPROC
 #endif
-END(nmi)
-	.previous .text
+KPROBE_END(nmi)
 
 KPROBE_ENTRY(int3)
  	INTR_FRAME
@@ -1072,8 +1069,7 @@ KPROBE_ENTRY(int3)
  	paranoidentry do_int3, DEBUG_STACK
  	jmp paranoid_exit1
  	CFI_ENDPROC
-END(int3)
-	.previous .text
+KPROBE_END(int3)
 
 ENTRY(overflow)
 	zeroentry do_overflow
@@ -1121,8 +1117,7 @@ END(stack_segment)
 
 KPROBE_ENTRY(general_protection)
 	errorentry do_general_protection
-END(general_protection)
-	.previous .text
+KPROBE_END(general_protection)
 
 ENTRY(alignment_check)
 	errorentry do_alignment_check
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 932021f872d5..6c9873f88287 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -35,9 +35,13 @@
 #endif
 
 #define KPROBE_ENTRY(name) \
-  .section .kprobes.text, "ax"; \
+  .pushsection .kprobes.text, "ax"; \
   ENTRY(name)
 
+#define KPROBE_END(name) \
+  END(name);		 \
+  .popsection
+
 #ifndef END
 #define END(name) \
   .size name, .-name
-- 
cgit v1.2.3


From 3d08a256da8aed5300bd0752200ece426f49b050 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Make enable_local_apic static

enable_local_apic can now become static.

Cc: len.brown@intel.com

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/apic.c | 13 ++++++++++++-
 include/asm-i386/apic.h | 12 ------------
 2 files changed, 12 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index ce75e71b694f..90faae5c5d30 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -52,7 +52,18 @@ static cpumask_t timer_bcast_ipi;
 /*
  * Knob to control our willingness to enable the local APIC.
  */
-int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
+static int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
+
+static inline void lapic_disable(void)
+{
+	enable_local_apic = -1;
+	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+}
+
+static inline void lapic_enable(void)
+{
+	enable_local_apic = 1;
+}
 
 /*
  * Debug level
diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h
index 8c5331ceca7b..3a42b7d6fc92 100644
--- a/include/asm-i386/apic.h
+++ b/include/asm-i386/apic.h
@@ -16,20 +16,8 @@
 #define APIC_VERBOSE 1
 #define APIC_DEBUG   2
 
-extern int enable_local_apic;
 extern int apic_verbosity;
 
-static inline void lapic_disable(void)
-{
-	enable_local_apic = -1;
-	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
-}
-
-static inline void lapic_enable(void)
-{
-	enable_local_apic = 1;
-}
-
 /*
  * Define the default level of output to be very little
  * This can be turned up by using apic=verbose for more
-- 
cgit v1.2.3


From 522e93e3fcdbf00ba85c72fde6df28cfc0486a65 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Descriptor and trap table cleanups.

The implementation comes from Zach's [RFC, PATCH 10/24] i386 Vmi
descriptor changes:

Descriptor and trap table cleanups.  Add cleanly written accessors for
IDT and GDT gates so the subarch may override them.  Note that this
allows the hypervisor to transparently tweak the DPL of the descriptors
as well as the RPL of segments in those descriptors, with no unnecessary
kernel code modification.  It also allows the hypervisor implementation
of the VMI to tweak the gates, allowing for custom exception frames or
extra layers of indirection above the guest fault / IRQ handlers.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c |  24 ++--------
 include/asm-i386/desc.h  | 121 +++++++++++++++++++++++++++++------------------
 2 files changed, 81 insertions(+), 64 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index bdf949c30c7c..00d643f3de41 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -1165,20 +1165,6 @@ void __init trap_init_f00f_bug(void)
 }
 #endif
 
-#define _set_gate(gate_addr,type,dpl,addr,seg) \
-do { \
-  int __d0, __d1; \
-  __asm__ __volatile__ ("movw %%dx,%%ax\n\t" \
-	"movw %4,%%dx\n\t" \
-	"movl %%eax,%0\n\t" \
-	"movl %%edx,%1" \
-	:"=m" (*((long *) (gate_addr))), \
-	 "=m" (*(1+(long *) (gate_addr))), "=&a" (__d0), "=&d" (__d1) \
-	:"i" ((short) (0x8000+(dpl<<13)+(type<<8))), \
-	 "3" ((char *) (addr)),"2" ((seg) << 16)); \
-} while (0)
-
-
 /*
  * This needs to use 'idt_table' rather than 'idt', and
  * thus use the _nonmapped_ version of the IDT, as the
@@ -1187,7 +1173,7 @@ do { \
  */
 void set_intr_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n,14,0,addr,__KERNEL_CS);
+	_set_gate(n, DESCTYPE_INT, addr, __KERNEL_CS);
 }
 
 /*
@@ -1195,22 +1181,22 @@ void set_intr_gate(unsigned int n, void *addr)
  */
 static inline void set_system_intr_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n, 14, 3, addr, __KERNEL_CS);
+	_set_gate(n, DESCTYPE_INT | DESCTYPE_DPL3, addr, __KERNEL_CS);
 }
 
 static void __init set_trap_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n,15,0,addr,__KERNEL_CS);
+	_set_gate(n, DESCTYPE_TRAP, addr, __KERNEL_CS);
 }
 
 static void __init set_system_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n,15,3,addr,__KERNEL_CS);
+	_set_gate(n, DESCTYPE_TRAP | DESCTYPE_DPL3, addr, __KERNEL_CS);
 }
 
 static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
 {
-	_set_gate(idt_table+n,5,0,0,(gdt_entry<<3));
+	_set_gate(n, DESCTYPE_TASK, (void *)0, (gdt_entry<<3));
 }
 
 
diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 89b8b82c82b3..5db9e96e8dc1 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -33,50 +33,99 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 	return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address;
 }
 
+/*
+ * This is the ldt that every process will get unless we need
+ * something other than this.
+ */
+extern struct desc_struct default_ldt[];
+extern struct desc_struct idt_table[];
+extern void set_intr_gate(unsigned int irq, void * addr);
+
+static inline void pack_descriptor(__u32 *a, __u32 *b,
+	unsigned long base, unsigned long limit, unsigned char type, unsigned char flags)
+{
+	*a = ((base & 0xffff) << 16) | (limit & 0xffff);
+	*b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
+	     ((type & 0xff) << 8) | ((flags & 0xf) << 12);
+}
+
+static inline void pack_gate(__u32 *a, __u32 *b,
+	unsigned long base, unsigned short seg, unsigned char type, unsigned char flags)
+{
+	*a = (seg << 16) | (base & 0xffff);
+	*b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
+}
+
+#define DESCTYPE_LDT 	0x82	/* present, system, DPL-0, LDT */
+#define DESCTYPE_TSS 	0x89	/* present, system, DPL-0, 32-bit TSS */
+#define DESCTYPE_TASK	0x85	/* present, system, DPL-0, task gate */
+#define DESCTYPE_INT	0x8e	/* present, system, DPL-0, interrupt gate */
+#define DESCTYPE_TRAP	0x8f	/* present, system, DPL-0, trap gate */
+#define DESCTYPE_DPL3	0x60	/* DPL-3 */
+#define DESCTYPE_S	0x10	/* !system */
+
 #define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8))
 #define load_LDT_desc() __asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8))
 
 #define load_gdt(dtr) __asm__ __volatile("lgdt %0"::"m" (*dtr))
 #define load_idt(dtr) __asm__ __volatile("lidt %0"::"m" (*dtr))
-#define load_tr(tr) __asm__ __volatile("ltr %0"::"mr" (tr))
-#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"mr" (ldt))
+#define load_tr(tr) __asm__ __volatile("ltr %0"::"m" (tr))
+#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"m" (ldt))
 
 #define store_gdt(dtr) __asm__ ("sgdt %0":"=m" (*dtr))
 #define store_idt(dtr) __asm__ ("sidt %0":"=m" (*dtr))
-#define store_tr(tr) __asm__ ("str %0":"=mr" (tr))
-#define store_ldt(ldt) __asm__ ("sldt %0":"=mr" (ldt))
+#define store_tr(tr) __asm__ ("str %0":"=m" (tr))
+#define store_ldt(ldt) __asm__ ("sldt %0":"=m" (ldt))
 
-/*
- * This is the ldt that every process will get unless we need
- * something other than this.
- */
-extern struct desc_struct default_ldt[];
-extern void set_intr_gate(unsigned int irq, void * addr);
+#if TLS_SIZE != 24
+# error update this code.
+#endif
 
-#define _set_tssldt_desc(n,addr,limit,type) \
-__asm__ __volatile__ ("movw %w3,0(%2)\n\t" \
-	"movw %w1,2(%2)\n\t" \
-	"rorl $16,%1\n\t" \
-	"movb %b1,4(%2)\n\t" \
-	"movb %4,5(%2)\n\t" \
-	"movb $0,6(%2)\n\t" \
-	"movb %h1,7(%2)\n\t" \
-	"rorl $16,%1" \
-	: "=m"(*(n)) : "q" (addr), "r"(n), "ir"(limit), "i"(type))
-
-static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, void *addr)
+static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 {
-	_set_tssldt_desc(&get_cpu_gdt_table(cpu)[entry], (int)addr,
-		offsetof(struct tss_struct, __cacheline_filler) - 1, 0x89);
+#define C(i) get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]
+	C(0); C(1); C(2);
+#undef C
 }
 
-#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
+static inline void write_dt_entry(void *dt, int entry, __u32 entry_a, __u32 entry_b)
+{
+	__u32 *lp = (__u32 *)((char *)dt + entry*8);
+	*lp = entry_a;
+	*(lp+1) = entry_b;
+}
+
+#define write_ldt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_gdt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_idt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+
+static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg)
+{
+	__u32 a, b;
+	pack_gate(&a, &b, (unsigned long)addr, seg, type, 0);
+	write_idt_entry(idt_table, gate, a, b);
+}
 
-static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int size)
+static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
 {
-	_set_tssldt_desc(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT], (int)addr, ((size << 3)-1), 0x82);
+	__u32 a, b;
+	pack_descriptor(&a, &b, (unsigned long)addr,
+			offsetof(struct tss_struct, __cacheline_filler) - 1,
+			DESCTYPE_TSS, 0);
+	write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b);
 }
 
+static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int entries)
+{
+	__u32 a, b;
+	pack_descriptor(&a, &b, (unsigned long)addr,
+			entries * sizeof(struct desc_struct) - 1,
+			DESCTYPE_LDT, 0);
+	write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, a, b);
+}
+
+#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
+
 #define LDT_entry_a(info) \
 	((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
 
@@ -102,24 +151,6 @@ static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int size)
 	(info)->seg_not_present	== 1	&& \
 	(info)->useable		== 0	)
 
-static inline void write_ldt_entry(void *ldt, int entry, __u32 entry_a, __u32 entry_b)
-{
-	__u32 *lp = (__u32 *)((char *)ldt + entry*8);
-	*lp = entry_a;
-	*(lp+1) = entry_b;
-}
-
-#if TLS_SIZE != 24
-# error update this code.
-#endif
-
-static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
-{
-#define C(i) get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]
-	C(0); C(1); C(2);
-#undef C
-}
-
 static inline void clear_LDT(void)
 {
 	int cpu = get_cpu();
-- 
cgit v1.2.3


From ba4d40bb5c465f0a4dcc30d02dab80c2cb7e1ff3 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] Auto size the per cpu area.

Now for a completely different but trivial approach.
I just boot tested it with 255 CPUS and everything worked.

Currently everything (except module data) we place in
the per cpu area we know about at compile time.  So
instead of allocating a fixed size for the per_cpu area
allocate the number of bytes we need plus a fixed constant
for to be used for modules.

It isn't perfect but it is much less of a pain to
work with than what we are doing now.

AK: fixed warning

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c |  7 ++-----
 include/asm-x86_64/percpu.h  | 10 ++++++++++
 2 files changed, 12 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index b09e60fa96b4..e85cfbb49b63 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -95,12 +95,9 @@ void __init setup_per_cpu_areas(void)
 #endif
 
 	/* Copy section for each CPU (we discard the original) */
-	size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
-#ifdef CONFIG_MODULES
-	if (size < PERCPU_ENOUGH_ROOM)
-		size = PERCPU_ENOUGH_ROOM;
-#endif
+	size = PERCPU_ENOUGH_ROOM;
 
+	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n", size);
 	for_each_cpu_mask (i, cpu_possible_map) {
 		char *ptr;
 
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 08dd9f9dda81..39d2bab9b520 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -11,6 +11,16 @@
 
 #include <asm/pda.h>
 
+#ifdef CONFIG_MODULES
+# define PERCPU_MODULE_RESERVE 8192
+#else
+# define PERCPU_MODULE_RESERVE 0
+#endif
+
+#define PERCPU_ENOUGH_ROOM \
+	(ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES) + \
+	 PERCPU_MODULE_RESERVE)
+
 #define __per_cpu_offset(cpu) (cpu_pda(cpu)->data_offset)
 #define __my_cpu_offset() read_pda(data_offset)
 
-- 
cgit v1.2.3


From 73fea175303926055440c06bc8894f0c5c58afc8 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Support physical cpu hotplug for x86_64

This patch enables ACPI based physical CPU hotplug support for x86_64.
Implements acpi_map_lsapic() and acpi_unmap_lsapic() to support physical cpu
hotplug.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/i386/kernel/acpi/boot.c | 69 +++++++++++++++++++++++++++++++++++++++++---
 arch/i386/kernel/mpparse.c   |  2 +-
 include/asm-i386/smp.h       |  1 +
 3 files changed, 67 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 0fba420d668a..8a6c5b412348 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -26,6 +26,7 @@
 #include <linux/init.h>
 #include <linux/acpi.h>
 #include <linux/efi.h>
+#include <linux/cpumask.h>
 #include <linux/module.h>
 #include <linux/dmi.h>
 #include <linux/irq.h>
@@ -512,16 +513,76 @@ EXPORT_SYMBOL(acpi_register_gsi);
 #ifdef CONFIG_ACPI_HOTPLUG_CPU
 int acpi_map_lsapic(acpi_handle handle, int *pcpu)
 {
-	/* TBD */
-	return -EINVAL;
+	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
+	union acpi_object *obj;
+	struct acpi_table_lapic *lapic;
+	cpumask_t tmp_map, new_map;
+	u8 physid;
+	int cpu;
+
+	if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer)))
+		return -EINVAL;
+
+	if (!buffer.length || !buffer.pointer)
+		return -EINVAL;
+
+	obj = buffer.pointer;
+	if (obj->type != ACPI_TYPE_BUFFER ||
+	    obj->buffer.length < sizeof(*lapic)) {
+		kfree(buffer.pointer);
+		return -EINVAL;
+	}
+
+	lapic = (struct acpi_table_lapic *)obj->buffer.pointer;
+
+	if ((lapic->header.type != ACPI_MADT_LAPIC) ||
+	    (!lapic->flags.enabled)) {
+		kfree(buffer.pointer);
+		return -EINVAL;
+	}
+
+	physid = lapic->id;
+
+	kfree(buffer.pointer);
+	buffer.length = ACPI_ALLOCATE_BUFFER;
+	buffer.pointer = NULL;
+
+	tmp_map = cpu_present_map;
+	mp_register_lapic(physid, lapic->flags.enabled);
+
+	/*
+	 * If mp_register_lapic successfully generates a new logical cpu
+	 * number, then the following will get us exactly what was mapped
+	 */
+	cpus_andnot(new_map, cpu_present_map, tmp_map);
+	if (cpus_empty(new_map)) {
+		printk ("Unable to map lapic to logical cpu number\n");
+		return -EINVAL;
+	}
+
+	cpu = first_cpu(new_map);
+
+	*pcpu = cpu;
+	return 0;
 }
 
 EXPORT_SYMBOL(acpi_map_lsapic);
 
 int acpi_unmap_lsapic(int cpu)
 {
-	/* TBD */
-	return -EINVAL;
+	int i;
+
+	for_each_possible_cpu(i) {
+		if (x86_acpiid_to_apicid[i] == x86_cpu_to_apicid[cpu]) {
+			x86_acpiid_to_apicid[i] = -1;
+			break;
+		}
+	}
+	x86_cpu_to_apicid[cpu] = -1;
+	cpu_clear(cpu, cpu_present_map);
+	num_processors--;
+
+	return (0);
 }
 
 EXPORT_SYMBOL(acpi_unmap_lsapic);
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 772a4233bfe6..442aaf8c77eb 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -69,7 +69,7 @@ unsigned int def_to_bigsmp = 0;
 /* Processor that is doing the boot up */
 unsigned int boot_cpu_physical_apicid = -1U;
 /* Internal processor count */
-static unsigned int __devinitdata num_processors;
+unsigned int __cpuinitdata num_processors;
 
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map;
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index f87826039a51..32ac8c91d5c5 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -84,6 +84,7 @@ static inline int hard_smp_processor_id(void)
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
+extern unsigned int num_processors;
 
 #endif /* !__ASSEMBLY__ */
 
-- 
cgit v1.2.3


From e07e23e1fd3000289fc7ccc6c71879070d3b19e0 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] non lazy "sleazy" fpu implementation

Right now the kernel on x86-64 has a 100% lazy fpu behavior: after *every*
context switch a trap is taken for the first FPU use to restore the FPU
context lazily.  This is of course great for applications that have very
sporadic or no FPU use (since then you avoid doing the expensive
save/restore all the time).  However for very frequent FPU users...  you
take an extra trap every context switch.

The patch below adds a simple heuristic to this code: After 5 consecutive
context switches of FPU use, the lazy behavior is disabled and the context
gets restored every context switch.  If the app indeed uses the FPU, the
trap is avoided.  (the chance of the 6th time slice using FPU after the
previous 5 having done so are quite high obviously).

After 256 switches, this is reset and lazy behavior is returned (until
there are 5 consecutive ones again).  The reason for this is to give apps
that do longer bursts of FPU use still the lazy behavior back after some
time.

[akpm@osdl.org: place new task_struct field next to jit_keyring to save space]
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/x86_64/kernel/process.c | 10 ++++++++++
 arch/x86_64/kernel/traps.c   |  1 +
 include/asm-x86_64/i387.h    |  5 ++++-
 include/linux/sched.h        |  9 +++++++++
 4 files changed, 24 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 6fbd19564e4e..9e9a70e50c72 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -552,6 +552,10 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	int cpu = smp_processor_id();  
 	struct tss_struct *tss = &per_cpu(init_tss, cpu);
 
+	/* we're going to use this soon, after a few expensive things */
+	if (next_p->fpu_counter>5)
+		prefetch(&next->i387.fxsave);
+
 	/*
 	 * Reload esp0, LDT and the page table pointer:
 	 */
@@ -629,6 +633,12 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	    || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))
 		__switch_to_xtra(prev_p, next_p, tss);
 
+	/* If the task has used fpu the last 5 timeslices, just do a full
+	 * restore of the math state immediately to avoid the trap; the
+	 * chances of needing FPU soon are obviously high now
+	 */
+	if (next_p->fpu_counter>5)
+		math_state_restore();
 	return prev_p;
 }
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 28e53342f294..ffc40cff1e07 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -1136,6 +1136,7 @@ asmlinkage void math_state_restore(void)
 		init_fpu(me);
 	restore_fpu_checking(&me->thread.i387.fxsave);
 	task_thread_info(me)->status |= TS_USEDFPU;
+	me->fpu_counter++;
 }
 
 void __init trap_init(void)
diff --git a/include/asm-x86_64/i387.h b/include/asm-x86_64/i387.h
index cba8a3b0cded..60c0f4853fdb 100644
--- a/include/asm-x86_64/i387.h
+++ b/include/asm-x86_64/i387.h
@@ -24,6 +24,7 @@ extern unsigned int mxcsr_feature_mask;
 extern void mxcsr_feature_mask_init(void);
 extern void init_fpu(struct task_struct *child);
 extern int save_i387(struct _fpstate __user *buf);
+extern asmlinkage void math_state_restore(void);
 
 /*
  * FPU lazy state save handling...
@@ -31,7 +32,9 @@ extern int save_i387(struct _fpstate __user *buf);
 
 #define unlazy_fpu(tsk) do { \
 	if (task_thread_info(tsk)->status & TS_USEDFPU) \
-		save_init_fpu(tsk); \
+		save_init_fpu(tsk); 			\
+	else						\
+		tsk->fpu_counter = 0;			\
 } while (0)
 
 /* Ignore delayed exceptions from user space */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 34ed0d99b1bd..807556c5bcd2 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -865,6 +865,15 @@ struct task_struct {
 	struct key *thread_keyring;	/* keyring private to this thread */
 	unsigned char jit_keyring;	/* default keyring to attach requested keys to */
 #endif
+	/*
+	 * fpu_counter contains the number of consecutive context switches
+	 * that the FPU is used. If this is over a threshold, the lazy fpu
+	 * saving becomes unlazy to save the trap. This is an unsigned char
+	 * so that after 256 times the counter wraps and the behavior turns
+	 * lazy again; this to deal with bursty apps that only use FPU for
+	 * a short time
+	 */
+	unsigned char fpu_counter;
 	int oomkilladj; /* OOM kill score adjustment (bit shift). */
 	char comm[TASK_COMM_LEN]; /* executable name excluding path
 				     - access with [gs]et_task_comm (which lock
-- 
cgit v1.2.3


From f704cb935006580db0495e54d3c82631f6e2a984 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] x86: remove config.h includes from asm-i386 & asm-x86_64

This is now automatically included by kbuild.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/dwarf2.h    | 2 --
 include/asm-i386/tsc.h       | 1 -
 include/asm-x86_64/calgary.h | 1 -
 3 files changed, 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h
index 2280f6272f80..fe2392f361ee 100644
--- a/include/asm-i386/dwarf2.h
+++ b/include/asm-i386/dwarf2.h
@@ -1,8 +1,6 @@
 #ifndef _DWARF2_H
 #define _DWARF2_H
 
-#include <linux/config.h>
-
 #ifndef __ASSEMBLY__
 #warning "asm/dwarf2.h should be only included in pure assembly files"
 #endif
diff --git a/include/asm-i386/tsc.h b/include/asm-i386/tsc.h
index 97b828ce31e0..c13933185c1c 100644
--- a/include/asm-i386/tsc.h
+++ b/include/asm-i386/tsc.h
@@ -6,7 +6,6 @@
 #ifndef _ASM_i386_TSC_H
 #define _ASM_i386_TSC_H
 
-#include <linux/config.h>
 #include <asm/processor.h>
 
 /*
diff --git a/include/asm-x86_64/calgary.h b/include/asm-x86_64/calgary.h
index 0a03bda94d03..6b93f5a3a5c8 100644
--- a/include/asm-x86_64/calgary.h
+++ b/include/asm-x86_64/calgary.h
@@ -24,7 +24,6 @@
 #ifndef _ASM_X86_64_CALGARY_H
 #define _ASM_X86_64_CALGARY_H
 
-#include <linux/config.h>
 #include <linux/spinlock.h>
 #include <linux/device.h>
 #include <linux/dma-mapping.h>
-- 
cgit v1.2.3


From 1bb4996bcebca1cde49d964b4e012699ce180e61 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Move compiler check for modules to ia64 only

Apparently IA64 needs it, but i386/x86-64 don't anymore
since gcc 2.95 support was dropped.  Nobody else on linux-arch
requested keeping it generically

Cc: tony.luck@intel.com
Cc: kaos@sgi.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-ia64/module.h | 3 ++-
 include/linux/vermagic.h  | 4 ++--
 2 files changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-ia64/module.h b/include/asm-ia64/module.h
index 85c82bd819f2..d2da61e4c49b 100644
--- a/include/asm-ia64/module.h
+++ b/include/asm-ia64/module.h
@@ -28,7 +28,8 @@ struct mod_arch_specific {
 #define Elf_Ehdr	Elf64_Ehdr
 
 #define MODULE_PROC_FAMILY	"ia64"
-#define MODULE_ARCH_VERMAGIC	MODULE_PROC_FAMILY
+#define MODULE_ARCH_VERMAGIC	MODULE_PROC_FAMILY \
+	"gcc-" __stringify(__GNUC__) "." __stringify(__GNUC_MINOR__)
 
 #define ARCH_SHF_SMALL	SHF_IA_64_SHORT
 
diff --git a/include/linux/vermagic.h b/include/linux/vermagic.h
index 46919f9f5eb3..4d0909e53595 100644
--- a/include/linux/vermagic.h
+++ b/include/linux/vermagic.h
@@ -24,5 +24,5 @@
 #define VERMAGIC_STRING 						\
 	UTS_RELEASE " "							\
 	MODULE_VERMAGIC_SMP MODULE_VERMAGIC_PREEMPT 			\
-	MODULE_VERMAGIC_MODULE_UNLOAD MODULE_ARCH_VERMAGIC 		\
-	"gcc-" __stringify(__GNUC__) "." __stringify(__GNUC_MINOR__)
+	MODULE_VERMAGIC_MODULE_UNLOAD MODULE_ARCH_VERMAGIC
+
-- 
cgit v1.2.3


From a549b86dd0f3cbffcd5f9343f4ae7fcd59f7e756 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] i386: annotate FIX_STACK() and the rest of nmi()

In i386's entry.S, FIX_STACK() needs annotation because it
replaces the stack pointer.  And the rest of nmi() needs
annotation in order to compile with these new annotations.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S  | 18 +++++++++++++++---
 include/asm-i386/dwarf2.h |  2 ++
 2 files changed, 17 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 0928f70639aa..4b0845249222 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -701,9 +701,15 @@ device_not_available_emulate:
 	jne ok;					\
 label:						\
 	movl TSS_sysenter_esp0+offset(%esp),%esp;	\
+	CFI_DEF_CFA esp, 0;			\
+	CFI_UNDEFINED eip;			\
 	pushfl;					\
+	CFI_ADJUST_CFA_OFFSET 4;		\
 	pushl $__KERNEL_CS;			\
-	pushl $sysenter_past_esp
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	pushl $sysenter_past_esp;		\
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	CFI_REL_OFFSET eip, 0
 
 KPROBE_ENTRY(debug)
 	RING0_INT_FRAME
@@ -754,6 +760,7 @@ KPROBE_ENTRY(nmi)
 	cmpl $sysenter_entry,12(%esp)
 	je nmi_debug_stack_check
 nmi_stack_correct:
+	/* We have a RING0_INT_FRAME here */
 	pushl %eax
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
@@ -764,9 +771,12 @@ nmi_stack_correct:
 	CFI_ENDPROC
 
 nmi_stack_fixup:
+	RING0_INT_FRAME
 	FIX_STACK(12,nmi_stack_correct, 1)
 	jmp nmi_stack_correct
+
 nmi_debug_stack_check:
+	/* We have a RING0_INT_FRAME here */
 	cmpw $__KERNEL_CS,16(%esp)
 	jne nmi_stack_correct
 	cmpl $debug,(%esp)
@@ -777,8 +787,10 @@ nmi_debug_stack_check:
 	jmp nmi_stack_correct
 
 nmi_16bit_stack:
-	RING0_INT_FRAME
-	/* create the pointer to lss back */
+	/* We have a RING0_INT_FRAME here.
+	 *
+	 * create the pointer to lss back
+	 */
 	pushl %ss
 	CFI_ADJUST_CFA_OFFSET 4
 	pushl %esp
diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h
index fe2392f361ee..5d1a8db5a9b0 100644
--- a/include/asm-i386/dwarf2.h
+++ b/include/asm-i386/dwarf2.h
@@ -26,6 +26,7 @@
 #define CFI_RESTORE .cfi_restore
 #define CFI_REMEMBER_STATE .cfi_remember_state
 #define CFI_RESTORE_STATE .cfi_restore_state
+#define CFI_UNDEFINED .cfi_undefined
 
 #else
 
@@ -46,6 +47,7 @@
 #define CFI_RESTORE	ignore
 #define CFI_REMEMBER_STATE ignore
 #define CFI_RESTORE_STATE ignore
+#define CFI_UNDEFINED ignore
 
 #endif
 
-- 
cgit v1.2.3


From 151f8cc1169f9052095b2be36183ab132d75c6c2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove safe_smp_processor_id()

And replace all users with ordinary smp_processor_id.  The function
was originally added to get some basic oops information out even
if the GS register was corrupted. However that didn't
work for some anymore because printk is needed to print the oops
and it uses smp_processor_id() already. Also GS register corruptions
are not particularly common anymore.

This also helps the Xen port which would otherwise need to
do this in a special way because it can't access the local APIC.

Cc: Chris Wright <chrisw@sous-sol.org>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mce.c   |  2 +-
 arch/x86_64/kernel/smp.c   | 23 -----------------------
 arch/x86_64/kernel/traps.c | 10 +++++-----
 include/asm-x86_64/smp.h   |  2 --
 4 files changed, 6 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 4e017fb30fb3..1a93c3738404 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -182,7 +182,7 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 		goto out2;
 
 	memset(&m, 0, sizeof(struct mce));
-	m.cpu = safe_smp_processor_id();
+	m.cpu = smp_processor_id();
 	rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus);
 	if (!(m.mcgstatus & MCG_STATUS_RIPV))
 		kill_it = 1;
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 2df988bbf312..4f67697f5036 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -522,26 +522,3 @@ asmlinkage void smp_call_function_interrupt(void)
 	}
 }
 
-int safe_smp_processor_id(void)
-{
-	unsigned apicid, i;
-
-	if (disable_apic || !apic_mapped)
-		return 0;
-
-	apicid = hard_smp_processor_id();
-	if (apicid < NR_CPUS && x86_cpu_to_apicid[apicid] == apicid)
-		return apicid;
-
-	for (i = 0; i < NR_CPUS; ++i) {
-		if (x86_cpu_to_apicid[i] == apicid)
-			return i;
-	}
-
-	/* No entries in x86_cpu_to_apicid?  Either no MPS|ACPI,
-	 * or called too early.  Either way, we must be CPU 0. */
-      	if (x86_cpu_to_apicid[0] == BAD_APICID)
-		return 0;
-
-	return 0; /* Should not happen */
-}
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 34660b1e2720..38bc821e457b 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -264,7 +264,7 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context)
 void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack,
 		struct stacktrace_ops *ops, void *data)
 {
-	const unsigned cpu = safe_smp_processor_id();
+	const unsigned cpu = smp_processor_id();
 	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
 	unsigned used = 0;
 
@@ -429,7 +429,7 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
 {
 	unsigned long *stack;
 	int i;
-	const int cpu = safe_smp_processor_id();
+	const int cpu = smp_processor_id();
 	unsigned long *irqstack_end = (unsigned long *) (cpu_pda(cpu)->irqstackptr);
 	unsigned long *irqstack = (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
 
@@ -483,7 +483,7 @@ void show_registers(struct pt_regs *regs)
 	int i;
 	int in_kernel = !user_mode(regs);
 	unsigned long rsp;
-	const int cpu = safe_smp_processor_id(); 
+	const int cpu = smp_processor_id();
 	struct task_struct *cur = cpu_pda(cpu)->pcurrent;
 
 		rsp = regs->rsp;
@@ -558,7 +558,7 @@ static unsigned int die_nest_count;
 
 unsigned __kprobes long oops_begin(void)
 {
-	int cpu = safe_smp_processor_id();
+	int cpu = smp_processor_id();
 	unsigned long flags;
 
 	oops_enter();
@@ -636,7 +636,7 @@ void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
 	 * We are in trouble anyway, lets at least try
 	 * to get a message out.
 	 */
-	printk(str, safe_smp_processor_id());
+	printk(str, smp_processor_id());
 	show_registers(regs);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 498fbc1fc179..58b5d6149a42 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -66,7 +66,6 @@ static inline int hard_smp_processor_id(void)
 	return GET_APIC_ID(*(unsigned int *)(APIC_BASE+APIC_ID));
 }
 
-extern int safe_smp_processor_id(void);
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 extern void prefill_possible_map(void);
@@ -100,7 +99,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 
 #ifndef CONFIG_SMP
 #define stack_smp_processor_id() 0
-#define safe_smp_processor_id() 0
 #define cpu_logical_map(x) (x)
 #else
 #include <asm/thread_info.h>
-- 
cgit v1.2.3


From df992848f5aa803fcacd2c5e7d67034bb89e3fa3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Fix pte_exec/mkexec and use it in change_page_attr()

Fix the pte_exec/mkexec page table accessor functions to really
use the NX bit. Previously they only checked the USER bit, but
weren't actually used for anything.

Then use them in change_page_attr() to manipulate the NX bit
properly.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/pageattr.c    | 8 +++++---
 include/asm-x86_64/pgtable.h | 4 ++--
 2 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c
index 2685b1f3671c..01591b649f4e 100644
--- a/arch/x86_64/mm/pageattr.c
+++ b/arch/x86_64/mm/pageattr.c
@@ -190,10 +190,12 @@ int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
 		 * lowmem */
 		if (__pa(address) < KERNEL_TEXT_SIZE) {
 			unsigned long addr2;
-			pgprot_t prot2 = prot;
+			pgprot_t prot2;
 			addr2 = __START_KERNEL_map + __pa(address);
- 			pgprot_val(prot2) &= ~_PAGE_NX;
-			err = __change_page_attr(addr2, pfn, prot2, PAGE_KERNEL_EXEC);
+			/* Make sure the kernel mappings stay executable */
+			prot2 = pte_pgprot(pte_mkexec(pfn_pte(0, prot)));
+			err = __change_page_attr(addr2, pfn, prot2,
+						 PAGE_KERNEL_EXEC);
 		} 
 	} 	
 	up_write(&init_mm.mmap_sem); 
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index d79e7441b513..f7614670c655 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -265,7 +265,7 @@ static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
 #define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
 static inline int pte_user(pte_t pte)		{ return pte_val(pte) & _PAGE_USER; }
 static inline int pte_read(pte_t pte)		{ return pte_val(pte) & _PAGE_USER; }
-static inline int pte_exec(pte_t pte)		{ return pte_val(pte) & _PAGE_USER; }
+static inline int pte_exec(pte_t pte)		{ return !(pte_val(pte) & _PAGE_NX); }
 static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
 static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
 static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
@@ -278,7 +278,7 @@ static inline pte_t pte_mkclean(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) &
 static inline pte_t pte_mkold(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_wrprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_RW)); return pte; }
 static inline pte_t pte_mkread(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_USER)); return pte; }
-static inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_USER)); return pte; }
+static inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_NX)); return pte; }
 static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
 static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
-- 
cgit v1.2.3


From 5e6b0bfe5b452957b7be4b6ef181cd41880f8359 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Use proper accessors to change PSE bits in change_page_attr()

Use normal pte accessors in change_page_attr() to access the PSE
bits.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/pageattr.c    | 16 ++++++----------
 include/asm-x86_64/pgtable.h |  1 +
 2 files changed, 7 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c
index 01591b649f4e..3e231d762aaa 100644
--- a/arch/x86_64/mm/pageattr.c
+++ b/arch/x86_64/mm/pageattr.c
@@ -108,8 +108,8 @@ static void revert_page(unsigned long address, pgprot_t ref_prot)
 	BUG_ON(pud_none(*pud));
 	pmd = pmd_offset(pud, address);
 	BUG_ON(pmd_val(*pmd) & _PAGE_PSE);
-	pgprot_val(ref_prot) |= _PAGE_PSE;
 	large_pte = mk_pte_phys(__pa(address) & LARGE_PAGE_MASK, ref_prot);
+	large_pte = pte_mkhuge(large_pte);
 	set_pte((pte_t *)pmd, large_pte);
 }      
 
@@ -119,32 +119,28 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
 { 
 	pte_t *kpte; 
 	struct page *kpte_page;
-	unsigned kpte_flags;
 	pgprot_t ref_prot2;
 	kpte = lookup_address(address);
 	if (!kpte) return 0;
 	kpte_page = virt_to_page(((unsigned long)kpte) & PAGE_MASK);
-	kpte_flags = pte_val(*kpte); 
 	if (pgprot_val(prot) != pgprot_val(ref_prot)) { 
-		if ((kpte_flags & _PAGE_PSE) == 0) { 
+		if (!pte_huge(*kpte)) {
 			set_pte(kpte, pfn_pte(pfn, prot));
 		} else {
  			/*
 			 * split_large_page will take the reference for this
 			 * change_page_attr on the split page.
  			 */
-
 			struct page *split;
-			ref_prot2 = __pgprot(pgprot_val(pte_pgprot(*lookup_address(address))) & ~(1<<_PAGE_BIT_PSE));
-
+			ref_prot2 = pte_pgprot(pte_clrhuge(*kpte));
 			split = split_large_page(address, prot, ref_prot2);
 			if (!split)
 				return -ENOMEM;
-			set_pte(kpte,mk_pte(split, ref_prot2));
+			set_pte(kpte, mk_pte(split, ref_prot2));
 			kpte_page = split;
-		}	
+		}
 		page_private(kpte_page)++;
-	} else if ((kpte_flags & _PAGE_PSE) == 0) { 
+	} else if (!pte_huge(*kpte)) {
 		set_pte(kpte, pfn_pte(pfn, ref_prot));
 		BUG_ON(page_private(kpte_page) == 0);
 		page_private(kpte_page)--;
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index f7614670c655..b34f43acdef1 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -283,6 +283,7 @@ static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) |
 static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
 static inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_PSE)); return pte; }
+static inline pte_t pte_clrhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_PSE)); return pte; }
 
 struct vm_area_struct;
 
-- 
cgit v1.2.3


From e4251e130deef9de5226cc36faa70a1c6671d3c5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove some cruft in apic id checking during processor setup

- Remove a define that was used only once
- Remove the too large APIC ID check because we always support
the full 8bit range of APICs.
- Restructure code a bit to be simpler.

Cc: len.brown@intel.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mpparse.c | 18 +-----------------
 include/asm-x86_64/acpi.h    |  2 --
 2 files changed, 1 insertion(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 32b6977f80c0..02eef8f6415f 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -100,7 +100,6 @@ static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 		disabled_cpus++;
 		return;
 	}
-
 	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
 		bootup_cpu = " (Bootup-CPU)";
 		boot_cpu_id = m->mpc_apicid;
@@ -118,13 +117,6 @@ static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 	cpus_complement(tmp_map, cpu_present_map);
 	cpu = first_cpu(tmp_map);
 
-#if MAX_APICS < 255	
-	if ((int)m->mpc_apicid > MAX_APICS) {
-		printk(KERN_ERR "Processor #%d INVALID. (Max ID: %d).\n",
-			m->mpc_apicid, MAX_APICS);
-		return;
-	}
-#endif
 	physid_set(m->mpc_apicid, phys_cpu_present_map);
  	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
  		/*
@@ -603,8 +595,6 @@ void __init mp_register_lapic_address(u64 address)
 	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
 	if (boot_cpu_id == -1U)
 		boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
-
-	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
 
 void __cpuinit mp_register_lapic (u8 id, u8 enabled)
@@ -612,13 +602,7 @@ void __cpuinit mp_register_lapic (u8 id, u8 enabled)
 	struct mpc_config_processor processor;
 	int			boot_cpu = 0;
 	
-	if (id >= MAX_APICS) {
-		printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
-			id, MAX_APICS);
-		return;
-	}
-
-	if (id == boot_cpu_physical_apicid)
+	if (id == boot_cpu_id)
 		boot_cpu = 1;
 
 	processor.mpc_type = MP_PROCESSOR;
diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h
index 2c95a319c056..ed59aa4c6ff9 100644
--- a/include/asm-x86_64/acpi.h
+++ b/include/asm-x86_64/acpi.h
@@ -155,8 +155,6 @@ extern void acpi_reserve_bootmem(void);
 
 #endif /*CONFIG_ACPI_SLEEP*/
 
-#define boot_cpu_physical_apicid boot_cpu_id
-
 extern int acpi_disabled;
 extern int acpi_pci_disabled;
 
-- 
cgit v1.2.3


From 8380aabb99719af583447133f19a4d8074b5c337 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove non e820 fallbacks in high level code

Drop support for non e820 BIOS calls to get the memory map.

The boot assembler code still has some support, but not the C code now.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/e820.c | 28 +++++++++-------------------
 include/asm-x86_64/e820.h |  4 ----
 2 files changed, 9 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index e06c2714ecf3..5d1275a93658 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -536,10 +536,14 @@ static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
 	return 0;
 }
 
-void __init setup_memory_region(void)
+void early_panic(char *msg)
 {
-	char *who = "BIOS-e820";
+	early_printk(msg);
+	panic(msg);
+}
 
+void __init setup_memory_region(void)
+{
 	/*
 	 * Try to copy the BIOS-supplied E820-map.
 	 *
@@ -547,24 +551,10 @@ void __init setup_memory_region(void)
 	 * the next section from 1mb->appropriate_mem_k
 	 */
 	sanitize_e820_map(E820_MAP, &E820_MAP_NR);
-	if (copy_e820_map(E820_MAP, E820_MAP_NR) < 0) {
-		unsigned long mem_size;
-
-		/* compare results from other methods and take the greater */
-		if (ALT_MEM_K < EXT_MEM_K) {
-			mem_size = EXT_MEM_K;
-			who = "BIOS-88";
-		} else {
-			mem_size = ALT_MEM_K;
-			who = "BIOS-e801";
-		}
-
-		e820.nr_map = 0;
-		add_memory_region(0, LOWMEMSIZE(), E820_RAM);
-		add_memory_region(HIGH_MEMORY, mem_size << 10, E820_RAM);
-  	}
+	if (copy_e820_map(E820_MAP, E820_MAP_NR) < 0)
+		early_panic("Cannot find a valid memory map");
 	printk(KERN_INFO "BIOS-provided physical RAM map:\n");
-	e820_print_map(who);
+	e820_print_map("BIOS-e820");
 }
 
 static int __init parse_memopt(char *p)
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index dba012151856..8dac397bf85e 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -22,10 +22,6 @@
 #define E820_ACPI	3 /* usable as RAM once ACPI tables have been read */
 #define E820_NVS	4
 
-#define HIGH_MEMORY	(1024*1024)
-
-#define LOWMEMSIZE()	(0x9f000)
-
 #ifndef __ASSEMBLY__
 struct e820entry {
 	u64 addr;	/* start of memory segment */
-- 
cgit v1.2.3


From 0136611c62e8650e354b95c76dff6d2ce6030eff Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] optimize hweight64 for x86_64

Based on patch from David Rientjes <rientjes@google.com>, but
changed by AK.

Optimizes the 64-bit hamming weight for x86_64 processors assuming they
have fast multiplication.  Uses five fewer bitops than the generic
hweight64.  Benchmark on one EMT64 showed ~25% speedup with 2^24
consecutive calls.

Define a new ARCH_HAS_FAST_MULTIPLIER that can be set by other
architectures that can also multiply fast.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/bitops.h |  2 ++
 lib/hweight.c               | 10 ++++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/bitops.h b/include/asm-x86_64/bitops.h
index f7ba57b1cc08..5b535eaf5309 100644
--- a/include/asm-x86_64/bitops.h
+++ b/include/asm-x86_64/bitops.h
@@ -399,6 +399,8 @@ static __inline__ int fls(int x)
 	return r+1;
 }
 
+#define ARCH_HAS_FAST_MULTIPLIER 1
+
 #include <asm-generic/bitops/hweight.h>
 
 #endif /* __KERNEL__ */
diff --git a/lib/hweight.c b/lib/hweight.c
index 438257671708..360556a7803d 100644
--- a/lib/hweight.c
+++ b/lib/hweight.c
@@ -1,5 +1,6 @@
 #include <linux/module.h>
 #include <asm/types.h>
+#include <asm/bitops.h>
 
 /**
  * hweightN - returns the hamming weight of a N-bit word
@@ -40,14 +41,19 @@ unsigned long hweight64(__u64 w)
 #if BITS_PER_LONG == 32
 	return hweight32((unsigned int)(w >> 32)) + hweight32((unsigned int)w);
 #elif BITS_PER_LONG == 64
+#ifdef ARCH_HAS_FAST_MULTIPLIER
+	w -= (w >> 1) & 0x5555555555555555ul;
+	w =  (w & 0x3333333333333333ul) + ((w >> 2) & 0x3333333333333333ul);
+	w =  (w + (w >> 4)) & 0x0f0f0f0f0f0f0f0ful;
+	return (w * 0x0101010101010101ul) >> 56;
+#else
 	__u64 res = w - ((w >> 1) & 0x5555555555555555ul);
 	res = (res & 0x3333333333333333ul) + ((res >> 2) & 0x3333333333333333ul);
 	res = (res + (res >> 4)) & 0x0F0F0F0F0F0F0F0Ful;
 	res = res + (res >> 8);
 	res = res + (res >> 16);
 	return (res + (res >> 32)) & 0x00000000000000FFul;
-#else
-#error BITS_PER_LONG not defined
+#endif
 #endif
 }
 EXPORT_SYMBOL(hweight64);
-- 
cgit v1.2.3


From 80d2679cbc8e170011c9649fb8fb684ffd7e5c8f Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] x86: Remove incorrect comment about ACPI e820 entries

They cannot be actually freed because the FACS table has a
shared-with-the-BIOS lock.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/e820.h   | 2 +-
 include/asm-x86_64/e820.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h
index ca82acb8cb1f..f7514fb6e8e4 100644
--- a/include/asm-i386/e820.h
+++ b/include/asm-i386/e820.h
@@ -18,7 +18,7 @@
 
 #define E820_RAM	1
 #define E820_RESERVED	2
-#define E820_ACPI	3 /* usable as RAM once ACPI tables have been read */
+#define E820_ACPI	3
 #define E820_NVS	4
 
 #define HIGH_MEMORY	(1024*1024)
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index 8dac397bf85e..fd4198b49637 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -19,7 +19,7 @@
 
 #define E820_RAM	1
 #define E820_RESERVED	2
-#define E820_ACPI	3 /* usable as RAM once ACPI tables have been read */
+#define E820_ACPI	3
 #define E820_NVS	4
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.3


From 53ee11ae0d73f28029a5f0d991bc4dcd7c817e7a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Optimize PDA accesses slightly

Based on a idea by Jeremy Fitzhardinge:

Replace the volatiles and memory clobbers in the PDA access with
telling gcc about access to a proxy PDA structure that doesn't
actually exist. But the dummy accesses give a defined ordering for
read/write accesses.

Also add some memory barriers to the early GS initialization to
make sure no PDA access is moved before it.

Advantage is some .text savings (probably most from better
code for accessing "current"):

   text    data     bss     dec     hex filename
4845647 1223688  615864 6685199  66020f vmlinux
4837780 1223688  615864 6677332  65e354 vmlinux-pda

1.2% smaller code

Cc:  Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c |  3 +++
 include/asm-x86_64/pda.h     | 41 ++++++++++++++++++++++-------------------
 2 files changed, 25 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index e85cfbb49b63..491361752c70 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -121,7 +121,10 @@ void pda_init(int cpu)
 
 	/* Setup up data that may be needed in __get_free_pages early */
 	asm volatile("movl %0,%%fs ; movl %0,%%gs" :: "r" (0)); 
+	/* Memory clobbers used to order PDA accessed */
+	mb();
 	wrmsrl(MSR_GS_BASE, pda);
+	mb();
 
 	pda->cpunumber = cpu; 
 	pda->irqcount = -1;
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index b47c3df9ed1d..55e21da96e7a 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -36,40 +36,43 @@ extern struct x8664_pda boot_cpu_pda[];
  * There is no fast way to get the base address of the PDA, all the accesses
  * have to mention %fs/%gs.  So it needs to be done this Torvaldian way.
  */ 
-#define sizeof_field(type,field)  (sizeof(((type *)0)->field))
-#define typeof_field(type,field)  typeof(((type *)0)->field)
-
 extern void __bad_pda_field(void);
 
+/* proxy_pda doesn't actually exist, but tell gcc it is accessed
+   for all PDA accesses so it gets read/write dependencies right. */
+extern struct x8664_pda _proxy_pda;
+
 #define pda_offset(field) offsetof(struct x8664_pda, field)
 
 #define pda_to_op(op,field,val) do { \
-	typedef typeof_field(struct x8664_pda, field) T__; \
-       switch (sizeof_field(struct x8664_pda, field)) { 		\
+	typedef typeof(_proxy_pda.field) T__; \
+       switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm volatile(op "w %0,%%gs:%P1"::"ri" ((T__)val),"i"(pda_offset(field)):"memory"); break; \
+asm(op "w %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 4: \
-asm volatile(op "l %0,%%gs:%P1"::"ri" ((T__)val),"i"(pda_offset(field)):"memory"); break; \
+asm(op "l %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 8: \
-asm volatile(op "q %0,%%gs:%P1"::"ri" ((T__)val),"i"(pda_offset(field)):"memory"); break; \
-       default: __bad_pda_field(); 					\
+asm(op "q %1,%%gs:%P2": "+m" (_proxy_pda.field) : \
+	 "ri" ((T__)val),"i"(pda_offset(field))); break; \
+default: __bad_pda_field(); 					\
        } \
        } while (0)
 
-/* 
- * AK: PDA read accesses should be neither volatile nor have an memory clobber.
- * Unfortunately removing them causes all hell to break lose currently.
- */
 #define pda_from_op(op,field) ({ \
-       typeof_field(struct x8664_pda, field) ret__; \
-       switch (sizeof_field(struct x8664_pda, field)) { 		\
+       typeof(_proxy_pda.field) ret__; \
+       switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm volatile(op "w %%gs:%P1,%0":"=r" (ret__):"i"(pda_offset(field)):"memory"); break;\
+asm(op "w %%gs:%P1,%0":"=r" (ret__):\
+	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 4: \
-asm volatile(op "l %%gs:%P1,%0":"=r" (ret__):"i"(pda_offset(field)):"memory"); break;\
+asm(op "l %%gs:%P1,%0":"=r" (ret__):\
+	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 8: \
-asm volatile(op "q %%gs:%P1,%0":"=r" (ret__):"i"(pda_offset(field)):"memory"); break;\
-       default: __bad_pda_field(); 					\
+asm(op "q %%gs:%P1,%0":"=r" (ret__):\
+	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
+default: __bad_pda_field(); 					\
        } \
        ret__; })
 
-- 
cgit v1.2.3


From 575400d1b483fbe9e03c68758059bfaf4e4768d1 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] i386: Fix the EDD code misparsing the command line

The EDD code would scan the command line as a fixed array, without
taking account of either whitespace, null-termination, the old
command-line protocol, late overrides early, or the fact that the
command line may not be reachable from INITSEG.

This should fix those problems, and enable us to use a longer command
line.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/boot/edd.S | 97 ++++++++++++++++++++++++++++++++++++++++------------
 include/linux/edd.h  |  1 +
 2 files changed, 76 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/i386/boot/edd.S b/arch/i386/boot/edd.S
index 4b84ea216f2b..34321368011a 100644
--- a/arch/i386/boot/edd.S
+++ b/arch/i386/boot/edd.S
@@ -15,42 +15,95 @@
 #include <asm/setup.h>
 
 #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
+
+# It is assumed that %ds == INITSEG here
+
 	movb	$0, (EDD_MBR_SIG_NR_BUF)
 	movb	$0, (EDDNR)
 
-# Check the command line for two options:
+# Check the command line for options:
 # edd=of  disables EDD completely  (edd=off)
 # edd=sk  skips the MBR test    (edd=skipmbr)
+# edd=on  re-enables EDD (edd=on)
+
 	pushl	%esi
-    	cmpl	$0, %cs:cmd_line_ptr
-	jz	done_cl
+	movw	$edd_mbr_sig_start, %di	# Default to edd=on
+
 	movl	%cs:(cmd_line_ptr), %esi
-# ds:esi has the pointer to the command line now
-	movl	$(COMMAND_LINE_SIZE-7), %ecx
-# loop through kernel command line one byte at a time
-cl_loop:
-	cmpl	$EDD_CL_EQUALS, (%si)
+	andl	%esi, %esi
+	jz	old_cl			# Old boot protocol?
+
+# Convert to a real-mode pointer in fs:si
+	movl	%esi, %eax
+	shrl	$4, %eax
+	movw	%ax, %fs
+	andw	$0xf, %si
+	jmp	have_cl_pointer
+
+# Old-style boot protocol?
+old_cl:
+	push	%ds			# aka INITSEG
+	pop	%fs
+
+	cmpw	$0xa33f, (0x20)
+	jne	done_cl			# No command line at all?
+	movw	(0x22), %si		# Pointer relative to INITSEG
+
+# fs:si has the pointer to the command line now
+have_cl_pointer:
+
+# Loop through kernel command line one byte at a time.  Just in
+# case the loader is buggy and failed to null-terminate the command line
+# terminate if we get close enough to the end of the segment that we
+# cannot fit "edd=XX"...
+cl_atspace:
+	cmpw	$-5, %si		# Watch for segment wraparound
+	jae	done_cl
+	movl	%fs:(%si), %eax
+	andb	%al, %al		# End of line?
+	jz	done_cl
+	cmpl	$EDD_CL_EQUALS, %eax
 	jz	found_edd_equals
-	incl	%esi
-	loop	cl_loop
-	jmp	done_cl
+	cmpb	$0x20, %al		# <= space consider whitespace
+	ja	cl_skipword
+	incw	%si
+	jmp	cl_atspace
+
+cl_skipword:
+	cmpw	$-5, %si		# Watch for segment wraparound
+	jae	done_cl
+	movb	%fs:(%si), %al		# End of string?
+	andb	%al, %al
+	jz	done_cl
+	cmpb	$0x20, %al
+	jbe	cl_atspace
+	incw	%si
+	jmp	cl_skipword
+
 found_edd_equals:
 # only looking at first two characters after equals
-    	addl	$4, %esi
-	cmpw	$EDD_CL_OFF, (%si)	# edd=of
-	jz	do_edd_off
-	cmpw	$EDD_CL_SKIP, (%si)	# edd=sk
-	jz	do_edd_skipmbr
-	jmp	done_cl
+# late overrides early on the command line, so keep going after finding something
+	movw	%fs:4(%si), %ax
+	cmpw	$EDD_CL_OFF, %ax	# edd=of
+	je	do_edd_off
+	cmpw	$EDD_CL_SKIP, %ax	# edd=sk
+	je	do_edd_skipmbr
+	cmpw	$EDD_CL_ON, %ax		# edd=on
+	je	do_edd_on
+	jmp	cl_skipword
 do_edd_skipmbr:
-    	popl	%esi
-	jmp	edd_start
+	movw	$edd_start, %di
+	jmp	cl_skipword
 do_edd_off:
-	popl	%esi
-	jmp	edd_done
+	movw	$edd_done, %di
+	jmp	cl_skipword
+do_edd_on:
+	movw	$edd_mbr_sig_start, %di
+	jmp	cl_skipword
+
 done_cl:
 	popl	%esi
-
+	jmpw	*%di
 
 # Read the first sector of each BIOS disk device and store the 4-byte signature
 edd_mbr_sig_start:
diff --git a/include/linux/edd.h b/include/linux/edd.h
index 162512b886f7..b2b3e68aa512 100644
--- a/include/linux/edd.h
+++ b/include/linux/edd.h
@@ -52,6 +52,7 @@
 #define EDD_CL_EQUALS   0x3d646465     /* "edd=" */
 #define EDD_CL_OFF      0x666f         /* "of" for off  */
 #define EDD_CL_SKIP     0x6b73         /* "sk" for skipmbr */
+#define EDD_CL_ON       0x6e6f	       /* "on" for on */
 
 #ifndef __ASSEMBLY__
 
-- 
cgit v1.2.3


From f574164491d00d28b727d713685fb5edc9138200 Mon Sep 17 00:00:00 2001
From: Keith Owens <kaos@ocs.com.au>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Remove most of the special cases for the debug IST stack

Remove most of the special cases for the debug IST stack.  This is a
follow on clean up patch, it requires the bug fix patch that adds
orig_ist.

Signed-off-by: Keith Owens <kaos@ocs.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c | 21 +++++----------------
 arch/x86_64/kernel/traps.c   | 21 +--------------------
 include/asm-x86_64/pda.h     |  3 ---
 3 files changed, 6 insertions(+), 39 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 491361752c70..9332d2361e08 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -237,28 +237,17 @@ void __cpuinit cpu_init (void)
 	 * set up and load the per-CPU TSS
 	 */
 	for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+		static const unsigned int order[N_EXCEPTION_STACKS] = {
+			[0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+			[DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+		};
 		if (cpu) {
-			static const unsigned int order[N_EXCEPTION_STACKS] = {
-				[0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
-				[DEBUG_STACK - 1] = DEBUG_STACK_ORDER
-			};
-
 			estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
 			if (!estacks)
 				panic("Cannot allocate exception stack %ld %d\n",
 				      v, cpu); 
 		}
-		switch (v + 1) {
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		case DEBUG_STACK:
-			cpu_pda(cpu)->debugstack = (unsigned long)estacks;
-			estacks += DEBUG_STKSZ;
-			break;
-#endif
-		default:
-			estacks += EXCEPTION_STKSZ;
-			break;
-		}
+		estacks += PAGE_SIZE << order[v];
 		orig_ist->ist[v] = t->ist[v] = (unsigned long)estacks;
 	}
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 38bc821e457b..fb8486eca1b6 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -162,26 +162,7 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 	 * 'stack' is in one of them:
 	 */
 	for (k = 0; k < N_EXCEPTION_STACKS; k++) {
-		unsigned long end;
-
-		/*
-		 * set 'end' to the end of the exception stack.
-		 */
-		switch (k + 1) {
-		/*
-		 * TODO: this block is not needed i think, because
-		 * setup64.c:cpu_init() sets up t->ist[DEBUG_STACK]
-		 * properly too.
-		 */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		case DEBUG_STACK:
-			end = cpu_pda(cpu)->debugstack + DEBUG_STKSZ;
-			break;
-#endif
-		default:
-			end = per_cpu(orig_ist, cpu).ist[k];
-			break;
-		}
+		unsigned long end = per_cpu(orig_ist, cpu).ist[k];
 		/*
 		 * Is 'stack' above this exception frame's end?
 		 * If yes then skip to the next frame.
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 55e21da96e7a..e2b23e337b94 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -13,9 +13,6 @@ struct x8664_pda {
 	unsigned long data_offset;	/* Per cpu data offset from linker address */
 	unsigned long kernelstack;  /* top of kernel stack for current */ 
 	unsigned long oldrsp; 	    /* user rsp for system call */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-	unsigned long debugstack;   /* #DB/#BP stack. */
-#endif
         int irqcount;		    /* Irq nesting counter. Starts with -1 */  	
 	int cpunumber;		    /* Logical CPU number */
 	char *irqstackptr;	/* top of irqstack */
-- 
cgit v1.2.3


From 4bfaaef01a1badb9e8ffb0c0a37cd2379008d21f Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Avoid overwriting the current pgd (V4, x86_64)

kexec: Avoid overwriting the current pgd (V4, x86_64)

This patch upgrades the x86_64-specific kexec code to avoid overwriting the
current pgd. Overwriting the current pgd is bad when CONFIG_CRASH_DUMP is used
to start a secondary kernel that dumps the memory of the previous kernel.

The code introduces a new set of page tables. These tables are used to provide
an executable identity mapping without overwriting the current pgd.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/machine_kexec.c   |  71 ++++++++-------
 arch/x86_64/kernel/relocate_kernel.S | 171 +++++++++++++++++++++++++++++++----
 include/asm-x86_64/kexec.h           |  29 ++++++
 3 files changed, 218 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index 2e94c072d84a..0497e3bd5bff 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -15,6 +15,15 @@
 #include <asm/mmu_context.h>
 #include <asm/io.h>
 
+#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
+static u64 kexec_pgd[512] PAGE_ALIGNED;
+static u64 kexec_pud0[512] PAGE_ALIGNED;
+static u64 kexec_pmd0[512] PAGE_ALIGNED;
+static u64 kexec_pte0[512] PAGE_ALIGNED;
+static u64 kexec_pud1[512] PAGE_ALIGNED;
+static u64 kexec_pmd1[512] PAGE_ALIGNED;
+static u64 kexec_pte1[512] PAGE_ALIGNED;
+
 static void init_level2_page(pmd_t *level2p, unsigned long addr)
 {
 	unsigned long end_addr;
@@ -144,32 +153,19 @@ static void load_segments(void)
 		);
 }
 
-typedef NORET_TYPE void (*relocate_new_kernel_t)(unsigned long indirection_page,
-					unsigned long control_code_buffer,
-					unsigned long start_address,
-					unsigned long pgtable) ATTRIB_NORET;
-
-extern const unsigned char relocate_new_kernel[];
-extern const unsigned long relocate_new_kernel_size;
-
 int machine_kexec_prepare(struct kimage *image)
 {
-	unsigned long start_pgtable, control_code_buffer;
+	unsigned long start_pgtable;
 	int result;
 
 	/* Calculate the offsets */
 	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
-	control_code_buffer = start_pgtable + PAGE_SIZE;
 
 	/* Setup the identity mapped 64bit page table */
 	result = init_pgtable(image, start_pgtable);
 	if (result)
 		return result;
 
-	/* Place the code in the reboot code buffer */
-	memcpy(__va(control_code_buffer), relocate_new_kernel,
-						relocate_new_kernel_size);
-
 	return 0;
 }
 
@@ -184,28 +180,34 @@ void machine_kexec_cleanup(struct kimage *image)
  */
 NORET_TYPE void machine_kexec(struct kimage *image)
 {
-	unsigned long page_list;
-	unsigned long control_code_buffer;
-	unsigned long start_pgtable;
-	relocate_new_kernel_t rnk;
+	unsigned long page_list[PAGES_NR];
+	void *control_page;
 
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
-	/* Calculate the offsets */
-	page_list = image->head;
-	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
-	control_code_buffer = start_pgtable + PAGE_SIZE;
-
-	/* Set the low half of the page table to my identity mapped
-	 * page table for kexec.  Leave the high half pointing at the
-	 * kernel pages.   Don't bother to flush the global pages
-	 * as that will happen when I fully switch to my identity mapped
-	 * page table anyway.
-	 */
-	memcpy(__va(read_cr3()), __va(start_pgtable), PAGE_SIZE/2);
-	__flush_tlb();
-
+	control_page = page_address(image->control_code_page) + PAGE_SIZE;
+	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+
+	page_list[PA_CONTROL_PAGE] = __pa(control_page);
+	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
+	page_list[PA_PGD] = __pa(kexec_pgd);
+	page_list[VA_PGD] = (unsigned long)kexec_pgd;
+	page_list[PA_PUD_0] = __pa(kexec_pud0);
+	page_list[VA_PUD_0] = (unsigned long)kexec_pud0;
+	page_list[PA_PMD_0] = __pa(kexec_pmd0);
+	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
+	page_list[PA_PTE_0] = __pa(kexec_pte0);
+	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
+	page_list[PA_PUD_1] = __pa(kexec_pud1);
+	page_list[VA_PUD_1] = (unsigned long)kexec_pud1;
+	page_list[PA_PMD_1] = __pa(kexec_pmd1);
+	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
+	page_list[PA_PTE_1] = __pa(kexec_pte1);
+	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
+
+	page_list[PA_TABLE_PAGE] =
+	  (unsigned long)__pa(page_address(image->control_code_page));
 
 	/* The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
@@ -222,9 +224,10 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	 */
 	set_gdt(phys_to_virt(0),0);
 	set_idt(phys_to_virt(0),0);
+
 	/* now call it */
-	rnk = (relocate_new_kernel_t) control_code_buffer;
-	(*rnk)(page_list, control_code_buffer, image->start, start_pgtable);
+	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
+			image->start);
 }
 
 /* crashkernel=size@addr specifies the location to reserve for
diff --git a/arch/x86_64/kernel/relocate_kernel.S b/arch/x86_64/kernel/relocate_kernel.S
index d24fa9b72a2b..14e95872c6a3 100644
--- a/arch/x86_64/kernel/relocate_kernel.S
+++ b/arch/x86_64/kernel/relocate_kernel.S
@@ -7,31 +7,169 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/page.h>
+#include <asm/kexec.h>
 
-	/*
-	 * Must be relocatable PIC code callable as a C function, that once
-	 * it starts can not use the previous processes stack.
-	 */
-	.globl relocate_new_kernel
+/*
+ * Must be relocatable PIC code callable as a C function
+ */
+
+#define PTR(x) (x << 3)
+#define PAGE_ALIGNED (1 << PAGE_SHIFT)
+#define PAGE_ATTR 0x63 /* _PAGE_PRESENT|_PAGE_RW|_PAGE_ACCESSED|_PAGE_DIRTY */
+
+	.text
+	.align PAGE_ALIGNED
 	.code64
+	.globl relocate_kernel
+relocate_kernel:
+	/* %rdi indirection_page
+	 * %rsi page_list
+	 * %rdx start address
+	 */
+
+	/* map the control page at its virtual address */
+
+	movq	$0x0000ff8000000000, %r10        /* mask */
+	mov	$(39 - 3), %cl                   /* bits to shift */
+	movq	PTR(VA_CONTROL_PAGE)(%rsi), %r11 /* address to map */
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PGD)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PUD_0)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PUD_0)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PMD_0)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PMD_0)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PTE_0)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PTE_0)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	/* identity map the control page at its physical address */
+
+	movq	$0x0000ff8000000000, %r10        /* mask */
+	mov	$(39 - 3), %cl                   /* bits to shift */
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r11 /* address to map */
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PGD)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PUD_1)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PUD_1)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PMD_1)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PMD_1)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PTE_1)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PTE_1)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
 relocate_new_kernel:
-	/* %rdi page_list
-	 * %rsi reboot_code_buffer
+	/* %rdi indirection_page
+	 * %rsi page_list
 	 * %rdx start address
-	 * %rcx page_table
-	 * %r8  arg5
-	 * %r9  arg6
 	 */
 
 	/* zero out flags, and disable interrupts */
 	pushq $0
 	popfq
 
-	/* set a new stack at the bottom of our page... */
-	lea   4096(%rsi), %rsp
+	/* get physical address of control page now */
+	/* this is impossible after page table switch */
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r8
+
+	/* get physical address of page table now too */
+	movq	PTR(PA_TABLE_PAGE)(%rsi), %rcx
 
-	/* store the parameters back on the stack */
-	pushq	%rdx /* store the start address */
+	/* switch to new set of page tables */
+	movq	PTR(PA_PGD)(%rsi), %r9
+	movq	%r9, %cr3
+
+	/* setup a new stack at the end of the physical control page */
+	lea	4096(%r8), %rsp
+
+	/* jump to identity mapped page */
+	addq	$(identity_mapped - relocate_kernel), %r8
+	pushq	%r8
+	ret
+
+identity_mapped:
+	/* store the start address on the stack */
+	pushq   %rdx
 
 	/* Set cr0 to a known state:
 	 * 31 1 == Paging enabled
@@ -136,8 +274,3 @@ relocate_new_kernel:
 	xorq	%r15, %r15
 
 	ret
-relocate_new_kernel_end:
-
-	.globl relocate_new_kernel_size
-relocate_new_kernel_size:
-	.quad relocate_new_kernel_end - relocate_new_kernel
diff --git a/include/asm-x86_64/kexec.h b/include/asm-x86_64/kexec.h
index c564bae03433..5fab957e1091 100644
--- a/include/asm-x86_64/kexec.h
+++ b/include/asm-x86_64/kexec.h
@@ -1,6 +1,27 @@
 #ifndef _X86_64_KEXEC_H
 #define _X86_64_KEXEC_H
 
+#define PA_CONTROL_PAGE  0
+#define VA_CONTROL_PAGE  1
+#define PA_PGD           2
+#define VA_PGD           3
+#define PA_PUD_0         4
+#define VA_PUD_0         5
+#define PA_PMD_0         6
+#define VA_PMD_0         7
+#define PA_PTE_0         8
+#define VA_PTE_0         9
+#define PA_PUD_1         10
+#define VA_PUD_1         11
+#define PA_PMD_1         12
+#define VA_PMD_1         13
+#define PA_PTE_1         14
+#define VA_PTE_1         15
+#define PA_TABLE_PAGE    16
+#define PAGES_NR         17
+
+#ifndef __ASSEMBLY__
+
 #include <linux/string.h>
 
 #include <asm/page.h>
@@ -64,4 +85,12 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
 		newregs->rip = (unsigned long)current_text_addr();
 	}
 }
+
+NORET_TYPE void
+relocate_kernel(unsigned long indirection_page,
+		unsigned long page_list,
+		unsigned long start_address) ATTRIB_NORET;
+
+#endif /* __ASSEMBLY__ */
+
 #endif /* _X86_64_KEXEC_H */
-- 
cgit v1.2.3


From 3566561bfadffcb5dbc85d576be80c0dbf2cccc9 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] i386: Avoid overwriting the current pgd (V4, i386)

kexec: Avoid overwriting the current pgd (V4, i386)

This patch upgrades the i386-specific kexec code to avoid overwriting the
current pgd. Overwriting the current pgd is bad when CONFIG_CRASH_DUMP is used
to start a secondary kernel that dumps the memory of the previous kernel.

The code introduces a new set of page tables. These tables are used to provide
an executable identity mapping without overwriting the current pgd.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/machine_kexec.c   | 117 +++++++--------------------
 arch/i386/kernel/relocate_kernel.S | 162 +++++++++++++++++++++++++++++++++----
 include/asm-i386/kexec.h           |  27 +++++++
 3 files changed, 201 insertions(+), 105 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c
index 66c3dc99a655..91966bafb3dc 100644
--- a/arch/i386/kernel/machine_kexec.c
+++ b/arch/i386/kernel/machine_kexec.c
@@ -21,70 +21,13 @@
 #include <asm/system.h>
 
 #define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
-
-#define L0_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define L1_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define L2_ATTR (_PAGE_PRESENT)
-
-#define LEVEL0_SIZE (1UL << 12UL)
-
-#ifndef CONFIG_X86_PAE
-#define LEVEL1_SIZE (1UL << 22UL)
-static u32 pgtable_level1[1024] PAGE_ALIGNED;
-
-static void identity_map_page(unsigned long address)
-{
-	unsigned long level1_index, level2_index;
-	u32 *pgtable_level2;
-
-	/* Find the current page table */
-	pgtable_level2 = __va(read_cr3());
-
-	/* Find the indexes of the physical address to identity map */
-	level1_index = (address % LEVEL1_SIZE)/LEVEL0_SIZE;
-	level2_index = address / LEVEL1_SIZE;
-
-	/* Identity map the page table entry */
-	pgtable_level1[level1_index] = address | L0_ATTR;
-	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
-
-	/* Flush the tlb so the new mapping takes effect.
-	 * Global tlb entries are not flushed but that is not an issue.
-	 */
-	load_cr3(pgtable_level2);
-}
-
-#else
-#define LEVEL1_SIZE (1UL << 21UL)
-#define LEVEL2_SIZE (1UL << 30UL)
-static u64 pgtable_level1[512] PAGE_ALIGNED;
-static u64 pgtable_level2[512] PAGE_ALIGNED;
-
-static void identity_map_page(unsigned long address)
-{
-	unsigned long level1_index, level2_index, level3_index;
-	u64 *pgtable_level3;
-
-	/* Find the current page table */
-	pgtable_level3 = __va(read_cr3());
-
-	/* Find the indexes of the physical address to identity map */
-	level1_index = (address % LEVEL1_SIZE)/LEVEL0_SIZE;
-	level2_index = (address % LEVEL2_SIZE)/LEVEL1_SIZE;
-	level3_index = address / LEVEL2_SIZE;
-
-	/* Identity map the page table entry */
-	pgtable_level1[level1_index] = address | L0_ATTR;
-	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
-	set_64bit(&pgtable_level3[level3_index],
-					       __pa(pgtable_level2) | L2_ATTR);
-
-	/* Flush the tlb so the new mapping takes effect.
-	 * Global tlb entries are not flushed but that is not an issue.
-	 */
-	load_cr3(pgtable_level3);
-}
+static u32 kexec_pgd[1024] PAGE_ALIGNED;
+#ifdef CONFIG_X86_PAE
+static u32 kexec_pmd0[1024] PAGE_ALIGNED;
+static u32 kexec_pmd1[1024] PAGE_ALIGNED;
 #endif
+static u32 kexec_pte0[1024] PAGE_ALIGNED;
+static u32 kexec_pte1[1024] PAGE_ALIGNED;
 
 static void set_idt(void *newidt, __u16 limit)
 {
@@ -128,16 +71,6 @@ static void load_segments(void)
 #undef __STR
 }
 
-typedef asmlinkage NORET_TYPE void (*relocate_new_kernel_t)(
-					unsigned long indirection_page,
-					unsigned long reboot_code_buffer,
-					unsigned long start_address,
-					unsigned int has_pae) ATTRIB_NORET;
-
-extern const unsigned char relocate_new_kernel[];
-extern void relocate_new_kernel_end(void);
-extern const unsigned int relocate_new_kernel_size;
-
 /*
  * A architecture hook called to validate the
  * proposed image and prepare the control pages
@@ -170,25 +103,29 @@ void machine_kexec_cleanup(struct kimage *image)
  */
 NORET_TYPE void machine_kexec(struct kimage *image)
 {
-	unsigned long page_list;
-	unsigned long reboot_code_buffer;
-
-	relocate_new_kernel_t rnk;
+	unsigned long page_list[PAGES_NR];
+	void *control_page;
 
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
-	/* Compute some offsets */
-	reboot_code_buffer = page_to_pfn(image->control_code_page)
-								<< PAGE_SHIFT;
-	page_list = image->head;
-
-	/* Set up an identity mapping for the reboot_code_buffer */
-	identity_map_page(reboot_code_buffer);
-
-	/* copy it out */
-	memcpy((void *)reboot_code_buffer, relocate_new_kernel,
-						relocate_new_kernel_size);
+	control_page = page_address(image->control_code_page);
+	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+
+	page_list[PA_CONTROL_PAGE] = __pa(control_page);
+	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
+	page_list[PA_PGD] = __pa(kexec_pgd);
+	page_list[VA_PGD] = (unsigned long)kexec_pgd;
+#ifdef CONFIG_X86_PAE
+	page_list[PA_PMD_0] = __pa(kexec_pmd0);
+	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
+	page_list[PA_PMD_1] = __pa(kexec_pmd1);
+	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
+#endif
+	page_list[PA_PTE_0] = __pa(kexec_pte0);
+	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
+	page_list[PA_PTE_1] = __pa(kexec_pte1);
+	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
 
 	/* The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
@@ -207,8 +144,8 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	set_idt(phys_to_virt(0),0);
 
 	/* now call it */
-	rnk = (relocate_new_kernel_t) reboot_code_buffer;
-	(*rnk)(page_list, reboot_code_buffer, image->start, cpu_has_pae);
+	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
+			image->start, cpu_has_pae);
 }
 
 /* crashkernel=size@addr specifies the location to reserve for
diff --git a/arch/i386/kernel/relocate_kernel.S b/arch/i386/kernel/relocate_kernel.S
index d312616effa1..f151d6fae462 100644
--- a/arch/i386/kernel/relocate_kernel.S
+++ b/arch/i386/kernel/relocate_kernel.S
@@ -7,16 +7,138 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/page.h>
+#include <asm/kexec.h>
+
+/*
+ * Must be relocatable PIC code callable as a C function
+ */
+
+#define PTR(x) (x << 2)
+#define PAGE_ALIGNED (1 << PAGE_SHIFT)
+#define PAGE_ATTR 0x63 /* _PAGE_PRESENT|_PAGE_RW|_PAGE_ACCESSED|_PAGE_DIRTY */
+#define PAE_PGD_ATTR 0x01 /* _PAGE_PRESENT */
+
+	.text
+	.align PAGE_ALIGNED
+	.globl relocate_kernel
+relocate_kernel:
+	movl	8(%esp), %ebp /* list of pages */
+
+#ifdef CONFIG_X86_PAE
+	/* map the control page at its virtual address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xc0000000, %eax
+	shrl	$27, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PMD_0)(%ebp), %edx
+	orl	$PAE_PGD_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PMD_0)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x3fe00000, %eax
+	shrl	$18, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_0)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_0)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x001ff000, %eax
+	shrl	$9, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	/* identity map the control page at its physical address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xc0000000, %eax
+	shrl	$27, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PMD_1)(%ebp), %edx
+	orl	$PAE_PGD_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PMD_1)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x3fe00000, %eax
+	shrl	$18, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_1)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_1)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x001ff000, %eax
+	shrl	$9, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+#else
+	/* map the control page at its virtual address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xffc00000, %eax
+	shrl	$20, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_0)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_0)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x003ff000, %eax
+	shrl	$10, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	/* identity map the control page at its physical address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xffc00000, %eax
+	shrl	$20, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_1)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_1)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x003ff000, %eax
+	shrl	$10, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+#endif
 
-	/*
-	 * Must be relocatable PIC code callable as a C function, that once
-	 * it starts can not use the previous processes stack.
-	 */
-	.globl relocate_new_kernel
 relocate_new_kernel:
 	/* read the arguments and say goodbye to the stack */
 	movl  4(%esp), %ebx /* page_list */
-	movl  8(%esp), %ebp /* reboot_code_buffer */
+	movl  8(%esp), %ebp /* list of pages */
 	movl  12(%esp), %edx /* start address */
 	movl  16(%esp), %ecx /* cpu_has_pae */
 
@@ -24,11 +146,26 @@ relocate_new_kernel:
 	pushl $0
 	popfl
 
-	/* set a new stack at the bottom of our page... */
-	lea   4096(%ebp), %esp
+	/* get physical address of control page now */
+	/* this is impossible after page table switch */
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edi
 
-	/* store the parameters back on the stack */
-	pushl   %edx /* store the start address */
+	/* switch to new set of page tables */
+	movl	PTR(PA_PGD)(%ebp), %eax
+	movl	%eax, %cr3
+
+	/* setup a new stack at the end of the physical control page */
+	lea	4096(%edi), %esp
+
+	/* jump to identity mapped page */
+	movl    %edi, %eax
+	addl    $(identity_mapped - relocate_kernel), %eax
+	pushl   %eax
+	ret
+
+identity_mapped:
+	/* store the start address on the stack */
+	pushl   %edx
 
 	/* Set cr0 to a known state:
 	 * 31 0 == Paging disabled
@@ -113,8 +250,3 @@ relocate_new_kernel:
 	xorl    %edi, %edi
 	xorl    %ebp, %ebp
 	ret
-relocate_new_kernel_end:
-
-	.globl relocate_new_kernel_size
-relocate_new_kernel_size:
-	.long relocate_new_kernel_end - relocate_new_kernel
diff --git a/include/asm-i386/kexec.h b/include/asm-i386/kexec.h
index 53f0e06672dc..4dfc9f5ed031 100644
--- a/include/asm-i386/kexec.h
+++ b/include/asm-i386/kexec.h
@@ -1,6 +1,26 @@
 #ifndef _I386_KEXEC_H
 #define _I386_KEXEC_H
 
+#define PA_CONTROL_PAGE  0
+#define VA_CONTROL_PAGE  1
+#define PA_PGD           2
+#define VA_PGD           3
+#define PA_PTE_0         4
+#define VA_PTE_0         5
+#define PA_PTE_1         6
+#define VA_PTE_1         7
+#ifdef CONFIG_X86_PAE
+#define PA_PMD_0         8
+#define VA_PMD_0         9
+#define PA_PMD_1         10
+#define VA_PMD_1         11
+#define PAGES_NR         12
+#else
+#define PAGES_NR         8
+#endif
+
+#ifndef __ASSEMBLY__
+
 #include <asm/fixmap.h>
 #include <asm/ptrace.h>
 #include <asm/string.h>
@@ -72,5 +92,12 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
                newregs->eip = (unsigned long)current_text_addr();
        }
 }
+asmlinkage NORET_TYPE void
+relocate_kernel(unsigned long indirection_page,
+		unsigned long control_page,
+		unsigned long start_address,
+		unsigned int has_pae) ATTRIB_NORET;
+
+#endif /* __ASSEMBLY__ */
 
 #endif /* _I386_KEXEC_H */
-- 
cgit v1.2.3


From 29a9af60e2120f874d0c600bf9e27617254a0488 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Add comments to the PDA structure to annotate offsets

Change the comments in the pda structure to make the first fields to have
their offset documented and to have the comments aligned.
The stack protector series needs a field at offset 40 (gcc ABI); annotate
upto 40 for that reason.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index e2b23e337b94..6794ffaae433 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -9,13 +9,14 @@
 
 /* Per processor datastructure. %gs points to it while the kernel runs */ 
 struct x8664_pda {
-	struct task_struct *pcurrent;	/* Current process */
-	unsigned long data_offset;	/* Per cpu data offset from linker address */
-	unsigned long kernelstack;  /* top of kernel stack for current */ 
-	unsigned long oldrsp; 	    /* user rsp for system call */
-        int irqcount;		    /* Irq nesting counter. Starts with -1 */  	
-	int cpunumber;		    /* Logical CPU number */
-	char *irqstackptr;	/* top of irqstack */
+	struct task_struct *pcurrent;	/* 0  Current process */
+	unsigned long data_offset;	/* 8 Per cpu data offset from linker
+					   address */
+	unsigned long kernelstack;  /* 16 top of kernel stack for current */
+	unsigned long oldrsp; 	    /* 24 user rsp for system call */
+        int irqcount;		    /* 32 Irq nesting counter. Starts with -1 */
+	int cpunumber;		    /* 36 Logical CPU number */
+	char *irqstackptr;	/* 40 top of irqstack */
 	int nodenumber;		    /* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
-- 
cgit v1.2.3


From 0a4254058037eb172758961d0a5b94f4320a1425 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Add the canary field to the PDA area and the task struct

This patch adds the per thread cookie field to the task struct and the PDA.
Also it makes sure that the PDA value gets the new cookie value at context
switch, and that a new task gets a new cookie at task creation time.

Signed-off-by: Arjan van Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/process.c | 8 ++++++++
 include/asm-x86_64/pda.h     | 7 ++++++-
 include/linux/sched.h        | 5 +++++
 kernel/fork.c                | 5 +++++
 4 files changed, 24 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 9e9a70e50c72..fba8dfeda67c 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -625,6 +625,14 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	unlazy_fpu(prev_p);
 	write_pda(kernelstack,
 		  task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
+#ifdef CONFIG_CC_STACKPROTECTOR
+	write_pda(stack_canary, next_p->stack_canary);
+	/*
+	 * Build time only check to make sure the stack_canary is at
+	 * offset 40 in the pda; this is a gcc ABI requirement
+	 */
+	BUILD_BUG_ON(offsetof(struct x8664_pda, stack_canary) != 40);
+#endif
 
 	/*
 	 * Now maybe reload the debug registers and handle I/O bitmaps
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 6794ffaae433..e7773e0af865 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -16,7 +16,12 @@ struct x8664_pda {
 	unsigned long oldrsp; 	    /* 24 user rsp for system call */
         int irqcount;		    /* 32 Irq nesting counter. Starts with -1 */
 	int cpunumber;		    /* 36 Logical CPU number */
-	char *irqstackptr;	/* 40 top of irqstack */
+#ifdef CONFIG_CC_STACKPROTECTOR
+	unsigned long stack_canary;	/* 40 stack canary value */
+					/* gcc-ABI: this canary MUST be at
+					   offset 40!!! */
+#endif
+	char *irqstackptr;
 	int nodenumber;		    /* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 807556c5bcd2..9d4aa7f95bc8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -819,6 +819,11 @@ struct task_struct {
 	unsigned did_exec:1;
 	pid_t pid;
 	pid_t tgid;
+
+#ifdef CONFIG_CC_STACKPROTECTOR
+	/* Canary value for the -fstack-protector gcc feature */
+	unsigned long stack_canary;
+#endif
 	/* 
 	 * pointers to (original) parent process, youngest child, younger sibling,
 	 * older sibling, respectively.  (p->father can be replaced with 
diff --git a/kernel/fork.c b/kernel/fork.c
index f9b014e3e700..a0dad84567c9 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -45,6 +45,7 @@
 #include <linux/cn_proc.h>
 #include <linux/delayacct.h>
 #include <linux/taskstats_kern.h>
+#include <linux/random.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -175,6 +176,10 @@ static struct task_struct *dup_task_struct(struct task_struct *orig)
 	tsk->thread_info = ti;
 	setup_thread_stack(tsk, orig);
 
+#ifdef CONFIG_CC_STACKPROTECTOR
+	tsk->stack_canary = get_random_int();
+#endif
+
 	/* One for us, one for whoever does the "release_task()" (usually parent) */
 	atomic_set(&tsk->usage,2);
 	atomic_set(&tsk->fs_excl, 0);
-- 
cgit v1.2.3


From baf5695dd1a49bb48a3daf08726d7f243f42e97e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Use %c instead of %P modifier in pda access

Apparently that is the more official way to get numbers without $ in inline
assembly

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index e7773e0af865..c2aac96ac323 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -51,13 +51,13 @@ extern struct x8664_pda _proxy_pda;
 	typedef typeof(_proxy_pda.field) T__; \
        switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm(op "w %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+asm(op "w %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
 	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 4: \
-asm(op "l %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+asm(op "l %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
 	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 8: \
-asm(op "q %1,%%gs:%P2": "+m" (_proxy_pda.field) : \
+asm(op "q %1,%%gs:%c2": "+m" (_proxy_pda.field) : \
 	 "ri" ((T__)val),"i"(pda_offset(field))); break; \
 default: __bad_pda_field(); 					\
        } \
@@ -67,13 +67,13 @@ default: __bad_pda_field(); 					\
        typeof(_proxy_pda.field) ret__; \
        switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm(op "w %%gs:%P1,%0":"=r" (ret__):\
+asm(op "w %%gs:%c1,%0":"=r" (ret__):\
 	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 4: \
-asm(op "l %%gs:%P1,%0":"=r" (ret__):\
+asm(op "l %%gs:%c1,%0":"=r" (ret__):\
 	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 8: \
-asm(op "q %%gs:%P1,%0":"=r" (ret__):\
+asm(op "q %%gs:%c1,%0":"=r" (ret__):\
 	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 default: __bad_pda_field(); 					\
        } \
-- 
cgit v1.2.3


From 85691f135db78f3548107a0abe383dfab3bc38fa Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Type checking for write_pda()

I just added type checking for assignments the PDA in the i386 PDA code.
Here's the x86-64 equivalent.  (Obviously this doesn't contain the latest
x86-64 PDA change.)

Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index c2aac96ac323..5dadb201f769 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -49,7 +49,8 @@ extern struct x8664_pda _proxy_pda;
 
 #define pda_to_op(op,field,val) do { \
 	typedef typeof(_proxy_pda.field) T__; \
-       switch (sizeof(_proxy_pda.field)) { 		\
+	if (0) { T__ tmp__; tmp__ = (val); }  \
+	switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
 asm(op "w %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
 	"ri" ((T__)val),"i"(pda_offset(field))); break; \
-- 
cgit v1.2.3


From 0da5db313317e3195482d3e660a1074857374a89 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] i386: Abstract sensitive instructions

Abstract sensitive instructions in assembler code, replacing them with macros
(which currently are #defined to the native versions).  We use long names:
assembler is case-insensitive, so if something goes wrong and macros do not
expand, it would assemble anyway.

Resulting object files are exactly the same as before.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S    | 38 ++++++++++++++++++++++----------------
 include/asm-i386/spinlock.h |  7 +++++--
 2 files changed, 27 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 4b0845249222..3872fca5c74a 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -76,8 +76,15 @@ DF_MASK		= 0x00000400
 NT_MASK		= 0x00004000
 VM_MASK		= 0x00020000
 
+/* These are replaces for paravirtualization */
+#define DISABLE_INTERRUPTS		cli
+#define ENABLE_INTERRUPTS		sti
+#define ENABLE_INTERRUPTS_SYSEXIT	sti; sysexit
+#define INTERRUPT_RETURN		iret
+#define GET_CR0_INTO_EAX		movl %cr0, %eax
+
 #ifdef CONFIG_PREEMPT
-#define preempt_stop		cli; TRACE_IRQS_OFF
+#define preempt_stop		DISABLE_INTERRUPTS; TRACE_IRQS_OFF
 #else
 #define preempt_stop
 #define resume_kernel		restore_nocheck
@@ -236,7 +243,7 @@ check_userspace:
 	testl $(VM_MASK | 3), %eax
 	jz resume_kernel
 ENTRY(resume_userspace)
- 	cli				# make sure we don't miss an interrupt
+ 	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
 	movl TI_flags(%ebp), %ecx
@@ -247,7 +254,7 @@ ENTRY(resume_userspace)
 
 #ifdef CONFIG_PREEMPT
 ENTRY(resume_kernel)
-	cli
+	DISABLE_INTERRUPTS
 	cmpl $0,TI_preempt_count(%ebp)	# non-zero preempt_count ?
 	jnz restore_nocheck
 need_resched:
@@ -275,7 +282,7 @@ sysenter_past_esp:
 	 * No need to follow this irqs on/off section: the syscall
 	 * disabled irqs and here we enable it straight after entry:
 	 */
-	sti
+	ENABLE_INTERRUPTS
 	pushl $(__USER_DS)
 	CFI_ADJUST_CFA_OFFSET 4
 	/*CFI_REL_OFFSET ss, 0*/
@@ -320,7 +327,7 @@ sysenter_past_esp:
 	jae syscall_badsys
 	call *sys_call_table(,%eax,4)
 	movl %eax,EAX(%esp)
-	cli
+	DISABLE_INTERRUPTS
 	TRACE_IRQS_OFF
 	movl TI_flags(%ebp), %ecx
 	testw $_TIF_ALLWORK_MASK, %cx
@@ -330,8 +337,7 @@ sysenter_past_esp:
 	movl OLDESP(%esp), %ecx
 	xorl %ebp,%ebp
 	TRACE_IRQS_ON
-	sti
-	sysexit
+	ENABLE_INTERRUPTS_SYSEXIT
 	CFI_ENDPROC
 
 
@@ -356,7 +362,7 @@ syscall_call:
 	call *sys_call_table(,%eax,4)
 	movl %eax,EAX(%esp)		# store the return value
 syscall_exit:
-	cli				# make sure we don't miss an interrupt
+	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
 	TRACE_IRQS_OFF
@@ -381,11 +387,11 @@ restore_nocheck_notrace:
 	RESTORE_REGS
 	addl $4, %esp
 	CFI_ADJUST_CFA_OFFSET -4
-1:	iret
+1:	INTERRUPT_RETURN
 .section .fixup,"ax"
 iret_exc:
 	TRACE_IRQS_ON
-	sti
+	ENABLE_INTERRUPTS
 	pushl $0			# no error code
 	pushl $do_iret_error
 	jmp error_code
@@ -409,7 +415,7 @@ ldt_ss:
 	 * dosemu and wine happy. */
 	subl $8, %esp		# reserve space for switch16 pointer
 	CFI_ADJUST_CFA_OFFSET 8
-	cli
+	DISABLE_INTERRUPTS
 	TRACE_IRQS_OFF
 	movl %esp, %eax
 	/* Set up the 16bit stack frame with switch32 pointer on top,
@@ -419,7 +425,7 @@ ldt_ss:
 	TRACE_IRQS_IRET
 	RESTORE_REGS
 	lss 20+4(%esp), %esp	# switch to 16bit stack
-1:	iret
+1:	INTERRUPT_RETURN
 .section __ex_table,"a"
 	.align 4
 	.long 1b,iret_exc
@@ -434,7 +440,7 @@ work_pending:
 	jz work_notifysig
 work_resched:
 	call schedule
-	cli				# make sure we don't miss an interrupt
+	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
 	TRACE_IRQS_OFF
@@ -490,7 +496,7 @@ syscall_exit_work:
 	testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl
 	jz work_pending
 	TRACE_IRQS_ON
-	sti				# could let do_syscall_trace() call
+	ENABLE_INTERRUPTS		# could let do_syscall_trace() call
 					# schedule() instead
 	movl %esp, %eax
 	movl $1, %edx
@@ -668,7 +674,7 @@ ENTRY(device_not_available)
 	pushl $-1			# mark this as an int
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
-	movl %cr0, %eax
+	GET_CR0_INTO_EAX
 	testl $0x4, %eax		# EM (math emulation bit)
 	jne device_not_available_emulate
 	preempt_stop
@@ -811,7 +817,7 @@ nmi_16bit_stack:
 	call do_nmi
 	RESTORE_REGS
 	lss 12+4(%esp), %esp		# back to 16bit stack
-1:	iret
+1:	INTERRUPT_RETURN
 	CFI_ENDPROC
 .section __ex_table,"a"
 	.align 4
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index 324329313af8..b0b3043f05e1 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -7,6 +7,9 @@
 #include <asm/processor.h>
 #include <linux/compiler.h>
 
+#define CLI_STRING	"cli"
+#define STI_STRING	"sti"
+
 /*
  * Your basic SMP spinlocks, allowing only a single CPU anywhere
  *
@@ -55,12 +58,12 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long fla
 		"2:\t"
 		"testl $0x200, %1\n\t"
 		"jz 4f\n\t"
-		"sti\n"
+		STI_STRING "\n"
 		"3:\t"
 		"rep;nop\n\t"
 		"cmpb $0, %0\n\t"
 		"jle 3b\n\t"
-		"cli\n\t"
+		CLI_STRING "\n\t"
 		"jmp 1b\n"
 		"4:\t"
 		"rep;nop\n\t"
-- 
cgit v1.2.3


From 78be3706b21a232310590fe00258b224177ac05f Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] i386: Allow a kernel not to be in ring 0

We allow for the fact that the guest kernel may not run in ring 0.  This
requires some abstraction in a few places when setting %cs or checking
privilege level (user vs kernel).

This is Chris' [RFC PATCH 15/33] move segment checks to subarch, except rather
than using #define USER_MODE_MASK which depends on a config option, we use
Zach's more flexible approach of assuming ring 3 == userspace.  I also used
"get_kernel_rpl()" over "get_kernel_cs()" because I think it reads better in
the code...

1) Remove the hardcoded 3 and introduce #define SEGMENT_RPL_MASK 3 2) Add a
get_kernel_rpl() macro, and don't assume it's zero.

And:

Clean up of patch for letting kernel run other than ring 0:

a. Add some comments about the SEGMENT_IS_*_CODE() macros.
b. Add a USER_RPL macro.  (Code was comparing a value to a mask
   in some places and to the magic number 3 in other places.)
c. Add macros for table indicator field and use them.
d. Change the entry.S tests for LDT stack segment to use the macros

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S   |  9 +++++----
 arch/i386/kernel/process.c |  2 +-
 arch/i386/mm/extable.c     |  2 +-
 arch/i386/mm/fault.c       | 11 ++++-------
 include/asm-i386/ptrace.h  |  5 +++--
 include/asm-i386/segment.h | 17 +++++++++++++++++
 6 files changed, 31 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 3872fca5c74a..284f2e908ad0 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -240,8 +240,9 @@ ret_from_intr:
 check_userspace:
 	movl EFLAGS(%esp), %eax		# mix EFLAGS and CS
 	movb CS(%esp), %al
-	testl $(VM_MASK | 3), %eax
-	jz resume_kernel
+	andl $(VM_MASK | SEGMENT_RPL_MASK), %eax
+	cmpl $USER_RPL, %eax
+	jb resume_kernel		# not returning to v8086 or userspace
 ENTRY(resume_userspace)
  	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
@@ -377,8 +378,8 @@ restore_all:
 	# See comments in process.c:copy_thread() for details.
 	movb OLDSS(%esp), %ah
 	movb CS(%esp), %al
-	andl $(VM_MASK | (4 << 8) | 3), %eax
-	cmpl $((4 << 8) | 3), %eax
+	andl $(VM_MASK | (SEGMENT_TI_MASK << 8) | SEGMENT_RPL_MASK), %eax
+	cmpl $((SEGMENT_LDT << 8) | USER_RPL), %eax
 	CFI_REMEMBER_STATE
 	je ldt_ss			# returning to user-space with LDT SS
 restore_nocheck:
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 220aeca59c3a..8c190ca7ae44 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -338,7 +338,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 	regs.xes = __USER_DS;
 	regs.orig_eax = -1;
 	regs.eip = (unsigned long) kernel_thread_helper;
-	regs.xcs = __KERNEL_CS;
+	regs.xcs = __KERNEL_CS | get_kernel_rpl();
 	regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
 
 	/* Ok, create the new process.. */
diff --git a/arch/i386/mm/extable.c b/arch/i386/mm/extable.c
index de03c5430abc..0ce4f22a2635 100644
--- a/arch/i386/mm/extable.c
+++ b/arch/i386/mm/extable.c
@@ -11,7 +11,7 @@ int fixup_exception(struct pt_regs *regs)
 	const struct exception_table_entry *fixup;
 
 #ifdef CONFIG_PNPBIOS
-	if (unlikely((regs->xcs & ~15) == (GDT_ENTRY_PNPBIOS_BASE << 3)))
+	if (unlikely(SEGMENT_IS_PNP_CODE(regs->xcs)))
 	{
 		extern u32 pnp_bios_fault_eip, pnp_bios_fault_esp;
 		extern u32 pnp_bios_is_utter_crap;
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index 0ce86168a0b1..5e17a3f43b41 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -27,6 +27,7 @@
 #include <asm/uaccess.h>
 #include <asm/desc.h>
 #include <asm/kdebug.h>
+#include <asm/segment.h>
 
 extern void die(const char *,struct pt_regs *,long);
 
@@ -113,10 +114,10 @@ static inline unsigned long get_segment_eip(struct pt_regs *regs,
 	}
 
 	/* The standard kernel/user address space limit. */
-	*eip_limit = (seg & 3) ? USER_DS.seg : KERNEL_DS.seg;
+	*eip_limit = user_mode(regs) ? USER_DS.seg : KERNEL_DS.seg;
 	
 	/* By far the most common cases. */
-	if (likely(seg == __USER_CS || seg == __KERNEL_CS))
+	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
 		return eip;
 
 	/* Check the segment exists, is within the current LDT/GDT size,
@@ -430,11 +431,7 @@ good_area:
 	write = 0;
 	switch (error_code & 3) {
 		default:	/* 3: write, present */
-#ifdef TEST_VERIFY_AREA
-			if (regs->cs == KERNEL_CS)
-				printk("WP fault at %08lx\n", regs->eip);
-#endif
-			/* fall through */
+				/* fall through */
 		case 2:		/* write, not present */
 			if (!(vma->vm_flags & VM_WRITE))
 				goto bad_area;
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index 30a442ec2059..21bb91679c82 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -60,6 +60,7 @@ struct pt_regs {
 #ifdef __KERNEL__
 
 #include <asm/vm86.h>
+#include <asm/segment.h>
 
 struct task_struct;
 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
@@ -73,11 +74,11 @@ extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int erro
  */
 static inline int user_mode(struct pt_regs *regs)
 {
-	return (regs->xcs & 3) != 0;
+	return (regs->xcs & SEGMENT_RPL_MASK) == USER_RPL;
 }
 static inline int user_mode_vm(struct pt_regs *regs)
 {
-	return ((regs->xcs & 3) | (regs->eflags & VM_MASK)) != 0;
+	return ((regs->xcs & SEGMENT_RPL_MASK) | (regs->eflags & VM_MASK)) >= USER_RPL;
 }
 #define instruction_pointer(regs) ((regs)->eip)
 extern unsigned long profile_pc(struct pt_regs *regs);
diff --git a/include/asm-i386/segment.h b/include/asm-i386/segment.h
index faf995307b9e..b7ab59685ba7 100644
--- a/include/asm-i386/segment.h
+++ b/include/asm-i386/segment.h
@@ -83,6 +83,11 @@
 
 #define GDT_SIZE (GDT_ENTRIES * 8)
 
+/* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */
+#define SEGMENT_IS_FLAT_CODE(x)  (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8)
+/* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */
+#define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8)
+
 /* Simple and small GDT entries for booting only */
 
 #define GDT_ENTRY_BOOT_CS		2
@@ -112,4 +117,16 @@
  */
 #define IDT_ENTRIES 256
 
+/* Bottom two bits of selector give the ring privilege level */
+#define SEGMENT_RPL_MASK	0x3
+/* Bit 2 is table indicator (LDT/GDT) */
+#define SEGMENT_TI_MASK		0x4
+
+/* User mode is privilege level 3 */
+#define USER_RPL		0x3
+/* LDT segment has TI set, GDT has it cleared */
+#define SEGMENT_LDT		0x4
+#define SEGMENT_GDT		0x0
+
+#define get_kernel_rpl()  0
 #endif
-- 
cgit v1.2.3


From 3022d734a54cbd2b65eea9a024564821101b4a9a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Fix zeroing on exception in copy_*_user

- Don't zero for __copy_from_user_inatomic following i386.
This will prevent spurious zeros for parallel file system writers when
one does a exception
- The string instruction version didn't zero the output on
exception. Oops.

Also I cleaned up the code a bit while I was at it and added a minor
optimization to the string instruction path.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/x8664_ksyms.c |   1 +
 arch/x86_64/lib/copy_user.S      | 124 +++++++++++++++++++++++++--------------
 include/asm-x86_64/uaccess.h     |   6 +-
 3 files changed, 83 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c
index 370952c4ff22..c3454af5e3a2 100644
--- a/arch/x86_64/kernel/x8664_ksyms.c
+++ b/arch/x86_64/kernel/x8664_ksyms.c
@@ -29,6 +29,7 @@ EXPORT_SYMBOL(__put_user_8);
 EXPORT_SYMBOL(copy_user_generic);
 EXPORT_SYMBOL(copy_from_user);
 EXPORT_SYMBOL(copy_to_user);
+EXPORT_SYMBOL(__copy_from_user_inatomic);
 
 EXPORT_SYMBOL(copy_page);
 EXPORT_SYMBOL(clear_page);
diff --git a/arch/x86_64/lib/copy_user.S b/arch/x86_64/lib/copy_user.S
index 962f3a693c5e..70bebd310408 100644
--- a/arch/x86_64/lib/copy_user.S
+++ b/arch/x86_64/lib/copy_user.S
@@ -9,10 +9,29 @@
 
 #define FIX_ALIGNMENT 1
 
-	#include <asm/current.h>
-	#include <asm/asm-offsets.h>
-	#include <asm/thread_info.h>
-	#include <asm/cpufeature.h>
+#include <asm/current.h>
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
+#include <asm/cpufeature.h>
+
+	.macro ALTERNATIVE_JUMP feature,orig,alt
+0:
+	.byte 0xe9	/* 32bit jump */
+	.long \orig-1f	/* by default jump to orig */
+1:
+	.section .altinstr_replacement,"ax"
+2:	.byte 0xe9	             /* near jump with 32bit immediate */
+	.long \alt-1b /* offset */   /* or alternatively to alt */
+	.previous
+	.section .altinstructions,"a"
+	.align 8
+	.quad  0b
+	.quad  2b
+	.byte  \feature		     /* when feature is set */
+	.byte  5
+	.byte  5
+	.previous
+	.endm
 
 /* Standard copy_to_user with segment limit checking */		
 ENTRY(copy_to_user)
@@ -23,25 +42,21 @@ ENTRY(copy_to_user)
 	jc  bad_to_user
 	cmpq threadinfo_addr_limit(%rax),%rcx
 	jae bad_to_user
-2:
-	.byte 0xe9	/* 32bit jump */
-	.long .Lcug-1f
-1:
+	xorl %eax,%eax	/* clear zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
 	CFI_ENDPROC
-ENDPROC(copy_to_user)
 
-	.section .altinstr_replacement,"ax"
-3:	.byte 0xe9			/* replacement jmp with 32 bit immediate */
-	.long copy_user_generic_c-1b	/* offset */
-	.previous
-	.section .altinstructions,"a"
-	.align 8
-	.quad  2b
-	.quad  3b
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  5
-	.byte  5
-	.previous
+ENTRY(copy_user_generic)
+	CFI_STARTPROC
+	movl $1,%ecx	/* set zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
+	CFI_ENDPROC
+
+ENTRY(__copy_from_user_inatomic)
+	CFI_STARTPROC
+	xorl %ecx,%ecx	/* clear zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
+	CFI_ENDPROC
 
 /* Standard copy_from_user with segment limit checking */	
 ENTRY(copy_from_user)
@@ -52,7 +67,8 @@ ENTRY(copy_from_user)
 	jc  bad_from_user
 	cmpq threadinfo_addr_limit(%rax),%rcx
 	jae  bad_from_user
-	/* FALL THROUGH to copy_user_generic */
+	movl $1,%ecx	/* set zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
 	CFI_ENDPROC
 ENDPROC(copy_from_user)
 	
@@ -73,37 +89,26 @@ END(bad_from_user)
 	
 		
 /*
- * copy_user_generic - memory copy with exception handling.
+ * copy_user_generic_unrolled - memory copy with exception handling.
+ * This version is for CPUs like P4 that don't have efficient micro code for rep movsq
  * 	
  * Input:	
  * rdi destination
  * rsi source
  * rdx count
+ * ecx zero flag -- if true zero destination on error
  *
  * Output:		
  * eax uncopied bytes or 0 if successful.
  */
-ENTRY(copy_user_generic)
+ENTRY(copy_user_generic_unrolled)
 	CFI_STARTPROC
-	.byte 0x66,0x66,0x90	/* 5 byte nop for replacement jump */
-	.byte 0x66,0x90
-1:
-	.section .altinstr_replacement,"ax"
-2:	.byte 0xe9	             /* near jump with 32bit immediate */
-	.long copy_user_generic_c-1b /* offset */
-	.previous
-	.section .altinstructions,"a"
-	.align 8
-	.quad  copy_user_generic
-	.quad  2b
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  5
-	.byte  5
-	.previous
-.Lcug:
 	pushq %rbx
 	CFI_ADJUST_CFA_OFFSET 8
 	CFI_REL_OFFSET rbx, 0
+	pushq %rcx
+	CFI_ADJUST_CFA_OFFSET 8
+	CFI_REL_OFFSET rcx, 0
 	xorl %eax,%eax		/*zero for the exception handler */
 
 #ifdef FIX_ALIGNMENT
@@ -179,6 +184,9 @@ ENTRY(copy_user_generic)
 
 	CFI_REMEMBER_STATE
 .Lende:
+	popq %rcx
+	CFI_ADJUST_CFA_OFFSET -8
+	CFI_RESTORE rcx
 	popq %rbx
 	CFI_ADJUST_CFA_OFFSET -8
 	CFI_RESTORE rbx
@@ -265,6 +273,8 @@ ENTRY(copy_user_generic)
 	addl %ecx,%edx
 	/* edx: bytes to zero, rdi: dest, eax:zero */
 .Lzero_rest:
+	cmpl $0,(%rsp)
+	jz   .Le_zero
 	movq %rdx,%rcx
 .Le_byte:
 	xorl %eax,%eax
@@ -286,6 +296,7 @@ ENDPROC(copy_user_generic)
  /* rdi	destination
   * rsi source
   * rdx count
+  * ecx zero flag
   *
   * Output:
   * eax uncopied bytes or 0 if successfull.
@@ -296,25 +307,48 @@ ENDPROC(copy_user_generic)
   * And more would be dangerous because both Intel and AMD have
   * errata with rep movsq > 4GB. If someone feels the need to fix
   * this please consider this.
-   */
-copy_user_generic_c:
+  */
+ENTRY(copy_user_generic_string)
 	CFI_STARTPROC
+	movl %ecx,%r8d		/* save zero flag */
 	movl %edx,%ecx
 	shrl $3,%ecx
 	andl $7,%edx	
+	jz   10f
 1:	rep 
 	movsq 
 	movl %edx,%ecx
 2:	rep
 	movsb
-4:	movl %ecx,%eax
+9:	movl %ecx,%eax
 	ret
-3:	lea (%rdx,%rcx,8),%rax
+
+	/* multiple of 8 byte */
+10:	rep
+	movsq
+	xor %eax,%eax
 	ret
+
+	/* exception handling */
+3:      lea (%rdx,%rcx,8),%rax	/* exception on quad loop */
+	jmp 6f
+5:	movl %ecx,%eax		/* exception on byte loop */
+	/* eax: left over bytes */
+6:	testl %r8d,%r8d		/* zero flag set? */
+	jz 7f
+	movl %eax,%ecx		/* initialize x86 loop counter */
+	push %rax
+	xorl %eax,%eax
+8:	rep
+	stosb 			/* zero the rest */
+11:	pop %rax
+7:	ret
 	CFI_ENDPROC
 END(copy_user_generic_c)
 
 	.section __ex_table,"a"
 	.quad 1b,3b
-	.quad 2b,4b
+	.quad 2b,5b
+	.quad 8b,11b
+	.quad 10b,3b
 	.previous
diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index 1e1fa003daa3..bc6812009bd7 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -238,6 +238,7 @@ do {									\
 
 /* Handles exceptions in both to and from, but doesn't do access_ok */
 extern unsigned long copy_user_generic(void *to, const void *from, unsigned len); 
+extern unsigned long copy_user_generic_dontzero(void *to, const void *from, unsigned len);
 
 extern unsigned long copy_to_user(void __user *to, const void *from, unsigned len); 
 extern unsigned long copy_from_user(void *to, const void __user *from, unsigned len); 
@@ -303,7 +304,6 @@ static __always_inline int __copy_to_user(void __user *dst, const void *src, uns
 	}
 }	
 
-
 static __always_inline int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 { 
        int ret = 0;
@@ -352,7 +352,7 @@ long strlen_user(const char __user *str);
 unsigned long clear_user(void __user *mem, unsigned long len);
 unsigned long __clear_user(void __user *mem, unsigned long len);
 
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+extern long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
+#define __copy_to_user_inatomic copy_user_generic
 
 #endif /* __X86_64_UACCESS_H */
-- 
cgit v1.2.3


From 95912008ba1fb9d0677c1ce5930aeb0e85ba5710 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Add __must_check to copy_*_user

Following i386.

And also fix the two occurrences that caused warnings in arch/x86_64/*

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ptrace32.c  |  6 ++++--
 include/asm-x86_64/i387.h    |  4 ++--
 include/asm-x86_64/uaccess.h | 46 ++++++++++++++++++++++++++------------------
 3 files changed, 33 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/ia32/ptrace32.c b/arch/x86_64/ia32/ptrace32.c
index 72bf92a9d375..d18198ed636b 100644
--- a/arch/x86_64/ia32/ptrace32.c
+++ b/arch/x86_64/ia32/ptrace32.c
@@ -375,8 +375,10 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 		ret = -EIO;
 		if (!access_ok(VERIFY_READ, u, sizeof(*u)))
 			break;
-		/* no checking to be bug-to-bug compatible with i386 */
-		__copy_from_user(&child->thread.i387.fxsave, u, sizeof(*u));
+		/* no checking to be bug-to-bug compatible with i386. */
+		/* but silence warning */
+		if (__copy_from_user(&child->thread.i387.fxsave, u, sizeof(*u)))
+			;
 		set_stopped_child_used_math(child);
 		child->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
 		ret = 0; 
diff --git a/include/asm-x86_64/i387.h b/include/asm-x86_64/i387.h
index 60c0f4853fdb..0217b74cc9fc 100644
--- a/include/asm-x86_64/i387.h
+++ b/include/asm-x86_64/i387.h
@@ -137,8 +137,8 @@ static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
 #else
 		     : [fx] "cdaSDb" (fx), "0" (0));
 #endif
-	if (unlikely(err))
-		__clear_user(fx, sizeof(struct i387_fxsave_struct));
+	if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct)))
+		err = -EFAULT;
 	/* No need to clear here because the caller clears USED_MATH */
 	return err;
 } 
diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index bc6812009bd7..802a4a068ef6 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -237,14 +237,18 @@ do {									\
  */
 
 /* Handles exceptions in both to and from, but doesn't do access_ok */
-extern unsigned long copy_user_generic(void *to, const void *from, unsigned len); 
-extern unsigned long copy_user_generic_dontzero(void *to, const void *from, unsigned len);
-
-extern unsigned long copy_to_user(void __user *to, const void *from, unsigned len); 
-extern unsigned long copy_from_user(void *to, const void __user *from, unsigned len); 
-extern unsigned long copy_in_user(void __user *to, const void __user *from, unsigned len); 
-
-static __always_inline int __copy_from_user(void *dst, const void __user *src, unsigned size)
+__must_check unsigned long
+copy_user_generic(void *to, const void *from, unsigned len);
+
+__must_check unsigned long
+copy_to_user(void __user *to, const void *from, unsigned len);
+__must_check unsigned long
+copy_from_user(void *to, const void __user *from, unsigned len);
+__must_check unsigned long
+copy_in_user(void __user *to, const void __user *from, unsigned len);
+
+static __always_inline __must_check
+int __copy_from_user(void *dst, const void __user *src, unsigned size)
 { 
        int ret = 0;
 	if (!__builtin_constant_p(size))
@@ -273,7 +277,8 @@ static __always_inline int __copy_from_user(void *dst, const void __user *src, u
 	}
 }	
 
-static __always_inline int __copy_to_user(void __user *dst, const void *src, unsigned size)
+static __always_inline __must_check
+int __copy_to_user(void __user *dst, const void *src, unsigned size)
 { 
        int ret = 0;
 	if (!__builtin_constant_p(size))
@@ -304,7 +309,8 @@ static __always_inline int __copy_to_user(void __user *dst, const void *src, uns
 	}
 }	
 
-static __always_inline int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
+static __always_inline __must_check
+int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 { 
        int ret = 0;
 	if (!__builtin_constant_p(size))
@@ -344,15 +350,17 @@ static __always_inline int __copy_in_user(void __user *dst, const void __user *s
 	}
 }	
 
-long strncpy_from_user(char *dst, const char __user *src, long count);
-long __strncpy_from_user(char *dst, const char __user *src, long count);
-long strnlen_user(const char __user *str, long n);
-long __strnlen_user(const char __user *str, long n);
-long strlen_user(const char __user *str);
-unsigned long clear_user(void __user *mem, unsigned long len);
-unsigned long __clear_user(void __user *mem, unsigned long len);
-
-extern long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
+__must_check long 
+strncpy_from_user(char *dst, const char __user *src, long count);
+__must_check long 
+__strncpy_from_user(char *dst, const char __user *src, long count);
+__must_check long strnlen_user(const char __user *str, long n);
+__must_check long __strnlen_user(const char __user *str, long n);
+__must_check long strlen_user(const char __user *str);
+__must_check unsigned long clear_user(void __user *mem, unsigned long len);
+__must_check unsigned long __clear_user(void __user *mem, unsigned long len);
+
+__must_check long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
 #define __copy_to_user_inatomic copy_user_generic
 
 #endif /* __X86_64_UACCESS_H */
-- 
cgit v1.2.3


From 383d079bfdfcfccd6a720405a510fe325b3e6576 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Fix some stylistic issues in uaccess.h

- Replace some broken white space.
- Replace __ keywords with standard names

No functional changes.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/uaccess.h | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index 802a4a068ef6..e856570c0689 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -84,7 +84,7 @@ struct exception_table_entry
  */
 
 #define __get_user_x(size,ret,x,ptr) \
-	__asm__ __volatile__("call __get_user_" #size \
+	asm volatile("call __get_user_" #size \
 		:"=a" (ret),"=d" (x) \
 		:"c" (ptr) \
 		:"r8")
@@ -101,7 +101,7 @@ struct exception_table_entry
 	case 8:  __get_user_x(8,__ret_gu,__val_gu,ptr); break;		\
 	default: __get_user_bad(); break;				\
 	}								\
-	(x) = (__typeof__(*(ptr)))__val_gu;				\
+	(x) = (typeof(*(ptr)))__val_gu;				\
 	__ret_gu;							\
 })
 
@@ -112,7 +112,7 @@ extern void __put_user_8(void);
 extern void __put_user_bad(void);
 
 #define __put_user_x(size,ret,x,ptr)					\
-	__asm__ __volatile__("call __put_user_" #size			\
+	asm volatile("call __put_user_" #size			\
 		:"=a" (ret)						\
 		:"c" (ptr),"d" (x)					\
 		:"r8")
@@ -139,7 +139,7 @@ extern void __put_user_bad(void);
 #define __put_user_check(x,ptr,size)			\
 ({							\
 	int __pu_err;					\
-	__typeof__(*(ptr)) __user *__pu_addr = (ptr);	\
+	typeof(*(ptr)) __user *__pu_addr = (ptr);	\
 	switch (size) { 				\
 	case 1: __put_user_x(1,__pu_err,x,__pu_addr); break;	\
 	case 2: __put_user_x(2,__pu_err,x,__pu_addr); break;	\
@@ -173,7 +173,7 @@ struct __large_struct { unsigned long buf[100]; };
  * aliasing issues.
  */
 #define __put_user_asm(x, addr, err, itype, rtype, ltype, errno)	\
-	__asm__ __volatile__(					\
+	asm volatile(					\
 		"1:	mov"itype" %"rtype"1,%2\n"		\
 		"2:\n"						\
 		".section .fixup,\"ax\"\n"			\
@@ -193,7 +193,7 @@ struct __large_struct { unsigned long buf[100]; };
 	int __gu_err;						\
 	unsigned long __gu_val;					\
 	__get_user_size(__gu_val,(ptr),(size),__gu_err);	\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
+	(x) = (typeof(*(ptr)))__gu_val;			\
 	__gu_err;						\
 })
 
@@ -217,7 +217,7 @@ do {									\
 } while (0)
 
 #define __get_user_asm(x, addr, err, itype, rtype, ltype, errno)	\
-	__asm__ __volatile__(					\
+	asm volatile(					\
 		"1:	mov"itype" %2,%"rtype"1\n"		\
 		"2:\n"						\
 		".section .fixup,\"ax\"\n"			\
@@ -250,7 +250,7 @@ copy_in_user(void __user *to, const void __user *from, unsigned len);
 static __always_inline __must_check
 int __copy_from_user(void *dst, const void __user *src, unsigned size)
 { 
-       int ret = 0;
+	int ret = 0;
 	if (!__builtin_constant_p(size))
 		return copy_user_generic(dst,(__force void *)src,size);
 	switch (size) { 
@@ -280,7 +280,7 @@ int __copy_from_user(void *dst, const void __user *src, unsigned size)
 static __always_inline __must_check
 int __copy_to_user(void __user *dst, const void *src, unsigned size)
 { 
-       int ret = 0;
+	int ret = 0;
 	if (!__builtin_constant_p(size))
 		return copy_user_generic((__force void *)dst,src,size);
 	switch (size) { 
@@ -312,7 +312,7 @@ int __copy_to_user(void __user *dst, const void *src, unsigned size)
 static __always_inline __must_check
 int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 { 
-       int ret = 0;
+	int ret = 0;
 	if (!__builtin_constant_p(size))
 		return copy_user_generic((__force void *)dst,(__force void *)src,size);
 	switch (size) { 
-- 
cgit v1.2.3


From c1a9d41f4f103bfef2ed0bea1e95b3190e39e448 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Reindent macros in pda.h

Reindent the macros in x86-64 pda.h, making them much more readable.
Follows Jeremy's i386 version of this.

No functional changes

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 85 ++++++++++++++++++++++++++++++------------------
 1 file changed, 53 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 5dadb201f769..9e3aaf74475d 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -41,46 +41,67 @@ extern struct x8664_pda boot_cpu_pda[];
  */ 
 extern void __bad_pda_field(void);
 
-/* proxy_pda doesn't actually exist, but tell gcc it is accessed
-   for all PDA accesses so it gets read/write dependencies right. */
+/*
+ * proxy_pda doesn't actually exist, but tell gcc it is accessed for
+ * all PDA accesses so it gets read/write dependencies right.
+ */
 extern struct x8664_pda _proxy_pda;
 
 #define pda_offset(field) offsetof(struct x8664_pda, field)
 
-#define pda_to_op(op,field,val) do { \
-	typedef typeof(_proxy_pda.field) T__; \
-	if (0) { T__ tmp__; tmp__ = (val); }  \
-	switch (sizeof(_proxy_pda.field)) { 		\
-case 2: \
-asm(op "w %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
-	"ri" ((T__)val),"i"(pda_offset(field))); break; \
-case 4: \
-asm(op "l %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
-	"ri" ((T__)val),"i"(pda_offset(field))); break; \
-case 8: \
-asm(op "q %1,%%gs:%c2": "+m" (_proxy_pda.field) : \
-	 "ri" ((T__)val),"i"(pda_offset(field))); break; \
-default: __bad_pda_field(); 					\
-       } \
+#define pda_to_op(op,field,val) do {		\
+	typedef typeof(_proxy_pda.field) T__;	\
+	if (0) { T__ tmp__; tmp__ = (val); }	/* type checking */ \
+	switch (sizeof(_proxy_pda.field)) {	\
+	case 2:					\
+		asm(op "w %1,%%gs:%c2" : 	\
+		    "+m" (_proxy_pda.field) :	\
+		    "ri" ((T__)val),		\
+		    "i"(pda_offset(field))); 	\
+ 		break;				\
+	case 4:					\
+		asm(op "l %1,%%gs:%c2" : 	\
+		    "+m" (_proxy_pda.field) :	\
+		    "ri" ((T__)val),		\
+		    "i" (pda_offset(field))); 	\
+		break;				\
+	case 8:					\
+		asm(op "q %1,%%gs:%c2": 	\
+		    "+m" (_proxy_pda.field) :	\
+		    "ri" ((T__)val),		\
+		    "i"(pda_offset(field))); 	\
+		break;				\
+       default: 				\
+		__bad_pda_field();		\
+       }					\
        } while (0)
 
-#define pda_from_op(op,field) ({ \
-       typeof(_proxy_pda.field) ret__; \
-       switch (sizeof(_proxy_pda.field)) { 		\
-case 2: \
-asm(op "w %%gs:%c1,%0":"=r" (ret__):\
-	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
-case 4: \
-asm(op "l %%gs:%c1,%0":"=r" (ret__):\
-	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
-case 8: \
-asm(op "q %%gs:%c1,%0":"=r" (ret__):\
-	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
-default: __bad_pda_field(); 					\
-       } \
+#define pda_from_op(op,field) ({		\
+	typeof(_proxy_pda.field) ret__;		\
+	switch (sizeof(_proxy_pda.field)) {	\
+       	case 2:					\
+		asm(op "w %%gs:%c1,%0" : 	\
+		    "=r" (ret__) :		\
+		    "i" (pda_offset(field)), 	\
+		    "m" (_proxy_pda.field)); 	\
+		 break;				\
+	case 4:					\
+		asm(op "l %%gs:%c1,%0":		\
+		    "=r" (ret__):		\
+		    "i" (pda_offset(field)), 	\
+		    "m" (_proxy_pda.field)); 	\
+		 break;				\
+       case 8:					\
+		asm(op "q %%gs:%c1,%0":		\
+		    "=r" (ret__) :		\
+		    "i" (pda_offset(field)), 	\
+		    "m" (_proxy_pda.field)); 	\
+		 break;				\
+       default: 				\
+		__bad_pda_field();		\
+       }					\
        ret__; })
 
-
 #define read_pda(field) pda_from_op("mov",field)
 #define write_pda(field,val) pda_to_op("mov",field,val)
 #define add_pda(field,val) pda_to_op("add",field,val)
-- 
cgit v1.2.3


From fd167e42b237e0688005b3dec380eb5a6e5f3585 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Define __bad_pda_field as noreturn

This quietens so warnings about uninitialized use of the return
value of the pda read operations.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 9e3aaf74475d..531f48a6c3af 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -39,7 +39,7 @@ extern struct x8664_pda boot_cpu_pda[];
  * There is no fast way to get the base address of the PDA, all the accesses
  * have to mention %fs/%gs.  So it needs to be done this Torvaldian way.
  */ 
-extern void __bad_pda_field(void);
+extern void __bad_pda_field(void) __attribute__((noreturn));
 
 /*
  * proxy_pda doesn't actually exist, but tell gcc it is accessed for
-- 
cgit v1.2.3


From 73bb5117a448bdf0b56232ca28451fe4c534cb3a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Remove unused asm-x86_64/mmx.h

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/mmx.h | 14 --------------
 1 file changed, 14 deletions(-)
 delete mode 100644 include/asm-x86_64/mmx.h

(limited to 'include')

diff --git a/include/asm-x86_64/mmx.h b/include/asm-x86_64/mmx.h
deleted file mode 100644
index 46b71da99869..000000000000
--- a/include/asm-x86_64/mmx.h
+++ /dev/null
@@ -1,14 +0,0 @@
-#ifndef _ASM_MMX_H
-#define _ASM_MMX_H
-
-/*
- *	MMX 3Dnow! helper operations
- */
-
-#include <linux/types.h>
- 
-extern void *_mmx_memcpy(void *to, const void *from, size_t size);
-extern void mmx_clear_page(void *page);
-extern void mmx_copy_page(void *to, void *from);
-
-#endif
-- 
cgit v1.2.3


From a15da49debaf7f09460a886b0ecd08588410715e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Fix idle notifiers

Previously exit_idle would be called more often than enter_idle

Now instead of using complicated tests just keep track of it
using the per CPU variable as a flip flop.  I moved the idle state into the
PDA to make the access more efficient.

Original bug report and an initial patch from Stephane Eranian,
but redone by AK.

Cc: Stephane Eranian <eranian@hpl.hp.com>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/process.c | 15 +++++++++------
 include/asm-x86_64/pda.h     |  3 ++-
 2 files changed, 11 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 885c318f76ab..458006ae19f3 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -80,25 +80,25 @@ void idle_notifier_unregister(struct notifier_block *n)
 }
 EXPORT_SYMBOL(idle_notifier_unregister);
 
-enum idle_state { CPU_IDLE, CPU_NOT_IDLE };
-static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE;
-
 void enter_idle(void)
 {
-	__get_cpu_var(idle_state) = CPU_IDLE;
+	write_pda(isidle, 1);
 	atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
 }
 
 static void __exit_idle(void)
 {
-	__get_cpu_var(idle_state) = CPU_NOT_IDLE;
+	if (read_pda(isidle) == 0)
+		return;
+	write_pda(isidle, 0);
 	atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL);
 }
 
 /* Called from interrupts to signify idle end */
 void exit_idle(void)
 {
-	if (current->pid | read_pda(irqcount))
+	/* idle loop has pid 0 */
+	if (current->pid)
 		return;
 	__exit_idle();
 }
@@ -220,6 +220,9 @@ void cpu_idle (void)
 				play_dead();
 			enter_idle();
 			idle();
+			/* In many cases the interrupt that ended idle
+			   has already called exit_idle. But some idle
+			   loops can be woken up without interrupt. */
 			__exit_idle();
 		}
 
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 531f48a6c3af..14996d962bac 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -25,7 +25,8 @@ struct x8664_pda {
 	int nodenumber;		    /* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
-	int mmu_state;     
+	short mmu_state;
+	short isidle;
 	struct mm_struct *active_mm;
 	unsigned apic_timer_irqs;
 } ____cacheline_aligned_in_smp;
-- 
cgit v1.2.3


From 2817716ace8c397685bd702223cc5a8a42c7e997 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] i386: Fix pack_descriptor()

Fix pack_descriptor:
 1. flags are bits 20-23 in the high word
 2. limit's 4 msb are bits 16-19 in the high word

These haven't mattered so far, because all users have had small limits
and a flags setting of 0.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>

===================================================================
---
 include/asm-i386/desc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 5db9e96e8dc1..5874ef119ffd 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -46,7 +46,7 @@ static inline void pack_descriptor(__u32 *a, __u32 *b,
 {
 	*a = ((base & 0xffff) << 16) | (limit & 0xffff);
 	*b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
-	     ((type & 0xff) << 8) | ((flags & 0xf) << 12);
+		(limit & 0x000f0000) | ((type & 0xff) << 8) | ((flags & 0xf) << 20);
 }
 
 static inline void pack_gate(__u32 *a, __u32 *b,
-- 
cgit v1.2.3


From adf1423698f00d00b267f7dca8231340ce7d65ef Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] i386/x86-64: Work around gcc bug with noreturn functions in
 unwinder

Current gcc generates calls not jumps to noreturn functions. When that happens the
return address can point to the next function, which confuses the unwinder.

This patch works around it by marking asynchronous exception
frames in contrast normal call frames in the unwind information.  Then teach
the unwinder to decode this.

For normal call frames the unwinder now subtracts one from the address which avoids
this problem.  The standard libgcc unwinder uses the same trick.

It doesn't include adjustment of the printed address (i.e. for the original
example, it'd still be kernel_math_error+0 that gets displayed, but the
unwinder wouldn't get confused anymore.

This only works with binutils 2.6.17+ and some versions of H.J.Lu's 2.6.16
unfortunately because earlier binutils don't support .cfi_signal_frame

[AK: added automatic detection of the new binutils and wrote description]

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/Makefile           |  4 ++++
 arch/i386/kernel/entry.S     |  4 ++++
 arch/x86_64/Makefile         |  4 ++++
 arch/x86_64/ia32/ia32entry.S |  4 ++++
 arch/x86_64/kernel/entry.S   |  4 ++++
 include/asm-i386/dwarf2.h    |  7 +++++++
 include/asm-i386/unwind.h    |  5 +++++
 include/asm-x86_64/dwarf2.h  |  6 ++++++
 include/asm-x86_64/unwind.h  |  5 +++++
 kernel/unwind.c              | 35 ++++++++++++++++++++++++++++-------
 scripts/Kbuild.include       |  4 ++--
 11 files changed, 73 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/i386/Makefile b/arch/i386/Makefile
index 508cdbeb3a09..7cc0b189b82b 100644
--- a/arch/i386/Makefile
+++ b/arch/i386/Makefile
@@ -50,6 +50,10 @@ CFLAGS				+= $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-op
 cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
+# is .cfi_signal_frame supported too?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+
 CFLAGS += $(cflags-y)
 
 # Default subarch .c files
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 284f2e908ad0..5a63d6fdb70e 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -183,18 +183,21 @@ VM_MASK		= 0x00020000
 
 #define RING0_INT_FRAME \
 	CFI_STARTPROC simple;\
+	CFI_SIGNAL_FRAME;\
 	CFI_DEF_CFA esp, 3*4;\
 	/*CFI_OFFSET cs, -2*4;*/\
 	CFI_OFFSET eip, -3*4
 
 #define RING0_EC_FRAME \
 	CFI_STARTPROC simple;\
+	CFI_SIGNAL_FRAME;\
 	CFI_DEF_CFA esp, 4*4;\
 	/*CFI_OFFSET cs, -2*4;*/\
 	CFI_OFFSET eip, -3*4
 
 #define RING0_PTREGS_FRAME \
 	CFI_STARTPROC simple;\
+	CFI_SIGNAL_FRAME;\
 	CFI_DEF_CFA esp, OLDESP-EBX;\
 	/*CFI_OFFSET cs, CS-OLDESP;*/\
 	CFI_OFFSET eip, EIP-OLDESP;\
@@ -275,6 +278,7 @@ need_resched:
 	# sysenter call handler stub
 ENTRY(sysenter_entry)
 	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA esp, 0
 	CFI_REGISTER esp, ebp
 	movl TSS_sysenter_esp0(%esp),%esp
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 2b8d07c70106..1c0f18d4f887 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -58,6 +58,10 @@ cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
 cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
+# is .cfi_signal_frame supported too?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,)
+
 cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector )
 cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector-all )
 
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 32fd32bea07c..b4aa875e175b 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -71,6 +71,7 @@
  */ 	
 ENTRY(ia32_sysenter_target)
 	CFI_STARTPROC32	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,0
 	CFI_REGISTER	rsp,rbp
 	swapgs
@@ -186,6 +187,7 @@ ENDPROC(ia32_sysenter_target)
  */ 	
 ENTRY(ia32_cstar_target)
 	CFI_STARTPROC32	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,PDA_STACKOFFSET
 	CFI_REGISTER	rip,rcx
 	/*CFI_REGISTER	rflags,r11*/
@@ -293,6 +295,7 @@ ia32_badarg:
 
 ENTRY(ia32_syscall)
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8-RIP
 	/*CFI_REL_OFFSET	ss,SS-RIP*/
 	CFI_REL_OFFSET	rsp,RSP-RIP
@@ -370,6 +373,7 @@ ENTRY(ia32_ptregs_common)
 	popq %r11
 	CFI_ENDPROC
 	CFI_STARTPROC32	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8-ARGOFFSET
 	CFI_REL_OFFSET	rax,RAX-ARGOFFSET
 	CFI_REL_OFFSET	rcx,RCX-ARGOFFSET
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index ea32688386fd..4cbc65290ae7 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -123,6 +123,7 @@
 	.macro	CFI_DEFAULT_STACK start=1
 	.if \start
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8
 	.else
 	CFI_DEF_CFA_OFFSET SS+8
@@ -207,6 +208,7 @@ END(ret_from_fork)
 
 ENTRY(system_call)
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,PDA_STACKOFFSET
 	CFI_REGISTER	rip,rcx
 	/*CFI_REGISTER	rflags,r11*/
@@ -324,6 +326,7 @@ END(system_call)
  */ 	
 ENTRY(int_ret_from_sys_call)
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8-ARGOFFSET
 	/*CFI_REL_OFFSET	ss,SS-ARGOFFSET*/
 	CFI_REL_OFFSET	rsp,RSP-ARGOFFSET
@@ -484,6 +487,7 @@ END(stub_rt_sigreturn)
  */
 	.macro _frame ref
 	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA rsp,SS+8-\ref
 	/*CFI_REL_OFFSET ss,SS-\ref*/
 	CFI_REL_OFFSET rsp,RSP-\ref
diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h
index 5d1a8db5a9b0..6d66398a307d 100644
--- a/include/asm-i386/dwarf2.h
+++ b/include/asm-i386/dwarf2.h
@@ -28,6 +28,12 @@
 #define CFI_RESTORE_STATE .cfi_restore_state
 #define CFI_UNDEFINED .cfi_undefined
 
+#ifdef CONFIG_AS_CFI_SIGNAL_FRAME
+#define CFI_SIGNAL_FRAME .cfi_signal_frame
+#else
+#define CFI_SIGNAL_FRAME
+#endif
+
 #else
 
 /* Due to the structure of pre-exisiting code, don't use assembler line
@@ -48,6 +54,7 @@
 #define CFI_REMEMBER_STATE ignore
 #define CFI_RESTORE_STATE ignore
 #define CFI_UNDEFINED ignore
+#define CFI_SIGNAL_FRAME ignore
 
 #endif
 
diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h
index f0ac399bae3c..5031d693b89d 100644
--- a/include/asm-i386/unwind.h
+++ b/include/asm-i386/unwind.h
@@ -18,6 +18,7 @@ struct unwind_frame_info
 {
 	struct pt_regs regs;
 	struct task_struct *task;
+	unsigned call_frame:1;
 };
 
 #define UNW_PC(frame)        (frame)->regs.eip
@@ -44,6 +45,10 @@ struct unwind_frame_info
 	PTREGS_INFO(edi), \
 	PTREGS_INFO(eip)
 
+#define UNW_DEFAULT_RA(raItem, dataAlign) \
+	((raItem).where == Memory && \
+	 !((raItem).value * (dataAlign) + 4))
+
 static inline void arch_unw_init_frame_info(struct unwind_frame_info *info,
                                             /*const*/ struct pt_regs *regs)
 {
diff --git a/include/asm-x86_64/dwarf2.h b/include/asm-x86_64/dwarf2.h
index 2b9368365fad..eedc08526b0b 100644
--- a/include/asm-x86_64/dwarf2.h
+++ b/include/asm-x86_64/dwarf2.h
@@ -28,6 +28,11 @@
 #define CFI_REMEMBER_STATE .cfi_remember_state
 #define CFI_RESTORE_STATE .cfi_restore_state
 #define CFI_UNDEFINED .cfi_undefined
+#ifdef CONFIG_AS_CFI_SIGNAL_FRAME
+#define CFI_SIGNAL_FRAME .cfi_signal_frame
+#else
+#define CFI_SIGNAL_FRAME
+#endif
 
 #else
 
@@ -45,6 +50,7 @@
 #define CFI_REMEMBER_STATE	#
 #define CFI_RESTORE_STATE	#
 #define CFI_UNDEFINED	#
+#define CFI_SIGNAL_FRAME	#
 
 #endif
 
diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h
index 1f6e9bfb569e..b8fa5cb7ff88 100644
--- a/include/asm-x86_64/unwind.h
+++ b/include/asm-x86_64/unwind.h
@@ -18,6 +18,7 @@ struct unwind_frame_info
 {
 	struct pt_regs regs;
 	struct task_struct *task;
+	unsigned call_frame:1;
 };
 
 #define UNW_PC(frame)        (frame)->regs.rip
@@ -57,6 +58,10 @@ struct unwind_frame_info
 	PTREGS_INFO(r15), \
 	PTREGS_INFO(rip)
 
+#define UNW_DEFAULT_RA(raItem, dataAlign) \
+	((raItem).where == Memory && \
+	 !((raItem).value * (dataAlign) + 8))
+
 static inline void arch_unw_init_frame_info(struct unwind_frame_info *info,
                                             /*const*/ struct pt_regs *regs)
 {
diff --git a/kernel/unwind.c b/kernel/unwind.c
index f69c804c8e62..3430475fcd88 100644
--- a/kernel/unwind.c
+++ b/kernel/unwind.c
@@ -603,6 +603,7 @@ int unwind(struct unwind_frame_info *frame)
 #define FRAME_REG(r, t) (((t *)frame)[reg_info[r].offs])
 	const u32 *fde = NULL, *cie = NULL;
 	const u8 *ptr = NULL, *end = NULL;
+	unsigned long pc = UNW_PC(frame) - frame->call_frame;
 	unsigned long startLoc = 0, endLoc = 0, cfa;
 	unsigned i;
 	signed ptrType = -1;
@@ -612,7 +613,7 @@ int unwind(struct unwind_frame_info *frame)
 
 	if (UNW_PC(frame) == 0)
 		return -EINVAL;
-	if ((table = find_table(UNW_PC(frame))) != NULL
+	if ((table = find_table(pc)) != NULL
 	    && !(table->size & (sizeof(*fde) - 1))) {
 		unsigned long tableSize = table->size;
 
@@ -647,7 +648,7 @@ int unwind(struct unwind_frame_info *frame)
 			                        ptrType & DW_EH_PE_indirect
 			                        ? ptrType
 			                        : ptrType & (DW_EH_PE_FORM|DW_EH_PE_signed));
-			if (UNW_PC(frame) >= startLoc && UNW_PC(frame) < endLoc)
+			if (pc >= startLoc && pc < endLoc)
 				break;
 			cie = NULL;
 		}
@@ -657,16 +658,28 @@ int unwind(struct unwind_frame_info *frame)
 		state.cieEnd = ptr; /* keep here temporarily */
 		ptr = (const u8 *)(cie + 2);
 		end = (const u8 *)(cie + 1) + *cie;
+		frame->call_frame = 1;
 		if ((state.version = *ptr) != 1)
 			cie = NULL; /* unsupported version */
 		else if (*++ptr) {
 			/* check if augmentation size is first (and thus present) */
 			if (*ptr == 'z') {
-				/* check for ignorable (or already handled)
-				 * nul-terminated augmentation string */
-				while (++ptr < end && *ptr)
-					if (strchr("LPR", *ptr) == NULL)
+				while (++ptr < end && *ptr) {
+					switch(*ptr) {
+					/* check for ignorable (or already handled)
+					 * nul-terminated augmentation string */
+					case 'L':
+					case 'P':
+					case 'R':
+						continue;
+					case 'S':
+						frame->call_frame = 0;
+						continue;
+					default:
 						break;
+					}
+					break;
+				}
 			}
 			if (ptr >= end || *ptr)
 				cie = NULL;
@@ -755,7 +768,7 @@ int unwind(struct unwind_frame_info *frame)
 	state.org = startLoc;
 	memcpy(&state.cfa, &badCFA, sizeof(state.cfa));
 	/* process instructions */
-	if (!processCFI(ptr, end, UNW_PC(frame), ptrType, &state)
+	if (!processCFI(ptr, end, pc, ptrType, &state)
 	   || state.loc > endLoc
 	   || state.regs[retAddrReg].where == Nowhere
 	   || state.cfa.reg >= ARRAY_SIZE(reg_info)
@@ -763,6 +776,11 @@ int unwind(struct unwind_frame_info *frame)
 	   || state.cfa.offs % sizeof(unsigned long))
 		return -EIO;
 	/* update frame */
+#ifndef CONFIG_AS_CFI_SIGNAL_FRAME
+	if(frame->call_frame
+	   && !UNW_DEFAULT_RA(state.regs[retAddrReg], state.dataAlign))
+		frame->call_frame = 0;
+#endif
 	cfa = FRAME_REG(state.cfa.reg, unsigned long) + state.cfa.offs;
 	startLoc = min((unsigned long)UNW_SP(frame), cfa);
 	endLoc = max((unsigned long)UNW_SP(frame), cfa);
@@ -866,6 +884,7 @@ int unwind_init_frame_info(struct unwind_frame_info *info,
                            /*const*/ struct pt_regs *regs)
 {
 	info->task = tsk;
+	info->call_frame = 0;
 	arch_unw_init_frame_info(info, regs);
 
 	return 0;
@@ -879,6 +898,7 @@ int unwind_init_blocked(struct unwind_frame_info *info,
                         struct task_struct *tsk)
 {
 	info->task = tsk;
+	info->call_frame = 0;
 	arch_unw_init_blocked(info);
 
 	return 0;
@@ -894,6 +914,7 @@ int unwind_init_running(struct unwind_frame_info *info,
                         void *arg)
 {
 	info->task = current;
+	info->call_frame = 0;
 
 	return arch_unwind_init_running(info, callback, arg);
 }
diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include
index 7adef12a0c26..4f5ff19b992b 100644
--- a/scripts/Kbuild.include
+++ b/scripts/Kbuild.include
@@ -66,8 +66,8 @@ as-option = $(shell if $(CC) $(CFLAGS) $(1) -Wa,-Z -c -o /dev/null \
 # as-instr
 # Usage: cflags-y += $(call as-instr, instr, option1, option2)
 
-as-instr = $(shell if echo -e "$(1)" | $(AS) -Z -o astest$$$$.out \
-		   2>&1 >/dev/null ; then echo "$(2)"; else echo "$(3)"; fi; \
+as-instr = $(shell if echo -e "$(1)" | $(AS) >/dev/null 2>&1 -W -Z -o astest$$$$.out ; \
+		   then echo "$(2)"; else echo "$(3)"; fi; \
 	           rm -f astest$$$$.out)
 
 # cc-option
-- 
cgit v1.2.3


From 658fdbef66e5e9be79b457edc2cbbb3add840aa9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Don't leak NT bit into next task

SYSENTER can cause a NT to be set which might cause crashes on the IRET
in the next task.

Following similar i386 patch from Linus.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S   | 4 ++++
 arch/x86_64/kernel/setup64.c | 4 ++++
 include/asm-x86_64/system.h  | 5 +++--
 3 files changed, 11 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 4cbc65290ae7..2802524104f3 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -155,6 +155,10 @@
 /* rdi:	prev */	
 ENTRY(ret_from_fork)
 	CFI_DEFAULT_STACK
+	push kernel_eflags(%rip)
+	CFI_ADJUST_CFA_OFFSET 4
+	popf				# reset kernel eflags
+	CFI_ADJUST_CFA_OFFSET -4
 	call schedule_tail
 	GET_THREAD_INFO(%rcx)
 	testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx)
diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 9332d2361e08..8c4b80fe71a1 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -180,6 +180,8 @@ void __cpuinit check_efer(void)
         }       
 }
 
+unsigned long kernel_eflags;
+
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
  * initialized (naturally) in the bootstrap process, such as the GDT
@@ -281,4 +283,6 @@ void __cpuinit cpu_init (void)
 	set_debugreg(0UL, 7);
 
 	fpu_init(); 
+
+	raw_local_save_flags(kernel_eflags);
 }
diff --git a/include/asm-x86_64/system.h b/include/asm-x86_64/system.h
index 6bf170bceae1..bd376bc8c4ab 100644
--- a/include/asm-x86_64/system.h
+++ b/include/asm-x86_64/system.h
@@ -14,12 +14,13 @@
 #define __RESTORE(reg,offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
 
 /* frame pointer must be last for get_wchan */
-#define SAVE_CONTEXT    "pushq %%rbp ; movq %%rsi,%%rbp\n\t"
-#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp\n\t"
+#define SAVE_CONTEXT    "pushf ; pushq %%rbp ; movq %%rsi,%%rbp\n\t"
+#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp ; popf\t"
 
 #define __EXTRA_CLOBBER  \
 	,"rcx","rbx","rdx","r8","r9","r10","r11","r12","r13","r14","r15"
 
+/* Save restore flags to clear handle leaking NT */
 #define switch_to(prev,next,last) \
 	asm volatile(SAVE_CONTEXT						    \
 		     "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */	  \
-- 
cgit v1.2.3


From 8f60774a116ced9b73ae3913d511687889efe725 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] x86: Move direct PCI scanning functions out of line

Saves about 200 bytes of code space.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/pci/Makefile          |  2 +-
 arch/i386/pci/early.c           | 44 +++++++++++++++++++++++++++++++++++++++++
 arch/x86_64/pci/Makefile        |  3 ++-
 include/asm-x86_64/pci-direct.h | 42 ++++-----------------------------------
 4 files changed, 51 insertions(+), 40 deletions(-)
 create mode 100644 arch/i386/pci/early.c

(limited to 'include')

diff --git a/arch/i386/pci/Makefile b/arch/i386/pci/Makefile
index 62ad75c57e6a..1594d2f55c8f 100644
--- a/arch/i386/pci/Makefile
+++ b/arch/i386/pci/Makefile
@@ -11,4 +11,4 @@ pci-y				+= legacy.o irq.o
 pci-$(CONFIG_X86_VISWS)		:= visws.o fixup.o
 pci-$(CONFIG_X86_NUMAQ)		:= numa.o irq.o
 
-obj-y				+= $(pci-y) common.o
+obj-y				+= $(pci-y) common.o early.o
diff --git a/arch/i386/pci/early.c b/arch/i386/pci/early.c
new file mode 100644
index 000000000000..b1f7f40d809b
--- /dev/null
+++ b/arch/i386/pci/early.c
@@ -0,0 +1,44 @@
+#include <linux/kernel.h>
+#include <asm/pci-direct.h>
+#include <asm/io.h>
+
+/* Direct PCI access. This is used for PCI accesses in early boot before
+   the PCI subsystem works. */
+
+#define PDprintk(x...)
+
+u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset)
+{
+	u32 v;
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	v = inl(0xcfc);
+	if (v != 0xffffffff)
+		PDprintk("%x reading 4 from %x: %x\n", slot, offset, v);
+	return v;
+}
+
+u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset)
+{
+	u8 v;
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	v = inb(0xcfc + (offset&3));
+	PDprintk("%x reading 1 from %x: %x\n", slot, offset, v);
+	return v;
+}
+
+u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset)
+{
+	u16 v;
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	v = inw(0xcfc + (offset&2));
+	PDprintk("%x reading 2 from %x: %x\n", slot, offset, v);
+	return v;
+}
+
+void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
+				    u32 val)
+{
+	PDprintk("%x writing to %x: %x\n", slot, offset, val);
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	outl(val, 0xcfc);
+}
diff --git a/arch/x86_64/pci/Makefile b/arch/x86_64/pci/Makefile
index a3f6ad570179..1eb18f421edf 100644
--- a/arch/x86_64/pci/Makefile
+++ b/arch/x86_64/pci/Makefile
@@ -9,7 +9,7 @@ obj-y		:= i386.o
 obj-$(CONFIG_PCI_DIRECT)+= direct.o
 obj-y		+= fixup.o init.o
 obj-$(CONFIG_ACPI)	+= acpi.o
-obj-y			+= legacy.o irq.o common.o
+obj-y			+= legacy.o irq.o common.o early.o
 # mmconfig has a 64bit special
 obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o direct.o
 
@@ -23,3 +23,4 @@ common-y += ../../i386/pci/common.o
 fixup-y  += ../../i386/pci/fixup.o
 i386-y  += ../../i386/pci/i386.o
 init-y += ../../i386/pci/init.o
+early-y += ../../i386/pci/early.o
diff --git a/include/asm-x86_64/pci-direct.h b/include/asm-x86_64/pci-direct.h
index 036b6ca5b53b..9d916cdaa18e 100644
--- a/include/asm-x86_64/pci-direct.h
+++ b/include/asm-x86_64/pci-direct.h
@@ -2,47 +2,13 @@
 #define ASM_PCI_DIRECT_H 1
 
 #include <linux/types.h>
-#include <asm/io.h>
 
 /* Direct PCI access. This is used for PCI accesses in early boot before
    the PCI subsystem works. */ 
 
-#define PDprintk(x...)
-
-static inline u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset)
-{
-	u32 v; 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	v = inl(0xcfc); 
-	if (v != 0xffffffff)
-		PDprintk("%x reading 4 from %x: %x\n", slot, offset, v);
-	return v;
-}
-
-static inline u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset)
-{
-	u8 v; 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	v = inb(0xcfc + (offset&3)); 
-	PDprintk("%x reading 1 from %x: %x\n", slot, offset, v);
-	return v;
-}
-
-static inline u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset)
-{
-	u16 v; 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	v = inw(0xcfc + (offset&2)); 
-	PDprintk("%x reading 2 from %x: %x\n", slot, offset, v);
-	return v;
-}
-
-static inline void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
-				    u32 val)
-{
-	PDprintk("%x writing to %x: %x\n", slot, offset, val); 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	outl(val, 0xcfc); 
-}
+extern u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset);
+extern u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset);
+extern u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset);
+extern void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset, u32 val);
 
 #endif
-- 
cgit v1.2.3


From 0637a70a5db98182d9ad3d6ae1ee30acf20afde9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] x86: Allow disabling early pci scans with pci=noearly or
 disallowing conf1

Some buggy systems can machine check when config space accesses
happen for some non existent devices.  i386/x86-64 do some early
device scans that might trigger this. Allow pci=noearly to disable
this. Also when type 1 is disabling also don't do any early
accesses which are always type1.

This moves the pci= configuration parsing to be a early parameter.
I don't think this can break anything because it only changes
a single global that is only used by PCI.

Cc: gregkh@suse.de
Cc: Trammell Hudson <hudson@osresearch.net>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/kernel-parameters.txt | 6 +++++-
 arch/i386/kernel/acpi/earlyquirk.c  | 6 +++++-
 arch/i386/pci/common.c              | 4 ++++
 arch/i386/pci/early.c               | 8 ++++++++
 arch/i386/pci/pci.h                 | 1 +
 arch/x86_64/kernel/aperture.c       | 2 +-
 arch/x86_64/kernel/early-quirks.c   | 4 ++++
 arch/x86_64/kernel/pci-calgary.c    | 3 +++
 arch/x86_64/kernel/vsmp.c           | 3 +++
 arch/x86_64/mm/k8topology.c         | 3 +++
 drivers/pci/pci.c                   | 5 ++---
 include/asm-x86_64/pci-direct.h     | 2 ++
 12 files changed, 41 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 71d05f481727..4ae99f6e9938 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1240,7 +1240,11 @@ running once the system is up.
 				bootloader. This is currently used on
 				IXP2000 systems where the bus has to be
 				configured a certain way for adjunct CPUs.
-
+		noearly		[X86] Don't do any early type 1 scanning.
+				This might help on some broken boards which
+				machine check when some devices' config space
+				is read. But various workarounds are disabled
+				and some IOMMU drivers will not work.
 	pcmv=		[HW,PCMCIA] BadgePAD 4
 
 	pd.		[PARIDE]
diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c
index 1649a175a206..fe799b11ac0a 100644
--- a/arch/i386/kernel/acpi/earlyquirk.c
+++ b/arch/i386/kernel/acpi/earlyquirk.c
@@ -48,7 +48,11 @@ void __init check_acpi_pci(void)
 	int num, slot, func;
 
 	/* Assume the machine supports type 1. If not it will 
-	   always read ffffffff and should not have any side effect. */
+	   always read ffffffff and should not have any side effect.
+	   Actually a few buggy systems can machine check. Allow the user
+	   to disable it by command line option at least -AK */
+	if (!early_pci_allowed())
+		return;
 
 	/* Poor man's PCI discovery */
 	for (num = 0; num < 32; num++) {
diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c
index 0a362e3aeac5..68bce194e688 100644
--- a/arch/i386/pci/common.c
+++ b/arch/i386/pci/common.c
@@ -242,6 +242,10 @@ char * __devinit  pcibios_setup(char *str)
 		acpi_noirq_set();
 		return NULL;
 	}
+	else if (!strcmp(str, "noearly")) {
+		pci_probe |= PCI_PROBE_NOEARLY;
+		return NULL;
+	}
 #ifndef CONFIG_X86_VISWS
 	else if (!strcmp(str, "usepirqmask")) {
 		pci_probe |= PCI_USE_PIRQ_MASK;
diff --git a/arch/i386/pci/early.c b/arch/i386/pci/early.c
index b1f7f40d809b..713d6c866cae 100644
--- a/arch/i386/pci/early.c
+++ b/arch/i386/pci/early.c
@@ -1,6 +1,8 @@
 #include <linux/kernel.h>
+#include <linux/pci.h>
 #include <asm/pci-direct.h>
 #include <asm/io.h>
+#include "pci.h"
 
 /* Direct PCI access. This is used for PCI accesses in early boot before
    the PCI subsystem works. */
@@ -42,3 +44,9 @@ void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
 	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
 	outl(val, 0xcfc);
 }
+
+int early_pci_allowed(void)
+{
+	return (pci_probe & (PCI_PROBE_CONF1|PCI_PROBE_NOEARLY)) ==
+			PCI_PROBE_CONF1;
+}
diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h
index 8a7cf1f23684..1814f74569c6 100644
--- a/arch/i386/pci/pci.h
+++ b/arch/i386/pci/pci.h
@@ -17,6 +17,7 @@
 #define PCI_PROBE_CONF2		0x0004
 #define PCI_PROBE_MMCONF	0x0008
 #define PCI_PROBE_MASK		0x000f
+#define PCI_PROBE_NOEARLY	0x0010
 
 #define PCI_NO_SORT		0x0100
 #define PCI_BIOS_SORT		0x0200
diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index f851e1f9c82a..b487396c4c5b 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -212,7 +212,7 @@ void __init iommu_hole_init(void)
 	u64 aper_base, last_aper_base = 0;
 	int valid_agp = 0;
 
-	if (iommu_aperture_disabled || !fix_aperture)
+	if (iommu_aperture_disabled || !fix_aperture || !early_pci_allowed())
 		return;
 
 	printk("Checking aperture...\n"); 
diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c
index d637cff1c4b1..208e38a372c1 100644
--- a/arch/x86_64/kernel/early-quirks.c
+++ b/arch/x86_64/kernel/early-quirks.c
@@ -82,6 +82,10 @@ static struct chipset early_qrk[] = {
 void __init early_quirks(void)
 {
 	int num, slot, func;
+
+	if (!early_pci_allowed())
+		return;
+
 	/* Poor man's PCI discovery */
 	for (num = 0; num < 32; num++) {
 		for (slot = 0; slot < 32; slot++) {
diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 466588f95601..cfb09b07ae99 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -924,6 +924,9 @@ void __init detect_calgary(void)
 	if (swiotlb || no_iommu || iommu_detected)
 		return;
 
+	if (!early_pci_allowed())
+		return;
+
 	specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE);
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
diff --git a/arch/x86_64/kernel/vsmp.c b/arch/x86_64/kernel/vsmp.c
index 92f70c74965f..044e852bd25e 100644
--- a/arch/x86_64/kernel/vsmp.c
+++ b/arch/x86_64/kernel/vsmp.c
@@ -20,6 +20,9 @@ static int __init vsmp_init(void)
 	void *address;
 	unsigned int cap, ctl;
 
+	if (!early_pci_allowed())
+		return 0;
+
 	/* Check if we are running on a ScaleMP vSMP box */
 	if ((read_pci_config_16(0, 0x1f, 0, PCI_VENDOR_ID) != PCI_VENDOR_ID_SCALEMP) ||
 	    (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) != PCI_DEVICE_ID_SCALEMP_VSMP_CTL))
diff --git a/arch/x86_64/mm/k8topology.c b/arch/x86_64/mm/k8topology.c
index 7c45c2d2b8b2..5cf594f9230d 100644
--- a/arch/x86_64/mm/k8topology.c
+++ b/arch/x86_64/mm/k8topology.c
@@ -54,6 +54,9 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 
 	nodes_clear(nodes_parsed);
 
+	if (!early_pci_allowed())
+		return -1;
+
 	nb = find_northbridge(); 
 	if (nb < 0) 
 		return nb;
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 9f79dd6d51ab..684deb6b03aa 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -953,13 +953,12 @@ static int __devinit pci_setup(char *str)
 		}
 		str = k;
 	}
-	return 1;
+	return 0;
 }
+early_param("pci", pci_setup);
 
 device_initcall(pci_init);
 
-__setup("pci=", pci_setup);
-
 #if defined(CONFIG_ISA) || defined(CONFIG_EISA)
 /* FIXME: Some boxes have multiple ISA bridges! */
 struct pci_dev *isa_bridge;
diff --git a/include/asm-x86_64/pci-direct.h b/include/asm-x86_64/pci-direct.h
index 9d916cdaa18e..eba9cb471df3 100644
--- a/include/asm-x86_64/pci-direct.h
+++ b/include/asm-x86_64/pci-direct.h
@@ -11,4 +11,6 @@ extern u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset);
 extern u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset);
 extern void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset, u32 val);
 
+extern int early_pci_allowed(void);
+
 #endif
-- 
cgit v1.2.3


From b89ebd0b0a65d5371aa9ad98e873c4616056ca68 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] Fix unwinder warning in traps.c

Fix

linux/arch/x86_64/kernel/traps.c: In function 'dump_trace':
linux/arch/x86_64/kernel/traps.c:275: warning: cast to pointer from integer of different size

with allnoconfig

Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/unwind.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h
index b8fa5cb7ff88..2e7ff10fd775 100644
--- a/include/asm-x86_64/unwind.h
+++ b/include/asm-x86_64/unwind.h
@@ -99,8 +99,8 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info)
 
 #else
 
-#define UNW_PC(frame) ((void)(frame), 0)
-#define UNW_SP(frame) ((void)(frame), 0)
+#define UNW_PC(frame) ((void)(frame), 0UL)
+#define UNW_SP(frame) ((void)(frame), 0UL)
 
 static inline int arch_unw_user_mode(const void *info)
 {
-- 
cgit v1.2.3


From 3b171672831b9633c2ed8fa94805255cd4d5af19 Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] Add 64bit jiffies compares (for use with get_jiffies_64)

The current time_before/time_after macros will fail typechecks
when passed u64 values (as returned by get_jiffies_64()). On 64bit
systems, this will just result in a warning about mismatching types
without explicit casts, but since unsigned long and u64
(unsigned long long) are of same size, it will still work.
On 32bit systems, a long is 32bits, so the value from get_jiffies_64()
will be truncated by the cast and thus lose all the precision gained by
64bit jiffies.

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/linux/jiffies.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to 'include')

diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h
index 329ebcffa106..c8d5f207c3d4 100644
--- a/include/linux/jiffies.h
+++ b/include/linux/jiffies.h
@@ -115,6 +115,21 @@ static inline u64 get_jiffies_64(void)
 	 ((long)(a) - (long)(b) >= 0))
 #define time_before_eq(a,b)	time_after_eq(b,a)
 
+/* Same as above, but does so with platform independent 64bit types.
+ * These must be used when utilizing jiffies_64 (i.e. return value of
+ * get_jiffies_64() */
+#define time_after64(a,b)	\
+	(typecheck(__u64, a) &&	\
+	 typecheck(__u64, b) && \
+	 ((__s64)(b) - (__s64)(a) < 0))
+#define time_before64(a,b)	time_after64(b,a)
+
+#define time_after_eq64(a,b)	\
+	(typecheck(__u64, a) && \
+	 typecheck(__u64, b) && \
+	 ((__s64)(a) - (__s64)(b) >= 0))
+#define time_before_eq64(a,b)	time_after_eq64(b,a)
+
 /*
  * Have the 32 bit jiffies value wrap 5 minutes after boot
  * so jiffies wrap bugs show up earlier.
-- 
cgit v1.2.3


From 15d5f8398311f565682959daaca30e3ca7aea600 Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] x86: Refactor thermal throttle processing

Refactor the event processing (syslog messaging and rate limiting)
into separate file therm_throt.c. This allows consistent reporting
of CPU thermal throttle events.

After ACK'ing the interrupt, if the event is current, the user
(p4.c/mce_intel.c) calls therm_throt_process to log (and rate limit)
the event. If that function returns 1, the user has the option to log
things further (such as to mce_log in x86_64).

AK: minor cleanup

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/mcheck/Makefile      |  2 +-
 arch/i386/kernel/cpu/mcheck/p4.c          | 23 ++++--------
 arch/i386/kernel/cpu/mcheck/therm_throt.c | 58 +++++++++++++++++++++++++++++++
 arch/x86_64/kernel/Makefile               |  4 +--
 arch/x86_64/kernel/mce.c                  | 27 ++++++++++++++
 arch/x86_64/kernel/mce_intel.c            | 27 ++++----------
 include/asm-i386/therm_throt.h            |  6 ++++
 include/asm-x86_64/mce.h                  |  2 ++
 include/asm-x86_64/therm_throt.h          |  1 +
 9 files changed, 109 insertions(+), 41 deletions(-)
 create mode 100644 arch/i386/kernel/cpu/mcheck/therm_throt.c
 create mode 100644 include/asm-i386/therm_throt.h
 create mode 100644 include/asm-x86_64/therm_throt.h

(limited to 'include')

diff --git a/arch/i386/kernel/cpu/mcheck/Makefile b/arch/i386/kernel/cpu/mcheck/Makefile
index 30808f3d6715..f1ebe1c1c17a 100644
--- a/arch/i386/kernel/cpu/mcheck/Makefile
+++ b/arch/i386/kernel/cpu/mcheck/Makefile
@@ -1,2 +1,2 @@
-obj-y	=	mce.o k7.o p4.o p5.o p6.o winchip.o
+obj-y	=	mce.o k7.o p4.o p5.o p6.o winchip.o therm_throt.o
 obj-$(CONFIG_X86_MCE_NONFATAL)	+=	non-fatal.o
diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c
index b95f1b3d53aa..d83a669d376f 100644
--- a/arch/i386/kernel/cpu/mcheck/p4.c
+++ b/arch/i386/kernel/cpu/mcheck/p4.c
@@ -13,6 +13,8 @@
 #include <asm/msr.h>
 #include <asm/apic.h>
 
+#include <asm/therm_throt.h>
+
 #include "mce.h"
 
 /* as supported by the P4/Xeon family */
@@ -44,25 +46,12 @@ static void unexpected_thermal_interrupt(struct pt_regs *regs)
 /* P4/Xeon Thermal transition interrupt handler */
 static void intel_thermal_interrupt(struct pt_regs *regs)
 {
-	u32 l, h;
-	unsigned int cpu = smp_processor_id();
-	static unsigned long next[NR_CPUS];
+	__u64 msr_val;
 
 	ack_APIC_irq();
 
-	if (time_after(next[cpu], jiffies))
-		return;
-
-	next[cpu] = jiffies + HZ*5;
-	rdmsr(MSR_IA32_THERM_STATUS, l, h);
-	if (l & 0x1) {
-		printk(KERN_EMERG "CPU%d: Temperature above threshold\n", cpu);
-		printk(KERN_EMERG "CPU%d: Running in modulated clock mode\n",
-				cpu);
-		add_taint(TAINT_MACHINE_CHECK);
-	} else {
-		printk(KERN_INFO "CPU%d: Temperature/speed normal\n", cpu);
-	}
+	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
+	therm_throt_process(msr_val & 0x1);
 }
 
 /* Thermal interrupt handler for this CPU setup */
@@ -122,7 +111,7 @@ static void intel_init_thermal(struct cpuinfo_x86 *c)
 	
 	rdmsr (MSR_IA32_MISC_ENABLE, l, h);
 	wrmsr (MSR_IA32_MISC_ENABLE, l | (1<<3), h);
-	
+
 	l = apic_read (APIC_LVTTHMR);
 	apic_write_around (APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 	printk (KERN_INFO "CPU%d: Thermal monitoring enabled\n", cpu);
diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c
new file mode 100644
index 000000000000..85eba00d680e
--- /dev/null
+++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c
@@ -0,0 +1,58 @@
+/*
+ * linux/arch/i386/kerne/cpu/mcheck/therm_throt.c
+ *
+ * Thermal throttle event support code.
+ *
+ * Author: Dmitriy Zavin (dmitriyz@google.com)
+ *
+ * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c.
+ *
+ */
+
+#include <linux/percpu.h>
+#include <linux/cpu.h>
+#include <asm/cpu.h>
+#include <linux/notifier.h>
+#include <asm/therm_throt.h>
+
+/* How long to wait between reporting thermal events */
+#define CHECK_INTERVAL              (300 * HZ)
+
+static DEFINE_PER_CPU(unsigned long, next_check);
+
+/***
+ * therm_throt_process - Process thermal throttling event
+ * @curr: Whether the condition is current or not (boolean), since the
+ *        thermal interrupt normally gets called both when the thermal
+ *        event begins and once the event has ended.
+ *
+ * This function is normally called by the thermal interrupt after the
+ * IRQ has been acknowledged.
+ *
+ * It will take care of rate limiting and printing messages to the syslog.
+ *
+ * Returns: 0 : Event should NOT be further logged, i.e. still in
+ *              "timeout" from previous log message.
+ *          1 : Event should be logged further, and a message has been
+ *              printed to the syslog.
+ */
+int therm_throt_process(int curr)
+{
+	unsigned int cpu = smp_processor_id();
+
+	if (time_before(jiffies, __get_cpu_var(next_check)))
+		return 0;
+
+	__get_cpu_var(next_check) = jiffies + CHECK_INTERVAL;
+
+	/* if we just entered the thermal event */
+	if (curr) {
+		printk(KERN_CRIT "CPU%d: Temperature above threshold, "
+		       "cpu clock throttled\n", cpu);
+		add_taint(TAINT_MACHINE_CHECK);
+	} else {
+		printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu);
+	}
+
+	return 1;
+}
diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 2466fbd035ee..3c7cbff04d3d 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -11,7 +11,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		pci-dma.o pci-nommu.o alternative.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
-obj-$(CONFIG_X86_MCE)         += mce.o
+obj-$(CONFIG_X86_MCE)		+= mce.o therm_throt.o
 obj-$(CONFIG_X86_MCE_INTEL)	+= mce_intel.o
 obj-$(CONFIG_X86_MCE_AMD)	+= mce_amd.o
 obj-$(CONFIG_MTRR)		+= ../../i386/kernel/cpu/mtrr/
@@ -46,6 +46,7 @@ obj-y				+= intel_cacheinfo.o
 
 CFLAGS_vsyscall.o		:= $(PROFILING) -g0
 
+therm_throt-y                   += ../../i386/kernel/cpu/mcheck/therm_throt.o
 bootflag-y			+= ../../i386/kernel/bootflag.o
 cpuid-$(subst m,y,$(CONFIG_X86_CPUID))  += ../../i386/kernel/cpuid.o
 topology-y                     += ../../i386/kernel/topology.o
@@ -55,4 +56,3 @@ quirks-y			+= ../../i386/kernel/quirks.o
 i8237-y				+= ../../i386/kernel/i8237.o
 msr-$(subst m,y,$(CONFIG_X86_MSR))  += ../../i386/kernel/msr.o
 alternative-y			+= ../../i386/kernel/alternative.o
-
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 1a93c3738404..bbea88801d88 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -274,6 +274,33 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 	atomic_dec(&mce_entry);
 }
 
+#ifdef CONFIG_X86_MCE_INTEL
+/***
+ * mce_log_therm_throt_event - Logs the thermal throttling event to mcelog
+ * @cpu: The CPU on which the event occured.
+ * @status: Event status information
+ *
+ * This function should be called by the thermal interrupt after the
+ * event has been processed and the decision was made to log the event
+ * further.
+ *
+ * The status parameter will be saved to the 'status' field of 'struct mce'
+ * and historically has been the register value of the
+ * MSR_IA32_THERMAL_STATUS (Intel) msr.
+ */
+void mce_log_therm_throt_event(unsigned int cpu, __u64 status)
+{
+	struct mce m;
+
+	memset(&m, 0, sizeof(m));
+	m.cpu = cpu;
+	m.bank = MCE_THERMAL_BANK;
+	m.status = status;
+	rdtscll(m.tsc);
+	mce_log(&m);
+}
+#endif /* CONFIG_X86_MCE_INTEL */
+
 /*
  * Periodic polling timer for "silent" machine check errors.
  */
diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c
index 8f533d2c40cb..dec11219e278 100644
--- a/arch/x86_64/kernel/mce_intel.c
+++ b/arch/x86_64/kernel/mce_intel.c
@@ -11,36 +11,21 @@
 #include <asm/mce.h>
 #include <asm/hw_irq.h>
 #include <asm/idle.h>
-
-static DEFINE_PER_CPU(unsigned long, next_check);
+#include <asm/therm_throt.h>
 
 asmlinkage void smp_thermal_interrupt(void)
 {
-	struct mce m;
+	__u64 msr_val;
 
 	ack_APIC_irq();
 
 	exit_idle();
 	irq_enter();
-	if (time_before(jiffies, __get_cpu_var(next_check)))
-		goto done;
-
-	__get_cpu_var(next_check) = jiffies + HZ*300;
-	memset(&m, 0, sizeof(m));
-	m.cpu = smp_processor_id();
-	m.bank = MCE_THERMAL_BANK;
-	rdtscll(m.tsc);
-	rdmsrl(MSR_IA32_THERM_STATUS, m.status);
-	if (m.status & 0x1) {
-		printk(KERN_EMERG
-			"CPU%d: Temperature above threshold, cpu clock throttled\n", m.cpu);
-		add_taint(TAINT_MACHINE_CHECK);
-	} else {
-		printk(KERN_EMERG "CPU%d: Temperature/speed normal\n", m.cpu);
-	}
 
-	mce_log(&m);
-done:
+	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
+	if (therm_throt_process(msr_val & 1))
+		mce_log_therm_throt_event(smp_processor_id(), msr_val);
+
 	irq_exit();
 }
 
diff --git a/include/asm-i386/therm_throt.h b/include/asm-i386/therm_throt.h
new file mode 100644
index 000000000000..3c9c22cc10c9
--- /dev/null
+++ b/include/asm-i386/therm_throt.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_I386_THERM_THROT_H__
+#define __ASM_I386_THERM_THROT_H__ 1
+
+int therm_throt_process(int curr);
+
+#endif /* __ASM_I386_THERM_THROT_H__ */
diff --git a/include/asm-x86_64/mce.h b/include/asm-x86_64/mce.h
index d13687dfd691..5a11146d6d9c 100644
--- a/include/asm-x86_64/mce.h
+++ b/include/asm-x86_64/mce.h
@@ -99,6 +99,8 @@ static inline void mce_amd_feature_init(struct cpuinfo_x86 *c)
 }
 #endif
 
+void mce_log_therm_throt_event(unsigned int cpu, __u64 status);
+
 extern atomic_t mce_entry;
 
 #endif
diff --git a/include/asm-x86_64/therm_throt.h b/include/asm-x86_64/therm_throt.h
new file mode 100644
index 000000000000..5aac059007ba
--- /dev/null
+++ b/include/asm-x86_64/therm_throt.h
@@ -0,0 +1 @@
+#include <asm-i386/therm_throt.h>
-- 
cgit v1.2.3


From 3222b36f46c22f46697a0a53fa8804153a32669f Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] x86: Add a cumulative thermal throttle event counter.

The counter is exported to /sys that keeps track of the
number of thermal events, such that the user knows how bad the
thermal problem might be (since the logging to syslog and mcelog
is rate limited).

AK: Fixed cpu hotplug locking

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/mcheck/p4.c          |   3 +
 arch/i386/kernel/cpu/mcheck/therm_throt.c | 133 ++++++++++++++++++++++++++++--
 arch/x86_64/kernel/mce_intel.c            |   3 +
 include/asm-i386/therm_throt.h            |   3 +
 4 files changed, 136 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c
index d83a669d376f..504434a46011 100644
--- a/arch/i386/kernel/cpu/mcheck/p4.c
+++ b/arch/i386/kernel/cpu/mcheck/p4.c
@@ -115,6 +115,9 @@ static void intel_init_thermal(struct cpuinfo_x86 *c)
 	l = apic_read (APIC_LVTTHMR);
 	apic_write_around (APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 	printk (KERN_INFO "CPU%d: Thermal monitoring enabled\n", cpu);
+
+	/* enable thermal throttle processing */
+	atomic_set(&therm_throt_en, 1);
 	return;
 }
 #endif /* CONFIG_X86_MCE_P4THERMAL */
diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c
index 101f7ace00ce..4f43047de406 100644
--- a/arch/i386/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c
@@ -1,15 +1,22 @@
 /*
  * linux/arch/i386/kerne/cpu/mcheck/therm_throt.c
  *
- * Thermal throttle event support code.
+ * Thermal throttle event support code (such as syslog messaging and rate
+ * limiting) that was factored out from x86_64 (mce_intel.c) and i386 (p4.c).
+ * This allows consistent reporting of CPU thermal throttle events.
+ *
+ * Maintains a counter in /sys that keeps track of the number of thermal
+ * events, such that the user knows how bad the thermal problem might be
+ * (since the logging to syslog and mcelog is rate limited).
  *
  * Author: Dmitriy Zavin (dmitriyz@google.com)
  *
  * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c.
- *
+ *          Inspired by Ross Biro's and Al Borchers' counter code.
  */
 
 #include <linux/percpu.h>
+#include <linux/sysdev.h>
 #include <linux/cpu.h>
 #include <asm/cpu.h>
 #include <linux/notifier.h>
@@ -18,15 +25,53 @@
 /* How long to wait between reporting thermal events */
 #define CHECK_INTERVAL              (300 * HZ)
 
-static DEFINE_PER_CPU(__u64, next_check);
+static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES;
+static DEFINE_PER_CPU(unsigned long, thermal_throttle_count);
+atomic_t therm_throt_en = ATOMIC_INIT(0);
+
+#ifdef CONFIG_SYSFS
+#define define_therm_throt_sysdev_one_ro(_name)                              \
+        static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL)
+
+#define define_therm_throt_sysdev_show_func(name)                            \
+static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev,        \
+                                              char *buf)                     \
+{                                                                            \
+	unsigned int cpu = dev->id;                                          \
+	ssize_t ret;                                                         \
+                                                                             \
+	preempt_disable();              /* CPU hotplug */                    \
+	if (cpu_online(cpu))                                                 \
+		ret = sprintf(buf, "%lu\n",                                  \
+			      per_cpu(thermal_throttle_##name, cpu));        \
+	else                                                                 \
+		ret = 0;                                                     \
+	preempt_enable();                                                    \
+                                                                             \
+	return ret;                                                          \
+}
+
+define_therm_throt_sysdev_show_func(count);
+define_therm_throt_sysdev_one_ro(count);
+
+static struct attribute *thermal_throttle_attrs[] = {
+	&attr_count.attr,
+	NULL
+};
+
+static struct attribute_group thermal_throttle_attr_group = {
+	.attrs = thermal_throttle_attrs,
+	.name = "thermal_throttle"
+};
+#endif /* CONFIG_SYSFS */
 
 /***
- * therm_throt_process - Process thermal throttling event
+ * therm_throt_process - Process thermal throttling event from interrupt
  * @curr: Whether the condition is current or not (boolean), since the
  *        thermal interrupt normally gets called both when the thermal
  *        event begins and once the event has ended.
  *
- * This function is normally called by the thermal interrupt after the
+ * This function is called by the thermal interrupt after the
  * IRQ has been acknowledged.
  *
  * It will take care of rate limiting and printing messages to the syslog.
@@ -41,6 +86,9 @@ int therm_throt_process(int curr)
 	unsigned int cpu = smp_processor_id();
 	__u64 tmp_jiffs = get_jiffies_64();
 
+	if (curr)
+		__get_cpu_var(thermal_throttle_count)++;
+
 	if (time_before64(tmp_jiffs, __get_cpu_var(next_check)))
 		return 0;
 
@@ -49,7 +97,9 @@ int therm_throt_process(int curr)
 	/* if we just entered the thermal event */
 	if (curr) {
 		printk(KERN_CRIT "CPU%d: Temperature above threshold, "
-		       "cpu clock throttled\n", cpu);
+		       "cpu clock throttled (total events = %lu)\n", cpu,
+		       __get_cpu_var(thermal_throttle_count));
+
 		add_taint(TAINT_MACHINE_CHECK);
 	} else {
 		printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu);
@@ -57,3 +107,74 @@ int therm_throt_process(int curr)
 
 	return 1;
 }
+
+#ifdef CONFIG_SYSFS
+/* Add/Remove thermal_throttle interface for CPU device */
+static __cpuinit int thermal_throttle_add_dev(struct sys_device * sys_dev)
+{
+	sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group);
+	return 0;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static __cpuinit int thermal_throttle_remove_dev(struct sys_device * sys_dev)
+{
+	sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group);
+	return 0;
+}
+
+/* Mutex protecting device creation against CPU hotplug */
+static DEFINE_MUTEX(therm_cpu_lock);
+
+/* Get notified when a cpu comes on/off. Be hotplug friendly. */
+static __cpuinit int thermal_throttle_cpu_callback(struct notifier_block *nfb,
+						   unsigned long action,
+						   void *hcpu)
+{
+	unsigned int cpu = (unsigned long)hcpu;
+	struct sys_device *sys_dev;
+
+	sys_dev = get_cpu_sysdev(cpu);
+	mutex_lock(&therm_cpu_lock);
+	switch (action) {
+	case CPU_ONLINE:
+		thermal_throttle_add_dev(sys_dev);
+		break;
+	case CPU_DEAD:
+		thermal_throttle_remove_dev(sys_dev);
+		break;
+	}
+	mutex_unlock(&therm_cpu_lock);
+	return NOTIFY_OK;
+}
+
+static struct notifier_block thermal_throttle_cpu_notifier =
+{
+	.notifier_call = thermal_throttle_cpu_callback,
+};
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static __init int thermal_throttle_init_device(void)
+{
+	unsigned int cpu = 0;
+
+	if (!atomic_read(&therm_throt_en))
+		return 0;
+
+	register_hotcpu_notifier(&thermal_throttle_cpu_notifier);
+
+#ifdef CONFIG_HOTPLUG_CPU
+	mutex_lock(&therm_cpu_lock);
+#endif
+	/* connect live CPUs to sysfs */
+	for_each_online_cpu(cpu)
+		thermal_throttle_add_dev(get_cpu_sysdev(cpu));
+#ifdef CONFIG_HOTPLUG_CPU
+	mutex_unlock(&therm_cpu_lock);
+#endif
+
+	return 0;
+}
+
+device_initcall(thermal_throttle_init_device);
+#endif /* CONFIG_SYSFS */
diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c
index dec11219e278..6551505d8a2c 100644
--- a/arch/x86_64/kernel/mce_intel.c
+++ b/arch/x86_64/kernel/mce_intel.c
@@ -77,6 +77,9 @@ static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c)
 	apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 	printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n",
 		cpu, tm2 ? "TM2" : "TM1");
+
+	/* enable thermal throttle processing */
+	atomic_set(&therm_throt_en, 1);
 	return;
 }
 
diff --git a/include/asm-i386/therm_throt.h b/include/asm-i386/therm_throt.h
index 3c9c22cc10c9..399bf6026b16 100644
--- a/include/asm-i386/therm_throt.h
+++ b/include/asm-i386/therm_throt.h
@@ -1,6 +1,9 @@
 #ifndef __ASM_I386_THERM_THROT_H__
 #define __ASM_I386_THERM_THROT_H__ 1
 
+#include <asm/atomic.h>
+
+extern atomic_t therm_throt_en;
 int therm_throt_process(int curr);
 
 #endif /* __ASM_I386_THERM_THROT_H__ */
-- 
cgit v1.2.3


From 38ee04f04340ffd6af868499862341d11ed2b331 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Date: Tue, 8 Aug 2006 09:10:01 -0300
Subject: V4L/DVB (4348): Fix: compile for radio aimslab and aztech with V4L2
 only

All radio devices use an obsolete mode of opening/release driver.
Since this is not V4L1 core, better to keep the method available for more
time than to rewrite open/release without a radio device to test, since the
newer method is much more complex than the previous one (although providing
support for multiple opens and multiple devices).

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/radio/radio-aimslab.c |  2 +-
 drivers/media/radio/radio-aztech.c  |  2 +-
 include/media/v4l2-dev.h            | 10 +++++++---
 3 files changed, 9 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/drivers/media/radio/radio-aimslab.c b/drivers/media/radio/radio-aimslab.c
index 7b9043f9d9bc..3368a89bfadb 100644
--- a/drivers/media/radio/radio-aimslab.c
+++ b/drivers/media/radio/radio-aimslab.c
@@ -372,7 +372,7 @@ static struct video_device rtrack_radio=
 	.owner		= THIS_MODULE,
 	.name		= "RadioTrack radio",
 	.type		= VID_TYPE_TUNER,
-	.hardware	= VID_HARDWARE_RTRACK,
+	.hardware	= 0,
 	.fops           = &rtrack_fops,
 };
 
diff --git a/drivers/media/radio/radio-aztech.c b/drivers/media/radio/radio-aztech.c
index f21d918b3f99..3ba5fa8cf7e6 100644
--- a/drivers/media/radio/radio-aztech.c
+++ b/drivers/media/radio/radio-aztech.c
@@ -328,7 +328,7 @@ static struct video_device aztech_radio=
 	.owner		= THIS_MODULE,
 	.name		= "Aztech radio",
 	.type		= VID_TYPE_TUNER,
-	.hardware	= VID_HARDWARE_AZTECH,
+	.hardware	= 0,
 	.fops           = &aztech_fops,
 };
 
diff --git a/include/media/v4l2-dev.h b/include/media/v4l2-dev.h
index 810462f8a374..c12d72d5f008 100644
--- a/include/media/v4l2-dev.h
+++ b/include/media/v4l2-dev.h
@@ -9,7 +9,8 @@
 #ifndef _V4L2_DEV_H
 #define _V4L2_DEV_H
 
-#define OBSOLETE_OWNER 1 /* to be removed soon */
+#define OBSOLETE_OWNER   1 /* to be removed soon */
+#define OBSOLETE_DEVDATA 1 /* to be removed soon */
 
 #include <linux/poll.h>
 #include <linux/fs.h>
@@ -338,8 +339,6 @@ extern int video_usercopy(struct inode *inode, struct file *file,
 #ifdef CONFIG_VIDEO_V4L1_COMPAT
 #include <linux/mm.h>
 
-extern struct video_device* video_devdata(struct file*);
-
 #define to_video_device(cd) container_of(cd, struct video_device, class_dev)
 static inline int
 video_device_create_file(struct video_device *vfd,
@@ -370,9 +369,14 @@ static inline void video_set_drvdata(struct video_device *dev, void *data)
 {
 	dev->priv = data;
 }
+
 #endif
 
+#ifdef OBSOLETE_DEVDATA /* to be removed soon */
+/* Obsolete stuff - Still needed for radio devices and obsolete drivers */
+extern struct video_device* video_devdata(struct file*);
 extern int video_exclusive_open(struct inode *inode, struct file *file);
 extern int video_exclusive_release(struct inode *inode, struct file *file);
+#endif
 
 #endif /* _V4L2_DEV_H */
-- 
cgit v1.2.3


From d1009bd733a9324baff74611e0635e17fce4dfa2 Mon Sep 17 00:00:00 2001
From: Peter Naulls <peter@chocky.org>
Date: Tue, 8 Aug 2006 09:10:05 -0300
Subject: V4L/DVB (4361): Cx88: add support for Norwood PCI TV Tuner (non-pro)

This patch adds support for Norwood PCI TV Tuner (non-pro)

Signed-off-by: Peter Naulls <peter@chocky.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.cx88 |  1 +
 drivers/media/common/ir-keymaps.c       | 48 +++++++++++++++++++++++++++++++++
 drivers/media/video/cx88/cx88-cards.c   | 20 ++++++++++++++
 drivers/media/video/cx88/cx88-core.c    |  2 +-
 drivers/media/video/cx88/cx88-input.c   | 17 +++++++++++-
 drivers/media/video/cx88/cx88.h         |  1 +
 include/media/ir-common.h               |  1 +
 7 files changed, 88 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.cx88 b/Documentation/video4linux/CARDLIST.cx88
index d692a35ef1cc..348b67ddc36b 100644
--- a/Documentation/video4linux/CARDLIST.cx88
+++ b/Documentation/video4linux/CARDLIST.cx88
@@ -52,3 +52,4 @@
  51 -> WinFast DTV2000 H                                   [107d:665e]
  52 -> Geniatech DVB-S                                     [14f1:0084]
  53 -> Hauppauge WinTV-HVR3000 TriMode Analog/DVB-S/DVB-T  [0070:1404]
+ 54 -> Norwood Micro TV Tuner
diff --git a/drivers/media/common/ir-keymaps.c b/drivers/media/common/ir-keymaps.c
index ca98d9478947..ce1281445dde 100644
--- a/drivers/media/common/ir-keymaps.c
+++ b/drivers/media/common/ir-keymaps.c
@@ -1473,3 +1473,51 @@ IR_KEYTAB_TYPE ir_codes_npgtech[IR_KEYTAB_SIZE] = {
 };
 
 EXPORT_SYMBOL_GPL(ir_codes_npgtech);
+
+/* Norwood Micro (non-Pro) TV Tuner
+   By Peter Naulls <peter@chocky.org>
+   Key comments are the functions given in the manual */
+IR_KEYTAB_TYPE ir_codes_norwood[IR_KEYTAB_SIZE] = {
+	/* Keys 0 to 9 */
+	[ 0x20 ] = KEY_0,
+	[ 0x21 ] = KEY_1,
+	[ 0x22 ] = KEY_2,
+	[ 0x23 ] = KEY_3,
+	[ 0x24 ] = KEY_4,
+	[ 0x25 ] = KEY_5,
+	[ 0x26 ] = KEY_6,
+	[ 0x27 ] = KEY_7,
+	[ 0x28 ] = KEY_8,
+	[ 0x29 ] = KEY_9,
+
+	[ 0x78 ] = KEY_TUNER,             /* Video Source        */
+	[ 0x2c ] = KEY_EXIT,              /* Open/Close software */
+	[ 0x2a ] = KEY_SELECT,            /* 2 Digit Select      */
+	[ 0x69 ] = KEY_AGAIN,             /* Recall              */
+
+	[ 0x32 ] = KEY_BRIGHTNESSUP,      /* Brightness increase */
+	[ 0x33 ] = KEY_BRIGHTNESSDOWN,    /* Brightness decrease */
+	[ 0x6b ] = KEY_KPPLUS,            /* (not named >>>>>)   */
+	[ 0x6c ] = KEY_KPMINUS,           /* (not named <<<<<)   */
+
+	[ 0x2d ] = KEY_MUTE,              /* Mute                */
+	[ 0x30 ] = KEY_VOLUMEUP,          /* Volume up           */
+	[ 0x31 ] = KEY_VOLUMEDOWN,        /* Volume down         */
+	[ 0x60 ] = KEY_CHANNELUP,         /* Channel up          */
+	[ 0x61 ] = KEY_CHANNELDOWN,       /* Channel down        */
+
+	[ 0x3f ] = KEY_RECORD,            /* Record              */
+	[ 0x37 ] = KEY_PLAY,              /* Play                */
+	[ 0x36 ] = KEY_PAUSE,             /* Pause               */
+	[ 0x2b ] = KEY_STOP,              /* Stop                */
+	[ 0x67 ] = KEY_FASTFORWARD,       /* Foward              */
+	[ 0x66 ] = KEY_REWIND,            /* Rewind              */
+	[ 0x3e ] = KEY_SEARCH,            /* Auto Scan           */
+	[ 0x2e ] = KEY_CAMERA,            /* Capture Video       */
+	[ 0x6d ] = KEY_MENU,              /* Show/Hide Control   */
+	[ 0x2f ] = KEY_ZOOM,              /* Full Screen         */
+	[ 0x34 ] = KEY_RADIO,             /* FM                  */
+	[ 0x65 ] = KEY_POWER,             /* Computer power      */
+};
+
+EXPORT_SYMBOL_GPL(ir_codes_norwood);
diff --git a/drivers/media/video/cx88/cx88-cards.c b/drivers/media/video/cx88/cx88-cards.c
index afd50b465df7..83bfcecee2d3 100644
--- a/drivers/media/video/cx88/cx88-cards.c
+++ b/drivers/media/video/cx88/cx88-cards.c
@@ -1231,6 +1231,26 @@ struct cx88_board cx88_boards[] = {
 			.gpio0  = 0x84bf,
 		}},
 	},
+	[CX88_BOARD_NORWOOD_MICRO] = {
+		.name           = "Norwood Micro TV Tuner",
+		.tuner_type     = TUNER_TNF_5335MF,
+		.radio_type     = UNSET,
+		.tuner_addr     = ADDR_UNSET,
+		.radio_addr     = ADDR_UNSET,
+		.input          = {{
+			.type   = CX88_VMUX_TELEVISION,
+			.vmux   = 0,
+			.gpio0  = 0x0709,
+		},{
+			.type   = CX88_VMUX_COMPOSITE1,
+			.vmux   = 1,
+			.gpio0  = 0x070b,
+		},{
+			.type   = CX88_VMUX_SVIDEO,
+			.vmux   = 2,
+			.gpio0  = 0x070b,
+		}},
+	},
 };
 const unsigned int cx88_bcount = ARRAY_SIZE(cx88_boards);
 
diff --git a/drivers/media/video/cx88/cx88-core.c b/drivers/media/video/cx88/cx88-core.c
index 973d3f39b2d5..d72e177607a7 100644
--- a/drivers/media/video/cx88/cx88-core.c
+++ b/drivers/media/video/cx88/cx88-core.c
@@ -105,7 +105,7 @@ static u32* cx88_risc_field(u32 *rp, struct scatterlist *sglist,
 			*(rp++)=cpu_to_le32(sg_dma_address(sg)+offset);
 			offset+=bpl;
 		} else {
-			/* scanline needs to be splitted */
+			/* scanline needs to be split */
 			todo = bpl;
 			*(rp++)=cpu_to_le32(RISC_WRITE|RISC_SOL|
 					    (sg_dma_len(sg)-offset));
diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c
index c25564648993..3049bd5b5a48 100644
--- a/drivers/media/video/cx88/cx88-input.c
+++ b/drivers/media/video/cx88/cx88-input.c
@@ -107,7 +107,15 @@ static void cx88_ir_handle_key(struct cx88_IR *ir)
 		   (gpio & ir->mask_keydown) ? " down" : "",
 		   (gpio & ir->mask_keyup) ? " up" : "");
 
-	if (ir->mask_keydown) {
+	if (ir->core->board == CX88_BOARD_NORWOOD_MICRO) {
+		u32 gpio_key = cx_read(MO_GP0_IO);
+
+		data = (data << 4) | ((gpio_key & 0xf0) >> 4);
+
+		ir_input_keydown(ir->input, &ir->ir, data, data);
+		ir_input_nokey(ir->input, &ir->ir);
+
+	} else if (ir->mask_keydown) {
 		/* bit set on keydown */
 		if (gpio & ir->mask_keydown) {
 			ir_input_keydown(ir->input, &ir->ir, data, data);
@@ -248,6 +256,13 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci)
 		ir_type = IR_TYPE_PD;
 		ir->sampling = 0xff00; /* address */
 		break;
+	case CX88_BOARD_NORWOOD_MICRO:
+		ir_codes         = ir_codes_norwood;
+		ir->gpio_addr    = MO_GP1_IO;
+		ir->mask_keycode = 0x0e;
+		ir->mask_keyup   = 0x80;
+		ir->polling      = 50; /* ms */
+		break;
 	case CX88_BOARD_NPGTECH_REALTV_TOP10FM:
 		ir_codes = ir_codes_npgtech;
 		ir->gpio_addr = MO_GP0_IO;
diff --git a/drivers/media/video/cx88/cx88.h b/drivers/media/video/cx88/cx88.h
index ce441e27f214..0405d6d0157a 100644
--- a/drivers/media/video/cx88/cx88.h
+++ b/drivers/media/video/cx88/cx88.h
@@ -198,6 +198,7 @@ extern struct sram_channel cx88_sram_channels[];
 #define CX88_BOARD_WINFAST_DTV2000H        51
 #define CX88_BOARD_GENIATECH_DVBS          52
 #define CX88_BOARD_HAUPPAUGE_HVR3000       53
+#define CX88_BOARD_NORWOOD_MICRO           54
 
 enum cx88_itype {
 	CX88_VMUX_COMPOSITE1 = 1,
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index 7bab09b0ed45..f883bc3aa70e 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -90,6 +90,7 @@ extern IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_pinnacle_color[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_hauppauge_new[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_npgtech[IR_KEYTAB_SIZE];
+extern IR_KEYTAB_TYPE ir_codes_norwood[IR_KEYTAB_SIZE];
 
 #endif
 
-- 
cgit v1.2.3


From d7304dee3b7e29e801ba59bbf9a47440c196263d Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Thu, 24 Aug 2006 22:43:45 -0300
Subject: V4L/DVB (4533): Tda9887: add configuration setting for L standard PLL
 gating

Add a tuner config parameter for TDA9887, default_pll_gating_18, that
changes the L standard PLL gating value from 36% to 0% (datasheet says
0%, tda9887 code says 18%).
Turn this on for Microtune 4049FM5, as recomended by tuner datasheet.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/tda9887.c      | 2 ++
 drivers/media/video/tuner-simple.c | 2 ++
 drivers/media/video/tuner-types.c  | 1 +
 include/media/tuner-types.h        | 3 +++
 include/media/tuner.h              | 1 +
 5 files changed, 9 insertions(+)

(limited to 'include')

diff --git a/drivers/media/video/tda9887.c b/drivers/media/video/tda9887.c
index 8dab481d384a..87ffb0e84a7a 100644
--- a/drivers/media/video/tda9887.c
+++ b/drivers/media/video/tda9887.c
@@ -480,6 +480,8 @@ static int tda9887_set_config(struct tuner *t, char *buf)
 	}
 	if ((t->tda9887_config & TDA9887_INTERCARRIER_NTSC) && (t->std & V4L2_STD_NTSC))
 		buf[1] &= ~cQSS;
+	if (t->tda9887_config & TDA9887_GATING_18)
+		buf[3] &= ~cGating_36;
 	return 0;
 }
 
diff --git a/drivers/media/video/tuner-simple.c b/drivers/media/video/tuner-simple.c
index 4cfe71ba7304..63db4e97ae6c 100644
--- a/drivers/media/video/tuner-simple.c
+++ b/drivers/media/video/tuner-simple.c
@@ -331,6 +331,8 @@ static void default_set_tv_freq(struct i2c_client *c, unsigned int freq)
 			else if (params->default_top_high)
 				config |= TDA9887_TOP(params->default_top_high);
 		}
+		if (params->default_pll_gating_18)
+			config |= TDA9887_GATING_18;
 		i2c_clients_command(c->adapter, TDA9887_SET_CONFIG, &config);
 	}
 	tuner_dbg("tv 0x%02x 0x%02x 0x%02x 0x%02x\n",
diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c
index 8b542599ed47..8fff642fad56 100644
--- a/drivers/media/video/tuner-types.c
+++ b/drivers/media/video/tuner-types.c
@@ -650,6 +650,7 @@ static struct tuner_params tuner_microtune_4049_fm5_params[] = {
 		.count  = ARRAY_SIZE(tuner_temic_4009f_5_pal_ranges),
 		.has_tda9887 = 1,
 		.port1_invert_for_secam_lc = 1,
+		.default_pll_gating_18 = 1,
 	},
 };
 
diff --git a/include/media/tuner-types.h b/include/media/tuner-types.h
index 3c43b95f4c0d..37dad07a8439 100644
--- a/include/media/tuner-types.h
+++ b/include/media/tuner-types.h
@@ -72,6 +72,9 @@ struct tuner_params {
 	unsigned int port2_invert_for_secam_lc:1;
 	/* Some cards require PORT1 to be 1 for mono Radio FM and 0 for stereo. */
 	unsigned int port1_set_for_fm_mono:1;
+	/* Select 18% (or according to datasheet 0%) L standard PLL gating,
+	   vs the driver default of 36%. */
+	unsigned int default_pll_gating_18:1;
 	/* Default tda9887 TOP value in dB for the low band. Default is 0.
 	   Range: -16:+15 */
 	signed int default_top_low:5;
diff --git a/include/media/tuner.h b/include/media/tuner.h
index 2f7b00b08e88..3116e750132f 100644
--- a/include/media/tuner.h
+++ b/include/media/tuner.h
@@ -144,6 +144,7 @@ extern int tuner_debug;
 #define TDA9887_DEEMPHASIS_50 		(2<<16)
 #define TDA9887_DEEMPHASIS_75 		(3<<16)
 #define TDA9887_AUTOMUTE 		(1<<18)
+#define TDA9887_GATING_18		(1<<19)
 
 #ifdef __KERNEL__
 
-- 
cgit v1.2.3


From 1739adea321788e380794c1072c810d445090bca Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sat, 26 Aug 2006 03:05:17 -0300
Subject: V4L/DVB (4545): Add missing v4l2_buf_type to struct
 v4l2_sliced_vbi_cap.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/linux/videodev2.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index e3715d774197..d5746d470c73 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -1135,7 +1135,8 @@ struct v4l2_sliced_vbi_cap
 				 (equals frame lines 313-336 for 625 line video
 				  standards, 263-286 for 525 line standards) */
 	__u16   service_lines[2][24];
-	__u32   reserved[4];    /* must be 0 */
+	enum v4l2_buf_type type;
+	__u32   reserved[3];    /* must be 0 */
 };
 
 struct v4l2_sliced_vbi_data
-- 
cgit v1.2.3


From 784e8fe417d45b526eeb74794b7df43e11000f70 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sat, 26 Aug 2006 03:17:58 -0300
Subject: V4L/DVB (4546): Add u32 argument to VIDIOC_INT_RESET.

The extra argument makes it possible to reset subsystems of a chip if
that is supported.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/em28xx/em28xx-video.c         | 2 +-
 drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c | 2 +-
 drivers/media/video/v4l2-common.c                 | 1 +
 include/media/v4l2-common.h                       | 5 +++--
 4 files changed, 6 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/em28xx/em28xx-video.c b/drivers/media/video/em28xx/em28xx-video.c
index 2a461dde480c..20df657b70c8 100644
--- a/drivers/media/video/em28xx/em28xx-video.c
+++ b/drivers/media/video/em28xx/em28xx-video.c
@@ -174,7 +174,7 @@ static void em28xx_config_i2c(struct em28xx *dev)
 
 	route.input = INPUT(dev->ctl_input)->vmux;
 	route.output = 0;
-	em28xx_i2c_call_clients(dev, VIDIOC_INT_RESET, NULL);
+	em28xx_i2c_call_clients(dev, VIDIOC_INT_RESET, 0);
 	em28xx_i2c_call_clients(dev, VIDIOC_INT_S_VIDEO_ROUTING, &route);
 	em28xx_i2c_call_clients(dev, VIDIOC_STREAMON, NULL);
 
diff --git a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
index c80c26be6e4d..df8feac16aee 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
@@ -221,7 +221,7 @@ static unsigned int decoder_describe(struct pvr2_v4l_cx2584x *ctxt,
 static void decoder_reset(struct pvr2_v4l_cx2584x *ctxt)
 {
 	int ret;
-	ret = pvr2_i2c_client_cmd(ctxt->client,VIDIOC_INT_RESET,NULL);
+	ret = pvr2_i2c_client_cmd(ctxt->client,VIDIOC_INT_RESET,0);
 	pvr2_trace(PVR2_TRACE_CHIPS,"i2c cx25840 decoder_reset (ret=%d)",ret);
 }
 
diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index 8d972ffdaf98..78d28b03ec93 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -938,6 +938,7 @@ void v4l_printk_ioctl_arg(char *s,unsigned int cmd, void *arg)
 	case VIDIOC_INT_AUDIO_CLOCK_FREQ:
 	case VIDIOC_INT_I2S_CLOCK_FREQ:
 	case VIDIOC_INT_S_STANDBY:
+	case VIDIOC_INT_RESET:
 	{
 		u32 *p=arg;
 
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index 5564db13c0d5..8721ac4a45a7 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -169,8 +169,9 @@ enum v4l2_chip_ident {
 #define	VIDIOC_INT_S_REGISTER 		_IOR ('d', 100, struct v4l2_register)
 #define	VIDIOC_INT_G_REGISTER 		_IOWR('d', 101, struct v4l2_register)
 
-/* Reset the I2C chip */
-#define VIDIOC_INT_RESET            	_IO  ('d', 102)
+/* Generic reset command. The argument selects which subsystems to reset.
+   Passing 0 will always reset the whole chip. */
+#define VIDIOC_INT_RESET            	_IOW ('d', 102, u32)
 
 /* Set the frequency (in Hz) of the audio clock output.
    Used to slave an audio processor to the video decoder, ensuring that audio
-- 
cgit v1.2.3


From 89f75ffc7e97d96ea76556671446d57d77c46beb Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@infradead.org>
Date: Tue, 29 Aug 2006 22:07:03 -0300
Subject: V4L/DVB (4553): Add support for saa7111 and partial support for
 saa7118

Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/saa7115.c | 252 +++++++++++++++++++++++++++++++-----------
 include/media/v4l2-common.h   |   9 +-
 2 files changed, 195 insertions(+), 66 deletions(-)

(limited to 'include')

diff --git a/drivers/media/video/saa7115.c b/drivers/media/video/saa7115.c
index d5052dbd0744..900c65772fa0 100644
--- a/drivers/media/video/saa7115.c
+++ b/drivers/media/video/saa7115.c
@@ -1,4 +1,6 @@
-/* saa7115 - Philips SAA7113/SAA7114/SAA7115 video decoder driver
+/* saa711x - Philips SAA711x video decoder driver
+ * This driver can work with saa7111, saa7111a, saa7113, saa7114,
+ *			     saa7115 and saa7118.
  *
  * Based on saa7114 driver by Maxim Yevtyushkin, which is based on
  * the saa7111 driver by Dave Perks.
@@ -16,7 +18,9 @@
  * (2/17/2003)
  *
  * VBI support (2004) and cleanups (2005) by Hans Verkuil <hverkuil@xs4all.nl>
- * SAA7113 support by Mauro Carvalho Chehab <mchehab@infradead.org>
+ *
+ * Copyright (c) 2005-2006 Mauro Carvalho Chehab <mchehab@infradead.org>
+ *	SAA7111, SAA7113 and SAA7118 support
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
@@ -44,7 +48,7 @@
 #include <media/saa7115.h>
 #include <asm/div64.h>
 
-MODULE_DESCRIPTION("Philips SAA7113/SAA7114/SAA7115 video decoder driver");
+MODULE_DESCRIPTION("Philips SAA7111/SAA7113/SAA7114/SAA7115/SAA7118 video decoder driver");
 MODULE_AUTHOR(  "Maxim Yevtyushkin, Kevin Thayer, Chris Kennedy, "
 		"Hans Verkuil, Mauro Carvalho Chehab");
 MODULE_LICENSE("GPL");
@@ -55,8 +59,8 @@ module_param(debug, bool, 0644);
 MODULE_PARM_DESC(debug, "Debug level (0-1)");
 
 static unsigned short normal_i2c[] = {
-		0x4a >> 1, 0x48 >> 1,	/* SAA7113 */
-		0x42 >> 1, 0x40 >> 1,	/* SAA7114 and SAA7115 */
+		0x4a >> 1, 0x48 >> 1,	/* SAA7111, SAA7111A and SAA7113 */
+		0x42 >> 1, 0x40 >> 1,	/* SAA7114, SAA7115 and SAA7118 */
 		I2C_CLIENT_END };
 
 
@@ -86,15 +90,58 @@ static inline int saa7115_write(struct i2c_client *client, u8 reg, u8 value)
 	return i2c_smbus_write_byte_data(client, reg, value);
 }
 
+/* Sanity routine to check if a register is present */
+static int saa711x_has_reg(const int id, const u8 reg)
+{
+	switch (id) {
+	case V4L2_IDENT_SAA7111:
+		if (reg>0x1f || reg==1 || reg==0x0f || reg==0x14 || reg==0x18
+				       || reg==0x19 || reg==0x1d || reg==0x1e)
+			return 0;
+	case V4L2_IDENT_SAA7113:
+		if (reg>0x62 || reg==0x14 || (reg>=0x18 && reg<=0x1e) ||
+				    (reg>=0x20 && reg<=0x3f) ||reg==0x5f )
+			return 0;
+	case V4L2_IDENT_SAA7114:
+		if (reg>=0xf0 || (reg>=0x1a && reg<=0x1e) ||
+				  (reg>=0x20 && reg<=0x2f) ||
+				  (reg>=0x63 && reg<=0x7f) )
+			return 0;
+	case V4L2_IDENT_SAA7115:
+		if ((reg>=0x20 && reg<=0x2f) || (reg==0x5c) ||
+				(reg>=0xfc && reg<=0xfe) )
+			return 0;
+	case V4L2_IDENT_SAA7118:
+		if (reg>=0xf0 || (reg>=0x1a && reg<=0x1d) ||
+					(reg>=0x63 && reg<=0x6f) )
+			return 0;
+	}
+
+	/* Those registers are reserved for all family */
+	if (unlikely((reg>=0x20 && reg<=0x22) ||
+			(reg>=0x26 && reg<=0x28) ||
+			(reg>=0x3b && reg<=0x3f) || (reg==0x5f) ||
+			(reg>=0x63 && reg<=0x6f) ) )
+		return 0;
+
+	return 1;
+}
+
 static int saa7115_writeregs(struct i2c_client *client, const unsigned char *regs)
 {
+	struct saa7115_state *state = i2c_get_clientdata(client);
 	unsigned char reg, data;
 
 	while (*regs != 0x00) {
 		reg = *(regs++);
 		data = *(regs++);
-		if (saa7115_write(client, reg, data) < 0)
-			return -1;
+
+		/* According with datasheets, reserved regs should be
+		   filled with 0 - seems better not to touch on they */
+		if (saa711x_has_reg(state->ident,reg)) {
+			if (saa7115_write(client, reg, data) < 0)
+				return -1;
+		}
 	}
 	return 0;
 }
@@ -106,10 +153,80 @@ static inline int saa7115_read(struct i2c_client *client, u8 reg)
 
 /* ----------------------------------------------------------------------- */
 
+/* SAA7111 initialization table */
+static const unsigned char saa7111_init_auto_input[] = {
+	R_01_INC_DELAY, 0x00,		/* reserved */
+
+	/*front end */
+	R_02_INPUT_CNTL_1, 0xd0,	/* FUSE=3, GUDL=2, MODE=0 */
+	R_03_INPUT_CNTL_2, 0x23,	/* HLNRS=0, VBSL=1, WPOFF=0, HOLDG=0,
+					 * GAFIX=0, GAI1=256, GAI2=256 */
+	R_04_INPUT_CNTL_3, 0x00,	/* GAI1=256 */
+	R_05_INPUT_CNTL_4, 0x00,	/* GAI2=256 */
+
+	/* decoder */
+	R_06_H_SYNC_START, 0xf3,	/* HSB at  13(50Hz) /  17(60Hz)
+					 * pixels after end of last line */
+	R_07_H_SYNC_STOP, 0xe8,		/* HSS seems to be needed to
+					 * work with NTSC, too */
+	R_08_SYNC_CNTL, 0xc8,		/* AUFD=1, FSEL=1, EXFIL=0,
+					 * VTRC=1, HPLL=0, VNOI=0 */
+	R_09_LUMA_CNTL, 0x01,		/* BYPS=0, PREF=0, BPSS=0,
+					 * VBLB=0, UPTCV=0, APER=1 */
+	R_0A_LUMA_BRIGHT_CNTL, 0x80,
+	R_0B_LUMA_CONTRAST_CNTL, 0x47,	/* 0b - CONT=1.109 */
+	R_0C_CHROMA_SAT_CNTL, 0x40,
+	R_0D_CHROMA_HUE_CNTL, 0x00,
+	R_0E_CHROMA_CNTL_1, 0x01,	/* 0e - CDTO=0, CSTD=0, DCCF=0,
+					 * FCTC=0, CHBW=1 */
+	R_0F_CHROMA_GAIN_CNTL, 0x00,	/* reserved */
+	R_10_CHROMA_CNTL_2, 0x48,	/* 10 - OFTS=1, HDEL=0, VRLN=1, YDEL=0 */
+	R_11_MODE_DELAY_CNTL, 0x1c,	/* 11 - GPSW=0, CM99=0, FECO=0, COMPO=1,
+					 * OEYC=1, OEHV=1, VIPB=0, COLO=0 */
+	R_12_RT_SIGNAL_CNTL, 0x00,	/* 12 - output control 2 */
+	R_13_RT_X_PORT_OUT_CNTL, 0x00,	/* 13 - output control 3 */
+	R_14_ANAL_ADC_COMPAT_CNTL, 0x00,
+	R_15_VGATE_START_FID_CHG, 0x00,
+	R_16_VGATE_STOP, 0x00,
+	R_17_MISC_VGATE_CONF_AND_MSB, 0x00,
+
+	0x00, 0x00
+};
+
+/* SAA7113 init codes */
+static const unsigned char saa7113_init_auto_input[] = {
+	R_01_INC_DELAY, 0x08,
+	R_02_INPUT_CNTL_1, 0xc2,
+	R_03_INPUT_CNTL_2, 0x30,
+	R_04_INPUT_CNTL_3, 0x00,
+	R_05_INPUT_CNTL_4, 0x00,
+	R_06_H_SYNC_START, 0x89,
+	R_07_H_SYNC_STOP, 0x0d,
+	R_08_SYNC_CNTL, 0x88,
+	R_09_LUMA_CNTL, 0x01,
+	R_0A_LUMA_BRIGHT_CNTL, 0x80,
+	R_0B_LUMA_CONTRAST_CNTL, 0x47,
+	R_0C_CHROMA_SAT_CNTL, 0x40,
+	R_0D_CHROMA_HUE_CNTL, 0x00,
+	R_0E_CHROMA_CNTL_1, 0x01,
+	R_0F_CHROMA_GAIN_CNTL, 0x2a,
+	R_10_CHROMA_CNTL_2, 0x08,
+	R_11_MODE_DELAY_CNTL, 0x0c,
+	R_12_RT_SIGNAL_CNTL, 0x07,
+	R_13_RT_X_PORT_OUT_CNTL, 0x00,
+	R_14_ANAL_ADC_COMPAT_CNTL, 0x00,
+	R_15_VGATE_START_FID_CHG, 0x00,
+	R_16_VGATE_STOP, 0x00,
+	R_17_MISC_VGATE_CONF_AND_MSB, 0x00,
+
+	0x00, 0x00
+};
+
 /* If a value differs from the Hauppauge driver values, then the comment starts with
    'was 0xXX' to denote the Hauppauge value. Otherwise the value is identical to what the
    Hauppauge driver sets. */
 
+/* SAA7114 and SAA7115 initialization table */
 static const unsigned char saa7115_init_auto_input[] = {
 		/* Front-End Part */
 	R_01_INC_DELAY, 0x48,			/* white peak control disabled */
@@ -142,6 +259,7 @@ static const unsigned char saa7115_init_auto_input[] = {
 	0x00, 0x00
 };
 
+/* Used to reset saa7113, saa7114 and saa7115 */
 static const unsigned char saa7115_cfg_reset_scaler[] = {
 	R_87_I_PORT_I_O_ENA_OUT_CLK_AND_GATED, 0x00,	/* disable I-port output */
 	R_88_POWER_SAVE_ADC_PORT_CNTL, 0xd0,		/* reset scaler */
@@ -152,6 +270,7 @@ static const unsigned char saa7115_cfg_reset_scaler[] = {
 
 /* ============== SAA7715 VIDEO templates =============  */
 
+/* Used on saa7114 and saa7115 */
 static const unsigned char saa7115_cfg_60hz_fullres_x[] = {
 	/* hsize = 0x2d0 = 720 */
 	R_CC_B_HORIZ_OUTPUT_WINDOW_LENGTH, 0xd0,
@@ -170,6 +289,7 @@ static const unsigned char saa7115_cfg_60hz_fullres_x[] = {
 	0x00, 0x00
 };
 
+/* Used on saa7114 and saa7115 */
 static const unsigned char saa7115_cfg_60hz_fullres_y[] = {
 	/* output window size = 248 (but 60hz is 240?) */
 	R_CE_B_VERT_OUTPUT_WINDOW_LENGTH, 0xf8,
@@ -408,33 +528,6 @@ static const unsigned char saa7115_cfg_vbi_off[] = {
 	0x00, 0x00
 };
 
-static const unsigned char saa7113_init_auto_input[] = {
-	R_01_INC_DELAY, 0x08,
-	R_02_INPUT_CNTL_1, 0xc2,
-	R_03_INPUT_CNTL_2, 0x30,
-	R_04_INPUT_CNTL_3, 0x00,
-	R_05_INPUT_CNTL_4, 0x00,
-	R_06_H_SYNC_START, 0x89,
-	R_07_H_SYNC_STOP, 0x0d,
-	R_08_SYNC_CNTL, 0x88,
-	R_09_LUMA_CNTL, 0x01,
-	R_0A_LUMA_BRIGHT_CNTL, 0x80,
-	R_0B_LUMA_CONTRAST_CNTL, 0x47,
-	R_0C_CHROMA_SAT_CNTL, 0x40,
-	R_0D_CHROMA_HUE_CNTL, 0x00,
-	R_0E_CHROMA_CNTL_1, 0x01,
-	R_0F_CHROMA_GAIN_CNTL, 0x2a,
-	R_10_CHROMA_CNTL_2, 0x08,
-	R_11_MODE_DELAY_CNTL, 0x0c,
-	R_12_RT_SIGNAL_CNTL, 0x07,
-	R_13_RT_X_PORT_OUT_CNTL, 0x00,
-	R_14_ANAL_ADC_COMPAT_CNTL, 0x00,
-	R_15_VGATE_START_FID_CHG, 0x00,
-	R_16_VGATE_STOP, 0x00,
-	R_17_MISC_VGATE_CONF_AND_MSB, 0x00,
-
-	0x00, 0x00
-};
 
 static const unsigned char saa7115_init_misc[] = {
 	R_81_V_SYNC_FLD_ID_SRC_SEL_AND_RETIMED_V_F, 0x01,
@@ -662,7 +755,6 @@ static int saa7115_decode_wss(u8 * p)
 	return wss;
 }
 
-
 static int saa7115_set_audio_clock_freq(struct i2c_client *client, u32 freq)
 {
 	struct saa7115_state *state = i2c_get_clientdata(client);
@@ -672,16 +764,16 @@ static int saa7115_set_audio_clock_freq(struct i2c_client *client, u32 freq)
 	u64 f;
 	u8 acc = 0; 	/* reg 0x3a, audio clock control */
 
+	/* Checks for chips that don't have audio clock (saa7111, saa7113) */
+	if (!saa711x_has_reg(state->ident,R_30_AUD_MAST_CLK_CYCLES_PER_FIELD))
+		return 0;
+
 	v4l_dbg(1, debug, client, "set audio clock freq: %d\n", freq);
 
 	/* sanity check */
 	if (freq < 32000 || freq > 48000)
 		return -EINVAL;
 
-	/* The saa7113 has no audio clock */
-	if (state->ident == V4L2_IDENT_SAA7113)
-		return 0;
-
 	/* hz is the refresh rate times 100 */
 	hz = (state->std & V4L2_STD_525_60) ? 5994 : 5000;
 	/* acpf = (256 * freq) / field_frequency == (256 * 100 * freq) / hz */
@@ -799,7 +891,6 @@ static int saa7115_get_v4lctrl(struct i2c_client *client, struct v4l2_control *c
 static void saa7115_set_v4lstd(struct i2c_client *client, v4l2_std_id std)
 {
 	struct saa7115_state *state = i2c_get_clientdata(client);
-	int taskb = saa7115_read(client, R_80_GLOBAL_CNTL_1) & 0x10;
 
 	/* Prevent unnecessary standard changes. During a standard
 	   change the I-Port is temporarily disabled. Any devices
@@ -821,7 +912,7 @@ static void saa7115_set_v4lstd(struct i2c_client *client, v4l2_std_id std)
 	}
 
 	/* Register 0E - Bits D6-D4 on NO-AUTO mode
-		(SAA7113 doesn't have auto mode)
+		(SAA7111 and SAA7113 doesn't have auto mode)
 	    50 Hz / 625 lines           60 Hz / 525 lines
 	000 PAL BGDHI (4.43Mhz)         NTSC M (3.58MHz)
 	001 NTSC 4.43 (50 Hz)           PAL 4.43 (60 Hz)
@@ -829,7 +920,8 @@ static void saa7115_set_v4lstd(struct i2c_client *client, v4l2_std_id std)
 	011 NTSC N (3.58MHz)            PAL M (3.58MHz)
 	100 reserved                    NTSC-Japan (3.58MHz)
 	*/
-	if (state->ident == V4L2_IDENT_SAA7113) {
+	if (state->ident == V4L2_IDENT_SAA7111 ||
+	    state->ident == V4L2_IDENT_SAA7113) {
 		u8 reg = saa7115_read(client, R_0E_CHROMA_CNTL_1) & 0x8f;
 
 		if (std == V4L2_STD_PAL_M) {
@@ -842,18 +934,19 @@ static void saa7115_set_v4lstd(struct i2c_client *client, v4l2_std_id std)
 			reg |= 0x40;
 		}
 		saa7115_write(client, R_0E_CHROMA_CNTL_1, reg);
-	}
-
+	} else {
+		/* restart task B if needed */
+		int taskb = saa7115_read(client, R_80_GLOBAL_CNTL_1) & 0x10;
 
-	state->std = std;
+		if (taskb && state->ident == V4L2_IDENT_SAA7114) {
+			saa7115_writeregs(client, saa7115_cfg_vbi_on);
+		}
 
-	/* restart task B if needed */
-	if (taskb && state->ident != V4L2_IDENT_SAA7115) {
-		saa7115_writeregs(client, saa7115_cfg_vbi_on);
+		/* switch audio mode too! */
+		saa7115_set_audio_clock_freq(client, state->audclk_freq);
 	}
 
-	/* switch audio mode too! */
-	saa7115_set_audio_clock_freq(client, state->audclk_freq);
+	state->std = std;
 }
 
 static v4l2_std_id saa7115_get_v4lstd(struct i2c_client *client)
@@ -919,9 +1012,16 @@ static void saa7115_set_lcr(struct i2c_client *client, struct v4l2_sliced_vbi_fo
 	u8 lcr[24];
 	int i, x;
 
-	/* saa7113/7114 doesn't yet support VBI */
+#if 1
+	/* saa7113/7114/7118 VBI support are experimental */
+	if (!saa711x_has_reg(state->ident,R_41_LCR_BASE))
+		return;
+
+#else
+	/* SAA7113 and SAA7118 also should support VBI - Need testing */
 	if (state->ident != V4L2_IDENT_SAA7115)
 		return;
+#endif
 
 	for (i = 0; i <= 23; i++)
 		lcr[i] = 0xff;
@@ -1021,8 +1121,9 @@ static int saa7115_set_v4lfmt(struct i2c_client *client, struct v4l2_format *fmt
 	struct saa7115_state *state = i2c_get_clientdata(client);
 	struct v4l2_pix_format *pix;
 	int HPSC, HFSC;
-	int VSCY, Vsrc;
+	int VSCY;
 	int is_50hz = state->std & V4L2_STD_625_50;
+	int Vsrc = is_50hz ? 576 : 480;
 
 	if (fmt->type == V4L2_BUF_TYPE_SLICED_VBI_CAPTURE) {
 		saa7115_set_lcr(client, &fmt->fmt.sliced);
@@ -1041,18 +1142,29 @@ static int saa7115_set_v4lfmt(struct i2c_client *client, struct v4l2_format *fmt
 	if ((pix->height < 1) || (pix->height > 960))
 		return -EINVAL;
 
+	if (!saa711x_has_reg(state->ident,R_D0_B_HORIZ_PRESCALING)) {
+		/* Decoder only supports 720 columns and 480 or 576 lines */
+		if (pix->width != 720)
+			return -EINVAL;
+		if (pix->height != Vsrc)
+			return -EINVAL;
+	}
+
 	/* probably have a valid size, let's set it */
 	/* Set output width/height */
 	/* width */
-	saa7115_write(client, R_CC_B_HORIZ_OUTPUT_WINDOW_LENGTH,
-				(u8) (pix->width & 0xff));
-	saa7115_write(client, R_CD_B_HORIZ_OUTPUT_WINDOW_LENGTH_MSB,
-				(u8) ((pix->width >> 8) & 0xff));
-	/* height */
-	saa7115_write(client, R_CE_B_VERT_OUTPUT_WINDOW_LENGTH,
-				(u8) (pix->height & 0xff));
-	saa7115_write(client, R_CF_B_VERT_OUTPUT_WINDOW_LENGTH_MSB,
-				(u8) ((pix->height >> 8) & 0xff));
+
+	if (!saa711x_has_reg(state->ident,R_CC_B_HORIZ_OUTPUT_WINDOW_LENGTH)) {
+		saa7115_write(client, R_CC_B_HORIZ_OUTPUT_WINDOW_LENGTH,
+					(u8) (pix->width & 0xff));
+		saa7115_write(client, R_CD_B_HORIZ_OUTPUT_WINDOW_LENGTH_MSB,
+					(u8) ((pix->width >> 8) & 0xff));
+		/* height */
+		saa7115_write(client, R_CE_B_VERT_OUTPUT_WINDOW_LENGTH,
+					(u8) (pix->height & 0xff));
+		saa7115_write(client, R_CF_B_VERT_OUTPUT_WINDOW_LENGTH_MSB,
+					(u8) ((pix->height >> 8) & 0xff));
+	}
 
 	/* Scaling settings */
 	/* Hprescaler is floor(inres/outres) */
@@ -1090,8 +1202,6 @@ static int saa7115_set_v4lfmt(struct i2c_client *client, struct v4l2_format *fmt
 		}
 	}
 
-	Vsrc = is_50hz ? 576 : 480;
-
 	if (pix->height != Vsrc) {
 		VSCY = (int)((1024 * Vsrc) / pix->height);
 		v4l_dbg(1, debug, client, "Vsrc: %d, Vscy: 0x%05x\n", Vsrc, VSCY);
@@ -1123,6 +1233,7 @@ static int saa7115_set_v4lfmt(struct i2c_client *client, struct v4l2_format *fmt
 	}
 
 	saa7115_writeregs(client, saa7115_cfg_reset_scaler);
+
 	return 0;
 }
 
@@ -1468,15 +1579,25 @@ static int saa7115_attach(struct i2c_adapter *adapter, int address, int kind)
 	state->hue = 0;
 	state->sat = 64;
 	switch (chip_id) {
+	case 1:
+		state->ident = V4L2_IDENT_SAA7111;
+		break;
 	case 3:
 		state->ident = V4L2_IDENT_SAA7113;
 		break;
 	case 4:
 		state->ident = V4L2_IDENT_SAA7114;
 		break;
-	default:
+	case 5:
 		state->ident = V4L2_IDENT_SAA7115;
 		break;
+	case 8:
+		state->ident = V4L2_IDENT_SAA7118;
+		break;
+	default:
+		state->ident = V4L2_IDENT_SAA7111;
+		v4l_info(client, "WARNING: Chip is not known - Falling back to saa7111\n");
+
 	}
 
 	state->audclk_freq = 48000;
@@ -1484,7 +1605,8 @@ static int saa7115_attach(struct i2c_adapter *adapter, int address, int kind)
 	v4l_dbg(1, debug, client, "writing init values\n");
 
 	/* init to 60hz/48khz */
-	if (state->ident == V4L2_IDENT_SAA7113) {
+	if (state->ident == V4L2_IDENT_SAA7111 ||
+			state->ident == V4L2_IDENT_SAA7113) {
 		state->crystal_freq = SAA7115_FREQ_24_576_MHZ;
 		saa7115_writeregs(client, saa7113_init_auto_input);
 	} else {
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index 8721ac4a45a7..ce24a6ad445c 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -121,10 +121,17 @@ enum v4l2_chip_ident {
 	/* general idents: reserved range 0-49 */
 	V4L2_IDENT_UNKNOWN = 0,
 
-	/* module saa7115: reserved range 100-149 */
+	/* module saa7110: just ident= 100 */
+	V4L2_IDENT_SAA7110 = 100,
+
+	/* module saa7111: just ident= 101 */
+	V4L2_IDENT_SAA7111 = 101,
+
+	/* module saa7115: reserved range 102-149 */
 	V4L2_IDENT_SAA7113 = 103,
 	V4L2_IDENT_SAA7114 = 104,
 	V4L2_IDENT_SAA7115 = 105,
+	V4L2_IDENT_SAA7118 = 108,
 
 	/* module saa7127: reserved range 150-199 */
 	V4L2_IDENT_SAA7127 = 157,
-- 
cgit v1.2.3


From f2a49bc72bd2993356b3640cc105646ae56b10b8 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Fri, 1 Sep 2006 18:32:55 -0300
Subject: V4L/DVB (4582): VIDIOC_INT_S_REGISTER is IOW, not IOR.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/media/v4l2-common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index ce24a6ad445c..aecc946980a3 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -173,7 +173,7 @@ enum v4l2_chip_ident {
 #define VIDIOC_INT_S_STANDBY 	     _IOW('d', 94, u32)
 
 /* only implemented if CONFIG_VIDEO_ADV_DEBUG is defined */
-#define	VIDIOC_INT_S_REGISTER 		_IOR ('d', 100, struct v4l2_register)
+#define	VIDIOC_INT_S_REGISTER 		_IOW ('d', 100, struct v4l2_register)
 #define	VIDIOC_INT_G_REGISTER 		_IOWR('d', 101, struct v4l2_register)
 
 /* Generic reset command. The argument selects which subsystems to reset.
-- 
cgit v1.2.3


From 616b8b639e6491cfa63f79238a5c6fbee383eb09 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Fri, 1 Sep 2006 18:36:48 -0300
Subject: V4L/DVB (4585): VIDIOC_G_SLICED_VBI_CAP now accepts a v4l2_buf_type,
 make it IOWR

The VIDIOC_G_SLICED_VBI_CAP needs to receive the v4l2_buf_type field before
it can return a result. Hence this ioctl must be IOWR, not IOR. Since this
ioctl is still marked experimental we can make this change.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 include/linux/videodev2.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index d5746d470c73..44c59da26ed2 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -1243,7 +1243,7 @@ struct v4l2_streamparm
 #define VIDIOC_G_PRIORITY       _IOR  ('V', 67, enum v4l2_priority)
 #define VIDIOC_S_PRIORITY       _IOW  ('V', 68, enum v4l2_priority)
 #if 1
-#define VIDIOC_G_SLICED_VBI_CAP _IOR  ('V', 69, struct v4l2_sliced_vbi_cap)
+#define VIDIOC_G_SLICED_VBI_CAP _IOWR ('V', 69, struct v4l2_sliced_vbi_cap)
 #endif
 #define VIDIOC_LOG_STATUS       _IO   ('V', 70)
 #define VIDIOC_G_EXT_CTRLS	_IOWR ('V', 71, struct v4l2_ext_controls)
-- 
cgit v1.2.3


From b04c1baf485f4c3a25d8dbc187156030f9247cb0 Mon Sep 17 00:00:00 2001
From: Michal Majchrowicz <mmajchrowicz@gmail.com>
Date: Wed, 13 Sep 2006 16:42:42 -0300
Subject: V4L/DVB (4629): Saa7134: add card support for Proteus Pro 2309

Add card support for Proteus Pro 2309, based on saa7130 bridge

Signed-off-by: Michal Majchrowicz <mmajchrowicz@gmail.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/CARDLIST.saa7134  |  1 +
 drivers/media/common/ir-keymaps.c           | 31 +++++++++++++++++++++++
 drivers/media/video/saa7134/saa7134-cards.c | 39 ++++++++++++++++++++++++++++-
 drivers/media/video/saa7134/saa7134-input.c |  6 +++++
 drivers/media/video/saa7134/saa7134.h       |  1 +
 include/media/ir-common.h                   |  1 +
 6 files changed, 78 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/Documentation/video4linux/CARDLIST.saa7134 b/Documentation/video4linux/CARDLIST.saa7134
index ac1357d47a27..94cf695b1378 100644
--- a/Documentation/video4linux/CARDLIST.saa7134
+++ b/Documentation/video4linux/CARDLIST.saa7134
@@ -96,3 +96,4 @@
  95 -> LifeView FlyVIDEO3000 (NTSC)             [5169:0138]
  96 -> Medion Md8800 Quadro                     [16be:0007,16be:0008]
  97 -> LifeView FlyDVB-S /Acorp TV134DS         [5168:0300,4e42:0300]
+ 98 -> Proteus Pro 2309                         [0919:2003]
diff --git a/drivers/media/common/ir-keymaps.c b/drivers/media/common/ir-keymaps.c
index ce1281445dde..db753443587a 100644
--- a/drivers/media/common/ir-keymaps.c
+++ b/drivers/media/common/ir-keymaps.c
@@ -32,6 +32,37 @@ IR_KEYTAB_TYPE ir_codes_empty[IR_KEYTAB_SIZE] = {
 
 EXPORT_SYMBOL_GPL(ir_codes_empty);
 
+/* Michal Majchrowicz <mmajchrowicz@gmail.com> */
+IR_KEYTAB_TYPE ir_codes_proteus_2309[IR_KEYTAB_SIZE] = {
+	/* numeric */
+	[ 0x00 ] = KEY_0,
+	[ 0x01 ] = KEY_1,
+	[ 0x02 ] = KEY_2,
+	[ 0x03 ] = KEY_3,
+	[ 0x04 ] = KEY_4,
+	[ 0x05 ] = KEY_5,
+	[ 0x06 ] = KEY_6,
+	[ 0x07 ] = KEY_7,
+	[ 0x08 ] = KEY_8,
+	[ 0x09 ] = KEY_9,
+
+	[ 0x5c ] = KEY_POWER,     /* power       */
+	[ 0x20 ] = KEY_F,         /* full screen */
+	[ 0x0f ] = KEY_BACKSPACE, /* recall      */
+	[ 0x1b ] = KEY_ENTER,     /* mute        */
+	[ 0x41 ] = KEY_RECORD,    /* record      */
+	[ 0x43 ] = KEY_STOP,      /* stop        */
+	[ 0x16 ] = KEY_S,
+	[ 0x1a ] = KEY_Q,         /* off         */
+	[ 0x2e ] = KEY_RED,
+	[ 0x1f ] = KEY_DOWN,      /* channel -   */
+	[ 0x1c ] = KEY_UP,        /* channel +   */
+	[ 0x10 ] = KEY_LEFT,      /* volume -    */
+	[ 0x1e ] = KEY_RIGHT,     /* volume +    */
+	[ 0x14 ] = KEY_F1,
+};
+
+EXPORT_SYMBOL_GPL(ir_codes_proteus_2309);
 /* Matt Jesson <dvb@jesson.eclipse.co.uk */
 IR_KEYTAB_TYPE ir_codes_avermedia_dvbt[IR_KEYTAB_SIZE] = {
 	[ 0x28 ] = KEY_0,         //'0' / 'enter'
diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c
index bd526ad79f5b..aa1db509f3d4 100644
--- a/drivers/media/video/saa7134/saa7134-cards.c
+++ b/drivers/media/video/saa7134/saa7134-cards.c
@@ -2934,7 +2934,37 @@ struct saa7134_board saa7134_boards[] = {
 			.amux = LINE1,
 		}},
 	},
-
+	[SAA7134_BOARD_PROTEUS_2309] = {
+		.name           = "Proteus Pro 2309",
+		.audio_clock    = 0x00187de7,
+		.tuner_type	= TUNER_PHILIPS_FM1216ME_MK3,
+		.radio_type     = UNSET,
+		.tuner_addr	= ADDR_UNSET,
+		.radio_addr	= ADDR_UNSET,
+		.tda9887_conf   = TDA9887_PRESENT,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 1,
+			.amux = LINE2,
+			.tv   = 1,
+		},{
+			.name = name_comp1,
+			.vmux = 0,
+			.amux = LINE2,
+		},{
+			.name = name_comp2,
+			.vmux = 3,
+			.amux = LINE2,
+		},{
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE2,
+		}},
+		.mute = {
+			.name = name_mute,
+			.amux = LINE1,
+		},
+	},
 };
 
 const unsigned int saa7134_bcount = ARRAY_SIZE(saa7134_boards);
@@ -3525,6 +3555,12 @@ struct pci_device_id saa7134_pci_tbl[] = {
 		.subvendor    = 0x1489,
 		.subdevice    = 0x0502,                /* Cardbus version */
 		.driver_data  = SAA7134_BOARD_FLYDVBT_DUO_CARDBUS,
+	},{
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x0919, /* Philips Proteus PRO 2309 */
+		.subdevice    = 0x2003,
+		.driver_data  = SAA7134_BOARD_PROTEUS_2309,
 	},{
 		/* --- boards without eeprom + subsystem ID --- */
 		.vendor       = PCI_VENDOR_ID_PHILIPS,
@@ -3628,6 +3664,7 @@ int saa7134_board_init1(struct saa7134_dev *dev)
 	case SAA7134_BOARD_SEDNA_PC_TV_CARDBUS:
 	case SAA7134_BOARD_FLYDVBT_LR301:
 	case SAA7134_BOARD_FLYDVBTDUO:
+	case SAA7134_BOARD_PROTEUS_2309:
 		dev->has_remote = SAA7134_REMOTE_GPIO;
 		break;
 	case SAA7134_BOARD_FLYDVBS_LR300:
diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c
index 7c595492c56b..f7ea857d5d73 100644
--- a/drivers/media/video/saa7134/saa7134-input.c
+++ b/drivers/media/video/saa7134/saa7134-input.c
@@ -228,6 +228,12 @@ int saa7134_input_init1(struct saa7134_dev *dev)
 		mask_keyup   = 0x400000;
 		polling      = 50; // ms
 		break;
+	case SAA7134_BOARD_PROTEUS_2309:
+		ir_codes     = ir_codes_proteus_2309;
+		mask_keycode = 0x00007F;
+		mask_keyup   = 0x000080;
+		polling      = 50; // ms
+		break;
 	case SAA7134_BOARD_VIDEOMATE_DVBT_300:
 	case SAA7134_BOARD_VIDEOMATE_DVBT_200:
 		ir_codes     = ir_codes_videomate_tv_pvr;
diff --git a/drivers/media/video/saa7134/saa7134.h b/drivers/media/video/saa7134/saa7134.h
index c2f5181c770c..7db7b9705953 100644
--- a/drivers/media/video/saa7134/saa7134.h
+++ b/drivers/media/video/saa7134/saa7134.h
@@ -225,6 +225,7 @@ struct saa7134_format {
 #define SAA7134_BOARD_FLYVIDEO3000_NTSC 95
 #define SAA7134_BOARD_MEDION_MD8800_QUADRO 96
 #define SAA7134_BOARD_FLYDVBS_LR300 97
+#define SAA7134_BOARD_PROTEUS_2309 98
 
 #define SAA7134_MAXBOARDS 8
 #define SAA7134_INPUT_MAX 8
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index f883bc3aa70e..8f58406533c6 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -91,6 +91,7 @@ extern IR_KEYTAB_TYPE ir_codes_pinnacle_color[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_hauppauge_new[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_npgtech[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_norwood[IR_KEYTAB_SIZE];
+extern IR_KEYTAB_TYPE ir_codes_proteus_2309[IR_KEYTAB_SIZE];
 
 #endif
 
-- 
cgit v1.2.3


From 632bbfeee4f042c05bc65150b4433a297d3fe387 Mon Sep 17 00:00:00 2001
From: Jan Blunck <jblunck@suse.de>
Date: Mon, 25 Sep 2006 23:30:53 -0700
Subject: [PATCH] trigger a syntax error if percpu macros are incorrectly used

get_cpu_var()/per_cpu()/__get_cpu_var() arguments must be simple
identifiers.  Otherwise the arch dependent implementations might break.

This patch enforces the correct usage of the macros by producing a syntax
error if the variable is not a simple identifier.

Signed-off-by: Jan Blunck <jblunck@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/percpu.h |  4 +++-
 include/asm-s390/percpu.h    | 20 +++++++++++---------
 include/asm-x86_64/percpu.h  | 12 +++++++++---
 include/linux/percpu.h       | 10 ++++++++--
 4 files changed, 31 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index e160e04290fb..6d45ee5472af 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -14,7 +14,9 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
     __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
 
 /* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]))
+#define per_cpu(var, cpu) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]); }))
 #define __get_cpu_var(var) per_cpu(var, smp_processor_id())
 #define __raw_get_cpu_var(var) per_cpu(var, raw_smp_processor_id())
 
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index 28b3517e787c..495ad99c7635 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -15,18 +15,20 @@
  */
 #if defined(__s390x__) && defined(MODULE)
 
-#define __reloc_hide(var,offset) \
-  (*({ unsigned long *__ptr; \
-       asm ( "larl %0,per_cpu__"#var"@GOTENT" \
-             : "=a" (__ptr) : "X" (per_cpu__##var) ); \
-       (typeof(&per_cpu__##var))((*__ptr) + (offset)); }))
+#define __reloc_hide(var,offset) (*({			\
+	extern int simple_indentifier_##var(void);	\
+	unsigned long *__ptr;				\
+	asm ( "larl %0,per_cpu__"#var"@GOTENT"		\
+	    : "=a" (__ptr) : "X" (per_cpu__##var) );	\
+	(typeof(&per_cpu__##var))((*__ptr) + (offset));	}))
 
 #else
 
-#define __reloc_hide(var, offset) \
-  (*({ unsigned long __ptr; \
-       asm ( "" : "=a" (__ptr) : "0" (&per_cpu__##var) ); \
-       (typeof(&per_cpu__##var)) (__ptr + (offset)); }))
+#define __reloc_hide(var, offset) (*({				\
+	extern int simple_indentifier_##var(void);		\
+	unsigned long __ptr;					\
+	asm ( "" : "=a" (__ptr) : "0" (&per_cpu__##var) );	\
+	(typeof(&per_cpu__##var)) (__ptr + (offset)); }))
 
 #endif
 
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 08dd9f9dda81..bffb2f886a51 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -21,9 +21,15 @@
     __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
 
 /* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)))
-#define __get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()))
-#define __raw_get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()))
+#define per_cpu(var, cpu) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)); }))
+#define __get_cpu_var(var) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()); }))
+#define __raw_get_cpu_var(var) (*({			\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()); }))
 
 /* A macro to avoid #include hell... */
 #define percpu_modcopy(pcpudst, src, size)			\
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index cb9039a21f2a..f926490a7d8b 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -11,8 +11,14 @@
 #define PERCPU_ENOUGH_ROOM 32768
 #endif
 
-/* Must be an lvalue. */
-#define get_cpu_var(var) (*({ preempt_disable(); &__get_cpu_var(var); }))
+/*
+ * Must be an lvalue. Since @var must be a simple identifier,
+ * we force a syntax error here if it isn't.
+ */
+#define get_cpu_var(var) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	preempt_disable();				\
+	&__get_cpu_var(var); }))
 #define put_cpu_var(var) preempt_enable()
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.3


From a6ca1b99ed434f3fb41bbed647ed36c0420501e5 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Mon, 25 Sep 2006 23:30:55 -0700
Subject: [PATCH] update to the kernel kmap/kunmap API

Give non-highmem architectures access to the kmap API for the purposes of
overriding (this is what the attached patch does).

The proposal is that we should now require all architectures with coherence
issues to manage data coherence via the kmap/kunmap API.  Thus driver
writers never have to write code like

    kmap(page)
    modify data in page
    flush_kernel_dcache_page(page)
    kunmap(page)

instead, kmap/kunmap will manage the coherence and driver (and filesystem)
writers don't need to worry about how to flush between kmap and kunmap.

For most architectures, the page only needs to be flushed if it was
actually written to *and* there are user mappings of it, so the best
implementation looks to be: clear the page dirty pte bit in the kernel page
tables on kmap and on kunmap, check page->mappings for user maps, and then
the dirty bit, and only flush if it both has user mappings and is dirty.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/highmem.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/highmem.h b/include/linux/highmem.h
index 85ce7ef9a512..42620e723abb 100644
--- a/include/linux/highmem.h
+++ b/include/linux/highmem.h
@@ -29,6 +29,7 @@ unsigned int nr_free_highpages(void);
 
 static inline unsigned int nr_free_highpages(void) { return 0; }
 
+#ifndef ARCH_HAS_KMAP
 static inline void *kmap(struct page *page)
 {
 	might_sleep();
@@ -41,6 +42,7 @@ static inline void *kmap(struct page *page)
 #define kunmap_atomic(addr, idx)	do { } while (0)
 #define kmap_atomic_pfn(pfn, idx)	page_address(pfn_to_page(pfn))
 #define kmap_atomic_to_page(ptr)	virt_to_page(ptr)
+#endif
 
 #endif /* CONFIG_HIGHMEM */
 
-- 
cgit v1.2.3


From 725d704ecaca4a43f067092c140d4f3271cf2856 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:30:55 -0700
Subject: [PATCH] mm: VM_BUG_ON

Introduce a VM_BUG_ON, which is turned on with CONFIG_DEBUG_VM.  Use this
in the lightweight, inline refcounting functions; PageLRU and PageActive
checks in vmscan, because they're pretty well confined to vmscan.  And in
page allocate/free fastpaths which can be the hottest parts of the kernel
for kbuilds.

Unlike BUG_ON, VM_BUG_ON must not be used to execute statements with
side-effects, and should not be used outside core mm code.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h | 10 +++++++++-
 mm/internal.h      |  4 ++--
 mm/page_alloc.c    | 23 +++++++++++------------
 mm/swap.c          | 12 ++++++------
 mm/vmscan.c        | 16 ++++++++--------
 5 files changed, 36 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 224178a000d2..7d20b25c58fc 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -278,6 +278,12 @@ struct page {
  */
 #include <linux/page-flags.h>
 
+#ifdef CONFIG_DEBUG_VM
+#define VM_BUG_ON(cond) BUG_ON(cond)
+#else
+#define VM_BUG_ON(condition) do { } while(0)
+#endif
+
 /*
  * Methods to modify the page usage count.
  *
@@ -297,7 +303,7 @@ struct page {
  */
 static inline int put_page_testzero(struct page *page)
 {
-	BUG_ON(atomic_read(&page->_count) == 0);
+	VM_BUG_ON(atomic_read(&page->_count) == 0);
 	return atomic_dec_and_test(&page->_count);
 }
 
@@ -307,6 +313,7 @@ static inline int put_page_testzero(struct page *page)
  */
 static inline int get_page_unless_zero(struct page *page)
 {
+	VM_BUG_ON(PageCompound(page));
 	return atomic_inc_not_zero(&page->_count);
 }
 
@@ -323,6 +330,7 @@ static inline void get_page(struct page *page)
 {
 	if (unlikely(PageCompound(page)))
 		page = (struct page *)page_private(page);
+	VM_BUG_ON(atomic_read(&page->_count) == 0);
 	atomic_inc(&page->_count);
 }
 
diff --git a/mm/internal.h b/mm/internal.h
index d20e3cc4aef0..d527b80b292f 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -24,8 +24,8 @@ static inline void set_page_count(struct page *page, int v)
  */
 static inline void set_page_refcounted(struct page *page)
 {
-	BUG_ON(PageCompound(page) && page_private(page) != (unsigned long)page);
-	BUG_ON(atomic_read(&page->_count));
+	VM_BUG_ON(PageCompound(page) && page_private(page) != (unsigned long)page);
+	VM_BUG_ON(atomic_read(&page->_count));
 	set_page_count(page, 1);
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8a52ba9fe693..4b33878e9488 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -127,7 +127,6 @@ static int bad_range(struct zone *zone, struct page *page)
 
 	return 0;
 }
-
 #else
 static inline int bad_range(struct zone *zone, struct page *page)
 {
@@ -218,12 +217,12 @@ static inline void prep_zero_page(struct page *page, int order, gfp_t gfp_flags)
 {
 	int i;
 
-	BUG_ON((gfp_flags & (__GFP_WAIT | __GFP_HIGHMEM)) == __GFP_HIGHMEM);
+	VM_BUG_ON((gfp_flags & (__GFP_WAIT | __GFP_HIGHMEM)) == __GFP_HIGHMEM);
 	/*
 	 * clear_highpage() will use KM_USER0, so it's a bug to use __GFP_ZERO
 	 * and __GFP_HIGHMEM from hard or soft interrupt context.
 	 */
-	BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt());
+	VM_BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt());
 	for (i = 0; i < (1 << order); i++)
 		clear_highpage(page + i);
 }
@@ -347,8 +346,8 @@ static inline void __free_one_page(struct page *page,
 
 	page_idx = page_to_pfn(page) & ((1 << MAX_ORDER) - 1);
 
-	BUG_ON(page_idx & (order_size - 1));
-	BUG_ON(bad_range(zone, page));
+	VM_BUG_ON(page_idx & (order_size - 1));
+	VM_BUG_ON(bad_range(zone, page));
 
 	zone->free_pages += order_size;
 	while (order < MAX_ORDER-1) {
@@ -421,7 +420,7 @@ static void free_pages_bulk(struct zone *zone, int count,
 	while (count--) {
 		struct page *page;
 
-		BUG_ON(list_empty(list));
+		VM_BUG_ON(list_empty(list));
 		page = list_entry(list->prev, struct page, lru);
 		/* have to delete it as __free_one_page list manipulates */
 		list_del(&page->lru);
@@ -512,7 +511,7 @@ static inline void expand(struct zone *zone, struct page *page,
 		area--;
 		high--;
 		size >>= 1;
-		BUG_ON(bad_range(zone, &page[size]));
+		VM_BUG_ON(bad_range(zone, &page[size]));
 		list_add(&page[size].lru, &area->free_list);
 		area->nr_free++;
 		set_page_order(&page[size], high);
@@ -761,8 +760,8 @@ void split_page(struct page *page, unsigned int order)
 {
 	int i;
 
-	BUG_ON(PageCompound(page));
-	BUG_ON(!page_count(page));
+	VM_BUG_ON(PageCompound(page));
+	VM_BUG_ON(!page_count(page));
 	for (i = 1; i < (1 << order); i++)
 		set_page_refcounted(page + i);
 }
@@ -809,7 +808,7 @@ again:
 	local_irq_restore(flags);
 	put_cpu();
 
-	BUG_ON(bad_range(zone, page));
+	VM_BUG_ON(bad_range(zone, page));
 	if (prep_new_page(page, order, gfp_flags))
 		goto again;
 	return page;
@@ -1083,7 +1082,7 @@ fastcall unsigned long get_zeroed_page(gfp_t gfp_mask)
 	 * get_zeroed_page() returns a 32-bit address, which cannot represent
 	 * a highmem page
 	 */
-	BUG_ON((gfp_mask & __GFP_HIGHMEM) != 0);
+	VM_BUG_ON((gfp_mask & __GFP_HIGHMEM) != 0);
 
 	page = alloc_pages(gfp_mask | __GFP_ZERO, 0);
 	if (page)
@@ -1116,7 +1115,7 @@ EXPORT_SYMBOL(__free_pages);
 fastcall void free_pages(unsigned long addr, unsigned int order)
 {
 	if (addr != 0) {
-		BUG_ON(!virt_addr_valid((void *)addr));
+		VM_BUG_ON(!virt_addr_valid((void *)addr));
 		__free_pages(virt_to_page((void *)addr), order);
 	}
 }
diff --git a/mm/swap.c b/mm/swap.c
index 687686a61f7c..600235e43704 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -233,7 +233,7 @@ void fastcall __page_cache_release(struct page *page)
 		struct zone *zone = page_zone(page);
 
 		spin_lock_irqsave(&zone->lru_lock, flags);
-		BUG_ON(!PageLRU(page));
+		VM_BUG_ON(!PageLRU(page));
 		__ClearPageLRU(page);
 		del_page_from_lru(zone, page);
 		spin_unlock_irqrestore(&zone->lru_lock, flags);
@@ -284,7 +284,7 @@ void release_pages(struct page **pages, int nr, int cold)
 				zone = pagezone;
 				spin_lock_irq(&zone->lru_lock);
 			}
-			BUG_ON(!PageLRU(page));
+			VM_BUG_ON(!PageLRU(page));
 			__ClearPageLRU(page);
 			del_page_from_lru(zone, page);
 		}
@@ -337,7 +337,7 @@ void __pagevec_release_nonlru(struct pagevec *pvec)
 	for (i = 0; i < pagevec_count(pvec); i++) {
 		struct page *page = pvec->pages[i];
 
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		if (put_page_testzero(page))
 			pagevec_add(&pages_to_free, page);
 	}
@@ -364,7 +364,7 @@ void __pagevec_lru_add(struct pagevec *pvec)
 			zone = pagezone;
 			spin_lock_irq(&zone->lru_lock);
 		}
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
 		add_page_to_inactive_list(zone, page);
 	}
@@ -391,9 +391,9 @@ void __pagevec_lru_add_active(struct pagevec *pvec)
 			zone = pagezone;
 			spin_lock_irq(&zone->lru_lock);
 		}
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
-		BUG_ON(PageActive(page));
+		VM_BUG_ON(PageActive(page));
 		SetPageActive(page);
 		add_page_to_active_list(zone, page);
 	}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5d4c4d02254d..41a3da3d6ccc 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -440,7 +440,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 		if (TestSetPageLocked(page))
 			goto keep;
 
-		BUG_ON(PageActive(page));
+		VM_BUG_ON(PageActive(page));
 
 		sc->nr_scanned++;
 
@@ -564,7 +564,7 @@ keep_locked:
 		unlock_page(page);
 keep:
 		list_add(&page->lru, &ret_pages);
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 	}
 	list_splice(&ret_pages, page_list);
 	if (pagevec_count(&freed_pvec))
@@ -603,7 +603,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
 		page = lru_to_page(src);
 		prefetchw_prev_lru_page(page, src, flags);
 
-		BUG_ON(!PageLRU(page));
+		VM_BUG_ON(!PageLRU(page));
 
 		list_del(&page->lru);
 		target = src;
@@ -674,7 +674,7 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 		 */
 		while (!list_empty(&page_list)) {
 			page = lru_to_page(&page_list);
-			BUG_ON(PageLRU(page));
+			VM_BUG_ON(PageLRU(page));
 			SetPageLRU(page);
 			list_del(&page->lru);
 			if (PageActive(page))
@@ -797,9 +797,9 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	while (!list_empty(&l_inactive)) {
 		page = lru_to_page(&l_inactive);
 		prefetchw_prev_lru_page(page, &l_inactive, flags);
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
-		BUG_ON(!PageActive(page));
+		VM_BUG_ON(!PageActive(page));
 		ClearPageActive(page);
 
 		list_move(&page->lru, &zone->inactive_list);
@@ -827,9 +827,9 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	while (!list_empty(&l_active)) {
 		page = lru_to_page(&l_active);
 		prefetchw_prev_lru_page(page, &l_active, flags);
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
-		BUG_ON(!PageActive(page));
+		VM_BUG_ON(!PageActive(page));
 		list_move(&page->lru, &zone->active_list);
 		pgmoved++;
 		if (!pagevec_add(&pvec, page)) {
-- 
cgit v1.2.3


From d08b3851da41d0ee60851f2c75b118e1f7a5fc89 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:30:57 -0700
Subject: [PATCH] mm: tracking shared dirty pages

Tracking of dirty pages in shared writeable mmap()s.

The idea is simple: write protect clean shared writeable pages, catch the
write-fault, make writeable and set dirty.  On page write-back clean all the
PTE dirty bits and write protect them once again.

The implementation is a tad harder, mainly because the default
backing_dev_info capabilities were too loosely maintained.  Hence it is not
enough to test the backing_dev_info for cap_account_dirty.

The current heuristic is as follows, a VMA is eligible when:
 - its shared writeable
    (vm_flags & (VM_WRITE|VM_SHARED)) == (VM_WRITE|VM_SHARED)
 - it is not a 'special' mapping
    (vm_flags & (VM_PFNMAP|VM_INSERTPAGE)) == 0
 - the backing_dev_info is cap_account_dirty
    mapping_cap_account_dirty(vma->vm_file->f_mapping)
 - f_op->mmap() didn't change the default page protection

Page from remap_pfn_range() are explicitly excluded because their COW
semantics are already horrid enough (see vm_normal_page() in do_wp_page()) and
because they don't have a backing store anyway.

mprotect() is taught about the new behaviour as well.  However it overrides
the last condition.

Cleaning the pages on write-back is done with page_mkclean() a new rmap call.
It can be called on any page, but is currently only implemented for mapped
pages, if the page is found the be of a VMA that accounts dirty pages it will
also wrprotect the PTE.

Finally, in fs/buffers.c:try_to_free_buffers(); remove clear_page_dirty() from
under ->private_lock.  This seems to be safe, since ->private_lock is used to
serialize access to the buffers, not the page itself.  This is needed because
clear_page_dirty() will call into page_mkclean() and would thereby violate
locking order.

[dhowells@redhat.com: Provide a page_mkclean() implementation for NOMMU]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/buffer.c          |  2 +-
 include/linux/mm.h   | 34 +++++++++++++++++++++++++++
 include/linux/rmap.h | 14 +++++++++++
 mm/memory.c          | 29 ++++++++++++++++++-----
 mm/mmap.c            | 10 ++++----
 mm/mprotect.c        | 21 +++++++----------
 mm/page-writeback.c  | 17 ++++++++++----
 mm/rmap.c            | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 8 files changed, 162 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/fs/buffer.c b/fs/buffer.c
index 71649ef9b658..3b6d701073e7 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -2987,6 +2987,7 @@ int try_to_free_buffers(struct page *page)
 
 	spin_lock(&mapping->private_lock);
 	ret = drop_buffers(page, &buffers_to_free);
+	spin_unlock(&mapping->private_lock);
 	if (ret) {
 		/*
 		 * If the filesystem writes its buffers by hand (eg ext3)
@@ -2998,7 +2999,6 @@ int try_to_free_buffers(struct page *page)
 		 */
 		clear_page_dirty(page);
 	}
-	spin_unlock(&mapping->private_lock);
 out:
 	if (buffers_to_free) {
 		struct buffer_head *bh = buffers_to_free;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7d20b25c58fc..449841413cf1 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -15,6 +15,7 @@
 #include <linux/fs.h>
 #include <linux/mutex.h>
 #include <linux/debug_locks.h>
+#include <linux/backing-dev.h>
 
 struct mempolicy;
 struct anon_vma;
@@ -810,6 +811,39 @@ struct shrinker;
 extern struct shrinker *set_shrinker(int, shrinker_t);
 extern void remove_shrinker(struct shrinker *shrinker);
 
+/*
+ * Some shared mappigns will want the pages marked read-only
+ * to track write events. If so, we'll downgrade vm_page_prot
+ * to the private version (using protection_map[] without the
+ * VM_SHARED bit).
+ */
+static inline int vma_wants_writenotify(struct vm_area_struct *vma)
+{
+	unsigned int vm_flags = vma->vm_flags;
+
+	/* If it was private or non-writable, the write bit is already clear */
+	if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED)))
+		return 0;
+
+	/* The backer wishes to know when pages are first written to? */
+	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
+		return 1;
+
+	/* The open routine did something to the protections already? */
+	if (pgprot_val(vma->vm_page_prot) !=
+	    pgprot_val(protection_map[vm_flags &
+		    (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]))
+		return 0;
+
+	/* Specialty mapping? */
+	if (vm_flags & (VM_PFNMAP|VM_INSERTPAGE))
+		return 0;
+
+	/* Can the mapping track the dirty pages? */
+	return vma->vm_file && vma->vm_file->f_mapping &&
+		mapping_cap_account_dirty(vma->vm_file->f_mapping);
+}
+
 extern pte_t *FASTCALL(get_locked_pte(struct mm_struct *mm, unsigned long addr, spinlock_t **ptl));
 
 int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address);
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index bf97b0900014..db2c1df4fef9 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -103,6 +103,14 @@ pte_t *page_check_address(struct page *, struct mm_struct *,
  */
 unsigned long page_address_in_vma(struct page *, struct vm_area_struct *);
 
+/*
+ * Cleans the PTEs of shared mappings.
+ * (and since clean PTEs should also be readonly, write protects them too)
+ *
+ * returns the number of cleaned PTEs.
+ */
+int page_mkclean(struct page *);
+
 #else	/* !CONFIG_MMU */
 
 #define anon_vma_init()		do {} while (0)
@@ -112,6 +120,12 @@ unsigned long page_address_in_vma(struct page *, struct vm_area_struct *);
 #define page_referenced(page,l) TestClearPageReferenced(page)
 #define try_to_unmap(page, refs) SWAP_FAIL
 
+static inline int page_mkclean(struct page *page)
+{
+	return 0;
+}
+
+
 #endif	/* CONFIG_MMU */
 
 /*
diff --git a/mm/memory.c b/mm/memory.c
index 109e9866237e..fa941b169071 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1458,14 +1458,19 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 {
 	struct page *old_page, *new_page;
 	pte_t entry;
-	int reuse, ret = VM_FAULT_MINOR;
+	int reuse = 0, ret = VM_FAULT_MINOR;
+	struct page *dirty_page = NULL;
 
 	old_page = vm_normal_page(vma, address, orig_pte);
 	if (!old_page)
 		goto gotten;
 
-	if (unlikely((vma->vm_flags & (VM_SHARED|VM_WRITE)) ==
-				(VM_SHARED|VM_WRITE))) {
+	/*
+	 * Only catch write-faults on shared writable pages, read-only
+	 * shared pages can get COWed by get_user_pages(.write=1, .force=1).
+	 */
+	if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
+					(VM_WRITE|VM_SHARED))) {
 		if (vma->vm_ops && vma->vm_ops->page_mkwrite) {
 			/*
 			 * Notify the address space that the page is about to
@@ -1494,13 +1499,12 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			if (!pte_same(*page_table, orig_pte))
 				goto unlock;
 		}
-
+		dirty_page = old_page;
+		get_page(dirty_page);
 		reuse = 1;
 	} else if (PageAnon(old_page) && !TestSetPageLocked(old_page)) {
 		reuse = can_share_swap_page(old_page);
 		unlock_page(old_page);
-	} else {
-		reuse = 0;
 	}
 
 	if (reuse) {
@@ -1566,6 +1570,10 @@ gotten:
 		page_cache_release(old_page);
 unlock:
 	pte_unmap_unlock(page_table, ptl);
+	if (dirty_page) {
+		set_page_dirty(dirty_page);
+		put_page(dirty_page);
+	}
 	return ret;
 oom:
 	if (old_page)
@@ -2098,6 +2106,7 @@ static int do_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	unsigned int sequence = 0;
 	int ret = VM_FAULT_MINOR;
 	int anon = 0;
+	struct page *dirty_page = NULL;
 
 	pte_unmap(page_table);
 	BUG_ON(vma->vm_flags & VM_PFNMAP);
@@ -2192,6 +2201,10 @@ retry:
 		} else {
 			inc_mm_counter(mm, file_rss);
 			page_add_file_rmap(new_page);
+			if (write_access) {
+				dirty_page = new_page;
+				get_page(dirty_page);
+			}
 		}
 	} else {
 		/* One of our sibling threads was faster, back out. */
@@ -2204,6 +2217,10 @@ retry:
 	lazy_mmu_prot_update(entry);
 unlock:
 	pte_unmap_unlock(page_table, ptl);
+	if (dirty_page) {
+		set_page_dirty(dirty_page);
+		put_page(dirty_page);
+	}
 	return ret;
 oom:
 	page_cache_release(new_page);
diff --git a/mm/mmap.c b/mm/mmap.c
index d799d896d74a..8507ee9cd573 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1105,12 +1105,6 @@ munmap_back:
 			goto free_vma;
 	}
 
-	/* Don't make the VMA automatically writable if it's shared, but the
-	 * backer wishes to know when pages are first written to */
-	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
-		vma->vm_page_prot =
-			protection_map[vm_flags & (VM_READ|VM_WRITE|VM_EXEC)];
-
 	/* We set VM_ACCOUNT in a shared mapping's vm_flags, to inform
 	 * shmem_zero_setup (perhaps called through /dev/zero's ->mmap)
 	 * that memory reservation must be checked; but that reservation
@@ -1128,6 +1122,10 @@ munmap_back:
 	pgoff = vma->vm_pgoff;
 	vm_flags = vma->vm_flags;
 
+	if (vma_wants_writenotify(vma))
+		vma->vm_page_prot =
+			protection_map[vm_flags & (VM_READ|VM_WRITE|VM_EXEC)];
+
 	if (!file || !vma_merge(mm, prev, addr, vma->vm_end,
 			vma->vm_flags, NULL, file, pgoff, vma_policy(vma))) {
 		file = vma->vm_file;
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 638edabaff71..367b7f6c0637 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -123,8 +123,6 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	unsigned long oldflags = vma->vm_flags;
 	long nrpages = (end - start) >> PAGE_SHIFT;
 	unsigned long charged = 0;
-	unsigned int mask;
-	pgprot_t newprot;
 	pgoff_t pgoff;
 	int error;
 
@@ -176,24 +174,21 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	}
 
 success:
-	/* Don't make the VMA automatically writable if it's shared, but the
-	 * backer wishes to know when pages are first written to */
-	mask = VM_READ|VM_WRITE|VM_EXEC|VM_SHARED;
-	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
-		mask &= ~VM_SHARED;
-
-	newprot = protection_map[newflags & mask];
-
 	/*
 	 * vm_flags and vm_page_prot are protected by the mmap_sem
 	 * held in write mode.
 	 */
 	vma->vm_flags = newflags;
-	vma->vm_page_prot = newprot;
+	vma->vm_page_prot = protection_map[newflags &
+		(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
+	if (vma_wants_writenotify(vma))
+		vma->vm_page_prot = protection_map[newflags &
+			(VM_READ|VM_WRITE|VM_EXEC)];
+
 	if (is_vm_hugetlb_page(vma))
-		hugetlb_change_protection(vma, start, end, newprot);
+		hugetlb_change_protection(vma, start, end, vma->vm_page_prot);
 	else
-		change_protection(vma, start, end, newprot);
+		change_protection(vma, start, end, vma->vm_page_prot);
 	vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
 	vm_stat_account(mm, newflags, vma->vm_file, nrpages);
 	return 0;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 77a0bc4e261a..1c87430b7a25 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -23,6 +23,7 @@
 #include <linux/backing-dev.h>
 #include <linux/blkdev.h>
 #include <linux/mpage.h>
+#include <linux/rmap.h>
 #include <linux/percpu.h>
 #include <linux/notifier.h>
 #include <linux/smp.h>
@@ -550,7 +551,7 @@ int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
 		return 0;
 	wbc->for_writepages = 1;
 	if (mapping->a_ops->writepages)
-		ret =  mapping->a_ops->writepages(mapping, wbc);
+		ret = mapping->a_ops->writepages(mapping, wbc);
 	else
 		ret = generic_writepages(mapping, wbc);
 	wbc->for_writepages = 0;
@@ -712,9 +713,15 @@ int test_clear_page_dirty(struct page *page)
 			radix_tree_tag_clear(&mapping->page_tree,
 						page_index(page),
 						PAGECACHE_TAG_DIRTY);
-			if (mapping_cap_account_dirty(mapping))
-				__dec_zone_page_state(page, NR_FILE_DIRTY);
 			write_unlock_irqrestore(&mapping->tree_lock, flags);
+			/*
+			 * We can continue to use `mapping' here because the
+			 * page is locked, which pins the address_space
+			 */
+			if (mapping_cap_account_dirty(mapping)) {
+				page_mkclean(page);
+				dec_zone_page_state(page, NR_FILE_DIRTY);
+			}
 			return 1;
 		}
 		write_unlock_irqrestore(&mapping->tree_lock, flags);
@@ -744,8 +751,10 @@ int clear_page_dirty_for_io(struct page *page)
 
 	if (mapping) {
 		if (TestClearPageDirty(page)) {
-			if (mapping_cap_account_dirty(mapping))
+			if (mapping_cap_account_dirty(mapping)) {
+				page_mkclean(page);
 				dec_zone_page_state(page, NR_FILE_DIRTY);
+			}
 			return 1;
 		}
 		return 0;
diff --git a/mm/rmap.c b/mm/rmap.c
index 40158b59729e..e2155d791d99 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -434,6 +434,71 @@ int page_referenced(struct page *page, int is_locked)
 	return referenced;
 }
 
+static int page_mkclean_one(struct page *page, struct vm_area_struct *vma)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	unsigned long address;
+	pte_t *pte, entry;
+	spinlock_t *ptl;
+	int ret = 0;
+
+	address = vma_address(page, vma);
+	if (address == -EFAULT)
+		goto out;
+
+	pte = page_check_address(page, mm, address, &ptl);
+	if (!pte)
+		goto out;
+
+	if (!pte_dirty(*pte) && !pte_write(*pte))
+		goto unlock;
+
+	entry = ptep_get_and_clear(mm, address, pte);
+	entry = pte_mkclean(entry);
+	entry = pte_wrprotect(entry);
+	ptep_establish(vma, address, pte, entry);
+	lazy_mmu_prot_update(entry);
+	ret = 1;
+
+unlock:
+	pte_unmap_unlock(pte, ptl);
+out:
+	return ret;
+}
+
+static int page_mkclean_file(struct address_space *mapping, struct page *page)
+{
+	pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
+	struct vm_area_struct *vma;
+	struct prio_tree_iter iter;
+	int ret = 0;
+
+	BUG_ON(PageAnon(page));
+
+	spin_lock(&mapping->i_mmap_lock);
+	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) {
+		if (vma->vm_flags & VM_SHARED)
+			ret += page_mkclean_one(page, vma);
+	}
+	spin_unlock(&mapping->i_mmap_lock);
+	return ret;
+}
+
+int page_mkclean(struct page *page)
+{
+	int ret = 0;
+
+	BUG_ON(!PageLocked(page));
+
+	if (page_mapped(page)) {
+		struct address_space *mapping = page_mapping(page);
+		if (mapping)
+			ret = page_mkclean_file(mapping, page);
+	}
+
+	return ret;
+}
+
 /**
  * page_set_anon_rmap - setup new anonymous rmap
  * @page:	the page to add the mapping to
-- 
cgit v1.2.3


From edc79b2a46ed854595e40edcf3f8b37f9f14aa3f Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:30:58 -0700
Subject: [PATCH] mm: balance dirty pages

Now that we can detect writers of shared mappings, throttle them.  Avoids OOM
by surprise.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/writeback.h |  1 +
 mm/memory.c               |  5 +++--
 mm/page-writeback.c       | 10 ++++++++++
 3 files changed, 14 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 0422036af4eb..56a23a0e7f2e 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -116,6 +116,7 @@ int sync_page_range(struct inode *inode, struct address_space *mapping,
 			loff_t pos, loff_t count);
 int sync_page_range_nolock(struct inode *inode, struct address_space *mapping,
 			   loff_t pos, loff_t count);
+void set_page_dirty_balance(struct page *page);
 
 /* pdflush.c */
 extern int nr_pdflush_threads;	/* Global so it can be exported to sysctl
diff --git a/mm/memory.c b/mm/memory.c
index fa941b169071..dd7d7fc5ed60 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -49,6 +49,7 @@
 #include <linux/module.h>
 #include <linux/delayacct.h>
 #include <linux/init.h>
+#include <linux/writeback.h>
 
 #include <asm/pgalloc.h>
 #include <asm/uaccess.h>
@@ -1571,7 +1572,7 @@ gotten:
 unlock:
 	pte_unmap_unlock(page_table, ptl);
 	if (dirty_page) {
-		set_page_dirty(dirty_page);
+		set_page_dirty_balance(dirty_page);
 		put_page(dirty_page);
 	}
 	return ret;
@@ -2218,7 +2219,7 @@ retry:
 unlock:
 	pte_unmap_unlock(page_table, ptl);
 	if (dirty_page) {
-		set_page_dirty(dirty_page);
+		set_page_dirty_balance(dirty_page);
 		put_page(dirty_page);
 	}
 	return ret;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 1c87430b7a25..b9f4c6f1be86 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -244,6 +244,16 @@ static void balance_dirty_pages(struct address_space *mapping)
 		pdflush_operation(background_writeout, 0);
 }
 
+void set_page_dirty_balance(struct page *page)
+{
+	if (set_page_dirty(page)) {
+		struct address_space *mapping = page_mapping(page);
+
+		if (mapping)
+			balance_dirty_pages_ratelimited(mapping);
+	}
+}
+
 /**
  * balance_dirty_pages_ratelimited_nr - balance dirty memory state
  * @mapping: address_space which was dirtied
-- 
cgit v1.2.3


From b221385bc41d6789edde3d2fa0cb20d5045730eb Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 25 Sep 2006 23:31:02 -0700
Subject: [PATCH] mm/: make functions static

This patch makes the following needlessly global functions static:
 - slab.c: kmem_find_general_cachep()
 - swap.c: __page_cache_release()
 - vmalloc.c: __vmalloc_node()

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h      |  2 --
 include/linux/slab.h    |  2 --
 include/linux/vmalloc.h |  2 --
 mm/slab.c               |  3 +--
 mm/swap.c               | 39 +++++++++++++++++++--------------------
 mm/vmalloc.c            |  8 +++++---
 6 files changed, 25 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 449841413cf1..45678b036955 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -318,8 +318,6 @@ static inline int get_page_unless_zero(struct page *page)
 	return atomic_inc_not_zero(&page->_count);
 }
 
-extern void FASTCALL(__page_cache_release(struct page *));
-
 static inline int page_count(struct page *page)
 {
 	if (unlikely(PageCompound(page)))
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 45ad55b70d1c..193c03c547ec 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -67,7 +67,6 @@ extern void *kmem_cache_zalloc(struct kmem_cache *, gfp_t);
 extern void kmem_cache_free(kmem_cache_t *, void *);
 extern unsigned int kmem_cache_size(kmem_cache_t *);
 extern const char *kmem_cache_name(kmem_cache_t *);
-extern kmem_cache_t *kmem_find_general_cachep(size_t size, gfp_t gfpflags);
 
 /* Size description struct for general caches. */
 struct cache_sizes {
@@ -223,7 +222,6 @@ extern int FASTCALL(kmem_ptr_validate(kmem_cache_t *cachep, void *ptr));
 /* SLOB allocator routines */
 
 void kmem_cache_init(void);
-struct kmem_cache *kmem_find_general_cachep(size_t, gfp_t gfpflags);
 struct kmem_cache *kmem_cache_create(const char *c, size_t, size_t,
 	unsigned long,
 	void (*)(void *, struct kmem_cache *, unsigned long),
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 71b6363caaaf..dee88c6b6fa7 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -44,8 +44,6 @@ extern void *vmalloc_32_user(unsigned long size);
 extern void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot);
 extern void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask,
 				pgprot_t prot);
-extern void *__vmalloc_node(unsigned long size, gfp_t gfp_mask,
-				pgprot_t prot, int node);
 extern void vfree(void *addr);
 
 extern void *vmap(struct page **pages, unsigned int count,
diff --git a/mm/slab.c b/mm/slab.c
index 21ba06035700..5870bcbd33cf 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -768,11 +768,10 @@ static inline struct kmem_cache *__find_general_cachep(size_t size,
 	return csizep->cs_cachep;
 }
 
-struct kmem_cache *kmem_find_general_cachep(size_t size, gfp_t gfpflags)
+static struct kmem_cache *kmem_find_general_cachep(size_t size, gfp_t gfpflags)
 {
 	return __find_general_cachep(size, gfpflags);
 }
-EXPORT_SYMBOL(kmem_find_general_cachep);
 
 static size_t slab_mgmt_size(size_t nr_objs, size_t align)
 {
diff --git a/mm/swap.c b/mm/swap.c
index 600235e43704..2e0e871f542f 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -34,6 +34,25 @@
 /* How many pages do we try to swap or page in/out together? */
 int page_cluster;
 
+/*
+ * This path almost never happens for VM activity - pages are normally
+ * freed via pagevecs.  But it gets used by networking.
+ */
+static void fastcall __page_cache_release(struct page *page)
+{
+	if (PageLRU(page)) {
+		unsigned long flags;
+		struct zone *zone = page_zone(page);
+
+		spin_lock_irqsave(&zone->lru_lock, flags);
+		VM_BUG_ON(!PageLRU(page));
+		__ClearPageLRU(page);
+		del_page_from_lru(zone, page);
+		spin_unlock_irqrestore(&zone->lru_lock, flags);
+	}
+	free_hot_page(page);
+}
+
 static void put_compound_page(struct page *page)
 {
 	page = (struct page *)page_private(page);
@@ -222,26 +241,6 @@ int lru_add_drain_all(void)
 }
 #endif
 
-/*
- * This path almost never happens for VM activity - pages are normally
- * freed via pagevecs.  But it gets used by networking.
- */
-void fastcall __page_cache_release(struct page *page)
-{
-	if (PageLRU(page)) {
-		unsigned long flags;
-		struct zone *zone = page_zone(page);
-
-		spin_lock_irqsave(&zone->lru_lock, flags);
-		VM_BUG_ON(!PageLRU(page));
-		__ClearPageLRU(page);
-		del_page_from_lru(zone, page);
-		spin_unlock_irqrestore(&zone->lru_lock, flags);
-	}
-	free_hot_page(page);
-}
-EXPORT_SYMBOL(__page_cache_release);
-
 /*
  * Batched page_cache_release().  Decrement the reference count on all the
  * passed pages.  If it fell to zero then remove the page from the LRU and
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 266162d2ba28..9aad8b0cc6ee 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -24,6 +24,9 @@
 DEFINE_RWLOCK(vmlist_lock);
 struct vm_struct *vmlist;
 
+static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
+			    int node);
+
 static void vunmap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end)
 {
 	pte_t *pte;
@@ -478,8 +481,8 @@ void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
  *	allocator with @gfp_mask flags.  Map them into contiguous
  *	kernel virtual space, using a pagetable protection of @prot.
  */
-void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
-			int node)
+static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
+			    int node)
 {
 	struct vm_struct *area;
 
@@ -493,7 +496,6 @@ void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
 
 	return __vmalloc_area_node(area, gfp_mask, prot, node);
 }
-EXPORT_SYMBOL(__vmalloc_node);
 
 void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
 {
-- 
cgit v1.2.3


From 91023300057e96de7f46e95166a3e02394ae72f9 Mon Sep 17 00:00:00 2001
From: keith mannthey <kmannth@us.ibm.com>
Date: Mon, 25 Sep 2006 23:31:03 -0700
Subject: [PATCH] convert i386 NUMA KVA space to bootmem

Address a long standing issue of booting with an initrd on an i386 numa
system.  Currently (and always) the numa kva area is mapped into low memory
by finding the end of low memory and moving that mark down (thus creating
space for the kva).  The issue with this is that Grub loads initrds into
this similar space so when the kernel check the initrd it finds it outside
max_low_pfn and disables it (it thinks the initrd is not mapped into usable
memory) thus initrd enabled kernels can't boot i386 numa :(

My solution to the problem just converts the numa kva area to use the
bootmem allocator to save it's area (instead of moving the end of low
memory).  Using bootmem allows the kva area to be mapped into more diverse
addresses (not just the end of low memory) and enables the kva area to be
mapped below the initrd if present.

I have tested this patch on numaq(no initrd) and summit(initrd) i386 numa
based systems.

[akpm@osdl.org: cleanups]
Signed-off-by: Keith Mannthey <kmannth@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/setup.c  |  3 ++-
 arch/i386/mm/discontig.c  | 29 +++++++++++++++++++++--------
 include/asm-i386/mmzone.h |  6 ++++++
 3 files changed, 29 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index f1682206d304..27d4dc0d3ef1 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -53,6 +53,7 @@
 #include <asm/apic.h>
 #include <asm/e820.h>
 #include <asm/mpspec.h>
+#include <asm/mmzone.h>
 #include <asm/setup.h>
 #include <asm/arch_hooks.h>
 #include <asm/sections.h>
@@ -1258,7 +1259,7 @@ void __init setup_bootmem_allocator(void)
 	 */
 	find_smp_config();
 #endif
-
+	numa_kva_reserve();
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (LOADER_TYPE && INITRD_START) {
 		if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) {
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 7c392dc553b8..2e36eff8aff9 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -117,7 +117,8 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags);
 
 void *node_remap_end_vaddr[MAX_NUMNODES];
 void *node_remap_alloc_vaddr[MAX_NUMNODES];
-
+static unsigned long kva_start_pfn;
+static unsigned long kva_pages;
 /*
  * FLAT - support for basic PC memory model with discontig enabled, essentially
  *        a single node with all available processors in it with a flat
@@ -286,7 +287,6 @@ unsigned long __init setup_memory(void)
 {
 	int nid;
 	unsigned long system_start_pfn, system_max_low_pfn;
-	unsigned long reserve_pages;
 
 	/*
 	 * When mapping a NUMA machine we allocate the node_mem_map arrays
@@ -298,14 +298,23 @@ unsigned long __init setup_memory(void)
 	find_max_pfn();
 	get_memcfg_numa();
 
-	reserve_pages = calculate_numa_remap_pages();
+	kva_pages = calculate_numa_remap_pages();
 
 	/* partially used pages are not usable - thus round upwards */
 	system_start_pfn = min_low_pfn = PFN_UP(init_pg_tables_end);
 
-	system_max_low_pfn = max_low_pfn = find_max_low_pfn() - reserve_pages;
-	printk("reserve_pages = %ld find_max_low_pfn() ~ %ld\n",
-			reserve_pages, max_low_pfn + reserve_pages);
+	kva_start_pfn = find_max_low_pfn() - kva_pages;
+
+#ifdef CONFIG_BLK_DEV_INITRD
+	/* Numa kva area is below the initrd */
+	if (LOADER_TYPE && INITRD_START)
+		kva_start_pfn = PFN_DOWN(INITRD_START)  - kva_pages;
+#endif
+	kva_start_pfn -= kva_start_pfn & (PTRS_PER_PTE-1);
+
+	system_max_low_pfn = max_low_pfn = find_max_low_pfn();
+	printk("kva_start_pfn ~ %ld find_max_low_pfn() ~ %ld\n",
+		kva_start_pfn, max_low_pfn);
 	printk("max_pfn = %ld\n", max_pfn);
 #ifdef CONFIG_HIGHMEM
 	highstart_pfn = highend_pfn = max_pfn;
@@ -323,7 +332,7 @@ unsigned long __init setup_memory(void)
 			(ulong) pfn_to_kaddr(max_low_pfn));
 	for_each_online_node(nid) {
 		node_remap_start_vaddr[nid] = pfn_to_kaddr(
-				highstart_pfn + node_remap_offset[nid]);
+				kva_start_pfn + node_remap_offset[nid]);
 		/* Init the node remap allocator */
 		node_remap_end_vaddr[nid] = node_remap_start_vaddr[nid] +
 			(node_remap_size[nid] * PAGE_SIZE);
@@ -338,7 +347,6 @@ unsigned long __init setup_memory(void)
 	}
 	printk("High memory starts at vaddr %08lx\n",
 			(ulong) pfn_to_kaddr(highstart_pfn));
-	vmalloc_earlyreserve = reserve_pages * PAGE_SIZE;
 	for_each_online_node(nid)
 		find_max_pfn_node(nid);
 
@@ -348,6 +356,11 @@ unsigned long __init setup_memory(void)
 	return max_low_pfn;
 }
 
+void __init numa_kva_reserve(void)
+{
+	reserve_bootmem(PFN_PHYS(kva_start_pfn),PFN_PHYS(kva_pages));
+}
+
 void __init zone_sizes_init(void)
 {
 	int nid;
diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h
index 22cb07cc8f32..61b073322006 100644
--- a/include/asm-i386/mmzone.h
+++ b/include/asm-i386/mmzone.h
@@ -38,10 +38,16 @@ static inline void get_memcfg_numa(void)
 }
 
 extern int early_pfn_to_nid(unsigned long pfn);
+extern void numa_kva_reserve(void);
 
 #else /* !CONFIG_NUMA */
+
 #define get_memcfg_numa get_memcfg_numa_flat
 #define get_zholes_size(n) (0)
+
+static inline void numa_kva_reserve(void)
+{
+}
 #endif /* CONFIG_NUMA */
 
 #ifdef CONFIG_DISCONTIGMEM
-- 
cgit v1.2.3


From 2d1a07d487d8b36658404839cdf03a974968cefd Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:03 -0700
Subject: [PATCH] bootmem: remove useless __init in header file

__init in headers is pretty useless because the compiler doesn't check it, and
they get out of sync relatively frequently.  So if you see an __init in a
header file, it's quite unreliable and you need to check the definition
anyway.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 46 +++++++++++++++++++++++-----------------------
 1 file changed, 23 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index e319c649e4fd..c7124d4ce7cf 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -41,23 +41,23 @@ typedef struct bootmem_data {
 	struct list_head list;
 } bootmem_data_t;
 
-extern unsigned long __init bootmem_bootmap_pages (unsigned long);
-extern unsigned long __init init_bootmem (unsigned long addr, unsigned long memend);
-extern void __init free_bootmem (unsigned long addr, unsigned long size);
-extern void * __init __alloc_bootmem (unsigned long size, unsigned long align, unsigned long goal);
-extern void * __init __alloc_bootmem_nopanic (unsigned long size, unsigned long align, unsigned long goal);
-extern void * __init __alloc_bootmem_low(unsigned long size,
+extern unsigned long bootmem_bootmap_pages (unsigned long);
+extern unsigned long init_bootmem (unsigned long addr, unsigned long memend);
+extern void free_bootmem (unsigned long addr, unsigned long size);
+extern void * __alloc_bootmem (unsigned long size, unsigned long align, unsigned long goal);
+extern void * __alloc_bootmem_nopanic (unsigned long size, unsigned long align, unsigned long goal);
+extern void * __alloc_bootmem_low(unsigned long size,
 					 unsigned long align,
 					 unsigned long goal);
-extern void * __init __alloc_bootmem_low_node(pg_data_t *pgdat,
+extern void * __alloc_bootmem_low_node(pg_data_t *pgdat,
 					      unsigned long size,
 					      unsigned long align,
 					      unsigned long goal);
-extern void * __init __alloc_bootmem_core(struct bootmem_data *bdata,
+extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
 		unsigned long size, unsigned long align, unsigned long goal,
 		unsigned long limit);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-extern void __init reserve_bootmem (unsigned long addr, unsigned long size);
+extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
 	__alloc_bootmem((x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
@@ -67,12 +67,12 @@ extern void __init reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem_low_pages(x) \
 	__alloc_bootmem_low((x), PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
-extern unsigned long __init free_all_bootmem (void);
-extern void * __init __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
-extern unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn);
-extern void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size);
-extern void __init free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned long size);
-extern unsigned long __init free_all_bootmem_node (pg_data_t *pgdat);
+extern unsigned long free_all_bootmem (void);
+extern void * __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
+extern unsigned long init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn);
+extern void reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size);
+extern void free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned long size);
+extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
 	__alloc_bootmem_node((pgdat), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
@@ -94,14 +94,14 @@ static inline void *alloc_remap(int nid, unsigned long size)
 extern unsigned long __meminitdata nr_kernel_pages;
 extern unsigned long nr_all_pages;
 
-extern void *__init alloc_large_system_hash(const char *tablename,
-					    unsigned long bucketsize,
-					    unsigned long numentries,
-					    int scale,
-					    int flags,
-					    unsigned int *_hash_shift,
-					    unsigned int *_hash_mask,
-					    unsigned long limit);
+extern void * alloc_large_system_hash(const char *tablename,
+				      unsigned long bucketsize,
+				      unsigned long numentries,
+				      int scale,
+				      int flags,
+				      unsigned int *_hash_shift,
+				      unsigned int *_hash_mask,
+				      unsigned long limit);
 
 #define HASH_HIGHMEM	0x00000001	/* Consider highmem? */
 #define HASH_EARLY	0x00000002	/* Allocating during early boot? */
-- 
cgit v1.2.3


From 71fb2e8f8753b66b1f4295aa264a2eb4e69381e8 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:05 -0700
Subject: [PATCH] bootmem: remove useless parentheses in bootmem header file

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index c7124d4ce7cf..fa2523f47628 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -59,13 +59,13 @@ extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
-	__alloc_bootmem((x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
-	__alloc_bootmem_low((x), SMP_CACHE_BYTES, 0)
+	__alloc_bootmem_low(x, SMP_CACHE_BYTES, 0)
 #define alloc_bootmem_pages(x) \
-	__alloc_bootmem((x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem(x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages(x) \
-	__alloc_bootmem_low((x), PAGE_SIZE, 0)
+	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 extern unsigned long free_all_bootmem (void);
 extern void * __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
@@ -75,11 +75,11 @@ extern void free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned lo
 extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
-	__alloc_bootmem_node((pgdat), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_pages_node(pgdat, x) \
-	__alloc_bootmem_node((pgdat), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem_node(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages_node(pgdat, x) \
-	__alloc_bootmem_low_node((pgdat), (x), PAGE_SIZE, 0)
+	__alloc_bootmem_low_node(pgdat, x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
 #ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP
-- 
cgit v1.2.3


From bb0923a66820718f636736b22ce47372f79e3400 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:05 -0700
Subject: [PATCH] bootmem: limit to 80 columns width

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 42 +++++++++++++++++++++++++++++-------------
 mm/bootmem.c            | 28 ++++++++++++++++++----------
 2 files changed, 47 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index fa2523f47628..1e8dddfb3083 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -44,18 +44,24 @@ typedef struct bootmem_data {
 extern unsigned long bootmem_bootmap_pages (unsigned long);
 extern unsigned long init_bootmem (unsigned long addr, unsigned long memend);
 extern void free_bootmem (unsigned long addr, unsigned long size);
-extern void * __alloc_bootmem (unsigned long size, unsigned long align, unsigned long goal);
-extern void * __alloc_bootmem_nopanic (unsigned long size, unsigned long align, unsigned long goal);
+extern void * __alloc_bootmem (unsigned long size,
+			       unsigned long align,
+			       unsigned long goal);
+extern void * __alloc_bootmem_nopanic (unsigned long size,
+				       unsigned long align,
+				       unsigned long goal);
 extern void * __alloc_bootmem_low(unsigned long size,
-					 unsigned long align,
-					 unsigned long goal);
+				  unsigned long align,
+				  unsigned long goal);
 extern void * __alloc_bootmem_low_node(pg_data_t *pgdat,
-					      unsigned long size,
-					      unsigned long align,
-					      unsigned long goal);
+				       unsigned long size,
+				       unsigned long align,
+				       unsigned long goal);
 extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
-		unsigned long size, unsigned long align, unsigned long goal,
-		unsigned long limit);
+				   unsigned long size,
+				   unsigned long align,
+				   unsigned long goal,
+				   unsigned long limit);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
@@ -68,10 +74,20 @@ extern void reserve_bootmem (unsigned long addr, unsigned long size);
 	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 extern unsigned long free_all_bootmem (void);
-extern void * __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
-extern unsigned long init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn);
-extern void reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size);
-extern void free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned long size);
+extern void * __alloc_bootmem_node (pg_data_t *pgdat,
+				    unsigned long size,
+				    unsigned long align,
+				    unsigned long goal);
+extern unsigned long init_bootmem_node (pg_data_t *pgdat,
+					unsigned long freepfn,
+					unsigned long startpfn,
+					unsigned long endpfn);
+extern void reserve_bootmem_node (pg_data_t *pgdat,
+				  unsigned long physaddr,
+				  unsigned long size);
+extern void free_bootmem_node (pg_data_t *pgdat,
+			       unsigned long addr,
+			       unsigned long size);
 extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 70f1528a3c8a..9083f5029673 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -102,7 +102,8 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
  * might be used for boot-time allocations - or it might get added
  * to the free page pool later on.
  */
-static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long addr, unsigned long size)
+static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
+					unsigned long size)
 {
 	unsigned long i;
 	/*
@@ -127,7 +128,8 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long add
 		}
 }
 
-static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, unsigned long size)
+static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
+				     unsigned long size)
 {
 	unsigned long i;
 	unsigned long start;
@@ -355,17 +357,20 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	return total;
 }
 
-unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn)
+unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn,
+				unsigned long startpfn, unsigned long endpfn)
 {
 	return(init_bootmem_core(pgdat, freepfn, startpfn, endpfn));
 }
 
-void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size)
+void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
+				  unsigned long size)
 {
 	reserve_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
-void __init free_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size)
+void __init free_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
+			       unsigned long size)
 {
 	free_bootmem_core(pgdat->bdata, physaddr, size);
 }
@@ -399,7 +404,8 @@ unsigned long __init free_all_bootmem (void)
 	return(free_all_bootmem_core(NODE_DATA(0)));
 }
 
-void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, unsigned long goal)
+void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
+				      unsigned long goal)
 {
 	bootmem_data_t *bdata;
 	void *ptr;
@@ -410,7 +416,8 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, u
 	return NULL;
 }
 
-void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned long goal)
+void * __init __alloc_bootmem(unsigned long size, unsigned long align,
+			      unsigned long goal)
 {
 	void *mem = __alloc_bootmem_nopanic(size,align,goal);
 	if (mem)
@@ -424,8 +431,8 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned
 }
 
 
-void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigned long align,
-				   unsigned long goal)
+void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
+				   unsigned long align, unsigned long goal)
 {
 	void *ptr;
 
@@ -438,7 +445,8 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigne
 
 #define LOW32LIMIT 0xffffffff
 
-void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, unsigned long goal)
+void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
+				  unsigned long goal)
 {
 	bootmem_data_t *bdata;
 	void *ptr;
-- 
cgit v1.2.3


From e786e86a542ccc1133f333402526ad00b9c088ae Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:06 -0700
Subject: [PATCH] bootmem: remove useless headers inclusions

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h |  5 +----
 mm/bootmem.c            | 10 +++-------
 2 files changed, 4 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 1e8dddfb3083..b2067e4a4486 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -4,11 +4,8 @@
 #ifndef _LINUX_BOOTMEM_H
 #define _LINUX_BOOTMEM_H
 
-#include <asm/pgtable.h>
-#include <asm/dma.h>
-#include <linux/cache.h>
-#include <linux/init.h>
 #include <linux/mmzone.h>
+#include <asm/dma.h>
 
 /*
  *  simple boot-time physical memory area allocator.
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 9083f5029673..e136c086fd98 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -8,17 +8,13 @@
  *  free memory collector. It's used to deal with reserved
  *  system memory and memory holes as well.
  */
-
-#include <linux/mm.h>
-#include <linux/kernel_stat.h>
-#include <linux/swap.h>
-#include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/bootmem.h>
-#include <linux/mmzone.h>
 #include <linux/module.h>
-#include <asm/dma.h>
+
+#include <asm/bug.h>
 #include <asm/io.h>
+
 #include "internal.h"
 
 /*
-- 
cgit v1.2.3


From f71bf0cac730ccb5ebcdf21747db75ae0445ccde Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:08 -0700
Subject: [PATCH] bootmem: miscellaneous coding style fixes

It fixes various coding style issues, specially when spaces are useless.  For
example '*' go next to the function name.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 95 +++++++++++++++++++++++++------------------------
 mm/bootmem.c            | 81 ++++++++++++++++++++++-------------------
 2 files changed, 93 insertions(+), 83 deletions(-)

(limited to 'include')

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index b2067e4a4486..31e9abb6d977 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -38,29 +38,30 @@ typedef struct bootmem_data {
 	struct list_head list;
 } bootmem_data_t;
 
-extern unsigned long bootmem_bootmap_pages (unsigned long);
-extern unsigned long init_bootmem (unsigned long addr, unsigned long memend);
-extern void free_bootmem (unsigned long addr, unsigned long size);
-extern void * __alloc_bootmem (unsigned long size,
-			       unsigned long align,
-			       unsigned long goal);
-extern void * __alloc_bootmem_nopanic (unsigned long size,
-				       unsigned long align,
-				       unsigned long goal);
-extern void * __alloc_bootmem_low(unsigned long size,
+extern unsigned long bootmem_bootmap_pages(unsigned long);
+extern unsigned long init_bootmem(unsigned long addr, unsigned long memend);
+extern void free_bootmem(unsigned long addr, unsigned long size);
+extern void *__alloc_bootmem(unsigned long size,
+			     unsigned long align,
+			     unsigned long goal);
+extern void *__alloc_bootmem_nopanic(unsigned long size,
+				     unsigned long align,
+				     unsigned long goal);
+extern void *__alloc_bootmem_low(unsigned long size,
+				 unsigned long align,
+				 unsigned long goal);
+extern void *__alloc_bootmem_low_node(pg_data_t *pgdat,
+				      unsigned long size,
+				      unsigned long align,
+				      unsigned long goal);
+extern void *__alloc_bootmem_core(struct bootmem_data *bdata,
+				  unsigned long size,
 				  unsigned long align,
-				  unsigned long goal);
-extern void * __alloc_bootmem_low_node(pg_data_t *pgdat,
-				       unsigned long size,
-				       unsigned long align,
-				       unsigned long goal);
-extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
-				   unsigned long size,
-				   unsigned long align,
-				   unsigned long goal,
-				   unsigned long limit);
+				  unsigned long goal,
+				  unsigned long limit);
+
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-extern void reserve_bootmem (unsigned long addr, unsigned long size);
+extern void reserve_bootmem(unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
 	__alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
@@ -70,22 +71,24 @@ extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem_low_pages(x) \
 	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
-extern unsigned long free_all_bootmem (void);
-extern void * __alloc_bootmem_node (pg_data_t *pgdat,
-				    unsigned long size,
-				    unsigned long align,
-				    unsigned long goal);
-extern unsigned long init_bootmem_node (pg_data_t *pgdat,
-					unsigned long freepfn,
-					unsigned long startpfn,
-					unsigned long endpfn);
-extern void reserve_bootmem_node (pg_data_t *pgdat,
-				  unsigned long physaddr,
-				  unsigned long size);
-extern void free_bootmem_node (pg_data_t *pgdat,
-			       unsigned long addr,
-			       unsigned long size);
-extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
+
+extern unsigned long free_all_bootmem(void);
+extern unsigned long free_all_bootmem_node(pg_data_t *pgdat);
+extern void *__alloc_bootmem_node(pg_data_t *pgdat,
+				  unsigned long size,
+				  unsigned long align,
+				  unsigned long goal);
+extern unsigned long init_bootmem_node(pg_data_t *pgdat,
+				       unsigned long freepfn,
+				       unsigned long startpfn,
+				       unsigned long endpfn);
+extern void reserve_bootmem_node(pg_data_t *pgdat,
+				 unsigned long physaddr,
+				 unsigned long size);
+extern void free_bootmem_node(pg_data_t *pgdat,
+			      unsigned long addr,
+			      unsigned long size);
+
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
 	__alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
@@ -102,19 +105,19 @@ static inline void *alloc_remap(int nid, unsigned long size)
 {
 	return NULL;
 }
-#endif
+#endif /* CONFIG_HAVE_ARCH_ALLOC_REMAP */
 
 extern unsigned long __meminitdata nr_kernel_pages;
 extern unsigned long nr_all_pages;
 
-extern void * alloc_large_system_hash(const char *tablename,
-				      unsigned long bucketsize,
-				      unsigned long numentries,
-				      int scale,
-				      int flags,
-				      unsigned int *_hash_shift,
-				      unsigned int *_hash_mask,
-				      unsigned long limit);
+extern void *alloc_large_system_hash(const char *tablename,
+				     unsigned long bucketsize,
+				     unsigned long numentries,
+				     int scale,
+				     int flags,
+				     unsigned int *_hash_shift,
+				     unsigned int *_hash_mask,
+				     unsigned long limit);
 
 #define HASH_HIGHMEM	0x00000001	/* Consider highmem? */
 #define HASH_EARLY	0x00000002	/* Allocating during early boot? */
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 23c3317c09ed..f0f85fa713ca 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -38,7 +38,7 @@ unsigned long saved_max_pfn;
 #endif
 
 /* return the number of _pages_ that will be allocated for the boot bitmap */
-unsigned long __init bootmem_bootmap_pages (unsigned long pages)
+unsigned long __init bootmem_bootmap_pages(unsigned long pages)
 {
 	unsigned long mapsize;
 
@@ -48,12 +48,14 @@ unsigned long __init bootmem_bootmap_pages (unsigned long pages)
 
 	return mapsize;
 }
+
 /*
  * link bdata in order
  */
 static void __init link_bootmem(bootmem_data_t *bdata)
 {
 	bootmem_data_t *ent;
+
 	if (list_empty(&bdata_list)) {
 		list_add(&bdata->list, &bdata_list);
 		return;
@@ -66,7 +68,6 @@ static void __init link_bootmem(bootmem_data_t *bdata)
 		}
 	}
 	list_add_tail(&bdata->list, &bdata_list);
-	return;
 }
 
 /*
@@ -85,7 +86,7 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata)
 /*
  * Called once to set up the allocator itself.
  */
-static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
+static unsigned long __init init_bootmem_core(pg_data_t *pgdat,
 	unsigned long mapstart, unsigned long start, unsigned long end)
 {
 	bootmem_data_t *bdata = pgdat->bdata;
@@ -185,7 +186,7 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	unsigned long i, start = 0, incr, eidx, end_pfn;
 	void *ret;
 
-	if(!size) {
+	if (!size) {
 		printk("__alloc_bootmem_core(): zero-sized request\n");
 		BUG();
 	}
@@ -234,7 +235,7 @@ restart_scan:
 		for (j = i + 1; j < i + areasize; ++j) {
 			if (j >= eidx)
 				goto fail_block;
-			if (test_bit (j, bdata->node_bootmem_map))
+			if (test_bit(j, bdata->node_bootmem_map))
 				goto fail_block;
 		}
 		start = i;
@@ -262,19 +263,21 @@ found:
 	    bdata->last_offset && bdata->last_pos+1 == start) {
 		offset = ALIGN(bdata->last_offset, align);
 		BUG_ON(offset > PAGE_SIZE);
-		remaining_size = PAGE_SIZE-offset;
+		remaining_size = PAGE_SIZE - offset;
 		if (size < remaining_size) {
 			areasize = 0;
 			/* last_pos unchanged */
-			bdata->last_offset = offset+size;
-			ret = phys_to_virt(bdata->last_pos*PAGE_SIZE + offset +
-						bdata->node_boot_start);
+			bdata->last_offset = offset + size;
+			ret = phys_to_virt(bdata->last_pos * PAGE_SIZE +
+					   offset +
+					   bdata->node_boot_start);
 		} else {
 			remaining_size = size - remaining_size;
-			areasize = (remaining_size+PAGE_SIZE-1)/PAGE_SIZE;
-			ret = phys_to_virt(bdata->last_pos*PAGE_SIZE + offset +
-						bdata->node_boot_start);
-			bdata->last_pos = start+areasize-1;
+			areasize = (remaining_size + PAGE_SIZE-1) / PAGE_SIZE;
+			ret = phys_to_virt(bdata->last_pos * PAGE_SIZE +
+					   offset +
+					   bdata->node_boot_start);
+			bdata->last_pos = start + areasize - 1;
 			bdata->last_offset = remaining_size;
 		}
 		bdata->last_offset &= ~PAGE_MASK;
@@ -287,7 +290,7 @@ found:
 	/*
 	 * Reserve the area now:
 	 */
-	for (i = start; i < start+areasize; i++)
+	for (i = start; i < start + areasize; i++)
 		if (unlikely(test_and_set_bit(i, bdata->node_bootmem_map)))
 			BUG();
 	memset(ret, 0, size);
@@ -338,7 +341,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 				}
 			}
 		} else {
-			i+=BITS_PER_LONG;
+			i += BITS_PER_LONG;
 		}
 		pfn += BITS_PER_LONG;
 	}
@@ -361,51 +364,51 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	return total;
 }
 
-unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn,
+unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,
 				unsigned long startpfn, unsigned long endpfn)
 {
-	return(init_bootmem_core(pgdat, freepfn, startpfn, endpfn));
+	return init_bootmem_core(pgdat, freepfn, startpfn, endpfn);
 }
 
-void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
-				  unsigned long size)
+void __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
+				 unsigned long size)
 {
 	reserve_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
-void __init free_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
-			       unsigned long size)
+void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
+			      unsigned long size)
 {
 	free_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
-unsigned long __init free_all_bootmem_node (pg_data_t *pgdat)
+unsigned long __init free_all_bootmem_node(pg_data_t *pgdat)
 {
-	return(free_all_bootmem_core(pgdat));
+	return free_all_bootmem_core(pgdat);
 }
 
-unsigned long __init init_bootmem (unsigned long start, unsigned long pages)
+unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
 {
 	max_low_pfn = pages;
 	min_low_pfn = start;
-	return(init_bootmem_core(NODE_DATA(0), start, 0, pages));
+	return init_bootmem_core(NODE_DATA(0), start, 0, pages);
 }
 
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-void __init reserve_bootmem (unsigned long addr, unsigned long size)
+void __init reserve_bootmem(unsigned long addr, unsigned long size)
 {
 	reserve_bootmem_core(NODE_DATA(0)->bdata, addr, size);
 }
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
-void __init free_bootmem (unsigned long addr, unsigned long size)
+void __init free_bootmem(unsigned long addr, unsigned long size)
 {
 	free_bootmem_core(NODE_DATA(0)->bdata, addr, size);
 }
 
-unsigned long __init free_all_bootmem (void)
+unsigned long __init free_all_bootmem(void)
 {
-	return(free_all_bootmem_core(NODE_DATA(0)));
+	return free_all_bootmem_core(NODE_DATA(0));
 }
 
 void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
@@ -414,9 +417,11 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 	bootmem_data_t *bdata;
 	void *ptr;
 
-	list_for_each_entry(bdata, &bdata_list, list)
-		if ((ptr = __alloc_bootmem_core(bdata, size, align, goal, 0)))
-			return(ptr);
+	list_for_each_entry(bdata, &bdata_list, list) {
+		ptr = __alloc_bootmem_core(bdata, size, align, goal, 0);
+		if (ptr)
+			return ptr;
+	}
 	return NULL;
 }
 
@@ -424,6 +429,7 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 			      unsigned long goal)
 {
 	void *mem = __alloc_bootmem_nopanic(size,align,goal);
+
 	if (mem)
 		return mem;
 	/*
@@ -442,7 +448,7 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 
 	ptr = __alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
 	if (ptr)
-		return (ptr);
+		return ptr;
 
 	return __alloc_bootmem(size, align, goal);
 }
@@ -455,10 +461,11 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 	bootmem_data_t *bdata;
 	void *ptr;
 
-	list_for_each_entry(bdata, &bdata_list, list)
-		if ((ptr = __alloc_bootmem_core(bdata, size,
-						 align, goal, LOW32LIMIT)))
-			return(ptr);
+	list_for_each_entry(bdata, &bdata_list, list) {
+		ptr = __alloc_bootmem_core(bdata, size, align, goal, LOW32LIMIT);
+		if (ptr)
+			return ptr;
+	}
 
 	/*
 	 * Whoops, we cannot satisfy the allocation request.
-- 
cgit v1.2.3


From c1f60a5a419cc60aff27daffb150f5a3a3a79ef4 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:11 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: move HIGHMEM counters into highmem.c/.h

Move totalhigh_pages and nr_free_highpages() into highmem.c/.h

Move the totalhigh_pages definition into highmem.c/.h.  Move the
nr_free_highpages function into highmem.c

[yoichi_yuasa@tripeaks.co.jp: build fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/sgi-ip27/ip27-memory.c |  1 +
 arch/um/kernel/mem.c             |  2 ++
 include/linux/highmem.h          |  3 +++
 include/linux/swap.h             |  1 -
 mm/highmem.c                     | 13 +++++++++++++
 mm/page_alloc.c                  | 15 ---------------
 mm/shmem.c                       |  1 +
 7 files changed, 20 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index 59bfc0fc3f45..16e5682b01f1 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -19,6 +19,7 @@
 #include <linux/swap.h>
 #include <linux/bootmem.h>
 #include <linux/pfn.h>
+#include <linux/highmem.h>
 #include <asm/page.h>
 #include <asm/sections.h>
 
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index 61280167c560..b39e624c3291 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -79,8 +79,10 @@ void mem_init(void)
 
 	/* this will put all low memory onto the freelists */
 	totalram_pages = free_all_bootmem();
+#ifdef CONFIG_HIGHMEM
 	totalhigh_pages = highmem >> PAGE_SHIFT;
 	totalram_pages += totalhigh_pages;
+#endif
 	num_physpages = totalram_pages;
 	max_pfn = totalram_pages;
 	printk(KERN_INFO "Memory: %luk available\n", 
diff --git a/include/linux/highmem.h b/include/linux/highmem.h
index 42620e723abb..fd7d12daa94f 100644
--- a/include/linux/highmem.h
+++ b/include/linux/highmem.h
@@ -24,11 +24,14 @@ static inline void flush_kernel_dcache_page(struct page *page)
 
 /* declarations for linux/mm/highmem.c */
 unsigned int nr_free_highpages(void);
+extern unsigned long totalhigh_pages;
 
 #else /* CONFIG_HIGHMEM */
 
 static inline unsigned int nr_free_highpages(void) { return 0; }
 
+#define totalhigh_pages 0
+
 #ifndef ARCH_HAS_KMAP
 static inline void *kmap(struct page *page)
 {
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 5e59184c9096..34a6bc3e6cf3 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -162,7 +162,6 @@ extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *
 
 /* linux/mm/page_alloc.c */
 extern unsigned long totalram_pages;
-extern unsigned long totalhigh_pages;
 extern unsigned long totalreserve_pages;
 extern long nr_swap_pages;
 extern unsigned int nr_free_pages(void);
diff --git a/mm/highmem.c b/mm/highmem.c
index 9b2a5403c447..ee5519b176ee 100644
--- a/mm/highmem.c
+++ b/mm/highmem.c
@@ -46,6 +46,19 @@ static void *mempool_alloc_pages_isa(gfp_t gfp_mask, void *data)
  */
 #ifdef CONFIG_HIGHMEM
 
+unsigned long totalhigh_pages __read_mostly;
+
+unsigned int nr_free_highpages (void)
+{
+	pg_data_t *pgdat;
+	unsigned int pages = 0;
+
+	for_each_online_pgdat(pgdat)
+		pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
+
+	return pages;
+}
+
 static int pkmap_count[LAST_PKMAP];
 static unsigned int last_pkmap_nr;
 static  __cacheline_aligned_in_smp DEFINE_SPINLOCK(kmap_lock);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e26491cb5a27..5cde54695cfb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -51,7 +51,6 @@ EXPORT_SYMBOL(node_online_map);
 nodemask_t node_possible_map __read_mostly = NODE_MASK_ALL;
 EXPORT_SYMBOL(node_possible_map);
 unsigned long totalram_pages __read_mostly;
-unsigned long totalhigh_pages __read_mostly;
 unsigned long totalreserve_pages __read_mostly;
 long nr_swap_pages;
 int percpu_pagelist_fraction;
@@ -1185,20 +1184,6 @@ unsigned int nr_free_pagecache_pages(void)
 {
 	return nr_free_zone_pages(gfp_zone(GFP_HIGHUSER));
 }
-
-#ifdef CONFIG_HIGHMEM
-unsigned int nr_free_highpages (void)
-{
-	pg_data_t *pgdat;
-	unsigned int pages = 0;
-
-	for_each_online_pgdat(pgdat)
-		pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
-
-	return pages;
-}
-#endif
-
 #ifdef CONFIG_NUMA
 static void show_node(struct zone *zone)
 {
diff --git a/mm/shmem.c b/mm/shmem.c
index db21c51531ca..8631be45b40d 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -45,6 +45,7 @@
 #include <linux/namei.h>
 #include <linux/ctype.h>
 #include <linux/migrate.h>
+#include <linux/highmem.h>
 
 #include <asm/uaccess.h>
 #include <asm/div64.h>
-- 
cgit v1.2.3


From 2f1b6248682f8b39ca3c7e549dfc216d26c4109b Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:13 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: use enum to define zones, reformat and
 comment

Use enum for zones and reformat zones dependent information

Add comments explaning the use of zones and add a zones_t type for zone
numbers.

Line up information that will be #ifdefd by the following patches.

[akpm@osdl.org: comment cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     |  7 +++---
 include/linux/mmzone.h | 66 +++++++++++++++++++++++++++++++++++---------------
 mm/page_alloc.c        | 24 +++++++++++++-----
 3 files changed, 69 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 45678b036955..2db4229a0066 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -470,7 +470,7 @@ void split_page(struct page *page, unsigned int order);
 #define SECTIONS_MASK		((1UL << SECTIONS_WIDTH) - 1)
 #define ZONETABLE_MASK		((1UL << ZONETABLE_SHIFT) - 1)
 
-static inline unsigned long page_zonenum(struct page *page)
+static inline enum zone_type page_zonenum(struct page *page)
 {
 	return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK;
 }
@@ -499,11 +499,12 @@ static inline unsigned long page_to_section(struct page *page)
 	return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK;
 }
 
-static inline void set_page_zone(struct page *page, unsigned long zone)
+static inline void set_page_zone(struct page *page, enum zone_type zone)
 {
 	page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT);
 	page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT;
 }
+
 static inline void set_page_node(struct page *page, unsigned long node)
 {
 	page->flags &= ~(NODES_MASK << NODES_PGSHIFT);
@@ -515,7 +516,7 @@ static inline void set_page_section(struct page *page, unsigned long section)
 	page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT;
 }
 
-static inline void set_page_links(struct page *page, unsigned long zone,
+static inline void set_page_links(struct page *page, enum zone_type zone,
 	unsigned long node, unsigned long pfn)
 {
 	set_page_zone(page, zone);
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index f45163c528e8..03a5a6eb0ffa 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -88,14 +88,53 @@ struct per_cpu_pageset {
 #define zone_pcp(__z, __cpu) (&(__z)->pageset[(__cpu)])
 #endif
 
-#define ZONE_DMA		0
-#define ZONE_DMA32		1
-#define ZONE_NORMAL		2
-#define ZONE_HIGHMEM		3
+enum zone_type {
+	/*
+	 * ZONE_DMA is used when there are devices that are not able
+	 * to do DMA to all of addressable memory (ZONE_NORMAL). Then we
+	 * carve out the portion of memory that is needed for these devices.
+	 * The range is arch specific.
+	 *
+	 * Some examples
+	 *
+	 * Architecture		Limit
+	 * ---------------------------
+	 * parisc, ia64, sparc	<4G
+	 * s390			<2G
+	 * arm26		<48M
+	 * arm			Various
+	 * alpha		Unlimited or 0-16MB.
+	 *
+	 * i386, x86_64 and multiple other arches
+	 * 			<16M.
+	 */
+	ZONE_DMA,
+	/*
+	 * x86_64 needs two ZONE_DMAs because it supports devices that are
+	 * only able to do DMA to the lower 16M but also 32 bit devices that
+	 * can only do DMA areas below 4G.
+	 */
+	ZONE_DMA32,
+	/*
+	 * Normal addressable memory is in ZONE_NORMAL. DMA operations can be
+	 * performed on pages in ZONE_NORMAL if the DMA devices support
+	 * transfers to all addressable memory.
+	 */
+	ZONE_NORMAL,
+	/*
+	 * A memory area that is only addressable by the kernel through
+	 * mapping portions into its own address space. This is for example
+	 * used by i386 to allow the kernel to address the memory beyond
+	 * 900MB. The kernel will set up special mappings (page
+	 * table entries on i386) for each page that the kernel needs to
+	 * access.
+	 */
+	ZONE_HIGHMEM,
 
-#define MAX_NR_ZONES		4	/* Sync this with ZONES_SHIFT */
-#define ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
+	MAX_NR_ZONES
+};
 
+#define	ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
 
 /*
  * When a memory allocation must conform to specific limitations (such
@@ -126,16 +165,6 @@ struct per_cpu_pageset {
 /* #define GFP_ZONETYPES       (GFP_ZONEMASK + 1) */           /* Non-loner */
 #define GFP_ZONETYPES  ((GFP_ZONEMASK + 1) / 2 + 1)            /* Loner */
 
-/*
- * On machines where it is needed (eg PCs) we divide physical memory
- * into multiple physical zones. On a 32bit PC we have 4 zones:
- *
- * ZONE_DMA	  < 16 MB	ISA DMA capable memory
- * ZONE_DMA32	     0 MB 	Empty
- * ZONE_NORMAL	16-896 MB	direct mapped by the kernel
- * ZONE_HIGHMEM	 > 896 MB	only page cache and user processes
- */
-
 struct zone {
 	/* Fields commonly accessed by the page allocator */
 	unsigned long		free_pages;
@@ -266,7 +295,6 @@ struct zone {
 	char			*name;
 } ____cacheline_internodealigned_in_smp;
 
-
 /*
  * The "priority" of VM scanning is how much of the queues we will scan in one
  * go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the
@@ -373,12 +401,12 @@ static inline int populated_zone(struct zone *zone)
 	return (!!zone->present_pages);
 }
 
-static inline int is_highmem_idx(int idx)
+static inline int is_highmem_idx(enum zone_type idx)
 {
 	return (idx == ZONE_HIGHMEM);
 }
 
-static inline int is_normal_idx(int idx)
+static inline int is_normal_idx(enum zone_type idx)
 {
 	return (idx == ZONE_NORMAL);
 }
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 13c102b95c57..2410a3cb1c53 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -68,7 +68,11 @@ static void __free_pages_ok(struct page *page, unsigned int order);
  * TBD: should special case ZONE_DMA32 machines here - in those we normally
  * don't need any ZONE_NORMAL reservation
  */
-int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = { 256, 256, 32 };
+int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = {
+	 256,
+	 256,
+	 32
+};
 
 EXPORT_SYMBOL(totalram_pages);
 
@@ -79,7 +83,13 @@ EXPORT_SYMBOL(totalram_pages);
 struct zone *zone_table[1 << ZONETABLE_SHIFT] __read_mostly;
 EXPORT_SYMBOL(zone_table);
 
-static char *zone_names[MAX_NR_ZONES] = { "DMA", "DMA32", "Normal", "HighMem" };
+static char *zone_names[MAX_NR_ZONES] = {
+	 "DMA",
+	 "DMA32",
+	 "Normal",
+	 "HighMem"
+};
+
 int min_free_kbytes = 1024;
 
 unsigned long __meminitdata nr_kernel_pages;
@@ -1487,7 +1497,9 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, j, k, node, local_node;
+	int i, node, local_node;
+	enum zone_type k;
+	enum zone_type j;
 
 	local_node = pgdat->node_id;
 	for (i = 0; i < GFP_ZONETYPES; i++) {
@@ -1675,8 +1687,8 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone,
 }
 
 #define ZONETABLE_INDEX(x, zone_nr)	((x << ZONES_SHIFT) | zone_nr)
-void zonetable_add(struct zone *zone, int nid, int zid, unsigned long pfn,
-		unsigned long size)
+void zonetable_add(struct zone *zone, int nid, enum zone_type zid,
+		unsigned long pfn, unsigned long size)
 {
 	unsigned long snum = pfn_to_section_nr(pfn);
 	unsigned long end = pfn_to_section_nr(pfn + size);
@@ -1960,7 +1972,7 @@ __meminit int init_currently_empty_zone(struct zone *zone,
 static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		unsigned long *zones_size, unsigned long *zholes_size)
 {
-	unsigned long j;
+	enum zone_type j;
 	int nid = pgdat->node_id;
 	unsigned long zone_start_pfn = pgdat->node_start_pfn;
 	int ret;
-- 
cgit v1.2.3


From fb0e7942bdcbbd2f90e61cb4cfa4fa892a873f8a Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:13 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: make ZONE_DMA32 optional

Make ZONE_DMA32 optional

- Add #ifdefs around ZONE_DMA32 specific code and definitions.

- Add CONFIG_ZONE_DMA32 config option and use that for x86_64
  that alone needs this zone.

- Remove the use of CONFIG_DMA_IS_DMA32 and CONFIG_DMA_IS_NORMAL
  for ia64 and fix up the way per node ZVCs are calculated.

- Fall back to prior GFP_ZONEMASK of 0x03 if there is no
  DMA32 zone.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig      |  9 ---------
 arch/x86_64/Kconfig    |  4 ++++
 include/linux/gfp.h    |  2 +-
 include/linux/mmzone.h | 16 +++++++++++++---
 include/linux/vmstat.h |  4 +---
 mm/page_alloc.c        |  6 ++++++
 6 files changed, 25 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index db274da7dba1..f521f2f60a78 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -66,15 +66,6 @@ config IA64_UNCACHED_ALLOCATOR
 	bool
 	select GENERIC_ALLOCATOR
 
-config DMA_IS_DMA32
-	bool
-	default y
-
-config DMA_IS_NORMAL
-	bool
-	depends on IA64_SGI_SN2
-	default y
-
 config AUDIT_ARCH
 	bool
 	default y
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 6cd4878625f1..581ce9af0ec8 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -24,6 +24,10 @@ config X86
 	bool
 	default y
 
+config ZONE_DMA32
+	bool
+	default y
+
 config LOCKDEP_SUPPORT
 	bool
 	default y
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index cc9e60844484..14610b56c132 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -13,7 +13,7 @@ struct vm_area_struct;
 /* Zone modifiers in GFP_ZONEMASK (see linux/mmzone.h - low three bits) */
 #define __GFP_DMA	((__force gfp_t)0x01u)
 #define __GFP_HIGHMEM	((__force gfp_t)0x02u)
-#ifdef CONFIG_DMA_IS_DMA32
+#ifndef CONFIG_ZONE_DMA32
 #define __GFP_DMA32	((__force gfp_t)0x01)	/* ZONE_DMA is ZONE_DMA32 */
 #elif BITS_PER_LONG < 64
 #define __GFP_DMA32	((__force gfp_t)0x00)	/* ZONE_NORMAL is ZONE_DMA32 */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 03a5a6eb0ffa..adae3c915938 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -109,12 +109,14 @@ enum zone_type {
 	 * 			<16M.
 	 */
 	ZONE_DMA,
+#ifdef CONFIG_ZONE_DMA32
 	/*
 	 * x86_64 needs two ZONE_DMAs because it supports devices that are
 	 * only able to do DMA to the lower 16M but also 32 bit devices that
 	 * can only do DMA areas below 4G.
 	 */
 	ZONE_DMA32,
+#endif
 	/*
 	 * Normal addressable memory is in ZONE_NORMAL. DMA operations can be
 	 * performed on pages in ZONE_NORMAL if the DMA devices support
@@ -161,9 +163,13 @@ enum zone_type {
  *
  * NOTE! Make sure this matches the zones in <linux/gfp.h>
  */
-#define GFP_ZONEMASK	0x07
-/* #define GFP_ZONETYPES       (GFP_ZONEMASK + 1) */           /* Non-loner */
-#define GFP_ZONETYPES  ((GFP_ZONEMASK + 1) / 2 + 1)            /* Loner */
+#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
+
+#ifdef CONFIG_ZONE_DMA32
+#define GFP_ZONEMASK		0x07
+#else
+#define GFP_ZONEMASK		0x03
+#endif
 
 struct zone {
 	/* Fields commonly accessed by the page allocator */
@@ -429,7 +435,11 @@ static inline int is_normal(struct zone *zone)
 
 static inline int is_dma32(struct zone *zone)
 {
+#ifdef CONFIG_ZONE_DMA32
 	return zone == zone->zone_pgdat->node_zones + ZONE_DMA32;
+#else
+	return 0;
+#endif
 }
 
 static inline int is_dma(struct zone *zone)
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 2d9b1b60798a..9c6e62c56ec2 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -124,12 +124,10 @@ static inline unsigned long node_page_state(int node,
 	struct zone *zones = NODE_DATA(node)->node_zones;
 
 	return
-#ifndef CONFIG_DMA_IS_NORMAL
-#if !defined(CONFIG_DMA_IS_DMA32) && BITS_PER_LONG >= 64
+#ifdef CONFIG_ZONE_DMA32
 		zone_page_state(&zones[ZONE_DMA32], item) +
 #endif
 		zone_page_state(&zones[ZONE_NORMAL], item) +
-#endif
 #ifdef CONFIG_HIGHMEM
 		zone_page_state(&zones[ZONE_HIGHMEM], item) +
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 2410a3cb1c53..5b5cbb5e1816 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -70,7 +70,9 @@ static void __free_pages_ok(struct page *page, unsigned int order);
  */
 int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = {
 	 256,
+#ifdef CONFIG_ZONE_DMA32
 	 256,
+#endif
 	 32
 };
 
@@ -85,7 +87,9 @@ EXPORT_SYMBOL(zone_table);
 
 static char *zone_names[MAX_NR_ZONES] = {
 	 "DMA",
+#ifdef CONFIG_ZONE_DMA32
 	 "DMA32",
+#endif
 	 "Normal",
 	 "HighMem"
 };
@@ -1373,8 +1377,10 @@ static inline int highest_zone(int zone_bits)
 	int res = ZONE_NORMAL;
 	if (zone_bits & (__force int)__GFP_HIGHMEM)
 		res = ZONE_HIGHMEM;
+#ifdef CONFIG_ZONE_DMA32
 	if (zone_bits & (__force int)__GFP_DMA32)
 		res = ZONE_DMA32;
+#endif
 	if (zone_bits & (__force int)__GFP_DMA)
 		res = ZONE_DMA;
 	return res;
-- 
cgit v1.2.3


From e53ef38d05dd59ed281a35590e4a5b64d8ff4c52 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:14 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: make ZONE_HIGHMEM optional

Make ZONE_HIGHMEM optional

- ifdef out code and definitions related to CONFIG_HIGHMEM

- __GFP_HIGHMEM falls back to normal allocations if there is no
  ZONE_HIGHMEM

- GFP_ZONEMASK becomes 0x01 if there is no DMA32 and no HIGHMEM
  zone.

[jdike@addtoit.com: build fix]
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/mem.c   |  2 ++
 include/linux/gfp.h    | 18 ++++++++++--------
 include/linux/mmzone.h | 36 +++++++++++++++++++++++++++++++++---
 mm/page_alloc.c        |  6 ++++++
 4 files changed, 51 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index b39e624c3291..b1cd5c6e468b 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -226,7 +226,9 @@ void paging_init(void)
 	for(i=0;i<sizeof(zones_size)/sizeof(zones_size[0]);i++) 
 		zones_size[i] = 0;
 	zones_size[ZONE_DMA] = (end_iomem >> PAGE_SHIFT) - (uml_physmem >> PAGE_SHIFT);
+#ifdef CONFIG_HIGHMEM
 	zones_size[ZONE_HIGHMEM] = highmem >> PAGE_SHIFT;
+#endif
 	free_area_init(zones_size);
 
 	/*
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 14610b56c132..2a2153ebfe0b 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -9,17 +9,19 @@ struct vm_area_struct;
 
 /*
  * GFP bitmasks..
+ *
+ * Zone modifiers (see linux/mmzone.h - low three bits)
+ *
+ * These may be masked by GFP_ZONEMASK to make allocations with this bit
+ * set fall back to ZONE_NORMAL.
+ *
+ * Do not put any conditional on these. If necessary modify the definitions
+ * without the underscores and use the consistently. The definitions here may
+ * be used in bit comparisons.
  */
-/* Zone modifiers in GFP_ZONEMASK (see linux/mmzone.h - low three bits) */
 #define __GFP_DMA	((__force gfp_t)0x01u)
 #define __GFP_HIGHMEM	((__force gfp_t)0x02u)
-#ifndef CONFIG_ZONE_DMA32
-#define __GFP_DMA32	((__force gfp_t)0x01)	/* ZONE_DMA is ZONE_DMA32 */
-#elif BITS_PER_LONG < 64
-#define __GFP_DMA32	((__force gfp_t)0x00)	/* ZONE_NORMAL is ZONE_DMA32 */
-#else
-#define __GFP_DMA32	((__force gfp_t)0x04)	/* Has own ZONE_DMA32 */
-#endif
+#define __GFP_DMA32	((__force gfp_t)0x04u)
 
 /*
  * Action modifiers - doesn't change the zoning
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index adae3c915938..76d33e688593 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -123,6 +123,7 @@ enum zone_type {
 	 * transfers to all addressable memory.
 	 */
 	ZONE_NORMAL,
+#ifdef CONFIG_HIGHMEM
 	/*
 	 * A memory area that is only addressable by the kernel through
 	 * mapping portions into its own address space. This is for example
@@ -132,11 +133,10 @@ enum zone_type {
 	 * access.
 	 */
 	ZONE_HIGHMEM,
-
+#endif
 	MAX_NR_ZONES
 };
 
-#define	ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
 
 /*
  * When a memory allocation must conform to specific limitations (such
@@ -163,12 +163,34 @@ enum zone_type {
  *
  * NOTE! Make sure this matches the zones in <linux/gfp.h>
  */
-#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
 
 #ifdef CONFIG_ZONE_DMA32
+
+#ifdef CONFIG_HIGHMEM
+#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
 #define GFP_ZONEMASK		0x07
+#define ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
 #else
+#define GFP_ZONETYPES		((0x07 + 1) / 2 + 1)    /* Loner */
+/* Mask __GFP_HIGHMEM */
+#define GFP_ZONEMASK		0x05
+#define ZONES_SHIFT		2
+#endif
+
+#else
+#ifdef CONFIG_HIGHMEM
+
 #define GFP_ZONEMASK		0x03
+#define ZONES_SHIFT		2
+#define GFP_ZONETYPES		3
+
+#else
+
+#define GFP_ZONEMASK		0x01
+#define ZONES_SHIFT		1
+#define GFP_ZONETYPES		2
+
+#endif
 #endif
 
 struct zone {
@@ -409,7 +431,11 @@ static inline int populated_zone(struct zone *zone)
 
 static inline int is_highmem_idx(enum zone_type idx)
 {
+#ifdef CONFIG_HIGHMEM
 	return (idx == ZONE_HIGHMEM);
+#else
+	return 0;
+#endif
 }
 
 static inline int is_normal_idx(enum zone_type idx)
@@ -425,7 +451,11 @@ static inline int is_normal_idx(enum zone_type idx)
  */
 static inline int is_highmem(struct zone *zone)
 {
+#ifdef CONFIG_HIGHMEM
 	return zone == zone->zone_pgdat->node_zones + ZONE_HIGHMEM;
+#else
+	return 0;
+#endif
 }
 
 static inline int is_normal(struct zone *zone)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5b5cbb5e1816..6c7c2dd1b3ed 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -73,7 +73,9 @@ int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = {
 #ifdef CONFIG_ZONE_DMA32
 	 256,
 #endif
+#ifdef CONFIG_HIGHMEM
 	 32
+#endif
 };
 
 EXPORT_SYMBOL(totalram_pages);
@@ -91,7 +93,9 @@ static char *zone_names[MAX_NR_ZONES] = {
 	 "DMA32",
 #endif
 	 "Normal",
+#ifdef CONFIG_HIGHMEM
 	 "HighMem"
+#endif
 };
 
 int min_free_kbytes = 1024;
@@ -1375,8 +1379,10 @@ static int __meminit build_zonelists_node(pg_data_t *pgdat,
 static inline int highest_zone(int zone_bits)
 {
 	int res = ZONE_NORMAL;
+#ifdef CONFIG_HIGHMEM
 	if (zone_bits & (__force int)__GFP_HIGHMEM)
 		res = ZONE_HIGHMEM;
+#endif
 #ifdef CONFIG_ZONE_DMA32
 	if (zone_bits & (__force int)__GFP_DMA32)
 		res = ZONE_DMA32;
-- 
cgit v1.2.3


From 27bf71c2a7e596ed34e9bf2d4a5030321a09a1ad Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:15 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: remove display of counters for
 unconfigured zones

eventcounters: Do not display counters for zones that are not available on an
arch

Do not define or display counters for the DMA32 and the HIGHMEM zone if such
zones were not configured.

[akpm@osdl.org: s390 fix]
[heiko.carstens@de.ibm.com: s390 fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/appldata/appldata_mem.c |  3 +--
 include/linux/vmstat.h            | 14 ++++++++++++-
 mm/vmstat.c                       | 43 ++++++++++++++++++---------------------
 3 files changed, 34 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/s390/appldata/appldata_mem.c b/arch/s390/appldata/appldata_mem.c
index ab3b0765a64e..8aea3698a77b 100644
--- a/arch/s390/appldata/appldata_mem.c
+++ b/arch/s390/appldata/appldata_mem.c
@@ -117,8 +117,7 @@ static void appldata_get_mem_data(void *data)
 	mem_data->pgpgout    = ev[PGPGOUT] >> 1;
 	mem_data->pswpin     = ev[PSWPIN];
 	mem_data->pswpout    = ev[PSWPOUT];
-	mem_data->pgalloc    = ev[PGALLOC_HIGH] + ev[PGALLOC_NORMAL] +
-			       ev[PGALLOC_DMA];
+	mem_data->pgalloc    = ev[PGALLOC_NORMAL] + ev[PGALLOC_DMA];
 	mem_data->pgfault    = ev[PGFAULT];
 	mem_data->pgmajfault = ev[PGMAJFAULT];
 
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 9c6e62c56ec2..176c7f797339 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -18,7 +18,19 @@
  * generated will simply be the increment of a global address.
  */
 
-#define FOR_ALL_ZONES(x) x##_DMA, x##_DMA32, x##_NORMAL, x##_HIGH
+#ifdef CONFIG_ZONE_DMA32
+#define DMA32_ZONE(xx) xx##_DMA32,
+#else
+#define DMA32_ZONE(xx)
+#endif
+
+#ifdef CONFIG_HIGHMEM
+#define HIGHMEM_ZONE(xx) , xx##_HIGH
+#else
+#define HIGHMEM_ZONE(xx)
+#endif
+
+#define FOR_ALL_ZONES(xx) xx##_DMA, DMA32_ZONE(xx) xx##_NORMAL HIGHMEM_ZONE(xx)
 
 enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
 		FOR_ALL_ZONES(PGALLOC),
diff --git a/mm/vmstat.c b/mm/vmstat.c
index c1b5f4106b38..04a9093f649e 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -435,6 +435,21 @@ struct seq_operations fragmentation_op = {
 	.show	= frag_show,
 };
 
+#ifdef CONFIG_ZONE_DMA32
+#define TEXT_FOR_DMA32(xx) xx "_dma32",
+#else
+#define TEXT_FOR_DMA32(xx)
+#endif
+
+#ifdef CONFIG_HIGHMEM
+#define TEXT_FOR_HIGHMEM(xx) xx "_high",
+#else
+#define TEXT_FOR_HIGHMEM(xx)
+#endif
+
+#define TEXTS_FOR_ZONES(xx) xx "_dma", TEXT_FOR_DMA32(xx) xx "_normal", \
+					TEXT_FOR_HIGHMEM(xx)
+
 static char *vmstat_text[] = {
 	/* Zoned VM counters */
 	"nr_anon_pages",
@@ -462,10 +477,7 @@ static char *vmstat_text[] = {
 	"pswpin",
 	"pswpout",
 
-	"pgalloc_dma",
-	"pgalloc_dma32",
-	"pgalloc_normal",
-	"pgalloc_high",
+	TEXTS_FOR_ZONES("pgalloc")
 
 	"pgfree",
 	"pgactivate",
@@ -474,25 +486,10 @@ static char *vmstat_text[] = {
 	"pgfault",
 	"pgmajfault",
 
-	"pgrefill_dma",
-	"pgrefill_dma32",
-	"pgrefill_normal",
-	"pgrefill_high",
-
-	"pgsteal_dma",
-	"pgsteal_dma32",
-	"pgsteal_normal",
-	"pgsteal_high",
-
-	"pgscan_kswapd_dma",
-	"pgscan_kswapd_dma32",
-	"pgscan_kswapd_normal",
-	"pgscan_kswapd_high",
-
-	"pgscan_direct_dma",
-	"pgscan_direct_dma32",
-	"pgscan_direct_normal",
-	"pgscan_direct_high",
+	TEXTS_FOR_ZONES("pgrefill")
+	TEXTS_FOR_ZONES("pgsteal")
+	TEXTS_FOR_ZONES("pgscan_kswapd")
+	TEXTS_FOR_ZONES("pgscan_direct")
 
 	"pginodesteal",
 	"slabs_scanned",
-- 
cgit v1.2.3


From 4e4785bcf0c8503224fa6c17d8e0228de781bff6 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:17 -0700
Subject: [PATCH] mempolicies: fix policy_zone check

There is a check in zonelist_policy that compares pieces of the bitmap
obtained from a gfp mask via GFP_ZONETYPES with a zone number in function
zonelist_policy().

The bitmap is an ORed mask of __GFP_DMA, __GFP_DMA32 and __GFP_HIGHMEM.
The policy_zone is a zone number with the possible values of ZONE_DMA,
ZONE_DMA32, ZONE_HIGHMEM and ZONE_NORMAL. These are two different domains
of values.

For some reason seemed to work before the zone reduction patchset (It
definitely works on SGI boxes since we just have one zone and the check
cannot fail).

With the zone reduction patchset this check definitely fails on systems
with two zones if the system actually has memory in both zones.

This is because ZONE_NORMAL is selected using no __GFP flag at
all and thus gfp_zone(gfpmask) == 0. ZONE_DMA is selected when __GFP_DMA
is set. __GFP_DMA is 0x01.  So gfp_zone(gfpmask) == 1.

policy_zone is set to ZONE_NORMAL (==1) if ZONE_NORMAL and ZONE_DMA are
populated.

For ZONE_NORMAL gfp_zone(<no _GFP_DMA>) yields 0 which is <
policy_zone(ZONE_NORMAL) and so policy is not applied to regular memory
allocations!

Instead gfp_zone(__GFP_DMA) == 1 which results in policy being applied
to DMA allocations!

What we realy want in that place is to establish the highest allowable
zone for a given gfp_mask. If the highest zone is higher or equal to the
policy_zone then memory policies need to be applied. We have such
a highest_zone() function in page_alloc.c.

So move the highest_zone() function from mm/page_alloc.c into
include/linux/gfp.h.  On the way we simplify the function and use the new
zone_type that was also introduced with the zone reduction patchset plus we
also specify the right type for the gfp flags parameter.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h | 15 +++++++++++++++
 mm/mempolicy.c      |  2 +-
 mm/page_alloc.c     | 16 ----------------
 3 files changed, 16 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 2a2153ebfe0b..a0992d392d79 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -85,6 +85,21 @@ static inline int gfp_zone(gfp_t gfp)
 	return zone;
 }
 
+static inline enum zone_type highest_zone(gfp_t flags)
+{
+	if (flags & __GFP_DMA)
+		return ZONE_DMA;
+#ifdef CONFIG_ZONE_DMA32
+	if (flags & __GFP_DMA32)
+		return ZONE_DMA32;
+#endif
+#ifdef CONFIG_HIGHMEM
+	if (flags & __GFP_HIGHMEM)
+		return ZONE_HIGHMEM;
+#endif
+	return ZONE_NORMAL;
+}
+
 /*
  * There is only one page-allocator function, and two main namespaces to
  * it. The alloc_page*() variants return 'struct page *' and as such
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index a9963ceddd65..9870624d72a6 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1096,7 +1096,7 @@ static struct zonelist *zonelist_policy(gfp_t gfp, struct mempolicy *policy)
 	case MPOL_BIND:
 		/* Lower zones don't get a policy applied */
 		/* Careful: current->mems_allowed might have moved */
-		if (gfp_zone(gfp) >= policy_zone)
+		if (highest_zone(gfp) >= policy_zone)
 			if (cpuset_zonelist_valid_mems_allowed(policy->v.zonelist))
 				return policy->v.zonelist;
 		/*FALL THROUGH*/
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 6c7c2dd1b3ed..25f39865e324 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1376,22 +1376,6 @@ static int __meminit build_zonelists_node(pg_data_t *pgdat,
 	return nr_zones;
 }
 
-static inline int highest_zone(int zone_bits)
-{
-	int res = ZONE_NORMAL;
-#ifdef CONFIG_HIGHMEM
-	if (zone_bits & (__force int)__GFP_HIGHMEM)
-		res = ZONE_HIGHMEM;
-#endif
-#ifdef CONFIG_ZONE_DMA32
-	if (zone_bits & (__force int)__GFP_DMA32)
-		res = ZONE_DMA32;
-#endif
-	if (zone_bits & (__force int)__GFP_DMA)
-		res = ZONE_DMA;
-	return res;
-}
-
 #ifdef CONFIG_NUMA
 #define MAX_NODE_LOAD (num_online_nodes())
 static int __meminitdata node_load[MAX_NUMNODES];
-- 
cgit v1.2.3


From 2f6726e54a9410e2e4cee864947c05e954051916 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:18 -0700
Subject: [PATCH] Apply type enum zone_type

After we have done this we can now do some typing cleanup.

The memory policy layer keeps a policy_zone that specifies
the zone that gets memory policies applied. This variable
can now be of type enum zone_type.

The check_highest_zone function and the build_zonelists funnctionm must
then also take a enum zone_type parameter.

Plus there are a number of loops over zones that also should use
zone_type.

We run into some troubles at some points with functions that need a
zone_type variable to become -1. Fix that up.

[pj@sgi.com: fix set_mempolicy() crash]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mempolicy.h |  4 ++--
 mm/mempolicy.c            | 11 ++++++++---
 mm/page_alloc.c           | 27 +++++++++++++++++----------
 3 files changed, 27 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h
index 72440f0a443d..09f0f575ddff 100644
--- a/include/linux/mempolicy.h
+++ b/include/linux/mempolicy.h
@@ -162,9 +162,9 @@ extern struct zonelist *huge_zonelist(struct vm_area_struct *vma,
 		unsigned long addr);
 extern unsigned slab_node(struct mempolicy *policy);
 
-extern int policy_zone;
+extern enum zone_type policy_zone;
 
-static inline void check_highest_zone(int k)
+static inline void check_highest_zone(enum zone_type k)
 {
 	if (k > policy_zone)
 		policy_zone = k;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 9870624d72a6..7da4142ce960 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -105,7 +105,7 @@ static struct kmem_cache *sn_cache;
 
 /* Highest zone. An specific allocation for a zone below that is not
    policied. */
-int policy_zone = ZONE_DMA;
+enum zone_type policy_zone = ZONE_DMA;
 
 struct mempolicy default_policy = {
 	.refcnt = ATOMIC_INIT(1), /* never free it */
@@ -137,7 +137,8 @@ static int mpol_check_policy(int mode, nodemask_t *nodes)
 static struct zonelist *bind_zonelist(nodemask_t *nodes)
 {
 	struct zonelist *zl;
-	int num, max, nd, k;
+	int num, max, nd;
+	enum zone_type k;
 
 	max = 1 + MAX_NR_ZONES * nodes_weight(*nodes);
 	zl = kmalloc(sizeof(struct zone *) * max, GFP_KERNEL);
@@ -148,12 +149,16 @@ static struct zonelist *bind_zonelist(nodemask_t *nodes)
 	   lower zones etc. Avoid empty zones because the memory allocator
 	   doesn't like them. If you implement node hot removal you
 	   have to fix that. */
-	for (k = policy_zone; k >= 0; k--) { 
+	k = policy_zone;
+	while (1) {
 		for_each_node_mask(nd, *nodes) { 
 			struct zone *z = &NODE_DATA(nd)->node_zones[k];
 			if (z->present_pages > 0) 
 				zl->zones[num++] = z;
 		}
+		if (k == 0)
+			break;
+		k--;
 	}
 	zl->zones[num] = NULL;
 	return zl;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 25f39865e324..9c44b9a39d30 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -637,7 +637,8 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
  */
 void drain_node_pages(int nodeid)
 {
-	int i, z;
+	int i;
+	enum zone_type z;
 	unsigned long flags;
 
 	for (z = 0; z < MAX_NR_ZONES; z++) {
@@ -1158,7 +1159,8 @@ EXPORT_SYMBOL(nr_free_pages);
 #ifdef CONFIG_NUMA
 unsigned int nr_free_pages_pgdat(pg_data_t *pgdat)
 {
-	unsigned int i, sum = 0;
+	unsigned int sum = 0;
+	enum zone_type i;
 
 	for (i = 0; i < MAX_NR_ZONES; i++)
 		sum += pgdat->node_zones[i].free_pages;
@@ -1358,21 +1360,22 @@ void show_free_areas(void)
  * Add all populated zones of a node to the zonelist.
  */
 static int __meminit build_zonelists_node(pg_data_t *pgdat,
-			struct zonelist *zonelist, int nr_zones, int zone_type)
+			struct zonelist *zonelist, int nr_zones, enum zone_type zone_type)
 {
 	struct zone *zone;
 
 	BUG_ON(zone_type >= MAX_NR_ZONES);
+	zone_type++;
 
 	do {
+		zone_type--;
 		zone = pgdat->node_zones + zone_type;
 		if (populated_zone(zone)) {
 			zonelist->zones[nr_zones++] = zone;
 			check_highest_zone(zone_type);
 		}
-		zone_type--;
 
-	} while (zone_type >= 0);
+	} while (zone_type);
 	return nr_zones;
 }
 
@@ -1441,10 +1444,11 @@ static int __meminit find_next_best_node(int node, nodemask_t *used_node_mask)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, j, k, node, local_node;
+	int i, j, node, local_node;
 	int prev_node, load;
 	struct zonelist *zonelist;
 	nodemask_t used_mask;
+	enum zone_type k;
 
 	/* initialize zonelists */
 	for (i = 0; i < GFP_ZONETYPES; i++) {
@@ -1628,7 +1632,7 @@ static void __init calculate_zone_totalpages(struct pglist_data *pgdat,
 		unsigned long *zones_size, unsigned long *zholes_size)
 {
 	unsigned long realtotalpages, totalpages = 0;
-	int i;
+	enum zone_type i;
 
 	for (i = 0; i < MAX_NR_ZONES; i++)
 		totalpages += zones_size[i];
@@ -2116,7 +2120,7 @@ static void calculate_totalreserve_pages(void)
 {
 	struct pglist_data *pgdat;
 	unsigned long reserve_pages = 0;
-	int i, j;
+	enum zone_type i, j;
 
 	for_each_online_pgdat(pgdat) {
 		for (i = 0; i < MAX_NR_ZONES; i++) {
@@ -2149,7 +2153,7 @@ static void calculate_totalreserve_pages(void)
 static void setup_per_zone_lowmem_reserve(void)
 {
 	struct pglist_data *pgdat;
-	int j, idx;
+	enum zone_type j, idx;
 
 	for_each_online_pgdat(pgdat) {
 		for (j = 0; j < MAX_NR_ZONES; j++) {
@@ -2158,9 +2162,12 @@ static void setup_per_zone_lowmem_reserve(void)
 
 			zone->lowmem_reserve[j] = 0;
 
-			for (idx = j-1; idx >= 0; idx--) {
+			idx = j;
+			while (idx) {
 				struct zone *lower_zone;
 
+				idx--;
+
 				if (sysctl_lowmem_reserve_ratio[idx] < 1)
 					sysctl_lowmem_reserve_ratio[idx] = 1;
 
-- 
cgit v1.2.3


From 19655d3487001d7df0e10e9cbfc27c758b77c2b5 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:19 -0700
Subject: [PATCH] linearly index zone->node_zonelists[]

I wonder why we need this bitmask indexing into zone->node_zonelists[]?

We always start with the highest zone and then include all lower zones
if we build zonelists.

Are there really cases where we need allocation from ZONE_DMA or
ZONE_HIGHMEM but not ZONE_NORMAL? It seems that the current implementation
of highest_zone() makes that already impossible.

If we go linear on the index then gfp_zone() == highest_zone() and a lot
of definitions fall by the wayside.

We can now revert back to the use of gfp_zone() in mempolicy.c ;-)

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h    | 12 +-----------
 include/linux/mmzone.h | 52 +++++---------------------------------------------
 mm/mempolicy.c         |  2 +-
 mm/page_alloc.c        | 27 +++++++++++---------------
 4 files changed, 18 insertions(+), 75 deletions(-)

(limited to 'include')

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index a0992d392d79..63ab88a36f39 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -12,9 +12,6 @@ struct vm_area_struct;
  *
  * Zone modifiers (see linux/mmzone.h - low three bits)
  *
- * These may be masked by GFP_ZONEMASK to make allocations with this bit
- * set fall back to ZONE_NORMAL.
- *
  * Do not put any conditional on these. If necessary modify the definitions
  * without the underscores and use the consistently. The definitions here may
  * be used in bit comparisons.
@@ -78,14 +75,7 @@ struct vm_area_struct;
 #define GFP_DMA32	__GFP_DMA32
 
 
-static inline int gfp_zone(gfp_t gfp)
-{
-	int zone = GFP_ZONEMASK & (__force int) gfp;
-	BUG_ON(zone >= GFP_ZONETYPES);
-	return zone;
-}
-
-static inline enum zone_type highest_zone(gfp_t flags)
+static inline enum zone_type gfp_zone(gfp_t flags)
 {
 	if (flags & __GFP_DMA)
 		return ZONE_DMA;
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 76d33e688593..7fe317164b73 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -137,60 +137,18 @@ enum zone_type {
 	MAX_NR_ZONES
 };
 
-
 /*
  * When a memory allocation must conform to specific limitations (such
  * as being suitable for DMA) the caller will pass in hints to the
  * allocator in the gfp_mask, in the zone modifier bits.  These bits
  * are used to select a priority ordered list of memory zones which
- * match the requested limits.  GFP_ZONEMASK defines which bits within
- * the gfp_mask should be considered as zone modifiers.  Each valid
- * combination of the zone modifier bits has a corresponding list
- * of zones (in node_zonelists).  Thus for two zone modifiers there
- * will be a maximum of 4 (2 ** 2) zonelists, for 3 modifiers there will
- * be 8 (2 ** 3) zonelists.  GFP_ZONETYPES defines the number of possible
- * combinations of zone modifiers in "zone modifier space".
- *
- * As an optimisation any zone modifier bits which are only valid when
- * no other zone modifier bits are set (loners) should be placed in
- * the highest order bits of this field.  This allows us to reduce the
- * extent of the zonelists thus saving space.  For example in the case
- * of three zone modifier bits, we could require up to eight zonelists.
- * If the left most zone modifier is a "loner" then the highest valid
- * zonelist would be four allowing us to allocate only five zonelists.
- * Use the first form for GFP_ZONETYPES when the left most bit is not
- * a "loner", otherwise use the second.
- *
- * NOTE! Make sure this matches the zones in <linux/gfp.h>
+ * match the requested limits. See gfp_zone() in include/linux/gfp.h
  */
 
-#ifdef CONFIG_ZONE_DMA32
-
-#ifdef CONFIG_HIGHMEM
-#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
-#define GFP_ZONEMASK		0x07
-#define ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
-#else
-#define GFP_ZONETYPES		((0x07 + 1) / 2 + 1)    /* Loner */
-/* Mask __GFP_HIGHMEM */
-#define GFP_ZONEMASK		0x05
-#define ZONES_SHIFT		2
-#endif
-
-#else
-#ifdef CONFIG_HIGHMEM
-
-#define GFP_ZONEMASK		0x03
-#define ZONES_SHIFT		2
-#define GFP_ZONETYPES		3
-
+#if !defined(CONFIG_ZONE_DMA32) && !defined(CONFIG_HIGHMEM)
+#define ZONES_SHIFT 1
 #else
-
-#define GFP_ZONEMASK		0x01
-#define ZONES_SHIFT		1
-#define GFP_ZONETYPES		2
-
-#endif
+#define ZONES_SHIFT 2
 #endif
 
 struct zone {
@@ -360,7 +318,7 @@ struct zonelist {
 struct bootmem_data;
 typedef struct pglist_data {
 	struct zone node_zones[MAX_NR_ZONES];
-	struct zonelist node_zonelists[GFP_ZONETYPES];
+	struct zonelist node_zonelists[MAX_NR_ZONES];
 	int nr_zones;
 #ifdef CONFIG_FLAT_NODE_MEM_MAP
 	struct page *node_mem_map;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 7da4142ce960..c3429a710ab1 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1101,7 +1101,7 @@ static struct zonelist *zonelist_policy(gfp_t gfp, struct mempolicy *policy)
 	case MPOL_BIND:
 		/* Lower zones don't get a policy applied */
 		/* Careful: current->mems_allowed might have moved */
-		if (highest_zone(gfp) >= policy_zone)
+		if (gfp_zone(gfp) >= policy_zone)
 			if (cpuset_zonelist_valid_mems_allowed(policy->v.zonelist))
 				return policy->v.zonelist;
 		/*FALL THROUGH*/
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9c44b9a39d30..208a6b03aa78 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1444,14 +1444,14 @@ static int __meminit find_next_best_node(int node, nodemask_t *used_node_mask)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, j, node, local_node;
+	int j, node, local_node;
+	enum zone_type i;
 	int prev_node, load;
 	struct zonelist *zonelist;
 	nodemask_t used_mask;
-	enum zone_type k;
 
 	/* initialize zonelists */
-	for (i = 0; i < GFP_ZONETYPES; i++) {
+	for (i = 0; i < MAX_NR_ZONES; i++) {
 		zonelist = pgdat->node_zonelists + i;
 		zonelist->zones[0] = NULL;
 	}
@@ -1481,13 +1481,11 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 			node_load[node] += load;
 		prev_node = node;
 		load--;
-		for (i = 0; i < GFP_ZONETYPES; i++) {
+		for (i = 0; i < MAX_NR_ZONES; i++) {
 			zonelist = pgdat->node_zonelists + i;
 			for (j = 0; zonelist->zones[j] != NULL; j++);
 
-			k = highest_zone(i);
-
-	 		j = build_zonelists_node(NODE_DATA(node), zonelist, j, k);
+	 		j = build_zonelists_node(NODE_DATA(node), zonelist, j, i);
 			zonelist->zones[j] = NULL;
 		}
 	}
@@ -1497,19 +1495,16 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, node, local_node;
-	enum zone_type k;
-	enum zone_type j;
+	int node, local_node;
+	enum zone_type i,j;
 
 	local_node = pgdat->node_id;
-	for (i = 0; i < GFP_ZONETYPES; i++) {
+	for (i = 0; i < MAX_NR_ZONES; i++) {
 		struct zonelist *zonelist;
 
 		zonelist = pgdat->node_zonelists + i;
 
-		j = 0;
-		k = highest_zone(i);
- 		j = build_zonelists_node(pgdat, zonelist, j, k);
+ 		j = build_zonelists_node(pgdat, zonelist, 0, i);
  		/*
  		 * Now we build the zonelist so that it contains the zones
  		 * of all the other nodes.
@@ -1521,12 +1516,12 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 		for (node = local_node + 1; node < MAX_NUMNODES; node++) {
 			if (!node_online(node))
 				continue;
-			j = build_zonelists_node(NODE_DATA(node), zonelist, j, k);
+			j = build_zonelists_node(NODE_DATA(node), zonelist, j, i);
 		}
 		for (node = 0; node < local_node; node++) {
 			if (!node_online(node))
 				continue;
-			j = build_zonelists_node(NODE_DATA(node), zonelist, j, k);
+			j = build_zonelists_node(NODE_DATA(node), zonelist, j, i);
 		}
 
 		zonelist->zones[j] = NULL;
-- 
cgit v1.2.3


From 8bc719d3cab8414938f9ea6e33b58d8810d18068 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Mon, 25 Sep 2006 23:31:20 -0700
Subject: [PATCH] out of memory notifier

Add a notifer chain to the out of memory killer.  If one of the registered
callbacks could release some memory, do not kill the process but return and
retry the allocation that forced the oom killer to run.

The purpose of the notifier is to add a safety net in the presence of
memory ballooners.  If the resource manager inflated the balloon to a size
where memory allocations can not be satisfied anymore, it is better to
deflate the balloon a bit instead of killing processes.

The implementation for the s390 ballooner is included.

[akpm@osdl.org: cleanups]
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/mm/cmm.c   | 155 +++++++++++++++++++++++++++++++--------------------
 include/linux/swap.h |   4 ++
 mm/oom_kill.c        |  22 ++++++++
 3 files changed, 121 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c
index 786a44dba5bf..f2e00df99bae 100644
--- a/arch/s390/mm/cmm.c
+++ b/arch/s390/mm/cmm.c
@@ -15,6 +15,7 @@
 #include <linux/sched.h>
 #include <linux/sysctl.h>
 #include <linux/ctype.h>
+#include <linux/swap.h>
 
 #include <asm/pgalloc.h>
 #include <asm/uaccess.h>
@@ -34,17 +35,18 @@ struct cmm_page_array {
 	unsigned long pages[CMM_NR_PAGES];
 };
 
-static long cmm_pages = 0;
-static long cmm_timed_pages = 0;
-static volatile long cmm_pages_target = 0;
-static volatile long cmm_timed_pages_target = 0;
-static long cmm_timeout_pages = 0;
-static long cmm_timeout_seconds = 0;
+static long cmm_pages;
+static long cmm_timed_pages;
+static volatile long cmm_pages_target;
+static volatile long cmm_timed_pages_target;
+static long cmm_timeout_pages;
+static long cmm_timeout_seconds;
 
-static struct cmm_page_array *cmm_page_list = NULL;
-static struct cmm_page_array *cmm_timed_page_list = NULL;
+static struct cmm_page_array *cmm_page_list;
+static struct cmm_page_array *cmm_timed_page_list;
+static DEFINE_SPINLOCK(cmm_lock);
 
-static unsigned long cmm_thread_active = 0;
+static unsigned long cmm_thread_active;
 static struct work_struct cmm_thread_starter;
 static wait_queue_head_t cmm_thread_wait;
 static struct timer_list cmm_timer;
@@ -53,58 +55,89 @@ static void cmm_timer_fn(unsigned long);
 static void cmm_set_timer(void);
 
 static long
-cmm_alloc_pages(long pages, long *counter, struct cmm_page_array **list)
+cmm_alloc_pages(long nr, long *counter, struct cmm_page_array **list)
 {
-	struct cmm_page_array *pa;
-	unsigned long page;
+	struct cmm_page_array *pa, *npa;
+	unsigned long addr;
 
-	pa = *list;
-	while (pages) {
-		page = __get_free_page(GFP_NOIO);
-		if (!page)
+	while (nr) {
+		addr = __get_free_page(GFP_NOIO);
+		if (!addr)
 			break;
+		spin_lock(&cmm_lock);
+		pa = *list;
 		if (!pa || pa->index >= CMM_NR_PAGES) {
 			/* Need a new page for the page list. */
-			pa = (struct cmm_page_array *)
+			spin_unlock(&cmm_lock);
+			npa = (struct cmm_page_array *)
 				__get_free_page(GFP_NOIO);
-			if (!pa) {
-				free_page(page);
+			if (!npa) {
+				free_page(addr);
 				break;
 			}
-			pa->next = *list;
-			pa->index = 0;
-			*list = pa;
+			spin_lock(&cmm_lock);
+			pa = *list;
+			if (!pa || pa->index >= CMM_NR_PAGES) {
+				npa->next = pa;
+				npa->index = 0;
+				pa = npa;
+				*list = pa;
+			} else
+				free_page((unsigned long) npa);
 		}
-		diag10(page);
-		pa->pages[pa->index++] = page;
+		diag10(addr);
+		pa->pages[pa->index++] = addr;
 		(*counter)++;
-		pages--;
+		spin_unlock(&cmm_lock);
+		nr--;
 	}
-	return pages;
+	return nr;
 }
 
-static void
-cmm_free_pages(long pages, long *counter, struct cmm_page_array **list)
+static long
+cmm_free_pages(long nr, long *counter, struct cmm_page_array **list)
 {
 	struct cmm_page_array *pa;
-	unsigned long page;
+	unsigned long addr;
 
+	spin_lock(&cmm_lock);
 	pa = *list;
-	while (pages) {
+	while (nr) {
 		if (!pa || pa->index <= 0)
 			break;
-		page = pa->pages[--pa->index];
+		addr = pa->pages[--pa->index];
 		if (pa->index == 0) {
 			pa = pa->next;
 			free_page((unsigned long) *list);
 			*list = pa;
 		}
-		free_page(page);
+		free_page(addr);
 		(*counter)--;
-		pages--;
+		nr--;
 	}
+	spin_unlock(&cmm_lock);
+	return nr;
 }
 
+static int cmm_oom_notify(struct notifier_block *self,
+			  unsigned long dummy, void *parm)
+{
+	unsigned long *freed = parm;
+	long nr = 256;
+
+	nr = cmm_free_pages(nr, &cmm_timed_pages, &cmm_timed_page_list);
+	if (nr > 0)
+		nr = cmm_free_pages(nr, &cmm_pages, &cmm_page_list);
+	cmm_pages_target = cmm_pages;
+	cmm_timed_pages_target = cmm_timed_pages;
+	*freed += 256 - nr;
+	return NOTIFY_OK;
+}
+
+static struct notifier_block cmm_oom_nb = {
+	.notifier_call = cmm_oom_notify
+};
+
 static int
 cmm_thread(void *dummy)
 {
@@ -177,21 +210,21 @@ cmm_set_timer(void)
 static void
 cmm_timer_fn(unsigned long ignored)
 {
-	long pages;
+	long nr;
 
-	pages = cmm_timed_pages_target - cmm_timeout_pages;
-	if (pages < 0)
+	nr = cmm_timed_pages_target - cmm_timeout_pages;
+	if (nr < 0)
 		cmm_timed_pages_target = 0;
 	else
-		cmm_timed_pages_target = pages;
+		cmm_timed_pages_target = nr;
 	cmm_kick_thread();
 	cmm_set_timer();
 }
 
 void
-cmm_set_pages(long pages)
+cmm_set_pages(long nr)
 {
-	cmm_pages_target = pages;
+	cmm_pages_target = nr;
 	cmm_kick_thread();
 }
 
@@ -202,9 +235,9 @@ cmm_get_pages(void)
 }
 
 void
-cmm_add_timed_pages(long pages)
+cmm_add_timed_pages(long nr)
 {
-	cmm_timed_pages_target += pages;
+	cmm_timed_pages_target += nr;
 	cmm_kick_thread();
 }
 
@@ -215,9 +248,9 @@ cmm_get_timed_pages(void)
 }
 
 void
-cmm_set_timeout(long pages, long seconds)
+cmm_set_timeout(long nr, long seconds)
 {
-	cmm_timeout_pages = pages;
+	cmm_timeout_pages = nr;
 	cmm_timeout_seconds = seconds;
 	cmm_set_timer();
 }
@@ -245,7 +278,7 @@ cmm_pages_handler(ctl_table *ctl, int write, struct file *filp,
 		  void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	char buf[16], *p;
-	long pages;
+	long nr;
 	int len;
 
 	if (!*lenp || (*ppos && !write)) {
@@ -260,17 +293,17 @@ cmm_pages_handler(ctl_table *ctl, int write, struct file *filp,
 			return -EFAULT;
 		buf[sizeof(buf) - 1] = '\0';
 		cmm_skip_blanks(buf, &p);
-		pages = simple_strtoul(p, &p, 0);
+		nr = simple_strtoul(p, &p, 0);
 		if (ctl == &cmm_table[0])
-			cmm_set_pages(pages);
+			cmm_set_pages(nr);
 		else
-			cmm_add_timed_pages(pages);
+			cmm_add_timed_pages(nr);
 	} else {
 		if (ctl == &cmm_table[0])
-			pages = cmm_get_pages();
+			nr = cmm_get_pages();
 		else
-			pages = cmm_get_timed_pages();
-		len = sprintf(buf, "%ld\n", pages);
+			nr = cmm_get_timed_pages();
+		len = sprintf(buf, "%ld\n", nr);
 		if (len > *lenp)
 			len = *lenp;
 		if (copy_to_user(buffer, buf, len))
@@ -286,7 +319,7 @@ cmm_timeout_handler(ctl_table *ctl, int write, struct file *filp,
 		    void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	char buf[64], *p;
-	long pages, seconds;
+	long nr, seconds;
 	int len;
 
 	if (!*lenp || (*ppos && !write)) {
@@ -301,10 +334,10 @@ cmm_timeout_handler(ctl_table *ctl, int write, struct file *filp,
 			return -EFAULT;
 		buf[sizeof(buf) - 1] = '\0';
 		cmm_skip_blanks(buf, &p);
-		pages = simple_strtoul(p, &p, 0);
+		nr = simple_strtoul(p, &p, 0);
 		cmm_skip_blanks(p, &p);
 		seconds = simple_strtoul(p, &p, 0);
-		cmm_set_timeout(pages, seconds);
+		cmm_set_timeout(nr, seconds);
 	} else {
 		len = sprintf(buf, "%ld %ld\n",
 			      cmm_timeout_pages, cmm_timeout_seconds);
@@ -357,7 +390,7 @@ static struct ctl_table cmm_dir_table[] = {
 static void
 cmm_smsg_target(char *from, char *msg)
 {
-	long pages, seconds;
+	long nr, seconds;
 
 	if (strlen(sender) > 0 && strcmp(from, sender) != 0)
 		return;
@@ -366,27 +399,27 @@ cmm_smsg_target(char *from, char *msg)
 	if (strncmp(msg, "SHRINK", 6) == 0) {
 		if (!cmm_skip_blanks(msg + 6, &msg))
 			return;
-		pages = simple_strtoul(msg, &msg, 0);
+		nr = simple_strtoul(msg, &msg, 0);
 		cmm_skip_blanks(msg, &msg);
 		if (*msg == '\0')
-			cmm_set_pages(pages);
+			cmm_set_pages(nr);
 	} else if (strncmp(msg, "RELEASE", 7) == 0) {
 		if (!cmm_skip_blanks(msg + 7, &msg))
 			return;
-		pages = simple_strtoul(msg, &msg, 0);
+		nr = simple_strtoul(msg, &msg, 0);
 		cmm_skip_blanks(msg, &msg);
 		if (*msg == '\0')
-			cmm_add_timed_pages(pages);
+			cmm_add_timed_pages(nr);
 	} else if (strncmp(msg, "REUSE", 5) == 0) {
 		if (!cmm_skip_blanks(msg + 5, &msg))
 			return;
-		pages = simple_strtoul(msg, &msg, 0);
+		nr = simple_strtoul(msg, &msg, 0);
 		if (!cmm_skip_blanks(msg, &msg))
 			return;
 		seconds = simple_strtoul(msg, &msg, 0);
 		cmm_skip_blanks(msg, &msg);
 		if (*msg == '\0')
-			cmm_set_timeout(pages, seconds);
+			cmm_set_timeout(nr, seconds);
 	}
 }
 #endif
@@ -402,6 +435,7 @@ cmm_init (void)
 #ifdef CONFIG_CMM_IUCV
 	smsg_register_callback(SMSG_PREFIX, cmm_smsg_target);
 #endif
+	register_oom_notifier(&cmm_oom_nb);
 	INIT_WORK(&cmm_thread_starter, (void *) cmm_start_thread, NULL);
 	init_waitqueue_head(&cmm_thread_wait);
 	init_timer(&cmm_timer);
@@ -411,6 +445,7 @@ cmm_init (void)
 static void
 cmm_exit(void)
 {
+	unregister_oom_notifier(&cmm_oom_nb);
 	cmm_free_pages(cmm_pages, &cmm_pages, &cmm_page_list);
 	cmm_free_pages(cmm_timed_pages, &cmm_timed_pages, &cmm_timed_page_list);
 #ifdef CONFIG_CMM_PROC
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 34a6bc3e6cf3..32db06c8ffe0 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -10,6 +10,8 @@
 #include <asm/atomic.h>
 #include <asm/page.h>
 
+struct notifier_block;
+
 #define SWAP_FLAG_PREFER	0x8000	/* set if swap priority specified */
 #define SWAP_FLAG_PRIO_MASK	0x7fff
 #define SWAP_FLAG_PRIO_SHIFT	0
@@ -156,6 +158,8 @@ struct swap_list_t {
 
 /* linux/mm/oom_kill.c */
 extern void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order);
+extern int register_oom_notifier(struct notifier_block *nb);
+extern int unregister_oom_notifier(struct notifier_block *nb);
 
 /* linux/mm/memory.c */
 extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index b9af136e5cfa..7d056843fa2d 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -21,6 +21,8 @@
 #include <linux/timex.h>
 #include <linux/jiffies.h>
 #include <linux/cpuset.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
 
 int sysctl_panic_on_oom;
 /* #define DEBUG */
@@ -306,6 +308,20 @@ static int oom_kill_process(struct task_struct *p, unsigned long points,
 	return oom_kill_task(p, message);
 }
 
+static BLOCKING_NOTIFIER_HEAD(oom_notify_list);
+
+int register_oom_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&oom_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(register_oom_notifier);
+
+int unregister_oom_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&oom_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(unregister_oom_notifier);
+
 /**
  * out_of_memory - kill the "best" process when we run out of memory
  *
@@ -318,6 +334,12 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 {
 	struct task_struct *p;
 	unsigned long points = 0;
+	unsigned long freed = 0;
+
+	blocking_notifier_call_chain(&oom_notify_list, 0, &freed);
+	if (freed > 0)
+		/* Got some memory back in the last second. */
+		return;
 
 	if (printk_ratelimit()) {
 		printk("oom-killer: gfp_mask=0x%x, order=%d\n",
-- 
cgit v1.2.3


From 7ff6f08295d90ab20d25200ef485ebb45b1b8d71 Mon Sep 17 00:00:00 2001
From: Martin Peschke <mp3@de.ibm.com>
Date: Mon, 25 Sep 2006 23:31:21 -0700
Subject: [PATCH] CPU hotplug compatible alloc_percpu()

This patch splits alloc_percpu() up into two phases.  Likewise for
free_percpu().  This allows clients to limit initial allocations to online
cpu's, and to populate or depopulate per-cpu data at run time as needed:

  struct my_struct *obj;

  /* initial allocation for online cpu's */
  obj = percpu_alloc(sizeof(struct my_struct), GFP_KERNEL);

  ...

  /* populate per-cpu data for cpu coming online */
  ptr = percpu_populate(obj, sizeof(struct my_struct), GFP_KERNEL, cpu);

  ...

  /* access per-cpu object */
  ptr = percpu_ptr(obj, smp_processor_id());

  ...

  /* depopulate per-cpu data for cpu going offline */
  percpu_depopulate(obj, cpu);

  ...

  /* final removal */
  percpu_free(obj);

Signed-off-by: Martin Peschke <mp3@de.ibm.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/percpu.h |  79 +++++++++++++++++------
 mm/slab.c              | 166 ++++++++++++++++++++++++++++++++-----------------
 2 files changed, 169 insertions(+), 76 deletions(-)

(limited to 'include')

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index f926490a7d8b..3835a9642f13 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -1,9 +1,12 @@
 #ifndef __LINUX_PERCPU_H
 #define __LINUX_PERCPU_H
+
 #include <linux/spinlock.h> /* For preempt_disable() */
 #include <linux/slab.h> /* For kmalloc() */
 #include <linux/smp.h>
 #include <linux/string.h> /* For memset() */
+#include <linux/cpumask.h>
+
 #include <asm/percpu.h>
 
 /* Enough to cover all DEFINE_PER_CPUs in kernel, including modules. */
@@ -27,39 +30,77 @@ struct percpu_data {
 	void *ptrs[NR_CPUS];
 };
 
+#define __percpu_disguise(pdata) (struct percpu_data *)~(unsigned long)(pdata)
 /* 
- * Use this to get to a cpu's version of the per-cpu object allocated using
- * alloc_percpu.  Non-atomic access to the current CPU's version should
+ * Use this to get to a cpu's version of the per-cpu object dynamically
+ * allocated. Non-atomic access to the current CPU's version should
  * probably be combined with get_cpu()/put_cpu().
  */ 
-#define per_cpu_ptr(ptr, cpu)                   \
-({                                              \
-        struct percpu_data *__p = (struct percpu_data *)~(unsigned long)(ptr); \
-        (__typeof__(ptr))__p->ptrs[(cpu)];	\
+#define percpu_ptr(ptr, cpu)                              \
+({                                                        \
+        struct percpu_data *__p = __percpu_disguise(ptr); \
+        (__typeof__(ptr))__p->ptrs[(cpu)];	          \
 })
 
-extern void *__alloc_percpu(size_t size);
-extern void free_percpu(const void *);
+extern void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu);
+extern void percpu_depopulate(void *__pdata, int cpu);
+extern int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+				  cpumask_t *mask);
+extern void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask);
+extern void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask);
+extern void percpu_free(void *__pdata);
 
 #else /* CONFIG_SMP */
 
-#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
+#define percpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
+
+static inline void percpu_depopulate(void *__pdata, int cpu)
+{
+}
+
+static inline void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
+{
+}
 
-static inline void *__alloc_percpu(size_t size)
+static inline void *percpu_populate(void *__pdata, size_t size, gfp_t gfp,
+				    int cpu)
 {
-	void *ret = kmalloc(size, GFP_KERNEL);
-	if (ret)
-		memset(ret, 0, size);
-	return ret;
+	return percpu_ptr(__pdata, cpu);
 }
-static inline void free_percpu(const void *ptr)
-{	
-	kfree(ptr);
+
+static inline int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+					 cpumask_t *mask)
+{
+	return 0;
+}
+
+static inline void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+{
+	return kzalloc(size, gfp);
+}
+
+static inline void percpu_free(void *__pdata)
+{
+	kfree(__pdata);
 }
 
 #endif /* CONFIG_SMP */
 
-/* Simple wrapper for the common case: zeros memory. */
-#define alloc_percpu(type)	((type *)(__alloc_percpu(sizeof(type))))
+#define percpu_populate_mask(__pdata, size, gfp, mask) \
+	__percpu_populate_mask((__pdata), (size), (gfp), &(mask))
+#define percpu_depopulate_mask(__pdata, mask) \
+	__percpu_depopulate_mask((__pdata), &(mask))
+#define percpu_alloc_mask(size, gfp, mask) \
+	__percpu_alloc_mask((size), (gfp), &(mask))
+
+#define percpu_alloc(size, gfp) percpu_alloc_mask((size), (gfp), cpu_online_map)
+
+/* (legacy) interface for use without CPU hotplug handling */
+
+#define __alloc_percpu(size)	percpu_alloc_mask((size), GFP_KERNEL, \
+						  cpu_possible_map)
+#define alloc_percpu(type)	(type *)__alloc_percpu(sizeof(type))
+#define free_percpu(ptr)	percpu_free((ptr))
+#define per_cpu_ptr(ptr, cpu)	percpu_ptr((ptr), (cpu))
 
 #endif /* __LINUX_PERCPU_H */
diff --git a/mm/slab.c b/mm/slab.c
index 5870bcbd33cf..00584dbbec03 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3371,52 +3371,127 @@ EXPORT_SYMBOL(__kmalloc_track_caller);
 
 #ifdef CONFIG_SMP
 /**
- * __alloc_percpu - allocate one copy of the object for every present
- * cpu in the system, zeroing them.
- * Objects should be dereferenced using the per_cpu_ptr macro only.
+ * percpu_depopulate - depopulate per-cpu data for given cpu
+ * @__pdata: per-cpu data to depopulate
+ * @cpu: depopulate per-cpu data for this cpu
  *
- * @size: how many bytes of memory are required.
+ * Depopulating per-cpu data for a cpu going offline would be a typical
+ * use case. You need to register a cpu hotplug handler for that purpose.
  */
-void *__alloc_percpu(size_t size)
+void percpu_depopulate(void *__pdata, int cpu)
 {
-	int i;
-	struct percpu_data *pdata = kmalloc(sizeof(*pdata), GFP_KERNEL);
+	struct percpu_data *pdata = __percpu_disguise(__pdata);
+	if (pdata->ptrs[cpu]) {
+		kfree(pdata->ptrs[cpu]);
+		pdata->ptrs[cpu] = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(percpu_depopulate);
 
-	if (!pdata)
-		return NULL;
+/**
+ * percpu_depopulate_mask - depopulate per-cpu data for some cpu's
+ * @__pdata: per-cpu data to depopulate
+ * @mask: depopulate per-cpu data for cpu's selected through mask bits
+ */
+void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
+{
+	int cpu;
+	for_each_cpu_mask(cpu, *mask)
+		percpu_depopulate(__pdata, cpu);
+}
+EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
 
-	/*
-	 * Cannot use for_each_online_cpu since a cpu may come online
-	 * and we have no way of figuring out how to fix the array
-	 * that we have allocated then....
-	 */
-	for_each_possible_cpu(i) {
-		int node = cpu_to_node(i);
+/**
+ * percpu_populate - populate per-cpu data for given cpu
+ * @__pdata: per-cpu data to populate further
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @cpu: populate per-data for this cpu
+ *
+ * Populating per-cpu data for a cpu coming online would be a typical
+ * use case. You need to register a cpu hotplug handler for that purpose.
+ * Per-cpu object is populated with zeroed buffer.
+ */
+void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
+{
+	struct percpu_data *pdata = __percpu_disguise(__pdata);
+	int node = cpu_to_node(cpu);
 
-		if (node_online(node))
-			pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL, node);
-		else
-			pdata->ptrs[i] = kmalloc(size, GFP_KERNEL);
+	BUG_ON(pdata->ptrs[cpu]);
+	if (node_online(node)) {
+		/* FIXME: kzalloc_node(size, gfp, node) */
+		pdata->ptrs[cpu] = kmalloc_node(size, gfp, node);
+		if (pdata->ptrs[cpu])
+			memset(pdata->ptrs[cpu], 0, size);
+	} else
+		pdata->ptrs[cpu] = kzalloc(size, gfp);
+	return pdata->ptrs[cpu];
+}
+EXPORT_SYMBOL_GPL(percpu_populate);
 
-		if (!pdata->ptrs[i])
-			goto unwind_oom;
-		memset(pdata->ptrs[i], 0, size);
-	}
+/**
+ * percpu_populate_mask - populate per-cpu data for more cpu's
+ * @__pdata: per-cpu data to populate further
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @mask: populate per-cpu data for cpu's selected through mask bits
+ *
+ * Per-cpu objects are populated with zeroed buffers.
+ */
+int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+			   cpumask_t *mask)
+{
+	cpumask_t populated = CPU_MASK_NONE;
+	int cpu;
+
+	for_each_cpu_mask(cpu, *mask)
+		if (unlikely(!percpu_populate(__pdata, size, gfp, cpu))) {
+			__percpu_depopulate_mask(__pdata, &populated);
+			return -ENOMEM;
+		} else
+			cpu_set(cpu, populated);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__percpu_populate_mask);
 
-	/* Catch derefs w/o wrappers */
-	return (void *)(~(unsigned long)pdata);
+/**
+ * percpu_alloc_mask - initial setup of per-cpu data
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @mask: populate per-data for cpu's selected through mask bits
+ *
+ * Populating per-cpu data for all online cpu's would be a typical use case,
+ * which is simplified by the percpu_alloc() wrapper.
+ * Per-cpu objects are populated with zeroed buffers.
+ */
+void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+{
+	void *pdata = kzalloc(sizeof(struct percpu_data), gfp);
+	void *__pdata = __percpu_disguise(pdata);
 
-unwind_oom:
-	while (--i >= 0) {
-		if (!cpu_possible(i))
-			continue;
-		kfree(pdata->ptrs[i]);
-	}
+	if (unlikely(!pdata))
+		return NULL;
+	if (likely(!__percpu_populate_mask(__pdata, size, gfp, mask)))
+		return __pdata;
 	kfree(pdata);
 	return NULL;
 }
-EXPORT_SYMBOL(__alloc_percpu);
-#endif
+EXPORT_SYMBOL_GPL(__percpu_alloc_mask);
+
+/**
+ * percpu_free - final cleanup of per-cpu data
+ * @__pdata: object to clean up
+ *
+ * We simply clean up any per-cpu object left. No need for the client to
+ * track and specify through a bis mask which per-cpu objects are to free.
+ */
+void percpu_free(void *__pdata)
+{
+	__percpu_depopulate_mask(__pdata, &cpu_possible_map);
+	kfree(__percpu_disguise(__pdata));
+}
+EXPORT_SYMBOL_GPL(percpu_free);
+#endif	/* CONFIG_SMP */
 
 /**
  * kmem_cache_free - Deallocate an object
@@ -3463,29 +3538,6 @@ void kfree(const void *objp)
 }
 EXPORT_SYMBOL(kfree);
 
-#ifdef CONFIG_SMP
-/**
- * free_percpu - free previously allocated percpu memory
- * @objp: pointer returned by alloc_percpu.
- *
- * Don't free memory not originally allocated by alloc_percpu()
- * The complemented objp is to check for that.
- */
-void free_percpu(const void *objp)
-{
-	int i;
-	struct percpu_data *p = (struct percpu_data *)(~(unsigned long)objp);
-
-	/*
-	 * We allocate for all cpus so we cannot use for online cpu here.
-	 */
-	for_each_possible_cpu(i)
-	    kfree(p->ptrs[i]);
-	kfree(p);
-}
-EXPORT_SYMBOL(free_percpu);
-#endif
-
 unsigned int kmem_cache_size(struct kmem_cache *cachep)
 {
 	return obj_size(cachep);
-- 
cgit v1.2.3


From db37648cd6ce9b828abd6d49aa3d269926ee7b7d Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:24 -0700
Subject: [PATCH] mm: non syncing lock_page()

lock_page needs the caller to have a reference on the page->mapping inode
due to sync_page, ergo set_page_dirty_lock is obviously buggy according to
its comments.

Solve it by introducing a new lock_page_nosync which does not do a sync_page.

akpm: unpleasant solution to an unpleasant problem.  If it goes wrong it could
cause great slowdowns while the lock_page() caller waits for kblockd to
perform the unplug.  And if a filesystem has special sync_page() requirements
(none presently do), permanent hangs are possible.

otoh, set_page_dirty_lock() is usually (always?) called against userspace
pages.  They are always up-to-date, so there shouldn't be any pending read I/O
against these pages.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/pagemap.h | 15 +++++++++++++++
 mm/filemap.c            | 17 +++++++++++++++++
 mm/page-writeback.c     |  2 +-
 3 files changed, 33 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 0a2f5d27f60e..64f950925151 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -130,14 +130,29 @@ static inline pgoff_t linear_page_index(struct vm_area_struct *vma,
 }
 
 extern void FASTCALL(__lock_page(struct page *page));
+extern void FASTCALL(__lock_page_nosync(struct page *page));
 extern void FASTCALL(unlock_page(struct page *page));
 
+/*
+ * lock_page may only be called if we have the page's inode pinned.
+ */
 static inline void lock_page(struct page *page)
 {
 	might_sleep();
 	if (TestSetPageLocked(page))
 		__lock_page(page);
 }
+
+/*
+ * lock_page_nosync should only be used if we can't pin the page's inode.
+ * Doesn't play quite so well with block device plugging.
+ */
+static inline void lock_page_nosync(struct page *page)
+{
+	might_sleep();
+	if (TestSetPageLocked(page))
+		__lock_page_nosync(page);
+}
 	
 /*
  * This is exported only for wait_on_page_locked/wait_on_page_writeback.
diff --git a/mm/filemap.c b/mm/filemap.c
index b9a60c43b61a..d5af1cab4268 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -488,6 +488,12 @@ struct page *page_cache_alloc_cold(struct address_space *x)
 EXPORT_SYMBOL(page_cache_alloc_cold);
 #endif
 
+static int __sleep_on_page_lock(void *word)
+{
+	io_schedule();
+	return 0;
+}
+
 /*
  * In order to wait for pages to become available there must be
  * waitqueues associated with pages. By using a hash table of
@@ -577,6 +583,17 @@ void fastcall __lock_page(struct page *page)
 }
 EXPORT_SYMBOL(__lock_page);
 
+/*
+ * Variant of lock_page that does not require the caller to hold a reference
+ * on the page's mapping.
+ */
+void fastcall __lock_page_nosync(struct page *page)
+{
+	DEFINE_WAIT_BIT(wait, &page->flags, PG_locked);
+	__wait_on_bit_lock(page_waitqueue(page), &wait, __sleep_on_page_lock,
+							TASK_UNINTERRUPTIBLE);
+}
+
 /**
  * find_get_page - find and get a page reference
  * @mapping: the address_space to search
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index b9f4c6f1be86..555752907dc3 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -701,7 +701,7 @@ int set_page_dirty_lock(struct page *page)
 {
 	int ret;
 
-	lock_page(page);
+	lock_page_nosync(page);
 	ret = set_page_dirty(page);
 	unlock_page(page);
 	return ret;
-- 
cgit v1.2.3


From dfd54cbcc0b834652389ce99b5e656ea5f44a3c1 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 25 Sep 2006 23:31:33 -0700
Subject: [PATCH] bootmem: use MAX_DMA_ADDRESS instead of LOW32LIMIT

Introduce ARCH_LOW_ADDRESS_LIMIT which can be set per architecture to
override the 4GB default limit used by the bootmem allocater within
__alloc_bootmem_low() and __alloc_bootmem_low_node().  E.g.  s390 needs a
2GB limit instead of 4GB.

Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-s390/processor.h |  2 ++
 mm/bootmem.c                 | 11 ++++++++---
 2 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index a3a4e5fd30d7..578c2209fa76 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -337,6 +337,8 @@ struct notifier_block;
 int register_idle_notifier(struct notifier_block *nb);
 int unregister_idle_notifier(struct notifier_block *nb);
 
+#define ARCH_LOW_ADDRESS_LIMIT	0x7fffffffUL
+
 #endif
 
 /*
diff --git a/mm/bootmem.c b/mm/bootmem.c
index f0f85fa713ca..d53112fcb404 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -15,6 +15,7 @@
 
 #include <asm/bug.h>
 #include <asm/io.h>
+#include <asm/processor.h>
 
 #include "internal.h"
 
@@ -453,7 +454,9 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 	return __alloc_bootmem(size, align, goal);
 }
 
-#define LOW32LIMIT 0xffffffff
+#ifndef ARCH_LOW_ADDRESS_LIMIT
+#define ARCH_LOW_ADDRESS_LIMIT	0xffffffffUL
+#endif
 
 void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 				  unsigned long goal)
@@ -462,7 +465,8 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 	void *ptr;
 
 	list_for_each_entry(bdata, &bdata_list, list) {
-		ptr = __alloc_bootmem_core(bdata, size, align, goal, LOW32LIMIT);
+		ptr = __alloc_bootmem_core(bdata, size, align, goal,
+						ARCH_LOW_ADDRESS_LIMIT);
 		if (ptr)
 			return ptr;
 	}
@@ -478,5 +482,6 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size,
 				       unsigned long align, unsigned long goal)
 {
-	return __alloc_bootmem_core(pgdat->bdata, size, align, goal, LOW32LIMIT);
+	return __alloc_bootmem_core(pgdat->bdata, size, align, goal,
+				    ARCH_LOW_ADDRESS_LIMIT);
 }
-- 
cgit v1.2.3


From da6052f7b33abe55fbfd7d2213815f58c00a88d4 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:35 -0700
Subject: [PATCH] update some mm/ comments

Let's try to keep mm/ comments more useful and up to date. This is a start.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h         | 68 +++++++++++++++++++++++++++-------------------
 include/linux/page-flags.h | 35 ++++++++++++++----------
 mm/filemap.c               |  8 +++---
 3 files changed, 64 insertions(+), 47 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 2db4229a0066..f2018775b995 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -219,7 +219,8 @@ struct inode;
  * Each physical page in the system has a struct page associated with
  * it to keep track of whatever it is we are using the page for at the
  * moment. Note that we have no way to track which tasks are using
- * a page.
+ * a page, though if it is a pagecache page, rmap structures can tell us
+ * who is mapping it.
  */
 struct page {
 	unsigned long flags;		/* Atomic flags, some possibly
@@ -299,8 +300,7 @@ struct page {
  */
 
 /*
- * Drop a ref, return true if the logical refcount fell to zero (the page has
- * no users)
+ * Drop a ref, return true if the refcount fell to zero (the page has no users)
  */
 static inline int put_page_testzero(struct page *page)
 {
@@ -356,43 +356,55 @@ void split_page(struct page *page, unsigned int order);
  * For the non-reserved pages, page_count(page) denotes a reference count.
  *   page_count() == 0 means the page is free. page->lru is then used for
  *   freelist management in the buddy allocator.
- *   page_count() == 1 means the page is used for exactly one purpose
- *   (e.g. a private data page of one process).
+ *   page_count() > 0  means the page has been allocated.
  *
- * A page may be used for kmalloc() or anyone else who does a
- * __get_free_page(). In this case the page_count() is at least 1, and
- * all other fields are unused but should be 0 or NULL. The
- * management of this page is the responsibility of the one who uses
- * it.
+ * Pages are allocated by the slab allocator in order to provide memory
+ * to kmalloc and kmem_cache_alloc. In this case, the management of the
+ * page, and the fields in 'struct page' are the responsibility of mm/slab.c
+ * unless a particular usage is carefully commented. (the responsibility of
+ * freeing the kmalloc memory is the caller's, of course).
  *
- * The other pages (we may call them "process pages") are completely
+ * A page may be used by anyone else who does a __get_free_page().
+ * In this case, page_count still tracks the references, and should only
+ * be used through the normal accessor functions. The top bits of page->flags
+ * and page->virtual store page management information, but all other fields
+ * are unused and could be used privately, carefully. The management of this
+ * page is the responsibility of the one who allocated it, and those who have
+ * subsequently been given references to it.
+ *
+ * The other pages (we may call them "pagecache pages") are completely
  * managed by the Linux memory manager: I/O, buffers, swapping etc.
  * The following discussion applies only to them.
  *
- * A page may belong to an inode's memory mapping. In this case,
- * page->mapping is the pointer to the inode, and page->index is the
- * file offset of the page, in units of PAGE_CACHE_SIZE.
+ * A pagecache page contains an opaque `private' member, which belongs to the
+ * page's address_space. Usually, this is the address of a circular list of
+ * the page's disk buffers. PG_private must be set to tell the VM to call
+ * into the filesystem to release these pages.
  *
- * A page contains an opaque `private' member, which belongs to the
- * page's address_space.  Usually, this is the address of a circular
- * list of the page's disk buffers.
+ * A page may belong to an inode's memory mapping. In this case, page->mapping
+ * is the pointer to the inode, and page->index is the file offset of the page,
+ * in units of PAGE_CACHE_SIZE.
  *
- * For pages belonging to inodes, the page_count() is the number of
- * attaches, plus 1 if `private' contains something, plus one for
- * the page cache itself.
+ * If pagecache pages are not associated with an inode, they are said to be
+ * anonymous pages. These may become associated with the swapcache, and in that
+ * case PG_swapcache is set, and page->private is an offset into the swapcache.
  *
- * Instead of keeping dirty/clean pages in per address-space lists, we instead
- * now tag pages as dirty/under writeback in the radix tree.
+ * In either case (swapcache or inode backed), the pagecache itself holds one
+ * reference to the page. Setting PG_private should also increment the
+ * refcount. The each user mapping also has a reference to the page.
  *
- * There is also a per-mapping radix tree mapping index to the page
- * in memory if present. The tree is rooted at mapping->root.  
+ * The pagecache pages are stored in a per-mapping radix tree, which is
+ * rooted at mapping->page_tree, and indexed by offset.
+ * Where 2.4 and early 2.6 kernels kept dirty/clean pages in per-address_space
+ * lists, we instead now tag pages as dirty/writeback in the radix tree.
  *
- * All process pages can do I/O:
+ * All pagecache pages may be subject to I/O:
  * - inode pages may need to be read from disk,
  * - inode pages which have been modified and are MAP_SHARED may need
- *   to be written to disk,
- * - private pages which have been modified may need to be swapped out
- *   to swap space and (later) to be read back into memory.
+ *   to be written back to the inode on disk,
+ * - anonymous pages (including MAP_PRIVATE file mappings) which have been
+ *   modified may need to be swapped out to swap space and (later) to be read
+ *   back into memory.
  */
 
 /*
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 5748642e9f36..9d7921dd50f0 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -13,24 +13,25 @@
  * PG_reserved is set for special pages, which can never be swapped out. Some
  * of them might not even exist (eg empty_bad_page)...
  *
- * The PG_private bitflag is set if page->private contains a valid value.
+ * The PG_private bitflag is set on pagecache pages if they contain filesystem
+ * specific data (which is normally at page->private). It can be used by
+ * private allocations for its own usage.
  *
- * During disk I/O, PG_locked is used. This bit is set before I/O and
- * reset when I/O completes. page_waitqueue(page) is a wait queue of all tasks
- * waiting for the I/O on this page to complete.
+ * During initiation of disk I/O, PG_locked is set. This bit is set before I/O
+ * and cleared when writeback _starts_ or when read _completes_. PG_writeback
+ * is set before writeback starts and cleared when it finishes.
+ *
+ * PG_locked also pins a page in pagecache, and blocks truncation of the file
+ * while it is held.
+ *
+ * page_waitqueue(page) is a wait queue of all tasks waiting for the page
+ * to become unlocked.
  *
  * PG_uptodate tells whether the page's contents is valid.  When a read
  * completes, the page becomes uptodate, unless a disk I/O error happened.
  *
- * For choosing which pages to swap out, inode pages carry a PG_referenced bit,
- * which is set any time the system accesses that page through the (mapping,
- * index) hash table.  This referenced bit, together with the referenced bit
- * in the page tables, is used to manipulate page->age and move the page across
- * the active, inactive_dirty and inactive_clean lists.
- *
- * Note that the referenced bit, the page->lru list_head and the active,
- * inactive_dirty and inactive_clean lists are protected by the
- * zone->lru_lock, and *NOT* by the usual PG_locked bit!
+ * PG_referenced, PG_reclaim are used for page reclaim for anonymous and
+ * file-backed pagecache (see mm/vmscan.c).
  *
  * PG_error is set to indicate that an I/O error occurred on this page.
  *
@@ -42,6 +43,10 @@
  * space, they need to be kmapped separately for doing IO on the pages.  The
  * struct page (these bits with information) are always mapped into kernel
  * address space...
+ *
+ * PG_buddy is set to indicate that the page is free and in the buddy system
+ * (see mm/page_alloc.c).
+ *
  */
 
 /*
@@ -74,7 +79,7 @@
 #define PG_checked		 8	/* kill me in 2.5.<early>. */
 #define PG_arch_1		 9
 #define PG_reserved		10
-#define PG_private		11	/* Has something at ->private */
+#define PG_private		11	/* If pagecache, has fs-private data */
 
 #define PG_writeback		12	/* Page is under writeback */
 #define PG_nosave		13	/* Used for system suspend/resume */
@@ -83,7 +88,7 @@
 
 #define PG_mappedtodisk		16	/* Has blocks allocated on-disk */
 #define PG_reclaim		17	/* To be reclaimed asap */
-#define PG_nosave_free		18	/* Free, should not be written */
+#define PG_nosave_free		18	/* Used for system suspend/resume */
 #define PG_buddy		19	/* Page is free, on buddy lists */
 
 
diff --git a/mm/filemap.c b/mm/filemap.c
index d5af1cab4268..afcdc72b5e90 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -599,8 +599,8 @@ void fastcall __lock_page_nosync(struct page *page)
  * @mapping: the address_space to search
  * @offset: the page index
  *
- * A rather lightweight function, finding and getting a reference to a
- * hashed page atomically.
+ * Is there a pagecache struct page at the given (mapping, offset) tuple?
+ * If yes, increment its refcount and return it; if no, return NULL.
  */
 struct page * find_get_page(struct address_space *mapping, unsigned long offset)
 {
@@ -987,7 +987,7 @@ page_not_up_to_date:
 		/* Get exclusive access to the page ... */
 		lock_page(page);
 
-		/* Did it get unhashed before we got the lock? */
+		/* Did it get truncated before we got the lock? */
 		if (!page->mapping) {
 			unlock_page(page);
 			page_cache_release(page);
@@ -1627,7 +1627,7 @@ no_cached_page:
 page_not_uptodate:
 	lock_page(page);
 
-	/* Did it get unhashed while we waited for it? */
+	/* Did it get truncated while we waited for it? */
 	if (!page->mapping) {
 		unlock_page(page);
 		goto err;
-- 
cgit v1.2.3


From dbe5e69d2d6e591996ea2b817b887d03b60bb143 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 25 Sep 2006 23:31:36 -0700
Subject: [PATCH] slab: optimize kmalloc_node the same way as kmalloc

[akpm@osdl.org: export fix]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/slab.h | 25 ++++++++++++++++++++++++-
 mm/slab.c            |  4 ++--
 2 files changed, 26 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 193c03c547ec..2f6bef6a98c9 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -202,7 +202,30 @@ extern int slab_is_available(void);
 
 #ifdef CONFIG_NUMA
 extern void *kmem_cache_alloc_node(kmem_cache_t *, gfp_t flags, int node);
-extern void *kmalloc_node(size_t size, gfp_t flags, int node);
+extern void *__kmalloc_node(size_t size, gfp_t flags, int node);
+
+static inline void *kmalloc_node(size_t size, gfp_t flags, int node)
+{
+	if (__builtin_constant_p(size)) {
+		int i = 0;
+#define CACHE(x) \
+		if (size <= x) \
+			goto found; \
+		else \
+			i++;
+#include "kmalloc_sizes.h"
+#undef CACHE
+		{
+			extern void __you_cannot_kmalloc_that_much(void);
+			__you_cannot_kmalloc_that_much();
+		}
+found:
+		return kmem_cache_alloc_node((flags & GFP_DMA) ?
+			malloc_sizes[i].cs_dmacachep :
+			malloc_sizes[i].cs_cachep, flags, node);
+	}
+	return __kmalloc_node(size, flags, node);
+}
 #else
 static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, gfp_t flags, int node)
 {
diff --git a/mm/slab.c b/mm/slab.c
index 3ad2f64998fd..5e59ce7a46c8 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3348,7 +3348,7 @@ void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 }
 EXPORT_SYMBOL(kmem_cache_alloc_node);
 
-void *kmalloc_node(size_t size, gfp_t flags, int node)
+void *__kmalloc_node(size_t size, gfp_t flags, int node)
 {
 	struct kmem_cache *cachep;
 
@@ -3357,7 +3357,7 @@ void *kmalloc_node(size_t size, gfp_t flags, int node)
 		return NULL;
 	return kmem_cache_alloc_node(cachep, flags, node);
 }
-EXPORT_SYMBOL(kmalloc_node);
+EXPORT_SYMBOL(__kmalloc_node);
 #endif
 
 /**
-- 
cgit v1.2.3


From 9b819d204cf602eab1a53a9ec4b8d2ca51e02a1d Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:40 -0700
Subject: [PATCH] Add __GFP_THISNODE to avoid fallback to other nodes and
 ignore cpuset/memory policy restrictions

Add a new gfp flag __GFP_THISNODE to avoid fallback to other nodes.  This
flag is essential if a kernel component requires memory to be located on a
certain node.  It will be needed for alloc_pages_node() to force allocation
on the indicated node and for alloc_pages() to force allocation on the
current node.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h | 3 ++-
 kernel/cpuset.c     | 2 +-
 mm/mempolicy.c      | 2 +-
 mm/page_alloc.c     | 3 +++
 4 files changed, 7 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 63ab88a36f39..0eda5b6dedbd 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -45,6 +45,7 @@ struct vm_area_struct;
 #define __GFP_ZERO	((__force gfp_t)0x8000u)/* Return zeroed page on success */
 #define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */
 #define __GFP_HARDWALL   ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */
+#define __GFP_THISNODE	((__force gfp_t)0x40000u)/* No fallback, no policies */
 
 #define __GFP_BITS_SHIFT 20	/* Room for 20 __GFP_FOO bits */
 #define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
@@ -53,7 +54,7 @@ struct vm_area_struct;
 #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \
 			__GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \
 			__GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \
-			__GFP_NOMEMALLOC|__GFP_HARDWALL)
+			__GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE)
 
 /* This equals 0, but use constants in case they ever change */
 #define GFP_NOWAIT	(GFP_ATOMIC & ~__GFP_HIGH)
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 4ea6f0dc2fc5..76940361273e 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2316,7 +2316,7 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
 	const struct cpuset *cs;	/* current cpuset ancestors */
 	int allowed;			/* is allocation in zone z allowed? */
 
-	if (in_interrupt())
+	if (in_interrupt() || (gfp_mask & __GFP_THISNODE))
 		return 1;
 	node = z->zone_pgdat->node_id;
 	might_sleep_if(!(gfp_mask & __GFP_HARDWALL));
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index c3429a710ab1..8002e1faccda 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1290,7 +1290,7 @@ struct page *alloc_pages_current(gfp_t gfp, unsigned order)
 
 	if ((gfp & __GFP_WAIT) && !in_interrupt())
 		cpuset_update_task_memory_state();
-	if (!pol || in_interrupt())
+	if (!pol || in_interrupt() || (gfp & __GFP_THISNODE))
 		pol = &default_policy;
 	if (pol->policy == MPOL_INTERLEAVE)
 		return alloc_page_interleave(gfp, order, interleave_nodes(pol));
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 208a6b03aa78..ea498788af53 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -893,6 +893,9 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order,
 	 * See also cpuset_zone_allowed() comment in kernel/cpuset.c.
 	 */
 	do {
+		if (unlikely((gfp_mask & __GFP_THISNODE) &&
+			(*z)->zone_pgdat != zonelist->zones[0]->zone_pgdat))
+				break;
 		if ((alloc_flags & ALLOC_CPUSET) &&
 				!cpuset_zone_allowed(*z, gfp_mask))
 			continue;
-- 
cgit v1.2.3


From 980128f223fa3c75e3ebdde650c9f1bcabd4c0a2 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:46 -0700
Subject: [PATCH] Define easier to handle GFP_THISNODE

In many places we will need to use the same combination of flags.  Specify
a single GFP_THISNODE definition for ease of use in gfp.h.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/uncached.c | 3 +--
 include/linux/gfp.h         | 2 ++
 mm/migrate.c                | 4 +---
 3 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index f813caa80570..c58e933694d5 100644
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -98,8 +98,7 @@ static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid)
 
 	/* attempt to allocate a granule's worth of cached memory pages */
 
-	page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO |
-				 __GFP_THISNODE | __GFP_NORETRY | __GFP_NOWARN,
+	page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
 				IA64_GRANULE_SHIFT-PAGE_SHIFT);
 	if (!page) {
 		mutex_unlock(&uc_pool->add_chunk_mutex);
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 0eda5b6dedbd..8b34aabfe4c6 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -67,6 +67,8 @@ struct vm_area_struct;
 #define GFP_HIGHUSER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
 			 __GFP_HIGHMEM)
 
+#define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
+
 /* Flag - indicates that the buffer will be suitable for DMA.  Ignored on some
    platforms, used as appropriate on others */
 
diff --git a/mm/migrate.c b/mm/migrate.c
index 6196f45c5263..20a8c2687b1e 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -741,9 +741,7 @@ static struct page *new_page_node(struct page *p, unsigned long private,
 
 	*result = &pm->status;
 
-	return alloc_pages_node(pm->node,
-		GFP_HIGHUSER | __GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY,
-		0);
+	return alloc_pages_node(pm->node, GFP_HIGHUSER | GFP_THISNODE, 0);
 }
 
 /*
-- 
cgit v1.2.3


From 46a82b2d5591335277ed2930611f6acb4ce654ed Mon Sep 17 00:00:00 2001
From: Dave McCracken <dmccr@us.ibm.com>
Date: Mon, 25 Sep 2006 23:31:48 -0700
Subject: [PATCH] Standardize pxx_page macros

One of the changes necessary for shared page tables is to standardize the
pxx_page macros.  pte_page and pmd_page have always returned the struct
page associated with their entry, while pte_page_kernel and pmd_page_kernel
have returned the kernel virtual address.  pud_page and pgd_page, on the
other hand, return the kernel virtual address.

Shared page tables needs pud_page and pgd_page to return the actual page
structures.  There are very few actual users of these functions, so it is
simple to standardize their usage.

Since this is basic cleanup, I am submitting these changes as a standalone
patch.  Per Hugh Dickins' comments about it, I am also changing the
pxx_page_kernel macros to pxx_page_vaddr to clarify their meaning.

Signed-off-by: Dave McCracken <dmccr@us.ibm.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/arm/mm/ioremap.c               |  2 +-
 arch/sh/mm/cache-sh7705.c           |  2 +-
 arch/sparc/mm/srmmu.c               |  2 +-
 arch/sparc/mm/sun4c.c               |  2 +-
 arch/um/kernel/skas/mmu.c           |  2 +-
 arch/x86_64/mm/fault.c              |  6 +++---
 include/asm-alpha/mmzone.h          |  1 +
 include/asm-alpha/pgtable.h         |  9 +++++----
 include/asm-arm/pgtable.h           |  8 ++++----
 include/asm-arm26/pgtable.h         |  8 ++++----
 include/asm-cris/pgtable.h          |  4 ++--
 include/asm-frv/pgtable.h           |  8 ++++----
 include/asm-generic/4level-fixup.h  |  4 ++++
 include/asm-generic/pgtable-nopmd.h |  2 +-
 include/asm-generic/pgtable-nopud.h |  2 +-
 include/asm-i386/pgtable-3level.h   |  2 +-
 include/asm-i386/pgtable.h          |  4 ++--
 include/asm-ia64/pgtable.h          | 14 ++++++++------
 include/asm-m32r/pgtable-2level.h   |  6 +++++-
 include/asm-m32r/pgtable.h          |  4 ++--
 include/asm-m68k/motorola_pgtable.h |  1 +
 include/asm-mips/pgtable-32.h       |  4 ++--
 include/asm-mips/pgtable-64.h       | 10 ++++++----
 include/asm-mips/pgtable.h          |  2 +-
 include/asm-parisc/pgtable.h        |  9 +++++----
 include/asm-powerpc/pgtable-4k.h    |  5 +++--
 include/asm-powerpc/pgtable.h       | 11 ++++++-----
 include/asm-ppc/pgtable.h           |  8 ++++----
 include/asm-s390/pgtable.h          | 10 ++++++----
 include/asm-sh/pgtable-2level.h     |  5 ++++-
 include/asm-sh/pgtable.h            |  4 ++--
 include/asm-sh64/pgtable.h          |  6 ++++--
 include/asm-sparc/pgtable.h         |  4 ++--
 include/asm-sparc64/pgtable.h       |  5 +++--
 include/asm-um/pgtable-2level.h     |  2 +-
 include/asm-um/pgtable-3level.h     |  5 +++--
 include/asm-um/pgtable.h            |  4 ++--
 include/asm-x86_64/pgtable.h        | 16 ++++++++--------
 include/asm-xtensa/pgtable.h        |  4 ++--
 39 files changed, 117 insertions(+), 90 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c
index 88a999df0ab3..591fc3187c7f 100644
--- a/arch/arm/mm/ioremap.c
+++ b/arch/arm/mm/ioremap.c
@@ -177,7 +177,7 @@ static void unmap_area_sections(unsigned long virt, unsigned long size)
 			 * Free the page table, if there was one.
 			 */
 			if ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_TABLE)
-				pte_free_kernel(pmd_page_kernel(pmd));
+				pte_free_kernel(pmd_page_vaddr(pmd));
 		}
 
 		addr += PGDIR_SIZE;
diff --git a/arch/sh/mm/cache-sh7705.c b/arch/sh/mm/cache-sh7705.c
index ad8ed7d41e16..bf94eedb0a8e 100644
--- a/arch/sh/mm/cache-sh7705.c
+++ b/arch/sh/mm/cache-sh7705.c
@@ -30,7 +30,7 @@
 
 #define __pte_offset(address) \
 		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset(dir, address) ((pte_t *) pmd_page_kernel(*(dir)) + \
+#define pte_offset(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
 		__pte_offset(address))
 
 static inline void cache_wback_all(void)
diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
index 16e13f663ab0..b27a506309ee 100644
--- a/arch/sparc/mm/srmmu.c
+++ b/arch/sparc/mm/srmmu.c
@@ -2175,7 +2175,7 @@ void __init ld_mmu_srmmu(void)
 
 	BTFIXUPSET_CALL(pte_pfn, srmmu_pte_pfn, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(pmd_page, srmmu_pmd_page, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(pgd_page, srmmu_pgd_page, BTFIXUPCALL_NORM);
+	BTFIXUPSET_CALL(pgd_page_vaddr, srmmu_pgd_page, BTFIXUPCALL_NORM);
 
 	BTFIXUPSET_SETHI(none_mask, 0xF0000000);
 
diff --git a/arch/sparc/mm/sun4c.c b/arch/sparc/mm/sun4c.c
index 7fdddf3c7e16..436021ceb2e7 100644
--- a/arch/sparc/mm/sun4c.c
+++ b/arch/sparc/mm/sun4c.c
@@ -2280,5 +2280,5 @@ void __init ld_mmu_sun4c(void)
 
 	/* These should _never_ get called with two level tables. */
 	BTFIXUPSET_CALL(pgd_set, sun4c_pgd_set, BTFIXUPCALL_NOP);
-	BTFIXUPSET_CALL(pgd_page, sun4c_pgd_page, BTFIXUPCALL_RETO0);
+	BTFIXUPSET_CALL(pgd_page_vaddr, sun4c_pgd_page, BTFIXUPCALL_RETO0);
 }
diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c
index 624ca238d1fd..79c22707a637 100644
--- a/arch/um/kernel/skas/mmu.c
+++ b/arch/um/kernel/skas/mmu.c
@@ -55,7 +55,7 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc,
 	 * destroy_context_skas.
 	 */
 
-        mm->context.skas.last_page_table = pmd_page_kernel(*pmd);
+        mm->context.skas.last_page_table = pmd_page_vaddr(*pmd);
 #ifdef CONFIG_3_LEVEL_PGTABLES
         mm->context.skas.last_pmd = (unsigned long) __va(pud_val(*pud));
 #endif
diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index ac8ea66ccb94..4198798e1469 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -299,7 +299,7 @@ static int vmalloc_fault(unsigned long address)
 	if (pgd_none(*pgd))
 		set_pgd(pgd, *pgd_ref);
 	else
-		BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
+		BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
 
 	/* Below here mismatches are bugs because these lower tables
 	   are shared */
@@ -308,7 +308,7 @@ static int vmalloc_fault(unsigned long address)
 	pud_ref = pud_offset(pgd_ref, address);
 	if (pud_none(*pud_ref))
 		return -1;
-	if (pud_none(*pud) || pud_page(*pud) != pud_page(*pud_ref))
+	if (pud_none(*pud) || pud_page_vaddr(*pud) != pud_page_vaddr(*pud_ref))
 		BUG();
 	pmd = pmd_offset(pud, address);
 	pmd_ref = pmd_offset(pud_ref, address);
@@ -641,7 +641,7 @@ void vmalloc_sync_all(void)
 				if (pgd_none(*pgd))
 					set_pgd(pgd, *pgd_ref);
 				else
-					BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
+					BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
 			}
 			spin_unlock(&pgd_lock);
 			set_bit(pgd_index(address), insync);
diff --git a/include/asm-alpha/mmzone.h b/include/asm-alpha/mmzone.h
index 64d0ab98fcd8..8af56ce346ad 100644
--- a/include/asm-alpha/mmzone.h
+++ b/include/asm-alpha/mmzone.h
@@ -75,6 +75,7 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n)
 #define VALID_PAGE(page)	(((page) - mem_map) < max_mapnr)
 
 #define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> 32))
+#define pgd_page(pgd)		(pfn_to_page(pgd_val(pgd) >> 32))
 #define pte_pfn(pte)		(pte_val(pte) >> 32)
 
 #define mk_pte(page, pgprot)						     \
diff --git a/include/asm-alpha/pgtable.h b/include/asm-alpha/pgtable.h
index 93eaa58b7961..49ac9bee7ced 100644
--- a/include/asm-alpha/pgtable.h
+++ b/include/asm-alpha/pgtable.h
@@ -230,16 +230,17 @@ extern inline void pgd_set(pgd_t * pgdp, pmd_t * pmdp)
 
 
 extern inline unsigned long
-pmd_page_kernel(pmd_t pmd)
+pmd_page_vaddr(pmd_t pmd)
 {
 	return ((pmd_val(pmd) & _PFN_MASK) >> (32-PAGE_SHIFT)) + PAGE_OFFSET;
 }
 
 #ifndef CONFIG_DISCONTIGMEM
 #define pmd_page(pmd)	(mem_map + ((pmd_val(pmd) & _PFN_MASK) >> 32))
+#define pgd_page(pgd)	(mem_map + ((pgd_val(pgd) & _PFN_MASK) >> 32))
 #endif
 
-extern inline unsigned long pgd_page(pgd_t pgd)
+extern inline unsigned long pgd_page_vaddr(pgd_t pgd)
 { return PAGE_OFFSET + ((pgd_val(pgd) & _PFN_MASK) >> (32-PAGE_SHIFT)); }
 
 extern inline int pte_none(pte_t pte)		{ return !pte_val(pte); }
@@ -293,13 +294,13 @@ extern inline pte_t pte_mkyoung(pte_t pte)	{ pte_val(pte) |= __ACCESS_BITS; retu
 /* Find an entry in the second-level page table.. */
 extern inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 {
-	return (pmd_t *) pgd_page(*dir) + ((address >> PMD_SHIFT) & (PTRS_PER_PAGE - 1));
+	return (pmd_t *) pgd_page_vaddr(*dir) + ((address >> PMD_SHIFT) & (PTRS_PER_PAGE - 1));
 }
 
 /* Find an entry in the third-level page table.. */
 extern inline pte_t * pte_offset_kernel(pmd_t * dir, unsigned long address)
 {
-	return (pte_t *) pmd_page_kernel(*dir)
+	return (pte_t *) pmd_page_vaddr(*dir)
 		+ ((address >> PAGE_SHIFT) & (PTRS_PER_PAGE - 1));
 }
 
diff --git a/include/asm-arm/pgtable.h b/include/asm-arm/pgtable.h
index 8d3919c6458c..4d10d319fa34 100644
--- a/include/asm-arm/pgtable.h
+++ b/include/asm-arm/pgtable.h
@@ -224,9 +224,9 @@ extern struct page *empty_zero_page;
 #define pte_none(pte)		(!pte_val(pte))
 #define pte_clear(mm,addr,ptep)	set_pte_at((mm),(addr),(ptep), __pte(0))
 #define pte_page(pte)		(pfn_to_page(pte_pfn(pte)))
-#define pte_offset_kernel(dir,addr)	(pmd_page_kernel(*(dir)) + __pte_index(addr))
-#define pte_offset_map(dir,addr)	(pmd_page_kernel(*(dir)) + __pte_index(addr))
-#define pte_offset_map_nested(dir,addr)	(pmd_page_kernel(*(dir)) + __pte_index(addr))
+#define pte_offset_kernel(dir,addr)	(pmd_page_vaddr(*(dir)) + __pte_index(addr))
+#define pte_offset_map(dir,addr)	(pmd_page_vaddr(*(dir)) + __pte_index(addr))
+#define pte_offset_map_nested(dir,addr)	(pmd_page_vaddr(*(dir)) + __pte_index(addr))
 #define pte_unmap(pte)		do { } while (0)
 #define pte_unmap_nested(pte)	do { } while (0)
 
@@ -291,7 +291,7 @@ PTE_BIT_FUNC(mkyoung,   |= L_PTE_YOUNG);
 		clean_pmd_entry(pmdp);	\
 	} while (0)
 
-static inline pte_t *pmd_page_kernel(pmd_t pmd)
+static inline pte_t *pmd_page_vaddr(pmd_t pmd)
 {
 	unsigned long ptr;
 
diff --git a/include/asm-arm26/pgtable.h b/include/asm-arm26/pgtable.h
index 19ac9101a6bb..63a8881fae13 100644
--- a/include/asm-arm26/pgtable.h
+++ b/include/asm-arm26/pgtable.h
@@ -186,12 +186,12 @@ extern struct page *empty_zero_page;
  * return a pointer to memory (no special alignment)
  */
 #define pmd_page(pmd)  ((struct page *)(pmd_val((pmd)) & ~_PMD_PRESENT))
-#define pmd_page_kernel(pmd) ((pte_t *)(pmd_val((pmd)) & ~_PMD_PRESENT))
+#define pmd_page_vaddr(pmd) ((pte_t *)(pmd_val((pmd)) & ~_PMD_PRESENT))
 
-#define pte_offset_kernel(dir,addr)     (pmd_page_kernel(*(dir)) + __pte_index(addr))
+#define pte_offset_kernel(dir,addr)     (pmd_page_vaddr(*(dir)) + __pte_index(addr))
 
-#define pte_offset_map(dir,addr)        (pmd_page_kernel(*(dir)) + __pte_index(addr))
-#define pte_offset_map_nested(dir,addr) (pmd_page_kernel(*(dir)) + __pte_index(addr))
+#define pte_offset_map(dir,addr)        (pmd_page_vaddr(*(dir)) + __pte_index(addr))
+#define pte_offset_map_nested(dir,addr) (pmd_page_vaddr(*(dir)) + __pte_index(addr))
 #define pte_unmap(pte)                  do { } while (0)
 #define pte_unmap_nested(pte)           do { } while (0)
 
diff --git a/include/asm-cris/pgtable.h b/include/asm-cris/pgtable.h
index 5d76c1c0d6c9..c94a7107019c 100644
--- a/include/asm-cris/pgtable.h
+++ b/include/asm-cris/pgtable.h
@@ -253,7 +253,7 @@ static inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
 { pmd_val(*pmdp) = _PAGE_TABLE | (unsigned long) ptep; }
 
 #define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
-#define pmd_page_kernel(pmd)	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd)	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /* to find an entry in a page-table-directory. */
 #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
@@ -271,7 +271,7 @@ static inline pgd_t * pgd_offset(struct mm_struct * mm, unsigned long address)
 #define __pte_offset(address) \
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  __pte_offset(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  __pte_offset(address))
 #define pte_offset_map(dir, address) \
 	((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
 #define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h
index 7af7485e889e..2fb3c6f05e03 100644
--- a/include/asm-frv/pgtable.h
+++ b/include/asm-frv/pgtable.h
@@ -217,7 +217,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
 }
 
 #define pgd_page(pgd)				(pud_page((pud_t){ pgd }))
-#define pgd_page_kernel(pgd)			(pud_page_kernel((pud_t){ pgd }))
+#define pgd_page_vaddr(pgd)			(pud_page_vaddr((pud_t){ pgd }))
 
 /*
  * allocating and freeing a pud is trivial: the 1-entry pud is
@@ -246,7 +246,7 @@ static inline void pud_clear(pud_t *pud)	{ }
 #define set_pud(pudptr, pudval)			set_pmd((pmd_t *)(pudptr), (pmd_t) { pudval })
 
 #define pud_page(pud)				(pmd_page((pmd_t){ pud }))
-#define pud_page_kernel(pud)			(pmd_page_kernel((pmd_t){ pud }))
+#define pud_page_vaddr(pud)			(pmd_page_vaddr((pmd_t){ pud }))
 
 /*
  * (pmds are folded into pgds so this doesn't get actually called,
@@ -362,7 +362,7 @@ static inline pmd_t *pmd_offset(pud_t *dir, unsigned long address)
 #define	pmd_bad(x)	(pmd_val(x) & xAMPRx_SS)
 #define pmd_clear(xp)	do { __set_pmd(xp, 0); } while(0)
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 #ifndef CONFIG_DISCONTIGMEM
@@ -458,7 +458,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
diff --git a/include/asm-generic/4level-fixup.h b/include/asm-generic/4level-fixup.h
index 68c6fea994d9..7b88d3931e34 100644
--- a/include/asm-generic/4level-fixup.h
+++ b/include/asm-generic/4level-fixup.h
@@ -21,6 +21,10 @@
 #define pud_present(pud)		1
 #define pud_ERROR(pud)			do { } while (0)
 #define pud_clear(pud)			pgd_clear(pud)
+#define pud_val(pud)			pgd_val(pud)
+#define pud_populate(mm, pud, pmd)	pgd_populate(mm, pud, pmd)
+#define pud_page(pud)			pgd_page(pud)
+#define pud_page_vaddr(pud)		pgd_page_vaddr(pud)
 
 #undef pud_free_tlb
 #define pud_free_tlb(tlb, x)            do { } while (0)
diff --git a/include/asm-generic/pgtable-nopmd.h b/include/asm-generic/pgtable-nopmd.h
index c8d53ba20e19..29ff5d84d8c3 100644
--- a/include/asm-generic/pgtable-nopmd.h
+++ b/include/asm-generic/pgtable-nopmd.h
@@ -47,7 +47,7 @@ static inline pmd_t * pmd_offset(pud_t * pud, unsigned long address)
 #define __pmd(x)				((pmd_t) { __pud(x) } )
 
 #define pud_page(pud)				(pmd_page((pmd_t){ pud }))
-#define pud_page_kernel(pud)			(pmd_page_kernel((pmd_t){ pud }))
+#define pud_page_vaddr(pud)			(pmd_page_vaddr((pmd_t){ pud }))
 
 /*
  * allocating and freeing a pmd is trivial: the 1-entry pmd is
diff --git a/include/asm-generic/pgtable-nopud.h b/include/asm-generic/pgtable-nopud.h
index 82e29f0ce467..566464500558 100644
--- a/include/asm-generic/pgtable-nopud.h
+++ b/include/asm-generic/pgtable-nopud.h
@@ -44,7 +44,7 @@ static inline pud_t * pud_offset(pgd_t * pgd, unsigned long address)
 #define __pud(x)				((pud_t) { __pgd(x) } )
 
 #define pgd_page(pgd)				(pud_page((pud_t){ pgd }))
-#define pgd_page_kernel(pgd)			(pud_page_kernel((pud_t){ pgd }))
+#define pgd_page_vaddr(pgd)			(pud_page_vaddr((pud_t){ pgd }))
 
 /*
  * allocating and freeing a pud is trivial: the 1-entry pud is
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index dccb1b3337ad..807ed9e366de 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -77,7 +77,7 @@ static inline void pud_clear (pud_t * pud) { }
 #define pud_page(pud) \
 ((struct page *) __va(pud_val(pud) & PAGE_MASK))
 
-#define pud_page_kernel(pud) \
+#define pud_page_vaddr(pud) \
 ((unsigned long) __va(pud_val(pud) & PAGE_MASK))
 
 
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 09697fec3d2b..c04b3d0f484e 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -364,11 +364,11 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 
 #define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 		((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h
index 228981cadf8f..553182747722 100644
--- a/include/asm-ia64/pgtable.h
+++ b/include/asm-ia64/pgtable.h
@@ -275,21 +275,23 @@ ia64_phys_addr_valid (unsigned long addr)
 #define pmd_bad(pmd)			(!ia64_phys_addr_valid(pmd_val(pmd)))
 #define pmd_present(pmd)		(pmd_val(pmd) != 0UL)
 #define pmd_clear(pmdp)			(pmd_val(*(pmdp)) = 0UL)
-#define pmd_page_kernel(pmd)		((unsigned long) __va(pmd_val(pmd) & _PFN_MASK))
+#define pmd_page_vaddr(pmd)		((unsigned long) __va(pmd_val(pmd) & _PFN_MASK))
 #define pmd_page(pmd)			virt_to_page((pmd_val(pmd) + PAGE_OFFSET))
 
 #define pud_none(pud)			(!pud_val(pud))
 #define pud_bad(pud)			(!ia64_phys_addr_valid(pud_val(pud)))
 #define pud_present(pud)		(pud_val(pud) != 0UL)
 #define pud_clear(pudp)			(pud_val(*(pudp)) = 0UL)
-#define pud_page(pud)			((unsigned long) __va(pud_val(pud) & _PFN_MASK))
+#define pud_page_vaddr(pud)		((unsigned long) __va(pud_val(pud) & _PFN_MASK))
+#define pud_page(pud)			virt_to_page((pud_val(pud) + PAGE_OFFSET))
 
 #ifdef CONFIG_PGTABLE_4
 #define pgd_none(pgd)			(!pgd_val(pgd))
 #define pgd_bad(pgd)			(!ia64_phys_addr_valid(pgd_val(pgd)))
 #define pgd_present(pgd)		(pgd_val(pgd) != 0UL)
 #define pgd_clear(pgdp)			(pgd_val(*(pgdp)) = 0UL)
-#define pgd_page(pgd)			((unsigned long) __va(pgd_val(pgd) & _PFN_MASK))
+#define pgd_page_vaddr(pgd)		((unsigned long) __va(pgd_val(pgd) & _PFN_MASK))
+#define pgd_page(pgd)			virt_to_page((pgd_val(pgd) + PAGE_OFFSET))
 #endif
 
 /*
@@ -360,19 +362,19 @@ pgd_offset (struct mm_struct *mm, unsigned long address)
 #ifdef CONFIG_PGTABLE_4
 /* Find an entry in the second-level page table.. */
 #define pud_offset(dir,addr) \
-	((pud_t *) pgd_page(*(dir)) + (((addr) >> PUD_SHIFT) & (PTRS_PER_PUD - 1)))
+	((pud_t *) pgd_page_vaddr(*(dir)) + (((addr) >> PUD_SHIFT) & (PTRS_PER_PUD - 1)))
 #endif
 
 /* Find an entry in the third-level page table.. */
 #define pmd_offset(dir,addr) \
-	((pmd_t *) pud_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+	((pmd_t *) pud_page_vaddr(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
 
 /*
  * Find an entry in the third-level page table.  This looks more complicated than it
  * should be because some platforms place page tables in high memory.
  */
 #define pte_index(addr)	 	(((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset_kernel(dir,addr)	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(addr))
+#define pte_offset_kernel(dir,addr)	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(addr))
 #define pte_offset_map(dir,addr)	pte_offset_kernel(dir, addr)
 #define pte_offset_map_nested(dir,addr)	pte_offset_map(dir, addr)
 #define pte_unmap(pte)			do { } while (0)
diff --git a/include/asm-m32r/pgtable-2level.h b/include/asm-m32r/pgtable-2level.h
index be0f167e344a..6a674e3d37a2 100644
--- a/include/asm-m32r/pgtable-2level.h
+++ b/include/asm-m32r/pgtable-2level.h
@@ -52,9 +52,13 @@ static inline int pgd_present(pgd_t pgd)	{ return 1; }
 #define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
 #define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)
 
-#define pgd_page(pgd) \
+#define pgd_page_vaddr(pgd) \
 ((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
 
+#ifndef CONFIG_DISCONTIGMEM
+#define pgd_page(pgd)	(mem_map + ((pgd_val(pgd) >> PAGE_SHIFT) - PFN_BASE))
+#endif /* !CONFIG_DISCONTIGMEM */
+
 static inline pmd_t *pmd_offset(pgd_t * dir, unsigned long address)
 {
 	return (pmd_t *) dir;
diff --git a/include/asm-m32r/pgtable.h b/include/asm-m32r/pgtable.h
index 1983b7f4527a..1c15ba7ce319 100644
--- a/include/asm-m32r/pgtable.h
+++ b/include/asm-m32r/pgtable.h
@@ -336,7 +336,7 @@ static inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
 	pmd_val(*pmdp) = (((unsigned long) ptep) & PAGE_MASK);
 }
 
-#define pmd_page_kernel(pmd)	\
+#define pmd_page_vaddr(pmd)	\
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 #ifndef CONFIG_DISCONTIGMEM
@@ -358,7 +358,7 @@ static inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
 #define pte_index(address)	\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address)	\
-	((pte_t *)pmd_page_kernel(*(dir)) + pte_index(address))
+	((pte_t *)pmd_page_vaddr(*(dir)) + pte_index(address))
 #define pte_offset_map(dir, address)	\
 	((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
 #define pte_offset_map_nested(dir, address)	pte_offset_map(dir, address)
diff --git a/include/asm-m68k/motorola_pgtable.h b/include/asm-m68k/motorola_pgtable.h
index 1ccc7338a54b..61e4406ed96a 100644
--- a/include/asm-m68k/motorola_pgtable.h
+++ b/include/asm-m68k/motorola_pgtable.h
@@ -150,6 +150,7 @@ static inline void pgd_set(pgd_t *pgdp, pmd_t *pmdp)
 #define pgd_bad(pgd)		((pgd_val(pgd) & _DESCTYPE_MASK) != _PAGE_TABLE)
 #define pgd_present(pgd)	(pgd_val(pgd) & _PAGE_TABLE)
 #define pgd_clear(pgdp)		({ pgd_val(*pgdp) = 0; })
+#define pgd_page(pgd)		(mem_map + ((unsigned long)(__va(pgd_val(pgd)) - PAGE_OFFSET) >> PAGE_SHIFT))
 
 #define pte_ERROR(e) \
 	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
diff --git a/include/asm-mips/pgtable-32.h b/include/asm-mips/pgtable-32.h
index 4b26d8528133..d20f2e9b28be 100644
--- a/include/asm-mips/pgtable-32.h
+++ b/include/asm-mips/pgtable-32.h
@@ -156,9 +156,9 @@ pfn_pte(unsigned long pfn, pgprot_t prot)
 #define __pte_offset(address)						\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset(dir, address)					\
-	((pte_t *) (pmd_page_kernel(*dir)) + __pte_offset(address))
+	((pte_t *) (pmd_page_vaddr(*dir)) + __pte_offset(address))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  __pte_offset(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  __pte_offset(address))
 
 #define pte_offset_map(dir, address)                                    \
 	((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
diff --git a/include/asm-mips/pgtable-64.h b/include/asm-mips/pgtable-64.h
index e3db93212eab..c59a1e21f5b0 100644
--- a/include/asm-mips/pgtable-64.h
+++ b/include/asm-mips/pgtable-64.h
@@ -178,24 +178,26 @@ static inline void pud_clear(pud_t *pudp)
 /* to find an entry in a page-table-directory */
 #define pgd_offset(mm,addr)	((mm)->pgd + pgd_index(addr))
 
-static inline unsigned long pud_page(pud_t pud)
+static inline unsigned long pud_page_vaddr(pud_t pud)
 {
 	return pud_val(pud);
 }
+#define pud_phys(pud)		(pud_val(pud) - PAGE_OFFSET)
+#define pud_page(pud)		(pfn_to_page(pud_phys(pud) >> PAGE_SHIFT))
 
 /* Find an entry in the second-level page table.. */
 static inline pmd_t *pmd_offset(pud_t * pud, unsigned long address)
 {
-	return (pmd_t *) pud_page(*pud) + pmd_index(address);
+	return (pmd_t *) pud_page_vaddr(*pud) + pmd_index(address);
 }
 
 /* Find an entry in the third-level page table.. */
 #define __pte_offset(address)						\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset(dir, address)					\
-	((pte_t *) (pmd_page_kernel(*dir)) + __pte_offset(address))
+	((pte_t *) (pmd_page_vaddr(*dir)) + __pte_offset(address))
 #define pte_offset_kernel(dir, address)					\
-	((pte_t *) pmd_page_kernel(*(dir)) +  __pte_offset(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  __pte_offset(address))
 #define pte_offset_map(dir, address)					\
 	((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
 #define pte_offset_map_nested(dir, address)				\
diff --git a/include/asm-mips/pgtable.h b/include/asm-mips/pgtable.h
index a36ca1be17f2..1ca4d1e185c7 100644
--- a/include/asm-mips/pgtable.h
+++ b/include/asm-mips/pgtable.h
@@ -87,7 +87,7 @@ extern void paging_init(void);
  */
 #define pmd_phys(pmd)		(pmd_val(pmd) - PAGE_OFFSET)
 #define pmd_page(pmd)		(pfn_to_page(pmd_phys(pmd) >> PAGE_SHIFT))
-#define pmd_page_kernel(pmd)	pmd_val(pmd)
+#define pmd_page_vaddr(pmd)	pmd_val(pmd)
 
 #if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32_R1)
 
diff --git a/include/asm-parisc/pgtable.h b/include/asm-parisc/pgtable.h
index 5066c54dae0a..c0b61e0d1497 100644
--- a/include/asm-parisc/pgtable.h
+++ b/include/asm-parisc/pgtable.h
@@ -303,7 +303,8 @@ static inline void pmd_clear(pmd_t *pmd) {
 
 
 #if PT_NLEVELS == 3
-#define pgd_page(pgd) ((unsigned long) __va(pgd_address(pgd)))
+#define pgd_page_vaddr(pgd) ((unsigned long) __va(pgd_address(pgd)))
+#define pgd_page(pgd)	virt_to_page((void *)pgd_page_vaddr(pgd))
 
 /* For 64 bit we have three level tables */
 
@@ -382,7 +383,7 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #define pte_page(pte)		(pfn_to_page(pte_pfn(pte)))
 
-#define pmd_page_kernel(pmd)	((unsigned long) __va(pmd_address(pmd)))
+#define pmd_page_vaddr(pmd)	((unsigned long) __va(pmd_address(pmd)))
 
 #define __pmd_page(pmd) ((unsigned long) __va(pmd_address(pmd)))
 #define pmd_page(pmd)	virt_to_page((void *)__pmd_page(pmd))
@@ -400,7 +401,7 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #if PT_NLEVELS == 3
 #define pmd_offset(dir,address) \
-((pmd_t *) pgd_page(*(dir)) + (((address)>>PMD_SHIFT) & (PTRS_PER_PMD-1)))
+((pmd_t *) pgd_page_vaddr(*(dir)) + (((address)>>PMD_SHIFT) & (PTRS_PER_PMD-1)))
 #else
 #define pmd_offset(dir,addr) ((pmd_t *) dir)
 #endif
@@ -408,7 +409,7 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 /* Find an entry in the third-level page table.. */ 
 #define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1))
 #define pte_offset_kernel(pmd, address) \
-	((pte_t *) pmd_page_kernel(*(pmd)) + pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(pmd)) + pte_index(address))
 #define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_offset_map_nested(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_unmap(pte) do { } while (0)
diff --git a/include/asm-powerpc/pgtable-4k.h b/include/asm-powerpc/pgtable-4k.h
index e7036155672e..345d9b07b3e2 100644
--- a/include/asm-powerpc/pgtable-4k.h
+++ b/include/asm-powerpc/pgtable-4k.h
@@ -88,10 +88,11 @@
 #define pgd_bad(pgd)		(pgd_val(pgd) == 0)
 #define pgd_present(pgd)	(pgd_val(pgd) != 0)
 #define pgd_clear(pgdp)		(pgd_val(*(pgdp)) = 0)
-#define pgd_page(pgd)		(pgd_val(pgd) & ~PGD_MASKED_BITS)
+#define pgd_page_vaddr(pgd)	(pgd_val(pgd) & ~PGD_MASKED_BITS)
+#define pgd_page(pgd)		virt_to_page(pgd_page_vaddr(pgd))
 
 #define pud_offset(pgdp, addr)	\
-  (((pud_t *) pgd_page(*(pgdp))) + \
+  (((pud_t *) pgd_page_vaddr(*(pgdp))) + \
     (((addr) >> PUD_SHIFT) & (PTRS_PER_PUD - 1)))
 
 #define pud_ERROR(e) \
diff --git a/include/asm-powerpc/pgtable.h b/include/asm-powerpc/pgtable.h
index 8dbf5ad8150f..10f52743f4ff 100644
--- a/include/asm-powerpc/pgtable.h
+++ b/include/asm-powerpc/pgtable.h
@@ -196,8 +196,8 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 				 || (pmd_val(pmd) & PMD_BAD_BITS))
 #define	pmd_present(pmd)	(pmd_val(pmd) != 0)
 #define	pmd_clear(pmdp)		(pmd_val(*(pmdp)) = 0)
-#define pmd_page_kernel(pmd)	(pmd_val(pmd) & ~PMD_MASKED_BITS)
-#define pmd_page(pmd)		virt_to_page(pmd_page_kernel(pmd))
+#define pmd_page_vaddr(pmd)	(pmd_val(pmd) & ~PMD_MASKED_BITS)
+#define pmd_page(pmd)		virt_to_page(pmd_page_vaddr(pmd))
 
 #define pud_set(pudp, pudval)	(pud_val(*(pudp)) = (pudval))
 #define pud_none(pud)		(!pud_val(pud))
@@ -205,7 +205,8 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 				 || (pud_val(pud) & PUD_BAD_BITS))
 #define pud_present(pud)	(pud_val(pud) != 0)
 #define pud_clear(pudp)		(pud_val(*(pudp)) = 0)
-#define pud_page(pud)		(pud_val(pud) & ~PUD_MASKED_BITS)
+#define pud_page_vaddr(pud)	(pud_val(pud) & ~PUD_MASKED_BITS)
+#define pud_page(pud)		virt_to_page(pud_page_vaddr(pud))
 
 #define pgd_set(pgdp, pudp)	({pgd_val(*(pgdp)) = (unsigned long)(pudp);})
 
@@ -219,10 +220,10 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 #define pgd_offset(mm, address)	 ((mm)->pgd + pgd_index(address))
 
 #define pmd_offset(pudp,addr) \
-  (((pmd_t *) pud_page(*(pudp))) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+  (((pmd_t *) pud_page_vaddr(*(pudp))) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
 
 #define pte_offset_kernel(dir,addr) \
-  (((pte_t *) pmd_page_kernel(*(dir))) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))
+  (((pte_t *) pmd_page_vaddr(*(dir))) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))
 
 #define pte_offset_map(dir,addr)	pte_offset_kernel((dir), (addr))
 #define pte_offset_map_nested(dir,addr)	pte_offset_kernel((dir), (addr))
diff --git a/include/asm-ppc/pgtable.h b/include/asm-ppc/pgtable.h
index 51fa7c662917..b1fdbf40dba2 100644
--- a/include/asm-ppc/pgtable.h
+++ b/include/asm-ppc/pgtable.h
@@ -526,7 +526,7 @@ static inline int pgd_bad(pgd_t pgd)		{ return 0; }
 static inline int pgd_present(pgd_t pgd)	{ return 1; }
 #define pgd_clear(xp)				do { } while (0)
 
-#define pgd_page(pgd) \
+#define pgd_page_vaddr(pgd) \
 	((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
 
 /*
@@ -720,12 +720,12 @@ extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
  * of the pte page.  -- paulus
  */
 #ifndef CONFIG_BOOKE
-#define pmd_page_kernel(pmd)	\
+#define pmd_page_vaddr(pmd)	\
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd)		\
 	(mem_map + (pmd_val(pmd) >> PAGE_SHIFT))
 #else
-#define pmd_page_kernel(pmd)	\
+#define pmd_page_vaddr(pmd)	\
 	((unsigned long) (pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd)		\
 	(mem_map + (__pa(pmd_val(pmd)) >> PAGE_SHIFT))
@@ -748,7 +748,7 @@ static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 #define pte_index(address)		\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, addr)	\
-	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(addr))
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(addr))
 #define pte_offset_map(dir, addr)		\
 	((pte_t *) kmap_atomic(pmd_page(*(dir)), KM_PTE0) + pte_index(addr))
 #define pte_offset_map_nested(dir, addr)	\
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index 1a07028d575e..e965309fedac 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -664,11 +664,13 @@ static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
 #define pte_pfn(x) (pte_val(x) >> PAGE_SHIFT)
 #define pte_page(x) pfn_to_page(pte_pfn(x))
 
-#define pmd_page_kernel(pmd) (pmd_val(pmd) & PAGE_MASK)
+#define pmd_page_vaddr(pmd) (pmd_val(pmd) & PAGE_MASK)
 
 #define pmd_page(pmd) (mem_map+(pmd_val(pmd) >> PAGE_SHIFT))
 
-#define pgd_page_kernel(pgd) (pgd_val(pgd) & PAGE_MASK)
+#define pgd_page_vaddr(pgd) (pgd_val(pgd) & PAGE_MASK)
+
+#define pgd_page(pgd) (mem_map+(pgd_val(pgd) >> PAGE_SHIFT))
 
 /* to find an entry in a page-table-directory */
 #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
@@ -690,14 +692,14 @@ static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 /* Find an entry in the second-level page table.. */
 #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
 #define pmd_offset(dir,addr) \
-	((pmd_t *) pgd_page_kernel(*(dir)) + pmd_index(addr))
+	((pmd_t *) pgd_page_vaddr(*(dir)) + pmd_index(addr))
 
 #endif /* __s390x__ */
 
 /* Find an entry in the third-level page table.. */
 #define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1))
 #define pte_offset_kernel(pmd, address) \
-	((pte_t *) pmd_page_kernel(*(pmd)) + pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(pmd)) + pte_index(address))
 #define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_offset_map_nested(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_unmap(pte) do { } while (0)
diff --git a/include/asm-sh/pgtable-2level.h b/include/asm-sh/pgtable-2level.h
index b0528aa3cb1f..b525db6f61c6 100644
--- a/include/asm-sh/pgtable-2level.h
+++ b/include/asm-sh/pgtable-2level.h
@@ -50,9 +50,12 @@ static inline void pgd_clear (pgd_t * pgdp) 	{ }
 #define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
 #define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)
 
-#define pgd_page(pgd) \
+#define pgd_page_vaddr(pgd) \
 ((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
 
+#define pgd_page(pgd) \
+	(phys_to_page(pgd_val(pgd)))
+
 static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 {
 	return (pmd_t *) dir;
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index dcd23a03683d..40d41a78041e 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -225,7 +225,7 @@ static inline pgprot_t pgprot_noncached(pgprot_t _prot)
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 { set_pte(&pte, __pte((pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot))); return pte; }
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 #define pmd_page(pmd) \
@@ -242,7 +242,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
 		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
 #define pte_offset_map(dir, address) pte_offset_kernel(dir, address)
 #define pte_offset_map_nested(dir, address) pte_offset_kernel(dir, address)
 #define pte_unmap(pte)		do { } while (0)
diff --git a/include/asm-sh64/pgtable.h b/include/asm-sh64/pgtable.h
index 54c7821893f5..6b97c4cb1d64 100644
--- a/include/asm-sh64/pgtable.h
+++ b/include/asm-sh64/pgtable.h
@@ -190,7 +190,9 @@ static inline int pgd_bad(pgd_t pgd)		{ return 0; }
 #endif
 
 
-#define pgd_page(pgd_entry)	((unsigned long) (pgd_val(pgd_entry) & PAGE_MASK))
+#define pgd_page_vaddr(pgd_entry)	((unsigned long) (pgd_val(pgd_entry) & PAGE_MASK))
+#define pgd_page(pgd)	(virt_to_page(pgd_val(pgd)))
+
 
 /*
  * PMD defines. Middle level.
@@ -219,7 +221,7 @@ static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 #define pmd_none(pmd_entry)	(pmd_val((pmd_entry)) == _PMD_EMPTY)
 #define pmd_bad(pmd_entry)	((pmd_val(pmd_entry) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
 
-#define pmd_page_kernel(pmd_entry) \
+#define pmd_page_vaddr(pmd_entry) \
 	((unsigned long) __va(pmd_val(pmd_entry) & PAGE_MASK))
 
 #define pmd_page(pmd) \
diff --git a/include/asm-sparc/pgtable.h b/include/asm-sparc/pgtable.h
index 226c6475c9a2..4f0a5ba0d6a0 100644
--- a/include/asm-sparc/pgtable.h
+++ b/include/asm-sparc/pgtable.h
@@ -143,10 +143,10 @@ extern unsigned long empty_zero_page;
 /*
  */
 BTFIXUPDEF_CALL_CONST(struct page *, pmd_page, pmd_t)
-BTFIXUPDEF_CALL_CONST(unsigned long, pgd_page, pgd_t)
+BTFIXUPDEF_CALL_CONST(unsigned long, pgd_page_vaddr, pgd_t)
 
 #define pmd_page(pmd) BTFIXUP_CALL(pmd_page)(pmd)
-#define pgd_page(pgd) BTFIXUP_CALL(pgd_page)(pgd)
+#define pgd_page_vaddr(pgd) BTFIXUP_CALL(pgd_page_vaddr)(pgd)
 
 BTFIXUPDEF_SETHI(none_mask)
 BTFIXUPDEF_CALL_CONST(int, pte_present, pte_t)
diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h
index ebfe395cfb87..b12be7a869f6 100644
--- a/include/asm-sparc64/pgtable.h
+++ b/include/asm-sparc64/pgtable.h
@@ -630,8 +630,9 @@ static inline unsigned long pte_present(pte_t pte)
 #define __pmd_page(pmd)		\
 	((unsigned long) __va((((unsigned long)pmd_val(pmd))<<11UL)))
 #define pmd_page(pmd) 			virt_to_page((void *)__pmd_page(pmd))
-#define pud_page(pud)		\
+#define pud_page_vaddr(pud)		\
 	((unsigned long) __va((((unsigned long)pud_val(pud))<<11UL)))
+#define pud_page(pud) 			virt_to_page((void *)pud_page_vaddr(pud))
 #define pmd_none(pmd)			(!pmd_val(pmd))
 #define pmd_bad(pmd)			(0)
 #define pmd_present(pmd)		(pmd_val(pmd) != 0U)
@@ -653,7 +654,7 @@ static inline unsigned long pte_present(pte_t pte)
 
 /* Find an entry in the second-level page table.. */
 #define pmd_offset(pudp, address)	\
-	((pmd_t *) pud_page(*(pudp)) + \
+	((pmd_t *) pud_page_vaddr(*(pudp)) + \
 	 (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)))
 
 /* Find an entry in the third-level page table.. */
diff --git a/include/asm-um/pgtable-2level.h b/include/asm-um/pgtable-2level.h
index ffe017f6b64b..6050e0eb257e 100644
--- a/include/asm-um/pgtable-2level.h
+++ b/include/asm-um/pgtable-2level.h
@@ -41,7 +41,7 @@ static inline void pgd_mkuptodate(pgd_t pgd)	{ }
 #define pfn_pte(pfn, prot) __pte(pfn_to_phys(pfn) | pgprot_val(prot))
 #define pfn_pmd(pfn, prot) __pmd(pfn_to_phys(pfn) | pgprot_val(prot))
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
diff --git a/include/asm-um/pgtable-3level.h b/include/asm-um/pgtable-3level.h
index 786c25727289..ca0c2a92a112 100644
--- a/include/asm-um/pgtable-3level.h
+++ b/include/asm-um/pgtable-3level.h
@@ -74,11 +74,12 @@ extern inline void pud_clear (pud_t *pud)
         set_pud(pud, __pud(0));
 }
 
-#define pud_page(pud) \
+#define pud_page(pud) phys_to_page(pud_val(pud) & PAGE_MASK)
+#define pud_page_vaddr(pud) \
 	((struct page *) __va(pud_val(pud) & PAGE_MASK))
 
 /* Find an entry in the second-level page table.. */
-#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \
+#define pmd_offset(pud, address) ((pmd_t *) pud_page_vaddr(*(pud)) + \
 			pmd_index(address))
 
 static inline unsigned long pte_pfn(pte_t pte)
diff --git a/include/asm-um/pgtable.h b/include/asm-um/pgtable.h
index ac64eb955868..4862daf8b906 100644
--- a/include/asm-um/pgtable.h
+++ b/include/asm-um/pgtable.h
@@ -349,7 +349,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 	return pte; 
 }
 
-#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
  * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
@@ -389,7 +389,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
  */
 #define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 #define pte_offset_map(dir, address) \
 	((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
 #define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index a31ab4e68a9b..51eba2395171 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -101,9 +101,6 @@ static inline void pgd_clear (pgd_t * pgd)
 	set_pgd(pgd, __pgd(0));
 }
 
-#define pud_page(pud) \
-((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
-
 #define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte, 0))
 
 struct mm_struct;
@@ -326,7 +323,8 @@ static inline int pmd_large(pmd_t pte) {
 /*
  * Level 4 access.
  */
-#define pgd_page(pgd) ((unsigned long) __va((unsigned long)pgd_val(pgd) & PTE_MASK))
+#define pgd_page_vaddr(pgd) ((unsigned long) __va((unsigned long)pgd_val(pgd) & PTE_MASK))
+#define pgd_page(pgd)		(pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT))
 #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
 #define pgd_offset(mm, addr) ((mm)->pgd + pgd_index(addr))
 #define pgd_offset_k(address) (init_level4_pgt + pgd_index(address))
@@ -335,16 +333,18 @@ static inline int pmd_large(pmd_t pte) {
 
 /* PUD - Level3 access */
 /* to find an entry in a page-table-directory. */
+#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
+#define pud_page(pud)		(pfn_to_page(pud_val(pud) >> PAGE_SHIFT))
 #define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))
-#define pud_offset(pgd, address) ((pud_t *) pgd_page(*(pgd)) + pud_index(address))
+#define pud_offset(pgd, address) ((pud_t *) pgd_page_vaddr(*(pgd)) + pud_index(address))
 #define pud_present(pud) (pud_val(pud) & _PAGE_PRESENT)
 
 /* PMD  - Level 2 access */
-#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
 #define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
 
 #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
-#define pmd_offset(dir, address) ((pmd_t *) pud_page(*(dir)) + \
+#define pmd_offset(dir, address) ((pmd_t *) pud_page_vaddr(*(dir)) + \
 			pmd_index(address))
 #define pmd_none(x)	(!pmd_val(x))
 #define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
@@ -382,7 +382,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_kernel(*(dir)) + \
+#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
 			pte_index(address))
 
 /* x86-64 always has all page tables mapped. */
diff --git a/include/asm-xtensa/pgtable.h b/include/asm-xtensa/pgtable.h
index 7b15afb70c56..a47cc734c20c 100644
--- a/include/asm-xtensa/pgtable.h
+++ b/include/asm-xtensa/pgtable.h
@@ -218,7 +218,7 @@ extern pgd_t swapper_pg_dir[PAGE_SIZE/sizeof(pgd_t)];
 /*
  * The pmd contains the kernel virtual address of the pte page.
  */
-#define pmd_page_kernel(pmd) ((unsigned long)(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long)(pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd) virt_to_page(pmd_val(pmd))
 
 /*
@@ -349,7 +349,7 @@ ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 /* Find an entry in the third-level page table.. */
 #define pte_index(address)	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir,addr) 					\
-	((pte_t*) pmd_page_kernel(*(dir)) + pte_index(addr))
+	((pte_t*) pmd_page_vaddr(*(dir)) + pte_index(addr))
 #define pte_offset_map(dir,addr)	pte_offset_kernel((dir),(addr))
 #define pte_offset_map_nested(dir,addr)	pte_offset_kernel((dir),(addr))
 
-- 
cgit v1.2.3


From 8417bba4b151346ed475fcc923693c9e3be89063 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:51 -0700
Subject: [PATCH] Replace min_unmapped_ratio by min_unmapped_pages in struct
 zone

*_pages is a better description of the role of the variable.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h | 2 +-
 mm/page_alloc.c        | 4 ++--
 mm/vmscan.c            | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 7fe317164b73..a703527e2b45 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -169,7 +169,7 @@ struct zone {
 	/*
 	 * zone reclaim becomes active if more unmapped pages exist.
 	 */
-	unsigned long		min_unmapped_ratio;
+	unsigned long		min_unmapped_pages;
 	struct per_cpu_pageset	*pageset[NR_CPUS];
 #else
 	struct per_cpu_pageset	pageset[NR_CPUS];
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f7ea020c23ea..5da6bc4e0a6b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2002,7 +2002,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		zone->spanned_pages = size;
 		zone->present_pages = realsize;
 #ifdef CONFIG_NUMA
-		zone->min_unmapped_ratio = (realsize*sysctl_min_unmapped_ratio)
+		zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio)
 						/ 100;
 #endif
 		zone->name = zone_names[j];
@@ -2313,7 +2313,7 @@ int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write,
 		return rc;
 
 	for_each_zone(zone)
-		zone->min_unmapped_ratio = (zone->present_pages *
+		zone->min_unmapped_pages = (zone->present_pages *
 				sysctl_min_unmapped_ratio) / 100;
 	return 0;
 }
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 8f35d7d585cb..5154c25e8440 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1618,7 +1618,7 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	 * unmapped file backed pages.
 	 */
 	if (zone_page_state(zone, NR_FILE_PAGES) -
-	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_ratio)
+	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages)
 		return 0;
 
 	/*
-- 
cgit v1.2.3


From 972d1a7b140569084439a81265a0f15b74e924e0 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:51 -0700
Subject: [PATCH] ZVC: Support NR_SLAB_RECLAIMABLE / NR_SLAB_UNRECLAIMABLE

Remove the atomic counter for slab_reclaim_pages and replace the counter
and NR_SLAB with two ZVC counter that account for unreclaimable and
reclaimable slab pages: NR_SLAB_RECLAIMABLE and NR_SLAB_UNRECLAIMABLE.

Change the check in vmscan.c to refer to to NR_SLAB_RECLAIMABLE.  The
intend seems to be to check for slab pages that could be freed.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/pgtable.c |  4 +++-
 drivers/base/node.c    |  9 +++++++--
 fs/proc/proc_misc.c    |  7 ++++++-
 include/linux/mmzone.h |  3 ++-
 include/linux/slab.h   |  2 --
 mm/mmap.c              |  2 +-
 mm/nommu.c             |  2 +-
 mm/page_alloc.c        |  3 ++-
 mm/slab.c              | 24 +++++++++++-------------
 mm/slob.c              |  4 ----
 mm/vmscan.c            |  2 +-
 mm/vmstat.c            |  3 ++-
 12 files changed, 36 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index bd98768d8764..a9f4910a22f8 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -60,7 +60,9 @@ void show_mem(void)
 	printk(KERN_INFO "%lu pages writeback\n",
 					global_page_state(NR_WRITEBACK));
 	printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_FILE_MAPPED));
-	printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
+	printk(KERN_INFO "%lu pages slab\n",
+		global_page_state(NR_SLAB_RECLAIMABLE) +
+		global_page_state(NR_SLAB_UNRECLAIMABLE));
 	printk(KERN_INFO "%lu pages pagetables\n",
 					global_page_state(NR_PAGETABLE));
 }
diff --git a/drivers/base/node.c b/drivers/base/node.c
index e09f5c2c11ee..001e6f6b9c1b 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -68,7 +68,9 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       "Node %d PageTables:   %8lu kB\n"
 		       "Node %d NFS_Unstable: %8lu kB\n"
 		       "Node %d Bounce:       %8lu kB\n"
-		       "Node %d Slab:         %8lu kB\n",
+		       "Node %d Slab:         %8lu kB\n"
+		       "Node %d SReclaimable: %8lu kB\n"
+		       "Node %d SUnreclaim:   %8lu kB\n",
 		       nid, K(i.totalram),
 		       nid, K(i.freeram),
 		       nid, K(i.totalram - i.freeram),
@@ -88,7 +90,10 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       nid, K(node_page_state(nid, NR_PAGETABLE)),
 		       nid, K(node_page_state(nid, NR_UNSTABLE_NFS)),
 		       nid, K(node_page_state(nid, NR_BOUNCE)),
-		       nid, K(node_page_state(nid, NR_SLAB)));
+		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE) +
+				node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE)));
 	n += hugetlb_report_node_meminfo(nid, buf + n);
 	return n;
 }
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index caa0a51560a0..5bbd60896050 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -170,6 +170,8 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		"AnonPages:    %8lu kB\n"
 		"Mapped:       %8lu kB\n"
 		"Slab:         %8lu kB\n"
+		"SReclaimable: %8lu kB\n"
+		"SUnreclaim:   %8lu kB\n"
 		"PageTables:   %8lu kB\n"
 		"NFS_Unstable: %8lu kB\n"
 		"Bounce:       %8lu kB\n"
@@ -197,7 +199,10 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		K(global_page_state(NR_WRITEBACK)),
 		K(global_page_state(NR_ANON_PAGES)),
 		K(global_page_state(NR_FILE_MAPPED)),
-		K(global_page_state(NR_SLAB)),
+		K(global_page_state(NR_SLAB_RECLAIMABLE) +
+				global_page_state(NR_SLAB_UNRECLAIMABLE)),
+		K(global_page_state(NR_SLAB_RECLAIMABLE)),
+		K(global_page_state(NR_SLAB_UNRECLAIMABLE)),
 		K(global_page_state(NR_PAGETABLE)),
 		K(global_page_state(NR_UNSTABLE_NFS)),
 		K(global_page_state(NR_BOUNCE)),
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index a703527e2b45..08c41b9f92e0 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -51,7 +51,8 @@ enum zone_stat_item {
 	NR_FILE_MAPPED,	/* pagecache pages mapped into pagetables.
 			   only modified from process context */
 	NR_FILE_PAGES,
-	NR_SLAB,	/* Pages used by slab allocator */
+	NR_SLAB_RECLAIMABLE,
+	NR_SLAB_UNRECLAIMABLE,
 	NR_PAGETABLE,	/* used for pagetables */
 	NR_FILE_DIRTY,
 	NR_WRITEBACK,
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 2f6bef6a98c9..66d6eb78d1c6 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -284,8 +284,6 @@ extern kmem_cache_t	*fs_cachep;
 extern kmem_cache_t	*sighand_cachep;
 extern kmem_cache_t	*bio_cachep;
 
-extern atomic_t slab_reclaim_pages;
-
 #endif	/* __KERNEL__ */
 
 #endif	/* _LINUX_SLAB_H */
diff --git a/mm/mmap.c b/mm/mmap.c
index 8507ee9cd573..eea8eefd51a8 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -116,7 +116,7 @@ int __vm_enough_memory(long pages, int cap_sys_admin)
 		 * which are reclaimable, under pressure.  The dentry
 		 * cache and most inode caches should fall into this
 		 */
-		free += atomic_read(&slab_reclaim_pages);
+		free += global_page_state(NR_SLAB_RECLAIMABLE);
 
 		/*
 		 * Leave the last 3% for root
diff --git a/mm/nommu.c b/mm/nommu.c
index c576df71e3bb..d99dea31e443 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1133,7 +1133,7 @@ int __vm_enough_memory(long pages, int cap_sys_admin)
 		 * which are reclaimable, under pressure.  The dentry
 		 * cache and most inode caches should fall into this
 		 */
-		free += atomic_read(&slab_reclaim_pages);
+		free += global_page_state(NR_SLAB_RECLAIMABLE);
 
 		/*
 		 * Leave the last 3% for root
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5da6bc4e0a6b..47e98423b30d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1304,7 +1304,8 @@ void show_free_areas(void)
 		global_page_state(NR_WRITEBACK),
 		global_page_state(NR_UNSTABLE_NFS),
 		nr_free_pages(),
-		global_page_state(NR_SLAB),
+		global_page_state(NR_SLAB_RECLAIMABLE) +
+			global_page_state(NR_SLAB_UNRECLAIMABLE),
 		global_page_state(NR_FILE_MAPPED),
 		global_page_state(NR_PAGETABLE));
 
diff --git a/mm/slab.c b/mm/slab.c
index 13b5050f84cc..7a48eb1a60c8 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -735,14 +735,6 @@ static inline void init_lock_keys(void)
 static DEFINE_MUTEX(cache_chain_mutex);
 static struct list_head cache_chain;
 
-/*
- * vm_enough_memory() looks at this to determine how many slab-allocated pages
- * are possibly freeable under pressure
- *
- * SLAB_RECLAIM_ACCOUNT turns this on per-slab
- */
-atomic_t slab_reclaim_pages;
-
 /*
  * chicken and egg problem: delay the per-cpu array allocation
  * until the general caches are up.
@@ -1580,8 +1572,11 @@ static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 
 	nr_pages = (1 << cachep->gfporder);
 	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
-		atomic_add(nr_pages, &slab_reclaim_pages);
-	add_zone_page_state(page_zone(page), NR_SLAB, nr_pages);
+		add_zone_page_state(page_zone(page),
+			NR_SLAB_RECLAIMABLE, nr_pages);
+	else
+		add_zone_page_state(page_zone(page),
+			NR_SLAB_UNRECLAIMABLE, nr_pages);
 	for (i = 0; i < nr_pages; i++)
 		__SetPageSlab(page + i);
 	return page_address(page);
@@ -1596,7 +1591,12 @@ static void kmem_freepages(struct kmem_cache *cachep, void *addr)
 	struct page *page = virt_to_page(addr);
 	const unsigned long nr_freed = i;
 
-	sub_zone_page_state(page_zone(page), NR_SLAB, nr_freed);
+	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
+		sub_zone_page_state(page_zone(page),
+				NR_SLAB_RECLAIMABLE, nr_freed);
+	else
+		sub_zone_page_state(page_zone(page),
+				NR_SLAB_UNRECLAIMABLE, nr_freed);
 	while (i--) {
 		BUG_ON(!PageSlab(page));
 		__ClearPageSlab(page);
@@ -1605,8 +1605,6 @@ static void kmem_freepages(struct kmem_cache *cachep, void *addr)
 	if (current->reclaim_state)
 		current->reclaim_state->reclaimed_slab += nr_freed;
 	free_pages((unsigned long)addr, cachep->gfporder);
-	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
-		atomic_sub(1 << cachep->gfporder, &slab_reclaim_pages);
 }
 
 static void kmem_rcu_free(struct rcu_head *head)
diff --git a/mm/slob.c b/mm/slob.c
index 4c28a421b270..20188627347c 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -339,7 +339,3 @@ void kmem_cache_init(void)
 
 	mod_timer(&slob_timer, jiffies + HZ);
 }
-
-atomic_t slab_reclaim_pages = ATOMIC_INIT(0);
-EXPORT_SYMBOL(slab_reclaim_pages);
-
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5154c25e8440..349797ba4bac 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1378,7 +1378,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages)
 	for_each_zone(zone)
 		lru_pages += zone->nr_active + zone->nr_inactive;
 
-	nr_slab = global_page_state(NR_SLAB);
+	nr_slab = global_page_state(NR_SLAB_RECLAIMABLE);
 	/* If slab caches are huge, it's better to hit them first */
 	while (nr_slab >= lru_pages) {
 		reclaim_state.reclaimed_slab = 0;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 968c0072e19a..490d8c1a0ded 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -458,7 +458,8 @@ static char *vmstat_text[] = {
 	"nr_anon_pages",
 	"nr_mapped",
 	"nr_file_pages",
-	"nr_slab",
+	"nr_slab_reclaimable",
+	"nr_slab_unreclaimable",
 	"nr_page_table_pages",
 	"nr_dirty",
 	"nr_writeback",
-- 
cgit v1.2.3


From 0ff38490c836dc379ff7ec45b10a15a662f4e5f6 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:52 -0700
Subject: [PATCH] zone_reclaim: dynamic slab reclaim

Currently one can enable slab reclaim by setting an explicit option in
/proc/sys/vm/zone_reclaim_mode.  Slab reclaim is then used as a final
option if the freeing of unmapped file backed pages is not enough to free
enough pages to allow a local allocation.

However, that means that the slab can grow excessively and that most memory
of a node may be used by slabs.  We have had a case where a machine with
46GB of memory was using 40-42GB for slab.  Zone reclaim was effective in
dealing with pagecache pages.  However, slab reclaim was only done during
global reclaim (which is a bit rare on NUMA systems).

This patch implements slab reclaim during zone reclaim.  Zone reclaim
occurs if there is a danger of an off node allocation.  At that point we

1. Shrink the per node page cache if the number of pagecache
   pages is more than min_unmapped_ratio percent of pages in a zone.

2. Shrink the slab cache if the number of the nodes reclaimable slab pages
   (patch depends on earlier one that implements that counter)
   are more than min_slab_ratio (a new /proc/sys/vm tunable).

The shrinking of the slab cache is a bit problematic since it is not node
specific.  So we simply calculate what point in the slab we want to reach
(current per node slab use minus the number of pages that neeed to be
allocated) and then repeately run the global reclaim until that is
unsuccessful or we have reached the limit.  I hope we will have zone based
slab reclaim at some point which will make that easier.

The default for the min_slab_ratio is 5%

Also remove the slab option from /proc/sys/vm/zone_reclaim_mode.

[akpm@osdl.org: cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/sysctl/vm.txt | 27 +++++++++++++++------
 include/linux/mmzone.h      |  3 +++
 include/linux/swap.h        |  1 +
 include/linux/sysctl.h      |  1 +
 kernel/sysctl.c             | 11 +++++++++
 mm/page_alloc.c             | 17 +++++++++++++
 mm/vmscan.c                 | 58 +++++++++++++++++++++++++++++----------------
 7 files changed, 90 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 7cee90223d3a..20d0d797f539 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -29,6 +29,7 @@ Currently, these files are in /proc/sys/vm:
 - drop-caches
 - zone_reclaim_mode
 - min_unmapped_ratio
+- min_slab_ratio
 - panic_on_oom
 
 ==============================================================
@@ -138,7 +139,6 @@ This is value ORed together of
 1	= Zone reclaim on
 2	= Zone reclaim writes dirty pages out
 4	= Zone reclaim swaps pages
-8	= Also do a global slab reclaim pass
 
 zone_reclaim_mode is set during bootup to 1 if it is determined that pages
 from remote zones will cause a measurable performance reduction. The
@@ -162,18 +162,13 @@ Allowing regular swap effectively restricts allocations to the local
 node unless explicitly overridden by memory policies or cpuset
 configurations.
 
-It may be advisable to allow slab reclaim if the system makes heavy
-use of files and builds up large slab caches. However, the slab
-shrink operation is global, may take a long time and free slabs
-in all nodes of the system.
-
 =============================================================
 
 min_unmapped_ratio:
 
 This is available only on NUMA kernels.
 
-A percentage of the file backed pages in each zone.  Zone reclaim will only
+A percentage of the total pages in each zone.  Zone reclaim will only
 occur if more than this percentage of pages are file backed and unmapped.
 This is to insure that a minimal amount of local pages is still available for
 file I/O even if the node is overallocated.
@@ -182,6 +177,24 @@ The default is 1 percent.
 
 =============================================================
 
+min_slab_ratio:
+
+This is available only on NUMA kernels.
+
+A percentage of the total pages in each zone.  On Zone reclaim
+(fallback from the local zone occurs) slabs will be reclaimed if more
+than this percentage of pages in a zone are reclaimable slab pages.
+This insures that the slab growth stays under control even in NUMA
+systems that rarely perform global reclaim.
+
+The default is 5 percent.
+
+Note that slab reclaim is triggered in a per zone / node fashion.
+The process of reclaiming slab memory is currently not node specific
+and may not be fast.
+
+=============================================================
+
 panic_on_oom
 
 This enables or disables panic on out-of-memory feature.  If this is set to 1,
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 08c41b9f92e0..3693f1a52788 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -171,6 +171,7 @@ struct zone {
 	 * zone reclaim becomes active if more unmapped pages exist.
 	 */
 	unsigned long		min_unmapped_pages;
+	unsigned long		min_slab_pages;
 	struct per_cpu_pageset	*pageset[NR_CPUS];
 #else
 	struct per_cpu_pageset	pageset[NR_CPUS];
@@ -448,6 +449,8 @@ int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *, int, struct file
 					void __user *, size_t *, loff_t *);
 int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *, int,
 			struct file *, void __user *, size_t *, loff_t *);
+int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *, int,
+			struct file *, void __user *, size_t *, loff_t *);
 
 #include <linux/topology.h>
 /* Returns the number of the current Node. */
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 32db06c8ffe0..a2f5ad7c2d2e 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -193,6 +193,7 @@ extern long vm_total_pages;
 #ifdef CONFIG_NUMA
 extern int zone_reclaim_mode;
 extern int sysctl_min_unmapped_ratio;
+extern int sysctl_min_slab_ratio;
 extern int zone_reclaim(struct zone *, gfp_t, unsigned int);
 #else
 #define zone_reclaim_mode 0
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 736ed917a4f8..eca555781d05 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -191,6 +191,7 @@ enum
 	VM_MIN_UNMAPPED=32,	/* Set min percent of unmapped pages */
 	VM_PANIC_ON_OOM=33,	/* panic at out-of-memory */
 	VM_VDSO_ENABLED=34,	/* map VDSO into new processes? */
+	VM_MIN_SLAB=35,		 /* Percent pages ignored by zone reclaim */
 };
 
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 362a0cc37138..fd43c3e6786b 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -943,6 +943,17 @@ static ctl_table vm_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one_hundred,
 	},
+	{
+		.ctl_name	= VM_MIN_SLAB,
+		.procname	= "min_slab_ratio",
+		.data		= &sysctl_min_slab_ratio,
+		.maxlen		= sizeof(sysctl_min_slab_ratio),
+		.mode		= 0644,
+		.proc_handler	= &sysctl_min_slab_ratio_sysctl_handler,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+		.extra2		= &one_hundred,
+	},
 #endif
 #ifdef CONFIG_X86_32
 	{
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 47e98423b30d..cf913bdd433e 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2005,6 +2005,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 #ifdef CONFIG_NUMA
 		zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio)
 						/ 100;
+		zone->min_slab_pages = (realsize * sysctl_min_slab_ratio) / 100;
 #endif
 		zone->name = zone_names[j];
 		spin_lock_init(&zone->lock);
@@ -2318,6 +2319,22 @@ int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write,
 				sysctl_min_unmapped_ratio) / 100;
 	return 0;
 }
+
+int sysctl_min_slab_ratio_sysctl_handler(ctl_table *table, int write,
+	struct file *file, void __user *buffer, size_t *length, loff_t *ppos)
+{
+	struct zone *zone;
+	int rc;
+
+	rc = proc_dointvec_minmax(table, write, file, buffer, length, ppos);
+	if (rc)
+		return rc;
+
+	for_each_zone(zone)
+		zone->min_slab_pages = (zone->present_pages *
+				sysctl_min_slab_ratio) / 100;
+	return 0;
+}
 #endif
 
 /*
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 349797ba4bac..089e943c4d38 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1527,7 +1527,6 @@ int zone_reclaim_mode __read_mostly;
 #define RECLAIM_ZONE (1<<0)	/* Run shrink_cache on the zone */
 #define RECLAIM_WRITE (1<<1)	/* Writeout pages during reclaim */
 #define RECLAIM_SWAP (1<<2)	/* Swap pages out during reclaim */
-#define RECLAIM_SLAB (1<<3)	/* Do a global slab shrink if the zone is out of memory */
 
 /*
  * Priority for ZONE_RECLAIM. This determines the fraction of pages
@@ -1542,6 +1541,12 @@ int zone_reclaim_mode __read_mostly;
  */
 int sysctl_min_unmapped_ratio = 1;
 
+/*
+ * If the number of slab pages in a zone grows beyond this percentage then
+ * slab reclaim needs to occur.
+ */
+int sysctl_min_slab_ratio = 5;
+
 /*
  * Try to free up some pages from this zone through reclaim.
  */
@@ -1573,29 +1578,37 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	reclaim_state.reclaimed_slab = 0;
 	p->reclaim_state = &reclaim_state;
 
-	/*
-	 * Free memory by calling shrink zone with increasing priorities
-	 * until we have enough memory freed.
-	 */
-	priority = ZONE_RECLAIM_PRIORITY;
-	do {
-		nr_reclaimed += shrink_zone(priority, zone, &sc);
-		priority--;
-	} while (priority >= 0 && nr_reclaimed < nr_pages);
+	if (zone_page_state(zone, NR_FILE_PAGES) -
+		zone_page_state(zone, NR_FILE_MAPPED) >
+		zone->min_unmapped_pages) {
+		/*
+		 * Free memory by calling shrink zone with increasing
+		 * priorities until we have enough memory freed.
+		 */
+		priority = ZONE_RECLAIM_PRIORITY;
+		do {
+			nr_reclaimed += shrink_zone(priority, zone, &sc);
+			priority--;
+		} while (priority >= 0 && nr_reclaimed < nr_pages);
+	}
 
-	if (nr_reclaimed < nr_pages && (zone_reclaim_mode & RECLAIM_SLAB)) {
+	if (zone_page_state(zone, NR_SLAB_RECLAIMABLE) > zone->min_slab_pages) {
 		/*
 		 * shrink_slab() does not currently allow us to determine how
-		 * many pages were freed in this zone. So we just shake the slab
-		 * a bit and then go off node for this particular allocation
-		 * despite possibly having freed enough memory to allocate in
-		 * this zone.  If we freed local memory then the next
-		 * allocations will be local again.
+		 * many pages were freed in this zone. So we take the current
+		 * number of slab pages and shake the slab until it is reduced
+		 * by the same nr_pages that we used for reclaiming unmapped
+		 * pages.
 		 *
-		 * shrink_slab will free memory on all zones and may take
-		 * a long time.
+		 * Note that shrink_slab will free memory on all zones and may
+		 * take a long time.
 		 */
-		shrink_slab(sc.nr_scanned, gfp_mask, order);
+		unsigned long limit = zone_page_state(zone,
+				NR_SLAB_RECLAIMABLE) - nr_pages;
+
+		while (shrink_slab(sc.nr_scanned, gfp_mask, order) &&
+			zone_page_state(zone, NR_SLAB_RECLAIMABLE) > limit)
+			;
 	}
 
 	p->reclaim_state = NULL;
@@ -1609,7 +1622,8 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	int node_id;
 
 	/*
-	 * Zone reclaim reclaims unmapped file backed pages.
+	 * Zone reclaim reclaims unmapped file backed pages and
+	 * slab pages if we are over the defined limits.
 	 *
 	 * A small portion of unmapped file backed pages is needed for
 	 * file I/O otherwise pages read by file I/O will be immediately
@@ -1618,7 +1632,9 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	 * unmapped file backed pages.
 	 */
 	if (zone_page_state(zone, NR_FILE_PAGES) -
-	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages)
+	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages
+	    && zone_page_state(zone, NR_SLAB_RECLAIMABLE)
+			<= zone->min_slab_pages)
 		return 0;
 
 	/*
-- 
cgit v1.2.3


From 89fa30242facca249aead2aac03c4c69764f911c Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:55 -0700
Subject: [PATCH] NUMA: Add zone_to_nid function

There are many places where we need to determine the node of a zone.
Currently we use a difficult to read sequence of pointer dereferencing.
Put that into an inline function and use throughout VM.  Maybe we can find
a way to optimize the lookup in the future.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/discontig.c | 2 +-
 arch/parisc/mm/init.c    | 2 +-
 include/linux/mm.h       | 7 ++++++-
 kernel/cpuset.c          | 4 ++--
 mm/hugetlb.c             | 2 +-
 mm/mempolicy.c           | 6 +++---
 mm/oom_kill.c            | 3 +--
 mm/page_alloc.c          | 2 +-
 mm/vmscan.c              | 2 +-
 9 files changed, 17 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 07c300f93764..fb5d8b747de4 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -422,7 +422,7 @@ void __init set_highmem_pages_init(int bad_ppro)
 		zone_end_pfn = zone_start_pfn + zone->spanned_pages;
 
 		printk("Initializing %s for node %d (%08lx:%08lx)\n",
-				zone->name, zone->zone_pgdat->node_id,
+				zone->name, zone_to_nid(zone),
 				zone_start_pfn, zone_end_pfn);
 
 		for (node_pfn = zone_start_pfn; node_pfn < zone_end_pfn; node_pfn++) {
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index c7329615ef94..25ad28d63e88 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -551,7 +551,7 @@ void show_mem(void)
 
 				printk("Zone list for zone %d on node %d: ", j, i);
 				for (k = 0; zl->zones[k] != NULL; k++) 
-					printk("[%d/%s] ", zl->zones[k]->zone_pgdat->node_id, zl->zones[k]->name);
+					printk("[%d/%s] ", zone_to_nid(zl->zones[k]), zl->zones[k]->name);
 				printk("\n");
 			}
 		}
diff --git a/include/linux/mm.h b/include/linux/mm.h
index f2018775b995..856f0ee7e84a 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -499,12 +499,17 @@ static inline struct zone *page_zone(struct page *page)
 	return zone_table[page_zone_id(page)];
 }
 
+static inline unsigned long zone_to_nid(struct zone *zone)
+{
+	return zone->zone_pgdat->node_id;
+}
+
 static inline unsigned long page_to_nid(struct page *page)
 {
 	if (FLAGS_HAS_NODE)
 		return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
 	else
-		return page_zone(page)->zone_pgdat->node_id;
+		return zone_to_nid(page_zone(page));
 }
 static inline unsigned long page_to_section(struct page *page)
 {
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 76940361273e..cff41511269f 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2245,7 +2245,7 @@ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl)
 	int i;
 
 	for (i = 0; zl->zones[i]; i++) {
-		int nid = zl->zones[i]->zone_pgdat->node_id;
+		int nid = zone_to_nid(zl->zones[i]);
 
 		if (node_isset(nid, current->mems_allowed))
 			return 1;
@@ -2318,7 +2318,7 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
 
 	if (in_interrupt() || (gfp_mask & __GFP_THISNODE))
 		return 1;
-	node = z->zone_pgdat->node_id;
+	node = zone_to_nid(z);
 	might_sleep_if(!(gfp_mask & __GFP_HARDWALL));
 	if (node_isset(node, current->mems_allowed))
 		return 1;
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 3aceadce1a76..7c7d03dbf73d 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -72,7 +72,7 @@ static struct page *dequeue_huge_page(struct vm_area_struct *vma,
 	struct zone **z;
 
 	for (z = zonelist->zones; *z; z++) {
-		nid = (*z)->zone_pgdat->node_id;
+		nid = zone_to_nid(*z);
 		if (cpuset_zone_allowed(*z, GFP_HIGHUSER) &&
 		    !list_empty(&hugepage_freelists[nid]))
 			break;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 8002e1faccda..38f89650bc84 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -487,7 +487,7 @@ static void get_zonemask(struct mempolicy *p, nodemask_t *nodes)
 	switch (p->policy) {
 	case MPOL_BIND:
 		for (i = 0; p->v.zonelist->zones[i]; i++)
-			node_set(p->v.zonelist->zones[i]->zone_pgdat->node_id,
+			node_set(zone_to_nid(p->v.zonelist->zones[i]),
 				*nodes);
 		break;
 	case MPOL_DEFAULT:
@@ -1145,7 +1145,7 @@ unsigned slab_node(struct mempolicy *policy)
 		 * Follow bind policy behavior and start allocation at the
 		 * first node.
 		 */
-		return policy->v.zonelist->zones[0]->zone_pgdat->node_id;
+		return zone_to_nid(policy->v.zonelist->zones[0]);
 
 	case MPOL_PREFERRED:
 		if (policy->v.preferred_node >= 0)
@@ -1649,7 +1649,7 @@ void mpol_rebind_policy(struct mempolicy *pol, const nodemask_t *newmask)
 
 		nodes_clear(nodes);
 		for (z = pol->v.zonelist->zones; *z; z++)
-			node_set((*z)->zone_pgdat->node_id, nodes);
+			node_set(zone_to_nid(*z), nodes);
 		nodes_remap(tmp, nodes, *mpolmask, *newmask);
 		nodes = tmp;
 
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index f1c0ef1fd21f..bada3d03119f 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -177,8 +177,7 @@ static inline int constrained_alloc(struct zonelist *zonelist, gfp_t gfp_mask)
 
 	for (z = zonelist->zones; *z; z++)
 		if (cpuset_zone_allowed(*z, gfp_mask))
-			node_clear((*z)->zone_pgdat->node_id,
-					nodes);
+			node_clear(zone_to_nid(*z), nodes);
 		else
 			return CONSTRAINT_CPUSET;
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cf913bdd433e..51070b6d593f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1217,7 +1217,7 @@ unsigned int nr_free_pagecache_pages(void)
 #ifdef CONFIG_NUMA
 static void show_node(struct zone *zone)
 {
-	printk("Node %d ", zone->zone_pgdat->node_id);
+	printk("Node %ld ", zone_to_nid(zone));
 }
 #else
 #define show_node(zone)	do { } while (0)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b950f193816e..87779dda4ec6 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1661,7 +1661,7 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	 * over remote processors and spread off node memory allocations
 	 * as wide as possible.
 	 */
-	node_id = zone->zone_pgdat->node_id;
+	node_id = zone_to_nid(zone);
 	mask = node_to_cpumask(node_id);
 	if (!cpus_empty(mask) && node_id != numa_node_id())
 		return 0;
-- 
cgit v1.2.3


From 62bac0185ad3dfef11d9602980445c54d45199c6 Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:56 -0700
Subject: [PATCH] selinux: eliminate selinux_task_ctxid

Eliminate selinux_task_ctxid since it duplicates selinux_task_get_sid.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/selinux.h    | 15 ---------------
 kernel/auditsc.c           |  2 +-
 security/selinux/exports.c |  9 ---------
 3 files changed, 1 insertion(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index aad4e390d6a5..79e4707ca772 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -69,16 +69,6 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
  */
 void selinux_audit_set_callback(int (*callback)(void));
 
-/**
- *	selinux_task_ctxid - determine a context ID for a process.
- *	@tsk: the task object
- *	@ctxid: ID value returned via this
- *
- *	On return, ctxid will contain an ID for the context.  This value
- *	should only be used opaquely.
- */
-void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid);
-
 /**
  *     selinux_ctxid_to_string - map a security context ID to a string
  *     @ctxid: security context ID to be converted.
@@ -166,11 +156,6 @@ static inline void selinux_audit_set_callback(int (*callback)(void))
 	return;
 }
 
-static inline void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid)
-{
-	*ctxid = 0;
-}
-
 static inline int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
 {
        *ctx = NULL;
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 1bd8827a0102..331e17010393 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -385,7 +385,7 @@ static int audit_filter_rules(struct task_struct *tsk,
 			   logged upon error */
 			if (f->se_rule) {
 				if (need_sid) {
-					selinux_task_ctxid(tsk, &sid);
+					selinux_get_task_sid(tsk, &sid);
 					need_sid = 0;
 				}
 				result = selinux_audit_rule_match(sid, f->type,
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index 9d7737db5e51..ee0fb47f81ae 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -21,15 +21,6 @@
 #include "security.h"
 #include "objsec.h"
 
-void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid)
-{
-	struct task_security_struct *tsec = tsk->security;
-	if (selinux_enabled)
-		*ctxid = tsec->sid;
-	else
-		*ctxid = 0;
-}
-
 int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
 {
 	if (selinux_enabled)
-- 
cgit v1.2.3


From 1a70cd40cb291c25b67ec0da715a49d76719329d Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:57 -0700
Subject: [PATCH] selinux: rename selinux_ctxid_to_string

Rename selinux_ctxid_to_string to selinux_sid_to_string to be
consistent with other interfaces.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/selinux.h    |  8 ++++----
 kernel/audit.c             | 14 +++++++-------
 kernel/auditfilter.c       |  2 +-
 kernel/auditsc.c           |  4 ++--
 security/selinux/exports.c |  4 ++--
 5 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include')

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 79e4707ca772..df9098de4c99 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -70,8 +70,8 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
 void selinux_audit_set_callback(int (*callback)(void));
 
 /**
- *     selinux_ctxid_to_string - map a security context ID to a string
- *     @ctxid: security context ID to be converted.
+ *     selinux_sid_to_string - map a security context ID to a string
+ *     @sid: security context ID to be converted.
  *     @ctx: address of context string to be returned
  *     @ctxlen: length of returned context string.
  *
@@ -79,7 +79,7 @@ void selinux_audit_set_callback(int (*callback)(void));
  *     string will be allocated internally, and the caller must call
  *     kfree() on it after use.
  */
-int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen);
+int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen);
 
 /**
  *     selinux_get_inode_sid - get the inode's security context ID
@@ -156,7 +156,7 @@ static inline void selinux_audit_set_callback(int (*callback)(void))
 	return;
 }
 
-static inline int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
+static inline int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
 {
        *ctx = NULL;
        *ctxlen = 0;
diff --git a/kernel/audit.c b/kernel/audit.c
index 963fd15c9621..f9889ee77825 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -244,7 +244,7 @@ static int audit_set_rate_limit(int limit, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -267,7 +267,7 @@ static int audit_set_backlog_limit(int limit, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -293,7 +293,7 @@ static int audit_set_enabled(int state, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -321,7 +321,7 @@ static int audit_set_failure(int state, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -538,7 +538,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		if (status_get->mask & AUDIT_STATUS_PID) {
 			int old   = audit_pid;
 			if (sid) {
-				if ((err = selinux_ctxid_to_string(
+				if ((err = selinux_sid_to_string(
 						sid, &ctx, &len)))
 					return err;
 				else
@@ -576,7 +576,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 						 "user pid=%d uid=%u auid=%u",
 						 pid, uid, loginuid);
 				if (sid) {
-					if (selinux_ctxid_to_string(
+					if (selinux_sid_to_string(
 							sid, &ctx, &len)) {
 						audit_log_format(ab, 
 							" ssid=%u", sid);
@@ -614,7 +614,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 					   loginuid, sid);
 		break;
 	case AUDIT_SIGNAL_INFO:
-		err = selinux_ctxid_to_string(audit_sig_sid, &ctx, &len);
+		err = selinux_sid_to_string(audit_sig_sid, &ctx, &len);
 		if (err)
 			return err;
 		sig_data = kmalloc(sizeof(*sig_data) + len, GFP_KERNEL);
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index a44879b0c72f..1a58a81fb09d 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1398,7 +1398,7 @@ static void audit_log_rule_change(uid_t loginuid, u32 sid, char *action,
 	if (sid) {
 		char *ctx = NULL;
 		u32 len;
-		if (selinux_ctxid_to_string(sid, &ctx, &len))
+		if (selinux_sid_to_string(sid, &ctx, &len))
 			audit_log_format(ab, " ssid=%u", sid);
 		else
 			audit_log_format(ab, " subj=%s", ctx);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 331e17010393..fb83c5cb8c32 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -898,7 +898,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 			if (axi->osid != 0) {
 				char *ctx = NULL;
 				u32 len;
-				if (selinux_ctxid_to_string(
+				if (selinux_sid_to_string(
 						axi->osid, &ctx, &len)) {
 					audit_log_format(ab, " osid=%u",
 							axi->osid);
@@ -1005,7 +1005,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 		if (n->osid != 0) {
 			char *ctx = NULL;
 			u32 len;
-			if (selinux_ctxid_to_string(
+			if (selinux_sid_to_string(
 				n->osid, &ctx, &len)) {
 				audit_log_format(ab, " osid=%u", n->osid);
 				call_panic = 2;
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index ee0fb47f81ae..b6f96943be1f 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -21,10 +21,10 @@
 #include "security.h"
 #include "objsec.h"
 
-int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
+int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
 {
 	if (selinux_enabled)
-		return security_sid_to_context(ctxid, ctx, ctxlen);
+		return security_sid_to_context(sid, ctx, ctxlen);
 	else {
 		*ctx = NULL;
 		*ctxlen = 0;
-- 
cgit v1.2.3


From 9a2f44f01a67a6ecca71515af999895b45a2aeb0 Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:58 -0700
Subject: [PATCH] selinux: replace ctxid with sid in selinux_audit_rule_match
 interface

Replace ctxid with sid in selinux_audit_rule_match interface for
consistency with other interfaces.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/selinux.h        | 6 +++---
 security/selinux/ss/services.c | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index df9098de4c99..d1b7ca6c1c57 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -46,7 +46,7 @@ void selinux_audit_rule_free(struct selinux_audit_rule *rule);
 
 /**
  *	selinux_audit_rule_match - determine if a context ID matches a rule.
- *	@ctxid: the context ID to check
+ *	@sid: the context ID to check
  *	@field: the field this rule refers to
  *	@op: the operater the rule uses
  *	@rule: pointer to the audit rule to check against
@@ -55,7 +55,7 @@ void selinux_audit_rule_free(struct selinux_audit_rule *rule);
  *	Returns 1 if the context id matches the rule, 0 if it does not, and
  *	-errno on failure.
  */
-int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
+int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
                              struct selinux_audit_rule *rule,
                              struct audit_context *actx);
 
@@ -144,7 +144,7 @@ static inline void selinux_audit_rule_free(struct selinux_audit_rule *rule)
 	return;
 }
 
-static inline int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
+static inline int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
                                            struct selinux_audit_rule *rule,
                                            struct audit_context *actx)
 {
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 22ed17c17718..988079f45294 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2003,7 +2003,7 @@ int selinux_audit_rule_init(u32 field, u32 op, char *rulestr,
 	return rc;
 }
 
-int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
+int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
                              struct selinux_audit_rule *rule,
                              struct audit_context *actx)
 {
@@ -2026,11 +2026,11 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
 		goto out;
 	}
 
-	ctxt = sidtab_search(&sidtab, ctxid);
+	ctxt = sidtab_search(&sidtab, sid);
 	if (!ctxt) {
 		audit_log(actx, GFP_ATOMIC, AUDIT_SELINUX_ERR,
 		          "selinux_audit_rule_match: unrecognized SID %d\n",
-		          ctxid);
+		          sid);
 		match = -ENOENT;
 		goto out;
 	}
-- 
cgit v1.2.3


From 1bcbba306048ed86b935d57a95d887c23d52c94b Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:04 -0700
Subject: [PATCH] FRV: Use the generic IRQ stuff

Make the FRV arch use the generic IRQ code rather than having its own
routines for doing so.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/Kconfig                    |   8 +-
 arch/frv/kernel/Makefile            |   5 +-
 arch/frv/kernel/irq-mb93091.c       | 157 +++++---
 arch/frv/kernel/irq-mb93093.c       | 115 ++++--
 arch/frv/kernel/irq-mb93493.c       | 167 +++++---
 arch/frv/kernel/irq-routing.c       | 291 --------------
 arch/frv/kernel/irq.c               | 750 ++++++------------------------------
 arch/frv/kernel/setup.c             |   1 -
 arch/frv/kernel/time.c              |   1 -
 arch/frv/mb93090-mb00/pci-irq.c     |   1 -
 include/asm-frv/cpu-irqs.h          |  54 ++-
 include/asm-frv/hardirq.h           |   5 +
 include/asm-frv/irq-routing.h       |  70 ----
 include/asm-frv/irq.h               |  26 +-
 include/asm-frv/mb93091-fpga-irqs.h |   6 +-
 include/asm-frv/mb93093-fpga-irqs.h |   6 +-
 include/asm-frv/mb93493-irqs.h      |   6 +-
 include/asm-frv/mb93493-regs.h      |   2 +
 18 files changed, 450 insertions(+), 1221 deletions(-)
 delete mode 100644 arch/frv/kernel/irq-routing.c
 delete mode 100644 include/asm-frv/irq-routing.h

(limited to 'include')

diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index a601a17cf568..5833808fb823 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -27,7 +27,7 @@ config GENERIC_CALIBRATE_DELAY
 
 config GENERIC_HARDIRQS
 	bool
-	default n
+	default y
 
 config GENERIC_TIME
 	bool
@@ -251,6 +251,12 @@ config MB93091_NO_MB
 endchoice
 endif
 
+config FUJITSU_MB93493
+	bool "MB93493 Multimedia chip"
+	help
+	  Select this option if the MB93493 multimedia chip is going to be
+	  used.
+
 choice
 	prompt "GP-Relative data support"
 	default GPREL_DATA_8
diff --git a/arch/frv/kernel/Makefile b/arch/frv/kernel/Makefile
index 5a827b349b5e..32db3499c461 100644
--- a/arch/frv/kernel/Makefile
+++ b/arch/frv/kernel/Makefile
@@ -10,15 +10,14 @@ extra-y:= head.o init_task.o vmlinux.lds
 obj-y := $(heads-y) entry.o entry-table.o break.o switch_to.o kernel_thread.o \
 	 process.o traps.o ptrace.o signal.o dma.o \
 	 sys_frv.o time.o semaphore.o setup.o frv_ksyms.o \
-	 debug-stub.o irq.o irq-routing.o sleep.o uaccess.o
+	 debug-stub.o irq.o sleep.o uaccess.o
 
 obj-$(CONFIG_GDBSTUB)		+= gdb-stub.o gdb-io.o
 
 obj-$(CONFIG_MB93091_VDK)	+= irq-mb93091.o
-obj-$(CONFIG_MB93093_PDK)	+= irq-mb93093.o
-obj-$(CONFIG_FUJITSU_MB93493)	+= irq-mb93493.o
 obj-$(CONFIG_PM)		+= pm.o cmode.o
 obj-$(CONFIG_MB93093_PDK)	+= pm-mb93093.o
+obj-$(CONFIG_FUJITSU_MB93493)	+= irq-mb93493.o
 obj-$(CONFIG_SYSCTL)		+= sysctl.o
 obj-$(CONFIG_FUTEX)		+= futex.o
 obj-$(CONFIG_MODULES)		+= module.o
diff --git a/arch/frv/kernel/irq-mb93091.c b/arch/frv/kernel/irq-mb93091.c
index 1381abcd5cc9..635d23437666 100644
--- a/arch/frv/kernel/irq-mb93091.c
+++ b/arch/frv/kernel/irq-mb93091.c
@@ -24,7 +24,6 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 
 #define __reg16(ADDR) (*(volatile unsigned short *)(ADDR))
 
@@ -33,83 +32,129 @@
 #define __get_IFR()	({ __reg16(0xffc0000c); })
 #define __clr_IFR(M)	do { __reg16(0xffc0000c) = ~(M); wmb(); } while(0)
 
-static void frv_fpga_doirq(struct irq_source *source);
-static void frv_fpga_control(struct irq_group *group, int irq, int on);
 
-/*****************************************************************************/
 /*
- * FPGA IRQ multiplexor
+ * on-motherboard FPGA PIC operations
  */
-static struct irq_source frv_fpga[4] = {
-#define __FPGA(X, M)					\
-	[X] = {						\
-		.muxname	= "fpga."#X,		\
-		.irqmask	= M,			\
-		.doirq		= frv_fpga_doirq,	\
-	}
+static void frv_fpga_enable(unsigned int irq)
+{
+	uint16_t imr = __get_IMR();
 
-	__FPGA(0, 0x0028),
-	__FPGA(1, 0x0050),
-	__FPGA(2, 0x1c00),
-	__FPGA(3, 0x6386),
-};
-
-static struct irq_group frv_fpga_irqs = {
-	.first_irq	= IRQ_BASE_FPGA,
-	.control	= frv_fpga_control,
-	.sources = {
-		[ 1] = &frv_fpga[3],
-		[ 2] = &frv_fpga[3],
-		[ 3] = &frv_fpga[0],
-		[ 4] = &frv_fpga[1],
-		[ 5] = &frv_fpga[0],
-		[ 6] = &frv_fpga[1],
-		[ 7] = &frv_fpga[3],
-		[ 8] = &frv_fpga[3],
-		[ 9] = &frv_fpga[3],
-		[10] = &frv_fpga[2],
-		[11] = &frv_fpga[2],
-		[12] = &frv_fpga[2],
-		[13] = &frv_fpga[3],
-		[14] = &frv_fpga[3],
-	},
-};
+	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
 
+	__set_IMR(imr);
+}
 
-static void frv_fpga_control(struct irq_group *group, int index, int on)
+static void frv_fpga_disable(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
-	if (on)
-		imr &= ~(1 << index);
-	else
-		imr |= 1 << index;
+	imr |= 1 << (irq - IRQ_BASE_FPGA);
 
 	__set_IMR(imr);
 }
 
-static void frv_fpga_doirq(struct irq_source *source)
+static void frv_fpga_ack(unsigned int irq)
+{
+	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
+}
+
+static void frv_fpga_end(unsigned int irq)
+{
+}
+
+static struct irq_chip frv_fpga_pic = {
+	.name		= "mb93091",
+	.enable		= frv_fpga_enable,
+	.disable	= frv_fpga_disable,
+	.ack		= frv_fpga_ack,
+	.mask		= frv_fpga_disable,
+	.unmask		= frv_fpga_enable,
+	.end		= frv_fpga_end,
+};
+
+/*
+ * FPGA PIC interrupt handler
+ */
+static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 {
-	uint16_t mask, imr;
+	uint16_t imr, mask = (unsigned long) _mask;
+	irqreturn_t iret = 0;
 
 	imr = __get_IMR();
-	mask = source->irqmask & ~imr & __get_IFR();
-	if (mask) {
-		__set_IMR(imr | mask);
-		__clr_IFR(mask);
-		distribute_irqs(&frv_fpga_irqs, mask);
-		__set_IMR(imr);
+	mask = mask & ~imr & __get_IFR();
+
+	/* poll all the triggered IRQs */
+	while (mask) {
+		int irq;
+
+		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(mask));
+		irq = 31 - irq;
+		mask &= ~(1 << irq);
+
+		if (__do_IRQ(IRQ_BASE_FPGA + irq, regs))
+			iret |= IRQ_HANDLED;
 	}
+
+	return iret;
 }
 
+/*
+ * define an interrupt action for each FPGA PIC output
+ * - use dev_id to indicate the FPGA PIC input to output mappings
+ */
+static struct irqaction fpga_irq[4]  = {
+	[0] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.0",
+		.dev_id		= (void *) 0x0028UL,
+	},
+	[1] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.1",
+		.dev_id		= (void *) 0x0050UL,
+	},
+	[2] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.2",
+		.dev_id		= (void *) 0x1c00UL,
+	},
+	[3] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.3",
+		.dev_id		= (void *) 0x6386UL,
+	}
+};
+
+/*
+ * initialise the motherboard FPGA's PIC
+ */
 void __init fpga_init(void)
 {
+	int irq;
+
+	/* all PIC inputs are all set to be low-level driven, apart from the
+	 * NMI button (15) which is fixed at falling-edge
+	 */
 	__set_IMR(0x7ffe);
 	__clr_IFR(0x0000);
 
-	frv_irq_route_external(&frv_fpga[0], IRQ_CPU_EXTERNAL0);
-	frv_irq_route_external(&frv_fpga[1], IRQ_CPU_EXTERNAL1);
-	frv_irq_route_external(&frv_fpga[2], IRQ_CPU_EXTERNAL2);
-	frv_irq_route_external(&frv_fpga[3], IRQ_CPU_EXTERNAL3);
-	frv_irq_set_group(&frv_fpga_irqs);
+	for (irq = IRQ_BASE_FPGA + 1; irq <= IRQ_BASE_FPGA + 14; irq++)
+		set_irq_chip_and_handler(irq, &frv_fpga_pic, handle_level_irq);
+
+	set_irq_chip_and_handler(IRQ_FPGA_NMI, &frv_fpga_pic, handle_edge_irq);
+
+	/* the FPGA drives the first four external IRQ inputs on the CPU PIC */
+	setup_irq(IRQ_CPU_EXTERNAL0, &fpga_irq[0]);
+	setup_irq(IRQ_CPU_EXTERNAL1, &fpga_irq[1]);
+	setup_irq(IRQ_CPU_EXTERNAL2, &fpga_irq[2]);
+	setup_irq(IRQ_CPU_EXTERNAL3, &fpga_irq[3]);
 }
diff --git a/arch/frv/kernel/irq-mb93093.c b/arch/frv/kernel/irq-mb93093.c
index 48b2a6420888..f60db7988e78 100644
--- a/arch/frv/kernel/irq-mb93093.c
+++ b/arch/frv/kernel/irq-mb93093.c
@@ -1,6 +1,6 @@
 /* irq-mb93093.c: MB93093 FPGA interrupt handling
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -24,7 +24,6 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 
 #define __reg16(ADDR) (*(volatile unsigned short *)(__region_CS2 + (ADDR)))
 
@@ -33,66 +32,100 @@
 #define __get_IFR()	({ __reg16(0x02); })
 #define __clr_IFR(M)	do { __reg16(0x02) = ~(M); wmb(); } while(0)
 
-static void frv_fpga_doirq(struct irq_source *source);
-static void frv_fpga_control(struct irq_group *group, int irq, int on);
-
-/*****************************************************************************/
 /*
- * FPGA IRQ multiplexor
+ * off-CPU FPGA PIC operations
  */
-static struct irq_source frv_fpga[4] = {
-#define __FPGA(X, M)					\
-	[X] = {						\
-		.muxname	= "fpga."#X,		\
-		.irqmask	= M,			\
-		.doirq		= frv_fpga_doirq,	\
-	}
-
-	__FPGA(0, 0x0700),
-};
+static void frv_fpga_enable(unsigned int irq)
+{
+	uint16_t imr = __get_IMR();
 
-static struct irq_group frv_fpga_irqs = {
-	.first_irq	= IRQ_BASE_FPGA,
-	.control	= frv_fpga_control,
-	.sources = {
-		[ 8] = &frv_fpga[0],
-		[ 9] = &frv_fpga[0],
-		[10] = &frv_fpga[0],
-	},
-};
+	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
 
+	__set_IMR(imr);
+}
 
-static void frv_fpga_control(struct irq_group *group, int index, int on)
+static void frv_fpga_disable(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
-	if (on)
-		imr &= ~(1 << index);
-	else
-		imr |= 1 << index;
+	imr |= 1 << (irq - IRQ_BASE_FPGA);
 
 	__set_IMR(imr);
 }
 
-static void frv_fpga_doirq(struct irq_source *source)
+static void frv_fpga_ack(unsigned int irq)
+{
+	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
+}
+
+static void frv_fpga_end(unsigned int irq)
+{
+}
+
+static struct irq_chip frv_fpga_pic = {
+	.name		= "mb93093",
+	.enable		= frv_fpga_enable,
+	.disable	= frv_fpga_disable,
+	.ack		= frv_fpga_ack,
+	.mask		= frv_fpga_disable,
+	.unmask		= frv_fpga_enable,
+	.end		= frv_fpga_end,
+};
+
+/*
+ * FPGA PIC interrupt handler
+ */
+static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 {
-	uint16_t mask, imr;
+	uint16_t imr, mask = (unsigned long) _mask;
+	irqreturn_t iret = 0;
 
 	imr = __get_IMR();
-	mask = source->irqmask & ~imr & __get_IFR();
-	if (mask) {
-		__set_IMR(imr | mask);
-		__clr_IFR(mask);
-		distribute_irqs(&frv_fpga_irqs, mask);
-		__set_IMR(imr);
+	mask = mask & ~imr & __get_IFR();
+
+	/* poll all the triggered IRQs */
+	while (mask) {
+		int irq;
+
+		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(mask));
+		irq = 31 - irq;
+		mask &= ~(1 << irq);
+
+		if (__do_IRQ(IRQ_BASE_FPGA + irq, regs))
+			iret |= IRQ_HANDLED;
 	}
+
+	return iret;
 }
 
+/*
+ * define an interrupt action for each FPGA PIC output
+ * - use dev_id to indicate the FPGA PIC input to output mappings
+ */
+static struct irqaction fpga_irq[1]  = {
+	[0] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.0",
+		.dev_id		= (void *) 0x0700UL,
+	}
+};
+
+/*
+ * initialise the motherboard FPGA's PIC
+ */
 void __init fpga_init(void)
 {
+	int irq;
+
+	/* all PIC inputs are all set to be edge triggered */
 	__set_IMR(0x0700);
 	__clr_IFR(0x0000);
 
-	frv_irq_route_external(&frv_fpga[0], IRQ_CPU_EXTERNAL2);
-	frv_irq_set_group(&frv_fpga_irqs);
+	for (irq = IRQ_BASE_FPGA + 8; irq <= IRQ_BASE_FPGA + 10; irq++)
+		set_irq_chip_and_handler(irq, &frv_fpga_pic, handle_edge_irq);
+
+	/* the FPGA drives external IRQ input #2 on the CPU PIC */
+	setup_irq(IRQ_CPU_EXTERNAL2, &fpga_irq[0]);
 }
diff --git a/arch/frv/kernel/irq-mb93493.c b/arch/frv/kernel/irq-mb93493.c
index 988d035640e1..8ad9abfc7c13 100644
--- a/arch/frv/kernel/irq-mb93493.c
+++ b/arch/frv/kernel/irq-mb93493.c
@@ -1,6 +1,6 @@
 /* irq-mb93493.c: MB93493 companion chip interrupt handler
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -24,84 +24,133 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 #include <asm/mb93493-irqs.h>
+#include <asm/mb93493-regs.h>
 
-static void frv_mb93493_doirq(struct irq_source *source);
+#define IRQ_ROUTE_ONE(X) (X##_ROUTE << (X - IRQ_BASE_MB93493))
+
+#define IRQ_ROUTING					\
+	(IRQ_ROUTE_ONE(IRQ_MB93493_VDC)		|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_VCC)		|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_AUDIO_OUT)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_I2C_0)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_I2C_1)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_USB)		|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_LOCAL_BUS)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_PCMCIA)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_GPIO)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_AUDIO_IN))
 
-/*****************************************************************************/
 /*
- * MB93493 companion chip IRQ multiplexor
+ * daughter board PIC operations
  */
-static struct irq_source frv_mb93493[2] = {
-	[0] = {
-		.muxname		= "mb93493.0",
-		.muxdata		= __region_CS3 + 0x3d0,
-		.doirq			= frv_mb93493_doirq,
-		.irqmask		= 0x0000,
-	},
-	[1] = {
-		.muxname		= "mb93493.1",
-		.muxdata		= __region_CS3 + 0x3d4,
-		.doirq			= frv_mb93493_doirq,
-		.irqmask		= 0x0000,
-	},
-};
-
-static void frv_mb93493_control(struct irq_group *group, int index, int on)
+static void frv_mb93493_enable(unsigned int irq)
 {
-	struct irq_source *source;
 	uint32_t iqsr;
+	volatile void *piqsr;
 
-	if ((frv_mb93493[0].irqmask & (1 << index)))
-		source = &frv_mb93493[0];
+	if (IRQ_ROUTING & (1 << (irq - IRQ_BASE_MB93493)))
+		piqsr = __addr_MB93493_IQSR(1);
 	else
-		source = &frv_mb93493[1];
+		piqsr = __addr_MB93493_IQSR(0);
 
-	iqsr = readl(source->muxdata);
-	if (on)
-		iqsr |= 1 << (index + 16);
+	iqsr = readl(piqsr);
+	iqsr |= 1 << (irq - IRQ_BASE_MB93493 + 16);
+	writel(iqsr, piqsr);
+}
+
+static void frv_mb93493_disable(unsigned int irq)
+{
+	uint32_t iqsr;
+	volatile void *piqsr;
+
+	if (IRQ_ROUTING & (1 << (irq - IRQ_BASE_MB93493)))
+		piqsr = __addr_MB93493_IQSR(1);
 	else
-		iqsr &= ~(1 << (index + 16));
+		piqsr = __addr_MB93493_IQSR(0);
 
-	writel(iqsr, source->muxdata);
+	iqsr = readl(piqsr);
+	iqsr &= ~(1 << (irq - IRQ_BASE_MB93493 + 16));
+	writel(iqsr, piqsr);
 }
 
-static struct irq_group frv_mb93493_irqs = {
-	.first_irq	= IRQ_BASE_MB93493,
-	.control	= frv_mb93493_control,
-};
-
-static void frv_mb93493_doirq(struct irq_source *source)
+static void frv_mb93493_ack(unsigned int irq)
 {
-	uint32_t mask = readl(source->muxdata);
-	mask = mask & (mask >> 16) & 0xffff;
+}
 
-	if (mask)
-		distribute_irqs(&frv_mb93493_irqs, mask);
+static void frv_mb93493_end(unsigned int irq)
+{
 }
 
-static void __init mb93493_irq_route(int irq, int source)
+static struct irq_chip frv_mb93493_pic = {
+	.name		= "mb93093",
+	.enable		= frv_mb93493_enable,
+	.disable	= frv_mb93493_disable,
+	.ack		= frv_mb93493_ack,
+	.mask		= frv_mb93493_disable,
+	.unmask		= frv_mb93493_enable,
+	.end		= frv_mb93493_end,
+};
+
+/*
+ * MB93493 PIC interrupt handler
+ */
+static irqreturn_t mb93493_interrupt(int irq, void *_piqsr, struct pt_regs *regs)
 {
-	frv_mb93493[source].irqmask |= 1 << (irq - IRQ_BASE_MB93493);
-	frv_mb93493_irqs.sources[irq - IRQ_BASE_MB93493] = &frv_mb93493[source];
+	volatile void *piqsr = _piqsr;
+	irqreturn_t iret = 0;
+	uint32_t iqsr;
+
+	iqsr = readl(piqsr);
+	iqsr = iqsr & (iqsr >> 16) & 0xffff;
+
+	/* poll all the triggered IRQs */
+	while (iqsr) {
+		int irq;
+
+		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(iqsr));
+		irq = 31 - irq;
+		iqsr &= ~(1 << irq);
+
+		if (__do_IRQ(IRQ_BASE_MB93493 + irq, regs))
+			iret |= IRQ_HANDLED;
+	}
+
+	return iret;
 }
 
-void __init route_mb93493_irqs(void)
+/*
+ * define an interrupt action for each MB93493 PIC output
+ * - use dev_id to indicate the MB93493 PIC input to output mappings
+ */
+static struct irqaction mb93493_irq[2]  = {
+	[0] = {
+		.handler	= mb93493_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "mb93493.0",
+		.dev_id		= (void *) __addr_MB93493_IQSR(0),
+	},
+	[1] = {
+		.handler	= mb93493_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "mb93493.1",
+		.dev_id		= (void *) __addr_MB93493_IQSR(1),
+	}
+};
+
+/*
+ * initialise the motherboard MB93493's PIC
+ */
+void __init mb93493_init(void)
 {
-	frv_irq_route_external(&frv_mb93493[0], IRQ_CPU_MB93493_0);
-	frv_irq_route_external(&frv_mb93493[1], IRQ_CPU_MB93493_1);
-
-	frv_irq_set_group(&frv_mb93493_irqs);
-
-	mb93493_irq_route(IRQ_MB93493_VDC,		IRQ_MB93493_VDC_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_VCC,		IRQ_MB93493_VCC_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_AUDIO_IN,		IRQ_MB93493_AUDIO_IN_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_I2C_0,		IRQ_MB93493_I2C_0_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_I2C_1,		IRQ_MB93493_I2C_1_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_USB,		IRQ_MB93493_USB_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_LOCAL_BUS,	IRQ_MB93493_LOCAL_BUS_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_PCMCIA,		IRQ_MB93493_PCMCIA_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_GPIO,		IRQ_MB93493_GPIO_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_AUDIO_OUT,	IRQ_MB93493_AUDIO_OUT_ROUTE);
+	int irq;
+
+	for (irq = IRQ_BASE_MB93493 + 0; irq <= IRQ_BASE_MB93493 + 10; irq++)
+		set_irq_chip_and_handler(irq, &frv_mb93493_pic, handle_edge_irq);
+
+	/* the MB93493 drives external IRQ inputs on the CPU PIC */
+	setup_irq(IRQ_CPU_MB93493_0, &mb93493_irq[0]);
+	setup_irq(IRQ_CPU_MB93493_1, &mb93493_irq[1]);
 }
diff --git a/arch/frv/kernel/irq-routing.c b/arch/frv/kernel/irq-routing.c
deleted file mode 100644
index 53886adf47de..000000000000
--- a/arch/frv/kernel/irq-routing.c
+++ /dev/null
@@ -1,291 +0,0 @@
-/* irq-routing.c: IRQ routing
- *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#include <linux/sched.h>
-#include <linux/random.h>
-#include <linux/init.h>
-#include <linux/serial_reg.h>
-#include <asm/io.h>
-#include <asm/irq-routing.h>
-#include <asm/irc-regs.h>
-#include <asm/serial-regs.h>
-#include <asm/dma.h>
-
-struct irq_level frv_irq_levels[16] = {
-	[0 ... 15] = {
-		.lock	= SPIN_LOCK_UNLOCKED,
-	}
-};
-
-struct irq_group *irq_groups[NR_IRQ_GROUPS];
-
-extern struct irq_group frv_cpu_irqs;
-
-void __init frv_irq_route(struct irq_source *source, int irqlevel)
-{
-	source->level = &frv_irq_levels[irqlevel];
-	source->next = frv_irq_levels[irqlevel].sources;
-	frv_irq_levels[irqlevel].sources = source;
-}
-
-void __init frv_irq_route_external(struct irq_source *source, int irq)
-{
-	int irqlevel = 0;
-
-	switch (irq) {
-	case IRQ_CPU_EXTERNAL0:	irqlevel = IRQ_XIRQ0_LEVEL; break;
-	case IRQ_CPU_EXTERNAL1:	irqlevel = IRQ_XIRQ1_LEVEL; break;
-	case IRQ_CPU_EXTERNAL2:	irqlevel = IRQ_XIRQ2_LEVEL; break;
-	case IRQ_CPU_EXTERNAL3:	irqlevel = IRQ_XIRQ3_LEVEL; break;
-	case IRQ_CPU_EXTERNAL4:	irqlevel = IRQ_XIRQ4_LEVEL; break;
-	case IRQ_CPU_EXTERNAL5:	irqlevel = IRQ_XIRQ5_LEVEL; break;
-	case IRQ_CPU_EXTERNAL6:	irqlevel = IRQ_XIRQ6_LEVEL; break;
-	case IRQ_CPU_EXTERNAL7:	irqlevel = IRQ_XIRQ7_LEVEL; break;
-	default: BUG();
-	}
-
-	source->level = &frv_irq_levels[irqlevel];
-	source->next = frv_irq_levels[irqlevel].sources;
-	frv_irq_levels[irqlevel].sources = source;
-}
-
-void __init frv_irq_set_group(struct irq_group *group)
-{
-	irq_groups[group->first_irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP] = group;
-}
-
-void distribute_irqs(struct irq_group *group, unsigned long irqmask)
-{
-	struct irqaction *action;
-	int irq;
-
-	while (irqmask) {
-		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(irqmask));
-		if (irq < 0 || irq > 31)
-			asm volatile("break");
-		irq = 31 - irq;
-
-		irqmask &= ~(1 << irq);
-		action = group->actions[irq];
-
-		irq += group->first_irq;
-
-		if (action) {
-			int status = 0;
-
-//			if (!(action->flags & IRQF_DISABLED))
-//				local_irq_enable();
-
-			do {
-				status |= action->flags;
-				action->handler(irq, action->dev_id, __frame);
-				action = action->next;
-			} while (action);
-
-			if (status & IRQF_SAMPLE_RANDOM)
-				add_interrupt_randomness(irq);
-			local_irq_disable();
-		}
-	}
-}
-
-/*****************************************************************************/
-/*
- * CPU UART interrupts
- */
-static void frv_cpuuart_doirq(struct irq_source *source)
-{
-//	uint8_t iir = readb(source->muxdata + UART_IIR * 8);
-//	if ((iir & 0x0f) != UART_IIR_NO_INT)
-		distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cpuuart[2] = {
-#define __CPUUART(X, A)						\
-	[X] = {							\
-		.muxname	= "uart",			\
-		.muxdata	= (volatile void __iomem *)(unsigned long)A,\
-		.irqmask	= 1 << IRQ_CPU_UART##X,		\
-		.doirq		= frv_cpuuart_doirq,		\
-	}
-
-	__CPUUART(0, UART0_BASE),
-	__CPUUART(1, UART1_BASE),
-};
-
-/*****************************************************************************/
-/*
- * CPU DMA interrupts
- */
-static void frv_cpudma_doirq(struct irq_source *source)
-{
-	uint32_t cstr = readl(source->muxdata + DMAC_CSTRx);
-	if (cstr & DMAC_CSTRx_INT)
-		distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cpudma[8] = {
-#define __CPUDMA(X, A)						\
-	[X] = {							\
-		.muxname	= "dma",			\
-		.muxdata	= (volatile void __iomem *)(unsigned long)A,\
-		.irqmask	= 1 << IRQ_CPU_DMA##X,		\
-		.doirq		= frv_cpudma_doirq,		\
-	}
-
-	__CPUDMA(0, 0xfe000900),
-	__CPUDMA(1, 0xfe000980),
-	__CPUDMA(2, 0xfe000a00),
-	__CPUDMA(3, 0xfe000a80),
-	__CPUDMA(4, 0xfe001000),
-	__CPUDMA(5, 0xfe001080),
-	__CPUDMA(6, 0xfe001100),
-	__CPUDMA(7, 0xfe001180),
-};
-
-/*****************************************************************************/
-/*
- * CPU timer interrupts - can't tell whether they've generated an interrupt or not
- */
-static void frv_cputimer_doirq(struct irq_source *source)
-{
-	distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cputimer[3] = {
-#define __CPUTIMER(X)						\
-	[X] = {							\
-		.muxname	= "timer",			\
-		.muxdata	= NULL,				\
-		.irqmask	= 1 << IRQ_CPU_TIMER##X,	\
-		.doirq		= frv_cputimer_doirq,		\
-	}
-
-	__CPUTIMER(0),
-	__CPUTIMER(1),
-	__CPUTIMER(2),
-};
-
-/*****************************************************************************/
-/*
- * external CPU interrupts - can't tell directly whether they've generated an interrupt or not
- */
-static void frv_cpuexternal_doirq(struct irq_source *source)
-{
-	distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cpuexternal[8] = {
-#define __CPUEXTERNAL(X)					\
-	[X] = {							\
-		.muxname	= "ext",			\
-		.muxdata	= NULL,				\
-		.irqmask	= 1 << IRQ_CPU_EXTERNAL##X,	\
-		.doirq		= frv_cpuexternal_doirq,	\
-	}
-
-	__CPUEXTERNAL(0),
-	__CPUEXTERNAL(1),
-	__CPUEXTERNAL(2),
-	__CPUEXTERNAL(3),
-	__CPUEXTERNAL(4),
-	__CPUEXTERNAL(5),
-	__CPUEXTERNAL(6),
-	__CPUEXTERNAL(7),
-};
-
-#define set_IRR(N,A,B,C,D) __set_IRR(N, (A << 28) | (B << 24) | (C << 20) | (D << 16))
-
-struct irq_group frv_cpu_irqs = {
-	.sources = {
-		[IRQ_CPU_UART0]		= &frv_cpuuart[0],
-		[IRQ_CPU_UART1]		= &frv_cpuuart[1],
-		[IRQ_CPU_TIMER0]	= &frv_cputimer[0],
-		[IRQ_CPU_TIMER1]	= &frv_cputimer[1],
-		[IRQ_CPU_TIMER2]	= &frv_cputimer[2],
-		[IRQ_CPU_DMA0]		= &frv_cpudma[0],
-		[IRQ_CPU_DMA1]		= &frv_cpudma[1],
-		[IRQ_CPU_DMA2]		= &frv_cpudma[2],
-		[IRQ_CPU_DMA3]		= &frv_cpudma[3],
-		[IRQ_CPU_DMA4]		= &frv_cpudma[4],
-		[IRQ_CPU_DMA5]		= &frv_cpudma[5],
-		[IRQ_CPU_DMA6]		= &frv_cpudma[6],
-		[IRQ_CPU_DMA7]		= &frv_cpudma[7],
-		[IRQ_CPU_EXTERNAL0]	= &frv_cpuexternal[0],
-		[IRQ_CPU_EXTERNAL1]	= &frv_cpuexternal[1],
-		[IRQ_CPU_EXTERNAL2]	= &frv_cpuexternal[2],
-		[IRQ_CPU_EXTERNAL3]	= &frv_cpuexternal[3],
-		[IRQ_CPU_EXTERNAL4]	= &frv_cpuexternal[4],
-		[IRQ_CPU_EXTERNAL5]	= &frv_cpuexternal[5],
-		[IRQ_CPU_EXTERNAL6]	= &frv_cpuexternal[6],
-		[IRQ_CPU_EXTERNAL7]	= &frv_cpuexternal[7],
-	},
-};
-
-/*****************************************************************************/
-/*
- * route the CPU's interrupt sources
- */
-void __init route_cpu_irqs(void)
-{
-	frv_irq_set_group(&frv_cpu_irqs);
-
-	__set_IITMR(0, 0x003f0000);	/* DMA0-3, TIMER0-2 IRQ detect levels */
-	__set_IITMR(1, 0x20000000);	/* ERR0-1, UART0-1, DMA4-7 IRQ detect levels */
-
-	/* route UART and error interrupts */
-	frv_irq_route(&frv_cpuuart[0],	IRQ_UART0_LEVEL);
-	frv_irq_route(&frv_cpuuart[1],	IRQ_UART1_LEVEL);
-
-	set_IRR(6, IRQ_GDBSTUB_LEVEL, IRQ_GDBSTUB_LEVEL, IRQ_UART1_LEVEL, IRQ_UART0_LEVEL);
-
-	/* route DMA channel interrupts */
-	frv_irq_route(&frv_cpudma[0],	IRQ_DMA0_LEVEL);
-	frv_irq_route(&frv_cpudma[1],	IRQ_DMA1_LEVEL);
-	frv_irq_route(&frv_cpudma[2],	IRQ_DMA2_LEVEL);
-	frv_irq_route(&frv_cpudma[3],	IRQ_DMA3_LEVEL);
-	frv_irq_route(&frv_cpudma[4],	IRQ_DMA4_LEVEL);
-	frv_irq_route(&frv_cpudma[5],	IRQ_DMA5_LEVEL);
-	frv_irq_route(&frv_cpudma[6],	IRQ_DMA6_LEVEL);
-	frv_irq_route(&frv_cpudma[7],	IRQ_DMA7_LEVEL);
-
-	set_IRR(4, IRQ_DMA3_LEVEL, IRQ_DMA2_LEVEL, IRQ_DMA1_LEVEL, IRQ_DMA0_LEVEL);
-	set_IRR(7, IRQ_DMA7_LEVEL, IRQ_DMA6_LEVEL, IRQ_DMA5_LEVEL, IRQ_DMA4_LEVEL);
-
-	/* route timer interrupts */
-	frv_irq_route(&frv_cputimer[0],	IRQ_TIMER0_LEVEL);
-	frv_irq_route(&frv_cputimer[1],	IRQ_TIMER1_LEVEL);
-	frv_irq_route(&frv_cputimer[2],	IRQ_TIMER2_LEVEL);
-
-	set_IRR(5, 0, IRQ_TIMER2_LEVEL, IRQ_TIMER1_LEVEL, IRQ_TIMER0_LEVEL);
-
-	/* route external interrupts */
-	frv_irq_route(&frv_cpuexternal[0], IRQ_XIRQ0_LEVEL);
-	frv_irq_route(&frv_cpuexternal[1], IRQ_XIRQ1_LEVEL);
-	frv_irq_route(&frv_cpuexternal[2], IRQ_XIRQ2_LEVEL);
-	frv_irq_route(&frv_cpuexternal[3], IRQ_XIRQ3_LEVEL);
-	frv_irq_route(&frv_cpuexternal[4], IRQ_XIRQ4_LEVEL);
-	frv_irq_route(&frv_cpuexternal[5], IRQ_XIRQ5_LEVEL);
-	frv_irq_route(&frv_cpuexternal[6], IRQ_XIRQ6_LEVEL);
-	frv_irq_route(&frv_cpuexternal[7], IRQ_XIRQ7_LEVEL);
-
-	set_IRR(2, IRQ_XIRQ7_LEVEL, IRQ_XIRQ6_LEVEL, IRQ_XIRQ5_LEVEL, IRQ_XIRQ4_LEVEL);
-	set_IRR(3, IRQ_XIRQ3_LEVEL, IRQ_XIRQ2_LEVEL, IRQ_XIRQ1_LEVEL, IRQ_XIRQ0_LEVEL);
-
-#if defined(CONFIG_MB93091_VDK)
-	__set_TM1(0x55550000);		/* XIRQ7-0 all active low */
-#elif defined(CONFIG_MB93093_PDK)
-	__set_TM1(0x15550000);		/* XIRQ7 active high, 6-0 all active low */
-#else
-#error dont know external IRQ trigger levels for this setup
-#endif
-
-} /* end route_cpu_irqs() */
diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c
index 08967010be04..e1ab9f2e43fb 100644
--- a/arch/frv/kernel/irq.c
+++ b/arch/frv/kernel/irq.c
@@ -1,6 +1,6 @@
 /* irq.c: FRV IRQ handling
  *
- * Copyright (C) 2003, 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2003, 2004, 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -9,13 +9,6 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-/*
- * (mostly architecture independent, will move to kernel/irq.c in 2.5.)
- *
- * IRQs are in fact implemented a bit like signal handlers for the kernel.
- * Naturally it's not a 1:1 relation, but there are similarities.
- */
-
 #include <linux/ptrace.h>
 #include <linux/errno.h>
 #include <linux/signal.h>
@@ -43,19 +36,16 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 #include <asm/gdb-stub.h>
 
-extern void __init fpga_init(void);
-extern void __init route_mb93493_irqs(void);
-
-static void register_irq_proc (unsigned int irq);
+#define set_IRR(N,A,B,C,D) __set_IRR(N, (A << 28) | (B << 24) | (C << 20) | (D << 16))
 
-/*
- * Special irq handlers.
- */
+extern void __init fpga_init(void);
+#ifdef CONFIG_FUJITSU_MB93493
+extern void __init mb93493_init(void);
+#endif
 
-irqreturn_t no_action(int cpl, void *dev_id, struct pt_regs *regs) { return IRQ_HANDLED; }
+#define __reg16(ADDR) (*(volatile unsigned short *)(ADDR))
 
 atomic_t irq_err_count;
 
@@ -64,215 +54,99 @@ atomic_t irq_err_count;
  */
 int show_interrupts(struct seq_file *p, void *v)
 {
-	struct irqaction *action;
-	struct irq_group *group;
+	int i = *(loff_t *) v, cpu;
+	struct irqaction * action;
 	unsigned long flags;
-	int level, grp, ix, i, j;
-
-	i = *(loff_t *) v;
-
-	switch (i) {
-	case 0:
-		seq_printf(p, "           ");
-		for_each_online_cpu(j)
-			seq_printf(p, "CPU%d       ",j);
-
-		seq_putc(p, '\n');
-		break;
 
-	case 1 ... NR_IRQ_GROUPS * NR_IRQ_ACTIONS_PER_GROUP:
-		local_irq_save(flags);
-
-		grp = (i - 1) / NR_IRQ_ACTIONS_PER_GROUP;
-		group = irq_groups[grp];
-		if (!group)
-			goto skip;
-
-		ix = (i - 1) % NR_IRQ_ACTIONS_PER_GROUP;
-		action = group->actions[ix];
-		if (!action)
-			goto skip;
-
-		seq_printf(p, "%3d: ", i - 1);
-
-#ifndef CONFIG_SMP
-		seq_printf(p, "%10u ", kstat_irqs(i));
-#else
-		for_each_online_cpu(j)
-			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i - 1]);
-#endif
-
-		level = group->sources[ix]->level - frv_irq_levels;
-
-		seq_printf(p, " %12s@%x", group->sources[ix]->muxname, level);
-		seq_printf(p, "  %s", action->name);
-
-		for (action = action->next; action; action = action->next)
-			seq_printf(p, ", %s", action->name);
+	if (i == 0) {
+		char cpuname[12];
 
+		seq_printf(p, "    ");
+		for_each_present_cpu(cpu) {
+			sprintf(cpuname, "CPU%d", cpu);
+			seq_printf(p, " %10s", cpuname);
+		}
 		seq_putc(p, '\n');
-skip:
-		local_irq_restore(flags);
-		break;
+	}
 
-	case NR_IRQ_GROUPS * NR_IRQ_ACTIONS_PER_GROUP + 1:
-		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
-		break;
+	if (i < NR_IRQS) {
+		spin_lock_irqsave(&irq_desc[i].lock, flags);
+		action = irq_desc[i].action;
+		if (action) {
+			seq_printf(p, "%3d: ", i);
+			for_each_present_cpu(cpu)
+				seq_printf(p, "%10u ", kstat_cpu(cpu).irqs[i]);
+			seq_printf(p, " %10s", irq_desc[i].chip->name ? : "-");
+			seq_printf(p, "  %s", action->name);
+			for (action = action->next;
+			     action;
+			     action = action->next)
+				seq_printf(p, ", %s", action->name);
+
+			seq_putc(p, '\n');
+		}
 
-	default:
-		break;
+		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
+	} else if (i == NR_IRQS) {
+		seq_printf(p, "Err: %10u\n", atomic_read(&irq_err_count));
 	}
 
 	return 0;
 }
 
-
 /*
- * Generic enable/disable code: this just calls
- * down into the PIC-specific version for the actual
- * hardware disable after having gotten the irq
- * controller lock.
+ * on-CPU PIC operations
  */
-
-/**
- *	disable_irq_nosync - disable an irq without waiting
- *	@irq: Interrupt to disable
- *
- *	Disable the selected interrupt line.  Disables and Enables are
- *	nested.
- *	Unlike disable_irq(), this function does not ensure existing
- *	instances of the IRQ handler have completed before returning.
- *
- *	This function may be called from IRQ context.
- */
-
-void disable_irq_nosync(unsigned int irq)
+static void frv_cpupic_enable(unsigned int irqlevel)
 {
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	unsigned long flags;
-	int idx = irq & (NR_IRQ_ACTIONS_PER_GROUP - 1);
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[idx];
-	if (!source)
-		BUG();
-
-	level = source->level;
-
-	spin_lock_irqsave(&level->lock, flags);
-
-	if (group->control) {
-		if (!group->disable_cnt[idx]++)
-			group->control(group, idx, 0);
-	} else if (!level->disable_count++) {
-		__set_MASK(level - frv_irq_levels);
-	}
-
-	spin_unlock_irqrestore(&level->lock, flags);
+	__clr_MASK(irqlevel);
 }
 
-EXPORT_SYMBOL(disable_irq_nosync);
-
-/**
- *	disable_irq - disable an irq and wait for completion
- *	@irq: Interrupt to disable
- *
- *	Disable the selected interrupt line.  Enables and Disables are
- *	nested.
- *	This function waits for any pending IRQ handlers for this interrupt
- *	to complete before returning. If you use this function while
- *	holding a resource the IRQ handler may need you will deadlock.
- *
- *	This function may be called - with care - from IRQ context.
- */
-
-void disable_irq(unsigned int irq)
+static void frv_cpupic_disable(unsigned int irqlevel)
 {
-	disable_irq_nosync(irq);
-
-#ifdef CONFIG_SMP
-	if (!local_irq_count(smp_processor_id())) {
-		do {
-			barrier();
-		} while (irq_desc[irq].status & IRQ_INPROGRESS);
-	}
-#endif
+	__set_MASK(irqlevel);
 }
 
-EXPORT_SYMBOL(disable_irq);
-
-/**
- *	enable_irq - enable handling of an irq
- *	@irq: Interrupt to enable
- *
- *	Undoes the effect of one call to disable_irq().  If this
- *	matches the last disable, processing of interrupts on this
- *	IRQ line is re-enabled.
- *
- *	This function may be called from IRQ context.
- */
-
-void enable_irq(unsigned int irq)
+static void frv_cpupic_ack(unsigned int irqlevel)
 {
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	unsigned long flags;
-	int idx = irq & (NR_IRQ_ACTIONS_PER_GROUP - 1);
-	int count;
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[idx];
-	if (!source)
-		BUG();
-
-	level = source->level;
-
-	spin_lock_irqsave(&level->lock, flags);
-
-	if (group->control)
-		count = group->disable_cnt[idx];
-	else
-		count = level->disable_count;
-
-	switch (count) {
-	case 1:
-		if (group->control) {
-			if (group->actions[idx])
-				group->control(group, idx, 1);
-		} else {
-			if (level->usage)
-				__clr_MASK(level - frv_irq_levels);
-		}
-		/* fall-through */
+	__set_MASK(irqlevel);
+	__clr_RC(irqlevel);
+	__clr_IRL();
+}
 
-	default:
-		count--;
-		break;
+static void frv_cpupic_mask(unsigned int irqlevel)
+{
+	__set_MASK(irqlevel);
+}
 
-	case 0:
-		printk("enable_irq(%u) unbalanced from %p\n", irq, __builtin_return_address(0));
-	}
+static void frv_cpupic_mask_ack(unsigned int irqlevel)
+{
+	__set_MASK(irqlevel);
+	__clr_RC(irqlevel);
+	__clr_IRL();
+}
 
-	if (group->control)
-		group->disable_cnt[idx] = count;
-	else
-		level->disable_count = count;
+static void frv_cpupic_unmask(unsigned int irqlevel)
+{
+	__clr_MASK(irqlevel);
+}
 
-	spin_unlock_irqrestore(&level->lock, flags);
+static void frv_cpupic_end(unsigned int irqlevel)
+{
+	__clr_MASK(irqlevel);
 }
 
-EXPORT_SYMBOL(enable_irq);
+static struct irq_chip frv_cpu_pic = {
+	.name		= "cpu",
+	.enable		= frv_cpupic_enable,
+	.disable	= frv_cpupic_disable,
+	.ack		= frv_cpupic_ack,
+	.mask		= frv_cpupic_mask,
+	.mask_ack	= frv_cpupic_mask_ack,
+	.unmask		= frv_cpupic_unmask,
+	.end		= frv_cpupic_end,
+};
 
-/*****************************************************************************/
 /*
  * handles all normal device IRQ's
  * - registers are referred to by the __frame variable (GR28)
@@ -281,463 +155,65 @@ EXPORT_SYMBOL(enable_irq);
  */
 asmlinkage void do_IRQ(void)
 {
-	struct irq_source *source;
-	int level, cpu;
-
 	irq_enter();
-
-	level = (__frame->tbr >> 4) & 0xf;
-	cpu = smp_processor_id();
-
-	if ((unsigned long) __frame - (unsigned long) (current + 1) < 512)
-		BUG();
-
-	__set_MASK(level);
-	__clr_RC(level);
-	__clr_IRL();
-
-	kstat_this_cpu.irqs[level]++;
-
-	for (source = frv_irq_levels[level].sources; source; source = source->next)
-		source->doirq(source);
-
-	__clr_MASK(level);
-
+	__do_IRQ(__get_IRL(), __frame);
 	irq_exit();
+}
 
-} /* end do_IRQ() */
-
-/*****************************************************************************/
 /*
  * handles all NMIs when not co-opted by the debugger
  * - registers are referred to by the __frame variable (GR28)
  */
 asmlinkage void do_NMI(void)
 {
-} /* end do_NMI() */
-
-/*****************************************************************************/
-/**
- *	request_irq - allocate an interrupt line
- *	@irq: Interrupt line to allocate
- *	@handler: Function to be called when the IRQ occurs
- *	@irqflags: Interrupt type flags
- *	@devname: An ascii name for the claiming device
- *	@dev_id: A cookie passed back to the handler function
- *
- *	This call allocates interrupt resources and enables the
- *	interrupt line and IRQ handling. From the point this
- *	call is made your handler function may be invoked. Since
- *	your handler function must clear any interrupt the board
- *	raises, you must take care both to initialise your hardware
- *	and to set up the interrupt handler in the right order.
- *
- *	Dev_id must be globally unique. Normally the address of the
- *	device data structure is used as the cookie. Since the handler
- *	receives this value it makes sense to use it.
- *
- *	If your interrupt is shared you must pass a non NULL dev_id
- *	as this is required when freeing the interrupt.
- *
- *	Flags:
- *
- *	IRQF_SHARED		Interrupt is shared
- *
- *	IRQF_DISABLED	Disable local interrupts while processing
- *
- *	IRQF_SAMPLE_RANDOM	The interrupt can be used for entropy
- *
- */
-
-int request_irq(unsigned int irq,
-		irqreturn_t (*handler)(int, void *, struct pt_regs *),
-		unsigned long irqflags,
-		const char * devname,
-		void *dev_id)
-{
-	int retval;
-	struct irqaction *action;
-
-#if 1
-	/*
-	 * Sanity-check: shared interrupts should REALLY pass in
-	 * a real dev-ID, otherwise we'll have trouble later trying
-	 * to figure out which interrupt is which (messes up the
-	 * interrupt freeing logic etc).
-	 */
-	if (irqflags & IRQF_SHARED) {
-		if (!dev_id)
-			printk("Bad boy: %s (at 0x%x) called us without a dev_id!\n",
-			       devname, (&irq)[-1]);
-	}
-#endif
-
-	if ((irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP) >= NR_IRQ_GROUPS)
-		return -EINVAL;
-	if (!handler)
-		return -EINVAL;
-
-	action = (struct irqaction *) kmalloc(sizeof(struct irqaction), GFP_KERNEL);
-	if (!action)
-		return -ENOMEM;
-
-	action->handler = handler;
-	action->flags = irqflags;
-	action->mask = CPU_MASK_NONE;
-	action->name = devname;
-	action->next = NULL;
-	action->dev_id = dev_id;
-
-	retval = setup_irq(irq, action);
-	if (retval)
-		kfree(action);
-	return retval;
-}
-
-EXPORT_SYMBOL(request_irq);
-
-/**
- *	free_irq - free an interrupt
- *	@irq: Interrupt line to free
- *	@dev_id: Device identity to free
- *
- *	Remove an interrupt handler. The handler is removed and if the
- *	interrupt line is no longer in use by any driver it is disabled.
- *	On a shared IRQ the caller must ensure the interrupt is disabled
- *	on the card it drives before calling this function. The function
- *	does not return until any executing interrupts for this IRQ
- *	have completed.
- *
- *	This function may be called from interrupt context.
- *
- *	Bugs: Attempting to free an irq in a handler for the same irq hangs
- *	      the machine.
- */
-
-void free_irq(unsigned int irq, void *dev_id)
-{
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	struct irqaction **p, **pp;
-	unsigned long flags;
-
-	if ((irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP) >= NR_IRQ_GROUPS)
-		return;
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-	if (!source)
-		BUG();
-
-	level = source->level;
-	p = &group->actions[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-
-	spin_lock_irqsave(&level->lock, flags);
-
-	for (pp = p; *pp; pp = &(*pp)->next) {
-		struct irqaction *action = *pp;
-
-		if (action->dev_id != dev_id)
-			continue;
-
-		/* found it - remove from the list of entries */
-		*pp = action->next;
-
-		level->usage--;
-
-		if (p == pp && group->control)
-			group->control(group, irq & (NR_IRQ_ACTIONS_PER_GROUP - 1), 0);
-
-		if (level->usage == 0)
-			__set_MASK(level - frv_irq_levels);
-
-		spin_unlock_irqrestore(&level->lock,flags);
-
-#ifdef CONFIG_SMP
-		/* Wait to make sure it's not being used on another CPU */
-		while (desc->status & IRQ_INPROGRESS)
-			barrier();
-#endif
-		kfree(action);
-		return;
-	}
-}
-
-EXPORT_SYMBOL(free_irq);
-
-/*
- * IRQ autodetection code..
- *
- * This depends on the fact that any interrupt that comes in on to an
- * unassigned IRQ will cause GxICR_DETECT to be set
- */
-
-static DECLARE_MUTEX(probe_sem);
-
-/**
- *	probe_irq_on	- begin an interrupt autodetect
- *
- *	Commence probing for an interrupt. The interrupts are scanned
- *	and a mask of potential interrupt lines is returned.
- *
- */
-
-unsigned long probe_irq_on(void)
-{
-	down(&probe_sem);
-	return 0;
 }
 
-EXPORT_SYMBOL(probe_irq_on);
-
-/*
- * Return a mask of triggered interrupts (this
- * can handle only legacy ISA interrupts).
- */
-
-/**
- *	probe_irq_mask - scan a bitmap of interrupt lines
- *	@val:	mask of interrupts to consider
- *
- *	Scan the ISA bus interrupt lines and return a bitmap of
- *	active interrupts. The interrupt probe logic state is then
- *	returned to its previous value.
- *
- *	Note: we need to scan all the irq's even though we will
- *	only return ISA irq numbers - just so that we reset them
- *	all to a known state.
- */
-unsigned int probe_irq_mask(unsigned long xmask)
-{
-	up(&probe_sem);
-	return 0;
-}
-
-EXPORT_SYMBOL(probe_irq_mask);
-
 /*
- * Return the one interrupt that triggered (this can
- * handle any interrupt source).
- */
-
-/**
- *	probe_irq_off	- end an interrupt autodetect
- *	@xmask: mask of potential interrupts (unused)
- *
- *	Scans the unused interrupt lines and returns the line which
- *	appears to have triggered the interrupt. If no interrupt was
- *	found then zero is returned. If more than one interrupt is
- *	found then minus the first candidate is returned to indicate
- *	their is doubt.
- *
- *	The interrupt probe logic state is returned to its previous
- *	value.
- *
- *	BUGS: When used in a module (which arguably shouldnt happen)
- *	nothing prevents two IRQ probe callers from overlapping. The
- *	results of this are non-optimal.
+ * initialise the interrupt system
  */
-
-int probe_irq_off(unsigned long xmask)
-{
-	up(&probe_sem);
-	return -1;
-}
-
-EXPORT_SYMBOL(probe_irq_off);
-
-/* this was setup_x86_irq but it seems pretty generic */
-int setup_irq(unsigned int irq, struct irqaction *new)
-{
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	struct irqaction **p, **pp;
-	unsigned long flags;
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-	if (!source)
-		BUG();
-
-	level = source->level;
-
-	p = &group->actions[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-
-	/*
-	 * Some drivers like serial.c use request_irq() heavily,
-	 * so we have to be careful not to interfere with a
-	 * running system.
-	 */
-	if (new->flags & IRQF_SAMPLE_RANDOM) {
-		/*
-		 * This function might sleep, we want to call it first,
-		 * outside of the atomic block.
-		 * Yes, this might clear the entropy pool if the wrong
-		 * driver is attempted to be loaded, without actually
-		 * installing a new handler, but is this really a problem,
-		 * only the sysadmin is able to do this.
-		 */
-		rand_initialize_irq(irq);
-	}
-
-	/* must juggle the interrupt processing stuff with interrupts disabled */
-	spin_lock_irqsave(&level->lock, flags);
-
-	/* can't share interrupts unless all parties agree to */
-	if (level->usage != 0 && !(level->flags & new->flags & IRQF_SHARED)) {
-		spin_unlock_irqrestore(&level->lock,flags);
-		return -EBUSY;
-	}
-
-	/* add new interrupt at end of irq queue */
-	pp = p;
-	while (*pp)
-		pp = &(*pp)->next;
-
-	*pp = new;
-
-	level->usage++;
-	level->flags = new->flags;
-
-	/* turn the interrupts on */
-	if (level->usage == 1)
-		__clr_MASK(level - frv_irq_levels);
-
-	if (p == pp && group->control)
-		group->control(group, irq & (NR_IRQ_ACTIONS_PER_GROUP - 1), 1);
-
-	spin_unlock_irqrestore(&level->lock, flags);
-	register_irq_proc(irq);
-	return 0;
-}
-
-static struct proc_dir_entry * root_irq_dir;
-static struct proc_dir_entry * irq_dir [NR_IRQS];
-
-#define HEX_DIGITS 8
-
-static unsigned int parse_hex_value (const char __user *buffer,
-				     unsigned long count, unsigned long *ret)
-{
-	unsigned char hexnum [HEX_DIGITS];
-	unsigned long value;
-	int i;
-
-	if (!count)
-		return -EINVAL;
-	if (count > HEX_DIGITS)
-		count = HEX_DIGITS;
-	if (copy_from_user(hexnum, buffer, count))
-		return -EFAULT;
-
-	/*
-	 * Parse the first 8 characters as a hex string, any non-hex char
-	 * is end-of-string. '00e1', 'e1', '00E1', 'E1' are all the same.
-	 */
-	value = 0;
-
-	for (i = 0; i < count; i++) {
-		unsigned int c = hexnum[i];
-
-		switch (c) {
-			case '0' ... '9': c -= '0'; break;
-			case 'a' ... 'f': c -= 'a'-10; break;
-			case 'A' ... 'F': c -= 'A'-10; break;
-		default:
-			goto out;
-		}
-		value = (value << 4) | c;
-	}
-out:
-	*ret = value;
-	return 0;
-}
-
-
-static int prof_cpu_mask_read_proc (char *page, char **start, off_t off,
-			int count, int *eof, void *data)
-{
-	unsigned long *mask = (unsigned long *) data;
-	if (count < HEX_DIGITS+1)
-		return -EINVAL;
-	return sprintf (page, "%08lx\n", *mask);
-}
-
-static int prof_cpu_mask_write_proc (struct file *file, const char __user *buffer,
-					unsigned long count, void *data)
-{
-	unsigned long *mask = (unsigned long *) data, full_count = count, err;
-	unsigned long new_value;
-
-	show_state();
-	err = parse_hex_value(buffer, count, &new_value);
-	if (err)
-		return err;
-
-	*mask = new_value;
-	return full_count;
-}
-
-#define MAX_NAMELEN 10
-
-static void register_irq_proc (unsigned int irq)
-{
-	char name [MAX_NAMELEN];
-
-	if (!root_irq_dir || irq_dir[irq])
-		return;
-
-	memset(name, 0, MAX_NAMELEN);
-	sprintf(name, "%d", irq);
-
-	/* create /proc/irq/1234 */
-	irq_dir[irq] = proc_mkdir(name, root_irq_dir);
-}
-
-unsigned long prof_cpu_mask = -1;
-
-void init_irq_proc (void)
+void __init init_IRQ(void)
 {
-	struct proc_dir_entry *entry;
-	int i;
+	int level;
 
-	/* create /proc/irq */
-	root_irq_dir = proc_mkdir("irq", NULL);
+	for (level = 1; level <= 14; level++)
+		set_irq_chip_and_handler(level, &frv_cpu_pic,
+					 handle_level_irq);
 
-	/* create /proc/irq/prof_cpu_mask */
-	entry = create_proc_entry("prof_cpu_mask", 0600, root_irq_dir);
-	if (!entry)
-	    return;
+	set_irq_handler(IRQ_CPU_TIMER0, handle_edge_irq);
 
-	entry->nlink = 1;
-	entry->data = (void *)&prof_cpu_mask;
-	entry->read_proc = prof_cpu_mask_read_proc;
-	entry->write_proc = prof_cpu_mask_write_proc;
-
-	/*
-	 * Create entries for all existing IRQs.
+	/* set the trigger levels for internal interrupt sources
+	 * - timers all falling-edge
+	 * - ERR0 is rising-edge
+	 * - all others are high-level
 	 */
-	for (i = 0; i < NR_IRQS; i++)
-		register_irq_proc(i);
-}
+	__set_IITMR(0, 0x003f0000);	/* DMA0-3, TIMER0-2 */
+	__set_IITMR(1, 0x20000000);	/* ERR0-1, UART0-1, DMA4-7 */
+
+	/* route internal interrupts */
+	set_IRR(4, IRQ_DMA3_LEVEL, IRQ_DMA2_LEVEL, IRQ_DMA1_LEVEL,
+		IRQ_DMA0_LEVEL);
+	set_IRR(5, 0, IRQ_TIMER2_LEVEL, IRQ_TIMER1_LEVEL, IRQ_TIMER0_LEVEL);
+	set_IRR(6, IRQ_GDBSTUB_LEVEL, IRQ_GDBSTUB_LEVEL,
+		IRQ_UART1_LEVEL, IRQ_UART0_LEVEL);
+	set_IRR(7, IRQ_DMA7_LEVEL, IRQ_DMA6_LEVEL, IRQ_DMA5_LEVEL,
+		IRQ_DMA4_LEVEL);
+
+	/* route external interrupts */
+	set_IRR(2, IRQ_XIRQ7_LEVEL, IRQ_XIRQ6_LEVEL, IRQ_XIRQ5_LEVEL,
+		IRQ_XIRQ4_LEVEL);
+	set_IRR(3, IRQ_XIRQ3_LEVEL, IRQ_XIRQ2_LEVEL, IRQ_XIRQ1_LEVEL,
+		IRQ_XIRQ0_LEVEL);
+
+#if defined(CONFIG_MB93091_VDK)
+	__set_TM1(0x55550000);		/* XIRQ7-0 all active low */
+#elif defined(CONFIG_MB93093_PDK)
+	__set_TM1(0x15550000);		/* XIRQ7 active high, 6-0 all active low */
+#else
+#error dont know external IRQ trigger levels for this setup
+#endif
 
-/*****************************************************************************/
-/*
- * initialise the interrupt system
- */
-void __init init_IRQ(void)
-{
-	route_cpu_irqs();
 	fpga_init();
 #ifdef CONFIG_FUJITSU_MB93493
-	route_mb93493_irqs();
+	mb93493_init();
 #endif
-} /* end init_IRQ() */
+}
diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c
index af08ccd4ed6e..d96a57e5f030 100644
--- a/arch/frv/kernel/setup.c
+++ b/arch/frv/kernel/setup.c
@@ -43,7 +43,6 @@
 #include <asm/mb-regs.h>
 #include <asm/mb93493-regs.h>
 #include <asm/gdb-stub.h>
-#include <asm/irq-routing.h>
 #include <asm/io.h>
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/frv/kernel/time.c b/arch/frv/kernel/time.c
index 68a77fe3bb40..3d0284bccb94 100644
--- a/arch/frv/kernel/time.c
+++ b/arch/frv/kernel/time.c
@@ -26,7 +26,6 @@
 #include <asm/timer-regs.h>
 #include <asm/mb-regs.h>
 #include <asm/mb86943a.h>
-#include <asm/irq-routing.h>
 
 #include <linux/timex.h>
 
diff --git a/arch/frv/mb93090-mb00/pci-irq.c b/arch/frv/mb93090-mb00/pci-irq.c
index 2278c80bd88c..ba587523c015 100644
--- a/arch/frv/mb93090-mb00/pci-irq.c
+++ b/arch/frv/mb93090-mb00/pci-irq.c
@@ -15,7 +15,6 @@
 
 #include <asm/io.h>
 #include <asm/smp.h>
-#include <asm/irq-routing.h>
 
 #include "pci-frv.h"
 
diff --git a/include/asm-frv/cpu-irqs.h b/include/asm-frv/cpu-irqs.h
index 5cd691e1f8c4..478f3498fcfe 100644
--- a/include/asm-frv/cpu-irqs.h
+++ b/include/asm-frv/cpu-irqs.h
@@ -14,36 +14,6 @@
 
 #ifndef __ASSEMBLY__
 
-#include <asm/irq-routing.h>
-
-#define IRQ_BASE_CPU		(NR_IRQ_ACTIONS_PER_GROUP * 0)
-
-/* IRQ IDs presented to drivers */
-enum {
-	IRQ_CPU__UNUSED = IRQ_BASE_CPU,
-	IRQ_CPU_UART0,
-	IRQ_CPU_UART1,
-	IRQ_CPU_TIMER0,
-	IRQ_CPU_TIMER1,
-	IRQ_CPU_TIMER2,
-	IRQ_CPU_DMA0,
-	IRQ_CPU_DMA1,
-	IRQ_CPU_DMA2,
-	IRQ_CPU_DMA3,
-	IRQ_CPU_DMA4,
-	IRQ_CPU_DMA5,
-	IRQ_CPU_DMA6,
-	IRQ_CPU_DMA7,
-	IRQ_CPU_EXTERNAL0,
-	IRQ_CPU_EXTERNAL1,
-	IRQ_CPU_EXTERNAL2,
-	IRQ_CPU_EXTERNAL3,
-	IRQ_CPU_EXTERNAL4,
-	IRQ_CPU_EXTERNAL5,
-	IRQ_CPU_EXTERNAL6,
-	IRQ_CPU_EXTERNAL7,
-};
-
 /* IRQ to level mappings */
 #define IRQ_GDBSTUB_LEVEL	15
 #define IRQ_UART_LEVEL		13
@@ -82,6 +52,30 @@ enum {
 #define IRQ_XIRQ6_LEVEL		7
 #define IRQ_XIRQ7_LEVEL		8
 
+/* IRQ IDs presented to drivers */
+#define IRQ_CPU__UNUSED		IRQ_BASE_CPU
+#define IRQ_CPU_UART0		(IRQ_BASE_CPU + IRQ_UART0_LEVEL)
+#define IRQ_CPU_UART1		(IRQ_BASE_CPU + IRQ_UART1_LEVEL)
+#define IRQ_CPU_TIMER0		(IRQ_BASE_CPU + IRQ_TIMER0_LEVEL)
+#define IRQ_CPU_TIMER1		(IRQ_BASE_CPU + IRQ_TIMER1_LEVEL)
+#define IRQ_CPU_TIMER2		(IRQ_BASE_CPU + IRQ_TIMER2_LEVEL)
+#define IRQ_CPU_DMA0		(IRQ_BASE_CPU + IRQ_DMA0_LEVEL)
+#define IRQ_CPU_DMA1		(IRQ_BASE_CPU + IRQ_DMA1_LEVEL)
+#define IRQ_CPU_DMA2		(IRQ_BASE_CPU + IRQ_DMA2_LEVEL)
+#define IRQ_CPU_DMA3		(IRQ_BASE_CPU + IRQ_DMA3_LEVEL)
+#define IRQ_CPU_DMA4		(IRQ_BASE_CPU + IRQ_DMA4_LEVEL)
+#define IRQ_CPU_DMA5		(IRQ_BASE_CPU + IRQ_DMA5_LEVEL)
+#define IRQ_CPU_DMA6		(IRQ_BASE_CPU + IRQ_DMA6_LEVEL)
+#define IRQ_CPU_DMA7		(IRQ_BASE_CPU + IRQ_DMA7_LEVEL)
+#define IRQ_CPU_EXTERNAL0	(IRQ_BASE_CPU + IRQ_XIRQ0_LEVEL)
+#define IRQ_CPU_EXTERNAL1	(IRQ_BASE_CPU + IRQ_XIRQ1_LEVEL)
+#define IRQ_CPU_EXTERNAL2	(IRQ_BASE_CPU + IRQ_XIRQ2_LEVEL)
+#define IRQ_CPU_EXTERNAL3	(IRQ_BASE_CPU + IRQ_XIRQ3_LEVEL)
+#define IRQ_CPU_EXTERNAL4	(IRQ_BASE_CPU + IRQ_XIRQ4_LEVEL)
+#define IRQ_CPU_EXTERNAL5	(IRQ_BASE_CPU + IRQ_XIRQ5_LEVEL)
+#define IRQ_CPU_EXTERNAL6	(IRQ_BASE_CPU + IRQ_XIRQ6_LEVEL)
+#define IRQ_CPU_EXTERNAL7	(IRQ_BASE_CPU + IRQ_XIRQ7_LEVEL)
+
 #endif /* !__ASSEMBLY__ */
 
 #endif /* _ASM_CPU_IRQS_H */
diff --git a/include/asm-frv/hardirq.h b/include/asm-frv/hardirq.h
index 7581b5a7559a..fc47515822a2 100644
--- a/include/asm-frv/hardirq.h
+++ b/include/asm-frv/hardirq.h
@@ -26,5 +26,10 @@ typedef struct {
 #error SMP not available on FR-V
 #endif /* CONFIG_SMP */
 
+extern atomic_t irq_err_count;
+static inline void ack_bad_irq(int irq)
+{
+	atomic_inc(&irq_err_count);
+}
 
 #endif
diff --git a/include/asm-frv/irq-routing.h b/include/asm-frv/irq-routing.h
deleted file mode 100644
index ac3ab900a1dc..000000000000
--- a/include/asm-frv/irq-routing.h
+++ /dev/null
@@ -1,70 +0,0 @@
-/* irq-routing.h: multiplexed IRQ routing
- *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#ifndef _ASM_IRQ_ROUTING_H
-#define _ASM_IRQ_ROUTING_H
-
-#ifndef __ASSEMBLY__
-
-#include <linux/spinlock.h>
-#include <asm/irq.h>
-
-struct irq_source;
-struct irq_level;
-
-/*
- * IRQ action distribution sets
- */
-struct irq_group {
-	int			first_irq;	/* first IRQ distributed here */
-	void (*control)(struct irq_group *group, int index, int on);
-
-	struct irqaction	*actions[NR_IRQ_ACTIONS_PER_GROUP];	/* IRQ action chains */
-	struct irq_source	*sources[NR_IRQ_ACTIONS_PER_GROUP];	/* IRQ sources */
-	int			disable_cnt[NR_IRQ_ACTIONS_PER_GROUP];	/* disable counts */
-};
-
-/*
- * IRQ source manager
- */
-struct irq_source {
-	struct irq_source	*next;
-	struct irq_level	*level;
-	const char		*muxname;
-	volatile void __iomem	*muxdata;
-	unsigned long		irqmask;
-
-	void (*doirq)(struct irq_source *source);
-};
-
-/*
- * IRQ level management (per CPU IRQ priority / entry vector)
- */
-struct irq_level {
-	int			usage;
-	int			disable_count;
-	unsigned long		flags;		/* current IRQF_DISABLED and IRQF_SHARED settings */
-	spinlock_t		lock;
-	struct irq_source	*sources;
-};
-
-extern struct irq_level frv_irq_levels[16];
-extern struct irq_group *irq_groups[NR_IRQ_GROUPS];
-
-extern void frv_irq_route(struct irq_source *source, int irqlevel);
-extern void frv_irq_route_external(struct irq_source *source, int irq);
-extern void frv_irq_set_group(struct irq_group *group);
-extern void distribute_irqs(struct irq_group *group, unsigned long irqmask);
-extern void route_cpu_irqs(void);
-
-#endif /* !__ASSEMBLY__ */
-
-#endif /* _ASM_IRQ_ROUTING_H */
diff --git a/include/asm-frv/irq.h b/include/asm-frv/irq.h
index 58b619215a50..8fefd6b827aa 100644
--- a/include/asm-frv/irq.h
+++ b/include/asm-frv/irq.h
@@ -1,6 +1,6 @@
 /* irq.h: FRV IRQ definitions
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -12,32 +12,22 @@
 #ifndef _ASM_IRQ_H_
 #define _ASM_IRQ_H_
 
-
-/*
- * the system has an on-CPU PIC and another PIC on the FPGA and other PICs on other peripherals,
- * so we do some routing in irq-routing.[ch] to reduce the number of false-positives seen by
- * drivers
- */
-
 /* this number is used when no interrupt has been assigned */
 #define NO_IRQ				(-1)
 
-#define NR_IRQ_LOG2_ACTIONS_PER_GROUP	5
-#define NR_IRQ_ACTIONS_PER_GROUP	(1 << NR_IRQ_LOG2_ACTIONS_PER_GROUP)
-#define NR_IRQ_GROUPS			4
-#define NR_IRQS				(NR_IRQ_ACTIONS_PER_GROUP * NR_IRQ_GROUPS)
+#define NR_IRQS				48
+#define IRQ_BASE_CPU			(0 * 16)
+#define IRQ_BASE_FPGA			(1 * 16)
+#define IRQ_BASE_MB93493		(2 * 16)
 
 /* probe returns a 32-bit IRQ mask:-/ */
-#define MIN_PROBE_IRQ	(NR_IRQS - 32)
+#define MIN_PROBE_IRQ			(NR_IRQS - 32)
 
+#ifndef __ASSEMBLY__
 static inline int irq_canonicalize(int irq)
 {
 	return irq;
 }
-
-extern void disable_irq_nosync(unsigned int irq);
-extern void disable_irq(unsigned int irq);
-extern void enable_irq(unsigned int irq);
-
+#endif
 
 #endif /* _ASM_IRQ_H_ */
diff --git a/include/asm-frv/mb93091-fpga-irqs.h b/include/asm-frv/mb93091-fpga-irqs.h
index 341bfc52a0eb..19778c5ba9d6 100644
--- a/include/asm-frv/mb93091-fpga-irqs.h
+++ b/include/asm-frv/mb93091-fpga-irqs.h
@@ -12,11 +12,9 @@
 #ifndef _ASM_MB93091_FPGA_IRQS_H
 #define _ASM_MB93091_FPGA_IRQS_H
 
-#ifndef __ASSEMBLY__
-
-#include <asm/irq-routing.h>
+#include <asm/irq.h>
 
-#define IRQ_BASE_FPGA		(NR_IRQ_ACTIONS_PER_GROUP * 1)
+#ifndef __ASSEMBLY__
 
 /* IRQ IDs presented to drivers */
 enum {
diff --git a/include/asm-frv/mb93093-fpga-irqs.h b/include/asm-frv/mb93093-fpga-irqs.h
index 1e0f11c2fcdb..590266b1a6d3 100644
--- a/include/asm-frv/mb93093-fpga-irqs.h
+++ b/include/asm-frv/mb93093-fpga-irqs.h
@@ -12,11 +12,9 @@
 #ifndef _ASM_MB93093_FPGA_IRQS_H
 #define _ASM_MB93093_FPGA_IRQS_H
 
-#ifndef __ASSEMBLY__
-
-#include <asm/irq-routing.h>
+#include <asm/irq.h>
 
-#define IRQ_BASE_FPGA		(NR_IRQ_ACTIONS_PER_GROUP * 1)
+#ifndef __ASSEMBLY__
 
 /* IRQ IDs presented to drivers */
 enum {
diff --git a/include/asm-frv/mb93493-irqs.h b/include/asm-frv/mb93493-irqs.h
index 15096e731325..82c7aeddd333 100644
--- a/include/asm-frv/mb93493-irqs.h
+++ b/include/asm-frv/mb93493-irqs.h
@@ -12,11 +12,9 @@
 #ifndef _ASM_MB93493_IRQS_H
 #define _ASM_MB93493_IRQS_H
 
-#ifndef __ASSEMBLY__
-
-#include <asm/irq-routing.h>
+#include <asm/irq.h>
 
-#define IRQ_BASE_MB93493	(NR_IRQ_ACTIONS_PER_GROUP * 2)
+#ifndef __ASSEMBLY__
 
 /* IRQ IDs presented to drivers */
 enum {
diff --git a/include/asm-frv/mb93493-regs.h b/include/asm-frv/mb93493-regs.h
index c54aa9d14468..8a1f6aac8cf1 100644
--- a/include/asm-frv/mb93493-regs.h
+++ b/include/asm-frv/mb93493-regs.h
@@ -15,6 +15,7 @@
 #include <asm/mb-regs.h>
 #include <asm/mb93493-irqs.h>
 
+#define __addr_MB93493(X)	((volatile unsigned long *)(__region_CS3 + (X)))
 #define __get_MB93493(X)	({ *(volatile unsigned long *)(__region_CS3 + (X)); })
 
 #define __set_MB93493(X,V)						\
@@ -26,6 +27,7 @@ do {									\
 #define __set_MB93493_STSR(X,V)	__set_MB93493(0x3c0 + (X) * 4, (V))
 #define MB93493_STSR_EN
 
+#define __addr_MB93493_IQSR(X)	__addr_MB93493(0x3d0 + (X) * 4)
 #define __get_MB93493_IQSR(X)	__get_MB93493(0x3d0 + (X) * 4)
 #define __set_MB93493_IQSR(X,V)	__set_MB93493(0x3d0 + (X) * 4, (V))
 
-- 
cgit v1.2.3


From af8c65b57aaa4ae321af34dbfc5ca7f5625263fe Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:07 -0700
Subject: [PATCH] FRV: permit __do_IRQ() to be dispensed with

Permit __do_IRQ() to be dispensed with based on a configuration option.

Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/Kconfig    | 4 ++++
 include/linux/irq.h | 6 ++++++
 kernel/irq/handle.c | 2 ++
 3 files changed, 12 insertions(+)

(limited to 'include')

diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 5833808fb823..f7b171b92ea2 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -29,6 +29,10 @@ config GENERIC_HARDIRQS
 	bool
 	default y
 
+config GENERIC_HARDIRQS_NO__DO_IRQ
+	bool
+	default y
+
 config GENERIC_TIME
 	bool
 	default y
diff --git a/include/linux/irq.h b/include/linux/irq.h
index fbf6d901e9c2..48d3cb3b6a47 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -320,7 +320,9 @@ handle_irq_name(void fastcall (*handle)(unsigned int, struct irq_desc *,
  * Monolithic do_IRQ implementation.
  * (is an explicit fastcall, because i386 4KSTACKS calls it from assembly)
  */
+#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
 extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs);
+#endif
 
 /*
  * Architectures call this to let the generic IRQ layer
@@ -332,10 +334,14 @@ static inline void generic_handle_irq(unsigned int irq, struct pt_regs *regs)
 {
 	struct irq_desc *desc = irq_desc + irq;
 
+#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
+	desc->handle_irq(irq, desc, regs);
+#else
 	if (likely(desc->handle_irq))
 		desc->handle_irq(irq, desc, regs);
 	else
 		__do_IRQ(irq, regs);
+#endif
 }
 
 /* Handling of unhandled and spurious interrupts: */
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 48a53f68af96..4c6cdbaed661 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -154,6 +154,7 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct pt_regs *regs,
 	return retval;
 }
 
+#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
 /**
  * __do_IRQ - original all in one highlevel IRQ handler
  * @irq:	the interrupt number
@@ -253,6 +254,7 @@ out:
 
 	return 1;
 }
+#endif
 
 #ifdef CONFIG_TRACE_IRQFLAGS
 
-- 
cgit v1.2.3


From 92fc707208bb2e601c24b5ab65db37bcb361b658 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:07 -0700
Subject: [PATCH] FRV: Fix fls() to handle bit 31 being set correctly

Fix FRV fls() to handle bit 31 being set correctly (it should return 32 not 0).

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/bitops.h | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h
index 980ae1b0cd28..97fb746f76c7 100644
--- a/include/asm-frv/bitops.h
+++ b/include/asm-frv/bitops.h
@@ -161,16 +161,29 @@ static inline int __test_bit(int nr, const volatile void * addr)
 #include <asm-generic/bitops/__ffs.h>
 #include <asm-generic/bitops/find.h>
 
-/*
- * fls: find last bit set.
+/**
+ * fls - find last bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as ffs:
+ * - return 32..1 to indicate bit 31..0 most significant bit set
+ * - return 0 to indicate no bits set
  */
 #define fls(x)						\
 ({							\
 	int bit;					\
 							\
-	asm("scan %1,gr0,%0" : "=r"(bit) : "r"(x));	\
+	asm("	subcc	%1,gr0,gr0,icc0		\n"	\
+	    "	ckne	icc0,cc4		\n"	\
+	    "	cscan.p	%1,gr0,%0	,cc4,#1	\n"	\
+	    "	csub	%0,%0,%0	,cc4,#0	\n"	\
+	    "   csub    %2,%0,%0	,cc4,#1	\n"	\
+	    : "=&r"(bit)				\
+	    : "r"(x), "r"(32)				\
+	    : "icc0", "cc4"				\
+	    );						\
 							\
-	bit ? 33 - bit : bit;				\
+	bit;						\
 })
 
 #include <asm-generic/bitops/fls64.h>
-- 
cgit v1.2.3


From a8ad27d03f17e6154c61e81d4a7028c56ca6390d Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:08 -0700
Subject: [PATCH] FRV: Implement fls64()

Implement fls64() for FRV without recource to conditional jumps.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/bitops.h | 42 +++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 41 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h
index 97fb746f76c7..591eecc1f8cd 100644
--- a/include/asm-frv/bitops.h
+++ b/include/asm-frv/bitops.h
@@ -186,7 +186,47 @@ static inline int __test_bit(int nr, const volatile void * addr)
 	bit;						\
 })
 
-#include <asm-generic/bitops/fls64.h>
+/**
+ * fls64 - find last bit set in a 64-bit value
+ * @n: the value to search
+ *
+ * This is defined the same way as ffs:
+ * - return 64..1 to indicate bit 63..0 most significant bit set
+ * - return 0 to indicate no bits set
+ */
+static inline __attribute__((const))
+int fls64(u64 n)
+{
+	union {
+		u64 ll;
+		struct { u32 h, l; };
+	} _;
+	int bit, x, y;
+
+	_.ll = n;
+
+	asm("	subcc.p		%3,gr0,gr0,icc0		\n"
+	    "	subcc		%4,gr0,gr0,icc1		\n"
+	    "	ckne		icc0,cc4		\n"
+	    "	ckne		icc1,cc5		\n"
+	    "	norcr		cc4,cc5,cc6		\n"
+	    "	csub.p		%0,%0,%0	,cc6,1	\n"
+	    "	orcr		cc5,cc4,cc4		\n"
+	    "	andcr		cc4,cc5,cc4		\n"
+	    "	cscan.p		%3,gr0,%0	,cc4,0	\n"
+	    "   setlos		#64,%1			\n"
+	    "	cscan.p		%4,gr0,%0	,cc4,1	\n"
+	    "   setlos		#32,%2			\n"
+	    "	csub.p		%1,%0,%0	,cc4,0	\n"
+	    "	csub		%2,%0,%0	,cc4,1	\n"
+	    : "=&r"(bit), "=r"(x), "=r"(y)
+	    : "0r"(_.h), "r"(_.l)
+	    : "icc0", "icc1", "cc4", "cc5", "cc6"
+	    );
+	return bit;
+
+}
+
 #include <asm-generic/bitops/sched.h>
 #include <asm-generic/bitops/hweight.h>
 
-- 
cgit v1.2.3


From cf134483b2cd657039b305777215c531a1009947 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:09 -0700
Subject: [PATCH] FRV: Optimise ffs()

Optimise ffs(x) by using fls(x & x - 1) which we optimise to use the SCAN
instruction.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/bitops.h | 33 +++++++++++++++++++++++++++++++--
 1 file changed, 31 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h
index 591eecc1f8cd..1f70d47148bd 100644
--- a/include/asm-frv/bitops.h
+++ b/include/asm-frv/bitops.h
@@ -157,8 +157,6 @@ static inline int __test_bit(int nr, const volatile void * addr)
  __constant_test_bit((nr),(addr)) : \
  __test_bit((nr),(addr)))
 
-#include <asm-generic/bitops/ffs.h>
-#include <asm-generic/bitops/__ffs.h>
 #include <asm-generic/bitops/find.h>
 
 /**
@@ -227,6 +225,37 @@ int fls64(u64 n)
 
 }
 
+/**
+ * ffs - find first bit set
+ * @x: the word to search
+ *
+ * - return 32..1 to indicate bit 31..0 most least significant bit set
+ * - return 0 to indicate no bits set
+ */
+static inline __attribute__((const))
+int ffs(int x)
+{
+	/* Note: (x & -x) gives us a mask that is the least significant
+	 * (rightmost) 1-bit of the value in x.
+	 */
+	return fls(x & -x);
+}
+
+/**
+ * __ffs - find first bit set
+ * @x: the word to search
+ *
+ * - return 31..0 to indicate bit 31..0 most least significant bit set
+ * - if no bits are set in x, the result is undefined
+ */
+static inline __attribute__((const))
+int __ffs(unsigned long x)
+{
+	int bit;
+	asm("scan %1,gr0,%0" : "=r"(bit) : "r"(x & -x));
+	return 31 - bit;
+}
+
 #include <asm-generic/bitops/sched.h>
 #include <asm-generic/bitops/hweight.h>
 
-- 
cgit v1.2.3


From 53e62d3aaa60590d4a69b4e07c29f448b5151047 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 25 Sep 2006 23:32:10 -0700
Subject: [PATCH] Alchemy: Delete unused pt_regs * argument from
 au1xxx_dbdma_chan_alloc

The third argument of au1xxx_dbdma_chan_alloc's callback function is not
used anywhere.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/au1000/common/dbdma.c             | 10 ++++++----
 drivers/ide/mips/au1xxx-ide.c               |  4 ++--
 drivers/mmc/au1xmmc.c                       |  2 +-
 include/asm-mips/mach-au1x00/au1xxx_dbdma.h |  6 +++---
 sound/oss/au1550_ac97.c                     |  6 ++----
 5 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/mips/au1000/common/dbdma.c b/arch/mips/au1000/common/dbdma.c
index 98244d51c154..c4fae8ff4671 100644
--- a/arch/mips/au1000/common/dbdma.c
+++ b/arch/mips/au1000/common/dbdma.c
@@ -230,7 +230,7 @@ EXPORT_SYMBOL(au1xxx_ddma_add_device);
 */
 u32
 au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
-       void (*callback)(int, void *, struct pt_regs *), void *callparam)
+       void (*callback)(int, void *), void *callparam)
 {
 	unsigned long   flags;
 	u32		used, chan, rv;
@@ -248,8 +248,10 @@ au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
 		au1xxx_dbdma_init();
 	dbdma_initialized = 1;
 
-	if ((stp = find_dbdev_id(srcid)) == NULL) return 0;
-	if ((dtp = find_dbdev_id(destid)) == NULL) return 0;
+	if ((stp = find_dbdev_id(srcid)) == NULL)
+		return 0;
+	if ((dtp = find_dbdev_id(destid)) == NULL)
+		return 0;
 
 	used = 0;
 	rv = 0;
@@ -869,7 +871,7 @@ dbdma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	au_sync();
 
 	if (ctp->chan_callback)
-		(ctp->chan_callback)(irq, ctp->chan_callparam, regs);
+		(ctp->chan_callback)(irq, ctp->chan_callparam);
 
 	ctp->cur_ptr = phys_to_virt(DSCR_GET_NXTPTR(dp->dscr_nxtptr));
 	return IRQ_RETVAL(1);
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 71f27e955d87..c7854ea57b52 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -476,13 +476,13 @@ static int auide_dma_lostirq(ide_drive_t *drive)
 	return 0;
 }
 
-static void auide_ddma_tx_callback(int irq, void *param, struct pt_regs *regs)
+static void auide_ddma_tx_callback(int irq, void *param)
 {
 	_auide_hwif *ahwif = (_auide_hwif*)param;
 	ahwif->drive->waiting_for_dma = 0;
 }
 
-static void auide_ddma_rx_callback(int irq, void *param, struct pt_regs *regs)
+static void auide_ddma_rx_callback(int irq, void *param)
 {
 	_auide_hwif *ahwif = (_auide_hwif*)param;
 	ahwif->drive->waiting_for_dma = 0;
diff --git a/drivers/mmc/au1xmmc.c b/drivers/mmc/au1xmmc.c
index fb606165af3b..61268da13957 100644
--- a/drivers/mmc/au1xmmc.c
+++ b/drivers/mmc/au1xmmc.c
@@ -731,7 +731,7 @@ static void au1xmmc_set_ios(struct mmc_host* mmc, struct mmc_ios* ios)
 	}
 }
 
-static void au1xmmc_dma_callback(int irq, void *dev_id, struct pt_regs *regs)
+static void au1xmmc_dma_callback(int irq, void *dev_id)
 {
 	struct au1xmmc_host *host = (struct au1xmmc_host *) dev_id;
 
diff --git a/include/asm-mips/mach-au1x00/au1xxx_dbdma.h b/include/asm-mips/mach-au1x00/au1xxx_dbdma.h
index d5b38a247e5a..eeb0c3115b6a 100644
--- a/include/asm-mips/mach-au1x00/au1xxx_dbdma.h
+++ b/include/asm-mips/mach-au1x00/au1xxx_dbdma.h
@@ -316,7 +316,7 @@ typedef struct dbdma_chan_config {
 	au1x_ddma_desc_t	*chan_desc_base;
 	au1x_ddma_desc_t	*get_ptr, *put_ptr, *cur_ptr;
 	void			*chan_callparam;
-	void (*chan_callback)(int, void *, struct pt_regs *);
+	void (*chan_callback)(int, void *);
 } chan_tab_t;
 
 #define DEV_FLAGS_INUSE		(1 << 0)
@@ -334,8 +334,8 @@ typedef struct dbdma_chan_config {
  * meaningful name.  The 'callback' is called during dma completion
  * interrupt.
  */
-u32 au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
-       void (*callback)(int, void *, struct pt_regs *), void *callparam);
+extern u32 au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
+	void (*callback)(int, void *), void *callparam);
 
 #define DBDMA_MEM_CHAN	DSCR_CMD0_ALWAYS
 
diff --git a/sound/oss/au1550_ac97.c b/sound/oss/au1550_ac97.c
index 4cdb86252d67..219795171c71 100644
--- a/sound/oss/au1550_ac97.c
+++ b/sound/oss/au1550_ac97.c
@@ -719,8 +719,7 @@ prog_dmabuf_dac(struct au1550_state *s)
 }
 
 
-static void
-dac_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+static void dac_dma_interrupt(int irq, void *dev_id)
 {
 	struct au1550_state *s = (struct au1550_state *) dev_id;
 	struct dmabuf  *db = &s->dma_dac;
@@ -754,8 +753,7 @@ dac_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 }
 
 
-static void
-adc_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+static void adc_dma_interrupt(int irq, void *dev_id)
 {
 	struct	au1550_state *s = (struct au1550_state *)dev_id;
 	struct	dmabuf  *dp = &s->dma_adc;
-- 
cgit v1.2.3


From 5f97f7f9400de47ae837170bb274e90ad3934386 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Mon, 25 Sep 2006 23:32:13 -0700
Subject: [PATCH] avr32 architecture

This adds support for the Atmel AVR32 architecture as well as the AT32AP7000
CPU and the AT32STK1000 development board.

AVR32 is a new high-performance 32-bit RISC microprocessor core, designed for
cost-sensitive embedded applications, with particular emphasis on low power
consumption and high code density.  The AVR32 architecture is not binary
compatible with earlier 8-bit AVR architectures.

The AVR32 architecture, including the instruction set, is described by the
AVR32 Architecture Manual, available from

http://www.atmel.com/dyn/resources/prod_documents/doc32000.pdf

The Atmel AT32AP7000 is the first CPU implementing the AVR32 architecture.  It
features a 7-stage pipeline, 16KB instruction and data caches and a full
Memory Management Unit.  It also comes with a large set of integrated
peripherals, many of which are shared with the AT91 ARM-based controllers from
Atmel.

Full data sheet is available from

http://www.atmel.com/dyn/resources/prod_documents/doc32003.pdf

while the CPU core implementation including caches and MMU is documented by
the AVR32 AP Technical Reference, available from

http://www.atmel.com/dyn/resources/prod_documents/doc32001.pdf

Information about the AT32STK1000 development board can be found at

http://www.atmel.com/dyn/products/tools_card.asp?tool_id=3918

including a BSP CD image with an earlier version of this patch, development
tools (binaries and source/patches) and a root filesystem image suitable for
booting from SD card.

Alternatively, there's a preliminary "getting started" guide available at
http://avr32linux.org/twiki/bin/view/Main/GettingStarted which provides links
to the sources and patches you will need in order to set up a cross-compiling
environment for avr32-linux.

This patch, as well as the other patches included with the BSP and the
toolchain patches, is actively supported by Atmel Corporation.

[dmccr@us.ibm.com: Fix more pxx_page macro locations]
[bunk@stusta.de: fix `make defconfig']
Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Dave McCracken <dmccr@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 MAINTAINERS                                      |  17 +
 arch/avr32/Kconfig                               | 196 +++++
 arch/avr32/Kconfig.debug                         |  19 +
 arch/avr32/Makefile                              |  84 +++
 arch/avr32/boards/atstk1000/Makefile             |   2 +
 arch/avr32/boards/atstk1000/atstk1002.c          |  37 +
 arch/avr32/boards/atstk1000/setup.c              |  59 ++
 arch/avr32/boards/atstk1000/spi.c                |  27 +
 arch/avr32/boot/images/Makefile                  |  62 ++
 arch/avr32/boot/u-boot/Makefile                  |   3 +
 arch/avr32/boot/u-boot/empty.S                   |   1 +
 arch/avr32/boot/u-boot/head.S                    |  60 ++
 arch/avr32/configs/atstk1002_defconfig           | 754 ++++++++++++++++++++
 arch/avr32/kernel/Makefile                       |  18 +
 arch/avr32/kernel/asm-offsets.c                  |  25 +
 arch/avr32/kernel/avr32_ksyms.c                  |  55 ++
 arch/avr32/kernel/cpu.c                          | 327 +++++++++
 arch/avr32/kernel/entry-avr32b.S                 | 678 ++++++++++++++++++
 arch/avr32/kernel/head.S                         |  45 ++
 arch/avr32/kernel/init_task.c                    |  38 +
 arch/avr32/kernel/irq.c                          |  71 ++
 arch/avr32/kernel/kprobes.c                      | 270 +++++++
 arch/avr32/kernel/module.c                       | 324 +++++++++
 arch/avr32/kernel/process.c                      | 276 ++++++++
 arch/avr32/kernel/ptrace.c                       | 371 ++++++++++
 arch/avr32/kernel/semaphore.c                    | 148 ++++
 arch/avr32/kernel/setup.c                        | 335 +++++++++
 arch/avr32/kernel/signal.c                       | 328 +++++++++
 arch/avr32/kernel/switch_to.S                    |  35 +
 arch/avr32/kernel/sys_avr32.c                    |  51 ++
 arch/avr32/kernel/syscall-stubs.S                | 102 +++
 arch/avr32/kernel/syscall_table.S                | 289 ++++++++
 arch/avr32/kernel/time.c                         | 238 +++++++
 arch/avr32/kernel/traps.c                        | 425 +++++++++++
 arch/avr32/kernel/vmlinux.lds.c                  | 139 ++++
 arch/avr32/lib/Makefile                          |  10 +
 arch/avr32/lib/__avr32_asr64.S                   |  31 +
 arch/avr32/lib/__avr32_lsl64.S                   |  31 +
 arch/avr32/lib/__avr32_lsr64.S                   |  31 +
 arch/avr32/lib/clear_user.S                      |  76 ++
 arch/avr32/lib/copy_user.S                       | 119 ++++
 arch/avr32/lib/csum_partial.S                    |  47 ++
 arch/avr32/lib/csum_partial_copy_generic.S       |  99 +++
 arch/avr32/lib/delay.c                           |  55 ++
 arch/avr32/lib/findbit.S                         | 154 ++++
 arch/avr32/lib/io-readsl.S                       |  24 +
 arch/avr32/lib/io-readsw.S                       |  43 ++
 arch/avr32/lib/io-writesl.S                      |  20 +
 arch/avr32/lib/io-writesw.S                      |  38 +
 arch/avr32/lib/libgcc.h                          |  33 +
 arch/avr32/lib/longlong.h                        |  98 +++
 arch/avr32/lib/memcpy.S                          |  62 ++
 arch/avr32/lib/memset.S                          |  72 ++
 arch/avr32/lib/strncpy_from_user.S               |  60 ++
 arch/avr32/lib/strnlen_user.S                    |  67 ++
 arch/avr32/mach-at32ap/Makefile                  |   2 +
 arch/avr32/mach-at32ap/at32ap.c                  |  90 +++
 arch/avr32/mach-at32ap/at32ap7000.c              | 866 +++++++++++++++++++++++
 arch/avr32/mach-at32ap/clock.c                   | 148 ++++
 arch/avr32/mach-at32ap/clock.h                   |  30 +
 arch/avr32/mach-at32ap/extint.c                  | 171 +++++
 arch/avr32/mach-at32ap/intc.c                    | 133 ++++
 arch/avr32/mach-at32ap/intc.h                    | 327 +++++++++
 arch/avr32/mach-at32ap/pio.c                     | 118 +++
 arch/avr32/mach-at32ap/pio.h                     | 178 +++++
 arch/avr32/mach-at32ap/sm.c                      | 289 ++++++++
 arch/avr32/mach-at32ap/sm.h                      | 240 +++++++
 arch/avr32/mm/Makefile                           |   6 +
 arch/avr32/mm/cache.c                            | 150 ++++
 arch/avr32/mm/clear_page.S                       |  25 +
 arch/avr32/mm/copy_page.S                        |  28 +
 arch/avr32/mm/dma-coherent.c                     | 139 ++++
 arch/avr32/mm/fault.c                            | 315 +++++++++
 arch/avr32/mm/init.c                             | 480 +++++++++++++
 arch/avr32/mm/ioremap.c                          | 197 ++++++
 arch/avr32/mm/tlb.c                              | 378 ++++++++++
 include/asm-avr32/Kbuild                         |   3 +
 include/asm-avr32/a.out.h                        |  26 +
 include/asm-avr32/addrspace.h                    |  43 ++
 include/asm-avr32/arch-at32ap/at91rm9200_pdc.h   |  36 +
 include/asm-avr32/arch-at32ap/at91rm9200_usart.h | 123 ++++
 include/asm-avr32/arch-at32ap/board.h            |  35 +
 include/asm-avr32/arch-at32ap/init.h             |  21 +
 include/asm-avr32/arch-at32ap/portmux.h          |  16 +
 include/asm-avr32/arch-at32ap/sm.h               |  27 +
 include/asm-avr32/asm.h                          | 102 +++
 include/asm-avr32/atomic.h                       | 201 ++++++
 include/asm-avr32/auxvec.h                       |   4 +
 include/asm-avr32/bitops.h                       | 296 ++++++++
 include/asm-avr32/bug.h                          |  47 ++
 include/asm-avr32/bugs.h                         |  15 +
 include/asm-avr32/byteorder.h                    |  25 +
 include/asm-avr32/cache.h                        |  29 +
 include/asm-avr32/cachectl.h                     |  11 +
 include/asm-avr32/cacheflush.h                   | 129 ++++
 include/asm-avr32/checksum.h                     | 156 ++++
 include/asm-avr32/cputime.h                      |   6 +
 include/asm-avr32/current.h                      |  15 +
 include/asm-avr32/delay.h                        |  26 +
 include/asm-avr32/div64.h                        |   6 +
 include/asm-avr32/dma-mapping.h                  | 320 +++++++++
 include/asm-avr32/dma.h                          |   8 +
 include/asm-avr32/elf.h                          | 110 +++
 include/asm-avr32/emergency-restart.h            |   6 +
 include/asm-avr32/errno.h                        |   6 +
 include/asm-avr32/fcntl.h                        |   6 +
 include/asm-avr32/futex.h                        |   6 +
 include/asm-avr32/hardirq.h                      |  34 +
 include/asm-avr32/hw_irq.h                       |   9 +
 include/asm-avr32/intc.h                         | 128 ++++
 include/asm-avr32/io.h                           | 253 +++++++
 include/asm-avr32/ioctl.h                        |   6 +
 include/asm-avr32/ioctls.h                       |  83 +++
 include/asm-avr32/ipcbuf.h                       |  29 +
 include/asm-avr32/irq.h                          |  10 +
 include/asm-avr32/irqflags.h                     |  68 ++
 include/asm-avr32/kdebug.h                       |  38 +
 include/asm-avr32/kmap_types.h                   |  30 +
 include/asm-avr32/kprobes.h                      |  34 +
 include/asm-avr32/linkage.h                      |   7 +
 include/asm-avr32/local.h                        |   6 +
 include/asm-avr32/mach/serial_at91.h             |  33 +
 include/asm-avr32/mman.h                         |  17 +
 include/asm-avr32/mmu.h                          |  10 +
 include/asm-avr32/mmu_context.h                  | 148 ++++
 include/asm-avr32/module.h                       |  28 +
 include/asm-avr32/msgbuf.h                       |  31 +
 include/asm-avr32/mutex.h                        |   9 +
 include/asm-avr32/namei.h                        |   7 +
 include/asm-avr32/numnodes.h                     |   7 +
 include/asm-avr32/ocd.h                          |  78 ++
 include/asm-avr32/page.h                         | 112 +++
 include/asm-avr32/param.h                        |  23 +
 include/asm-avr32/pci.h                          |   8 +
 include/asm-avr32/percpu.h                       |   6 +
 include/asm-avr32/pgalloc.h                      |  96 +++
 include/asm-avr32/pgtable-2level.h               |  47 ++
 include/asm-avr32/pgtable.h                      | 408 +++++++++++
 include/asm-avr32/poll.h                         |  27 +
 include/asm-avr32/posix_types.h                  | 129 ++++
 include/asm-avr32/processor.h                    | 147 ++++
 include/asm-avr32/ptrace.h                       | 154 ++++
 include/asm-avr32/resource.h                     |   6 +
 include/asm-avr32/scatterlist.h                  |  21 +
 include/asm-avr32/sections.h                     |   6 +
 include/asm-avr32/semaphore.h                    | 109 +++
 include/asm-avr32/sembuf.h                       |  25 +
 include/asm-avr32/setup.h                        | 141 ++++
 include/asm-avr32/shmbuf.h                       |  42 ++
 include/asm-avr32/shmparam.h                     |   6 +
 include/asm-avr32/sigcontext.h                   |  34 +
 include/asm-avr32/siginfo.h                      |   6 +
 include/asm-avr32/signal.h                       | 168 +++++
 include/asm-avr32/socket.h                       |  53 ++
 include/asm-avr32/sockios.h                      |  12 +
 include/asm-avr32/stat.h                         |  79 +++
 include/asm-avr32/statfs.h                       |   6 +
 include/asm-avr32/string.h                       |  17 +
 include/asm-avr32/sysreg.h                       | 332 +++++++++
 include/asm-avr32/system.h                       | 155 ++++
 include/asm-avr32/termbits.h                     | 173 +++++
 include/asm-avr32/termios.h                      |  80 +++
 include/asm-avr32/thread_info.h                  | 106 +++
 include/asm-avr32/timex.h                        |  40 ++
 include/asm-avr32/tlb.h                          |  32 +
 include/asm-avr32/tlbflush.h                     |  40 ++
 include/asm-avr32/topology.h                     |   6 +
 include/asm-avr32/traps.h                        |  23 +
 include/asm-avr32/types.h                        |  70 ++
 include/asm-avr32/uaccess.h                      | 335 +++++++++
 include/asm-avr32/ucontext.h                     |  12 +
 include/asm-avr32/unaligned.h                    |  25 +
 include/asm-avr32/unistd.h                       | 387 ++++++++++
 include/asm-avr32/user.h                         |  65 ++
 include/linux/elf-em.h                           |   1 +
 lib/Kconfig.debug                                |   4 +-
 176 files changed, 18124 insertions(+), 2 deletions(-)
 create mode 100644 arch/avr32/Kconfig
 create mode 100644 arch/avr32/Kconfig.debug
 create mode 100644 arch/avr32/Makefile
 create mode 100644 arch/avr32/boards/atstk1000/Makefile
 create mode 100644 arch/avr32/boards/atstk1000/atstk1002.c
 create mode 100644 arch/avr32/boards/atstk1000/setup.c
 create mode 100644 arch/avr32/boards/atstk1000/spi.c
 create mode 100644 arch/avr32/boot/images/Makefile
 create mode 100644 arch/avr32/boot/u-boot/Makefile
 create mode 100644 arch/avr32/boot/u-boot/empty.S
 create mode 100644 arch/avr32/boot/u-boot/head.S
 create mode 100644 arch/avr32/configs/atstk1002_defconfig
 create mode 100644 arch/avr32/kernel/Makefile
 create mode 100644 arch/avr32/kernel/asm-offsets.c
 create mode 100644 arch/avr32/kernel/avr32_ksyms.c
 create mode 100644 arch/avr32/kernel/cpu.c
 create mode 100644 arch/avr32/kernel/entry-avr32b.S
 create mode 100644 arch/avr32/kernel/head.S
 create mode 100644 arch/avr32/kernel/init_task.c
 create mode 100644 arch/avr32/kernel/irq.c
 create mode 100644 arch/avr32/kernel/kprobes.c
 create mode 100644 arch/avr32/kernel/module.c
 create mode 100644 arch/avr32/kernel/process.c
 create mode 100644 arch/avr32/kernel/ptrace.c
 create mode 100644 arch/avr32/kernel/semaphore.c
 create mode 100644 arch/avr32/kernel/setup.c
 create mode 100644 arch/avr32/kernel/signal.c
 create mode 100644 arch/avr32/kernel/switch_to.S
 create mode 100644 arch/avr32/kernel/sys_avr32.c
 create mode 100644 arch/avr32/kernel/syscall-stubs.S
 create mode 100644 arch/avr32/kernel/syscall_table.S
 create mode 100644 arch/avr32/kernel/time.c
 create mode 100644 arch/avr32/kernel/traps.c
 create mode 100644 arch/avr32/kernel/vmlinux.lds.c
 create mode 100644 arch/avr32/lib/Makefile
 create mode 100644 arch/avr32/lib/__avr32_asr64.S
 create mode 100644 arch/avr32/lib/__avr32_lsl64.S
 create mode 100644 arch/avr32/lib/__avr32_lsr64.S
 create mode 100644 arch/avr32/lib/clear_user.S
 create mode 100644 arch/avr32/lib/copy_user.S
 create mode 100644 arch/avr32/lib/csum_partial.S
 create mode 100644 arch/avr32/lib/csum_partial_copy_generic.S
 create mode 100644 arch/avr32/lib/delay.c
 create mode 100644 arch/avr32/lib/findbit.S
 create mode 100644 arch/avr32/lib/io-readsl.S
 create mode 100644 arch/avr32/lib/io-readsw.S
 create mode 100644 arch/avr32/lib/io-writesl.S
 create mode 100644 arch/avr32/lib/io-writesw.S
 create mode 100644 arch/avr32/lib/libgcc.h
 create mode 100644 arch/avr32/lib/longlong.h
 create mode 100644 arch/avr32/lib/memcpy.S
 create mode 100644 arch/avr32/lib/memset.S
 create mode 100644 arch/avr32/lib/strncpy_from_user.S
 create mode 100644 arch/avr32/lib/strnlen_user.S
 create mode 100644 arch/avr32/mach-at32ap/Makefile
 create mode 100644 arch/avr32/mach-at32ap/at32ap.c
 create mode 100644 arch/avr32/mach-at32ap/at32ap7000.c
 create mode 100644 arch/avr32/mach-at32ap/clock.c
 create mode 100644 arch/avr32/mach-at32ap/clock.h
 create mode 100644 arch/avr32/mach-at32ap/extint.c
 create mode 100644 arch/avr32/mach-at32ap/intc.c
 create mode 100644 arch/avr32/mach-at32ap/intc.h
 create mode 100644 arch/avr32/mach-at32ap/pio.c
 create mode 100644 arch/avr32/mach-at32ap/pio.h
 create mode 100644 arch/avr32/mach-at32ap/sm.c
 create mode 100644 arch/avr32/mach-at32ap/sm.h
 create mode 100644 arch/avr32/mm/Makefile
 create mode 100644 arch/avr32/mm/cache.c
 create mode 100644 arch/avr32/mm/clear_page.S
 create mode 100644 arch/avr32/mm/copy_page.S
 create mode 100644 arch/avr32/mm/dma-coherent.c
 create mode 100644 arch/avr32/mm/fault.c
 create mode 100644 arch/avr32/mm/init.c
 create mode 100644 arch/avr32/mm/ioremap.c
 create mode 100644 arch/avr32/mm/tlb.c
 create mode 100644 include/asm-avr32/Kbuild
 create mode 100644 include/asm-avr32/a.out.h
 create mode 100644 include/asm-avr32/addrspace.h
 create mode 100644 include/asm-avr32/arch-at32ap/at91rm9200_pdc.h
 create mode 100644 include/asm-avr32/arch-at32ap/at91rm9200_usart.h
 create mode 100644 include/asm-avr32/arch-at32ap/board.h
 create mode 100644 include/asm-avr32/arch-at32ap/init.h
 create mode 100644 include/asm-avr32/arch-at32ap/portmux.h
 create mode 100644 include/asm-avr32/arch-at32ap/sm.h
 create mode 100644 include/asm-avr32/asm.h
 create mode 100644 include/asm-avr32/atomic.h
 create mode 100644 include/asm-avr32/auxvec.h
 create mode 100644 include/asm-avr32/bitops.h
 create mode 100644 include/asm-avr32/bug.h
 create mode 100644 include/asm-avr32/bugs.h
 create mode 100644 include/asm-avr32/byteorder.h
 create mode 100644 include/asm-avr32/cache.h
 create mode 100644 include/asm-avr32/cachectl.h
 create mode 100644 include/asm-avr32/cacheflush.h
 create mode 100644 include/asm-avr32/checksum.h
 create mode 100644 include/asm-avr32/cputime.h
 create mode 100644 include/asm-avr32/current.h
 create mode 100644 include/asm-avr32/delay.h
 create mode 100644 include/asm-avr32/div64.h
 create mode 100644 include/asm-avr32/dma-mapping.h
 create mode 100644 include/asm-avr32/dma.h
 create mode 100644 include/asm-avr32/elf.h
 create mode 100644 include/asm-avr32/emergency-restart.h
 create mode 100644 include/asm-avr32/errno.h
 create mode 100644 include/asm-avr32/fcntl.h
 create mode 100644 include/asm-avr32/futex.h
 create mode 100644 include/asm-avr32/hardirq.h
 create mode 100644 include/asm-avr32/hw_irq.h
 create mode 100644 include/asm-avr32/intc.h
 create mode 100644 include/asm-avr32/io.h
 create mode 100644 include/asm-avr32/ioctl.h
 create mode 100644 include/asm-avr32/ioctls.h
 create mode 100644 include/asm-avr32/ipcbuf.h
 create mode 100644 include/asm-avr32/irq.h
 create mode 100644 include/asm-avr32/irqflags.h
 create mode 100644 include/asm-avr32/kdebug.h
 create mode 100644 include/asm-avr32/kmap_types.h
 create mode 100644 include/asm-avr32/kprobes.h
 create mode 100644 include/asm-avr32/linkage.h
 create mode 100644 include/asm-avr32/local.h
 create mode 100644 include/asm-avr32/mach/serial_at91.h
 create mode 100644 include/asm-avr32/mman.h
 create mode 100644 include/asm-avr32/mmu.h
 create mode 100644 include/asm-avr32/mmu_context.h
 create mode 100644 include/asm-avr32/module.h
 create mode 100644 include/asm-avr32/msgbuf.h
 create mode 100644 include/asm-avr32/mutex.h
 create mode 100644 include/asm-avr32/namei.h
 create mode 100644 include/asm-avr32/numnodes.h
 create mode 100644 include/asm-avr32/ocd.h
 create mode 100644 include/asm-avr32/page.h
 create mode 100644 include/asm-avr32/param.h
 create mode 100644 include/asm-avr32/pci.h
 create mode 100644 include/asm-avr32/percpu.h
 create mode 100644 include/asm-avr32/pgalloc.h
 create mode 100644 include/asm-avr32/pgtable-2level.h
 create mode 100644 include/asm-avr32/pgtable.h
 create mode 100644 include/asm-avr32/poll.h
 create mode 100644 include/asm-avr32/posix_types.h
 create mode 100644 include/asm-avr32/processor.h
 create mode 100644 include/asm-avr32/ptrace.h
 create mode 100644 include/asm-avr32/resource.h
 create mode 100644 include/asm-avr32/scatterlist.h
 create mode 100644 include/asm-avr32/sections.h
 create mode 100644 include/asm-avr32/semaphore.h
 create mode 100644 include/asm-avr32/sembuf.h
 create mode 100644 include/asm-avr32/setup.h
 create mode 100644 include/asm-avr32/shmbuf.h
 create mode 100644 include/asm-avr32/shmparam.h
 create mode 100644 include/asm-avr32/sigcontext.h
 create mode 100644 include/asm-avr32/siginfo.h
 create mode 100644 include/asm-avr32/signal.h
 create mode 100644 include/asm-avr32/socket.h
 create mode 100644 include/asm-avr32/sockios.h
 create mode 100644 include/asm-avr32/stat.h
 create mode 100644 include/asm-avr32/statfs.h
 create mode 100644 include/asm-avr32/string.h
 create mode 100644 include/asm-avr32/sysreg.h
 create mode 100644 include/asm-avr32/system.h
 create mode 100644 include/asm-avr32/termbits.h
 create mode 100644 include/asm-avr32/termios.h
 create mode 100644 include/asm-avr32/thread_info.h
 create mode 100644 include/asm-avr32/timex.h
 create mode 100644 include/asm-avr32/tlb.h
 create mode 100644 include/asm-avr32/tlbflush.h
 create mode 100644 include/asm-avr32/topology.h
 create mode 100644 include/asm-avr32/traps.h
 create mode 100644 include/asm-avr32/types.h
 create mode 100644 include/asm-avr32/uaccess.h
 create mode 100644 include/asm-avr32/ucontext.h
 create mode 100644 include/asm-avr32/unaligned.h
 create mode 100644 include/asm-avr32/unistd.h
 create mode 100644 include/asm-avr32/user.h

(limited to 'include')

diff --git a/MAINTAINERS b/MAINTAINERS
index 23348c0d37bc..767a43434ae8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -443,6 +443,23 @@ W:	http://people.redhat.com/sgrubb/audit/
 T:	git kernel.org:/pub/scm/linux/kernel/git/dwmw2/audit-2.6.git
 S:	Maintained
 
+AVR32 ARCHITECTURE
+P:	Atmel AVR32 Support Team
+M:	avr32@atmel.com
+P:	Haavard Skinnemoen
+M:	hskinnemoen@atmel.com
+W:	http://www.atmel.com/products/AVR32/
+W:	http://avr32linux.org/
+W:	http://avrfreaks.net/
+S:	Supported
+
+AVR32/AT32AP MACHINE SUPPORT
+P:	Atmel AVR32 Support Team
+M:	avr32@atmel.com
+P:	Haavard Skinnemoen
+M:	hskinnemoen@atmel.com
+S:	Supported
+
 AX.25 NETWORK LAYER
 P:	Ralf Baechle
 M:	ralf@linux-mips.org
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
new file mode 100644
index 000000000000..5f1694eea842
--- /dev/null
+++ b/arch/avr32/Kconfig
@@ -0,0 +1,196 @@
+#
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/kconfig-language.txt.
+#
+
+mainmenu "Linux Kernel Configuration"
+
+config AVR32
+	bool
+	default y
+	# With EMBEDDED=n, we get lots of stuff automatically selected
+	# that we usually don't need on AVR32.
+	select EMBEDDED
+	help
+	  AVR32 is a high-performance 32-bit RISC microprocessor core,
+	  designed for cost-sensitive embedded applications, with particular
+	  emphasis on low power consumption and high code density.
+
+	  There is an AVR32 Linux project with a web page at
+	  http://avr32linux.org/.
+
+config UID16
+	bool
+
+config GENERIC_HARDIRQS
+	bool
+	default y
+
+config HARDIRQS_SW_RESEND
+	bool
+	default y
+
+config GENERIC_IRQ_PROBE
+	bool
+	default y
+
+config RWSEM_GENERIC_SPINLOCK
+	bool
+	default y
+
+config GENERIC_TIME
+	bool
+	default y
+
+config RWSEM_XCHGADD_ALGORITHM
+	bool
+
+config GENERIC_BUST_SPINLOCK
+	bool
+
+config GENERIC_HWEIGHT
+	bool
+	default y
+
+config GENERIC_CALIBRATE_DELAY
+	bool
+	default y
+
+source "init/Kconfig"
+
+menu "System Type and features"
+
+config SUBARCH_AVR32B
+	bool
+config MMU
+	bool
+config PERFORMANCE_COUNTERS
+	bool
+
+config PLATFORM_AT32AP
+	bool
+	select SUBARCH_AVR32B
+	select MMU
+	select PERFORMANCE_COUNTERS
+
+choice
+	prompt "AVR32 CPU type"
+	default CPU_AT32AP7000
+
+config CPU_AT32AP7000
+	bool "AT32AP7000"
+	select PLATFORM_AT32AP
+endchoice
+
+#
+# CPU Daughterboards for ATSTK1000
+config BOARD_ATSTK1002
+	bool
+
+choice
+	prompt "AVR32 board type"
+	default BOARD_ATSTK1000
+
+config BOARD_ATSTK1000
+	bool "ATSTK1000 evaluation board"
+	select BOARD_ATSTK1002 if CPU_AT32AP7000
+endchoice
+
+choice
+	prompt "Boot loader type"
+	default LOADER_U_BOOT
+
+config	LOADER_U_BOOT
+	bool "U-Boot (or similar) bootloader"
+endchoice
+
+config LOAD_ADDRESS
+	hex
+	default 0x10000000 if LOADER_U_BOOT=y && CPU_AT32AP7000=y
+
+config ENTRY_ADDRESS
+	hex
+	default 0x90000000 if LOADER_U_BOOT=y && CPU_AT32AP7000=y
+
+config PHYS_OFFSET
+	hex
+	default 0x10000000 if CPU_AT32AP7000=y
+
+source "kernel/Kconfig.preempt"
+
+config HAVE_ARCH_BOOTMEM_NODE
+	bool
+	default n
+
+config ARCH_HAVE_MEMORY_PRESENT
+	bool
+	default n
+
+config NEED_NODE_MEMMAP_SIZE
+	bool
+	default n
+
+config ARCH_FLATMEM_ENABLE
+	bool
+	default y
+
+config ARCH_DISCONTIGMEM_ENABLE
+	bool
+	default n
+
+config ARCH_SPARSEMEM_ENABLE
+	bool
+	default n
+
+source "mm/Kconfig"
+
+config OWNERSHIP_TRACE
+	bool "Ownership trace support"
+	default y
+	help
+	  Say Y to generate an Ownership Trace message on every context switch,
+	  enabling Nexus-compliant debuggers to keep track of the PID of the
+	  currently executing task.
+
+# FPU emulation goes here
+
+source "kernel/Kconfig.hz"
+
+config CMDLINE
+	string "Default kernel command line"
+	default ""
+	help
+	  If you don't have a boot loader capable of passing a command line string
+	  to the kernel, you may specify one here. As a minimum, you should specify
+	  the memory size and the root device (e.g., mem=8M, root=/dev/nfs).
+
+endmenu
+
+menu "Bus options"
+
+config PCI
+	bool
+
+source "drivers/pci/Kconfig"
+
+source "drivers/pcmcia/Kconfig"
+
+endmenu
+
+menu "Executable file formats"
+source "fs/Kconfig.binfmt"
+endmenu
+
+source "net/Kconfig"
+
+source "drivers/Kconfig"
+
+source "fs/Kconfig"
+
+source "arch/avr32/Kconfig.debug"
+
+source "security/Kconfig"
+
+source "crypto/Kconfig"
+
+source "lib/Kconfig"
diff --git a/arch/avr32/Kconfig.debug b/arch/avr32/Kconfig.debug
new file mode 100644
index 000000000000..64ace00fe6cb
--- /dev/null
+++ b/arch/avr32/Kconfig.debug
@@ -0,0 +1,19 @@
+menu "Kernel hacking"
+
+config TRACE_IRQFLAGS_SUPPORT
+	bool
+	default y
+
+source "lib/Kconfig.debug"
+
+config KPROBES
+	bool "Kprobes"
+	depends on DEBUG_KERNEL
+	help
+	  Kprobes allows you to trap at almost any kernel address and
+          execute a callback function.  register_kprobe() establishes
+          a probepoint and specifies the callback.  Kprobes is useful
+          for kernel debugging, non-intrusive instrumentation and testing.
+          If in doubt, say "N".
+
+endmenu
diff --git a/arch/avr32/Makefile b/arch/avr32/Makefile
new file mode 100644
index 000000000000..cefc95a73980
--- /dev/null
+++ b/arch/avr32/Makefile
@@ -0,0 +1,84 @@
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+# Copyright (C) 2004-2006 Atmel Corporation.
+
+# Default target when executing plain make
+.PHONY: all
+all: uImage vmlinux.elf linux.lst
+
+KBUILD_DEFCONFIG	:= atstk1002_defconfig
+
+CFLAGS		+= -pipe -fno-builtin -mno-pic
+AFLAGS		+= -mrelax -mno-pic
+CFLAGS_MODULE	+= -mno-relax
+LDFLAGS_vmlinux	+= --relax
+
+cpuflags-$(CONFIG_CPU_AP7000)	+= -mcpu=ap7000
+
+CFLAGS		+= $(cpuflags-y)
+AFLAGS		+= $(cpuflags-y)
+
+CHECKFLAGS	+= -D__avr32__
+
+LIBGCC		:= $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
+
+head-$(CONFIG_LOADER_U_BOOT)		+= arch/avr32/boot/u-boot/head.o
+head-y					+= arch/avr32/kernel/head.o
+core-$(CONFIG_PLATFORM_AT32AP)		+= arch/avr32/mach-at32ap/
+core-$(CONFIG_BOARD_ATSTK1000)		+= arch/avr32/boards/atstk1000/
+core-$(CONFIG_LOADER_U_BOOT)		+= arch/avr32/boot/u-boot/
+core-y					+= arch/avr32/kernel/
+core-y					+= arch/avr32/mm/
+libs-y					+= arch/avr32/lib/ #$(LIBGCC)
+
+archincdir-$(CONFIG_PLATFORM_AT32AP)	:= arch-at32ap
+
+include/asm-avr32/.arch: $(wildcard include/config/platform/*.h) include/config/auto.conf
+	@echo '  SYMLINK include/asm-avr32/arch -> include/asm-avr32/$(archincdir-y)'
+ifneq ($(KBUILD_SRC),)
+	$(Q)mkdir -p include/asm-avr32
+	$(Q)ln -fsn $(srctree)/include/asm-avr32/$(archincdir-y) include/asm-avr32/arch
+else
+	$(Q)ln -fsn $(archincdir-y) include/asm-avr32/arch
+endif
+	@touch $@
+
+archprepare: include/asm-avr32/.arch
+
+BOOT_TARGETS := vmlinux.elf vmlinux.bin uImage uImage.srec
+
+.PHONY: $(BOOT_TARGETS) install
+
+boot := arch/$(ARCH)/boot/images
+
+             KBUILD_IMAGE := $(boot)/uImage
+vmlinux.elf: KBUILD_IMAGE := $(boot)/vmlinux.elf
+vmlinux.cso: KBUILD_IMAGE := $(boot)/vmlinux.cso
+uImage.srec: KBUILD_IMAGE := $(boot)/uImage.srec
+uImage:      KBUILD_IMAGE := $(boot)/uImage
+
+quiet_cmd_listing = LST     $@
+      cmd_listing = avr32-linux-objdump $(OBJDUMPFLAGS) -lS $< > $@
+quiet_cmd_disasm  = DIS     $@
+      cmd_disasm  = avr32-linux-objdump $(OBJDUMPFLAGS) -d $< > $@
+
+vmlinux.elf vmlinux.bin uImage.srec uImage vmlinux.cso: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+
+install: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@
+
+linux.s: vmlinux
+	$(call if_changed,disasm)
+
+linux.lst: vmlinux
+	$(call if_changed,listing)
+
+define archhelp
+  @echo '* vmlinux.elf		- ELF image with load address 0'
+  @echo '  vmlinux.cso		- PathFinder CSO image'
+  @echo '  uImage		- Create a bootable image for U-Boot'
+endef
diff --git a/arch/avr32/boards/atstk1000/Makefile b/arch/avr32/boards/atstk1000/Makefile
new file mode 100644
index 000000000000..c3e36ece8651
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/Makefile
@@ -0,0 +1,2 @@
+obj-y				+= setup.o spi.o
+obj-$(CONFIG_BOARD_ATSTK1002)	+= atstk1002.o
diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c
new file mode 100644
index 000000000000..49164e9aadd6
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/atstk1002.c
@@ -0,0 +1,37 @@
+/*
+ * ATSTK1002 daughterboard-specific init code
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+
+#include <asm/arch/board.h>
+
+struct eth_platform_data __initdata eth0_data = {
+	.valid		= 1,
+	.mii_phy_addr	= 0x10,
+	.is_rmii	= 0,
+	.hw_addr	= { 0x6a, 0x87, 0x71, 0x14, 0xcd, 0xcb },
+};
+
+extern struct lcdc_platform_data atstk1000_fb0_data;
+
+static int __init atstk1002_init(void)
+{
+	at32_add_system_devices();
+
+	at32_add_device_usart(1);	/* /dev/ttyS0 */
+	at32_add_device_usart(2);	/* /dev/ttyS1 */
+	at32_add_device_usart(3);	/* /dev/ttyS2 */
+
+	at32_add_device_eth(0, &eth0_data);
+	at32_add_device_spi(0);
+	at32_add_device_lcdc(0, &atstk1000_fb0_data);
+
+	return 0;
+}
+postcore_initcall(atstk1002_init);
diff --git a/arch/avr32/boards/atstk1000/setup.c b/arch/avr32/boards/atstk1000/setup.c
new file mode 100644
index 000000000000..191ab85de9a3
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/setup.c
@@ -0,0 +1,59 @@
+/*
+ * ATSTK1000 board-specific setup code.
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/bootmem.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/linkage.h>
+
+#include <asm/setup.h>
+
+#include <asm/arch/board.h>
+
+/* Initialized by bootloader-specific startup code. */
+struct tag *bootloader_tags __initdata;
+
+struct lcdc_platform_data __initdata atstk1000_fb0_data;
+
+asmlinkage void __init board_early_init(void)
+{
+	extern void sdram_init(void);
+
+#ifdef CONFIG_LOADER_STANDALONE
+	sdram_init();
+#endif
+}
+
+void __init board_setup_fbmem(unsigned long fbmem_start,
+			      unsigned long fbmem_size)
+{
+	if (!fbmem_size)
+		return;
+
+	if (!fbmem_start) {
+		void *fbmem;
+
+		fbmem = alloc_bootmem_low_pages(fbmem_size);
+		fbmem_start = __pa(fbmem);
+	} else {
+		pg_data_t *pgdat;
+
+		for_each_online_pgdat(pgdat) {
+			if (fbmem_start >= pgdat->bdata->node_boot_start
+			    && fbmem_start <= pgdat->bdata->node_low_pfn)
+				reserve_bootmem_node(pgdat, fbmem_start,
+						     fbmem_size);
+		}
+	}
+
+	printk("%luKiB framebuffer memory at address 0x%08lx\n",
+	       fbmem_size >> 10, fbmem_start);
+	atstk1000_fb0_data.fbmem_start = fbmem_start;
+	atstk1000_fb0_data.fbmem_size = fbmem_size;
+}
diff --git a/arch/avr32/boards/atstk1000/spi.c b/arch/avr32/boards/atstk1000/spi.c
new file mode 100644
index 000000000000..567726c82c6e
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/spi.c
@@ -0,0 +1,27 @@
+/*
+ * ATSTK1000 SPI devices
+ *
+ * Copyright (C) 2005 Atmel Norway
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/device.h>
+#include <linux/spi/spi.h>
+
+static struct spi_board_info spi_board_info[] __initdata = {
+	{
+		.modalias	= "ltv350qv",
+		.max_speed_hz	= 16000000,
+		.bus_num	= 0,
+		.chip_select	= 1,
+	},
+};
+
+static int board_init_spi(void)
+{
+	spi_register_board_info(spi_board_info, ARRAY_SIZE(spi_board_info));
+	return 0;
+}
+arch_initcall(board_init_spi);
diff --git a/arch/avr32/boot/images/Makefile b/arch/avr32/boot/images/Makefile
new file mode 100644
index 000000000000..ccd74eeecec3
--- /dev/null
+++ b/arch/avr32/boot/images/Makefile
@@ -0,0 +1,62 @@
+#
+# Copyright (C) 2004-2006 Atmel Corporation
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+
+MKIMAGE		:= $(srctree)/scripts/mkuboot.sh
+
+extra-y		:= vmlinux.bin vmlinux.gz
+
+OBJCOPYFLAGS_vmlinux.bin := -O binary
+$(obj)/vmlinux.bin: vmlinux FORCE
+	$(call if_changed,objcopy)
+
+$(obj)/vmlinux.gz: $(obj)/vmlinux.bin FORCE
+	$(call if_changed,gzip)
+
+quiet_cmd_uimage = UIMAGE $@
+      cmd_uimage = $(CONFIG_SHELL) $(MKIMAGE) -A avr32 -O linux -T kernel	\
+		-C gzip -a $(CONFIG_LOAD_ADDRESS) -e $(CONFIG_ENTRY_ADDRESS)	\
+		-n 'Linux-$(KERNELRELEASE)' -d $< $@
+
+targets += uImage uImage.srec
+$(obj)/uImage: $(obj)/vmlinux.gz
+	$(call if_changed,uimage)
+	@echo '  Image $@ is ready'
+
+OBJCOPYFLAGS_uImage.srec := -I binary -O srec
+$(obj)/uImage.srec: $(obj)/uImage
+	$(call if_changed,objcopy)
+
+OBJCOPYFLAGS_vmlinux.elf := --change-section-lma .text-0x80000000 \
+			    --change-section-lma __ex_table-0x80000000 \
+			    --change-section-lma .rodata-0x80000000 \
+			    --change-section-lma .data-0x80000000 \
+			    --change-section-lma .init-0x80000000 \
+			    --change-section-lma .bss-0x80000000 \
+			    --change-section-lma .initrd-0x80000000 \
+			    --change-section-lma __param-0x80000000 \
+			    --change-section-lma __ksymtab-0x80000000 \
+			    --change-section-lma __ksymtab_gpl-0x80000000 \
+			    --change-section-lma __kcrctab-0x80000000 \
+			    --change-section-lma __kcrctab_gpl-0x80000000 \
+			    --change-section-lma __ksymtab_strings-0x80000000 \
+			    --change-section-lma .got-0x80000000 \
+			    --set-start 0xa0000000
+$(obj)/vmlinux.elf: vmlinux FORCE
+	$(call if_changed,objcopy)
+
+quiet_cmd_sfdwarf = SFDWARF $@
+      cmd_sfdwarf = sfdwarf $< TO $@ GNUAVR IW $(SFDWARF_FLAGS) > $(obj)/sfdwarf.log
+
+$(obj)/vmlinux.cso: $(obj)/vmlinux.elf FORCE
+	$(call if_changed,sfdwarf)
+
+install: $(BOOTIMAGE)
+	sh $(srctree)/install-kernel.sh $<
+
+# Generated files to be removed upon make clean
+clean-files	:= vmlinux* uImage uImage.srec
diff --git a/arch/avr32/boot/u-boot/Makefile b/arch/avr32/boot/u-boot/Makefile
new file mode 100644
index 000000000000..125ddc96c275
--- /dev/null
+++ b/arch/avr32/boot/u-boot/Makefile
@@ -0,0 +1,3 @@
+extra-y		:= head.o
+
+obj-y		:= empty.o
diff --git a/arch/avr32/boot/u-boot/empty.S b/arch/avr32/boot/u-boot/empty.S
new file mode 100644
index 000000000000..8ac91a5f12f0
--- /dev/null
+++ b/arch/avr32/boot/u-boot/empty.S
@@ -0,0 +1 @@
+/* Empty file */
diff --git a/arch/avr32/boot/u-boot/head.S b/arch/avr32/boot/u-boot/head.S
new file mode 100644
index 000000000000..4488fa27fe94
--- /dev/null
+++ b/arch/avr32/boot/u-boot/head.S
@@ -0,0 +1,60 @@
+/*
+ * Startup code for use with the u-boot bootloader.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/setup.h>
+
+	/*
+	 * The kernel is loaded where we want it to be and all caches
+	 * have just been flushed. We get two parameters from u-boot:
+	 *
+	 * r12 contains a magic number (ATAG_MAGIC)
+	 * r11 points to a tag table providing information about
+	 *     the system.
+	 */
+	.section .init.text,"ax"
+	.global _start
+_start:
+	/* Check if the boot loader actually provided a tag table */
+	lddpc	r0, magic_number
+	cp.w	r12, r0
+	brne	no_tag_table
+
+	/* Initialize .bss */
+	lddpc	r2, bss_start_addr
+	lddpc   r3, end_addr
+	mov	r0, 0
+	mov	r1, 0
+1:      st.d    r2++, r0
+	cp      r2, r3
+	brlo    1b
+
+	/*
+	 * Save the tag table address for later use. This must be done
+	 * _after_ .bss has been initialized...
+	 */
+	lddpc	r0, tag_table_addr
+	st.w	r0[0], r11
+
+	/* Jump to loader-independent setup code */
+	rjmp	kernel_entry
+
+	.align	2
+magic_number:
+	.long	ATAG_MAGIC
+tag_table_addr:
+	.long	bootloader_tags
+bss_start_addr:
+	.long   __bss_start
+end_addr:
+	.long   _end
+
+no_tag_table:
+	sub	r12, pc, (. - 2f)
+	bral	panic
+2:	.asciz	"Boot loader didn't provide correct magic number\n"
diff --git a/arch/avr32/configs/atstk1002_defconfig b/arch/avr32/configs/atstk1002_defconfig
new file mode 100644
index 000000000000..1d22255009fd
--- /dev/null
+++ b/arch/avr32/configs/atstk1002_defconfig
@@ -0,0 +1,754 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.18-rc1
+# Tue Jul 11 12:41:36 2006
+#
+CONFIG_AVR32=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_HARDIRQS_SW_RESEND=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_SWAP=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_BASE_FULL is not set
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+CONFIG_SHMEM=y
+# CONFIG_SLAB is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=1
+CONFIG_SLOB=y
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+
+#
+# Block layer
+#
+# CONFIG_BLK_DEV_IO_TRACE is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+
+#
+# System Type and features
+#
+CONFIG_SUBARCH_AVR32B=y
+CONFIG_MMU=y
+CONFIG_PERFORMANCE_COUNTERS=y
+CONFIG_PLATFORM_AT32AP=y
+CONFIG_CPU_AT32AP7000=y
+CONFIG_BOARD_ATSTK1002=y
+CONFIG_BOARD_ATSTK1000=y
+CONFIG_LOADER_U_BOOT=y
+CONFIG_LOAD_ADDRESS=0x10000000
+CONFIG_ENTRY_ADDRESS=0x90000000
+CONFIG_PHYS_OFFSET=0x10000000
+CONFIG_PREEMPT_NONE=y
+# CONFIG_PREEMPT_VOLUNTARY is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_HAVE_ARCH_BOOTMEM_NODE is not set
+# CONFIG_ARCH_HAVE_MEMORY_PRESENT is not set
+# CONFIG_NEED_NODE_MEMMAP_SIZE is not set
+CONFIG_ARCH_FLATMEM_ENABLE=y
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+# CONFIG_ARCH_SPARSEMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+# CONFIG_OWNERSHIP_TRACE is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+CONFIG_CMDLINE=""
+
+#
+# Bus options
+#
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_NET_TCPPROBE is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+# CONFIG_PREVENT_FIRMWARE_BUILD is not set
+# CONFIG_FW_LOADER is not set
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_SYS_HYPERVISOR is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=m
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+CONFIG_BLK_DEV_NBD=m
+CONFIG_BLK_DEV_RAM=m
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_INITRD=y
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+CONFIG_DUMMY=y
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+CONFIG_MACB=y
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+CONFIG_PPP=m
+# CONFIG_PPP_MULTILINK is not set
+# CONFIG_PPP_FILTER is not set
+CONFIG_PPP_ASYNC=m
+# CONFIG_PPP_SYNC_TTY is not set
+CONFIG_PPP_DEFLATE=m
+# CONFIG_PPP_BSDCOMP is not set
+# CONFIG_PPP_MPPE is not set
+# CONFIG_PPPOE is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_AT91=y
+CONFIG_SERIAL_AT91_CONSOLE=y
+# CONFIG_SERIAL_AT91_TTYAT is not set
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+# CONFIG_LEGACY_PTYS is not set
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# SPI support
+#
+CONFIG_SPI=y
+# CONFIG_SPI_DEBUG is not set
+CONFIG_SPI_MASTER=y
+
+#
+# SPI Master Controller Drivers
+#
+CONFIG_SPI_ATMEL=m
+# CONFIG_SPI_BITBANG is not set
+
+#
+# SPI Protocol Masters
+#
+
+#
+# Dallas's 1-wire bus
+#
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+CONFIG_VIDEO_V4L2=y
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FIRMWARE_EDID is not set
+CONFIG_FB=m
+CONFIG_FB_CFB_FILLRECT=m
+CONFIG_FB_CFB_COPYAREA=m
+CONFIG_FB_CFB_IMAGEBLIT=m
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+CONFIG_FB_SIDSA=m
+CONFIG_FB_SIDSA_DEFAULT_BPP=24
+# CONFIG_FB_S1D13XXX is not set
+# CONFIG_FB_VIRTUAL is not set
+
+#
+# Logo configuration
+#
+# CONFIG_LOGO is not set
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+# CONFIG_BACKLIGHT_CLASS_DEVICE is not set
+CONFIG_LCD_CLASS_DEVICE=m
+CONFIG_LCD_DEVICE=y
+CONFIG_LCD_LTV350QV=m
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# DMA Engine support
+#
+# CONFIG_DMA_ENGINE is not set
+
+#
+# DMA Clients
+#
+
+#
+# DMA Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+CONFIG_MINIX_FS=m
+CONFIG_ROMFS_FS=m
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+CONFIG_CONFIGFS_FS=m
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+CONFIG_CIFS=m
+# CONFIG_CIFS_STATS is not set
+# CONFIG_CIFS_WEAK_PW_HASH is not set
+# CONFIG_CIFS_XATTR is not set
+# CONFIG_CIFS_DEBUG2 is not set
+# CONFIG_CIFS_EXPERIMENTAL is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=m
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=m
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+CONFIG_NLS_CODEPAGE_850=m
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+CONFIG_NLS_ISO8859_1=m
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=m
+
+#
+# Kernel hacking
+#
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
+CONFIG_PRINTK_TIME=y
+CONFIG_MAGIC_SYSRQ=y
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_DEBUG_KERNEL=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_RWSEMS is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+CONFIG_DEBUG_BUGVERBOSE=y
+# CONFIG_DEBUG_INFO is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_DEBUG_VM is not set
+CONFIG_FRAME_POINTER=y
+# CONFIG_UNWIND_INFO is not set
+CONFIG_FORCED_INLINING=y
+# CONFIG_RCU_TORTURE_TEST is not set
+CONFIG_KPROBES=y
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+CONFIG_CRC_CCITT=m
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=m
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=m
+CONFIG_ZLIB_DEFLATE=m
diff --git a/arch/avr32/kernel/Makefile b/arch/avr32/kernel/Makefile
new file mode 100644
index 000000000000..90e5afff54a2
--- /dev/null
+++ b/arch/avr32/kernel/Makefile
@@ -0,0 +1,18 @@
+#
+# Makefile for the Linux/AVR32 kernel.
+#
+
+extra-y				:= head.o vmlinux.lds
+
+obj-$(CONFIG_SUBARCH_AVR32B)	+= entry-avr32b.o
+obj-y				+= syscall_table.o syscall-stubs.o irq.o
+obj-y				+= setup.o traps.o semaphore.o ptrace.o
+obj-y				+= signal.o sys_avr32.o process.o time.o
+obj-y				+= init_task.o switch_to.o cpu.o
+obj-$(CONFIG_MODULES)		+= module.o avr32_ksyms.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o
+
+USE_STANDARD_AS_RULE		:= true
+
+%.lds: %.lds.c FORCE
+	$(call if_changed_dep,cpp_lds_S)
diff --git a/arch/avr32/kernel/asm-offsets.c b/arch/avr32/kernel/asm-offsets.c
new file mode 100644
index 000000000000..97d865865667
--- /dev/null
+++ b/arch/avr32/kernel/asm-offsets.c
@@ -0,0 +1,25 @@
+/*
+ * Generate definitions needed by assembly language modules.
+ * This code generates raw asm output which is post-processed
+ * to extract and format the required data.
+ */
+
+#include <linux/thread_info.h>
+
+#define DEFINE(sym, val) \
+        asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+
+#define BLANK() asm volatile("\n->" : : )
+
+#define OFFSET(sym, str, mem) \
+        DEFINE(sym, offsetof(struct str, mem));
+
+void foo(void)
+{
+	OFFSET(TI_task, thread_info, task);
+	OFFSET(TI_exec_domain, thread_info, exec_domain);
+	OFFSET(TI_flags, thread_info, flags);
+	OFFSET(TI_cpu, thread_info, cpu);
+	OFFSET(TI_preempt_count, thread_info, preempt_count);
+	OFFSET(TI_restart_block, thread_info, restart_block);
+}
diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c
new file mode 100644
index 000000000000..04f767a272b7
--- /dev/null
+++ b/arch/avr32/kernel/avr32_ksyms.c
@@ -0,0 +1,55 @@
+/*
+ * Export AVR32-specific functions for loadable modules.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+
+#include <asm/checksum.h>
+#include <asm/uaccess.h>
+#include <asm/delay.h>
+
+/*
+ * GCC functions
+ */
+extern unsigned long long __avr32_lsl64(unsigned long long u, unsigned long b);
+extern unsigned long long __avr32_lsr64(unsigned long long u, unsigned long b);
+extern unsigned long long __avr32_asr64(unsigned long long u, unsigned long b);
+EXPORT_SYMBOL(__avr32_lsl64);
+EXPORT_SYMBOL(__avr32_lsr64);
+EXPORT_SYMBOL(__avr32_asr64);
+
+/*
+ * String functions
+ */
+EXPORT_SYMBOL(memset);
+EXPORT_SYMBOL(memcpy);
+
+/*
+ * Userspace access stuff.
+ */
+EXPORT_SYMBOL(copy_from_user);
+EXPORT_SYMBOL(copy_to_user);
+EXPORT_SYMBOL(__copy_user);
+EXPORT_SYMBOL(strncpy_from_user);
+EXPORT_SYMBOL(__strncpy_from_user);
+EXPORT_SYMBOL(clear_user);
+EXPORT_SYMBOL(__clear_user);
+EXPORT_SYMBOL(csum_partial);
+EXPORT_SYMBOL(csum_partial_copy_generic);
+
+/* Delay loops (lib/delay.S) */
+EXPORT_SYMBOL(__ndelay);
+EXPORT_SYMBOL(__udelay);
+EXPORT_SYMBOL(__const_udelay);
+
+/* Bit operations (lib/findbit.S) */
+EXPORT_SYMBOL(find_first_zero_bit);
+EXPORT_SYMBOL(find_next_zero_bit);
+EXPORT_SYMBOL(find_first_bit);
+EXPORT_SYMBOL(find_next_bit);
+EXPORT_SYMBOL(generic_find_next_zero_le_bit);
diff --git a/arch/avr32/kernel/cpu.c b/arch/avr32/kernel/cpu.c
new file mode 100644
index 000000000000..342452ba2049
--- /dev/null
+++ b/arch/avr32/kernel/cpu.c
@@ -0,0 +1,327 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/sysdev.h>
+#include <linux/seq_file.h>
+#include <linux/cpu.h>
+#include <linux/percpu.h>
+#include <linux/param.h>
+#include <linux/errno.h>
+
+#include <asm/setup.h>
+#include <asm/sysreg.h>
+
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+
+#ifdef CONFIG_PERFORMANCE_COUNTERS
+
+/*
+ * XXX: If/when a SMP-capable implementation of AVR32 will ever be
+ * made, we must make sure that the code executes on the correct CPU.
+ */
+static ssize_t show_pc0event(struct sys_device *dev, char *buf)
+{
+	unsigned long pccr;
+
+	pccr = sysreg_read(PCCR);
+	return sprintf(buf, "0x%lx\n", (pccr >> 12) & 0x3f);
+}
+static ssize_t store_pc0event(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf || val > 0x3f)
+		return -EINVAL;
+	val = (val << 12) | (sysreg_read(PCCR) & 0xfffc0fff);
+	sysreg_write(PCCR, val);
+	return count;
+}
+static ssize_t show_pc0count(struct sys_device *dev, char *buf)
+{
+	unsigned long pcnt0;
+
+	pcnt0 = sysreg_read(PCNT0);
+	return sprintf(buf, "%lu\n", pcnt0);
+}
+static ssize_t store_pc0count(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	sysreg_write(PCNT0, val);
+
+	return count;
+}
+
+static ssize_t show_pc1event(struct sys_device *dev, char *buf)
+{
+	unsigned long pccr;
+
+	pccr = sysreg_read(PCCR);
+	return sprintf(buf, "0x%lx\n", (pccr >> 18) & 0x3f);
+}
+static ssize_t store_pc1event(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf || val > 0x3f)
+		return -EINVAL;
+	val = (val << 18) | (sysreg_read(PCCR) & 0xff03ffff);
+	sysreg_write(PCCR, val);
+	return count;
+}
+static ssize_t show_pc1count(struct sys_device *dev, char *buf)
+{
+	unsigned long pcnt1;
+
+	pcnt1 = sysreg_read(PCNT1);
+	return sprintf(buf, "%lu\n", pcnt1);
+}
+static ssize_t store_pc1count(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	sysreg_write(PCNT1, val);
+
+	return count;
+}
+
+static ssize_t show_pccycles(struct sys_device *dev, char *buf)
+{
+	unsigned long pccnt;
+
+	pccnt = sysreg_read(PCCNT);
+	return sprintf(buf, "%lu\n", pccnt);
+}
+static ssize_t store_pccycles(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	sysreg_write(PCCNT, val);
+
+	return count;
+}
+
+static ssize_t show_pcenable(struct sys_device *dev, char *buf)
+{
+	unsigned long pccr;
+
+	pccr = sysreg_read(PCCR);
+	return sprintf(buf, "%c\n", (pccr & 1)?'1':'0');
+}
+static ssize_t store_pcenable(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long pccr, val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	if (val)
+		val = 1;
+
+	pccr = sysreg_read(PCCR);
+	pccr = (pccr & ~1UL) | val;
+	sysreg_write(PCCR, pccr);
+
+	return count;
+}
+
+static SYSDEV_ATTR(pc0event, 0600, show_pc0event, store_pc0event);
+static SYSDEV_ATTR(pc0count, 0600, show_pc0count, store_pc0count);
+static SYSDEV_ATTR(pc1event, 0600, show_pc1event, store_pc1event);
+static SYSDEV_ATTR(pc1count, 0600, show_pc1count, store_pc1count);
+static SYSDEV_ATTR(pccycles, 0600, show_pccycles, store_pccycles);
+static SYSDEV_ATTR(pcenable, 0600, show_pcenable, store_pcenable);
+
+#endif /* CONFIG_PERFORMANCE_COUNTERS */
+
+static int __init topology_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		struct cpu *c = &per_cpu(cpu_devices, cpu);
+
+		register_cpu(c, cpu);
+
+#ifdef CONFIG_PERFORMANCE_COUNTERS
+		sysdev_create_file(&c->sysdev, &attr_pc0event);
+		sysdev_create_file(&c->sysdev, &attr_pc0count);
+		sysdev_create_file(&c->sysdev, &attr_pc1event);
+		sysdev_create_file(&c->sysdev, &attr_pc1count);
+		sysdev_create_file(&c->sysdev, &attr_pccycles);
+		sysdev_create_file(&c->sysdev, &attr_pcenable);
+#endif
+	}
+
+	return 0;
+}
+
+subsys_initcall(topology_init);
+
+static const char *cpu_names[] = {
+	"Morgan",
+	"AP7000",
+};
+#define NR_CPU_NAMES ARRAY_SIZE(cpu_names)
+
+static const char *arch_names[] = {
+	"AVR32A",
+	"AVR32B",
+};
+#define NR_ARCH_NAMES ARRAY_SIZE(arch_names)
+
+static const char *mmu_types[] = {
+	"No MMU",
+	"ITLB and DTLB",
+	"Shared TLB",
+	"MPU"
+};
+
+void __init setup_processor(void)
+{
+	unsigned long config0, config1;
+	unsigned cpu_id, cpu_rev, arch_id, arch_rev, mmu_type;
+	unsigned tmp;
+
+	config0 = sysreg_read(CONFIG0); /* 0x0000013e; */
+	config1 = sysreg_read(CONFIG1); /* 0x01f689a2; */
+	cpu_id = config0 >> 24;
+	cpu_rev = (config0 >> 16) & 0xff;
+	arch_id = (config0 >> 13) & 0x07;
+	arch_rev = (config0 >> 10) & 0x07;
+	mmu_type = (config0 >> 7) & 0x03;
+
+	boot_cpu_data.arch_type = arch_id;
+	boot_cpu_data.cpu_type = cpu_id;
+	boot_cpu_data.arch_revision = arch_rev;
+	boot_cpu_data.cpu_revision = cpu_rev;
+	boot_cpu_data.tlb_config = mmu_type;
+
+	tmp = (config1 >> 13) & 0x07;
+	if (tmp) {
+		boot_cpu_data.icache.ways = 1 << ((config1 >> 10) & 0x07);
+		boot_cpu_data.icache.sets = 1 << ((config1 >> 16) & 0x0f);
+		boot_cpu_data.icache.linesz = 1 << (tmp + 1);
+	}
+	tmp = (config1 >> 3) & 0x07;
+	if (tmp) {
+		boot_cpu_data.dcache.ways = 1 << (config1 & 0x07);
+		boot_cpu_data.dcache.sets = 1 << ((config1 >> 6) & 0x0f);
+		boot_cpu_data.dcache.linesz = 1 << (tmp + 1);
+	}
+
+	if ((cpu_id >= NR_CPU_NAMES) || (arch_id >= NR_ARCH_NAMES)) {
+		printk ("Unknown CPU configuration (ID %02x, arch %02x), "
+			"continuing anyway...\n",
+			cpu_id, arch_id);
+		return;
+	}
+
+	printk ("CPU: %s [%02x] revision %d (%s revision %d)\n",
+		cpu_names[cpu_id], cpu_id, cpu_rev,
+		arch_names[arch_id], arch_rev);
+	printk ("CPU: MMU configuration: %s\n", mmu_types[mmu_type]);
+	printk ("CPU: features:");
+	if (config0 & (1 << 6))
+		printk(" fpu");
+	if (config0 & (1 << 5))
+		printk(" java");
+	if (config0 & (1 << 4))
+		printk(" perfctr");
+	if (config0 & (1 << 3))
+		printk(" ocd");
+	printk("\n");
+}
+
+#ifdef CONFIG_PROC_FS
+static int c_show(struct seq_file *m, void *v)
+{
+	unsigned int icache_size, dcache_size;
+	unsigned int cpu = smp_processor_id();
+
+	icache_size = boot_cpu_data.icache.ways *
+		boot_cpu_data.icache.sets *
+		boot_cpu_data.icache.linesz;
+	dcache_size = boot_cpu_data.dcache.ways *
+		boot_cpu_data.dcache.sets *
+		boot_cpu_data.dcache.linesz;
+
+	seq_printf(m, "processor\t: %d\n", cpu);
+
+	if (boot_cpu_data.arch_type < NR_ARCH_NAMES)
+		seq_printf(m, "cpu family\t: %s revision %d\n",
+			   arch_names[boot_cpu_data.arch_type],
+			   boot_cpu_data.arch_revision);
+	if (boot_cpu_data.cpu_type < NR_CPU_NAMES)
+		seq_printf(m, "cpu type\t: %s revision %d\n",
+			   cpu_names[boot_cpu_data.cpu_type],
+			   boot_cpu_data.cpu_revision);
+
+	seq_printf(m, "i-cache\t\t: %dK (%u ways x %u sets x %u)\n",
+		   icache_size >> 10,
+		   boot_cpu_data.icache.ways,
+		   boot_cpu_data.icache.sets,
+		   boot_cpu_data.icache.linesz);
+	seq_printf(m, "d-cache\t\t: %dK (%u ways x %u sets x %u)\n",
+		   dcache_size >> 10,
+		   boot_cpu_data.dcache.ways,
+		   boot_cpu_data.dcache.sets,
+		   boot_cpu_data.dcache.linesz);
+	seq_printf(m, "bogomips\t: %lu.%02lu\n",
+		   boot_cpu_data.loops_per_jiffy / (500000/HZ),
+		   (boot_cpu_data.loops_per_jiffy / (5000/HZ)) % 100);
+
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	return *pos < 1 ? (void *)1 : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return NULL;
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+
+}
+
+struct seq_operations cpuinfo_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= c_show
+};
+#endif /* CONFIG_PROC_FS */
diff --git a/arch/avr32/kernel/entry-avr32b.S b/arch/avr32/kernel/entry-avr32b.S
new file mode 100644
index 000000000000..eeb66792bc37
--- /dev/null
+++ b/arch/avr32/kernel/entry-avr32b.S
@@ -0,0 +1,678 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * This file contains the low-level entry-points into the kernel, that is,
+ * exception handlers, debug trap handlers, interrupt handlers and the
+ * system call handler.
+ */
+#include <linux/errno.h>
+
+#include <asm/asm.h>
+#include <asm/hardirq.h>
+#include <asm/irq.h>
+#include <asm/ocd.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/ptrace.h>
+#include <asm/sysreg.h>
+#include <asm/thread_info.h>
+#include <asm/unistd.h>
+
+#ifdef CONFIG_PREEMPT
+# define preempt_stop		mask_interrupts
+#else
+# define preempt_stop
+# define fault_resume_kernel	fault_restore_all
+#endif
+
+#define __MASK(x)	((1 << (x)) - 1)
+#define IRQ_MASK	((__MASK(SOFTIRQ_BITS) << SOFTIRQ_SHIFT) | \
+			 (__MASK(HARDIRQ_BITS) << HARDIRQ_SHIFT))
+
+	.section .ex.text,"ax",@progbits
+	.align	2
+exception_vectors:
+	bral	handle_critical
+	.align	2
+	bral	handle_critical
+	.align	2
+	bral	do_bus_error_write
+	.align	2
+	bral	do_bus_error_read
+	.align	2
+	bral	do_nmi_ll
+	.align	2
+	bral	handle_address_fault
+	.align	2
+	bral	handle_protection_fault
+	.align	2
+	bral	handle_debug
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	do_fpe_ll
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	handle_address_fault
+	.align	2
+	bral	handle_address_fault
+	.align	2
+	bral	handle_protection_fault
+	.align	2
+	bral	handle_protection_fault
+	.align	2
+	bral	do_dtlb_modified
+
+	/*
+	 * r0 :	PGD/PT/PTE
+	 * r1 : Offending address
+	 * r2 : Scratch register
+	 * r3 : Cause (5, 12 or 13)
+	 */
+#define	tlbmiss_save	pushm	r0-r3
+#define tlbmiss_restore	popm	r0-r3
+
+	.section .tlbx.ex.text,"ax",@progbits
+	.global	itlb_miss
+itlb_miss:
+	tlbmiss_save
+	rjmp	tlb_miss_common
+
+	.section .tlbr.ex.text,"ax",@progbits
+dtlb_miss_read:
+	tlbmiss_save
+	rjmp	tlb_miss_common
+
+	.section .tlbw.ex.text,"ax",@progbits
+dtlb_miss_write:
+	tlbmiss_save
+
+	.global	tlb_miss_common
+tlb_miss_common:
+	mfsr	r0, SYSREG_PTBR
+	mfsr	r1, SYSREG_TLBEAR
+
+	/* Is it the vmalloc space? */
+	bld	r1, 31
+	brcs	handle_vmalloc_miss
+
+	/* First level lookup */
+pgtbl_lookup:
+	lsr	r2, r1, PGDIR_SHIFT
+	ld.w	r0, r0[r2 << 2]
+	bld	r0, _PAGE_BIT_PRESENT
+	brcc	page_table_not_present
+
+	/* TODO: Check access rights on page table if necessary */
+
+	/* Translate to virtual address in P1. */
+	andl	r0, 0xf000
+	sbr	r0, 31
+
+	/* Second level lookup */
+	lsl	r1, (32 - PGDIR_SHIFT)
+	lsr	r1, (32 - PGDIR_SHIFT) + PAGE_SHIFT
+	add	r2, r0, r1 << 2
+	ld.w	r1, r2[0]
+	bld	r1, _PAGE_BIT_PRESENT
+	brcc	page_not_present
+
+	/* Mark the page as accessed */
+	sbr	r1, _PAGE_BIT_ACCESSED
+	st.w	r2[0], r1
+
+	/* Drop software flags */
+	andl	r1, _PAGE_FLAGS_HARDWARE_MASK & 0xffff
+	mtsr	SYSREG_TLBELO, r1
+
+	/* Figure out which entry we want to replace */
+	mfsr	r0, SYSREG_TLBARLO
+	clz	r2, r0
+	brcc	1f
+	mov	r1, -1			/* All entries have been accessed, */
+	mtsr	SYSREG_TLBARLO, r1	/* so reset TLBAR */
+	mov	r2, 0			/* and start at 0 */
+1:	mfsr	r1, SYSREG_MMUCR
+	lsl	r2, 14
+	andl	r1, 0x3fff, COH
+	or	r1, r2
+	mtsr	SYSREG_MMUCR, r1
+
+	tlbw
+
+	tlbmiss_restore
+	rete
+
+handle_vmalloc_miss:
+	/* Simply do the lookup in init's page table */
+	mov	r0, lo(swapper_pg_dir)
+	orh	r0, hi(swapper_pg_dir)
+	rjmp	pgtbl_lookup
+
+
+	/* ---                    System Call                    --- */
+
+	.section .scall.text,"ax",@progbits
+system_call:
+	pushm	r12		/* r12_orig */
+	stmts	--sp, r0-lr
+	zero_fp
+	mfsr	r0, SYSREG_RAR_SUP
+	mfsr	r1, SYSREG_RSR_SUP
+	stm	--sp, r0-r1
+
+	/* check for syscall tracing */
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+	bld	r1, TIF_SYSCALL_TRACE
+	brcs	syscall_trace_enter
+
+syscall_trace_cont:
+	cp.w	r8, NR_syscalls
+	brhs	syscall_badsys
+
+	lddpc   lr, syscall_table_addr
+	ld.w    lr, lr[r8 << 2]
+	mov	r8, r5		/* 5th argument (6th is pushed by stub) */
+	icall   lr
+
+	.global	syscall_return
+syscall_return:
+	get_thread_info r0
+	mask_interrupts		/* make sure we don't miss an interrupt
+				   setting need_resched or sigpending
+				   between sampling and the rets */
+
+	/* Store the return value so that the correct value is loaded below */
+	stdsp   sp[REG_R12], r12
+
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_ALLWORK_MASK, COH
+	brne	syscall_exit_work
+
+syscall_exit_cont:
+	popm	r8-r9
+	mtsr	SYSREG_RAR_SUP, r8
+	mtsr	SYSREG_RSR_SUP, r9
+	ldmts	sp++, r0-lr
+	sub	sp, -4		/* r12_orig */
+	rets
+
+	.align	2
+syscall_table_addr:
+	.long   sys_call_table
+
+syscall_badsys:
+	mov	r12, -ENOSYS
+	rjmp	syscall_return
+
+	.global ret_from_fork
+ret_from_fork:
+	rcall   schedule_tail
+
+	/* check for syscall tracing */
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_ALLWORK_MASK, COH
+	brne	syscall_exit_work
+	rjmp    syscall_exit_cont
+
+syscall_trace_enter:
+	pushm	r8-r12
+	rcall	syscall_trace
+	popm	r8-r12
+	rjmp	syscall_trace_cont
+
+syscall_exit_work:
+	bld	r1, TIF_SYSCALL_TRACE
+	brcc	1f
+	unmask_interrupts
+	rcall	syscall_trace
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+
+1:	bld	r1, TIF_NEED_RESCHED
+	brcc	2f
+	unmask_interrupts
+	rcall	schedule
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+2:	mov	r2, _TIF_SIGPENDING | _TIF_RESTORE_SIGMASK
+	tst	r1, r2
+	breq	3f
+	unmask_interrupts
+	mov	r12, sp
+	mov	r11, r0
+	rcall	do_notify_resume
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+3:	bld	r1, TIF_BREAKPOINT
+	brcc	syscall_exit_cont
+	mfsr	r3, SYSREG_TLBEHI
+	lddsp	r2, sp[REG_PC]
+	andl	r3, 0xff, COH
+	lsl	r3, 1
+	sbr	r3, 30
+	sbr	r3, 0
+	mtdr	DBGREG_BWA2A, r2
+	mtdr	DBGREG_BWC2A, r3
+	rjmp	syscall_exit_cont
+
+
+	/* The slow path of the TLB miss handler */
+page_table_not_present:
+page_not_present:
+	tlbmiss_restore
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_page_fault
+	rjmp	ret_from_exception
+
+	/* This function expects to find offending PC in SYSREG_RAR_EX */
+save_full_context_ex:
+	mfsr	r8, SYSREG_RSR_EX
+	mov	r12, r8
+	andh	r8, (MODE_MASK >> 16), COH
+	mfsr	r11, SYSREG_RAR_EX
+	brne	2f
+
+1:	pushm	r11, r12	/* PC and SR */
+	unmask_exceptions
+	ret	r12
+
+2:	sub	r10, sp, -(FRAME_SIZE_FULL - REG_LR)
+	stdsp	sp[4], r10	/* replace saved SP */
+	rjmp	1b
+
+	/* Low-level exception handlers */
+handle_critical:
+	pushm	r12
+	pushm	r0-r12
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_critical_exception
+
+	/* We should never get here... */
+bad_return:
+	sub	r12, pc, (. - 1f)
+	bral	panic
+	.align	2
+1:	.asciz	"Return from critical exception!"
+
+	.align	1
+do_bus_error_write:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mov	r11, 1
+	rjmp	1f
+
+do_bus_error_read:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mov	r11, 0
+1:	mfsr	r12, SYSREG_BEAR
+	mov	r10, sp
+	rcall	do_bus_error
+	rjmp	ret_from_exception
+
+	.align	1
+do_nmi_ll:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	/* FIXME: Make sure RAR_NMI and RSR_NMI are pushed instead of *_EX */
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_nmi
+	rjmp	bad_return
+
+handle_address_fault:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_address_exception
+	rjmp	ret_from_exception
+
+handle_protection_fault:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_page_fault
+	rjmp	ret_from_exception
+
+	.align	1
+do_illegal_opcode_ll:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_illegal_opcode
+	rjmp	ret_from_exception
+
+do_dtlb_modified:
+	pushm	r0-r3
+	mfsr	r1, SYSREG_TLBEAR
+	mfsr	r0, SYSREG_PTBR
+	lsr	r2, r1, PGDIR_SHIFT
+	ld.w	r0, r0[r2 << 2]
+	lsl	r1, (32 - PGDIR_SHIFT)
+	lsr	r1, (32 - PGDIR_SHIFT) + PAGE_SHIFT
+
+	/* Translate to virtual address in P1 */
+	andl	r0, 0xf000
+	sbr	r0, 31
+	add	r2, r0, r1 << 2
+	ld.w	r3, r2[0]
+	sbr	r3, _PAGE_BIT_DIRTY
+	mov	r0, r3
+	st.w	r2[0], r3
+
+	/* The page table is up-to-date. Update the TLB entry as well */
+	andl	r0, lo(_PAGE_FLAGS_HARDWARE_MASK)
+	mtsr	SYSREG_TLBELO, r0
+
+	/* MMUCR[DRP] is updated automatically, so let's go... */
+	tlbw
+
+	popm	r0-r3
+	rete
+
+do_fpe_ll:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	unmask_interrupts
+	mov	r12, 26
+	mov	r11, sp
+	rcall	do_fpe
+	rjmp	ret_from_exception
+
+ret_from_exception:
+	mask_interrupts
+	lddsp	r4, sp[REG_SR]
+	andh	r4, (MODE_MASK >> 16), COH
+	brne	fault_resume_kernel
+
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_WORK_MASK, COH
+	brne	fault_exit_work
+
+fault_resume_user:
+	popm	r8-r9
+	mask_exceptions
+	mtsr	SYSREG_RAR_EX, r8
+	mtsr	SYSREG_RSR_EX, r9
+	ldmts	sp++, r0-lr
+	sub	sp, -4
+	rete
+
+fault_resume_kernel:
+#ifdef CONFIG_PREEMPT
+	get_thread_info	r0
+	ld.w	r2, r0[TI_preempt_count]
+	cp.w	r2, 0
+	brne	1f
+	ld.w	r1, r0[TI_flags]
+	bld	r1, TIF_NEED_RESCHED
+	brcc	1f
+	lddsp	r4, sp[REG_SR]
+	bld	r4, SYSREG_GM_OFFSET
+	brcs	1f
+	rcall	preempt_schedule_irq
+1:
+#endif
+
+	popm	r8-r9
+	mask_exceptions
+	mfsr	r1, SYSREG_SR
+	mtsr	SYSREG_RAR_EX, r8
+	mtsr	SYSREG_RSR_EX, r9
+	popm	lr
+	sub	sp, -4		/* ignore SP */
+	popm	r0-r12
+	sub	sp, -4		/* ignore r12_orig */
+	rete
+
+irq_exit_work:
+	/* Switch to exception mode so that we can share the same code. */
+	mfsr	r8, SYSREG_SR
+	cbr	r8, SYSREG_M0_OFFSET
+	orh	r8, hi(SYSREG_BIT(M1) | SYSREG_BIT(M2))
+	mtsr	SYSREG_SR, r8
+	sub	pc, -2
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+
+fault_exit_work:
+	bld	r1, TIF_NEED_RESCHED
+	brcc	1f
+	unmask_interrupts
+	rcall	schedule
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	fault_exit_work
+
+1:	mov	r2, _TIF_SIGPENDING | _TIF_RESTORE_SIGMASK
+	tst	r1, r2
+	breq	2f
+	unmask_interrupts
+	mov	r12, sp
+	mov	r11, r0
+	rcall	do_notify_resume
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	fault_exit_work
+
+2:	bld	r1, TIF_BREAKPOINT
+	brcc	fault_resume_user
+	mfsr	r3, SYSREG_TLBEHI
+	lddsp	r2, sp[REG_PC]
+	andl	r3, 0xff, COH
+	lsl	r3, 1
+	sbr	r3, 30
+	sbr	r3, 0
+	mtdr	DBGREG_BWA2A, r2
+	mtdr	DBGREG_BWC2A, r3
+	rjmp	fault_resume_user
+
+	/* If we get a debug trap from privileged context we end up here */
+handle_debug_priv:
+	/* Fix up LR and SP in regs. r11 contains the mode we came from */
+	mfsr	r8, SYSREG_SR
+	mov	r9, r8
+	andh	r8, hi(~MODE_MASK)
+	or	r8, r11
+	mtsr	SYSREG_SR, r8
+	sub	pc, -2
+	stdsp	sp[REG_LR], lr
+	mtsr	SYSREG_SR, r9
+	sub	pc, -2
+	sub	r10, sp, -FRAME_SIZE_FULL
+	stdsp	sp[REG_SP], r10
+	mov	r12, sp
+	rcall	do_debug_priv
+
+	/* Now, put everything back */
+	ssrf	SR_EM_BIT
+	popm	r10, r11
+	mtsr	SYSREG_RAR_DBG, r10
+	mtsr	SYSREG_RSR_DBG, r11
+	mfsr	r8, SYSREG_SR
+	mov	r9, r8
+	andh	r8, hi(~MODE_MASK)
+	andh	r11, hi(MODE_MASK)
+	or	r8, r11
+	mtsr	SYSREG_SR, r8
+	sub	pc, -2
+	popm	lr
+	mtsr	SYSREG_SR, r9
+	sub	pc, -2
+	sub	sp, -4		/* skip SP */
+	popm	r0-r12
+	sub	sp, -4
+	retd
+
+	/*
+	 * At this point, everything is masked, that is, interrupts,
+	 * exceptions and debugging traps. We might get called from
+	 * interrupt or exception context in some rare cases, but this
+	 * will be taken care of by do_debug(), so we're not going to
+	 * do a 100% correct context save here.
+	 */
+handle_debug:
+	sub	sp, 4		/* r12_orig */
+	stmts	--sp, r0-lr
+	mfsr	r10, SYSREG_RAR_DBG
+	mfsr	r11, SYSREG_RSR_DBG
+	unmask_exceptions
+	pushm	r10,r11
+	andh	r11, (MODE_MASK >> 16), COH
+	brne	handle_debug_priv
+
+	mov	r12, sp
+	rcall	do_debug
+
+	lddsp	r10, sp[REG_SR]
+	andh	r10, (MODE_MASK >> 16), COH
+	breq	debug_resume_user
+
+debug_restore_all:
+	popm	r10,r11
+	mask_exceptions
+	mtsr	SYSREG_RSR_DBG, r11
+	mtsr	SYSREG_RAR_DBG, r10
+	ldmts	sp++, r0-lr
+	sub	sp, -4
+	retd
+
+debug_resume_user:
+	get_thread_info r0
+	mask_interrupts
+
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_DBGWORK_MASK, COH
+	breq	debug_restore_all
+
+1:	bld	r1, TIF_NEED_RESCHED
+	brcc	2f
+	unmask_interrupts
+	rcall	schedule
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+2:	mov	r2, _TIF_SIGPENDING | _TIF_RESTORE_SIGMASK
+	tst	r1, r2
+	breq	3f
+	unmask_interrupts
+	mov	r12, sp
+	mov	r11, r0
+	rcall	do_notify_resume
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+3:	bld	r1, TIF_SINGLE_STEP
+	brcc	debug_restore_all
+	mfdr	r2, DBGREG_DC
+	sbr	r2, DC_SS_BIT
+	mtdr	DBGREG_DC, r2
+	rjmp	debug_restore_all
+
+	.set	rsr_int0,	SYSREG_RSR_INT0
+	.set	rsr_int1,	SYSREG_RSR_INT1
+	.set	rsr_int2,	SYSREG_RSR_INT2
+	.set	rsr_int3,	SYSREG_RSR_INT3
+	.set	rar_int0,	SYSREG_RAR_INT0
+	.set	rar_int1,	SYSREG_RAR_INT1
+	.set	rar_int2,	SYSREG_RAR_INT2
+	.set	rar_int3,	SYSREG_RAR_INT3
+
+	.macro	IRQ_LEVEL level
+	.type	irq_level\level, @function
+irq_level\level:
+	sub	sp, 4		/* r12_orig */
+	stmts	--sp,r0-lr
+	mfsr	r8, rar_int\level
+	mfsr	r9, rsr_int\level
+	pushm	r8-r9
+
+	mov	r11, sp
+	mov	r12, \level
+
+	rcall	do_IRQ
+
+	lddsp	r4, sp[REG_SR]
+	andh	r4, (MODE_MASK >> 16), COH
+#ifdef CONFIG_PREEMPT
+	brne	2f
+#else
+	brne	1f
+#endif
+
+	get_thread_info	r0
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_WORK_MASK, COH
+	brne	irq_exit_work
+
+1:	popm	r8-r9
+	mtsr	rar_int\level, r8
+	mtsr	rsr_int\level, r9
+	ldmts	sp++,r0-lr
+	sub	sp, -4		/* ignore r12_orig */
+	rete
+
+#ifdef CONFIG_PREEMPT
+2:
+	get_thread_info	r0
+	ld.w	r2, r0[TI_preempt_count]
+	cp.w	r2, 0
+	brne	1b
+	ld.w	r1, r0[TI_flags]
+	bld	r1, TIF_NEED_RESCHED
+	brcc	1b
+	lddsp	r4, sp[REG_SR]
+	bld	r4, SYSREG_GM_OFFSET
+	brcs	1b
+	rcall	preempt_schedule_irq
+	rjmp	1b
+#endif
+	.endm
+
+	.section .irq.text,"ax",@progbits
+
+	.global	irq_level0
+	.global	irq_level1
+	.global	irq_level2
+	.global	irq_level3
+	IRQ_LEVEL 0
+	IRQ_LEVEL 1
+	IRQ_LEVEL 2
+	IRQ_LEVEL 3
diff --git a/arch/avr32/kernel/head.S b/arch/avr32/kernel/head.S
new file mode 100644
index 000000000000..773b7ad87be9
--- /dev/null
+++ b/arch/avr32/kernel/head.S
@@ -0,0 +1,45 @@
+/*
+ * Non-board-specific low-level startup code
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/sysreg.h>
+
+	.section .init.text,"ax"
+	.global kernel_entry
+kernel_entry:
+	/* Initialize status register */
+	lddpc   r0, init_sr
+	mtsr	SYSREG_SR, r0
+
+	/* Set initial stack pointer */
+	lddpc   sp, stack_addr
+	sub	sp, -THREAD_SIZE
+
+#ifdef CONFIG_FRAME_POINTER
+	/* Mark last stack frame */
+	mov	lr, 0
+	mov	r7, 0
+#endif
+
+	/* Set up the PIO, SDRAM controller, early printk, etc. */
+	rcall	board_early_init
+
+	/* Start the show */
+	lddpc   pc, kernel_start_addr
+
+	.align  2
+init_sr:
+	.long   0x007f0000	/* Supervisor mode, everything masked */
+stack_addr:
+	.long   init_thread_union
+kernel_start_addr:
+	.long   start_kernel
diff --git a/arch/avr32/kernel/init_task.c b/arch/avr32/kernel/init_task.c
new file mode 100644
index 000000000000..effcacf9d1a2
--- /dev/null
+++ b/arch/avr32/kernel/init_task.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/sched.h>
+#include <linux/init_task.h>
+#include <linux/mqueue.h>
+
+#include <asm/pgtable.h>
+
+static struct fs_struct init_fs = INIT_FS;
+static struct files_struct init_files = INIT_FILES;
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+struct mm_struct init_mm = INIT_MM(init_mm);
+
+EXPORT_SYMBOL(init_mm);
+
+/*
+ * Initial thread structure. Must be aligned on an 8192-byte boundary.
+ */
+union thread_union init_thread_union
+	__attribute__((__section__(".data.init_task"))) =
+		{ INIT_THREAD_INFO(init_task) };
+
+/*
+ * Initial task structure.
+ *
+ * All other task structs will be allocated on slabs in fork.c
+ */
+struct task_struct init_task = INIT_TASK(init_task);
+
+EXPORT_SYMBOL(init_task);
diff --git a/arch/avr32/kernel/irq.c b/arch/avr32/kernel/irq.c
new file mode 100644
index 000000000000..856f3548e664
--- /dev/null
+++ b/arch/avr32/kernel/irq.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on arch/i386/kernel/irq.c
+ *   Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This file contains the code used by various IRQ handling routines:
+ * asking for different IRQ's should be done through these routines
+ * instead of just grabbing them. Thus setups with different IRQ numbers
+ * shouldn't result in any weird surprises, and installing new handlers
+ * should be easier.
+ *
+ * IRQ's are in fact implemented a bit like signal handlers for the kernel.
+ * Naturally it's not a 1:1 relation, but there are similarities.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel_stat.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/sysdev.h>
+
+/*
+ * 'what should we do if we get a hw irq event on an illegal vector'.
+ * each architecture has to answer this themselves.
+ */
+void ack_bad_irq(unsigned int irq)
+{
+	printk("unexpected IRQ %u\n", irq);
+}
+
+#ifdef CONFIG_PROC_FS
+int show_interrupts(struct seq_file *p, void *v)
+{
+	int i = *(loff_t *)v, cpu;
+	struct irqaction *action;
+	unsigned long flags;
+
+	if (i == 0) {
+		seq_puts(p, "           ");
+		for_each_online_cpu(cpu)
+			seq_printf(p, "CPU%d       ", cpu);
+		seq_putc(p, '\n');
+	}
+
+	if (i < NR_IRQS) {
+		spin_lock_irqsave(&irq_desc[i].lock, flags);
+		action = irq_desc[i].action;
+		if (!action)
+			goto unlock;
+
+		seq_printf(p, "%3d: ", i);
+		for_each_online_cpu(cpu)
+			seq_printf(p, "%10u ", kstat_cpu(cpu).irqs[i]);
+		seq_printf(p, "  %s", action->name);
+		for (action = action->next; action; action = action->next)
+			seq_printf(p, ", %s", action->name);
+
+		seq_putc(p, '\n');
+	unlock:
+		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
+	}
+
+	return 0;
+}
+#endif
diff --git a/arch/avr32/kernel/kprobes.c b/arch/avr32/kernel/kprobes.c
new file mode 100644
index 000000000000..6caf9e8d8080
--- /dev/null
+++ b/arch/avr32/kernel/kprobes.c
@@ -0,0 +1,270 @@
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * Based on arch/ppc64/kernel/kprobes.c
+ *  Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+
+#include <asm/cacheflush.h>
+#include <asm/kdebug.h>
+#include <asm/ocd.h>
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe);
+static unsigned long kprobe_status;
+static struct pt_regs jprobe_saved_regs;
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	int ret = 0;
+
+	if ((unsigned long)p->addr & 0x01) {
+		printk("Attempt to register kprobe at an unaligned address\n");
+		ret = -EINVAL;
+	}
+
+	/* XXX: Might be a good idea to check if p->addr is a valid
+	 * kernel address as well... */
+
+	if (!ret) {
+		pr_debug("copy kprobe at %p\n", p->addr);
+		memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+		p->opcode = *p->addr;
+	}
+
+	return ret;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	pr_debug("arming kprobe at %p\n", p->addr);
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	pr_debug("disarming kprobe at %p\n", p->addr);
+	*p->addr = p->opcode;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+{
+	unsigned long dc;
+
+	pr_debug("preparing to singlestep over %p (PC=%08lx)\n",
+		 p->addr, regs->pc);
+
+	BUG_ON(!(sysreg_read(SR) & SYSREG_BIT(SR_D)));
+
+	dc = __mfdr(DBGREG_DC);
+	dc |= DC_SS;
+	__mtdr(DBGREG_DC, dc);
+
+	/*
+	 * We must run the instruction from its original location
+	 * since it may actually reference PC.
+	 *
+	 * TODO: Do the instruction replacement directly in icache.
+	 */
+	*p->addr = p->opcode;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
+{
+	unsigned long dc;
+
+	pr_debug("resuming execution at PC=%08lx\n", regs->pc);
+
+	dc = __mfdr(DBGREG_DC);
+	dc &= ~DC_SS;
+	__mtdr(DBGREG_DC, dc);
+
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p)
+{
+	__get_cpu_var(current_kprobe) = p;
+}
+
+static int __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p;
+	void *addr = (void *)regs->pc;
+	int ret = 0;
+
+	pr_debug("kprobe_handler: kprobe_running=%d\n",
+		 kprobe_running());
+
+	/*
+	 * We don't want to be preempted for the entire
+	 * duration of kprobe processing
+	 */
+	preempt_disable();
+
+	/* Check that we're not recursing */
+	if (kprobe_running()) {
+		p = get_kprobe(addr);
+		if (p) {
+			if (kprobe_status == KPROBE_HIT_SS) {
+				printk("FIXME: kprobe hit while single-stepping!\n");
+				goto no_kprobe;
+			}
+
+			printk("FIXME: kprobe hit while handling another kprobe\n");
+			goto no_kprobe;
+		} else {
+			p = kprobe_running();
+			if (p->break_handler && p->break_handler(p, regs))
+				goto ss_probe;
+		}
+		/* If it's not ours, can't be delete race, (we hold lock). */
+		goto no_kprobe;
+	}
+
+	p = get_kprobe(addr);
+	if (!p)
+		goto no_kprobe;
+
+	kprobe_status = KPROBE_HIT_ACTIVE;
+	set_current_kprobe(p);
+	if (p->pre_handler && p->pre_handler(p, regs))
+		/* handler has already set things up, so skip ss setup */
+		return 1;
+
+ss_probe:
+	prepare_singlestep(p, regs);
+	kprobe_status = KPROBE_HIT_SS;
+	return 1;
+
+no_kprobe:
+	return ret;
+}
+
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *cur = kprobe_running();
+
+	pr_debug("post_kprobe_handler, cur=%p\n", cur);
+
+	if (!cur)
+		return 0;
+
+	if (cur->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
+		cur->post_handler(cur, regs, 0);
+	}
+
+	resume_execution(cur, regs);
+	reset_current_kprobe();
+	preempt_enable_no_resched();
+
+	return 1;
+}
+
+static int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+{
+	struct kprobe *cur = kprobe_running();
+
+	pr_debug("kprobe_fault_handler: trapnr=%d\n", trapnr);
+
+	if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
+		return 1;
+
+	if (kprobe_status & KPROBE_HIT_SS) {
+		resume_execution(cur, regs);
+		preempt_enable_no_resched();
+	}
+	return 0;
+}
+
+/*
+ * Wrapper routine to for handling exceptions.
+ */
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	pr_debug("kprobe_exceptions_notify: val=%lu, data=%p\n",
+		 val, data);
+
+	switch (val) {
+	case DIE_BREAKPOINT:
+		if (kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_SSTEP:
+		if (post_kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_FAULT:
+		if (kprobe_running()
+		    && kprobe_fault_handler(args->regs, args->trapnr))
+			ret = NOTIFY_STOP;
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+
+	memcpy(&jprobe_saved_regs, regs, sizeof(struct pt_regs));
+
+	/*
+	 * TODO: We should probably save some of the stack here as
+	 * well, since gcc may pass arguments on the stack for certain
+	 * functions (lots of arguments, large aggregates, varargs)
+	 */
+
+	/* setup return addr to the jprobe handler routine */
+	regs->pc = (unsigned long)jp->entry;
+	return 1;
+}
+
+void __kprobes jprobe_return(void)
+{
+	asm volatile("breakpoint" ::: "memory");
+}
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	/*
+	 * FIXME - we should ideally be validating that we got here 'cos
+	 * of the "trap" in jprobe_return() above, before restoring the
+	 * saved regs...
+	 */
+	memcpy(regs, &jprobe_saved_regs, sizeof(struct pt_regs));
+	return 1;
+}
+
+int __init arch_init_kprobes(void)
+{
+	printk("KPROBES: Enabling monitor mode (MM|DBE)...\n");
+	__mtdr(DBGREG_DC, DC_MM | DC_DBE);
+
+	/* TODO: Register kretprobe trampoline */
+	return 0;
+}
diff --git a/arch/avr32/kernel/module.c b/arch/avr32/kernel/module.c
new file mode 100644
index 000000000000..dfc32f2817b6
--- /dev/null
+++ b/arch/avr32/kernel/module.c
@@ -0,0 +1,324 @@
+/*
+ * AVR32-specific kernel module loader
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * GOT initialization parts are based on the s390 version
+ *   Copyright (C) 2002, 2003 IBM Deutschland Entwicklung GmbH,
+ *                            IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/moduleloader.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/elf.h>
+#include <linux/vmalloc.h>
+
+void *module_alloc(unsigned long size)
+{
+	if (size == 0)
+		return NULL;
+	return vmalloc(size);
+}
+
+void module_free(struct module *mod, void *module_region)
+{
+	vfree(mod->arch.syminfo);
+	mod->arch.syminfo = NULL;
+
+	vfree(module_region);
+	/* FIXME: if module_region == mod->init_region, trim exception
+	 * table entries. */
+}
+
+static inline int check_rela(Elf32_Rela *rela, struct module *module,
+			     char *strings, Elf32_Sym *symbols)
+{
+	struct mod_arch_syminfo *info;
+
+	info = module->arch.syminfo + ELF32_R_SYM(rela->r_info);
+	switch (ELF32_R_TYPE(rela->r_info)) {
+	case R_AVR32_GOT32:
+	case R_AVR32_GOT16:
+	case R_AVR32_GOT8:
+	case R_AVR32_GOT21S:
+	case R_AVR32_GOT18SW:	/* mcall */
+	case R_AVR32_GOT16S:	/* ld.w */
+		if (rela->r_addend != 0) {
+			printk(KERN_ERR
+			       "GOT relocation against %s at offset %u with addend\n",
+			       strings + symbols[ELF32_R_SYM(rela->r_info)].st_name,
+			       rela->r_offset);
+			return -ENOEXEC;
+		}
+		if (info->got_offset == -1UL) {
+			info->got_offset = module->arch.got_size;
+			module->arch.got_size += sizeof(void *);
+		}
+		pr_debug("GOT[%3lu] %s\n", info->got_offset,
+			 strings + symbols[ELF32_R_SYM(rela->r_info)].st_name);
+		break;
+	}
+
+	return 0;
+}
+
+int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
+			      char *secstrings, struct module *module)
+{
+	Elf32_Shdr *symtab;
+	Elf32_Sym *symbols;
+	Elf32_Rela *rela;
+	char *strings;
+	int nrela, i, j;
+	int ret;
+
+	/* Find the symbol table */
+	symtab = NULL;
+	for (i = 0; i < hdr->e_shnum; i++)
+		switch (sechdrs[i].sh_type) {
+		case SHT_SYMTAB:
+			symtab = &sechdrs[i];
+			break;
+		}
+	if (!symtab) {
+		printk(KERN_ERR "module %s: no symbol table\n", module->name);
+		return -ENOEXEC;
+	}
+
+	/* Allocate room for one syminfo structure per symbol. */
+	module->arch.nsyms = symtab->sh_size / sizeof(Elf_Sym);
+	module->arch.syminfo = vmalloc(module->arch.nsyms
+				   * sizeof(struct mod_arch_syminfo));
+	if (!module->arch.syminfo)
+		return -ENOMEM;
+
+	symbols = (void *)hdr + symtab->sh_offset;
+	strings = (void *)hdr + sechdrs[symtab->sh_link].sh_offset;
+	for (i = 0; i < module->arch.nsyms; i++) {
+		if (symbols[i].st_shndx == SHN_UNDEF &&
+		    strcmp(strings + symbols[i].st_name,
+			   "_GLOBAL_OFFSET_TABLE_") == 0)
+			/* "Define" it as absolute. */
+			symbols[i].st_shndx = SHN_ABS;
+		module->arch.syminfo[i].got_offset = -1UL;
+		module->arch.syminfo[i].got_initialized = 0;
+	}
+
+	/* Allocate GOT entries for symbols that need it. */
+	module->arch.got_size = 0;
+	for (i = 0; i < hdr->e_shnum; i++) {
+		if (sechdrs[i].sh_type != SHT_RELA)
+			continue;
+		nrela = sechdrs[i].sh_size / sizeof(Elf32_Rela);
+		rela = (void *)hdr + sechdrs[i].sh_offset;
+		for (j = 0; j < nrela; j++) {
+			ret = check_rela(rela + j, module,
+					 strings, symbols);
+			if (ret)
+				goto out_free_syminfo;
+		}
+	}
+
+	/*
+	 * Increase core size to make room for GOT and set start
+	 * offset for GOT.
+	 */
+	module->core_size = ALIGN(module->core_size, 4);
+	module->arch.got_offset = module->core_size;
+	module->core_size += module->arch.got_size;
+
+	return 0;
+
+out_free_syminfo:
+	vfree(module->arch.syminfo);
+	module->arch.syminfo = NULL;
+
+	return ret;
+}
+
+static inline int reloc_overflow(struct module *module, const char *reloc_name,
+				 Elf32_Addr relocation)
+{
+	printk(KERN_ERR "module %s: Value %lx does not fit relocation %s\n",
+	       module->name, (unsigned long)relocation, reloc_name);
+	return -ENOEXEC;
+}
+
+#define get_u16(loc)		(*((uint16_t *)loc))
+#define put_u16(loc, val)	(*((uint16_t *)loc) = (val))
+
+int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab,
+		       unsigned int symindex, unsigned int relindex,
+		       struct module *module)
+{
+	Elf32_Shdr *symsec = sechdrs + symindex;
+	Elf32_Shdr *relsec = sechdrs + relindex;
+	Elf32_Shdr *dstsec = sechdrs + relsec->sh_info;
+	Elf32_Rela *rel = (void *)relsec->sh_addr;
+	unsigned int i;
+	int ret = 0;
+
+	for (i = 0; i < relsec->sh_size / sizeof(Elf32_Rela); i++, rel++) {
+		struct mod_arch_syminfo *info;
+		Elf32_Sym *sym;
+		Elf32_Addr relocation;
+		uint32_t *location;
+		uint32_t value;
+
+		location = (void *)dstsec->sh_addr + rel->r_offset;
+		sym = (Elf32_Sym *)symsec->sh_addr + ELF32_R_SYM(rel->r_info);
+		relocation = sym->st_value + rel->r_addend;
+
+		info = module->arch.syminfo + ELF32_R_SYM(rel->r_info);
+
+		/* Initialize GOT entry if necessary */
+		switch (ELF32_R_TYPE(rel->r_info)) {
+		case R_AVR32_GOT32:
+		case R_AVR32_GOT16:
+		case R_AVR32_GOT8:
+		case R_AVR32_GOT21S:
+		case R_AVR32_GOT18SW:
+		case R_AVR32_GOT16S:
+			if (!info->got_initialized) {
+				Elf32_Addr *gotent;
+
+				gotent = (module->module_core
+					  + module->arch.got_offset
+					  + info->got_offset);
+				*gotent = relocation;
+				info->got_initialized = 1;
+			}
+
+			relocation = info->got_offset;
+			break;
+		}
+
+		switch (ELF32_R_TYPE(rel->r_info)) {
+		case R_AVR32_32:
+		case R_AVR32_32_CPENT:
+			*location = relocation;
+			break;
+		case R_AVR32_22H_PCREL:
+			relocation -= (Elf32_Addr)location;
+			if ((relocation & 0xffe00001) != 0
+			    && (relocation & 0xffc00001) != 0xffc00000)
+				return reloc_overflow(module,
+						      "R_AVR32_22H_PCREL",
+						      relocation);
+			relocation >>= 1;
+
+			value = *location;
+			value = ((value & 0xe1ef0000)
+				 | (relocation & 0xffff)
+				 | ((relocation & 0x10000) << 4)
+				 | ((relocation & 0x1e0000) << 8));
+			*location = value;
+			break;
+		case R_AVR32_11H_PCREL:
+			relocation -= (Elf32_Addr)location;
+			if ((relocation & 0xfffffc01) != 0
+			    && (relocation & 0xfffff801) != 0xfffff800)
+				return reloc_overflow(module,
+						      "R_AVR32_11H_PCREL",
+						      relocation);
+			value = get_u16(location);
+			value = ((value & 0xf00c)
+				 | ((relocation & 0x1fe) << 3)
+				 | ((relocation & 0x600) >> 9));
+			put_u16(location, value);
+			break;
+		case R_AVR32_9H_PCREL:
+			relocation -= (Elf32_Addr)location;
+			if ((relocation & 0xffffff01) != 0
+			    && (relocation & 0xfffffe01) != 0xfffffe00)
+				return reloc_overflow(module,
+						      "R_AVR32_9H_PCREL",
+						      relocation);
+			value = get_u16(location);
+			value = ((value & 0xf00f)
+				 | ((relocation & 0x1fe) << 3));
+			put_u16(location, value);
+			break;
+		case R_AVR32_9UW_PCREL:
+			relocation -= ((Elf32_Addr)location) & 0xfffffffc;
+			if ((relocation & 0xfffffc03) != 0)
+				return reloc_overflow(module,
+						      "R_AVR32_9UW_PCREL",
+						      relocation);
+			value = get_u16(location);
+			value = ((value & 0xf80f)
+				 | ((relocation & 0x1fc) << 2));
+			put_u16(location, value);
+			break;
+		case R_AVR32_GOTPC:
+			/*
+			 * R6 = PC - (PC - GOT)
+			 *
+			 * At this point, relocation contains the
+			 * value of PC.  Just subtract the value of
+			 * GOT, and we're done.
+			 */
+			pr_debug("GOTPC: PC=0x%lx, got_offset=0x%lx, core=0x%p\n",
+				 relocation, module->arch.got_offset,
+				 module->module_core);
+			relocation -= ((unsigned long)module->module_core
+				       + module->arch.got_offset);
+			*location = relocation;
+			break;
+		case R_AVR32_GOT18SW:
+			if ((relocation & 0xfffe0003) != 0
+			    && (relocation & 0xfffc0003) != 0xffff0000)
+				return reloc_overflow(module, "R_AVR32_GOT18SW",
+						     relocation);
+			relocation >>= 2;
+			/* fall through */
+		case R_AVR32_GOT16S:
+			if ((relocation & 0xffff8000) != 0
+			    && (relocation & 0xffff0000) != 0xffff0000)
+				return reloc_overflow(module, "R_AVR32_GOT16S",
+						      relocation);
+			pr_debug("GOT reloc @ 0x%lx -> %lu\n",
+				 rel->r_offset, relocation);
+			value = *location;
+			value = ((value & 0xffff0000)
+				 | (relocation & 0xffff));
+			*location = value;
+			break;
+
+		default:
+			printk(KERN_ERR "module %s: Unknown relocation: %u\n",
+			       module->name, ELF32_R_TYPE(rel->r_info));
+			return -ENOEXEC;
+		}
+	}
+
+	return ret;
+}
+
+int apply_relocate(Elf32_Shdr *sechdrs, const char *strtab,
+		   unsigned int symindex, unsigned int relindex,
+		   struct module *module)
+{
+	printk(KERN_ERR "module %s: REL relocations are not supported\n",
+		module->name);
+	return -ENOEXEC;
+}
+
+int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
+		    struct module *module)
+{
+	vfree(module->arch.syminfo);
+	module->arch.syminfo = NULL;
+
+	return 0;
+}
+
+void module_arch_cleanup(struct module *module)
+{
+
+}
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
new file mode 100644
index 000000000000..317dc50945f2
--- /dev/null
+++ b/arch/avr32/kernel/process.c
@@ -0,0 +1,276 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+#include <linux/fs.h>
+#include <linux/ptrace.h>
+#include <linux/reboot.h>
+#include <linux/unistd.h>
+
+#include <asm/sysreg.h>
+#include <asm/ocd.h>
+
+void (*pm_power_off)(void) = NULL;
+EXPORT_SYMBOL(pm_power_off);
+
+/*
+ * This file handles the architecture-dependent parts of process handling..
+ */
+
+void cpu_idle(void)
+{
+	/* endless idle loop with no priority at all */
+	while (1) {
+		/* TODO: Enter sleep mode */
+		while (!need_resched())
+			cpu_relax();
+		preempt_enable_no_resched();
+		schedule();
+		preempt_disable();
+	}
+}
+
+void machine_halt(void)
+{
+}
+
+void machine_power_off(void)
+{
+}
+
+void machine_restart(char *cmd)
+{
+	__mtdr(DBGREG_DC, DC_DBE);
+	__mtdr(DBGREG_DC, DC_RES);
+	while (1) ;
+}
+
+/*
+ * PC is actually discarded when returning from a system call -- the
+ * return address must be stored in LR. This function will make sure
+ * LR points to do_exit before starting the thread.
+ *
+ * Also, when returning from fork(), r12 is 0, so we must copy the
+ * argument as well.
+ *
+ *  r0 : The argument to the main thread function
+ *  r1 : The address of do_exit
+ *  r2 : The address of the main thread function
+ */
+asmlinkage extern void kernel_thread_helper(void);
+__asm__("	.type	kernel_thread_helper, @function\n"
+	"kernel_thread_helper:\n"
+	"	mov	r12, r0\n"
+	"	mov	lr, r2\n"
+	"	mov	pc, r1\n"
+	"	.size	kernel_thread_helper, . - kernel_thread_helper");
+
+int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
+{
+	struct pt_regs regs;
+
+	memset(&regs, 0, sizeof(regs));
+
+	regs.r0 = (unsigned long)arg;
+	regs.r1 = (unsigned long)fn;
+	regs.r2 = (unsigned long)do_exit;
+	regs.lr = (unsigned long)kernel_thread_helper;
+	regs.pc = (unsigned long)kernel_thread_helper;
+	regs.sr = MODE_SUPERVISOR;
+
+	return do_fork(flags | CLONE_VM | CLONE_UNTRACED,
+		       0, &regs, 0, NULL, NULL);
+}
+EXPORT_SYMBOL(kernel_thread);
+
+/*
+ * Free current thread data structures etc
+ */
+void exit_thread(void)
+{
+	/* nothing to do */
+}
+
+void flush_thread(void)
+{
+	/* nothing to do */
+}
+
+void release_thread(struct task_struct *dead_task)
+{
+	/* do nothing */
+}
+
+static const char *cpu_modes[] = {
+	"Application", "Supervisor", "Interrupt level 0", "Interrupt level 1",
+	"Interrupt level 2", "Interrupt level 3", "Exception", "NMI"
+};
+
+void show_regs(struct pt_regs *regs)
+{
+	unsigned long sp = regs->sp;
+	unsigned long lr = regs->lr;
+	unsigned long mode = (regs->sr & MODE_MASK) >> MODE_SHIFT;
+
+	if (!user_mode(regs))
+		sp = (unsigned long)regs + FRAME_SIZE_FULL;
+
+	print_symbol("PC is at %s\n", instruction_pointer(regs));
+	print_symbol("LR is at %s\n", lr);
+	printk("pc : [<%08lx>]    lr : [<%08lx>]    %s\n"
+	       "sp : %08lx  r12: %08lx  r11: %08lx\n",
+	       instruction_pointer(regs),
+	       lr, print_tainted(), sp, regs->r12, regs->r11);
+	printk("r10: %08lx  r9 : %08lx  r8 : %08lx\n",
+	       regs->r10, regs->r9, regs->r8);
+	printk("r7 : %08lx  r6 : %08lx  r5 : %08lx  r4 : %08lx\n",
+	       regs->r7, regs->r6, regs->r5, regs->r4);
+	printk("r3 : %08lx  r2 : %08lx  r1 : %08lx  r0 : %08lx\n",
+	       regs->r3, regs->r2, regs->r1, regs->r0);
+	printk("Flags: %c%c%c%c%c\n",
+	       regs->sr & SR_Q ? 'Q' : 'q',
+	       regs->sr & SR_V ? 'V' : 'v',
+	       regs->sr & SR_N ? 'N' : 'n',
+	       regs->sr & SR_Z ? 'Z' : 'z',
+	       regs->sr & SR_C ? 'C' : 'c');
+	printk("Mode bits: %c%c%c%c%c%c%c%c%c\n",
+	       regs->sr & SR_H ? 'H' : 'h',
+	       regs->sr & SR_R ? 'R' : 'r',
+	       regs->sr & SR_J ? 'J' : 'j',
+	       regs->sr & SR_EM ? 'E' : 'e',
+	       regs->sr & SR_I3M ? '3' : '.',
+	       regs->sr & SR_I2M ? '2' : '.',
+	       regs->sr & SR_I1M ? '1' : '.',
+	       regs->sr & SR_I0M ? '0' : '.',
+	       regs->sr & SR_GM ? 'G' : 'g');
+	printk("CPU Mode: %s\n", cpu_modes[mode]);
+
+	show_trace(NULL, (unsigned long *)sp, regs);
+}
+EXPORT_SYMBOL(show_regs);
+
+/* Fill in the fpu structure for a core dump. This is easy -- we don't have any */
+int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu)
+{
+	/* Not valid */
+	return 0;
+}
+
+asmlinkage void ret_from_fork(void);
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
+		unsigned long unused,
+		struct task_struct *p, struct pt_regs *regs)
+{
+	struct pt_regs *childregs;
+
+	childregs = ((struct pt_regs *)(THREAD_SIZE + (unsigned long)p->thread_info)) - 1;
+	*childregs = *regs;
+
+	if (user_mode(regs))
+		childregs->sp = usp;
+	else
+		childregs->sp = (unsigned long)p->thread_info + THREAD_SIZE;
+
+	childregs->r12 = 0; /* Set return value for child */
+
+	p->thread.cpu_context.sr = MODE_SUPERVISOR | SR_GM;
+	p->thread.cpu_context.ksp = (unsigned long)childregs;
+	p->thread.cpu_context.pc = (unsigned long)ret_from_fork;
+
+	return 0;
+}
+
+/* r12-r8 are dummy parameters to force the compiler to use the stack */
+asmlinkage int sys_fork(struct pt_regs *regs)
+{
+	return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
+}
+
+asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
+			 unsigned long parent_tidptr,
+			 unsigned long child_tidptr, struct pt_regs *regs)
+{
+	if (!newsp)
+		newsp = regs->sp;
+	return do_fork(clone_flags, newsp, regs, 0,
+		       (int __user *)parent_tidptr,
+		       (int __user *)child_tidptr);
+}
+
+asmlinkage int sys_vfork(struct pt_regs *regs)
+{
+	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs,
+		       0, NULL, NULL);
+}
+
+asmlinkage int sys_execve(char __user *ufilename, char __user *__user *uargv,
+			  char __user *__user *uenvp, struct pt_regs *regs)
+{
+	int error;
+	char *filename;
+
+	filename = getname(ufilename);
+	error = PTR_ERR(filename);
+	if (IS_ERR(filename))
+		goto out;
+
+	error = do_execve(filename, uargv, uenvp, regs);
+	if (error == 0)
+		current->ptrace &= ~PT_DTRACE;
+	putname(filename);
+
+out:
+	return error;
+}
+
+
+/*
+ * This function is supposed to answer the question "who called
+ * schedule()?"
+ */
+unsigned long get_wchan(struct task_struct *p)
+{
+	unsigned long pc;
+	unsigned long stack_page;
+
+	if (!p || p == current || p->state == TASK_RUNNING)
+		return 0;
+
+	stack_page = (unsigned long)p->thread_info;
+	BUG_ON(!stack_page);
+
+	/*
+	 * The stored value of PC is either the address right after
+	 * the call to __switch_to() or ret_from_fork.
+	 */
+	pc = thread_saved_pc(p);
+	if (in_sched_functions(pc)) {
+#ifdef CONFIG_FRAME_POINTER
+		unsigned long fp = p->thread.cpu_context.r7;
+		BUG_ON(fp < stack_page || fp > (THREAD_SIZE + stack_page));
+		pc = *(unsigned long *)fp;
+#else
+		/*
+		 * We depend on the frame size of schedule here, which
+		 * is actually quite ugly. It might be possible to
+		 * determine the frame size automatically at build
+		 * time by doing this:
+		 *   - compile sched.c
+		 *   - disassemble the resulting sched.o
+		 *   - look for 'sub sp,??' shortly after '<schedule>:'
+		 */
+		unsigned long sp = p->thread.cpu_context.ksp + 16;
+		BUG_ON(sp < stack_page || sp > (THREAD_SIZE + stack_page));
+		pc = *(unsigned long *)sp;
+#endif
+	}
+
+	return pc;
+}
diff --git a/arch/avr32/kernel/ptrace.c b/arch/avr32/kernel/ptrace.c
new file mode 100644
index 000000000000..3c89e59029ab
--- /dev/null
+++ b/arch/avr32/kernel/ptrace.c
@@ -0,0 +1,371 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#undef DEBUG
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp_lock.h>
+#include <linux/ptrace.h>
+#include <linux/errno.h>
+#include <linux/user.h>
+#include <linux/security.h>
+#include <linux/unistd.h>
+#include <linux/notifier.h>
+
+#include <asm/traps.h>
+#include <asm/uaccess.h>
+#include <asm/ocd.h>
+#include <asm/mmu_context.h>
+#include <asm/kdebug.h>
+
+static struct pt_regs *get_user_regs(struct task_struct *tsk)
+{
+	return (struct pt_regs *)((unsigned long) tsk->thread_info +
+				  THREAD_SIZE - sizeof(struct pt_regs));
+}
+
+static void ptrace_single_step(struct task_struct *tsk)
+{
+	pr_debug("ptrace_single_step: pid=%u, SR=0x%08lx\n",
+		 tsk->pid, tsk->thread.cpu_context.sr);
+	if (!(tsk->thread.cpu_context.sr & SR_D)) {
+		/*
+		 * Set a breakpoint at the current pc to force the
+		 * process into debug mode.  The syscall/exception
+		 * exit code will set a breakpoint at the return
+		 * address when this flag is set.
+		 */
+		pr_debug("ptrace_single_step: Setting TIF_BREAKPOINT\n");
+		set_tsk_thread_flag(tsk, TIF_BREAKPOINT);
+	}
+
+	/* The monitor code will do the actual step for us */
+	set_tsk_thread_flag(tsk, TIF_SINGLE_STEP);
+}
+
+/*
+ * Called by kernel/ptrace.c when detaching
+ *
+ * Make sure any single step bits, etc. are not set
+ */
+void ptrace_disable(struct task_struct *child)
+{
+	clear_tsk_thread_flag(child, TIF_SINGLE_STEP);
+}
+
+/*
+ * Handle hitting a breakpoint
+ */
+static void ptrace_break(struct task_struct *tsk, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code  = TRAP_BRKPT;
+	info.si_addr  = (void __user *)instruction_pointer(regs);
+
+	pr_debug("ptrace_break: Sending SIGTRAP to PID %u (pc = 0x%p)\n",
+		 tsk->pid, info.si_addr);
+	force_sig_info(SIGTRAP, &info, tsk);
+}
+
+/*
+ * Read the word at offset "offset" into the task's "struct user". We
+ * actually access the pt_regs struct stored on the kernel stack.
+ */
+static int ptrace_read_user(struct task_struct *tsk, unsigned long offset,
+			    unsigned long __user *data)
+{
+	unsigned long *regs;
+	unsigned long value;
+
+	pr_debug("ptrace_read_user(%p, %#lx, %p)\n",
+		 tsk, offset, data);
+
+	if (offset & 3 || offset >= sizeof(struct user)) {
+		printk("ptrace_read_user: invalid offset 0x%08lx\n", offset);
+		return -EIO;
+	}
+
+	regs = (unsigned long *)get_user_regs(tsk);
+
+	value = 0;
+	if (offset < sizeof(struct pt_regs))
+		value = regs[offset / sizeof(regs[0])];
+
+	return put_user(value, data);
+}
+
+/*
+ * Write the word "value" to offset "offset" into the task's "struct
+ * user". We actually access the pt_regs struct stored on the kernel
+ * stack.
+ */
+static int ptrace_write_user(struct task_struct *tsk, unsigned long offset,
+			     unsigned long value)
+{
+	unsigned long *regs;
+
+	if (offset & 3 || offset >= sizeof(struct user)) {
+		printk("ptrace_write_user: invalid offset 0x%08lx\n", offset);
+		return -EIO;
+	}
+
+	if (offset >= sizeof(struct pt_regs))
+		return 0;
+
+	regs = (unsigned long *)get_user_regs(tsk);
+	regs[offset / sizeof(regs[0])] = value;
+
+	return 0;
+}
+
+static int ptrace_getregs(struct task_struct *tsk, void __user *uregs)
+{
+	struct pt_regs *regs = get_user_regs(tsk);
+
+	return copy_to_user(uregs, regs, sizeof(*regs)) ? -EFAULT : 0;
+}
+
+static int ptrace_setregs(struct task_struct *tsk, const void __user *uregs)
+{
+	struct pt_regs newregs;
+	int ret;
+
+	ret = -EFAULT;
+	if (copy_from_user(&newregs, uregs, sizeof(newregs)) == 0) {
+		struct pt_regs *regs = get_user_regs(tsk);
+
+		ret = -EINVAL;
+		if (valid_user_regs(&newregs)) {
+			*regs = newregs;
+			ret = 0;
+		}
+	}
+
+	return ret;
+}
+
+long arch_ptrace(struct task_struct *child, long request, long addr, long data)
+{
+	unsigned long tmp;
+	int ret;
+
+	pr_debug("arch_ptrace(%ld, %ld, %#lx, %#lx)\n",
+		 request, child->pid, addr, data);
+
+	pr_debug("ptrace: Enabling monitor mode...\n");
+	__mtdr(DBGREG_DC, __mfdr(DBGREG_DC) | DC_MM | DC_DBE);
+
+	switch (request) {
+	/* Read the word at location addr in the child process */
+	case PTRACE_PEEKTEXT:
+	case PTRACE_PEEKDATA:
+		ret = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
+		if (ret == sizeof(tmp))
+			ret = put_user(tmp, (unsigned long __user *)data);
+		else
+			ret = -EIO;
+		break;
+
+	case PTRACE_PEEKUSR:
+		ret = ptrace_read_user(child, addr,
+				       (unsigned long __user *)data);
+		break;
+
+	/* Write the word in data at location addr */
+	case PTRACE_POKETEXT:
+	case PTRACE_POKEDATA:
+		ret = access_process_vm(child, addr, &data, sizeof(data), 1);
+		if (ret == sizeof(data))
+			ret = 0;
+		else
+			ret = -EIO;
+		break;
+
+	case PTRACE_POKEUSR:
+		ret = ptrace_write_user(child, addr, data);
+		break;
+
+	/* continue and stop at next (return from) syscall */
+	case PTRACE_SYSCALL:
+	/* restart after signal */
+	case PTRACE_CONT:
+		ret = -EIO;
+		if (!valid_signal(data))
+			break;
+		if (request == PTRACE_SYSCALL)
+			set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		else
+			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		child->exit_code = data;
+		/* XXX: Are we sure no breakpoints are active here? */
+		wake_up_process(child);
+		ret = 0;
+		break;
+
+	/*
+	 * Make the child exit. Best I can do is send it a
+	 * SIGKILL. Perhaps it should be put in the status that it
+	 * wants to exit.
+	 */
+	case PTRACE_KILL:
+		ret = 0;
+		if (child->exit_state == EXIT_ZOMBIE)
+			break;
+		child->exit_code = SIGKILL;
+		wake_up_process(child);
+		break;
+
+	/*
+	 * execute single instruction.
+	 */
+	case PTRACE_SINGLESTEP:
+		ret = -EIO;
+		if (!valid_signal(data))
+			break;
+		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		ptrace_single_step(child);
+		child->exit_code = data;
+		wake_up_process(child);
+		ret = 0;
+		break;
+
+	/* Detach a process that was attached */
+	case PTRACE_DETACH:
+		ret = ptrace_detach(child, data);
+		break;
+
+	case PTRACE_GETREGS:
+		ret = ptrace_getregs(child, (void __user *)data);
+		break;
+
+	case PTRACE_SETREGS:
+		ret = ptrace_setregs(child, (const void __user *)data);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+		break;
+	}
+
+	pr_debug("sys_ptrace returning %d (DC = 0x%08lx)\n", ret, __mfdr(DBGREG_DC));
+	return ret;
+}
+
+asmlinkage void syscall_trace(void)
+{
+	pr_debug("syscall_trace called\n");
+	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+		return;
+	if (!(current->ptrace & PT_PTRACED))
+		return;
+
+	pr_debug("syscall_trace: notifying parent\n");
+	/* The 0x80 provides a way for the tracing parent to
+	 * distinguish between a syscall stop and SIGTRAP delivery */
+	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
+				 ? 0x80 : 0));
+
+	/*
+	 * this isn't the same as continuing with a signal, but it
+	 * will do for normal use.  strace only continues with a
+	 * signal if the stopping signal is not SIGTRAP.  -brl
+	 */
+	if (current->exit_code) {
+		pr_debug("syscall_trace: sending signal %d to PID %u\n",
+			 current->exit_code, current->pid);
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+}
+
+asmlinkage void do_debug_priv(struct pt_regs *regs)
+{
+	unsigned long dc, ds;
+	unsigned long die_val;
+
+	ds = __mfdr(DBGREG_DS);
+
+	pr_debug("do_debug_priv: pc = %08lx, ds = %08lx\n", regs->pc, ds);
+
+	if (ds & DS_SSS)
+		die_val = DIE_SSTEP;
+	else
+		die_val = DIE_BREAKPOINT;
+
+	if (notify_die(die_val, regs, 0, SIGTRAP) == NOTIFY_STOP)
+		return;
+
+	if (likely(ds & DS_SSS)) {
+		extern void itlb_miss(void);
+		extern void tlb_miss_common(void);
+		struct thread_info *ti;
+
+		dc = __mfdr(DBGREG_DC);
+		dc &= ~DC_SS;
+		__mtdr(DBGREG_DC, dc);
+
+		ti = current_thread_info();
+		ti->flags |= _TIF_BREAKPOINT;
+
+		/* The TLB miss handlers don't check thread flags */
+		if ((regs->pc >= (unsigned long)&itlb_miss)
+		    && (regs->pc <= (unsigned long)&tlb_miss_common)) {
+			__mtdr(DBGREG_BWA2A, sysreg_read(RAR_EX));
+			__mtdr(DBGREG_BWC2A, 0x40000001 | (get_asid() << 1));
+		}
+
+		/*
+		 * If we're running in supervisor mode, the breakpoint
+		 * will take us where we want directly, no need to
+		 * single step.
+		 */
+		if ((regs->sr & MODE_MASK) != MODE_SUPERVISOR)
+			ti->flags |= TIF_SINGLE_STEP;
+	} else {
+		panic("Unable to handle debug trap at pc = %08lx\n",
+		      regs->pc);
+	}
+}
+
+/*
+ * Handle breakpoints, single steps and other debuggy things. To keep
+ * things simple initially, we run with interrupts and exceptions
+ * disabled all the time.
+ */
+asmlinkage void do_debug(struct pt_regs *regs)
+{
+	unsigned long dc, ds;
+
+	ds = __mfdr(DBGREG_DS);
+	pr_debug("do_debug: pc = %08lx, ds = %08lx\n", regs->pc, ds);
+
+	if (test_thread_flag(TIF_BREAKPOINT)) {
+		pr_debug("TIF_BREAKPOINT set\n");
+		/* We're taking care of it */
+		clear_thread_flag(TIF_BREAKPOINT);
+		__mtdr(DBGREG_BWC2A, 0);
+	}
+
+	if (test_thread_flag(TIF_SINGLE_STEP)) {
+		pr_debug("TIF_SINGLE_STEP set, ds = 0x%08lx\n", ds);
+		if (ds & DS_SSS) {
+			dc = __mfdr(DBGREG_DC);
+			dc &= ~DC_SS;
+			__mtdr(DBGREG_DC, dc);
+
+			clear_thread_flag(TIF_SINGLE_STEP);
+			ptrace_break(current, regs);
+		}
+	} else {
+		/* regular breakpoint */
+		ptrace_break(current, regs);
+	}
+}
diff --git a/arch/avr32/kernel/semaphore.c b/arch/avr32/kernel/semaphore.c
new file mode 100644
index 000000000000..1e2705a05016
--- /dev/null
+++ b/arch/avr32/kernel/semaphore.c
@@ -0,0 +1,148 @@
+/*
+ * AVR32 sempahore implementation.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/i386/kernel/semaphore.c
+ *  Copyright (C) 1999 Linus Torvalds
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+
+#include <asm/semaphore.h>
+#include <asm/atomic.h>
+
+/*
+ * Semaphores are implemented using a two-way counter:
+ * The "count" variable is decremented for each process
+ * that tries to acquire the semaphore, while the "sleeping"
+ * variable is a count of such acquires.
+ *
+ * Notably, the inline "up()" and "down()" functions can
+ * efficiently test if they need to do any extra work (up
+ * needs to do something only if count was negative before
+ * the increment operation.
+ *
+ * "sleeping" and the contention routine ordering is protected
+ * by the spinlock in the semaphore's waitqueue head.
+ *
+ * Note that these functions are only called when there is
+ * contention on the lock, and as such all this is the
+ * "non-critical" part of the whole semaphore business. The
+ * critical part is the inline stuff in <asm/semaphore.h>
+ * where we want to avoid any extra jumps and calls.
+ */
+
+/*
+ * Logic:
+ *  - only on a boundary condition do we need to care. When we go
+ *    from a negative count to a non-negative, we wake people up.
+ *  - when we go from a non-negative count to a negative do we
+ *    (a) synchronize with the "sleeper" count and (b) make sure
+ *    that we're on the wakeup list before we synchronize so that
+ *    we cannot lose wakeup events.
+ */
+
+void __up(struct semaphore *sem)
+{
+	wake_up(&sem->wait);
+}
+EXPORT_SYMBOL(__up);
+
+void __sched __down(struct semaphore *sem)
+{
+	struct task_struct *tsk = current;
+        DECLARE_WAITQUEUE(wait, tsk);
+        unsigned long flags;
+
+        tsk->state = TASK_UNINTERRUPTIBLE;
+        spin_lock_irqsave(&sem->wait.lock, flags);
+        add_wait_queue_exclusive_locked(&sem->wait, &wait);
+
+        sem->sleepers++;
+        for (;;) {
+                int sleepers = sem->sleepers;
+
+                /*
+                 * Add "everybody else" into it. They aren't
+                 * playing, because we own the spinlock in
+                 * the wait_queue_head.
+                 */
+                if (atomic_add_return(sleepers - 1, &sem->count) >= 0) {
+                        sem->sleepers = 0;
+                        break;
+                }
+                sem->sleepers = 1;      /* us - see -1 above */
+                spin_unlock_irqrestore(&sem->wait.lock, flags);
+
+                schedule();
+
+                spin_lock_irqsave(&sem->wait.lock, flags);
+                tsk->state = TASK_UNINTERRUPTIBLE;
+        }
+        remove_wait_queue_locked(&sem->wait, &wait);
+        wake_up_locked(&sem->wait);
+        spin_unlock_irqrestore(&sem->wait.lock, flags);
+        tsk->state = TASK_RUNNING;
+}
+EXPORT_SYMBOL(__down);
+
+int __sched __down_interruptible(struct semaphore *sem)
+{
+	int retval = 0;
+	struct task_struct *tsk = current;
+        DECLARE_WAITQUEUE(wait, tsk);
+        unsigned long flags;
+
+        tsk->state = TASK_INTERRUPTIBLE;
+        spin_lock_irqsave(&sem->wait.lock, flags);
+        add_wait_queue_exclusive_locked(&sem->wait, &wait);
+
+        sem->sleepers++;
+        for (;;) {
+                int sleepers = sem->sleepers;
+
+		/*
+		 * With signals pending, this turns into the trylock
+		 * failure case - we won't be sleeping, and we can't
+		 * get the lock as it has contention. Just correct the
+		 * count and exit.
+		 */
+		if (signal_pending(current)) {
+			retval = -EINTR;
+			sem->sleepers = 0;
+			atomic_add(sleepers, &sem->count);
+			break;
+		}
+
+                /*
+                 * Add "everybody else" into it. They aren't
+                 * playing, because we own the spinlock in
+                 * the wait_queue_head.
+                 */
+                if (atomic_add_return(sleepers - 1, &sem->count) >= 0) {
+                        sem->sleepers = 0;
+                        break;
+                }
+                sem->sleepers = 1;      /* us - see -1 above */
+                spin_unlock_irqrestore(&sem->wait.lock, flags);
+
+                schedule();
+
+                spin_lock_irqsave(&sem->wait.lock, flags);
+                tsk->state = TASK_INTERRUPTIBLE;
+        }
+        remove_wait_queue_locked(&sem->wait, &wait);
+        wake_up_locked(&sem->wait);
+        spin_unlock_irqrestore(&sem->wait.lock, flags);
+
+        tsk->state = TASK_RUNNING;
+	return retval;
+}
+EXPORT_SYMBOL(__down_interruptible);
diff --git a/arch/avr32/kernel/setup.c b/arch/avr32/kernel/setup.c
new file mode 100644
index 000000000000..5d68f3c6990b
--- /dev/null
+++ b/arch/avr32/kernel/setup.c
@@ -0,0 +1,335 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/console.h>
+#include <linux/ioport.h>
+#include <linux/bootmem.h>
+#include <linux/fs.h>
+#include <linux/module.h>
+#include <linux/root_dev.h>
+#include <linux/cpu.h>
+
+#include <asm/sections.h>
+#include <asm/processor.h>
+#include <asm/pgtable.h>
+#include <asm/setup.h>
+#include <asm/sysreg.h>
+
+#include <asm/arch/board.h>
+#include <asm/arch/init.h>
+
+extern int root_mountflags;
+
+/*
+ * Bootloader-provided information about physical memory
+ */
+struct tag_mem_range *mem_phys;
+struct tag_mem_range *mem_reserved;
+struct tag_mem_range *mem_ramdisk;
+
+/*
+ * Initialize loops_per_jiffy as 5000000 (500MIPS).
+ * Better make it too large than too small...
+ */
+struct avr32_cpuinfo boot_cpu_data = {
+	.loops_per_jiffy = 5000000
+};
+EXPORT_SYMBOL(boot_cpu_data);
+
+static char command_line[COMMAND_LINE_SIZE];
+
+/*
+ * Should be more than enough, but if you have a _really_ complex
+ * setup, you might need to increase the size of this...
+ */
+static struct tag_mem_range __initdata mem_range_cache[32];
+static unsigned mem_range_next_free;
+
+/*
+ * Standard memory resources
+ */
+static struct resource mem_res[] = {
+	{
+		.name	= "Kernel code",
+		.start	= 0,
+		.end	= 0,
+		.flags	= IORESOURCE_MEM
+	},
+	{
+		.name	= "Kernel data",
+		.start	= 0,
+		.end	= 0,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+#define kernel_code	mem_res[0]
+#define kernel_data	mem_res[1]
+
+/*
+ * Early framebuffer allocation. Works as follows:
+ *   - If fbmem_size is zero, nothing will be allocated or reserved.
+ *   - If fbmem_start is zero when setup_bootmem() is called,
+ *     fbmem_size bytes will be allocated from the bootmem allocator.
+ *   - If fbmem_start is nonzero, an area of size fbmem_size will be
+ *     reserved at the physical address fbmem_start if necessary. If
+ *     the area isn't in a memory region known to the kernel, it will
+ *     be left alone.
+ *
+ * Board-specific code may use these variables to set up platform data
+ * for the framebuffer driver if fbmem_size is nonzero.
+ */
+static unsigned long __initdata fbmem_start;
+static unsigned long __initdata fbmem_size;
+
+/*
+ * "fbmem=xxx[kKmM]" allocates the specified amount of boot memory for
+ * use as framebuffer.
+ *
+ * "fbmem=xxx[kKmM]@yyy[kKmM]" defines a memory region of size xxx and
+ * starting at yyy to be reserved for use as framebuffer.
+ *
+ * The kernel won't verify that the memory region starting at yyy
+ * actually contains usable RAM.
+ */
+static int __init early_parse_fbmem(char *p)
+{
+	fbmem_size = memparse(p, &p);
+	if (*p == '@')
+		fbmem_start = memparse(p, &p);
+	return 0;
+}
+early_param("fbmem", early_parse_fbmem);
+
+static inline void __init resource_init(void)
+{
+	struct tag_mem_range *region;
+
+	kernel_code.start = __pa(init_mm.start_code);
+	kernel_code.end = __pa(init_mm.end_code - 1);
+	kernel_data.start = __pa(init_mm.end_code);
+	kernel_data.end = __pa(init_mm.brk - 1);
+
+	for (region = mem_phys; region; region = region->next) {
+		struct resource *res;
+		unsigned long phys_start, phys_end;
+
+		if (region->size == 0)
+			continue;
+
+		phys_start = region->addr;
+		phys_end = phys_start + region->size - 1;
+
+		res = alloc_bootmem_low(sizeof(*res));
+		res->name = "System RAM";
+		res->start = phys_start;
+		res->end = phys_end;
+		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+
+		request_resource (&iomem_resource, res);
+
+		if (kernel_code.start >= res->start &&
+		    kernel_code.end <= res->end)
+			request_resource (res, &kernel_code);
+		if (kernel_data.start >= res->start &&
+		    kernel_data.end <= res->end)
+			request_resource (res, &kernel_data);
+	}
+}
+
+static int __init parse_tag_core(struct tag *tag)
+{
+	if (tag->hdr.size > 2) {
+		if ((tag->u.core.flags & 1) == 0)
+			root_mountflags &= ~MS_RDONLY;
+		ROOT_DEV = new_decode_dev(tag->u.core.rootdev);
+	}
+	return 0;
+}
+__tagtable(ATAG_CORE, parse_tag_core);
+
+static int __init parse_tag_mem_range(struct tag *tag,
+				      struct tag_mem_range **root)
+{
+	struct tag_mem_range *cur, **pprev;
+	struct tag_mem_range *new;
+
+	/*
+	 * Ignore zero-sized entries. If we're running standalone, the
+	 * SDRAM code may emit such entries if something goes
+	 * wrong...
+	 */
+	if (tag->u.mem_range.size == 0)
+		return 0;
+
+	/*
+	 * Copy the data so the bootmem init code doesn't need to care
+	 * about it.
+	 */
+	if (mem_range_next_free >=
+	    (sizeof(mem_range_cache) / sizeof(mem_range_cache[0])))
+		panic("Physical memory map too complex!\n");
+
+	new = &mem_range_cache[mem_range_next_free++];
+	*new = tag->u.mem_range;
+
+	pprev = root;
+	cur = *root;
+	while (cur) {
+		pprev = &cur->next;
+		cur = cur->next;
+	}
+
+	*pprev = new;
+	new->next = NULL;
+
+	return 0;
+}
+
+static int __init parse_tag_mem(struct tag *tag)
+{
+	return parse_tag_mem_range(tag, &mem_phys);
+}
+__tagtable(ATAG_MEM, parse_tag_mem);
+
+static int __init parse_tag_cmdline(struct tag *tag)
+{
+	strlcpy(saved_command_line, tag->u.cmdline.cmdline, COMMAND_LINE_SIZE);
+	return 0;
+}
+__tagtable(ATAG_CMDLINE, parse_tag_cmdline);
+
+static int __init parse_tag_rdimg(struct tag *tag)
+{
+	return parse_tag_mem_range(tag, &mem_ramdisk);
+}
+__tagtable(ATAG_RDIMG, parse_tag_rdimg);
+
+static int __init parse_tag_clock(struct tag *tag)
+{
+	/*
+	 * We'll figure out the clocks by peeking at the system
+	 * manager regs directly.
+	 */
+	return 0;
+}
+__tagtable(ATAG_CLOCK, parse_tag_clock);
+
+static int __init parse_tag_rsvd_mem(struct tag *tag)
+{
+	return parse_tag_mem_range(tag, &mem_reserved);
+}
+__tagtable(ATAG_RSVD_MEM, parse_tag_rsvd_mem);
+
+static int __init parse_tag_ethernet(struct tag *tag)
+{
+#if 0
+	const struct platform_device *pdev;
+
+	/*
+	 * We really need a bus type that supports "classes"...this
+	 * will do for now (until we must handle other kinds of
+	 * ethernet controllers)
+	 */
+	pdev = platform_get_device("macb", tag->u.ethernet.mac_index);
+	if (pdev && pdev->dev.platform_data) {
+		struct eth_platform_data *data = pdev->dev.platform_data;
+
+		data->valid = 1;
+		data->mii_phy_addr = tag->u.ethernet.mii_phy_addr;
+		memcpy(data->hw_addr, tag->u.ethernet.hw_address,
+		       sizeof(data->hw_addr));
+	}
+#endif
+	return 0;
+}
+__tagtable(ATAG_ETHERNET, parse_tag_ethernet);
+
+/*
+ * Scan the tag table for this tag, and call its parse function. The
+ * tag table is built by the linker from all the __tagtable
+ * declarations.
+ */
+static int __init parse_tag(struct tag *tag)
+{
+	extern struct tagtable __tagtable_begin, __tagtable_end;
+	struct tagtable *t;
+
+	for (t = &__tagtable_begin; t < &__tagtable_end; t++)
+		if (tag->hdr.tag == t->tag) {
+			t->parse(tag);
+			break;
+		}
+
+	return t < &__tagtable_end;
+}
+
+/*
+ * Parse all tags in the list we got from the boot loader
+ */
+static void __init parse_tags(struct tag *t)
+{
+	for (; t->hdr.tag != ATAG_NONE; t = tag_next(t))
+		if (!parse_tag(t))
+			printk(KERN_WARNING
+			       "Ignoring unrecognised tag 0x%08x\n",
+			       t->hdr.tag);
+}
+
+void __init setup_arch (char **cmdline_p)
+{
+	struct clk *cpu_clk;
+
+	parse_tags(bootloader_tags);
+
+	setup_processor();
+	setup_platform();
+
+	cpu_clk = clk_get(NULL, "cpu");
+	if (IS_ERR(cpu_clk)) {
+		printk(KERN_WARNING "Warning: Unable to get CPU clock\n");
+	} else {
+		unsigned long cpu_hz = clk_get_rate(cpu_clk);
+
+		/*
+		 * Well, duh, but it's probably a good idea to
+		 * increment the use count.
+		 */
+		clk_enable(cpu_clk);
+
+		boot_cpu_data.clk = cpu_clk;
+		boot_cpu_data.loops_per_jiffy = cpu_hz * 4;
+		printk("CPU: Running at %lu.%03lu MHz\n",
+		       ((cpu_hz + 500) / 1000) / 1000,
+		       ((cpu_hz + 500) / 1000) % 1000);
+	}
+
+	init_mm.start_code = (unsigned long) &_text;
+	init_mm.end_code = (unsigned long) &_etext;
+	init_mm.end_data = (unsigned long) &_edata;
+	init_mm.brk = (unsigned long) &_end;
+
+	strlcpy(command_line, saved_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
+	parse_early_param();
+
+	setup_bootmem();
+
+	board_setup_fbmem(fbmem_start, fbmem_size);
+
+#ifdef CONFIG_VT
+	conswitchp = &dummy_con;
+#endif
+
+	paging_init();
+
+	resource_init();
+}
diff --git a/arch/avr32/kernel/signal.c b/arch/avr32/kernel/signal.c
new file mode 100644
index 000000000000..33096651c24f
--- /dev/null
+++ b/arch/avr32/kernel/signal.c
@@ -0,0 +1,328 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/sh/kernel/signal.c
+ *  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/suspend.h>
+
+#include <asm/uaccess.h>
+#include <asm/ucontext.h>
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+asmlinkage int sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+			       struct pt_regs *regs)
+{
+	return do_sigaltstack(uss, uoss, regs->sp);
+}
+
+struct rt_sigframe
+{
+	struct siginfo info;
+	struct ucontext uc;
+	unsigned long retcode;
+};
+
+static int
+restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	int err = 0;
+
+#define COPY(x)		err |= __get_user(regs->x, &sc->x)
+	COPY(sr);
+	COPY(pc);
+	COPY(lr);
+	COPY(sp);
+	COPY(r12);
+	COPY(r11);
+	COPY(r10);
+	COPY(r9);
+	COPY(r8);
+	COPY(r7);
+	COPY(r6);
+	COPY(r5);
+	COPY(r4);
+	COPY(r3);
+	COPY(r2);
+	COPY(r1);
+	COPY(r0);
+#undef	COPY
+
+	/*
+	 * Don't allow anyone to pretend they're running in supervisor
+	 * mode or something...
+	 */
+	err |= !valid_user_regs(regs);
+
+	return err;
+}
+
+
+asmlinkage int sys_rt_sigreturn(struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	sigset_t set;
+
+	frame = (struct rt_sigframe __user *)regs->sp;
+	pr_debug("SIG return: frame = %p\n", frame);
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext))
+		goto badframe;
+
+	pr_debug("Context restored: pc = %08lx, lr = %08lx, sp = %08lx\n",
+		 regs->pc, regs->lr, regs->sp);
+
+	return regs->r12;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+static int
+setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs)
+{
+	int err = 0;
+
+#define COPY(x)		err |= __put_user(regs->x, &sc->x)
+	COPY(sr);
+	COPY(pc);
+	COPY(lr);
+	COPY(sp);
+	COPY(r12);
+	COPY(r11);
+	COPY(r10);
+	COPY(r9);
+	COPY(r8);
+	COPY(r7);
+	COPY(r6);
+	COPY(r5);
+	COPY(r4);
+	COPY(r3);
+	COPY(r2);
+	COPY(r1);
+	COPY(r0);
+#undef	COPY
+
+	return err;
+}
+
+static inline void __user *
+get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, int framesize)
+{
+	unsigned long sp = regs->sp;
+
+	if ((ka->sa.sa_flags & SA_ONSTACK) && !sas_ss_flags(sp))
+		sp = current->sas_ss_sp + current->sas_ss_size;
+
+	return (void __user *)((sp - framesize) & ~3);
+}
+
+static int
+setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+	       sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ka, regs, sizeof(*frame));
+	err = -EFAULT;
+	if (!access_ok(VERIFY_WRITE, frame, sizeof (*frame)))
+		goto out;
+
+	/*
+	 * Set up the return code:
+	 *
+	 *	mov	r8, __NR_rt_sigreturn
+	 *	scall
+	 *
+	 * Note: This will blow up since we're using a non-executable
+	 * stack. Better use SA_RESTORER.
+	 */
+#if __NR_rt_sigreturn > 127
+# error __NR_rt_sigreturn must be < 127 to fit in a short mov
+#endif
+	err = __put_user(0x3008d733 | (__NR_rt_sigreturn << 20),
+			 &frame->retcode);
+
+	err |= copy_siginfo_to_user(&frame->info, info);
+
+	/* Set up the ucontext */
+	err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(NULL, &frame->uc.uc_link);
+	err |= __put_user((void __user *)current->sas_ss_sp,
+			  &frame->uc.uc_stack.ss_sp);
+	err |= __put_user(sas_ss_flags(regs->sp),
+			  &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(current->sas_ss_size,
+			  &frame->uc.uc_stack.ss_size);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	if (err)
+		goto out;
+
+	regs->r12 = sig;
+	regs->r11 = (unsigned long) &frame->info;
+	regs->r10 = (unsigned long) &frame->uc;
+	regs->sp = (unsigned long) frame;
+	if (ka->sa.sa_flags & SA_RESTORER)
+		regs->lr = (unsigned long)ka->sa.sa_restorer;
+	else {
+		printk(KERN_NOTICE "[%s:%d] did not set SA_RESTORER\n",
+		       current->comm, current->pid);
+		regs->lr = (unsigned long) &frame->retcode;
+	}
+
+	pr_debug("SIG deliver [%s:%d]: sig=%d sp=0x%lx pc=0x%lx->0x%p lr=0x%lx\n",
+		 current->comm, current->pid, sig, regs->sp,
+		 regs->pc, ka->sa.sa_handler, regs->lr);
+
+	regs->pc = (unsigned long) ka->sa.sa_handler;
+
+out:
+	return err;
+}
+
+static inline void restart_syscall(struct pt_regs *regs)
+{
+	if (regs->r12 == -ERESTART_RESTARTBLOCK)
+		regs->r8 = __NR_restart_syscall;
+	else
+		regs->r12 = regs->r12_orig;
+	regs->pc -= 2;
+}
+
+static inline void
+handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
+	      sigset_t *oldset, struct pt_regs *regs, int syscall)
+{
+	int ret;
+
+	/*
+	 * Set up the stack frame
+	 */
+	ret = setup_rt_frame(sig, ka, info, oldset, regs);
+
+	/*
+	 * Check that the resulting registers are sane
+	 */
+	ret |= !valid_user_regs(regs);
+
+	/*
+	 * Block the signal if we were unsuccessful.
+	 */
+	if (ret != 0 || !(ka->sa.sa_flags & SA_NODEFER)) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked, &current->blocked,
+			  &ka->sa.sa_mask);
+		sigaddset(&current->blocked, sig);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	if (ret == 0)
+		return;
+
+	force_sigsegv(sig, current);
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it
+ * doesn't want to handle. Thus you cannot kill init even with a
+ * SIGKILL even by mistake.
+ */
+int do_signal(struct pt_regs *regs, sigset_t *oldset, int syscall)
+{
+	siginfo_t info;
+	int signr;
+	struct k_sigaction ka;
+
+	/*
+	 * We want the common case to go fast, which is why we may in
+	 * certain cases get here from kernel mode. Just return
+	 * without doing anything if so.
+	 */
+	if (!user_mode(regs))
+		return 0;
+
+	if (try_to_freeze()) {
+		signr = 0;
+		if (!signal_pending(current))
+			goto no_signal;
+	}
+
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else if (!oldset)
+		oldset = &current->blocked;
+
+	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+no_signal:
+	if (syscall) {
+		switch (regs->r12) {
+		case -ERESTART_RESTARTBLOCK:
+		case -ERESTARTNOHAND:
+			if (signr > 0) {
+				regs->r12 = -EINTR;
+				break;
+			}
+			/* fall through */
+		case -ERESTARTSYS:
+			if (signr > 0 && !(ka.sa.sa_flags & SA_RESTART)) {
+				regs->r12 = -EINTR;
+				break;
+			}
+			/* fall through */
+		case -ERESTARTNOINTR:
+			restart_syscall(regs);
+		}
+	}
+
+	if (signr == 0) {
+		/* No signal to deliver -- put the saved sigmask back */
+		if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+			clear_thread_flag(TIF_RESTORE_SIGMASK);
+			sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+		}
+		return 0;
+	}
+
+	handle_signal(signr, &ka, &info, oldset, regs, syscall);
+	return 1;
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, struct thread_info *ti)
+{
+	int syscall = 0;
+
+	if ((sysreg_read(SR) & MODE_MASK) == MODE_SUPERVISOR)
+		syscall = 1;
+
+	if (ti->flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+		do_signal(regs, &current->blocked, syscall);
+}
diff --git a/arch/avr32/kernel/switch_to.S b/arch/avr32/kernel/switch_to.S
new file mode 100644
index 000000000000..a48d046723c5
--- /dev/null
+++ b/arch/avr32/kernel/switch_to.S
@@ -0,0 +1,35 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/sysreg.h>
+
+	.text
+	.global	__switch_to
+	.type	__switch_to, @function
+
+	/* Switch thread context from "prev" to "next", returning "last"
+	 *   r12 :	prev
+	 *   r11 :	&prev->thread + 1
+	 *   r10 :	&next->thread
+	 */
+__switch_to:
+	stm	--r11, r0,r1,r2,r3,r4,r5,r6,r7,sp,lr
+	mfsr	r9, SYSREG_SR
+	st.w	--r11, r9
+	ld.w	r8, r10++
+	/*
+	 * schedule() may have been called from a mode with a different
+	 * set of registers. Make sure we don't lose anything here.
+	 */
+	pushm	r10,r12
+	mtsr	SYSREG_SR, r8
+	frs			/* flush the return stack */
+	sub	pc, -2		/* flush the pipeline */
+	popm	r10,r12
+	ldm	r10++, r0,r1,r2,r3,r4,r5,r6,r7,sp,pc
+	.size	__switch_to, . - __switch_to
diff --git a/arch/avr32/kernel/sys_avr32.c b/arch/avr32/kernel/sys_avr32.c
new file mode 100644
index 000000000000..6ec5693da448
--- /dev/null
+++ b/arch/avr32/kernel/sys_avr32.c
@@ -0,0 +1,51 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mm.h>
+#include <linux/unistd.h>
+
+#include <asm/mman.h>
+#include <asm/uaccess.h>
+
+asmlinkage int sys_pipe(unsigned long __user *filedes)
+{
+	int fd[2];
+	int error;
+
+	error = do_pipe(fd);
+	if (!error) {
+		if (copy_to_user(filedes, fd, sizeof(fd)))
+			error = -EFAULT;
+	}
+	return error;
+}
+
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+			  unsigned long prot, unsigned long flags,
+			  unsigned long fd, off_t offset)
+{
+	int error = -EBADF;
+	struct file *file = NULL;
+
+	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
+	if (!(flags & MAP_ANONYMOUS)) {
+		file = fget(fd);
+		if (!file)
+			return error;
+	}
+
+	down_write(&current->mm->mmap_sem);
+	error = do_mmap_pgoff(file, addr, len, prot, flags, offset);
+	up_write(&current->mm->mmap_sem);
+
+	if (file)
+		fput(file);
+	return error;
+}
diff --git a/arch/avr32/kernel/syscall-stubs.S b/arch/avr32/kernel/syscall-stubs.S
new file mode 100644
index 000000000000..7589a9b426cb
--- /dev/null
+++ b/arch/avr32/kernel/syscall-stubs.S
@@ -0,0 +1,102 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * Stubs for syscalls that require access to pt_regs or that take more
+ * than five parameters.
+ */
+
+#define ARG6	r3
+
+	.text
+	.global __sys_rt_sigsuspend
+	.type	__sys_rt_sigsuspend,@function
+__sys_rt_sigsuspend:
+	mov	r10, sp
+	rjmp	sys_rt_sigsuspend
+
+	.global	__sys_sigaltstack
+	.type	__sys_sigaltstack,@function
+__sys_sigaltstack:
+	mov	r10, sp
+	rjmp	sys_sigaltstack
+
+	.global	__sys_rt_sigreturn
+	.type	__sys_rt_sigreturn,@function
+__sys_rt_sigreturn:
+	mov	r12, sp
+	rjmp	sys_rt_sigreturn
+
+	.global	__sys_fork
+	.type	__sys_fork,@function
+__sys_fork:
+	mov	r12, sp
+	rjmp	sys_fork
+
+	.global	__sys_clone
+	.type	__sys_clone,@function
+__sys_clone:
+	mov	r8, sp
+	rjmp	sys_clone
+
+	.global	__sys_vfork
+	.type	__sys_vfork,@function
+__sys_vfork:
+	mov	r12, sp
+	rjmp	sys_vfork
+
+	.global	__sys_execve
+	.type	__sys_execve,@function
+__sys_execve:
+	mov	r9, sp
+	rjmp	sys_execve
+
+	.global	__sys_mmap2
+	.type	__sys_mmap2,@function
+__sys_mmap2:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_mmap2
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_sendto
+	.type	__sys_sendto,@function
+__sys_sendto:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_sendto
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_recvfrom
+	.type	__sys_recvfrom,@function
+__sys_recvfrom:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_recvfrom
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_pselect6
+	.type	__sys_pselect6,@function
+__sys_pselect6:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_pselect6
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_splice
+	.type	__sys_splice,@function
+__sys_splice:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_splice
+	sub	sp, -4
+	popm	pc
diff --git a/arch/avr32/kernel/syscall_table.S b/arch/avr32/kernel/syscall_table.S
new file mode 100644
index 000000000000..63b206965d05
--- /dev/null
+++ b/arch/avr32/kernel/syscall_table.S
@@ -0,0 +1,289 @@
+/*
+ * AVR32 system call table
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
+#define sys_nfsservctl sys_ni_syscall
+#endif
+
+#if !defined(CONFIG_SYSV_IPC)
+# define sys_ipc	sys_ni_syscall
+#endif
+
+	.section .rodata,"a",@progbits
+	.type	sys_call_table,@object
+	.global	sys_call_table
+	.align	2
+sys_call_table:
+	.long	sys_restart_syscall
+	.long	sys_exit
+	.long	__sys_fork
+	.long	sys_read
+	.long	sys_write
+	.long	sys_open		/* 5 */
+	.long	sys_close
+	.long	sys_umask
+	.long	sys_creat
+	.long	sys_link
+	.long	sys_unlink		/* 10 */
+	.long	__sys_execve
+	.long	sys_chdir
+	.long	sys_time
+	.long	sys_mknod
+	.long	sys_chmod		/* 15 */
+	.long	sys_chown
+	.long	sys_lchown
+	.long	sys_lseek
+	.long	sys_llseek
+	.long	sys_getpid		/* 20 */
+	.long	sys_mount
+	.long	sys_umount
+	.long	sys_setuid
+	.long	sys_getuid
+	.long	sys_stime		/* 25 */
+	.long	sys_ptrace
+	.long	sys_alarm
+	.long	sys_pause
+	.long	sys_utime
+	.long	sys_newstat		/* 30 */
+	.long	sys_newfstat
+	.long	sys_newlstat
+	.long	sys_access
+	.long	sys_chroot
+	.long	sys_sync		/* 35 */
+	.long	sys_fsync
+	.long	sys_kill
+	.long	sys_rename
+	.long	sys_mkdir
+	.long	sys_rmdir		/* 40 */
+	.long	sys_dup
+	.long	sys_pipe
+	.long	sys_times
+	.long	__sys_clone
+	.long	sys_brk			/* 45 */
+	.long	sys_setgid
+	.long	sys_getgid
+	.long	sys_getcwd
+	.long	sys_geteuid
+	.long	sys_getegid		/* 50 */
+	.long	sys_acct
+	.long	sys_setfsuid
+	.long	sys_setfsgid
+	.long	sys_ioctl
+	.long	sys_fcntl		/* 55 */
+	.long	sys_setpgid
+	.long	sys_mremap
+	.long	sys_setresuid
+	.long	sys_getresuid
+	.long	sys_setreuid		/* 60 */
+	.long	sys_setregid
+	.long	sys_ustat
+	.long	sys_dup2
+	.long	sys_getppid
+	.long	sys_getpgrp		/* 65 */
+	.long	sys_setsid
+	.long	sys_rt_sigaction
+	.long	__sys_rt_sigreturn
+	.long	sys_rt_sigprocmask
+	.long	sys_rt_sigpending	/* 70 */
+	.long	sys_rt_sigtimedwait
+	.long	sys_rt_sigqueueinfo
+	.long	__sys_rt_sigsuspend
+	.long	sys_sethostname
+	.long	sys_setrlimit		/* 75 */
+	.long	sys_getrlimit
+	.long	sys_getrusage
+	.long	sys_gettimeofday
+	.long	sys_settimeofday
+	.long	sys_getgroups		/* 80 */
+	.long	sys_setgroups
+	.long	sys_select
+	.long	sys_symlink
+	.long	sys_fchdir
+	.long	sys_readlink		/* 85 */
+	.long	sys_pread64
+	.long	sys_pwrite64
+	.long	sys_swapon
+	.long	sys_reboot
+	.long	__sys_mmap2		/* 90 */
+	.long	sys_munmap
+	.long	sys_truncate
+	.long	sys_ftruncate
+	.long	sys_fchmod
+	.long	sys_fchown		/* 95 */
+	.long	sys_getpriority
+	.long	sys_setpriority
+	.long	sys_wait4
+	.long	sys_statfs
+	.long	sys_fstatfs		/* 100 */
+	.long	sys_vhangup
+	.long	__sys_sigaltstack
+	.long	sys_syslog
+	.long	sys_setitimer
+	.long	sys_getitimer		/* 105 */
+	.long	sys_swapoff
+	.long	sys_sysinfo
+	.long	sys_ipc
+	.long	sys_sendfile
+	.long	sys_setdomainname	/* 110 */
+	.long	sys_newuname
+	.long	sys_adjtimex
+	.long	sys_mprotect
+	.long	__sys_vfork
+	.long	sys_init_module		/* 115 */
+	.long	sys_delete_module
+	.long	sys_quotactl
+	.long	sys_getpgid
+	.long	sys_bdflush
+	.long	sys_sysfs		/* 120 */
+	.long	sys_personality
+	.long	sys_ni_syscall		/* reserved for afs_syscall */
+	.long	sys_getdents
+	.long	sys_flock
+	.long	sys_msync		/* 125 */
+	.long	sys_readv
+	.long	sys_writev
+	.long	sys_getsid
+	.long	sys_fdatasync
+	.long	sys_sysctl		/* 130 */
+	.long	sys_mlock
+	.long	sys_munlock
+	.long	sys_mlockall
+	.long	sys_munlockall
+	.long	sys_sched_setparam		/* 135 */
+	.long	sys_sched_getparam
+	.long	sys_sched_setscheduler
+	.long	sys_sched_getscheduler
+	.long	sys_sched_yield
+	.long	sys_sched_get_priority_max	/* 140 */
+	.long	sys_sched_get_priority_min
+	.long	sys_sched_rr_get_interval
+	.long	sys_nanosleep
+	.long	sys_poll
+	.long	sys_nfsservctl		/* 145 */
+	.long	sys_setresgid
+	.long	sys_getresgid
+	.long	sys_prctl
+	.long	sys_socket
+	.long	sys_bind		/* 150 */
+	.long	sys_connect
+	.long	sys_listen
+	.long	sys_accept
+	.long	sys_getsockname
+	.long	sys_getpeername		/* 155 */
+	.long	sys_socketpair
+	.long	sys_send
+	.long	sys_recv
+	.long	__sys_sendto
+	.long	__sys_recvfrom		/* 160 */
+	.long	sys_shutdown
+	.long	sys_setsockopt
+	.long	sys_getsockopt
+	.long	sys_sendmsg
+	.long	sys_recvmsg		/* 165 */
+	.long	sys_truncate64
+	.long	sys_ftruncate64
+	.long	sys_stat64
+	.long	sys_lstat64
+	.long	sys_fstat64		/* 170 */
+	.long	sys_pivot_root
+	.long	sys_mincore
+	.long	sys_madvise
+	.long	sys_getdents64
+	.long	sys_fcntl64		/* 175 */
+	.long	sys_gettid
+	.long	sys_readahead
+	.long	sys_setxattr
+	.long	sys_lsetxattr
+	.long	sys_fsetxattr		/* 180 */
+	.long	sys_getxattr
+	.long	sys_lgetxattr
+	.long	sys_fgetxattr
+	.long	sys_listxattr
+	.long	sys_llistxattr		/* 185 */
+	.long	sys_flistxattr
+	.long	sys_removexattr
+	.long	sys_lremovexattr
+	.long	sys_fremovexattr
+	.long	sys_tkill		/* 190 */
+	.long	sys_sendfile64
+	.long	sys_futex
+	.long	sys_sched_setaffinity
+	.long	sys_sched_getaffinity
+	.long	sys_capget		/* 195 */
+	.long	sys_capset
+	.long	sys_io_setup
+	.long	sys_io_destroy
+	.long	sys_io_getevents
+	.long	sys_io_submit		/* 200 */
+	.long	sys_io_cancel
+	.long	sys_fadvise64
+	.long	sys_exit_group
+	.long	sys_lookup_dcookie
+	.long	sys_epoll_create	/* 205 */
+	.long	sys_epoll_ctl
+	.long	sys_epoll_wait
+	.long	sys_remap_file_pages
+	.long	sys_set_tid_address
+	.long	sys_timer_create	/* 210 */
+	.long	sys_timer_settime
+	.long	sys_timer_gettime
+	.long	sys_timer_getoverrun
+	.long	sys_timer_delete
+	.long	sys_clock_settime	/* 215 */
+	.long	sys_clock_gettime
+	.long	sys_clock_getres
+	.long	sys_clock_nanosleep
+	.long	sys_statfs64
+	.long	sys_fstatfs64		/* 220 */
+	.long	sys_tgkill
+	.long	sys_ni_syscall		/* reserved for TUX */
+	.long	sys_utimes
+	.long	sys_fadvise64_64
+	.long	sys_cacheflush		/* 225 */
+	.long	sys_ni_syscall		/* sys_vserver */
+	.long	sys_mq_open
+	.long	sys_mq_unlink
+	.long	sys_mq_timedsend
+	.long	sys_mq_timedreceive	/* 230 */
+	.long	sys_mq_notify
+	.long	sys_mq_getsetattr
+	.long	sys_kexec_load
+	.long	sys_waitid
+	.long	sys_add_key		/* 235 */
+	.long	sys_request_key
+	.long	sys_keyctl
+	.long	sys_ioprio_set
+	.long	sys_ioprio_get
+	.long	sys_inotify_init	/* 240 */
+	.long	sys_inotify_add_watch
+	.long	sys_inotify_rm_watch
+	.long	sys_openat
+	.long	sys_mkdirat
+	.long	sys_mknodat		/* 245 */
+	.long	sys_fchownat
+	.long	sys_futimesat
+	.long	sys_fstatat64
+	.long	sys_unlinkat
+	.long	sys_renameat		/* 250 */
+	.long	sys_linkat
+	.long	sys_symlinkat
+	.long	sys_readlinkat
+	.long	sys_fchmodat
+	.long	sys_faccessat		/* 255 */
+	.long	__sys_pselect6
+	.long	sys_ppoll
+	.long	sys_unshare
+	.long	sys_set_robust_list
+	.long	sys_get_robust_list	/* 260 */
+	.long	__sys_splice
+	.long	sys_sync_file_range
+	.long	sys_tee
+	.long	sys_vmsplice
+	.long	sys_ni_syscall		/* r8 is saturated at nr_syscalls */
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
new file mode 100644
index 000000000000..b0e6b5855a38
--- /dev/null
+++ b/arch/avr32/kernel/time.c
@@ -0,0 +1,238 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on MIPS implementation arch/mips/kernel/time.c
+ *   Copyright 2001 MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/clocksource.h>
+#include <linux/time.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel_stat.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/profile.h>
+#include <linux/sysdev.h>
+
+#include <asm/div64.h>
+#include <asm/sysreg.h>
+#include <asm/io.h>
+#include <asm/sections.h>
+
+static cycle_t read_cycle_count(void)
+{
+	return (cycle_t)sysreg_read(COUNT);
+}
+
+static struct clocksource clocksource_avr32 = {
+	.name		= "avr32",
+	.rating		= 350,
+	.read		= read_cycle_count,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.shift		= 16,
+	.is_continuous	= 1,
+};
+
+/*
+ * By default we provide the null RTC ops
+ */
+static unsigned long null_rtc_get_time(void)
+{
+	return mktime(2004, 1, 1, 0, 0, 0);
+}
+
+static int null_rtc_set_time(unsigned long sec)
+{
+	return 0;
+}
+
+static unsigned long (*rtc_get_time)(void) = null_rtc_get_time;
+static int (*rtc_set_time)(unsigned long) = null_rtc_set_time;
+
+/* how many counter cycles in a jiffy? */
+static unsigned long cycles_per_jiffy;
+
+/* cycle counter value at the previous timer interrupt */
+static unsigned int timerhi, timerlo;
+
+/* the count value for the next timer interrupt */
+static unsigned int expirelo;
+
+static void avr32_timer_ack(void)
+{
+	unsigned int count;
+
+	/* Ack this timer interrupt and set the next one */
+	expirelo += cycles_per_jiffy;
+	if (expirelo == 0) {
+		printk(KERN_DEBUG "expirelo == 0\n");
+		sysreg_write(COMPARE, expirelo + 1);
+	} else {
+		sysreg_write(COMPARE, expirelo);
+	}
+
+	/* Check to see if we have missed any timer interrupts */
+	count = sysreg_read(COUNT);
+	if ((count - expirelo) < 0x7fffffff) {
+		expirelo = count + cycles_per_jiffy;
+		sysreg_write(COMPARE, expirelo);
+	}
+}
+
+static unsigned int avr32_hpt_read(void)
+{
+	return sysreg_read(COUNT);
+}
+
+/*
+ * Taken from MIPS c0_hpt_timer_init().
+ *
+ * Why is it so complicated, and what is "count"?  My assumption is
+ * that `count' specifies the "reference cycle", i.e. the cycle since
+ * reset that should mean "zero". The reason COUNT is written twice is
+ * probably to make sure we don't get any timer interrupts while we
+ * are messing with the counter.
+ */
+static void avr32_hpt_init(unsigned int count)
+{
+	count = sysreg_read(COUNT) - count;
+	expirelo = (count / cycles_per_jiffy + 1) * cycles_per_jiffy;
+	sysreg_write(COUNT, expirelo - cycles_per_jiffy);
+	sysreg_write(COMPARE, expirelo);
+	sysreg_write(COUNT, count);
+}
+
+/*
+ * Scheduler clock - returns current time in nanosec units.
+ */
+unsigned long long sched_clock(void)
+{
+	/* There must be better ways...? */
+	return (unsigned long long)jiffies * (1000000000 / HZ);
+}
+
+/*
+ * local_timer_interrupt() does profiling and process accounting on a
+ * per-CPU basis.
+ *
+ * In UP mode, it is invoked from the (global) timer_interrupt.
+ */
+static void local_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	if (current->pid)
+		profile_tick(CPU_PROFILING, regs);
+	update_process_times(user_mode(regs));
+}
+
+static irqreturn_t
+timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	unsigned int count;
+
+	/* ack timer interrupt and try to set next interrupt */
+	count = avr32_hpt_read();
+	avr32_timer_ack();
+
+	/* Update timerhi/timerlo for intra-jiffy calibration */
+	timerhi += count < timerlo;	/* Wrap around */
+	timerlo = count;
+
+	/*
+	 * Call the generic timer interrupt handler
+	 */
+	write_seqlock(&xtime_lock);
+	do_timer(regs);
+	write_sequnlock(&xtime_lock);
+
+	/*
+	 * In UP mode, we call local_timer_interrupt() to do profiling
+	 * and process accounting.
+	 *
+	 * SMP is not supported yet.
+	 */
+	local_timer_interrupt(irq, dev_id, regs);
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction timer_irqaction = {
+	.handler	= timer_interrupt,
+	.flags		= IRQF_DISABLED,
+	.name		= "timer",
+};
+
+void __init time_init(void)
+{
+	unsigned long mult, shift, count_hz;
+	int ret;
+
+	xtime.tv_sec = rtc_get_time();
+	xtime.tv_nsec = 0;
+
+	set_normalized_timespec(&wall_to_monotonic,
+				-xtime.tv_sec, -xtime.tv_nsec);
+
+	printk("Before time_init: count=%08lx, compare=%08lx\n",
+	       (unsigned long)sysreg_read(COUNT),
+	       (unsigned long)sysreg_read(COMPARE));
+
+	count_hz = clk_get_rate(boot_cpu_data.clk);
+	shift = clocksource_avr32.shift;
+	mult = clocksource_hz2mult(count_hz, shift);
+	clocksource_avr32.mult = mult;
+
+	printk("Cycle counter: mult=%lu, shift=%lu\n", mult, shift);
+
+	{
+		u64 tmp;
+
+		tmp = TICK_NSEC;
+		tmp <<= shift;
+		tmp += mult / 2;
+		do_div(tmp, mult);
+
+		cycles_per_jiffy = tmp;
+	}
+
+	/* This sets up the high precision timer for the first interrupt. */
+	avr32_hpt_init(avr32_hpt_read());
+
+	printk("After time_init: count=%08lx, compare=%08lx\n",
+	       (unsigned long)sysreg_read(COUNT),
+	       (unsigned long)sysreg_read(COMPARE));
+
+	ret = clocksource_register(&clocksource_avr32);
+	if (ret)
+		printk(KERN_ERR
+		       "timer: could not register clocksource: %d\n", ret);
+
+	ret = setup_irq(0, &timer_irqaction);
+	if (ret)
+		printk("timer: could not request IRQ 0: %d\n", ret);
+}
+
+static struct sysdev_class timer_class = {
+	set_kset_name("timer"),
+};
+
+static struct sys_device timer_device = {
+	.id	= 0,
+	.cls	= &timer_class,
+};
+
+static int __init init_timer_sysfs(void)
+{
+	int err = sysdev_class_register(&timer_class);
+	if (!err)
+		err = sysdev_register(&timer_device);
+	return err;
+}
+
+device_initcall(init_timer_sysfs);
diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c
new file mode 100644
index 000000000000..7e803f4d7a12
--- /dev/null
+++ b/arch/avr32/kernel/traps.c
@@ -0,0 +1,425 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#undef DEBUG
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+#include <linux/notifier.h>
+
+#include <asm/traps.h>
+#include <asm/sysreg.h>
+#include <asm/addrspace.h>
+#include <asm/ocd.h>
+#include <asm/mmu_context.h>
+#include <asm/uaccess.h>
+
+static void dump_mem(const char *str, unsigned long bottom, unsigned long top)
+{
+	unsigned long p;
+	int i;
+
+	printk("%s(0x%08lx to 0x%08lx)\n", str, bottom, top);
+
+	for (p = bottom & ~31; p < top; ) {
+		printk("%04lx: ", p & 0xffff);
+
+		for (i = 0; i < 8; i++, p += 4) {
+			unsigned int val;
+
+			if (p < bottom || p >= top)
+				printk("         ");
+			else {
+				if (__get_user(val, (unsigned int __user *)p)) {
+					printk("\n");
+					goto out;
+				}
+				printk("%08x ", val);
+			}
+		}
+		printk("\n");
+	}
+
+out:
+	return;
+}
+
+#ifdef CONFIG_FRAME_POINTER
+static inline void __show_trace(struct task_struct *tsk, unsigned long *sp,
+				struct pt_regs *regs)
+{
+	unsigned long __user *fp;
+	unsigned long __user *last_fp = NULL;
+
+	if (regs) {
+		fp = (unsigned long __user *)regs->r7;
+	} else if (tsk == current) {
+		register unsigned long __user *real_fp __asm__("r7");
+		fp = real_fp;
+	} else {
+		fp = (unsigned long __user *)tsk->thread.cpu_context.r7;
+	}
+
+	/*
+	 * Walk the stack until (a) we get an exception, (b) the frame
+	 * pointer becomes zero, or (c) the frame pointer gets stuck
+	 * at the same value.
+	 */
+	while (fp && fp != last_fp) {
+		unsigned long lr, new_fp = 0;
+
+		last_fp = fp;
+		if (__get_user(lr, fp))
+			break;
+		if (fp && __get_user(new_fp, fp + 1))
+			break;
+		fp = (unsigned long __user *)new_fp;
+
+		printk(" [<%08lx>] ", lr);
+		print_symbol("%s\n", lr);
+	}
+	printk("\n");
+}
+#else
+static inline void __show_trace(struct task_struct *tsk, unsigned long *sp,
+				struct pt_regs *regs)
+{
+	unsigned long addr;
+
+	while (!kstack_end(sp)) {
+		addr = *sp++;
+		if (kernel_text_address(addr)) {
+			printk(" [<%08lx>] ", addr);
+			print_symbol("%s\n", addr);
+		}
+	}
+}
+#endif
+
+void show_trace(struct task_struct *tsk, unsigned long *sp,
+		       struct pt_regs *regs)
+{
+	if (regs &&
+	    (((regs->sr & MODE_MASK) == MODE_EXCEPTION) ||
+	     ((regs->sr & MODE_MASK) == MODE_USER)))
+		return;
+
+	printk ("Call trace:");
+#ifdef CONFIG_KALLSYMS
+	printk("\n");
+#endif
+
+	__show_trace(tsk, sp, regs);
+	printk("\n");
+}
+
+void show_stack(struct task_struct *tsk, unsigned long *sp)
+{
+	unsigned long stack;
+
+	if (!tsk)
+		tsk = current;
+	if (sp == 0) {
+		if (tsk == current) {
+			register unsigned long *real_sp __asm__("sp");
+			sp = real_sp;
+		} else {
+			sp = (unsigned long *)tsk->thread.cpu_context.ksp;
+		}
+	}
+
+	stack = (unsigned long)sp;
+	dump_mem("Stack: ", stack,
+		 THREAD_SIZE + (unsigned long)tsk->thread_info);
+	show_trace(tsk, sp, NULL);
+}
+
+void dump_stack(void)
+{
+	show_stack(NULL, NULL);
+}
+EXPORT_SYMBOL(dump_stack);
+
+ATOMIC_NOTIFIER_HEAD(avr32_die_chain);
+
+int register_die_notifier(struct notifier_block *nb)
+{
+	pr_debug("register_die_notifier: %p\n", nb);
+
+	return atomic_notifier_chain_register(&avr32_die_chain, nb);
+}
+EXPORT_SYMBOL(register_die_notifier);
+
+int unregister_die_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_unregister(&avr32_die_chain, nb);
+}
+EXPORT_SYMBOL(unregister_die_notifier);
+
+static DEFINE_SPINLOCK(die_lock);
+
+void __die(const char *str, struct pt_regs *regs, unsigned long err,
+	   const char *file, const char *func, unsigned long line)
+{
+	struct task_struct *tsk = current;
+	static int die_counter;
+
+	console_verbose();
+	spin_lock_irq(&die_lock);
+	bust_spinlocks(1);
+
+	printk(KERN_ALERT "%s", str);
+	if (file && func)
+		printk(" in %s:%s, line %ld", file, func, line);
+	printk("[#%d]:\n", ++die_counter);
+	print_modules();
+	show_regs(regs);
+	printk("Process %s (pid: %d, stack limit = 0x%p)\n",
+	       tsk->comm, tsk->pid, tsk->thread_info + 1);
+
+	if (!user_mode(regs) || in_interrupt()) {
+		dump_mem("Stack: ", regs->sp,
+			 THREAD_SIZE + (unsigned long)tsk->thread_info);
+	}
+
+	bust_spinlocks(0);
+	spin_unlock_irq(&die_lock);
+	do_exit(SIGSEGV);
+}
+
+void __die_if_kernel(const char *str, struct pt_regs *regs, unsigned long err,
+		     const char *file, const char *func, unsigned long line)
+{
+	if (!user_mode(regs))
+		__die(str, regs, err, file, func, line);
+}
+
+asmlinkage void do_nmi(unsigned long ecr, struct pt_regs *regs)
+{
+#ifdef CONFIG_SUBARCH_AVR32B
+	/*
+	 * The exception entry always saves RSR_EX. For NMI, this is
+	 * wrong; it should be RSR_NMI
+	 */
+	regs->sr = sysreg_read(RSR_NMI);
+#endif
+
+	printk("NMI taken!!!!\n");
+	die("NMI", regs, ecr);
+	BUG();
+}
+
+asmlinkage void do_critical_exception(unsigned long ecr, struct pt_regs *regs)
+{
+	printk("Unable to handle critical exception %lu at pc = %08lx!\n",
+	       ecr, regs->pc);
+	die("Oops", regs, ecr);
+	BUG();
+}
+
+asmlinkage void do_address_exception(unsigned long ecr, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	die_if_kernel("Oops: Address exception in kernel mode", regs, ecr);
+
+#ifdef DEBUG
+	if (ecr == ECR_ADDR_ALIGN_X)
+		pr_debug("Instruction Address Exception at pc = %08lx\n",
+			 regs->pc);
+	else if (ecr == ECR_ADDR_ALIGN_R)
+		pr_debug("Data Address Exception (Read) at pc = %08lx\n",
+			 regs->pc);
+	else if (ecr == ECR_ADDR_ALIGN_W)
+		pr_debug("Data Address Exception (Write) at pc = %08lx\n",
+			 regs->pc);
+	else
+		BUG();
+
+	show_regs(regs);
+#endif
+
+	info.si_signo = SIGBUS;
+	info.si_errno = 0;
+	info.si_code = BUS_ADRALN;
+	info.si_addr = (void __user *)regs->pc;
+
+	force_sig_info(SIGBUS, &info, current);
+}
+
+/* This way of handling undefined instructions is stolen from ARM */
+static LIST_HEAD(undef_hook);
+static spinlock_t undef_lock = SPIN_LOCK_UNLOCKED;
+
+void register_undef_hook(struct undef_hook *hook)
+{
+	spin_lock_irq(&undef_lock);
+	list_add(&hook->node, &undef_hook);
+	spin_unlock_irq(&undef_lock);
+}
+
+void unregister_undef_hook(struct undef_hook *hook)
+{
+	spin_lock_irq(&undef_lock);
+	list_del(&hook->node);
+	spin_unlock_irq(&undef_lock);
+}
+
+static int do_cop_absent(u32 insn)
+{
+	int cop_nr;
+	u32 cpucr;
+	if ( (insn & 0xfdf00000) == 0xf1900000 )
+		/* LDC0 */
+		cop_nr = 0;
+	else
+		cop_nr = (insn >> 13) & 0x7;
+
+	/* Try enabling the coprocessor */
+	cpucr = sysreg_read(CPUCR);
+	cpucr |= (1 << (24 + cop_nr));
+	sysreg_write(CPUCR, cpucr);
+
+	cpucr = sysreg_read(CPUCR);
+	if ( !(cpucr & (1 << (24 + cop_nr))) ){
+		printk("Coprocessor #%i not found!\n", cop_nr);
+		return -1;
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_BUG
+#ifdef CONFIG_DEBUG_BUGVERBOSE
+static inline void do_bug_verbose(struct pt_regs *regs, u32 insn)
+{
+	char *file;
+	u16 line;
+	char c;
+
+	if (__get_user(line, (u16 __user *)(regs->pc + 2)))
+		return;
+	if (__get_user(file, (char * __user *)(regs->pc + 4))
+	    || (unsigned long)file < PAGE_OFFSET
+	    || __get_user(c, file))
+		file = "<bad filename>";
+
+	printk(KERN_ALERT "kernel BUG at %s:%d!\n", file, line);
+}
+#else
+static inline void do_bug_verbose(struct pt_regs *regs, u32 insn)
+{
+
+}
+#endif
+#endif
+
+asmlinkage void do_illegal_opcode(unsigned long ecr, struct pt_regs *regs)
+{
+	u32 insn;
+	struct undef_hook *hook;
+	siginfo_t info;
+	void __user *pc;
+
+	if (!user_mode(regs))
+		goto kernel_trap;
+
+	local_irq_enable();
+
+	pc = (void __user *)instruction_pointer(regs);
+	if (__get_user(insn, (u32 __user *)pc))
+		goto invalid_area;
+
+        if (ecr == ECR_COPROC_ABSENT) {
+		if (do_cop_absent(insn) == 0)
+			return;
+        }
+
+	spin_lock_irq(&undef_lock);
+	list_for_each_entry(hook, &undef_hook, node) {
+		if ((insn & hook->insn_mask) == hook->insn_val) {
+			if (hook->fn(regs, insn) == 0) {
+				spin_unlock_irq(&undef_lock);
+				return;
+			}
+		}
+	}
+	spin_unlock_irq(&undef_lock);
+
+invalid_area:
+
+#ifdef DEBUG
+	printk("Illegal instruction at pc = %08lx\n", regs->pc);
+	if (regs->pc < TASK_SIZE) {
+		unsigned long ptbr, pgd, pte, *p;
+
+		ptbr = sysreg_read(PTBR);
+		p = (unsigned long *)ptbr;
+		pgd = p[regs->pc >> 22];
+		p = (unsigned long *)((pgd & 0x1ffff000) | 0x80000000);
+		pte = p[(regs->pc >> 12) & 0x3ff];
+		printk("page table: 0x%08lx -> 0x%08lx -> 0x%08lx\n", ptbr, pgd, pte);
+	}
+#endif
+
+	info.si_signo = SIGILL;
+	info.si_errno = 0;
+	info.si_addr = (void __user *)regs->pc;
+	switch (ecr) {
+	case ECR_ILLEGAL_OPCODE:
+	case ECR_UNIMPL_INSTRUCTION:
+		info.si_code = ILL_ILLOPC;
+		break;
+	case ECR_PRIVILEGE_VIOLATION:
+		info.si_code = ILL_PRVOPC;
+		break;
+	case ECR_COPROC_ABSENT:
+		info.si_code = ILL_COPROC;
+		break;
+	default:
+		BUG();
+	}
+
+	force_sig_info(SIGILL, &info, current);
+	return;
+
+kernel_trap:
+#ifdef CONFIG_BUG
+	if (__kernel_text_address(instruction_pointer(regs))) {
+		insn = *(u16 *)instruction_pointer(regs);
+		if (insn == AVR32_BUG_OPCODE) {
+			do_bug_verbose(regs, insn);
+			die("Kernel BUG", regs, 0);
+			return;
+		}
+	}
+#endif
+
+	die("Oops: Illegal instruction in kernel code", regs, ecr);
+}
+
+asmlinkage void do_fpe(unsigned long ecr, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	printk("Floating-point exception at pc = %08lx\n", regs->pc);
+
+	/* We have no FPU... */
+	info.si_signo = SIGILL;
+	info.si_errno = 0;
+	info.si_addr = (void __user *)regs->pc;
+	info.si_code = ILL_COPROC;
+
+	force_sig_info(SIGILL, &info, current);
+}
+
+
+void __init trap_init(void)
+{
+
+}
diff --git a/arch/avr32/kernel/vmlinux.lds.c b/arch/avr32/kernel/vmlinux.lds.c
new file mode 100644
index 000000000000..cdd627c6b7dc
--- /dev/null
+++ b/arch/avr32/kernel/vmlinux.lds.c
@@ -0,0 +1,139 @@
+/*
+ * AVR32 linker script for the Linux kernel
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#define LOAD_OFFSET 0x00000000
+#include <asm-generic/vmlinux.lds.h>
+
+OUTPUT_FORMAT("elf32-avr32", "elf32-avr32", "elf32-avr32")
+OUTPUT_ARCH(avr32)
+ENTRY(_start)
+
+/* Big endian */
+jiffies = jiffies_64 + 4;
+
+SECTIONS
+{
+	. = CONFIG_ENTRY_ADDRESS;
+	.init		: AT(ADDR(.init) - LOAD_OFFSET) {
+		_stext = .;
+		__init_begin = .;
+			_sinittext = .;
+			*(.text.reset)
+			*(.init.text)
+			_einittext = .;
+		. = ALIGN(4);
+		__tagtable_begin = .;
+			*(.taglist)
+		__tagtable_end = .;
+			*(.init.data)
+		. = ALIGN(16);
+		__setup_start = .;
+			*(.init.setup)
+		__setup_end = .;
+		. = ALIGN(4);
+		__initcall_start = .;
+			*(.initcall1.init)
+			*(.initcall2.init)
+			*(.initcall3.init)
+			*(.initcall4.init)
+			*(.initcall5.init)
+			*(.initcall6.init)
+			*(.initcall7.init)
+		__initcall_end = .;
+		__con_initcall_start = .;
+			*(.con_initcall.init)
+		__con_initcall_end = .;
+		__security_initcall_start = .;
+			*(.security_initcall.init)
+		__security_initcall_end = .;
+		. = ALIGN(32);
+		__initramfs_start = .;
+			*(.init.ramfs)
+		__initramfs_end = .;
+		. = ALIGN(4096);
+		__init_end = .;
+	}
+
+	. = ALIGN(8192);
+	.text		: AT(ADDR(.text) - LOAD_OFFSET) {
+		_evba = .;
+		_text = .;
+		*(.ex.text)
+		. = 0x50;
+		*(.tlbx.ex.text)
+		. = 0x60;
+		*(.tlbr.ex.text)
+		. = 0x70;
+		*(.tlbw.ex.text)
+		. = 0x100;
+		*(.scall.text)
+		*(.irq.text)
+		*(.text)
+		SCHED_TEXT
+		LOCK_TEXT
+		KPROBES_TEXT
+		*(.fixup)
+		*(.gnu.warning)
+		_etext = .;
+	} = 0xd703d703
+
+	. = ALIGN(4);
+	__ex_table	: AT(ADDR(__ex_table) - LOAD_OFFSET) {
+		__start___ex_table = .;
+		*(__ex_table)
+		__stop___ex_table = .;
+	}
+
+	RODATA
+
+	. = ALIGN(8192);
+
+	.data		: AT(ADDR(.data) - LOAD_OFFSET) {
+		_data = .;
+		_sdata = .;
+		/*
+		 * First, the init task union, aligned to an 8K boundary.
+		 */
+		*(.data.init_task)
+
+		/* Then, the cacheline aligned data */
+		. = ALIGN(32);
+		*(.data.cacheline_aligned)
+
+		/* And the rest... */
+		*(.data.rel*)
+		*(.data)
+		CONSTRUCTORS
+
+		_edata = .;
+	}
+
+
+	. = ALIGN(8);
+	.bss    	: AT(ADDR(.bss) - LOAD_OFFSET) {
+		__bss_start = .;
+		*(.bss)
+		*(COMMON)
+		. = ALIGN(8);
+		__bss_stop = .;
+		_end = .;
+	}
+
+	/* When something in the kernel is NOT compiled as a module, the module
+	 * cleanup code and data are put into these segments. Both can then be
+	 * thrown away, as cleanup code is never called unless it's a module.
+	 */
+	/DISCARD/       	: {
+		*(.exit.text)
+		*(.exit.data)
+		*(.exitcall.exit)
+	}
+
+	DWARF_DEBUG
+}
diff --git a/arch/avr32/lib/Makefile b/arch/avr32/lib/Makefile
new file mode 100644
index 000000000000..09ac43e40522
--- /dev/null
+++ b/arch/avr32/lib/Makefile
@@ -0,0 +1,10 @@
+#
+# Makefile for AVR32-specific library files
+#
+
+lib-y	:= copy_user.o clear_user.o
+lib-y	+= strncpy_from_user.o strnlen_user.o
+lib-y	+= delay.o memset.o memcpy.o findbit.o
+lib-y	+= csum_partial.o csum_partial_copy_generic.o
+lib-y	+= io-readsw.o io-readsl.o io-writesw.o io-writesl.o
+lib-y	+= __avr32_lsl64.o __avr32_lsr64.o __avr32_asr64.o
diff --git a/arch/avr32/lib/__avr32_asr64.S b/arch/avr32/lib/__avr32_asr64.S
new file mode 100644
index 000000000000..368b6bca4c76
--- /dev/null
+++ b/arch/avr32/lib/__avr32_asr64.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * DWtype __avr32_asr64(DWtype u, word_type b)
+	 */
+	.text
+	.global	__avr32_asr64
+	.type	__avr32_asr64,@function
+__avr32_asr64:
+	cp.w	r12, 0
+	reteq	r12
+
+	rsub	r9, r12, 32
+	brle	1f
+
+	lsl	r8, r11, r9
+	lsr	r10, r10, r12
+	asr	r11, r11, r12
+	or	r10, r8
+	retal	r12
+
+1:	neg	r9
+	asr	r10, r11, r9
+	asr	r11, 31
+	retal	r12
diff --git a/arch/avr32/lib/__avr32_lsl64.S b/arch/avr32/lib/__avr32_lsl64.S
new file mode 100644
index 000000000000..f1dbc2b36257
--- /dev/null
+++ b/arch/avr32/lib/__avr32_lsl64.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * DWtype __avr32_lsl64(DWtype u, word_type b)
+	 */
+	.text
+	.global	__avr32_lsl64
+	.type	__avr32_lsl64,@function
+__avr32_lsl64:
+	cp.w	r12, 0
+	reteq	r12
+
+	rsub	r9, r12, 32
+	brle	1f
+
+	lsr	r8, r10, r9
+	lsl	r10, r10, r12
+	lsl	r11, r11, r12
+	or	r11, r8
+	retal	r12
+
+1:	neg	r9
+	lsl	r11, r10, r9
+	mov	r10, 0
+	retal	r12
diff --git a/arch/avr32/lib/__avr32_lsr64.S b/arch/avr32/lib/__avr32_lsr64.S
new file mode 100644
index 000000000000..e65bb7f0d24c
--- /dev/null
+++ b/arch/avr32/lib/__avr32_lsr64.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * DWtype __avr32_lsr64(DWtype u, word_type b)
+	 */
+	.text
+	.global	__avr32_lsr64
+	.type	__avr32_lsr64,@function
+__avr32_lsr64:
+	cp.w	r12, 0
+	reteq	r12
+
+	rsub	r9, r12, 32
+	brle	1f
+
+	lsl	r8, r11, r9
+	lsr	r11, r11, r12
+	lsr	r10, r10, r12
+	or	r10, r8
+	retal	r12
+
+1:	neg	r9
+	lsr	r10, r11, r9
+	mov	r11, 0
+	retal	r12
diff --git a/arch/avr32/lib/clear_user.S b/arch/avr32/lib/clear_user.S
new file mode 100644
index 000000000000..d8991b6f8eb7
--- /dev/null
+++ b/arch/avr32/lib/clear_user.S
@@ -0,0 +1,76 @@
+/*
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/asm.h>
+
+	.text
+	.align	1
+	.global	clear_user
+	.type	clear_user, "function"
+clear_user:
+	branch_if_kernel r8, __clear_user
+	ret_if_privileged r8, r12, r11, r11
+
+	.global	__clear_user
+	.type	__clear_user, "function"
+__clear_user:
+	mov	r9, r12
+	mov	r8, 0
+	andl	r9, 3, COH
+	brne	5f
+
+1:	sub	r11, 4
+	brlt	2f
+
+10:	st.w	r12++, r8
+	sub	r11, 4
+	brge	10b
+
+2:	sub	r11, -4
+	reteq	0
+
+	/* Unaligned count or address */
+	bld	r11, 1
+	brcc	12f
+11:	st.h	r12++, r8
+	sub	r11, 2
+	reteq	0
+12:	st.b	r12++, r8
+	retal	0
+
+	/* Unaligned address */
+5:	cp.w	r11, 4
+	brlt	2b
+
+	lsl	r9, 2
+	add	pc, pc, r9
+13:	st.b	r12++, r8
+	sub	r11, 1
+14:	st.b	r12++, r8
+	sub	r11, 1
+15:	st.b	r12++, r8
+	sub	r11, 1
+	rjmp	1b
+
+	.size	clear_user, . - clear_user
+	.size	__clear_user, . - __clear_user
+
+	.section .fixup, "ax"
+	.align	1
+18:	sub	r11, -4
+19:	retal	r11
+
+	.section __ex_table, "a"
+	.align	2
+	.long	10b, 18b
+	.long	11b, 19b
+	.long	12b, 19b
+	.long	13b, 19b
+	.long	14b, 19b
+	.long	15b, 19b
diff --git a/arch/avr32/lib/copy_user.S b/arch/avr32/lib/copy_user.S
new file mode 100644
index 000000000000..ea59c04b07de
--- /dev/null
+++ b/arch/avr32/lib/copy_user.S
@@ -0,0 +1,119 @@
+/*
+ * Copy to/from userspace with optional address space checking.
+ *
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/asm.h>
+
+	/*
+	 * __kernel_size_t
+	 * __copy_user(void *to, const void *from, __kernel_size_t n)
+	 *
+	 * Returns the number of bytes not copied. Might be off by
+	 * max 3 bytes if we get a fault in the main loop.
+	 *
+	 * The address-space checking functions simply fall through to
+	 * the non-checking version.
+	 */
+	.text
+	.align	1
+	.global	copy_from_user
+	.type	copy_from_user, @function
+copy_from_user:
+	branch_if_kernel r8, __copy_user
+	ret_if_privileged r8, r11, r10, r10
+	rjmp	__copy_user
+	.size	copy_from_user, . - copy_from_user
+
+	.global	copy_to_user
+	.type	copy_to_user, @function
+copy_to_user:
+	branch_if_kernel r8, __copy_user
+	ret_if_privileged r8, r12, r10, r10
+	.size	copy_to_user, . - copy_to_user
+
+	.global	__copy_user
+	.type	__copy_user, @function
+__copy_user:
+	mov	r9, r11
+	andl	r9, 3, COH
+	brne	6f
+
+	/* At this point, from is word-aligned */
+1:	sub	r10, 4
+	brlt	3f
+
+2:
+10:	ld.w	r8, r11++
+11:	st.w	r12++, r8
+	sub	r10, 4
+	brge	2b
+
+3:	sub	r10, -4
+	reteq	0
+
+	/*
+	 * Handle unaligned count. Need to be careful with r10 here so
+	 * that we return the correct value even if we get a fault
+	 */
+4:
+20:	ld.ub	r8, r11++
+21:	st.b	r12++, r8
+	sub	r10, 1
+	reteq	0
+22:	ld.ub	r8, r11++
+23:	st.b	r12++, r8
+	sub	r10, 1
+	reteq	0
+24:	ld.ub	r8, r11++
+25:	st.b	r12++, r8
+	retal	0
+
+	/* Handle unaligned from-pointer */
+6:	cp.w	r10, 4
+	brlt	4b
+	rsub	r9, r9, 4
+
+30:	ld.ub	r8, r11++
+31:	st.b	r12++, r8
+	sub	r10, 1
+	sub	r9, 1
+	breq	1b
+32:	ld.ub	r8, r11++
+33:	st.b	r12++, r8
+	sub	r10, 1
+	sub	r9, 1
+	breq	1b
+34:	ld.ub	r8, r11++
+35:	st.b	r12++, r8
+	sub	r10, 1
+	rjmp	1b
+	.size	__copy_user, . - __copy_user
+
+	.section .fixup,"ax"
+	.align	1
+19:	sub	r10, -4
+29:	retal	r10
+
+	.section __ex_table,"a"
+	.align	2
+	.long	10b, 19b
+	.long	11b, 19b
+	.long	20b, 29b
+	.long	21b, 29b
+	.long	22b, 29b
+	.long	23b, 29b
+	.long	24b, 29b
+	.long	25b, 29b
+	.long	30b, 29b
+	.long	31b, 29b
+	.long	32b, 29b
+	.long	33b, 29b
+	.long	34b, 29b
+	.long	35b, 29b
diff --git a/arch/avr32/lib/csum_partial.S b/arch/avr32/lib/csum_partial.S
new file mode 100644
index 000000000000..6a262b528eb7
--- /dev/null
+++ b/arch/avr32/lib/csum_partial.S
@@ -0,0 +1,47 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * unsigned int csum_partial(const unsigned char *buff,
+	 *			     int len, unsigned int sum)
+	 */
+	.text
+	.global	csum_partial
+	.type	csum_partial,"function"
+	.align	1
+csum_partial:
+	/* checksum complete words, aligned or not */
+3:	sub	r11, 4
+	brlt	5f
+4:	ld.w	r9, r12++
+	add	r10, r9
+	acr	r10
+	sub	r11, 4
+	brge	4b
+
+	/* return if we had a whole number of words */
+5:	sub	r11, -4
+	reteq	r10
+
+	/* checksum any remaining bytes at the end */
+	mov	r9, 0
+	mov	r8, 0
+	cp	r11, 2
+	brlt	6f
+	ld.uh	r9, r12++
+	sub	r11, 2
+	breq	7f
+	lsl	r9, 16
+6:	ld.ub	r8, r12++
+	lsl	r8, 8
+7:	or	r9, r8
+	add	r10, r9
+	acr	r10
+
+	retal	r10
+	.size	csum_partial, . - csum_partial
diff --git a/arch/avr32/lib/csum_partial_copy_generic.S b/arch/avr32/lib/csum_partial_copy_generic.S
new file mode 100644
index 000000000000..a3a0f9b8929c
--- /dev/null
+++ b/arch/avr32/lib/csum_partial_copy_generic.S
@@ -0,0 +1,99 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/errno.h>
+#include <asm/asm.h>
+
+	/*
+	 * unsigned int csum_partial_copy_generic(const char *src, char *dst, int len
+	 *					  int sum, int *src_err_ptr,
+	 *					  int *dst_err_ptr)
+	 *
+	 * Copy src to dst while checksumming, otherwise like csum_partial.
+	 */
+
+	.macro ld_src size, reg, ptr
+9999:	ld.\size \reg, \ptr
+	.section __ex_table, "a"
+	.long	9999b, fixup_ld_src
+	.previous
+	.endm
+
+	.macro st_dst size, ptr, reg
+9999:	st.\size \ptr, \reg
+	.section __ex_table, "a"
+	.long	9999b, fixup_st_dst
+	.previous
+	.endm
+
+	.text
+	.global	csum_partial_copy_generic
+	.type	csum_partial_copy_generic,"function"
+	.align	1
+csum_partial_copy_generic:
+	pushm	r4-r7,lr
+
+	/* The inner loop */
+1:	sub	r10, 4
+	brlt	5f
+2:	ld_src	w, r5, r12++
+	st_dst	w, r11++, r5
+	add	r9, r5
+	acr	r9
+	sub	r10, 4
+	brge	2b
+
+	/* return if we had a whole number of words */
+5:	sub	r10, -4
+	brne	7f
+
+6:	mov	r12, r9
+	popm	r4-r7,pc
+
+	/* handle additional bytes at the tail */
+7:	mov	r5, 0
+	mov	r4, 32
+8:	ld_src	ub, r6, r12++
+	st_dst	b, r11++, r6
+	lsl	r5, 8
+	sub	r4, 8
+	bfins	r5, r6, 0, 8
+	sub	r10, 1
+	brne	8b
+
+	lsl	r5, r5, r4
+	add	r9, r5
+	acr	r9
+	rjmp	6b
+
+	/* Exception handler */
+	.section .fixup,"ax"
+	.align	1
+fixup_ld_src:
+	mov	r9, -EFAULT
+	cp.w	r8, 0
+	breq	1f
+	st.w	r8[0], r9
+
+1:	/*
+	 * TODO: zero the complete destination - computing the rest
+	 * is too much work
+	 */
+
+	mov	r9, 0
+	rjmp	6b
+
+fixup_st_dst:
+	mov	r9, -EFAULT
+	lddsp	r8, sp[20]
+	cp.w	r8, 0
+	breq	1f
+	st.w	r8[0], r9
+1:	mov	r9, 0
+	rjmp	6b
+
+	.previous
diff --git a/arch/avr32/lib/delay.c b/arch/avr32/lib/delay.c
new file mode 100644
index 000000000000..462c8307b680
--- /dev/null
+++ b/arch/avr32/lib/delay.c
@@ -0,0 +1,55 @@
+/*
+ *      Precise Delay Loops for avr32
+ *
+ *      Copyright (C) 1993 Linus Torvalds
+ *      Copyright (C) 1997 Martin Mares <mj@atrey.karlin.mff.cuni.cz>
+ *	Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/types.h>
+
+#include <asm/delay.h>
+#include <asm/processor.h>
+#include <asm/sysreg.h>
+
+int read_current_timer(unsigned long *timer_value)
+{
+	*timer_value = sysreg_read(COUNT);
+	return 0;
+}
+
+void __delay(unsigned long loops)
+{
+	unsigned bclock, now;
+
+	bclock = sysreg_read(COUNT);
+	do {
+		now = sysreg_read(COUNT);
+	} while ((now - bclock) < loops);
+}
+
+inline void __const_udelay(unsigned long xloops)
+{
+	unsigned long long loops;
+
+	asm("mulu.d %0, %1, %2"
+	    : "=r"(loops)
+	    : "r"(current_cpu_data.loops_per_jiffy * HZ), "r"(xloops));
+	__delay(loops >> 32);
+}
+
+void __udelay(unsigned long usecs)
+{
+	__const_udelay(usecs * 0x000010c7); /* 2**32 / 1000000 (rounded up) */
+}
+
+void __ndelay(unsigned long nsecs)
+{
+	__const_udelay(nsecs * 0x00005); /* 2**32 / 1000000000 (rounded up) */
+}
diff --git a/arch/avr32/lib/findbit.S b/arch/avr32/lib/findbit.S
new file mode 100644
index 000000000000..2b4856f4bf7c
--- /dev/null
+++ b/arch/avr32/lib/findbit.S
@@ -0,0 +1,154 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+
+	.text
+	/*
+	 * unsigned long find_first_zero_bit(const unsigned long *addr,
+	 *				     unsigned long size)
+	 */
+ENTRY(find_first_zero_bit)
+	cp.w	r11, 0
+	reteq	r11
+	mov	r9, r11
+1:	ld.w	r8, r12[0]
+	com	r8
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+	/*
+	 * unsigned long find_next_zero_bit(const unsigned long *addr,
+	 *				    unsigned long size,
+	 *				    unsigned long offset)
+	 */
+ENTRY(find_next_zero_bit)
+	lsr	r8, r10, 5
+	sub	r9, r11, r10
+	retle	r11
+
+	lsl	r8, 2
+	add	r12, r8
+	andl	r10, 31, COH
+	breq	1f
+
+	/* offset is not word-aligned. Handle the first (32 - r10) bits */
+	ld.w	r8, r12[0]
+	com	r8
+	sub	r12, -4
+	lsr	r8, r8, r10
+	brne	.L_found
+
+	/* r9 = r9 - (32 - r10) = r9 + r10 - 32 */
+	add	r9, r10
+	sub	r9, 32
+	retle	r11
+
+	/* Main loop. offset must be word-aligned */
+1:	ld.w	r8, r12[0]
+	com	r8
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+	/* Common return path for when a bit is actually found. */
+.L_found:
+	brev	r8
+	clz	r10, r8
+	rsub	r9, r11
+	add	r10, r9
+
+	/* XXX: If we don't have to return exactly "size" when the bit
+	   is not found, we may drop this "min" thing */
+	min	r12, r11, r10
+	retal	r12
+
+	/*
+	 * unsigned long find_first_bit(const unsigned long *addr,
+	 *				unsigned long size)
+	 */
+ENTRY(find_first_bit)
+	cp.w	r11, 0
+	reteq	r11
+	mov	r9, r11
+1:	ld.w	r8, r12[0]
+	cp.w	r8, 0
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+	/*
+	 * unsigned long find_next_bit(const unsigned long *addr,
+	 *			       unsigned long size,
+	 *			       unsigned long offset)
+	 */
+ENTRY(find_next_bit)
+	lsr	r8, r10, 5
+	sub	r9, r11, r10
+	retle	r11
+
+	lsl	r8, 2
+	add	r12, r8
+	andl	r10, 31, COH
+	breq	1f
+
+	/* offset is not word-aligned. Handle the first (32 - r10) bits */
+	ld.w	r8, r12[0]
+	sub	r12, -4
+	lsr	r8, r8, r10
+	brne	.L_found
+
+	/* r9 = r9 - (32 - r10) = r9 + r10 - 32 */
+	add	r9, r10
+	sub	r9, 32
+	retle	r11
+
+	/* Main loop. offset must be word-aligned */
+1:	ld.w	r8, r12[0]
+	cp.w	r8, 0
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+ENTRY(generic_find_next_zero_le_bit)
+	lsr	r8, r10, 5
+	sub	r9, r11, r10
+	retle	r11
+
+	lsl	r8, 2
+	add	r12, r8
+	andl	r10, 31, COH
+	breq	1f
+
+	/* offset is not word-aligned. Handle the first (32 - r10) bits */
+	ldswp.w	r8, r12[0]
+	sub	r12, -4
+	lsr	r8, r8, r10
+	brne	.L_found
+
+	/* r9 = r9 - (32 - r10) = r9 + r10 - 32 */
+	add	r9, r10
+	sub	r9, 32
+	retle	r11
+
+	/* Main loop. offset must be word-aligned */
+1:	ldswp.w	r8, r12[0]
+	cp.w	r8, 0
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
diff --git a/arch/avr32/lib/io-readsl.S b/arch/avr32/lib/io-readsl.S
new file mode 100644
index 000000000000..b103511ed6c4
--- /dev/null
+++ b/arch/avr32/lib/io-readsl.S
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	.global	__raw_readsl
+	.type	__raw_readsl,@function
+__raw_readsl:
+	cp.w	r10, 0
+	reteq	r12
+
+	/*
+	 * If r11 isn't properly aligned, we might get an exception on
+	 * some implementations. But there's not much we can do about it.
+	 */
+1:	ld.w	r8, r12[0]
+	sub	r10, 1
+	st.w	r11++, r8
+	brne	1b
+
+	retal	r12
diff --git a/arch/avr32/lib/io-readsw.S b/arch/avr32/lib/io-readsw.S
new file mode 100644
index 000000000000..456be9909027
--- /dev/null
+++ b/arch/avr32/lib/io-readsw.S
@@ -0,0 +1,43 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+.Lnot_word_aligned:
+	/*
+	 * Bad alignment will cause a hardware exception, which is as
+	 * good as anything. No need for us to check for proper alignment.
+	 */
+	ld.uh	r8, r12[0]
+	sub	r10, 1
+	st.h	r11++, r8
+
+	/* fall through */
+
+	.global	__raw_readsw
+	.type	__raw_readsw,@function
+__raw_readsw:
+	cp.w	r10, 0
+	reteq	r12
+	mov	r9, 3
+	tst	r11, r9
+	brne	.Lnot_word_aligned
+
+	sub	r10, 2
+	brlt	2f
+
+1:	ldins.h	r8:t, r12[0]
+	ldins.h	r8:b, r12[0]
+	st.w	r11++, r8
+	sub	r10, 2
+	brge	1b
+
+2:	sub	r10, -2
+	reteq	r12
+
+	ld.uh	r8, r12[0]
+	st.h	r11++, r8
+	retal	r12
diff --git a/arch/avr32/lib/io-writesl.S b/arch/avr32/lib/io-writesl.S
new file mode 100644
index 000000000000..22138b3a16e5
--- /dev/null
+++ b/arch/avr32/lib/io-writesl.S
@@ -0,0 +1,20 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	.global	__raw_writesl
+	.type	__raw_writesl,@function
+__raw_writesl:
+	cp.w	r10, 0
+	reteq	r12
+
+1:	ld.w	r8, r11++
+	sub	r10, 1
+	st.w	r12[0], r8
+	brne	1b
+
+	retal	r12
diff --git a/arch/avr32/lib/io-writesw.S b/arch/avr32/lib/io-writesw.S
new file mode 100644
index 000000000000..8c4a53f1c52a
--- /dev/null
+++ b/arch/avr32/lib/io-writesw.S
@@ -0,0 +1,38 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+.Lnot_word_aligned:
+	ld.uh	r8, r11++
+	sub	r10, 1
+	st.h	r12[0], r8
+
+	.global	__raw_writesw
+	.type	__raw_writesw,@function
+__raw_writesw:
+	cp.w	r10, 0
+	mov	r9, 3
+	reteq	r12
+	tst	r11, r9
+	brne	.Lnot_word_aligned
+
+	sub	r10, 2
+	brlt	2f
+
+1:	ld.w	r8, r11++
+	bfextu	r9, r8, 16, 16
+	st.h	r12[0], r9
+	st.h	r12[0], r8
+	sub	r10, 2
+	brge	1b
+
+2:	sub	r10, -2
+	reteq	r12
+
+	ld.uh	r8, r11++
+	st.h	r12[0], r8
+	retal	r12
diff --git a/arch/avr32/lib/libgcc.h b/arch/avr32/lib/libgcc.h
new file mode 100644
index 000000000000..5a091b5e3618
--- /dev/null
+++ b/arch/avr32/lib/libgcc.h
@@ -0,0 +1,33 @@
+/* Definitions for various functions 'borrowed' from gcc-3.4.3 */
+
+#define BITS_PER_UNIT	8
+
+typedef		 int QItype	__attribute__ ((mode (QI)));
+typedef unsigned int UQItype	__attribute__ ((mode (QI)));
+typedef		 int HItype	__attribute__ ((mode (HI)));
+typedef unsigned int UHItype	__attribute__ ((mode (HI)));
+typedef 	 int SItype	__attribute__ ((mode (SI)));
+typedef unsigned int USItype	__attribute__ ((mode (SI)));
+typedef		 int DItype	__attribute__ ((mode (DI)));
+typedef unsigned int UDItype	__attribute__ ((mode (DI)));
+typedef 	float SFtype	__attribute__ ((mode (SF)));
+typedef		float DFtype	__attribute__ ((mode (DF)));
+typedef int word_type __attribute__ ((mode (__word__)));
+
+#define W_TYPE_SIZE (4 * BITS_PER_UNIT)
+#define Wtype	SItype
+#define UWtype	USItype
+#define HWtype	SItype
+#define UHWtype	USItype
+#define DWtype	DItype
+#define UDWtype	UDItype
+#define __NW(a,b)	__ ## a ## si ## b
+#define __NDW(a,b)	__ ## a ## di ## b
+
+struct DWstruct {Wtype high, low;};
+
+typedef union
+{
+  struct DWstruct s;
+  DWtype ll;
+} DWunion;
diff --git a/arch/avr32/lib/longlong.h b/arch/avr32/lib/longlong.h
new file mode 100644
index 000000000000..cd5e369ac437
--- /dev/null
+++ b/arch/avr32/lib/longlong.h
@@ -0,0 +1,98 @@
+/* longlong.h -- definitions for mixed size 32/64 bit arithmetic.
+   Copyright (C) 1991, 1992, 1994, 1995, 1996, 1997, 1998, 1999, 2000
+   Free Software Foundation, Inc.
+
+   This definition file is free software; you can redistribute it
+   and/or modify it under the terms of the GNU General Public
+   License as published by the Free Software Foundation; either
+   version 2, or (at your option) any later version.
+
+   This definition file is distributed in the hope that it will be
+   useful, but WITHOUT ANY WARRANTY; without even the implied
+   warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+   See the GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, write to the Free Software
+   Foundation, Inc., 59 Temple Place - Suite 330,
+   Boston, MA 02111-1307, USA.  */
+
+/* Borrowed from gcc-3.4.3 */
+
+#define __BITS4 (W_TYPE_SIZE / 4)
+#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
+#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
+#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))
+
+#define count_leading_zeros(count, x) ((count) = __builtin_clz(x))
+
+#define __udiv_qrnnd_c(q, r, n1, n0, d) \
+  do {									\
+    UWtype __d1, __d0, __q1, __q0;					\
+    UWtype __r1, __r0, __m;						\
+    __d1 = __ll_highpart (d);						\
+    __d0 = __ll_lowpart (d);						\
+									\
+    __r1 = (n1) % __d1;							\
+    __q1 = (n1) / __d1;							\
+    __m = (UWtype) __q1 * __d0;						\
+    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
+    if (__r1 < __m)							\
+      {									\
+	__q1--, __r1 += (d);						\
+	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
+	  if (__r1 < __m)						\
+	    __q1--, __r1 += (d);					\
+      }									\
+    __r1 -= __m;							\
+									\
+    __r0 = __r1 % __d1;							\
+    __q0 = __r1 / __d1;							\
+    __m = (UWtype) __q0 * __d0;						\
+    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
+    if (__r0 < __m)							\
+      {									\
+	__q0--, __r0 += (d);						\
+	if (__r0 >= (d))						\
+	  if (__r0 < __m)						\
+	    __q0--, __r0 += (d);					\
+      }									\
+    __r0 -= __m;							\
+									\
+    (q) = (UWtype) __q1 * __ll_B | __q0;				\
+    (r) = __r0;								\
+  } while (0)
+
+#define udiv_qrnnd __udiv_qrnnd_c
+
+#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
+  do {									\
+    UWtype __x;								\
+    __x = (al) - (bl);							\
+    (sh) = (ah) - (bh) - (__x > (al));					\
+    (sl) = __x;								\
+  } while (0)
+
+#define umul_ppmm(w1, w0, u, v)						\
+  do {									\
+    UWtype __x0, __x1, __x2, __x3;					\
+    UHWtype __ul, __vl, __uh, __vh;					\
+									\
+    __ul = __ll_lowpart (u);						\
+    __uh = __ll_highpart (u);						\
+    __vl = __ll_lowpart (v);						\
+    __vh = __ll_highpart (v);						\
+									\
+    __x0 = (UWtype) __ul * __vl;					\
+    __x1 = (UWtype) __ul * __vh;					\
+    __x2 = (UWtype) __uh * __vl;					\
+    __x3 = (UWtype) __uh * __vh;					\
+									\
+    __x1 += __ll_highpart (__x0);/* this can't give carry */		\
+    __x1 += __x2;		/* but this indeed can */		\
+    if (__x1 < __x2)		/* did we get it? */			\
+      __x3 += __ll_B;		/* yes, add it in the proper pos.  */	\
+									\
+    (w1) = __x3 + __ll_highpart (__x1);					\
+    (w0) = __ll_lowpart (__x1) * __ll_B + __ll_lowpart (__x0);		\
+  } while (0)
diff --git a/arch/avr32/lib/memcpy.S b/arch/avr32/lib/memcpy.S
new file mode 100644
index 000000000000..0abb26142b64
--- /dev/null
+++ b/arch/avr32/lib/memcpy.S
@@ -0,0 +1,62 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * void *memcpy(void *to, const void *from, unsigned long n)
+	 *
+	 * This implementation does word-aligned loads in the main loop,
+	 * possibly sacrificing alignment of stores.
+	 *
+	 * Hopefully, in most cases, both "to" and "from" will be
+	 * word-aligned to begin with.
+	 */
+	.text
+	.global	memcpy
+	.type	memcpy, @function
+memcpy:
+	mov	r9, r11
+	andl	r9, 3, COH
+	brne	1f
+
+	/* At this point, "from" is word-aligned */
+2:	sub	r10, 4
+	mov	r9, r12
+	brlt	4f
+
+3:	ld.w	r8, r11++
+	sub	r10, 4
+	st.w	r12++, r8
+	brge	3b
+
+4:	neg	r10
+	reteq	r9
+
+	/* Handle unaligned count */
+	lsl	r10, 2
+	add	pc, pc, r10
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	retal	r9
+
+	/* Handle unaligned "from" pointer */
+1:	sub	r10, 4
+	brlt	4b
+	add	r10, r9
+	lsl	r9, 2
+	add	pc, pc, r9
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	rjmp	2b
diff --git a/arch/avr32/lib/memset.S b/arch/avr32/lib/memset.S
new file mode 100644
index 000000000000..40da32c0480c
--- /dev/null
+++ b/arch/avr32/lib/memset.S
@@ -0,0 +1,72 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/arm/lib/memset.S
+ *   Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * ASM optimised string functions
+ */
+#include <asm/asm.h>
+
+	/*
+	 * r12:	void *b
+	 * r11:	int c
+	 * r10:	size_t len
+	 *
+	 * Returns b in r12
+	 */
+	.text
+	.global	memset
+	.type	memset, @function
+	.align	5
+memset:
+	mov	r9, r12
+	mov	r8, r12
+	or	r11, r11, r11 << 8
+	andl	r9, 3, COH
+	brne	1f
+
+2:	or	r11, r11, r11 << 16
+	sub	r10, 4
+	brlt	5f
+
+	/* Let's do some real work */
+4:	st.w	r8++, r11
+	sub	r10, 4
+	brge	4b
+
+	/*
+	 * When we get here, we've got less than 4 bytes to set. r10
+	 * might be negative.
+	 */
+5:	sub	r10, -4
+	reteq	r12
+
+	/* Fastpath ends here, exactly 32 bytes from memset */
+
+	/* Handle unaligned count or pointer */
+	bld	r10, 1
+	brcc	6f
+	st.b	r8++, r11
+	st.b	r8++, r11
+	bld	r10, 0
+	retcc	r12
+6:	st.b	r8++, r11
+	retal	r12
+
+	/* Handle unaligned pointer */
+1:	sub	r10, 4
+	brlt	5b
+	add	r10, r9
+	lsl	r9, 1
+	add	pc, r9
+	st.b	r8++, r11
+	st.b	r8++, r11
+	st.b	r8++, r11
+	rjmp	2b
+
+	.size	memset, . - memset
diff --git a/arch/avr32/lib/strncpy_from_user.S b/arch/avr32/lib/strncpy_from_user.S
new file mode 100644
index 000000000000..72bd50599ec6
--- /dev/null
+++ b/arch/avr32/lib/strncpy_from_user.S
@@ -0,0 +1,60 @@
+/*
+ * Copy to/from userspace with optional address space checking.
+ *
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/errno.h>
+
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/asm.h>
+
+	/*
+	 * long strncpy_from_user(char *dst, const char *src, long count)
+	 *
+	 * On success, returns the length of the string, not including
+	 * the terminating NUL.
+	 *
+	 * If the string is longer than count, returns count
+	 *
+	 * If userspace access fails, returns -EFAULT
+	 */
+	.text
+	.align	1
+	.global	strncpy_from_user
+	.type	strncpy_from_user, "function"
+strncpy_from_user:
+	mov	r9, -EFAULT
+	branch_if_kernel r8, __strncpy_from_user
+	ret_if_privileged r8, r11, r10, r9
+
+	.global	__strncpy_from_user
+	.type	__strncpy_from_user, "function"
+__strncpy_from_user:
+	cp.w	r10, 0
+	reteq	0
+
+	mov	r9, r10
+
+1:	ld.ub	r8, r11++
+	st.b	r12++, r8
+	cp.w	r8, 0
+	breq	2f
+	sub	r9, 1
+	brne	1b
+
+2:	sub	r10, r9
+	retal	r10
+
+	.section .fixup, "ax"
+	.align	1
+3:	mov	r12, -EFAULT
+	retal	r12
+
+	.section __ex_table, "a"
+	.align	2
+	.long	1b, 3b
diff --git a/arch/avr32/lib/strnlen_user.S b/arch/avr32/lib/strnlen_user.S
new file mode 100644
index 000000000000..65ce11afa66a
--- /dev/null
+++ b/arch/avr32/lib/strnlen_user.S
@@ -0,0 +1,67 @@
+/*
+ * Copy to/from userspace with optional address space checking.
+ *
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/processor.h>
+#include <asm/asm.h>
+
+	.text
+	.align	1
+	.global	strnlen_user
+	.type	strnlen_user, "function"
+strnlen_user:
+	branch_if_kernel r8, __strnlen_user
+	sub	r8, r11, 1
+	add	r8, r12
+	retcs	0
+	brmi	adjust_length	/* do a closer inspection */
+
+	.global	__strnlen_user
+	.type	__strnlen_user, "function"
+__strnlen_user:
+	mov	r10, r12
+
+10:	ld.ub	r8, r12++
+	cp.w	r8, 0
+	breq	2f
+	sub	r11, 1
+	brne	10b
+
+	sub	r12, -1
+2:	sub	r12, r10
+	retal	r12
+
+
+	.type	adjust_length, "function"
+adjust_length:
+	cp.w	r12, 0		/* addr must always be < TASK_SIZE */
+	retmi	0
+
+	pushm	lr
+	lddpc	lr, _task_size
+	sub	r11, lr, r12
+	mov	r9, r11
+	rcall	__strnlen_user
+	cp.w	r12, r9
+	brgt	1f
+	popm	pc
+1:	popm	pc, r12=0
+
+	.align	2
+_task_size:
+	.long	TASK_SIZE
+
+	.section .fixup, "ax"
+	.align	1
+19:	retal	0
+
+	.section __ex_table, "a"
+	.align	2
+	.long	10b, 19b
diff --git a/arch/avr32/mach-at32ap/Makefile b/arch/avr32/mach-at32ap/Makefile
new file mode 100644
index 000000000000..4b10853eb614
--- /dev/null
+++ b/arch/avr32/mach-at32ap/Makefile
@@ -0,0 +1,2 @@
+obj-y				+= at32ap.o clock.o pio.o intc.o extint.o
+obj-$(CONFIG_CPU_AT32AP7000)	+= at32ap7000.o
diff --git a/arch/avr32/mach-at32ap/at32ap.c b/arch/avr32/mach-at32ap/at32ap.c
new file mode 100644
index 000000000000..f7cedf5aabea
--- /dev/null
+++ b/arch/avr32/mach-at32ap/at32ap.c
@@ -0,0 +1,90 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/init.h>
+#include <asm/arch/sm.h>
+
+struct at32_sm system_manager;
+
+static int __init at32_sm_init(void)
+{
+	struct resource *regs;
+	struct at32_sm *sm = &system_manager;
+	int ret = -ENXIO;
+
+	regs = platform_get_resource(&at32_sm_device, IORESOURCE_MEM, 0);
+	if (!regs)
+		goto fail;
+
+	spin_lock_init(&sm->lock);
+	sm->pdev = &at32_sm_device;
+
+	ret = -ENOMEM;
+	sm->regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!sm->regs)
+		goto fail;
+
+	return 0;
+
+fail:
+	printk(KERN_ERR "Failed to initialize System Manager: %d\n", ret);
+	return ret;
+}
+
+void __init setup_platform(void)
+{
+	at32_sm_init();
+	at32_clock_init();
+	at32_portmux_init();
+
+	/* FIXME: This doesn't belong here */
+	at32_setup_serial_console(1);
+}
+
+static int __init pdc_probe(struct platform_device *pdev)
+{
+	struct clk *pclk, *hclk;
+
+	pclk = clk_get(&pdev->dev, "pclk");
+	if (IS_ERR(pclk)) {
+		dev_err(&pdev->dev, "no pclk defined\n");
+		return PTR_ERR(pclk);
+	}
+	hclk = clk_get(&pdev->dev, "hclk");
+	if (IS_ERR(hclk)) {
+		dev_err(&pdev->dev, "no hclk defined\n");
+		clk_put(pclk);
+		return PTR_ERR(hclk);
+	}
+
+	clk_enable(pclk);
+	clk_enable(hclk);
+
+	dev_info(&pdev->dev, "Atmel Peripheral DMA Controller enabled\n");
+	return 0;
+}
+
+static struct platform_driver pdc_driver = {
+	.probe		= pdc_probe,
+	.driver		= {
+		.name	= "pdc",
+	},
+};
+
+static int __init pdc_init(void)
+{
+	return platform_driver_register(&pdc_driver);
+}
+arch_initcall(pdc_init);
diff --git a/arch/avr32/mach-at32ap/at32ap7000.c b/arch/avr32/mach-at32ap/at32ap7000.c
new file mode 100644
index 000000000000..e8c6893a1c23
--- /dev/null
+++ b/arch/avr32/mach-at32ap/at32ap7000.c
@@ -0,0 +1,866 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/board.h>
+#include <asm/arch/portmux.h>
+#include <asm/arch/sm.h>
+
+#include "clock.h"
+#include "pio.h"
+#include "sm.h"
+
+#define PBMEM(base)					\
+	{						\
+		.start		= base,			\
+		.end		= base + 0x3ff,		\
+		.flags		= IORESOURCE_MEM,	\
+	}
+#define IRQ(num)					\
+	{						\
+		.start		= num,			\
+		.end		= num,			\
+		.flags		= IORESOURCE_IRQ,	\
+	}
+#define NAMED_IRQ(num, _name)				\
+	{						\
+		.start		= num,			\
+		.end		= num,			\
+		.name		= _name,		\
+		.flags		= IORESOURCE_IRQ,	\
+	}
+
+#define DEFINE_DEV(_name, _id)					\
+static struct platform_device _name##_id##_device = {		\
+	.name		= #_name,				\
+	.id		= _id,					\
+	.resource	= _name##_id##_resource,		\
+	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
+}
+#define DEFINE_DEV_DATA(_name, _id)				\
+static struct platform_device _name##_id##_device = {		\
+	.name		= #_name,				\
+	.id		= _id,					\
+	.dev		= {					\
+		.platform_data	= &_name##_id##_data,		\
+	},							\
+	.resource	= _name##_id##_resource,		\
+	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
+}
+
+#define DEV_CLK(_name, devname, bus, _index)			\
+static struct clk devname##_##_name = {				\
+	.name		= #_name,				\
+	.dev		= &devname##_device.dev,		\
+	.parent		= &bus##_clk,				\
+	.mode		= bus##_clk_mode,			\
+	.get_rate	= bus##_clk_get_rate,			\
+	.index		= _index,				\
+}
+
+enum {
+	PIOA,
+	PIOB,
+	PIOC,
+	PIOD,
+};
+
+enum {
+	FUNC_A,
+	FUNC_B,
+};
+
+unsigned long at32ap7000_osc_rates[3] = {
+	[0] = 32768,
+	/* FIXME: these are ATSTK1002-specific */
+	[1] = 20000000,
+	[2] = 12000000,
+};
+
+static unsigned long osc_get_rate(struct clk *clk)
+{
+	return at32ap7000_osc_rates[clk->index];
+}
+
+static unsigned long pll_get_rate(struct clk *clk, unsigned long control)
+{
+	unsigned long div, mul, rate;
+
+	if (!(control & SM_BIT(PLLEN)))
+		return 0;
+
+	div = SM_BFEXT(PLLDIV, control) + 1;
+	mul = SM_BFEXT(PLLMUL, control) + 1;
+
+	rate = clk->parent->get_rate(clk->parent);
+	rate = (rate + div / 2) / div;
+	rate *= mul;
+
+	return rate;
+}
+
+static unsigned long pll0_get_rate(struct clk *clk)
+{
+	u32 control;
+
+	control = sm_readl(&system_manager, PM_PLL0);
+
+	return pll_get_rate(clk, control);
+}
+
+static unsigned long pll1_get_rate(struct clk *clk)
+{
+	u32 control;
+
+	control = sm_readl(&system_manager, PM_PLL1);
+
+	return pll_get_rate(clk, control);
+}
+
+/*
+ * The AT32AP7000 has five primary clock sources: One 32kHz
+ * oscillator, two crystal oscillators and two PLLs.
+ */
+static struct clk osc32k = {
+	.name		= "osc32k",
+	.get_rate	= osc_get_rate,
+	.users		= 1,
+	.index		= 0,
+};
+static struct clk osc0 = {
+	.name		= "osc0",
+	.get_rate	= osc_get_rate,
+	.users		= 1,
+	.index		= 1,
+};
+static struct clk osc1 = {
+	.name		= "osc1",
+	.get_rate	= osc_get_rate,
+	.index		= 2,
+};
+static struct clk pll0 = {
+	.name		= "pll0",
+	.get_rate	= pll0_get_rate,
+	.parent		= &osc0,
+};
+static struct clk pll1 = {
+	.name		= "pll1",
+	.get_rate	= pll1_get_rate,
+	.parent		= &osc0,
+};
+
+/*
+ * The main clock can be either osc0 or pll0.  The boot loader may
+ * have chosen one for us, so we don't really know which one until we
+ * have a look at the SM.
+ */
+static struct clk *main_clock;
+
+/*
+ * Synchronous clocks are generated from the main clock. The clocks
+ * must satisfy the constraint
+ *   fCPU >= fHSB >= fPB
+ * i.e. each clock must not be faster than its parent.
+ */
+static unsigned long bus_clk_get_rate(struct clk *clk, unsigned int shift)
+{
+	return main_clock->get_rate(main_clock) >> shift;
+};
+
+static void cpu_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_CPU_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_CPU_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long cpu_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(CPUDIV))
+		shift = SM_BFEXT(CPUSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void hsb_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_HSB_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_HSB_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long hsb_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(HSBDIV))
+		shift = SM_BFEXT(HSBSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void pba_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_PBA_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_PBA_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long pba_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(PBADIV))
+		shift = SM_BFEXT(PBASEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void pbb_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_PBB_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_PBB_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long pbb_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(PBBDIV))
+		shift = SM_BFEXT(PBBSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static struct clk cpu_clk = {
+	.name		= "cpu",
+	.get_rate	= cpu_clk_get_rate,
+	.users		= 1,
+};
+static struct clk hsb_clk = {
+	.name		= "hsb",
+	.parent		= &cpu_clk,
+	.get_rate	= hsb_clk_get_rate,
+};
+static struct clk pba_clk = {
+	.name		= "pba",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= pba_clk_get_rate,
+	.index		= 1,
+};
+static struct clk pbb_clk = {
+	.name		= "pbb",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.users		= 1,
+	.index		= 2,
+};
+
+/* --------------------------------------------------------------------
+ *  Generic Clock operations
+ * -------------------------------------------------------------------- */
+
+static void genclk_mode(struct clk *clk, int enabled)
+{
+	u32 control;
+
+	BUG_ON(clk->index > 7);
+
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+	if (enabled)
+		control |= SM_BIT(CEN);
+	else
+		control &= ~SM_BIT(CEN);
+	sm_writel(&system_manager, PM_GCCTRL + 4 * clk->index, control);
+}
+
+static unsigned long genclk_get_rate(struct clk *clk)
+{
+	u32 control;
+	unsigned long div = 1;
+
+	BUG_ON(clk->index > 7);
+
+	if (!clk->parent)
+		return 0;
+
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+	if (control & SM_BIT(DIVEN))
+		div = 2 * (SM_BFEXT(DIV, control) + 1);
+
+	return clk->parent->get_rate(clk->parent) / div;
+}
+
+static long genclk_set_rate(struct clk *clk, unsigned long rate, int apply)
+{
+	u32 control;
+	unsigned long parent_rate, actual_rate, div;
+
+	BUG_ON(clk->index > 7);
+
+	if (!clk->parent)
+		return 0;
+
+	parent_rate = clk->parent->get_rate(clk->parent);
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+
+	if (rate > 3 * parent_rate / 4) {
+		actual_rate = parent_rate;
+		control &= ~SM_BIT(DIVEN);
+	} else {
+		div = (parent_rate + rate) / (2 * rate) - 1;
+		control = SM_BFINS(DIV, div, control) | SM_BIT(DIVEN);
+		actual_rate = parent_rate / (2 * (div + 1));
+	}
+
+	printk("clk %s: new rate %lu (actual rate %lu)\n",
+	       clk->name, rate, actual_rate);
+
+	if (apply)
+		sm_writel(&system_manager, PM_GCCTRL + 4 * clk->index,
+			  control);
+
+	return actual_rate;
+}
+
+int genclk_set_parent(struct clk *clk, struct clk *parent)
+{
+	u32 control;
+
+	BUG_ON(clk->index > 7);
+
+	printk("clk %s: new parent %s (was %s)\n",
+	       clk->name, parent->name,
+	       clk->parent ? clk->parent->name : "(null)");
+
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+
+	if (parent == &osc1 || parent == &pll1)
+		control |= SM_BIT(OSCSEL);
+	else if (parent == &osc0 || parent == &pll0)
+		control &= ~SM_BIT(OSCSEL);
+	else
+		return -EINVAL;
+
+	if (parent == &pll0 || parent == &pll1)
+		control |= SM_BIT(PLLSEL);
+	else
+		control &= ~SM_BIT(PLLSEL);
+
+	sm_writel(&system_manager, PM_GCCTRL + 4 * clk->index, control);
+	clk->parent = parent;
+
+	return 0;
+}
+
+/* --------------------------------------------------------------------
+ *  System peripherals
+ * -------------------------------------------------------------------- */
+static struct resource sm_resource[] = {
+	PBMEM(0xfff00000),
+	NAMED_IRQ(19, "eim"),
+	NAMED_IRQ(20, "pm"),
+	NAMED_IRQ(21, "rtc"),
+};
+struct platform_device at32_sm_device = {
+	.name		= "sm",
+	.id		= 0,
+	.resource	= sm_resource,
+	.num_resources	= ARRAY_SIZE(sm_resource),
+};
+DEV_CLK(pclk, at32_sm, pbb, 0);
+
+static struct resource intc0_resource[] = {
+	PBMEM(0xfff00400),
+};
+struct platform_device at32_intc0_device = {
+	.name		= "intc",
+	.id		= 0,
+	.resource	= intc0_resource,
+	.num_resources	= ARRAY_SIZE(intc0_resource),
+};
+DEV_CLK(pclk, at32_intc0, pbb, 1);
+
+static struct clk ebi_clk = {
+	.name		= "ebi",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.users		= 1,
+};
+static struct clk hramc_clk = {
+	.name		= "hramc",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.users		= 1,
+};
+
+static struct platform_device pdc_device = {
+	.name		= "pdc",
+	.id		= 0,
+};
+DEV_CLK(hclk, pdc, hsb, 4);
+DEV_CLK(pclk, pdc, pba, 16);
+
+static struct clk pico_clk = {
+	.name		= "pico",
+	.parent		= &cpu_clk,
+	.mode		= cpu_clk_mode,
+	.get_rate	= cpu_clk_get_rate,
+	.users		= 1,
+};
+
+/* --------------------------------------------------------------------
+ *  PIO
+ * -------------------------------------------------------------------- */
+
+static struct resource pio0_resource[] = {
+	PBMEM(0xffe02800),
+	IRQ(13),
+};
+DEFINE_DEV(pio, 0);
+DEV_CLK(mck, pio0, pba, 10);
+
+static struct resource pio1_resource[] = {
+	PBMEM(0xffe02c00),
+	IRQ(14),
+};
+DEFINE_DEV(pio, 1);
+DEV_CLK(mck, pio1, pba, 11);
+
+static struct resource pio2_resource[] = {
+	PBMEM(0xffe03000),
+	IRQ(15),
+};
+DEFINE_DEV(pio, 2);
+DEV_CLK(mck, pio2, pba, 12);
+
+static struct resource pio3_resource[] = {
+	PBMEM(0xffe03400),
+	IRQ(16),
+};
+DEFINE_DEV(pio, 3);
+DEV_CLK(mck, pio3, pba, 13);
+
+void __init at32_add_system_devices(void)
+{
+	system_manager.eim_first_irq = NR_INTERNAL_IRQS;
+
+	platform_device_register(&at32_sm_device);
+	platform_device_register(&at32_intc0_device);
+	platform_device_register(&pdc_device);
+
+	platform_device_register(&pio0_device);
+	platform_device_register(&pio1_device);
+	platform_device_register(&pio2_device);
+	platform_device_register(&pio3_device);
+}
+
+/* --------------------------------------------------------------------
+ *  USART
+ * -------------------------------------------------------------------- */
+
+static struct resource usart0_resource[] = {
+	PBMEM(0xffe00c00),
+	IRQ(7),
+};
+DEFINE_DEV(usart, 0);
+DEV_CLK(usart, usart0, pba, 4);
+
+static struct resource usart1_resource[] = {
+	PBMEM(0xffe01000),
+	IRQ(7),
+};
+DEFINE_DEV(usart, 1);
+DEV_CLK(usart, usart1, pba, 4);
+
+static struct resource usart2_resource[] = {
+	PBMEM(0xffe01400),
+	IRQ(8),
+};
+DEFINE_DEV(usart, 2);
+DEV_CLK(usart, usart2, pba, 5);
+
+static struct resource usart3_resource[] = {
+	PBMEM(0xffe01800),
+	IRQ(9),
+};
+DEFINE_DEV(usart, 3);
+DEV_CLK(usart, usart3, pba, 6);
+
+static inline void configure_usart0_pins(void)
+{
+	portmux_set_func(PIOA,  8, FUNC_B);	/* RXD	*/
+	portmux_set_func(PIOA,  9, FUNC_B);	/* TXD	*/
+}
+
+static inline void configure_usart1_pins(void)
+{
+	portmux_set_func(PIOA, 17, FUNC_A);	/* RXD	*/
+	portmux_set_func(PIOA, 18, FUNC_A);	/* TXD	*/
+}
+
+static inline void configure_usart2_pins(void)
+{
+	portmux_set_func(PIOB, 26, FUNC_B);	/* RXD	*/
+	portmux_set_func(PIOB, 27, FUNC_B);	/* TXD	*/
+}
+
+static inline void configure_usart3_pins(void)
+{
+	portmux_set_func(PIOB, 18, FUNC_B);	/* RXD	*/
+	portmux_set_func(PIOB, 17, FUNC_B);	/* TXD	*/
+}
+
+static struct platform_device *setup_usart(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &usart0_device;
+		configure_usart0_pins();
+		break;
+	case 1:
+		pdev = &usart1_device;
+		configure_usart1_pins();
+		break;
+	case 2:
+		pdev = &usart2_device;
+		configure_usart2_pins();
+		break;
+	case 3:
+		pdev = &usart3_device;
+		configure_usart3_pins();
+		break;
+	default:
+		pdev = NULL;
+		break;
+	}
+
+	return pdev;
+}
+
+struct platform_device *__init at32_add_device_usart(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	pdev = setup_usart(id);
+	if (pdev)
+		platform_device_register(pdev);
+
+	return pdev;
+}
+
+struct platform_device *at91_default_console_device;
+
+void __init at32_setup_serial_console(unsigned int usart_id)
+{
+	at91_default_console_device = setup_usart(usart_id);
+}
+
+/* --------------------------------------------------------------------
+ *  Ethernet
+ * -------------------------------------------------------------------- */
+
+static struct eth_platform_data macb0_data;
+static struct resource macb0_resource[] = {
+	PBMEM(0xfff01800),
+	IRQ(25),
+};
+DEFINE_DEV_DATA(macb, 0);
+DEV_CLK(hclk, macb0, hsb, 8);
+DEV_CLK(pclk, macb0, pbb, 6);
+
+struct platform_device *__init
+at32_add_device_eth(unsigned int id, struct eth_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &macb0_device;
+
+		portmux_set_func(PIOC,  3, FUNC_A);	/* TXD0	*/
+		portmux_set_func(PIOC,  4, FUNC_A);	/* TXD1	*/
+		portmux_set_func(PIOC,  7, FUNC_A);	/* TXEN	*/
+		portmux_set_func(PIOC,  8, FUNC_A);	/* TXCK */
+		portmux_set_func(PIOC,  9, FUNC_A);	/* RXD0	*/
+		portmux_set_func(PIOC, 10, FUNC_A);	/* RXD1	*/
+		portmux_set_func(PIOC, 13, FUNC_A);	/* RXER	*/
+		portmux_set_func(PIOC, 15, FUNC_A);	/* RXDV	*/
+		portmux_set_func(PIOC, 16, FUNC_A);	/* MDC	*/
+		portmux_set_func(PIOC, 17, FUNC_A);	/* MDIO	*/
+
+		if (!data->is_rmii) {
+			portmux_set_func(PIOC,  0, FUNC_A);	/* COL	*/
+			portmux_set_func(PIOC,  1, FUNC_A);	/* CRS	*/
+			portmux_set_func(PIOC,  2, FUNC_A);	/* TXER	*/
+			portmux_set_func(PIOC,  5, FUNC_A);	/* TXD2	*/
+			portmux_set_func(PIOC,  6, FUNC_A);	/* TXD3 */
+			portmux_set_func(PIOC, 11, FUNC_A);	/* RXD2	*/
+			portmux_set_func(PIOC, 12, FUNC_A);	/* RXD3	*/
+			portmux_set_func(PIOC, 14, FUNC_A);	/* RXCK	*/
+			portmux_set_func(PIOC, 18, FUNC_A);	/* SPD	*/
+		}
+		break;
+
+	default:
+		return NULL;
+	}
+
+	memcpy(pdev->dev.platform_data, data, sizeof(struct eth_platform_data));
+	platform_device_register(pdev);
+
+	return pdev;
+}
+
+/* --------------------------------------------------------------------
+ *  SPI
+ * -------------------------------------------------------------------- */
+static struct resource spi0_resource[] = {
+	PBMEM(0xffe00000),
+	IRQ(3),
+};
+DEFINE_DEV(spi, 0);
+DEV_CLK(mck, spi0, pba, 0);
+
+struct platform_device *__init at32_add_device_spi(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &spi0_device;
+		portmux_set_func(PIOA,  0, FUNC_A);	/* MISO	 */
+		portmux_set_func(PIOA,  1, FUNC_A);	/* MOSI	 */
+		portmux_set_func(PIOA,  2, FUNC_A);	/* SCK	 */
+		portmux_set_func(PIOA,  3, FUNC_A);	/* NPCS0 */
+		portmux_set_func(PIOA,  4, FUNC_A);	/* NPCS1 */
+		portmux_set_func(PIOA,  5, FUNC_A);	/* NPCS2 */
+		break;
+
+	default:
+		return NULL;
+	}
+
+	platform_device_register(pdev);
+	return pdev;
+}
+
+/* --------------------------------------------------------------------
+ *  LCDC
+ * -------------------------------------------------------------------- */
+static struct lcdc_platform_data lcdc0_data;
+static struct resource lcdc0_resource[] = {
+	{
+		.start		= 0xff000000,
+		.end		= 0xff000fff,
+		.flags		= IORESOURCE_MEM,
+	},
+	IRQ(1),
+};
+DEFINE_DEV_DATA(lcdc, 0);
+DEV_CLK(hclk, lcdc0, hsb, 7);
+static struct clk lcdc0_pixclk = {
+	.name		= "pixclk",
+	.dev		= &lcdc0_device.dev,
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 7,
+};
+
+struct platform_device *__init
+at32_add_device_lcdc(unsigned int id, struct lcdc_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &lcdc0_device;
+		portmux_set_func(PIOC, 19, FUNC_A);	/* CC	  */
+		portmux_set_func(PIOC, 20, FUNC_A);	/* HSYNC  */
+		portmux_set_func(PIOC, 21, FUNC_A);	/* PCLK	  */
+		portmux_set_func(PIOC, 22, FUNC_A);	/* VSYNC  */
+		portmux_set_func(PIOC, 23, FUNC_A);	/* DVAL	  */
+		portmux_set_func(PIOC, 24, FUNC_A);	/* MODE	  */
+		portmux_set_func(PIOC, 25, FUNC_A);	/* PWR	  */
+		portmux_set_func(PIOC, 26, FUNC_A);	/* DATA0  */
+		portmux_set_func(PIOC, 27, FUNC_A);	/* DATA1  */
+		portmux_set_func(PIOC, 28, FUNC_A);	/* DATA2  */
+		portmux_set_func(PIOC, 29, FUNC_A);	/* DATA3  */
+		portmux_set_func(PIOC, 30, FUNC_A);	/* DATA4  */
+		portmux_set_func(PIOC, 31, FUNC_A);	/* DATA5  */
+		portmux_set_func(PIOD,  0, FUNC_A);	/* DATA6  */
+		portmux_set_func(PIOD,  1, FUNC_A);	/* DATA7  */
+		portmux_set_func(PIOD,  2, FUNC_A);	/* DATA8  */
+		portmux_set_func(PIOD,  3, FUNC_A);	/* DATA9  */
+		portmux_set_func(PIOD,  4, FUNC_A);	/* DATA10 */
+		portmux_set_func(PIOD,  5, FUNC_A);	/* DATA11 */
+		portmux_set_func(PIOD,  6, FUNC_A);	/* DATA12 */
+		portmux_set_func(PIOD,  7, FUNC_A);	/* DATA13 */
+		portmux_set_func(PIOD,  8, FUNC_A);	/* DATA14 */
+		portmux_set_func(PIOD,  9, FUNC_A);	/* DATA15 */
+		portmux_set_func(PIOD, 10, FUNC_A);	/* DATA16 */
+		portmux_set_func(PIOD, 11, FUNC_A);	/* DATA17 */
+		portmux_set_func(PIOD, 12, FUNC_A);	/* DATA18 */
+		portmux_set_func(PIOD, 13, FUNC_A);	/* DATA19 */
+		portmux_set_func(PIOD, 14, FUNC_A);	/* DATA20 */
+		portmux_set_func(PIOD, 15, FUNC_A);	/* DATA21 */
+		portmux_set_func(PIOD, 16, FUNC_A);	/* DATA22 */
+		portmux_set_func(PIOD, 17, FUNC_A);	/* DATA23 */
+
+		clk_set_parent(&lcdc0_pixclk, &pll0);
+		clk_set_rate(&lcdc0_pixclk, clk_get_rate(&pll0));
+		break;
+
+	default:
+		return NULL;
+	}
+
+	memcpy(pdev->dev.platform_data, data,
+	       sizeof(struct lcdc_platform_data));
+
+	platform_device_register(pdev);
+	return pdev;
+}
+
+struct clk *at32_clock_list[] = {
+	&osc32k,
+	&osc0,
+	&osc1,
+	&pll0,
+	&pll1,
+	&cpu_clk,
+	&hsb_clk,
+	&pba_clk,
+	&pbb_clk,
+	&at32_sm_pclk,
+	&at32_intc0_pclk,
+	&ebi_clk,
+	&hramc_clk,
+	&pdc_hclk,
+	&pdc_pclk,
+	&pico_clk,
+	&pio0_mck,
+	&pio1_mck,
+	&pio2_mck,
+	&pio3_mck,
+	&usart0_usart,
+	&usart1_usart,
+	&usart2_usart,
+	&usart3_usart,
+	&macb0_hclk,
+	&macb0_pclk,
+	&spi0_mck,
+	&lcdc0_hclk,
+	&lcdc0_pixclk,
+};
+unsigned int at32_nr_clocks = ARRAY_SIZE(at32_clock_list);
+
+void __init at32_portmux_init(void)
+{
+	at32_init_pio(&pio0_device);
+	at32_init_pio(&pio1_device);
+	at32_init_pio(&pio2_device);
+	at32_init_pio(&pio3_device);
+}
+
+void __init at32_clock_init(void)
+{
+	struct at32_sm *sm = &system_manager;
+	u32 cpu_mask = 0, hsb_mask = 0, pba_mask = 0, pbb_mask = 0;
+	int i;
+
+	if (sm_readl(sm, PM_MCCTRL) & SM_BIT(PLLSEL))
+		main_clock = &pll0;
+	else
+		main_clock = &osc0;
+
+	if (sm_readl(sm, PM_PLL0) & SM_BIT(PLLOSC))
+		pll0.parent = &osc1;
+	if (sm_readl(sm, PM_PLL1) & SM_BIT(PLLOSC))
+		pll1.parent = &osc1;
+
+	/*
+	 * Turn on all clocks that have at least one user already, and
+	 * turn off everything else. We only do this for module
+	 * clocks, and even though it isn't particularly pretty to
+	 * check the address of the mode function, it should do the
+	 * trick...
+	 */
+	for (i = 0; i < ARRAY_SIZE(at32_clock_list); i++) {
+		struct clk *clk = at32_clock_list[i];
+
+		if (clk->mode == &cpu_clk_mode)
+			cpu_mask |= 1 << clk->index;
+		else if (clk->mode == &hsb_clk_mode)
+			hsb_mask |= 1 << clk->index;
+		else if (clk->mode == &pba_clk_mode)
+			pba_mask |= 1 << clk->index;
+		else if (clk->mode == &pbb_clk_mode)
+			pbb_mask |= 1 << clk->index;
+	}
+
+	sm_writel(sm, PM_CPU_MASK, cpu_mask);
+	sm_writel(sm, PM_HSB_MASK, hsb_mask);
+	sm_writel(sm, PM_PBA_MASK, pba_mask);
+	sm_writel(sm, PM_PBB_MASK, pbb_mask);
+}
diff --git a/arch/avr32/mach-at32ap/clock.c b/arch/avr32/mach-at32ap/clock.c
new file mode 100644
index 000000000000..3d0d1097389f
--- /dev/null
+++ b/arch/avr32/mach-at32ap/clock.c
@@ -0,0 +1,148 @@
+/*
+ * Clock management for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Based on arch/arm/mach-at91rm9200/clock.c
+ *   Copyright (C) 2005 David Brownell
+ *   Copyright (C) 2005 Ivan Kokshaysky
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/device.h>
+#include <linux/string.h>
+
+#include "clock.h"
+
+static spinlock_t clk_lock = SPIN_LOCK_UNLOCKED;
+
+struct clk *clk_get(struct device *dev, const char *id)
+{
+	int i;
+
+	for (i = 0; i < at32_nr_clocks; i++) {
+		struct clk *clk = at32_clock_list[i];
+
+		if (clk->dev == dev && strcmp(id, clk->name) == 0)
+			return clk;
+	}
+
+	return ERR_PTR(-ENOENT);
+}
+EXPORT_SYMBOL(clk_get);
+
+void clk_put(struct clk *clk)
+{
+	/* clocks are static for now, we can't free them */
+}
+EXPORT_SYMBOL(clk_put);
+
+static void __clk_enable(struct clk *clk)
+{
+	if (clk->parent)
+		__clk_enable(clk->parent);
+	if (clk->users++ == 0 && clk->mode)
+		clk->mode(clk, 1);
+}
+
+int clk_enable(struct clk *clk)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	__clk_enable(clk);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL(clk_enable);
+
+static void __clk_disable(struct clk *clk)
+{
+	BUG_ON(clk->users == 0);
+
+	if (--clk->users == 0 && clk->mode)
+		clk->mode(clk, 0);
+	if (clk->parent)
+		__clk_disable(clk->parent);
+}
+
+void clk_disable(struct clk *clk)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	__clk_disable(clk);
+	spin_unlock_irqrestore(&clk_lock, flags);
+}
+EXPORT_SYMBOL(clk_disable);
+
+unsigned long clk_get_rate(struct clk *clk)
+{
+	unsigned long flags;
+	unsigned long rate;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	rate = clk->get_rate(clk);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return rate;
+}
+EXPORT_SYMBOL(clk_get_rate);
+
+long clk_round_rate(struct clk *clk, unsigned long rate)
+{
+	unsigned long flags, actual_rate;
+
+	if (!clk->set_rate)
+		return -ENOSYS;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	actual_rate = clk->set_rate(clk, rate, 0);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return actual_rate;
+}
+EXPORT_SYMBOL(clk_round_rate);
+
+int clk_set_rate(struct clk *clk, unsigned long rate)
+{
+	unsigned long flags;
+	long ret;
+
+	if (!clk->set_rate)
+		return -ENOSYS;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	ret = clk->set_rate(clk, rate, 1);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return (ret < 0) ? ret : 0;
+}
+EXPORT_SYMBOL(clk_set_rate);
+
+int clk_set_parent(struct clk *clk, struct clk *parent)
+{
+	unsigned long flags;
+	int ret;
+
+	if (!clk->set_parent)
+		return -ENOSYS;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	ret = clk->set_parent(clk, parent);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(clk_set_parent);
+
+struct clk *clk_get_parent(struct clk *clk)
+{
+	return clk->parent;
+}
+EXPORT_SYMBOL(clk_get_parent);
diff --git a/arch/avr32/mach-at32ap/clock.h b/arch/avr32/mach-at32ap/clock.h
new file mode 100644
index 000000000000..f953f044ba4d
--- /dev/null
+++ b/arch/avr32/mach-at32ap/clock.h
@@ -0,0 +1,30 @@
+/*
+ * Clock management for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Based on arch/arm/mach-at91rm9200/clock.c
+ *   Copyright (C) 2005 David Brownell
+ *   Copyright (C) 2005 Ivan Kokshaysky
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+
+struct clk {
+	const char	*name;		/* Clock name/function */
+	struct device	*dev;		/* Device the clock is used by */
+	struct clk	*parent;	/* Parent clock, if any */
+	void		(*mode)(struct clk *clk, int enabled);
+	unsigned long	(*get_rate)(struct clk *clk);
+	long		(*set_rate)(struct clk *clk, unsigned long rate,
+				    int apply);
+	int		(*set_parent)(struct clk *clk, struct clk *parent);
+	u16		users;		/* Enabled if non-zero */
+	u16		index;		/* Sibling index */
+};
+
+extern struct clk *at32_clock_list[];
+extern unsigned int at32_nr_clocks;
diff --git a/arch/avr32/mach-at32ap/extint.c b/arch/avr32/mach-at32ap/extint.c
new file mode 100644
index 000000000000..7da9c5f7a0eb
--- /dev/null
+++ b/arch/avr32/mach-at32ap/extint.c
@@ -0,0 +1,171 @@
+/*
+ * External interrupt handling for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+#include <linux/random.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/sm.h>
+
+#include "sm.h"
+
+static void eim_ack_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_ICR, 1 << (irq - sm->eim_first_irq));
+}
+
+static void eim_mask_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_IDR, 1 << (irq - sm->eim_first_irq));
+}
+
+static void eim_mask_ack_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_ICR, 1 << (irq - sm->eim_first_irq));
+	sm_writel(sm, EIM_IDR, 1 << (irq - sm->eim_first_irq));
+}
+
+static void eim_unmask_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_IER, 1 << (irq - sm->eim_first_irq));
+}
+
+static int eim_set_irq_type(unsigned int irq, unsigned int flow_type)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	unsigned int i = irq - sm->eim_first_irq;
+	u32 mode, edge, level;
+	unsigned long flags;
+	int ret = 0;
+
+	flow_type &= IRQ_TYPE_SENSE_MASK;
+
+	spin_lock_irqsave(&sm->lock, flags);
+
+	mode = sm_readl(sm, EIM_MODE);
+	edge = sm_readl(sm, EIM_EDGE);
+	level = sm_readl(sm, EIM_LEVEL);
+
+	switch (flow_type) {
+	case IRQ_TYPE_LEVEL_LOW:
+		mode |= 1 << i;
+		level &= ~(1 << i);
+		break;
+	case IRQ_TYPE_LEVEL_HIGH:
+		mode |= 1 << i;
+		level |= 1 << i;
+		break;
+	case IRQ_TYPE_EDGE_RISING:
+		mode &= ~(1 << i);
+		edge |= 1 << i;
+		break;
+	case IRQ_TYPE_EDGE_FALLING:
+		mode &= ~(1 << i);
+		edge &= ~(1 << i);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	sm_writel(sm, EIM_MODE, mode);
+	sm_writel(sm, EIM_EDGE, edge);
+	sm_writel(sm, EIM_LEVEL, level);
+
+	spin_unlock_irqrestore(&sm->lock, flags);
+
+	return ret;
+}
+
+struct irq_chip eim_chip = {
+	.name		= "eim",
+	.ack		= eim_ack_irq,
+	.mask		= eim_mask_irq,
+	.mask_ack	= eim_mask_ack_irq,
+	.unmask		= eim_unmask_irq,
+	.set_type	= eim_set_irq_type,
+};
+
+static void demux_eim_irq(unsigned int irq, struct irq_desc *desc,
+			  struct pt_regs *regs)
+{
+	struct at32_sm *sm = desc->handler_data;
+	struct irq_desc *ext_desc;
+	unsigned long status, pending;
+	unsigned int i, ext_irq;
+
+	spin_lock(&sm->lock);
+
+	status = sm_readl(sm, EIM_ISR);
+	pending = status & sm_readl(sm, EIM_IMR);
+
+	while (pending) {
+		i = fls(pending) - 1;
+		pending &= ~(1 << i);
+
+		ext_irq = i + sm->eim_first_irq;
+		ext_desc = irq_desc + ext_irq;
+		ext_desc->handle_irq(ext_irq, ext_desc, regs);
+	}
+
+	spin_unlock(&sm->lock);
+}
+
+static int __init eim_init(void)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned int i;
+	unsigned int nr_irqs;
+	unsigned int int_irq;
+	u32 pattern;
+
+	/*
+	 * The EIM is really the same module as SM, so register
+	 * mapping, etc. has been taken care of already.
+	 */
+
+	/*
+	 * Find out how many interrupt lines that are actually
+	 * implemented in hardware.
+	 */
+	sm_writel(sm, EIM_IDR, ~0UL);
+	sm_writel(sm, EIM_MODE, ~0UL);
+	pattern = sm_readl(sm, EIM_MODE);
+	nr_irqs = fls(pattern);
+
+	sm->eim_chip = &eim_chip;
+
+	for (i = 0; i < nr_irqs; i++) {
+		set_irq_chip(sm->eim_first_irq + i, &eim_chip);
+		set_irq_chip_data(sm->eim_first_irq + i, sm);
+	}
+
+	int_irq = platform_get_irq_byname(sm->pdev, "eim");
+
+	set_irq_chained_handler(int_irq, demux_eim_irq);
+	set_irq_data(int_irq, sm);
+
+	printk("EIM: External Interrupt Module at 0x%p, IRQ %u\n",
+	       sm->regs, int_irq);
+	printk("EIM: Handling %u external IRQs, starting with IRQ %u\n",
+	       nr_irqs, sm->eim_first_irq);
+
+	return 0;
+}
+arch_initcall(eim_init);
diff --git a/arch/avr32/mach-at32ap/intc.c b/arch/avr32/mach-at32ap/intc.c
new file mode 100644
index 000000000000..74f8c9f2f03d
--- /dev/null
+++ b/arch/avr32/mach-at32ap/intc.c
@@ -0,0 +1,133 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include "intc.h"
+
+struct intc {
+	void __iomem	*regs;
+	struct irq_chip	chip;
+};
+
+extern struct platform_device at32_intc0_device;
+
+/*
+ * TODO: We may be able to implement mask/unmask by setting IxM flags
+ * in the status register.
+ */
+static void intc_mask_irq(unsigned int irq)
+{
+
+}
+
+static void intc_unmask_irq(unsigned int irq)
+{
+
+}
+
+static struct intc intc0 = {
+	.chip = {
+		.name		= "intc",
+		.mask		= intc_mask_irq,
+		.unmask		= intc_unmask_irq,
+	},
+};
+
+/*
+ * All interrupts go via intc at some point.
+ */
+asmlinkage void do_IRQ(int level, struct pt_regs *regs)
+{
+	struct irq_desc *desc;
+	unsigned int irq;
+	unsigned long status_reg;
+
+	local_irq_disable();
+
+	irq_enter();
+
+	irq = intc_readl(&intc0, INTCAUSE0 - 4 * level);
+	desc = irq_desc + irq;
+	desc->handle_irq(irq, desc, regs);
+
+	/*
+	 * Clear all interrupt level masks so that we may handle
+	 * interrupts during softirq processing.  If this is a nested
+	 * interrupt, interrupts must stay globally disabled until we
+	 * return.
+	 */
+	status_reg = sysreg_read(SR);
+	status_reg &= ~(SYSREG_BIT(I0M) | SYSREG_BIT(I1M)
+			| SYSREG_BIT(I2M) | SYSREG_BIT(I3M));
+	sysreg_write(SR, status_reg);
+
+	irq_exit();
+}
+
+void __init init_IRQ(void)
+{
+	extern void _evba(void);
+	extern void irq_level0(void);
+	struct resource *regs;
+	struct clk *pclk;
+	unsigned int i;
+	u32 offset, readback;
+
+	regs = platform_get_resource(&at32_intc0_device, IORESOURCE_MEM, 0);
+	if (!regs) {
+		printk(KERN_EMERG "intc: no mmio resource defined\n");
+		goto fail;
+	}
+	pclk = clk_get(&at32_intc0_device.dev, "pclk");
+	if (IS_ERR(pclk)) {
+		printk(KERN_EMERG "intc: no clock defined\n");
+		goto fail;
+	}
+
+	clk_enable(pclk);
+
+	intc0.regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!intc0.regs) {
+		printk(KERN_EMERG "intc: failed to map registers (0x%08lx)\n",
+		       (unsigned long)regs->start);
+		goto fail;
+	}
+
+	/*
+	 * Initialize all interrupts to level 0 (lowest priority). The
+	 * priority level may be changed by calling
+	 * irq_set_priority().
+	 *
+	 */
+	offset = (unsigned long)&irq_level0 - (unsigned long)&_evba;
+	for (i = 0; i < NR_INTERNAL_IRQS; i++) {
+		intc_writel(&intc0, INTPR0 + 4 * i, offset);
+		readback = intc_readl(&intc0, INTPR0 + 4 * i);
+		if (readback == offset)
+			set_irq_chip_and_handler(i, &intc0.chip,
+						 handle_simple_irq);
+	}
+
+	/* Unmask all interrupt levels */
+	sysreg_write(SR, (sysreg_read(SR)
+			  & ~(SR_I3M | SR_I2M | SR_I1M | SR_I0M)));
+
+	return;
+
+fail:
+	panic("Interrupt controller initialization failed!\n");
+}
+
diff --git a/arch/avr32/mach-at32ap/intc.h b/arch/avr32/mach-at32ap/intc.h
new file mode 100644
index 000000000000..d289ca2fff13
--- /dev/null
+++ b/arch/avr32/mach-at32ap/intc.h
@@ -0,0 +1,327 @@
+/*
+ * Automatically generated by gen-header.xsl
+ */
+#ifndef __ASM_AVR32_PERIHP_INTC_H__
+#define __ASM_AVR32_PERIHP_INTC_H__
+
+#define INTC_NUM_INT_GRPS            33
+
+#define INTC_INTPR0                  0x0
+# define INTC_INTPR0_INTLEV_OFFSET   30
+# define INTC_INTPR0_INTLEV_SIZE     2
+# define INTC_INTPR0_OFFSET_OFFSET   0
+# define INTC_INTPR0_OFFSET_SIZE     24
+#define INTC_INTREQ0                 0x100
+# define INTC_INTREQ0_IREQUEST0_OFFSET 0
+# define INTC_INTREQ0_IREQUEST0_SIZE 1
+# define INTC_INTREQ0_IREQUEST1_OFFSET 1
+# define INTC_INTREQ0_IREQUEST1_SIZE 1
+#define INTC_INTPR1                  0x4
+# define INTC_INTPR1_INTLEV_OFFSET   30
+# define INTC_INTPR1_INTLEV_SIZE     2
+# define INTC_INTPR1_OFFSET_OFFSET   0
+# define INTC_INTPR1_OFFSET_SIZE     24
+#define INTC_INTREQ1                 0x104
+# define INTC_INTREQ1_IREQUEST32_OFFSET 0
+# define INTC_INTREQ1_IREQUEST32_SIZE 1
+# define INTC_INTREQ1_IREQUEST33_OFFSET 1
+# define INTC_INTREQ1_IREQUEST33_SIZE 1
+# define INTC_INTREQ1_IREQUEST34_OFFSET 2
+# define INTC_INTREQ1_IREQUEST34_SIZE 1
+# define INTC_INTREQ1_IREQUEST35_OFFSET 3
+# define INTC_INTREQ1_IREQUEST35_SIZE 1
+# define INTC_INTREQ1_IREQUEST36_OFFSET 4
+# define INTC_INTREQ1_IREQUEST36_SIZE 1
+# define INTC_INTREQ1_IREQUEST37_OFFSET 5
+# define INTC_INTREQ1_IREQUEST37_SIZE 1
+#define INTC_INTPR2                  0x8
+# define INTC_INTPR2_INTLEV_OFFSET   30
+# define INTC_INTPR2_INTLEV_SIZE     2
+# define INTC_INTPR2_OFFSET_OFFSET   0
+# define INTC_INTPR2_OFFSET_SIZE     24
+#define INTC_INTREQ2                 0x108
+# define INTC_INTREQ2_IREQUEST64_OFFSET 0
+# define INTC_INTREQ2_IREQUEST64_SIZE 1
+# define INTC_INTREQ2_IREQUEST65_OFFSET 1
+# define INTC_INTREQ2_IREQUEST65_SIZE 1
+# define INTC_INTREQ2_IREQUEST66_OFFSET 2
+# define INTC_INTREQ2_IREQUEST66_SIZE 1
+# define INTC_INTREQ2_IREQUEST67_OFFSET 3
+# define INTC_INTREQ2_IREQUEST67_SIZE 1
+# define INTC_INTREQ2_IREQUEST68_OFFSET 4
+# define INTC_INTREQ2_IREQUEST68_SIZE 1
+#define INTC_INTPR3                  0xc
+# define INTC_INTPR3_INTLEV_OFFSET   30
+# define INTC_INTPR3_INTLEV_SIZE     2
+# define INTC_INTPR3_OFFSET_OFFSET   0
+# define INTC_INTPR3_OFFSET_SIZE     24
+#define INTC_INTREQ3                 0x10c
+# define INTC_INTREQ3_IREQUEST96_OFFSET 0
+# define INTC_INTREQ3_IREQUEST96_SIZE 1
+#define INTC_INTPR4                  0x10
+# define INTC_INTPR4_INTLEV_OFFSET   30
+# define INTC_INTPR4_INTLEV_SIZE     2
+# define INTC_INTPR4_OFFSET_OFFSET   0
+# define INTC_INTPR4_OFFSET_SIZE     24
+#define INTC_INTREQ4                 0x110
+# define INTC_INTREQ4_IREQUEST128_OFFSET 0
+# define INTC_INTREQ4_IREQUEST128_SIZE 1
+#define INTC_INTPR5                  0x14
+# define INTC_INTPR5_INTLEV_OFFSET   30
+# define INTC_INTPR5_INTLEV_SIZE     2
+# define INTC_INTPR5_OFFSET_OFFSET   0
+# define INTC_INTPR5_OFFSET_SIZE     24
+#define INTC_INTREQ5                 0x114
+# define INTC_INTREQ5_IREQUEST160_OFFSET 0
+# define INTC_INTREQ5_IREQUEST160_SIZE 1
+#define INTC_INTPR6                  0x18
+# define INTC_INTPR6_INTLEV_OFFSET   30
+# define INTC_INTPR6_INTLEV_SIZE     2
+# define INTC_INTPR6_OFFSET_OFFSET   0
+# define INTC_INTPR6_OFFSET_SIZE     24
+#define INTC_INTREQ6                 0x118
+# define INTC_INTREQ6_IREQUEST192_OFFSET 0
+# define INTC_INTREQ6_IREQUEST192_SIZE 1
+#define INTC_INTPR7                  0x1c
+# define INTC_INTPR7_INTLEV_OFFSET   30
+# define INTC_INTPR7_INTLEV_SIZE     2
+# define INTC_INTPR7_OFFSET_OFFSET   0
+# define INTC_INTPR7_OFFSET_SIZE     24
+#define INTC_INTREQ7                 0x11c
+# define INTC_INTREQ7_IREQUEST224_OFFSET 0
+# define INTC_INTREQ7_IREQUEST224_SIZE 1
+#define INTC_INTPR8                  0x20
+# define INTC_INTPR8_INTLEV_OFFSET   30
+# define INTC_INTPR8_INTLEV_SIZE     2
+# define INTC_INTPR8_OFFSET_OFFSET   0
+# define INTC_INTPR8_OFFSET_SIZE     24
+#define INTC_INTREQ8                 0x120
+# define INTC_INTREQ8_IREQUEST256_OFFSET 0
+# define INTC_INTREQ8_IREQUEST256_SIZE 1
+#define INTC_INTPR9                  0x24
+# define INTC_INTPR9_INTLEV_OFFSET   30
+# define INTC_INTPR9_INTLEV_SIZE     2
+# define INTC_INTPR9_OFFSET_OFFSET   0
+# define INTC_INTPR9_OFFSET_SIZE     24
+#define INTC_INTREQ9                 0x124
+# define INTC_INTREQ9_IREQUEST288_OFFSET 0
+# define INTC_INTREQ9_IREQUEST288_SIZE 1
+#define INTC_INTPR10                 0x28
+# define INTC_INTPR10_INTLEV_OFFSET  30
+# define INTC_INTPR10_INTLEV_SIZE    2
+# define INTC_INTPR10_OFFSET_OFFSET  0
+# define INTC_INTPR10_OFFSET_SIZE    24
+#define INTC_INTREQ10                0x128
+# define INTC_INTREQ10_IREQUEST320_OFFSET 0
+# define INTC_INTREQ10_IREQUEST320_SIZE 1
+#define INTC_INTPR11                 0x2c
+# define INTC_INTPR11_INTLEV_OFFSET  30
+# define INTC_INTPR11_INTLEV_SIZE    2
+# define INTC_INTPR11_OFFSET_OFFSET  0
+# define INTC_INTPR11_OFFSET_SIZE    24
+#define INTC_INTREQ11                0x12c
+# define INTC_INTREQ11_IREQUEST352_OFFSET 0
+# define INTC_INTREQ11_IREQUEST352_SIZE 1
+#define INTC_INTPR12                 0x30
+# define INTC_INTPR12_INTLEV_OFFSET  30
+# define INTC_INTPR12_INTLEV_SIZE    2
+# define INTC_INTPR12_OFFSET_OFFSET  0
+# define INTC_INTPR12_OFFSET_SIZE    24
+#define INTC_INTREQ12                0x130
+# define INTC_INTREQ12_IREQUEST384_OFFSET 0
+# define INTC_INTREQ12_IREQUEST384_SIZE 1
+#define INTC_INTPR13                 0x34
+# define INTC_INTPR13_INTLEV_OFFSET  30
+# define INTC_INTPR13_INTLEV_SIZE    2
+# define INTC_INTPR13_OFFSET_OFFSET  0
+# define INTC_INTPR13_OFFSET_SIZE    24
+#define INTC_INTREQ13                0x134
+# define INTC_INTREQ13_IREQUEST416_OFFSET 0
+# define INTC_INTREQ13_IREQUEST416_SIZE 1
+#define INTC_INTPR14                 0x38
+# define INTC_INTPR14_INTLEV_OFFSET  30
+# define INTC_INTPR14_INTLEV_SIZE    2
+# define INTC_INTPR14_OFFSET_OFFSET  0
+# define INTC_INTPR14_OFFSET_SIZE    24
+#define INTC_INTREQ14                0x138
+# define INTC_INTREQ14_IREQUEST448_OFFSET 0
+# define INTC_INTREQ14_IREQUEST448_SIZE 1
+#define INTC_INTPR15                 0x3c
+# define INTC_INTPR15_INTLEV_OFFSET  30
+# define INTC_INTPR15_INTLEV_SIZE    2
+# define INTC_INTPR15_OFFSET_OFFSET  0
+# define INTC_INTPR15_OFFSET_SIZE    24
+#define INTC_INTREQ15                0x13c
+# define INTC_INTREQ15_IREQUEST480_OFFSET 0
+# define INTC_INTREQ15_IREQUEST480_SIZE 1
+#define INTC_INTPR16                 0x40
+# define INTC_INTPR16_INTLEV_OFFSET  30
+# define INTC_INTPR16_INTLEV_SIZE    2
+# define INTC_INTPR16_OFFSET_OFFSET  0
+# define INTC_INTPR16_OFFSET_SIZE    24
+#define INTC_INTREQ16                0x140
+# define INTC_INTREQ16_IREQUEST512_OFFSET 0
+# define INTC_INTREQ16_IREQUEST512_SIZE 1
+#define INTC_INTPR17                 0x44
+# define INTC_INTPR17_INTLEV_OFFSET  30
+# define INTC_INTPR17_INTLEV_SIZE    2
+# define INTC_INTPR17_OFFSET_OFFSET  0
+# define INTC_INTPR17_OFFSET_SIZE    24
+#define INTC_INTREQ17                0x144
+# define INTC_INTREQ17_IREQUEST544_OFFSET 0
+# define INTC_INTREQ17_IREQUEST544_SIZE 1
+#define INTC_INTPR18                 0x48
+# define INTC_INTPR18_INTLEV_OFFSET  30
+# define INTC_INTPR18_INTLEV_SIZE    2
+# define INTC_INTPR18_OFFSET_OFFSET  0
+# define INTC_INTPR18_OFFSET_SIZE    24
+#define INTC_INTREQ18                0x148
+# define INTC_INTREQ18_IREQUEST576_OFFSET 0
+# define INTC_INTREQ18_IREQUEST576_SIZE 1
+#define INTC_INTPR19                 0x4c
+# define INTC_INTPR19_INTLEV_OFFSET  30
+# define INTC_INTPR19_INTLEV_SIZE    2
+# define INTC_INTPR19_OFFSET_OFFSET  0
+# define INTC_INTPR19_OFFSET_SIZE    24
+#define INTC_INTREQ19                0x14c
+# define INTC_INTREQ19_IREQUEST608_OFFSET 0
+# define INTC_INTREQ19_IREQUEST608_SIZE 1
+# define INTC_INTREQ19_IREQUEST609_OFFSET 1
+# define INTC_INTREQ19_IREQUEST609_SIZE 1
+# define INTC_INTREQ19_IREQUEST610_OFFSET 2
+# define INTC_INTREQ19_IREQUEST610_SIZE 1
+# define INTC_INTREQ19_IREQUEST611_OFFSET 3
+# define INTC_INTREQ19_IREQUEST611_SIZE 1
+#define INTC_INTPR20                 0x50
+# define INTC_INTPR20_INTLEV_OFFSET  30
+# define INTC_INTPR20_INTLEV_SIZE    2
+# define INTC_INTPR20_OFFSET_OFFSET  0
+# define INTC_INTPR20_OFFSET_SIZE    24
+#define INTC_INTREQ20                0x150
+# define INTC_INTREQ20_IREQUEST640_OFFSET 0
+# define INTC_INTREQ20_IREQUEST640_SIZE 1
+#define INTC_INTPR21                 0x54
+# define INTC_INTPR21_INTLEV_OFFSET  30
+# define INTC_INTPR21_INTLEV_SIZE    2
+# define INTC_INTPR21_OFFSET_OFFSET  0
+# define INTC_INTPR21_OFFSET_SIZE    24
+#define INTC_INTREQ21                0x154
+# define INTC_INTREQ21_IREQUEST672_OFFSET 0
+# define INTC_INTREQ21_IREQUEST672_SIZE 1
+#define INTC_INTPR22                 0x58
+# define INTC_INTPR22_INTLEV_OFFSET  30
+# define INTC_INTPR22_INTLEV_SIZE    2
+# define INTC_INTPR22_OFFSET_OFFSET  0
+# define INTC_INTPR22_OFFSET_SIZE    24
+#define INTC_INTREQ22                0x158
+# define INTC_INTREQ22_IREQUEST704_OFFSET 0
+# define INTC_INTREQ22_IREQUEST704_SIZE 1
+# define INTC_INTREQ22_IREQUEST705_OFFSET 1
+# define INTC_INTREQ22_IREQUEST705_SIZE 1
+# define INTC_INTREQ22_IREQUEST706_OFFSET 2
+# define INTC_INTREQ22_IREQUEST706_SIZE 1
+#define INTC_INTPR23                 0x5c
+# define INTC_INTPR23_INTLEV_OFFSET  30
+# define INTC_INTPR23_INTLEV_SIZE    2
+# define INTC_INTPR23_OFFSET_OFFSET  0
+# define INTC_INTPR23_OFFSET_SIZE    24
+#define INTC_INTREQ23                0x15c
+# define INTC_INTREQ23_IREQUEST736_OFFSET 0
+# define INTC_INTREQ23_IREQUEST736_SIZE 1
+# define INTC_INTREQ23_IREQUEST737_OFFSET 1
+# define INTC_INTREQ23_IREQUEST737_SIZE 1
+# define INTC_INTREQ23_IREQUEST738_OFFSET 2
+# define INTC_INTREQ23_IREQUEST738_SIZE 1
+#define INTC_INTPR24                 0x60
+# define INTC_INTPR24_INTLEV_OFFSET  30
+# define INTC_INTPR24_INTLEV_SIZE    2
+# define INTC_INTPR24_OFFSET_OFFSET  0
+# define INTC_INTPR24_OFFSET_SIZE    24
+#define INTC_INTREQ24                0x160
+# define INTC_INTREQ24_IREQUEST768_OFFSET 0
+# define INTC_INTREQ24_IREQUEST768_SIZE 1
+#define INTC_INTPR25                 0x64
+# define INTC_INTPR25_INTLEV_OFFSET  30
+# define INTC_INTPR25_INTLEV_SIZE    2
+# define INTC_INTPR25_OFFSET_OFFSET  0
+# define INTC_INTPR25_OFFSET_SIZE    24
+#define INTC_INTREQ25                0x164
+# define INTC_INTREQ25_IREQUEST800_OFFSET 0
+# define INTC_INTREQ25_IREQUEST800_SIZE 1
+#define INTC_INTPR26                 0x68
+# define INTC_INTPR26_INTLEV_OFFSET  30
+# define INTC_INTPR26_INTLEV_SIZE    2
+# define INTC_INTPR26_OFFSET_OFFSET  0
+# define INTC_INTPR26_OFFSET_SIZE    24
+#define INTC_INTREQ26                0x168
+# define INTC_INTREQ26_IREQUEST832_OFFSET 0
+# define INTC_INTREQ26_IREQUEST832_SIZE 1
+#define INTC_INTPR27                 0x6c
+# define INTC_INTPR27_INTLEV_OFFSET  30
+# define INTC_INTPR27_INTLEV_SIZE    2
+# define INTC_INTPR27_OFFSET_OFFSET  0
+# define INTC_INTPR27_OFFSET_SIZE    24
+#define INTC_INTREQ27                0x16c
+# define INTC_INTREQ27_IREQUEST864_OFFSET 0
+# define INTC_INTREQ27_IREQUEST864_SIZE 1
+#define INTC_INTPR28                 0x70
+# define INTC_INTPR28_INTLEV_OFFSET  30
+# define INTC_INTPR28_INTLEV_SIZE    2
+# define INTC_INTPR28_OFFSET_OFFSET  0
+# define INTC_INTPR28_OFFSET_SIZE    24
+#define INTC_INTREQ28                0x170
+# define INTC_INTREQ28_IREQUEST896_OFFSET 0
+# define INTC_INTREQ28_IREQUEST896_SIZE 1
+#define INTC_INTPR29                 0x74
+# define INTC_INTPR29_INTLEV_OFFSET  30
+# define INTC_INTPR29_INTLEV_SIZE    2
+# define INTC_INTPR29_OFFSET_OFFSET  0
+# define INTC_INTPR29_OFFSET_SIZE    24
+#define INTC_INTREQ29                0x174
+# define INTC_INTREQ29_IREQUEST928_OFFSET 0
+# define INTC_INTREQ29_IREQUEST928_SIZE 1
+#define INTC_INTPR30                 0x78
+# define INTC_INTPR30_INTLEV_OFFSET  30
+# define INTC_INTPR30_INTLEV_SIZE    2
+# define INTC_INTPR30_OFFSET_OFFSET  0
+# define INTC_INTPR30_OFFSET_SIZE    24
+#define INTC_INTREQ30                0x178
+# define INTC_INTREQ30_IREQUEST960_OFFSET 0
+# define INTC_INTREQ30_IREQUEST960_SIZE 1
+#define INTC_INTPR31                 0x7c
+# define INTC_INTPR31_INTLEV_OFFSET  30
+# define INTC_INTPR31_INTLEV_SIZE    2
+# define INTC_INTPR31_OFFSET_OFFSET  0
+# define INTC_INTPR31_OFFSET_SIZE    24
+#define INTC_INTREQ31                0x17c
+# define INTC_INTREQ31_IREQUEST992_OFFSET 0
+# define INTC_INTREQ31_IREQUEST992_SIZE 1
+#define INTC_INTPR32                 0x80
+# define INTC_INTPR32_INTLEV_OFFSET  30
+# define INTC_INTPR32_INTLEV_SIZE    2
+# define INTC_INTPR32_OFFSET_OFFSET  0
+# define INTC_INTPR32_OFFSET_SIZE    24
+#define INTC_INTREQ32                0x180
+# define INTC_INTREQ32_IREQUEST1024_OFFSET 0
+# define INTC_INTREQ32_IREQUEST1024_SIZE 1
+#define INTC_INTCAUSE0               0x20c
+# define INTC_INTCAUSE0_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE0_CAUSEGRP_SIZE 6
+#define INTC_INTCAUSE1               0x208
+# define INTC_INTCAUSE1_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE1_CAUSEGRP_SIZE 6
+#define INTC_INTCAUSE2               0x204
+# define INTC_INTCAUSE2_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE2_CAUSEGRP_SIZE 6
+#define INTC_INTCAUSE3               0x200
+# define INTC_INTCAUSE3_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE3_CAUSEGRP_SIZE 6
+
+#define INTC_BIT(name)               (1 << INTC_##name##_OFFSET)
+#define INTC_MKBF(name, value)       (((value) & ((1 << INTC_##name##_SIZE) - 1)) << INTC_##name##_OFFSET)
+#define INTC_GETBF(name, value)      (((value) >> INTC_##name##_OFFSET) & ((1 << INTC_##name##_SIZE) - 1))
+
+#define intc_readl(port,reg)         readl((port)->regs + INTC_##reg)
+#define intc_writel(port,reg,value)  writel((value), (port)->regs + INTC_##reg)
+
+#endif /* __ASM_AVR32_PERIHP_INTC_H__ */
diff --git a/arch/avr32/mach-at32ap/pio.c b/arch/avr32/mach-at32ap/pio.c
new file mode 100644
index 000000000000..d3aabfca8598
--- /dev/null
+++ b/arch/avr32/mach-at32ap/pio.c
@@ -0,0 +1,118 @@
+/*
+ * Atmel PIO2 Port Multiplexer support
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/debugfs.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/portmux.h>
+
+#include "pio.h"
+
+#define MAX_NR_PIO_DEVICES		8
+
+struct pio_device {
+	void __iomem *regs;
+	const struct platform_device *pdev;
+	struct clk *clk;
+	u32 alloc_mask;
+	char name[32];
+};
+
+static struct pio_device pio_dev[MAX_NR_PIO_DEVICES];
+
+void portmux_set_func(unsigned int portmux_id, unsigned int pin_id,
+		      unsigned int function_id)
+{
+	struct pio_device *pio;
+	u32 mask = 1 << pin_id;
+
+	BUG_ON(portmux_id >= MAX_NR_PIO_DEVICES);
+
+	pio = &pio_dev[portmux_id];
+
+	if (function_id)
+		pio_writel(pio, BSR, mask);
+	else
+		pio_writel(pio, ASR, mask);
+	pio_writel(pio, PDR, mask);
+}
+
+static int __init pio_probe(struct platform_device *pdev)
+{
+	struct pio_device *pio = NULL;
+
+	BUG_ON(pdev->id >= MAX_NR_PIO_DEVICES);
+	pio = &pio_dev[pdev->id];
+	BUG_ON(!pio->regs);
+
+	/* TODO: Interrupts */
+
+	platform_set_drvdata(pdev, pio);
+
+	printk(KERN_INFO "%s: Atmel Port Multiplexer at 0x%p (irq %d)\n",
+	       pio->name, pio->regs, platform_get_irq(pdev, 0));
+
+	return 0;
+}
+
+static struct platform_driver pio_driver = {
+	.probe		= pio_probe,
+	.driver		= {
+		.name		= "pio",
+	},
+};
+
+static int __init pio_init(void)
+{
+	return platform_driver_register(&pio_driver);
+}
+subsys_initcall(pio_init);
+
+void __init at32_init_pio(struct platform_device *pdev)
+{
+	struct resource *regs;
+	struct pio_device *pio;
+
+	if (pdev->id > MAX_NR_PIO_DEVICES) {
+		dev_err(&pdev->dev, "only %d PIO devices supported\n",
+			MAX_NR_PIO_DEVICES);
+		return;
+	}
+
+	pio = &pio_dev[pdev->id];
+	snprintf(pio->name, sizeof(pio->name), "pio%d", pdev->id);
+
+	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!regs) {
+		dev_err(&pdev->dev, "no mmio resource defined\n");
+		return;
+	}
+
+	pio->clk = clk_get(&pdev->dev, "mck");
+	if (IS_ERR(pio->clk))
+		/*
+		 * This is a fatal error, but if we continue we might
+		 * be so lucky that we manage to initialize the
+		 * console and display this message...
+		 */
+		dev_err(&pdev->dev, "no mck clock defined\n");
+	else
+		clk_enable(pio->clk);
+
+	pio->pdev = pdev;
+	pio->regs = ioremap(regs->start, regs->end - regs->start + 1);
+
+	pio_writel(pio, ODR, ~0UL);
+	pio_writel(pio, PER, ~0UL);
+}
diff --git a/arch/avr32/mach-at32ap/pio.h b/arch/avr32/mach-at32ap/pio.h
new file mode 100644
index 000000000000..cfea12351599
--- /dev/null
+++ b/arch/avr32/mach-at32ap/pio.h
@@ -0,0 +1,178 @@
+/*
+ * Atmel PIO2 Port Multiplexer support
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ARCH_AVR32_AT32AP_PIO_H__
+#define __ARCH_AVR32_AT32AP_PIO_H__
+
+/* PIO register offsets */
+#define PIO_PER                                0x0000
+#define PIO_PDR                                0x0004
+#define PIO_PSR                                0x0008
+#define PIO_OER                                0x0010
+#define PIO_ODR                                0x0014
+#define PIO_OSR                                0x0018
+#define PIO_IFER                               0x0020
+#define PIO_IFDR                               0x0024
+#define PIO_ISFR                               0x0028
+#define PIO_SODR                               0x0030
+#define PIO_CODR                               0x0034
+#define PIO_ODSR                               0x0038
+#define PIO_PDSR                               0x003c
+#define PIO_IER                                0x0040
+#define PIO_IDR                                0x0044
+#define PIO_IMR                                0x0048
+#define PIO_ISR                                0x004c
+#define PIO_MDER                               0x0050
+#define PIO_MDDR                               0x0054
+#define PIO_MDSR                               0x0058
+#define PIO_PUDR                               0x0060
+#define PIO_PUER                               0x0064
+#define PIO_PUSR                               0x0068
+#define PIO_ASR                                0x0070
+#define PIO_BSR                                0x0074
+#define PIO_ABSR                               0x0078
+#define PIO_OWER                               0x00a0
+#define PIO_OWDR                               0x00a4
+#define PIO_OWSR                               0x00a8
+
+/* Bitfields in PER */
+
+/* Bitfields in PDR */
+
+/* Bitfields in PSR */
+
+/* Bitfields in OER */
+
+/* Bitfields in ODR */
+
+/* Bitfields in OSR */
+
+/* Bitfields in IFER */
+
+/* Bitfields in IFDR */
+
+/* Bitfields in ISFR */
+
+/* Bitfields in SODR */
+
+/* Bitfields in CODR */
+
+/* Bitfields in ODSR */
+
+/* Bitfields in PDSR */
+
+/* Bitfields in IER */
+
+/* Bitfields in IDR */
+
+/* Bitfields in IMR */
+
+/* Bitfields in ISR */
+
+/* Bitfields in MDER */
+
+/* Bitfields in MDDR */
+
+/* Bitfields in MDSR */
+
+/* Bitfields in PUDR */
+
+/* Bitfields in PUER */
+
+/* Bitfields in PUSR */
+
+/* Bitfields in ASR */
+
+/* Bitfields in BSR */
+
+/* Bitfields in ABSR */
+#define PIO_P0_OFFSET                          0
+#define PIO_P0_SIZE                            1
+#define PIO_P1_OFFSET                          1
+#define PIO_P1_SIZE                            1
+#define PIO_P2_OFFSET                          2
+#define PIO_P2_SIZE                            1
+#define PIO_P3_OFFSET                          3
+#define PIO_P3_SIZE                            1
+#define PIO_P4_OFFSET                          4
+#define PIO_P4_SIZE                            1
+#define PIO_P5_OFFSET                          5
+#define PIO_P5_SIZE                            1
+#define PIO_P6_OFFSET                          6
+#define PIO_P6_SIZE                            1
+#define PIO_P7_OFFSET                          7
+#define PIO_P7_SIZE                            1
+#define PIO_P8_OFFSET                          8
+#define PIO_P8_SIZE                            1
+#define PIO_P9_OFFSET                          9
+#define PIO_P9_SIZE                            1
+#define PIO_P10_OFFSET                         10
+#define PIO_P10_SIZE                           1
+#define PIO_P11_OFFSET                         11
+#define PIO_P11_SIZE                           1
+#define PIO_P12_OFFSET                         12
+#define PIO_P12_SIZE                           1
+#define PIO_P13_OFFSET                         13
+#define PIO_P13_SIZE                           1
+#define PIO_P14_OFFSET                         14
+#define PIO_P14_SIZE                           1
+#define PIO_P15_OFFSET                         15
+#define PIO_P15_SIZE                           1
+#define PIO_P16_OFFSET                         16
+#define PIO_P16_SIZE                           1
+#define PIO_P17_OFFSET                         17
+#define PIO_P17_SIZE                           1
+#define PIO_P18_OFFSET                         18
+#define PIO_P18_SIZE                           1
+#define PIO_P19_OFFSET                         19
+#define PIO_P19_SIZE                           1
+#define PIO_P20_OFFSET                         20
+#define PIO_P20_SIZE                           1
+#define PIO_P21_OFFSET                         21
+#define PIO_P21_SIZE                           1
+#define PIO_P22_OFFSET                         22
+#define PIO_P22_SIZE                           1
+#define PIO_P23_OFFSET                         23
+#define PIO_P23_SIZE                           1
+#define PIO_P24_OFFSET                         24
+#define PIO_P24_SIZE                           1
+#define PIO_P25_OFFSET                         25
+#define PIO_P25_SIZE                           1
+#define PIO_P26_OFFSET                         26
+#define PIO_P26_SIZE                           1
+#define PIO_P27_OFFSET                         27
+#define PIO_P27_SIZE                           1
+#define PIO_P28_OFFSET                         28
+#define PIO_P28_SIZE                           1
+#define PIO_P29_OFFSET                         29
+#define PIO_P29_SIZE                           1
+#define PIO_P30_OFFSET                         30
+#define PIO_P30_SIZE                           1
+#define PIO_P31_OFFSET                         31
+#define PIO_P31_SIZE                           1
+
+/* Bitfields in OWER */
+
+/* Bitfields in OWDR */
+
+/* Bitfields in OWSR */
+
+/* Bit manipulation macros */
+#define PIO_BIT(name)                          (1 << PIO_##name##_OFFSET)
+#define PIO_BF(name,value)                     (((value) & ((1 << PIO_##name##_SIZE) - 1)) << PIO_##name##_OFFSET)
+#define PIO_BFEXT(name,value)                  (((value) >> PIO_##name##_OFFSET) & ((1 << PIO_##name##_SIZE) - 1))
+#define PIO_BFINS(name,value,old)              (((old) & ~(((1 << PIO_##name##_SIZE) - 1) << PIO_##name##_OFFSET)) | PIO_BF(name,value))
+
+/* Register access macros */
+#define pio_readl(port,reg)                    readl((port)->regs + PIO_##reg)
+#define pio_writel(port,reg,value)             writel((value), (port)->regs + PIO_##reg)
+
+void at32_init_pio(struct platform_device *pdev);
+
+#endif /* __ARCH_AVR32_AT32AP_PIO_H__ */
diff --git a/arch/avr32/mach-at32ap/sm.c b/arch/avr32/mach-at32ap/sm.c
new file mode 100644
index 000000000000..03306eb0345e
--- /dev/null
+++ b/arch/avr32/mach-at32ap/sm.c
@@ -0,0 +1,289 @@
+/*
+ * System Manager driver for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/random.h>
+#include <linux/spinlock.h>
+
+#include <asm/intc.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+
+#include <asm/arch/sm.h>
+
+#include "sm.h"
+
+#define SM_EIM_IRQ_RESOURCE	1
+#define SM_PM_IRQ_RESOURCE	2
+#define SM_RTC_IRQ_RESOURCE	3
+
+#define to_eim(irqc) container_of(irqc, struct at32_sm, irqc)
+
+struct at32_sm system_manager;
+
+int __init at32_sm_init(void)
+{
+	struct resource *regs;
+	struct at32_sm *sm = &system_manager;
+	int ret = -ENXIO;
+
+	regs = platform_get_resource(&at32_sm_device, IORESOURCE_MEM, 0);
+	if (!regs)
+		goto fail;
+
+	spin_lock_init(&sm->lock);
+	sm->pdev = &at32_sm_device;
+
+	ret = -ENOMEM;
+	sm->regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!sm->regs)
+		goto fail;
+
+	return 0;
+
+fail:
+	printk(KERN_ERR "Failed to initialize System Manager: %d\n", ret);
+	return ret;
+}
+
+/*
+ * External Interrupt Module (EIM).
+ *
+ * EIM gets level- or edge-triggered interrupts of either polarity
+ * from the outside and converts it to active-high level-triggered
+ * interrupts that the internal interrupt controller can handle. EIM
+ * also provides masking/unmasking of interrupts, as well as
+ * acknowledging of edge-triggered interrupts.
+ */
+
+static irqreturn_t spurious_eim_interrupt(int irq, void *dev_id,
+					  struct pt_regs *regs)
+{
+	printk(KERN_WARNING "Spurious EIM interrupt %d\n", irq);
+	disable_irq(irq);
+	return IRQ_NONE;
+}
+
+static struct irqaction eim_spurious_action = {
+	.handler = spurious_eim_interrupt,
+};
+
+static irqreturn_t eim_handle_irq(int irq, void *dev_id, struct pt_regs *regs)
+{
+	struct irq_controller * irqc = dev_id;
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned long pending;
+
+	/*
+	 * No need to disable interrupts globally.  The interrupt
+	 * level relevant to this group must be masked all the time,
+	 * so we know that this particular EIM instance will not be
+	 * re-entered.
+	 */
+	spin_lock(&sm->lock);
+
+	pending = intc_get_pending(sm->irqc.irq_group);
+	if (unlikely(!pending)) {
+		printk(KERN_ERR "EIM (group %u): No interrupts pending!\n",
+		       sm->irqc.irq_group);
+		goto unlock;
+	}
+
+	do {
+		struct irqaction *action;
+		unsigned int i;
+
+		i = fls(pending) - 1;
+		pending &= ~(1 << i);
+		action = sm->action[i];
+
+		/* Acknowledge the interrupt */
+		sm_writel(sm, EIM_ICR, 1 << i);
+
+		spin_unlock(&sm->lock);
+
+		if (action->flags & SA_INTERRUPT)
+			local_irq_disable();
+		action->handler(sm->irqc.first_irq + i, action->dev_id, regs);
+		local_irq_enable();
+		spin_lock(&sm->lock);
+		if (action->flags & SA_SAMPLE_RANDOM)
+			add_interrupt_randomness(sm->irqc.first_irq + i);
+	} while (pending);
+
+unlock:
+	spin_unlock(&sm->lock);
+	return IRQ_HANDLED;
+}
+
+static void eim_mask(struct irq_controller *irqc, unsigned int irq)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned int i;
+
+	i = irq - sm->irqc.first_irq;
+	sm_writel(sm, EIM_IDR, 1 << i);
+}
+
+static void eim_unmask(struct irq_controller *irqc, unsigned int irq)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned int i;
+
+	i = irq - sm->irqc.first_irq;
+	sm_writel(sm, EIM_IER, 1 << i);
+}
+
+static int eim_setup(struct irq_controller *irqc, unsigned int irq,
+		struct irqaction *action)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	sm->action[irq - sm->irqc.first_irq] = action;
+	/* Acknowledge earlier interrupts */
+	sm_writel(sm, EIM_ICR, (1<<(irq - sm->irqc.first_irq)));
+	eim_unmask(irqc, irq);
+	return 0;
+}
+
+static void eim_free(struct irq_controller *irqc, unsigned int irq,
+		void *dev)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	eim_mask(irqc, irq);
+	sm->action[irq - sm->irqc.first_irq] = &eim_spurious_action;
+}
+
+static int eim_set_type(struct irq_controller *irqc, unsigned int irq,
+			unsigned int type)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned long flags;
+	u32 value, pattern;
+
+	spin_lock_irqsave(&sm->lock, flags);
+
+	pattern = 1 << (irq - sm->irqc.first_irq);
+
+	value = sm_readl(sm, EIM_MODE);
+	if (type & IRQ_TYPE_LEVEL)
+		value |= pattern;
+	else
+		value &= ~pattern;
+	sm_writel(sm, EIM_MODE, value);
+	value = sm_readl(sm, EIM_EDGE);
+	if (type & IRQ_EDGE_RISING)
+		value |= pattern;
+	else
+		value &= ~pattern;
+	sm_writel(sm, EIM_EDGE, value);
+	value = sm_readl(sm, EIM_LEVEL);
+	if (type & IRQ_LEVEL_HIGH)
+		value |= pattern;
+	else
+		value &= ~pattern;
+	sm_writel(sm, EIM_LEVEL, value);
+
+	spin_unlock_irqrestore(&sm->lock, flags);
+
+	return 0;
+}
+
+static unsigned int eim_get_type(struct irq_controller *irqc,
+				 unsigned int irq)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned long flags;
+	unsigned int type = 0;
+	u32 mode, edge, level, pattern;
+
+	pattern = 1 << (irq - sm->irqc.first_irq);
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mode = sm_readl(sm, EIM_MODE);
+	edge = sm_readl(sm, EIM_EDGE);
+	level = sm_readl(sm, EIM_LEVEL);
+	spin_unlock_irqrestore(&sm->lock, flags);
+
+	if (mode & pattern)
+		type |= IRQ_TYPE_LEVEL;
+	if (edge & pattern)
+		type |= IRQ_EDGE_RISING;
+	if (level & pattern)
+		type |= IRQ_LEVEL_HIGH;
+
+	return type;
+}
+
+static struct irq_controller_class eim_irq_class = {
+	.typename	= "EIM",
+	.handle		= eim_handle_irq,
+	.setup		= eim_setup,
+	.free		= eim_free,
+	.mask		= eim_mask,
+	.unmask		= eim_unmask,
+	.set_type	= eim_set_type,
+	.get_type	= eim_get_type,
+};
+
+static int __init eim_init(void)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned int i;
+	u32 pattern;
+	int ret;
+
+	/*
+	 * The EIM is really the same module as SM, so register
+	 * mapping, etc. has been taken care of already.
+	 */
+
+	/*
+	 * Find out how many interrupt lines that are actually
+	 * implemented in hardware.
+	 */
+	sm_writel(sm, EIM_IDR, ~0UL);
+	sm_writel(sm, EIM_MODE, ~0UL);
+	pattern = sm_readl(sm, EIM_MODE);
+	sm->irqc.nr_irqs = fls(pattern);
+
+	ret = -ENOMEM;
+	sm->action = kmalloc(sizeof(*sm->action) * sm->irqc.nr_irqs,
+			     GFP_KERNEL);
+	if (!sm->action)
+		goto out;
+
+	for (i = 0; i < sm->irqc.nr_irqs; i++)
+		sm->action[i] = &eim_spurious_action;
+
+	spin_lock_init(&sm->lock);
+	sm->irqc.irq_group = sm->pdev->resource[SM_EIM_IRQ_RESOURCE].start;
+	sm->irqc.class = &eim_irq_class;
+
+	ret = intc_register_controller(&sm->irqc);
+	if (ret < 0)
+		goto out_free_actions;
+
+	printk("EIM: External Interrupt Module at 0x%p, IRQ group %u\n",
+	       sm->regs, sm->irqc.irq_group);
+	printk("EIM: Handling %u external IRQs, starting with IRQ%u\n",
+	       sm->irqc.nr_irqs, sm->irqc.first_irq);
+
+	return 0;
+
+out_free_actions:
+	kfree(sm->action);
+out:
+	return ret;
+}
+arch_initcall(eim_init);
diff --git a/arch/avr32/mach-at32ap/sm.h b/arch/avr32/mach-at32ap/sm.h
new file mode 100644
index 000000000000..27565822ae2a
--- /dev/null
+++ b/arch/avr32/mach-at32ap/sm.h
@@ -0,0 +1,240 @@
+/*
+ * Register definitions for SM
+ *
+ * System Manager
+ */
+#ifndef __ASM_AVR32_SM_H__
+#define __ASM_AVR32_SM_H__
+
+/* SM register offsets */
+#define SM_PM_MCCTRL                            0x0000
+#define SM_PM_CKSEL                             0x0004
+#define SM_PM_CPU_MASK                          0x0008
+#define SM_PM_HSB_MASK                          0x000c
+#define SM_PM_PBA_MASK                         0x0010
+#define SM_PM_PBB_MASK                         0x0014
+#define SM_PM_PLL0                              0x0020
+#define SM_PM_PLL1                              0x0024
+#define SM_PM_VCTRL                             0x0030
+#define SM_PM_VMREF                             0x0034
+#define SM_PM_VMV                               0x0038
+#define SM_PM_IER                               0x0040
+#define SM_PM_IDR                               0x0044
+#define SM_PM_IMR                               0x0048
+#define SM_PM_ISR                               0x004c
+#define SM_PM_ICR                               0x0050
+#define SM_PM_GCCTRL                            0x0060
+#define SM_RTC_CTRL                             0x0080
+#define SM_RTC_VAL                              0x0084
+#define SM_RTC_TOP                              0x0088
+#define SM_RTC_IER                              0x0090
+#define SM_RTC_IDR                              0x0094
+#define SM_RTC_IMR                              0x0098
+#define SM_RTC_ISR                              0x009c
+#define SM_RTC_ICR                              0x00a0
+#define SM_WDT_CTRL                             0x00b0
+#define SM_WDT_CLR                              0x00b4
+#define SM_WDT_EXT                              0x00b8
+#define SM_RC_RCAUSE                            0x00c0
+#define SM_EIM_IER                              0x0100
+#define SM_EIM_IDR                              0x0104
+#define SM_EIM_IMR                              0x0108
+#define SM_EIM_ISR                              0x010c
+#define SM_EIM_ICR                              0x0110
+#define SM_EIM_MODE                             0x0114
+#define SM_EIM_EDGE                             0x0118
+#define SM_EIM_LEVEL                            0x011c
+#define SM_EIM_TEST                             0x0120
+#define SM_EIM_NMIC                             0x0124
+
+/* Bitfields in PM_MCCTRL */
+
+/* Bitfields in PM_CKSEL */
+#define SM_CPUSEL_OFFSET                        0
+#define SM_CPUSEL_SIZE                          3
+#define SM_CPUDIV_OFFSET                        7
+#define SM_CPUDIV_SIZE                          1
+#define SM_HSBSEL_OFFSET                        8
+#define SM_HSBSEL_SIZE                          3
+#define SM_HSBDIV_OFFSET                        15
+#define SM_HSBDIV_SIZE                          1
+#define SM_PBASEL_OFFSET                       16
+#define SM_PBASEL_SIZE                         3
+#define SM_PBADIV_OFFSET                       23
+#define SM_PBADIV_SIZE                         1
+#define SM_PBBSEL_OFFSET                       24
+#define SM_PBBSEL_SIZE                         3
+#define SM_PBBDIV_OFFSET                       31
+#define SM_PBBDIV_SIZE                         1
+
+/* Bitfields in PM_CPU_MASK */
+
+/* Bitfields in PM_HSB_MASK */
+
+/* Bitfields in PM_PBA_MASK */
+
+/* Bitfields in PM_PBB_MASK */
+
+/* Bitfields in PM_PLL0 */
+#define SM_PLLEN_OFFSET                         0
+#define SM_PLLEN_SIZE                           1
+#define SM_PLLOSC_OFFSET                        1
+#define SM_PLLOSC_SIZE                          1
+#define SM_PLLOPT_OFFSET                        2
+#define SM_PLLOPT_SIZE                          3
+#define SM_PLLDIV_OFFSET                        8
+#define SM_PLLDIV_SIZE                          8
+#define SM_PLLMUL_OFFSET                        16
+#define SM_PLLMUL_SIZE                          8
+#define SM_PLLCOUNT_OFFSET                      24
+#define SM_PLLCOUNT_SIZE                        6
+#define SM_PLLTEST_OFFSET                       31
+#define SM_PLLTEST_SIZE                         1
+
+/* Bitfields in PM_PLL1 */
+
+/* Bitfields in PM_VCTRL */
+#define SM_VAUTO_OFFSET                         0
+#define SM_VAUTO_SIZE                           1
+#define SM_PM_VCTRL_VAL_OFFSET                  8
+#define SM_PM_VCTRL_VAL_SIZE                    7
+
+/* Bitfields in PM_VMREF */
+#define SM_REFSEL_OFFSET                        0
+#define SM_REFSEL_SIZE                          4
+
+/* Bitfields in PM_VMV */
+#define SM_PM_VMV_VAL_OFFSET                    0
+#define SM_PM_VMV_VAL_SIZE                      8
+
+/* Bitfields in PM_IER */
+
+/* Bitfields in PM_IDR */
+
+/* Bitfields in PM_IMR */
+
+/* Bitfields in PM_ISR */
+
+/* Bitfields in PM_ICR */
+#define SM_LOCK0_OFFSET                         0
+#define SM_LOCK0_SIZE                           1
+#define SM_LOCK1_OFFSET                         1
+#define SM_LOCK1_SIZE                           1
+#define SM_WAKE_OFFSET                          2
+#define SM_WAKE_SIZE                            1
+#define SM_VOK_OFFSET                           3
+#define SM_VOK_SIZE                             1
+#define SM_VMRDY_OFFSET                         4
+#define SM_VMRDY_SIZE                           1
+#define SM_CKRDY_OFFSET                         5
+#define SM_CKRDY_SIZE                           1
+
+/* Bitfields in PM_GCCTRL */
+#define SM_OSCSEL_OFFSET                        0
+#define SM_OSCSEL_SIZE                          1
+#define SM_PLLSEL_OFFSET                        1
+#define SM_PLLSEL_SIZE                          1
+#define SM_CEN_OFFSET                           2
+#define SM_CEN_SIZE                             1
+#define SM_CPC_OFFSET                           3
+#define SM_CPC_SIZE                             1
+#define SM_DIVEN_OFFSET                         4
+#define SM_DIVEN_SIZE                           1
+#define SM_DIV_OFFSET                           8
+#define SM_DIV_SIZE                             8
+
+/* Bitfields in RTC_CTRL */
+#define SM_PCLR_OFFSET                          1
+#define SM_PCLR_SIZE                            1
+#define SM_TOPEN_OFFSET                         2
+#define SM_TOPEN_SIZE                           1
+#define SM_CLKEN_OFFSET                         3
+#define SM_CLKEN_SIZE                           1
+#define SM_PSEL_OFFSET                          8
+#define SM_PSEL_SIZE                            16
+
+/* Bitfields in RTC_VAL */
+#define SM_RTC_VAL_VAL_OFFSET                   0
+#define SM_RTC_VAL_VAL_SIZE                     31
+
+/* Bitfields in RTC_TOP */
+#define SM_RTC_TOP_VAL_OFFSET                   0
+#define SM_RTC_TOP_VAL_SIZE                     32
+
+/* Bitfields in RTC_IER */
+
+/* Bitfields in RTC_IDR */
+
+/* Bitfields in RTC_IMR */
+
+/* Bitfields in RTC_ISR */
+
+/* Bitfields in RTC_ICR */
+#define SM_TOPI_OFFSET                          0
+#define SM_TOPI_SIZE                            1
+
+/* Bitfields in WDT_CTRL */
+#define SM_KEY_OFFSET                           24
+#define SM_KEY_SIZE                             8
+
+/* Bitfields in WDT_CLR */
+
+/* Bitfields in WDT_EXT */
+
+/* Bitfields in RC_RCAUSE */
+#define SM_POR_OFFSET                           0
+#define SM_POR_SIZE                             1
+#define SM_BOD_OFFSET                           1
+#define SM_BOD_SIZE                             1
+#define SM_EXT_OFFSET                           2
+#define SM_EXT_SIZE                             1
+#define SM_WDT_OFFSET                           3
+#define SM_WDT_SIZE                             1
+#define SM_NTAE_OFFSET                          4
+#define SM_NTAE_SIZE                            1
+#define SM_SERP_OFFSET                          5
+#define SM_SERP_SIZE                            1
+
+/* Bitfields in EIM_IER */
+
+/* Bitfields in EIM_IDR */
+
+/* Bitfields in EIM_IMR */
+
+/* Bitfields in EIM_ISR */
+
+/* Bitfields in EIM_ICR */
+
+/* Bitfields in EIM_MODE */
+
+/* Bitfields in EIM_EDGE */
+#define SM_INT0_OFFSET                          0
+#define SM_INT0_SIZE                            1
+#define SM_INT1_OFFSET                          1
+#define SM_INT1_SIZE                            1
+#define SM_INT2_OFFSET                          2
+#define SM_INT2_SIZE                            1
+#define SM_INT3_OFFSET                          3
+#define SM_INT3_SIZE                            1
+
+/* Bitfields in EIM_LEVEL */
+
+/* Bitfields in EIM_TEST */
+#define SM_TESTEN_OFFSET                        31
+#define SM_TESTEN_SIZE                          1
+
+/* Bitfields in EIM_NMIC */
+#define SM_EN_OFFSET                            0
+#define SM_EN_SIZE                              1
+
+/* Bit manipulation macros */
+#define SM_BIT(name)                            (1 << SM_##name##_OFFSET)
+#define SM_BF(name,value)                       (((value) & ((1 << SM_##name##_SIZE) - 1)) << SM_##name##_OFFSET)
+#define SM_BFEXT(name,value)                    (((value) >> SM_##name##_OFFSET) & ((1 << SM_##name##_SIZE) - 1))
+#define SM_BFINS(name,value,old)                (((old) & ~(((1 << SM_##name##_SIZE) - 1) << SM_##name##_OFFSET)) | SM_BF(name,value))
+
+/* Register access macros */
+#define sm_readl(port,reg)                      readl((port)->regs + SM_##reg)
+#define sm_writel(port,reg,value)               writel((value), (port)->regs + SM_##reg)
+
+#endif /* __ASM_AVR32_SM_H__ */
diff --git a/arch/avr32/mm/Makefile b/arch/avr32/mm/Makefile
new file mode 100644
index 000000000000..0066491f90d4
--- /dev/null
+++ b/arch/avr32/mm/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the Linux/AVR32 kernel.
+#
+
+obj-y				+= init.o clear_page.o copy_page.o dma-coherent.o
+obj-y				+= ioremap.o cache.o fault.o tlb.o
diff --git a/arch/avr32/mm/cache.c b/arch/avr32/mm/cache.c
new file mode 100644
index 000000000000..450515b245a0
--- /dev/null
+++ b/arch/avr32/mm/cache.c
@@ -0,0 +1,150 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/highmem.h>
+#include <linux/unistd.h>
+
+#include <asm/cacheflush.h>
+#include <asm/cachectl.h>
+#include <asm/processor.h>
+#include <asm/uaccess.h>
+
+/*
+ * If you attempt to flush anything more than this, you need superuser
+ * privileges.  The value is completely arbitrary.
+ */
+#define CACHEFLUSH_MAX_LEN	1024
+
+void invalidate_dcache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+
+	//printk("invalidate dcache: %p + %u\n", start, size);
+
+	/* You asked for it, you got it */
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		invalidate_dcache_line((void *)v);
+}
+
+void clean_dcache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		clean_dcache_line((void *)v);
+	flush_write_buffer();
+}
+
+void flush_dcache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		flush_dcache_line((void *)v);
+	flush_write_buffer();
+}
+
+void invalidate_icache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.icache.linesz;
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		invalidate_icache_line((void *)v);
+}
+
+static inline void __flush_icache_range(unsigned long start, unsigned long end)
+{
+	unsigned long v, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	for (v = start; v < end; v += linesz) {
+		clean_dcache_line((void *)v);
+		invalidate_icache_line((void *)v);
+	}
+
+	flush_write_buffer();
+}
+
+/*
+ * This one is called after a module has been loaded.
+ */
+void flush_icache_range(unsigned long start, unsigned long end)
+{
+	unsigned long linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	__flush_icache_range(start & ~(linesz - 1),
+			     (end + linesz - 1) & ~(linesz - 1));
+}
+
+/*
+ * This one is called from do_no_page(), do_swap_page() and install_page().
+ */
+void flush_icache_page(struct vm_area_struct *vma, struct page *page)
+{
+	if (vma->vm_flags & VM_EXEC) {
+		void *v = kmap(page);
+		__flush_icache_range((unsigned long)v, (unsigned long)v + PAGE_SIZE);
+		kunmap(v);
+	}
+}
+
+/*
+ * This one is used by copy_to_user_page()
+ */
+void flush_icache_user_range(struct vm_area_struct *vma, struct page *page,
+			     unsigned long addr, int len)
+{
+	if (vma->vm_flags & VM_EXEC)
+		flush_icache_range(addr, addr + len);
+}
+
+asmlinkage int sys_cacheflush(int operation, void __user *addr, size_t len)
+{
+	int ret;
+
+	if (len > CACHEFLUSH_MAX_LEN) {
+		ret = -EPERM;
+		if (!capable(CAP_SYS_ADMIN))
+			goto out;
+	}
+
+	ret = -EFAULT;
+	if (!access_ok(VERIFY_WRITE, addr, len))
+		goto out;
+
+	switch (operation) {
+	case CACHE_IFLUSH:
+		flush_icache_range((unsigned long)addr,
+				   (unsigned long)addr + len);
+		ret = 0;
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+out:
+	return ret;
+}
diff --git a/arch/avr32/mm/clear_page.S b/arch/avr32/mm/clear_page.S
new file mode 100644
index 000000000000..5d70dca00699
--- /dev/null
+++ b/arch/avr32/mm/clear_page.S
@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/page.h>
+
+/*
+ * clear_page
+ * r12: P1 address (to)
+ */
+	.text
+	.global clear_page
+clear_page:
+	sub	r9, r12, -PAGE_SIZE
+	mov     r10, 0
+	mov	r11, 0
+0:      st.d    r12++, r10
+	cp      r12, r9
+	brne	0b
+	mov     pc, lr
diff --git a/arch/avr32/mm/copy_page.S b/arch/avr32/mm/copy_page.S
new file mode 100644
index 000000000000..c2b3752946b8
--- /dev/null
+++ b/arch/avr32/mm/copy_page.S
@@ -0,0 +1,28 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/page.h>
+
+/*
+ * copy_page
+ *
+ * r12		to (P1 address)
+ * r11		from (P1 address)
+ * r8-r10	scratch
+ */
+	.text
+	.global copy_page
+copy_page:
+	sub	r10, r11, -(1 << PAGE_SHIFT)
+	/* pref	r11[0] */
+1:	/* pref	r11[8] */
+	ld.d	r8, r11++
+	st.d	r12++, r8
+	cp	r11, r10
+	brlo	1b
+	mov	pc, lr
diff --git a/arch/avr32/mm/dma-coherent.c b/arch/avr32/mm/dma-coherent.c
new file mode 100644
index 000000000000..44ab8a7bdae2
--- /dev/null
+++ b/arch/avr32/mm/dma-coherent.c
@@ -0,0 +1,139 @@
+/*
+ *  Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/dma-mapping.h>
+
+#include <asm/addrspace.h>
+#include <asm/cacheflush.h>
+
+void dma_cache_sync(void *vaddr, size_t size, int direction)
+{
+	/*
+	 * No need to sync an uncached area
+	 */
+	if (PXSEG(vaddr) == P2SEG)
+		return;
+
+	switch (direction) {
+	case DMA_FROM_DEVICE:		/* invalidate only */
+		dma_cache_inv(vaddr, size);
+		break;
+	case DMA_TO_DEVICE:		/* writeback only */
+		dma_cache_wback(vaddr, size);
+		break;
+	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
+		dma_cache_wback_inv(vaddr, size);
+		break;
+	default:
+		BUG();
+	}
+}
+EXPORT_SYMBOL(dma_cache_sync);
+
+static struct page *__dma_alloc(struct device *dev, size_t size,
+				dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page, *free, *end;
+	int order;
+
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+
+	page = alloc_pages(gfp, order);
+	if (!page)
+		return NULL;
+	split_page(page, order);
+
+	/*
+	 * When accessing physical memory with valid cache data, we
+	 * get a cache hit even if the virtual memory region is marked
+	 * as uncached.
+	 *
+	 * Since the memory is newly allocated, there is no point in
+	 * doing a writeback. If the previous owner cares, he should
+	 * have flushed the cache before releasing the memory.
+	 */
+	invalidate_dcache_region(phys_to_virt(page_to_phys(page)), size);
+
+	*handle = page_to_bus(page);
+	free = page + (size >> PAGE_SHIFT);
+	end = page + (1 << order);
+
+	/*
+	 * Free any unused pages
+	 */
+	while (free < end) {
+		__free_page(free);
+		free++;
+	}
+
+	return page;
+}
+
+static void __dma_free(struct device *dev, size_t size,
+		       struct page *page, dma_addr_t handle)
+{
+	struct page *end = page + (PAGE_ALIGN(size) >> PAGE_SHIFT);
+
+	while (page < end)
+		__free_page(page++);
+}
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			 dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page;
+	void *ret = NULL;
+
+	page = __dma_alloc(dev, size, handle, gfp);
+	if (page)
+		ret = phys_to_uncached(page_to_phys(page));
+
+	return ret;
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
+
+void dma_free_coherent(struct device *dev, size_t size,
+		       void *cpu_addr, dma_addr_t handle)
+{
+	void *addr = phys_to_cached(uncached_to_phys(cpu_addr));
+	struct page *page;
+
+	pr_debug("dma_free_coherent addr %p (phys %08lx) size %u\n",
+		 cpu_addr, (unsigned long)handle, (unsigned)size);
+	BUG_ON(!virt_addr_valid(addr));
+	page = virt_to_page(addr);
+	__dma_free(dev, size, page, handle);
+}
+EXPORT_SYMBOL(dma_free_coherent);
+
+#if 0
+void *dma_alloc_writecombine(struct device *dev, size_t size,
+			     dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page;
+
+	page = __dma_alloc(dev, size, handle, gfp);
+
+	/* Now, map the page into P3 with write-combining turned on */
+	return __ioremap(page_to_phys(page), size, _PAGE_BUFFER);
+}
+EXPORT_SYMBOL(dma_alloc_writecombine);
+
+void dma_free_writecombine(struct device *dev, size_t size,
+			   void *cpu_addr, dma_addr_t handle)
+{
+	struct page *page;
+
+	iounmap(cpu_addr);
+
+	page = bus_to_page(handle);
+	__dma_free(dev, size, page, handle);
+}
+EXPORT_SYMBOL(dma_free_writecombine);
+#endif
diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c
new file mode 100644
index 000000000000..678557260a35
--- /dev/null
+++ b/arch/avr32/mm/fault.c
@@ -0,0 +1,315 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/sh/mm/fault.c:
+ *   Copyright (C) 1999  Niibe Yutaka
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/pagemap.h>
+
+#include <asm/kdebug.h>
+#include <asm/mmu_context.h>
+#include <asm/sysreg.h>
+#include <asm/uaccess.h>
+#include <asm/tlb.h>
+
+#ifdef DEBUG
+static void dump_code(unsigned long pc)
+{
+	char *p = (char *)pc;
+	char val;
+	int i;
+
+
+	printk(KERN_DEBUG "Code:");
+	for (i = 0; i < 16; i++) {
+		if (__get_user(val, p + i))
+			break;
+		printk(" %02x", val);
+	}
+	printk("\n");
+}
+#endif
+
+#ifdef CONFIG_KPROBES
+ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+
+/* Hook to register for page fault notifications */
+int register_page_fault_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
+}
+
+int unregister_page_fault_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
+}
+
+static inline int notify_page_fault(enum die_val val, struct pt_regs *regs,
+				    int trap, int sig)
+{
+	struct die_args args = {
+		.regs = regs,
+		.trapnr = trap,
+	};
+	return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
+}
+#else
+static inline int notify_page_fault(enum die_val val, struct pt_regs *regs,
+				    int trap, int sig)
+{
+	return NOTIFY_DONE;
+}
+#endif
+
+/*
+ * This routine handles page faults. It determines the address and the
+ * problem, and then passes it off to one of the appropriate routines.
+ *
+ * ecr is the Exception Cause Register. Possible values are:
+ *   5:  Page not found (instruction access)
+ *   6:  Protection fault (instruction access)
+ *   12: Page not found (read access)
+ *   13: Page not found (write access)
+ *   14: Protection fault (read access)
+ *   15: Protection fault (write access)
+ */
+asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs)
+{
+	struct task_struct *tsk;
+	struct mm_struct *mm;
+	struct vm_area_struct *vma;
+	const struct exception_table_entry *fixup;
+	unsigned long address;
+	unsigned long page;
+	int writeaccess = 0;
+
+	if (notify_page_fault(DIE_PAGE_FAULT, regs,
+			      ecr, SIGSEGV) == NOTIFY_STOP)
+		return;
+
+	address = sysreg_read(TLBEAR);
+
+	tsk = current;
+	mm = tsk->mm;
+
+	/*
+	 * If we're in an interrupt or have no user context, we must
+	 * not take the fault...
+	 */
+	if (in_atomic() || !mm || regs->sr & SYSREG_BIT(GM))
+		goto no_context;
+
+	local_irq_enable();
+
+	down_read(&mm->mmap_sem);
+
+	vma = find_vma(mm, address);
+	if (!vma)
+		goto bad_area;
+	if (vma->vm_start <= address)
+		goto good_area;
+	if (!(vma->vm_flags & VM_GROWSDOWN))
+		goto bad_area;
+	if (expand_stack(vma, address))
+		goto bad_area;
+
+	/*
+	 * Ok, we have a good vm_area for this memory access, so we
+	 * can handle it...
+	 */
+good_area:
+	//pr_debug("good area: vm_flags = 0x%lx\n", vma->vm_flags);
+	switch (ecr) {
+	case ECR_PROTECTION_X:
+	case ECR_TLB_MISS_X:
+		if (!(vma->vm_flags & VM_EXEC))
+			goto bad_area;
+		break;
+	case ECR_PROTECTION_R:
+	case ECR_TLB_MISS_R:
+		if (!(vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)))
+			goto bad_area;
+		break;
+	case ECR_PROTECTION_W:
+	case ECR_TLB_MISS_W:
+		if (!(vma->vm_flags & VM_WRITE))
+			goto bad_area;
+		writeaccess = 1;
+		break;
+	default:
+		panic("Unhandled case %lu in do_page_fault!", ecr);
+	}
+
+	/*
+	 * If for any reason at all we couldn't handle the fault, make
+	 * sure we exit gracefully rather than endlessly redo the
+	 * fault.
+	 */
+survive:
+	switch (handle_mm_fault(mm, vma, address, writeaccess)) {
+	case VM_FAULT_MINOR:
+		tsk->min_flt++;
+		break;
+	case VM_FAULT_MAJOR:
+		tsk->maj_flt++;
+		break;
+	case VM_FAULT_SIGBUS:
+		goto do_sigbus;
+	case VM_FAULT_OOM:
+		goto out_of_memory;
+	default:
+		BUG();
+	}
+
+	up_read(&mm->mmap_sem);
+	return;
+
+	/*
+	 * Something tried to access memory that isn't in our memory
+	 * map. Fix it, but check if it's kernel or user first...
+	 */
+bad_area:
+	pr_debug("Bad area [%s:%u]: addr %08lx, ecr %lu\n",
+		 tsk->comm, tsk->pid, address, ecr);
+
+	up_read(&mm->mmap_sem);
+
+	if (user_mode(regs)) {
+		/* Hmm...we have to pass address and ecr somehow... */
+		/* tsk->thread.address = address;
+		   tsk->thread.error_code = ecr; */
+#ifdef DEBUG
+		show_regs(regs);
+		dump_code(regs->pc);
+
+		page = sysreg_read(PTBR);
+		printk("ptbr = %08lx", page);
+		if (page) {
+			page = ((unsigned long *)page)[address >> 22];
+			printk(" pgd = %08lx", page);
+			if (page & _PAGE_PRESENT) {
+				page &= PAGE_MASK;
+				address &= 0x003ff000;
+				page = ((unsigned long *)__va(page))[address >> PAGE_SHIFT];
+				printk(" pte = %08lx\n", page);
+			}
+		}
+#endif
+		pr_debug("Sending SIGSEGV to PID %d...\n",
+			tsk->pid);
+		force_sig(SIGSEGV, tsk);
+		return;
+	}
+
+no_context:
+	pr_debug("No context\n");
+
+	/* Are we prepared to handle this kernel fault? */
+	fixup = search_exception_tables(regs->pc);
+	if (fixup) {
+		regs->pc = fixup->fixup;
+		pr_debug("Found fixup at %08lx\n", fixup->fixup);
+		return;
+	}
+
+	/*
+	 * Oops. The kernel tried to access some bad page. We'll have
+	 * to terminate things with extreme prejudice.
+	 */
+	if (address < PAGE_SIZE)
+		printk(KERN_ALERT
+		       "Unable to handle kernel NULL pointer dereference");
+	else
+		printk(KERN_ALERT
+		       "Unable to handle kernel paging request");
+	printk(" at virtual address %08lx\n", address);
+	printk(KERN_ALERT "pc = %08lx\n", regs->pc);
+
+	page = sysreg_read(PTBR);
+	printk(KERN_ALERT "ptbr = %08lx", page);
+	if (page) {
+		page = ((unsigned long *)page)[address >> 22];
+		printk(" pgd = %08lx", page);
+		if (page & _PAGE_PRESENT) {
+			page &= PAGE_MASK;
+			address &= 0x003ff000;
+			page = ((unsigned long *)__va(page))[address >> PAGE_SHIFT];
+			printk(" pte = %08lx\n", page);
+		}
+	}
+	die("\nOops", regs, ecr);
+	do_exit(SIGKILL);
+
+	/*
+	 * We ran out of memory, or some other thing happened to us
+	 * that made us unable to handle the page fault gracefully.
+	 */
+out_of_memory:
+	printk("Out of memory\n");
+	up_read(&mm->mmap_sem);
+	if (current->pid == 1) {
+		yield();
+		down_read(&mm->mmap_sem);
+		goto survive;
+	}
+	printk("VM: Killing process %s\n", tsk->comm);
+	if (user_mode(regs))
+		do_exit(SIGKILL);
+	goto no_context;
+
+do_sigbus:
+	up_read(&mm->mmap_sem);
+
+	/*
+	 * Send a sigbus, regardless of whether we were in kernel or
+	 * user mode.
+	 */
+	/* address, error_code, trap_no, ... */
+#ifdef DEBUG
+	show_regs(regs);
+	dump_code(regs->pc);
+#endif
+	pr_debug("Sending SIGBUS to PID %d...\n", tsk->pid);
+	force_sig(SIGBUS, tsk);
+
+	/* Kernel mode? Handle exceptions or die */
+	if (!user_mode(regs))
+		goto no_context;
+}
+
+asmlinkage void do_bus_error(unsigned long addr, int write_access,
+			     struct pt_regs *regs)
+{
+	printk(KERN_ALERT
+	       "Bus error at physical address 0x%08lx (%s access)\n",
+	       addr, write_access ? "write" : "read");
+	printk(KERN_INFO "DTLB dump:\n");
+	dump_dtlb();
+	die("Bus Error", regs, write_access);
+	do_exit(SIGKILL);
+}
+
+/*
+ * This functionality is currently not possible to implement because
+ * we're using segmentation to ensure a fixed mapping of the kernel
+ * virtual address space.
+ *
+ * It would be possible to implement this, but it would require us to
+ * disable segmentation at startup and load the kernel mappings into
+ * the TLB like any other pages. There will be lots of trickery to
+ * avoid recursive invocation of the TLB miss handler, though...
+ */
+#ifdef CONFIG_DEBUG_PAGEALLOC
+void kernel_map_pages(struct page *page, int numpages, int enable)
+{
+
+}
+EXPORT_SYMBOL(kernel_map_pages);
+#endif
diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c
new file mode 100644
index 000000000000..3e6c41039808
--- /dev/null
+++ b/arch/avr32/mm/init.c
@@ -0,0 +1,480 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/init.h>
+#include <linux/initrd.h>
+#include <linux/mmzone.h>
+#include <linux/bootmem.h>
+#include <linux/pagemap.h>
+#include <linux/pfn.h>
+#include <linux/nodemask.h>
+
+#include <asm/page.h>
+#include <asm/mmu_context.h>
+#include <asm/tlb.h>
+#include <asm/io.h>
+#include <asm/dma.h>
+#include <asm/setup.h>
+#include <asm/sections.h>
+
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+
+pgd_t swapper_pg_dir[PTRS_PER_PGD];
+
+struct page *empty_zero_page;
+
+/*
+ * Cache of MMU context last used.
+ */
+unsigned long mmu_context_cache = NO_CONTEXT;
+
+#define START_PFN	(NODE_DATA(0)->bdata->node_boot_start >> PAGE_SHIFT)
+#define MAX_LOW_PFN	(NODE_DATA(0)->bdata->node_low_pfn)
+
+void show_mem(void)
+{
+	int total = 0, reserved = 0, cached = 0;
+	int slab = 0, free = 0, shared = 0;
+	pg_data_t *pgdat;
+
+	printk("Mem-info:\n");
+	show_free_areas();
+
+	for_each_online_pgdat(pgdat) {
+		struct page *page, *end;
+
+		page = pgdat->node_mem_map;
+		end = page + pgdat->node_spanned_pages;
+
+		do {
+			total++;
+			if (PageReserved(page))
+				reserved++;
+			else if (PageSwapCache(page))
+				cached++;
+			else if (PageSlab(page))
+				slab++;
+			else if (!page_count(page))
+				free++;
+			else
+				shared += page_count(page) - 1;
+			page++;
+		} while (page < end);
+	}
+
+	printk ("%d pages of RAM\n", total);
+	printk ("%d free pages\n", free);
+	printk ("%d reserved pages\n", reserved);
+	printk ("%d slab pages\n", slab);
+	printk ("%d pages shared\n", shared);
+	printk ("%d pages swap cached\n", cached);
+}
+
+static void __init print_memory_map(const char *what,
+				    struct tag_mem_range *mem)
+{
+	printk ("%s:\n", what);
+	for (; mem; mem = mem->next) {
+		printk ("  %08lx - %08lx\n",
+			(unsigned long)mem->addr,
+			(unsigned long)(mem->addr + mem->size));
+	}
+}
+
+#define MAX_LOWMEM	HIGHMEM_START
+#define MAX_LOWMEM_PFN	PFN_DOWN(MAX_LOWMEM)
+
+/*
+ * Sort a list of memory regions in-place by ascending address.
+ *
+ * We're using bubble sort because we only have singly linked lists
+ * with few elements.
+ */
+static void __init sort_mem_list(struct tag_mem_range **pmem)
+{
+	int done;
+	struct tag_mem_range **a, **b;
+
+	if (!*pmem)
+		return;
+
+	do {
+		done = 1;
+		a = pmem, b = &(*pmem)->next;
+		while (*b) {
+			if ((*a)->addr > (*b)->addr) {
+				struct tag_mem_range *tmp;
+				tmp = (*b)->next;
+				(*b)->next = *a;
+				*a = *b;
+				*b = tmp;
+				done = 0;
+			}
+			a = &(*a)->next;
+			b = &(*a)->next;
+		}
+	} while (!done);
+}
+
+/*
+ * Find a free memory region large enough for storing the
+ * bootmem bitmap.
+ */
+static unsigned long __init
+find_bootmap_pfn(const struct tag_mem_range *mem)
+{
+	unsigned long bootmap_pages, bootmap_len;
+	unsigned long node_pages = PFN_UP(mem->size);
+	unsigned long bootmap_addr = mem->addr;
+	struct tag_mem_range *reserved = mem_reserved;
+	struct tag_mem_range *ramdisk = mem_ramdisk;
+	unsigned long kern_start = virt_to_phys(_stext);
+	unsigned long kern_end = virt_to_phys(_end);
+
+	bootmap_pages = bootmem_bootmap_pages(node_pages);
+	bootmap_len = bootmap_pages << PAGE_SHIFT;
+
+	/*
+	 * Find a large enough region without reserved pages for
+	 * storing the bootmem bitmap. We can take advantage of the
+	 * fact that all lists have been sorted.
+	 *
+	 * We have to check explicitly reserved regions as well as the
+	 * kernel image and any RAMDISK images...
+	 *
+	 * Oh, and we have to make sure we don't overwrite the taglist
+	 * since we're going to use it until the bootmem allocator is
+	 * fully up and running.
+	 */
+	while (1) {
+		if ((bootmap_addr < kern_end) &&
+		    ((bootmap_addr + bootmap_len) > kern_start))
+			bootmap_addr = kern_end;
+
+		while (reserved &&
+		       (bootmap_addr >= (reserved->addr + reserved->size)))
+			reserved = reserved->next;
+
+		if (reserved &&
+		    ((bootmap_addr + bootmap_len) >= reserved->addr)) {
+			bootmap_addr = reserved->addr + reserved->size;
+			continue;
+		}
+
+		while (ramdisk &&
+		       (bootmap_addr >= (ramdisk->addr + ramdisk->size)))
+			ramdisk = ramdisk->next;
+
+		if (!ramdisk ||
+		    ((bootmap_addr + bootmap_len) < ramdisk->addr))
+			break;
+
+		bootmap_addr = ramdisk->addr + ramdisk->size;
+	}
+
+	if ((PFN_UP(bootmap_addr) + bootmap_len) >= (mem->addr + mem->size))
+		return ~0UL;
+
+	return PFN_UP(bootmap_addr);
+}
+
+void __init setup_bootmem(void)
+{
+	unsigned bootmap_size;
+	unsigned long first_pfn, bootmap_pfn, pages;
+	unsigned long max_pfn, max_low_pfn;
+	unsigned long kern_start = virt_to_phys(_stext);
+	unsigned long kern_end = virt_to_phys(_end);
+	unsigned node = 0;
+	struct tag_mem_range *bank, *res;
+
+	sort_mem_list(&mem_phys);
+	sort_mem_list(&mem_reserved);
+
+	print_memory_map("Physical memory", mem_phys);
+	print_memory_map("Reserved memory", mem_reserved);
+
+	nodes_clear(node_online_map);
+
+	if (mem_ramdisk) {
+#ifdef CONFIG_BLK_DEV_INITRD
+		initrd_start = __va(mem_ramdisk->addr);
+		initrd_end = initrd_start + mem_ramdisk->size;
+
+		print_memory_map("RAMDISK images", mem_ramdisk);
+		if (mem_ramdisk->next)
+			printk(KERN_WARNING
+			       "Warning: Only the first RAMDISK image "
+			       "will be used\n");
+		sort_mem_list(&mem_ramdisk);
+#else
+		printk(KERN_WARNING "RAM disk image present, but "
+		       "no initrd support in kernel!\n");
+#endif
+	}
+
+	if (mem_phys->next)
+		printk(KERN_WARNING "Only using first memory bank\n");
+
+	for (bank = mem_phys; bank; bank = NULL) {
+		first_pfn = PFN_UP(bank->addr);
+		max_low_pfn = max_pfn = PFN_DOWN(bank->addr + bank->size);
+		bootmap_pfn = find_bootmap_pfn(bank);
+		if (bootmap_pfn > max_pfn)
+			panic("No space for bootmem bitmap!\n");
+
+		if (max_low_pfn > MAX_LOWMEM_PFN) {
+			max_low_pfn = MAX_LOWMEM_PFN;
+#ifndef CONFIG_HIGHMEM
+			/*
+			 * Lowmem is memory that can be addressed
+			 * directly through P1/P2
+			 */
+			printk(KERN_WARNING
+			       "Node %u: Only %ld MiB of memory will be used.\n",
+			       node, MAX_LOWMEM >> 20);
+			printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
+#else
+#error HIGHMEM is not supported by AVR32 yet
+#endif
+		}
+
+		/* Initialize the boot-time allocator with low memory only. */
+		bootmap_size = init_bootmem_node(NODE_DATA(node), bootmap_pfn,
+						 first_pfn, max_low_pfn);
+
+		printk("Node %u: bdata = %p, bdata->node_bootmem_map = %p\n",
+		       node, NODE_DATA(node)->bdata,
+		       NODE_DATA(node)->bdata->node_bootmem_map);
+
+		/*
+		 * Register fully available RAM pages with the bootmem
+		 * allocator.
+		 */
+		pages = max_low_pfn - first_pfn;
+		free_bootmem_node (NODE_DATA(node), PFN_PHYS(first_pfn),
+				   PFN_PHYS(pages));
+
+		/*
+		 * Reserve space for the kernel image (if present in
+		 * this node)...
+		 */
+		if ((kern_start >= PFN_PHYS(first_pfn)) &&
+		    (kern_start < PFN_PHYS(max_pfn))) {
+			printk("Node %u: Kernel image %08lx - %08lx\n",
+			       node, kern_start, kern_end);
+			reserve_bootmem_node(NODE_DATA(node), kern_start,
+					     kern_end - kern_start);
+		}
+
+		/* ...the bootmem bitmap... */
+		reserve_bootmem_node(NODE_DATA(node),
+				     PFN_PHYS(bootmap_pfn),
+				     bootmap_size);
+
+		/* ...any RAMDISK images... */
+		for (res = mem_ramdisk; res; res = res->next) {
+			if (res->addr > PFN_PHYS(max_pfn))
+				break;
+
+			if (res->addr >= PFN_PHYS(first_pfn)) {
+				printk("Node %u: RAMDISK %08lx - %08lx\n",
+				       node,
+				       (unsigned long)res->addr,
+				       (unsigned long)(res->addr + res->size));
+				reserve_bootmem_node(NODE_DATA(node),
+						     res->addr, res->size);
+			}
+		}
+
+		/* ...and any other reserved regions. */
+		for (res = mem_reserved; res; res = res->next) {
+			if (res->addr > PFN_PHYS(max_pfn))
+				break;
+
+			if (res->addr >= PFN_PHYS(first_pfn)) {
+				printk("Node %u: Reserved %08lx - %08lx\n",
+				       node,
+				       (unsigned long)res->addr,
+				       (unsigned long)(res->addr + res->size));
+				reserve_bootmem_node(NODE_DATA(node),
+						     res->addr, res->size);
+			}
+		}
+
+		node_set_online(node);
+	}
+}
+
+/*
+ * paging_init() sets up the page tables
+ *
+ * This routine also unmaps the page at virtual kernel address 0, so
+ * that we can trap those pesky NULL-reference errors in the kernel.
+ */
+void __init paging_init(void)
+{
+	extern unsigned long _evba;
+	void *zero_page;
+	int nid;
+
+	/*
+	 * Make sure we can handle exceptions before enabling
+	 * paging. Not that we should ever _get_ any exceptions this
+	 * early, but you never know...
+	 */
+	printk("Exception vectors start at %p\n", &_evba);
+	sysreg_write(EVBA, (unsigned long)&_evba);
+
+	/*
+	 * Since we are ready to handle exceptions now, we should let
+	 * the CPU generate them...
+	 */
+	__asm__ __volatile__ ("csrf %0" : : "i"(SR_EM_BIT));
+
+	/*
+	 * Allocate the zero page. The allocator will panic if it
+	 * can't satisfy the request, so no need to check.
+	 */
+	zero_page = alloc_bootmem_low_pages_node(NODE_DATA(0),
+						 PAGE_SIZE);
+
+	{
+		pgd_t *pg_dir;
+		int i;
+
+		pg_dir = swapper_pg_dir;
+		sysreg_write(PTBR, (unsigned long)pg_dir);
+
+		for (i = 0; i < PTRS_PER_PGD; i++)
+			pgd_val(pg_dir[i]) = 0;
+
+		enable_mmu();
+		printk ("CPU: Paging enabled\n");
+	}
+
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		unsigned long zones_size[MAX_NR_ZONES];
+		unsigned long low, start_pfn;
+
+		start_pfn = pgdat->bdata->node_boot_start;
+		start_pfn >>= PAGE_SHIFT;
+		low = pgdat->bdata->node_low_pfn;
+
+		memset(zones_size, 0, sizeof(zones_size));
+		zones_size[ZONE_NORMAL] = low - start_pfn;
+
+		printk("Node %u: start_pfn = 0x%lx, low = 0x%lx\n",
+		       nid, start_pfn, low);
+
+		free_area_init_node(nid, pgdat, zones_size, start_pfn, NULL);
+
+		printk("Node %u: mem_map starts at %p\n",
+		       pgdat->node_id, pgdat->node_mem_map);
+	}
+
+	mem_map = NODE_DATA(0)->node_mem_map;
+
+	memset(zero_page, 0, PAGE_SIZE);
+	empty_zero_page = virt_to_page(zero_page);
+	flush_dcache_page(empty_zero_page);
+}
+
+void __init mem_init(void)
+{
+	int codesize, reservedpages, datasize, initsize;
+	int nid, i;
+
+	reservedpages = 0;
+	high_memory = NULL;
+
+	/* this will put all low memory onto the freelists */
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		unsigned long node_pages = 0;
+		void *node_high_memory;
+
+		num_physpages += pgdat->node_present_pages;
+
+		if (pgdat->node_spanned_pages != 0)
+			node_pages = free_all_bootmem_node(pgdat);
+
+		totalram_pages += node_pages;
+
+		for (i = 0; i < node_pages; i++)
+			if (PageReserved(pgdat->node_mem_map + i))
+				reservedpages++;
+
+		node_high_memory = (void *)((pgdat->node_start_pfn
+					     + pgdat->node_spanned_pages)
+					    << PAGE_SHIFT);
+		if (node_high_memory > high_memory)
+			high_memory = node_high_memory;
+	}
+
+	max_mapnr = MAP_NR(high_memory);
+
+	codesize = (unsigned long)_etext - (unsigned long)_text;
+	datasize = (unsigned long)_edata - (unsigned long)_data;
+	initsize = (unsigned long)__init_end - (unsigned long)__init_begin;
+
+	printk ("Memory: %luk/%luk available (%dk kernel code, "
+		"%dk reserved, %dk data, %dk init)\n",
+		(unsigned long)nr_free_pages() << (PAGE_SHIFT - 10),
+		totalram_pages << (PAGE_SHIFT - 10),
+		codesize >> 10,
+		reservedpages << (PAGE_SHIFT - 10),
+		datasize >> 10,
+		initsize >> 10);
+}
+
+static inline void free_area(unsigned long addr, unsigned long end, char *s)
+{
+	unsigned int size = (end - addr) >> 10;
+
+	for (; addr < end; addr += PAGE_SIZE) {
+		struct page *page = virt_to_page(addr);
+		ClearPageReserved(page);
+		init_page_count(page);
+		free_page(addr);
+		totalram_pages++;
+	}
+
+	if (size && s)
+		printk(KERN_INFO "Freeing %s memory: %dK (%lx - %lx)\n",
+		       s, size, end - (size << 10), end);
+}
+
+void free_initmem(void)
+{
+	free_area((unsigned long)__init_begin, (unsigned long)__init_end,
+		  "init");
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+
+static int keep_initrd;
+
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+	if (!keep_initrd)
+		free_area(start, end, "initrd");
+}
+
+static int __init keepinitrd_setup(char *__unused)
+{
+	keep_initrd = 1;
+	return 1;
+}
+
+__setup("keepinitrd", keepinitrd_setup);
+#endif
diff --git a/arch/avr32/mm/ioremap.c b/arch/avr32/mm/ioremap.c
new file mode 100644
index 000000000000..536021877df6
--- /dev/null
+++ b/arch/avr32/mm/ioremap.c
@@ -0,0 +1,197 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/vmalloc.h>
+#include <linux/module.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+#include <asm/addrspace.h>
+
+static inline int remap_area_pte(pte_t *pte, unsigned long address,
+				  unsigned long end, unsigned long phys_addr,
+				  pgprot_t prot)
+{
+	unsigned long pfn;
+
+	pfn = phys_addr >> PAGE_SHIFT;
+	do {
+		WARN_ON(!pte_none(*pte));
+
+		set_pte(pte, pfn_pte(pfn, prot));
+		address += PAGE_SIZE;
+		pfn++;
+		pte++;
+	} while (address && (address < end));
+
+	return 0;
+}
+
+static inline int remap_area_pmd(pmd_t *pmd, unsigned long address,
+				 unsigned long end, unsigned long phys_addr,
+				 pgprot_t prot)
+{
+	unsigned long next;
+
+	phys_addr -= address;
+
+	do {
+		pte_t *pte = pte_alloc_kernel(pmd, address);
+		if (!pte)
+			return -ENOMEM;
+
+		next = (address + PMD_SIZE) & PMD_MASK;
+		if (remap_area_pte(pte, address, next,
+				   address + phys_addr, prot))
+			return -ENOMEM;
+
+		address = next;
+		pmd++;
+	} while (address && (address < end));
+	return 0;
+}
+
+static int remap_area_pud(pud_t *pud, unsigned long address,
+			  unsigned long end, unsigned long phys_addr,
+			  pgprot_t prot)
+{
+	unsigned long next;
+
+	phys_addr -= address;
+
+	do {
+		pmd_t *pmd = pmd_alloc(&init_mm, pud, address);
+		if (!pmd)
+			return -ENOMEM;
+		next = (address + PUD_SIZE) & PUD_MASK;
+		if (remap_area_pmd(pmd, address, next,
+				   phys_addr + address, prot))
+			return -ENOMEM;
+
+		address = next;
+		pud++;
+	} while (address && address < end);
+
+	return 0;
+}
+
+static int remap_area_pages(unsigned long address, unsigned long phys_addr,
+			    size_t size, pgprot_t prot)
+{
+	unsigned long end = address + size;
+	unsigned long next;
+	pgd_t *pgd;
+	int err = 0;
+
+	phys_addr -= address;
+
+	pgd = pgd_offset_k(address);
+	flush_cache_all();
+	BUG_ON(address >= end);
+
+	spin_lock(&init_mm.page_table_lock);
+	do {
+		pud_t *pud = pud_alloc(&init_mm, pgd, address);
+
+		err = -ENOMEM;
+		if (!pud)
+			break;
+
+		next = (address + PGDIR_SIZE) & PGDIR_MASK;
+		if (next < address || next > end)
+			next = end;
+		err = remap_area_pud(pud, address, next,
+				     phys_addr + address, prot);
+		if (err)
+			break;
+
+		address = next;
+		pgd++;
+	} while (address && (address < end));
+
+	spin_unlock(&init_mm.page_table_lock);
+	flush_tlb_all();
+	return err;
+}
+
+/*
+ * Re-map an arbitrary physical address space into the kernel virtual
+ * address space. Needed when the kernel wants to access physical
+ * memory directly.
+ */
+void __iomem *__ioremap(unsigned long phys_addr, size_t size,
+			unsigned long flags)
+{
+	void *addr;
+	struct vm_struct *area;
+	unsigned long offset, last_addr;
+	pgprot_t prot;
+
+	/*
+	 * Check if we can simply use the P4 segment. This area is
+	 * uncacheable, so if caching/buffering is requested, we can't
+	 * use it.
+	 */
+	if ((phys_addr >= P4SEG) && (flags == 0))
+		return (void __iomem *)phys_addr;
+
+	/* Don't allow wraparound or zero size */
+	last_addr = phys_addr + size - 1;
+	if (!size || last_addr < phys_addr)
+		return NULL;
+
+	/*
+	 * XXX: When mapping regular RAM, we'd better make damn sure
+	 * it's never used for anything else.  But this is really the
+	 * caller's responsibility...
+	 */
+	if (PHYSADDR(P2SEGADDR(phys_addr)) == phys_addr)
+		return (void __iomem *)P2SEGADDR(phys_addr);
+
+	/* Mappings have to be page-aligned */
+	offset = phys_addr & ~PAGE_MASK;
+	phys_addr &= PAGE_MASK;
+	size = PAGE_ALIGN(last_addr + 1) - phys_addr;
+
+	prot = __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY
+			| _PAGE_ACCESSED | _PAGE_TYPE_SMALL | flags);
+
+	/*
+	 * Ok, go for it..
+	 */
+	area = get_vm_area(size, VM_IOREMAP);
+	if (!area)
+		return NULL;
+	area->phys_addr = phys_addr;
+	addr = area->addr;
+	if (remap_area_pages((unsigned long)addr, phys_addr, size, prot)) {
+		vunmap(addr);
+		return NULL;
+	}
+
+	return (void __iomem *)(offset + (char *)addr);
+}
+EXPORT_SYMBOL(__ioremap);
+
+void __iounmap(void __iomem *addr)
+{
+	struct vm_struct *p;
+
+	if ((unsigned long)addr >= P4SEG)
+		return;
+
+	p = remove_vm_area((void *)(PAGE_MASK & (unsigned long __force)addr));
+	if (unlikely(!p)) {
+		printk (KERN_ERR "iounmap: bad address %p\n", addr);
+		return;
+	}
+
+	kfree (p);
+}
+EXPORT_SYMBOL(__iounmap);
diff --git a/arch/avr32/mm/tlb.c b/arch/avr32/mm/tlb.c
new file mode 100644
index 000000000000..5d0523bbe298
--- /dev/null
+++ b/arch/avr32/mm/tlb.c
@@ -0,0 +1,378 @@
+/*
+ * AVR32 TLB operations
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/mm.h>
+
+#include <asm/mmu_context.h>
+
+#define _TLBEHI_I	0x100
+
+void show_dtlb_entry(unsigned int index)
+{
+	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save, flags;
+
+	local_irq_save(flags);
+	mmucr_save = sysreg_read(MMUCR);
+	tlbehi_save = sysreg_read(TLBEHI);
+	mmucr = mmucr_save & 0x13;
+	mmucr |= index << 14;
+	sysreg_write(MMUCR, mmucr);
+
+	asm volatile("tlbr" : : : "memory");
+	cpu_sync_pipeline();
+
+	tlbehi = sysreg_read(TLBEHI);
+	tlbelo = sysreg_read(TLBELO);
+
+	printk("%2u: %c %c %02x   %05x %05x %o  %o  %c %c %c %c\n",
+	       index,
+	       (tlbehi & 0x200)?'1':'0',
+	       (tlbelo & 0x100)?'1':'0',
+	       (tlbehi & 0xff),
+	       (tlbehi >> 12), (tlbelo >> 12),
+	       (tlbelo >> 4) & 7, (tlbelo >> 2) & 3,
+	       (tlbelo & 0x200)?'1':'0',
+	       (tlbelo & 0x080)?'1':'0',
+	       (tlbelo & 0x001)?'1':'0',
+	       (tlbelo & 0x002)?'1':'0');
+
+	sysreg_write(MMUCR, mmucr_save);
+	sysreg_write(TLBEHI, tlbehi_save);
+	cpu_sync_pipeline();
+	local_irq_restore(flags);
+}
+
+void dump_dtlb(void)
+{
+	unsigned int i;
+
+	printk("ID  V G ASID VPN   PFN   AP SZ C B W D\n");
+	for (i = 0; i < 32; i++)
+		show_dtlb_entry(i);
+}
+
+static unsigned long last_mmucr;
+
+static inline void set_replacement_pointer(unsigned shift)
+{
+	unsigned long mmucr, mmucr_save;
+
+	mmucr = mmucr_save = sysreg_read(MMUCR);
+
+	/* Does this mapping already exist? */
+	__asm__ __volatile__(
+		"	tlbs\n"
+		"	mfsr %0, %1"
+		: "=r"(mmucr)
+		: "i"(SYSREG_MMUCR));
+
+	if (mmucr & SYSREG_BIT(MMUCR_N)) {
+		/* Not found -- pick a not-recently-accessed entry */
+		unsigned long rp;
+		unsigned long tlbar = sysreg_read(TLBARLO);
+
+		rp = 32 - fls(tlbar);
+		if (rp == 32) {
+			rp = 0;
+			sysreg_write(TLBARLO, -1L);
+		}
+
+		mmucr &= 0x13;
+		mmucr |= (rp << shift);
+
+		sysreg_write(MMUCR, mmucr);
+	}
+
+	last_mmucr = mmucr;
+}
+
+static void update_dtlb(unsigned long address, pte_t pte, unsigned long asid)
+{
+	unsigned long vpn;
+
+	vpn = (address & MMU_VPN_MASK) | _TLBEHI_VALID | asid;
+	sysreg_write(TLBEHI, vpn);
+	cpu_sync_pipeline();
+
+	set_replacement_pointer(14);
+
+	sysreg_write(TLBELO, pte_val(pte) & _PAGE_FLAGS_HARDWARE_MASK);
+
+	/* Let's go */
+	asm volatile("nop\n\ttlbw" : : : "memory");
+	cpu_sync_pipeline();
+}
+
+void update_mmu_cache(struct vm_area_struct *vma,
+		      unsigned long address, pte_t pte)
+{
+	unsigned long flags;
+
+	/* ptrace may call this routine */
+	if (vma && current->active_mm != vma->vm_mm)
+		return;
+
+	local_irq_save(flags);
+	update_dtlb(address, pte, get_asid());
+	local_irq_restore(flags);
+}
+
+void __flush_tlb_page(unsigned long asid, unsigned long page)
+{
+	unsigned long mmucr, tlbehi;
+
+	page |= asid;
+	sysreg_write(TLBEHI, page);
+	cpu_sync_pipeline();
+	asm volatile("tlbs");
+	mmucr = sysreg_read(MMUCR);
+
+	if (!(mmucr & SYSREG_BIT(MMUCR_N))) {
+		unsigned long tlbarlo;
+		unsigned long entry;
+
+		/* Clear the "valid" bit */
+		tlbehi = sysreg_read(TLBEHI);
+		tlbehi &= ~_TLBEHI_VALID;
+		sysreg_write(TLBEHI, tlbehi);
+		cpu_sync_pipeline();
+
+		/* mark the entry as "not accessed" */
+		entry = (mmucr >> 14) & 0x3f;
+		tlbarlo = sysreg_read(TLBARLO);
+		tlbarlo |= (0x80000000 >> entry);
+		sysreg_write(TLBARLO, tlbarlo);
+
+		/* update the entry with valid bit clear */
+		asm volatile("tlbw");
+		cpu_sync_pipeline();
+	}
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
+{
+	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
+		unsigned long flags, asid;
+		unsigned long saved_asid = MMU_NO_ASID;
+
+		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
+		page &= PAGE_MASK;
+
+		local_irq_save(flags);
+		if (vma->vm_mm != current->mm) {
+			saved_asid = get_asid();
+			set_asid(asid);
+		}
+
+		__flush_tlb_page(asid, page);
+
+		if (saved_asid != MMU_NO_ASID)
+			set_asid(saved_asid);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+		     unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+		int size;
+
+		local_irq_save(flags);
+		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+		if (size > (MMU_DTLB_ENTRIES / 4)) { /* Too many entries to flush */
+			mm->context = NO_CONTEXT;
+			if (mm == current->mm)
+				activate_context(mm);
+		} else {
+			unsigned long asid = mm->context & MMU_CONTEXT_ASID_MASK;
+			unsigned long saved_asid = MMU_NO_ASID;
+
+			start &= PAGE_MASK;
+			end += (PAGE_SIZE - 1);
+			end &= PAGE_MASK;
+			if (mm != current->mm) {
+				saved_asid = get_asid();
+				set_asid(asid);
+			}
+
+			while (start < end) {
+				__flush_tlb_page(asid, start);
+				start += PAGE_SIZE;
+			}
+			if (saved_asid != MMU_NO_ASID)
+				set_asid(saved_asid);
+		}
+		local_irq_restore(flags);
+	}
+}
+
+/*
+ * TODO: If this is only called for addresses > TASK_SIZE, we can probably
+ * skip the ASID stuff and just use the Global bit...
+ */
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	unsigned long flags;
+	int size;
+
+	local_irq_save(flags);
+	size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	if (size > (MMU_DTLB_ENTRIES / 4)) { /* Too many entries to flush */
+		flush_tlb_all();
+	} else {
+		unsigned long asid = init_mm.context & MMU_CONTEXT_ASID_MASK;
+		unsigned long saved_asid = get_asid();
+
+		start &= PAGE_MASK;
+		end += (PAGE_SIZE - 1);
+		end &= PAGE_MASK;
+		set_asid(asid);
+		while (start < end) {
+			__flush_tlb_page(asid, start);
+			start += PAGE_SIZE;
+		}
+		set_asid(saved_asid);
+	}
+	local_irq_restore(flags);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	/* Invalidate all TLB entries of this process by getting a new ASID */
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+
+		local_irq_save(flags);
+		mm->context = NO_CONTEXT;
+		if (mm == current->mm)
+			activate_context(mm);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_all(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	sysreg_write(MMUCR, sysreg_read(MMUCR) | SYSREG_BIT(MMUCR_I));
+	local_irq_restore(flags);
+}
+
+#ifdef CONFIG_PROC_FS
+
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+
+static void *tlb_start(struct seq_file *tlb, loff_t *pos)
+{
+	static unsigned long tlb_index;
+
+	if (*pos >= 32)
+		return NULL;
+
+	tlb_index = 0;
+	return &tlb_index;
+}
+
+static void *tlb_next(struct seq_file *tlb, void *v, loff_t *pos)
+{
+	unsigned long *index = v;
+
+	if (*index >= 31)
+		return NULL;
+
+	++*pos;
+	++*index;
+	return index;
+}
+
+static void tlb_stop(struct seq_file *tlb, void *v)
+{
+
+}
+
+static int tlb_show(struct seq_file *tlb, void *v)
+{
+	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save, flags;
+	unsigned long *index = v;
+
+	if (*index == 0)
+		seq_puts(tlb, "ID  V G ASID VPN   PFN   AP SZ C B W D\n");
+
+	BUG_ON(*index >= 32);
+
+	local_irq_save(flags);
+	mmucr_save = sysreg_read(MMUCR);
+	tlbehi_save = sysreg_read(TLBEHI);
+	mmucr = mmucr_save & 0x13;
+	mmucr |= *index << 14;
+	sysreg_write(MMUCR, mmucr);
+
+	asm volatile("tlbr" : : : "memory");
+	cpu_sync_pipeline();
+
+	tlbehi = sysreg_read(TLBEHI);
+	tlbelo = sysreg_read(TLBELO);
+
+	sysreg_write(MMUCR, mmucr_save);
+	sysreg_write(TLBEHI, tlbehi_save);
+	cpu_sync_pipeline();
+	local_irq_restore(flags);
+
+	seq_printf(tlb, "%2lu: %c %c %02x   %05x %05x %o  %o  %c %c %c %c\n",
+	       *index,
+	       (tlbehi & 0x200)?'1':'0',
+	       (tlbelo & 0x100)?'1':'0',
+	       (tlbehi & 0xff),
+	       (tlbehi >> 12), (tlbelo >> 12),
+	       (tlbelo >> 4) & 7, (tlbelo >> 2) & 3,
+	       (tlbelo & 0x200)?'1':'0',
+	       (tlbelo & 0x080)?'1':'0',
+	       (tlbelo & 0x001)?'1':'0',
+	       (tlbelo & 0x002)?'1':'0');
+
+	return 0;
+}
+
+static struct seq_operations tlb_ops = {
+	.start		= tlb_start,
+	.next		= tlb_next,
+	.stop		= tlb_stop,
+	.show		= tlb_show,
+};
+
+static int tlb_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &tlb_ops);
+}
+
+static struct file_operations proc_tlb_operations = {
+	.open		= tlb_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int __init proctlb_init(void)
+{
+	struct proc_dir_entry *entry;
+
+	entry = create_proc_entry("tlb", 0, NULL);
+	if (entry)
+		entry->proc_fops = &proc_tlb_operations;
+	return 0;
+}
+late_initcall(proctlb_init);
+#endif /* CONFIG_PROC_FS */
diff --git a/include/asm-avr32/Kbuild b/include/asm-avr32/Kbuild
new file mode 100644
index 000000000000..8770e73ce938
--- /dev/null
+++ b/include/asm-avr32/Kbuild
@@ -0,0 +1,3 @@
+include include/asm-generic/Kbuild.asm
+
+headers-y	+= cachectl.h
diff --git a/include/asm-avr32/a.out.h b/include/asm-avr32/a.out.h
new file mode 100644
index 000000000000..50bf6e31a143
--- /dev/null
+++ b/include/asm-avr32/a.out.h
@@ -0,0 +1,26 @@
+#ifndef __ASM_AVR32_A_OUT_H
+#define __ASM_AVR32_A_OUT_H
+
+struct exec
+{
+  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
+  unsigned a_text;		/* length of text, in bytes */
+  unsigned a_data;		/* length of data, in bytes */
+  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
+  unsigned a_syms;		/* length of symbol table data in file, in bytes */
+  unsigned a_entry;		/* start address */
+  unsigned a_trsize;		/* length of relocation info for text, in bytes */
+  unsigned a_drsize;		/* length of relocation info for data, in bytes */
+};
+
+#define N_TRSIZE(a)	((a).a_trsize)
+#define N_DRSIZE(a)	((a).a_drsize)
+#define N_SYMSIZE(a)	((a).a_syms)
+
+#ifdef __KERNEL__
+
+#define STACK_TOP	TASK_SIZE
+
+#endif
+
+#endif /* __ASM_AVR32_A_OUT_H */
diff --git a/include/asm-avr32/addrspace.h b/include/asm-avr32/addrspace.h
new file mode 100644
index 000000000000..366794858ec7
--- /dev/null
+++ b/include/asm-avr32/addrspace.h
@@ -0,0 +1,43 @@
+/*
+ * Defitions for the address spaces of the AVR32 CPUs. Heavily based on
+ * include/asm-sh/addrspace.h
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_ADDRSPACE_H
+#define __ASM_AVR32_ADDRSPACE_H
+
+#ifdef CONFIG_MMU
+
+/* Memory segments when segmentation is enabled */
+#define P0SEG		0x00000000
+#define P1SEG		0x80000000
+#define P2SEG		0xa0000000
+#define P3SEG		0xc0000000
+#define P4SEG		0xe0000000
+
+/* Returns the privileged segment base of a given address */
+#define PXSEG(a)	(((unsigned long)(a)) & 0xe0000000)
+
+/* Returns the physical address of a PnSEG (n=1,2) address */
+#define PHYSADDR(a)	(((unsigned long)(a)) & 0x1fffffff)
+
+/*
+ * Map an address to a certain privileged segment
+ */
+#define P1SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P1SEG))
+#define P2SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P2SEG))
+#define P3SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P3SEG))
+#define P4SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P4SEG))
+
+#endif /* CONFIG_MMU */
+
+#endif /* __ASM_AVR32_ADDRSPACE_H */
diff --git a/include/asm-avr32/arch-at32ap/at91rm9200_pdc.h b/include/asm-avr32/arch-at32ap/at91rm9200_pdc.h
new file mode 100644
index 000000000000..ce1150d4438d
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/at91rm9200_pdc.h
@@ -0,0 +1,36 @@
+/*
+ * include/asm-arm/arch-at91rm9200/at91rm9200_pdc.h
+ *
+ * Copyright (C) 2005 Ivan Kokshaysky
+ * Copyright (C) SAN People
+ *
+ * Peripheral Data Controller (PDC) registers.
+ * Based on AT91RM9200 datasheet revision E.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91RM9200_PDC_H
+#define AT91RM9200_PDC_H
+
+#define AT91_PDC_RPR		0x100	/* Receive Pointer Register */
+#define AT91_PDC_RCR		0x104	/* Receive Counter Register */
+#define AT91_PDC_TPR		0x108	/* Transmit Pointer Register */
+#define AT91_PDC_TCR		0x10c	/* Transmit Counter Register */
+#define AT91_PDC_RNPR		0x110	/* Receive Next Pointer Register */
+#define AT91_PDC_RNCR		0x114	/* Receive Next Counter Register */
+#define AT91_PDC_TNPR		0x118	/* Transmit Next Pointer Register */
+#define AT91_PDC_TNCR		0x11c	/* Transmit Next Counter Register */
+
+#define AT91_PDC_PTCR		0x120	/* Transfer Control Register */
+#define		AT91_PDC_RXTEN		(1 << 0)	/* Receiver Transfer Enable */
+#define		AT91_PDC_RXTDIS		(1 << 1)	/* Receiver Transfer Disable */
+#define		AT91_PDC_TXTEN		(1 << 8)	/* Transmitter Transfer Enable */
+#define		AT91_PDC_TXTDIS		(1 << 9)	/* Transmitter Transfer Disable */
+
+#define AT91_PDC_PTSR		0x124	/* Transfer Status Register */
+
+#endif
diff --git a/include/asm-avr32/arch-at32ap/at91rm9200_usart.h b/include/asm-avr32/arch-at32ap/at91rm9200_usart.h
new file mode 100644
index 000000000000..79f851e31b9c
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/at91rm9200_usart.h
@@ -0,0 +1,123 @@
+/*
+ * include/asm-arm/arch-at91rm9200/at91rm9200_usart.h
+ *
+ * Copyright (C) 2005 Ivan Kokshaysky
+ * Copyright (C) SAN People
+ *
+ * USART registers.
+ * Based on AT91RM9200 datasheet revision E.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91RM9200_USART_H
+#define AT91RM9200_USART_H
+
+#define AT91_US_CR		0x00			/* Control Register */
+#define		AT91_US_RSTRX		(1 <<  2)		/* Reset Receiver */
+#define		AT91_US_RSTTX		(1 <<  3)		/* Reset Transmitter */
+#define		AT91_US_RXEN		(1 <<  4)		/* Receiver Enable */
+#define		AT91_US_RXDIS		(1 <<  5)		/* Receiver Disable */
+#define		AT91_US_TXEN		(1 <<  6)		/* Transmitter Enable */
+#define		AT91_US_TXDIS		(1 <<  7)		/* Transmitter Disable */
+#define		AT91_US_RSTSTA		(1 <<  8)		/* Reset Status Bits */
+#define		AT91_US_STTBRK		(1 <<  9)		/* Start Break */
+#define		AT91_US_STPBRK		(1 << 10)		/* Stop Break */
+#define		AT91_US_STTTO		(1 << 11)		/* Start Time-out */
+#define		AT91_US_SENDA		(1 << 12)		/* Send Address */
+#define		AT91_US_RSTIT		(1 << 13)		/* Reset Iterations */
+#define		AT91_US_RSTNACK		(1 << 14)		/* Reset Non Acknowledge */
+#define		AT91_US_RETTO		(1 << 15)		/* Rearm Time-out */
+#define		AT91_US_DTREN		(1 << 16)		/* Data Terminal Ready Enable */
+#define		AT91_US_DTRDIS		(1 << 17)		/* Data Terminal Ready Disable */
+#define		AT91_US_RTSEN		(1 << 18)		/* Request To Send Enable */
+#define		AT91_US_RTSDIS		(1 << 19)		/* Request To Send Disable */
+
+#define AT91_US_MR		0x04			/* Mode Register */
+#define		AT91_US_USMODE		(0xf <<  0)		/* Mode of the USART */
+#define			AT91_US_USMODE_NORMAL		0
+#define			AT91_US_USMODE_RS485		1
+#define			AT91_US_USMODE_HWHS		2
+#define			AT91_US_USMODE_MODEM		3
+#define			AT91_US_USMODE_ISO7816_T0	4
+#define			AT91_US_USMODE_ISO7816_T1	6
+#define			AT91_US_USMODE_IRDA		8
+#define		AT91_US_USCLKS		(3   <<  4)		/* Clock Selection */
+#define		AT91_US_CHRL		(3   <<  6)		/* Character Length */
+#define			AT91_US_CHRL_5			(0 <<  6)
+#define			AT91_US_CHRL_6			(1 <<  6)
+#define			AT91_US_CHRL_7			(2 <<  6)
+#define			AT91_US_CHRL_8			(3 <<  6)
+#define		AT91_US_SYNC		(1 <<  8)		/* Synchronous Mode Select */
+#define		AT91_US_PAR		(7 <<  9)		/* Parity Type */
+#define			AT91_US_PAR_EVEN		(0 <<  9)
+#define			AT91_US_PAR_ODD			(1 <<  9)
+#define			AT91_US_PAR_SPACE		(2 <<  9)
+#define			AT91_US_PAR_MARK		(3 <<  9)
+#define			AT91_US_PAR_NONE		(4 <<  9)
+#define			AT91_US_PAR_MULTI_DROP		(6 <<  9)
+#define		AT91_US_NBSTOP		(3 << 12)		/* Number of Stop Bits */
+#define			AT91_US_NBSTOP_1		(0 << 12)
+#define			AT91_US_NBSTOP_1_5		(1 << 12)
+#define			AT91_US_NBSTOP_2		(2 << 12)
+#define		AT91_US_CHMODE		(3 << 14)		/* Channel Mode */
+#define			AT91_US_CHMODE_NORMAL		(0 << 14)
+#define			AT91_US_CHMODE_ECHO		(1 << 14)
+#define			AT91_US_CHMODE_LOC_LOOP		(2 << 14)
+#define			AT91_US_CHMODE_REM_LOOP		(3 << 14)
+#define		AT91_US_MSBF		(1 << 16)		/* Bit Order */
+#define		AT91_US_MODE9		(1 << 17)		/* 9-bit Character Length */
+#define		AT91_US_CLKO		(1 << 18)		/* Clock Output Select */
+#define		AT91_US_OVER		(1 << 19)		/* Oversampling Mode */
+#define		AT91_US_INACK		(1 << 20)		/* Inhibit Non Acknowledge */
+#define		AT91_US_DSNACK		(1 << 21)		/* Disable Successive NACK */
+#define		AT91_US_MAX_ITER	(7 << 24)		/* Max Iterations */
+#define		AT91_US_FILTER		(1 << 28)		/* Infrared Receive Line Filter */
+
+#define AT91_US_IER		0x08			/* Interrupt Enable Register */
+#define		AT91_US_RXRDY		(1 <<  0)		/* Receiver Ready */
+#define		AT91_US_TXRDY		(1 <<  1)		/* Transmitter Ready */
+#define		AT91_US_RXBRK		(1 <<  2)		/* Break Received / End of Break */
+#define		AT91_US_ENDRX		(1 <<  3)		/* End of Receiver Transfer */
+#define		AT91_US_ENDTX		(1 <<  4)		/* End of Transmitter Transfer */
+#define		AT91_US_OVRE		(1 <<  5)		/* Overrun Error */
+#define		AT91_US_FRAME		(1 <<  6)		/* Framing Error */
+#define		AT91_US_PARE		(1 <<  7)		/* Parity Error */
+#define		AT91_US_TIMEOUT		(1 <<  8)		/* Receiver Time-out */
+#define		AT91_US_TXEMPTY		(1 <<  9)		/* Transmitter Empty */
+#define		AT91_US_ITERATION	(1 << 10)		/* Max number of Repetitions Reached */
+#define		AT91_US_TXBUFE		(1 << 11)		/* Transmission Buffer Empty */
+#define		AT91_US_RXBUFF		(1 << 12)		/* Reception Buffer Full */
+#define		AT91_US_NACK		(1 << 13)		/* Non Acknowledge */
+#define		AT91_US_RIIC		(1 << 16)		/* Ring Indicator Input Change */
+#define		AT91_US_DSRIC		(1 << 17)		/* Data Set Ready Input Change */
+#define		AT91_US_DCDIC		(1 << 18)		/* Data Carrier Detect Input Change */
+#define		AT91_US_CTSIC		(1 << 19)		/* Clear to Send Input Change */
+#define		AT91_US_RI		(1 << 20)		/* RI */
+#define		AT91_US_DSR		(1 << 21)		/* DSR */
+#define		AT91_US_DCD		(1 << 22)		/* DCD */
+#define		AT91_US_CTS		(1 << 23)		/* CTS */
+
+#define AT91_US_IDR		0x0c			/* Interrupt Disable Register */
+#define AT91_US_IMR		0x10			/* Interrupt Mask Register */
+#define AT91_US_CSR		0x14			/* Channel Status Register */
+#define AT91_US_RHR		0x18			/* Receiver Holding Register */
+#define AT91_US_THR		0x1c			/* Transmitter Holding Register */
+
+#define AT91_US_BRGR		0x20			/* Baud Rate Generator Register */
+#define		AT91_US_CD		(0xffff << 0)		/* Clock Divider */
+
+#define AT91_US_RTOR		0x24			/* Receiver Time-out Register */
+#define		AT91_US_TO		(0xffff << 0)		/* Time-out Value */
+
+#define AT91_US_TTGR		0x28			/* Transmitter Timeguard Register */
+#define		AT91_US_TG		(0xff << 0)		/* Timeguard Value */
+
+#define AT91_US_FIDI		0x40			/* FI DI Ratio Register */
+#define AT91_US_NER		0x44			/* Number of Errors Register */
+#define AT91_US_IF		0x4c			/* IrDA Filter Register */
+
+#endif
diff --git a/include/asm-avr32/arch-at32ap/board.h b/include/asm-avr32/arch-at32ap/board.h
new file mode 100644
index 000000000000..39368e18ab20
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/board.h
@@ -0,0 +1,35 @@
+/*
+ * Platform data definitions.
+ */
+#ifndef __ASM_ARCH_BOARD_H
+#define __ASM_ARCH_BOARD_H
+
+#include <linux/types.h>
+
+/* Add basic devices: system manager, interrupt controller, portmuxes, etc. */
+void at32_add_system_devices(void);
+
+#define AT91_NR_UART	4
+extern struct platform_device *at91_default_console_device;
+
+struct platform_device *at32_add_device_usart(unsigned int id);
+
+struct eth_platform_data {
+	u8	valid;
+	u8	mii_phy_addr;
+	u8	is_rmii;
+	u8	hw_addr[6];
+};
+struct platform_device *
+at32_add_device_eth(unsigned int id, struct eth_platform_data *data);
+
+struct platform_device *at32_add_device_spi(unsigned int id);
+
+struct lcdc_platform_data {
+	unsigned long fbmem_start;
+	unsigned long fbmem_size;
+};
+struct platform_device *
+at32_add_device_lcdc(unsigned int id, struct lcdc_platform_data *data);
+
+#endif /* __ASM_ARCH_BOARD_H */
diff --git a/include/asm-avr32/arch-at32ap/init.h b/include/asm-avr32/arch-at32ap/init.h
new file mode 100644
index 000000000000..43722634e069
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/init.h
@@ -0,0 +1,21 @@
+/*
+ * AT32AP platform initialization calls.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_AT32AP_INIT_H__
+#define __ASM_AVR32_AT32AP_INIT_H__
+
+void setup_platform(void);
+
+/* Called by setup_platform */
+void at32_clock_init(void);
+void at32_portmux_init(void);
+
+void at32_setup_serial_console(unsigned int usart_id);
+
+#endif /* __ASM_AVR32_AT32AP_INIT_H__ */
diff --git a/include/asm-avr32/arch-at32ap/portmux.h b/include/asm-avr32/arch-at32ap/portmux.h
new file mode 100644
index 000000000000..4d50421262a1
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/portmux.h
@@ -0,0 +1,16 @@
+/*
+ * AT32 portmux interface.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_AT32_PORTMUX_H__
+#define __ASM_AVR32_AT32_PORTMUX_H__
+
+void portmux_set_func(unsigned int portmux_id, unsigned int pin_id,
+		      unsigned int function_id);
+
+#endif /* __ASM_AVR32_AT32_PORTMUX_H__ */
diff --git a/include/asm-avr32/arch-at32ap/sm.h b/include/asm-avr32/arch-at32ap/sm.h
new file mode 100644
index 000000000000..265a9ead20bf
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/sm.h
@@ -0,0 +1,27 @@
+/*
+ * AT32 System Manager interface.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_AT32_SM_H__
+#define __ASM_AVR32_AT32_SM_H__
+
+struct irq_chip;
+struct platform_device;
+
+struct at32_sm {
+	spinlock_t lock;
+	void __iomem *regs;
+	struct irq_chip *eim_chip;
+	unsigned int eim_first_irq;
+	struct platform_device *pdev;
+};
+
+extern struct platform_device at32_sm_device;
+extern struct at32_sm system_manager;
+
+#endif /* __ASM_AVR32_AT32_SM_H__ */
diff --git a/include/asm-avr32/asm.h b/include/asm-avr32/asm.h
new file mode 100644
index 000000000000..515c7618952b
--- /dev/null
+++ b/include/asm-avr32/asm.h
@@ -0,0 +1,102 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_ASM_H__
+#define __ASM_AVR32_ASM_H__
+
+#include <asm/sysreg.h>
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
+
+#define mask_interrupts		ssrf	SR_GM_BIT
+#define mask_exceptions		ssrf	SR_EM_BIT
+#define unmask_interrupts	csrf	SR_GM_BIT
+#define unmask_exceptions	csrf	SR_EM_BIT
+
+#ifdef CONFIG_FRAME_POINTER
+	.macro	save_fp
+	st.w	--sp, r7
+	.endm
+	.macro	restore_fp
+	ld.w	r7, sp++
+	.endm
+	.macro	zero_fp
+	mov	r7, 0
+	.endm
+#else
+	.macro	save_fp
+	.endm
+	.macro	restore_fp
+	.endm
+	.macro	zero_fp
+	.endm
+#endif
+	.macro	get_thread_info reg
+	mov	\reg, sp
+	andl	\reg, ~(THREAD_SIZE - 1) & 0xffff
+	.endm
+
+	/* Save and restore registers */
+	.macro	save_min sr, tmp=lr
+	pushm	lr
+	mfsr	\tmp, \sr
+	zero_fp
+	st.w	--sp, \tmp
+	.endm
+
+	.macro	restore_min sr, tmp=lr
+	ld.w	\tmp, sp++
+	mtsr	\sr, \tmp
+	popm	lr
+	.endm
+
+	.macro	save_half sr, tmp=lr
+	save_fp
+	pushm	r8-r9,r10,r11,r12,lr
+	zero_fp
+	mfsr	\tmp, \sr
+	st.w	--sp, \tmp
+	.endm
+
+	.macro	restore_half sr, tmp=lr
+	ld.w	\tmp, sp++
+	mtsr	\sr, \tmp
+	popm	r8-r9,r10,r11,r12,lr
+	restore_fp
+	.endm
+
+	.macro	save_full_user sr, tmp=lr
+	stmts	--sp, r0,r1,r2,r3,r4,r5,r6,r7,r8,r9,r10,r11,r12,sp,lr
+	st.w	--sp, lr
+	zero_fp
+	mfsr	\tmp, \sr
+	st.w	--sp, \tmp
+	.endm
+
+	.macro	restore_full_user sr, tmp=lr
+	ld.w	\tmp, sp++
+	mtsr	\sr, \tmp
+	ld.w	lr, sp++
+	ldmts	sp++, r0,r1,r2,r3,r4,r5,r6,r7,r8,r9,r10,r11,r12,sp,lr
+	.endm
+
+	/* uaccess macros */
+	.macro branch_if_kernel scratch, label
+	get_thread_info \scratch
+	ld.w	\scratch, \scratch[TI_flags]
+	bld	\scratch, TIF_USERSPACE
+	brcc	\label
+	.endm
+
+	.macro ret_if_privileged scratch, addr, size, ret
+	sub	\scratch, \size, 1
+	add	\scratch, \addr
+	retcs	\ret
+	retmi	\ret
+	.endm
+
+#endif /* __ASM_AVR32_ASM_H__ */
diff --git a/include/asm-avr32/atomic.h b/include/asm-avr32/atomic.h
new file mode 100644
index 000000000000..e0b9c44c126c
--- /dev/null
+++ b/include/asm-avr32/atomic.h
@@ -0,0 +1,201 @@
+/*
+ * Atomic operations that C can't guarantee us.  Useful for
+ * resource counting etc.
+ *
+ * But use these as seldom as possible since they are slower than
+ * regular operations.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_ATOMIC_H
+#define __ASM_AVR32_ATOMIC_H
+
+#include <asm/system.h>
+
+typedef struct { volatile int counter; } atomic_t;
+#define ATOMIC_INIT(i)  { (i) }
+
+#define atomic_read(v)		((v)->counter)
+#define atomic_set(v, i)	(((v)->counter) = i)
+
+/*
+ * atomic_sub_return - subtract the atomic variable
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically subtracts @i from @v. Returns the resulting value.
+ */
+static inline int atomic_sub_return(int i, atomic_t *v)
+{
+	int result;
+
+	asm volatile(
+		"/* atomic_sub_return */\n"
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %2\n"
+		"	sub	%0, %3\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b"
+		: "=&r"(result), "=o"(v->counter)
+		: "m"(v->counter), "ir"(i)
+		: "cc");
+
+	return result;
+}
+
+/*
+ * atomic_add_return - add integer to atomic variable
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v. Returns the resulting value.
+ */
+static inline int atomic_add_return(int i, atomic_t *v)
+{
+	int result;
+
+	if (__builtin_constant_p(i))
+		result = atomic_sub_return(-i, v);
+	else
+		asm volatile(
+			"/* atomic_add_return */\n"
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %1\n"
+			"	add	%0, %3\n"
+			"	stcond	%2, %0\n"
+			"	brne	1b"
+			: "=&r"(result), "=o"(v->counter)
+			: "m"(v->counter), "r"(i)
+			: "cc", "memory");
+
+	return result;
+}
+
+/*
+ * atomic_sub_unless - sub unless the number is a given value
+ * @v: pointer of type atomic_t
+ * @a: the amount to add to v...
+ * @u: ...unless v is equal to u.
+ *
+ * If the atomic value v is not equal to u, this function subtracts a
+ * from v, and returns non zero. If v is equal to u then it returns
+ * zero. This is done as an atomic operation.
+*/
+static inline int atomic_sub_unless(atomic_t *v, int a, int u)
+{
+	int tmp, result = 0;
+
+	asm volatile(
+		"/* atomic_sub_unless */\n"
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %3\n"
+		"	cp.w	%0, %5\n"
+		"	breq	1f\n"
+		"	sub	%0, %4\n"
+		"	stcond	%2, %0\n"
+		"	brne	1b\n"
+		"	mov	%1, 1\n"
+		"1:"
+		: "=&r"(tmp), "=&r"(result), "=o"(v->counter)
+		: "m"(v->counter), "ir"(a), "ir"(u)
+		: "cc", "memory");
+
+	return result;
+}
+
+/*
+ * atomic_add_unless - add unless the number is a given value
+ * @v: pointer of type atomic_t
+ * @a: the amount to add to v...
+ * @u: ...unless v is equal to u.
+ *
+ * If the atomic value v is not equal to u, this function adds a to v,
+ * and returns non zero. If v is equal to u then it returns zero. This
+ * is done as an atomic operation.
+*/
+static inline int atomic_add_unless(atomic_t *v, int a, int u)
+{
+	int tmp, result;
+
+	if (__builtin_constant_p(a))
+		result = atomic_sub_unless(v, -a, u);
+	else {
+		result = 0;
+		asm volatile(
+			"/* atomic_add_unless */\n"
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	cp.w	%0, %5\n"
+			"	breq	1f\n"
+			"	add	%0, %4\n"
+			"	stcond	%2, %0\n"
+			"	brne	1b\n"
+			"	mov	%1, 1\n"
+			"1:"
+			: "=&r"(tmp), "=&r"(result), "=o"(v->counter)
+			: "m"(v->counter), "r"(a), "ir"(u)
+			: "cc", "memory");
+	}
+
+	return result;
+}
+
+/*
+ * atomic_sub_if_positive - conditionally subtract integer from atomic variable
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically test @v and subtract @i if @v is greater or equal than @i.
+ * The function returns the old value of @v minus @i.
+ */
+static inline int atomic_sub_if_positive(int i, atomic_t *v)
+{
+	int result;
+
+	asm volatile(
+		"/* atomic_sub_if_positive */\n"
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %2\n"
+		"	sub	%0, %3\n"
+		"	brlt	1f\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b\n"
+		"1:"
+		: "=&r"(result), "=o"(v->counter)
+		: "m"(v->counter), "ir"(i)
+		: "cc", "memory");
+
+	return result;
+}
+
+#define atomic_xchg(v, new)	(xchg(&((v)->counter), new))
+#define atomic_cmpxchg(v, o, n)	((int)cmpxchg(&((v)->counter), (o), (n)))
+
+#define atomic_sub(i, v)	(void)atomic_sub_return(i, v)
+#define atomic_add(i, v)	(void)atomic_add_return(i, v)
+#define atomic_dec(v)		atomic_sub(1, (v))
+#define atomic_inc(v)		atomic_add(1, (v))
+
+#define atomic_dec_return(v)	atomic_sub_return(1, v)
+#define atomic_inc_return(v)	atomic_add_return(1, v)
+
+#define atomic_sub_and_test(i, v) (atomic_sub_return(i, v) == 0)
+#define atomic_inc_and_test(v) (atomic_add_return(1, v) == 0)
+#define atomic_dec_and_test(v) (atomic_sub_return(1, v) == 0)
+#define atomic_add_negative(i, v) (atomic_add_return(i, v) < 0)
+
+#define atomic_inc_not_zero(v)	atomic_add_unless(v, 1, 0)
+#define atomic_dec_if_positive(v) atomic_sub_if_positive(1, v)
+
+#define smp_mb__before_atomic_dec()	barrier()
+#define smp_mb__after_atomic_dec()	barrier()
+#define smp_mb__before_atomic_inc()	barrier()
+#define smp_mb__after_atomic_inc()	barrier()
+
+#include <asm-generic/atomic.h>
+
+#endif /*  __ASM_AVR32_ATOMIC_H */
diff --git a/include/asm-avr32/auxvec.h b/include/asm-avr32/auxvec.h
new file mode 100644
index 000000000000..d5dd435bf8f4
--- /dev/null
+++ b/include/asm-avr32/auxvec.h
@@ -0,0 +1,4 @@
+#ifndef __ASM_AVR32_AUXVEC_H
+#define __ASM_AVR32_AUXVEC_H
+
+#endif /* __ASM_AVR32_AUXVEC_H */
diff --git a/include/asm-avr32/bitops.h b/include/asm-avr32/bitops.h
new file mode 100644
index 000000000000..5299f8c8e11d
--- /dev/null
+++ b/include/asm-avr32/bitops.h
@@ -0,0 +1,296 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_BITOPS_H
+#define __ASM_AVR32_BITOPS_H
+
+#include <asm/byteorder.h>
+#include <asm/system.h>
+
+/*
+ * clear_bit() doesn't provide any barrier for the compiler
+ */
+#define smp_mb__before_clear_bit()	barrier()
+#define smp_mb__after_clear_bit()	barrier()
+
+/*
+ * set_bit - Atomically set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * This function is atomic and may not be reordered.  See __set_bit()
+ * if you do not require the atomic guarantees.
+ *
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void set_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long tmp;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	sbr	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "i"(nr)
+			: "cc");
+	} else {
+		unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	or	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "r"(mask)
+			: "cc");
+	}
+}
+
+/*
+ * clear_bit - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * clear_bit() is atomic and may not be reordered.  However, it does
+ * not contain a memory barrier, so if it is used for locking purposes,
+ * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
+ * in order to ensure changes are visible on other processors.
+ */
+static inline void clear_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long tmp;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	cbr	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "i"(nr)
+			: "cc");
+	} else {
+		unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	andn	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "r"(mask)
+			: "cc");
+	}
+}
+
+/*
+ * change_bit - Toggle a bit in memory
+ * @nr: Bit to change
+ * @addr: Address to start counting from
+ *
+ * change_bit() is atomic and may not be reordered.
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void change_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp;
+
+	asm volatile(
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %2\n"
+		"	eor	%0, %3\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b"
+		: "=&r"(tmp), "=o"(*p)
+		: "m"(*p), "r"(mask)
+		: "cc");
+}
+
+/*
+ * test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_set_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp, old;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	mov	%2, %0\n"
+			"	sbr	%0, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "i"(nr)
+			: "memory", "cc");
+	} else {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%2, %3\n"
+			"	or	%0, %2, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "r"(mask)
+			: "memory", "cc");
+	}
+
+	return (old & mask) != 0;
+}
+
+/*
+ * test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_clear_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp, old;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	mov	%2, %0\n"
+			"	cbr	%0, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "i"(nr)
+			: "memory", "cc");
+	} else {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	mov	%2, %0\n"
+			"	andn	%0, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "r"(mask)
+			: "memory", "cc");
+	}
+
+	return (old & mask) != 0;
+}
+
+/*
+ * test_and_change_bit - Change a bit and return its old value
+ * @nr: Bit to change
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_change_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp, old;
+
+	asm volatile(
+		"1:	ssrf	5\n"
+		"	ld.w	%2, %3\n"
+		"	eor	%0, %2, %4\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b"
+		: "=&r"(tmp), "=o"(*p), "=&r"(old)
+		: "m"(*p), "r"(mask)
+		: "memory", "cc");
+
+	return (old & mask) != 0;
+}
+
+#include <asm-generic/bitops/non-atomic.h>
+
+/* Find First bit Set */
+static inline unsigned long __ffs(unsigned long word)
+{
+	unsigned long result;
+
+	asm("brev %1\n\t"
+	    "clz %0,%1"
+	    : "=r"(result), "=&r"(word)
+	    : "1"(word));
+	return result;
+}
+
+/* Find First Zero */
+static inline unsigned long ffz(unsigned long word)
+{
+	return __ffs(~word);
+}
+
+/* Find Last bit Set */
+static inline int fls(unsigned long word)
+{
+	unsigned long result;
+
+	asm("clz %0,%1" : "=r"(result) : "r"(word));
+	return 32 - result;
+}
+
+unsigned long find_first_zero_bit(const unsigned long *addr,
+				  unsigned long size);
+unsigned long find_next_zero_bit(const unsigned long *addr,
+				 unsigned long size,
+				 unsigned long offset);
+unsigned long find_first_bit(const unsigned long *addr,
+			     unsigned long size);
+unsigned long find_next_bit(const unsigned long *addr,
+				 unsigned long size,
+				 unsigned long offset);
+
+/*
+ * ffs: find first bit set. This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ *
+ * The difference is that bit numbering starts at 1, and if no bit is set,
+ * the function returns 0.
+ */
+static inline int ffs(unsigned long word)
+{
+	if(word == 0)
+		return 0;
+	return __ffs(word) + 1;
+}
+
+#include <asm-generic/bitops/fls64.h>
+#include <asm-generic/bitops/sched.h>
+#include <asm-generic/bitops/hweight.h>
+
+#include <asm-generic/bitops/ext2-non-atomic.h>
+#include <asm-generic/bitops/ext2-atomic.h>
+#include <asm-generic/bitops/minix-le.h>
+
+#endif /* __ASM_AVR32_BITOPS_H */
diff --git a/include/asm-avr32/bug.h b/include/asm-avr32/bug.h
new file mode 100644
index 000000000000..521766bc9366
--- /dev/null
+++ b/include/asm-avr32/bug.h
@@ -0,0 +1,47 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_BUG_H
+#define __ASM_AVR32_BUG_H
+
+#ifdef CONFIG_BUG
+
+/*
+ * According to our Chief Architect, this compact opcode is very
+ * unlikely to ever be implemented.
+ */
+#define AVR32_BUG_OPCODE	0x5df0
+
+#ifdef CONFIG_DEBUG_BUGVERBOSE
+
+#define BUG()								\
+	do {								\
+		asm volatile(".hword	%0\n\t"				\
+			     ".hword	%1\n\t"				\
+			     ".long	%2"				\
+			     :						\
+			     : "n"(AVR32_BUG_OPCODE),			\
+			       "i"(__LINE__), "X"(__FILE__));		\
+	} while (0)
+
+#else
+
+#define BUG()								\
+	do {								\
+		asm volatile(".hword	%0\n\t"				\
+			     : : "n"(AVR32_BUG_OPCODE));		\
+	} while (0)
+
+#endif /* CONFIG_DEBUG_BUGVERBOSE */
+
+#define HAVE_ARCH_BUG
+
+#endif /* CONFIG_BUG */
+
+#include <asm-generic/bug.h>
+
+#endif /* __ASM_AVR32_BUG_H */
diff --git a/include/asm-avr32/bugs.h b/include/asm-avr32/bugs.h
new file mode 100644
index 000000000000..7635e770622e
--- /dev/null
+++ b/include/asm-avr32/bugs.h
@@ -0,0 +1,15 @@
+/*
+ * This is included by init/main.c to check for architecture-dependent bugs.
+ *
+ * Needs:
+ *      void check_bugs(void);
+ */
+#ifndef __ASM_AVR32_BUGS_H
+#define __ASM_AVR32_BUGS_H
+
+static void __init check_bugs(void)
+{
+	cpu_data->loops_per_jiffy = loops_per_jiffy;
+}
+
+#endif /* __ASM_AVR32_BUGS_H */
diff --git a/include/asm-avr32/byteorder.h b/include/asm-avr32/byteorder.h
new file mode 100644
index 000000000000..402ff4125cdc
--- /dev/null
+++ b/include/asm-avr32/byteorder.h
@@ -0,0 +1,25 @@
+/*
+ * AVR32 endian-conversion functions.
+ */
+#ifndef __ASM_AVR32_BYTEORDER_H
+#define __ASM_AVR32_BYTEORDER_H
+
+#include <asm/types.h>
+#include <linux/compiler.h>
+
+#ifdef __CHECKER__
+extern unsigned long __builtin_bswap_32(unsigned long x);
+extern unsigned short __builtin_bswap_16(unsigned short x);
+#endif
+
+#define __arch__swab32(x) __builtin_bswap_32(x)
+#define __arch__swab16(x) __builtin_bswap_16(x)
+
+#if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
+# define __BYTEORDER_HAS_U64__
+# define __SWAB_64_THRU_32__
+#endif
+
+#include <linux/byteorder/big_endian.h>
+
+#endif /* __ASM_AVR32_BYTEORDER_H */
diff --git a/include/asm-avr32/cache.h b/include/asm-avr32/cache.h
new file mode 100644
index 000000000000..dabb955f3c00
--- /dev/null
+++ b/include/asm-avr32/cache.h
@@ -0,0 +1,29 @@
+#ifndef __ASM_AVR32_CACHE_H
+#define __ASM_AVR32_CACHE_H
+
+#define L1_CACHE_SHIFT 5
+#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT)
+
+#ifndef __ASSEMBLER__
+struct cache_info {
+	unsigned int ways;
+	unsigned int sets;
+	unsigned int linesz;
+};
+#endif /* __ASSEMBLER */
+
+/* Cache operation constants */
+#define ICACHE_FLUSH		0x00
+#define ICACHE_INVALIDATE	0x01
+#define ICACHE_LOCK		0x02
+#define ICACHE_UNLOCK		0x03
+#define ICACHE_PREFETCH		0x04
+
+#define DCACHE_FLUSH		0x08
+#define DCACHE_LOCK		0x09
+#define DCACHE_UNLOCK		0x0a
+#define DCACHE_INVALIDATE	0x0b
+#define DCACHE_CLEAN		0x0c
+#define DCACHE_CLEAN_INVAL	0x0d
+
+#endif /* __ASM_AVR32_CACHE_H */
diff --git a/include/asm-avr32/cachectl.h b/include/asm-avr32/cachectl.h
new file mode 100644
index 000000000000..4faf1ce60061
--- /dev/null
+++ b/include/asm-avr32/cachectl.h
@@ -0,0 +1,11 @@
+#ifndef __ASM_AVR32_CACHECTL_H
+#define __ASM_AVR32_CACHECTL_H
+
+/*
+ * Operations that can be performed through the cacheflush system call
+ */
+
+/* Clean the data cache, then invalidate the icache */
+#define CACHE_IFLUSH	0
+
+#endif /* __ASM_AVR32_CACHECTL_H */
diff --git a/include/asm-avr32/cacheflush.h b/include/asm-avr32/cacheflush.h
new file mode 100644
index 000000000000..f1bf1708980e
--- /dev/null
+++ b/include/asm-avr32/cacheflush.h
@@ -0,0 +1,129 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_CACHEFLUSH_H
+#define __ASM_AVR32_CACHEFLUSH_H
+
+/* Keep includes the same across arches.  */
+#include <linux/mm.h>
+
+#define CACHE_OP_ICACHE_INVALIDATE	0x01
+#define CACHE_OP_DCACHE_INVALIDATE	0x0b
+#define CACHE_OP_DCACHE_CLEAN		0x0c
+#define CACHE_OP_DCACHE_CLEAN_INVAL	0x0d
+
+/*
+ * Invalidate any cacheline containing virtual address vaddr without
+ * writing anything back to memory.
+ *
+ * Note that this function may corrupt unrelated data structures when
+ * applied on buffers that are not cacheline aligned in both ends.
+ */
+static inline void invalidate_dcache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_DCACHE_INVALIDATE)
+		     : "memory");
+}
+
+/*
+ * Make sure any cacheline containing virtual address vaddr is written
+ * to memory.
+ */
+static inline void clean_dcache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_DCACHE_CLEAN)
+		     : "memory");
+}
+
+/*
+ * Make sure any cacheline containing virtual address vaddr is written
+ * to memory and then invalidate it.
+ */
+static inline void flush_dcache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_DCACHE_CLEAN_INVAL)
+		     : "memory");
+}
+
+/*
+ * Invalidate any instruction cacheline containing virtual address
+ * vaddr.
+ */
+static inline void invalidate_icache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_ICACHE_INVALIDATE)
+		     : "memory");
+}
+
+/*
+ * Applies the above functions on all lines that are touched by the
+ * specified virtual address range.
+ */
+void invalidate_dcache_region(void *start, size_t len);
+void clean_dcache_region(void *start, size_t len);
+void flush_dcache_region(void *start, size_t len);
+void invalidate_icache_region(void *start, size_t len);
+
+/*
+ * Make sure any pending writes are completed before continuing.
+ */
+#define flush_write_buffer() asm volatile("sync 0" : : : "memory")
+
+/*
+ * The following functions are called when a virtual mapping changes.
+ * We do not need to flush anything in this case.
+ */
+#define flush_cache_all()			do { } while (0)
+#define flush_cache_mm(mm)			do { } while (0)
+#define flush_cache_range(vma, start, end)	do { } while (0)
+#define flush_cache_page(vma, vmaddr, pfn)	do { } while (0)
+#define flush_cache_vmap(start, end)		do { } while (0)
+#define flush_cache_vunmap(start, end)		do { } while (0)
+
+/*
+ * I think we need to implement this one to be able to reliably
+ * execute pages from RAMDISK. However, if we implement the
+ * flush_dcache_*() functions, it might not be needed anymore.
+ *
+ * #define flush_icache_page(vma, page)		do { } while (0)
+ */
+extern void flush_icache_page(struct vm_area_struct *vma, struct page *page);
+
+/*
+ * These are (I think) related to D-cache aliasing.  We might need to
+ * do something here, but only for certain configurations.  No such
+ * configurations exist at this time.
+ */
+#define flush_dcache_page(page)			do { } while (0)
+#define flush_dcache_mmap_lock(page)		do { } while (0)
+#define flush_dcache_mmap_unlock(page)		do { } while (0)
+
+/*
+ * These are for I/D cache coherency. In this case, we do need to
+ * flush with all configurations.
+ */
+extern void flush_icache_range(unsigned long start, unsigned long end);
+extern void flush_icache_user_range(struct vm_area_struct *vma,
+				    struct page *page,
+				    unsigned long addr, int len);
+
+#define copy_to_user_page(vma, page, vaddr, dst, src, len) do {	\
+	memcpy(dst, src, len);					\
+	flush_icache_user_range(vma, page, vaddr, len);		\
+} while(0)
+#define copy_from_user_page(vma, page, vaddr, dst, src, len)	\
+	memcpy(dst, src, len)
+
+#endif /* __ASM_AVR32_CACHEFLUSH_H */
diff --git a/include/asm-avr32/checksum.h b/include/asm-avr32/checksum.h
new file mode 100644
index 000000000000..41b7af09edc4
--- /dev/null
+++ b/include/asm-avr32/checksum.h
@@ -0,0 +1,156 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_CHECKSUM_H
+#define __ASM_AVR32_CHECKSUM_H
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+unsigned int csum_partial(const unsigned char * buff, int len,
+			  unsigned int sum);
+
+/*
+ * the same as csum_partial, but copies from src while it
+ * checksums, and handles user-space pointer exceptions correctly, when needed.
+ *
+ * here even more important to align src and dst on a 32-bit (or even
+ * better 64-bit) boundary
+ */
+unsigned int csum_partial_copy_generic(const char *src, char *dst, int len,
+				       int sum, int *src_err_ptr,
+				       int *dst_err_ptr);
+
+/*
+ *	Note: when you get a NULL pointer exception here this means someone
+ *	passed in an incorrect kernel address to one of these functions.
+ *
+ *	If you use these functions directly please don't forget the
+ *	verify_area().
+ */
+static inline
+unsigned int csum_partial_copy_nocheck(const char *src, char *dst,
+				       int len, int sum)
+{
+	return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
+}
+
+static inline
+unsigned int csum_partial_copy_from_user (const char __user *src, char *dst,
+					  int len, int sum, int *err_ptr)
+{
+	return csum_partial_copy_generic((const char __force *)src, dst, len,
+					 sum, err_ptr, NULL);
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ */
+static inline unsigned short ip_fast_csum(unsigned char *iph,
+					  unsigned int ihl)
+{
+	unsigned int sum, tmp;
+
+	__asm__ __volatile__(
+		"	ld.w	%0, %1++\n"
+		"	ld.w	%3, %1++\n"
+		"	sub	%2, 4\n"
+		"	add	%0, %3\n"
+		"	ld.w	%3, %1++\n"
+		"	adc	%0, %0, %3\n"
+		"	ld.w	%3, %1++\n"
+		"	adc	%0, %0, %3\n"
+		"	acr	%0\n"
+		"1:	ld.w	%3, %1++\n"
+		"	add	%0, %3\n"
+		"	acr	%0\n"
+		"	sub	%2, 1\n"
+		"	brne	1b\n"
+		"	lsl	%3, %0, 16\n"
+		"	andl	%0, 0\n"
+		"	mov	%2, 0xffff\n"
+		"	add	%0, %3\n"
+		"	adc	%0, %0, %2\n"
+		"	com	%0\n"
+		"	lsr	%0, 16\n"
+		: "=r"(sum), "=r"(iph), "=r"(ihl), "=r"(tmp)
+		: "1"(iph), "2"(ihl)
+		: "memory", "cc");
+	return sum;
+}
+
+/*
+ *	Fold a partial checksum
+ */
+
+static inline unsigned int csum_fold(unsigned int sum)
+{
+	unsigned int tmp;
+
+	asm("	bfextu	%1, %0, 0, 16\n"
+	    "	lsr	%0, 16\n"
+	    "	add	%0, %1\n"
+	    "	bfextu	%1, %0, 16, 16\n"
+	    "	add	%0, %1"
+	    : "=&r"(sum), "=&r"(tmp)
+	    : "0"(sum));
+
+	return ~sum;
+}
+
+static inline unsigned long csum_tcpudp_nofold(unsigned long saddr,
+					       unsigned long daddr,
+					       unsigned short len,
+					       unsigned short proto,
+					       unsigned int sum)
+{
+	asm("	add	%0, %1\n"
+	    "	adc	%0, %0, %2\n"
+	    "	adc	%0, %0, %3\n"
+	    "	acr	%0"
+	    : "=r"(sum)
+	    : "r"(daddr), "r"(saddr), "r"(ntohs(len) | (proto << 16)),
+	      "0"(sum)
+	    : "cc");
+
+	return sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static inline unsigned short int csum_tcpudp_magic(unsigned long saddr,
+						   unsigned long daddr,
+						   unsigned short len,
+						   unsigned short proto,
+						   unsigned int sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+
+static inline unsigned short ip_compute_csum(unsigned char * buff, int len)
+{
+    return csum_fold(csum_partial(buff, len, 0));
+}
+
+#endif /* __ASM_AVR32_CHECKSUM_H */
diff --git a/include/asm-avr32/cputime.h b/include/asm-avr32/cputime.h
new file mode 100644
index 000000000000..e87e0f81cbeb
--- /dev/null
+++ b/include/asm-avr32/cputime.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_CPUTIME_H
+#define __ASM_AVR32_CPUTIME_H
+
+#include <asm-generic/cputime.h>
+
+#endif /* __ASM_AVR32_CPUTIME_H */
diff --git a/include/asm-avr32/current.h b/include/asm-avr32/current.h
new file mode 100644
index 000000000000..c7b0549eab8a
--- /dev/null
+++ b/include/asm-avr32/current.h
@@ -0,0 +1,15 @@
+#ifndef __ASM_AVR32_CURRENT_H
+#define __ASM_AVR32_CURRENT_H
+
+#include <linux/thread_info.h>
+
+struct task_struct;
+
+inline static struct task_struct * get_current(void)
+{
+	return current_thread_info()->task;
+}
+
+#define current get_current()
+
+#endif /* __ASM_AVR32_CURRENT_H */
diff --git a/include/asm-avr32/delay.h b/include/asm-avr32/delay.h
new file mode 100644
index 000000000000..cc3b2e3343b3
--- /dev/null
+++ b/include/asm-avr32/delay.h
@@ -0,0 +1,26 @@
+#ifndef __ASM_AVR32_DELAY_H
+#define __ASM_AVR32_DELAY_H
+
+/*
+ * Copyright (C) 1993 Linus Torvalds
+ *
+ * Delay routines calling functions in arch/avr32/lib/delay.c
+ */
+
+extern void __bad_udelay(void);
+extern void __bad_ndelay(void);
+
+extern void __udelay(unsigned long usecs);
+extern void __ndelay(unsigned long nsecs);
+extern void __const_udelay(unsigned long usecs);
+extern void __delay(unsigned long loops);
+
+#define udelay(n) (__builtin_constant_p(n) ? \
+	((n) > 20000 ? __bad_udelay() : __const_udelay((n) * 0x10c6ul)) : \
+	__udelay(n))
+
+#define ndelay(n) (__builtin_constant_p(n) ? \
+	((n) > 20000 ? __bad_ndelay() : __const_udelay((n) * 5ul)) : \
+	__ndelay(n))
+
+#endif /* __ASM_AVR32_DELAY_H */
diff --git a/include/asm-avr32/div64.h b/include/asm-avr32/div64.h
new file mode 100644
index 000000000000..d7ddd4fdeca6
--- /dev/null
+++ b/include/asm-avr32/div64.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_DIV64_H
+#define __ASM_AVR32_DIV64_H
+
+#include <asm-generic/div64.h>
+
+#endif /* __ASM_AVR32_DIV64_H */
diff --git a/include/asm-avr32/dma-mapping.h b/include/asm-avr32/dma-mapping.h
new file mode 100644
index 000000000000..4c40cb41cdf8
--- /dev/null
+++ b/include/asm-avr32/dma-mapping.h
@@ -0,0 +1,320 @@
+#ifndef __ASM_AVR32_DMA_MAPPING_H
+#define __ASM_AVR32_DMA_MAPPING_H
+
+#include <linux/mm.h>
+#include <linux/device.h>
+#include <asm/scatterlist.h>
+#include <asm/processor.h>
+#include <asm/cacheflush.h>
+#include <asm/io.h>
+
+extern void dma_cache_sync(void *vaddr, size_t size, int direction);
+
+/*
+ * Return whether the given device DMA address mask can be supported
+ * properly.  For example, if your device can only drive the low 24-bits
+ * during bus mastering, then you would pass 0x00ffffff as the mask
+ * to this function.
+ */
+static inline int dma_supported(struct device *dev, u64 mask)
+{
+	/* Fix when needed. I really don't know of any limitations */
+	return 1;
+}
+
+static inline int dma_set_mask(struct device *dev, u64 dma_mask)
+{
+	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
+		return -EIO;
+
+	*dev->dma_mask = dma_mask;
+	return 0;
+}
+
+/**
+ * dma_alloc_coherent - allocate consistent memory for DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: required memory size
+ * @handle: bus-specific DMA address
+ *
+ * Allocate some uncached, unbuffered memory for a device for
+ * performing DMA.  This function allocates pages, and will
+ * return the CPU-viewed address, and sets @handle to be the
+ * device-viewed address.
+ */
+extern void *dma_alloc_coherent(struct device *dev, size_t size,
+				dma_addr_t *handle, gfp_t gfp);
+
+/**
+ * dma_free_coherent - free memory allocated by dma_alloc_coherent
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: size of memory originally requested in dma_alloc_coherent
+ * @cpu_addr: CPU-view address returned from dma_alloc_coherent
+ * @handle: device-view address returned from dma_alloc_coherent
+ *
+ * Free (and unmap) a DMA buffer previously allocated by
+ * dma_alloc_coherent().
+ *
+ * References to memory and mappings associated with cpu_addr/handle
+ * during and after this call executing are illegal.
+ */
+extern void dma_free_coherent(struct device *dev, size_t size,
+			      void *cpu_addr, dma_addr_t handle);
+
+/**
+ * dma_alloc_writecombine - allocate write-combining memory for DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: required memory size
+ * @handle: bus-specific DMA address
+ *
+ * Allocate some uncached, buffered memory for a device for
+ * performing DMA.  This function allocates pages, and will
+ * return the CPU-viewed address, and sets @handle to be the
+ * device-viewed address.
+ */
+extern void *dma_alloc_writecombine(struct device *dev, size_t size,
+				    dma_addr_t *handle, gfp_t gfp);
+
+/**
+ * dma_free_coherent - free memory allocated by dma_alloc_writecombine
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: size of memory originally requested in dma_alloc_writecombine
+ * @cpu_addr: CPU-view address returned from dma_alloc_writecombine
+ * @handle: device-view address returned from dma_alloc_writecombine
+ *
+ * Free (and unmap) a DMA buffer previously allocated by
+ * dma_alloc_writecombine().
+ *
+ * References to memory and mappings associated with cpu_addr/handle
+ * during and after this call executing are illegal.
+ */
+extern void dma_free_writecombine(struct device *dev, size_t size,
+				  void *cpu_addr, dma_addr_t handle);
+
+/**
+ * dma_map_single - map a single buffer for streaming DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @cpu_addr: CPU direct mapped address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Ensure that any data held in the cache is appropriately discarded
+ * or written back.
+ *
+ * The device owns this memory once this call has completed.  The CPU
+ * can regain ownership by calling dma_unmap_single() or dma_sync_single().
+ */
+static inline dma_addr_t
+dma_map_single(struct device *dev, void *cpu_addr, size_t size,
+	       enum dma_data_direction direction)
+{
+	dma_cache_sync(cpu_addr, size, direction);
+	return virt_to_bus(cpu_addr);
+}
+
+/**
+ * dma_unmap_single - unmap a single buffer previously mapped
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @handle: DMA address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Unmap a single streaming mode DMA translation.  The handle and size
+ * must match what was provided in the previous dma_map_single() call.
+ * All other usages are undefined.
+ *
+ * After this call, reads by the CPU to the buffer are guaranteed to see
+ * whatever the device wrote there.
+ */
+static inline void
+dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
+		 enum dma_data_direction direction)
+{
+
+}
+
+/**
+ * dma_map_page - map a portion of a page for streaming DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @page: page that buffer resides in
+ * @offset: offset into page for start of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Ensure that any data held in the cache is appropriately discarded
+ * or written back.
+ *
+ * The device owns this memory once this call has completed.  The CPU
+ * can regain ownership by calling dma_unmap_page() or dma_sync_single().
+ */
+static inline dma_addr_t
+dma_map_page(struct device *dev, struct page *page,
+	     unsigned long offset, size_t size,
+	     enum dma_data_direction direction)
+{
+	return dma_map_single(dev, page_address(page) + offset,
+			      size, direction);
+}
+
+/**
+ * dma_unmap_page - unmap a buffer previously mapped through dma_map_page()
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @handle: DMA address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Unmap a single streaming mode DMA translation.  The handle and size
+ * must match what was provided in the previous dma_map_single() call.
+ * All other usages are undefined.
+ *
+ * After this call, reads by the CPU to the buffer are guaranteed to see
+ * whatever the device wrote there.
+ */
+static inline void
+dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
+	       enum dma_data_direction direction)
+{
+	dma_unmap_single(dev, dma_address, size, direction);
+}
+
+/**
+ * dma_map_sg - map a set of SG buffers for streaming mode DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @sg: list of buffers
+ * @nents: number of buffers to map
+ * @dir: DMA transfer direction
+ *
+ * Map a set of buffers described by scatterlist in streaming
+ * mode for DMA.  This is the scatter-gather version of the
+ * above pci_map_single interface.  Here the scatter gather list
+ * elements are each tagged with the appropriate dma address
+ * and length.  They are obtained via sg_dma_{address,length}(SG).
+ *
+ * NOTE: An implementation may be able to use a smaller number of
+ *       DMA address/length pairs than there are SG table elements.
+ *       (for example via virtual mapping capabilities)
+ *       The routine returns the number of addr/length pairs actually
+ *       used, at most nents.
+ *
+ * Device ownership issues as mentioned above for pci_map_single are
+ * the same here.
+ */
+static inline int
+dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
+	   enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++) {
+		char *virt;
+
+		sg[i].dma_address = page_to_bus(sg[i].page) + sg[i].offset;
+		virt = page_address(sg[i].page) + sg[i].offset;
+		dma_cache_sync(virt, sg[i].length, direction);
+	}
+
+	return nents;
+}
+
+/**
+ * dma_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @sg: list of buffers
+ * @nents: number of buffers to map
+ * @dir: DMA transfer direction
+ *
+ * Unmap a set of streaming mode DMA translations.
+ * Again, CPU read rules concerning calls here are the same as for
+ * pci_unmap_single() above.
+ */
+static inline void
+dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
+	     enum dma_data_direction direction)
+{
+
+}
+
+/**
+ * dma_sync_single_for_cpu
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @handle: DMA address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Make physical memory consistent for a single streaming mode DMA
+ * translation after a transfer.
+ *
+ * If you perform a dma_map_single() but wish to interrogate the
+ * buffer using the cpu, yet do not wish to teardown the DMA mapping,
+ * you must call this function before doing so.  At the next point you
+ * give the DMA address back to the card, you must first perform a
+ * dma_sync_single_for_device, and then the device again owns the
+ * buffer.
+ */
+static inline void
+dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
+			size_t size, enum dma_data_direction direction)
+{
+	dma_cache_sync(bus_to_virt(dma_handle), size, direction);
+}
+
+static inline void
+dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
+			   size_t size, enum dma_data_direction direction)
+{
+	dma_cache_sync(bus_to_virt(dma_handle), size, direction);
+}
+
+/**
+ * dma_sync_sg_for_cpu
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @sg: list of buffers
+ * @nents: number of buffers to map
+ * @dir: DMA transfer direction
+ *
+ * Make physical memory consistent for a set of streaming
+ * mode DMA translations after a transfer.
+ *
+ * The same as dma_sync_single_for_* but for a scatter-gather list,
+ * same rules and usage.
+ */
+static inline void
+dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
+		    int nents, enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++) {
+		dma_cache_sync(page_address(sg[i].page) + sg[i].offset,
+			       sg[i].length, direction);
+	}
+}
+
+static inline void
+dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
+		       int nents, enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++) {
+		dma_cache_sync(page_address(sg[i].page) + sg[i].offset,
+			       sg[i].length, direction);
+	}
+}
+
+/* Now for the API extensions over the pci_ one */
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+static inline int dma_is_consistent(dma_addr_t dma_addr)
+{
+	return 1;
+}
+
+static inline int dma_get_cache_alignment(void)
+{
+	return boot_cpu_data.dcache.linesz;
+}
+
+#endif /* __ASM_AVR32_DMA_MAPPING_H */
diff --git a/include/asm-avr32/dma.h b/include/asm-avr32/dma.h
new file mode 100644
index 000000000000..9e91205590ac
--- /dev/null
+++ b/include/asm-avr32/dma.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_AVR32_DMA_H
+#define __ASM_AVR32_DMA_H
+
+/* The maximum address that we can perform a DMA transfer to on this platform.
+ * Not really applicable to AVR32, but some functions need it. */
+#define MAX_DMA_ADDRESS		0xffffffff
+
+#endif /* __ASM_AVR32_DMA_H */
diff --git a/include/asm-avr32/elf.h b/include/asm-avr32/elf.h
new file mode 100644
index 000000000000..d334b4994d2d
--- /dev/null
+++ b/include/asm-avr32/elf.h
@@ -0,0 +1,110 @@
+#ifndef __ASM_AVR32_ELF_H
+#define __ASM_AVR32_ELF_H
+
+/* AVR32 relocation numbers */
+#define R_AVR32_NONE		0
+#define R_AVR32_32		1
+#define R_AVR32_16		2
+#define R_AVR32_8		3
+#define R_AVR32_32_PCREL	4
+#define R_AVR32_16_PCREL	5
+#define R_AVR32_8_PCREL		6
+#define R_AVR32_DIFF32		7
+#define R_AVR32_DIFF16		8
+#define R_AVR32_DIFF8		9
+#define R_AVR32_GOT32		10
+#define R_AVR32_GOT16		11
+#define R_AVR32_GOT8		12
+#define R_AVR32_21S		13
+#define R_AVR32_16U		14
+#define R_AVR32_16S		15
+#define R_AVR32_8S		16
+#define R_AVR32_8S_EXT		17
+#define R_AVR32_22H_PCREL	18
+#define R_AVR32_18W_PCREL	19
+#define R_AVR32_16B_PCREL	20
+#define R_AVR32_16N_PCREL	21
+#define R_AVR32_14UW_PCREL	22
+#define R_AVR32_11H_PCREL	23
+#define R_AVR32_10UW_PCREL	24
+#define R_AVR32_9H_PCREL	25
+#define R_AVR32_9UW_PCREL	26
+#define R_AVR32_HI16		27
+#define R_AVR32_LO16		28
+#define R_AVR32_GOTPC		29
+#define R_AVR32_GOTCALL		30
+#define R_AVR32_LDA_GOT		31
+#define R_AVR32_GOT21S		32
+#define R_AVR32_GOT18SW		33
+#define R_AVR32_GOT16S		34
+#define R_AVR32_GOT7UW		35
+#define R_AVR32_32_CPENT	36
+#define R_AVR32_CPCALL		37
+#define R_AVR32_16_CP		38
+#define R_AVR32_9W_CP		39
+#define R_AVR32_RELATIVE	40
+#define R_AVR32_GLOB_DAT	41
+#define R_AVR32_JMP_SLOT	42
+#define R_AVR32_ALIGN		43
+
+/*
+ * ELF register definitions..
+ */
+
+#include <asm/ptrace.h>
+#include <asm/user.h>
+
+typedef unsigned long elf_greg_t;
+
+#define ELF_NGREG (sizeof (struct pt_regs) / sizeof (elf_greg_t))
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+typedef struct user_fpu_struct elf_fpregset_t;
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x) ( (x)->e_machine == EM_AVR32 )
+
+/*
+ * These are used to set parameters in the core dumps.
+ */
+#define ELF_CLASS	ELFCLASS32
+#ifdef __LITTLE_ENDIAN__
+#define ELF_DATA	ELFDATA2LSB
+#else
+#define ELF_DATA	ELFDATA2MSB
+#endif
+#define ELF_ARCH	EM_AVR32
+
+#define USE_ELF_CORE_DUMP
+#define ELF_EXEC_PAGESIZE	4096
+
+/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
+   use of this is to invoke "./ld.so someprog" to test out a new version of
+   the loader.  We need to make sure that it is out of the way of the program
+   that it will "exec", and that there is sufficient room for the brk.  */
+
+#define ELF_ET_DYN_BASE         (2 * TASK_SIZE / 3)
+
+
+/* This yields a mask that user programs can use to figure out what
+   instruction set this CPU supports.  This could be done in user space,
+   but it's not easy, and we've already done it here.  */
+
+#define ELF_HWCAP	(0)
+
+/* This yields a string that ld.so will use to load implementation
+   specific libraries for optimization.  This is more specific in
+   intent than poking at uname or /proc/cpuinfo.
+
+   For the moment, we have only optimizations for the Intel generations,
+   but that could change... */
+
+#define ELF_PLATFORM  (NULL)
+
+#ifdef __KERNEL__
+#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
+#endif
+
+#endif /* __ASM_AVR32_ELF_H */
diff --git a/include/asm-avr32/emergency-restart.h b/include/asm-avr32/emergency-restart.h
new file mode 100644
index 000000000000..3e7e014776ba
--- /dev/null
+++ b/include/asm-avr32/emergency-restart.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_EMERGENCY_RESTART_H
+#define __ASM_AVR32_EMERGENCY_RESTART_H
+
+#include <asm-generic/emergency-restart.h>
+
+#endif /* __ASM_AVR32_EMERGENCY_RESTART_H */
diff --git a/include/asm-avr32/errno.h b/include/asm-avr32/errno.h
new file mode 100644
index 000000000000..558a7249f06d
--- /dev/null
+++ b/include/asm-avr32/errno.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_ERRNO_H
+#define __ASM_AVR32_ERRNO_H
+
+#include <asm-generic/errno.h>
+
+#endif /* __ASM_AVR32_ERRNO_H */
diff --git a/include/asm-avr32/fcntl.h b/include/asm-avr32/fcntl.h
new file mode 100644
index 000000000000..14c0c4402b11
--- /dev/null
+++ b/include/asm-avr32/fcntl.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_FCNTL_H
+#define __ASM_AVR32_FCNTL_H
+
+#include <asm-generic/fcntl.h>
+
+#endif /* __ASM_AVR32_FCNTL_H */
diff --git a/include/asm-avr32/futex.h b/include/asm-avr32/futex.h
new file mode 100644
index 000000000000..10419f14a68a
--- /dev/null
+++ b/include/asm-avr32/futex.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_FUTEX_H
+#define __ASM_AVR32_FUTEX_H
+
+#include <asm-generic/futex.h>
+
+#endif /* __ASM_AVR32_FUTEX_H */
diff --git a/include/asm-avr32/hardirq.h b/include/asm-avr32/hardirq.h
new file mode 100644
index 000000000000..267354356f60
--- /dev/null
+++ b/include/asm-avr32/hardirq.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_AVR32_HARDIRQ_H
+#define __ASM_AVR32_HARDIRQ_H
+
+#include <linux/threads.h>
+#include <asm/irq.h>
+
+#ifndef __ASSEMBLY__
+
+#include <linux/cache.h>
+
+/* entry.S is sensitive to the offsets of these fields */
+typedef struct {
+	unsigned int __softirq_pending;
+} ____cacheline_aligned irq_cpustat_t;
+
+void ack_bad_irq(unsigned int irq);
+
+/* Standard mappings for irq_cpustat_t above */
+#include <linux/irq_cpustat.h>
+
+#endif /* __ASSEMBLY__ */
+
+#define HARDIRQ_BITS	12
+
+/*
+ * The hardirq mask has to be large enough to have
+ * space for potentially all IRQ sources in the system
+ * nesting on a single CPU:
+ */
+#if (1 << HARDIRQ_BITS) < NR_IRQS
+# error HARDIRQ_BITS is too low!
+#endif
+
+#endif /* __ASM_AVR32_HARDIRQ_H */
diff --git a/include/asm-avr32/hw_irq.h b/include/asm-avr32/hw_irq.h
new file mode 100644
index 000000000000..218b0a6bfd1b
--- /dev/null
+++ b/include/asm-avr32/hw_irq.h
@@ -0,0 +1,9 @@
+#ifndef __ASM_AVR32_HW_IRQ_H
+#define __ASM_AVR32_HW_IRQ_H
+
+static inline void hw_resend_irq(struct hw_interrupt_type *h, unsigned int i)
+{
+	/* Nothing to do */
+}
+
+#endif /* __ASM_AVR32_HW_IRQ_H */
diff --git a/include/asm-avr32/intc.h b/include/asm-avr32/intc.h
new file mode 100644
index 000000000000..1ac9ca75e8fd
--- /dev/null
+++ b/include/asm-avr32/intc.h
@@ -0,0 +1,128 @@
+#ifndef __ASM_AVR32_INTC_H
+#define __ASM_AVR32_INTC_H
+
+#include <linux/sysdev.h>
+#include <linux/interrupt.h>
+
+struct irq_controller;
+struct irqaction;
+struct pt_regs;
+
+struct platform_device;
+
+/* Information about the internal interrupt controller */
+struct intc_device {
+	/* ioremapped address of configuration block */
+	void __iomem *regs;
+
+	/* the physical device */
+	struct platform_device *pdev;
+
+	/* Number of interrupt lines per group. */
+	unsigned int irqs_per_group;
+
+	/* The highest group ID + 1 */
+	unsigned int nr_groups;
+
+	/*
+	 * Bitfield indicating which groups are actually in use.  The
+	 * size of the array is
+	 * ceil(group_max / (8 * sizeof(unsigned int))).
+	 */
+	unsigned int group_mask[];
+};
+
+struct irq_controller_class {
+	/*
+	 * A short name identifying this kind of controller.
+	 */
+	const char *typename;
+	/*
+	 * Handle the IRQ.  Must do any necessary acking and masking.
+	 */
+	irqreturn_t (*handle)(int irq, void *dev_id, struct pt_regs *regs);
+	/*
+	 * Register a new IRQ handler.
+	 */
+	int (*setup)(struct irq_controller *ctrl, unsigned int irq,
+		     struct irqaction *action);
+	/*
+	 * Unregister a IRQ handler.
+	 */
+	void (*free)(struct irq_controller *ctrl, unsigned int irq,
+		     void *dev_id);
+	/*
+	 * Mask the IRQ in the interrupt controller.
+	 */
+	void (*mask)(struct irq_controller *ctrl, unsigned int irq);
+	/*
+	 * Unmask the IRQ in the interrupt controller.
+	 */
+	void (*unmask)(struct irq_controller *ctrl, unsigned int irq);
+	/*
+	 * Set the type of the IRQ. See below for possible types.
+	 * Return -EINVAL if a given type is not supported
+	 */
+	int (*set_type)(struct irq_controller *ctrl, unsigned int irq,
+			unsigned int type);
+	/*
+	 * Return the IRQ type currently set
+	 */
+	unsigned int (*get_type)(struct irq_controller *ctrl, unsigned int irq);
+};
+
+struct irq_controller {
+	struct irq_controller_class *class;
+	unsigned int irq_group;
+	unsigned int first_irq;
+	unsigned int nr_irqs;
+	struct list_head list;
+};
+
+struct intc_group_desc {
+	struct irq_controller *ctrl;
+	irqreturn_t (*handle)(int, void *, struct pt_regs *);
+	unsigned long flags;
+	void *dev_id;
+	const char *devname;
+};
+
+/*
+ * The internal interrupt controller.  Defined in board/part-specific
+ * devices.c.
+ * TODO: Should probably be defined per-cpu.
+ */
+extern struct intc_device intc;
+
+extern int request_internal_irq(unsigned int irq,
+				irqreturn_t (*handler)(int, void *, struct pt_regs *),
+				unsigned long irqflags,
+				const char *devname, void *dev_id);
+extern void free_internal_irq(unsigned int irq);
+
+/* Only used by time_init() */
+extern int setup_internal_irq(unsigned int irq, struct intc_group_desc *desc);
+
+/*
+ * Set interrupt priority for a given group. `group' can be found by
+ * using irq_to_group(irq). Priority can be from 0 (lowest) to 3
+ * (highest). Higher-priority interrupts will preempt lower-priority
+ * interrupts (unless interrupts are masked globally).
+ *
+ * This function does not check for conflicts within a group.
+ */
+extern int intc_set_priority(unsigned int group,
+			     unsigned int priority);
+
+/*
+ * Returns a bitmask of pending interrupts in a group.
+ */
+extern unsigned long intc_get_pending(unsigned int group);
+
+/*
+ * Register a new external interrupt controller.  Returns the first
+ * external IRQ number that is assigned to the new controller.
+ */
+extern int intc_register_controller(struct irq_controller *ctrl);
+
+#endif /* __ASM_AVR32_INTC_H */
diff --git a/include/asm-avr32/io.h b/include/asm-avr32/io.h
new file mode 100644
index 000000000000..2fc8f111dce9
--- /dev/null
+++ b/include/asm-avr32/io.h
@@ -0,0 +1,253 @@
+#ifndef __ASM_AVR32_IO_H
+#define __ASM_AVR32_IO_H
+
+#include <linux/string.h>
+
+#ifdef __KERNEL__
+
+#include <asm/addrspace.h>
+#include <asm/byteorder.h>
+
+/* virt_to_phys will only work when address is in P1 or P2 */
+static __inline__ unsigned long virt_to_phys(volatile void *address)
+{
+	return PHYSADDR(address);
+}
+
+static __inline__ void * phys_to_virt(unsigned long address)
+{
+	return (void *)P1SEGADDR(address);
+}
+
+#define cached_to_phys(addr)	((unsigned long)PHYSADDR(addr))
+#define uncached_to_phys(addr)	((unsigned long)PHYSADDR(addr))
+#define phys_to_cached(addr)	((void *)P1SEGADDR(addr))
+#define phys_to_uncached(addr)	((void *)P2SEGADDR(addr))
+
+/*
+ * Generic IO read/write.  These perform native-endian accesses.  Note
+ * that some architectures will want to re-define __raw_{read,write}w.
+ */
+extern void __raw_writesb(unsigned int addr, const void *data, int bytelen);
+extern void __raw_writesw(unsigned int addr, const void *data, int wordlen);
+extern void __raw_writesl(unsigned int addr, const void *data, int longlen);
+
+extern void __raw_readsb(unsigned int addr, void *data, int bytelen);
+extern void __raw_readsw(unsigned int addr, void *data, int wordlen);
+extern void __raw_readsl(unsigned int addr, void *data, int longlen);
+
+static inline void writeb(unsigned char b, volatile void __iomem *addr)
+{
+	*(volatile unsigned char __force *)addr = b;
+}
+static inline void writew(unsigned short b, volatile void __iomem *addr)
+{
+	*(volatile unsigned short __force *)addr = b;
+}
+static inline void writel(unsigned int b, volatile void __iomem *addr)
+{
+	*(volatile unsigned int __force *)addr = b;
+}
+#define __raw_writeb writeb
+#define __raw_writew writew
+#define __raw_writel writel
+
+static inline unsigned char readb(const volatile void __iomem *addr)
+{
+	return *(const volatile unsigned char __force *)addr;
+}
+static inline unsigned short readw(const volatile void __iomem *addr)
+{
+	return *(const volatile unsigned short __force *)addr;
+}
+static inline unsigned int readl(const volatile void __iomem *addr)
+{
+	return *(const volatile unsigned int __force *)addr;
+}
+#define __raw_readb readb
+#define __raw_readw readw
+#define __raw_readl readl
+
+#define writesb(p, d, l)	__raw_writesb((unsigned int)p, d, l)
+#define writesw(p, d, l)	__raw_writesw((unsigned int)p, d, l)
+#define writesl(p, d, l)	__raw_writesl((unsigned int)p, d, l)
+
+#define readsb(p, d, l)		__raw_readsb((unsigned int)p, d, l)
+#define readsw(p, d, l)		__raw_readsw((unsigned int)p, d, l)
+#define readsl(p, d, l)		__raw_readsl((unsigned int)p, d, l)
+
+/*
+ * These two are only here because ALSA _thinks_ it needs them...
+ */
+static inline void memcpy_fromio(void * to, const volatile void __iomem *from,
+				 unsigned long count)
+{
+	char *p = to;
+	while (count) {
+		count--;
+		*p = readb(from);
+		p++;
+		from++;
+	}
+}
+
+static inline void  memcpy_toio(volatile void __iomem *to, const void * from,
+				unsigned long count)
+{
+	const char *p = from;
+	while (count) {
+		count--;
+		writeb(*p, to);
+		p++;
+		to++;
+	}
+}
+
+static inline void memset_io(volatile void __iomem *addr, unsigned char val,
+			     unsigned long count)
+{
+	memset((void __force *)addr, val, count);
+}
+
+/*
+ * Bad read/write accesses...
+ */
+extern void __readwrite_bug(const char *fn);
+
+#define IO_SPACE_LIMIT	0xffffffff
+
+/* Convert I/O port address to virtual address */
+#define __io(p)		((void __iomem *)phys_to_uncached(p))
+
+/*
+ *  IO port access primitives
+ *  -------------------------
+ *
+ * The AVR32 doesn't have special IO access instructions; all IO is memory
+ * mapped. Note that these are defined to perform little endian accesses
+ * only. Their primary purpose is to access PCI and ISA peripherals.
+ *
+ * Note that for a big endian machine, this implies that the following
+ * big endian mode connectivity is in place.
+ *
+ * The machine specific io.h include defines __io to translate an "IO"
+ * address to a memory address.
+ *
+ * Note that we prevent GCC re-ordering or caching values in expressions
+ * by introducing sequence points into the in*() definitions.  Note that
+ * __raw_* do not guarantee this behaviour.
+ *
+ * The {in,out}[bwl] macros are for emulating x86-style PCI/ISA IO space.
+ */
+#define outb(v, p)		__raw_writeb(v, __io(p))
+#define outw(v, p)		__raw_writew(cpu_to_le16(v), __io(p))
+#define outl(v, p)		__raw_writel(cpu_to_le32(v), __io(p))
+
+#define inb(p)			__raw_readb(__io(p))
+#define inw(p)			le16_to_cpu(__raw_readw(__io(p)))
+#define inl(p)			le32_to_cpu(__raw_readl(__io(p)))
+
+static inline void __outsb(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		outb(*(u8 *)addr, port);
+		addr++;
+	}
+}
+
+static inline void __insb(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		*(u8 *)addr = inb(port);
+		addr++;
+	}
+}
+
+static inline void __outsw(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		outw(*(u16 *)addr, port);
+		addr += 2;
+	}
+}
+
+static inline void __insw(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		*(u16 *)addr = inw(port);
+		addr += 2;
+	}
+}
+
+static inline void __outsl(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		outl(*(u32 *)addr, port);
+		addr += 4;
+	}
+}
+
+static inline void __insl(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		*(u32 *)addr = inl(port);
+		addr += 4;
+	}
+}
+
+#define outsb(port, addr, count)	__outsb(port, addr, count)
+#define insb(port, addr, count)		__insb(port, addr, count)
+#define outsw(port, addr, count)	__outsw(port, addr, count)
+#define insw(port, addr, count)		__insw(port, addr, count)
+#define outsl(port, addr, count)	__outsl(port, addr, count)
+#define insl(port, addr, count)		__insl(port, addr, count)
+
+extern void __iomem *__ioremap(unsigned long offset, size_t size,
+			       unsigned long flags);
+extern void __iounmap(void __iomem *addr);
+
+/*
+ * ioremap	-   map bus memory into CPU space
+ * @offset	bus address of the memory
+ * @size	size of the resource to map
+ *
+ * ioremap performs a platform specific sequence of operations to make
+ * bus memory CPU accessible via the readb/.../writel functions and
+ * the other mmio helpers. The returned address is not guaranteed to
+ * be usable directly as a virtual address.
+ */
+#define ioremap(offset, size)			\
+	__ioremap((offset), (size), 0)
+
+#define iounmap(addr)				\
+	__iounmap(addr)
+
+#define cached(addr) P1SEGADDR(addr)
+#define uncached(addr) P2SEGADDR(addr)
+
+#define virt_to_bus virt_to_phys
+#define bus_to_virt phys_to_virt
+#define page_to_bus page_to_phys
+#define bus_to_page phys_to_page
+
+#define dma_cache_wback_inv(_start, _size)	\
+	flush_dcache_region(_start, _size)
+#define dma_cache_inv(_start, _size)		\
+	invalidate_dcache_region(_start, _size)
+#define dma_cache_wback(_start, _size)		\
+	clean_dcache_region(_start, _size)
+
+/*
+ * Convert a physical pointer to a virtual kernel pointer for /dev/mem
+ * access
+ */
+#define xlate_dev_mem_ptr(p)    __va(p)
+
+/*
+ * Convert a virtual cached pointer to an uncached pointer
+ */
+#define xlate_dev_kmem_ptr(p)   p
+
+#endif /* __KERNEL__ */
+
+#endif /* __ASM_AVR32_IO_H */
diff --git a/include/asm-avr32/ioctl.h b/include/asm-avr32/ioctl.h
new file mode 100644
index 000000000000..c8472c1398ef
--- /dev/null
+++ b/include/asm-avr32/ioctl.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_IOCTL_H
+#define __ASM_AVR32_IOCTL_H
+
+#include <asm-generic/ioctl.h>
+
+#endif /* __ASM_AVR32_IOCTL_H */
diff --git a/include/asm-avr32/ioctls.h b/include/asm-avr32/ioctls.h
new file mode 100644
index 000000000000..0500426b7186
--- /dev/null
+++ b/include/asm-avr32/ioctls.h
@@ -0,0 +1,83 @@
+#ifndef __ASM_AVR32_IOCTLS_H
+#define __ASM_AVR32_IOCTLS_H
+
+#include <asm/ioctl.h>
+
+/* 0x54 is just a magic number to make these relatively unique ('T') */
+
+#define TCGETS		0x5401
+#define TCSETS		0x5402 /* Clashes with SNDCTL_TMR_START sound ioctl */
+#define TCSETSW		0x5403
+#define TCSETSF		0x5404
+#define TCGETA		0x5405
+#define TCSETA		0x5406
+#define TCSETAW		0x5407
+#define TCSETAF		0x5408
+#define TCSBRK		0x5409
+#define TCXONC		0x540A
+#define TCFLSH		0x540B
+#define TIOCEXCL	0x540C
+#define TIOCNXCL	0x540D
+#define TIOCSCTTY	0x540E
+#define TIOCGPGRP	0x540F
+#define TIOCSPGRP	0x5410
+#define TIOCOUTQ	0x5411
+#define TIOCSTI		0x5412
+#define TIOCGWINSZ	0x5413
+#define TIOCSWINSZ	0x5414
+#define TIOCMGET	0x5415
+#define TIOCMBIS	0x5416
+#define TIOCMBIC	0x5417
+#define TIOCMSET	0x5418
+#define TIOCGSOFTCAR	0x5419
+#define TIOCSSOFTCAR	0x541A
+#define FIONREAD	0x541B
+#define TIOCINQ		FIONREAD
+#define TIOCLINUX	0x541C
+#define TIOCCONS	0x541D
+#define TIOCGSERIAL	0x541E
+#define TIOCSSERIAL	0x541F
+#define TIOCPKT		0x5420
+#define FIONBIO		0x5421
+#define TIOCNOTTY	0x5422
+#define TIOCSETD	0x5423
+#define TIOCGETD	0x5424
+#define TCSBRKP		0x5425	/* Needed for POSIX tcsendbreak() */
+/* #define TIOCTTYGSTRUCT 0x5426 - Former debugging-only ioctl */
+#define TIOCSBRK	0x5427  /* BSD compatibility */
+#define TIOCCBRK	0x5428  /* BSD compatibility */
+#define TIOCGSID	0x5429  /* Return the session ID of FD */
+#define TIOCGPTN	_IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
+#define TIOCSPTLCK	_IOW('T',0x31, int)  /* Lock/unlock Pty */
+
+#define FIONCLEX	0x5450
+#define FIOCLEX		0x5451
+#define FIOASYNC	0x5452
+#define TIOCSERCONFIG	0x5453
+#define TIOCSERGWILD	0x5454
+#define TIOCSERSWILD	0x5455
+#define TIOCGLCKTRMIOS	0x5456
+#define TIOCSLCKTRMIOS	0x5457
+#define TIOCSERGSTRUCT	0x5458 /* For debugging only */
+#define TIOCSERGETLSR   0x5459 /* Get line status register */
+#define TIOCSERGETMULTI 0x545A /* Get multiport config  */
+#define TIOCSERSETMULTI 0x545B /* Set multiport config */
+
+#define TIOCMIWAIT	0x545C	/* wait for a change on serial input line(s) */
+#define TIOCGICOUNT	0x545D	/* read serial port inline interrupt counts */
+#define TIOCGHAYESESP   0x545E  /* Get Hayes ESP configuration */
+#define TIOCSHAYESESP   0x545F  /* Set Hayes ESP configuration */
+#define FIOQSIZE	0x5460
+
+/* Used for packet mode */
+#define TIOCPKT_DATA		 0
+#define TIOCPKT_FLUSHREAD	 1
+#define TIOCPKT_FLUSHWRITE	 2
+#define TIOCPKT_STOP		 4
+#define TIOCPKT_START		 8
+#define TIOCPKT_NOSTOP		16
+#define TIOCPKT_DOSTOP		32
+
+#define TIOCSER_TEMT    0x01	/* Transmitter physically empty */
+
+#endif /* __ASM_AVR32_IOCTLS_H */
diff --git a/include/asm-avr32/ipcbuf.h b/include/asm-avr32/ipcbuf.h
new file mode 100644
index 000000000000..1552c9698f5e
--- /dev/null
+++ b/include/asm-avr32/ipcbuf.h
@@ -0,0 +1,29 @@
+#ifndef __ASM_AVR32_IPCBUF_H
+#define __ASM_AVR32_IPCBUF_H
+
+/*
+* The user_ipc_perm structure for AVR32 architecture.
+* Note extra padding because this structure is passed back and forth
+* between kernel and user space.
+*
+* Pad space is left for:
+* - 32-bit mode_t and seq
+* - 2 miscellaneous 32-bit values
+*/
+
+struct ipc64_perm
+{
+        __kernel_key_t          key;
+        __kernel_uid32_t        uid;
+        __kernel_gid32_t        gid;
+        __kernel_uid32_t        cuid;
+        __kernel_gid32_t        cgid;
+        __kernel_mode_t         mode;
+        unsigned short          __pad1;
+        unsigned short          seq;
+        unsigned short          __pad2;
+        unsigned long           __unused1;
+        unsigned long           __unused2;
+};
+
+#endif /* __ASM_AVR32_IPCBUF_H */
diff --git a/include/asm-avr32/irq.h b/include/asm-avr32/irq.h
new file mode 100644
index 000000000000..f7e725707dd7
--- /dev/null
+++ b/include/asm-avr32/irq.h
@@ -0,0 +1,10 @@
+#ifndef __ASM_AVR32_IRQ_H
+#define __ASM_AVR32_IRQ_H
+
+#define NR_INTERNAL_IRQS	64
+#define NR_EXTERNAL_IRQS	64
+#define NR_IRQS			(NR_INTERNAL_IRQS + NR_EXTERNAL_IRQS)
+
+#define irq_canonicalize(i)	(i)
+
+#endif /* __ASM_AVR32_IOCTLS_H */
diff --git a/include/asm-avr32/irqflags.h b/include/asm-avr32/irqflags.h
new file mode 100644
index 000000000000..93570daac38a
--- /dev/null
+++ b/include/asm-avr32/irqflags.h
@@ -0,0 +1,68 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_IRQFLAGS_H
+#define __ASM_AVR32_IRQFLAGS_H
+
+#include <asm/sysreg.h>
+
+static inline unsigned long __raw_local_save_flags(void)
+{
+	return sysreg_read(SR);
+}
+
+#define raw_local_save_flags(x)					\
+	do { (x) = __raw_local_save_flags(); } while (0)
+
+/*
+ * This will restore ALL status register flags, not only the interrupt
+ * mask flag.
+ *
+ * The empty asm statement informs the compiler of this fact while
+ * also serving as a barrier.
+ */
+static inline void raw_local_irq_restore(unsigned long flags)
+{
+	sysreg_write(SR, flags);
+	asm volatile("" : : : "memory", "cc");
+}
+
+static inline void raw_local_irq_disable(void)
+{
+	asm volatile("ssrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory");
+}
+
+static inline void raw_local_irq_enable(void)
+{
+	asm volatile("csrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory");
+}
+
+static inline int raw_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & SYSREG_BIT(GM)) != 0;
+}
+
+static inline int raw_irqs_disabled(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	return raw_irqs_disabled_flags(flags);
+}
+
+static inline unsigned long __raw_local_irq_save(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	raw_local_irq_disable();
+
+	return flags;
+}
+
+#define raw_local_irq_save(flags)				\
+	do { (flags) = __raw_local_irq_save(); } while (0)
+
+#endif /* __ASM_AVR32_IRQFLAGS_H */
diff --git a/include/asm-avr32/kdebug.h b/include/asm-avr32/kdebug.h
new file mode 100644
index 000000000000..f583b643ffb2
--- /dev/null
+++ b/include/asm-avr32/kdebug.h
@@ -0,0 +1,38 @@
+#ifndef __ASM_AVR32_KDEBUG_H
+#define __ASM_AVR32_KDEBUG_H
+
+#include <linux/notifier.h>
+
+struct pt_regs;
+
+struct die_args {
+	struct pt_regs *regs;
+	int trapnr;
+};
+
+int register_die_notifier(struct notifier_block *nb);
+int unregister_die_notifier(struct notifier_block *nb);
+int register_page_fault_notifier(struct notifier_block *nb);
+int unregister_page_fault_notifier(struct notifier_block *nb);
+extern struct atomic_notifier_head avr32_die_chain;
+
+/* Grossly misnamed. */
+enum die_val {
+	DIE_FAULT,
+	DIE_BREAKPOINT,
+	DIE_SSTEP,
+	DIE_PAGE_FAULT,
+};
+
+static inline int notify_die(enum die_val val, struct pt_regs *regs,
+			     int trap, int sig)
+{
+	struct die_args args = {
+		.regs = regs,
+		.trapnr = trap,
+	};
+
+	return atomic_notifier_call_chain(&avr32_die_chain, val, &args);
+}
+
+#endif /* __ASM_AVR32_KDEBUG_H */
diff --git a/include/asm-avr32/kmap_types.h b/include/asm-avr32/kmap_types.h
new file mode 100644
index 000000000000..b7f5c6870107
--- /dev/null
+++ b/include/asm-avr32/kmap_types.h
@@ -0,0 +1,30 @@
+#ifndef __ASM_AVR32_KMAP_TYPES_H
+#define __ASM_AVR32_KMAP_TYPES_H
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+# define D(n) __KM_FENCE_##n ,
+#else
+# define D(n)
+#endif
+
+enum km_type {
+D(0)	KM_BOUNCE_READ,
+D(1)	KM_SKB_SUNRPC_DATA,
+D(2)	KM_SKB_DATA_SOFTIRQ,
+D(3)	KM_USER0,
+D(4)	KM_USER1,
+D(5)	KM_BIO_SRC_IRQ,
+D(6)	KM_BIO_DST_IRQ,
+D(7)	KM_PTE0,
+D(8)	KM_PTE1,
+D(9)	KM_PTE2,
+D(10)	KM_IRQ0,
+D(11)	KM_IRQ1,
+D(12)	KM_SOFTIRQ0,
+D(13)	KM_SOFTIRQ1,
+D(14)	KM_TYPE_NR
+};
+
+#undef D
+
+#endif /* __ASM_AVR32_KMAP_TYPES_H */
diff --git a/include/asm-avr32/kprobes.h b/include/asm-avr32/kprobes.h
new file mode 100644
index 000000000000..09a5cbe2f896
--- /dev/null
+++ b/include/asm-avr32/kprobes.h
@@ -0,0 +1,34 @@
+/*
+ * Kernel Probes (KProbes)
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ * Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_KPROBES_H
+#define __ASM_AVR32_KPROBES_H
+
+#include <linux/types.h>
+
+typedef u16	kprobe_opcode_t;
+#define BREAKPOINT_INSTRUCTION	0xd673	/* breakpoint */
+#define MAX_INSN_SIZE		2
+
+#define ARCH_INACTIVE_KPROBE_COUNT 1
+
+#define arch_remove_kprobe(p)	do { } while (0)
+
+/* Architecture specific copy of original instruction */
+struct arch_specific_insn {
+	kprobe_opcode_t	insn[MAX_INSN_SIZE];
+};
+
+extern int kprobe_exceptions_notify(struct notifier_block *self,
+				    unsigned long val, void *data);
+
+#define flush_insn_slot(p)	do { } while (0)
+
+#endif /* __ASM_AVR32_KPROBES_H */
diff --git a/include/asm-avr32/linkage.h b/include/asm-avr32/linkage.h
new file mode 100644
index 000000000000..f7b285e910d4
--- /dev/null
+++ b/include/asm-avr32/linkage.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_LINKAGE_H
+#define __ASM_LINKAGE_H
+
+#define __ALIGN .balign 2
+#define __ALIGN_STR ".balign 2"
+
+#endif /* __ASM_LINKAGE_H */
diff --git a/include/asm-avr32/local.h b/include/asm-avr32/local.h
new file mode 100644
index 000000000000..1c1619694da3
--- /dev/null
+++ b/include/asm-avr32/local.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_LOCAL_H
+#define __ASM_AVR32_LOCAL_H
+
+#include <asm-generic/local.h>
+
+#endif /* __ASM_AVR32_LOCAL_H */
diff --git a/include/asm-avr32/mach/serial_at91.h b/include/asm-avr32/mach/serial_at91.h
new file mode 100644
index 000000000000..1290bb32802d
--- /dev/null
+++ b/include/asm-avr32/mach/serial_at91.h
@@ -0,0 +1,33 @@
+/*
+ *  linux/include/asm-arm/mach/serial_at91.h
+ *
+ *  Based on serial_sa1100.h  by Nicolas Pitre
+ *
+ *  Copyright (C) 2002 ATMEL Rousset
+ *
+ *  Low level machine dependent UART functions.
+ */
+
+struct uart_port;
+
+/*
+ * This is a temporary structure for registering these
+ * functions; it is intended to be discarded after boot.
+ */
+struct at91_port_fns {
+	void	(*set_mctrl)(struct uart_port *, u_int);
+	u_int	(*get_mctrl)(struct uart_port *);
+	void	(*enable_ms)(struct uart_port *);
+	void	(*pm)(struct uart_port *, u_int, u_int);
+	int	(*set_wake)(struct uart_port *, u_int);
+	int	(*open)(struct uart_port *);
+	void	(*close)(struct uart_port *);
+};
+
+#if defined(CONFIG_SERIAL_AT91)
+void at91_register_uart_fns(struct at91_port_fns *fns);
+#else
+#define at91_register_uart_fns(fns) do { } while (0)
+#endif
+
+
diff --git a/include/asm-avr32/mman.h b/include/asm-avr32/mman.h
new file mode 100644
index 000000000000..648f91e7187a
--- /dev/null
+++ b/include/asm-avr32/mman.h
@@ -0,0 +1,17 @@
+#ifndef __ASM_AVR32_MMAN_H__
+#define __ASM_AVR32_MMAN_H__
+
+#include <asm-generic/mman.h>
+
+#define MAP_GROWSDOWN	0x0100		/* stack-like segment */
+#define MAP_DENYWRITE	0x0800		/* ETXTBSY */
+#define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
+#define MAP_LOCKED	0x2000		/* pages are locked */
+#define MAP_NORESERVE	0x4000		/* don't check for reservations */
+#define MAP_POPULATE	0x8000		/* populate (prefault) page tables */
+#define MAP_NONBLOCK	0x10000		/* do not block on IO */
+
+#define MCL_CURRENT	1		/* lock all current mappings */
+#define MCL_FUTURE	2		/* lock all future mappings */
+
+#endif /* __ASM_AVR32_MMAN_H__ */
diff --git a/include/asm-avr32/mmu.h b/include/asm-avr32/mmu.h
new file mode 100644
index 000000000000..60c2d2650d32
--- /dev/null
+++ b/include/asm-avr32/mmu.h
@@ -0,0 +1,10 @@
+#ifndef __ASM_AVR32_MMU_H
+#define __ASM_AVR32_MMU_H
+
+/* Default "unsigned long" context */
+typedef unsigned long mm_context_t;
+
+#define MMU_ITLB_ENTRIES	64
+#define MMU_DTLB_ENTRIES	64
+
+#endif /* __ASM_AVR32_MMU_H */
diff --git a/include/asm-avr32/mmu_context.h b/include/asm-avr32/mmu_context.h
new file mode 100644
index 000000000000..31add1ae8089
--- /dev/null
+++ b/include/asm-avr32/mmu_context.h
@@ -0,0 +1,148 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * ASID handling taken from SH implementation.
+ *   Copyright (C) 1999 Niibe Yutaka
+ *   Copyright (C) 2003 Paul Mundt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_MMU_CONTEXT_H
+#define __ASM_AVR32_MMU_CONTEXT_H
+
+#include <asm/tlbflush.h>
+#include <asm/pgalloc.h>
+#include <asm/sysreg.h>
+
+/*
+ * The MMU "context" consists of two things:
+ *    (a) TLB cache version
+ *    (b) ASID (Address Space IDentifier)
+ */
+#define MMU_CONTEXT_ASID_MASK		0x000000ff
+#define MMU_CONTEXT_VERSION_MASK	0xffffff00
+#define MMU_CONTEXT_FIRST_VERSION       0x00000100
+#define NO_CONTEXT			0
+
+#define MMU_NO_ASID			0x100
+
+/* Virtual Page Number mask */
+#define MMU_VPN_MASK	0xfffff000
+
+/* Cache of MMU context last used */
+extern unsigned long mmu_context_cache;
+
+/*
+ * Get MMU context if needed
+ */
+static inline void
+get_mmu_context(struct mm_struct *mm)
+{
+	unsigned long mc = mmu_context_cache;
+
+	if (((mm->context ^ mc) & MMU_CONTEXT_VERSION_MASK) == 0)
+		/* It's up to date, do nothing */
+		return;
+
+	/* It's old, we need to get new context with new version */
+	mc = ++mmu_context_cache;
+	if (!(mc & MMU_CONTEXT_ASID_MASK)) {
+		/*
+		 * We have exhausted all ASIDs of this version.
+		 * Flush the TLB and start new cycle.
+		 */
+		flush_tlb_all();
+		/*
+		 * Fix version. Note that we avoid version #0
+		 * to distinguish NO_CONTEXT.
+		 */
+		if (!mc)
+			mmu_context_cache = mc = MMU_CONTEXT_FIRST_VERSION;
+	}
+	mm->context = mc;
+}
+
+/*
+ * Initialize the context related info for a new mm_struct
+ * instance.
+ */
+static inline int init_new_context(struct task_struct *tsk,
+				       struct mm_struct *mm)
+{
+	mm->context = NO_CONTEXT;
+	return 0;
+}
+
+/*
+ * Destroy context related info for an mm_struct that is about
+ * to be put to rest.
+ */
+static inline void destroy_context(struct mm_struct *mm)
+{
+	/* Do nothing */
+}
+
+static inline void set_asid(unsigned long asid)
+{
+	/* XXX: We're destroying TLBEHI[8:31] */
+	sysreg_write(TLBEHI, asid & MMU_CONTEXT_ASID_MASK);
+	cpu_sync_pipeline();
+}
+
+static inline unsigned long get_asid(void)
+{
+	unsigned long asid;
+
+	asid = sysreg_read(TLBEHI);
+	return asid & MMU_CONTEXT_ASID_MASK;
+}
+
+static inline void activate_context(struct mm_struct *mm)
+{
+	get_mmu_context(mm);
+	set_asid(mm->context & MMU_CONTEXT_ASID_MASK);
+}
+
+static inline void switch_mm(struct mm_struct *prev,
+				 struct mm_struct *next,
+				 struct task_struct *tsk)
+{
+	if (likely(prev != next)) {
+		unsigned long __pgdir = (unsigned long)next->pgd;
+
+		sysreg_write(PTBR, __pgdir);
+		activate_context(next);
+	}
+}
+
+#define deactivate_mm(tsk,mm) do { } while(0)
+
+#define activate_mm(prev, next) switch_mm((prev), (next), NULL)
+
+static inline void
+enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+}
+
+
+static inline void enable_mmu(void)
+{
+	sysreg_write(MMUCR, (SYSREG_BIT(MMUCR_S)
+			     | SYSREG_BIT(E)
+			     | SYSREG_BIT(MMUCR_I)));
+	nop(); nop(); nop(); nop(); nop(); nop(); nop(); nop();
+
+	if (mmu_context_cache == NO_CONTEXT)
+		mmu_context_cache = MMU_CONTEXT_FIRST_VERSION;
+
+	set_asid(mmu_context_cache & MMU_CONTEXT_ASID_MASK);
+}
+
+static inline void disable_mmu(void)
+{
+	sysreg_write(MMUCR, SYSREG_BIT(MMUCR_S));
+}
+
+#endif /* __ASM_AVR32_MMU_CONTEXT_H */
diff --git a/include/asm-avr32/module.h b/include/asm-avr32/module.h
new file mode 100644
index 000000000000..451444538a1b
--- /dev/null
+++ b/include/asm-avr32/module.h
@@ -0,0 +1,28 @@
+#ifndef __ASM_AVR32_MODULE_H
+#define __ASM_AVR32_MODULE_H
+
+struct mod_arch_syminfo {
+	unsigned long got_offset;
+	int got_initialized;
+};
+
+struct mod_arch_specific {
+	/* Starting offset of got in the module core memory. */
+	unsigned long got_offset;
+	/* Size of the got. */
+	unsigned long got_size;
+	/* Number of symbols in syminfo. */
+	int nsyms;
+	/* Additional symbol information (got offsets). */
+	struct mod_arch_syminfo *syminfo;
+};
+
+#define Elf_Shdr		Elf32_Shdr
+#define Elf_Sym			Elf32_Sym
+#define Elf_Ehdr		Elf32_Ehdr
+
+#define MODULE_PROC_FAMILY "AVR32v1"
+
+#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
+
+#endif /* __ASM_AVR32_MODULE_H */
diff --git a/include/asm-avr32/msgbuf.h b/include/asm-avr32/msgbuf.h
new file mode 100644
index 000000000000..ac18bc4da7f7
--- /dev/null
+++ b/include/asm-avr32/msgbuf.h
@@ -0,0 +1,31 @@
+#ifndef __ASM_AVR32_MSGBUF_H
+#define __ASM_AVR32_MSGBUF_H
+
+/*
+ * The msqid64_ds structure for i386 architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ */
+
+struct msqid64_ds {
+	struct ipc64_perm msg_perm;
+	__kernel_time_t msg_stime;	/* last msgsnd time */
+	unsigned long	__unused1;
+	__kernel_time_t msg_rtime;	/* last msgrcv time */
+	unsigned long	__unused2;
+	__kernel_time_t msg_ctime;	/* last change time */
+	unsigned long	__unused3;
+	unsigned long  msg_cbytes;	/* current number of bytes on queue */
+	unsigned long  msg_qnum;	/* number of messages in queue */
+	unsigned long  msg_qbytes;	/* max number of bytes on queue */
+	__kernel_pid_t msg_lspid;	/* pid of last msgsnd */
+	__kernel_pid_t msg_lrpid;	/* last receive pid */
+	unsigned long  __unused4;
+	unsigned long  __unused5;
+};
+
+#endif /* __ASM_AVR32_MSGBUF_H */
diff --git a/include/asm-avr32/mutex.h b/include/asm-avr32/mutex.h
new file mode 100644
index 000000000000..458c1f7fbc18
--- /dev/null
+++ b/include/asm-avr32/mutex.h
@@ -0,0 +1,9 @@
+/*
+ * Pull in the generic implementation for the mutex fastpath.
+ *
+ * TODO: implement optimized primitives instead, or leave the generic
+ * implementation in place, or pick the atomic_xchg() based generic
+ * implementation. (see asm-generic/mutex-xchg.h for details)
+ */
+
+#include <asm-generic/mutex-dec.h>
diff --git a/include/asm-avr32/namei.h b/include/asm-avr32/namei.h
new file mode 100644
index 000000000000..f0a26de06cab
--- /dev/null
+++ b/include/asm-avr32/namei.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_AVR32_NAMEI_H
+#define __ASM_AVR32_NAMEI_H
+
+/* This dummy routine may be changed to something useful */
+#define __emul_prefix() NULL
+
+#endif /* __ASM_AVR32_NAMEI_H */
diff --git a/include/asm-avr32/numnodes.h b/include/asm-avr32/numnodes.h
new file mode 100644
index 000000000000..0b864d7ce330
--- /dev/null
+++ b/include/asm-avr32/numnodes.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_AVR32_NUMNODES_H
+#define __ASM_AVR32_NUMNODES_H
+
+/* Max 4 nodes */
+#define NODES_SHIFT	2
+
+#endif /* __ASM_AVR32_NUMNODES_H */
diff --git a/include/asm-avr32/ocd.h b/include/asm-avr32/ocd.h
new file mode 100644
index 000000000000..46f73180a127
--- /dev/null
+++ b/include/asm-avr32/ocd.h
@@ -0,0 +1,78 @@
+/*
+ * AVR32 OCD Registers
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_OCD_H
+#define __ASM_AVR32_OCD_H
+
+/* Debug Registers */
+#define DBGREG_DID		  0
+#define DBGREG_DC		  8
+#define DBGREG_DS		 16
+#define DBGREG_RWCS		 28
+#define DBGREG_RWA		 36
+#define DBGREG_RWD		 40
+#define DBGREG_WT		 44
+#define DBGREG_DTC		 52
+#define DBGREG_DTSA0		 56
+#define DBGREG_DTSA1		 60
+#define DBGREG_DTEA0		 72
+#define DBGREG_DTEA1		 76
+#define DBGREG_BWC0A		 88
+#define DBGREG_BWC0B		 92
+#define DBGREG_BWC1A		 96
+#define DBGREG_BWC1B		100
+#define DBGREG_BWC2A		104
+#define DBGREG_BWC2B		108
+#define DBGREG_BWC3A		112
+#define DBGREG_BWC3B		116
+#define DBGREG_BWA0A		120
+#define DBGREG_BWA0B		124
+#define DBGREG_BWA1A		128
+#define DBGREG_BWA1B		132
+#define DBGREG_BWA2A		136
+#define DBGREG_BWA2B		140
+#define DBGREG_BWA3A		144
+#define DBGREG_BWA3B		148
+#define DBGREG_BWD3A		153
+#define DBGREG_BWD3B		156
+
+#define DBGREG_PID		284
+
+#define SABAH_OCD		0x01
+#define SABAH_ICACHE		0x02
+#define SABAH_MEM_CACHED	0x04
+#define SABAH_MEM_UNCACHED	0x05
+
+/* Fields in the Development Control register */
+#define DC_SS_BIT		8
+
+#define DC_SS			(1 <<  DC_SS_BIT)
+#define DC_DBE			(1 << 13)
+#define DC_RID			(1 << 27)
+#define DC_ORP			(1 << 28)
+#define DC_MM			(1 << 29)
+#define DC_RES			(1 << 30)
+
+/* Fields in the Development Status register */
+#define DS_SSS			(1 <<  0)
+#define DS_SWB			(1 <<  1)
+#define DS_HWB			(1 <<  2)
+#define DS_BP_SHIFT		8
+#define DS_BP_MASK		(0xff << DS_BP_SHIFT)
+
+#define __mfdr(addr)							\
+({									\
+	register unsigned long value;					\
+	asm volatile("mfdr	%0, %1" : "=r"(value) : "i"(addr));	\
+	value;								\
+})
+#define __mtdr(addr, value)						\
+	asm volatile("mtdr	%0, %1" : : "i"(addr), "r"(value))
+
+#endif /* __ASM_AVR32_OCD_H */
diff --git a/include/asm-avr32/page.h b/include/asm-avr32/page.h
new file mode 100644
index 000000000000..0f630b3e9932
--- /dev/null
+++ b/include/asm-avr32/page.h
@@ -0,0 +1,112 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PAGE_H
+#define __ASM_AVR32_PAGE_H
+
+#ifdef __KERNEL__
+
+/* PAGE_SHIFT determines the page size */
+#define PAGE_SHIFT	12
+#ifdef __ASSEMBLY__
+#define PAGE_SIZE	(1 << PAGE_SHIFT)
+#else
+#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#endif
+#define PAGE_MASK	(~(PAGE_SIZE-1))
+#define PTE_MASK	PAGE_MASK
+
+#ifndef __ASSEMBLY__
+
+#include <asm/addrspace.h>
+
+extern void clear_page(void *to);
+extern void copy_page(void *to, void *from);
+
+#define clear_user_page(page, vaddr, pg)	clear_page(page)
+#define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef struct { unsigned long pte; } pte_t;
+typedef struct { unsigned long pgd; } pgd_t;
+typedef struct { unsigned long pgprot; } pgprot_t;
+
+#define pte_val(x)		((x).pte)
+#define pgd_val(x)		((x).pgd)
+#define pgprot_val(x)		((x).pgprot)
+
+#define __pte(x)		((pte_t) { (x) })
+#define __pgd(x)		((pgd_t) { (x) })
+#define __pgprot(x)		((pgprot_t) { (x) })
+
+/* FIXME: These should be removed soon */
+extern unsigned long memory_start, memory_end;
+
+/* Pure 2^n version of get_order */
+static inline int get_order(unsigned long size)
+{
+	unsigned lz;
+
+	size = (size - 1) >> PAGE_SHIFT;
+	asm("clz %0, %1" : "=r"(lz) : "r"(size));
+	return 32 - lz;
+}
+
+#endif /* !__ASSEMBLY__ */
+
+/* Align the pointer to the (next) page boundary */
+#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
+
+/*
+ * The hardware maps the virtual addresses 0x80000000 -> 0x9fffffff
+ * permanently to the physical addresses 0x00000000 -> 0x1fffffff when
+ * segmentation is enabled. We want to make use of this in order to
+ * minimize TLB pressure.
+ */
+#define PAGE_OFFSET		(0x80000000UL)
+
+/*
+ * ALSA uses virt_to_page() on DMA pages, which I'm not entirely sure
+ * is a good idea. Anyway, we can't simply subtract PAGE_OFFSET here
+ * in that case, so we'll have to mask out the three most significant
+ * bits of the address instead...
+ *
+ * What's the difference between __pa() and virt_to_phys() anyway?
+ */
+#define __pa(x)		PHYSADDR(x)
+#define __va(x)		((void *)(P1SEGADDR(x)))
+
+#define MAP_NR(addr)	(((unsigned long)(addr) - PAGE_OFFSET) >> PAGE_SHIFT)
+
+#define phys_to_page(phys)	(pfn_to_page(phys >> PAGE_SHIFT))
+#define page_to_phys(page)	(page_to_pfn(page) << PAGE_SHIFT)
+
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+
+#define PHYS_PFN_OFFSET		(CONFIG_PHYS_OFFSET >> PAGE_SHIFT)
+
+#define pfn_to_page(pfn)	(mem_map + ((pfn) - PHYS_PFN_OFFSET))
+#define page_to_pfn(page)	((unsigned long)((page) - mem_map) + PHYS_PFN_OFFSET)
+#define pfn_valid(pfn)		((pfn) >= PHYS_PFN_OFFSET && (pfn) < (PHYS_PFN_OFFSET + max_mapnr))
+#endif /* CONFIG_NEED_MULTIPLE_NODES */
+
+#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
+#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+
+#define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE |	\
+				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+/*
+ * Memory above this physical address will be considered highmem.
+ */
+#define HIGHMEM_START		0x20000000UL
+
+#endif /* __KERNEL__ */
+
+#endif /* __ASM_AVR32_PAGE_H */
diff --git a/include/asm-avr32/param.h b/include/asm-avr32/param.h
new file mode 100644
index 000000000000..34bc8d4c3b29
--- /dev/null
+++ b/include/asm-avr32/param.h
@@ -0,0 +1,23 @@
+#ifndef __ASM_AVR32_PARAM_H
+#define __ASM_AVR32_PARAM_H
+
+#ifdef __KERNEL__
+# define HZ		CONFIG_HZ
+# define USER_HZ	100		/* User interfaces are in "ticks" */
+# define CLOCKS_PER_SEC	(USER_HZ)	/* frequency at which times() counts */
+#endif
+
+#ifndef HZ
+# define HZ		100
+#endif
+
+/* TODO: Should be configurable */
+#define EXEC_PAGESIZE	4096
+
+#ifndef NOGROUP
+# define NOGROUP	(-1)
+#endif
+
+#define MAXHOSTNAMELEN	64
+
+#endif /* __ASM_AVR32_PARAM_H */
diff --git a/include/asm-avr32/pci.h b/include/asm-avr32/pci.h
new file mode 100644
index 000000000000..0f5f134b896a
--- /dev/null
+++ b/include/asm-avr32/pci.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_AVR32_PCI_H__
+#define __ASM_AVR32_PCI_H__
+
+/* We don't support PCI yet, but some drivers require this file anyway */
+
+#define PCI_DMA_BUS_IS_PHYS	(1)
+
+#endif /* __ASM_AVR32_PCI_H__ */
diff --git a/include/asm-avr32/percpu.h b/include/asm-avr32/percpu.h
new file mode 100644
index 000000000000..69227b4cd0d4
--- /dev/null
+++ b/include/asm-avr32/percpu.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_PERCPU_H
+#define __ASM_AVR32_PERCPU_H
+
+#include <asm-generic/percpu.h>
+
+#endif /* __ASM_AVR32_PERCPU_H */
diff --git a/include/asm-avr32/pgalloc.h b/include/asm-avr32/pgalloc.h
new file mode 100644
index 000000000000..7492cfb92ced
--- /dev/null
+++ b/include/asm-avr32/pgalloc.h
@@ -0,0 +1,96 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PGALLOC_H
+#define __ASM_AVR32_PGALLOC_H
+
+#include <asm/processor.h>
+#include <linux/threads.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+
+#define pmd_populate_kernel(mm, pmd, pte) \
+	set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
+
+static __inline__ void pmd_populate(struct mm_struct *mm, pmd_t *pmd,
+				    struct page *pte)
+{
+	set_pmd(pmd, __pmd(_PAGE_TABLE + page_to_phys(pte)));
+}
+
+/*
+ * Allocate and free page tables
+ */
+static __inline__ pgd_t *pgd_alloc(struct mm_struct *mm)
+{
+	unsigned int pgd_size = (USER_PTRS_PER_PGD * sizeof(pgd_t));
+	pgd_t *pgd = (pgd_t *)kmalloc(pgd_size, GFP_KERNEL);
+
+	if (pgd)
+		memset(pgd, 0, pgd_size);
+
+	return pgd;
+}
+
+static inline void pgd_free(pgd_t *pgd)
+{
+	kfree(pgd);
+}
+
+static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
+					  unsigned long address)
+{
+	int count = 0;
+	pte_t *pte;
+
+	do {
+		pte = (pte_t *) __get_free_page(GFP_KERNEL | __GFP_REPEAT);
+		if (pte)
+			clear_page(pte);
+		else {
+			current->state = TASK_UNINTERRUPTIBLE;
+			schedule_timeout(HZ);
+		}
+	} while (!pte && (count++ < 10));
+
+	return pte;
+}
+
+static inline struct page *pte_alloc_one(struct mm_struct *mm,
+					 unsigned long address)
+{
+	int count = 0;
+	struct page *pte;
+
+	do {
+		pte = alloc_pages(GFP_KERNEL, 0);
+		if (pte)
+			clear_page(page_address(pte));
+		else {
+			current->state = TASK_UNINTERRUPTIBLE;
+			schedule_timeout(HZ);
+		}
+	} while (!pte && (count++ < 10));
+
+	return pte;
+}
+
+static inline void pte_free_kernel(pte_t *pte)
+{
+	free_page((unsigned long)pte);
+}
+
+static inline void pte_free(struct page *pte)
+{
+	__free_page(pte);
+}
+
+#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
+
+#define check_pgt_cache() do { } while(0)
+
+#endif /* __ASM_AVR32_PGALLOC_H */
diff --git a/include/asm-avr32/pgtable-2level.h b/include/asm-avr32/pgtable-2level.h
new file mode 100644
index 000000000000..425dd567b5b9
--- /dev/null
+++ b/include/asm-avr32/pgtable-2level.h
@@ -0,0 +1,47 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PGTABLE_2LEVEL_H
+#define __ASM_AVR32_PGTABLE_2LEVEL_H
+
+#include <asm-generic/pgtable-nopmd.h>
+
+/*
+ * Traditional 2-level paging structure
+ */
+#define PGDIR_SHIFT	22
+#define PTRS_PER_PGD	1024
+
+#define PTRS_PER_PTE	1024
+
+#ifndef __ASSEMBLY__
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+
+/*
+ * Certain architectures need to do special things when PTEs
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep, pteval)
+
+/*
+ * (pmds are folded into pgds so this doesn't get actually called,
+ * but the define is needed for a generic inline function.)
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+#define pte_pfn(x)		((unsigned long)(((x).pte >> PAGE_SHIFT)))
+#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PGTABLE_2LEVEL_H */
diff --git a/include/asm-avr32/pgtable.h b/include/asm-avr32/pgtable.h
new file mode 100644
index 000000000000..6b8ca9db2bd5
--- /dev/null
+++ b/include/asm-avr32/pgtable.h
@@ -0,0 +1,408 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PGTABLE_H
+#define __ASM_AVR32_PGTABLE_H
+
+#include <asm/addrspace.h>
+
+#ifndef __ASSEMBLY__
+#include <linux/sched.h>
+
+#endif /* !__ASSEMBLY__ */
+
+/*
+ * Use two-level page tables just as the i386 (without PAE)
+ */
+#include <asm/pgtable-2level.h>
+
+/*
+ * The following code might need some cleanup when the values are
+ * final...
+ */
+#define PMD_SIZE	(1UL << PMD_SHIFT)
+#define PMD_MASK	(~(PMD_SIZE-1))
+#define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
+#define PGDIR_MASK	(~(PGDIR_SIZE-1))
+
+#define USER_PTRS_PER_PGD	(TASK_SIZE / PGDIR_SIZE)
+#define FIRST_USER_ADDRESS	0
+
+#define PTE_PHYS_MASK	0x1ffff000
+
+#ifndef __ASSEMBLY__
+extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
+extern void paging_init(void);
+
+/*
+ * ZERO_PAGE is a global shared page that is always zero: used for
+ * zero-mapped memory areas etc.
+ */
+extern struct page *empty_zero_page;
+#define ZERO_PAGE(vaddr) (empty_zero_page)
+
+/*
+ * Just any arbitrary offset to the start of the vmalloc VM area: the
+ * current 8 MiB value just means that there will be a 8 MiB "hole"
+ * after the uncached physical memory (P2 segment) until the vmalloc
+ * area starts. That means that any out-of-bounds memory accesses will
+ * hopefully be caught; we don't know if the end of the P1/P2 segments
+ * are actually used for anything, but it is anyway safer to let the
+ * MMU catch these kinds of errors than to rely on the memory bus.
+ *
+ * A "hole" of the same size is added to the end of the P3 segment as
+ * well. It might seem wasteful to use 16 MiB of virtual address space
+ * on this, but we do have 512 MiB of it...
+ *
+ * The vmalloc() routines leave a hole of 4 KiB between each vmalloced
+ * area for the same reason.
+ */
+#define VMALLOC_OFFSET	(8 * 1024 * 1024)
+#define VMALLOC_START	(P3SEG + VMALLOC_OFFSET)
+#define VMALLOC_END	(P4SEG - VMALLOC_OFFSET)
+#endif /* !__ASSEMBLY__ */
+
+/*
+ * Page flags. Some of these flags are not directly supported by
+ * hardware, so we have to emulate them.
+ */
+#define _TLBEHI_BIT_VALID	9
+#define _TLBEHI_VALID		(1 << _TLBEHI_BIT_VALID)
+
+#define _PAGE_BIT_WT		0  /* W-bit   : write-through */
+#define _PAGE_BIT_DIRTY		1  /* D-bit   : page changed */
+#define _PAGE_BIT_SZ0		2  /* SZ0-bit : Size of page */
+#define _PAGE_BIT_SZ1		3  /* SZ1-bit : Size of page */
+#define _PAGE_BIT_EXECUTE	4  /* X-bit   : execute access allowed */
+#define _PAGE_BIT_RW		5  /* AP0-bit : write access allowed */
+#define _PAGE_BIT_USER		6  /* AP1-bit : user space access allowed */
+#define _PAGE_BIT_BUFFER	7  /* B-bit   : bufferable */
+#define _PAGE_BIT_GLOBAL	8  /* G-bit   : global (ignore ASID) */
+#define _PAGE_BIT_CACHABLE	9  /* C-bit   : cachable */
+
+/* If we drop support for 1K pages, we get two extra bits */
+#define _PAGE_BIT_PRESENT	10
+#define _PAGE_BIT_ACCESSED	11 /* software: page was accessed */
+
+/* The following flags are only valid when !PRESENT */
+#define _PAGE_BIT_FILE		0 /* software: pagecache or swap? */
+
+#define _PAGE_WT		(1 << _PAGE_BIT_WT)
+#define _PAGE_DIRTY		(1 << _PAGE_BIT_DIRTY)
+#define _PAGE_EXECUTE		(1 << _PAGE_BIT_EXECUTE)
+#define _PAGE_RW		(1 << _PAGE_BIT_RW)
+#define _PAGE_USER		(1 << _PAGE_BIT_USER)
+#define _PAGE_BUFFER		(1 << _PAGE_BIT_BUFFER)
+#define _PAGE_GLOBAL		(1 << _PAGE_BIT_GLOBAL)
+#define _PAGE_CACHABLE		(1 << _PAGE_BIT_CACHABLE)
+
+/* Software flags */
+#define _PAGE_ACCESSED		(1 << _PAGE_BIT_ACCESSED)
+#define _PAGE_PRESENT		(1 << _PAGE_BIT_PRESENT)
+#define _PAGE_FILE		(1 << _PAGE_BIT_FILE)
+
+/*
+ * Page types, i.e. sizes. _PAGE_TYPE_NONE corresponds to what is
+ * usually called _PAGE_PROTNONE on other architectures.
+ *
+ * XXX: Find out if _PAGE_PROTNONE is equivalent with !_PAGE_USER. If
+ * so, we can encode all possible page sizes (although we can't really
+ * support 1K pages anyway due to the _PAGE_PRESENT and _PAGE_ACCESSED
+ * bits)
+ *
+ */
+#define _PAGE_TYPE_MASK		((1 << _PAGE_BIT_SZ0) | (1 << _PAGE_BIT_SZ1))
+#define _PAGE_TYPE_NONE		(0 << _PAGE_BIT_SZ0)
+#define _PAGE_TYPE_SMALL	(1 << _PAGE_BIT_SZ0)
+#define _PAGE_TYPE_MEDIUM	(2 << _PAGE_BIT_SZ0)
+#define _PAGE_TYPE_LARGE	(3 << _PAGE_BIT_SZ0)
+
+/*
+ * Mask which drop software flags. We currently can't handle more than
+ * 512 MiB of physical memory, so we can use bits 29-31 for other
+ * stuff.  With a fixed 4K page size, we can use bits 10-11 as well as
+ * bits 2-3 (SZ)
+ */
+#define _PAGE_FLAGS_HARDWARE_MASK	0xfffff3ff
+
+#define _PAGE_FLAGS_CACHE_MASK	(_PAGE_CACHABLE | _PAGE_BUFFER | _PAGE_WT)
+
+/* TODO: Check for saneness */
+/* User-mode page table flags (to be set in a pgd or pmd entry) */
+#define _PAGE_TABLE		(_PAGE_PRESENT | _PAGE_TYPE_SMALL | _PAGE_RW \
+				 | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
+/* Kernel-mode page table flags */
+#define _KERNPG_TABLE		(_PAGE_PRESENT | _PAGE_TYPE_SMALL | _PAGE_RW \
+				 | _PAGE_ACCESSED | _PAGE_DIRTY)
+/* Flags that may be modified by software */
+#define _PAGE_CHG_MASK		(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY \
+				 | _PAGE_FLAGS_CACHE_MASK)
+
+#define _PAGE_FLAGS_READ	(_PAGE_CACHABLE	| _PAGE_BUFFER)
+#define _PAGE_FLAGS_WRITE	(_PAGE_FLAGS_READ | _PAGE_RW | _PAGE_DIRTY)
+
+#define _PAGE_NORMAL(x)	__pgprot((x) | _PAGE_PRESENT | _PAGE_TYPE_SMALL	\
+				 | _PAGE_ACCESSED)
+
+#define PAGE_NONE	(_PAGE_ACCESSED | _PAGE_TYPE_NONE)
+#define PAGE_READ	(_PAGE_FLAGS_READ | _PAGE_USER)
+#define PAGE_EXEC	(_PAGE_FLAGS_READ | _PAGE_EXECUTE | _PAGE_USER)
+#define PAGE_WRITE	(_PAGE_FLAGS_WRITE | _PAGE_USER)
+#define PAGE_KERNEL	_PAGE_NORMAL(_PAGE_FLAGS_WRITE | _PAGE_EXECUTE | _PAGE_GLOBAL)
+#define PAGE_KERNEL_RO	_PAGE_NORMAL(_PAGE_FLAGS_READ | _PAGE_EXECUTE | _PAGE_GLOBAL)
+
+#define _PAGE_P(x)	_PAGE_NORMAL((x) & ~(_PAGE_RW | _PAGE_DIRTY))
+#define _PAGE_S(x)	_PAGE_NORMAL(x)
+
+#define PAGE_COPY	_PAGE_P(PAGE_WRITE | PAGE_READ)
+
+#ifndef __ASSEMBLY__
+/*
+ * The hardware supports flags for write- and execute access. Read is
+ * always allowed if the page is loaded into the TLB, so the "-w-",
+ * "--x" and "-wx" mappings are implemented as "rw-", "r-x" and "rwx",
+ * respectively.
+ *
+ * The "---" case is handled by software; the page will simply not be
+ * loaded into the TLB if the page type is _PAGE_TYPE_NONE.
+ */
+
+#define __P000	__pgprot(PAGE_NONE)
+#define __P001	_PAGE_P(PAGE_READ)
+#define __P010	_PAGE_P(PAGE_WRITE)
+#define __P011	_PAGE_P(PAGE_WRITE | PAGE_READ)
+#define __P100	_PAGE_P(PAGE_EXEC)
+#define __P101	_PAGE_P(PAGE_EXEC | PAGE_READ)
+#define __P110	_PAGE_P(PAGE_EXEC | PAGE_WRITE)
+#define __P111	_PAGE_P(PAGE_EXEC | PAGE_WRITE | PAGE_READ)
+
+#define __S000	__pgprot(PAGE_NONE)
+#define __S001	_PAGE_S(PAGE_READ)
+#define __S010	_PAGE_S(PAGE_WRITE)
+#define __S011	_PAGE_S(PAGE_WRITE | PAGE_READ)
+#define __S100	_PAGE_S(PAGE_EXEC)
+#define __S101	_PAGE_S(PAGE_EXEC | PAGE_READ)
+#define __S110	_PAGE_S(PAGE_EXEC | PAGE_WRITE)
+#define __S111	_PAGE_S(PAGE_EXEC | PAGE_WRITE | PAGE_READ)
+
+#define pte_none(x)	(!pte_val(x))
+#define pte_present(x)	(pte_val(x) & _PAGE_PRESENT)
+
+#define pte_clear(mm,addr,xp)					\
+	do {							\
+		set_pte_at(mm, addr, xp, __pte(0));		\
+	} while (0)
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+static inline int pte_read(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_USER;
+}
+static inline int pte_write(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_RW;
+}
+static inline int pte_exec(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_EXECUTE;
+}
+static inline int pte_dirty(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_DIRTY;
+}
+static inline int pte_young(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_ACCESSED;
+}
+
+/*
+ * The following only work if pte_present() is not true.
+ */
+static inline int pte_file(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_FILE;
+}
+
+/* Mutator functions for PTE bits */
+static inline pte_t pte_rdprotect(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_USER));
+	return pte;
+}
+static inline pte_t pte_wrprotect(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_RW));
+	return pte;
+}
+static inline pte_t pte_exprotect(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_EXECUTE));
+	return pte;
+}
+static inline pte_t pte_mkclean(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY));
+	return pte;
+}
+static inline pte_t pte_mkold(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED));
+	return pte;
+}
+static inline pte_t pte_mkread(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_USER));
+	return pte;
+}
+static inline pte_t pte_mkwrite(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW));
+	return pte;
+}
+static inline pte_t pte_mkexec(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_EXECUTE));
+	return pte;
+}
+static inline pte_t pte_mkdirty(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY));
+	return pte;
+}
+static inline pte_t pte_mkyoung(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED));
+	return pte;
+}
+
+#define pmd_none(x)	(!pmd_val(x))
+#define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
+#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
+#define	pmd_bad(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER))	\
+			 != _KERNPG_TABLE)
+
+/*
+ * Permanent address of a page. We don't support highmem, so this is
+ * trivial.
+ */
+#define pages_to_mb(x)	((x) >> (20-PAGE_SHIFT))
+#define pte_page(x) 	phys_to_page(pte_val(x) & PTE_PHYS_MASK)
+
+/*
+ * Mark the prot value as uncacheable and unbufferable
+ */
+#define pgprot_noncached(prot)						\
+	__pgprot(pgprot_val(prot) & ~(_PAGE_BUFFER | _PAGE_CACHABLE))
+
+/*
+ * Mark the prot value as uncacheable but bufferable
+ */
+#define pgprot_writecombine(prot)					\
+	__pgprot((pgprot_val(prot) & ~_PAGE_CACHABLE) | _PAGE_BUFFER)
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ *
+ * extern pte_t mk_pte(struct page *page, pgprot_t pgprot)
+ */
+#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+	set_pte(&pte, __pte((pte_val(pte) & _PAGE_CHG_MASK)
+			    | pgprot_val(newprot)));
+	return pte;
+}
+
+#define page_pte(page)	page_pte_prot(page, __pgprot(0))
+
+#define pmd_page_vaddr(pmd)					\
+	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+
+#define pmd_page(pmd)	(phys_to_page(pmd_val(pmd)))
+
+/* to find an entry in a page-table-directory. */
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+#define pgd_offset_current(address)				\
+	((pgd_t *)__mfsr(SYSREG_PTBR) + pgd_index(address))
+
+/* to find an entry in a kernel page-table-directory */
+#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+
+/* Find an entry in the third-level page table.. */
+#define pte_index(address)				\
+	((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset(dir, address)					\
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
+#define pte_offset_kernel(dir, address)					\
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
+#define pte_offset_map(dir, address) pte_offset_kernel(dir, address)
+#define pte_offset_map_nested(dir, address) pte_offset_kernel(dir, address)
+#define pte_unmap(pte)		do { } while (0)
+#define pte_unmap_nested(pte)	do { } while (0)
+
+struct vm_area_struct;
+extern void update_mmu_cache(struct vm_area_struct * vma,
+			     unsigned long address, pte_t pte);
+
+/*
+ * Encode and decode a swap entry
+ *
+ * Constraints:
+ *   _PAGE_FILE at bit 0
+ *   _PAGE_TYPE_* at bits 2-3 (for emulating _PAGE_PROTNONE)
+ *   _PAGE_PRESENT at bit 10
+ *
+ * We encode the type into bits 4-9 and offset into bits 11-31. This
+ * gives us a 21 bits offset, or 2**21 * 4K = 8G usable swap space per
+ * device, and 64 possible types.
+ *
+ * NOTE: We should set ZEROs at the position of _PAGE_PRESENT
+ *       and _PAGE_PROTNONE bits
+ */
+#define __swp_type(x)		(((x).val >> 4) & 0x3f)
+#define __swp_offset(x)		((x).val >> 11)
+#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 4) | ((offset) << 11) })
+#define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
+#define __swp_entry_to_pte(x)	((pte_t) { (x).val })
+
+/*
+ * Encode and decode a nonlinear file mapping entry. We have to
+ * preserve _PAGE_FILE and _PAGE_PRESENT here. _PAGE_TYPE_* isn't
+ * necessary, since _PAGE_FILE implies !_PAGE_PROTNONE (?)
+ */
+#define PTE_FILE_MAX_BITS	30
+#define pte_to_pgoff(pte)	(((pte_val(pte) >> 1) & 0x1ff)		\
+				 | ((pte_val(pte) >> 11) << 9))
+#define pgoff_to_pte(off)	((pte_t) { ((((off) & 0x1ff) << 1)	\
+					    | (((off) >> 9) << 11)	\
+					    | _PAGE_FILE) })
+
+typedef pte_t *pte_addr_t;
+
+#define kern_addr_valid(addr)	(1)
+
+#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)	\
+	remap_pfn_range(vma, vaddr, pfn, size, prot)
+
+#define MK_IOSPACE_PFN(space, pfn)	(pfn)
+#define GET_IOSPACE(pfn)		0
+#define GET_PFN(pfn)			(pfn)
+
+/* No page table caches to initialize (?) */
+#define pgtable_cache_init()	do { } while(0)
+
+#include <asm-generic/pgtable.h>
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PGTABLE_H */
diff --git a/include/asm-avr32/poll.h b/include/asm-avr32/poll.h
new file mode 100644
index 000000000000..736e29755dfc
--- /dev/null
+++ b/include/asm-avr32/poll.h
@@ -0,0 +1,27 @@
+#ifndef __ASM_AVR32_POLL_H
+#define __ASM_AVR32_POLL_H
+
+/* These are specified by iBCS2 */
+#define POLLIN		0x0001
+#define POLLPRI		0x0002
+#define POLLOUT		0x0004
+#define POLLERR		0x0008
+#define POLLHUP		0x0010
+#define POLLNVAL	0x0020
+
+/* The rest seem to be more-or-less nonstandard. Check them! */
+#define POLLRDNORM	0x0040
+#define POLLRDBAND	0x0080
+#define POLLWRNORM	0x0100
+#define POLLWRBAND	0x0200
+#define POLLMSG		0x0400
+#define POLLREMOVE	0x1000
+#define POLLRDHUP	0x2000
+
+struct pollfd {
+	int fd;
+	short events;
+	short revents;
+};
+
+#endif /* __ASM_AVR32_POLL_H */
diff --git a/include/asm-avr32/posix_types.h b/include/asm-avr32/posix_types.h
new file mode 100644
index 000000000000..2831b039b349
--- /dev/null
+++ b/include/asm-avr32/posix_types.h
@@ -0,0 +1,129 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_POSIX_TYPES_H
+#define __ASM_AVR32_POSIX_TYPES_H
+
+/*
+ * This file is generally used by user-level software, so you need to
+ * be a little careful about namespace pollution etc.  Also, we cannot
+ * assume GCC is being used.
+ */
+
+typedef unsigned long   __kernel_ino_t;
+typedef unsigned short  __kernel_mode_t;
+typedef unsigned short  __kernel_nlink_t;
+typedef long            __kernel_off_t;
+typedef int             __kernel_pid_t;
+typedef unsigned short  __kernel_ipc_pid_t;
+typedef unsigned int	__kernel_uid_t;
+typedef unsigned int	__kernel_gid_t;
+typedef unsigned long	__kernel_size_t;
+typedef int             __kernel_ssize_t;
+typedef int             __kernel_ptrdiff_t;
+typedef long            __kernel_time_t;
+typedef long            __kernel_suseconds_t;
+typedef long            __kernel_clock_t;
+typedef int             __kernel_timer_t;
+typedef int             __kernel_clockid_t;
+typedef int             __kernel_daddr_t;
+typedef char *          __kernel_caddr_t;
+typedef unsigned short  __kernel_uid16_t;
+typedef unsigned short  __kernel_gid16_t;
+typedef unsigned int    __kernel_uid32_t;
+typedef unsigned int    __kernel_gid32_t;
+
+typedef unsigned short  __kernel_old_uid_t;
+typedef unsigned short  __kernel_old_gid_t;
+typedef unsigned short  __kernel_old_dev_t;
+
+#ifdef __GNUC__
+typedef long long       __kernel_loff_t;
+#endif
+
+typedef struct {
+#if defined(__KERNEL__) || defined(__USE_ALL)
+    int     val[2];
+#else /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+    int     __val[2];
+#endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+} __kernel_fsid_t;
+
+#if defined(__KERNEL__)
+
+#undef  __FD_SET
+static __inline__ void __FD_SET(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+    unsigned long __tmp = __fd / __NFDBITS;
+    unsigned long __rem = __fd % __NFDBITS;
+    __fdsetp->fds_bits[__tmp] |= (1UL<<__rem);
+}
+
+#undef  __FD_CLR
+static __inline__ void __FD_CLR(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+    unsigned long __tmp = __fd / __NFDBITS;
+    unsigned long __rem = __fd % __NFDBITS;
+    __fdsetp->fds_bits[__tmp] &= ~(1UL<<__rem);
+}
+
+
+#undef  __FD_ISSET
+static __inline__ int __FD_ISSET(unsigned long __fd, const __kernel_fd_set *__p)
+{
+    unsigned long __tmp = __fd / __NFDBITS;
+    unsigned long __rem = __fd % __NFDBITS;
+    return (__p->fds_bits[__tmp] & (1UL<<__rem)) != 0;
+}
+
+/*
+ * This will unroll the loop for the normal constant case (8 ints,
+ * for a 256-bit fd_set)
+ */
+#undef  __FD_ZERO
+static __inline__ void __FD_ZERO(__kernel_fd_set *__p)
+{
+    unsigned long *__tmp = __p->fds_bits;
+    int __i;
+
+    if (__builtin_constant_p(__FDSET_LONGS)) {
+        switch (__FDSET_LONGS) {
+            case 16:
+                __tmp[ 0] = 0; __tmp[ 1] = 0;
+                __tmp[ 2] = 0; __tmp[ 3] = 0;
+                __tmp[ 4] = 0; __tmp[ 5] = 0;
+                __tmp[ 6] = 0; __tmp[ 7] = 0;
+                __tmp[ 8] = 0; __tmp[ 9] = 0;
+                __tmp[10] = 0; __tmp[11] = 0;
+                __tmp[12] = 0; __tmp[13] = 0;
+                __tmp[14] = 0; __tmp[15] = 0;
+                return;
+
+            case 8:
+                __tmp[ 0] = 0; __tmp[ 1] = 0;
+                __tmp[ 2] = 0; __tmp[ 3] = 0;
+                __tmp[ 4] = 0; __tmp[ 5] = 0;
+                __tmp[ 6] = 0; __tmp[ 7] = 0;
+                return;
+
+            case 4:
+                __tmp[ 0] = 0; __tmp[ 1] = 0;
+                __tmp[ 2] = 0; __tmp[ 3] = 0;
+                return;
+        }
+    }
+    __i = __FDSET_LONGS;
+    while (__i) {
+        __i--;
+        *__tmp = 0;
+        __tmp++;
+    }
+}
+
+#endif /* defined(__KERNEL__) */
+
+#endif /* __ASM_AVR32_POSIX_TYPES_H */
diff --git a/include/asm-avr32/processor.h b/include/asm-avr32/processor.h
new file mode 100644
index 000000000000..f6913778a45f
--- /dev/null
+++ b/include/asm-avr32/processor.h
@@ -0,0 +1,147 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PROCESSOR_H
+#define __ASM_AVR32_PROCESSOR_H
+
+#include <asm/page.h>
+#include <asm/cache.h>
+
+#define TASK_SIZE	0x80000000
+
+#ifndef __ASSEMBLY__
+
+static inline void *current_text_addr(void)
+{
+	register void *pc asm("pc");
+	return pc;
+}
+
+enum arch_type {
+	ARCH_AVR32A,
+	ARCH_AVR32B,
+	ARCH_MAX
+};
+
+enum cpu_type {
+	CPU_MORGAN,
+	CPU_AT32AP,
+	CPU_MAX
+};
+
+enum tlb_config {
+	TLB_NONE,
+	TLB_SPLIT,
+	TLB_UNIFIED,
+	TLB_INVALID
+};
+
+struct avr32_cpuinfo {
+	struct clk *clk;
+	unsigned long loops_per_jiffy;
+	enum arch_type arch_type;
+	enum cpu_type cpu_type;
+	unsigned short arch_revision;
+	unsigned short cpu_revision;
+	enum tlb_config tlb_config;
+
+	struct cache_info icache;
+	struct cache_info dcache;
+};
+
+extern struct avr32_cpuinfo boot_cpu_data;
+
+#ifdef CONFIG_SMP
+extern struct avr32_cpuinfo cpu_data[];
+#define current_cpu_data cpu_data[smp_processor_id()]
+#else
+#define cpu_data (&boot_cpu_data)
+#define current_cpu_data boot_cpu_data
+#endif
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's
+ */
+#define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
+
+#define cpu_relax()		barrier()
+#define cpu_sync_pipeline()	asm volatile("sub pc, -2" : : : "memory")
+
+struct cpu_context {
+	unsigned long sr;
+	unsigned long pc;
+	unsigned long ksp;	/* Kernel stack pointer */
+	unsigned long r7;
+	unsigned long r6;
+	unsigned long r5;
+	unsigned long r4;
+	unsigned long r3;
+	unsigned long r2;
+	unsigned long r1;
+	unsigned long r0;
+};
+
+/* This struct contains the CPU context as stored by switch_to() */
+struct thread_struct {
+	struct cpu_context cpu_context;
+	unsigned long single_step_addr;
+	u16 single_step_insn;
+};
+
+#define INIT_THREAD {						\
+	.cpu_context = {					\
+		.ksp = sizeof(init_stack) + (long)&init_stack,	\
+	},							\
+}
+
+/*
+ * Do necessary setup to start up a newly executed thread.
+ */
+#define start_thread(regs, new_pc, new_sp)	 \
+	do {					 \
+		set_fs(USER_DS);		 \
+		memset(regs, 0, sizeof(*regs));	 \
+		regs->sr = MODE_USER;		 \
+		regs->pc = new_pc & ~1;		 \
+		regs->sp = new_sp;		 \
+	} while(0)
+
+struct task_struct;
+
+/* Free all resources held by a thread */
+extern void release_thread(struct task_struct *);
+
+/* Create a kernel thread without removing it from tasklists */
+extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
+
+/* Prepare to copy thread state - unlazy all lazy status */
+#define prepare_to_copy(tsk) do { } while(0)
+
+/* Return saved PC of a blocked thread */
+#define thread_saved_pc(tsk)    ((tsk)->thread.cpu_context.pc)
+
+struct pt_regs;
+void show_trace(struct task_struct *task, unsigned long *stack,
+		struct pt_regs *regs);
+
+extern unsigned long get_wchan(struct task_struct *p);
+
+#define KSTK_EIP(tsk)	((tsk)->thread.cpu_context.pc)
+#define KSTK_ESP(tsk)	((tsk)->thread.cpu_context.ksp)
+
+#define ARCH_HAS_PREFETCH
+
+static inline void prefetch(const void *x)
+{
+	const char *c = x;
+	asm volatile("pref %0" : : "r"(c));
+}
+#define PREFETCH_STRIDE	L1_CACHE_BYTES
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PROCESSOR_H */
diff --git a/include/asm-avr32/ptrace.h b/include/asm-avr32/ptrace.h
new file mode 100644
index 000000000000..60f0f19a81f1
--- /dev/null
+++ b/include/asm-avr32/ptrace.h
@@ -0,0 +1,154 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PTRACE_H
+#define __ASM_AVR32_PTRACE_H
+
+#define PTRACE_GETREGS		12
+#define PTRACE_SETREGS		13
+
+/*
+ * Status Register bits
+ */
+#define SR_H		0x40000000
+#define SR_R		0x20000000
+#define SR_J		0x10000000
+#define SR_DM		0x08000000
+#define SR_D		0x04000000
+#define MODE_NMI	0x01c00000
+#define MODE_EXCEPTION	0x01800000
+#define MODE_INT3	0x01400000
+#define MODE_INT2	0x01000000
+#define MODE_INT1	0x00c00000
+#define MODE_INT0	0x00800000
+#define MODE_SUPERVISOR	0x00400000
+#define MODE_USER	0x00000000
+#define MODE_MASK	0x01c00000
+#define SR_EM		0x00200000
+#define SR_I3M		0x00100000
+#define SR_I2M		0x00080000
+#define SR_I1M		0x00040000
+#define SR_I0M		0x00020000
+#define SR_GM		0x00010000
+
+#define SR_H_BIT	30
+#define SR_R_BIT	29
+#define SR_J_BIT	28
+#define SR_DM_BIT	27
+#define SR_D_BIT	26
+#define MODE_SHIFT	22
+#define SR_EM_BIT	21
+#define SR_I3M_BIT	20
+#define SR_I2M_BIT	19
+#define SR_I1M_BIT	18
+#define SR_I0M_BIT	17
+#define SR_GM_BIT	16
+
+/* The user-visible part */
+#define SR_L		0x00000020
+#define SR_Q		0x00000010
+#define SR_V		0x00000008
+#define SR_N		0x00000004
+#define SR_Z		0x00000002
+#define SR_C		0x00000001
+
+#define SR_L_BIT	5
+#define SR_Q_BIT	4
+#define SR_V_BIT	3
+#define SR_N_BIT	2
+#define SR_Z_BIT	1
+#define SR_C_BIT	0
+
+/*
+ * The order is defined by the stmts instruction. r0 is stored first,
+ * so it gets the highest address.
+ *
+ * Registers 0-12 are general-purpose registers (r12 is normally used for
+ * the function return value).
+ * Register 13 is the stack pointer
+ * Register 14 is the link register
+ * Register 15 is the program counter (retrieved from the RAR sysreg)
+ */
+#define FRAME_SIZE_FULL 72
+#define REG_R12_ORIG	68
+#define REG_R0		64
+#define REG_R1		60
+#define REG_R2		56
+#define REG_R3		52
+#define REG_R4		48
+#define REG_R5		44
+#define REG_R6		40
+#define REG_R7		36
+#define REG_R8		32
+#define REG_R9		28
+#define REG_R10		24
+#define REG_R11		20
+#define REG_R12		16
+#define REG_SP		12
+#define REG_LR		 8
+
+#define FRAME_SIZE_MIN	 8
+#define REG_PC		 4
+#define REG_SR		 0
+
+#ifndef __ASSEMBLY__
+struct pt_regs {
+	/* These are always saved */
+	unsigned long sr;
+	unsigned long pc;
+
+	/* These are sometimes saved */
+	unsigned long lr;
+	unsigned long sp;
+	unsigned long r12;
+	unsigned long r11;
+	unsigned long r10;
+	unsigned long r9;
+	unsigned long r8;
+	unsigned long r7;
+	unsigned long r6;
+	unsigned long r5;
+	unsigned long r4;
+	unsigned long r3;
+	unsigned long r2;
+	unsigned long r1;
+	unsigned long r0;
+
+	/* Only saved on system call */
+	unsigned long r12_orig;
+};
+
+#ifdef __KERNEL__
+# define user_mode(regs) (((regs)->sr & MODE_MASK) == MODE_USER)
+extern void show_regs (struct pt_regs *);
+
+static __inline__ int valid_user_regs(struct pt_regs *regs)
+{
+	/*
+	 * Some of the Java bits might be acceptable if/when we
+	 * implement some support for that stuff...
+	 */
+	if ((regs->sr & 0xffff0000) == 0)
+		return 1;
+
+	/*
+	 * Force status register flags to be sane and report this
+	 * illegal behaviour...
+	 */
+	regs->sr &= 0x0000ffff;
+	return 0;
+}
+
+#define instruction_pointer(regs) ((regs)->pc)
+
+#define profile_pc(regs) instruction_pointer(regs)
+
+#endif /* __KERNEL__ */
+
+#endif /* ! __ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PTRACE_H */
diff --git a/include/asm-avr32/resource.h b/include/asm-avr32/resource.h
new file mode 100644
index 000000000000..c6dd101472b1
--- /dev/null
+++ b/include/asm-avr32/resource.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_RESOURCE_H
+#define __ASM_AVR32_RESOURCE_H
+
+#include <asm-generic/resource.h>
+
+#endif /* __ASM_AVR32_RESOURCE_H */
diff --git a/include/asm-avr32/scatterlist.h b/include/asm-avr32/scatterlist.h
new file mode 100644
index 000000000000..bfe7d753423c
--- /dev/null
+++ b/include/asm-avr32/scatterlist.h
@@ -0,0 +1,21 @@
+#ifndef __ASM_AVR32_SCATTERLIST_H
+#define __ASM_AVR32_SCATTERLIST_H
+
+struct scatterlist {
+    struct page		*page;
+    unsigned int	offset;
+    dma_addr_t		dma_address;
+    unsigned int	length;
+};
+
+/* These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
+#define sg_dma_len(sg)		((sg)->length)
+
+#define ISA_DMA_THRESHOLD (0xffffffff)
+
+#endif /* __ASM_AVR32_SCATTERLIST_H */
diff --git a/include/asm-avr32/sections.h b/include/asm-avr32/sections.h
new file mode 100644
index 000000000000..aa14252e4181
--- /dev/null
+++ b/include/asm-avr32/sections.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_SECTIONS_H
+#define __ASM_AVR32_SECTIONS_H
+
+#include <asm-generic/sections.h>
+
+#endif /* __ASM_AVR32_SECTIONS_H */
diff --git a/include/asm-avr32/semaphore.h b/include/asm-avr32/semaphore.h
new file mode 100644
index 000000000000..ef99ddccc10c
--- /dev/null
+++ b/include/asm-avr32/semaphore.h
@@ -0,0 +1,109 @@
+/*
+ * SMP- and interrupt-safe semaphores.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Based on include/asm-i386/semaphore.h
+ *   Copyright (C) 1996 Linus Torvalds
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SEMAPHORE_H
+#define __ASM_AVR32_SEMAPHORE_H
+
+#include <linux/linkage.h>
+
+#include <asm/system.h>
+#include <asm/atomic.h>
+#include <linux/wait.h>
+#include <linux/rwsem.h>
+
+struct semaphore {
+	atomic_t count;
+	int sleepers;
+	wait_queue_head_t wait;
+};
+
+#define __SEMAPHORE_INITIALIZER(name, n)				\
+{									\
+	.count		= ATOMIC_INIT(n),				\
+	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
+}
+
+#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
+	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
+
+#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
+#define DECLARE_MUTEX_LOCKED(name) __DECLARE_SEMAPHORE_GENERIC(name,0)
+
+static inline void sema_init (struct semaphore *sem, int val)
+{
+	atomic_set(&sem->count, val);
+	sem->sleepers = 0;
+	init_waitqueue_head(&sem->wait);
+}
+
+static inline void init_MUTEX (struct semaphore *sem)
+{
+	sema_init(sem, 1);
+}
+
+static inline void init_MUTEX_LOCKED (struct semaphore *sem)
+{
+	sema_init(sem, 0);
+}
+
+void __down(struct semaphore * sem);
+int  __down_interruptible(struct semaphore * sem);
+void __up(struct semaphore * sem);
+
+/*
+ * This is ugly, but we want the default case to fall through.
+ * "__down_failed" is a special asm handler that calls the C
+ * routine that actually waits. See arch/i386/kernel/semaphore.c
+ */
+static inline void down(struct semaphore * sem)
+{
+	might_sleep();
+	if (unlikely(atomic_dec_return (&sem->count) < 0))
+		__down (sem);
+}
+
+/*
+ * Interruptible try to acquire a semaphore.  If we obtained
+ * it, return zero.  If we were interrupted, returns -EINTR
+ */
+static inline int down_interruptible(struct semaphore * sem)
+{
+	int ret = 0;
+
+	might_sleep();
+	if (unlikely(atomic_dec_return (&sem->count) < 0))
+		ret = __down_interruptible (sem);
+	return ret;
+}
+
+/*
+ * Non-blockingly attempt to down() a semaphore.
+ * Returns zero if we acquired it
+ */
+static inline int down_trylock(struct semaphore * sem)
+{
+	return atomic_dec_if_positive(&sem->count) < 0;
+}
+
+/*
+ * Note! This is subtle. We jump to wake people up only if
+ * the semaphore was negative (== somebody was waiting on it).
+ * The default case (no contention) will result in NO
+ * jumps for both down() and up().
+ */
+static inline void up(struct semaphore * sem)
+{
+	if (unlikely(atomic_inc_return (&sem->count) <= 0))
+		__up (sem);
+}
+
+#endif /*__ASM_AVR32_SEMAPHORE_H */
diff --git a/include/asm-avr32/sembuf.h b/include/asm-avr32/sembuf.h
new file mode 100644
index 000000000000..e472216e0c97
--- /dev/null
+++ b/include/asm-avr32/sembuf.h
@@ -0,0 +1,25 @@
+#ifndef __ASM_AVR32_SEMBUF_H
+#define __ASM_AVR32_SEMBUF_H
+
+/*
+* The semid64_ds structure for AVR32 architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ */
+
+struct semid64_ds {
+        struct ipc64_perm sem_perm;             /* permissions .. see ipc.h */
+        __kernel_time_t sem_otime;              /* last semop time */
+        unsigned long   __unused1;
+        __kernel_time_t sem_ctime;              /* last change time */
+        unsigned long   __unused2;
+        unsigned long   sem_nsems;              /* no. of semaphores in array */
+        unsigned long   __unused3;
+        unsigned long   __unused4;
+};
+
+#endif /* __ASM_AVR32_SEMBUF_H */
diff --git a/include/asm-avr32/setup.h b/include/asm-avr32/setup.h
new file mode 100644
index 000000000000..10193da4113b
--- /dev/null
+++ b/include/asm-avr32/setup.h
@@ -0,0 +1,141 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/include/asm-arm/setup.h
+ *   Copyright (C) 1997-1999 Russel King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SETUP_H__
+#define __ASM_AVR32_SETUP_H__
+
+#define COMMAND_LINE_SIZE 256
+
+/* Magic number indicating that a tag table is present */
+#define ATAG_MAGIC	0xa2a25441
+
+#ifndef __ASSEMBLY__
+
+/*
+ * Generic memory range, used by several tags.
+ *
+ *   addr is always physical.
+ *   size is measured in bytes.
+ *   next is for use by the OS, e.g. for grouping regions into
+ *        linked lists.
+ */
+struct tag_mem_range {
+	u32			addr;
+	u32			size;
+	struct tag_mem_range *	next;
+};
+
+/* The list ends with an ATAG_NONE node. */
+#define ATAG_NONE	0x00000000
+
+struct tag_header {
+	u32 size;
+	u32 tag;
+};
+
+/* The list must start with an ATAG_CORE node */
+#define ATAG_CORE	0x54410001
+
+struct tag_core {
+	u32 flags;
+	u32 pagesize;
+	u32 rootdev;
+};
+
+/* it is allowed to have multiple ATAG_MEM nodes */
+#define ATAG_MEM	0x54410002
+/* ATAG_MEM uses tag_mem_range */
+
+/* command line: \0 terminated string */
+#define ATAG_CMDLINE	0x54410003
+
+struct tag_cmdline {
+	char	cmdline[1];	/* this is the minimum size */
+};
+
+/* Ramdisk image (may be compressed) */
+#define ATAG_RDIMG	0x54410004
+/* ATAG_RDIMG uses tag_mem_range */
+
+/* Information about various clocks present in the system */
+#define ATAG_CLOCK	0x54410005
+
+struct tag_clock {
+	u32	clock_id;	/* Which clock are we talking about? */
+	u32	clock_flags;	/* Special features */
+	u64	clock_hz;	/* Clock speed in Hz */
+};
+
+/* The clock types we know about */
+#define CLOCK_BOOTCPU	0
+
+/* Memory reserved for the system (e.g. the bootloader) */
+#define ATAG_RSVD_MEM	0x54410006
+/* ATAG_RSVD_MEM uses tag_mem_range */
+
+/* Ethernet information */
+
+#define ATAG_ETHERNET	0x54410007
+
+struct tag_ethernet {
+	u8	mac_index;
+	u8	mii_phy_addr;
+	u8	hw_address[6];
+};
+
+#define ETH_INVALID_PHY	0xff
+
+struct tag {
+	struct tag_header hdr;
+	union {
+		struct tag_core core;
+		struct tag_mem_range mem_range;
+		struct tag_cmdline cmdline;
+		struct tag_clock clock;
+		struct tag_ethernet ethernet;
+	} u;
+};
+
+struct tagtable {
+	u32	tag;
+	int	(*parse)(struct tag *);
+};
+
+#define __tag __attribute_used__ __attribute__((__section__(".taglist")))
+#define __tagtable(tag, fn)						\
+	static struct tagtable __tagtable_##fn __tag = { tag, fn }
+
+#define tag_member_present(tag,member)					\
+	((unsigned long)(&((struct tag *)0L)->member + 1)		\
+	 <= (tag)->hdr.size * 4)
+
+#define tag_next(t)	((struct tag *)((u32 *)(t) + (t)->hdr.size))
+#define tag_size(type)	((sizeof(struct tag_header) + sizeof(struct type)) >> 2)
+
+#define for_each_tag(t,base)						\
+	for (t = base; t->hdr.size; t = tag_next(t))
+
+extern struct tag_mem_range *mem_phys;
+extern struct tag_mem_range *mem_reserved;
+extern struct tag_mem_range *mem_ramdisk;
+
+extern struct tag *bootloader_tags;
+
+extern void setup_bootmem(void);
+extern void setup_processor(void);
+extern void board_setup_fbmem(unsigned long fbmem_start,
+			      unsigned long fbmem_size);
+
+/* Chip-specific hook to enable the use of SDRAM */
+void chip_enable_sdram(void);
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_SETUP_H__ */
diff --git a/include/asm-avr32/shmbuf.h b/include/asm-avr32/shmbuf.h
new file mode 100644
index 000000000000..c62fba41739a
--- /dev/null
+++ b/include/asm-avr32/shmbuf.h
@@ -0,0 +1,42 @@
+#ifndef __ASM_AVR32_SHMBUF_H
+#define __ASM_AVR32_SHMBUF_H
+
+/*
+ * The shmid64_ds structure for i386 architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ */
+
+struct shmid64_ds {
+	struct ipc64_perm	shm_perm;	/* operation perms */
+	size_t			shm_segsz;	/* size of segment (bytes) */
+	__kernel_time_t		shm_atime;	/* last attach time */
+	unsigned long		__unused1;
+	__kernel_time_t		shm_dtime;	/* last detach time */
+	unsigned long		__unused2;
+	__kernel_time_t		shm_ctime;	/* last change time */
+	unsigned long		__unused3;
+	__kernel_pid_t		shm_cpid;	/* pid of creator */
+	__kernel_pid_t		shm_lpid;	/* pid of last operator */
+	unsigned long		shm_nattch;	/* no. of current attaches */
+	unsigned long		__unused4;
+	unsigned long		__unused5;
+};
+
+struct shminfo64 {
+	unsigned long	shmmax;
+	unsigned long	shmmin;
+	unsigned long	shmmni;
+	unsigned long	shmseg;
+	unsigned long	shmall;
+	unsigned long	__unused1;
+	unsigned long	__unused2;
+	unsigned long	__unused3;
+	unsigned long	__unused4;
+};
+
+#endif /* __ASM_AVR32_SHMBUF_H */
diff --git a/include/asm-avr32/shmparam.h b/include/asm-avr32/shmparam.h
new file mode 100644
index 000000000000..3681266c77f7
--- /dev/null
+++ b/include/asm-avr32/shmparam.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_SHMPARAM_H
+#define __ASM_AVR32_SHMPARAM_H
+
+#define SHMLBA PAGE_SIZE	/* attach addr a multiple of this */
+
+#endif /* __ASM_AVR32_SHMPARAM_H */
diff --git a/include/asm-avr32/sigcontext.h b/include/asm-avr32/sigcontext.h
new file mode 100644
index 000000000000..e04062b5f39f
--- /dev/null
+++ b/include/asm-avr32/sigcontext.h
@@ -0,0 +1,34 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SIGCONTEXT_H
+#define __ASM_AVR32_SIGCONTEXT_H
+
+struct sigcontext {
+	unsigned long	oldmask;
+
+	/* CPU registers */
+	unsigned long	sr;
+	unsigned long	pc;
+	unsigned long	lr;
+	unsigned long	sp;
+	unsigned long	r12;
+	unsigned long	r11;
+	unsigned long	r10;
+	unsigned long	r9;
+	unsigned long	r8;
+	unsigned long	r7;
+	unsigned long	r6;
+	unsigned long	r5;
+	unsigned long	r4;
+	unsigned long	r3;
+	unsigned long	r2;
+	unsigned long	r1;
+	unsigned long	r0;
+};
+
+#endif /* __ASM_AVR32_SIGCONTEXT_H */
diff --git a/include/asm-avr32/siginfo.h b/include/asm-avr32/siginfo.h
new file mode 100644
index 000000000000..5ee93f40a8a8
--- /dev/null
+++ b/include/asm-avr32/siginfo.h
@@ -0,0 +1,6 @@
+#ifndef _AVR32_SIGINFO_H
+#define _AVR32_SIGINFO_H
+
+#include <asm-generic/siginfo.h>
+
+#endif
diff --git a/include/asm-avr32/signal.h b/include/asm-avr32/signal.h
new file mode 100644
index 000000000000..caffefeeba1f
--- /dev/null
+++ b/include/asm-avr32/signal.h
@@ -0,0 +1,168 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SIGNAL_H
+#define __ASM_AVR32_SIGNAL_H
+
+#include <linux/types.h>
+
+/* Avoid too many header ordering problems.  */
+struct siginfo;
+
+#ifdef __KERNEL__
+/* Most things should be clean enough to redefine this at will, if care
+   is taken to make libc match.  */
+
+#define _NSIG		64
+#define _NSIG_BPW	32
+#define _NSIG_WORDS	(_NSIG / _NSIG_BPW)
+
+typedef unsigned long old_sigset_t;		/* at least 32 bits */
+
+typedef struct {
+	unsigned long sig[_NSIG_WORDS];
+} sigset_t;
+
+#else
+/* Here we must cater to libcs that poke about in kernel headers.  */
+
+#define NSIG		32
+typedef unsigned long sigset_t;
+
+#endif /* __KERNEL__ */
+
+#define SIGHUP		 1
+#define SIGINT		 2
+#define SIGQUIT		 3
+#define SIGILL		 4
+#define SIGTRAP		 5
+#define SIGABRT		 6
+#define SIGIOT		 6
+#define SIGBUS		 7
+#define SIGFPE		 8
+#define SIGKILL		 9
+#define SIGUSR1		10
+#define SIGSEGV		11
+#define SIGUSR2		12
+#define SIGPIPE		13
+#define SIGALRM		14
+#define SIGTERM		15
+#define SIGSTKFLT	16
+#define SIGCHLD		17
+#define SIGCONT		18
+#define SIGSTOP		19
+#define SIGTSTP		20
+#define SIGTTIN		21
+#define SIGTTOU		22
+#define SIGURG		23
+#define SIGXCPU		24
+#define SIGXFSZ		25
+#define SIGVTALRM	26
+#define SIGPROF		27
+#define SIGWINCH	28
+#define SIGIO		29
+#define SIGPOLL		SIGIO
+/*
+#define SIGLOST		29
+*/
+#define SIGPWR		30
+#define SIGSYS		31
+#define	SIGUNUSED	31
+
+/* These should not be considered constants from userland.  */
+#define SIGRTMIN	32
+#define SIGRTMAX	(_NSIG-1)
+
+/*
+ * SA_FLAGS values:
+ *
+ * SA_NOCLDSTOP		flag to turn off SIGCHLD when children stop.
+ * SA_NOCLDWAIT		flag on SIGCHLD to inhibit zombies.
+ * SA_SIGINFO		deliver the signal with SIGINFO structs
+ * SA_ONSTACK		indicates that a registered stack_t will be used.
+ * SA_RESTART		flag to get restarting signals (which were the default long ago)
+ * SA_NODEFER		prevents the current signal from being masked in the handler.
+ * SA_RESETHAND		clears the handler when the signal is delivered.
+ *
+ * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single
+ * Unix names RESETHAND and NODEFER respectively.
+ */
+#define SA_NOCLDSTOP	0x00000001
+#define SA_NOCLDWAIT	0x00000002
+#define SA_SIGINFO	0x00000004
+#define SA_RESTORER	0x04000000
+#define SA_ONSTACK	0x08000000
+#define SA_RESTART	0x10000000
+#define SA_NODEFER	0x40000000
+#define SA_RESETHAND	0x80000000
+
+#define SA_NOMASK	SA_NODEFER
+#define SA_ONESHOT	SA_RESETHAND
+
+/*
+ * sigaltstack controls
+ */
+#define SS_ONSTACK	1
+#define SS_DISABLE	2
+
+#define MINSIGSTKSZ	2048
+#define SIGSTKSZ	8192
+
+#include <asm-generic/signal.h>
+
+#ifdef __KERNEL__
+struct old_sigaction {
+	__sighandler_t sa_handler;
+	old_sigset_t sa_mask;
+	unsigned long sa_flags;
+	__sigrestore_t sa_restorer;
+};
+
+struct sigaction {
+	__sighandler_t sa_handler;
+	unsigned long sa_flags;
+	__sigrestore_t sa_restorer;
+	sigset_t sa_mask;		/* mask last for extensibility */
+};
+
+struct k_sigaction {
+	struct sigaction sa;
+};
+#else
+/* Here we must cater to libcs that poke about in kernel headers.  */
+
+struct sigaction {
+	union {
+		__sighandler_t _sa_handler;
+		void (*_sa_sigaction)(int, struct siginfo *, void *);
+	} _u;
+	sigset_t sa_mask;
+	unsigned long sa_flags;
+	void (*sa_restorer)(void);
+};
+
+#define sa_handler	_u._sa_handler
+#define sa_sigaction	_u._sa_sigaction
+
+#endif /* __KERNEL__ */
+
+typedef struct sigaltstack {
+	void __user *ss_sp;
+	int ss_flags;
+	size_t ss_size;
+} stack_t;
+
+#ifdef __KERNEL__
+
+#include <asm/sigcontext.h>
+#undef __HAVE_ARCH_SIG_BITOPS
+
+#define ptrace_signal_deliver(regs, cookie) do { } while (0)
+
+#endif /* __KERNEL__ */
+
+#endif
diff --git a/include/asm-avr32/socket.h b/include/asm-avr32/socket.h
new file mode 100644
index 000000000000..543229de8173
--- /dev/null
+++ b/include/asm-avr32/socket.h
@@ -0,0 +1,53 @@
+#ifndef __ASM_AVR32_SOCKET_H
+#define __ASM_AVR32_SOCKET_H
+
+#include <asm/sockios.h>
+
+/* For setsockopt(2) */
+#define SOL_SOCKET	1
+
+#define SO_DEBUG	1
+#define SO_REUSEADDR	2
+#define SO_TYPE		3
+#define SO_ERROR	4
+#define SO_DONTROUTE	5
+#define SO_BROADCAST	6
+#define SO_SNDBUF	7
+#define SO_RCVBUF	8
+#define SO_SNDBUFFORCE	32
+#define SO_RCVBUFFORCE	33
+#define SO_KEEPALIVE	9
+#define SO_OOBINLINE	10
+#define SO_NO_CHECK	11
+#define SO_PRIORITY	12
+#define SO_LINGER	13
+#define SO_BSDCOMPAT	14
+/* To add :#define SO_REUSEPORT 15 */
+#define SO_PASSCRED	16
+#define SO_PEERCRED	17
+#define SO_RCVLOWAT	18
+#define SO_SNDLOWAT	19
+#define SO_RCVTIMEO	20
+#define SO_SNDTIMEO	21
+
+/* Security levels - as per NRL IPv6 - don't actually do anything */
+#define SO_SECURITY_AUTHENTICATION		22
+#define SO_SECURITY_ENCRYPTION_TRANSPORT	23
+#define SO_SECURITY_ENCRYPTION_NETWORK		24
+
+#define SO_BINDTODEVICE	25
+
+/* Socket filtering */
+#define SO_ATTACH_FILTER        26
+#define SO_DETACH_FILTER        27
+
+#define SO_PEERNAME		28
+#define SO_TIMESTAMP		29
+#define SCM_TIMESTAMP		SO_TIMESTAMP
+
+#define SO_ACCEPTCONN		30
+
+#define SO_PEERSEC		31
+#define SO_PASSSEC		34
+
+#endif /* __ASM_AVR32_SOCKET_H */
diff --git a/include/asm-avr32/sockios.h b/include/asm-avr32/sockios.h
new file mode 100644
index 000000000000..84f3d65b3b3b
--- /dev/null
+++ b/include/asm-avr32/sockios.h
@@ -0,0 +1,12 @@
+#ifndef __ASM_AVR32_SOCKIOS_H
+#define __ASM_AVR32_SOCKIOS_H
+
+/* Socket-level I/O control calls. */
+#define FIOSETOWN 	0x8901
+#define SIOCSPGRP	0x8902
+#define FIOGETOWN	0x8903
+#define SIOCGPGRP	0x8904
+#define SIOCATMARK	0x8905
+#define SIOCGSTAMP	0x8906		/* Get stamp */
+
+#endif /* __ASM_AVR32_SOCKIOS_H */
diff --git a/include/asm-avr32/stat.h b/include/asm-avr32/stat.h
new file mode 100644
index 000000000000..e72881e10230
--- /dev/null
+++ b/include/asm-avr32/stat.h
@@ -0,0 +1,79 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_STAT_H
+#define __ASM_AVR32_STAT_H
+
+struct __old_kernel_stat {
+        unsigned short st_dev;
+        unsigned short st_ino;
+        unsigned short st_mode;
+        unsigned short st_nlink;
+        unsigned short st_uid;
+        unsigned short st_gid;
+        unsigned short st_rdev;
+        unsigned long  st_size;
+        unsigned long  st_atime;
+        unsigned long  st_mtime;
+        unsigned long  st_ctime;
+};
+
+struct stat {
+        unsigned long st_dev;
+        unsigned long st_ino;
+        unsigned short st_mode;
+        unsigned short st_nlink;
+        unsigned short st_uid;
+        unsigned short st_gid;
+        unsigned long  st_rdev;
+        unsigned long  st_size;
+        unsigned long  st_blksize;
+        unsigned long  st_blocks;
+        unsigned long  st_atime;
+        unsigned long  st_atime_nsec;
+        unsigned long  st_mtime;
+        unsigned long  st_mtime_nsec;
+        unsigned long  st_ctime;
+        unsigned long  st_ctime_nsec;
+        unsigned long  __unused4;
+        unsigned long  __unused5;
+};
+
+#define STAT_HAVE_NSEC 1
+
+struct stat64 {
+	unsigned long long st_dev;
+
+	unsigned long long st_ino;
+	unsigned int	st_mode;
+	unsigned int	st_nlink;
+
+	unsigned long	st_uid;
+	unsigned long	st_gid;
+
+	unsigned long long st_rdev;
+
+	long long	st_size;
+	unsigned long	__pad1;		/* align 64-bit st_blocks */
+	unsigned long	st_blksize;
+
+	unsigned long long st_blocks;	/* Number 512-byte blocks allocated. */
+
+	unsigned long	st_atime;
+	unsigned long	st_atime_nsec;
+
+	unsigned long	st_mtime;
+	unsigned long	st_mtime_nsec;
+
+	unsigned long	st_ctime;
+	unsigned long	st_ctime_nsec;
+
+	unsigned long	__unused1;
+	unsigned long	__unused2;
+};
+
+#endif /* __ASM_AVR32_STAT_H */
diff --git a/include/asm-avr32/statfs.h b/include/asm-avr32/statfs.h
new file mode 100644
index 000000000000..2961bd18c50e
--- /dev/null
+++ b/include/asm-avr32/statfs.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_STATFS_H
+#define __ASM_AVR32_STATFS_H
+
+#include <asm-generic/statfs.h>
+
+#endif /* __ASM_AVR32_STATFS_H */
diff --git a/include/asm-avr32/string.h b/include/asm-avr32/string.h
new file mode 100644
index 000000000000..c91a623cd585
--- /dev/null
+++ b/include/asm-avr32/string.h
@@ -0,0 +1,17 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_STRING_H
+#define __ASM_AVR32_STRING_H
+
+#define __HAVE_ARCH_MEMSET
+extern void *memset(void *b, int c, size_t len);
+
+#define __HAVE_ARCH_MEMCPY
+extern void *memcpy(void *to, const void *from, size_t len);
+
+#endif /* __ASM_AVR32_STRING_H */
diff --git a/include/asm-avr32/sysreg.h b/include/asm-avr32/sysreg.h
new file mode 100644
index 000000000000..f91975f330f6
--- /dev/null
+++ b/include/asm-avr32/sysreg.h
@@ -0,0 +1,332 @@
+/*
+ * AVR32 System Registers
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SYSREG_H__
+#define __ASM_AVR32_SYSREG_H__
+
+/* sysreg register offsets */
+#define SYSREG_SR                               0x0000
+#define SYSREG_EVBA                             0x0004
+#define SYSREG_ACBA                             0x0008
+#define SYSREG_CPUCR                            0x000c
+#define SYSREG_ECR                              0x0010
+#define SYSREG_RSR_SUP                          0x0014
+#define SYSREG_RSR_INT0                         0x0018
+#define SYSREG_RSR_INT1                         0x001c
+#define SYSREG_RSR_INT2                         0x0020
+#define SYSREG_RSR_INT3                         0x0024
+#define SYSREG_RSR_EX                           0x0028
+#define SYSREG_RSR_NMI                          0x002c
+#define SYSREG_RSR_DBG                          0x0030
+#define SYSREG_RAR_SUP                          0x0034
+#define SYSREG_RAR_INT0                         0x0038
+#define SYSREG_RAR_INT1                         0x003c
+#define SYSREG_RAR_INT2                         0x0040
+#define SYSREG_RAR_INT3                         0x0044
+#define SYSREG_RAR_EX                           0x0048
+#define SYSREG_RAR_NMI                          0x004c
+#define SYSREG_RAR_DBG                          0x0050
+#define SYSREG_JECR                             0x0054
+#define SYSREG_JOSP                             0x0058
+#define SYSREG_JAVA_LV0                         0x005c
+#define SYSREG_JAVA_LV1                         0x0060
+#define SYSREG_JAVA_LV2                         0x0064
+#define SYSREG_JAVA_LV3                         0x0068
+#define SYSREG_JAVA_LV4                         0x006c
+#define SYSREG_JAVA_LV5                         0x0070
+#define SYSREG_JAVA_LV6                         0x0074
+#define SYSREG_JAVA_LV7                         0x0078
+#define SYSREG_JTBA                             0x007c
+#define SYSREG_JBCR                             0x0080
+#define SYSREG_CONFIG0                          0x0100
+#define SYSREG_CONFIG1                          0x0104
+#define SYSREG_COUNT                            0x0108
+#define SYSREG_COMPARE                          0x010c
+#define SYSREG_TLBEHI                           0x0110
+#define SYSREG_TLBELO                           0x0114
+#define SYSREG_PTBR                             0x0118
+#define SYSREG_TLBEAR                           0x011c
+#define SYSREG_MMUCR                            0x0120
+#define SYSREG_TLBARLO                          0x0124
+#define SYSREG_TLBARHI                          0x0128
+#define SYSREG_PCCNT                            0x012c
+#define SYSREG_PCNT0                            0x0130
+#define SYSREG_PCNT1                            0x0134
+#define SYSREG_PCCR                             0x0138
+#define SYSREG_BEAR                             0x013c
+
+/* Bitfields in SR */
+#define SYSREG_SR_C_OFFSET                      0
+#define SYSREG_SR_C_SIZE                        1
+#define SYSREG_Z_OFFSET                         1
+#define SYSREG_Z_SIZE                           1
+#define SYSREG_SR_N_OFFSET                      2
+#define SYSREG_SR_N_SIZE                        1
+#define SYSREG_SR_V_OFFSET                      3
+#define SYSREG_SR_V_SIZE                        1
+#define SYSREG_Q_OFFSET                         4
+#define SYSREG_Q_SIZE                           1
+#define SYSREG_GM_OFFSET                        16
+#define SYSREG_GM_SIZE                          1
+#define SYSREG_I0M_OFFSET                       17
+#define SYSREG_I0M_SIZE                         1
+#define SYSREG_I1M_OFFSET                       18
+#define SYSREG_I1M_SIZE                         1
+#define SYSREG_I2M_OFFSET                       19
+#define SYSREG_I2M_SIZE                         1
+#define SYSREG_I3M_OFFSET                       20
+#define SYSREG_I3M_SIZE                         1
+#define SYSREG_EM_OFFSET                        21
+#define SYSREG_EM_SIZE                          1
+#define SYSREG_M0_OFFSET                        22
+#define SYSREG_M0_SIZE                          1
+#define SYSREG_M1_OFFSET                        23
+#define SYSREG_M1_SIZE                          1
+#define SYSREG_M2_OFFSET                        24
+#define SYSREG_M2_SIZE                          1
+#define SYSREG_SR_D_OFFSET                      26
+#define SYSREG_SR_D_SIZE                        1
+#define SYSREG_DM_OFFSET                        27
+#define SYSREG_DM_SIZE                          1
+#define SYSREG_SR_J_OFFSET                      28
+#define SYSREG_SR_J_SIZE                        1
+#define SYSREG_R_OFFSET                         29
+#define SYSREG_R_SIZE                           1
+#define SYSREG_H_OFFSET                         30
+#define SYSREG_H_SIZE                           1
+
+/* Bitfields in EVBA */
+
+/* Bitfields in ACBA */
+
+/* Bitfields in CPUCR */
+#define SYSREG_BI_OFFSET                        0
+#define SYSREG_BI_SIZE                          1
+#define SYSREG_BE_OFFSET                        1
+#define SYSREG_BE_SIZE                          1
+#define SYSREG_FE_OFFSET                        2
+#define SYSREG_FE_SIZE                          1
+#define SYSREG_RE_OFFSET                        3
+#define SYSREG_RE_SIZE                          1
+#define SYSREG_IBE_OFFSET                       4
+#define SYSREG_IBE_SIZE                         1
+#define SYSREG_IEE_OFFSET                       5
+#define SYSREG_IEE_SIZE                         1
+
+/* Bitfields in ECR */
+#define SYSREG_ECR_OFFSET                       0
+#define SYSREG_ECR_SIZE                         32
+
+/* Bitfields in RSR_SUP */
+
+/* Bitfields in RSR_INT0 */
+
+/* Bitfields in RSR_INT1 */
+
+/* Bitfields in RSR_INT2 */
+
+/* Bitfields in RSR_INT3 */
+
+/* Bitfields in RSR_EX */
+
+/* Bitfields in RSR_NMI */
+
+/* Bitfields in RSR_DBG */
+
+/* Bitfields in RAR_SUP */
+
+/* Bitfields in RAR_INT0 */
+
+/* Bitfields in RAR_INT1 */
+
+/* Bitfields in RAR_INT2 */
+
+/* Bitfields in RAR_INT3 */
+
+/* Bitfields in RAR_EX */
+
+/* Bitfields in RAR_NMI */
+
+/* Bitfields in RAR_DBG */
+
+/* Bitfields in JECR */
+
+/* Bitfields in JOSP */
+
+/* Bitfields in JAVA_LV0 */
+
+/* Bitfields in JAVA_LV1 */
+
+/* Bitfields in JAVA_LV2 */
+
+/* Bitfields in JAVA_LV3 */
+
+/* Bitfields in JAVA_LV4 */
+
+/* Bitfields in JAVA_LV5 */
+
+/* Bitfields in JAVA_LV6 */
+
+/* Bitfields in JAVA_LV7 */
+
+/* Bitfields in JTBA */
+
+/* Bitfields in JBCR */
+
+/* Bitfields in CONFIG0 */
+#define SYSREG_CONFIG0_D_OFFSET                 1
+#define SYSREG_CONFIG0_D_SIZE                   1
+#define SYSREG_CONFIG0_S_OFFSET                 2
+#define SYSREG_CONFIG0_S_SIZE                   1
+#define SYSREG_O_OFFSET                         3
+#define SYSREG_O_SIZE                           1
+#define SYSREG_P_OFFSET                         4
+#define SYSREG_P_SIZE                           1
+#define SYSREG_CONFIG0_J_OFFSET                 5
+#define SYSREG_CONFIG0_J_SIZE                   1
+#define SYSREG_F_OFFSET                         6
+#define SYSREG_F_SIZE                           1
+#define SYSREG_MMUT_OFFSET                      7
+#define SYSREG_MMUT_SIZE                        3
+#define SYSREG_AR_OFFSET                        10
+#define SYSREG_AR_SIZE                          3
+#define SYSREG_AT_OFFSET                        13
+#define SYSREG_AT_SIZE                          3
+#define SYSREG_PROCESSORREVISION_OFFSET         16
+#define SYSREG_PROCESSORREVISION_SIZE           8
+#define SYSREG_PROCESSORID_OFFSET               24
+#define SYSREG_PROCESSORID_SIZE                 8
+
+/* Bitfields in CONFIG1 */
+#define SYSREG_DASS_OFFSET                      0
+#define SYSREG_DASS_SIZE                        3
+#define SYSREG_DLSZ_OFFSET                      3
+#define SYSREG_DLSZ_SIZE                        3
+#define SYSREG_DSET_OFFSET                      6
+#define SYSREG_DSET_SIZE                        4
+#define SYSREG_IASS_OFFSET                      10
+#define SYSREG_IASS_SIZE                        2
+#define SYSREG_ILSZ_OFFSET                      13
+#define SYSREG_ILSZ_SIZE                        3
+#define SYSREG_ISET_OFFSET                      16
+#define SYSREG_ISET_SIZE                        4
+#define SYSREG_DMMUSZ_OFFSET                    20
+#define SYSREG_DMMUSZ_SIZE                      6
+#define SYSREG_IMMUSZ_OFFSET                    26
+#define SYSREG_IMMUSZ_SIZE                      6
+
+/* Bitfields in COUNT */
+
+/* Bitfields in COMPARE */
+
+/* Bitfields in TLBEHI */
+#define SYSREG_ASID_OFFSET                      0
+#define SYSREG_ASID_SIZE                        8
+#define SYSREG_TLBEHI_I_OFFSET                  8
+#define SYSREG_TLBEHI_I_SIZE                    1
+#define SYSREG_TLBEHI_V_OFFSET                  9
+#define SYSREG_TLBEHI_V_SIZE                    1
+#define SYSREG_VPN_OFFSET                       10
+#define SYSREG_VPN_SIZE                         22
+
+/* Bitfields in TLBELO */
+#define SYSREG_W_OFFSET                         0
+#define SYSREG_W_SIZE                           1
+#define SYSREG_TLBELO_D_OFFSET                  1
+#define SYSREG_TLBELO_D_SIZE                    1
+#define SYSREG_SZ_OFFSET                        2
+#define SYSREG_SZ_SIZE                          2
+#define SYSREG_AP_OFFSET                        4
+#define SYSREG_AP_SIZE                          3
+#define SYSREG_B_OFFSET                         7
+#define SYSREG_B_SIZE                           1
+#define SYSREG_G_OFFSET                         8
+#define SYSREG_G_SIZE                           1
+#define SYSREG_TLBELO_C_OFFSET                  9
+#define SYSREG_TLBELO_C_SIZE                    1
+#define SYSREG_PFN_OFFSET                       10
+#define SYSREG_PFN_SIZE                         22
+
+/* Bitfields in PTBR */
+
+/* Bitfields in TLBEAR */
+
+/* Bitfields in MMUCR */
+#define SYSREG_E_OFFSET                         0
+#define SYSREG_E_SIZE                           1
+#define SYSREG_M_OFFSET                         1
+#define SYSREG_M_SIZE                           1
+#define SYSREG_MMUCR_I_OFFSET                   2
+#define SYSREG_MMUCR_I_SIZE                     1
+#define SYSREG_MMUCR_N_OFFSET                   3
+#define SYSREG_MMUCR_N_SIZE                     1
+#define SYSREG_MMUCR_S_OFFSET                   4
+#define SYSREG_MMUCR_S_SIZE                     1
+#define SYSREG_DLA_OFFSET                       8
+#define SYSREG_DLA_SIZE                         6
+#define SYSREG_DRP_OFFSET                       14
+#define SYSREG_DRP_SIZE                         6
+#define SYSREG_ILA_OFFSET                       20
+#define SYSREG_ILA_SIZE                         6
+#define SYSREG_IRP_OFFSET                       26
+#define SYSREG_IRP_SIZE                         6
+
+/* Bitfields in TLBARLO */
+
+/* Bitfields in TLBARHI */
+
+/* Bitfields in PCCNT */
+
+/* Bitfields in PCNT0 */
+
+/* Bitfields in PCNT1 */
+
+/* Bitfields in PCCR */
+
+/* Bitfields in BEAR */
+
+/* Constants for ECR */
+#define ECR_UNRECOVERABLE                       0
+#define ECR_TLB_MULTIPLE                        1
+#define ECR_BUS_ERROR_WRITE                     2
+#define ECR_BUS_ERROR_READ                      3
+#define ECR_NMI                                 4
+#define ECR_ADDR_ALIGN_X                        5
+#define ECR_PROTECTION_X                        6
+#define ECR_DEBUG                               7
+#define ECR_ILLEGAL_OPCODE                      8
+#define ECR_UNIMPL_INSTRUCTION                  9
+#define ECR_PRIVILEGE_VIOLATION                 10
+#define ECR_FPE                                 11
+#define ECR_COPROC_ABSENT                       12
+#define ECR_ADDR_ALIGN_R                        13
+#define ECR_ADDR_ALIGN_W                        14
+#define ECR_PROTECTION_R                        15
+#define ECR_PROTECTION_W                        16
+#define ECR_DTLB_MODIFIED                       17
+#define ECR_TLB_MISS_X                          20
+#define ECR_TLB_MISS_R                          24
+#define ECR_TLB_MISS_W                          28
+
+/* Bit manipulation macros */
+#define SYSREG_BIT(name)                        (1 << SYSREG_##name##_OFFSET)
+#define SYSREG_BF(name,value)                   (((value) & ((1 << SYSREG_##name##_SIZE) - 1)) << SYSREG_##name##_OFFSET)
+#define SYSREG_BFEXT(name,value)                (((value) >> SYSREG_##name##_OFFSET) & ((1 << SYSREG_##name##_SIZE) - 1))
+#define SYSREG_BFINS(name,value,old)            (((old) & ~(((1 << SYSREG_##name##_SIZE) - 1) << SYSREG_##name##_OFFSET)) | SYSREG_BF(name,value))
+
+#ifdef __CHECKER__
+extern unsigned long __builtin_mfsr(unsigned long reg);
+extern void __builtin_mtsr(unsigned long reg, unsigned long value);
+#endif
+
+/* Register access macros */
+#define sysreg_read(reg)                        __builtin_mfsr(SYSREG_##reg)
+#define sysreg_write(reg, value)                __builtin_mtsr(SYSREG_##reg, value)
+
+#endif /* __ASM_AVR32_SYSREG_H__ */
diff --git a/include/asm-avr32/system.h b/include/asm-avr32/system.h
new file mode 100644
index 000000000000..ac596058697d
--- /dev/null
+++ b/include/asm-avr32/system.h
@@ -0,0 +1,155 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SYSTEM_H
+#define __ASM_AVR32_SYSTEM_H
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+
+#include <asm/ptrace.h>
+#include <asm/sysreg.h>
+
+#define xchg(ptr,x) \
+	((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
+
+#define nop() asm volatile("nop")
+
+#define mb()			asm volatile("" : : : "memory")
+#define rmb()			mb()
+#define wmb()			asm volatile("sync 0" : : : "memory")
+#define read_barrier_depends()  do { } while(0)
+#define set_mb(var, value)      do { var = value; mb(); } while(0)
+
+/*
+ * Help PathFinder and other Nexus-compliant debuggers keep track of
+ * the current PID by emitting an Ownership Trace Message each time we
+ * switch task.
+ */
+#ifdef CONFIG_OWNERSHIP_TRACE
+#include <asm/ocd.h>
+#define finish_arch_switch(prev)			\
+	do {						\
+		__mtdr(DBGREG_PID, prev->pid);		\
+		__mtdr(DBGREG_PID, current->pid);	\
+	} while(0)
+#endif
+
+/*
+ * switch_to(prev, next, last) should switch from task `prev' to task
+ * `next'. `prev' will never be the same as `next'.
+ *
+ * We just delegate everything to the __switch_to assembly function,
+ * which is implemented in arch/avr32/kernel/switch_to.S
+ *
+ * mb() tells GCC not to cache `current' across this call.
+ */
+struct cpu_context;
+struct task_struct;
+extern struct task_struct *__switch_to(struct task_struct *,
+				       struct cpu_context *,
+				       struct cpu_context *);
+#define switch_to(prev, next, last)					\
+	do {								\
+		last = __switch_to(prev, &prev->thread.cpu_context + 1,	\
+				   &next->thread.cpu_context);		\
+	} while (0)
+
+#ifdef CONFIG_SMP
+# error "The AVR32 port does not support SMP"
+#else
+# define smp_mb()		barrier()
+# define smp_rmb()		barrier()
+# define smp_wmb()		barrier()
+# define smp_read_barrier_depends() do { } while(0)
+#endif
+
+#include <linux/irqflags.h>
+
+extern void __xchg_called_with_bad_pointer(void);
+
+#ifdef __CHECKER__
+extern unsigned long __builtin_xchg(void *ptr, unsigned long x);
+#endif
+
+#define xchg_u32(val, m) __builtin_xchg((void *)m, val)
+
+static inline unsigned long __xchg(unsigned long x,
+				       volatile void *ptr,
+				       int size)
+{
+	switch(size) {
+	case 4:
+		return xchg_u32(x, ptr);
+	default:
+		__xchg_called_with_bad_pointer();
+		return x;
+	}
+}
+
+static inline unsigned long __cmpxchg_u32(volatile int *m, unsigned long old,
+					  unsigned long new)
+{
+	__u32 ret;
+
+	asm volatile(
+		"1:	ssrf	5\n"
+		"	ld.w	%[ret], %[m]\n"
+		"	cp.w	%[ret], %[old]\n"
+		"	brne	2f\n"
+		"	stcond	%[m], %[new]\n"
+		"	brne	1b\n"
+		"2:\n"
+		: [ret] "=&r"(ret), [m] "=m"(*m)
+		: "m"(m), [old] "ir"(old), [new] "r"(new)
+		: "memory", "cc");
+	return ret;
+}
+
+extern unsigned long __cmpxchg_u64_unsupported_on_32bit_kernels(
+        volatile int * m, unsigned long old, unsigned long new);
+#define __cmpxchg_u64 __cmpxchg_u64_unsupported_on_32bit_kernels
+
+/* This function doesn't exist, so you'll get a linker error
+   if something tries to do an invalid cmpxchg().  */
+extern void __cmpxchg_called_with_bad_pointer(void);
+
+#define __HAVE_ARCH_CMPXCHG 1
+
+static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
+				      unsigned long new, int size)
+{
+	switch (size) {
+	case 4:
+		return __cmpxchg_u32(ptr, old, new);
+	case 8:
+		return __cmpxchg_u64(ptr, old, new);
+	}
+
+	__cmpxchg_called_with_bad_pointer();
+	return old;
+}
+
+#define cmpxchg(ptr, old, new)					\
+	((typeof(*(ptr)))__cmpxchg((ptr), (unsigned long)(old),	\
+				   (unsigned long)(new),	\
+				   sizeof(*(ptr))))
+
+struct pt_regs;
+extern void __die(const char *, struct pt_regs *, unsigned long,
+		  const char *, const char *, unsigned long);
+extern void __die_if_kernel(const char *, struct pt_regs *, unsigned long,
+			    const char *, const char *, unsigned long);
+
+#define die(msg, regs, err)					\
+	__die(msg, regs, err, __FILE__ ":", __FUNCTION__, __LINE__)
+#define die_if_kernel(msg, regs, err)					\
+	__die_if_kernel(msg, regs, err, __FILE__ ":", __FUNCTION__, __LINE__)
+
+#define arch_align_stack(x)	(x)
+
+#endif /* __ASM_AVR32_SYSTEM_H */
diff --git a/include/asm-avr32/termbits.h b/include/asm-avr32/termbits.h
new file mode 100644
index 000000000000..9dc6eacafa33
--- /dev/null
+++ b/include/asm-avr32/termbits.h
@@ -0,0 +1,173 @@
+#ifndef __ASM_AVR32_TERMBITS_H
+#define __ASM_AVR32_TERMBITS_H
+
+#include <linux/posix_types.h>
+
+typedef unsigned char	cc_t;
+typedef unsigned int	speed_t;
+typedef unsigned int	tcflag_t;
+
+#define NCCS 19
+struct termios {
+	tcflag_t c_iflag;		/* input mode flags */
+	tcflag_t c_oflag;		/* output mode flags */
+	tcflag_t c_cflag;		/* control mode flags */
+	tcflag_t c_lflag;		/* local mode flags */
+	cc_t c_line;			/* line discipline */
+	cc_t c_cc[NCCS];		/* control characters */
+};
+
+/* c_cc characters */
+#define VINTR 0
+#define VQUIT 1
+#define VERASE 2
+#define VKILL 3
+#define VEOF 4
+#define VTIME 5
+#define VMIN 6
+#define VSWTC 7
+#define VSTART 8
+#define VSTOP 9
+#define VSUSP 10
+#define VEOL 11
+#define VREPRINT 12
+#define VDISCARD 13
+#define VWERASE 14
+#define VLNEXT 15
+#define VEOL2 16
+
+/* c_iflag bits */
+#define IGNBRK	0000001
+#define BRKINT	0000002
+#define IGNPAR	0000004
+#define PARMRK	0000010
+#define INPCK	0000020
+#define ISTRIP	0000040
+#define INLCR	0000100
+#define IGNCR	0000200
+#define ICRNL	0000400
+#define IUCLC	0001000
+#define IXON	0002000
+#define IXANY	0004000
+#define IXOFF	0010000
+#define IMAXBEL	0020000
+#define IUTF8	0040000
+
+/* c_oflag bits */
+#define OPOST	0000001
+#define OLCUC	0000002
+#define ONLCR	0000004
+#define OCRNL	0000010
+#define ONOCR	0000020
+#define ONLRET	0000040
+#define OFILL	0000100
+#define OFDEL	0000200
+#define NLDLY	0000400
+#define   NL0	0000000
+#define   NL1	0000400
+#define CRDLY	0003000
+#define   CR0	0000000
+#define   CR1	0001000
+#define   CR2	0002000
+#define   CR3	0003000
+#define TABDLY	0014000
+#define   TAB0	0000000
+#define   TAB1	0004000
+#define   TAB2	0010000
+#define   TAB3	0014000
+#define   XTABS	0014000
+#define BSDLY	0020000
+#define   BS0	0000000
+#define   BS1	0020000
+#define VTDLY	0040000
+#define   VT0	0000000
+#define   VT1	0040000
+#define FFDLY	0100000
+#define   FF0	0000000
+#define   FF1	0100000
+
+/* c_cflag bit meaning */
+#define CBAUD	0010017
+#define  B0	0000000		/* hang up */
+#define  B50	0000001
+#define  B75	0000002
+#define  B110	0000003
+#define  B134	0000004
+#define  B150	0000005
+#define  B200	0000006
+#define  B300	0000007
+#define  B600	0000010
+#define  B1200	0000011
+#define  B1800	0000012
+#define  B2400	0000013
+#define  B4800	0000014
+#define  B9600	0000015
+#define  B19200	0000016
+#define  B38400	0000017
+#define EXTA B19200
+#define EXTB B38400
+#define CSIZE	0000060
+#define   CS5	0000000
+#define   CS6	0000020
+#define   CS7	0000040
+#define   CS8	0000060
+#define CSTOPB	0000100
+#define CREAD	0000200
+#define PARENB	0000400
+#define PARODD	0001000
+#define HUPCL	0002000
+#define CLOCAL	0004000
+#define CBAUDEX 0010000
+#define    B57600 0010001
+#define   B115200 0010002
+#define   B230400 0010003
+#define   B460800 0010004
+#define   B500000 0010005
+#define   B576000 0010006
+#define   B921600 0010007
+#define  B1000000 0010010
+#define  B1152000 0010011
+#define  B1500000 0010012
+#define  B2000000 0010013
+#define  B2500000 0010014
+#define  B3000000 0010015
+#define  B3500000 0010016
+#define  B4000000 0010017
+#define CIBAUD	  002003600000	/* input baud rate (not used) */
+#define CMSPAR	  010000000000		/* mark or space (stick) parity */
+#define CRTSCTS	  020000000000		/* flow control */
+
+/* c_lflag bits */
+#define ISIG	0000001
+#define ICANON	0000002
+#define XCASE	0000004
+#define ECHO	0000010
+#define ECHOE	0000020
+#define ECHOK	0000040
+#define ECHONL	0000100
+#define NOFLSH	0000200
+#define TOSTOP	0000400
+#define ECHOCTL	0001000
+#define ECHOPRT	0002000
+#define ECHOKE	0004000
+#define FLUSHO	0010000
+#define PENDIN	0040000
+#define IEXTEN	0100000
+
+/* tcflow() and TCXONC use these */
+#define	TCOOFF		0
+#define	TCOON		1
+#define	TCIOFF		2
+#define	TCION		3
+
+/* tcflush() and TCFLSH use these */
+#define	TCIFLUSH	0
+#define	TCOFLUSH	1
+#define	TCIOFLUSH	2
+
+/* tcsetattr uses these */
+#define	TCSANOW		0
+#define	TCSADRAIN	1
+#define	TCSAFLUSH	2
+
+#endif /* __ASM_AVR32_TERMBITS_H */
diff --git a/include/asm-avr32/termios.h b/include/asm-avr32/termios.h
new file mode 100644
index 000000000000..615bc0639e5c
--- /dev/null
+++ b/include/asm-avr32/termios.h
@@ -0,0 +1,80 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TERMIOS_H
+#define __ASM_AVR32_TERMIOS_H
+
+#include <asm/termbits.h>
+#include <asm/ioctls.h>
+
+struct winsize {
+	unsigned short ws_row;
+	unsigned short ws_col;
+	unsigned short ws_xpixel;
+	unsigned short ws_ypixel;
+};
+
+#define NCC 8
+struct termio {
+	unsigned short c_iflag;		/* input mode flags */
+	unsigned short c_oflag;		/* output mode flags */
+	unsigned short c_cflag;		/* control mode flags */
+	unsigned short c_lflag;		/* local mode flags */
+	unsigned char c_line;		/* line discipline */
+	unsigned char c_cc[NCC];	/* control characters */
+};
+
+/* modem lines */
+#define TIOCM_LE	0x001
+#define TIOCM_DTR	0x002
+#define TIOCM_RTS	0x004
+#define TIOCM_ST	0x008
+#define TIOCM_SR	0x010
+#define TIOCM_CTS	0x020
+#define TIOCM_CAR	0x040
+#define TIOCM_RNG	0x080
+#define TIOCM_DSR	0x100
+#define TIOCM_CD	TIOCM_CAR
+#define TIOCM_RI	TIOCM_RNG
+#define TIOCM_OUT1	0x2000
+#define TIOCM_OUT2	0x4000
+#define TIOCM_LOOP	0x8000
+
+/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
+
+/* line disciplines */
+#define N_TTY		0
+#define N_SLIP		1
+#define N_MOUSE		2
+#define N_PPP		3
+#define N_STRIP		4
+#define N_AX25		5
+#define N_X25		6	/* X.25 async */
+#define N_6PACK		7
+#define N_MASC		8	/* Reserved for Mobitex module <kaz@cafe.net> */
+#define N_R3964		9	/* Reserved for Simatic R3964 module */
+#define N_PROFIBUS_FDL	10	/* Reserved for Profibus <Dave@mvhi.com> */
+#define N_IRDA		11	/* Linux IR - http://irda.sourceforge.net/ */
+#define N_SMSBLOCK	12	/* SMS block mode - for talking to GSM data cards about SMS messages */
+#define N_HDLC		13	/* synchronous HDLC */
+#define N_SYNC_PPP	14	/* synchronous PPP */
+#define N_HCI		15  /* Bluetooth HCI UART */
+
+#ifdef __KERNEL__
+/*	intr=^C		quit=^\		erase=del	kill=^U
+	eof=^D		vtime=\0	vmin=\1		sxtc=\0
+	start=^Q	stop=^S		susp=^Z		eol=\0
+	reprint=^R	discard=^U	werase=^W	lnext=^V
+	eol2=\0
+*/
+#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0"
+
+#include <asm-generic/termios.h>
+
+#endif	/* __KERNEL__ */
+
+#endif	/* __ASM_AVR32_TERMIOS_H */
diff --git a/include/asm-avr32/thread_info.h b/include/asm-avr32/thread_info.h
new file mode 100644
index 000000000000..d1f5b35ebd54
--- /dev/null
+++ b/include/asm-avr32/thread_info.h
@@ -0,0 +1,106 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_THREAD_INFO_H
+#define __ASM_AVR32_THREAD_INFO_H
+
+#include <asm/page.h>
+
+#define THREAD_SIZE_ORDER	1
+#define THREAD_SIZE		(PAGE_SIZE << THREAD_SIZE_ORDER)
+
+#ifndef __ASSEMBLY__
+#include <asm/types.h>
+
+struct task_struct;
+struct exec_domain;
+
+struct thread_info {
+	struct task_struct	*task;		/* main task structure */
+	struct exec_domain	*exec_domain;	/* execution domain */
+	unsigned long		flags;		/* low level flags */
+	__u32			cpu;
+	__s32			preempt_count;	/* 0 => preemptable, <0 => BUG */
+	struct restart_block	restart_block;
+	__u8			supervisor_stack[0];
+};
+
+#define INIT_THREAD_INFO(tsk)						\
+{									\
+	.task		= &tsk,						\
+	.exec_domain	= &default_exec_domain,				\
+	.flags		= 0,						\
+	.cpu		= 0,						\
+	.preempt_count	= 1,						\
+	.restart_block	= {						\
+		.fn	= do_no_restart_syscall				\
+	}								\
+}
+
+#define init_thread_info	(init_thread_union.thread_info)
+#define init_stack		(init_thread_union.stack)
+
+/*
+ * Get the thread information struct from C.
+ * We do the usual trick and use the lower end of the stack for this
+ */
+static inline struct thread_info *current_thread_info(void)
+{
+	unsigned long addr = ~(THREAD_SIZE - 1);
+
+	asm("and %0, sp" : "=r"(addr) : "0"(addr));
+	return (struct thread_info *)addr;
+}
+
+/* thread information allocation */
+#define alloc_thread_info(ti) \
+	((struct thread_info *) __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER))
+#define free_thread_info(ti) free_pages((unsigned long)(ti), 1)
+#define get_thread_info(ti) get_task_struct((ti)->task)
+#define put_thread_info(ti) put_task_struct((ti)->task)
+
+#endif /* !__ASSEMBLY__ */
+
+#define PREEMPT_ACTIVE		0x40000000
+
+/*
+ * Thread information flags
+ * - these are process state flags that various assembly files may need to access
+ * - pending work-to-be-done flags are in LSW
+ * - other flags in MSW
+ */
+#define TIF_SYSCALL_TRACE       0       /* syscall trace active */
+#define TIF_NOTIFY_RESUME       1       /* resumption notification requested */
+#define TIF_SIGPENDING          2       /* signal pending */
+#define TIF_NEED_RESCHED        3       /* rescheduling necessary */
+#define TIF_POLLING_NRFLAG      4       /* true if poll_idle() is polling
+					   TIF_NEED_RESCHED */
+#define TIF_BREAKPOINT		5	/* true if we should break after return */
+#define TIF_SINGLE_STEP		6	/* single step after next break */
+#define TIF_MEMDIE		7
+#define TIF_RESTORE_SIGMASK	8	/* restore signal mask in do_signal */
+#define TIF_USERSPACE		31      /* true if FS sets userspace */
+
+#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
+#define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
+#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
+#define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
+#define _TIF_BREAKPOINT		(1 << TIF_BREAKPOINT)
+#define _TIF_SINGLE_STEP	(1 << TIF_SINGLE_STEP)
+#define _TIF_MEMDIE		(1 << TIF_MEMDIE)
+#define _TIF_RESTORE_SIGMASK	(1 << TIF_RESTORE_SIGMASK)
+
+/* XXX: These two masks must never span more than 16 bits! */
+/* work to do on interrupt/exception return */
+#define _TIF_WORK_MASK		0x0000013e
+/* work to do on any return to userspace */
+#define _TIF_ALLWORK_MASK	0x0000013f
+/* work to do on return from debug mode */
+#define _TIF_DBGWORK_MASK	0x0000017e
+
+#endif /* __ASM_AVR32_THREAD_INFO_H */
diff --git a/include/asm-avr32/timex.h b/include/asm-avr32/timex.h
new file mode 100644
index 000000000000..5e44ecb3ce0c
--- /dev/null
+++ b/include/asm-avr32/timex.h
@@ -0,0 +1,40 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TIMEX_H
+#define __ASM_AVR32_TIMEX_H
+
+/*
+ * This is the frequency of the timer used for Linux's timer interrupt.
+ * The value should be defined as accurate as possible or under certain
+ * circumstances Linux timekeeping might become inaccurate or fail.
+ *
+ * For many system the exact clockrate of the timer isn't known but due to
+ * the way this value is used we can get away with a wrong value as long
+ * as this value is:
+ *
+ *  - a multiple of HZ
+ *  - a divisor of the actual rate
+ *
+ * 500000 is a good such cheat value.
+ *
+ * The obscure number 1193182 is the same as used by the original i8254
+ * time in legacy PC hardware; the chip is never found in AVR32 systems.
+ */
+#define CLOCK_TICK_RATE		500000	/* Underlying HZ */
+
+typedef unsigned long cycles_t;
+
+static inline cycles_t get_cycles (void)
+{
+	return 0;
+}
+
+extern int read_current_timer(unsigned long *timer_value);
+#define ARCH_HAS_READ_CURRENT_TIMER	1
+
+#endif /* __ASM_AVR32_TIMEX_H */
diff --git a/include/asm-avr32/tlb.h b/include/asm-avr32/tlb.h
new file mode 100644
index 000000000000..5c55f9ce7c7d
--- /dev/null
+++ b/include/asm-avr32/tlb.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TLB_H
+#define __ASM_AVR32_TLB_H
+
+#define tlb_start_vma(tlb, vma) \
+	flush_cache_range(vma, vma->vm_start, vma->vm_end)
+
+#define tlb_end_vma(tlb, vma) \
+	flush_tlb_range(vma, vma->vm_start, vma->vm_end)
+
+#define __tlb_remove_tlb_entry(tlb, pte, address) do { } while(0)
+
+/*
+ * Flush whole TLB for MM
+ */
+#define tlb_flush(tlb) flush_tlb_mm((tlb)->mm)
+
+#include <asm-generic/tlb.h>
+
+/*
+ * For debugging purposes
+ */
+extern void show_dtlb_entry(unsigned int index);
+extern void dump_dtlb(void);
+
+#endif /* __ASM_AVR32_TLB_H */
diff --git a/include/asm-avr32/tlbflush.h b/include/asm-avr32/tlbflush.h
new file mode 100644
index 000000000000..730e268f81f3
--- /dev/null
+++ b/include/asm-avr32/tlbflush.h
@@ -0,0 +1,40 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TLBFLUSH_H
+#define __ASM_AVR32_TLBFLUSH_H
+
+#include <asm/mmu.h>
+
+/*
+ * TLB flushing:
+ *
+ *  - flush_tlb() flushes the current mm struct TLBs
+ *  - flush_tlb_all() flushes all processes' TLB entries
+ *  - flush_tlb_mm(mm) flushes the specified mm context TLBs
+ *  - flush_tlb_page(vma, vmaddr) flushes one page
+ *  - flush_tlb_range(vma, start, end) flushes a range of pages
+ *  - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
+ *  - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
+ */
+extern void flush_tlb(void);
+extern void flush_tlb_all(void);
+extern void flush_tlb_mm(struct mm_struct *mm);
+extern void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+			    unsigned long end);
+extern void flush_tlb_page(struct vm_area_struct *vma, unsigned long page);
+extern void __flush_tlb_page(unsigned long asid, unsigned long page);
+
+static inline void flush_tlb_pgtables(struct mm_struct *mm,
+				      unsigned long start, unsigned long end)
+{
+	/* Nothing to do */
+}
+
+extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
+
+#endif /* __ASM_AVR32_TLBFLUSH_H */
diff --git a/include/asm-avr32/topology.h b/include/asm-avr32/topology.h
new file mode 100644
index 000000000000..5b766cbb4806
--- /dev/null
+++ b/include/asm-avr32/topology.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_TOPOLOGY_H
+#define __ASM_AVR32_TOPOLOGY_H
+
+#include <asm-generic/topology.h>
+
+#endif /* __ASM_AVR32_TOPOLOGY_H */
diff --git a/include/asm-avr32/traps.h b/include/asm-avr32/traps.h
new file mode 100644
index 000000000000..6a8fb944f414
--- /dev/null
+++ b/include/asm-avr32/traps.h
@@ -0,0 +1,23 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TRAPS_H
+#define __ASM_AVR32_TRAPS_H
+
+#include <linux/list.h>
+
+struct undef_hook {
+	struct list_head node;
+	u32 insn_mask;
+	u32 insn_val;
+	int (*fn)(struct pt_regs *regs, u32 insn);
+};
+
+void register_undef_hook(struct undef_hook *hook);
+void unregister_undef_hook(struct undef_hook *hook);
+
+#endif /* __ASM_AVR32_TRAPS_H */
diff --git a/include/asm-avr32/types.h b/include/asm-avr32/types.h
new file mode 100644
index 000000000000..3f47db9675af
--- /dev/null
+++ b/include/asm-avr32/types.h
@@ -0,0 +1,70 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TYPES_H
+#define __ASM_AVR32_TYPES_H
+
+#ifndef __ASSEMBLY__
+
+typedef unsigned short umode_t;
+
+/*
+ * __xx is ok: it doesn't pollute the POSIX namespace. Use these in the
+ * header files exported to user space
+ */
+typedef __signed__ char __s8;
+typedef unsigned char __u8;
+
+typedef __signed__ short __s16;
+typedef unsigned short __u16;
+
+typedef __signed__ int __s32;
+typedef unsigned int __u32;
+
+#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
+typedef __signed__ long long __s64;
+typedef unsigned long long __u64;
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+/*
+ * These aren't exported outside the kernel to avoid name space clashes
+ */
+#ifdef __KERNEL__
+
+#define BITS_PER_LONG 32
+
+#ifndef __ASSEMBLY__
+
+typedef signed char s8;
+typedef unsigned char u8;
+
+typedef signed short s16;
+typedef unsigned short u16;
+
+typedef signed int s32;
+typedef unsigned int u32;
+
+typedef signed long long s64;
+typedef unsigned long long u64;
+
+/* Dma addresses are 32-bits wide.  */
+
+typedef u32 dma_addr_t;
+
+#ifdef CONFIG_LBD
+typedef u64 sector_t;
+#define HAVE_SECTOR_T
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __KERNEL__ */
+
+
+#endif /* __ASM_AVR32_TYPES_H */
diff --git a/include/asm-avr32/uaccess.h b/include/asm-avr32/uaccess.h
new file mode 100644
index 000000000000..821deb5a9d28
--- /dev/null
+++ b/include/asm-avr32/uaccess.h
@@ -0,0 +1,335 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_UACCESS_H
+#define __ASM_AVR32_UACCESS_H
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+
+#define VERIFY_READ	0
+#define VERIFY_WRITE	1
+
+typedef struct {
+	unsigned int is_user_space;
+} mm_segment_t;
+
+/*
+ * The fs value determines whether argument validity checking should be
+ * performed or not.  If get_fs() == USER_DS, checking is performed, with
+ * get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons (Data Segment Register?), these macros are misnamed.
+ */
+#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
+#define segment_eq(a,b)	((a).is_user_space == (b).is_user_space)
+
+#define USER_ADDR_LIMIT 0x80000000
+
+#define KERNEL_DS	MAKE_MM_SEG(0)
+#define USER_DS		MAKE_MM_SEG(1)
+
+#define get_ds()	(KERNEL_DS)
+
+static inline mm_segment_t get_fs(void)
+{
+	return MAKE_MM_SEG(test_thread_flag(TIF_USERSPACE));
+}
+
+static inline void set_fs(mm_segment_t s)
+{
+	if (s.is_user_space)
+		set_thread_flag(TIF_USERSPACE);
+	else
+		clear_thread_flag(TIF_USERSPACE);
+}
+
+/*
+ * Test whether a block of memory is a valid user space address.
+ * Returns 0 if the range is valid, nonzero otherwise.
+ *
+ * We do the following checks:
+ *   1. Is the access from kernel space?
+ *   2. Does (addr + size) set the carry bit?
+ *   3. Is (addr + size) a negative number (i.e. >= 0x80000000)?
+ *
+ * If yes on the first check, access is granted.
+ * If no on any of the others, access is denied.
+ */
+#define __range_ok(addr, size)						\
+	(test_thread_flag(TIF_USERSPACE)				\
+	 && (((unsigned long)(addr) >= 0x80000000)			\
+	     || ((unsigned long)(size) > 0x80000000)			\
+	     || (((unsigned long)(addr) + (unsigned long)(size)) > 0x80000000)))
+
+#define access_ok(type, addr, size) (likely(__range_ok(addr, size) == 0))
+
+static inline int
+verify_area(int type, const void __user *addr, unsigned long size)
+{
+	return access_ok(type, addr, size) ? 0 : -EFAULT;
+}
+
+/* Generic arbitrary sized copy. Return the number of bytes NOT copied */
+extern __kernel_size_t __copy_user(void *to, const void *from,
+				   __kernel_size_t n);
+
+extern __kernel_size_t copy_to_user(void __user *to, const void *from,
+				    __kernel_size_t n);
+extern __kernel_size_t copy_from_user(void *to, const void __user *from,
+				      __kernel_size_t n);
+
+static inline __kernel_size_t __copy_to_user(void __user *to, const void *from,
+					     __kernel_size_t n)
+{
+	return __copy_user((void __force *)to, from, n);
+}
+static inline __kernel_size_t __copy_from_user(void *to,
+					       const void __user *from,
+					       __kernel_size_t n)
+{
+	return __copy_user(to, (const void __force *)from, n);
+}
+
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+/*
+ * put_user: - Write a simple value into user space.
+ * @x:   Value to copy to user space.
+ * @ptr: Destination address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple value from kernel space to user
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and @x must be assignable
+ * to the result of dereferencing @ptr.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ */
+#define put_user(x,ptr)	\
+	__put_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * get_user: - Get a simple variable from user space.
+ * @x:   Variable to store result.
+ * @ptr: Source address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple variable from user space to kernel
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and the result of
+ * dereferencing @ptr must be assignable to @x without a cast.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ * On error, the variable @x is set to zero.
+ */
+#define get_user(x,ptr) \
+	__get_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * __put_user: - Write a simple value into user space, with less checking.
+ * @x:   Value to copy to user space.
+ * @ptr: Destination address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple value from kernel space to user
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and @x must be assignable
+ * to the result of dereferencing @ptr.
+ *
+ * Caller must check the pointer with access_ok() before calling this
+ * function.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ */
+#define __put_user(x,ptr) \
+	__put_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * __get_user: - Get a simple variable from user space, with less checking.
+ * @x:   Variable to store result.
+ * @ptr: Source address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple variable from user space to kernel
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and the result of
+ * dereferencing @ptr must be assignable to @x without a cast.
+ *
+ * Caller must check the pointer with access_ok() before calling this
+ * function.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ * On error, the variable @x is set to zero.
+ */
+#define __get_user(x,ptr) \
+	__get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+extern int __get_user_bad(void);
+extern int __put_user_bad(void);
+
+#define __get_user_nocheck(x, ptr, size)				\
+({									\
+	typeof(*(ptr)) __gu_val = (typeof(*(ptr)) __force)0;		\
+	int __gu_err = 0;						\
+									\
+	switch (size) {							\
+	case 1: __get_user_asm("ub", __gu_val, ptr, __gu_err); break;	\
+	case 2: __get_user_asm("uh", __gu_val, ptr, __gu_err); break;	\
+	case 4: __get_user_asm("w", __gu_val, ptr, __gu_err); break;	\
+	case 8: __get_user_asm("d", __gu_val, ptr, __gu_err); break;	\
+	default: __gu_err = __get_user_bad(); break;			\
+	}								\
+									\
+	x = __gu_val;							\
+	__gu_err;							\
+})
+
+#define __get_user_check(x, ptr, size)					\
+({									\
+	typeof(*(ptr)) __gu_val = (typeof(*(ptr)) __force)0;		\
+	const typeof(*(ptr)) __user * __gu_addr = (ptr);		\
+	int __gu_err = 0;						\
+									\
+	if (access_ok(VERIFY_READ, __gu_addr, size)) {			\
+		switch (size) {						\
+		case 1:							\
+			__get_user_asm("ub", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		case 2:							\
+			__get_user_asm("uh", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		case 4:							\
+			__get_user_asm("w", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		case 8:							\
+			__get_user_asm("d", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		default:						\
+			__gu_err = __get_user_bad();			\
+			break;						\
+		}							\
+	} else {							\
+		__gu_err = -EFAULT;					\
+	}								\
+	x = __gu_val;							\
+	__gu_err;							\
+})
+
+#define __get_user_asm(suffix, __gu_val, ptr, __gu_err)			\
+	asm volatile(							\
+		"1:	ld." suffix "	%1, %3			\n"	\
+		"2:						\n"	\
+		"	.section .fixup, \"ax\"			\n"	\
+		"3:	mov	%0, %4				\n"	\
+		"	rjmp	2b				\n"	\
+		"	.previous				\n"	\
+		"	.section __ex_table, \"a\"		\n"	\
+		"	.long	1b, 3b				\n"	\
+		"	.previous				\n"	\
+		: "=r"(__gu_err), "=r"(__gu_val)			\
+		: "0"(__gu_err), "m"(*(ptr)), "i"(-EFAULT))
+
+#define __put_user_nocheck(x, ptr, size)				\
+({									\
+	typeof(*(ptr)) __pu_val;					\
+	int __pu_err = 0;						\
+									\
+	__pu_val = (x);							\
+	switch (size) {							\
+	case 1: __put_user_asm("b", ptr, __pu_val, __pu_err); break;	\
+	case 2: __put_user_asm("h", ptr, __pu_val, __pu_err); break;	\
+	case 4: __put_user_asm("w", ptr, __pu_val, __pu_err); break;	\
+	case 8: __put_user_asm("d", ptr, __pu_val, __pu_err); break;	\
+	default: __pu_err = __put_user_bad(); break;			\
+	}								\
+	__pu_err;							\
+})
+
+#define __put_user_check(x, ptr, size)					\
+({									\
+	typeof(*(ptr)) __pu_val;					\
+	typeof(*(ptr)) __user *__pu_addr = (ptr);			\
+	int __pu_err = 0;						\
+									\
+	__pu_val = (x);							\
+	if (access_ok(VERIFY_WRITE, __pu_addr, size)) {			\
+		switch (size) {						\
+		case 1:							\
+			__put_user_asm("b", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 2:							\
+			__put_user_asm("h", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 4:							\
+			__put_user_asm("w", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 8:							\
+			__put_user_asm("d", __pu_addr, __pu_val,		\
+				       __pu_err);			\
+			break;						\
+		default:						\
+			__pu_err = __put_user_bad();			\
+			break;						\
+		}							\
+	} else {							\
+		__pu_err = -EFAULT;					\
+	}								\
+	__pu_err;							\
+})
+
+#define __put_user_asm(suffix, ptr, __pu_val, __gu_err)			\
+	asm volatile(							\
+		"1:	st." suffix "	%1, %3			\n"	\
+		"2:						\n"	\
+		"	.section .fixup, \"ax\"			\n"	\
+		"3:	mov	%0, %4				\n"	\
+		"	rjmp	2b				\n"	\
+		"	.previous				\n"	\
+		"	.section __ex_table, \"a\"		\n"	\
+		"	.long	1b, 3b				\n"	\
+		"	.previous				\n"	\
+		: "=r"(__gu_err), "=m"(*(ptr))				\
+		: "0"(__gu_err), "r"(__pu_val), "i"(-EFAULT))
+
+extern __kernel_size_t clear_user(void __user *addr, __kernel_size_t size);
+extern __kernel_size_t __clear_user(void __user *addr, __kernel_size_t size);
+
+extern long strncpy_from_user(char *dst, const char __user *src, long count);
+extern long __strncpy_from_user(char *dst, const char __user *src, long count);
+
+extern long strnlen_user(const char __user *__s, long __n);
+extern long __strnlen_user(const char __user *__s, long __n);
+
+#define strlen_user(s) strnlen_user(s, ~0UL >> 1)
+
+struct exception_table_entry
+{
+	unsigned long insn, fixup;
+};
+
+#endif /* __ASM_AVR32_UACCESS_H */
diff --git a/include/asm-avr32/ucontext.h b/include/asm-avr32/ucontext.h
new file mode 100644
index 000000000000..ac7259c2a799
--- /dev/null
+++ b/include/asm-avr32/ucontext.h
@@ -0,0 +1,12 @@
+#ifndef __ASM_AVR32_UCONTEXT_H
+#define __ASM_AVR32_UCONTEXT_H
+
+struct ucontext {
+	unsigned long		uc_flags;
+	struct ucontext	*	uc_link;
+	stack_t			uc_stack;
+	struct sigcontext	uc_mcontext;
+	sigset_t		uc_sigmask;
+};
+
+#endif /* __ASM_AVR32_UCONTEXT_H */
diff --git a/include/asm-avr32/unaligned.h b/include/asm-avr32/unaligned.h
new file mode 100644
index 000000000000..3042723fcbfd
--- /dev/null
+++ b/include/asm-avr32/unaligned.h
@@ -0,0 +1,25 @@
+#ifndef __ASM_AVR32_UNALIGNED_H
+#define __ASM_AVR32_UNALIGNED_H
+
+/*
+ * AVR32 can handle some unaligned accesses, depending on the
+ * implementation.  The AVR32 AP implementation can handle unaligned
+ * words, but halfwords must be halfword-aligned, and doublewords must
+ * be word-aligned.
+ *
+ * TODO: Make all this CPU-specific and optimize.
+ */
+
+#include <linux/string.h>
+
+/* Use memmove here, so gcc does not insert a __builtin_memcpy. */
+
+#define get_unaligned(ptr) \
+  ({ __typeof__(*(ptr)) __tmp; memmove(&__tmp, (ptr), sizeof(*(ptr))); __tmp; })
+
+#define put_unaligned(val, ptr)				\
+  ({ __typeof__(*(ptr)) __tmp = (val);			\
+     memmove((ptr), &__tmp, sizeof(*(ptr)));		\
+     (void)0; })
+
+#endif /* __ASM_AVR32_UNALIGNED_H */
diff --git a/include/asm-avr32/unistd.h b/include/asm-avr32/unistd.h
new file mode 100644
index 000000000000..1f528f92690d
--- /dev/null
+++ b/include/asm-avr32/unistd.h
@@ -0,0 +1,387 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_UNISTD_H
+#define __ASM_AVR32_UNISTD_H
+
+/*
+ * This file contains the system call numbers.
+ */
+
+#define __NR_restart_syscall      0
+#define __NR_exit		  1
+#define __NR_fork		  2
+#define __NR_read		  3
+#define __NR_write		  4
+#define __NR_open		  5
+#define __NR_close		  6
+#define __NR_umask		  7
+#define __NR_creat		  8
+#define __NR_link		  9
+#define __NR_unlink		 10
+#define __NR_execve		 11
+#define __NR_chdir		 12
+#define __NR_time		 13
+#define __NR_mknod		 14
+#define __NR_chmod		 15
+#define __NR_chown		 16
+#define __NR_lchown		 17
+#define __NR_lseek		 18
+#define __NR__llseek		 19
+#define __NR_getpid		 20
+#define __NR_mount		 21
+#define __NR_umount2		 22
+#define __NR_setuid		 23
+#define __NR_getuid		 24
+#define __NR_stime		 25
+#define __NR_ptrace		 26
+#define __NR_alarm		 27
+#define __NR_pause		 28
+#define __NR_utime		 29
+#define __NR_stat		 30
+#define __NR_fstat		 31
+#define __NR_lstat		 32
+#define __NR_access		 33
+#define __NR_chroot		 34
+#define __NR_sync		 35
+#define __NR_fsync		 36
+#define __NR_kill		 37
+#define __NR_rename		 38
+#define __NR_mkdir		 39
+#define __NR_rmdir		 40
+#define __NR_dup		 41
+#define __NR_pipe		 42
+#define __NR_times		 43
+#define __NR_clone		 44
+#define __NR_brk		 45
+#define __NR_setgid		 46
+#define __NR_getgid		 47
+#define __NR_getcwd		 48
+#define __NR_geteuid		 49
+#define __NR_getegid		 50
+#define __NR_acct		 51
+#define __NR_setfsuid		 52
+#define __NR_setfsgid		 53
+#define __NR_ioctl		 54
+#define __NR_fcntl		 55
+#define __NR_setpgid		 56
+#define __NR_mremap		 57
+#define __NR_setresuid		 58
+#define __NR_getresuid		 59
+#define __NR_setreuid		 60
+#define __NR_setregid		 61
+#define __NR_ustat		 62
+#define __NR_dup2		 63
+#define __NR_getppid		 64
+#define __NR_getpgrp		 65
+#define __NR_setsid		 66
+#define __NR_rt_sigaction	 67
+#define __NR_rt_sigreturn	 68
+#define __NR_rt_sigprocmask	 69
+#define __NR_rt_sigpending	 70
+#define __NR_rt_sigtimedwait	 71
+#define __NR_rt_sigqueueinfo	 72
+#define __NR_rt_sigsuspend	 73
+#define __NR_sethostname	 74
+#define __NR_setrlimit		 75
+#define __NR_getrlimit		 76	/* SuS compliant getrlimit */
+#define __NR_getrusage		 77
+#define __NR_gettimeofday	 78
+#define __NR_settimeofday	 79
+#define __NR_getgroups		 80
+#define __NR_setgroups		 81
+#define __NR_select		 82
+#define __NR_symlink		 83
+#define __NR_fchdir		 84
+#define __NR_readlink		 85
+#define __NR_pread		 86
+#define __NR_pwrite		 87
+#define __NR_swapon		 88
+#define __NR_reboot		 89
+#define __NR_mmap2		 90
+#define __NR_munmap		 91
+#define __NR_truncate		 92
+#define __NR_ftruncate		 93
+#define __NR_fchmod		 94
+#define __NR_fchown		 95
+#define __NR_getpriority	 96
+#define __NR_setpriority	 97
+#define __NR_wait4		 98
+#define __NR_statfs		 99
+#define __NR_fstatfs		100
+#define __NR_vhangup		101
+#define __NR_sigaltstack	102
+#define __NR_syslog		103
+#define __NR_setitimer		104
+#define __NR_getitimer		105
+#define __NR_swapoff		106
+#define __NR_sysinfo		107
+#define __NR_ipc		108
+#define __NR_sendfile		109
+#define __NR_setdomainname	110
+#define __NR_uname		111
+#define __NR_adjtimex		112
+#define __NR_mprotect		113
+#define __NR_vfork		114
+#define __NR_init_module	115
+#define __NR_delete_module	116
+#define __NR_quotactl		117
+#define __NR_getpgid		118
+#define __NR_bdflush		119
+#define __NR_sysfs		120
+#define __NR_personality	121
+#define __NR_afs_syscall	122 /* Syscall for Andrew File System */
+#define __NR_getdents		123
+#define __NR_flock		124
+#define __NR_msync		125
+#define __NR_readv		126
+#define __NR_writev		127
+#define __NR_getsid		128
+#define __NR_fdatasync		129
+#define __NR__sysctl		130
+#define __NR_mlock		131
+#define __NR_munlock		132
+#define __NR_mlockall		133
+#define __NR_munlockall		134
+#define __NR_sched_setparam		135
+#define __NR_sched_getparam		136
+#define __NR_sched_setscheduler		137
+#define __NR_sched_getscheduler		138
+#define __NR_sched_yield		139
+#define __NR_sched_get_priority_max	140
+#define __NR_sched_get_priority_min	141
+#define __NR_sched_rr_get_interval	142
+#define __NR_nanosleep		143
+#define __NR_poll		144
+#define __NR_nfsservctl		145
+#define __NR_setresgid		146
+#define __NR_getresgid		147
+#define __NR_prctl              148
+#define __NR_socket		149
+#define __NR_bind		150
+#define __NR_connect		151
+#define __NR_listen		152
+#define __NR_accept		153
+#define __NR_getsockname	154
+#define __NR_getpeername	155
+#define __NR_socketpair		156
+#define __NR_send		157
+#define __NR_recv		158
+#define __NR_sendto		159
+#define __NR_recvfrom		160
+#define __NR_shutdown		161
+#define __NR_setsockopt		162
+#define __NR_getsockopt		163
+#define __NR_sendmsg		164
+#define __NR_recvmsg		165
+#define __NR_truncate64		166
+#define __NR_ftruncate64	167
+#define __NR_stat64		168
+#define __NR_lstat64		169
+#define __NR_fstat64		170
+#define __NR_pivot_root		171
+#define __NR_mincore		172
+#define __NR_madvise		173
+#define __NR_getdents64		174
+#define __NR_fcntl64		175
+#define __NR_gettid		176
+#define __NR_readahead		177
+#define __NR_setxattr		178
+#define __NR_lsetxattr		179
+#define __NR_fsetxattr		180
+#define __NR_getxattr		181
+#define __NR_lgetxattr		182
+#define __NR_fgetxattr		183
+#define __NR_listxattr		184
+#define __NR_llistxattr		185
+#define __NR_flistxattr		186
+#define __NR_removexattr	187
+#define __NR_lremovexattr	188
+#define __NR_fremovexattr	189
+#define __NR_tkill		190
+#define __NR_sendfile64		191
+#define __NR_futex		192
+#define __NR_sched_setaffinity	193
+#define __NR_sched_getaffinity	194
+#define __NR_capget		195
+#define __NR_capset		196
+#define __NR_io_setup		197
+#define __NR_io_destroy		198
+#define __NR_io_getevents	199
+#define __NR_io_submit		200
+#define __NR_io_cancel		201
+#define __NR_fadvise64		202
+#define __NR_exit_group		203
+#define __NR_lookup_dcookie	204
+#define __NR_epoll_create	205
+#define __NR_epoll_ctl		206
+#define __NR_epoll_wait		207
+#define __NR_remap_file_pages	208
+#define __NR_set_tid_address	209
+
+#define __NR_timer_create	210
+#define __NR_timer_settime	211
+#define __NR_timer_gettime	212
+#define __NR_timer_getoverrun	213
+#define __NR_timer_delete	214
+#define __NR_clock_settime	215
+#define __NR_clock_gettime	216
+#define __NR_clock_getres	217
+#define __NR_clock_nanosleep	218
+#define __NR_statfs64		219
+#define __NR_fstatfs64		220
+#define __NR_tgkill		221
+				/* 222 reserved for tux */
+#define __NR_utimes		223
+#define __NR_fadvise64_64	224
+
+#define __NR_cacheflush		225
+
+#define __NR_vserver		226
+#define __NR_mq_open		227
+#define __NR_mq_unlink		228
+#define __NR_mq_timedsend	229
+#define __NR_mq_timedreceive	230
+#define __NR_mq_notify		231
+#define __NR_mq_getsetattr	232
+#define __NR_kexec_load		233
+#define __NR_waitid		234
+#define __NR_add_key		235
+#define __NR_request_key	236
+#define __NR_keyctl		237
+#define __NR_ioprio_set		238
+#define __NR_ioprio_get		239
+#define __NR_inotify_init	240
+#define __NR_inotify_add_watch	241
+#define __NR_inotify_rm_watch	242
+#define __NR_openat		243
+#define __NR_mkdirat		244
+#define __NR_mknodat		245
+#define __NR_fchownat		246
+#define __NR_futimesat		247
+#define __NR_fstatat64		248
+#define __NR_unlinkat		249
+#define __NR_renameat		250
+#define __NR_linkat		251
+#define __NR_symlinkat		252
+#define __NR_readlinkat		253
+#define __NR_fchmodat		254
+#define __NR_faccessat		255
+#define __NR_pselect6		256
+#define __NR_ppoll		257
+#define __NR_unshare		258
+#define __NR_set_robust_list	259
+#define __NR_get_robust_list	260
+#define __NR_splice		261
+#define __NR_sync_file_range	262
+#define __NR_tee		263
+#define __NR_vmsplice		264
+
+#define NR_syscalls		265
+
+
+/*
+ * AVR32 calling convention for system calls:
+ *   - System call number in r8
+ *   - Parameters in r12 and downwards to r9 as well as r6 and r5.
+ *   - Return value in r12
+ */
+
+/*
+ * user-visible error numbers are in the range -1 - -124: see
+ * <asm-generic/errno.h>
+ */
+
+#define __syscall_return(type, res) do {				\
+		if ((unsigned long)(res) >= (unsigned long)(-125)) {	\
+			errno = -(res);					\
+			res = -1;					\
+		}							\
+		return (type) (res);					\
+	} while (0)
+
+#ifdef __KERNEL__
+#define __ARCH_WANT_IPC_PARSE_VERSION
+#define __ARCH_WANT_STAT64
+#define __ARCH_WANT_SYS_ALARM
+#define __ARCH_WANT_SYS_GETHOSTNAME
+#define __ARCH_WANT_SYS_PAUSE
+#define __ARCH_WANT_SYS_TIME
+#define __ARCH_WANT_SYS_UTIME
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
+#endif
+
+#if defined(__KERNEL_SYSCALLS__) || defined(__CHECKER__)
+
+#include <linux/types.h>
+#include <linux/linkage.h>
+#include <asm/signal.h>
+
+struct pt_regs;
+
+/*
+ * we need this inline - forking from kernel space will result
+ * in NO COPY ON WRITE (!!!), until an execve is executed. This
+ * is no problem, but for the stack. This is handled by not letting
+ * main() use the stack at all after fork(). Thus, no function
+ * calls - which means inline code for fork too, as otherwise we
+ * would use the stack upon exit from 'fork()'.
+ *
+ * Actually only pause and fork are needed inline, so that there
+ * won't be any messing with the stack from main(), but we define
+ * some others too.
+ */
+static inline int execve(const char *file, char **argv, char **envp)
+{
+	register long scno asm("r8") = __NR_execve;
+	register long sc1 asm("r12") = (long)file;
+	register long sc2 asm("r11") = (long)argv;
+	register long sc3 asm("r10") = (long)envp;
+	int res;
+
+	asm volatile("scall"
+		     : "=r"(sc1)
+		     : "r"(scno), "0"(sc1), "r"(sc2), "r"(sc3)
+		     : "lr", "memory");
+	res = sc1;
+	__syscall_return(int, res);
+}
+
+asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize);
+asmlinkage int sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+			       struct pt_regs *regs);
+asmlinkage int sys_rt_sigreturn(struct pt_regs *regs);
+asmlinkage int sys_pipe(unsigned long __user *filedes);
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+			  unsigned long prot, unsigned long flags,
+			  unsigned long fd, off_t offset);
+asmlinkage int sys_cacheflush(int operation, void __user *addr, size_t len);
+asmlinkage int sys_fork(struct pt_regs *regs);
+asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
+			 unsigned long parent_tidptr,
+			 unsigned long child_tidptr, struct pt_regs *regs);
+asmlinkage int sys_vfork(struct pt_regs *regs);
+asmlinkage int sys_execve(char __user *ufilename, char __user *__user *uargv,
+			  char __user *__user *uenvp, struct pt_regs *regs);
+
+#endif
+
+/*
+ * "Conditional" syscalls
+ *
+ * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
+ * but it doesn't work on all toolchains, so we just do it by hand
+ */
+#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall");
+
+#endif /* __ASM_AVR32_UNISTD_H */
diff --git a/include/asm-avr32/user.h b/include/asm-avr32/user.h
new file mode 100644
index 000000000000..060fb3acee49
--- /dev/null
+++ b/include/asm-avr32/user.h
@@ -0,0 +1,65 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Note: We may not need these definitions for AVR32, as we don't
+ * support a.out.
+ */
+#ifndef __ASM_AVR32_USER_H
+#define __ASM_AVR32_USER_H
+
+#include <linux/types.h>
+#include <asm/ptrace.h>
+#include <asm/page.h>
+
+/*
+ * Core file format: The core file is written in such a way that gdb
+ * can understand it and provide useful information to the user (under
+ * linux we use the `trad-core' bfd).  The file contents are as follows:
+ *
+ *  upage: 1 page consisting of a user struct that tells gdb
+ *	what is present in the file.  Directly after this is a
+ *	copy of the task_struct, which is currently not used by gdb,
+ *	but it may come in handy at some point.  All of the registers
+ *	are stored as part of the upage.  The upage should always be
+ *	only one page long.
+ *  data: The data segment follows next.  We use current->end_text to
+ *	current->brk to pick up all of the user variables, plus any memory
+ *	that may have been sbrk'ed.  No attempt is made to determine if a
+ *	page is demand-zero or if a page is totally unused, we just cover
+ *	the entire range.  All of the addresses are rounded in such a way
+ *	that an integral number of pages is written.
+ *  stack: We need the stack information in order to get a meaningful
+ *	backtrace.  We need to write the data from usp to
+ *	current->start_stack, so we round each of these in order to be able
+ *	to write an integer number of pages.
+ */
+
+struct user_fpu_struct {
+	/* We have no FPU (yet) */
+};
+
+struct user {
+	struct pt_regs	regs;			/* entire machine state */
+	size_t		u_tsize;		/* text size (pages) */
+	size_t		u_dsize;		/* data size (pages) */
+	size_t		u_ssize;		/* stack size (pages) */
+	unsigned long	start_code;		/* text starting address */
+	unsigned long	start_data;		/* data starting address */
+	unsigned long	start_stack;		/* stack starting address */
+	long int	signal;			/* signal causing core dump */
+	struct regs *	u_ar0;			/* help gdb find registers */
+	unsigned long	magic;			/* identifies a core file */
+	char		u_comm[32];		/* user command name */
+};
+
+#define NBPG			PAGE_SIZE
+#define UPAGES			1
+#define HOST_TEXT_START_ADDR	(u.start_code)
+#define HOST_DATA_START_ADDR	(u.start_data)
+#define HOST_STACK_END_ADDR	(u.start_stack + u.u_ssize * NBPG)
+
+#endif /* __ASM_AVR32_USER_H */
diff --git a/include/linux/elf-em.h b/include/linux/elf-em.h
index 6a5796c81c90..666e0a5f00fc 100644
--- a/include/linux/elf-em.h
+++ b/include/linux/elf-em.h
@@ -31,6 +31,7 @@
 #define EM_M32R		88	/* Renesas M32R */
 #define EM_H8_300	46	/* Renesas H8/300,300H,H8S */
 #define EM_FRV		0x5441	/* Fujitsu FR-V */
+#define EM_AVR32	0x18ad	/* Atmel AVR32 */
 
 /*
  * This is an interim value that we will use until the committee comes
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 554ee688a9f8..3f21cc79a134 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -277,7 +277,7 @@ config DEBUG_HIGHMEM
 config DEBUG_BUGVERBOSE
 	bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED
 	depends on BUG
-	depends on ARM || ARM26 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV
+	depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV
 	default !EMBEDDED
 	help
 	  Say Y here to make BUG() panics output the file name and line number
@@ -315,7 +315,7 @@ config DEBUG_VM
 
 config FRAME_POINTER
 	bool "Compile the kernel with frame pointers"
-	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390)
+	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390 || AVR32)
 	default y if DEBUG_INFO && UML
 	help
 	  If you say Y here the resulting kernel image will be slightly larger
-- 
cgit v1.2.3


From bc157b75960f1f33566074e820342690216629b9 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Mon, 25 Sep 2006 23:32:16 -0700
Subject: [PATCH] AVR32 MTD: Static Memory Controller driver

This patchset adds the necessary drivers and infrastructure to access the
external flash on the ATSTK1000 board through the MTD subsystem.  With this
stuff in place, it will be possible to use a jffs2 filesystem stored in the
external flash as a root filesystem.  It might also be possible to update the
boot loader if you drop the write protection of partition 0.

As suggested by David Woodhouse, I reworked the patches to use the physmap
driver instead of introducing a separate mapping driver for the ATSTK1000.
I've also cleaned up the hsmc header by removing useless comments and
converting spaces to tabs (my headerfile generator needs some work.)

Unfortunately, I couldn't unlock the flash in fixup_use_atmel_lock because the
erase regions hadn't been set up yet, so I had to do it from cfi_amdstd_setup
instead.

This patch:

This adds a simple API for configuring the static memory controller along with
an implementation for the Atmel HSMC.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/avr32/mach-at32ap/Makefile     |   2 +-
 arch/avr32/mach-at32ap/at32ap7000.c |  10 +++
 arch/avr32/mach-at32ap/hsmc.c       | 164 ++++++++++++++++++++++++++++++++++++
 arch/avr32/mach-at32ap/hsmc.h       | 127 ++++++++++++++++++++++++++++
 include/asm-avr32/arch-at32ap/smc.h |  60 +++++++++++++
 5 files changed, 362 insertions(+), 1 deletion(-)
 create mode 100644 arch/avr32/mach-at32ap/hsmc.c
 create mode 100644 arch/avr32/mach-at32ap/hsmc.h
 create mode 100644 include/asm-avr32/arch-at32ap/smc.h

(limited to 'include')

diff --git a/arch/avr32/mach-at32ap/Makefile b/arch/avr32/mach-at32ap/Makefile
index 4b10853eb614..f62eb6915510 100644
--- a/arch/avr32/mach-at32ap/Makefile
+++ b/arch/avr32/mach-at32ap/Makefile
@@ -1,2 +1,2 @@
-obj-y				+= at32ap.o clock.o pio.o intc.o extint.o
+obj-y				+= at32ap.o clock.o pio.o intc.o extint.o hsmc.o
 obj-$(CONFIG_CPU_AT32AP7000)	+= at32ap7000.o
diff --git a/arch/avr32/mach-at32ap/at32ap7000.c b/arch/avr32/mach-at32ap/at32ap7000.c
index e8c6893a1c23..37982b60398e 100644
--- a/arch/avr32/mach-at32ap/at32ap7000.c
+++ b/arch/avr32/mach-at32ap/at32ap7000.c
@@ -450,6 +450,13 @@ static struct clk hramc_clk = {
 	.users		= 1,
 };
 
+static struct resource smc0_resource[] = {
+	PBMEM(0xfff03400),
+};
+DEFINE_DEV(smc, 0);
+DEV_CLK(pclk, smc0, pbb, 13);
+DEV_CLK(mck, smc0, hsb, 0);
+
 static struct platform_device pdc_device = {
 	.name		= "pdc",
 	.id		= 0,
@@ -503,6 +510,7 @@ void __init at32_add_system_devices(void)
 
 	platform_device_register(&at32_sm_device);
 	platform_device_register(&at32_intc0_device);
+	platform_device_register(&smc0_device);
 	platform_device_register(&pdc_device);
 
 	platform_device_register(&pio0_device);
@@ -796,6 +804,8 @@ struct clk *at32_clock_list[] = {
 	&at32_intc0_pclk,
 	&ebi_clk,
 	&hramc_clk,
+	&smc0_pclk,
+	&smc0_mck,
 	&pdc_hclk,
 	&pdc_pclk,
 	&pico_clk,
diff --git a/arch/avr32/mach-at32ap/hsmc.c b/arch/avr32/mach-at32ap/hsmc.c
new file mode 100644
index 000000000000..7691721928a7
--- /dev/null
+++ b/arch/avr32/mach-at32ap/hsmc.c
@@ -0,0 +1,164 @@
+/*
+ * Static Memory Controller for AT32 chips
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#define DEBUG
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+#include <asm/arch/smc.h>
+
+#include "hsmc.h"
+
+#define NR_CHIP_SELECTS 6
+
+struct hsmc {
+	void __iomem *regs;
+	struct clk *pclk;
+	struct clk *mck;
+};
+
+static struct hsmc *hsmc;
+
+int smc_set_configuration(int cs, const struct smc_config *config)
+{
+	unsigned long mul;
+	unsigned long offset;
+	u32 setup, pulse, cycle, mode;
+
+	if (!hsmc)
+		return -ENODEV;
+	if (cs >= NR_CHIP_SELECTS)
+		return -EINVAL;
+
+	/*
+	 * cycles = x / T = x * f
+	 *   = ((x * 1000000000) * ((f * 65536) / 1000000000)) / 65536
+	 *   = ((x * 1000000000) * (((f / 10000) * 65536) / 100000)) / 65536
+	 */
+	mul = (clk_get_rate(hsmc->mck) / 10000) << 16;
+	mul /= 100000;
+
+#define ns2cyc(x) ((((x) * mul) + 65535) >> 16)
+
+	setup = (HSMC_BF(NWE_SETUP, ns2cyc(config->nwe_setup))
+		 | HSMC_BF(NCS_WR_SETUP, ns2cyc(config->ncs_write_setup))
+		 | HSMC_BF(NRD_SETUP, ns2cyc(config->nrd_setup))
+		 | HSMC_BF(NCS_RD_SETUP, ns2cyc(config->ncs_read_setup)));
+	pulse = (HSMC_BF(NWE_PULSE, ns2cyc(config->nwe_pulse))
+		 | HSMC_BF(NCS_WR_PULSE, ns2cyc(config->ncs_write_pulse))
+		 | HSMC_BF(NRD_PULSE, ns2cyc(config->nrd_pulse))
+		 | HSMC_BF(NCS_RD_PULSE, ns2cyc(config->ncs_read_pulse)));
+	cycle = (HSMC_BF(NWE_CYCLE, ns2cyc(config->write_cycle))
+		 | HSMC_BF(NRD_CYCLE, ns2cyc(config->read_cycle)));
+
+	switch (config->bus_width) {
+	case 1:
+		mode = HSMC_BF(DBW, HSMC_DBW_8_BITS);
+		break;
+	case 2:
+		mode = HSMC_BF(DBW, HSMC_DBW_16_BITS);
+		break;
+	case 4:
+		mode = HSMC_BF(DBW, HSMC_DBW_32_BITS);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (config->nrd_controlled)
+		mode |= HSMC_BIT(READ_MODE);
+	if (config->nwe_controlled)
+		mode |= HSMC_BIT(WRITE_MODE);
+	if (config->byte_write)
+		mode |= HSMC_BIT(BAT);
+
+	pr_debug("smc cs%d: setup/%08x pulse/%08x cycle/%08x mode/%08x\n",
+		 cs, setup, pulse, cycle, mode);
+
+	offset = cs * 0x10;
+	hsmc_writel(hsmc, SETUP0 + offset, setup);
+	hsmc_writel(hsmc, PULSE0 + offset, pulse);
+	hsmc_writel(hsmc, CYCLE0 + offset, cycle);
+	hsmc_writel(hsmc, MODE0 + offset, mode);
+	hsmc_readl(hsmc, MODE0); /* I/O barrier */
+
+	return 0;
+}
+EXPORT_SYMBOL(smc_set_configuration);
+
+static int hsmc_probe(struct platform_device *pdev)
+{
+	struct resource *regs;
+	struct clk *pclk, *mck;
+	int ret;
+
+	if (hsmc)
+		return -EBUSY;
+
+	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!regs)
+		return -ENXIO;
+	pclk = clk_get(&pdev->dev, "pclk");
+	if (IS_ERR(pclk))
+		return PTR_ERR(pclk);
+	mck = clk_get(&pdev->dev, "mck");
+	if (IS_ERR(mck)) {
+		ret = PTR_ERR(mck);
+		goto out_put_pclk;
+	}
+
+	ret = -ENOMEM;
+	hsmc = kzalloc(sizeof(struct hsmc), GFP_KERNEL);
+	if (!hsmc)
+		goto out_put_clocks;
+
+	clk_enable(pclk);
+	clk_enable(mck);
+
+	hsmc->pclk = pclk;
+	hsmc->mck = mck;
+	hsmc->regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!hsmc->regs)
+		goto out_disable_clocks;
+
+	dev_info(&pdev->dev, "Atmel Static Memory Controller at 0x%08lx\n",
+		 (unsigned long)regs->start);
+
+	platform_set_drvdata(pdev, hsmc);
+
+	return 0;
+
+out_disable_clocks:
+	clk_disable(mck);
+	clk_disable(pclk);
+	kfree(hsmc);
+out_put_clocks:
+	clk_put(mck);
+out_put_pclk:
+	clk_put(pclk);
+	hsmc = NULL;
+	return ret;
+}
+
+static struct platform_driver hsmc_driver = {
+	.probe		= hsmc_probe,
+	.driver		= {
+		.name	= "smc",
+	},
+};
+
+static int __init hsmc_init(void)
+{
+	return platform_driver_register(&hsmc_driver);
+}
+arch_initcall(hsmc_init);
diff --git a/arch/avr32/mach-at32ap/hsmc.h b/arch/avr32/mach-at32ap/hsmc.h
new file mode 100644
index 000000000000..5681276fafdb
--- /dev/null
+++ b/arch/avr32/mach-at32ap/hsmc.h
@@ -0,0 +1,127 @@
+/*
+ * Register definitions for Atmel Static Memory Controller (SMC)
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_HSMC_H__
+#define __ASM_AVR32_HSMC_H__
+
+/* HSMC register offsets */
+#define HSMC_SETUP0				0x0000
+#define HSMC_PULSE0				0x0004
+#define HSMC_CYCLE0				0x0008
+#define HSMC_MODE0				0x000c
+#define HSMC_SETUP1				0x0010
+#define HSMC_PULSE1				0x0014
+#define HSMC_CYCLE1				0x0018
+#define HSMC_MODE1				0x001c
+#define HSMC_SETUP2				0x0020
+#define HSMC_PULSE2				0x0024
+#define HSMC_CYCLE2				0x0028
+#define HSMC_MODE2				0x002c
+#define HSMC_SETUP3				0x0030
+#define HSMC_PULSE3				0x0034
+#define HSMC_CYCLE3				0x0038
+#define HSMC_MODE3				0x003c
+#define HSMC_SETUP4				0x0040
+#define HSMC_PULSE4				0x0044
+#define HSMC_CYCLE4				0x0048
+#define HSMC_MODE4				0x004c
+#define HSMC_SETUP5				0x0050
+#define HSMC_PULSE5				0x0054
+#define HSMC_CYCLE5				0x0058
+#define HSMC_MODE5				0x005c
+
+/* Bitfields in SETUP0 */
+#define HSMC_NWE_SETUP_OFFSET			0
+#define HSMC_NWE_SETUP_SIZE			6
+#define HSMC_NCS_WR_SETUP_OFFSET		8
+#define HSMC_NCS_WR_SETUP_SIZE			6
+#define HSMC_NRD_SETUP_OFFSET			16
+#define HSMC_NRD_SETUP_SIZE			6
+#define HSMC_NCS_RD_SETUP_OFFSET		24
+#define HSMC_NCS_RD_SETUP_SIZE			6
+
+/* Bitfields in PULSE0 */
+#define HSMC_NWE_PULSE_OFFSET			0
+#define HSMC_NWE_PULSE_SIZE			7
+#define HSMC_NCS_WR_PULSE_OFFSET		8
+#define HSMC_NCS_WR_PULSE_SIZE			7
+#define HSMC_NRD_PULSE_OFFSET			16
+#define HSMC_NRD_PULSE_SIZE			7
+#define HSMC_NCS_RD_PULSE_OFFSET		24
+#define HSMC_NCS_RD_PULSE_SIZE			7
+
+/* Bitfields in CYCLE0 */
+#define HSMC_NWE_CYCLE_OFFSET			0
+#define HSMC_NWE_CYCLE_SIZE			9
+#define HSMC_NRD_CYCLE_OFFSET			16
+#define HSMC_NRD_CYCLE_SIZE			9
+
+/* Bitfields in MODE0 */
+#define HSMC_READ_MODE_OFFSET			0
+#define HSMC_READ_MODE_SIZE			1
+#define HSMC_WRITE_MODE_OFFSET			1
+#define HSMC_WRITE_MODE_SIZE			1
+#define HSMC_EXNW_MODE_OFFSET			4
+#define HSMC_EXNW_MODE_SIZE			2
+#define HSMC_BAT_OFFSET				8
+#define HSMC_BAT_SIZE				1
+#define HSMC_DBW_OFFSET				12
+#define HSMC_DBW_SIZE				2
+#define HSMC_TDF_CYCLES_OFFSET			16
+#define HSMC_TDF_CYCLES_SIZE			4
+#define HSMC_TDF_MODE_OFFSET			20
+#define HSMC_TDF_MODE_SIZE			1
+#define HSMC_PMEN_OFFSET			24
+#define HSMC_PMEN_SIZE				1
+#define HSMC_PS_OFFSET				28
+#define HSMC_PS_SIZE				2
+
+/* Constants for READ_MODE */
+#define HSMC_READ_MODE_NCS_CONTROLLED		0
+#define HSMC_READ_MODE_NRD_CONTROLLED		1
+
+/* Constants for WRITE_MODE */
+#define HSMC_WRITE_MODE_NCS_CONTROLLED		0
+#define HSMC_WRITE_MODE_NWE_CONTROLLED		1
+
+/* Constants for EXNW_MODE */
+#define HSMC_EXNW_MODE_DISABLED			0
+#define HSMC_EXNW_MODE_RESERVED			1
+#define HSMC_EXNW_MODE_FROZEN			2
+#define HSMC_EXNW_MODE_READY			3
+
+/* Constants for BAT */
+#define HSMC_BAT_BYTE_SELECT			0
+#define HSMC_BAT_BYTE_WRITE			1
+
+/* Constants for DBW */
+#define HSMC_DBW_8_BITS				0
+#define HSMC_DBW_16_BITS			1
+#define HSMC_DBW_32_BITS			2
+
+/* Bit manipulation macros */
+#define HSMC_BIT(name)							\
+	(1 << HSMC_##name##_OFFSET)
+#define HSMC_BF(name,value)						\
+	(((value) & ((1 << HSMC_##name##_SIZE) - 1))			\
+	 << HSMC_##name##_OFFSET)
+#define HSMC_BFEXT(name,value)						\
+	(((value) >> HSMC_##name##_OFFSET)				\
+	 & ((1 << HSMC_##name##_SIZE) - 1))
+#define HSMC_BFINS(name,value,old)					\
+	(((old) & ~(((1 << HSMC_##name##_SIZE) - 1)			\
+		    << HSMC_##name##_OFFSET)) | HSMC_BF(name,value))
+
+/* Register access macros */
+#define hsmc_readl(port,reg)						\
+	readl((port)->regs + HSMC_##reg)
+#define hsmc_writel(port,reg,value)					\
+	writel((value), (port)->regs + HSMC_##reg)
+
+#endif /* __ASM_AVR32_HSMC_H__ */
diff --git a/include/asm-avr32/arch-at32ap/smc.h b/include/asm-avr32/arch-at32ap/smc.h
new file mode 100644
index 000000000000..3732b328303d
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/smc.h
@@ -0,0 +1,60 @@
+/*
+ * Static Memory Controller for AT32 chips
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Inspired by the OMAP2 General-Purpose Memory Controller interface
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ARCH_AT32AP_SMC_H
+#define __ARCH_AT32AP_SMC_H
+
+/*
+ * All timing parameters are in nanoseconds.
+ */
+struct smc_config {
+	/* Delay from address valid to assertion of given strobe */
+	u16		ncs_read_setup;
+	u16		nrd_setup;
+	u16		ncs_write_setup;
+	u16		nwe_setup;
+
+	/* Pulse length of given strobe */
+	u16		ncs_read_pulse;
+	u16		nrd_pulse;
+	u16		ncs_write_pulse;
+	u16		nwe_pulse;
+
+	/* Total cycle length of given operation */
+	u16		read_cycle;
+	u16		write_cycle;
+
+	/* Bus width in bytes */
+	u8		bus_width;
+
+	/*
+	 * 0: Data is sampled on rising edge of NCS
+	 * 1: Data is sampled on rising edge of NRD
+	 */
+	unsigned int	nrd_controlled:1;
+
+	/*
+	 * 0: Data is driven on falling edge of NCS
+	 * 1: Data is driven on falling edge of NWR
+	 */
+	unsigned int	nwe_controlled:1;
+
+	/*
+	 * 0: Byte select access type
+	 * 1: Byte write access type
+	 */
+	unsigned int	byte_write:1;
+};
+
+extern int smc_set_configuration(int cs, const struct smc_config *config);
+extern struct smc_config *smc_get_configuration(int cs);
+
+#endif /* __ARCH_AT32AP_SMC_H */
-- 
cgit v1.2.3


From 3a750363e6075a28e5542ce93a69c620c0cfd605 Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Mon, 25 Sep 2006 23:32:20 -0700
Subject: [PATCH] Use BUG_ON(foo) instead of "if (foo) BUG()" in
 include/asm-i386/dma-mapping.h

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/dma-mapping.h | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/dma-mapping.h b/include/asm-i386/dma-mapping.h
index 9cf20cacf76e..576ae01d71c8 100644
--- a/include/asm-i386/dma-mapping.h
+++ b/include/asm-i386/dma-mapping.h
@@ -21,8 +21,7 @@ static inline dma_addr_t
 dma_map_single(struct device *dev, void *ptr, size_t size,
 	       enum dma_data_direction direction)
 {
-	if (direction == DMA_NONE)
-		BUG();
+	BUG_ON(direction == DMA_NONE);
 	WARN_ON(size == 0);
 	flush_write_buffers();
 	return virt_to_phys(ptr);
@@ -32,8 +31,7 @@ static inline void
 dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
 		 enum dma_data_direction direction)
 {
-	if (direction == DMA_NONE)
-		BUG();
+	BUG_ON(direction == DMA_NONE);
 }
 
 static inline int
@@ -42,8 +40,7 @@ dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
 {
 	int i;
 
-	if (direction == DMA_NONE)
-		BUG();
+	BUG_ON(direction == DMA_NONE);
 	WARN_ON(nents == 0 || sg[0].length == 0);
 
 	for (i = 0; i < nents; i++ ) {
-- 
cgit v1.2.3


From 027a8c7e6067a1bcdef6775d1b1c08729dfbae51 Mon Sep 17 00:00:00 2001
From: Chris Wright <chrisw@sous-sol.org>
Date: Mon, 25 Sep 2006 23:32:23 -0700
Subject: [PATCH] x86: implement always-locked bit ops, for memory shared with
 an SMP hypervisor

Add "always lock'd" implementations of set_bit, clear_bit and change_bit and
the corresponding test_and_ functions.  Also add "always lock'd"
implementation of cmpxchg.  These give guaranteed strong synchronisation and
are required for non-SMP kernels running on an SMP hypervisor.

Signed-off-by: Ian Pratt <ian.pratt@xensource.com>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/sync_bitops.h | 156 +++++++++++++++++++++++++++++++++++++++++
 include/asm-i386/system.h      |  36 ++++++++++
 2 files changed, 192 insertions(+)
 create mode 100644 include/asm-i386/sync_bitops.h

(limited to 'include')

diff --git a/include/asm-i386/sync_bitops.h b/include/asm-i386/sync_bitops.h
new file mode 100644
index 000000000000..c94d51c993ee
--- /dev/null
+++ b/include/asm-i386/sync_bitops.h
@@ -0,0 +1,156 @@
+#ifndef _I386_SYNC_BITOPS_H
+#define _I386_SYNC_BITOPS_H
+
+/*
+ * Copyright 1992, Linus Torvalds.
+ */
+
+/*
+ * These have to be done with inline assembly: that way the bit-setting
+ * is guaranteed to be atomic. All bit operations return 0 if the bit
+ * was cleared before the operation and != 0 if it was not.
+ *
+ * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
+ */
+
+#define ADDR (*(volatile long *) addr)
+
+/**
+ * sync_set_bit - Atomically set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * This function is atomic and may not be reordered.  See __set_bit()
+ * if you do not require the atomic guarantees.
+ *
+ * Note: there are no guarantees that this function will not be reordered
+ * on non x86 architectures, so if you are writting portable code,
+ * make sure not to rely on its reordering guarantees.
+ *
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void sync_set_bit(int nr, volatile unsigned long * addr)
+{
+	__asm__ __volatile__("lock; btsl %1,%0"
+			     :"+m" (ADDR)
+			     :"Ir" (nr)
+			     : "memory");
+}
+
+/**
+ * sync_clear_bit - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * sync_clear_bit() is atomic and may not be reordered.  However, it does
+ * not contain a memory barrier, so if it is used for locking purposes,
+ * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
+ * in order to ensure changes are visible on other processors.
+ */
+static inline void sync_clear_bit(int nr, volatile unsigned long * addr)
+{
+	__asm__ __volatile__("lock; btrl %1,%0"
+			     :"+m" (ADDR)
+			     :"Ir" (nr)
+			     : "memory");
+}
+
+/**
+ * sync_change_bit - Toggle a bit in memory
+ * @nr: Bit to change
+ * @addr: Address to start counting from
+ *
+ * change_bit() is atomic and may not be reordered. It may be
+ * reordered on other architectures than x86.
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void sync_change_bit(int nr, volatile unsigned long * addr)
+{
+	__asm__ __volatile__("lock; btcl %1,%0"
+			     :"+m" (ADDR)
+			     :"Ir" (nr)
+			     : "memory");
+}
+
+/**
+ * sync_test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It may be reordered on other architectures than x86.
+ * It also implies a memory barrier.
+ */
+static inline int sync_test_and_set_bit(int nr, volatile unsigned long * addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("lock; btsl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit),"+m" (ADDR)
+			     :"Ir" (nr) : "memory");
+	return oldbit;
+}
+
+/**
+ * sync_test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It can be reorderdered on other architectures other than x86.
+ * It also implies a memory barrier.
+ */
+static inline int sync_test_and_clear_bit(int nr, volatile unsigned long * addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("lock; btrl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit),"+m" (ADDR)
+			     :"Ir" (nr) : "memory");
+	return oldbit;
+}
+
+/**
+ * sync_test_and_change_bit - Change a bit and return its old value
+ * @nr: Bit to change
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int sync_test_and_change_bit(int nr, volatile unsigned long* addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("lock; btcl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit),"+m" (ADDR)
+			     :"Ir" (nr) : "memory");
+	return oldbit;
+}
+
+static __always_inline int sync_const_test_bit(int nr, const volatile unsigned long *addr)
+{
+	return ((1UL << (nr & 31)) &
+		(((const volatile unsigned int *)addr)[nr >> 5])) != 0;
+}
+
+static inline int sync_var_test_bit(int nr, const volatile unsigned long * addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("btl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit)
+			     :"m" (ADDR),"Ir" (nr));
+	return oldbit;
+}
+
+#define sync_test_bit(nr,addr)			\
+	(__builtin_constant_p(nr) ?		\
+	 sync_constant_test_bit((nr),(addr)) :	\
+	 sync_var_test_bit((nr),(addr)))
+
+#undef ADDR
+
+#endif /* _I386_SYNC_BITOPS_H */
diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h
index 098bcee94e38..a6dabbcd6e6a 100644
--- a/include/asm-i386/system.h
+++ b/include/asm-i386/system.h
@@ -267,6 +267,9 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz
 #define cmpxchg(ptr,o,n)\
 	((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
 					(unsigned long)(n),sizeof(*(ptr))))
+#define sync_cmpxchg(ptr,o,n)\
+	((__typeof__(*(ptr)))__sync_cmpxchg((ptr),(unsigned long)(o),\
+					(unsigned long)(n),sizeof(*(ptr))))
 #endif
 
 static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
@@ -296,6 +299,39 @@ static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 	return old;
 }
 
+/*
+ * Always use locked operations when touching memory shared with a
+ * hypervisor, since the system may be SMP even if the guest kernel
+ * isn't.
+ */
+static inline unsigned long __sync_cmpxchg(volatile void *ptr,
+					    unsigned long old,
+					    unsigned long new, int size)
+{
+	unsigned long prev;
+	switch (size) {
+	case 1:
+		__asm__ __volatile__("lock; cmpxchgb %b1,%2"
+				     : "=a"(prev)
+				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+				     : "memory");
+		return prev;
+	case 2:
+		__asm__ __volatile__("lock; cmpxchgw %w1,%2"
+				     : "=a"(prev)
+				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+				     : "memory");
+		return prev;
+	case 4:
+		__asm__ __volatile__("lock; cmpxchgl %1,%2"
+				     : "=a"(prev)
+				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+				     : "memory");
+		return prev;
+	}
+	return old;
+}
+
 #ifndef CONFIG_X86_CMPXCHG
 /*
  * Building a kernel capable running on 80386. It may be necessary to
-- 
cgit v1.2.3


From 9f093394d75cd9c5df82c7a99c5eb5d7ce7ba199 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:24 -0700
Subject: [PATCH] x86: roll all the cpuid asm into one __cpuid call

It's a little neater, and also means only one place to patch for
paravirtualization.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/processor.h | 60 ++++++++++++++++++++------------------------
 1 file changed, 27 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index b32346d62e10..2277127696d2 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -143,6 +143,18 @@ static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #define X86_EFLAGS_VIP	0x00100000 /* Virtual Interrupt Pending */
 #define X86_EFLAGS_ID	0x00200000 /* CPUID detection flag */
 
+static inline void __cpuid(unsigned int *eax, unsigned int *ebx,
+			   unsigned int *ecx, unsigned int *edx)
+{
+	/* ecx is often an input as well as an output. */
+	__asm__("cpuid"
+		: "=a" (*eax),
+		  "=b" (*ebx),
+		  "=c" (*ecx),
+		  "=d" (*edx)
+		: "0" (*eax), "2" (*ecx));
+}
+
 /*
  * Generic CPUID function
  * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx
@@ -150,24 +162,18 @@ static inline void detect_ht(struct cpuinfo_x86 *c) {}
  */
 static inline void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx)
 {
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (op), "c"(0));
+	*eax = op;
+	*ecx = 0;
+	__cpuid(eax, ebx, ecx, edx);
 }
 
 /* Some CPUID calls want 'count' to be placed in ecx */
 static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
-	       	int *edx)
+			       int *edx)
 {
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (op), "c" (count));
+	*eax = op;
+	*ecx = count;
+	__cpuid(eax, ebx, ecx, edx);
 }
 
 /*
@@ -175,42 +181,30 @@ static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
  */
 static inline unsigned int cpuid_eax(unsigned int op)
 {
-	unsigned int eax;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax)
-		: "0" (op)
-		: "bx", "cx", "dx");
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return eax;
 }
 static inline unsigned int cpuid_ebx(unsigned int op)
 {
-	unsigned int eax, ebx;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax), "=b" (ebx)
-		: "0" (op)
-		: "cx", "dx" );
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return ebx;
 }
 static inline unsigned int cpuid_ecx(unsigned int op)
 {
-	unsigned int eax, ecx;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax), "=c" (ecx)
-		: "0" (op)
-		: "bx", "dx" );
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return ecx;
 }
 static inline unsigned int cpuid_edx(unsigned int op)
 {
-	unsigned int eax, edx;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax), "=d" (edx)
-		: "0" (op)
-		: "bx", "cx");
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return edx;
 }
 
-- 
cgit v1.2.3


From 052e79941a042e5be4feffa03b1fd60d93fb9e9a Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@xensource.com>
Date: Mon, 25 Sep 2006 23:32:25 -0700
Subject: [PATCH] x86: make __FIXADDR_TOP variable to allow it to make space
 for a hypervisor

Make __FIXADDR_TOP a variable, so that it can be set to not get in the way of
address space a hypervisor may want to reserve.

Original patch by Gerd Hoffmann <kraxel@suse.de>

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig         |  1 +
 arch/i386/mm/init.c       | 42 ++++++++++++++++++++++++++++++++++++++++++
 arch/i386/mm/pgtable.c    | 26 ++++++++++++++++++++++++++
 include/asm-i386/fixmap.h |  7 ++++++-
 4 files changed, 75 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index b2751eadbc56..1f83efb7a60e 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -794,6 +794,7 @@ config HOTPLUG_CPU
 config COMPAT_VDSO
 	bool "Compat VDSO support"
 	default y
+	depends on !PARAVIRT
 	help
 	  Map the VDSO to the predictable old-style address too.
 	---help---
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 353453d550a4..efd0bcdac65d 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -629,6 +629,48 @@ void __init mem_init(void)
 		(unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
 	       );
 
+#if 1 /* double-sanity-check paranoia */
+	printk("virtual kernel memory layout:\n"
+	       "    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+#ifdef CONFIG_HIGHMEM
+	       "    pkmap   : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+#endif
+	       "    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+	       "    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+	       "      .init : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+	       "      .data : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+	       "      .text : 0x%08lx - 0x%08lx   (%4ld kB)\n",
+	       FIXADDR_START, FIXADDR_TOP,
+	       (FIXADDR_TOP - FIXADDR_START) >> 10,
+
+#ifdef CONFIG_HIGHMEM
+	       PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE,
+	       (LAST_PKMAP*PAGE_SIZE) >> 10,
+#endif
+
+	       VMALLOC_START, VMALLOC_END,
+	       (VMALLOC_END - VMALLOC_START) >> 20,
+
+	       (unsigned long)__va(0), (unsigned long)high_memory,
+	       ((unsigned long)high_memory - (unsigned long)__va(0)) >> 20,
+
+	       (unsigned long)&__init_begin, (unsigned long)&__init_end,
+	       ((unsigned long)&__init_end - (unsigned long)&__init_begin) >> 10,
+
+	       (unsigned long)&_etext, (unsigned long)&_edata,
+	       ((unsigned long)&_edata - (unsigned long)&_etext) >> 10,
+
+	       (unsigned long)&_text, (unsigned long)&_etext,
+	       ((unsigned long)&_etext - (unsigned long)&_text) >> 10);
+
+#ifdef CONFIG_HIGHMEM
+	BUG_ON(PKMAP_BASE+LAST_PKMAP*PAGE_SIZE > FIXADDR_START);
+	BUG_ON(VMALLOC_END                     > PKMAP_BASE);
+#endif
+	BUG_ON(VMALLOC_START                   > VMALLOC_END);
+	BUG_ON((unsigned long)high_memory      > VMALLOC_START);
+#endif /* double-sanity-check paranoia */
+
 #ifdef CONFIG_X86_PAE
 	if (!cpu_has_pae)
 		panic("cannot execute a PAE-enabled kernel on a PAE-less CPU!");
diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index a9f4910a22f8..10126e3f8174 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -12,6 +12,7 @@
 #include <linux/slab.h>
 #include <linux/pagemap.h>
 #include <linux/spinlock.h>
+#include <linux/module.h>
 
 #include <asm/system.h>
 #include <asm/pgtable.h>
@@ -139,6 +140,12 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
 	__flush_tlb_one(vaddr);
 }
 
+static int fixmaps;
+#ifndef CONFIG_COMPAT_VDSO
+unsigned long __FIXADDR_TOP = 0xfffff000;
+EXPORT_SYMBOL(__FIXADDR_TOP);
+#endif
+
 void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags)
 {
 	unsigned long address = __fix_to_virt(idx);
@@ -148,6 +155,25 @@ void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags)
 		return;
 	}
 	set_pte_pfn(address, phys >> PAGE_SHIFT, flags);
+	fixmaps++;
+}
+
+/**
+ * reserve_top_address - reserves a hole in the top of kernel address space
+ * @reserve - size of hole to reserve
+ *
+ * Can be used to relocate the fixmap area and poke a hole in the top
+ * of kernel address space to make room for a hypervisor.
+ */
+void reserve_top_address(unsigned long reserve)
+{
+	BUG_ON(fixmaps > 0);
+#ifdef CONFIG_COMPAT_VDSO
+	BUG_ON(reserve != 0);
+#else
+	__FIXADDR_TOP = -reserve - PAGE_SIZE;
+	__VMALLOC_RESERVE += reserve;
+#endif
 }
 
 pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
diff --git a/include/asm-i386/fixmap.h b/include/asm-i386/fixmap.h
index a48cc3f7ccc6..02428cb36621 100644
--- a/include/asm-i386/fixmap.h
+++ b/include/asm-i386/fixmap.h
@@ -19,7 +19,11 @@
  * Leave one empty page between vmalloc'ed areas and
  * the start of the fixmap.
  */
-#define __FIXADDR_TOP	0xfffff000
+#ifndef CONFIG_COMPAT_VDSO
+extern unsigned long __FIXADDR_TOP;
+#else
+#define __FIXADDR_TOP  0xfffff000
+#endif
 
 #ifndef __ASSEMBLY__
 #include <linux/kernel.h>
@@ -93,6 +97,7 @@ enum fixed_addresses {
 
 extern void __set_fixmap (enum fixed_addresses idx,
 					unsigned long phys, pgprot_t flags);
+extern void reserve_top_address(unsigned long reserve);
 
 #define set_fixmap(idx, phys) \
 		__set_fixmap(idx, phys, PAGE_KERNEL)
-- 
cgit v1.2.3


From 9c9b8b388296ad5a306ab238dc677cfe6ff4cb12 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@xensource.com>
Date: Mon, 25 Sep 2006 23:32:26 -0700
Subject: [PATCH] x86: put .note.* sections into a PT_NOTE segment in vmlinux

This patch will pack any .note.* section into a PT_NOTE segment in the output
file.

To do this, we tell ld that we need a PT_NOTE segment.  This requires us to
start explicitly mapping sections to segments, so we also need to explicitly
create PT_LOAD segments for text and data, and map the sections to them
appropriately.  Fortunately, each section will default to its previous
section's segment, so it doesn't take many changes to vmlinux.lds.S.

This only changes i386 for now, but I presume the corresponding changes for
other architectures will be as simple.

This change also adds <linux/elfnote.h>, which defines C and Assembler macros
for actually creating ELF notes.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/vmlinux.lds.S    | 12 +++++-
 include/asm-generic/vmlinux.lds.h |  3 ++
 include/linux/elfnote.h           | 88 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 101 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/elfnote.h

(limited to 'include')

diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index 2d4f1386e2b1..1e7ac1c44ddc 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -13,6 +13,12 @@ OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
 OUTPUT_ARCH(i386)
 ENTRY(phys_startup_32)
 jiffies = jiffies_64;
+
+PHDRS {
+	text PT_LOAD FLAGS(5);	/* R_E */
+	data PT_LOAD FLAGS(7);	/* RWE */
+	note PT_NOTE FLAGS(4);	/* R__ */
+}
 SECTIONS
 {
   . = __KERNEL_START;
@@ -26,7 +32,7 @@ SECTIONS
 	KPROBES_TEXT
 	*(.fixup)
 	*(.gnu.warning)
-	} = 0x9090
+	} :text = 0x9090
 
   _etext = .;			/* End of text section */
 
@@ -48,7 +54,7 @@ SECTIONS
   .data : AT(ADDR(.data) - LOAD_OFFSET) {	/* Data */
 	*(.data)
 	CONSTRUCTORS
-	}
+	} :data
 
   . = ALIGN(4096);
   __nosave_begin = .;
@@ -184,4 +190,6 @@ SECTIONS
   STABS_DEBUG
 
   DWARF_DEBUG
+
+  NOTES
 }
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index db5a3732f106..253ae1328271 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -194,3 +194,6 @@
 		.stab.index 0 : { *(.stab.index) }			\
 		.stab.indexstr 0 : { *(.stab.indexstr) }		\
 		.comment 0 : { *(.comment) }
+
+#define NOTES								\
+		.notes : { *(.note.*) } :note
diff --git a/include/linux/elfnote.h b/include/linux/elfnote.h
new file mode 100644
index 000000000000..16f9f8ebffd9
--- /dev/null
+++ b/include/linux/elfnote.h
@@ -0,0 +1,88 @@
+#ifndef _LINUX_ELFNOTE_H
+#define _LINUX_ELFNOTE_H
+/*
+ * Helper macros to generate ELF Note structures, which are put into a
+ * PT_NOTE segment of the final vmlinux image.  These are useful for
+ * including name-value pairs of metadata into the kernel binary (or
+ * modules?) for use by external programs.
+ *
+ * Each note has three parts: a name, a type and a desc.  The name is
+ * intended to distinguish the note's originator, so it would be a
+ * company, project, subsystem, etc; it must be in a suitable form for
+ * use in a section name.  The type is an integer which is used to tag
+ * the data, and is considered to be within the "name" namespace (so
+ * "FooCo"'s type 42 is distinct from "BarProj"'s type 42).  The
+ * "desc" field is the actual data.  There are no constraints on the
+ * desc field's contents, though typically they're fairly small.
+ *
+ * All notes from a given NAME are put into a section named
+ * .note.NAME.  When the kernel image is finally linked, all the notes
+ * are packed into a single .notes section, which is mapped into the
+ * PT_NOTE segment.  Because notes for a given name are grouped into
+ * the same section, they'll all be adjacent the output file.
+ *
+ * This file defines macros for both C and assembler use.  Their
+ * syntax is slightly different, but they're semantically similar.
+ *
+ * See the ELF specification for more detail about ELF notes.
+ */
+
+#ifdef __ASSEMBLER__
+/*
+ * Generate a structure with the same shape as Elf{32,64}_Nhdr (which
+ * turn out to be the same size and shape), followed by the name and
+ * desc data with appropriate padding.  The 'desc' argument includes
+ * the assembler pseudo op defining the type of the data: .asciz
+ * "hello, world"
+ */
+.macro ELFNOTE name type desc:vararg
+.pushsection ".note.\name"
+  .align 4
+  .long 2f - 1f			/* namesz */
+  .long 4f - 3f			/* descsz */
+  .long \type
+1:.asciz "\name"
+2:.align 4
+3:\desc
+4:.align 4
+.popsection
+.endm
+#else	/* !__ASSEMBLER__ */
+#include <linux/elf.h>
+/*
+ * Use an anonymous structure which matches the shape of
+ * Elf{32,64}_Nhdr, but includes the name and desc data.  The size and
+ * type of name and desc depend on the macro arguments.  "name" must
+ * be a literal string, and "desc" must be passed by value.  You may
+ * only define one note per line, since __LINE__ is used to generate
+ * unique symbols.
+ */
+#define _ELFNOTE_PASTE(a,b)	a##b
+#define _ELFNOTE(size, name, unique, type, desc)			\
+	static const struct {						\
+		struct elf##size##_note _nhdr;				\
+		unsigned char _name[sizeof(name)]			\
+		__attribute__((aligned(sizeof(Elf##size##_Word))));	\
+		typeof(desc) _desc					\
+			     __attribute__((aligned(sizeof(Elf##size##_Word)))); \
+	} _ELFNOTE_PASTE(_note_, unique)				\
+		__attribute_used__					\
+		__attribute__((section(".note." name),			\
+			       aligned(sizeof(Elf##size##_Word)),	\
+			       unused)) = {				\
+		{							\
+			sizeof(name),					\
+			sizeof(desc),					\
+			type,						\
+		},							\
+		name,							\
+		desc							\
+	}
+#define ELFNOTE(size, name, type, desc)		\
+	_ELFNOTE(size, name, __LINE__, type, desc)
+
+#define ELFNOTE32(name, type, desc) ELFNOTE(32, name, type, desc)
+#define ELFNOTE64(name, type, desc) ELFNOTE(64, name, type, desc)
+#endif	/* __ASSEMBLER__ */
+
+#endif /* _LINUX_ELFNOTE_H */
-- 
cgit v1.2.3


From 5091e746848f74c9a2c0579b4ef8d8cd1a6b135d Mon Sep 17 00:00:00 2001
From: Ian Campbell <Ian.Campbell@xensource.com>
Date: Mon, 25 Sep 2006 23:32:28 -0700
Subject: [PATCH] Translate asm version of ELFNOTE macro into preprocessor
 macro

I've come across some problems with the assembly version of the ELFNOTE
macro currently in -mm. (in
x86-put-note-sections-into-a-pt_note-segment-in-vmlinux.patch)

The first is that older gas does not support :varargs in .macro
definitions (in my testing 2.17 does while 2.15 does not, I don't know
when it became supported). The Changes file says binutils >= 2.12 so I
think we need to avoid using it. There are no other uses in mainline or
-mm. Old gas appears to just ignore it so you get "too many arguments"
type errors.

Secondly it seems that passing strings as arguments to assembler macros
is broken without varargs. It looks like they get unquoted or each
character is treated as a separate argument or something and this causes
all manner of grief. I think this is because of the use of -traditional
when compiling assembly files.

Therefore I have translated the assembler macro into a pre-processor
macro.

I added the desctype as a separate argument instead of including it with
the descdata as the previous version did since -traditional means the
ELFNOTE definition after the #else needs to have the same number of
arguments (I think so anyway, the -traditional CPP semantics are pretty
fscking strange!).

With this patch I am able to define elfnotes in assembly like this with
both old and new assemblers.

	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS,       .asciz, "linux")
	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION,  .asciz, "2.6")
	ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION,    .asciz, "xen-3.0")
	ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE,      .long,  __PAGE_OFFSET)

Which seems reasonable enough.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/elfnote.h | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/elfnote.h b/include/linux/elfnote.h
index 16f9f8ebffd9..67396db141e8 100644
--- a/include/linux/elfnote.h
+++ b/include/linux/elfnote.h
@@ -31,22 +31,24 @@
 /*
  * Generate a structure with the same shape as Elf{32,64}_Nhdr (which
  * turn out to be the same size and shape), followed by the name and
- * desc data with appropriate padding.  The 'desc' argument includes
- * the assembler pseudo op defining the type of the data: .asciz
- * "hello, world"
+ * desc data with appropriate padding.  The 'desctype' argument is the
+ * assembler pseudo op defining the type of the data e.g. .asciz while
+ * 'descdata' is the data itself e.g.  "hello, world".
+ *
+ * e.g. ELFNOTE(XYZCo, 42, .asciz, "forty-two")
+ *      ELFNOTE(XYZCo, 12, .long, 0xdeadbeef)
  */
-.macro ELFNOTE name type desc:vararg
-.pushsection ".note.\name"
-  .align 4
-  .long 2f - 1f			/* namesz */
-  .long 4f - 3f			/* descsz */
-  .long \type
-1:.asciz "\name"
-2:.align 4
-3:\desc
-4:.align 4
-.popsection
-.endm
+#define ELFNOTE(name, type, desctype, descdata)	\
+.pushsection .note.name			;	\
+  .align 4				;	\
+  .long 2f - 1f		/* namesz */	;	\
+  .long 4f - 3f		/* descsz */	;	\
+  .long type				;	\
+1:.asciz "name"				;	\
+2:.align 4				;	\
+3:desctype descdata			;	\
+4:.align 4				;	\
+.popsection				;
 #else	/* !__ASSEMBLER__ */
 #include <linux/elf.h>
 /*
-- 
cgit v1.2.3


From 673eae8230a192f07b8715b872d6925521e9738d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:29 -0700
Subject: [PATCH] x86: trivial pgtable.h __ASSEMBLY__ move

Parsing generic pgtable.h in assembler is simply crazy.  None of this file is
needed in assembler code, and C inline functions and structures routine break
one or more different compiles.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/pgtable.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index c2059a3a0621..349260cd86ed 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_GENERIC_PGTABLE_H
 #define _ASM_GENERIC_PGTABLE_H
 
+#ifndef __ASSEMBLY__
+
 #ifndef __HAVE_ARCH_PTEP_ESTABLISH
 /*
  * Establish a new mapping:
@@ -188,7 +190,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addres
 })
 #endif
 
-#ifndef __ASSEMBLY__
 /*
  * When walking page tables, we usually want to skip any p?d_none entries;
  * and any p?d_bad entries - reporting the error before resetting to none.
-- 
cgit v1.2.3


From 6049742dbcecf170e903638a029f4dc280b9d53d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:30 -0700
Subject: [PATCH] x86: trivial move of __HAVE macros in i386 pagetable headers

Move the __HAVE_ARCH_PTEP defines to accompany the function definitions.
Anything else is just a complete nightmare to track through the 2/3-level
paging code, and this caused duplicate definitions to be needed (pte_same),
which could have easily been taken care of with the asm-generic pgtable
functions.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/pgtable-2level.h |  3 ++-
 include/asm-i386/pgtable-3level.h |  2 ++
 include/asm-i386/pgtable.h        | 10 ++++------
 3 files changed, 8 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/pgtable-2level.h b/include/asm-i386/pgtable-2level.h
index 2756d4b04c27..201c86a6711e 100644
--- a/include/asm-i386/pgtable-2level.h
+++ b/include/asm-i386/pgtable-2level.h
@@ -21,8 +21,9 @@
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 #define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
 #define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte_low, 0))
-#define pte_same(a, b)		((a).pte_low == (b).pte_low)
+
 #define pte_page(x)		pfn_to_page(pte_pfn(x))
 #define pte_none(x)		(!(x).pte_low)
 #define pte_pfn(x)		((unsigned long)(((x).pte_low >> PAGE_SHIFT)))
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index 807ed9e366de..0d899173232e 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -105,6 +105,7 @@ static inline void pmd_clear(pmd_t *pmd)
 	*(tmp + 1) = 0;
 }
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
 static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	pte_t res;
@@ -117,6 +118,7 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 	return res;
 }
 
+#define __HAVE_ARCH_PTE_SAME
 static inline int pte_same(pte_t a, pte_t b)
 {
 	return a.pte_low == b.pte_low && a.pte_high == b.pte_high;
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index c04b3d0f484e..38bf97514436 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -246,6 +246,7 @@ static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return p
 # include <asm/pgtable-2level.h>
 #endif
 
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
 static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
 {
 	if (!pte_dirty(*ptep))
@@ -253,6 +254,7 @@ static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned
 	return test_and_clear_bit(_PAGE_BIT_DIRTY, &ptep->pte_low);
 }
 
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
 static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
 {
 	if (!pte_young(*ptep))
@@ -260,6 +262,7 @@ static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned
 	return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte_low);
 }
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
 static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
 {
 	pte_t pte;
@@ -272,6 +275,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 	return pte;
 }
 
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
 static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	clear_bit(_PAGE_BIT_RW, &ptep->pte_low);
@@ -441,12 +445,6 @@ extern void noexec_setup(const char *str);
 #define GET_IOSPACE(pfn)		0
 #define GET_PFN(pfn)			(pfn)
 
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-#define __HAVE_ARCH_PTEP_SET_WRPROTECT
-#define __HAVE_ARCH_PTE_SAME
 #include <asm-generic/pgtable.h>
 
 #endif /* _I386_PGTABLE_H */
-- 
cgit v1.2.3


From 2965a0e6da0ccd8971ccf2c00a02bfa6e212acdb Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:31 -0700
Subject: [PATCH] x86: trivial move of ptep_set_access_flags

Move ptep_set_access_flags to be closer to the other ptep accessors, and make
the indentation standard.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/pgtable.h | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 38bf97514436..0dc051a8078b 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -246,6 +246,22 @@ static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return p
 # include <asm/pgtable-2level.h>
 #endif
 
+/*
+ * We only update the dirty/accessed state if we set
+ * the dirty bit by hand in the kernel, since the hardware
+ * will do the accessed bit for us, and we don't want to
+ * race with other CPU's that might be updating the dirty
+ * bit at the same time.
+ */
+#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+#define ptep_set_access_flags(vma, address, ptep, entry, dirty)		\
+do {									\
+	if (dirty) {							\
+		(ptep)->pte_low = (entry).pte_low;			\
+		flush_tlb_page(vma, address);				\
+	}								\
+} while (0)
+
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
 static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
 {
@@ -415,23 +431,8 @@ extern void noexec_setup(const char *str);
 /*
  * The i386 doesn't have any external MMU info: the kernel page
  * tables contain all the necessary information.
- *
- * Also, we only update the dirty/accessed state if we set
- * the dirty bit by hand in the kernel, since the hardware
- * will do the accessed bit for us, and we don't want to
- * race with other CPU's that might be updating the dirty
- * bit at the same time.
  */
 #define update_mmu_cache(vma,address,pte) do { } while (0)
-#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
-#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
-	do {								  \
-		if (__dirty) {						  \
-			(__ptep)->pte_low = (__entry).pte_low;	  	  \
-			flush_tlb_page(__vma, __address);		  \
-		}							  \
-	} while (0)
-
 #endif /* !__ASSEMBLY__ */
 
 #ifdef CONFIG_FLATMEM
-- 
cgit v1.2.3


From a3bc0dbc81d36fd38991b4373f6de8e1a507605a Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:33 -0700
Subject: [PATCH] smp_call_function_single() cleanup

If we're going to implement smp_call_function_single() on three architecture
with the same prototype then it should have a declaration in a
non-arch-specific header file.

Move it into <linux/smp.h>.

Cc: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smp.c              | 4 ++--
 arch/ia64/kernel/perfmon.c          | 1 +
 arch/ia64/sn/kernel/sn2/sn_hwperf.c | 3 ++-
 arch/x86_64/kernel/smpboot.c        | 3 ++-
 include/asm-ia64/smp.h              | 2 --
 include/asm-x86_64/smp.h            | 2 --
 include/linux/smp.h                 | 3 +++
 7 files changed, 10 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index 304243ed7a30..465188e2d701 100644
--- a/arch/i386/kernel/smp.c
+++ b/arch/i386/kernel/smp.c
@@ -683,8 +683,8 @@ __smp_call_function_single(int cpu, void (*func) (void *info), void *info,
  * or is or has executed.
  */
 
-int smp_call_function_single (int cpu, void (*func) (void *info), void *info,
-	int nonatomic, int wait)
+int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
+			int nonatomic, int wait)
 {
 	/* prevent preemption and reschedule on another processor */
 	int me = get_cpu();
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 84a7e52f56f6..7bb7696e4ce2 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -34,6 +34,7 @@
 #include <linux/file.h>
 #include <linux/poll.h>
 #include <linux/vfs.h>
+#include <linux/smp.h>
 #include <linux/pagemap.h>
 #include <linux/mount.h>
 #include <linux/bitops.h>
diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
index 9a8a29339d2d..b632b9c1e3b3 100644
--- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c
+++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
@@ -32,9 +32,10 @@
 #include <linux/cpumask.h>
 #include <linux/smp_lock.h>
 #include <linux/nodemask.h>
+#include <linux/smp.h>
+
 #include <asm/processor.h>
 #include <asm/topology.h>
-#include <asm/smp.h>
 #include <asm/semaphore.h>
 #include <asm/uaccess.h>
 #include <asm/sal.h>
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 975380207b46..3ae9ffddddc0 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -46,9 +46,10 @@
 #include <linux/bootmem.h>
 #include <linux/thread_info.h>
 #include <linux/module.h>
-
 #include <linux/delay.h>
 #include <linux/mc146818rtc.h>
+#include <linux/smp.h>
+
 #include <asm/mtrr.h>
 #include <asm/pgalloc.h>
 #include <asm/desc.h>
diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h
index 719ff309ce09..74bde1c2bb1a 100644
--- a/include/asm-ia64/smp.h
+++ b/include/asm-ia64/smp.h
@@ -122,8 +122,6 @@ extern void __init smp_build_cpu_map(void);
 extern void __init init_smp_config (void);
 extern void smp_do_timer (struct pt_regs *regs);
 
-extern int smp_call_function_single (int cpuid, void (*func) (void *info), void *info,
-				     int retry, int wait);
 extern void smp_send_reschedule (int cpu);
 extern void lock_ipi_calllock(void);
 extern void unlock_ipi_calllock(void);
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6805e1feb300..ce97f65e1d10 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -48,8 +48,6 @@ extern void unlock_ipi_call_lock(void);
 extern int smp_num_siblings;
 extern void smp_send_reschedule(int cpu);
 void smp_stop_cpu(void);
-extern int smp_call_function_single(int cpuid, void (*func) (void *info),
-				void *info, int retry, int wait);
 
 extern cpumask_t cpu_sibling_map[NR_CPUS];
 extern cpumask_t cpu_core_map[NR_CPUS];
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 837e8bce1349..51649987f691 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -53,6 +53,9 @@ extern void smp_cpus_done(unsigned int max_cpus);
  */
 int smp_call_function(void(*func)(void *info), void *info, int retry, int wait);
 
+int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
+				int retry, int wait);
+
 /*
  * Call a function on all processors
  */
-- 
cgit v1.2.3


From 930631edd4b1fe2781d9fe90edbe35d89dfc94cc Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Mon, 25 Sep 2006 23:32:40 -0700
Subject: [PATCH] add DIV_ROUND_UP()

Add the DIV_ROUND_UP() helper macro: divide `n' by `d', rounding up.

Stolen from the gfs2 tree(!) because the swsusp patches need it.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kernel.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b2ae4fdce8b..e44a37e2c71c 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -33,6 +33,7 @@ extern const char linux_banner[];
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
 #define ALIGN(x,a) (((x)+(a)-1UL)&~((a)-1UL))
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
 #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
 
 #define	KERN_EMERG	"<0>"	/* system is unusable			*/
-- 
cgit v1.2.3


From ab954160350c91c77ae03740ef90458c3ad5412c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:42 -0700
Subject: [PATCH] swsusp: write speedup

Switch the swsusp writeout code from 4k-at-a-time to 4MB-at-a-time.

Crufty old PIII testbox:
	12.9 MB/s -> 20.9 MB/s

Sony Vaio:
	14.7 MB/s -> 26.5 MB/s

The implementation is crude.  A better one would use larger BIOs, but wouldn't
gain any performance.

The memcpys will be mostly pipelined with the IO and basically come for free.

The ENOMEM path has not been tested.  It should be.

Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/swap.h |  5 ++-
 kernel/power/swap.c  | 93 ++++++++++++++++++++++++++++++++++++++++++----------
 mm/page_io.c         | 25 ++++++++++----
 3 files changed, 98 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index a2f5ad7c2d2e..3d434cbffe26 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -12,6 +12,8 @@
 
 struct notifier_block;
 
+struct bio;
+
 #define SWAP_FLAG_PREFER	0x8000	/* set if swap priority specified */
 #define SWAP_FLAG_PRIO_MASK	0x7fff
 #define SWAP_FLAG_PRIO_SHIFT	0
@@ -216,7 +218,8 @@ extern void swap_unplug_io_fn(struct backing_dev_info *, struct page *);
 /* linux/mm/page_io.c */
 extern int swap_readpage(struct file *, struct page *);
 extern int swap_writepage(struct page *page, struct writeback_control *wbc);
-extern int rw_swap_page_sync(int, swp_entry_t, struct page *);
+extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page,
+				struct bio **bio_chain);
 
 /* linux/mm/swap_state.c */
 extern struct address_space swapper_space;
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 79b66e734bdb..2dc883d361d5 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -49,18 +49,16 @@ static int mark_swapfiles(swp_entry_t start)
 {
 	int error;
 
-	rw_swap_page_sync(READ,
-			  swp_entry(root_swap, 0),
-			  virt_to_page((unsigned long)&swsusp_header));
+	rw_swap_page_sync(READ, swp_entry(root_swap, 0),
+			  virt_to_page((unsigned long)&swsusp_header), NULL);
 	if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
 	    !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
 		memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
 		memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
 		swsusp_header.image = start;
-		error = rw_swap_page_sync(WRITE,
-					  swp_entry(root_swap, 0),
-					  virt_to_page((unsigned long)
-						       &swsusp_header));
+		error = rw_swap_page_sync(WRITE, swp_entry(root_swap, 0),
+				virt_to_page((unsigned long)&swsusp_header),
+				NULL);
 	} else {
 		pr_debug("swsusp: Partition is not swap space.\n");
 		error = -ENODEV;
@@ -88,16 +86,37 @@ static int swsusp_swap_check(void) /* This is called before saving image */
  *	write_page - Write one page to given swap location.
  *	@buf:		Address we're writing.
  *	@offset:	Offset of the swap page we're writing to.
+ *	@bio_chain:	Link the next write BIO here
  */
 
-static int write_page(void *buf, unsigned long offset)
+static int write_page(void *buf, unsigned long offset, struct bio **bio_chain)
 {
 	swp_entry_t entry;
 	int error = -ENOSPC;
 
 	if (offset) {
+		struct page *page = virt_to_page(buf);
+
+		if (bio_chain) {
+			/*
+			 * Whether or not we successfully allocated a copy page,
+			 * we take a ref on the page here.  It gets undone in
+			 * wait_on_bio_chain().
+			 */
+			struct page *page_copy;
+			page_copy = alloc_page(GFP_ATOMIC);
+			if (page_copy == NULL) {
+				WARN_ON_ONCE(1);
+				bio_chain = NULL;	/* Go synchronous */
+				get_page(page);
+			} else {
+				memcpy(page_address(page_copy),
+					page_address(page), PAGE_SIZE);
+				page = page_copy;
+			}
+		}
 		entry = swp_entry(root_swap, offset);
-		error = rw_swap_page_sync(WRITE, entry, virt_to_page(buf));
+		error = rw_swap_page_sync(WRITE, entry, page, bio_chain);
 	}
 	return error;
 }
@@ -185,37 +204,68 @@ static int get_swap_writer(struct swap_map_handle *handle)
 	return 0;
 }
 
-static int swap_write_page(struct swap_map_handle *handle, void *buf)
+static int wait_on_bio_chain(struct bio **bio_chain)
 {
-	int error;
+	struct bio *bio;
+	struct bio *next_bio;
+	int ret = 0;
+
+	if (bio_chain == NULL)
+		return 0;
+
+	bio = *bio_chain;
+	while (bio) {
+		struct page *page;
+
+		next_bio = bio->bi_private;
+		page = bio->bi_io_vec[0].bv_page;
+		wait_on_page_locked(page);
+		if (!PageUptodate(page) || PageError(page))
+			ret = -EIO;
+		put_page(page);
+		bio_put(bio);
+		bio = next_bio;
+	}
+	*bio_chain = NULL;
+	return ret;
+}
+
+static int swap_write_page(struct swap_map_handle *handle, void *buf,
+				struct bio **bio_chain)
+{
+	int error = 0;
 	unsigned long offset;
 
 	if (!handle->cur)
 		return -EINVAL;
 	offset = alloc_swap_page(root_swap, handle->bitmap);
-	error = write_page(buf, offset);
+	error = write_page(buf, offset, bio_chain);
 	if (error)
 		return error;
 	handle->cur->entries[handle->k++] = offset;
 	if (handle->k >= MAP_PAGE_ENTRIES) {
+		error = wait_on_bio_chain(bio_chain);
+		if (error)
+			goto out;
 		offset = alloc_swap_page(root_swap, handle->bitmap);
 		if (!offset)
 			return -ENOSPC;
 		handle->cur->next_swap = offset;
-		error = write_page(handle->cur, handle->cur_swap);
+		error = write_page(handle->cur, handle->cur_swap, NULL);
 		if (error)
-			return error;
+			goto out;
 		memset(handle->cur, 0, PAGE_SIZE);
 		handle->cur_swap = offset;
 		handle->k = 0;
 	}
-	return 0;
+out:
+	return error;
 }
 
 static int flush_swap_writer(struct swap_map_handle *handle)
 {
 	if (handle->cur && handle->cur_swap)
-		return write_page(handle->cur, handle->cur_swap);
+		return write_page(handle->cur, handle->cur_swap, NULL);
 	else
 		return -EINVAL;
 }
@@ -232,6 +282,8 @@ static int save_image(struct swap_map_handle *handle,
 	int ret;
 	int error = 0;
 	int nr_pages;
+	int err2;
+	struct bio *bio;
 	struct timeval start;
 	struct timeval stop;
 
@@ -240,11 +292,13 @@ static int save_image(struct swap_map_handle *handle,
 	if (!m)
 		m = 1;
 	nr_pages = 0;
+	bio = NULL;
 	do_gettimeofday(&start);
 	do {
 		ret = snapshot_read_next(snapshot, PAGE_SIZE);
 		if (ret > 0) {
-			error = swap_write_page(handle, data_of(*snapshot));
+			error = swap_write_page(handle, data_of(*snapshot),
+						&bio);
 			if (error)
 				break;
 			if (!(nr_pages % m))
@@ -252,7 +306,10 @@ static int save_image(struct swap_map_handle *handle,
 			nr_pages++;
 		}
 	} while (ret > 0);
+	err2 = wait_on_bio_chain(&bio);
 	do_gettimeofday(&stop);
+	if (!error)
+		error = err2;
 	if (!error)
 		printk("\b\b\b\bdone\n");
 	show_speed(&start, &stop, nr_to_write, "Wrote");
@@ -307,7 +364,7 @@ int swsusp_write(void)
 	error = get_swap_writer(&handle);
 	if (!error) {
 		unsigned long start = handle.cur_swap;
-		error = swap_write_page(&handle, header);
+		error = swap_write_page(&handle, header, NULL);
 		if (!error)
 			error = save_image(&handle, &snapshot,
 					header->pages - 1);
diff --git a/mm/page_io.c b/mm/page_io.c
index d2f0a5783370..f46a9862b7ef 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -156,10 +156,12 @@ out:
  * We use end_swap_bio_read() even for writes, because it happens to do what
  * we want.
  */
-int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page)
+int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page,
+			struct bio **bio_chain)
 {
 	struct bio *bio;
 	int ret = 0;
+	int bio_rw;
 
 	lock_page(page);
 
@@ -170,11 +172,22 @@ int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page)
 		goto out;
 	}
 
-	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	wait_on_page_locked(page);
-
-	if (!PageUptodate(page) || PageError(page))
-		ret = -EIO;
+	bio_rw = rw;
+	if (!bio_chain)
+		bio_rw |= (1 << BIO_RW_SYNC);
+	if (bio_chain)
+		bio_get(bio);
+	submit_bio(bio_rw, bio);
+	if (bio_chain == NULL) {
+		wait_on_page_locked(page);
+
+		if (!PageUptodate(page) || PageError(page))
+			ret = -EIO;
+	}
+	if (bio_chain) {
+		bio->bi_private = *bio_chain;
+		*bio_chain = bio;
+	}
 out:
 	return ret;
 }
-- 
cgit v1.2.3


From 546e0d271941dd1ff6961e2a1f7eac75f1fc277e Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:44 -0700
Subject: [PATCH] swsusp: read speedup

Implement async reads for swsusp resuming.

Crufty old PIII testbox:
	15.7 MB/s -> 20.3 MB/s

Sony Vaio:
	14.6 MB/s -> 33.3 MB/s

I didn't implement the post-resume bio_set_pages_dirty().  I don't really
understand why resume needs to run set_page_dirty() against these pages.

It might be a worry that this code modifies PG_Uptodate, PG_Error and
PG_Locked against the image pages.  Can this possibly affect the resumed-into
kernel?  Hopefully not, if we're atomically restoring its mem_map?

Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Jens Axboe <axboe@suse.de>
Cc: Laurent Riffard <laurent.riffard@free.fr>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/swap.h    |   1 +
 kernel/power/power.h    |   1 +
 kernel/power/snapshot.c |  11 ++--
 kernel/power/swap.c     | 138 ++++++++++++++++++++++++------------------------
 mm/page_io.c            |   2 +-
 5 files changed, 81 insertions(+), 72 deletions(-)

(limited to 'include')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 3d434cbffe26..e7c36ba2a2db 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -220,6 +220,7 @@ extern int swap_readpage(struct file *, struct page *);
 extern int swap_writepage(struct page *page, struct writeback_control *wbc);
 extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page,
 				struct bio **bio_chain);
+extern int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err);
 
 /* linux/mm/swap_state.c */
 extern struct address_space swapper_space;
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 57a792982fb9..59ce712f082d 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -58,6 +58,7 @@ struct snapshot_handle {
 	struct pbe	*pbe, *last_pbe;
 	void		*buffer;
 	unsigned int	buf_offset;
+	int		sync_read;
 };
 
 #define data_of(handle)	((handle).buffer + (handle).buf_offset)
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 75d4886e648e..591301ae8b7d 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -314,7 +314,7 @@ static unsigned int unsafe_pages;
  *	and we count them using unsafe_pages
  */
 
-static inline void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
+static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 {
 	void *res;
 
@@ -828,13 +828,16 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 	}
 	if (!handle->offset)
 		handle->buffer = buffer;
+	handle->sync_read = 1;
 	if (handle->prev < handle->page) {
 		if (!handle->prev) {
-			error = load_header(handle, (struct swsusp_info *)buffer);
+			error = load_header(handle,
+					(struct swsusp_info *)buffer);
 			if (error)
 				return error;
 		} else if (handle->prev <= nr_meta_pages) {
-			handle->pbe = unpack_orig_addresses(buffer, handle->pbe);
+			handle->pbe = unpack_orig_addresses(buffer,
+							handle->pbe);
 			if (!handle->pbe) {
 				error = prepare_image(handle);
 				if (error)
@@ -842,10 +845,12 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 				handle->pbe = pagedir_nosave;
 				handle->last_pbe = NULL;
 				handle->buffer = get_buffer(handle);
+				handle->sync_read = 0;
 			}
 		} else {
 			handle->pbe = handle->pbe->next;
 			handle->buffer = get_buffer(handle);
+			handle->sync_read = 0;
 		}
 		handle->prev = handle->page;
 	}
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 9ab989572164..8309d20b2563 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -22,6 +22,7 @@
 #include <linux/device.h>
 #include <linux/buffer_head.h>
 #include <linux/bio.h>
+#include <linux/blkdev.h>
 #include <linux/swap.h>
 #include <linux/swapops.h>
 #include <linux/pm.h>
@@ -214,6 +215,8 @@ static int wait_on_bio_chain(struct bio **bio_chain)
 		return 0;
 
 	bio = *bio_chain;
+	if (bio == NULL)
+		return 0;
 	while (bio) {
 		struct page *page;
 
@@ -349,7 +352,8 @@ int swsusp_write(void)
 	int error;
 
 	if ((error = swsusp_swap_check())) {
-		printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
+		printk(KERN_ERR "swsusp: Cannot find swap device, try "
+				"swapon -a.\n");
 		return error;
 	}
 	memset(&snapshot, 0, sizeof(struct snapshot_handle));
@@ -381,27 +385,6 @@ int swsusp_write(void)
 	return error;
 }
 
-/*
- *	Using bio to read from swap.
- *	This code requires a bit more work than just using buffer heads
- *	but, it is the recommended way for 2.5/2.6.
- *	The following are to signal the beginning and end of I/O. Bios
- *	finish asynchronously, while we want them to happen synchronously.
- *	A simple atomic_t, and a wait loop take care of this problem.
- */
-
-static atomic_t io_done = ATOMIC_INIT(0);
-
-static int end_io(struct bio *bio, unsigned int num, int err)
-{
-	if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) {
-		printk(KERN_ERR "I/O error reading swsusp image.\n");
-		return -EIO;
-	}
-	atomic_set(&io_done, 0);
-	return 0;
-}
-
 static struct block_device *resume_bdev;
 
 /**
@@ -409,15 +392,15 @@ static struct block_device *resume_bdev;
  *	@rw:	READ or WRITE.
  *	@off	physical offset of page.
  *	@page:	page we're reading or writing.
+ *	@bio_chain: list of pending biod (for async reading)
  *
  *	Straight from the textbook - allocate and initialize the bio.
- *	If we're writing, make sure the page is marked as dirty.
- *	Then submit it and wait.
+ *	If we're reading, make sure the page is marked as dirty.
+ *	Then submit it and, if @bio_chain == NULL, wait.
  */
-
-static int submit(int rw, pgoff_t page_off, void *page)
+static int submit(int rw, pgoff_t page_off, struct page *page,
+			struct bio **bio_chain)
 {
-	int error = 0;
 	struct bio *bio;
 
 	bio = bio_alloc(GFP_ATOMIC, 1);
@@ -425,33 +408,40 @@ static int submit(int rw, pgoff_t page_off, void *page)
 		return -ENOMEM;
 	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
 	bio->bi_bdev = resume_bdev;
-	bio->bi_end_io = end_io;
+	bio->bi_end_io = end_swap_bio_read;
 
-	if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
-		printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
-		error = -EFAULT;
-		goto Done;
+	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
+		printk("swsusp: ERROR: adding page to bio at %ld\n", page_off);
+		bio_put(bio);
+		return -EFAULT;
 	}
 
-	atomic_set(&io_done, 1);
-	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	while (atomic_read(&io_done))
-		yield();
-	if (rw == READ)
-		bio_set_pages_dirty(bio);
- Done:
-	bio_put(bio);
-	return error;
+	lock_page(page);
+	bio_get(bio);
+
+	if (bio_chain == NULL) {
+		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+		wait_on_page_locked(page);
+		if (rw == READ)
+			bio_set_pages_dirty(bio);
+		bio_put(bio);
+	} else {
+		get_page(page);
+		bio->bi_private = *bio_chain;
+		*bio_chain = bio;
+		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+	}
+	return 0;
 }
 
-static int bio_read_page(pgoff_t page_off, void *page)
+static int bio_read_page(pgoff_t page_off, void *addr, struct bio **bio_chain)
 {
-	return submit(READ, page_off, page);
+	return submit(READ, page_off, virt_to_page(addr), bio_chain);
 }
 
-static int bio_write_page(pgoff_t page_off, void *page)
+static int bio_write_page(pgoff_t page_off, void *addr)
 {
-	return submit(WRITE, page_off, page);
+	return submit(WRITE, page_off, virt_to_page(addr), NULL);
 }
 
 /**
@@ -476,7 +466,7 @@ static int get_swap_reader(struct swap_map_handle *handle,
 	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
 	if (!handle->cur)
 		return -ENOMEM;
-	error = bio_read_page(swp_offset(start), handle->cur);
+	error = bio_read_page(swp_offset(start), handle->cur, NULL);
 	if (error) {
 		release_swap_reader(handle);
 		return error;
@@ -485,7 +475,8 @@ static int get_swap_reader(struct swap_map_handle *handle,
 	return 0;
 }
 
-static int swap_read_page(struct swap_map_handle *handle, void *buf)
+static int swap_read_page(struct swap_map_handle *handle, void *buf,
+				struct bio **bio_chain)
 {
 	unsigned long offset;
 	int error;
@@ -495,16 +486,17 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf)
 	offset = handle->cur->entries[handle->k];
 	if (!offset)
 		return -EFAULT;
-	error = bio_read_page(offset, buf);
+	error = bio_read_page(offset, buf, bio_chain);
 	if (error)
 		return error;
 	if (++handle->k >= MAP_PAGE_ENTRIES) {
+		error = wait_on_bio_chain(bio_chain);
 		handle->k = 0;
 		offset = handle->cur->next_swap;
 		if (!offset)
 			release_swap_reader(handle);
-		else
-			error = bio_read_page(offset, handle->cur);
+		else if (!error)
+			error = bio_read_page(offset, handle->cur, NULL);
 	}
 	return error;
 }
@@ -517,38 +509,48 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf)
 
 static int load_image(struct swap_map_handle *handle,
                       struct snapshot_handle *snapshot,
-                      unsigned int nr_pages)
+                      unsigned int nr_to_read)
 {
 	unsigned int m;
-	int ret;
 	int error = 0;
 	struct timeval start;
 	struct timeval stop;
+	struct bio *bio;
+	int err2;
+	unsigned nr_pages;
 
-	printk("Loading image data pages (%u pages) ...     ", nr_pages);
-	m = nr_pages / 100;
+	printk("Loading image data pages (%u pages) ...     ", nr_to_read);
+	m = nr_to_read / 100;
 	if (!m)
 		m = 1;
 	nr_pages = 0;
+	bio = NULL;
 	do_gettimeofday(&start);
-	do {
-		ret = snapshot_write_next(snapshot, PAGE_SIZE);
-		if (ret > 0) {
-			error = swap_read_page(handle, data_of(*snapshot));
-			if (error)
-				break;
-			if (!(nr_pages % m))
-				printk("\b\b\b\b%3d%%", nr_pages / m);
-			nr_pages++;
-		}
-	} while (ret > 0);
+	for ( ; ; ) {
+		error = snapshot_write_next(snapshot, PAGE_SIZE);
+		if (error <= 0)
+			break;
+		error = swap_read_page(handle, data_of(*snapshot), &bio);
+		if (error)
+			break;
+		if (snapshot->sync_read)
+			error = wait_on_bio_chain(&bio);
+		if (error)
+			break;
+		if (!(nr_pages % m))
+			printk("\b\b\b\b%3d%%", nr_pages / m);
+		nr_pages++;
+	}
+	err2 = wait_on_bio_chain(&bio);
 	do_gettimeofday(&stop);
+	if (!error)
+		error = err2;
 	if (!error) {
 		printk("\b\b\b\bdone\n");
 		if (!snapshot_image_loaded(snapshot))
 			error = -ENODATA;
 	}
-	show_speed(&start, &stop, nr_pages, "Read");
+	show_speed(&start, &stop, nr_to_read, "Read");
 	return error;
 }
 
@@ -571,7 +573,7 @@ int swsusp_read(void)
 	header = (struct swsusp_info *)data_of(snapshot);
 	error = get_swap_reader(&handle, swsusp_header.image);
 	if (!error)
-		error = swap_read_page(&handle, header);
+		error = swap_read_page(&handle, header, NULL);
 	if (!error)
 		error = load_image(&handle, &snapshot, header->pages - 1);
 	release_swap_reader(&handle);
@@ -597,7 +599,7 @@ int swsusp_check(void)
 	if (!IS_ERR(resume_bdev)) {
 		set_blocksize(resume_bdev, PAGE_SIZE);
 		memset(&swsusp_header, 0, sizeof(swsusp_header));
-		if ((error = bio_read_page(0, &swsusp_header)))
+		if ((error = bio_read_page(0, &swsusp_header, NULL)))
 			return error;
 		if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
 			memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
diff --git a/mm/page_io.c b/mm/page_io.c
index f46a9862b7ef..d4840ecbf8f9 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -74,7 +74,7 @@ static int end_swap_bio_write(struct bio *bio, unsigned int bytes_done, int err)
 	return 0;
 }
 
-static int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err)
+int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err)
 {
 	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	struct page *page = bio->bi_io_vec[0].bv_page;
-- 
cgit v1.2.3


From e8eff5ac294e12531c4195e0c15a222d3c9015e5 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:46 -0700
Subject: [PATCH] Make swsusp avoid memory holes and reserved memory regions on
 x86_64

On x86_64 machines with more than 2 GB of RAM there are large memory gaps
(with no corresponding kernel virtual addresses) and reserved memory
regions between areas of usable physical RAM.  Moreover, if CONFIG_FLATMEM
is set, they appear within the normal zone.  swsusp should not try to save
them, so the corresponding page structs have to be marked as 'nosave'.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Mel Gorman <mel@csn.ul.ie>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/e820.c  | 48 ++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/setup.c |  1 +
 include/asm-x86_64/e820.h  |  1 +
 3 files changed, 50 insertions(+)

(limited to 'include')

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index d6d7f731f6f0..708a3cd9a27e 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -16,6 +16,7 @@
 #include <linux/string.h>
 #include <linux/kexec.h>
 #include <linux/module.h>
+#include <linux/mm.h>
 
 #include <asm/pgtable.h>
 #include <asm/page.h>
@@ -297,6 +298,53 @@ void __init e820_reserve_resources(void)
 	}
 }
 
+/* Mark pages corresponding to given address range as nosave */
+static void __init
+e820_mark_nosave_range(unsigned long start, unsigned long end)
+{
+	unsigned long pfn, max_pfn;
+
+	if (start >= end)
+		return;
+
+	printk("Nosave address range: %016lx - %016lx\n", start, end);
+	max_pfn = end >> PAGE_SHIFT;
+	for (pfn = start >> PAGE_SHIFT; pfn < max_pfn; pfn++)
+		if (pfn_valid(pfn))
+			SetPageNosave(pfn_to_page(pfn));
+}
+
+/*
+ * Find the ranges of physical addresses that do not correspond to
+ * e820 RAM areas and mark the corresponding pages as nosave for software
+ * suspend and suspend to RAM.
+ *
+ * This function requires the e820 map to be sorted and without any
+ * overlapping entries and assumes the first e820 area to be RAM.
+ */
+void __init e820_mark_nosave_regions(void)
+{
+	int i;
+	unsigned long paddr;
+
+	paddr = round_down(e820.map[0].addr + e820.map[0].size, PAGE_SIZE);
+	for (i = 1; i < e820.nr_map; i++) {
+		struct e820entry *ei = &e820.map[i];
+
+		if (paddr < ei->addr)
+			e820_mark_nosave_range(paddr,
+					round_up(ei->addr, PAGE_SIZE));
+
+		paddr = round_down(ei->addr + ei->size, PAGE_SIZE);
+		if (ei->type != E820_RAM)
+			e820_mark_nosave_range(round_up(ei->addr, PAGE_SIZE),
+					paddr);
+
+		if (paddr >= (end_pfn << PAGE_SHIFT))
+			break;
+	}
+}
+
 /* 
  * Add a memory region to the kernel e820 map.
  */ 
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 34afad704824..4b39f0da17f3 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -689,6 +689,7 @@ void __init setup_arch(char **cmdline_p)
 	 */
 	probe_roms();
 	e820_reserve_resources(); 
+	e820_mark_nosave_regions();
 
 	request_resource(&iomem_resource, &video_ram_resource);
 
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index 670a3388e70a..f65674832318 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -46,6 +46,7 @@ extern void setup_memory_region(void);
 extern void contig_e820_setup(void); 
 extern unsigned long e820_end_of_ram(void);
 extern void e820_reserve_resources(void);
+extern void e820_mark_nosave_regions(void);
 extern void e820_print_map(char *who);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
-- 
cgit v1.2.3


From e3920fb42c8ddfe63befb54d95c0e13eabacea9b Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:48 -0700
Subject: [PATCH] Disable CPU hotplug during suspend

The current suspend code has to be run on one CPU, so we use the CPU
hotplug to take the non-boot CPUs offline on SMP machines.  However, we
should also make sure that these CPUs will not be enabled by someone else
after we have disabled them.

The functions disable_nonboot_cpus() and enable_nonboot_cpus() are moved to
kernel/cpu.c, because they now refer to some stuff in there that should
better be static.  Also it's better if disable_nonboot_cpus() returns an
error instead of panicking if something goes wrong, and
enable_nonboot_cpus() has no reason to panic(), because the CPUs may have
been enabled by the userland before it tries to take them online.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/cpu.h     |   8 +++
 include/linux/suspend.h |   8 ---
 kernel/cpu.c            | 138 +++++++++++++++++++++++++++++++++++++++++-------
 kernel/power/Makefile   |   2 -
 kernel/power/disk.c     |   7 ++-
 kernel/power/main.c     |  10 ++--
 kernel/power/smp.c      |  62 ----------------------
 kernel/power/user.c     |  14 +++--
 8 files changed, 145 insertions(+), 104 deletions(-)
 delete mode 100644 kernel/power/smp.c

(limited to 'include')

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 8fb344a9abd8..3fef7d67aedc 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -89,4 +89,12 @@ int cpu_down(unsigned int cpu);
 static inline int cpu_is_offline(int cpu) { return 0; }
 #endif
 
+#ifdef CONFIG_SUSPEND_SMP
+extern int disable_nonboot_cpus(void);
+extern void enable_nonboot_cpus(void);
+#else
+static inline int disable_nonboot_cpus(void) { return 0; }
+static inline void enable_nonboot_cpus(void) {}
+#endif
+
 #endif /* _LINUX_CPU_H_ */
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 96e31aa64cc7..6e8a06c950f4 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -57,14 +57,6 @@ static inline int software_suspend(void)
 }
 #endif /* CONFIG_PM */
 
-#ifdef CONFIG_SUSPEND_SMP
-extern void disable_nonboot_cpus(void);
-extern void enable_nonboot_cpus(void);
-#else
-static inline void disable_nonboot_cpus(void) {}
-static inline void enable_nonboot_cpus(void) {}
-#endif
-
 void save_processor_state(void);
 void restore_processor_state(void);
 struct saved_context;
diff --git a/kernel/cpu.c b/kernel/cpu.c
index f230f9ae01c2..32c96628463e 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -21,6 +21,11 @@ static DEFINE_MUTEX(cpu_bitmask_lock);
 
 static __cpuinitdata BLOCKING_NOTIFIER_HEAD(cpu_chain);
 
+/* If set, cpu_up and cpu_down will return -EBUSY and do nothing.
+ * Should always be manipulated under cpu_add_remove_lock
+ */
+static int cpu_hotplug_disabled;
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 /* Crappy recursive lock-takers in cpufreq! Complain loudly about idiots */
@@ -108,30 +113,25 @@ static int take_cpu_down(void *unused)
 	return 0;
 }
 
-int cpu_down(unsigned int cpu)
+/* Requires cpu_add_remove_lock to be held */
+static int _cpu_down(unsigned int cpu)
 {
 	int err;
 	struct task_struct *p;
 	cpumask_t old_allowed, tmp;
 
-	mutex_lock(&cpu_add_remove_lock);
-	if (num_online_cpus() == 1) {
-		err = -EBUSY;
-		goto out;
-	}
+	if (num_online_cpus() == 1)
+		return -EBUSY;
 
-	if (!cpu_online(cpu)) {
-		err = -EINVAL;
-		goto out;
-	}
+	if (!cpu_online(cpu))
+		return -EINVAL;
 
 	err = blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE,
 						(void *)(long)cpu);
 	if (err == NOTIFY_BAD) {
 		printk("%s: attempt to take down CPU %u failed\n",
 				__FUNCTION__, cpu);
-		err = -EINVAL;
-		goto out;
+		return -EINVAL;
 	}
 
 	/* Ensure that we are not runnable on dying cpu */
@@ -179,22 +179,32 @@ out_thread:
 	err = kthread_stop(p);
 out_allowed:
 	set_cpus_allowed(current, old_allowed);
-out:
+	return err;
+}
+
+int cpu_down(unsigned int cpu)
+{
+	int err = 0;
+
+	mutex_lock(&cpu_add_remove_lock);
+	if (cpu_hotplug_disabled)
+		err = -EBUSY;
+	else
+		err = _cpu_down(cpu);
+
 	mutex_unlock(&cpu_add_remove_lock);
 	return err;
 }
 #endif /*CONFIG_HOTPLUG_CPU*/
 
-int __devinit cpu_up(unsigned int cpu)
+/* Requires cpu_add_remove_lock to be held */
+static int __devinit _cpu_up(unsigned int cpu)
 {
 	int ret;
 	void *hcpu = (void *)(long)cpu;
 
-	mutex_lock(&cpu_add_remove_lock);
-	if (cpu_online(cpu) || !cpu_present(cpu)) {
-		ret = -EINVAL;
-		goto out;
-	}
+	if (cpu_online(cpu) || !cpu_present(cpu))
+		return -EINVAL;
 
 	ret = blocking_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu);
 	if (ret == NOTIFY_BAD) {
@@ -219,7 +229,95 @@ out_notify:
 	if (ret != 0)
 		blocking_notifier_call_chain(&cpu_chain,
 				CPU_UP_CANCELED, hcpu);
+
+	return ret;
+}
+
+int __devinit cpu_up(unsigned int cpu)
+{
+	int err = 0;
+
+	mutex_lock(&cpu_add_remove_lock);
+	if (cpu_hotplug_disabled)
+		err = -EBUSY;
+	else
+		err = _cpu_up(cpu);
+
+	mutex_unlock(&cpu_add_remove_lock);
+	return err;
+}
+
+#ifdef CONFIG_SUSPEND_SMP
+static cpumask_t frozen_cpus;
+
+int disable_nonboot_cpus(void)
+{
+	int cpu, first_cpu, error;
+
+	mutex_lock(&cpu_add_remove_lock);
+	first_cpu = first_cpu(cpu_present_map);
+	if (!cpu_online(first_cpu)) {
+		error = _cpu_up(first_cpu);
+		if (error) {
+			printk(KERN_ERR "Could not bring CPU%d up.\n",
+				first_cpu);
+			goto out;
+		}
+	}
+	error = set_cpus_allowed(current, cpumask_of_cpu(first_cpu));
+	if (error) {
+		printk(KERN_ERR "Could not run on CPU%d\n", first_cpu);
+		goto out;
+	}
+	/* We take down all of the non-boot CPUs in one shot to avoid races
+	 * with the userspace trying to use the CPU hotplug at the same time
+	 */
+	cpus_clear(frozen_cpus);
+	printk("Disabling non-boot CPUs ...\n");
+	for_each_online_cpu(cpu) {
+		if (cpu == first_cpu)
+			continue;
+		error = _cpu_down(cpu);
+		if (!error) {
+			cpu_set(cpu, frozen_cpus);
+			printk("CPU%d is down\n", cpu);
+		} else {
+			printk(KERN_ERR "Error taking CPU%d down: %d\n",
+				cpu, error);
+			break;
+		}
+	}
+	if (!error) {
+		BUG_ON(num_online_cpus() > 1);
+		/* Make sure the CPUs won't be enabled by someone else */
+		cpu_hotplug_disabled = 1;
+	} else {
+		printk(KERN_ERR "Non-boot CPUs are not disabled");
+	}
 out:
 	mutex_unlock(&cpu_add_remove_lock);
-	return ret;
+	return error;
+}
+
+void enable_nonboot_cpus(void)
+{
+	int cpu, error;
+
+	/* Allow everyone to use the CPU hotplug again */
+	mutex_lock(&cpu_add_remove_lock);
+	cpu_hotplug_disabled = 0;
+	mutex_unlock(&cpu_add_remove_lock);
+
+	printk("Enabling non-boot CPUs ...\n");
+	for_each_cpu_mask(cpu, frozen_cpus) {
+		error = cpu_up(cpu);
+		if (!error) {
+			printk("CPU%d is up\n", cpu);
+			continue;
+		}
+		printk(KERN_WARNING "Error taking CPU%d up: %d\n",
+			cpu, error);
+	}
+	cpus_clear(frozen_cpus);
 }
+#endif
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index 8d0af3d37a4b..38725f526afc 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -7,6 +7,4 @@ obj-y				:= main.o process.o console.o
 obj-$(CONFIG_PM_LEGACY)		+= pm.o
 obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o disk.o snapshot.o swap.o user.o
 
-obj-$(CONFIG_SUSPEND_SMP)	+= smp.o
-
 obj-$(CONFIG_MAGIC_SYSRQ)	+= poweroff.o
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index e13e74067845..7c7b9b65e365 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -18,6 +18,7 @@
 #include <linux/fs.h>
 #include <linux/mount.h>
 #include <linux/pm.h>
+#include <linux/cpu.h>
 
 #include "power.h"
 
@@ -72,7 +73,10 @@ static int prepare_processes(void)
 	int error;
 
 	pm_prepare_console();
-	disable_nonboot_cpus();
+
+	error = disable_nonboot_cpus();
+	if (error)
+		goto enable_cpus;
 
 	if (freeze_processes()) {
 		error = -EBUSY;
@@ -84,6 +88,7 @@ static int prepare_processes(void)
 		return 0;
 thaw:
 	thaw_processes();
+enable_cpus:
 	enable_nonboot_cpus();
 	pm_restore_console();
 	return error;
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6d295c776794..4d403323a7bb 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -16,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/pm.h>
 #include <linux/console.h>
+#include <linux/cpu.h>
 
 #include "power.h"
 
@@ -51,7 +52,7 @@ void pm_set_ops(struct pm_ops * ops)
 
 static int suspend_prepare(suspend_state_t state)
 {
-	int error = 0;
+	int error;
 	unsigned int free_pages;
 
 	if (!pm_ops || !pm_ops->enter)
@@ -59,12 +60,9 @@ static int suspend_prepare(suspend_state_t state)
 
 	pm_prepare_console();
 
-	disable_nonboot_cpus();
-
-	if (num_online_cpus() != 1) {
-		error = -EPERM;
+	error = disable_nonboot_cpus();
+	if (error)
 		goto Enable_cpu;
-	}
 
 	if (freeze_processes()) {
 		error = -EAGAIN;
diff --git a/kernel/power/smp.c b/kernel/power/smp.c
deleted file mode 100644
index 5957312b2d68..000000000000
--- a/kernel/power/smp.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * drivers/power/smp.c - Functions for stopping other CPUs.
- *
- * Copyright 2004 Pavel Machek <pavel@suse.cz>
- * Copyright (C) 2002-2003 Nigel Cunningham <ncunningham@clear.net.nz>
- *
- * This file is released under the GPLv2.
- */
-
-#undef DEBUG
-
-#include <linux/smp_lock.h>
-#include <linux/interrupt.h>
-#include <linux/suspend.h>
-#include <linux/module.h>
-#include <linux/cpu.h>
-#include <asm/atomic.h>
-#include <asm/tlbflush.h>
-
-/* This is protected by pm_sem semaphore */
-static cpumask_t frozen_cpus;
-
-void disable_nonboot_cpus(void)
-{
-	int cpu, error;
-
-	error = 0;
-	cpus_clear(frozen_cpus);
-	printk("Freezing cpus ...\n");
-	for_each_online_cpu(cpu) {
-		if (cpu == 0)
-			continue;
-		error = cpu_down(cpu);
-		if (!error) {
-			cpu_set(cpu, frozen_cpus);
-			printk("CPU%d is down\n", cpu);
-			continue;
-		}
-		printk("Error taking cpu %d down: %d\n", cpu, error);
-	}
-	BUG_ON(raw_smp_processor_id() != 0);
-	if (error)
-		panic("cpus not sleeping");
-}
-
-void enable_nonboot_cpus(void)
-{
-	int cpu, error;
-
-	printk("Thawing cpus ...\n");
-	for_each_cpu_mask(cpu, frozen_cpus) {
-		error = cpu_up(cpu);
-		if (!error) {
-			printk("CPU%d is up\n", cpu);
-			continue;
-		}
-		printk("Error taking cpu %d up: %d\n", cpu, error);
-		panic("Not enough cpus");
-	}
-	cpus_clear(frozen_cpus);
-}
-
diff --git a/kernel/power/user.c b/kernel/power/user.c
index 3f1539fbe48a..0ef5e4ba39e5 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -19,6 +19,7 @@
 #include <linux/swapops.h>
 #include <linux/pm.h>
 #include <linux/fs.h>
+#include <linux/cpu.h>
 
 #include <asm/uaccess.h>
 
@@ -139,12 +140,15 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		if (data->frozen)
 			break;
 		down(&pm_sem);
-		disable_nonboot_cpus();
-		if (freeze_processes()) {
-			thaw_processes();
-			enable_nonboot_cpus();
-			error = -EBUSY;
+		error = disable_nonboot_cpus();
+		if (!error) {
+			error = freeze_processes();
+			if (error) {
+				thaw_processes();
+				error = -EBUSY;
+			}
 		}
+		enable_nonboot_cpus();
 		up(&pm_sem);
 		if (!error)
 			data->frozen = 1;
-- 
cgit v1.2.3


From dcbb5a54f6e3984efa24772394f2225b11495c55 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:51 -0700
Subject: [PATCH] swsusp: clean up suspend header

Remove some things that are no longer used or defined elsewhere from suspend.h
and make the inline version of software_suspend() return the right error code.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/suspend.h | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 6e8a06c950f4..c11cacf1a13b 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -10,11 +10,11 @@
 #include <linux/pm.h>
 
 /* page backup entry */
-typedef struct pbe {
+struct pbe {
 	unsigned long address;		/* address of the copy */
 	unsigned long orig_address;	/* original address of page */
 	struct pbe *next;
-} suspend_pagedir_t;
+};
 
 #define for_each_pbe(pbe, pblist) \
 	for (pbe = pblist ; pbe ; pbe = pbe->next)
@@ -25,15 +25,6 @@ typedef struct pbe {
 #define for_each_pb_page(pbe, pblist) \
 	for (pbe = pblist ; pbe ; pbe = (pbe+PB_PAGE_SKIP)->next)
 
-
-#define SWAP_FILENAME_MAXLENGTH	32
-
-
-extern dev_t swsusp_resume_device;
-   
-/* mm/vmscan.c */
-extern int shrink_mem(void);
-
 /* mm/page_alloc.c */
 extern void drain_local_pages(void);
 extern void mark_free_pages(struct zone *zone);
@@ -53,7 +44,7 @@ static inline void pm_restore_console(void) {}
 static inline int software_suspend(void)
 {
 	printk("Warning: fake suspend called\n");
-	return -EPERM;
+	return -ENOSYS;
 }
 #endif /* CONFIG_PM */
 
-- 
cgit v1.2.3


From 940864ddabdb180e02041c4dcd46ba6f9eee732f Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:55 -0700
Subject: [PATCH] swsusp: Use memory bitmaps during resume

Make swsusp use memory bitmaps to store its internal information during the
resume phase of the suspend-resume cycle.

If the pfns of saveable pages are saved during the suspend phase instead of
the kernel virtual addresses of these pages, we can use them during the resume
phase directly to set the corresponding bits in a memory bitmap.  Then, this
bitmap is used to mark the page frames corresponding to the pages that were
saveable before the suspend (aka "unsafe" page frames).

Next, we allocate as many page frames as needed to store the entire suspend
image and make sure that there will be some extra free "safe" page frames for
the list of PBEs constructed later.  Subsequently, the image is loaded and, if
possible, the data loaded from it are written into their "original" page
frames (ie.  the ones they had occupied before the suspend).

The image data that cannot be written into their "original" page frames are
loaded into "safe" page frames and their "original" kernel virtual addresses,
as well as the addresses of the "safe" pages containing their copies, are
stored in a list of PBEs.  Finally, the list of PBEs is used to copy the
remaining image data into their "original" page frames (this is done
atomically, by the architecture-dependent parts of swsusp).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/suspend.h |   9 --
 kernel/power/power.h    |  11 +-
 kernel/power/snapshot.c | 416 +++++++++++++++++++++---------------------------
 kernel/power/swap.c     |   4 +-
 kernel/power/user.c     |   1 +
 5 files changed, 186 insertions(+), 255 deletions(-)

(limited to 'include')

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index c11cacf1a13b..b1237f16ecde 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -16,15 +16,6 @@ struct pbe {
 	struct pbe *next;
 };
 
-#define for_each_pbe(pbe, pblist) \
-	for (pbe = pblist ; pbe ; pbe = pbe->next)
-
-#define PBES_PER_PAGE      (PAGE_SIZE/sizeof(struct pbe))
-#define PB_PAGE_SKIP       (PBES_PER_PAGE-1)
-
-#define for_each_pb_page(pbe, pblist) \
-	for (pbe = pblist ; pbe ; pbe = (pbe+PB_PAGE_SKIP)->next)
-
 /* mm/page_alloc.c */
 extern void drain_local_pages(void);
 extern void mark_free_pages(struct zone *zone);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 6e9e2acc34f8..bfe999f7b272 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -81,16 +81,6 @@ struct snapshot_handle {
 	unsigned int	prev;	/* number of the block of PAGE_SIZE bytes that
 				 * was the current one previously
 				 */
-	struct pbe	*pbe;	/* PBE that corresponds to 'buffer' */
-	struct pbe	*last_pbe;	/* When the image is restored (eg. read
-					 * from disk) we can store some image
-					 * data directly in the page frames
-					 * in which they were before suspend.
-					 * In such a case the PBEs that
-					 * correspond to them will be unused.
-					 * This is the last PBE, so far, that
-					 * does not correspond to such data.
-					 */
 	void		*buffer;	/* address of the block to read from
 					 * or write to
 					 */
@@ -113,6 +103,7 @@ extern unsigned int snapshot_additional_pages(struct zone *zone);
 extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
 extern int snapshot_write_next(struct snapshot_handle *handle, size_t count);
 extern int snapshot_image_loaded(struct snapshot_handle *handle);
+extern void snapshot_free_unused_memory(struct snapshot_handle *handle);
 
 #define SNAPSHOT_IOC_MAGIC	'3'
 #define SNAPSHOT_FREEZE			_IO(SNAPSHOT_IOC_MAGIC, 1)
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 852e0df41719..1b84313cbab5 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -39,7 +39,7 @@ struct pbe *restore_pblist;
 
 static unsigned int nr_copy_pages;
 static unsigned int nr_meta_pages;
-static unsigned long *buffer;
+static void *buffer;
 
 #ifdef CONFIG_HIGHMEM
 unsigned int count_highmem_pages(void)
@@ -172,7 +172,7 @@ static inline int restore_highmem(void) {return 0;}
 #define PG_UNSAFE_CLEAR	1
 #define PG_UNSAFE_KEEP	0
 
-static unsigned int unsafe_pages;
+static unsigned int allocated_unsafe_pages;
 
 static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 {
@@ -183,7 +183,7 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 		while (res && PageNosaveFree(virt_to_page(res))) {
 			/* The page is unsafe, mark it for swsusp_free() */
 			SetPageNosave(virt_to_page(res));
-			unsafe_pages++;
+			allocated_unsafe_pages++;
 			res = (void *)get_zeroed_page(gfp_mask);
 		}
 	if (res) {
@@ -772,101 +772,10 @@ copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm)
 }
 
 /**
- *	free_pagedir - free pages allocated with alloc_pagedir()
- */
-
-static void free_pagedir(struct pbe *pblist, int clear_nosave_free)
-{
-	struct pbe *pbe;
-
-	while (pblist) {
-		pbe = (pblist + PB_PAGE_SKIP)->next;
-		free_image_page(pblist, clear_nosave_free);
-		pblist = pbe;
-	}
-}
-
-/**
- *	fill_pb_page - Create a list of PBEs on a given memory page
- */
-
-static inline void fill_pb_page(struct pbe *pbpage, unsigned int n)
-{
-	struct pbe *p;
-
-	p = pbpage;
-	pbpage += n - 1;
-	do
-		p->next = p + 1;
-	while (++p < pbpage);
-}
-
-/**
- *	create_pbe_list - Create a list of PBEs on top of a given chain
- *	of memory pages allocated with alloc_pagedir()
+ *	swsusp_free - free pages allocated for the suspend.
  *
- *	This function assumes that pages allocated by alloc_image_page() will
- *	always be zeroed.
- */
-
-static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages)
-{
-	struct pbe *pbpage;
-	unsigned int num = PBES_PER_PAGE;
-
-	for_each_pb_page (pbpage, pblist) {
-		if (num >= nr_pages)
-			break;
-
-		fill_pb_page(pbpage, PBES_PER_PAGE);
-		num += PBES_PER_PAGE;
-	}
-	if (pbpage) {
-		num -= PBES_PER_PAGE;
-		fill_pb_page(pbpage, nr_pages - num);
-	}
-}
-
-/**
- *	alloc_pagedir - Allocate the page directory.
- *
- *	First, determine exactly how many pages we need and
- *	allocate them.
- *
- *	We arrange the pages in a chain: each page is an array of PBES_PER_PAGE
- *	struct pbe elements (pbes) and the last element in the page points
- *	to the next page.
- *
- *	On each page we set up a list of struct_pbe elements.
- */
-
-static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask,
-				 int safe_needed)
-{
-	unsigned int num;
-	struct pbe *pblist, *pbe;
-
-	if (!nr_pages)
-		return NULL;
-
-	pblist = alloc_image_page(gfp_mask, safe_needed);
-	pbe = pblist;
-	for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) {
-		if (!pbe) {
-			free_pagedir(pblist, PG_UNSAFE_CLEAR);
-			return NULL;
-		}
-		pbe += PB_PAGE_SKIP;
-		pbe->next = alloc_image_page(gfp_mask, safe_needed);
-		pbe = pbe->next;
-	}
-	create_pbe_list(pblist, nr_pages);
-	return pblist;
-}
-
-/**
- * Free pages we allocated for suspend. Suspend pages are alocated
- * before atomic copy, so we need to free them after resume.
+ *	Suspend pages are alocated before the atomic copy is made, so we
+ *	need to release them after the resume.
  */
 
 void swsusp_free(void)
@@ -904,14 +813,18 @@ void swsusp_free(void)
 static int enough_free_mem(unsigned int nr_pages)
 {
 	struct zone *zone;
-	unsigned int n = 0;
+	unsigned int free = 0, meta = 0;
 
 	for_each_zone (zone)
-		if (!is_highmem(zone))
-			n += zone->free_pages;
-	pr_debug("swsusp: available memory: %u pages\n", n);
-	return n > (nr_pages + PAGES_FOR_IO +
-		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+		if (!is_highmem(zone)) {
+			free += zone->free_pages;
+			meta += snapshot_additional_pages(zone);
+		}
+
+	pr_debug("swsusp: pages needed: %u + %u + %u, available pages: %u\n",
+		nr_pages, PAGES_FOR_IO, meta, free);
+
+	return free > nr_pages + PAGES_FOR_IO + meta;
 }
 
 static int
@@ -961,11 +874,6 @@ asmlinkage int swsusp_save(void)
 	nr_pages = count_data_pages();
 	printk("swsusp: Need to copy %u pages\n", nr_pages);
 
-	pr_debug("swsusp: pages needed: %u + %lu + %u, free: %u\n",
-		 nr_pages,
-		 (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE,
-		 PAGES_FOR_IO, nr_free_pages());
-
 	if (!enough_free_mem(nr_pages)) {
 		printk(KERN_ERR "swsusp: Not enough free memory\n");
 		return -ENOMEM;
@@ -1007,22 +915,19 @@ static void init_header(struct swsusp_info *info)
 }
 
 /**
- *	pack_addresses - the addresses corresponding to pfns found in the
- *	bitmap @bm are stored in the array @buf[] (1 page)
+ *	pack_pfns - pfns corresponding to the set bits found in the bitmap @bm
+ *	are stored in the array @buf[] (1 page at a time)
  */
 
 static inline void
-pack_addresses(unsigned long *buf, struct memory_bitmap *bm)
+pack_pfns(unsigned long *buf, struct memory_bitmap *bm)
 {
 	int j;
 
 	for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
-		unsigned long pfn = memory_bm_next_pfn(bm);
-
-		if (unlikely(pfn == BM_END_OF_MAP))
+		buf[j] = memory_bm_next_pfn(bm);
+		if (unlikely(buf[j] == BM_END_OF_MAP))
 			break;
-
-		buf[j] = (unsigned long)page_address(pfn_to_page(pfn));
 	}
 }
 
@@ -1068,7 +973,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 	if (handle->prev < handle->cur) {
 		if (handle->cur <= nr_meta_pages) {
 			memset(buffer, 0, PAGE_SIZE);
-			pack_addresses(buffer, &orig_bm);
+			pack_pfns(buffer, &orig_bm);
 		} else {
 			unsigned long pfn = memory_bm_next_pfn(&copy_bm);
 
@@ -1094,14 +999,10 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
  *	had been used before suspend
  */
 
-static int mark_unsafe_pages(struct pbe *pblist)
+static int mark_unsafe_pages(struct memory_bitmap *bm)
 {
 	struct zone *zone;
 	unsigned long pfn, max_zone_pfn;
-	struct pbe *p;
-
-	if (!pblist) /* a sanity check */
-		return -EINVAL;
 
 	/* Clear page flags */
 	for_each_zone (zone) {
@@ -1111,30 +1012,37 @@ static int mark_unsafe_pages(struct pbe *pblist)
 				ClearPageNosaveFree(pfn_to_page(pfn));
 	}
 
-	/* Mark orig addresses */
-	for_each_pbe (p, pblist) {
-		if (virt_addr_valid(p->orig_address))
-			SetPageNosaveFree(virt_to_page(p->orig_address));
-		else
-			return -EFAULT;
-	}
+	/* Mark pages that correspond to the "original" pfns as "unsafe" */
+	memory_bm_position_reset(bm);
+	do {
+		pfn = memory_bm_next_pfn(bm);
+		if (likely(pfn != BM_END_OF_MAP)) {
+			if (likely(pfn_valid(pfn)))
+				SetPageNosaveFree(pfn_to_page(pfn));
+			else
+				return -EFAULT;
+		}
+	} while (pfn != BM_END_OF_MAP);
 
-	unsafe_pages = 0;
+	allocated_unsafe_pages = 0;
 
 	return 0;
 }
 
-static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
+static void
+duplicate_memory_bitmap(struct memory_bitmap *dst, struct memory_bitmap *src)
 {
-	/* We assume both lists contain the same number of elements */
-	while (src) {
-		dst->orig_address = src->orig_address;
-		dst = dst->next;
-		src = src->next;
+	unsigned long pfn;
+
+	memory_bm_position_reset(src);
+	pfn = memory_bm_next_pfn(src);
+	while (pfn != BM_END_OF_MAP) {
+		memory_bm_set_bit(dst, pfn);
+		pfn = memory_bm_next_pfn(src);
 	}
 }
 
-static int check_header(struct swsusp_info *info)
+static inline int check_header(struct swsusp_info *info)
 {
 	char *reason = NULL;
 
@@ -1161,19 +1069,14 @@ static int check_header(struct swsusp_info *info)
  *	load header - check the image header and copy data from it
  */
 
-static int load_header(struct snapshot_handle *handle,
-                              struct swsusp_info *info)
+static int
+load_header(struct swsusp_info *info)
 {
 	int error;
-	struct pbe *pblist;
 
+	restore_pblist = NULL;
 	error = check_header(info);
 	if (!error) {
-		pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, PG_ANY);
-		if (!pblist)
-			return -ENOMEM;
-		restore_pblist = pblist;
-		handle->pbe = pblist;
 		nr_copy_pages = info->image_pages;
 		nr_meta_pages = info->pages - info->image_pages - 1;
 	}
@@ -1181,108 +1084,137 @@ static int load_header(struct snapshot_handle *handle,
 }
 
 /**
- *	unpack_orig_addresses - copy the elements of @buf[] (1 page) to
- *	the PBEs in the list starting at @pbe
+ *	unpack_orig_pfns - for each element of @buf[] (1 page at a time) set
+ *	the corresponding bit in the memory bitmap @bm
  */
 
-static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
-                                                struct pbe *pbe)
+static inline void
+unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm)
 {
 	int j;
 
-	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-		pbe->orig_address = buf[j];
-		pbe = pbe->next;
+	for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
+		if (unlikely(buf[j] == BM_END_OF_MAP))
+			break;
+
+		memory_bm_set_bit(bm, buf[j]);
 	}
-	return pbe;
 }
 
 /**
- *	prepare_image - use metadata contained in the PBE list
- *	pointed to by restore_pblist to mark the pages that will
- *	be overwritten in the process of restoring the system
- *	memory state from the image ("unsafe" pages) and allocate
- *	memory for the image
+ *	prepare_image - use the memory bitmap @bm to mark the pages that will
+ *	be overwritten in the process of restoring the system memory state
+ *	from the suspend image ("unsafe" pages) and allocate memory for the
+ *	image.
  *
- *	The idea is to allocate the PBE list first and then
- *	allocate as many pages as it's needed for the image data,
- *	but not to assign these pages to the PBEs initially.
- *	Instead, we just mark them as allocated and create a list
- *	of "safe" which will be used later
+ *	The idea is to allocate a new memory bitmap first and then allocate
+ *	as many pages as needed for the image data, but not to assign these
+ *	pages to specific tasks initially.  Instead, we just mark them as
+ *	allocated and create a list of "safe" pages that will be used later.
  */
 
-static struct linked_page *safe_pages;
+#define PBES_PER_LINKED_PAGE	(LINKED_PAGE_DATA_SIZE / sizeof(struct pbe))
+
+static struct linked_page *safe_pages_list;
 
-static int prepare_image(struct snapshot_handle *handle)
+static int
+prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 {
-	int error = 0;
-	unsigned int nr_pages = nr_copy_pages;
-	struct pbe *p, *pblist = NULL;
+	unsigned int nr_pages;
+	struct linked_page *sp_list, *lp;
+	int error;
 
-	p = restore_pblist;
-	error = mark_unsafe_pages(p);
-	if (!error) {
-		pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, PG_SAFE);
-		if (pblist)
-			copy_page_backup_list(pblist, p);
-		free_pagedir(p, PG_UNSAFE_KEEP);
-		if (!pblist)
+	error = mark_unsafe_pages(bm);
+	if (error)
+		goto Free;
+
+	error = memory_bm_create(new_bm, GFP_ATOMIC, PG_SAFE);
+	if (error)
+		goto Free;
+
+	duplicate_memory_bitmap(new_bm, bm);
+	memory_bm_free(bm, PG_UNSAFE_KEEP);
+	/* Reserve some safe pages for potential later use.
+	 *
+	 * NOTE: This way we make sure there will be enough safe pages for the
+	 * chain_alloc() in get_buffer().  It is a bit wasteful, but
+	 * nr_copy_pages cannot be greater than 50% of the memory anyway.
+	 */
+	sp_list = NULL;
+	/* nr_copy_pages cannot be lesser than allocated_unsafe_pages */
+	nr_pages = nr_copy_pages - allocated_unsafe_pages;
+	nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE);
+	while (nr_pages > 0) {
+		lp = alloc_image_page(GFP_ATOMIC, PG_SAFE);
+		if (!lp) {
 			error = -ENOMEM;
+			goto Free;
+		}
+		lp->next = sp_list;
+		sp_list = lp;
+		nr_pages--;
 	}
-	safe_pages = NULL;
-	if (!error && nr_pages > unsafe_pages) {
-		nr_pages -= unsafe_pages;
-		while (nr_pages--) {
-			struct linked_page *ptr;
-
-			ptr = (void *)get_zeroed_page(GFP_ATOMIC);
-			if (!ptr) {
-				error = -ENOMEM;
-				break;
-			}
-			if (!PageNosaveFree(virt_to_page(ptr))) {
-				/* The page is "safe", add it to the list */
-				ptr->next = safe_pages;
-				safe_pages = ptr;
-			}
-			/* Mark the page as allocated */
-			SetPageNosave(virt_to_page(ptr));
-			SetPageNosaveFree(virt_to_page(ptr));
+	/* Preallocate memory for the image */
+	safe_pages_list = NULL;
+	nr_pages = nr_copy_pages - allocated_unsafe_pages;
+	while (nr_pages > 0) {
+		lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC);
+		if (!lp) {
+			error = -ENOMEM;
+			goto Free;
+		}
+		if (!PageNosaveFree(virt_to_page(lp))) {
+			/* The page is "safe", add it to the list */
+			lp->next = safe_pages_list;
+			safe_pages_list = lp;
 		}
+		/* Mark the page as allocated */
+		SetPageNosave(virt_to_page(lp));
+		SetPageNosaveFree(virt_to_page(lp));
+		nr_pages--;
 	}
-	if (!error) {
-		restore_pblist = pblist;
-	} else {
-		handle->pbe = NULL;
-		swsusp_free();
+	/* Free the reserved safe pages so that chain_alloc() can use them */
+	while (sp_list) {
+		lp = sp_list->next;
+		free_image_page(sp_list, PG_UNSAFE_CLEAR);
+		sp_list = lp;
 	}
+	return 0;
+
+Free:
+	swsusp_free();
 	return error;
 }
 
-static void *get_buffer(struct snapshot_handle *handle)
+/**
+ *	get_buffer - compute the address that snapshot_write_next() should
+ *	set for its caller to write to.
+ */
+
+static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca)
 {
-	struct pbe *pbe = handle->pbe, *last = handle->last_pbe;
-	struct page *page = virt_to_page(pbe->orig_address);
+	struct pbe *pbe;
+	struct page *page = pfn_to_page(memory_bm_next_pfn(bm));
 
-	if (PageNosave(page) && PageNosaveFree(page)) {
-		/*
-		 * We have allocated the "original" page frame and we can
-		 * use it directly to store the read page
+	if (PageNosave(page) && PageNosaveFree(page))
+		/* We have allocated the "original" page frame and we can
+		 * use it directly to store the loaded page.
 		 */
-		pbe->address = 0;
-		if (last && last->next)
-			last->next = NULL;
-		return (void *)pbe->orig_address;
-	}
-	/*
-	 * The "original" page frame has not been allocated and we have to
-	 * use a "safe" page frame to store the read page
+		return page_address(page);
+
+	/* The "original" page frame has not been allocated and we have to
+	 * use a "safe" page frame to store the loaded page.
 	 */
-	pbe->address = (unsigned long)safe_pages;
-	safe_pages = safe_pages->next;
-	if (last)
-		last->next = pbe;
-	handle->last_pbe = pbe;
+	pbe = chain_alloc(ca, sizeof(struct pbe));
+	if (!pbe) {
+		swsusp_free();
+		return NULL;
+	}
+	pbe->orig_address = (unsigned long)page_address(page);
+	pbe->address = (unsigned long)safe_pages_list;
+	safe_pages_list = safe_pages_list->next;
+	pbe->next = restore_pblist;
+	restore_pblist = pbe;
 	return (void *)pbe->address;
 }
 
@@ -1310,10 +1242,13 @@ static void *get_buffer(struct snapshot_handle *handle)
 
 int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 {
+	static struct chain_allocator ca;
 	int error = 0;
 
+	/* Check if we have already loaded the entire image */
 	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
 		return 0;
+
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
 		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
@@ -1324,26 +1259,32 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 		handle->buffer = buffer;
 	handle->sync_read = 1;
 	if (handle->prev < handle->cur) {
-		if (!handle->prev) {
-			error = load_header(handle,
-					(struct swsusp_info *)buffer);
+		if (handle->prev == 0) {
+			error = load_header(buffer);
+			if (error)
+				return error;
+
+			error = memory_bm_create(&copy_bm, GFP_ATOMIC, PG_ANY);
 			if (error)
 				return error;
+
 		} else if (handle->prev <= nr_meta_pages) {
-			handle->pbe = unpack_orig_addresses(buffer,
-							handle->pbe);
-			if (!handle->pbe) {
-				error = prepare_image(handle);
+			unpack_orig_pfns(buffer, &copy_bm);
+			if (handle->prev == nr_meta_pages) {
+				error = prepare_image(&orig_bm, &copy_bm);
 				if (error)
 					return error;
-				handle->pbe = restore_pblist;
-				handle->last_pbe = NULL;
-				handle->buffer = get_buffer(handle);
+
+				chain_init(&ca, GFP_ATOMIC, PG_SAFE);
+				memory_bm_position_reset(&orig_bm);
+				restore_pblist = NULL;
+				handle->buffer = get_buffer(&orig_bm, &ca);
 				handle->sync_read = 0;
+				if (!handle->buffer)
+					return -ENOMEM;
 			}
 		} else {
-			handle->pbe = handle->pbe->next;
-			handle->buffer = get_buffer(handle);
+			handle->buffer = get_buffer(&orig_bm, &ca);
 			handle->sync_read = 0;
 		}
 		handle->prev = handle->cur;
@@ -1362,6 +1303,13 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 
 int snapshot_image_loaded(struct snapshot_handle *handle)
 {
-	return !(!handle->pbe || handle->pbe->next || !nr_copy_pages ||
-		handle->cur <= nr_meta_pages + nr_copy_pages);
+	return !(!nr_copy_pages ||
+			handle->cur <= nr_meta_pages + nr_copy_pages);
+}
+
+void snapshot_free_unused_memory(struct snapshot_handle *handle)
+{
+	/* Free only if we have loaded the image entirely */
+	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
+		memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR);
 }
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 8309d20b2563..9b2ee5344dee 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -331,8 +331,7 @@ static int enough_swap(unsigned int nr_pages)
 	unsigned int free_swap = count_swap_pages(root_swap, 1);
 
 	pr_debug("swsusp: free swap pages: %u\n", free_swap);
-	return free_swap > (nr_pages + PAGES_FOR_IO +
-		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+	return free_swap > nr_pages + PAGES_FOR_IO;
 }
 
 /**
@@ -547,6 +546,7 @@ static int load_image(struct swap_map_handle *handle,
 		error = err2;
 	if (!error) {
 		printk("\b\b\b\bdone\n");
+		snapshot_free_unused_memory(snapshot);
 		if (!snapshot_image_loaded(snapshot))
 			error = -ENODATA;
 	}
diff --git a/kernel/power/user.c b/kernel/power/user.c
index 0ef5e4ba39e5..2e4499f3e4d9 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -193,6 +193,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			error = -EPERM;
 			break;
 		}
+		snapshot_free_unused_memory(&data->handle);
 		down(&pm_sem);
 		pm_prepare_console();
 		error = device_suspend(PMSG_FREEZE);
-- 
cgit v1.2.3


From c8eb8b4025175f967af0ba8e933f23aa9954dc35 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:56 -0700
Subject: [PATCH] PM: make it possible to disable console suspending

Change suspend_console() so that it waits for all consoles to flush the
remaining messages and make it possible to switch the console suspending off
with the help of a Kconfig option.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: Stefan Seyfried <seife@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/console.h |  5 +++++
 kernel/power/Kconfig    | 11 +++++++++++
 kernel/printk.c         |  3 +++
 3 files changed, 19 insertions(+)

(limited to 'include')

diff --git a/include/linux/console.h b/include/linux/console.h
index 3bdf2155e565..76a1807726eb 100644
--- a/include/linux/console.h
+++ b/include/linux/console.h
@@ -120,9 +120,14 @@ extern void console_stop(struct console *);
 extern void console_start(struct console *);
 extern int is_console_locked(void);
 
+#ifndef CONFIG_DISABLE_CONSOLE_SUSPEND
 /* Suspend and resume console messages over PM events */
 extern void suspend_console(void);
 extern void resume_console(void);
+#else
+static inline void suspend_console(void) {}
+static inline void resume_console(void) {}
+#endif /* CONFIG_DISABLE_CONSOLE_SUSPEND */
 
 /* Some debug stub to catch some of the obvious races in the VT code */
 #if 1
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 619ecabf7c58..4b6e2f18e056 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -36,6 +36,17 @@ config PM_DEBUG
 	code. This is helpful when debugging and reporting various PM bugs, 
 	like suspend support.
 
+config DISABLE_CONSOLE_SUSPEND
+	bool "Keep console(s) enabled during suspend/resume (DANGEROUS)"
+	depends on PM && PM_DEBUG
+	default n
+	---help---
+	This option turns off the console suspend mechanism that prevents
+	debug messages from reaching the console during the suspend/resume
+	operations.  This may be helpful when debugging device drivers'
+	suspend/resume routines, but may itself lead to problems, for example
+	if netconsole is used.
+
 config PM_TRACE
 	bool "Suspend/resume event tracing"
 	depends on PM && PM_DEBUG && X86_32 && EXPERIMENTAL
diff --git a/kernel/printk.c b/kernel/printk.c
index 1149365e989e..771f5e861bcd 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -721,6 +721,7 @@ int __init add_preferred_console(char *name, int idx, char *options)
 	return 0;
 }
 
+#ifndef CONFIG_DISABLE_CONSOLE_SUSPEND
 /**
  * suspend_console - suspend the console subsystem
  *
@@ -728,6 +729,7 @@ int __init add_preferred_console(char *name, int idx, char *options)
  */
 void suspend_console(void)
 {
+	printk("Suspending console(s)\n");
 	acquire_console_sem();
 	console_suspended = 1;
 }
@@ -737,6 +739,7 @@ void resume_console(void)
 	console_suspended = 0;
 	release_console_sem();
 }
+#endif /* CONFIG_DISABLE_CONSOLE_SUSPEND */
 
 /**
  * acquire_console_sem - lock the console system for exclusive use.
-- 
cgit v1.2.3


From c5c6ba4e08ab9c9e390a0f3a7d9a5c332f5cc6ef Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:58 -0700
Subject: [PATCH] PM: Add pm_trace switch

Add the pm_trace attribute in /sys/power which has to be explicitly set to
one to really enable the "PM tracing" code compiled in when CONFIG_PM_TRACE
is set (which modifies the machine's CMOS clock in unpredictable ways).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/power/interface.txt | 15 +++++++++++++++
 include/linux/resume-trace.h      | 24 ++++++++++++++----------
 kernel/power/main.c               | 30 ++++++++++++++++++++++++++++++
 3 files changed, 59 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/Documentation/power/interface.txt b/Documentation/power/interface.txt
index 4117802af0f8..a66bec222b16 100644
--- a/Documentation/power/interface.txt
+++ b/Documentation/power/interface.txt
@@ -52,3 +52,18 @@ suspend image will be as small as possible.
 
 Reading from this file will display the current image size limit, which
 is set to 500 MB by default.
+
+/sys/power/pm_trace controls the code which saves the last PM event point in
+the RTC across reboots, so that you can debug a machine that just hangs
+during suspend (or more commonly, during resume).  Namely, the RTC is only
+used to save the last PM event point if this file contains '1'.  Initially it
+contains '0' which may be changed to '1' by writing a string representing a
+nonzero integer into it.
+
+To use this debugging feature you should attempt to suspend the machine, then
+reboot it and run
+
+	dmesg -s 1000000 | grep 'hash matches'
+
+CAUTION: Using it will cause your machine's real-time (CMOS) clock to be
+set to a random invalid time after a resume.
diff --git a/include/linux/resume-trace.h b/include/linux/resume-trace.h
index a376bd4ade39..81e9299ca148 100644
--- a/include/linux/resume-trace.h
+++ b/include/linux/resume-trace.h
@@ -3,21 +3,25 @@
 
 #ifdef CONFIG_PM_TRACE
 
+extern int pm_trace_enabled;
+
 struct device;
 extern void set_trace_device(struct device *);
 extern void generate_resume_trace(void *tracedata, unsigned int user);
 
 #define TRACE_DEVICE(dev) set_trace_device(dev)
-#define TRACE_RESUME(user) do {				\
-	void *tracedata;				\
-	asm volatile("movl $1f,%0\n"			\
-		".section .tracedata,\"a\"\n"		\
-		"1:\t.word %c1\n"			\
-		"\t.long %c2\n"				\
-		".previous"				\
-		:"=r" (tracedata)			\
-		: "i" (__LINE__), "i" (__FILE__));	\
-	generate_resume_trace(tracedata, user);		\
+#define TRACE_RESUME(user) do {					\
+	if (pm_trace_enabled) {					\
+		void *tracedata;				\
+		asm volatile("movl $1f,%0\n"			\
+			".section .tracedata,\"a\"\n"		\
+			"1:\t.word %c1\n"			\
+			"\t.long %c2\n"				\
+			".previous"				\
+			:"=r" (tracedata)			\
+			: "i" (__LINE__), "i" (__FILE__));	\
+		generate_resume_trace(tracedata, user);		\
+	}							\
 } while (0)
 
 #else
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 4d403323a7bb..873228c71dab 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -17,6 +17,7 @@
 #include <linux/pm.h>
 #include <linux/console.h>
 #include <linux/cpu.h>
+#include <linux/resume-trace.h>
 
 #include "power.h"
 
@@ -281,10 +282,39 @@ static ssize_t state_store(struct subsystem * subsys, const char * buf, size_t n
 
 power_attr(state);
 
+#ifdef CONFIG_PM_TRACE
+int pm_trace_enabled;
+
+static ssize_t pm_trace_show(struct subsystem * subsys, char * buf)
+{
+	return sprintf(buf, "%d\n", pm_trace_enabled);
+}
+
+static ssize_t
+pm_trace_store(struct subsystem * subsys, const char * buf, size_t n)
+{
+	int val;
+
+	if (sscanf(buf, "%d", &val) == 1) {
+		pm_trace_enabled = !!val;
+		return n;
+	}
+	return -EINVAL;
+}
+
+power_attr(pm_trace);
+
+static struct attribute * g[] = {
+	&state_attr.attr,
+	&pm_trace_attr.attr,
+	NULL,
+};
+#else
 static struct attribute * g[] = {
 	&state_attr.attr,
 	NULL,
 };
+#endif /* CONFIG_PM_TRACE */
 
 static struct attribute_group attr_group = {
 	.attrs = g,
-- 
cgit v1.2.3


From 75e29b18d9a46bf3193278e92dc95609a8cca2ab Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:08 -0700
Subject: [PATCH] uml: stack usage reduction

The KSTK_* macros used an inordinate amount of stack.  In order to overcome
an impedance mismatch between their interface, which just returns a single
register value, and the interface of get_thread_regs, which took a full
pt_regs, the implementation created an on-stack pt_regs, filled it in, and
returned one field.  do_task_stat calls KSTK_* twice, resulting in two
local pt_regs, blowing out the stack.

This patch changes the interface (and name) of get_thread_regs to just
return a single register from a jmp_buf.

The include of archsetjmp.h" in registers.h to get the definition of
jmp_buf exposed a bogus include of <setjmp.h> in start_up.c.  <setjmp.h>
shouldn't be used anywhere any more since UML uses the klibc
setjmp/longjmp.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/registers.h             |  3 ++-
 arch/um/os-Linux/start_up.c             |  1 -
 arch/um/os-Linux/sys-i386/registers.c   | 15 +++++++++------
 arch/um/os-Linux/sys-x86_64/registers.c | 15 +++++++++------
 include/asm-um/processor-generic.h      |  4 +---
 5 files changed, 21 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/arch/um/include/registers.h b/arch/um/include/registers.h
index 83b688ca198f..f845b3629a6d 100644
--- a/arch/um/include/registers.h
+++ b/arch/um/include/registers.h
@@ -7,6 +7,7 @@
 #define __REGISTERS_H
 
 #include "sysdep/ptrace.h"
+#include "sysdep/archsetjmp.h"
 
 extern void init_thread_registers(union uml_pt_regs *to);
 extern int save_fp_registers(int pid, unsigned long *fp_regs);
@@ -15,6 +16,6 @@ extern void save_registers(int pid, union uml_pt_regs *regs);
 extern void restore_registers(int pid, union uml_pt_regs *regs);
 extern void init_registers(int pid);
 extern void get_safe_registers(unsigned long * regs, unsigned long * fp_regs);
-extern void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer);
+extern unsigned long get_thread_reg(int reg, jmp_buf *buf);
 
 #endif
diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c
index 503148504009..7fe92680c7dd 100644
--- a/arch/um/os-Linux/start_up.c
+++ b/arch/um/os-Linux/start_up.c
@@ -14,7 +14,6 @@
 #include <sched.h>
 #include <fcntl.h>
 #include <errno.h>
-#include <setjmp.h>
 #include <sys/time.h>
 #include <sys/wait.h>
 #include <sys/mman.h>
diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c
index 1f90a2d71386..7cd0369e02b3 100644
--- a/arch/um/os-Linux/sys-i386/registers.c
+++ b/arch/um/os-Linux/sys-i386/registers.c
@@ -130,11 +130,14 @@ void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
 		       HOST_FP_SIZE * sizeof(unsigned long));
 }
 
-void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer)
+unsigned long get_thread_reg(int reg, jmp_buf *buf)
 {
-	struct __jmp_buf *jmpbuf = buffer;
-
-	UPT_SET(uml_regs, EIP, jmpbuf->__eip);
-	UPT_SET(uml_regs, UESP, jmpbuf->__esp);
-	UPT_SET(uml_regs, EBP, jmpbuf->__ebp);
+	switch(reg){
+	case EIP: return buf[0]->__eip;
+	case UESP: return buf[0]->__esp;
+	case EBP: return buf[0]->__ebp;
+	default:
+		printk("get_thread_regs - unknown register %d\n", reg);
+		return 0;
+	}
 }
diff --git a/arch/um/os-Linux/sys-x86_64/registers.c b/arch/um/os-Linux/sys-x86_64/registers.c
index e730447d6c02..cb8e8a263280 100644
--- a/arch/um/os-Linux/sys-x86_64/registers.c
+++ b/arch/um/os-Linux/sys-x86_64/registers.c
@@ -78,11 +78,14 @@ void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
 		       HOST_FP_SIZE * sizeof(unsigned long));
 }
 
-void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer)
+unsigned long get_thread_reg(int reg, jmp_buf *buf)
 {
-	struct __jmp_buf *jmpbuf = buffer;
-
-	UPT_SET(uml_regs, RIP, jmpbuf->__rip);
-	UPT_SET(uml_regs, RSP, jmpbuf->__rsp);
-	UPT_SET(uml_regs, RBP, jmpbuf->__rbp);
+	switch(reg){
+	case RIP: return buf[0]->__rip;
+	case RSP: return buf[0]->__rsp;
+	case RBP: return buf[0]->__rbp;
+	default:
+		printk("get_thread_regs - unknown register %d\n", reg);
+		return 0;
+	}
 }
diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h
index 824c28896382..afa4fe1ca9f1 100644
--- a/include/asm-um/processor-generic.h
+++ b/include/asm-um/processor-generic.h
@@ -138,9 +138,7 @@ extern struct cpuinfo_um cpu_data[];
 
 #ifdef CONFIG_MODE_SKAS
 #define KSTK_REG(tsk, reg) \
-	({ union uml_pt_regs regs; \
-	   get_thread_regs(&regs, tsk->thread.mode.skas.switch_buf); \
-	   UPT_REG(&regs, reg); })
+	get_thread_reg(reg, tsk->thread.mode.skas.switch_buf)
 #else
 #define KSTK_REG(tsk, reg) (0xbadbabe)
 #endif
-- 
cgit v1.2.3


From 70e0eb8ef143f3729065c504177413ffe165af22 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:09 -0700
Subject: [PATCH] Split i386 and x86_64 ptrace.h

The use of SEGMENT_RPL_MASK in the i386 ptrace.h introduced by
x86-allow-a-kernel-to-not-be-in-ring-0.patch broke the UML build, as UML
includes the underlying architecture's ptrace.h, but has no easy access to the
x86 segment definitions.

Rather than kludging around this, as in the past, this patch splits the
userspace-usable parts, which are the bits that UML needs, of ptrace.h into
ptrace-abi.h, which is included back into ptrace.h.  Thus, there is no net
effect on i386.

As a side-effect, this creates a ptrace header which is close to being usable
in /usr/include.

x86_64 is also treated in this way for consistency.  There was some trailing
whitespace there, which is cleaned up.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/Kbuild         |  1 +
 include/asm-i386/ptrace-abi.h   | 39 +++++++++++++++++++++++++++
 include/asm-i386/ptrace.h       | 35 +-----------------------
 include/asm-x86_64/Kbuild       |  1 +
 include/asm-x86_64/ptrace-abi.h | 51 +++++++++++++++++++++++++++++++++++
 include/asm-x86_64/ptrace.h     | 59 +++++------------------------------------
 6 files changed, 100 insertions(+), 86 deletions(-)
 create mode 100644 include/asm-i386/ptrace-abi.h
 create mode 100644 include/asm-x86_64/ptrace-abi.h

(limited to 'include')

diff --git a/include/asm-i386/Kbuild b/include/asm-i386/Kbuild
index b75a348d0c1c..147e4ac1ebf0 100644
--- a/include/asm-i386/Kbuild
+++ b/include/asm-i386/Kbuild
@@ -3,6 +3,7 @@ include include/asm-generic/Kbuild.asm
 header-y += boot.h
 header-y += debugreg.h
 header-y += ldt.h
+header-y += ptrace-abi.h
 header-y += ucontext.h
 
 unifdef-y += mtrr.h
diff --git a/include/asm-i386/ptrace-abi.h b/include/asm-i386/ptrace-abi.h
new file mode 100644
index 000000000000..a44901817a26
--- /dev/null
+++ b/include/asm-i386/ptrace-abi.h
@@ -0,0 +1,39 @@
+#ifndef I386_PTRACE_ABI_H
+#define I386_PTRACE_ABI_H
+
+#define EBX 0
+#define ECX 1
+#define EDX 2
+#define ESI 3
+#define EDI 4
+#define EBP 5
+#define EAX 6
+#define DS 7
+#define ES 8
+#define FS 9
+#define GS 10
+#define ORIG_EAX 11
+#define EIP 12
+#define CS  13
+#define EFL 14
+#define UESP 15
+#define SS   16
+#define FRAME_SIZE 17
+
+/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
+#define PTRACE_GETREGS            12
+#define PTRACE_SETREGS            13
+#define PTRACE_GETFPREGS          14
+#define PTRACE_SETFPREGS          15
+#define PTRACE_GETFPXREGS         18
+#define PTRACE_SETFPXREGS         19
+
+#define PTRACE_OLDSETOPTIONS         21
+
+#define PTRACE_GET_THREAD_AREA    25
+#define PTRACE_SET_THREAD_AREA    26
+
+#define PTRACE_SYSEMU		  31
+#define PTRACE_SYSEMU_SINGLESTEP  32
+
+#endif
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index f324c53b6f9a..1910880fcd40 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -1,24 +1,7 @@
 #ifndef _I386_PTRACE_H
 #define _I386_PTRACE_H
 
-#define EBX 0
-#define ECX 1
-#define EDX 2
-#define ESI 3
-#define EDI 4
-#define EBP 5
-#define EAX 6
-#define DS 7
-#define ES 8
-#define FS 9
-#define GS 10
-#define ORIG_EAX 11
-#define EIP 12
-#define CS  13
-#define EFL 14
-#define UESP 15
-#define SS   16
-#define FRAME_SIZE 17
+#include <asm/ptrace-abi.h>
 
 /* this struct defines the way the registers are stored on the 
    stack during a system call. */
@@ -41,22 +24,6 @@ struct pt_regs {
 	int  xss;
 };
 
-/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
-#define PTRACE_GETREGS            12
-#define PTRACE_SETREGS            13
-#define PTRACE_GETFPREGS          14
-#define PTRACE_SETFPREGS          15
-#define PTRACE_GETFPXREGS         18
-#define PTRACE_SETFPXREGS         19
-
-#define PTRACE_OLDSETOPTIONS         21
-
-#define PTRACE_GET_THREAD_AREA    25
-#define PTRACE_SET_THREAD_AREA    26
-
-#define PTRACE_SYSEMU		  31
-#define PTRACE_SYSEMU_SINGLESTEP  32
-
 #ifdef __KERNEL__
 
 #include <asm/vm86.h>
diff --git a/include/asm-x86_64/Kbuild b/include/asm-x86_64/Kbuild
index 40f2f13fe174..1ee9b07f3fe6 100644
--- a/include/asm-x86_64/Kbuild
+++ b/include/asm-x86_64/Kbuild
@@ -11,6 +11,7 @@ header-y += debugreg.h
 header-y += ldt.h
 header-y += msr.h
 header-y += prctl.h
+header-y += ptrace-abi.h
 header-y += setup.h
 header-y += sigcontext32.h
 header-y += ucontext.h
diff --git a/include/asm-x86_64/ptrace-abi.h b/include/asm-x86_64/ptrace-abi.h
new file mode 100644
index 000000000000..19184b0806b1
--- /dev/null
+++ b/include/asm-x86_64/ptrace-abi.h
@@ -0,0 +1,51 @@
+#ifndef _X86_64_PTRACE_ABI_H
+#define _X86_64_PTRACE_ABI_H
+
+#if defined(__ASSEMBLY__) || defined(__FRAME_OFFSETS)
+#define R15 0
+#define R14 8
+#define R13 16
+#define R12 24
+#define RBP 32
+#define RBX 40
+/* arguments: interrupts/non tracing syscalls only save upto here*/
+#define R11 48
+#define R10 56
+#define R9 64
+#define R8 72
+#define RAX 80
+#define RCX 88
+#define RDX 96
+#define RSI 104
+#define RDI 112
+#define ORIG_RAX 120       /* = ERROR */
+/* end of arguments */
+/* cpu exception frame or undefined in case of fast syscall. */
+#define RIP 128
+#define CS 136
+#define EFLAGS 144
+#define RSP 152
+#define SS 160
+#define ARGOFFSET R11
+#endif /* __ASSEMBLY__ */
+
+/* top of stack page */
+#define FRAME_SIZE 168
+
+#define PTRACE_OLDSETOPTIONS         21
+
+/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
+#define PTRACE_GETREGS            12
+#define PTRACE_SETREGS            13
+#define PTRACE_GETFPREGS          14
+#define PTRACE_SETFPREGS          15
+#define PTRACE_GETFPXREGS         18
+#define PTRACE_SETFPXREGS         19
+
+/* only useful for access 32bit programs */
+#define PTRACE_GET_THREAD_AREA    25
+#define PTRACE_SET_THREAD_AREA    26
+
+#define PTRACE_ARCH_PRCTL	  30	/* arch_prctl for child */
+
+#endif
diff --git a/include/asm-x86_64/ptrace.h b/include/asm-x86_64/ptrace.h
index ca6f15ff61d4..ab827dc381d7 100644
--- a/include/asm-x86_64/ptrace.h
+++ b/include/asm-x86_64/ptrace.h
@@ -1,40 +1,9 @@
 #ifndef _X86_64_PTRACE_H
 #define _X86_64_PTRACE_H
 
-#if defined(__ASSEMBLY__) || defined(__FRAME_OFFSETS) 
-#define R15 0
-#define R14 8
-#define R13 16
-#define R12 24
-#define RBP 32
-#define RBX 40
-/* arguments: interrupts/non tracing syscalls only save upto here*/
-#define R11 48
-#define R10 56	
-#define R9 64
-#define R8 72
-#define RAX 80
-#define RCX 88
-#define RDX 96
-#define RSI 104
-#define RDI 112
-#define ORIG_RAX 120       /* = ERROR */ 
-/* end of arguments */ 	
-/* cpu exception frame or undefined in case of fast syscall. */
-#define RIP 128
-#define CS 136
-#define EFLAGS 144
-#define RSP 152
-#define SS 160
-#define ARGOFFSET R11
-#endif /* __ASSEMBLY__ */
+#include <asm/ptrace-abi.h>
 
-/* top of stack page */ 
-#define FRAME_SIZE 168
-
-#define PTRACE_OLDSETOPTIONS         21
-
-#ifndef __ASSEMBLY__ 
+#ifndef __ASSEMBLY__
 
 struct pt_regs {
 	unsigned long r15;
@@ -45,7 +14,7 @@ struct pt_regs {
 	unsigned long rbx;
 /* arguments: non interrupts/non tracing syscalls only save upto here*/
  	unsigned long r11;
-	unsigned long r10;	
+	unsigned long r10;
 	unsigned long r9;
 	unsigned long r8;
 	unsigned long rax;
@@ -54,32 +23,18 @@ struct pt_regs {
 	unsigned long rsi;
 	unsigned long rdi;
 	unsigned long orig_rax;
-/* end of arguments */ 	
+/* end of arguments */
 /* cpu exception frame or undefined */
 	unsigned long rip;
 	unsigned long cs;
-	unsigned long eflags; 
-	unsigned long rsp; 
+	unsigned long eflags;
+	unsigned long rsp;
 	unsigned long ss;
-/* top of stack page */ 
+/* top of stack page */
 };
 
 #endif
 
-/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
-#define PTRACE_GETREGS            12
-#define PTRACE_SETREGS            13
-#define PTRACE_GETFPREGS          14
-#define PTRACE_SETFPREGS          15
-#define PTRACE_GETFPXREGS         18
-#define PTRACE_SETFPXREGS         19
-
-/* only useful for access 32bit programs */
-#define PTRACE_GET_THREAD_AREA    25
-#define PTRACE_SET_THREAD_AREA    26
-
-#define PTRACE_ARCH_PRCTL	  30	/* arch_prctl for child */
-
 #if defined(__KERNEL__) && !defined(__ASSEMBLY__) 
 #define user_mode(regs) (!!((regs)->cs & 3))
 #define user_mode_vm(regs) user_mode(regs)
-- 
cgit v1.2.3


From e8df8c3304cfc7d68e9a6a7b6df6c64783cd6991 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:10 -0700
Subject: [PATCH] Make UML use ptrace-abi.h

Include the host architecture's ptrace-abi.h instead of ptrace.h.

There was some cpp mangling of names around the ptrace.h include to avoid
symbol clashes between UML and the host architecture.  Most of these can go
away.  The exception is struct pt_regs, which is convenient to have in
userspace, but must be renamed in order that UML can define its own.

ptrace-x86_64.h needed to have some now-obsolete cpp cruft and a declaration
removed.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-um/ptrace-generic.h | 14 +-------------
 include/asm-um/ptrace-x86_64.h  |  4 ----
 2 files changed, 1 insertion(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-um/ptrace-generic.h b/include/asm-um/ptrace-generic.h
index a36f5371b36b..99c87c5ce994 100644
--- a/include/asm-um/ptrace-generic.h
+++ b/include/asm-um/ptrace-generic.h
@@ -8,19 +8,7 @@
 
 #ifndef __ASSEMBLY__
 
-
-#define pt_regs pt_regs_subarch
-#define show_regs show_regs_subarch
-#define send_sigtrap send_sigtrap_subarch
-
-#include "asm/arch/ptrace.h"
-
-#undef pt_regs
-#undef show_regs
-#undef send_sigtrap
-#undef user_mode
-#undef instruction_pointer
-
+#include "asm/arch/ptrace-abi.h"
 #include "sysdep/ptrace.h"
 
 struct pt_regs {
diff --git a/include/asm-um/ptrace-x86_64.h b/include/asm-um/ptrace-x86_64.h
index c894e68b1f96..2074483e6ca4 100644
--- a/include/asm-um/ptrace-x86_64.h
+++ b/include/asm-um/ptrace-x86_64.h
@@ -11,15 +11,11 @@
 #include "asm/errno.h"
 #include "asm/host_ldt.h"
 
-#define signal_fault signal_fault_x86_64
 #define __FRAME_OFFSETS /* Needed to get the R* macros */
 #include "asm/ptrace-generic.h"
-#undef signal_fault
 
 #define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64
 
-void signal_fault(struct pt_regs_subarch *regs, void *frame, char *where);
-
 #define FS_BASE (21 * sizeof(unsigned long))
 #define GS_BASE (22 * sizeof(unsigned long))
 #define DS (23 * sizeof(unsigned long))
-- 
cgit v1.2.3


From 214ddde2f95037e129eff7e895869771719c7c1b Mon Sep 17 00:00:00 2001
From: bibo mao <bibo.mao@intel.com>
Date: Tue, 26 Sep 2006 11:20:37 -0700
Subject: [IA64] kprobe opcode 16 bytes alignment on IA64

On IA64 instruction opcode must be 16 bytes alignment, in kprobe structure
there is one element to save original instruction, currently saved opcode
is not statically allocated in kprobe structure, that can not assure
16 bytes alignment. This patch dynamically allocated kprobe instruction
opcode to assure 16 bytes alignment.

Signed-off-by: bibo mao <bibo.mao@intel.com>
Acked-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/kprobes.c | 55 ++++++++++++++++++++++++----------------------
 include/asm-ia64/kprobes.h |  9 ++++----
 2 files changed, 34 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 781960f80b6f..320cb7a91c9d 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -136,10 +136,8 @@ static void __kprobes update_kprobe_inst_flag(uint template, uint  slot,
 static int __kprobes unsupported_inst(uint template, uint  slot,
 				      uint major_opcode,
 				      unsigned long kprobe_inst,
-				      struct kprobe *p)
+				      unsigned long addr)
 {
-	unsigned long addr = (unsigned long)p->addr;
-
 	if (bundle_encoding[template][slot] == I) {
 		switch (major_opcode) {
 			case 0x0: //I_UNIT_MISC_OPCODE:
@@ -217,7 +215,7 @@ static void __kprobes prepare_break_inst(uint template, uint  slot,
 					 struct kprobe *p)
 {
 	unsigned long break_inst = BREAK_INST;
-	bundle_t *bundle = &p->ainsn.insn.bundle;
+	bundle_t *bundle = &p->opcode.bundle;
 
 	/*
 	 * Copy the original kprobe_inst qualifying predicate(qp)
@@ -423,11 +421,9 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 	unsigned long *kprobe_addr = (unsigned long *)(addr & ~0xFULL);
 	unsigned long kprobe_inst=0;
 	unsigned int slot = addr & 0xf, template, major_opcode = 0;
-	bundle_t *bundle = &p->ainsn.insn.bundle;
-
-	memcpy(&p->opcode.bundle, kprobe_addr, sizeof(bundle_t));
-	memcpy(&p->ainsn.insn.bundle, kprobe_addr, sizeof(bundle_t));
+	bundle_t *bundle;
 
+	bundle = &((kprobe_opcode_t *)kprobe_addr)->bundle;
  	template = bundle->quad0.template;
 
 	if(valid_kprobe_addr(template, slot, addr))
@@ -440,20 +436,19 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 	/* Get kprobe_inst and major_opcode from the bundle */
 	get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode);
 
-	if (unsupported_inst(template, slot, major_opcode, kprobe_inst, p))
+	if (unsupported_inst(template, slot, major_opcode, kprobe_inst, addr))
 			return -EINVAL;
 
-	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p);
 
-	return 0;
-}
+	p->ainsn.insn = get_insn_slot();
+	if (!p->ainsn.insn)
+		return -ENOMEM;
+	memcpy(&p->opcode, kprobe_addr, sizeof(kprobe_opcode_t));
+	memcpy(p->ainsn.insn, kprobe_addr, sizeof(kprobe_opcode_t));
 
-void __kprobes flush_insn_slot(struct kprobe *p)
-{
-	unsigned long arm_addr;
+	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p);
 
-	arm_addr = ((unsigned long)&p->opcode.bundle) & ~0xFULL;
-	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+	return 0;
 }
 
 void __kprobes arch_arm_kprobe(struct kprobe *p)
@@ -461,9 +456,10 @@ void __kprobes arch_arm_kprobe(struct kprobe *p)
 	unsigned long addr = (unsigned long)p->addr;
 	unsigned long arm_addr = addr & ~0xFULL;
 
-	flush_insn_slot(p);
-	memcpy((char *)arm_addr, &p->ainsn.insn.bundle, sizeof(bundle_t));
-	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+	flush_icache_range((unsigned long)p->ainsn.insn,
+			(unsigned long)p->ainsn.insn + sizeof(kprobe_opcode_t));
+	memcpy((char *)arm_addr, &p->opcode, sizeof(kprobe_opcode_t));
+	flush_icache_range(arm_addr, arm_addr + sizeof(kprobe_opcode_t));
 }
 
 void __kprobes arch_disarm_kprobe(struct kprobe *p)
@@ -471,11 +467,18 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 	unsigned long addr = (unsigned long)p->addr;
 	unsigned long arm_addr = addr & ~0xFULL;
 
-	/* p->opcode contains the original unaltered bundle */
-	memcpy((char *) arm_addr, (char *) &p->opcode.bundle, sizeof(bundle_t));
-	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+	/* p->ainsn.insn contains the original unaltered kprobe_opcode_t */
+	memcpy((char *) arm_addr, (char *) p->ainsn.insn,
+					 sizeof(kprobe_opcode_t));
+	flush_icache_range(arm_addr, arm_addr + sizeof(kprobe_opcode_t));
 }
 
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	mutex_lock(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	mutex_unlock(&kprobe_mutex);
+}
 /*
  * We are resuming execution after a single step fault, so the pt_regs
  * structure reflects the register state after we executed the instruction
@@ -486,12 +489,12 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
  */
 static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
-  	unsigned long bundle_addr = ((unsigned long) (&p->opcode.bundle)) & ~0xFULL;
+  	unsigned long bundle_addr = (unsigned long) (&p->ainsn.insn->bundle);
   	unsigned long resume_addr = (unsigned long)p->addr & ~0xFULL;
  	unsigned long template;
  	int slot = ((unsigned long)p->addr & 0xf);
 
-	template = p->opcode.bundle.quad0.template;
+	template = p->ainsn.insn->bundle.quad0.template;
 
  	if (slot == 1 && bundle_encoding[template][1] == L)
  		slot = 2;
@@ -553,7 +556,7 @@ turn_ss_off:
 
 static void __kprobes prepare_ss(struct kprobe *p, struct pt_regs *regs)
 {
-	unsigned long bundle_addr = (unsigned long) &p->opcode.bundle;
+	unsigned long bundle_addr = (unsigned long) &p->ainsn.insn->bundle;
 	unsigned long slot = (unsigned long)p->addr & 0xf;
 
 	/* single step inline if break instruction */
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index 938904910115..1b45b71c79b9 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -29,7 +29,8 @@
 #include <linux/percpu.h>
 #include <asm/break.h>
 
-#define MAX_INSN_SIZE   16
+#define __ARCH_WANT_KPROBES_INSN_SLOT
+#define MAX_INSN_SIZE   1
 #define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
 
 typedef union cmp_inst {
@@ -94,7 +95,7 @@ struct kprobe_ctlblk {
 #define IP_RELATIVE_PREDICT_OPCODE	(7)
 #define LONG_BRANCH_OPCODE		(0xC)
 #define LONG_CALL_OPCODE		(0xD)
-#define arch_remove_kprobe(p)		do {} while (0)
+#define flush_insn_slot(p)		do { } while (0)
 
 typedef struct kprobe_opcode {
 	bundle_t bundle;
@@ -108,7 +109,7 @@ struct fnptr {
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
 	/* copy of the instruction to be emulated */
-	kprobe_opcode_t insn;
+	kprobe_opcode_t *insn;
  #define INST_FLAG_FIX_RELATIVE_IP_ADDR		1
  #define INST_FLAG_FIX_BRANCH_REG		2
  #define INST_FLAG_BREAK_INST			4
@@ -125,6 +126,6 @@ static inline void jprobe_return(void)
 }
 extern void invalidate_stacked_regs(void);
 extern void flush_register_stack(void);
-extern void flush_insn_slot(struct kprobe *p);
+extern void arch_remove_kprobe(struct kprobe *p);
 
 #endif				/* _ASM_KPROBES_H */
-- 
cgit v1.2.3


From dd562c05410e13e878a3ee0deb8ac06db2e132c7 Mon Sep 17 00:00:00 2001
From: Stephane Eranian <eranian@hpl.hp.com>
Date: Thu, 21 Sep 2006 10:35:44 -0700
Subject: [IA64] Add interface so modules can discover whether multithreading
 is on.

Add is_multithreading_enabled() to check whether multi-threading
is enabled independently of which cpu is currently online

Signed-off-by: stephane eranian <eranian@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smpboot.c | 24 ++++++++++++++++++++++++
 include/asm-ia64/smp.h     |  1 +
 2 files changed, 25 insertions(+)

(limited to 'include')

diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 6203ed4ec8cf..f7d7f5668144 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -879,3 +879,27 @@ identify_siblings(struct cpuinfo_ia64 *c)
 	c->core_id = info.log1_cid;
 	c->thread_id = info.log1_tid;
 }
+
+/*
+ * returns non zero, if multi-threading is enabled
+ * on at least one physical package. Due to hotplug cpu
+ * and (maxcpus=), all threads may not necessarily be enabled
+ * even though the processor supports multi-threading.
+ */
+int is_multithreading_enabled(void)
+{
+	int i, j;
+
+	for_each_present_cpu(i) {
+		for_each_present_cpu(j) {
+			if (j == i)
+				continue;
+			if ((cpu_data(j)->socket_id == cpu_data(i)->socket_id)) {
+				if (cpu_data(j)->core_id == cpu_data(i)->core_id)
+					return 1;
+			}
+		}
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(is_multithreading_enabled);
diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h
index 719ff309ce09..6533eb4e86ad 100644
--- a/include/asm-ia64/smp.h
+++ b/include/asm-ia64/smp.h
@@ -128,6 +128,7 @@ extern void smp_send_reschedule (int cpu);
 extern void lock_ipi_calllock(void);
 extern void unlock_ipi_calllock(void);
 extern void identify_siblings (struct cpuinfo_ia64 *);
+extern int is_multithreading_enabled(void);
 
 #else
 
-- 
cgit v1.2.3


From 35589a8fa8138244e7f2ef9317c440aa580c9335 Mon Sep 17 00:00:00 2001
From: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>
Date: Tue, 26 Sep 2006 12:03:13 -0700
Subject: [IA64] Move perfmon tables from thread_struct to pfm_context

This patch renders thread_struct->pmcs[] and thread_struct->pmds[]
OBSOLETE. The actual table is moved to pfm_context structure which
saves space in thread_struct (in turn saving space in task_struct
which frees up more space for kernel stacks).

Signed-off-by: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/perfmon.c   | 113 +++++++++++++++++++------------------------
 include/asm-ia64/processor.h |  12 +----
 2 files changed, 50 insertions(+), 75 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 84a7e52f56f6..45000d5d0cfa 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -62,6 +62,9 @@
 
 #define PFM_INVALID_ACTIVATION	(~0UL)
 
+#define PFM_NUM_PMC_REGS	64	/* PMC save area for ctxsw */
+#define PFM_NUM_PMD_REGS	64	/* PMD save area for ctxsw */
+
 /*
  * depth of message queue
  */
@@ -296,14 +299,17 @@ typedef struct pfm_context {
 	unsigned long		ctx_reload_pmcs[4];	/* bitmask of force reload PMC on ctxsw in */
 	unsigned long		ctx_used_monitors[4];	/* bitmask of monitor PMC being used */
 
-	unsigned long		ctx_pmcs[IA64_NUM_PMC_REGS];	/*  saved copies of PMC values */
+	unsigned long		ctx_pmcs[PFM_NUM_PMC_REGS];	/*  saved copies of PMC values */
 
 	unsigned int		ctx_used_ibrs[1];		/* bitmask of used IBR (speedup ctxsw in) */
 	unsigned int		ctx_used_dbrs[1];		/* bitmask of used DBR (speedup ctxsw in) */
 	unsigned long		ctx_dbrs[IA64_NUM_DBG_REGS];	/* DBR values (cache) when not loaded */
 	unsigned long		ctx_ibrs[IA64_NUM_DBG_REGS];	/* IBR values (cache) when not loaded */
 
-	pfm_counter_t		ctx_pmds[IA64_NUM_PMD_REGS]; /* software state for PMDS */
+	pfm_counter_t		ctx_pmds[PFM_NUM_PMD_REGS]; /* software state for PMDS */
+
+	unsigned long		th_pmcs[PFM_NUM_PMC_REGS];	/* PMC thread save state */
+	unsigned long		th_pmds[PFM_NUM_PMD_REGS];	/* PMD thread save state */
 
 	u64			ctx_saved_psr_up;	/* only contains psr.up value */
 
@@ -867,7 +873,6 @@ static void
 pfm_mask_monitoring(struct task_struct *task)
 {
 	pfm_context_t *ctx = PFM_GET_CTX(task);
-	struct thread_struct *th = &task->thread;
 	unsigned long mask, val, ovfl_mask;
 	int i;
 
@@ -888,7 +893,7 @@ pfm_mask_monitoring(struct task_struct *task)
 	 * So in both cases, the live register contains the owner's
 	 * state. We can ONLY touch the PMU registers and NOT the PSR.
 	 *
-	 * As a consequence to this call, the thread->pmds[] array
+	 * As a consequence to this call, the ctx->th_pmds[] array
 	 * contains stale information which must be ignored
 	 * when context is reloaded AND monitoring is active (see
 	 * pfm_restart).
@@ -923,9 +928,9 @@ pfm_mask_monitoring(struct task_struct *task)
 	mask = ctx->ctx_used_monitors[0] >> PMU_FIRST_COUNTER;
 	for(i= PMU_FIRST_COUNTER; mask; i++, mask>>=1) {
 		if ((mask & 0x1) == 0UL) continue;
-		ia64_set_pmc(i, th->pmcs[i] & ~0xfUL);
-		th->pmcs[i] &= ~0xfUL;
-		DPRINT_ovfl(("pmc[%d]=0x%lx\n", i, th->pmcs[i]));
+		ia64_set_pmc(i, ctx->th_pmcs[i] & ~0xfUL);
+		ctx->th_pmcs[i] &= ~0xfUL;
+		DPRINT_ovfl(("pmc[%d]=0x%lx\n", i, ctx->th_pmcs[i]));
 	}
 	/*
 	 * make all of this visible
@@ -942,7 +947,6 @@ static void
 pfm_restore_monitoring(struct task_struct *task)
 {
 	pfm_context_t *ctx = PFM_GET_CTX(task);
-	struct thread_struct *th = &task->thread;
 	unsigned long mask, ovfl_mask;
 	unsigned long psr, val;
 	int i, is_system;
@@ -1008,9 +1012,9 @@ pfm_restore_monitoring(struct task_struct *task)
 	mask = ctx->ctx_used_monitors[0] >> PMU_FIRST_COUNTER;
 	for(i= PMU_FIRST_COUNTER; mask; i++, mask>>=1) {
 		if ((mask & 0x1) == 0UL) continue;
-		th->pmcs[i] = ctx->ctx_pmcs[i];
-		ia64_set_pmc(i, th->pmcs[i]);
-		DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, th->pmcs[i]));
+		ctx->th_pmcs[i] = ctx->ctx_pmcs[i];
+		ia64_set_pmc(i, ctx->th_pmcs[i]);
+		DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, ctx->th_pmcs[i]));
 	}
 	ia64_srlz_d();
 
@@ -1069,7 +1073,6 @@ pfm_restore_pmds(unsigned long *pmds, unsigned long mask)
 static inline void
 pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx)
 {
-	struct thread_struct *thread = &task->thread;
 	unsigned long ovfl_val = pmu_conf->ovfl_val;
 	unsigned long mask = ctx->ctx_all_pmds[0];
 	unsigned long val;
@@ -1091,11 +1094,11 @@ pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx)
 			ctx->ctx_pmds[i].val = val & ~ovfl_val;
 			 val &= ovfl_val;
 		}
-		thread->pmds[i] = val;
+		ctx->th_pmds[i] = val;
 
 		DPRINT(("pmd[%d]=0x%lx soft_val=0x%lx\n",
 			i,
-			thread->pmds[i],
+			ctx->th_pmds[i],
 			ctx->ctx_pmds[i].val));
 	}
 }
@@ -1106,7 +1109,6 @@ pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx)
 static inline void
 pfm_copy_pmcs(struct task_struct *task, pfm_context_t *ctx)
 {
-	struct thread_struct *thread = &task->thread;
 	unsigned long mask = ctx->ctx_all_pmcs[0];
 	int i;
 
@@ -1114,8 +1116,8 @@ pfm_copy_pmcs(struct task_struct *task, pfm_context_t *ctx)
 
 	for (i=0; mask; i++, mask>>=1) {
 		/* masking 0 with ovfl_val yields 0 */
-		thread->pmcs[i] = ctx->ctx_pmcs[i];
-		DPRINT(("pmc[%d]=0x%lx\n", i, thread->pmcs[i]));
+		ctx->th_pmcs[i] = ctx->ctx_pmcs[i];
+		DPRINT(("pmc[%d]=0x%lx\n", i, ctx->th_pmcs[i]));
 	}
 }
 
@@ -2859,7 +2861,6 @@ pfm_reset_regs(pfm_context_t *ctx, unsigned long *ovfl_regs, int is_long_reset)
 static int
 pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 {
-	struct thread_struct *thread = NULL;
 	struct task_struct *task;
 	pfarg_reg_t *req = (pfarg_reg_t *)arg;
 	unsigned long value, pmc_pm;
@@ -2880,7 +2881,6 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	if (state == PFM_CTX_ZOMBIE) return -EINVAL;
 
 	if (is_loaded) {
-		thread = &task->thread;
 		/*
 		 * In system wide and when the context is loaded, access can only happen
 		 * when the caller is running on the CPU being monitored by the session.
@@ -3035,7 +3035,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 		 *
 		 * The value in ctx_pmcs[] can only be changed in pfm_write_pmcs().
 		 *
-		 * The value in thread->pmcs[] may be modified on overflow, i.e.,  when
+		 * The value in th_pmcs[] may be modified on overflow, i.e.,  when
 		 * monitoring needs to be stopped.
 		 */
 		if (is_monitor) CTX_USED_MONITOR(ctx, 1UL << cnum);
@@ -3049,7 +3049,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			/*
 			 * write thread state
 			 */
-			if (is_system == 0) thread->pmcs[cnum] = value;
+			if (is_system == 0) ctx->th_pmcs[cnum] = value;
 
 			/*
 			 * write hardware register if we can
@@ -3101,7 +3101,6 @@ error:
 static int
 pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 {
-	struct thread_struct *thread = NULL;
 	struct task_struct *task;
 	pfarg_reg_t *req = (pfarg_reg_t *)arg;
 	unsigned long value, hw_value, ovfl_mask;
@@ -3125,7 +3124,6 @@ pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	 * the owner of the local PMU.
 	 */
 	if (likely(is_loaded)) {
-		thread = &task->thread;
 		/*
 		 * In system wide and when the context is loaded, access can only happen
 		 * when the caller is running on the CPU being monitored by the session.
@@ -3233,7 +3231,7 @@ pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			/*
 		 	 * write thread state
 		 	 */
-			if (is_system == 0) thread->pmds[cnum] = hw_value;
+			if (is_system == 0) ctx->th_pmds[cnum] = hw_value;
 
 			/*
 			 * write hardware register if we can
@@ -3299,7 +3297,6 @@ abort_mission:
 static int
 pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 {
-	struct thread_struct *thread = NULL;
 	struct task_struct *task;
 	unsigned long val = 0UL, lval, ovfl_mask, sval;
 	pfarg_reg_t *req = (pfarg_reg_t *)arg;
@@ -3323,7 +3320,6 @@ pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	if (state == PFM_CTX_ZOMBIE) return -EINVAL;
 
 	if (likely(is_loaded)) {
-		thread = &task->thread;
 		/*
 		 * In system wide and when the context is loaded, access can only happen
 		 * when the caller is running on the CPU being monitored by the session.
@@ -3385,7 +3381,7 @@ pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			 * if context is zombie, then task does not exist anymore.
 			 * In this case, we use the full value saved in the context (pfm_flush_regs()).
 			 */
-			val = is_loaded ? thread->pmds[cnum] : 0UL;
+			val = is_loaded ? ctx->th_pmds[cnum] : 0UL;
 		}
 		rd_func = pmu_conf->pmd_desc[cnum].read_check;
 
@@ -4354,8 +4350,8 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	pfm_copy_pmds(task, ctx);
 	pfm_copy_pmcs(task, ctx);
 
-	pmcs_source = thread->pmcs;
-	pmds_source = thread->pmds;
+	pmcs_source = ctx->th_pmcs;
+	pmds_source = ctx->th_pmds;
 
 	/*
 	 * always the case for system-wide
@@ -5864,14 +5860,12 @@ void
 pfm_save_regs(struct task_struct *task)
 {
 	pfm_context_t *ctx;
-	struct thread_struct *t;
 	unsigned long flags;
 	u64 psr;
 
 
 	ctx = PFM_GET_CTX(task);
 	if (ctx == NULL) return;
-	t = &task->thread;
 
 	/*
  	 * we always come here with interrupts ALREADY disabled by
@@ -5929,19 +5923,19 @@ pfm_save_regs(struct task_struct *task)
 	 * guarantee we will be schedule at that same
 	 * CPU again.
 	 */
-	pfm_save_pmds(t->pmds, ctx->ctx_used_pmds[0]);
+	pfm_save_pmds(ctx->th_pmds, ctx->ctx_used_pmds[0]);
 
 	/*
 	 * save pmc0 ia64_srlz_d() done in pfm_save_pmds()
 	 * we will need it on the restore path to check
 	 * for pending overflow.
 	 */
-	t->pmcs[0] = ia64_get_pmc(0);
+	ctx->th_pmcs[0] = ia64_get_pmc(0);
 
 	/*
 	 * unfreeze PMU if had pending overflows
 	 */
-	if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
+	if (ctx->th_pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
 
 	/*
 	 * finally, allow context access.
@@ -5986,7 +5980,6 @@ static void
 pfm_lazy_save_regs (struct task_struct *task)
 {
 	pfm_context_t *ctx;
-	struct thread_struct *t;
 	unsigned long flags;
 
 	{ u64 psr  = pfm_get_psr();
@@ -5994,7 +5987,6 @@ pfm_lazy_save_regs (struct task_struct *task)
 	}
 
 	ctx = PFM_GET_CTX(task);
-	t   = &task->thread;
 
 	/*
 	 * we need to mask PMU overflow here to
@@ -6019,19 +6011,19 @@ pfm_lazy_save_regs (struct task_struct *task)
 	/*
 	 * save all the pmds we use
 	 */
-	pfm_save_pmds(t->pmds, ctx->ctx_used_pmds[0]);
+	pfm_save_pmds(ctx->th_pmds, ctx->ctx_used_pmds[0]);
 
 	/*
 	 * save pmc0 ia64_srlz_d() done in pfm_save_pmds()
 	 * it is needed to check for pended overflow
 	 * on the restore path
 	 */
-	t->pmcs[0] = ia64_get_pmc(0);
+	ctx->th_pmcs[0] = ia64_get_pmc(0);
 
 	/*
 	 * unfreeze PMU if had pending overflows
 	 */
-	if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
+	if (ctx->th_pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
 
 	/*
 	 * now get can unmask PMU interrupts, they will
@@ -6050,7 +6042,6 @@ void
 pfm_load_regs (struct task_struct *task)
 {
 	pfm_context_t *ctx;
-	struct thread_struct *t;
 	unsigned long pmc_mask = 0UL, pmd_mask = 0UL;
 	unsigned long flags;
 	u64 psr, psr_up;
@@ -6061,11 +6052,10 @@ pfm_load_regs (struct task_struct *task)
 
 	BUG_ON(GET_PMU_OWNER());
 
-	t     = &task->thread;
 	/*
 	 * possible on unload
 	 */
-	if (unlikely((t->flags & IA64_THREAD_PM_VALID) == 0)) return;
+	if (unlikely((task->thread.flags & IA64_THREAD_PM_VALID) == 0)) return;
 
 	/*
  	 * we always come here with interrupts ALREADY disabled by
@@ -6147,21 +6137,21 @@ pfm_load_regs (struct task_struct *task)
 	 *
 	 * XXX: optimize here
 	 */
-	if (pmd_mask) pfm_restore_pmds(t->pmds, pmd_mask);
-	if (pmc_mask) pfm_restore_pmcs(t->pmcs, pmc_mask);
+	if (pmd_mask) pfm_restore_pmds(ctx->th_pmds, pmd_mask);
+	if (pmc_mask) pfm_restore_pmcs(ctx->th_pmcs, pmc_mask);
 
 	/*
 	 * check for pending overflow at the time the state
 	 * was saved.
 	 */
-	if (unlikely(PMC0_HAS_OVFL(t->pmcs[0]))) {
+	if (unlikely(PMC0_HAS_OVFL(ctx->th_pmcs[0]))) {
 		/*
 		 * reload pmc0 with the overflow information
 		 * On McKinley PMU, this will trigger a PMU interrupt
 		 */
-		ia64_set_pmc(0, t->pmcs[0]);
+		ia64_set_pmc(0, ctx->th_pmcs[0]);
 		ia64_srlz_d();
-		t->pmcs[0] = 0UL;
+		ctx->th_pmcs[0] = 0UL;
 
 		/*
 		 * will replay the PMU interrupt
@@ -6214,7 +6204,6 @@ pfm_load_regs (struct task_struct *task)
 void
 pfm_load_regs (struct task_struct *task)
 {
-	struct thread_struct *t;
 	pfm_context_t *ctx;
 	struct task_struct *owner;
 	unsigned long pmd_mask, pmc_mask;
@@ -6223,7 +6212,6 @@ pfm_load_regs (struct task_struct *task)
 
 	owner = GET_PMU_OWNER();
 	ctx   = PFM_GET_CTX(task);
-	t     = &task->thread;
 	psr   = pfm_get_psr();
 
 	BUG_ON(psr & (IA64_PSR_UP|IA64_PSR_PP));
@@ -6286,22 +6274,22 @@ pfm_load_regs (struct task_struct *task)
 	 */
 	pmc_mask = ctx->ctx_all_pmcs[0];
 
-	pfm_restore_pmds(t->pmds, pmd_mask);
-	pfm_restore_pmcs(t->pmcs, pmc_mask);
+	pfm_restore_pmds(ctx->th_pmds, pmd_mask);
+	pfm_restore_pmcs(ctx->th_pmcs, pmc_mask);
 
 	/*
 	 * check for pending overflow at the time the state
 	 * was saved.
 	 */
-	if (unlikely(PMC0_HAS_OVFL(t->pmcs[0]))) {
+	if (unlikely(PMC0_HAS_OVFL(ctx->th_pmcs[0]))) {
 		/*
 		 * reload pmc0 with the overflow information
 		 * On McKinley PMU, this will trigger a PMU interrupt
 		 */
-		ia64_set_pmc(0, t->pmcs[0]);
+		ia64_set_pmc(0, ctx->th_pmcs[0]);
 		ia64_srlz_d();
 
-		t->pmcs[0] = 0UL;
+		ctx->th_pmcs[0] = 0UL;
 
 		/*
 		 * will replay the PMU interrupt
@@ -6376,11 +6364,11 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 		 */
 		pfm_unfreeze_pmu();
 	} else {
-		pmc0 = task->thread.pmcs[0];
+		pmc0 = ctx->th_pmcs[0];
 		/*
 		 * clear whatever overflow status bits there were
 		 */
-		task->thread.pmcs[0] = 0;
+		ctx->th_pmcs[0] = 0;
 	}
 	ovfl_val = pmu_conf->ovfl_val;
 	/*
@@ -6401,7 +6389,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 		/*
 		 * can access PMU always true in system wide mode
 		 */
-		val = pmd_val = can_access_pmu ? ia64_get_pmd(i) : task->thread.pmds[i];
+		val = pmd_val = can_access_pmu ? ia64_get_pmd(i) : ctx->th_pmds[i];
 
 		if (PMD_IS_COUNTING(i)) {
 			DPRINT(("[%d] pmd[%d] ctx_pmd=0x%lx hw_pmd=0x%lx\n",
@@ -6433,7 +6421,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 
 		DPRINT(("[%d] ctx_pmd[%d]=0x%lx  pmd_val=0x%lx\n", task->pid, i, val, pmd_val));
 
-		if (is_self) task->thread.pmds[i] = pmd_val;
+		if (is_self) ctx->th_pmds[i] = pmd_val;
 
 		ctx->ctx_pmds[i].val = val;
 	}
@@ -6677,7 +6665,7 @@ pfm_init(void)
 	       ffz(pmu_conf->ovfl_val));
 
 	/* sanity check */
-	if (pmu_conf->num_pmds >= IA64_NUM_PMD_REGS || pmu_conf->num_pmcs >= IA64_NUM_PMC_REGS) {
+	if (pmu_conf->num_pmds >= PFM_NUM_PMD_REGS || pmu_conf->num_pmcs >= PFM_NUM_PMC_REGS) {
 		printk(KERN_ERR "perfmon: not enough pmc/pmd, perfmon disabled\n");
 		pmu_conf = NULL;
 		return -1;
@@ -6752,7 +6740,6 @@ void
 dump_pmu_state(const char *from)
 {
 	struct task_struct *task;
-	struct thread_struct *t;
 	struct pt_regs *regs;
 	pfm_context_t *ctx;
 	unsigned long psr, dcr, info, flags;
@@ -6797,16 +6784,14 @@ dump_pmu_state(const char *from)
 	ia64_psr(regs)->up = 0;
 	ia64_psr(regs)->pp = 0;
 
-	t = &current->thread;
-
 	for (i=1; PMC_IS_LAST(i) == 0; i++) {
 		if (PMC_IS_IMPL(i) == 0) continue;
-		printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, t->pmcs[i]);
+		printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, ctx->th_pmcs[i]);
 	}
 
 	for (i=1; PMD_IS_LAST(i) == 0; i++) {
 		if (PMD_IS_IMPL(i) == 0) continue;
-		printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, t->pmds[i]);
+		printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, ctx->th_pmds[i]);
 	}
 
 	if (ctx) {
diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h
index ead68f8eb883..5830d36fd8e6 100644
--- a/include/asm-ia64/processor.h
+++ b/include/asm-ia64/processor.h
@@ -20,12 +20,6 @@
 #include <asm/ustack.h>
 
 #define IA64_NUM_DBG_REGS	8
-/*
- * Limits for PMC and PMD are set to less than maximum architected values
- * but should be sufficient for a while
- */
-#define IA64_NUM_PMC_REGS	64
-#define IA64_NUM_PMD_REGS	64
 
 #define DEFAULT_MAP_BASE	__IA64_UL_CONST(0x2000000000000000)
 #define DEFAULT_TASK_SIZE	__IA64_UL_CONST(0xa000000000000000)
@@ -263,13 +257,9 @@ struct thread_struct {
 # define INIT_THREAD_IA32
 #endif /* CONFIG_IA32_SUPPORT */
 #ifdef CONFIG_PERFMON
-	__u64 pmcs[IA64_NUM_PMC_REGS];
-	__u64 pmds[IA64_NUM_PMD_REGS];
 	void *pfm_context;		     /* pointer to detailed PMU context */
 	unsigned long pfm_needs_checking;    /* when >0, pending perfmon work on kernel exit */
-# define INIT_THREAD_PM		.pmcs =			{0UL, },  \
-				.pmds =			{0UL, },  \
-				.pfm_context =		NULL,     \
+# define INIT_THREAD_PM		.pfm_context =		NULL,     \
 				.pfm_needs_checking =	0UL,
 #else
 # define INIT_THREAD_PM
-- 
cgit v1.2.3


From a192dc16000241dc02990a36b6830839b73c44de Mon Sep 17 00:00:00 2001
From: Jakub Jelinek <jakub@redhat.com>
Date: Tue, 26 Sep 2006 14:00:56 -0700
Subject: [IA64] Implement futex primitives

Implement futex_atomic_op_inuser() and futex_atomic_cmpxchg_inatomic()
on IA64 in order to fully support all futex functionality.

Signed-off-by: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/futex.h | 122 ++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 120 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-ia64/futex.h b/include/asm-ia64/futex.h
index 6a332a9f099c..07d77f3a8cbe 100644
--- a/include/asm-ia64/futex.h
+++ b/include/asm-ia64/futex.h
@@ -1,6 +1,124 @@
 #ifndef _ASM_FUTEX_H
 #define _ASM_FUTEX_H
 
-#include <asm-generic/futex.h>
+#include <linux/futex.h>
+#include <asm/errno.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
 
-#endif
+#define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \
+do {									\
+	register unsigned long r8 __asm ("r8") = 0;			\
+	__asm__ __volatile__(						\
+		"	mf;;					\n"	\
+		"[1:] "	insn ";;				\n"	\
+		"	.xdata4 \"__ex_table\", 1b-., 2f-.	\n"	\
+		"[2:]"							\
+		: "+r" (r8), "=r" (oldval)				\
+		: "r" (uaddr), "r" (oparg)				\
+		: "memory");						\
+	ret = r8;							\
+} while (0)
+
+#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \
+do {									\
+	register unsigned long r8 __asm ("r8") = 0;			\
+	int val, newval;						\
+	do {								\
+		__asm__ __volatile__(					\
+			"	mf;;				  \n"	\
+			"[1:]	ld4 %3=[%4];;			  \n"	\
+			"	mov %2=%3			  \n"	\
+				insn	";;			  \n"	\
+			"	mov ar.ccv=%2;;			  \n"	\
+			"[2:]	cmpxchg4.acq %1=[%4],%3,ar.ccv;;  \n"	\
+			"	.xdata4 \"__ex_table\", 1b-., 3f-.\n"	\
+			"	.xdata4 \"__ex_table\", 2b-., 3f-.\n"	\
+			"[3:]"						\
+			: "+r" (r8), "=r" (val), "=&r" (oldval),	\
+			   "=&r" (newval)				\
+			: "r" (uaddr), "r" (oparg)			\
+			: "memory");					\
+		if (unlikely (r8))					\
+			break;						\
+	} while (unlikely (val != oldval));				\
+	ret = r8;							\
+} while (0)
+
+static inline int
+futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
+{
+	int op = (encoded_op >> 28) & 7;
+	int cmp = (encoded_op >> 24) & 15;
+	int oparg = (encoded_op << 8) >> 20;
+	int cmparg = (encoded_op << 20) >> 20;
+	int oldval = 0, ret;
+	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+		oparg = 1 << oparg;
+
+	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
+		return -EFAULT;
+
+	inc_preempt_count();
+
+	switch (op) {
+	case FUTEX_OP_SET:
+		__futex_atomic_op1("xchg4 %1=[%2],%3", ret, oldval, uaddr,
+				   oparg);
+		break;
+	case FUTEX_OP_ADD:
+		__futex_atomic_op2("add %3=%3,%5", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_atomic_op2("or %3=%3,%5", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_atomic_op2("and %3=%3,%5", ret, oldval, uaddr,
+				   ~oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_atomic_op2("xor %3=%3,%5", ret, oldval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
+	}
+
+	dec_preempt_count();
+
+	if (!ret) {
+		switch (cmp) {
+		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
+		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
+		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
+		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
+		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
+		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
+		default: ret = -ENOSYS;
+		}
+	}
+	return ret;
+}
+
+static inline int
+futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+{
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+		return -EFAULT;
+
+	{
+		register unsigned long r8 __asm ("r8");
+		__asm__ __volatile__(
+			"	mf;;					\n"
+			"	mov ar.ccv=%3;;				\n"
+			"[1:]	cmpxchg4.acq %0=[%1],%2,ar.ccv		\n"
+			"	.xdata4 \"__ex_table\", 1b-., 2f-.	\n"
+			"[2:]"
+			: "=r" (r8)
+			: "r" (uaddr), "r" (newval),
+			  "rO" ((long) (unsigned) oldval)
+			: "memory");
+		return r8;
+	}
+}
+
+#endif /* _ASM_FUTEX_H */
-- 
cgit v1.2.3


From 5c55cd63a77a85f603c98c2171a8054ca34b6a9f Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Tue, 26 Sep 2006 14:04:42 -0700
Subject: Revert "[IA64] Unwire set/get_robust_list"

This reverts commit 2636255488484e04d6d54303d2b0ec30f7ef7e02.

Jakub Jelinek provided the missing futex_atomic_cmpxchg_inatomic()
function, so now it should be safe to re-enable these syscalls.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S  | 4 ++--
 include/asm-ia64/unistd.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index fef06571be99..12701cf32d99 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1605,8 +1605,8 @@ sys_call_table:
 	data8 sys_ni_syscall			// 1295 reserved for ppoll
 	data8 sys_unshare
 	data8 sys_splice
-	data8 sys_ni_syscall			// reserved for set_robust_list
-	data8 sys_ni_syscall			// reserved for get_robust_list
+	data8 sys_set_robust_list
+	data8 sys_get_robust_list
 	data8 sys_sync_file_range		// 1300
 	data8 sys_tee
 	data8 sys_vmsplice
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index f581662c5ab8..bb0eb727dcd0 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -286,7 +286,8 @@
 /* 1294, 1295 reserved for pselect/ppoll */
 #define __NR_unshare			1296
 #define __NR_splice			1297
-/* 1298, 1299 reserved for set_robust_list/get_robust_list */
+#define __NR_set_robust_list		1298
+#define __NR_get_robust_list		1299
 #define __NR_sync_file_range		1300
 #define __NR_tee			1301
 #define __NR_vmsplice			1302
-- 
cgit v1.2.3


From f5a3f3dc189485d607fbd42678cc23958acc0a6e Mon Sep 17 00:00:00 2001
From: Zou Nan hai <nanhai.zou@intel.com>
Date: Thu, 14 Sep 2006 08:25:15 +0800
Subject: [IA64] Make gp value point to Region 5 in mca handler

MCA dispatch code take physical address of GP passed from SAL, then call
DATA_PA_TO_VA twice on GP before call into C code.  The first time is
in ia64_set_kernel_register, the second time is in VIRTUAL_MODE_ENTER.
The gp is changed to a virtual address in region 7 because DATA_PA_TO_VA
is implemented by dep instruction.

However when notify blocks were called from MCA handler code, because
notify blocks are supported by callback function pointers, gp value
value was switched to region 5 again.

The patch set gp register to kernel gp of region 5 at entry of MCA
dispatch.

Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/mca_asm.S | 5 -----
 include/asm-ia64/mca_asm.h | 2 +-
 2 files changed, 1 insertion(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S
index 96047491d1b9..ee3ff76c8a35 100644
--- a/arch/ia64/kernel/mca_asm.S
+++ b/arch/ia64/kernel/mca_asm.S
@@ -1025,18 +1025,13 @@ ia64_old_stack:
 
 ia64_set_kernel_registers:
 	add temp3=MCA_SP_OFFSET, r3
-	add temp4=MCA_SOS_OFFSET+SOS(OS_GP), r3
 	mov b0=r2		// save return address
 	GET_IA64_MCA_DATA(temp1)
 	;;
-	add temp4=temp4, temp1	// &struct ia64_sal_os_state.os_gp
 	add r12=temp1, temp3	// kernel stack pointer on MCA/INIT stack
 	add r13=temp1, r3	// set current to start of MCA/INIT stack
 	add r20=temp1, r3	// physical start of MCA/INIT stack
 	;;
-	ld8 r1=[temp4]		// OS GP from SAL OS state
-	;;
-	DATA_PA_TO_VA(r1,temp1)
 	DATA_PA_TO_VA(r12,temp2)
 	DATA_PA_TO_VA(r13,temp3)
 	;;
diff --git a/include/asm-ia64/mca_asm.h b/include/asm-ia64/mca_asm.h
index 27c9203d8ce3..76203f9a8718 100644
--- a/include/asm-ia64/mca_asm.h
+++ b/include/asm-ia64/mca_asm.h
@@ -197,9 +197,9 @@
 	movl	temp2 = start_addr;				\
 	;;							\
 	mov	cr.iip = temp2;					\
+	movl	gp = __gp					\
 	;;							\
 	DATA_PA_TO_VA(sp, temp1);				\
-	DATA_PA_TO_VA(gp, temp2);				\
 	srlz.i;							\
 	;;							\
 	nop	1;						\
-- 
cgit v1.2.3


From eb2a2fd91f7c8a53b15063d6f08cf22b9a56cbfb Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Tue, 26 Sep 2006 23:23:45 +0200
Subject: [PATCH] Modularize generic HDLC

This patch enables building of individual WAN protocol support
routines (parts of generic HDLC) as separate modules.
All protocol-private definitions are moved from hdlc.h file
to protocol drivers. User-space interface and interface
between generic HDLC and underlying low-level HDLC drivers
are unchanged.

Signed-off-by: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/wan/Kconfig        |  12 +-
 drivers/net/wan/Makefile       |  19 +-
 drivers/net/wan/hdlc.c         | 368 ++++++++++++++++++++++++++++++++++++++
 drivers/net/wan/hdlc_cisco.c   | 198 ++++++++++++++-------
 drivers/net/wan/hdlc_fr.c      | 389 +++++++++++++++++++++++++----------------
 drivers/net/wan/hdlc_generic.c | 339 -----------------------------------
 drivers/net/wan/hdlc_ppp.c     |  77 ++++++--
 drivers/net/wan/hdlc_raw.c     |  50 +++++-
 drivers/net/wan/hdlc_raw_eth.c |  49 +++++-
 drivers/net/wan/hdlc_x25.c     |  54 ++++--
 include/linux/hdlc.h           | 201 +++++----------------
 include/linux/hdlc/ioctl.h     |  33 ++++
 12 files changed, 1012 insertions(+), 777 deletions(-)
 create mode 100644 drivers/net/wan/hdlc.c
 delete mode 100644 drivers/net/wan/hdlc_generic.c

(limited to 'include')

diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig
index 54b8e492ef97..58b7efbb0750 100644
--- a/drivers/net/wan/Kconfig
+++ b/drivers/net/wan/Kconfig
@@ -154,7 +154,7 @@ config HDLC
 	  If unsure, say N.
 
 config HDLC_RAW
-	bool "Raw HDLC support"
+	tristate "Raw HDLC support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting raw HDLC over WAN connections.
@@ -162,7 +162,7 @@ config HDLC_RAW
 	  If unsure, say N.
 
 config HDLC_RAW_ETH
-	bool "Raw HDLC Ethernet device support"
+	tristate "Raw HDLC Ethernet device support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting raw HDLC Ethernet device emulation
@@ -173,7 +173,7 @@ config HDLC_RAW_ETH
 	  If unsure, say N.
 
 config HDLC_CISCO
-	bool "Cisco HDLC support"
+	tristate "Cisco HDLC support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting Cisco HDLC over WAN connections.
@@ -181,7 +181,7 @@ config HDLC_CISCO
 	  If unsure, say N.
 
 config HDLC_FR
-	bool "Frame Relay support"
+	tristate "Frame Relay support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting Frame Relay over WAN connections.
@@ -189,7 +189,7 @@ config HDLC_FR
 	  If unsure, say N.
 
 config HDLC_PPP
-	bool "Synchronous Point-to-Point Protocol (PPP) support"
+	tristate "Synchronous Point-to-Point Protocol (PPP) support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting PPP over WAN connections.
@@ -197,7 +197,7 @@ config HDLC_PPP
 	  If unsure, say N.
 
 config HDLC_X25
-	bool "X.25 protocol support"
+	tristate "X.25 protocol support"
 	depends on HDLC && (LAPB=m && HDLC=m || LAPB=y)
 	help
 	  Generic HDLC driver supporting X.25 over WAN connections.
diff --git a/drivers/net/wan/Makefile b/drivers/net/wan/Makefile
index 316ca6869d5e..83ec2c87ba3f 100644
--- a/drivers/net/wan/Makefile
+++ b/drivers/net/wan/Makefile
@@ -9,14 +9,13 @@ cyclomx-y                       := cycx_main.o
 cyclomx-$(CONFIG_CYCLOMX_X25)	+= cycx_x25.o
 cyclomx-objs			:= $(cyclomx-y)  
 
-hdlc-y				:= hdlc_generic.o
-hdlc-$(CONFIG_HDLC_RAW)		+= hdlc_raw.o
-hdlc-$(CONFIG_HDLC_RAW_ETH)	+= hdlc_raw_eth.o
-hdlc-$(CONFIG_HDLC_CISCO)	+= hdlc_cisco.o
-hdlc-$(CONFIG_HDLC_FR)		+= hdlc_fr.o
-hdlc-$(CONFIG_HDLC_PPP)		+= hdlc_ppp.o
-hdlc-$(CONFIG_HDLC_X25)		+= hdlc_x25.o
-hdlc-objs			:= $(hdlc-y)
+obj-$(CONFIG_HDLC)		+= hdlc.o
+obj-$(CONFIG_HDLC_RAW)		+= hdlc_raw.o
+obj-$(CONFIG_HDLC_RAW_ETH)	+= hdlc_raw_eth.o
+obj-$(CONFIG_HDLC_CISCO)	+= hdlc_cisco.o
+obj-$(CONFIG_HDLC_FR)		+= hdlc_fr.o
+obj-$(CONFIG_HDLC_PPP)		+= hdlc_ppp.o	syncppp.o
+obj-$(CONFIG_HDLC_X25)		+= hdlc_x25.o
 
 pc300-y				:= pc300_drv.o
 pc300-$(CONFIG_PC300_MLPPP)	+= pc300_tty.o
@@ -38,10 +37,6 @@ obj-$(CONFIG_CYCLADES_SYNC)	+= cycx_drv.o cyclomx.o
 obj-$(CONFIG_LAPBETHER)		+= lapbether.o
 obj-$(CONFIG_SBNI)		+= sbni.o
 obj-$(CONFIG_PC300)		+= pc300.o
-obj-$(CONFIG_HDLC)		+= hdlc.o
-ifeq ($(CONFIG_HDLC_PPP),y)
-  obj-$(CONFIG_HDLC)		+= syncppp.o
-endif
 obj-$(CONFIG_N2)		+= n2.o
 obj-$(CONFIG_C101)		+= c101.o
 obj-$(CONFIG_WANXL)		+= wanxl.o
diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c
new file mode 100644
index 000000000000..db354e0edbe5
--- /dev/null
+++ b/drivers/net/wan/hdlc.c
@@ -0,0 +1,368 @@
+/*
+ * Generic HDLC support routines for Linux
+ *
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Currently supported:
+ *	* raw IP-in-HDLC
+ *	* Cisco HDLC
+ *	* Frame Relay with ANSI or CCITT LMI (both user and network side)
+ *	* PPP
+ *	* X.25
+ *
+ * Use sethdlc utility to set line parameters, protocol and PVCs
+ *
+ * How does it work:
+ * - proto->open(), close(), start(), stop() calls are serialized.
+ *   The order is: open, [ start, stop ... ] close ...
+ * - proto->start() and stop() are called with spin_lock_irq held.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/poll.h>
+#include <linux/errno.h>
+#include <linux/if_arp.h>
+#include <linux/init.h>
+#include <linux/skbuff.h>
+#include <linux/pkt_sched.h>
+#include <linux/inetdevice.h>
+#include <linux/lapb.h>
+#include <linux/rtnetlink.h>
+#include <linux/notifier.h>
+#include <linux/hdlc.h>
+
+
+static const char* version = "HDLC support module revision 1.20";
+
+#undef DEBUG_LINK
+
+static struct hdlc_proto *first_proto = NULL;
+
+
+static int hdlc_change_mtu(struct net_device *dev, int new_mtu)
+{
+	if ((new_mtu < 68) || (new_mtu > HDLC_MAX_MTU))
+		return -EINVAL;
+	dev->mtu = new_mtu;
+	return 0;
+}
+
+
+
+static struct net_device_stats *hdlc_get_stats(struct net_device *dev)
+{
+	return hdlc_stats(dev);
+}
+
+
+
+static int hdlc_rcv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *p, struct net_device *orig_dev)
+{
+	struct hdlc_device_desc *desc = dev_to_desc(dev);
+	if (desc->netif_rx)
+		return desc->netif_rx(skb);
+
+	desc->stats.rx_dropped++; /* Shouldn't happen */
+	dev_kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+
+
+static inline void hdlc_proto_start(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+	if (hdlc->proto->start)
+		return hdlc->proto->start(dev);
+}
+
+
+
+static inline void hdlc_proto_stop(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+	if (hdlc->proto->stop)
+		return hdlc->proto->stop(dev);
+}
+
+
+
+static int hdlc_device_event(struct notifier_block *this, unsigned long event,
+			     void *ptr)
+{
+	struct net_device *dev = ptr;
+	hdlc_device *hdlc;
+	unsigned long flags;
+	int on;
+ 
+	if (dev->get_stats != hdlc_get_stats)
+		return NOTIFY_DONE; /* not an HDLC device */
+ 
+	if (event != NETDEV_CHANGE)
+		return NOTIFY_DONE; /* Only interrested in carrier changes */
+
+	on = netif_carrier_ok(dev);
+
+#ifdef DEBUG_LINK
+	printk(KERN_DEBUG "%s: hdlc_device_event NETDEV_CHANGE, carrier %i\n",
+	       dev->name, on);
+#endif
+
+	hdlc = dev_to_hdlc(dev);
+	spin_lock_irqsave(&hdlc->state_lock, flags);
+
+	if (hdlc->carrier == on)
+		goto carrier_exit; /* no change in DCD line level */
+
+	hdlc->carrier = on;
+
+	if (!hdlc->open)
+		goto carrier_exit;
+
+	if (hdlc->carrier) {
+		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
+		hdlc_proto_start(dev);
+	} else {
+		printk(KERN_INFO "%s: Carrier lost\n", dev->name);
+		hdlc_proto_stop(dev);
+	}
+
+carrier_exit:
+	spin_unlock_irqrestore(&hdlc->state_lock, flags);
+	return NOTIFY_DONE;
+}
+
+
+
+/* Must be called by hardware driver when HDLC device is being opened */
+int hdlc_open(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+#ifdef DEBUG_LINK
+	printk(KERN_DEBUG "%s: hdlc_open() carrier %i open %i\n", dev->name,
+	       hdlc->carrier, hdlc->open);
+#endif
+
+	if (hdlc->proto == NULL)
+		return -ENOSYS;	/* no protocol attached */
+
+	if (hdlc->proto->open) {
+		int result = hdlc->proto->open(dev);
+		if (result)
+			return result;
+	}
+
+	spin_lock_irq(&hdlc->state_lock);
+
+	if (hdlc->carrier) {
+		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
+		hdlc_proto_start(dev);
+	} else
+		printk(KERN_INFO "%s: No carrier\n", dev->name);
+
+	hdlc->open = 1;
+
+	spin_unlock_irq(&hdlc->state_lock);
+	return 0;
+}
+
+
+
+/* Must be called by hardware driver when HDLC device is being closed */
+void hdlc_close(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+#ifdef DEBUG_LINK
+	printk(KERN_DEBUG "%s: hdlc_close() carrier %i open %i\n", dev->name,
+	       hdlc->carrier, hdlc->open);
+#endif
+
+	spin_lock_irq(&hdlc->state_lock);
+
+	hdlc->open = 0;
+	if (hdlc->carrier)
+		hdlc_proto_stop(dev);
+
+	spin_unlock_irq(&hdlc->state_lock);
+
+	if (hdlc->proto->close)
+		hdlc->proto->close(dev);
+}
+
+
+
+int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct hdlc_proto *proto = first_proto;
+	int result;
+
+	if (cmd != SIOCWANDEV)
+		return -EINVAL;
+
+	if (dev_to_hdlc(dev)->proto) {
+		result = dev_to_hdlc(dev)->proto->ioctl(dev, ifr);
+		if (result != -EINVAL)
+			return result;
+	}
+
+	/* Not handled by currently attached protocol (if any) */
+
+	while (proto) {
+		if ((result = proto->ioctl(dev, ifr)) != -EINVAL)
+			return result;
+		proto = proto->next;
+	}
+	return -EINVAL;
+}
+
+void hdlc_setup(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+
+	dev->get_stats = hdlc_get_stats;
+	dev->change_mtu = hdlc_change_mtu;
+	dev->mtu = HDLC_MAX_MTU;
+
+	dev->type = ARPHRD_RAWHDLC;
+	dev->hard_header_len = 16;
+
+	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
+
+	hdlc->carrier = 1;
+	hdlc->open = 0;
+	spin_lock_init(&hdlc->state_lock);
+}
+
+struct net_device *alloc_hdlcdev(void *priv)
+{
+	struct net_device *dev;
+	dev = alloc_netdev(sizeof(struct hdlc_device_desc) +
+			   sizeof(hdlc_device), "hdlc%d", hdlc_setup);
+	if (dev)
+		dev_to_hdlc(dev)->priv = priv;
+	return dev;
+}
+
+void unregister_hdlc_device(struct net_device *dev)
+{
+	rtnl_lock();
+	unregister_netdevice(dev);
+	detach_hdlc_protocol(dev);
+	rtnl_unlock();
+}
+
+
+
+int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto,
+			 int (*rx)(struct sk_buff *skb), size_t size)
+{
+	detach_hdlc_protocol(dev);
+
+	if (!try_module_get(proto->module))
+		return -ENOSYS;
+
+	if (size)
+		if ((dev_to_hdlc(dev)->state = kmalloc(size,
+						       GFP_KERNEL)) == NULL) {
+			printk(KERN_WARNING "Memory squeeze on"
+			       " hdlc_proto_attach()\n");
+			module_put(proto->module);
+			return -ENOBUFS;
+		}
+	dev_to_hdlc(dev)->proto = proto;
+	dev_to_desc(dev)->netif_rx = rx;
+	return 0;
+}
+
+
+void detach_hdlc_protocol(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+
+	if (hdlc->proto) {
+		if (hdlc->proto->detach)
+			hdlc->proto->detach(dev);
+		module_put(hdlc->proto->module);
+		hdlc->proto = NULL;
+	}
+	kfree(hdlc->state);
+	hdlc->state = NULL;
+}
+
+
+void register_hdlc_protocol(struct hdlc_proto *proto)
+{
+	proto->next = first_proto;
+	first_proto = proto;
+}
+
+
+void unregister_hdlc_protocol(struct hdlc_proto *proto)
+{
+	struct hdlc_proto **p = &first_proto;
+	while (*p) {
+		if (*p == proto) {
+			*p = proto->next;
+			return;
+		}
+		p = &((*p)->next);
+	}
+}
+
+
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("HDLC support module");
+MODULE_LICENSE("GPL v2");
+
+EXPORT_SYMBOL(hdlc_open);
+EXPORT_SYMBOL(hdlc_close);
+EXPORT_SYMBOL(hdlc_ioctl);
+EXPORT_SYMBOL(hdlc_setup);
+EXPORT_SYMBOL(alloc_hdlcdev);
+EXPORT_SYMBOL(unregister_hdlc_device);
+EXPORT_SYMBOL(register_hdlc_protocol);
+EXPORT_SYMBOL(unregister_hdlc_protocol);
+EXPORT_SYMBOL(attach_hdlc_protocol);
+EXPORT_SYMBOL(detach_hdlc_protocol);
+
+static struct packet_type hdlc_packet_type = {
+	.type = __constant_htons(ETH_P_HDLC),
+	.func = hdlc_rcv,
+};
+
+
+static struct notifier_block hdlc_notifier = {
+        .notifier_call = hdlc_device_event,
+};
+
+
+static int __init hdlc_module_init(void)
+{
+	int result;
+
+	printk(KERN_INFO "%s\n", version);
+	if ((result = register_netdevice_notifier(&hdlc_notifier)) != 0)
+                return result;
+        dev_add_pack(&hdlc_packet_type);
+	return 0;
+}
+
+
+
+static void __exit hdlc_module_exit(void)
+{
+	dev_remove_pack(&hdlc_packet_type);
+	unregister_netdevice_notifier(&hdlc_notifier);
+}
+
+
+module_init(hdlc_module_init);
+module_exit(hdlc_module_exit);
diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index f289daba0c7b..7ec2b2f9b7ee 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * Cisco HDLC support
  *
- * Copyright (C) 2000 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 2000 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -34,17 +34,56 @@
 #define CISCO_KEEPALIVE_REQ	2	/* Cisco keepalive request */
 
 
+struct hdlc_header {
+	u8 address;
+	u8 control;
+	u16 protocol;
+}__attribute__ ((packed));
+
+
+struct cisco_packet {
+	u32 type;		/* code */
+	u32 par1;
+	u32 par2;
+	u16 rel;		/* reliability */
+	u32 time;
+}__attribute__ ((packed));
+#define	CISCO_PACKET_LEN	18
+#define	CISCO_BIG_PACKET_LEN	20
+
+
+struct cisco_state {
+	cisco_proto settings;
+
+	struct timer_list timer;
+	unsigned long last_poll;
+	int up;
+	int request_sent;
+	u32 txseq; /* TX sequence number */
+	u32 rxseq; /* RX sequence number */
+};
+
+
+static int cisco_ioctl(struct net_device *dev, struct ifreq *ifr);
+
+
+static inline struct cisco_state * state(hdlc_device *hdlc)
+{
+	return(struct cisco_state *)(hdlc->state);
+}
+
+
 static int cisco_hard_header(struct sk_buff *skb, struct net_device *dev,
 			     u16 type, void *daddr, void *saddr,
 			     unsigned int len)
 {
-	hdlc_header *data;
+	struct hdlc_header *data;
 #ifdef DEBUG_HARD_HEADER
 	printk(KERN_DEBUG "%s: cisco_hard_header called\n", dev->name);
 #endif
 
-	skb_push(skb, sizeof(hdlc_header));
-	data = (hdlc_header*)skb->data;
+	skb_push(skb, sizeof(struct hdlc_header));
+	data = (struct hdlc_header*)skb->data;
 	if (type == CISCO_KEEPALIVE)
 		data->address = CISCO_MULTICAST;
 	else
@@ -52,7 +91,7 @@ static int cisco_hard_header(struct sk_buff *skb, struct net_device *dev,
 	data->control = 0;
 	data->protocol = htons(type);
 
-	return sizeof(hdlc_header);
+	return sizeof(struct hdlc_header);
 }
 
 
@@ -61,9 +100,10 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 				 u32 par1, u32 par2)
 {
 	struct sk_buff *skb;
-	cisco_packet *data;
+	struct cisco_packet *data;
 
-	skb = dev_alloc_skb(sizeof(hdlc_header) + sizeof(cisco_packet));
+	skb = dev_alloc_skb(sizeof(struct hdlc_header) +
+			    sizeof(struct cisco_packet));
 	if (!skb) {
 		printk(KERN_WARNING
 		       "%s: Memory squeeze on cisco_keepalive_send()\n",
@@ -72,7 +112,7 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 	}
 	skb_reserve(skb, 4);
 	cisco_hard_header(skb, dev, CISCO_KEEPALIVE, NULL, NULL, 0);
-	data = (cisco_packet*)(skb->data + 4);
+	data = (struct cisco_packet*)(skb->data + 4);
 
 	data->type = htonl(type);
 	data->par1 = htonl(par1);
@@ -81,7 +121,7 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 	/* we will need do_div here if 1000 % HZ != 0 */
 	data->time = htonl((jiffies - INITIAL_JIFFIES) * (1000 / HZ));
 
-	skb_put(skb, sizeof(cisco_packet));
+	skb_put(skb, sizeof(struct cisco_packet));
 	skb->priority = TC_PRIO_CONTROL;
 	skb->dev = dev;
 	skb->nh.raw = skb->data;
@@ -93,9 +133,9 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 
 static __be16 cisco_type_trans(struct sk_buff *skb, struct net_device *dev)
 {
-	hdlc_header *data = (hdlc_header*)skb->data;
+	struct hdlc_header *data = (struct hdlc_header*)skb->data;
 
-	if (skb->len < sizeof(hdlc_header))
+	if (skb->len < sizeof(struct hdlc_header))
 		return __constant_htons(ETH_P_HDLC);
 
 	if (data->address != CISCO_MULTICAST &&
@@ -106,7 +146,7 @@ static __be16 cisco_type_trans(struct sk_buff *skb, struct net_device *dev)
 	case __constant_htons(ETH_P_IP):
 	case __constant_htons(ETH_P_IPX):
 	case __constant_htons(ETH_P_IPV6):
-		skb_pull(skb, sizeof(hdlc_header));
+		skb_pull(skb, sizeof(struct hdlc_header));
 		return data->protocol;
 	default:
 		return __constant_htons(ETH_P_HDLC);
@@ -118,12 +158,12 @@ static int cisco_rx(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	hdlc_header *data = (hdlc_header*)skb->data;
-	cisco_packet *cisco_data;
+	struct hdlc_header *data = (struct hdlc_header*)skb->data;
+	struct cisco_packet *cisco_data;
 	struct in_device *in_dev;
 	u32 addr, mask;
 
-	if (skb->len < sizeof(hdlc_header))
+	if (skb->len < sizeof(struct hdlc_header))
 		goto rx_error;
 
 	if (data->address != CISCO_MULTICAST &&
@@ -137,15 +177,17 @@ static int cisco_rx(struct sk_buff *skb)
 		return NET_RX_SUCCESS;
 
 	case CISCO_KEEPALIVE:
-		if (skb->len != sizeof(hdlc_header) + CISCO_PACKET_LEN &&
-		    skb->len != sizeof(hdlc_header) + CISCO_BIG_PACKET_LEN) {
-			printk(KERN_INFO "%s: Invalid length of Cisco "
-			       "control packet (%d bytes)\n",
-			       dev->name, skb->len);
+		if ((skb->len != sizeof(struct hdlc_header) +
+		     CISCO_PACKET_LEN) &&
+		    (skb->len != sizeof(struct hdlc_header) +
+		     CISCO_BIG_PACKET_LEN)) {
+			printk(KERN_INFO "%s: Invalid length of Cisco control"
+			       " packet (%d bytes)\n", dev->name, skb->len);
 			goto rx_error;
 		}
 
-		cisco_data = (cisco_packet*)(skb->data + sizeof(hdlc_header));
+		cisco_data = (struct cisco_packet*)(skb->data + sizeof
+						    (struct hdlc_header));
 
 		switch(ntohl (cisco_data->type)) {
 		case CISCO_ADDR_REQ: /* Stolen from syncppp.c :-) */
@@ -178,11 +220,11 @@ static int cisco_rx(struct sk_buff *skb)
 			goto rx_error;
 
 		case CISCO_KEEPALIVE_REQ:
-			hdlc->state.cisco.rxseq = ntohl(cisco_data->par1);
-			if (hdlc->state.cisco.request_sent &&
-			    ntohl(cisco_data->par2)==hdlc->state.cisco.txseq) {
-				hdlc->state.cisco.last_poll = jiffies;
-				if (!hdlc->state.cisco.up) {
+			state(hdlc)->rxseq = ntohl(cisco_data->par1);
+			if (state(hdlc)->request_sent &&
+			    ntohl(cisco_data->par2) == state(hdlc)->txseq) {
+				state(hdlc)->last_poll = jiffies;
+				if (!state(hdlc)->up) {
 					u32 sec, min, hrs, days;
 					sec = ntohl(cisco_data->time) / 1000;
 					min = sec / 60; sec -= min * 60;
@@ -193,7 +235,7 @@ static int cisco_rx(struct sk_buff *skb)
 					       dev->name, days, hrs,
 					       min, sec);
 					netif_dormant_off(dev);
-					hdlc->state.cisco.up = 1;
+					state(hdlc)->up = 1;
 				}
 			}
 
@@ -208,7 +250,7 @@ static int cisco_rx(struct sk_buff *skb)
 	return NET_RX_DROP;
 
  rx_error:
-	hdlc->stats.rx_errors++; /* Mark error */
+	dev_to_desc(dev)->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -220,23 +262,22 @@ static void cisco_timer(unsigned long arg)
 	struct net_device *dev = (struct net_device *)arg;
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 
-	if (hdlc->state.cisco.up &&
-	    time_after(jiffies, hdlc->state.cisco.last_poll +
-		       hdlc->state.cisco.settings.timeout * HZ)) {
-		hdlc->state.cisco.up = 0;
+	if (state(hdlc)->up &&
+	    time_after(jiffies, state(hdlc)->last_poll +
+		       state(hdlc)->settings.timeout * HZ)) {
+		state(hdlc)->up = 0;
 		printk(KERN_INFO "%s: Link down\n", dev->name);
 		netif_dormant_on(dev);
 	}
 
-	cisco_keepalive_send(dev, CISCO_KEEPALIVE_REQ,
-			     ++hdlc->state.cisco.txseq,
-			     hdlc->state.cisco.rxseq);
-	hdlc->state.cisco.request_sent = 1;
-	hdlc->state.cisco.timer.expires = jiffies +
-		hdlc->state.cisco.settings.interval * HZ;
-	hdlc->state.cisco.timer.function = cisco_timer;
-	hdlc->state.cisco.timer.data = arg;
-	add_timer(&hdlc->state.cisco.timer);
+	cisco_keepalive_send(dev, CISCO_KEEPALIVE_REQ, ++state(hdlc)->txseq,
+			     state(hdlc)->rxseq);
+	state(hdlc)->request_sent = 1;
+	state(hdlc)->timer.expires = jiffies +
+		state(hdlc)->settings.interval * HZ;
+	state(hdlc)->timer.function = cisco_timer;
+	state(hdlc)->timer.data = arg;
+	add_timer(&state(hdlc)->timer);
 }
 
 
@@ -244,15 +285,15 @@ static void cisco_timer(unsigned long arg)
 static void cisco_start(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	hdlc->state.cisco.up = 0;
-	hdlc->state.cisco.request_sent = 0;
-	hdlc->state.cisco.txseq = hdlc->state.cisco.rxseq = 0;
-
-	init_timer(&hdlc->state.cisco.timer);
-	hdlc->state.cisco.timer.expires = jiffies + HZ; /*First poll after 1s*/
-	hdlc->state.cisco.timer.function = cisco_timer;
-	hdlc->state.cisco.timer.data = (unsigned long)dev;
-	add_timer(&hdlc->state.cisco.timer);
+	state(hdlc)->up = 0;
+	state(hdlc)->request_sent = 0;
+	state(hdlc)->txseq = state(hdlc)->rxseq = 0;
+
+	init_timer(&state(hdlc)->timer);
+	state(hdlc)->timer.expires = jiffies + HZ; /*First poll after 1s*/
+	state(hdlc)->timer.function = cisco_timer;
+	state(hdlc)->timer.data = (unsigned long)dev;
+	add_timer(&state(hdlc)->timer);
 }
 
 
@@ -260,15 +301,24 @@ static void cisco_start(struct net_device *dev)
 static void cisco_stop(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	del_timer_sync(&hdlc->state.cisco.timer);
+	del_timer_sync(&state(hdlc)->timer);
 	netif_dormant_on(dev);
-	hdlc->state.cisco.up = 0;
-	hdlc->state.cisco.request_sent = 0;
+	state(hdlc)->up = 0;
+	state(hdlc)->request_sent = 0;
 }
 
 
 
-int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.start		= cisco_start,
+	.stop		= cisco_stop,
+	.type_trans	= cisco_type_trans,
+	.ioctl		= cisco_ioctl,
+	.module		= THIS_MODULE,
+};
+ 
+ 
+static int cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	cisco_proto __user *cisco_s = ifr->ifr_settings.ifs_ifsu.cisco;
 	const size_t size = sizeof(cisco_proto);
@@ -278,12 +328,14 @@ int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_CISCO;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(cisco_s, &hdlc->state.cisco.settings, size))
+		if (copy_to_user(cisco_s, &state(hdlc)->settings, size))
 			return -EFAULT;
 		return 0;
 
@@ -302,19 +354,15 @@ int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 			return -EINVAL;
 
 		result=hdlc->attach(dev, ENCODING_NRZ,PARITY_CRC16_PR1_CCITT);
-
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memcpy(&hdlc->state.cisco.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
+		result = attach_hdlc_protocol(dev, &proto, cisco_rx,
+					      sizeof(struct cisco_state));
+		if (result)
+			return result;
 
-		hdlc->proto.start = cisco_start;
-		hdlc->proto.stop = cisco_stop;
-		hdlc->proto.netif_rx = cisco_rx;
-		hdlc->proto.type_trans = cisco_type_trans;
-		hdlc->proto.id = IF_PROTO_CISCO;
+		memcpy(&state(hdlc)->settings, &new_settings, size);
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = cisco_hard_header;
 		dev->hard_header_cache = NULL;
@@ -327,3 +375,25 @@ int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Cisco HDLC protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c
index 7bb737bbdeb9..b45ab680d2d6 100644
--- a/drivers/net/wan/hdlc_fr.c
+++ b/drivers/net/wan/hdlc_fr.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * Frame Relay support
  *
- * Copyright (C) 1999 - 2005 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -52,6 +52,8 @@
 #undef DEBUG_PKT
 #undef DEBUG_ECN
 #undef DEBUG_LINK
+#undef DEBUG_PROTO
+#undef DEBUG_PVC
 
 #define FR_UI			0x03
 #define FR_PAD			0x00
@@ -115,13 +117,53 @@ typedef struct {
 }__attribute__ ((packed)) fr_hdr;
 
 
+typedef struct pvc_device_struct {
+	struct net_device *frad;
+	struct net_device *main;
+	struct net_device *ether;	/* bridged Ethernet interface	*/
+	struct pvc_device_struct *next;	/* Sorted in ascending DLCI order */
+	int dlci;
+	int open_count;
+
+	struct {
+		unsigned int new: 1;
+		unsigned int active: 1;
+		unsigned int exist: 1;
+		unsigned int deleted: 1;
+		unsigned int fecn: 1;
+		unsigned int becn: 1;
+		unsigned int bandwidth;	/* Cisco LMI reporting only */
+	}state;
+}pvc_device;
+
+
+struct frad_state {
+	fr_proto settings;
+	pvc_device *first_pvc;
+	int dce_pvc_count;
+
+	struct timer_list timer;
+	unsigned long last_poll;
+	int reliable;
+	int dce_changed;
+	int request;
+	int fullrep_sent;
+	u32 last_errors; /* last errors bit list */
+	u8 n391cnt;
+	u8 txseq; /* TX sequence number */
+	u8 rxseq; /* RX sequence number */
+};
+
+
+static int fr_ioctl(struct net_device *dev, struct ifreq *ifr);
+
+
 static inline u16 q922_to_dlci(u8 *hdr)
 {
 	return ((hdr[0] & 0xFC) << 2) | ((hdr[1] & 0xF0) >> 4);
 }
 
 
-
 static inline void dlci_to_q922(u8 *hdr, u16 dlci)
 {
 	hdr[0] = (dlci >> 2) & 0xFC;
@@ -129,10 +171,21 @@ static inline void dlci_to_q922(u8 *hdr, u16 dlci)
 }
 
 
+static inline struct frad_state * state(hdlc_device *hdlc)
+{
+	return(struct frad_state *)(hdlc->state);
+}
+
+
+static __inline__ pvc_device* dev_to_pvc(struct net_device *dev)
+{
+	return dev->priv;
+}
+
 
 static inline pvc_device* find_pvc(hdlc_device *hdlc, u16 dlci)
 {
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
+	pvc_device *pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {
 		if (pvc->dlci == dlci)
@@ -146,10 +199,10 @@ static inline pvc_device* find_pvc(hdlc_device *hdlc, u16 dlci)
 }
 
 
-static inline pvc_device* add_pvc(struct net_device *dev, u16 dlci)
+static pvc_device* add_pvc(struct net_device *dev, u16 dlci)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	pvc_device *pvc, **pvc_p = &hdlc->state.fr.first_pvc;
+	pvc_device *pvc, **pvc_p = &state(hdlc)->first_pvc;
 
 	while (*pvc_p) {
 		if ((*pvc_p)->dlci == dlci)
@@ -160,12 +213,15 @@ static inline pvc_device* add_pvc(struct net_device *dev, u16 dlci)
 	}
 
 	pvc = kmalloc(sizeof(pvc_device), GFP_ATOMIC);
+#ifdef DEBUG_PVC
+	printk(KERN_DEBUG "add_pvc: allocated pvc %p, frad %p\n", pvc, dev);
+#endif
 	if (!pvc)
 		return NULL;
 
 	memset(pvc, 0, sizeof(pvc_device));
 	pvc->dlci = dlci;
-	pvc->master = dev;
+	pvc->frad = dev;
 	pvc->next = *pvc_p;	/* Put it in the chain */
 	*pvc_p = pvc;
 	return pvc;
@@ -174,7 +230,7 @@ static inline pvc_device* add_pvc(struct net_device *dev, u16 dlci)
 
 static inline int pvc_is_used(pvc_device *pvc)
 {
-	return pvc->main != NULL || pvc->ether != NULL;
+	return pvc->main || pvc->ether;
 }
 
 
@@ -200,11 +256,14 @@ static inline void pvc_carrier(int on, pvc_device *pvc)
 
 static inline void delete_unused_pvcs(hdlc_device *hdlc)
 {
-	pvc_device **pvc_p = &hdlc->state.fr.first_pvc;
+	pvc_device **pvc_p = &state(hdlc)->first_pvc;
 
 	while (*pvc_p) {
 		if (!pvc_is_used(*pvc_p)) {
 			pvc_device *pvc = *pvc_p;
+#ifdef DEBUG_PVC
+			printk(KERN_DEBUG "freeing unused pvc: %p\n", pvc);
+#endif
 			*pvc_p = pvc->next;
 			kfree(pvc);
 			continue;
@@ -295,16 +354,16 @@ static int pvc_open(struct net_device *dev)
 {
 	pvc_device *pvc = dev_to_pvc(dev);
 
-	if ((pvc->master->flags & IFF_UP) == 0)
-		return -EIO;  /* Master must be UP in order to activate PVC */
+	if ((pvc->frad->flags & IFF_UP) == 0)
+		return -EIO;  /* Frad must be UP in order to activate PVC */
 
 	if (pvc->open_count++ == 0) {
-		hdlc_device *hdlc = dev_to_hdlc(pvc->master);
-		if (hdlc->state.fr.settings.lmi == LMI_NONE)
-			pvc->state.active = netif_carrier_ok(pvc->master);
+		hdlc_device *hdlc = dev_to_hdlc(pvc->frad);
+		if (state(hdlc)->settings.lmi == LMI_NONE)
+			pvc->state.active = netif_carrier_ok(pvc->frad);
 
 		pvc_carrier(pvc->state.active, pvc);
-		hdlc->state.fr.dce_changed = 1;
+		state(hdlc)->dce_changed = 1;
 	}
 	return 0;
 }
@@ -316,12 +375,12 @@ static int pvc_close(struct net_device *dev)
 	pvc_device *pvc = dev_to_pvc(dev);
 
 	if (--pvc->open_count == 0) {
-		hdlc_device *hdlc = dev_to_hdlc(pvc->master);
-		if (hdlc->state.fr.settings.lmi == LMI_NONE)
+		hdlc_device *hdlc = dev_to_hdlc(pvc->frad);
+		if (state(hdlc)->settings.lmi == LMI_NONE)
 			pvc->state.active = 0;
 
-		if (hdlc->state.fr.settings.dce) {
-			hdlc->state.fr.dce_changed = 1;
+		if (state(hdlc)->settings.dce) {
+			state(hdlc)->dce_changed = 1;
 			pvc->state.active = 0;
 		}
 	}
@@ -348,7 +407,7 @@ static int pvc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		}
 
 		info.dlci = pvc->dlci;
-		memcpy(info.master, pvc->master->name, IFNAMSIZ);
+		memcpy(info.master, pvc->frad->name, IFNAMSIZ);
 		if (copy_to_user(ifr->ifr_settings.ifs_ifsu.fr_pvc_info,
 				 &info, sizeof(info)))
 			return -EFAULT;
@@ -361,7 +420,7 @@ static int pvc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 static inline struct net_device_stats *pvc_get_stats(struct net_device *dev)
 {
-	return netdev_priv(dev);
+	return &dev_to_desc(dev)->stats;
 }
 
 
@@ -393,7 +452,7 @@ static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 			stats->tx_packets++;
 			if (pvc->state.fecn) /* TX Congestion counter */
 				stats->tx_compressed++;
-			skb->dev = pvc->master;
+			skb->dev = pvc->frad;
 			dev_queue_xmit(skb);
 			return 0;
 		}
@@ -419,7 +478,7 @@ static int pvc_change_mtu(struct net_device *dev, int new_mtu)
 static inline void fr_log_dlci_active(pvc_device *pvc)
 {
 	printk(KERN_INFO "%s: DLCI %d [%s%s%s]%s %s\n",
-	       pvc->master->name,
+	       pvc->frad->name,
 	       pvc->dlci,
 	       pvc->main ? pvc->main->name : "",
 	       pvc->main && pvc->ether ? " " : "",
@@ -438,21 +497,20 @@ static inline u8 fr_lmi_nextseq(u8 x)
 }
 
 
-
 static void fr_lmi_send(struct net_device *dev, int fullrep)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	struct sk_buff *skb;
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
-	int lmi = hdlc->state.fr.settings.lmi;
-	int dce = hdlc->state.fr.settings.dce;
+	pvc_device *pvc = state(hdlc)->first_pvc;
+	int lmi = state(hdlc)->settings.lmi;
+	int dce = state(hdlc)->settings.dce;
 	int len = lmi == LMI_ANSI ? LMI_ANSI_LENGTH : LMI_CCITT_CISCO_LENGTH;
 	int stat_len = (lmi == LMI_CISCO) ? 6 : 3;
 	u8 *data;
 	int i = 0;
 
 	if (dce && fullrep) {
-		len += hdlc->state.fr.dce_pvc_count * (2 + stat_len);
+		len += state(hdlc)->dce_pvc_count * (2 + stat_len);
 		if (len > HDLC_MAX_MRU) {
 			printk(KERN_WARNING "%s: Too many PVCs while sending "
 			       "LMI full report\n", dev->name);
@@ -486,8 +544,9 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)
 	data[i++] = fullrep ? LMI_FULLREP : LMI_INTEGRITY;
 	data[i++] = lmi == LMI_CCITT ? LMI_CCITT_ALIVE : LMI_ANSI_CISCO_ALIVE;
 	data[i++] = LMI_INTEG_LEN;
-	data[i++] = hdlc->state.fr.txseq =fr_lmi_nextseq(hdlc->state.fr.txseq);
-	data[i++] = hdlc->state.fr.rxseq;
+	data[i++] = state(hdlc)->txseq =
+		fr_lmi_nextseq(state(hdlc)->txseq);
+	data[i++] = state(hdlc)->rxseq;
 
 	if (dce && fullrep) {
 		while (pvc) {
@@ -496,7 +555,7 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)
 			data[i++] = stat_len;
 
 			/* LMI start/restart */
-			if (hdlc->state.fr.reliable && !pvc->state.exist) {
+			if (state(hdlc)->reliable && !pvc->state.exist) {
 				pvc->state.exist = pvc->state.new = 1;
 				fr_log_dlci_active(pvc);
 			}
@@ -541,15 +600,15 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)
 static void fr_set_link_state(int reliable, struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
+	pvc_device *pvc = state(hdlc)->first_pvc;
 
-	hdlc->state.fr.reliable = reliable;
+	state(hdlc)->reliable = reliable;
 	if (reliable) {
 		netif_dormant_off(dev);
-		hdlc->state.fr.n391cnt = 0; /* Request full status */
-		hdlc->state.fr.dce_changed = 1;
+		state(hdlc)->n391cnt = 0; /* Request full status */
+		state(hdlc)->dce_changed = 1;
 
-		if (hdlc->state.fr.settings.lmi == LMI_NONE) {
+		if (state(hdlc)->settings.lmi == LMI_NONE) {
 			while (pvc) {	/* Activate all PVCs */
 				pvc_carrier(1, pvc);
 				pvc->state.exist = pvc->state.active = 1;
@@ -563,7 +622,7 @@ static void fr_set_link_state(int reliable, struct net_device *dev)
 			pvc_carrier(0, pvc);
 			pvc->state.exist = pvc->state.active = 0;
 			pvc->state.new = 0;
-			if (!hdlc->state.fr.settings.dce)
+			if (!state(hdlc)->settings.dce)
 				pvc->state.bandwidth = 0;
 			pvc = pvc->next;
 		}
@@ -571,7 +630,6 @@ static void fr_set_link_state(int reliable, struct net_device *dev)
 }
 
 
-
 static void fr_timer(unsigned long arg)
 {
 	struct net_device *dev = (struct net_device *)arg;
@@ -579,62 +637,61 @@ static void fr_timer(unsigned long arg)
 	int i, cnt = 0, reliable;
 	u32 list;
 
-	if (hdlc->state.fr.settings.dce) {
-		reliable = hdlc->state.fr.request &&
-			time_before(jiffies, hdlc->state.fr.last_poll +
-				    hdlc->state.fr.settings.t392 * HZ);
-		hdlc->state.fr.request = 0;
+	if (state(hdlc)->settings.dce) {
+		reliable = state(hdlc)->request &&
+			time_before(jiffies, state(hdlc)->last_poll +
+				    state(hdlc)->settings.t392 * HZ);
+		state(hdlc)->request = 0;
 	} else {
-		hdlc->state.fr.last_errors <<= 1; /* Shift the list */
-		if (hdlc->state.fr.request) {
-			if (hdlc->state.fr.reliable)
+		state(hdlc)->last_errors <<= 1; /* Shift the list */
+		if (state(hdlc)->request) {
+			if (state(hdlc)->reliable)
 				printk(KERN_INFO "%s: No LMI status reply "
 				       "received\n", dev->name);
-			hdlc->state.fr.last_errors |= 1;
+			state(hdlc)->last_errors |= 1;
 		}
 
-		list = hdlc->state.fr.last_errors;
-		for (i = 0; i < hdlc->state.fr.settings.n393; i++, list >>= 1)
+		list = state(hdlc)->last_errors;
+		for (i = 0; i < state(hdlc)->settings.n393; i++, list >>= 1)
 			cnt += (list & 1);	/* errors count */
 
-		reliable = (cnt < hdlc->state.fr.settings.n392);
+		reliable = (cnt < state(hdlc)->settings.n392);
 	}
 
-	if (hdlc->state.fr.reliable != reliable) {
+	if (state(hdlc)->reliable != reliable) {
 		printk(KERN_INFO "%s: Link %sreliable\n", dev->name,
 		       reliable ? "" : "un");
 		fr_set_link_state(reliable, dev);
 	}
 
-	if (hdlc->state.fr.settings.dce)
-		hdlc->state.fr.timer.expires = jiffies +
-			hdlc->state.fr.settings.t392 * HZ;
+	if (state(hdlc)->settings.dce)
+		state(hdlc)->timer.expires = jiffies +
+			state(hdlc)->settings.t392 * HZ;
 	else {
-		if (hdlc->state.fr.n391cnt)
-			hdlc->state.fr.n391cnt--;
+		if (state(hdlc)->n391cnt)
+			state(hdlc)->n391cnt--;
 
-		fr_lmi_send(dev, hdlc->state.fr.n391cnt == 0);
+		fr_lmi_send(dev, state(hdlc)->n391cnt == 0);
 
-		hdlc->state.fr.last_poll = jiffies;
-		hdlc->state.fr.request = 1;
-		hdlc->state.fr.timer.expires = jiffies +
-			hdlc->state.fr.settings.t391 * HZ;
+		state(hdlc)->last_poll = jiffies;
+		state(hdlc)->request = 1;
+		state(hdlc)->timer.expires = jiffies +
+			state(hdlc)->settings.t391 * HZ;
 	}
 
-	hdlc->state.fr.timer.function = fr_timer;
-	hdlc->state.fr.timer.data = arg;
-	add_timer(&hdlc->state.fr.timer);
+	state(hdlc)->timer.function = fr_timer;
+	state(hdlc)->timer.data = arg;
+	add_timer(&state(hdlc)->timer);
 }
 
 
-
 static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	pvc_device *pvc;
 	u8 rxseq, txseq;
-	int lmi = hdlc->state.fr.settings.lmi;
-	int dce = hdlc->state.fr.settings.dce;
+	int lmi = state(hdlc)->settings.lmi;
+	int dce = state(hdlc)->settings.dce;
 	int stat_len = (lmi == LMI_CISCO) ? 6 : 3, reptype, error, no_ram, i;
 
 	if (skb->len < (lmi == LMI_ANSI ? LMI_ANSI_LENGTH :
@@ -645,8 +702,8 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 
 	if (skb->data[3] != (lmi == LMI_CISCO ? NLPID_CISCO_LMI :
 			     NLPID_CCITT_ANSI_LMI)) {
-		printk(KERN_INFO "%s: Received non-LMI frame with LMI"
-		       " DLCI\n", dev->name);
+		printk(KERN_INFO "%s: Received non-LMI frame with LMI DLCI\n",
+		       dev->name);
 		return 1;
 	}
 
@@ -706,53 +763,53 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 	}
 	i++;
 
-	hdlc->state.fr.rxseq = skb->data[i++]; /* TX sequence from peer */
+	state(hdlc)->rxseq = skb->data[i++]; /* TX sequence from peer */
 	rxseq = skb->data[i++];	/* Should confirm our sequence */
 
-	txseq = hdlc->state.fr.txseq;
+	txseq = state(hdlc)->txseq;
 
 	if (dce)
-		hdlc->state.fr.last_poll = jiffies;
+		state(hdlc)->last_poll = jiffies;
 
 	error = 0;
-	if (!hdlc->state.fr.reliable)
+	if (!state(hdlc)->reliable)
 		error = 1;
 
-	if (rxseq == 0 || rxseq != txseq) {
-		hdlc->state.fr.n391cnt = 0; /* Ask for full report next time */
+	if (rxseq == 0 || rxseq != txseq) { /* Ask for full report next time */
+		state(hdlc)->n391cnt = 0;
 		error = 1;
 	}
 
 	if (dce) {
-		if (hdlc->state.fr.fullrep_sent && !error) {
+		if (state(hdlc)->fullrep_sent && !error) {
 /* Stop sending full report - the last one has been confirmed by DTE */
-			hdlc->state.fr.fullrep_sent = 0;
-			pvc = hdlc->state.fr.first_pvc;
+			state(hdlc)->fullrep_sent = 0;
+			pvc = state(hdlc)->first_pvc;
 			while (pvc) {
 				if (pvc->state.new) {
 					pvc->state.new = 0;
 
 /* Tell DTE that new PVC is now active */
-					hdlc->state.fr.dce_changed = 1;
+					state(hdlc)->dce_changed = 1;
 				}
 				pvc = pvc->next;
 			}
 		}
 
-		if (hdlc->state.fr.dce_changed) {
+		if (state(hdlc)->dce_changed) {
 			reptype = LMI_FULLREP;
-			hdlc->state.fr.fullrep_sent = 1;
-			hdlc->state.fr.dce_changed = 0;
+			state(hdlc)->fullrep_sent = 1;
+			state(hdlc)->dce_changed = 0;
 		}
 
-		hdlc->state.fr.request = 1; /* got request */
+		state(hdlc)->request = 1; /* got request */
 		fr_lmi_send(dev, reptype == LMI_FULLREP ? 1 : 0);
 		return 0;
 	}
 
 	/* DTE */
 
-	hdlc->state.fr.request = 0; /* got response, no request pending */
+	state(hdlc)->request = 0; /* got response, no request pending */
 
 	if (error)
 		return 0;
@@ -760,7 +817,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 	if (reptype != LMI_FULLREP)
 		return 0;
 
-	pvc = hdlc->state.fr.first_pvc;
+	pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {
 		pvc->state.deleted = 1;
@@ -827,7 +884,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 		i += stat_len;
 	}
 
-	pvc = hdlc->state.fr.first_pvc;
+	pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {
 		if (pvc->state.deleted && pvc->state.exist) {
@@ -841,17 +898,16 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 	}
 
 	/* Next full report after N391 polls */
-	hdlc->state.fr.n391cnt = hdlc->state.fr.settings.n391;
+	state(hdlc)->n391cnt = state(hdlc)->settings.n391;
 
 	return 0;
 }
 
 
-
 static int fr_rx(struct sk_buff *skb)
 {
-	struct net_device *ndev = skb->dev;
-	hdlc_device *hdlc = dev_to_hdlc(ndev);
+	struct net_device *frad = skb->dev;
+	hdlc_device *hdlc = dev_to_hdlc(frad);
 	fr_hdr *fh = (fr_hdr*)skb->data;
 	u8 *data = skb->data;
 	u16 dlci;
@@ -864,11 +920,11 @@ static int fr_rx(struct sk_buff *skb)
 	dlci = q922_to_dlci(skb->data);
 
 	if ((dlci == LMI_CCITT_ANSI_DLCI &&
-	     (hdlc->state.fr.settings.lmi == LMI_ANSI ||
-	      hdlc->state.fr.settings.lmi == LMI_CCITT)) ||
+	     (state(hdlc)->settings.lmi == LMI_ANSI ||
+	      state(hdlc)->settings.lmi == LMI_CCITT)) ||
 	    (dlci == LMI_CISCO_DLCI &&
-	     hdlc->state.fr.settings.lmi == LMI_CISCO)) {
-		if (fr_lmi_recv(ndev, skb))
+	     state(hdlc)->settings.lmi == LMI_CISCO)) {
+		if (fr_lmi_recv(frad, skb))
 			goto rx_error;
 		dev_kfree_skb_any(skb);
 		return NET_RX_SUCCESS;
@@ -878,7 +934,7 @@ static int fr_rx(struct sk_buff *skb)
 	if (!pvc) {
 #ifdef DEBUG_PKT
 		printk(KERN_INFO "%s: No PVC for received frame's DLCI %d\n",
-		       ndev->name, dlci);
+		       frad->name, dlci);
 #endif
 		dev_kfree_skb_any(skb);
 		return NET_RX_DROP;
@@ -886,7 +942,7 @@ static int fr_rx(struct sk_buff *skb)
 
 	if (pvc->state.fecn != fh->fecn) {
 #ifdef DEBUG_ECN
-		printk(KERN_DEBUG "%s: DLCI %d FECN O%s\n", ndev->name,
+		printk(KERN_DEBUG "%s: DLCI %d FECN O%s\n", frad->name,
 		       dlci, fh->fecn ? "N" : "FF");
 #endif
 		pvc->state.fecn ^= 1;
@@ -894,7 +950,7 @@ static int fr_rx(struct sk_buff *skb)
 
 	if (pvc->state.becn != fh->becn) {
 #ifdef DEBUG_ECN
-		printk(KERN_DEBUG "%s: DLCI %d BECN O%s\n", ndev->name,
+		printk(KERN_DEBUG "%s: DLCI %d BECN O%s\n", frad->name,
 		       dlci, fh->becn ? "N" : "FF");
 #endif
 		pvc->state.becn ^= 1;
@@ -902,7 +958,7 @@ static int fr_rx(struct sk_buff *skb)
 
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		hdlc->stats.rx_dropped++;
+		dev_to_desc(frad)->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
@@ -938,13 +994,13 @@ static int fr_rx(struct sk_buff *skb)
 
 		default:
 			printk(KERN_INFO "%s: Unsupported protocol, OUI=%x "
-			       "PID=%x\n", ndev->name, oui, pid);
+			       "PID=%x\n", frad->name, oui, pid);
 			dev_kfree_skb_any(skb);
 			return NET_RX_DROP;
 		}
 	} else {
 		printk(KERN_INFO "%s: Unsupported protocol, NLPID=%x "
-		       "length = %i\n", ndev->name, data[3], skb->len);
+		       "length = %i\n", frad->name, data[3], skb->len);
 		dev_kfree_skb_any(skb);
 		return NET_RX_DROP;
 	}
@@ -964,7 +1020,7 @@ static int fr_rx(struct sk_buff *skb)
 	}
 
  rx_error:
-	hdlc->stats.rx_errors++; /* Mark error */
+	dev_to_desc(frad)->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -977,44 +1033,42 @@ static void fr_start(struct net_device *dev)
 #ifdef DEBUG_LINK
 	printk(KERN_DEBUG "fr_start\n");
 #endif
-	if (hdlc->state.fr.settings.lmi != LMI_NONE) {
-		hdlc->state.fr.reliable = 0;
-		hdlc->state.fr.dce_changed = 1;
-		hdlc->state.fr.request = 0;
-		hdlc->state.fr.fullrep_sent = 0;
-		hdlc->state.fr.last_errors = 0xFFFFFFFF;
-		hdlc->state.fr.n391cnt = 0;
-		hdlc->state.fr.txseq = hdlc->state.fr.rxseq = 0;
-
-		init_timer(&hdlc->state.fr.timer);
+	if (state(hdlc)->settings.lmi != LMI_NONE) {
+		state(hdlc)->reliable = 0;
+		state(hdlc)->dce_changed = 1;
+		state(hdlc)->request = 0;
+		state(hdlc)->fullrep_sent = 0;
+		state(hdlc)->last_errors = 0xFFFFFFFF;
+		state(hdlc)->n391cnt = 0;
+		state(hdlc)->txseq = state(hdlc)->rxseq = 0;
+
+		init_timer(&state(hdlc)->timer);
 		/* First poll after 1 s */
-		hdlc->state.fr.timer.expires = jiffies + HZ;
-		hdlc->state.fr.timer.function = fr_timer;
-		hdlc->state.fr.timer.data = (unsigned long)dev;
-		add_timer(&hdlc->state.fr.timer);
+		state(hdlc)->timer.expires = jiffies + HZ;
+		state(hdlc)->timer.function = fr_timer;
+		state(hdlc)->timer.data = (unsigned long)dev;
+		add_timer(&state(hdlc)->timer);
 	} else
 		fr_set_link_state(1, dev);
 }
 
 
-
 static void fr_stop(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 #ifdef DEBUG_LINK
 	printk(KERN_DEBUG "fr_stop\n");
 #endif
-	if (hdlc->state.fr.settings.lmi != LMI_NONE)
-		del_timer_sync(&hdlc->state.fr.timer);
+	if (state(hdlc)->settings.lmi != LMI_NONE)
+		del_timer_sync(&state(hdlc)->timer);
 	fr_set_link_state(0, dev);
 }
 
 
-
 static void fr_close(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
+	pvc_device *pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {		/* Shutdown all PVCs for this FRAD */
 		if (pvc->main)
@@ -1025,7 +1079,8 @@ static void fr_close(struct net_device *dev)
 	}
 }
 
-static void dlci_setup(struct net_device *dev)
+
+static void pvc_setup(struct net_device *dev)
 {
 	dev->type = ARPHRD_DLCI;
 	dev->flags = IFF_POINTOPOINT;
@@ -1033,9 +1088,9 @@ static void dlci_setup(struct net_device *dev)
 	dev->addr_len = 2;
 }
 
-static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
+static int fr_add_pvc(struct net_device *frad, unsigned int dlci, int type)
 {
-	hdlc_device *hdlc = dev_to_hdlc(master);
+	hdlc_device *hdlc = dev_to_hdlc(frad);
 	pvc_device *pvc = NULL;
 	struct net_device *dev;
 	int result, used;
@@ -1044,9 +1099,9 @@ static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
 	if (type == ARPHRD_ETHER)
 		prefix = "pvceth%d";
 
-	if ((pvc = add_pvc(master, dlci)) == NULL) {
+	if ((pvc = add_pvc(frad, dlci)) == NULL) {
 		printk(KERN_WARNING "%s: Memory squeeze on fr_add_pvc()\n",
-		       master->name);
+		       frad->name);
 		return -ENOBUFS;
 	}
 
@@ -1060,11 +1115,11 @@ static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
 				   "pvceth%d", ether_setup);
 	else
 		dev = alloc_netdev(sizeof(struct net_device_stats),
-				   "pvc%d", dlci_setup);
+				   "pvc%d", pvc_setup);
 
 	if (!dev) {
 		printk(KERN_WARNING "%s: Memory squeeze on fr_pvc()\n",
-		       master->name);
+		       frad->name);
 		delete_unused_pvcs(hdlc);
 		return -ENOBUFS;
 	}
@@ -1102,8 +1157,8 @@ static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
 	dev->destructor = free_netdev;
 	*get_dev_p(pvc, type) = dev;
 	if (!used) {
-		hdlc->state.fr.dce_changed = 1;
-		hdlc->state.fr.dce_pvc_count++;
+		state(hdlc)->dce_changed = 1;
+		state(hdlc)->dce_pvc_count++;
 	}
 	return 0;
 }
@@ -1128,8 +1183,8 @@ static int fr_del_pvc(hdlc_device *hdlc, unsigned int dlci, int type)
 	*get_dev_p(pvc, type) = NULL;
 
 	if (!pvc_is_used(pvc)) {
-		hdlc->state.fr.dce_pvc_count--;
-		hdlc->state.fr.dce_changed = 1;
+		state(hdlc)->dce_pvc_count--;
+		state(hdlc)->dce_changed = 1;
 	}
 	delete_unused_pvcs(hdlc);
 	return 0;
@@ -1137,14 +1192,13 @@ static int fr_del_pvc(hdlc_device *hdlc, unsigned int dlci, int type)
 
 
 
-static void fr_destroy(hdlc_device *hdlc)
+static void fr_destroy(struct net_device *frad)
 {
-	pvc_device *pvc;
-
-	pvc = hdlc->state.fr.first_pvc;
-	hdlc->state.fr.first_pvc = NULL; /* All PVCs destroyed */
-	hdlc->state.fr.dce_pvc_count = 0;
-	hdlc->state.fr.dce_changed = 1;
+	hdlc_device *hdlc = dev_to_hdlc(frad);
+	pvc_device *pvc = state(hdlc)->first_pvc;
+	state(hdlc)->first_pvc = NULL; /* All PVCs destroyed */
+	state(hdlc)->dce_pvc_count = 0;
+	state(hdlc)->dce_changed = 1;
 
 	while (pvc) {
 		pvc_device *next = pvc->next;
@@ -1161,8 +1215,17 @@ static void fr_destroy(hdlc_device *hdlc)
 }
 
 
+static struct hdlc_proto proto = {
+	.close		= fr_close,
+	.start		= fr_start,
+	.stop		= fr_stop,
+	.detach		= fr_destroy,
+	.ioctl		= fr_ioctl,
+	.module		= THIS_MODULE,
+};
+
 
-int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
+static int fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	fr_proto __user *fr_s = ifr->ifr_settings.ifs_ifsu.fr;
 	const size_t size = sizeof(fr_proto);
@@ -1173,12 +1236,14 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto) /* Different proto */
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_FR;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(fr_s, &hdlc->state.fr.settings, size))
+		if (copy_to_user(fr_s, &state(hdlc)->settings, size))
 			return -EFAULT;
 		return 0;
 
@@ -1213,20 +1278,16 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		if (hdlc->proto.id != IF_PROTO_FR) {
-			hdlc_proto_detach(hdlc);
-			hdlc->state.fr.first_pvc = NULL;
-			hdlc->state.fr.dce_pvc_count = 0;
+		if (dev_to_hdlc(dev)->proto != &proto) { /* Different proto */
+			result = attach_hdlc_protocol(dev, &proto, fr_rx,
+						      sizeof(struct frad_state));
+			if (result)
+				return result;
+			state(hdlc)->first_pvc = NULL;
+			state(hdlc)->dce_pvc_count = 0;
 		}
-		memcpy(&hdlc->state.fr.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.close = fr_close;
-		hdlc->proto.start = fr_start;
-		hdlc->proto.stop = fr_stop;
-		hdlc->proto.detach = fr_destroy;
-		hdlc->proto.netif_rx = fr_rx;
-		hdlc->proto.id = IF_PROTO_FR;
+		memcpy(&state(hdlc)->settings, &new_settings, size);
+
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_FRAD;
@@ -1238,6 +1299,9 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 	case IF_PROTO_FR_DEL_PVC:
 	case IF_PROTO_FR_ADD_ETH_PVC:
 	case IF_PROTO_FR_DEL_ETH_PVC:
+		if (dev_to_hdlc(dev)->proto != &proto) /* Different proto */
+			return -EINVAL;
+
 		if(!capable(CAP_NET_ADMIN))
 			return -EPERM;
 
@@ -1263,3 +1327,24 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Frame-Relay protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_generic.c b/drivers/net/wan/hdlc_generic.c
deleted file mode 100644
index 04ca1f7b6424..000000000000
--- a/drivers/net/wan/hdlc_generic.c
+++ /dev/null
@@ -1,339 +0,0 @@
-/*
- * Generic HDLC support routines for Linux
- *
- * Copyright (C) 1999 - 2005 Krzysztof Halasa <khc@pm.waw.pl>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
- *
- * Currently supported:
- *	* raw IP-in-HDLC
- *	* Cisco HDLC
- *	* Frame Relay with ANSI or CCITT LMI (both user and network side)
- *	* PPP
- *	* X.25
- *
- * Use sethdlc utility to set line parameters, protocol and PVCs
- *
- * How does it work:
- * - proto.open(), close(), start(), stop() calls are serialized.
- *   The order is: open, [ start, stop ... ] close ...
- * - proto.start() and stop() are called with spin_lock_irq held.
- */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/poll.h>
-#include <linux/errno.h>
-#include <linux/if_arp.h>
-#include <linux/init.h>
-#include <linux/skbuff.h>
-#include <linux/pkt_sched.h>
-#include <linux/inetdevice.h>
-#include <linux/lapb.h>
-#include <linux/rtnetlink.h>
-#include <linux/notifier.h>
-#include <linux/hdlc.h>
-
-
-static const char* version = "HDLC support module revision 1.19";
-
-#undef DEBUG_LINK
-
-
-static int hdlc_change_mtu(struct net_device *dev, int new_mtu)
-{
-	if ((new_mtu < 68) || (new_mtu > HDLC_MAX_MTU))
-		return -EINVAL;
-	dev->mtu = new_mtu;
-	return 0;
-}
-
-
-
-static struct net_device_stats *hdlc_get_stats(struct net_device *dev)
-{
-	return hdlc_stats(dev);
-}
-
-
-
-static int hdlc_rcv(struct sk_buff *skb, struct net_device *dev,
-		    struct packet_type *p, struct net_device *orig_dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	if (hdlc->proto.netif_rx)
-		return hdlc->proto.netif_rx(skb);
-
-	hdlc->stats.rx_dropped++; /* Shouldn't happen */
-	dev_kfree_skb(skb);
-	return NET_RX_DROP;
-}
-
-
-
-static inline void hdlc_proto_start(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	if (hdlc->proto.start)
-		return hdlc->proto.start(dev);
-}
-
-
-
-static inline void hdlc_proto_stop(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	if (hdlc->proto.stop)
-		return hdlc->proto.stop(dev);
-}
-
-
-
-static int hdlc_device_event(struct notifier_block *this, unsigned long event,
-			     void *ptr)
-{
-	struct net_device *dev = ptr;
-	hdlc_device *hdlc;
-	unsigned long flags;
-	int on;
- 
-	if (dev->get_stats != hdlc_get_stats)
-		return NOTIFY_DONE; /* not an HDLC device */
- 
-	if (event != NETDEV_CHANGE)
-		return NOTIFY_DONE; /* Only interrested in carrier changes */
-
-	on = netif_carrier_ok(dev);
-
-#ifdef DEBUG_LINK
-	printk(KERN_DEBUG "%s: hdlc_device_event NETDEV_CHANGE, carrier %i\n",
-	       dev->name, on);
-#endif
-
-	hdlc = dev_to_hdlc(dev);
-	spin_lock_irqsave(&hdlc->state_lock, flags);
-
-	if (hdlc->carrier == on)
-		goto carrier_exit; /* no change in DCD line level */
-
-	hdlc->carrier = on;
-
-	if (!hdlc->open)
-		goto carrier_exit;
-
-	if (hdlc->carrier) {
-		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
-		hdlc_proto_start(dev);
-	} else {
-		printk(KERN_INFO "%s: Carrier lost\n", dev->name);
-		hdlc_proto_stop(dev);
-	}
-
-carrier_exit:
-	spin_unlock_irqrestore(&hdlc->state_lock, flags);
-	return NOTIFY_DONE;
-}
-
-
-
-/* Must be called by hardware driver when HDLC device is being opened */
-int hdlc_open(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-#ifdef DEBUG_LINK
-	printk(KERN_DEBUG "hdlc_open() carrier %i open %i\n",
-	       hdlc->carrier, hdlc->open);
-#endif
-
-	if (hdlc->proto.id == -1)
-		return -ENOSYS;	/* no protocol attached */
-
-	if (hdlc->proto.open) {
-		int result = hdlc->proto.open(dev);
-		if (result)
-			return result;
-	}
-
-	spin_lock_irq(&hdlc->state_lock);
-
-	if (hdlc->carrier) {
-		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
-		hdlc_proto_start(dev);
-	} else
-		printk(KERN_INFO "%s: No carrier\n", dev->name);
-
-	hdlc->open = 1;
-
-	spin_unlock_irq(&hdlc->state_lock);
-	return 0;
-}
-
-
-
-/* Must be called by hardware driver when HDLC device is being closed */
-void hdlc_close(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-#ifdef DEBUG_LINK
-	printk(KERN_DEBUG "hdlc_close() carrier %i open %i\n",
-	       hdlc->carrier, hdlc->open);
-#endif
-
-	spin_lock_irq(&hdlc->state_lock);
-
-	hdlc->open = 0;
-	if (hdlc->carrier)
-		hdlc_proto_stop(dev);
-
-	spin_unlock_irq(&hdlc->state_lock);
-
-	if (hdlc->proto.close)
-		hdlc->proto.close(dev);
-}
-
-
-
-#ifndef CONFIG_HDLC_RAW
-#define hdlc_raw_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_RAW_ETH
-#define hdlc_raw_eth_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_PPP
-#define hdlc_ppp_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_CISCO
-#define hdlc_cisco_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_FR
-#define hdlc_fr_ioctl(dev, ifr)		-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_X25
-#define hdlc_x25_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-
-int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	unsigned int proto;
-
-	if (cmd != SIOCWANDEV)
-		return -EINVAL;
-
-	switch(ifr->ifr_settings.type) {
-	case IF_PROTO_HDLC:
-	case IF_PROTO_HDLC_ETH:
-	case IF_PROTO_PPP:
-	case IF_PROTO_CISCO:
-	case IF_PROTO_FR:
-	case IF_PROTO_X25:
-		proto = ifr->ifr_settings.type;
-		break;
-
-	default:
-		proto = hdlc->proto.id;
-	}
-
-	switch(proto) {
-	case IF_PROTO_HDLC:	return hdlc_raw_ioctl(dev, ifr);
-	case IF_PROTO_HDLC_ETH:	return hdlc_raw_eth_ioctl(dev, ifr);
-	case IF_PROTO_PPP:	return hdlc_ppp_ioctl(dev, ifr);
-	case IF_PROTO_CISCO:	return hdlc_cisco_ioctl(dev, ifr);
-	case IF_PROTO_FR:	return hdlc_fr_ioctl(dev, ifr);
-	case IF_PROTO_X25:	return hdlc_x25_ioctl(dev, ifr);
-	default:		return -EINVAL;
-	}
-}
-
-void hdlc_setup(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-
-	dev->get_stats = hdlc_get_stats;
-	dev->change_mtu = hdlc_change_mtu;
-	dev->mtu = HDLC_MAX_MTU;
-
-	dev->type = ARPHRD_RAWHDLC;
-	dev->hard_header_len = 16;
-
-	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
-
-	hdlc->proto.id = -1;
-	hdlc->proto.detach = NULL;
-	hdlc->carrier = 1;
-	hdlc->open = 0;
-	spin_lock_init(&hdlc->state_lock);
-}
-
-struct net_device *alloc_hdlcdev(void *priv)
-{
-	struct net_device *dev;
-	dev = alloc_netdev(sizeof(hdlc_device), "hdlc%d", hdlc_setup);
-	if (dev)
-		dev_to_hdlc(dev)->priv = priv;
-	return dev;
-}
-
-void unregister_hdlc_device(struct net_device *dev)
-{
-	rtnl_lock();
-	hdlc_proto_detach(dev_to_hdlc(dev));
-	unregister_netdevice(dev);
-	rtnl_unlock();
-}
-
-
-
-MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
-MODULE_DESCRIPTION("HDLC support module");
-MODULE_LICENSE("GPL v2");
-
-EXPORT_SYMBOL(hdlc_open);
-EXPORT_SYMBOL(hdlc_close);
-EXPORT_SYMBOL(hdlc_ioctl);
-EXPORT_SYMBOL(hdlc_setup);
-EXPORT_SYMBOL(alloc_hdlcdev);
-EXPORT_SYMBOL(unregister_hdlc_device);
-
-static struct packet_type hdlc_packet_type = {
-	.type = __constant_htons(ETH_P_HDLC),
-	.func = hdlc_rcv,
-};
-
-
-static struct notifier_block hdlc_notifier = {
-        .notifier_call = hdlc_device_event,
-};
-
-
-static int __init hdlc_module_init(void)
-{
-	int result;
-
-	printk(KERN_INFO "%s\n", version);
-	if ((result = register_netdevice_notifier(&hdlc_notifier)) != 0)
-                return result;
-        dev_add_pack(&hdlc_packet_type);
-	return 0;
-}
-
-
-
-static void __exit hdlc_module_exit(void)
-{
-	dev_remove_pack(&hdlc_packet_type);
-	unregister_netdevice_notifier(&hdlc_notifier);
-}
-
-
-module_init(hdlc_module_init);
-module_exit(hdlc_module_exit);
diff --git a/drivers/net/wan/hdlc_ppp.c b/drivers/net/wan/hdlc_ppp.c
index fbaab5bf71eb..e9f717070fde 100644
--- a/drivers/net/wan/hdlc_ppp.c
+++ b/drivers/net/wan/hdlc_ppp.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * Point-to-point protocol support
  *
- * Copyright (C) 1999 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -22,6 +22,21 @@
 #include <linux/lapb.h>
 #include <linux/rtnetlink.h>
 #include <linux/hdlc.h>
+#include <net/syncppp.h>
+
+struct ppp_state {
+	struct ppp_device pppdev;
+	struct ppp_device *syncppp_ptr;
+	int (*old_change_mtu)(struct net_device *dev, int new_mtu);
+};
+
+static int ppp_ioctl(struct net_device *dev, struct ifreq *ifr);
+
+
+static inline struct ppp_state* state(hdlc_device *hdlc)
+{
+	return(struct ppp_state *)(hdlc->state);
+}
 
 
 static int ppp_open(struct net_device *dev)
@@ -30,16 +45,16 @@ static int ppp_open(struct net_device *dev)
 	void *old_ioctl;
 	int result;
 
-	dev->priv = &hdlc->state.ppp.syncppp_ptr;
-	hdlc->state.ppp.syncppp_ptr = &hdlc->state.ppp.pppdev;
-	hdlc->state.ppp.pppdev.dev = dev;
+	dev->priv = &state(hdlc)->syncppp_ptr;
+	state(hdlc)->syncppp_ptr = &state(hdlc)->pppdev;
+	state(hdlc)->pppdev.dev = dev;
 
 	old_ioctl = dev->do_ioctl;
-	hdlc->state.ppp.old_change_mtu = dev->change_mtu;
-	sppp_attach(&hdlc->state.ppp.pppdev);
+	state(hdlc)->old_change_mtu = dev->change_mtu;
+	sppp_attach(&state(hdlc)->pppdev);
 	/* sppp_attach nukes them. We don't need syncppp's ioctl */
 	dev->do_ioctl = old_ioctl;
-	hdlc->state.ppp.pppdev.sppp.pp_flags &= ~PP_CISCO;
+	state(hdlc)->pppdev.sppp.pp_flags &= ~PP_CISCO;
 	dev->type = ARPHRD_PPP;
 	result = sppp_open(dev);
 	if (result) {
@@ -59,7 +74,7 @@ static void ppp_close(struct net_device *dev)
 	sppp_close(dev);
 	sppp_detach(dev);
 	dev->rebuild_header = NULL;
-	dev->change_mtu = hdlc->state.ppp.old_change_mtu;
+	dev->change_mtu = state(hdlc)->old_change_mtu;
 	dev->mtu = HDLC_MAX_MTU;
 	dev->hard_header_len = 16;
 }
@@ -73,13 +88,24 @@ static __be16 ppp_type_trans(struct sk_buff *skb, struct net_device *dev)
 
 
 
-int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.open		= ppp_open,
+	.close		= ppp_close,
+	.type_trans	= ppp_type_trans,
+	.ioctl		= ppp_ioctl,
+	.module		= THIS_MODULE,
+};
+
+
+static int ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	int result;
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_PPP;
 		return 0; /* return protocol only, no settable parameters */
 
@@ -96,13 +122,10 @@ int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.open = ppp_open;
-		hdlc->proto.close = ppp_close;
-		hdlc->proto.type_trans = ppp_type_trans;
-		hdlc->proto.id = IF_PROTO_PPP;
+		result = attach_hdlc_protocol(dev, &proto, NULL,
+					      sizeof(struct ppp_state));
+		if (result)
+			return result;
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_PPP;
@@ -113,3 +136,25 @@ int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("PPP protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_raw.c b/drivers/net/wan/hdlc_raw.c
index f15aa6ba77f1..fe3cae5c6b9d 100644
--- a/drivers/net/wan/hdlc_raw.c
+++ b/drivers/net/wan/hdlc_raw.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * HDLC support
  *
- * Copyright (C) 1999 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -24,6 +24,8 @@
 #include <linux/hdlc.h>
 
 
+static int raw_ioctl(struct net_device *dev, struct ifreq *ifr);
+
 static __be16 raw_type_trans(struct sk_buff *skb, struct net_device *dev)
 {
 	return __constant_htons(ETH_P_IP);
@@ -31,7 +33,14 @@ static __be16 raw_type_trans(struct sk_buff *skb, struct net_device *dev)
 
 
 
-int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.type_trans	= raw_type_trans,
+	.ioctl		= raw_ioctl,
+	.module		= THIS_MODULE,
+};
+
+
+static int raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	raw_hdlc_proto __user *raw_s = ifr->ifr_settings.ifs_ifsu.raw_hdlc;
 	const size_t size = sizeof(raw_hdlc_proto);
@@ -41,12 +50,14 @@ int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_HDLC;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(raw_s, &hdlc->state.raw_hdlc.settings, size))
+		if (copy_to_user(raw_s, hdlc->state, size))
 			return -EFAULT;
 		return 0;
 
@@ -71,12 +82,11 @@ int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memcpy(&hdlc->state.raw_hdlc.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.type_trans = raw_type_trans;
-		hdlc->proto.id = IF_PROTO_HDLC;
+		result = attach_hdlc_protocol(dev, &proto, NULL,
+					      sizeof(raw_hdlc_proto));
+		if (result)
+			return result;
+		memcpy(hdlc->state, &new_settings, size);
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_RAWHDLC;
@@ -88,3 +98,25 @@ int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Raw HDLC protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_raw_eth.c b/drivers/net/wan/hdlc_raw_eth.c
index d1884987f94e..1a69a9aaa9b9 100644
--- a/drivers/net/wan/hdlc_raw_eth.c
+++ b/drivers/net/wan/hdlc_raw_eth.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * HDLC Ethernet emulation support
  *
- * Copyright (C) 2002-2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 2002-2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -25,6 +25,7 @@
 #include <linux/etherdevice.h>
 #include <linux/hdlc.h>
 
+static int raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr);
 
 static int eth_tx(struct sk_buff *skb, struct net_device *dev)
 {
@@ -44,7 +45,14 @@ static int eth_tx(struct sk_buff *skb, struct net_device *dev)
 }
 
 
-int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.type_trans	= eth_type_trans,
+	.ioctl		= raw_eth_ioctl,
+	.module		= THIS_MODULE,
+};
+
+
+static int raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	raw_hdlc_proto __user *raw_s = ifr->ifr_settings.ifs_ifsu.raw_hdlc;
 	const size_t size = sizeof(raw_hdlc_proto);
@@ -56,12 +64,14 @@ int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_HDLC_ETH;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(raw_s, &hdlc->state.raw_hdlc.settings, size))
+		if (copy_to_user(raw_s, hdlc->state, size))
 			return -EFAULT;
 		return 0;
 
@@ -86,12 +96,11 @@ int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memcpy(&hdlc->state.raw_hdlc.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.type_trans = eth_type_trans;
-		hdlc->proto.id = IF_PROTO_HDLC_ETH;
+		result = attach_hdlc_protocol(dev, &proto, NULL,
+					      sizeof(raw_hdlc_proto));
+		if (result)
+			return result;
+		memcpy(hdlc->state, &new_settings, size);
 		dev->hard_start_xmit = eth_tx;
 		old_ch_mtu = dev->change_mtu;
 		old_qlen = dev->tx_queue_len;
@@ -106,3 +115,25 @@ int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Ethernet encapsulation support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index a867fb411f89..e4bb9f8ad433 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * X.25 support
  *
- * Copyright (C) 1999 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -25,6 +25,8 @@
 
 #include <net/x25device.h>
 
+static int x25_ioctl(struct net_device *dev, struct ifreq *ifr);
+
 /* These functions are callbacks called by LAPB layer */
 
 static void x25_connect_disconnect(struct net_device *dev, int reason, int code)
@@ -162,30 +164,39 @@ static void x25_close(struct net_device *dev)
 
 static int x25_rx(struct sk_buff *skb)
 {
-	hdlc_device *hdlc = dev_to_hdlc(skb->dev);
+	struct hdlc_device_desc *desc = dev_to_desc(skb->dev);
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		hdlc->stats.rx_dropped++;
+		desc->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
 	if (lapb_data_received(skb->dev, skb) == LAPB_OK)
 		return NET_RX_SUCCESS;
 
-	hdlc->stats.rx_errors++;
+	desc->stats.rx_errors++;
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
 
 
+static struct hdlc_proto proto = {
+	.open		= x25_open,
+	.close		= x25_close,
+	.ioctl		= x25_ioctl,
+	.module		= THIS_MODULE,
+};
+
 
-int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr)
+static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	int result;
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_X25;
 		return 0; /* return protocol only, no settable parameters */
 
@@ -200,14 +211,9 @@ int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.open = x25_open;
-		hdlc->proto.close = x25_close;
-		hdlc->proto.netif_rx = x25_rx;
-		hdlc->proto.type_trans = NULL;
-		hdlc->proto.id = IF_PROTO_X25;
+		if ((result = attach_hdlc_protocol(dev, &proto,
+						   x25_rx, 0)) != 0)
+			return result;
 		dev->hard_start_xmit = x25_xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_X25;
@@ -218,3 +224,25 @@ int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("X.25 protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h
index d5ebbb29aeae..d4b333938f73 100644
--- a/include/linux/hdlc.h
+++ b/include/linux/hdlc.h
@@ -11,95 +11,46 @@
 #ifndef __HDLC_H
 #define __HDLC_H
 
-#define GENERIC_HDLC_VERSION 4	/* For synchronization with sethdlc utility */
-
-#define CLOCK_DEFAULT   0	/* Default setting */
-#define CLOCK_EXT	1	/* External TX and RX clock - DTE */
-#define CLOCK_INT	2	/* Internal TX and RX clock - DCE */
-#define CLOCK_TXINT	3	/* Internal TX and external RX clock */
-#define CLOCK_TXFROMRX	4	/* TX clock derived from external RX clock */
-
-
-#define ENCODING_DEFAULT	0 /* Default setting */
-#define ENCODING_NRZ		1
-#define ENCODING_NRZI		2
-#define ENCODING_FM_MARK	3
-#define ENCODING_FM_SPACE	4
-#define ENCODING_MANCHESTER	5
-
-
-#define PARITY_DEFAULT		0 /* Default setting */
-#define PARITY_NONE		1 /* No parity */
-#define PARITY_CRC16_PR0	2 /* CRC16, initial value 0x0000 */
-#define PARITY_CRC16_PR1	3 /* CRC16, initial value 0xFFFF */
-#define PARITY_CRC16_PR0_CCITT	4 /* CRC16, initial 0x0000, ITU-T version */
-#define PARITY_CRC16_PR1_CCITT	5 /* CRC16, initial 0xFFFF, ITU-T version */
-#define PARITY_CRC32_PR0_CCITT	6 /* CRC32, initial value 0x00000000 */
-#define PARITY_CRC32_PR1_CCITT	7 /* CRC32, initial value 0xFFFFFFFF */
-
-#define LMI_DEFAULT		0 /* Default setting */
-#define LMI_NONE		1 /* No LMI, all PVCs are static */
-#define LMI_ANSI		2 /* ANSI Annex D */
-#define LMI_CCITT		3 /* ITU-T Annex A */
-#define LMI_CISCO		4 /* The "original" LMI, aka Gang of Four */
 
 #define HDLC_MAX_MTU 1500	/* Ethernet 1500 bytes */
+#if 0
 #define HDLC_MAX_MRU (HDLC_MAX_MTU + 10 + 14 + 4) /* for ETH+VLAN over FR */
+#else
+#define HDLC_MAX_MRU 1600 /* as required for FR network */
+#endif
 
 
 #ifdef __KERNEL__
 
 #include <linux/skbuff.h>
 #include <linux/netdevice.h>
-#include <net/syncppp.h>
 #include <linux/hdlc/ioctl.h>
 
 
-typedef struct {		/* Used in Cisco and PPP mode */
-	u8 address;
-	u8 control;
-	u16 protocol;
-}__attribute__ ((packed)) hdlc_header;
-
-
-
-typedef struct {
-	u32 type;		/* code */
-	u32 par1;
-	u32 par2;
-	u16 rel;		/* reliability */
-	u32 time;
-}__attribute__ ((packed)) cisco_packet;
-#define	CISCO_PACKET_LEN	18
-#define	CISCO_BIG_PACKET_LEN	20
-
-
-
-typedef struct pvc_device_struct {
-	struct net_device *master;
-	struct net_device *main;
-	struct net_device *ether; /* bridged Ethernet interface */
-	struct pvc_device_struct *next;	/* Sorted in ascending DLCI order */
-	int dlci;
-	int open_count;
-
-	struct {
-		unsigned int new: 1;
-		unsigned int active: 1;
-		unsigned int exist: 1;
-		unsigned int deleted: 1;
-		unsigned int fecn: 1;
-		unsigned int becn: 1;
-		unsigned int bandwidth;	/* Cisco LMI reporting only */
-	}state;
-}pvc_device;
-
-
-
-typedef struct hdlc_device_struct {
-	/* To be initialized by hardware driver */
+/* Used by all network devices here, pointed to by netdev_priv(dev) */
+struct hdlc_device_desc {
+	int (*netif_rx)(struct sk_buff *skb);
 	struct net_device_stats stats;
-
+};
+
+/* This structure is a private property of HDLC protocols.
+   Hardware drivers have no interest here */
+
+struct hdlc_proto {
+	int (*open)(struct net_device *dev);
+	void (*close)(struct net_device *dev);
+	void (*start)(struct net_device *dev); /* if open & DCD */
+	void (*stop)(struct net_device *dev); /* if open & !DCD */
+	void (*detach)(struct net_device *dev);
+	int (*ioctl)(struct net_device *dev, struct ifreq *ifr);
+	unsigned short (*type_trans)(struct sk_buff *skb,
+				     struct net_device *dev);
+	struct module *module;
+	struct hdlc_proto *next; /* next protocol in the list */
+};
+
+
+typedef struct hdlc_device {
 	/* used by HDLC layer to take control over HDLC device from hw driver*/
 	int (*attach)(struct net_device *dev,
 		      unsigned short encoding, unsigned short parity);
@@ -107,82 +58,18 @@ typedef struct hdlc_device_struct {
 	/* hardware driver must handle this instead of dev->hard_start_xmit */
 	int (*xmit)(struct sk_buff *skb, struct net_device *dev);
 
-
 	/* Things below are for HDLC layer internal use only */
-	struct {
-		int (*open)(struct net_device *dev);
-		void (*close)(struct net_device *dev);
-
-		/* if open & DCD */
-		void (*start)(struct net_device *dev);
-		/* if open & !DCD */
-		void (*stop)(struct net_device *dev);
-
-		void (*detach)(struct hdlc_device_struct *hdlc);
-		int (*netif_rx)(struct sk_buff *skb);
-		unsigned short (*type_trans)(struct sk_buff *skb,
-					     struct net_device *dev);
-		int id;		/* IF_PROTO_HDLC/CISCO/FR/etc. */
-	}proto;
-
+	const struct hdlc_proto *proto;
 	int carrier;
 	int open;
 	spinlock_t state_lock;
-
-	union {
-		struct {
-			fr_proto settings;
-			pvc_device *first_pvc;
-			int dce_pvc_count;
-
-			struct timer_list timer;
-			unsigned long last_poll;
-			int reliable;
-			int dce_changed;
-			int request;
-			int fullrep_sent;
-			u32 last_errors; /* last errors bit list */
-			u8 n391cnt;
-			u8 txseq; /* TX sequence number */
-			u8 rxseq; /* RX sequence number */
-		}fr;
-
-		struct {
-			cisco_proto settings;
-
-			struct timer_list timer;
-			unsigned long last_poll;
-			int up;
-			int request_sent;
-			u32 txseq; /* TX sequence number */
-			u32 rxseq; /* RX sequence number */
-		}cisco;
-
-		struct {
-			raw_hdlc_proto settings;
-		}raw_hdlc;
-
-		struct {
-			struct ppp_device pppdev;
-			struct ppp_device *syncppp_ptr;
-			int (*old_change_mtu)(struct net_device *dev,
-					      int new_mtu);
-		}ppp;
-	}state;
+	void *state;
 	void *priv;
 }hdlc_device;
 
 
 
-int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr);
-
-
-/* Exported from hdlc.o */
+/* Exported from hdlc module */
 
 /* Called by hardware driver when a user requests HDLC service */
 int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
@@ -191,17 +78,21 @@ int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
 #define register_hdlc_device(dev)	register_netdev(dev)
 void unregister_hdlc_device(struct net_device *dev);
 
+
+void register_hdlc_protocol(struct hdlc_proto *proto);
+void unregister_hdlc_protocol(struct hdlc_proto *proto);
+
 struct net_device *alloc_hdlcdev(void *priv);
 
-static __inline__ hdlc_device* dev_to_hdlc(struct net_device *dev)
+
+static __inline__ struct hdlc_device_desc* dev_to_desc(struct net_device *dev)
 {
 	return netdev_priv(dev);
 }
 
-
-static __inline__ pvc_device* dev_to_pvc(struct net_device *dev)
+static __inline__ hdlc_device* dev_to_hdlc(struct net_device *dev)
 {
-	return (pvc_device*)dev->priv;
+	return netdev_priv(dev) + sizeof(struct hdlc_device_desc);
 }
 
 
@@ -225,18 +116,14 @@ int hdlc_open(struct net_device *dev);
 /* Must be called by hardware driver when HDLC device is being closed */
 void hdlc_close(struct net_device *dev);
 
+int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto,
+			 int (*rx)(struct sk_buff *skb), size_t size);
 /* May be used by hardware driver to gain control over HDLC device */
-static __inline__ void hdlc_proto_detach(hdlc_device *hdlc)
-{
-	if (hdlc->proto.detach)
-		hdlc->proto.detach(hdlc);
-	hdlc->proto.detach = NULL;
-}
-
+void detach_hdlc_protocol(struct net_device *dev);
 
 static __inline__ struct net_device_stats *hdlc_stats(struct net_device *dev)
 {
-	return &dev_to_hdlc(dev)->stats;
+	return &dev_to_desc(dev)->stats;
 }
 
 
@@ -248,8 +135,8 @@ static __inline__ __be16 hdlc_type_trans(struct sk_buff *skb,
 	skb->mac.raw  = skb->data;
 	skb->dev      = dev;
 
-	if (hdlc->proto.type_trans)
-		return hdlc->proto.type_trans(skb, dev);
+	if (hdlc->proto->type_trans)
+		return hdlc->proto->type_trans(skb, dev);
 	else
 		return htons(ETH_P_HDLC);
 }
diff --git a/include/linux/hdlc/ioctl.h b/include/linux/hdlc/ioctl.h
index 78430ba3ea69..583972364357 100644
--- a/include/linux/hdlc/ioctl.h
+++ b/include/linux/hdlc/ioctl.h
@@ -1,6 +1,39 @@
 #ifndef __HDLC_IOCTL_H__
 #define __HDLC_IOCTL_H__
 
+
+#define GENERIC_HDLC_VERSION 4	/* For synchronization with sethdlc utility */
+
+#define CLOCK_DEFAULT   0	/* Default setting */
+#define CLOCK_EXT	1	/* External TX and RX clock - DTE */
+#define CLOCK_INT	2	/* Internal TX and RX clock - DCE */
+#define CLOCK_TXINT	3	/* Internal TX and external RX clock */
+#define CLOCK_TXFROMRX	4	/* TX clock derived from external RX clock */
+
+
+#define ENCODING_DEFAULT	0 /* Default setting */
+#define ENCODING_NRZ		1
+#define ENCODING_NRZI		2
+#define ENCODING_FM_MARK	3
+#define ENCODING_FM_SPACE	4
+#define ENCODING_MANCHESTER	5
+
+
+#define PARITY_DEFAULT		0 /* Default setting */
+#define PARITY_NONE		1 /* No parity */
+#define PARITY_CRC16_PR0	2 /* CRC16, initial value 0x0000 */
+#define PARITY_CRC16_PR1	3 /* CRC16, initial value 0xFFFF */
+#define PARITY_CRC16_PR0_CCITT	4 /* CRC16, initial 0x0000, ITU-T version */
+#define PARITY_CRC16_PR1_CCITT	5 /* CRC16, initial 0xFFFF, ITU-T version */
+#define PARITY_CRC32_PR0_CCITT	6 /* CRC32, initial value 0x00000000 */
+#define PARITY_CRC32_PR1_CCITT	7 /* CRC32, initial value 0xFFFFFFFF */
+
+#define LMI_DEFAULT		0 /* Default setting */
+#define LMI_NONE		1 /* No LMI, all PVCs are static */
+#define LMI_ANSI		2 /* ANSI Annex D */
+#define LMI_CCITT		3 /* ITU-T Annex A */
+#define LMI_CISCO		4 /* The "original" LMI, aka Gang of Four */
+
 typedef struct { 
 	unsigned int clock_rate; /* bits per second */
 	unsigned int clock_type; /* internal, external, TX-internal etc. */
-- 
cgit v1.2.3


From b29e7132b5a9f2496beed37beef7ba4d010afb2c Mon Sep 17 00:00:00 2001
From: Russ Anderson <rja@sgi.com>
Date: Tue, 26 Sep 2006 14:47:48 -0500
Subject: [IA64] PAL calls need physical mode, stacked

PAL_CACHE_READ and PAL_CACHE_WRITE need to be called in physical
mode with stacked registers.

Signed-off-by: Russ Anderson (rja@sgi.com)
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/pal.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h
index d1587e4f5759..2c8fd92d0ece 100644
--- a/include/asm-ia64/pal.h
+++ b/include/asm-ia64/pal.h
@@ -964,7 +964,8 @@ static inline s64
 ia64_pal_cache_read (pal_cache_line_id_u_t line_id, u64 physical_addr)
 {
 	struct ia64_pal_retval iprv;
-	PAL_CALL(iprv, PAL_CACHE_READ, line_id.pclid_data, physical_addr, 0);
+	PAL_CALL_PHYS_STK(iprv, PAL_CACHE_READ, line_id.pclid_data,
+				physical_addr, 0);
 	return iprv.status;
 }
 
@@ -986,7 +987,8 @@ static inline s64
 ia64_pal_cache_write (pal_cache_line_id_u_t line_id, u64 physical_addr, u64 data)
 {
 	struct ia64_pal_retval iprv;
-	PAL_CALL(iprv, PAL_CACHE_WRITE, line_id.pclid_data, physical_addr, data);
+	PAL_CALL_PHYS_STK(iprv, PAL_CACHE_WRITE, line_id.pclid_data,
+				physical_addr, data);
 	return iprv.status;
 }
 
-- 
cgit v1.2.3


From 51c3711704b66986373408cbc0540abea43d2380 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:33:16 +0200
Subject: i2c-algo-sibyte: Merge into i2c-sibyte

i2c-algo-sibyte: Merge into i2c-sibyte

Merge i2c-algo-sibyte into i2c-sibyte, as this is a complete,
hardware-dependent SMBus implementation and not a reusable algorithm.

Perform some basic coding style cleanups while we're here (mainly
space-based indentation replaced by tabulations.)

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/algos/Kconfig           |   6 --
 drivers/i2c/algos/Makefile          |   1 -
 drivers/i2c/algos/i2c-algo-sibyte.c | 186 ------------------------------------
 drivers/i2c/busses/i2c-sibyte.c     | 158 +++++++++++++++++++++++++++++-
 include/linux/i2c-algo-sibyte.h     |  33 -------
 5 files changed, 155 insertions(+), 229 deletions(-)
 delete mode 100644 drivers/i2c/algos/i2c-algo-sibyte.c
 delete mode 100644 include/linux/i2c-algo-sibyte.h

(limited to 'include')

diff --git a/drivers/i2c/algos/Kconfig b/drivers/i2c/algos/Kconfig
index 30408015d231..c034820615bb 100644
--- a/drivers/i2c/algos/Kconfig
+++ b/drivers/i2c/algos/Kconfig
@@ -53,12 +53,6 @@ config I2C_ALGO8XX
 	tristate "MPC8xx CPM I2C interface"
 	depends on 8xx && I2C
 
-config I2C_ALGO_SIBYTE
-	tristate "SiByte SMBus interface"
-	depends on SIBYTE_SB1xxx_SOC && I2C
-	help
-	  Supports the SiByte SOC on-chip I2C interfaces (2 channels).
-
 config I2C_ALGO_SGI
 	tristate "I2C SGI interfaces"
 	depends on I2C && (SGI_IP22 || SGI_IP32 || X86_VISWS)
diff --git a/drivers/i2c/algos/Makefile b/drivers/i2c/algos/Makefile
index 867fe1f67401..208be04a3dbd 100644
--- a/drivers/i2c/algos/Makefile
+++ b/drivers/i2c/algos/Makefile
@@ -6,7 +6,6 @@ obj-$(CONFIG_I2C_ALGOBIT)	+= i2c-algo-bit.o
 obj-$(CONFIG_I2C_ALGOPCF)	+= i2c-algo-pcf.o
 obj-$(CONFIG_I2C_ALGOPCA)	+= i2c-algo-pca.o
 obj-$(CONFIG_I2C_ALGOITE)	+= i2c-algo-ite.o
-obj-$(CONFIG_I2C_ALGO_SIBYTE)	+= i2c-algo-sibyte.o
 obj-$(CONFIG_I2C_ALGO_SGI)	+= i2c-algo-sgi.o
 
 ifeq ($(CONFIG_I2C_DEBUG_ALGO),y)
diff --git a/drivers/i2c/algos/i2c-algo-sibyte.c b/drivers/i2c/algos/i2c-algo-sibyte.c
deleted file mode 100644
index 16d666fa16f3..000000000000
--- a/drivers/i2c/algos/i2c-algo-sibyte.c
+++ /dev/null
@@ -1,186 +0,0 @@
-/* ------------------------------------------------------------------------- */
-/* i2c-algo-sibyte.c i2c driver algorithms for bit-shift adapters		     */
-/* ------------------------------------------------------------------------- */
-/*   Copyright (C) 2001,2002,2003 Broadcom Corporation
-     Copyright (C) 1995-2000 Simon G. Vogl
-
-    This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.		     */
-/* ------------------------------------------------------------------------- */
-
-/* With some changes from Kyösti Mälkki <kmalkki@cc.hut.fi> and even
-   Frodo Looijaard <frodol@dds.nl>.  */
-
-/* Ported for SiByte SOCs by Broadcom Corporation.  */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/sibyte/sb1250_regs.h>
-#include <asm/sibyte/sb1250_smbus.h>
-
-#include <linux/i2c.h>
-#include <linux/i2c-algo-sibyte.h>
-
-/* ----- global defines ----------------------------------------------- */
-#define SMB_CSR(a,r) ((long)(a->reg_base + r))
-
-/* ----- global variables ---------------------------------------------	*/
-
-/* module parameters:
- */
-static int bit_scan;	/* have a look at what's hanging 'round		*/
-
-
-static int smbus_xfer(struct i2c_adapter *i2c_adap, u16 addr, 
-                      unsigned short flags, char read_write,
-                      u8 command, int size, union i2c_smbus_data * data)
-{
-	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
-        int data_bytes = 0;
-        int error;
-
-        while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
-                ;
-
-        switch (size) {
-        case I2C_SMBUS_QUICK:
-                csr_out32((V_SMB_ADDR(addr) | (read_write == I2C_SMBUS_READ ? M_SMB_QDATA : 0) |
-			   V_SMB_TT_QUICKCMD), SMB_CSR(adap, R_SMB_START));
-                break;
-        case I2C_SMBUS_BYTE:
-                if (read_write == I2C_SMBUS_READ) {
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_RD1BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                        data_bytes = 1;
-                } else {
-                        csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR1BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                }
-                break;
-        case I2C_SMBUS_BYTE_DATA:
-                csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
-                if (read_write == I2C_SMBUS_READ) {
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD1BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                        data_bytes = 1;
-                } else {
-                        csr_out32(V_SMB_LB(data->byte), SMB_CSR(adap, R_SMB_DATA));
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                }
-                break;
-        case I2C_SMBUS_WORD_DATA:
-                csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
-                if (read_write == I2C_SMBUS_READ) {
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD2BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                        data_bytes = 2;
-                } else {
-                        csr_out32(V_SMB_LB(data->word & 0xff), SMB_CSR(adap, R_SMB_DATA));
-                        csr_out32(V_SMB_MB(data->word >> 8), SMB_CSR(adap, R_SMB_DATA));
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                }
-                break;
-        default:
-                return -1;      /* XXXKW better error code? */
-        }
-
-        while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
-                ;
-
-        error = csr_in32(SMB_CSR(adap, R_SMB_STATUS));
-        if (error & M_SMB_ERROR) {
-                /* Clear error bit by writing a 1 */
-                csr_out32(M_SMB_ERROR, SMB_CSR(adap, R_SMB_STATUS));
-                return -1;      /* XXXKW better error code? */
-        }
-
-        if (data_bytes == 1)
-                data->byte = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xff;
-        if (data_bytes == 2)
-                data->word = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xffff;
-
-        return 0;
-}
-
-static u32 bit_func(struct i2c_adapter *adap)
-{
-	return (I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
-                I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA);
-}
-
-
-/* -----exported algorithm data: -------------------------------------	*/
-
-static struct i2c_algorithm i2c_sibyte_algo = {
-	.smbus_xfer	= smbus_xfer,
-	.functionality	= bit_func,
-};
-
-/* 
- * registering functions to load algorithms at runtime 
- */
-int i2c_sibyte_add_bus(struct i2c_adapter *i2c_adap, int speed)
-{
-	int i;
-	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
-
-	/* register new adapter to i2c module... */
-	i2c_adap->algo = &i2c_sibyte_algo;
-        
-        /* Set the frequency to 100 kHz */
-        csr_out32(speed, SMB_CSR(adap,R_SMB_FREQ));
-        csr_out32(0, SMB_CSR(adap,R_SMB_CONTROL));
-
-	/* scan bus */
-	if (bit_scan) {
-                union i2c_smbus_data data;
-                int rc;
-		printk(KERN_INFO " i2c-algo-sibyte.o: scanning bus %s.\n",
-		       i2c_adap->name);
-		for (i = 0x00; i < 0x7f; i++) {
-                        /* XXXKW is this a realistic probe? */
-                        rc = smbus_xfer(i2c_adap, i, 0, I2C_SMBUS_READ, 0,
-                                        I2C_SMBUS_BYTE_DATA, &data);
-			if (!rc) {
-				printk("(%02x)",i); 
-			} else 
-				printk("."); 
-		}
-		printk("\n");
-	}
-
-	return i2c_add_adapter(i2c_adap);
-}
-
-
-int i2c_sibyte_del_bus(struct i2c_adapter *adap)
-{
-	return i2c_del_adapter(adap);
-}
-
-
-EXPORT_SYMBOL(i2c_sibyte_add_bus);
-EXPORT_SYMBOL(i2c_sibyte_del_bus);
-
-MODULE_AUTHOR("Kip Walker, Broadcom Corp.");
-MODULE_DESCRIPTION("SiByte I2C-Bus algorithm");
-module_param(bit_scan, int, 0);
-MODULE_PARM_DESC(bit_scan, "Scan for active chips on the bus");
-MODULE_LICENSE("GPL");
diff --git a/drivers/i2c/busses/i2c-sibyte.c b/drivers/i2c/busses/i2c-sibyte.c
index fa503ed9f86d..f516eb7a23f7 100644
--- a/drivers/i2c/busses/i2c-sibyte.c
+++ b/drivers/i2c/busses/i2c-sibyte.c
@@ -1,6 +1,7 @@
 /*
  * Copyright (C) 2004 Steven J. Hill
  * Copyright (C) 2001,2002,2003 Broadcom Corporation
+ * Copyright (C) 1995-2000 Simon G. Vogl
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
@@ -17,11 +18,162 @@
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
+#include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/i2c-algo-sibyte.h>
+#include <linux/init.h>
+#include <linux/i2c.h>
+#include <asm/io.h>
 #include <asm/sibyte/sb1250_regs.h>
 #include <asm/sibyte/sb1250_smbus.h>
 
+
+struct i2c_algo_sibyte_data {
+	void *data;		/* private data */
+	int   bus;		/* which bus */
+	void *reg_base;		/* CSR base */
+};
+
+/* ----- global defines ----------------------------------------------- */
+#define SMB_CSR(a,r) ((long)(a->reg_base + r))
+
+/* ----- global variables --------------------------------------------- */
+
+/* module parameters:
+ */
+static int bit_scan;	/* have a look at what's hanging 'round */
+module_param(bit_scan, int, 0);
+MODULE_PARM_DESC(bit_scan, "Scan for active chips on the bus");
+
+
+static int smbus_xfer(struct i2c_adapter *i2c_adap, u16 addr,
+		      unsigned short flags, char read_write,
+		      u8 command, int size, union i2c_smbus_data * data)
+{
+	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
+	int data_bytes = 0;
+	int error;
+
+	while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
+		;
+
+	switch (size) {
+	case I2C_SMBUS_QUICK:
+		csr_out32((V_SMB_ADDR(addr) |
+			   (read_write == I2C_SMBUS_READ ? M_SMB_QDATA : 0) |
+			   V_SMB_TT_QUICKCMD), SMB_CSR(adap, R_SMB_START));
+		break;
+	case I2C_SMBUS_BYTE:
+		if (read_write == I2C_SMBUS_READ) {
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_RD1BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+			data_bytes = 1;
+		} else {
+			csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR1BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+		}
+		break;
+	case I2C_SMBUS_BYTE_DATA:
+		csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
+		if (read_write == I2C_SMBUS_READ) {
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD1BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+			data_bytes = 1;
+		} else {
+			csr_out32(V_SMB_LB(data->byte),
+				  SMB_CSR(adap, R_SMB_DATA));
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+		}
+		break;
+	case I2C_SMBUS_WORD_DATA:
+		csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
+		if (read_write == I2C_SMBUS_READ) {
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD2BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+			data_bytes = 2;
+		} else {
+			csr_out32(V_SMB_LB(data->word & 0xff),
+				  SMB_CSR(adap, R_SMB_DATA));
+			csr_out32(V_SMB_MB(data->word >> 8),
+				  SMB_CSR(adap, R_SMB_DATA));
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+		}
+		break;
+	default:
+		return -1;      /* XXXKW better error code? */
+	}
+
+	while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
+		;
+
+	error = csr_in32(SMB_CSR(adap, R_SMB_STATUS));
+	if (error & M_SMB_ERROR) {
+		/* Clear error bit by writing a 1 */
+		csr_out32(M_SMB_ERROR, SMB_CSR(adap, R_SMB_STATUS));
+		return -1;      /* XXXKW better error code? */
+	}
+
+	if (data_bytes == 1)
+		data->byte = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xff;
+	if (data_bytes == 2)
+		data->word = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xffff;
+
+	return 0;
+}
+
+static u32 bit_func(struct i2c_adapter *adap)
+{
+	return (I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
+		I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA);
+}
+
+
+/* -----exported algorithm data: -------------------------------------	*/
+
+static struct i2c_algorithm i2c_sibyte_algo = {
+	.smbus_xfer	= smbus_xfer,
+	.functionality	= bit_func,
+};
+
+/*
+ * registering functions to load algorithms at runtime
+ */
+int i2c_sibyte_add_bus(struct i2c_adapter *i2c_adap, int speed)
+{
+	int i;
+	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
+
+	/* register new adapter to i2c module... */
+	i2c_adap->algo = &i2c_sibyte_algo;
+
+	/* Set the frequency to 100 kHz */
+	csr_out32(speed, SMB_CSR(adap,R_SMB_FREQ));
+	csr_out32(0, SMB_CSR(adap,R_SMB_CONTROL));
+
+	/* scan bus */
+	if (bit_scan) {
+		union i2c_smbus_data data;
+		int rc;
+		printk(KERN_INFO " i2c-algo-sibyte.o: scanning bus %s.\n",
+		       i2c_adap->name);
+		for (i = 0x00; i < 0x7f; i++) {
+			/* XXXKW is this a realistic probe? */
+			rc = smbus_xfer(i2c_adap, i, 0, I2C_SMBUS_READ, 0,
+					I2C_SMBUS_BYTE_DATA, &data);
+			if (!rc) {
+				printk("(%02x)",i);
+			} else
+				printk(".");
+		}
+		printk("\n");
+	}
+
+	return i2c_add_adapter(i2c_adap);
+}
+
+
 static struct i2c_algo_sibyte_data sibyte_board_data[2] = {
 	{ NULL, 0, (void *) (CKSEG1+A_SMB_BASE(0)) },
 	{ NULL, 1, (void *) (CKSEG1+A_SMB_BASE(1)) }
@@ -58,8 +210,8 @@ static int __init i2c_sibyte_init(void)
 
 static void __exit i2c_sibyte_exit(void)
 {
-	i2c_sibyte_del_bus(&sibyte_board_adapter[0]);
-	i2c_sibyte_del_bus(&sibyte_board_adapter[1]);
+	i2c_del_bus(&sibyte_board_adapter[0]);
+	i2c_del_bus(&sibyte_board_adapter[1]);
 }
 
 module_init(i2c_sibyte_init);
diff --git a/include/linux/i2c-algo-sibyte.h b/include/linux/i2c-algo-sibyte.h
deleted file mode 100644
index 03914ded8614..000000000000
--- a/include/linux/i2c-algo-sibyte.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/*
- * Copyright (C) 2001,2002,2003 Broadcom Corporation
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
- * of the License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
- */
-
-#ifndef I2C_ALGO_SIBYTE_H
-#define I2C_ALGO_SIBYTE_H 1
-
-#include <linux/i2c.h>
-
-struct i2c_algo_sibyte_data {
-	void *data;		/* private data */
-        int   bus;		/* which bus */
-        void *reg_base;		/* CSR base */
-};
-
-int i2c_sibyte_add_bus(struct i2c_adapter *, int speed);
-int i2c_sibyte_del_bus(struct i2c_adapter *);
-
-#endif /* I2C_ALGO_SIBYTE_H */
-- 
cgit v1.2.3


From a0d9c63d3640bd4fc90a408e8334754ef44bcf48 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Aug 2006 11:46:49 +0200
Subject: i2c-algo-bit: Discard the mdelay data struct member

i2c-algo-bit: Discard the mdelay data struct member

The i2c_algo_bit_data structure has an mdelay member, which is not
used by the algorithm code (the code has always been ifdef'd out.)
Let's discard it to save some code and memory.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/acorn/char/i2c.c                   |  1 -
 drivers/i2c/algos/i2c-algo-bit.c           |  4 ----
 drivers/i2c/busses/i2c-hydra.c             |  1 -
 drivers/i2c/busses/i2c-i810.c              |  2 --
 drivers/i2c/busses/i2c-ixp2000.c           |  1 -
 drivers/i2c/busses/i2c-ixp4xx.c            |  1 -
 drivers/i2c/busses/i2c-parport-light.c     |  1 -
 drivers/i2c/busses/i2c-parport.c           |  1 -
 drivers/i2c/busses/i2c-prosavage.c         |  1 -
 drivers/i2c/busses/i2c-savage4.c           |  1 -
 drivers/i2c/busses/i2c-via.c               |  1 -
 drivers/i2c/busses/i2c-voodoo3.c           |  2 --
 drivers/i2c/busses/scx200_i2c.c            | 12 ++++++------
 drivers/ieee1394/pcilynx.c                 |  1 -
 drivers/media/video/bt8xx/bttv-i2c.c       |  1 -
 drivers/media/video/cx88/cx88-i2c.c        |  1 -
 drivers/media/video/cx88/cx88-vp3054-i2c.c |  1 -
 drivers/media/video/zoran_card.c           |  1 -
 drivers/video/i810/i810-i2c.c              |  1 -
 drivers/video/matrox/i2c-matroxfb.c        |  1 -
 drivers/video/savage/savagefb-i2c.c        |  1 -
 include/linux/i2c-algo-bit.h               |  1 -
 22 files changed, 6 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/drivers/acorn/char/i2c.c b/drivers/acorn/char/i2c.c
index c26c08b36829..bdb9c8b78ed8 100644
--- a/drivers/acorn/char/i2c.c
+++ b/drivers/acorn/char/i2c.c
@@ -308,7 +308,6 @@ static struct i2c_algo_bit_data ioc_data = {
 	.getsda		= ioc_getsda,
 	.getscl		= ioc_getscl,
 	.udelay		= 80,
-	.mdelay		= 80,
 	.timeout	= 100
 };
 
diff --git a/drivers/i2c/algos/i2c-algo-bit.c b/drivers/i2c/algos/i2c-algo-bit.c
index ab230c033f99..761df16838b4 100644
--- a/drivers/i2c/algos/i2c-algo-bit.c
+++ b/drivers/i2c/algos/i2c-algo-bit.c
@@ -354,10 +354,6 @@ static int sendbytes(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
 			return (retval<0)? retval : -EFAULT;
 			        /* got a better one ?? */
 		}
-#if 0
-		/* from asm/delay.h */
-		__delay(adap->mdelay * (loops_per_sec / 1000) );
-#endif
 	}
 	return wrcount;
 }
diff --git a/drivers/i2c/busses/i2c-hydra.c b/drivers/i2c/busses/i2c-hydra.c
index e0cb3b0f92fa..457d48a0ab9d 100644
--- a/drivers/i2c/busses/i2c-hydra.c
+++ b/drivers/i2c/busses/i2c-hydra.c
@@ -99,7 +99,6 @@ static struct i2c_algo_bit_data hydra_bit_data = {
 	.getsda		= hydra_bit_getsda,
 	.getscl		= hydra_bit_getscl,
 	.udelay		= 5,
-	.mdelay		= 5,
 	.timeout	= HZ
 };
 
diff --git a/drivers/i2c/busses/i2c-i810.c b/drivers/i2c/busses/i2c-i810.c
index 748be30f2bae..b66fb6bb1870 100644
--- a/drivers/i2c/busses/i2c-i810.c
+++ b/drivers/i2c/busses/i2c-i810.c
@@ -166,7 +166,6 @@ static struct i2c_algo_bit_data i810_i2c_bit_data = {
 	.getsda		= bit_i810i2c_getsda,
 	.getscl		= bit_i810i2c_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT,
 };
 
@@ -182,7 +181,6 @@ static struct i2c_algo_bit_data i810_ddc_bit_data = {
 	.getsda		= bit_i810ddc_getsda,
 	.getscl		= bit_i810ddc_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT,
 };
 
diff --git a/drivers/i2c/busses/i2c-ixp2000.c b/drivers/i2c/busses/i2c-ixp2000.c
index cd6f45d186ab..dd3f4cd3aa68 100644
--- a/drivers/i2c/busses/i2c-ixp2000.c
+++ b/drivers/i2c/busses/i2c-ixp2000.c
@@ -114,7 +114,6 @@ static int ixp2000_i2c_probe(struct platform_device *plat_dev)
 	drv_data->algo_data.getsda = ixp2000_bit_getsda;
 	drv_data->algo_data.getscl = ixp2000_bit_getscl;
 	drv_data->algo_data.udelay = 6;
-	drv_data->algo_data.mdelay = 6;
 	drv_data->algo_data.timeout = 100;
 
 	drv_data->adapter.id = I2C_HW_B_IXP2000,
diff --git a/drivers/i2c/busses/i2c-ixp4xx.c b/drivers/i2c/busses/i2c-ixp4xx.c
index 2ed07112d683..ab573254a8aa 100644
--- a/drivers/i2c/busses/i2c-ixp4xx.c
+++ b/drivers/i2c/busses/i2c-ixp4xx.c
@@ -122,7 +122,6 @@ static int ixp4xx_i2c_probe(struct platform_device *plat_dev)
 	drv_data->algo_data.getsda = ixp4xx_bit_getsda;
 	drv_data->algo_data.getscl = ixp4xx_bit_getscl;
 	drv_data->algo_data.udelay = 10;
-	drv_data->algo_data.mdelay = 10;
 	drv_data->algo_data.timeout = 100;
 
 	drv_data->adapter.id = I2C_HW_B_IXP4XX;
diff --git a/drivers/i2c/busses/i2c-parport-light.c b/drivers/i2c/busses/i2c-parport-light.c
index e09ebbb2f9f0..5eb2bd294fd9 100644
--- a/drivers/i2c/busses/i2c-parport-light.c
+++ b/drivers/i2c/busses/i2c-parport-light.c
@@ -103,7 +103,6 @@ static struct i2c_algo_bit_data parport_algo_data = {
 	.getsda		= parport_getsda,
 	.getscl		= parport_getscl,
 	.udelay		= 50,
-	.mdelay		= 50,
 	.timeout	= HZ,
 }; 
 
diff --git a/drivers/i2c/busses/i2c-parport.c b/drivers/i2c/busses/i2c-parport.c
index 934bd55bae15..48a829431c7b 100644
--- a/drivers/i2c/busses/i2c-parport.c
+++ b/drivers/i2c/busses/i2c-parport.c
@@ -138,7 +138,6 @@ static struct i2c_algo_bit_data parport_algo_data = {
 	.getsda		= parport_getsda,
 	.getscl		= parport_getscl,
 	.udelay		= 60,
-	.mdelay		= 60,
 	.timeout	= HZ,
 }; 
 
diff --git a/drivers/i2c/busses/i2c-prosavage.c b/drivers/i2c/busses/i2c-prosavage.c
index 9479525892e3..7745e21874a8 100644
--- a/drivers/i2c/busses/i2c-prosavage.c
+++ b/drivers/i2c/busses/i2c-prosavage.c
@@ -180,7 +180,6 @@ static int i2c_register_bus(struct pci_dev *dev, struct s_i2c_bus *p, void __iom
 	p->algo.getsda	  = bit_s3via_getsda;
 	p->algo.getscl	  = bit_s3via_getscl;
 	p->algo.udelay	  = CYCLE_DELAY;
-	p->algo.mdelay	  = CYCLE_DELAY;
 	p->algo.timeout	  = TIMEOUT;
 	p->algo.data	  = p;
 	p->mmvga	  = mmvga;
diff --git a/drivers/i2c/busses/i2c-savage4.c b/drivers/i2c/busses/i2c-savage4.c
index 0c8518298e4d..209f47ea1750 100644
--- a/drivers/i2c/busses/i2c-savage4.c
+++ b/drivers/i2c/busses/i2c-savage4.c
@@ -140,7 +140,6 @@ static struct i2c_algo_bit_data sav_i2c_bit_data = {
 	.getsda		= bit_savi2c_getsda,
 	.getscl		= bit_savi2c_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT
 };
 
diff --git a/drivers/i2c/busses/i2c-via.c b/drivers/i2c/busses/i2c-via.c
index 484bbacfce6b..910e200ad500 100644
--- a/drivers/i2c/busses/i2c-via.c
+++ b/drivers/i2c/busses/i2c-via.c
@@ -81,7 +81,6 @@ static struct i2c_algo_bit_data bit_data = {
 	.getsda		= bit_via_getsda,
 	.getscl		= bit_via_getscl,
 	.udelay		= 5,
-	.mdelay		= 5,
 	.timeout	= HZ
 };
 
diff --git a/drivers/i2c/busses/i2c-voodoo3.c b/drivers/i2c/busses/i2c-voodoo3.c
index b675773b0cc1..6c8d25183382 100644
--- a/drivers/i2c/busses/i2c-voodoo3.c
+++ b/drivers/i2c/busses/i2c-voodoo3.c
@@ -160,7 +160,6 @@ static struct i2c_algo_bit_data voo_i2c_bit_data = {
 	.getsda		= bit_vooi2c_getsda,
 	.getscl		= bit_vooi2c_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT
 };
 
@@ -177,7 +176,6 @@ static struct i2c_algo_bit_data voo_ddc_bit_data = {
 	.getsda		= bit_vooddc_getsda,
 	.getscl		= bit_vooddc_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT
 };
 
diff --git a/drivers/i2c/busses/scx200_i2c.c b/drivers/i2c/busses/scx200_i2c.c
index cb3ef5ac99fd..8b65a5cf8251 100644
--- a/drivers/i2c/busses/scx200_i2c.c
+++ b/drivers/i2c/busses/scx200_i2c.c
@@ -71,12 +71,12 @@ static int scx200_i2c_getsda(void *data)
  */
 
 static struct i2c_algo_bit_data scx200_i2c_data = {
-	NULL,
-	scx200_i2c_setsda,
-	scx200_i2c_setscl,
-	scx200_i2c_getsda,
-	scx200_i2c_getscl,
-	10, 10, 100,		/* waits, timeout */
+	.setsda		= scx200_i2c_setsda,
+	.setscl		= scx200_i2c_setscl,
+	.getsda		= scx200_i2c_getsda,
+	.getscl		= scx200_i2c_getscl,
+	.udelay		= 10,
+	.timeout	= 100,
 };
 
 static struct i2c_adapter scx200_i2c_ops = {
diff --git a/drivers/ieee1394/pcilynx.c b/drivers/ieee1394/pcilynx.c
index e6f41238f5e8..b4f146f2c951 100644
--- a/drivers/ieee1394/pcilynx.c
+++ b/drivers/ieee1394/pcilynx.c
@@ -137,7 +137,6 @@ static struct i2c_algo_bit_data bit_data = {
 	.getsda			= bit_getsda,
 	.getscl			= bit_getscl,
 	.udelay			= 5,
-	.mdelay			= 5,
 	.timeout		= 100,
 };
 
diff --git a/drivers/media/video/bt8xx/bttv-i2c.c b/drivers/media/video/bt8xx/bttv-i2c.c
index 4b562b386fcf..0dfbcc85ebb9 100644
--- a/drivers/media/video/bt8xx/bttv-i2c.c
+++ b/drivers/media/video/bt8xx/bttv-i2c.c
@@ -100,7 +100,6 @@ static struct i2c_algo_bit_data bttv_i2c_algo_bit_template = {
 	.getsda  = bttv_bit_getsda,
 	.getscl  = bttv_bit_getscl,
 	.udelay  = 16,
-	.mdelay  = 10,
 	.timeout = 200,
 };
 
diff --git a/drivers/media/video/cx88/cx88-i2c.c b/drivers/media/video/cx88/cx88-i2c.c
index 70663805cc30..7bea34714861 100644
--- a/drivers/media/video/cx88/cx88-i2c.c
+++ b/drivers/media/video/cx88/cx88-i2c.c
@@ -155,7 +155,6 @@ static struct i2c_algo_bit_data cx8800_i2c_algo_template = {
 	.getsda  = cx8800_bit_getsda,
 	.getscl  = cx8800_bit_getscl,
 	.udelay  = 16,
-	.mdelay  = 10,
 	.timeout = 200,
 };
 
diff --git a/drivers/media/video/cx88/cx88-vp3054-i2c.c b/drivers/media/video/cx88/cx88-vp3054-i2c.c
index 751a754a45e9..2b4f1970c7df 100644
--- a/drivers/media/video/cx88/cx88-vp3054-i2c.c
+++ b/drivers/media/video/cx88/cx88-vp3054-i2c.c
@@ -100,7 +100,6 @@ static struct i2c_algo_bit_data vp3054_i2c_algo_template = {
 	.getsda  = vp3054_bit_getsda,
 	.getscl  = vp3054_bit_getscl,
 	.udelay  = 16,
-	.mdelay  = 10,
 	.timeout = 200,
 };
 
diff --git a/drivers/media/video/zoran_card.c b/drivers/media/video/zoran_card.c
index f2249ed25273..29f59c36f001 100644
--- a/drivers/media/video/zoran_card.c
+++ b/drivers/media/video/zoran_card.c
@@ -820,7 +820,6 @@ static struct i2c_algo_bit_data zoran_i2c_bit_data_template = {
 	.getsda = zoran_i2c_getsda,
 	.getscl = zoran_i2c_getscl,
 	.udelay = 10,
-	.mdelay = 0,
 	.timeout = 100,
 };
 
diff --git a/drivers/video/i810/i810-i2c.c b/drivers/video/i810/i810-i2c.c
index c1f7b49975dd..7d06b38e80a0 100644
--- a/drivers/video/i810/i810-i2c.c
+++ b/drivers/video/i810/i810-i2c.c
@@ -98,7 +98,6 @@ static int i810_setup_i2c_bus(struct i810fb_i2c_chan *chan, const char *name)
 	chan->algo.getsda               = i810i2c_getsda;
 	chan->algo.getscl               = i810i2c_getscl;
 	chan->algo.udelay               = 10;
-	chan->algo.mdelay               = 10;
         chan->algo.timeout              = (HZ/2);
         chan->algo.data                 = chan;
 
diff --git a/drivers/video/matrox/i2c-matroxfb.c b/drivers/video/matrox/i2c-matroxfb.c
index 9842042d2af5..795c1a99a680 100644
--- a/drivers/video/matrox/i2c-matroxfb.c
+++ b/drivers/video/matrox/i2c-matroxfb.c
@@ -100,7 +100,6 @@ static struct i2c_algo_bit_data matrox_i2c_algo_template =
 	.getsda		= matroxfb_gpio_getsda,
 	.getscl		= matroxfb_gpio_getscl,
 	.udelay		= 10,
-	.mdelay		= 10,
 	.timeout	= 100,
 };
 
diff --git a/drivers/video/savage/savagefb-i2c.c b/drivers/video/savage/savagefb-i2c.c
index e83befd16d63..d7d810dbf0bd 100644
--- a/drivers/video/savage/savagefb-i2c.c
+++ b/drivers/video/savage/savagefb-i2c.c
@@ -148,7 +148,6 @@ static int savage_setup_i2c_bus(struct savagefb_i2c_chan *chan,
 		chan->adapter.algo_data		= &chan->algo;
 		chan->adapter.dev.parent	= &chan->par->pcidev->dev;
 		chan->algo.udelay		= 40;
-		chan->algo.mdelay               = 5;
 		chan->algo.timeout		= 20;
 		chan->algo.data 		= chan;
 
diff --git a/include/linux/i2c-algo-bit.h b/include/linux/i2c-algo-bit.h
index c0e7fab28ce3..c8f8df25c7e0 100644
--- a/include/linux/i2c-algo-bit.h
+++ b/include/linux/i2c-algo-bit.h
@@ -40,7 +40,6 @@ struct i2c_algo_bit_data {
 	/* local settings */
 	int udelay;		/* half-clock-cycle time in microsecs */
 				/* i.e. clock is (500 / udelay) KHz */
-	int mdelay;		/* in millisecs, unused */
 	int timeout;		/* in jiffies */
 };
 
-- 
cgit v1.2.3


From 9b4ccb86b4abe644ffd218720da2f942b6a20fc2 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sun, 3 Sep 2006 22:22:50 +0200
Subject: i2c-algo-pcf: Discard the mdelay data struct member

i2c-algo-pcf: Discard the mdelay data struct member

Just as i2c-algo-bit, i2c-algo-pcf has an unused mdelay struct member,
which we can get rid of to spare some code and memory.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-elektor.c | 1 -
 include/linux/i2c-algo-pcf.h     | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include')

diff --git a/drivers/i2c/busses/i2c-elektor.c b/drivers/i2c/busses/i2c-elektor.c
index 59f8308c2356..caa8e5c8bfbb 100644
--- a/drivers/i2c/busses/i2c-elektor.c
+++ b/drivers/i2c/busses/i2c-elektor.c
@@ -196,7 +196,6 @@ static struct i2c_algo_pcf_data pcf_isa_data = {
 	.getclock   = pcf_isa_getclock,
 	.waitforpin = pcf_isa_waitforpin,
 	.udelay	    = 10,
-	.mdelay	    = 10,
 	.timeout    = 100,
 };
 
diff --git a/include/linux/i2c-algo-pcf.h b/include/linux/i2c-algo-pcf.h
index 18b0adf57a3d..9908f3fc4839 100644
--- a/include/linux/i2c-algo-pcf.h
+++ b/include/linux/i2c-algo-pcf.h
@@ -35,7 +35,6 @@ struct i2c_algo_pcf_data {
 
 	/* local settings */
 	int udelay;
-	int mdelay;
 	int timeout;
 };
 
-- 
cgit v1.2.3


From af71ff690b92894f66ccede27f731150dc10d80d Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 3 Sep 2006 22:37:11 +0200
Subject: i2c: Let drivers constify i2c_algorithm data

i2c: Let drivers constify i2c_algorithm data

Let drivers constify I2C algorithm method operations tables,
moving them from ".data" to ".rodata".

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/i2c.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index eb0628a7ecc6..23ad1ee42a4c 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -220,7 +220,7 @@ struct i2c_adapter {
 	struct module *owner;
 	unsigned int id;
 	unsigned int class;
-	struct i2c_algorithm *algo;/* the algorithm to access the bus	*/
+	const struct i2c_algorithm *algo; /* the algorithm to access the bus */
 	void *algo_data;
 
 	/* --- administration stuff. */
-- 
cgit v1.2.3


From 6d3aae9d74221b00e2cbf50a353527e5a71a58ba Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:41:08 +0200
Subject: i2c: Drop unimplemented slave functions

i2c: Drop unimplemented slave functions

Drop the function declarations for slave mode support of i2c adapters.
This was never implemented, and by the time it is I bet we will want
something different anyway.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/i2c.h | 12 ------------
 1 file changed, 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 23ad1ee42a4c..9b5d04768c2c 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -64,14 +64,6 @@ extern int i2c_master_recv(struct i2c_client *,char* ,int);
  */
 extern int i2c_transfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num);
 
-/*
- * Some adapter types (i.e. PCF 8584 based ones) may support slave behaviuor. 
- * This is not tested/implemented yet and will change in the future.
- */
-extern int i2c_slave_send(struct i2c_client *,char*,int);
-extern int i2c_slave_recv(struct i2c_client *,char*,int);
-
-
 
 /* This is the very generalized SMBus access routine. You probably do not
    want to use this, though; one of the functions below may be much easier,
@@ -201,10 +193,6 @@ struct i2c_algorithm {
 	                   unsigned short flags, char read_write,
 	                   u8 command, int size, union i2c_smbus_data * data);
 
-	/* --- these optional/future use for some adapter types.*/
-	int (*slave_send)(struct i2c_adapter *,char*,int);
-	int (*slave_recv)(struct i2c_adapter *,char*,int);
-
 	/* --- ioctl like call to set div. parameters. */
 	int (*algo_control)(struct i2c_adapter *, unsigned int, unsigned long);
 
-- 
cgit v1.2.3


From 46ff34633ed09f36ebc4b5c40ac37e592172df74 Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:55:24 -0400
Subject: MSI: Rename PCI_CAP_ID_HT_IRQCONF into PCI_CAP_ID_HT

0x08 is the HT capability, while PCI_CAP_ID_HT_IRQCONF would be
the subtype 0x80 that mpic_scan_ht_pic() uses.
Rename PCI_CAP_ID_HT_IRQCONF into PCI_CAP_ID_HT.

And by the way, use it in the ipath driver instead of defining its
own HT_CAPABILITY_ID.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/powerpc/sysdev/mpic.c                  | 2 +-
 drivers/infiniband/hw/ipath/ipath_iba6110.c | 5 ++---
 include/linux/pci_regs.h                    | 2 +-
 3 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index b604926401f5..723972bb5bd9 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -339,7 +339,7 @@ static void __init mpic_scan_ht_pic(struct mpic *mpic, u8 __iomem *devbase,
 	for (pos = readb(devbase + PCI_CAPABILITY_LIST); pos != 0;
 	     pos = readb(devbase + pos + PCI_CAP_LIST_NEXT)) {
 		u8 id = readb(devbase + pos + PCI_CAP_LIST_ID);
-		if (id == PCI_CAP_ID_HT_IRQCONF) {
+		if (id == PCI_CAP_ID_HT) {
 			id = readb(devbase + pos + 3);
 			if (id == 0x80)
 				break;
diff --git a/drivers/infiniband/hw/ipath/ipath_iba6110.c b/drivers/infiniband/hw/ipath/ipath_iba6110.c
index bf2455a6d562..5c9b509e40e4 100644
--- a/drivers/infiniband/hw/ipath/ipath_iba6110.c
+++ b/drivers/infiniband/hw/ipath/ipath_iba6110.c
@@ -742,7 +742,6 @@ static int ipath_setup_ht_reset(struct ipath_devdata *dd)
 	return 0;
 }
 
-#define HT_CAPABILITY_ID   0x08	/* HT capabilities not defined in kernel */
 #define HT_INTR_DISC_CONFIG  0x80	/* HT interrupt and discovery cap */
 #define HT_INTR_REG_INDEX    2	/* intconfig requires indirect accesses */
 
@@ -973,7 +972,7 @@ static int ipath_setup_ht_config(struct ipath_devdata *dd,
 	 * do this early, before we ever enable errors or hardware errors,
 	 * mostly to avoid causing the chip to enter freeze mode.
 	 */
-	pos = pci_find_capability(pdev, HT_CAPABILITY_ID);
+	pos = pci_find_capability(pdev, PCI_CAP_ID_HT);
 	if (!pos) {
 		ipath_dev_err(dd, "Couldn't find HyperTransport "
 			      "capability; no interrupts\n");
@@ -996,7 +995,7 @@ static int ipath_setup_ht_config(struct ipath_devdata *dd,
 		else if (cap_type == HT_INTR_DISC_CONFIG)
 			ihandler = set_int_handler(dd, pdev, pos);
 	} while ((pos = pci_find_next_capability(pdev, pos,
-						 HT_CAPABILITY_ID)));
+						 PCI_CAP_ID_HT)));
 
 	if (!ihandler) {
 		ipath_dev_err(dd, "Couldn't find interrupt handler in "
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index 96930cb5927c..7d0e26cba420 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -196,7 +196,7 @@
 #define  PCI_CAP_ID_MSI		0x05	/* Message Signalled Interrupts */
 #define  PCI_CAP_ID_CHSWP	0x06	/* CompactPCI HotSwap */
 #define  PCI_CAP_ID_PCIX	0x07	/* PCI-X */
-#define  PCI_CAP_ID_HT_IRQCONF	0x08	/* HyperTransport IRQ Configuration */
+#define  PCI_CAP_ID_HT		0x08	/* HyperTransport */
 #define  PCI_CAP_ID_VNDR	0x09	/* Vendor specific capability */
 #define  PCI_CAP_ID_SHPC 	0x0C	/* PCI Standard Hot-Plug Controller */
 #define  PCI_CAP_ID_EXP 	0x10	/* PCI Express */
-- 
cgit v1.2.3


From 6397c75cbc4d7dbc3d07278b57c82a47dafb21b5 Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:55:32 -0400
Subject: MSI: Blacklist PCI-E chipsets depending on Hypertransport MSI
 capability

Introduce msi_ht_cap_enabled() to check the MSI capability in the
Hypertransport configuration space.
It is used in a generic quirk quirk_msi_ht_cap() to check whether
MSI is enabled on hypertransport chipset, and a nVidia specific quirk
quirk_nvidia_ck804_msi_ht_cap() where two 2 HT MSI mappings have to
be checked.
Both quirks set the PCI_BUS_FLAGS_NO_MSI bus flag when MSI is disabled.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/quirks.c    | 59 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h |  1 +
 2 files changed, 60 insertions(+)

(limited to 'include')

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 8385b815ecb1..08cd86a6dd66 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -1703,6 +1703,65 @@ static void __devinit quirk_disable_msi(struct pci_dev *dev)
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_disable_msi);
+
+/* Go through the list of Hypertransport capabilities and
+ * return 1 if a HT MSI capability is found and enabled */
+static int __devinit msi_ht_cap_enabled(struct pci_dev *dev)
+{
+	u8 pos;
+	int ttl;
+	for (pos = pci_find_capability(dev, PCI_CAP_ID_HT), ttl = 48;
+	     pos && ttl;
+	     pos = pci_find_next_capability(dev, pos, PCI_CAP_ID_HT), ttl--) {
+		u32 cap_hdr;
+		/* MSI mapping section according to Hypertransport spec */
+		if (pci_read_config_dword(dev, pos, &cap_hdr) == 0
+		    && (cap_hdr & 0xf8000000) == 0xa8000000 /* MSI mapping */) {
+			printk(KERN_INFO "PCI: Found HT MSI mapping on %s with capability %s\n",
+			       pci_name(dev), cap_hdr & 0x10000 ? "enabled" : "disabled");
+			return (cap_hdr & 0x10000) != 0; /* MSI mapping cap enabled */
+		}
+	}
+	return 0;
+}
+
+/* Check the hypertransport MSI mapping to know whether MSI is enabled or not */
+static void __devinit quirk_msi_ht_cap(struct pci_dev *dev)
+{
+	if (dev->subordinate && !msi_ht_cap_enabled(dev)) {
+		printk(KERN_WARNING "PCI: MSI quirk detected. "
+		       "MSI disabled on chipset %s.\n",
+		       pci_name(dev));
+		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
+	}
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_HT2000_PCIE,
+			quirk_msi_ht_cap);
+
+/* The nVidia CK804 chipset may have 2 HT MSI mappings.
+ * MSI are supported if the MSI capability set in any of these mappings.
+ */
+static void __devinit quirk_nvidia_ck804_msi_ht_cap(struct pci_dev *dev)
+{
+	struct pci_dev *pdev;
+
+	if (!dev->subordinate)
+		return;
+
+	/* check HT MSI cap on this chipset and the root one.
+	 * a single one having MSI is enough to be sure that MSI are supported.
+	 */
+	pdev = pci_find_slot(dev->bus->number, 0);
+	if (dev->subordinate && !msi_ht_cap_enabled(dev)
+	    && !msi_ht_cap_enabled(pdev)) {
+		printk(KERN_WARNING "PCI: MSI quirk detected. "
+		       "MSI disabled on chipset %s.\n",
+		       pci_name(dev));
+		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
+	}
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE,
+			quirk_nvidia_ck804_msi_ht_cap);
 #endif /* CONFIG_PCI_MSI */
 
 EXPORT_SYMBOL(pcie_mch_quirk);
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 6a1e09834559..b9e263adebab 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1411,6 +1411,7 @@
 #define PCI_DEVICE_ID_SERVERWORKS_LE	  0x0009
 #define PCI_DEVICE_ID_SERVERWORKS_GCNB_LE 0x0017
 #define PCI_DEVICE_ID_SERVERWORKS_EPB	  0x0103
+#define PCI_DEVICE_ID_SERVERWORKS_HT2000_PCIE	0x0132
 #define PCI_DEVICE_ID_SERVERWORKS_OSB4	  0x0200
 #define PCI_DEVICE_ID_SERVERWORKS_CSB5	  0x0201
 #define PCI_DEVICE_ID_SERVERWORKS_CSB6    0x0203
-- 
cgit v1.2.3


From 6c2b374d74857e892080ee726184ec1d15e7d4e4 Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin.zhang@intel.com>
Date: Mon, 31 Jul 2006 15:21:33 +0800
Subject: PCI-Express AER implemetation: AER core and aerdriver

Patch 3 implements the core part of PCI-Express AER and aerdrv
port service driver.

When a root port service device is probed, the aerdrv will call
request_irq to register irq handler for AER error interrupt.

When a device sends an PCI-Express error message to the root port,
the root port will trigger an interrupt, by either MSI or IO-APIC,
then kernel would run the irq handler. The handler collects root
error status register and schedules a work. The work will call
the core part to process the error based on its type
(Correctable/non-fatal/fatal).

As for Correctable errors, the patch chooses to just clear the correctable
error status register of the device.

As for the non-fatal error, the patch follows generic PCI error handler
rules to call the error callback functions of the endpoint's driver. If
the device is a bridge, the patch chooses to broadcast the error to
downstream devices.

As for the fatal error, the patch resets the pci-express link and
follows generic PCI error handler rules to call the error callback
functions of the endpoint's driver. If the device is a bridge, the patch
chooses to broadcast the error to downstream devices.

Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pcie/Kconfig               |   1 +
 drivers/pci/pcie/Makefile              |   3 +
 drivers/pci/pcie/aer/Kconfig           |  12 +
 drivers/pci/pcie/aer/Makefile          |   8 +
 drivers/pci/pcie/aer/aerdrv.c          | 346 +++++++++++++++
 drivers/pci/pcie/aer/aerdrv.h          | 125 ++++++
 drivers/pci/pcie/aer/aerdrv_acpi.c     |  68 +++
 drivers/pci/pcie/aer/aerdrv_core.c     | 757 +++++++++++++++++++++++++++++++++
 drivers/pci/pcie/aer/aerdrv_errprint.c | 248 +++++++++++
 include/linux/aer.h                    |  24 ++
 include/linux/pcieport_if.h            |   6 +
 11 files changed, 1598 insertions(+)
 create mode 100644 drivers/pci/pcie/aer/Kconfig
 create mode 100644 drivers/pci/pcie/aer/Makefile
 create mode 100644 drivers/pci/pcie/aer/aerdrv.c
 create mode 100644 drivers/pci/pcie/aer/aerdrv.h
 create mode 100644 drivers/pci/pcie/aer/aerdrv_acpi.c
 create mode 100644 drivers/pci/pcie/aer/aerdrv_core.c
 create mode 100644 drivers/pci/pcie/aer/aerdrv_errprint.c
 create mode 100644 include/linux/aer.h

(limited to 'include')

diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig
index 1012db8b8b2c..0ad92a8ad8b1 100644
--- a/drivers/pci/pcie/Kconfig
+++ b/drivers/pci/pcie/Kconfig
@@ -34,3 +34,4 @@ config HOTPLUG_PCI_PCIE_POLL_EVENT_MODE
 	   
 	  When in doubt, say N.
 
+source "drivers/pci/pcie/aer/Kconfig"
diff --git a/drivers/pci/pcie/Makefile b/drivers/pci/pcie/Makefile
index 984fa87283e3..e00fb99acf44 100644
--- a/drivers/pci/pcie/Makefile
+++ b/drivers/pci/pcie/Makefile
@@ -5,3 +5,6 @@
 pcieportdrv-y			:= portdrv_core.o portdrv_pci.o portdrv_bus.o
 
 obj-$(CONFIG_PCIEPORTBUS)	+= pcieportdrv.o
+
+# Build PCI Express AER if needed
+obj-$(CONFIG_PCIEAER)		+= aer/
diff --git a/drivers/pci/pcie/aer/Kconfig b/drivers/pci/pcie/aer/Kconfig
new file mode 100644
index 000000000000..3f37a60a6438
--- /dev/null
+++ b/drivers/pci/pcie/aer/Kconfig
@@ -0,0 +1,12 @@
+#
+# PCI Express Root Port Device AER Configuration
+#
+
+config PCIEAER
+	boolean "Root Port Advanced Error Reporting support"
+	depends on PCIEPORTBUS && ACPI
+	default y
+	help
+	  This enables PCI Express Root Port Advanced Error Reporting
+	  (AER) driver support. Error reporting messages sent to Root
+	  Port will be handled by PCI Express AER driver.
diff --git a/drivers/pci/pcie/aer/Makefile b/drivers/pci/pcie/aer/Makefile
new file mode 100644
index 000000000000..15a4f40d520b
--- /dev/null
+++ b/drivers/pci/pcie/aer/Makefile
@@ -0,0 +1,8 @@
+#
+# Makefile for PCI-Express Root Port Advanced Error Reporting Driver
+#
+
+obj-$(CONFIG_PCIEAER) += aerdriver.o
+
+aerdriver-objs := aerdrv_errprint.o aerdrv_core.o aerdrv.o aerdrv_acpi.o
+
diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c
new file mode 100644
index 000000000000..0d4ac027d53e
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv.c
@@ -0,0 +1,346 @@
+/*
+ * drivers/pci/pcie/aer/aerdrv.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * This file implements the AER root port service driver. The driver will
+ * register an irq handler. When root port triggers an AER interrupt, the irq
+ * handler will collect root port status and schedule a work.
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/pcieport_if.h>
+
+#include "aerdrv.h"
+
+/*
+ * Version Information
+ */
+#define DRIVER_VERSION "v1.0"
+#define DRIVER_AUTHOR "tom.l.nguyen@intel.com"
+#define DRIVER_DESC "Root Port Advanced Error Reporting Driver"
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+static int __devinit aer_probe (struct pcie_device *dev,
+	const struct pcie_port_service_id *id );
+static void aer_remove(struct pcie_device *dev);
+static int aer_suspend(struct pcie_device *dev, pm_message_t state)
+{return 0;}
+static int aer_resume(struct pcie_device *dev) {return 0;}
+static pci_ers_result_t aer_error_detected(struct pci_dev *dev,
+	enum pci_channel_state error);
+static void aer_error_resume(struct pci_dev *dev);
+static pci_ers_result_t aer_root_reset(struct pci_dev *dev);
+
+/*
+ * PCI Express bus's AER Root service driver data structure
+ */
+static struct pcie_port_service_id aer_id[] = {
+	{
+	.vendor 	= PCI_ANY_ID,
+	.device 	= PCI_ANY_ID,
+	.port_type 	= PCIE_RC_PORT,
+	.service_type 	= PCIE_PORT_SERVICE_AER,
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_error_handlers aer_error_handlers = {
+	.error_detected = aer_error_detected,
+	.resume = aer_error_resume,
+};
+
+static struct pcie_port_service_driver aerdrv = {
+	.name		= "aer",
+	.id_table	= &aer_id[0],
+
+	.probe		= aer_probe,
+	.remove		= aer_remove,
+
+	.suspend	= aer_suspend,
+	.resume		= aer_resume,
+
+	.err_handler	= &aer_error_handlers,
+
+	.reset_link	= aer_root_reset,
+};
+
+/**
+ * aer_irq - Root Port's ISR
+ * @irq: IRQ assigned to Root Port
+ * @context: pointer to Root Port data structure
+ * @r: pointer struct pt_regs
+ *
+ * Invoked when Root Port detects AER messages.
+ **/
+static irqreturn_t aer_irq(int irq, void *context, struct pt_regs * r)
+{
+	unsigned int status, id;
+	struct pcie_device *pdev = (struct pcie_device *)context;
+	struct aer_rpc *rpc = get_service_data(pdev);
+	int next_prod_idx;
+	unsigned long flags;
+	int pos;
+
+	pos = pci_find_aer_capability(pdev->port);
+	/*
+	 * Must lock access to Root Error Status Reg, Root Error ID Reg,
+	 * and Root error producer/consumer index
+	 */
+	spin_lock_irqsave(&rpc->e_lock, flags);
+
+	/* Read error status */
+	pci_read_config_dword(pdev->port, pos + PCI_ERR_ROOT_STATUS, &status);
+	if (!(status & ROOT_ERR_STATUS_MASKS)) {
+		spin_unlock_irqrestore(&rpc->e_lock, flags);
+		return IRQ_NONE;
+	}
+
+	/* Read error source and clear error status */
+	pci_read_config_dword(pdev->port, pos + PCI_ERR_ROOT_COR_SRC, &id);
+	pci_write_config_dword(pdev->port, pos + PCI_ERR_ROOT_STATUS, status);
+
+	/* Store error source for later DPC handler */
+	next_prod_idx = rpc->prod_idx + 1;
+	if (next_prod_idx == AER_ERROR_SOURCES_MAX)
+		next_prod_idx = 0;
+	if (next_prod_idx == rpc->cons_idx) {
+		/*
+		 * Error Storm Condition - possibly the same error occurred.
+		 * Drop the error.
+		 */
+		spin_unlock_irqrestore(&rpc->e_lock, flags);
+		return IRQ_HANDLED;
+	}
+	rpc->e_sources[rpc->prod_idx].status =  status;
+	rpc->e_sources[rpc->prod_idx].id = id;
+	rpc->prod_idx = next_prod_idx;
+	spin_unlock_irqrestore(&rpc->e_lock, flags);
+
+	/*  Invoke DPC handler */
+	schedule_work(&rpc->dpc_handler);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * aer_alloc_rpc - allocate Root Port data structure
+ * @dev: pointer to the pcie_dev data structure
+ *
+ * Invoked when Root Port's AER service is loaded.
+ **/
+static struct aer_rpc* aer_alloc_rpc(struct pcie_device *dev)
+{
+	struct aer_rpc *rpc;
+
+	if (!(rpc = (struct aer_rpc *)kmalloc(sizeof(struct aer_rpc),
+		GFP_KERNEL)))
+		return NULL;
+
+	memset(rpc, 0, sizeof(struct aer_rpc));
+	/*
+	 * Initialize Root lock access, e_lock, to Root Error Status Reg,
+	 * Root Error ID Reg, and Root error producer/consumer index.
+	 */
+	rpc->e_lock = SPIN_LOCK_UNLOCKED;
+
+	rpc->rpd = dev;
+	INIT_WORK(&rpc->dpc_handler, aer_isr, (void *)dev);
+	rpc->prod_idx = rpc->cons_idx = 0;
+	mutex_init(&rpc->rpc_mutex);
+	init_waitqueue_head(&rpc->wait_release);
+
+	/* Use PCIE bus function to store rpc into PCIE device */
+	set_service_data(dev, rpc);
+
+	return rpc;
+}
+
+/**
+ * aer_remove - clean up resources
+ * @dev: pointer to the pcie_dev data structure
+ *
+ * Invoked when PCI Express bus unloads or AER probe fails.
+ **/
+static void aer_remove(struct pcie_device *dev)
+{
+	struct aer_rpc *rpc = get_service_data(dev);
+
+	if (rpc) {
+		/* If register interrupt service, it must be free. */
+		if (rpc->isr)
+			free_irq(dev->irq, dev);
+
+		wait_event(rpc->wait_release, rpc->prod_idx == rpc->cons_idx);
+
+		aer_delete_rootport(rpc);
+		set_service_data(dev, NULL);
+	}
+}
+
+/**
+ * aer_probe - initialize resources
+ * @dev: pointer to the pcie_dev data structure
+ * @id: pointer to the service id data structure
+ *
+ * Invoked when PCI Express bus loads AER service driver.
+ **/
+static int __devinit aer_probe (struct pcie_device *dev,
+				const struct pcie_port_service_id *id )
+{
+	int status;
+	struct aer_rpc *rpc;
+	struct device *device = &dev->device;
+
+	/* Init */
+	if ((status = aer_init(dev)))
+		return status;
+
+	/* Alloc rpc data structure */
+	if (!(rpc = aer_alloc_rpc(dev))) {
+		printk(KERN_DEBUG "%s: Alloc rpc fails on PCIE device[%s]\n",
+			__FUNCTION__, device->bus_id);
+		aer_remove(dev);
+		return -ENOMEM;
+	}
+
+	/* Request IRQ ISR */
+	if ((status = request_irq(dev->irq, aer_irq, SA_SHIRQ, "aerdrv",
+				dev))) {
+		printk(KERN_DEBUG "%s: Request ISR fails on PCIE device[%s]\n",
+			__FUNCTION__, device->bus_id);
+		aer_remove(dev);
+		return status;
+	}
+
+	rpc->isr = 1;
+
+	aer_enable_rootport(rpc);
+
+	return status;
+}
+
+/**
+ * aer_root_reset - reset link on Root Port
+ * @dev: pointer to Root Port's pci_dev data structure
+ *
+ * Invoked by Port Bus driver when performing link reset at Root Port.
+ **/
+static pci_ers_result_t aer_root_reset(struct pci_dev *dev)
+{
+	u16 p2p_ctrl;
+	u32 status;
+	int pos;
+
+	pos = pci_find_aer_capability(dev);
+
+	/* Disable Root's interrupt in response to error messages */
+	pci_write_config_dword(dev, pos + PCI_ERR_ROOT_COMMAND, 0);
+
+	/* Assert Secondary Bus Reset */
+	pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &p2p_ctrl);
+	p2p_ctrl |= PCI_CB_BRIDGE_CTL_CB_RESET;
+	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, p2p_ctrl);
+
+	/* De-assert Secondary Bus Reset */
+	p2p_ctrl &= ~PCI_CB_BRIDGE_CTL_CB_RESET;
+	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, p2p_ctrl);
+
+	/*
+	 * System software must wait for at least 100ms from the end
+	 * of a reset of one or more device before it is permitted
+	 * to issue Configuration Requests to those devices.
+	 */
+	msleep(200);
+	printk(KERN_DEBUG "Complete link reset at Root[%s]\n", dev->dev.bus_id);
+
+	/* Enable Root Port's interrupt in response to error messages */
+	pci_read_config_dword(dev, pos + PCI_ERR_ROOT_STATUS, &status);
+	pci_write_config_dword(dev, pos + PCI_ERR_ROOT_STATUS, status);
+	pci_write_config_dword(dev,
+		pos + PCI_ERR_ROOT_COMMAND,
+		ROOT_PORT_INTR_ON_MESG_MASK);
+
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+/**
+ * aer_error_detected - update severity status
+ * @dev: pointer to Root Port's pci_dev data structure
+ * @error: error severity being notified by port bus
+ *
+ * Invoked by Port Bus driver during error recovery.
+ **/
+static pci_ers_result_t aer_error_detected(struct pci_dev *dev,
+			enum pci_channel_state error)
+{
+	/* Root Port has no impact. Always recovers. */
+	return PCI_ERS_RESULT_CAN_RECOVER;
+}
+
+/**
+ * aer_error_resume - clean up corresponding error status bits
+ * @dev: pointer to Root Port's pci_dev data structure
+ *
+ * Invoked by Port Bus driver during nonfatal recovery.
+ **/
+static void aer_error_resume(struct pci_dev *dev)
+{
+	int pos;
+	u32 status, mask;
+	u16 reg16;
+
+	/* Clean up Root device status */
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &reg16);
+	pci_write_config_word(dev, pos + PCI_EXP_DEVSTA, reg16);
+
+	/* Clean AER Root Error Status */
+	pos = pci_find_aer_capability(dev);
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, &mask);
+	if (dev->error_state == pci_channel_io_normal)
+		status &= ~mask; /* Clear corresponding nonfatal bits */
+	else
+		status &= mask; /* Clear corresponding fatal bits */
+	pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, status);
+}
+
+/**
+ * aer_service_init - register AER root service driver
+ *
+ * Invoked when AER root service driver is loaded.
+ **/
+static int __init aer_service_init(void)
+{
+	return pcie_port_service_register(&aerdrv);
+}
+
+/**
+ * aer_service_exit - unregister AER root service driver
+ *
+ * Invoked when AER root service driver is unloaded.
+ **/
+static void __exit aer_service_exit(void)
+{
+	pcie_port_service_unregister(&aerdrv);
+}
+
+module_init(aer_service_init);
+module_exit(aer_service_exit);
diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
new file mode 100644
index 000000000000..daf0cad88fc8
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -0,0 +1,125 @@
+/*
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#ifndef _AERDRV_H_
+#define _AERDRV_H_
+
+#include <linux/pcieport_if.h>
+#include <linux/aer.h>
+
+#define AER_NONFATAL			0
+#define AER_FATAL			1
+#define AER_CORRECTABLE			2
+#define AER_UNCORRECTABLE		4
+#define AER_ERROR_MASK			0x001fffff
+#define AER_ERROR(d)			(d & AER_ERROR_MASK)
+
+#define OSC_METHOD_RUN_SUCCESS		0
+#define OSC_METHOD_NOT_SUPPORTED	1
+#define OSC_METHOD_RUN_FAILURE		2
+
+/* Root Error Status Register Bits */
+#define ROOT_ERR_STATUS_MASKS			0x0f
+
+#define SYSTEM_ERROR_INTR_ON_MESG_MASK	(PCI_EXP_RTCTL_SECEE|	\
+					PCI_EXP_RTCTL_SENFEE|	\
+					PCI_EXP_RTCTL_SEFEE)
+#define ROOT_PORT_INTR_ON_MESG_MASK	(PCI_ERR_ROOT_CMD_COR_EN|	\
+					PCI_ERR_ROOT_CMD_NONFATAL_EN|	\
+					PCI_ERR_ROOT_CMD_FATAL_EN)
+#define ERR_COR_ID(d)			(d & 0xffff)
+#define ERR_UNCOR_ID(d)			(d >> 16)
+
+#define AER_SUCCESS			0
+#define AER_UNSUCCESS			1
+#define AER_ERROR_SOURCES_MAX		100
+
+#define AER_LOG_TLP_MASKS		(PCI_ERR_UNC_POISON_TLP|	\
+					PCI_ERR_UNC_ECRC|		\
+					PCI_ERR_UNC_UNSUP|		\
+					PCI_ERR_UNC_COMP_ABORT|		\
+					PCI_ERR_UNC_UNX_COMP|		\
+					PCI_ERR_UNC_MALF_TLP)
+
+/* AER Error Info Flags */
+#define AER_TLP_HEADER_VALID_FLAG	0x00000001
+#define AER_MULTI_ERROR_VALID_FLAG	0x00000002
+
+#define ERR_CORRECTABLE_ERROR_MASK	0x000031c1
+#define ERR_UNCORRECTABLE_ERROR_MASK	0x001ff010
+
+struct header_log_regs {
+	unsigned int dw0;
+	unsigned int dw1;
+	unsigned int dw2;
+	unsigned int dw3;
+};
+
+struct aer_err_info {
+	int severity;			/* 0:NONFATAL | 1:FATAL | 2:COR */
+	int flags;
+	unsigned int status;		/* COR/UNCOR Error Status */
+	struct header_log_regs tlp; 	/* TLP Header */
+};
+
+struct aer_err_source {
+	unsigned int status;
+	unsigned int id;
+};
+
+struct aer_rpc {
+	struct pcie_device *rpd;	/* Root Port device */
+	struct work_struct dpc_handler;
+	struct aer_err_source e_sources[AER_ERROR_SOURCES_MAX];
+	unsigned short prod_idx;	/* Error Producer Index */
+	unsigned short cons_idx;	/* Error Consumer Index */
+	int isr;
+	spinlock_t e_lock;		/*
+					 * Lock access to Error Status/ID Regs
+					 * and error producer/consumer index
+					 */
+	struct mutex rpc_mutex;		/*
+					 * only one thread could do
+					 * recovery on the same
+					 * root port hierachy
+					 */
+	wait_queue_head_t wait_release;
+};
+
+struct aer_broadcast_data {
+	enum pci_channel_state state;
+	enum pci_ers_result result;
+};
+
+static inline pci_ers_result_t merge_result(enum pci_ers_result orig,
+		enum pci_ers_result new)
+{
+	switch (orig) {
+	case PCI_ERS_RESULT_CAN_RECOVER:
+	case PCI_ERS_RESULT_RECOVERED:
+		orig = new;
+		break;
+	case PCI_ERS_RESULT_DISCONNECT:
+		if (new == PCI_ERS_RESULT_NEED_RESET)
+			orig = new;
+		break;
+	default:
+		break;
+	}
+
+	return orig;
+}
+
+extern struct bus_type pcie_port_bus_type;
+extern void aer_enable_rootport(struct aer_rpc *rpc);
+extern void aer_delete_rootport(struct aer_rpc *rpc);
+extern int aer_init(struct pcie_device *dev);
+extern void aer_isr(void *context);
+extern void aer_print_error(struct pci_dev *dev, struct aer_err_info *info);
+extern int aer_osc_setup(struct pci_dev *dev);
+
+#endif //_AERDRV_H_
diff --git a/drivers/pci/pcie/aer/aerdrv_acpi.c b/drivers/pci/pcie/aer/aerdrv_acpi.c
new file mode 100644
index 000000000000..fa68e89ebec9
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv_acpi.c
@@ -0,0 +1,68 @@
+/*
+ * Access ACPI _OSC method
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+#include <linux/acpi.h>
+#include <linux/pci-acpi.h>
+#include <linux/delay.h>
+#include "aerdrv.h"
+
+/**
+ * aer_osc_setup - run ACPI _OSC method
+ *
+ * Return:
+ *	Zero if success. Nonzero for otherwise.
+ *
+ * Invoked when PCIE bus loads AER service driver. To avoid conflict with
+ * BIOS AER support requires BIOS to yield AER control to OS native driver.
+ **/
+int aer_osc_setup(struct pci_dev *dev)
+{
+	int retval = OSC_METHOD_RUN_SUCCESS;
+	acpi_status status;
+	acpi_handle handle = DEVICE_ACPI_HANDLE(&dev->dev);
+	struct pci_dev *pdev = dev;
+	struct pci_bus *parent;
+
+	while (!handle) {
+		if (!pdev || !pdev->bus->parent)
+			break;
+		parent = pdev->bus->parent;
+		if (!parent->self)
+			/* Parent must be a host bridge */
+			handle = acpi_get_pci_rootbridge_handle(
+					pci_domain_nr(parent),
+					parent->number);
+		else
+			handle = DEVICE_ACPI_HANDLE(
+					&(parent->self->dev));
+		pdev = parent->self;
+	}
+
+	if (!handle)
+		return OSC_METHOD_NOT_SUPPORTED;
+
+	pci_osc_support_set(OSC_EXT_PCI_CONFIG_SUPPORT);
+	status = pci_osc_control_set(handle, OSC_PCI_EXPRESS_AER_CONTROL |
+		OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL);
+	if (ACPI_FAILURE(status)) {
+		if (status == AE_SUPPORT)
+			retval = OSC_METHOD_NOT_SUPPORTED;
+	 	else
+			retval = OSC_METHOD_RUN_FAILURE;
+	}
+
+	return retval;
+}
+
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
new file mode 100644
index 000000000000..5591043acea7
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -0,0 +1,757 @@
+/*
+ * drivers/pci/pcie/aer/aerdrv_core.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * This file implements the core part of PCI-Express AER. When an pci-express
+ * error is delivered, an error message will be collected and printed to
+ * console, then, an error recovery procedure will be executed by following
+ * the pci error recovery rules.
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+#include <linux/acpi.h>
+#include <linux/pci-acpi.h>
+#include <linux/delay.h>
+#include "aerdrv.h"
+
+static int forceload;
+module_param(forceload, bool, 0);
+
+#define PCI_CFG_SPACE_SIZE	(0x100)
+int pci_find_aer_capability(struct pci_dev *dev)
+{
+	int pos;
+	u32 reg32 = 0;
+
+	/* Check if it's a pci-express device */
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return 0;
+
+	/* Check if it supports pci-express AER */
+	pos = PCI_CFG_SPACE_SIZE;
+	while (pos) {
+		if (pci_read_config_dword(dev, pos, &reg32))
+			return 0;
+
+		/* some broken boards return ~0 */
+		if (reg32 == 0xffffffff)
+			return 0;
+
+		if (PCI_EXT_CAP_ID(reg32) == PCI_EXT_CAP_ID_ERR)
+			break;
+
+		pos = reg32 >> 20;
+	}
+
+	return pos;
+}
+
+int pci_enable_pcie_error_reporting(struct pci_dev *dev)
+{
+	u16 reg16 = 0;
+	int pos;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return -EIO;
+
+	pci_read_config_word(dev, pos+PCI_EXP_DEVCTL, &reg16);
+	reg16 = reg16 |
+		PCI_EXP_DEVCTL_CERE |
+		PCI_EXP_DEVCTL_NFERE |
+		PCI_EXP_DEVCTL_FERE |
+		PCI_EXP_DEVCTL_URRE;
+	pci_write_config_word(dev, pos+PCI_EXP_DEVCTL,
+			reg16);
+	return 0;
+}
+
+int pci_disable_pcie_error_reporting(struct pci_dev *dev)
+{
+	u16 reg16 = 0;
+	int pos;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return -EIO;
+
+	pci_read_config_word(dev, pos+PCI_EXP_DEVCTL, &reg16);
+	reg16 = reg16 & ~(PCI_EXP_DEVCTL_CERE |
+			PCI_EXP_DEVCTL_NFERE |
+			PCI_EXP_DEVCTL_FERE |
+			PCI_EXP_DEVCTL_URRE);
+	pci_write_config_word(dev, pos+PCI_EXP_DEVCTL,
+			reg16);
+	return 0;
+}
+
+int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
+{
+	int pos;
+	u32 status, mask;
+
+	pos = pci_find_aer_capability(dev);
+	if (!pos)
+		return -EIO;
+
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, &mask);
+	if (dev->error_state == pci_channel_io_normal)
+		status &= ~mask; /* Clear corresponding nonfatal bits */
+	else
+		status &= mask; /* Clear corresponding fatal bits */
+	pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, status);
+
+	return 0;
+}
+
+static int find_device_iter(struct device *device, void *data)
+{
+	struct pci_dev *dev;
+	u16 id = *(unsigned long *)data;
+	u8 secondary, subordinate, d_bus = id >> 8;
+
+	if (device->bus == &pci_bus_type) {
+		dev = to_pci_dev(device);
+		if (id == ((dev->bus->number << 8) | dev->devfn)) {
+			/*
+			 * Device ID match
+			 */
+			*(unsigned long*)data = (unsigned long)device;
+			return 1;
+		}
+
+		/*
+		 * If device is P2P, check if it is an upstream?
+		 */
+		if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
+			pci_read_config_byte(dev, PCI_SECONDARY_BUS,
+				&secondary);
+			pci_read_config_byte(dev, PCI_SUBORDINATE_BUS,
+				&subordinate);
+			if (d_bus >= secondary && d_bus <= subordinate) {
+				*(unsigned long*)data = (unsigned long)device;
+				return 1;
+			}
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * find_source_device - search through device hierarchy for source device
+ * @p_dev: pointer to Root Port pci_dev data structure
+ * @id: device ID of agent who sends an error message to this Root Port
+ *
+ * Invoked when error is detected at the Root Port.
+ **/
+static struct device* find_source_device(struct pci_dev *parent, u16 id)
+{
+	struct pci_dev *dev = parent;
+	struct device *device;
+	unsigned long device_addr;
+	int status;
+
+	/* Is Root Port an agent that sends error message? */
+	if (id == ((dev->bus->number << 8) | dev->devfn))
+		return &dev->dev;
+
+	do {
+		device_addr = id;
+ 		if ((status = device_for_each_child(&dev->dev,
+			&device_addr, find_device_iter))) {
+			device = (struct device*)device_addr;
+			dev = to_pci_dev(device);
+			if (id == ((dev->bus->number << 8) | dev->devfn))
+				return device;
+		}
+ 	}while (status);
+
+	return NULL;
+}
+
+static void report_error_detected(struct pci_dev *dev, void *data)
+{
+	pci_ers_result_t vote;
+	struct pci_error_handlers *err_handler;
+	struct aer_broadcast_data *result_data;
+	result_data = (struct aer_broadcast_data *) data;
+
+	dev->error_state = result_data->state;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->error_detected) {
+		if (result_data->state == pci_channel_io_frozen &&
+			!(dev->hdr_type & PCI_HEADER_TYPE_BRIDGE)) {
+			/*
+			 * In case of fatal recovery, if one of down-
+			 * stream device has no driver. We might be
+			 * unable to recover because a later insmod
+			 * of a driver for this device is unaware of
+			 * its hw state.
+			 */
+			printk(KERN_DEBUG "Device ID[%s] has %s\n",
+					dev->dev.bus_id, (dev->driver) ?
+					"no AER-aware driver" : "no driver");
+		}
+		return;
+	}
+
+	err_handler = dev->driver->err_handler;
+	vote = err_handler->error_detected(dev, result_data->state);
+	result_data->result = merge_result(result_data->result, vote);
+	return;
+}
+
+static void report_mmio_enabled(struct pci_dev *dev, void *data)
+{
+	pci_ers_result_t vote;
+	struct pci_error_handlers *err_handler;
+	struct aer_broadcast_data *result_data;
+	result_data = (struct aer_broadcast_data *) data;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->mmio_enabled)
+		return;
+
+	err_handler = dev->driver->err_handler;
+	vote = err_handler->mmio_enabled(dev);
+	result_data->result = merge_result(result_data->result, vote);
+	return;
+}
+
+static void report_slot_reset(struct pci_dev *dev, void *data)
+{
+	pci_ers_result_t vote;
+	struct pci_error_handlers *err_handler;
+	struct aer_broadcast_data *result_data;
+	result_data = (struct aer_broadcast_data *) data;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->slot_reset)
+		return;
+
+	err_handler = dev->driver->err_handler;
+	vote = err_handler->slot_reset(dev);
+	result_data->result = merge_result(result_data->result, vote);
+	return;
+}
+
+static void report_resume(struct pci_dev *dev, void *data)
+{
+	struct pci_error_handlers *err_handler;
+
+	dev->error_state = pci_channel_io_normal;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->slot_reset)
+		return;
+
+	err_handler = dev->driver->err_handler;
+	err_handler->resume(dev);
+	return;
+}
+
+/**
+ * broadcast_error_message - handle message broadcast to downstream drivers
+ * @device: pointer to from where in a hierarchy message is broadcasted down
+ * @api: callback to be broadcasted
+ * @state: error state
+ *
+ * Invoked during error recovery process. Once being invoked, the content
+ * of error severity will be broadcasted to all downstream drivers in a
+ * hierarchy in question.
+ **/
+static pci_ers_result_t broadcast_error_message(struct pci_dev *dev,
+	enum pci_channel_state state,
+	char *error_mesg,
+	void (*cb)(struct pci_dev *, void *))
+{
+	struct aer_broadcast_data result_data;
+
+	printk(KERN_DEBUG "Broadcast %s message\n", error_mesg);
+	result_data.state = state;
+	if (cb == report_error_detected)
+		result_data.result = PCI_ERS_RESULT_CAN_RECOVER;
+	else
+		result_data.result = PCI_ERS_RESULT_RECOVERED;
+
+	if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
+		/*
+		 * If the error is reported by a bridge, we think this error
+		 * is related to the downstream link of the bridge, so we
+		 * do error recovery on all subordinates of the bridge instead
+		 * of the bridge and clear the error status of the bridge.
+		 */
+		if (cb == report_error_detected)
+			dev->error_state = state;
+		pci_walk_bus(dev->subordinate, cb, &result_data);
+		if (cb == report_resume) {
+			pci_cleanup_aer_uncorrect_error_status(dev);
+			dev->error_state = pci_channel_io_normal;
+		}
+	}
+	else {
+		/*
+		 * If the error is reported by an end point, we think this
+		 * error is related to the upstream link of the end point.
+		 */
+		pci_walk_bus(dev->bus, cb, &result_data);
+	}
+
+	return result_data.result;
+}
+
+struct find_aer_service_data {
+	struct pcie_port_service_driver *aer_driver;
+	int is_downstream;
+};
+
+static int find_aer_service_iter(struct device *device, void *data)
+{
+	struct device_driver *driver;
+	struct pcie_port_service_driver *service_driver;
+	struct pcie_device *pcie_dev;
+	struct find_aer_service_data *result;
+
+	result = (struct find_aer_service_data *) data;
+
+	if (device->bus == &pcie_port_bus_type) {
+		pcie_dev = to_pcie_device(device);
+		if (pcie_dev->id.port_type == PCIE_SW_DOWNSTREAM_PORT)
+			result->is_downstream = 1;
+
+		driver = device->driver;
+		if (driver) {
+			service_driver = to_service_driver(driver);
+			if (service_driver->id_table->service_type ==
+					PCIE_PORT_SERVICE_AER) {
+				result->aer_driver = service_driver;
+				return 1;
+			}
+		}
+	}
+
+	return 0;
+}
+
+static void find_aer_service(struct pci_dev *dev,
+		struct find_aer_service_data *data)
+{
+	device_for_each_child(&dev->dev, data, find_aer_service_iter);
+}
+
+static pci_ers_result_t reset_link(struct pcie_device *aerdev,
+		struct pci_dev *dev)
+{
+	struct pci_dev *udev;
+	pci_ers_result_t status;
+	struct find_aer_service_data data;
+
+	if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE)
+		udev = dev;
+	else
+		udev= dev->bus->self;
+
+	data.is_downstream = 0;
+	data.aer_driver = NULL;
+	find_aer_service(udev, &data);
+
+	/*
+	 * Use the aer driver of the error agent firstly.
+	 * If it hasn't the aer driver, use the root port's
+	 */
+	if (!data.aer_driver || !data.aer_driver->reset_link) {
+		if (data.is_downstream &&
+			aerdev->device.driver &&
+			to_service_driver(aerdev->device.driver)->reset_link) {
+			data.aer_driver =
+				to_service_driver(aerdev->device.driver);
+		} else {
+			printk(KERN_DEBUG "No link-reset support to Device ID"
+				"[%s]\n",
+				dev->dev.bus_id);
+			return PCI_ERS_RESULT_DISCONNECT;
+		}
+	}
+
+	status = data.aer_driver->reset_link(udev);
+	if (status != PCI_ERS_RESULT_RECOVERED) {
+		printk(KERN_DEBUG "Link reset at upstream Device ID"
+			"[%s] failed\n",
+			udev->dev.bus_id);
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	return status;
+}
+
+/**
+ * do_recovery - handle nonfatal/fatal error recovery process
+ * @aerdev: pointer to a pcie_device data structure of root port
+ * @dev: pointer to a pci_dev data structure of agent detecting an error
+ * @severity: error severity type
+ *
+ * Invoked when an error is nonfatal/fatal. Once being invoked, broadcast
+ * error detected message to all downstream drivers within a hierarchy in
+ * question and return the returned code.
+ **/
+static pci_ers_result_t do_recovery(struct pcie_device *aerdev,
+		struct pci_dev *dev,
+		int severity)
+{
+	pci_ers_result_t status, result = PCI_ERS_RESULT_RECOVERED;
+	enum pci_channel_state state;
+
+	if (severity == AER_FATAL)
+		state = pci_channel_io_frozen;
+	else
+		state = pci_channel_io_normal;
+
+	status = broadcast_error_message(dev,
+			state,
+			"error_detected",
+			report_error_detected);
+
+	if (severity == AER_FATAL) {
+		result = reset_link(aerdev, dev);
+		if (result != PCI_ERS_RESULT_RECOVERED) {
+			/* TODO: Should panic here? */
+			return result;
+		}
+	}
+
+	if (status == PCI_ERS_RESULT_CAN_RECOVER)
+		status = broadcast_error_message(dev,
+				state,
+				"mmio_enabled",
+				report_mmio_enabled);
+
+	if (status == PCI_ERS_RESULT_NEED_RESET) {
+		/*
+		 * TODO: Should call platform-specific
+		 * functions to reset slot before calling
+		 * drivers' slot_reset callbacks?
+		 */
+		status = broadcast_error_message(dev,
+				state,
+				"slot_reset",
+				report_slot_reset);
+	}
+
+	if (status == PCI_ERS_RESULT_RECOVERED)
+		broadcast_error_message(dev,
+				state,
+				"resume",
+				report_resume);
+
+	return status;
+}
+
+/**
+ * handle_error_source - handle logging error into an event log
+ * @aerdev: pointer to pcie_device data structure of the root port
+ * @dev: pointer to pci_dev data structure of error source device
+ * @info: comprehensive error information
+ *
+ * Invoked when an error being detected by Root Port.
+ **/
+static void handle_error_source(struct pcie_device * aerdev,
+	struct pci_dev *dev,
+	struct aer_err_info info)
+{
+	pci_ers_result_t status = 0;
+	int pos;
+
+	if (info.severity == AER_CORRECTABLE) {
+		/*
+		 * Correctable error does not need software intevention.
+		 * No need to go through error recovery process.
+		 */
+		pos = pci_find_aer_capability(dev);
+		if (pos)
+			pci_write_config_dword(dev, pos + PCI_ERR_COR_STATUS,
+					info.status);
+	} else {
+		status = do_recovery(aerdev, dev, info.severity);
+		if (status == PCI_ERS_RESULT_RECOVERED) {
+			printk(KERN_DEBUG "AER driver successfully recovered\n");
+		} else {
+			/* TODO: Should kernel panic here? */
+			printk(KERN_DEBUG "AER driver didn't recover\n");
+		}
+	}
+}
+
+/**
+ * aer_enable_rootport - enable Root Port's interrupts when receiving messages
+ * @rpc: pointer to a Root Port data structure
+ *
+ * Invoked when PCIE bus loads AER service driver.
+ **/
+void aer_enable_rootport(struct aer_rpc *rpc)
+{
+	struct pci_dev *pdev = rpc->rpd->port;
+	int pos, aer_pos;
+	u16 reg16;
+	u32 reg32;
+
+	pos = pci_find_capability(pdev, PCI_CAP_ID_EXP);
+	/* Clear PCIE Capability's Device Status */
+	pci_read_config_word(pdev, pos+PCI_EXP_DEVSTA, &reg16);
+	pci_write_config_word(pdev, pos+PCI_EXP_DEVSTA, reg16);
+
+	/* Disable system error generation in response to error messages */
+	pci_read_config_word(pdev, pos + PCI_EXP_RTCTL, &reg16);
+	reg16 &= ~(SYSTEM_ERROR_INTR_ON_MESG_MASK);
+	pci_write_config_word(pdev, pos + PCI_EXP_RTCTL, reg16);
+
+	aer_pos = pci_find_aer_capability(pdev);
+	/* Clear error status */
+	pci_read_config_dword(pdev, aer_pos + PCI_ERR_ROOT_STATUS, &reg32);
+	pci_write_config_dword(pdev, aer_pos + PCI_ERR_ROOT_STATUS, reg32);
+	pci_read_config_dword(pdev, aer_pos + PCI_ERR_COR_STATUS, &reg32);
+	pci_write_config_dword(pdev, aer_pos + PCI_ERR_COR_STATUS, reg32);
+	pci_read_config_dword(pdev, aer_pos + PCI_ERR_UNCOR_STATUS, &reg32);
+	pci_write_config_dword(pdev, aer_pos + PCI_ERR_UNCOR_STATUS, reg32);
+
+	/* Enable Root Port device reporting error itself */
+	pci_read_config_word(pdev, pos+PCI_EXP_DEVCTL, &reg16);
+	reg16 = reg16 |
+		PCI_EXP_DEVCTL_CERE |
+		PCI_EXP_DEVCTL_NFERE |
+		PCI_EXP_DEVCTL_FERE |
+		PCI_EXP_DEVCTL_URRE;
+	pci_write_config_word(pdev, pos+PCI_EXP_DEVCTL,
+		reg16);
+
+	/* Enable Root Port's interrupt in response to error messages */
+	pci_write_config_dword(pdev,
+		aer_pos + PCI_ERR_ROOT_COMMAND,
+		ROOT_PORT_INTR_ON_MESG_MASK);
+}
+
+/**
+ * disable_root_aer - disable Root Port's interrupts when receiving messages
+ * @rpc: pointer to a Root Port data structure
+ *
+ * Invoked when PCIE bus unloads AER service driver.
+ **/
+static void disable_root_aer(struct aer_rpc *rpc)
+{
+	struct pci_dev *pdev = rpc->rpd->port;
+	u32 reg32;
+	int pos;
+
+	pos = pci_find_aer_capability(pdev);
+	/* Disable Root's interrupt in response to error messages */
+	pci_write_config_dword(pdev, pos + PCI_ERR_ROOT_COMMAND, 0);
+
+	/* Clear Root's error status reg */
+	pci_read_config_dword(pdev, pos + PCI_ERR_ROOT_STATUS, &reg32);
+	pci_write_config_dword(pdev, pos + PCI_ERR_ROOT_STATUS, reg32);
+}
+
+/**
+ * get_e_source - retrieve an error source
+ * @rpc: pointer to the root port which holds an error
+ *
+ * Invoked by DPC handler to consume an error.
+ **/
+static struct aer_err_source* get_e_source(struct aer_rpc *rpc)
+{
+	struct aer_err_source *e_source;
+	unsigned long flags;
+
+	/* Lock access to Root error producer/consumer index */
+	spin_lock_irqsave(&rpc->e_lock, flags);
+	if (rpc->prod_idx == rpc->cons_idx) {
+		spin_unlock_irqrestore(&rpc->e_lock, flags);
+		return NULL;
+	}
+	e_source = &rpc->e_sources[rpc->cons_idx];
+	rpc->cons_idx++;
+	if (rpc->cons_idx == AER_ERROR_SOURCES_MAX)
+		rpc->cons_idx = 0;
+	spin_unlock_irqrestore(&rpc->e_lock, flags);
+
+	return e_source;
+}
+
+static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info)
+{
+	int pos;
+
+	pos = pci_find_aer_capability(dev);
+
+	/* The device might not support AER */
+	if (!pos)
+		return AER_SUCCESS;
+
+	if (info->severity == AER_CORRECTABLE) {
+		pci_read_config_dword(dev, pos + PCI_ERR_COR_STATUS,
+			&info->status);
+		if (!(info->status & ERR_CORRECTABLE_ERROR_MASK))
+			return AER_UNSUCCESS;
+	} else if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE ||
+		info->severity == AER_NONFATAL) {
+
+		/* Link is still healthy for IO reads */
+		pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS,
+			&info->status);
+		if (!(info->status & ERR_UNCORRECTABLE_ERROR_MASK))
+			return AER_UNSUCCESS;
+
+		if (info->status & AER_LOG_TLP_MASKS) {
+			info->flags |= AER_TLP_HEADER_VALID_FLAG;
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG, &info->tlp.dw0);
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG + 4, &info->tlp.dw1);
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG + 8, &info->tlp.dw2);
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG + 12, &info->tlp.dw3);
+		}
+	}
+
+	return AER_SUCCESS;
+}
+
+/**
+ * aer_isr_one_error - consume an error detected by root port
+ * @p_device: pointer to error root port service device
+ * @e_src: pointer to an error source
+ **/
+static void aer_isr_one_error(struct pcie_device *p_device,
+		struct aer_err_source *e_src)
+{
+	struct device *s_device;
+	struct aer_err_info e_info = {0, 0, 0,};
+	int i;
+	u16 id;
+
+	/*
+	 * There is a possibility that both correctable error and
+	 * uncorrectable error being logged. Report correctable error first.
+	 */
+	for (i = 1; i & ROOT_ERR_STATUS_MASKS ; i <<= 2) {
+		if (i > 4)
+			break;
+		if (!(e_src->status & i))
+			continue;
+
+		/* Init comprehensive error information */
+		if (i & PCI_ERR_ROOT_COR_RCV) {
+			id = ERR_COR_ID(e_src->id);
+			e_info.severity = AER_CORRECTABLE;
+		} else {
+			id = ERR_UNCOR_ID(e_src->id);
+			e_info.severity = ((e_src->status >> 6) & 1);
+		}
+		if (e_src->status &
+			(PCI_ERR_ROOT_MULTI_COR_RCV |
+			 PCI_ERR_ROOT_MULTI_UNCOR_RCV))
+			e_info.flags |= AER_MULTI_ERROR_VALID_FLAG;
+		if (!(s_device = find_source_device(p_device->port, id))) {
+			printk(KERN_DEBUG "%s->can't find device of ID%04x\n",
+				__FUNCTION__, id);
+			continue;
+		}
+		if (get_device_error_info(to_pci_dev(s_device), &e_info) ==
+				AER_SUCCESS) {
+			aer_print_error(to_pci_dev(s_device), &e_info);
+			handle_error_source(p_device,
+				to_pci_dev(s_device),
+				e_info);
+		}
+	}
+}
+
+/**
+ * aer_isr - consume errors detected by root port
+ * @context: pointer to a private data of pcie device
+ *
+ * Invoked, as DPC, when root port records new detected error
+ **/
+void aer_isr(void *context)
+{
+	struct pcie_device *p_device = (struct pcie_device *) context;
+	struct aer_rpc *rpc = get_service_data(p_device);
+	struct aer_err_source *e_src;
+
+	mutex_lock(&rpc->rpc_mutex);
+	e_src = get_e_source(rpc);
+	while (e_src) {
+		aer_isr_one_error(p_device, e_src);
+		e_src = get_e_source(rpc);
+	}
+	mutex_unlock(&rpc->rpc_mutex);
+
+	wake_up(&rpc->wait_release);
+}
+
+/**
+ * aer_delete_rootport - disable root port aer and delete service data
+ * @rpc: pointer to a root port device being deleted
+ *
+ * Invoked when AER service unloaded on a specific Root Port
+ **/
+void aer_delete_rootport(struct aer_rpc *rpc)
+{
+	/* Disable root port AER itself */
+	disable_root_aer(rpc);
+
+	kfree(rpc);
+}
+
+/**
+ * aer_init - provide AER initialization
+ * @dev: pointer to AER pcie device
+ *
+ * Invoked when AER service driver is loaded.
+ **/
+int aer_init(struct pcie_device *dev)
+{
+	int status;
+
+	/* Run _OSC Method */
+	status = aer_osc_setup(dev->port);
+
+	if(status != OSC_METHOD_RUN_SUCCESS) {
+		printk(KERN_DEBUG "%s: AER service init fails - %s\n",
+		__FUNCTION__,
+		(status == OSC_METHOD_NOT_SUPPORTED) ?
+			"No ACPI _OSC support" : "Run ACPI _OSC fails");
+
+		if (!forceload)
+			return status;
+	}
+
+	return AER_SUCCESS;
+}
+
+EXPORT_SYMBOL_GPL(pci_find_aer_capability);
+EXPORT_SYMBOL_GPL(pci_enable_pcie_error_reporting);
+EXPORT_SYMBOL_GPL(pci_disable_pcie_error_reporting);
+EXPORT_SYMBOL_GPL(pci_cleanup_aer_uncorrect_error_status);
+
diff --git a/drivers/pci/pcie/aer/aerdrv_errprint.c b/drivers/pci/pcie/aer/aerdrv_errprint.c
new file mode 100644
index 000000000000..3933d4f30e8c
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv_errprint.c
@@ -0,0 +1,248 @@
+/*
+ * drivers/pci/pcie/aer/aerdrv_errprint.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Format error messages and print them to console.
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+
+#include "aerdrv.h"
+
+#define AER_AGENT_RECEIVER		0
+#define AER_AGENT_REQUESTER		1
+#define AER_AGENT_COMPLETER		2
+#define AER_AGENT_TRANSMITTER		3
+
+#define AER_AGENT_REQUESTER_MASK	(PCI_ERR_UNC_COMP_TIME|	\
+					PCI_ERR_UNC_UNSUP)
+
+#define AER_AGENT_COMPLETER_MASK	PCI_ERR_UNC_COMP_ABORT
+
+#define AER_AGENT_TRANSMITTER_MASK(t, e) (e & (PCI_ERR_COR_REP_ROLL| \
+	((t == AER_CORRECTABLE) ? PCI_ERR_COR_REP_TIMER: 0)))
+
+#define AER_GET_AGENT(t, e)						\
+	((e & AER_AGENT_COMPLETER_MASK) ? AER_AGENT_COMPLETER :		\
+	(e & AER_AGENT_REQUESTER_MASK) ? AER_AGENT_REQUESTER :		\
+	(AER_AGENT_TRANSMITTER_MASK(t, e)) ? AER_AGENT_TRANSMITTER :	\
+	AER_AGENT_RECEIVER)
+
+#define AER_PHYSICAL_LAYER_ERROR_MASK	PCI_ERR_COR_RCVR
+#define AER_DATA_LINK_LAYER_ERROR_MASK(t, e)	\
+		(PCI_ERR_UNC_DLP|		\
+		PCI_ERR_COR_BAD_TLP| 		\
+		PCI_ERR_COR_BAD_DLLP|		\
+		PCI_ERR_COR_REP_ROLL| 		\
+		((t == AER_CORRECTABLE) ?	\
+		PCI_ERR_COR_REP_TIMER: 0))
+
+#define AER_PHYSICAL_LAYER_ERROR	0
+#define AER_DATA_LINK_LAYER_ERROR	1
+#define AER_TRANSACTION_LAYER_ERROR	2
+
+#define AER_GET_LAYER_ERROR(t, e)				\
+	((e & AER_PHYSICAL_LAYER_ERROR_MASK) ?			\
+	AER_PHYSICAL_LAYER_ERROR :				\
+	(e & AER_DATA_LINK_LAYER_ERROR_MASK(t, e)) ?		\
+		AER_DATA_LINK_LAYER_ERROR : 			\
+		AER_TRANSACTION_LAYER_ERROR)
+
+/*
+ * AER error strings
+ */
+static char* aer_error_severity_string[] = {
+	"Uncorrected (Non-Fatal)",
+	"Uncorrected (Fatal)",
+	"Corrected"
+};
+
+static char* aer_error_layer[] = {
+	"Physical Layer",
+	"Data Link Layer",
+	"Transaction Layer"
+};
+static char* aer_correctable_error_string[] = {
+	"Receiver Error        ",	/* Bit Position 0 	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	"Bad TLP               ",	/* Bit Position 6 	*/
+	"Bad DLLP              ",	/* Bit Position 7 	*/
+	"RELAY_NUM Rollover    ",	/* Bit Position 8 	*/
+	NULL,
+	NULL,
+	NULL,
+	"Replay Timer Timeout  ",	/* Bit Position 12 	*/
+	"Advisory Non-Fatal    ", 	/* Bit Position 13	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+};
+
+static char* aer_uncorrectable_error_string[] = {
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	"Data Link Protocol    ",	/* Bit Position 4	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	"Poisoned TLP          ",	/* Bit Position 12 	*/
+	"Flow Control Protocol ",	/* Bit Position 13	*/
+	"Completion Timeout    ",	/* Bit Position 14 	*/
+	"Completer Abort       ",	/* Bit Position 15 	*/
+	"Unexpected Completion ",	/* Bit Position 16	*/
+	"Receiver Overflow     ",	/* Bit Position 17	*/
+	"Malformed TLP         ",	/* Bit Position 18	*/
+	"ECRC                  ",	/* Bit Position 19	*/
+	"Unsupported Request   ",	/* Bit Position 20	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+};
+
+static char* aer_agent_string[] = {
+	"Receiver ID",
+	"Requester ID",
+	"Completer ID",
+	"Transmitter ID"
+};
+
+static char * aer_get_error_source_name(int severity,
+			unsigned int status,
+			char errmsg_buff[])
+{
+	int i;
+	char * errmsg = NULL;
+
+	for (i = 0; i < 32; i++) {
+		if (!(status & (1 << i)))
+			continue;
+
+		if (severity == AER_CORRECTABLE)
+			errmsg = aer_correctable_error_string[i];
+		else
+			errmsg = aer_uncorrectable_error_string[i];
+
+		if (!errmsg) {
+			sprintf(errmsg_buff, "Unknown Error Bit %2d  ", i);
+			errmsg = errmsg_buff;
+		}
+
+		break;
+	}
+
+	return errmsg;
+}
+
+static DEFINE_SPINLOCK(logbuf_lock);
+static char errmsg_buff[100];
+void aer_print_error(struct pci_dev *dev, struct aer_err_info *info)
+{
+	char * errmsg;
+	int err_layer, agent;
+	char * loglevel;
+
+	if (info->severity == AER_CORRECTABLE)
+		loglevel = KERN_WARNING;
+	else
+		loglevel = KERN_ERR;
+
+	printk("%s+------ PCI-Express Device Error ------+\n", loglevel);
+	printk("%sError Severity\t\t: %s\n", loglevel,
+		aer_error_severity_string[info->severity]);
+
+	if ( info->status == 0) {
+		printk("%sPCIE Bus Error type\t: (Unaccessible)\n", loglevel);
+		printk("%sUnaccessible Received\t: %s\n", loglevel,
+			info->flags & AER_MULTI_ERROR_VALID_FLAG ?
+				"Multiple" : "First");
+		printk("%sUnregistered Agent ID\t: %04x\n", loglevel,
+			(dev->bus->number << 8) | dev->devfn);
+	} else {
+		err_layer = AER_GET_LAYER_ERROR(info->severity, info->status);
+		printk("%sPCIE Bus Error type\t: %s\n", loglevel,
+			aer_error_layer[err_layer]);
+
+		spin_lock(&logbuf_lock);
+		errmsg = aer_get_error_source_name(info->severity,
+				info->status,
+				errmsg_buff);
+		printk("%s%s\t: %s\n", loglevel, errmsg,
+			info->flags & AER_MULTI_ERROR_VALID_FLAG ?
+				"Multiple" : "First");
+		spin_unlock(&logbuf_lock);
+
+		agent = AER_GET_AGENT(info->severity, info->status);
+		printk("%s%s\t\t: %04x\n", loglevel,
+			aer_agent_string[agent],
+			(dev->bus->number << 8) | dev->devfn);
+
+		printk("%sVendorID=%04xh, DeviceID=%04xh,"
+			" Bus=%02xh, Device=%02xh, Function=%02xh\n",
+			loglevel,
+			dev->vendor,
+			dev->device,
+			dev->bus->number,
+			PCI_SLOT(dev->devfn),
+			PCI_FUNC(dev->devfn));
+
+		if (info->flags & AER_TLP_HEADER_VALID_FLAG) {
+			unsigned char *tlp = (unsigned char *) &info->tlp;
+			printk("%sTLB Header:\n", loglevel);
+			printk("%s%02x%02x%02x%02x %02x%02x%02x%02x"
+				" %02x%02x%02x%02x %02x%02x%02x%02x\n",
+				loglevel,
+				*(tlp + 3), *(tlp + 2), *(tlp + 1), *tlp,
+				*(tlp + 7), *(tlp + 6), *(tlp + 5), *(tlp + 4),
+				*(tlp + 11), *(tlp + 10), *(tlp + 9),
+				*(tlp + 8), *(tlp + 15), *(tlp + 14),
+				*(tlp + 13), *(tlp + 12));
+		}
+	}
+}
+
diff --git a/include/linux/aer.h b/include/linux/aer.h
new file mode 100644
index 000000000000..402e178b38eb
--- /dev/null
+++ b/include/linux/aer.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2006 Intel Corp.
+ *     Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *     Zhang Yanmin (yanmin.zhang@intel.com)
+ */
+
+#ifndef _AER_H_
+#define _AER_H_
+
+#if defined(CONFIG_PCIEAER)
+/* pci-e port driver needs this function to enable aer */
+extern int pci_enable_pcie_error_reporting(struct pci_dev *dev);
+extern int pci_find_aer_capability(struct pci_dev *dev);
+extern int pci_disable_pcie_error_reporting(struct pci_dev *dev);
+extern int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev);
+#else
+#define pci_enable_pcie_error_reporting(dev)		do { } while (0)
+#define pci_find_aer_capability(dev)			do { } while (0)
+#define pci_disable_pcie_error_reporting(dev)		do { } while (0)
+#define pci_cleanup_aer_uncorrect_error_status(dev)	do { } while (0)
+#endif
+
+#endif //_AER_H_
+
diff --git a/include/linux/pcieport_if.h b/include/linux/pcieport_if.h
index b44e01a70914..6cd91e3f9820 100644
--- a/include/linux/pcieport_if.h
+++ b/include/linux/pcieport_if.h
@@ -62,6 +62,12 @@ struct pcie_port_service_driver {
 	int (*suspend) (struct pcie_device *dev, pm_message_t state);
 	int (*resume) (struct pcie_device *dev);
 
+	/* Service Error Recovery Handler */
+	struct pci_error_handlers *err_handler;
+
+	/* Link Reset Capability - AER service driver specific */
+	pci_ers_result_t (*reset_link) (struct pci_dev *dev);
+
 	const struct pcie_port_service_id *id_table;
 	struct device_driver driver;
 };
-- 
cgit v1.2.3


From b19441af185559118e8247382ea4f2f76ebffc6d Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: PCI: fix __must_check warnings

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/bus.c                  |  22 ++++++--
 drivers/pci/hotplug/fakephp.c      |  18 ++++--
 drivers/pci/pci-driver.c           |   5 +-
 drivers/pci/pci-sysfs.c            | 112 ++++++++++++++++++++++++-------------
 drivers/pci/pcie/aer/aerdrv_core.c |   3 +-
 drivers/pci/pcie/portdrv_core.c    |   6 +-
 drivers/pci/pcie/portdrv_pci.c     |  16 ++++--
 drivers/pci/probe.c                |  16 +++++-
 include/linux/pci.h                |   2 +-
 9 files changed, 138 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 5f7db9d2436e..aadaa3c8096b 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -77,9 +77,12 @@ pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
  * This adds a single pci device to the global
  * device list and adds sysfs and procfs entries
  */
-void __devinit pci_bus_add_device(struct pci_dev *dev)
+int __devinit pci_bus_add_device(struct pci_dev *dev)
 {
-	device_add(&dev->dev);
+	int retval;
+	retval = device_add(&dev->dev);
+	if (retval)
+		return retval;
 
 	down_write(&pci_bus_sem);
 	list_add_tail(&dev->global_list, &pci_devices);
@@ -87,6 +90,7 @@ void __devinit pci_bus_add_device(struct pci_dev *dev)
 
 	pci_proc_attach_device(dev);
 	pci_create_sysfs_dev_files(dev);
+	return 0;
 }
 
 /**
@@ -104,6 +108,7 @@ void __devinit pci_bus_add_device(struct pci_dev *dev)
 void __devinit pci_bus_add_devices(struct pci_bus *bus)
 {
 	struct pci_dev *dev;
+	int retval;
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
 		/*
@@ -112,7 +117,9 @@ void __devinit pci_bus_add_devices(struct pci_bus *bus)
 		 */
 		if (!list_empty(&dev->global_list))
 			continue;
-		pci_bus_add_device(dev);
+		retval = pci_bus_add_device(dev);
+		if (retval)
+			dev_err(&dev->dev, "Error adding device, continuing\n");
 	}
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
@@ -129,10 +136,13 @@ void __devinit pci_bus_add_devices(struct pci_bus *bus)
 			       list_add_tail(&dev->subordinate->node,
 					       &dev->bus->children);
 			       up_write(&pci_bus_sem);
-		       }
+			}
 			pci_bus_add_devices(dev->subordinate);
-
-			sysfs_create_link(&dev->subordinate->class_dev.kobj, &dev->dev.kobj, "bridge");
+			retval = sysfs_create_link(&dev->subordinate->class_dev.kobj,
+						   &dev->dev.kobj, "bridge");
+			if (retval)
+				dev_err(&dev->dev, "Error creating sysfs "
+					"bridge symlink, continuing...\n");
 		}
 	}
 }
diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c
index dd2b762777c4..05a4f0f90186 100644
--- a/drivers/pci/hotplug/fakephp.c
+++ b/drivers/pci/hotplug/fakephp.c
@@ -176,7 +176,9 @@ static void pci_rescan_slot(struct pci_dev *temp)
 	struct pci_bus *bus = temp->bus;
 	struct pci_dev *dev;
 	int func;
+	int retval;
 	u8 hdr_type;
+
 	if (!pci_read_config_byte(temp, PCI_HEADER_TYPE, &hdr_type)) {
 		temp->hdr_type = hdr_type & 0x7f;
 		if (!pci_find_slot(bus->number, temp->devfn)) {
@@ -185,8 +187,12 @@ static void pci_rescan_slot(struct pci_dev *temp)
 				dbg("New device on %s function %x:%x\n",
 					bus->name, temp->devfn >> 3,
 					temp->devfn & 7);
-				pci_bus_add_device(dev);
-				add_slot(dev);
+				retval = pci_bus_add_device(dev);
+				if (retval)
+					dev_err(&dev->dev, "error adding "
+						"device, continuing.\n");
+				else
+					add_slot(dev);
 			}
 		}
 		/* multifunction device? */
@@ -205,8 +211,12 @@ static void pci_rescan_slot(struct pci_dev *temp)
 					dbg("New device on %s function %x:%x\n",
 						bus->name, temp->devfn >> 3,
 						temp->devfn & 7);
-					pci_bus_add_device(dev);
-					add_slot(dev);
+					retval = pci_bus_add_device(dev);
+					if (retval)
+						dev_err(&dev->dev, "error adding "
+							"device, continuing.\n");
+					else
+						add_slot(dev);
 				}
 			}
 		}
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index d8ace1f90dd2..309629e03bae 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -56,6 +56,7 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count)
 		subdevice=PCI_ANY_ID, class=0, class_mask=0;
 	unsigned long driver_data=0;
 	int fields=0;
+	int retval = 0;
 
 	fields = sscanf(buf, "%x %x %x %x %x %x %lux",
 			&vendor, &device, &subvendor, &subdevice,
@@ -82,10 +83,12 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count)
 	spin_unlock(&pdrv->dynids.lock);
 
 	if (get_driver(&pdrv->driver)) {
-		driver_attach(&pdrv->driver);
+		retval = driver_attach(&pdrv->driver);
 		put_driver(&pdrv->driver);
 	}
 
+	if (retval)
+		return retval;
 	return count;
 }
 static DRIVER_ATTR(new_id, S_IWUSR, NULL, store_new_id);
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 010e01c4bd43..a1d2e979b17f 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -117,6 +117,7 @@ is_enabled_store(struct device *dev, struct device_attribute *attr,
 		const char *buf, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
+	int retval = 0;
 
 	/* this can crash the machine when done on the "wrong" device */
 	if (!capable(CAP_SYS_ADMIN))
@@ -126,8 +127,10 @@ is_enabled_store(struct device *dev, struct device_attribute *attr,
 		pci_disable_device(pdev);
 
 	if (*buf == '1')
-		pci_enable_device(pdev);
+		retval = pci_enable_device(pdev);
 
+	if (retval)
+		return retval;
 	return count;
 }
 
@@ -425,16 +428,39 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 	return pci_mmap_page_range(pdev, vma, mmap_type, 0);
 }
 
+/**
+ * pci_remove_resource_files - cleanup resource files
+ * @dev: dev to cleanup
+ *
+ * If we created resource files for @dev, remove them from sysfs and
+ * free their resources.
+ */
+static void
+pci_remove_resource_files(struct pci_dev *pdev)
+{
+	int i;
+
+	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
+		struct bin_attribute *res_attr;
+
+		res_attr = pdev->res_attr[i];
+		if (res_attr) {
+			sysfs_remove_bin_file(&pdev->dev.kobj, res_attr);
+			kfree(res_attr);
+		}
+	}
+}
+
 /**
  * pci_create_resource_files - create resource files in sysfs for @dev
  * @dev: dev in question
  *
  * Walk the resources in @dev creating files for each resource available.
  */
-static void
-pci_create_resource_files(struct pci_dev *pdev)
+static int pci_create_resource_files(struct pci_dev *pdev)
 {
 	int i;
+	int retval;
 
 	/* Expose the PCI resources from this device as files */
 	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
@@ -457,35 +483,19 @@ pci_create_resource_files(struct pci_dev *pdev)
 			res_attr->size = pci_resource_len(pdev, i);
 			res_attr->mmap = pci_mmap_resource;
 			res_attr->private = &pdev->resource[i];
-			sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
-		}
-	}
-}
-
-/**
- * pci_remove_resource_files - cleanup resource files
- * @dev: dev to cleanup
- *
- * If we created resource files for @dev, remove them from sysfs and
- * free their resources.
- */
-static void
-pci_remove_resource_files(struct pci_dev *pdev)
-{
-	int i;
-
-	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
-		struct bin_attribute *res_attr;
-
-		res_attr = pdev->res_attr[i];
-		if (res_attr) {
-			sysfs_remove_bin_file(&pdev->dev.kobj, res_attr);
-			kfree(res_attr);
+			retval = sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
+			if (retval) {
+				pci_remove_resource_files(pdev);
+				return retval;
+			}
+		} else {
+			return -ENOMEM;
 		}
 	}
+	return 0;
 }
 #else /* !HAVE_PCI_MMAP */
-static inline void pci_create_resource_files(struct pci_dev *dev) { return; }
+static inline int pci_create_resource_files(struct pci_dev *dev) { return 0; }
 static inline void pci_remove_resource_files(struct pci_dev *dev) { return; }
 #endif /* HAVE_PCI_MMAP */
 
@@ -570,22 +580,27 @@ static struct bin_attribute pcie_config_attr = {
 	.write = pci_write_config,
 };
 
-int pci_create_sysfs_dev_files (struct pci_dev *pdev)
+int __must_check pci_create_sysfs_dev_files (struct pci_dev *pdev)
 {
+	struct bin_attribute *rom_attr = NULL;
+	int retval;
+
 	if (!sysfs_initialized)
 		return -EACCES;
 
 	if (pdev->cfg_size < 4096)
-		sysfs_create_bin_file(&pdev->dev.kobj, &pci_config_attr);
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pci_config_attr);
 	else
-		sysfs_create_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+	if (retval)
+		goto err;
 
-	pci_create_resource_files(pdev);
+	retval = pci_create_resource_files(pdev);
+	if (retval)
+		goto err_bin_file;
 
 	/* If the device has a ROM, try to expose it in sysfs. */
 	if (pci_resource_len(pdev, PCI_ROM_RESOURCE)) {
-		struct bin_attribute *rom_attr;
-		
 		rom_attr = kzalloc(sizeof(*rom_attr), GFP_ATOMIC);
 		if (rom_attr) {
 			pdev->rom_attr = rom_attr;
@@ -595,13 +610,28 @@ int pci_create_sysfs_dev_files (struct pci_dev *pdev)
 			rom_attr->attr.owner = THIS_MODULE;
 			rom_attr->read = pci_read_rom;
 			rom_attr->write = pci_write_rom;
-			sysfs_create_bin_file(&pdev->dev.kobj, rom_attr);
+			retval = sysfs_create_bin_file(&pdev->dev.kobj, rom_attr);
+			if (retval)
+				goto err_rom;
+		} else {
+			retval = -ENOMEM;
+			goto err_bin_file;
 		}
 	}
 	/* add platform-specific attributes */
 	pcibios_add_platform_entries(pdev);
-	
+
 	return 0;
+
+err_rom:
+	kfree(rom_attr);
+err_bin_file:
+	if (pdev->cfg_size < 4096)
+		sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr);
+	else
+		sysfs_remove_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+err:
+	return retval;
 }
 
 /**
@@ -630,10 +660,14 @@ void pci_remove_sysfs_dev_files(struct pci_dev *pdev)
 static int __init pci_sysfs_init(void)
 {
 	struct pci_dev *pdev = NULL;
-	
+	int retval;
+
 	sysfs_initialized = 1;
-	for_each_pci_dev(pdev)
-		pci_create_sysfs_dev_files(pdev);
+	for_each_pci_dev(pdev) {
+		retval = pci_create_sysfs_dev_files(pdev);
+		if (retval)
+			return retval;
+	}
 
 	return 0;
 }
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index 5591043acea7..1c7e660d6535 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -357,7 +357,8 @@ static int find_aer_service_iter(struct device *device, void *data)
 static void find_aer_service(struct pci_dev *dev,
 		struct find_aer_service_data *data)
 {
-	device_for_each_child(&dev->dev, data, find_aer_service_iter);
+	int retval;
+	retval = device_for_each_child(&dev->dev, data, find_aer_service_iter);
 }
 
 static pci_ers_result_t reset_link(struct pcie_device *aerdev,
diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c
index cf9e810b4bf8..bd6615b4d40e 100644
--- a/drivers/pci/pcie/portdrv_core.c
+++ b/drivers/pci/pcie/portdrv_core.c
@@ -340,8 +340,7 @@ static int suspend_iter(struct device *dev, void *data)
 
 int pcie_port_device_suspend(struct pci_dev *dev, pm_message_t state)
 {
-	device_for_each_child(&dev->dev, &state, suspend_iter);
-	return 0;
+	return device_for_each_child(&dev->dev, &state, suspend_iter);
 }
 
 static int resume_iter(struct device *dev, void *data)
@@ -359,8 +358,7 @@ static int resume_iter(struct device *dev, void *data)
 
 int pcie_port_device_resume(struct pci_dev *dev)
 {
-	device_for_each_child(&dev->dev, NULL, resume_iter);
-	return 0;
+	return device_for_each_child(&dev->dev, NULL, resume_iter);
 }
 #endif
 
diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c
index e4a2429986f0..037690e08f5f 100644
--- a/drivers/pci/pcie/portdrv_pci.c
+++ b/drivers/pci/pcie/portdrv_pci.c
@@ -147,8 +147,10 @@ static pci_ers_result_t pcie_portdrv_error_detected(struct pci_dev *dev,
 {
 	struct aer_broadcast_data result_data =
 			{error, PCI_ERS_RESULT_CAN_RECOVER};
+	int retval;
 
-	device_for_each_child(&dev->dev, &result_data, error_detected_iter);
+	/* can not fail */
+	retval = device_for_each_child(&dev->dev, &result_data, error_detected_iter);
 
 	return result_data.result;
 }
@@ -181,8 +183,10 @@ static int mmio_enabled_iter(struct device *device, void *data)
 static pci_ers_result_t pcie_portdrv_mmio_enabled(struct pci_dev *dev)
 {
 	pci_ers_result_t status = PCI_ERS_RESULT_RECOVERED;
+	int retval;
 
-	device_for_each_child(&dev->dev, &status, mmio_enabled_iter);
+	/* get true return value from &status */
+	retval = device_for_each_child(&dev->dev, &status, mmio_enabled_iter);
 	return status;
 }
 
@@ -214,6 +218,7 @@ static int slot_reset_iter(struct device *device, void *data)
 static pci_ers_result_t pcie_portdrv_slot_reset(struct pci_dev *dev)
 {
 	pci_ers_result_t status;
+	int retval;
 
 	/* If fatal, restore cfg space for possible link reset at upstream */
 	if (dev->error_state == pci_channel_io_frozen) {
@@ -221,7 +226,8 @@ static pci_ers_result_t pcie_portdrv_slot_reset(struct pci_dev *dev)
 		pci_enable_pcie_error_reporting(dev);
 	}
 
-	device_for_each_child(&dev->dev, &status, slot_reset_iter);
+	/* get true return value from &status */
+	retval = device_for_each_child(&dev->dev, &status, slot_reset_iter);
 
 	return status;
 }
@@ -248,7 +254,9 @@ static int resume_iter(struct device *device, void *data)
 
 static void pcie_portdrv_err_resume(struct pci_dev *dev)
 {
-	device_for_each_child(&dev->dev, NULL, resume_iter);
+	int retval;
+	/* nothing to do with error value, if it ever happens */
+	retval = device_for_each_child(&dev->dev, NULL, resume_iter);
 }
 
 /*
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index c5a58d1c6c1c..a3b0a5eb5054 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -339,6 +339,7 @@ pci_alloc_child_bus(struct pci_bus *parent, struct pci_dev *bridge, int busnr)
 {
 	struct pci_bus *child;
 	int i;
+	int retval;
 
 	/*
 	 * Allocate a new bus, and inherit stuff from the parent..
@@ -356,8 +357,13 @@ pci_alloc_child_bus(struct pci_bus *parent, struct pci_dev *bridge, int busnr)
 
 	child->class_dev.class = &pcibus_class;
 	sprintf(child->class_dev.class_id, "%04x:%02x", pci_domain_nr(child), busnr);
-	class_device_register(&child->class_dev);
-	class_device_create_file(&child->class_dev, &class_device_attr_cpuaffinity);
+	retval = class_device_register(&child->class_dev);
+	if (retval)
+		goto error_register;
+	retval = class_device_create_file(&child->class_dev,
+					  &class_device_attr_cpuaffinity);
+	if (retval)
+		goto error_file_create;
 
 	/*
 	 * Set up the primary, secondary and subordinate
@@ -375,6 +381,12 @@ pci_alloc_child_bus(struct pci_bus *parent, struct pci_dev *bridge, int busnr)
 	bridge->subordinate = child;
 
 	return child;
+
+error_file_create:
+	class_device_unregister(&child->class_dev);
+error_register:
+	kfree(child);
+	return NULL;
 }
 
 struct pci_bus * __devinit pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev, int busnr)
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 3ec72551ac31..c9bb7bee52c7 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -431,7 +431,7 @@ int pci_scan_slot(struct pci_bus *bus, int devfn);
 struct pci_dev * pci_scan_single_device(struct pci_bus *bus, int devfn);
 void pci_device_add(struct pci_dev *dev, struct pci_bus *bus);
 unsigned int pci_scan_child_bus(struct pci_bus *bus);
-void pci_bus_add_device(struct pci_dev *dev);
+int __must_check pci_bus_add_device(struct pci_dev *dev);
 void pci_read_bridge_bases(struct pci_bus *child);
 struct resource *pci_find_parent_resource(const struct pci_dev *dev, struct resource *res);
 int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge);
-- 
cgit v1.2.3


From 50b0075520a0acba9cabab5203bbce918b966d9a Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Wed, 16 Aug 2006 17:42:18 +0100
Subject: PCI: Multiprobe sanitizer

There are numerous drivers that can use multithreaded probing but having
some kind of global flag as the way to control this makes migration to
threaded probing hard and since it enables it everywhere and is almost
as likely to cause serious pain as holding a clog dance in a minefield.

If we have a pci_driver multithread_probe flag to inherit you can turn
it on for one driver at a time.

From playing so far however I think we need a different model at the
device layer which serializes until the called probe function says "ok
you can start another one now". That would need some kind of flag and
semaphore plus a helper function.

Anyway in the absence of that this is a starting point to usefully play
with this stuff

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci-driver.c | 6 +++++-
 include/linux/pci.h      | 2 ++
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 309629e03bae..b1c0c707d96c 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -421,7 +421,11 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner)
 	drv->driver.bus = &pci_bus_type;
 	drv->driver.owner = owner;
 	drv->driver.kobj.ktype = &pci_driver_kobj_type;
-	drv->driver.multithread_probe = pci_multithread_probe;
+
+	if (pci_multithread_probe)
+		drv->driver.multithread_probe = pci_multithread_probe;
+	else
+		drv->driver.multithread_probe = drv->multithread_probe;
 
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index c9bb7bee52c7..549d8410974b 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -356,6 +356,8 @@ struct pci_driver {
 	struct pci_error_handlers *err_handler;
 	struct device_driver	driver;
 	struct pci_dynids dynids;
+
+	int multithread_probe;
 };
 
 #define	to_pci_driver(drv) container_of(drv,struct pci_driver, driver)
-- 
cgit v1.2.3


From 24f8aa9b464b73e0553f092b747770940ee0ea54 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:16:36 -0700
Subject: PCI: add pci_stop_bus_device

This patch adds pci_stop_bus_device() which stops a PCI device (detach
the driver, remove from the global list and so on) and any children.
This is needed for ACPI based PCI-to-PCI bridge hot-remove, and it will
be also needed for ACPI based PCI root bridge hot-remove.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/remove.c | 37 ++++++++++++++++++++++++++++++++++++-
 include/linux/pci.h  |  1 +
 2 files changed, 37 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
index 99ffbd478b29..430281b2e921 100644
--- a/drivers/pci/remove.c
+++ b/drivers/pci/remove.c
@@ -16,8 +16,11 @@ static void pci_free_resources(struct pci_dev *dev)
 	}
 }
 
-static void pci_destroy_dev(struct pci_dev *dev)
+static void pci_stop_dev(struct pci_dev *dev)
 {
+	if (!dev->global_list.next)
+		return;
+
 	if (!list_empty(&dev->global_list)) {
 		pci_proc_detach_device(dev);
 		pci_remove_sysfs_dev_files(dev);
@@ -27,6 +30,11 @@ static void pci_destroy_dev(struct pci_dev *dev)
 		dev->global_list.next = dev->global_list.prev = NULL;
 		up_write(&pci_bus_sem);
 	}
+}
+
+static void pci_destroy_dev(struct pci_dev *dev)
+{
+	pci_stop_dev(dev);
 
 	/* Remove the device from the device lists, and prevent any further
 	 * list accesses from this device */
@@ -119,5 +127,32 @@ void pci_remove_behind_bridge(struct pci_dev *dev)
 	}
 }
 
+static void pci_stop_bus_devices(struct pci_bus *bus)
+{
+	struct list_head *l, *n;
+
+	list_for_each_safe(l, n, &bus->devices) {
+		struct pci_dev *dev = pci_dev_b(l);
+		pci_stop_bus_device(dev);
+	}
+}
+
+/**
+ * pci_stop_bus_device - stop a PCI device and any children
+ * @dev: the device to stop
+ *
+ * Stop a PCI device (detach the driver, remove from the global list
+ * and so on). This also stop any subordinate buses and children in a
+ * depth-first manner.
+ */
+void pci_stop_bus_device(struct pci_dev *dev)
+{
+	if (dev->subordinate)
+		pci_stop_bus_devices(dev->subordinate);
+
+	pci_stop_dev(dev);
+}
+
 EXPORT_SYMBOL(pci_remove_bus_device);
 EXPORT_SYMBOL(pci_remove_behind_bridge);
+EXPORT_SYMBOL_GPL(pci_stop_bus_device);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 549d8410974b..5c3a4176eb64 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -441,6 +441,7 @@ extern struct pci_dev *pci_dev_get(struct pci_dev *dev);
 extern void pci_dev_put(struct pci_dev *dev);
 extern void pci_remove_bus(struct pci_bus *b);
 extern void pci_remove_bus_device(struct pci_dev *dev);
+extern void pci_stop_bus_device(struct pci_dev *dev);
 void pci_setup_cardbus(struct pci_bus *bus);
 
 /* Generic PCI functions exported to card drivers */
-- 
cgit v1.2.3


From 0c91c1a7012911e4b0180a1c1ae258e2b706f987 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:16:20 +0900
Subject: sh: Move smc37c93x.h for SystemH board use.

SystemH needs this header as well, not just 770x SE.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/se/770x/setup.c |   2 +-
 include/asm-sh/se/smc37c93x.h  | 190 -----------------------------------------
 include/asm-sh/smc37c93x.h     | 190 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 191 insertions(+), 191 deletions(-)
 delete mode 100644 include/asm-sh/se/smc37c93x.h
 create mode 100644 include/asm-sh/smc37c93x.h

(limited to 'include')

diff --git a/arch/sh/boards/se/770x/setup.c b/arch/sh/boards/se/770x/setup.c
index 7d1a071727cc..b6730ea57502 100644
--- a/arch/sh/boards/se/770x/setup.c
+++ b/arch/sh/boards/se/770x/setup.c
@@ -15,7 +15,7 @@
 #include <linux/ide.h>
 #include <asm/io.h>
 #include <asm/se/se.h>
-#include <asm/se/smc37c93x.h>
+#include <asm/smc37c93x.h>
 
 /*
  * Configure the Super I/O chip
diff --git a/include/asm-sh/se/smc37c93x.h b/include/asm-sh/se/smc37c93x.h
deleted file mode 100644
index 585da2a8fc45..000000000000
--- a/include/asm-sh/se/smc37c93x.h
+++ /dev/null
@@ -1,190 +0,0 @@
-#ifndef __ASM_SH_SMC37C93X_H
-#define __ASM_SH_SMC37C93X_H
-
-/*
- * linux/include/asm-sh/smc37c93x.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * SMSC 37C93x Super IO Chip support
- */
-
-/* Default base I/O address */
-#define FDC_PRIMARY_BASE	0x3f0
-#define IDE1_PRIMARY_BASE	0x1f0
-#define IDE1_SECONDARY_BASE	0x170
-#define PARPORT_PRIMARY_BASE	0x378
-#define COM1_PRIMARY_BASE	0x2f8
-#define COM2_PRIMARY_BASE	0x3f8
-#define RTC_PRIMARY_BASE	0x070
-#define KBC_PRIMARY_BASE	0x060
-#define AUXIO_PRIMARY_BASE	0x000	/* XXX */
-
-/* Logical device number */
-#define LDN_FDC			0
-#define LDN_IDE1		1
-#define LDN_IDE2		2
-#define LDN_PARPORT		3
-#define LDN_COM1		4
-#define LDN_COM2		5
-#define LDN_RTC			6
-#define LDN_KBC			7
-#define LDN_AUXIO		8
-
-/* Configuration port and key */
-#define CONFIG_PORT		0x3f0
-#define INDEX_PORT		CONFIG_PORT
-#define DATA_PORT		0x3f1
-#define CONFIG_ENTER		0x55
-#define CONFIG_EXIT		0xaa
-
-/* Configuration index */
-#define CURRENT_LDN_INDEX	0x07
-#define POWER_CONTROL_INDEX	0x22
-#define ACTIVATE_INDEX		0x30
-#define IO_BASE_HI_INDEX	0x60
-#define IO_BASE_LO_INDEX	0x61
-#define IRQ_SELECT_INDEX	0x70
-#define DMA_SELECT_INDEX	0x74
-
-#define GPIO46_INDEX		0xc6
-#define GPIO47_INDEX		0xc7
-
-/* UART stuff. Only for debugging.  */
-/* UART Register */
-
-#define UART_RBR	0x0	/* Receiver Buffer Register (Read Only) */
-#define UART_THR	0x0	/* Transmitter Holding Register (Write Only) */
-#define UART_IER	0x2	/* Interrupt Enable Register */
-#define UART_IIR	0x4	/* Interrupt Ident Register (Read Only) */
-#define UART_FCR	0x4	/* FIFO Control Register (Write Only) */
-#define UART_LCR	0x6	/* Line Control Register */
-#define UART_MCR	0x8	/* MODEM Control Register */
-#define UART_LSR	0xa	/* Line Status Register */
-#define UART_MSR	0xc	/* MODEM Status Register */
-#define UART_SCR	0xe	/* Scratch Register */
-#define UART_DLL	0x0	/* Divisor Latch (LS) */
-#define UART_DLM	0x2	/* Divisor Latch (MS) */
-
-#ifndef __ASSEMBLY__
-typedef struct uart_reg {
-	volatile __u16 rbr;
-	volatile __u16 ier;
-	volatile __u16 iir;
-	volatile __u16 lcr;
-	volatile __u16 mcr;
-	volatile __u16 lsr;
-	volatile __u16 msr;
-	volatile __u16 scr;
-} uart_reg;
-#endif /* ! __ASSEMBLY__ */
-
-/* Alias for Write Only Register */
-
-#define thr	rbr
-#define tcr	iir
-
-/* Alias for Divisor Latch Register */
-
-#define dll	rbr
-#define dlm	ier
-#define fcr	iir
-
-/* Interrupt Enable Register */
-
-#define IER_ERDAI	0x0100	/* Enable Received Data Available Interrupt */
-#define IER_ETHREI	0x0200	/* Enable Transmitter Holding Register Empty Interrupt */
-#define IER_ELSI	0x0400	/* Enable Receiver Line Status Interrupt */
-#define IER_EMSI	0x0800	/* Enable MODEM Status Interrupt */
-
-/* Interrupt Ident Register */
-
-#define IIR_IP		0x0100	/* "0" if Interrupt Pending */
-#define IIR_IIB0	0x0200	/* Interrupt ID Bit 0 */
-#define IIR_IIB1	0x0400	/* Interrupt ID Bit 1 */
-#define IIR_IIB2	0x0800	/* Interrupt ID Bit 2 */
-#define IIR_FIFO	0xc000	/* FIFOs enabled */
-
-/* FIFO Control Register */
-
-#define FCR_FEN		0x0100	/* FIFO enable */
-#define FCR_RFRES	0x0200	/* Receiver FIFO reset */
-#define FCR_TFRES	0x0400	/* Transmitter FIFO reset */
-#define FCR_DMA		0x0800	/* DMA mode select */
-#define FCR_RTL		0x4000	/* Receiver triger (LSB) */
-#define FCR_RTM		0x8000	/* Receiver triger (MSB) */
-
-/* Line Control Register */
-
-#define LCR_WLS0	0x0100	/* Word Length Select Bit 0 */
-#define LCR_WLS1	0x0200	/* Word Length Select Bit 1 */
-#define LCR_STB		0x0400	/* Number of Stop Bits */
-#define LCR_PEN		0x0800	/* Parity Enable */
-#define LCR_EPS		0x1000	/* Even Parity Select */
-#define LCR_SP		0x2000	/* Stick Parity */
-#define LCR_SB		0x4000	/* Set Break */
-#define LCR_DLAB	0x8000	/* Divisor Latch Access Bit */
-
-/* MODEM Control Register */
-
-#define MCR_DTR		0x0100	/* Data Terminal Ready */
-#define MCR_RTS		0x0200	/* Request to Send */
-#define MCR_OUT1	0x0400	/* Out 1 */
-#define MCR_IRQEN	0x0800	/* IRQ Enable */
-#define MCR_LOOP	0x1000	/* Loop */
-
-/* Line Status Register */
-
-#define LSR_DR		0x0100	/* Data Ready */
-#define LSR_OE		0x0200	/* Overrun Error */
-#define LSR_PE		0x0400	/* Parity Error */
-#define LSR_FE		0x0800	/* Framing Error */
-#define LSR_BI		0x1000	/* Break Interrupt */
-#define LSR_THRE	0x2000	/* Transmitter Holding Register Empty */
-#define LSR_TEMT	0x4000	/* Transmitter Empty */
-#define LSR_FIFOE	0x8000	/* Receiver FIFO error */
-
-/* MODEM Status Register */
-
-#define MSR_DCTS	0x0100	/* Delta Clear to Send */
-#define MSR_DDSR	0x0200	/* Delta Data Set Ready */
-#define MSR_TERI	0x0400	/* Trailing Edge Ring Indicator */
-#define MSR_DDCD	0x0800	/* Delta Data Carrier Detect */
-#define MSR_CTS		0x1000	/* Clear to Send */
-#define MSR_DSR		0x2000	/* Data Set Ready */
-#define MSR_RI		0x4000	/* Ring Indicator */
-#define MSR_DCD		0x8000	/* Data Carrier Detect */
-
-/* Baud Rate Divisor */
-
-#define UART_CLK	(1843200)	/* 1.8432 MHz */
-#define UART_BAUD(x)	(UART_CLK / (16 * (x)))
-
-/* RTC register definition */
-#define RTC_SECONDS             0
-#define RTC_SECONDS_ALARM       1
-#define RTC_MINUTES             2
-#define RTC_MINUTES_ALARM       3
-#define RTC_HOURS               4
-#define RTC_HOURS_ALARM         5
-#define RTC_DAY_OF_WEEK         6
-#define RTC_DAY_OF_MONTH        7
-#define RTC_MONTH               8
-#define RTC_YEAR                9
-#define RTC_FREQ_SELECT		10
-# define RTC_UIP 0x80
-# define RTC_DIV_CTL 0x70
-/* This RTC can work under 32.768KHz clock only.  */
-# define RTC_OSC_ENABLE 0x20
-# define RTC_OSC_DISABLE 0x00
-#define RTC_CONTROL     	11
-# define RTC_SET 0x80
-# define RTC_PIE 0x40
-# define RTC_AIE 0x20
-# define RTC_UIE 0x10
-# define RTC_SQWE 0x08
-# define RTC_DM_BINARY 0x04
-# define RTC_24H 0x02
-# define RTC_DST_EN 0x01
-
-#endif  /* __ASM_SH_SMC37C93X_H */
diff --git a/include/asm-sh/smc37c93x.h b/include/asm-sh/smc37c93x.h
new file mode 100644
index 000000000000..585da2a8fc45
--- /dev/null
+++ b/include/asm-sh/smc37c93x.h
@@ -0,0 +1,190 @@
+#ifndef __ASM_SH_SMC37C93X_H
+#define __ASM_SH_SMC37C93X_H
+
+/*
+ * linux/include/asm-sh/smc37c93x.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * SMSC 37C93x Super IO Chip support
+ */
+
+/* Default base I/O address */
+#define FDC_PRIMARY_BASE	0x3f0
+#define IDE1_PRIMARY_BASE	0x1f0
+#define IDE1_SECONDARY_BASE	0x170
+#define PARPORT_PRIMARY_BASE	0x378
+#define COM1_PRIMARY_BASE	0x2f8
+#define COM2_PRIMARY_BASE	0x3f8
+#define RTC_PRIMARY_BASE	0x070
+#define KBC_PRIMARY_BASE	0x060
+#define AUXIO_PRIMARY_BASE	0x000	/* XXX */
+
+/* Logical device number */
+#define LDN_FDC			0
+#define LDN_IDE1		1
+#define LDN_IDE2		2
+#define LDN_PARPORT		3
+#define LDN_COM1		4
+#define LDN_COM2		5
+#define LDN_RTC			6
+#define LDN_KBC			7
+#define LDN_AUXIO		8
+
+/* Configuration port and key */
+#define CONFIG_PORT		0x3f0
+#define INDEX_PORT		CONFIG_PORT
+#define DATA_PORT		0x3f1
+#define CONFIG_ENTER		0x55
+#define CONFIG_EXIT		0xaa
+
+/* Configuration index */
+#define CURRENT_LDN_INDEX	0x07
+#define POWER_CONTROL_INDEX	0x22
+#define ACTIVATE_INDEX		0x30
+#define IO_BASE_HI_INDEX	0x60
+#define IO_BASE_LO_INDEX	0x61
+#define IRQ_SELECT_INDEX	0x70
+#define DMA_SELECT_INDEX	0x74
+
+#define GPIO46_INDEX		0xc6
+#define GPIO47_INDEX		0xc7
+
+/* UART stuff. Only for debugging.  */
+/* UART Register */
+
+#define UART_RBR	0x0	/* Receiver Buffer Register (Read Only) */
+#define UART_THR	0x0	/* Transmitter Holding Register (Write Only) */
+#define UART_IER	0x2	/* Interrupt Enable Register */
+#define UART_IIR	0x4	/* Interrupt Ident Register (Read Only) */
+#define UART_FCR	0x4	/* FIFO Control Register (Write Only) */
+#define UART_LCR	0x6	/* Line Control Register */
+#define UART_MCR	0x8	/* MODEM Control Register */
+#define UART_LSR	0xa	/* Line Status Register */
+#define UART_MSR	0xc	/* MODEM Status Register */
+#define UART_SCR	0xe	/* Scratch Register */
+#define UART_DLL	0x0	/* Divisor Latch (LS) */
+#define UART_DLM	0x2	/* Divisor Latch (MS) */
+
+#ifndef __ASSEMBLY__
+typedef struct uart_reg {
+	volatile __u16 rbr;
+	volatile __u16 ier;
+	volatile __u16 iir;
+	volatile __u16 lcr;
+	volatile __u16 mcr;
+	volatile __u16 lsr;
+	volatile __u16 msr;
+	volatile __u16 scr;
+} uart_reg;
+#endif /* ! __ASSEMBLY__ */
+
+/* Alias for Write Only Register */
+
+#define thr	rbr
+#define tcr	iir
+
+/* Alias for Divisor Latch Register */
+
+#define dll	rbr
+#define dlm	ier
+#define fcr	iir
+
+/* Interrupt Enable Register */
+
+#define IER_ERDAI	0x0100	/* Enable Received Data Available Interrupt */
+#define IER_ETHREI	0x0200	/* Enable Transmitter Holding Register Empty Interrupt */
+#define IER_ELSI	0x0400	/* Enable Receiver Line Status Interrupt */
+#define IER_EMSI	0x0800	/* Enable MODEM Status Interrupt */
+
+/* Interrupt Ident Register */
+
+#define IIR_IP		0x0100	/* "0" if Interrupt Pending */
+#define IIR_IIB0	0x0200	/* Interrupt ID Bit 0 */
+#define IIR_IIB1	0x0400	/* Interrupt ID Bit 1 */
+#define IIR_IIB2	0x0800	/* Interrupt ID Bit 2 */
+#define IIR_FIFO	0xc000	/* FIFOs enabled */
+
+/* FIFO Control Register */
+
+#define FCR_FEN		0x0100	/* FIFO enable */
+#define FCR_RFRES	0x0200	/* Receiver FIFO reset */
+#define FCR_TFRES	0x0400	/* Transmitter FIFO reset */
+#define FCR_DMA		0x0800	/* DMA mode select */
+#define FCR_RTL		0x4000	/* Receiver triger (LSB) */
+#define FCR_RTM		0x8000	/* Receiver triger (MSB) */
+
+/* Line Control Register */
+
+#define LCR_WLS0	0x0100	/* Word Length Select Bit 0 */
+#define LCR_WLS1	0x0200	/* Word Length Select Bit 1 */
+#define LCR_STB		0x0400	/* Number of Stop Bits */
+#define LCR_PEN		0x0800	/* Parity Enable */
+#define LCR_EPS		0x1000	/* Even Parity Select */
+#define LCR_SP		0x2000	/* Stick Parity */
+#define LCR_SB		0x4000	/* Set Break */
+#define LCR_DLAB	0x8000	/* Divisor Latch Access Bit */
+
+/* MODEM Control Register */
+
+#define MCR_DTR		0x0100	/* Data Terminal Ready */
+#define MCR_RTS		0x0200	/* Request to Send */
+#define MCR_OUT1	0x0400	/* Out 1 */
+#define MCR_IRQEN	0x0800	/* IRQ Enable */
+#define MCR_LOOP	0x1000	/* Loop */
+
+/* Line Status Register */
+
+#define LSR_DR		0x0100	/* Data Ready */
+#define LSR_OE		0x0200	/* Overrun Error */
+#define LSR_PE		0x0400	/* Parity Error */
+#define LSR_FE		0x0800	/* Framing Error */
+#define LSR_BI		0x1000	/* Break Interrupt */
+#define LSR_THRE	0x2000	/* Transmitter Holding Register Empty */
+#define LSR_TEMT	0x4000	/* Transmitter Empty */
+#define LSR_FIFOE	0x8000	/* Receiver FIFO error */
+
+/* MODEM Status Register */
+
+#define MSR_DCTS	0x0100	/* Delta Clear to Send */
+#define MSR_DDSR	0x0200	/* Delta Data Set Ready */
+#define MSR_TERI	0x0400	/* Trailing Edge Ring Indicator */
+#define MSR_DDCD	0x0800	/* Delta Data Carrier Detect */
+#define MSR_CTS		0x1000	/* Clear to Send */
+#define MSR_DSR		0x2000	/* Data Set Ready */
+#define MSR_RI		0x4000	/* Ring Indicator */
+#define MSR_DCD		0x8000	/* Data Carrier Detect */
+
+/* Baud Rate Divisor */
+
+#define UART_CLK	(1843200)	/* 1.8432 MHz */
+#define UART_BAUD(x)	(UART_CLK / (16 * (x)))
+
+/* RTC register definition */
+#define RTC_SECONDS             0
+#define RTC_SECONDS_ALARM       1
+#define RTC_MINUTES             2
+#define RTC_MINUTES_ALARM       3
+#define RTC_HOURS               4
+#define RTC_HOURS_ALARM         5
+#define RTC_DAY_OF_WEEK         6
+#define RTC_DAY_OF_MONTH        7
+#define RTC_MONTH               8
+#define RTC_YEAR                9
+#define RTC_FREQ_SELECT		10
+# define RTC_UIP 0x80
+# define RTC_DIV_CTL 0x70
+/* This RTC can work under 32.768KHz clock only.  */
+# define RTC_OSC_ENABLE 0x20
+# define RTC_OSC_DISABLE 0x00
+#define RTC_CONTROL     	11
+# define RTC_SET 0x80
+# define RTC_PIE 0x40
+# define RTC_AIE 0x20
+# define RTC_UIE 0x10
+# define RTC_SQWE 0x08
+# define RTC_DM_BINARY 0x04
+# define RTC_24H 0x02
+# define RTC_DST_EN 0x01
+
+#endif  /* __ASM_SH_SMC37C93X_H */
-- 
cgit v1.2.3


From e4e3b5ccd77226c9c4dbb0737106b868dfc182d9 Mon Sep 17 00:00:00 2001
From: Tom Rini <trini@kernel.crashing.org>
Date: Wed, 27 Sep 2006 11:28:20 +0900
Subject: sh: Add a simple cmpxchg().

We didn't have one of these before, a simple implementation
borrowed from MIPS as well as the __HAVE_ARCH_CMPXCHG bits.

Signed-off-by: Tom Rini <trini@kernel.crashing.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index ad35ad4958f4..e89728d405d8 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -6,6 +6,7 @@
  * Copyright (C) 2002 Paul Mundt
  */
 
+#include <asm/types.h>
 
 /*
  *	switch_to() should switch tasks to task nr n, first
@@ -260,6 +261,45 @@ static __inline__ unsigned long __xchg(unsigned long x, volatile void * ptr, int
 	return x;
 }
 
+static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
+	unsigned long new)
+{
+	__u32 retval;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	retval = *m;
+	if (retval == old)
+		*m = new;
+	local_irq_restore(flags);       /* implies memory barrier  */
+	return retval;
+}
+
+/* This function doesn't exist, so you'll get a linker error
+ * if something tries to do an invalid cmpxchg(). */
+extern void __cmpxchg_called_with_bad_pointer(void);
+
+#define __HAVE_ARCH_CMPXCHG 1
+
+static inline unsigned long __cmpxchg(volatile void * ptr, unsigned long old,
+		unsigned long new, int size)
+{
+	switch (size) {
+	case 4:
+		return __cmpxchg_u32(ptr, old, new);
+	}
+	__cmpxchg_called_with_bad_pointer();
+	return old;
+}
+
+#define cmpxchg(ptr,o,n)						 \
+  ({									 \
+     __typeof__(*(ptr)) _o_ = (o);					 \
+     __typeof__(*(ptr)) _n_ = (n);					 \
+     (__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_,		 \
+				    (unsigned long)_n_, sizeof(*(ptr))); \
+  })
+
 /* XXX
  * disable hlt during certain critical i/o operations
  */
-- 
cgit v1.2.3


From e4c2cfee5d5cf3e4c16b423be23551aeddf2717b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 12:31:01 +0900
Subject: sh: Various cosmetic cleanups.

We had quite a bit of whitespace damage, clean most of it up..

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Arthur Othieno <a.othieno@bluewin.ch>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/dreamcast/irq.c              |  8 ++--
 arch/sh/boards/dreamcast/rtc.c              | 18 ++++-----
 arch/sh/boards/dreamcast/setup.c            |  6 +--
 arch/sh/boards/renesas/edosk7705/Makefile   |  4 --
 arch/sh/boards/renesas/hs7751rvoip/Makefile |  4 --
 arch/sh/boards/renesas/hs7751rvoip/io.c     |  4 +-
 arch/sh/boards/renesas/rts7751r2d/Makefile  |  4 --
 arch/sh/boards/renesas/rts7751r2d/led.c     |  2 -
 arch/sh/boards/renesas/rts7751r2d/setup.c   |  3 --
 arch/sh/drivers/dma/dma-pvr2.c              |  5 +--
 arch/sh/kernel/cpu/clock.c                  |  2 +-
 arch/sh/kernel/cpu/sh4/sq.c                 |  3 +-
 arch/sh/kernel/entry.S                      | 18 ---------
 arch/sh/kernel/ptrace.c                     |  1 -
 arch/sh/kernel/sh_ksyms.c                   |  2 +
 drivers/char/watchdog/shwdt.c               | 58 ++++++++++++-----------------
 include/asm-sh/atomic.h                     |  1 +
 include/asm-sh/bitops.h                     | 16 ++++----
 include/asm-sh/checksum.h                   |  2 +
 include/asm-sh/fixmap.h                     |  2 +-
 include/asm-sh/watchdog.h                   |  3 --
 21 files changed, 60 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boards/dreamcast/irq.c b/arch/sh/boards/dreamcast/irq.c
index b10a6b11c034..373a22e48ac4 100644
--- a/arch/sh/boards/dreamcast/irq.c
+++ b/arch/sh/boards/dreamcast/irq.c
@@ -26,10 +26,10 @@
    event.
 
    There are three 32-bit ESRs located at 0xa05f8900 - 0xa05f6908.  Event
-   types can be found in include/asm-sh/dc_sysasic.h.  There are three groups
-   of EMRs that parallel the ESRs.  Each EMR group corresponds to an IRQ, so
-   0xa05f6910 - 0xa05f6918 triggers IRQ 13, 0xa05f6920 - 0xa05f6928 triggers
-   IRQ 11, and 0xa05f6930 - 0xa05f6938 triggers IRQ 9.
+   types can be found in include/asm-sh/dreamcast/sysasic.h. There are three
+   groups of EMRs that parallel the ESRs.  Each EMR group corresponds to an
+   IRQ, so 0xa05f6910 - 0xa05f6918 triggers IRQ 13, 0xa05f6920 - 0xa05f6928
+   triggers IRQ 11, and 0xa05f6930 - 0xa05f6938 triggers IRQ 9.
 
    In the kernel, these events are mapped to virtual IRQs so that drivers can
    respond to them as they would a normal interrupt.  In order to keep this
diff --git a/arch/sh/boards/dreamcast/rtc.c b/arch/sh/boards/dreamcast/rtc.c
index 379de1629134..44420494dc22 100644
--- a/arch/sh/boards/dreamcast/rtc.c
+++ b/arch/sh/boards/dreamcast/rtc.c
@@ -1,4 +1,5 @@
-/* arch/sh/kernel/rtc-aica.c
+/*
+ * arch/sh/boards/dreamcast/rtc.c
  *
  * Dreamcast AICA RTC routines.
  *
@@ -10,15 +11,12 @@
  */
 
 #include <linux/time.h>
-
+#include <asm/rtc.h>
 #include <asm/io.h>
 
-extern void (*rtc_get_time)(struct timespec *);
-extern int (*rtc_set_time)(const time_t);
-
 /* The AICA RTC has an Epoch of 1/1/1950, so we must subtract 20 years (in
-   seconds to get the standard Unix Epoch when getting the time, and add 20
-   years when setting the time. */
+   seconds) to get the standard Unix Epoch when getting the time, and add
+   20 years when setting the time. */
 #define TWENTY_YEARS ((20 * 365LU + 5) * 86400)
 
 /* The AICA RTC is represented by a 32-bit seconds counter stored in 2 16-bit
@@ -32,7 +30,8 @@ extern int (*rtc_set_time)(const time_t);
  *
  * Grabs the current RTC seconds counter and adjusts it to the Unix Epoch.
  */
-void aica_rtc_gettimeofday(struct timespec *ts) {
+void aica_rtc_gettimeofday(struct timespec *ts)
+{
 	unsigned long val1, val2;
 
 	do {
@@ -55,7 +54,8 @@ void aica_rtc_gettimeofday(struct timespec *ts) {
  *
  * Adjusts the given @tv to the AICA Epoch and sets the RTC seconds counter.
  */
-int aica_rtc_settimeofday(const time_t secs) {
+int aica_rtc_settimeofday(const time_t secs)
+{
 	unsigned long val1, val2;
 	unsigned long adj = secs + TWENTY_YEARS;
 
diff --git a/arch/sh/boards/dreamcast/setup.c b/arch/sh/boards/dreamcast/setup.c
index 0027b80a2343..a00ac94beb05 100644
--- a/arch/sh/boards/dreamcast/setup.c
+++ b/arch/sh/boards/dreamcast/setup.c
@@ -25,18 +25,16 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
+#include <asm/rtc.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
 #include <asm/mach/sysasic.h>
 
 extern struct hw_interrupt_type systemasic_int;
-/* XXX: Move this into it's proper header. */
-extern void (*board_time_init)(void);
 extern void aica_time_init(void);
 extern int gapspci_init(void);
 extern int systemasic_irq_demux(int);
 
-void *dreamcast_consistent_alloc(struct device *, size_t, dma_addr_t *, int);
+void *dreamcast_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
 int dreamcast_consistent_free(struct device *, size_t, void *, dma_addr_t);
 
 const char *get_system_type(void)
diff --git a/arch/sh/boards/renesas/edosk7705/Makefile b/arch/sh/boards/renesas/edosk7705/Makefile
index 7fccbf2e4a1d..14bdd531f116 100644
--- a/arch/sh/boards/renesas/edosk7705/Makefile
+++ b/arch/sh/boards/renesas/edosk7705/Makefile
@@ -1,10 +1,6 @@
 #
 # Makefile for the EDOSK7705 specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
 
 obj-y	 := setup.o io.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Makefile b/arch/sh/boards/renesas/hs7751rvoip/Makefile
index e8b4109ace11..3ef8cbaa0b37 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/Makefile
+++ b/arch/sh/boards/renesas/hs7751rvoip/Makefile
@@ -1,10 +1,6 @@
 #
 # Makefile for the HS7751RVoIP specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
 
 obj-y	 := mach.o setup.o io.o irq.o led.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index 3a1abfa2fefb..09fb77ffb835 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -167,7 +167,7 @@ void hs7751rvoip_outb(unsigned char value, unsigned long port)
                 *(volatile unsigned char *)port = value;
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned cjar *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
         	*(unsigned char *)PCI_IOMAP(port) = value;
@@ -181,7 +181,7 @@ void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
                 *(volatile unsigned char *)port = value;
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned cjar *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
         	*(unsigned char *)PCI_IOMAP(port) = value;
diff --git a/arch/sh/boards/renesas/rts7751r2d/Makefile b/arch/sh/boards/renesas/rts7751r2d/Makefile
index daa53334bdc3..eee1ed6f5727 100644
--- a/arch/sh/boards/renesas/rts7751r2d/Makefile
+++ b/arch/sh/boards/renesas/rts7751r2d/Makefile
@@ -1,10 +1,6 @@
 #
 # Makefile for the RTS7751R2D specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
 
 obj-y	 := mach.o setup.o io.o irq.o led.o
 
diff --git a/arch/sh/boards/renesas/rts7751r2d/led.c b/arch/sh/boards/renesas/rts7751r2d/led.c
index 4d16de71fac1..cf35f90dfe03 100644
--- a/arch/sh/boards/renesas/rts7751r2d/led.c
+++ b/arch/sh/boards/renesas/rts7751r2d/led.c
@@ -12,8 +12,6 @@
 #include <asm/io.h>
 #include <asm/rts7751r2d/rts7751r2d.h>
 
-extern unsigned int debug_counter;
-
 #ifdef CONFIG_HEARTBEAT
 
 #include <linux/sched.h>
diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index 7953dde0b91c..60907f574f34 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -14,8 +14,6 @@
 #include <asm/io.h>
 #include <asm/rts7751r2d/rts7751r2d.h>
 
-unsigned int debug_counter;
-
 const char *get_system_type(void)
 {
 	return "RTS7751R2D";
@@ -34,5 +32,4 @@ void __init platform_setup(void)
 	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
 	ctrl_outw(0x0000, PA_OUTPORT);
 	pm_power_off = rts7751r2d_power_off;
-	debug_counter = 0;
 }
diff --git a/arch/sh/drivers/dma/dma-pvr2.c b/arch/sh/drivers/dma/dma-pvr2.c
index 30a580aa7cbd..3b0b0f60bb3c 100644
--- a/arch/sh/drivers/dma/dma-pvr2.c
+++ b/arch/sh/drivers/dma/dma-pvr2.c
@@ -18,8 +18,8 @@
 #include <asm/dma.h>
 #include <asm/io.h>
 
-static unsigned int xfer_complete = 0;
-static int count = 0;
+static unsigned int xfer_complete;
+static int count;
 
 static irqreturn_t pvr2_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
@@ -107,4 +107,3 @@ module_exit(pvr2_dma_exit);
 MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>");
 MODULE_DESCRIPTION("NEC PowerVR 2 DMA driver");
 MODULE_LICENSE("GPL");
-
diff --git a/arch/sh/kernel/cpu/clock.c b/arch/sh/kernel/cpu/clock.c
index 97fa37f42b84..e7eef07536f5 100644
--- a/arch/sh/kernel/cpu/clock.c
+++ b/arch/sh/kernel/cpu/clock.c
@@ -225,7 +225,7 @@ int __init clk_init(void)
 {
 	int i, ret = 0;
 
-	BUG_ON(unlikely(!master_clk.rate));
+	BUG_ON(!master_clk.rate);
 
 	for (i = 0; i < ARRAY_SIZE(onchip_clocks); i++) {
 		struct clk *clk = onchip_clocks[i];
diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index b09805f3ee23..00f6a3c8c43b 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -24,9 +24,10 @@
 #include <linux/proc_fs.h>
 #include <linux/miscdevice.h>
 #include <linux/vmalloc.h>
-
+#include <linux/mm.h>
 #include <asm/io.h>
 #include <asm/page.h>
+#include <asm/cacheflush.h>
 #include <asm/mmu_context.h>
 #include <asm/cpu/sq.h>
 
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 1a05804caf37..306eb8d83de5 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -18,24 +18,6 @@
 #include <asm/cpu/mmu_context.h>
 #include <asm/unistd.h>
 
-#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
-#define sys_nfsservctl		sys_ni_syscall
-#endif
-
-#if !defined(CONFIG_MMU)
-#define sys_madvise		sys_ni_syscall
-#define sys_readahead		sys_ni_syscall
-#define sys_mprotect		sys_ni_syscall
-#define sys_msync		sys_ni_syscall
-#define sys_mlock		sys_ni_syscall
-#define sys_munlock		sys_ni_syscall
-#define sys_mlockall		sys_ni_syscall
-#define sys_munlockall		sys_ni_syscall
-#define sys_mremap		sys_ni_syscall
-#define sys_mincore		sys_ni_syscall
-#define sys_remap_file_pages	sys_ni_syscall
-#endif
-
 ! NOTE:
 ! GNU as (as of 2.9.1) changes bf/s into bt/s and bra, when the address
 ! to be jumped is too far, but it causes illegal slot exception.
diff --git a/arch/sh/kernel/ptrace.c b/arch/sh/kernel/ptrace.c
index f7eebbde3291..04ca13a041c1 100644
--- a/arch/sh/kernel/ptrace.c
+++ b/arch/sh/kernel/ptrace.c
@@ -224,7 +224,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_SETDSPREGS: {
 		unsigned long dp;
-		int i;
 
 		ret = -EIO;
 		dp = ((unsigned long) child) + THREAD_SIZE -
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 245ed8f945e8..8cad80cec9a6 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -117,6 +117,8 @@ EXPORT_SYMBOL(synchronize_irq);
 #endif
 
 EXPORT_SYMBOL(csum_partial);
+#ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
+#endif
 EXPORT_SYMBOL(consistent_sync);
 EXPORT_SYMBOL(clear_page);
diff --git a/drivers/char/watchdog/shwdt.c b/drivers/char/watchdog/shwdt.c
index 1355038f1044..7c39f39effdf 100644
--- a/drivers/char/watchdog/shwdt.c
+++ b/drivers/char/watchdog/shwdt.c
@@ -125,7 +125,6 @@ static void sh_wdt_start(void)
 
 /**
  * 	sh_wdt_stop - Stop the Watchdog
- *
  * 	Stops the watchdog.
  */
 static void sh_wdt_stop(void)
@@ -141,22 +140,20 @@ static void sh_wdt_stop(void)
 
 /**
  * 	sh_wdt_keepalive - Keep the Userspace Watchdog Alive
- *
  * 	The Userspace watchdog got a KeepAlive: schedule the next heartbeat.
  */
-static void sh_wdt_keepalive(void)
+static inline void sh_wdt_keepalive(void)
 {
 	next_heartbeat = jiffies + (heartbeat * HZ);
 }
 
 /**
  * 	sh_wdt_set_heartbeat - Set the Userspace Watchdog heartbeat
- *
  * 	Set the Userspace Watchdog heartbeat
  */
 static int sh_wdt_set_heartbeat(int t)
 {
-	if ((t < 1) || (t > 3600)) /* arbitrary upper limit */
+	if (unlikely((t < 1) || (t > 3600))) /* arbitrary upper limit */
 		return -EINVAL;
 
 	heartbeat = t;
@@ -165,7 +162,6 @@ static int sh_wdt_set_heartbeat(int t)
 
 /**
  * 	sh_wdt_ping - Ping the Watchdog
- *
  *	@data: Unused
  *
  * 	Clears overflow bit, resets timer counter.
@@ -182,14 +178,13 @@ static void sh_wdt_ping(unsigned long data)
 		sh_wdt_write_cnt(0);
 
 		mod_timer(&timer, next_ping_period(clock_division_ratio));
-	} else {
-		printk(KERN_WARNING PFX "Heartbeat lost! Will not ping the watchdog\n");
-	}
+	} else
+		printk(KERN_WARNING PFX "Heartbeat lost! Will not ping "
+		       "the watchdog\n");
 }
 
 /**
  * 	sh_wdt_open - Open the Device
- *
  * 	@inode: inode of device
  * 	@file: file handle of device
  *
@@ -209,7 +204,6 @@ static int sh_wdt_open(struct inode *inode, struct file *file)
 
 /**
  * 	sh_wdt_close - Close the Device
- *
  * 	@inode: inode of device
  * 	@file: file handle of device
  *
@@ -220,7 +214,8 @@ static int sh_wdt_close(struct inode *inode, struct file *file)
 	if (shwdt_expect_close == 42) {
 		sh_wdt_stop();
 	} else {
-		printk(KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n");
+		printk(KERN_CRIT PFX "Unexpected close, not "
+		       "stopping watchdog!\n");
 		sh_wdt_keepalive();
 	}
 
@@ -232,7 +227,6 @@ static int sh_wdt_close(struct inode *inode, struct file *file)
 
 /**
  * 	sh_wdt_write - Write to Device
- *
  * 	@file: file handle of device
  * 	@buf: buffer to write
  * 	@count: length of buffer
@@ -265,7 +259,6 @@ static ssize_t sh_wdt_write(struct file *file, const char *buf,
 
 /**
  * 	sh_wdt_ioctl - Query Device
- *
  * 	@inode: inode of device
  * 	@file: file handle of device
  * 	@cmd: watchdog command
@@ -326,7 +319,6 @@ static int sh_wdt_ioctl(struct inode *inode, struct file *file,
 
 /**
  * 	sh_wdt_notify_sys - Notifier Handler
- *
  * 	@this: notifier block
  * 	@code: notifier event
  * 	@unused: unused
@@ -337,9 +329,8 @@ static int sh_wdt_ioctl(struct inode *inode, struct file *file,
 static int sh_wdt_notify_sys(struct notifier_block *this,
 			     unsigned long code, void *unused)
 {
-	if (code == SYS_DOWN || code == SYS_HALT) {
+	if (code == SYS_DOWN || code == SYS_HALT)
 		sh_wdt_stop();
-	}
 
 	return NOTIFY_DONE;
 }
@@ -354,7 +345,8 @@ static const struct file_operations sh_wdt_fops = {
 };
 
 static struct watchdog_info sh_wdt_info = {
-	.options		= WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE,
+	.options		= WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT |
+				  WDIOF_MAGICCLOSE,
 	.firmware_version	= 1,
 	.identity		= "SH WDT",
 };
@@ -371,7 +363,6 @@ static struct miscdevice sh_wdt_miscdev = {
 
 /**
  * 	sh_wdt_init - Initialize module
- *
  * 	Registers the device and notifier handler. Actual device
  * 	initialization is handled by sh_wdt_open().
  */
@@ -381,15 +372,15 @@ static int __init sh_wdt_init(void)
 
 	if ((clock_division_ratio < 0x5) || (clock_division_ratio > 0x7)) {
 		clock_division_ratio = WTCSR_CKS_4096;
-		printk(KERN_INFO PFX "clock_division_ratio value must be 0x5<=x<=0x7, using %d\n",
-			clock_division_ratio);
+		printk(KERN_INFO PFX "clock_division_ratio value must "
+		       "be 0x5<=x<=0x7, using %d\n", clock_division_ratio);
 	}
 
-	if (sh_wdt_set_heartbeat(heartbeat))
-	{
+	rc = sh_wdt_set_heartbeat(heartbeat);
+	if (unlikely(rc)) {
 		heartbeat = WATCHDOG_HEARTBEAT;
-		printk(KERN_INFO PFX "heartbeat value must be 1<=x<=3600, using %d\n",
-			heartbeat);
+		printk(KERN_INFO PFX "heartbeat value must "
+		       "be 1<=x<=3600, using %d\n", heartbeat);
 	}
 
 	init_timer(&timer);
@@ -397,15 +388,16 @@ static int __init sh_wdt_init(void)
 	timer.data = 0;
 
 	rc = register_reboot_notifier(&sh_wdt_notifier);
-	if (rc) {
-		printk(KERN_ERR PFX "Can't register reboot notifier (err=%d)\n", rc);
+	if (unlikely(rc)) {
+		printk(KERN_ERR PFX "Can't register reboot notifier (err=%d)\n",
+		       rc);
 		return rc;
 	}
 
 	rc = misc_register(&sh_wdt_miscdev);
-	if (rc) {
-		printk(KERN_ERR PFX "Can't register miscdev on minor=%d (err=%d)\n",
-			sh_wdt_miscdev.minor, rc);
+	if (unlikely(rc)) {
+		printk(KERN_ERR PFX "Can't register miscdev on "
+		       "minor=%d (err=%d)\n", sh_wdt_miscdev.minor, rc);
 		unregister_reboot_notifier(&sh_wdt_notifier);
 		return rc;
 	}
@@ -418,7 +410,6 @@ static int __init sh_wdt_init(void)
 
 /**
  * 	sh_wdt_exit - Deinitialize module
- *
  * 	Unregisters the device and notifier handler. Actual device
  * 	deinitialization is handled by sh_wdt_close().
  */
@@ -434,14 +425,13 @@ MODULE_LICENSE("GPL");
 MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(clock_division_ratio, int, 0);
-MODULE_PARM_DESC(clock_division_ratio, "Clock division ratio. Valid ranges are from 0x5 (1.31ms) to 0x7 (5.25ms). Defaults to 0x7.");
+MODULE_PARM_DESC(clock_division_ratio, "Clock division ratio. Valid ranges are from 0x5 (1.31ms) to 0x7 (5.25ms). (default=" __MODULE_STRING(clock_division_ratio) ")");
 
 module_param(heartbeat, int, 0);
 MODULE_PARM_DESC(heartbeat, "Watchdog heartbeat in seconds. (1<=heartbeat<=3600, default=" __MODULE_STRING(WATCHDOG_HEARTBEAT) ")");
 
 module_param(nowayout, int, 0);
-MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CONFIG_WATCHDOG_NOWAYOUT)");
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
 
 module_init(sh_wdt_init);
 module_exit(sh_wdt_exit);
-
diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h
index fb627de217f2..049eb2dda6b6 100644
--- a/include/asm-sh/atomic.h
+++ b/include/asm-sh/atomic.h
@@ -14,6 +14,7 @@ typedef struct { volatile int counter; } atomic_t;
 #define atomic_read(v)		((v)->counter)
 #define atomic_set(v,i)		((v)->counter = (i))
 
+#include <linux/compiler.h>
 #include <asm/system.h>
 
 /*
diff --git a/include/asm-sh/bitops.h b/include/asm-sh/bitops.h
index e34f82508568..1c16792cee1d 100644
--- a/include/asm-sh/bitops.h
+++ b/include/asm-sh/bitops.h
@@ -6,7 +6,7 @@
 /* For __swab32 */
 #include <asm/byteorder.h>
 
-static __inline__ void set_bit(int nr, volatile void * addr)
+static inline void set_bit(int nr, volatile void * addr)
 {
 	int	mask;
 	volatile unsigned int *a = addr;
@@ -24,7 +24,7 @@ static __inline__ void set_bit(int nr, volatile void * addr)
  */
 #define smp_mb__before_clear_bit()	barrier()
 #define smp_mb__after_clear_bit()	barrier()
-static __inline__ void clear_bit(int nr, volatile void * addr)
+static inline void clear_bit(int nr, volatile void * addr)
 {
 	int	mask;
 	volatile unsigned int *a = addr;
@@ -37,7 +37,7 @@ static __inline__ void clear_bit(int nr, volatile void * addr)
 	local_irq_restore(flags);
 }
 
-static __inline__ void change_bit(int nr, volatile void * addr)
+static inline void change_bit(int nr, volatile void * addr)
 {
 	int	mask;
 	volatile unsigned int *a = addr;
@@ -50,7 +50,7 @@ static __inline__ void change_bit(int nr, volatile void * addr)
 	local_irq_restore(flags);
 }
 
-static __inline__ int test_and_set_bit(int nr, volatile void * addr)
+static inline int test_and_set_bit(int nr, volatile void * addr)
 {
 	int	mask, retval;
 	volatile unsigned int *a = addr;
@@ -66,7 +66,7 @@ static __inline__ int test_and_set_bit(int nr, volatile void * addr)
 	return retval;
 }
 
-static __inline__ int test_and_clear_bit(int nr, volatile void * addr)
+static inline int test_and_clear_bit(int nr, volatile void * addr)
 {
 	int	mask, retval;
 	volatile unsigned int *a = addr;
@@ -82,7 +82,7 @@ static __inline__ int test_and_clear_bit(int nr, volatile void * addr)
 	return retval;
 }
 
-static __inline__ int test_and_change_bit(int nr, volatile void * addr)
+static inline int test_and_change_bit(int nr, volatile void * addr)
 {
 	int	mask, retval;
 	volatile unsigned int *a = addr;
@@ -100,7 +100,7 @@ static __inline__ int test_and_change_bit(int nr, volatile void * addr)
 
 #include <asm-generic/bitops/non-atomic.h>
 
-static __inline__ unsigned long ffz(unsigned long word)
+static inline unsigned long ffz(unsigned long word)
 {
 	unsigned long result;
 
@@ -120,7 +120,7 @@ static __inline__ unsigned long ffz(unsigned long word)
  *
  * Undefined if no bit exists, so code should check against 0 first.
  */
-static __inline__ unsigned long __ffs(unsigned long word)
+static inline unsigned long __ffs(unsigned long word)
 {
 	unsigned long result;
 
diff --git a/include/asm-sh/checksum.h b/include/asm-sh/checksum.h
index fa03b30c4269..08168afe6746 100644
--- a/include/asm-sh/checksum.h
+++ b/include/asm-sh/checksum.h
@@ -159,6 +159,7 @@ static __inline__ unsigned short ip_compute_csum(unsigned char * buff, int len)
 }
 
 #define _HAVE_ARCH_IPV6_CSUM
+#ifdef CONFIG_IPV6
 static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr,
 						     struct in6_addr *daddr,
 						     __u32 len,
@@ -194,6 +195,7 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr,
 
 	return csum_fold(sum);
 }
+#endif
 
 /* 
  *	Copy and checksum to user
diff --git a/include/asm-sh/fixmap.h b/include/asm-sh/fixmap.h
index 412bccaa07e6..458e9fa59545 100644
--- a/include/asm-sh/fixmap.h
+++ b/include/asm-sh/fixmap.h
@@ -25,7 +25,7 @@
  * addresses. The point is to have a constant address at
  * compile time, but to set the physical address only
  * in the boot process. We allocate these special  addresses
- * from the end of virtual memory (0xfffff000) backwards.
+ * from the end of P3 backwards.
  * Also this lets us do fail-safe vmalloc(), we
  * can guarantee that these special addresses and
  * vmalloc()-ed addresses never overlap.
diff --git a/include/asm-sh/watchdog.h b/include/asm-sh/watchdog.h
index 09ca41972a11..d19ea62ef8c6 100644
--- a/include/asm-sh/watchdog.h
+++ b/include/asm-sh/watchdog.h
@@ -62,7 +62,6 @@
 
 /**
  * 	sh_wdt_read_cnt - Read from Counter
- *
  * 	Reads back the WTCNT value.
  */
 static inline __u8 sh_wdt_read_cnt(void)
@@ -72,7 +71,6 @@ static inline __u8 sh_wdt_read_cnt(void)
 
 /**
  *	sh_wdt_write_cnt - Write to Counter
- *
  *	@val: Value to write
  *
  *	Writes the given value @val to the lower byte of the timer counter.
@@ -95,7 +93,6 @@ static inline __u8 sh_wdt_read_csr(void)
 
 /**
  * 	sh_wdt_write_csr - Write to Control/Status Register
- *
  * 	@val: Value to write
  *
  * 	Writes the given value @val to the lower byte of the control/status
-- 
cgit v1.2.3


From 3f787fe2e077ecfe43ad9ad56f12a8e21cffafc3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:11:57 +0900
Subject: sh: hugetlb updates.

For some of the larger sizes we permitted spanning pages
across several PTEs, but this turned out to not be generally
useful. This reverts the sh hugetlbpage interface to something
more sensible using huge pages at single PTE granularity.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/hugetlbpage.c | 52 +++++++++++++++---------------------------------
 include/asm-sh/page.h    |  1 -
 2 files changed, 16 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c
index 2a85bc15a412..329059d6b54a 100644
--- a/arch/sh/mm/hugetlbpage.c
+++ b/arch/sh/mm/hugetlbpage.c
@@ -26,61 +26,41 @@
 pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte = NULL;
 
 	pgd = pgd_offset(mm, addr);
 	if (pgd) {
-		pmd = pmd_alloc(mm, pgd, addr);
-		if (pmd)
-			pte = pte_alloc_map(mm, pmd, addr);
+		pud = pud_alloc(mm, pgd, addr);
+		if (pud) {
+			pmd = pmd_alloc(mm, pud, addr);
+			if (pmd)
+				pte = pte_alloc_map(mm, pmd, addr);
+		}
 	}
+
 	return pte;
 }
 
 pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte = NULL;
 
 	pgd = pgd_offset(mm, addr);
 	if (pgd) {
-		pmd = pmd_offset(pgd, addr);
-		if (pmd)
-			pte = pte_offset_map(pmd, addr);
-	}
-	return pte;
-}
-
-void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
-		     pte_t *ptep, pte_t entry)
-{
-	int i;
-
-	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-		set_pte_at(mm, addr, ptep, entry);
-		ptep++;
-		addr += PAGE_SIZE;
-		pte_val(entry) += PAGE_SIZE;
+		pud = pud_offset(pgd, addr);
+		if (pud) {
+			pmd = pmd_offset(pud, addr);
+			if (pmd)
+				pte = pte_offset_map(pmd, addr);
+		}
 	}
-}
-
-pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
-			      pte_t *ptep)
-{
-	pte_t entry;
-	int i;
-
-	entry = *ptep;
 
-	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-		pte_clear(mm, addr, ptep);
-		addr += PAGE_SIZE;
-		ptep++;
-	}
-
-	return entry;
+	return pte;
 }
 
 struct page *follow_huge_addr(struct mm_struct *mm,
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 6f7eb8a3aba5..4811d410d123 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -30,7 +30,6 @@
 #define HPAGE_SIZE		(1UL << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE-1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT-PAGE_SHIFT)
-#define ARCH_HAS_SETCLEAR_HUGE_PTE
 #endif
 
 #ifdef __KERNEL__
-- 
cgit v1.2.3


From d95fb13c960ae19e9fd4a95807eb68fa20caf537 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:30:08 +0900
Subject: sh: Fixup TMU_TOCR definition for SH7300.

SH7300 has a different TMU_TOCR, make the TMU code work again.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh3/timer.h | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh3/timer.h b/include/asm-sh/cpu-sh3/timer.h
index 3d8e95e8d10c..2082ad956f21 100644
--- a/include/asm-sh/cpu-sh3/timer.h
+++ b/include/asm-sh/cpu-sh3/timer.h
@@ -23,6 +23,10 @@
  * ---------------------------------------------------------------------------
  */
 
+#if !defined(CONFIG_CPU_SUBTYPE_SH7727)
+#define TMU_TOCR	0xfffffe90	/* Byte access */
+#endif
+
 #if defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define TMU_TSTR	0xa412fe92	/* Byte access */
 
@@ -39,9 +43,6 @@
 #define TMU2_TCR	0xa412feb4	/* Word access */
 
 #else
-#if !defined(CONFIG_CPU_SUBTYPE_SH7727)
-#define TMU_TOCR	0xfffffe90	/* Byte access */
-#endif
 #define TMU_TSTR	0xfffffe92	/* Byte access */
 
 #define TMU0_TCOR	0xfffffe94	/* Long access */
-- 
cgit v1.2.3


From 6d75e650f1d0d59fd97c7629f0903ef18e8dfb7b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:42:57 +0900
Subject: sh: Move hd64461.h to a more sensible location.

With the I/O rework for hd64461 we're down to a single header,
so move it by itself and get rid of the directory.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/cchips/hd6446x/hd64461/io.c    |  11 +-
 arch/sh/cchips/hd6446x/hd64461/setup.c |   3 +-
 include/asm-sh/hd64461.h               | 203 +++++++++++++++++++++++++++++++++
 include/asm-sh/hd64461/hd64461.h       | 202 --------------------------------
 include/asm-sh/hd64461/io.h            |  43 -------
 include/asm-sh/hp6xx/io.h              |   2 +-
 6 files changed, 210 insertions(+), 254 deletions(-)
 create mode 100644 include/asm-sh/hd64461.h
 delete mode 100644 include/asm-sh/hd64461/hd64461.h
 delete mode 100644 include/asm-sh/hd64461/io.h

(limited to 'include')

diff --git a/arch/sh/cchips/hd6446x/hd64461/io.c b/arch/sh/cchips/hd6446x/hd64461/io.c
index ac3062671db7..f77f18f2ba7f 100644
--- a/arch/sh/cchips/hd6446x/hd64461/io.c
+++ b/arch/sh/cchips/hd6446x/hd64461/io.c
@@ -1,11 +1,10 @@
 /*
- *	$Id: io.c,v 1.6 2004/03/16 00:07:50 lethal Exp $
  *	Copyright (C) 2000 YAEGASHI Takeshi
  *	Typical I/O routines for HD64461 system.
  */
 
 #include <asm/io.h>
-#include <asm/hd64461/hd64461.h>
+#include <asm/hd64461.h>
 
 #define MEM_BASE (CONFIG_HD64461_IOBASE - HD64461_STBCR)
 
@@ -144,13 +143,13 @@ void hd64461_outsl(unsigned long port, const void *buffer, unsigned long count)
 	while(count--) *addr=*buf++;
 }
 
-unsigned short hd64461_readw(unsigned long addr)
+unsigned short hd64461_readw(void __iomem *addr)
 {
-	return *(volatile unsigned short*)(MEM_BASE+addr);
+	return ctrl_inw(MEM_BASE+(unsigned long __force)addr);
 }
 
-void hd64461_writew(unsigned short b, unsigned long addr)
+void hd64461_writew(unsigned short b, void __iomem *addr)
 {
-	*(volatile unsigned short*)(MEM_BASE+addr) = b;
+	ctrl_outw(b, MEM_BASE+(unsigned long __force)addr);
 }
 
diff --git a/arch/sh/cchips/hd6446x/hd64461/setup.c b/arch/sh/cchips/hd6446x/hd64461/setup.c
index ad126016720f..6c961273a23a 100644
--- a/arch/sh/cchips/hd6446x/hd64461/setup.c
+++ b/arch/sh/cchips/hd6446x/hd64461/setup.c
@@ -14,8 +14,7 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
-
-#include <asm/hd64461/hd64461.h>
+#include <asm/hd64461.h>
 
 static void disable_hd64461_irq(unsigned int irq)
 {
diff --git a/include/asm-sh/hd64461.h b/include/asm-sh/hd64461.h
new file mode 100644
index 000000000000..0f2e2132cc35
--- /dev/null
+++ b/include/asm-sh/hd64461.h
@@ -0,0 +1,203 @@
+#ifndef __ASM_SH_HD64461
+#define __ASM_SH_HD64461
+/*
+ *	$Id: hd64461.h,v 1.5 2004/03/16 00:07:51 lethal Exp $
+ *	Copyright (C) 2000 YAEGASHI Takeshi
+ *	Hitachi HD64461 companion chip support
+ */
+
+/* Constants for PCMCIA mappings */
+#define HD64461_PCC_WINDOW	0x01000000
+
+#define HD64461_PCC0_BASE	0xb8000000	/* area 6 */
+#define HD64461_PCC0_ATTR	(HD64461_PCC0_BASE)
+#define HD64461_PCC0_COMM	(HD64461_PCC0_BASE+HD64461_PCC_WINDOW)
+#define HD64461_PCC0_IO		(HD64461_PCC0_BASE+2*HD64461_PCC_WINDOW)
+
+#define HD64461_PCC1_BASE	0xb4000000	/* area 5 */
+#define HD64461_PCC1_ATTR	(HD64461_PCC1_BASE)
+#define HD64461_PCC1_COMM	(HD64461_PCC1_BASE+HD64461_PCC_WINDOW)
+
+#define HD64461_STBCR	0x10000
+#define HD64461_STBCR_CKIO_STBY			0x2000
+#define HD64461_STBCR_SAFECKE_IST		0x1000
+#define HD64461_STBCR_SLCKE_IST			0x0800
+#define HD64461_STBCR_SAFECKE_OST		0x0400
+#define HD64461_STBCR_SLCKE_OST			0x0200
+#define HD64461_STBCR_SMIAST			0x0100
+#define HD64461_STBCR_SLCDST			0x0080
+#define HD64461_STBCR_SPC0ST			0x0040
+#define HD64461_STBCR_SPC1ST			0x0020
+#define HD64461_STBCR_SAFEST			0x0010
+#define HD64461_STBCR_STM0ST			0x0008
+#define HD64461_STBCR_STM1ST			0x0004
+#define HD64461_STBCR_SIRST				0x0002
+#define HD64461_STBCR_SURTST			0x0001
+
+#define HD64461_SYSCR	0x10002
+#define HD64461_SCPUCR	0x10004
+
+#define HD64461_LCDCBAR		0x11000
+#define HD64461_LCDCLOR		0x11002
+#define HD64461_LCDCCR		0x11004
+#define HD64461_LCDCCR_MOFF	0x80
+
+#define	HD64461_LDR1		0x11010
+#define	HD64461_LDR1_DON	0x01
+#define	HD64461_LDR1_DINV	0x80
+
+#define	HD64461_LDR2		0x11012
+#define	HD64461_LDHNCR		0x11014
+#define	HD64461_LDHNSR		0x11016
+#define HD64461_LDVNTR		0x11018
+#define HD64461_LDVNDR		0x1101a
+#define HD64461_LDVSPR		0x1101c
+#define HD64461_LDR3		0x1101e
+
+#define HD64461_CPTWAR		0x11030
+#define HD64461_CPTWDR		0x11032
+#define HD64461_CPTRAR		0x11034
+#define HD64461_CPTRDR		0x11036
+
+#define HD64461_GRDOR		0x11040
+#define HD64461_GRSCR		0x11042
+#define HD64461_GRCFGR		0x11044
+#define HD64461_GRCFGR_ACCSTATUS		0x10
+#define HD64461_GRCFGR_ACCRESET			0x08
+#define HD64461_GRCFGR_ACCSTART_BITBLT	0x06
+#define HD64461_GRCFGR_ACCSTART_LINE	0x04
+#define HD64461_GRCFGR_COLORDEPTH16		0x01
+
+#define HD64461_LNSARH		0x11046
+#define HD64461_LNSARL		0x11048
+#define HD64461_LNAXLR		0x1104a
+#define HD64461_LNDGR		0x1104c
+#define HD64461_LNAXR		0x1104e
+#define HD64461_LNERTR		0x11050
+#define HD64461_LNMDR		0x11052
+#define HD64461_BBTSSARH	0x11054
+#define HD64461_BBTSSARL	0x11056
+#define HD64461_BBTDSARH	0x11058
+#define HD64461_BBTDSARL	0x1105a
+#define HD64461_BBTDWR		0x1105c
+#define HD64461_BBTDHR		0x1105e
+#define HD64461_BBTPARH		0x11060
+#define HD64461_BBTPARL		0x11062
+#define HD64461_BBTMARH		0x11064
+#define HD64461_BBTMARL		0x11066
+#define HD64461_BBTROPR		0x11068
+#define HD64461_BBTMDR		0x1106a
+
+/* PC Card Controller Registers */
+#define HD64461_PCC0ISR         0x12000 /* socket 0 interface status */
+#define HD64461_PCC0GCR         0x12002 /* socket 0 general control */
+#define HD64461_PCC0CSCR        0x12004 /* socket 0 card status change */
+#define HD64461_PCC0CSCIER      0x12006 /* socket 0 card status change interrupt enable */
+#define HD64461_PCC0SCR         0x12008 /* socket 0 software control */
+#define HD64461_PCC1ISR         0x12010 /* socket 1 interface status */
+#define HD64461_PCC1GCR         0x12012 /* socket 1 general control */
+#define HD64461_PCC1CSCR        0x12014 /* socket 1 card status change */
+#define HD64461_PCC1CSCIER      0x12016 /* socket 1 card status change interrupt enable */
+#define HD64461_PCC1SCR         0x12018 /* socket 1 software control */
+
+/* PCC Interface Status Register */
+#define HD64461_PCCISR_READY		0x80	/* card ready */
+#define HD64461_PCCISR_MWP		0x40	/* card write-protected */
+#define HD64461_PCCISR_VS2		0x20	/* voltage select pin 2 */
+#define HD64461_PCCISR_VS1		0x10	/* voltage select pin 1 */
+#define HD64461_PCCISR_CD2		0x08	/* card detect 2 */
+#define HD64461_PCCISR_CD1		0x04	/* card detect 1 */
+#define HD64461_PCCISR_BVD2		0x02	/* battery 1 */
+#define HD64461_PCCISR_BVD1		0x01	/* battery 1 */
+
+#define HD64461_PCCISR_PCD_MASK		0x0c    /* card detect */
+#define HD64461_PCCISR_BVD_MASK		0x03    /* battery voltage */
+#define HD64461_PCCISR_BVD_BATGOOD	0x03    /* battery good */
+#define HD64461_PCCISR_BVD_BATWARN	0x01    /* battery low warning */
+#define HD64461_PCCISR_BVD_BATDEAD1	0x02    /* battery dead */
+#define HD64461_PCCISR_BVD_BATDEAD2	0x00    /* battery dead */
+
+/* PCC General Control Register */
+#define HD64461_PCCGCR_DRVE		0x80    /* output drive */
+#define HD64461_PCCGCR_PCCR		0x40    /* PC card reset */
+#define HD64461_PCCGCR_PCCT		0x20    /* PC card type, 1=IO&mem, 0=mem */
+#define HD64461_PCCGCR_VCC0		0x10    /* voltage control pin VCC0SEL0 */
+#define HD64461_PCCGCR_PMMOD		0x08    /* memory mode */
+#define HD64461_PCCGCR_PA25		0x04    /* pin A25 */
+#define HD64461_PCCGCR_PA24		0x02    /* pin A24 */
+#define HD64461_PCCGCR_REG		0x01    /* pin PCC0REG# */
+
+/* PCC Card Status Change Register */
+#define HD64461_PCCCSCR_SCDI		0x80    /* sw card detect intr */
+#define HD64461_PCCCSCR_SRV1		0x40    /* reserved */
+#define HD64461_PCCCSCR_IREQ		0x20    /* IREQ intr req */
+#define HD64461_PCCCSCR_SC		0x10    /* STSCHG (status change) pin */
+#define HD64461_PCCCSCR_CDC		0x08    /* CD (card detect) change */
+#define HD64461_PCCCSCR_RC		0x04    /* READY change */
+#define HD64461_PCCCSCR_BW		0x02    /* battery warning change */
+#define HD64461_PCCCSCR_BD		0x01    /* battery dead change */
+
+/* PCC Card Status Change Interrupt Enable Register */
+#define HD64461_PCCCSCIER_CRE		0x80    /* change reset enable */
+#define HD64461_PCCCSCIER_IREQE_MASK	0x60   /* IREQ enable */
+#define HD64461_PCCCSCIER_IREQE_DISABLED	0x00   /* IREQ disabled */
+#define HD64461_PCCCSCIER_IREQE_LEVEL	0x20   /* IREQ level-triggered */
+#define HD64461_PCCCSCIER_IREQE_FALLING	0x40   /* IREQ falling-edge-trig */
+#define HD64461_PCCCSCIER_IREQE_RISING	0x60   /* IREQ rising-edge-trig */
+
+#define HD64461_PCCCSCIER_SCE		0x10    /* status change enable */
+#define HD64461_PCCCSCIER_CDE		0x08    /* card detect change enable */
+#define HD64461_PCCCSCIER_RE		0x04    /* ready change enable */
+#define HD64461_PCCCSCIER_BWE		0x02    /* battery warn change enable */
+#define HD64461_PCCCSCIER_BDE		0x01    /* battery dead change enable*/
+
+/* PCC Software Control Register */
+#define HD64461_PCCSCR_VCC1		0x02	/* voltage control pin 1 */
+#define HD64461_PCCSCR_SWP		0x01    /* write protect */
+
+#define HD64461_P0OCR           0x1202a
+#define HD64461_P1OCR           0x1202c
+#define HD64461_PGCR            0x1202e
+
+#define HD64461_GPACR		0x14000
+#define HD64461_GPBCR		0x14002
+#define HD64461_GPCCR		0x14004
+#define HD64461_GPDCR		0x14006
+#define HD64461_GPADR		0x14010
+#define HD64461_GPBDR		0x14012
+#define HD64461_GPCDR		0x14014
+#define HD64461_GPDDR		0x14016
+#define HD64461_GPAICR		0x14020
+#define HD64461_GPBICR		0x14022
+#define HD64461_GPCICR		0x14024
+#define HD64461_GPDICR		0x14026
+#define HD64461_GPAISR		0x14040
+#define HD64461_GPBISR		0x14042
+#define HD64461_GPCISR		0x14044
+#define HD64461_GPDISR		0x14046
+
+#define HD64461_NIRR		0x15000
+#define HD64461_NIMR		0x15002
+
+#define HD64461_IRQBASE		OFFCHIP_IRQ_BASE
+#define HD64461_IRQ_NUM		16
+
+#define HD64461_IRQ_UART	(HD64461_IRQBASE+5)
+#define HD64461_IRQ_IRDA	(HD64461_IRQBASE+6)
+#define HD64461_IRQ_TMU1	(HD64461_IRQBASE+9)
+#define HD64461_IRQ_TMU0	(HD64461_IRQBASE+10)
+#define HD64461_IRQ_GPIO	(HD64461_IRQBASE+11)
+#define HD64461_IRQ_AFE		(HD64461_IRQBASE+12)
+#define HD64461_IRQ_PCC1	(HD64461_IRQBASE+13)
+#define HD64461_IRQ_PCC0	(HD64461_IRQBASE+14)
+
+#define __IO_PREFIX	hd64461
+#include <asm/io_generic.h>
+
+/* arch/sh/cchips/hd6446x/hd64461/setup.c */
+int hd64461_irq_demux(int irq);
+void hd64461_register_irq_demux(int irq,
+				int (*demux) (int irq, void *dev), void *dev);
+void hd64461_unregister_irq_demux(int irq);
+
+#endif
diff --git a/include/asm-sh/hd64461/hd64461.h b/include/asm-sh/hd64461/hd64461.h
deleted file mode 100644
index 87f13d24c630..000000000000
--- a/include/asm-sh/hd64461/hd64461.h
+++ /dev/null
@@ -1,202 +0,0 @@
-#ifndef __ASM_SH_HD64461
-#define __ASM_SH_HD64461
-/*
- *	$Id: hd64461.h,v 1.5 2004/03/16 00:07:51 lethal Exp $
- *	Copyright (C) 2000 YAEGASHI Takeshi
- *	Hitachi HD64461 companion chip support
- */
-
-/* Constants for PCMCIA mappings */
-#define HD64461_PCC_WINDOW	0x01000000
-
-#define HD64461_PCC0_BASE	0xb8000000	/* area 6 */
-#define HD64461_PCC0_ATTR	(HD64461_PCC0_BASE)
-#define HD64461_PCC0_COMM	(HD64461_PCC0_BASE+HD64461_PCC_WINDOW)
-#define HD64461_PCC0_IO		(HD64461_PCC0_BASE+2*HD64461_PCC_WINDOW)
-
-#define HD64461_PCC1_BASE	0xb4000000	/* area 5 */
-#define HD64461_PCC1_ATTR	(HD64461_PCC1_BASE)
-#define HD64461_PCC1_COMM	(HD64461_PCC1_BASE+HD64461_PCC_WINDOW)
-
-#define HD64461_STBCR	0x10000
-#define HD64461_STBCR_CKIO_STBY			0x2000
-#define HD64461_STBCR_SAFECKE_IST		0x1000
-#define HD64461_STBCR_SLCKE_IST			0x0800
-#define HD64461_STBCR_SAFECKE_OST		0x0400
-#define HD64461_STBCR_SLCKE_OST			0x0200
-#define HD64461_STBCR_SMIAST			0x0100
-#define HD64461_STBCR_SLCDST			0x0080
-#define HD64461_STBCR_SPC0ST			0x0040
-#define HD64461_STBCR_SPC1ST			0x0020
-#define HD64461_STBCR_SAFEST			0x0010
-#define HD64461_STBCR_STM0ST			0x0008
-#define HD64461_STBCR_STM1ST			0x0004
-#define HD64461_STBCR_SIRST				0x0002
-#define HD64461_STBCR_SURTST			0x0001
-
-#define HD64461_SYSCR	0x10002
-#define HD64461_SCPUCR	0x10004
-
-#define HD64461_LCDCBAR		0x11000
-#define HD64461_LCDCLOR		0x11002
-#define HD64461_LCDCCR		0x11004
-#define HD64461_LCDCCR_MOFF	0x80
-
-#define	HD64461_LDR1		0x11010
-#define	HD64461_LDR1_DON	0x01
-#define	HD64461_LDR1_DINV	0x80
-
-#define	HD64461_LDR2		0x11012
-#define	HD64461_LDHNCR		0x11014
-#define	HD64461_LDHNSR		0x11016
-#define HD64461_LDVNTR		0x11018
-#define HD64461_LDVNDR		0x1101a
-#define HD64461_LDVSPR		0x1101c
-#define HD64461_LDR3		0x1101e
-
-#define HD64461_CPTWAR		0x11030	
-#define HD64461_CPTWDR		0x11032
-#define HD64461_CPTRAR		0x11034	
-#define HD64461_CPTRDR		0x11036
-
-#define HD64461_GRDOR		0x11040
-#define HD64461_GRSCR		0x11042
-#define HD64461_GRCFGR		0x11044
-#define HD64461_GRCFGR_ACCSTATUS		0x10
-#define HD64461_GRCFGR_ACCRESET			0x08
-#define HD64461_GRCFGR_ACCSTART_BITBLT	0x06
-#define HD64461_GRCFGR_ACCSTART_LINE	0x04
-#define HD64461_GRCFGR_COLORDEPTH16		0x01
-
-#define HD64461_LNSARH		0x11046
-#define HD64461_LNSARL		0x11048
-#define HD64461_LNAXLR		0x1104a
-#define HD64461_LNDGR		0x1104c
-#define HD64461_LNAXR		0x1104e
-#define HD64461_LNERTR		0x11050
-#define HD64461_LNMDR		0x11052
-#define HD64461_BBTSSARH	0x11054
-#define HD64461_BBTSSARL	0x11056
-#define HD64461_BBTDSARH	0x11058
-#define HD64461_BBTDSARL	0x1105a
-#define HD64461_BBTDWR		0x1105c
-#define HD64461_BBTDHR		0x1105e
-#define HD64461_BBTPARH		0x11060
-#define HD64461_BBTPARL		0x11062
-#define HD64461_BBTMARH		0x11064
-#define HD64461_BBTMARL		0x11066
-#define HD64461_BBTROPR		0x11068
-#define HD64461_BBTMDR		0x1106a
-
-/* PC Card Controller Registers */
-#define HD64461_PCC0ISR         0x12000 /* socket 0 interface status */
-#define HD64461_PCC0GCR         0x12002 /* socket 0 general control */
-#define HD64461_PCC0CSCR        0x12004 /* socket 0 card status change */
-#define HD64461_PCC0CSCIER      0x12006 /* socket 0 card status change interrupt enable */
-#define HD64461_PCC0SCR         0x12008 /* socket 0 software control */
-#define HD64461_PCC1ISR         0x12010 /* socket 1 interface status */
-#define HD64461_PCC1GCR         0x12012 /* socket 1 general control */
-#define HD64461_PCC1CSCR        0x12014 /* socket 1 card status change */
-#define HD64461_PCC1CSCIER      0x12016 /* socket 1 card status change interrupt enable */
-#define HD64461_PCC1SCR         0x12018 /* socket 1 software control */
-
-/* PCC Interface Status Register */
-#define HD64461_PCCISR_READY		0x80	/* card ready */
-#define HD64461_PCCISR_MWP		0x40	/* card write-protected */
-#define HD64461_PCCISR_VS2		0x20	/* voltage select pin 2 */
-#define HD64461_PCCISR_VS1		0x10	/* voltage select pin 1 */
-#define HD64461_PCCISR_CD2		0x08	/* card detect 2 */
-#define HD64461_PCCISR_CD1		0x04	/* card detect 1 */
-#define HD64461_PCCISR_BVD2		0x02	/* battery 1 */
-#define HD64461_PCCISR_BVD1		0x01	/* battery 1 */
-
-#define HD64461_PCCISR_PCD_MASK		0x0c    /* card detect */
-#define HD64461_PCCISR_BVD_MASK	0x03    /* battery voltage */
-#define HD64461_PCCISR_BVD_BATGOOD	0x03    /* battery good */
-#define HD64461_PCCISR_BVD_BATWARN	0x01    /* battery low warning */
-#define HD64461_PCCISR_BVD_BATDEAD1	0x02    /* battery dead */
-#define HD64461_PCCISR_BVD_BATDEAD2	0x00    /* battery dead */
-
-/* PCC General Control Register */
-#define HD64461_PCCGCR_DRVE		0x80    /* output drive */
-#define HD64461_PCCGCR_PCCR		0x40    /* PC card reset */
-#define HD64461_PCCGCR_PCCT		0x20    /* PC card type, 1=IO&mem, 0=mem */
-#define HD64461_PCCGCR_VCC0		0x10    /* voltage control pin VCC0SEL0 */
-#define HD64461_PCCGCR_PMMOD		0x08    /* memory mode */
-#define HD64461_PCCGCR_PA25		0x04    /* pin A25 */
-#define HD64461_PCCGCR_PA24		0x02    /* pin A24 */
-#define HD64461_PCCGCR_REG		0x01    /* pin PCC0REG# */
-
-/* PCC Card Status Change Register */
-#define HD64461_PCCCSCR_SCDI		0x80    /* sw card detect intr */
-#define HD64461_PCCCSCR_SRV1		0x40    /* reserved */
-#define HD64461_PCCCSCR_IREQ		0x20    /* IREQ intr req */
-#define HD64461_PCCCSCR_SC		0x10    /* STSCHG (status change) pin */
-#define HD64461_PCCCSCR_CDC		0x08    /* CD (card detect) change */
-#define HD64461_PCCCSCR_RC		0x04    /* READY change */
-#define HD64461_PCCCSCR_BW		0x02    /* battery warning change */
-#define HD64461_PCCCSCR_BD		0x01    /* battery dead change */
-
-/* PCC Card Status Change Interrupt Enable Register */
-#define HD64461_PCCCSCIER_CRE		0x80    /* change reset enable */
-#define HD64461_PCCCSCIER_IREQE_MASK   	0x60   /* IREQ enable */
-#define HD64461_PCCCSCIER_IREQE_DISABLED	0x00   /* IREQ disabled */
-#define HD64461_PCCCSCIER_IREQE_LEVEL  	0x20   /* IREQ level-triggered */
-#define HD64461_PCCCSCIER_IREQE_FALLING	0x40   /* IREQ falling-edge-trig */
-#define HD64461_PCCCSCIER_IREQE_RISING 	0x60   /* IREQ rising-edge-trig */
-
-#define HD64461_PCCCSCIER_SCE		0x10    /* status change enable */
-#define HD64461_PCCCSCIER_CDE		0x08    /* card detect change enable */
-#define HD64461_PCCCSCIER_RE		0x04    /* ready change enable */
-#define HD64461_PCCCSCIER_BWE		0x02    /* battery warn change enable */
-#define HD64461_PCCCSCIER_BDE		0x01    /* battery dead change enable*/
-
-/* PCC Software Control Register */
-#define HD64461_PCCSCR_VCC1		0x02	/* voltage control pin 1 */
-#define HD64461_PCCSCR_SWP		0x01    /* write protect */
-
-
-#define HD64461_P0OCR           0x1202a
-#define HD64461_P1OCR           0x1202c
-#define HD64461_PGCR            0x1202e
-
-#define HD64461_GPACR		0x14000
-#define HD64461_GPBCR		0x14002
-#define HD64461_GPCCR		0x14004
-#define HD64461_GPDCR		0x14006
-#define HD64461_GPADR		0x14010
-#define HD64461_GPBDR		0x14012
-#define HD64461_GPCDR		0x14014
-#define HD64461_GPDDR		0x14016
-#define HD64461_GPAICR		0x14020
-#define HD64461_GPBICR		0x14022
-#define HD64461_GPCICR		0x14024
-#define HD64461_GPDICR		0x14026
-#define HD64461_GPAISR		0x14040
-#define HD64461_GPBISR		0x14042
-#define HD64461_GPCISR		0x14044
-#define HD64461_GPDISR		0x14046
-
-#define HD64461_NIRR		0x15000
-#define HD64461_NIMR		0x15002
-
-#ifndef CONFIG_HD64461_IOBASE
-#define CONFIG_HD64461_IOBASE	0xb0000000
-#endif
-#ifndef CONFIG_HD64461_IRQ
-#define CONFIG_HD64461_IRQ	36
-#endif
-
-#define HD64461_IRQBASE		OFFCHIP_IRQ_BASE
-#define HD64461_IRQ_NUM 	16
-
-#define HD64461_IRQ_UART    	(HD64461_IRQBASE+5)
-#define HD64461_IRQ_IRDA    	(HD64461_IRQBASE+6)
-#define HD64461_IRQ_TMU1   	(HD64461_IRQBASE+9)
-#define HD64461_IRQ_TMU0  	(HD64461_IRQBASE+10)
-#define HD64461_IRQ_GPIO    	(HD64461_IRQBASE+11)
-#define HD64461_IRQ_AFE     	(HD64461_IRQBASE+12)
-#define HD64461_IRQ_PCC1 	(HD64461_IRQBASE+13)
-#define HD64461_IRQ_PCC0 	(HD64461_IRQBASE+14)
-
-#endif
diff --git a/include/asm-sh/hd64461/io.h b/include/asm-sh/hd64461/io.h
deleted file mode 100644
index 67f2489088d9..000000000000
--- a/include/asm-sh/hd64461/io.h
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * include/asm-sh/io_hd64461.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an HD64461
- */
-
-#ifndef _ASM_SH_IO_HD64461_H
-#define _ASM_SH_IO_HD64461_H
-
-extern unsigned char hd64461_inb(unsigned long port);
-extern unsigned short hd64461_inw(unsigned long port);
-extern unsigned int hd64461_inl(unsigned long port);
-
-extern void hd64461_outb(unsigned char value, unsigned long port);
-extern void hd64461_outw(unsigned short value, unsigned long port);
-extern void hd64461_outl(unsigned int value, unsigned long port);
-
-extern unsigned char hd64461_inb_p(unsigned long port);
-extern void hd64461_outb_p(unsigned char value, unsigned long port);
-
-extern void hd64461_insb(unsigned long port, void *addr, unsigned long count);
-extern void hd64461_insw(unsigned long port, void *addr, unsigned long count);
-extern void hd64461_insl(unsigned long port, void *addr, unsigned long count);
-
-extern void hd64461_outsb(unsigned long port, const void *buffer, unsigned long count);
-extern void hd64461_outsw(unsigned long port, const void *buffer, unsigned long count);
-extern void hd64461_outsl(unsigned long port, const void *buffer, unsigned long count);
-
-extern unsigned short hd64461_readw(unsigned long addr);
-extern void hd64461_writew(unsigned short b, unsigned long addr);
-
-
-extern int hd64461_irq_demux(int irq);
-extern void hd64461_register_irq_demux(int irq,
-		int (*demux)(int irq, void *dev), void *dev);
-extern void hd64461_unregister_irq_demux(int irq);
-
-#endif /* _ASM_SH_IO_HD64461_H */
diff --git a/include/asm-sh/hp6xx/io.h b/include/asm-sh/hp6xx/io.h
index 731798003550..2044476ab199 100644
--- a/include/asm-sh/hp6xx/io.h
+++ b/include/asm-sh/hp6xx/io.h
@@ -4,7 +4,7 @@
 /*
  * Nothing special here.. just use the generic cchip io routines.
  */
-#include <asm/hd64461/io.h>
+#include <asm/hd64461.h>
 
 #endif /* __ASM_SH_HP6XX_IO_H */
 
-- 
cgit v1.2.3


From e8fb67f8e05bb1f4c07c3585967cfc6d44822ab0 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:56:28 +0900
Subject: sh: HS7751RVoIP board updates.

Various cleanups for HS7751RVoIP. Mostly just getting
rid of the old mach.c and splitting codec configuration
in to its own Kconfig.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                             |  2 +-
 arch/sh/boards/renesas/hs7751rvoip/Kconfig  | 12 ++++++
 arch/sh/boards/renesas/hs7751rvoip/Makefile |  2 +-
 arch/sh/boards/renesas/hs7751rvoip/mach.c   | 54 -------------------------
 arch/sh/boards/renesas/hs7751rvoip/setup.c  | 63 +++++++++++++++++++++++------
 include/asm-sh/hs7751rvoip/hs7751rvoip.h    |  6 +++
 6 files changed, 71 insertions(+), 68 deletions(-)
 create mode 100644 arch/sh/boards/renesas/hs7751rvoip/Kconfig
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/mach.c

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 1a0db1d4c952..cbf0d527e991 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -400,7 +400,7 @@ config SH_TMU
 
 endmenu
 
-#source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
+source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
 
 #source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Kconfig b/arch/sh/boards/renesas/hs7751rvoip/Kconfig
new file mode 100644
index 000000000000..1743be477be5
--- /dev/null
+++ b/arch/sh/boards/renesas/hs7751rvoip/Kconfig
@@ -0,0 +1,12 @@
+if SH_HS7751RVOIP
+
+menu "HS7751RVoIP options"
+
+config HS7751RVOIP_CODEC
+	bool "Support VoIP Codec section"
+	help
+	  Selecting this option will support CODEC section.
+
+endmenu
+
+endif
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Makefile b/arch/sh/boards/renesas/hs7751rvoip/Makefile
index 3ef8cbaa0b37..bede2d54fd6c 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/Makefile
+++ b/arch/sh/boards/renesas/hs7751rvoip/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the HS7751RVoIP specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o led.o
+obj-y	 := setup.o io.o irq.o led.o
 
 obj-$(CONFIG_PCI) += pci.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/mach.c b/arch/sh/boards/renesas/hs7751rvoip/mach.c
deleted file mode 100644
index caf967f77c61..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/mach.c
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_hs7751rvoip.c
- *
- * Minor tweak of mach_se.c file to reference hs7751rvoip-specific items.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Renesas Technology sales HS7751RVoIP
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/irq.h>
-#include <asm/hs7751rvoip/io.h>
-
-extern void init_hs7751rvoip_IRQ(void);
-extern void *hs7751rvoip_ioremap(unsigned long, unsigned long);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_hs7751rvoip __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= hs7751rvoip_inb,
-	.mv_inw			= hs7751rvoip_inw,
-	.mv_inl			= hs7751rvoip_inl,
-	.mv_outb		= hs7751rvoip_outb,
-	.mv_outw		= hs7751rvoip_outw,
-	.mv_outl		= hs7751rvoip_outl,
-
-	.mv_inb_p		= hs7751rvoip_inb_p,
-	.mv_inw_p		= hs7751rvoip_inw,
-	.mv_inl_p		= hs7751rvoip_inl,
-	.mv_outb_p		= hs7751rvoip_outb_p,
-	.mv_outw_p		= hs7751rvoip_outw,
-	.mv_outl_p		= hs7751rvoip_outl,
-
-	.mv_insb		= hs7751rvoip_insb,
-	.mv_insw		= hs7751rvoip_insw,
-	.mv_insl		= hs7751rvoip_insl,
-	.mv_outsb		= hs7751rvoip_outsb,
-	.mv_outsw		= hs7751rvoip_outsw,
-	.mv_outsl		= hs7751rvoip_outsl,
-
-	.mv_ioremap		= hs7751rvoip_ioremap,
-	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
-	.mv_init_irq		= init_hs7751rvoip_IRQ,
-};
-ALIAS_MV(hs7751rvoip)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index 7ce577822dc0..813fc4d5862a 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -17,14 +17,55 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/pm.h>
-#include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
-
-/* defined in mm/ioremap.c */
-extern void * p3_ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags);
+#include <asm/hs7751rvoip/io.h>
+#include <asm/io.h>
+#include <asm/machvec.h>
+#include <asm/rtc.h>
+#include <asm/irq.h>
 
 unsigned int debug_counter;
 
+static void __init hs7751rvoip_init_irq(void)
+{
+#if defined(CONFIG_HS7751RVOIP_CODEC)
+	make_ipr_irq(DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+	make_ipr_irq(DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+#endif
+
+	init_hs7751rvoip_IRQ();
+}
+
+struct sh_machine_vector mv_hs7751rvoip __initmv = {
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= hs7751rvoip_inb,
+	.mv_inw			= hs7751rvoip_inw,
+	.mv_inl			= hs7751rvoip_inl,
+	.mv_outb		= hs7751rvoip_outb,
+	.mv_outw		= hs7751rvoip_outw,
+	.mv_outl		= hs7751rvoip_outl,
+
+	.mv_inb_p		= hs7751rvoip_inb_p,
+	.mv_inw_p		= hs7751rvoip_inw,
+	.mv_inl_p		= hs7751rvoip_inl,
+	.mv_outb_p		= hs7751rvoip_outb_p,
+	.mv_outw_p		= hs7751rvoip_outw,
+	.mv_outl_p		= hs7751rvoip_outl,
+
+	.mv_insb		= hs7751rvoip_insb,
+	.mv_insw		= hs7751rvoip_insw,
+	.mv_insl		= hs7751rvoip_insl,
+	.mv_outsb		= hs7751rvoip_outsb,
+	.mv_outsw		= hs7751rvoip_outsw,
+	.mv_outsl		= hs7751rvoip_outsl,
+
+	.mv_ioremap		= hs7751rvoip_ioremap,
+	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
+	.mv_init_irq		= hs7751rvoip_init_irq,
+};
+ALIAS_MV(hs7751rvoip)
+
 const char *get_system_type(void)
 {
 	return "HS7751RVoIP";
@@ -51,16 +92,15 @@ void *area6_io8_base;
 void *area5_io16_base;
 void *area6_io16_base;
 
-int __init cf_init(void)
+static int __init hs7751rvoip_cf_init(void)
 {
 	pgprot_t prot;
-	unsigned long paddrbase, psize;
+	unsigned long paddrbase;
 
 	/* open I/O area window */
 	paddrbase = virt_to_phys((void *)(PA_AREA5_IO+0x00000800));
-	psize = PAGE_SIZE;
 	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_COM16);
-	area5_io16_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	area5_io16_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
 	if (!area5_io16_base) {
 		printk("allocate_cf_area : can't open CF I/O window!\n");
 		return -ENOMEM;
@@ -69,19 +109,18 @@ int __init cf_init(void)
 	/* XXX : do we need attribute and common-memory area also? */
 
 	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
-	psize = PAGE_SIZE;
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_COM8);
 #else
 	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO8);
 #endif
-	area6_io8_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	area6_io8_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
 	if (!area6_io8_base) {
 		printk("allocate_cf_area : can't open CODEC I/O 8bit window!\n");
 		return -ENOMEM;
 	}
 	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
-	area6_io16_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	area6_io16_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
 	if (!area6_io16_base) {
 		printk("allocate_cf_area : can't open CODEC I/O 16bit window!\n");
 		return -ENOMEM;
@@ -90,4 +129,4 @@ int __init cf_init(void)
 	return 0;
 }
 
-__initcall (cf_init);
+__initcall(hs7751rvoip_cf_init);
diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
index 5f995f937a44..857cc2ebbc69 100644
--- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h
+++ b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
@@ -44,4 +44,10 @@
 #define	IRQ_RINGING	4		/* Ringing IRQ */
 #define	IRQ_CODEC	5		/* CODEC IRQ */
 
+/* arch/sh/boards/renesas/hs7751rvoip/irq.c */
+void init_hs7751rvoip_IRQ(void);
+
+/* arch/sh/boards/renesas/hs7751rvoip/io.c */
+void *hs7751rvoip_ioremap(unsigned long, unsigned long);
+
 #endif  /* __ASM_SH_RENESAS_HS7751RVOIP */
-- 
cgit v1.2.3


From fdfc74f9fcebdda14609159d5010b758a9409acf Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:05:52 +0900
Subject: sh: Support for SH-4A memory barriers.

SH-4A supports 'synco' as a barrier, sprinkle it around
the cache ops as necessary..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c  | 5 +++++
 include/asm-sh/system.h | 7 +++++++
 2 files changed, 12 insertions(+)

(limited to 'include')

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 94c05d09c3f7..846b63d6f5e8 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -184,6 +184,7 @@ void flush_cache_sigtramp(unsigned long addr)
 	     i++, index += cpu_data->icache.way_incr)
 		ctrl_outl(0, index);	/* Clear out Valid-bit */
 	back_to_P1();
+	wmb();
 	local_irq_restore(flags);
 }
 
@@ -223,6 +224,8 @@ void flush_dcache_page(struct page *page)
 		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x2000, phys);
 		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x3000, phys);
 	}
+
+	wmb();
 }
 
 static inline void flush_icache_all(void)
@@ -247,6 +250,7 @@ void flush_dcache_all(void)
 		__flush_dcache_all();
 	else
 		__flush_dcache_all_ex();
+	wmb();
 }
 
 void flush_cache_all(void)
@@ -377,5 +381,6 @@ void flush_icache_user_range(struct vm_area_struct *vma,
 			     struct page *page, unsigned long addr, int len)
 {
 	flush_cache_page(vma, addr, page_to_pfn(page));
+	mb();
 }
 
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index e89728d405d8..eb4902ed920a 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -84,10 +84,17 @@ static __inline__ unsigned long tas(volatile int *m)
 
 extern void __xchg_called_with_bad_pointer(void);
 
+#ifdef CONFIG_CPU_SH4A
+#define mb()	__asm__ __volatile__ ("synco": : :"memory")
+#define rmb()	mb()
+#define wmb()	__asm__ __volatile__ ("synco": : :"memory")
+#define read_barrier_depends()	do { } while(0)
+#else
 #define mb()	__asm__ __volatile__ ("": : :"memory")
 #define rmb()	mb()
 #define wmb()	__asm__ __volatile__ ("": : :"memory")
 #define read_barrier_depends()	do { } while(0)
+#endif
 
 #ifdef CONFIG_SMP
 #define smp_mb()	mb()
-- 
cgit v1.2.3


From b638d0b921dc95229af0dfd09cd24850336a2f75 Mon Sep 17 00:00:00 2001
From: Richard Curnow <richard.curnow@st.com>
Date: Wed, 27 Sep 2006 14:09:26 +0900
Subject: sh: Optimized cache handling for SH-4/SH-4A caches.

This reworks some of the SH-4 cache handling code to more easily
accomodate newer-style caches (particularly for the > direct-mapped
case), as well as optimizing some of the old code.

Signed-off-by: Richard Curnow <richard.curnow@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/init.c      |  16 +-
 arch/sh/kernel/cpu/sh4/probe.c |  11 +
 arch/sh/mm/cache-sh4.c         | 517 ++++++++++++++++++++++++++++++++++-------
 arch/sh/mm/clear_page.S        |  99 +-------
 include/asm-sh/cache.h         |  22 +-
 5 files changed, 475 insertions(+), 190 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index 868e68b28880..731dd61419dd 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -4,6 +4,7 @@
  * CPU init code
  *
  * Copyright (C) 2002, 2003  Paul Mundt
+ * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -51,7 +52,15 @@ static void __init cache_init(void)
 	ccr = ctrl_inl(CCR);
 
 	/*
-	 * If the cache is already enabled .. flush it.
+	 * At this point we don't know whether the cache is enabled or not - a
+	 * bootloader may have enabled it.  There are at least 2 things that
+	 * could be dirty in the cache at this point:
+	 * 1. kernel command line set up by boot loader
+	 * 2. spilled registers from the prolog of this function
+	 * => before re-initialising the cache, we must do a purge of the whole
+	 * cache out to memory for safety.  As long as nothing is spilled
+	 * during the loop to lines that have already been done, this is safe.
+	 * - RPC
 	 */
 	if (ccr & CCR_CACHE_ENABLE) {
 		unsigned long ways, waysize, addrstart;
@@ -98,6 +107,8 @@ static void __init cache_init(void)
 	/* Force EMODE if possible */
 	if (cpu_data->dcache.ways > 1)
 		flags |= CCR_CACHE_EMODE;
+	else
+		flags &= ~CCR_CACHE_EMODE;
 #endif
 
 #ifdef CONFIG_SH_WRITETHROUGH
@@ -112,6 +123,9 @@ static void __init cache_init(void)
 	/* Turn on OCRAM -- halve the OC */
 	flags |= CCR_CACHE_ORA;
 	cpu_data->dcache.sets >>= 1;
+
+	cpu_data->dcache.way_size = cpu_data->dcache.sets *
+				    cpu_data->dcache.linesz;
 #endif
 
 	ctrl_outl(flags, CCR);
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 42427b79697b..1208da8fe5db 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -113,6 +113,11 @@ int __init detect_cpu_and_cache_system(void)
 		break;
 	}
 
+#ifdef CONFIG_SH_DIRECT_MAPPED
+	cpu_data->icache.ways = 1;
+	cpu_data->dcache.ways = 1;
+#endif
+
 	/*
 	 * On anything that's not a direct-mapped cache, look to the CVR
 	 * for I/D-cache specifics.
@@ -125,6 +130,9 @@ int __init detect_cpu_and_cache_system(void)
 			(cpu_data->icache.way_incr - (1 << 5));
 	}
 
+	cpu_data->icache.way_size = cpu_data->icache.sets *
+				    cpu_data->icache.linesz;
+
 	if (cpu_data->dcache.ways > 1) {
 		size = sizes[(cvr >> 16) & 0xf];
 		cpu_data->dcache.way_incr	= (size >> 1);
@@ -133,6 +141,9 @@ int __init detect_cpu_and_cache_system(void)
 			(cpu_data->dcache.way_incr - (1 << 5));
 	}
 
+	cpu_data->dcache.way_size = cpu_data->dcache.sets *
+				    cpu_data->dcache.linesz;
+
 	return 0;
 }
 
diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 846b63d6f5e8..c036c2b4ac2b 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -25,28 +25,95 @@
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 
-extern void __flush_cache_4096(unsigned long addr, unsigned long phys,
+static void __flush_dcache_segment_1way(unsigned long start,
+					unsigned long extent);
+static void __flush_dcache_segment_2way(unsigned long start,
+					unsigned long extent);
+static void __flush_dcache_segment_4way(unsigned long start,
+					unsigned long extent);
+
+static void __flush_cache_4096(unsigned long addr, unsigned long phys,
 			       unsigned long exec_offset);
-extern void __flush_cache_4096_all(unsigned long start);
-static void __flush_cache_4096_all_ex(unsigned long start);
-extern void __flush_dcache_all(void);
-static void __flush_dcache_all_ex(void);
+
+/*
+ * This is initialised here to ensure that it is not placed in the BSS.  If
+ * that were to happen, note that cache_init gets called before the BSS is
+ * cleared, so this would get nulled out which would be hopeless.
+ */
+static void (*__flush_dcache_segment_fn)(unsigned long, unsigned long) =
+	(void (*)(unsigned long, unsigned long))0xdeadbeef;
+
+static void compute_alias(struct cache_info *c)
+{
+	c->alias_mask = ((c->sets - 1) << c->entry_shift) & ~(PAGE_SIZE - 1);
+	c->n_aliases = (c->alias_mask >> PAGE_SHIFT) + 1;
+}
+
+static void __init emit_cache_params(void)
+{
+	printk("PVR=%08x CVR=%08x PRR=%08x\n",
+		ctrl_inl(CCN_PVR),
+		ctrl_inl(CCN_CVR),
+		ctrl_inl(CCN_PRR));
+	printk("I-cache : n_ways=%d n_sets=%d way_incr=%d\n",
+		cpu_data->icache.ways,
+		cpu_data->icache.sets,
+		cpu_data->icache.way_incr);
+	printk("I-cache : entry_mask=0x%08x alias_mask=0x%08x n_aliases=%d\n",
+		cpu_data->icache.entry_mask,
+		cpu_data->icache.alias_mask,
+		cpu_data->icache.n_aliases);
+	printk("D-cache : n_ways=%d n_sets=%d way_incr=%d\n",
+		cpu_data->dcache.ways,
+		cpu_data->dcache.sets,
+		cpu_data->dcache.way_incr);
+	printk("D-cache : entry_mask=0x%08x alias_mask=0x%08x n_aliases=%d\n",
+		cpu_data->dcache.entry_mask,
+		cpu_data->dcache.alias_mask,
+		cpu_data->dcache.n_aliases);
+
+	if (!__flush_dcache_segment_fn)
+		panic("unknown number of cache ways\n");
+}
 
 /*
  * SH-4 has virtually indexed and physically tagged cache.
  */
 
-struct semaphore p3map_sem[4];
+/* Worst case assumed to be 64k cache, direct-mapped i.e. 4 synonym bits. */
+#define MAX_P3_SEMAPHORES 16
+
+struct semaphore p3map_sem[MAX_P3_SEMAPHORES];
 
 void __init p3_cache_init(void)
 {
-	if (remap_area_pages(P3SEG, 0, PAGE_SIZE*4, _PAGE_CACHABLE))
+	int i;
+
+	compute_alias(&cpu_data->icache);
+	compute_alias(&cpu_data->dcache);
+
+	switch (cpu_data->dcache.ways) {
+	case 1:
+		__flush_dcache_segment_fn = __flush_dcache_segment_1way;
+		break;
+	case 2:
+		__flush_dcache_segment_fn = __flush_dcache_segment_2way;
+		break;
+	case 4:
+		__flush_dcache_segment_fn = __flush_dcache_segment_4way;
+		break;
+	default:
+		__flush_dcache_segment_fn = NULL;
+		break;
+	}
+
+	emit_cache_params();
+
+	if (remap_area_pages(P3SEG, 0, PAGE_SIZE * 4, _PAGE_CACHABLE))
 		panic("%s failed.", __FUNCTION__);
 
-	sema_init (&p3map_sem[0], 1);
-	sema_init (&p3map_sem[1], 1);
-	sema_init (&p3map_sem[2], 1);
-	sema_init (&p3map_sem[3], 1);
+	for (i = 0; i < cpu_data->dcache.n_aliases; i++)
+		sema_init(&p3map_sem[i], 1);
 }
 
 /*
@@ -91,7 +158,6 @@ void __flush_purge_region(void *start, int size)
 	}
 }
 
-
 /*
  * No write back please
  */
@@ -110,46 +176,6 @@ void __flush_invalidate_region(void *start, int size)
 	}
 }
 
-static void __flush_dcache_all_ex(void)
-{
-	unsigned long addr, end_addr, entry_offset;
-
-	end_addr = CACHE_OC_ADDRESS_ARRAY +
-		(cpu_data->dcache.sets << cpu_data->dcache.entry_shift) *
-		 cpu_data->dcache.ways;
-
-	entry_offset = 1 << cpu_data->dcache.entry_shift;
-	for (addr = CACHE_OC_ADDRESS_ARRAY;
-	     addr < end_addr;
-	     addr += entry_offset) {
-		ctrl_outl(0, addr);
-	}
-}
-
-static void __flush_cache_4096_all_ex(unsigned long start)
-{
-	unsigned long addr, entry_offset;
-	int i;
-
-	entry_offset = 1 << cpu_data->dcache.entry_shift;
-	for (i = 0; i < cpu_data->dcache.ways;
-	     i++, start += cpu_data->dcache.way_incr) {
-		for (addr = CACHE_OC_ADDRESS_ARRAY + start;
-		     addr < CACHE_OC_ADDRESS_ARRAY + 4096 + start;
-		     addr += entry_offset) {
-			ctrl_outl(0, addr);
-		}
-	}
-}
-
-void flush_cache_4096_all(unsigned long start)
-{
-	if (cpu_data->dcache.ways == 1)
-		__flush_cache_4096_all(start);
-	else
-		__flush_cache_4096_all_ex(start);
-}
-
 /*
  * Write back the range of D-cache, and purge the I-cache.
  *
@@ -180,9 +206,11 @@ void flush_cache_sigtramp(unsigned long addr)
 
 	local_irq_save(flags);
 	jump_to_P2();
+
 	for (i = 0; i < cpu_data->icache.ways;
 	     i++, index += cpu_data->icache.way_incr)
 		ctrl_outl(0, index);	/* Clear out Valid-bit */
+
 	back_to_P1();
 	wmb();
 	local_irq_restore(flags);
@@ -194,8 +222,8 @@ static inline void flush_cache_4096(unsigned long start,
 	unsigned long flags;
 
 	/*
-	 * SH7751, SH7751R, and ST40 have no restriction to handle cache.
-	 * (While SH7750 must do that at P2 area.)
+	 * All types of SH-4 require PC to be in P2 to operate on the I-cache.
+	 * Some types of SH-4 require PC to be in P2 to operate on the D-cache.
 	 */
 	if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG)
 	   || start < CACHE_OC_ADDRESS_ARRAY) {
@@ -217,12 +245,13 @@ void flush_dcache_page(struct page *page)
 {
 	if (test_bit(PG_mapped, &page->flags)) {
 		unsigned long phys = PHYSADDR(page_address(page));
+		unsigned long addr = CACHE_OC_ADDRESS_ARRAY;
+		int i, n;
 
 		/* Loop all the D-cache */
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY,          phys);
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x1000, phys);
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x2000, phys);
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x3000, phys);
+		n = cpu_data->dcache.n_aliases;
+		for (i = 0; i < n; i++, addr += PAGE_SIZE)
+			flush_cache_4096(addr, phys);
 	}
 
 	wmb();
@@ -246,10 +275,7 @@ static inline void flush_icache_all(void)
 
 void flush_dcache_all(void)
 {
-	if (cpu_data->dcache.ways == 1)
-		__flush_dcache_all();
-	else
-		__flush_dcache_all_ex();
+	(*__flush_dcache_segment_fn)(0UL, cpu_data->dcache.way_size);
 	wmb();
 }
 
@@ -261,6 +287,16 @@ void flush_cache_all(void)
 
 void flush_cache_mm(struct mm_struct *mm)
 {
+	/*
+	 * Note : (RPC) since the caches are physically tagged, the only point
+	 * of flush_cache_mm for SH-4 is to get rid of aliases from the
+	 * D-cache.  The assumption elsewhere, e.g. flush_cache_range, is that
+	 * lines can stay resident so long as the virtual address they were
+	 * accessed with (hence cache set) is in accord with the physical
+	 * address (i.e. tag).  It's no different here.  So I reckon we don't
+	 * need to flush the I-cache, since aliases don't matter for that.  We
+	 * should try that.
+	 */
 	flush_cache_all();
 }
 
@@ -273,24 +309,36 @@ void flush_cache_mm(struct mm_struct *mm)
 void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigned long pfn)
 {
 	unsigned long phys = pfn << PAGE_SHIFT;
+	unsigned int alias_mask;
+
+	alias_mask = cpu_data->dcache.alias_mask;
 
 	/* We only need to flush D-cache when we have alias */
-	if ((address^phys) & CACHE_ALIAS) {
+	if ((address^phys) & alias_mask) {
 		/* Loop 4K of the D-cache */
 		flush_cache_4096(
-			CACHE_OC_ADDRESS_ARRAY | (address & CACHE_ALIAS),
+			CACHE_OC_ADDRESS_ARRAY | (address & alias_mask),
 			phys);
 		/* Loop another 4K of the D-cache */
 		flush_cache_4096(
-			CACHE_OC_ADDRESS_ARRAY | (phys & CACHE_ALIAS),
+			CACHE_OC_ADDRESS_ARRAY | (phys & alias_mask),
 			phys);
 	}
 
-	if (vma->vm_flags & VM_EXEC)
-		/* Loop 4K (half) of the I-cache */
+	alias_mask = cpu_data->icache.alias_mask;
+	if (vma->vm_flags & VM_EXEC) {
+		/*
+		 * Evict entries from the portion of the cache from which code
+		 * may have been executed at this address (virtual).  There's
+		 * no need to evict from the portion corresponding to the
+		 * physical address as for the D-cache, because we know the
+		 * kernel has never executed the code through its identity
+		 * translation.
+		 */
 		flush_cache_4096(
-			CACHE_IC_ADDRESS_ARRAY | (address & 0x1000),
+			CACHE_IC_ADDRESS_ARRAY | (address & alias_mask),
 			phys);
+	}
 }
 
 /*
@@ -305,14 +353,28 @@ void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigne
 void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 		       unsigned long end)
 {
-	unsigned long p = start & PAGE_MASK;
+	unsigned long d = 0, p = start & PAGE_MASK;
+	unsigned long alias_mask = cpu_data->dcache.alias_mask;
+	unsigned long n_aliases = cpu_data->dcache.n_aliases;
+	unsigned long select_bit;
+	unsigned long all_aliases_mask;
+	unsigned long addr_offset;
+	unsigned long phys;
 	pgd_t *dir;
 	pmd_t *pmd;
 	pud_t *pud;
 	pte_t *pte;
 	pte_t entry;
-	unsigned long phys;
-	unsigned long d = 0;
+	int i;
+
+	/*
+	 * If cache is only 4k-per-way, there are never any 'aliases'.  Since
+	 * the cache is physically tagged, the data can just be left in there.
+	 */
+	if (n_aliases == 0)
+		return;
+
+	all_aliases_mask = (1 << n_aliases) - 1;
 
 	/*
 	 * Don't bother with the lookup and alias check if we have a
@@ -335,39 +397,52 @@ void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 
 	do {
 		if (pmd_none(*pmd) || pmd_bad(*pmd)) {
-			p &= ~((1 << PMD_SHIFT) -1);
+			p &= ~((1 << PMD_SHIFT) - 1);
 			p += (1 << PMD_SHIFT);
 			pmd++;
+
 			continue;
 		}
+
 		pte = pte_offset_kernel(pmd, p);
+
 		do {
 			entry = *pte;
+
 			if ((pte_val(entry) & _PAGE_PRESENT)) {
-				phys = pte_val(entry)&PTE_PHYS_MASK;
-				if ((p^phys) & CACHE_ALIAS) {
-					d |= 1 << ((p & CACHE_ALIAS)>>12);
-					d |= 1 << ((phys & CACHE_ALIAS)>>12);
-					if (d == 0x0f)
+				phys = pte_val(entry) & PTE_PHYS_MASK;
+
+				if ((p ^ phys) & alias_mask) {
+					d |= 1 << ((p & alias_mask) >> PAGE_SHIFT);
+					d |= 1 << ((phys & alias_mask) >> PAGE_SHIFT);
+
+					if (d == all_aliases_mask)
 						goto loop_exit;
 				}
 			}
+
 			pte++;
 			p += PAGE_SIZE;
 		} while (p < end && ((unsigned long)pte & ~PAGE_MASK));
 		pmd++;
 	} while (p < end);
- loop_exit:
-	if (d & 1)
-		flush_cache_4096_all(0);
-	if (d & 2)
-		flush_cache_4096_all(0x1000);
-	if (d & 4)
-		flush_cache_4096_all(0x2000);
-	if (d & 8)
-		flush_cache_4096_all(0x3000);
-	if (vma->vm_flags & VM_EXEC)
+
+loop_exit:
+	for (i = 0, select_bit = 0x1, addr_offset = 0x0; i < n_aliases;
+	     i++, select_bit <<= 1, addr_offset += PAGE_SIZE)
+		if (d & select_bit) {
+			(*__flush_dcache_segment_fn)(addr_offset, PAGE_SIZE);
+			wmb();
+		}
+
+	if (vma->vm_flags & VM_EXEC) {
+		/*
+		 * TODO: Is this required???  Need to look at how I-cache
+		 * coherency is assured when new programs are loaded to see if
+		 * this matters.
+		 */
 		flush_icache_all();
+	}
 }
 
 /*
@@ -384,3 +459,271 @@ void flush_icache_user_range(struct vm_area_struct *vma,
 	mb();
 }
 
+/**
+ * __flush_cache_4096
+ *
+ * @addr:  address in memory mapped cache array
+ * @phys:  P1 address to flush (has to match tags if addr has 'A' bit
+ *         set i.e. associative write)
+ * @exec_offset: set to 0x20000000 if flush has to be executed from P2
+ *               region else 0x0
+ *
+ * The offset into the cache array implied by 'addr' selects the
+ * 'colour' of the virtual address range that will be flushed.  The
+ * operation (purge/write-back) is selected by the lower 2 bits of
+ * 'phys'.
+ */
+static void __flush_cache_4096(unsigned long addr, unsigned long phys,
+			       unsigned long exec_offset)
+{
+	int way_count;
+	unsigned long base_addr = addr;
+	struct cache_info *dcache;
+	unsigned long way_incr;
+	unsigned long a, ea, p;
+	unsigned long temp_pc;
+
+	dcache = &cpu_data->dcache;
+	/* Write this way for better assembly. */
+	way_count = dcache->ways;
+	way_incr = dcache->way_incr;
+
+	/*
+	 * Apply exec_offset (i.e. branch to P2 if required.).
+	 *
+	 * FIXME:
+	 *
+	 *	If I write "=r" for the (temp_pc), it puts this in r6 hence
+	 *	trashing exec_offset before it's been added on - why?  Hence
+	 *	"=&r" as a 'workaround'
+	 */
+	asm volatile("mov.l 1f, %0\n\t"
+		     "add   %1, %0\n\t"
+		     "jmp   @%0\n\t"
+		     "nop\n\t"
+		     ".balign 4\n\t"
+		     "1:  .long 2f\n\t"
+		     "2:\n" : "=&r" (temp_pc) : "r" (exec_offset));
+
+	/*
+	 * We know there will be >=1 iteration, so write as do-while to avoid
+	 * pointless nead-of-loop check for 0 iterations.
+	 */
+	do {
+		ea = base_addr + PAGE_SIZE;
+		a = base_addr;
+		p = phys;
+
+		do {
+			*(volatile unsigned long *)a = p;
+			/*
+			 * Next line: intentionally not p+32, saves an add, p
+			 * will do since only the cache tag bits need to
+			 * match.
+			 */
+			*(volatile unsigned long *)(a+32) = p;
+			a += 64;
+			p += 64;
+		} while (a < ea);
+
+		base_addr += way_incr;
+	} while (--way_count != 0);
+}
+
+/*
+ * Break the 1, 2 and 4 way variants of this out into separate functions to
+ * avoid nearly all the overhead of having the conditional stuff in the function
+ * bodies (+ the 1 and 2 way cases avoid saving any registers too).
+ */
+static void __flush_dcache_segment_1way(unsigned long start,
+					unsigned long extent_per_way)
+{
+	unsigned long orig_sr, sr_with_bl;
+	unsigned long base_addr;
+	unsigned long way_incr, linesz, way_size;
+	struct cache_info *dcache;
+	register unsigned long a0, a0e;
+
+	asm volatile("stc sr, %0" : "=r" (orig_sr));
+	sr_with_bl = orig_sr | (1<<28);
+	base_addr = ((unsigned long)&empty_zero_page[0]);
+
+	/*
+	 * The previous code aligned base_addr to 16k, i.e. the way_size of all
+	 * existing SH-4 D-caches.  Whilst I don't see a need to have this
+	 * aligned to any better than the cache line size (which it will be
+	 * anyway by construction), let's align it to at least the way_size of
+	 * any existing or conceivable SH-4 D-cache.  -- RPC
+	 */
+	base_addr = ((base_addr >> 16) << 16);
+	base_addr |= start;
+
+	dcache = &cpu_data->dcache;
+	linesz = dcache->linesz;
+	way_incr = dcache->way_incr;
+	way_size = dcache->way_size;
+
+	a0 = base_addr;
+	a0e = base_addr + extent_per_way;
+	do {
+		asm volatile("ldc %0, sr" : : "r" (sr_with_bl));
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		a0 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		a0 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		a0 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		asm volatile("ldc %0, sr" : : "r" (orig_sr));
+		a0 += linesz;
+	} while (a0 < a0e);
+}
+
+static void __flush_dcache_segment_2way(unsigned long start,
+					unsigned long extent_per_way)
+{
+	unsigned long orig_sr, sr_with_bl;
+	unsigned long base_addr;
+	unsigned long way_incr, linesz, way_size;
+	struct cache_info *dcache;
+	register unsigned long a0, a1, a0e;
+
+	asm volatile("stc sr, %0" : "=r" (orig_sr));
+	sr_with_bl = orig_sr | (1<<28);
+	base_addr = ((unsigned long)&empty_zero_page[0]);
+
+	/* See comment under 1-way above */
+	base_addr = ((base_addr >> 16) << 16);
+	base_addr |= start;
+
+	dcache = &cpu_data->dcache;
+	linesz = dcache->linesz;
+	way_incr = dcache->way_incr;
+	way_size = dcache->way_size;
+
+	a0 = base_addr;
+	a1 = a0 + way_incr;
+	a0e = base_addr + extent_per_way;
+	do {
+		asm volatile("ldc %0, sr" : : "r" (sr_with_bl));
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		a0 += linesz;
+		a1 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		a0 += linesz;
+		a1 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		a0 += linesz;
+		a1 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		asm volatile("ldc %0, sr" : : "r" (orig_sr));
+		a0 += linesz;
+		a1 += linesz;
+	} while (a0 < a0e);
+}
+
+static void __flush_dcache_segment_4way(unsigned long start,
+					unsigned long extent_per_way)
+{
+	unsigned long orig_sr, sr_with_bl;
+	unsigned long base_addr;
+	unsigned long way_incr, linesz, way_size;
+	struct cache_info *dcache;
+	register unsigned long a0, a1, a2, a3, a0e;
+
+	asm volatile("stc sr, %0" : "=r" (orig_sr));
+	sr_with_bl = orig_sr | (1<<28);
+	base_addr = ((unsigned long)&empty_zero_page[0]);
+
+	/* See comment under 1-way above */
+	base_addr = ((base_addr >> 16) << 16);
+	base_addr |= start;
+
+	dcache = &cpu_data->dcache;
+	linesz = dcache->linesz;
+	way_incr = dcache->way_incr;
+	way_size = dcache->way_size;
+
+	a0 = base_addr;
+	a1 = a0 + way_incr;
+	a2 = a1 + way_incr;
+	a3 = a2 + way_incr;
+	a0e = base_addr + extent_per_way;
+	do {
+		asm volatile("ldc %0, sr" : : "r" (sr_with_bl));
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		asm volatile("ldc %0, sr" : : "r" (orig_sr));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+	} while (a0 < a0e);
+}
+
diff --git a/arch/sh/mm/clear_page.S b/arch/sh/mm/clear_page.S
index 08acead7b2a1..7b96425ae270 100644
--- a/arch/sh/mm/clear_page.S
+++ b/arch/sh/mm/clear_page.S
@@ -193,102 +193,5 @@ ENTRY(__clear_user_page)
 	 nop
 .L4096:	.word	4096
 
-ENTRY(__flush_cache_4096)
-	mov.l	1f,r3
-	add	r6,r3
-	mov	r4,r0
-	mov	#64,r2
-	shll	r2
-	mov	#64,r6
-	jmp	@r3
-	 mov	#96,r7
-	.align	2
-1:	.long	2f
-2:
-	.rept	32
-	mov.l	r5,@r0
-	mov.l	r5,@(32,r0)
-	mov.l	r5,@(r0,r6)
-	mov.l	r5,@(r0,r7)
-	add	r2,r5
-	add	r2,r0
-	.endr
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	rts
-	 nop
-
-ENTRY(__flush_dcache_all)
-	mov.l	2f,r0
-	mov.l	3f,r4
-	and	r0,r4		! r4 = (unsigned long)&empty_zero_page[0] & ~0xffffc000
-	stc	sr,r1		! save SR
-	mov.l	4f,r2
-	or	r1,r2
-	mov	#32,r3
-	shll2	r3
-1:
-	ldc	r2,sr		! set BL bit
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	ldc	r1,sr		! restore SR
-	dt	r3
-	bf/s	1b
-	 add	#32,r4
-
-	rts
-	 nop
-	.align	2
-2:	.long	0xffffc000
-3:	.long	empty_zero_page
-4:	.long	0x10000000	! BL bit
-
-/* __flush_cache_4096_all(unsigned long addr) */
-ENTRY(__flush_cache_4096_all)
-	mov.l	2f,r0
-	mov.l	3f,r2
-	and	r0,r2
-	or	r2,r4		! r4 = addr | (unsigned long)&empty_zero_page[0] & ~0x3fff
-	stc	sr,r1		! save SR
-	mov.l	4f,r2
-	or	r1,r2
-	mov	#32,r3
-1:
-	ldc	r2,sr		! set BL bit
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	ldc	r1,sr		! restore SR
-	dt	r3
-	bf/s	1b
-	 add	#32,r4
-
-	rts
-	 nop
-	.align	2
-2:	.long	0xffffc000
-3:	.long	empty_zero_page
-4:	.long	0x10000000	! BL bit
 #endif
+
diff --git a/include/asm-sh/cache.h b/include/asm-sh/cache.h
index 656fdfe9e8b4..33f13367054b 100644
--- a/include/asm-sh/cache.h
+++ b/include/asm-sh/cache.h
@@ -23,15 +23,29 @@
 #define L1_CACHE_ALIGN(x)	(((x)+(L1_CACHE_BYTES-1))&~(L1_CACHE_BYTES-1))
 
 struct cache_info {
-	unsigned int ways;
-	unsigned int sets;
-	unsigned int linesz;
+	unsigned int ways;		/* Number of cache ways */
+	unsigned int sets;		/* Number of cache sets */
+	unsigned int linesz;		/* Cache line size (bytes) */
 
-	unsigned int way_incr;
+	unsigned int way_size;		/* sets * line size */
 
+	/*
+	 * way_incr is the address offset for accessing the next way
+	 * in memory mapped cache array ops.
+	 */
+	unsigned int way_incr;
 	unsigned int entry_shift;
 	unsigned int entry_mask;
 
+	/*
+	 * Compute a mask which selects the address bits which overlap between
+	 * 1. those used to select the cache set during indexing
+	 * 2. those in the physical page number.
+	 */
+	unsigned int alias_mask;
+
+	unsigned int n_aliases;		/* Number of aliases */
+
 	unsigned long flags;
 };
 
-- 
cgit v1.2.3


From e86d6b66f5b38680746b2cb71186d90af17f150f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:20:54 +0900
Subject: sh: prefetch()/prefetchw() support.

SH-2/3/4 are able to prefetch, add support for it..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/processor.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index eeb0f48bb99e..e99aff706cf7 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -266,5 +266,18 @@ extern unsigned long get_wchan(struct task_struct *p);
 #define cpu_sleep()	__asm__ __volatile__ ("sleep" : : : "memory")
 #define cpu_relax()	barrier()
 
+#if defined(CONFIG_CPU_SH2A) || defined(CONFIG_CPU_SH3) || \
+    defined(CONFIG_CPU_SH4)
+#define PREFETCH_STRIDE		L1_CACHE_BYTES
+#define ARCH_HAS_PREFETCH
+#define ARCH_HAS_PREFETCHW
+static inline void prefetch(void *x)
+{
+	__asm__ __volatile__ ("pref @%0\n\t" : : "r" (x) : "memory");
+}
+
+#define prefetchw(x)	prefetch(x)
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PROCESSOR_H */
-- 
cgit v1.2.3


From a80fd21e52cc09ff1e4d36de5781173a5b87b2dc Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:26:53 +0900
Subject: sh: earlyprintk= support and cleanups.

Allow multiple early printk consoles via earlyprintk=.

With this change earlyprintk is no longer enabled by default,
it must be specified on the kernel command line. Optionally
with ,keep to prevent unreg by tty_io.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/early_printk.c | 106 +++++++++++++++++++++++++++++++-----------
 arch/sh/kernel/setup.c        |  22 +++++++--
 include/asm-sh/setup.h        |   2 +
 3 files changed, 100 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/early_printk.c b/arch/sh/kernel/early_printk.c
index 1378db375e17..a00022722e9e 100644
--- a/arch/sh/kernel/early_printk.c
+++ b/arch/sh/kernel/early_printk.c
@@ -3,7 +3,7 @@
  *
  *  Copyright (C) 1999, 2000  Niibe Yutaka
  *  Copyright (C) 2002  M. R. Brown
- *  Copyright (C) 2004  Paul Mundt
+ *  Copyright (C) 2004 - 2006  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -49,7 +49,7 @@ static int __init sh_console_setup(struct console *co, char *options)
 	return 0;
 }
 
-static struct console early_console = {
+static struct console bios_console = {
 	.name		= "bios",
 	.write		= sh_console_write,
 	.setup		= sh_console_setup,
@@ -59,34 +59,43 @@ static struct console early_console = {
 #endif
 
 #ifdef CONFIG_EARLY_SCIF_CONSOLE
+#include <linux/serial_core.h>
+#include "../../../drivers/serial/sh-sci.h"
+
+#ifdef CONFIG_CPU_SH4
 #define SCIF_REG	0xffe80000
+#elif defined(CONFIG_CPU_SUBTYPE_SH72060)
+#define SCIF_REG	0xfffe9800
+#else
+#error "Undefined SCIF for this subtype"
+#endif
+
+static struct uart_port scif_port = {
+	.mapbase	= SCIF_REG,
+	.membase	= (char __iomem *)SCIF_REG,
+};
 
 static void scif_sercon_putc(int c)
 {
-	while (!(ctrl_inw(SCIF_REG + 0x10) & 0x20)) ;
+	while (((sci_in(&scif_port, SCFDR) & 0x1f00 >> 8) == 16))
+		;
 
-	ctrl_outb(c, SCIF_REG + 12);
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0x9f), SCIF_REG + 0x10);
+	sci_out(&scif_port, SCxTDR, c);
+	sci_in(&scif_port, SCxSR);
+	sci_out(&scif_port, SCxSR, 0xf3 & ~(0x20 | 0x40));
+
+	while ((sci_in(&scif_port, SCxSR) & 0x40) == 0);
+		;
 
 	if (c == '\n')
 		scif_sercon_putc('\r');
 }
 
-static void scif_sercon_flush(void)
-{
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0xbf), SCIF_REG + 0x10);
-
-	while (!(ctrl_inw(SCIF_REG + 0x10) & 0x40)) ;
-
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0xbf), SCIF_REG + 0x10);
-}
-
-static void scif_sercon_write(struct console *con, const char *s, unsigned count)
+static void scif_sercon_write(struct console *con, const char *s,
+			      unsigned count)
 {
 	while (count-- > 0)
 		scif_sercon_putc(*s++);
-
-	scif_sercon_flush();
 }
 
 static int __init scif_sercon_setup(struct console *con, char *options)
@@ -96,7 +105,7 @@ static int __init scif_sercon_setup(struct console *con, char *options)
 	return 0;
 }
 
-static struct console early_console = {
+static struct console scif_console = {
 	.name		= "sercon",
 	.write		= scif_sercon_write,
 	.setup		= scif_sercon_setup,
@@ -104,7 +113,7 @@ static struct console early_console = {
 	.index		= -1,
 };
 
-void scif_sercon_init(int baud)
+static void scif_sercon_init(int baud)
 {
 	ctrl_outw(0, SCIF_REG + 8);
 	ctrl_outw(0, SCIF_REG);
@@ -122,16 +131,61 @@ void scif_sercon_init(int baud)
 }
 #endif
 
-void __init enable_early_printk(void)
+/*
+ * Setup a default console, if more than one is compiled in, rely on the
+ * earlyprintk= parsing to give priority.
+ */
+static struct console *early_console =
+#ifdef CONFIG_SH_STANDARD_BIOS
+	&bios_console
+#elif defined(CONFIG_EARLY_SCIF_CONSOLE)
+	&scif_console
+#else
+	NULL
+#endif
+	;
+
+static int __initdata keep_early;
+
+int __init setup_early_printk(char *opt)
 {
-#ifdef CONFIG_EARLY_SCIF_CONSOLE
-	scif_sercon_init(115200);
+	char *space;
+	char buf[256];
+
+	strlcpy(buf, opt, sizeof(buf));
+	space = strchr(buf, ' ');
+	if (space)
+		*space = 0;
+
+	if (strstr(buf, "keep"))
+		keep_early = 1;
+
+#ifdef CONFIG_SH_STANDARD_BIOS
+	if (!strncmp(buf, "bios", 4))
+		early_console = &bios_console;
+#endif
+#if defined(CONFIG_EARLY_SCIF_CONSOLE)
+	if (!strncmp(buf, "serial", 6)) {
+		early_console = &scif_console;
+
+#ifdef CONFIG_CPU_SH4
+		scif_sercon_init(115200);
+#endif
+	}
 #endif
-	register_console(&early_console);
+
+	if (likely(early_console))
+		register_console(early_console);
+
+	return 1;
 }
+__setup("earlyprintk=", setup_early_printk);
 
-void disable_early_printk(void)
+void __init disable_early_printk(void)
 {
-	unregister_console(&early_console);
+	if (!keep_early) {
+		printk("disabling early console\n");
+		unregister_console(early_console);
+	} else
+		printk("keeping early console\n");
 }
-
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index e75189cb1db7..cff8d36f91b0 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -145,6 +145,24 @@ static inline void parse_cmdline (char ** cmdline_p, char mv_name[MV_NAME_SIZE],
 				memory_end = memory_start + mem_size;
 			}
 		}
+
+#ifdef CONFIG_EARLY_PRINTK
+		if (c == ' ' && !memcmp(from, "earlyprintk=", 12)) {
+			char *ep_end;
+
+			if (to != command_line)
+				to--;
+
+			from += 12;
+			ep_end = strchr(from, ' ');
+
+			setup_early_printk(from);
+			printk("early console enabled\n");
+
+			from = ep_end;
+		}
+#endif
+
 		if (c == ' ' && !memcmp(from, "sh_mv=", 6)) {
 			char* mv_end;
 			char* mv_comma;
@@ -245,11 +263,7 @@ void __init setup_arch(char **cmdline_p)
 	unsigned long bootmap_size;
 	unsigned long start_pfn, max_pfn, max_low_pfn;
 
-#ifdef CONFIG_EARLY_PRINTK
-	extern void enable_early_printk(void);
 
-	enable_early_printk();
-#endif
 #ifdef CONFIG_CMDLINE_BOOL
         strcpy(COMMAND_LINE, CONFIG_CMDLINE);
 #endif
diff --git a/include/asm-sh/setup.h b/include/asm-sh/setup.h
index d19de7c8df4e..34ca8a7f06ba 100644
--- a/include/asm-sh/setup.h
+++ b/include/asm-sh/setup.h
@@ -4,5 +4,7 @@
 
 #define COMMAND_LINE_SIZE 256
 
+int setup_early_printk(char *);
+
 #endif /* _SH_SETUP_H */
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From 5b19c9081fbd0882c936ec087bf9055a20251dec Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:31:40 +0900
Subject: sh: Support for SH7770/SH7780 CPU subtypes.

Merge support for SH7770 and SH7780 SH-4A subtypes.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/ex.S    | 164 +++++++++++++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/probe.c |  15 ++++
 arch/sh/kernel/cpu/sh4/sq.c    |   2 +
 arch/sh/mm/tlb-sh4.c           |   4 +
 include/asm-sh/bugs.h          |   4 +
 include/asm-sh/cpu-sh4/cache.h |   2 +
 6 files changed, 191 insertions(+)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S
index 26a27df06505..af5ecbddea55 100644
--- a/arch/sh/kernel/cpu/sh4/ex.S
+++ b/arch/sh/kernel/cpu/sh4/ex.S
@@ -72,6 +72,7 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	! 1110
 	.long	exception_error		
 	! Internal hardware
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	.long	do_IRQ	! TMU0 tuni0	/* 400 */
 	.long	do_IRQ	! TMU1 tuni1
 	.long	do_IRQ	! TMU2 tuni2
@@ -379,5 +380,168 @@ ENTRY(interrupt_table)
 	.long	exception_error			! 141 0x13a0
 	.long	exception_error			! 142 0x13c0
 	.long	exception_error			! 143 0x13e0
+#elif defined(CONFIG_CPU_SUBTYPE_SH7770)
+	.long	do_IRQ	!  50 0x840
+	.long	do_IRQ	!  51 0x860
+	.long	do_IRQ	!  52 0x880
+	.long	do_IRQ	!  53 0x8a0
+	.long	do_IRQ	!  54 0x8c0
+	.long	do_IRQ	!  55 0x8e0
+	.long	do_IRQ	!  56 0x900
+	.long	do_IRQ	!  57 0x920
+	.long	do_IRQ	!  58 0x940
+	.long	do_IRQ	!  59 0x960
+	.long	do_IRQ	!  60 0x980
+	.long	do_IRQ	!  61 0x9a0
+	.long	do_IRQ	!  62 0x9c0
+	.long	do_IRQ	!  63 0x9e0
+	.long	do_IRQ	!  64 0xa00
+	.long	do_IRQ	!  65 0xa20
+	.long	do_IRQ	!  66 0xa4d
+	.long	do_IRQ	!  67 0xa60
+	.long	do_IRQ	!  68 0xa80
+	.long	do_IRQ	!  69 0xaa0
+	.long	do_IRQ	!  70 0xac0
+	.long	do_IRQ	!  71 0xae0
+	.long	do_IRQ	!  72 0xb00
+	.long	do_IRQ	!  73 0xb20
+	.long	do_IRQ	!  74 0xb40
+	.long	do_IRQ	!  75 0xb60
+	.long	do_IRQ	!  76 0xb80
+	.long	do_IRQ	!  77 0xba0
+	.long	do_IRQ	!  78 0xbc0
+	.long	do_IRQ	!  79 0xbe0
+	.long	do_IRQ	!  80 0xc00
+	.long	do_IRQ	!  81 0xc20
+	.long	do_IRQ	!  82 0xc40
+	.long	do_IRQ	!  83 0xc60
+	.long	do_IRQ	!  84 0xc80
+	.long	do_IRQ	!  85 0xca0
+	.long	do_IRQ	!  86 0xcc0
+	.long	do_IRQ	!  87 0xce0
+	.long	do_IRQ	!  88 0xd00
+	.long	do_IRQ	!  89 0xd20
+	.long	do_IRQ	!  90 0xd40
+	.long	do_IRQ	!  91 0xd60
+	.long	do_IRQ	!  92 0xd80
+	.long	do_IRQ	!  93 0xda0
+	.long	do_IRQ	!  94 0xdc0
+	.long	do_IRQ	!  95 0xde0
+	.long	do_IRQ	!  96 0xe00
+	.long	do_IRQ	!  97 0xe20
+	.long	do_IRQ	!  98 0xe40
+	.long	do_IRQ	!  99 0xe60
+	.long	do_IRQ	! 100 0xe80
+	.long	do_IRQ	! 101 0xea0
+	.long	do_IRQ	! 102 0xec0
+	.long	do_IRQ	! 103 0xee0
+	.long	do_IRQ	! 104 0xf00
+	.long	do_IRQ	! 105 0xf20
+	.long	do_IRQ	! 106 0xf40
+	.long	do_IRQ	! 107 0xf60
+	.long	do_IRQ	! 108 0xf80
+#endif
+#else
+	.long	exception_error		/* 400 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! RTC	ati
+	.long	do_IRQ	!	pri
+	.long	do_IRQ	!	cui
+	.long	exception_error
+	.long	exception_error		/* 500 */
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! WDT	iti	/* 560 */
+	.long	do_IRQ	! TMU-ch0
+	.long	do_IRQ	! TMU-ch1
+	.long	do_IRQ	! TMU-ch2
+	.long	do_IRQ	! ticpi2	/* 5E0 */
+	.long	do_IRQ	! 32 Hitachi UDI	/* 600 */
+	.long	exception_error
+	.long	do_IRQ	! 34 DMAC dmte0
+	.long	do_IRQ	! 35	  dmte1
+	.long	do_IRQ	! 36	  dmte2
+	.long	do_IRQ	! 37	  dmte3
+	.long	do_IRQ	! 38	  dmae
+	.long	exception_error			! 39	/* 6E0 */
+	.long	do_IRQ	! 40 SCIF-ch0 eri		/* 700 */
+	.long	do_IRQ	! 41	      rxi
+	.long	do_IRQ	! 42	      bri
+	.long	do_IRQ	! 43	      txi
+	.long	do_IRQ	! 44 DMAC dmte4		/* 780 */
+	.long	do_IRQ	! 45	  dmte5
+	.long	do_IRQ	! 46	  dmte6
+	.long	do_IRQ	! 47	  dmte7		/* 7E0 */
+#if defined(CONFIG_SH_FPU)
+	.long	do_fpu_state_restore	! 48	/* 800 */
+	.long	do_fpu_state_restore	! 49	/* 820 */
+#else
+	.long	exception_error
+	.long	exception_error
+#endif
+	.long	exception_error			/* 840 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! 56 CMT	/* 900 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! 60 HAC
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! PCI serr	/* A00 */
+	.long	do_IRQ	!     INTA
+	.long	do_IRQ	!     INTB
+	.long	do_IRQ	!     INTC
+	.long	do_IRQ	!     INTD
+	.long	do_IRQ	!     err
+	.long	do_IRQ	!     pwd3
+	.long	do_IRQ	!     pwd2
+	.long	do_IRQ	!     pwd1	/* B00 */
+	.long	do_IRQ	!     pwd0
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! SCIF-ch1 eri	/* B80 */
+	.long	do_IRQ	!	   rxi
+	.long	do_IRQ	!	   bri
+	.long	do_IRQ	!	   txi
+	.long	do_IRQ	! SIOF		/* C00 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! HSPI		/* C80 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! MMCIF	fatat	/* D00 */
+	.long	do_IRQ	!	tran
+	.long	do_IRQ	!	err
+	.long	do_IRQ	!	frdy
+	.long	do_IRQ	! DMAC dmint8	/* D80 */
+	.long	do_IRQ	!      dmint9
+	.long	do_IRQ	!      dmint10
+	.long	do_IRQ	!      dmint11
+	.long	do_IRQ	! TMU-ch3	/* E00 */
+	.long	do_IRQ	! TMU-ch4
+	.long	do_IRQ	! TMU-ch5
+	.long	exception_error
+	.long	do_IRQ	! SSI
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! FLCTL	flste	/* F00 */
+	.long	do_IRQ	!	fltend
+	.long	do_IRQ	!	fltrq0
+	.long	do_IRQ	!	fltrq1
+	.long	do_IRQ	! GPIO gpioi0	/* F80 */
+	.long	do_IRQ	!      gpioi1
+	.long	do_IRQ	!      gpioi2
+	.long	do_IRQ	!      gpioi3
 #endif
 
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 89986e70d041..85ff48c1533a 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -78,6 +78,21 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->dcache.ways = 4;
 		cpu_data->flags &= ~CPU_HAS_FPU;
 		break;
+	case 0x2001:
+	case 0x2004:
+		cpu_data->type = CPU_SH7770;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		break;
+	case 0x2006:
+	case 0x200A:
+		if (prr == 0x61)
+			cpu_data->type = CPU_SH7781;
+		else
+			cpu_data->type = CPU_SH7780;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
 		break;
diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index 00f6a3c8c43b..32c93b781b51 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -166,7 +166,9 @@ static struct sq_mapping *__sq_remap(struct sq_mapping *map)
 	ctrl_outl((pteh & MMU_VPN_MASK) | get_asid(), MMU_PTEH);
 
 	ptel = map->addr & PAGE_MASK;
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
+#endif
 
 	pgprot = pgprot_noncached(PAGE_KERNEL);
 
diff --git a/arch/sh/mm/tlb-sh4.c b/arch/sh/mm/tlb-sh4.c
index 115b1b6be40b..96e5fb0ac4fa 100644
--- a/arch/sh/mm/tlb-sh4.c
+++ b/arch/sh/mm/tlb-sh4.c
@@ -36,7 +36,9 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	unsigned long vpn;
 	struct page *page;
 	unsigned long pfn;
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	unsigned long ptea;
+#endif
 
 	/* Ptrace may call this routine. */
 	if (vma && current->active_mm != vma->vm_mm)
@@ -59,10 +61,12 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	ctrl_outl(vpn, MMU_PTEH);
 
 	pteval = pte_val(pte);
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	/* Set PTEA register */
 	/* TODO: make this look less hacky */
 	ptea = ((pteval >> 28) & 0xe) | (pteval & 0x1);
 	ctrl_outl(ptea, MMU_PTEA);
+#endif
 
 	/* Set PTEL register */
 	pteval &= _PAGE_FLAGS_HARDWARE_MASK; /* drop software flags */
diff --git a/include/asm-sh/bugs.h b/include/asm-sh/bugs.h
index a6de3d06a3d9..b4000c8bf31b 100644
--- a/include/asm-sh/bugs.h
+++ b/include/asm-sh/bugs.h
@@ -32,6 +32,10 @@ static void __init check_bugs(void)
 	case CPU_SH7750 ... CPU_SH4_501:
 		*p++ = '4';
 		break;
+	case CPU_SH7770 ... CPU_SH7781:
+		*p++ = '4';
+		*p++ = 'a';
+		break;
 	default:
 		*p++ = '?';
 		*p++ = '!';
diff --git a/include/asm-sh/cpu-sh4/cache.h b/include/asm-sh/cpu-sh4/cache.h
index 1fe20359312c..6e9c7e6ee8e4 100644
--- a/include/asm-sh/cpu-sh4/cache.h
+++ b/include/asm-sh/cpu-sh4/cache.h
@@ -22,7 +22,9 @@
 #define CCR_CACHE_ICE	0x0100	/* Instruction Cache Enable */
 #define CCR_CACHE_ICI	0x0800	/* IC Invalidate */
 #define CCR_CACHE_IIX	0x8000	/* IC Index Enable */
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 #define CCR_CACHE_EMODE	0x80000000	/* EMODE Enable */
+#endif
 
 /* Default CCR setup: 8k+16k-byte cache,P1-wb,enable */
 #define CCR_CACHE_ENABLE	(CCR_CACHE_OCE|CCR_CACHE_ICE)
-- 
cgit v1.2.3


From ef9a1d4c0c383f75710f6adf2abb8cc264877e2c Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Wed, 27 Sep 2006 14:32:57 +0900
Subject: sh: remove cpu_online() definition from <asm/smp.h>

It's defined in <linux/cpumask.h> and log is horribly flooded by
"redefined" messages.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/smp.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/smp.h b/include/asm-sh/smp.h
index f57c4fe9692a..71ecddf70db3 100644
--- a/include/asm-sh/smp.h
+++ b/include/asm-sh/smp.h
@@ -19,11 +19,6 @@
 #include <asm/atomic.h>
 #include <asm/current.h>
 
-extern cpumask_t cpu_online_map;
-extern cpumask_t cpu_possible_map;
-
-#define cpu_online(cpu)		cpu_isset(cpu, cpu_online_map)
-
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 /* I've no idea what the real meaning of this is */
-- 
cgit v1.2.3


From 75c92acdd5b19a5e3536ed670e1122d73c635b4a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:36:44 +0900
Subject: sh: Wire up new syscalls.

The syscall table has lagged behind a bit, wire up the new ones..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/syscalls.S | 28 ++++++++++++++++++++++++++++
 include/asm-sh/unistd.h   | 26 ++++++++++++++++++++++++--
 2 files changed, 52 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
index f500304cbd69..ada61b0d9a73 100644
--- a/arch/sh/kernel/syscalls.S
+++ b/arch/sh/kernel/syscalls.S
@@ -322,3 +322,31 @@ ENTRY(sys_call_table)
 	.long sys_add_key		/* 285 */
 	.long sys_request_key
 	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get
+	.long sys_inotify_init		/* 290 */
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch
+	.long sys_migrate_pages
+	.long sys_openat
+	.long sys_mkdirat		/* 295 */
+	.long sys_mknodat
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_fstatat64
+	.long sys_unlinkat		/* 300 */
+	.long sys_renameat
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat
+	.long sys_fchmodat		/* 305 */
+	.long sys_faccessat
+	.long sys_ni_syscall		/* Reserved for pselect6 */
+	.long sys_ni_syscall		/* Reserved for ppoll */
+	.long sys_unshare
+	.long sys_set_robust_list	/* 310 */
+	.long sys_get_robust_list
+	.long sys_splice
+	.long sys_sync_file_range
+	.long sys_tee
+	.long sys_vmsplice		/* 315 */
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index 76b5430cb458..49c6a6ee0ab2 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -300,9 +300,31 @@
 #define __NR_inotify_init	290
 #define __NR_inotify_add_watch	291
 #define __NR_inotify_rm_watch	292
+#define __NR_migrate_pages	293
+#define __NR_openat		294
+#define __NR_mkdirat		295
+#define __NR_mknodat		296
+#define __NR_fchownat		297
+#define __NR_futimesat		298
+#define __NR_fstatat64		299
+#define __NR_unlinkat		300
+#define __NR_renameat		301
+#define __NR_linkat		302
+#define __NR_symlinkat		303
+#define __NR_readlinkat		304
+#define __NR_fchmodat		305
+#define __NR_faccessat		305
+#define __NR_pselect6		307
+#define __NR_ppoll		308
+#define __NR_unshare		309
+#define __NR_set_robust_list	310
+#define __NR_get_robust_list	311
+#define __NR_splice		312
+#define __NR_sync_file_range	313
+#define __NR_tee		314
+#define __NR_vmsplice		315
 
-
-#define NR_syscalls 293
+#define NR_syscalls 316
 
 #ifdef __KERNEL__
 
-- 
cgit v1.2.3


From 8b395265f81817385f12e62f03f795efb732a445 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:38:02 +0900
Subject: sh: Fix fatal oops in copy_user_page() on sh4a (SH7780).

We had a pretty interesting oops happening, where copy_user_page()
was down()'ing p3map_sem[] with a bogus offset (particularly, an
offset that hadn't been initialized with sema_init(), due to the
mismatch between cpu_data->dcache.n_aliases and what was assumed
based off of the old CACHE_ALIAS value).

Luckily, spinlock debugging caught this for us, and so we drop
the old hardcoded CACHE_ALIAS for sh4 completely and rely on the
run-time probed cpu_data->dcache.alias_mask. This in turn gets
the p3map_sem[] index right, and everything works again.

While we're at it, also convert to 4-level page tables..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/pg-sh4.c                 | 24 ++++++++++++++----------
 include/asm-sh/cpu-sh4/cacheflush.h |  4 ----
 2 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/arch/sh/mm/pg-sh4.c b/arch/sh/mm/pg-sh4.c
index c776b60fc250..07371ed7a313 100644
--- a/arch/sh/mm/pg-sh4.c
+++ b/arch/sh/mm/pg-sh4.c
@@ -2,7 +2,7 @@
  * arch/sh/mm/pg-sh4.c
  *
  * Copyright (C) 1999, 2000, 2002  Niibe Yutaka
- * Copyright (C) 2002  Paul Mundt
+ * Copyright (C) 2002 - 2005  Paul Mundt
  *
  * Released under the terms of the GNU GPL v2.0.
  */
@@ -23,6 +23,8 @@
 
 extern struct semaphore p3map_sem[];
 
+#define CACHE_ALIAS (cpu_data->dcache.alias_mask)
+
 /*
  * clear_user_page
  * @to: P1 address
@@ -35,14 +37,15 @@ void clear_user_page(void *to, unsigned long address, struct page *page)
 	if (((address ^ (unsigned long)to) & CACHE_ALIAS) == 0)
 		clear_page(to);
 	else {
-		pgprot_t pgprot = __pgprot(_PAGE_PRESENT | 
+		pgprot_t pgprot = __pgprot(_PAGE_PRESENT |
 					   _PAGE_RW | _PAGE_CACHABLE |
-					   _PAGE_DIRTY | _PAGE_ACCESSED | 
+					   _PAGE_DIRTY | _PAGE_ACCESSED |
 					   _PAGE_HW_SHARED | _PAGE_FLAGS_HARD);
 		unsigned long phys_addr = PHYSADDR(to);
 		unsigned long p3_addr = P3SEG + (address & CACHE_ALIAS);
-		pgd_t *dir = pgd_offset_k(p3_addr);
-		pmd_t *pmd = pmd_offset(dir, p3_addr);
+		pgd_t *pgd = pgd_offset_k(p3_addr);
+		pud_t *pud = pud_offset(pgd, p3_addr);
+		pmd_t *pmd = pmd_offset(pud, p3_addr);
 		pte_t *pte = pte_offset_kernel(pmd, p3_addr);
 		pte_t entry;
 		unsigned long flags;
@@ -67,21 +70,22 @@ void clear_user_page(void *to, unsigned long address, struct page *page)
  * @address: U0 address to be mapped
  * @page: page (virt_to_page(to))
  */
-void copy_user_page(void *to, void *from, unsigned long address, 
+void copy_user_page(void *to, void *from, unsigned long address,
 		    struct page *page)
 {
 	__set_bit(PG_mapped, &page->flags);
 	if (((address ^ (unsigned long)to) & CACHE_ALIAS) == 0)
 		copy_page(to, from);
 	else {
-		pgprot_t pgprot = __pgprot(_PAGE_PRESENT | 
+		pgprot_t pgprot = __pgprot(_PAGE_PRESENT |
 					   _PAGE_RW | _PAGE_CACHABLE |
-					   _PAGE_DIRTY | _PAGE_ACCESSED | 
+					   _PAGE_DIRTY | _PAGE_ACCESSED |
 					   _PAGE_HW_SHARED | _PAGE_FLAGS_HARD);
 		unsigned long phys_addr = PHYSADDR(to);
 		unsigned long p3_addr = P3SEG + (address & CACHE_ALIAS);
-		pgd_t *dir = pgd_offset_k(p3_addr);
-		pmd_t *pmd = pmd_offset(dir, p3_addr);
+		pgd_t *pgd = pgd_offset_k(p3_addr);
+		pud_t *pud = pud_offset(pgd, p3_addr);
+		pmd_t *pmd = pmd_offset(pud, p3_addr);
 		pte_t *pte = pte_offset_kernel(pmd, p3_addr);
 		pte_t entry;
 		unsigned long flags;
diff --git a/include/asm-sh/cpu-sh4/cacheflush.h b/include/asm-sh/cpu-sh4/cacheflush.h
index f323567e085f..ea58c4c5944d 100644
--- a/include/asm-sh/cpu-sh4/cacheflush.h
+++ b/include/asm-sh/cpu-sh4/cacheflush.h
@@ -16,10 +16,6 @@
  *  caching; in which case they're only semi-broken),
  *  so we need them.
  */
-
-/* Page is 4K, OC size is 16K, there are four lines. */
-#define CACHE_ALIAS 0x00003000
-
 struct page;
 struct mm_struct;
 struct vm_area_struct;
-- 
cgit v1.2.3


From 634bf4f69b925950ddb09ef99ad7516a449a4333 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:48:09 +0900
Subject: sh: Fix libata build.

Drop virt_to_bus() from sg_dma_address() so libata builds.
While we're at it, move sg_dma_address() and sg_dma_len()
from pci.h to scatterlist.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pci.h         | 9 ---------
 include/asm-sh/scatterlist.h | 9 +++++++++
 2 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 0a523c85b11c..18a109de0f2e 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -87,15 +87,6 @@ static inline void pcibios_penalize_isa_irq(int irq, int active)
  */
 #define pci_dac_dma_supported(pci_dev, mask) (0)
 
-/* These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns, or alternatively stop on the first sg_dma_len(sg) which
- * is 0.
- */
-#define sg_dma_address(sg)	(virt_to_bus((sg)->dma_address))
-#define sg_dma_len(sg)		((sg)->length)
-
 #ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
diff --git a/include/asm-sh/scatterlist.h b/include/asm-sh/scatterlist.h
index 7b91df140022..d19e7cd3b023 100644
--- a/include/asm-sh/scatterlist.h
+++ b/include/asm-sh/scatterlist.h
@@ -10,4 +10,13 @@ struct scatterlist {
 
 #define ISA_DMA_THRESHOLD (0x1fffffff)
 
+/* These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns, or alternatively stop on the first sg_dma_len(sg) which
+ * is 0.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
+#define sg_dma_len(sg)		((sg)->length)
+
 #endif /* !(__ASM_SH_SCATTERLIST_H) */
-- 
cgit v1.2.3


From 94c0fa520cc169ccf661e9c03b5b95f74d1520b8 Mon Sep 17 00:00:00 2001
From: kogiidena <kogiidena@eggplant.ddo.jp>
Date: Wed, 27 Sep 2006 14:53:35 +0900
Subject: sh: landisk board support.

This adds support for the I-O DATA Landisk.

Signed-off-by: kogiidena <kogiidena@eggplant.ddo.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                        |    1 +
 arch/sh/boards/landisk/Makefile         |    7 +
 arch/sh/boards/landisk/io.c             |  322 ++++++++
 arch/sh/boards/landisk/irq.c            |  105 +++
 arch/sh/boards/landisk/landisk_pwb.c    |  348 ++++++++
 arch/sh/boards/landisk/rtc.c            |  102 +++
 arch/sh/boards/landisk/setup.c          |  218 +++++
 arch/sh/configs/landisk_defconfig       | 1373 +++++++++++++++++++++++++++++++
 arch/sh/tools/mach-types                |    2 +-
 include/asm-sh/landisk/gio.h            |   45 +
 include/asm-sh/landisk/ide.h            |   14 +
 include/asm-sh/landisk/iodata_landisk.h |   78 ++
 12 files changed, 2614 insertions(+), 1 deletion(-)
 create mode 100644 arch/sh/boards/landisk/Makefile
 create mode 100644 arch/sh/boards/landisk/io.c
 create mode 100644 arch/sh/boards/landisk/irq.c
 create mode 100644 arch/sh/boards/landisk/landisk_pwb.c
 create mode 100644 arch/sh/boards/landisk/rtc.c
 create mode 100644 arch/sh/boards/landisk/setup.c
 create mode 100644 arch/sh/configs/landisk_defconfig
 create mode 100644 include/asm-sh/landisk/gio.h
 create mode 100644 include/asm-sh/landisk/ide.h
 create mode 100644 include/asm-sh/landisk/iodata_landisk.h

(limited to 'include')

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 66cf0ea6d38f..e9112def48dd 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -105,6 +105,7 @@ machdir-$(CONFIG_SH_RTS7751R2D)			:= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
+machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
diff --git a/arch/sh/boards/landisk/Makefile b/arch/sh/boards/landisk/Makefile
new file mode 100644
index 000000000000..60e75c6db0eb
--- /dev/null
+++ b/arch/sh/boards/landisk/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for I-O DATA DEVICE, INC. "LANDISK Series"
+#
+
+obj-y	 := setup.o io.o irq.o rtc.o landisk_pwb.o
+
+obj-$(CONFIG_PCI) += pci.o
diff --git a/arch/sh/boards/landisk/io.c b/arch/sh/boards/landisk/io.c
new file mode 100644
index 000000000000..1f1679af09d0
--- /dev/null
+++ b/arch/sh/boards/landisk/io.c
@@ -0,0 +1,322 @@
+/*
+ * arch/sh/boards/landisk/io.c
+ *
+ * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
+ * Based largely on io_se.c.
+ *
+ * I/O routine for I-O Data Device, Inc. LANDISK.
+ *
+ * Initial version only to support LAN access; some
+ * placeholder code from io_landisk.c left in with the
+ * expectation of later SuperIO and PCMCIA access.
+ */
+/*
+ * modifed by kogiidena
+ * 2005.03.03
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/io.h>
+#include <asm/landisk/iodata_landisk.h>
+#include <asm/addrspace.h>
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include "../../drivers/pci/pci-sh7751.h"
+
+extern void *area5_io_base;	/* Area 5 I/O Base address */
+extern void *area6_io_base;	/* Area 6 I/O Base address */
+
+/*
+ * The 7751R LANDISK uses the built-in PCI controller (PCIC)
+ * of the 7751R processor, and has a SuperIO accessible via the PCI.
+ * The board also includes a PCMCIA controller on its memory bus,
+ * like the other Solution Engine boards.
+ */
+
+#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
+#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
+#define PCI_IO_AREA	SH7751_PCI_IO_BASE
+#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
+
+#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
+
+#define maybebadio(name,port) \
+  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
+	 #name, (port), (__u32) __builtin_return_address(0))
+
+static inline void delay(void)
+{
+	ctrl_inw(0xa0000000);
+}
+
+static inline unsigned long port2adr(unsigned int port)
+{
+	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
+		if (port == 0x3f6)
+			return ((unsigned long)area5_io_base + 0x2c);
+		else
+			return ((unsigned long)area5_io_base + PA_PIDE_OFFSET +
+				((port - 0x1f0) << 1));
+	else if ((0x170 <= port && port < 0x178) || port == 0x376)
+		if (port == 0x376)
+			return ((unsigned long)area6_io_base + 0x2c);
+		else
+			return ((unsigned long)area6_io_base + PA_SIDE_OFFSET +
+				((port - 0x170) << 1));
+	else
+		maybebadio(port2adr, (unsigned long)port);
+
+	return port;
+}
+
+/* In case someone configures the kernel w/o PCI support: in that */
+/* scenario, don't ever bother to check for PCI-window addresses */
+
+/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
+#if defined(CONFIG_PCI)
+#define CHECK_SH7751_PCIIO(port) \
+  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
+#else
+#define CHECK_SH_7751_PCIIO(port) (0)
+#endif
+
+/*
+ * General outline: remap really low stuff [eventually] to SuperIO,
+ * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
+ * is mapped through the PCI IO window.  Stuff with high bits (PXSEG)
+ * should be way beyond the window, and is used  w/o translation for
+ * compatibility.
+ */
+unsigned char landisk_inb(unsigned long port)
+{
+	if (PXSEG(port))
+		return *(volatile unsigned char *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else
+		return (*(volatile unsigned short *)port2adr(port) & 0xff);
+}
+
+unsigned char landisk_inb_p(unsigned long port)
+{
+	unsigned char v;
+
+	if (PXSEG(port))
+		v = *(volatile unsigned char *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else
+		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
+	delay();
+
+	return v;
+}
+
+unsigned short landisk_inw(unsigned long port)
+{
+	if (PXSEG(port))
+		return *(volatile unsigned short *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return *(volatile unsigned short *)PCI_IOMAP(port);
+	else
+		maybebadio(inw, port);
+
+	return 0;
+}
+
+unsigned int landisk_inl(unsigned long port)
+{
+	if (PXSEG(port))
+		return *(volatile unsigned long *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return *(volatile unsigned long *)PCI_IOMAP(port);
+	else
+		maybebadio(inl, port);
+
+	return 0;
+}
+
+void landisk_outb(unsigned char value, unsigned long port)
+{
+
+	if (PXSEG(port))
+		*(volatile unsigned char *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else
+		*(volatile unsigned short *)port2adr(port) = value;
+}
+
+void landisk_outb_p(unsigned char value, unsigned long port)
+{
+	if (PXSEG(port))
+		*(volatile unsigned char *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else
+		*(volatile unsigned short *)port2adr(port) = value;
+	delay();
+}
+
+void landisk_outw(unsigned short value, unsigned long port)
+{
+	if (PXSEG(port))
+		*(volatile unsigned short *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned short *)PCI_IOMAP(port) = value;
+	else
+		maybebadio(outw, port);
+}
+
+void landisk_outl(unsigned int value, unsigned long port)
+{
+	if (PXSEG(port))
+		*(volatile unsigned long *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned long *)PCI_IOMAP(port) = value;
+	else
+		maybebadio(outl, port);
+}
+
+void landisk_insb(unsigned long port, void *addr, unsigned long count)
+{
+	if (PXSEG(port))
+		while (count--)
+			*((unsigned char *)addr)++ =
+			    *(volatile unsigned char *)port;
+	else if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+
+		while (count--)
+			*((volatile unsigned char *)addr)++ = *bp;
+	} else {
+		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+
+		while (count--)
+			*((unsigned char *)addr)++ = *p;
+	}
+}
+
+void landisk_insw(unsigned long port, void *addr, unsigned long count)
+{
+	volatile __u16 *p;
+
+	if (PXSEG(port))
+		p = (volatile unsigned short *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else
+		p = (volatile unsigned short *)port2adr(port);
+	while (count--)
+		*((__u16 *) addr)++ = *p;
+}
+
+void landisk_insl(unsigned long port, void *addr, unsigned long count)
+{
+	if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
+
+		while (count--)
+			*((__u32 *) addr)++ = *p;
+	} else
+		maybebadio(insl, port);
+}
+
+void landisk_outsb(unsigned long port, const void *addr, unsigned long count)
+{
+	if (PXSEG(port))
+		while (count--)
+			*(volatile unsigned char *)port =
+			    *((unsigned char *)addr)++;
+	else if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+
+		while (count--)
+			*bp = *((volatile unsigned char *)addr)++;
+	} else {
+		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+
+		while (count--)
+			*p = *((unsigned char *)addr)++;
+	}
+}
+
+void landisk_outsw(unsigned long port, const void *addr, unsigned long count)
+{
+	volatile __u16 *p;
+
+	if (PXSEG(port))
+		p = (volatile unsigned short *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else
+		p = (volatile unsigned short *)port2adr(port);
+	while (count--)
+		*p = *((__u16 *) addr)++;
+}
+
+void landisk_outsl(unsigned long port, const void *addr, unsigned long count)
+{
+	if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
+
+		while (count--)
+			*p = *((__u32 *) addr)++;
+	} else
+		maybebadio(outsl, port);
+}
+
+/* For read/write calls, just copy generic (pass-thru); PCIMBR is  */
+/* already set up.  For a larger memory space, these would need to */
+/* reset PCIMBR as needed on a per-call basis...                   */
+
+unsigned char landisk_readb(unsigned long addr)
+{
+	return *(volatile unsigned char *)addr;
+}
+
+unsigned short landisk_readw(unsigned long addr)
+{
+	return *(volatile unsigned short *)addr;
+}
+
+unsigned int landisk_readl(unsigned long addr)
+{
+	return *(volatile unsigned long *)addr;
+}
+
+void landisk_writeb(unsigned char b, unsigned long addr)
+{
+	*(volatile unsigned char *)addr = b;
+}
+
+void landisk_writew(unsigned short b, unsigned long addr)
+{
+	*(volatile unsigned short *)addr = b;
+}
+
+void landisk_writel(unsigned int b, unsigned long addr)
+{
+	*(volatile unsigned long *)addr = b;
+}
+
+void *landisk_ioremap(unsigned long offset, unsigned long size)
+{
+	if (offset >= 0xfd000000)
+		return (void *)offset;
+	else
+		return (void *)P2SEGADDR(offset);
+}
+
+void landisk_iounmap(void *addr)
+{
+}
+
+/* Map ISA bus address to the real address. Only for PCMCIA.  */
+
+unsigned long landisk_isa_port2addr(unsigned long offset)
+{
+	return port2adr(offset);
+}
diff --git a/arch/sh/boards/landisk/irq.c b/arch/sh/boards/landisk/irq.c
new file mode 100644
index 000000000000..1dcc3feb7a44
--- /dev/null
+++ b/arch/sh/boards/landisk/irq.c
@@ -0,0 +1,105 @@
+/*
+ * arch/sh/boards/landisk/irq.c
+ *
+ * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
+ * Based largely on io_se.c.
+ *
+ * I/O routine for I-O Data Device, Inc. LANDISK.
+ *
+ * Initial version only to support LAN access; some
+ * placeholder code from io_landisk.c left in with the
+ * expectation of later SuperIO and PCMCIA access.
+ */
+/*
+ * modified by kogiidena
+ * 2005.03.03
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/landisk/iodata_landisk.h>
+
+static void enable_landisk_irq(unsigned int irq);
+static void disable_landisk_irq(unsigned int irq);
+
+/* shutdown is same as "disable" */
+#define shutdown_landisk_irq disable_landisk_irq
+
+static void ack_landisk_irq(unsigned int irq);
+static void end_landisk_irq(unsigned int irq);
+
+static unsigned int startup_landisk_irq(unsigned int irq)
+{
+	enable_landisk_irq(irq);
+	return 0;		/* never anything pending */
+}
+
+static void disable_landisk_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned char val;
+	unsigned char mask = 0xff ^ (0x01 << (irq - 5));
+
+	/* Set the priority in IPR to 0 */
+	local_irq_save(flags);
+	val = ctrl_inb(PA_IMASK);
+	val &= mask;
+	ctrl_outb(val, PA_IMASK);
+	local_irq_restore(flags);
+}
+
+static void enable_landisk_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned char val;
+	unsigned char value = (0x01 << (irq - 5));
+
+	/* Set priority in IPR back to original value */
+	local_irq_save(flags);
+	val = ctrl_inb(PA_IMASK);
+	val |= value;
+	ctrl_outb(val, PA_IMASK);
+	local_irq_restore(flags);
+}
+
+static void ack_landisk_irq(unsigned int irq)
+{
+	disable_landisk_irq(irq);
+}
+
+static void end_landisk_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+		enable_landisk_irq(irq);
+}
+
+static struct hw_interrupt_type landisk_irq_type = {
+	.typename = "LANDISK IRQ",
+	.startup = startup_landisk_irq,
+	.shutdown = shutdown_landisk_irq,
+	.enable = enable_landisk_irq,
+	.disable = disable_landisk_irq,
+	.ack = ack_landisk_irq,
+	.end = end_landisk_irq
+};
+
+static void make_landisk_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &landisk_irq_type;
+	disable_landisk_irq(irq);
+}
+
+/*
+ * Initialize IRQ setting
+ */
+void __init init_landisk_IRQ(void)
+{
+	int i;
+
+	for (i = 5; i < 14; i++)
+		make_landisk_irq(i);
+}
diff --git a/arch/sh/boards/landisk/landisk_pwb.c b/arch/sh/boards/landisk/landisk_pwb.c
new file mode 100644
index 000000000000..e75cb578a28b
--- /dev/null
+++ b/arch/sh/boards/landisk/landisk_pwb.c
@@ -0,0 +1,348 @@
+/*
+ * arch/sh/boards/landisk/landisk_pwb.c -- driver for the Power control switch.
+ *
+ * This driver will also support the I-O DATA Device, Inc. LANDISK Board.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copylight (C) 2002 Atom Create Engineering Co., Ltd.
+ *
+ * LED control drive function added by kogiidena
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/major.h>
+#include <linux/poll.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/interrupt.h>
+
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+#include <asm/landisk/iodata_landisk.h>
+
+#define SHUTDOWN_BTN_MINOR	1	/* Shutdown button device minor no. */
+#define LED_MINOR	       21	/* LED minor no. */
+#define BTN_MINOR	       22	/* BUTTON minor no. */
+#define GIO_MINOR	       40	/* GIO minor no. */
+
+static int openCnt;
+static int openCntLED;
+static int openCntGio;
+static int openCntBtn;
+static int landisk_btn;
+static int landisk_btnctrlpid;
+/*
+ * Functions prototypes
+ */
+
+static int gio_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
+		     unsigned long arg);
+
+static int swdrv_open(struct inode *inode, struct file *filp)
+{
+	int minor;
+
+	minor = MINOR(inode->i_rdev);
+	filp->private_data = (void *)minor;
+
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		if (openCnt > 0) {
+			return -EALREADY;
+		} else {
+			openCnt++;
+			return 0;
+		}
+	} else if (minor == LED_MINOR) {
+		if (openCntLED > 0) {
+			return -EALREADY;
+		} else {
+			openCntLED++;
+			return 0;
+		}
+	} else if (minor == BTN_MINOR) {
+		if (openCntBtn > 0) {
+			return -EALREADY;
+		} else {
+			openCntBtn++;
+			return 0;
+		}
+	} else if (minor == GIO_MINOR) {
+		if (openCntGio > 0) {
+			return -EALREADY;
+		} else {
+			openCntGio++;
+			return 0;
+		}
+	}
+	return -ENOENT;
+
+}
+
+static int swdrv_close(struct inode *inode, struct file *filp)
+{
+	int minor;
+
+	minor = MINOR(inode->i_rdev);
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		openCnt--;
+	} else if (minor == LED_MINOR) {
+		openCntLED--;
+	} else if (minor == BTN_MINOR) {
+		openCntBtn--;
+	} else if (minor == GIO_MINOR) {
+		openCntGio--;
+	}
+	return 0;
+}
+
+static int swdrv_read(struct file *filp, char *buff, size_t count,
+		      loff_t * ppos)
+{
+	int minor;
+	minor = (int)(filp->private_data);
+
+	if (!access_ok(VERIFY_WRITE, (void *)buff, count))
+		return -EFAULT;
+
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		if (landisk_btn & 0x10) {
+			put_user(1, buff);
+			return 1;
+		} else {
+			return 0;
+		}
+	}
+	return 0;
+}
+
+static int swdrv_write(struct file *filp, const char *buff, size_t count,
+		       loff_t * ppos)
+{
+	int minor;
+	minor = (int)(filp->private_data);
+
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		return count;
+	}
+	return count;
+}
+
+static irqreturn_t sw_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	landisk_btn = (0x0ff & (~ctrl_inb(PA_STATUS)));
+	disable_irq(IRQ_BUTTON);
+	disable_irq(IRQ_POWER);
+	ctrl_outb(0x00, PA_PWRINT_CLR);
+
+	if (landisk_btnctrlpid != 0) {
+		kill_proc(landisk_btnctrlpid, SIGUSR1, 1);
+		landisk_btnctrlpid = 0;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static struct file_operations swdrv_fops = {
+	.read = swdrv_read,	/* read */
+	.write = swdrv_write,	/* write */
+	.open = swdrv_open,	/* open */
+	.release = swdrv_close,	/* release */
+	.ioctl = gio_ioctl,	/* ioctl */
+
+};
+
+static char banner[] __initdata =
+    KERN_INFO "LANDISK and USL-5P Button, LED and GIO driver initialized\n";
+
+int __init swdrv_init(void)
+{
+	int error;
+
+	printk("%s", banner);
+
+	openCnt = 0;
+	openCntLED = 0;
+	openCntBtn = 0;
+	openCntGio = 0;
+	landisk_btn = 0;
+	landisk_btnctrlpid = 0;
+
+	if ((error = register_chrdev(SHUTDOWN_BTN_MAJOR, "swdrv", &swdrv_fops))) {
+		printk(KERN_ERR
+		       "Button, LED and GIO driver:Couldn't register driver, error=%d\n",
+		       error);
+		return 1;
+	}
+
+	if (request_irq(IRQ_POWER, sw_interrupt, 0, "SHUTDOWNSWITCH", NULL)) {
+		printk(KERN_ERR "Unable to get IRQ 11.\n");
+		return 1;
+	}
+	if (request_irq(IRQ_BUTTON, sw_interrupt, 0, "USL-5P BUTTON", NULL)) {
+		printk(KERN_ERR "Unable to get IRQ 12.\n");
+		return 1;
+	}
+	ctrl_outb(0x00, PA_PWRINT_CLR);
+
+	return 0;
+}
+
+module_init(swdrv_init);
+
+/*
+ * gio driver
+ *
+ */
+
+#include <asm/landisk/gio.h>
+
+static int gio_ioctl(struct inode *inode, struct file *filp,
+		     unsigned int cmd, unsigned long arg)
+{
+	int minor;
+	unsigned int data, mask;
+	static unsigned int addr = 0;
+
+	minor = (int)(filp->private_data);
+
+	/* access control */
+	if (minor == GIO_MINOR) {
+		;
+	} else if (minor == LED_MINOR) {
+		if (((cmd & 0x0ff) >= 9) && ((cmd & 0x0ff) < 20)) {
+			;
+		} else {
+			return -EINVAL;
+		}
+	} else if (minor == BTN_MINOR) {
+		if (((cmd & 0x0ff) >= 20) && ((cmd & 0x0ff) < 30)) {
+			;
+		} else {
+			return -EINVAL;
+		}
+	} else {
+		return -EINVAL;
+	}
+
+	if (cmd & 0x01) {	/* write */
+		if (copy_from_user(&data, (int *)arg, sizeof(int))) {
+			return -EFAULT;
+		}
+	}
+
+	switch (cmd) {
+	case GIODRV_IOCSGIOSETADDR:	/* addres set */
+		addr = data;
+		break;
+
+	case GIODRV_IOCSGIODATA1:	/* write byte */
+		ctrl_outb((unsigned char)(0x0ff & data), addr);
+		break;
+
+	case GIODRV_IOCSGIODATA2:	/* write word */
+		if (addr & 0x01) {
+			return -EFAULT;
+		}
+		ctrl_outw((unsigned short int)(0x0ffff & data), addr);
+		break;
+
+	case GIODRV_IOCSGIODATA4:	/* write long */
+		if (addr & 0x03) {
+			return -EFAULT;
+		}
+		ctrl_outl(data, addr);
+		break;
+
+	case GIODRV_IOCGGIODATA1:	/* read byte */
+		data = ctrl_inb(addr);
+		break;
+
+	case GIODRV_IOCGGIODATA2:	/* read word */
+		if (addr & 0x01) {
+			return -EFAULT;
+		}
+		data = ctrl_inw(addr);
+		break;
+
+	case GIODRV_IOCGGIODATA4:	/* read long */
+		if (addr & 0x03) {
+			return -EFAULT;
+		}
+		data = ctrl_inl(addr);
+		break;
+	case GIODRV_IOCSGIO_LED:	/* write */
+		mask = ((data & 0x00ffffff) << 8)
+		    | ((data & 0x0000ffff) << 16)
+		    | ((data & 0x000000ff) << 24);
+		landisk_ledparam = data & (~mask);
+		if (landisk_arch == 0) {	/* arch == landisk */
+			landisk_ledparam &= 0x03030303;
+			mask = (~(landisk_ledparam >> 22)) & 0x000c;
+			landisk_ledparam |= mask;
+		} else {	                /* arch == usl-5p */
+			mask = (landisk_ledparam >> 24) & 0x0001;
+			landisk_ledparam |= mask;
+			landisk_ledparam &= 0x007f7f7f;
+		}
+		landisk_ledparam |= 0x80;
+		break;
+	case GIODRV_IOCGGIO_LED:	/* read */
+		data = landisk_ledparam;
+		if (landisk_arch == 0) {	/* arch == landisk */
+			data &= 0x03030303;
+		} else {	                /* arch == usl-5p */
+			;
+		}
+		data &= (~0x080);
+		break;
+	case GIODRV_IOCSGIO_BUZZER:	/* write */
+		landisk_buzzerparam = data;
+		landisk_ledparam |= 0x80;
+		break;
+	case GIODRV_IOCGGIO_LANDISK:	/* read */
+		data = landisk_arch & 0x01;
+		break;
+	case GIODRV_IOCGGIO_BTN:	/* read */
+		data = (0x0ff & ctrl_inb(PA_PWRINT_CLR));
+		data <<= 8;
+		data |= (0x0ff & ctrl_inb(PA_IMASK));
+		data <<= 8;
+		data |= (0x0ff & landisk_btn);
+		data <<= 8;
+		data |= (0x0ff & (~ctrl_inb(PA_STATUS)));
+		break;
+	case GIODRV_IOCSGIO_BTNPID:	/* write */
+		landisk_btnctrlpid = data;
+		landisk_btn = 0;
+		if (irq_desc[IRQ_BUTTON].depth) {
+			enable_irq(IRQ_BUTTON);
+		}
+		if (irq_desc[IRQ_POWER].depth) {
+			enable_irq(IRQ_POWER);
+		}
+		break;
+	case GIODRV_IOCGGIO_BTNPID:	/* read */
+		data = landisk_btnctrlpid;
+		break;
+	default:
+		return -EFAULT;
+		break;
+	}
+
+	if ((cmd & 0x01) == 0) {	/* read */
+		if (copy_to_user((int *)arg, &data, sizeof(int))) {
+			return -EFAULT;
+		}
+	}
+	return 0;
+}
diff --git a/arch/sh/boards/landisk/rtc.c b/arch/sh/boards/landisk/rtc.c
new file mode 100644
index 000000000000..d666aa63303f
--- /dev/null
+++ b/arch/sh/boards/landisk/rtc.c
@@ -0,0 +1,102 @@
+/*
+ * arch/sh/boards/landisk/rtc.c --  RTC support
+ *
+ *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
+ *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
+ */
+/*
+ * modifed by kogiidena
+ * 2005.09.16
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/time.h>
+#include <linux/delay.h>
+#include <linux/spinlock.h>
+
+#ifndef BCD_TO_BIN
+#define BCD_TO_BIN(val) ((val)=((val)&15) + ((val)>>4)*10)
+#endif
+
+#ifndef BIN_TO_BCD
+#define BIN_TO_BCD(val) ((val)=(((val)/10)<<4) + (val)%10)
+#endif
+
+extern void (*rtc_get_time) (struct timespec *);
+extern int (*rtc_set_time) (const time_t);
+extern spinlock_t rtc_lock;
+
+extern void
+rs5c313_set_cmos_time(unsigned int BCD_yr, unsigned int BCD_mon,
+		      unsigned int BCD_day, unsigned int BCD_hr,
+		      unsigned int BCD_min, unsigned int BCD_sec);
+
+extern unsigned long
+rs5c313_get_cmos_time(unsigned int *BCD_yr, unsigned int *BCD_mon,
+		      unsigned int *BCD_day, unsigned int *BCD_hr,
+		      unsigned int *BCD_min, unsigned int *BCD_sec);
+
+void landisk_rtc_gettimeofday(struct timespec *tv)
+{
+	unsigned int BCD_yr, BCD_mon, BCD_day, BCD_hr, BCD_min, BCD_sec;
+	unsigned long flags;
+
+	spin_lock_irqsave(&rtc_lock, flags);
+	tv->tv_sec = rs5c313_get_cmos_time
+	    (&BCD_yr, &BCD_mon, &BCD_day, &BCD_hr, &BCD_min, &BCD_sec);
+	tv->tv_nsec = 0;
+	spin_unlock_irqrestore(&rtc_lock, flags);
+}
+
+int landisk_rtc_settimeofday(const time_t secs)
+{
+	int retval = 0;
+	int real_seconds, real_minutes, cmos_minutes;
+	unsigned long flags;
+	unsigned long nowtime = secs;
+	unsigned int BCD_yr, BCD_mon, BCD_day, BCD_hr, BCD_min, BCD_sec;
+
+	spin_lock_irqsave(&rtc_lock, flags);
+
+	rs5c313_get_cmos_time
+	  (&BCD_yr, &BCD_mon, &BCD_day, &BCD_hr, &BCD_min, &BCD_sec);
+	cmos_minutes = BCD_min;
+	BCD_TO_BIN(cmos_minutes);
+
+	/*
+	 * since we're only adjusting minutes and seconds,
+	 * don't interfere with hour overflow. This avoids
+	 * messing with unknown time zones but requires your
+	 * RTC not to be off by more than 15 minutes
+	 */
+	real_seconds = nowtime % 60;
+	real_minutes = nowtime / 60;
+	if (((abs(real_minutes - cmos_minutes) + 15) / 30) & 1)
+		real_minutes += 30;	/* correct for half hour time zone */
+	real_minutes %= 60;
+
+	if (abs(real_minutes - cmos_minutes) < 30) {
+		BIN_TO_BCD(real_seconds);
+		BIN_TO_BCD(real_minutes);
+		rs5c313_set_cmos_time(BCD_yr, BCD_mon, BCD_day, BCD_hr,
+				      real_minutes, real_seconds);
+	} else {
+		printk(KERN_WARNING
+		       "set_rtc_time: can't update from %d to %d\n",
+		       cmos_minutes, real_minutes);
+		retval = -1;
+	}
+
+	spin_unlock_irqrestore(&rtc_lock, flags);
+	return retval;
+}
+
+
+void landisk_time_init(void)
+{
+	rtc_get_time = landisk_rtc_gettimeofday;
+	rtc_set_time = landisk_rtc_settimeofday;
+}
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
new file mode 100644
index 000000000000..0c60eaa10ba7
--- /dev/null
+++ b/arch/sh/boards/landisk/setup.c
@@ -0,0 +1,218 @@
+/*
+ * arch/sh/boards/landisk/setup.c
+ *
+ * Copyright (C) 2002 Paul Mundt
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * Setup code for an unknown machine (internal peripherials only)
+ */
+/*
+ * linux/arch/sh/kernel/setup_landisk.c
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * I-O DATA Device, Inc. LANDISK Support.
+ *
+ * Modified for LANDISK by
+ * Atom Create Engineering Co., Ltd. 2002.
+ */
+/*
+ * modifed by kogiidena
+ * 2005.09.16
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/pm.h>
+
+#include <linux/hdreg.h>
+#include <linux/ide.h>
+#include <linux/pci.h>
+
+#include <asm/machvec.h>
+#include <asm/rtc.h>
+#include <asm/machvec_init.h>
+#include <asm/io.h>
+#include <asm/landisk/iodata_landisk.h>
+#include <asm/landisk/io.h>
+
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
+
+extern void (*board_time_init) (void);
+void landisk_time_init(void);
+extern void init_landisk_IRQ(void);
+
+int landisk_ledparam;
+int landisk_buzzerparam;
+int landisk_arch;
+
+/* defined in mm/ioremap.c */
+extern void *p3_ioremap(unsigned long phys_addr, unsigned long size,
+			unsigned long flags);
+
+/*
+ * Initialize the board
+ */
+
+const char *get_system_type(void)
+{
+	return "LANDISK";
+}
+
+static void landisk_power_off(void)
+{
+	ctrl_outb(0x01, PA_SHUTDOWN);
+}
+
+void check_usl5p(void)
+{
+	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
+	unsigned char tmp1, tmp2;
+	tmp1 = *p;
+	*p = 0x40;
+	tmp2 = *p;
+	*p = tmp1;
+	landisk_arch = (tmp2 == 0x40) ? 1 : 0;
+	if (landisk_arch == 1) {	/* arch == usl-5p */
+		landisk_ledparam = 0x00000380;
+		landisk_ledparam |= (tmp1 & 0x07c);
+	} else {                        /* arch == landisk */
+		landisk_ledparam = 0x02000180;
+		landisk_ledparam |= 0x04;
+	}
+	return;
+}
+
+void __init platform_setup(void)
+{
+
+	landisk_buzzerparam = 0;
+	check_usl5p();
+
+	printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
+	board_time_init = landisk_time_init;
+	pm_power_off = landisk_power_off;
+
+}
+
+void *area5_io_base;
+void *area6_io_base;
+
+int __init cf_init(void)
+{
+	pgprot_t prot;
+	unsigned long paddrbase, psize;
+
+	/* open I/O area window */
+	paddrbase = virt_to_phys((void *)PA_AREA5_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_IO16);
+	area5_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area5_io_base) {
+		printk("allocate_cf_area : can't open CF I/O window!\n");
+		return -ENOMEM;
+	}
+
+	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
+	area6_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area6_io_base) {
+		printk("allocate_cf_area : can't open HDD I/O window!\n");
+		return -ENOMEM;
+	}
+
+	printk(KERN_INFO "Allocate Area5/6 success.\n");
+
+	/* XXX : do we need attribute and common-memory area also? */
+
+	return 0;
+}
+
+__initcall(cf_init);
+
+#include <linux/sched.h>
+
+/* Cycle the LED's in the clasic knightrider/Sun pattern */
+
+void heartbeat_landisk(void)
+{
+	static unsigned int cnt = 0, blink = 0x00, period = 25;
+	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
+	char data;
+
+	if ((landisk_ledparam & 0x080) == 0) {
+		return;
+	}
+	cnt += 1;
+	if (cnt < period) {
+		return;
+	}
+	cnt = 0;
+	blink++;
+
+	data = (blink & 0x01) ? (landisk_ledparam >> 16) : 0;
+	data |= (blink & 0x02) ? (landisk_ledparam >> 8) : 0;
+	data |= landisk_ledparam;
+
+	/* buzzer */
+	if (landisk_buzzerparam & 0x1) {
+		data |= 0x80;
+	} else {
+		data &= 0x7f;
+	}
+	*p = data;
+
+	if (((landisk_ledparam & 0x007f7f00) == 0)
+	    && (landisk_buzzerparam == 0)) {
+		landisk_ledparam &= (~0x0080);
+	}
+	landisk_buzzerparam >>= 1;
+}
+
+/*
+ * The Machine Vector
+ */
+
+struct sh_machine_vector mv_landisk __initmv = {
+	.mv_nr_irqs = 72,
+	.mv_inb = landisk_inb,
+	.mv_inw = landisk_inw,
+	.mv_inl = landisk_inl,
+	.mv_outb = landisk_outb,
+	.mv_outw = landisk_outw,
+	.mv_outl = landisk_outl,
+	.mv_inb_p = landisk_inb_p,
+	.mv_inw_p = landisk_inw,
+	.mv_inl_p = landisk_inl,
+	.mv_outb_p = landisk_outb_p,
+	.mv_outw_p = landisk_outw,
+	.mv_outl_p = landisk_outl,
+	.mv_insb = landisk_insb,
+	.mv_insw = landisk_insw,
+	.mv_insl = landisk_insl,
+	.mv_outsb = landisk_outsb,
+	.mv_outsw = landisk_outsw,
+	.mv_outsl = landisk_outsl,
+	.mv_readb = landisk_readb,
+	.mv_readw = landisk_readw,
+	.mv_readl = landisk_readl,
+	.mv_writeb = landisk_writeb,
+	.mv_writew = landisk_writew,
+	.mv_writel = landisk_writel,
+	.mv_ioremap = landisk_ioremap,
+	.mv_iounmap = landisk_iounmap,
+	.mv_isa_port2addr = landisk_isa_port2addr,
+	.mv_init_irq = init_landisk_IRQ,
+
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat = heartbeat_landisk,
+#endif
+
+};
+
+ALIAS_MV(landisk)
diff --git a/arch/sh/configs/landisk_defconfig b/arch/sh/configs/landisk_defconfig
new file mode 100644
index 000000000000..6b43316d03cf
--- /dev/null
+++ b/arch/sh/configs/landisk_defconfig
@@ -0,0 +1,1373 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.13-sh
+# Sun Sep 11 13:00:46 2005
+#
+CONFIG_SUPERH=y
+CONFIG_UID16=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_IOMAP=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+# CONFIG_IKCONFIG is not set
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_EMBEDDED is not set
+CONFIG_KALLSYMS=y
+CONFIG_KALLSYMS_EXTRA_PASS=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_EPOLL=y
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+CONFIG_SH_LANDISK=y
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+CONFIG_CPU_SUBTYPE_SH7751=y
+CONFIG_CPU_SUBTYPE_SH7751R=y
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x04000000
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_SH_FPU=y
+# CONFIG_SH_STORE_QUEUES is not set
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ_BOOL=y
+CONFIG_SH_PCLK_FREQ=33333333
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=4
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+CONFIG_HEARTBEAT=y
+
+#
+# Kernel features
+#
+CONFIG_KEXEC=y
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+# CONFIG_CMDLINE_BOOL is not set
+
+#
+# Bus options
+#
+CONFIG_ISA=y
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+CONFIG_PCI_LEGACY_PROC=y
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+CONFIG_PCCARD=y
+# CONFIG_PCMCIA_DEBUG is not set
+CONFIG_PCMCIA=y
+CONFIG_PCMCIA_LOAD_CIS=y
+CONFIG_PCMCIA_IOCTL=y
+CONFIG_CARDBUS=y
+
+#
+# PC-card bridges
+#
+CONFIG_YENTA=y
+# CONFIG_PD6729 is not set
+# CONFIG_I82092 is not set
+# CONFIG_I82365 is not set
+# CONFIG_TCIC is not set
+CONFIG_PCMCIA_PROBE=y
+CONFIG_PCCARD_NONSTATIC=y
+
+#
+# PCI Hotplug Support
+#
+# CONFIG_HOTPLUG_PCI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_MULTIPLE_TABLES is not set
+# CONFIG_IP_ROUTE_MULTIPATH is not set
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+
+#
+# IP: Virtual Server Configuration
+#
+# CONFIG_IP_VS is not set
+# CONFIG_IPV6 is not set
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+# CONFIG_NETFILTER_NETLINK is not set
+
+#
+# IP: Netfilter Configuration
+#
+CONFIG_IP_NF_CONNTRACK=m
+CONFIG_IP_NF_CT_ACCT=y
+CONFIG_IP_NF_CONNTRACK_MARK=y
+# CONFIG_IP_NF_CONNTRACK_EVENTS is not set
+# CONFIG_IP_NF_CT_PROTO_SCTP is not set
+CONFIG_IP_NF_FTP=m
+CONFIG_IP_NF_IRC=m
+# CONFIG_IP_NF_NETBIOS_NS is not set
+CONFIG_IP_NF_TFTP=m
+CONFIG_IP_NF_AMANDA=m
+# CONFIG_IP_NF_PPTP is not set
+CONFIG_IP_NF_QUEUE=m
+CONFIG_IP_NF_IPTABLES=m
+CONFIG_IP_NF_MATCH_LIMIT=m
+CONFIG_IP_NF_MATCH_IPRANGE=m
+CONFIG_IP_NF_MATCH_MAC=m
+CONFIG_IP_NF_MATCH_PKTTYPE=m
+CONFIG_IP_NF_MATCH_MARK=m
+CONFIG_IP_NF_MATCH_MULTIPORT=m
+CONFIG_IP_NF_MATCH_TOS=m
+CONFIG_IP_NF_MATCH_RECENT=m
+CONFIG_IP_NF_MATCH_ECN=m
+CONFIG_IP_NF_MATCH_DSCP=m
+CONFIG_IP_NF_MATCH_AH_ESP=m
+CONFIG_IP_NF_MATCH_LENGTH=m
+CONFIG_IP_NF_MATCH_TTL=m
+CONFIG_IP_NF_MATCH_TCPMSS=m
+CONFIG_IP_NF_MATCH_HELPER=m
+CONFIG_IP_NF_MATCH_STATE=m
+CONFIG_IP_NF_MATCH_CONNTRACK=m
+CONFIG_IP_NF_MATCH_OWNER=m
+CONFIG_IP_NF_MATCH_ADDRTYPE=m
+CONFIG_IP_NF_MATCH_REALM=m
+CONFIG_IP_NF_MATCH_SCTP=m
+# CONFIG_IP_NF_MATCH_DCCP is not set
+CONFIG_IP_NF_MATCH_COMMENT=m
+CONFIG_IP_NF_MATCH_CONNMARK=m
+# CONFIG_IP_NF_MATCH_CONNBYTES is not set
+CONFIG_IP_NF_MATCH_HASHLIMIT=m
+# CONFIG_IP_NF_MATCH_STRING is not set
+CONFIG_IP_NF_FILTER=m
+CONFIG_IP_NF_TARGET_REJECT=m
+CONFIG_IP_NF_TARGET_LOG=m
+CONFIG_IP_NF_TARGET_ULOG=m
+CONFIG_IP_NF_TARGET_TCPMSS=m
+# CONFIG_IP_NF_TARGET_NFQUEUE is not set
+CONFIG_IP_NF_NAT=m
+CONFIG_IP_NF_NAT_NEEDED=y
+CONFIG_IP_NF_TARGET_MASQUERADE=m
+CONFIG_IP_NF_TARGET_REDIRECT=m
+CONFIG_IP_NF_TARGET_NETMAP=m
+CONFIG_IP_NF_TARGET_SAME=m
+# CONFIG_IP_NF_NAT_SNMP_BASIC is not set
+CONFIG_IP_NF_NAT_IRC=m
+CONFIG_IP_NF_NAT_FTP=m
+CONFIG_IP_NF_NAT_TFTP=m
+CONFIG_IP_NF_NAT_AMANDA=m
+CONFIG_IP_NF_MANGLE=m
+CONFIG_IP_NF_TARGET_TOS=m
+CONFIG_IP_NF_TARGET_ECN=m
+CONFIG_IP_NF_TARGET_DSCP=m
+CONFIG_IP_NF_TARGET_MARK=m
+CONFIG_IP_NF_TARGET_CLASSIFY=m
+# CONFIG_IP_NF_TARGET_TTL is not set
+CONFIG_IP_NF_TARGET_CONNMARK=m
+# CONFIG_IP_NF_TARGET_CLUSTERIP is not set
+CONFIG_IP_NF_RAW=m
+CONFIG_IP_NF_TARGET_NOTRACK=m
+CONFIG_IP_NF_ARPTABLES=m
+CONFIG_IP_NF_ARPFILTER=m
+CONFIG_IP_NF_ARP_MANGLE=m
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+CONFIG_LLC=m
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+CONFIG_ATALK=m
+# CONFIG_DEV_APPLETALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+CONFIG_NET_CLS_ROUTE=y
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+# CONFIG_PNP is not set
+
+#
+# Block devices
+#
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_LBD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+CONFIG_IDE=y
+CONFIG_IDE_MAX_HWIFS=4
+CONFIG_BLK_DEV_IDE=y
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+# CONFIG_BLK_DEV_IDE_SATA is not set
+CONFIG_BLK_DEV_IDEDISK=y
+# CONFIG_IDEDISK_MULTI_MODE is not set
+# CONFIG_BLK_DEV_IDECS is not set
+CONFIG_BLK_DEV_IDECD=y
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+CONFIG_BLK_DEV_IDESCSI=y
+# CONFIG_IDE_TASK_IOCTL is not set
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=y
+CONFIG_BLK_DEV_IDEPCI=y
+CONFIG_IDEPCI_SHARE_IRQ=y
+CONFIG_BLK_DEV_OFFBOARD=y
+CONFIG_BLK_DEV_GENERIC=y
+# CONFIG_BLK_DEV_OPTI621 is not set
+CONFIG_BLK_DEV_IDEDMA_PCI=y
+# CONFIG_BLK_DEV_IDEDMA_FORCED is not set
+CONFIG_IDEDMA_PCI_AUTO=y
+CONFIG_IDEDMA_ONLYDISK=y
+CONFIG_BLK_DEV_AEC62XX=y
+# CONFIG_BLK_DEV_ALI15X3 is not set
+# CONFIG_BLK_DEV_AMD74XX is not set
+# CONFIG_BLK_DEV_CMD64X is not set
+# CONFIG_BLK_DEV_TRIFLEX is not set
+# CONFIG_BLK_DEV_CY82C693 is not set
+# CONFIG_BLK_DEV_CS5520 is not set
+# CONFIG_BLK_DEV_CS5530 is not set
+# CONFIG_BLK_DEV_HPT34X is not set
+# CONFIG_BLK_DEV_HPT366 is not set
+# CONFIG_BLK_DEV_SC1200 is not set
+# CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
+# CONFIG_BLK_DEV_NS87415 is not set
+# CONFIG_BLK_DEV_PDC202XX_OLD is not set
+# CONFIG_BLK_DEV_PDC202XX_NEW is not set
+# CONFIG_BLK_DEV_SVWKS is not set
+# CONFIG_BLK_DEV_SIIMAGE is not set
+# CONFIG_BLK_DEV_SLC90E66 is not set
+# CONFIG_BLK_DEV_TRM290 is not set
+# CONFIG_BLK_DEV_VIA82CXXX is not set
+CONFIG_IDE_SH=y
+# CONFIG_IDE_ARM is not set
+# CONFIG_IDE_CHIPSETS is not set
+CONFIG_BLK_DEV_IDEDMA=y
+# CONFIG_IDEDMA_IVB is not set
+CONFIG_IDEDMA_AUTO=y
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=y
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=y
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+# CONFIG_BLK_DEV_SR is not set
+# CONFIG_CHR_DEV_SG is not set
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+CONFIG_SCSI_MULTI_LUN=y
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+
+#
+# SCSI Transport Attributes
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_ATTRS is not set
+
+#
+# SCSI low-level drivers
+#
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AHA152X is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_SCSI_IN2000 is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_DTC3280 is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_GENERIC_NCR5380 is not set
+# CONFIG_SCSI_GENERIC_NCR5380_MMIO is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_NCR53C406A is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_PAS16 is not set
+# CONFIG_SCSI_PSI240I is not set
+# CONFIG_SCSI_QLOGIC_FAS is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+CONFIG_SCSI_QLA2XXX=y
+# CONFIG_SCSI_QLA21XX is not set
+# CONFIG_SCSI_QLA22XX is not set
+# CONFIG_SCSI_QLA2300 is not set
+# CONFIG_SCSI_QLA2322 is not set
+# CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_SYM53C416 is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_T128 is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+
+#
+# PCMCIA SCSI adapter support
+#
+# CONFIG_PCMCIA_AHA152X is not set
+# CONFIG_PCMCIA_FDOMAIN is not set
+# CONFIG_PCMCIA_NINJA_SCSI is not set
+# CONFIG_PCMCIA_QLOGIC is not set
+# CONFIG_PCMCIA_SYM53C500 is not set
+
+#
+# Old CD-ROM drivers (not SCSI, not IDE)
+#
+# CONFIG_CD_NO_IDESCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+CONFIG_MD=y
+CONFIG_BLK_DEV_MD=m
+CONFIG_MD_LINEAR=m
+CONFIG_MD_RAID0=m
+CONFIG_MD_RAID1=m
+# CONFIG_MD_RAID10 is not set
+# CONFIG_MD_RAID5 is not set
+# CONFIG_MD_RAID6 is not set
+# CONFIG_MD_MULTIPATH is not set
+# CONFIG_MD_FAULTY is not set
+# CONFIG_BLK_DEV_DM is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
+# CONFIG_FUSION_SAS is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_CASSINI is not set
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_NET_VENDOR_SMC is not set
+# CONFIG_SMC91X is not set
+# CONFIG_NET_VENDOR_RACAL is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_AT1700 is not set
+# CONFIG_DEPCA is not set
+# CONFIG_HP100 is not set
+# CONFIG_NET_ISA is not set
+# CONFIG_NE2000 is not set
+CONFIG_NET_PCI=y
+# CONFIG_PCNET32 is not set
+# CONFIG_AMD8111_ETH is not set
+# CONFIG_ADAPTEC_STARFIRE is not set
+# CONFIG_AC3200 is not set
+# CONFIG_APRICOT is not set
+# CONFIG_B44 is not set
+# CONFIG_FORCEDETH is not set
+# CONFIG_CS89x0 is not set
+# CONFIG_DGRS is not set
+# CONFIG_EEPRO100 is not set
+# CONFIG_E100 is not set
+# CONFIG_FEALNX is not set
+# CONFIG_NATSEMI is not set
+# CONFIG_NE2K_PCI is not set
+CONFIG_8139CP=y
+# CONFIG_8139TOO is not set
+# CONFIG_SIS900 is not set
+# CONFIG_EPIC100 is not set
+# CONFIG_SUNDANCE is not set
+# CONFIG_TLAN is not set
+# CONFIG_VIA_RHINE is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+# CONFIG_R8169 is not set
+# CONFIG_SIS190 is not set
+# CONFIG_SKGE is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# PCMCIA network device support
+#
+# CONFIG_NET_PCMCIA is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+CONFIG_RS5C313_RTC=y
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+
+#
+# PCMCIA character devices
+#
+# CONFIG_SYNCLINK_CS is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia Capabilities Port drivers
+#
+
+#
+# Multimedia devices
+#
+CONFIG_VIDEO_DEV=m
+
+#
+# Video For Linux
+#
+
+#
+# Video Adapters
+#
+# CONFIG_VIDEO_PMS is not set
+# CONFIG_VIDEO_CPIA is not set
+# CONFIG_VIDEO_STRADIS is not set
+# CONFIG_VIDEO_MXB is not set
+# CONFIG_VIDEO_DPC is not set
+# CONFIG_VIDEO_HEXIUM_ORION is not set
+# CONFIG_VIDEO_HEXIUM_GEMINI is not set
+
+#
+# Radio Adapters
+#
+# CONFIG_RADIO_CADET is not set
+# CONFIG_RADIO_RTRACK is not set
+# CONFIG_RADIO_RTRACK2 is not set
+# CONFIG_RADIO_AZTECH is not set
+# CONFIG_RADIO_GEMTEK is not set
+# CONFIG_RADIO_GEMTEK_PCI is not set
+# CONFIG_RADIO_MAXIRADIO is not set
+# CONFIG_RADIO_MAESTRO is not set
+# CONFIG_RADIO_SF16FMI is not set
+# CONFIG_RADIO_SF16FMR2 is not set
+# CONFIG_RADIO_TERRATEC is not set
+# CONFIG_RADIO_TRUST is not set
+# CONFIG_RADIO_TYPHOON is not set
+# CONFIG_RADIO_ZOLTRIX is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Console display driver support
+#
+# CONFIG_MDA_CONSOLE is not set
+CONFIG_DUMMY_CONSOLE=y
+CONFIG_FONT_8x16=y
+
+#
+# Sound
+#
+CONFIG_SOUND=m
+
+#
+# Advanced Linux Sound Architecture
+#
+# CONFIG_SND is not set
+
+#
+# Open Sound System
+#
+CONFIG_SOUND_PRIME=m
+# CONFIG_SOUND_BT878 is not set
+# CONFIG_SOUND_CMPCI is not set
+# CONFIG_SOUND_EMU10K1 is not set
+# CONFIG_SOUND_FUSION is not set
+# CONFIG_SOUND_CS4281 is not set
+# CONFIG_SOUND_ES1370 is not set
+# CONFIG_SOUND_ES1371 is not set
+# CONFIG_SOUND_ESSSOLO1 is not set
+# CONFIG_SOUND_MAESTRO is not set
+# CONFIG_SOUND_MAESTRO3 is not set
+# CONFIG_SOUND_ICH is not set
+# CONFIG_SOUND_SONICVIBES is not set
+# CONFIG_SOUND_TRIDENT is not set
+# CONFIG_SOUND_MSNDCLAS is not set
+# CONFIG_SOUND_MSNDPIN is not set
+# CONFIG_SOUND_VIA82CXXX is not set
+# CONFIG_SOUND_ALI5455 is not set
+# CONFIG_SOUND_FORTE is not set
+# CONFIG_SOUND_RME96XX is not set
+# CONFIG_SOUND_AD1980 is not set
+
+#
+# USB support
+#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_BANDWIDTH is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+CONFIG_USB_EHCI_HCD=y
+# CONFIG_USB_EHCI_SPLIT_ISO is not set
+# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
+CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
+# CONFIG_USB_UHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+
+#
+# USB Device Class drivers
+#
+CONFIG_OBSOLETE_OSS_USB_DRIVER=y
+CONFIG_USB_AUDIO=m
+# CONFIG_USB_BLUETOOTH_TTY is not set
+CONFIG_USB_MIDI=m
+# CONFIG_USB_ACM is not set
+CONFIG_USB_PRINTER=m
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=m
+# CONFIG_USB_STORAGE_DEBUG is not set
+CONFIG_USB_STORAGE_DATAFAB=y
+CONFIG_USB_STORAGE_FREECOM=y
+CONFIG_USB_STORAGE_ISD200=y
+CONFIG_USB_STORAGE_DPCM=y
+# CONFIG_USB_STORAGE_USBAT is not set
+CONFIG_USB_STORAGE_SDDR09=y
+CONFIG_USB_STORAGE_SDDR55=y
+CONFIG_USB_STORAGE_JUMPSHOT=y
+
+#
+# USB Input Devices
+#
+CONFIG_USB_HID=m
+CONFIG_USB_HIDINPUT=y
+# CONFIG_HID_FF is not set
+# CONFIG_USB_HIDDEV is not set
+
+#
+# USB HID Boot Protocol drivers
+#
+# CONFIG_USB_KBD is not set
+# CONFIG_USB_MOUSE is not set
+# CONFIG_USB_AIPTEK is not set
+# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
+# CONFIG_USB_KBTAB is not set
+# CONFIG_USB_POWERMATE is not set
+# CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
+# CONFIG_USB_EGALAX is not set
+# CONFIG_USB_YEALINK is not set
+# CONFIG_USB_XPAD is not set
+# CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_KEYSPAN_REMOTE is not set
+# CONFIG_USB_APPLETOUCH is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+
+#
+# USB Multimedia devices
+#
+CONFIG_USB_DABUSB=m
+CONFIG_USB_VICAM=m
+CONFIG_USB_DSBR=m
+CONFIG_USB_IBMCAM=m
+CONFIG_USB_KONICAWC=m
+CONFIG_USB_OV511=m
+CONFIG_USB_SE401=m
+CONFIG_USB_SN9C102=m
+CONFIG_USB_STV680=m
+CONFIG_USB_PWC=m
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+CONFIG_USB_PEGASUS=m
+CONFIG_USB_RTL8150=m
+# CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+CONFIG_USB_SERIAL=m
+# CONFIG_USB_SERIAL_GENERIC is not set
+# CONFIG_USB_SERIAL_AIRPRIME is not set
+# CONFIG_USB_SERIAL_BELKIN is not set
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
+# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
+# CONFIG_USB_SERIAL_CP2101 is not set
+# CONFIG_USB_SERIAL_CYPRESS_M8 is not set
+# CONFIG_USB_SERIAL_EMPEG is not set
+CONFIG_USB_SERIAL_FTDI_SIO=m
+# CONFIG_USB_SERIAL_VISOR is not set
+# CONFIG_USB_SERIAL_IPAQ is not set
+# CONFIG_USB_SERIAL_IR is not set
+# CONFIG_USB_SERIAL_EDGEPORT is not set
+# CONFIG_USB_SERIAL_EDGEPORT_TI is not set
+# CONFIG_USB_SERIAL_GARMIN is not set
+# CONFIG_USB_SERIAL_IPW is not set
+# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
+# CONFIG_USB_SERIAL_KEYSPAN is not set
+# CONFIG_USB_SERIAL_KLSI is not set
+# CONFIG_USB_SERIAL_KOBIL_SCT is not set
+# CONFIG_USB_SERIAL_MCT_U232 is not set
+CONFIG_USB_SERIAL_PL2303=m
+# CONFIG_USB_SERIAL_HP4X is not set
+# CONFIG_USB_SERIAL_SAFE is not set
+# CONFIG_USB_SERIAL_TI is not set
+# CONFIG_USB_SERIAL_CYBERJACK is not set
+# CONFIG_USB_SERIAL_XIRCOM is not set
+# CONFIG_USB_SERIAL_OPTION is not set
+# CONFIG_USB_SERIAL_OMNINET is not set
+
+#
+# USB Miscellaneous drivers
+#
+CONFIG_USB_EMI62=m
+CONFIG_USB_EMI26=m
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGETKIT is not set
+# CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+CONFIG_USB_SISUSBVGA=m
+CONFIG_USB_SISUSBVGA_CON=y
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TEST is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+CONFIG_REISERFS_FS=y
+# CONFIG_REISERFS_CHECK is not set
+# CONFIG_REISERFS_PROC_INFO is not set
+# CONFIG_REISERFS_FS_XATTR is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+CONFIG_ROMFS_FS=y
+CONFIG_INOTIFY=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=m
+# CONFIG_JOLIET is not set
+# CONFIG_ZISOFS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+CONFIG_NTFS_FS=m
+# CONFIG_NTFS_DEBUG is not set
+CONFIG_NTFS_RW=y
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+# CONFIG_RELAYFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+CONFIG_UFS_FS=m
+CONFIG_UFS_FS_WRITE=y
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=m
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=m
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_NFSD_TCP=y
+CONFIG_LOCKD=m
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=m
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=m
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+CONFIG_SMB_FS=m
+# CONFIG_SMB_NLS_DEFAULT is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=y
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+CONFIG_NLS_CODEPAGE_932=y
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+# CONFIG_NLS_ISO8859_1 is not set
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+# CONFIG_NLS_UTF8 is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_FRAME_POINTER is not set
+CONFIG_SH_STANDARD_BIOS=y
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_EARLY_PRINTK is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 182fe9092577..c503ca072e11 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -25,4 +25,4 @@ RTS7751R2D		SH_RTS7751R2D
 EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
-
+LANDISK			SH_LANDISK
diff --git a/include/asm-sh/landisk/gio.h b/include/asm-sh/landisk/gio.h
new file mode 100644
index 000000000000..3fce4c451a46
--- /dev/null
+++ b/include/asm-sh/landisk/gio.h
@@ -0,0 +1,45 @@
+#ifndef __ASM_SH_LANDISK_GIO_H
+#define __ASM_SH_LANDISK_GIO_H
+
+#include <linux/ioctl.h>
+
+/* version */
+#define VERSION_STR	"1.00"
+
+/* Driver name */
+#define GIO_DRIVER_NAME		"/dev/giodrv"
+
+/* Use 'k' as magic number */
+#define GIODRV_IOC_MAGIC  'k'
+
+#define GIODRV_IOCRESET    _IO(GIODRV_IOC_MAGIC, 0)
+/*
+ * S means "Set" through a ptr,
+ * T means "Tell" directly
+ * G means "Get" (to a pointed var)
+ * Q means "Query", response is on the return value
+ * X means "eXchange": G and S atomically
+ * H means "sHift": T and Q atomically
+ */
+#define GIODRV_IOCSGIODATA1   _IOW(GIODRV_IOC_MAGIC,  1, unsigned char *)
+#define GIODRV_IOCGGIODATA1   _IOR(GIODRV_IOC_MAGIC,  2, unsigned char *)
+#define GIODRV_IOCSGIODATA2   _IOW(GIODRV_IOC_MAGIC,  3, unsigned short *)
+#define GIODRV_IOCGGIODATA2   _IOR(GIODRV_IOC_MAGIC,  4, unsigned short *)
+#define GIODRV_IOCSGIODATA4   _IOW(GIODRV_IOC_MAGIC,  5, unsigned long *)
+#define GIODRV_IOCGGIODATA4   _IOR(GIODRV_IOC_MAGIC,  6, unsigned long *)
+#define GIODRV_IOCSGIOSETADDR _IOW(GIODRV_IOC_MAGIC,  7, unsigned long *)
+#define GIODRV_IOCHARDRESET   _IO(GIODRV_IOC_MAGIC, 8) /* debugging tool */
+
+#define GIODRV_IOCSGIO_LED    _IOW(GIODRV_IOC_MAGIC,  9, unsigned long *)
+#define GIODRV_IOCGGIO_LED    _IOR(GIODRV_IOC_MAGIC,  10, unsigned long *)
+#define GIODRV_IOCSGIO_BUZZER _IOW(GIODRV_IOC_MAGIC,  11, unsigned long *)
+#define GIODRV_IOCGGIO_LANDISK _IOR(GIODRV_IOC_MAGIC,  14, unsigned long *)
+#define GIODRV_IOCGGIO_BTN _IOR(GIODRV_IOC_MAGIC,  22, unsigned long *)
+#define GIODRV_IOCSGIO_BTNPID _IOW(GIODRV_IOC_MAGIC,  23, unsigned long *)
+#define GIODRV_IOCGGIO_BTNPID _IOR(GIODRV_IOC_MAGIC,  24, unsigned long *)
+
+#define GIODRV_IOC_MAXNR 8
+#define GIO_READ 0x00000000
+#define GIO_WRITE 0x00000001
+
+#endif /* __ASM_SH_LANDISK_GIO_H  */
diff --git a/include/asm-sh/landisk/ide.h b/include/asm-sh/landisk/ide.h
new file mode 100644
index 000000000000..6490e28415ed
--- /dev/null
+++ b/include/asm-sh/landisk/ide.h
@@ -0,0 +1,14 @@
+/*
+ * modifed by kogiidena
+ * 2005.03.03
+ */
+
+#ifndef __ASM_SH_LANDISK_IDE_H
+#define __ASM_SH_LANDISK_IDE_H
+
+/* Nothing to see here.. */
+#include <asm/landisk/iodata_landisk.h>
+#define IRQ_CFCARD	IRQ_FATA	/* CF Card IRQ */
+#define IRQ_PCMCIA	IRQ_ATA		/* PCMCIA IRQ */
+
+#endif /* __ASM_SH_LANDISK_IDE_H  */
diff --git a/include/asm-sh/landisk/iodata_landisk.h b/include/asm-sh/landisk/iodata_landisk.h
new file mode 100644
index 000000000000..7189d3a36384
--- /dev/null
+++ b/include/asm-sh/landisk/iodata_landisk.h
@@ -0,0 +1,78 @@
+#ifndef __ASM_SH_IODATA_LANDISK_H
+#define __ASM_SH_IODATA_LANDISK_H
+
+/*
+ * linux/include/asm-sh/landisk/iodata_landisk.h
+ *
+ * Copyright (C) 2000  Atom Create Engineering Co., Ltd.
+ *
+ * IO-DATA LANDISK support
+ */
+
+/* Box specific addresses.  */
+
+#define PA_USB		0xa4000000	/* USB Controller M66590 */
+
+#define PA_ATARST	0xb0000000	/* ATA/FATA Access Control Register */
+#define PA_LED		0xb0000001	/* LED Control Register */
+#define PA_STATUS	0xb0000002	/* Switch Status Register */
+#define PA_SHUTDOWN	0xb0000003	/* Shutdown Control Register */
+#define PA_PCIPME	0xb0000004	/* PCI PME Status Register */
+#define PA_IMASK	0xb0000005	/* Interrupt Mask Register */
+/* 2003.10.31 I-O DATA NSD NWG	add.	for shutdown port clear */
+#define PA_PWRINT_CLR	0xb0000006	/* Shutdown Interrupt clear Register */
+
+#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
+#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
+#define PA_LCD_CLRDSP	0x00		/* LCD Clear Display Offset */
+#define PA_LCD_RTNHOME	0x00		/* LCD Return Home Offset */
+#define PA_LCD_ENTMODE	0x00		/* LCD Entry Mode Offset */
+#define PA_LCD_DSPCTL	0x00		/* LCD Display ON/OFF Control Offset */
+#define PA_LCD_FUNC	0x00		/* LCD Function Set Offset */
+#define PA_LCD_CGRAM	0x00		/* LCD Set CGRAM Address Offset */
+#define PA_LCD_DDRAM	0x00		/* LCD Set DDRAM Address Offset */
+#define PA_LCD_RDFLAG	0x01		/* LCD Read Busy Flag Offset */
+#define PA_LCD_WTDATA	0x02		/* LCD Write Datat to RAM Offset */
+#define PA_LCD_RDDATA	0x03		/* LCD Read Data from RAM Offset */
+#define PA_PIDE_OFFSET	0x40		/* CF IDE Offset */
+#define PA_SIDE_OFFSET	0x40		/* HDD IDE Offset */
+
+#define IRQ_PCIINTA	5		/* PCI INTA IRQ */
+#define IRQ_PCIINTB	6		/* PCI INTB IRQ */
+#define IRQ_PCIINDC	7		/* PCI INTC IRQ */
+#define IRQ_PCIINTD	8		/* PCI INTD IRQ */
+#define IRQ_ATA		9		/* ATA IRQ */
+#define IRQ_FATA	10		/* FATA IRQ */
+#define IRQ_POWER	11		/* Power Switch IRQ */
+#define IRQ_BUTTON	12		/* USL-5P Button IRQ */
+#define IRQ_FAULT	13		/* USL-5P Fault  IRQ */
+
+#define SHUTDOWN_BTN_MAJOR	99	/* Shutdown button device major no. */
+
+#define SHUTDOWN_LOOP_CNT	5	/* Shutdown button Detection loop */
+#define SHUTDOWN_DELAY		200	/* Shutdown button delay value(ms) */
+
+
+/* added by kogiidena */
+/*
+ *  landisk_ledparam
+ *
+ * led  ------10 -6543210 -6543210 -6543210
+ *     |000000..|0.......|0.......|U.......|
+ *     |  HARD  |fastblik| blink  |   on   |
+ *
+ *   led0: power       U:update flag
+ *   led1: error
+ *   led2: usb1
+ *   led3: usb2
+ *   led4: usb3
+ *   led5: usb4
+ *   led6: usb5
+ *
+ */
+extern int landisk_ledparam;    /* from setup.c */
+extern int landisk_buzzerparam; /* from setup.c */
+extern int landisk_arch;        /* from setup.c */
+
+#endif  /* __ASM_SH_IODATA_LANDISK_H */
+
-- 
cgit v1.2.3


From 298476220d1f793ca0ac6c9e5dc817e1ad3e9851 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:57:44 +0900
Subject: sh: Add control register barriers.

Currently when making changes to control registers, we
typically need some time for changes to take effect (8
nops, generally).  However, for sh4a we simply need to
do an icbi..

This is a simple patch for implementing a general purpose
ctrl_barrier() which functions as a control register write
barrier. There's some additional documentation in the patch
itself, but it's pretty self explanatory.

There were also some places where we were not doing the
barrier, which didn't seem to have any adverse effects on
legacy parts, but certainly did on sh4a. It's safer to have
the barrier in place for legacy parts as well in these cases,
though this does make flush_tlb_all() more expensive (by an
order of 8 nops).  We can ifdef around the flush_tlb_all()
case for now if it's clear that all legacy parts won't have
a problem with this.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c       |  5 +++++
 arch/sh/mm/fault.c           |  5 +++--
 include/asm-sh/mmu_context.h |  7 +++----
 include/asm-sh/system.h      | 43 ++++++++++++++++++++++++++++++++++---------
 4 files changed, 45 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index c036c2b4ac2b..2203bd6aadb3 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -269,6 +269,11 @@ static inline void flush_icache_all(void)
 	ccr |= CCR_CACHE_ICI;
 	ctrl_outl(ccr, CCR);
 
+	/*
+	 * back_to_P1() will take care of the barrier for us, don't add
+	 * another one!
+	 */
+
 	back_to_P1();
 	local_irq_restore(flags);
 }
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 775f86cd3fe8..364181f27b79 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -369,12 +369,13 @@ void flush_tlb_all(void)
 	 * Flush all the TLB.
 	 *
 	 * Write to the MMU control register's bit:
-	 * 	TF-bit for SH-3, TI-bit for SH-4.
+	 *	TF-bit for SH-3, TI-bit for SH-4.
 	 *      It's same position, bit #2.
 	 */
 	local_irq_save(flags);
 	status = ctrl_inl(MMUCR);
-	status |= 0x04;		
+	status |= 0x04;
 	ctrl_outl(status, MMUCR);
+	ctrl_barrier();
 	local_irq_restore(flags);
 }
diff --git a/include/asm-sh/mmu_context.h b/include/asm-sh/mmu_context.h
index 6760d064bd02..87678ba8d6b6 100644
--- a/include/asm-sh/mmu_context.h
+++ b/include/asm-sh/mmu_context.h
@@ -174,9 +174,7 @@ static inline void enable_mmu(void)
 {
 	/* Enable MMU */
 	ctrl_outl(MMU_CONTROL_INIT, MMUCR);
-
-	/* The manual suggests doing some nops after turning on the MMU */
-	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop\n\t");
+	ctrl_barrier();
 
 	if (mmu_context_cache == NO_CONTEXT)
 		mmu_context_cache = MMU_CONTEXT_FIRST_VERSION;
@@ -191,7 +189,8 @@ static inline void disable_mmu(void)
 	cr = ctrl_inl(MMUCR);
 	cr &= ~MMU_CONTROL_INIT;
 	ctrl_outl(cr, MMUCR);
-	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop\n\t");
+
+	ctrl_barrier();
 }
 #else
 /*
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index eb4902ed920a..1630a5411e5f 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -67,8 +67,17 @@ static inline void sched_cacheflush(void)
 {
 }
 
-#define nop() __asm__ __volatile__ ("nop")
-
+#ifdef CONFIG_CPU_SH4A
+#define __icbi()			\
+{					\
+	unsigned long __addr;		\
+	__addr = 0xa8000000;		\
+	__asm__ __volatile__(		\
+		"icbi   %0\n\t"		\
+		: /* no output */	\
+		: "m" (__m(__addr)));	\
+}
+#endif
 
 #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
 
@@ -84,15 +93,31 @@ static __inline__ unsigned long tas(volatile int *m)
 
 extern void __xchg_called_with_bad_pointer(void);
 
+/*
+ * A brief note on ctrl_barrier(), the control register write barrier.
+ *
+ * Legacy SH cores typically require a sequence of 8 nops after
+ * modification of a control register in order for the changes to take
+ * effect. On newer cores (like the sh4a and sh5) this is accomplished
+ * with icbi.
+ *
+ * Also note that on sh4a in the icbi case we can forego a synco for the
+ * write barrier, as it's not necessary for control registers.
+ *
+ * Historically we have only done this type of barrier for the MMUCR, but
+ * it's also necessary for the CCR, so we make it generic here instead.
+ */
 #ifdef CONFIG_CPU_SH4A
-#define mb()	__asm__ __volatile__ ("synco": : :"memory")
-#define rmb()	mb()
-#define wmb()	__asm__ __volatile__ ("synco": : :"memory")
+#define mb()		__asm__ __volatile__ ("synco": : :"memory")
+#define rmb()		mb()
+#define wmb()		__asm__ __volatile__ ("synco": : :"memory")
+#define ctrl_barrier()	__icbi()
 #define read_barrier_depends()	do { } while(0)
 #else
-#define mb()	__asm__ __volatile__ ("": : :"memory")
-#define rmb()	mb()
-#define wmb()	__asm__ __volatile__ ("": : :"memory")
+#define mb()		__asm__ __volatile__ ("": : :"memory")
+#define rmb()		mb()
+#define wmb()		__asm__ __volatile__ ("": : :"memory")
+#define ctrl_barrier()	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop")
 #define read_barrier_depends()	do { } while(0)
 #endif
 
@@ -218,8 +243,8 @@ do {					\
 #define back_to_P1()					\
 do {							\
 	unsigned long __dummy;				\
+	ctrl_barrier();					\
 	__asm__ __volatile__(				\
-		"nop;nop;nop;nop;nop;nop;nop\n\t"	\
 		"mov.l	1f, %0\n\t"			\
 		"jmp	@%0\n\t"			\
 		" nop\n\t"				\
-- 
cgit v1.2.3


From a09749dd86e9e93de10f12ab4ce4e90815b5650a Mon Sep 17 00:00:00 2001
From: Jamie Lenehan <lenehan@twibble.org>
Date: Wed, 27 Sep 2006 15:05:39 +0900
Subject: sh: Titan board support.

Add support for the titan board.

Signed-off-by: Jamie Lenehan <lenehan@twibble.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                |    1 +
 arch/sh/boards/titan/Makefile   |    5 +
 arch/sh/boards/titan/io.c       |  156 +++++
 arch/sh/boards/titan/setup.c    |   60 ++
 arch/sh/configs/titan_defconfig | 1367 +++++++++++++++++++++++++++++++++++++++
 arch/sh/drivers/pci/Makefile    |    1 +
 arch/sh/drivers/pci/ops-titan.c |   84 +++
 arch/sh/drivers/pci/pci.c       |   58 +-
 arch/sh/tools/mach-types        |    1 +
 include/asm-sh/titan.h          |   43 ++
 10 files changed, 1764 insertions(+), 12 deletions(-)
 create mode 100644 arch/sh/boards/titan/Makefile
 create mode 100644 arch/sh/boards/titan/io.c
 create mode 100644 arch/sh/boards/titan/setup.c
 create mode 100644 arch/sh/configs/titan_defconfig
 create mode 100644 arch/sh/drivers/pci/ops-titan.c
 create mode 100644 include/asm-sh/titan.h

(limited to 'include')

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index e9112def48dd..55dcba2c06fc 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -106,6 +106,7 @@ machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
+machdir-$(CONFIG_SH_TITAN)			:= titan
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
diff --git a/arch/sh/boards/titan/Makefile b/arch/sh/boards/titan/Makefile
new file mode 100644
index 000000000000..08d753700062
--- /dev/null
+++ b/arch/sh/boards/titan/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the Nimble Microsystems TITAN specific parts of the kernel
+#
+
+obj-y	 := setup.o io.o
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
new file mode 100644
index 000000000000..d66900c99a11
--- /dev/null
+++ b/arch/sh/boards/titan/io.c
@@ -0,0 +1,156 @@
+/*
+ *	I/O routines for Titan
+ */
+
+#include <linux/pci.h>
+#include <asm/machvec.h>
+#include <asm/addrspace.h>
+#include <asm/titan.h>
+#include <asm/io.h>
+#include "../../drivers/pci/pci-sh7751.h"
+
+#define PCIIOBR         (volatile long *)PCI_REG(SH7751_PCIIOBR)
+#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
+#define PCI_IO_AREA     SH7751_PCI_IO_BASE
+#define PCI_MEM_AREA    SH7751_PCI_CONFIG_BASE
+
+#define PCI_IOMAP(adr)  (PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
+
+#if defined(CONFIG_PCI)
+#define CHECK_SH7751_PCIIO(port) \
+  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
+#define CHECK_SH7751_PCIMEMIO(port) \
+  ((port >= PCIBIOS_MIN_MEM) && (port < (PCIBIOS_MIN_MEM + SH7751_PCI_MEM_SIZE)))
+#else
+#define CHECK_SH7751_PCIIO(port) (0)
+#endif
+
+static inline void delay(void)
+{
+        ctrl_inw(0xa0000000);
+}
+
+static inline volatile u16 *port2adr(unsigned int port)
+{
+        maybebadio((unsigned long)port);
+        return (volatile u16*)port;
+}
+
+u8 titan_inb(unsigned long port)
+{
+        if (PXSEG(port))
+                return ctrl_inb(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return ctrl_inb(PCI_IOMAP(port));
+        return ctrl_inw(port2adr(port)) & 0xff;
+}
+
+u8 titan_inb_p(unsigned long port)
+{
+        u8 v;
+
+        if (PXSEG(port))
+                v = ctrl_inb(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                v = ctrl_inb(PCI_IOMAP(port));
+        else
+                v = ctrl_inw(port2adr(port)) & 0xff;
+        delay();
+        return v;
+}
+
+u16 titan_inw(unsigned long port)
+{
+        if (PXSEG(port))
+                return ctrl_inw(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return ctrl_inw(PCI_IOMAP(port));
+        else if (port >= 0x2000)
+                return ctrl_inw(port2adr(port));
+        else
+                maybebadio(port);
+        return 0;
+}
+
+u32 titan_inl(unsigned long port)
+{
+        if (PXSEG(port))
+                return ctrl_inl(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return ctrl_inl(PCI_IOMAP(port));
+        else if (port >= 0x2000)
+                return ctrl_inw(port2adr(port));
+        else
+                maybebadio(port);
+        return 0;
+}
+
+void titan_outb(u8 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outb(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outb(value, PCI_IOMAP(port));
+        else
+                ctrl_outw(value, port2adr(port));
+}
+
+void titan_outb_p(u8 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outb(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outb(value, PCI_IOMAP(port));
+        else
+                ctrl_outw(value, port2adr(port));
+        delay();
+}
+
+void titan_outw(u16 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outw(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outw(value, PCI_IOMAP(port));
+        else if (port >= 0x2000)
+                ctrl_outw(value, port2adr(port));
+        else
+                maybebadio(port);
+}
+
+void titan_outl(u32 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outl(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outl(value, PCI_IOMAP(port));
+        else
+                maybebadio(port);
+}
+
+void titan_insl(unsigned long port, void *dst, unsigned long count)
+{
+        maybebadio(port);
+}
+
+void titan_outsl(unsigned long port, const void *src, unsigned long count)
+{
+        maybebadio(port);
+}
+
+void *titan_ioremap(unsigned long offset, unsigned long size) {
+	if (CHECK_SH7751_PCIIO(offset) || CHECK_SH7751_PCIMEMIO(offset))
+		return (void *)offset;
+}
+
+void __iomem *titan_ioport_map(unsigned long port, unsigned int size)
+{
+	if (PXSEG(port))
+		return (void __iomem *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return (void __iomem *)PCI_IOMAP(port);
+
+	return (void __iomem *)port2adr(port);
+}
+
+EXPORT_SYMBOL(titan_ioremap);
diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c
new file mode 100644
index 000000000000..6ac5c8d7b3fb
--- /dev/null
+++ b/arch/sh/boards/titan/setup.c
@@ -0,0 +1,60 @@
+/*
+ *	Setup for Titan
+ */
+
+#include <linux/init.h>
+#include <asm/irq.h>
+#include <asm/titan.h>
+#include <asm/io.h>
+
+extern void __init pcibios_init_platform(void);
+
+static void __init init_titan_irq(void)
+{
+	/* enable individual interrupt mode for externals */
+	ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR);
+
+	make_ipr_irq( TITAN_IRQ_WAN,   IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY); /* PCIRQ0 */
+	make_ipr_irq( TITAN_IRQ_LAN,   IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY); /* PCIRQ1 */
+	make_ipr_irq( TITAN_IRQ_MPCIA, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY); /* PCIRQ2 */
+	make_ipr_irq( TITAN_IRQ_USB,   IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); /* PCIRQ3 */
+}
+
+const char *get_system_type(void)
+{
+	return "Titan";
+}
+
+int __init platform_setup(void)
+{
+	printk("%s Platform Setup\n", get_system_type());
+	return 0;
+}
+
+struct sh_machine_vector mv_titan __initmv = {
+	.mv_nr_irqs =	NR_IRQS,
+
+	.mv_inb =	titan_inb,
+	.mv_inw =	titan_inw,
+	.mv_inl =	titan_inl,
+	.mv_outb =	titan_outb,
+	.mv_outw =	titan_outw,
+	.mv_outl =	titan_outl,
+
+	.mv_inb_p =	titan_inb_p,
+	.mv_inw_p =	titan_inw,
+	.mv_inl_p =	titan_inl,
+	.mv_outb_p =	titan_outb_p,
+	.mv_outw_p =	titan_outw,
+	.mv_outl_p =	titan_outl,
+
+	.mv_insl =	titan_insl,
+	.mv_outsl =	titan_outsl,
+
+	.mv_ioremap =	titan_ioremap,
+	.mv_ioport_map = titan_ioport_map,
+
+	.mv_init_irq =	init_titan_irq,
+	.mv_init_pci =	pcibios_init_platform,
+};
+ALIAS_MV(titan)
diff --git a/arch/sh/configs/titan_defconfig b/arch/sh/configs/titan_defconfig
new file mode 100644
index 000000000000..1db2904de9e5
--- /dev/null
+++ b/arch/sh/configs/titan_defconfig
@@ -0,0 +1,1367 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.14-sh
+# Wed Nov  9 00:35:56 2005
+#
+CONFIG_SUPERH=y
+CONFIG_UID16=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_IOMAP=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_EMBEDDED is not set
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_EPOLL=y
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+CONFIG_SH_TITAN=y
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+CONFIG_CPU_SUBTYPE_SH7751=y
+CONFIG_CPU_SUBTYPE_SH7751R=y
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+CONFIG_MEMORY_START=0x08030000
+CONFIG_MEMORY_SIZE=0x7fd0000
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_SH_RTC=y
+CONFIG_SH_FPU=y
+# CONFIG_SH_STORE_QUEUES is not set
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ_BOOL=y
+CONFIG_SH_PCLK_FREQ=30000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=8
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_KEXEC is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x009e0000
+# CONFIG_UBC_WAKEUP is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttySC1,38400N81 root=/dev/nfs ip=:::::eth1:autoconf"
+
+#
+# Bus options
+#
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+CONFIG_PCI_LEGACY_PROC=y
+#CONFIG_PCI_DEBUG is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_HOTPLUG_PCI_FAKE is not set
+# CONFIG_HOTPLUG_PCI_CPCI is not set
+# CONFIG_HOTPLUG_PCI_SHPC is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+# CONFIG_XFRM_USER is not set
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_MULTIPLE_TABLES=y
+# CONFIG_IP_ROUTE_FWMARK is not set
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_MULTIPATH_CACHED=y
+CONFIG_IP_ROUTE_MULTIPATH_RR=m
+CONFIG_IP_ROUTE_MULTIPATH_RANDOM=m
+CONFIG_IP_ROUTE_MULTIPATH_WRANDOM=m
+CONFIG_IP_ROUTE_MULTIPATH_DRR=m
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+CONFIG_NET_IPIP=m
+CONFIG_NET_IPGRE=m
+CONFIG_NET_IPGRE_BROADCAST=y
+CONFIG_IP_MROUTE=y
+CONFIG_IP_PIMSM_V1=y
+CONFIG_IP_PIMSM_V2=y
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+CONFIG_INET_AH=m
+CONFIG_INET_ESP=m
+CONFIG_INET_IPCOMP=m
+CONFIG_INET_TUNNEL=m
+CONFIG_INET_DIAG=m
+CONFIG_INET_TCP_DIAG=m
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+
+#
+# IP: Virtual Server Configuration
+#
+# CONFIG_IP_VS is not set
+CONFIG_IPV6=m
+CONFIG_IPV6_PRIVACY=y
+CONFIG_INET6_AH=m
+CONFIG_INET6_ESP=m
+CONFIG_INET6_IPCOMP=m
+CONFIG_INET6_TUNNEL=m
+CONFIG_IPV6_TUNNEL=m
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+CONFIG_BRIDGE_NETFILTER=y
+CONFIG_NETFILTER_NETLINK=m
+CONFIG_NETFILTER_NETLINK_QUEUE=m
+CONFIG_NETFILTER_NETLINK_LOG=m
+
+#
+# IP: Netfilter Configuration
+#
+CONFIG_IP_NF_CONNTRACK=m
+CONFIG_IP_NF_CT_ACCT=y
+CONFIG_IP_NF_CONNTRACK_MARK=y
+CONFIG_IP_NF_CONNTRACK_EVENTS=y
+CONFIG_IP_NF_CONNTRACK_NETLINK=m
+# CONFIG_IP_NF_CT_PROTO_SCTP is not set
+CONFIG_IP_NF_FTP=m
+CONFIG_IP_NF_IRC=m
+CONFIG_IP_NF_NETBIOS_NS=m
+CONFIG_IP_NF_TFTP=m
+# CONFIG_IP_NF_AMANDA is not set
+CONFIG_IP_NF_PPTP=m
+CONFIG_IP_NF_QUEUE=m
+CONFIG_IP_NF_IPTABLES=m
+CONFIG_IP_NF_MATCH_LIMIT=m
+CONFIG_IP_NF_MATCH_IPRANGE=m
+CONFIG_IP_NF_MATCH_MAC=m
+CONFIG_IP_NF_MATCH_PKTTYPE=m
+CONFIG_IP_NF_MATCH_MARK=m
+CONFIG_IP_NF_MATCH_MULTIPORT=m
+CONFIG_IP_NF_MATCH_TOS=m
+CONFIG_IP_NF_MATCH_RECENT=m
+CONFIG_IP_NF_MATCH_ECN=m
+CONFIG_IP_NF_MATCH_DSCP=m
+CONFIG_IP_NF_MATCH_AH_ESP=m
+CONFIG_IP_NF_MATCH_LENGTH=m
+CONFIG_IP_NF_MATCH_TTL=m
+CONFIG_IP_NF_MATCH_TCPMSS=m
+CONFIG_IP_NF_MATCH_HELPER=m
+CONFIG_IP_NF_MATCH_STATE=m
+CONFIG_IP_NF_MATCH_CONNTRACK=m
+CONFIG_IP_NF_MATCH_OWNER=m
+# CONFIG_IP_NF_MATCH_PHYSDEV is not set
+CONFIG_IP_NF_MATCH_ADDRTYPE=m
+CONFIG_IP_NF_MATCH_REALM=m
+# CONFIG_IP_NF_MATCH_SCTP is not set
+# CONFIG_IP_NF_MATCH_DCCP is not set
+CONFIG_IP_NF_MATCH_COMMENT=m
+CONFIG_IP_NF_MATCH_CONNMARK=m
+CONFIG_IP_NF_MATCH_CONNBYTES=m
+CONFIG_IP_NF_MATCH_HASHLIMIT=m
+CONFIG_IP_NF_MATCH_STRING=m
+CONFIG_IP_NF_FILTER=m
+CONFIG_IP_NF_TARGET_REJECT=m
+CONFIG_IP_NF_TARGET_LOG=m
+CONFIG_IP_NF_TARGET_ULOG=m
+CONFIG_IP_NF_TARGET_TCPMSS=m
+CONFIG_IP_NF_TARGET_NFQUEUE=m
+CONFIG_IP_NF_NAT=m
+CONFIG_IP_NF_NAT_NEEDED=y
+CONFIG_IP_NF_TARGET_MASQUERADE=m
+CONFIG_IP_NF_TARGET_REDIRECT=m
+CONFIG_IP_NF_TARGET_NETMAP=m
+CONFIG_IP_NF_TARGET_SAME=m
+CONFIG_IP_NF_NAT_SNMP_BASIC=m
+CONFIG_IP_NF_NAT_IRC=m
+CONFIG_IP_NF_NAT_FTP=m
+CONFIG_IP_NF_NAT_TFTP=m
+CONFIG_IP_NF_NAT_PPTP=m
+CONFIG_IP_NF_MANGLE=m
+CONFIG_IP_NF_TARGET_TOS=m
+CONFIG_IP_NF_TARGET_ECN=m
+CONFIG_IP_NF_TARGET_DSCP=m
+CONFIG_IP_NF_TARGET_MARK=m
+CONFIG_IP_NF_TARGET_CLASSIFY=m
+CONFIG_IP_NF_TARGET_TTL=m
+CONFIG_IP_NF_TARGET_CONNMARK=m
+CONFIG_IP_NF_TARGET_CLUSTERIP=m
+CONFIG_IP_NF_RAW=m
+CONFIG_IP_NF_TARGET_NOTRACK=m
+CONFIG_IP_NF_ARPTABLES=m
+CONFIG_IP_NF_ARPFILTER=m
+CONFIG_IP_NF_ARP_MANGLE=m
+
+#
+# IPv6: Netfilter Configuration (EXPERIMENTAL)
+#
+CONFIG_IP6_NF_QUEUE=m
+CONFIG_IP6_NF_IPTABLES=m
+CONFIG_IP6_NF_MATCH_LIMIT=m
+CONFIG_IP6_NF_MATCH_MAC=m
+CONFIG_IP6_NF_MATCH_RT=m
+CONFIG_IP6_NF_MATCH_OPTS=m
+CONFIG_IP6_NF_MATCH_FRAG=m
+CONFIG_IP6_NF_MATCH_HL=m
+CONFIG_IP6_NF_MATCH_MULTIPORT=m
+CONFIG_IP6_NF_MATCH_OWNER=m
+CONFIG_IP6_NF_MATCH_MARK=m
+CONFIG_IP6_NF_MATCH_IPV6HEADER=m
+CONFIG_IP6_NF_MATCH_AHESP=m
+CONFIG_IP6_NF_MATCH_LENGTH=m
+CONFIG_IP6_NF_MATCH_EUI64=m
+# CONFIG_IP6_NF_MATCH_PHYSDEV is not set
+CONFIG_IP6_NF_FILTER=m
+CONFIG_IP6_NF_TARGET_LOG=m
+CONFIG_IP6_NF_TARGET_REJECT=m
+CONFIG_IP6_NF_TARGET_NFQUEUE=m
+CONFIG_IP6_NF_MANGLE=m
+CONFIG_IP6_NF_TARGET_MARK=m
+CONFIG_IP6_NF_TARGET_HL=m
+CONFIG_IP6_NF_RAW=m
+
+#
+# Bridge: Netfilter Configuration
+#
+# CONFIG_BRIDGE_NF_EBTABLES is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+CONFIG_BRIDGE=m
+CONFIG_VLAN_8021Q=m
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_CLK_JIFFIES=y
+# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set
+# CONFIG_NET_SCH_CLK_CPU is not set
+CONFIG_NET_SCH_CBQ=m
+CONFIG_NET_SCH_HTB=m
+CONFIG_NET_SCH_HFSC=m
+CONFIG_NET_SCH_PRIO=m
+CONFIG_NET_SCH_RED=m
+CONFIG_NET_SCH_SFQ=m
+CONFIG_NET_SCH_TEQL=m
+CONFIG_NET_SCH_TBF=m
+CONFIG_NET_SCH_GRED=m
+CONFIG_NET_SCH_DSMARK=m
+CONFIG_NET_SCH_NETEM=m
+CONFIG_NET_SCH_INGRESS=m
+CONFIG_NET_QOS=y
+CONFIG_NET_ESTIMATOR=y
+CONFIG_NET_CLS=y
+CONFIG_NET_CLS_BASIC=m
+CONFIG_NET_CLS_TCINDEX=m
+CONFIG_NET_CLS_ROUTE4=m
+CONFIG_NET_CLS_ROUTE=y
+CONFIG_NET_CLS_FW=m
+CONFIG_NET_CLS_U32=m
+CONFIG_CLS_U32_PERF=y
+CONFIG_NET_CLS_IND=y
+CONFIG_CLS_U32_MARK=y
+CONFIG_NET_CLS_RSVP=m
+CONFIG_NET_CLS_RSVP6=m
+CONFIG_NET_EMATCH=y
+CONFIG_NET_EMATCH_STACK=32
+CONFIG_NET_EMATCH_CMP=m
+CONFIG_NET_EMATCH_NBYTE=m
+CONFIG_NET_EMATCH_U32=m
+CONFIG_NET_EMATCH_META=m
+CONFIG_NET_EMATCH_TEXT=m
+CONFIG_NET_CLS_ACT=y
+CONFIG_NET_ACT_POLICE=m
+CONFIG_NET_ACT_GACT=m
+CONFIG_GACT_PROB=y
+CONFIG_NET_ACT_MIRRED=m
+CONFIG_NET_ACT_IPT=m
+CONFIG_NET_ACT_PEDIT=m
+# CONFIG_NET_ACT_SIMP is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+CONFIG_CONNECTOR=m
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+CONFIG_BLK_SSFDC=y
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=m
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_INITRD=y
+# CONFIG_LBD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=m
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=m
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+CONFIG_BLK_DEV_SR=m
+# CONFIG_BLK_DEV_SR_VENDOR is not set
+CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+# CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+
+#
+# SCSI Transport Attributes
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_ATTRS is not set
+
+#
+# SCSI low-level drivers
+#
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+CONFIG_SCSI_QLA2XXX=m
+# CONFIG_SCSI_QLA21XX is not set
+# CONFIG_SCSI_QLA22XX is not set
+# CONFIG_SCSI_QLA2300 is not set
+# CONFIG_SCSI_QLA2322 is not set
+# CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
+# CONFIG_FUSION_SAS is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# PHY device support
+#
+CONFIG_PHYLIB=m
+CONFIG_PHYCONTROL=y
+
+#
+# MII PHY device drivers
+#
+CONFIG_MARVELL_PHY=m
+CONFIG_DAVICOM_PHY=m
+CONFIG_QSEMI_PHY=m
+CONFIG_LXT_PHY=m
+CONFIG_CICADA_PHY=m
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+CONFIG_CASSINI=m
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_SMC91X is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+# CONFIG_NE2000 is not set
+CONFIG_NET_PCI=y
+# CONFIG_PCNET32 is not set
+# CONFIG_AMD8111_ETH is not set
+# CONFIG_ADAPTEC_STARFIRE is not set
+# CONFIG_B44 is not set
+# CONFIG_FORCEDETH is not set
+# CONFIG_DGRS is not set
+# CONFIG_EEPRO100 is not set
+# CONFIG_E100 is not set
+# CONFIG_FEALNX is not set
+# CONFIG_NATSEMI is not set
+# CONFIG_NE2K_PCI is not set
+# CONFIG_8139CP is not set
+CONFIG_8139TOO=y
+# CONFIG_8139TOO_PIO is not set
+CONFIG_8139TOO_TUNE_TWISTER=y
+# CONFIG_8139TOO_8129 is not set
+CONFIG_8139_OLD_RX_RESET=y
+# CONFIG_SIS900 is not set
+# CONFIG_EPIC100 is not set
+# CONFIG_SUNDANCE is not set
+# CONFIG_TLAN is not set
+# CONFIG_VIA_RHINE is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+# CONFIG_R8169 is not set
+# CONFIG_SIS190 is not set
+# CONFIG_SKGE is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+CONFIG_NET_RADIO=y
+
+#
+# Obsolete Wireless cards support (pre-802.11)
+#
+# CONFIG_STRIP is not set
+
+#
+# Wireless 802.11b ISA/PCI cards support
+#
+# CONFIG_HERMES is not set
+# CONFIG_ATMEL is not set
+
+#
+# Prism GT/Duette 802.11(a/b/g) PCI/Cardbus support
+#
+CONFIG_PRISM54=m
+# CONFIG_HOSTAP is not set
+CONFIG_NET_WIRELESS=y
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+CONFIG_PPP=m
+CONFIG_PPP_MULTILINK=y
+CONFIG_PPP_FILTER=y
+CONFIG_PPP_ASYNC=m
+CONFIG_PPP_SYNC_TTY=m
+CONFIG_PPP_DEFLATE=m
+CONFIG_PPP_BSDCOMP=m
+CONFIG_PPPOE=m
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia Capabilities Port drivers
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Console display driver support
+#
+CONFIG_DUMMY_CONSOLE=y
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB=m
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_BANDWIDTH is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+CONFIG_USB_EHCI_HCD=m
+# CONFIG_USB_EHCI_SPLIT_ISO is not set
+# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
+CONFIG_USB_OHCI_HCD=m
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
+# CONFIG_USB_UHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_BLUETOOTH_TTY is not set
+# CONFIG_USB_ACM is not set
+# CONFIG_USB_PRINTER is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=m
+# CONFIG_USB_STORAGE_DEBUG is not set
+# CONFIG_USB_STORAGE_DATAFAB is not set
+# CONFIG_USB_STORAGE_FREECOM is not set
+# CONFIG_USB_STORAGE_DPCM is not set
+# CONFIG_USB_STORAGE_USBAT is not set
+# CONFIG_USB_STORAGE_SDDR09 is not set
+# CONFIG_USB_STORAGE_SDDR55 is not set
+# CONFIG_USB_STORAGE_JUMPSHOT is not set
+
+#
+# USB Input Devices
+#
+# CONFIG_USB_HID is not set
+
+#
+# USB HID Boot Protocol drivers
+#
+# CONFIG_USB_KBD is not set
+# CONFIG_USB_MOUSE is not set
+# CONFIG_USB_AIPTEK is not set
+# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
+# CONFIG_USB_KBTAB is not set
+# CONFIG_USB_POWERMATE is not set
+# CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
+# CONFIG_USB_EGALAX is not set
+# CONFIG_USB_YEALINK is not set
+# CONFIG_USB_XPAD is not set
+# CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_KEYSPAN_REMOTE is not set
+# CONFIG_USB_APPLETOUCH is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+
+#
+# USB Multimedia devices
+#
+# CONFIG_USB_DABUSB is not set
+
+#
+# Video4Linux support is needed for USB Multimedia device support
+#
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_USB_ZD1201 is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+CONFIG_USB_SERIAL=m
+CONFIG_USB_SERIAL_GENERIC=y
+# CONFIG_USB_SERIAL_AIRPRIME is not set
+# CONFIG_USB_SERIAL_BELKIN is not set
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
+# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
+# CONFIG_USB_SERIAL_CP2101 is not set
+# CONFIG_USB_SERIAL_CYPRESS_M8 is not set
+# CONFIG_USB_SERIAL_EMPEG is not set
+# CONFIG_USB_SERIAL_FTDI_SIO is not set
+# CONFIG_USB_SERIAL_VISOR is not set
+# CONFIG_USB_SERIAL_IPAQ is not set
+# CONFIG_USB_SERIAL_IR is not set
+# CONFIG_USB_SERIAL_EDGEPORT is not set
+# CONFIG_USB_SERIAL_EDGEPORT_TI is not set
+# CONFIG_USB_SERIAL_GARMIN is not set
+# CONFIG_USB_SERIAL_IPW is not set
+# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
+# CONFIG_USB_SERIAL_KEYSPAN is not set
+# CONFIG_USB_SERIAL_KLSI is not set
+# CONFIG_USB_SERIAL_KOBIL_SCT is not set
+# CONFIG_USB_SERIAL_MCT_U232 is not set
+CONFIG_USB_SERIAL_PL2303=m
+# CONFIG_USB_SERIAL_HP4X is not set
+# CONFIG_USB_SERIAL_SAFE is not set
+# CONFIG_USB_SERIAL_TI is not set
+# CONFIG_USB_SERIAL_CYBERJACK is not set
+# CONFIG_USB_SERIAL_XIRCOM is not set
+# CONFIG_USB_SERIAL_OMNINET is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGETKIT is not set
+# CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_SISUSBVGA is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TEST is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_EXT2_FS_POSIX_ACL=y
+CONFIG_EXT2_FS_SECURITY=y
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+CONFIG_EXT3_FS_POSIX_ACL=y
+CONFIG_EXT3_FS_SECURITY=y
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+CONFIG_REISERFS_FS=m
+# CONFIG_REISERFS_CHECK is not set
+# CONFIG_REISERFS_PROC_INFO is not set
+# CONFIG_REISERFS_FS_XATTR is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+CONFIG_FUSE_FS=m
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=m
+CONFIG_JOLIET=y
+CONFIG_ZISOFS=y
+CONFIG_ZISOFS_FS=m
+CONFIG_UDF_FS=m
+CONFIG_UDF_NLS=y
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+CONFIG_RELAYFS_FS=m
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+CONFIG_SMB_FS=m
+# CONFIG_SMB_NLS_DEFAULT is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+CONFIG_9P_FS=m
+
+#
+# Partition Types
+#
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_ACORN_PARTITION is not set
+# CONFIG_OSF_PARTITION is not set
+# CONFIG_AMIGA_PARTITION is not set
+# CONFIG_ATARI_PARTITION is not set
+# CONFIG_MAC_PARTITION is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_BSD_DISKLABEL is not set
+# CONFIG_MINIX_SUBPARTITION is not set
+# CONFIG_SOLARIS_X86_PARTITION is not set
+# CONFIG_UNIXWARE_DISKLABEL is not set
+# CONFIG_LDM_PARTITION is not set
+# CONFIG_SGI_PARTITION is not set
+# CONFIG_ULTRIX_PARTITION is not set
+# CONFIG_SUN_PARTITION is not set
+# CONFIG_EFI_PARTITION is not set
+
+#
+# Native Language Support
+#
+CONFIG_NLS=m
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=m
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+CONFIG_NLS_ASCII=m
+CONFIG_NLS_ISO8859_1=m
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=m
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_DEBUG_KERNEL=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_LOG_BUF_SHIFT=16
+# CONFIG_DETECT_SOFTLOCKUP is not set
+# CONFIG_SCHEDSTATS is not set
+CONFIG_DEBUG_SLAB=y
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_KOBJECT is not set
+# CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_FRAME_POINTER is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+CONFIG_EARLY_SCIF_CONSOLE=y
+# CONFIG_EARLY_PRINTK is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+CONFIG_CRYPTO=y
+CONFIG_CRYPTO_HMAC=y
+# CONFIG_CRYPTO_NULL is not set
+CONFIG_CRYPTO_MD4=m
+CONFIG_CRYPTO_MD5=m
+CONFIG_CRYPTO_SHA1=m
+CONFIG_CRYPTO_SHA256=m
+CONFIG_CRYPTO_SHA512=m
+CONFIG_CRYPTO_WP512=m
+CONFIG_CRYPTO_TGR192=m
+CONFIG_CRYPTO_DES=m
+CONFIG_CRYPTO_BLOWFISH=m
+CONFIG_CRYPTO_TWOFISH=m
+CONFIG_CRYPTO_SERPENT=m
+CONFIG_CRYPTO_AES=m
+CONFIG_CRYPTO_CAST5=m
+CONFIG_CRYPTO_CAST6=m
+CONFIG_CRYPTO_TEA=m
+CONFIG_CRYPTO_ARC4=m
+CONFIG_CRYPTO_KHAZAD=m
+CONFIG_CRYPTO_ANUBIS=m
+CONFIG_CRYPTO_DEFLATE=m
+CONFIG_CRYPTO_MICHAEL_MIC=m
+CONFIG_CRYPTO_CRC32C=m
+# CONFIG_CRYPTO_TEST is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+CONFIG_CRC_CCITT=m
+CONFIG_CRC16=m
+CONFIG_CRC32=y
+CONFIG_LIBCRC32C=m
+CONFIG_ZLIB_INFLATE=m
+CONFIG_ZLIB_DEFLATE=m
+CONFIG_TEXTSEARCH=y
+CONFIG_TEXTSEARCH_KMP=m
+CONFIG_TEXTSEARCH_BM=m
+CONFIG_TEXTSEARCH_FSM=m
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 365bc16a4a83..86d6d0660d4c 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -14,3 +14,4 @@ obj-$(CONFIG_SH_SECUREEDGE5410)		+= ops-snapgear.o
 obj-$(CONFIG_SH_BIGSUR)			+= ops-bigsur.o
 obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
+obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
diff --git a/arch/sh/drivers/pci/ops-titan.c b/arch/sh/drivers/pci/ops-titan.c
new file mode 100644
index 000000000000..f76e4e963ac1
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-titan.c
@@ -0,0 +1,84 @@
+/*
+ * arch/sh/drivers/pci/ops-titan.c
+ *
+ * Ported to new API by Paul Mundt <lethal@linux-sh.org>
+ *
+ * Modified from ops-snapgear.c written by  David McCullough
+ * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * PCI initialization for the Titan boards
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <asm/io.h>
+#include <asm/titan.h>
+#include "pci-sh7751.h"
+
+int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+{
+	int irq = -1;
+
+	switch (slot) {
+	case 0: irq = TITAN_IRQ_WAN;   break;	/* eth0 (WAN) */
+	case 1: irq = TITAN_IRQ_LAN;   break;	/* eth1 (LAN) */
+	case 2: irq = TITAN_IRQ_MPCIA; break;	/* mPCI A */
+	case 3: irq = TITAN_IRQ_MPCIB; break;	/* mPCI B */
+	case 4: irq = TITAN_IRQ_USB;   break;	/* USB */
+	default:
+		printk(KERN_INFO "PCI: Bad IRQ mapping request for slot %d\n", slot);
+		return -1;
+	}
+
+	printk("PCI: Mapping TITAN IRQ for slot %d, pin %c to irq %d\n",
+		slot, pin - 1 + 'A', irq);
+
+	return irq;
+}
+
+static struct resource sh7751_io_resource = {
+	.name	= "SH7751_IO",
+	.start	= SH7751_PCI_IO_BASE,
+	.end	= SH7751_PCI_IO_BASE + SH7751_PCI_IO_SIZE - 1,
+	.flags	= IORESOURCE_IO
+};
+
+static struct resource sh7751_mem_resource = {
+	.name	= "SH7751_mem",
+	.start	= SH7751_PCI_MEMORY_BASE,
+	.end	= SH7751_PCI_MEMORY_BASE + SH7751_PCI_MEM_SIZE - 1,
+	.flags	= IORESOURCE_MEM
+};
+
+extern struct pci_ops sh7751_pci_ops;
+
+struct pci_channel board_pci_channels[] = {
+	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ NULL, NULL, NULL, 0, 0 },
+};
+EXPORT_SYMBOL(board_pci_channels);
+
+static struct sh7751_pci_address_map sh7751_pci_map = {
+	.window0	= {
+		.base	= SH7751_CS2_BASE_ADDR,
+		.size	= SH7751_MEM_REGION_SIZE*2,	/* cs2 and cs3 */
+	},
+
+	.window1	= {
+		.base	= SH7751_CS2_BASE_ADDR,
+		.size	= SH7751_MEM_REGION_SIZE*2,
+	},
+
+	.flags	= SH7751_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	return sh7751_pcic_init(&sh7751_pci_map);
+}
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 3d546ba329cf..1f5e23e8b163 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -1,21 +1,25 @@
-/* arch/sh/kernel/pci.c
- * $Id: pci.c,v 1.1 2003/08/24 19:15:45 lethal Exp $
+/*
+ * arch/sh/drivers/pci/pci.c
  *
  * Copyright (c) 2002 M. R. Brown  <mrbrown@linux-sh.org>
- * 
- * 
+ * Copyright (c) 2004, 2005 Paul Mundt  <lethal@linux-sh.org>
+ *
  * These functions are collected here to reduce duplication of common
  * code amongst the many platform-specific PCI support code files.
- * 
+ *
  * These routines require the following board-specific routines:
  * void pcibios_fixup_irqs();
  *
  * See include/asm-sh/pci.h for more information.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <linux/init.h>
+#include <asm/io.h>
 
 static int __init pcibios_init(void)
 {
@@ -26,9 +30,8 @@ static int __init pcibios_init(void)
 #ifdef CONFIG_PCI_AUTO
 	/* assign resources */
 	busno = 0;
-	for (p = board_pci_channels; p->pci_ops != NULL; p++) {
+	for (p = board_pci_channels; p->pci_ops != NULL; p++)
 		busno = pciauto_assign_resources(busno, p) + 1;
-	}
 #endif
 
 	/* scan the buses */
@@ -61,13 +64,17 @@ pcibios_update_resource(struct pci_dev *dev, struct resource *root,
 		new |= PCI_ROM_ADDRESS_ENABLE;
 		reg = dev->rom_base_reg;
 	} else {
-		/* Somebody might have asked allocation of a non-standard resource */
+		/*
+		 * Somebody might have asked allocation of a non-standard
+		 * resource
+		 */
 		return;
 	}
-	
+
 	pci_write_config_dword(dev, reg, new);
 	pci_read_config_dword(dev, reg, &check);
-	if ((new ^ check) & ((new & PCI_BASE_ADDRESS_SPACE_IO) ? PCI_BASE_ADDRESS_IO_MASK : PCI_BASE_ADDRESS_MEM_MASK)) {
+	if ((new ^ check) & ((new & PCI_BASE_ADDRESS_SPACE_IO) ?
+		PCI_BASE_ADDRESS_IO_MASK : PCI_BASE_ADDRESS_MEM_MASK)) {
 		printk(KERN_ERR "PCI: Error while updating region "
 		       "%s/%d (%08x != %08x)\n", pci_name(dev), resource,
 		       new, check);
@@ -145,7 +152,8 @@ void pcibios_set_master(struct pci_dev *dev)
 		lat = pcibios_max_latency;
 	else
 		return;
-	printk(KERN_INFO "PCI: Setting latency timer of device %s to %d\n", pci_name(dev), lat);
+	printk(KERN_INFO "PCI: Setting latency timer of device %s to %d\n",
+	       pci_name(dev), lat);
 	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
 }
 
@@ -153,3 +161,29 @@ void __init pcibios_update_irq(struct pci_dev *dev, int irq)
 {
 	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
 }
+
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
+{
+	unsigned long start = pci_resource_start(dev, bar);
+	unsigned long len = pci_resource_len(dev, bar);
+	unsigned long flags = pci_resource_flags(dev, bar);
+
+	if (!len || !start)
+		return NULL;
+	if (maxlen && len > maxlen)
+		len = maxlen;
+	if (flags & IORESOURCE_IO)
+		return ioport_map(start, len);
+	if (flags & IORESOURCE_MEM)
+		return ioremap(start, len);
+
+	return NULL;
+}
+
+void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
+{
+	iounmap(addr);
+}
+
+EXPORT_SYMBOL(pci_iomap);
+EXPORT_SYMBOL(pci_iounmap);
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index c503ca072e11..01ae394a6599 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -26,3 +26,4 @@ EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
 LANDISK			SH_LANDISK
+TITAN			SH_TITAN
diff --git a/include/asm-sh/titan.h b/include/asm-sh/titan.h
new file mode 100644
index 000000000000..270a4f4bc8a9
--- /dev/null
+++ b/include/asm-sh/titan.h
@@ -0,0 +1,43 @@
+/*
+ * Platform defintions for Titan
+ */
+
+#ifndef _ASM_SH_TITAN_TITAN_H
+#define _ASM_SH_TITAN_TITAN_H
+
+#define __IO_PREFIX titan
+#include <asm/io_generic.h>
+
+/* IRQ assignments */
+#define TITAN_IRQ_WAN		2	/* eth0 (WAN) */
+#define TITAN_IRQ_LAN		5	/* eth1 (LAN) */
+#define TITAN_IRQ_MPCIA		8	/* mPCI A */
+#define TITAN_IRQ_MPCIB		11	/* mPCI B */
+#define TITAN_IRQ_USB		11	/* USB */
+
+/*
+ * The external interrupt lines, these take up ints 0 - 15 inclusive
+ * depending on the priority for the interrupt.  In fact the priority
+ * is the interrupt :-)
+ */
+#define IRL0_IRQ	0
+#define IRL0_IPR_ADDR	INTC_IPRD
+#define IRL0_IPR_POS	3
+#define IRL0_PRIORITY	8
+
+#define IRL1_IRQ	1
+#define IRL1_IPR_ADDR	INTC_IPRD
+#define IRL1_IPR_POS	2
+#define IRL1_PRIORITY	8
+
+#define IRL2_IRQ	2
+#define IRL2_IPR_ADDR	INTC_IPRD
+#define IRL2_IPR_POS	1
+#define IRL2_PRIORITY	8
+
+#define IRL3_IRQ	3
+#define IRL3_IPR_ADDR	INTC_IPRD
+#define IRL3_IPR_POS	0
+#define IRL3_PRIORITY	8
+
+#endif
-- 
cgit v1.2.3


From 0c7b1df69c62209db19d1279dd882b37c04c5c2f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:08:07 +0900
Subject: sh: SH-4A Privileged Space Mapping Buffer (PMB) support.

Add support for 32-bit physical addressing through the SH-4A
Privileged Space Mapping Buffer (PMB).

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/Makefile  |   3 +-
 arch/sh/mm/pmb.c     | 269 +++++++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/mmu.h |  48 ++++++++-
 3 files changed, 318 insertions(+), 2 deletions(-)
 create mode 100644 arch/sh/mm/pmb.c

(limited to 'include')

diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index 9489a1424644..f4e32b3d24dc 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -22,4 +22,5 @@ obj-$(CONFIG_CPU_SH4)	+= tlb-sh4.o ioremap.o
 obj-$(CONFIG_SH7705_CACHE_32KB) += pg-sh7705.o
 endif
 
-obj-$(CONFIG_SH7705_CACHE_32KB) += cache-sh7705.o
+obj-$(CONFIG_SH7705_CACHE_32KB)	+= cache-sh7705.o
+obj-$(CONFIG_32BIT)		+= pmb.o
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
new file mode 100644
index 000000000000..ff5bde745647
--- /dev/null
+++ b/arch/sh/mm/pmb.c
@@ -0,0 +1,269 @@
+/*
+ * arch/sh/mm/pmb.c
+ *
+ * Privileged Space Mapping Buffer (PMB) Support.
+ *
+ * Copyright (C) 2005 Paul Mundt
+ *
+ * P1/P2 Section mapping definitions from map32.h, which was:
+ *
+ *	Copyright 2003 (c) Lineo Solutions,Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/bitops.h>
+#include <linux/debugfs.h>
+#include <linux/fs.h>
+#include <linux/seq_file.h>
+#include <linux/err.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/mmu.h>
+#include <asm/io.h>
+
+#define NR_PMB_ENTRIES	16
+
+static kmem_cache_t *pmb_cache;
+static unsigned long pmb_map;
+
+static struct pmb_entry pmb_init_map[] = {
+	/* vpn         ppn         flags (ub/sz/c/wt) */
+
+	/* P1 Section Mappings */
+	{ 0x80000000, 0x00000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x84000000, 0x04000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x88000000, 0x08000000, PMB_SZ_128M | PMB_C, },
+	{ 0x90000000, 0x10000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x94000000, 0x14000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x98000000, 0x18000000, PMB_SZ_64M  | PMB_C, },
+
+	/* P2 Section Mappings */
+	{ 0xa0000000, 0x00000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xa4000000, 0x04000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xa8000000, 0x08000000, PMB_UB | PMB_SZ_128M | PMB_WT, },
+	{ 0xb0000000, 0x10000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xb4000000, 0x14000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xb8000000, 0x18000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+};
+
+static inline unsigned long mk_pmb_entry(unsigned int entry)
+{
+	return (entry & PMB_E_MASK) << PMB_E_SHIFT;
+}
+
+static inline unsigned long mk_pmb_addr(unsigned int entry)
+{
+	return mk_pmb_entry(entry) | PMB_ADDR;
+}
+
+static inline unsigned long mk_pmb_data(unsigned int entry)
+{
+	return mk_pmb_entry(entry) | PMB_DATA;
+}
+
+struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
+			    unsigned long flags)
+{
+	struct pmb_entry *pmbe;
+
+	pmbe = kmem_cache_alloc(pmb_cache, GFP_KERNEL);
+	if (!pmbe)
+		return ERR_PTR(-ENOMEM);
+
+	pmbe->vpn	= vpn;
+	pmbe->ppn	= ppn;
+	pmbe->flags	= flags;
+
+	return pmbe;
+}
+
+void pmb_free(struct pmb_entry *pmbe)
+{
+	kmem_cache_free(pmb_cache, pmbe);
+}
+
+/*
+ * Must be in P2 for __set_pmb_entry()
+ */
+int __set_pmb_entry(unsigned long vpn, unsigned long ppn,
+		    unsigned long flags, int *entry)
+{
+	unsigned int pos = *entry;
+
+	if (unlikely(pos == PMB_NO_ENTRY))
+		pos = find_first_zero_bit(&pmb_map, NR_PMB_ENTRIES);
+
+repeat:
+	if (unlikely(pos > NR_PMB_ENTRIES))
+		return -ENOSPC;
+
+	if (test_and_set_bit(pos, &pmb_map)) {
+		pos = find_first_zero_bit(&pmb_map, NR_PMB_ENTRIES);
+		goto repeat;
+	}
+
+	ctrl_outl(vpn | PMB_V, mk_pmb_addr(pos));
+
+#ifdef CONFIG_SH_WRITETHROUGH
+	/*
+	 * When we are in 32-bit address extended mode, CCR.CB becomes
+	 * invalid, so care must be taken to manually adjust cacheable
+	 * translations.
+	 */
+	if (likely(flags & PMB_C))
+		flags |= PMB_WT;
+#endif
+
+	ctrl_outl(ppn | flags | PMB_V, mk_pmb_data(pos));
+
+	*entry = pos;
+
+	return 0;
+}
+
+void set_pmb_entry(struct pmb_entry *pmbe)
+{
+	jump_to_P2();
+	__set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
+	back_to_P1();
+}
+
+void clear_pmb_entry(struct pmb_entry *pmbe)
+{
+	unsigned int entry = pmbe->entry;
+	unsigned long addr;
+
+	/*
+	 * Don't allow clearing of wired init entries, P1 or P2 access
+	 * without a corresponding mapping in the PMB will lead to reset
+	 * by the TLB.
+	 */
+	if (unlikely(entry < ARRAY_SIZE(pmb_init_map) ||
+		     entry >= NR_PMB_ENTRIES))
+		return;
+
+	jump_to_P2();
+
+	/* Clear V-bit */
+	addr = mk_pmb_addr(entry);
+	ctrl_outl(ctrl_inl(addr) & ~PMB_V, addr);
+
+	addr = mk_pmb_data(entry);
+	ctrl_outl(ctrl_inl(addr) & ~PMB_V, addr);
+
+	back_to_P1();
+
+	clear_bit(entry, &pmb_map);
+}
+
+static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags)
+{
+	memset(pmb, 0, sizeof(struct pmb_entry));
+
+	((struct pmb_entry *)pmb)->entry = PMB_NO_ENTRY;
+}
+
+static int __init pmb_init(void)
+{
+	unsigned int nr_entries = ARRAY_SIZE(pmb_init_map);
+	unsigned int entry;
+
+	BUG_ON(unlikely(nr_entries >= NR_PMB_ENTRIES));
+
+	pmb_cache = kmem_cache_create("pmb", sizeof(struct pmb_entry),
+				      0, 0, pmb_cache_ctor, NULL);
+	BUG_ON(!pmb_cache);
+
+	jump_to_P2();
+
+	/*
+	 * Ordering is important, P2 must be mapped in the PMB before we
+	 * can set PMB.SE, and P1 must be mapped before we jump back to
+	 * P1 space.
+	 */
+	for (entry = 0; entry < nr_entries; entry++) {
+		struct pmb_entry *pmbe = pmb_init_map + entry;
+
+		__set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &entry);
+	}
+
+	ctrl_outl(0, PMB_IRMCR);
+
+	/* PMB.SE and UB[7] */
+	ctrl_outl((1 << 31) | (1 << 7), PMB_PASCR);
+
+	back_to_P1();
+
+	return 0;
+}
+
+arch_initcall(pmb_init);
+
+#ifdef CONFIG_DEBUG_FS
+static int pmb_seq_show(struct seq_file *file, void *iter)
+{
+	int i;
+
+	seq_printf(file, "V: Valid, C: Cacheable, WT: Write-Through\n"
+			 "CB: Copy-Back, B: Buffered, UB: Unbuffered\n");
+	seq_printf(file, "ety   vpn  ppn  size   flags\n");
+
+	for (i = 0; i < NR_PMB_ENTRIES; i++) {
+		unsigned long addr, data;
+		unsigned int size;
+		char *sz_str = NULL;
+
+		addr = ctrl_inl(mk_pmb_addr(i));
+		data = ctrl_inl(mk_pmb_data(i));
+
+		size = data & PMB_SZ_MASK;
+		sz_str = (size == PMB_SZ_16M)  ? " 16MB":
+			 (size == PMB_SZ_64M)  ? " 64MB":
+			 (size == PMB_SZ_128M) ? "128MB":
+					         "512MB";
+
+		/* 02: V 0x88 0x08 128MB C CB  B */
+		seq_printf(file, "%02d: %c 0x%02lx 0x%02lx %s %c %s %s\n",
+			   i, ((addr & PMB_V) && (data & PMB_V)) ? 'V' : ' ',
+			   (addr >> 24) & 0xff, (data >> 24) & 0xff,
+			   sz_str, (data & PMB_C) ? 'C' : ' ',
+			   (data & PMB_WT) ? "WT" : "CB",
+			   (data & PMB_UB) ? "UB" : " B");
+	}
+
+	return 0;
+}
+
+static int pmb_debugfs_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, pmb_seq_show, NULL);
+}
+
+static struct file_operations pmb_debugfs_fops = {
+	.owner		= THIS_MODULE,
+	.open		= pmb_debugfs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int __init pmb_debugfs_init(void)
+{
+	struct dentry *dentry;
+
+	dentry = debugfs_create_file("pmb", S_IFREG | S_IRUGO,
+				     NULL, NULL, &pmb_debugfs_fops);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
+
+	return 0;
+}
+
+postcore_initcall(pmb_debugfs_init);
+#endif
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index 72f07be6117f..91c884634276 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -25,5 +25,51 @@ typedef struct {
 typedef unsigned long mm_context_t;
 
 #endif /* CONFIG_MMU */
-#endif /* __MMH_H */
+
+/*
+ * Privileged Space Mapping Buffer (PMB) definitions
+ */
+#define PMB_PASCR		0xff000070
+#define PMB_IRMCR		0xff000078
+
+#define PMB_ADDR		0xf6100000
+#define PMB_DATA		0xf7100000
+#define PMB_ENTRY_MAX		16
+#define PMB_E_MASK		0x0000000f
+#define PMB_E_SHIFT		8
+
+#define PMB_SZ_16M		0x00000000
+#define PMB_SZ_64M		0x00000010
+#define PMB_SZ_128M		0x00000080
+#define PMB_SZ_512M		0x00000090
+#define PMB_SZ_MASK		PMB_SZ_512M
+#define PMB_C			0x00000008
+#define PMB_WT			0x00000001
+#define PMB_UB			0x00000200
+#define PMB_V			0x00000100
+
+#define PMB_NO_ENTRY		(-1)
+
+struct pmb_entry {
+	unsigned long vpn;
+	unsigned long ppn;
+	unsigned long flags;
+
+	/*
+	 * 0 .. NR_PMB_ENTRIES for specific entry selection, or
+	 * PMB_NO_ENTRY to search for a free one
+	 */
+	int entry;
+};
+
+/* arch/sh/mm/pmb.c */
+int __set_pmb_entry(unsigned long vpn, unsigned long ppn,
+		    unsigned long flags, int *entry);
+void set_pmb_entry(struct pmb_entry *pmbe);
+void clear_pmb_entry(struct pmb_entry *pmbe);
+struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
+			    unsigned long flags);
+void pmb_free(struct pmb_entry *pmbe);
+
+#endif /* __MMU_H */
 
-- 
cgit v1.2.3


From 26ff6c11ef38e08990c1e417c299246e6ab18ff7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:13:36 +0900
Subject: sh: page table alloc cleanups and page fault optimizations.

Cleanup of page table allocators, using generic folded PMD and PUD
helpers. TLB flushing operations are moved to a more sensible spot.

The page fault handler is also optimized slightly, we no longer waste
cycles on IRQ disabling for flushing of the page from the ITLB, since
we're already under CLI protection by the initial exception handler.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/sys_sh.c             |   2 +-
 arch/sh/mm/Makefile                 |   2 +-
 arch/sh/mm/consistent.c             |   2 +
 arch/sh/mm/fault.c                  | 202 +++++++-----------------------------
 arch/sh/mm/init.c                   |  13 ++-
 arch/sh/mm/tlb-flush.c              | 132 +++++++++++++++++++++++
 include/asm-sh/cache.h              |   8 --
 include/asm-sh/cacheflush.h         |   1 +
 include/asm-sh/cpu-sh3/cacheflush.h |   8 +-
 include/asm-sh/cpu-sh4/cacheflush.h |  29 +++---
 include/asm-sh/page.h               |  12 +--
 include/asm-sh/pgalloc.h            |  37 +------
 include/asm-sh/pgtable.h            |  80 +++++++++-----
 13 files changed, 261 insertions(+), 267 deletions(-)
 create mode 100644 arch/sh/mm/tlb-flush.c

(limited to 'include')

diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index 917b2f32f260..d8bcd8a22327 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -21,7 +21,7 @@
 #include <linux/mman.h>
 #include <linux/file.h>
 #include <linux/utsname.h>
-
+#include <asm/cacheflush.h>
 #include <asm/uaccess.h>
 #include <asm/ipc.h>
 
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index f4e32b3d24dc..d90906367c5f 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -12,7 +12,7 @@ obj-$(CONFIG_DMA_PAGE_OPS)	+= pg-dma.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
 
 mmu-y			:= fault-nommu.o tlb-nommu.o pg-nommu.o
-mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o
+mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o
 
 obj-y			+= $(mmu-y)
 
diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c
index ee73e30263af..c81e6b67ad30 100644
--- a/arch/sh/mm/consistent.c
+++ b/arch/sh/mm/consistent.c
@@ -9,6 +9,8 @@
  */
 #include <linux/mm.h>
 #include <linux/dma-mapping.h>
+#include <asm/cacheflush.h>
+#include <asm/addrspace.h>
 #include <asm/io.h>
 
 void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle)
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 364181f27b79..7a03ffe6dadd 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -1,33 +1,20 @@
-/* $Id: fault.c,v 1.14 2004/01/13 05:52:11 kkojima Exp $
+/*
+ * Page fault handler for SH with an MMU.
  *
- *  linux/arch/sh/mm/fault.c
  *  Copyright (C) 1999  Niibe Yutaka
  *  Copyright (C) 2003  Paul Mundt
  *
  *  Based on linux/arch/i386/mm/fault.c:
  *   Copyright (C) 1995  Linus Torvalds
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
-#include <linux/signal.h>
-#include <linux/sched.h>
 #include <linux/kernel.h>
-#include <linux/errno.h>
-#include <linux/string.h>
-#include <linux/types.h>
-#include <linux/ptrace.h>
-#include <linux/mman.h>
 #include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/interrupt.h>
-#include <linux/module.h>
-
 #include <asm/system.h>
-#include <asm/io.h>
-#include <asm/uaccess.h>
-#include <asm/pgalloc.h>
 #include <asm/mmu_context.h>
-#include <asm/cacheflush.h>
 #include <asm/kgdb.h>
 
 extern void die(const char *,struct pt_regs *,long);
@@ -187,14 +174,25 @@ do_sigbus:
 		goto no_context;
 }
 
+#ifdef CONFIG_SH_STORE_QUEUES
 /*
- * Called with interrupt disabled.
+ * This is a special case for the SH-4 store queues, as pages for this
+ * space still need to be faulted in before it's possible to flush the
+ * store queue cache for writeout to the remapped region.
+ */
+#define P3_ADDR_MAX		(P4SEG_STORE_QUE + 0x04000000)
+#else
+#define P3_ADDR_MAX		P4SEG
+#endif
+
+/*
+ * Called with interrupts disabled.
  */
 asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 			       unsigned long address)
 {
-	unsigned long addrmax = P4SEG;
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 	pte_t entry;
@@ -207,31 +205,36 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 		kgdb_bus_err_hook();
 #endif
 
-#ifdef CONFIG_SH_STORE_QUEUES
-	addrmax = P4SEG_STORE_QUE + 0x04000000;
-#endif
-
-	if (address >= P3SEG && address < addrmax) {
+	/*
+	 * We don't take page faults for P1, P2, and parts of P4, these
+	 * are always mapped, whether it be due to legacy behaviour in
+	 * 29-bit mode, or due to PMB configuration in 32-bit mode.
+	 */
+	if (address >= P3SEG && address < P3_ADDR_MAX)
 		pgd = pgd_offset_k(address);
-		mm = NULL;
-	} else if (address >= TASK_SIZE)
-		return 1;
-	else if (!(mm = current->mm))
-		return 1;
-	else
-		pgd = pgd_offset(mm, address);
+	else {
+		if (unlikely(address >= TASK_SIZE || !current->mm))
+			return 1;
+
+		pgd = pgd_offset(current->mm, address);
+	}
 
-	pmd = pmd_offset(pgd, address);
+	pud = pud_offset(pgd, address);
+	if (pud_none_or_clear_bad(pud))
+		return 1;
+	pmd = pmd_offset(pud, address);
 	if (pmd_none_or_clear_bad(pmd))
 		return 1;
+
 	if (mm)
 		pte = pte_offset_map_lock(mm, pmd, address, &ptl);
 	else
 		pte = pte_offset_kernel(pmd, address);
 
 	entry = *pte;
-	if (pte_none(entry) || pte_not_present(entry)
-	    || (writeaccess && !pte_write(entry)))
+	if (unlikely(pte_none(entry) || pte_not_present(entry)))
+		goto unlock;
+	if (unlikely(writeaccess && !pte_write(entry)))
 		goto unlock;
 
 	if (writeaccess)
@@ -243,13 +246,7 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 	 * ITLB is not affected by "ldtlb" instruction.
 	 * So, we need to flush the entry by ourselves.
 	 */
-
-	{
-		unsigned long flags;
-		local_irq_save(flags);
-		__flush_tlb_page(get_asid(), address&PAGE_MASK);
-		local_irq_restore(flags);
-	}
+	__flush_tlb_page(get_asid(), address & PAGE_MASK);
 #endif
 
 	set_pte(pte, entry);
@@ -260,122 +257,3 @@ unlock:
 		pte_unmap_unlock(pte, ptl);
 	return ret;
 }
-
-void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
-{
-	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
-		unsigned long flags;
-		unsigned long asid;
-		unsigned long saved_asid = MMU_NO_ASID;
-
-		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
-		page &= PAGE_MASK;
-
-		local_irq_save(flags);
-		if (vma->vm_mm != current->mm) {
-			saved_asid = get_asid();
-			set_asid(asid);
-		}
-		__flush_tlb_page(asid, page);
-		if (saved_asid != MMU_NO_ASID)
-			set_asid(saved_asid);
-		local_irq_restore(flags);
-	}
-}
-
-void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
-		     unsigned long end)
-{
-	struct mm_struct *mm = vma->vm_mm;
-
-	if (mm->context != NO_CONTEXT) {
-		unsigned long flags;
-		int size;
-
-		local_irq_save(flags);
-		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
-		if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
-			mm->context = NO_CONTEXT;
-			if (mm == current->mm)
-				activate_context(mm);
-		} else {
-			unsigned long asid = mm->context&MMU_CONTEXT_ASID_MASK;
-			unsigned long saved_asid = MMU_NO_ASID;
-
-			start &= PAGE_MASK;
-			end += (PAGE_SIZE - 1);
-			end &= PAGE_MASK;
-			if (mm != current->mm) {
-				saved_asid = get_asid();
-				set_asid(asid);
-			}
-			while (start < end) {
-				__flush_tlb_page(asid, start);
-				start += PAGE_SIZE;
-			}
-			if (saved_asid != MMU_NO_ASID)
-				set_asid(saved_asid);
-		}
-		local_irq_restore(flags);
-	}
-}
-
-void flush_tlb_kernel_range(unsigned long start, unsigned long end)
-{
-	unsigned long flags;
-	int size;
-
-	local_irq_save(flags);
-	size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
-	if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
-		flush_tlb_all();
-	} else {
-		unsigned long asid = init_mm.context&MMU_CONTEXT_ASID_MASK;
-		unsigned long saved_asid = get_asid();
-
-		start &= PAGE_MASK;
-		end += (PAGE_SIZE - 1);
-		end &= PAGE_MASK;
-		set_asid(asid);
-		while (start < end) {
-			__flush_tlb_page(asid, start);
-			start += PAGE_SIZE;
-		}
-		set_asid(saved_asid);
-	}
-	local_irq_restore(flags);
-}
-
-void flush_tlb_mm(struct mm_struct *mm)
-{
-	/* Invalidate all TLB of this process. */
-	/* Instead of invalidating each TLB, we get new MMU context. */
-	if (mm->context != NO_CONTEXT) {
-		unsigned long flags;
-
-		local_irq_save(flags);
-		mm->context = NO_CONTEXT;
-		if (mm == current->mm)
-			activate_context(mm);
-		local_irq_restore(flags);
-	}
-}
-
-void flush_tlb_all(void)
-{
-	unsigned long flags, status;
-
-	/*
-	 * Flush all the TLB.
-	 *
-	 * Write to the MMU control register's bit:
-	 *	TF-bit for SH-3, TI-bit for SH-4.
-	 *      It's same position, bit #2.
-	 */
-	local_irq_save(flags);
-	status = ctrl_inl(MMUCR);
-	status |= 0x04;
-	ctrl_outl(status, MMUCR);
-	ctrl_barrier();
-	local_irq_restore(flags);
-}
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 8ea27ca4b700..d1a979eab656 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -80,6 +80,7 @@ void show_mem(void)
 static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 {
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 
@@ -89,7 +90,17 @@ static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 		return;
 	}
 
-	pmd = pmd_offset(pgd, addr);
+	pud = pud_offset(pgd, addr);
+	if (pud_none(*pud)) {
+		pmd = (pmd_t *)get_zeroed_page(GFP_ATOMIC);
+		set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
+		if (pmd != pmd_offset(pud, 0)) {
+			pud_ERROR(*pud);
+			return;
+		}
+	}
+
+	pmd = pmd_offset(pud, addr);
 	if (pmd_none(*pmd)) {
 		pte = (pte_t *)get_zeroed_page(GFP_ATOMIC);
 		set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
diff --git a/arch/sh/mm/tlb-flush.c b/arch/sh/mm/tlb-flush.c
new file mode 100644
index 000000000000..fd7e42bcaa40
--- /dev/null
+++ b/arch/sh/mm/tlb-flush.c
@@ -0,0 +1,132 @@
+/*
+ * TLB flushing operations for SH with an MMU.
+ *
+ *  Copyright (C) 1999  Niibe Yutaka
+ *  Copyright (C) 2003  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/mm.h>
+#include <asm/mmu_context.h>
+#include <asm/tlbflush.h>
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
+{
+	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
+		unsigned long flags;
+		unsigned long asid;
+		unsigned long saved_asid = MMU_NO_ASID;
+
+		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
+		page &= PAGE_MASK;
+
+		local_irq_save(flags);
+		if (vma->vm_mm != current->mm) {
+			saved_asid = get_asid();
+			set_asid(asid);
+		}
+		__flush_tlb_page(asid, page);
+		if (saved_asid != MMU_NO_ASID)
+			set_asid(saved_asid);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+		     unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+		int size;
+
+		local_irq_save(flags);
+		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+		if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
+			mm->context = NO_CONTEXT;
+			if (mm == current->mm)
+				activate_context(mm);
+		} else {
+			unsigned long asid = mm->context&MMU_CONTEXT_ASID_MASK;
+			unsigned long saved_asid = MMU_NO_ASID;
+
+			start &= PAGE_MASK;
+			end += (PAGE_SIZE - 1);
+			end &= PAGE_MASK;
+			if (mm != current->mm) {
+				saved_asid = get_asid();
+				set_asid(asid);
+			}
+			while (start < end) {
+				__flush_tlb_page(asid, start);
+				start += PAGE_SIZE;
+			}
+			if (saved_asid != MMU_NO_ASID)
+				set_asid(saved_asid);
+		}
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	unsigned long flags;
+	int size;
+
+	local_irq_save(flags);
+	size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
+		flush_tlb_all();
+	} else {
+		unsigned long asid = init_mm.context&MMU_CONTEXT_ASID_MASK;
+		unsigned long saved_asid = get_asid();
+
+		start &= PAGE_MASK;
+		end += (PAGE_SIZE - 1);
+		end &= PAGE_MASK;
+		set_asid(asid);
+		while (start < end) {
+			__flush_tlb_page(asid, start);
+			start += PAGE_SIZE;
+		}
+		set_asid(saved_asid);
+	}
+	local_irq_restore(flags);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	/* Invalidate all TLB of this process. */
+	/* Instead of invalidating each TLB, we get new MMU context. */
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+
+		local_irq_save(flags);
+		mm->context = NO_CONTEXT;
+		if (mm == current->mm)
+			activate_context(mm);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_all(void)
+{
+	unsigned long flags, status;
+
+	/*
+	 * Flush all the TLB.
+	 *
+	 * Write to the MMU control register's bit:
+	 *	TF-bit for SH-3, TI-bit for SH-4.
+	 *      It's same position, bit #2.
+	 */
+	local_irq_save(flags);
+	status = ctrl_inl(MMUCR);
+	status |= 0x04;
+	ctrl_outl(status, MMUCR);
+	ctrl_barrier();
+	local_irq_restore(flags);
+}
diff --git a/include/asm-sh/cache.h b/include/asm-sh/cache.h
index 33f13367054b..e3a180cf5062 100644
--- a/include/asm-sh/cache.h
+++ b/include/asm-sh/cache.h
@@ -10,7 +10,6 @@
 #ifdef __KERNEL__
 
 #include <asm/cpu/cache.h>
-#include <asm/cpu/cacheflush.h>
 
 #define SH_CACHE_VALID		1
 #define SH_CACHE_UPDATED	2
@@ -49,12 +48,5 @@ struct cache_info {
 	unsigned long flags;
 };
 
-/* Flush (write-back only) a region (smaller than a page) */
-extern void __flush_wback_region(void *start, int size);
-/* Flush (write-back & invalidate) a region (smaller than a page) */
-extern void __flush_purge_region(void *start, int size);
-/* Flush (invalidate only) a region (smaller than a page) */
-extern void __flush_invalidate_region(void *start, int size);
-
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_CACHE_H */
diff --git a/include/asm-sh/cacheflush.h b/include/asm-sh/cacheflush.h
index 9dfb33edb008..92930b4a40d4 100644
--- a/include/asm-sh/cacheflush.h
+++ b/include/asm-sh/cacheflush.h
@@ -2,6 +2,7 @@
 #define __ASM_SH_CACHEFLUSH_H
 #ifdef __KERNEL__
 
+#include <linux/mm.h>
 #include <asm/cpu/cacheflush.h>
 
 /* Flush (write-back only) a region (smaller than a page) */
diff --git a/include/asm-sh/cpu-sh3/cacheflush.h b/include/asm-sh/cpu-sh3/cacheflush.h
index f51aed00c68f..db0cb071ea8e 100644
--- a/include/asm-sh/cpu-sh3/cacheflush.h
+++ b/include/asm-sh/cpu-sh3/cacheflush.h
@@ -10,7 +10,7 @@
 #ifndef __ASM_CPU_SH3_CACHEFLUSH_H
 #define __ASM_CPU_SH3_CACHEFLUSH_H
 
-/* 
+/*
  * Cache flushing:
  *
  *  - flush_cache_all() flushes entire cache
@@ -35,10 +35,6 @@
  /* 32KB cache, 4kb PAGE sizes need to check bit 12 */
 #define CACHE_ALIAS 0x00001000
 
-struct page;
-struct mm_struct;
-struct vm_area_struct;
-
 extern void flush_cache_all(void);
 extern void flush_cache_mm(struct mm_struct *mm);
 extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
@@ -79,8 +75,6 @@ extern void flush_icache_page(struct vm_area_struct *vma, struct page *page);
 
 #define p3_cache_init()				do { } while (0)
 
-#define HAVE_ARCH_UNMAPPED_AREA
-
 #endif
 
 #endif /* __ASM_CPU_SH3_CACHEFLUSH_H */
diff --git a/include/asm-sh/cpu-sh4/cacheflush.h b/include/asm-sh/cpu-sh4/cacheflush.h
index ea58c4c5944d..a95fc951aff6 100644
--- a/include/asm-sh/cpu-sh4/cacheflush.h
+++ b/include/asm-sh/cpu-sh4/cacheflush.h
@@ -16,30 +16,26 @@
  *  caching; in which case they're only semi-broken),
  *  so we need them.
  */
-struct page;
-struct mm_struct;
-struct vm_area_struct;
-
-extern void flush_cache_all(void);
-extern void flush_cache_mm(struct mm_struct *mm);
-extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
-			      unsigned long end);
-extern void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
-extern void flush_dcache_page(struct page *pg);
+void flush_cache_all(void);
+void flush_cache_mm(struct mm_struct *mm);
+void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
+		       unsigned long end);
+void flush_cache_page(struct vm_area_struct *vma, unsigned long addr,
+		      unsigned long pfn);
+void flush_dcache_page(struct page *pg);
 
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 
-extern void flush_icache_range(unsigned long start, unsigned long end);
-extern void flush_cache_sigtramp(unsigned long addr);
-extern void flush_icache_user_range(struct vm_area_struct *vma,
-				    struct page *page, unsigned long addr,
-				    int len);
+void flush_icache_range(unsigned long start, unsigned long end);
+void flush_cache_sigtramp(unsigned long addr);
+void flush_icache_user_range(struct vm_area_struct *vma, struct page *page,
+			     unsigned long addr, int len);
 
 #define flush_icache_page(vma,pg)		do { } while (0)
 
 /* Initialization of P3 area for copy_user_page */
-extern void p3_cache_init(void);
+void p3_cache_init(void);
 
 #define PG_mapped	PG_arch_1
 
@@ -57,4 +53,3 @@ static inline int remap_area_pages(unsigned long addr, unsigned long phys_addr,
 }
 #endif /* CONFIG_MMU */
 #endif /* __ASM_CPU_SH4_CACHEFLUSH_H */
-
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 4811d410d123..51d7281a546a 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -41,7 +41,8 @@ extern void (*copy_page)(void *to, void *from);
 extern void clear_page_slow(void *to);
 extern void copy_page_slow(void *to, void *from);
 
-#if defined(CONFIG_SH7705_CACHE_32KB) && defined(CONFIG_MMU)
+#if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
+	defined(CONFIG_SH7705_CACHE_32KB))
 struct page;
 extern void clear_user_page(void *to, unsigned long address, struct page *pg);
 extern void copy_user_page(void *to, void *from, unsigned long address, struct page *pg);
@@ -50,29 +51,20 @@ extern void __copy_user_page(void *to, void *from, void *orig_to);
 #elif defined(CONFIG_CPU_SH2) || defined(CONFIG_CPU_SH3) || !defined(CONFIG_MMU)
 #define clear_user_page(page, vaddr, pg)	clear_page(page)
 #define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
-#elif defined(CONFIG_CPU_SH4)
-struct page;
-extern void clear_user_page(void *to, unsigned long address, struct page *pg);
-extern void copy_user_page(void *to, void *from, unsigned long address, struct page *pg);
-extern void __clear_user_page(void *to, void *orig_to);
-extern void __copy_user_page(void *to, void *from, void *orig_to);
 #endif
 
 /*
  * These are used to make use of C type-checking..
  */
 typedef struct { unsigned long pte; } pte_t;
-typedef struct { unsigned long pmd; } pmd_t;
 typedef struct { unsigned long pgd; } pgd_t;
 typedef struct { unsigned long pgprot; } pgprot_t;
 
 #define pte_val(x)	((x).pte)
-#define pmd_val(x)	((x).pmd)
 #define pgd_val(x)	((x).pgd)
 #define pgprot_val(x)	((x).pgprot)
 
 #define __pte(x) ((pte_t) { (x) } )
-#define __pmd(x) ((pmd_t) { (x) } )
 #define __pgd(x) ((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
diff --git a/include/asm-sh/pgalloc.h b/include/asm-sh/pgalloc.h
index f4f233f7a4f5..e841465ab4d2 100644
--- a/include/asm-sh/pgalloc.h
+++ b/include/asm-sh/pgalloc.h
@@ -1,15 +1,6 @@
 #ifndef __ASM_SH_PGALLOC_H
 #define __ASM_SH_PGALLOC_H
 
-#include <linux/threads.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-
-#define pgd_quicklist ((unsigned long *)0)
-#define pmd_quicklist ((unsigned long *)0)
-#define pte_quicklist ((unsigned long *)0)
-#define pgtable_cache_size 0L
-
 #define pmd_populate_kernel(mm, pmd, pte) \
 		set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
 
@@ -24,38 +15,24 @@ static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd,
  */
 static inline pgd_t *pgd_alloc(struct mm_struct *mm)
 {
-	unsigned int pgd_size = (USER_PTRS_PER_PGD * sizeof(pgd_t));
-	pgd_t *pgd = (pgd_t *)kmalloc(pgd_size, GFP_KERNEL);
-
-	if (pgd)
-		memset(pgd, 0, pgd_size);
-
-	return pgd;
+	return (pgd_t *)__get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
 }
 
 static inline void pgd_free(pgd_t *pgd)
 {
-	kfree(pgd);
+	free_page((unsigned long)pgd);
 }
 
 static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
 					  unsigned long address)
 {
-	pte_t *pte;
-
-	pte = (pte_t *) __get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
-
-	return pte;
+	return (pte_t *)__get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
 }
 
 static inline struct page *pte_alloc_one(struct mm_struct *mm,
 					 unsigned long address)
 {
-	struct page *pte;
-
-   	pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
-
-	return pte;
+	return alloc_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
 }
 
 static inline void pte_free_kernel(pte_t *pte)
@@ -75,14 +52,8 @@ static inline void pte_free(struct page *pte)
  * inside the pgd, so has no extra memory associated with it.
  */
 
-#define pmd_alloc_one(mm, addr)		({ BUG(); ((pmd_t *)2); })
 #define pmd_free(x)			do { } while (0)
 #define __pmd_free_tlb(tlb,x)		do { } while (0)
-#define pgd_populate(mm, pmd, pte)	BUG()
 #define check_pgt_cache()		do { } while (0)
 
-#ifdef CONFIG_CPU_SH4
-#define PG_mapped			PG_arch_1
-#endif
-
 #endif /* __ASM_SH_PGALLOC_H */
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 40d41a78041e..9728b58f7c13 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -1,42 +1,42 @@
-#ifndef __ASM_SH_PGTABLE_H
-#define __ASM_SH_PGTABLE_H
-
-#include <asm-generic/4level-fixup.h>
-
 /*
+ * This file contains the functions and defines necessary to modify and
+ * use the SuperH page table tree.
+ *
  * Copyright (C) 1999 Niibe Yutaka
- * Copyright (C) 2002, 2003, 2004 Paul Mundt
+ * Copyright (C) 2002 - 2005 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of this
+ * archive for more details.
  */
+#ifndef __ASM_SH_PGTABLE_H
+#define __ASM_SH_PGTABLE_H
 
-#include <asm/pgtable-2level.h>
+#include <asm-generic/pgtable-nopmd.h>
+#include <asm/page.h>
+
+#define PTRS_PER_PGD		1024
 
-/*
- * This file contains the functions and defines necessary to modify and use
- * the SuperH page table tree.
- */
 #ifndef __ASSEMBLY__
-#include <asm/processor.h>
 #include <asm/addrspace.h>
 #include <asm/fixmap.h>
-#include <linux/threads.h>
 
 extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
 extern void paging_init(void);
 
-/*
- * Basically we have the same two-level (which is the logical three level
- * Linux page table layout folded) page tables as the i386.
- */
-
 /*
  * ZERO_PAGE is a global shared page that is always zero: used
  * for zero-mapped memory areas etc..
  */
-extern unsigned long empty_zero_page[1024];
+extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
 
 #endif /* !__ASSEMBLY__ */
 
+/* traditional two-level paging structure */
+#define PGDIR_SHIFT	22
+#define PTRS_PER_PMD	1
+#define PTRS_PER_PTE	1024
 #define PMD_SIZE	(1UL << PMD_SHIFT)
 #define PMD_MASK	(~(PMD_SIZE-1))
 #define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
@@ -47,7 +47,6 @@ extern unsigned long empty_zero_page[1024];
 
 #define PTE_PHYS_MASK	0x1ffff000
 
-#ifndef __ASSEMBLY__
 /*
  * First 1MB map is used by fixed purpose.
  * Currently only 4-enty (16kB) is used (see arch/sh/mm/cache.c)
@@ -65,7 +64,7 @@ extern unsigned long empty_zero_page[1024];
 #define _PAGE_SZ1	0x080  /* SZ1-bit : Size of page (on SH-4) */
 #define _PAGE_PRESENT	0x100  /* V-bit   : page is valid */
 #define _PAGE_PROTNONE	0x200  /* software: if not present  */
-#define _PAGE_ACCESSED 	0x400  /* software: page referenced */
+#define _PAGE_ACCESSED	0x400  /* software: page referenced */
 #define _PAGE_U0_SHARED 0x800  /* software: page is shared in user space */
 
 #define	_PAGE_FILE	_PAGE_WT  /* software: pagecache or swap? */
@@ -83,7 +82,6 @@ extern unsigned long empty_zero_page[1024];
 #define _PAGE_PCC_ATR8	0x60000000	/* Attribute Memory space, 8 bit bus */
 #define _PAGE_PCC_ATR16	0x60000001	/* Attribute Memory space, 6 bit bus */
 
-
 /* Mask which drop software flags
  * We also drop WT bit since it is used for _PAGE_FILE
  * bit in this implementation.
@@ -115,6 +113,8 @@ extern unsigned long empty_zero_page[1024];
 #define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY | _PAGE_SHARED)
 
+#ifndef __ASSEMBLY__
+
 #ifdef CONFIG_MMU
 #define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_SHARED | _PAGE_FLAGS_HARD)
@@ -137,12 +137,13 @@ extern unsigned long empty_zero_page[1024];
 #define PAGE_KERNEL_PCC		__pgprot(0)
 #endif
 
+#endif /* __ASSEMBLY__ */
+
 /*
  * As i386 and MIPS, SuperH can't do page protection for execute, and
  * considers that the same as a read.  Also, write permissions imply
- * read permissions. This is the closest we can get..  
+ * read permissions. This is the closest we can get..
  */
-
 #define __P000	PAGE_NONE
 #define __P001	PAGE_READONLY
 #define __P010	PAGE_COPY
@@ -161,6 +162,26 @@ extern unsigned long empty_zero_page[1024];
 #define __S110	PAGE_SHARED
 #define __S111	PAGE_SHARED
 
+#ifndef __ASSEMBLY__
+
+/*
+ * Certain architectures need to do special things when PTEs
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+
+/*
+ * (pmds are folded into pgds so this doesn't get actually called,
+ * but the define is needed for a generic inline function.)
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+#define pte_pfn(x)		((unsigned long)(((x).pte >> PAGE_SHIFT)))
+#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+
 #define pte_none(x)	(!pte_val(x))
 #define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
@@ -171,7 +192,7 @@ extern unsigned long empty_zero_page[1024];
 #define	pmd_bad(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
 
 #define pages_to_mb(x)	((x) >> (20-PAGE_SHIFT))
-#define pte_page(x) 	phys_to_page(pte_val(x)&PTE_PHYS_MASK)
+#define pte_page(x)	phys_to_page(pte_val(x)&PTE_PHYS_MASK)
 
 /*
  * The following only work if pte_present() is true.
@@ -248,6 +269,11 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_unmap(pte)		do { } while (0)
 #define pte_unmap_nested(pte)	do { } while (0)
 
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+
 struct vm_area_struct;
 extern void update_mmu_cache(struct vm_area_struct * vma,
 			     unsigned long address, pte_t pte);
@@ -272,8 +298,6 @@ extern void update_mmu_cache(struct vm_area_struct * vma,
 
 typedef pte_t *pte_addr_t;
 
-#endif /* !__ASSEMBLY__ */
-
 #define kern_addr_valid(addr)	(1)
 
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
@@ -301,5 +325,7 @@ extern pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t
 
 #include <asm-generic/pgtable.h>
 
+#endif /* !__ASSEMBLY__ */
+
 #endif /* __ASM_SH_PAGE_H */
 
-- 
cgit v1.2.3


From d7cdc9e8ac82c43fdcd4fde6b5b53d2dcba7f707 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:16:42 +0900
Subject: sh: ioremap() overhaul.

ioremap() overhaul. Add support for transparent PMB mapping, get rid of
p3_ioremap(), etc. Also drop ioremap() and iounmap() routines from the
machvec, as everyone can use the generic ioremap() API instead. For PCI
memory apertures and other special cases, use the pci_iomap() API, as
boards are already required to get the mapping right there.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/landisk/io.c                | 236 ++++++++++++----------------
 arch/sh/boards/landisk/setup.c             | 237 ++++++++++++-----------------
 arch/sh/boards/renesas/hs7751rvoip/io.c    |   9 --
 arch/sh/boards/renesas/hs7751rvoip/setup.c |   1 -
 arch/sh/boards/renesas/rts7751r2d/io.c     |   9 --
 arch/sh/boards/renesas/rts7751r2d/mach.c   |   2 -
 arch/sh/boards/titan/io.c                  |   9 +-
 arch/sh/boards/titan/setup.c               |   1 -
 arch/sh/drivers/pci/pci.c                  |  21 ++-
 arch/sh/kernel/cf-enabler.c                |   5 +-
 arch/sh/kernel/sh_ksyms.c                  |   1 -
 arch/sh/mm/Makefile                        |   7 +-
 arch/sh/mm/pmb.c                           | 145 +++++++++++++++++-
 include/asm-sh/landisk/iodata_landisk.h    |   3 +
 include/asm-sh/mmu.h                       |  11 +-
 15 files changed, 374 insertions(+), 323 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boards/landisk/io.c b/arch/sh/boards/landisk/io.c
index 1f1679af09d0..aa6b145c9e8f 100644
--- a/arch/sh/boards/landisk/io.c
+++ b/arch/sh/boards/landisk/io.c
@@ -17,9 +17,9 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <asm/io.h>
 #include <asm/landisk/iodata_landisk.h>
 #include <asm/addrspace.h>
+#include <asm/io.h>
 
 #include <linux/module.h>
 #include <linux/pci.h>
@@ -42,10 +42,6 @@ extern void *area6_io_base;	/* Area 6 I/O Base address */
 
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -66,7 +62,7 @@ static inline unsigned long port2adr(unsigned int port)
 			return ((unsigned long)area6_io_base + PA_SIDE_OFFSET +
 				((port - 0x170) << 1));
 	else
-		maybebadio(port2adr, (unsigned long)port);
+		maybebadio((unsigned long)port);
 
 	return port;
 }
@@ -89,234 +85,200 @@ static inline unsigned long port2adr(unsigned int port)
  * should be way beyond the window, and is used  w/o translation for
  * compatibility.
  */
-unsigned char landisk_inb(unsigned long port)
+u8 landisk_inb(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned char *)port;
+		return ctrl_inb(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
-	else
-		return (*(volatile unsigned short *)port2adr(port) & 0xff);
+		return ctrl_inb(PCI_IOMAP(port));
+
+	return ctrl_inw(port2adr(port)) & 0xff;
 }
 
-unsigned char landisk_inb_p(unsigned long port)
+u8 landisk_inb_p(unsigned long port)
 {
-	unsigned char v;
+	u8 v;
 
 	if (PXSEG(port))
-		v = *(volatile unsigned char *)port;
+		v = ctrl_inb(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		v = *(volatile unsigned char *)PCI_IOMAP(port);
+		v = ctrl_inb(PCI_IOMAP(port));
 	else
-		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
+		v = ctrl_inw(port2adr(port)) & 0xff;
+
 	delay();
 
 	return v;
 }
 
-unsigned short landisk_inw(unsigned long port)
+u16 landisk_inw(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned short *)port;
+		return ctrl_inw(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned short *)PCI_IOMAP(port);
+		return ctrl_inw(PCI_IOMAP(port));
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 
 	return 0;
 }
 
-unsigned int landisk_inl(unsigned long port)
+u32 landisk_inl(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned long *)port;
+		return ctrl_inl(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned long *)PCI_IOMAP(port);
+		return ctrl_inl(PCI_IOMAP(port));
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 
 	return 0;
 }
 
-void landisk_outb(unsigned char value, unsigned long port)
+void landisk_outb(u8 value, unsigned long port)
 {
-
 	if (PXSEG(port))
-		*(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outw(value, port2adr(port));
 }
 
-void landisk_outb_p(unsigned char value, unsigned long port)
+void landisk_outb_p(u8 value, unsigned long port)
 {
 	if (PXSEG(port))
-		*(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outw(value, port2adr(port));
 	delay();
 }
 
-void landisk_outw(unsigned short value, unsigned long port)
+void landisk_outw(u16 value, unsigned long port)
 {
 	if (PXSEG(port))
-		*(volatile unsigned short *)port = value;
+		ctrl_outw(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned short *)PCI_IOMAP(port) = value;
+		ctrl_outw(value, PCI_IOMAP(port));
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
-void landisk_outl(unsigned int value, unsigned long port)
+void landisk_outl(u32 value, unsigned long port)
 {
 	if (PXSEG(port))
-		*(volatile unsigned long *)port = value;
+		ctrl_outl(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned long *)PCI_IOMAP(port) = value;
+		ctrl_outl(value, PCI_IOMAP(port));
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
-void landisk_insb(unsigned long port, void *addr, unsigned long count)
+void landisk_insb(unsigned long port, void *dst, unsigned long count)
 {
-	if (PXSEG(port))
-		while (count--)
-			*((unsigned char *)addr)++ =
-			    *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+        volatile u16 *p;
+        u8 *buf = dst;
 
-		while (count--)
-			*((volatile unsigned char *)addr)++ = *bp;
-	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+        if (PXSEG(port)) {
+                while (count--)
+                        *buf++ = *(volatile u8 *)port;
+	} else if (CHECK_SH7751_PCIIO(port)) {
+                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--)
-			*((unsigned char *)addr)++ = *p;
+                while (count--)
+                        *buf++ = *bp;
+	} else {
+                p = (volatile u16 *)port2adr(port);
+                while (count--)
+                        *buf++ = *p & 0xff;
 	}
 }
 
-void landisk_insw(unsigned long port, void *addr, unsigned long count)
+void landisk_insw(unsigned long port, void *dst, unsigned long count)
 {
-	volatile __u16 *p;
+        volatile u16 *p;
+        u16 *buf = dst;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+		p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+		p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
+		p = (volatile u16 *)port2adr(port);
 	while (count--)
-		*((__u16 *) addr)++ = *p;
+		*buf++ = *p;
 }
 
-void landisk_insl(unsigned long port, void *addr, unsigned long count)
+void landisk_insl(unsigned long port, void *dst, unsigned long count)
 {
+        u32 *buf = dst;
+
 	if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
+                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
 
-		while (count--)
-			*((__u32 *) addr)++ = *p;
+                while (count--)
+                        *buf++ = *p;
 	} else
-		maybebadio(insl, port);
+		maybebadio(port);
 }
 
-void landisk_outsb(unsigned long port, const void *addr, unsigned long count)
+void landisk_outsb(unsigned long port, const void *src, unsigned long count)
 {
+        volatile u16 *p;
+        const u8 *buf = src;
+
 	if (PXSEG(port))
-		while (count--)
-			*(volatile unsigned char *)port =
-			    *((unsigned char *)addr)++;
+                while (count--)
+                        ctrl_outb(*buf++, port);
 	else if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--)
-			*bp = *((volatile unsigned char *)addr)++;
+                while (count--)
+                        *bp = *buf++;
 	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
-
-		while (count--)
-			*p = *((unsigned char *)addr)++;
+                p = (volatile u16 *)port2adr(port);
+                while (count--)
+                        *p = *buf++;
 	}
 }
 
-void landisk_outsw(unsigned long port, const void *addr, unsigned long count)
+void landisk_outsw(unsigned long port, const void *src, unsigned long count)
 {
-	volatile __u16 *p;
+        volatile u16 *p;
+        const u16 *buf = src;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+                p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+                p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
-	while (count--)
-		*p = *((__u16 *) addr)++;
-}
+                p = (volatile u16 *)port2adr(port);
 
-void landisk_outsl(unsigned long port, const void *addr, unsigned long count)
-{
-	if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
-
-		while (count--)
-			*p = *((__u32 *) addr)++;
-	} else
-		maybebadio(outsl, port);
+        while (count--)
+                *p = *buf++;
 }
 
-/* For read/write calls, just copy generic (pass-thru); PCIMBR is  */
-/* already set up.  For a larger memory space, these would need to */
-/* reset PCIMBR as needed on a per-call basis...                   */
-
-unsigned char landisk_readb(unsigned long addr)
+void landisk_outsl(unsigned long port, const void *src, unsigned long count)
 {
-	return *(volatile unsigned char *)addr;
-}
+        const u32 *buf = src;
 
-unsigned short landisk_readw(unsigned long addr)
-{
-	return *(volatile unsigned short *)addr;
-}
-
-unsigned int landisk_readl(unsigned long addr)
-{
-	return *(volatile unsigned long *)addr;
-}
-
-void landisk_writeb(unsigned char b, unsigned long addr)
-{
-	*(volatile unsigned char *)addr = b;
-}
-
-void landisk_writew(unsigned short b, unsigned long addr)
-{
-	*(volatile unsigned short *)addr = b;
-}
-
-void landisk_writel(unsigned int b, unsigned long addr)
-{
-	*(volatile unsigned long *)addr = b;
-}
+	if (CHECK_SH7751_PCIIO(port)) {
+                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
 
-void *landisk_ioremap(unsigned long offset, unsigned long size)
-{
-	if (offset >= 0xfd000000)
-		return (void *)offset;
-	else
-		return (void *)P2SEGADDR(offset);
+                while (count--)
+                        *p = *buf++;
+	} else
+		maybebadio(port);
 }
 
-void landisk_iounmap(void *addr)
+void __iomem *landisk_ioport_map(unsigned long port, unsigned int size)
 {
-}
+        if (PXSEG(port))
+                return (void __iomem *)port;
+        else if (CHECK_SH7751_PCIIO(port))
+                return (void __iomem *)PCI_IOMAP(port);
 
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-unsigned long landisk_isa_port2addr(unsigned long offset)
-{
-	return port2adr(offset);
+        return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
index 0c60eaa10ba7..3a795cfb1eda 100644
--- a/arch/sh/boards/landisk/setup.c
+++ b/arch/sh/boards/landisk/setup.c
@@ -1,157 +1,52 @@
 /*
  * arch/sh/boards/landisk/setup.c
  *
+ * Copyright (C) 2000 Kazumoto Kojima
  * Copyright (C) 2002 Paul Mundt
  *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Setup code for an unknown machine (internal peripherials only)
- */
-/*
- * linux/arch/sh/kernel/setup_landisk.c
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
  * I-O DATA Device, Inc. LANDISK Support.
  *
  * Modified for LANDISK by
  * Atom Create Engineering Co., Ltd. 2002.
- */
-/*
+ *
  * modifed by kogiidena
  * 2005.09.16
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/config.h>
 #include <linux/init.h>
-#include <linux/irq.h>
 #include <linux/pm.h>
-
-#include <linux/hdreg.h>
-#include <linux/ide.h>
-#include <linux/pci.h>
-
+#include <linux/mm.h>
 #include <asm/machvec.h>
 #include <asm/rtc.h>
-#include <asm/machvec_init.h>
-#include <asm/io.h>
 #include <asm/landisk/iodata_landisk.h>
-#include <asm/landisk/io.h>
-
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
+#include <asm/io.h>
 
-extern void (*board_time_init) (void);
 void landisk_time_init(void);
-extern void init_landisk_IRQ(void);
+void init_landisk_IRQ(void);
 
 int landisk_ledparam;
 int landisk_buzzerparam;
 int landisk_arch;
 
-/* defined in mm/ioremap.c */
-extern void *p3_ioremap(unsigned long phys_addr, unsigned long size,
-			unsigned long flags);
-
-/*
- * Initialize the board
- */
-
-const char *get_system_type(void)
-{
-	return "LANDISK";
-}
-
-static void landisk_power_off(void)
-{
-	ctrl_outb(0x01, PA_SHUTDOWN);
-}
-
-void check_usl5p(void)
-{
-	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
-	unsigned char tmp1, tmp2;
-	tmp1 = *p;
-	*p = 0x40;
-	tmp2 = *p;
-	*p = tmp1;
-	landisk_arch = (tmp2 == 0x40) ? 1 : 0;
-	if (landisk_arch == 1) {	/* arch == usl-5p */
-		landisk_ledparam = 0x00000380;
-		landisk_ledparam |= (tmp1 & 0x07c);
-	} else {                        /* arch == landisk */
-		landisk_ledparam = 0x02000180;
-		landisk_ledparam |= 0x04;
-	}
-	return;
-}
-
-void __init platform_setup(void)
-{
-
-	landisk_buzzerparam = 0;
-	check_usl5p();
-
-	printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
-	board_time_init = landisk_time_init;
-	pm_power_off = landisk_power_off;
-
-}
-
-void *area5_io_base;
-void *area6_io_base;
-
-int __init cf_init(void)
-{
-	pgprot_t prot;
-	unsigned long paddrbase, psize;
-
-	/* open I/O area window */
-	paddrbase = virt_to_phys((void *)PA_AREA5_IO);
-	psize = PAGE_SIZE;
-	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_IO16);
-	area5_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
-	if (!area5_io_base) {
-		printk("allocate_cf_area : can't open CF I/O window!\n");
-		return -ENOMEM;
-	}
-
-	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
-	psize = PAGE_SIZE;
-	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
-	area6_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
-	if (!area6_io_base) {
-		printk("allocate_cf_area : can't open HDD I/O window!\n");
-		return -ENOMEM;
-	}
-
-	printk(KERN_INFO "Allocate Area5/6 success.\n");
-
-	/* XXX : do we need attribute and common-memory area also? */
-
-	return 0;
-}
-
-__initcall(cf_init);
-
-#include <linux/sched.h>
-
-/* Cycle the LED's in the clasic knightrider/Sun pattern */
-
-void heartbeat_landisk(void)
+/* cycle the led's in the clasic knightrider/sun pattern */
+static void heartbeat_landisk(void)
 {
 	static unsigned int cnt = 0, blink = 0x00, period = 25;
-	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
+        volatile u8 *p = (volatile u8 *)PA_LED;
 	char data;
 
-	if ((landisk_ledparam & 0x080) == 0) {
+        if ((landisk_ledparam & 0x080) == 0)
 		return;
-	}
+
 	cnt += 1;
-	if (cnt < period) {
+
+        if (cnt < period)
 		return;
-	}
+
 	cnt = 0;
 	blink++;
 
@@ -167,17 +62,16 @@ void heartbeat_landisk(void)
 	}
 	*p = data;
 
-	if (((landisk_ledparam & 0x007f7f00) == 0)
-	    && (landisk_buzzerparam == 0)) {
+        if (((landisk_ledparam & 0x007f7f00) == 0) &&
+             (landisk_buzzerparam == 0))
 		landisk_ledparam &= (~0x0080);
-	}
+
 	landisk_buzzerparam >>= 1;
 }
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_landisk __initmv = {
 	.mv_nr_irqs = 72,
 	.mv_inb = landisk_inb,
@@ -198,21 +92,88 @@ struct sh_machine_vector mv_landisk __initmv = {
 	.mv_outsb = landisk_outsb,
 	.mv_outsw = landisk_outsw,
 	.mv_outsl = landisk_outsl,
-	.mv_readb = landisk_readb,
-	.mv_readw = landisk_readw,
-	.mv_readl = landisk_readl,
-	.mv_writeb = landisk_writeb,
-	.mv_writew = landisk_writew,
-	.mv_writel = landisk_writel,
-	.mv_ioremap = landisk_ioremap,
-	.mv_iounmap = landisk_iounmap,
-	.mv_isa_port2addr = landisk_isa_port2addr,
+        .mv_ioport_map = landisk_ioport_map,
 	.mv_init_irq = init_landisk_IRQ,
-
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat = heartbeat_landisk,
 #endif
-
 };
-
 ALIAS_MV(landisk)
+
+const char *get_system_type(void)
+{
+        return "LANDISK";
+}
+
+static void landisk_power_off(void)
+{
+        ctrl_outb(0x01, PA_SHUTDOWN);
+}
+
+static void check_usl5p(void)
+{
+        volatile u8 *p = (volatile u8 *)PA_LED;
+        u8 tmp1, tmp2;
+
+        tmp1 = *p;
+        *p = 0x40;
+        tmp2 = *p;
+        *p = tmp1;
+
+        landisk_arch = (tmp2 == 0x40);
+        if (landisk_arch == 1) {
+                /* arch == usl-5p */
+                landisk_ledparam = 0x00000380;
+                landisk_ledparam |= (tmp1 & 0x07c);
+        } else {
+                /* arch == landisk */
+                landisk_ledparam = 0x02000180;
+                landisk_ledparam |= 0x04;
+        }
+}
+
+void __init platform_setup(void)
+{
+        landisk_buzzerparam = 0;
+        check_usl5p();
+
+        printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
+        board_time_init = landisk_time_init;
+        pm_power_off = landisk_power_off;
+}
+
+void *area5_io_base;
+void *area6_io_base;
+
+static int __init landisk_cf_init(void)
+{
+	pgprot_t prot;
+	unsigned long paddrbase, psize;
+
+	/* open I/O area window */
+	paddrbase = virt_to_phys((void *)PA_AREA5_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_IO16);
+	area5_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area5_io_base) {
+		printk("allocate_cf_area : can't open CF I/O window!\n");
+		return -ENOMEM;
+	}
+
+	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
+	area6_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area6_io_base) {
+		printk("allocate_cf_area : can't open HDD I/O window!\n");
+		return -ENOMEM;
+	}
+
+	printk(KERN_INFO "Allocate Area5/6 success.\n");
+
+	/* XXX : do we need attribute and common-memory area also? */
+
+	return 0;
+}
+
+__initcall(landisk_cf_init);
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index 09fb77ffb835..edecf107fc13 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -294,15 +294,6 @@ void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count
 		maybebadio(outsl, port);
 }
 
-void *hs7751rvoip_ioremap(unsigned long offset, unsigned long size)
-{
-	if (offset >= 0xfd000000)
-		return (void *)offset;
-	else
-		return (void *)P2SEGADDR(offset);
-}
-EXPORT_SYMBOL(hs7751rvoip_ioremap);
-
 unsigned long hs7751rvoip_isa_port2addr(unsigned long offset)
 {
 	return port2adr(offset);
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index 813fc4d5862a..aa51bda931f6 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -60,7 +60,6 @@ struct sh_machine_vector mv_hs7751rvoip __initmv = {
 	.mv_outsw		= hs7751rvoip_outsw,
 	.mv_outsl		= hs7751rvoip_outsl,
 
-	.mv_ioremap		= hs7751rvoip_ioremap,
 	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
 	.mv_init_irq		= hs7751rvoip_init_irq,
 };
diff --git a/arch/sh/boards/renesas/rts7751r2d/io.c b/arch/sh/boards/renesas/rts7751r2d/io.c
index 9e7fa726a86d..8dc2a2e2e5df 100644
--- a/arch/sh/boards/renesas/rts7751r2d/io.c
+++ b/arch/sh/boards/renesas/rts7751r2d/io.c
@@ -321,15 +321,6 @@ void rts7751r2d_outsl(unsigned long port, const void *addr, unsigned long count)
 		maybebadio(port);
 }
 
-void *rts7751r2d_ioremap(unsigned long offset, unsigned long size)
-{
-	if (offset >= 0xfd000000)
-		return (void *)offset;
-	else
-		return (void *)P2SEGADDR(offset);
-}
-EXPORT_SYMBOL(rts7751r2d_ioremap);
-
 unsigned long rts7751r2d_isa_port2addr(unsigned long offset)
 {
 	return port2adr(offset);
diff --git a/arch/sh/boards/renesas/rts7751r2d/mach.c b/arch/sh/boards/renesas/rts7751r2d/mach.c
index 175a93d726e8..fe3e8735e9f8 100644
--- a/arch/sh/boards/renesas/rts7751r2d/mach.c
+++ b/arch/sh/boards/renesas/rts7751r2d/mach.c
@@ -19,7 +19,6 @@
 
 extern void heartbeat_rts7751r2d(void);
 extern void init_rts7751r2d_IRQ(void);
-extern void *rts7751r2d_ioremap(unsigned long, unsigned long);
 extern int rts7751r2d_irq_demux(int irq);
 
 extern void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
@@ -53,7 +52,6 @@ struct sh_machine_vector mv_rts7751r2d __initmv = {
 	.mv_outsw		= rts7751r2d_outsw,
 	.mv_outsl		= rts7751r2d_outsl,
 
-	.mv_ioremap		= rts7751r2d_ioremap,
 	.mv_init_irq		= init_rts7751r2d_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_rts7751r2d,
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
index d66900c99a11..b886fd233a66 100644
--- a/arch/sh/boards/titan/io.c
+++ b/arch/sh/boards/titan/io.c
@@ -138,19 +138,12 @@ void titan_outsl(unsigned long port, const void *src, unsigned long count)
         maybebadio(port);
 }
 
-void *titan_ioremap(unsigned long offset, unsigned long size) {
-	if (CHECK_SH7751_PCIIO(offset) || CHECK_SH7751_PCIMEMIO(offset))
-		return (void *)offset;
-}
-
 void __iomem *titan_ioport_map(unsigned long port, unsigned int size)
 {
-	if (PXSEG(port))
+	if (PXSEG(port) || CHECK_SH7751_PCIMEMIO(port))
 		return (void __iomem *)port;
 	else if (CHECK_SH7751_PCIIO(port))
 		return (void __iomem *)PCI_IOMAP(port);
 
 	return (void __iomem *)port2adr(port);
 }
-
-EXPORT_SYMBOL(titan_ioremap);
diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c
index 6ac5c8d7b3fb..c8b431c1d0fd 100644
--- a/arch/sh/boards/titan/setup.c
+++ b/arch/sh/boards/titan/setup.c
@@ -51,7 +51,6 @@ struct sh_machine_vector mv_titan __initmv = {
 	.mv_insl =	titan_insl,
 	.mv_outsl =	titan_outsl,
 
-	.mv_ioremap =	titan_ioremap,
 	.mv_ioport_map = titan_ioport_map,
 
 	.mv_init_irq =	init_titan_irq,
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 1f5e23e8b163..285dffd12bd8 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -2,7 +2,7 @@
  * arch/sh/drivers/pci/pci.c
  *
  * Copyright (c) 2002 M. R. Brown  <mrbrown@linux-sh.org>
- * Copyright (c) 2004, 2005 Paul Mundt  <lethal@linux-sh.org>
+ * Copyright (c) 2004 - 2006 Paul Mundt  <lethal@linux-sh.org>
  *
  * These functions are collected here to reduce duplication of common
  * code amongst the many platform-specific PCI support code files.
@@ -172,10 +172,23 @@ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 		return NULL;
 	if (maxlen && len > maxlen)
 		len = maxlen;
-	if (flags & IORESOURCE_IO)
+
+	/*
+	 * Presently the IORESOURCE_MEM case is a bit special, most
+	 * SH7751 style PCI controllers have PCI memory at a fixed
+	 * location in the address space where no remapping is desired
+	 * (traditionally at 0xfd000000). Once this changes, the
+	 * IORESOURCE_MEM case will have to switch to using ioremap() and
+	 * more care will have to be taken to inhibit page table mapping
+	 * for legacy cores.
+	 *
+	 * For now everything wraps to ioport_map(), since boards that
+	 * have PCI will be able to check the address range properly on
+	 * their own.
+	 *			-- PFM.
+	 */
+	if (flags & (IORESOURCE_IO | IORESOURCE_MEM))
 		return ioport_map(start, len);
-	if (flags & IORESOURCE_MEM)
-		return ioremap(start, len);
 
 	return NULL;
 }
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index f1f9ab87f0b0..c9b823d1d073 100644
--- a/arch/sh/kernel/cf-enabler.c
+++ b/arch/sh/kernel/cf-enabler.c
@@ -41,9 +41,6 @@
 #define slot_no 1
 #endif
 
-/* defined in mm/ioremap.c */
-extern void * p3_ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags);
-
 /* use this pointer to access to directly connected compact flash io area*/
 void *cf_io_base;
 
@@ -62,7 +59,7 @@ static int __init allocate_cf_area(void)
 		return -ENOMEM;
 	}
 /*	printk("p3_ioremap(paddr=0x%08lx, psize=0x%08lx, prot=0x%08lx)=0x%08lx\n",
-	    	paddrbase, psize, prot.pgprot, cf_io_base);*/
+		paddrbase, psize, prot.pgprot, cf_io_base);*/
 
 	/* XXX : do we need attribute and common-memory area also? */
 
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 8a6dd06fd071..b73feafcd06e 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -27,7 +27,6 @@ EXPORT_SYMBOL(sh_mv);
 
 /* platform dependent support */
 EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL(iounmap);
 EXPORT_SYMBOL(enable_irq);
 EXPORT_SYMBOL(disable_irq);
 EXPORT_SYMBOL(probe_irq_mask);
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index d90906367c5f..87a7c07265c0 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -12,13 +12,14 @@ obj-$(CONFIG_DMA_PAGE_OPS)	+= pg-dma.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
 
 mmu-y			:= fault-nommu.o tlb-nommu.o pg-nommu.o
-mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o
+mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o	\
+			   ioremap.o
 
 obj-y			+= $(mmu-y)
 
 ifdef CONFIG_MMU
-obj-$(CONFIG_CPU_SH3)	+= tlb-sh3.o
-obj-$(CONFIG_CPU_SH4)	+= tlb-sh4.o ioremap.o
+obj-$(CONFIG_CPU_SH3)		+= tlb-sh3.o
+obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o
 obj-$(CONFIG_SH7705_CACHE_32KB) += pg-sh7705.o
 endif
 
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index ff5bde745647..819fd0faf022 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -3,7 +3,7 @@
  *
  * Privileged Space Mapping Buffer (PMB) Support.
  *
- * Copyright (C) 2005 Paul Mundt
+ * Copyright (C) 2005, 2006 Paul Mundt
  *
  * P1/P2 Section mapping definitions from map32.h, which was:
  *
@@ -24,6 +24,7 @@
 #include <linux/err.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
+#include <asm/pgtable.h>
 #include <asm/mmu.h>
 #include <asm/io.h>
 
@@ -127,11 +128,15 @@ repeat:
 	return 0;
 }
 
-void set_pmb_entry(struct pmb_entry *pmbe)
+int set_pmb_entry(struct pmb_entry *pmbe)
 {
+	int ret;
+
 	jump_to_P2();
-	__set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
+	ret = __set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
 	back_to_P1();
+
+	return ret;
 }
 
 void clear_pmb_entry(struct pmb_entry *pmbe)
@@ -162,11 +167,141 @@ void clear_pmb_entry(struct pmb_entry *pmbe)
 	clear_bit(entry, &pmb_map);
 }
 
+static DEFINE_SPINLOCK(pmb_list_lock);
+static struct pmb_entry *pmb_list;
+
+static inline void pmb_list_add(struct pmb_entry *pmbe)
+{
+	struct pmb_entry **p, *tmp;
+
+	p = &pmb_list;
+	while ((tmp = *p) != NULL)
+		p = &tmp->next;
+
+	pmbe->next = tmp;
+	*p = pmbe;
+}
+
+static inline void pmb_list_del(struct pmb_entry *pmbe)
+{
+	struct pmb_entry **p, *tmp;
+
+	for (p = &pmb_list; (tmp = *p); p = &tmp->next)
+		if (tmp == pmbe) {
+			*p = tmp->next;
+			return;
+		}
+}
+
+static struct {
+	unsigned long size;
+	int flag;
+} pmb_sizes[] = {
+	{ .size	= 0x20000000, .flag = PMB_SZ_512M, },
+	{ .size = 0x08000000, .flag = PMB_SZ_128M, },
+	{ .size = 0x04000000, .flag = PMB_SZ_64M,  },
+	{ .size = 0x01000000, .flag = PMB_SZ_16M,  },
+};
+
+long pmb_remap(unsigned long vaddr, unsigned long phys,
+	       unsigned long size, unsigned long flags)
+{
+	struct pmb_entry *pmbp;
+	unsigned long wanted;
+	int pmb_flags, i;
+
+	/* Convert typical pgprot value to the PMB equivalent */
+	if (flags & _PAGE_CACHABLE) {
+		if (flags & _PAGE_WT)
+			pmb_flags = PMB_WT;
+		else
+			pmb_flags = PMB_C;
+	} else
+		pmb_flags = PMB_WT | PMB_UB;
+
+	pmbp = NULL;
+	wanted = size;
+
+again:
+	for (i = 0; i < ARRAY_SIZE(pmb_sizes); i++) {
+		struct pmb_entry *pmbe;
+		int ret;
+
+		if (size < pmb_sizes[i].size)
+			continue;
+
+		pmbe = pmb_alloc(vaddr, phys, pmb_flags | pmb_sizes[i].flag);
+		if (IS_ERR(pmbe))
+			return PTR_ERR(pmbe);
+
+		ret = set_pmb_entry(pmbe);
+		if (ret != 0) {
+			pmb_free(pmbe);
+			return -EBUSY;
+		}
+
+		phys	+= pmb_sizes[i].size;
+		vaddr	+= pmb_sizes[i].size;
+		size	-= pmb_sizes[i].size;
+
+		/*
+		 * Link adjacent entries that span multiple PMB entries
+		 * for easier tear-down.
+		 */
+		if (likely(pmbp))
+			pmbp->link = pmbe;
+
+		pmbp = pmbe;
+	}
+
+	if (size >= 0x1000000)
+		goto again;
+
+	return wanted - size;
+}
+
+void pmb_unmap(unsigned long addr)
+{
+	struct pmb_entry **p, *pmbe;
+
+	for (p = &pmb_list; (pmbe = *p); p = &pmbe->next)
+		if (pmbe->vpn == addr)
+			break;
+
+	if (unlikely(!pmbe))
+		return;
+
+	WARN_ON(!test_bit(pmbe->entry, &pmb_map));
+
+	do {
+		struct pmb_entry *pmblink = pmbe;
+
+		clear_pmb_entry(pmbe);
+		pmbe = pmblink->link;
+
+		pmb_free(pmblink);
+	} while (pmbe);
+}
+
 static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags)
 {
+	struct pmb_entry *pmbe = pmb;
+
 	memset(pmb, 0, sizeof(struct pmb_entry));
 
-	((struct pmb_entry *)pmb)->entry = PMB_NO_ENTRY;
+	spin_lock_irq(&pmb_list_lock);
+
+	pmbe->entry = PMB_NO_ENTRY;
+	pmb_list_add(pmbe);
+
+	spin_unlock_irq(&pmb_list_lock);
+}
+
+static void pmb_cache_dtor(void *pmb, kmem_cache_t *cachep, unsigned long flags)
+{
+	spin_lock_irq(&pmb_list_lock);
+	pmb_list_del(pmb);
+	spin_unlock_irq(&pmb_list_lock);
 }
 
 static int __init pmb_init(void)
@@ -177,7 +312,7 @@ static int __init pmb_init(void)
 	BUG_ON(unlikely(nr_entries >= NR_PMB_ENTRIES));
 
 	pmb_cache = kmem_cache_create("pmb", sizeof(struct pmb_entry),
-				      0, 0, pmb_cache_ctor, NULL);
+				      0, 0, pmb_cache_ctor, pmb_cache_dtor);
 	BUG_ON(!pmb_cache);
 
 	jump_to_P2();
diff --git a/include/asm-sh/landisk/iodata_landisk.h b/include/asm-sh/landisk/iodata_landisk.h
index 7189d3a36384..9db3cdfe6776 100644
--- a/include/asm-sh/landisk/iodata_landisk.h
+++ b/include/asm-sh/landisk/iodata_landisk.h
@@ -74,5 +74,8 @@ extern int landisk_ledparam;    /* from setup.c */
 extern int landisk_buzzerparam; /* from setup.c */
 extern int landisk_arch;        /* from setup.c */
 
+#define __IO_PREFIX landisk
+#include <asm/io_generic.h>
+
 #endif  /* __ASM_SH_IODATA_LANDISK_H */
 
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index 91c884634276..ec09589fa6ca 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -50,6 +50,8 @@ typedef unsigned long mm_context_t;
 
 #define PMB_NO_ENTRY		(-1)
 
+struct pmb_entry;
+
 struct pmb_entry {
 	unsigned long vpn;
 	unsigned long ppn;
@@ -60,16 +62,23 @@ struct pmb_entry {
 	 * PMB_NO_ENTRY to search for a free one
 	 */
 	int entry;
+
+	struct pmb_entry *next;
+	/* Adjacent entry link for contiguous multi-entry mappings */
+	struct pmb_entry *link;
 };
 
 /* arch/sh/mm/pmb.c */
 int __set_pmb_entry(unsigned long vpn, unsigned long ppn,
 		    unsigned long flags, int *entry);
-void set_pmb_entry(struct pmb_entry *pmbe);
+int set_pmb_entry(struct pmb_entry *pmbe);
 void clear_pmb_entry(struct pmb_entry *pmbe);
 struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
 			    unsigned long flags);
 void pmb_free(struct pmb_entry *pmbe);
+long pmb_remap(unsigned long virt, unsigned long phys,
+	       unsigned long size, unsigned long flags);
+void pmb_unmap(unsigned long addr);
 
 #endif /* __MMU_H */
 
-- 
cgit v1.2.3


From c470662854ff94d44bf8c192cefac3efa33db676 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:29:18 +0900
Subject: sh: Fixup SHMLBA definition for SH7705.

We need this set to something sensible anywhere were we have
an aliasing dcache..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh2/shmparam.h   | 16 -------------
 include/asm-sh/cpu-sh3/cacheflush.h | 48 ++++++++++++++++---------------------
 include/asm-sh/cpu-sh3/shmparam.h   | 16 -------------
 include/asm-sh/cpu-sh4/shmparam.h   | 19 ---------------
 include/asm-sh/shmparam.h           | 20 +++++++++++++---
 5 files changed, 38 insertions(+), 81 deletions(-)
 delete mode 100644 include/asm-sh/cpu-sh2/shmparam.h
 delete mode 100644 include/asm-sh/cpu-sh3/shmparam.h
 delete mode 100644 include/asm-sh/cpu-sh4/shmparam.h

(limited to 'include')

diff --git a/include/asm-sh/cpu-sh2/shmparam.h b/include/asm-sh/cpu-sh2/shmparam.h
deleted file mode 100644
index 817c1821ee4b..000000000000
--- a/include/asm-sh/cpu-sh2/shmparam.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- * include/asm-sh/cpu-sh2/shmparam.h
- *
- * Copyright (C) 2003  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_CPU_SH2_SHMPARAM_H
-#define __ASM_CPU_SH2_SHMPARAM_H
-
-#define	SHMLBA PAGE_SIZE		 /* attach addr a multiple of this */
-
-#endif /* __ASM_CPU_SH2_SHMPARAM_H */
-
diff --git a/include/asm-sh/cpu-sh3/cacheflush.h b/include/asm-sh/cpu-sh3/cacheflush.h
index db0cb071ea8e..97f5a64c2ab8 100644
--- a/include/asm-sh/cpu-sh3/cacheflush.h
+++ b/include/asm-sh/cpu-sh3/cacheflush.h
@@ -35,47 +35,41 @@
  /* 32KB cache, 4kb PAGE sizes need to check bit 12 */
 #define CACHE_ALIAS 0x00001000
 
-extern void flush_cache_all(void);
-extern void flush_cache_mm(struct mm_struct *mm);
-extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
-                              unsigned long end);
-extern void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
-extern void flush_dcache_page(struct page *pg);
-extern void flush_icache_range(unsigned long start, unsigned long end);
-extern void flush_icache_page(struct vm_area_struct *vma, struct page *page);
-
-#define flush_dcache_mmap_lock(mapping)		do { } while (0)
-#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
-
-/* SH3 has unified cache so no special action needed here */
-#define flush_cache_sigtramp(vaddr)		do { } while (0)
-#define flush_page_to_ram(page)			do { } while (0)
-#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
-
-#define p3_cache_init()				do { } while (0)
-
 #define PG_mapped	PG_arch_1
 
-/* We provide our own get_unmapped_area to avoid cache alias issue */
-#define HAVE_ARCH_UNMAPPED_AREA
-
+void flush_cache_all(void);
+void flush_cache_mm(struct mm_struct *mm);
+void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
+                              unsigned long end);
+void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
+void flush_dcache_page(struct page *pg);
+void flush_icache_range(unsigned long start, unsigned long end);
+void flush_icache_page(struct vm_area_struct *vma, struct page *page);
 #else
-
 #define flush_cache_all()			do { } while (0)
 #define flush_cache_mm(mm)			do { } while (0)
 #define flush_cache_range(vma, start, end)	do { } while (0)
 #define flush_cache_page(vma, vmaddr, pfn)	do { } while (0)
 #define flush_dcache_page(page)			do { } while (0)
-#define flush_dcache_mmap_lock(mapping)		do { } while (0)
-#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 #define flush_icache_range(start, end)		do { } while (0)
 #define flush_icache_page(vma,pg)		do { } while (0)
-#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
+#endif
+
+#define flush_dcache_mmap_lock(mapping)		do { } while (0)
+#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
+
+/* SH3 has unified cache so no special action needed here */
 #define flush_cache_sigtramp(vaddr)		do { } while (0)
+#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
 
 #define p3_cache_init()				do { } while (0)
 
-#endif
+/*
+ * We provide our own get_unmapped_area to avoid cache aliasing issues
+ * on SH7705 with a 32KB cache, and to page align addresses in the
+ * non-aliasing case.
+ */
+#define HAVE_ARCH_UNMAPPED_AREA
 
 #endif /* __ASM_CPU_SH3_CACHEFLUSH_H */
 
diff --git a/include/asm-sh/cpu-sh3/shmparam.h b/include/asm-sh/cpu-sh3/shmparam.h
deleted file mode 100644
index da5b5eec81ee..000000000000
--- a/include/asm-sh/cpu-sh3/shmparam.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- * include/asm-sh/cpu-sh3/shmparam.h
- *
- * Copyright (C) 1999 Niibe Yutaka
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_CPU_SH3_SHMPARAM_H
-#define __ASM_CPU_SH3_SHMPARAM_H
-
-#define	SHMLBA PAGE_SIZE		 /* attach addr a multiple of this */
-
-#endif /* __ASM_CPU_SH3_SHMPARAM_H */
-
diff --git a/include/asm-sh/cpu-sh4/shmparam.h b/include/asm-sh/cpu-sh4/shmparam.h
deleted file mode 100644
index a5a0aa9425fe..000000000000
--- a/include/asm-sh/cpu-sh4/shmparam.h
+++ /dev/null
@@ -1,19 +0,0 @@
-/*
- * include/asm-sh/cpu-sh4/shmparam.h
- *
- * Copyright (C) 1999 Niibe Yutaka
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_CPU_SH4_SHMPARAM_H
-#define __ASM_CPU_SH4_SHMPARAM_H
-
-/*
- * SH-4 has D-cache alias issue
- */
-#define	SHMLBA (PAGE_SIZE*4)		 /* attach addr a multiple of this */
-
-#endif /* __ASM_CPU_SH4_SHMPARAM_H */
-
diff --git a/include/asm-sh/shmparam.h b/include/asm-sh/shmparam.h
index 0a95604b9b66..ba1758d90106 100644
--- a/include/asm-sh/shmparam.h
+++ b/include/asm-sh/shmparam.h
@@ -1,8 +1,22 @@
+/*
+ * include/asm-sh/shmparam.h
+ *
+ * Copyright (C) 1999 Niibe Yutaka
+ * Copyright (C) 2006 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
 #ifndef __ASM_SH_SHMPARAM_H
 #define __ASM_SH_SHMPARAM_H
-#ifdef __KERNEL__
 
-#include <asm/cpu/shmparam.h>
+/*
+ * SH-4 and SH-3 7705 have an aliasing dcache. Bump this up to a sensible value
+ * for everyone, and work out the specifics from the probed cache descriptor.
+ */
+#define	SHMLBA	0x4000		 /* attach addr a multiple of this */
+
+#define __ARCH_FORCE_SHMLBA
 
-#endif /* __KERNEL__ */
 #endif /* __ASM_SH_SHMPARAM_H */
-- 
cgit v1.2.3


From 373e68b5472d421cbd2703e7a77caf053f78c005 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:41:24 +0900
Subject: sh: Board updates for I/O routine rework.

This updates the various boards for some of the recent I/O routine
updates.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                           |   3 -
 arch/sh/boards/renesas/hs7751rvoip/io.c    | 159 +++++++++++++++++------------
 arch/sh/boards/renesas/hs7751rvoip/setup.c |   5 +-
 arch/sh/boards/renesas/systemh/io.c        |  96 ++---------------
 arch/sh/boards/renesas/systemh/irq.c       |   2 +-
 arch/sh/boards/renesas/systemh/setup.c     |  14 +--
 arch/sh/boards/se/7300/io.c                |   8 +-
 arch/sh/boards/se/7300/irq.c               |   2 +-
 arch/sh/boards/se/7300/led.c               |  18 +---
 arch/sh/boards/se/7300/setup.c             |   3 +-
 arch/sh/boards/se/73180/setup.c            |   3 +-
 arch/sh/boards/se/770x/io.c                |  52 ++--------
 arch/sh/boards/se/770x/irq.c               |   2 +-
 arch/sh/boards/se/770x/led.c               |   2 +-
 arch/sh/boards/se/770x/mach.c              |   7 +-
 arch/sh/boards/se/770x/setup.c             |   2 +-
 arch/sh/boards/se/7751/io.c                |  26 ++---
 arch/sh/boards/se/7751/irq.c               |   2 +-
 arch/sh/boards/se/7751/led.c               |   3 +-
 arch/sh/boards/se/7751/mach.c              |   8 +-
 arch/sh/boards/se/7751/setup.c             |   6 +-
 arch/sh/boards/sh03/setup.c                |  25 ++---
 arch/sh/boards/snapgear/io.c               |  63 ++----------
 arch/sh/boards/snapgear/rtc.c              |   3 -
 arch/sh/boards/snapgear/setup.c            |  97 +-----------------
 arch/sh/boards/titan/io.c                  |   4 +-
 arch/sh/cchips/Kconfig                     |   6 +-
 arch/sh/kernel/cf-enabler.c                |   7 +-
 drivers/net/stnic.c                        |   2 +-
 include/asm-sh/hs7751rvoip/hs7751rvoip.h   |   3 +
 include/asm-sh/mc146818rtc.h               |   2 +-
 include/asm-sh/se.h                        |  80 +++++++++++++++
 include/asm-sh/se/io.h                     |  35 -------
 include/asm-sh/se/se.h                     |  77 --------------
 include/asm-sh/se7300.h                    |  64 ++++++++++++
 include/asm-sh/se7300/io.h                 |  29 ------
 include/asm-sh/se7300/se7300.h             |  61 -----------
 include/asm-sh/se73180.h                   |  65 ++++++++++++
 include/asm-sh/se73180/io.h                |  32 ------
 include/asm-sh/se73180/se73180.h           |  62 -----------
 include/asm-sh/se7751.h                    |  71 +++++++++++++
 include/asm-sh/se7751/io.h                 |  42 --------
 include/asm-sh/se7751/se7751.h             |  68 ------------
 include/asm-sh/sh03/io.h                   |  10 +-
 include/asm-sh/snapgear.h                  |  79 ++++++++++++++
 include/asm-sh/snapgear/io.h               |  92 -----------------
 include/asm-sh/systemh/7751systemh.h       |  68 ------------
 include/asm-sh/systemh/io.h                |  43 --------
 include/asm-sh/systemh7751.h               |  71 +++++++++++++
 49 files changed, 607 insertions(+), 1077 deletions(-)
 create mode 100644 include/asm-sh/se.h
 delete mode 100644 include/asm-sh/se/io.h
 delete mode 100644 include/asm-sh/se/se.h
 create mode 100644 include/asm-sh/se7300.h
 delete mode 100644 include/asm-sh/se7300/io.h
 delete mode 100644 include/asm-sh/se7300/se7300.h
 create mode 100644 include/asm-sh/se73180.h
 delete mode 100644 include/asm-sh/se73180/io.h
 delete mode 100644 include/asm-sh/se73180/se73180.h
 create mode 100644 include/asm-sh/se7751.h
 delete mode 100644 include/asm-sh/se7751/io.h
 delete mode 100644 include/asm-sh/se7751/se7751.h
 create mode 100644 include/asm-sh/snapgear.h
 delete mode 100644 include/asm-sh/snapgear/io.h
 delete mode 100644 include/asm-sh/systemh/7751systemh.h
 delete mode 100644 include/asm-sh/systemh/io.h
 create mode 100644 include/asm-sh/systemh7751.h

(limited to 'include')

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 55dcba2c06fc..65676c33822e 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -111,10 +111,7 @@ machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
 
-incdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se
 incdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se7751
-incdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)        := se7300
-incdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se73180
 incdir-$(CONFIG_SH_HP6XX)			:= hp6xx
 
 ifneq ($(machdir-y),)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index edecf107fc13..ecdce7ef6a34 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -21,10 +21,8 @@
 #include <linux/pci.h>
 #include "../../../drivers/pci/pci-sh7751.h"
 
-extern void *area5_io8_base;	/* Area 5 8bit I/O Base address */
 extern void *area6_io8_base;	/* Area 6 8bit I/O Base address */
 extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
-extern void *area6_io16_base;	/* Area 6 16bit I/O Base address */
 
 /*
  * The 7751R HS7751RVoIP uses the built-in PCI controller (PCIC)
@@ -37,16 +35,10 @@ extern void *area6_io16_base;	/* Area 6 16bit I/O Base address */
 #define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
 #define PCI_IO_AREA	SH7751_PCI_IO_BASE
 #define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 #define CODEC_IO_BASE	0x1000
-#endif
-
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
+#define CODEC_IOMAP(a)	((unsigned long)area6_io8_base + ((a) - CODEC_IO_BASE))
 
 static inline void delay(void)
 {
@@ -61,7 +53,7 @@ static inline unsigned long port2adr(unsigned int port)
 		else
 			return ((unsigned long)area5_io16_base + 0x800 + ((port-0x1f0) << 1));
 	else
-		maybebadio(port2adr, (unsigned long)port);
+		maybebadio((unsigned long)port);
 	return port;
 }
 
@@ -109,15 +101,15 @@ codec_port(unsigned long port)
 unsigned char hs7751rvoip_inb(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned char *)port;
+		return ctrl_inb(port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		return *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE));
+		return ctrl_inb(CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+		return ctrl_inb(PCI_IOMAP(port));
 	else
-		return (*(volatile unsigned short *)port2adr(port) & 0xff);
+		return ctrl_inw(port2adr(port)) & 0xff;
 }
 
 unsigned char hs7751rvoip_inb_p(unsigned long port)
@@ -125,15 +117,15 @@ unsigned char hs7751rvoip_inb_p(unsigned long port)
 	unsigned char v;
 
         if (PXSEG(port))
-                v = *(volatile unsigned char *)port;
+		v = ctrl_inb(port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		v = *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE));
+		v = ctrl_inb(CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-                v = *(volatile unsigned char *)PCI_IOMAP(port);
+		v = ctrl_inb(PCI_IOMAP(port));
 	else
-		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
+		v = ctrl_inw(port2adr(port)) & 0xff;
 	delay();
 	return v;
 }
@@ -141,22 +133,22 @@ unsigned char hs7751rvoip_inb_p(unsigned long port)
 unsigned short hs7751rvoip_inw(unsigned long port)
 {
         if (PXSEG(port))
-                return *(volatile unsigned short *)port;
+		return ctrl_inw(port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-                return *(volatile unsigned short *)PCI_IOMAP(port);
+		return ctrl_inw(PCI_IOMAP(port));
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
 unsigned int hs7751rvoip_inl(unsigned long port)
 {
         if (PXSEG(port))
-                return *(volatile unsigned long *)port;
+		return ctrl_inl(port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-                return *(volatile unsigned long *)PCI_IOMAP(port);
+		return ctrl_inl(PCI_IOMAP(port));
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -164,137 +156,168 @@ void hs7751rvoip_outb(unsigned char value, unsigned long port)
 {
 
         if (PXSEG(port))
-                *(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		ctrl_outb(value, CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*(unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outb(value, port2adr(port));
 }
 
 void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
-                *(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		ctrl_outb(value, CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*(unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outw(value, port2adr(port));
+
 	delay();
 }
 
 void hs7751rvoip_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
-                *(volatile unsigned short *)port = value;
+		ctrl_outw(value, port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*(unsigned short *)PCI_IOMAP(port) = value;
+		ctrl_outw(value, PCI_IOMAP(port));
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void hs7751rvoip_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
-                *(volatile unsigned long *)port = value;
+		ctrl_outl(value, port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*((unsigned long *)PCI_IOMAP(port)) = value;
+		ctrl_outl(value, PCI_IOMAP(port));
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void hs7751rvoip_insb(unsigned long port, void *addr, unsigned long count)
 {
+	u8 *buf = addr;
+
 	if (PXSEG(port))
-		while (count--) *((unsigned char *) addr)++ = *(volatile unsigned char *)port;
+		while (count--)
+			*buf++ = ctrl_inb(port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		while (count--) *((unsigned char *) addr)++ = *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE));
+		while (count--)
+			*buf++ = ctrl_inb(CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u8 *bp = (__u8 *)PCI_IOMAP(port);
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--) *((volatile unsigned char *) addr)++ = *bp;
+		while (count--)
+			*buf++ = *bp;
 	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+		volatile u16 *p = (volatile u16 *)port2adr(port);
 
-		while (count--) *((unsigned char *) addr)++ = *p & 0xff;
+		while (count--)
+			*buf++ = *p & 0xff;
 	}
 }
 
 void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
 {
-	volatile __u16 *p;
+	volatile u16 *p;
+	u16 *buf = addr;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+		p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+		p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
-	while (count--) *((__u16 *) addr)++ = *p;
+		p = (volatile u16 *)port2adr(port);
+	while (count--)
+		*buf++ = *p;
 }
 
 void hs7751rvoip_insl(unsigned long port, void *addr, unsigned long count)
 {
+
 	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u32 *p = (__u32 *)PCI_IOMAP(port);
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+		u32 *buf = addr;
 
-		while (count--) *((__u32 *) addr)++ = *p;
+		while (count--)
+			*buf++ = *p;
 	} else
-		maybebadio(insl, port);
+		maybebadio(port);
 }
 
 void hs7751rvoip_outsb(unsigned long port, const void *addr, unsigned long count)
 {
+	const u8 *buf = addr;
+
 	if (PXSEG(port))
-		while (count--) *(volatile unsigned char *)port = *((unsigned char *) addr)++;
+		while (count--)
+			ctrl_outb(*buf++, port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		while (count--) *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = *((unsigned char *) addr)++;
+		while (count--)
+			ctrl_outb(*buf++, CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u8 *bp = (__u8 *)PCI_IOMAP(port);
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--) *bp = *((volatile unsigned char *) addr)++;
+		while (count--)
+			*bp = *buf++;
 	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+		volatile u16 *p = (volatile u16 *)port2adr(port);
 
-		while (count--) *p = *((unsigned char *) addr)++;
+		while (count--)
+			*p = *buf++;
 	}
 }
 
 void hs7751rvoip_outsw(unsigned long port, const void *addr, unsigned long count)
 {
-	volatile __u16 *p;
+	volatile u16 *p;
+	const u16 *buf = addr;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+		p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+		p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
-	while (count--) *p = *((__u16 *) addr)++;
+		p = (volatile u16 *)port2adr(port);
+
+	while (count--)
+		*p = *buf++;
 }
 
 void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count)
 {
+	const u32 *buf = addr;
+
 	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u32 *p = (__u32 *)PCI_IOMAP(port);
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
 
-		while (count--) *p = *((__u32 *) addr)++;
+		while (count--)
+			*p = *buf++;
 	} else
-		maybebadio(outsl, port);
+		maybebadio(port);
 }
 
-unsigned long hs7751rvoip_isa_port2addr(unsigned long offset)
+void __iomem *hs7751rvoip_ioport_map(unsigned long port, unsigned int size)
 {
-	return port2adr(offset);
+        if (PXSEG(port))
+                return (void __iomem *)port;
+	else if (unlikely(codec_port(port) && (size == 1)))
+		return (void __iomem *)CODEC_IOMAP(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return (void __iomem *)PCI_IOMAP(port);
+
+        return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index aa51bda931f6..a2cbcc5d530f 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -17,9 +17,8 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/pm.h>
-#include <asm/hs7751rvoip/hs7751rvoip.h>
-#include <asm/hs7751rvoip/io.h>
 #include <asm/io.h>
+#include <asm/hs7751rvoip/hs7751rvoip.h>
 #include <asm/machvec.h>
 #include <asm/rtc.h>
 #include <asm/irq.h>
@@ -60,8 +59,8 @@ struct sh_machine_vector mv_hs7751rvoip __initmv = {
 	.mv_outsw		= hs7751rvoip_outsw,
 	.mv_outsl		= hs7751rvoip_outsl,
 
-	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
 	.mv_init_irq		= hs7751rvoip_init_irq,
+	.mv_ioport_map		= hs7751rvoip_ioport_map,
 };
 ALIAS_MV(hs7751rvoip)
 
diff --git a/arch/sh/boards/renesas/systemh/io.c b/arch/sh/boards/renesas/systemh/io.c
index 85511576d414..0befd4f9894c 100644
--- a/arch/sh/boards/renesas/systemh/io.c
+++ b/arch/sh/boards/renesas/systemh/io.c
@@ -10,11 +10,10 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <asm/systemh/7751systemh.h>
+#include <linux/pci.h>
+#include <asm/systemh7751.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
-
-#include <linux/pci.h>
 #include "../../../drivers/pci/pci-sh7751.h"
 
 /*
@@ -31,11 +30,6 @@
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 #define ETHER_IOMAP(adr) (0xB3000000 + (adr)) /*map to 16bits access area
                                                 of smc lan chip*/
-
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -46,11 +40,7 @@ port2adr(unsigned int port)
 {
 	if (port >= 0x2000)
 		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
-#if 0
-	else
-		return (volatile __u16 *) (PA_SUPERIO + (port << 1));
-#endif
-	maybebadio(name,(unsigned long)port);
+	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
@@ -111,7 +101,7 @@ unsigned short sh7751systemh_inw(unsigned long port)
 	else if (port <= 0x3F1)
 		return *(volatile unsigned int *)ETHER_IOMAP(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -126,7 +116,7 @@ unsigned int sh7751systemh_inl(unsigned long port)
 	else if (port <= 0x3F1)
 		return *(volatile unsigned int *)ETHER_IOMAP(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -167,7 +157,7 @@ void sh7751systemh_outw(unsigned short value, unsigned long port)
 	else if (port <= 0x3F1)
 		*(volatile unsigned short *)ETHER_IOMAP(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void sh7751systemh_outl(unsigned int value, unsigned long port)
@@ -177,7 +167,7 @@ void sh7751systemh_outl(unsigned int value, unsigned long port)
 	else if (CHECK_SH7751_PCIIO(port))
         	*((unsigned long*)PCI_IOMAP(port)) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void sh7751systemh_insb(unsigned long port, void *addr, unsigned long count)
@@ -194,7 +184,7 @@ void sh7751systemh_insw(unsigned long port, void *addr, unsigned long count)
 
 void sh7751systemh_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void sh7751systemh_outsb(unsigned long port, const void *addr, unsigned long count)
@@ -211,73 +201,5 @@ void sh7751systemh_outsw(unsigned long port, const void *addr, unsigned long cou
 
 void sh7751systemh_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
-}
-
-/* For read/write calls, just copy generic (pass-thru); PCIMBR is  */
-/* already set up.  For a larger memory space, these would need to */
-/* reset PCIMBR as needed on a per-call basis...                   */
-
-unsigned char sh7751systemh_readb(unsigned long addr)
-{
-	return *(volatile unsigned char*)addr;
-}
-
-unsigned short sh7751systemh_readw(unsigned long addr)
-{
-	return *(volatile unsigned short*)addr;
-}
-
-unsigned int sh7751systemh_readl(unsigned long addr)
-{
-	return *(volatile unsigned long*)addr;
-}
-
-void sh7751systemh_writeb(unsigned char b, unsigned long addr)
-{
-	*(volatile unsigned char*)addr = b;
-}
-
-void sh7751systemh_writew(unsigned short b, unsigned long addr)
-{
-	*(volatile unsigned short*)addr = b;
-}
-
-void sh7751systemh_writel(unsigned int b, unsigned long addr)
-{
-        *(volatile unsigned long*)addr = b;
-}
-
-
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-#if 0
-static int
-sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-	return 0;
-}
-#endif
-
-unsigned long
-sh7751systemh_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
+	maybebadio(port);
 }
diff --git a/arch/sh/boards/renesas/systemh/irq.c b/arch/sh/boards/renesas/systemh/irq.c
index 8372d967f601..53731a0abb8f 100644
--- a/arch/sh/boards/renesas/systemh/irq.c
+++ b/arch/sh/boards/renesas/systemh/irq.c
@@ -15,7 +15,7 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
-#include <asm/mach/7751systemh.h>
+#include <asm/systemh7751.h>
 #include <asm/smc37c93x.h>
 
 /* address of external interrupt mask register
diff --git a/arch/sh/boards/renesas/systemh/setup.c b/arch/sh/boards/renesas/systemh/setup.c
index 826fa3d7669c..433c7c3f35e2 100644
--- a/arch/sh/boards/renesas/systemh/setup.c
+++ b/arch/sh/boards/renesas/systemh/setup.c
@@ -15,9 +15,8 @@
  * for more details.
  */
 #include <linux/init.h>
-#include <asm/mach/7751systemh.h>
-#include <asm/mach/io.h>
 #include <asm/machvec.h>
+#include <asm/systemh7751.h>
 
 extern void make_systemh_irq(unsigned int irq);
 
@@ -31,8 +30,6 @@ const char *get_system_type(void)
  */
 void __init init_7751systemh_IRQ(void)
 {
-/*  	make_ipr_irq(10, BCR_ILCRD, 1, 0x0f-10); LAN */
-/*  	make_ipr_irq(14, BCR_ILCRA, 2, 0x0f-4); */
 	make_systemh_irq(0xb);	/* Ethernet interrupt */
 }
 
@@ -60,15 +57,6 @@ struct sh_machine_vector mv_7751systemh __initmv = {
 	.mv_outsw		= sh7751systemh_outsw,
 	.mv_outsl		= sh7751systemh_outsl,
 
-	.mv_readb		= sh7751systemh_readb,
-	.mv_readw		= sh7751systemh_readw,
-	.mv_readl		= sh7751systemh_readl,
-	.mv_writeb		= sh7751systemh_writeb,
-	.mv_writew		= sh7751systemh_writew,
-	.mv_writel		= sh7751systemh_writel,
-
-	.mv_isa_port2addr	= sh7751systemh_isa_port2addr,
-
 	.mv_init_irq		= init_7751systemh_IRQ,
 };
 ALIAS_MV(7751systemh)
diff --git a/arch/sh/boards/se/7300/io.c b/arch/sh/boards/se/7300/io.c
index f449a94ddffd..8a03d7a52a7c 100644
--- a/arch/sh/boards/se/7300/io.c
+++ b/arch/sh/boards/se/7300/io.c
@@ -9,8 +9,8 @@
  */
 
 #include <linux/kernel.h>
-#include <asm/mach/se7300.h>
 #include <asm/io.h>
+#include <asm/se7300.h>
 
 #define badio(fn, a) panic("bad i/o operation %s for %08lx.", #fn, a)
 
@@ -99,6 +99,7 @@ bad_outb(struct iop *p, unsigned char value, unsigned long port)
 	badio(inw, port);
 }
 
+#ifdef CONFIG_SMC91X
 /* MSTLANEX01 LAN at 0xb400:0000 */
 static struct iop laniop = {
 	.start = 0x300,
@@ -110,6 +111,7 @@ static struct iop laniop = {
 	.outb = simple_outb,
 	.outw = simple_outw,
 };
+#endif
 
 /* NE2000 pc card NIC */
 static struct iop neiop = {
@@ -123,6 +125,7 @@ static struct iop neiop = {
 	.outw = simple_outw,
 };
 
+#ifdef CONFIG_IDE
 /* CF in CF slot */
 static struct iop cfiop = {
 	.base = 0xb0600000,
@@ -132,12 +135,13 @@ static struct iop cfiop = {
 	.outb = pcc_outb,
 	.outw = simple_outw,
 };
+#endif
 
 static __inline__ struct iop *
 port2iop(unsigned long port)
 {
 	if (0) ;
-#if defined(CONFIG_SMC91111)
+#if defined(CONFIG_SMC91X)
 	else if (laniop.check(&laniop, port))
 		return &laniop;
 #endif
diff --git a/arch/sh/boards/se/7300/irq.c b/arch/sh/boards/se/7300/irq.c
index 216a78d1a108..ad1034f98a29 100644
--- a/arch/sh/boards/se/7300/irq.c
+++ b/arch/sh/boards/se/7300/irq.c
@@ -11,7 +11,7 @@
 #include <linux/irq.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/mach/se7300.h>
+#include <asm/se7300.h>
 
 /*
  * Initialize IRQ setting
diff --git a/arch/sh/boards/se/7300/led.c b/arch/sh/boards/se/7300/led.c
index ad51f0a9c1e3..4d03bb7774be 100644
--- a/arch/sh/boards/se/7300/led.c
+++ b/arch/sh/boards/se/7300/led.c
@@ -12,24 +12,10 @@
  */
 
 #include <linux/sched.h>
-#include <asm/mach/se7300.h>
-
-static void
-mach_led(int position, int value)
-{
-	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
-
-	if (value) {
-		*p |= (1 << 8);
-	} else {
-		*p &= ~(1 << 8);
-	}
-}
-
+#include <asm/se7300.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
-void
-heartbeat_7300se(void)
+void heartbeat_7300se(void)
 {
 	static unsigned int cnt = 0, period = 0;
 	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
diff --git a/arch/sh/boards/se/7300/setup.c b/arch/sh/boards/se/7300/setup.c
index ebcd98d4c081..bb7e1a189be8 100644
--- a/arch/sh/boards/se/7300/setup.c
+++ b/arch/sh/boards/se/7300/setup.c
@@ -9,8 +9,7 @@
 
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
-#include <asm/mach/io.h>
+#include <asm/se7300.h>
 
 void heartbeat_7300se(void);
 void init_7300se_IRQ(void);
diff --git a/arch/sh/boards/se/73180/setup.c b/arch/sh/boards/se/73180/setup.c
index cdb7b5f8d942..cddc7a2b858f 100644
--- a/arch/sh/boards/se/73180/setup.c
+++ b/arch/sh/boards/se/73180/setup.c
@@ -11,8 +11,7 @@
 
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
-#include <asm/mach/io.h>
+#include <asm/se73180.h>
 
 void heartbeat_73180se(void);
 void init_73180se_IRQ(void);
diff --git a/arch/sh/boards/se/770x/io.c b/arch/sh/boards/se/770x/io.c
index 9a39ee963143..5102201c97ab 100644
--- a/arch/sh/boards/se/770x/io.c
+++ b/arch/sh/boards/se/770x/io.c
@@ -1,4 +1,4 @@
-/* $Id: io.c,v 1.5 2004/02/22 23:08:43 kkojima Exp $
+/* $Id: io.c,v 1.6 2006/01/04 17:53:54 lethal Exp $
  *
  * linux/arch/sh/kernel/io_se.c
  *
@@ -11,7 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /* SH pcmcia io window base, start and end.  */
 int sh_pcic_io_wbase = 0xb8400000;
@@ -52,10 +52,6 @@ shifted_port(unsigned long port)
 		return 1;
 }
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 unsigned char se_inb(unsigned long port)
 {
 	if (sh_pcic_io_start <= port && port <= sh_pcic_io_stop)
@@ -86,13 +82,13 @@ unsigned short se_inw(unsigned long port)
 	    (sh_pcic_io_start <= port && port <= sh_pcic_io_stop))
 		return *port2adr(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
 unsigned int se_inl(unsigned long port)
 {
-	maybebadio(inl, port);
+	maybebadio(port);
 	return 0;
 }
 
@@ -123,12 +119,12 @@ void se_outw(unsigned short value, unsigned long port)
 	    (sh_pcic_io_start <= port && port <= sh_pcic_io_stop))
 		*port2adr(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void se_outl(unsigned int value, unsigned long port)
 {
-	maybebadio(outl, port);
+	maybebadio(port);
 }
 
 void se_insb(unsigned long port, void *addr, unsigned long count)
@@ -159,7 +155,7 @@ void se_insw(unsigned long port, void *addr, unsigned long count)
 
 void se_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void se_outsb(unsigned long port, const void *addr, unsigned long count)
@@ -190,37 +186,5 @@ void se_outsw(unsigned long port, const void *addr, unsigned long count)
 
 void se_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
-}
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-static int
-sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-#if 0
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-#endif
-	return 0;
-}
-
-unsigned long
-se_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
+	maybebadio(port);
 }
diff --git a/arch/sh/boards/se/770x/irq.c b/arch/sh/boards/se/770x/irq.c
index 3e558716ce10..cff6700bbafd 100644
--- a/arch/sh/boards/se/770x/irq.c
+++ b/arch/sh/boards/se/770x/irq.c
@@ -11,7 +11,7 @@
 #include <linux/irq.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /*
  * Initialize IRQ setting
diff --git a/arch/sh/boards/se/770x/led.c b/arch/sh/boards/se/770x/led.c
index 73e9848387ea..daf7b1ee786a 100644
--- a/arch/sh/boards/se/770x/led.c
+++ b/arch/sh/boards/se/770x/led.c
@@ -10,7 +10,7 @@
  */
 
 #include <linux/sched.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
 void heartbeat_se(void)
diff --git a/arch/sh/boards/se/770x/mach.c b/arch/sh/boards/se/770x/mach.c
index 6ec07bd3dcf1..e8968b71c353 100644
--- a/arch/sh/boards/se/770x/mach.c
+++ b/arch/sh/boards/se/770x/mach.c
@@ -13,12 +13,9 @@
 
 #include <asm/machvec.h>
 #include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/se/io.h>
+#include <asm/se.h>
 
 void heartbeat_se(void);
-void setup_se(void);
 void init_se_IRQ(void);
 
 /*
@@ -57,8 +54,6 @@ struct sh_machine_vector mv_se __initmv = {
 	.mv_outsw		= se_outsw,
 	.mv_outsl		= se_outsl,
 
-	.mv_isa_port2addr	= se_isa_port2addr,
-
 	.mv_init_irq		= init_se_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_se,
diff --git a/arch/sh/boards/se/770x/setup.c b/arch/sh/boards/se/770x/setup.c
index b6730ea57502..c9f75272e751 100644
--- a/arch/sh/boards/se/770x/setup.c
+++ b/arch/sh/boards/se/770x/setup.c
@@ -14,7 +14,7 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 #include <asm/smc37c93x.h>
 
 /*
diff --git a/arch/sh/boards/se/7751/io.c b/arch/sh/boards/se/7751/io.c
index 99041b269261..0e8a3ba48316 100644
--- a/arch/sh/boards/se/7751/io.c
+++ b/arch/sh/boards/se/7751/io.c
@@ -1,4 +1,4 @@
-/* 
+/*
  * linux/arch/sh/kernel/io_7751se.c
  *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
@@ -14,7 +14,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <asm/io.h>
-#include <asm/se7751/se7751.h>
+#include <asm/se7751.h>
 #include <asm/addrspace.h>
 
 #include <linux/pci.h>
@@ -52,10 +52,6 @@ int sh_pcic_io_dummy;
 
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -66,11 +62,7 @@ port2adr(unsigned int port)
 {
 	if (port >= 0x2000)
 		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
-#if 0
-	else
-		return (volatile __u16 *) (PA_SUPERIO + (port << 1));
-#endif
-	maybebadio(name,(unsigned long)port);
+	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
@@ -140,7 +132,7 @@ unsigned short sh7751se_inw(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -153,7 +145,7 @@ unsigned int sh7751se_inl(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -188,7 +180,7 @@ void sh7751se_outw(unsigned short value, unsigned long port)
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void sh7751se_outl(unsigned int value, unsigned long port)
@@ -198,17 +190,17 @@ void sh7751se_outl(unsigned int value, unsigned long port)
 	else if (CHECK_SH7751_PCIIO(port))
         	*((unsigned long*)PCI_IOMAP(port)) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void sh7751se_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
+	maybebadio(port);
 }
 
 /* Map ISA bus address to the real address. Only for PCMCIA.  */
diff --git a/arch/sh/boards/se/7751/irq.c b/arch/sh/boards/se/7751/irq.c
index bf6c023615df..c607b0a48479 100644
--- a/arch/sh/boards/se/7751/irq.c
+++ b/arch/sh/boards/se/7751/irq.c
@@ -12,7 +12,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <asm/irq.h>
-#include <asm/se7751/se7751.h>
+#include <asm/se7751.h>
 
 /*
  * Initialize IRQ setting
diff --git a/arch/sh/boards/se/7751/led.c b/arch/sh/boards/se/7751/led.c
index 4405e26cf866..ff0355dea81b 100644
--- a/arch/sh/boards/se/7751/led.c
+++ b/arch/sh/boards/se/7751/led.c
@@ -8,9 +8,8 @@
  *
  * This file contains Solution Engine specific LED code.
  */
-
-#include <asm/se7751/se7751.h>
 #include <linux/sched.h>
+#include <asm/se7751.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
 void heartbeat_7751se(void)
diff --git a/arch/sh/boards/se/7751/mach.c b/arch/sh/boards/se/7751/mach.c
index 62d8d3e62590..1bb9047d863b 100644
--- a/arch/sh/boards/se/7751/mach.c
+++ b/arch/sh/boards/se/7751/mach.c
@@ -10,12 +10,8 @@
  */
 
 #include <linux/init.h>
-
 #include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/se7751/io.h>
+#include <asm/se7751.h>
 
 void heartbeat_7751se(void);
 void init_7751se_IRQ(void);
@@ -44,8 +40,6 @@ struct sh_machine_vector mv_7751se __initmv = {
 	.mv_insl		= sh7751se_insl,
 	.mv_outsl		= sh7751se_outsl,
 
-	.mv_isa_port2addr	= sh7751se_isa_port2addr,
-
 	.mv_init_irq		= init_7751se_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_7751se,
diff --git a/arch/sh/boards/se/7751/setup.c b/arch/sh/boards/se/7751/setup.c
index 48dc5aee67d4..8b693105893c 100644
--- a/arch/sh/boards/se/7751/setup.c
+++ b/arch/sh/boards/se/7751/setup.c
@@ -1,4 +1,4 @@
-/* 
+/*
  * linux/arch/sh/kernel/setup_7751se.c
  *
  * Copyright (C) 2000  Kazumoto Kojima
@@ -11,11 +11,9 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
-
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
-#include <asm/se7751/se7751.h>
+#include <asm/se7751.h>
 
 #ifdef CONFIG_SH_KGDB
 #include <asm/kgdb.h>
diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c
index 60290f8f289c..dab742a00c6f 100644
--- a/arch/sh/boards/sh03/setup.c
+++ b/arch/sh/boards/sh03/setup.c
@@ -7,22 +7,19 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
-#include <linux/hdreg.h>
-#include <linux/ide.h>
 #include <asm/io.h>
+#include <asm/rtc.h>
 #include <asm/sh03/io.h>
 #include <asm/sh03/sh03.h>
 #include <asm/addrspace.h>
 #include "../../drivers/pci/pci-sh7751.h"
 
-extern void (*board_time_init)(void);
-
 const char *get_system_type(void)
 {
 	return "Interface CTP/PCI-SH03)";
 }
 
-void init_sh03_IRQ(void)
+static void init_sh03_IRQ(void)
 {
 	ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR);
 
@@ -34,24 +31,20 @@ void init_sh03_IRQ(void)
 
 extern void *cf_io_base;
 
-unsigned long sh03_isa_port2addr(unsigned long port)
+static void __iomem *sh03_ioport_map(unsigned long port, unsigned int size)
 {
 	if (PXSEG(port))
-		return port;
+		return (void __iomem *)port;
 	/* CompactFlash (IDE) */
-	if (((port >= 0x1f0) && (port <= 0x1f7)) || (port == 0x3f6)) {
-		return (unsigned long)cf_io_base + port;
-	}
-        return port + SH7751_PCI_IO_BASE;
-}
+	if (((port >= 0x1f0) && (port <= 0x1f7)) || (port == 0x3f6))
+		return (void __iomem *)((unsigned long)cf_io_base + port);
 
-/*
- * The Machine Vector
- */
+        return (void __iomem *)(port + SH7751_PCI_IO_BASE);
+}
 
 struct sh_machine_vector mv_sh03 __initmv = {
 	.mv_nr_irqs		= 48,
-	.mv_isa_port2addr	= sh03_isa_port2addr,
+	.mv_ioport_map		= sh03_ioport_map,
 	.mv_init_irq		= init_sh03_IRQ,
 
 #ifdef CONFIG_HEARTBEAT
diff --git a/arch/sh/boards/snapgear/io.c b/arch/sh/boards/snapgear/io.c
index e2eb78fc381d..9f700b8392bb 100644
--- a/arch/sh/boards/snapgear/io.c
+++ b/arch/sh/boards/snapgear/io.c
@@ -28,39 +28,26 @@ unsigned short secureedge5410_ioport;
 /*
  * The SnapGear uses the built-in PCI controller (PCIC)
  * of the 7751 processor
- */ 
+ */
 
 #define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
 #define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
 #define PCI_IO_AREA	SH7751_PCI_IO_BASE
 #define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
 
-
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
 }
 
-
 static inline volatile __u16 *port2adr(unsigned int port)
 {
-#if 0
-	if (port >= 0x2000)
-		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
-#endif
-	maybebadio(name,(unsigned long)port);
+	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
-
 /* In case someone configures the kernel w/o PCI support: in that */
 /* scenario, don't ever bother to check for PCI-window addresses */
 
@@ -115,7 +102,7 @@ unsigned short snapgear_inw(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -129,7 +116,7 @@ unsigned int snapgear_inl(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -167,7 +154,7 @@ void snapgear_outw(unsigned short value, unsigned long port)
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 
@@ -178,49 +165,15 @@ void snapgear_outl(unsigned int value, unsigned long port)
 	else if (CHECK_SH7751_PCIIO(port))
 		*((unsigned long*)PCI_IOMAP(port)) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void snapgear_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void snapgear_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
-}
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-
-#if 0
-static int sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-#if 0
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-#endif
-	return 0;
-}
-#endif
-
-unsigned long snapgear_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
+	maybebadio(port);
 }
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index b71e009da35c..287bc4ed3ac7 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -17,10 +17,7 @@
 #include <linux/time.h>
 #include <linux/rtc.h>
 #include <linux/mc146818rtc.h>
-
 #include <asm/io.h>
-#include <asm/rtc.h>
-#include <asm/mc146818rtc.h>
 
 /****************************************************************************/
 
diff --git a/arch/sh/boards/snapgear/setup.c b/arch/sh/boards/snapgear/setup.c
index f1f7c70c9402..66ce32f8b13c 100644
--- a/arch/sh/boards/snapgear/setup.c
+++ b/arch/sh/boards/snapgear/setup.c
@@ -1,5 +1,4 @@
-/****************************************************************************/
-/* 
+/*
  * linux/arch/sh/boards/snapgear/setup.c
  *
  * Copyright (C) 2002  David McCullough <davidm@snapgear.com>
@@ -12,8 +11,6 @@
  *           Modified for 7751 Solution Engine by
  *           Ian da Silva and Jeremy Siegel, 2001.
  */
-/****************************************************************************/
-
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/interrupt.h>
@@ -21,14 +18,13 @@
 #include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/sched.h>
-
 #include <asm/machvec.h>
-#include <asm/mach/io.h>
+#include <asm/snapgear.h>
 #include <asm/irq.h>
 #include <asm/io.h>
+#include <asm/rtc.h>
 #include <asm/cpu/timer.h>
 
-extern void (*board_time_init)(void);
 extern void secureedge5410_rtc_init(void);
 extern void pcibios_init(void);
 
@@ -85,91 +81,6 @@ static void __init init_snapgear_IRQ(void)
 	make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY);
 }
 
-/****************************************************************************/
-/*
- *	Fast poll interrupt simulator.
- */
-
-/*
- * Leave all of the fast timer/fast poll stuff commented out for now, since
- * it's not clear whether it actually works or not. Since it wasn't being used
- * at all in 2.4, we'll assume it's not sane for 2.6 either.. -- PFM
- */
-#if 0
-#define FAST_POLL	1000
-//#define FAST_POLL_INTR
-
-#define FASTTIMER_IRQ   17
-#define FASTTIMER_IPR_ADDR  INTC_IPRA
-#define FASTTIMER_IPR_POS    2
-#define FASTTIMER_PRIORITY   3
-
-#ifdef FAST_POLL_INTR
-#define TMU1_TCR_INIT	0x0020
-#else
-#define TMU1_TCR_INIT	0
-#endif
-#define TMU_TSTR_INIT	1
-#define TMU1_TCR_CALIB	0x0000
-
-
-#ifdef FAST_POLL_INTR
-static void fast_timer_irq(int irq, void *dev_instance, struct pt_regs *regs)
-{
-	unsigned long timer_status;
-    timer_status = ctrl_inw(TMU1_TCR);
-	timer_status &= ~0x100;
-	ctrl_outw(timer_status, TMU1_TCR);
-}
-#endif
-
-/*
- * return the current ticks on the fast timer
- */
-
-unsigned long fast_timer_count(void)
-{
-	return(ctrl_inl(TMU1_TCNT));
-}
-
-/*
- * setup a fast timer for profiling etc etc
- */
-
-static void setup_fast_timer()
-{
-	unsigned long interval;
-
-#ifdef FAST_POLL_INTR
-	interval = (current_cpu_data.module_clock/4 + FAST_POLL/2) / FAST_POLL;
-
-	make_ipr_irq(FASTTIMER_IRQ, FASTTIMER_IPR_ADDR, FASTTIMER_IPR_POS,
-			FASTTIMER_PRIORITY);
-
-	printk("SnapGear: %dHz fast timer on IRQ %d\n",FAST_POLL,FASTTIMER_IRQ);
-
-	if (request_irq(FASTTIMER_IRQ, fast_timer_irq, 0, "SnapGear fast timer",
-			NULL) != 0)
-		printk("%s(%d): request_irq() failed?\n", __FILE__, __LINE__);
-#else
-	printk("SnapGear: fast timer running\n",FAST_POLL,FASTTIMER_IRQ);
-	interval = 0xffffffff;
-#endif
-
-	ctrl_outb(ctrl_inb(TMU_TSTR) & ~0x2, TMU_TSTR); /* disable timer 1 */
-	ctrl_outw(TMU1_TCR_INIT, TMU1_TCR);
-	ctrl_outl(interval, TMU1_TCOR);
-	ctrl_outl(interval, TMU1_TCNT);
-	ctrl_outb(ctrl_inb(TMU_TSTR) | 0x2, TMU_TSTR); /* enable timer 1 */
-
-	printk("Timer count 1 = 0x%x\n", fast_timer_count());
-	udelay(1000);
-	printk("Timer count 2 = 0x%x\n", fast_timer_count());
-}
-#endif
-
-/****************************************************************************/
-
 const char *get_system_type(void)
 {
 	return "SnapGear SecureEdge5410";
@@ -196,8 +107,6 @@ struct sh_machine_vector mv_snapgear __initmv = {
 	.mv_outw_p		= snapgear_outw,
 	.mv_outl_p		= snapgear_outl,
 
-	.mv_isa_port2addr	= snapgear_isa_port2addr,
-
 	.mv_init_irq		= init_snapgear_IRQ,
 };
 ALIAS_MV(snapgear)
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
index b886fd233a66..48f3494f55b1 100644
--- a/arch/sh/boards/titan/io.c
+++ b/arch/sh/boards/titan/io.c
@@ -30,10 +30,10 @@ static inline void delay(void)
         ctrl_inw(0xa0000000);
 }
 
-static inline volatile u16 *port2adr(unsigned int port)
+static inline unsigned int port2adr(unsigned int port)
 {
         maybebadio((unsigned long)port);
-        return (volatile u16*)port;
+        return port;
 }
 
 u8 titan_inb(unsigned long port)
diff --git a/arch/sh/cchips/Kconfig b/arch/sh/cchips/Kconfig
index 155d139884c3..0582ca8346b6 100644
--- a/arch/sh/cchips/Kconfig
+++ b/arch/sh/cchips/Kconfig
@@ -65,6 +65,11 @@ config HD64461_IRQ
 
 	  Do not change this unless you know what you are doing.
 
+config HD64461_IOBASE
+	hex "HD64461 start address"
+	depends on HD64461
+	default "0xb0000000"
+
 config HD64461_ENABLER
 	bool "HD64461 PCMCIA enabler"
 	depends on HD64461
@@ -73,7 +78,6 @@ config HD64461_ENABLER
 	  via the HD64461 companion chip.
 	  Otherwise, say N.
 
-
 config HD64465_IOBASE
 	hex "HD64465 start address"
 	depends on HD64465
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index c9b823d1d073..3e5fa1e24df0 100644
--- a/arch/sh/kernel/cf-enabler.c
+++ b/arch/sh/kernel/cf-enabler.c
@@ -10,7 +10,8 @@
  */
 
 #include <linux/init.h>
-
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
 #include <asm/io.h>
 #include <asm/irq.h>
 
@@ -32,8 +33,6 @@
 /* SH4 can't access PCMCIA interface through P2 area.
  * we must remap it with appropreate attribute bit of the page set.
  * this part is based on Greg Banks' hd64465_ss.c implementation - Masahiro Abe */
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
 
 #if defined(CONFIG_CF_AREA6)
 #define slot_no 0
@@ -84,7 +83,7 @@ static int __init cf_init_default(void)
 }
 
 #if defined(CONFIG_SH_SOLUTION_ENGINE)
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /*
  * SolutionEngine
diff --git a/drivers/net/stnic.c b/drivers/net/stnic.c
index 3fd7a4fee665..e6f90427160c 100644
--- a/drivers/net/stnic.c
+++ b/drivers/net/stnic.c
@@ -19,7 +19,7 @@
 
 #include <asm/system.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 #include <asm/machvec.h>
 #ifdef CONFIG_SH_STANDARD_BIOS
 #include <asm/sh_bios.h>
diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
index 857cc2ebbc69..69faf0171473 100644
--- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h
+++ b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
@@ -44,6 +44,9 @@
 #define	IRQ_RINGING	4		/* Ringing IRQ */
 #define	IRQ_CODEC	5		/* CODEC IRQ */
 
+#define __IO_PREFIX	hs7751rvoip
+#include <asm/io_generic.h>
+
 /* arch/sh/boards/renesas/hs7751rvoip/irq.c */
 void init_hs7751rvoip_IRQ(void);
 
diff --git a/include/asm-sh/mc146818rtc.h b/include/asm-sh/mc146818rtc.h
index 1707cfb2915d..adc6e67c6b75 100644
--- a/include/asm-sh/mc146818rtc.h
+++ b/include/asm-sh/mc146818rtc.h
@@ -24,7 +24,7 @@
 #define CMOS_WRITE(val,addr)	__CMOS_WRITE(val,addr,b)
 
 #elif defined(CONFIG_SH_SECUREEDGE5410)
-#include <asm/snapgear/io.h>
+#include <asm/snapgear.h>
 
 #define RTC_PORT(n)             SECUREEDGE_IOPORT_ADDR
 #define CMOS_READ(addr)         secureedge5410_cmos_read(addr)
diff --git a/include/asm-sh/se.h b/include/asm-sh/se.h
new file mode 100644
index 000000000000..a1832154a3aa
--- /dev/null
+++ b/include/asm-sh/se.h
@@ -0,0 +1,80 @@
+#ifndef __ASM_SH_HITACHI_SE_H
+#define __ASM_SH_HITACHI_SE_H
+
+/*
+ * linux/include/asm-sh/hitachi_se.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Hitachi SolutionEngine support
+ */
+
+/* Box specific addresses.  */
+
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_FROM		0x01000000	/* EPROM */
+#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+
+#define PA_EXT4		0x12000000
+#define PA_EXT4_SIZE	0x02000000
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
+
+#define PA_83902	0xb0000000	/* DP83902A */
+#define PA_83902_IF	0xb0040000	/* DP83902A remote io port */
+#define PA_83902_RST	0xb0080000	/* DP83902A reset port */
+
+#define PA_SUPERIO	0xb0400000	/* SMC37C935A super io chip */
+#define PA_DIPSW0	0xb0800000	/* Dip switch 5,6 */
+#define PA_DIPSW1	0xb0800002	/* Dip switch 7,8 */
+#define PA_LED		0xb0c00000	/* LED */
+#if defined(CONFIG_CPU_SUBTYPE_SH7705)
+#define PA_BCR		0xb0e00000
+#else
+#define PA_BCR		0xb1400000	/* FPGA */
+#endif
+
+#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+
+#define BCR_ILCRA	(PA_BCR + 0)
+#define BCR_ILCRB	(PA_BCR + 2)
+#define BCR_ILCRC	(PA_BCR + 4)
+#define BCR_ILCRD	(PA_BCR + 6)
+#define BCR_ILCRE	(PA_BCR + 8)
+#define BCR_ILCRF	(PA_BCR + 10)
+#define BCR_ILCRG	(PA_BCR + 12)
+
+#if defined(CONFIG_CPU_SUBTYPE_SH7705)
+#define IRQ_STNIC   12
+#else
+#define IRQ_STNIC	10
+#endif
+
+#define __IO_PREFIX	se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_SE_H */
diff --git a/include/asm-sh/se/io.h b/include/asm-sh/se/io.h
deleted file mode 100644
index 9eeb86cd6cef..000000000000
--- a/include/asm-sh/se/io.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * include/asm-sh/io_se.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an Hitachi SolutionEngine
- */
-
-#ifndef _ASM_SH_IO_SE_H
-#define _ASM_SH_IO_SE_H
-
-extern unsigned char se_inb(unsigned long port);
-extern unsigned short se_inw(unsigned long port);
-extern unsigned int se_inl(unsigned long port);
-
-extern void se_outb(unsigned char value, unsigned long port);
-extern void se_outw(unsigned short value, unsigned long port);
-extern void se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char se_inb_p(unsigned long port);
-extern void se_outb_p(unsigned char value, unsigned long port);
-
-extern void se_insb(unsigned long port, void *addr, unsigned long count);
-extern void se_insw(unsigned long port, void *addr, unsigned long count);
-extern void se_insl(unsigned long port, void *addr, unsigned long count);
-extern void se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned long se_isa_port2addr(unsigned long offset);
-
-#endif /* _ASM_SH_IO_SE_H */
diff --git a/include/asm-sh/se/se.h b/include/asm-sh/se/se.h
deleted file mode 100644
index 791c5da0388a..000000000000
--- a/include/asm-sh/se/se.h
+++ /dev/null
@@ -1,77 +0,0 @@
-#ifndef __ASM_SH_HITACHI_SE_H
-#define __ASM_SH_HITACHI_SE_H
-
-/*
- * linux/include/asm-sh/hitachi_se.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Hitachi SolutionEngine support
- */
-
-/* Box specific addresses.  */
-
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_FROM		0x01000000	/* EPROM */
-#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-
-#define PA_EXT4		0x12000000
-#define PA_EXT4_SIZE	0x02000000
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
-
-#define PA_83902	0xb0000000	/* DP83902A */
-#define PA_83902_IF	0xb0040000	/* DP83902A remote io port */
-#define PA_83902_RST	0xb0080000	/* DP83902A reset port */
-
-#define PA_SUPERIO	0xb0400000	/* SMC37C935A super io chip */
-#define PA_DIPSW0	0xb0800000	/* Dip switch 5,6 */
-#define PA_DIPSW1	0xb0800002	/* Dip switch 7,8 */
-#define PA_LED		0xb0c00000	/* LED */
-#if defined(CONFIG_CPU_SUBTYPE_SH7705)
-#define PA_BCR		0xb0e00000
-#else
-#define PA_BCR		0xb1400000	/* FPGA */
-#endif
-
-#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA	(PA_BCR + 0)
-#define BCR_ILCRB	(PA_BCR + 2)
-#define BCR_ILCRC	(PA_BCR + 4)
-#define BCR_ILCRD	(PA_BCR + 6)
-#define BCR_ILCRE	(PA_BCR + 8)
-#define BCR_ILCRF	(PA_BCR + 10)
-#define BCR_ILCRG	(PA_BCR + 12)
-
-#if defined(CONFIG_CPU_SUBTYPE_SH7705)
-#define IRQ_STNIC   12
-#else
-#define IRQ_STNIC	10
-#endif
-
-#endif  /* __ASM_SH_HITACHI_SE_H */
diff --git a/include/asm-sh/se7300.h b/include/asm-sh/se7300.h
new file mode 100644
index 000000000000..4e24edccb30d
--- /dev/null
+++ b/include/asm-sh/se7300.h
@@ -0,0 +1,64 @@
+#ifndef __ASM_SH_HITACHI_SE7300_H
+#define __ASM_SH_HITACHI_SE7300_H
+
+/*
+ * linux/include/asm-sh/se/se7300.h
+ *
+ * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
+ *
+ * SH-Mobile SolutionEngine 7300 support
+ */
+
+/* Box specific addresses.  */
+
+/* Area 0 */
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
+#define PA_FROM		0x00400000	/* Flash ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
+#define PA_SRAM		0x00800000	/* SRAM */
+#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
+/* Area 1 */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+/* Area 2 */
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+/* Area 3 */
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+/* Area 4 */
+#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
+#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+#define PA_LED		0xb0800000	/* LED */
+#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
+#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
+#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
+#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
+/* Area 5 */
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+/* Area 6 */
+#define PA_LCD1		0xb8000000
+#define PA_LCD2		0xb8800000
+
+#define __IO_PREFIX	sh7300se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_SE7300_H */
diff --git a/include/asm-sh/se7300/io.h b/include/asm-sh/se7300/io.h
deleted file mode 100644
index c6af85529714..000000000000
--- a/include/asm-sh/se7300/io.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * include/asm-sh/se7300/io.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- * IO functions for SH-Mobile(SH7300) SolutionEngine
- */
-
-#ifndef _ASM_SH_IO_7300SE_H
-#define _ASM_SH_IO_7300SE_H
-
-extern unsigned char sh7300se_inb(unsigned long port);
-extern unsigned short sh7300se_inw(unsigned long port);
-extern unsigned int sh7300se_inl(unsigned long port);
-
-extern void sh7300se_outb(unsigned char value, unsigned long port);
-extern void sh7300se_outw(unsigned short value, unsigned long port);
-extern void sh7300se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh7300se_inb_p(unsigned long port);
-extern void sh7300se_outb_p(unsigned char value, unsigned long port);
-
-extern void sh7300se_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh7300se_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh7300se_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh7300se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh7300se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh7300se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-#endif /* _ASM_SH_IO_7300SE_H */
diff --git a/include/asm-sh/se7300/se7300.h b/include/asm-sh/se7300/se7300.h
deleted file mode 100644
index 3ec1ded86c97..000000000000
--- a/include/asm-sh/se7300/se7300.h
+++ /dev/null
@@ -1,61 +0,0 @@
-#ifndef __ASM_SH_HITACHI_SE7300_H
-#define __ASM_SH_HITACHI_SE7300_H
-
-/*
- * linux/include/asm-sh/se/se7300.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- *
- * SH-Mobile SolutionEngine 7300 support
- */
-
-/* Box specific addresses.  */
-
-/* Area 0 */
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
-#define PA_FROM		0x00400000	/* Flash ROM */
-#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
-#define PA_SRAM		0x00800000	/* SRAM */
-#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
-/* Area 1 */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-/* Area 2 */
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-/* Area 3 */
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-/* Area 4 */
-#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
-#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-#define PA_LED		0xb0800000	/* LED */
-#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
-#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
-#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
-#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
-/* Area 5 */
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-/* Area 6 */
-#define PA_LCD1		0xb8000000
-#define PA_LCD2		0xb8800000
-
-#endif  /* __ASM_SH_HITACHI_SE7300_H */
diff --git a/include/asm-sh/se73180.h b/include/asm-sh/se73180.h
new file mode 100644
index 000000000000..3a4acb3e38a1
--- /dev/null
+++ b/include/asm-sh/se73180.h
@@ -0,0 +1,65 @@
+#ifndef __ASM_SH_HITACHI_SE73180_H
+#define __ASM_SH_HITACHI_SE73180_H
+
+/*
+ * include/asm-sh/se/se73180.h
+ *
+ * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
+ *
+ * SH-Mobile SolutionEngine 73180 support
+ */
+
+/* Box specific addresses.  */
+
+/* Area 0 */
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
+#define PA_FROM		0x00400000	/* Flash ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
+#define PA_SRAM		0x00800000	/* SRAM */
+#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
+/* Area 1 */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+/* Area 2 */
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+/* Area 3 */
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+/* Area 4 */
+#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
+#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+#define PA_LED		0xb0C00000	/* LED */
+#define LED_SHIFT       0
+#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
+#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
+#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
+#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
+/* Area 5 */
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+/* Area 6 */
+#define PA_LCD1		0xb8000000
+#define PA_LCD2		0xb8800000
+
+#define __IO_PREFIX	sh73180se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_SE73180_H */
diff --git a/include/asm-sh/se73180/io.h b/include/asm-sh/se73180/io.h
deleted file mode 100644
index c9cb1b9412c6..000000000000
--- a/include/asm-sh/se73180/io.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * include/asm-sh/se73180/io.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- * Based on include/asm-sh/se7300/io.h
- *
- * IO functions for SH-Mobile3(SH73180) SolutionEngine
- *
- */
-
-#ifndef _ASM_SH_IO_73180SE_H
-#define _ASM_SH_IO_73180SE_H
-
-extern unsigned char sh73180se_inb(unsigned long port);
-extern unsigned short sh73180se_inw(unsigned long port);
-extern unsigned int sh73180se_inl(unsigned long port);
-
-extern void sh73180se_outb(unsigned char value, unsigned long port);
-extern void sh73180se_outw(unsigned short value, unsigned long port);
-extern void sh73180se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh73180se_inb_p(unsigned long port);
-extern void sh73180se_outb_p(unsigned char value, unsigned long port);
-
-extern void sh73180se_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh73180se_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh73180se_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh73180se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh73180se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh73180se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-#endif /* _ASM_SH_IO_73180SE_H */
diff --git a/include/asm-sh/se73180/se73180.h b/include/asm-sh/se73180/se73180.h
deleted file mode 100644
index f5b93e39e768..000000000000
--- a/include/asm-sh/se73180/se73180.h
+++ /dev/null
@@ -1,62 +0,0 @@
-#ifndef __ASM_SH_HITACHI_SE73180_H
-#define __ASM_SH_HITACHI_SE73180_H
-
-/*
- * include/asm-sh/se/se73180.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- *
- * SH-Mobile SolutionEngine 73180 support
- */
-
-/* Box specific addresses.  */
-
-/* Area 0 */
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
-#define PA_FROM		0x00400000	/* Flash ROM */
-#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
-#define PA_SRAM		0x00800000	/* SRAM */
-#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
-/* Area 1 */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-/* Area 2 */
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-/* Area 3 */
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-/* Area 4 */
-#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
-#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-#define PA_LED		0xb0C00000	/* LED */
-#define LED_SHIFT       0
-#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
-#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
-#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
-#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
-/* Area 5 */
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-/* Area 6 */
-#define PA_LCD1		0xb8000000
-#define PA_LCD2		0xb8800000
-
-#endif  /* __ASM_SH_HITACHI_SE73180_H */
diff --git a/include/asm-sh/se7751.h b/include/asm-sh/se7751.h
new file mode 100644
index 000000000000..88cd379d9084
--- /dev/null
+++ b/include/asm-sh/se7751.h
@@ -0,0 +1,71 @@
+#ifndef __ASM_SH_HITACHI_7751SE_H
+#define __ASM_SH_HITACHI_7751SE_H
+
+/*
+ * linux/include/asm-sh/hitachi_7751se.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Hitachi SolutionEngine support
+
+ * Modified for 7751 Solution Engine by
+ * Ian da Silva and Jeremy Siegel, 2001.
+ */
+
+/* Box specific addresses.  */
+
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_FROM		0x01000000	/* EPROM */
+#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+
+#define PA_EXT4		0x12000000
+#define PA_EXT4_SIZE	0x02000000
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
+
+#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
+#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
+#define PA_LED		0xba000000	/* LED */
+#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
+
+#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
+#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
+#define MRSHPC_MODE     (PA_MRSHPC + 4)
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+
+#define BCR_ILCRA	(PA_BCR + 0)
+#define BCR_ILCRB	(PA_BCR + 2)
+#define BCR_ILCRC	(PA_BCR + 4)
+#define BCR_ILCRD	(PA_BCR + 6)
+#define BCR_ILCRE	(PA_BCR + 8)
+#define BCR_ILCRF	(PA_BCR + 10)
+#define BCR_ILCRG	(PA_BCR + 12)
+
+#define IRQ_79C973	13
+
+#define __IO_PREFIX	sh7751se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_7751SE_H */
diff --git a/include/asm-sh/se7751/io.h b/include/asm-sh/se7751/io.h
deleted file mode 100644
index 78d8f5744bc5..000000000000
--- a/include/asm-sh/se7751/io.h
+++ /dev/null
@@ -1,42 +0,0 @@
-/*
- * include/asm-sh/io_7751se.h
- *
- * Modified version of io_se.h for the 7751se-specific functions.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an Hitachi SolutionEngine
- */
-
-#ifndef _ASM_SH_IO_7751SE_H
-#define _ASM_SH_IO_7751SE_H
-
-extern unsigned char sh7751se_inb(unsigned long port);
-extern unsigned short sh7751se_inw(unsigned long port);
-extern unsigned int sh7751se_inl(unsigned long port);
-
-extern void sh7751se_outb(unsigned char value, unsigned long port);
-extern void sh7751se_outw(unsigned short value, unsigned long port);
-extern void sh7751se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh7751se_inb_p(unsigned long port);
-extern void sh7751se_outb_p(unsigned char value, unsigned long port);
-
-extern void sh7751se_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh7751se_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh7751se_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh7751se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned char sh7751se_readb(unsigned long addr);
-extern unsigned short sh7751se_readw(unsigned long addr);
-extern unsigned int sh7751se_readl(unsigned long addr);
-extern void sh7751se_writeb(unsigned char b, unsigned long addr);
-extern void sh7751se_writew(unsigned short b, unsigned long addr);
-extern void sh7751se_writel(unsigned int b, unsigned long addr);
-
-extern unsigned long sh7751se_isa_port2addr(unsigned long offset);
-
-#endif /* _ASM_SH_IO_7751SE_H */
diff --git a/include/asm-sh/se7751/se7751.h b/include/asm-sh/se7751/se7751.h
deleted file mode 100644
index 738e22bebdfb..000000000000
--- a/include/asm-sh/se7751/se7751.h
+++ /dev/null
@@ -1,68 +0,0 @@
-#ifndef __ASM_SH_HITACHI_7751SE_H
-#define __ASM_SH_HITACHI_7751SE_H
-
-/*
- * linux/include/asm-sh/hitachi_7751se.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Hitachi SolutionEngine support
-
- * Modified for 7751 Solution Engine by
- * Ian da Silva and Jeremy Siegel, 2001.
- */
-
-/* Box specific addresses.  */
-
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_FROM		0x01000000	/* EPROM */
-#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-
-#define PA_EXT4		0x12000000
-#define PA_EXT4_SIZE	0x02000000
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
-
-#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
-#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
-#define PA_LED		0xba000000	/* LED */
-#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
-
-#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
-#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
-#define MRSHPC_MODE     (PA_MRSHPC + 4)
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA	(PA_BCR + 0)
-#define BCR_ILCRB	(PA_BCR + 2)
-#define BCR_ILCRC	(PA_BCR + 4)
-#define BCR_ILCRD	(PA_BCR + 6)
-#define BCR_ILCRE	(PA_BCR + 8)
-#define BCR_ILCRF	(PA_BCR + 10)
-#define BCR_ILCRG	(PA_BCR + 12)
-
-#define IRQ_79C973	13
-
-#endif  /* __ASM_SH_HITACHI_7751SE_H */
diff --git a/include/asm-sh/sh03/io.h b/include/asm-sh/sh03/io.h
index 25792e9831ea..df3b187ef883 100644
--- a/include/asm-sh/sh03/io.h
+++ b/include/asm-sh/sh03/io.h
@@ -33,14 +33,6 @@
 #define IRL3_IPR_POS	0
 #define IRL3_PRIORITY	4
 
-
-extern unsigned long sh03_isa_port2addr(unsigned long offset);
-
-extern void setup_sh03(void);
-extern void init_sh03_IRQ(void);
-extern void heartbeat_sh03(void);
-
-extern void sh03_rtc_gettimeofday(struct timeval *tv);
-extern int sh03_rtc_settimeofday(const struct timeval *tv);
+void heartbeat_sh03(void);
 
 #endif /* _ASM_SH_IO_SH03_H */
diff --git a/include/asm-sh/snapgear.h b/include/asm-sh/snapgear.h
new file mode 100644
index 000000000000..6b5e4ddc073a
--- /dev/null
+++ b/include/asm-sh/snapgear.h
@@ -0,0 +1,79 @@
+/*
+ * include/asm-sh/snapgear/io.h
+ *
+ * Modified version of io_se.h for the snapgear-specific functions.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * IO functions for a SnapGear
+ */
+
+#ifndef _ASM_SH_IO_SNAPGEAR_H
+#define _ASM_SH_IO_SNAPGEAR_H
+
+#if defined(CONFIG_CPU_SH4)
+/*
+ * The external interrupt lines, these take up ints 0 - 15 inclusive
+ * depending on the priority for the interrupt.  In fact the priority
+ * is the interrupt :-)
+ */
+
+#define IRL0_IRQ		2
+#define IRL0_IPR_ADDR	INTC_IPRD
+#define IRL0_IPR_POS	3
+#define IRL0_PRIORITY	13
+
+#define IRL1_IRQ		5
+#define IRL1_IPR_ADDR	INTC_IPRD
+#define IRL1_IPR_POS	2
+#define IRL1_PRIORITY	10
+
+#define IRL2_IRQ		8
+#define IRL2_IPR_ADDR	INTC_IPRD
+#define IRL2_IPR_POS	1
+#define IRL2_PRIORITY	7
+
+#define IRL3_IRQ		11
+#define IRL3_IPR_ADDR	INTC_IPRD
+#define IRL3_IPR_POS	0
+#define IRL3_PRIORITY	4
+#endif
+
+#define __IO_PREFIX	snapgear
+#include <asm/io_generic.h>
+
+#ifdef CONFIG_SH_SECUREEDGE5410
+/*
+ * We need to remember what was written to the ioport as some bits
+ * are shared with other functions and you cannot read back what was
+ * written :-|
+ *
+ * Bit        Read                   Write
+ * -----------------------------------------------
+ * D0         DCD on ttySC1          power
+ * D1         Reset Switch           heatbeat
+ * D2         ttySC0 CTS (7100)      LAN
+ * D3         -                      WAN
+ * D4         ttySC0 DCD (7100)      CONSOLE
+ * D5         -                      ONLINE
+ * D6         -                      VPN
+ * D7         -                      DTR on ttySC1
+ * D8         -                      ttySC0 RTS (7100)
+ * D9         -                      ttySC0 DTR (7100)
+ * D10        -                      RTC SCLK
+ * D11        RTC DATA               RTC DATA
+ * D12        -                      RTS RESET
+ */
+
+#define SECUREEDGE_IOPORT_ADDR ((volatile short *) 0xb0000000)
+extern unsigned short secureedge5410_ioport;
+
+#define SECUREEDGE_WRITE_IOPORT(val, mask) (*SECUREEDGE_IOPORT_ADDR = \
+	 (secureedge5410_ioport = \
+			((secureedge5410_ioport & ~(mask)) | ((val) & (mask)))))
+#define SECUREEDGE_READ_IOPORT() \
+	 ((*SECUREEDGE_IOPORT_ADDR&0x0817) | (secureedge5410_ioport&~0x0817))
+#endif
+
+#endif /* _ASM_SH_IO_SNAPGEAR_H */
diff --git a/include/asm-sh/snapgear/io.h b/include/asm-sh/snapgear/io.h
deleted file mode 100644
index bfa97ac06280..000000000000
--- a/include/asm-sh/snapgear/io.h
+++ /dev/null
@@ -1,92 +0,0 @@
-/*
- * include/asm-sh/snapgear/io.h
- *
- * Modified version of io_se.h for the snapgear-specific functions.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for a SnapGear
- */
-
-#ifndef _ASM_SH_IO_SNAPGEAR_H
-#define _ASM_SH_IO_SNAPGEAR_H
-
-#if defined(CONFIG_CPU_SH4)
-/*
- * The external interrupt lines, these take up ints 0 - 15 inclusive
- * depending on the priority for the interrupt.  In fact the priority
- * is the interrupt :-)
- */
-
-#define IRL0_IRQ		2
-#define IRL0_IPR_ADDR	INTC_IPRD
-#define IRL0_IPR_POS	3
-#define IRL0_PRIORITY	13
-
-#define IRL1_IRQ		5
-#define IRL1_IPR_ADDR	INTC_IPRD
-#define IRL1_IPR_POS	2
-#define IRL1_PRIORITY	10
-
-#define IRL2_IRQ		8
-#define IRL2_IPR_ADDR	INTC_IPRD
-#define IRL2_IPR_POS	1
-#define IRL2_PRIORITY	7
-
-#define IRL3_IRQ		11
-#define IRL3_IPR_ADDR	INTC_IPRD
-#define IRL3_IPR_POS	0
-#define IRL3_PRIORITY	4
-#endif
-
-extern unsigned char snapgear_inb(unsigned long port);
-extern unsigned short snapgear_inw(unsigned long port);
-extern unsigned int snapgear_inl(unsigned long port);
-
-extern void snapgear_outb(unsigned char value, unsigned long port);
-extern void snapgear_outw(unsigned short value, unsigned long port);
-extern void snapgear_outl(unsigned int value, unsigned long port);
-
-extern unsigned char snapgear_inb_p(unsigned long port);
-extern void snapgear_outb_p(unsigned char value, unsigned long port);
-
-extern void snapgear_insl(unsigned long port, void *addr, unsigned long count);
-extern void snapgear_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned long snapgear_isa_port2addr(unsigned long offset);
-
-#ifdef CONFIG_SH_SECUREEDGE5410
-/*
- * We need to remember what was written to the ioport as some bits
- * are shared with other functions and you cannot read back what was
- * written :-|
- *
- * Bit        Read                   Write
- * -----------------------------------------------
- * D0         DCD on ttySC1          power
- * D1         Reset Switch           heatbeat
- * D2         ttySC0 CTS (7100)      LAN
- * D3         -                      WAN
- * D4         ttySC0 DCD (7100)      CONSOLE
- * D5         -                      ONLINE
- * D6         -                      VPN
- * D7         -                      DTR on ttySC1
- * D8         -                      ttySC0 RTS (7100)
- * D9         -                      ttySC0 DTR (7100)
- * D10        -                      RTC SCLK
- * D11        RTC DATA               RTC DATA
- * D12        -                      RTS RESET
- */
-
- #define SECUREEDGE_IOPORT_ADDR ((volatile short *) 0xb0000000)
- extern unsigned short secureedge5410_ioport;
-
- #define SECUREEDGE_WRITE_IOPORT(val, mask) (*SECUREEDGE_IOPORT_ADDR = \
-		 (secureedge5410_ioport = \
-		 		((secureedge5410_ioport & ~(mask)) | ((val) & (mask)))))
- #define SECUREEDGE_READ_IOPORT() \
- 		 ((*SECUREEDGE_IOPORT_ADDR&0x0817) | (secureedge5410_ioport&~0x0817))
-#endif
-
-#endif /* _ASM_SH_IO_SNAPGEAR_H */
diff --git a/include/asm-sh/systemh/7751systemh.h b/include/asm-sh/systemh/7751systemh.h
deleted file mode 100644
index 4170531bdbd9..000000000000
--- a/include/asm-sh/systemh/7751systemh.h
+++ /dev/null
@@ -1,68 +0,0 @@
-#ifndef __ASM_SH_SYSTEMH_7751SYSTEMH_H
-#define __ASM_SH_SYSTEMH_7751SYSTEMH_H
-
-/*
- * linux/include/asm-sh/systemh/7751systemh.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Hitachi SystemH support
-
- * Modified for 7751 SystemH by
- * Jonathan Short, 2002.
- */
-
-/* Box specific addresses.  */
-
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_FROM		0x01000000	/* EPROM */
-#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-
-#define PA_EXT4		0x12000000
-#define PA_EXT4_SIZE	0x02000000
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
-
-#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
-#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
-#define PA_LED		0xba000000	/* LED */
-#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
-
-#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
-#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
-#define MRSHPC_MODE     (PA_MRSHPC + 4)
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA	(PA_BCR + 0)
-#define BCR_ILCRB	(PA_BCR + 2)
-#define BCR_ILCRC	(PA_BCR + 4)
-#define BCR_ILCRD	(PA_BCR + 6)
-#define BCR_ILCRE	(PA_BCR + 8)
-#define BCR_ILCRF	(PA_BCR + 10)
-#define BCR_ILCRG	(PA_BCR + 12)
-
-#define IRQ_79C973	13
-
-#endif  /* __ASM_SH_SYSTEMH_7751SYSTEMH_H */
diff --git a/include/asm-sh/systemh/io.h b/include/asm-sh/systemh/io.h
deleted file mode 100644
index 327849b49db8..000000000000
--- a/include/asm-sh/systemh/io.h
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * include/asm-sh/systemh/io.h
- *
- * Stupid I/O definitions for SystemH, cloned from SE7751.
- *
- * Copyright (C) 2003  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_SH_SYSTEMH_IO_H
-#define __ASM_SH_SYSTEMH_IO_H
-
-extern unsigned char sh7751systemh_inb(unsigned long port);
-extern unsigned short sh7751systemh_inw(unsigned long port);
-extern unsigned int sh7751systemh_inl(unsigned long port);
-
-extern void sh7751systemh_outb(unsigned char value, unsigned long port);
-extern void sh7751systemh_outw(unsigned short value, unsigned long port);
-extern void sh7751systemh_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh7751systemh_inb_p(unsigned long port);
-extern void sh7751systemh_outb_p(unsigned char value, unsigned long port);
-
-extern void sh7751systemh_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh7751systemh_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh7751systemh_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh7751systemh_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751systemh_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751systemh_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned char sh7751systemh_readb(unsigned long addr);
-extern unsigned short sh7751systemh_readw(unsigned long addr);
-extern unsigned int sh7751systemh_readl(unsigned long addr);
-extern void sh7751systemh_writeb(unsigned char b, unsigned long addr);
-extern void sh7751systemh_writew(unsigned short b, unsigned long addr);
-extern void sh7751systemh_writel(unsigned int b, unsigned long addr);
-
-extern unsigned long sh7751systemh_isa_port2addr(unsigned long offset);
-
-#endif /* __ASM_SH_SYSTEMH_IO_H */
-
diff --git a/include/asm-sh/systemh7751.h b/include/asm-sh/systemh7751.h
new file mode 100644
index 000000000000..b143bb2a2ca7
--- /dev/null
+++ b/include/asm-sh/systemh7751.h
@@ -0,0 +1,71 @@
+#ifndef __ASM_SH_SYSTEMH_7751SYSTEMH_H
+#define __ASM_SH_SYSTEMH_7751SYSTEMH_H
+
+/*
+ * linux/include/asm-sh/systemh/7751systemh.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Hitachi SystemH support
+
+ * Modified for 7751 SystemH by
+ * Jonathan Short, 2002.
+ */
+
+/* Box specific addresses.  */
+
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_FROM		0x01000000	/* EPROM */
+#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+
+#define PA_EXT4		0x12000000
+#define PA_EXT4_SIZE	0x02000000
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
+
+#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
+#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
+#define PA_LED		0xba000000	/* LED */
+#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
+
+#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
+#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
+#define MRSHPC_MODE     (PA_MRSHPC + 4)
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+
+#define BCR_ILCRA	(PA_BCR + 0)
+#define BCR_ILCRB	(PA_BCR + 2)
+#define BCR_ILCRC	(PA_BCR + 4)
+#define BCR_ILCRD	(PA_BCR + 6)
+#define BCR_ILCRE	(PA_BCR + 8)
+#define BCR_ILCRF	(PA_BCR + 10)
+#define BCR_ILCRG	(PA_BCR + 12)
+
+#define IRQ_79C973	13
+
+#define __IO_PREFIX	sh7751systemh
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_SYSTEMH_7751SYSTEMH_H */
-- 
cgit v1.2.3


From d7c30c682a278abe1a52db83f69efec1a9d8f8c2 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:49:57 +0900
Subject: sh: Store Queue API rework.

Rewrite the store queue API for a per-cpu interface in the driver
model. The old miscdevice is dropped, due to TASK_SIZE limitations,
and no one was using it anyways.

Carve up and allocate store queue space with a bitmap, back sq
mapping objects with a slab cache, and let userspace worry about
its own prefetching.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/sq.c | 542 ++++++++++++++++++++------------------------
 include/asm-sh/cpu-sh4/sq.h |  23 +-
 2 files changed, 251 insertions(+), 314 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index b148966dd7c7..7bcc73f9b8df 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -1,50 +1,52 @@
 /*
- * arch/sh/kernel/cpu/sq.c
+ * arch/sh/kernel/cpu/sh4/sq.c
  *
  * General management API for SH-4 integrated Store Queues
  *
- * Copyright (C) 2001, 2002, 2003, 2004  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2001, 2002  M. R. Brown
  *
- * Some of this code has been adopted directly from the old arch/sh/mm/sq.c
- * hack that was part of the LinuxDC project. For all intents and purposes,
- * this is a completely new interface that really doesn't have much in common
- * with the old zone-based approach at all. In fact, it's only listed here for
- * general completeness.
- *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
 #include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/bitmap.h>
+#include <linux/sysdev.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/slab.h>
-#include <linux/list.h>
-#include <linux/proc_fs.h>
-#include <linux/miscdevice.h>
 #include <linux/vmalloc.h>
 #include <linux/mm.h>
 #include <asm/io.h>
 #include <asm/page.h>
 #include <asm/cacheflush.h>
-#include <asm/mmu_context.h>
 #include <asm/cpu/sq.h>
 
-static LIST_HEAD(sq_mapping_list);
+struct sq_mapping;
+
+struct sq_mapping {
+	const char *name;
+
+	unsigned long sq_addr;
+	unsigned long addr;
+	unsigned int size;
+
+	struct sq_mapping *next;
+};
+
+static struct sq_mapping *sq_mapping_list;
 static DEFINE_SPINLOCK(sq_mapping_lock);
+static kmem_cache_t *sq_cache;
+static unsigned long *sq_bitmap;
 
-/**
- * sq_flush - Flush (prefetch) the store queue cache
- * @addr: the store queue address to flush
- *
- * Executes a prefetch instruction on the specified store queue cache,
- * so that the cached data is written to physical memory.
- */
-inline void sq_flush(void *addr)
-{
-	__asm__ __volatile__ ("pref @%0" : : "r" (addr) : "memory");
-}
+#define store_queue_barrier()			\
+do {						\
+	(void)ctrl_inl(P4SEG_STORE_QUE);	\
+	ctrl_outl(0, P4SEG_STORE_QUE + 0);	\
+	ctrl_outl(0, P4SEG_STORE_QUE + 8);	\
+} while (0);
 
 /**
  * sq_flush_range - Flush (prefetch) a specific SQ range
@@ -57,154 +59,73 @@ inline void sq_flush(void *addr)
 void sq_flush_range(unsigned long start, unsigned int len)
 {
 	volatile unsigned long *sq = (unsigned long *)start;
-	unsigned long dummy;
 
 	/* Flush the queues */
 	for (len >>= 5; len--; sq += 8)
-		sq_flush((void *)sq);
+		prefetchw((void *)sq);
 
 	/* Wait for completion */
-	dummy = ctrl_inl(P4SEG_STORE_QUE);
-
-	ctrl_outl(0, P4SEG_STORE_QUE + 0);
-	ctrl_outl(0, P4SEG_STORE_QUE + 8);
+	store_queue_barrier();
 }
 
-static struct sq_mapping *__sq_alloc_mapping(unsigned long virt, unsigned long phys, unsigned long size, const char *name)
+static inline void sq_mapping_list_add(struct sq_mapping *map)
 {
-	struct sq_mapping *map;
-
-	if (virt + size > SQ_ADDRMAX)
-		return ERR_PTR(-ENOSPC);
+	struct sq_mapping **p, *tmp;
 
-	map = kmalloc(sizeof(struct sq_mapping), GFP_KERNEL);
-	if (!map)
-		return ERR_PTR(-ENOMEM);
+	spin_lock_irq(&sq_mapping_lock);
 
-	INIT_LIST_HEAD(&map->list);
+	p = &sq_mapping_list;
+	while ((tmp = *p) != NULL)
+		p = &tmp->next;
 
-	map->sq_addr	= virt;
-	map->addr	= phys;
-	map->size	= size + 1;
-	map->name	= name;
+	map->next = tmp;
+	*p = map;
 
-	list_add(&map->list, &sq_mapping_list);
-
-	return map;
+	spin_unlock_irq(&sq_mapping_lock);
 }
 
-static unsigned long __sq_get_next_addr(void)
+static inline void sq_mapping_list_del(struct sq_mapping *map)
 {
-	if (!list_empty(&sq_mapping_list)) {
-		struct list_head *pos, *tmp;
-
-		/*
-		 * Read one off the list head, as it will have the highest
-		 * mapped allocation. Set the next one up right above it.
-		 *
-		 * This is somewhat sub-optimal, as we don't look at
-		 * gaps between allocations or anything lower then the
-		 * highest-level allocation.
-		 *
-		 * However, in the interest of performance and the general
-		 * lack of desire to do constant list rebalancing, we don't
-		 * worry about it.
-		 */
-		list_for_each_safe(pos, tmp, &sq_mapping_list) {
-			struct sq_mapping *entry;
-
-			entry = list_entry(pos, typeof(*entry), list);
-
-			return entry->sq_addr + entry->size;
+	struct sq_mapping **p, *tmp;
+
+	spin_lock_irq(&sq_mapping_lock);
+
+	for (p = &sq_mapping_list; (tmp = *p); p = &tmp->next)
+		if (tmp == map) {
+			*p = tmp->next;
+			break;
 		}
-	}
 
-	return P4SEG_STORE_QUE;
+	spin_unlock_irq(&sq_mapping_lock);
 }
 
-/**
- * __sq_remap - Perform a translation from the SQ to a phys addr
- * @map: sq mapping containing phys and store queue addresses.
- *
- * Maps the store queue address specified in the mapping to the physical
- * address specified in the mapping.
- */
-static struct sq_mapping *__sq_remap(struct sq_mapping *map)
+static int __sq_remap(struct sq_mapping *map, unsigned long flags)
 {
-	unsigned long flags, pteh, ptel;
+#if defined(CONFIG_MMU)
 	struct vm_struct *vma;
-	pgprot_t pgprot;
 
-	/*
-	 * Without an MMU (or with it turned off), this is much more
-	 * straightforward, as we can just load up each queue's QACR with
-	 * the physical address appropriately masked.
-	 */
-
-	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR0);
-	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR1);
-
-#ifdef CONFIG_MMU
-	/*
-	 * With an MMU on the other hand, things are slightly more involved.
-	 * Namely, we have to have a direct mapping between the SQ addr and
-	 * the associated physical address in the UTLB by way of setting up
-	 * a virt<->phys translation by hand. We do this by simply specifying
-	 * the SQ addr in UTLB.VPN and the associated physical address in
-	 * UTLB.PPN.
-	 *
-	 * Notably, even though this is a special case translation, and some
-	 * of the configuration bits are meaningless, we're still required
-	 * to have a valid ASID context in PTEH.
-	 *
-	 * We could also probably get by without explicitly setting PTEA, but
-	 * we do it here just for good measure.
-	 */
-	spin_lock_irqsave(&sq_mapping_lock, flags);
-
-	pteh = map->sq_addr;
-	ctrl_outl((pteh & MMU_VPN_MASK) | get_asid(), MMU_PTEH);
-
-	ptel = map->addr & PAGE_MASK;
-
-	if (cpu_data->flags & CPU_HAS_PTEA)
-		ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
-
-	pgprot = pgprot_noncached(PAGE_KERNEL);
-
-	ptel &= _PAGE_FLAGS_HARDWARE_MASK;
-	ptel |= pgprot_val(pgprot);
-	ctrl_outl(ptel, MMU_PTEL);
-
-	__asm__ __volatile__ ("ldtlb" : : : "memory");
-
-	spin_unlock_irqrestore(&sq_mapping_lock, flags);
-
-	/*
-	 * Next, we need to map ourselves in the kernel page table, so that
-	 * future accesses after a TLB flush will be handled when we take a
-	 * page fault.
-	 *
-	 * Theoretically we could just do this directly and not worry about
-	 * setting up the translation by hand ahead of time, but for the
-	 * cases where we want a one-shot SQ mapping followed by a quick
-	 * writeout before we hit the TLB flush, we do it anyways. This way
-	 * we at least save ourselves the initial page fault overhead.
-	 */
 	vma = __get_vm_area(map->size, VM_ALLOC, map->sq_addr, SQ_ADDRMAX);
 	if (!vma)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
 	vma->phys_addr = map->addr;
 
 	if (remap_area_pages((unsigned long)vma->addr, vma->phys_addr,
-			     map->size, pgprot_val(pgprot))) {
+			     map->size, flags)) {
 		vunmap(vma->addr);
-		return NULL;
+		return -EAGAIN;
 	}
-#endif /* CONFIG_MMU */
+#else
+	/*
+	 * Without an MMU (or with it turned off), this is much more
+	 * straightforward, as we can just load up each queue's QACR with
+	 * the physical address appropriately masked.
+	 */
+	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR0);
+	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR1);
+#endif
 
-	return map;
+	return 0;
 }
 
 /**
@@ -212,42 +133,65 @@ static struct sq_mapping *__sq_remap(struct sq_mapping *map)
  * @phys: Physical address of mapping.
  * @size: Length of mapping.
  * @name: User invoking mapping.
+ * @flags: Protection flags.
  *
  * Remaps the physical address @phys through the next available store queue
  * address of @size length. @name is logged at boot time as well as through
- * the procfs interface.
- *
- * A pre-allocated and filled sq_mapping pointer is returned, and must be
- * cleaned up with a call to sq_unmap() when the user is done with the
- * mapping.
+ * the sysfs interface.
  */
-struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *name)
+unsigned long sq_remap(unsigned long phys, unsigned int size,
+		       const char *name, unsigned long flags)
 {
 	struct sq_mapping *map;
-	unsigned long virt, end;
+	unsigned long end;
 	unsigned int psz;
+	int ret, page;
 
 	/* Don't allow wraparound or zero size */
 	end = phys + size - 1;
-	if (!size || end < phys)
-		return NULL;
+	if (unlikely(!size || end < phys))
+		return -EINVAL;
 	/* Don't allow anyone to remap normal memory.. */
-	if (phys < virt_to_phys(high_memory))
-		return NULL;
+	if (unlikely(phys < virt_to_phys(high_memory)))
+		return -EINVAL;
 
 	phys &= PAGE_MASK;
+	size = PAGE_ALIGN(end + 1) - phys;
+
+	map = kmem_cache_alloc(sq_cache, GFP_KERNEL);
+	if (unlikely(!map))
+		return -ENOMEM;
+
+	map->addr = phys;
+	map->size = size;
+	map->name = name;
+
+	page = bitmap_find_free_region(sq_bitmap, 0x04000000,
+				       get_order(map->size));
+	if (unlikely(page < 0)) {
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	map->sq_addr = P4SEG_STORE_QUE + (page << PAGE_SHIFT);
+
+	ret = __sq_remap(map, flags);
+	if (unlikely(ret != 0))
+		goto out;
+
+	psz = (size + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	pr_info("sqremap: %15s  [%4d page%s]  va 0x%08lx   pa 0x%08lx\n",
+		likely(map->name) ? map->name : "???",
+		psz, psz == 1 ? " " : "s",
+		map->sq_addr, map->addr);
 
-	size  = PAGE_ALIGN(end + 1) - phys;
-	virt  = __sq_get_next_addr();
-	psz   = (size + (PAGE_SIZE - 1)) / PAGE_SIZE;
-	map   = __sq_alloc_mapping(virt, phys, size, name);
+	sq_mapping_list_add(map);
 
-	printk("sqremap: %15s  [%4d page%s]  va 0x%08lx   pa 0x%08lx\n",
-	       map->name ? map->name : "???",
-	       psz, psz == 1 ? " " : "s",
-	       map->sq_addr, map->addr);
+	return map->sq_addr;
 
-	return __sq_remap(map);
+out:
+	kmem_cache_free(sq_cache, map);
+	return ret;
 }
 
 /**
@@ -258,188 +202,198 @@ struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *n
  * sq_remap(). Also frees up the pte that was previously inserted into
  * the kernel page table and discards the UTLB translation.
  */
-void sq_unmap(struct sq_mapping *map)
+void sq_unmap(unsigned long vaddr)
 {
-	if (map->sq_addr > (unsigned long)high_memory)
-		vfree((void *)(map->sq_addr & PAGE_MASK));
+	struct sq_mapping **p, *map;
+	struct vm_struct *vma;
+	int page;
 
-	list_del(&map->list);
-	kfree(map);
-}
+	for (p = &sq_mapping_list; (map = *p); p = &map->next)
+		if (map->sq_addr == vaddr)
+			break;
 
-/**
- * sq_clear - Clear a store queue range
- * @addr: Address to start clearing from.
- * @len: Length to clear.
- *
- * A quick zero-fill implementation for clearing out memory that has been
- * remapped through the store queues.
- */
-void sq_clear(unsigned long addr, unsigned int len)
-{
-	int i;
+	if (unlikely(!map)) {
+		printk("%s: bad store queue address 0x%08lx\n",
+		       __FUNCTION__, vaddr);
+		return;
+	}
 
-	/* Clear out both queues linearly */
-	for (i = 0; i < 8; i++) {
-		ctrl_outl(0, addr + i + 0);
-		ctrl_outl(0, addr + i + 8);
+	page = (map->sq_addr - P4SEG_STORE_QUE) >> PAGE_SHIFT;
+	bitmap_release_region(sq_bitmap, page, get_order(map->size));
+
+#ifdef CONFIG_MMU
+	vma = remove_vm_area((void *)(map->sq_addr & PAGE_MASK));
+	if (!vma) {
+		printk(KERN_ERR "%s: bad address 0x%08lx\n",
+		       __FUNCTION__, map->sq_addr);
+		return;
 	}
+#endif
+
+	sq_mapping_list_del(map);
 
-	sq_flush_range(addr, len);
+	kmem_cache_free(sq_cache, map);
 }
 
-/**
- * sq_vma_unmap - Unmap a VMA range
- * @area: VMA containing range.
- * @addr: Start of range.
- * @len: Length of range.
+/*
+ * Needlessly complex sysfs interface. Unfortunately it doesn't seem like
+ * there is any other easy way to add things on a per-cpu basis without
+ * putting the directory entries somewhere stupid and having to create
+ * links in sysfs by hand back in to the per-cpu directories.
  *
- * Searches the sq_mapping_list for a mapping matching the sq addr @addr,
- * and subsequently frees up the entry. Further cleanup is done by generic
- * code.
+ * Some day we may want to have an additional abstraction per store
+ * queue, but considering the kobject hell we already have to deal with,
+ * it's simply not worth the trouble.
  */
-static void sq_vma_unmap(struct vm_area_struct *area,
-			 unsigned long addr, size_t len)
-{
-	struct list_head *pos, *tmp;
+static struct kobject *sq_kobject[NR_CPUS];
 
-	list_for_each_safe(pos, tmp, &sq_mapping_list) {
-		struct sq_mapping *entry;
+struct sq_sysfs_attr {
+	struct attribute attr;
+	ssize_t (*show)(char *buf);
+	ssize_t (*store)(const char *buf, size_t count);
+};
 
-		entry = list_entry(pos, typeof(*entry), list);
+#define to_sq_sysfs_attr(attr)	container_of(attr, struct sq_sysfs_attr, attr)
 
-		if (entry->sq_addr == addr) {
-			/*
-			 * We could probably get away without doing the tlb flush
-			 * here, as generic code should take care of most of this
-			 * when unmapping the rest of the VMA range for us. Leave
-			 * it in for added sanity for the time being..
-			 */
-			__flush_tlb_page(get_asid(), entry->sq_addr & PAGE_MASK);
+static ssize_t sq_sysfs_show(struct kobject *kobj, struct attribute *attr,
+			     char *buf)
+{
+	struct sq_sysfs_attr *sattr = to_sq_sysfs_attr(attr);
 
-			list_del(&entry->list);
-			kfree(entry);
+	if (likely(sattr->show))
+		return sattr->show(buf);
 
-			return;
-		}
-	}
+	return -EIO;
 }
 
-/**
- * sq_vma_sync - Sync a VMA range
- * @area: VMA containing range.
- * @start: Start of range.
- * @len: Length of range.
- * @flags: Additional flags.
- *
- * Synchronizes an sq mapped range by flushing the store queue cache for
- * the duration of the mapping.
- *
- * Used internally for user mappings, which must use msync() to prefetch
- * the store queue cache.
- */
-static int sq_vma_sync(struct vm_area_struct *area,
-		       unsigned long start, size_t len, unsigned int flags)
+static ssize_t sq_sysfs_store(struct kobject *kobj, struct attribute *attr,
+			      const char *buf, size_t count)
 {
-	sq_flush_range(start, len);
+	struct sq_sysfs_attr *sattr = to_sq_sysfs_attr(attr);
 
-	return 0;
+	if (likely(sattr->store))
+		return sattr->store(buf, count);
+
+	return -EIO;
 }
 
-static struct vm_operations_struct sq_vma_ops = {
-	.unmap	= sq_vma_unmap,
-	.sync	= sq_vma_sync,
-};
+static ssize_t mapping_show(char *buf)
+{
+	struct sq_mapping **list, *entry;
+	char *p = buf;
 
-/**
- * sq_mmap - mmap() for /dev/cpu/sq
- * @file: unused.
- * @vma: VMA to remap.
- *
- * Remap the specified vma @vma through the store queues, and setup associated
- * information for the new mapping. Also build up the page tables for the new
- * area.
- */
-static int sq_mmap(struct file *file, struct vm_area_struct *vma)
+	for (list = &sq_mapping_list; (entry = *list); list = &entry->next)
+		p += sprintf(p, "%08lx-%08lx [%08lx]: %s\n",
+			     entry->sq_addr, entry->sq_addr + entry->size,
+			     entry->addr, entry->name);
+
+	return p - buf;
+}
+
+static ssize_t mapping_store(const char *buf, size_t count)
 {
-	unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
-	unsigned long size = vma->vm_end - vma->vm_start;
-	struct sq_mapping *map;
+	unsigned long base = 0, len = 0;
 
-	/*
-	 * We're not interested in any arbitrary virtual address that has
-	 * been stuck in the VMA, as we already know what addresses we
-	 * want. Save off the size, and reposition the VMA to begin at
-	 * the next available sq address.
-	 */
-	vma->vm_start = __sq_get_next_addr();
-	vma->vm_end   = vma->vm_start + size;
+	sscanf(buf, "%lx %lx", &base, &len);
+	if (!base)
+		return -EIO;
 
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	if (likely(len)) {
+		int ret = sq_remap(base, len, "Userspace",
+				   pgprot_val(PAGE_SHARED));
+		if (ret < 0)
+			return ret;
+	} else
+		sq_unmap(base);
 
-	vma->vm_flags |= VM_IO | VM_RESERVED;
+	return count;
+}
 
-	map = __sq_alloc_mapping(vma->vm_start, offset, size, "Userspace");
+static struct sq_sysfs_attr mapping_attr =
+	__ATTR(mapping, 0644, mapping_show, mapping_store);
 
-	if (io_remap_pfn_range(vma, map->sq_addr, map->addr >> PAGE_SHIFT,
-				size, vma->vm_page_prot))
-		return -EAGAIN;
+static struct attribute *sq_sysfs_attrs[] = {
+	&mapping_attr.attr,
+	NULL,
+};
 
-	vma->vm_ops = &sq_vma_ops;
+static struct sysfs_ops sq_sysfs_ops = {
+	.show	= sq_sysfs_show,
+	.store	= sq_sysfs_store,
+};
 
-	return 0;
-}
+static struct kobj_type ktype_percpu_entry = {
+	.sysfs_ops	= &sq_sysfs_ops,
+	.default_attrs	= sq_sysfs_attrs,
+};
 
-#ifdef CONFIG_PROC_FS
-static int sq_mapping_read_proc(char *buf, char **start, off_t off,
-				int len, int *eof, void *data)
+static int __devinit sq_sysdev_add(struct sys_device *sysdev)
 {
-	struct list_head *pos;
-	char *p = buf;
+	unsigned int cpu = sysdev->id;
+	struct kobject *kobj;
 
-	list_for_each_prev(pos, &sq_mapping_list) {
-		struct sq_mapping *entry;
+	sq_kobject[cpu] = kzalloc(sizeof(struct kobject), GFP_KERNEL);
+	if (unlikely(!sq_kobject[cpu]))
+		return -ENOMEM;
 
-		entry = list_entry(pos, typeof(*entry), list);
+	kobj = sq_kobject[cpu];
+	kobj->parent = &sysdev->kobj;
+	kobject_set_name(kobj, "%s", "sq");
+	kobj->ktype = &ktype_percpu_entry;
 
-		p += sprintf(p, "%08lx-%08lx [%08lx]: %s\n", entry->sq_addr,
-			     entry->sq_addr + entry->size - 1, entry->addr,
-			     entry->name);
-	}
-
-	return p - buf;
+	return kobject_register(kobj);
 }
-#endif
 
-static struct file_operations sq_fops = {
-	.owner		= THIS_MODULE,
-	.mmap		= sq_mmap,
-};
+static int __devexit sq_sysdev_remove(struct sys_device *sysdev)
+{
+	unsigned int cpu = sysdev->id;
+	struct kobject *kobj = sq_kobject[cpu];
 
-static struct miscdevice sq_dev = {
-	.minor		= STORE_QUEUE_MINOR,
-	.name		= "sq",
-	.fops		= &sq_fops,
+	kobject_unregister(kobj);
+	return 0;
+}
+
+static struct sysdev_driver sq_sysdev_driver = {
+	.add		= sq_sysdev_add,
+	.remove		= __devexit_p(sq_sysdev_remove),
 };
 
 static int __init sq_api_init(void)
 {
-	int ret;
+	unsigned int nr_pages = 0x04000000 >> PAGE_SHIFT;
+	unsigned int size = (nr_pages + (BITS_PER_LONG - 1)) / BITS_PER_LONG;
+	int ret = -ENOMEM;
+
 	printk(KERN_NOTICE "sq: Registering store queue API.\n");
 
-	create_proc_read_entry("sq_mapping", 0, 0, sq_mapping_read_proc, 0);
+	sq_cache = kmem_cache_create("store_queue_cache",
+				sizeof(struct sq_mapping), 0, 0,
+				NULL, NULL);
+	if (unlikely(!sq_cache))
+		return ret;
 
-	ret = misc_register(&sq_dev);
-	if (ret)
-		remove_proc_entry("sq_mapping", NULL);
+	sq_bitmap = kzalloc(size, GFP_KERNEL);
+	if (unlikely(!sq_bitmap))
+		goto out;
+
+	ret = sysdev_driver_register(&cpu_sysdev_class, &sq_sysdev_driver);
+	if (unlikely(ret != 0))
+		goto out;
+
+	return 0;
+
+out:
+	kfree(sq_bitmap);
+	kmem_cache_destroy(sq_cache);
 
 	return ret;
 }
 
 static void __exit sq_api_exit(void)
 {
-	misc_deregister(&sq_dev);
-	remove_proc_entry("sq_mapping", NULL);
+	sysdev_driver_unregister(&cpu_sysdev_class, &sq_sysdev_driver);
+	kfree(sq_bitmap);
+	kmem_cache_destroy(sq_cache);
 }
 
 module_init(sq_api_init);
@@ -448,11 +402,7 @@ module_exit(sq_api_exit);
 MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>, M. R. Brown <mrbrown@0xd6.org>");
 MODULE_DESCRIPTION("Simple API for SH-4 integrated Store Queues");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(STORE_QUEUE_MINOR);
 
 EXPORT_SYMBOL(sq_remap);
 EXPORT_SYMBOL(sq_unmap);
-EXPORT_SYMBOL(sq_clear);
-EXPORT_SYMBOL(sq_flush);
 EXPORT_SYMBOL(sq_flush_range);
-
diff --git a/include/asm-sh/cpu-sh4/sq.h b/include/asm-sh/cpu-sh4/sq.h
index 366b09166d3b..586d6491816a 100644
--- a/include/asm-sh/cpu-sh4/sq.h
+++ b/include/asm-sh/cpu-sh4/sq.h
@@ -17,7 +17,7 @@
  * Store queues range from e0000000-e3fffffc, allowing approx. 64MB to be
  * mapped to any physical address space. Since data is written (and aligned)
  * to 32-byte boundaries, we need to be sure that all allocations are aligned.
- */ 
+ */
 #define SQ_SIZE                 32
 #define SQ_ALIGN_MASK           (~(SQ_SIZE - 1))
 #define SQ_ALIGN(addr)          (((addr)+SQ_SIZE-1) & SQ_ALIGN_MASK)
@@ -26,23 +26,10 @@
 #define SQ_QACR1		(P4SEG_REG_BASE  + 0x3c)
 #define SQ_ADDRMAX              (P4SEG_STORE_QUE + 0x04000000)
 
-struct sq_mapping {
-	const char *name;
-
-	unsigned long sq_addr;
-	unsigned long addr;
-	unsigned int size;
-
-	struct list_head list;
-};
-
 /* arch/sh/kernel/cpu/sh4/sq.c */
-extern struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *name);
-extern void sq_unmap(struct sq_mapping *map);
-
-extern void sq_clear(unsigned long addr, unsigned int len);
-extern void sq_flush(void *addr);
-extern void sq_flush_range(unsigned long start, unsigned int len);
+unsigned long sq_remap(unsigned long phys, unsigned int size,
+		       const char *name, unsigned long flags);
+void sq_unmap(unsigned long vaddr);
+void sq_flush_range(unsigned long start, unsigned int len);
 
 #endif /* __ASM_CPU_SH4_SQ_H */
-
-- 
cgit v1.2.3


From 5283ecb5ccbdb90d49fce6488d3944bba63a591c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:59:17 +0900
Subject: sh: Add support for R7780RP and R7780MP boards.

This adds support for the Renesas SH7780 development boards,
R7780RP and R7780MP.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                          |    2 +
 arch/sh/Makefile                         |    1 +
 arch/sh/boards/renesas/r7780rp/Kconfig   |   14 +
 arch/sh/boards/renesas/r7780rp/Makefile  |    6 +
 arch/sh/boards/renesas/r7780rp/io.c      |  338 +++++++++
 arch/sh/boards/renesas/r7780rp/irq.c     |  123 ++++
 arch/sh/boards/renesas/r7780rp/led.c     |   45 ++
 arch/sh/boards/renesas/r7780rp/setup.c   |  164 +++++
 arch/sh/configs/r7780rp_defconfig        | 1099 ++++++++++++++++++++++++++++++
 arch/sh/drivers/dma/Kconfig              |    3 +-
 arch/sh/drivers/dma/dma-sh.c             |    4 -
 arch/sh/drivers/pci/Makefile             |    2 +
 arch/sh/drivers/pci/fixups-r7780rp.c     |   39 ++
 arch/sh/drivers/pci/ops-r7780rp.c        |   77 +++
 arch/sh/drivers/pci/pci-auto.c           |   44 +-
 arch/sh/drivers/pci/pci-sh7751.c         |    4 +-
 arch/sh/drivers/pci/pci-sh7780.c         |  341 +++++++++
 arch/sh/drivers/pci/pci-sh7780.h         |  168 +++++
 arch/sh/kernel/cpu/irq/intc2.c           |    6 +-
 arch/sh/tools/mach-types                 |    2 +
 include/asm-sh/cpu-sh4/addrspace.h       |    3 +
 include/asm-sh/cpu-sh4/dma-sh7780.h      |   39 ++
 include/asm-sh/cpu-sh4/dma.h             |   11 +-
 include/asm-sh/hs7751rvoip/hs7751rvoip.h |    2 -
 include/asm-sh/irq-sh7780.h              |    5 -
 include/asm-sh/irq.h                     |    2 +-
 include/asm-sh/landisk/iodata_landisk.h  |    2 -
 include/asm-sh/r7780rp/ide.h             |    8 +
 include/asm-sh/r7780rp/r7780rp.h         |  177 +++++
 include/asm-sh/rts7751r2d/rts7751r2d.h   |    2 -
 30 files changed, 2684 insertions(+), 49 deletions(-)
 create mode 100644 arch/sh/boards/renesas/r7780rp/Kconfig
 create mode 100644 arch/sh/boards/renesas/r7780rp/Makefile
 create mode 100644 arch/sh/boards/renesas/r7780rp/io.c
 create mode 100644 arch/sh/boards/renesas/r7780rp/irq.c
 create mode 100644 arch/sh/boards/renesas/r7780rp/led.c
 create mode 100644 arch/sh/boards/renesas/r7780rp/setup.c
 create mode 100644 arch/sh/configs/r7780rp_defconfig
 create mode 100644 arch/sh/drivers/pci/fixups-r7780rp.c
 create mode 100644 arch/sh/drivers/pci/ops-r7780rp.c
 create mode 100644 arch/sh/drivers/pci/pci-sh7780.c
 create mode 100644 arch/sh/drivers/pci/pci-sh7780.h
 create mode 100644 include/asm-sh/cpu-sh4/dma-sh7780.h
 create mode 100644 include/asm-sh/r7780rp/ide.h
 create mode 100644 include/asm-sh/r7780rp/r7780rp.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index a66b9c44075f..fe982f0eba23 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -404,6 +404,8 @@ source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
 
 source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
 
+source "arch/sh/boards/renesas/r7780rp/Kconfig"
+
 config SH_PCLK_FREQ
 	int "Peripheral clock frequency (in Hz)"
 	default "50000000" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7780
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 65676c33822e..c143510d71fc 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -104,6 +104,7 @@ machdir-$(CONFIG_SH_HS7751RVOIP)		:= renesas/hs7751rvoip
 machdir-$(CONFIG_SH_RTS7751R2D)			:= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
+machdir-$(CONFIG_SH_R7780RP)			:= renesas/r7780rp
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_TITAN)			:= titan
diff --git a/arch/sh/boards/renesas/r7780rp/Kconfig b/arch/sh/boards/renesas/r7780rp/Kconfig
new file mode 100644
index 000000000000..c26d9813d239
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/Kconfig
@@ -0,0 +1,14 @@
+if SH_R7780RP
+
+menu "R7780RP options"
+
+config SH_R7780MP
+	bool "R7780MP board support"
+	default y
+	help
+	  Selecting this option will enable support for the mass-production
+	  version of the R7780RP. If in doubt, say Y.
+
+endmenu
+
+endif
diff --git a/arch/sh/boards/renesas/r7780rp/Makefile b/arch/sh/boards/renesas/r7780rp/Makefile
new file mode 100644
index 000000000000..f1776d027978
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the R7780RP-1 specific parts of the kernel
+#
+
+obj-y	 := setup.o io.o irq.o
+obj-$(CONFIG_HEARTBEAT)	+= led.o
diff --git a/arch/sh/boards/renesas/r7780rp/io.c b/arch/sh/boards/renesas/r7780rp/io.c
new file mode 100644
index 000000000000..f73ca3f0f5a1
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/io.c
@@ -0,0 +1,338 @@
+/*
+ * linux/arch/sh/kernel/io_r7780rp.c
+ *
+ * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
+ * Based largely on io_se.c.
+ *
+ * I/O routine for Renesas Solutions Highlander R7780RP-1
+ *
+ * Initial version only to support LAN access; some
+ * placeholder code from io_r7780rp.c left in with the
+ * expectation of later SuperIO and PCMCIA access.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/r7780rp/r7780rp.h>
+#include <asm/addrspace.h>
+#include <asm/io.h>
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include "../../../drivers/pci/pci-sh7780.h"
+
+/*
+ * The 7780 R7780RP-1 uses the built-in PCI controller (PCIC)
+ * of the 7780 processor, and has a SuperIO accessible via the PCI.
+ * The board also includes a PCMCIA controller on its memory bus,
+ * like the other Solution Engine boards.
+ */
+
+#define	SH7780_PCIIOBR_MASK	0xFFFC0000	/* IO Space Mask */
+#define PCIIOBR		(volatile long *)PCI_REG(SH7780_PCIIOBR)
+#define PCIMBR          (volatile long *)PCI_REG(SH7780_PCIMBR)
+#define PCI_IO_AREA	SH7780_PCI_IO_BASE
+#define PCI_MEM_AREA	SH7780_PCI_CONFIG_BASE
+
+#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7780_PCIIOBR_MASK))
+
+static inline void delay(void)
+{
+	ctrl_inw(0xa0000000);
+}
+
+static inline unsigned long port2adr(unsigned int port)
+{
+	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
+		if (port == 0x3f6)
+			return (PA_AREA5_IO + 0x80c);
+		else
+			return (PA_AREA5_IO + 0x1000 + ((port-0x1f0) << 1));
+	else
+		maybebadio((unsigned long)port);
+
+	return port;
+}
+
+static inline unsigned long port88796l(unsigned int port, int flag)
+{
+	unsigned long addr;
+
+	if (flag)
+		addr = PA_AX88796L + ((port - AX88796L_IO_BASE) << 1);
+	else
+		addr = PA_AX88796L + ((port - AX88796L_IO_BASE) << 1) + 0x1000;
+
+	return addr;
+}
+
+/* The 7780 R7780RP-1 seems to have everything hooked */
+/* up pretty normally (nothing on high-bytes only...) so this */
+/* shouldn't be needed */
+static inline int shifted_port(unsigned long port)
+{
+	/* For IDE registers, value is not shifted */
+	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
+		return 0;
+	else
+		return 1;
+}
+
+/* In case someone configures the kernel w/o PCI support: in that */
+/* scenario, don't ever bother to check for PCI-window addresses */
+
+/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
+#if defined(CONFIG_PCI)
+#define CHECK_SH7780_PCIIO(port) \
+  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7780_PCI_IO_SIZE)))
+#else
+#define CHECK_SH7780_PCIIO(port) (0)
+#endif
+
+#if defined(CONFIG_NE2000) || defined(CONFIG_NE2000_MODULE)
+#define CHECK_AX88796L_PORT(port) \
+  ((port >= AX88796L_IO_BASE) && (port < (AX88796L_IO_BASE+0x20)))
+#else
+#define CHECK_AX88796L_PORT(port) (0)
+#endif
+
+/*
+ * General outline: remap really low stuff [eventually] to SuperIO,
+ * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
+ * is mapped through the PCI IO window.  Stuff with high bits (PXSEG)
+ * should be way beyond the window, and is used  w/o translation for
+ * compatibility.
+ */
+u8 r7780rp_inb(unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		return ctrl_inw(port88796l(port, 0)) & 0xff;
+	else if (PXSEG(port))
+		return ctrl_inb(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return ctrl_inb(PCI_IOMAP(port));
+
+	return ctrl_inw(port2adr(port)) & 0xff;
+}
+
+u8 r7780rp_inb_p(unsigned long port)
+{
+	u8 v;
+
+	if (CHECK_AX88796L_PORT(port))
+		v = ctrl_inw(port88796l(port, 0)) & 0xff;
+	else if (PXSEG(port))
+		v = ctrl_inb(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		v = ctrl_inb(PCI_IOMAP(port));
+	else
+		v = ctrl_inw(port2adr(port)) & 0xff;
+
+	delay();
+
+	return v;
+}
+
+u16 r7780rp_inw(unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		return ctrl_inw(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return ctrl_inw(PCI_IOMAP(port));
+	else
+		maybebadio(port);
+
+	return 0;
+}
+
+u32 r7780rp_inl(unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		return ctrl_inl(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return ctrl_inl(PCI_IOMAP(port));
+	else
+		maybebadio(port);
+
+	return 0;
+}
+
+void r7780rp_outb(u8 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		ctrl_outw(value, port88796l(port, 0));
+	else if (PXSEG(port))
+		ctrl_outb(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outb(value, PCI_IOMAP(port));
+	else
+		ctrl_outw(value, port2adr(port));
+}
+
+void r7780rp_outb_p(u8 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		ctrl_outw(value, port88796l(port, 0));
+	else if (PXSEG(port))
+		ctrl_outb(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outb(value, PCI_IOMAP(port));
+	else
+		ctrl_outw(value, port2adr(port));
+
+	delay();
+}
+
+void r7780rp_outw(u16 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		ctrl_outw(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outw(value, PCI_IOMAP(port));
+	else
+		maybebadio(port);
+}
+
+void r7780rp_outl(u32 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		ctrl_outl(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outl(value, PCI_IOMAP(port));
+	else
+		maybebadio(port);
+}
+
+void r7780rp_insb(unsigned long port, void *dst, unsigned long count)
+{
+	volatile u16 *p;
+	u8 *buf = dst;
+
+	if (CHECK_AX88796L_PORT(port)) {
+		p = (volatile u16 *)port88796l(port, 0);
+		while (count--)
+			*buf++ = *p & 0xff;
+	} else if (PXSEG(port)) {
+		while (count--)
+			*buf++ = *(volatile u8 *)port;
+	} else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+
+		while (count--)
+			*buf++ = *bp;
+	} else {
+		p = (volatile u16 *)port2adr(port);
+		while (count--)
+			*buf++ = *p & 0xff;
+	}
+}
+
+void r7780rp_insw(unsigned long port, void *dst, unsigned long count)
+{
+	volatile u16 *p;
+	u16 *buf = dst;
+
+	if (CHECK_AX88796L_PORT(port))
+		p = (volatile u16 *)port88796l(port, 1);
+	else if (PXSEG(port))
+		p = (volatile u16 *)port;
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		p = (volatile u16 *)PCI_IOMAP(port);
+	else
+		p = (volatile u16 *)port2adr(port);
+
+	while (count--)
+		*buf++ = *p;
+}
+
+void r7780rp_insl(unsigned long port, void *dst, unsigned long count)
+{
+	u32 *buf = dst;
+
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+
+		while (count--)
+			*buf++ = *p;
+	} else
+		maybebadio(port);
+}
+
+void r7780rp_outsb(unsigned long port, const void *src, unsigned long count)
+{
+	volatile u16 *p;
+	const u8 *buf = src;
+
+	if (CHECK_AX88796L_PORT(port)) {
+		p = (volatile u16 *)port88796l(port, 0);
+		while (count--)
+			*p = *buf++;
+	} else if (PXSEG(port))
+		while (count--)
+			ctrl_outb(*buf++, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+
+		while (count--)
+			*bp = *buf++;
+	} else {
+		p = (volatile u16 *)port2adr(port);
+		while (count--)
+			*p = *buf++;
+	}
+}
+
+void r7780rp_outsw(unsigned long port, const void *src, unsigned long count)
+{
+	volatile u16 *p;
+	const u16 *buf = src;
+
+	if (CHECK_AX88796L_PORT(port))
+		p = (volatile u16 *)port88796l(port, 1);
+	else if (PXSEG(port))
+		p = (volatile u16 *)port;
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		p = (volatile u16 *)PCI_IOMAP(port);
+	else
+		p = (volatile u16 *)port2adr(port);
+
+	while (count--)
+		*p = *buf++;
+}
+
+void r7780rp_outsl(unsigned long port, const void *src, unsigned long count)
+{
+	const u32 *buf = src;
+
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+
+		while (count--)
+			*p = *buf++;
+	} else
+		maybebadio(port);
+}
+
+void __iomem *r7780rp_ioport_map(unsigned long port, unsigned int size)
+{
+	if (CHECK_AX88796L_PORT(port))
+		return (void __iomem *)port88796l(port, size > 1);
+	else if (PXSEG(port))
+		return (void __iomem *)port;
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return (void __iomem *)PCI_IOMAP(port);
+
+	return (void __iomem *)port2adr(port);
+}
diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c
new file mode 100644
index 000000000000..c7b9fe6e640d
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/irq.c
@@ -0,0 +1,123 @@
+/*
+ * linux/arch/sh/boards/renesas/r7780rp/irq.c
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Renesas Solutions Highlander R7780RP-1 Support.
+ *
+ * Modified for R7780RP-1 by
+ * Atom Create Engineering Co., Ltd. 2002.
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/r7780rp/r7780rp.h>
+
+#ifdef CONFIG_SH_R7780MP
+static int mask_pos[] = {12, 11, 9, 14, 15, 8, 13, 6, 5, 4, 3, 2, 0, 0, 1, 0};
+#else
+static int mask_pos[] = {15, 14, 13, 12, 11, 10, 9, 8, 7, 5, 6, 4, 0, 1, 2, 0};
+#endif
+
+static void enable_r7780rp_irq(unsigned int irq);
+static void disable_r7780rp_irq(unsigned int irq);
+
+/* shutdown is same as "disable" */
+#define shutdown_r7780rp_irq disable_r7780rp_irq
+
+static void ack_r7780rp_irq(unsigned int irq);
+static void end_r7780rp_irq(unsigned int irq);
+
+static unsigned int startup_r7780rp_irq(unsigned int irq)
+{
+	enable_r7780rp_irq(irq);
+	return 0; /* never anything pending */
+}
+
+static void disable_r7780rp_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned short val;
+	unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]);
+
+	/* Set the priority in IPR to 0 */
+	local_irq_save(flags);
+	val = ctrl_inw(IRLCNTR1);
+	val &= mask;
+	ctrl_outw(val, IRLCNTR1);
+	local_irq_restore(flags);
+}
+
+static void enable_r7780rp_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned short val;
+	unsigned short value = (0x0001 << mask_pos[irq]);
+
+	/* Set priority in IPR back to original value */
+	local_irq_save(flags);
+	val = ctrl_inw(IRLCNTR1);
+	val |= value;
+	ctrl_outw(val, IRLCNTR1);
+	local_irq_restore(flags);
+}
+
+static void ack_r7780rp_irq(unsigned int irq)
+{
+	disable_r7780rp_irq(irq);
+}
+
+static void end_r7780rp_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
+		enable_r7780rp_irq(irq);
+}
+
+static struct hw_interrupt_type r7780rp_irq_type = {
+	.typename = "R7780RP-IRQ",
+	.startup = startup_r7780rp_irq,
+	.shutdown = shutdown_r7780rp_irq,
+	.enable = enable_r7780rp_irq,
+	.disable = disable_r7780rp_irq,
+	.ack = ack_r7780rp_irq,
+	.end = end_r7780rp_irq,
+};
+
+static void make_r7780rp_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &r7780rp_irq_type;
+	disable_r7780rp_irq(irq);
+}
+
+/*
+ * Initialize IRQ setting
+ */
+void __init init_r7780rp_IRQ(void)
+{
+	int i;
+
+	/* IRL0=PCI Slot #A
+	 * IRL1=PCI Slot #B
+	 * IRL2=PCI Slot #C
+	 * IRL3=PCI Slot #D
+	 * IRL4=CF Card
+	 * IRL5=CF Card Insert
+	 * IRL6=M66596
+	 * IRL7=SD Card
+	 * IRL8=Touch Panel
+	 * IRL9=SCI
+	 * IRL10=Serial
+	 * IRL11=Extention #A
+	 * IRL11=Extention #B
+	 * IRL12=Debug LAN
+	 * IRL13=Push Switch
+	 * IRL14=ZiggBee IO
+	 */
+
+	for (i=0; i<15; i++)
+		make_r7780rp_irq(i);
+}
diff --git a/arch/sh/boards/renesas/r7780rp/led.c b/arch/sh/boards/renesas/r7780rp/led.c
new file mode 100644
index 000000000000..9f02766b6f53
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/led.c
@@ -0,0 +1,45 @@
+/*
+ * Copyright (C) Atom Create Engineering Co., Ltd.
+ *
+ * May be copied or modified under the terms of GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * This file contains Renesas Solutions HIGHLANDER R7780RP-1 specific LED code.
+ */
+
+#include <linux/config.h>
+#include <linux/sched.h>
+#include <asm/io.h>
+#include <asm/r7780rp/r7780rp.h>
+
+/* Cycle the LED's in the clasic Knightriger/Sun pattern */
+void heartbeat_r7780rp(void)
+{
+	static unsigned int cnt = 0, period = 0;
+	volatile unsigned short *p = (volatile unsigned short *)PA_OBLED;
+	static unsigned bit = 0, up = 1;
+	unsigned bit_pos[] = {2, 1, 0, 3, 6, 5, 4, 7};
+
+	cnt += 1;
+	if (cnt < period)
+		return;
+
+	cnt = 0;
+
+	/* Go through the points (roughly!):
+	 * f(0)=10, f(1)=16, f(2)=20, f(5)=35, f(int)->110
+	 */
+	period = 110 - ((300 << FSHIFT)/((avenrun[0]/5) + (3<<FSHIFT)));
+
+	*p = 1 << bit_pos[bit];
+	if (up)
+		if (bit == 7) {
+			bit--;
+			up = 0;
+		} else
+			bit++;
+	else if (bit == 0)
+		up = 1;
+	else
+		bit--;
+}
diff --git a/arch/sh/boards/renesas/r7780rp/setup.c b/arch/sh/boards/renesas/r7780rp/setup.c
new file mode 100644
index 000000000000..0e4d5e1b541c
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/setup.c
@@ -0,0 +1,164 @@
+/*
+ * arch/sh/boards/renesas/r7780rp/setup.c
+ *
+ * Copyright (C) 2002 Atom Create Engineering Co., Ltd.
+ * Copyright (C) 2005, 2006 Paul Mundt
+ *
+ * Renesas Solutions Highlander R7780RP-1 Support.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <asm/machvec.h>
+#include <asm/r7780rp/r7780rp.h>
+#include <asm/clock.h>
+#include <asm/io.h>
+
+extern void heartbeat_r7780rp(void);
+extern void init_r7780rp_IRQ(void);
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_r7780rp __initmv = {
+	.mv_nr_irqs		= 109,
+
+	.mv_inb			= r7780rp_inb,
+	.mv_inw			= r7780rp_inw,
+	.mv_inl			= r7780rp_inl,
+	.mv_outb		= r7780rp_outb,
+	.mv_outw		= r7780rp_outw,
+	.mv_outl		= r7780rp_outl,
+
+	.mv_inb_p		= r7780rp_inb_p,
+	.mv_inw_p		= r7780rp_inw,
+	.mv_inl_p		= r7780rp_inl,
+	.mv_outb_p		= r7780rp_outb_p,
+	.mv_outw_p		= r7780rp_outw,
+	.mv_outl_p		= r7780rp_outl,
+
+	.mv_insb		= r7780rp_insb,
+	.mv_insw		= r7780rp_insw,
+	.mv_insl		= r7780rp_insl,
+	.mv_outsb		= r7780rp_outsb,
+	.mv_outsw		= r7780rp_outsw,
+	.mv_outsl		= r7780rp_outsl,
+
+	.mv_ioport_map		= r7780rp_ioport_map,
+	.mv_init_irq		= init_r7780rp_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_r7780rp,
+#endif
+};
+ALIAS_MV(r7780rp)
+
+static struct resource m66596_usb_host_resources[] = {
+	[0] = {
+		.start	= 0xa4800000,
+		.end	= 0xa4ffffff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= 6,		/* irq number */
+		.end	= 6,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device m66596_usb_host_device = {
+	.name		= "m66596-hcd",
+	.id		= 0,
+	.dev = {
+		.dma_mask		= NULL,		/* don't use dma */
+		.coherent_dma_mask	= 0xffffffff,
+	},
+	.num_resources	= ARRAY_SIZE(m66596_usb_host_resources),
+	.resource	= m66596_usb_host_resources,
+};
+
+static struct platform_device *r7780rp_devices[] __initdata = {
+	&m66596_usb_host_device,
+};
+
+static int __init r7780rp_devices_setup(void)
+{
+	return platform_add_devices(r7780rp_devices,
+				    ARRAY_SIZE(r7780rp_devices));
+}
+__initcall(r7780rp_devices_setup);
+
+/*
+ * Platform specific clocks
+ */
+static void ivdr_clk_enable(struct clk *clk)
+{
+	ctrl_outw(ctrl_inw(PA_IVDRCTL) | (1 << 8), PA_IVDRCTL);
+}
+
+static void ivdr_clk_disable(struct clk *clk)
+{
+	ctrl_outw(ctrl_inw(PA_IVDRCTL) & ~(1 << 8), PA_IVDRCTL);
+}
+
+static struct clk_ops ivdr_clk_ops = {
+	.enable		= ivdr_clk_enable,
+	.disable	= ivdr_clk_disable,
+};
+
+static struct clk ivdr_clk = {
+	.name		= "ivdr_clk",
+	.ops		= &ivdr_clk_ops,
+};
+
+static struct clk *r7780rp_clocks[] = {
+	&ivdr_clk,
+};
+
+const char *get_system_type(void)
+{
+	return "Highlander R7780RP-1";
+}
+
+static void r7780rp_power_off(void)
+{
+#ifdef CONFIG_SH_R7780MP
+	ctrl_outw(0x0001, PA_POFF);
+#endif
+}
+
+/*
+ * Initialize the board
+ */
+void __init platform_setup(void)
+{
+	u16 ver = ctrl_inw(PA_VERREG);
+	int i;
+
+	printk(KERN_INFO "Renesas Solutions Highlander R7780RP-1 support.\n");
+
+	printk(KERN_INFO "Board version: %d (revision %d), "
+			 "FPGA version: %d (revision %d)\n",
+			 (ver >> 12) & 0xf, (ver >> 8) & 0xf,
+			 (ver >>  4) & 0xf, ver & 0xf);
+
+	/*
+	 * Enable the important clocks right away..
+	 */
+	for (i = 0; i < ARRAY_SIZE(r7780rp_clocks); i++) {
+		struct clk *clk = r7780rp_clocks[i];
+
+		clk_register(clk);
+		clk_enable(clk);
+	}
+
+	ctrl_outw(0x0000, PA_OBLED);	/* Clear LED. */
+#ifndef CONFIG_SH_R7780MP
+	ctrl_outw(0x0001, PA_SDPOW);	/* SD Power ON */
+#endif
+	ctrl_outw(ctrl_inw(PA_IVDRCTL) | 0x0100, PA_IVDRCTL);	/* Si13112 */
+
+	pm_power_off = r7780rp_power_off;
+}
diff --git a/arch/sh/configs/r7780rp_defconfig b/arch/sh/configs/r7780rp_defconfig
new file mode 100644
index 000000000000..d597fc571549
--- /dev/null
+++ b/arch/sh/configs/r7780rp_defconfig
@@ -0,0 +1,1099 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.15-sh
+# Sat Jan  7 19:47:53 2006
+#
+CONFIG_SUPERH=y
+CONFIG_UID16=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+CONFIG_BSD_PROCESS_ACCT=y
+# CONFIG_BSD_PROCESS_ACCT_V3 is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R77703DRP is not set
+CONFIG_SH_R7780RP=y
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+CONFIG_CPU_SH4A=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+CONFIG_CPU_SUBTYPE_SH7780=y
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_32BIT=y
+CONFIG_HUGETLB_PAGE_SIZE_64K=y
+# CONFIG_HUGETLB_PAGE_SIZE_1MB is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+CONFIG_MEMORY_START=0x08000000
+CONFIG_MEMORY_SIZE=0x08000000
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_SH_FPU=y
+CONFIG_SH_STORE_QUEUES=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ_BOOL=y
+CONFIG_SH_PCLK_FREQ=32000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=6
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_KEXEC is not set
+CONFIG_PREEMPT=y
+# CONFIG_SMP is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_INTC2_IRQ=y
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="mem=128M console=ttySC0,115200 root=/dev/hda1"
+
+#
+# Bus options
+#
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+CONFIG_PCI_LEGACY_PROC=y
+# CONFIG_PCI_DEBUG is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_HOTPLUG_PCI_FAKE is not set
+# CONFIG_HOTPLUG_PCI_CPCI is not set
+# CONFIG_HOTPLUG_PCI_SHPC is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_MULTIPLE_TABLES is not set
+# CONFIG_IP_ROUTE_MULTIPATH is not set
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+CONFIG_BRIDGE=m
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+CONFIG_IDE=m
+CONFIG_IDE_MAX_HWIFS=4
+CONFIG_BLK_DEV_IDE=m
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+CONFIG_BLK_DEV_IDE_SATA=y
+CONFIG_BLK_DEV_IDEDISK=m
+CONFIG_IDEDISK_MULTI_MODE=y
+# CONFIG_BLK_DEV_IDECD is not set
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+CONFIG_BLK_DEV_IDESCSI=m
+# CONFIG_IDE_TASK_IOCTL is not set
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=m
+CONFIG_BLK_DEV_IDEPCI=y
+CONFIG_IDEPCI_SHARE_IRQ=y
+# CONFIG_BLK_DEV_OFFBOARD is not set
+CONFIG_BLK_DEV_GENERIC=m
+# CONFIG_BLK_DEV_OPTI621 is not set
+CONFIG_BLK_DEV_IDEDMA_PCI=y
+# CONFIG_BLK_DEV_IDEDMA_FORCED is not set
+CONFIG_IDEDMA_PCI_AUTO=y
+# CONFIG_IDEDMA_ONLYDISK is not set
+CONFIG_BLK_DEV_AEC62XX=m
+# CONFIG_BLK_DEV_ALI15X3 is not set
+# CONFIG_BLK_DEV_AMD74XX is not set
+# CONFIG_BLK_DEV_CMD64X is not set
+# CONFIG_BLK_DEV_TRIFLEX is not set
+# CONFIG_BLK_DEV_CY82C693 is not set
+# CONFIG_BLK_DEV_CS5520 is not set
+# CONFIG_BLK_DEV_CS5530 is not set
+# CONFIG_BLK_DEV_HPT34X is not set
+# CONFIG_BLK_DEV_HPT366 is not set
+# CONFIG_BLK_DEV_SC1200 is not set
+# CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
+# CONFIG_BLK_DEV_NS87415 is not set
+# CONFIG_BLK_DEV_PDC202XX_OLD is not set
+CONFIG_BLK_DEV_PDC202XX_NEW=m
+# CONFIG_PDC202XX_FORCE is not set
+# CONFIG_BLK_DEV_SVWKS is not set
+CONFIG_BLK_DEV_SIIMAGE=m
+# CONFIG_BLK_DEV_SLC90E66 is not set
+# CONFIG_BLK_DEV_TRM290 is not set
+# CONFIG_BLK_DEV_VIA82CXXX is not set
+CONFIG_IDE_SH=y
+# CONFIG_IDE_ARM is not set
+CONFIG_BLK_DEV_IDEDMA=y
+# CONFIG_IDEDMA_IVB is not set
+CONFIG_IDEDMA_AUTO=y
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=m
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=m
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+# CONFIG_BLK_DEV_SR is not set
+CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+# CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+
+#
+# SCSI Transport Attributes
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_ATTRS is not set
+
+#
+# SCSI low-level drivers
+#
+# CONFIG_ISCSI_TCP is not set
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+CONFIG_SCSI_QLA2XXX=m
+# CONFIG_SCSI_QLA21XX is not set
+# CONFIG_SCSI_QLA22XX is not set
+# CONFIG_SCSI_QLA2300 is not set
+# CONFIG_SCSI_QLA2322 is not set
+# CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
+# CONFIG_FUSION_SAS is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_CASSINI is not set
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_SMC91X is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+CONFIG_NE2000=y
+CONFIG_NET_PCI=y
+CONFIG_PCNET32=m
+# CONFIG_AMD8111_ETH is not set
+# CONFIG_ADAPTEC_STARFIRE is not set
+# CONFIG_B44 is not set
+# CONFIG_FORCEDETH is not set
+# CONFIG_DGRS is not set
+# CONFIG_EEPRO100 is not set
+# CONFIG_E100 is not set
+# CONFIG_FEALNX is not set
+# CONFIG_NATSEMI is not set
+# CONFIG_NE2K_PCI is not set
+CONFIG_8139CP=m
+CONFIG_8139TOO=m
+# CONFIG_8139TOO_PIO is not set
+# CONFIG_8139TOO_TUNE_TWISTER is not set
+CONFIG_8139TOO_8129=y
+# CONFIG_8139_OLD_RX_RESET is not set
+# CONFIG_SIS900 is not set
+# CONFIG_EPIC100 is not set
+# CONFIG_SUNDANCE is not set
+# CONFIG_TLAN is not set
+CONFIG_VIA_RHINE=m
+CONFIG_VIA_RHINE_MMIO=y
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+CONFIG_E1000=m
+# CONFIG_E1000_NAPI is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+CONFIG_R8169=y
+# CONFIG_R8169_NAPI is not set
+# CONFIG_SIS190 is not set
+# CONFIG_SKGE is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+CONFIG_NET_RADIO=y
+
+#
+# Obsolete Wireless cards support (pre-802.11)
+#
+# CONFIG_STRIP is not set
+
+#
+# Wireless 802.11b ISA/PCI cards support
+#
+CONFIG_HERMES=m
+# CONFIG_PLX_HERMES is not set
+# CONFIG_TMD_HERMES is not set
+# CONFIG_NORTEL_HERMES is not set
+# CONFIG_PCI_HERMES is not set
+# CONFIG_ATMEL is not set
+
+#
+# Prism GT/Duette 802.11(a/b/g) PCI/Cardbus support
+#
+CONFIG_PRISM54=m
+# CONFIG_HOSTAP is not set
+CONFIG_NET_WIRELESS=y
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+CONFIG_INPUT_KEYBOARD=y
+CONFIG_KEYBOARD_ATKBD=y
+# CONFIG_KEYBOARD_SUNKBD is not set
+# CONFIG_KEYBOARD_LKKBD is not set
+# CONFIG_KEYBOARD_XTKBD is not set
+# CONFIG_KEYBOARD_NEWTON is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_PCIPS2 is not set
+CONFIG_SERIO_LIBPS2=y
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia Capabilities Port drivers
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+CONFIG_SOUND=m
+
+#
+# Advanced Linux Sound Architecture
+#
+# CONFIG_SND is not set
+
+#
+# Open Sound System
+#
+CONFIG_SOUND_PRIME=m
+# CONFIG_OBSOLETE_OSS_DRIVER is not set
+# CONFIG_SOUND_FUSION is not set
+# CONFIG_SOUND_ICH is not set
+# CONFIG_SOUND_TRIDENT is not set
+# CONFIG_SOUND_MSNDCLAS is not set
+# CONFIG_SOUND_MSNDPIN is not set
+
+#
+# USB support
+#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+# CONFIG_USB is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+# CONFIG_XFS_FS is not set
+CONFIG_MINIX_FS=y
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+CONFIG_NTFS_FS=y
+# CONFIG_NTFS_DEBUG is not set
+CONFIG_NTFS_RW=y
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+CONFIG_HUGETLBFS=y
+CONFIG_HUGETLB_PAGE=y
+CONFIG_RAMFS=y
+# CONFIG_RELAYFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+CONFIG_NFS_V4=y
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+CONFIG_NFSD_V4=y
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+CONFIG_SUNRPC_GSS=y
+CONFIG_RPCSEC_GSS_KRB5=y
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=y
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+CONFIG_NLS_CODEPAGE_932=y
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+CONFIG_NLS_ISO8859_1=y
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+# CONFIG_NLS_UTF8 is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_MAGIC_SYSRQ is not set
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_DEBUG_SLAB is not set
+CONFIG_DEBUG_PREEMPT=y
+CONFIG_DEBUG_SPINLOCK=y
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_KOBJECT is not set
+# CONFIG_DEBUG_INFO is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_DEBUG_VM is not set
+CONFIG_FRAME_POINTER=y
+# CONFIG_RCU_TORTURE_TEST is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+CONFIG_CRYPTO=y
+CONFIG_CRYPTO_HMAC=y
+# CONFIG_CRYPTO_NULL is not set
+# CONFIG_CRYPTO_MD4 is not set
+CONFIG_CRYPTO_MD5=y
+# CONFIG_CRYPTO_SHA1 is not set
+# CONFIG_CRYPTO_SHA256 is not set
+# CONFIG_CRYPTO_SHA512 is not set
+# CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
+CONFIG_CRYPTO_DES=y
+# CONFIG_CRYPTO_BLOWFISH is not set
+# CONFIG_CRYPTO_TWOFISH is not set
+# CONFIG_CRYPTO_SERPENT is not set
+# CONFIG_CRYPTO_AES is not set
+# CONFIG_CRYPTO_CAST5 is not set
+# CONFIG_CRYPTO_CAST6 is not set
+# CONFIG_CRYPTO_TEA is not set
+# CONFIG_CRYPTO_ARC4 is not set
+# CONFIG_CRYPTO_KHAZAD is not set
+# CONFIG_CRYPTO_ANUBIS is not set
+# CONFIG_CRYPTO_DEFLATE is not set
+# CONFIG_CRYPTO_MICHAEL_MIC is not set
+# CONFIG_CRYPTO_CRC32C is not set
+# CONFIG_CRYPTO_TEST is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
diff --git a/arch/sh/drivers/dma/Kconfig b/arch/sh/drivers/dma/Kconfig
index 0f15216cd39d..defc13c37d48 100644
--- a/arch/sh/drivers/dma/Kconfig
+++ b/arch/sh/drivers/dma/Kconfig
@@ -11,6 +11,8 @@ config SH_DMA
 config NR_ONCHIP_DMA_CHANNELS
 	depends on SH_DMA
 	int "Number of on-chip DMAC channels"
+	default "8" if CPU_SUBTYPE_SH7750R || CPU_SUBTYPE_SH7751R
+	default "12" if CPU_SUBTYPE_SH7780
 	default "4"
 	help
 	  This allows you to specify the number of channels that the on-chip
@@ -52,4 +54,3 @@ config DMA_PAGE_OPS_CHANNEL
 	  are dual-address capable.
 
 endmenu
-
diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c
index e028a2d2a4ea..4428ee809651 100644
--- a/arch/sh/drivers/dma/dma-sh.c
+++ b/arch/sh/drivers/dma/dma-sh.c
@@ -11,14 +11,10 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-
 #include <linux/init.h>
-#include <linux/irq.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <asm/dreamcast/dma.h>
-#include <asm/signal.h>
-#include <asm/irq.h>
 #include <asm/dma.h>
 #include <asm/io.h>
 #include "dma-sh.h"
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 86d6d0660d4c..3d8078f1c05f 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_PCI_AUTO)			+= pci-auto.o
 
 obj-$(CONFIG_CPU_SUBTYPE_ST40STB1)	+= pci-st40.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o 
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o
 
 obj-$(CONFIG_SH_DREAMCAST)		+= ops-dreamcast.o fixups-dreamcast.o \
 					   dma-dreamcast.o
@@ -14,4 +15,5 @@ obj-$(CONFIG_SH_SECUREEDGE5410)		+= ops-snapgear.o
 obj-$(CONFIG_SH_BIGSUR)			+= ops-bigsur.o
 obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
+obj-$(CONFIG_SH_R7780RP)		+= ops-r7780rp.o fixups-r7780rp.o
 obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
diff --git a/arch/sh/drivers/pci/fixups-r7780rp.c b/arch/sh/drivers/pci/fixups-r7780rp.c
new file mode 100644
index 000000000000..b656b562ec99
--- /dev/null
+++ b/arch/sh/drivers/pci/fixups-r7780rp.c
@@ -0,0 +1,39 @@
+/*
+ * arch/sh/drivers/pci/fixups-r7780rp.c
+ *
+ * Highlander R7780RP-1 PCI fixups
+ *
+ * Copyright (C) 2003  Lineo uSolutions, Inc.
+ * Copyright (C) 2004  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include "pci-sh7780.h"
+#include <asm/io.h>
+
+int pci_fixup_pcic(void)
+{
+	outl(0x000043ff, PCI_REG(SH7780_PCIIMR));
+	outl(0x0000380f, PCI_REG(SH7780_PCIAINTM));
+
+	outl(0xfbb00047, PCI_REG(SH7780_PCICMD));
+	outl(0x00000000, PCI_REG(SH7780_PCIIBAR));
+
+	outl(0x00011912, PCI_REG(SH7780_PCISVID));
+	outl(0x08000000, PCI_REG(SH7780_PCICSCR0));
+	outl(0x0000001b, PCI_REG(SH7780_PCICSAR0));
+	outl(0xfd000000, PCI_REG(SH7780_PCICSCR1));
+	outl(0x0000000f, PCI_REG(SH7780_PCICSAR1));
+
+	outl(0xfd000000, PCI_REG(SH7780_PCIMBR0));
+	outl(0x00fc0000, PCI_REG(SH7780_PCIMBMR0));
+
+	/* Set IOBR for windows containing area specified in pci.h */
+	outl((PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE-1)), PCI_REG(SH7780_PCIIOBR));
+	outl(((SH7780_PCI_IO_SIZE-1) & (7<<18)), PCI_REG(SH7780_PCIIOBMR));
+
+	return 0;
+}
+
diff --git a/arch/sh/drivers/pci/ops-r7780rp.c b/arch/sh/drivers/pci/ops-r7780rp.c
new file mode 100644
index 000000000000..3254c4e917a9
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-r7780rp.c
@@ -0,0 +1,77 @@
+/*
+ * Author:  Ian DaSilva (idasilva@mvista.com)
+ *
+ * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * PCI initialization for the Renesas SH7780 Highlander R7780RP-1 board
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+
+#include <asm/io.h>
+#include "pci-sh7780.h"
+#include <asm/r7780rp/r7780rp.h>
+
+int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+{
+        switch (slot) {
+	case 0: return IRQ_PCISLOT1;		/* PCI Interrupt #1 */
+	case 1: return IRQ_PCISLOT2;		/* PCI Interrupt #2 */
+	case 2: return IRQ_PCISLOT3;		/* PCI Interrupt #3 */
+	case 3: return IRQ_PCISLOT4;		/* PCI Interrupt E4 */
+	default:
+		printk("PCI: Bad IRQ mapping request for slot %d, func %d\n", slot, pin-1);
+		return -1;
+	}
+}
+
+static struct resource sh7780_io_resource = {
+	.name	= "SH7780_IO",
+	.start	= 0x2000,
+	.end	= 0x2000 + SH7780_PCI_IO_SIZE - 1,
+	.flags	= IORESOURCE_IO
+};
+
+static struct resource sh7780_mem_resource = {
+	.name	= "SH7780_mem",
+	.start	= SH7780_PCI_MEMORY_BASE,
+	.end	= SH7780_PCI_MEMORY_BASE + SH7780_PCI_MEM_SIZE - 1,
+	.flags	= IORESOURCE_MEM
+};
+
+extern struct pci_ops sh7780_pci_ops;
+
+struct pci_channel board_pci_channels[] = {
+	{ &sh7780_pci_ops, &sh7780_io_resource, &sh7780_mem_resource, 0, 0xff },
+	{ NULL, NULL, NULL, 0, 0 },
+};
+EXPORT_SYMBOL(board_pci_channels);
+
+static struct sh7780_pci_address_map sh7780_pci_map = {
+	.window0	= {
+		.base	= SH7780_CS2_BASE_ADDR,
+		.size	= 0x04000000,
+	},
+
+	.window1	= {
+		.base	= SH7780_CS3_BASE_ADDR,
+		.size	= 0x04000000,
+	},
+
+	.flags	= SH7780_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	return sh7780_pcic_init(&sh7780_pci_map);
+}
+
diff --git a/arch/sh/drivers/pci/pci-auto.c b/arch/sh/drivers/pci/pci-auto.c
index 4cef4d1d8c84..d55e46618549 100644
--- a/arch/sh/drivers/pci/pci-auto.c
+++ b/arch/sh/drivers/pci/pci-auto.c
@@ -45,11 +45,11 @@
 #include <linux/types.h>
 #include <linux/pci.h>
 
-#undef	DEBUG
-#ifdef 	DEBUG
+#define	DEBUG
+#ifdef	DEBUG
 #define	DBG(x...)	printk(x)
 #else
-#define	DBG(x...)	
+#define	DBG(x...)
 #endif
 
 /*
@@ -102,7 +102,7 @@ static u32 pciauto_upper_iospc;
 static u32 pciauto_lower_memspc;
 static u32 pciauto_upper_memspc;
 
-static void __init 
+static void __init
 pciauto_setup_bars(struct pci_channel *hose,
 		   int top_bus,
 		   int current_bus,
@@ -116,7 +116,6 @@ pciauto_setup_bars(struct pci_channel *hose,
 	int found_mem64 = 0;
 
 	for (bar = PCI_BASE_ADDRESS_0; bar <= bar_limit; bar+=4) {
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
 		u32 bar_addr;
 
 		/* Read the old BAR value */
@@ -125,7 +124,6 @@ pciauto_setup_bars(struct pci_channel *hose,
 					pci_devfn,
 					bar,
 					&bar_addr);
-#endif
 
 		/* Tickle the BAR and get the response */
 		early_write_config_dword(hose, top_bus,
@@ -140,8 +138,7 @@ pciauto_setup_bars(struct pci_channel *hose,
 					bar,
 					&bar_response);
 
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
-		/* 
+		/*
 		 * Write the old BAR value back out, only update the BAR
 		 * if we implicitly want resources to be updated, which
 		 * is done by the generic code further down. -- PFM.
@@ -151,7 +148,6 @@ pciauto_setup_bars(struct pci_channel *hose,
 					 pci_devfn,
 					 bar,
 					 bar_addr);
-#endif
 
 		/* If BAR is not implemented go to the next BAR */
 		if (!bar_response)
@@ -177,7 +173,7 @@ retry:
 			    PCI_BASE_ADDRESS_MEM_TYPE_64)
 				found_mem64 = 1;
 
-			addr_mask = PCI_BASE_ADDRESS_MEM_MASK;		
+			addr_mask = PCI_BASE_ADDRESS_MEM_MASK;
 			upper_limit = &pciauto_upper_memspc;
 			lower_limit = &pciauto_lower_memspc;
 			DBG("        Mem");
@@ -193,22 +189,22 @@ retry:
 		if ((bar_value + bar_size) > *upper_limit) {
 			if (bar_response & PCI_BASE_ADDRESS_SPACE) {
 				if (io_resource_inuse->child) {
-					io_resource_inuse = 
+					io_resource_inuse =
 						io_resource_inuse->child;
-					pciauto_lower_iospc = 
+					pciauto_lower_iospc =
 						io_resource_inuse->start;
-					pciauto_upper_iospc = 
+					pciauto_upper_iospc =
 						io_resource_inuse->end + 1;
 					goto retry;
 				}
 
 			} else {
 				if (mem_resource_inuse->child) {
-					mem_resource_inuse = 
+					mem_resource_inuse =
 						mem_resource_inuse->child;
-					pciauto_lower_memspc = 
+					pciauto_lower_memspc =
 						mem_resource_inuse->start;
-					pciauto_upper_memspc = 
+					pciauto_upper_memspc =
 						mem_resource_inuse->end + 1;
 					goto retry;
 				}
@@ -230,7 +226,7 @@ retry:
 		 * If we are a 64-bit decoder then increment to the
 		 * upper 32 bits of the bar and force it to locate
 		 * in the lower 4GB of memory.
-		 */ 
+		 */
 		if (found_mem64) {
 			bar += 4;
 			early_write_config_dword(hose, top_bus,
@@ -362,7 +358,7 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 {
 	u32 temp;
 
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
+#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D) && !defined(CONFIG_SH_R7780RP)
 	/*
 	 * [jsun] we always bump up baselines a little, so that if there
 	 * nothing behind P2P bridge, we don't wind up overlapping IO/MEM
@@ -396,7 +392,7 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 	 * configured by this routine to happily live behind a
 	 * P2P bridge in a system.
 	 */
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D)
+#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
 	pciauto_lower_memspc += 0x00400000;
 	pciauto_lower_iospc += 0x00004000;
 #endif
@@ -433,12 +429,12 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 	int devfn_stop = 0xff;
 
 	sub_bus = current_bus;
-	
+
 	if (hose->first_devfn)
 		devfn_start = hose->first_devfn;
 	if (hose->last_devfn)
 		devfn_stop = hose->last_devfn;
-	
+
 	for (pci_devfn=devfn_start; pci_devfn<devfn_stop; pci_devfn++) {
 
 		if (PCI_FUNC(pci_devfn) && !found_multi)
@@ -471,7 +467,7 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 		if ((pci_class >> 16) == PCI_CLASS_BRIDGE_PCI) {
 			DBG("        Bridge: primary=%.2x, secondary=%.2x\n",
 				current_bus, sub_bus + 1);
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D)
+#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
 			pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_1);
 #endif
 			pciauto_prescan_setup_bridge(hose, top_bus, current_bus,
@@ -490,10 +486,10 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 			DBG("PCI Autoconfig: Found CardBus bridge, device %d function %d\n", PCI_SLOT(pci_devfn), PCI_FUNC(pci_devfn));
 			/* Place CardBus Socket/ExCA registers */
 			pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_0);
- 
+
 			pciauto_prescan_setup_cardbus_bridge(hose, top_bus,
 					current_bus, pci_devfn, sub_bus);
- 
+
 			DBG("Scanning sub bus %.2x, I/O 0x%.8x, Mem 0x%.8x\n",
 				sub_bus + 1,
 				pciauto_lower_iospc, pciauto_lower_memspc);
diff --git a/arch/sh/drivers/pci/pci-sh7751.c b/arch/sh/drivers/pci/pci-sh7751.c
index 682f3dae305d..65093ec1b55e 100644
--- a/arch/sh/drivers/pci/pci-sh7751.c
+++ b/arch/sh/drivers/pci/pci-sh7751.c
@@ -223,7 +223,7 @@ static int __init __area_sdram_check(unsigned int area)
 
 	word = inl(SH7751_BCR1);
 	/* check BCR for SDRAM in area */
-	if(((word >> area) & 1) == 0) {
+	if (((word >> area) & 1) == 0) {
 		printk("PCI: Area %d is not configured for SDRAM. BCR1=0x%x\n",
 		       area, word);
 		return 0;
@@ -232,7 +232,7 @@ static int __init __area_sdram_check(unsigned int area)
 
 	word = (u16)inw(SH7751_BCR2);
 	/* check BCR2 for 32bit SDRAM interface*/
-	if(((word >> (area << 1)) & 0x3) != 0x3) {
+	if (((word >> (area << 1)) & 0x3) != 0x3) {
 		printk("PCI: Area %d is not 32 bit SDRAM. BCR2=0x%x\n",
 		       area, word);
 		return 0;
diff --git a/arch/sh/drivers/pci/pci-sh7780.c b/arch/sh/drivers/pci/pci-sh7780.c
new file mode 100644
index 000000000000..e09721330ac2
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh7780.c
@@ -0,0 +1,341 @@
+/*
+ *	Low-Level PCI Support for the SH7780
+ *
+ *  Dustin McIntire (dustin@sensoria.com)
+ *	Derived from arch/i386/kernel/pci-*.c which bore the message:
+ *	(c) 1999--2000 Martin Mares <mj@ucw.cz>
+ *
+ *  Ported to the new API by Paul Mundt <lethal@linux-sh.org>
+ *  With cleanup by Paul van Gool <pvangool@mimotech.com>
+ *
+ *  May be copied or modified under the terms of the GNU General Public
+ *  License.  See linux/COPYING for more information.
+ *
+ */
+
+#undef DEBUG
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/sched.h>
+#include <linux/ioport.h>
+#include <linux/errno.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+
+#include <asm/machvec.h>
+#include <asm/io.h>
+#include "pci-sh7780.h"
+
+static unsigned int pci_probe = PCI_PROBE_CONF1;
+extern int pci_fixup_pcic(void);
+
+/*
+ * Direct access to PCI hardware...
+ */
+
+#define CONFIG_CMD(bus, devfn, where) (0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
+
+/*
+ * Functions for accessing PCI configuration space with type 1 accesses
+ */
+static int sh7780_pci_read(struct pci_bus *bus, unsigned int devfn,
+			   int where, int size, u32 *val)
+{
+	unsigned long flags;
+	u32 data;
+
+	/*
+	 * PCIPDR may only be accessed as 32 bit words,
+	 * so we must do byte alignment by hand
+	 */
+	local_irq_save(flags);
+	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
+	data = inl(PCI_REG(SH7780_PCIPDR));
+	local_irq_restore(flags);
+
+	switch (size) {
+	case 1:
+		*val = (data >> ((where & 3) << 3)) & 0xff;
+		break;
+	case 2:
+		*val = (data >> ((where & 2) << 3)) & 0xffff;
+		break;
+	case 4:
+		*val = data;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+/*
+ * Since SH7780 only does 32bit access we'll have to do a read,
+ * mask,write operation.
+ * We'll allow an odd byte offset, though it should be illegal.
+ */
+static int sh7780_pci_write(struct pci_bus *bus, unsigned int devfn,
+			    int where, int size, u32 val)
+{
+	unsigned long flags;
+	int shift;
+	u32 data;
+
+	local_irq_save(flags);
+	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
+	data = inl(PCI_REG(SH7780_PCIPDR));
+	local_irq_restore(flags);
+
+	switch (size) {
+	case 1:
+		shift = (where & 3) << 3;
+		data &= ~(0xff << shift);
+		data |= ((val & 0xff) << shift);
+		break;
+	case 2:
+		shift = (where & 2) << 3;
+		data &= ~(0xffff << shift);
+		data |= ((val & 0xffff) << shift);
+		break;
+	case 4:
+		data = val;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	outl(data, PCI_REG(SH7780_PCIPDR));
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+#undef CONFIG_CMD
+
+struct pci_ops sh7780_pci_ops = {
+	.read		= sh7780_pci_read,
+	.write		= sh7780_pci_write,
+};
+
+static int __init pci_check_direct(void)
+{
+	unsigned int tmp, id;
+
+	outl(0x00000001, SH7780_PCI_VCR2); /* Enable PCIC */
+
+	/* check for SH7780/SH7780R hardware */
+	id = inl(PCI_REG(SH7780_PCIVID));
+	if ((id != ((SH7780_DEVICE_ID << 16) | SH7780_VENDOR_ID)) &&
+	    (id != ((SH7781_DEVICE_ID << 16) | SH7780_VENDOR_ID))) {
+		printk(KERN_ERR "PCI: This is not an SH7780 (%x)\n", id);
+		return -ENODEV;
+	}
+
+	/*
+	 * Check if configuration works.
+	 */
+	if (pci_probe & PCI_PROBE_CONF1) {
+		tmp = inl(PCI_REG(SH7780_PCIPAR));
+		outl(0x80000000, PCI_REG(SH7780_PCIPAR));
+		if (inl(PCI_REG(SH7780_PCIPAR)) == 0x80000000) {
+			outl(tmp, PCI_REG(SH7780_PCIPAR));
+			printk(KERN_INFO "PCI: Using configuration type 1\n");
+			request_region(PCI_REG(SH7780_PCIPAR), 8, "PCI conf1");
+			return 0;
+		}
+		outl(tmp, PCI_REG(SH7780_PCIPAR));
+	}
+
+	pr_debug("PCI: pci_check_direct failed\n");
+	return -EINVAL;
+}
+
+/***************************************************************************************/
+
+/*
+ *  Handle bus scanning and fixups ....
+ */
+
+static void __init pci_fixup_ide_bases(struct pci_dev *d)
+{
+	int i;
+
+	/*
+	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
+	 */
+	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return;
+	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
+	for(i=0; i<4; i++) {
+		struct resource *r = &d->resource[i];
+		if ((r->start & ~0x80) == 0x374) {
+			r->start |= 2;
+			r->end = r->start;
+		}
+	}
+}
+
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+
+/*
+ *  Called after each bus is probed, but before its children
+ *  are examined.
+ */
+
+void __init pcibios_fixup_bus(struct pci_bus *b)
+{
+	pci_read_bridge_bases(b);
+}
+
+/*
+ * Initialization. Try all known PCI access methods. Note that we support
+ * using both PCI BIOS and direct access: in such cases, we use I/O ports
+ * to access config space.
+ *
+ * Note that the platform specific initialization (BSC registers, and memory
+ * space mapping) will be called via the machine vectors (sh_mv.mv_pci_init()) if it
+ * exists and via the platform defined function pcibios_init_platform().
+ * See pci_bigsur.c for implementation;
+ *
+ * The BIOS version of the pci functions is not yet implemented but it is left
+ * in for completeness.  Currently an error will be genereated at compile time.
+ */
+
+static int __init sh7780_pci_init(void)
+{
+	int ret;
+
+	pr_debug("PCI: Starting intialization.\n");
+
+	/* Setup the INTC */
+	ctrl_outl(0x00200000, INTC_ICR0);	/* INTC SH-4 Mode */
+	ctrl_outl(0x00078000, INTC_INT2MSKCR);	/* enable PCIINTA - PCIINTD */
+	ctrl_outl(0x40000000, INTC_INTMSK1);	/* disable IRL4-7 Interrupt */
+	ctrl_outl(0x0000fffe, INTC_INTMSK2);	/* disable IRL4-7 Interrupt */
+	ctrl_outl(0x80000000, INTC_INTMSKCLR1);	/* enable IRL0-3 Interrupt */
+	ctrl_outl(0xfffe0000, INTC_INTMSKCLR2);	/* enable IRL0-3 Interrupt */
+
+	if ((ret = pci_check_direct()) != 0)
+		return ret;
+
+	return pcibios_init_platform();
+}
+
+core_initcall(sh7780_pci_init);
+
+int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
+{
+	u32 word;
+
+	/*
+	 * This code is unused for some boards as it is done in the
+	 * bootloader and doing it here means the MAC addresses loaded
+	 * by the bootloader get lost.
+	 */
+	if (!(map->flags & SH7780_PCIC_NO_RESET)) {
+		/* toggle PCI reset pin */
+		word = SH7780_PCICR_PREFIX | SH7780_PCICR_PRST;
+		outl(word,PCI_REG(SH7780_PCICR));
+		/* Wait for a long time... not 1 sec. but long enough */
+		mdelay(100);
+		word = SH7780_PCICR_PREFIX;
+		outl(word,PCI_REG(SH7780_PCICR));
+	}
+
+	/* set the command/status bits to:
+	 * Wait Cycle Control + Parity Enable + Bus Master +
+	 * Mem space enable
+	 */
+	outl(0x00000046, PCI_REG(SH7780_PCICMD));
+
+	/* define this host as the host bridge */
+	word = SH7780_PCI_HOST_BRIDGE << 24;
+	outl(word, PCI_REG(SH7780_PCIRID));
+
+	/* Set IO and Mem windows to local address
+	 * Make PCI and local address the same for easy 1 to 1 mapping
+	 * Window0 = map->window0.size @ non-cached area base = SDRAM
+	 * Window1 = map->window1.size @ cached area base = SDRAM
+	 */
+	word = ((map->window0.size - 1) & 0x1ff00001) | 0x01;
+	outl(0x07f00001, PCI_REG(SH7780_PCILSR0));
+	word = ((map->window1.size - 1) & 0x1ff00001) | 0x01;
+	outl(0x00000001, PCI_REG(SH7780_PCILSR1));
+	/* Set the values on window 0 PCI config registers */
+	word = P2SEGADDR(map->window0.base);
+	outl(0xa8000000, PCI_REG(SH7780_PCILAR0));
+	outl(0x08000000, PCI_REG(SH7780_PCIMBAR0));
+	/* Set the values on window 1 PCI config registers */
+	word = P2SEGADDR(map->window1.base);
+	outl(0x00000000, PCI_REG(SH7780_PCILAR1));
+	outl(0x00000000, PCI_REG(SH7780_PCIMBAR1));
+
+	/* Map IO space into PCI IO window
+	 * The IO window is 64K-PCIBIOS_MIN_IO in size
+	 * IO addresses will be translated to the
+	 * PCI IO window base address
+	 */
+	PCIDBG(3,"PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO,
+	    (64*1024), SH7780_PCI_IO_BASE+PCIBIOS_MIN_IO);
+
+	/* NOTE: I'm ignoring the PCI error IRQs for now..
+	 * TODO: add support for the internal error interrupts and
+	 * DMA interrupts...
+	 */
+
+#ifdef CONFIG_SH_R7780RP
+	pci_fixup_pcic();
+#endif
+
+	/* SH7780 init done, set central function init complete */
+	/* use round robin mode to stop a device starving/overruning */
+	word = SH7780_PCICR_PREFIX | SH7780_PCICR_CFIN | /* SH7780_PCICR_ARBM |*/ SH7780_PCICR_FTO;
+	outl(word, PCI_REG(SH7780_PCICR));
+
+	return 1;
+}
+
+char * __init pcibios_setup(char *str)
+{
+	if (!strcmp(str, "off")) {
+		pci_probe = 0;
+		return NULL;
+	}
+
+	return str;
+}
+
+/*
+ *	IRQ functions
+ */
+static u8 __init sh7780_no_swizzle(struct pci_dev *dev, u8 *pin)
+{
+	/* no swizzling */
+	return PCI_SLOT(dev->devfn);
+}
+
+static int sh7780_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq = -1;
+
+	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
+	irq = pcibios_map_platform_irq(slot,pin);
+	if( irq < 0 ) {
+		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
+		return irq;
+	}
+
+	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
+
+	return irq;
+}
+
+void __init pcibios_fixup_irqs(void)
+{
+	pci_fixup_irqs(sh7780_no_swizzle, sh7780_pci_lookup_irq);
+}
+
diff --git a/arch/sh/drivers/pci/pci-sh7780.h b/arch/sh/drivers/pci/pci-sh7780.h
new file mode 100644
index 000000000000..750d5d7753a2
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh7780.h
@@ -0,0 +1,168 @@
+/*
+ *	Low-Level PCI Support for SH7780 targets
+ *
+ *  Dustin McIntire (dustin@sensoria.com) (c) 2001
+ *  Paul Mundt (lethal@linux-sh.org) (c) 2003
+ *
+ *  May be copied or modified under the terms of the GNU General Public
+ *  License.  See linux/COPYING for more information.
+ *
+ */
+
+#ifndef _PCI_SH7780_H_
+#define _PCI_SH7780_H_
+
+#include <linux/pci.h>
+
+/* set debug level 4=verbose...1=terse */
+//#define DEBUG_PCI 3
+#undef DEBUG_PCI
+
+#ifdef DEBUG_PCI
+#define PCIDBG(n, x...) { if(DEBUG_PCI>=n) printk(x); }
+#else
+#define PCIDBG(n, x...)
+#endif
+
+/* startup values */
+#define PCI_PROBE_BIOS		1
+#define PCI_PROBE_CONF1		2
+#define PCI_PROBE_CONF2		4
+#define PCI_NO_SORT		0x100
+#define PCI_BIOS_SORT		0x200
+#define PCI_NO_CHECKS		0x400
+#define PCI_ASSIGN_ROMS		0x1000
+#define PCI_BIOS_IRQ_SCAN	0x2000
+
+/* Platform Specific Values */
+#define SH7780_VENDOR_ID	0x1912
+#define SH7780_DEVICE_ID	0x0002
+#define SH7781_DEVICE_ID	0x0001
+
+/* SH7780 Control Registers */
+#define	SH7780_PCI_VCR0		0xFE000000
+#define	SH7780_PCI_VCR1		0xFE000004
+#define	SH7780_PCI_VCR2		0xFE000008
+
+/* SH7780 Specific Values */
+#define SH7780_PCI_CONFIG_BASE	0xFD000000	/* Config space base addr */
+#define SH7780_PCI_CONFIG_SIZE	0x01000000	/* Config space size */
+#define SH7780_PCI_MEMORY_BASE	0xFD000000	/* Memory space base addr */
+#define SH7780_PCI_MEM_SIZE	0x01000000	/* Size of Memory window */
+#if 1
+#define SH7780_PCI_IO_BASE	0xFE400000	/* IO space base address */
+#define SH7780_PCI_IO_SIZE	0x00400000	/* Size of IO window */
+#else
+#define SH7780_PCI_IO_BASE	0xFE200000	/* IO space base address */
+#define SH7780_PCI_IO_SIZE	0x00200000	/* Size of IO window */
+#endif
+
+#define SH7780_PCIREG_BASE	0xFE040000	/* PCI regs base address */
+#define PCI_REG(n)		(SH7780_PCIREG_BASE+n)
+
+/* SH7780 PCI Config Registers */
+#define SH7780_PCIVID		0x000		/* Vendor ID */
+#define SH7780_PCIDID		0x002		/* Device ID */
+#define SH7780_PCICMD		0x004		/* Command */
+#define SH7780_PCISTATUS	0x006		/* Status */
+#define SH7780_PCIRID		0x008		/* Revision ID */
+#define SH7780_PCIPIF		0x009		/* Program Interface */
+#define SH7780_PCISUB		0x00a		/* Sub class code */
+#define SH7780_PCIBCC		0x00b		/* Base class code */
+#define SH7780_PCICLS		0x00c		/* Cache line size */
+#define SH7780_PCILTM		0x00d		/* latency timer */
+#define SH7780_PCIHDR		0x00e		/* Header type */
+#define SH7780_PCIBIST		0x00f		/* BIST */
+#define SH7780_PCIIBAR		0x010		/* IO Base address */
+#define SH7780_PCIMBAR0		0x014		/* Memory base address0 */
+#define SH7780_PCIMBAR1		0x018		/* Memory base address1 */
+#define SH7780_PCISVID		0x02c		/* Sub system vendor ID */
+#define SH7780_PCISID		0x02e		/* Sub system ID */
+#define SH7780_PCICP		0x034
+#define SH7780_PCIINTLINE	0x03c		/* Interrupt line */
+#define SH7780_PCIINTPIN	0x03d		/* Interrupt pin */
+#define SH7780_PCIMINGNT	0x03e		/* Minumum grand */
+#define SH7780_PCIMAXLAT	0x03f		/* Maxmum latency */
+#define SH7780_PCICID		0x040
+#define SH7780_PCINIP		0x041
+#define SH7780_PCIPMC		0x042
+#define SH7780_PCIPMCSR		0x044
+#define SH7780_PCIPMCSR_BSE	0x046
+#define SH7780_PCICDD		0x047
+
+/* SH7780 PCI Local Registers */
+#define SH7780_PCICR		0x100		/* PCI Control Register */
+  #define SH7780_PCICR_PREFIX	0xA5000000	/* CR prefix for write */
+  #define SH7780_PCICR_PFCS	0x00000800	/* TRDY/IRDY Enable */
+  #define SH7780_PCICR_FTO	0x00000400	/* TRDY/IRDY Enable */
+  #define SH7780_PCICR_PFE	0x00000200	/* Target Read Single */
+  #define SH7780_PCICR_TBS	0x00000100	/* Target Byte Swap */
+  #define SH7780_PCICR_ARBM	0x00000040	/* PCI Arbitration Mode */
+  #define SH7780_PCICR_IOCS	0x00000004	/* INTA output assert */
+  #define SH7780_PCICR_PRST	0x00000002	/* PCI Reset Assert */
+  #define SH7780_PCICR_CFIN	0x00000001	/* Central Fun. Init Done */
+#define SH7780_PCILSR0		0x104		/* PCI Local Space Register0 */
+#define SH7780_PCILSR1		0x108		/* PCI Local Space Register1 */
+#define SH7780_PCILAR0		0x10C	/* PCI Local Address Register1 */
+#define SH7780_PCILAR1		0x110	/* PCI Local Address Register1 */
+#define SH7780_PCIIR		0x114		/* PCI Interrupt Register */
+#define SH7780_PCIIMR		0x118	/* PCI Interrupt Mask Register */
+#define SH7780_PCIAIR		0x11C		/* Error Address Register */
+#define SH7780_PCICIR		0x120	/* Error Command/Data Register */
+#define SH7780_PCIAINT		0x130		/* Arbiter Interrupt Register */
+#define SH7780_PCIAINTM		0x134		/* Arbiter Int. Mask Register */
+#define SH7780_PCIBMIR		0x138		/* Error Bus Master Register */
+#define SH7780_PCIPAR		0x1C0		/* PIO Address Register */
+#define SH7780_PCIPINT		0x1CC	/* Power Management Int. Register */
+#define SH7780_PCIPINTM		0x1D0	/* Power Management Mask Register */
+#define SH7780_PCIMBR0		0x1E0		/* Memory Bank0 Register */
+#define SH7780_PCIMBMR0		0x1E4		/* Memory Bank0 Mask Register */
+#define SH7780_PCIMBR1		0x1E8		/* Memory Bank1  Register */
+#define SH7780_PCIMBMR1		0x1EC		/* Memory Bank1 Mask Register */
+#define SH7780_PCIMBR2		0x1F0		/* Memory Bank2 Register */
+#define SH7780_PCIMBMR2		0x1F4		/* Memory Bank2 Mask Register */
+#define SH7780_PCIIOBR		0x1F8		/* Bank Register */
+#define SH7780_PCIIOBMR		0x1FC		/* Bank Mask Register */
+#define SH7780_PCICSCR0		0x210		/* Cache Snoop1 Cnt. Register */
+#define SH7780_PCICSCR1		0x214		/* Cache Snoop2 Cnt. Register */
+#define SH7780_PCICSAR0		0x218	/* Cache Snoop1 Addr. Register */
+#define SH7780_PCICSAR1		0x21C	/* Cache Snoop2 Addr. Register */
+#define SH7780_PCIPDR		0x220		/* Port IO Data Register */
+
+/* General Memory Config Addresses */
+#define SH7780_CS0_BASE_ADDR	0x0
+#define SH7780_MEM_REGION_SIZE	0x04000000
+#define SH7780_CS1_BASE_ADDR	(SH7780_CS0_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS2_BASE_ADDR	(SH7780_CS1_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS3_BASE_ADDR	(SH7780_CS2_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS4_BASE_ADDR	(SH7780_CS3_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS5_BASE_ADDR	(SH7780_CS4_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS6_BASE_ADDR	(SH7780_CS5_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+
+/* General PCI values */
+#define SH7780_PCI_HOST_BRIDGE	0x6
+
+/* Flags */
+#define SH7780_PCIC_NO_RESET	0x0001
+
+/* External functions defined per platform i.e. Big Sur, SE... (these could be routed
+ * through the machine vectors... */
+extern int pcibios_init_platform(void);
+extern int pcibios_map_platform_irq(u8 slot, u8 pin);
+
+struct sh7780_pci_address_space {
+	unsigned long base;
+	unsigned long size;
+};
+
+struct sh7780_pci_address_map {
+	struct sh7780_pci_address_space window0;
+	struct sh7780_pci_address_space window1;
+	unsigned long flags;
+};
+
+/* arch/sh/drivers/pci/pci-sh7780.c */
+extern int sh7780_pcic_init(struct sh7780_pci_address_map *map);
+
+#endif /* _PCI_SH7780_H_ */
+
diff --git a/arch/sh/kernel/cpu/irq/intc2.c b/arch/sh/kernel/cpu/irq/intc2.c
index 30064bf6e154..e30e4b7aa70e 100644
--- a/arch/sh/kernel/cpu/irq/intc2.c
+++ b/arch/sh/kernel/cpu/irq/intc2.c
@@ -241,9 +241,9 @@ static struct intc2_init {
 	/* 110-111 reserved/unused */
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 	{ TIMER_IRQ, 0, 24, 0, INTC_TMU0_MSK, 2},
-#ifdef CONFIG_SH_RTC
-	{ RTC_IRQ, 4, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY },
-#endif
+	{ 21, 1, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY },
+	{ 22, 1, 1, 0, INTC_RTC_MSK, TIMER_PRIORITY },
+	{ 23, 1, 2, 0, INTC_RTC_MSK, TIMER_PRIORITY },
 	{ SCIF0_ERI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
 	{ SCIF0_RXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
 	{ SCIF0_BRI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 01ae394a6599..1f7f2c255ca2 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -26,4 +26,6 @@ EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
 LANDISK			SH_LANDISK
+R7780RP			SH_R7780RP
+R7780MP			SH_R7780MP
 TITAN			SH_TITAN
diff --git a/include/asm-sh/cpu-sh4/addrspace.h b/include/asm-sh/cpu-sh4/addrspace.h
index 727634d886ce..bb2e1b03060c 100644
--- a/include/asm-sh/cpu-sh4/addrspace.h
+++ b/include/asm-sh/cpu-sh4/addrspace.h
@@ -22,5 +22,8 @@
 #define P4SEG_TLB_DATA	0xf7000000
 #define P4SEG_REG_BASE	0xff000000
 
+#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
+#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
+
 #endif /* __ASM_CPU_SH4_ADDRSPACE_H */
 
diff --git a/include/asm-sh/cpu-sh4/dma-sh7780.h b/include/asm-sh/cpu-sh4/dma-sh7780.h
new file mode 100644
index 000000000000..6c90d28331b2
--- /dev/null
+++ b/include/asm-sh/cpu-sh4/dma-sh7780.h
@@ -0,0 +1,39 @@
+#ifndef __ASM_SH_CPU_SH4_DMA_SH7780_H
+#define __ASM_SH_CPU_SH4_DMA_SH7780_H
+
+#define REQ_HE	0x000000C0
+#define REQ_H	0x00000080
+#define REQ_LE	0x00000040
+#define TM_BURST 0x0000020
+#define TS_8	0x00000000
+#define TS_16	0x00000008
+#define TS_32	0x00000010
+#define TS_16BLK	0x00000018
+#define TS_32BLK	0x00100000
+
+/*
+ * The SuperH DMAC supports a number of transmit sizes, we list them here,
+ * with their respective values as they appear in the CHCR registers.
+ *
+ * Defaults to a 64-bit transfer size.
+ */
+enum {
+	XMIT_SZ_8BIT,
+	XMIT_SZ_16BIT,
+	XMIT_SZ_32BIT,
+	XMIT_SZ_128BIT,
+	XMIT_SZ_256BIT,
+};
+
+/*
+ * The DMA count is defined as the number of bytes to transfer.
+ */
+static unsigned int __attribute__ ((used)) ts_shift[] = {
+	[XMIT_SZ_8BIT]		= 0,
+	[XMIT_SZ_16BIT]		= 1,
+	[XMIT_SZ_32BIT]		= 2,
+	[XMIT_SZ_128BIT]	= 4,
+	[XMIT_SZ_256BIT]	= 5,
+};
+
+#endif /* __ASM_SH_CPU_SH4_DMA_SH7780_H */
diff --git a/include/asm-sh/cpu-sh4/dma.h b/include/asm-sh/cpu-sh4/dma.h
index 0dfe61f14802..3e4b3e6d80c0 100644
--- a/include/asm-sh/cpu-sh4/dma.h
+++ b/include/asm-sh/cpu-sh4/dma.h
@@ -1,11 +1,17 @@
 #ifndef __ASM_CPU_SH4_DMA_H
 #define __ASM_CPU_SH4_DMA_H
 
+#define DMAOR_INIT	( 0x8000 | DMAOR_DME )
+
 #ifdef CONFIG_CPU_SH4A
 #define SH_DMAC_BASE	0xfc808020
+
+#define CHCR_TS_MASK	0x18
+#define CHCR_TS_SHIFT	3
+
+#include <asm/cpu/dma-sh7780.h>
 #else
 #define SH_DMAC_BASE	0xffa00000
-#endif
 
 /* Definitions for the SuperH DMAC */
 #define TM_BURST	0x0000080
@@ -19,8 +25,6 @@
 
 #define DMAOR_COD	0x00000008
 
-#define DMAOR_INIT	( 0x8000 | DMAOR_DME )
-
 /*
  * The SuperH DMAC supports a number of transmit sizes, we list them here,
  * with their respective values as they appear in the CHCR registers.
@@ -45,5 +49,6 @@ static unsigned int ts_shift[] __attribute__ ((used)) = {
 	[XMIT_SZ_32BIT]		= 2,
 	[XMIT_SZ_256BIT]	= 5,
 };
+#endif
 
 #endif /* __ASM_CPU_SH4_DMA_H */
diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
index 69faf0171473..c4cff9d33927 100644
--- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h
+++ b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
@@ -19,8 +19,6 @@
 #define PA_OUTPORTR	0xa400000e	/* Output Port Reguster */
 #define PA_VERREG	0xa4000014	/* FPGA Version Register */
 
-#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
-#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
 #define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
 
 #define IRLCNTR1	(PA_BCR + 0)	/* Interrupt Control Register1 */
diff --git a/include/asm-sh/irq-sh7780.h b/include/asm-sh/irq-sh7780.h
index 7f90315cd830..895c5780e454 100644
--- a/include/asm-sh/irq-sh7780.h
+++ b/include/asm-sh/irq-sh7780.h
@@ -145,11 +145,6 @@
 #define	TMU_CH5_IPR_POS		1
 #define TMU_CH5_PRIORITY	2
 
-#define	RTC_IRQ		22
-#define	RTC_IPR_ADDR	INTC_INT2PRI1
-#define	RTC_IPR_POS	0
-#define	RTC_PRIORITY	TIMER_PRIORITY
-
 /* SCIF0 */
 #define SCIF0_ERI_IRQ	40
 #define SCIF0_RXI_IRQ	41
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 611e67cd0627..7e8455b1cb44 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -577,7 +577,7 @@ extern int ipr_irq_demux(int irq);
 #define NR_INTC2_IRQS	64
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 #define INTC2_BASE	0xffd40000
-#define INTC2_FIRST_IRQ	22
+#define INTC2_FIRST_IRQ	21
 #define INTC2_INTMSK_OFFSET	(0x38)
 #define INTC2_INTMSKCLR_OFFSET	(0x3c)
 #define NR_INTC2_IRQS	60
diff --git a/include/asm-sh/landisk/iodata_landisk.h b/include/asm-sh/landisk/iodata_landisk.h
index 9db3cdfe6776..c74d3c73f377 100644
--- a/include/asm-sh/landisk/iodata_landisk.h
+++ b/include/asm-sh/landisk/iodata_landisk.h
@@ -22,8 +22,6 @@
 /* 2003.10.31 I-O DATA NSD NWG	add.	for shutdown port clear */
 #define PA_PWRINT_CLR	0xb0000006	/* Shutdown Interrupt clear Register */
 
-#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
-#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
 #define PA_LCD_CLRDSP	0x00		/* LCD Clear Display Offset */
 #define PA_LCD_RTNHOME	0x00		/* LCD Return Home Offset */
 #define PA_LCD_ENTMODE	0x00		/* LCD Entry Mode Offset */
diff --git a/include/asm-sh/r7780rp/ide.h b/include/asm-sh/r7780rp/ide.h
new file mode 100644
index 000000000000..a1ed78e0f617
--- /dev/null
+++ b/include/asm-sh/r7780rp/ide.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_R7780RP_IDE_H
+#define __ASM_SH_R7780RP_IDE_H
+
+/* Nothing to see here.. */
+#include <asm/mach/r7780rp.h>
+
+#endif /* __ASM_SH_R7780RP_IDE_H */
+
diff --git a/include/asm-sh/r7780rp/r7780rp.h b/include/asm-sh/r7780rp/r7780rp.h
new file mode 100644
index 000000000000..f95d9dba31a2
--- /dev/null
+++ b/include/asm-sh/r7780rp/r7780rp.h
@@ -0,0 +1,177 @@
+#ifndef __ASM_SH_RENESAS_R7780RP_H
+#define __ASM_SH_RENESAS_R7780RP_H
+
+/*
+ * linux/include/asm-sh/r7780rp.h
+ *
+ * Copyright (C) 2000  Atom Create Engineering Co., Ltd.
+ *
+ * Renesas Solutions Highlander R7780RP support
+ */
+
+/* Box specific addresses.  */
+#if defined(CONFIG_SH_R7780MP)
+#define PA_BCR          0xa4000000      /* FPGA */
+#define PA_IRLMSK       (PA_BCR+0x0000) /* Interrupt Mask control */
+#define PA_IRLMON       (PA_BCR+0x0002) /* Interrupt Status control */
+#define PA_IRLPRI1      (PA_BCR+0x0004) /* Interrupt Priorty 1 */
+#define PA_IRLPRI2      (PA_BCR+0x0006) /* Interrupt Priorty 2 */
+#define PA_IRLPRI3      (PA_BCR+0x0008) /* Interrupt Priorty 3 */
+#define PA_IRLPRI4      (PA_BCR+0x000a) /* Interrupt Priorty 4 */
+#define PA_RSTCTL       (PA_BCR+0x000c) /* Reset Control */
+#define PA_PCIBD        (PA_BCR+0x000e) /* PCI Board detect control */
+#define PA_PCICD        (PA_BCR+0x0010) /* PCI Conector detect control */
+#define PA_EXTGIO       (PA_BCR+0x0016) /* Extension GPIO Control */
+#define PA_IVDRMON      (PA_BCR+0x0018) /* iVDR Moniter control */
+#define PA_IVDRCTL      (PA_BCR+0x001a) /* iVDR control */
+#define PA_OBLED        (PA_BCR+0x001c) /* On Board LED control */
+#define PA_OBSW         (PA_BCR+0x001e) /* On Board Switch control */
+#define PA_AUDIOSEL     (PA_BCR+0x0020) /* Sound Interface Select control */
+#define PA_EXTPLR       (PA_BCR+0x001e) /* Extention Pin Polarity control */
+#define PA_TPCTL        (PA_BCR+0x0100) /* Touch Panel Access control */
+#define PA_TPDCKCTL     (PA_BCR+0x0102) /* Touch Panel Access data control */
+#define PA_TPCTLCLR     (PA_BCR+0x0104) /* Touch Panel Access control */
+#define PA_TPXPOS       (PA_BCR+0x0106) /* Touch Panel X position control */
+#define PA_TPYPOS       (PA_BCR+0x0108) /* Touch Panel Y position control */
+#define PA_DBSW         (PA_BCR+0x0200) /* Debug Board Switch control */
+#define PA_CFCTL        (PA_BCR+0x0300) /* CF Timing control */
+#define PA_CFPOW        (PA_BCR+0x0302) /* CF Power control */
+#define PA_CFCDINTCLR   (PA_BCR+0x0304) /* CF Insert Interrupt clear */
+#define PA_SCSMR0       (PA_BCR+0x0400) /* SCIF0 Serial mode control */
+#define PA_SCBRR0       (PA_BCR+0x0404) /* SCIF0 Bit rate control */
+#define PA_SCSCR0       (PA_BCR+0x0408) /* SCIF0 Serial control */
+#define PA_SCFTDR0      (PA_BCR+0x040c) /* SCIF0 Send FIFO control */
+#define PA_SCFSR0       (PA_BCR+0x0410) /* SCIF0 Serial status control */
+#define PA_SCFRDR0      (PA_BCR+0x0414) /* SCIF0 Receive FIFO control */
+#define PA_SCFCR0       (PA_BCR+0x0418) /* SCIF0 FIFO control */
+#define PA_SCTFDR0      (PA_BCR+0x041c) /* SCIF0 Send FIFO data control */
+#define PA_SCRFDR0      (PA_BCR+0x0420) /* SCIF0 Receive FIFO data control */
+#define PA_SCSPTR0      (PA_BCR+0x0424) /* SCIF0 Serial Port control */
+#define PA_SCLSR0       (PA_BCR+0x0428) /* SCIF0 Line Status control */
+#define PA_SCRER0       (PA_BCR+0x042c) /* SCIF0 Serial Error control */
+#define PA_SCSMR1       (PA_BCR+0x0500) /* SCIF1 Serial mode control */
+#define PA_SCBRR1       (PA_BCR+0x0504) /* SCIF1 Bit rate control */
+#define PA_SCSCR1       (PA_BCR+0x0508) /* SCIF1 Serial control */
+#define PA_SCFTDR1      (PA_BCR+0x050c) /* SCIF1 Send FIFO control */
+#define PA_SCFSR1       (PA_BCR+0x0510) /* SCIF1 Serial status control */
+#define PA_SCFRDR1      (PA_BCR+0x0514) /* SCIF1 Receive FIFO control */
+#define PA_SCFCR1       (PA_BCR+0x0518) /* SCIF1 FIFO control */
+#define PA_SCTFDR1      (PA_BCR+0x051c) /* SCIF1 Send FIFO data control */
+#define PA_SCRFDR1      (PA_BCR+0x0520) /* SCIF1 Receive FIFO data control */
+#define PA_SCSPTR1      (PA_BCR+0x0524) /* SCIF1 Serial Port control */
+#define PA_SCLSR1       (PA_BCR+0x0528) /* SCIF1 Line Status control */
+#define PA_SCRER1       (PA_BCR+0x052c) /* SCIF1 Serial Error control */
+#define PA_ICCR         (PA_BCR+0x0600) /* Serial control */
+#define PA_SAR          (PA_BCR+0x0602) /* Serial Slave control */
+#define PA_MDR          (PA_BCR+0x0604) /* Serial Mode control */
+#define PA_ADR1         (PA_BCR+0x0606) /* Serial Address1 control */
+#define PA_DAR1         (PA_BCR+0x0646) /* Serial Data1 control */
+#define PA_VERREG       (PA_BCR+0x0700) /* FPGA Version Register */
+#define PA_POFF         (PA_BCR+0x0800) /* System Power Off control */
+#define PA_PMR          (PA_BCR+0x0900) /*  */
+
+#define PA_AX88796L     0xa4100400      /* AX88796L Area */
+#define PA_SC1602BSLB   0xa6000000      /* SC1602BSLB Area */
+#define PA_AREA5_IO     0xb4000000      /* Area 5 IO Memory */
+#define PA_AREA6_IO     0xb8000000      /* Area 6 IO Memory */
+#define PA_IDE_OFFSET   0x1f0           /* CF IDE Offset */
+#define AX88796L_IO_BASE        0x1000  /* AX88796L IO Base Address */
+
+#define IRLCNTR1        (PA_BCR + 0)    /* Interrupt Control Register1 */
+
+#define IRQ_PCISLOT1    65              /* PCI Slot #1 IRQ */
+#define IRQ_PCISLOT2    66              /* PCI Slot #2 IRQ */
+#define IRQ_PCISLOT3    67              /* PCI Slot #3 IRQ */
+#define IRQ_PCISLOT4    68              /* PCI Slot #4 IRQ */
+#define IRQ_CFCARD      1               /* CF Card IRQ */
+// #define IRQ_CFINST   0               /* CF Card Insert IRQ */
+#define IRQ_TP          2               /* Touch Panel IRQ */
+#define IRQ_SCI1        3               /* SCI1 IRQ */
+#define IRQ_SCI0        4               /* SCI0 IRQ */
+#define IRQ_2SERIAL     5               /* Serial IRQ */
+#define IRQ_RTC         6               /* RTC A / B IRQ */
+#define IRQ_EXTENTION6  7               /* EXT6n IRQ */
+#define IRQ_EXTENTION5  8               /* EXT5n IRQ */
+#define IRQ_EXTENTION4  9               /* EXT4n IRQ */
+#define IRQ_EXTENTION2  10              /* EXT2n IRQ */
+#define IRQ_EXTENTION1  11              /* EXT1n IRQ */
+#define IRQ_ONETH       13              /* On board Ethernet IRQ */
+#define IRQ_PSW         14              /* Push Switch IRQ */
+
+#else /* R7780RP */
+
+#define PA_BCR		0xa5000000	/* FPGA */
+#define	PA_IRLMSK	(PA_BCR+0x0000)	/* Interrupt Mask control */
+#define PA_IRLMON	(PA_BCR+0x0002)	/* Interrupt Status control */
+#define	PA_SDPOW	(PA_BCR+0x0004)	/* SD Power control */
+#define	PA_RSTCTL	(PA_BCR+0x0006)	/* Device Reset control */
+#define	PA_PCIBD	(PA_BCR+0x0008)	/* PCI Board detect control */
+#define	PA_PCICD	(PA_BCR+0x000a)	/* PCI Conector detect control */
+#define	PA_ZIGIO1	(PA_BCR+0x000c)	/* Zigbee IO control 1 */
+#define	PA_ZIGIO2	(PA_BCR+0x000e)	/* Zigbee IO control 2 */
+#define	PA_ZIGIO3	(PA_BCR+0x0010)	/* Zigbee IO control 3 */
+#define	PA_ZIGIO4	(PA_BCR+0x0012)	/* Zigbee IO control 4 */
+#define	PA_IVDRMON	(PA_BCR+0x0014)	/* iVDR Moniter control */
+#define	PA_IVDRCTL	(PA_BCR+0x0016)	/* iVDR control */
+#define PA_OBLED	(PA_BCR+0x0018)	/* On Board LED control */
+#define PA_OBSW		(PA_BCR+0x001a)	/* On Board Switch control */
+#define PA_AUDIOSEL	(PA_BCR+0x001c)	/* Sound Interface Select control */
+#define PA_EXTPLR	(PA_BCR+0x001e)	/* Extention Pin Polarity control */
+#define PA_TPCTL	(PA_BCR+0x0100)	/* Touch Panel Access control */
+#define PA_TPDCKCTL	(PA_BCR+0x0102)	/* Touch Panel Access data control */
+#define PA_TPCTLCLR	(PA_BCR+0x0104)	/* Touch Panel Access control */
+#define PA_TPXPOS	(PA_BCR+0x0106)	/* Touch Panel X position control */
+#define PA_TPYPOS	(PA_BCR+0x0108)	/* Touch Panel Y position control */
+#define PA_DBDET	(PA_BCR+0x0200)	/* Debug Board detect control */
+#define PA_DBDISPCTL	(PA_BCR+0x0202)	/* Debug Board Dot timing control */
+#define PA_DBSW		(PA_BCR+0x0204)	/* Debug Board Switch control */
+#define PA_CFCTL	(PA_BCR+0x0300)	/* CF Timing control */
+#define PA_CFPOW	(PA_BCR+0x0302)	/* CF Power control */
+#define PA_CFCDINTCLR	(PA_BCR+0x0304)	/* CF Insert Interrupt clear */
+#define PA_SCSMR	(PA_BCR+0x0400)	/* SCIF Serial mode control */
+#define PA_SCBRR	(PA_BCR+0x0402)	/* SCIF Bit rate control */
+#define PA_SCSCR	(PA_BCR+0x0404)	/* SCIF Serial control */
+#define PA_SCFDTR	(PA_BCR+0x0406)	/* SCIF Send FIFO control */
+#define PA_SCFSR	(PA_BCR+0x0408)	/* SCIF Serial status control */
+#define PA_SCFRDR	(PA_BCR+0x040a)	/* SCIF Receive FIFO control */
+#define PA_SCFCR	(PA_BCR+0x040c)	/* SCIF FIFO control */
+#define PA_SCFDR	(PA_BCR+0x040e)	/* SCIF FIFO data control */
+#define PA_SCLSR	(PA_BCR+0x0412)	/* SCIF Line Status control */
+#define PA_ICCR		(PA_BCR+0x0500)	/* Serial control */
+#define PA_SAR		(PA_BCR+0x0502)	/* Serial Slave control */
+#define PA_MDR		(PA_BCR+0x0504)	/* Serial Mode control */
+#define PA_ADR1		(PA_BCR+0x0506)	/* Serial Address1 control */
+#define PA_DAR1		(PA_BCR+0x0546)	/* Serial Data1 control */
+#define PA_VERREG	(PA_BCR+0x0600)	/* FPGA Version Register */
+
+#define PA_AX88796L	0xa5800400	/* AX88796L Area */
+#define PA_SC1602BSLB	0xa6000000	/* SC1602BSLB Area */
+#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
+#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
+#define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
+#define AX88796L_IO_BASE	0x1000	/* AX88796L IO Base Address */
+
+#define IRLCNTR1	(PA_BCR + 0)	/* Interrupt Control Register1 */
+
+#define IRQ_PCISLOT1	0		/* PCI Slot #1 IRQ */
+#define IRQ_PCISLOT2	1		/* PCI Slot #2 IRQ */
+#define IRQ_PCISLOT3	2		/* PCI Slot #3 IRQ */
+#define IRQ_PCISLOT4	3		/* PCI Slot #4 IRQ */
+#define IRQ_CFCARD	4		/* CF Card IRQ */
+#define IRQ_CFINST	5		/* CF Card Insert IRQ */
+#define IRQ_M66596	6		/* M66596 IRQ */
+#define IRQ_SDCARD	7		/* SD Card IRQ */
+#define IRQ_TUCHPANEL	8		/* Touch Panel IRQ */
+#define IRQ_SCI		9		/* SCI IRQ */
+#define IRQ_2SERIAL	10		/* Serial IRQ */
+#define	IRQ_EXTENTION	11		/* EXTn IRQ */
+#define IRQ_ONETH	12		/* On board Ethernet IRQ */
+#define IRQ_PSW		13		/* Push Switch IRQ */
+#define IRQ_ZIGBEE	14		/* Ziggbee IO IRQ */
+
+#endif  /* CONFIG_SH_R7780MP */
+
+#define __IO_PREFIX	r7780rp
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_RENESAS_R7780RP */
diff --git a/include/asm-sh/rts7751r2d/rts7751r2d.h b/include/asm-sh/rts7751r2d/rts7751r2d.h
index 4e09ba597e9a..b112ae221fd1 100644
--- a/include/asm-sh/rts7751r2d/rts7751r2d.h
+++ b/include/asm-sh/rts7751r2d/rts7751r2d.h
@@ -41,8 +41,6 @@
 
 #define PA_AX88796L	0xaa000400	/* AX88796L Area */
 #define PA_VOYAGER	0xab000000	/* VOYAGER GX Area */
-#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
-#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
 #define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
 #define AX88796L_IO_BASE	0x1000	/* AX88796L IO Base Address */
 
-- 
cgit v1.2.3


From f1517494407b1f1ca0063a756cc30d75e96d433c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:01:12 +0900
Subject: sh: Cleanup and document register bank usage.

Initial register bank cleanup. Make SR.RB configurable, and add some
preliminary documentation on register bank usage within the kernel.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 Documentation/sh/register-banks.txt | 33 +++++++++++++++++++++++++++++++++
 include/asm-sh/system.h             | 17 ++++++++++++++++-
 include/asm-sh/thread_info.h        | 17 +++++++++++++++--
 3 files changed, 64 insertions(+), 3 deletions(-)
 create mode 100644 Documentation/sh/register-banks.txt

(limited to 'include')

diff --git a/Documentation/sh/register-banks.txt b/Documentation/sh/register-banks.txt
new file mode 100644
index 000000000000..a6719f2f6594
--- /dev/null
+++ b/Documentation/sh/register-banks.txt
@@ -0,0 +1,33 @@
+	Notes on register bank usage in the kernel
+	==========================================
+
+Introduction
+------------
+
+The SH-3 and SH-4 CPU families traditionally include a single partial register
+bank (selected by SR.RB, only r0 ... r7 are banked), whereas other families
+may have more full-featured banking or simply no such capabilities at all.
+
+SR.RB banking
+-------------
+
+In the case of this type of banking, banked registers are mapped directly to
+r0 ... r7 if SR.RB is set to the bank we are interested in, otherwise ldc/stc
+can still be used to reference the banked registers (as r0_bank ... r7_bank)
+when in the context of another bank. The developer must keep the SR.RB value
+in mind when writing code that utilizes these banked registers, for obvious
+reasons. Userspace is also not able to poke at the bank1 values, so these can
+be used rather effectively as scratch registers by the kernel.
+
+Presently the kernel uses several of these registers.
+
+	- r0_bank, r1_bank (referenced as k0 and k1, used for scratch
+	  registers when doing exception handling).
+	- r2_bank (used to track the EXPEVT/INTEVT code)
+		- Used by do_IRQ() and friends for doing irq mapping based off
+		  of the interrupt exception vector jump table offset
+	- r6_bank (global interrupt mask)
+		- The SR.IMASK interrupt handler makes use of this to set the
+		  interrupt priority level (used by local_irq_enable())
+	- r7_bank (current)
+
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 1630a5411e5f..198d17e3069a 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -136,7 +136,8 @@ extern void __xchg_called_with_bad_pointer(void);
 #define set_mb(var, value) do { xchg(&var, value); } while (0)
 
 /* Interrupt Control */
-static __inline__ void local_irq_enable(void)
+#ifdef CONFIG_CPU_HAS_SR_RB
+static inline void local_irq_enable(void)
 {
 	unsigned long __dummy0, __dummy1;
 
@@ -149,6 +150,20 @@ static __inline__ void local_irq_enable(void)
 			     : "1" (~0x000000f0)
 			     : "memory");
 }
+#else
+static inline void local_irq_enable(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"and	%1, %0\n\t"
+		"ldc	%0, sr\n\t"
+		: "=&r" (__dummy0), "=r" (__dummy1)
+		: "1" (~0x000000f0)
+		: "memory");
+}
+#endif
 
 static __inline__ void local_irq_disable(void)
 {
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 7345350d98c0..f64dd803a014 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -48,16 +48,29 @@ struct thread_info {
 #define init_thread_info	(init_thread_union.thread_info)
 #define init_stack		(init_thread_union.stack)
 
+#define THREAD_SIZE (2*PAGE_SIZE)
+
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
 	struct thread_info *ti;
+#ifdef CONFIG_CPU_HAS_SR_RB
 	__asm__("stc	r7_bank, %0" : "=r" (ti));
+#else
+	unsigned long __dummy;
+
+	__asm__ __volatile__ (
+		"mov	r15, %0\n\t"
+		"and	%1, %0\n\t"
+		: "=&r" (ti), "=r" (__dummy)
+		: "1" (~(THREAD_SIZE - 1))
+		: "memory");
+#endif
+
 	return ti;
 }
 
 /* thread information allocation */
-#define THREAD_SIZE (2*PAGE_SIZE)
 #define alloc_thread_info(ti) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
 #define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
 
@@ -65,7 +78,7 @@ static inline struct thread_info *current_thread_info(void)
 
 /* how to get the thread information struct from ASM */
 #define GET_THREAD_INFO(reg) \
-	stc	r7_bank, reg
+	stc     r7_bank, reg
 
 #endif
 
-- 
cgit v1.2.3


From 00b3aa3fc9bd827caaa859de90d9eba831b77d40 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:05:56 +0900
Subject: sh: xchg()/__xchg() always_inline fixes for gcc4.

Make __xchg() a macro, so that gcc 4.0 doesn't blow up thanks to
always_inline..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h | 57 +++++++++++++++++++++++++++----------------------
 1 file changed, 32 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 198d17e3069a..bd7dc0554b11 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -79,10 +79,8 @@ static inline void sched_cacheflush(void)
 }
 #endif
 
-#define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
-
 static __inline__ unsigned long tas(volatile int *m)
-{ /* #define tas(ptr) (xchg((ptr),1)) */
+{
 	unsigned long retval;
 
 	__asm__ __volatile__ ("tas.b	@%1\n\t"
@@ -91,8 +89,6 @@ static __inline__ unsigned long tas(volatile int *m)
 	return retval;
 }
 
-extern void __xchg_called_with_bad_pointer(void);
-
 /*
  * A brief note on ctrl_barrier(), the control register write barrier.
  *
@@ -220,17 +216,17 @@ static __inline__ void  local_irq_restore(unsigned long x)
 	}
 }
 #else
-#define local_irq_restore(x) do { 			\
+#define local_irq_restore(x) do {			\
 	if ((x & 0x000000f0) != 0x000000f0)		\
-		local_irq_enable();				\
+		local_irq_enable();			\
 } while (0)
 #endif
 
-#define really_restore_flags(x) do { 			\
+#define really_restore_flags(x) do {			\
 	if ((x & 0x000000f0) != 0x000000f0)		\
-		local_irq_enable();				\
+		local_irq_enable();			\
 	else						\
-		local_irq_disable();				\
+		local_irq_disable();			\
 } while (0)
 
 /*
@@ -272,7 +268,7 @@ do {							\
 /* For spinlocks etc */
 #define local_irq_save(x)	x = local_irq_save()
 
-static __inline__ unsigned long xchg_u32(volatile int * m, unsigned long val)
+static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
 {
 	unsigned long flags, retval;
 
@@ -283,7 +279,7 @@ static __inline__ unsigned long xchg_u32(volatile int * m, unsigned long val)
 	return retval;
 }
 
-static __inline__ unsigned long xchg_u8(volatile unsigned char * m, unsigned long val)
+static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
 {
 	unsigned long flags, retval;
 
@@ -294,19 +290,30 @@ static __inline__ unsigned long xchg_u8(volatile unsigned char * m, unsigned lon
 	return retval;
 }
 
-static __inline__ unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
-{
-	switch (size) {
-	case 4:
-		return xchg_u32(ptr, x);
-		break;
-	case 1:
-		return xchg_u8(ptr, x);
-		break;
-	}
-	__xchg_called_with_bad_pointer();
-	return x;
-}
+extern void __xchg_called_with_bad_pointer(void);
+
+#define __xchg(ptr, x, size)				\
+({							\
+	unsigned long __xchg__res;			\
+	volatile void *__xchg_ptr = (ptr);		\
+	switch (size) {					\
+	case 4:						\
+		__xchg__res = xchg_u32(__xchg_ptr, x);	\
+		break;					\
+	case 1:						\
+		__xchg__res = xchg_u8(__xchg_ptr, x);	\
+		break;					\
+	default:					\
+		__xchg_called_with_bad_pointer();	\
+		__xchg__res = x;			\
+		break;					\
+	}						\
+							\
+	__xchg__res;					\
+})
+
+#define xchg(ptr,x)	\
+	((__typeof__(*(ptr)))__xchg((ptr),(unsigned long)(x), sizeof(*(ptr))))
 
 static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
 	unsigned long new)
-- 
cgit v1.2.3


From ef48e8e3498605351f91f195cc9af0ef981b0dde Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:17:17 +0900
Subject: sh: Free up some and document PTEL flags.

Drop _PAGE_SHARED/_PAGE_U0_SHARED and document Linux PTE encodings in
the PTEL value. Preserve the swap cache entry encoding semantics for
now, though it will need rework to free up _PAGE_WT from _PAGE_FILE.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pgtable.h | 76 ++++++++++++++++++++++++++++--------------------
 1 file changed, 45 insertions(+), 31 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 9728b58f7c13..41c559d8ba87 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -54,20 +54,41 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define VMALLOC_START	(P3SEG+0x00100000)
 #define VMALLOC_END	(FIXADDR_START-2*PAGE_SIZE)
 
-#define	_PAGE_WT	0x001  /* WT-bit on SH-4, 0 on SH-3 */
-#define _PAGE_HW_SHARED	0x002  /* SH-bit  : page is shared among processes */
-#define _PAGE_DIRTY	0x004  /* D-bit   : page changed */
-#define _PAGE_CACHABLE	0x008  /* C-bit   : cachable */
-#define _PAGE_SZ0	0x010  /* SZ0-bit : Size of page */
-#define _PAGE_RW	0x020  /* PR0-bit : write access allowed */
-#define _PAGE_USER	0x040  /* PR1-bit : user space access allowed */
-#define _PAGE_SZ1	0x080  /* SZ1-bit : Size of page (on SH-4) */
-#define _PAGE_PRESENT	0x100  /* V-bit   : page is valid */
-#define _PAGE_PROTNONE	0x200  /* software: if not present  */
-#define _PAGE_ACCESSED	0x400  /* software: page referenced */
-#define _PAGE_U0_SHARED 0x800  /* software: page is shared in user space */
-
-#define	_PAGE_FILE	_PAGE_WT  /* software: pagecache or swap? */
+/*
+ * Linux PTEL encoding.
+ *
+ * Hardware and software bit definitions for the PTEL value:
+ *
+ * - Bits 0 and 7 are reserved on SH-3 (_PAGE_WT and _PAGE_SZ1 on SH-4).
+ *
+ * - Bit 1 is the SH-bit, but is unused on SH-3 due to an MMU bug (the
+ *   hardware PTEL value can't have the SH-bit set when MMUCR.IX is set,
+ *   which is the default in cpu-sh3/mmu_context.h:MMU_CONTROL_INIT).
+ *
+ *   In order to keep this relatively clean, do not use these for defining
+ *   SH-3 specific flags until all of the other unused bits have been
+ *   exhausted.
+ *
+ * - Bit 9 is reserved by everyone and used by _PAGE_PROTNONE.
+ *
+ * - Bits 10 and 11 are low bits of the PPN that are reserved on >= 4K pages.
+ *   Bit 10 is used for _PAGE_ACCESSED, bit 11 remains unused.
+ *
+ * - Bits 31, 30, and 29 remain unused by everyone and can be used for future
+ *   software flags, although care must be taken to update _PAGE_CLEAR_FLAGS.
+ */
+#define	_PAGE_WT	0x001		/* WT-bit on SH-4, 0 on SH-3 */
+#define _PAGE_HW_SHARED	0x002		/* SH-bit  : shared among processes */
+#define _PAGE_DIRTY	0x004		/* D-bit   : page changed */
+#define _PAGE_CACHABLE	0x008		/* C-bit   : cachable */
+#define _PAGE_SZ0	0x010		/* SZ0-bit : Size of page */
+#define _PAGE_RW	0x020		/* PR0-bit : write access allowed */
+#define _PAGE_USER	0x040		/* PR1-bit : user space access allowed */
+#define _PAGE_SZ1	0x080		/* SZ1-bit : Size of page (on SH-4) */
+#define _PAGE_PRESENT	0x100		/* V-bit   : page is valid */
+#define _PAGE_PROTNONE	0x200		/* software: if not present  */
+#define _PAGE_ACCESSED	0x400		/* software: page referenced */
+#define _PAGE_FILE	_PAGE_WT	/* software: pagecache or swap? */
 
 /* software: moves to PTEA.TC (Timing Control) */
 #define _PAGE_PCC_AREA5	0x00000000	/* use BSC registers for area5 */
@@ -82,22 +103,17 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define _PAGE_PCC_ATR8	0x60000000	/* Attribute Memory space, 8 bit bus */
 #define _PAGE_PCC_ATR16	0x60000001	/* Attribute Memory space, 6 bit bus */
 
-/* Mask which drop software flags
- * We also drop WT bit since it is used for _PAGE_FILE
- * bit in this implementation.
- */
-#define _PAGE_CLEAR_FLAGS	(_PAGE_WT | _PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_U0_SHARED)
-
-#if defined(CONFIG_CPU_SH3)
-/*
- * MMU on SH-3 has bug on SH-bit: We can't use it if MMUCR.IX=1.
- * Work around: Just drop SH-bit.
- */
-#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS | _PAGE_HW_SHARED))
+/* Mask which drops unused bits from the PTEL value */
+#ifdef CONFIG_CPU_SH3
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED| \
+				 _PAGE_FILE	| _PAGE_SZ1	| \
+				 _PAGE_HW_SHARED)
 #else
-#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_FILE)
 #endif
 
+#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
+
 /* Hardware flags: SZ0=1 (4k-byte) */
 #define _PAGE_FLAGS_HARD	_PAGE_SZ0
 
@@ -107,17 +123,15 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define _PAGE_SZHUGE	(_PAGE_SZ0 | _PAGE_SZ1)
 #endif
 
-#define _PAGE_SHARED	_PAGE_U0_SHARED
-
 #define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY | _PAGE_SHARED)
+#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY)
 
 #ifndef __ASSEMBLY__
 
 #ifdef CONFIG_MMU
 #define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_FLAGS_HARD)
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_SHARED | _PAGE_FLAGS_HARD)
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_COPY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_CACHABLE | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_HW_SHARED | _PAGE_FLAGS_HARD)
-- 
cgit v1.2.3


From 3aa770e7972723f479122cf66b529017d2175289 Mon Sep 17 00:00:00 2001
From: Andriy Skulysh <askulysh@gmail.com>
Date: Wed, 27 Sep 2006 16:20:22 +0900
Subject: sh: APM/PM support.

This adds some simple PM stubs and the basic APM interfaces,
primarily for use by hp6xx, where the existing userland
expects it.

Signed-off-by: Andriy Skulysh <askulysh@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                            |  10 +
 arch/sh/boards/hp6xx/Makefile              |   5 +-
 arch/sh/boards/hp6xx/hp6xx_apm.c           | 123 +++++++
 arch/sh/boards/hp6xx/pm.c                  |  88 +++++
 arch/sh/boards/hp6xx/pm_wakeup.S           |  58 ++++
 arch/sh/boards/hp6xx/setup.c               |  14 +
 arch/sh/kernel/Makefile                    |   2 +
 arch/sh/kernel/apm.c                       | 539 +++++++++++++++++++++++++++++
 arch/sh/kernel/entry.S                     |   4 +-
 arch/sh/kernel/pm.c                        |  88 +++++
 arch/sh/kernel/sh_ksyms.c                  |   4 +
 arch/sh/kernel/time.c                      |  25 ++
 arch/sh/kernel/timers/timer-tmu.c          |  18 +-
 drivers/input/touchscreen/hp680_ts_input.c |  14 -
 include/asm-sh/apm.h                       |  46 +++
 include/asm-sh/cpu-sh3/freq.h              |   4 +
 include/asm-sh/hd64461.h                   |   7 +-
 include/asm-sh/hp6xx/hp6xx.h               |  53 ++-
 include/asm-sh/pm.h                        |  17 +
 include/asm-sh/system.h                    |  25 ++
 include/asm-sh/timer.h                     |   2 +
 21 files changed, 1124 insertions(+), 22 deletions(-)
 create mode 100644 arch/sh/boards/hp6xx/hp6xx_apm.c
 create mode 100644 arch/sh/boards/hp6xx/pm.c
 create mode 100644 arch/sh/boards/hp6xx/pm_wakeup.S
 create mode 100644 arch/sh/kernel/apm.c
 create mode 100644 arch/sh/kernel/pm.c
 create mode 100644 include/asm-sh/apm.h
 create mode 100644 include/asm-sh/pm.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index fe982f0eba23..d1d724447826 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -638,6 +638,16 @@ source "fs/Kconfig.binfmt"
 
 endmenu
 
+menu "Power management options (EXPERIMENTAL)"
+depends on EXPERIMENTAL
+
+source kernel/power/Kconfig
+
+config APM
+	bool "Advanced Power Management Emulation"
+	depends on PM
+endmenu
+
 source "net/Kconfig"
 
 source "drivers/Kconfig"
diff --git a/arch/sh/boards/hp6xx/Makefile b/arch/sh/boards/hp6xx/Makefile
index 927fe0aa5dfa..fdf98ffcf437 100644
--- a/arch/sh/boards/hp6xx/Makefile
+++ b/arch/sh/boards/hp6xx/Makefile
@@ -2,5 +2,8 @@
 # Makefile for the HP6xx specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o
+obj-y	 		:= mach.o setup.o
+obj-$(CONFIG_PM)	+= pm.o pm_wakeup.o
+obj-$(CONFIG_APM)	+= hp6xx_apm.o
+
 
diff --git a/arch/sh/boards/hp6xx/hp6xx_apm.c b/arch/sh/boards/hp6xx/hp6xx_apm.c
new file mode 100644
index 000000000000..ad0e712c29f6
--- /dev/null
+++ b/arch/sh/boards/hp6xx/hp6xx_apm.c
@@ -0,0 +1,123 @@
+/*
+ * bios-less APM driver for hp680
+ *
+ * Copyright 2005 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/apm_bios.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <asm/io.h>
+#include <asm/apm.h>
+#include <asm/adc.h>
+#include <asm/hp6xx/hp6xx.h>
+
+#define SH7709_PGDR			0xa400012c
+
+#define APM_CRITICAL			10
+#define APM_LOW				30
+
+#define HP680_BATTERY_MAX		875
+#define HP680_BATTERY_MIN		600
+#define HP680_BATTERY_AC_ON		900
+
+#define MODNAME "hp6x0_apm"
+
+static int hp6x0_apm_get_info(char *buf, char **start, off_t fpos, int length)
+{
+	u8 pgdr;
+	char *p;
+	int battery_status;
+	int battery_flag;
+	int ac_line_status;
+	int time_units = APM_BATTERY_LIFE_UNKNOWN;
+
+	int battery = adc_single(ADC_CHANNEL_BATTERY);
+	int backup = adc_single(ADC_CHANNEL_BACKUP);
+	int charging = adc_single(ADC_CHANNEL_CHARGE);
+	int percentage;
+
+	percentage = 100 * (battery - HP680_BATTERY_MIN) /
+			   (HP680_BATTERY_MAX - HP680_BATTERY_MIN);
+
+	ac_line_status = (battery > HP680_BATTERY_AC_ON) ?
+			 APM_AC_ONLINE : APM_AC_OFFLINE;
+
+	p = buf;
+
+	pgdr = ctrl_inb(SH7709_PGDR);
+	if (pgdr & PGDR_MAIN_BATTERY_OUT) {
+		battery_status = APM_BATTERY_STATUS_NOT_PRESENT;
+		battery_flag = 0x80;
+		percentage = -1;
+	} else if (charging < 8 ) {
+		battery_status = APM_BATTERY_STATUS_CHARGING;
+		battery_flag = 0x08;
+		ac_line_status = 0xff;
+	} else if (percentage <= APM_CRITICAL) {
+		battery_status = APM_BATTERY_STATUS_CRITICAL;
+		battery_flag = 0x04;
+	} else if (percentage <= APM_LOW) {
+		battery_status = APM_BATTERY_STATUS_LOW;
+		battery_flag = 0x02;
+	} else {
+		battery_status = APM_BATTERY_STATUS_HIGH;
+		battery_flag = 0x01;
+	}
+
+	p += sprintf(p, "1.0 1.2 0x%02x 0x%02x 0x%02x 0x%02x %d%% %d %s\n",
+		     APM_32_BIT_SUPPORT,
+		     ac_line_status,
+		     battery_status,
+		     battery_flag,
+		     percentage,
+		     time_units,
+		     "min");
+	p += sprintf(p, "bat=%d backup=%d charge=%d\n",
+		     battery, backup, charging);
+
+	return p - buf;
+}
+
+static irqreturn_t hp6x0_apm_interrupt(int irq, void *dev, struct pt_regs *regs)
+{
+	if (!apm_suspended)
+		apm_queue_event(APM_USER_SUSPEND);
+
+	return IRQ_HANDLED;
+}
+
+static int __init hp6x0_apm_init(void)
+{
+	int ret;
+
+	ret = request_irq(HP680_BTN_IRQ, hp6x0_apm_interrupt,
+			  SA_INTERRUPT, MODNAME, 0);
+	if (unlikely(ret < 0)) {
+		printk(KERN_ERR MODNAME ": IRQ %d request failed\n",
+		       HP680_BTN_IRQ);
+		return ret;
+	}
+
+	apm_get_info = hp6x0_apm_get_info;
+
+	return ret;
+}
+
+static void __exit hp6x0_apm_exit(void)
+{
+	free_irq(HP680_BTN_IRQ, 0);
+	apm_get_info = 0;
+}
+
+module_init(hp6x0_apm_init);
+module_exit(hp6x0_apm_exit);
+
+MODULE_AUTHOR("Adriy Skulysh");
+MODULE_DESCRIPTION("hp6xx Advanced Power Management");
+MODULE_LICENSE("GPL");
diff --git a/arch/sh/boards/hp6xx/pm.c b/arch/sh/boards/hp6xx/pm.c
new file mode 100644
index 000000000000..0e501bcbd7a9
--- /dev/null
+++ b/arch/sh/boards/hp6xx/pm.c
@@ -0,0 +1,88 @@
+/*
+ * hp6x0 Power Management Routines
+ *
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/suspend.h>
+#include <linux/errno.h>
+#include <linux/time.h>
+#include <asm/io.h>
+#include <asm/hd64461.h>
+#include <asm/hp6xx/hp6xx.h>
+#include <asm/cpu/dac.h>
+#include <asm/pm.h>
+
+#define STBCR		0xffffff82
+#define STBCR2		0xffffff88
+
+static int hp6x0_pm_enter(suspend_state_t state)
+{
+	u8 stbcr, stbcr2;
+#ifdef CONFIG_HD64461_ENABLER
+	u8 scr;
+	u16 hd64461_stbcr;
+#endif
+
+	if (state != PM_SUSPEND_MEM)
+		return -EINVAL;
+
+#ifdef CONFIG_HD64461_ENABLER
+	outb(0, HD64461_PCC1CSCIER);
+
+	scr = inb(HD64461_PCC1SCR);
+	scr |= HD64461_PCCSCR_VCC1;
+	outb(scr, HD64461_PCC1SCR);
+
+	hd64461_stbcr = inw(HD64461_STBCR);
+	hd64461_stbcr |= HD64461_STBCR_SPC1ST;
+	outw(hd64461_stbcr, HD64461_STBCR);
+#endif
+
+	ctrl_outb(0x1f, DACR);
+
+	stbcr = ctrl_inb(STBCR);
+	ctrl_outb(0x01, STBCR);
+
+	stbcr2 = ctrl_inb(STBCR2);
+	ctrl_outb(0x7f , STBCR2);
+
+	outw(0xf07f, HD64461_SCPUCR);
+
+	pm_enter();
+
+	outw(0, HD64461_SCPUCR);
+	ctrl_outb(stbcr, STBCR);
+	ctrl_outb(stbcr2, STBCR2);
+
+#ifdef CONFIG_HD64461_ENABLER
+	hd64461_stbcr = inw(HD64461_STBCR);
+	hd64461_stbcr &= ~HD64461_STBCR_SPC1ST;
+	outw(hd64461_stbcr, HD64461_STBCR);
+
+	outb(0x4c, HD64461_PCC1CSCIER);
+	outb(0x00, HD64461_PCC1CSCR);
+#endif
+
+	return 0;
+}
+
+/*
+ * Set to PM_DISK_FIRMWARE so we can quickly veto suspend-to-disk.
+ */
+static struct pm_ops hp6x0_pm_ops = {
+	.pm_disk_mode	= PM_DISK_FIRMWARE,
+	.enter		= hp6x0_pm_enter,
+};
+
+static int __init hp6x0_pm_init(void)
+{
+	pm_set_ops(&hp6x0_pm_ops);
+	return 0;
+}
+
+late_initcall(hp6x0_pm_init);
diff --git a/arch/sh/boards/hp6xx/pm_wakeup.S b/arch/sh/boards/hp6xx/pm_wakeup.S
new file mode 100644
index 000000000000..45e9bf0b9115
--- /dev/null
+++ b/arch/sh/boards/hp6xx/pm_wakeup.S
@@ -0,0 +1,58 @@
+/*
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/cpu/mmu_context.h>
+
+#define k0	r0
+#define k1	r1
+#define k2	r2
+#define k3	r3
+#define k4	r4
+
+/*
+ * Kernel mode register usage:
+ *	k0	scratch
+ *	k1	scratch
+ *	k2	scratch (Exception code)
+ *	k3	scratch (Return address)
+ *	k4	scratch
+ *	k5	reserved
+ *	k6	Global Interrupt Mask (0--15 << 4)
+ *	k7	CURRENT_THREAD_INFO (pointer to current thread info)
+ */
+
+ENTRY(wakeup_start)
+! clear STBY bit
+	mov	#-126, k2
+   	and	#127, k0
+	mov.b	k0, @k2
+! enable refresh
+	mov.l	5f, k1
+	mov.w	6f, k0
+  	mov.w	k0, @k1
+! jump to handler
+	mov.l	2f, k2
+	mov.l	3f, k3
+	mov.l	@k2, k2
+
+	mov.l	4f, k1
+	jmp	@k1
+	nop
+
+	.align	2
+1:	.long	EXPEVT
+2:	.long	INTEVT
+3:	.long	ret_from_irq
+4:	.long	handle_exception
+5:	.long	0xffffff68
+6:	.word	0x0524
+
+ENTRY(wakeup_end)
+	nop
diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c
index 71f315663cc9..5fc00f0727c4 100644
--- a/arch/sh/boards/hp6xx/setup.c
+++ b/arch/sh/boards/hp6xx/setup.c
@@ -15,6 +15,9 @@
 #include <asm/hp6xx/hp6xx.h>
 #include <asm/cpu/dac.h>
 
+#define	SCPCR	0xa4000116
+#define SCPDR	0xa4000136
+
 const char *get_system_type(void)
 {
 	return "HP6xx";
@@ -24,6 +27,7 @@ int __init platform_setup(void)
 {
 	u8 v8;
 	u16 v;
+
 	v = inw(HD64461_STBCR);
 	v |= HD64461_STBCR_SURTST | HD64461_STBCR_SIRST |
 	    HD64461_STBCR_STM1ST | HD64461_STBCR_STM0ST |
@@ -50,5 +54,15 @@ int __init platform_setup(void)
 	v8 &= ~DACR_DAE;
 	ctrl_outb(v8,DACR);
 
+	v8 = ctrl_inb(SCPDR);
+	v8 |= SCPDR_TS_SCAN_X | SCPDR_TS_SCAN_Y;
+	v8 &= ~SCPDR_TS_SCAN_ENABLE;
+	ctrl_outb(v8, SCPDR);
+
+	v = ctrl_inw(SCPCR);
+	v &= ~SCPCR_TS_MASK;
+	v |= SCPCR_TS_ENABLE;
+	ctrl_outw(v, SCPCR);
+
 	return 0;
 }
diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index 2de82b66af4a..0e2148f63e7e 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -18,3 +18,5 @@ obj-$(CONFIG_SH_CPU_FREQ)	+= cpufreq.o
 obj-$(CONFIG_MODULES)		+= module.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
+obj-$(CONFIG_APM)		+= apm.o
+obj-$(CONFIG_PM)		+= pm.o
diff --git a/arch/sh/kernel/apm.c b/arch/sh/kernel/apm.c
new file mode 100644
index 000000000000..871e7d640002
--- /dev/null
+++ b/arch/sh/kernel/apm.c
@@ -0,0 +1,539 @@
+/*
+ * bios-less APM driver for hp680
+ *
+ * Copyright 2005 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * based on ARM APM driver by
+ *  Jamey Hicks <jamey@crl.dec.com>
+ *
+ * adapted from the APM BIOS driver for Linux by
+ *  Stephen Rothwell (sfr@linuxcare.com)
+ *
+ * APM 1.2 Reference:
+ *   Intel Corporation, Microsoft Corporation. Advanced Power Management
+ *   (APM) BIOS Interface Specification, Revision 1.2, February 1996.
+ *
+ * [This document is available from Microsoft at:
+ *    http://www.microsoft.com/hwdev/busbios/amp_12.htm]
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/proc_fs.h>
+#include <linux/miscdevice.h>
+#include <linux/apm_bios.h>
+#include <linux/pm.h>
+#include <linux/pm_legacy.h>
+#include <asm/apm.h>
+
+#define MODNAME "apm"
+
+/*
+ * The apm_bios device is one of the misc char devices.
+ * This is its minor number.
+ */
+#define APM_MINOR_DEV			134
+
+/*
+ * Maximum number of events stored
+ */
+#define APM_MAX_EVENTS			16
+
+struct apm_queue {
+	unsigned int		event_head;
+	unsigned int		event_tail;
+	apm_event_t		events[APM_MAX_EVENTS];
+};
+
+/*
+ * The per-file APM data
+ */
+struct apm_user {
+	struct list_head	list;
+
+	unsigned int		suser: 1;
+	unsigned int		writer: 1;
+	unsigned int		reader: 1;
+
+	int			suspend_result;
+	unsigned int		suspend_state;
+#define SUSPEND_NONE	0		/* no suspend pending */
+#define SUSPEND_PENDING	1		/* suspend pending read */
+#define SUSPEND_READ	2		/* suspend read, pending ack */
+#define SUSPEND_ACKED	3		/* suspend acked */
+#define SUSPEND_DONE	4		/* suspend completed */
+
+	struct apm_queue	queue;
+};
+
+/*
+ * Local variables
+ */
+static int suspends_pending;
+
+static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
+static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
+
+/*
+ * This is a list of everyone who has opened /dev/apm_bios
+ */
+static DECLARE_RWSEM(user_list_lock);
+static LIST_HEAD(apm_user_list);
+
+/*
+ * kapmd info.  kapmd provides us a process context to handle
+ * "APM" events within - specifically necessary if we're going
+ * to be suspending the system.
+ */
+static DECLARE_WAIT_QUEUE_HEAD(kapmd_wait);
+static DECLARE_COMPLETION(kapmd_exit);
+static DEFINE_SPINLOCK(kapmd_queue_lock);
+static struct apm_queue kapmd_queue;
+
+int apm_suspended;
+EXPORT_SYMBOL(apm_suspended);
+
+/* Platform-specific apm_read_proc(). */
+int (*apm_get_info)(char *buf, char **start, off_t fpos, int length);
+EXPORT_SYMBOL(apm_get_info);
+
+/*
+ * APM event queue management.
+ */
+static inline int queue_empty(struct apm_queue *q)
+{
+	return q->event_head == q->event_tail;
+}
+
+static inline apm_event_t queue_get_event(struct apm_queue *q)
+{
+	q->event_tail = (q->event_tail + 1) % APM_MAX_EVENTS;
+	return q->events[q->event_tail];
+}
+
+static void queue_add_event(struct apm_queue *q, apm_event_t event)
+{
+	q->event_head = (q->event_head + 1) % APM_MAX_EVENTS;
+	if (q->event_head == q->event_tail) {
+		static int notified;
+
+		if (notified++ == 0)
+			printk(KERN_ERR "apm: an event queue overflowed\n");
+
+		q->event_tail = (q->event_tail + 1) % APM_MAX_EVENTS;
+	}
+	q->events[q->event_head] = event;
+}
+
+static void queue_event_one_user(struct apm_user *as, apm_event_t event)
+{
+	if (as->suser && as->writer) {
+		switch (event) {
+		case APM_SYS_SUSPEND:
+		case APM_USER_SUSPEND:
+			/*
+			 * If this user already has a suspend pending,
+			 * don't queue another one.
+			 */
+			if (as->suspend_state != SUSPEND_NONE)
+				return;
+
+			as->suspend_state = SUSPEND_PENDING;
+			suspends_pending++;
+			break;
+		}
+	}
+	queue_add_event(&as->queue, event);
+}
+
+static void queue_event(apm_event_t event, struct apm_user *sender)
+{
+	struct apm_user *as;
+
+	down_read(&user_list_lock);
+
+	list_for_each_entry(as, &apm_user_list, list)
+		if (as != sender && as->reader)
+			queue_event_one_user(as, event);
+
+	up_read(&user_list_lock);
+	wake_up_interruptible(&apm_waitqueue);
+}
+
+/**
+ * apm_queue_event - queue an APM event for kapmd
+ * @event: APM event
+ *
+ * Queue an APM event for kapmd to process and ultimately take the
+ * appropriate action.  Only a subset of events are handled:
+ *   %APM_LOW_BATTERY
+ *   %APM_POWER_STATUS_CHANGE
+ *   %APM_USER_SUSPEND
+ *   %APM_SYS_SUSPEND
+ *   %APM_CRITICAL_SUSPEND
+ */
+void apm_queue_event(apm_event_t event)
+{
+	spin_lock_irq(&kapmd_queue_lock);
+	queue_add_event(&kapmd_queue, event);
+	spin_unlock_irq(&kapmd_queue_lock);
+
+	wake_up_interruptible(&kapmd_wait);
+}
+EXPORT_SYMBOL(apm_queue_event);
+
+static void apm_suspend(void)
+{
+	struct apm_user *as;
+	int err;
+
+	apm_suspended = 1;
+	err = pm_suspend(PM_SUSPEND_MEM);
+
+	/*
+	 * Anyone on the APM queues will think we're still suspended.
+	 * Send a message so everyone knows we're now awake again.
+	 */
+	queue_event(APM_NORMAL_RESUME, NULL);
+
+	/*
+	 * Finally, wake up anyone who is sleeping on the suspend.
+	 */
+	down_read(&user_list_lock);
+	list_for_each_entry(as, &apm_user_list, list) {
+		as->suspend_result = err;
+		as->suspend_state = SUSPEND_DONE;
+	}
+	up_read(&user_list_lock);
+
+	wake_up(&apm_suspend_waitqueue);
+	apm_suspended = 0;
+}
+
+static ssize_t apm_read(struct file *fp, char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	struct apm_user *as = fp->private_data;
+	apm_event_t event;
+	int i = count, ret = 0;
+
+	if (count < sizeof(apm_event_t))
+		return -EINVAL;
+
+	if (queue_empty(&as->queue) && fp->f_flags & O_NONBLOCK)
+		return -EAGAIN;
+
+	wait_event_interruptible(apm_waitqueue, !queue_empty(&as->queue));
+
+	while ((i >= sizeof(event)) && !queue_empty(&as->queue)) {
+		event = queue_get_event(&as->queue);
+
+		ret = -EFAULT;
+		if (copy_to_user(buf, &event, sizeof(event)))
+			break;
+
+		if (event == APM_SYS_SUSPEND || event == APM_USER_SUSPEND)
+			as->suspend_state = SUSPEND_READ;
+
+		buf += sizeof(event);
+		i -= sizeof(event);
+	}
+
+	if (i < count)
+		ret = count - i;
+
+	return ret;
+}
+
+static unsigned int apm_poll(struct file *fp, poll_table * wait)
+{
+	struct apm_user *as = fp->private_data;
+
+	poll_wait(fp, &apm_waitqueue, wait);
+	return queue_empty(&as->queue) ? 0 : POLLIN | POLLRDNORM;
+}
+
+/*
+ * apm_ioctl - handle APM ioctl
+ *
+ * APM_IOC_SUSPEND
+ *   This IOCTL is overloaded, and performs two functions.  It is used to:
+ *     - initiate a suspend
+ *     - acknowledge a suspend read from /dev/apm_bios.
+ *   Only when everyone who has opened /dev/apm_bios with write permission
+ *   has acknowledge does the actual suspend happen.
+ */
+static int
+apm_ioctl(struct inode * inode, struct file *filp, u_int cmd, u_long arg)
+{
+	struct apm_user *as = filp->private_data;
+	unsigned long flags;
+	int err = -EINVAL;
+
+	if (!as->suser || !as->writer)
+		return -EPERM;
+
+	switch (cmd) {
+	case APM_IOC_SUSPEND:
+		as->suspend_result = -EINTR;
+
+		if (as->suspend_state == SUSPEND_READ) {
+			/*
+			 * If we read a suspend command from /dev/apm_bios,
+			 * then the corresponding APM_IOC_SUSPEND ioctl is
+			 * interpreted as an acknowledge.
+			 */
+			as->suspend_state = SUSPEND_ACKED;
+			suspends_pending--;
+		} else {
+			/*
+			 * Otherwise it is a request to suspend the system.
+			 * Queue an event for all readers, and expect an
+			 * acknowledge from all writers who haven't already
+			 * acknowledged.
+			 */
+			queue_event(APM_USER_SUSPEND, as);
+		}
+
+		/*
+		 * If there are no further acknowledges required, suspend
+		 * the system.
+		 */
+		if (suspends_pending == 0)
+			apm_suspend();
+
+		/*
+		 * Wait for the suspend/resume to complete.  If there are
+		 * pending acknowledges, we wait here for them.
+		 *
+		 * Note that we need to ensure that the PM subsystem does
+		 * not kick us out of the wait when it suspends the threads.
+		 */
+		flags = current->flags;
+		current->flags |= PF_NOFREEZE;
+
+		/*
+		 * Note: do not allow a thread which is acking the suspend
+		 * to escape until the resume is complete.
+		 */
+		if (as->suspend_state == SUSPEND_ACKED)
+			wait_event(apm_suspend_waitqueue,
+					 as->suspend_state == SUSPEND_DONE);
+		else
+			wait_event_interruptible(apm_suspend_waitqueue,
+					 as->suspend_state == SUSPEND_DONE);
+
+		current->flags = flags;
+		err = as->suspend_result;
+		as->suspend_state = SUSPEND_NONE;
+		break;
+	}
+
+	return err;
+}
+
+static int apm_release(struct inode * inode, struct file * filp)
+{
+	struct apm_user *as = filp->private_data;
+	filp->private_data = NULL;
+
+	down_write(&user_list_lock);
+	list_del(&as->list);
+	up_write(&user_list_lock);
+
+	/*
+	 * We are now unhooked from the chain.  As far as new
+	 * events are concerned, we no longer exist.  However, we
+	 * need to balance suspends_pending, which means the
+	 * possibility of sleeping.
+	 */
+	if (as->suspend_state != SUSPEND_NONE) {
+		suspends_pending -= 1;
+		if (suspends_pending == 0)
+			apm_suspend();
+	}
+
+	kfree(as);
+	return 0;
+}
+
+static int apm_open(struct inode * inode, struct file * filp)
+{
+	struct apm_user *as;
+
+	as = kzalloc(sizeof(*as), GFP_KERNEL);
+	if (as) {
+		/*
+		 * XXX - this is a tiny bit broken, when we consider BSD
+		 * process accounting. If the device is opened by root, we
+		 * instantly flag that we used superuser privs. Who knows,
+		 * we might close the device immediately without doing a
+		 * privileged operation -- cevans
+		 */
+		as->suser = capable(CAP_SYS_ADMIN);
+		as->writer = (filp->f_mode & FMODE_WRITE) == FMODE_WRITE;
+		as->reader = (filp->f_mode & FMODE_READ) == FMODE_READ;
+
+		down_write(&user_list_lock);
+		list_add(&as->list, &apm_user_list);
+		up_write(&user_list_lock);
+
+		filp->private_data = as;
+	}
+
+	return as ? 0 : -ENOMEM;
+}
+
+static struct file_operations apm_bios_fops = {
+	.owner		= THIS_MODULE,
+	.read		= apm_read,
+	.poll		= apm_poll,
+	.ioctl		= apm_ioctl,
+	.open		= apm_open,
+	.release	= apm_release,
+};
+
+static struct miscdevice apm_device = {
+	.minor		= APM_MINOR_DEV,
+	.name		= "apm_bios",
+	.fops		= &apm_bios_fops
+};
+
+
+#ifdef CONFIG_PROC_FS
+/*
+ * Arguments, with symbols from linux/apm_bios.h.
+ *
+ *   0) Linux driver version (this will change if format changes)
+ *   1) APM BIOS Version.  Usually 1.0, 1.1 or 1.2.
+ *   2) APM flags from APM Installation Check (0x00):
+ *	bit 0: APM_16_BIT_SUPPORT
+ *	bit 1: APM_32_BIT_SUPPORT
+ *	bit 2: APM_IDLE_SLOWS_CLOCK
+ *	bit 3: APM_BIOS_DISABLED
+ *	bit 4: APM_BIOS_DISENGAGED
+ *   3) AC line status
+ *	0x00: Off-line
+ *	0x01: On-line
+ *	0x02: On backup power (BIOS >= 1.1 only)
+ *	0xff: Unknown
+ *   4) Battery status
+ *	0x00: High
+ *	0x01: Low
+ *	0x02: Critical
+ *	0x03: Charging
+ *	0x04: Selected battery not present (BIOS >= 1.2 only)
+ *	0xff: Unknown
+ *   5) Battery flag
+ *	bit 0: High
+ *	bit 1: Low
+ *	bit 2: Critical
+ *	bit 3: Charging
+ *	bit 7: No system battery
+ *	0xff: Unknown
+ *   6) Remaining battery life (percentage of charge):
+ *	0-100: valid
+ *	-1: Unknown
+ *   7) Remaining battery life (time units):
+ *	Number of remaining minutes or seconds
+ *	-1: Unknown
+ *   8) min = minutes; sec = seconds
+ */
+static int apm_read_proc(char *buf, char **start, off_t fpos, int length)
+{
+	if (likely(apm_get_info))
+		return apm_get_info(buf, start, fpos, length);
+
+	return -EINVAL;
+}
+#endif
+
+static int kapmd(void *arg)
+{
+	daemonize("kapmd");
+	current->flags |= PF_NOFREEZE;
+
+	do {
+		apm_event_t event;
+
+		wait_event_interruptible(kapmd_wait,
+				!queue_empty(&kapmd_queue) || !pm_active);
+
+		if (!pm_active)
+			break;
+
+		spin_lock_irq(&kapmd_queue_lock);
+		event = 0;
+		if (!queue_empty(&kapmd_queue))
+			event = queue_get_event(&kapmd_queue);
+		spin_unlock_irq(&kapmd_queue_lock);
+
+		switch (event) {
+		case 0:
+			break;
+
+		case APM_LOW_BATTERY:
+		case APM_POWER_STATUS_CHANGE:
+			queue_event(event, NULL);
+			break;
+
+		case APM_USER_SUSPEND:
+		case APM_SYS_SUSPEND:
+			queue_event(event, NULL);
+			if (suspends_pending == 0)
+				apm_suspend();
+			break;
+
+		case APM_CRITICAL_SUSPEND:
+			apm_suspend();
+			break;
+		}
+	} while (1);
+
+	complete_and_exit(&kapmd_exit, 0);
+}
+
+static int __init apm_init(void)
+{
+	int ret;
+
+	pm_active = 1;
+
+	ret = kernel_thread(kapmd, NULL, CLONE_KERNEL);
+	if (unlikely(ret < 0)) {
+		pm_active = 0;
+		return ret;
+	}
+
+	create_proc_info_entry("apm", 0, NULL, apm_read_proc);
+
+	ret = misc_register(&apm_device);
+	if (unlikely(ret != 0)) {
+		remove_proc_entry("apm", NULL);
+
+		pm_active = 0;
+		wake_up(&kapmd_wait);
+		wait_for_completion(&kapmd_exit);
+	}
+
+	return ret;
+}
+
+static void __exit apm_exit(void)
+{
+	misc_deregister(&apm_device);
+	remove_proc_entry("apm", NULL);
+
+	pm_active = 0;
+	wake_up(&kapmd_wait);
+	wait_for_completion(&kapmd_exit);
+}
+
+module_init(apm_init);
+module_exit(apm_exit);
+
+MODULE_AUTHOR("Stephen Rothwell, Andriy Skulysh");
+MODULE_DESCRIPTION("Advanced Power Management");
+MODULE_LICENSE("GPL");
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 306eb8d83de5..b087d34dba35 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -308,7 +308,7 @@ ENTRY(exception_error)
 	.align	2
 ret_from_exception:
 	preempt_stop()
-ret_from_irq:
+ENTRY(ret_from_irq)
 	!
 	mov	#OFF_SR, r0
 	mov.l	@(r0,r15), r0	! get status register
@@ -704,7 +704,7 @@ interrupt:
 !
 !
 	.align	2
-handle_exception:
+ENTRY(handle_exception)
 	! Using k0, k1 for scratch registers (r0_bank1, r1_bank),
 	! save all registers onto stack.
 	!
diff --git a/arch/sh/kernel/pm.c b/arch/sh/kernel/pm.c
new file mode 100644
index 000000000000..10ab62c9aede
--- /dev/null
+++ b/arch/sh/kernel/pm.c
@@ -0,0 +1,88 @@
+/*
+ * Generic Power Management Routine
+ *
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/suspend.h>
+#include <linux/delay.h>
+#include <linux/gfp.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+#include <asm/watchdog.h>
+#include <asm/pm.h>
+
+#define INTR_OFFSET	0x600
+
+#define STBCR		0xffffff82
+#define STBCR2		0xffffff88
+
+#define STBCR_STBY	0x80
+#define STBCR_MSTP2	0x04
+
+#define MCR		0xffffff68
+#define RTCNT		0xffffff70
+
+#define MCR_RMODE	2
+#define MCR_RFSH	4
+
+void pm_enter(void)
+{
+	u8 stbcr, csr;
+	u16 frqcr, mcr;
+	u32 vbr_new, vbr_old;
+
+	set_bl_bit();
+
+	/* set wdt */
+	csr = sh_wdt_read_csr();
+	csr &= ~WTCSR_TME;
+	csr |= WTCSR_CKS_4096;
+	sh_wdt_write_csr(csr);
+	csr = sh_wdt_read_csr();
+	sh_wdt_write_cnt(0);
+
+	/* disable PLL1 */
+	frqcr = ctrl_inw(FRQCR);
+	frqcr &= ~(FRQCR_PLLEN | FRQCR_PSTBY);
+	ctrl_outw(frqcr, FRQCR);
+
+	/* enable standby */
+	stbcr = ctrl_inb(STBCR);
+	ctrl_outb(stbcr | STBCR_STBY | STBCR_MSTP2, STBCR);
+
+	/* set self-refresh */
+	mcr = ctrl_inw(MCR);
+	ctrl_outw(mcr & ~MCR_RFSH, MCR);
+
+	/* set interrupt handler */
+	asm volatile("stc vbr, %0" : "=r" (vbr_old));
+	vbr_new = get_zeroed_page(GFP_ATOMIC);
+	udelay(50);
+	memcpy((void*)(vbr_new + INTR_OFFSET),
+	       &wakeup_start, &wakeup_end - &wakeup_start);
+	asm volatile("ldc %0, vbr" : : "r" (vbr_new));
+
+	ctrl_outw(0, RTCNT);
+	ctrl_outw(mcr | MCR_RFSH | MCR_RMODE, MCR);
+
+	cpu_sleep();
+
+	asm volatile("ldc %0, vbr" : : "r" (vbr_old));
+
+	free_page(vbr_new);
+
+	/* enable PLL1 */
+	frqcr = ctrl_inw(FRQCR);
+	frqcr |= FRQCR_PSTBY;
+	ctrl_outw(frqcr, FRQCR);
+	udelay(50);
+	frqcr |= FRQCR_PLLEN;
+	ctrl_outw(frqcr, FRQCR);
+
+	ctrl_outb(stbcr, STBCR);
+
+	clear_bl_bit();
+}
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index b73feafcd06e..fd73ab0326e9 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -116,6 +116,10 @@ EXPORT_SYMBOL(__down_trylock);
 EXPORT_SYMBOL(synchronize_irq);
 #endif
 
+#ifdef CONFIG_PM
+EXPORT_SYMBOL(pm_suspend);
+#endif
+
 EXPORT_SYMBOL(csum_partial);
 #ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index a1589f85499d..c8db6ca4f9d1 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -143,8 +143,33 @@ void handle_timer_tick(struct pt_regs *regs)
 	}
 }
 
+#ifdef CONFIG_PM
+int timer_suspend(struct sys_device *dev, pm_message_t state)
+{
+	struct sys_timer *sys_timer = container_of(dev, struct sys_timer, dev);
+
+	sys_timer->ops->stop();
+
+	return 0;
+}
+
+int timer_resume(struct sys_device *dev)
+{
+	struct sys_timer *sys_timer = container_of(dev, struct sys_timer, dev);
+
+	sys_timer->ops->start();
+
+	return 0;
+}
+#else
+#define timer_suspend NULL
+#define timer_resume NULL
+#endif
+
 static struct sysdev_class timer_sysclass = {
 	set_kset_name("timer"),
+	.suspend = timer_suspend,
+	.resume	 = timer_resume,
 };
 
 static int __init timer_init_sysfs(void)
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index d4212add53b2..ea4bdf8cdf0d 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -188,6 +188,18 @@ static struct clk tmu0_clk = {
 	.ops		= &tmu_clk_ops,
 };
 
+static int tmu_timer_start(void)
+{
+	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+	return 0;
+}
+
+static int tmu_timer_stop(void)
+{
+	ctrl_outb(0, TMU_TSTR);
+	return 0;
+}
+
 static int tmu_timer_init(void)
 {
 	unsigned long interval;
@@ -197,7 +209,7 @@ static int tmu_timer_init(void)
 	tmu0_clk.parent = clk_get("module_clk");
 
 	/* Start TMU0 */
-	ctrl_outb(0, TMU_TSTR);
+	tmu_timer_stop();
 #if !defined(CONFIG_CPU_SUBTYPE_SH7300) && !defined(CONFIG_CPU_SUBTYPE_SH7760)
 	ctrl_outb(TMU_TOCR_INIT, TMU_TOCR);
 #endif
@@ -211,13 +223,15 @@ static int tmu_timer_init(void)
 	ctrl_outl(interval, TMU0_TCOR);
 	ctrl_outl(interval, TMU0_TCNT);
 
-	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+	tmu_timer_start();
 
 	return 0;
 }
 
 struct sys_timer_ops tmu_timer_ops = {
 	.init		= tmu_timer_init,
+	.start		= tmu_timer_start,
+	.stop		= tmu_timer_stop,
 	.get_frequency	= tmu_timer_get_frequency,
 	.get_offset	= tmu_timer_get_offset,
 };
diff --git a/drivers/input/touchscreen/hp680_ts_input.c b/drivers/input/touchscreen/hp680_ts_input.c
index fa97e0f79e7e..ee6c2f40cdf6 100644
--- a/drivers/input/touchscreen/hp680_ts_input.c
+++ b/drivers/input/touchscreen/hp680_ts_input.c
@@ -15,7 +15,6 @@
 #define HP680_TS_ABS_Y_MIN	80
 #define HP680_TS_ABS_Y_MAX	910
 
-#define	SCPCR	0xa4000116
 #define	PHDR	0xa400012e
 #define SCPDR	0xa4000136
 
@@ -77,19 +76,6 @@ static irqreturn_t hp680_ts_interrupt(int irq, void *dev, struct pt_regs *regs)
 
 static int __init hp680_ts_init(void)
 {
-	u8 scpdr;
-	u16 scpcr;
-
-	scpdr = ctrl_inb(SCPDR);
-	scpdr |= SCPDR_TS_SCAN_X | SCPDR_TS_SCAN_Y;
-	scpdr &= ~SCPDR_TS_SCAN_ENABLE;
-	ctrl_outb(scpdr, SCPDR);
-
-	scpcr = ctrl_inw(SCPCR);
-	scpcr &= ~SCPCR_TS_MASK;
-	scpcr |= SCPCR_TS_ENABLE;
-	ctrl_outw(scpcr, SCPCR);
-
 	hp680_ts_dev = input_allocate_device();
 	if (!hp680_ts_dev)
 		return -ENOMEM;
diff --git a/include/asm-sh/apm.h b/include/asm-sh/apm.h
new file mode 100644
index 000000000000..8b091e93651f
--- /dev/null
+++ b/include/asm-sh/apm.h
@@ -0,0 +1,46 @@
+/*
+ * Copyright 2006 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+
+#ifndef __ASM_SH_APM_H
+#define __ASM_SH_APM_H
+
+#define APM_AC_OFFLINE			0
+#define APM_AC_ONLINE			1
+#define APM_AC_BACKUP			2
+#define APM_AC_UNKNOWN			0xff
+
+#define APM_BATTERY_STATUS_HIGH		0
+#define APM_BATTERY_STATUS_LOW		1
+#define APM_BATTERY_STATUS_CRITICAL	2
+#define APM_BATTERY_STATUS_CHARGING	3
+#define APM_BATTERY_STATUS_NOT_PRESENT	4
+#define APM_BATTERY_STATUS_UNKNOWN	0xff
+
+#define APM_BATTERY_LIFE_UNKNOWN	0xFFFF
+#define APM_BATTERY_LIFE_MINUTES	0x8000
+#define APM_BATTERY_LIFE_VALUE_MASK	0x7FFF
+
+#define APM_BATTERY_FLAG_HIGH		(1 << 0)
+#define APM_BATTERY_FLAG_LOW		(1 << 1)
+#define APM_BATTERY_FLAG_CRITICAL	(1 << 2)
+#define APM_BATTERY_FLAG_CHARGING	(1 << 3)
+#define APM_BATTERY_FLAG_NOT_PRESENT	(1 << 7)
+#define APM_BATTERY_FLAG_UNKNOWN	0xff
+
+#define APM_UNITS_MINS			0
+#define APM_UNITS_SECS			1
+#define APM_UNITS_UNKNOWN		-1
+
+
+extern int (*apm_get_info)(char *buf, char **start, off_t fpos, int length);
+extern int apm_suspended;
+
+void apm_queue_event(apm_event_t event);
+
+#endif
diff --git a/include/asm-sh/cpu-sh3/freq.h b/include/asm-sh/cpu-sh3/freq.h
index b61b6e331df0..273f3229785c 100644
--- a/include/asm-sh/cpu-sh3/freq.h
+++ b/include/asm-sh/cpu-sh3/freq.h
@@ -18,5 +18,9 @@
 #define MIN_DIVISOR_NR		0
 #define MAX_DIVISOR_NR		4
 
+#define FRQCR_CKOEN	0x0100
+#define FRQCR_PLLEN	0x0080
+#define FRQCR_PSTBY	0x0040
+
 #endif /* __ASM_CPU_SH3_FREQ_H */
 
diff --git a/include/asm-sh/hd64461.h b/include/asm-sh/hd64461.h
index 0f2e2132cc35..27e5c34e2659 100644
--- a/include/asm-sh/hd64461.h
+++ b/include/asm-sh/hd64461.h
@@ -40,7 +40,12 @@
 #define HD64461_LCDCBAR		0x11000
 #define HD64461_LCDCLOR		0x11002
 #define HD64461_LCDCCR		0x11004
-#define HD64461_LCDCCR_MOFF	0x80
+#define HD64461_LCDCCR_STBACK	0x0400
+#define HD64461_LCDCCR_STREQ	0x0100
+#define HD64461_LCDCCR_MOFF	0x0080
+#define HD64461_LCDCCR_REFSEL	0x0040
+#define HD64461_LCDCCR_EPON	0x0020
+#define HD64461_LCDCCR_SPON	0x0010
 
 #define	HD64461_LDR1		0x11010
 #define	HD64461_LDR1_DON	0x01
diff --git a/include/asm-sh/hp6xx/hp6xx.h b/include/asm-sh/hp6xx/hp6xx.h
index a26247fd3d87..f35134c159dd 100644
--- a/include/asm-sh/hp6xx/hp6xx.h
+++ b/include/asm-sh/hp6xx/hp6xx.h
@@ -2,16 +2,33 @@
 #define __ASM_SH_HP6XX_H
 
 /*
- * Copyright (C) 2003  Andriy Skulysh
+ * Copyright (C) 2003, 2004, 2005  Andriy Skulysh
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
  */
 
-#define HP680_TS_IRQ IRQ3_IRQ
+#define HP680_BTN_IRQ		IRQ0_IRQ
+#define HP680_TS_IRQ		IRQ3_IRQ
+#define HP680_HD64461_IRQ	IRQ4_IRQ
 
 #define DAC_LCD_BRIGHTNESS	0
 #define DAC_SPEAKER_VOLUME	1
 
+#define PGDR_OPENED		0x01
+#define PGDR_MAIN_BATTERY_OUT	0x04
+#define PGDR_PLAY_BUTTON	0x08
+#define PGDR_REWIND_BUTTON	0x10
+#define PGDR_RECORD_BUTTON	0x20
+
 #define PHDR_TS_PEN_DOWN	0x08
 
+#define PJDR_LED_BLINK		0x02
+
+#define PKDR_LED_GREEN		0x10
+
 #define SCPDR_TS_SCAN_ENABLE	0x20
 #define SCPDR_TS_SCAN_Y		0x02
 #define SCPDR_TS_SCAN_X		0x01
@@ -21,11 +38,43 @@
 
 #define ADC_CHANNEL_TS_Y	1
 #define ADC_CHANNEL_TS_X	2
+#define ADC_CHANNEL_BATTERY	3
+#define ADC_CHANNEL_BACKUP	4
+#define ADC_CHANNEL_CHARGE	5
 
 #define HD64461_GPADR_SPEAKER	0x01
 #define HD64461_GPADR_PCMCIA0	(0x02|0x08)
+
 #define HD64461_GPBDR_LCDOFF	0x01
+#define HD64461_GPBDR_LCD_CONTRAST_MASK	0x78
 #define HD64461_GPBDR_LED_RED	0x80
 
+#include <asm/hd64461.h>
+#include <asm/io.h>
+
+#define PJDR	0xa4000130
+#define PKDR	0xa4000132
+
+static inline void hp6xx_led_red(int on)
+{
+	u16 v16;
+	v16 = ctrl_inw(CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000);
+	if (on)
+	    ctrl_outw(v16 & (~HD64461_GPBDR_LED_RED), CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000);
+	else
+	    ctrl_outw(v16 | HD64461_GPBDR_LED_RED, CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000);
+}
+
+static inline void hp6xx_led_green(int on)
+{
+	u8 v8;
+
+	v8 = ctrl_inb(PKDR);
+	if (on)
+	    ctrl_outb(v8 & (~PKDR_LED_GREEN), PKDR);
+	else
+	    ctrl_outb(v8 | PKDR_LED_GREEN, PKDR);
+}
+
 
 #endif /* __ASM_SH_HP6XX_H */
diff --git a/include/asm-sh/pm.h b/include/asm-sh/pm.h
new file mode 100644
index 000000000000..56fdbd6b1c94
--- /dev/null
+++ b/include/asm-sh/pm.h
@@ -0,0 +1,17 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright 2006 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ */
+#ifndef __ASM_SH_PM_H
+#define __ASM_SH_PM_H
+
+extern u8 wakeup_start;
+extern u8 wakeup_end;
+
+void pm_enter(void);
+
+#endif
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index bd7dc0554b11..477422afeb0d 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -172,6 +172,31 @@ static __inline__ void local_irq_disable(void)
 			     : "memory");
 }
 
+static __inline__ void set_bl_bit(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ ("stc	sr, %0\n\t"
+			     "or	%2, %0\n\t"
+			     "and	%3, %0\n\t"
+			     "ldc	%0, sr"
+			     : "=&r" (__dummy0), "=r" (__dummy1)
+			     : "r" (0x10000000), "r" (0xffffff0f)
+			     : "memory");
+}
+
+static __inline__ void clear_bl_bit(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ ("stc	sr, %0\n\t"
+			     "and	%2, %0\n\t"
+			     "ldc	%0, sr"
+			     : "=&r" (__dummy0), "=r" (__dummy1)
+			     : "1" (~0x10000000)
+			     : "memory");
+}
+
 #define local_save_flags(x) \
 	__asm__("stc sr, %0; and #0xf0, %0" : "=&z" (x) :/**/: "memory" )
 
diff --git a/include/asm-sh/timer.h b/include/asm-sh/timer.h
index dd6579c0b04c..c7ab28095ba0 100644
--- a/include/asm-sh/timer.h
+++ b/include/asm-sh/timer.h
@@ -6,6 +6,8 @@
 
 struct sys_timer_ops {
 	int (*init)(void);
+	int (*start)(void);
+	int (*stop)(void);
 	unsigned long (*get_offset)(void);
 	unsigned long (*get_frequency)(void);
 };
-- 
cgit v1.2.3


From 56e8d7b5786dc2f7d1f701500f8914fd2c52b111 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:24:55 +0900
Subject: sh: kgdb stub cleanups.

Some kgdb cleanup. Move hexchars/highhex/lowhex to the header, so it can
be reused by sh-sci. Also drop silly ctrl_inl/outl() overloading being
done by the kgdb stub.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/kgdb_stub.c | 33 +++++----------------------------
 include/asm-sh/kgdb.h      | 15 +++++++++++++++
 2 files changed, 20 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/kgdb_stub.c b/arch/sh/kernel/kgdb_stub.c
index 42638b92b51c..9c6315f0335d 100644
--- a/arch/sh/kernel/kgdb_stub.c
+++ b/arch/sh/kernel/kgdb_stub.c
@@ -101,16 +101,17 @@
 #include <linux/linkage.h>
 #include <linux/init.h>
 
+#ifdef CONFIG_SH_KGDB_CONSOLE
+#include <linux/console.h>
+#endif
+
 #include <asm/system.h>
 #include <asm/current.h>
 #include <asm/signal.h>
 #include <asm/pgtable.h>
 #include <asm/ptrace.h>
 #include <asm/kgdb.h>
-
-#ifdef CONFIG_SH_KGDB_CONSOLE
-#include <linux/console.h>
-#endif
+#include <asm/io.h>
 
 /* Function pointers for linkage */
 kgdb_debug_hook_t *kgdb_debug_hook;
@@ -240,7 +241,6 @@ static jmp_buf rem_com_env;
 /* Misc static */
 static int stepped_address;
 static short stepped_opcode;
-static const char hexchars[] = "0123456789abcdef";
 static char in_buffer[BUFMAX];
 static char out_buffer[OUTBUFMAX];
 
@@ -253,29 +253,6 @@ typedef unsigned char threadref[8];
 #define BUF_THREAD_ID_SIZE 16
 #endif
 
-/* Return addr as a real volatile address */
-static inline unsigned int ctrl_inl(const unsigned long addr)
-{
-	return *(volatile unsigned long *) addr;
-}
-
-/* Correctly set *addr using volatile */
-static inline void ctrl_outl(const unsigned int b, unsigned long addr)
-{
-	*(volatile unsigned long *) addr = b;
-}
-
-/* Get high hex bits */
-static char highhex(const int x)
-{
-	return hexchars[(x >> 4) & 0xf];
-}
-
-/* Get low hex bits */
-static char lowhex(const int x)
-{
-	return hexchars[x & 0xf];
-}
 
 /* Convert ch to hex */
 static int hex(const char ch)
diff --git a/include/asm-sh/kgdb.h b/include/asm-sh/kgdb.h
index 1653ffb75fbe..7b26f53fe343 100644
--- a/include/asm-sh/kgdb.h
+++ b/include/asm-sh/kgdb.h
@@ -128,4 +128,19 @@ extern int     setjmp(jmp_buf __jmpb);
 #define KGDB_ASSERT(condition, message)
 #endif
 
+/* Taken from sh-stub.c of GDB 4.18 */
+static const char hexchars[] = "0123456789abcdef";
+
+/* Get high hex bits */
+static inline char highhex(const int x)
+{
+	return hexchars[(x >> 4) & 0xf];
+}
+
+/* Get low hex bits */
+static inline char lowhex(const int x)
+{
+	return hexchars[x & 0xf];
+}
+
 #endif
-- 
cgit v1.2.3


From 959f85f8a3223c116bbe95dd8a9b207790b5d4d3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:43:28 +0900
Subject: sh: Consolidated SH7751/SH7780 PCI support.

This cleans up quite a lot of the PCI mess that we
currently have, and attempts to consolidate the
duplication in the SH7780 and SH7751 PCI controllers.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/landisk/Makefile         |   2 -
 arch/sh/boards/landisk/io.c             | 100 ++++------
 arch/sh/boards/mpc1211/setup.c          |  15 +-
 arch/sh/boards/renesas/hs7751rvoip/io.c | 116 ++++--------
 arch/sh/boards/renesas/r7780rp/io.c     | 103 ++++------
 arch/sh/boards/renesas/rts7751r2d/io.c  |  89 ++++-----
 arch/sh/boards/renesas/systemh/io.c     |  67 ++-----
 arch/sh/boards/se/770x/io.c             |  11 +-
 arch/sh/boards/se/7751/io.c             | 145 +++-----------
 arch/sh/boards/sh03/setup.c             |   6 +-
 arch/sh/boards/snapgear/io.c            |  84 +++------
 arch/sh/boards/titan/io.c               |  65 +++----
 arch/sh/cchips/hd6446x/hd64461/io.c     |   9 +-
 arch/sh/drivers/pci/Makefile            |   5 +-
 arch/sh/drivers/pci/fixups-dreamcast.c  |  38 +---
 arch/sh/drivers/pci/fixups-r7780rp.c    |  38 ++--
 arch/sh/drivers/pci/fixups-rts7751r2d.c |  24 +--
 arch/sh/drivers/pci/fixups-sh03.c       |  38 +---
 arch/sh/drivers/pci/ops-bigsur.c        |  18 +-
 arch/sh/drivers/pci/ops-landisk.c       |  68 +++++++
 arch/sh/drivers/pci/ops-r7780rp.c       |  18 +-
 arch/sh/drivers/pci/ops-rts7751r2d.c    |  13 +-
 arch/sh/drivers/pci/ops-sh4.c           | 164 ++++++++++++++++
 arch/sh/drivers/pci/ops-snapgear.c      |  21 +--
 arch/sh/drivers/pci/ops-titan.c         |  16 +-
 arch/sh/drivers/pci/pci-auto.c          |  10 -
 arch/sh/drivers/pci/pci-sh4.h           | 180 ++++++++++++++++++
 arch/sh/drivers/pci/pci-sh7751.c        | 322 ++++++--------------------------
 arch/sh/drivers/pci/pci-sh7751.h        | 174 +----------------
 arch/sh/drivers/pci/pci-sh7780.c        | 270 ++++----------------------
 arch/sh/drivers/pci/pci-sh7780.h        |  94 +---------
 arch/sh/drivers/pci/pci-st40.c          |  19 --
 arch/sh/drivers/pci/pci.c               |  40 +++-
 arch/sh/kernel/cpu/sh4/Makefile         |   3 +
 arch/sh/kernel/cpu/sh4/setup-sh7780.c   |  79 ++++++++
 include/asm-sh/io.h                     |   5 +
 include/asm-sh/pci.h                    |  35 +++-
 37 files changed, 984 insertions(+), 1520 deletions(-)
 create mode 100644 arch/sh/drivers/pci/ops-landisk.c
 create mode 100644 arch/sh/drivers/pci/ops-sh4.c
 create mode 100644 arch/sh/drivers/pci/pci-sh4.h
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7780.c

(limited to 'include')

diff --git a/arch/sh/boards/landisk/Makefile b/arch/sh/boards/landisk/Makefile
index 60e75c6db0eb..89e4beb2ad47 100644
--- a/arch/sh/boards/landisk/Makefile
+++ b/arch/sh/boards/landisk/Makefile
@@ -3,5 +3,3 @@
 #
 
 obj-y	 := setup.o io.o irq.o rtc.o landisk_pwb.o
-
-obj-$(CONFIG_PCI) += pci.o
diff --git a/arch/sh/boards/landisk/io.c b/arch/sh/boards/landisk/io.c
index aa6b145c9e8f..92498b4947d5 100644
--- a/arch/sh/boards/landisk/io.c
+++ b/arch/sh/boards/landisk/io.c
@@ -14,39 +14,16 @@
  * modifed by kogiidena
  * 2005.03.03
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/pci.h>
 #include <asm/landisk/iodata_landisk.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
 
-#include <linux/module.h>
-#include <linux/pci.h>
-#include "../../drivers/pci/pci-sh7751.h"
-
 extern void *area5_io_base;	/* Area 5 I/O Base address */
 extern void *area6_io_base;	/* Area 6 I/O Base address */
 
-/*
- * The 7751R LANDISK uses the built-in PCI controller (PCIC)
- * of the 7751R processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
@@ -67,17 +44,6 @@ static inline unsigned long port2adr(unsigned int port)
 	return port;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH_7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -89,8 +55,8 @@ u8 landisk_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inb(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		return ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		return ctrl_inb(pci_ioaddr(port));
 
 	return ctrl_inw(port2adr(port)) & 0xff;
 }
@@ -101,12 +67,12 @@ u8 landisk_inb_p(unsigned long port)
 
 	if (PXSEG(port))
 		v = ctrl_inb(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		v = ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		v = ctrl_inb(pci_ioaddr(port));
 	else
 		v = ctrl_inw(port2adr(port)) & 0xff;
 
-	delay();
+	ctrl_delay();
 
 	return v;
 }
@@ -115,8 +81,8 @@ u16 landisk_inw(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inw(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		return ctrl_inw(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		return ctrl_inw(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -127,8 +93,8 @@ u32 landisk_inl(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inl(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		return ctrl_inl(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		return ctrl_inl(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -139,8 +105,8 @@ void landisk_outb(u8 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 }
@@ -149,19 +115,19 @@ void landisk_outb_p(u8 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
-	delay();
+	ctrl_delay();
 }
 
 void landisk_outw(u16 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outw(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outw(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outw(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -170,8 +136,8 @@ void landisk_outl(u32 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outl(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outl(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outl(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -184,8 +150,8 @@ void landisk_insb(unsigned long port, void *dst, unsigned long count)
         if (PXSEG(port)) {
                 while (count--)
                         *buf++ = *(volatile u8 *)port;
-	} else if (CHECK_SH7751_PCIIO(port)) {
-                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	} else if (is_pci_ioaddr(port)) {
+                volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
                 while (count--)
                         *buf++ = *bp;
@@ -203,8 +169,8 @@ void landisk_insw(unsigned long port, void *dst, unsigned long count)
 
 	if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 	while (count--)
@@ -215,8 +181,8 @@ void landisk_insl(unsigned long port, void *dst, unsigned long count)
 {
         u32 *buf = dst;
 
-	if (CHECK_SH7751_PCIIO(port)) {
-                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port)) {
+                volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
                 while (count--)
                         *buf++ = *p;
@@ -232,8 +198,8 @@ void landisk_outsb(unsigned long port, const void *src, unsigned long count)
 	if (PXSEG(port))
                 while (count--)
                         ctrl_outb(*buf++, port);
-	else if (CHECK_SH7751_PCIIO(port)) {
-                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port)) {
+                volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
                 while (count--)
                         *bp = *buf++;
@@ -251,8 +217,8 @@ void landisk_outsw(unsigned long port, const void *src, unsigned long count)
 
 	if (PXSEG(port))
                 p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                p = (volatile u16 *)pci_ioaddr(port);
 	else
                 p = (volatile u16 *)port2adr(port);
 
@@ -264,8 +230,8 @@ void landisk_outsl(unsigned long port, const void *src, unsigned long count)
 {
         const u32 *buf = src;
 
-	if (CHECK_SH7751_PCIIO(port)) {
-                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port)) {
+                volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
                 while (count--)
                         *p = *buf++;
@@ -277,8 +243,8 @@ void __iomem *landisk_ioport_map(unsigned long port, unsigned int size)
 {
         if (PXSEG(port))
                 return (void __iomem *)port;
-        else if (CHECK_SH7751_PCIIO(port))
-                return (void __iomem *)PCI_IOMAP(port);
+        else if (is_pci_ioaddr(port))
+                return (void __iomem *)pci_ioaddr(port);
 
         return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/mpc1211/setup.c b/arch/sh/boards/mpc1211/setup.c
index 2bfb221cc35c..a8c5180ae219 100644
--- a/arch/sh/boards/mpc1211/setup.c
+++ b/arch/sh/boards/mpc1211/setup.c
@@ -255,23 +255,12 @@ void __init init_mpc1211_IRQ(void)
 	}
 }
 
-/*
-  Initialize the board
-*/
-
-
-static void delay (void)
-{
-	volatile unsigned short tmp;
-	tmp = *(volatile unsigned short *) 0xa0000000;
-}
-
-static void delay1000 (void)
+static void delay1000(void)
 {
 	int i;
 
 	for (i=0; i<1000; i++)
-		delay ();
+		ctrl_delay();
 }
 
 static int put_smb_blk(unsigned char *p, int address, int command, int no)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index ecdce7ef6a34..8c26550ca2e4 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -13,14 +13,11 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/module.h>
 #include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
 #include <asm/addrspace.h>
 
-#include <linux/module.h>
-#include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-
 extern void *area6_io8_base;	/* Area 6 8bit I/O Base address */
 extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
 
@@ -31,27 +28,17 @@ extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
  * like the other Solution Engine boards.
  */
 
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
 #define CODEC_IO_BASE	0x1000
 #define CODEC_IOMAP(a)	((unsigned long)area6_io8_base + ((a) - CODEC_IO_BASE))
 
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
 		if (port == 0x3f6)
 			return ((unsigned long)area5_io16_base + 0x0c);
 		else
-			return ((unsigned long)area5_io16_base + 0x800 + ((port-0x1f0) << 1));
+			return ((unsigned long)area5_io16_base + 0x800 +
+				((port-0x1f0) << 1));
 	else
 		maybebadio((unsigned long)port);
 	return port;
@@ -70,25 +57,10 @@ static inline int shifted_port(unsigned long port)
 }
 
 #if defined(CONFIG_HS7751RVOIP_CODEC)
-static inline int
-codec_port(unsigned long port)
-{
-	if (CODEC_IO_BASE <= port && port < (CODEC_IO_BASE+0x20))
-		return 1;
-	else
-		return 0;
-}
-#endif
-
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
+#define codec_port(port)	\
+	((CODEC_IO_BASE <= (port)) && ((port) < (CODEC_IO_BASE + 0x20)))
 #else
-#define CHECK_SH7751_PCIIO(port) (0)
+#define codec_port(port)	(0)
 #endif
 
 /*
@@ -102,12 +74,10 @@ unsigned char hs7751rvoip_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inb(port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		return ctrl_inb(CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inb(pci_ioaddr(port));
 	else
 		return ctrl_inw(port2adr(port)) & 0xff;
 }
@@ -118,15 +88,13 @@ unsigned char hs7751rvoip_inb_p(unsigned long port)
 
         if (PXSEG(port))
 		v = ctrl_inb(port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		v = ctrl_inb(CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		v = ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		v = ctrl_inb(pci_ioaddr(port));
 	else
 		v = ctrl_inw(port2adr(port)) & 0xff;
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -134,8 +102,8 @@ unsigned short hs7751rvoip_inw(unsigned long port)
 {
         if (PXSEG(port))
 		return ctrl_inw(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return ctrl_inw(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inw(pci_ioaddr(port));
 	else
 		maybebadio(port);
 	return 0;
@@ -145,8 +113,8 @@ unsigned int hs7751rvoip_inl(unsigned long port)
 {
         if (PXSEG(port))
 		return ctrl_inl(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return ctrl_inl(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inl(pci_ioaddr(port));
 	else
 		maybebadio(port);
 	return 0;
@@ -157,12 +125,10 @@ void hs7751rvoip_outb(unsigned char value, unsigned long port)
 
         if (PXSEG(port))
 		ctrl_outb(value, port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		ctrl_outb(value, CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outb(value, port2adr(port));
 }
@@ -171,24 +137,22 @@ void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
 		ctrl_outb(value, port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		ctrl_outb(value, CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 
-	delay();
+	ctrl_delay();
 }
 
 void hs7751rvoip_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
 		ctrl_outw(value, port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outw(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outw(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -197,8 +161,8 @@ void hs7751rvoip_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
 		ctrl_outl(value, port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outl(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outl(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -210,13 +174,11 @@ void hs7751rvoip_insb(unsigned long port, void *addr, unsigned long count)
 	if (PXSEG(port))
 		while (count--)
 			*buf++ = ctrl_inb(port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		while (count--)
 			*buf++ = ctrl_inb(CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*buf++ = *bp;
@@ -235,8 +197,8 @@ void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
 
 	if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 	while (count--)
@@ -246,8 +208,8 @@ void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
 void hs7751rvoip_insl(unsigned long port, void *addr, unsigned long count)
 {
 
-	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 		u32 *buf = addr;
 
 		while (count--)
@@ -263,13 +225,11 @@ void hs7751rvoip_outsb(unsigned long port, const void *addr, unsigned long count
 	if (PXSEG(port))
 		while (count--)
 			ctrl_outb(*buf++, port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		while (count--)
 			ctrl_outb(*buf++, CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*bp = *buf++;
@@ -288,8 +248,8 @@ void hs7751rvoip_outsw(unsigned long port, const void *addr, unsigned long count
 
 	if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 
@@ -301,8 +261,8 @@ void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count
 {
 	const u32 *buf = addr;
 
-	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
 		while (count--)
 			*p = *buf++;
@@ -316,8 +276,8 @@ void __iomem *hs7751rvoip_ioport_map(unsigned long port, unsigned int size)
                 return (void __iomem *)port;
 	else if (unlikely(codec_port(port) && (size == 1)))
 		return (void __iomem *)CODEC_IOMAP(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return (void __iomem *)PCI_IOMAP(port);
+        else if (is_pci_ioaddr(port))
+                return (void __iomem *)pci_ioaddr(port);
 
         return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/renesas/r7780rp/io.c b/arch/sh/boards/renesas/r7780rp/io.c
index f73ca3f0f5a1..db92d6e6ae99 100644
--- a/arch/sh/boards/renesas/r7780rp/io.c
+++ b/arch/sh/boards/renesas/r7780rp/io.c
@@ -1,6 +1,4 @@
 /*
- * linux/arch/sh/kernel/io_r7780rp.c
- *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
  *
@@ -10,37 +8,13 @@
  * placeholder code from io_r7780rp.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
+#include <linux/pci.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <asm/r7780rp/r7780rp.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
 
-#include <linux/module.h>
-#include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7780.h"
-
-/*
- * The 7780 R7780RP-1 uses the built-in PCI controller (PCIC)
- * of the 7780 processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */
-
-#define	SH7780_PCIIOBR_MASK	0xFFFC0000	/* IO Space Mask */
-#define PCIIOBR		(volatile long *)PCI_REG(SH7780_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7780_PCIMBR)
-#define PCI_IO_AREA	SH7780_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7780_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7780_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
@@ -78,17 +52,6 @@ static inline int shifted_port(unsigned long port)
 		return 1;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7780_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7780_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7780_PCIIO(port) (0)
-#endif
-
 #if defined(CONFIG_NE2000) || defined(CONFIG_NE2000_MODULE)
 #define CHECK_AX88796L_PORT(port) \
   ((port >= AX88796L_IO_BASE) && (port < (AX88796L_IO_BASE+0x20)))
@@ -109,8 +72,8 @@ u8 r7780rp_inb(unsigned long port)
 		return ctrl_inw(port88796l(port, 0)) & 0xff;
 	else if (PXSEG(port))
 		return ctrl_inb(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inb(pci_ioaddr(port));
 
 	return ctrl_inw(port2adr(port)) & 0xff;
 }
@@ -123,12 +86,12 @@ u8 r7780rp_inb_p(unsigned long port)
 		v = ctrl_inw(port88796l(port, 0)) & 0xff;
 	else if (PXSEG(port))
 		v = ctrl_inb(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		v = ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		v = ctrl_inb(pci_ioaddr(port));
 	else
 		v = ctrl_inw(port2adr(port)) & 0xff;
 
-	delay();
+	ctrl_delay();
 
 	return v;
 }
@@ -139,8 +102,8 @@ u16 r7780rp_inw(unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		return ctrl_inw(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return ctrl_inw(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inw(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -153,8 +116,8 @@ u32 r7780rp_inl(unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		return ctrl_inl(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return ctrl_inl(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inl(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -167,8 +130,8 @@ void r7780rp_outb(u8 value, unsigned long port)
 		ctrl_outw(value, port88796l(port, 0));
 	else if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 }
@@ -179,12 +142,12 @@ void r7780rp_outb_p(u8 value, unsigned long port)
 		ctrl_outw(value, port88796l(port, 0));
 	else if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 
-	delay();
+	ctrl_delay();
 }
 
 void r7780rp_outw(u16 value, unsigned long port)
@@ -193,8 +156,8 @@ void r7780rp_outw(u16 value, unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		ctrl_outw(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outw(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outw(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -205,8 +168,8 @@ void r7780rp_outl(u32 value, unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		ctrl_outl(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outl(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outl(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -223,8 +186,8 @@ void r7780rp_insb(unsigned long port, void *dst, unsigned long count)
 	} else if (PXSEG(port)) {
 		while (count--)
 			*buf++ = *(volatile u8 *)port;
-	} else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	} else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*buf++ = *bp;
@@ -244,8 +207,8 @@ void r7780rp_insw(unsigned long port, void *dst, unsigned long count)
 		p = (volatile u16 *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 
@@ -259,8 +222,8 @@ void r7780rp_insl(unsigned long port, void *dst, unsigned long count)
 
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
 		while (count--)
 			*buf++ = *p;
@@ -280,8 +243,8 @@ void r7780rp_outsb(unsigned long port, const void *src, unsigned long count)
 	} else if (PXSEG(port))
 		while (count--)
 			ctrl_outb(*buf++, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*bp = *buf++;
@@ -301,8 +264,8 @@ void r7780rp_outsw(unsigned long port, const void *src, unsigned long count)
 		p = (volatile u16 *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 
@@ -316,8 +279,8 @@ void r7780rp_outsl(unsigned long port, const void *src, unsigned long count)
 
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
 		while (count--)
 			*p = *buf++;
@@ -331,8 +294,8 @@ void __iomem *r7780rp_ioport_map(unsigned long port, unsigned int size)
 		return (void __iomem *)port88796l(port, size > 1);
 	else if (PXSEG(port))
 		return (void __iomem *)port;
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return (void __iomem *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return (void __iomem *)pci_ioaddr(port);
 
 	return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/renesas/rts7751r2d/io.c b/arch/sh/boards/renesas/rts7751r2d/io.c
index 8dc2a2e2e5df..135aa0b5e62d 100644
--- a/arch/sh/boards/renesas/rts7751r2d/io.c
+++ b/arch/sh/boards/renesas/rts7751r2d/io.c
@@ -1,6 +1,4 @@
 /*
- * linux/arch/sh/kernel/io_rts7751r2d.c
- *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
  *
@@ -10,11 +8,9 @@
  * placeholder code from io_rts7751r2d.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7751.h"
 #include <asm/rts7751r2d/rts7751r2d.h>
 #include <asm/io.h>
 #include <asm/addrspace.h>
@@ -26,18 +22,6 @@
  * like the other Solution Engine boards.
  */
 
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
@@ -75,17 +59,6 @@ static inline int shifted_port(unsigned long port)
 		return 1;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 #if defined(CONFIG_NE2000) || defined(CONFIG_NE2000_MODULE)
 #define CHECK_AX88796L_PORT(port) \
   ((port >= AX88796L_IO_BASE) && (port < (AX88796L_IO_BASE+0x20)))
@@ -106,8 +79,8 @@ unsigned char rts7751r2d_inb(unsigned long port)
 		return (*(volatile unsigned short *)port88796l(port, 0)) & 0xff;
 	else if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else
 		return (*(volatile unsigned short *)port2adr(port) & 0xff);
 }
@@ -120,11 +93,12 @@ unsigned char rts7751r2d_inb_p(unsigned long port)
 		v = (*(volatile unsigned short *)port88796l(port, 0)) & 0xff;
         else if (PXSEG(port))
 		v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		v = *(volatile unsigned char *)pci_ioaddr(port);
 	else
 		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
-	delay();
+
+	ctrl_delay();
 
 	return v;
 }
@@ -135,8 +109,8 @@ unsigned short rts7751r2d_inw(unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return *(volatile unsigned short *)pci_ioaddr(port);
 	else
 		maybebadio(port);
 
@@ -149,8 +123,8 @@ unsigned int rts7751r2d_inl(unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned long *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return *(volatile unsigned long *)pci_ioaddr(port);
 	else
 		maybebadio(port);
 
@@ -163,8 +137,8 @@ void rts7751r2d_outb(unsigned char value, unsigned long port)
 		*((volatile unsigned short *)port88796l(port, 0)) = value;
         else if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned char *)pci_ioaddr(port) = value;
 	else
 		*(volatile unsigned short *)port2adr(port) = value;
 }
@@ -175,11 +149,12 @@ void rts7751r2d_outb_p(unsigned char value, unsigned long port)
 		*((volatile unsigned short *)port88796l(port, 0)) = value;
         else if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned char *)pci_ioaddr(port) = value;
 	else
 		*(volatile unsigned short *)port2adr(port) = value;
-	delay();
+
+	ctrl_delay();
 }
 
 void rts7751r2d_outw(unsigned short value, unsigned long port)
@@ -188,8 +163,8 @@ void rts7751r2d_outw(unsigned short value, unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		*(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned short *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned short *)pci_ioaddr(port) = value;
 	else
 		maybebadio(port);
 }
@@ -200,8 +175,8 @@ void rts7751r2d_outl(unsigned int value, unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		*(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned long *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned long *)pci_ioaddr(port) = value;
 	else
 		maybebadio(port);
 }
@@ -219,8 +194,8 @@ void rts7751r2d_insb(unsigned long port, void *addr, unsigned long count)
 	} else if (PXSEG(port))
 		while (count--)
 			ctrl_outb(ctrl_inb(port), a++);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		bp = (__u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		bp = (__u8 *)pci_ioaddr(port);
 		while (count--)
 			ctrl_outb(*bp, a++);
 	} else {
@@ -239,8 +214,8 @@ void rts7751r2d_insw(unsigned long port, void *addr, unsigned long count)
 		p = (volatile unsigned short *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile unsigned short *)pci_ioaddr(port);
 	else
 		p = (volatile unsigned short *)port2adr(port);
 	while (count--)
@@ -251,11 +226,11 @@ void rts7751r2d_insl(unsigned long port, void *addr, unsigned long count)
 {
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
 		unsigned long a = (unsigned long)addr;
 
 		while (count--) {
-			ctrl_outl(ctrl_inl(PCI_IOMAP(port)), a);
+			ctrl_outl(ctrl_inl(pci_ioaddr(port)), a);
 			a += 4;
 		}
 	} else
@@ -275,8 +250,8 @@ void rts7751r2d_outsb(unsigned long port, const void *addr, unsigned long count)
 	} else if (PXSEG(port))
 		while (count--)
 			ctrl_outb(a++, port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		bp = (__u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		bp = (__u8 *)pci_ioaddr(port);
 		while (count--)
 			*bp = ctrl_inb(a++);
 	} else {
@@ -295,8 +270,8 @@ void rts7751r2d_outsw(unsigned long port, const void *addr, unsigned long count)
 		p = (volatile unsigned short *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile unsigned short *)pci_ioaddr(port);
 	else
 		p = (volatile unsigned short *)port2adr(port);
 
@@ -310,11 +285,11 @@ void rts7751r2d_outsl(unsigned long port, const void *addr, unsigned long count)
 {
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
 		unsigned long a = (unsigned long)addr;
 
 		while (count--) {
-			ctrl_outl(ctrl_inl(a), PCI_IOMAP(port));
+			ctrl_outl(ctrl_inl(a), pci_ioaddr(port));
 			a += 4;
 		}
 	} else
diff --git a/arch/sh/boards/renesas/systemh/io.c b/arch/sh/boards/renesas/systemh/io.c
index 0befd4f9894c..cde6e5d192c4 100644
--- a/arch/sh/boards/renesas/systemh/io.c
+++ b/arch/sh/boards/renesas/systemh/io.c
@@ -5,36 +5,16 @@
  * Based largely on io_se.c.
  *
  * I/O routine for Hitachi 7751 Systemh.
- *
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <asm/systemh7751.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-
-/*
- * The 7751 SystemH Engine uses the built-in PCI controller (PCIC)
- * of the 7751 processor, and has a SuperIO accessible on its memory
- * bus.
- */
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
 
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 #define ETHER_IOMAP(adr) (0xB3000000 + (adr)) /*map to 16bits access area
                                                 of smc lan chip*/
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline volatile __u16 *
 port2adr(unsigned int port)
 {
@@ -44,17 +24,6 @@ port2adr(unsigned int port)
 	return (volatile __u16*)port;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -66,8 +35,8 @@ unsigned char sh7751systemh_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else if (port <= 0x3F1)
 		return *(volatile unsigned char *)ETHER_IOMAP(port);
 	else
@@ -80,13 +49,13 @@ unsigned char sh7751systemh_inb_p(unsigned long port)
 
         if (PXSEG(port))
                 v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                v = *(volatile unsigned char *)pci_ioaddr(port);
 	else if (port <= 0x3F1)
 		v = *(volatile unsigned char *)ETHER_IOMAP(port);
 	else
 		v = (*port2adr(port))&0xff;
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -94,8 +63,8 @@ unsigned short sh7751systemh_inw(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned short *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else if (port <= 0x3F1)
@@ -109,8 +78,8 @@ unsigned int sh7751systemh_inl(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned int *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned int *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else if (port <= 0x3F1)
@@ -125,8 +94,8 @@ void sh7751systemh_outb(unsigned char value, unsigned long port)
 
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else if (port <= 0x3F1)
 		*(volatile unsigned char *)ETHER_IOMAP(port) = value;
 	else
@@ -137,21 +106,21 @@ void sh7751systemh_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else if (port <= 0x3F1)
 		*(volatile unsigned char *)ETHER_IOMAP(port) = value;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
 void sh7751systemh_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned short *)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned short *)pci_ioaddr(port)) = value;
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else if (port <= 0x3F1)
@@ -164,8 +133,8 @@ void sh7751systemh_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned long*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned long*)pci_ioaddr(port)) = value;
 	else
 		maybebadio(port);
 }
diff --git a/arch/sh/boards/se/770x/io.c b/arch/sh/boards/se/770x/io.c
index 5102201c97ab..9941949331ab 100644
--- a/arch/sh/boards/se/770x/io.c
+++ b/arch/sh/boards/se/770x/io.c
@@ -1,4 +1,4 @@
-/* $Id: io.c,v 1.6 2006/01/04 17:53:54 lethal Exp $
+/* $Id: io.c,v 1.7 2006/02/05 21:55:29 lethal Exp $
  *
  * linux/arch/sh/kernel/io_se.c
  *
@@ -20,11 +20,6 @@ int sh_pcic_io_stop;
 int sh_pcic_io_type;
 int sh_pcic_io_dummy;
 
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 /* MS7750 requires special versions of in*, out* routines, since
    PC-like io ports are located at upper half byte of 16-bit word which
    can be accessed only with 16-bit wide.  */
@@ -72,7 +67,7 @@ unsigned char se_inb_p(unsigned long port)
 		v = (*port2adr(port) >> 8); 
 	else
 		v = (*port2adr(port))&0xff; 
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -110,7 +105,7 @@ void se_outb_p(unsigned char value, unsigned long port)
 		*(port2adr(port)) = value << 8;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
 void se_outw(unsigned short value, unsigned long port)
diff --git a/arch/sh/boards/se/7751/io.c b/arch/sh/boards/se/7751/io.c
index 0e8a3ba48316..e8d846cec89d 100644
--- a/arch/sh/boards/se/7751/io.c
+++ b/arch/sh/boards/se/7751/io.c
@@ -1,6 +1,4 @@
 /*
- * linux/arch/sh/kernel/io_7751se.c
- *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
  *
@@ -10,55 +8,14 @@
  * placeholder code from io_se.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/se7751.h>
 #include <asm/addrspace.h>
 
-#include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-
-#if 0
-/******************************************************************
- * Variables from io_se.c, related to PCMCIA (not PCI); we're not
- * compiling them in, and have removed references from functions
- * which follow.  [Many checked for IO ports in the range bounded
- * by sh_pcic_io_start/stop, and used sh_pcic_io_wbase as offset.
- * As start/stop are uninitialized, only port 0x0 would match?]
- * When used, remember to adjust names to avoid clash with io_se?
- *****************************************************************/
-/* SH pcmcia io window base, start and end.  */
-int sh_pcic_io_wbase = 0xb8400000;
-int sh_pcic_io_start;
-int sh_pcic_io_stop;
-int sh_pcic_io_type;
-int sh_pcic_io_dummy;
-/*************************************************************/
-#endif
-
-/*
- * The 7751 Solution Engine uses the built-in PCI controller (PCIC)
- * of the 7751 processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */ 
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
-static inline volatile __u16 *
-port2adr(unsigned int port)
+static inline volatile u16 *port2adr(unsigned int port)
 {
 	if (port >= 0x2000)
 		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
@@ -66,32 +23,6 @@ port2adr(unsigned int port)
 	return (volatile __u16*)port;
 }
 
-#if 0
-/* The 7751 Solution Engine seems to have everything hooked */
-/* up pretty normally (nothing on high-bytes only...) so this */
-/* shouldn't be needed */
-static inline int
-shifted_port(unsigned long port)
-{
-	/* For IDE registers, value is not shifted */
-	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
-		return 0;
-	else
-		return 1;
-}
-#endif
-
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -103,10 +34,10 @@ unsigned char sh7751se_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		return (*port2adr(port))&0xff; 
+		return (*port2adr(port)) & 0xff;
 }
 
 unsigned char sh7751se_inb_p(unsigned long port)
@@ -115,11 +46,11 @@ unsigned char sh7751se_inb_p(unsigned long port)
 
         if (PXSEG(port))
                 v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                v = *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		v = (*port2adr(port))&0xff; 
-	delay();
+		v = (*port2adr(port)) & 0xff;
+	ctrl_delay();
 	return v;
 }
 
@@ -127,8 +58,8 @@ unsigned short sh7751se_inw(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned short *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -140,8 +71,8 @@ unsigned int sh7751se_inl(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned int *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned int *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -154,8 +85,8 @@ void sh7751se_outb(unsigned char value, unsigned long port)
 
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
 }
@@ -164,19 +95,19 @@ void sh7751se_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
 void sh7751se_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned short *)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned short *)pci_ioaddr(port)) = value;
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
@@ -187,8 +118,8 @@ void sh7751se_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned long*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned long*)pci_ioaddr(port)) = value;
 	else
 		maybebadio(port);
 }
@@ -202,35 +133,3 @@ void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count)
 {
 	maybebadio(port);
 }
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-#if 0
-static int
-sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-	return 0;
-}
-#endif
-
-unsigned long
-sh7751se_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
-}
diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c
index dab742a00c6f..7d31d6aed8a5 100644
--- a/arch/sh/boards/sh03/setup.c
+++ b/arch/sh/boards/sh03/setup.c
@@ -12,11 +12,10 @@
 #include <asm/sh03/io.h>
 #include <asm/sh03/sh03.h>
 #include <asm/addrspace.h>
-#include "../../drivers/pci/pci-sh7751.h"
 
 const char *get_system_type(void)
 {
-	return "Interface CTP/PCI-SH03)";
+	return "Interface (CTP/PCI-SH03)";
 }
 
 static void init_sh03_IRQ(void)
@@ -39,7 +38,7 @@ static void __iomem *sh03_ioport_map(unsigned long port, unsigned int size)
 	if (((port >= 0x1f0) && (port <= 0x1f7)) || (port == 0x3f6))
 		return (void __iomem *)((unsigned long)cf_io_base + port);
 
-        return (void __iomem *)(port + SH7751_PCI_IO_BASE);
+        return (void __iomem *)(port + PCI_IO_BASE);
 }
 
 struct sh_machine_vector mv_sh03 __initmv = {
@@ -51,7 +50,6 @@ struct sh_machine_vector mv_sh03 __initmv = {
 	.mv_heartbeat		= heartbeat_sh03,
 #endif
 };
-
 ALIAS_MV(sh03)
 
 /* arch/sh/boards/sh03/rtc.c */
diff --git a/arch/sh/boards/snapgear/io.c b/arch/sh/boards/snapgear/io.c
index 9f700b8392bb..0f4824264557 100644
--- a/arch/sh/boards/snapgear/io.c
+++ b/arch/sh/boards/snapgear/io.c
@@ -1,6 +1,4 @@
-/* 
- * linux/arch/sh/kernel/io_7751se.c
- *
+/*
  * Copyright (C) 2002  David McCullough <davidm@snapgear.com>
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
@@ -11,54 +9,22 @@
  * placeholder code from io_se.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/addrspace.h>
 
-#include <asm/pci.h>
-#include "../../drivers/pci/pci-sh7751.h"
-
 #ifdef CONFIG_SH_SECUREEDGE5410
 unsigned short secureedge5410_ioport;
 #endif
 
-/*
- * The SnapGear uses the built-in PCI controller (PCIC)
- * of the 7751 processor
- */
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline volatile __u16 *port2adr(unsigned int port)
 {
 	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -66,39 +32,36 @@ static inline volatile __u16 *port2adr(unsigned int port)
  * should be way beyond the window, and is used  w/o translation for
  * compatibility.
  */
-
 unsigned char snapgear_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		return (*port2adr(port))&0xff; 
+		return (*port2adr(port)) & 0xff;
 }
 
-
 unsigned char snapgear_inb_p(unsigned long port)
 {
 	unsigned char v;
 
 	if (PXSEG(port))
 		v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		v = *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		v = (*port2adr(port))&0xff; 
-	delay();
+		v = (*port2adr(port))&0xff;
+	ctrl_delay();
 	return v;
 }
 
-
 unsigned short snapgear_inw(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned short *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -106,13 +69,12 @@ unsigned short snapgear_inw(unsigned long port)
 	return 0;
 }
 
-
 unsigned int snapgear_inl(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned int *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned int *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -120,50 +82,46 @@ unsigned int snapgear_inl(unsigned long port)
 	return 0;
 }
 
-
 void snapgear_outb(unsigned char value, unsigned long port)
 {
 
 	if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
 }
 
-
 void snapgear_outb_p(unsigned char value, unsigned long port)
 {
 	if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
-
 void snapgear_outw(unsigned short value, unsigned long port)
 {
 	if (PXSEG(port))
 		*(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned short *)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned short *)pci_ioaddr(port)) = value;
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
 		maybebadio(port);
 }
 
-
 void snapgear_outl(unsigned int value, unsigned long port)
 {
 	if (PXSEG(port))
 		*(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned long*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned long*)pci_ioaddr(port)) = value;
 	else
 		maybebadio(port);
 }
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
index 48f3494f55b1..4730c1dd697d 100644
--- a/arch/sh/boards/titan/io.c
+++ b/arch/sh/boards/titan/io.c
@@ -1,34 +1,11 @@
 /*
  *	I/O routines for Titan
  */
-
 #include <linux/pci.h>
 #include <asm/machvec.h>
 #include <asm/addrspace.h>
 #include <asm/titan.h>
 #include <asm/io.h>
-#include "../../drivers/pci/pci-sh7751.h"
-
-#define PCIIOBR         (volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA     SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA    SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)  (PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#define CHECK_SH7751_PCIMEMIO(port) \
-  ((port >= PCIBIOS_MIN_MEM) && (port < (PCIBIOS_MIN_MEM + SH7751_PCI_MEM_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
-static inline void delay(void)
-{
-        ctrl_inw(0xa0000000);
-}
 
 static inline unsigned int port2adr(unsigned int port)
 {
@@ -40,8 +17,8 @@ u8 titan_inb(unsigned long port)
 {
         if (PXSEG(port))
                 return ctrl_inb(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return ctrl_inb(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                return ctrl_inb(pci_ioaddr(port));
         return ctrl_inw(port2adr(port)) & 0xff;
 }
 
@@ -51,11 +28,11 @@ u8 titan_inb_p(unsigned long port)
 
         if (PXSEG(port))
                 v = ctrl_inb(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                v = ctrl_inb(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                v = ctrl_inb(pci_ioaddr(port));
         else
                 v = ctrl_inw(port2adr(port)) & 0xff;
-        delay();
+        ctrl_delay();
         return v;
 }
 
@@ -63,8 +40,8 @@ u16 titan_inw(unsigned long port)
 {
         if (PXSEG(port))
                 return ctrl_inw(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return ctrl_inw(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                return ctrl_inw(pci_ioaddr(port));
         else if (port >= 0x2000)
                 return ctrl_inw(port2adr(port));
         else
@@ -76,8 +53,8 @@ u32 titan_inl(unsigned long port)
 {
         if (PXSEG(port))
                 return ctrl_inl(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return ctrl_inl(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                return ctrl_inl(pci_ioaddr(port));
         else if (port >= 0x2000)
                 return ctrl_inw(port2adr(port));
         else
@@ -89,8 +66,8 @@ void titan_outb(u8 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outb(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outb(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outb(value, pci_ioaddr(port));
         else
                 ctrl_outw(value, port2adr(port));
 }
@@ -99,19 +76,19 @@ void titan_outb_p(u8 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outb(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outb(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outb(value, pci_ioaddr(port));
         else
                 ctrl_outw(value, port2adr(port));
-        delay();
+        ctrl_delay();
 }
 
 void titan_outw(u16 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outw(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outw(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outw(value, pci_ioaddr(port));
         else if (port >= 0x2000)
                 ctrl_outw(value, port2adr(port));
         else
@@ -122,8 +99,8 @@ void titan_outl(u32 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outl(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outl(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outl(value, pci_ioaddr(port));
         else
                 maybebadio(port);
 }
@@ -140,10 +117,10 @@ void titan_outsl(unsigned long port, const void *src, unsigned long count)
 
 void __iomem *titan_ioport_map(unsigned long port, unsigned int size)
 {
-	if (PXSEG(port) || CHECK_SH7751_PCIMEMIO(port))
+	if (PXSEG(port) || is_pci_memaddr(port))
 		return (void __iomem *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return (void __iomem *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return (void __iomem *)pci_ioaddr(port);
 
 	return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/cchips/hd6446x/hd64461/io.c b/arch/sh/cchips/hd6446x/hd64461/io.c
index f77f18f2ba7f..7909a1b7b512 100644
--- a/arch/sh/cchips/hd6446x/hd64461/io.c
+++ b/arch/sh/cchips/hd6446x/hd64461/io.c
@@ -53,11 +53,6 @@ static __inline__ unsigned long PORT2ADDR(unsigned long port)
 	return 0xa0000000 + (port & 0x1fffffff);
 }
 
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 unsigned char hd64461_inb(unsigned long port)
 {
 	return *(volatile unsigned char*)PORT2ADDR(port);
@@ -66,7 +61,7 @@ unsigned char hd64461_inb(unsigned long port)
 unsigned char hd64461_inb_p(unsigned long port)
 {
 	unsigned long v = *(volatile unsigned char*)PORT2ADDR(port);
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -88,7 +83,7 @@ void hd64461_outb(unsigned char b, unsigned long port)
 void hd64461_outb_p(unsigned char b, unsigned long port)
 {
 	*(volatile unsigned char*)PORT2ADDR(port) = b;
-	delay();
+	ctrl_delay();
 }
 
 void hd64461_outw(unsigned short b, unsigned long port)
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 3d8078f1c05f..9e00cb8a39e9 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -6,8 +6,8 @@ obj-y					+= pci.o
 obj-$(CONFIG_PCI_AUTO)			+= pci-auto.o
 
 obj-$(CONFIG_CPU_SUBTYPE_ST40STB1)	+= pci-st40.o
-obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o 
-obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o ops-sh4.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o ops-sh4.o
 
 obj-$(CONFIG_SH_DREAMCAST)		+= ops-dreamcast.o fixups-dreamcast.o \
 					   dma-dreamcast.o
@@ -17,3 +17,4 @@ obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
 obj-$(CONFIG_SH_R7780RP)		+= ops-r7780rp.o fixups-r7780rp.o
 obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
+obj-$(CONFIG_SH_LANDISK)		+= ops-landisk.o
diff --git a/arch/sh/drivers/pci/fixups-dreamcast.c b/arch/sh/drivers/pci/fixups-dreamcast.c
index 63b1c6f4b8d2..c0af5f7ef414 100644
--- a/arch/sh/drivers/pci/fixups-dreamcast.c
+++ b/arch/sh/drivers/pci/fixups-dreamcast.c
@@ -4,7 +4,7 @@
  * PCI fixups for the Sega Dreamcast
  *
  * Copyright (C) 2001, 2002  M. R. Brown
- * Copyright (C) 2002, 2003  Paul Mundt
+ * Copyright (C) 2002, 2003, 2006  Paul Mundt
  *
  * This file originally bore the message (with enclosed-$):
  *	Id: pci.c,v 1.3 2003/05/04 19:29:46 lethal Exp
@@ -45,36 +45,16 @@ static void __init gapspci_fixup_resources(struct pci_dev *dev)
 		printk("PCI: Failed resource fixup\n");
 	}
 }
-
 DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, gapspci_fixup_resources);
 
-void __init pcibios_fixup_bus(struct pci_bus *bus)
+int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
 {
-	/* 
-	 * We don't have any sub bus to fix up, and this is a rather
-	 * stupid place to put general device fixups. Don't do it.
-	 * Use the pcibios_fixups table or suffer the consequences.
+	/*
+	 * The interrupt routing semantics here are quite trivial.
+	 *
+	 * We basically only support one interrupt, so we only bother
+	 * updating a device's interrupt line with this single shared
+	 * interrupt. Keeps routing quite simple, doesn't it?
 	 */
+	return GAPSPCI_IRQ;
 }
-
-void __init pcibios_fixup_irqs(void)
-{
-	struct pci_dev *dev = 0;
-
-	for_each_pci_dev(dev) {
-		/*
-		 * The interrupt routing semantics here are quite trivial.
-		 *
-		 * We basically only support one interrupt, so we only bother
-		 * updating a device's interrupt line with this single shared
-		 * interrupt. Keeps routing quite simple, doesn't it?
-		 */
-		printk(KERN_NOTICE "PCI: Fixing up IRQ routing for device %s\n",
-		       pci_name(dev));
-
-		dev->irq = GAPSPCI_IRQ;
-
-		pci_write_config_byte(dev, PCI_INTERRUPT_LINE, dev->irq);
-	}
-}
-
diff --git a/arch/sh/drivers/pci/fixups-r7780rp.c b/arch/sh/drivers/pci/fixups-r7780rp.c
index b656b562ec99..3e321df65d22 100644
--- a/arch/sh/drivers/pci/fixups-r7780rp.c
+++ b/arch/sh/drivers/pci/fixups-r7780rp.c
@@ -4,36 +4,42 @@
  * Highlander R7780RP-1 PCI fixups
  *
  * Copyright (C) 2003  Lineo uSolutions, Inc.
- * Copyright (C) 2004  Paul Mundt
+ * Copyright (C) 2004 - 2006  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-#include "pci-sh7780.h"
+#include <linux/pci.h>
+#include "pci-sh4.h"
 #include <asm/io.h>
 
 int pci_fixup_pcic(void)
 {
-	outl(0x000043ff, PCI_REG(SH7780_PCIIMR));
-	outl(0x0000380f, PCI_REG(SH7780_PCIAINTM));
+	pci_write_reg(0x000043ff, SH4_PCIINTM);
+	pci_write_reg(0x0000380f, SH4_PCIAINTM);
 
-	outl(0xfbb00047, PCI_REG(SH7780_PCICMD));
-	outl(0x00000000, PCI_REG(SH7780_PCIIBAR));
+	pci_write_reg(0xfbb00047, SH7780_PCICMD);
+	pci_write_reg(0x00000000, SH7780_PCIIBAR);
 
-	outl(0x00011912, PCI_REG(SH7780_PCISVID));
-	outl(0x08000000, PCI_REG(SH7780_PCICSCR0));
-	outl(0x0000001b, PCI_REG(SH7780_PCICSAR0));
-	outl(0xfd000000, PCI_REG(SH7780_PCICSCR1));
-	outl(0x0000000f, PCI_REG(SH7780_PCICSAR1));
+	pci_write_reg(0x00011912, SH7780_PCISVID);
+	pci_write_reg(0x08000000, SH7780_PCICSCR0);
+	pci_write_reg(0x0000001b, SH7780_PCICSAR0);
+	pci_write_reg(0xfd000000, SH7780_PCICSCR1);
+	pci_write_reg(0x0000000f, SH7780_PCICSAR1);
 
-	outl(0xfd000000, PCI_REG(SH7780_PCIMBR0));
-	outl(0x00fc0000, PCI_REG(SH7780_PCIMBMR0));
+	pci_write_reg(0xfd000000, SH7780_PCIMBR0);
+	pci_write_reg(0x00fc0000, SH7780_PCIMBMR0);
+
+#ifdef CONFIG_32BIT
+	pci_write_reg(0xc0000000, SH7780_PCIMBR2);
+	pci_write_reg(0x20000000 - SH7780_PCI_IO_SIZE, SH7780_PCIMBMR2);
+#endif
 
 	/* Set IOBR for windows containing area specified in pci.h */
-	outl((PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE-1)), PCI_REG(SH7780_PCIIOBR));
-	outl(((SH7780_PCI_IO_SIZE-1) & (7<<18)), PCI_REG(SH7780_PCIIOBMR));
+	pci_write_reg((PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE - 1)),
+		      SH7780_PCIIOBR);
+	pci_write_reg(((SH7780_PCI_IO_SIZE-1) & (7<<18)), SH7780_PCIIOBMR);
 
 	return 0;
 }
-
diff --git a/arch/sh/drivers/pci/fixups-rts7751r2d.c b/arch/sh/drivers/pci/fixups-rts7751r2d.c
index 0c590fc7a081..e72ceb560d5b 100644
--- a/arch/sh/drivers/pci/fixups-rts7751r2d.c
+++ b/arch/sh/drivers/pci/fixups-rts7751r2d.c
@@ -10,8 +10,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-#include "pci-sh7751.h"
-#include <asm/io.h>
+#include "pci-sh4.h"
 
 #define PCIMCR_MRSET_OFF	0xBFFFFFFF
 #define PCIMCR_RFSH_OFF		0xFFFFFFFB
@@ -22,22 +21,23 @@ int pci_fixup_pcic(void)
 
 	bcr1 = inl(SH7751_BCR1);
 	bcr1 |= 0x40080000;	/* Enable Bit 19 BREQEN, set PCIC to slave */
-	outl(bcr1, PCI_REG(SH7751_PCIBCR1));
+	pci_write_reg(bcr1, SH4_PCIBCR1);
 
 	/* Enable all interrupts, so we known what to fix */
-	outl(0x0000c3ff, PCI_REG(SH7751_PCIINTM));
-	outl(0x0000380f, PCI_REG(SH7751_PCIAINTM));
+	pci_write_reg(0x0000c3ff, SH4_PCIINTM);
+	pci_write_reg(0x0000380f, SH4_PCIAINTM);
 
-	outl(0xfb900047, PCI_REG(SH7751_PCICONF1));
-	outl(0xab000001, PCI_REG(SH7751_PCICONF4));
+	pci_write_reg(0xfb900047, SH7751_PCICONF1);
+	pci_write_reg(0xab000001, SH7751_PCICONF4);
 
 	mcr = inl(SH7751_MCR);
 	mcr = (mcr & PCIMCR_MRSET_OFF) & PCIMCR_RFSH_OFF;
-	outl(mcr, PCI_REG(SH7751_PCIMCR));
+	pci_write_reg(mcr, SH4_PCIMCR);
+
+	pci_write_reg(0x0c000000, SH7751_PCICONF5);
+	pci_write_reg(0xd0000000, SH7751_PCICONF6);
+	pci_write_reg(0x0c000000, SH4_PCILAR0);
+	pci_write_reg(0x00000000, SH4_PCILAR1);
 
-	outl(0x0c000000, PCI_REG(SH7751_PCICONF5));
-	outl(0xd0000000, PCI_REG(SH7751_PCICONF6));
-	outl(0x0c000000, PCI_REG(SH7751_PCILAR0));
-	outl(0x00000000, PCI_REG(SH7751_PCILAR1));
 	return 0;
 }
diff --git a/arch/sh/drivers/pci/fixups-sh03.c b/arch/sh/drivers/pci/fixups-sh03.c
index 57ac26c2171f..2e8a18b7ee53 100644
--- a/arch/sh/drivers/pci/fixups-sh03.c
+++ b/arch/sh/drivers/pci/fixups-sh03.c
@@ -3,11 +3,7 @@
 #include <linux/types.h>
 #include <linux/pci.h>
 
-/*
- * 	IRQ functions
- */
-
-int __init pcibios_map_platform_irq(u8 slot, u8 pin, struct pci_dev *dev)
+int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
 {
 	int irq;
 
@@ -17,8 +13,9 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin, struct pci_dev *dev)
 		case 8: return 5;	/* eth1       */
 		case 6: return 2;	/* PCI bridge */
 		default:
-                	printk("PCI: Bad IRQ mapping request for slot %d\n", slot);
-                	return 2;
+			printk(KERN_ERR "PCI: Bad IRQ mapping request "
+					"for slot %d\n", slot);
+			return 2;
 		}
 	} else {
 		switch (pin) {
@@ -32,30 +29,3 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin, struct pci_dev *dev)
 	}
 	return irq;
 }
-
-static u8 __init sh03_no_swizzle(struct pci_dev *dev, u8 *pin)
-{
-	/* no swizzling */
-	return PCI_SLOT(dev->devfn);
-}
-
-static int sh03_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int irq = -1;
-
-	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
-	irq = pcibios_map_platform_irq(slot, pin, dev);
-	if( irq < 0 ) {
-		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
-		return irq;
-	}
-
-	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
-
-	return irq;
-}
-
-void __init pcibios_fixup_irqs(void)
-{
-	pci_fixup_irqs(sh03_no_swizzle, sh03_pci_lookup_irq);
-}
diff --git a/arch/sh/drivers/pci/ops-bigsur.c b/arch/sh/drivers/pci/ops-bigsur.c
index ae82c6ca05e5..5da501bd77b5 100644
--- a/arch/sh/drivers/pci/ops-bigsur.c
+++ b/arch/sh/drivers/pci/ops-bigsur.c
@@ -10,15 +10,12 @@
  *
  * PCI initialization for the Hitachi Big Sur Evaluation Board
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/delay.h>
 #include <linux/pci.h>
-
 #include <asm/io.h>
-#include "pci-sh7751.h"
+#include "pci-sh4.h"
 #include <asm/bigsur/bigsur.h>
 
 #define BIGSUR_PCI_IO	0x4000
@@ -41,11 +38,11 @@ static struct resource sh7751_mem_resource = {
 extern struct pci_ops sh7751_pci_ops;
 
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ 0, }
 };
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS3_BASE_ADDR,
 		.size	= BIGSUR_LSR0_SIZE,
@@ -58,7 +55,7 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 };
 
 /*
- * Initialize the Big Sur PCI interface 
+ * Initialize the Big Sur PCI interface
  * Setup hardware to be Central Funtion
  * Copy the BSR regs to the PCI interface
  * Setup PCI windows into local RAM
@@ -68,15 +65,15 @@ int __init pcibios_init_platform(void)
 	return sh7751_pcic_init(&sh7751_pci_map);
 }
 
-int pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
-	/* 
+	/*
 	 * The Big Sur can be used in a CPCI chassis, but the SH7751 PCI
 	 * interface is on the wrong end of the board so that it can also
 	 * support a V320 CPI interface chip...  Therefor the IRQ mapping is
 	 * somewhat use dependent... I'l assume a linear map for now, i.e.
 	 * INTA=slot0,pin0... INTD=slot3,pin0...
-	 */ 
+	 */
 	int irq = (slot + pin-1) % 4 + BIGSUR_SH7751_PCI_IRQ_BASE;
 
 	PCIDBG(2, "PCI: Mapping Big Sur IRQ for slot %d, pin %c to irq %d\n",
@@ -84,4 +81,3 @@ int pcibios_map_platform_irq(u8 slot, u8 pin)
 
 	return irq;
 }
-
diff --git a/arch/sh/drivers/pci/ops-landisk.c b/arch/sh/drivers/pci/ops-landisk.c
new file mode 100644
index 000000000000..ada301c21fe7
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-landisk.c
@@ -0,0 +1,68 @@
+/*
+ * arch/sh/drivers/pci/ops-landisk.c
+ *
+ * PCI initialization for the I-O DATA Device, Inc. LANDISK board
+ *
+ * Copyright (C) 2006 kogiidena
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ */
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include "pci-sh4.h"
+
+static struct resource sh7751_io_resource = {
+	.name = "SH7751 IO",
+	.start = 0x4000,
+	.end = 0x4000 + SH7751_PCI_IO_SIZE - 1,
+	.flags = IORESOURCE_IO
+};
+
+static struct resource sh7751_mem_resource = {
+	.name = "SH7751 mem",
+	.start = SH7751_PCI_MEMORY_BASE,
+	.end = SH7751_PCI_MEMORY_BASE + SH7751_PCI_MEM_SIZE - 1,
+	.flags = IORESOURCE_MEM
+};
+
+struct pci_channel board_pci_channels[] = {
+	{&sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0x3ff},
+	{NULL, NULL, NULL, 0, 0},
+};
+
+static struct sh4_pci_address_map sh7751_pci_map = {
+	.window0 = {
+		.base	= SH7751_CS3_BASE_ADDR,
+		.size	= (64 << 20),	/* 64MB */
+	},
+
+	.flags = SH4_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	return sh7751_pcic_init(&sh7751_pci_map);
+}
+
+int pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
+{
+	/*
+	 * slot0: pin1-4 = irq5,6,7,8
+	 * slot1: pin1-4 = irq6,7,8,5
+	 * slot2: pin1-4 = irq7,8,5,6
+	 * slot3: pin1-4 = irq8,5,6,7
+	 */
+	int irq = ((slot + pin - 1) & 0x3) + 5;
+
+	if ((slot | (pin - 1)) > 0x3) {
+		printk("PCI: Bad IRQ mapping request for slot %d pin %c\n",
+		       slot, pin - 1 + 'A');
+		return -1;
+	}
+	return irq;
+}
diff --git a/arch/sh/drivers/pci/ops-r7780rp.c b/arch/sh/drivers/pci/ops-r7780rp.c
index 3254c4e917a9..554d5ed2c586 100644
--- a/arch/sh/drivers/pci/ops-r7780rp.c
+++ b/arch/sh/drivers/pci/ops-r7780rp.c
@@ -15,13 +15,11 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/pci.h>
-#include <linux/module.h>
-
-#include <asm/io.h>
-#include "pci-sh7780.h"
 #include <asm/r7780rp/r7780rp.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
         switch (slot) {
 	case 0: return IRQ_PCISLOT1;		/* PCI Interrupt #1 */
@@ -29,7 +27,8 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin)
 	case 2: return IRQ_PCISLOT3;		/* PCI Interrupt #3 */
 	case 3: return IRQ_PCISLOT4;		/* PCI Interrupt E4 */
 	default:
-		printk("PCI: Bad IRQ mapping request for slot %d, func %d\n", slot, pin-1);
+		printk(KERN_ERR "PCI: Bad IRQ mapping "
+		       "request for slot %d, func %d\n", slot, pin-1);
 		return -1;
 	}
 }
@@ -51,12 +50,12 @@ static struct resource sh7780_mem_resource = {
 extern struct pci_ops sh7780_pci_ops;
 
 struct pci_channel board_pci_channels[] = {
-	{ &sh7780_pci_ops, &sh7780_io_resource, &sh7780_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7780_io_resource, &sh7780_mem_resource, 0, 0xff },
 	{ NULL, NULL, NULL, 0, 0 },
 };
 EXPORT_SYMBOL(board_pci_channels);
 
-static struct sh7780_pci_address_map sh7780_pci_map = {
+static struct sh4_pci_address_map sh7780_pci_map = {
 	.window0	= {
 		.base	= SH7780_CS2_BASE_ADDR,
 		.size	= 0x04000000,
@@ -67,11 +66,10 @@ static struct sh7780_pci_address_map sh7780_pci_map = {
 		.size	= 0x04000000,
 	},
 
-	.flags	= SH7780_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 int __init pcibios_init_platform(void)
 {
 	return sh7780_pcic_init(&sh7780_pci_map);
 }
-
diff --git a/arch/sh/drivers/pci/ops-rts7751r2d.c b/arch/sh/drivers/pci/ops-rts7751r2d.c
index 83171d10141a..88f44e245424 100644
--- a/arch/sh/drivers/pci/ops-rts7751r2d.c
+++ b/arch/sh/drivers/pci/ops-rts7751r2d.c
@@ -17,12 +17,11 @@
 #include <linux/delay.h>
 #include <linux/pci.h>
 #include <linux/module.h>
-
-#include <asm/io.h>
-#include "pci-sh7751.h"
 #include <asm/rts7751r2d/rts7751r2d.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
         switch (slot) {
 	case 0: return IRQ_PCISLOT1;	/* PCI Extend slot #1 */
@@ -52,12 +51,12 @@ static struct resource sh7751_mem_resource = {
 extern struct pci_ops sh7751_pci_ops;
 
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ NULL, NULL, NULL, 0, 0 },
 };
 EXPORT_SYMBOL(board_pci_channels);
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS3_BASE_ADDR,
 		.size	= 0x04000000,
@@ -68,7 +67,7 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 		.size	= 0x00000000,	/* Unused */
 	},
 
-	.flags	= SH7751_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 int __init pcibios_init_platform(void)
diff --git a/arch/sh/drivers/pci/ops-sh4.c b/arch/sh/drivers/pci/ops-sh4.c
new file mode 100644
index 000000000000..2d4371009a5e
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-sh4.c
@@ -0,0 +1,164 @@
+/*
+ * Generic SH-4 / SH-4A PCIC operations (SH7751, SH7780).
+ *
+ * Copyright (C) 2002 - 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License v2. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/pci.h>
+#include <asm/addrspace.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
+
+/*
+ * Direct access to PCI hardware...
+ */
+#define CONFIG_CMD(bus, devfn, where) \
+	P1SEGADDR((bus->number << 16) | (devfn << 8) | (where & ~3))
+
+static DEFINE_SPINLOCK(sh4_pci_lock);
+
+/*
+ * Functions for accessing PCI configuration space with type 1 accesses
+ */
+static int sh4_pci_read(struct pci_bus *bus, unsigned int devfn,
+			   int where, int size, u32 *val)
+{
+	unsigned long flags;
+	u32 data;
+
+	/*
+	 * PCIPDR may only be accessed as 32 bit words,
+	 * so we must do byte alignment by hand
+	 */
+	spin_lock_irqsave(&sh4_pci_lock, flags);
+	pci_write_reg(CONFIG_CMD(bus, devfn, where), SH4_PCIPAR);
+	data = pci_read_reg(SH4_PCIPDR);
+	spin_unlock_irqrestore(&sh4_pci_lock, flags);
+
+	switch (size) {
+	case 1:
+		*val = (data >> ((where & 3) << 3)) & 0xff;
+		break;
+	case 2:
+		*val = (data >> ((where & 2) << 3)) & 0xffff;
+		break;
+	case 4:
+		*val = data;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+/*
+ * Since SH4 only does 32bit access we'll have to do a read,
+ * mask,write operation.
+ * We'll allow an odd byte offset, though it should be illegal.
+ */
+static int sh4_pci_write(struct pci_bus *bus, unsigned int devfn,
+			 int where, int size, u32 val)
+{
+	unsigned long flags;
+	int shift;
+	u32 data;
+
+	spin_lock_irqsave(&sh4_pci_lock, flags);
+	pci_write_reg(CONFIG_CMD(bus, devfn, where), SH4_PCIPAR);
+	data = pci_read_reg(SH4_PCIPDR);
+	spin_unlock_irqrestore(&sh4_pci_lock, flags);
+
+	switch (size) {
+	case 1:
+		shift = (where & 3) << 3;
+		data &= ~(0xff << shift);
+		data |= ((val & 0xff) << shift);
+		break;
+	case 2:
+		shift = (where & 2) << 3;
+		data &= ~(0xffff << shift);
+		data |= ((val & 0xffff) << shift);
+		break;
+	case 4:
+		data = val;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	pci_write_reg(data, SH4_PCIPDR);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+struct pci_ops sh4_pci_ops = {
+	.read		= sh4_pci_read,
+	.write		= sh4_pci_write,
+};
+
+/*
+ * Not really related to pci_ops, but it's common and not worth shoving
+ * somewhere else for now..
+ */
+static unsigned int pci_probe = PCI_PROBE_CONF1;
+
+int __init sh4_pci_check_direct(void)
+{
+	/*
+	 * Check if configuration works.
+	 */
+	if (pci_probe & PCI_PROBE_CONF1) {
+		unsigned int tmp = pci_read_reg(SH4_PCIPAR);
+
+		pci_write_reg(P1SEG, SH4_PCIPAR);
+
+		if (pci_read_reg(SH4_PCIPAR) == P1SEG) {
+			pci_write_reg(tmp, SH4_PCIPAR);
+			printk(KERN_INFO "PCI: Using configuration type 1\n");
+			request_region(PCI_REG(SH4_PCIPAR), 8, "PCI conf1");
+
+			return 0;
+		}
+
+		pci_write_reg(tmp, SH4_PCIPAR);
+	}
+
+	pr_debug("PCI: pci_check_direct failed\n");
+	return -EINVAL;
+}
+
+/* Handle generic fixups */
+static void __init pci_fixup_ide_bases(struct pci_dev *d)
+{
+	int i;
+
+	/*
+	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
+	 */
+	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return;
+	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
+	for(i = 0; i < 4; i++) {
+		struct resource *r = &d->resource[i];
+
+		if ((r->start & ~0x80) == 0x374) {
+			r->start |= 2;
+			r->end = r->start;
+		}
+	}
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+
+char * __init pcibios_setup(char *str)
+{
+	if (!strcmp(str, "off")) {
+		pci_probe = 0;
+		return NULL;
+	}
+
+	return str;
+}
diff --git a/arch/sh/drivers/pci/ops-snapgear.c b/arch/sh/drivers/pci/ops-snapgear.c
index 3cbd14dd28fe..53dd893d4e54 100644
--- a/arch/sh/drivers/pci/ops-snapgear.c
+++ b/arch/sh/drivers/pci/ops-snapgear.c
@@ -2,7 +2,7 @@
  * arch/sh/drivers/pci/ops-snapgear.c
  *
  * Author:  David McCullough <davidm@snapgear.com>
- * 
+ *
  * Ported to new API by Paul Mundt <lethal@linux-sh.org>
  *
  * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
@@ -12,15 +12,11 @@
  *
  * PCI initialization for the SnapGear boards
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/delay.h>
 #include <linux/pci.h>
-
-#include <asm/io.h>
-#include "pci-sh7751.h"
+#include "pci-sh4.h"
 
 #define SNAPGEAR_PCI_IO		0x4000
 #define SNAPGEAR_PCI_MEM	0xfd000000
@@ -43,14 +39,12 @@ static struct resource sh7751_mem_resource = {
 	.flags		= IORESOURCE_MEM,
 };
 
-extern struct pci_ops sh7751_pci_ops;
-
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ 0, }
 };
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS2_BASE_ADDR,
 		.size	= SNAPGEAR_LSR0_SIZE,
@@ -61,11 +55,11 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 		.size	= SNAPGEAR_LSR1_SIZE,
 	},
 
-	.flags	= SH7751_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 /*
- * Initialize the SnapGear PCI interface 
+ * Initialize the SnapGear PCI interface
  * Setup hardware to be Central Funtion
  * Copy the BSR regs to the PCI interface
  * Setup PCI windows into local RAM
@@ -75,7 +69,7 @@ int __init pcibios_init_platform(void)
 	return sh7751_pcic_init(&sh7751_pci_map);
 }
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
 	int irq = -1;
 
@@ -98,4 +92,3 @@ void __init pcibios_fixup(void)
 {
 	/* Nothing to fixup .. */
 }
-
diff --git a/arch/sh/drivers/pci/ops-titan.c b/arch/sh/drivers/pci/ops-titan.c
index f76e4e963ac1..9c8b2027c35d 100644
--- a/arch/sh/drivers/pci/ops-titan.c
+++ b/arch/sh/drivers/pci/ops-titan.c
@@ -16,12 +16,11 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/delay.h>
 #include <asm/io.h>
 #include <asm/titan.h>
-#include "pci-sh7751.h"
+#include "pci-sh4.h"
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
 	int irq = -1;
 
@@ -32,7 +31,8 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin)
 	case 3: irq = TITAN_IRQ_MPCIB; break;	/* mPCI B */
 	case 4: irq = TITAN_IRQ_USB;   break;	/* USB */
 	default:
-		printk(KERN_INFO "PCI: Bad IRQ mapping request for slot %d\n", slot);
+		printk(KERN_INFO "PCI: Bad IRQ mapping "
+				 "request for slot %d\n", slot);
 		return -1;
 	}
 
@@ -56,15 +56,13 @@ static struct resource sh7751_mem_resource = {
 	.flags	= IORESOURCE_MEM
 };
 
-extern struct pci_ops sh7751_pci_ops;
-
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ NULL, NULL, NULL, 0, 0 },
 };
 EXPORT_SYMBOL(board_pci_channels);
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS2_BASE_ADDR,
 		.size	= SH7751_MEM_REGION_SIZE*2,	/* cs2 and cs3 */
@@ -75,7 +73,7 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 		.size	= SH7751_MEM_REGION_SIZE*2,
 	},
 
-	.flags	= SH7751_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 int __init pcibios_init_platform(void)
diff --git a/arch/sh/drivers/pci/pci-auto.c b/arch/sh/drivers/pci/pci-auto.c
index d55e46618549..ecf16344f94a 100644
--- a/arch/sh/drivers/pci/pci-auto.c
+++ b/arch/sh/drivers/pci/pci-auto.c
@@ -358,7 +358,6 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 {
 	u32 temp;
 
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D) && !defined(CONFIG_SH_R7780RP)
 	/*
 	 * [jsun] we always bump up baselines a little, so that if there
 	 * nothing behind P2P bridge, we don't wind up overlapping IO/MEM
@@ -366,7 +365,6 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 	 */
 	pciauto_lower_memspc += 1;
 	pciauto_lower_iospc += 1;
-#endif
 
 	/*
 	 * Configure subordinate bus number.  The PCI subsystem
@@ -392,11 +390,6 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 	 * configured by this routine to happily live behind a
 	 * P2P bridge in a system.
 	 */
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
-	pciauto_lower_memspc += 0x00400000;
-	pciauto_lower_iospc += 0x00004000;
-#endif
-
 	/* Align memory and I/O to 4KB and 4 byte boundaries. */
 	pciauto_lower_memspc = (pciauto_lower_memspc + (0x1000 - 1))
 		& ~(0x1000 - 1);
@@ -467,9 +460,6 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 		if ((pci_class >> 16) == PCI_CLASS_BRIDGE_PCI) {
 			DBG("        Bridge: primary=%.2x, secondary=%.2x\n",
 				current_bus, sub_bus + 1);
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
-			pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_1);
-#endif
 			pciauto_prescan_setup_bridge(hose, top_bus, current_bus,
 						     pci_devfn, sub_bus);
 			DBG("Scanning sub bus %.2x, I/O 0x%.8x, Mem 0x%.8x\n",
diff --git a/arch/sh/drivers/pci/pci-sh4.h b/arch/sh/drivers/pci/pci-sh4.h
new file mode 100644
index 000000000000..5a61d6041f2c
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh4.h
@@ -0,0 +1,180 @@
+#ifndef __PCI_SH4_H
+#define __PCI_SH4_H
+
+#ifdef CONFIG_CPU_SUBTYPE_SH7780
+#include "pci-sh7780.h"
+#else
+#include "pci-sh7751.h"
+#endif
+
+#include <asm/io.h>
+
+/* startup values */
+#define PCI_PROBE_BIOS		1
+#define PCI_PROBE_CONF1		2
+#define PCI_PROBE_CONF2		4
+#define PCI_NO_SORT		0x100
+#define PCI_BIOS_SORT		0x200
+#define PCI_NO_CHECKS		0x400
+#define PCI_ASSIGN_ROMS		0x1000
+#define PCI_BIOS_IRQ_SCAN	0x2000
+
+#define SH4_PCICR		0x100		/* PCI Control Register */
+  #define SH4_PCICR_PREFIX	  0xA5000000	/* CR prefix for write */
+  #define SH4_PCICR_FTO		  0x00000400	/* TRDY/IRDY Enable */
+  #define SH4_PCICR_TRSB	  0x00000200	/* Target Read Single */
+  #define SH4_PCICR_BSWP	  0x00000100	/* Target Byte Swap */
+  #define SH4_PCICR_PLUP	  0x00000080	/* Enable PCI Pullup */
+  #define SH4_PCICR_ARBM	  0x00000040	/* PCI Arbitration Mode */
+  #define SH4_PCICR_MD		  0x00000030	/* MD9 and MD10 status */
+  #define SH4_PCICR_SERR	  0x00000008	/* SERR output assert */
+  #define SH4_PCICR_INTA	  0x00000004	/* INTA output assert */
+  #define SH4_PCICR_PRST	  0x00000002	/* PCI Reset Assert */
+  #define SH4_PCICR_CFIN	  0x00000001	/* Central Fun. Init Done */
+#define SH4_PCILSR0		0x104		/* PCI Local Space Register0 */
+#define SH4_PCILSR1		0x108		/* PCI Local Space Register1 */
+#define SH4_PCILAR0		0x10C		/* PCI Local Addr Register1 */
+#define SH4_PCILAR1		0x110		/* PCI Local Addr Register1 */
+#define SH4_PCIINT		0x114		/* PCI Interrupt Register */
+  #define SH4_PCIINT_MLCK	  0x00008000	/* Master Lock Error */
+  #define SH4_PCIINT_TABT	  0x00004000	/* Target Abort Error */
+  #define SH4_PCIINT_TRET	  0x00000200	/* Target Retry Error */
+  #define SH4_PCIINT_MFDE	  0x00000100	/* Master Func. Disable Error */
+  #define SH4_PCIINT_PRTY	  0x00000080	/* Address Parity Error */
+  #define SH4_PCIINT_SERR	  0x00000040	/* SERR Detection Error */
+  #define SH4_PCIINT_TWDP	  0x00000020	/* Tgt. Write Parity Error */
+  #define SH4_PCIINT_TRDP	  0x00000010	/* Tgt. Read Parity Err Det. */
+  #define SH4_PCIINT_MTABT	  0x00000008	/* Master-Tgt. Abort Error */
+  #define SH4_PCIINT_MMABT	  0x00000004	/* Master-Master Abort Error */
+  #define SH4_PCIINT_MWPD	  0x00000002	/* Master Write PERR Detect */
+  #define SH4_PCIINT_MRPD	  0x00000001	/* Master Read PERR Detect */
+#define SH4_PCIINTM		0x118		/* PCI Interrupt Mask */
+#define SH4_PCIALR		0x11C		/* Error Address Register */
+#define SH4_PCICLR		0x120		/* Error Command/Data */
+  #define SH4_PCICLR_MPIO	  0x80000000
+  #define SH4_PCICLR_MDMA0	  0x40000000	/* DMA0 Transfer Error */
+  #define SH4_PCICLR_MDMA1	  0x20000000	/* DMA1 Transfer Error */
+  #define SH4_PCICLR_MDMA2	  0x10000000	/* DMA2 Transfer Error */
+  #define SH4_PCICLR_MDMA3	  0x08000000	/* DMA3 Transfer Error */
+  #define SH4_PCICLR_TGT	  0x04000000	/* Target Transfer Error */
+  #define SH4_PCICLR_CMDL	  0x0000000F	/* PCI Command at Error */
+#define SH4_PCIAINT		0x130		/* Arbiter Interrupt Register */
+  #define SH4_PCIAINT_MBKN	  0x00002000	/* Master Broken Interrupt */
+  #define SH4_PCIAINT_TBTO	  0x00001000	/* Target Bus Time Out */
+  #define SH4_PCIAINT_MBTO	  0x00001000	/* Master Bus Time Out */
+  #define SH4_PCIAINT_TABT	  0x00000008	/* Target Abort */
+  #define SH4_PCIAINT_MABT	  0x00000004	/* Master Abort */
+  #define SH4_PCIAINT_RDPE	  0x00000002	/* Read Data Parity Error */
+  #define SH4_PCIAINT_WDPE	  0x00000001	/* Write Data Parity Error */
+#define SH4_PCIAINTM            0x134		/* Arbiter Int. Mask Register */
+#define SH4_PCIBMLR		0x138		/* Error Bus Master Register */
+  #define SH4_PCIBMLR_REQ4	  0x00000010	/* REQ4 bus master at error */
+  #define SH4_PCIBMLR_REQ3	  0x00000008	/* REQ3 bus master at error */
+  #define SH4_PCIBMLR_REQ2	  0x00000004	/* REQ2 bus master at error */
+  #define SH4_PCIBMLR_REQ1	  0x00000002	/* REQ1 bus master at error */
+  #define SH4_PCIBMLR_REQ0	  0x00000001	/* REQ0 bus master at error */
+#define SH4_PCIDMABT		0x140		/* DMA Transfer Arb. Register */
+  #define SH4_PCIDMABT_RRBN	  0x00000001	/* DMA Arbitor Round-Robin */
+#define SH4_PCIDPA0		0x180		/* DMA0 Transfer Addr. */
+#define SH4_PCIDLA0		0x184		/* DMA0 Local Addr. */
+#define SH4_PCIDTC0		0x188		/* DMA0 Transfer Cnt. */
+#define SH4_PCIDCR0		0x18C		/* DMA0 Control Register */
+  #define SH4_PCIDCR_ALGN	  0x00000600	/* DMA Alignment Mode */
+  #define SH4_PCIDCR_MAST	  0x00000100	/* DMA Termination Type */
+  #define SH4_PCIDCR_INTM	  0x00000080	/* DMA Interrupt Done Mask*/
+  #define SH4_PCIDCR_INTS	  0x00000040	/* DMA Interrupt Done Status */
+  #define SH4_PCIDCR_LHLD	  0x00000020	/* Local Address Control */
+  #define SH4_PCIDCR_PHLD	  0x00000010	/* PCI Address Control*/
+  #define SH4_PCIDCR_IOSEL	  0x00000008	/* PCI Address Space Type */
+  #define SH4_PCIDCR_DIR	  0x00000004	/* DMA Transfer Direction */
+  #define SH4_PCIDCR_STOP	  0x00000002	/* Force DMA Stop */
+  #define SH4_PCIDCR_STRT	  0x00000001	/* DMA Start */
+#define SH4_PCIDPA1		0x190		/* DMA1 Transfer Addr. */
+#define SH4_PCIDLA1		0x194		/* DMA1 Local Addr. */
+#define SH4_PCIDTC1		0x198		/* DMA1 Transfer Cnt. */
+#define SH4_PCIDCR1		0x19C		/* DMA1 Control Register */
+#define SH4_PCIDPA2		0x1A0		/* DMA2 Transfer Addr. */
+#define SH4_PCIDLA2		0x1A4		/* DMA2 Local Addr. */
+#define SH4_PCIDTC2		0x1A8		/* DMA2 Transfer Cnt. */
+#define SH4_PCIDCR2		0x1AC		/* DMA2 Control Register */
+#define SH4_PCIDPA3		0x1B0		/* DMA3 Transfer Addr. */
+#define SH4_PCIDLA3		0x1B4		/* DMA3 Local Addr. */
+#define SH4_PCIDTC3		0x1B8		/* DMA3 Transfer Cnt. */
+#define SH4_PCIDCR3		0x1BC		/* DMA3 Control Register */
+#define SH4_PCIPAR		0x1C0		/* PIO Address Register */
+  #define SH4_PCIPAR_CFGEN	  0x80000000	/* Configuration Enable */
+  #define SH4_PCIPAR_BUSNO	  0x00FF0000	/* Config. Bus Number */
+  #define SH4_PCIPAR_DEVNO	  0x0000FF00	/* Config. Device Number */
+  #define SH4_PCIPAR_REGAD	  0x000000FC	/* Register Address Number */
+#define SH4_PCIMBR		0x1C4		/* Memory Base Address */
+  #define SH4_PCIMBR_MASK	  0xFF000000	/* Memory Space Mask */
+  #define SH4_PCIMBR_LOCK	  0x00000001	/* Lock Memory Space */
+#define SH4_PCIIOBR		0x1C8		/* I/O Base Address Register */
+  #define SH4_PCIIOBR_MASK	  0xFFFC0000	/* IO Space Mask */
+  #define SH4_PCIIOBR_LOCK	  0x00000001	/* Lock IO Space */
+#define SH4_PCIPINT		0x1CC		/* Power Mgmnt Int. Register */
+  #define SH4_PCIPINT_D3	  0x00000002	/* D3 Pwr Mgmt. Interrupt */
+  #define SH4_PCIPINT_D0	  0x00000001	/* D0 Pwr Mgmt. Interrupt */
+#define SH4_PCIPINTM		0x1D0		/* Power Mgmnt Mask Register */
+#define SH4_PCICLKR		0x1D4		/* Clock Ctrl. Register */
+  #define SH4_PCICLKR_PCSTP	  0x00000002	/* PCI Clock Stop */
+  #define SH4_PCICLKR_BCSTP	  0x00000001	/* BCLK Clock Stop */
+/* For definitions of BCR, MCR see ... */
+#define SH4_PCIBCR1		0x1E0		/* Memory BCR1 Register */
+  #define SH4_PCIMBR0		SH4_PCIBCR1
+#define SH4_PCIBCR2		0x1E4		/* Memory BCR2 Register */
+  #define SH4_PCIMBMR0		SH4_PCIBCR2
+#define SH4_PCIWCR1		0x1E8		/* Wait Control 1 Register */
+#define SH4_PCIWCR2		0x1EC		/* Wait Control 2 Register */
+#define SH4_PCIWCR3		0x1F0		/* Wait Control 3 Register */
+  #define SH4_PCIMBR2		SH4_PCIWCR3
+#define SH4_PCIMCR		0x1F4		/* Memory Control Register */
+#define SH4_PCIBCR3		0x1f8		/* Memory BCR3 Register */
+#define SH4_PCIPCTR             0x200		/* Port Control Register */
+  #define SH4_PCIPCTR_P2EN	  0x000400000	/* Port 2 Enable */
+  #define SH4_PCIPCTR_P1EN	  0x000200000	/* Port 1 Enable */
+  #define SH4_PCIPCTR_P0EN	  0x000100000	/* Port 0 Enable */
+  #define SH4_PCIPCTR_P2UP	  0x000000020	/* Port2 Pull Up Enable */
+  #define SH4_PCIPCTR_P2IO	  0x000000010	/* Port2 Output Enable */
+  #define SH4_PCIPCTR_P1UP	  0x000000008	/* Port1 Pull Up Enable */
+  #define SH4_PCIPCTR_P1IO	  0x000000004	/* Port1 Output Enable */
+  #define SH4_PCIPCTR_P0UP	  0x000000002	/* Port0 Pull Up Enable */
+  #define SH4_PCIPCTR_P0IO	  0x000000001	/* Port0 Output Enable */
+#define SH4_PCIPDTR		0x204		/* Port Data Register */
+  #define SH4_PCIPDTR_PB5	  0x000000020	/* Port 5 Enable */
+  #define SH4_PCIPDTR_PB4	  0x000000010	/* Port 4 Enable */
+  #define SH4_PCIPDTR_PB3	  0x000000008	/* Port 3 Enable */
+  #define SH4_PCIPDTR_PB2	  0x000000004	/* Port 2 Enable */
+  #define SH4_PCIPDTR_PB1	  0x000000002	/* Port 1 Enable */
+  #define SH4_PCIPDTR_PB0	  0x000000001	/* Port 0 Enable */
+#define SH4_PCIPDR		0x220		/* Port IO Data Register */
+
+/* Flags */
+#define SH4_PCIC_NO_RESET	0x0001
+
+/* arch/sh/kernel/drivers/pci/ops-sh4.c */
+extern struct pci_ops sh4_pci_ops;
+int sh4_pci_check_direct(void);
+int pci_fixup_pcic(void);
+
+struct sh4_pci_address_space {
+	unsigned long base;
+	unsigned long size;
+};
+
+struct sh4_pci_address_map {
+	struct sh4_pci_address_space window0;
+	struct sh4_pci_address_space window1;
+	unsigned long flags;
+};
+
+static inline void pci_write_reg(unsigned long val, unsigned long reg)
+{
+	outl(val, PCI_REG(reg));
+}
+
+static inline unsigned long pci_read_reg(unsigned long reg)
+{
+	return inl(PCI_REG(reg));
+}
+#endif /* __PCI_SH4_H */
diff --git a/arch/sh/drivers/pci/pci-sh7751.c b/arch/sh/drivers/pci/pci-sh7751.c
index 65093ec1b55e..dbe837884983 100644
--- a/arch/sh/drivers/pci/pci-sh7751.c
+++ b/arch/sh/drivers/pci/pci-sh7751.c
@@ -15,180 +15,14 @@
 
 #undef DEBUG
 
-#include <linux/types.h>
-#include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/pci.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
+#include <linux/types.h>
 #include <linux/errno.h>
-#include <linux/irq.h>
 #include <linux/delay.h>
-
-#include <asm/machvec.h>
+#include "pci-sh4.h"
+#include <asm/addrspace.h>
 #include <asm/io.h>
-#include "pci-sh7751.h"
-
-static unsigned int pci_probe = PCI_PROBE_CONF1;
-extern int pci_fixup_pcic(void);
-
-void pcibios_fixup_irqs(void) __attribute__ ((weak));
-
-/*
- * Direct access to PCI hardware...
- */
-
-#define CONFIG_CMD(bus, devfn, where) (0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
-
-/*
- * Functions for accessing PCI configuration space with type 1 accesses
- */
-static int sh7751_pci_read(struct pci_bus *bus, unsigned int devfn,
-			   int where, int size, u32 *val)
-{
-	unsigned long flags;
-	u32 data;
-
-	/* 
-	 * PCIPDR may only be accessed as 32 bit words, 
-	 * so we must do byte alignment by hand 
-	 */
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus,devfn,where), PCI_REG(SH7751_PCIPAR));
-	data = inl(PCI_REG(SH7751_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		*val = (data >> ((where & 3) << 3)) & 0xff;
-		break;
-	case 2:
-		*val = (data >> ((where & 2) << 3)) & 0xffff;
-		break;
-	case 4:
-		*val = data;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-/* 
- * Since SH7751 only does 32bit access we'll have to do a read,
- * mask,write operation.
- * We'll allow an odd byte offset, though it should be illegal.
- */ 
-static int sh7751_pci_write(struct pci_bus *bus, unsigned int devfn,
-			    int where, int size, u32 val)
-{
-	unsigned long flags;
-	int shift;
-	u32 data;
-
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus,devfn,where), PCI_REG(SH7751_PCIPAR));
-	data = inl(PCI_REG(SH7751_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		shift = (where & 3) << 3;
-		data &= ~(0xff << shift);
-		data |= ((val & 0xff) << shift);
-		break;
-	case 2:
-		shift = (where & 2) << 3;
-		data &= ~(0xffff << shift);
-		data |= ((val & 0xffff) << shift);
-		break;
-	case 4:
-		data = val;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	outl(data, PCI_REG(SH7751_PCIPDR));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-#undef CONFIG_CMD
-
-struct pci_ops sh7751_pci_ops = {
-	.read 		= sh7751_pci_read,
-	.write		= sh7751_pci_write,
-};
-
-static int __init pci_check_direct(void)
-{
-	unsigned int tmp, id;
-
-	/* check for SH7751/SH7751R hardware */
-	id = inl(SH7751_PCIREG_BASE+SH7751_PCICONF0);
-	if (id != ((SH7751_DEVICE_ID << 16) | SH7751_VENDOR_ID) &&
-	    id != ((SH7751R_DEVICE_ID << 16) | SH7751_VENDOR_ID)) {
-		pr_debug("PCI: This is not an SH7751(R) (%x)\n", id);
-		return -ENODEV;
-	}
-
-	/*
-	 * Check if configuration works.
-	 */
-	if (pci_probe & PCI_PROBE_CONF1) {
-		tmp = inl (PCI_REG(SH7751_PCIPAR));
-		outl (0x80000000, PCI_REG(SH7751_PCIPAR));
-		if (inl (PCI_REG(SH7751_PCIPAR)) == 0x80000000) {
-			outl (tmp, PCI_REG(SH7751_PCIPAR));
-			printk(KERN_INFO "PCI: Using configuration type 1\n");
-			request_region(PCI_REG(SH7751_PCIPAR), 8, "PCI conf1");
-			return 0;
-		}
-		outl (tmp, PCI_REG(SH7751_PCIPAR));
-	}
-
-	pr_debug("PCI: pci_check_direct failed\n");
-	return -EINVAL;
-}
-
-/***************************************************************************************/
-
-/*
- *  Handle bus scanning and fixups ....
- */
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
-
-/*
- *  Called after each bus is probed, but before its children
- *  are examined.
- */
-
-void __init pcibios_fixup_bus(struct pci_bus *b)
-{
-	pci_read_bridge_bases(b);
-}
 
 /*
  * Initialization. Try all known PCI access methods. Note that we support
@@ -196,25 +30,29 @@ void __init pcibios_fixup_bus(struct pci_bus *b)
  * to access config space.
  * 
  * Note that the platform specific initialization (BSC registers, and memory
- * space mapping) will be called via the machine vectors (sh_mv.mv_pci_init()) if it
- * exitst and via the platform defined function pcibios_init_platform().  
- * See pci_bigsur.c for implementation;
- * 
- * The BIOS version of the pci functions is not yet implemented but it is left
- * in for completeness.  Currently an error will be genereated at compile time. 
+ * space mapping) will be called via the platform defined function
+ * pcibios_init_platform().
  */
-
 static int __init sh7751_pci_init(void)
 {
+	unsigned int id;
 	int ret;
 
 	pr_debug("PCI: Starting intialization.\n");
-	if ((ret = pci_check_direct()) != 0)
+
+	/* check for SH7751/SH7751R hardware */
+	id = pci_read_reg(SH7751_PCICONF0);
+	if (id != ((SH7751_DEVICE_ID << 16) | SH7751_VENDOR_ID) &&
+	    id != ((SH7751R_DEVICE_ID << 16) | SH7751_VENDOR_ID)) {
+		pr_debug("PCI: This is not an SH7751(R) (%x)\n", id);
+		return -ENODEV;
+	}
+
+	if ((ret = sh4_pci_check_direct()) != 0)
 		return ret;
 
 	return pcibios_init_platform();
 }
-
 subsys_initcall(sh7751_pci_init);
 
 static int __init __area_sdram_check(unsigned int area)
@@ -228,7 +66,7 @@ static int __init __area_sdram_check(unsigned int area)
 		       area, word);
 		return 0;
 	}
-	outl(word, PCI_REG(SH7751_PCIBCR1));
+	pci_write_reg(word, SH4_PCIBCR1);
 
 	word = (u16)inw(SH7751_BCR2);
 	/* check BCR2 for 32bit SDRAM interface*/
@@ -237,12 +75,12 @@ static int __init __area_sdram_check(unsigned int area)
 		       area, word);
 		return 0;
 	}
-	outl(word, PCI_REG(SH7751_PCIBCR2));
+	pci_write_reg(word, SH4_PCIBCR2);
 
 	return 1;
 }
 
-int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
+int __init sh7751_pcic_init(struct sh4_pci_address_map *map)
 {
 	u32 reg;
 	u32 word;
@@ -251,39 +89,39 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 	reg = inl(SH7751_BCR1);
 	reg |= 0x80000;
 	outl(reg, SH7751_BCR1);
-	
+
 	/* Turn the clocks back on (not done in reset)*/
-	outl(0, PCI_REG(SH7751_PCICLKR));
+	pci_write_reg(0, SH4_PCICLKR);
 	/* Clear Powerdown IRQ's (not done in reset) */
-	word = SH7751_PCIPINT_D3 | SH7751_PCIPINT_D0;
-	outl(word, PCI_REG(SH7751_PCIPINT));
+	word = SH4_PCIPINT_D3 | SH4_PCIPINT_D0;
+	pci_write_reg(word, SH4_PCIPINT);
 
 	/*
 	 * This code is unused for some boards as it is done in the
 	 * bootloader and doing it here means the MAC addresses loaded
 	 * by the bootloader get lost.
 	 */
-	if (!(map->flags & SH7751_PCIC_NO_RESET)) {
+	if (!(map->flags & SH4_PCIC_NO_RESET)) {
 		/* toggle PCI reset pin */
-		word = SH7751_PCICR_PREFIX | SH7751_PCICR_PRST;
-		outl(word,PCI_REG(SH7751_PCICR));
+		word = SH4_PCICR_PREFIX | SH4_PCICR_PRST;
+		pci_write_reg(word, SH4_PCICR);
 		/* Wait for a long time... not 1 sec. but long enough */
 		mdelay(100);
-		word = SH7751_PCICR_PREFIX;
-		outl(word,PCI_REG(SH7751_PCICR));
+		word = SH4_PCICR_PREFIX;
+		pci_write_reg(word, SH4_PCICR);
 	}
-	
+
 	/* set the command/status bits to:
 	 * Wait Cycle Control + Parity Enable + Bus Master +
 	 * Mem space enable
 	 */
 	word = SH7751_PCICONF1_WCC | SH7751_PCICONF1_PER | 
 	       SH7751_PCICONF1_BUM | SH7751_PCICONF1_MES;
-	outl(word, PCI_REG(SH7751_PCICONF1));
+	pci_write_reg(word, SH7751_PCICONF1);
 
 	/* define this host as the host bridge */
-	word = SH7751_PCI_HOST_BRIDGE << 24;
-	outl(word, PCI_REG(SH7751_PCICONF2));
+	word = PCI_BASE_CLASS_BRIDGE << 24;
+	pci_write_reg(word, SH7751_PCICONF2);
 
 	/* Set IO and Mem windows to local address 
 	 * Make PCI and local address the same for easy 1 to 1 mapping 
@@ -291,46 +129,49 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 	 * Window1 = map->window1.size @ cached area base = SDRAM 
 	 */
 	word = map->window0.size - 1;
-	outl(word, PCI_REG(SH7751_PCILSR0));
+	pci_write_reg(word, SH4_PCILSR0);
 	word = map->window1.size - 1;
-	outl(word, PCI_REG(SH7751_PCILSR1));
+	pci_write_reg(word, SH4_PCILSR1);
 	/* Set the values on window 0 PCI config registers */
 	word = P2SEGADDR(map->window0.base);
-	outl(word, PCI_REG(SH7751_PCILAR0));
-	outl(word, PCI_REG(SH7751_PCICONF5));
+	pci_write_reg(word, SH4_PCILAR0);
+	pci_write_reg(word, SH7751_PCICONF5);
 	/* Set the values on window 1 PCI config registers */
 	word =  PHYSADDR(map->window1.base);
-	outl(word, PCI_REG(SH7751_PCILAR1));
-	outl(word, PCI_REG(SH7751_PCICONF6));
+	pci_write_reg(word, SH4_PCILAR1);
+	pci_write_reg(word, SH7751_PCICONF6);
 
-	/* Set the local 16MB PCI memory space window to 
+	/* Set the local 16MB PCI memory space window to
 	 * the lowest PCI mapped address
 	 */
-	word = PCIBIOS_MIN_MEM & SH7751_PCIMBR_MASK;
-	PCIDBG(2,"PCI: Setting upper bits of Memory window to 0x%x\n", word);
-	outl(word , PCI_REG(SH7751_PCIMBR));
+	word = PCIBIOS_MIN_MEM & SH4_PCIMBR_MASK;
+	pr_debug("PCI: Setting upper bits of Memory window to 0x%x\n", word);
+	pci_write_reg(word , SH4_PCIMBR);
 
 	/* Map IO space into PCI IO window
 	 * The IO window is 64K-PCIBIOS_MIN_IO in size
-	 * IO addresses will be translated to the 
+	 * IO addresses will be translated to the
 	 * PCI IO window base address
 	 */
-	PCIDBG(3,"PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO,
-	    (64*1024), SH7751_PCI_IO_BASE+PCIBIOS_MIN_IO);
+	pr_debug("PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n",
+		 PCIBIOS_MIN_IO, (64 << 10),
+		 SH4_PCI_IO_BASE + PCIBIOS_MIN_IO);
 
-	/* 
+	/*
 	 * XXX: For now, leave this board-specific. In the event we have other
 	 * boards that need to do similar work, this can be wrapped.
 	 */
 #ifdef CONFIG_SH_BIGSUR
-	bigsur_port_map(PCIBIOS_MIN_IO, (64*1024), SH7751_PCI_IO_BASE+PCIBIOS_MIN_IO,0);
+	bigsur_port_map(PCIBIOS_MIN_IO, (64 << 10),
+			SH4_PCI_IO_BASE + PCIBIOS_MIN_IO, 0);
 #endif
 
-	/* Make sure the MSB's of IO window are set to access PCI space correctly */
-	word = PCIBIOS_MIN_IO & SH7751_PCIIOBR_MASK;
-	PCIDBG(2,"PCI: Setting upper bits of IO window to 0x%x\n", word);
-	outl(word, PCI_REG(SH7751_PCIIOBR));
-	
+	/* Make sure the MSB's of IO window are set to access PCI space
+	 * correctly */
+	word = PCIBIOS_MIN_IO & SH4_PCIIOBR_MASK;
+	pr_debug("PCI: Setting upper bits of IO window to 0x%x\n", word);
+	pci_write_reg(word, SH4_PCIIOBR);
+
 	/* Set PCI WCRx, BCRx's, copy from BSC locations */
 
 	/* check BCR for SDRAM in specified area */
@@ -349,13 +190,13 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 
 	/* configure the wait control registers */
 	word = inl(SH7751_WCR1);
-	outl(word, PCI_REG(SH7751_PCIWCR1));
+	pci_write_reg(word, SH4_PCIWCR1);
 	word = inl(SH7751_WCR2);
-	outl(word, PCI_REG(SH7751_PCIWCR2));
+	pci_write_reg(word, SH4_PCIWCR2);
 	word = inl(SH7751_WCR3);
-	outl(word, PCI_REG(SH7751_PCIWCR3));
+	pci_write_reg(word, SH4_PCIWCR3);
 	word = inl(SH7751_MCR);
-	outl(word, PCI_REG(SH7751_PCIMCR));
+	pci_write_reg(word, SH4_PCIMCR);
 
 	/* NOTE: I'm ignoring the PCI error IRQs for now..
 	 * TODO: add support for the internal error interrupts and
@@ -368,49 +209,8 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 
 	/* SH7751 init done, set central function init complete */
 	/* use round robin mode to stop a device starving/overruning */
-	word = SH7751_PCICR_PREFIX | SH7751_PCICR_CFIN | SH7751_PCICR_ARBM;
-	outl(word,PCI_REG(SH7751_PCICR)); 
+	word = SH4_PCICR_PREFIX | SH4_PCICR_CFIN | SH4_PCICR_ARBM;
+	pci_write_reg(word, SH4_PCICR);
 
 	return 1;
 }
-
-char * __init pcibios_setup(char *str)
-{
-	if (!strcmp(str, "off")) {
-		pci_probe = 0;
-		return NULL;
-	}
-
-	return str;
-}
-
-/* 
- * 	IRQ functions 
- */
-static u8 __init sh7751_no_swizzle(struct pci_dev *dev, u8 *pin)
-{
-	/* no swizzling */
-	return PCI_SLOT(dev->devfn);
-}
-
-static int sh7751_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int irq = -1;
-
-	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
-	irq = pcibios_map_platform_irq(slot,pin);
-	if( irq < 0 ) {
-		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
-		return irq;
-	}
-
-	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
-
-	return irq;
-}
-
-void __init pcibios_fixup_irqs(void)
-{
-	pci_fixup_irqs(sh7751_no_swizzle, sh7751_pci_lookup_irq);
-}
-
diff --git a/arch/sh/drivers/pci/pci-sh7751.h b/arch/sh/drivers/pci/pci-sh7751.h
index 1fee5cae10d1..68e3cb5e6bec 100644
--- a/arch/sh/drivers/pci/pci-sh7751.h
+++ b/arch/sh/drivers/pci/pci-sh7751.h
@@ -3,7 +3,7 @@
  *
  *  Dustin McIntire (dustin@sensoria.com) (c) 2001
  *  Paul Mundt (lethal@linux-sh.org) (c) 2003
- *	
+ *
  *  May be copied or modified under the terms of the GNU General Public
  *  License.  See linux/COPYING for more information.
  *
@@ -12,28 +12,6 @@
 #ifndef _PCI_SH7751_H_
 #define _PCI_SH7751_H_
 
-#include <linux/pci.h>
-
-/* set debug level 4=verbose...1=terse */
-//#define DEBUG_PCI 3
-#undef DEBUG_PCI
-
-#ifdef DEBUG_PCI
-#define PCIDBG(n, x...) { if(DEBUG_PCI>=n) printk(x); }
-#else
-#define PCIDBG(n, x...)
-#endif
-
-/* startup values */
-#define PCI_PROBE_BIOS 1
-#define PCI_PROBE_CONF1 2
-#define PCI_PROBE_CONF2 4
-#define PCI_NO_SORT 0x100
-#define PCI_BIOS_SORT 0x200
-#define PCI_NO_CHECKS 0x400
-#define PCI_ASSIGN_ROMS 0x1000
-#define PCI_BIOS_IRQ_SCAN 0x2000
-
 /* Platform Specific Values */
 #define SH7751_VENDOR_ID             0x1054
 #define SH7751_DEVICE_ID             0x3505
@@ -128,131 +106,6 @@
   #define SH7751_PCICONF17_PMEN      0x00010000  /* PME Enable */
   #define SH7751_PCICONF17_PWST      0x00000003  /* Power State */
 /* SH7715 Internal PCI Registers */
-#define SH7751_PCICR               0x100         /* PCI Control Register */
-  #define SH7751_PCICR_PREFIX        0xA5000000  /* CR prefix for write */
-  #define SH7751_PCICR_TRSB          0x00000200  /* Target Read Single */
-  #define SH7751_PCICR_BSWP          0x00000100  /* Target Byte Swap */
-  #define SH7751_PCICR_PLUP          0x00000080  /* Enable PCI Pullup */
-  #define SH7751_PCICR_ARBM          0x00000040  /* PCI Arbitration Mode */
-  #define SH7751_PCICR_MD            0x00000030  /* MD9 and MD10 status */
-  #define SH7751_PCICR_SERR          0x00000008  /* SERR output assert */
-  #define SH7751_PCICR_INTA          0x00000004  /* INTA output assert */
-  #define SH7751_PCICR_PRST          0x00000002  /* PCI Reset Assert */
-  #define SH7751_PCICR_CFIN          0x00000001  /* Central Fun. Init Done */
-#define SH7751_PCILSR0             0x104         /* PCI Local Space Register0 */
-#define SH7751_PCILSR1             0x108         /* PCI Local Space Register1 */
-#define SH7751_PCILAR0             0x10C         /* PCI Local Address Register1 */
-#define SH7751_PCILAR1             0x110         /* PCI Local Address Register1 */
-#define SH7751_PCIINT              0x114         /* PCI Interrupt Register */
-  #define SH7751_PCIINT_MLCK         0x00008000  /* Master Lock Error */
-  #define SH7751_PCIINT_TABT         0x00004000  /* Target Abort Error */
-  #define SH7751_PCIINT_TRET         0x00000200  /* Target Retry Error */
-  #define SH7751_PCIINT_MFDE         0x00000100  /* Master Func. Disable Error */
-  #define SH7751_PCIINT_PRTY         0x00000080  /* Address Parity Error */
-  #define SH7751_PCIINT_SERR         0x00000040  /* SERR Detection Error */
-  #define SH7751_PCIINT_TWDP         0x00000020  /* Tgt. Write Parity Error */
-  #define SH7751_PCIINT_TRDP         0x00000010  /* Tgt. Read Parity Error Det. */
-  #define SH7751_PCIINT_MTABT        0x00000008  /* Master-Tgt. Abort Error */
-  #define SH7751_PCIINT_MMABT        0x00000004  /* Master-Master Abort Error */
-  #define SH7751_PCIINT_MWPD         0x00000002  /* Master Write PERR Detect */
-  #define SH7751_PCIINT_MRPD         0x00000002  /* Master Read PERR Detect */
-#define SH7751_PCIINTM             0x118         /* PCI Interrupt Mask Register */
-#define SH7751_PCIALR              0x11C         /* Error Address Register */
-#define SH7751_PCICLR              0x120         /* Error Command/Data Register */
-  #define SH7751_PCICLR_MPIO         0x80000000  /* Error Command/Data Register */
-  #define SH7751_PCICLR_MDMA0        0x40000000  /* DMA0 Transfer Error */
-  #define SH7751_PCICLR_MDMA1        0x20000000  /* DMA1 Transfer Error */
-  #define SH7751_PCICLR_MDMA2        0x10000000  /* DMA2 Transfer Error */
-  #define SH7751_PCICLR_MDMA3        0x08000000  /* DMA3 Transfer Error */
-  #define SH7751_PCICLR_TGT          0x04000000  /* Target Transfer Error */
-  #define SH7751_PCICLR_CMDL         0x0000000F  /* PCI Command at Error */
-#define SH7751_PCIAINT             0x130         /* Arbiter Interrupt Register */
-  #define SH7751_PCIAINT_MBKN        0x00002000  /* Master Broken Interrupt */
-  #define SH7751_PCIAINT_TBTO        0x00001000  /* Target Bus Time Out */
-  #define SH7751_PCIAINT_MBTO        0x00001000  /* Master Bus Time Out */
-  #define SH7751_PCIAINT_TABT        0x00000008  /* Target Abort */
-  #define SH7751_PCIAINT_MABT        0x00000004  /* Master Abort */
-  #define SH7751_PCIAINT_RDPE        0x00000002  /* Read Data Parity Error */
-  #define SH7751_PCIAINT_WDPE        0x00000002  /* Write Data Parity Error */
-#define SH7751_PCIAINTM            0x134         /* Arbiter Int. Mask Register */
-#define SH7751_PCIBMLR             0x138         /* Error Bus Master Register */
-  #define SH7751_PCIBMLR_REQ4        0x00000010  /* REQ4 bus master at error */
-  #define SH7751_PCIBMLR_REQ3        0x00000008  /* REQ3 bus master at error */
-  #define SH7751_PCIBMLR_REQ2        0x00000004  /* REQ2 bus master at error */
-  #define SH7751_PCIBMLR_REQ1        0x00000002  /* REQ1 bus master at error */
-  #define SH7751_PCIBMLR_REQ0        0x00000001  /* REQ0 bus master at error */
-#define SH7751_PCIDMABT            0x140         /* DMA Transfer Arb. Register */
-  #define SH7751_PCIDMABT_RRBN       0x00000001  /* DMA Arbitor Round-Robin */
-#define SH7751_PCIDPA0             0x180         /* DMA0 Transfer Addr. Register */
-#define SH7751_PCIDLA0             0x184         /* DMA0 Local Addr. Register */
-#define SH7751_PCIDTC0             0x188         /* DMA0 Transfer Cnt. Register */
-#define SH7751_PCIDCR0             0x18C         /* DMA0 Control Register */
-  #define SH7751_PCIDCR_ALGN         0x00000600  /* DMA Alignment Mode */
-  #define SH7751_PCIDCR_MAST         0x00000100  /* DMA Termination Type */
-  #define SH7751_PCIDCR_INTM         0x00000080  /* DMA Interrupt Done Mask*/
-  #define SH7751_PCIDCR_INTS         0x00000040  /* DMA Interrupt Done Status */
-  #define SH7751_PCIDCR_LHLD         0x00000020  /* Local Address Control */
-  #define SH7751_PCIDCR_PHLD         0x00000010  /* PCI Address Control*/
-  #define SH7751_PCIDCR_IOSEL        0x00000008  /* PCI Address Space Type */
-  #define SH7751_PCIDCR_DIR          0x00000004  /* DMA Transfer Direction */
-  #define SH7751_PCIDCR_STOP         0x00000002  /* Force DMA Stop */
-  #define SH7751_PCIDCR_STRT         0x00000001  /* DMA Start */
-#define SH7751_PCIDPA1             0x190         /* DMA1 Transfer Addr. Register */
-#define SH7751_PCIDLA1             0x194         /* DMA1 Local Addr. Register */
-#define SH7751_PCIDTC1             0x198         /* DMA1 Transfer Cnt. Register */
-#define SH7751_PCIDCR1             0x19C         /* DMA1 Control Register */
-#define SH7751_PCIDPA2             0x1A0         /* DMA2 Transfer Addr. Register */
-#define SH7751_PCIDLA2             0x1A4         /* DMA2 Local Addr. Register */
-#define SH7751_PCIDTC2             0x1A8         /* DMA2 Transfer Cnt. Register */
-#define SH7751_PCIDCR2             0x1AC         /* DMA2 Control Register */
-#define SH7751_PCIDPA3             0x1B0         /* DMA3 Transfer Addr. Register */
-#define SH7751_PCIDLA3             0x1B4         /* DMA3 Local Addr. Register */
-#define SH7751_PCIDTC3             0x1B8         /* DMA3 Transfer Cnt. Register */
-#define SH7751_PCIDCR3             0x1BC         /* DMA3 Control Register */
-#define SH7751_PCIPAR              0x1C0         /* PIO Address Register */
-  #define SH7751_PCIPAR_CFGEN        0x80000000  /* Configuration Enable */
-  #define SH7751_PCIPAR_BUSNO        0x00FF0000  /* Config. Bus Number */
-  #define SH7751_PCIPAR_DEVNO        0x0000FF00  /* Config. Device Number */
-  #define SH7751_PCIPAR_REGAD        0x000000FC  /* Register Address Number */
-#define SH7751_PCIMBR              0x1C4         /* Memory Base Address Register */
-  #define SH7751_PCIMBR_MASK         0xFF000000  /* Memory Space Mask */
-  #define SH7751_PCIMBR_LOCK         0x00000001  /* Lock Memory Space */
-#define SH7751_PCIIOBR             0x1C8         /* I/O Base Address Register */
-  #define SH7751_PCIIOBR_MASK         0xFFFC0000 /* IO Space Mask */
-  #define SH7751_PCIIOBR_LOCK         0x00000001 /* Lock IO Space */
-#define SH7751_PCIPINT             0x1CC         /* Power Mgmnt Int. Register */
-  #define SH7751_PCIPINT_D3           0x00000002 /* D3 Pwr Mgmt. Interrupt */
-  #define SH7751_PCIPINT_D0           0x00000001 /* D0 Pwr Mgmt. Interrupt */  
-#define SH7751_PCIPINTM            0x1D0         /* Power Mgmnt Mask Register */
-#define SH7751_PCICLKR             0x1D4         /* Clock Ctrl. Register */
-  #define SH7751_PCICLKR_PCSTP        0x00000002 /* PCI Clock Stop */
-  #define SH7751_PCICLKR_BCSTP        0x00000002 /* BCLK Clock Stop */
-/* For definitions of BCR, MCR see ... */
-#define SH7751_PCIBCR1             0x1E0         /* Memory BCR1 Register */
-#define SH7751_PCIBCR2             0x1E4         /* Memory BCR2 Register */
-#define SH7751_PCIWCR1             0x1E8         /* Wait Control 1 Register */
-#define SH7751_PCIWCR2             0x1EC         /* Wait Control 2 Register */
-#define SH7751_PCIWCR3             0x1F0         /* Wait Control 3 Register */
-#define SH7751_PCIMCR              0x1F4         /* Memory Control Register */
-#define SH7751_PCIBCR3		   0x1f8	 /* Memory BCR3 Register */
-#define SH7751_PCIPCTR             0x200         /* Port Control Register */
-  #define SH7751_PCIPCTR_P2EN        0x000400000 /* Port 2 Enable */
-  #define SH7751_PCIPCTR_P1EN        0x000200000 /* Port 1 Enable */
-  #define SH7751_PCIPCTR_P0EN        0x000100000 /* Port 0 Enable */
-  #define SH7751_PCIPCTR_P2UP        0x000000020 /* Port2 Pull Up Enable */
-  #define SH7751_PCIPCTR_P2IO        0x000000010 /* Port2 Output Enable */
-  #define SH7751_PCIPCTR_P1UP        0x000000008 /* Port1 Pull Up Enable */
-  #define SH7751_PCIPCTR_P1IO        0x000000004 /* Port1 Output Enable */
-  #define SH7751_PCIPCTR_P0UP        0x000000002 /* Port0 Pull Up Enable */
-  #define SH7751_PCIPCTR_P0IO        0x000000001 /* Port0 Output Enable */
-#define SH7751_PCIPDTR             0x204         /* Port Data Register */
-  #define SH7751_PCIPDTR_PB5         0x000000020 /* Port 5 Enable */
-  #define SH7751_PCIPDTR_PB4         0x000000010 /* Port 4 Enable */
-  #define SH7751_PCIPDTR_PB3         0x000000008 /* Port 3 Enable */
-  #define SH7751_PCIPDTR_PB2         0x000000004 /* Port 2 Enable */
-  #define SH7751_PCIPDTR_PB1         0x000000002 /* Port 1 Enable */
-  #define SH7751_PCIPDTR_PB0         0x000000001 /* Port 0 Enable */
-#define SH7751_PCIPDR              0x220         /* Port IO Data Register */
 
 /* Memory Control Registers */
 #define SH7751_BCR1                0xFF800000    /* Memory BCR1 Register */
@@ -274,30 +127,9 @@
 #define SH7751_CS5_BASE_ADDR       (SH7751_CS4_BASE_ADDR + SH7751_MEM_REGION_SIZE)
 #define SH7751_CS6_BASE_ADDR       (SH7751_CS5_BASE_ADDR + SH7751_MEM_REGION_SIZE)
 
-/* General PCI values */
-#define SH7751_PCI_HOST_BRIDGE		0x6
-
-/* Flags */
-#define SH7751_PCIC_NO_RESET	0x0001
-
-/* External functions defined per platform i.e. Big Sur, SE... (these could be routed 
- * through the machine vectors... */
-extern int pcibios_init_platform(void);
-extern int pcibios_map_platform_irq(u8 slot, u8 pin);
-
-struct sh7751_pci_address_space {
-	unsigned long base;
-	unsigned long size;
-};
-
-struct sh7751_pci_address_map {
-	struct sh7751_pci_address_space window0;
-	struct sh7751_pci_address_space window1;
-	unsigned long flags;
-};
+struct sh4_pci_address_map;
 
 /* arch/sh/drivers/pci/pci-sh7751.c */
-extern int sh7751_pcic_init(struct sh7751_pci_address_map *map);
+int sh7751_pcic_init(struct sh4_pci_address_map *map);
 
 #endif /* _PCI_SH7751_H_ */
-
diff --git a/arch/sh/drivers/pci/pci-sh7780.c b/arch/sh/drivers/pci/pci-sh7780.c
index e09721330ac2..bd3064a82087 100644
--- a/arch/sh/drivers/pci/pci-sh7780.c
+++ b/arch/sh/drivers/pci/pci-sh7780.c
@@ -20,197 +20,36 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/pci.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
 #include <linux/errno.h>
-#include <linux/irq.h>
 #include <linux/delay.h>
-
-#include <asm/machvec.h>
-#include <asm/io.h>
-#include "pci-sh7780.h"
-
-static unsigned int pci_probe = PCI_PROBE_CONF1;
-extern int pci_fixup_pcic(void);
-
-/*
- * Direct access to PCI hardware...
- */
-
-#define CONFIG_CMD(bus, devfn, where) (0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
-
-/*
- * Functions for accessing PCI configuration space with type 1 accesses
- */
-static int sh7780_pci_read(struct pci_bus *bus, unsigned int devfn,
-			   int where, int size, u32 *val)
-{
-	unsigned long flags;
-	u32 data;
-
-	/*
-	 * PCIPDR may only be accessed as 32 bit words,
-	 * so we must do byte alignment by hand
-	 */
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
-	data = inl(PCI_REG(SH7780_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		*val = (data >> ((where & 3) << 3)) & 0xff;
-		break;
-	case 2:
-		*val = (data >> ((where & 2) << 3)) & 0xffff;
-		break;
-	case 4:
-		*val = data;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
+#include "pci-sh4.h"
 
 /*
- * Since SH7780 only does 32bit access we'll have to do a read,
- * mask,write operation.
- * We'll allow an odd byte offset, though it should be illegal.
+ * Initialization. Try all known PCI access methods. Note that we support
+ * using both PCI BIOS and direct access: in such cases, we use I/O ports
+ * to access config space.
+ *
+ * Note that the platform specific initialization (BSC registers, and memory
+ * space mapping) will be called via the platform defined function
+ * pcibios_init_platform().
  */
-static int sh7780_pci_write(struct pci_bus *bus, unsigned int devfn,
-			    int where, int size, u32 val)
+static int __init sh7780_pci_init(void)
 {
-	unsigned long flags;
-	int shift;
-	u32 data;
-
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
-	data = inl(PCI_REG(SH7780_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		shift = (where & 3) << 3;
-		data &= ~(0xff << shift);
-		data |= ((val & 0xff) << shift);
-		break;
-	case 2:
-		shift = (where & 2) << 3;
-		data &= ~(0xffff << shift);
-		data |= ((val & 0xffff) << shift);
-		break;
-	case 4:
-		data = val;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	outl(data, PCI_REG(SH7780_PCIPDR));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-#undef CONFIG_CMD
-
-struct pci_ops sh7780_pci_ops = {
-	.read		= sh7780_pci_read,
-	.write		= sh7780_pci_write,
-};
+	unsigned int id;
+	int ret;
 
-static int __init pci_check_direct(void)
-{
-	unsigned int tmp, id;
+	pr_debug("PCI: Starting intialization.\n");
 
 	outl(0x00000001, SH7780_PCI_VCR2); /* Enable PCIC */
 
 	/* check for SH7780/SH7780R hardware */
-	id = inl(PCI_REG(SH7780_PCIVID));
+	id = pci_read_reg(SH7780_PCIVID);
 	if ((id != ((SH7780_DEVICE_ID << 16) | SH7780_VENDOR_ID)) &&
 	    (id != ((SH7781_DEVICE_ID << 16) | SH7780_VENDOR_ID))) {
 		printk(KERN_ERR "PCI: This is not an SH7780 (%x)\n", id);
 		return -ENODEV;
 	}
 
-	/*
-	 * Check if configuration works.
-	 */
-	if (pci_probe & PCI_PROBE_CONF1) {
-		tmp = inl(PCI_REG(SH7780_PCIPAR));
-		outl(0x80000000, PCI_REG(SH7780_PCIPAR));
-		if (inl(PCI_REG(SH7780_PCIPAR)) == 0x80000000) {
-			outl(tmp, PCI_REG(SH7780_PCIPAR));
-			printk(KERN_INFO "PCI: Using configuration type 1\n");
-			request_region(PCI_REG(SH7780_PCIPAR), 8, "PCI conf1");
-			return 0;
-		}
-		outl(tmp, PCI_REG(SH7780_PCIPAR));
-	}
-
-	pr_debug("PCI: pci_check_direct failed\n");
-	return -EINVAL;
-}
-
-/***************************************************************************************/
-
-/*
- *  Handle bus scanning and fixups ....
- */
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
-
-/*
- *  Called after each bus is probed, but before its children
- *  are examined.
- */
-
-void __init pcibios_fixup_bus(struct pci_bus *b)
-{
-	pci_read_bridge_bases(b);
-}
-
-/*
- * Initialization. Try all known PCI access methods. Note that we support
- * using both PCI BIOS and direct access: in such cases, we use I/O ports
- * to access config space.
- *
- * Note that the platform specific initialization (BSC registers, and memory
- * space mapping) will be called via the machine vectors (sh_mv.mv_pci_init()) if it
- * exists and via the platform defined function pcibios_init_platform().
- * See pci_bigsur.c for implementation;
- *
- * The BIOS version of the pci functions is not yet implemented but it is left
- * in for completeness.  Currently an error will be genereated at compile time.
- */
-
-static int __init sh7780_pci_init(void)
-{
-	int ret;
-
-	pr_debug("PCI: Starting intialization.\n");
-
 	/* Setup the INTC */
 	ctrl_outl(0x00200000, INTC_ICR0);	/* INTC SH-4 Mode */
 	ctrl_outl(0x00078000, INTC_INT2MSKCR);	/* enable PCIINTA - PCIINTD */
@@ -219,15 +58,14 @@ static int __init sh7780_pci_init(void)
 	ctrl_outl(0x80000000, INTC_INTMSKCLR1);	/* enable IRL0-3 Interrupt */
 	ctrl_outl(0xfffe0000, INTC_INTMSKCLR2);	/* enable IRL0-3 Interrupt */
 
-	if ((ret = pci_check_direct()) != 0)
+	if ((ret = sh4_pci_check_direct()) != 0)
 		return ret;
 
 	return pcibios_init_platform();
 }
-
 core_initcall(sh7780_pci_init);
 
-int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
+int __init sh7780_pcic_init(struct sh4_pci_address_map *map)
 {
 	u32 word;
 
@@ -236,25 +74,25 @@ int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
 	 * bootloader and doing it here means the MAC addresses loaded
 	 * by the bootloader get lost.
 	 */
-	if (!(map->flags & SH7780_PCIC_NO_RESET)) {
+	if (!(map->flags & SH4_PCIC_NO_RESET)) {
 		/* toggle PCI reset pin */
-		word = SH7780_PCICR_PREFIX | SH7780_PCICR_PRST;
-		outl(word,PCI_REG(SH7780_PCICR));
+		word = SH4_PCICR_PREFIX | SH4_PCICR_PRST;
+		pci_write_reg(word, SH4_PCICR);
 		/* Wait for a long time... not 1 sec. but long enough */
 		mdelay(100);
-		word = SH7780_PCICR_PREFIX;
-		outl(word,PCI_REG(SH7780_PCICR));
+		word = SH4_PCICR_PREFIX;
+		pci_write_reg(word, SH4_PCICR);
 	}
 
 	/* set the command/status bits to:
 	 * Wait Cycle Control + Parity Enable + Bus Master +
 	 * Mem space enable
 	 */
-	outl(0x00000046, PCI_REG(SH7780_PCICMD));
+	pci_write_reg(0x00000046, SH7780_PCICMD);
 
 	/* define this host as the host bridge */
-	word = SH7780_PCI_HOST_BRIDGE << 24;
-	outl(word, PCI_REG(SH7780_PCIRID));
+	word = PCI_BASE_CLASS_BRIDGE << 24;
+	pci_write_reg(word, SH7780_PCIRID);
 
 	/* Set IO and Mem windows to local address
 	 * Make PCI and local address the same for easy 1 to 1 mapping
@@ -262,25 +100,26 @@ int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
 	 * Window1 = map->window1.size @ cached area base = SDRAM
 	 */
 	word = ((map->window0.size - 1) & 0x1ff00001) | 0x01;
-	outl(0x07f00001, PCI_REG(SH7780_PCILSR0));
+	pci_write_reg(0x07f00001, SH4_PCILSR0);
 	word = ((map->window1.size - 1) & 0x1ff00001) | 0x01;
-	outl(0x00000001, PCI_REG(SH7780_PCILSR1));
+	pci_write_reg(0x00000001, SH4_PCILSR1);
 	/* Set the values on window 0 PCI config registers */
 	word = P2SEGADDR(map->window0.base);
-	outl(0xa8000000, PCI_REG(SH7780_PCILAR0));
-	outl(0x08000000, PCI_REG(SH7780_PCIMBAR0));
+	pci_write_reg(0xa8000000, SH4_PCILAR0);
+	pci_write_reg(0x08000000, SH7780_PCIMBAR0);
 	/* Set the values on window 1 PCI config registers */
 	word = P2SEGADDR(map->window1.base);
-	outl(0x00000000, PCI_REG(SH7780_PCILAR1));
-	outl(0x00000000, PCI_REG(SH7780_PCIMBAR1));
+	pci_write_reg(0x00000000, SH4_PCILAR1);
+	pci_write_reg(0x00000000, SH7780_PCIMBAR1);
 
 	/* Map IO space into PCI IO window
 	 * The IO window is 64K-PCIBIOS_MIN_IO in size
 	 * IO addresses will be translated to the
 	 * PCI IO window base address
 	 */
-	PCIDBG(3,"PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO,
-	    (64*1024), SH7780_PCI_IO_BASE+PCIBIOS_MIN_IO);
+	pr_debug("PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n",
+		 PCIBIOS_MIN_IO, (64 << 10),
+		 SH7780_PCI_IO_BASE + PCIBIOS_MIN_IO);
 
 	/* NOTE: I'm ignoring the PCI error IRQs for now..
 	 * TODO: add support for the internal error interrupts and
@@ -293,49 +132,8 @@ int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
 
 	/* SH7780 init done, set central function init complete */
 	/* use round robin mode to stop a device starving/overruning */
-	word = SH7780_PCICR_PREFIX | SH7780_PCICR_CFIN | /* SH7780_PCICR_ARBM |*/ SH7780_PCICR_FTO;
-	outl(word, PCI_REG(SH7780_PCICR));
+	word = SH4_PCICR_PREFIX | SH4_PCICR_CFIN | SH4_PCICR_FTO;
+	pci_write_reg(word, SH4_PCICR);
 
 	return 1;
 }
-
-char * __init pcibios_setup(char *str)
-{
-	if (!strcmp(str, "off")) {
-		pci_probe = 0;
-		return NULL;
-	}
-
-	return str;
-}
-
-/*
- *	IRQ functions
- */
-static u8 __init sh7780_no_swizzle(struct pci_dev *dev, u8 *pin)
-{
-	/* no swizzling */
-	return PCI_SLOT(dev->devfn);
-}
-
-static int sh7780_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int irq = -1;
-
-	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
-	irq = pcibios_map_platform_irq(slot,pin);
-	if( irq < 0 ) {
-		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
-		return irq;
-	}
-
-	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
-
-	return irq;
-}
-
-void __init pcibios_fixup_irqs(void)
-{
-	pci_fixup_irqs(sh7780_no_swizzle, sh7780_pci_lookup_irq);
-}
-
diff --git a/arch/sh/drivers/pci/pci-sh7780.h b/arch/sh/drivers/pci/pci-sh7780.h
index 750d5d7753a2..f02d2180a4bc 100644
--- a/arch/sh/drivers/pci/pci-sh7780.h
+++ b/arch/sh/drivers/pci/pci-sh7780.h
@@ -12,28 +12,6 @@
 #ifndef _PCI_SH7780_H_
 #define _PCI_SH7780_H_
 
-#include <linux/pci.h>
-
-/* set debug level 4=verbose...1=terse */
-//#define DEBUG_PCI 3
-#undef DEBUG_PCI
-
-#ifdef DEBUG_PCI
-#define PCIDBG(n, x...) { if(DEBUG_PCI>=n) printk(x); }
-#else
-#define PCIDBG(n, x...)
-#endif
-
-/* startup values */
-#define PCI_PROBE_BIOS		1
-#define PCI_PROBE_CONF1		2
-#define PCI_PROBE_CONF2		4
-#define PCI_NO_SORT		0x100
-#define PCI_BIOS_SORT		0x200
-#define PCI_NO_CHECKS		0x400
-#define PCI_ASSIGN_ROMS		0x1000
-#define PCI_BIOS_IRQ_SCAN	0x2000
-
 /* Platform Specific Values */
 #define SH7780_VENDOR_ID	0x1912
 #define SH7780_DEVICE_ID	0x0002
@@ -47,15 +25,12 @@
 /* SH7780 Specific Values */
 #define SH7780_PCI_CONFIG_BASE	0xFD000000	/* Config space base addr */
 #define SH7780_PCI_CONFIG_SIZE	0x01000000	/* Config space size */
+
 #define SH7780_PCI_MEMORY_BASE	0xFD000000	/* Memory space base addr */
 #define SH7780_PCI_MEM_SIZE	0x01000000	/* Size of Memory window */
-#if 1
+
 #define SH7780_PCI_IO_BASE	0xFE400000	/* IO space base address */
 #define SH7780_PCI_IO_SIZE	0x00400000	/* Size of IO window */
-#else
-#define SH7780_PCI_IO_BASE	0xFE200000	/* IO space base address */
-#define SH7780_PCI_IO_SIZE	0x00200000	/* Size of IO window */
-#endif
 
 #define SH7780_PCIREG_BASE	0xFE040000	/* PCI regs base address */
 #define PCI_REG(n)		(SH7780_PCIREG_BASE+n)
@@ -90,44 +65,16 @@
 #define SH7780_PCIPMCSR_BSE	0x046
 #define SH7780_PCICDD		0x047
 
-/* SH7780 PCI Local Registers */
-#define SH7780_PCICR		0x100		/* PCI Control Register */
-  #define SH7780_PCICR_PREFIX	0xA5000000	/* CR prefix for write */
-  #define SH7780_PCICR_PFCS	0x00000800	/* TRDY/IRDY Enable */
-  #define SH7780_PCICR_FTO	0x00000400	/* TRDY/IRDY Enable */
-  #define SH7780_PCICR_PFE	0x00000200	/* Target Read Single */
-  #define SH7780_PCICR_TBS	0x00000100	/* Target Byte Swap */
-  #define SH7780_PCICR_ARBM	0x00000040	/* PCI Arbitration Mode */
-  #define SH7780_PCICR_IOCS	0x00000004	/* INTA output assert */
-  #define SH7780_PCICR_PRST	0x00000002	/* PCI Reset Assert */
-  #define SH7780_PCICR_CFIN	0x00000001	/* Central Fun. Init Done */
-#define SH7780_PCILSR0		0x104		/* PCI Local Space Register0 */
-#define SH7780_PCILSR1		0x108		/* PCI Local Space Register1 */
-#define SH7780_PCILAR0		0x10C	/* PCI Local Address Register1 */
-#define SH7780_PCILAR1		0x110	/* PCI Local Address Register1 */
-#define SH7780_PCIIR		0x114		/* PCI Interrupt Register */
-#define SH7780_PCIIMR		0x118	/* PCI Interrupt Mask Register */
-#define SH7780_PCIAIR		0x11C		/* Error Address Register */
-#define SH7780_PCICIR		0x120	/* Error Command/Data Register */
-#define SH7780_PCIAINT		0x130		/* Arbiter Interrupt Register */
-#define SH7780_PCIAINTM		0x134		/* Arbiter Int. Mask Register */
-#define SH7780_PCIBMIR		0x138		/* Error Bus Master Register */
-#define SH7780_PCIPAR		0x1C0		/* PIO Address Register */
-#define SH7780_PCIPINT		0x1CC	/* Power Management Int. Register */
-#define SH7780_PCIPINTM		0x1D0	/* Power Management Mask Register */
-#define SH7780_PCIMBR0		0x1E0		/* Memory Bank0 Register */
-#define SH7780_PCIMBMR0		0x1E4		/* Memory Bank0 Mask Register */
-#define SH7780_PCIMBR1		0x1E8		/* Memory Bank1  Register */
-#define SH7780_PCIMBMR1		0x1EC		/* Memory Bank1 Mask Register */
-#define SH7780_PCIMBR2		0x1F0		/* Memory Bank2 Register */
-#define SH7780_PCIMBMR2		0x1F4		/* Memory Bank2 Mask Register */
-#define SH7780_PCIIOBR		0x1F8		/* Bank Register */
-#define SH7780_PCIIOBMR		0x1FC		/* Bank Mask Register */
+#define SH7780_PCIMBR0		0x1E0
+#define SH7780_PCIMBMR0		0x1E4
+#define SH7780_PCIMBR2		0x1F0
+#define SH7780_PCIMBMR2		0x1F4
+#define SH7780_PCIIOBR		0x1F8
+#define SH7780_PCIIOBMR		0x1FC
 #define SH7780_PCICSCR0		0x210		/* Cache Snoop1 Cnt. Register */
 #define SH7780_PCICSCR1		0x214		/* Cache Snoop2 Cnt. Register */
 #define SH7780_PCICSAR0		0x218	/* Cache Snoop1 Addr. Register */
 #define SH7780_PCICSAR1		0x21C	/* Cache Snoop2 Addr. Register */
-#define SH7780_PCIPDR		0x220		/* Port IO Data Register */
 
 /* General Memory Config Addresses */
 #define SH7780_CS0_BASE_ADDR	0x0
@@ -139,30 +86,9 @@
 #define SH7780_CS5_BASE_ADDR	(SH7780_CS4_BASE_ADDR + SH7780_MEM_REGION_SIZE)
 #define SH7780_CS6_BASE_ADDR	(SH7780_CS5_BASE_ADDR + SH7780_MEM_REGION_SIZE)
 
-/* General PCI values */
-#define SH7780_PCI_HOST_BRIDGE	0x6
-
-/* Flags */
-#define SH7780_PCIC_NO_RESET	0x0001
-
-/* External functions defined per platform i.e. Big Sur, SE... (these could be routed
- * through the machine vectors... */
-extern int pcibios_init_platform(void);
-extern int pcibios_map_platform_irq(u8 slot, u8 pin);
-
-struct sh7780_pci_address_space {
-	unsigned long base;
-	unsigned long size;
-};
-
-struct sh7780_pci_address_map {
-	struct sh7780_pci_address_space window0;
-	struct sh7780_pci_address_space window1;
-	unsigned long flags;
-};
+struct sh4_pci_address_map;
 
 /* arch/sh/drivers/pci/pci-sh7780.c */
-extern int sh7780_pcic_init(struct sh7780_pci_address_map *map);
+int sh7780_pcic_init(struct sh4_pci_address_map *map);
 
 #endif /* _PCI_SH7780_H_ */
-
diff --git a/arch/sh/drivers/pci/pci-st40.c b/arch/sh/drivers/pci/pci-st40.c
index 7c81b8b65bb5..4ab5ea6b35fb 100644
--- a/arch/sh/drivers/pci/pci-st40.c
+++ b/arch/sh/drivers/pci/pci-st40.c
@@ -70,12 +70,6 @@
 static void pci_set_rbar_region(unsigned int region,     unsigned long localAddr,
 			 unsigned long pciOffset, unsigned long regionSize);
 
-/*
- * The pcibios_map_platform_irq function is defined in the appropriate
- * board specific code and referenced here
- */
-extern int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin);
-
 static __init void SetPCIPLL(void)
 {
 	{
@@ -422,13 +416,6 @@ struct pci_ops st40pci_config_ops = {
 /* Everything hangs off this */
 static struct pci_bus *pci_root_bus;
 
-
-static u8 __init no_swizzle(struct pci_dev *dev, u8 * pin)
-{
-	return PCI_SLOT(dev->devfn);
-}
-
-
 static int __init pcibios_init(void)
 {
 	extern unsigned long memory_start, memory_end;
@@ -465,17 +452,11 @@ static int __init pcibios_init(void)
 	/* ok, do the scan man */
 	pci_root_bus = pci_scan_bus(0, &st40pci_config_ops, NULL);
 	pci_assign_unassigned_resources();
-	pci_fixup_irqs(no_swizzle, pcibios_map_platform_irq);
 
 	return 0;
 }
-
 subsys_initcall(pcibios_init);
 
-void __init pcibios_fixup_bus(struct pci_bus *bus)
-{
-}
-
 /*
  * Publish a region of local address space over the PCI bus
  * to other devices.
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 285dffd12bd8..7377a8a8e161 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -21,6 +21,26 @@
 #include <linux/init.h>
 #include <asm/io.h>
 
+static inline u8 bridge_swizzle(u8 pin, u8 slot)
+{
+	return (((pin - 1) + slot) % 4) + 1;
+}
+
+static u8 __init simple_swizzle(struct pci_dev *dev, u8 *pinp)
+{
+	u8 pin = *pinp;
+
+	while (dev->bus->parent) {
+		pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
+		/* Move up the chain of bridges. */
+		dev = dev->bus->self;
+	}
+	*pinp = pin;
+
+	/* The slot is the slot of the last bridge. */
+	return PCI_SLOT(dev->devfn);
+}
+
 static int __init pcibios_init(void)
 {
 	struct pci_channel *p;
@@ -36,19 +56,26 @@ static int __init pcibios_init(void)
 
 	/* scan the buses */
 	busno = 0;
-	for (p= board_pci_channels; p->pci_ops != NULL; p++) {
+	for (p = board_pci_channels; p->pci_ops != NULL; p++) {
 		bus = pci_scan_bus(busno, p->pci_ops, p);
-		busno = bus->subordinate+1;
+		busno = bus->subordinate + 1;
 	}
 
-	/* board-specific fixups */
-	pcibios_fixup_irqs();
+	pci_fixup_irqs(simple_swizzle, pcibios_map_platform_irq);
 
 	return 0;
 }
-
 subsys_initcall(pcibios_init);
 
+/*
+ *  Called after each bus is probed, but before its children
+ *  are examined.
+ */
+void __init pcibios_fixup_bus(struct pci_bus *bus)
+{
+	pci_read_bridge_bases(bus);
+}
+
 void
 pcibios_update_resource(struct pci_dev *dev, struct resource *root,
 			struct resource *res, int resource)
@@ -192,11 +219,10 @@ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 
 	return NULL;
 }
+EXPORT_SYMBOL(pci_iomap);
 
 void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
 {
 	iounmap(addr);
 }
-
-EXPORT_SYMBOL(pci_iomap);
 EXPORT_SYMBOL(pci_iounmap);
diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index 3d5cafc71ae3..9611fab03b55 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -7,6 +7,9 @@ obj-y	:= ex.o probe.o
 obj-$(CONFIG_SH_FPU)                    += fpu.o
 obj-$(CONFIG_SH_STORE_QUEUES)		+= sq.o
 
+# CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
+
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH4)			:= clock-sh4.o
 clock-$(CONFIG_CPU_SUBTYPE_SH73180)	:= clock-sh73180.o
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7780.c b/arch/sh/kernel/cpu/sh4/setup-sh7780.c
new file mode 100644
index 000000000000..72493f259edc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7780.c
@@ -0,0 +1,79 @@
+/*
+ * SH7780 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct resource rtc_resources[] = {
+	[0] = {
+		.start	= 0xffe80000,
+		.end	= 0xffe80000 + 0x58 - 1,
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		/* Period IRQ */
+		.start	= 21,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		/* Carry IRQ */
+		.start	= 22,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[3] = {
+		/* Alarm IRQ */
+		.start	= 23,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device rtc_device = {
+	.name		= "sh-rtc",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(rtc_resources),
+	.resource	= rtc_resources,
+};
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.mapbase	= 0xffe10000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 76, 77, 79, 78 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7780_devices[] __initdata = {
+	&rtc_device,
+	&sci_device,
+};
+
+static int __init sh7780_devices_setup(void)
+{
+	return platform_add_devices(sh7780_devices,
+				    ARRAY_SIZE(sh7780_devices));
+}
+__initcall(sh7780_devices_setup);
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 894e64b2d5f0..27dba653cbe7 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -209,6 +209,11 @@ static inline void ctrl_outl(unsigned int b, unsigned long addr)
         *(volatile unsigned long*)addr = b;
 }
 
+static inline void ctrl_delay(void)
+{
+	ctrl_inw(P2SEG);
+}
+
 #define IO_SPACE_LIMIT 0xffffffff
 
 /*
diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 18a109de0f2e..6ccc948fe216 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -32,6 +32,34 @@ extern struct pci_channel board_pci_channels[];
 #define PCIBIOS_MIN_IO		board_pci_channels->io_resource->start
 #define PCIBIOS_MIN_MEM		board_pci_channels->mem_resource->start
 
+/*
+ * I/O routine helpers
+ */
+#ifdef CONFIG_CPU_SUBTYPE_SH7780
+#define PCI_IO_AREA		0xFE400000
+#define PCI_IO_SIZE		0x00400000
+#else
+#define PCI_IO_AREA		0xFE240000
+#define PCI_IO_SIZE		0X00040000
+#endif
+
+#define PCI_MEM_SIZE		0x01000000
+
+#define SH4_PCIIOBR_MASK	0xFFFC0000
+#define pci_ioaddr(addr)	(PCI_IO_AREA + (addr & ~SH4_PCIIOBR_MASK))
+
+#if defined(CONFIG_PCI)
+#define is_pci_ioaddr(port)		\
+	(((port) >= PCIBIOS_MIN_IO) &&	\
+	 ((port) < (PCIBIOS_MIN_IO + PCI_IO_SIZE)))
+#define is_pci_memaddr(port)		\
+	(((port) >= PCIBIOS_MIN_MEM) &&	\
+	 ((port) < (PCIBIOS_MIN_MEM + PCI_MEM_SIZE)))
+#else
+#define is_pci_ioaddr(port)	(0)
+#define is_pci_memaddr(port)	(0)
+#endif
+
 struct pci_dev;
 
 extern void pcibios_set_master(struct pci_dev *dev);
@@ -98,11 +126,12 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 #endif
 
 /* Board-specific fixup routines. */
-extern void pcibios_fixup(void);
-extern void pcibios_fixup_irqs(void);
+void pcibios_fixup(void);
+int pcibios_init_platform(void);
+int pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin);
 
 #ifdef CONFIG_PCI_AUTO
-extern int pciauto_assign_resources(int busno, struct pci_channel *hose);
+int pciauto_assign_resources(int busno, struct pci_channel *hose);
 #endif
 
 static inline void pcibios_add_platform_entries(struct pci_dev *dev)
-- 
cgit v1.2.3


From 0f08f338083cc1d68788ccbccc44bd0502fc57ae Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:03:56 +0900
Subject: sh: More cosmetic cleanups and trivial fixes.

Nothing exciting here, just trivial fixes..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/drivers/dma/dma-sh.c    | 10 ++++++----
 arch/sh/drivers/pci/ops-titan.c |  1 +
 arch/sh/kernel/setup.c          |  2 +-
 arch/sh/kernel/sh_ksyms.c       | 15 ---------------
 arch/sh/mm/cache-sh7705.c       | 19 ++++++-------------
 arch/sh/mm/fault.c              | 13 ++++++++-----
 arch/sh/mm/pmb.c                |  4 ----
 include/asm-sh/dma.h            |  1 +
 include/asm-sh/irq.h            |  5 +++++
 include/asm-sh/kexec.h          |  2 ++
 include/asm-sh/system.h         | 36 +++++-------------------------------
 11 files changed, 35 insertions(+), 73 deletions(-)

(limited to 'include')

diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c
index 40a480d20aaf..cbbe8bce3d67 100644
--- a/arch/sh/drivers/dma/dma-sh.c
+++ b/arch/sh/drivers/dma/dma-sh.c
@@ -80,21 +80,23 @@ static irqreturn_t dma_tei(int irq, void *dev_id, struct pt_regs *regs)
 
 static int sh_dmac_request_dma(struct dma_channel *chan)
 {
-	char name[32];
-
 	if (unlikely(!chan->flags & DMA_TEI_CAPABLE))
 		return 0;
 
-	snprintf(name, sizeof(name), "DMAC Transfer End (Channel %d)",
+	chan->name = kzalloc(32, GFP_KERNEL);
+	if (unlikely(chan->name == NULL))
+		return -ENOMEM;
+	snprintf(chan->name, 32, "DMAC Transfer End (Channel %d)",
 		 chan->chan);
 
 	return request_irq(get_dmte_irq(chan->chan), dma_tei,
-			   IRQF_DISABLED, name, chan);
+			   IRQF_DISABLED, chan->name, chan);
 }
 
 static void sh_dmac_free_dma(struct dma_channel *chan)
 {
 	free_irq(get_dmte_irq(chan->chan), chan);
+	kfree(chan->name);
 }
 
 static void
diff --git a/arch/sh/drivers/pci/ops-titan.c b/arch/sh/drivers/pci/ops-titan.c
index 9c8b2027c35d..c6097bcd97fd 100644
--- a/arch/sh/drivers/pci/ops-titan.c
+++ b/arch/sh/drivers/pci/ops-titan.c
@@ -16,6 +16,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/titan.h>
 #include "pci-sh4.h"
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index cff8d36f91b0..4afdec071700 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -507,7 +507,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	 * unified cache on the SH-2 and SH-3, as well as the harvard
 	 * style cache on the SH-4.
 	 */
-	if (test_bit(SH_CACHE_COMBINED, &(boot_cpu_data.icache.flags))) {
+	if (boot_cpu_data.icache.flags & SH_CACHE_COMBINED) {
 		seq_printf(m, "unified\n");
 		show_cacheinfo(m, "cache", boot_cpu_data.icache);
 	} else {
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index fd73ab0326e9..7f3a42244cab 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -27,20 +27,11 @@ EXPORT_SYMBOL(sh_mv);
 
 /* platform dependent support */
 EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL(enable_irq);
-EXPORT_SYMBOL(disable_irq);
-EXPORT_SYMBOL(probe_irq_mask);
 EXPORT_SYMBOL(kernel_thread);
-EXPORT_SYMBOL(disable_irq_nosync);
 EXPORT_SYMBOL(irq_desc);
 EXPORT_SYMBOL(no_irq_type);
 
-EXPORT_SYMBOL(strstr);
 EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strnlen);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strncat);
 
 /* PCI exports */
 #ifdef CONFIG_PCI
@@ -51,13 +42,8 @@ EXPORT_SYMBOL(pci_free_consistent);
 /* mem exports */
 EXPORT_SYMBOL(memchr);
 EXPORT_SYMBOL(memcpy);
-EXPORT_SYMBOL(memcpy_fromio);
-EXPORT_SYMBOL(memcpy_toio);
 EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(memset_io);
 EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memcmp);
-EXPORT_SYMBOL(memscan);
 EXPORT_SYMBOL(__copy_user);
 EXPORT_SYMBOL(boot_cpu_data);
 
@@ -124,5 +110,4 @@ EXPORT_SYMBOL(csum_partial);
 #ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
 #endif
-EXPORT_SYMBOL(consistent_sync);
 EXPORT_SYMBOL(clear_page);
diff --git a/arch/sh/mm/cache-sh7705.c b/arch/sh/mm/cache-sh7705.c
index bf94eedb0a8e..045abdf078f5 100644
--- a/arch/sh/mm/cache-sh7705.c
+++ b/arch/sh/mm/cache-sh7705.c
@@ -9,7 +9,6 @@
  * for more details.
  *
  */
-
 #include <linux/init.h>
 #include <linux/mman.h>
 #include <linux/mm.h>
@@ -25,14 +24,10 @@
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 
-/* The 32KB cache on the SH7705 suffers from the same synonym problem
- * as SH4 CPUs */
-
-#define __pte_offset(address) \
-		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
-		__pte_offset(address))
-
+/*
+ * The 32KB cache on the SH7705 suffers from the same synonym problem
+ * as SH4 CPUs
+ */
 static inline void cache_wback_all(void)
 {
 	unsigned long ways, waysize, addrstart;
@@ -73,7 +68,6 @@ void flush_icache_range(unsigned long start, unsigned long end)
 	__flush_wback_region((void *)start, end - start);
 }
 
-
 /*
  * Writeback&Invalidate the D-cache of the page
  */
@@ -128,7 +122,6 @@ static void __flush_dcache_page(unsigned long phys)
 	local_irq_restore(flags);
 }
 
-
 /*
  * Write back & invalidate the D-cache of the page.
  * (To avoid "alias" issues)
@@ -186,7 +179,8 @@ void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
  *
  * ADDRESS: Virtual Address (U0 address)
  */
-void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigned long pfn)
+void flush_cache_page(struct vm_area_struct *vma, unsigned long address,
+		      unsigned long pfn)
 {
 	__flush_dcache_page(pfn << PAGE_SHIFT);
 }
@@ -203,4 +197,3 @@ void flush_icache_page(struct vm_area_struct *vma, struct page *page)
 {
 	__flush_purge_region(page_address(page), PAGE_SIZE);
 }
-
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index dc461d2bc183..c69fd603226a 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -13,6 +13,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/hardirq.h>
+#include <linux/kprobes.h>
 #include <asm/system.h>
 #include <asm/mmu_context.h>
 #include <asm/kgdb.h>
@@ -188,15 +190,16 @@ do_sigbus:
 /*
  * Called with interrupts disabled.
  */
-asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
-			       unsigned long address)
+asmlinkage int __kprobes __do_page_fault(struct pt_regs *regs,
+					 unsigned long writeaccess,
+					 unsigned long address)
 {
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 	pte_t entry;
-	struct mm_struct *mm;
+	struct mm_struct *mm = current->mm;
 	spinlock_t *ptl;
 	int ret = 1;
 
@@ -214,10 +217,10 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 		pgd = pgd_offset_k(address);
 		mm = NULL;
 	} else {
-		if (unlikely(address >= TASK_SIZE || !(mm = current->mm)))
+		if (unlikely(address >= TASK_SIZE || !mm))
 			return 1;
 
-		pgd = pgd_offset(current->mm, address);
+		pgd = pgd_offset(mm, address);
 	}
 
 	pud = pud_offset(pgd, address);
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index 819fd0faf022..92e745341e4d 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -337,10 +337,8 @@ static int __init pmb_init(void)
 
 	return 0;
 }
-
 arch_initcall(pmb_init);
 
-#ifdef CONFIG_DEBUG_FS
 static int pmb_seq_show(struct seq_file *file, void *iter)
 {
 	int i;
@@ -399,6 +397,4 @@ static int __init pmb_debugfs_init(void)
 
 	return 0;
 }
-
 postcore_initcall(pmb_debugfs_init);
-#endif
diff --git a/include/asm-sh/dma.h b/include/asm-sh/dma.h
index e62a6d0ed932..d9daa028689f 100644
--- a/include/asm-sh/dma.h
+++ b/include/asm-sh/dma.h
@@ -89,6 +89,7 @@ struct dma_channel {
 	wait_queue_head_t wait_queue;
 
 	struct sys_device dev;
+	char *name;
 };
 
 struct dma_info {
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 7e8455b1cb44..648102e9236f 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -334,6 +334,11 @@ extern void enable_irq(unsigned int);
 extern void make_maskreg_irq(unsigned int irq);
 extern unsigned short *irq_mask_register;
 
+/*
+ * PINT IRQs
+ */
+void init_IRQ_pint(void);
+
 /*
  * Function for "on chip support modules".
  */
diff --git a/include/asm-sh/kexec.h b/include/asm-sh/kexec.h
index 9dfe59f6fcb5..a5f85e9e428d 100644
--- a/include/asm-sh/kexec.h
+++ b/include/asm-sh/kexec.h
@@ -23,6 +23,8 @@
 /* The native architecture */
 #define KEXEC_ARCH KEXEC_ARCH_SH
 
+#define MAX_NOTE_BYTES 1024
+
 #ifndef __ASSEMBLY__
 
 extern void machine_shutdown(void);
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 477422afeb0d..6c1f8fde5ac4 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -79,7 +79,7 @@ static inline void sched_cacheflush(void)
 }
 #endif
 
-static __inline__ unsigned long tas(volatile int *m)
+static inline unsigned long tas(volatile int *m)
 {
 	unsigned long retval;
 
@@ -161,7 +161,7 @@ static inline void local_irq_enable(void)
 }
 #endif
 
-static __inline__ void local_irq_disable(void)
+static inline void local_irq_disable(void)
 {
 	unsigned long __dummy;
 	__asm__ __volatile__("stc	sr, %0\n\t"
@@ -172,7 +172,7 @@ static __inline__ void local_irq_disable(void)
 			     : "memory");
 }
 
-static __inline__ void set_bl_bit(void)
+static inline void set_bl_bit(void)
 {
 	unsigned long __dummy0, __dummy1;
 
@@ -185,7 +185,7 @@ static __inline__ void set_bl_bit(void)
 			     : "memory");
 }
 
-static __inline__ void clear_bl_bit(void)
+static inline void clear_bl_bit(void)
 {
 	unsigned long __dummy0, __dummy1;
 
@@ -207,7 +207,7 @@ static __inline__ void clear_bl_bit(void)
 	(flags != 0);			\
 })
 
-static __inline__ unsigned long local_irq_save(void)
+static inline unsigned long local_irq_save(void)
 {
 	unsigned long flags, __dummy;
 
@@ -223,36 +223,10 @@ static __inline__ unsigned long local_irq_save(void)
 	return flags;
 }
 
-#ifdef DEBUG_CLI_STI
-static __inline__ void  local_irq_restore(unsigned long x)
-{
-	if ((x & 0x000000f0) != 0x000000f0)
-		local_irq_enable();
-	else {
-		unsigned long flags;
-		local_save_flags(flags);
-
-		if (flags == 0) {
-			extern void dump_stack(void);
-			printk(KERN_ERR "BUG!\n");
-			dump_stack();
-			local_irq_disable();
-		}
-	}
-}
-#else
 #define local_irq_restore(x) do {			\
 	if ((x & 0x000000f0) != 0x000000f0)		\
 		local_irq_enable();			\
 } while (0)
-#endif
-
-#define really_restore_flags(x) do {			\
-	if ((x & 0x000000f0) != 0x000000f0)		\
-		local_irq_enable();			\
-	else						\
-		local_irq_disable();			\
-} while (0)
 
 /*
  * Jump to P2 area.
-- 
cgit v1.2.3


From 2991be725260d6fec11691a6138b9d71de949956 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:07:07 +0900
Subject: sh: Fixup __strnlen_user() behaviour.

Drop TIF_USERSPACE and add addr_limit to the thread_info struct.
Subsequently, use that for address checking in strnlen_user() to
ward off bogus -EFAULTs.

Make __strnlen_user() return 0 on exception, rather than -EFAULT.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/process.c     |  3 ++
 include/asm-sh/processor.h   |  4 +++
 include/asm-sh/thread_info.h |  4 +--
 include/asm-sh/uaccess.h     | 65 +++++++++-----------------------------------
 4 files changed, 22 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index f2031314cb2b..4a32550fd7c6 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -263,6 +263,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 		unsigned long unused,
 		struct task_struct *p, struct pt_regs *regs)
 {
+	struct thread_info *ti = task_thread_info(p);
 	struct pt_regs *childregs;
 #if defined(CONFIG_SH_FPU)
 	struct task_struct *tsk = current;
@@ -277,8 +278,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 
 	if (user_mode(regs)) {
 		childregs->regs[15] = usp;
+		ti->addr_limit = USER_DS;
 	} else {
 		childregs->regs[15] = (unsigned long)task_stack_page(p) + THREAD_SIZE;
+		ti->addr_limit = KERNEL_DS;
 	}
         if (clone_flags & CLONE_SETTLS) {
 		childregs->gbr = childregs->regs[0];
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index e99aff706cf7..a22732007dd4 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -149,6 +149,10 @@ struct thread_struct {
 	union sh_fpu_union fpu;
 };
 
+typedef struct {
+	unsigned long seg;
+} mm_segment_t;
+
 /* Count of active tasks with UBC settings */
 extern int ubc_usercnt;
 
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index f64dd803a014..b986a1914472 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -21,6 +21,7 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;
 	int			preempt_count; /* 0 => preemptable, <0 => BUG */
+	mm_segment_t		addr_limit;	/* thread address space */
 	struct restart_block	restart_block;
 	__u8			supervisor_stack[0];
 };
@@ -40,6 +41,7 @@ struct thread_info {
 	.flags		= 0,			\
 	.cpu		= 0,			\
 	.preempt_count	= 1,			\
+	.addr_limit	= KERNEL_DS,		\
 	.restart_block	= {			\
 		.fn = do_no_restart_syscall,	\
 	},					\
@@ -95,7 +97,6 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		18
-#define TIF_USERSPACE		31	/* true if FS sets userspace */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
@@ -103,7 +104,6 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
-#define _TIF_USERSPACE		(1<<TIF_USERSPACE)
 
 #define _TIF_WORK_MASK		0x000000FE	/* work to do on interrupt/exception return */
 #define _TIF_ALLWORK_MASK	0x000000FF	/* work to do on any return to u-space */
diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 2cb01861e7c5..6c0014dd2ef7 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -16,21 +16,9 @@
 #include <linux/errno.h>
 #include <linux/sched.h>
 
-/*
- * NOTE: Macro/functions in this file depends on threads_info.h implementation.
- * Assumes:
- * TI_FLAGS == 8
- * TIF_USERSPACE == 31
- * USER_ADDR_LIMIT == 0x80000000
- */
-
 #define VERIFY_READ    0
 #define VERIFY_WRITE   1
 
-typedef struct {
-	unsigned int is_user_space;
-} mm_segment_t;
-
 /*
  * The fs value determines whether argument validity checking should be
  * performed or not.  If get_fs() == USER_DS, checking is performed, with
@@ -40,12 +28,14 @@ typedef struct {
  */
 
 #define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-#define segment_eq(a,b)	((a).is_user_space == (b).is_user_space)
 
-#define USER_ADDR_LIMIT	0x80000000
+#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFFUL)
+#define USER_DS		MAKE_MM_SEG(PAGE_OFFSET)
+
+#define segment_eq(a,b)	((a).seg == (b).seg)
 
-#define KERNEL_DS	MAKE_MM_SEG(0)
-#define USER_DS		MAKE_MM_SEG(1)
+#define __addr_ok(addr) \
+	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
 
 #define get_ds()	(KERNEL_DS)
 
@@ -76,31 +66,8 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 	return ((addr >= memory_start) && ((addr + size) < memory_end));
 }
 #else /* CONFIG_MMU */
-static inline mm_segment_t get_fs(void)
-{
-	return MAKE_MM_SEG(test_thread_flag(TIF_USERSPACE));
-}
-
-static inline void set_fs(mm_segment_t s)
-{
-	unsigned long ti, flag;
-	__asm__ __volatile__(
-		"stc	r7_bank, %0\n\t"
-		"mov.l	@(8,%0), %1\n\t"
-		"shal	%1\n\t"
-		"cmp/pl	%2\n\t"
-		"rotcr	%1\n\t"
-		"mov.l	%1, @(8,%0)"
-		: "=&r" (ti), "=&r" (flag)
-		: "r" (s.is_user_space)
-		: "t");
-/****
-	if (s.is_user_space)
-		set_thread_flag(TIF_USERSPACE);
-	else
-		clear_thread_flag(TIF_USERSPACE);
-****/
-}
+#define get_fs()	(current_thread_info()->addr_limit)
+#define set_fs(x)	(current_thread_info()->addr_limit = (x))
 
 /*
  * __access_ok: Check if address with size is OK or not.
@@ -108,7 +75,7 @@ static inline void set_fs(mm_segment_t s)
  * We do three checks:
  * (1) is it user space? 
  * (2) addr + size --> carry?
- * (3) addr + size >= 0x80000000  (USER_ADDR_LIMIT)
+ * (3) addr + size >= 0x80000000  (PAGE_OFFSET)
  *
  * (1) (2) (3) | RESULT
  *  0   0   0  |  ok
@@ -541,7 +508,7 @@ static __inline__ long __strnlen_user(const char __user *__s, long __n)
 		"3:\n\t"
 		"mov.l	4f, %1\n\t"
 		"jmp	@%1\n\t"
-		" mov	%5, %0\n"
+		" mov	#0, %0\n"
 		".balign 4\n"
 		"4:	.long 2b\n"
 		".previous\n"
@@ -550,26 +517,20 @@ static __inline__ long __strnlen_user(const char __user *__s, long __n)
 		"	.long 1b,3b\n"
 		".previous"
 		: "=z" (res), "=&r" (__dummy)
-		: "0" (0), "r" (__s), "r" (__n), "i" (-EFAULT)
+		: "0" (0), "r" (__s), "r" (__n)
 		: "t");
 	return res;
 }
 
 static __inline__ long strnlen_user(const char __user *s, long n)
 {
-	if (!access_ok(VERIFY_READ, s, n))
+	if (!__addr_ok(s))
 		return 0;
 	else
 		return __strnlen_user(s, n);
 }
 
-static __inline__ long strlen_user(const char __user *s)
-{
-	if (!access_ok(VERIFY_READ, s, 0))
-		return 0;
-	else
-		return __strnlen_user(s, ~0UL >> 1);
-}
+#define strlen_user(str)	strnlen_user(str, ~0UL >> 1)
 
 /*
  * The exception table consists of pairs of addresses: the first is the
-- 
cgit v1.2.3


From af514ca7d27b31e3c278e1331f0ebdb3ad385a90 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:11:32 +0900
Subject: sh: Rename rtc_get/set_time() to avoid RTC_CLASS conflict.

We have a clash with RTC_CLASS over these names, so we
change them..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/dreamcast/rtc.c    |  4 ++--
 arch/sh/boards/landisk/rtc.c      | 17 ++++-------------
 arch/sh/boards/mpc1211/rtc.c      |  4 ++--
 arch/sh/boards/sh03/rtc.c         |  9 ++++-----
 arch/sh/boards/snapgear/rtc.c     |  8 ++++----
 arch/sh/kernel/time.c             | 14 +++++++-------
 arch/sh/kernel/timers/timer-tmu.c |  6 +++---
 include/asm-sh/rtc.h              |  4 ++--
 8 files changed, 28 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boards/dreamcast/rtc.c b/arch/sh/boards/dreamcast/rtc.c
index 44420494dc22..b3a876a3b859 100644
--- a/arch/sh/boards/dreamcast/rtc.c
+++ b/arch/sh/boards/dreamcast/rtc.c
@@ -75,7 +75,7 @@ int aica_rtc_settimeofday(const time_t secs)
 
 void aica_time_init(void)
 {
-	rtc_get_time = aica_rtc_gettimeofday;
-	rtc_set_time = aica_rtc_settimeofday;
+	rtc_sh_get_time = aica_rtc_gettimeofday;
+	rtc_sh_set_time = aica_rtc_settimeofday;
 }
 
diff --git a/arch/sh/boards/landisk/rtc.c b/arch/sh/boards/landisk/rtc.c
index d666aa63303f..35ba726a0979 100644
--- a/arch/sh/boards/landisk/rtc.c
+++ b/arch/sh/boards/landisk/rtc.c
@@ -16,17 +16,9 @@
 #include <linux/time.h>
 #include <linux/delay.h>
 #include <linux/spinlock.h>
+#include <linux/bcd.h>
+#include <asm/rtc.h>
 
-#ifndef BCD_TO_BIN
-#define BCD_TO_BIN(val) ((val)=((val)&15) + ((val)>>4)*10)
-#endif
-
-#ifndef BIN_TO_BCD
-#define BIN_TO_BCD(val) ((val)=(((val)/10)<<4) + (val)%10)
-#endif
-
-extern void (*rtc_get_time) (struct timespec *);
-extern int (*rtc_set_time) (const time_t);
 extern spinlock_t rtc_lock;
 
 extern void
@@ -94,9 +86,8 @@ int landisk_rtc_settimeofday(const time_t secs)
 	return retval;
 }
 
-
 void landisk_time_init(void)
 {
-	rtc_get_time = landisk_rtc_gettimeofday;
-	rtc_set_time = landisk_rtc_settimeofday;
+	rtc_sh_get_time = landisk_rtc_gettimeofday;
+	rtc_sh_set_time = landisk_rtc_settimeofday;
 }
diff --git a/arch/sh/boards/mpc1211/rtc.c b/arch/sh/boards/mpc1211/rtc.c
index a76c655dceee..03b123a4bba4 100644
--- a/arch/sh/boards/mpc1211/rtc.c
+++ b/arch/sh/boards/mpc1211/rtc.c
@@ -130,7 +130,7 @@ int mpc1211_rtc_settimeofday(const struct timeval *tv)
 
 void mpc1211_time_init(void)
 {
-	rtc_get_time = mpc1211_rtc_gettimeofday;
-	rtc_set_time = mpc1211_rtc_settimeofday;
+	rtc_sh_get_time = mpc1211_rtc_gettimeofday;
+	rtc_sh_set_time = mpc1211_rtc_settimeofday;
 }
 
diff --git a/arch/sh/boards/sh03/rtc.c b/arch/sh/boards/sh03/rtc.c
index d609863cfe53..0a9266bb51c5 100644
--- a/arch/sh/boards/sh03/rtc.c
+++ b/arch/sh/boards/sh03/rtc.c
@@ -10,9 +10,10 @@
 #include <linux/sched.h>
 #include <linux/time.h>
 #include <linux/bcd.h>
-#include <asm/io.h>
 #include <linux/rtc.h>
 #include <linux/spinlock.h>
+#include <asm/io.h>
+#include <asm/rtc.h>
 
 #define RTC_BASE	0xb0000000
 #define RTC_SEC1	(RTC_BASE + 0)
@@ -34,8 +35,6 @@
 #define RTC_BUSY	1
 #define RTC_STOP	2
 
-extern void (*rtc_get_time)(struct timespec *);
-extern int (*rtc_set_time)(const time_t);
 extern spinlock_t rtc_lock;
 
 unsigned long get_cmos_time(void)
@@ -128,6 +127,6 @@ int sh03_rtc_settimeofday(const time_t secs)
 
 void sh03_time_init(void)
 {
-	rtc_get_time = sh03_rtc_gettimeofday;
-	rtc_set_time = sh03_rtc_settimeofday;
+	rtc_sh_get_time = sh03_rtc_gettimeofday;
+	rtc_sh_set_time = sh03_rtc_settimeofday;
 }
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index 287bc4ed3ac7..722479deb55a 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -165,11 +165,11 @@ void __init secureedge5410_rtc_init(void)
 		}
 
 	if (use_ds1302) {
-		rtc_get_time = snapgear_rtc_gettimeofday;
-		rtc_set_time = snapgear_rtc_settimeofday;
+		rtc_sh_get_time = snapgear_rtc_gettimeofday;
+		rtc_sh_set_time = snapgear_rtc_settimeofday;
 	} else {
-		rtc_get_time = sh_rtc_gettimeofday;
-		rtc_set_time = sh_rtc_settimeofday;
+		rtc_sh_get_time = sh_rtc_gettimeofday;
+		rtc_sh_set_time = sh_rtc_settimeofday;
 	}
 		
 	printk("SnapGear RTC: using %s rtc.\n", use_ds1302 ? "ds1302" : "internal");
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index c8db6ca4f9d1..8acd70bffe76 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -29,11 +29,11 @@ EXPORT_SYMBOL(rtc_lock);
 /* XXX: Can we initialize this in a routine somewhere?  Dreamcast doesn't want
  * these routines anywhere... */
 #ifdef CONFIG_SH_RTC
-void (*rtc_get_time)(struct timespec *) = sh_rtc_gettimeofday;
-int (*rtc_set_time)(const time_t) = sh_rtc_settimeofday;
+void (*rtc_sh_get_time)(struct timespec *) = sh_rtc_gettimeofday;
+int (*rtc_sh_set_time)(const time_t) = sh_rtc_settimeofday;
 #else
-void (*rtc_get_time)(struct timespec *);
-int (*rtc_set_time)(const time_t);
+void (*rtc_sh_get_time)(struct timespec *);
+int (*rtc_sh_set_time)(const time_t);
 #endif
 
 /*
@@ -135,7 +135,7 @@ void handle_timer_tick(struct pt_regs *regs)
 	    xtime.tv_sec > last_rtc_update + 660 &&
 	    (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 &&
 	    (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) {
-		if (rtc_set_time(xtime.tv_sec) == 0)
+		if (rtc_sh_set_time(xtime.tv_sec) == 0)
 			last_rtc_update = xtime.tv_sec;
 		else
 			/* do it again in 60s */
@@ -193,8 +193,8 @@ void __init time_init(void)
 
 	clk_init();
 
-	if (rtc_get_time) {
-		rtc_get_time(&xtime);
+	if (rtc_sh_get_time) {
+		rtc_sh_get_time(&xtime);
 	} else {
 		xtime.tv_sec = mktime(2000, 1, 1, 0, 0, 0);
 		xtime.tv_nsec = 0;
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index ea4bdf8cdf0d..205816fcf0da 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -132,17 +132,17 @@ static unsigned long tmu_timer_get_frequency(void)
 	ctrl_outl(0xffffffff, TMU0_TCOR);
 	ctrl_outl(0xffffffff, TMU0_TCNT);
 
-	rtc_get_time(&ts2);
+	rtc_sh_get_time(&ts2);
 
 	do {
-		rtc_get_time(&ts1);
+		rtc_sh_get_time(&ts1);
 	} while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec);
 
 	/* actually start the timer */
 	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
 
 	do {
-		rtc_get_time(&ts2);
+		rtc_sh_get_time(&ts2);
 	} while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec);
 
 	freq = 0xffffffff - ctrl_inl(TMU0_TCNT);
diff --git a/include/asm-sh/rtc.h b/include/asm-sh/rtc.h
index cea9cdf9b925..4c7be859c9c0 100644
--- a/include/asm-sh/rtc.h
+++ b/include/asm-sh/rtc.h
@@ -8,8 +8,8 @@
 extern void sh_rtc_gettimeofday(struct timespec *ts);
 extern int sh_rtc_settimeofday(const time_t secs);
 extern void (*board_time_init)(void);
-extern void (*rtc_get_time)(struct timespec *);
-extern int (*rtc_set_time)(const time_t);
+extern void (*rtc_sh_get_time)(struct timespec *);
+extern int (*rtc_sh_set_time)(const time_t);
 
 /* RCR1 Bits */
 #define RCR1_CF		0x80	/* Carry Flag             */
-- 
cgit v1.2.3


From 4b565680d16300acab0ff167e24f0ea289a6bd5d Mon Sep 17 00:00:00 2001
From: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Date: Wed, 27 Sep 2006 17:15:32 +0900
Subject: sh: math-emu support

This implements initial math-emu support, aimed primarily at SH-3.

Signed-off-by: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig              |   9 +
 arch/sh/Makefile             |   1 +
 arch/sh/kernel/traps.c       | 164 +++++++++---
 arch/sh/math-emu/Makefile    |   1 +
 arch/sh/math-emu/math.c      | 624 +++++++++++++++++++++++++++++++++++++++++++
 arch/sh/math-emu/sfp-util.h  |  72 +++++
 include/asm-sh/sfp-machine.h |  86 ++++++
 7 files changed, 918 insertions(+), 39 deletions(-)
 create mode 100644 arch/sh/math-emu/Makefile
 create mode 100644 arch/sh/math-emu/math.c
 create mode 100644 arch/sh/math-emu/sfp-util.h
 create mode 100644 include/asm-sh/sfp-machine.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index d1d724447826..07d6d699e9c2 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -339,6 +339,15 @@ config SH_FPU
 
 	  This option must be set in order to enable the FPU.
 
+config SH_FPU_EMU
+	bool "FPU emulation support"
+	depends on !SH_FPU && EXPERIMENTAL
+	default n
+	help
+	  Selecting this option will enable support for software FPU emulation.
+	  Most SH-3 users will want to say Y here, whereas most SH-4 users will
+	  want to say N.
+
 config SH_DSP
 	bool "DSP support"
 	depends on !CPU_SH4
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 859285d47a29..13b688b13e1c 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -79,6 +79,7 @@ head-y := arch/sh/kernel/head.o arch/sh/kernel/init_task.o
 LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
 core-y				+= arch/sh/kernel/ arch/sh/mm/
+core-$(CONFIG_SH_FPU_EMU)	+= arch/sh/math-emu/
 
 # Boards
 machdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se/770x
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
index 21bef1b5f991..95c810b3c97e 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -36,40 +36,15 @@
 
 #ifdef CONFIG_SH_KGDB
 #include <asm/kgdb.h>
-#define CHK_REMOTE_DEBUG(regs)                                               \
-{                                                                            \
-  if ((kgdb_debug_hook != (kgdb_debug_hook_t *) NULL) && (!user_mode(regs))) \
-  {                                                                          \
-    (*kgdb_debug_hook)(regs);                                                \
-  }                                                                          \
+#define CHK_REMOTE_DEBUG(regs)                 	\
+{       					\
+	if (kgdb_debug_hook && !user_mode(regs))\
+		(*kgdb_debug_hook)(regs);       \
 }
 #else
 #define CHK_REMOTE_DEBUG(regs)
 #endif
 
-#define DO_ERROR(trapnr, signr, str, name, tsk)				\
-asmlinkage void do_##name(unsigned long r4, unsigned long r5,		\
-			  unsigned long r6, unsigned long r7,		\
-			  struct pt_regs regs)				\
-{									\
-	unsigned long error_code;					\
- 									\
-	/* Check if it's a DSP instruction */				\
- 	if (is_dsp_inst(&regs)) {					\
-		/* Enable DSP mode, and restart instruction. */		\
-		regs.sr |= SR_DSP;					\
-		return;							\
-	}								\
-									\
-	asm volatile("stc	r2_bank, %0": "=r" (error_code));	\
-	local_irq_enable();						\
-	tsk->thread.error_code = error_code;				\
-	tsk->thread.trap_no = trapnr;					\
-        CHK_REMOTE_DEBUG(&regs);					\
-	force_sig(signr, tsk);						\
-	die_if_no_fixup(str,&regs,error_code);				\
-}
-
 #ifdef CONFIG_CPU_SH2
 #define TRAP_RESERVED_INST	4
 #define TRAP_ILLEGAL_SLOT_INST	6
@@ -575,8 +550,117 @@ int is_dsp_inst(struct pt_regs *regs)
 #define is_dsp_inst(regs)	(0)
 #endif /* CONFIG_SH_DSP */
 
-DO_ERROR(TRAP_RESERVED_INST, SIGILL, "reserved instruction", reserved_inst, current)
-DO_ERROR(TRAP_ILLEGAL_SLOT_INST, SIGILL, "illegal slot instruction", illegal_slot_inst, current)
+extern int do_fpu_inst(unsigned short, struct pt_regs*);
+
+asmlinkage void do_reserved_inst(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs regs)
+{
+	unsigned long error_code;
+	struct task_struct *tsk = current;
+
+#ifdef CONFIG_SH_FPU_EMU
+	unsigned short inst;
+	int err;
+
+	get_user(inst, (unsigned short*)regs.pc);
+
+	err = do_fpu_inst(inst, &regs);
+	if (!err) {
+		regs.pc += 2;
+		return;
+	}
+	/* not a FPU inst. */
+#endif
+
+#ifdef CONFIG_SH_DSP
+	/* Check if it's a DSP instruction */
+ 	if (is_dsp_inst(&regs)) {
+		/* Enable DSP mode, and restart instruction. */
+		regs.sr |= SR_DSP;
+		return;
+	}
+#endif
+
+	asm volatile("stc	r2_bank, %0": "=r" (error_code));
+	local_irq_enable();
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = TRAP_RESERVED_INST;
+	CHK_REMOTE_DEBUG(&regs);
+	force_sig(SIGILL, tsk);
+	die_if_no_fixup("reserved instruction", &regs, error_code);
+}
+
+#ifdef CONFIG_SH_FPU_EMU
+static int emulate_branch(unsigned short inst, struct pt_regs* regs)
+{
+	/*
+	 * bfs: 8fxx: PC+=d*2+4;
+	 * bts: 8dxx: PC+=d*2+4;
+	 * bra: axxx: PC+=D*2+4;
+	 * bsr: bxxx: PC+=D*2+4  after PR=PC+4;
+	 * braf:0x23: PC+=Rn*2+4;
+	 * bsrf:0x03: PC+=Rn*2+4 after PR=PC+4;
+	 * jmp: 4x2b: PC=Rn;
+	 * jsr: 4x0b: PC=Rn      after PR=PC+4;
+	 * rts: 000b: PC=PR;
+	 */
+	if ((inst & 0xfd00) == 0x8d00) {
+		regs->pc += SH_PC_8BIT_OFFSET(inst);
+		return 0;
+	}
+
+	if ((inst & 0xe000) == 0xa000) {
+		regs->pc += SH_PC_12BIT_OFFSET(inst);
+		return 0;
+	}
+
+	if ((inst & 0xf0df) == 0x0003) {
+		regs->pc += regs->regs[(inst & 0x0f00) >> 8] + 4;
+		return 0;
+	}
+
+	if ((inst & 0xf0df) == 0x400b) {
+		regs->pc = regs->regs[(inst & 0x0f00) >> 8];
+		return 0;
+	}
+
+	if ((inst & 0xffff) == 0x000b) {
+		regs->pc = regs->pr;
+		return 0;
+	}
+
+	return 1;
+}
+#endif
+
+asmlinkage void do_illegal_slot_inst(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs regs)
+{
+	unsigned long error_code;
+	struct task_struct *tsk = current;
+#ifdef CONFIG_SH_FPU_EMU
+	unsigned short inst;
+
+	get_user(inst, (unsigned short *)regs.pc + 1);
+	if (!do_fpu_inst(inst, &regs)) {
+		get_user(inst, (unsigned short *)regs.pc);
+		if (!emulate_branch(inst, &regs))
+			return;
+		/* fault in branch.*/
+	}
+	/* not a FPU inst. */
+#endif
+
+	asm volatile("stc	r2_bank, %0": "=r" (error_code));
+	local_irq_enable();
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = TRAP_RESERVED_INST;
+	CHK_REMOTE_DEBUG(&regs);
+	force_sig(SIGILL, tsk);
+	die_if_no_fixup("illegal slot instruction", &regs, error_code);
+}
 
 asmlinkage void do_exception_error(unsigned long r4, unsigned long r5,
 				   unsigned long r6, unsigned long r7,
@@ -634,14 +718,16 @@ void __init trap_init(void)
 	exception_handling_table[TRAP_ILLEGAL_SLOT_INST]
 		= (void *)do_illegal_slot_inst;
 
-#ifdef CONFIG_CPU_SH4
-	if (!(cpu_data->flags & CPU_HAS_FPU)) {
-		/* For SH-4 lacking an FPU, treat floating point instructions
-		   as reserved. */
-		/* entry 64 corresponds to EXPEVT=0x800 */
-		exception_handling_table[64] = (void *)do_reserved_inst;
-		exception_handling_table[65] = (void *)do_illegal_slot_inst;
-	}
+#if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SH_FPU) || \
+    defined(CONFIG_SH_FPU_EMU)
+	/*
+	 * For SH-4 lacking an FPU, treat floating point instructions as
+	 * reserved. They'll be handled in the math-emu case, or faulted on
+	 * otherwise.
+	 */
+	/* entry 64 corresponds to EXPEVT=0x800 */
+	exception_handling_table[64] = (void *)do_reserved_inst;
+	exception_handling_table[65] = (void *)do_illegal_slot_inst;
 #endif
 		
 	/* Setup VBR for boot cpu */
diff --git a/arch/sh/math-emu/Makefile b/arch/sh/math-emu/Makefile
new file mode 100644
index 000000000000..638b342c781a
--- /dev/null
+++ b/arch/sh/math-emu/Makefile
@@ -0,0 +1 @@
+obj-y	:= math.o
diff --git a/arch/sh/math-emu/math.c b/arch/sh/math-emu/math.c
new file mode 100644
index 000000000000..26b6046814fd
--- /dev/null
+++ b/arch/sh/math-emu/math.c
@@ -0,0 +1,624 @@
+/*
+ * arch/sh/math-emu/math.c
+ *
+ * Copyright (C) 2006 Takashi YOSHII <takasi-y@ops.dti.ne.jp>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/signal.h>
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/io.h>
+
+#include "sfp-util.h"
+#include <math-emu/soft-fp.h>
+#include <math-emu/single.h>
+#include <math-emu/double.h>
+
+#define	FPUL		(fregs->fpul)
+#define FPSCR		(fregs->fpscr)
+#define FPSCR_RM	(FPSCR&3)
+#define FPSCR_DN	((FPSCR>>18)&1)
+#define FPSCR_PR	((FPSCR>>19)&1)
+#define FPSCR_SZ	((FPSCR>>20)&1)
+#define FPSCR_FR	((FPSCR>>21)&1)
+#define FPSCR_MASK	0x003fffffUL
+
+#define BANK(n)	(n^(FPSCR_FR?16:0))
+#define FR	((unsigned long*)(fregs->fp_regs))
+#define FR0	(FR[BANK(0)])
+#define FRn	(FR[BANK(n)])
+#define FRm	(FR[BANK(m)])
+#define DR	((unsigned long long*)(fregs->fp_regs))
+#define DRn	(DR[BANK(n)/2])
+#define DRm	(DR[BANK(m)/2])
+
+#define XREG(n)	(n^16)
+#define XFn	(FR[BANK(XREG(n))])
+#define XFm	(FR[BANK(XREG(m))])
+#define XDn	(DR[BANK(XREG(n))/2])
+#define XDm	(DR[BANK(XREG(m))/2])
+
+#define R0	(regs->regs[0])
+#define Rn	(regs->regs[n])
+#define Rm	(regs->regs[m])
+
+#define WRITE(d,a)	({if(put_user(d, (typeof (d)*)a)) return -EFAULT;})
+#define READ(d,a)	({if(get_user(d, (typeof (d)*)a)) return -EFAULT;})
+
+#define PACK_S(r,f)	FP_PACK_SP(&r,f)
+#define UNPACK_S(f,r)	FP_UNPACK_SP(f,&r)
+#define PACK_D(r,f) \
+	{u32 t[2]; FP_PACK_DP(t,f); ((u32*)&r)[0]=t[1]; ((u32*)&r)[1]=t[0];}
+#define UNPACK_D(f,r) \
+	{u32 t[2]; t[0]=((u32*)&r)[1]; t[1]=((u32*)&r)[0]; FP_UNPACK_DP(f,t);}
+
+// 2 args instructions.
+#define BOTH_PRmn(op,x) \
+	FP_DECL_EX; if(FPSCR_PR) op(D,x,DRm,DRn); else op(S,x,FRm,FRn);
+
+#define CMP_X(SZ,R,M,N) do{ \
+	FP_DECL_##SZ(Fm); FP_DECL_##SZ(Fn); \
+	UNPACK_##SZ(Fm, M); UNPACK_##SZ(Fn, N); \
+	FP_CMP_##SZ(R, Fn, Fm, 2); }while(0)
+#define EQ_X(SZ,R,M,N) do{ \
+	FP_DECL_##SZ(Fm); FP_DECL_##SZ(Fn); \
+	UNPACK_##SZ(Fm, M); UNPACK_##SZ(Fn, N); \
+	FP_CMP_EQ_##SZ(R, Fn, Fm); }while(0)
+#define CMP(OP) ({ int r; BOTH_PRmn(OP##_X,r); r; })
+
+static int
+fcmp_gt(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	if (CMP(CMP) > 0)
+		regs->sr |= 1;
+	else
+		regs->sr &= ~1;
+
+	return 0;
+}
+
+static int
+fcmp_eq(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	if (CMP(CMP /*EQ*/) == 0)
+		regs->sr |= 1;
+	else
+		regs->sr &= ~1;
+	return 0;
+}
+
+#define ARITH_X(SZ,OP,M,N) do{ \
+	FP_DECL_##SZ(Fm); FP_DECL_##SZ(Fn); FP_DECL_##SZ(Fr); \
+	UNPACK_##SZ(Fm, M); UNPACK_##SZ(Fn, N); \
+	FP_##OP##_##SZ(Fr, Fn, Fm); \
+	PACK_##SZ(N, Fr); }while(0)
+
+static int
+fadd(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, ADD);
+	return 0;
+}
+
+static int
+fsub(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, SUB);
+	return 0;
+}
+
+static int
+fmul(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, MUL);
+	return 0;
+}
+
+static int
+fdiv(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, DIV);
+	return 0;
+}
+
+static int
+fmac(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	FP_DECL_EX;
+	FP_DECL_S(Fr);
+	FP_DECL_S(Ft);
+	FP_DECL_S(F0);
+	FP_DECL_S(Fm);
+	FP_DECL_S(Fn);
+	UNPACK_S(F0, FR0);
+	UNPACK_S(Fm, FRm);
+	UNPACK_S(Fn, FRn);
+	FP_MUL_S(Ft, Fm, F0);
+	FP_ADD_S(Fr, Fn, Ft);
+	PACK_S(FRn, Fr);
+	return 0;
+}
+
+// to process fmov's extention (odd n for DR access XD).
+#define FMOV_EXT(x) if(x&1) x+=16-1
+
+static int
+fmov_idx_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(n);
+		READ(FRn, Rm + R0 + 4);
+		n++;
+		READ(FRn, Rm + R0);
+	} else {
+		READ(FRn, Rm + R0);
+	}
+
+	return 0;
+}
+
+static int
+fmov_mem_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(n);
+		READ(FRn, Rm + 4);
+		n++;
+		READ(FRn, Rm);
+	} else {
+		READ(FRn, Rm);
+	}
+
+	return 0;
+}
+
+static int
+fmov_inc_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(n);
+		READ(FRn, Rm + 4);
+		n++;
+		READ(FRn, Rm);
+		Rm += 8;
+	} else {
+		READ(FRn, Rm);
+		Rm += 4;
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_idx(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		WRITE(FRm, Rn + R0 + 4);
+		m++;
+		WRITE(FRm, Rn + R0);
+	} else {
+		WRITE(FRm, Rn + R0);
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_mem(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		WRITE(FRm, Rn + 4);
+		m++;
+		WRITE(FRm, Rn);
+	} else {
+		WRITE(FRm, Rn);
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_dec(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		Rn -= 8;
+		WRITE(FRm, Rn + 4);
+		m++;
+		WRITE(FRm, Rn);
+	} else {
+		Rn -= 4;
+		WRITE(FRm, Rn);
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		FMOV_EXT(n);
+		DRn = DRm;
+	} else {
+		FRn = FRm;
+	}
+
+	return 0;
+}
+
+static int
+fnop_mn(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	return -EINVAL;
+}
+
+// 1 arg instructions.
+#define NOTYETn(i) static int i(struct sh_fpu_soft_struct *fregs, int n) \
+	{ printk( #i " not yet done.\n"); return 0; }
+
+NOTYETn(ftrv)
+NOTYETn(fsqrt)
+NOTYETn(fipr)
+NOTYETn(fsca)
+NOTYETn(fsrra)
+
+#define EMU_FLOAT_X(SZ,N) do { \
+	FP_DECL_##SZ(Fn); \
+	FP_FROM_INT_##SZ(Fn, FPUL, 32, int); \
+	PACK_##SZ(N, Fn); }while(0)
+static int ffloat(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+
+	if (FPSCR_PR)
+		EMU_FLOAT_X(D, DRn);
+	else
+		EMU_FLOAT_X(S, FRn);
+
+	return 0;
+}
+
+#define EMU_FTRC_X(SZ,N) do { \
+	FP_DECL_##SZ(Fn); \
+	UNPACK_##SZ(Fn, N); \
+	FP_TO_INT_##SZ(FPUL, Fn, 32, 1); }while(0)
+static int ftrc(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+
+	if (FPSCR_PR)
+		EMU_FTRC_X(D, DRn);
+	else
+		EMU_FTRC_X(S, FRn);
+
+	return 0;
+}
+
+static int fcnvsd(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+	FP_DECL_S(Fn);
+	FP_DECL_D(Fr);
+	UNPACK_S(Fn, FPUL);
+	FP_CONV(D, S, 2, 1, Fr, Fn);
+	PACK_D(DRn, Fr);
+	return 0;
+}
+
+static int fcnvds(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+	FP_DECL_D(Fn);
+	FP_DECL_S(Fr);
+	UNPACK_D(Fn, DRn);
+	FP_CONV(S, D, 1, 2, Fr, Fn);
+	PACK_S(FPUL, Fr);
+	return 0;
+}
+
+static int fxchg(struct sh_fpu_soft_struct *fregs, int flag)
+{
+	FPSCR ^= flag;
+	return 0;
+}
+
+static int fsts(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn = FPUL;
+	return 0;
+}
+
+static int flds(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FPUL = FRn;
+	return 0;
+}
+
+static int fneg(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn ^= (1 << (_FP_W_TYPE_SIZE - 1));
+	return 0;
+}
+
+static int fabs(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn &= ~(1 << (_FP_W_TYPE_SIZE - 1));
+	return 0;
+}
+
+static int fld0(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn = 0;
+	return 0;
+}
+
+static int fld1(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn = (_FP_EXPBIAS_S << (_FP_FRACBITS_S - 1));
+	return 0;
+}
+
+static int fnop_n(struct sh_fpu_soft_struct *fregs, int n)
+{
+	return -EINVAL;
+}
+
+/// Instruction decoders.
+
+static int id_fxfd(struct sh_fpu_soft_struct *, int);
+static int id_fnxd(struct sh_fpu_soft_struct *, struct pt_regs *, int, int);
+
+static int (*fnxd[])(struct sh_fpu_soft_struct *, int) = {
+	fsts, flds, ffloat, ftrc, fneg, fabs, fsqrt, fsrra,
+	fld0, fld1, fcnvsd, fcnvds, fnop_n, fnop_n, fipr, id_fxfd
+};
+
+static int (*fnmx[])(struct sh_fpu_soft_struct *, struct pt_regs *, int, int) = {
+	fadd, fsub, fmul, fdiv, fcmp_eq, fcmp_gt, fmov_idx_reg, fmov_reg_idx,
+	fmov_mem_reg, fmov_inc_reg, fmov_reg_mem, fmov_reg_dec,
+	fmov_reg_reg, id_fnxd, fmac, fnop_mn};
+
+static int id_fxfd(struct sh_fpu_soft_struct *fregs, int x)
+{
+	const int flag[] = { FPSCR_SZ, FPSCR_PR, FPSCR_FR, 0 };
+	switch (x & 3) {
+	case 3:
+		fxchg(fregs, flag[x >> 2]);
+		break;
+	case 1:
+		ftrv(fregs, x - 1);
+		break;
+	default:
+		fsca(fregs, x);
+	}
+	return 0;
+}
+
+static int
+id_fnxd(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int x, int n)
+{
+	return (fnxd[x])(fregs, n);
+}
+
+static int
+id_fnmx(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, u16 code)
+{
+	int n = (code >> 8) & 0xf, m = (code >> 4) & 0xf, x = code & 0xf;
+	return (fnmx[x])(fregs, regs, m, n);
+}
+
+static int
+id_sys(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, u16 code)
+{
+	int n = ((code >> 8) & 0xf);
+	unsigned long *reg = (code & 0x0010) ? &FPUL : &FPSCR;
+
+	switch (code & 0xf0ff) {
+	case 0x005a:
+	case 0x006a:
+		Rn = *reg;
+		break;
+	case 0x405a:
+	case 0x406a:
+		*reg = Rn;
+		break;
+	case 0x4052:
+	case 0x4062:
+		Rn -= 4;
+		WRITE(*reg, Rn);
+		break;
+	case 0x4056:
+	case 0x4066:
+		READ(*reg, Rn);
+		Rn += 4;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int fpu_emulate(u16 code, struct sh_fpu_soft_struct *fregs, struct pt_regs *regs)
+{
+	if ((code & 0xf000) == 0xf000)
+		return id_fnmx(fregs, regs, code);
+	else
+		return id_sys(fregs, regs, code);
+}
+
+/**
+ *	denormal_to_double - Given denormalized float number,
+ *	                     store double float
+ *
+ *	@fpu: Pointer to sh_fpu_hard structure
+ *	@n: Index to FP register
+ */
+static void denormal_to_double(struct sh_fpu_hard_struct *fpu, int n)
+{
+	unsigned long du, dl;
+	unsigned long x = fpu->fpul;
+	int exp = 1023 - 126;
+
+	if (x != 0 && (x & 0x7f800000) == 0) {
+		du = (x & 0x80000000);
+		while ((x & 0x00800000) == 0) {
+			x <<= 1;
+			exp--;
+		}
+		x &= 0x007fffff;
+		du |= (exp << 20) | (x >> 3);
+		dl = x << 29;
+
+		fpu->fp_regs[n] = du;
+		fpu->fp_regs[n+1] = dl;
+	}
+}
+
+/**
+ *	ieee_fpe_handler - Handle denormalized number exception
+ *
+ *	@regs: Pointer to register structure
+ *
+ *	Returns 1 when it's handled (should not cause exception).
+ */
+static int ieee_fpe_handler(struct pt_regs *regs)
+{
+	unsigned short insn = *(unsigned short *)regs->pc;
+	unsigned short finsn;
+	unsigned long nextpc;
+	int nib[4] = {
+		(insn >> 12) & 0xf,
+		(insn >> 8) & 0xf,
+		(insn >> 4) & 0xf,
+		insn & 0xf};
+
+	if (nib[0] == 0xb ||
+	    (nib[0] == 0x4 && nib[2] == 0x0 && nib[3] == 0xb)) /* bsr & jsr */
+		regs->pr = regs->pc + 4;
+
+	if (nib[0] == 0xa || nib[0] == 0xb) { /* bra & bsr */
+		nextpc = regs->pc + 4 + ((short) ((insn & 0xfff) << 4) >> 3);
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xd) { /* bt/s */
+		if (regs->sr & 1)
+			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+		else
+			nextpc = regs->pc + 4;
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xf) { /* bf/s */
+		if (regs->sr & 1)
+			nextpc = regs->pc + 4;
+		else
+			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x4 && nib[3] == 0xb &&
+		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* jmp & jsr */
+		nextpc = regs->regs[nib[1]];
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x0 && nib[3] == 0x3 &&
+		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* braf & bsrf */
+		nextpc = regs->pc + 4 + regs->regs[nib[1]];
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (insn == 0x000b) { /* rts */
+		nextpc = regs->pr;
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else {
+		nextpc = regs->pc + 2;
+		finsn = insn;
+	}
+
+	if ((finsn & 0xf1ff) == 0xf0ad) { /* fcnvsd */
+		struct task_struct *tsk = current;
+
+		if ((tsk->thread.fpu.hard.fpscr & (1 << 17))) {
+			/* FPU error */
+			denormal_to_double (&tsk->thread.fpu.hard,
+					    (finsn >> 8) & 0xf);
+			tsk->thread.fpu.hard.fpscr &=
+				~(FPSCR_CAUSE_MASK | FPSCR_FLAG_MASK);
+			set_tsk_thread_flag(tsk, TIF_USEDFPU);
+		} else {
+			tsk->thread.trap_no = 11;
+			tsk->thread.error_code = 0;
+			force_sig(SIGFPE, tsk);
+		}
+
+		regs->pc = nextpc;
+		return 1;
+	}
+
+	return 0;
+}
+
+asmlinkage void do_fpu_error(unsigned long r4, unsigned long r5,
+			     unsigned long r6, unsigned long r7,
+			     struct pt_regs regs)
+{
+	struct task_struct *tsk = current;
+
+	if (ieee_fpe_handler (&regs))
+		return;
+
+	regs.pc += 2;
+	tsk->thread.trap_no = 11;
+	tsk->thread.error_code = 0;
+	force_sig(SIGFPE, tsk);
+}
+
+/**
+ * fpu_init - Initialize FPU registers
+ * @fpu: Pointer to software emulated FPU registers.
+ */
+static void fpu_init(struct sh_fpu_soft_struct *fpu)
+{
+	int i;
+
+	fpu->fpscr = FPSCR_INIT;
+	fpu->fpul = 0;
+
+	for (i = 0; i < 16; i++) {
+		fpu->fp_regs[i] = 0;
+		fpu->xfp_regs[i]= 0;
+	}
+}
+
+/**
+ * do_fpu_inst - Handle reserved instructions for FPU emulation
+ * @inst: instruction code.
+ * @regs: registers on stack.
+ */
+int do_fpu_inst(unsigned short inst, struct pt_regs *regs)
+{
+	struct task_struct *tsk = current;
+	struct sh_fpu_soft_struct *fpu = &(tsk->thread.fpu.soft);
+
+	if (!test_tsk_thread_flag(tsk, TIF_USEDFPU)) {
+		/* initialize once. */
+		fpu_init(fpu);
+		set_tsk_thread_flag(tsk, TIF_USEDFPU);
+	}
+
+	return fpu_emulate(inst, fpu, regs);
+}
diff --git a/arch/sh/math-emu/sfp-util.h b/arch/sh/math-emu/sfp-util.h
new file mode 100644
index 000000000000..8ae1bd310ad0
--- /dev/null
+++ b/arch/sh/math-emu/sfp-util.h
@@ -0,0 +1,72 @@
+/*
+ * These are copied from glibc/stdlib/longlong.h
+ */
+
+#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
+  do {                                                                  \
+    UWtype __x;                                                         \
+    __x = (al) + (bl);                                                  \
+    (sh) = (ah) + (bh) + (__x < (al));                                  \
+    (sl) = __x;                                                         \
+  } while (0)
+
+#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
+  do {                                                                  \
+    UWtype __x;                                                         \
+    __x = (al) - (bl);                                                  \
+    (sh) = (ah) - (bh) - (__x > (al));                                  \
+    (sl) = __x;                                                         \
+  } while (0)
+
+#define umul_ppmm(w1, w0, u, v) \
+  __asm__ ("dmulu.l %2,%3\n\tsts    macl,%1\n\tsts  mach,%0"	\
+	: "=r" ((u32)(w1)), "=r" ((u32)(w0))	\
+	:  "r" ((u32)(u)),   "r" ((u32)(v))	\
+	: "macl", "mach")
+
+#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
+#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
+#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))
+
+#define udiv_qrnnd(q, r, n1, n0, d) \
+  do {									\
+    UWtype __d1, __d0, __q1, __q0;					\
+    UWtype __r1, __r0, __m;						\
+    __d1 = __ll_highpart (d);						\
+    __d0 = __ll_lowpart (d);						\
+									\
+    __r1 = (n1) % __d1;							\
+    __q1 = (n1) / __d1;							\
+    __m = (UWtype) __q1 * __d0;						\
+    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
+    if (__r1 < __m)							\
+      {									\
+	__q1--, __r1 += (d);						\
+	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
+	  if (__r1 < __m)						\
+	    __q1--, __r1 += (d);					\
+      }									\
+    __r1 -= __m;							\
+									\
+    __r0 = __r1 % __d1;							\
+    __q0 = __r1 / __d1;							\
+    __m = (UWtype) __q0 * __d0;						\
+    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
+    if (__r0 < __m)							\
+      {									\
+	__q0--, __r0 += (d);						\
+	if (__r0 >= (d))						\
+	  if (__r0 < __m)						\
+	    __q0--, __r0 += (d);					\
+      }									\
+    __r0 -= __m;							\
+									\
+    (q) = (UWtype) __q1 * __ll_B | __q0;				\
+    (r) = __r0;								\
+  } while (0)
+
+#define abort()	return 0
+
+#define __BYTE_ORDER __LITTLE_ENDIAN
+
+
diff --git a/include/asm-sh/sfp-machine.h b/include/asm-sh/sfp-machine.h
new file mode 100644
index 000000000000..8a6399a8cfe0
--- /dev/null
+++ b/include/asm-sh/sfp-machine.h
@@ -0,0 +1,86 @@
+/* Machine-dependent software floating-point definitions.
+   SuperH kernel version.
+   Copyright (C) 1997,1998,1999 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+   Contributed by Richard Henderson (rth@cygnus.com),
+		  Jakub Jelinek (jj@ultra.linux.cz),
+		  David S. Miller (davem@redhat.com) and
+		  Peter Maydell (pmaydell@chiark.greenend.org.uk).
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Library General Public License as
+   published by the Free Software Foundation; either version 2 of the
+   License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Library General Public License for more details.
+
+   You should have received a copy of the GNU Library General Public
+   License along with the GNU C Library; see the file COPYING.LIB.  If
+   not, write to the Free Software Foundation, Inc.,
+   59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
+
+#ifndef _SFP_MACHINE_H
+#define _SFP_MACHINE_H
+
+#include <linux/config.h>
+
+#define _FP_W_TYPE_SIZE		32
+#define _FP_W_TYPE		unsigned long
+#define _FP_WS_TYPE		signed long
+#define _FP_I_TYPE		long
+
+#define _FP_MUL_MEAT_S(R,X,Y)					\
+  _FP_MUL_MEAT_1_wide(_FP_WFRACBITS_S,R,X,Y,umul_ppmm)
+#define _FP_MUL_MEAT_D(R,X,Y)					\
+  _FP_MUL_MEAT_2_wide(_FP_WFRACBITS_D,R,X,Y,umul_ppmm)
+#define _FP_MUL_MEAT_Q(R,X,Y)					\
+  _FP_MUL_MEAT_4_wide(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm)
+
+#define _FP_DIV_MEAT_S(R,X,Y)	_FP_DIV_MEAT_1_udiv(S,R,X,Y)
+#define _FP_DIV_MEAT_D(R,X,Y)	_FP_DIV_MEAT_2_udiv(D,R,X,Y)
+#define _FP_DIV_MEAT_Q(R,X,Y)	_FP_DIV_MEAT_4_udiv(Q,R,X,Y)
+
+#define _FP_NANFRAC_S		((_FP_QNANBIT_S << 1) - 1)
+#define _FP_NANFRAC_D		((_FP_QNANBIT_D << 1) - 1), -1
+#define _FP_NANFRAC_Q		((_FP_QNANBIT_Q << 1) - 1), -1, -1, -1
+#define _FP_NANSIGN_S		0
+#define _FP_NANSIGN_D		0
+#define _FP_NANSIGN_Q		0
+
+#define _FP_KEEPNANFRACP 1
+
+/*
+ * If one NaN is signaling and the other is not,
+ * we choose that one, otherwise we choose X.
+ */
+#define _FP_CHOOSENAN(fs, wc, R, X, Y, OP)                      \
+  do {                                                          \
+    if ((_FP_FRAC_HIGH_RAW_##fs(X) & _FP_QNANBIT_##fs)          \
+        && !(_FP_FRAC_HIGH_RAW_##fs(Y) & _FP_QNANBIT_##fs))     \
+      {                                                         \
+        R##_s = Y##_s;                                          \
+        _FP_FRAC_COPY_##wc(R,Y);                                \
+      }                                                         \
+    else                                                        \
+      {                                                         \
+        R##_s = X##_s;                                          \
+        _FP_FRAC_COPY_##wc(R,X);                                \
+      }                                                         \
+    R##_c = FP_CLS_NAN;                                         \
+  } while (0)
+
+//#define FP_ROUNDMODE		FPSCR_RM
+#define FP_DENORM_ZERO		1/*FPSCR_DN*/
+
+/* Exception flags. */
+#define FP_EX_INVALID		(1<<4)
+#define FP_EX_DIVZERO		(1<<3)
+#define FP_EX_OVERFLOW		(1<<2)
+#define FP_EX_UNDERFLOW		(1<<1)
+#define FP_EX_INEXACT		(1<<0)
+
+#endif
+
-- 
cgit v1.2.3


From adf1890b0cd63f754b2171b73e4d845c0950d407 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:17:27 +0900
Subject: sh: Move voyagergx_reg.h to a more sensible place.

Other boards require this as well, so move it out of the
rts7751r2d directory.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/rts7751r2d/setup.c |   2 +-
 arch/sh/cchips/voyagergx/irq.c            |   2 +-
 arch/sh/cchips/voyagergx/setup.c          |   2 +-
 include/asm-sh/rts7751r2d/voyagergx_reg.h | 313 ------------------------------
 include/asm-sh/voyagergx.h                | 313 ++++++++++++++++++++++++++++++
 5 files changed, 316 insertions(+), 316 deletions(-)
 delete mode 100644 include/asm-sh/rts7751r2d/voyagergx_reg.h
 create mode 100644 include/asm-sh/voyagergx.h

(limited to 'include')

diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index f6ce7021feba..bb56c018f2ee 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -14,7 +14,7 @@
 #include <linux/pm.h>
 #include <asm/io.h>
 #include <asm/mach/rts7751r2d.h>
-#include <asm/mach/voyagergx_reg.h>
+#include <asm/voyagergx.h>
 
 static struct plat_serial8250_port uart_platform_data[] = {
 	{
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
index 036e8f900013..4cfa7c990553 100644
--- a/arch/sh/cchips/voyagergx/irq.c
+++ b/arch/sh/cchips/voyagergx/irq.c
@@ -32,7 +32,7 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/mach/voyagergx_reg.h>
+#include <asm/voyagergx.h>
 
 static void disable_voyagergx_irq(unsigned int irq)
 {
diff --git a/arch/sh/cchips/voyagergx/setup.c b/arch/sh/cchips/voyagergx/setup.c
index 3a4cfaa9a214..66b2fedd7ad9 100644
--- a/arch/sh/cchips/voyagergx/setup.c
+++ b/arch/sh/cchips/voyagergx/setup.c
@@ -13,7 +13,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <asm/io.h>
-#include <asm/mach/voyagergx_reg.h>
+#include <asm/voyagergx.h>
 
 static int __init setup_voyagergx(void)
 {
diff --git a/include/asm-sh/rts7751r2d/voyagergx_reg.h b/include/asm-sh/rts7751r2d/voyagergx_reg.h
deleted file mode 100644
index f031b5d6cf54..000000000000
--- a/include/asm-sh/rts7751r2d/voyagergx_reg.h
+++ /dev/null
@@ -1,313 +0,0 @@
-/* -------------------------------------------------------------------- */
-/* voyagergx_reg.h                                                      */
-/* -------------------------------------------------------------------- */
-/*  This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-
-    Copyright 2003 (c) Lineo uSolutions,Inc.
-*/
-/* -------------------------------------------------------------------- */
-
-#ifndef _VOYAGER_GX_REG_H
-#define _VOYAGER_GX_REG_H
-
-#define VOYAGER_BASE			0xb3e00000
-#define VOYAGER_USBH_BASE		(0x40000 + VOYAGER_BASE)
-#define VOYAGER_UART_BASE		(0x30000 + VOYAGER_BASE)
-#define	VOYAGER_AC97_BASE		(0xa0000 + VOYAGER_BASE)
-
-#define VOYAGER_IRQ_NUM			32
-#define VOYAGER_IRQ_BASE		50
-#define VOYAGER_USBH_IRQ		VOYAGER_IRQ_BASE + 6
-#define VOYAGER_8051_IRQ		VOYAGER_IRQ_BASE + 10
-#define VOYAGER_UART0_IRQ		VOYAGER_IRQ_BASE + 12
-#define VOYAGER_UART1_IRQ		VOYAGER_IRQ_BASE + 13
-#define	VOYAGER_AC97_IRQ		VOYAGER_IRQ_BASE + 17
-
-/* ----- MISC controle  register ------------------------------ */
-#define MISC_CTRL			(0x000004 + VOYAGER_BASE)
-#define MISC_CTRL_USBCLK_48		(3 << 28)
-#define MISC_CTRL_USBCLK_96		(2 << 28)
-#define MISC_CTRL_USBCLK_CRYSTAL	(1 << 28)
-
-/* ----- GPIO[31:0] register --------------------------------- */
-#define GPIO_MUX_LOW			(0x000008 + VOYAGER_BASE)
-#define GPIO_MUX_LOW_AC97		0x1F000000
-#define GPIO_MUX_LOW_8051		0x0000ffff
-#define GPIO_MUX_LOW_PWM		(1 << 29)
-
-/* ----- GPIO[63:32] register --------------------------------- */
-#define GPIO_MUX_HIGH			(0x00000C + VOYAGER_BASE)
-
-/* ----- DRAM controle  register ------------------------------- */
-#define DRAM_CTRL			(0x000010 + VOYAGER_BASE)
-#define DRAM_CTRL_EMBEDDED		(1 << 31)
-#define DRAM_CTRL_CPU_BURST_1		(0 << 28)
-#define DRAM_CTRL_CPU_BURST_2		(1 << 28)
-#define DRAM_CTRL_CPU_BURST_4		(2 << 28)
-#define DRAM_CTRL_CPU_BURST_8		(3 << 28)
-#define DRAM_CTRL_CPU_CAS_LATENCY	(1 << 27)
-#define DRAM_CTRL_CPU_SIZE_2		(0 << 24)
-#define DRAM_CTRL_CPU_SIZE_4		(1 << 24)
-#define DRAM_CTRL_CPU_SIZE_64		(4 << 24)
-#define DRAM_CTRL_CPU_SIZE_32		(5 << 24)
-#define DRAM_CTRL_CPU_SIZE_16		(6 << 24)
-#define DRAM_CTRL_CPU_SIZE_8		(7 << 24)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_1024	(0 << 22)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_512	(2 << 22)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_256	(3 << 22)
-#define DRAM_CTRL_CPU_ACTIVE_PRECHARGE	(1 << 21)
-#define DRAM_CTRL_CPU_RESET		(1 << 20)
-#define DRAM_CTRL_CPU_BANKS		(1 << 19)
-#define DRAM_CTRL_CPU_WRITE_PRECHARGE	(1 << 18)
-#define DRAM_CTRL_BLOCK_WRITE		(1 << 17)
-#define DRAM_CTRL_REFRESH_COMMAND	(1 << 16)
-#define DRAM_CTRL_SIZE_4		(0 << 13)
-#define DRAM_CTRL_SIZE_8		(1 << 13)
-#define DRAM_CTRL_SIZE_16		(2 << 13)
-#define DRAM_CTRL_SIZE_32		(3 << 13)
-#define DRAM_CTRL_SIZE_64		(4 << 13)
-#define DRAM_CTRL_SIZE_2		(5 << 13)
-#define DRAM_CTRL_COLUMN_SIZE_256	(0 << 11)
-#define DRAM_CTRL_COLUMN_SIZE_512	(2 << 11)
-#define DRAM_CTRL_COLUMN_SIZE_1024	(3 << 11)
-#define DRAM_CTRL_BLOCK_WRITE_TIME	(1 << 10)
-#define DRAM_CTRL_BLOCK_WRITE_PRECHARGE	(1 << 9)
-#define DRAM_CTRL_ACTIVE_PRECHARGE	(1 << 8)
-#define DRAM_CTRL_RESET			(1 << 7)
-#define DRAM_CTRL_REMAIN_ACTIVE		(1 << 6)
-#define DRAM_CTRL_BANKS			(1 << 1)
-#define DRAM_CTRL_WRITE_PRECHARGE	(1 << 0)
-
-/* ----- Arvitration control register -------------------------- */
-#define ARBITRATION_CTRL		(0x000014 + VOYAGER_BASE)
-#define ARBITRATION_CTRL_CPUMEM		(1 << 29)
-#define ARBITRATION_CTRL_INTMEM		(1 << 28)
-#define ARBITRATION_CTRL_USB_OFF	(0 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_1	(1 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_2	(2 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_3	(3 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_4	(4 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_5	(5 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_6	(6 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_7	(7 << 24)
-#define ARBITRATION_CTRL_PANEL_OFF	(0 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_1	(1 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_2	(2 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_3	(3 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_4	(4 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_5	(5 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_6	(6 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_7	(7 << 20)
-#define ARBITRATION_CTRL_ZVPORT_OFF	(0 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_1	(1 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_2	(2 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_3	(3 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_4	(4 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_5	(5 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_6	(6 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_7	(7 << 16)
-#define ARBITRATION_CTRL_CMD_INTPR_OFF	(0 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_1	(1 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_2	(2 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_3	(3 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_4	(4 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_5	(5 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_6	(6 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_7	(7 << 12)
-#define ARBITRATION_CTRL_DMA_OFF	(0 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_1	(1 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_2	(2 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_3	(3 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_4	(4 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_5	(5 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_6	(6 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_7	(7 << 8)
-#define ARBITRATION_CTRL_VIDEO_OFF	(0 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_1	(1 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_2	(2 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_3	(3 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_4	(4 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_5	(5 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_6	(6 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_7	(7 << 4)
-#define ARBITRATION_CTRL_CRT_OFF	(0 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_1	(1 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_2	(2 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_3	(3 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_4	(4 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_5	(5 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_6	(6 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_7	(7 << 0)
-
-/* ----- Command list status register -------------------------- */
-#define CMD_INTPR_STATUS		(0x000024 + VOYAGER_BASE)
-
-/* ----- Interrupt status register ----------------------------- */
-#define INT_STATUS			(0x00002c + VOYAGER_BASE)
-#define INT_STATUS_UH			(1 << 6)
-#define INT_STATUS_MC			(1 << 10)
-#define INT_STATUS_U0			(1 << 12)
-#define INT_STATUS_U1			(1 << 13)
-#define	INT_STATUS_AC			(1 << 17)
-
-/* ----- Interrupt mask register ------------------------------ */
-#define VOYAGER_INT_MASK		(0x000030 + VOYAGER_BASE)
-#define VOYAGER_INT_MASK_AC		(1 << 17)
-
-/* ----- Current Gate register ---------------------------------*/
-#define CURRENT_GATE			(0x000038 + VOYAGER_BASE)
-
-/* ----- Power mode 0 gate register --------------------------- */
-#define POWER_MODE0_GATE		(0x000040 + VOYAGER_BASE)
-#define POWER_MODE0_GATE_G		(1 << 6)
-#define POWER_MODE0_GATE_U0		(1 << 7)
-#define POWER_MODE0_GATE_U1		(1 << 8)
-#define POWER_MODE0_GATE_UH		(1 << 11)
-#define	POWER_MODE0_GATE_AC		(1 << 18)
-
-/* ----- Power mode 1 gate register --------------------------- */
-#define POWER_MODE1_GATE		(0x000048 + VOYAGER_BASE)
-#define POWER_MODE1_GATE_G		(1 << 6)
-#define POWER_MODE1_GATE_U0		(1 << 7)
-#define POWER_MODE1_GATE_U1		(1 << 8)
-#define POWER_MODE1_GATE_UH		(1 << 11)
-#define	POWER_MODE1_GATE_AC		(1 << 18)
-
-/* ----- Power mode 0 clock register -------------------------- */
-#define POWER_MODE0_CLOCK		(0x000044 + VOYAGER_BASE)
-
-/* ----- Power mode 1 clock register -------------------------- */
-#define POWER_MODE1_CLOCK		(0x00004C + VOYAGER_BASE)
-
-/* ----- Power mode controll register ------------------------- */
-#define POWER_MODE_CTRL			(0x000054 + VOYAGER_BASE)
-
-/* ----- Miscellaneous Timing register ------------------------ */
-#define SYSTEM_DRAM_CTRL		(0x000068 + VOYAGER_BASE)
-
-/* ----- PWM register ------------------------------------------*/
-#define PWM_0				(0x010020 + VOYAGER_BASE)
-#define PWM_0_HC(x)			(((x)&0x0fff)<<20)
-#define PWM_0_LC(x)			(((x)&0x0fff)<<8 )
-#define PWM_0_CLK_DEV(x)		(((x)&0x000f)<<4 )
-#define PWM_0_EN			(1<<0)
-
-/* ----- I2C register ----------------------------------------- */
-#define I2C_BYTECOUNT			(0x010040 + VOYAGER_BASE)
-#define I2C_CONTROL			(0x010041 + VOYAGER_BASE)
-#define I2C_STATUS			(0x010042 + VOYAGER_BASE)
-#define I2C_RESET			(0x010042 + VOYAGER_BASE)
-#define I2C_SADDRESS			(0x010043 + VOYAGER_BASE)
-#define I2C_DATA			(0x010044 + VOYAGER_BASE)
-
-/* ----- Controle register bits ----------------------------------------- */
-#define I2C_CONTROL_E			(1 << 0)
-#define I2C_CONTROL_MODE		(1 << 1)
-#define I2C_CONTROL_STATUS		(1 << 2)
-#define I2C_CONTROL_INT			(1 << 4)
-#define I2C_CONTROL_INTACK		(1 << 5)
-#define I2C_CONTROL_REPEAT		(1 << 6)
-
-/* ----- Status register bits ----------------------------------------- */
-#define I2C_STATUS_BUSY			(1 << 0)
-#define I2C_STATUS_ACK			(1 << 1)
-#define I2C_STATUS_ERROR		(1 << 2)
-#define I2C_STATUS_COMPLETE		(1 << 3)
-
-/* ----- Reset register  ---------------------------------------------- */
-#define I2C_RESET_ERROR			(1 << 2)
-
-/* ----- transmission frequencies ------------------------------------- */
-#define I2C_SADDRESS_SELECT		(1 << 0)
-
-/* ----- Display Controll register ----------------------------------------- */
-#define PANEL_DISPLAY_CTRL		(0x080000 + VOYAGER_BASE)
-#define PANEL_DISPLAY_CTRL_BIAS         (1<<26)
-#define PANEL_PAN_CTRL			(0x080004 + VOYAGER_BASE)
-#define PANEL_COLOR_KEY			(0x080008 + VOYAGER_BASE)
-#define PANEL_FB_ADDRESS		(0x08000C + VOYAGER_BASE)
-#define PANEL_FB_WIDTH			(0x080010 + VOYAGER_BASE)
-#define PANEL_WINDOW_WIDTH		(0x080014 + VOYAGER_BASE)
-#define PANEL_WINDOW_HEIGHT		(0x080018 + VOYAGER_BASE)
-#define PANEL_PLANE_TL			(0x08001C + VOYAGER_BASE)
-#define PANEL_PLANE_BR			(0x080020 + VOYAGER_BASE)
-#define PANEL_HORIZONTAL_TOTAL		(0x080024 + VOYAGER_BASE)
-#define PANEL_HORIZONTAL_SYNC		(0x080028 + VOYAGER_BASE)
-#define PANEL_VERTICAL_TOTAL		(0x08002C + VOYAGER_BASE)
-#define PANEL_VERTICAL_SYNC		(0x080030 + VOYAGER_BASE)
-#define PANEL_CURRENT_LINE		(0x080034 + VOYAGER_BASE)
-#define VIDEO_DISPLAY_CTRL		(0x080040 + VOYAGER_BASE)
-#define VIDEO_FB_0_ADDRESS		(0x080044 + VOYAGER_BASE)
-#define VIDEO_FB_WIDTH			(0x080048 + VOYAGER_BASE)
-#define VIDEO_FB_0_LAST_ADDRESS		(0x08004C + VOYAGER_BASE)
-#define VIDEO_PLANE_TL			(0x080050 + VOYAGER_BASE)
-#define VIDEO_PLANE_BR			(0x080054 + VOYAGER_BASE)
-#define VIDEO_SCALE			(0x080058 + VOYAGER_BASE)
-#define VIDEO_INITIAL_SCALE		(0x08005C + VOYAGER_BASE)
-#define VIDEO_YUV_CONSTANTS		(0x080060 + VOYAGER_BASE)
-#define VIDEO_FB_1_ADDRESS		(0x080064 + VOYAGER_BASE)
-#define VIDEO_FB_1_LAST_ADDRESS		(0x080068 + VOYAGER_BASE)
-#define VIDEO_ALPHA_DISPLAY_CTRL	(0x080080 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_ADDRESS		(0x080084 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_WIDTH		(0x080088 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_LAST_ADDRESS	(0x08008C + VOYAGER_BASE)
-#define VIDEO_ALPHA_PLANE_TL		(0x080090 + VOYAGER_BASE)
-#define VIDEO_ALPHA_PLANE_BR		(0x080094 + VOYAGER_BASE)
-#define VIDEO_ALPHA_SCALE		(0x080098 + VOYAGER_BASE)
-#define VIDEO_ALPHA_INITIAL_SCALE	(0x08009C + VOYAGER_BASE)
-#define VIDEO_ALPHA_CHROMA_KEY		(0x0800A0 + VOYAGER_BASE)
-#define PANEL_HWC_ADDRESS		(0x0800F0 + VOYAGER_BASE)
-#define PANEL_HWC_LOCATION		(0x0800F4 + VOYAGER_BASE)
-#define PANEL_HWC_COLOR_12		(0x0800F8 + VOYAGER_BASE)
-#define PANEL_HWC_COLOR_3		(0x0800FC + VOYAGER_BASE)
-#define ALPHA_DISPLAY_CTRL		(0x080100 + VOYAGER_BASE)
-#define ALPHA_FB_ADDRESS		(0x080104 + VOYAGER_BASE)
-#define ALPHA_FB_WIDTH			(0x080108 + VOYAGER_BASE)
-#define ALPHA_PLANE_TL			(0x08010C + VOYAGER_BASE)
-#define ALPHA_PLANE_BR			(0x080110 + VOYAGER_BASE)
-#define ALPHA_CHROMA_KEY		(0x080114 + VOYAGER_BASE)
-#define CRT_DISPLAY_CTRL		(0x080200 + VOYAGER_BASE)
-#define CRT_FB_ADDRESS			(0x080204 + VOYAGER_BASE)
-#define CRT_FB_WIDTH			(0x080208 + VOYAGER_BASE)
-#define CRT_HORIZONTAL_TOTAL		(0x08020C + VOYAGER_BASE)
-#define CRT_HORIZONTAL_SYNC		(0x080210 + VOYAGER_BASE)
-#define CRT_VERTICAL_TOTAL		(0x080214 + VOYAGER_BASE)
-#define CRT_VERTICAL_SYNC		(0x080218 + VOYAGER_BASE)
-#define CRT_SIGNATURE_ANALYZER		(0x08021C + VOYAGER_BASE)
-#define CRT_CURRENT_LINE		(0x080220 + VOYAGER_BASE)
-#define CRT_MONITOR_DETECT		(0x080224 + VOYAGER_BASE)
-#define CRT_HWC_ADDRESS			(0x080230 + VOYAGER_BASE)
-#define CRT_HWC_LOCATION		(0x080234 + VOYAGER_BASE)
-#define CRT_HWC_COLOR_12		(0x080238 + VOYAGER_BASE)
-#define CRT_HWC_COLOR_3			(0x08023C + VOYAGER_BASE)
-#define CRT_PALETTE_RAM			(0x080400 + VOYAGER_BASE)
-#define PANEL_PALETTE_RAM		(0x080800 + VOYAGER_BASE)
-#define VIDEO_PALETTE_RAM		(0x080C00 + VOYAGER_BASE)
-
-/* ----- 8051 Controle register ----------------------------------------- */
-#define VOYAGER_8051_BASE		(0x000c0000 + VOYAGER_BASE)
-#define VOYAGER_8051_RESET		(0x000b0000 + VOYAGER_BASE)
-#define VOYAGER_8051_SELECT		(0x000b0004 + VOYAGER_BASE)
-#define VOYAGER_8051_CPU_INT		(0x000b000c + VOYAGER_BASE)
-
-/* ----- AC97 Controle register ----------------------------------------- */
-#define AC97_TX_SLOT0			(0x00000000 + VOYAGER_AC97_BASE)
-#define AC97_CONTROL_STATUS		(0x00000080 + VOYAGER_AC97_BASE)
-#define AC97C_READ			(1 << 19)
-#define AC97C_WD_BIT			(1 << 2)
-#define AC97C_INDEX_MASK		0x7f
-/* -------------------------------------------------------------------- */
-
-#endif /* _VOYAGER_GX_REG_H */
diff --git a/include/asm-sh/voyagergx.h b/include/asm-sh/voyagergx.h
new file mode 100644
index 000000000000..99b0807d1c9f
--- /dev/null
+++ b/include/asm-sh/voyagergx.h
@@ -0,0 +1,313 @@
+/* -------------------------------------------------------------------- */
+/* voyagergx.h	                                                      */
+/* -------------------------------------------------------------------- */
+/*  This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+    Copyright 2003 (c) Lineo uSolutions,Inc.
+*/
+/* -------------------------------------------------------------------- */
+
+#ifndef _VOYAGER_GX_REG_H
+#define _VOYAGER_GX_REG_H
+
+#define VOYAGER_BASE			0xb3e00000
+#define VOYAGER_USBH_BASE		(0x40000 + VOYAGER_BASE)
+#define VOYAGER_UART_BASE		(0x30000 + VOYAGER_BASE)
+#define	VOYAGER_AC97_BASE		(0xa0000 + VOYAGER_BASE)
+
+#define VOYAGER_IRQ_NUM			32
+#define VOYAGER_IRQ_BASE		50
+#define VOYAGER_USBH_IRQ		VOYAGER_IRQ_BASE + 6
+#define VOYAGER_8051_IRQ		VOYAGER_IRQ_BASE + 10
+#define VOYAGER_UART0_IRQ		VOYAGER_IRQ_BASE + 12
+#define VOYAGER_UART1_IRQ		VOYAGER_IRQ_BASE + 13
+#define	VOYAGER_AC97_IRQ		VOYAGER_IRQ_BASE + 17
+
+/* ----- MISC controle  register ------------------------------ */
+#define MISC_CTRL			(0x000004 + VOYAGER_BASE)
+#define MISC_CTRL_USBCLK_48		(3 << 28)
+#define MISC_CTRL_USBCLK_96		(2 << 28)
+#define MISC_CTRL_USBCLK_CRYSTAL	(1 << 28)
+
+/* ----- GPIO[31:0] register --------------------------------- */
+#define GPIO_MUX_LOW			(0x000008 + VOYAGER_BASE)
+#define GPIO_MUX_LOW_AC97		0x1F000000
+#define GPIO_MUX_LOW_8051		0x0000ffff
+#define GPIO_MUX_LOW_PWM		(1 << 29)
+
+/* ----- GPIO[63:32] register --------------------------------- */
+#define GPIO_MUX_HIGH			(0x00000C + VOYAGER_BASE)
+
+/* ----- DRAM controle  register ------------------------------- */
+#define DRAM_CTRL			(0x000010 + VOYAGER_BASE)
+#define DRAM_CTRL_EMBEDDED		(1 << 31)
+#define DRAM_CTRL_CPU_BURST_1		(0 << 28)
+#define DRAM_CTRL_CPU_BURST_2		(1 << 28)
+#define DRAM_CTRL_CPU_BURST_4		(2 << 28)
+#define DRAM_CTRL_CPU_BURST_8		(3 << 28)
+#define DRAM_CTRL_CPU_CAS_LATENCY	(1 << 27)
+#define DRAM_CTRL_CPU_SIZE_2		(0 << 24)
+#define DRAM_CTRL_CPU_SIZE_4		(1 << 24)
+#define DRAM_CTRL_CPU_SIZE_64		(4 << 24)
+#define DRAM_CTRL_CPU_SIZE_32		(5 << 24)
+#define DRAM_CTRL_CPU_SIZE_16		(6 << 24)
+#define DRAM_CTRL_CPU_SIZE_8		(7 << 24)
+#define DRAM_CTRL_CPU_COLUMN_SIZE_1024	(0 << 22)
+#define DRAM_CTRL_CPU_COLUMN_SIZE_512	(2 << 22)
+#define DRAM_CTRL_CPU_COLUMN_SIZE_256	(3 << 22)
+#define DRAM_CTRL_CPU_ACTIVE_PRECHARGE	(1 << 21)
+#define DRAM_CTRL_CPU_RESET		(1 << 20)
+#define DRAM_CTRL_CPU_BANKS		(1 << 19)
+#define DRAM_CTRL_CPU_WRITE_PRECHARGE	(1 << 18)
+#define DRAM_CTRL_BLOCK_WRITE		(1 << 17)
+#define DRAM_CTRL_REFRESH_COMMAND	(1 << 16)
+#define DRAM_CTRL_SIZE_4		(0 << 13)
+#define DRAM_CTRL_SIZE_8		(1 << 13)
+#define DRAM_CTRL_SIZE_16		(2 << 13)
+#define DRAM_CTRL_SIZE_32		(3 << 13)
+#define DRAM_CTRL_SIZE_64		(4 << 13)
+#define DRAM_CTRL_SIZE_2		(5 << 13)
+#define DRAM_CTRL_COLUMN_SIZE_256	(0 << 11)
+#define DRAM_CTRL_COLUMN_SIZE_512	(2 << 11)
+#define DRAM_CTRL_COLUMN_SIZE_1024	(3 << 11)
+#define DRAM_CTRL_BLOCK_WRITE_TIME	(1 << 10)
+#define DRAM_CTRL_BLOCK_WRITE_PRECHARGE	(1 << 9)
+#define DRAM_CTRL_ACTIVE_PRECHARGE	(1 << 8)
+#define DRAM_CTRL_RESET			(1 << 7)
+#define DRAM_CTRL_REMAIN_ACTIVE		(1 << 6)
+#define DRAM_CTRL_BANKS			(1 << 1)
+#define DRAM_CTRL_WRITE_PRECHARGE	(1 << 0)
+
+/* ----- Arvitration control register -------------------------- */
+#define ARBITRATION_CTRL		(0x000014 + VOYAGER_BASE)
+#define ARBITRATION_CTRL_CPUMEM		(1 << 29)
+#define ARBITRATION_CTRL_INTMEM		(1 << 28)
+#define ARBITRATION_CTRL_USB_OFF	(0 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_1	(1 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_2	(2 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_3	(3 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_4	(4 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_5	(5 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_6	(6 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_7	(7 << 24)
+#define ARBITRATION_CTRL_PANEL_OFF	(0 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_1	(1 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_2	(2 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_3	(3 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_4	(4 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_5	(5 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_6	(6 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_7	(7 << 20)
+#define ARBITRATION_CTRL_ZVPORT_OFF	(0 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_1	(1 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_2	(2 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_3	(3 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_4	(4 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_5	(5 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_6	(6 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_7	(7 << 16)
+#define ARBITRATION_CTRL_CMD_INTPR_OFF	(0 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_1	(1 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_2	(2 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_3	(3 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_4	(4 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_5	(5 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_6	(6 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_7	(7 << 12)
+#define ARBITRATION_CTRL_DMA_OFF	(0 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_1	(1 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_2	(2 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_3	(3 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_4	(4 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_5	(5 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_6	(6 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_7	(7 << 8)
+#define ARBITRATION_CTRL_VIDEO_OFF	(0 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_1	(1 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_2	(2 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_3	(3 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_4	(4 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_5	(5 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_6	(6 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_7	(7 << 4)
+#define ARBITRATION_CTRL_CRT_OFF	(0 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_1	(1 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_2	(2 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_3	(3 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_4	(4 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_5	(5 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_6	(6 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_7	(7 << 0)
+
+/* ----- Command list status register -------------------------- */
+#define CMD_INTPR_STATUS		(0x000024 + VOYAGER_BASE)
+
+/* ----- Interrupt status register ----------------------------- */
+#define INT_STATUS			(0x00002c + VOYAGER_BASE)
+#define INT_STATUS_UH			(1 << 6)
+#define INT_STATUS_MC			(1 << 10)
+#define INT_STATUS_U0			(1 << 12)
+#define INT_STATUS_U1			(1 << 13)
+#define	INT_STATUS_AC			(1 << 17)
+
+/* ----- Interrupt mask register ------------------------------ */
+#define VOYAGER_INT_MASK		(0x000030 + VOYAGER_BASE)
+#define VOYAGER_INT_MASK_AC		(1 << 17)
+
+/* ----- Current Gate register ---------------------------------*/
+#define CURRENT_GATE			(0x000038 + VOYAGER_BASE)
+
+/* ----- Power mode 0 gate register --------------------------- */
+#define POWER_MODE0_GATE		(0x000040 + VOYAGER_BASE)
+#define POWER_MODE0_GATE_G		(1 << 6)
+#define POWER_MODE0_GATE_U0		(1 << 7)
+#define POWER_MODE0_GATE_U1		(1 << 8)
+#define POWER_MODE0_GATE_UH		(1 << 11)
+#define	POWER_MODE0_GATE_AC		(1 << 18)
+
+/* ----- Power mode 1 gate register --------------------------- */
+#define POWER_MODE1_GATE		(0x000048 + VOYAGER_BASE)
+#define POWER_MODE1_GATE_G		(1 << 6)
+#define POWER_MODE1_GATE_U0		(1 << 7)
+#define POWER_MODE1_GATE_U1		(1 << 8)
+#define POWER_MODE1_GATE_UH		(1 << 11)
+#define	POWER_MODE1_GATE_AC		(1 << 18)
+
+/* ----- Power mode 0 clock register -------------------------- */
+#define POWER_MODE0_CLOCK		(0x000044 + VOYAGER_BASE)
+
+/* ----- Power mode 1 clock register -------------------------- */
+#define POWER_MODE1_CLOCK		(0x00004C + VOYAGER_BASE)
+
+/* ----- Power mode controll register ------------------------- */
+#define POWER_MODE_CTRL			(0x000054 + VOYAGER_BASE)
+
+/* ----- Miscellaneous Timing register ------------------------ */
+#define SYSTEM_DRAM_CTRL		(0x000068 + VOYAGER_BASE)
+
+/* ----- PWM register ------------------------------------------*/
+#define PWM_0				(0x010020 + VOYAGER_BASE)
+#define PWM_0_HC(x)			(((x)&0x0fff)<<20)
+#define PWM_0_LC(x)			(((x)&0x0fff)<<8 )
+#define PWM_0_CLK_DEV(x)		(((x)&0x000f)<<4 )
+#define PWM_0_EN			(1<<0)
+
+/* ----- I2C register ----------------------------------------- */
+#define I2C_BYTECOUNT			(0x010040 + VOYAGER_BASE)
+#define I2C_CONTROL			(0x010041 + VOYAGER_BASE)
+#define I2C_STATUS			(0x010042 + VOYAGER_BASE)
+#define I2C_RESET			(0x010042 + VOYAGER_BASE)
+#define I2C_SADDRESS			(0x010043 + VOYAGER_BASE)
+#define I2C_DATA			(0x010044 + VOYAGER_BASE)
+
+/* ----- Controle register bits ----------------------------------------- */
+#define I2C_CONTROL_E			(1 << 0)
+#define I2C_CONTROL_MODE		(1 << 1)
+#define I2C_CONTROL_STATUS		(1 << 2)
+#define I2C_CONTROL_INT			(1 << 4)
+#define I2C_CONTROL_INTACK		(1 << 5)
+#define I2C_CONTROL_REPEAT		(1 << 6)
+
+/* ----- Status register bits ----------------------------------------- */
+#define I2C_STATUS_BUSY			(1 << 0)
+#define I2C_STATUS_ACK			(1 << 1)
+#define I2C_STATUS_ERROR		(1 << 2)
+#define I2C_STATUS_COMPLETE		(1 << 3)
+
+/* ----- Reset register  ---------------------------------------------- */
+#define I2C_RESET_ERROR			(1 << 2)
+
+/* ----- transmission frequencies ------------------------------------- */
+#define I2C_SADDRESS_SELECT		(1 << 0)
+
+/* ----- Display Controll register ----------------------------------------- */
+#define PANEL_DISPLAY_CTRL		(0x080000 + VOYAGER_BASE)
+#define PANEL_DISPLAY_CTRL_BIAS         (1<<26)
+#define PANEL_PAN_CTRL			(0x080004 + VOYAGER_BASE)
+#define PANEL_COLOR_KEY			(0x080008 + VOYAGER_BASE)
+#define PANEL_FB_ADDRESS		(0x08000C + VOYAGER_BASE)
+#define PANEL_FB_WIDTH			(0x080010 + VOYAGER_BASE)
+#define PANEL_WINDOW_WIDTH		(0x080014 + VOYAGER_BASE)
+#define PANEL_WINDOW_HEIGHT		(0x080018 + VOYAGER_BASE)
+#define PANEL_PLANE_TL			(0x08001C + VOYAGER_BASE)
+#define PANEL_PLANE_BR			(0x080020 + VOYAGER_BASE)
+#define PANEL_HORIZONTAL_TOTAL		(0x080024 + VOYAGER_BASE)
+#define PANEL_HORIZONTAL_SYNC		(0x080028 + VOYAGER_BASE)
+#define PANEL_VERTICAL_TOTAL		(0x08002C + VOYAGER_BASE)
+#define PANEL_VERTICAL_SYNC		(0x080030 + VOYAGER_BASE)
+#define PANEL_CURRENT_LINE		(0x080034 + VOYAGER_BASE)
+#define VIDEO_DISPLAY_CTRL		(0x080040 + VOYAGER_BASE)
+#define VIDEO_FB_0_ADDRESS		(0x080044 + VOYAGER_BASE)
+#define VIDEO_FB_WIDTH			(0x080048 + VOYAGER_BASE)
+#define VIDEO_FB_0_LAST_ADDRESS		(0x08004C + VOYAGER_BASE)
+#define VIDEO_PLANE_TL			(0x080050 + VOYAGER_BASE)
+#define VIDEO_PLANE_BR			(0x080054 + VOYAGER_BASE)
+#define VIDEO_SCALE			(0x080058 + VOYAGER_BASE)
+#define VIDEO_INITIAL_SCALE		(0x08005C + VOYAGER_BASE)
+#define VIDEO_YUV_CONSTANTS		(0x080060 + VOYAGER_BASE)
+#define VIDEO_FB_1_ADDRESS		(0x080064 + VOYAGER_BASE)
+#define VIDEO_FB_1_LAST_ADDRESS		(0x080068 + VOYAGER_BASE)
+#define VIDEO_ALPHA_DISPLAY_CTRL	(0x080080 + VOYAGER_BASE)
+#define VIDEO_ALPHA_FB_ADDRESS		(0x080084 + VOYAGER_BASE)
+#define VIDEO_ALPHA_FB_WIDTH		(0x080088 + VOYAGER_BASE)
+#define VIDEO_ALPHA_FB_LAST_ADDRESS	(0x08008C + VOYAGER_BASE)
+#define VIDEO_ALPHA_PLANE_TL		(0x080090 + VOYAGER_BASE)
+#define VIDEO_ALPHA_PLANE_BR		(0x080094 + VOYAGER_BASE)
+#define VIDEO_ALPHA_SCALE		(0x080098 + VOYAGER_BASE)
+#define VIDEO_ALPHA_INITIAL_SCALE	(0x08009C + VOYAGER_BASE)
+#define VIDEO_ALPHA_CHROMA_KEY		(0x0800A0 + VOYAGER_BASE)
+#define PANEL_HWC_ADDRESS		(0x0800F0 + VOYAGER_BASE)
+#define PANEL_HWC_LOCATION		(0x0800F4 + VOYAGER_BASE)
+#define PANEL_HWC_COLOR_12		(0x0800F8 + VOYAGER_BASE)
+#define PANEL_HWC_COLOR_3		(0x0800FC + VOYAGER_BASE)
+#define ALPHA_DISPLAY_CTRL		(0x080100 + VOYAGER_BASE)
+#define ALPHA_FB_ADDRESS		(0x080104 + VOYAGER_BASE)
+#define ALPHA_FB_WIDTH			(0x080108 + VOYAGER_BASE)
+#define ALPHA_PLANE_TL			(0x08010C + VOYAGER_BASE)
+#define ALPHA_PLANE_BR			(0x080110 + VOYAGER_BASE)
+#define ALPHA_CHROMA_KEY		(0x080114 + VOYAGER_BASE)
+#define CRT_DISPLAY_CTRL		(0x080200 + VOYAGER_BASE)
+#define CRT_FB_ADDRESS			(0x080204 + VOYAGER_BASE)
+#define CRT_FB_WIDTH			(0x080208 + VOYAGER_BASE)
+#define CRT_HORIZONTAL_TOTAL		(0x08020C + VOYAGER_BASE)
+#define CRT_HORIZONTAL_SYNC		(0x080210 + VOYAGER_BASE)
+#define CRT_VERTICAL_TOTAL		(0x080214 + VOYAGER_BASE)
+#define CRT_VERTICAL_SYNC		(0x080218 + VOYAGER_BASE)
+#define CRT_SIGNATURE_ANALYZER		(0x08021C + VOYAGER_BASE)
+#define CRT_CURRENT_LINE		(0x080220 + VOYAGER_BASE)
+#define CRT_MONITOR_DETECT		(0x080224 + VOYAGER_BASE)
+#define CRT_HWC_ADDRESS			(0x080230 + VOYAGER_BASE)
+#define CRT_HWC_LOCATION		(0x080234 + VOYAGER_BASE)
+#define CRT_HWC_COLOR_12		(0x080238 + VOYAGER_BASE)
+#define CRT_HWC_COLOR_3			(0x08023C + VOYAGER_BASE)
+#define CRT_PALETTE_RAM			(0x080400 + VOYAGER_BASE)
+#define PANEL_PALETTE_RAM		(0x080800 + VOYAGER_BASE)
+#define VIDEO_PALETTE_RAM		(0x080C00 + VOYAGER_BASE)
+
+/* ----- 8051 Controle register ----------------------------------------- */
+#define VOYAGER_8051_BASE		(0x000c0000 + VOYAGER_BASE)
+#define VOYAGER_8051_RESET		(0x000b0000 + VOYAGER_BASE)
+#define VOYAGER_8051_SELECT		(0x000b0004 + VOYAGER_BASE)
+#define VOYAGER_8051_CPU_INT		(0x000b000c + VOYAGER_BASE)
+
+/* ----- AC97 Controle register ----------------------------------------- */
+#define AC97_TX_SLOT0			(0x00000000 + VOYAGER_AC97_BASE)
+#define AC97_CONTROL_STATUS		(0x00000080 + VOYAGER_AC97_BASE)
+#define AC97C_READ			(1 << 19)
+#define AC97C_WD_BIT			(1 << 2)
+#define AC97C_INDEX_MASK		0x7f
+/* -------------------------------------------------------------------- */
+
+#endif /* _VOYAGER_GX_REG_H */
-- 
cgit v1.2.3


From e7f93a355c7e32c26eab8910cf53b7506bb046c5 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:19:13 +0900
Subject: sh: Make PAGE_OFFSET configurable.

nommu needs to be able to shift PAGE_OFFSET, so we switch it to a
non-user-visible CONFIG_PAGE_OFFSET and use that in the few places
where it matters.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  | 26 --------------------------
 arch/sh/boot/compressed/Makefile |  3 ++-
 arch/sh/kernel/vmlinux.lds.S     |  2 +-
 arch/sh/mm/Kconfig               | 31 +++++++++++++++++++++++++++++++
 include/asm-sh/page.h            |  2 +-
 5 files changed, 35 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 07d6d699e9c2..035df7fb2237 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -238,32 +238,6 @@ endchoice
 
 source "arch/sh/mm/Kconfig"
 
-config MEMORY_START
-	hex "Physical memory start address"
-	default "0x08000000"
-	---help---
-	  Computers built with Hitachi SuperH processors always
-	  map the ROM starting at address zero.  But the processor
-	  does not specify the range that RAM takes.
-
-	  The physical memory (RAM) start address will be automatically
-	  set to 08000000. Other platforms, such as the Solution Engine
-	  boards typically map RAM at 0C000000.
-
-	  Tweak this only when porting to a new machine which does not
-	  already have a defconfig. Changing it from the known correct
-	  value on any of the known systems will only lead to disaster.
-
-config MEMORY_SIZE
-	hex "Physical memory size"
-	default "0x00400000"
-	help
-	  This sets the default memory size assumed by your SH kernel. It can
-	  be overridden as normal by the 'mem=' argument on the kernel command
-	  line. If unsure, consult your board specifications or just leave it
-	  as 0x00400000 which was the default value before this became
-	  configurable.
-
 config CF_ENABLER
 	bool "Compact Flash Enabler support"
 	depends on SH_ADX || SH_SOLUTION_ENGINE || SH_UNKNOWN || SH_CAT68701 || SH_SH03
diff --git a/arch/sh/boot/compressed/Makefile b/arch/sh/boot/compressed/Makefile
index 75a6876bf6c6..7074267c01b7 100644
--- a/arch/sh/boot/compressed/Makefile
+++ b/arch/sh/boot/compressed/Makefile
@@ -18,9 +18,10 @@ endif
 # Assign dummy values if these 2 variables are not defined,
 # in order to suppress error message.
 #
+CONFIG_PAGE_OFFSET	?= 0x80000000
 CONFIG_MEMORY_START     ?= 0x0c000000
 CONFIG_BOOT_LINK_OFFSET ?= 0x00800000
-IMAGE_OFFSET := $(shell printf "0x%8x" $$[0x80000000+$(CONFIG_MEMORY_START)+$(CONFIG_BOOT_LINK_OFFSET)])
+IMAGE_OFFSET := $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)+$(CONFIG_MEMORY_START)+$(CONFIG_BOOT_LINK_OFFSET)])
 
 LDFLAGS_vmlinux := -Ttext $(IMAGE_OFFSET) -e startup -T $(obj)/../../kernel/vmlinux.lds
 
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index 3f3e1e0735f5..eb860c51c697 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -13,7 +13,7 @@ OUTPUT_ARCH(sh)
 ENTRY(_start)
 SECTIONS
 {
-  . = 0x80000000 + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
+  . = CONFIG_PAGE_OFFSET + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
   _text = .;			/* Text and read-only data */
   text = .;			/* Text and read-only data */
   .empty_zero_page : {
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index 9064dfc91e6d..f25903aa607b 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -161,6 +161,37 @@ config MMU
 	  turning this off will boot the kernel on these machines with the
 	  MMU implicitly switched off.
 
+config PAGE_OFFSET
+	hex
+	default "0x80000000" if MMU
+	default "0x00000000"
+
+config MEMORY_START
+	hex "Physical memory start address"
+	default "0x08000000"
+	---help---
+	  Computers built with Hitachi SuperH processors always
+	  map the ROM starting at address zero.  But the processor
+	  does not specify the range that RAM takes.
+
+	  The physical memory (RAM) start address will be automatically
+	  set to 08000000. Other platforms, such as the Solution Engine
+	  boards typically map RAM at 0C000000.
+
+	  Tweak this only when porting to a new machine which does not
+	  already have a defconfig. Changing it from the known correct
+	  value on any of the known systems will only lead to disaster.
+
+config MEMORY_SIZE
+	hex "Physical memory size"
+	default "0x00400000"
+	help
+	  This sets the default memory size assumed by your SH kernel. It can
+	  be overridden as normal by the 'mem=' argument on the kernel command
+	  line. If unsure, consult your board specifications or just leave it
+	  as 0x00400000 which was the default value before this became
+	  configurable.
+
 config 32BIT
 	bool "Support 32-bit physical addressing through PMB"
 	depends on CPU_SH4A
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 51d7281a546a..1b3cfd165a66 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -84,7 +84,7 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define __MEMORY_START		CONFIG_MEMORY_START
 #define __MEMORY_SIZE		CONFIG_MEMORY_SIZE
 
-#define PAGE_OFFSET		(0x80000000UL)
+#define PAGE_OFFSET		CONFIG_PAGE_OFFSET
 #define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
 #define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
 
-- 
cgit v1.2.3


From e96636ccfa373a00a0ee0558e1971baa7856d8b5 Mon Sep 17 00:00:00 2001
From: Yoshinori Sato <ysato@users.sourceforge.jp>
Date: Wed, 27 Sep 2006 17:21:02 +0900
Subject: sh: Various nommu fixes.

This fixes up some of the various outstanding nommu bugs on
SH.

Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boot/compressed/Makefile | 10 ++++++++--
 arch/sh/mm/init.c                |  5 +++++
 arch/sh/mm/pg-nommu.c            | 17 +++++++++--------
 include/asm-sh/addrspace.h       |  8 ++++++++
 include/asm-sh/flat.h            |  2 +-
 include/asm-sh/mmu.h             | 13 +------------
 include/asm-sh/page.h            |  5 +++++
 include/asm-sh/uaccess.h         | 13 +++++++++++++
 8 files changed, 50 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boot/compressed/Makefile b/arch/sh/boot/compressed/Makefile
index 7074267c01b7..903470429cb4 100644
--- a/arch/sh/boot/compressed/Makefile
+++ b/arch/sh/boot/compressed/Makefile
@@ -21,11 +21,17 @@ endif
 CONFIG_PAGE_OFFSET	?= 0x80000000
 CONFIG_MEMORY_START     ?= 0x0c000000
 CONFIG_BOOT_LINK_OFFSET ?= 0x00800000
-IMAGE_OFFSET := $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)+$(CONFIG_MEMORY_START)+$(CONFIG_BOOT_LINK_OFFSET)])
+
+IMAGE_OFFSET	:= $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)  + \
+					     $(CONFIG_MEMORY_START) + \
+					     $(CONFIG_BOOT_LINK_OFFSET)])
+
+LIBGCC	:= $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
 LDFLAGS_vmlinux := -Ttext $(IMAGE_OFFSET) -e startup -T $(obj)/../../kernel/vmlinux.lds
 
-$(obj)/vmlinux: $(OBJECTS) $(obj)/piggy.o FORCE
+
+$(obj)/vmlinux: $(OBJECTS) $(obj)/piggy.o $(LIBGCC) FORCE
 	$(call if_changed,ld)
 	@:
 
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index d1a979eab656..2c6dc3d8df2c 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -248,8 +248,13 @@ void __init mem_init(void)
 	 * Setup wrappers for copy/clear_page(), these will get overridden
 	 * later in the boot process if a better method is available.
 	 */
+#ifdef CONFIG_MMU
 	copy_page = copy_page_slow;
 	clear_page = clear_page_slow;
+#else
+	copy_page = copy_page_nommu;
+	clear_page = clear_page_nommu;
+#endif
 
 	/* this will put all low memory onto the freelists */
 	totalram_pages += free_all_bootmem_node(NODE_DATA(0));
diff --git a/arch/sh/mm/pg-nommu.c b/arch/sh/mm/pg-nommu.c
index 8f9165a4e333..d15221beaa16 100644
--- a/arch/sh/mm/pg-nommu.c
+++ b/arch/sh/mm/pg-nommu.c
@@ -14,23 +14,24 @@
 #include <linux/string.h>
 #include <asm/page.h>
 
-static void copy_page_nommu(void *to, void *from)
+void copy_page_nommu(void *to, void *from)
 {
 	memcpy(to, from, PAGE_SIZE);
 }
 
-static void clear_page_nommu(void *to)
+void clear_page_nommu(void *to)
 {
 	memset(to, 0, PAGE_SIZE);
 }
 
-static int __init pg_nommu_init(void)
+__kernel_size_t __copy_user(void *to, const void *from, __kernel_size_t n)
 {
-	copy_page = copy_page_nommu;
-	clear_page = clear_page_nommu;
-
+	memcpy(to, from, n);
 	return 0;
 }
 
-subsys_initcall(pg_nommu_init);
-
+__kernel_size_t __clear_user(void *to, __kernel_size_t n)
+{
+	memset(to, 0, n);
+	return 0;
+}
diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index 720afc11c2ca..4207368267b1 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -14,11 +14,19 @@
 #include <asm/cpu/addrspace.h>
 
 /* Memory segments (32bit Privileged mode addresses)  */
+#ifdef CONFIG_MMU
 #define P0SEG		0x00000000
 #define P1SEG		0x80000000
 #define P2SEG		0xa0000000
 #define P3SEG		0xc0000000
 #define P4SEG		0xe0000000
+#else
+#define P0SEG		0x00000000
+#define P1SEG		0x00000000
+#define P2SEG		0x20000000
+#define P3SEG		0x40000000
+#define P4SEG 		0x80000000
+#endif
 
 /* Returns the privileged segment base of a given address  */
 #define PXSEG(a)	(((unsigned long)(a)) & 0xe0000000)
diff --git a/include/asm-sh/flat.h b/include/asm-sh/flat.h
index f29072e1c87e..0d5cc04ab005 100644
--- a/include/asm-sh/flat.h
+++ b/include/asm-sh/flat.h
@@ -13,7 +13,7 @@
 #define __ASM_SH_FLAT_H
 
 #define	flat_stack_align(sp)			/* nothing needed */
-#define	flat_argvp_envp_on_stack()		1
+#define	flat_argvp_envp_on_stack()		0
 #define	flat_old_ram_flag(flags)		(flags)
 #define	flat_reloc_valid(reloc, size)		((reloc) <= (size))
 #define	flat_get_addr_from_rp(rp, relval, flags)	get_unaligned(rp)
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index ec09589fa6ca..6383dc84501e 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -3,19 +3,8 @@
 
 #if !defined(CONFIG_MMU)
 
-struct mm_rblock_struct {
-	int	size;
-	int	refcount;
-	void	*kblock;
-};
-
-struct mm_tblock_struct {
-	struct mm_rblock_struct *rblock;
-	struct mm_tblock_struct *next;
-};
-
 typedef struct {
-	struct mm_tblock_struct tblock;
+	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 } mm_context_t;
 
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 1b3cfd165a66..e9135532d00c 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -38,8 +38,13 @@
 extern void (*clear_page)(void *to);
 extern void (*copy_page)(void *to, void *from);
 
+#ifdef CONFIG_MMU
 extern void clear_page_slow(void *to);
 extern void copy_page_slow(void *to, void *from);
+#else
+extern void clear_page_nommu(void *to);
+extern void copy_page_nommu(void *to, void *from);
+#endif
 
 #if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
 	defined(CONFIG_SH7705_CACHE_32KB))
diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 6c0014dd2ef7..5c3b00c2f107 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -168,6 +168,7 @@ do {								\
 	__gu_err;						\
 })
 
+#ifdef CONFIG_MMU
 #define __get_user_check(x,ptr,size)				\
 ({								\
 	long __gu_err, __gu_val;				\
@@ -257,6 +258,18 @@ __asm__("stc	r7_bank, %1\n\t"		\
 	: "r" (addr)				\
 	: "t");					\
 })
+#else /* CONFIG_MMU */
+#define __get_user_check(x,ptr,size)					\
+({									\
+	long __gu_err, __gu_val;					\
+	if (__access_ok((unsigned long)(ptr), (size))) {		\
+		__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
+		(x) = (__typeof__(*(ptr)))__gu_val;			\
+	} else								\
+		__gu_err = -EFAULT;					\
+	__gu_err;							\
+})
+#endif
 
 #define __get_user_asm(x, addr, err, insn) \
 ({ \
-- 
cgit v1.2.3


From a2d1a5fae6296c2a3ac1aaa982c95464c46c0585 Mon Sep 17 00:00:00 2001
From: Yoshinori Sato <ysato@users.sourceforge.jp>
Date: Wed, 27 Sep 2006 17:25:07 +0900
Subject: sh: __addr_ok() and other misc nommu fixups.

A few more outstanding nommu fixups..

Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boot/compressed/Makefile |  6 +++---
 arch/sh/kernel/process.c         | 13 +++++++++----
 arch/sh/kernel/sh_ksyms.c        | 10 ++++------
 arch/sh/kernel/sys_sh.c          |  2 +-
 arch/sh/mm/Kconfig               |  2 +-
 arch/sh/mm/Makefile              |  4 ++--
 include/asm-sh/addrspace.h       |  4 ++--
 include/asm-sh/io.h              |  5 +++++
 include/asm-sh/uaccess.h         |  9 ++++++---
 9 files changed, 33 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/sh/boot/compressed/Makefile b/arch/sh/boot/compressed/Makefile
index 903470429cb4..e5f443790079 100644
--- a/arch/sh/boot/compressed/Makefile
+++ b/arch/sh/boot/compressed/Makefile
@@ -22,9 +22,9 @@ CONFIG_PAGE_OFFSET	?= 0x80000000
 CONFIG_MEMORY_START     ?= 0x0c000000
 CONFIG_BOOT_LINK_OFFSET ?= 0x00800000
 
-IMAGE_OFFSET	:= $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)  + \
-					     $(CONFIG_MEMORY_START) + \
-					     $(CONFIG_BOOT_LINK_OFFSET)])
+IMAGE_OFFSET	:= $(shell printf "0x%08x" $$[$(CONFIG_PAGE_OFFSET)  + \
+					      $(CONFIG_MEMORY_START) + \
+					      $(CONFIG_BOOT_LINK_OFFSET)])
 
 LIBGCC	:= $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 4a32550fd7c6..c4aa687ba26a 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -302,9 +302,11 @@ ubc_set_tracing(int asid, unsigned long pc)
 {
 	ctrl_outl(pc, UBC_BARA);
 
+#ifdef CONFIG_MMU
 	/* We don't have any ASID settings for the SH-2! */
 	if (cpu_data->type != CPU_SH7604)
 		ctrl_outb(asid, UBC_BASRA);
+#endif
 
 	ctrl_outl(0, UBC_BAMRA);
 
@@ -347,6 +349,7 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	}
 #endif
 
+#ifdef CONFIG_MMU
 	/*
 	 * Restore the kernel mode register
 	 *   	k7 (r7_bank1)
@@ -354,19 +357,21 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	asm volatile("ldc	%0, r7_bank"
 		     : /* no output */
 		     : "r" (task_thread_info(next)));
+#endif
 
-#ifdef CONFIG_MMU
 	/* If no tasks are using the UBC, we're done */
 	if (ubc_usercnt == 0)
 		/* If no tasks are using the UBC, we're done */;
 	else if (next->thread.ubc_pc && next->mm) {
-		ubc_set_tracing(next->mm->context & MMU_CONTEXT_ASID_MASK,
-				next->thread.ubc_pc);
+		int asid = 0;
+#ifdef CONFIG_MMU
+		asid |= next->mm->context & MMU_CONTEXT_ASID_MASK;
+#endif
+		ubc_set_tracing(asid, next->thread.ubc_pc);
 	} else {
 		ctrl_outw(0, UBC_BBRA);
 		ctrl_outw(0, UBC_BBRB);
 	}
-#endif
 
 	return prev;
 }
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 7f3a42244cab..bf59d73415d7 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -79,20 +79,18 @@ EXPORT_SYMBOL(strcpy);
 DECLARE_EXPORT(__movstr_i4_even);
 DECLARE_EXPORT(__movstr_i4_odd);
 DECLARE_EXPORT(__movstrSI12_i4);
+#endif
 
+#if defined(CONFIG_CPU_SH4) || defined(CONFIG_SH7705_CACHE_32KB)
 /* needed by some modules */
 EXPORT_SYMBOL(flush_cache_all);
 EXPORT_SYMBOL(flush_cache_range);
 EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_purge_region);
-EXPORT_SYMBOL(clear_user_page);
 #endif
 
-#if defined(CONFIG_SH7705_CACHE_32KB)
-EXPORT_SYMBOL(flush_cache_all);
-EXPORT_SYMBOL(flush_cache_range);
-EXPORT_SYMBOL(flush_dcache_page);
-EXPORT_SYMBOL(__flush_purge_region);
+#ifdef CONFIG_MMU
+EXPORT_SYMBOL(clear_user_page);
 #endif
 
 EXPORT_SYMBOL(flush_tlb_page);
diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index d8bcd8a22327..0ee7bf4cb238 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -44,7 +44,7 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5,
 	return error;
 }
 
-#if defined(HAVE_ARCH_UNMAPPED_AREA)
+#if defined(HAVE_ARCH_UNMAPPED_AREA) && defined(CONFIG_MMU)
 /*
  * To avoid cache alias, we map the shard page with same color.
  */
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index f25903aa607b..bed697c0dc19 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -194,7 +194,7 @@ config MEMORY_SIZE
 
 config 32BIT
 	bool "Support 32-bit physical addressing through PMB"
-	depends on CPU_SH4A
+	depends on CPU_SH4A && MMU
 	default y
 	help
 	  If you say Y here, physical addressing will be extended to
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index 87a7c07265c0..da37d86e65eb 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -6,7 +6,7 @@ obj-y			:= init.o extable.o consistent.o
 
 obj-$(CONFIG_CPU_SH2)	+= cache-sh2.o
 obj-$(CONFIG_CPU_SH3)	+= cache-sh3.o
-obj-$(CONFIG_CPU_SH4)	+= cache-sh4.o pg-sh4.o
+obj-$(CONFIG_CPU_SH4)	+= cache-sh4.o
 
 obj-$(CONFIG_DMA_PAGE_OPS)	+= pg-dma.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
@@ -19,7 +19,7 @@ obj-y			+= $(mmu-y)
 
 ifdef CONFIG_MMU
 obj-$(CONFIG_CPU_SH3)		+= tlb-sh3.o
-obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o
+obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o pg-sh4.o
 obj-$(CONFIG_SH7705_CACHE_32KB) += pg-sh7705.o
 endif
 
diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index 4207368267b1..b860218e402e 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -14,7 +14,7 @@
 #include <asm/cpu/addrspace.h>
 
 /* Memory segments (32bit Privileged mode addresses)  */
-#ifdef CONFIG_MMU
+#ifndef CONFIG_CPU_SH2A
 #define P0SEG		0x00000000
 #define P1SEG		0x80000000
 #define P2SEG		0xa0000000
@@ -24,7 +24,7 @@
 #define P0SEG		0x00000000
 #define P1SEG		0x00000000
 #define P2SEG		0x20000000
-#define P3SEG		0x40000000
+#define P3SEG		0x00000000
 #define P4SEG 		0x80000000
 #endif
 
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 27dba653cbe7..377160b86295 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -216,6 +216,7 @@ static inline void ctrl_delay(void)
 
 #define IO_SPACE_LIMIT 0xffffffff
 
+#ifdef CONFIG_MMU
 /*
  * Change virtual addresses to physical addresses and vv.
  * These are trivial on the 1:1 Linux/SuperH mapping
@@ -229,6 +230,10 @@ static inline void *phys_to_virt(unsigned long address)
 {
 	return (void *)P1SEGADDR(address);
 }
+#else
+#define phys_to_virt(address)	((void *)(address))
+#define virt_to_phys(address)	((unsigned long)(address))
+#endif
 
 #define virt_to_bus virt_to_phys
 #define bus_to_virt phys_to_virt
diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 5c3b00c2f107..5c49ed6715f2 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -34,12 +34,12 @@
 
 #define segment_eq(a,b)	((a).seg == (b).seg)
 
-#define __addr_ok(addr) \
-	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
-
 #define get_ds()	(KERNEL_DS)
 
 #if !defined(CONFIG_MMU)
+/* NOMMU is always true */
+#define __addr_ok(addr) (1)
+
 static inline mm_segment_t get_fs(void)
 {
 	return USER_DS;
@@ -66,6 +66,9 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 	return ((addr >= memory_start) && ((addr + size) < memory_end));
 }
 #else /* CONFIG_MMU */
+#define __addr_ok(addr) \
+	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
+
 #define get_fs()	(current_thread_info()->addr_limit)
 #define set_fs(x)	(current_thread_info()->addr_limit = (x))
 
-- 
cgit v1.2.3


From 9f23e7e94f7083d9705b595cbd6b30972be6fbbb Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:27:00 +0900
Subject: sh: pselect6 and ppoll, along with signal trampoline rework.

This implements support for ppoll() and pselect6()..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/entry.S       |   8 +--
 arch/sh/kernel/signal.c      | 137 ++++++++++++++++++++-----------------------
 arch/sh/kernel/syscalls.S    |  31 +++++-----
 include/asm-sh/thread_info.h |   2 +
 include/asm-sh/unistd.h      |  66 +++++++++++----------
 5 files changed, 121 insertions(+), 123 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index fcbf50d4ee10..fd5fe2349f20 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -371,12 +371,12 @@ work_pending:
 	! r8: current_thread_info
 	! t:  result of "tst	#_TIF_NEED_RESCHED, r0"
 	bf/s	work_resched
-	 tst	#_TIF_SIGPENDING, r0
+	 tst	#(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), r0
 work_notifysig:
 	bt/s	restore_all
 	 mov	r15, r4
-	mov	#0, r5
-	mov	r12, r6		! set arg2(save_r0)
+	mov	r12, r5		! set arg1(save_r0)
+	mov	r0, r6
 	mov.l	2f, r1
 	mova	restore_all, r0
 	jmp	@r1
@@ -414,7 +414,7 @@ work_resched:
 
 	.align	2
 1:	.long	schedule
-2:	.long	do_signal
+2:	.long	do_notify_resume
 
 	.align	2
 syscall_exit_work:
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index eb6a3b97e46c..2f1c9545b49f 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -29,13 +29,10 @@
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
-#define DEBUG_SIG 0
+#undef DEBUG
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset,
-			 unsigned int save_r0);
-
 /*
  * Atomically swap in the new signal mask, and wait for a signal.
  */
@@ -44,51 +41,17 @@ sys_sigsuspend(old_sigset_t mask,
 	       unsigned long r5, unsigned long r6, unsigned long r7,
 	       struct pt_regs regs)
 {
-	sigset_t saveset;
-
 	mask &= _BLOCKABLE;
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	siginitset(&current->blocked, mask);
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs.regs[0] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(&regs, &saveset, regs.regs[0]))
-			return -EINTR;
-	}
-}
-
-asmlinkage int
-sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
-		  unsigned long r6, unsigned long r7,
-		  struct pt_regs regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	regs.regs[0] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(&regs, &saveset, regs.regs[0]))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 asmlinkage int 
@@ -349,7 +312,7 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
 	return (void __user *)((sp - frame_size) & -8ul);
 }
 
-static void setup_frame(int sig, struct k_sigaction *ka,
+static int setup_frame(int sig, struct k_sigaction *ka,
 			sigset_t *set, struct pt_regs *regs)
 {
 	struct sigframe __user *frame;
@@ -369,10 +332,9 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 
 	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
 
-	if (_NSIG_WORDS > 1) {
+	if (_NSIG_WORDS > 1)
 		err |= __copy_to_user(frame->extramask, &set->sig[1],
 				      sizeof(frame->extramask));
-	}
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
@@ -403,21 +365,22 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 
 	set_fs(USER_DS);
 
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, current->pid, frame, regs->pc, regs->pr);
 
 	flush_cache_sigtramp(regs->pr);
+
 	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
 		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-	return;
+
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return -EFAULT;
 }
 
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			   sigset_t *set, struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
@@ -478,28 +441,31 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	set_fs(USER_DS);
 
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, current->pid, frame, regs->pc, regs->pr);
 
 	flush_cache_sigtramp(regs->pr);
+
 	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
 		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-	return;
+
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return -EFAULT;
 }
 
 /*
  * OK, we're invoking a handler
  */
 
-static void
+static int
 handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 	      sigset_t *oldset, struct pt_regs *regs)
 {
+	int ret;
+
 	/* Are we from a system call? */
 	if (regs->tra >= 0) {
 		/* If so, check system call restarting.. */
@@ -540,19 +506,23 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 
 	/* Set up the stack frame */
 	if (ka->sa.sa_flags & SA_SIGINFO)
-		setup_rt_frame(sig, ka, info, oldset, regs);
+		ret = setup_rt_frame(sig, ka, info, oldset, regs);
 	else
-		setup_frame(sig, ka, oldset, regs);
+		ret = setup_frame(sig, ka, oldset, regs);
 
 	if (ka->sa.sa_flags & SA_ONESHOT)
 		ka->sa.sa_handler = SIG_DFL;
 
-	spin_lock_irq(&current->sighand->siglock);
-	sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-	if (!(ka->sa.sa_flags & SA_NODEFER))
-		sigaddset(&current->blocked,sig);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
+	if (ret == 0) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+		if (!(ka->sa.sa_flags & SA_NODEFER))
+			sigaddset(&current->blocked,sig);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	return ret;
 }
 
 /*
@@ -564,11 +534,12 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
  * the kernel can handle, and then we build all the user-level signal handling
  * stack-frames in one go after that.
  */
-int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
+static void do_signal(struct pt_regs *regs, unsigned int save_r0)
 {
 	siginfo_t info;
 	int signr;
 	struct k_sigaction ka;
+	sigset_t *oldset;
 
 	/*
 	 * We want the common case to go fast, which
@@ -577,19 +548,27 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
 	if (try_to_freeze())
 		goto no_signal;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 	if (signr > 0) {
 		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &ka, &info, oldset, regs);
-		return 1;
+		if (handle_signal(signr, &ka, &info, oldset, regs) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
 	}
 
  no_signal:
@@ -606,5 +585,19 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
 			regs->regs[3] = __NR_restart_syscall;
 		}
 	}
-	return 0;
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	 * back */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned int save_r0,
+				 __u32 thread_info_flags)
+{
+	/* deal with pending signal delivery */
+	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+		do_signal(regs, save_r0);
 }
diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
index ea23b213c77f..768334e95075 100644
--- a/arch/sh/kernel/syscalls.S
+++ b/arch/sh/kernel/syscalls.S
@@ -319,34 +319,35 @@ ENTRY(sys_call_table)
 	.long sys_mq_getsetattr
 	.long sys_kexec_load
 	.long sys_waitid
-	.long sys_add_key		/* 285 */
+	.long sys_ni_syscall		/* 285 */
+	.long sys_add_key
 	.long sys_request_key
 	.long sys_keyctl
 	.long sys_ioprio_set
-	.long sys_ioprio_get
-	.long sys_inotify_init		/* 290 */
+	.long sys_ioprio_get		/* 290 */
+	.long sys_inotify_init
 	.long sys_inotify_add_watch
 	.long sys_inotify_rm_watch
 	.long sys_migrate_pages
-	.long sys_openat
-	.long sys_mkdirat		/* 295 */
+	.long sys_openat		/* 295 */
+	.long sys_mkdirat
 	.long sys_mknodat
 	.long sys_fchownat
 	.long sys_futimesat
-	.long sys_fstatat64
-	.long sys_unlinkat		/* 300 */
+	.long sys_fstatat64		/* 300 */
+	.long sys_unlinkat
 	.long sys_renameat
 	.long sys_linkat
 	.long sys_symlinkat
-	.long sys_readlinkat
-	.long sys_fchmodat		/* 305 */
+	.long sys_readlinkat		/* 305 */
+	.long sys_fchmodat
 	.long sys_faccessat
-	.long sys_ni_syscall		/* Reserved for pselect6 */
-	.long sys_ni_syscall		/* Reserved for ppoll */
-	.long sys_unshare
-	.long sys_set_robust_list	/* 310 */
+	.long sys_pselect6
+	.long sys_ppoll
+	.long sys_unshare		/* 310 */
+	.long sys_set_robust_list
 	.long sys_get_robust_list
 	.long sys_splice
 	.long sys_sync_file_range
-	.long sys_tee
-	.long sys_vmsplice		/* 315 */
+	.long sys_tee			/* 315 */
+	.long sys_vmsplice
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index b986a1914472..5eb874b065a6 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -94,6 +94,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_NOTIFY_RESUME	1	/* resumption notification requested */
 #define TIF_SIGPENDING		2	/* signal pending */
 #define TIF_NEED_RESCHED	3	/* rescheduling necessary */
+#define TIF_RESTORE_SIGMASK	4	/* restore signal mask in do_signal() */
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		18
@@ -102,6 +103,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
+#define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index 49c6a6ee0ab2..080e5729f17d 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -292,39 +292,40 @@
 #define __NR_mq_getsetattr      (__NR_mq_open+5)
 #define __NR_kexec_load		283
 #define __NR_waitid		284
-#define __NR_add_key		285
-#define __NR_request_key	286
-#define __NR_keyctl		287
-#define __NR_ioprio_set		288
-#define __NR_ioprio_get		289
-#define __NR_inotify_init	290
-#define __NR_inotify_add_watch	291
-#define __NR_inotify_rm_watch	292
-#define __NR_migrate_pages	293
-#define __NR_openat		294
-#define __NR_mkdirat		295
-#define __NR_mknodat		296
-#define __NR_fchownat		297
-#define __NR_futimesat		298
-#define __NR_fstatat64		299
-#define __NR_unlinkat		300
-#define __NR_renameat		301
-#define __NR_linkat		302
-#define __NR_symlinkat		303
-#define __NR_readlinkat		304
-#define __NR_fchmodat		305
-#define __NR_faccessat		305
-#define __NR_pselect6		307
-#define __NR_ppoll		308
-#define __NR_unshare		309
-#define __NR_set_robust_list	310
-#define __NR_get_robust_list	311
-#define __NR_splice		312
-#define __NR_sync_file_range	313
-#define __NR_tee		314
-#define __NR_vmsplice		315
+/* #define __NR_sys_setaltroot	285 */
+#define __NR_add_key		286
+#define __NR_request_key	287
+#define __NR_keyctl		288
+#define __NR_ioprio_set		289
+#define __NR_ioprio_get		290
+#define __NR_inotify_init	291
+#define __NR_inotify_add_watch	292
+#define __NR_inotify_rm_watch	293
+#define __NR_migrate_pages	294
+#define __NR_openat		295
+#define __NR_mkdirat		296
+#define __NR_mknodat		297
+#define __NR_fchownat		298
+#define __NR_futimesat		299
+#define __NR_newfstatat		300
+#define __NR_unlinkat		301
+#define __NR_renameat		302
+#define __NR_linkat		303
+#define __NR_symlinkat		304
+#define __NR_readlinkat		305
+#define __NR_fchmodat		306
+#define __NR_faccessat		307
+#define __NR_pselect6		308
+#define __NR_ppoll		309
+#define __NR_unshare		310
+#define __NR_set_robust_list	311
+#define __NR_get_robust_list	312
+#define __NR_splice		313
+#define __NR_sync_file_range	314
+#define __NR_tee		315
+#define __NR_vmsplice		316
 
-#define NR_syscalls 316
+#define NR_syscalls 317
 
 #ifdef __KERNEL__
 
@@ -466,6 +467,7 @@ __syscall_return(type,__sc0); \
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
 
 #ifdef __KERNEL_SYSCALLS__
 
-- 
cgit v1.2.3


From ecd9561687a0952a96a0a705f618e59cb6f3189b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:32:30 +0900
Subject: serial: Add SERIAL_SH_SCI_NR_UARTS for sh-sci.

sh-sci needs to be able to define its number of ports to
support, we do this with a config option, like most other
ports do.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/serial/Kconfig |  9 +++++++--
 include/asm-sh/sci.h   | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+), 2 deletions(-)
 create mode 100644 include/asm-sh/sci.h

(limited to 'include')

diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 5b48ac22c9c5..261eaa442953 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -642,12 +642,17 @@ config V850E_UART_CONSOLE
 	select SERIAL_CORE_CONSOLE
 
 config SERIAL_SH_SCI
-	tristate "SH SCI(F) serial port support"
+	tristate "SuperH SCI(F) serial port support"
 	depends on SUPERH || H8300
 	select SERIAL_CORE
 
+config SERIAL_SH_SCI_NR_UARTS
+	int "Maximum number of SCI(F) serial ports"
+	depends on SERIAL_SH_SCI
+	default "2"
+
 config SERIAL_SH_SCI_CONSOLE
-	bool "Support for console on SH SCI(F)"
+	bool "Support for console on SuperH SCI(F)"
 	depends on SERIAL_SH_SCI=y
 	select SERIAL_CORE_CONSOLE
 
diff --git a/include/asm-sh/sci.h b/include/asm-sh/sci.h
new file mode 100644
index 000000000000..52e73660c129
--- /dev/null
+++ b/include/asm-sh/sci.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_SH_SCI_H
+#define __ASM_SH_SCI_H
+
+#include <linux/serial_core.h>
+
+/*
+ * Generic header for SuperH SCI(F)
+ *
+ * Do not place SH-specific parts in here, sh64 and h8300 depend on this too.
+ */
+
+/* Offsets into the sci_port->irqs array */
+enum {
+	SCIx_ERI_IRQ,
+	SCIx_RXI_IRQ,
+	SCIx_TXI_IRQ,
+	SCIx_BRI_IRQ,
+	SCIx_NR_IRQS,
+};
+
+/*
+ * Platform device specific platform_data struct
+ */
+struct plat_sci_port {
+	void __iomem	*membase;		/* io cookie */
+	unsigned long	mapbase;		/* resource base */
+	unsigned int	irqs[SCIx_NR_IRQS];	/* ERI, RXI, TXI, BRI */
+	unsigned int	type;			/* SCI / SCIF / IRDA */
+	upf_t		flags;			/* UPF_* flags */
+};
+
+int early_sci_setup(struct uart_port *port);
+
+#endif /* __ASM_SH_SCI_H */
-- 
cgit v1.2.3


From 4052ebb7a2729bd7c28260cdf8e470c0d81b9c56 Mon Sep 17 00:00:00 2001
From: "George G. Davis" <davis_g@mvista.com>
Date: Fri, 22 Sep 2006 18:36:38 +0100
Subject: [ARM] 3859/1: Fix devicemaps_init() XIP_KERNEL odd 1MiB XIP_PHYS_ADDR
 translation error

The ARM XIP_KERNEL map created in devicemaps_init() is wrong.
The map.pfn is rounded down to an even 1MiB section boundary
which results in va/pa translations errors when XIP_PHYS_ADDR
starts on an odd 1MiB boundary and this causes the kernel to
hang.  This patch fixes ARM XIP_KERNEL translation errors for
the odd 1MiB XIP_PHYS_ADDR boundary case.

Signed-off-by: George G. Davis <gdavis@mvista.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c        | 4 ++--
 include/asm-arm/pgtable.h | 7 +++++++
 2 files changed, 9 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 1099af6d470a..64262bda8e54 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -434,9 +434,9 @@ static void __init devicemaps_init(struct machine_desc *mdesc)
 	 * It is always first in the modulearea.
 	 */
 #ifdef CONFIG_XIP_KERNEL
-	map.pfn = __phys_to_pfn(CONFIG_XIP_PHYS_ADDR & PGDIR_MASK);
+	map.pfn = __phys_to_pfn(CONFIG_XIP_PHYS_ADDR & SECTION_MASK);
 	map.virtual = MODULE_START;
-	map.length = ((unsigned long)&_etext - map.virtual + ~PGDIR_MASK) & PGDIR_MASK;
+	map.length = ((unsigned long)&_etext - map.virtual + ~SECTION_MASK) & SECTION_MASK;
 	map.type = MT_ROM;
 	create_mapping(&map);
 #endif
diff --git a/include/asm-arm/pgtable.h b/include/asm-arm/pgtable.h
index 8d3919c6458c..38b2b55688e2 100644
--- a/include/asm-arm/pgtable.h
+++ b/include/asm-arm/pgtable.h
@@ -135,6 +135,13 @@ extern void __pgd_error(const char *file, int line, unsigned long val);
 #define FIRST_USER_PGD_NR	1
 #define USER_PTRS_PER_PGD	((TASK_SIZE/PGDIR_SIZE) - FIRST_USER_PGD_NR)
 
+/*
+ * section address mask and size definitions.
+ */
+#define SECTION_SHIFT		20
+#define SECTION_SIZE		(1UL << SECTION_SHIFT)
+#define SECTION_MASK		(~(SECTION_SIZE-1))
+
 /*
  * ARMv6 supersection address mask and size definitions.
  */
-- 
cgit v1.2.3


From e5723e0eeb2dc16629e86d66785024ead9169000 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:38:11 +0900
Subject: sh: Add support for SH7706/SH7710/SH7343 CPUs.

This adds support for the aforementioned CPU subtypes, and cleans
up some build issues encountered as a result.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                         |  12 +-
 arch/sh/Makefile                        |   7 +-
 arch/sh/boards/hp6xx/setup.c            |   5 +-
 arch/sh/boards/renesas/hs7751rvoip/io.c |   2 +-
 arch/sh/kernel/cpu/irq/ipr.c            |   2 +
 arch/sh/kernel/cpu/sh3/Makefile         |   5 +-
 arch/sh/kernel/cpu/sh3/clock-sh7706.c   |  84 +++++++++
 arch/sh/kernel/cpu/sh3/ex.S             |  54 +++++-
 arch/sh/kernel/cpu/sh3/probe.c          |   6 +
 arch/sh/kernel/cpu/sh3/setup-sh7710.c   |  43 +++++
 arch/sh/kernel/cpu/sh4/Makefile         |   1 +
 arch/sh/kernel/cpu/sh4/ex.S             |  12 +-
 arch/sh/kernel/cpu/sh4/probe.c          |  17 +-
 arch/sh/kernel/cpu/sh4/setup-sh7343.c   |  43 +++++
 arch/sh/kernel/process.c                |   2 +-
 arch/sh/kernel/setup.c                  |  31 ++--
 arch/sh/kernel/sh_ksyms.c               |   3 +-
 arch/sh/mm/Kconfig                      |  30 ++-
 include/asm-sh/cpu-sh3/cache.h          |   4 +-
 include/asm-sh/cpu-sh3/mmu_context.h    |   8 +-
 include/asm-sh/cpu-sh3/timer.h          |   1 +
 include/asm-sh/cpu-sh3/ubc.h            |  15 +-
 include/asm-sh/irq-sh73180.h            |   2 -
 include/asm-sh/irq-sh7343.h             | 317 ++++++++++++++++++++++++++++++++
 include/asm-sh/irq.h                    | 119 +++++++++++-
 include/asm-sh/processor.h              |   8 +-
 26 files changed, 767 insertions(+), 66 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh3/clock-sh7706.c
 create mode 100644 arch/sh/kernel/cpu/sh3/setup-sh7710.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7343.c
 create mode 100644 include/asm-sh/irq-sh7343.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 035df7fb2237..557e7232141c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -324,12 +324,11 @@ config SH_FPU_EMU
 
 config SH_DSP
 	bool "DSP support"
-	depends on !CPU_SH4
-	default y
+	default y if SH4AL_DSP || !CPU_SH4
+	default n
 	help
 	  Selecting this option will enable support for SH processors that
-	  have DSP units (ie, SH2-DSP and SH3-DSP). It is safe to say Y here
-	  by default, as the existance of the DSP will be probed at runtime.
+	  have DSP units (ie, SH2-DSP, SH3-DSP, and SH4AL-DSP).
 
 	  This option must be set in order to enable the DSP.
 
@@ -393,8 +392,9 @@ config SH_PCLK_FREQ
 	int "Peripheral clock frequency (in Hz)"
 	default "50000000" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7780
 	default "60000000" if CPU_SUBTYPE_SH7751
-	default "33333333" if CPU_SUBTYPE_SH7300 || CPU_SUBTYPE_SH7770 || CPU_SUBTYPE_SH7760
-	default "27000000" if CPU_SUBTYPE_SH73180
+	default "33333333" if CPU_SUBTYPE_SH7300 || CPU_SUBTYPE_SH7770 || \
+			      CPU_SUBTYPE_SH7760
+	default "27000000" if CPU_SUBTYPE_SH73180 || CPU_SUBTYPE_SH7343
 	default "66000000" if CPU_SUBTYPE_SH4_202
 	help
 	  This option is used to specify the peripheral clock frequency.
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 13b688b13e1c..d118e48ea1ba 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -18,11 +18,13 @@ cflags-y				:= -mb
 cflags-$(CONFIG_CPU_LITTLE_ENDIAN)	:= -ml
 
 isa-y					:= any
+isa-$(CONFIG_SH_DSP)			:= sh
 isa-$(CONFIG_CPU_SH2)			:= sh2
+isa-$(CONFIG_CPU_SH2A)			:= sh2a
 isa-$(CONFIG_CPU_SH3)			:= sh3
 isa-$(CONFIG_CPU_SH4)			:= sh4
 isa-$(CONFIG_CPU_SH4A)			:= sh4a
-isa-$(CONFIG_CPU_SH2A)			:= sh2a
+isa-$(CONFIG_CPU_SH4AL_DSP)		:= sh4al
 
 isa-$(CONFIG_SH_DSP)			:= $(isa-y)-dsp
 
@@ -30,9 +32,11 @@ ifndef CONFIG_MMU
 isa-y			:= $(isa-y)-nommu
 endif
 
+ifndef CONFIG_SH_DSP
 ifndef CONFIG_SH_FPU
 isa-y			:= $(isa-y)-nofpu
 endif
+endif
 
 cflags-y	+= $(call as-option,-Wa$(comma)-isa=$(isa-y),)
 
@@ -188,4 +192,3 @@ CLEAN_FILES += include/asm-sh/machtypes.h
 define archhelp
 	@echo '  zImage 	           - Compressed kernel image (arch/sh/boot/zImage)'
 endef
-
diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c
index 5fc00f0727c4..629016bec809 100644
--- a/arch/sh/boards/hp6xx/setup.c
+++ b/arch/sh/boards/hp6xx/setup.c
@@ -8,10 +8,11 @@
  *
  * Setup code for an HP680  (internal peripherials only)
  */
-
+#include <linux/types.h>
 #include <linux/init.h>
-#include <asm/io.h>
 #include <asm/hd64461.h>
+#include <asm/io.h>
+#include <asm/irq.h>
 #include <asm/hp6xx/hp6xx.h>
 #include <asm/cpu/dac.h>
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index 8c26550ca2e4..9ea1136b219b 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -10,10 +10,10 @@
  * placeholder code from io_hs7751rvoip.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/module.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
 #include <asm/addrspace.h>
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index 0f545941fb4f..d2b715a04c7e 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -89,6 +89,7 @@ static void mask_and_ack_ipr(unsigned int irq)
 	disable_ipr_irq(irq);
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
 	/* This is needed when we use edge triggered setting */
 	/* XXX: Is it really needed? */
@@ -162,6 +163,7 @@ void __init init_IRQ(void)
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
 	/*
 	 * Initialize the Interrupt Controller (INTC)
diff --git a/arch/sh/kernel/cpu/sh3/Makefile b/arch/sh/kernel/cpu/sh3/Makefile
index 1b292ae16f07..58d3815695ff 100644
--- a/arch/sh/kernel/cpu/sh3/Makefile
+++ b/arch/sh/kernel/cpu/sh3/Makefile
@@ -6,16 +6,19 @@ obj-y	:= ex.o probe.o
 
 # CPU subtype setup
 obj-$(CONFIG_CPU_SUBTYPE_SH7705)	+= setup-sh7705.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7706)	+= setup-sh7709.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7707)	+= setup-sh7709.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7708)	+= setup-sh7708.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7709)	+= setup-sh7709.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7300)	+= setup-sh7300.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7710)	+= setup-sh7710.o
 
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH3)			:= clock-sh3.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7300)	:= clock-sh7300.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7705)	:= clock-sh7705.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7706)	:= clock-sh7706.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7709)	:= clock-sh7709.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7710)	:= clock-sh7300.o
 
 obj-y	+= $(clock-y)
-
diff --git a/arch/sh/kernel/cpu/sh3/clock-sh7706.c b/arch/sh/kernel/cpu/sh3/clock-sh7706.c
new file mode 100644
index 000000000000..0cf96f9833bc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/clock-sh7706.c
@@ -0,0 +1,84 @@
+/*
+ * arch/sh/kernel/cpu/sh3/clock-sh7706.c
+ *
+ * SH7706 support for the clock framework
+ *
+ *  Copyright (C) 2006  Takashi YOSHII
+ *
+ * Based on arch/sh/kernel/cpu/sh3/clock-sh7709.c
+ *  Copyright (C) 2005  Andriy Skulysh
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/clock.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+
+static int stc_multipliers[] = { 1, 2, 4, 1, 3, 6, 1, 1 };
+static int ifc_divisors[]    = { 1, 2, 4, 1, 3, 1, 1, 1 };
+static int pfc_divisors[]    = { 1, 2, 4, 1, 3, 6, 1, 1 };
+
+static void master_clk_init(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x2000) >> 11) | (frqcr & 0x0003);
+
+	clk->rate *= pfc_divisors[idx];
+}
+
+static struct clk_ops sh7706_master_clk_ops = {
+	.init		= master_clk_init,
+};
+
+static void module_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x2000) >> 11) | (frqcr & 0x0003);
+
+	clk->rate = clk->parent->rate / pfc_divisors[idx];
+}
+
+static struct clk_ops sh7706_module_clk_ops = {
+	.recalc		= module_clk_recalc,
+};
+
+static void bus_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x8000) >> 13) | ((frqcr & 0x0030) >> 4);
+
+	clk->rate = clk->parent->rate / stc_multipliers[idx];
+}
+
+static struct clk_ops sh7706_bus_clk_ops = {
+	.recalc		= bus_clk_recalc,
+};
+
+static void cpu_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x4000) >> 12) | ((frqcr & 0x000c) >> 2);
+
+	clk->rate = clk->parent->rate / ifc_divisors[idx];
+}
+
+static struct clk_ops sh7706_cpu_clk_ops = {
+	.recalc		= cpu_clk_recalc,
+};
+
+static struct clk_ops *sh7706_clk_ops[] = {
+	&sh7706_master_clk_ops,
+	&sh7706_module_clk_ops,
+	&sh7706_bus_clk_ops,
+	&sh7706_cpu_clk_ops,
+};
+
+void __init arch_init_clk_ops(struct clk_ops **ops, int idx)
+{
+	if (idx < ARRAY_SIZE(sh7706_clk_ops))
+		*ops = sh7706_clk_ops[idx];
+}
diff --git a/arch/sh/kernel/cpu/sh3/ex.S b/arch/sh/kernel/cpu/sh3/ex.S
index cc04e9e239ff..44daf44833f9 100644
--- a/arch/sh/kernel/cpu/sh3/ex.S
+++ b/arch/sh/kernel/cpu/sh3/ex.S
@@ -84,8 +84,12 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	!      rovi
 	.long	do_IRQ			
 	.long	do_IRQ			/* 5E0 */
-#if  defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
-     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if  defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7706) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7710)
 	.long	do_IRQ	! 32 IRQ  irq0	/* 600 */
 	.long	do_IRQ	! 33      irq1
 	.long	do_IRQ	! 34      irq2
@@ -147,6 +151,51 @@ ENTRY(interrupt_table)
 	.long   do_IRQ	! 62 PCC  pcc0i
 	.long   do_IRQ	! 63      pcc1i	/* 9E0 */
 #endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+	.long   exception_none	! 61 	/* 9A0 */
+	.long   exception_none	! 62
+	.long   exception_none	! 63
+	.long   exception_none	! 64	/* A00 */
+	.long   exception_none	! 65
+	.long   exception_none	! 66
+	.long   exception_none	! 67
+	.long   exception_none	! 68
+	.long   exception_none	! 69
+	.long   exception_none	! 70
+	.long   exception_none	! 71
+	.long   exception_none	! 72	/* B00 */
+	.long   exception_none	! 73
+	.long   exception_none	! 74
+	.long   exception_none	! 75
+	.long   do_IRQ	! 76 DMAC2 dei4	/* B80 */
+	.long   do_IRQ	! 77 DMAC2 dei5
+	.long   exception_none	! 78
+	.long   do_IRQ	! 79 IPSEC ipseci /* BE0 */
+	.long   do_IRQ	! 80 EDMAC eint0 /* C00 */
+	.long   do_IRQ	! 81 EDMAC eint1
+	.long   do_IRQ	! 82 EDMAC eint2
+	.long   exception_none	! 83	/* C60 */
+	.long   exception_none	! 84
+	.long   exception_none	! 85
+	.long   exception_none	! 86
+	.long   exception_none	! 87
+	.long   exception_none	! 88	/* D00 */
+	.long   exception_none	! 89
+	.long   exception_none	! 90
+	.long   exception_none	! 91
+	.long   exception_none	! 92
+	.long   exception_none	! 93
+	.long   exception_none	! 94
+	.long   exception_none	! 95
+	.long   do_IRQ	! 96 SIOF eri0	/* E00 */
+	.long   do_IRQ	! 97      txi0
+	.long   do_IRQ	! 98      rxi0
+	.long   do_IRQ	! 99      cci0
+	.long   do_IRQ	! 100     eri1	/* E80 */
+	.long   do_IRQ	! 101     txi1
+	.long   do_IRQ	! 102     rxi2
+	.long   do_IRQ	! 103     cci3
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7300)
 	.long   do_IRQ	! 64
 	.long   do_IRQ	! 65
@@ -195,4 +244,3 @@ ENTRY(interrupt_table)
 	.long   do_IRQ	! 108
 #endif
 #endif
-
diff --git a/arch/sh/kernel/cpu/sh3/probe.c b/arch/sh/kernel/cpu/sh3/probe.c
index 5cdc88638601..e67098836290 100644
--- a/arch/sh/kernel/cpu/sh3/probe.c
+++ b/arch/sh/kernel/cpu/sh3/probe.c
@@ -72,6 +72,12 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->dcache.sets		= 256;
 		cpu_data->type = CPU_SH7729;
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7706)
+		cpu_data->type = CPU_SH7706;
+#endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+		cpu_data->type = CPU_SH7710;
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7705)
 		cpu_data->type = CPU_SH7705;
 
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7710.c b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
new file mode 100644
index 000000000000..895f99ee6a95
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
@@ -0,0 +1,43 @@
+/*
+ * SH7710 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4400000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7710_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7710_devices_setup(void)
+{
+	return platform_add_devices(sh7710_devices,
+				    ARRAY_SIZE(sh7710_devices));
+}
+__initcall(sh7710_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index 48946d54a2cd..8dbf3895ece7 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_CPU_SUBTYPE_SH7760)	+= setup-sh7760.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7770)	+= setup-sh7770.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
 obj-$(CONFIG_CPU_SUBTYPE_SH73180)	+= setup-sh73180.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7343)	+= setup-sh7343.o
 obj-$(CONFIG_CPU_SUBTYPE_SH4_202)	+= setup-sh4-202.o
 
 # Primary on-chip clocks (common)
diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S
index af5ecbddea55..7146893a6cca 100644
--- a/arch/sh/kernel/cpu/sh4/ex.S
+++ b/arch/sh/kernel/cpu/sh4/ex.S
@@ -123,6 +123,13 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	! 45      dmte5
 	.long	do_IRQ	! 46      dmte6
 	.long	do_IRQ	! 47      dmte7		/* 7E0 */
+#elif defined(CONFIG_CPU_SUBTYPE_SH7343)
+	.long	do_IRQ	! 44 IIC1 ali		/* 780 */
+	.long	do_IRQ	! 45      tacki
+	.long	do_IRQ	! 46      waiti
+	.long	do_IRQ	! 47      dtei		/* 7E0 */
+	.long	do_IRQ	! 48 DMAC dei0		/* 800 */
+	.long	do_IRQ	! 49      dei1		/* 820 */
 #else
 	.long	exception_error			! 44	/* 780 */
 	.long	exception_error			! 45
@@ -132,7 +139,8 @@ ENTRY(interrupt_table)
 #if defined(CONFIG_SH_FPU)
 	.long	do_fpu_state_restore	! 48	/* 800 */
 	.long	do_fpu_state_restore	! 49	/* 820 */
-#else
+#elif !defined(CONFIG_CPU_SUBTYPE_SH7343) && \
+      !defined(CONFIG_CPU_SUBTYPE_SH73180)
 	.long	exception_error
 	.long	exception_error
 #endif
@@ -225,7 +233,7 @@ ENTRY(interrupt_table)
 	.long	exception_error
 	.long	do_IRQ	! ADC	adi
 	.long	do_IRQ	! CMT	cmti	/* FA0 */
-#elif defined(CONFIG_CPU_SUBTYPE_SH73180)
+#elif defined(CONFIG_CPU_SUBTYPE_SH73180) || defined(CONFIG_CPU_SUBTYPE_SH7343)
 	.long	do_IRQ	!  50 0x840
 	.long	do_IRQ	!  51 0x860
 	.long	do_IRQ	!  52 0x880
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 2a7707a81d8f..6e8a2b5268e8 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -3,7 +3,7 @@
  *
  * CPU Subtype Probing for SH-4.
  *
- * Copyright (C) 2001 - 2005  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -76,12 +76,6 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->type = CPU_SH73180;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
-
-		/*
-		 * XXX: Double check this, none of the SH-4A/SH-4AL processors
-		 * should have this, as it's essentially a legacy thing.
-		 */
-		cpu_data->flags |= CPU_HAS_PTEA;
 		break;
 	case 0x2001:
 	case 0x2004:
@@ -89,8 +83,7 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
 
-		/* Same note as above applies here for PTEA */
-		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
+		cpu_data->flags |= CPU_HAS_FPU;
 		break;
 	case 0x2006:
 	case 0x200A:
@@ -104,6 +97,12 @@ int __init detect_cpu_and_cache_system(void)
 
 		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER;
 		break;
+	case 0x3000:
+	case 0x3003:
+		cpu_data->type = CPU_SH7343;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
 		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7343.c b/arch/sh/kernel/cpu/sh4/setup-sh7343.c
new file mode 100644
index 000000000000..91d61cf91ba1
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7343.c
@@ -0,0 +1,43 @@
+/*
+ * SH7343 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 80, 81, 83, 82 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7343_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7343_devices_setup(void)
+{
+	return platform_add_devices(sh7343_devices,
+				    ARRAY_SIZE(sh7343_devices));
+}
+__initcall(sh7343_devices_setup);
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index c4aa687ba26a..2167746e88f1 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -310,7 +310,7 @@ ubc_set_tracing(int asid, unsigned long pc)
 
 	ctrl_outl(0, UBC_BAMRA);
 
-	if (cpu_data->type == CPU_SH7729) {
+	if (cpu_data->type == CPU_SH7729 || cpu_data->type == CPU_SH7710) {
 		ctrl_outw(BBR_INST | BBR_READ | BBR_CPU, UBC_BBRA);
 		ctrl_outl(BRCR_PCBA | BRCR_PCTE, UBC_BRCR);
 	} else {
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 4afdec071700..86ef17fe48b5 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -424,25 +424,18 @@ static int __init topology_init(void)
 subsys_initcall(topology_init);
 
 static const char *cpu_name[] = {
-	[CPU_SH7604]	= "SH7604",
-	[CPU_SH7705]	= "SH7705",
-	[CPU_SH7708]	= "SH7708",
-	[CPU_SH7729]	= "SH7729",
-	[CPU_SH7300]	= "SH7300",
-	[CPU_SH7750]	= "SH7750",
-	[CPU_SH7750S]	= "SH7750S",
-	[CPU_SH7750R]	= "SH7750R",
-	[CPU_SH7751]	= "SH7751",
-	[CPU_SH7751R]	= "SH7751R",
-	[CPU_SH7760]	= "SH7760",
-	[CPU_SH73180]	= "SH73180",
-	[CPU_ST40RA]	= "ST40RA",
-	[CPU_ST40GX1]	= "ST40GX1",
-	[CPU_SH4_202]	= "SH4-202",
-	[CPU_SH4_501]	= "SH4-501",
-	[CPU_SH7770]	= "SH7770",
-	[CPU_SH7780]	= "SH7780",
-	[CPU_SH7781]	= "SH7781",
+	[CPU_SH7604]	= "SH7604",	[CPU_SH7300]	= "SH7300",
+	[CPU_SH7705]	= "SH7705",	[CPU_SH7706]	= "SH7706",
+	[CPU_SH7707]	= "SH7707",	[CPU_SH7708]	= "SH7708",
+	[CPU_SH7709]	= "SH7709",	[CPU_SH7710]	= "SH7710",
+	[CPU_SH7729]	= "SH7729",	[CPU_SH7750]	= "SH7750",
+	[CPU_SH7750S]	= "SH7750S",	[CPU_SH7750R]	= "SH7750R",
+	[CPU_SH7751]	= "SH7751",	[CPU_SH7751R]	= "SH7751R",
+	[CPU_SH7760]	= "SH7760",	[CPU_SH73180]	= "SH73180",
+	[CPU_ST40RA]	= "ST40RA",	[CPU_ST40GX1]	= "ST40GX1",
+	[CPU_SH4_202]	= "SH4-202",	[CPU_SH4_501]	= "SH4-501",
+	[CPU_SH7770]	= "SH7770",	[CPU_SH7780]	= "SH7780",
+	[CPU_SH7781]	= "SH7781",	[CPU_SH7343]	= "SH7343",
 	[CPU_SH_NONE]	= "Unknown"
 };
 
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index bf59d73415d7..d3cbfa2ad4a7 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -89,7 +89,8 @@ EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_purge_region);
 #endif
 
-#ifdef CONFIG_MMU
+#if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
+	defined(CONFIG_SH7705_CACHE_32KB))
 EXPORT_SYMBOL(clear_user_page);
 #endif
 
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index bed697c0dc19..b445d02075e8 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -21,6 +21,10 @@ config CPU_SH4A
 	bool
 	select CPU_SH4
 
+config CPU_SH4AL_DSP
+	bool
+	select CPU_SH4A
+
 config CPU_SUBTYPE_ST40
 	bool
 	select CPU_SH4
@@ -47,6 +51,12 @@ config CPU_SUBTYPE_SH7705
 	select CPU_SH3
 	select CPU_HAS_PINT_IRQ
 
+config CPU_SUBTYPE_SH7706
+	bool "Support SH7706 processor"
+	select CPU_SH3
+	help
+	  Select SH7706 if you have a 133 Mhz SH-3 HD6417706 CPU.
+
 config CPU_SUBTYPE_SH7707
 	bool "Support SH7707 processor"
 	select CPU_SH3
@@ -68,6 +78,12 @@ config CPU_SUBTYPE_SH7709
 	help
 	  Select SH7709 if you have a  80 Mhz SH-3 HD6417709 CPU.
 
+config CPU_SUBTYPE_SH7710
+	bool "Support SH7710 processor"
+	select CPU_SH3
+	help
+	  Select SH7710 if you have a SH3-DSP SH7710 CPU.
+
 comment "SH-4 Processor Support"
 
 config CPU_SUBTYPE_SH7750
@@ -132,10 +148,6 @@ config CPU_SUBTYPE_ST40GX1
 
 comment "SH-4A Processor Support"
 
-config CPU_SUBTYPE_SH73180
-	bool "Support SH73180 processor"
-	select CPU_SH4A
-
 config CPU_SUBTYPE_SH7770
 	bool "Support SH7770 processor"
 	select CPU_SH4A
@@ -145,6 +157,16 @@ config CPU_SUBTYPE_SH7780
 	select CPU_SH4A
 	select CPU_HAS_INTC2_IRQ
 
+comment "SH4AL-DSP Processor Support"
+
+config CPU_SUBTYPE_SH73180
+	bool "Support SH73180 processor"
+	select CPU_SH4AL_DSP
+
+config CPU_SUBTYPE_SH7343
+	bool "Support SH7343 processor"
+	select CPU_SH4AL_DSP
+
 endmenu
 
 menu "Memory management options"
diff --git a/include/asm-sh/cpu-sh3/cache.h b/include/asm-sh/cpu-sh3/cache.h
index 406aa8d9b947..ffe08d2813f9 100644
--- a/include/asm-sh/cpu-sh3/cache.h
+++ b/include/asm-sh/cpu-sh3/cache.h
@@ -26,12 +26,10 @@
 #define CCR_CACHE_ENABLE	CCR_CACHE_CE
 #define CCR_CACHE_INVALIDATE	CCR_CACHE_CF
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if defined(CONFIG_CPU_SUBTYPE_SH7705) || defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define CCR3	0xa40000b4
 #define CCR_CACHE_16KB  0x00010000
 #define CCR_CACHE_32KB	0x00020000
 #endif
 
-
 #endif /* __ASM_CPU_SH3_CACHE_H */
-
diff --git a/include/asm-sh/cpu-sh3/mmu_context.h b/include/asm-sh/cpu-sh3/mmu_context.h
index a844ea0965b6..bccb7ddb438b 100644
--- a/include/asm-sh/cpu-sh3/mmu_context.h
+++ b/include/asm-sh/cpu-sh3/mmu_context.h
@@ -27,8 +27,12 @@
 #define TRA	0xffffffd0
 #define EXPEVT	0xffffffd4
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define INTEVT	0xa4000000	/* INTEVTE2(0xa4000000) */
 #else
 #define INTEVT	0xffffffd8
diff --git a/include/asm-sh/cpu-sh3/timer.h b/include/asm-sh/cpu-sh3/timer.h
index 2082ad956f21..b2394cf76f49 100644
--- a/include/asm-sh/cpu-sh3/timer.h
+++ b/include/asm-sh/cpu-sh3/timer.h
@@ -20,6 +20,7 @@
  *	SH7710
  *	SH7720
  *	SH7300
+ *	SH7710
  * ---------------------------------------------------------------------------
  */
 
diff --git a/include/asm-sh/cpu-sh3/ubc.h b/include/asm-sh/cpu-sh3/ubc.h
index 0f809dec4e17..9d308cbe9b29 100644
--- a/include/asm-sh/cpu-sh3/ubc.h
+++ b/include/asm-sh/cpu-sh3/ubc.h
@@ -11,6 +11,19 @@
 #ifndef __ASM_CPU_SH3_UBC_H
 #define __ASM_CPU_SH3_UBC_H
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+#define UBC_BARA		0xa4ffffb0
+#define UBC_BAMRA		0xa4ffffb4
+#define UBC_BBRA		0xa4ffffb8
+#define UBC_BASRA		0xffffffe4
+#define UBC_BARB		0xa4ffffa0
+#define UBC_BAMRB		0xa4ffffa4
+#define UBC_BBRB		0xa4ffffa8
+#define UBC_BASRB		0xffffffe8
+#define UBC_BDRB		0xa4ffff90
+#define UBC_BDMRB		0xa4ffff94
+#define UBC_BRCR		0xa4ffff98
+#else
 #define UBC_BARA                0xffffffb0
 #define UBC_BAMRA               0xffffffb4
 #define UBC_BBRA                0xffffffb8
@@ -22,6 +35,6 @@
 #define UBC_BDRB                0xffffff90
 #define UBC_BDMRB               0xffffff94
 #define UBC_BRCR                0xffffff98
+#endif
 
 #endif /* __ASM_CPU_SH3_UBC_H */
-
diff --git a/include/asm-sh/irq-sh73180.h b/include/asm-sh/irq-sh73180.h
index d705252be260..b28af9a69d72 100644
--- a/include/asm-sh/irq-sh73180.h
+++ b/include/asm-sh/irq-sh73180.h
@@ -311,6 +311,4 @@
 #define IRQ6_PRIORITY	1
 #define IRQ7_PRIORITY	1
 
-int shmse_irq_demux(int irq);
-
 #endif /* __ASM_SH_IRQ_SH73180_H */
diff --git a/include/asm-sh/irq-sh7343.h b/include/asm-sh/irq-sh7343.h
new file mode 100644
index 000000000000..5d15419b53b0
--- /dev/null
+++ b/include/asm-sh/irq-sh7343.h
@@ -0,0 +1,317 @@
+#ifndef __ASM_SH_IRQ_SH7343_H
+#define __ASM_SH_IRQ_SH7343_H
+
+/*
+ * linux/include/asm-sh/irq-sh7343.h
+ *
+ * Copyright (C) 2006 Kenati Technologies Inc.
+ * Andre Mccurdy <andre@kenati.com>
+ * Ranjit Deshpande <ranjit@kenati.com>
+ */
+
+#undef INTC_IPRA
+#undef INTC_IPRB
+#undef INTC_IPRC
+#undef INTC_IPRD
+
+#undef DMTE0_IRQ
+#undef DMTE1_IRQ
+#undef DMTE2_IRQ
+#undef DMTE3_IRQ
+#undef DMTE4_IRQ
+#undef DMTE5_IRQ
+#undef DMTE6_IRQ
+#undef DMTE7_IRQ
+#undef DMAE_IRQ
+#undef DMA_IPR_ADDR
+#undef DMA_IPR_POS
+#undef DMA_PRIORITY
+
+#undef INTC_IMCR0
+#undef INTC_IMCR1
+#undef INTC_IMCR2
+#undef INTC_IMCR3
+#undef INTC_IMCR4
+#undef INTC_IMCR5
+#undef INTC_IMCR6
+#undef INTC_IMCR7
+#undef INTC_IMCR8
+#undef INTC_IMCR9
+#undef INTC_IMCR10
+
+
+#define INTC_IPRA  	0xA4080000UL
+#define INTC_IPRB  	0xA4080004UL
+#define INTC_IPRC  	0xA4080008UL
+#define INTC_IPRD  	0xA408000CUL
+#define INTC_IPRE  	0xA4080010UL
+#define INTC_IPRF  	0xA4080014UL
+#define INTC_IPRG  	0xA4080018UL
+#define INTC_IPRH  	0xA408001CUL
+#define INTC_IPRI  	0xA4080020UL
+#define INTC_IPRJ  	0xA4080024UL
+#define INTC_IPRK  	0xA4080028UL
+#define INTC_IPRL  	0xA408002CUL
+
+#define INTC_IMR0	0xA4080080UL
+#define INTC_IMR1	0xA4080084UL
+#define INTC_IMR2	0xA4080088UL
+#define INTC_IMR3	0xA408008CUL
+#define INTC_IMR4	0xA4080090UL
+#define INTC_IMR5	0xA4080094UL
+#define INTC_IMR6	0xA4080098UL
+#define INTC_IMR7	0xA408009CUL
+#define INTC_IMR8	0xA40800A0UL
+#define INTC_IMR9	0xA40800A4UL
+#define INTC_IMR10	0xA40800A8UL
+#define INTC_IMR11	0xA40800ACUL
+
+#define INTC_IMCR0	0xA40800C0UL
+#define INTC_IMCR1	0xA40800C4UL
+#define INTC_IMCR2	0xA40800C8UL
+#define INTC_IMCR3	0xA40800CCUL
+#define INTC_IMCR4	0xA40800D0UL
+#define INTC_IMCR5	0xA40800D4UL
+#define INTC_IMCR6	0xA40800D8UL
+#define INTC_IMCR7	0xA40800DCUL
+#define INTC_IMCR8	0xA40800E0UL
+#define INTC_IMCR9	0xA40800E4UL
+#define INTC_IMCR10	0xA40800E8UL
+#define INTC_IMCR11	0xA40800ECUL
+
+#define INTC_ICR0	0xA4140000UL
+#define INTC_ICR1	0xA414001CUL
+
+#define INTMSK0		0xa4140044
+#define INTMSKCLR0	0xa4140064
+#define INTC_INTPRI0	0xa4140010
+
+/*
+  NOTE:
+
+  *_IRQ = (INTEVT2 - 0x200)/0x20
+*/
+
+/* TMU0 */
+#define TMU0_IRQ	16
+#define TMU0_IPR_ADDR	INTC_IPRA
+#define TMU0_IPR_POS	 3
+#define TMU0_PRIORITY	 2
+
+#define TIMER_IRQ       16
+#define TIMER_IPR_ADDR  INTC_IPRA
+#define TIMER_IPR_POS    3
+#define TIMER_PRIORITY   2
+
+/* TMU1 */
+#define TMU1_IRQ	17
+#define TMU1_IPR_ADDR	INTC_IPRA
+#define TMU1_IPR_POS	 2
+#define TMU1_PRIORITY	 2
+
+/* TMU2 */
+#define TMU2_IRQ	18
+#define TMU2_IPR_ADDR	INTC_IPRA
+#define TMU2_IPR_POS	 1
+#define TMU2_PRIORITY	 2
+
+/* LCDC */
+#define LCDC_IRQ	28
+#define LCDC_IPR_ADDR	INTC_IPRB
+#define LCDC_IPR_POS	 2
+#define LCDC_PRIORITY	 2
+
+/* VIO (Video I/O) */
+#define CEU_IRQ		52
+#define BEU_IRQ		53
+#define VEU_IRQ		54
+#define VOU_IRQ		55
+#define VIO_IPR_ADDR	INTC_IPRE
+#define VIO_IPR_POS	 2
+#define VIO_PRIORITY	 2
+
+/* MFI (Multi Functional Interface) */
+#define MFI_IRQ		56
+#define MFI_IPR_ADDR	INTC_IPRE
+#define MFI_IPR_POS	 1
+#define MFI_PRIORITY	 2
+
+/* VPU (Video Processing Unit) */
+#define VPU_IRQ		60
+#define VPU_IPR_ADDR	INTC_IPRE
+#define VPU_IPR_POS	 0
+#define VPU_PRIORITY	 2
+
+/* 3DG */
+#define TDG_IRQ		63
+#define TDG_IPR_ADDR	INTC_IPRJ
+#define TDG_IPR_POS	 2
+#define TDG_PRIORITY	 2
+
+/* DMAC(1) */
+#define DMTE0_IRQ	48
+#define DMTE1_IRQ	49
+#define DMTE2_IRQ	50
+#define DMTE3_IRQ	51
+#define DMA1_IPR_ADDR	INTC_IPRE
+#define DMA1_IPR_POS	3
+#define DMA1_PRIORITY	7
+
+/* DMAC(2) */
+#define DMTE4_IRQ	76
+#define DMTE5_IRQ	77
+#define DMA2_IPR_ADDR	INTC_IPRF
+#define DMA2_IPR_POS	2
+#define DMA2_PRIORITY	7
+
+/* SCIF0 */
+#define SCIF_ERI_IRQ	80
+#define SCIF_RXI_IRQ	81
+#define SCIF_BRI_IRQ	82
+#define SCIF_TXI_IRQ	83
+#define SCIF_IPR_ADDR	INTC_IPRG
+#define SCIF_IPR_POS	3
+#define SCIF_PRIORITY	3
+
+/* SIOF0 */
+#define SIOF0_IRQ	84
+#define SIOF0_IPR_ADDR	INTC_IPRH
+#define SIOF0_IPR_POS	3
+#define SIOF0_PRIORITY	3
+
+/* FLCTL (Flash Memory Controller) */
+#define FLSTE_IRQ	92
+#define FLTEND_IRQ	93
+#define FLTRQ0_IRQ	94
+#define FLTRQ1_IRQ	95
+#define FLCTL_IPR_ADDR	INTC_IPRH
+#define FLCTL_IPR_POS	1
+#define FLCTL_PRIORITY	3
+
+/* IIC(0) (IIC Bus Interface) */
+#define IIC0_ALI_IRQ	96
+#define IIC0_TACKI_IRQ	97
+#define IIC0_WAITI_IRQ	98
+#define IIC0_DTEI_IRQ	99
+#define IIC0_IPR_ADDR	INTC_IPRH
+#define IIC0_IPR_POS	0
+#define IIC0_PRIORITY	3
+
+/* IIC(1) (IIC Bus Interface) */
+#define IIC1_ALI_IRQ	44
+#define IIC1_TACKI_IRQ	45
+#define IIC1_WAITI_IRQ	46
+#define IIC1_DTEI_IRQ	47
+#define IIC1_IPR_ADDR	INTC_IPRI
+#define IIC1_IPR_POS	0
+#define IIC1_PRIORITY	3
+
+/* SIO0 */
+#define SIO0_IRQ	88
+#define SIO0_IPR_ADDR	INTC_IPRI
+#define SIO0_IPR_POS	3
+#define SIO0_PRIORITY	3
+
+/* SDHI */
+#define SDHI_SDHII0_IRQ	100
+#define SDHI_SDHII1_IRQ	101
+#define SDHI_SDHII2_IRQ	102
+#define SDHI_SDHII3_IRQ	103
+#define SDHI_IPR_ADDR	INTC_IPRK
+#define SDHI_IPR_POS	0
+#define SDHI_PRIORITY	3
+
+/* SIU (Sound Interface Unit) */
+#define SIU_IRQ		108
+#define SIU_IPR_ADDR	INTC_IPRJ
+#define SIU_IPR_POS	1
+#define SIU_PRIORITY	3
+
+#define PORT_PACR	0xA4050100UL
+#define PORT_PBCR	0xA4050102UL
+#define PORT_PCCR	0xA4050104UL
+#define PORT_PDCR	0xA4050106UL
+#define PORT_PECR	0xA4050108UL
+#define PORT_PFCR	0xA405010AUL
+#define PORT_PGCR	0xA405010CUL
+#define PORT_PHCR	0xA405010EUL
+#define PORT_PJCR	0xA4050110UL
+#define PORT_PKCR	0xA4050112UL
+#define PORT_PLCR	0xA4050114UL
+#define PORT_SCPCR	0xA4050116UL
+#define PORT_PMCR	0xA4050118UL
+#define PORT_PNCR	0xA405011AUL
+#define PORT_PQCR	0xA405011CUL
+#define PORT_PRCR	0xA405011EUL
+#define PORT_PTCR	0xA405014CUL
+#define PORT_PUCR	0xA405014EUL
+#define PORT_PVCR	0xA4050150UL
+
+#define PORT_PSELA	0xA4050140UL
+#define PORT_PSELB	0xA4050142UL
+#define PORT_PSELC	0xA4050144UL
+#define PORT_PSELE	0xA4050158UL
+
+#define PORT_HIZCRA	0xA4050146UL
+#define PORT_HIZCRB	0xA4050148UL
+#define PORT_DRVCR	0xA405014AUL
+
+#define PORT_PADR  	0xA4050120UL
+#define PORT_PBDR  	0xA4050122UL
+#define PORT_PCDR  	0xA4050124UL
+#define PORT_PDDR  	0xA4050126UL
+#define PORT_PEDR  	0xA4050128UL
+#define PORT_PFDR  	0xA405012AUL
+#define PORT_PGDR  	0xA405012CUL
+#define PORT_PHDR  	0xA405012EUL
+#define PORT_PJDR  	0xA4050130UL
+#define PORT_PKDR  	0xA4050132UL
+#define PORT_PLDR  	0xA4050134UL
+#define PORT_SCPDR  	0xA4050136UL
+#define PORT_PMDR  	0xA4050138UL
+#define PORT_PNDR  	0xA405013AUL
+#define PORT_PQDR  	0xA405013CUL
+#define PORT_PRDR  	0xA405013EUL
+#define PORT_PTDR  	0xA405016CUL
+#define PORT_PUDR  	0xA405016EUL
+#define PORT_PVDR  	0xA4050170UL
+
+#define IRQ0_IRQ	32
+#define IRQ1_IRQ	33
+#define IRQ2_IRQ	34
+#define IRQ3_IRQ	35
+#define IRQ4_IRQ	36
+#define IRQ5_IRQ	37
+#define IRQ6_IRQ	38
+#define IRQ7_IRQ	39
+
+#define INTPRI00	0xA4140010UL
+
+#define IRQ0_IPR_ADDR	INTPRI00
+#define IRQ1_IPR_ADDR	INTPRI00
+#define IRQ2_IPR_ADDR	INTPRI00
+#define IRQ3_IPR_ADDR	INTPRI00
+#define IRQ4_IPR_ADDR	INTPRI00
+#define IRQ5_IPR_ADDR	INTPRI00
+#define IRQ6_IPR_ADDR	INTPRI00
+#define IRQ7_IPR_ADDR	INTPRI00
+
+#define IRQ0_IPR_POS	7
+#define IRQ1_IPR_POS	6
+#define IRQ2_IPR_POS	5
+#define IRQ3_IPR_POS	4
+#define IRQ4_IPR_POS	3
+#define IRQ5_IPR_POS	2
+#define IRQ6_IPR_POS	1
+#define IRQ7_IPR_POS	0
+
+#define IRQ0_PRIORITY	1
+#define IRQ1_PRIORITY	1
+#define IRQ2_PRIORITY	1
+#define IRQ3_PRIORITY	1
+#define IRQ4_PRIORITY	1
+#define IRQ5_PRIORITY	1
+#define IRQ6_PRIORITY	1
+#define IRQ7_PRIORITY	1
+
+#endif /* __ASM_SH_IRQ_SH7343_H */
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 648102e9236f..00886f9adb4d 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -192,7 +192,7 @@
 
 #if defined (CONFIG_CPU_SUBTYPE_SH7707) || defined (CONFIG_CPU_SUBTYPE_SH7708) || \
     defined (CONFIG_CPU_SUBTYPE_SH7709) || defined (CONFIG_CPU_SUBTYPE_SH7750) || \
-    defined (CONFIG_CPU_SUBTYPE_SH7751)
+    defined (CONFIG_CPU_SUBTYPE_SH7751) || defined (CONFIG_CPU_SUBTYPE_SH7706)
 #define SCI_ERI_IRQ	23
 #define SCI_RXI_IRQ	24
 #define SCI_TXI_IRQ	25
@@ -207,6 +207,7 @@
 #define SCIF0_IPR_POS	3
 #define SCIF0_PRIORITY	3
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706) || \
       defined(CONFIG_CPU_SUBTYPE_SH7707) || \
       defined(CONFIG_CPU_SUBTYPE_SH7709)
 #define SCIF_ERI_IRQ	56
@@ -261,9 +262,12 @@
 #elif defined(CONFIG_CPU_SUBTYPE_SH7708)
 # define ONCHIP_NR_IRQS 32
 #elif defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706) || \
       defined(CONFIG_CPU_SUBTYPE_SH7705)
 # define ONCHIP_NR_IRQS 64	// Actually 61
 # define PINT_NR_IRQS   16
+#elif defined(CONFIG_CPU_SUBTYPE_SH7710)
+# define ONCHIP_NR_IRQS 104
 #elif defined(CONFIG_CPU_SUBTYPE_SH7750)
 # define ONCHIP_NR_IRQS 48	// Actually 44
 #elif defined(CONFIG_CPU_SUBTYPE_SH7751)
@@ -275,7 +279,8 @@
 #elif defined(CONFIG_CPU_SUBTYPE_ST40STB1)
 # define ONCHIP_NR_IRQS 144
 #elif defined(CONFIG_CPU_SUBTYPE_SH7300) || \
-      defined(CONFIG_CPU_SUBTYPE_SH73180)
+      defined(CONFIG_CPU_SUBTYPE_SH73180) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7343)
 # define ONCHIP_NR_IRQS 109
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 # define ONCHIP_NR_IRQS 111
@@ -476,8 +481,10 @@ extern int ipr_irq_demux(int irq);
 
 #define INTC_ICR	0xfffffee0UL
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706) || \
       defined(CONFIG_CPU_SUBTYPE_SH7707) || \
-      defined(CONFIG_CPU_SUBTYPE_SH7709)
+      defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define INTC_IRR0	0xa4000004UL
 #define INTC_IRR1	0xa4000006UL
 #define INTC_IRR2	0xa4000008UL
@@ -496,8 +503,105 @@ extern int ipr_irq_demux(int irq);
 #define INTC_IPRF	0xa4080000UL
 #define INTC_IPRG	0xa4080002UL
 #define INTC_IPRH	0xa4080004UL
-#endif
+#elif defined(CONFIG_CPU_SUBTYPE_SH7710)
+/* Interrupt Controller Registers */
+#undef INTC_IPRA
+#undef INTC_IPRB
+#define INTC_IPRA  	0xA414FEE2UL
+#define INTC_IPRB  	0xA414FEE4UL
+#define INTC_IPRF  	0xA4080000UL
+#define INTC_IPRG  	0xA4080002UL
+#define INTC_IPRH  	0xA4080004UL
+#define INTC_IPRI  	0xA4080006UL
+
+#undef INTC_ICR0
+#undef INTC_ICR1
+#define INTC_ICR0	0xA414FEE0UL
+#define INTC_ICR1	0xA4140010UL
+
+#define INTC_IRR0	0xa4000004UL
+#define INTC_IRR1	0xa4000006UL
+#define INTC_IRR2	0xa4000008UL
+#define INTC_IRR3	0xa400000AUL
+#define INTC_IRR4	0xa400000CUL
+#define INTC_IRR5	0xa4080020UL
+#define INTC_IRR7	0xa4080024UL
+#define INTC_IRR8	0xa4080026UL
+
+/* Interrupt numbers */
+#define TIMER2_IRQ      18
+#define TIMER2_IPR_ADDR INTC_IPRA
+#define TIMER2_IPR_POS   1
+#define TIMER2_PRIORITY  2
 
+/* WDT */
+#define WDT_IRQ		27
+#define WDT_IPR_ADDR	INTC_IPRB
+#define WDT_IPR_POS	 3
+#define WDT_PRIORITY	 2
+
+#define SCIF0_ERI_IRQ	52
+#define SCIF0_RXI_IRQ	53
+#define SCIF0_BRI_IRQ	54
+#define SCIF0_TXI_IRQ	55
+#define SCIF0_IPR_ADDR	INTC_IPRE
+#define SCIF0_IPR_POS	2
+#define SCIF0_PRIORITY	3
+
+#define DMTE4_IRQ	76
+#define DMTE5_IRQ	77
+#define DMA2_IPR_ADDR	INTC_IPRF
+#define DMA2_IPR_POS	2
+#define DMA2_PRIORITY	7
+
+#define IPSEC_IRQ	79
+#define IPSEC_IPR_ADDR	INTC_IPRF
+#define IPSEC_IPR_POS	3
+#define IPSEC_PRIORITY	3
+
+/* EDMAC */
+#define EDMAC0_IRQ	80
+#define EDMAC0_IPR_ADDR	INTC_IPRG
+#define EDMAC0_IPR_POS	3
+#define EDMAC0_PRIORITY	3
+
+#define EDMAC1_IRQ	81
+#define EDMAC1_IPR_ADDR	INTC_IPRG
+#define EDMAC1_IPR_POS	2
+#define EDMAC1_PRIORITY	3
+
+#define EDMAC2_IRQ	82
+#define EDMAC2_IPR_ADDR	INTC_IPRG
+#define EDMAC2_IPR_POS	1
+#define EDMAC2_PRIORITY	3
+
+/* SIOF */
+#define SIOF0_ERI_IRQ	96
+#define SIOF0_TXI_IRQ	97
+#define SIOF0_RXI_IRQ	98
+#define SIOF0_CCI_IRQ	99
+#define SIOF0_IPR_ADDR	INTC_IPRH
+#define SIOF0_IPR_POS	0
+#define SIOF0_PRIORITY	7
+
+#define SIOF1_ERI_IRQ	100
+#define SIOF1_TXI_IRQ	101
+#define SIOF1_RXI_IRQ	102
+#define SIOF1_CCI_IRQ	103
+#define SIOF1_IPR_ADDR	INTC_IPRI
+#define SIOF1_IPR_POS	1
+#define SIOF1_PRIORITY	7
+#endif /* CONFIG_CPU_SUBTYPE_SH7710 */
+
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+#define PORT_PACR	0xa4050100UL
+#define PORT_PBCR	0xa4050102UL
+#define PORT_PCCR	0xa4050104UL
+#define PORT_PETCR	0xa4050106UL
+#define PORT_PADR  	0xa4050120UL
+#define PORT_PBDR  	0xa4050122UL
+#define PORT_PCDR  	0xa4050124UL
+#else
 #define PORT_PACR	0xa4000100UL
 #define PORT_PBCR	0xa4000102UL
 #define PORT_PCCR	0xa4000104UL
@@ -506,6 +610,7 @@ extern int ipr_irq_demux(int irq);
 #define PORT_PBDR  	0xa4000122UL
 #define PORT_PCDR  	0xa4000124UL
 #define PORT_PFDR  	0xa400012aUL
+#endif
 
 #define IRQ0_IRQ	32
 #define IRQ1_IRQ	33
@@ -599,6 +704,8 @@ void intc2_add_clear_irq(int irq, int (*fn)(int));
 
 #endif
 
+extern int shmse_irq_demux(int irq);
+
 static inline int generic_irq_demux(int irq)
 {
 	return irq;
@@ -614,4 +721,8 @@ static inline int generic_irq_demux(int irq)
 #include <asm/irq-sh73180.h>
 #endif
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7343)
+#include <asm/irq-sh7343.h>
+#endif
+
 #endif /* __ASM_SH_IRQ_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index a22732007dd4..3b3ef4f2bf31 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -38,13 +38,15 @@ enum cpu_type {
 	CPU_SH7604,
 
 	/* SH-3 types */
-	CPU_SH7705, CPU_SH7707,  CPU_SH7708, CPU_SH7708S, CPU_SH7708R,
-	CPU_SH7709, CPU_SH7709A, CPU_SH7729, CPU_SH7300,
+	CPU_SH7705, CPU_SH7706, CPU_SH7707,
+	CPU_SH7708, CPU_SH7708S, CPU_SH7708R,
+	CPU_SH7709, CPU_SH7709A, CPU_SH7710,
+	CPU_SH7729, CPU_SH7300,
 
 	/* SH-4 types */
 	CPU_SH7750, CPU_SH7750S, CPU_SH7750R, CPU_SH7751, CPU_SH7751R,
 	CPU_SH7760, CPU_ST40RA, CPU_ST40GX1, CPU_SH4_202, CPU_SH4_501,
-	CPU_SH73180, CPU_SH7770, CPU_SH7780, CPU_SH7781,
+	CPU_SH73180, CPU_SH7343, CPU_SH7770, CPU_SH7780, CPU_SH7781,
 
 	/* Unknown subtype */
 	CPU_SH_NONE
-- 
cgit v1.2.3


From 51e22e7a05c1c6f2e38ac7459d3404e32e543b75 Mon Sep 17 00:00:00 2001
From: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Date: Wed, 27 Sep 2006 17:41:31 +0900
Subject: sh: SHMIN board support.

This adds support for the SHMIN SH7706 board.

Signed-off-by: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                 |   8 +-
 arch/sh/Makefile                |   1 +
 arch/sh/boards/shmin/Makefile   |   5 +
 arch/sh/boards/shmin/setup.c    |  49 +++
 arch/sh/configs/shmin_defconfig | 827 ++++++++++++++++++++++++++++++++++++++++
 arch/sh/tools/mach-types        |   1 +
 include/asm-sh/shmin/shmin.h    |   9 +
 7 files changed, 899 insertions(+), 1 deletion(-)
 create mode 100644 arch/sh/boards/shmin/Makefile
 create mode 100644 arch/sh/boards/shmin/setup.c
 create mode 100644 arch/sh/configs/shmin_defconfig
 create mode 100644 include/asm-sh/shmin/shmin.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 557e7232141c..5bec3af164e5 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -222,6 +222,12 @@ config SH_TITAN
 	  Select Titan if you are configuring for a Nimble Microsystems
 	  NetEngine NP51R.
 
+config SH_SHMIN
+	bool "SHMIN"
+	select CPU_SUBTYPE_SH7706
+	help
+	  Select SHMIN if configureing for the SHMIN board
+
 config SH_UNKNOWN
 	bool "BareCPU"
 	help
@@ -295,7 +301,7 @@ config SH_RTC
 	bool
 	depends on !SH_DREAMCAST && !SH_SATURN && !SH_7300_SOLUTION_ENGINE && \
 		   !SH_73180_SOLUTION_ENGINE && !SH_LANDISK && \
-		   !SH_R7780RP
+		   !SH_R7780RP && !SH_SHMIN
 	default y
 	help
 	  Selecting this option will allow the Linux kernel to emulate
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index d118e48ea1ba..8ad0a286ff8f 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -113,6 +113,7 @@ machdir-$(CONFIG_SH_R7780RP)			:= renesas/r7780rp
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_TITAN)			:= titan
+machdir-$(CONFIG_SH_SHMIN)			:= shmin
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
diff --git a/arch/sh/boards/shmin/Makefile b/arch/sh/boards/shmin/Makefile
new file mode 100644
index 000000000000..3190cc72430e
--- /dev/null
+++ b/arch/sh/boards/shmin/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the SHMIN board.
+#
+
+obj-y	 := setup.o
diff --git a/arch/sh/boards/shmin/setup.c b/arch/sh/boards/shmin/setup.c
new file mode 100644
index 000000000000..a950ec4d765d
--- /dev/null
+++ b/arch/sh/boards/shmin/setup.c
@@ -0,0 +1,49 @@
+/*
+ * arch/sh/boards/shmin/setup.c
+ *
+ * Copyright (C) 2006 Takashi YOSHII
+ *
+ * SHMIN Support.
+ */
+#include <linux/init.h>
+#include <asm/machvec.h>
+#include <asm/shmin/shmin.h>
+#include <asm/clock.h>
+#include <asm/irq.h>
+#include <asm/io.h>
+
+#define PFC_PHCR	0xa400010e
+
+const char *get_system_type(void)
+{
+	return "SHMIN";
+}
+
+static void __init init_shmin_irq(void)
+{
+	ctrl_outw(0x2a00, PFC_PHCR);	// IRQ0-3=IRQ
+	ctrl_outw(0x0aaa, INTC_ICR1);	// IRQ0-3=IRQ-mode,Low-active.
+}
+
+void __init platform_setup(void)
+{
+}
+
+static void __iomem *shmin_ioport_map(unsigned long port, unsigned int size)
+{
+	static int dummy;
+
+	if ((port & ~0x1f) == SHMIN_NE_BASE)
+		return (void __iomem *)(SHMIN_IO_BASE + port);
+
+	dummy = 0;
+
+	return &dummy;
+
+}
+
+struct sh_machine_vector mv_shmin __initmv = {
+	.mv_init_irq	= init_shmin_irq,
+	.mv_ioport_map	= shmin_ioport_map,
+};
+ALIAS_MV(shmin)
diff --git a/arch/sh/configs/shmin_defconfig b/arch/sh/configs/shmin_defconfig
new file mode 100644
index 000000000000..382b3bd3963b
--- /dev/null
+++ b/arch/sh/configs/shmin_defconfig
@@ -0,0 +1,827 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.17
+# Wed Aug  2 01:45:03 2006
+#
+CONFIG_SUPERH=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_PTRACE=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SWAP is not set
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_SYSCTL is not set
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_UID16 is not set
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+# CONFIG_BUG is not set
+# CONFIG_ELF_CORE is not set
+# CONFIG_BASE_FULL is not set
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SHMEM is not set
+# CONFIG_SLAB is not set
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=1
+CONFIG_SLOB=y
+CONFIG_OBSOLETE_INTERMODULE=y
+
+#
+# Loadable module support
+#
+# CONFIG_MODULES is not set
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+# CONFIG_LSF is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7709_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R7780RP is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+CONFIG_SH_SHMIN=y
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH3=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+CONFIG_CPU_SUBTYPE_SH7706=y
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_PAGE_OFFSET=0x80000000
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x00800000
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+# CONFIG_SH_FPU_EMU is not set
+# CONFIG_SH_DSP is not set
+# CONFIG_SH_ADC is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_SR_RB=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ=32000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+# CONFIG_SH_DMA is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_KEXEC is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00210000
+# CONFIG_UBC_WAKEUP is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttySC1,115200 root=1f01 mtdparts=phys_mapped_flash:64k(firm)ro,-(sys) netdev=34,0x300,eth0 "
+
+#
+# Bus options
+#
+# CONFIG_PCI is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options (EXPERIMENTAL)
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+# CONFIG_PACKET is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_FW_LOADER is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+CONFIG_MTD_CMDLINE_PARTS=y
+
+#
+# User Modules And Translation Layers
+#
+# CONFIG_MTD_CHAR is not set
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+CONFIG_MTD_JEDECPROBE=y
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+# CONFIG_MTD_RAM is not set
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_OBSOLETE_CHIPS is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+CONFIG_MTD_PHYSMAP=y
+CONFIG_MTD_PHYSMAP_START=0xa0000000
+CONFIG_MTD_PHYSMAP_LEN=0x80000
+CONFIG_MTD_PHYSMAP_BANKWIDTH=1
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+
+#
+# NAND Flash Device Drivers
+#
+# CONFIG_MTD_NAND is not set
+
+#
+# OneNAND Flash Device Drivers
+#
+# CONFIG_MTD_ONENAND is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_RAM is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_STNIC is not set
+# CONFIG_SMC91X is not set
+CONFIG_NE2000=y
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_NR_UARTS=2
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+CONFIG_VIDEO_V4L2=y
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+# CONFIG_SYSFS is not set
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+CONFIG_CRAMFS=y
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_UNWIND_INFO is not set
+CONFIG_SH_STANDARD_BIOS=y
+CONFIG_EARLY_PRINTK=y
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 1f7f2c255ca2..ffd218083876 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -29,3 +29,4 @@ LANDISK			SH_LANDISK
 R7780RP			SH_R7780RP
 R7780MP			SH_R7780MP
 TITAN			SH_TITAN
+SHMIN			SH_SHMIN
diff --git a/include/asm-sh/shmin/shmin.h b/include/asm-sh/shmin/shmin.h
new file mode 100644
index 000000000000..36ba138a81fb
--- /dev/null
+++ b/include/asm-sh/shmin/shmin.h
@@ -0,0 +1,9 @@
+#ifndef __ASM_SH_SHMIN_H
+#define __ASM_SH_SHMIN_H
+
+#define SHMIN_IO_BASE 0xb0000000UL
+
+#define SHMIN_NE_IRQ IRQ2_IRQ
+#define SHMIN_NE_BASE 0x300
+
+#endif
-- 
cgit v1.2.3


From 91550f715b7f7707b5ab5b9b0cd455bda8505954 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:45:01 +0900
Subject: sh: Kill off the rest of the legacy rtc mess.

With the new RTC class driver, we can get rid of most of the
old left over cruft.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig               |  13 ----
 arch/sh/boards/snapgear/rtc.c |  27 +------
 arch/sh/kernel/cpu/Makefile   |   1 -
 arch/sh/kernel/cpu/irq/ipr.c  |   2 +-
 arch/sh/kernel/cpu/rtc.c      | 128 --------------------------------
 arch/sh/kernel/time.c         |  41 +++++-----
 drivers/char/Kconfig          |   2 +-
 include/asm-sh/mc146818rtc.h  | 169 ------------------------------------------
 include/asm-sh/rtc.h          |  21 ------
 9 files changed, 23 insertions(+), 381 deletions(-)
 delete mode 100644 arch/sh/kernel/cpu/rtc.c

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 5bec3af164e5..c7e3596dcde1 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -296,19 +296,6 @@ config CPU_LITTLE_ENDIAN
 	  endian byte order. These modes require different kernels. Say Y if
 	  your machine is little endian, N if it's a big endian machine.
 
-# The SH7750 RTC module is disabled in the Dreamcast
-config SH_RTC
-	bool
-	depends on !SH_DREAMCAST && !SH_SATURN && !SH_7300_SOLUTION_ENGINE && \
-		   !SH_73180_SOLUTION_ENGINE && !SH_LANDISK && \
-		   !SH_R7780RP && !SH_SHMIN
-	default y
-	help
-	  Selecting this option will allow the Linux kernel to emulate
-	  PC's RTC.
-
-	  If unsure, say N.
-
 config SH_FPU
 	bool "FPU support"
 	depends on !CPU_SH3
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index 722479deb55a..1659fdd6695a 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -19,9 +19,7 @@
 #include <linux/mc146818rtc.h>
 #include <asm/io.h>
 
-/****************************************************************************/
-
-static int use_ds1302 = 0;
+static int use_ds1302;
 
 /****************************************************************************/
 /*
@@ -79,10 +77,6 @@ static unsigned int ds1302_readbyte(unsigned int addr)
 	unsigned int	val;
 	unsigned long	flags;
 
-#if 0
-	printk("SnapGear RTC: ds1302_readbyte(addr=%x)\n", addr);
-#endif
-
 	local_irq_save(flags);
 	set_dirp(get_dirp() | RTC_RESET | RTC_IODATA | RTC_SCLK);
 	set_dp(get_dp() & ~(RTC_RESET | RTC_IODATA | RTC_SCLK));
@@ -101,10 +95,6 @@ static void ds1302_writebyte(unsigned int addr, unsigned int val)
 {
 	unsigned long	flags;
 
-#if 0
-	printk("SnapGear RTC: ds1302_writebyte(addr=%x)\n", addr);
-#endif
-
 	local_irq_save(flags);
 	set_dirp(get_dirp() | RTC_RESET | RTC_IODATA | RTC_SCLK);
 	set_dp(get_dp() & ~(RTC_RESET | RTC_IODATA | RTC_SCLK));
@@ -167,9 +157,6 @@ void __init secureedge5410_rtc_init(void)
 	if (use_ds1302) {
 		rtc_sh_get_time = snapgear_rtc_gettimeofday;
 		rtc_sh_set_time = snapgear_rtc_settimeofday;
-	} else {
-		rtc_sh_get_time = sh_rtc_gettimeofday;
-		rtc_sh_set_time = sh_rtc_settimeofday;
 	}
 		
 	printk("SnapGear RTC: using %s rtc.\n", use_ds1302 ? "ds1302" : "internal");
@@ -184,10 +171,8 @@ void snapgear_rtc_gettimeofday(struct timespec *ts)
 {
 	unsigned int sec, min, hr, day, mon, yr;
 
-	if (!use_ds1302) {
-		sh_rtc_gettimeofday(ts);
+	if (!use_ds1302)
 		return;
-	}
 
  	sec = bcd2int(ds1302_readbyte(RTC_ADDR_SEC));
  	min = bcd2int(ds1302_readbyte(RTC_ADDR_MIN));
@@ -228,7 +213,7 @@ int snapgear_rtc_settimeofday(const time_t secs)
 	unsigned long nowtime;
 
 	if (!use_ds1302)
-		return sh_rtc_settimeofday(secs);
+		return 0;
 
 /*
  *	This is called direct from the kernel timer handling code.
@@ -237,10 +222,6 @@ int snapgear_rtc_settimeofday(const time_t secs)
 
 	nowtime = secs;
 
-#if 1
-	printk("SnapGear RTC: snapgear_rtc_settimeofday(nowtime=%ld)\n", nowtime);
-#endif
-
 	/* STOP RTC */
 	ds1302_writebyte(RTC_ADDR_SEC, ds1302_readbyte(RTC_ADDR_SEC) | 0x80);
 
@@ -326,5 +307,3 @@ void secureedge5410_cmos_write(unsigned char val, int addr)
 	default:                                                      break;
 	}
 }
-
-/****************************************************************************/
diff --git a/arch/sh/kernel/cpu/Makefile b/arch/sh/kernel/cpu/Makefile
index 59d5b748752f..fb5dac069382 100644
--- a/arch/sh/kernel/cpu/Makefile
+++ b/arch/sh/kernel/cpu/Makefile
@@ -8,6 +8,5 @@ obj-$(CONFIG_CPU_SH2)		+= sh2/
 obj-$(CONFIG_CPU_SH3)		+= sh3/
 obj-$(CONFIG_CPU_SH4)		+= sh4/
 
-obj-$(CONFIG_SH_RTC)		+= rtc.o
 obj-$(CONFIG_UBC_WAKEUP)	+= ubc.o
 obj-$(CONFIG_SH_ADC)		+= adc.o
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index d2b715a04c7e..00edee971bcb 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -124,7 +124,7 @@ void __init init_IRQ(void)
 #ifndef CONFIG_CPU_SUBTYPE_SH7780
 	make_ipr_irq(TIMER_IRQ, TIMER_IPR_ADDR, TIMER_IPR_POS, TIMER_PRIORITY);
 	make_ipr_irq(TIMER1_IRQ, TIMER1_IPR_ADDR, TIMER1_IPR_POS, TIMER1_PRIORITY);
-#if defined(CONFIG_SH_RTC)
+#ifdef RTC_IRQ
 	make_ipr_irq(RTC_IRQ, RTC_IPR_ADDR, RTC_IPR_POS, RTC_PRIORITY);
 #endif
 
diff --git a/arch/sh/kernel/cpu/rtc.c b/arch/sh/kernel/cpu/rtc.c
deleted file mode 100644
index 4304cf75cfa2..000000000000
--- a/arch/sh/kernel/cpu/rtc.c
+++ /dev/null
@@ -1,128 +0,0 @@
-/*
- * linux/arch/sh/kernel/rtc.c -- SH3 / SH4 on-chip RTC support
- *
- *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
- *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
- */
-
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/time.h>
-#include <linux/bcd.h>
-#include <asm/io.h>
-#include <asm/rtc.h>
-
-void sh_rtc_gettimeofday(struct timespec *ts)
-{
-	unsigned int sec128, sec, sec2, min, hr, wk, day, mon, yr, yr100, cf_bit;
-	unsigned long flags;
-
- again:
-	do {
-		local_irq_save(flags);
-		ctrl_outb(0, RCR1);  /* Clear CF-bit */
-		sec128 = ctrl_inb(R64CNT);
-		sec = ctrl_inb(RSECCNT);
-		min = ctrl_inb(RMINCNT);
-		hr  = ctrl_inb(RHRCNT);
-		wk  = ctrl_inb(RWKCNT);
-		day = ctrl_inb(RDAYCNT);
-		mon = ctrl_inb(RMONCNT);
-#if defined(CONFIG_CPU_SH4)
-		yr  = ctrl_inw(RYRCNT);
-		yr100 = (yr >> 8);
-		yr &= 0xff;
-#else
-		yr  = ctrl_inb(RYRCNT);
-		yr100 = (yr == 0x99) ? 0x19 : 0x20;
-#endif
-		sec2 = ctrl_inb(R64CNT);
-		cf_bit = ctrl_inb(RCR1) & RCR1_CF;
-		local_irq_restore(flags);
-	} while (cf_bit != 0 || ((sec128 ^ sec2) & RTC_BIT_INVERTED) != 0);
-
-	BCD_TO_BIN(yr100);
-	BCD_TO_BIN(yr);
-	BCD_TO_BIN(mon);
-	BCD_TO_BIN(day);
-	BCD_TO_BIN(hr);
-	BCD_TO_BIN(min);
-	BCD_TO_BIN(sec);
-
-	if (yr > 99 || mon < 1 || mon > 12 || day > 31 || day < 1 ||
-	    hr > 23 || min > 59 || sec > 59) {
-		printk(KERN_ERR
-		       "SH RTC: invalid value, resetting to 1 Jan 2000\n");
-		local_irq_save(flags);
-		ctrl_outb(RCR2_RESET, RCR2);  /* Reset & Stop */
-		ctrl_outb(0, RSECCNT);
-		ctrl_outb(0, RMINCNT);
-		ctrl_outb(0, RHRCNT);
-		ctrl_outb(6, RWKCNT);
-		ctrl_outb(1, RDAYCNT);
-		ctrl_outb(1, RMONCNT);
-#if defined(CONFIG_CPU_SH4)
-		ctrl_outw(0x2000, RYRCNT);
-#else
-		ctrl_outb(0, RYRCNT);
-#endif
-		ctrl_outb(RCR2_RTCEN|RCR2_START, RCR2);  /* Start */
-		goto again;
-	}
-
-#if RTC_BIT_INVERTED != 0
-	if ((sec128 & RTC_BIT_INVERTED))
-		sec--;
-#endif
-
-	ts->tv_sec = mktime(yr100 * 100 + yr, mon, day, hr, min, sec);
-	ts->tv_nsec = ((sec128 * 1000000) / 128) * 1000;
-}
-
-/*
- * Changed to only care about tv_sec, and not the full timespec struct
- * (i.e. tv_nsec).  It can easily be switched to timespec for future cpus
- * that support setting usec or nsec RTC values.
- */
-int sh_rtc_settimeofday(const time_t secs)
-{
-	int retval = 0;
-	int real_seconds, real_minutes, cmos_minutes;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ctrl_outb(RCR2_RESET, RCR2);  /* Reset pre-scaler & stop RTC */
-
-	cmos_minutes = ctrl_inb(RMINCNT);
-	BCD_TO_BIN(cmos_minutes);
-
-	/*
-	 * since we're only adjusting minutes and seconds,
-	 * don't interfere with hour overflow. This avoids
-	 * messing with unknown time zones but requires your
-	 * RTC not to be off by more than 15 minutes
-	 */
-	real_seconds = secs % 60;
-	real_minutes = secs / 60;
-	if (((abs(real_minutes - cmos_minutes) + 15)/30) & 1)
-		real_minutes += 30;	/* correct for half hour time zone */
-	real_minutes %= 60;
-
-	if (abs(real_minutes - cmos_minutes) < 30) {
-		BIN_TO_BCD(real_seconds);
-		BIN_TO_BCD(real_minutes);
-		ctrl_outb(real_seconds, RSECCNT);
-		ctrl_outb(real_minutes, RMINCNT);
-	} else {
-		printk(KERN_WARNING
-		       "set_rtc_time: can't update from %d to %d\n",
-		       cmos_minutes, real_minutes);
-		retval = -1;
-	}
-
-	ctrl_outb(RCR2_RTCEN|RCR2_START, RCR2);  /* Start RTC */
-	local_irq_restore(flags);
-
-	return retval;
-}
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index 8acd70bffe76..149d9713eddf 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -3,13 +3,12 @@
  *
  *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
  *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
- *  Copyright (C) 2002, 2003, 2004, 2005  Paul Mundt
+ *  Copyright (C) 2002 - 2006  Paul Mundt
  *  Copyright (C) 2002  M. R. Brown  <mrbrown@linux-sh.org>
  *
  *  Some code taken from i386 version.
  *    Copyright (C) 1991, 1992, 1995  Linus Torvalds
  */
-
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -26,15 +25,20 @@ struct sys_timer *sys_timer;
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
 
-/* XXX: Can we initialize this in a routine somewhere?  Dreamcast doesn't want
- * these routines anywhere... */
-#ifdef CONFIG_SH_RTC
-void (*rtc_sh_get_time)(struct timespec *) = sh_rtc_gettimeofday;
-int (*rtc_sh_set_time)(const time_t) = sh_rtc_settimeofday;
-#else
-void (*rtc_sh_get_time)(struct timespec *);
-int (*rtc_sh_set_time)(const time_t);
-#endif
+/* Dummy RTC ops */
+static void null_rtc_get_time(struct timespec *tv)
+{
+	tv->tv_sec = mktime(2000, 1, 1, 0, 0, 0);
+	tv->tv_nsec = 0;
+}
+
+static int null_rtc_set_time(const time_t secs)
+{
+	return 0;
+}
+
+void (*rtc_sh_get_time)(struct timespec *) = null_rtc_get_time;
+int (*rtc_sh_set_time)(const time_t) = null_rtc_set_time;
 
 /*
  * Scheduler clock - returns current time in nanosec units.
@@ -70,7 +74,6 @@ void do_gettimeofday(struct timeval *tv)
 	tv->tv_sec = sec;
 	tv->tv_usec = usec;
 }
-
 EXPORT_SYMBOL(do_gettimeofday);
 
 int do_settimeofday(struct timespec *tv)
@@ -103,7 +106,6 @@ int do_settimeofday(struct timespec *tv)
 
 	return 0;
 }
-
 EXPORT_SYMBOL(do_settimeofday);
 
 /* last time the RTC clock got updated */
@@ -181,7 +183,6 @@ static int __init timer_init_sysfs(void)
 	sys_timer->dev.cls = &timer_sysclass;
 	return sysdev_register(&sys_timer->dev);
 }
-
 device_initcall(timer_init_sysfs);
 
 void (*board_time_init)(void);
@@ -193,15 +194,9 @@ void __init time_init(void)
 
 	clk_init();
 
-	if (rtc_sh_get_time) {
-		rtc_sh_get_time(&xtime);
-	} else {
-		xtime.tv_sec = mktime(2000, 1, 1, 0, 0, 0);
-		xtime.tv_nsec = 0;
-	}
-
-        set_normalized_timespec(&wall_to_monotonic,
-                                -xtime.tv_sec, -xtime.tv_nsec);
+	rtc_sh_get_time(&xtime);
+	set_normalized_timespec(&wall_to_monotonic,
+				-xtime.tv_sec, -xtime.tv_nsec);
 
 	/*
 	 * Find the timer to use as the system timer, it will be
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 52ea94b891f5..7f0d323b4725 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -739,7 +739,7 @@ config NVRAM
 
 config RTC
 	tristate "Enhanced Real Time Clock Support"
-	depends on !PPC && !PARISC && !IA64 && !M68K && (!SPARC || PCI) && !FRV && !ARM
+	depends on !PPC && !PARISC && !IA64 && !M68K && (!SPARC || PCI) && !FRV && !ARM && !SUPERH
 	---help---
 	  If you say Y here and create a character special file /dev/rtc with
 	  major number 10 and minor number 135 using mknod ("man mknod"), you
diff --git a/include/asm-sh/mc146818rtc.h b/include/asm-sh/mc146818rtc.h
index adc6e67c6b75..0aee96a97330 100644
--- a/include/asm-sh/mc146818rtc.h
+++ b/include/asm-sh/mc146818rtc.h
@@ -4,173 +4,4 @@
 #ifndef _ASM_MC146818RTC_H
 #define _ASM_MC146818RTC_H
 
-#ifdef CONFIG_SH_MPC1211
-#undef  _ASM_MC146818RTC_H
-#undef  RTC_IRQ
-#include <asm/mpc1211/mc146818rtc.h>
-#else
-
-#include <asm/rtc.h>
-
-#define RTC_ALWAYS_BCD	1
-
-/* FIXME:RTC Interrupt feature is not implemented yet. */
-#undef  RTC_IRQ
-#define RTC_IRQ		0
-
-#if defined(CONFIG_CPU_SH3)
-#define RTC_PORT(n)		(R64CNT+(n)*2)
-#define CMOS_READ(addr)		__CMOS_READ(addr,b)
-#define CMOS_WRITE(val,addr)	__CMOS_WRITE(val,addr,b)
-
-#elif defined(CONFIG_SH_SECUREEDGE5410)
-#include <asm/snapgear.h>
-
-#define RTC_PORT(n)             SECUREEDGE_IOPORT_ADDR
-#define CMOS_READ(addr)         secureedge5410_cmos_read(addr)
-#define CMOS_WRITE(val,addr)    secureedge5410_cmos_write(val,addr)
-extern unsigned char secureedge5410_cmos_read(int addr);
-extern void secureedge5410_cmos_write(unsigned char val, int addr);
-
-#elif defined(CONFIG_CPU_SH4)
-#define RTC_PORT(n)		(R64CNT+(n)*4)
-#define CMOS_READ(addr)		__CMOS_READ(addr,w)
-#define CMOS_WRITE(val,addr)	__CMOS_WRITE(val,addr,w)
-#endif
-
-#define __CMOS_READ(addr, s) ({						\
-	unsigned char val=0, rcr1, rcr2, r64cnt, retry;			\
-	switch(addr) {							\
-		case RTC_SECONDS:					\
-			val = ctrl_inb(RSECCNT);			\
-			break;						\
-		case RTC_SECONDS_ALARM:					\
-			val = ctrl_inb(RSECAR);				\
-			break;						\
-		case RTC_MINUTES:					\
-			val = ctrl_inb(RMINCNT);			\
-			break;						\
-		case RTC_MINUTES_ALARM:					\
-			val = ctrl_inb(RMINAR);				\
-			break;						\
-		case RTC_HOURS:						\
-			val = ctrl_inb(RHRCNT);				\
-			break;						\
-		case RTC_HOURS_ALARM:					\
-			val = ctrl_inb(RHRAR);				\
-			break;						\
-		case RTC_DAY_OF_WEEK:					\
-			val = ctrl_inb(RWKCNT);				\
-			break;						\
-		case RTC_DAY_OF_MONTH:					\
-			val = ctrl_inb(RDAYCNT);			\
-			break;						\
-		case RTC_MONTH:						\
-			val = ctrl_inb(RMONCNT);			\
-			break;						\
-		case RTC_YEAR:						\
-			val = ctrl_in##s(RYRCNT);			\
-			break;						\
-		case RTC_REG_A: /* RTC_FREQ_SELECT */			\
-			rcr2 = ctrl_inb(RCR2);				\
-			val = (rcr2 & RCR2_PESMASK) >> 4;		\
-			rcr1 = ctrl_inb(RCR1);				\
-			rcr1 = (rcr1 & (RCR1_CIE | RCR1_AIE)) | RCR1_AF;\
-			retry = 0;					\
-			do {						\
-				ctrl_outb(rcr1, RCR1); /* clear CF */	\
-				r64cnt = ctrl_inb(R64CNT);		\
-			} while((ctrl_inb(RCR1) & RCR1_CF) && retry++ < 1000);\
-			r64cnt ^= RTC_BIT_INVERTED;			\
-			if(r64cnt == 0x7f || r64cnt == 0)		\
-				val |= RTC_UIP;				\
-			break;						\
-		case RTC_REG_B:	/* RTC_CONTROL */			\
-			rcr1 = ctrl_inb(RCR1);				\
-			rcr2 = ctrl_inb(RCR2);				\
-			if(rcr1 & RCR1_CIE)	val |= RTC_UIE;		\
-			if(rcr1 & RCR1_AIE)	val |= RTC_AIE;		\
-			if(rcr2 & RCR2_PESMASK)	val |= RTC_PIE;		\
-			if(!(rcr2 & RCR2_START))val |= RTC_SET;		\
-			val |= RTC_24H;					\
-			break;						\
-		case RTC_REG_C:	/* RTC_INTR_FLAGS */			\
-			rcr1 = ctrl_inb(RCR1);				\
-			rcr1 &= ~(RCR1_CF | RCR1_AF);			\
-			ctrl_outb(rcr1, RCR1);				\
-			rcr2 = ctrl_inb(RCR2);				\
-			rcr2 &= ~RCR2_PEF;				\
-			ctrl_outb(rcr2, RCR2);				\
-			break;						\
-		case RTC_REG_D:	/* RTC_VALID */				\
-			/* Always valid ... */				\
-			val = RTC_VRT;					\
-			break;						\
-		default:						\
-			break;						\
-	}								\
-	val;								\
-})
-
-#define __CMOS_WRITE(val, addr, s) ({					\
-	unsigned char rcr1,rcr2;					\
-	switch(addr) {							\
-		case RTC_SECONDS:					\
-			ctrl_outb(val, RSECCNT);			\
-			break;						\
-		case RTC_SECONDS_ALARM:					\
-			ctrl_outb(val, RSECAR);				\
-			break;						\
-		case RTC_MINUTES:					\
-			ctrl_outb(val, RMINCNT);			\
-			break;						\
-		case RTC_MINUTES_ALARM:					\
-			ctrl_outb(val, RMINAR);				\
-			break;						\
-		case RTC_HOURS:						\
-			ctrl_outb(val, RHRCNT);				\
-			break;						\
-		case RTC_HOURS_ALARM:					\
-			ctrl_outb(val, RHRAR);				\
-			break;						\
-		case RTC_DAY_OF_WEEK:					\
-			ctrl_outb(val, RWKCNT);				\
-			break;						\
-		case RTC_DAY_OF_MONTH:					\
-			ctrl_outb(val, RDAYCNT);			\
-			break;						\
-		case RTC_MONTH:						\
-			ctrl_outb(val, RMONCNT);			\
-			break;						\
-		case RTC_YEAR:						\
-			ctrl_out##s((ctrl_in##s(RYRCNT) & 0xff00) | (val & 0xff), RYRCNT);\
-			break;						\
-		case RTC_REG_A: /* RTC_FREQ_SELECT */			\
-			rcr2 = ctrl_inb(RCR2);				\
-			if((val & RTC_DIV_CTL) == RTC_DIV_RESET2)	\
-				rcr2 |= RCR2_RESET;			\
-			ctrl_outb(rcr2, RCR2);				\
-			break;						\
-		case RTC_REG_B:	/* RTC_CONTROL */			\
-			rcr1 = (ctrl_inb(RCR1) & 0x99) | RCR1_AF;	\
-			if(val & RTC_AIE) rcr1 |= RCR1_AIE;		\
-			else              rcr1 &= ~RCR1_AIE;		\
-			if(val & RTC_UIE) rcr1 |= RCR1_CIE;		\
-			else              rcr1 &= ~RCR1_CIE;		\
-			ctrl_outb(rcr1, RCR1);				\
-			rcr2 = ctrl_inb(RCR2);				\
-			if(val & RTC_SET) rcr2 &= ~RCR2_START;		\
-			else              rcr2 |= RCR2_START;		\
-			ctrl_outb(rcr2, RCR2);				\
-			break;						\
-		case RTC_REG_C:	/* RTC_INTR_FLAGS */			\
-			break;						\
-		case RTC_REG_D:	/* RTC_VALID */				\
-			break;						\
-		default:						\
-			break;						\
-	}								\
-})
-
-#endif /* CONFIG_SH_MPC1211 */
 #endif /* _ASM_MC146818RTC_H */
diff --git a/include/asm-sh/rtc.h b/include/asm-sh/rtc.h
index 4c7be859c9c0..91aacc96151b 100644
--- a/include/asm-sh/rtc.h
+++ b/include/asm-sh/rtc.h
@@ -1,29 +1,8 @@
 #ifndef _ASM_RTC_H
 #define _ASM_RTC_H
-#ifdef __KERNEL__
 
-#include <asm/machvec.h>
-#include <asm/cpu/rtc.h>
-
-extern void sh_rtc_gettimeofday(struct timespec *ts);
-extern int sh_rtc_settimeofday(const time_t secs);
 extern void (*board_time_init)(void);
 extern void (*rtc_sh_get_time)(struct timespec *);
 extern int (*rtc_sh_set_time)(const time_t);
 
-/* RCR1 Bits */
-#define RCR1_CF		0x80	/* Carry Flag             */
-#define RCR1_CIE	0x10	/* Carry Interrupt Enable */
-#define RCR1_AIE	0x08	/* Alarm Interrupt Enable */
-#define RCR1_AF		0x01	/* Alarm Flag             */
-
-/* RCR2 Bits */
-#define RCR2_PEF	0x80	/* PEriodic interrupt Flag */
-#define RCR2_PESMASK	0x70	/* Periodic interrupt Set  */
-#define RCR2_RTCEN	0x08	/* ENable RTC              */
-#define RCR2_ADJ	0x04	/* ADJustment (30-second)  */
-#define RCR2_RESET	0x02	/* Reset bit               */
-#define RCR2_START	0x01	/* Start bit               */
-
-#endif /* __KERNEL__ */
 #endif /* _ASM_RTC_H */
-- 
cgit v1.2.3


From 781125ca58dfbd47635cfc0e408f1f9d7e10b227 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:52:19 +0900
Subject: sh: New atomic ops for SH-4A movli.l/movco.l

SH-4A implements LL/SC instructions, so we implement a simple
set of atomic operations using these.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/atomic.h | 105 +++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 96 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h
index 049eb2dda6b6..8bdc1ba56f73 100644
--- a/include/asm-sh/atomic.h
+++ b/include/asm-sh/atomic.h
@@ -22,49 +22,110 @@ typedef struct { volatile int counter; } atomic_t;
  * forward to code at the end of this object's .text section, then
  * branch back to restart the operation.
  */
-
-static __inline__ void atomic_add(int i, atomic_t * v)
+static inline void atomic_add(int i, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_add	\n"
+"	add	%2, %0				\n"
+"	movco.l	%0, @%3				\n"
+"	bf	1b				\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v += i;
 	local_irq_restore(flags);
+#endif
 }
 
-static __inline__ void atomic_sub(int i, atomic_t *v)
+static inline void atomic_sub(int i, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_sub	\n"
+"	sub	%2, %0				\n"
+"	movco.l	%0, @%3				\n"
+"	bf	1b				\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v -= i;
 	local_irq_restore(flags);
+#endif
 }
 
-static __inline__ int atomic_add_return(int i, atomic_t * v)
+/*
+ * SH-4A note:
+ *
+ * We basically get atomic_xxx_return() for free compared with
+ * atomic_xxx(). movli.l/movco.l require r0 due to the instruction
+ * encoding, so the retval is automatically set without having to
+ * do any special work.
+ */
+static inline int atomic_add_return(int i, atomic_t *v)
 {
-	unsigned long temp, flags;
+	unsigned long temp;
+
+#ifdef CONFIG_CPU_SH4A
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_add_return	\n"
+"	add	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+"	synco						\n"
+	: "=&z" (temp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
+	unsigned long flags;
 
 	local_irq_save(flags);
 	temp = *(long *)v;
 	temp += i;
 	*(long *)v = temp;
 	local_irq_restore(flags);
+#endif
 
 	return temp;
 }
 
 #define atomic_add_negative(a, v)	(atomic_add_return((a), (v)) < 0)
 
-static __inline__ int atomic_sub_return(int i, atomic_t * v)
+static inline int atomic_sub_return(int i, atomic_t *v)
 {
-	unsigned long temp, flags;
+	unsigned long temp;
+
+#ifdef CONFIG_CPU_SH4A
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_sub_return	\n"
+"	sub	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+"	synco						\n"
+	: "=&z" (temp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
+	unsigned long flags;
 
 	local_irq_save(flags);
 	temp = *(long *)v;
 	temp -= i;
 	*(long *)v = temp;
 	local_irq_restore(flags);
+#endif
 
 	return temp;
 }
@@ -119,22 +180,48 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u)
 }
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
-static __inline__ void atomic_clear_mask(unsigned int mask, atomic_t *v)
+static inline void atomic_clear_mask(unsigned int mask, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_clear_mask	\n"
+"	and	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (~mask), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v &= ~mask;
 	local_irq_restore(flags);
+#endif
 }
 
-static __inline__ void atomic_set_mask(unsigned int mask, atomic_t *v)
+static inline void atomic_set_mask(unsigned int mask, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_set_mask	\n"
+"	or	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (mask), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v |= mask;
 	local_irq_restore(flags);
+#endif
 }
 
 /* Atomic operations are already serializing on SH */
-- 
cgit v1.2.3


From 5a4053b23262afefa748e1e4c439931d4c27693b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:00:19 +0900
Subject: sh: Kill off dead boards.

None of these have been maintained in years, and no one seems to
be interested in doing so, so just get rid of them.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                      |  43 +--
 arch/sh/Makefile                     |   7 -
 arch/sh/boards/adx/Makefile          |   6 -
 arch/sh/boards/adx/irq.c             |  31 --
 arch/sh/boards/adx/irq_maskreg.c     | 106 -------
 arch/sh/boards/adx/setup.c           |  56 ----
 arch/sh/boards/cat68701/Makefile     |   6 -
 arch/sh/boards/cat68701/irq.c        |  28 --
 arch/sh/boards/cat68701/setup.c      |  85 -----
 arch/sh/boards/cqreek/Makefile       |   6 -
 arch/sh/boards/cqreek/irq.c          | 128 --------
 arch/sh/boards/cqreek/setup.c        | 100 ------
 arch/sh/boards/dmida/Makefile        |   7 -
 arch/sh/boards/dmida/mach.c          |  59 ----
 arch/sh/boards/harp/Makefile         |   8 -
 arch/sh/boards/harp/irq.c            | 147 ---------
 arch/sh/boards/harp/led.c            |  51 ---
 arch/sh/boards/harp/mach.c           |  62 ----
 arch/sh/boards/harp/pcidma.c         |  42 ---
 arch/sh/boards/harp/setup.c          |  90 ------
 arch/sh/boards/overdrive/Makefile    |   8 -
 arch/sh/boards/overdrive/fpga.c      | 133 --------
 arch/sh/boards/overdrive/galileo.c   | 587 -----------------------------------
 arch/sh/boards/overdrive/io.c        | 172 ----------
 arch/sh/boards/overdrive/irq.c       | 191 ------------
 arch/sh/boards/overdrive/led.c       |  58 ----
 arch/sh/boards/overdrive/mach.c      |  62 ----
 arch/sh/boards/overdrive/pcidma.c    |  46 ---
 arch/sh/boards/overdrive/setup.c     |  36 ---
 arch/sh/boards/sh2000/Makefile       |   6 -
 arch/sh/boards/sh2000/setup.c        |  70 -----
 arch/sh/tools/mach-types             |   2 -
 include/asm-sh/adx/io.h              |  86 -----
 include/asm-sh/cat68701/io.h         |  22 --
 include/asm-sh/cqreek/cqreek.h       |  27 --
 include/asm-sh/dmida/io.h            |  10 -
 include/asm-sh/harp/harp.h           |  43 ---
 include/asm-sh/harp/io.h             |  10 -
 include/asm-sh/overdrive/fpga.h      |  15 -
 include/asm-sh/overdrive/gt64111.h   | 109 -------
 include/asm-sh/overdrive/io.h        |  39 ---
 include/asm-sh/overdrive/overdrive.h |  88 ------
 include/asm-sh/sh2000/sh2000.h       |   8 -
 43 files changed, 4 insertions(+), 2892 deletions(-)
 delete mode 100644 arch/sh/boards/adx/Makefile
 delete mode 100644 arch/sh/boards/adx/irq.c
 delete mode 100644 arch/sh/boards/adx/irq_maskreg.c
 delete mode 100644 arch/sh/boards/adx/setup.c
 delete mode 100644 arch/sh/boards/cat68701/Makefile
 delete mode 100644 arch/sh/boards/cat68701/irq.c
 delete mode 100644 arch/sh/boards/cat68701/setup.c
 delete mode 100644 arch/sh/boards/cqreek/Makefile
 delete mode 100644 arch/sh/boards/cqreek/irq.c
 delete mode 100644 arch/sh/boards/cqreek/setup.c
 delete mode 100644 arch/sh/boards/dmida/Makefile
 delete mode 100644 arch/sh/boards/dmida/mach.c
 delete mode 100644 arch/sh/boards/harp/Makefile
 delete mode 100644 arch/sh/boards/harp/irq.c
 delete mode 100644 arch/sh/boards/harp/led.c
 delete mode 100644 arch/sh/boards/harp/mach.c
 delete mode 100644 arch/sh/boards/harp/pcidma.c
 delete mode 100644 arch/sh/boards/harp/setup.c
 delete mode 100644 arch/sh/boards/overdrive/Makefile
 delete mode 100644 arch/sh/boards/overdrive/fpga.c
 delete mode 100644 arch/sh/boards/overdrive/galileo.c
 delete mode 100644 arch/sh/boards/overdrive/io.c
 delete mode 100644 arch/sh/boards/overdrive/irq.c
 delete mode 100644 arch/sh/boards/overdrive/led.c
 delete mode 100644 arch/sh/boards/overdrive/mach.c
 delete mode 100644 arch/sh/boards/overdrive/pcidma.c
 delete mode 100644 arch/sh/boards/overdrive/setup.c
 delete mode 100644 arch/sh/boards/sh2000/Makefile
 delete mode 100644 arch/sh/boards/sh2000/setup.c
 delete mode 100644 include/asm-sh/adx/io.h
 delete mode 100644 include/asm-sh/cat68701/io.h
 delete mode 100644 include/asm-sh/cqreek/cqreek.h
 delete mode 100644 include/asm-sh/dmida/io.h
 delete mode 100644 include/asm-sh/harp/harp.h
 delete mode 100644 include/asm-sh/harp/io.h
 delete mode 100644 include/asm-sh/overdrive/fpga.h
 delete mode 100644 include/asm-sh/overdrive/gt64111.h
 delete mode 100644 include/asm-sh/overdrive/io.h
 delete mode 100644 include/asm-sh/overdrive/overdrive.h
 delete mode 100644 include/asm-sh/sh2000/sh2000.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index c7e3596dcde1..ddf2cc56868c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -89,12 +89,6 @@ config SH_7751_SYSTEMH
 	  Select SystemH if you are configuring for a Renesas SystemH
 	  7751R evaluation board.
 
-config SH_STB1_HARP
-	bool "STB1_Harp"
-
-config SH_STB1_OVERDRIVE
-	bool "STB1_Overdrive"
-
 config SH_HP6XX
 	bool "HP6XX"
 	help
@@ -102,19 +96,6 @@ config SH_HP6XX
 	  More information (hardware only) at
 	  <http://www.hp.com/jornada/>.
 
-config SH_CQREEK
-	bool "CqREEK"
-	help
-	  Select CqREEK if configuring for a CqREEK SH7708 or SH7750.
-	  More information at
-	  <http://sources.redhat.com/ecos/hardware.html#SuperH>.
-
-config SH_DMIDA
-	bool "DMIDA"
-	help
-	  Select DMIDA if configuring for a DataMyte 4000 Industrial
-	  Digital Assistant. More information at <http://www.dmida.com/>.
-
 config SH_EC3104
 	bool "EC3104"
 	help
@@ -136,25 +117,9 @@ config SH_DREAMCAST
 	  <http://www.m17n.org/linux-sh/dreamcast/>.  There is a
 	  Dreamcast project is at <http://linuxdc.sourceforge.net/>.
 
-config SH_CAT68701
-	bool "CAT68701"
-
 config SH_BIGSUR
 	bool "BigSur"
 
-config SH_SH2000
-	bool "SH2000"
-	select CPU_SUBTYPE_SH7709
-	help
-	  SH-2000 is a single-board computer based around SH7709A chip
-	  intended for embedded applications.
-	  It has an Ethernet interface (CS8900A), direct connected
-	  Compact Flash socket, three serial ports and PC-104 bus.
-	  More information at <http://sh2000.sh-linux.org>.
-
-config SH_ADX
-	bool "ADX"
-
 config SH_MPC1211
 	bool "Interface MPC1211"
 	help
@@ -246,7 +211,7 @@ source "arch/sh/mm/Kconfig"
 
 config CF_ENABLER
 	bool "Compact Flash Enabler support"
-	depends on SH_ADX || SH_SOLUTION_ENGINE || SH_UNKNOWN || SH_CAT68701 || SH_SH03
+	depends on SH_SOLUTION_ENGINE || SH_UNKNOWN || SH_SH03
 	---help---
 	  Compact Flash is a small, removable mass storage device introduced
 	  in 1994 originally as a PCMCIA device.  If you say `Y' here, you
@@ -274,7 +239,7 @@ config CF_AREA5
 	  - "Area5" if CompactFlash is connected to Area 5 (0x14000000)
 	  - "Area6" if it is connected to Area 6 (0x18000000)
 
-	  "Area6" will work for most boards. For ADX, select "Area5".
+	  "Area6" will work for most boards.
 
 config CF_AREA6
 	bool "Area6"
@@ -418,8 +383,8 @@ source "arch/sh/cchips/Kconfig"
 
 config HEARTBEAT
 	bool "Heartbeat LED"
-	depends on SH_MPC1211 || SH_SH03 || SH_CAT68701 || \
-		   SH_STB1_HARP || SH_STB1_OVERDRIVE || SH_BIGSUR || \
+	depends on SH_MPC1211 || SH_SH03 || \
+		   SH_BIGSUR || \
 		   SH_7751_SOLUTION_ENGINE || SH_7300_SOLUTION_ENGINE || \
 		   SH_73180_SOLUTION_ENGINE || SH_SOLUTION_ENGINE || \
 		   SH_RTS7751R2D || SH_SH4202_MICRODEV || SH_LANDISK
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 8ad0a286ff8f..d0168844c3c7 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -90,18 +90,11 @@ machdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se/770x
 machdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se/7751
 machdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)	:= se/7300
 machdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se/73180
-machdir-$(CONFIG_SH_STB1_HARP)			:= harp
-machdir-$(CONFIG_SH_STB1_OVERDRIVE)		:= overdrive
 machdir-$(CONFIG_SH_HP6XX)			:= hp6xx
-machdir-$(CONFIG_SH_CQREEK)			:= cqreek
-machdir-$(CONFIG_SH_DMIDA)			:= dmida
 machdir-$(CONFIG_SH_EC3104)			:= ec3104
 machdir-$(CONFIG_SH_SATURN)			:= saturn
 machdir-$(CONFIG_SH_DREAMCAST)			:= dreamcast
-machdir-$(CONFIG_SH_CAT68701)			:= cat68701
 machdir-$(CONFIG_SH_BIGSUR)			:= bigsur
-machdir-$(CONFIG_SH_SH2000)			:= sh2000
-machdir-$(CONFIG_SH_ADX)			:= adx
 machdir-$(CONFIG_SH_MPC1211)			:= mpc1211
 machdir-$(CONFIG_SH_SH03)			:= sh03
 machdir-$(CONFIG_SH_SECUREEDGE5410)		:= snapgear
diff --git a/arch/sh/boards/adx/Makefile b/arch/sh/boards/adx/Makefile
deleted file mode 100644
index 5b1c531b3991..000000000000
--- a/arch/sh/boards/adx/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for ADX boards
-#
-
-obj-y	 := setup.o irq.o irq_maskreq.o
-
diff --git a/arch/sh/boards/adx/irq.c b/arch/sh/boards/adx/irq.c
deleted file mode 100644
index c6ca409dff98..000000000000
--- a/arch/sh/boards/adx/irq.c
+++ /dev/null
@@ -1,31 +0,0 @@
-/*
- * linux/arch/sh/boards/adx/irq.c
- *
- * Copyright (C) 2001 A&D Co., Ltd.
- *
- * I/O routine and setup routines for A&D ADX Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/irq.h>
-
-void init_adx_IRQ(void)
-{
-        int i;
-
-/*      printk("init_adx_IRQ()\n");*/
-        /* setup irq_mask_register */
-        irq_mask_register = (unsigned short *)0xa6000008;
-
-        /* cover all external interrupt area by maskreg_irq_type
-         * (Actually, irq15 doesn't exist)
-         */
-        for (i = 0; i < 16; i++) {
-                make_maskreg_irq(i);
-                disable_irq(i);
-        }
-}
diff --git a/arch/sh/boards/adx/irq_maskreg.c b/arch/sh/boards/adx/irq_maskreg.c
deleted file mode 100644
index 4b2abe5eb165..000000000000
--- a/arch/sh/boards/adx/irq_maskreg.c
+++ /dev/null
@@ -1,106 +0,0 @@
-/*
- * linux/arch/sh/kernel/irq_maskreg.c
- *
- * Copyright (C) 2001 A&D Co., Ltd. <http://www.aandd.co.jp>
- *
- * This file may be copied or modified under the terms of the GNU
- * General Public License.  See linux/COPYING for more information.
- *
- * Interrupt handling for Simple external interrupt mask register
- *
- * This is for the machine which have single 16 bit register
- * for masking external IRQ individually.
- * Each bit of the register is for masking each interrupt.  
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/machvec.h>
-
-/* address of external interrupt mask register
- * address must be set prior to use these (maybe in init_XXX_irq())
- * XXX : is it better to use .config than specifying it in code? */
-unsigned short *irq_mask_register = 0;
-
-/* forward declaration */
-static unsigned int startup_maskreg_irq(unsigned int irq);
-static void shutdown_maskreg_irq(unsigned int irq);
-static void enable_maskreg_irq(unsigned int irq);
-static void disable_maskreg_irq(unsigned int irq);
-static void mask_and_ack_maskreg(unsigned int);
-static void end_maskreg_irq(unsigned int irq);
-
-/* hw_interrupt_type */
-static struct hw_interrupt_type maskreg_irq_type = {
-	.typename = " Mask Register",
-	.startup = startup_maskreg_irq,
-	.shutdown = shutdown_maskreg_irq,
-	.enable = enable_maskreg_irq,
-	.disable = disable_maskreg_irq,
-	.ack = mask_and_ack_maskreg,
-	.end = end_maskreg_irq
-};
-
-/* actual implementatin */
-static unsigned int startup_maskreg_irq(unsigned int irq)
-{ 
-	enable_maskreg_irq(irq);
-	return 0; /* never anything pending */
-}
-
-static void shutdown_maskreg_irq(unsigned int irq)
-{
-	disable_maskreg_irq(irq);
-}
-
-static void disable_maskreg_irq(unsigned int irq)
-{
-	if (irq_mask_register) {
-		unsigned long flags;
-		unsigned short val, mask = 0x01 << irq;
-
-		/* Set "irq"th bit */
-		local_irq_save(flags);
-		val = ctrl_inw((unsigned long)irq_mask_register);
-		val |= mask;
-		ctrl_outw(val, (unsigned long)irq_mask_register);
-		local_irq_restore(flags);
-	}
-}
-
-static void enable_maskreg_irq(unsigned int irq)
-{
-	if (irq_mask_register) {
-		unsigned long flags;
-		unsigned short val, mask = ~(0x01 << irq);
-
-		/* Clear "irq"th bit */
-		local_irq_save(flags);
-		val = ctrl_inw((unsigned long)irq_mask_register);
-		val &= mask;
-		ctrl_outw(val, (unsigned long)irq_mask_register);
-		local_irq_restore(flags);
-	}
-}
-
-static void mask_and_ack_maskreg(unsigned int irq)
-{
-	disable_maskreg_irq(irq);
-}
-
-static void end_maskreg_irq(unsigned int irq)
-{
-	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
-		enable_maskreg_irq(irq);
-}
-
-void make_maskreg_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &maskreg_irq_type;
-	disable_maskreg_irq(irq);
-}
diff --git a/arch/sh/boards/adx/setup.c b/arch/sh/boards/adx/setup.c
deleted file mode 100644
index 4938d9592343..000000000000
--- a/arch/sh/boards/adx/setup.c
+++ /dev/null
@@ -1,56 +0,0 @@
-/* 
- * linux/arch/sh/board/adx/setup.c
- *
- * Copyright (C) 2001 A&D Co., Ltd.
- *
- * I/O routine and setup routines for A&D ADX Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/machvec.h>
-#include <linux/module.h>
-
-extern void init_adx_IRQ(void);
-extern void *cf_io_base;
-
-const char *get_system_type(void)
-{
-	return "A&D ADX";
-}
-
-unsigned long adx_isa_port2addr(unsigned long offset)
-{
-	/* CompactFlash (IDE) */
-	if (((offset >= 0x1f0) && (offset <= 0x1f7)) || (offset == 0x3f6)) {
-		return (unsigned long)cf_io_base + offset;
-	}
-
-	/* eth0 */
-	if ((offset >= 0x300) && (offset <= 0x30f)) {
-		return 0xa5000000 + offset;	/* COMM BOARD (AREA1) */
-	}
-
-	return offset + 0xb0000000; /* IOBUS (AREA 4)*/
-}
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_adx __initmv = {
-	.mv_nr_irqs		= 48,
-	.mv_isa_port2addr	= adx_isa_port2addr,
-	.mv_init_irq		= init_adx_IRQ,
-};
-ALIAS_MV(adx)
-
-int __init platform_setup(void)
-{
-	/* Nothing to see here .. */
-	return 0;
-}
-
diff --git a/arch/sh/boards/cat68701/Makefile b/arch/sh/boards/cat68701/Makefile
deleted file mode 100644
index 52c1de0a6dfd..000000000000
--- a/arch/sh/boards/cat68701/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for the CAT-68701 specific parts of the kernel
-#
-
-obj-y	 := setup.o irq.o
-
diff --git a/arch/sh/boards/cat68701/irq.c b/arch/sh/boards/cat68701/irq.c
deleted file mode 100644
index f9a6d185fb8b..000000000000
--- a/arch/sh/boards/cat68701/irq.c
+++ /dev/null
@@ -1,28 +0,0 @@
-/*
- * linux/arch/sh/boards/cat68701/irq.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *               2001  Yutaro Ebihara
- *
- * Setup routines for A-ONE Corp CAT-68701 SH7708 Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/irq.h>
-
-int cat68701_irq_demux(int irq)
-{
-        if(irq==13) return 14;
-        if(irq==7)  return 10;
-        return irq;
-}
-
-void init_cat68701_IRQ()
-{
-        make_imask_irq(10);
-        make_imask_irq(14);
-}
diff --git a/arch/sh/boards/cat68701/setup.c b/arch/sh/boards/cat68701/setup.c
deleted file mode 100644
index 90e5175df227..000000000000
--- a/arch/sh/boards/cat68701/setup.c
+++ /dev/null
@@ -1,85 +0,0 @@
-/* 
- * linux/arch/sh/boards/cat68701/setup.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *               2001  Yutaro Ebihara
- *
- * Setup routines for A-ONE Corp CAT-68701 SH7708 Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/io.h>
-#include <asm/machvec.h>
-#include <asm/mach/io.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/sched.h>
-
-const char *get_system_type(void)
-{
-	return "CAT-68701";
-}
-
-#ifdef CONFIG_HEARTBEAT
-void heartbeat_cat68701()
-{
-        static unsigned int cnt = 0, period = 0 , bit = 0;
-        cnt += 1;
-        if (cnt < period) {
-                return;
-        }
-        cnt = 0;
-
-        /* Go through the points (roughly!):
-         * f(0)=10, f(1)=16, f(2)=20, f(5)=35,f(inf)->110
-         */
-        period = 110 - ( (300<<FSHIFT)/
-                         ((avenrun[0]/5) + (3<<FSHIFT)) );
-
-	if(bit){ bit=0; }else{ bit=1; }
-	outw(bit<<15,0x3fe);
-}
-#endif /* CONFIG_HEARTBEAT */
-
-unsigned long cat68701_isa_port2addr(unsigned long offset)
-{
-	/* CompactFlash (IDE) */
-	if (((offset >= 0x1f0) && (offset <= 0x1f7)) || (offset==0x3f6))
-		return 0xba000000 + offset;
-
-	/* INPUT PORT */
-	if ((offset >= 0x3fc) && (offset <= 0x3fd))
-		return 0xb4007000 + offset;
-
-	/* OUTPUT PORT */
-	if ((offset >= 0x3fe) && (offset <= 0x3ff))
-		return 0xb4007400 + offset;
-
-	return offset + 0xb4000000; /* other I/O (EREA 5)*/
-}
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_cat68701 __initmv = {
-	.mv_nr_irqs		= 32,
-	.mv_isa_port2addr	= cat68701_isa_port2addr,
-	.mv_irq_demux		= cat68701_irq_demux,
-
-	.mv_init_irq		= init_cat68701_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_cat68701,
-#endif
-};
-ALIAS_MV(cat68701)
-
-int __init platform_setup(void)
-{
-	/* dummy read erea5 (CS8900A) */
-}
-
diff --git a/arch/sh/boards/cqreek/Makefile b/arch/sh/boards/cqreek/Makefile
deleted file mode 100644
index 1a788a85eba3..000000000000
--- a/arch/sh/boards/cqreek/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for the CqREEK specific parts of the kernel
-#
-
-obj-y	 := setup.o irq.o
-
diff --git a/arch/sh/boards/cqreek/irq.c b/arch/sh/boards/cqreek/irq.c
deleted file mode 100644
index 2955adc52310..000000000000
--- a/arch/sh/boards/cqreek/irq.c
+++ /dev/null
@@ -1,128 +0,0 @@
-/* $Id: irq.c,v 1.1.2.4 2002/11/04 20:33:56 lethal Exp $
- *
- * arch/sh/boards/cqreek/irq.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *
- * CqREEK IDE/ISA Bridge Support.
- *
- */
-
-#include <linux/irq.h>
-#include <linux/init.h>
-
-#include <asm/cqreek/cqreek.h>
-#include <asm/io.h>
-#include <asm/io_generic.h>
-#include <asm/irq.h>
-#include <asm/machvec.h>
-#include <asm/machvec_init.h>
-#include <asm/rtc.h>
-
-struct cqreek_irq_data {
-	unsigned short mask_port;	/* Port of Interrupt Mask Register */
-	unsigned short stat_port;	/* Port of Interrupt Status Register */
-	unsigned short bit;		/* Value of the bit */
-};
-static struct cqreek_irq_data cqreek_irq_data[NR_IRQS];
-
-static void disable_cqreek_irq(unsigned int irq)
-{
-	unsigned long flags;
-	unsigned short mask;
-	unsigned short mask_port = cqreek_irq_data[irq].mask_port;
-	unsigned short bit = cqreek_irq_data[irq].bit;
-
-	local_irq_save(flags);
-	/* Disable IRQ */
-	mask = inw(mask_port) & ~bit;
-	outw_p(mask, mask_port);
-	local_irq_restore(flags);
-}
-
-static void enable_cqreek_irq(unsigned int irq)
-{
-	unsigned long flags;
-	unsigned short mask;
-	unsigned short mask_port = cqreek_irq_data[irq].mask_port;
-	unsigned short bit = cqreek_irq_data[irq].bit;
-
-	local_irq_save(flags);
-	/* Enable IRQ */
-	mask = inw(mask_port) | bit;
-	outw_p(mask, mask_port);
-	local_irq_restore(flags);
-}
-
-static void mask_and_ack_cqreek(unsigned int irq)
-{
-	unsigned short stat_port = cqreek_irq_data[irq].stat_port;
-	unsigned short bit = cqreek_irq_data[irq].bit;
-
-	disable_cqreek_irq(irq);
-	/* Clear IRQ (it might be edge IRQ) */
-	inw(stat_port);
-	outw_p(bit, stat_port);
-}
-
-static void end_cqreek_irq(unsigned int irq)
-{
-	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
-		enable_cqreek_irq(irq);
-}
-
-static unsigned int startup_cqreek_irq(unsigned int irq)
-{ 
-	enable_cqreek_irq(irq);
-	return 0;
-}
-
-static void shutdown_cqreek_irq(unsigned int irq)
-{
-	disable_cqreek_irq(irq);
-}
-
-static struct hw_interrupt_type cqreek_irq_type = {
-	.typename = "CqREEK-IRQ",
-	.startup = startup_cqreek_irq,
-	.shutdown = shutdown_cqreek_irq,
-	.enable = enable_cqreek_irq,
-	.disable = disable_cqreek_irq,
-	.ack = mask_and_ack_cqreek,
-	.end = end_cqreek_irq
-};
-
-int cqreek_has_ide, cqreek_has_isa;
-
-/* XXX: This is just for test for my NE2000 ISA board
-   What we really need is virtualized IRQ and demultiplexer like HP600 port */
-void __init init_cqreek_IRQ(void)
-{
-	if (cqreek_has_ide) {
-		cqreek_irq_data[14].mask_port = BRIDGE_IDE_INTR_MASK;
-		cqreek_irq_data[14].stat_port = BRIDGE_IDE_INTR_STAT;
-		cqreek_irq_data[14].bit = 1;
-
-		irq_desc[14].chip = &cqreek_irq_type;
-		irq_desc[14].status = IRQ_DISABLED;
-		irq_desc[14].action = 0;
-		irq_desc[14].depth = 1;
-
-		disable_cqreek_irq(14);
-	}
-
-	if (cqreek_has_isa) {
-		cqreek_irq_data[10].mask_port = BRIDGE_ISA_INTR_MASK;
-		cqreek_irq_data[10].stat_port = BRIDGE_ISA_INTR_STAT;
-		cqreek_irq_data[10].bit = (1 << 10);
-
-		/* XXX: Err... we may need demultiplexer for ISA irq... */
-		irq_desc[10].chip = &cqreek_irq_type;
-		irq_desc[10].status = IRQ_DISABLED;
-		irq_desc[10].action = 0;
-		irq_desc[10].depth = 1;
-
-		disable_cqreek_irq(10);
-	}
-}
-
diff --git a/arch/sh/boards/cqreek/setup.c b/arch/sh/boards/cqreek/setup.c
deleted file mode 100644
index eff4ed93599f..000000000000
--- a/arch/sh/boards/cqreek/setup.c
+++ /dev/null
@@ -1,100 +0,0 @@
-/* $Id: setup.c,v 1.5 2003/08/04 01:51:58 lethal Exp $
- *
- * arch/sh/kernel/setup_cqreek.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *
- * CqREEK IDE/ISA Bridge Support.
- *
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/mach/cqreek.h>
-#include <asm/machvec.h>
-#include <asm/io.h>
-#include <asm/io_generic.h>
-#include <asm/irq.h>
-#include <asm/rtc.h>
-
-#define IDE_OFFSET 0xA4000000UL
-#define ISA_OFFSET 0xA4A00000UL
-
-const char *get_system_type(void)
-{
-	return "CqREEK";
-}
-
-static unsigned long cqreek_port2addr(unsigned long port)
-{
-	if (0x0000<=port && port<=0x0040)
-		return IDE_OFFSET + port;
-	if ((0x01f0<=port && port<=0x01f7) || port == 0x03f6)
-		return IDE_OFFSET + port;
-
-	return ISA_OFFSET + port;
-}
-
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_cqreek __initmv = {
-#if defined(CONFIG_CPU_SH4)
-	.mv_nr_irqs		= 48,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7708)
-	.mv_nr_irqs		= 32,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7709)
-	.mv_nr_irqs		= 61,
-#endif
-
-	.mv_init_irq		= init_cqreek_IRQ,
-
-	.mv_isa_port2addr	= cqreek_port2addr,
-};
-ALIAS_MV(cqreek)
-
-/*
- * Initialize the board
- */
-void __init platform_setup(void)
-{
-	int i;
-/* udelay is not available at setup time yet... */
-#define DELAY() do {for (i=0; i<10000; i++) ctrl_inw(0xa0000000);} while(0)
-
-	if ((inw (BRIDGE_FEATURE) & 1)) { /* We have IDE interface */
-		outw_p(0, BRIDGE_IDE_INTR_LVL);
-		outw_p(0, BRIDGE_IDE_INTR_MASK);
-
-		outw_p(0, BRIDGE_IDE_CTRL);
-		DELAY();
-
-		outw_p(0x8000, BRIDGE_IDE_CTRL);
-		DELAY();
-
-		outw_p(0xffff, BRIDGE_IDE_INTR_STAT); /* Clear interrupt status */
-		outw_p(0x0f-14, BRIDGE_IDE_INTR_LVL); /* Use 14 IPR */
-		outw_p(1, BRIDGE_IDE_INTR_MASK); /* Enable interrupt */
-		cqreek_has_ide=1;
-	}
-
-	if ((inw (BRIDGE_FEATURE) & 2)) { /* We have ISA interface */
-		outw_p(0, BRIDGE_ISA_INTR_LVL);
-		outw_p(0, BRIDGE_ISA_INTR_MASK);
-
-		outw_p(0, BRIDGE_ISA_CTRL);
-		DELAY();
-		outw_p(0x8000, BRIDGE_ISA_CTRL);
-		DELAY();
-
-		outw_p(0xffff, BRIDGE_ISA_INTR_STAT); /* Clear interrupt status */
-		outw_p(0x0f-10, BRIDGE_ISA_INTR_LVL); /* Use 10 IPR */
-		outw_p(0xfff8, BRIDGE_ISA_INTR_MASK); /* Enable interrupt */
-		cqreek_has_isa=1;
-	}
-
-	printk(KERN_INFO "CqREEK Setup (IDE=%d, ISA=%d)...done\n", cqreek_has_ide, cqreek_has_isa);
-}
-
diff --git a/arch/sh/boards/dmida/Makefile b/arch/sh/boards/dmida/Makefile
deleted file mode 100644
index 75999aa0a2d9..000000000000
--- a/arch/sh/boards/dmida/Makefile
+++ /dev/null
@@ -1,7 +0,0 @@
-#
-# Makefile for the DataMyte Industrial Digital Assistant(tm) specific parts
-# of the kernel
-#
-
-obj-y	 := mach.o
-
diff --git a/arch/sh/boards/dmida/mach.c b/arch/sh/boards/dmida/mach.c
deleted file mode 100644
index d03a25f989c2..000000000000
--- a/arch/sh/boards/dmida/mach.c
+++ /dev/null
@@ -1,59 +0,0 @@
-/*
- * linux/arch/sh/boards/dmida/mach.c
- *
- * by Greg Banks <gbanks@pocketpenguins.com>
- * (c) 2000 PocketPenguins Inc
- *
- * Derived from mach_hp600.c, which bore the message:
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the DataMyte Industrial Digital Assistant(tm).
- * See http://www.dmida.com
- *
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/io.h>
-#include <asm/hd64465/hd64465.h>
-#include <asm/irq.h>
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_dmida __initmv = {
-	.mv_nr_irqs		= HD64465_IRQ_BASE+HD64465_IRQ_NUM,
-
-	.mv_inb			= hd64465_inb,
-	.mv_inw			= hd64465_inw,
-	.mv_inl			= hd64465_inl,
-	.mv_outb		= hd64465_outb,
-	.mv_outw		= hd64465_outw,
-	.mv_outl		= hd64465_outl,
-
-	.mv_inb_p		= hd64465_inb_p,
-	.mv_inw_p		= hd64465_inw,
-	.mv_inl_p		= hd64465_inl,
-	.mv_outb_p		= hd64465_outb_p,
-	.mv_outw_p		= hd64465_outw,
-	.mv_outl_p		= hd64465_outl,
-
-	.mv_insb		= hd64465_insb,
-	.mv_insw		= hd64465_insw,
-	.mv_insl		= hd64465_insl,
-	.mv_outsb		= hd64465_outsb,
-	.mv_outsw		= hd64465_outsw,
-	.mv_outsl		= hd64465_outsl,
-
-	.mv_irq_demux		= hd64465_irq_demux,
-};
-ALIAS_MV(dmida)
-
diff --git a/arch/sh/boards/harp/Makefile b/arch/sh/boards/harp/Makefile
deleted file mode 100644
index eb753d31812e..000000000000
--- a/arch/sh/boards/harp/Makefile
+++ /dev/null
@@ -1,8 +0,0 @@
-#
-# Makefile for STMicroelectronics board specific parts of the kernel
-#
-
-obj-y := irq.o setup.o mach.o led.o
-
-obj-$(CONFIG_PCI) += pcidma.o
-
diff --git a/arch/sh/boards/harp/irq.c b/arch/sh/boards/harp/irq.c
deleted file mode 100644
index 96bb41c9fc55..000000000000
--- a/arch/sh/boards/harp/irq.c
+++ /dev/null
@@ -1,147 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Looks after interrupts on the HARP board.
- *
- * Bases on the IPR irq system
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/harp/harp.h>
-
-
-#define NUM_EXTERNAL_IRQS 16
-
-// Early versions of the STB1 Overdrive required this nasty frig
-//#define INVERT_INTMASK_WRITES
-
-static void enable_harp_irq(unsigned int irq);
-static void disable_harp_irq(unsigned int irq);
-
-/* shutdown is same as "disable" */
-#define shutdown_harp_irq disable_harp_irq
-
-static void mask_and_ack_harp(unsigned int);
-static void end_harp_irq(unsigned int irq);
-
-static unsigned int startup_harp_irq(unsigned int irq)
-{
-	enable_harp_irq(irq);
-	return 0;		/* never anything pending */
-}
-
-static struct hw_interrupt_type harp_irq_type = {
-	.typename = "Harp-IRQ",
-	.startup = startup_harp_irq,
-	.shutdown = shutdown_harp_irq,
-	.enable = enable_harp_irq,
-	.disable = disable_harp_irq,
-	.ack = mask_and_ack_harp,
-	.end = end_harp_irq
-};
-
-static void disable_harp_irq(unsigned int irq)
-{
-	unsigned val, flags;
-	unsigned maskReg;
-	unsigned mask;
-	int pri;
-
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-	pri = 15 - irq;
-
-	if (pri < 8) {
-		maskReg = EPLD_INTMASK0;
-	} else {
-		maskReg = EPLD_INTMASK1;
-		pri -= 8;
-	}
-
-	local_irq_save(flags);
-	mask = ctrl_inl(maskReg);
-	mask &= (~(1 << pri));
-#if defined(INVERT_INTMASK_WRITES)
-	mask ^= 0xff;
-#endif
-	ctrl_outl(mask, maskReg);
-	local_irq_restore(flags);
-}
-
-static void enable_harp_irq(unsigned int irq)
-{
-	unsigned flags;
-	unsigned maskReg;
-	unsigned mask;
-	int pri;
-
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-	pri = 15 - irq;
-
-	if (pri < 8) {
-		maskReg = EPLD_INTMASK0;
-	} else {
-		maskReg = EPLD_INTMASK1;
-		pri -= 8;
-	}
-
-	local_irq_save(flags);
-	mask = ctrl_inl(maskReg);
-
-
-	mask |= (1 << pri);
-
-#if defined(INVERT_INTMASK_WRITES)
-	mask ^= 0xff;
-#endif
-	ctrl_outl(mask, maskReg);
-
-	local_irq_restore(flags);
-}
-
-/* This functions sets the desired irq handler to be an overdrive type */
-static void __init make_harp_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &harp_irq_type;
-	disable_harp_irq(irq);
-}
-
-static void mask_and_ack_harp(unsigned int irq)
-{
-	disable_harp_irq(irq);
-}
-
-static void end_harp_irq(unsigned int irq)
-{
-	enable_harp_irq(irq);
-}
-
-void __init init_harp_irq(void)
-{
-	int i;
-
-#if !defined(INVERT_INTMASK_WRITES)
-	// On the harp these are set to enable an interrupt
-	ctrl_outl(0x00, EPLD_INTMASK0);
-	ctrl_outl(0x00, EPLD_INTMASK1);
-#else
-	// On the Overdrive the data is inverted before being stored in the reg
-	ctrl_outl(0xff, EPLD_INTMASK0);
-	ctrl_outl(0xff, EPLD_INTMASK1);
-#endif
-
-	for (i = 0; i < NUM_EXTERNAL_IRQS; i++) {
-		make_harp_irq(i);
-	}
-}
diff --git a/arch/sh/boards/harp/led.c b/arch/sh/boards/harp/led.c
deleted file mode 100644
index aeb7b392b190..000000000000
--- a/arch/sh/boards/harp/led.c
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * linux/arch/sh/stboards/led.c
- *
- * Copyright (C) 2000 Stuart Menefy <stuart.menefy@st.com>
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * This file contains ST40STB1 HARP and compatible code.
- */
-
-#include <asm/io.h>
-#include <asm/harp/harp.h>
-
-/* Harp: Flash LD10 (front pannel) connected to EPLD (IC8) */
-/* Overdrive: Flash LD1 (front panel) connected to EPLD (IC4) */
-/* Works for HARP and overdrive */
-static void mach_led(int position, int value)
-{
-	if (value) {
-		ctrl_outl(EPLD_LED_ON, EPLD_LED);
-	} else {
-		ctrl_outl(EPLD_LED_OFF, EPLD_LED);
-	}
-}
-
-#ifdef CONFIG_HEARTBEAT
-
-#include <linux/sched.h>
-
-/* acts like an actual heart beat -- ie thump-thump-pause... */
-void heartbeat_harp(void)
-{
-	static unsigned cnt = 0, period = 0, dist = 0;
-
-	if (cnt == 0 || cnt == dist)
-		mach_led( -1, 1);
-	else if (cnt == 7 || cnt == dist+7)
-		mach_led( -1, 0);
-
-	if (++cnt > period) {
-		cnt = 0;
-		/* The hyperbolic function below modifies the heartbeat period
-		 * length in dependency of the current (5min) load. It goes
-		 * through the points f(0)=126, f(1)=86, f(5)=51,
-		 * f(inf)->30. */
-		period = ((672<<FSHIFT)/(5*avenrun[0]+(7<<FSHIFT))) + 30;
-		dist = period / 4;
-	}
-}
-#endif
diff --git a/arch/sh/boards/harp/mach.c b/arch/sh/boards/harp/mach.c
deleted file mode 100644
index a946dd1674ca..000000000000
--- a/arch/sh/boards/harp/mach.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * linux/arch/sh/boards/harp/mach.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the STMicroelectronics STB1 HARP and compatible boards
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-#include <asm/hd64465/io.h>
-#include <asm/hd64465/hd64465.h>
-
-void setup_harp(void);
-void init_harp_irq(void);
-void heartbeat_harp(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_harp __initmv = {
-	.mv_nr_irqs		= 89 + HD64465_IRQ_NUM,
-
-	.mv_inb			= hd64465_inb,
-	.mv_inw			= hd64465_inw,
-	.mv_inl			= hd64465_inl,
-	.mv_outb		= hd64465_outb,
-	.mv_outw		= hd64465_outw,
-	.mv_outl		= hd64465_outl,
-
-	.mv_inb_p		= hd64465_inb_p,
-	.mv_inw_p		= hd64465_inw,
-	.mv_inl_p		= hd64465_inl,
-	.mv_outb_p		= hd64465_outb_p,
-	.mv_outw_p		= hd64465_outw,
-	.mv_outl_p		= hd64465_outl,
-
-	.mv_insb		= hd64465_insb,
-	.mv_insw		= hd64465_insw,
-	.mv_insl		= hd64465_insl,
-	.mv_outsb		= hd64465_outsb,
-	.mv_outsw		= hd64465_outsw,
-	.mv_outsl		= hd64465_outsl,
-
-        .mv_isa_port2addr       = hd64465_isa_port2addr,
-
-#ifdef CONFIG_PCI
-	.mv_init_irq		= init_harp_irq,
-#endif
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_harp,
-#endif
-};
-
-ALIAS_MV(harp)
diff --git a/arch/sh/boards/harp/pcidma.c b/arch/sh/boards/harp/pcidma.c
deleted file mode 100644
index 475311390fd6..000000000000
--- a/arch/sh/boards/harp/pcidma.c
+++ /dev/null
@@ -1,42 +0,0 @@
-/* 
- * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Dynamic DMA mapping support.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-#include <asm/addrspace.h>
-
-
-void *pci_alloc_consistent(struct pci_dev *hwdev, size_t size,
-			   dma_addr_t * dma_handle)
-{
-	void *ret;
-	int gfp = GFP_ATOMIC;
-
-	ret = (void *) __get_free_pages(gfp, get_order(size));
-
-	if (ret != NULL) {
-	        /* Is it neccessary to do the memset? */
-		memset(ret, 0, size);
-		*dma_handle = virt_to_bus(ret);
-	}
-	/* We must flush the cache before we pass it on to the device */
-	flush_cache_all();
-	return  P2SEGADDR(ret);
-}
-
-void pci_free_consistent(struct pci_dev *hwdev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
-{
-        unsigned long p1addr=P1SEGADDR((unsigned long)vaddr);
-
-	free_pages(p1addr, get_order(size));
-}
diff --git a/arch/sh/boards/harp/setup.c b/arch/sh/boards/harp/setup.c
deleted file mode 100644
index 886e450ab63e..000000000000
--- a/arch/sh/boards/harp/setup.c
+++ /dev/null
@@ -1,90 +0,0 @@
-/*
- * arch/sh/stboard/setup.c
- *
- * Copyright (C) 2001 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * STMicroelectronics ST40STB1 HARP and compatible support.
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <asm/io.h>
-#include <asm/harp/harp.h>
-
-const char *get_system_type(void)
-{
-	return "STB1 Harp";
-}
-
-/*
- * Initialize the board
- */
-int __init platform_setup(void)
-{
-#ifdef CONFIG_SH_STB1_HARP
-	unsigned long ic8_version, ic36_version;
-
-	ic8_version = ctrl_inl(EPLD_REVID2);
-	ic36_version = ctrl_inl(EPLD_REVID1);
-
-        printk("STMicroelectronics STB1 HARP initialisaton\n");
-        printk("EPLD versions: IC8: %d.%02d, IC36: %d.%02d\n",
-               (ic8_version >> 4) & 0xf, ic8_version & 0xf,
-               (ic36_version >> 4) & 0xf, ic36_version & 0xf);
-#elif defined(CONFIG_SH_STB1_OVERDRIVE)
-	unsigned long version;
-
-	version = ctrl_inl(EPLD_REVID);
-
-        printk("STMicroelectronics STB1 Overdrive initialisaton\n");
-        printk("EPLD version: %d.%02d\n",
-	       (version >> 4) & 0xf, version & 0xf);
-#else
-#error Undefined machine
-#endif
- 
-        /* Currently all STB1 chips have problems with the sleep instruction,
-         * so disable it here.
-         */
-	disable_hlt();
-
-	return 0;
-}
-
-/*
- * pcibios_map_platform_irq
- *
- * This is board specific and returns the IRQ for a given PCI device.
- * It is used by the PCI code (arch/sh/kernel/st40_pci*)
- *
- */
-
-#define HARP_PCI_IRQ    1
-#define HARP_BRIDGE_IRQ 2
-#define OVERDRIVE_SLOT0_IRQ 0
-
-
-int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	switch (slot) {
-#ifdef CONFIG_SH_STB1_HARP
-	case 2:		/*This is the PCI slot on the */
-		return HARP_PCI_IRQ;
-	case 1:		/* this is the bridge */
-		return HARP_BRIDGE_IRQ;
-#elif defined(CONFIG_SH_STB1_OVERDRIVE)
-	case 1:
-	case 2:
-	case 3:
-		return slot - 1;
-#else
-#error Unknown board
-#endif
-	default:
-		return -1;
-	}
-}
-
diff --git a/arch/sh/boards/overdrive/Makefile b/arch/sh/boards/overdrive/Makefile
deleted file mode 100644
index 245f03baf762..000000000000
--- a/arch/sh/boards/overdrive/Makefile
+++ /dev/null
@@ -1,8 +0,0 @@
-#
-# Makefile for the STMicroelectronics Overdrive specific parts of the kernel
-#
-
-obj-y	 := mach.o setup.o io.o irq.o led.o
-
-obj-$(CONFIG_PCI) += fpga.o galileo.o pcidma.o
-
diff --git a/arch/sh/boards/overdrive/fpga.c b/arch/sh/boards/overdrive/fpga.c
deleted file mode 100644
index 956c23901228..000000000000
--- a/arch/sh/boards/overdrive/fpga.c
+++ /dev/null
@@ -1,133 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * This file handles programming up the Altera Flex10K that interfaces to
- * the Galileo, and does the PS/2 keyboard and mouse
- *
- */
-
-
-#include <linux/kernel.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/init.h>
-#include <linux/errno.h>
-#include <linux/pci.h>
-#include <linux/delay.h>
-
-
-#include <asm/overdriver/gt64111.h>
-#include <asm/overdrive/overdrive.h>
-#include <asm/overdrive/fpga.h>
-
-#define FPGA_NotConfigHigh()  (*FPGA_ControlReg) = (*FPGA_ControlReg) | ENABLE_FPGA_BIT
-#define FPGA_NotConfigLow()   (*FPGA_ControlReg) = (*FPGA_ControlReg) & RESET_FPGA_MASK
-
-/* I need to find out what (if any) the real delay factor here is */
-/* The delay is definately not critical */
-#define long_delay() {int i;for(i=0;i<10000;i++);}
-#define short_delay() {int i;for(i=0;i<100;i++);}
-
-static void __init program_overdrive_fpga(const unsigned char *fpgacode,
-					  int size)
-{
-	int timeout = 0;
-	int i, j;
-	unsigned char b;
-	static volatile unsigned char *FPGA_ControlReg =
-	    (volatile unsigned char *) (OVERDRIVE_CTRL);
-	static volatile unsigned char *FPGA_ProgramReg =
-	    (volatile unsigned char *) (FPGA_DCLK_ADDRESS);
-
-	printk("FPGA:  Commencing FPGA Programming\n");
-
-	/* The PCI reset but MUST be low when programming the FPGA !!! */
-	b = (*FPGA_ControlReg) & RESET_PCI_MASK;
-
-	(*FPGA_ControlReg) = b;
-
-	/* Prepare FPGA to program */
-
-	FPGA_NotConfigHigh();
-	long_delay();
-
-	FPGA_NotConfigLow();
-	short_delay();
-
-	while ((*FPGA_ProgramReg & FPGA_NOT_STATUS) != 0) {
-		printk("FPGA:  Waiting for NotStatus to go Low ... \n");
-	}
-
-	FPGA_NotConfigHigh();
-
-	/* Wait for FPGA "ready to be programmed" signal */
-	printk("FPGA:  Waiting for NotStatus to go high (FPGA ready)... \n");
-
-	for (timeout = 0;
-	     (((*FPGA_ProgramReg & FPGA_NOT_STATUS) == 0)
-	      && (timeout < FPGA_TIMEOUT)); timeout++);
-
-	/* Check if timeout condition occured - i.e. an error */
-
-	if (timeout == FPGA_TIMEOUT) {
-		printk
-		    ("FPGA:  Failed to program - Timeout waiting for notSTATUS to go high\n");
-		return;
-	}
-
-	printk("FPGA:  Copying data to FPGA ... %d bytes\n", size);
-
-	/* Copy array to FPGA - bit at a time */
-
-	for (i = 0; i < size; i++) {
-		volatile unsigned w = 0;
-
-		for (j = 0; j < 8; j++) {
-			*FPGA_ProgramReg = (fpgacode[i] >> j) & 0x01;
-			short_delay();
-		}
-		if ((i & 0x3ff) == 0) {
-			printk(".");
-		}
-	}
-
-	/* Waiting for CONFDONE to go high - means the program is complete  */
-
-	for (timeout = 0;
-	     (((*FPGA_ProgramReg & FPGA_CONFDONE) == 0)
-	      && (timeout < FPGA_TIMEOUT)); timeout++) {
-
-		*FPGA_ProgramReg = 0x0;
-		long_delay();
-	}
-
-	if (timeout == FPGA_TIMEOUT) {
-		printk
-		    ("FPGA:  Failed to program - Timeout waiting for CONFDONE to go high\n");
-		return;
-	} else {		/* Clock another 10 times - gets the device into a working state      */
-		for (i = 0; i < 10; i++) {
-			*FPGA_ProgramReg = 0x0;
-			short_delay();
-		}
-	}
-
-	printk("FPGA:  Programming complete\n");
-}
-
-
-static const unsigned char __init fpgacode[] = {
-#include "./overdrive.ttf"	/* Code from maxplus2 compiler */
-	, 0, 0
-};
-
-
-int __init init_overdrive_fpga(void)
-{
-	program_overdrive_fpga(fpgacode, sizeof(fpgacode));
-
-	return 0;
-}
diff --git a/arch/sh/boards/overdrive/galileo.c b/arch/sh/boards/overdrive/galileo.c
deleted file mode 100644
index 29e48971bba0..000000000000
--- a/arch/sh/boards/overdrive/galileo.c
+++ /dev/null
@@ -1,587 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * This file contains the PCI routines required for the Galileo GT6411 
- * PCI bridge as used on the Orion and Overdrive boards.
- *
- */
-
-#include <linux/kernel.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/init.h>
-#include <linux/errno.h>
-#include <linux/pci.h>
-#include <linux/delay.h>
-#include <linux/types.h>
-#include <linux/ioport.h>
-
-#include <asm/overdrive/overdrive.h>
-#include <asm/overdrive/gt64111.h>
-
-
-/* After boot, we shift the Galileo registers so that they appear 
- * in BANK6, along with IO space. This means we can have one contingous
- * lump of PCI address space without these registers appearing in the 
- * middle of them 
- */
-
-#define GT64111_BASE_ADDRESS  0xbb000000
-#define GT64111_IO_BASE_ADDRESS  0x1000
-/* The GT64111 registers appear at this address to the SH4 after reset */
-#define RESET_GT64111_BASE_ADDRESS           0xb4000000
-
-/* Macros used to access the Galileo registers */
-#define RESET_GT64111_REG(x) (RESET_GT64111_BASE_ADDRESS+x)
-#define GT64111_REG(x) (GT64111_BASE_ADDRESS+x)
-
-#define RESET_GT_WRITE(x,v) writel((v),RESET_GT64111_REG(x))
-
-#define RESET_GT_READ(x) readl(RESET_GT64111_REG(x))
-
-#define GT_WRITE(x,v) writel((v),GT64111_REG(x))
-#define GT_WRITE_BYTE(x,v) writeb((v),GT64111_REG(x))
-#define GT_WRITE_SHORT(x,v) writew((v),GT64111_REG(x))
-
-#define GT_READ(x)    readl(GT64111_REG(x))
-#define GT_READ_BYTE(x)  readb(GT64111_REG(x))
-#define GT_READ_SHORT(x) readw(GT64111_REG(x))
-
-
-/* Where the various SH banks start at */
-#define SH_BANK4_ADR 0xb0000000
-#define SH_BANK5_ADR 0xb4000000
-#define SH_BANK6_ADR 0xb8000000
-
-/* Masks out everything but lines 28,27,26 */
-#define BANK_SELECT_MASK 0x1c000000
-
-#define SH4_TO_BANK(x) ( (x) & BANK_SELECT_MASK)
-
-/* 
- * Masks used for address conversaion. Bank 6 is used for IO and 
- * has all the address bits zeroed by the FPGA. Special case this
- */
-#define MEMORY_BANK_MASK 0x1fffffff
-#define IO_BANK_MASK  0x03ffffff
-
-/* Mark bank 6 as the bank used for IO. You can change this in the FPGA code
- * if you want 
- */
-#define IO_BANK_ADR PCI_GTIO_BASE
-
-/* Will select the correct mask to apply depending on the SH$ address */
-#define SELECT_BANK_MASK(x) \
-   ( (SH4_TO_BANK(x)==SH4_TO_BANK(IO_BANK_ADR)) ? IO_BANK_MASK : MEMORY_BANK_MASK)
-
-/* Converts between PCI space and P2 region */
-#define SH4_TO_PCI(x) ((x)&SELECT_BANK_MASK(x))
-
-/* Various macros for figuring out what to stick in the Galileo registers. 
- * You *really* don't want to figure this stuff out by hand, you always get
- * it wrong
- */
-#define GT_MEM_LO_ADR(x) ((((unsigned)((x)&SELECT_BANK_MASK(x)))>>21)&0x7ff)
-#define GT_MEM_HI_ADR(x) ((((unsigned)((x)&SELECT_BANK_MASK(x)))>>21)&0x7f)
-#define GT_MEM_SUB_ADR(x) ((((unsigned)((x)&SELECT_BANK_MASK(x)))>>20)&0xff)
-
-#define PROGRAM_HI_LO(block,a,s) \
-    GT_WRITE(block##_LO_DEC_ADR,GT_MEM_LO_ADR(a));\
-    GT_WRITE(block##_HI_DEC_ADR,GT_MEM_HI_ADR(a+s-1))
-
-#define PROGRAM_SUB_HI_LO(block,a,s) \
-    GT_WRITE(block##_LO_DEC_ADR,GT_MEM_SUB_ADR(a));\
-    GT_WRITE(block##_HI_DEC_ADR,GT_MEM_SUB_ADR(a+s-1))
-
-/* We need to set the size, and the offset register */
-
-#define GT_BAR_MASK(x) ((x)&~0xfff)
-
-/* Macro to set up the BAR in the Galileo. Essentially used for the DRAM */
-#define PROGRAM_GT_BAR(block,a,s) \
-  GT_WRITE(PCI_##block##_BANK_SIZE,GT_BAR_MASK((s-1)));\
-  write_config_to_galileo(PCI_CONFIG_##block##_BASE_ADR,\
-			     GT_BAR_MASK(a))
-
-#define DISABLE_GT_BAR(block) \
-  GT_WRITE(PCI_##block##_BANK_SIZE,0),\
-  GT_CONFIG_WRITE(PCI_CONFIG_##block##_BASE_ADR,\
-    0x80000000)
-
-/* Macros to disable things we are not going to use */
-#define DISABLE_DECODE(x) GT_WRITE(x##_LO_DEC_ADR,0x7ff);\
-                          GT_WRITE(x##_HI_DEC_ADR,0x00)
-
-#define DISABLE_SUB_DECODE(x) GT_WRITE(x##_LO_DEC_ADR,0xff);\
-                              GT_WRITE(x##_HI_DEC_ADR,0x00)
-
-static void __init reset_pci(void)
-{
-	/* Set RESET_PCI bit high */
-	writeb(readb(OVERDRIVE_CTRL) | ENABLE_PCI_BIT, OVERDRIVE_CTRL);
-	udelay(250);
-
-	/* Set RESET_PCI bit low */
-	writeb(readb(OVERDRIVE_CTRL) & RESET_PCI_MASK, OVERDRIVE_CTRL);
-	udelay(250);
-
-	writeb(readb(OVERDRIVE_CTRL) | ENABLE_PCI_BIT, OVERDRIVE_CTRL);
-	udelay(250);
-}
-
-static int write_config_to_galileo(int where, u32 val);
-#define GT_CONFIG_WRITE(where,val) write_config_to_galileo(where,val)
-
-#define ENABLE_PCI_DRAM
-
-
-#ifdef TEST_DRAM
-/* Test function to check out if the PCI DRAM is working OK */
-static int  /* __init */ test_dram(unsigned *base, unsigned size)
-{
-	unsigned *p = base;
-	unsigned *end = (unsigned *) (((unsigned) base) + size);
-	unsigned w;
-
-	for (p = base; p < end; p++) {
-		*p = 0xffffffff;
-		if (*p != 0xffffffff) {
-			printk("AAARGH -write failed!!! at %p is %x\n", p,
-			       *p);
-			return 0;
-		}
-		*p = 0x0;
-		if (*p != 0x0) {
-			printk("AAARGH -write failed!!!\n");
-			return 0;
-		}
-	}
-
-	for (p = base; p < end; p++) {
-		*p = (unsigned) p;
-		if (*p != (unsigned) p) {
-			printk("Failed at 0x%p, actually is 0x%x\n", p,
-			       *p);
-			return 0;
-		}
-	}
-
-	for (p = base; p < end; p++) {
-		w = ((unsigned) p & 0xffff0000);
-		*p = w | (w >> 16);
-	}
-
-	for (p = base; p < end; p++) {
-		w = ((unsigned) p & 0xffff0000);
-		w |= (w >> 16);
-		if (*p != w) {
-			printk
-			    ("Failed at 0x%p, should be 0x%x actually is 0x%x\n",
-			     p, w, *p);
-			return 0;
-		}
-	}
-
-	return 1;
-}
-#endif
-
-
-/* Function to set up and initialise the galileo. This sets up the BARS,
- * maps the DRAM into the address space etc,etc
- */
-int __init galileo_init(void)
-{
-	reset_pci();
-
-	/* Now shift the galileo regs into this block */
-	RESET_GT_WRITE(INTERNAL_SPACE_DEC,
-		       GT_MEM_LO_ADR(GT64111_BASE_ADDRESS));
-
-	/* Should have a sanity check here, that you can read back  at the new
-	 * address what you just wrote 
-	 */
-
-	/* Disable decode for all regions */
-	DISABLE_DECODE(RAS10);
-	DISABLE_DECODE(RAS32);
-	DISABLE_DECODE(CS20);
-	DISABLE_DECODE(CS3);
-	DISABLE_DECODE(PCI_IO);
-	DISABLE_DECODE(PCI_MEM0);
-	DISABLE_DECODE(PCI_MEM1);
-
-	/* Disable all BARS */
-	GT_WRITE(BAR_ENABLE_ADR, 0x1ff);
-	DISABLE_GT_BAR(RAS10);
-	DISABLE_GT_BAR(RAS32);
-	DISABLE_GT_BAR(CS20);
-	DISABLE_GT_BAR(CS3);
-
-	/* Tell the BAR where the IO registers now are */
-	GT_CONFIG_WRITE(PCI_CONFIG_INT_REG_IO_ADR,GT_BAR_MASK(
-					    (GT64111_IO_BASE_ADDRESS &
-					     IO_BANK_MASK)));
-	/* set up a 112 Mb decode */
-	PROGRAM_HI_LO(PCI_MEM0, SH_BANK4_ADR, 112 * 1024 * 1024);
-
-	/* Set up a 32 MB io space decode */
-	PROGRAM_HI_LO(PCI_IO, IO_BANK_ADR, 32 * 1024 * 1024);
-
-#ifdef ENABLE_PCI_DRAM
-	/* Program up the DRAM configuration - there is DRAM only in bank 0 */
-	/* Now set up the DRAM decode */
-	PROGRAM_HI_LO(RAS10, PCI_DRAM_BASE, PCI_DRAM_SIZE);
-	/* And the sub decode */
-	PROGRAM_SUB_HI_LO(RAS0, PCI_DRAM_BASE, PCI_DRAM_SIZE);
-
-	DISABLE_SUB_DECODE(RAS1);
-
-	/* Set refresh rate */
-	GT_WRITE(DRAM_BANK0_PARMS, 0x3f);
-	GT_WRITE(DRAM_CFG, 0x100);
-
-	/* we have to lob off the top bits rememeber!! */
-	PROGRAM_GT_BAR(RAS10, SH4_TO_PCI(PCI_DRAM_BASE), PCI_DRAM_SIZE);
-
-#endif
-
-	/* We are only interested in decoding RAS10 and the Galileo's internal 
-	 * registers (as IO) on the PCI bus
-	 */
-#ifdef ENABLE_PCI_DRAM
-	GT_WRITE(BAR_ENABLE_ADR, (~((1 << 8) | (1 << 3))) & 0x1ff);
-#else
-	GT_WRITE(BAR_ENABLE_ADR, (~(1 << 3)) & 0x1ff);
-#endif
-
-	/* Change the class code to host bridge, it actually powers up 
-	 * as a memory controller
-         */
-	GT_CONFIG_WRITE(8, 0x06000011);
-
-	/* Allow the galileo to master the PCI bus */
-	GT_CONFIG_WRITE(PCI_COMMAND,
-			PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER |
-			PCI_COMMAND_IO);
-
-
-#if 0
-        printk("Testing PCI DRAM - ");
-	if(test_dram(PCI_DRAM_BASE,PCI_DRAM_SIZE)) {
-		printk("Passed\n");
-	}else {
-		printk("FAILED\n");
-	}
-#endif
-	return 0;
-
-}
-
-
-#define SET_CONFIG_BITS(bus,devfn,where)\
-  ((1<<31) | ((bus) << 16) | ((devfn) << 8) | ((where) & ~3))
-
-#define CONFIG_CMD(dev, where) SET_CONFIG_BITS((dev)->bus->number,(dev)->devfn,where)
-
-/* This write to the galileo config registers, unlike the functions below, can
- * be used before the PCI subsystem has started up
- */
-static int __init write_config_to_galileo(int where, u32 val)
-{
-	GT_WRITE(PCI_CFG_ADR, SET_CONFIG_BITS(0, 0, where));
-
-	GT_WRITE(PCI_CFG_DATA, val);
-	return 0;
-}
-
-/* We exclude the galileo and slot 31, the galileo because I don't know how to stop
- * the setup code shagging up the setup I have done on it, and 31 because the whole
- * thing locks up if you try to access that slot (which doesn't exist of course anyway
- */
-
-#define EXCLUDED_DEV(dev) ((dev->bus->number==0) && ((PCI_SLOT(dev->devfn)==0) || (PCI_SLOT(dev->devfn) == 31)))
-
-static int galileo_read_config_byte(struct pci_dev *dev, int where,
-				    u8 * val)
-{
-
-        
-	/* I suspect this doesn't work because this drives a special cycle ? */
-	if (EXCLUDED_DEV(dev)) {
-		*val = 0xff;
-		return PCIBIOS_SUCCESSFUL;
-	}
-	/* Start the config cycle */
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-	/* Read back the result */
-	*val = GT_READ_BYTE(PCI_CFG_DATA + (where & 3));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-
-static int galileo_read_config_word(struct pci_dev *dev, int where,
-				    u16 * val)
-{
-
-        if (EXCLUDED_DEV(dev)) {
-		*val = 0xffff;
-		return PCIBIOS_SUCCESSFUL;
-	}
-
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-	*val = GT_READ_SHORT(PCI_CFG_DATA + (where & 2));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-
-static int galileo_read_config_dword(struct pci_dev *dev, int where,
-				     u32 * val)
-{
-	if (EXCLUDED_DEV(dev)) {
-		*val = 0xffffffff;
-		return PCIBIOS_SUCCESSFUL;
-	}
-
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-	*val = GT_READ(PCI_CFG_DATA);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int galileo_write_config_byte(struct pci_dev *dev, int where,
-				     u8 val)
-{
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-
-	GT_WRITE_BYTE(PCI_CFG_DATA + (where & 3), val);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-
-static int galileo_write_config_word(struct pci_dev *dev, int where,
-				     u16 val)
-{
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-
-	GT_WRITE_SHORT(PCI_CFG_DATA + (where & 2), val);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int galileo_write_config_dword(struct pci_dev *dev, int where,
-				      u32 val)
-{
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-
-	GT_WRITE(PCI_CFG_DATA, val);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops pci_config_ops = {
-	galileo_read_config_byte,
-	galileo_read_config_word,
-	galileo_read_config_dword,
-	galileo_write_config_byte,
-	galileo_write_config_word,
-	galileo_write_config_dword
-};
-
-
-/* Everything hangs off this */
-static struct pci_bus *pci_root_bus;
-
-
-static u8 __init no_swizzle(struct pci_dev *dev, u8 * pin)
-{
-	return PCI_SLOT(dev->devfn);
-}
-
-static int __init map_od_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	/* Slot 1: Galileo 
-	 * Slot 2: PCI Slot 1
-	 * Slot 3: PCI Slot 2
-	 * Slot 4: ESS
-	 */
-	switch (slot) {
-	case 2: 
-		return OVERDRIVE_PCI_IRQ1;
-	case 3:
-		/* Note this assumes you have a hacked card in slot 2 */
-		return OVERDRIVE_PCI_IRQ2;
-	case 4:
-		return OVERDRIVE_ESS_IRQ;
-	default:
-		/* printk("PCI: Unexpected IRQ mapping request for slot %d\n", slot); */
-		return -1;
-	}
-}
-
-
-
-void __init
-pcibios_fixup_pbus_ranges(struct pci_bus *bus, struct pbus_set_ranges_data *ranges)
-{
-        ranges->io_start -= bus->resource[0]->start;
-        ranges->io_end -= bus->resource[0]->start;
-        ranges->mem_start -= bus->resource[1]->start;
-        ranges->mem_end -= bus->resource[1]->start;
-}                                                                                
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	printk("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
-
-void __init pcibios_init(void)
-{
-	static struct resource galio,galmem;
-
-        /* Allocate the registers used by the Galileo */
-        galio.flags = IORESOURCE_IO;
-        galio.name  = "Galileo GT64011";
-        galmem.flags = IORESOURCE_MEM|IORESOURCE_PREFETCH;
-        galmem.name  = "Galileo GT64011 DRAM";
-
-        allocate_resource(&ioport_resource, &galio, 256,
-		    GT64111_IO_BASE_ADDRESS,GT64111_IO_BASE_ADDRESS+256, 256, NULL, NULL);
-        allocate_resource(&iomem_resource, &galmem,PCI_DRAM_SIZE,
-		    PHYSADDR(PCI_DRAM_BASE), PHYSADDR(PCI_DRAM_BASE)+PCI_DRAM_SIZE, 
-			     PCI_DRAM_SIZE, NULL, NULL);
-
-  	/* ok, do the scan man */
-	pci_root_bus = pci_scan_bus(0, &pci_config_ops, NULL);
-
-        pci_assign_unassigned_resources();
-	pci_fixup_irqs(no_swizzle, map_od_irq);
-
-#ifdef TEST_DRAM
-        printk("Testing PCI DRAM - ");
-	if(test_dram(PCI_DRAM_BASE,PCI_DRAM_SIZE)) {
-		printk("Passed\n");
-	}else {
-		printk("FAILED\n");
-	}
-#endif
-
-}
-
-char * __init pcibios_setup(char *str)
-{
-	return str;
-}
-
-
-
-int pcibios_enable_device(struct pci_dev *dev)
-{
-
-	u16 cmd, old_cmd;
-	int idx;
-	struct resource *r;
-
-	pci_read_config_word(dev, PCI_COMMAND, &cmd);
-	old_cmd = cmd;
-	for (idx = 0; idx < 6; idx++) {
-		r = dev->resource + idx;
-		if (!r->start && r->end) {
-			printk(KERN_ERR
-			       "PCI: Device %s not available because"
-			       " of resource collisions\n",
-			       pci_name(dev));
-			return -EINVAL;
-		}
-		if (r->flags & IORESOURCE_IO)
-			cmd |= PCI_COMMAND_IO;
-		if (r->flags & IORESOURCE_MEM)
-			cmd |= PCI_COMMAND_MEMORY;
-	}
-	if (cmd != old_cmd) {
-		printk("PCI: enabling device %s (%04x -> %04x)\n",
-		       pci_name(dev), old_cmd, cmd);
-		pci_write_config_word(dev, PCI_COMMAND, cmd);
-	}
-	return 0;
-
-}
-
-/* We should do some optimisation work here I think. Ok for now though */
-void __init pcibios_fixup_bus(struct pci_bus *bus)
-{
-
-}
-
-void pcibios_align_resource(void *data, struct resource *res,
-			    resource_size_t size)
-{
-}
-
-void __init pcibios_update_resource(struct pci_dev *dev, struct resource *root,
-			     struct resource *res, int resource)
-{
-
-	unsigned long where, size;
-	u32 reg;
-	
-
-	printk("PCI: Assigning %3s %08lx to %s\n",
-	       res->flags & IORESOURCE_IO ? "IO" : "MEM",
-	       res->start, dev->name);
-
-	where = PCI_BASE_ADDRESS_0 + resource * 4;
-	size = res->end - res->start;
-
-	pci_read_config_dword(dev, where, &reg);
-	reg = (reg & size) | (((u32) (res->start - root->start)) & ~size);
-	pci_write_config_dword(dev, where, reg);
-}
-
-
-void __init pcibios_update_irq(struct pci_dev *dev, int irq)
-{
-	printk("PCI: Assigning IRQ %02d to %s\n", irq, dev->name);
-	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
-}
-
-/*
- *  If we set up a device for bus mastering, we need to check the latency
- *  timer as certain crappy BIOSes forget to set it properly.
- */
-unsigned int pcibios_max_latency = 255;
-
-void pcibios_set_master(struct pci_dev *dev)
-{
-	u8 lat;
-	pci_read_config_byte(dev, PCI_LATENCY_TIMER, &lat);
-	if (lat < 16)
-		lat = (64 <= pcibios_max_latency) ? 64 : pcibios_max_latency;
-	else if (lat > pcibios_max_latency)
-		lat = pcibios_max_latency;
-	else
-		return;
-	printk("PCI: Setting latency timer of device %s to %d\n", pci_name(dev), lat);
-	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
-}
diff --git a/arch/sh/boards/overdrive/io.c b/arch/sh/boards/overdrive/io.c
deleted file mode 100644
index 4671b6b047bb..000000000000
--- a/arch/sh/boards/overdrive/io.c
+++ /dev/null
@@ -1,172 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * This file contains the I/O routines for use on the overdrive board
- *
- */
-
-#include <linux/types.h>
-#include <linux/delay.h>
-#include <asm/processor.h>
-#include <asm/io.h>
-#include <asm/addrspace.h>
-
-#include <asm/overdrive/overdrive.h>
-
-/*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the SuperH architecture, we just read/write the
- * memory location directly.
- */
-
-#define dprintk(x...)
-
-/* Translates an IO address to where it is mapped in memory */
-
-#define io_addr(x) (((unsigned)(x))|PCI_GTIO_BASE)
-
-unsigned char od_inb(unsigned long port)
-{
-dprintk("od_inb(%x)\n", port);
-	return readb(io_addr(port)) & 0xff;
-}
-
-
-unsigned short od_inw(unsigned long port)
-{
-dprintk("od_inw(%x)\n", port);
-	return readw(io_addr(port)) & 0xffff;
-}
-
-unsigned int od_inl(unsigned long port)
-{
-dprintk("od_inl(%x)\n", port);
-	return readl(io_addr(port));
-}
-
-void od_outb(unsigned char value, unsigned long port)
-{
-dprintk("od_outb(%x, %x)\n", value, port);
-	writeb(value, io_addr(port));
-}
-
-void od_outw(unsigned short value, unsigned long port)
-{
-dprintk("od_outw(%x, %x)\n", value, port);
-	writew(value, io_addr(port));
-}
-
-void od_outl(unsigned int value, unsigned long port)
-{
-dprintk("od_outl(%x, %x)\n", value, port);
-	writel(value, io_addr(port));
-}
-
-/* This is horrible at the moment - needs more work to do something sensible */
-#define IO_DELAY() udelay(10)
-
-#define OUT_DELAY(x,type) \
-void od_out##x##_p(unsigned type value,unsigned long port){out##x(value,port);IO_DELAY();}
-
-#define IN_DELAY(x,type) \
-unsigned type od_in##x##_p(unsigned long port) {unsigned type tmp=in##x(port);IO_DELAY();return tmp;}
-
-
-OUT_DELAY(b,char)
-OUT_DELAY(w,short)
-OUT_DELAY(l,int)
-
-IN_DELAY(b,char)
-IN_DELAY(w,short)
-IN_DELAY(l,int)
-
-
-/*  Now for the string version of these functions */
-void od_outsb(unsigned long port, const void *addr, unsigned long count)
-{
-	int i;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p++) {
-		outb(*p, port);
-	}
-}
-
-
-void od_insb(unsigned long port, void *addr, unsigned long count)
-{
-	int i;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p++) {
-		*p = inb(port);
-	}
-}
-
-/* For the 16 and 32 bit string functions, we have to worry about alignment.
- * The SH does not do unaligned accesses, so we have to read as bytes and
- * then write as a word or dword. 
- * This can be optimised a lot more, especially in the case where the data
- * is aligned
- */
-
-void od_outsw(unsigned long port, const void *addr, unsigned long count)
-{
-	int i;
-	unsigned short tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 2) {
-		tmp = (*p) | ((*(p + 1)) << 8);
-		outw(tmp, port);
-	}
-}
-
-
-void od_insw(unsigned long port, void *addr, unsigned long count)
-{
-	int i;
-	unsigned short tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 2) {
-		tmp = inw(port);
-		p[0] = tmp & 0xff;
-		p[1] = (tmp >> 8) & 0xff;
-	}
-}
-
-
-void od_outsl(unsigned long port, const void *addr, unsigned long count)
-{
-	int i;
-	unsigned tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 4) {
-		tmp = (*p) | ((*(p + 1)) << 8) | ((*(p + 2)) << 16) |
-		      ((*(p + 3)) << 24);
-		outl(tmp, port);
-	}
-}
-
-
-void od_insl(unsigned long port, void *addr, unsigned long count)
-{
-	int i;
-	unsigned tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 4) {
-		tmp = inl(port);
-		p[0] = tmp & 0xff;
-		p[1] = (tmp >> 8) & 0xff;
-		p[2] = (tmp >> 16) & 0xff;
-		p[3] = (tmp >> 24) & 0xff;
-
-	}
-}
diff --git a/arch/sh/boards/overdrive/irq.c b/arch/sh/boards/overdrive/irq.c
deleted file mode 100644
index 5d730c70389e..000000000000
--- a/arch/sh/boards/overdrive/irq.c
+++ /dev/null
@@ -1,191 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Looks after interrupts on the overdrive board.
- *
- * Bases on the IPR irq system
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-
-#include <asm/overdrive/overdrive.h>
-
-struct od_data {
-	int overdrive_irq;
-	int irq_mask;
-};
-
-#define NUM_EXTERNAL_IRQS 16
-#define EXTERNAL_IRQ_NOT_IN_USE (-1)
-#define EXTERNAL_IRQ_NOT_ASSIGNED (-1)
-
-/*
- * This table is used to determine what to program into the FPGA's CT register
- * for the specified Linux IRQ.
- *
- * The irq_mask gives the interrupt number from the PCI board (PCI_Int(6:0))
- * but is one greater than that because the because the FPGA treats 0
- * as disabled, a value of 1 asserts PCI_Int0, and so on.
- *
- * The overdrive_irq specifies which of the eight interrupt sources generates
- * that interrupt, and but is multiplied by four to give the bit offset into
- * the CT register.
- *
- * The seven interrupts levels (SH4 IRL's) we have available here is hardwired
- * by the EPLD. The assignments here of which PCI interrupt generates each
- * level is arbitary.
- */
-static struct od_data od_data_table[NUM_EXTERNAL_IRQS] = {
-	/*    overdrive_irq       , irq_mask */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 0 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 7},	/* 1 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 6},	/* 2 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 3 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 5},	/* 4 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 5 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 6 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 4},	/* 7 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 8 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 9 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 3},	/* 10 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 2},	/* 11 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 12 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 1},	/* 13 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 14 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE}	/* 15 */
-};
-
-static void set_od_data(int overdrive_irq, int irq)
-{
-	if (irq >= NUM_EXTERNAL_IRQS || irq < 0)
-		return;
-	od_data_table[irq].overdrive_irq = overdrive_irq << 2;
-}
-
-static void enable_od_irq(unsigned int irq);
-void disable_od_irq(unsigned int irq);
-
-/* shutdown is same as "disable" */
-#define shutdown_od_irq disable_od_irq
-
-static void mask_and_ack_od(unsigned int);
-static void end_od_irq(unsigned int irq);
-
-static unsigned int startup_od_irq(unsigned int irq)
-{
-	enable_od_irq(irq);
-	return 0;		/* never anything pending */
-}
-
-static struct hw_interrupt_type od_irq_type = {
-	.typename = "Overdrive-IRQ",
-	.startup = startup_od_irq,
-	.shutdown = shutdown_od_irq,
-	.enable = enable_od_irq,
-	.disable = disable_od_irq,
-	.ack = mask_and_ack_od,
-	.end = end_od_irq
-};
-
-static void disable_od_irq(unsigned int irq)
-{
-	unsigned val, flags;
-	int overdrive_irq;
-	unsigned mask;
-
-	/* Not a valid interrupt */
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-        /* Is is necessary to use a cli here? Would a spinlock not be 
-         * mroe efficient?
-         */
-	local_irq_save(flags);
-	overdrive_irq = od_data_table[irq].overdrive_irq;
-	if (overdrive_irq != EXTERNAL_IRQ_NOT_ASSIGNED) {
-		mask = ~(0x7 << overdrive_irq);
-		val = ctrl_inl(OVERDRIVE_INT_CT);
-		val &= mask;
-		ctrl_outl(val, OVERDRIVE_INT_CT);
-	}
-	local_irq_restore(flags);
-}
-
-static void enable_od_irq(unsigned int irq)
-{
-	unsigned val, flags;
-	int overdrive_irq;
-	unsigned mask;
-
-	/* Not a valid interrupt */
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-	/* Set priority in OD back to original value */
-	local_irq_save(flags);
-	/* This one is not in use currently */
-	overdrive_irq = od_data_table[irq].overdrive_irq;
-	if (overdrive_irq != EXTERNAL_IRQ_NOT_ASSIGNED) {
-		val = ctrl_inl(OVERDRIVE_INT_CT);
-		mask = ~(0x7 << overdrive_irq);
-		val &= mask;
-		mask = od_data_table[irq].irq_mask << overdrive_irq;
-		val |= mask;
-		ctrl_outl(val, OVERDRIVE_INT_CT);
-	}
-	local_irq_restore(flags);
-}
-
-
-
-/* this functions sets the desired irq handler to be an overdrive type */
-static void __init make_od_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &od_irq_type;
-	disable_od_irq(irq);
-}
-
-
-static void mask_and_ack_od(unsigned int irq)
-{
-	disable_od_irq(irq);
-}
-
-static void end_od_irq(unsigned int irq)
-{
-	enable_od_irq(irq);
-}
-
-void __init init_overdrive_irq(void)
-{
-	int i;
-
-	/* Disable all interrupts */
-	ctrl_outl(0, OVERDRIVE_INT_CT);
-
-	/* Update interrupt pin mode to use encoded interrupts */
-	i = ctrl_inw(INTC_ICR);
-	i &= ~INTC_ICR_IRLM;
-	ctrl_outw(i, INTC_ICR);
-
-	for (i = 0; i < NUM_EXTERNAL_IRQS; i++) {
-		if (od_data_table[i].irq_mask != EXTERNAL_IRQ_NOT_IN_USE) {
-			make_od_irq(i);
-		} else if (i != 15) {	// Cannot use imask on level 15
-			make_imask_irq(i);
-		}
-	}
-
-	/* Set up the interrupts */
-	set_od_data(OVERDRIVE_PCI_INTA, OVERDRIVE_PCI_IRQ1);
-	set_od_data(OVERDRIVE_PCI_INTB, OVERDRIVE_PCI_IRQ2);
-	set_od_data(OVERDRIVE_AUDIO_INT, OVERDRIVE_ESS_IRQ);
-}
diff --git a/arch/sh/boards/overdrive/led.c b/arch/sh/boards/overdrive/led.c
deleted file mode 100644
index 860d7f204a4e..000000000000
--- a/arch/sh/boards/overdrive/led.c
+++ /dev/null
@@ -1,58 +0,0 @@
-/*
- * linux/arch/sh/overdrive/led.c
- *
- * Copyright (C) 1999 Stuart Menefy <stuart.menefy@st.com>
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * This file contains an Overdrive specific LED feature.
- */
-
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/overdrive/overdrive.h>
-
-static void mach_led(int position, int value)
-{
-	unsigned long flags;
-	unsigned long reg;
-
-	local_irq_save(flags);
-	
-	reg = readl(OVERDRIVE_CTRL);
-	if (value) {
-		reg |= (1<<3);
-	} else {
-		reg &= ~(1<<3);
-	}
-	writel(reg, OVERDRIVE_CTRL);
-
-	local_irq_restore(flags);
-}
-
-#ifdef CONFIG_HEARTBEAT
-
-#include <linux/sched.h>
-
-/* acts like an actual heart beat -- ie thump-thump-pause... */
-void heartbeat_od(void)
-{
-	static unsigned cnt = 0, period = 0, dist = 0;
-
-	if (cnt == 0 || cnt == dist)
-		mach_led( -1, 1);
-	else if (cnt == 7 || cnt == dist+7)
-		mach_led( -1, 0);
-
-	if (++cnt > period) {
-		cnt = 0;
-		/* The hyperbolic function below modifies the heartbeat period
-		 * length in dependency of the current (5min) load. It goes
-		 * through the points f(0)=126, f(1)=86, f(5)=51,
-		 * f(inf)->30. */
-		period = ((672<<FSHIFT)/(5*avenrun[0]+(7<<FSHIFT))) + 30;
-		dist = period / 4;
-	}
-}
-#endif /* CONFIG_HEARTBEAT */
diff --git a/arch/sh/boards/overdrive/mach.c b/arch/sh/boards/overdrive/mach.c
deleted file mode 100644
index 2834a03ae477..000000000000
--- a/arch/sh/boards/overdrive/mach.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * linux/arch/sh/overdrive/mach.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the STMicroelectronics Overdrive
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/io_unknown.h>
-#include <asm/io_generic.h>
-#include <asm/overdrive/io.h>
-
-void heartbeat_od(void);
-void init_overdrive_irq(void);
-void galileo_pcibios_init(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_od __initmv = {
-	.mv_nr_irqs		= 48,
-
-	.mv_inb			= od_inb,
-	.mv_inw			= od_inw,
-	.mv_inl			= od_inl,
-	.mv_outb		= od_outb,
-	.mv_outw		= od_outw,
-	.mv_outl		= od_outl,
-
-	.mv_inb_p		= od_inb_p,
-	.mv_inw_p		= od_inw_p,
-	.mv_inl_p		= od_inl_p,
-	.mv_outb_p		= od_outb_p,
-	.mv_outw_p		= od_outw_p,
-	.mv_outl_p		= od_outl_p,
-
-	.mv_insb		= od_insb,
-	.mv_insw		= od_insw,
-	.mv_insl		= od_insl,
-	.mv_outsb		= od_outsb,
-	.mv_outsw		= od_outsw,
-	.mv_outsl		= od_outsl,
-
-#ifdef CONFIG_PCI
-	.mv_init_irq		= init_overdrive_irq,
-#endif
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_od,
-#endif
-};
-
-ALIAS_MV(od)
diff --git a/arch/sh/boards/overdrive/pcidma.c b/arch/sh/boards/overdrive/pcidma.c
deleted file mode 100644
index 1c9bfeda00b7..000000000000
--- a/arch/sh/boards/overdrive/pcidma.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Dynamic DMA mapping support.
- *
- * On the overdrive, we can only DMA from memory behind the PCI bus!
- * this means that all DMA'able memory must come from there. 
- * this restriction will not apply to later boards.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-
-void *pci_alloc_consistent(struct pci_dev *hwdev, size_t size,
-			   dma_addr_t * dma_handle)
-{
-	void *ret;
-	int gfp = GFP_ATOMIC;
-
-        printk("BUG: pci_alloc_consistent() called - not yet supported\n");
-	/* We ALWAYS need DMA memory on the overdrive hardware,
-	 * due to it's extreme weirdness
-	 * Need to flush the cache here as well, since the memory
-	 * can still be seen through the cache!
-	 */
-	gfp |= GFP_DMA;
-	ret = (void *) __get_free_pages(gfp, get_order(size));
-
-	if (ret != NULL) {
-		memset(ret, 0, size);
-		*dma_handle = virt_to_bus(ret);
-	}
-	return ret;
-}
-
-void pci_free_consistent(struct pci_dev *hwdev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
-{
-	free_pages((unsigned long) vaddr, get_order(size));
-}
diff --git a/arch/sh/boards/overdrive/setup.c b/arch/sh/boards/overdrive/setup.c
deleted file mode 100644
index a3a7744c2047..000000000000
--- a/arch/sh/boards/overdrive/setup.c
+++ /dev/null
@@ -1,36 +0,0 @@
-/*
- * arch/sh/overdrive/setup.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * STMicroelectronics Overdrive Support.
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <asm/io.h>
-
-#include <asm/overdrive/overdrive.h>
-#include <asm/overdrive/fpga.h>
-
-const char *get_system_type(void)
-{
-	return "SH7750 Overdrive";
-}
-
-/*
- * Initialize the board
- */
-int __init platform_setup(void)
-{
-#ifdef CONFIG_PCI
-	init_overdrive_fpga();
-	galileo_init();
-#endif
-
-        /* Enable RS232 receive buffers */
-	writel(0x1e, OVERDRIVE_CTRL);
-}
diff --git a/arch/sh/boards/sh2000/Makefile b/arch/sh/boards/sh2000/Makefile
deleted file mode 100644
index 05d390c3599c..000000000000
--- a/arch/sh/boards/sh2000/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for the SH2000 specific parts of the kernel
-#
-
-obj-y	 := setup.o
-
diff --git a/arch/sh/boards/sh2000/setup.c b/arch/sh/boards/sh2000/setup.c
deleted file mode 100644
index 2fe6a11765e9..000000000000
--- a/arch/sh/boards/sh2000/setup.c
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- * linux/arch/sh/kernel/setup_sh2000.c
- *
- * Copyright (C) 2001  SUGIOKA Tochinobu
- *
- * SH-2000 Support.
- *
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/io.h>
-#include <asm/machvec.h>
-#include <asm/mach/sh2000.h>
-
-#define CF_CIS_BASE	0xb4200000
-
-#define PORT_PECR	0xa4000108
-#define PORT_PHCR	0xa400010E
-#define	PORT_ICR1	0xa4000010
-#define	PORT_IRR0	0xa4000004
-
-#define IDE_OFFSET	0xb6200000
-#define NIC_OFFSET	0xb6000000
-#define EXTBUS_OFFSET	0xba000000
-
-
-const char *get_system_type(void)
-{
-	return "sh2000";
-}
-
-static unsigned long sh2000_isa_port2addr(unsigned long offset)
-{
-	if((offset & ~7) == 0x1f0 || offset == 0x3f6)
-		return IDE_OFFSET + offset;
-	else if((offset & ~0x1f) == 0x300)
-		return NIC_OFFSET + offset;
-	return EXTBUS_OFFSET + offset;
-}
-
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_sh2000 __initmv = {
-        .mv_nr_irqs		= 80,
-        .mv_isa_port2addr	= sh2000_isa_port2addr,
-};
-ALIAS_MV(sh2000)
-
-/*
- * Initialize the board
- */
-int __init platform_setup(void)
-{
-	/* XXX: RTC setting comes here */
-
-	/* These should be done by BIOS/IPL ... */
-	/* Enable nCE2A, nCE2B output */
-	ctrl_outw(ctrl_inw(PORT_PECR) & ~0xf00, PORT_PECR);
-	/* Enable the Compact Flash card, and set the level interrupt */
-	ctrl_outw(0x0042, CF_CIS_BASE+0x0200);
-	/* Enable interrupt */
-	ctrl_outw(ctrl_inw(PORT_PHCR) & ~0x03f3, PORT_PHCR);
-	ctrl_outw(1, PORT_ICR1);
-	ctrl_outw(ctrl_inw(PORT_IRR0) & ~0xff3f, PORT_IRR0);
-	printk(KERN_INFO "SH-2000 Setup...done\n");
-	return 0;
-}
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index ffd218083876..b30e784b3a31 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -13,11 +13,9 @@ SE			SH_SOLUTION_ENGINE
 HP6XX			SH_HP6XX
 HD64461			HD64461
 HD64465			HD64465
-SH2000			SH_SH2000
 SATURN			SH_SATURN
 DREAMCAST		SH_DREAMCAST
 BIGSUR			SH_BIGSUR
-ADX			SH_ADX
 MPC1211			SH_MPC1211
 SNAPGEAR		SH_SECUREEDGE5410
 HS7751RVOIP		SH_HS7751RVOIP
diff --git a/include/asm-sh/adx/io.h b/include/asm-sh/adx/io.h
deleted file mode 100644
index ab1225f1d557..000000000000
--- a/include/asm-sh/adx/io.h
+++ /dev/null
@@ -1,86 +0,0 @@
-/*
- * include/asm-sh/io_adx.h
- *
- * Copyright (C) 2001 A&D Co., Ltd.
- *
- * This file may be copied or modified under the terms of the GNU
- * General Public License.  See linux/COPYING for more information.
- *
- * IO functions for an A&D ADX Board
- */
-
-#ifndef _ASM_SH_IO_ADX_H
-#define _ASM_SH_IO_ADX_H
-
-#include <asm/io_generic.h>
-
-extern unsigned char adx_inb(unsigned long port);
-extern unsigned short adx_inw(unsigned long port);
-extern unsigned int adx_inl(unsigned long port);
-
-extern void adx_outb(unsigned char value, unsigned long port);
-extern void adx_outw(unsigned short value, unsigned long port);
-extern void adx_outl(unsigned int value, unsigned long port);
-
-extern unsigned char adx_inb_p(unsigned long port);
-extern void adx_outb_p(unsigned char value, unsigned long port);
-
-extern void adx_insb(unsigned long port, void *addr, unsigned long count);
-extern void adx_insw(unsigned long port, void *addr, unsigned long count);
-extern void adx_insl(unsigned long port, void *addr, unsigned long count);
-extern void adx_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void adx_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void adx_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned char adx_readb(unsigned long addr);
-extern unsigned short adx_readw(unsigned long addr);
-extern unsigned int adx_readl(unsigned long addr);
-extern void adx_writeb(unsigned char b, unsigned long addr);
-extern void adx_writew(unsigned short b, unsigned long addr);
-extern void adx_writel(unsigned int b, unsigned long addr);
-
-extern void * adx_ioremap(unsigned long offset, unsigned long size);
-extern void adx_iounmap(void *addr);
-
-extern unsigned long adx_isa_port2addr(unsigned long offset);
-
-extern void setup_adx(void);
-extern void init_adx_IRQ(void);
-
-#ifdef __WANT_IO_DEF
-
-#define __inb		adx_inb
-#define __inw		adx_inw
-#define __inl		adx_inl
-#define __outb		adx_outb
-#define __outw		adx_outw
-#define __outl		adx_outl
-
-#define __inb_p		adx_inb_p
-#define __inw_p		adx_inw
-#define __inl_p		adx_inl
-#define __outb_p	adx_outb_p
-#define __outw_p	adx_outw
-#define __outl_p	adx_outl
-
-#define __insb		adx_insb
-#define __insw		adx_insw
-#define __insl		adx_insl
-#define __outsb		adx_outsb
-#define __outsw		adx_outsw
-#define __outsl		adx_outsl
-
-#define __readb		adx_readb
-#define __readw		adx_readw
-#define __readl		adx_readl
-#define __writeb	adx_writeb
-#define __writew	adx_writew
-#define __writel	adx_writel
-
-#define __isa_port2addr	adx_isa_port2addr
-#define __ioremap	adx_ioremap
-#define __iounmap	adx_iounmap
-
-#endif
-
-#endif /* _ASM_SH_IO_AANDD_H */
diff --git a/include/asm-sh/cat68701/io.h b/include/asm-sh/cat68701/io.h
deleted file mode 100644
index 753b8466ad11..000000000000
--- a/include/asm-sh/cat68701/io.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- * include/asm-sh/io_cat68701.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *           2001 Yutarou Ebihar (ebihara@si-linux.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an AONE Corp. CAT-68701 SH7708 Borad
- */
-
-#ifndef _ASM_SH_IO_CAT68701_H
-#define _ASM_SH_IO_CAT68701_H
-
-extern unsigned long cat68701_isa_port2addr(unsigned long offset);
-extern int cat68701_irq_demux(int irq);
-
-extern void init_cat68701_IRQ(void);
-extern void heartbeat_cat68701(void);
-
-#endif /* _ASM_SH_IO_CAT68701_H */
diff --git a/include/asm-sh/cqreek/cqreek.h b/include/asm-sh/cqreek/cqreek.h
deleted file mode 100644
index 09aecc06693e..000000000000
--- a/include/asm-sh/cqreek/cqreek.h
+++ /dev/null
@@ -1,27 +0,0 @@
-#ifndef __ASM_SH_CQREEK_CQREEK_H
-#define __ASM_SH_CQREEK_CQREEK_H
-
-#define BRIDGE_FEATURE		0x0002
-
-#define BRIDGE_IDE_CTRL		0x0018
-#define BRIDGE_IDE_INTR_LVL    	0x001A
-#define BRIDGE_IDE_INTR_MASK	0x001C
-#define BRIDGE_IDE_INTR_STAT	0x001E
-
-#define BRIDGE_ISA_CTRL		0x0028
-#define BRIDGE_ISA_INTR_LVL    	0x002A
-#define BRIDGE_ISA_INTR_MASK	0x002C
-#define BRIDGE_ISA_INTR_STAT	0x002E
-
-/* arch/sh/boards/cqreek/setup.c */
-extern void setup_cqreek(void);
-
-/* arch/sh/boards/cqreek/irq.c */
-extern int cqreek_has_ide, cqreek_has_isa;
-extern void init_cqreek_IRQ(void);
-
-/* arch/sh/boards/cqreek/io.c */
-extern unsigned long cqreek_port2addr(unsigned long port);
-
-#endif /* __ASM_SH_CQREEK_CQREEK_H */
-
diff --git a/include/asm-sh/dmida/io.h b/include/asm-sh/dmida/io.h
deleted file mode 100644
index 21bd416c01c3..000000000000
--- a/include/asm-sh/dmida/io.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef __ASM_SH_DMIDA_IO_H
-#define __ASM_SH_DMIDA_IO_H
-
-/*
- * Nothing special here.. just use the generic cchip io routines.
- */
-#include <asm/hd64465/io.h>
-
-#endif /* __ASM_SH_DMIDA_IO_H */
-
diff --git a/include/asm-sh/harp/harp.h b/include/asm-sh/harp/harp.h
deleted file mode 100644
index b2fbcfae9940..000000000000
--- a/include/asm-sh/harp/harp.h
+++ /dev/null
@@ -1,43 +0,0 @@
-/* 
- * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Defintions applicable to the STMicroelectronics ST40STB1 HARP and
- * compatible boards.
- */
-
-#if defined(CONFIG_SH_STB1_HARP)
-
-#define EPLD_BASE     0xa0800000
-
-#define EPLD_LED      (EPLD_BASE+0x000c0000)
-#define EPLD_INTSTAT0 (EPLD_BASE+0x00200000)
-#define EPLD_INTSTAT1 (EPLD_BASE+0x00240000)
-#define EPLD_INTMASK0 (EPLD_BASE+0x00280000)
-#define EPLD_INTMASK1 (EPLD_BASE+0x002c0000)
-#define EPLD_PAGEADDR (EPLD_BASE+0x00300000)
-#define EPLD_REVID1   (EPLD_BASE+0x00380000)
-#define EPLD_REVID2   (EPLD_BASE+0x003c0000)
-
-#define EPLD_LED_ON  1
-#define EPLD_LED_OFF 0
-
-#elif defined(CONFIG_SH_STB1_OVERDRIVE)
-
-#define EPLD_BASE     0xa7000000
-
-#define EPLD_REVID    (EPLD_BASE+0x00000000)
-#define EPLD_LED      (EPLD_BASE+0x00040000)
-#define EPLD_INTMASK0 (EPLD_BASE+0x001c0000)
-#define EPLD_INTMASK1 (EPLD_BASE+0x00200000)
-#define EPLD_INTSTAT0 (EPLD_BASE+0x00240000)
-#define EPLD_INTSTAT1 (EPLD_BASE+0x00280000)
-
-#define EPLD_LED_ON  0
-#define EPLD_LED_OFF 1
-
-#else
-#error Unknown board
-#endif
diff --git a/include/asm-sh/harp/io.h b/include/asm-sh/harp/io.h
deleted file mode 100644
index 68f39e0b39de..000000000000
--- a/include/asm-sh/harp/io.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef __ASM_SH_HARP_IO_H
-#define __ASM_SH_HARP_IO_H
-
-/*
- * Nothing special here.. just use the generic cchip io routines.
- */
-#include <asm/hd64465/io.h>
-
-#endif /* __ASM_SH_HARP_IO_H */
-
diff --git a/include/asm-sh/overdrive/fpga.h b/include/asm-sh/overdrive/fpga.h
deleted file mode 100644
index 1cd87992c124..000000000000
--- a/include/asm-sh/overdrive/fpga.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- */
-
-#ifndef __FPGA_OD_H__
-#define __FPGA_OD_H__
-
-/* This routine will program up the fpga which interfaces to the galileo */
-int init_overdrive_fpga(void);
-
-#endif
diff --git a/include/asm-sh/overdrive/gt64111.h b/include/asm-sh/overdrive/gt64111.h
deleted file mode 100644
index 01d58bc13a44..000000000000
--- a/include/asm-sh/overdrive/gt64111.h
+++ /dev/null
@@ -1,109 +0,0 @@
-#ifndef _GT64111_H_
-#define _GT64111_H_
-
-#define MASTER_INTERFACE         0x0
-#define RAS10_LO_DEC_ADR         0x8        
-#define RAS10_HI_DEC_ADR         0x10
-#define RAS32_LO_DEC_ADR         0x18
-#define RAS32_HI_DEC_ADR         0x20
-#define CS20_LO_DEC_ADR          0x28
-#define CS20_HI_DEC_ADR          0x30
-#define CS3_LO_DEC_ADR           0x38
-#define CS3_HI_DEC_ADR           0x40
-#define PCI_IO_LO_DEC_ADR        0x48
-#define PCI_IO_HI_DEC_ADR        0x50
-#define PCI_MEM0_LO_DEC_ADR      0x58
-#define PCI_MEM0_HI_DEC_ADR      0x60
-#define INTERNAL_SPACE_DEC       0x68
-#define BUS_ERR_ADR_LO_CPU       0x70
-#define READONLY0                0x78
-#define PCI_MEM1_LO_DEC_ADR      0x80
-#define PCI_MEM1_HI_DEC_ADR      0x88
-#define RAS0_LO_DEC_ADR          0x400   
-#define RAS0_HI_DEC_ADR          0x404
-#define RAS1_LO_DEC_ADR          0x408
-#define RAS1_HI_DEC_ADR          0x40c
-#define RAS2_LO_DEC_ADR          0x410
-#define RAS2_HI_DEC_ADR          0x414
-#define RAS3_LO_DEC_ADR          0x418
-#define RAS3_HI_DEC_ADR          0x41c
-#define DEV_CS0_LO_DEC_ADR       0x420
-#define DEV_CS0_HI_DEC_ADR       0x424
-#define DEV_CS1_LO_DEC_ADR       0x428
-#define DEV_CS1_HI_DEC_ADR       0x42c
-#define DEV_CS2_LO_DEC_ADR       0x430
-#define DEV_CS2_HI_DEC_ADR       0x434
-#define DEV_CS3_LO_DEC_ADR       0x438
-#define DEV_CS3_HI_DEC_ADR       0x43c
-#define DEV_BOOTCS_LO_DEC_ADR    0x440
-#define DEV_BOOTCS_HI_DEC_ADR    0x444
-#define DEV_ADR_DEC_ERR          0x470
-#define DRAM_CFG                 0x448   
-#define DRAM_BANK0_PARMS         0x44c   
-#define DRAM_BANK1_PARMS         0x450
-#define DRAM_BANK2_PARMS         0x454
-#define DRAM_BANK3_PARMS         0x458
-#define DEV_BANK0_PARMS          0x45c
-#define DEV_BANK1_PARMS          0x460
-#define DEV_BANK2_PARMS          0x464
-#define DEV_BANK3_PARMS          0x468
-#define DEV_BOOT_BANK_PARMS      0x46c
-#define CH0_DMA_BYTECOUNT        0x800
-#define CH1_DMA_BYTECOUNT        0x804
-#define CH2_DMA_BYTECOUNT        0x808
-#define CH3_DMA_BYTECOUNT        0x80c
-#define CH0_DMA_SRC_ADR          0x810
-#define CH1_DMA_SRC_ADR          0x814
-#define CH2_DMA_SRC_ADR          0x818
-#define CH3_DMA_SRC_ADR          0x81c
-#define CH0_DMA_DST_ADR          0x820
-#define CH1_DMA_DST_ADR          0x824
-#define CH2_DMA_DST_ADR          0x828
-#define CH3_DMA_DST_ADR          0x82c
-#define CH0_NEXT_REC_PTR         0x830
-#define CH1_NEXT_REC_PTR         0x834
-#define CH2_NEXT_REC_PTR         0x838
-#define CH3_NEXT_REC_PTR         0x83c
-#define CH0_CTRL                 0x840
-#define CH1_CTRL                 0x844
-#define CH2_CTRL                 0x848
-#define CH3_CTRL                 0x84c
-#define DMA_ARBITER              0x860
-#define TIMER0                   0x850
-#define TIMER1                   0x854
-#define TIMER2                   0x858
-#define TIMER3                   0x85c
-#define TIMER_CTRL               0x864
-#define PCI_CMD                  0xc00
-#define PCI_TIMEOUT              0xc04
-#define PCI_RAS10_BANK_SIZE      0xc08
-#define PCI_RAS32_BANK_SIZE      0xc0c
-#define PCI_CS20_BANK_SIZE       0xc10
-#define PCI_CS3_BANK_SIZE        0xc14
-#define PCI_SERRMASK             0xc28
-#define PCI_INTACK               0xc34
-#define PCI_BAR_EN               0xc3c
-#define PCI_CFG_ADR              0xcf8
-#define PCI_CFG_DATA             0xcfc
-#define PCI_INTCAUSE             0xc18
-#define PCI_MAST_MASK            0xc1c
-#define PCI_PCIMASK              0xc24
-#define BAR_ENABLE_ADR           0xc3c
-
-/* These are config registers, accessible via PCI space */
-#define PCI_CONFIG_RAS10_BASE_ADR   0x010
-#define PCI_CONFIG_RAS32_BASE_ADR   0x014
-#define PCI_CONFIG_CS20_BASE_ADR    0x018
-#define PCI_CONFIG_CS3_BASE_ADR     0x01c
-#define PCI_CONFIG_INT_REG_MM_ADR   0x020
-#define PCI_CONFIG_INT_REG_IO_ADR   0x024
-#define PCI_CONFIG_BOARD_VENDOR     0x02c
-#define PCI_CONFIG_ROM_ADR          0x030
-#define PCI_CONFIG_INT_PIN_LINE     0x03c
-
-
-
-
-
-#endif
-
diff --git a/include/asm-sh/overdrive/io.h b/include/asm-sh/overdrive/io.h
deleted file mode 100644
index 0dba700e9643..000000000000
--- a/include/asm-sh/overdrive/io.h
+++ /dev/null
@@ -1,39 +0,0 @@
-/*
- * include/asm-sh/io_od.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an STMicroelectronics Overdrive
- */
-
-#ifndef _ASM_SH_IO_OD_H
-#define _ASM_SH_IO_OD_H
-
-extern unsigned char od_inb(unsigned long port);
-extern unsigned short od_inw(unsigned long port);
-extern unsigned int od_inl(unsigned long port);
-
-extern void od_outb(unsigned char value, unsigned long port);
-extern void od_outw(unsigned short value, unsigned long port);
-extern void od_outl(unsigned int value, unsigned long port);
-
-extern unsigned char od_inb_p(unsigned long port);
-extern unsigned short od_inw_p(unsigned long port);
-extern unsigned int od_inl_p(unsigned long port);
-extern void od_outb_p(unsigned char value, unsigned long port);
-extern void od_outw_p(unsigned short value, unsigned long port);
-extern void od_outl_p(unsigned int value, unsigned long port);
-
-extern void od_insb(unsigned long port, void *addr, unsigned long count);
-extern void od_insw(unsigned long port, void *addr, unsigned long count);
-extern void od_insl(unsigned long port, void *addr, unsigned long count);
-extern void od_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void od_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void od_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned long od_isa_port2addr(unsigned long offset);
-
-#endif /* _ASM_SH_IO_OD_H */
diff --git a/include/asm-sh/overdrive/overdrive.h b/include/asm-sh/overdrive/overdrive.h
deleted file mode 100644
index fc746c244f83..000000000000
--- a/include/asm-sh/overdrive/overdrive.h
+++ /dev/null
@@ -1,88 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- */
-
-
-#ifndef __OVERDRIVE_H__
-#define __OVERDRIVE_H__
-
-#define OVERDRIVE_INT_CT 0xa3a00000
-#define OVERDRIVE_INT_DT 0xa3b00000
-
-#define OVERDRIVE_CTRL    0xa3000000
-
-/* Shoving all these bits into the same register is not a good idea. 
- * As soon as I get a spare moment, I'll change the FPGA and put each 
- * bit in a separate register
- */
-
-#define VALID_CTRL_BITS		          0x1f
-
-#define ENABLE_RS232_MASK	  	  0x1e
-#define DISABLE_RS232_BIT		  0x01
-
-#define ENABLE_NMI_MASK			  0x1d
-#define DISABLE_NMI_BIT			  0x02
-
-#define RESET_PCI_MASK			  0x1b
-#define ENABLE_PCI_BIT			  0x04
-
-#define ENABLE_LED_MASK			  0x17
-#define DISABLE_LED_BIT			  0x08
-
-#define RESET_FPGA_MASK			  0x0f
-#define ENABLE_FPGA_BIT			  0x10
-
-
-#define FPGA_DCLK_ADDRESS           0xA3C00000
-
-#define FPGA_DATA        0x01	/*   W */
-#define FPGA_CONFDONE    0x02	/* R   */
-#define FPGA_NOT_STATUS  0x04	/* R   */
-#define FPGA_INITDONE    0x08	/* R   */
-
-#define FPGA_TIMEOUT     100000
-
-
-/* Interrupts for the overdrive. Note that these numbers have 
- * nothing to do with the actual IRQ numbers they appear on, 
- * this is all programmable. This is simply the position in the 
- * INT_CT register.
- */
-
-#define OVERDRIVE_PCI_INTA              0
-#define OVERDRIVE_PCI_INTB              1
-#define OVERDRIVE_PCI_INTC              2
-#define OVERDRIVE_PCI_INTD              3
-#define OVERDRIVE_GALILEO_INT           4
-#define OVERDRIVE_GALILEO_LOCAL_INT     5
-#define OVERDRIVE_AUDIO_INT             6
-#define OVERDRIVE_KEYBOARD_INT          7
-
-/* Which Linux IRQ should we assign to each interrupt source? */
-#define OVERDRIVE_PCI_IRQ1              2
-#ifdef CONFIG_HACKED_NE2K
-#define OVERDRIVE_PCI_IRQ2              7
-#else
-#define OVERDRIVE_PCI_IRQ2              2
-#undef OVERDRIVE_PCI_INTB 
-#define OVERDRIVE_PCI_INTB OVERDRIVE_PCI_INTA
-
-#endif
-
-/* Put the ESS solo audio chip on IRQ 4 */
-#define OVERDRIVE_ESS_IRQ               4
-
-/* Where the memory behind the PCI bus appears */
-#define PCI_DRAM_BASE   0xb7000000
-#define PCI_DRAM_SIZE (16*1024*1024)
-#define PCI_DRAM_FINISH (PCI_DRAM_BASE+PCI_DRAM_SIZE-1)
-
-/* Where the IO region appears in the memory */
-#define PCI_GTIO_BASE   0xb8000000
-
-#endif
diff --git a/include/asm-sh/sh2000/sh2000.h b/include/asm-sh/sh2000/sh2000.h
deleted file mode 100644
index 8d547324d59a..000000000000
--- a/include/asm-sh/sh2000/sh2000.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef __ASM_SH_SH2000_SH2000_H
-#define __ASM_SH_SH2000_SH2000_H
-
-/* arch/sh/boards/sh2000/setup.c */
-extern int setup_sh2000(void);
-
-#endif /* __ASM_SH_SH2000_SH2000_H */
-
-- 
cgit v1.2.3


From bc8fb5d0471473f775378d09db712dcb8eeece75 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:09:34 +0900
Subject: sh: Solution Engine SH7343 board support.

This adds support for the SE7343 board.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  |  35 +-
 arch/sh/Makefile                 |   1 +
 arch/sh/boards/se/7343/Makefile  |   7 +
 arch/sh/boards/se/7343/io.c      | 275 +++++++++++
 arch/sh/boards/se/7343/irq.c     | 193 ++++++++
 arch/sh/boards/se/7343/led.c     |  46 ++
 arch/sh/boards/se/7343/setup.c   |  84 ++++
 arch/sh/configs/se7343_defconfig | 997 +++++++++++++++++++++++++++++++++++++++
 arch/sh/tools/mach-types         |   1 +
 include/asm-sh/irq.h             |   2 +
 include/asm-sh/se7343.h          |  82 ++++
 11 files changed, 1710 insertions(+), 13 deletions(-)
 create mode 100644 arch/sh/boards/se/7343/Makefile
 create mode 100644 arch/sh/boards/se/7343/io.c
 create mode 100644 arch/sh/boards/se/7343/irq.c
 create mode 100644 arch/sh/boards/se/7343/led.c
 create mode 100644 arch/sh/boards/se/7343/setup.c
 create mode 100644 arch/sh/configs/se7343_defconfig
 create mode 100644 include/asm-sh/se7343.h

(limited to 'include')

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 2549bd82c57f..9e5272e84fd8 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -51,18 +51,23 @@ source "init/Kconfig"
 
 menu "System type"
 
+config SOLUTION_ENGINE
+	bool
+
 choice
 	prompt "SuperH system type"
 	default SH_UNKNOWN
 
 config SH_SOLUTION_ENGINE
 	bool "SolutionEngine"
+	select SOLUTION_ENGINE
 	help
 	  Select SolutionEngine if configuring for a Hitachi SH7709
 	  or SH7750 evaluation board.
 
 config SH_7751_SOLUTION_ENGINE
 	bool "SolutionEngine7751"
+	select SOLUTION_ENGINE
 	select CPU_SUBTYPE_SH7751
 	help
 	  Select 7751 SolutionEngine if configuring for a Hitachi SH7751
@@ -70,17 +75,27 @@ config SH_7751_SOLUTION_ENGINE
 
 config SH_7300_SOLUTION_ENGINE
 	bool "SolutionEngine7300"
+	select SOLUTION_ENGINE
 	select CPU_SUBTYPE_SH7300
 	help
-	  Select 7300 SolutionEngine if configuring for a Hitachi SH7300(SH-Mobile V)
-	  evaluation board.
+	  Select 7300 SolutionEngine if configuring for a Hitachi
+	  SH7300(SH-Mobile V) evaluation board.
+
+config SH_7343_SOLUTION_ENGINE
+	bool "SolutionEngine7343"
+	select SOLUTION_ENGINE
+	select CPU_SUBTYPE_SH7343
+	help
+	  Select 7343 SolutionEngine if configuring for a Hitachi
+	  SH7343 (SH-Mobile 3AS) evaluation board.
 
 config SH_73180_SOLUTION_ENGINE
        bool "SolutionEngine73180"
-       select CPU_SUBTYPE_SH73180
-       help
-         Select 73180 SolutionEngine if configuring for a Hitachi SH73180(SH-Mobile 3)
-         evaluation board.
+	select SOLUTION_ENGINE
+	select CPU_SUBTYPE_SH73180
+	help
+	  Select 73180 SolutionEngine if configuring for a Hitachi
+	  SH73180(SH-Mobile 3) evaluation board.
 
 config SH_7751_SYSTEMH
 	bool "SystemH7751R"
@@ -394,9 +409,7 @@ source "arch/sh/cchips/Kconfig"
 config HEARTBEAT
 	bool "Heartbeat LED"
 	depends on SH_MPC1211 || SH_SH03 || \
-		   SH_BIGSUR || \
-		   SH_7751_SOLUTION_ENGINE || SH_7300_SOLUTION_ENGINE || \
-		   SH_73180_SOLUTION_ENGINE || SH_SOLUTION_ENGINE || \
+		   SH_BIGSUR || SOLUTION_ENGINE || \
 		   SH_RTS7751R2D || SH_SH4202_MICRODEV || SH_LANDISK
 	help
 	  Use the power-on LED on your machine as a load meter.  The exact
@@ -431,10 +444,6 @@ config KEXEC
 	  support.  As of this writing the exact hardware interface is
 	  strongly in flux, so no good recommendation can be made.
 
-config PREEMPT
-	bool "Preemptible Kernel (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
-
 config SMP
 	bool "Symmetric multi-processing support"
 	---help---
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index f0fc88e5f1ad..26d62ff51a64 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -89,6 +89,7 @@ core-$(CONFIG_SH_FPU_EMU)	+= arch/sh/math-emu/
 machdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se/770x
 machdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se/7751
 machdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)	:= se/7300
+machdir-$(CONFIG_SH_7343_SOLUTION_ENGINE)	:= se/7343
 machdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se/73180
 machdir-$(CONFIG_SH_HP6XX)			:= hp6xx
 machdir-$(CONFIG_SH_EC3104)			:= ec3104
diff --git a/arch/sh/boards/se/7343/Makefile b/arch/sh/boards/se/7343/Makefile
new file mode 100644
index 000000000000..4291069c0b4f
--- /dev/null
+++ b/arch/sh/boards/se/7343/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for the 7343 SolutionEngine specific parts of the kernel
+#
+
+obj-y	 := setup.o io.o irq.o
+
+obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/se/7343/io.c b/arch/sh/boards/se/7343/io.c
new file mode 100644
index 000000000000..646661a146ad
--- /dev/null
+++ b/arch/sh/boards/se/7343/io.c
@@ -0,0 +1,275 @@
+/*
+ * arch/sh/boards/se/7343/io.c
+ *
+ * I/O routine for SH-Mobile3AS 7343 SolutionEngine.
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <asm/io.h>
+#include <asm/mach/se7343.h>
+
+#define badio(fn, a) panic("bad i/o operation %s for %08lx.", #fn, a)
+
+struct iop {
+	unsigned long start, end;
+	unsigned long base;
+	struct iop *(*check) (struct iop * p, unsigned long port);
+	unsigned char (*inb) (struct iop * p, unsigned long port);
+	unsigned short (*inw) (struct iop * p, unsigned long port);
+	void (*outb) (struct iop * p, unsigned char value, unsigned long port);
+	void (*outw) (struct iop * p, unsigned short value, unsigned long port);
+};
+
+struct iop *
+simple_check(struct iop *p, unsigned long port)
+{
+	static int count;
+
+	if (count < 100)
+		count++;
+
+	port &= 0xFFFF;
+
+	if ((p->start <= port) && (port <= p->end))
+		return p;
+	else
+		badio(check, port);
+}
+
+struct iop *
+ide_check(struct iop *p, unsigned long port)
+{
+	if (((0x1f0 <= port) && (port <= 0x1f7)) || (port == 0x3f7))
+		return p;
+	return NULL;
+}
+
+unsigned char
+simple_inb(struct iop *p, unsigned long port)
+{
+	return *(unsigned char *) (p->base + port);
+}
+
+unsigned short
+simple_inw(struct iop *p, unsigned long port)
+{
+	return *(unsigned short *) (p->base + port);
+}
+
+void
+simple_outb(struct iop *p, unsigned char value, unsigned long port)
+{
+	*(unsigned char *) (p->base + port) = value;
+}
+
+void
+simple_outw(struct iop *p, unsigned short value, unsigned long port)
+{
+	*(unsigned short *) (p->base + port) = value;
+}
+
+unsigned char
+pcc_inb(struct iop *p, unsigned long port)
+{
+	unsigned long addr = p->base + port + 0x40000;
+	unsigned long v;
+
+	if (port & 1)
+		addr += 0x00400000;
+	v = *(volatile unsigned char *) addr;
+	return v;
+}
+
+void
+pcc_outb(struct iop *p, unsigned char value, unsigned long port)
+{
+	unsigned long addr = p->base + port + 0x40000;
+
+	if (port & 1)
+		addr += 0x00400000;
+	*(volatile unsigned char *) addr = value;
+}
+
+unsigned char
+bad_inb(struct iop *p, unsigned long port)
+{
+	badio(inb, port);
+}
+
+void
+bad_outb(struct iop *p, unsigned char value, unsigned long port)
+{
+	badio(inw, port);
+}
+
+#ifdef CONFIG_SMC91X
+/* MSTLANEX01 LAN at 0xb400:0000 */
+static struct iop laniop = {
+	.start = 0x00,
+	.end = 0x0F,
+	.base = 0x04000000,
+	.check = simple_check,
+	.inb = simple_inb,
+	.inw = simple_inw,
+	.outb = simple_outb,
+	.outw = simple_outw,
+};
+#endif
+
+#ifdef CONFIG_NE2000
+/* NE2000 pc card NIC */
+static struct iop neiop = {
+	.start = 0x280,
+	.end = 0x29f,
+	.base = 0xb0600000 + 0x80,	/* soft 0x280 -> hard 0x300 */
+	.check = simple_check,
+	.inb = pcc_inb,
+	.inw = simple_inw,
+	.outb = pcc_outb,
+	.outw = simple_outw,
+};
+#endif
+
+#ifdef CONFIG_IDE
+/* CF in CF slot */
+static struct iop cfiop = {
+	.base = 0xb0600000,
+	.check = ide_check,
+	.inb = pcc_inb,
+	.inw = simple_inw,
+	.outb = pcc_outb,
+	.outw = simple_outw,
+};
+#endif
+
+static __inline__ struct iop *
+port2iop(unsigned long port)
+{
+	if (0) ;
+#if defined(CONFIG_SMC91X)
+	else if (laniop.check(&laniop, port))
+		return &laniop;
+#endif
+#if defined(CONFIG_NE2000)
+	else if (neiop.check(&neiop, port))
+		return &neiop;
+#endif
+#if defined(CONFIG_IDE)
+	else if (cfiop.check(&cfiop, port))
+		return &cfiop;
+#endif
+	else
+		return NULL;
+}
+
+static inline void
+delay(void)
+{
+	ctrl_inw(0xac000000);
+	ctrl_inw(0xac000000);
+}
+
+unsigned char
+sh7343se_inb(unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	return (p->inb) (p, port);
+}
+
+unsigned char
+sh7343se_inb_p(unsigned long port)
+{
+	unsigned char v = sh7343se_inb(port);
+	delay();
+	return v;
+}
+
+unsigned short
+sh7343se_inw(unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	return (p->inw) (p, port);
+}
+
+unsigned int
+sh7343se_inl(unsigned long port)
+{
+	badio(inl, port);
+}
+
+void
+sh7343se_outb(unsigned char value, unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	(p->outb) (p, value, port);
+}
+
+void
+sh7343se_outb_p(unsigned char value, unsigned long port)
+{
+	sh7343se_outb(value, port);
+	delay();
+}
+
+void
+sh7343se_outw(unsigned short value, unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	(p->outw) (p, value, port);
+}
+
+void
+sh7343se_outl(unsigned int value, unsigned long port)
+{
+	badio(outl, port);
+}
+
+void
+sh7343se_insb(unsigned long port, void *addr, unsigned long count)
+{
+	unsigned char *a = addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		*a++ = (p->inb) (p, port);
+}
+
+void
+sh7343se_insw(unsigned long port, void *addr, unsigned long count)
+{
+	unsigned short *a = addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		*a++ = (p->inw) (p, port);
+}
+
+void
+sh7343se_insl(unsigned long port, void *addr, unsigned long count)
+{
+	badio(insl, port);
+}
+
+void
+sh7343se_outsb(unsigned long port, const void *addr, unsigned long count)
+{
+	unsigned char *a = (unsigned char *) addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		(p->outb) (p, *a++, port);
+}
+
+void
+sh7343se_outsw(unsigned long port, const void *addr, unsigned long count)
+{
+	unsigned short *a = (unsigned short *) addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		(p->outw) (p, *a++, port);
+}
+
+void
+sh7343se_outsl(unsigned long port, const void *addr, unsigned long count)
+{
+	badio(outsw, port);
+}
diff --git a/arch/sh/boards/se/7343/irq.c b/arch/sh/boards/se/7343/irq.c
new file mode 100644
index 000000000000..b41e3d4ea37c
--- /dev/null
+++ b/arch/sh/boards/se/7343/irq.c
@@ -0,0 +1,193 @@
+/*
+ * arch/sh/boards/se/7343/irq.c
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <asm/irq.h>
+#include <asm/io.h>
+#include <asm/mach/se7343.h>
+
+static void
+disable_intreq_irq(unsigned int irq)
+{
+	int bit = irq - OFFCHIP_IRQ_BASE;
+	u16 val;
+
+	val = ctrl_inw(PA_CPLD_IMSK);
+	val |= 1 << bit;
+	ctrl_outw(val, PA_CPLD_IMSK);
+}
+
+static void
+enable_intreq_irq(unsigned int irq)
+{
+	int bit = irq - OFFCHIP_IRQ_BASE;
+	u16 val;
+
+	val = ctrl_inw(PA_CPLD_IMSK);
+	val &= ~(1 << bit);
+	ctrl_outw(val, PA_CPLD_IMSK);
+}
+
+static void
+mask_and_ack_intreq_irq(unsigned int irq)
+{
+	disable_intreq_irq(irq);
+}
+
+static unsigned int
+startup_intreq_irq(unsigned int irq)
+{
+	enable_intreq_irq(irq);
+	return 0;
+}
+
+static void
+shutdown_intreq_irq(unsigned int irq)
+{
+	disable_intreq_irq(irq);
+}
+
+static void
+end_intreq_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+		enable_intreq_irq(irq);
+}
+
+static struct hw_interrupt_type intreq_irq_type = {
+	.typename = "FPGA-IRQ",
+	.startup = startup_intreq_irq,
+	.shutdown = shutdown_intreq_irq,
+	.enable = enable_intreq_irq,
+	.disable = disable_intreq_irq,
+	.ack = mask_and_ack_intreq_irq,
+	.end = end_intreq_irq
+};
+
+static void
+make_intreq_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &intreq_irq_type;
+	disable_intreq_irq(irq);
+}
+
+int
+shmse_irq_demux(int irq)
+{
+	int bit;
+	volatile u16 val;
+
+	if (irq == IRQ5_IRQ) {
+		/* Read status Register */
+		val = ctrl_inw(PA_CPLD_ST);
+		bit = ffs(val);
+		if (bit != 0)
+			return OFFCHIP_IRQ_BASE + bit - 1;
+	}
+	return irq;
+}
+
+/* IRQ5 is multiplexed between the following sources:
+ * 1. PC Card socket
+ * 2. Extension slot
+ * 3. USB Controller
+ * 4. Serial Controller
+ *
+ * We configure IRQ5 as a cascade IRQ.
+ */
+static struct irqaction irq5 = { no_action, 0, CPU_MASK_NONE, "IRQ5-cascade",
+				NULL, NULL};
+
+/*
+ * Initialize IRQ setting
+ */
+void __init
+init_7343se_IRQ(void)
+{
+	/* Setup Multiplexed interrupts */
+	ctrl_outw(8, PA_CPLD_MODESET);	/* Set all CPLD interrupts to active
+					 * low.
+					 */
+	/* Mask all CPLD controller interrupts */
+	ctrl_outw(0x0fff, PA_CPLD_IMSK);
+
+	/* PC Card interrupts */
+	make_intreq_irq(PC_IRQ0);
+	make_intreq_irq(PC_IRQ1);
+	make_intreq_irq(PC_IRQ2);
+	make_intreq_irq(PC_IRQ3);
+
+	/* Extension Slot Interrupts */
+	make_intreq_irq(EXT_IRQ0);
+	make_intreq_irq(EXT_IRQ1);
+	make_intreq_irq(EXT_IRQ2);
+	make_intreq_irq(EXT_IRQ3);
+
+	/* USB Controller interrupts */
+	make_intreq_irq(USB_IRQ0);
+	make_intreq_irq(USB_IRQ1);
+
+	/* Serial Controller interrupts */
+	make_intreq_irq(UART_IRQ0);
+	make_intreq_irq(UART_IRQ1);
+
+	/* Setup all external interrupts to be active low */
+	ctrl_outw(0xaaaa, INTC_ICR1);
+
+	make_ipr_irq(IRQ5_IRQ, IRQ5_IPR_ADDR+2, IRQ5_IPR_POS, IRQ5_PRIORITY);
+	setup_irq(IRQ5_IRQ, &irq5);
+	/* Set port control to use IRQ5 */
+	*(u16 *)0xA4050108 &= ~0xc;
+
+	make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY);
+	make_ipr_irq(VPU_IRQ, VPU_IPR_ADDR, VPU_IPR_POS, 8);
+
+	ctrl_outb(0x0f, INTC_IMCR5);	/* enable SCIF IRQ */
+
+	make_ipr_irq(DMTE0_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE1_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY);
+	make_ipr_irq(DMTE5_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY);
+
+	/* I2C block */
+	make_ipr_irq(IIC0_ALI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY);
+	make_ipr_irq(IIC0_TACKI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS,
+		     IIC0_PRIORITY);
+	make_ipr_irq(IIC0_WAITI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS,
+		     IIC0_PRIORITY);
+	make_ipr_irq(IIC0_DTEI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY);
+
+	make_ipr_irq(IIC1_ALI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY);
+	make_ipr_irq(IIC1_TACKI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS,
+		     IIC1_PRIORITY);
+	make_ipr_irq(IIC1_WAITI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS,
+		     IIC1_PRIORITY);
+	make_ipr_irq(IIC1_DTEI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY);
+
+	/* SIOF */
+	make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY);
+
+	/* SIU */
+	make_ipr_irq(SIU_IRQ, SIU_IPR_ADDR, SIU_IPR_POS, SIU_PRIORITY);
+
+	/* VIO interrupt */
+	make_ipr_irq(CEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY);
+	make_ipr_irq(BEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY);
+	make_ipr_irq(VEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY);
+
+	/*MFI interrupt*/
+
+	make_ipr_irq(MFI_IRQ, MFI_IPR_ADDR, MFI_IPR_POS, MFI_PRIORITY);
+
+	/* LCD controller */
+	make_ipr_irq(LCDC_IRQ, LCDC_IPR_ADDR, LCDC_IPR_POS, LCDC_PRIORITY);
+	ctrl_outw(0x2000, PA_MRSHPC + 0x0c);	/* mrshpc irq enable */
+}
diff --git a/arch/sh/boards/se/7343/led.c b/arch/sh/boards/se/7343/led.c
new file mode 100644
index 000000000000..6a439cf83e46
--- /dev/null
+++ b/arch/sh/boards/se/7343/led.c
@@ -0,0 +1,46 @@
+/*
+ * arch/sh/boards/se/7343/led.c
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/sched.h>
+#include <asm/mach/se7343.h>
+
+/* Cycle the LED's in the clasic Knightrider/Sun pattern */
+void heartbeat_7343se(void)
+{
+	static unsigned int cnt = 0, period = 0;
+	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
+	static unsigned bit = 0, up = 1;
+
+	cnt += 1;
+	if (cnt < period) {
+		return;
+	}
+
+	cnt = 0;
+
+	/* Go through the points (roughly!):
+	 * f(0)=10, f(1)=16, f(2)=20, f(5)=35,f(inf)->110
+	 */
+	period = 110 - ((300 << FSHIFT) / ((avenrun[0] / 5) + (3 << FSHIFT)));
+
+	if (up) {
+		if (bit == 7) {
+			bit--;
+			up = 0;
+		} else {
+			bit++;
+		}
+	} else {
+		if (bit == 0) {
+			bit++;
+			up = 1;
+		} else {
+			bit--;
+		}
+	}
+	*p = 1 << (bit + LED_SHIFT);
+
+}
diff --git a/arch/sh/boards/se/7343/setup.c b/arch/sh/boards/se/7343/setup.c
new file mode 100644
index 000000000000..787322291fb3
--- /dev/null
+++ b/arch/sh/boards/se/7343/setup.c
@@ -0,0 +1,84 @@
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <asm/machvec.h>
+#include <asm/mach/se7343.h>
+#include <asm/irq.h>
+
+void heartbeat_7343se(void);
+void init_7343se_IRQ(void);
+
+static struct resource smc91x_resources[] = {
+	[0] = {
+		.start	= 0x10000000,
+		.end	= 0x1000000F,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		/*
+		 * shared with other devices via externel
+		 * interrupt controller in FPGA...
+		 */
+		.start	= EXT_IRQ2,
+		.end	= EXT_IRQ2,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device smc91x_device = {
+	.name		= "smc91x",
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(smc91x_resources),
+	.resource	= smc91x_resources,
+};
+
+static struct platform_device *smc91x_platform_devices[] __initdata = {
+	&smc91x_device,
+};
+
+static int __init sh7343se_devices_setup(void)
+{
+	return platform_add_devices(smc91x_platform_devices,
+				    ARRAY_SIZE(smc91x_platform_devices));
+}
+
+static void __init sh7343se_setup(char **cmdline_p)
+{
+	device_initcall(sh7343se_devices_setup);
+}
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_7343se __initmv = {
+	.mv_name = "SolutionEngine 7343",
+	.mv_setup = sh7343se_setup,
+	.mv_nr_irqs = 108,
+	.mv_inb = sh7343se_inb,
+	.mv_inw = sh7343se_inw,
+	.mv_inl = sh7343se_inl,
+	.mv_outb = sh7343se_outb,
+	.mv_outw = sh7343se_outw,
+	.mv_outl = sh7343se_outl,
+
+	.mv_inb_p = sh7343se_inb_p,
+	.mv_inw_p = sh7343se_inw,
+	.mv_inl_p = sh7343se_inl,
+	.mv_outb_p = sh7343se_outb_p,
+	.mv_outw_p = sh7343se_outw,
+	.mv_outl_p = sh7343se_outl,
+
+	.mv_insb = sh7343se_insb,
+	.mv_insw = sh7343se_insw,
+	.mv_insl = sh7343se_insl,
+	.mv_outsb = sh7343se_outsb,
+	.mv_outsw = sh7343se_outsw,
+	.mv_outsl = sh7343se_outsl,
+
+	.mv_init_irq = init_7343se_IRQ,
+	.mv_irq_demux = shmse_irq_demux,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat = heartbeat_7343se,
+#endif
+};
+ALIAS_MV(7343se)
diff --git a/arch/sh/configs/se7343_defconfig b/arch/sh/configs/se7343_defconfig
new file mode 100644
index 000000000000..948e507b52be
--- /dev/null
+++ b/arch/sh/configs/se7343_defconfig
@@ -0,0 +1,997 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.17
+# Mon Aug  7 20:14:44 2006
+#
+CONFIG_SUPERH=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SWAP is not set
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_UID16=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SHMEM is not set
+CONFIG_SLAB=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+# CONFIG_SLOB is not set
+CONFIG_OBSOLETE_INTERMODULE=y
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+CONFIG_IOSCHED_DEADLINE=y
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+CONFIG_DEFAULT_DEADLINE=y
+# CONFIG_DEFAULT_CFQ is not set
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="deadline"
+
+#
+# System type
+#
+CONFIG_SOLUTION_ENGINE=y
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+CONFIG_SH_7343_SOLUTION_ENGINE=y
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_7710VOIPGW is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R7780RP is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_SHMIN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+CONFIG_CPU_SH4A=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7706 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+# CONFIG_CPU_SUBTYPE_SH7710 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+CONFIG_CPU_SUBTYPE_SH7343=y
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_PAGE_OFFSET=0x80000000
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x01000000
+CONFIG_32BIT=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+# CONFIG_SH_FPU is not set
+# CONFIG_SH_FPU_EMU is not set
+CONFIG_SH_DSP=y
+# CONFIG_SH_STORE_QUEUES is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_SR_RB=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ=27000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+# CONFIG_SH_DMA is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+CONFIG_HEARTBEAT=y
+
+#
+# Kernel features
+#
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+# CONFIG_KEXEC is not set
+# CONFIG_SMP is not set
+CONFIG_PREEMPT_NONE=y
+# CONFIG_PREEMPT_VOLUNTARY is not set
+# CONFIG_PREEMPT is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+# CONFIG_CMDLINE_BOOL is not set
+
+#
+# Bus options
+#
+# CONFIG_PCI is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options (EXPERIMENTAL)
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+CONFIG_MTD_CONCAT=y
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+CONFIG_MTD_CFI=y
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_OBSOLETE_CHIPS is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+# CONFIG_MTD_PHYSMAP is not set
+# CONFIG_MTD_SOLUTIONENGINE is not set
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+
+#
+# NAND Flash Device Drivers
+#
+# CONFIG_MTD_NAND is not set
+
+#
+# OneNAND Flash Device Drivers
+#
+# CONFIG_MTD_ONENAND is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_RAM is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+CONFIG_SMC91X=y
+# CONFIG_NE2000 is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_NR_UARTS=2
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+CONFIG_I2C=y
+CONFIG_I2C_CHARDEV=y
+
+#
+# I2C Algorithms
+#
+# CONFIG_I2C_ALGOBIT is not set
+# CONFIG_I2C_ALGOPCF is not set
+# CONFIG_I2C_ALGOPCA is not set
+
+#
+# I2C Hardware Bus support
+#
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_STUB is not set
+# CONFIG_I2C_PCA_ISA is not set
+CONFIG_I2C_SH7343=y
+
+#
+# Miscellaneous I2C Chip support
+#
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
+# CONFIG_SENSORS_EEPROM is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+CONFIG_VIDEO_DEV=y
+CONFIG_VIDEO_V4L1=y
+CONFIG_VIDEO_V4L1_COMPAT=y
+CONFIG_VIDEO_V4L2=y
+
+#
+# Video Capture Adapters
+#
+
+#
+# Video Capture Adapters
+#
+# CONFIG_VIDEO_ADV_DEBUG is not set
+# CONFIG_VIDEO_VIVI is not set
+# CONFIG_VIDEO_CPIA is not set
+# CONFIG_VIDEO_SAA5246A is not set
+# CONFIG_VIDEO_SAA5249 is not set
+# CONFIG_TUNER_3036 is not set
+# CONFIG_VIDEO_OVCAMCHIP is not set
+
+#
+# Encoders and Decoders
+#
+# CONFIG_VIDEO_MSP3400 is not set
+# CONFIG_VIDEO_CS53L32A is not set
+# CONFIG_VIDEO_WM8775 is not set
+# CONFIG_VIDEO_WM8739 is not set
+# CONFIG_VIDEO_CX25840 is not set
+# CONFIG_VIDEO_SAA711X is not set
+# CONFIG_VIDEO_SAA7127 is not set
+# CONFIG_VIDEO_UPD64031A is not set
+# CONFIG_VIDEO_UPD64083 is not set
+
+#
+# Radio Adapters
+#
+# CONFIG_RADIO_MAESTRO is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+CONFIG_FB=y
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_MACMODES is not set
+CONFIG_FB_FIRMWARE_EDID=y
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+# CONFIG_FB_EPSON1355 is not set
+# CONFIG_FB_S1D13XXX is not set
+# CONFIG_FB_VIRTUAL is not set
+
+#
+# Console display driver support
+#
+CONFIG_DUMMY_CONSOLE=y
+# CONFIG_FRAMEBUFFER_CONSOLE is not set
+
+#
+# Logo configuration
+#
+# CONFIG_LOGO is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Sound
+#
+CONFIG_SOUND=y
+
+#
+# Advanced Linux Sound Architecture
+#
+CONFIG_SND=y
+CONFIG_SND_TIMER=y
+CONFIG_SND_PCM=y
+CONFIG_SND_SEQUENCER=y
+# CONFIG_SND_SEQ_DUMMY is not set
+CONFIG_SND_OSSEMUL=y
+# CONFIG_SND_MIXER_OSS is not set
+CONFIG_SND_PCM_OSS=y
+CONFIG_SND_PCM_OSS_PLUGINS=y
+# CONFIG_SND_SEQUENCER_OSS is not set
+# CONFIG_SND_DYNAMIC_MINORS is not set
+CONFIG_SND_SUPPORT_OLD_API=y
+CONFIG_SND_VERBOSE_PROCFS=y
+# CONFIG_SND_VERBOSE_PRINTK is not set
+# CONFIG_SND_DEBUG is not set
+
+#
+# Generic devices
+#
+# CONFIG_SND_DUMMY is not set
+# CONFIG_SND_VIRMIDI is not set
+# CONFIG_SND_MTPAV is not set
+# CONFIG_SND_SERIAL_U16550 is not set
+# CONFIG_SND_MPU401 is not set
+
+#
+# SuperH devices
+#
+CONFIG_SH7343_SIU=m
+CONFIG_AK4537_CODEC=y
+
+#
+# Open Sound System
+#
+# CONFIG_SOUND_PRIME is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS_FS is not set
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_FS_DEBUG=0
+CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_SUMMARY is not set
+# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
+CONFIG_JFFS2_ZLIB=y
+CONFIG_JFFS2_RTIME=y
+# CONFIG_JFFS2_RUBIN is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+# CONFIG_NFSD_V3 is not set
+# CONFIG_NFSD_TCP is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
+CONFIG_ZLIB_DEFLATE=y
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 06df309f1a6d..ac57638977ee 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -8,6 +8,7 @@
 SE			SH_SOLUTION_ENGINE
 7751SE			SH_7751_SOLUTION_ENGINE		
 7300SE			SH_7300_SOLUTION_ENGINE
+7343SE			SH_7343_SOLUTION_ENGINE
 73180SE			SH_73180_SOLUTION_ENGINE
 7751SYSTEMH		SH_7751_SYSTEMH
 HP6XX			SH_HP6XX
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 00886f9adb4d..dd05e102fc0e 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -316,6 +316,8 @@
 # define OFFCHIP_NR_IRQS 4
 #elif defined(CONFIG_SH_R7780RP)
 # define OFFCHIP_NR_IRQS 16
+#elif defined(CONFIG_SH_7343_SOLUTION_ENGINE)
+# define OFFCHIP_NR_IRQS 12
 #elif defined(CONFIG_SH_UNKNOWN)
 # define OFFCHIP_NR_IRQS 16	/* Must also be last */
 #else
diff --git a/include/asm-sh/se7343.h b/include/asm-sh/se7343.h
new file mode 100644
index 000000000000..e7914a54aa96
--- /dev/null
+++ b/include/asm-sh/se7343.h
@@ -0,0 +1,82 @@
+#ifndef __ASM_SH_HITACHI_SE7343_H
+#define __ASM_SH_HITACHI_SE7343_H
+
+/*
+ * include/asm-sh/se/se7343.h
+ *
+ * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
+ *
+ * SH-Mobile SolutionEngine 7343 support
+ */
+
+/* Box specific addresses.  */
+
+/* Area 0 */
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
+#define PA_FROM		0x00400000	/* Flash ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
+#define PA_SRAM		0x00800000	/* SRAM */
+#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
+/* Area 1 */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+/* Area 2 */
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+/* Area 3 */
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+/* Area 4 */
+#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
+#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+#define PA_LED		0xb0C00000	/* LED */
+#define LED_SHIFT       0
+#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
+#define PA_CPLD_MODESET	0xb1400004	/* CPLD Mode set register */
+#define PA_CPLD_ST	0xb1400008	/* CPLD Interrupt status register */
+#define PA_CPLD_IMSK	0xb140000a	/* CPLD Interrupt mask register */
+/* Area 5 */
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+/* Area 6 */
+#define PA_LCD1		0xb8000000
+#define PA_LCD2		0xb8800000
+
+#define __IO_PREFIX	sh7343se
+#include <asm/io_generic.h>
+
+/* External Multiplexed interrupts */
+#define PC_IRQ0		OFFCHIP_IRQ_BASE
+#define PC_IRQ1		(PC_IRQ0 + 1)
+#define PC_IRQ2		(PC_IRQ1 + 1)
+#define PC_IRQ3		(PC_IRQ2 + 1)
+
+#define EXT_IRQ0	(PC_IRQ3 + 1)
+#define EXT_IRQ1	(EXT_IRQ0 + 1)
+#define EXT_IRQ2	(EXT_IRQ1 + 1)
+#define EXT_IRQ3	(EXT_IRQ2 + 1)
+
+#define USB_IRQ0	(EXT_IRQ3 + 1)
+#define USB_IRQ1	(USB_IRQ0 + 1)
+
+#define UART_IRQ0	(USB_IRQ1 + 1)
+#define UART_IRQ1	(UART_IRQ0 + 1)
+
+#endif  /* __ASM_SH_HITACHI_SE7343_H */
-- 
cgit v1.2.3


From 456335e2072fb35bf290b45e61d51916c322c145 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 27 Sep 2006 10:00:54 +0100
Subject: [ARM] Separate page table manipulation code from bootmem
 initialisation

nommu does not require the page table manipulation code in the
bootmem initialisation paths.  Move this into separate inline
functions.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c      | 73 +++++++++++++++++++++++++++++--------------------
 include/asm-arm/setup.h | 12 ++++----
 2 files changed, 51 insertions(+), 34 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 64262bda8e54..83145d1d3389 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -226,6 +226,44 @@ static __init void reserve_node_zero(pg_data_t *pgdat)
 		reserve_bootmem_node(pgdat, PHYS_OFFSET, res_size);
 }
 
+static inline void prepare_page_table(struct meminfo *mi)
+{
+	unsigned long addr;
+
+	/*
+	 * Clear out all the mappings below the kernel image.
+	 */
+	for (addr = 0; addr < MODULE_START; addr += PGDIR_SIZE)
+		pmd_clear(pmd_off_k(addr));
+
+#ifdef CONFIG_XIP_KERNEL
+	/* The XIP kernel is mapped in the module area -- skip over it */
+	addr = ((unsigned long)&_etext + PGDIR_SIZE - 1) & PGDIR_MASK;
+#endif
+	for ( ; addr < PAGE_OFFSET; addr += PGDIR_SIZE)
+		pmd_clear(pmd_off_k(addr));
+
+	/*
+	 * Clear out all the kernel space mappings, except for the first
+	 * memory bank, up to the end of the vmalloc region.
+	 */
+	for (addr = __phys_to_virt(mi->bank[0].start + mi->bank[0].size);
+	     addr < VMALLOC_END; addr += PGDIR_SIZE)
+		pmd_clear(pmd_off_k(addr));
+}
+
+static inline void map_memory_bank(struct membank *bank)
+{
+	struct map_desc map;
+
+	map.pfn = __phys_to_pfn(bank->start);
+	map.virtual = __phys_to_virt(bank->start);
+	map.length = bank->size;
+	map.type = MT_MEMORY;
+
+	create_mapping(&map);
+}
+
 static unsigned long __init
 bootmem_init_node(int node, int initrd_node, struct meminfo *mi)
 {
@@ -242,23 +280,18 @@ bootmem_init_node(int node, int initrd_node, struct meminfo *mi)
 	 * Calculate the pfn range, and map the memory banks for this node.
 	 */
 	for_each_nodebank(i, mi, node) {
+		struct membank *bank = &mi->bank[i];
 		unsigned long start, end;
-		struct map_desc map;
 
-		start = mi->bank[i].start >> PAGE_SHIFT;
-		end = (mi->bank[i].start + mi->bank[i].size) >> PAGE_SHIFT;
+		start = bank->start >> PAGE_SHIFT;
+		end = (bank->start + bank->size) >> PAGE_SHIFT;
 
 		if (start_pfn > start)
 			start_pfn = start;
 		if (end_pfn < end)
 			end_pfn = end;
 
-		map.pfn = __phys_to_pfn(mi->bank[i].start);
-		map.virtual = __phys_to_virt(mi->bank[i].start);
-		map.length = mi->bank[i].size;
-		map.type = MT_MEMORY;
-
-		create_mapping(&map);
+		map_memory_bank(bank);
 	}
 
 	/*
@@ -342,7 +375,7 @@ bootmem_init_node(int node, int initrd_node, struct meminfo *mi)
 
 static void __init bootmem_init(struct meminfo *mi)
 {
-	unsigned long addr, memend_pfn = 0;
+	unsigned long memend_pfn = 0;
 	int node, initrd_node, i;
 
 	/*
@@ -354,25 +387,7 @@ static void __init bootmem_init(struct meminfo *mi)
 
 	memcpy(&meminfo, mi, sizeof(meminfo));
 
-	/*
-	 * Clear out all the mappings below the kernel image.
-	 */
-	for (addr = 0; addr < MODULE_START; addr += PGDIR_SIZE)
-		pmd_clear(pmd_off_k(addr));
-#ifdef CONFIG_XIP_KERNEL
-	/* The XIP kernel is mapped in the module area -- skip over it */
-	addr = ((unsigned long)&_etext + PGDIR_SIZE - 1) & PGDIR_MASK;
-#endif
-	for ( ; addr < PAGE_OFFSET; addr += PGDIR_SIZE)
-		pmd_clear(pmd_off_k(addr));
-
-	/*
-	 * Clear out all the kernel space mappings, except for the first
-	 * memory bank, up to the end of the vmalloc region.
-	 */
-	for (addr = __phys_to_virt(mi->bank[0].start + mi->bank[0].size);
-	     addr < VMALLOC_END; addr += PGDIR_SIZE)
-		pmd_clear(pmd_off_k(addr));
+	prepare_page_table(mi);
 
 	/*
 	 * Locate which node contains the ramdisk image, if any.
diff --git a/include/asm-arm/setup.h b/include/asm-arm/setup.h
index ea3ed2465233..aa4b5782f0c9 100644
--- a/include/asm-arm/setup.h
+++ b/include/asm-arm/setup.h
@@ -194,13 +194,15 @@ static struct tagtable __tagtable_##fn __tag = { tag, fn }
 # define NR_BANKS 8
 #endif
 
+struct membank {
+	unsigned long start;
+	unsigned long size;
+	int           node;
+};
+
 struct meminfo {
 	int nr_banks;
-	struct {
-		unsigned long start;
-		unsigned long size;
-		int           node;
-	} bank[NR_BANKS];
+	struct membank bank[NR_BANKS];
 };
 
 /*
-- 
cgit v1.2.3


From 2c7834a6f15fe6c50ed4766f1bb6f9183b9e2740 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:17:31 +0900
Subject: sh: machvec rework.

Some more machvec overhauling and setup code cleanup. Kill off
get_system_type() and platform_setup(), we can do these both
through the machvec. While we're add it, kill off more useless
mach.c's and drop some legacy cruft from setup.c.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/bigsur/setup.c              |  47 +++++-------
 arch/sh/boards/dreamcast/setup.c           |  34 ++++-----
 arch/sh/boards/ec3104/setup.c              |  49 +++++--------
 arch/sh/boards/hp6xx/Makefile              |   4 +-
 arch/sh/boards/hp6xx/setup.c               |  47 ++++++++++--
 arch/sh/boards/landisk/setup.c             |  92 ++++++++++++-----------
 arch/sh/boards/mpc1211/setup.c             |  42 +++++------
 arch/sh/boards/renesas/edosk7705/setup.c   |  29 ++------
 arch/sh/boards/renesas/hs7751rvoip/setup.c |  96 +++++++++++-------------
 arch/sh/boards/renesas/r7780rp/setup.c     |  83 +++++++++++----------
 arch/sh/boards/renesas/rts7751r2d/Makefile |   4 +-
 arch/sh/boards/renesas/rts7751r2d/mach.c   |  66 -----------------
 arch/sh/boards/renesas/rts7751r2d/setup.c  |  62 ++++++++++++++--
 arch/sh/boards/renesas/systemh/setup.c     |  16 +---
 arch/sh/boards/saturn/setup.c              |  14 +---
 arch/sh/boards/se/7300/setup.c             |  17 +----
 arch/sh/boards/se/73180/setup.c            |  17 +----
 arch/sh/boards/se/770x/Makefile            |   3 +-
 arch/sh/boards/se/770x/mach.c              |  62 ----------------
 arch/sh/boards/se/770x/setup.c             |  61 ++++++++++++----
 arch/sh/boards/se/7751/Makefile            |   3 +-
 arch/sh/boards/se/7751/mach.c              |  48 ------------
 arch/sh/boards/se/7751/setup.c             | 103 ++++++++++----------------
 arch/sh/boards/sh03/setup.c                |  26 +++----
 arch/sh/boards/shmin/setup.c               |  10 +--
 arch/sh/boards/snapgear/setup.c            |  22 ++----
 arch/sh/boards/superh/microdev/setup.c     | 113 +++++++++++++----------------
 arch/sh/boards/titan/setup.c               |  13 +---
 arch/sh/boards/unknown/setup.c             |  13 +---
 arch/sh/kernel/setup.c                     |  54 +++-----------
 include/asm-sh/machvec.h                   |   7 +-
 31 files changed, 481 insertions(+), 776 deletions(-)
 delete mode 100644 arch/sh/boards/renesas/rts7751r2d/mach.c
 delete mode 100644 arch/sh/boards/se/770x/mach.c
 delete mode 100644 arch/sh/boards/se/7751/mach.c

(limited to 'include')

diff --git a/arch/sh/boards/bigsur/setup.c b/arch/sh/boards/bigsur/setup.c
index dfeede9da50f..9711c20fc9e4 100644
--- a/arch/sh/boards/bigsur/setup.c
+++ b/arch/sh/boards/bigsur/setup.c
@@ -41,31 +41,7 @@
 //		Big Sur Init Routines	
 /*===========================================================*/
 
-const char *get_system_type(void)
-{
-	return "Big Sur";
-}
-
-/*
- * The Machine Vector
- */
-extern void heartbeat_bigsur(void);
-extern void init_bigsur_IRQ(void);
-
-struct sh_machine_vector mv_bigsur __initmv = {
-	.mv_nr_irqs		= NR_IRQS,     // Defined in <asm/irq.h>
-
-	.mv_isa_port2addr	= bigsur_isa_port2addr,
-	.mv_irq_demux       	= bigsur_irq_demux,
-
-	.mv_init_irq		= init_bigsur_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_bigsur,
-#endif
-};
-ALIAS_MV(bigsur)
-
-int __init platform_setup(void)
+static void __init bigsur_setup(char **cmdline_p)
 {
 	/* Mask all 2nd level IRQ's */
 	outb(-1,BIGSUR_IMR0);
@@ -89,7 +65,24 @@ int __init platform_setup(void)
 	outw(1, BIGSUR_ETHR+0xe);
 	/* set the IO port to BIGSUR_ETHER_IOPORT */
 	outw(BIGSUR_ETHER_IOPORT<<3, BIGSUR_ETHR+0x2);
-
-	return 0;
 }
 
+/*
+ * The Machine Vector
+ */
+extern void heartbeat_bigsur(void);
+extern void init_bigsur_IRQ(void);
+
+struct sh_machine_vector mv_bigsur __initmv = {
+	.mv_name		= "Big Sur",
+	.mv_setup		= bigsur_setup,
+
+	.mv_isa_port2addr	= bigsur_isa_port2addr,
+	.mv_irq_demux       	= bigsur_irq_demux,
+
+	.mv_init_irq		= init_bigsur_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_bigsur,
+#endif
+};
+ALIAS_MV(bigsur)
diff --git a/arch/sh/boards/dreamcast/setup.c b/arch/sh/boards/dreamcast/setup.c
index a00ac94beb05..f13017eeeb27 100644
--- a/arch/sh/boards/dreamcast/setup.c
+++ b/arch/sh/boards/dreamcast/setup.c
@@ -22,7 +22,6 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/device.h>
-
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/rtc.h>
@@ -37,24 +36,7 @@ extern int systemasic_irq_demux(int);
 void *dreamcast_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
 int dreamcast_consistent_free(struct device *, size_t, void *, dma_addr_t);
 
-const char *get_system_type(void)
-{
-	return "Sega Dreamcast";
-}
-
-struct sh_machine_vector mv_dreamcast __initmv = {
-	.mv_nr_irqs		= NR_IRQS,
-
-	.mv_irq_demux		= systemasic_irq_demux,
-
-#ifdef CONFIG_PCI
-	.mv_consistent_alloc	= dreamcast_consistent_alloc,
-	.mv_consistent_free	= dreamcast_consistent_free,
-#endif
-};
-ALIAS_MV(dreamcast)
-
-int __init platform_setup(void)
+static void __init dreamcast_setup(char **cmdline_p)
 {
 	int i;
 
@@ -76,6 +58,16 @@ int __init platform_setup(void)
 	if (gapspci_init() < 0)
 		printk(KERN_WARNING "GAPSPCI was not detected.\n");
 #endif
-
-	return 0;
 }
+
+struct sh_machine_vector mv_dreamcast __initmv = {
+	.mv_name		= "Sega Dreamcast",
+	.mv_setup		= dreamcast_setup,
+	.mv_irq_demux		= systemasic_irq_demux,
+
+#ifdef CONFIG_PCI
+	.mv_consistent_alloc	= dreamcast_consistent_alloc,
+	.mv_consistent_free	= dreamcast_consistent_free,
+#endif
+};
+ALIAS_MV(dreamcast)
diff --git a/arch/sh/boards/ec3104/setup.c b/arch/sh/boards/ec3104/setup.c
index 4b3ef16a0e96..902bc975a13e 100644
--- a/arch/sh/boards/ec3104/setup.c
+++ b/arch/sh/boards/ec3104/setup.c
@@ -21,22 +21,36 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/types.h>
-
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/machvec.h>
 #include <asm/mach/ec3104.h>
 
-const char *get_system_type(void)
+static void __init ec3104_setup(char **cmdline_p)
 {
-	return "EC3104";
+	char str[8];
+	int i;
+
+	for (i=0; i<8; i++)
+		str[i] = ctrl_readb(EC3104_BASE + i);
+
+	for (i = EC3104_IRQBASE; i < EC3104_IRQBASE + 32; i++)
+		irq_desc[i].handler = &ec3104_int;
+
+	printk("initializing EC3104 \"%.8s\" at %08x, IRQ %d, IRQ base %d\n",
+	       str, EC3104_BASE, EC3104_IRQ, EC3104_IRQBASE);
+
+	/* mask all interrupts.  this should have been done by the boot
+	 * loader for us but we want to be sure ... */
+	ctrl_writel(0xffffffff, EC3104_IMR);
 }
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_ec3104 __initmv = {
+	.mv_name	= "EC3104",
+	.mv_setup	= ec3104_setup,
 	.mv_nr_irqs	= 96,
 
 	.mv_inb		= ec3104_inb,
@@ -48,31 +62,4 @@ struct sh_machine_vector mv_ec3104 __initmv = {
 
 	.mv_irq_demux	= ec3104_irq_demux,
 };
-
 ALIAS_MV(ec3104)
-
-int __init platform_setup(void)
-{
-	char str[8];
-	int i;
-	
-	if (0)
-		return 0;
-
-	for (i=0; i<8; i++)
-		str[i] = ctrl_readb(EC3104_BASE + i);
-
-	for (i = EC3104_IRQBASE; i < EC3104_IRQBASE + 32; i++)
-		irq_desc[i].chip = &ec3104_int;
-
-	printk("initializing EC3104 \"%.8s\" at %08x, IRQ %d, IRQ base %d\n",
-	       str, EC3104_BASE, EC3104_IRQ, EC3104_IRQBASE);
-
-
-	/* mask all interrupts.  this should have been done by the boot
-	 * loader for us but we want to be sure ... */
-	ctrl_writel(0xffffffff, EC3104_IMR);
-	
-	return 0;
-}
-
diff --git a/arch/sh/boards/hp6xx/Makefile b/arch/sh/boards/hp6xx/Makefile
index fdf98ffcf437..ff1b7f5b4e91 100644
--- a/arch/sh/boards/hp6xx/Makefile
+++ b/arch/sh/boards/hp6xx/Makefile
@@ -2,8 +2,6 @@
 # Makefile for the HP6xx specific parts of the kernel
 #
 
-obj-y	 		:= mach.o setup.o
+obj-y	 		:= setup.o
 obj-$(CONFIG_PM)	+= pm.o pm_wakeup.o
 obj-$(CONFIG_APM)	+= hp6xx_apm.o
-
-
diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c
index 629016bec809..60ab17ad6054 100644
--- a/arch/sh/boards/hp6xx/setup.c
+++ b/arch/sh/boards/hp6xx/setup.c
@@ -19,12 +19,7 @@
 #define	SCPCR	0xa4000116
 #define SCPDR	0xa4000136
 
-const char *get_system_type(void)
-{
-	return "HP6xx";
-}
-
-int __init platform_setup(void)
+static void __init hp6xx_setup(char **cmdline_p)
 {
 	u8 v8;
 	u16 v;
@@ -64,6 +59,42 @@ int __init platform_setup(void)
 	v &= ~SCPCR_TS_MASK;
 	v |= SCPCR_TS_ENABLE;
 	ctrl_outw(v, SCPCR);
-
-	return 0;
 }
+
+/*
+ * XXX: This is stupid, we should have a generic machine vector for the cchips
+ * and just wrap the platform setup code in to this, as it's the only thing
+ * that ends up being different.
+ */
+struct sh_machine_vector mv_hp6xx __initmv = {
+	.mv_name = "hp6xx",
+	.mv_setup = hp6xx_setup,
+	.mv_nr_irqs = HD64461_IRQBASE + HD64461_IRQ_NUM,
+
+	.mv_inb = hd64461_inb,
+	.mv_inw = hd64461_inw,
+	.mv_inl = hd64461_inl,
+	.mv_outb = hd64461_outb,
+	.mv_outw = hd64461_outw,
+	.mv_outl = hd64461_outl,
+
+	.mv_inb_p = hd64461_inb_p,
+	.mv_inw_p = hd64461_inw,
+	.mv_inl_p = hd64461_inl,
+	.mv_outb_p = hd64461_outb_p,
+	.mv_outw_p = hd64461_outw,
+	.mv_outl_p = hd64461_outl,
+
+	.mv_insb = hd64461_insb,
+	.mv_insw = hd64461_insw,
+	.mv_insl = hd64461_insl,
+	.mv_outsb = hd64461_outsb,
+	.mv_outsw = hd64461_outsw,
+	.mv_outsl = hd64461_outsl,
+
+	.mv_readw = hd64461_readw,
+	.mv_writew = hd64461_writew,
+
+	.mv_irq_demux = hd64461_irq_demux,
+};
+ALIAS_MV(hp6xx)
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
index 3a795cfb1eda..127b9e020e00 100644
--- a/arch/sh/boards/landisk/setup.c
+++ b/arch/sh/boards/landisk/setup.c
@@ -69,42 +69,6 @@ static void heartbeat_landisk(void)
 	landisk_buzzerparam >>= 1;
 }
 
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_landisk __initmv = {
-	.mv_nr_irqs = 72,
-	.mv_inb = landisk_inb,
-	.mv_inw = landisk_inw,
-	.mv_inl = landisk_inl,
-	.mv_outb = landisk_outb,
-	.mv_outw = landisk_outw,
-	.mv_outl = landisk_outl,
-	.mv_inb_p = landisk_inb_p,
-	.mv_inw_p = landisk_inw,
-	.mv_inl_p = landisk_inl,
-	.mv_outb_p = landisk_outb_p,
-	.mv_outw_p = landisk_outw,
-	.mv_outl_p = landisk_outl,
-	.mv_insb = landisk_insb,
-	.mv_insw = landisk_insw,
-	.mv_insl = landisk_insl,
-	.mv_outsb = landisk_outsb,
-	.mv_outsw = landisk_outsw,
-	.mv_outsl = landisk_outsl,
-        .mv_ioport_map = landisk_ioport_map,
-	.mv_init_irq = init_landisk_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat = heartbeat_landisk,
-#endif
-};
-ALIAS_MV(landisk)
-
-const char *get_system_type(void)
-{
-        return "LANDISK";
-}
-
 static void landisk_power_off(void)
 {
         ctrl_outb(0x01, PA_SHUTDOWN);
@@ -132,16 +96,6 @@ static void check_usl5p(void)
         }
 }
 
-void __init platform_setup(void)
-{
-        landisk_buzzerparam = 0;
-        check_usl5p();
-
-        printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
-        board_time_init = landisk_time_init;
-        pm_power_off = landisk_power_off;
-}
-
 void *area5_io_base;
 void *area6_io_base;
 
@@ -176,4 +130,48 @@ static int __init landisk_cf_init(void)
 	return 0;
 }
 
-__initcall(landisk_cf_init);
+static void __init landisk_setup(char **cmdline_p)
+{
+	device_initcall(landisk_cf_init);
+
+	landisk_buzzerparam = 0;
+	check_usl5p();
+
+	printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
+
+	board_time_init = landisk_time_init;
+	pm_power_off = landisk_power_off;
+}
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_landisk __initmv = {
+	.mv_name = "LANDISK",
+	.mv_setup = landisk_setup,
+	.mv_nr_irqs = 72,
+	.mv_inb = landisk_inb,
+	.mv_inw = landisk_inw,
+	.mv_inl = landisk_inl,
+	.mv_outb = landisk_outb,
+	.mv_outw = landisk_outw,
+	.mv_outl = landisk_outl,
+	.mv_inb_p = landisk_inb_p,
+	.mv_inw_p = landisk_inw,
+	.mv_inl_p = landisk_inl,
+	.mv_outb_p = landisk_outb_p,
+	.mv_outw_p = landisk_outw,
+	.mv_outl_p = landisk_outl,
+	.mv_insb = landisk_insb,
+	.mv_insw = landisk_insw,
+	.mv_insl = landisk_insl,
+	.mv_outsb = landisk_outsb,
+	.mv_outsw = landisk_outsw,
+	.mv_outsl = landisk_outsl,
+	.mv_ioport_map = landisk_ioport_map,
+	.mv_init_irq = init_landisk_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat = heartbeat_landisk,
+#endif
+};
+ALIAS_MV(landisk)
diff --git a/arch/sh/boards/mpc1211/setup.c b/arch/sh/boards/mpc1211/setup.c
index e9c8ff221dda..8eb5d4303972 100644
--- a/arch/sh/boards/mpc1211/setup.c
+++ b/arch/sh/boards/mpc1211/setup.c
@@ -10,14 +10,12 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/interrupt.h>
-
 #include <asm/io.h>
 #include <asm/machvec.h>
 #include <asm/mpc1211/mpc1211.h>
 #include <asm/mpc1211/pci.h>
 #include <asm/mpc1211/m1543c.h>
 
-
 /* ALI15X3 SMBus address offsets */
 #define SMBHSTSTS   (0 + 0x3100)
 #define SMBHSTCNT   (1 + 0x3100)
@@ -50,11 +48,6 @@
 #define ALI15X3_STS_TERM	0x80	/* terminated by abort */
 #define ALI15X3_STS_ERR		0xE0	/* all the bad error bits */
 
-const char *get_system_type(void)
-{
-	return "Interface MPC-1211(CTP/PCI/MPC-SH02)";
-}
-
 static void __init pci_write_config(unsigned long busNo,
 				    unsigned long devNo,
 				    unsigned long fncNo,
@@ -205,7 +198,7 @@ int mpc1211_irq_demux(int irq)
 	return irq;
 }
 
-void __init init_mpc1211_IRQ(void)
+static void __init init_mpc1211_IRQ(void)
 {
 	int i;
 	/*
@@ -289,26 +282,10 @@ static int put_smb_blk(unsigned char *p, int address, int command, int no)
 	return 0;
 }
 
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_mpc1211 __initmv = {
-	.mv_nr_irqs		= 48,
-	.mv_irq_demux		= mpc1211_irq_demux,
-	.mv_init_irq		= init_mpc1211_IRQ,
-
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_mpc1211,
-#endif
-};
-
-ALIAS_MV(mpc1211)
-
 /* arch/sh/boards/mpc1211/rtc.c */
 void mpc1211_time_init(void);
 
-int __init platform_setup(void)
+static void __init mpc1211_setup(char **cmdline_p)
 {
 	unsigned char spd_buf[128];
 
@@ -332,3 +309,18 @@ int __init platform_setup(void)
 	return 0;
 }
 
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_mpc1211 __initmv = {
+	.mv_name		= "Interface MPC-1211(CTP/PCI/MPC-SH02)",
+	.mv_setup		= mpc1211_setup,
+	.mv_nr_irqs		= 48,
+	.mv_irq_demux		= mpc1211_irq_demux,
+	.mv_init_irq		= init_mpc1211_IRQ,
+
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_mpc1211,
+#endif
+};
+ALIAS_MV(mpc1211)
diff --git a/arch/sh/boards/renesas/edosk7705/setup.c b/arch/sh/boards/renesas/edosk7705/setup.c
index ba143fa4afaa..ec5be0107719 100644
--- a/arch/sh/boards/renesas/edosk7705/setup.c
+++ b/arch/sh/boards/renesas/edosk7705/setup.c
@@ -8,19 +8,21 @@
  * Modified for edosk7705 development
  * board by S. Dunn, 2003.
  */
-
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
 #include <asm/edosk7705/io.h>
 
-static void init_edosk7705(void);
+static void __init sh_edosk7705_init_irq(void)
+{
+	/* This is the Ethernet interrupt */
+	make_imask_irq(0x09);
+}
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_edosk7705 __initmv = {
+	.mv_name		= "EDOSK7705",
 	.mv_nr_irqs		= 80,
 
 	.mv_inb			= sh_edosk7705_inb,
@@ -37,23 +39,6 @@ struct sh_machine_vector mv_edosk7705 __initmv = {
 	.mv_outsl		= sh_edosk7705_outsl,
 
 	.mv_isa_port2addr	= sh_edosk7705_isa_port2addr,
-	.mv_init_irq		= init_edosk7705,
+	.mv_init_irq		= sh_edosk7705_init_irq,
 };
 ALIAS_MV(edosk7705)
-
-static void __init init_edosk7705(void)
-{
-	/* This is the Ethernet interrupt */
-	make_imask_irq(0x09);
-}
-
-const char *get_system_type(void)
-{
-	return "EDOSK7705";
-}
-
-void __init platform_setup(void)
-{
-	/* Nothing .. */
-}
-
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index a2cbcc5d530f..0414c15c3458 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -1,15 +1,12 @@
 /*
- * linux/arch/sh/kernel/setup_hs7751rvoip.c
+ * Renesas Technology Sales HS7751RVoIP Support.
  *
  * Copyright (C) 2000  Kazumoto Kojima
  *
- * Renesas Technology Sales HS7751RVoIP Support.
- *
  * Modified for HS7751RVoIP by
  * Atom Create Engineering Co., Ltd. 2002.
  * Lineo uSolutions, Inc. 2003.
  */
-
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/mm.h>
@@ -23,8 +20,6 @@
 #include <asm/rtc.h>
 #include <asm/irq.h>
 
-unsigned int debug_counter;
-
 static void __init hs7751rvoip_init_irq(void)
 {
 #if defined(CONFIG_HS7751RVOIP_CODEC)
@@ -35,56 +30,11 @@ static void __init hs7751rvoip_init_irq(void)
 	init_hs7751rvoip_IRQ();
 }
 
-struct sh_machine_vector mv_hs7751rvoip __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= hs7751rvoip_inb,
-	.mv_inw			= hs7751rvoip_inw,
-	.mv_inl			= hs7751rvoip_inl,
-	.mv_outb		= hs7751rvoip_outb,
-	.mv_outw		= hs7751rvoip_outw,
-	.mv_outl		= hs7751rvoip_outl,
-
-	.mv_inb_p		= hs7751rvoip_inb_p,
-	.mv_inw_p		= hs7751rvoip_inw,
-	.mv_inl_p		= hs7751rvoip_inl,
-	.mv_outb_p		= hs7751rvoip_outb_p,
-	.mv_outw_p		= hs7751rvoip_outw,
-	.mv_outl_p		= hs7751rvoip_outl,
-
-	.mv_insb		= hs7751rvoip_insb,
-	.mv_insw		= hs7751rvoip_insw,
-	.mv_insl		= hs7751rvoip_insl,
-	.mv_outsb		= hs7751rvoip_outsb,
-	.mv_outsw		= hs7751rvoip_outsw,
-	.mv_outsl		= hs7751rvoip_outsl,
-
-	.mv_init_irq		= hs7751rvoip_init_irq,
-	.mv_ioport_map		= hs7751rvoip_ioport_map,
-};
-ALIAS_MV(hs7751rvoip)
-
-const char *get_system_type(void)
-{
-	return "HS7751RVoIP";
-}
-
 static void hs7751rvoip_power_off(void)
 {
 	ctrl_outw(ctrl_inw(PA_OUTPORTR) & 0xffdf, PA_OUTPORTR);
 }
 
-/*
- * Initialize the board
- */
-void __init platform_setup(void)
-{
-	printk(KERN_INFO "Renesas Technology Sales HS7751RVoIP-2 support.\n");
-	ctrl_outb(0xf0, PA_OUTPORTR);
-	pm_power_off = hs7751rvoip_power_off;
-	debug_counter = 0;
-}
-
 void *area5_io8_base;
 void *area6_io8_base;
 void *area5_io16_base;
@@ -127,4 +77,46 @@ static int __init hs7751rvoip_cf_init(void)
 	return 0;
 }
 
-__initcall(hs7751rvoip_cf_init);
+/*
+ * Initialize the board
+ */
+static void __init hs7751rvoip_setup(char **cmdline_p)
+{
+	device_initcall(hs7751rvoip_cf_init);
+
+	ctrl_outb(0xf0, PA_OUTPORTR);
+	pm_power_off = hs7751rvoip_power_off;
+
+	printk(KERN_INFO "Renesas Technology Sales HS7751RVoIP-2 support.\n");
+}
+
+struct sh_machine_vector mv_hs7751rvoip __initmv = {
+	.mv_name		= "HS7751RVoIP",
+	.mv_setup		= hs7751rvoip_setup,
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= hs7751rvoip_inb,
+	.mv_inw			= hs7751rvoip_inw,
+	.mv_inl			= hs7751rvoip_inl,
+	.mv_outb		= hs7751rvoip_outb,
+	.mv_outw		= hs7751rvoip_outw,
+	.mv_outl		= hs7751rvoip_outl,
+
+	.mv_inb_p		= hs7751rvoip_inb_p,
+	.mv_inw_p		= hs7751rvoip_inw,
+	.mv_inl_p		= hs7751rvoip_inl,
+	.mv_outb_p		= hs7751rvoip_outb_p,
+	.mv_outw_p		= hs7751rvoip_outw,
+	.mv_outl_p		= hs7751rvoip_outl,
+
+	.mv_insb		= hs7751rvoip_insb,
+	.mv_insw		= hs7751rvoip_insw,
+	.mv_insl		= hs7751rvoip_insl,
+	.mv_outsb		= hs7751rvoip_outsb,
+	.mv_outsw		= hs7751rvoip_outsw,
+	.mv_outsl		= hs7751rvoip_outsl,
+
+	.mv_init_irq		= hs7751rvoip_init_irq,
+	.mv_ioport_map		= hs7751rvoip_ioport_map,
+};
+ALIAS_MV(hs7751rvoip)
diff --git a/arch/sh/boards/renesas/r7780rp/setup.c b/arch/sh/boards/renesas/r7780rp/setup.c
index 0e4d5e1b541c..b941aa0aa34e 100644
--- a/arch/sh/boards/renesas/r7780rp/setup.c
+++ b/arch/sh/boards/renesas/r7780rp/setup.c
@@ -20,41 +20,6 @@
 extern void heartbeat_r7780rp(void);
 extern void init_r7780rp_IRQ(void);
 
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_r7780rp __initmv = {
-	.mv_nr_irqs		= 109,
-
-	.mv_inb			= r7780rp_inb,
-	.mv_inw			= r7780rp_inw,
-	.mv_inl			= r7780rp_inl,
-	.mv_outb		= r7780rp_outb,
-	.mv_outw		= r7780rp_outw,
-	.mv_outl		= r7780rp_outl,
-
-	.mv_inb_p		= r7780rp_inb_p,
-	.mv_inw_p		= r7780rp_inw,
-	.mv_inl_p		= r7780rp_inl,
-	.mv_outb_p		= r7780rp_outb_p,
-	.mv_outw_p		= r7780rp_outw,
-	.mv_outl_p		= r7780rp_outl,
-
-	.mv_insb		= r7780rp_insb,
-	.mv_insw		= r7780rp_insw,
-	.mv_insl		= r7780rp_insl,
-	.mv_outsb		= r7780rp_outsb,
-	.mv_outsw		= r7780rp_outsw,
-	.mv_outsl		= r7780rp_outsl,
-
-	.mv_ioport_map		= r7780rp_ioport_map,
-	.mv_init_irq		= init_r7780rp_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_r7780rp,
-#endif
-};
-ALIAS_MV(r7780rp)
-
 static struct resource m66596_usb_host_resources[] = {
 	[0] = {
 		.start	= 0xa4800000,
@@ -88,7 +53,6 @@ static int __init r7780rp_devices_setup(void)
 	return platform_add_devices(r7780rp_devices,
 				    ARRAY_SIZE(r7780rp_devices));
 }
-__initcall(r7780rp_devices_setup);
 
 /*
  * Platform specific clocks
@@ -117,11 +81,6 @@ static struct clk *r7780rp_clocks[] = {
 	&ivdr_clk,
 };
 
-const char *get_system_type(void)
-{
-	return "Highlander R7780RP-1";
-}
-
 static void r7780rp_power_off(void)
 {
 #ifdef CONFIG_SH_R7780MP
@@ -132,11 +91,13 @@ static void r7780rp_power_off(void)
 /*
  * Initialize the board
  */
-void __init platform_setup(void)
+static void __init r7780rp_setup(char **cmdline_p)
 {
 	u16 ver = ctrl_inw(PA_VERREG);
 	int i;
 
+	device_initcall(r7780rp_devices_setup);
+
 	printk(KERN_INFO "Renesas Solutions Highlander R7780RP-1 support.\n");
 
 	printk(KERN_INFO "Board version: %d (revision %d), "
@@ -162,3 +123,41 @@ void __init platform_setup(void)
 
 	pm_power_off = r7780rp_power_off;
 }
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_r7780rp __initmv = {
+	.mv_name		= "Highlander R7780RP-1",
+	.mv_setup		= r7780rp_setup,
+
+	.mv_nr_irqs		= 109,
+
+	.mv_inb			= r7780rp_inb,
+	.mv_inw			= r7780rp_inw,
+	.mv_inl			= r7780rp_inl,
+	.mv_outb		= r7780rp_outb,
+	.mv_outw		= r7780rp_outw,
+	.mv_outl		= r7780rp_outl,
+
+	.mv_inb_p		= r7780rp_inb_p,
+	.mv_inw_p		= r7780rp_inw,
+	.mv_inl_p		= r7780rp_inl,
+	.mv_outb_p		= r7780rp_outb_p,
+	.mv_outw_p		= r7780rp_outw,
+	.mv_outl_p		= r7780rp_outl,
+
+	.mv_insb		= r7780rp_insb,
+	.mv_insw		= r7780rp_insw,
+	.mv_insl		= r7780rp_insl,
+	.mv_outsb		= r7780rp_outsb,
+	.mv_outsw		= r7780rp_outsw,
+	.mv_outsl		= r7780rp_outsl,
+
+	.mv_ioport_map		= r7780rp_ioport_map,
+	.mv_init_irq		= init_r7780rp_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_r7780rp,
+#endif
+};
+ALIAS_MV(r7780rp)
diff --git a/arch/sh/boards/renesas/rts7751r2d/Makefile b/arch/sh/boards/renesas/rts7751r2d/Makefile
index eee1ed6f5727..686fc9ea5989 100644
--- a/arch/sh/boards/renesas/rts7751r2d/Makefile
+++ b/arch/sh/boards/renesas/rts7751r2d/Makefile
@@ -2,5 +2,5 @@
 # Makefile for the RTS7751R2D specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o led.o
-
+obj-y	 := setup.o io.o irq.o
+obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/renesas/rts7751r2d/mach.c b/arch/sh/boards/renesas/rts7751r2d/mach.c
deleted file mode 100644
index fe3e8735e9f8..000000000000
--- a/arch/sh/boards/renesas/rts7751r2d/mach.c
+++ /dev/null
@@ -1,66 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_rts7751r2d.c
- *
- * Minor tweak of mach_se.c file to reference rts7751r2d-specific items.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Renesas Technology sales RTS7751R2D
- */
-
-#include <linux/init.h>
-#include <linux/types.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/irq.h>
-#include <asm/mach/rts7751r2d.h>
-
-extern void heartbeat_rts7751r2d(void);
-extern void init_rts7751r2d_IRQ(void);
-extern int rts7751r2d_irq_demux(int irq);
-
-extern void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
-extern int voyagergx_consistent_free(struct device *, size_t, void *, dma_addr_t);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_rts7751r2d __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= rts7751r2d_inb,
-	.mv_inw			= rts7751r2d_inw,
-	.mv_inl			= rts7751r2d_inl,
-	.mv_outb		= rts7751r2d_outb,
-	.mv_outw		= rts7751r2d_outw,
-	.mv_outl		= rts7751r2d_outl,
-
-	.mv_inb_p		= rts7751r2d_inb_p,
-	.mv_inw_p		= rts7751r2d_inw,
-	.mv_inl_p		= rts7751r2d_inl,
-	.mv_outb_p		= rts7751r2d_outb_p,
-	.mv_outw_p		= rts7751r2d_outw,
-	.mv_outl_p		= rts7751r2d_outl,
-
-	.mv_insb		= rts7751r2d_insb,
-	.mv_insw		= rts7751r2d_insw,
-	.mv_insl		= rts7751r2d_insl,
-	.mv_outsb		= rts7751r2d_outsb,
-	.mv_outsw		= rts7751r2d_outsw,
-	.mv_outsl		= rts7751r2d_outsl,
-
-	.mv_init_irq		= init_rts7751r2d_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_rts7751r2d,
-#endif
-	.mv_irq_demux		= rts7751r2d_irq_demux,
-
-#ifdef CONFIG_USB_OHCI_HCD
-	.mv_consistent_alloc	= voyagergx_consistent_alloc,
-	.mv_consistent_free	= voyagergx_consistent_free,
-#endif
-};
-ALIAS_MV(rts7751r2d)
diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index bb56c018f2ee..20597a6e6702 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -13,9 +13,17 @@
 #include <linux/serial_8250.h>
 #include <linux/pm.h>
 #include <asm/io.h>
+#include <asm/machvec.h>
 #include <asm/mach/rts7751r2d.h>
 #include <asm/voyagergx.h>
 
+extern void heartbeat_rts7751r2d(void);
+extern void init_rts7751r2d_IRQ(void);
+extern int rts7751r2d_irq_demux(int irq);
+
+extern void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
+extern int voyagergx_consistent_free(struct device *, size_t, void *, dma_addr_t);
+
 static struct plat_serial8250_port uart_platform_data[] = {
 	{
 		.membase	= (void *)VOYAGER_UART_BASE,
@@ -70,12 +78,6 @@ static int __init rts7751r2d_devices_setup(void)
 	return platform_add_devices(rts7751r2d_devices,
 				    ARRAY_SIZE(rts7751r2d_devices));
 }
-__initcall(rts7751r2d_devices_setup);
-
-const char *get_system_type(void)
-{
-	return "RTS7751R2D";
-}
 
 static void rts7751r2d_power_off(void)
 {
@@ -85,12 +87,56 @@ static void rts7751r2d_power_off(void)
 /*
  * Initialize the board
  */
-void __init platform_setup(void)
+static void __init rts7751r2d_setup(char **cmdline_p)
 {
-	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
+	device_initcall(rts7751r2d_devices_setup);
+
 	ctrl_outw(0x0000, PA_OUTPORT);
 	pm_power_off = rts7751r2d_power_off;
 
 	voyagergx_serial_init();
 
+	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
 }
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_rts7751r2d __initmv = {
+	.mv_name		= "RTS7751R2D",
+	.mv_setup		= rts7751r2d_setup,
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= rts7751r2d_inb,
+	.mv_inw			= rts7751r2d_inw,
+	.mv_inl			= rts7751r2d_inl,
+	.mv_outb		= rts7751r2d_outb,
+	.mv_outw		= rts7751r2d_outw,
+	.mv_outl		= rts7751r2d_outl,
+
+	.mv_inb_p		= rts7751r2d_inb_p,
+	.mv_inw_p		= rts7751r2d_inw,
+	.mv_inl_p		= rts7751r2d_inl,
+	.mv_outb_p		= rts7751r2d_outb_p,
+	.mv_outw_p		= rts7751r2d_outw,
+	.mv_outl_p		= rts7751r2d_outl,
+
+	.mv_insb		= rts7751r2d_insb,
+	.mv_insw		= rts7751r2d_insw,
+	.mv_insl		= rts7751r2d_insl,
+	.mv_outsb		= rts7751r2d_outsb,
+	.mv_outsw		= rts7751r2d_outsw,
+	.mv_outsl		= rts7751r2d_outsl,
+
+	.mv_init_irq		= init_rts7751r2d_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_rts7751r2d,
+#endif
+	.mv_irq_demux		= rts7751r2d_irq_demux,
+
+#ifdef CONFIG_USB_SM501
+	.mv_consistent_alloc	= voyagergx_consistent_alloc,
+	.mv_consistent_free	= voyagergx_consistent_free,
+#endif
+};
+ALIAS_MV(rts7751r2d)
diff --git a/arch/sh/boards/renesas/systemh/setup.c b/arch/sh/boards/renesas/systemh/setup.c
index 433c7c3f35e2..a8467bf90c25 100644
--- a/arch/sh/boards/renesas/systemh/setup.c
+++ b/arch/sh/boards/renesas/systemh/setup.c
@@ -20,20 +20,16 @@
 
 extern void make_systemh_irq(unsigned int irq);
 
-const char *get_system_type(void)
-{
-	return "7751 SystemH";
-}
-
 /*
  * Initialize IRQ setting
  */
-void __init init_7751systemh_IRQ(void)
+static void __init sh7751systemh_init_irq(void)
 {
 	make_systemh_irq(0xb);	/* Ethernet interrupt */
 }
 
 struct sh_machine_vector mv_7751systemh __initmv = {
+	.mv_name		= "7751 SystemH",
 	.mv_nr_irqs		= 72,
 
 	.mv_inb			= sh7751systemh_inb,
@@ -57,12 +53,6 @@ struct sh_machine_vector mv_7751systemh __initmv = {
 	.mv_outsw		= sh7751systemh_outsw,
 	.mv_outsl		= sh7751systemh_outsl,
 
-	.mv_init_irq		= init_7751systemh_IRQ,
+	.mv_init_irq		= sh7751system_init_irq,
 };
 ALIAS_MV(7751systemh)
-
-int __init platform_setup(void)
-{
-	return 0;
-}
-
diff --git a/arch/sh/boards/saturn/setup.c b/arch/sh/boards/saturn/setup.c
index bea6c572ad82..a3a37c9aad2e 100644
--- a/arch/sh/boards/saturn/setup.c
+++ b/arch/sh/boards/saturn/setup.c
@@ -9,22 +9,17 @@
  */
 #include <linux/kernel.h>
 #include <linux/init.h>
-
 #include <asm/io.h>
 #include <asm/machvec.h>
 #include <asm/mach/io.h>
 
 extern int saturn_irq_demux(int irq_nr);
 
-const char *get_system_type(void)
-{
-	return "Sega Saturn";
-}
-
 /*
  * The Machine Vector
  */
 struct sh_machine_vector mv_saturn __initmv = {
+	.mv_name		= "Sega Saturn",
 	.mv_nr_irqs		= 80,	/* Fix this later */
 
 	.mv_isa_port2addr	= saturn_isa_port2addr,
@@ -33,11 +28,4 @@ struct sh_machine_vector mv_saturn __initmv = {
 	.mv_ioremap		= saturn_ioremap,
 	.mv_iounmap		= saturn_iounmap,
 };
-
 ALIAS_MV(saturn)
-
-int __init platform_setup(void)
-{
-	return 0;
-}
-
diff --git a/arch/sh/boards/se/7300/setup.c b/arch/sh/boards/se/7300/setup.c
index bb7e1a189be8..6f082a722d42 100644
--- a/arch/sh/boards/se/7300/setup.c
+++ b/arch/sh/boards/se/7300/setup.c
@@ -14,17 +14,11 @@
 void heartbeat_7300se(void);
 void init_7300se_IRQ(void);
 
-const char *
-get_system_type(void)
-{
-	return "SolutionEngine 7300";
-}
-
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_7300se __initmv = {
+	.mv_name = "SolutionEngine 7300",
 	.mv_nr_irqs = 109,
 	.mv_inb = sh7300se_inb,
 	.mv_inw = sh7300se_inw,
@@ -52,13 +46,4 @@ struct sh_machine_vector mv_7300se __initmv = {
 	.mv_heartbeat = heartbeat_7300se,
 #endif
 };
-
 ALIAS_MV(7300se)
-/*
- * Initialize the board
- */
-void __init
-platform_setup(void)
-{
-
-}
diff --git a/arch/sh/boards/se/73180/setup.c b/arch/sh/boards/se/73180/setup.c
index 4daf53b1457f..b38ef50a160a 100644
--- a/arch/sh/boards/se/73180/setup.c
+++ b/arch/sh/boards/se/73180/setup.c
@@ -17,17 +17,11 @@
 void heartbeat_73180se(void);
 void init_73180se_IRQ(void);
 
-const char *
-get_system_type(void)
-{
-	return "SolutionEngine 73180";
-}
-
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_73180se __initmv = {
+	.mv_name = "SolutionEngine 73180",
 	.mv_nr_irqs = 108,
 	.mv_inb = sh73180se_inb,
 	.mv_inw = sh73180se_inw,
@@ -56,13 +50,4 @@ struct sh_machine_vector mv_73180se __initmv = {
 	.mv_heartbeat = heartbeat_73180se,
 #endif
 };
-
 ALIAS_MV(73180se)
-/*
- * Initialize the board
- */
-void __init
-platform_setup(void)
-{
-
-}
diff --git a/arch/sh/boards/se/770x/Makefile b/arch/sh/boards/se/770x/Makefile
index c96ed6933fe1..9a5035f80ec0 100644
--- a/arch/sh/boards/se/770x/Makefile
+++ b/arch/sh/boards/se/770x/Makefile
@@ -2,6 +2,5 @@
 # Makefile for the 770x SolutionEngine specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o
-
+obj-y	 := setup.o io.o irq.o
 obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/se/770x/mach.c b/arch/sh/boards/se/770x/mach.c
deleted file mode 100644
index e8968b71c353..000000000000
--- a/arch/sh/boards/se/770x/mach.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_se.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Hitachi SolutionEngine
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/se.h>
-
-void heartbeat_se(void);
-void init_se_IRQ(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_se __initmv = {
-#if defined(CONFIG_CPU_SH4)
-	.mv_nr_irqs		= 48,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7708)
-	.mv_nr_irqs		= 32,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7709)
-	.mv_nr_irqs		= 61,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7705)
-	.mv_nr_irqs		= 86,
-#endif
-
-	.mv_inb			= se_inb,
-	.mv_inw			= se_inw,
-	.mv_inl			= se_inl,
-	.mv_outb		= se_outb,
-	.mv_outw		= se_outw,
-	.mv_outl		= se_outl,
-
-	.mv_inb_p		= se_inb_p,
-	.mv_inw_p		= se_inw,
-	.mv_inl_p		= se_inl,
-	.mv_outb_p		= se_outb_p,
-	.mv_outw_p		= se_outw,
-	.mv_outl_p		= se_outl,
-
-	.mv_insb		= se_insb,
-	.mv_insw		= se_insw,
-	.mv_insl		= se_insl,
-	.mv_outsb		= se_outsb,
-	.mv_outsw		= se_outsw,
-	.mv_outsl		= se_outsl,
-
-	.mv_init_irq		= init_se_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_se,
-#endif
-};
-ALIAS_MV(se)
diff --git a/arch/sh/boards/se/770x/setup.c b/arch/sh/boards/se/770x/setup.c
index c9f75272e751..f3f82b7c8217 100644
--- a/arch/sh/boards/se/770x/setup.c
+++ b/arch/sh/boards/se/770x/setup.c
@@ -7,15 +7,17 @@
  * Hitachi SolutionEngine Support.
  *
  */
-
 #include <linux/init.h>
 #include <linux/irq.h>
-
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
 #include <asm/se.h>
 #include <asm/smc37c93x.h>
+#include <asm/machvec.h>
+
+void heartbeat_se(void);
+void init_se_IRQ(void);
 
 /*
  * Configure the Super I/O chip
@@ -26,7 +28,8 @@ static void __init smsc_config(int index, int data)
 	outb_p(data, DATA_PORT);
 }
 
-static void __init init_smsc(void)
+/* XXX: Another candidate for a more generic cchip machine vector */
+static void __init smsc_setup(char **cmdline_p)
 {
 	outb_p(CONFIG_ENTER, CONFIG_PORT);
 	outb_p(CONFIG_ENTER, CONFIG_PORT);
@@ -69,16 +72,46 @@ static void __init init_smsc(void)
 	outb_p(CONFIG_EXIT, CONFIG_PORT);
 }
 
-const char *get_system_type(void)
-{
-	return "SolutionEngine";
-}
-
 /*
- * Initialize the board
+ * The Machine Vector
  */
-void __init platform_setup(void)
-{
-	init_smsc();
-	/* XXX: RTC setting comes here */
-}
+struct sh_machine_vector mv_se __initmv = {
+	.mv_name		= "SolutionEngine",
+	.mv_setup		= smsc_setup,
+#if defined(CONFIG_CPU_SH4)
+	.mv_nr_irqs		= 48,
+#elif defined(CONFIG_CPU_SUBTYPE_SH7708)
+	.mv_nr_irqs		= 32,
+#elif defined(CONFIG_CPU_SUBTYPE_SH7709)
+	.mv_nr_irqs		= 61,
+#elif defined(CONFIG_CPU_SUBTYPE_SH7705)
+	.mv_nr_irqs		= 86,
+#endif
+
+	.mv_inb			= se_inb,
+	.mv_inw			= se_inw,
+	.mv_inl			= se_inl,
+	.mv_outb		= se_outb,
+	.mv_outw		= se_outw,
+	.mv_outl		= se_outl,
+
+	.mv_inb_p		= se_inb_p,
+	.mv_inw_p		= se_inw,
+	.mv_inl_p		= se_inl,
+	.mv_outb_p		= se_outb_p,
+	.mv_outw_p		= se_outw,
+	.mv_outl_p		= se_outl,
+
+	.mv_insb		= se_insb,
+	.mv_insw		= se_insw,
+	.mv_insl		= se_insl,
+	.mv_outsb		= se_outsb,
+	.mv_outsw		= se_outsw,
+	.mv_outsl		= se_outsl,
+
+	.mv_init_irq		= init_se_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_se,
+#endif
+};
+ALIAS_MV(se)
diff --git a/arch/sh/boards/se/7751/Makefile b/arch/sh/boards/se/7751/Makefile
index 07bf17a4a19b..188900c48321 100644
--- a/arch/sh/boards/se/7751/Makefile
+++ b/arch/sh/boards/se/7751/Makefile
@@ -2,8 +2,7 @@
 # Makefile for the 7751 SolutionEngine specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o
+obj-y	 := setup.o io.o irq.o
 
 obj-$(CONFIG_PCI) += pci.o
 obj-$(CONFIG_HEARTBEAT) += led.o
-
diff --git a/arch/sh/boards/se/7751/mach.c b/arch/sh/boards/se/7751/mach.c
deleted file mode 100644
index 1bb9047d863b..000000000000
--- a/arch/sh/boards/se/7751/mach.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_7751se.c
- *
- * Minor tweak of mach_se.c file to reference 7751se-specific items.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Hitachi 7751 SolutionEngine
- */
-
-#include <linux/init.h>
-#include <asm/machvec.h>
-#include <asm/se7751.h>
-
-void heartbeat_7751se(void);
-void init_7751se_IRQ(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_7751se __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= sh7751se_inb,
-	.mv_inw			= sh7751se_inw,
-	.mv_inl			= sh7751se_inl,
-	.mv_outb		= sh7751se_outb,
-	.mv_outw		= sh7751se_outw,
-	.mv_outl		= sh7751se_outl,
-
-	.mv_inb_p		= sh7751se_inb_p,
-	.mv_inw_p		= sh7751se_inw,
-	.mv_inl_p		= sh7751se_inl,
-	.mv_outb_p		= sh7751se_outb_p,
-	.mv_outw_p		= sh7751se_outw,
-	.mv_outl_p		= sh7751se_outl,
-
-	.mv_insl		= sh7751se_insl,
-	.mv_outsl		= sh7751se_outsl,
-
-	.mv_init_irq		= init_7751se_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_7751se,
-#endif
-};
-ALIAS_MV(7751se)
diff --git a/arch/sh/boards/se/7751/setup.c b/arch/sh/boards/se/7751/setup.c
index 8b693105893c..73e826310ba8 100644
--- a/arch/sh/boards/se/7751/setup.c
+++ b/arch/sh/boards/se/7751/setup.c
@@ -15,72 +15,11 @@
 #include <asm/io.h>
 #include <asm/se7751.h>
 
-#ifdef CONFIG_SH_KGDB
-#include <asm/kgdb.h>
-#endif
-
-/*
- * Configure the Super I/O chip
- */
-#if 0
-/* Leftover code from regular Solution Engine, for reference. */
-/* The SH7751 Solution Engine has a different SuperIO. */
-static void __init smsc_config(int index, int data)
-{
-	outb_p(index, INDEX_PORT);
-	outb_p(data, DATA_PORT);
-}
-
-static void __init init_smsc(void)
-{
-	outb_p(CONFIG_ENTER, CONFIG_PORT);
-	outb_p(CONFIG_ENTER, CONFIG_PORT);
-
-	/* FDC */
-	smsc_config(CURRENT_LDN_INDEX, LDN_FDC);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IRQ_SELECT_INDEX, 6); /* IRQ6 */
-
-	/* IDE1 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_IDE1);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IRQ_SELECT_INDEX, 14); /* IRQ14 */
-
-	/* AUXIO (GPIO): to use IDE1 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_AUXIO);
-	smsc_config(GPIO46_INDEX, 0x00); /* nIOROP */
-	smsc_config(GPIO47_INDEX, 0x00); /* nIOWOP */
-
-	/* COM1 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_COM1);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IO_BASE_HI_INDEX, 0x03);
-	smsc_config(IO_BASE_LO_INDEX, 0xf8);
-	smsc_config(IRQ_SELECT_INDEX, 4); /* IRQ4 */
-
-	/* COM2 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_COM2);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IO_BASE_HI_INDEX, 0x02);
-	smsc_config(IO_BASE_LO_INDEX, 0xf8);
-	smsc_config(IRQ_SELECT_INDEX, 3); /* IRQ3 */
-
-	/* RTC */
-	smsc_config(CURRENT_LDN_INDEX, LDN_RTC);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IRQ_SELECT_INDEX, 8); /* IRQ8 */
-
-	/* XXX: PARPORT, KBD, and MOUSE will come here... */
-	outb_p(CONFIG_EXIT, CONFIG_PORT);
-}
-#endif
-
-const char *get_system_type(void)
-{
-	return "7751 SolutionEngine";
-}
+void heartbeat_7751se(void);
+void init_7751se_IRQ(void);
 
 #ifdef CONFIG_SH_KGDB
+#include <asm/kgdb.h>
 static int kgdb_uart_setup(void);
 static struct kgdb_sermap kgdb_uart_sermap = 
 { "ttyS", 0, kgdb_uart_setup, NULL };
@@ -89,7 +28,7 @@ static struct kgdb_sermap kgdb_uart_sermap =
 /*
  * Initialize the board
  */
-void __init platform_setup(void)
+static void __init sh7751se_setup(char **cmdline_p)
 {
 	/* Call init_smsc() replacement to set up SuperIO. */
 	/* XXX: RTC setting comes here */
@@ -223,3 +162,37 @@ static int kgdb_uart_setup(void)
 	return 0;
 }
 #endif /* CONFIG_SH_KGDB */
+
+
+/*
+ * The Machine Vector
+ */
+
+struct sh_machine_vector mv_7751se __initmv = {
+	.mv_name		= "7751 SolutionEngine",
+	.mv_setup		= sh7751se_setup,
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= sh7751se_inb,
+	.mv_inw			= sh7751se_inw,
+	.mv_inl			= sh7751se_inl,
+	.mv_outb		= sh7751se_outb,
+	.mv_outw		= sh7751se_outw,
+	.mv_outl		= sh7751se_outl,
+
+	.mv_inb_p		= sh7751se_inb_p,
+	.mv_inw_p		= sh7751se_inw,
+	.mv_inl_p		= sh7751se_inl,
+	.mv_outb_p		= sh7751se_outb_p,
+	.mv_outw_p		= sh7751se_outw,
+	.mv_outl_p		= sh7751se_outl,
+
+	.mv_insl		= sh7751se_insl,
+	.mv_outsl		= sh7751se_outsl,
+
+	.mv_init_irq		= init_7751se_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_7751se,
+#endif
+};
+ALIAS_MV(7751se)
diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c
index 7d31d6aed8a5..6c310587ddfe 100644
--- a/arch/sh/boards/sh03/setup.c
+++ b/arch/sh/boards/sh03/setup.c
@@ -13,12 +13,7 @@
 #include <asm/sh03/sh03.h>
 #include <asm/addrspace.h>
 
-const char *get_system_type(void)
-{
-	return "Interface (CTP/PCI-SH03)";
-}
-
-static void init_sh03_IRQ(void)
+static void __init init_sh03_IRQ(void)
 {
 	ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR);
 
@@ -41,7 +36,17 @@ static void __iomem *sh03_ioport_map(unsigned long port, unsigned int size)
         return (void __iomem *)(port + PCI_IO_BASE);
 }
 
+/* arch/sh/boards/sh03/rtc.c */
+void sh03_time_init(void);
+
+static void __init sh03_setup(char **cmdline_p)
+{
+	board_time_init = sh03_time_init;
+}
+
 struct sh_machine_vector mv_sh03 __initmv = {
+	.mv_name		= "Interface (CTP/PCI-SH03)",
+	.mv_setup		= sh03_setup,
 	.mv_nr_irqs		= 48,
 	.mv_ioport_map		= sh03_ioport_map,
 	.mv_init_irq		= init_sh03_IRQ,
@@ -51,12 +56,3 @@ struct sh_machine_vector mv_sh03 __initmv = {
 #endif
 };
 ALIAS_MV(sh03)
-
-/* arch/sh/boards/sh03/rtc.c */
-void sh03_time_init(void);
-
-int __init platform_setup(void)
-{
-	board_time_init = sh03_time_init;
-	return 0;
-}
diff --git a/arch/sh/boards/shmin/setup.c b/arch/sh/boards/shmin/setup.c
index a950ec4d765d..2f0c19706cf9 100644
--- a/arch/sh/boards/shmin/setup.c
+++ b/arch/sh/boards/shmin/setup.c
@@ -14,21 +14,12 @@
 
 #define PFC_PHCR	0xa400010e
 
-const char *get_system_type(void)
-{
-	return "SHMIN";
-}
-
 static void __init init_shmin_irq(void)
 {
 	ctrl_outw(0x2a00, PFC_PHCR);	// IRQ0-3=IRQ
 	ctrl_outw(0x0aaa, INTC_ICR1);	// IRQ0-3=IRQ-mode,Low-active.
 }
 
-void __init platform_setup(void)
-{
-}
-
 static void __iomem *shmin_ioport_map(unsigned long port, unsigned int size)
 {
 	static int dummy;
@@ -43,6 +34,7 @@ static void __iomem *shmin_ioport_map(unsigned long port, unsigned int size)
 }
 
 struct sh_machine_vector mv_shmin __initmv = {
+	.mv_name	= "SHMIN",
 	.mv_init_irq	= init_shmin_irq,
 	.mv_ioport_map	= shmin_ioport_map,
 };
diff --git a/arch/sh/boards/snapgear/setup.c b/arch/sh/boards/snapgear/setup.c
index 66ce32f8b13c..f5e98c56b530 100644
--- a/arch/sh/boards/snapgear/setup.c
+++ b/arch/sh/boards/snapgear/setup.c
@@ -81,16 +81,20 @@ static void __init init_snapgear_IRQ(void)
 	make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY);
 }
 
-const char *get_system_type(void)
+/*
+ * Initialize the board
+ */
+static void __init snapgear_setup(char **cmdline_p)
 {
-	return "SnapGear SecureEdge5410";
+	board_time_init = secureedge5410_rtc_init;
 }
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_snapgear __initmv = {
+	.mv_name		= "SnapGear SecureEdge5410",
+	.mv_setup		= snapgear_setup,
 	.mv_nr_irqs		= 72,
 
 	.mv_inb			= snapgear_inb,
@@ -110,15 +114,3 @@ struct sh_machine_vector mv_snapgear __initmv = {
 	.mv_init_irq		= init_snapgear_IRQ,
 };
 ALIAS_MV(snapgear)
-
-/*
- * Initialize the board
- */
-
-int __init platform_setup(void)
-{
-	board_time_init = secureedge5410_rtc_init;
-
-	return 0;
-}
-
diff --git a/arch/sh/boards/superh/microdev/setup.c b/arch/sh/boards/superh/microdev/setup.c
index 61b402a3f5d7..031c814e6e76 100644
--- a/arch/sh/boards/superh/microdev/setup.c
+++ b/arch/sh/boards/superh/microdev/setup.c
@@ -10,7 +10,6 @@
  * May be copied or modified under the terms of the GNU General Public
  * License.  See linux/COPYING for more information.
  */
-
 #include <linux/init.h>
 #include <linux/platform_device.h>
 #include <linux/ioport.h>
@@ -21,41 +20,6 @@
 
 extern void microdev_heartbeat(void);
 
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_sh4202_microdev __initmv = {
-	.mv_nr_irqs		= 72,		/* QQQ need to check this - use the MACRO */
-
-	.mv_inb			= microdev_inb,
-	.mv_inw			= microdev_inw,
-	.mv_inl			= microdev_inl,
-	.mv_outb		= microdev_outb,
-	.mv_outw		= microdev_outw,
-	.mv_outl		= microdev_outl,
-
-	.mv_inb_p		= microdev_inb_p,
-	.mv_inw_p		= microdev_inw_p,
-	.mv_inl_p		= microdev_inl_p,
-	.mv_outb_p		= microdev_outb_p,
-	.mv_outw_p		= microdev_outw_p,
-	.mv_outl_p		= microdev_outl_p,
-
-	.mv_insb		= microdev_insb,
-	.mv_insw		= microdev_insw,
-	.mv_insl		= microdev_insl,
-	.mv_outsb		= microdev_outsb,
-	.mv_outsw		= microdev_outsw,
-	.mv_outsl		= microdev_outsl,
-
-	.mv_init_irq		= init_microdev_irq,
-
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= microdev_heartbeat,
-#endif
-};
-ALIAS_MV(sh4202_microdev)
 
 /****************************************************************************/
 
@@ -113,11 +77,6 @@ ALIAS_MV(sh4202_microdev)
 	/* assume a Keyboard Controller is present */
 int microdev_kbd_controller_present = 1;
 
-const char *get_system_type(void)
-{
-	return "SH4-202 MicroDev";
-}
-
 static struct resource smc91x_resources[] = {
 	[0] = {
 		.start		= 0x300,
@@ -291,25 +250,9 @@ static int __init microdev_devices_setup(void)
 	return platform_add_devices(microdev_devices, ARRAY_SIZE(microdev_devices));
 }
 
-__initcall(microdev_devices_setup);
-
-void __init platform_setup(void)
-{
-	int * const fpgaRevisionRegister = (int*)(MICRODEV_FPGA_GP_BASE + 0x8ul);
-	const int fpgaRevision = *fpgaRevisionRegister;
-	int * const CacheControlRegister = (int*)CCR;
-
-	printk("SuperH %s board (FPGA rev: 0x%0x, CCR: 0x%0x)\n",
-		get_system_type(), fpgaRevision, *CacheControlRegister);
-}
-
-
-/****************************************************************************/
-
-
-	/*
-	 * Setup for the SMSC FDC37C93xAPM
-	 */
+/*
+ * Setup for the SMSC FDC37C93xAPM
+ */
 static int __init smsc_superio_setup(void)
 {
 
@@ -412,8 +355,52 @@ static int __init smsc_superio_setup(void)
 	return 0;
 }
 
+static void __init microdev_setup(char **cmdline_p)
+{
+	int * const fpgaRevisionRegister = (int*)(MICRODEV_FPGA_GP_BASE + 0x8ul);
+	const int fpgaRevision = *fpgaRevisionRegister;
+	int * const CacheControlRegister = (int*)CCR;
+
+	device_initcall(microdev_devices_setup);
+	device_initcall(smsc_superio_setup);
 
-/* This is grotty, but, because kernel is always referenced on the link line
- * before any devices, this is safe.
+	printk("SuperH %s board (FPGA rev: 0x%0x, CCR: 0x%0x)\n",
+		get_system_type(), fpgaRevision, *CacheControlRegister);
+}
+
+/*
+ * The Machine Vector
  */
-__initcall(smsc_superio_setup);
+struct sh_machine_vector mv_sh4202_microdev __initmv = {
+	.mv_name		= "SH4-202 MicroDev",
+	.mv_setup		= microdev_setup,
+	.mv_nr_irqs		= 72,		/* QQQ need to check this - use the MACRO */
+
+	.mv_inb			= microdev_inb,
+	.mv_inw			= microdev_inw,
+	.mv_inl			= microdev_inl,
+	.mv_outb		= microdev_outb,
+	.mv_outw		= microdev_outw,
+	.mv_outl		= microdev_outl,
+
+	.mv_inb_p		= microdev_inb_p,
+	.mv_inw_p		= microdev_inw_p,
+	.mv_inl_p		= microdev_inl_p,
+	.mv_outb_p		= microdev_outb_p,
+	.mv_outw_p		= microdev_outw_p,
+	.mv_outl_p		= microdev_outl_p,
+
+	.mv_insb		= microdev_insb,
+	.mv_insw		= microdev_insw,
+	.mv_insl		= microdev_insl,
+	.mv_outsb		= microdev_outsb,
+	.mv_outsw		= microdev_outsw,
+	.mv_outsl		= microdev_outsl,
+
+	.mv_init_irq		= init_microdev_irq,
+
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= microdev_heartbeat,
+#endif
+};
+ALIAS_MV(sh4202_microdev)
diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c
index c8b431c1d0fd..52b66d8b8d2a 100644
--- a/arch/sh/boards/titan/setup.c
+++ b/arch/sh/boards/titan/setup.c
@@ -20,19 +20,8 @@ static void __init init_titan_irq(void)
 	make_ipr_irq( TITAN_IRQ_USB,   IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); /* PCIRQ3 */
 }
 
-const char *get_system_type(void)
-{
-	return "Titan";
-}
-
-int __init platform_setup(void)
-{
-	printk("%s Platform Setup\n", get_system_type());
-	return 0;
-}
-
 struct sh_machine_vector mv_titan __initmv = {
-	.mv_nr_irqs =	NR_IRQS,
+	.mv_name =	"Titan",
 
 	.mv_inb =	titan_inb,
 	.mv_inw =	titan_inw,
diff --git a/arch/sh/boards/unknown/setup.c b/arch/sh/boards/unknown/setup.c
index c5e4ed10876b..1c941370a2e3 100644
--- a/arch/sh/boards/unknown/setup.c
+++ b/arch/sh/boards/unknown/setup.c
@@ -14,19 +14,8 @@
  */
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/irq.h>
 
 struct sh_machine_vector mv_unknown __initmv = {
-	.mv_nr_irqs		= NR_IRQS,
+	.mv_name		= "Unknown",
 };
 ALIAS_MV(unknown)
-
-const char *get_system_type(void)
-{
-	return "Unknown";
-}
-
-void __init platform_setup(void)
-{
-}
-
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 86ef17fe48b5..de8df969d6af 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -43,27 +43,14 @@ extern void * __rd_start, * __rd_end;
  * The bigger value means no problem.
  */
 struct sh_cpuinfo boot_cpu_data = { CPU_SH_NONE, 10000000, };
+#ifdef CONFIG_VT
 struct screen_info screen_info;
+#endif
 
 #if defined(CONFIG_SH_UNKNOWN)
 struct sh_machine_vector sh_mv;
 #endif
 
-/* We need this to satisfy some external references. */
-struct screen_info screen_info = {
-        0, 25,                  /* orig-x, orig-y */
-        0,                      /* unused */
-        0,                      /* orig-video-page */
-        0,                      /* orig-video-mode */
-        80,                     /* orig-video-cols */
-        0,0,0,                  /* ega_ax, ega_bx, ega_cx */
-        25,                     /* orig-video-lines */
-        0,                      /* orig-video-isVGA */
-        16                      /* orig-video-points */
-};
-
-extern void platform_setup(void);
-extern char *get_system_type(void);
 extern int root_mountflags;
 
 #define MV_NAME_SIZE 32
@@ -90,29 +77,8 @@ static struct sh_machine_vector* __init get_mv_byname(const char* name);
 
 static char command_line[COMMAND_LINE_SIZE] = { 0, };
 
-struct resource standard_io_resources[] = {
-	{ "dma1", 0x00, 0x1f },
-	{ "pic1", 0x20, 0x3f },
-	{ "timer", 0x40, 0x5f },
-	{ "keyboard", 0x60, 0x6f },
-	{ "dma page reg", 0x80, 0x8f },
-	{ "pic2", 0xa0, 0xbf },
-	{ "dma2", 0xc0, 0xdf },
-	{ "fpu", 0xf0, 0xff }
-};
-
-#define STANDARD_IO_RESOURCES (sizeof(standard_io_resources)/sizeof(struct resource))
-
-/* System RAM - interrupted by the 640kB-1M hole */
-#define code_resource (ram_resources[3])
-#define data_resource (ram_resources[4])
-static struct resource ram_resources[] = {
-	{ "System RAM", 0x000000, 0x09ffff, IORESOURCE_BUSY },
-	{ "System RAM", 0x100000, 0x100000, IORESOURCE_BUSY },
-	{ "Video RAM area", 0x0a0000, 0x0bffff },
-	{ "Kernel code", 0x100000, 0 },
-	{ "Kernel data", 0, 0 }
-};
+static struct resource code_resource = { .name = "Kernel code", };
+static struct resource data_resource = { .name = "Kernel data", };
 
 unsigned long memory_start, memory_end;
 
@@ -255,6 +221,9 @@ static int __init sh_mv_setup(char **cmdline_p)
 	__set_io_port_base(mv_io_base);
 #endif
 
+	if (!sh_mv.mv_nr_irqs)
+		sh_mv.mv_nr_irqs = NR_IRQS;
+
 	return 0;
 }
 
@@ -263,7 +232,6 @@ void __init setup_arch(char **cmdline_p)
 	unsigned long bootmap_size;
 	unsigned long start_pfn, max_pfn, max_low_pfn;
 
-
 #ifdef CONFIG_CMDLINE_BOOL
         strcpy(COMMAND_LINE, CONFIG_CMDLINE);
 #endif
@@ -382,14 +350,14 @@ void __init setup_arch(char **cmdline_p)
 #endif
 
 	/* Perform the machine specific initialisation */
-	platform_setup();
+	if (likely(sh_mv.mv_setup))
+		sh_mv.mv_setup(cmdline_p);
 
 	paging_init();
 }
 
 struct sh_machine_vector* __init get_mv_byname(const char* name)
 {
-	extern int strcasecmp(const char *, const char *);
 	extern long __machvec_start, __machvec_end;
 	struct sh_machine_vector *all_vecs =
 		(struct sh_machine_vector *)&__machvec_start;
@@ -467,7 +435,8 @@ static void show_cpuflags(struct seq_file *m)
 	seq_printf(m, "\n");
 }
 
-static void show_cacheinfo(struct seq_file *m, const char *type, struct cache_info info)
+static void show_cacheinfo(struct seq_file *m, const char *type,
+			   struct cache_info info)
 {
 	unsigned int cache_size;
 
@@ -624,4 +593,3 @@ static int __init kgdb_parse_options(char *options)
 }
 __setup("kgdb=", kgdb_parse_options);
 #endif /* CONFIG_SH_KGDB */
-
diff --git a/include/asm-sh/machvec.h b/include/asm-sh/machvec.h
index 550501fa4fed..70389b72ffef 100644
--- a/include/asm-sh/machvec.h
+++ b/include/asm-sh/machvec.h
@@ -8,17 +8,18 @@
  */
 
 #ifndef _ASM_SH_MACHVEC_H
-#define _ASM_SH_MACHVEC_H 1
+#define _ASM_SH_MACHVEC_H
 
 #include <linux/types.h>
 #include <linux/time.h>
-
 #include <asm/machtypes.h>
 #include <asm/machvec_init.h>
 
 struct device;
 
 struct sh_machine_vector {
+	void (*mv_setup)(char **cmdline_p);
+	const char *mv_name;
 	int mv_nr_irqs;
 
 	u8 (*mv_inb)(unsigned long);
@@ -65,4 +66,6 @@ struct sh_machine_vector {
 
 extern struct sh_machine_vector sh_mv;
 
+#define get_system_type()	sh_mv.mv_name
+
 #endif /* _ASM_SH_MACHVEC_H */
-- 
cgit v1.2.3


From d153ea88dccf003173315b5d21acabebb897fb4a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:20:16 +0900
Subject: sh: stack debugging support.

This adds a DEBUG_STACK_USAGE and DEBUG_STACKOVERFLOW for SH.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig.debug        | 16 ++++++++++++++++
 arch/sh/kernel/head.S        |  4 +++-
 arch/sh/kernel/irq.c         | 17 ++++++++++++++++-
 arch/sh/kernel/vmlinux.lds.S | 14 ++++++++------
 include/asm-sh/page.h        |  2 +-
 include/asm-sh/thread_info.h | 12 ++++++++----
 6 files changed, 52 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 75ed1be8a49e..f0188e626be0 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -30,6 +30,22 @@ config EARLY_PRINTK
 	  when the kernel may crash or hang before the serial console is
 	  initialised. If unsure, say N.
 
+config DEBUG_STACKOVERFLOW
+	bool "Check for stack overflows"
+	depends on DEBUG_KERNEL
+	help
+	  This option will cause messages to be printed if free stack space
+	  drops below a certain limit.
+
+config DEBUG_STACK_USAGE
+	bool "Stack utilization instrumentation"
+	depends on DEBUG_KERNEL
+	help
+	  Enables the display of the minimum amount of free stack which each
+	  task has ever had available in the sysrq-T and sysrq-P debug output.
+
+	  This option will slow down process creation somewhat.
+
 config KGDB
 	bool "Include KGDB kernel debugger"
 	select FRAME_POINTER
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index c5e363872b91..3e7d00b7985a 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -11,6 +11,8 @@
  * Head.S contains the SH exception handlers and startup code.
  */
 #include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <asm/page.h>
 
 #ifdef CONFIG_CPU_SH4A
 #define SYNCO()		synco
@@ -95,7 +97,7 @@ ENTRY(_stext)
 
 	.balign 4
 1:	.long	0x400080F0		! MD=1, RB=0, BL=0, FD=1, IMASK=0xF
-2:	.long	init_thread_union+8192
+2:	.long	init_thread_union+THREAD_SIZE
 3:	.long	__bss_start
 4:	.long	_end
 5:	.long	start_kernel
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index c2e07f7f3496..7066611aeb72 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -60,7 +60,6 @@ unlock:
 }
 #endif
 
-
 asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 		      unsigned long r6, unsigned long r7,
 		      struct pt_regs regs)
@@ -69,6 +68,22 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 
 	irq_enter();
 
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+	/* Debugging check for stack overflow: is there less than 1KB free? */
+	{
+		long sp;
+
+		__asm__ __volatile__ ("and r15, %0" :
+					"=r" (sp) : "0" (THREAD_SIZE - 1));
+
+		if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
+			printk("do_IRQ: stack overflow: %ld\n",
+			       sp - sizeof(struct thread_info));
+			dump_stack();
+		}
+	}
+#endif
+
 #ifdef CONFIG_CPU_HAS_INTEVT
 	__asm__ __volatile__ (
 #ifdef CONFIG_CPU_HAS_SR_RB
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index eb860c51c697..0220d8a838a7 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -2,6 +2,8 @@
  * ld script to make SuperH Linux kernel
  * Written by Niibe Yutaka
  */
+#include <asm/thread_info.h>
+#include <asm/page.h>
 #include <asm-generic/vmlinux.lds.h>
 
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
@@ -40,16 +42,16 @@ SECTIONS
 	*(.data)
 
  	 /* Align the initial ramdisk image (INITRD) on page boundaries. */
- 	 . = ALIGN(4096);
+ 	 . = ALIGN(PAGE_SIZE);
  	 __rd_start = .;
  	 *(.initrd)
- 	 . = ALIGN(4096);
+ 	 . = ALIGN(PAGE_SIZE);
  	 __rd_end = .;
 
 	CONSTRUCTORS
 	}
 
-  . = ALIGN(4096);
+  . = ALIGN(PAGE_SIZE);
   .data.page_aligned : { *(.data.idt) }
 
   . = ALIGN(32);
@@ -60,10 +62,10 @@ SECTIONS
 
   _edata = .;			/* End of data section */
 
-  . = ALIGN(8192);		/* init_task */
+  . = ALIGN(THREAD_SIZE);		/* init_task */
   .data.init_task : { *(.data.init_task) }
 
-  . = ALIGN(4096);		/* Init code and data */
+  . = ALIGN(PAGE_SIZE);		/* Init code and data */
   __init_begin = .;
   _sinittext = .;
   .init.text : { *(.init.text) }
@@ -94,7 +96,7 @@ SECTIONS
   __machvec_start = .;
   .init.machvec : { *(.init.machvec) }
   __machvec_end = .;
-  . = ALIGN(4096);
+  . = ALIGN(PAGE_SIZE);
   __init_end = .;
 
   . = ALIGN(4);
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index e9135532d00c..888d6fe0030e 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -16,7 +16,7 @@
 
 /* PAGE_SHIFT determines the page size */
 #define PAGE_SHIFT	12
-#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#define PAGE_SIZE	(1 << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 #define PTE_MASK	PAGE_MASK
 
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 5eb874b065a6..605259f88113 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -29,6 +29,8 @@ struct thread_info {
 #endif
 
 #define PREEMPT_ACTIVE		0x10000000
+#define THREAD_SIZE		(PAGE_SIZE * 2)
+#define STACK_WARN		(THREAD_SIZE / 8)
 
 /*
  * macros/functions for gaining access to the thread information structure
@@ -50,8 +52,6 @@ struct thread_info {
 #define init_thread_info	(init_thread_union.thread_info)
 #define init_stack		(init_thread_union.stack)
 
-#define THREAD_SIZE (2*PAGE_SIZE)
-
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
@@ -73,8 +73,12 @@ static inline struct thread_info *current_thread_info(void)
 }
 
 /* thread information allocation */
-#define alloc_thread_info(ti) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
-#define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
+#ifdef CONFIG_DEBUG_STACK_USAGE
+#define alloc_thread_info(ti)	kzalloc(THREAD_SIZE, GFP_KERNEL)
+#else
+#define alloc_thread_info(ti)	kmalloc(THREAD_SIZE, GFP_KERNEL)
+#endif
+#define free_thread_info(ti)	kfree(ti)
 
 #else /* !__ASSEMBLY__ */
 
-- 
cgit v1.2.3


From a6a31139897a5e539efe7ad3b7bd351fa9673ce8 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:22:14 +0900
Subject: sh: Add support for 4K stacks.

This enables support for 4K stacks on SH.

Currently this depends on DEBUG_KERNEL, but likely all boards
will switch to this as the default in the future.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig.debug        |  10 +++
 arch/sh/kernel/cpu/irq/ipr.c |   2 +
 arch/sh/kernel/entry.S       |   4 +-
 arch/sh/kernel/head.S        |   5 +-
 arch/sh/kernel/irq.c         | 153 +++++++++++++++++++++++++++++++++++++++++--
 arch/sh/kernel/traps.c       |  18 ++---
 arch/sh/kernel/vmlinux.lds.S |   1 -
 include/asm-sh/irq.h         |   9 +++
 include/asm-sh/thread_info.h |  12 +++-
 9 files changed, 190 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index f0188e626be0..48479e014dac 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -46,6 +46,16 @@ config DEBUG_STACK_USAGE
 
 	  This option will slow down process creation somewhat.
 
+config 4KSTACKS
+	bool "Use 4Kb for kernel stacks instead of 8Kb"
+	depends on DEBUG_KERNEL
+	help
+	  If you say Y here the kernel will use a 4Kb stacksize for the
+	  kernel stack attached to each process/thread. This facilitates
+	  running more threads on a system and also reduces the pressure
+	  on the VM subsystem for higher order allocations. This option
+	  will also use IRQ stacks to compensate for the reduced stackspace.
+
 config KGDB
 	bool "Include KGDB kernel debugger"
 	select FRAME_POINTER
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index 4c4fd4118d1a..f785822cd5de 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -190,6 +190,8 @@ void __init init_IRQ(void)
 	/* Perform the machine specific initialisation */
 	if (sh_mv.mv_init_irq != NULL)
 		sh_mv.mv_init_irq();
+
+	irq_ctx_init(smp_processor_id());
 }
 
 #if !defined(CONFIG_CPU_HAS_PINT_IRQ)
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index fd5fe2349f20..fe8221855b28 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -716,8 +716,8 @@ ENTRY(handle_exception)
 	bt/s	1f		! It's a kernel to kernel transition.
 	 mov	r15, k0		! save original stack to k0
 	/* User space to kernel */
-	mov	#0x20, k1
-	shll8	k1		! k1 := 8192 (== THREAD_SIZE)
+	mov	#(THREAD_SIZE >> 8), k1
+	shll8	k1		! k1 := THREAD_SIZE
 	add	current, k1
 	mov	k1, r15		! change to kernel stack
 	!
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index 3e7d00b7985a..f5f53d14f245 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -12,7 +12,6 @@
  */
 #include <linux/linkage.h>
 #include <asm/thread_info.h>
-#include <asm/page.h>
 
 #ifdef CONFIG_CPU_SH4A
 #define SYNCO()		synco
@@ -69,8 +68,8 @@ ENTRY(_stext)
 	!
 	mov.l	2f, r0
 	mov	r0, r15		! Set initial r15 (stack pointer)
-	mov	#0x20, r1	!
-	shll8	r1		! r1 = 8192
+	mov	#(THREAD_SIZE >> 8), r1
+	shll8	r1		! r1 = THREAD_SIZE
 	sub	r1, r0		!
 	ldc	r0, r7_bank	! ... and initial thread_info
 
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index 7066611aeb72..c7ebd6aec951 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -1,5 +1,4 @@
-/* $Id: irq.c,v 1.20 2004/01/13 05:52:11 kkojima Exp $
- *
+/*
  * linux/arch/sh/kernel/irq.c
  *
  *	Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
@@ -7,13 +6,15 @@
  *
  * SuperH version:  Copyright (C) 1999  Niibe Yutaka
  */
-
 #include <linux/irq.h>
 #include <linux/interrupt.h>
+#include <linux/module.h>
 #include <linux/kernel_stat.h>
 #include <linux/seq_file.h>
 #include <asm/irq.h>
 #include <asm/processor.h>
+#include <asm/uaccess.h>
+#include <asm/thread_info.h>
 #include <asm/cpu/mmu_context.h>
 
 /*
@@ -60,11 +61,27 @@ unlock:
 }
 #endif
 
+#ifdef CONFIG_4KSTACKS
+/*
+ * per-CPU IRQ handling contexts (thread information and stack)
+ */
+union irq_ctx {
+	struct thread_info	tinfo;
+	u32			stack[THREAD_SIZE/sizeof(u32)];
+};
+
+static union irq_ctx *hardirq_ctx[NR_CPUS];
+static union irq_ctx *softirq_ctx[NR_CPUS];
+#endif
+
 asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 		      unsigned long r6, unsigned long r7,
 		      struct pt_regs regs)
 {
 	int irq = r4;
+#ifdef CONFIG_4KSTACKS
+	union irq_ctx *curctx, *irqctx;
+#endif
 
 	irq_enter();
 
@@ -102,7 +119,135 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 #endif
 
 	irq = irq_demux(irq);
-	__do_IRQ(irq, &regs);
+
+#ifdef CONFIG_4KSTACKS
+	curctx = (union irq_ctx *)current_thread_info();
+	irqctx = hardirq_ctx[smp_processor_id()];
+
+	/*
+	 * this is where we switch to the IRQ stack. However, if we are
+	 * already using the IRQ stack (because we interrupted a hardirq
+	 * handler) we can't do that and just have to keep using the
+	 * current stack (which is the irq stack already after all)
+	 */
+	if (curctx != irqctx) {
+		u32 *isp;
+
+		isp = (u32 *)((char *)irqctx + sizeof(*irqctx));
+		irqctx->tinfo.task = curctx->tinfo.task;
+		irqctx->tinfo.previous_sp = current_stack_pointer;
+
+		__asm__ __volatile__ (
+			"mov	%0, r4		\n"
+			"mov	%1, r5		\n"
+			"mov	r15, r9		\n"
+			"jsr	@%2		\n"
+			/* swith to the irq stack */
+			" mov	%3, r15		\n"
+			/* restore the stack (ring zero) */
+			"mov	r9, r15		\n"
+			: /* no outputs */
+			: "r" (irq), "r" (&regs), "r" (__do_IRQ), "r" (isp)
+			/* XXX: A somewhat excessive clobber list? -PFM */
+			: "memory", "r0", "r1", "r2", "r3", "r4",
+			  "r5", "r6", "r7", "r8", "t", "pr"
+		);
+	} else
+#endif
+		__do_IRQ(irq, &regs);
+
 	irq_exit();
+
 	return 1;
 }
+
+#ifdef CONFIG_4KSTACKS
+/*
+ * These should really be __section__(".bss.page_aligned") as well, but
+ * gcc's 3.0 and earlier don't handle that correctly.
+ */
+static char softirq_stack[NR_CPUS * THREAD_SIZE]
+		__attribute__((__aligned__(THREAD_SIZE)));
+
+static char hardirq_stack[NR_CPUS * THREAD_SIZE]
+		__attribute__((__aligned__(THREAD_SIZE)));
+
+/*
+ * allocate per-cpu stacks for hardirq and for softirq processing
+ */
+void irq_ctx_init(int cpu)
+{
+	union irq_ctx *irqctx;
+
+	if (hardirq_ctx[cpu])
+		return;
+
+	irqctx = (union irq_ctx *)&hardirq_stack[cpu * THREAD_SIZE];
+	irqctx->tinfo.task		= NULL;
+	irqctx->tinfo.exec_domain	= NULL;
+	irqctx->tinfo.cpu		= cpu;
+	irqctx->tinfo.preempt_count	= HARDIRQ_OFFSET;
+	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+
+	hardirq_ctx[cpu] = irqctx;
+
+	irqctx = (union irq_ctx *)&softirq_stack[cpu * THREAD_SIZE];
+	irqctx->tinfo.task		= NULL;
+	irqctx->tinfo.exec_domain	= NULL;
+	irqctx->tinfo.cpu		= cpu;
+	irqctx->tinfo.preempt_count	= SOFTIRQ_OFFSET;
+	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+
+	softirq_ctx[cpu] = irqctx;
+
+	printk("CPU %u irqstacks, hard=%p soft=%p\n",
+		cpu, hardirq_ctx[cpu], softirq_ctx[cpu]);
+}
+
+void irq_ctx_exit(int cpu)
+{
+	hardirq_ctx[cpu] = NULL;
+}
+
+extern asmlinkage void __do_softirq(void);
+
+asmlinkage void do_softirq(void)
+{
+	unsigned long flags;
+	struct thread_info *curctx;
+	union irq_ctx *irqctx;
+	u32 *isp;
+
+	if (in_interrupt())
+		return;
+
+	local_irq_save(flags);
+
+	if (local_softirq_pending()) {
+		curctx = current_thread_info();
+		irqctx = softirq_ctx[smp_processor_id()];
+		irqctx->tinfo.task = curctx->task;
+		irqctx->tinfo.previous_sp = current_stack_pointer;
+
+		/* build the stack frame on the softirq stack */
+		isp = (u32 *)((char *)irqctx + sizeof(*irqctx));
+
+		__asm__ __volatile__ (
+			"mov	r15, r9		\n"
+			"jsr	@%0		\n"
+			/* switch to the softirq stack */
+			" mov	%1, r15		\n"
+			/* restore the thread stack */
+			"mov	r9, r15		\n"
+			: /* no outputs */
+			: "r" (__do_softirq), "r" (isp)
+			/* XXX: A somewhat excessive clobber list? -PFM */
+			: "memory", "r0", "r1", "r2", "r3", "r4",
+			  "r5", "r6", "r7", "r8", "r9", "r15", "t", "pr"
+		);
+	}
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(do_softirq);
+#endif
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
index 95c810b3c97e..c2c597e09482 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -741,20 +741,12 @@ void show_stack(struct task_struct *tsk, unsigned long *sp)
 	unsigned long module_end = VMALLOC_END;
 	int i = 1;
 
-	if (tsk && !sp) {
+	if (!tsk)
+		tsk = current;
+	if (tsk == current)
+		sp = (unsigned long *)current_stack_pointer;
+	else
 		sp = (unsigned long *)tsk->thread.sp;
-	}
-
-	if (!sp) {
-		__asm__ __volatile__ (
-			"mov r15, %0\n\t"
-			"stc r7_bank, %1\n\t"
-			: "=r" (module_start),
-			  "=r" (module_end)
-		);
-		
-		sp = (unsigned long *)module_start;
-	}
 
 	stack = sp;
 
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index 0220d8a838a7..5eb930918186 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -3,7 +3,6 @@
  * Written by Niibe Yutaka
  */
 #include <asm/thread_info.h>
-#include <asm/page.h>
 #include <asm-generic/vmlinux.lds.h>
 
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index dd05e102fc0e..0e5f365aff70 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -719,6 +719,15 @@ static inline int generic_irq_demux(int irq)
 #define irq_canonicalize(irq)	(irq)
 #define irq_demux(irq)		__irq_demux(sh_mv.mv_irq_demux(irq))
 
+#ifdef CONFIG_4KSTACKS
+extern void irq_ctx_init(int cpu);
+extern void irq_ctx_exit(int cpu);
+# define __ARCH_HAS_DO_SOFTIRQ
+#else
+# define irq_ctx_init(cpu) do { } while (0)
+# define irq_ctx_exit(cpu) do { } while (0)
+#endif
+
 #if defined(CONFIG_CPU_SUBTYPE_SH73180)
 #include <asm/irq-sh73180.h>
 #endif
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 605259f88113..3ebc3f9039eb 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -9,8 +9,8 @@
  *  Copyright (C) 2002  David Howells (dhowells@redhat.com)
  *  - Incorporating suggestions made by Linus Torvalds and Dave Miller
  */
-
 #ifdef __KERNEL__
+#include <asm/page.h>
 
 #ifndef __ASSEMBLY__
 #include <asm/processor.h>
@@ -23,13 +23,20 @@ struct thread_info {
 	int			preempt_count; /* 0 => preemptable, <0 => BUG */
 	mm_segment_t		addr_limit;	/* thread address space */
 	struct restart_block	restart_block;
+	unsigned long		previous_sp;	/* sp of previous stack in case
+						   of nested IRQ stacks */
 	__u8			supervisor_stack[0];
 };
 
 #endif
 
 #define PREEMPT_ACTIVE		0x10000000
+
+#ifdef CONFIG_4KSTACKS
+#define THREAD_SIZE		(PAGE_SIZE)
+#else
 #define THREAD_SIZE		(PAGE_SIZE * 2)
+#endif
 #define STACK_WARN		(THREAD_SIZE / 8)
 
 /*
@@ -52,6 +59,9 @@ struct thread_info {
 #define init_thread_info	(init_thread_union.thread_info)
 #define init_stack		(init_thread_union.stack)
 
+/* how to get the current stack pointer from C */
+register unsigned long current_stack_pointer asm("r15") __attribute_used__;
+
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
-- 
cgit v1.2.3


From 315bb96824149614efe4844ded077a13fc908880 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:22:53 +0900
Subject: sh: CPU flags in AT_HWCAP in ELF auxvt.

Encode processor flags in AT_HWCAP in the ELF auxiliary vector.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c |  8 +++++---
 include/asm-sh/cpu-features.h  | 15 +++++++++++++++
 include/asm-sh/elf.h           | 10 ++++++----
 include/asm-sh/processor.h     | 12 +-----------
 4 files changed, 27 insertions(+), 18 deletions(-)
 create mode 100644 include/asm-sh/cpu-features.h

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 6e8a2b5268e8..0e65aa6ddcaa 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -76,6 +76,7 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->type = CPU_SH73180;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
+		cpu_data->flags |= CPU_HAS_LLSC;
 		break;
 	case 0x2001:
 	case 0x2004:
@@ -83,7 +84,7 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
 
-		cpu_data->flags |= CPU_HAS_FPU;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_LLSC;
 		break;
 	case 0x2006:
 	case 0x200A:
@@ -95,13 +96,15 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
 
-		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER |
+				   CPU_HAS_LLSC;
 		break;
 	case 0x3000:
 	case 0x3003:
 		cpu_data->type = CPU_SH7343;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
+		cpu_data->flags |= CPU_HAS_LLSC;
 		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
@@ -180,4 +183,3 @@ int __init detect_cpu_and_cache_system(void)
 
 	return 0;
 }
-
diff --git a/include/asm-sh/cpu-features.h b/include/asm-sh/cpu-features.h
new file mode 100644
index 000000000000..e398947ec01d
--- /dev/null
+++ b/include/asm-sh/cpu-features.h
@@ -0,0 +1,15 @@
+#ifndef __ASM_SH_CPU_FEATURES_H
+#define __ASM_SH_CPU_FEATURES_H
+
+/*
+ * Processor flags
+ */
+#define CPU_HAS_FPU		0x0001	/* Hardware FPU support */
+#define CPU_HAS_P2_FLUSH_BUG	0x0002	/* Need to flush the cache in P2 area */
+#define CPU_HAS_MMU_PAGE_ASSOC	0x0004	/* SH3: TLB way selection bit support */
+#define CPU_HAS_DSP		0x0008	/* SH-DSP: DSP support */
+#define CPU_HAS_PERF_COUNTER	0x0010	/* Hardware performance counters */
+#define CPU_HAS_PTEA		0x0020	/* PTEA register */
+#define CPU_HAS_LLSC		0x0040	/* movli.l/movco.l */
+
+#endif /* __ASM_SH_CPU_FEATURES_H */
diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index 1b63dfeea4f2..cc8e5e767345 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -1,6 +1,11 @@
 #ifndef __ASM_SH_ELF_H
 #define __ASM_SH_ELF_H
 
+#include <asm/processor.h>
+#include <asm/auxvec.h>
+#include <asm/ptrace.h>
+#include <asm/user.h>
+
 /* SH relocation types  */
 #define	R_SH_NONE		0
 #define	R_SH_DIR32		1
@@ -46,9 +51,6 @@
  * ELF register definitions..
  */
 
-#include <asm/ptrace.h>
-#include <asm/user.h>
-
 typedef unsigned long elf_greg_t;
 
 #define ELF_NGREG (sizeof (struct pt_regs) / sizeof(elf_greg_t))
@@ -91,7 +93,7 @@ typedef struct user_fpu_struct elf_fpregset_t;
    instruction set this CPU supports.  This could be done in user space,
    but it's not easy, and we've already done it here.  */
 
-#define ELF_HWCAP	(0)
+#define ELF_HWCAP	(boot_cpu_data.flags)
 
 /* This yields a string that ld.so will use to load implementation
    specific libraries for optimization.  This is more specific in
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index 3b3ef4f2bf31..bdd472705546 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -14,6 +14,7 @@
 #include <asm/types.h>
 #include <asm/cache.h>
 #include <asm/ptrace.h>
+#include <asm/cpu-features.h>
 
 /*
  * Default implementation of macro that returns current
@@ -127,17 +128,6 @@ union sh_fpu_union {
 	struct sh_fpu_soft_struct soft;
 };
 
-/*
- * Processor flags
- */
-
-#define CPU_HAS_FPU		0x0001	/* Hardware FPU support */
-#define CPU_HAS_P2_FLUSH_BUG	0x0002	/* Need to flush the cache in P2 area */
-#define CPU_HAS_MMU_PAGE_ASSOC	0x0004	/* SH3: TLB way selection bit support */
-#define CPU_HAS_DSP		0x0008	/* SH-DSP: DSP support */
-#define CPU_HAS_PERF_COUNTER	0x0010	/* Hardware performance counters */
-#define CPU_HAS_PTEA		0x0020	/* PTEA register */
-
 struct thread_struct {
 	unsigned long sp;
 	unsigned long pc;
-- 
cgit v1.2.3


From 2220d164933a8776d1336c814e3c2e5573256d34 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:24:28 +0900
Subject: sh: Report movli.l/movco.l capabilities.

Add llsc to cpu_flags[] and comment cpu-features.h.

Signed-off-by: Jamie Lenehan <nynaeve@twibble.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/setup.c        | 4 +++-
 include/asm-sh/cpu-features.h | 8 ++++++++
 2 files changed, 11 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index de8df969d6af..6810de3f8ed2 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -413,8 +413,10 @@ const char *get_cpu_subtype(void)
 }
 
 #ifdef CONFIG_PROC_FS
+/* Symbolic CPU flags, keep in sync with asm/cpu-features.h */
 static const char *cpu_flags[] = {
-	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr", "ptea", NULL
+	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr",
+	"ptea", "llsc", NULL
 };
 
 static void show_cpuflags(struct seq_file *m)
diff --git a/include/asm-sh/cpu-features.h b/include/asm-sh/cpu-features.h
index e398947ec01d..e1260aae3ee3 100644
--- a/include/asm-sh/cpu-features.h
+++ b/include/asm-sh/cpu-features.h
@@ -3,6 +3,14 @@
 
 /*
  * Processor flags
+ *
+ * Note: When adding a new flag, keep cpu_flags[] in
+ * arch/sh/kernel/setup.c in sync so symbolic name
+ * mapping of the processor flags has a chance of being
+ * reasonably accurate.
+ *
+ * These flags are also available through the ELF
+ * auxiliary vector as AT_HWCAP.
  */
 #define CPU_HAS_FPU		0x0001	/* Hardware FPU support */
 #define CPU_HAS_P2_FLUSH_BUG	0x0002	/* Need to flush the cache in P2 area */
-- 
cgit v1.2.3


From 05ae91585167410dadd1bc8f2e207a062e638a16 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:25:24 +0900
Subject: sh: Optimized readsl()/writesl() support.

Implement optimized copies of readsl()/writesl().

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/io.c | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/io.h |  6 +++++
 2 files changed, 73 insertions(+)

(limited to 'include')

diff --git a/arch/sh/kernel/io.c b/arch/sh/kernel/io.c
index 71c9fde2fd90..501fe03e3715 100644
--- a/arch/sh/kernel/io.c
+++ b/arch/sh/kernel/io.c
@@ -61,6 +61,73 @@ void memset_io(volatile void __iomem *dst, int c, unsigned long count)
 }
 EXPORT_SYMBOL(memset_io);
 
+void __raw_readsl(unsigned long addr, void *datap, int len)
+{
+	u32 *data;
+
+	for (data = datap; (len != 0) && (((u32)data & 0x1f) != 0); len--)
+		*data++ = ctrl_inl(addr);
+
+	if (likely(len >= (0x20 >> 2))) {
+		int tmp2, tmp3, tmp4, tmp5, tmp6;
+
+		__asm__ __volatile__(
+			"1:			\n\t"
+			"mov.l	@%7, r0		\n\t"
+			"mov.l	@%7, %2		\n\t"
+#ifdef CONFIG_CPU_SH4
+			"movca.l r0, @%0	\n\t"
+#else
+			"mov.l	r0, @%0		\n\t"
+#endif
+			"mov.l	@%7, %3		\n\t"
+			"mov.l	@%7, %4		\n\t"
+			"mov.l	@%7, %5		\n\t"
+			"mov.l	@%7, %6		\n\t"
+			"mov.l	@%7, r7		\n\t"
+			"mov.l	@%7, r0		\n\t"
+			"mov.l	%2, @(0x04,%0)	\n\t"
+			"mov	#0x20>>2, %2	\n\t"
+			"mov.l	%3, @(0x08,%0)	\n\t"
+			"sub	%2, %1		\n\t"
+			"mov.l	%4, @(0x0c,%0)	\n\t"
+			"cmp/hi	%1, %2		! T if 32 > len	\n\t"
+			"mov.l	%5, @(0x10,%0)	\n\t"
+			"mov.l	%6, @(0x14,%0)	\n\t"
+			"mov.l	r7, @(0x18,%0)	\n\t"
+			"mov.l	r0, @(0x1c,%0)	\n\t"
+			"bf.s	1b		\n\t"
+			" add	#0x20, %0	\n\t"
+			: "=&r" (data), "=&r" (len),
+			  "=&r" (tmp2), "=&r" (tmp3), "=&r" (tmp4),
+			  "=&r" (tmp5), "=&r" (tmp6)
+			: "r"(addr), "0" (data), "1" (len)
+			: "r0", "r7", "t", "memory");
+	}
+
+	for (; len != 0; len--)
+		*data++ = ctrl_inl(addr);
+}
+EXPORT_SYMBOL(__raw_readsl);
+
+void __raw_writesl(unsigned long addr, const void *data, int len)
+{
+	if (likely(len != 0)) {
+		int tmp1;
+
+		__asm__ __volatile__ (
+			"1:				\n\t"
+			"mov.l	@%0+, %1	\n\t"
+			"dt		%3		\n\t"
+			"bf.s		1b		\n\t"
+			" mov.l	%1, @%4		\n\t"
+			: "=&r" (data), "=&r" (tmp1)
+			: "0" (data), "r" (len), "r"(addr)
+			: "t", "memory");
+	}
+}
+EXPORT_SYMBOL(__raw_writesl);
+
 void __iomem *ioport_map(unsigned long port, unsigned int nr)
 {
 	return sh_mv.mv_ioport_map(port, nr);
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 377160b86295..ed12d38e8c00 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -107,6 +107,9 @@
 #define __raw_writew(v, a)	__writew(v, (void __iomem *)(a))
 #define __raw_writel(v, a)	__writel(v, (void __iomem *)(a))
 
+void __raw_writesl(unsigned long addr, const void *data, int longlen);
+void __raw_readsl(unsigned long addr, void *data, int longlen);
+
 /*
  * The platform header files may define some of these macros to use
  * the inlined versions where appropriate.  These macros may also be
@@ -132,6 +135,9 @@
 # define writel(v,a)	({ __raw_writel((v),(a)); mb(); })
 #endif
 
+#define writesl __raw_writesl
+#define readsl  __raw_readsl
+
 #define readb_relaxed(a) readb(a)
 #define readw_relaxed(a) readw(a)
 #define readl_relaxed(a) readl(a)
-- 
cgit v1.2.3


From 9d549a7d8ef71f684a35cf1e438543957cf81d12 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:26:05 +0900
Subject: sh: Update kexec support for API changes.

This was falling a bit behind..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/machine_kexec.c |  6 ------
 arch/sh/kernel/process.c       | 10 ----------
 include/asm-sh/kexec.h         |  9 +++------
 3 files changed, 3 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/machine_kexec.c b/arch/sh/kernel/machine_kexec.c
index 6bcd8d92399f..08587cdb64d6 100644
--- a/arch/sh/kernel/machine_kexec.c
+++ b/arch/sh/kernel/machine_kexec.c
@@ -29,12 +29,6 @@ extern const unsigned char relocate_new_kernel[];
 extern const unsigned int relocate_new_kernel_size;
 extern void *gdb_vbr_vector;
 
-/*
- * Provide a dummy crash_notes definition while crash dump arrives to ppc.
- * This prevents breakage of crash_notes attribute in kernel/ksysfs.c.
- */
-void *crash_notes = NULL;
-
 void machine_shutdown(void)
 {
 }
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 2167746e88f1..630ec1af2483 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -81,16 +81,6 @@ void cpu_idle(void)
 
 void machine_restart(char * __unused)
 {
-
-#ifdef CONFIG_KEXEC
-	struct kimage *image;
-	image = xchg(&kexec_image, 0);
-	if (image) {
-		machine_shutdown();
-		machine_kexec(image);
-	}
-#endif
-
 	/* SR.BL=1 and invoke address error to let CPU reset (manual reset) */
 	asm volatile("ldc %0, sr\n\t"
 		     "mov.l @%1, %0" : : "r" (0x10000000), "r" (0x80000001));
diff --git a/include/asm-sh/kexec.h b/include/asm-sh/kexec.h
index a5f85e9e428d..9d235af20cdd 100644
--- a/include/asm-sh/kexec.h
+++ b/include/asm-sh/kexec.h
@@ -25,11 +25,8 @@
 
 #define MAX_NOTE_BYTES 1024
 
-#ifndef __ASSEMBLY__
-
-extern void machine_shutdown(void);
-extern void *crash_notes;
-
-#endif /* __ASSEMBLY__ */
+/* Provide a dummy definition to avoid build failures. */
+static inline void crash_setup_regs(struct pt_regs *newregs,
+					struct pt_regs *oldregs) { }
 
 #endif /* _SH_KEXEC_H */
-- 
cgit v1.2.3


From 72c35543f8cf1316773ffbd9619575bb84ac44fb Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:27:43 +0900
Subject: sh: Support for L2 cache on newer SH-4A CPUs.

This implements preliminary support for the L2 caches found
on newer SH-4A CPUs.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c | 47 +++++++++++++++++++++++++++++++++++++++++-
 arch/sh/kernel/setup.c         |  6 +++++-
 include/asm-sh/cpu-features.h  |  1 +
 include/asm-sh/processor.h     |  9 ++++----
 4 files changed, 57 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 0e65aa6ddcaa..bee00cac0b16 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -29,7 +29,7 @@ int __init detect_cpu_and_cache_system(void)
 		[9] = (1 << 16)
 	};
 
-	pvr = (ctrl_inl(CCN_PVR) >> 8) & 0xffff;
+	pvr = (ctrl_inl(CCN_PVR) >> 8) & 0xffffff;
 	prr = (ctrl_inl(CCN_PRR) >> 4) & 0xff;
 	cvr = (ctrl_inl(CCN_CVR));
 
@@ -53,6 +53,26 @@ int __init detect_cpu_and_cache_system(void)
 	cpu_data->dcache.ways		= 1;
 	cpu_data->dcache.linesz		= L1_CACHE_BYTES;
 
+	/*
+	 * Setup some generic flags we can probe
+	 * (L2 and DSP detection only work on SH-4A)
+	 */
+	if (((pvr >> 16) & 0xff) == 0x10) {
+		if ((cvr & 0x02000000) == 0)
+			cpu_data->flags |= CPU_HAS_L2_CACHE;
+		if ((cvr & 0x10000000) == 0)
+			cpu_data->flags |= CPU_HAS_DSP;
+
+		cpu_data->flags |= CPU_HAS_LLSC;
+	}
+
+	/* FPU detection works for everyone */
+	if ((cvr & 0x20000000) == 1)
+		cpu_data->flags |= CPU_HAS_FPU;
+
+	/* Mask off the upper chip ID */
+	pvr &= 0xffff;
+
 	/*
 	 * Probe the underlying processor version/revision and
 	 * adjust cpu_data setup accordingly.
@@ -181,5 +201,30 @@ int __init detect_cpu_and_cache_system(void)
 	cpu_data->dcache.way_size = cpu_data->dcache.sets *
 				    cpu_data->dcache.linesz;
 
+	/*
+	 * Setup the L2 cache desc
+	 *
+	 * SH-4A's have an optional PIPT L2.
+	 */
+	if (cpu_data->flags & CPU_HAS_L2_CACHE) {
+		/*
+		 * Size calculation is much more sensible
+		 * than it is for the L1.
+		 *
+		 * Sizes are 128KB, 258KB, 512KB, and 1MB.
+		 */
+		size = (cvr & 0xf) << 17;
+
+		BUG_ON(!size);
+
+		cpu_data->scache.way_incr	= (1 << 16);
+		cpu_data->scache.entry_shift	= 5;
+		cpu_data->scache.entry_mask	= 0xffe0;
+		cpu_data->scache.ways		= 4;
+		cpu_data->scache.linesz		= L1_CACHE_BYTES;
+		cpu_data->scache.sets		= size /
+			(cpu_data->scache.linesz * cpu_data->scache.ways);
+	}
+
 	return 0;
 }
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 6810de3f8ed2..5f587332234a 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -416,7 +416,7 @@ const char *get_cpu_subtype(void)
 /* Symbolic CPU flags, keep in sync with asm/cpu-features.h */
 static const char *cpu_flags[] = {
 	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr",
-	"ptea", "llsc", NULL
+	"ptea", "llsc", "l2", NULL
 };
 
 static void show_cpuflags(struct seq_file *m)
@@ -480,6 +480,10 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		show_cacheinfo(m, "dcache", boot_cpu_data.dcache);
 	}
 
+	/* Optional secondary cache */
+	if (boot_cpu_data.flags & CPU_HAS_L2_CACHE)
+		show_cacheinfo(m, "scache", boot_cpu_data.scache);
+
 	seq_printf(m, "bogomips\t: %lu.%02lu\n",
 		     boot_cpu_data.loops_per_jiffy/(500000/HZ),
 		     (boot_cpu_data.loops_per_jiffy/(5000/HZ)) % 100);
diff --git a/include/asm-sh/cpu-features.h b/include/asm-sh/cpu-features.h
index e1260aae3ee3..4bccd7c032f9 100644
--- a/include/asm-sh/cpu-features.h
+++ b/include/asm-sh/cpu-features.h
@@ -19,5 +19,6 @@
 #define CPU_HAS_PERF_COUNTER	0x0010	/* Hardware performance counters */
 #define CPU_HAS_PTEA		0x0020	/* PTEA register */
 #define CPU_HAS_LLSC		0x0040	/* movli.l/movco.l */
+#define CPU_HAS_L2_CACHE	0x0080	/* Secondary cache / URAM */
 
 #endif /* __ASM_SH_CPU_FEATURES_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index bdd472705546..b7cba4e91a72 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -54,14 +54,15 @@ enum cpu_type {
 };
 
 struct sh_cpuinfo {
-	enum cpu_type type;
+	unsigned int type;
 	unsigned long loops_per_jiffy;
 
-	struct cache_info icache;
-	struct cache_info dcache;
+	struct cache_info icache;	/* Primary I-cache */
+	struct cache_info dcache;	/* Primary D-cache */
+	struct cache_info scache;	/* Secondary cache */
 
 	unsigned long flags;
-};
+} __attribute__ ((aligned(SMP_CACHE_BYTES)));
 
 extern struct sh_cpuinfo boot_cpu_data;
 
-- 
cgit v1.2.3


From 8c12b5dc13bf8516303a8224ab4e9708b33d5b00 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:31:06 +0900
Subject: sh: Clean up PAGE_SIZE definition for assembly use.

We want to be able to use PAGE_SIZE all over the place,
this is the same approach adopted by other architectures..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/page.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include')

diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 888d6fe0030e..acf6977b4042 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -16,7 +16,13 @@
 
 /* PAGE_SHIFT determines the page size */
 #define PAGE_SHIFT	12
+
+#ifdef __ASSEMBLY__
 #define PAGE_SIZE	(1 << PAGE_SHIFT)
+#else
+#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#endif
+
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 #define PTE_MASK	PAGE_MASK
 
-- 
cgit v1.2.3


From 19f9a34f87c48bbd270d617d1c986d0c23866a1a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:33:49 +0900
Subject: sh: Initial vsyscall page support.

This implements initial support for the vsyscall page on SH.
At the moment we leave it configurable due to having nommu
to support from the same code base. We hook it up for the
signal trampoline return at present, with more to be added
later, once uClibc catches up.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile                      |   1 +
 arch/sh/kernel/process.c                     |   2 +-
 arch/sh/kernel/signal.c                      |  17 ++-
 arch/sh/kernel/vsyscall/Makefile             |  36 +++++++
 arch/sh/kernel/vsyscall/vsyscall-note.S      |  25 +++++
 arch/sh/kernel/vsyscall/vsyscall-sigreturn.S |  39 +++++++
 arch/sh/kernel/vsyscall/vsyscall-syscall.S   |  10 ++
 arch/sh/kernel/vsyscall/vsyscall-trapa.S     |  42 ++++++++
 arch/sh/kernel/vsyscall/vsyscall.c           | 150 +++++++++++++++++++++++++++
 arch/sh/kernel/vsyscall/vsyscall.lds.S       |  74 +++++++++++++
 arch/sh/mm/Kconfig                           |  13 +++
 arch/sh/mm/init.c                            |   3 +
 arch/sh/mm/tlb-flush.c                       |  18 ++--
 include/asm-sh/auxvec.h                      |  14 +++
 include/asm-sh/elf.h                         |  20 ++++
 include/asm-sh/mmu.h                         |   7 +-
 include/asm-sh/mmu_context.h                 |   8 +-
 include/asm-sh/page.h                        |   5 +
 include/asm-sh/processor.h                   |   6 ++
 19 files changed, 473 insertions(+), 17 deletions(-)
 create mode 100644 arch/sh/kernel/vsyscall/Makefile
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-note.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-syscall.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-trapa.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall.c
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall.lds.S

(limited to 'include')

diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index 0e2148f63e7e..5da88a43d350 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -9,6 +9,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 	io.o io_generic.o sh_ksyms.o syscalls.o
 
 obj-y				+= cpu/ timers/
+obj-$(CONFIG_VSYSCALL)		+= vsyscall/
 
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_CF_ENABLER)	+= cf-enabler.o
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 630ec1af2483..0b1d5dd7a93b 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -355,7 +355,7 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	else if (next->thread.ubc_pc && next->mm) {
 		int asid = 0;
 #ifdef CONFIG_MMU
-		asid |= next->mm->context & MMU_CONTEXT_ASID_MASK;
+		asid |= next->mm->context.id & MMU_CONTEXT_ASID_MASK;
 #endif
 		ubc_set_tracing(asid, next->thread.ubc_pc);
 	} else {
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index 2f1c9545b49f..5213f5bc6ce0 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -8,7 +8,6 @@
  *  SuperH version:  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
  *
  */
-
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
@@ -21,6 +20,7 @@
 #include <linux/unistd.h>
 #include <linux/stddef.h>
 #include <linux/tty.h>
+#include <linux/elf.h>
 #include <linux/personality.h>
 #include <linux/binfmts.h>
 
@@ -29,8 +29,6 @@
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
-#undef DEBUG
-
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
 /*
@@ -312,6 +310,11 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
 	return (void __user *)((sp - frame_size) & -8ul);
 }
 
+/* These symbols are defined with the addresses in the vsyscall page.
+   See vsyscall-trapa.S.  */
+extern void __user __kernel_sigreturn;
+extern void __user __kernel_rt_sigreturn;
+
 static int setup_frame(int sig, struct k_sigaction *ka,
 			sigset_t *set, struct pt_regs *regs)
 {
@@ -340,6 +343,10 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
 		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_sigreturn);
+#endif
 	} else {
 		/* Generate return code (system call to sigreturn) */
 		err |= __put_user(MOVW(7), &frame->retcode[0]);
@@ -416,6 +423,10 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
 		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_rt_sigreturn);
+#endif
 	} else {
 		/* Generate return code (system call to rt_sigreturn) */
 		err |= __put_user(MOVW(7), &frame->retcode[0]);
diff --git a/arch/sh/kernel/vsyscall/Makefile b/arch/sh/kernel/vsyscall/Makefile
new file mode 100644
index 000000000000..4bbce1cfa359
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/Makefile
@@ -0,0 +1,36 @@
+obj-y += vsyscall.o vsyscall-syscall.o
+
+$(obj)/vsyscall-syscall.o: \
+	$(foreach F,trapa,$(obj)/vsyscall-$F.so)
+
+# Teach kbuild about targets
+targets += $(foreach F,trapa,vsyscall-$F.o vsyscall-$F.so)
+targets += vsyscall-note.o vsyscall.lds
+
+# The DSO images are built using a special linker script
+quiet_cmd_syscall = SYSCALL $@
+      cmd_syscall = $(CC) -nostdlib $(SYSCFLAGS_$(@F)) \
+			   -Wl,-T,$(filter-out FORCE,$^) -o $@
+
+export CPPFLAGS_vsyscall.lds += -P -C -Ush
+
+vsyscall-flags = -shared -s -Wl,-soname=linux-gate.so.1 \
+		$(call ld-option, -Wl$(comma)--hash-style=sysv)
+
+SYSCFLAGS_vsyscall-trapa.so	= $(vsyscall-flags)
+
+$(obj)/vsyscall-trapa.so: \
+$(obj)/vsyscall-%.so: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE
+	$(call if_changed,syscall)
+
+# We also create a special relocatable object that should mirror the symbol
+# table and layout of the linked DSO.  With ld -R we can then refer to
+# these symbols in the kernel code rather than hand-coded addresses.
+extra-y += vsyscall-syms.o
+$(obj)/built-in.o: $(obj)/vsyscall-syms.o
+$(obj)/built-in.o: ld_flags += -R $(obj)/vsyscall-syms.o
+
+SYSCFLAGS_vsyscall-syms.o = -r
+$(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
+			$(obj)/vsyscall-trapa.o $(obj)/vsyscall-note.o FORCE
+	$(call if_changed,syscall)
diff --git a/arch/sh/kernel/vsyscall/vsyscall-note.S b/arch/sh/kernel/vsyscall/vsyscall-note.S
new file mode 100644
index 000000000000..d4b5be4f3d5f
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-note.S
@@ -0,0 +1,25 @@
+/*
+ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
+ * Here we can supply some information useful to userland.
+ */
+
+#include <linux/uts.h>
+#include <linux/version.h>
+
+#define ASM_ELF_NOTE_BEGIN(name, flags, vendor, type)			      \
+	.section name, flags;						      \
+	.balign 4;							      \
+	.long 1f - 0f;		/* name length */			      \
+	.long 3f - 2f;		/* data length */			      \
+	.long type;		/* note type */				      \
+0:	.asciz vendor;		/* vendor name */			      \
+1:	.balign 4;							      \
+2:
+
+#define ASM_ELF_NOTE_END						      \
+3:	.balign 4;		/* pad out section */			      \
+	.previous
+
+	ASM_ELF_NOTE_BEGIN(".note.kernel-version", "a", UTS_SYSNAME, 0)
+	.long LINUX_VERSION_CODE
+	ASM_ELF_NOTE_END
diff --git a/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S b/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
new file mode 100644
index 000000000000..555a64f124ca
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
@@ -0,0 +1,39 @@
+#include <asm/unistd.h>
+
+	.text
+	.balign 32
+	.globl __kernel_sigreturn
+	.type __kernel_sigreturn,@function
+__kernel_sigreturn:
+.LSTART_sigreturn:
+	mov.w	1f, r3
+	trapa	#0x10
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+
+1:	.short	__NR_sigreturn
+.LEND_sigreturn:
+	.size __kernel_sigreturn,.-.LSTART_sigreturn
+
+	.balign 32
+	.globl __kernel_rt_sigreturn
+	.type __kernel_rt_sigreturn,@function
+__kernel_rt_sigreturn:
+.LSTART_rt_sigreturn:
+	mov.w	1f, r3
+	trapa	#0x10
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+
+1:	.short	__NR_rt_sigreturn
+.LEND_rt_sigreturn:
+	.size __kernel_rt_sigreturn,.-.LSTART_rt_sigreturn
+
+	.section .eh_frame,"a",@progbits
+	.previous
diff --git a/arch/sh/kernel/vsyscall/vsyscall-syscall.S b/arch/sh/kernel/vsyscall/vsyscall-syscall.S
new file mode 100644
index 000000000000..c2ac7f0282b3
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-syscall.S
@@ -0,0 +1,10 @@
+#include <linux/init.h>
+
+__INITDATA
+
+	.globl vsyscall_trapa_start, vsyscall_trapa_end
+vsyscall_trapa_start:
+	.incbin "arch/sh/kernel/vsyscall/vsyscall-trapa.so"
+vsyscall_trapa_end:
+
+__FINIT
diff --git a/arch/sh/kernel/vsyscall/vsyscall-trapa.S b/arch/sh/kernel/vsyscall/vsyscall-trapa.S
new file mode 100644
index 000000000000..3b6eb34c43fa
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-trapa.S
@@ -0,0 +1,42 @@
+	.text
+	.globl __kernel_vsyscall
+	.type __kernel_vsyscall,@function
+__kernel_vsyscall:
+.LSTART_vsyscall:
+	/* XXX: We'll have to do something here once we opt to use the vDSO
+	 * page for something other than the signal trampoline.. as well as
+	 * fill out .eh_frame -- PFM. */
+.LEND_vsyscall:
+	.size __kernel_vsyscall,.-.LSTART_vsyscall
+	.previous
+
+	.section .eh_frame,"a",@progbits
+.LCIE:
+	.ualong	.LCIE_end - .LCIE_start
+.LCIE_start:
+	.ualong	0		/* CIE ID */
+	.byte	0x1		/* Version number */
+	.string	"zRS"		/* NUL-terminated augmentation string */
+	.uleb128 0x1		/* Code alignment factor */
+	.sleb128 -4		/* Data alignment factor */
+	.byte	0x11		/* Return address register column */
+				/* Augmentation length and data (none) */
+	.byte	0xc		/* DW_CFA_def_cfa */
+	.uleb128 0xf		/* r15 */
+	.uleb128 0x0		/* offset 0 */
+
+	.align 2
+.LCIE_end:
+
+	.ualong	.LFDE_end-.LFDE_start	/* Length FDE */
+.LFDE_start:
+	.ualong	.LCIE			/* CIE pointer */
+	.ualong	.LSTART_vsyscall-.	/* start address */
+	.ualong	.LEND_vsyscall-.LSTART_vsyscall
+	.uleb128 0
+	.align 2
+.LFDE_end:
+	.previous
+
+/* Get the common code for the sigreturn entry points */
+#include "vsyscall-sigreturn.S"
diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c
new file mode 100644
index 000000000000..075d6cc1a2d7
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall.c
@@ -0,0 +1,150 @@
+/*
+ * arch/sh/kernel/vsyscall.c
+ *
+ *  Copyright (C) 2006 Paul Mundt
+ *
+ * vDSO randomization
+ * Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/gfp.h>
+#include <linux/module.h>
+#include <linux/elf.h>
+
+/*
+ * Should the kernel map a VDSO page into processes and pass its
+ * address down to glibc upon exec()?
+ */
+unsigned int __read_mostly vdso_enabled = 1;
+EXPORT_SYMBOL_GPL(vdso_enabled);
+
+static int __init vdso_setup(char *s)
+{
+	vdso_enabled = simple_strtoul(s, NULL, 0);
+	return 1;
+}
+__setup("vdso=", vdso_setup);
+
+/*
+ * These symbols are defined by vsyscall.o to mark the bounds
+ * of the ELF DSO images included therein.
+ */
+extern const char vsyscall_trapa_start, vsyscall_trapa_end;
+static void *syscall_page;
+
+int __init vsyscall_init(void)
+{
+	syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
+
+	/*
+	 * XXX: Map this page to a fixmap entry if we get around
+	 * to adding the page to ELF core dumps
+	 */
+
+	memcpy(syscall_page,
+	       &vsyscall_trapa_start,
+	       &vsyscall_trapa_end - &vsyscall_trapa_start);
+
+	return 0;
+}
+
+static struct page *syscall_vma_nopage(struct vm_area_struct *vma,
+				       unsigned long address, int *type)
+{
+	unsigned long offset = address - vma->vm_start;
+	struct page *page;
+
+	if (address < vma->vm_start || address > vma->vm_end)
+		return NOPAGE_SIGBUS;
+
+	page = virt_to_page(syscall_page + offset);
+
+	get_page(page);
+
+	return page;
+}
+
+/* Prevent VMA merging */
+static void syscall_vma_close(struct vm_area_struct *vma)
+{
+}
+
+static struct vm_operations_struct syscall_vm_ops = {
+	.nopage	= syscall_vma_nopage,
+	.close	= syscall_vma_close,
+};
+
+/* Setup a VMA at program startup for the vsyscall page */
+int arch_setup_additional_pages(struct linux_binprm *bprm,
+				int executable_stack)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	unsigned long addr;
+	int ret;
+
+	down_write(&mm->mmap_sem);
+	addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0);
+	if (IS_ERR_VALUE(addr)) {
+		ret = addr;
+		goto up_fail;
+	}
+
+	vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL);
+	if (!vma) {
+		ret = -ENOMEM;
+		goto up_fail;
+	}
+
+	vma->vm_start = addr;
+	vma->vm_end = addr + PAGE_SIZE;
+	/* MAYWRITE to allow gdb to COW and set breakpoints */
+	vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE;
+	vma->vm_flags |= mm->def_flags;
+	vma->vm_page_prot = protection_map[vma->vm_flags & 7];
+	vma->vm_ops = &syscall_vm_ops;
+	vma->vm_mm = mm;
+
+	ret = insert_vm_struct(mm, vma);
+	if (unlikely(ret)) {
+		kmem_cache_free(vm_area_cachep, vma);
+		goto up_fail;
+	}
+
+	current->mm->context.vdso = (void *)addr;
+
+	mm->total_vm++;
+up_fail:
+	up_write(&mm->mmap_sem);
+	return ret;
+}
+
+const char *arch_vma_name(struct vm_area_struct *vma)
+{
+	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
+		return "[vdso]";
+
+	return NULL;
+}
+
+struct vm_area_struct *get_gate_vma(struct task_struct *task)
+{
+	return NULL;
+}
+
+int in_gate_area(struct task_struct *task, unsigned long address)
+{
+	return 0;
+}
+
+int in_gate_area_no_task(unsigned long address)
+{
+	return 0;
+}
diff --git a/arch/sh/kernel/vsyscall/vsyscall.lds.S b/arch/sh/kernel/vsyscall/vsyscall.lds.S
new file mode 100644
index 000000000000..b13c3d439fee
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall.lds.S
@@ -0,0 +1,74 @@
+/*
+ * Linker script for vsyscall DSO.  The vsyscall page is an ELF shared
+ * object prelinked to its virtual address, and with only one read-only
+ * segment (that fits in one page).  This script controls its layout.
+ */
+#include <asm/asm-offsets.h>
+
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+OUTPUT_FORMAT("elf32-sh-linux", "elf32-sh-linux", "elf32-sh-linux")
+#else
+OUTPUT_FORMAT("elf32-shbig-linux", "elf32-shbig-linux", "elf32-shbig-linux")
+#endif
+OUTPUT_ARCH(sh)
+
+/* The ELF entry point can be used to set the AT_SYSINFO value.  */
+ENTRY(__kernel_vsyscall);
+
+SECTIONS
+{
+  . = SIZEOF_HEADERS;
+
+  .hash           : { *(.hash) }		:text
+  .gnu.hash       : { *(.gnu.hash) }
+  .dynsym         : { *(.dynsym) }
+  .dynstr         : { *(.dynstr) }
+  .gnu.version    : { *(.gnu.version) }
+  .gnu.version_d  : { *(.gnu.version_d) }
+  .gnu.version_r  : { *(.gnu.version_r) }
+
+  /* This linker script is used both with -r and with -shared.
+     For the layouts to match, we need to skip more than enough
+     space for the dynamic symbol table et al.  If this amount
+     is insufficient, ld -shared will barf.  Just increase it here.  */
+  . = 0x400;
+
+  .text           : { *(.text) }		:text =0x90909090
+  .note		  : { *(.note.*) }		:text :note
+  .eh_frame_hdr   : { *(.eh_frame_hdr) }	:text :eh_frame_hdr
+  .eh_frame       : { KEEP (*(.eh_frame)) }	:text
+  .dynamic        : { *(.dynamic) }		:text :dynamic
+  .useless        : {
+  	*(.got.plt) *(.got)
+	*(.data .data.* .gnu.linkonce.d.*)
+	*(.dynbss)
+	*(.bss .bss.* .gnu.linkonce.b.*)
+  }						:text
+}
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+  text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
+  dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
+  note PT_NOTE FLAGS(4); /* PF_R */
+  eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */
+}
+
+/*
+ * This controls what symbols we export from the DSO.
+ */
+VERSION
+{
+  LINUX_2.6 {
+    global:
+    	__kernel_vsyscall;
+    	__kernel_sigreturn;
+    	__kernel_rt_sigreturn;
+
+    local: *;
+  };
+}
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index b445d02075e8..9dd606464d23 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -223,6 +223,19 @@ config 32BIT
 	  32-bits through the SH-4A PMB. If this is not set, legacy
 	  29-bit physical addressing will be used.
 
+config VSYSCALL
+	bool "Support vsyscall page"
+	depends on MMU
+	default y
+	help
+	  This will enable support for the kernel mapping a vDSO page
+	  in process space, and subsequently handing down the entry point
+	  to the libc through the ELF auxiliary vector.
+
+	  From the kernel side this is used for the signal trampoline.
+	  For systems with an MMU that can afford to give up a page,
+	  (the default value) say Y.
+
 choice
 	prompt "HugeTLB page size"
 	depends on HUGETLB_PAGE && CPU_SH4 && MMU
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index ad182b31d846..7154d1ce9785 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -286,6 +286,9 @@ void __init mem_init(void)
 		initsize >> 10);
 
 	p3_cache_init();
+
+	/* Initialize the vDSO */
+	vsyscall_init();
 }
 
 void free_initmem(void)
diff --git a/arch/sh/mm/tlb-flush.c b/arch/sh/mm/tlb-flush.c
index fd7e42bcaa40..73ec7f6084fa 100644
--- a/arch/sh/mm/tlb-flush.c
+++ b/arch/sh/mm/tlb-flush.c
@@ -14,12 +14,12 @@
 
 void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
 {
-	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
+	if (vma->vm_mm && vma->vm_mm->context.id != NO_CONTEXT) {
 		unsigned long flags;
 		unsigned long asid;
 		unsigned long saved_asid = MMU_NO_ASID;
 
-		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
+		asid = vma->vm_mm->context.id & MMU_CONTEXT_ASID_MASK;
 		page &= PAGE_MASK;
 
 		local_irq_save(flags);
@@ -39,20 +39,21 @@ void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 {
 	struct mm_struct *mm = vma->vm_mm;
 
-	if (mm->context != NO_CONTEXT) {
+	if (mm->context.id != NO_CONTEXT) {
 		unsigned long flags;
 		int size;
 
 		local_irq_save(flags);
 		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
 		if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
-			mm->context = NO_CONTEXT;
+			mm->context.id = NO_CONTEXT;
 			if (mm == current->mm)
 				activate_context(mm);
 		} else {
-			unsigned long asid = mm->context&MMU_CONTEXT_ASID_MASK;
+			unsigned long asid;
 			unsigned long saved_asid = MMU_NO_ASID;
 
+			asid = mm->context.id & MMU_CONTEXT_ASID_MASK;
 			start &= PAGE_MASK;
 			end += (PAGE_SIZE - 1);
 			end &= PAGE_MASK;
@@ -81,9 +82,10 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 	if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
 		flush_tlb_all();
 	} else {
-		unsigned long asid = init_mm.context&MMU_CONTEXT_ASID_MASK;
+		unsigned long asid;
 		unsigned long saved_asid = get_asid();
 
+		asid = init_mm.context.id & MMU_CONTEXT_ASID_MASK;
 		start &= PAGE_MASK;
 		end += (PAGE_SIZE - 1);
 		end &= PAGE_MASK;
@@ -101,11 +103,11 @@ void flush_tlb_mm(struct mm_struct *mm)
 {
 	/* Invalidate all TLB of this process. */
 	/* Instead of invalidating each TLB, we get new MMU context. */
-	if (mm->context != NO_CONTEXT) {
+	if (mm->context.id != NO_CONTEXT) {
 		unsigned long flags;
 
 		local_irq_save(flags);
-		mm->context = NO_CONTEXT;
+		mm->context.id = NO_CONTEXT;
 		if (mm == current->mm)
 			activate_context(mm);
 		local_irq_restore(flags);
diff --git a/include/asm-sh/auxvec.h b/include/asm-sh/auxvec.h
index fc21e4db5881..1b6916e63e90 100644
--- a/include/asm-sh/auxvec.h
+++ b/include/asm-sh/auxvec.h
@@ -1,4 +1,18 @@
 #ifndef __ASM_SH_AUXVEC_H
 #define __ASM_SH_AUXVEC_H
 
+/*
+ * Architecture-neutral AT_ values in 0-17, leave some room
+ * for more of them.
+ */
+
+#ifdef CONFIG_VSYSCALL
+/*
+ * Only define this in the vsyscall case, the entry point to
+ * the vsyscall page gets placed here. The kernel will attempt
+ * to build a gate VMA we don't care about otherwise..
+ */
+#define AT_SYSINFO_EHDR		33
+#endif
+
 #endif /* __ASM_SH_AUXVEC_H */
diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index cc8e5e767345..3a07ab40ac4d 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -121,4 +121,24 @@ extern int dump_task_fpu (struct task_struct *, elf_fpregset_t *);
 #define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs)
 #endif
 
+#ifdef CONFIG_VSYSCALL
+/* vDSO has arch_setup_additional_pages */
+#define ARCH_HAS_SETUP_ADDITIONAL_PAGES
+struct linux_binprm;
+extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+				       int executable_stack);
+
+extern unsigned int vdso_enabled;
+extern void __kernel_vsyscall;
+
+#define VDSO_BASE		((unsigned long)current->mm->context.vdso)
+#define VDSO_SYM(x)		(VDSO_BASE + (unsigned long)(x))
+
+#define ARCH_DLINFO						\
+do {								\
+	if (vdso_enabled)					\
+		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_BASE);	\
+} while (0)
+#endif /* CONFIG_VSYSCALL */
+
 #endif /* __ASM_SH_ELF_H */
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index 6383dc84501e..cf47df79bb94 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -11,7 +11,12 @@ typedef struct {
 #else
 
 /* Default "unsigned long" context */
-typedef unsigned long mm_context_t;
+typedef unsigned long mm_context_id_t;
+
+typedef struct {
+	mm_context_id_t id;
+	void *vdso;
+} mm_context_t;
 
 #endif /* CONFIG_MMU */
 
diff --git a/include/asm-sh/mmu_context.h b/include/asm-sh/mmu_context.h
index 87678ba8d6b6..c7088efe579a 100644
--- a/include/asm-sh/mmu_context.h
+++ b/include/asm-sh/mmu_context.h
@@ -49,7 +49,7 @@ get_mmu_context(struct mm_struct *mm)
 	unsigned long mc = mmu_context_cache;
 
 	/* Check if we have old version of context. */
-	if (((mm->context ^ mc) & MMU_CONTEXT_VERSION_MASK) == 0)
+	if (((mm->context.id ^ mc) & MMU_CONTEXT_VERSION_MASK) == 0)
 		/* It's up to date, do nothing */
 		return;
 
@@ -68,7 +68,7 @@ get_mmu_context(struct mm_struct *mm)
 		if (!mc)
 			mmu_context_cache = mc = MMU_CONTEXT_FIRST_VERSION;
 	}
-	mm->context = mc;
+	mm->context.id = mc;
 }
 
 /*
@@ -78,7 +78,7 @@ get_mmu_context(struct mm_struct *mm)
 static __inline__ int init_new_context(struct task_struct *tsk,
 				       struct mm_struct *mm)
 {
-	mm->context = NO_CONTEXT;
+	mm->context.id = NO_CONTEXT;
 
 	return 0;
 }
@@ -123,7 +123,7 @@ static __inline__ unsigned long get_asid(void)
 static __inline__ void activate_context(struct mm_struct *mm)
 {
 	get_mmu_context(mm);
-	set_asid(mm->context & MMU_CONTEXT_ASID_MASK);
+	set_asid(mm->context.id & MMU_CONTEXT_ASID_MASK);
 }
 
 /* MMU_TTB can be used for optimizing the fault handling.
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index acf6977b4042..3d8dae31a6f6 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -117,5 +117,10 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+/* vDSO support */
+#ifdef CONFIG_VSYSCALL
+#define __HAVE_ARCH_GATE_AREA
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PAGE_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index b7cba4e91a72..474773853cd1 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -276,5 +276,11 @@ static inline void prefetch(void *x)
 #define prefetchw(x)	prefetch(x)
 #endif
 
+#ifdef CONFIG_VSYSCALL
+extern int vsyscall_init(void);
+#else
+#define vsyscall_init() do { } while (0)
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PROCESSOR_H */
-- 
cgit v1.2.3


From 87b0ef91b6f27c07bf7dcce8584437481f473092 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:34:41 +0900
Subject: sh: dma-mapping compile fixes.

Silly bug, make it build again..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/dma-mapping.h | 42 +++++++++++++++++++++++++-----------------
 1 file changed, 25 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/asm-sh/dma-mapping.h b/include/asm-sh/dma-mapping.h
index 124968f9866e..56cd4b977232 100644
--- a/include/asm-sh/dma-mapping.h
+++ b/include/asm-sh/dma-mapping.h
@@ -141,25 +141,35 @@ static inline void dma_sync_sg(struct device *dev, struct scatterlist *sg,
 	}
 }
 
-static void dma_sync_single_for_cpu(struct device *dev,
-				    dma_addr_t dma_handle, size_t size,
-				    enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_single")));
+static inline void dma_sync_single_for_cpu(struct device *dev,
+					   dma_addr_t dma_handle, size_t size,
+					   enum dma_data_direction dir)
+{
+	dma_sync_single(dev, dma_handle, size, dir);
+}
+
+static inline void dma_sync_single_for_device(struct device *dev,
+					      dma_addr_t dma_handle,
+					      size_t size,
+					      enum dma_data_direction dir)
+{
+	dma_sync_single(dev, dma_handle, size, dir);
+}
 
-static void dma_sync_single_for_device(struct device *dev,
-				       dma_addr_t dma_handle, size_t size,
+static inline void dma_sync_sg_for_cpu(struct device *dev,
+				       struct scatterlist *sg, int nelems,
 				       enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_single")));
+{
+	dma_sync_sg(dev, sg, nelems, dir);
+}
 
-static void dma_sync_sg_for_cpu(struct device *dev,
-				struct scatterlist *sg, int nelems,
-				enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_sg")));
+static inline void dma_sync_sg_for_device(struct device *dev,
+					  struct scatterlist *sg, int nelems,
+					  enum dma_data_direction dir)
+{
+	dma_sync_sg(dev, sg, nelems, dir);
+}
 
-static void dma_sync_sg_for_device(struct device *dev,
-				   struct scatterlist *sg, int nelems,
-				   enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_sg")));
 
 static inline int dma_get_cache_alignment(void)
 {
@@ -174,6 +184,4 @@ static inline int dma_mapping_error(dma_addr_t dma_addr)
 {
 	return dma_addr == 0;
 }
-
 #endif /* __ASM_SH_DMA_MAPPING_H */
-
-- 
cgit v1.2.3


From f3c2575818fab45f8609e4aef2e43ab02b3a142e Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:36:17 +0900
Subject: sh: Calculate shm alignment at runtime.

Set the SHM alignment at runtime, based off of probed cache desc.
Optimize get_unmapped_area() to only colour align shared mappings.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/init.c           |  5 ++++
 arch/sh/kernel/sys_sh.c             | 54 +++++++++++++++++++++++--------------
 include/asm-sh/cacheflush.h         |  2 ++
 include/asm-sh/cpu-sh3/cacheflush.h |  8 ------
 include/asm-sh/cpu-sh4/cacheflush.h |  3 ---
 include/asm-sh/page.h               |  2 ++
 include/asm-sh/pgtable.h            |  2 --
 7 files changed, 43 insertions(+), 33 deletions(-)

(limited to 'include')

diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index 731dd61419dd..bfb90eb0b7a6 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -14,6 +14,7 @@
 #include <linux/kernel.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
+#include <asm/page.h>
 #include <asm/system.h>
 #include <asm/cacheflush.h>
 #include <asm/cache.h>
@@ -198,6 +199,10 @@ asmlinkage void __init sh_cpu_init(void)
 	/* Init the cache */
 	cache_init();
 
+	shm_align_mask = max_t(unsigned long,
+			       cpu_data->dcache.way_size - 1,
+			       PAGE_SIZE - 1);
+
 	/* Disable the FPU */
 	if (fpu_disabled) {
 		printk("FPU Disabled\n");
diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index 0ee7bf4cb238..b68ff705f067 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -21,6 +21,7 @@
 #include <linux/mman.h>
 #include <linux/file.h>
 #include <linux/utsname.h>
+#include <linux/module.h>
 #include <asm/cacheflush.h>
 #include <asm/uaccess.h>
 #include <asm/ipc.h>
@@ -44,11 +45,16 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5,
 	return error;
 }
 
-#if defined(HAVE_ARCH_UNMAPPED_AREA) && defined(CONFIG_MMU)
+unsigned long shm_align_mask = PAGE_SIZE - 1;	/* Sane caches */
+
+EXPORT_SYMBOL(shm_align_mask);
+
 /*
- * To avoid cache alias, we map the shard page with same color.
+ * To avoid cache aliases, we map the shared page with same color.
  */
-#define COLOUR_ALIGN(addr)	(((addr)+SHMLBA-1)&~(SHMLBA-1))
+#define COLOUR_ALIGN(addr, pgoff)				\
+	((((addr) + shm_align_mask) & ~shm_align_mask) +	\
+	 (((pgoff) << PAGE_SHIFT) & shm_align_mask))
 
 unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	unsigned long len, unsigned long pgoff, unsigned long flags)
@@ -56,43 +62,52 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
+	int do_colour_align;
 
 	if (flags & MAP_FIXED) {
 		/* We do not accept a shared mapping if it would violate
 		 * cache aliasing constraints.
 		 */
-		if ((flags & MAP_SHARED) && (addr & (SHMLBA - 1)))
+		if ((flags & MAP_SHARED) && (addr & shm_align_mask))
 			return -EINVAL;
 		return addr;
 	}
 
-	if (len > TASK_SIZE)
+	if (unlikely(len > TASK_SIZE))
 		return -ENOMEM;
 
+	do_colour_align = 0;
+	if (filp || (flags & MAP_SHARED))
+		do_colour_align = 1;
+
 	if (addr) {
-		if (flags & MAP_PRIVATE)
-			addr = PAGE_ALIGN(addr);
+		if (do_colour_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
 		else
-			addr = COLOUR_ALIGN(addr);
+			addr = PAGE_ALIGN(addr);
+
 		vma = find_vma(mm, addr);
 		if (TASK_SIZE - len >= addr &&
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
-	if (len <= mm->cached_hole_size) {
+
+	if (len > mm->cached_hole_size) {
+		start_addr = addr = mm->free_area_cache;
+	} else {
 	        mm->cached_hole_size = 0;
-		mm->free_area_cache = TASK_UNMAPPED_BASE;
+		start_addr = addr = TASK_UNMAPPED_BASE;
 	}
-	if (flags & MAP_PRIVATE)
-		addr = PAGE_ALIGN(mm->free_area_cache);
-	else
-		addr = COLOUR_ALIGN(mm->free_area_cache);
-	start_addr = addr;
 
 full_search:
+	if (do_colour_align)
+		addr = COLOUR_ALIGN(addr, pgoff);
+	else
+		addr = PAGE_ALIGN(mm->free_area_cache);
+
 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
 		/* At this point:  (!vma || addr < vma->vm_end). */
-		if (TASK_SIZE - len < addr) {
+		if (unlikely(TASK_SIZE - len < addr)) {
 			/*
 			 * Start a new search - just in case we missed
 			 * some holes.
@@ -104,7 +119,7 @@ full_search:
 			}
 			return -ENOMEM;
 		}
-		if (!vma || addr + len <= vma->vm_start) {
+		if (likely(!vma || addr + len <= vma->vm_start)) {
 			/*
 			 * Remember the place where we stopped the search:
 			 */
@@ -115,11 +130,10 @@ full_search:
 		        mm->cached_hole_size = vma->vm_start - addr;
 
 		addr = vma->vm_end;
-		if (!(flags & MAP_PRIVATE))
-			addr = COLOUR_ALIGN(addr);
+		if (do_colour_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
 	}
 }
-#endif
 
 static inline long
 do_mmap2(unsigned long addr, unsigned long len, unsigned long prot, 
diff --git a/include/asm-sh/cacheflush.h b/include/asm-sh/cacheflush.h
index 92930b4a40d4..07f62ec9ff0c 100644
--- a/include/asm-sh/cacheflush.h
+++ b/include/asm-sh/cacheflush.h
@@ -28,5 +28,7 @@ extern void __flush_invalidate_region(void *start, int size);
 		memcpy(dst, src, len);				\
 	} while (0)
 
+#define HAVE_ARCH_UNMAPPED_AREA
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_CACHEFLUSH_H */
diff --git a/include/asm-sh/cpu-sh3/cacheflush.h b/include/asm-sh/cpu-sh3/cacheflush.h
index 97f5a64c2ab8..03fde97a7fd0 100644
--- a/include/asm-sh/cpu-sh3/cacheflush.h
+++ b/include/asm-sh/cpu-sh3/cacheflush.h
@@ -64,12 +64,4 @@ void flush_icache_page(struct vm_area_struct *vma, struct page *page);
 
 #define p3_cache_init()				do { } while (0)
 
-/*
- * We provide our own get_unmapped_area to avoid cache aliasing issues
- * on SH7705 with a 32KB cache, and to page align addresses in the
- * non-aliasing case.
- */
-#define HAVE_ARCH_UNMAPPED_AREA
-
 #endif /* __ASM_CPU_SH3_CACHEFLUSH_H */
-
diff --git a/include/asm-sh/cpu-sh4/cacheflush.h b/include/asm-sh/cpu-sh4/cacheflush.h
index a95fc951aff6..515fd574267c 100644
--- a/include/asm-sh/cpu-sh4/cacheflush.h
+++ b/include/asm-sh/cpu-sh4/cacheflush.h
@@ -39,9 +39,6 @@ void p3_cache_init(void);
 
 #define PG_mapped	PG_arch_1
 
-/* We provide our own get_unmapped_area to avoid cache alias issue */
-#define HAVE_ARCH_UNMAPPED_AREA
-
 #ifdef CONFIG_MMU
 extern int remap_area_pages(unsigned long addr, unsigned long phys_addr,
 			    unsigned long size, unsigned long flags);
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 3d8dae31a6f6..ca8b26d90475 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -44,6 +44,8 @@
 extern void (*clear_page)(void *to);
 extern void (*copy_page)(void *to, void *from);
 
+extern unsigned long shm_align_mask;
+
 #ifdef CONFIG_MMU
 extern void clear_page_slow(void *to);
 extern void copy_page_slow(void *to, void *from);
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 41c559d8ba87..2c8682ad1012 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -340,6 +340,4 @@ extern pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t
 #include <asm-generic/pgtable.h>
 
 #endif /* !__ASSEMBLY__ */
-
 #endif /* __ASM_SH_PAGE_H */
-
-- 
cgit v1.2.3


From b4b30a5a0a270e6b3fef88373ad35d235a047fc1 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 3 Aug 2006 17:34:24 +0100
Subject: [MIPS] Cleanup leftovers of ARCH_HAS_IRQ_PER_CPU

CONFIG_IRQ_PER_CPU now controls the IRQ_PER_CPU stuff.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig      | 3 +--
 include/asm-mips/irq.h | 4 ----
 2 files changed, 1 insertion(+), 6 deletions(-)

(limited to 'include')

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 330f6abc7703..18e15a651d51 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1649,9 +1649,7 @@ config GENERIC_IRQ_PROBE
 	default y
 
 config IRQ_PER_CPU
-	depends on SMP
 	bool
-	default y
 
 #
 # - Highmem only makes sense for the 32-bit kernel.
@@ -1719,6 +1717,7 @@ source "mm/Kconfig"
 config SMP
 	bool "Multi-Processing support"
 	depends on SYS_SUPPORTS_SMP
+	select IRQ_PER_CPU
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/include/asm-mips/irq.h b/include/asm-mips/irq.h
index 896550bad322..d35c61776a02 100644
--- a/include/asm-mips/irq.h
+++ b/include/asm-mips/irq.h
@@ -76,8 +76,4 @@ extern int setup_irq_smtc(unsigned int irq, struct irqaction * new,
                           unsigned long hwmask);
 #endif /* CONFIG_MIPS_MT_SMTC */
 
-#ifdef CONFIG_SMP
-#define ARCH_HAS_IRQ_PER_CPU
-#endif
-
 #endif /* _ASM_IRQ_H */
-- 
cgit v1.2.3


From 585fa72493edd7d5acb308806e7bb609412c6228 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 12 Aug 2006 16:40:08 +0100
Subject: [MIPS] Retire flush_icache_page from mm use.

On the 34K the redundant cache operations were causing excessive stalls
resulting in realtime code running on the second VPE missing its deadline.
For all other platforms this patch is just a significant performance
improvment as illustrated by below benchmark numbers.

Processor, Processes - times in microseconds - smaller is better
------------------------------------------------------------------------------
Host                 OS  Mhz null null      open slct sig  sig  fork exec sh
                             call  I/O stat clos TCP  inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
25Kf      2.6.18-rc4     533 0.49 1.16 7.57 33.4 30.5 1.34 12.4 5497 17.K 54.K
25Kf      2.6.18-rc4-p   533 0.49 1.16 6.68 23.0 30.7 1.36 8.55 5030 16.K 48.K
4Kc       2.6.18-rc4      80 4.21 15.0 131. 289. 261. 16.5 258. 18.K 70.K 227K
4Kc       2.6.18-rc4-p    80 4.34 13.1 128. 285. 262. 18.2 258. 12.K 52.K 176K
34Kc      2.6.18-rc4      40 5.01 14.0 61.6 90.0 477. 17.9 94.7 29.K 108K 342K
34Kc      2.6.18-rc4-p    40 4.98 13.9 61.2 89.7 475. 17.6 93.7 8758 44.K 158K
BCM1480   2.6.18-rc4     700 0.28 0.60 3.68 5.92 16.0 0.78 5.08 931. 3163 15.K
BCM1480   2.6.18-rc4-p   700 0.28 0.61 3.65 5.85 16.0 0.79 5.20 395. 1464 8385
TX49-16K  2.6.18-rc3     197 0.73 2.41 19.0 37.8 82.9 2.94 17.5 4438 14.K 56.K
TX49-16K  2.6.18-rc3-p   197 0.73 2.40 19.9 36.3 82.9 2.94 23.4 2577 9103 38.K
TX49-32K  2.6.18-rc3     396 0.36 1.19 6.80 11.8 41.0 1.46 8.17 2738 8465 32.K
TX49-32K  2.6.18-rc3-p   396 0.36 1.19 6.82 10.2 41.0 1.46 8.18 1330 4638 18.K

Original patch by me with enhancements by Atsushi Nemoto.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
---
 arch/mips/mm/c-r3k.c          |  2 +-
 arch/mips/mm/c-r4k.c          |  2 +-
 arch/mips/mm/c-sb1.c          |  2 +-
 arch/mips/mm/c-tx39.c         |  4 ++--
 arch/mips/mm/cache.c          | 18 ++++++++++--------
 include/asm-mips/cacheflush.h | 12 ++++++++----
 include/asm-mips/page.h       |  7 +++++--
 7 files changed, 28 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/arch/mips/mm/c-r3k.c b/arch/mips/mm/c-r3k.c
index bb041a22f20a..e1f35ef81145 100644
--- a/arch/mips/mm/c-r3k.c
+++ b/arch/mips/mm/c-r3k.c
@@ -335,7 +335,7 @@ void __init r3k_cache_init(void)
 	flush_cache_mm = r3k_flush_cache_mm;
 	flush_cache_range = r3k_flush_cache_range;
 	flush_cache_page = r3k_flush_cache_page;
-	flush_icache_page = r3k_flush_icache_page;
+	__flush_icache_page = r3k_flush_icache_page;
 	flush_icache_range = r3k_flush_icache_range;
 
 	flush_cache_sigtramp = r3k_flush_cache_sigtramp;
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index b3ccb8738d29..4e1498246343 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -1291,7 +1291,7 @@ void __init r4k_cache_init(void)
 	__flush_cache_all	= r4k___flush_cache_all;
 	flush_cache_mm		= r4k_flush_cache_mm;
 	flush_cache_page	= r4k_flush_cache_page;
-	flush_icache_page	= r4k_flush_icache_page;
+	__flush_icache_page	= r4k_flush_icache_page;
 	flush_cache_range	= r4k_flush_cache_range;
 
 	flush_cache_sigtramp	= r4k_flush_cache_sigtramp;
diff --git a/arch/mips/mm/c-sb1.c b/arch/mips/mm/c-sb1.c
index 2d71efb82ac5..4bd9ad8a5e07 100644
--- a/arch/mips/mm/c-sb1.c
+++ b/arch/mips/mm/c-sb1.c
@@ -520,7 +520,7 @@ void sb1_cache_init(void)
 
 	/* These routines are for Icache coherence with the Dcache */
 	flush_icache_range = sb1_flush_icache_range;
-	flush_icache_page = sb1_flush_icache_page;
+	__flush_icache_page = sb1_flush_icache_page;
 	flush_icache_all = __sb1_flush_icache_all; /* local only */
 
 	/* This implies an Icache flush too, so can't be nop'ed */
diff --git a/arch/mips/mm/c-tx39.c b/arch/mips/mm/c-tx39.c
index 5dfc9b1901f6..932a09d7ef84 100644
--- a/arch/mips/mm/c-tx39.c
+++ b/arch/mips/mm/c-tx39.c
@@ -382,7 +382,7 @@ void __init tx39_cache_init(void)
 		flush_cache_mm		= (void *) tx39h_flush_icache_all;
 		flush_cache_range	= (void *) tx39h_flush_icache_all;
 		flush_cache_page	= (void *) tx39h_flush_icache_all;
-		flush_icache_page	= (void *) tx39h_flush_icache_all;
+		__flush_icache_page	= (void *) tx39h_flush_icache_all;
 		flush_icache_range	= (void *) tx39h_flush_icache_all;
 
 		flush_cache_sigtramp	= (void *) tx39h_flush_icache_all;
@@ -408,7 +408,7 @@ void __init tx39_cache_init(void)
 		flush_cache_mm = tx39_flush_cache_mm;
 		flush_cache_range = tx39_flush_cache_range;
 		flush_cache_page = tx39_flush_cache_page;
-		flush_icache_page = tx39_flush_icache_page;
+		__flush_icache_page = tx39_flush_icache_page;
 		flush_icache_range = tx39_flush_icache_range;
 
 		flush_cache_sigtramp = tx39_flush_cache_sigtramp;
diff --git a/arch/mips/mm/cache.c b/arch/mips/mm/cache.c
index ddd3a2de1d73..40c8b0235183 100644
--- a/arch/mips/mm/cache.c
+++ b/arch/mips/mm/cache.c
@@ -25,7 +25,7 @@ void (*flush_cache_range)(struct vm_area_struct *vma, unsigned long start,
 void (*flush_cache_page)(struct vm_area_struct *vma, unsigned long page,
 	unsigned long pfn);
 void (*flush_icache_range)(unsigned long start, unsigned long end);
-void (*flush_icache_page)(struct vm_area_struct *vma, struct page *page);
+void (*__flush_icache_page)(struct vm_area_struct *vma, struct page *page);
 
 /* MIPS specific cache operations */
 void (*flush_cache_sigtramp)(unsigned long addr);
@@ -70,6 +70,8 @@ void __flush_dcache_page(struct page *page)
 	struct address_space *mapping = page_mapping(page);
 	unsigned long addr;
 
+	if (PageHighMem(page))
+		return;
 	if (mapping && !mapping_mapped(mapping)) {
 		SetPageDcacheDirty(page);
 		return;
@@ -91,16 +93,16 @@ void __update_cache(struct vm_area_struct *vma, unsigned long address,
 {
 	struct page *page;
 	unsigned long pfn, addr;
+	int exec = (vma->vm_flags & VM_EXEC) && !cpu_has_ic_fills_f_dc;
 
 	pfn = pte_pfn(pte);
-	if (pfn_valid(pfn) && (page = pfn_to_page(pfn), page_mapping(page)) &&
-	    Page_dcache_dirty(page)) {
-		if (pages_do_alias((unsigned long)page_address(page),
-		                   address & PAGE_MASK)) {
-			addr = (unsigned long) page_address(page);
+	if (unlikely(!pfn_valid(pfn)))
+		return;
+	page = pfn_to_page(pfn);
+	if (page_mapping(page) && Page_dcache_dirty(page)) {
+		addr = (unsigned long) page_address(page);
+		if (exec || pages_do_alias(addr, address & PAGE_MASK))
 			flush_data_cache_page(addr);
-		}
-
 		ClearPageDcacheDirty(page);
 	}
 }
diff --git a/include/asm-mips/cacheflush.h b/include/asm-mips/cacheflush.h
index 47bc8f6c20d2..36416fdfcf68 100644
--- a/include/asm-mips/cacheflush.h
+++ b/include/asm-mips/cacheflush.h
@@ -21,7 +21,6 @@
  *  - flush_cache_range(vma, start, end) flushes a range of pages
  *  - flush_icache_range(start, end) flush a range of instructions
  *  - flush_dcache_page(pg) flushes(wback&invalidates) a page for dcache
- *  - flush_icache_page(vma, pg) flushes(invalidates) a page for icache
  *
  * MIPS specific flush operations:
  *
@@ -39,7 +38,7 @@ extern void __flush_dcache_page(struct page *page);
 
 static inline void flush_dcache_page(struct page *page)
 {
-	if (cpu_has_dc_aliases)
+	if (cpu_has_dc_aliases || !cpu_has_ic_fills_f_dc)
 		__flush_dcache_page(page);
 
 }
@@ -47,8 +46,13 @@ static inline void flush_dcache_page(struct page *page)
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 
-extern void (*flush_icache_page)(struct vm_area_struct *vma,
+extern void (*__flush_icache_page)(struct vm_area_struct *vma,
 	struct page *page);
+static inline void flush_icache_page(struct vm_area_struct *vma,
+	struct page *page)
+{
+}
+
 extern void (*flush_icache_range)(unsigned long start, unsigned long end);
 #define flush_cache_vmap(start, end)		flush_cache_all()
 #define flush_cache_vunmap(start, end)		flush_cache_all()
@@ -60,7 +64,7 @@ static inline void copy_to_user_page(struct vm_area_struct *vma,
 	if (cpu_has_dc_aliases)
 		flush_cache_page(vma, vaddr, page_to_pfn(page));
 	memcpy(dst, src, len);
-	flush_icache_page(vma, page);
+	__flush_icache_page(vma, page);
 }
 
 static inline void copy_from_user_page(struct vm_area_struct *vma,
diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h
index 219d359861f3..c3b872b047d7 100644
--- a/include/asm-mips/page.h
+++ b/include/asm-mips/page.h
@@ -34,6 +34,8 @@
 
 #ifndef __ASSEMBLY__
 
+#include <asm/cpu-features.h>
+
 extern void clear_page(void * page);
 extern void copy_page(void * to, void * from);
 
@@ -53,7 +55,7 @@ static inline void clear_user_page(void *addr, unsigned long vaddr,
 	extern void (*flush_data_cache_page)(unsigned long addr);
 
 	clear_page(addr);
-	if (pages_do_alias((unsigned long) addr, vaddr))
+	if (pages_do_alias((unsigned long) addr, vaddr & PAGE_MASK))
 		flush_data_cache_page((unsigned long)addr);
 }
 
@@ -63,7 +65,8 @@ static inline void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
 	extern void (*flush_data_cache_page)(unsigned long addr);
 
 	copy_page(vto, vfrom);
-	if (pages_do_alias((unsigned long)vto, vaddr))
+	if (!cpu_has_ic_fills_f_dc ||
+	    pages_do_alias((unsigned long)vto, vaddr & PAGE_MASK))
 		flush_data_cache_page((unsigned long)vto);
 }
 
-- 
cgit v1.2.3


From 65316fd13ad9d82560edbad0a940d684380f7461 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 31 Aug 2006 14:16:06 +0100
Subject: [MIPS] Replace generic__raw_read_trylock usage

generic__raw_read_trylock() is a defect generic function actually doing
a __raw_read_lock ...

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/spinlock.h | 47 ++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 46 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-mips/spinlock.h b/include/asm-mips/spinlock.h
index 669b8e349ff2..4c1a1b53aeaf 100644
--- a/include/asm-mips/spinlock.h
+++ b/include/asm-mips/spinlock.h
@@ -239,7 +239,51 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw)
 	: "memory");
 }
 
-#define __raw_read_trylock(lock) generic__raw_read_trylock(lock)
+static inline int __raw_read_trylock(raw_rwlock_t *rw)
+{
+	unsigned int tmp;
+	int ret;
+
+	if (R10000_LLSC_WAR) {
+		__asm__ __volatile__(
+		"	.set	noreorder	# __raw_read_trylock	\n"
+		"	li	%2, 0					\n"
+		"1:	ll	%1, %3					\n"
+		"	bnez	%1, 2f					\n"
+		"	 addu	%1, 1					\n"
+		"	sc	%1, %0					\n"
+		"	beqzl	%1, 1b					\n"
+		"	.set	reorder					\n"
+#ifdef CONFIG_SMP
+		"	 sync						\n"
+#endif
+		"	li	%2, 1					\n"
+		"2:							\n"
+		: "=m" (rw->lock), "=&r" (tmp), "=&r" (ret)
+		: "m" (rw->lock)
+		: "memory");
+	} else {
+		__asm__ __volatile__(
+		"	.set	noreorder	# __raw_read_trylock	\n"
+		"	li	%2, 0					\n"
+		"1:	ll	%1, %3					\n"
+		"	bnez	%1, 2f					\n"
+		"	 addu	%1, 1					\n"
+		"	sc	%1, %0					\n"
+		"	beqz	%1, 1b					\n"
+		"	.set	reorder					\n"
+#ifdef CONFIG_SMP
+		"	 sync						\n"
+#endif
+		"	li	%2, 1					\n"
+		"2:							\n"
+		: "=m" (rw->lock), "=&r" (tmp), "=&r" (ret)
+		: "m" (rw->lock)
+		: "memory");
+	}
+
+	return ret;
+}
 
 static inline int __raw_write_trylock(raw_rwlock_t *rw)
 {
@@ -283,4 +327,5 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 	return ret;
 }
 
+
 #endif /* _ASM_SPINLOCK_H */
-- 
cgit v1.2.3


From d34555fb20e7abf33f86d7aa3ec0826343f38256 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 31 Aug 2006 19:39:09 +0100
Subject: [MIPS] Do not lose upper 32-bit on MIPS32 with 64-bit addresses in
 __pte().

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/page.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h
index c3b872b047d7..85b258ee7090 100644
--- a/include/asm-mips/page.h
+++ b/include/asm-mips/page.h
@@ -77,15 +77,17 @@ static inline void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
   #ifdef CONFIG_CPU_MIPS32
     typedef struct { unsigned long pte_low, pte_high; } pte_t;
     #define pte_val(x)    ((x).pte_low | ((unsigned long long)(x).pte_high << 32))
+    #define __pte(x)      ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; })
   #else
      typedef struct { unsigned long long pte; } pte_t;
      #define pte_val(x)	((x).pte)
+     #define __pte(x)	((pte_t) { (x) } )
   #endif
 #else
 typedef struct { unsigned long pte; } pte_t;
 #define pte_val(x)	((x).pte)
-#endif
 #define __pte(x)	((pte_t) { (x) } )
+#endif
 
 /*
  * For 3-level pagetables we defines these ourselves, for 2-level the
-- 
cgit v1.2.3


From 8f9a2b324644d3f8c233287f0a7764d61655cda4 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Thu, 7 Sep 2006 01:00:22 +0900
Subject: [MIPS] Fix errors detected by "make headers_check"

* export asm/sgidefs.h
* include asm/isadep.h only if in kernel
* do not export contents of asm/timex.h and asm/user.h

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/Kbuild   | 2 ++
 include/asm-mips/ptrace.h | 3 +--
 include/asm-mips/timex.h  | 4 ++++
 include/asm-mips/user.h   | 4 ++++
 4 files changed, 11 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/Kbuild b/include/asm-mips/Kbuild
index c68e1680da01..01ea3e045652 100644
--- a/include/asm-mips/Kbuild
+++ b/include/asm-mips/Kbuild
@@ -1 +1,3 @@
 include include/asm-generic/Kbuild.asm
+
+header-y += sgidefs.h
diff --git a/include/asm-mips/ptrace.h b/include/asm-mips/ptrace.h
index 4113316ee0da..4fb0fc43ffd7 100644
--- a/include/asm-mips/ptrace.h
+++ b/include/asm-mips/ptrace.h
@@ -10,8 +10,6 @@
 #define _ASM_PTRACE_H
 
 
-#include <asm/isadep.h>
-
 /* 0 - 31 are integer registers, 32 - 63 are fp registers.  */
 #define FPR_BASE	32
 #define PC		64
@@ -73,6 +71,7 @@ struct pt_regs {
 #ifdef __KERNEL__
 
 #include <linux/linkage.h>
+#include <asm/isadep.h>
 
 /*
  * Does the process account for user or for system time?
diff --git a/include/asm-mips/timex.h b/include/asm-mips/timex.h
index 98aa737b34aa..b80de8e0fbbd 100644
--- a/include/asm-mips/timex.h
+++ b/include/asm-mips/timex.h
@@ -8,6 +8,8 @@
 #ifndef _ASM_TIMEX_H
 #define _ASM_TIMEX_H
 
+#ifdef __KERNEL__
+
 #include <asm/mipsregs.h>
 
 /*
@@ -51,4 +53,6 @@ static inline cycles_t get_cycles (void)
 	return read_c0_count();
 }
 
+#endif /* __KERNEL__ */
+
 #endif /*  _ASM_TIMEX_H */
diff --git a/include/asm-mips/user.h b/include/asm-mips/user.h
index 89bf8b4cab3c..61f2a093b91b 100644
--- a/include/asm-mips/user.h
+++ b/include/asm-mips/user.h
@@ -8,6 +8,8 @@
 #ifndef _ASM_USER_H
 #define _ASM_USER_H
 
+#ifdef __KERNEL__
+
 #include <asm/page.h>
 #include <asm/reg.h>
 
@@ -55,4 +57,6 @@ struct user {
 #define HOST_DATA_START_ADDR	(u.start_data)
 #define HOST_STACK_END_ADDR	(u.start_stack + u.u_ssize * NBPG)
 
+#endif /* __KERNEL__ */
+
 #endif /* _ASM_USER_H */
-- 
cgit v1.2.3


From 7fdeb048141b363a23b8cf6f6a226d74aca4d724 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 6 Sep 2006 22:42:02 +0900
Subject: [MIPS] Wire up set_robust_list(2) and get_robust_list(2)

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/scall32-o32.S |  2 ++
 arch/mips/kernel/scall64-64.S  |  2 ++
 arch/mips/kernel/scall64-n32.S |  4 +++-
 arch/mips/kernel/scall64-o32.S |  2 ++
 include/asm-mips/unistd.h      | 18 ++++++++++++------
 5 files changed, 21 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index ba1bcd83c7d3..e71785102206 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -662,6 +662,8 @@ einval:	li	v0, -EINVAL
 	sys	sys_tee			4
 	sys	sys_vmsplice		4
 	sys	sys_move_pages		6
+	sys	sys_set_robust_list	2
+	sys	sys_get_robust_list	3
 	.endm
 
 	/* We pre-compute the number of _instruction_ bytes needed to
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index 939e172db953..4c22d0b4825d 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -466,3 +466,5 @@ sys_call_table:
 	PTR	sys_tee				/* 5265 */
 	PTR	sys_vmsplice
 	PTR	sys_move_pages
+	PTR	sys_set_robust_list
+	PTR	sys_get_robust_list
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 549b4bcb53a5..f25c2a2f1038 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -390,5 +390,7 @@ EXPORT(sysn32_call_table)
 	PTR	sys_splice
 	PTR	sys_sync_file_range
 	PTR	sys_tee
-	PTR	sys_vmsplice			/* 6271 */
+	PTR	sys_vmsplice			/* 6270 */
 	PTR	sys_move_pages
+	PTR	compat_sys_set_robust_list
+	PTR	compat_sys_get_robust_list
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 505c9ee54009..2ac01412f747 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -514,4 +514,6 @@ sys_call_table:
 	PTR	sys_tee
 	PTR	sys_vmsplice
 	PTR	compat_sys_move_pages
+	PTR	compat_sys_set_robust_list
+	PTR	compat_sys_get_robust_list	/* 4310 */
 	.size	sys_call_table,.-sys_call_table
diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h
index 610ccb8a50b3..558e3cba4f92 100644
--- a/include/asm-mips/unistd.h
+++ b/include/asm-mips/unistd.h
@@ -329,16 +329,18 @@
 #define __NR_tee			(__NR_Linux + 306)
 #define __NR_vmsplice			(__NR_Linux + 307)
 #define __NR_move_pages			(__NR_Linux + 308)
+#define __NR_set_robust_list		(__NR_Linux + 309)
+#define __NR_get_robust_list		(__NR_Linux + 310)
 
 /*
  * Offset of the last Linux o32 flavoured syscall
  */
-#define __NR_Linux_syscalls		308
+#define __NR_Linux_syscalls		310
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */
 
 #define __NR_O32_Linux			4000
-#define __NR_O32_Linux_syscalls		308
+#define __NR_O32_Linux_syscalls		310
 
 #if _MIPS_SIM == _MIPS_SIM_ABI64
 
@@ -614,16 +616,18 @@
 #define __NR_tee			(__NR_Linux + 265)
 #define __NR_vmsplice			(__NR_Linux + 266)
 #define __NR_move_pages			(__NR_Linux + 267)
+#define __NR_set_robust_list		(__NR_Linux + 268)
+#define __NR_get_robust_list		(__NR_Linux + 269)
 
 /*
  * Offset of the last Linux 64-bit flavoured syscall
  */
-#define __NR_Linux_syscalls		267
+#define __NR_Linux_syscalls		269
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */
 
 #define __NR_64_Linux			5000
-#define __NR_64_Linux_syscalls		267
+#define __NR_64_Linux_syscalls		269
 
 #if _MIPS_SIM == _MIPS_SIM_NABI32
 
@@ -903,16 +907,18 @@
 #define __NR_tee			(__NR_Linux + 269)
 #define __NR_vmsplice			(__NR_Linux + 270)
 #define __NR_move_pages			(__NR_Linux + 271)
+#define __NR_set_robust_list		(__NR_Linux + 272)
+#define __NR_get_robust_list		(__NR_Linux + 273)
 
 /*
  * Offset of the last N32 flavoured syscall
  */
-#define __NR_Linux_syscalls		271
+#define __NR_Linux_syscalls		273
 
 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */
 
 #define __NR_N32_Linux			6000
-#define __NR_N32_Linux_syscalls		271
+#define __NR_N32_Linux_syscalls		273
 
 #ifdef __KERNEL__
 
-- 
cgit v1.2.3


From 3ee24e1b1e0b5ae413a85ba63677a7110915e3af Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@mips.com>
Date: Tue, 12 Sep 2006 19:02:44 +0100
Subject: [MIPS] Atlas: Fix building the RTC driver

Atlas maps its RTC chip in the host mmio space rather than using the
"traditional" location in the PCI/ISA port space.  A change that has
happened to the generic RTC header requires to define ARCH_RTC_LOCATION
now.

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mach-atlas/mc146818rtc.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/asm-mips/mach-atlas/mc146818rtc.h b/include/asm-mips/mach-atlas/mc146818rtc.h
index 397522ea5565..fc09928571d6 100644
--- a/include/asm-mips/mach-atlas/mc146818rtc.h
+++ b/include/asm-mips/mach-atlas/mc146818rtc.h
@@ -28,6 +28,8 @@
 #include <asm/mips-boards/atlas.h>
 #include <asm/mips-boards/atlasint.h>
 
+#define ARCH_RTC_LOCATION
+
 #define RTC_PORT(x)	(ATLAS_RTC_ADR_REG + (x) * 8)
 #define RTC_IO_EXTENT	0x100
 #define RTC_IOMAPPED	0
-- 
cgit v1.2.3


From fc095a902181b72ce77a10feb7b36ba1cbacd736 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@mips.com>
Date: Tue, 12 Sep 2006 19:12:18 +0100
Subject: [MIPS] Atlas: update interrupt handling

The following change updates the Atlas interrupt handling to match that
of Malta.  Tested with a 5Kc and a 34Kf successfully.

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mips-boards/atlas/atlas_int.c   | 105 ++++++++++++++++++++-----
 arch/mips/mips-boards/atlas/atlas_setup.c |   2 +-
 arch/mips/mips-boards/generic/time.c      |   7 +-
 arch/mips/pci/fixup-atlas.c               |  20 ++---
 include/asm-mips/mach-atlas/mc146818rtc.h |   2 +-
 include/asm-mips/mips-boards/atlasint.h   | 124 +++++++++++++++++++++---------
 6 files changed, 189 insertions(+), 71 deletions(-)

(limited to 'include')

diff --git a/arch/mips/mips-boards/atlas/atlas_int.c b/arch/mips/mips-boards/atlas/atlas_int.c
index fb25e0377f11..a020a3cb4f4b 100644
--- a/arch/mips/mips-boards/atlas/atlas_int.c
+++ b/arch/mips/mips-boards/atlas/atlas_int.c
@@ -1,6 +1,8 @@
 /*
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 1999, 2000, 2006  MIPS Technologies, Inc.
+ *	All rights reserved.
+ *	Authors: Carsten Langgaard <carstenl@mips.com>
+ *		 Maciej W. Rozycki <macro@mips.com>
  *
  * ########################################################################
  *
@@ -25,17 +27,20 @@
  */
 #include <linux/compiler.h>
 #include <linux/init.h>
+#include <linux/irq.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
 
-#include <asm/irq.h>
+#include <asm/gdb-stub.h>
 #include <asm/io.h>
+#include <asm/irq_cpu.h>
+#include <asm/msc01_ic.h>
+
 #include <asm/mips-boards/atlas.h>
 #include <asm/mips-boards/atlasint.h>
-#include <asm/gdb-stub.h>
-
+#include <asm/mips-boards/generic.h>
 
 static struct atlas_ictrl_regs *atlas_hw0_icregs;
 
@@ -47,13 +52,13 @@ static struct atlas_ictrl_regs *atlas_hw0_icregs;
 
 void disable_atlas_irq(unsigned int irq_nr)
 {
-	atlas_hw0_icregs->intrsten = (1 << (irq_nr-ATLASINT_BASE));
+	atlas_hw0_icregs->intrsten = 1 << (irq_nr - ATLAS_INT_BASE);
 	iob();
 }
 
 void enable_atlas_irq(unsigned int irq_nr)
 {
-	atlas_hw0_icregs->intseten = (1 << (irq_nr-ATLASINT_BASE));
+	atlas_hw0_icregs->intseten = 1 << (irq_nr - ATLAS_INT_BASE);
 	iob();
 }
 
@@ -107,7 +112,7 @@ static inline void atlas_hw0_irqdispatch(struct pt_regs *regs)
 	if (unlikely(int_status == 0))
 		return;
 
-	irq = ATLASINT_BASE + ls1bit32(int_status);
+	irq = ATLAS_INT_BASE + ls1bit32(int_status);
 
 	DEBUG_INT("atlas_hw0_irqdispatch: irq=%d\n", irq);
 
@@ -161,15 +166,14 @@ static inline unsigned int irq_ffs(unsigned int pending)
 }
 
 /*
- * IRQs on the Atlas board look basically (barring software IRQs which we
- * don't use at all and all external interrupt sources are combined together
- * on hardware interrupt 0 (MIPS IRQ 2)) like:
+ * IRQs on the Atlas board look basically like (all external interrupt
+ * sources are combined together on hardware interrupt 0 (MIPS IRQ 2)):
  *
- *	MIPS IRQ	Source
+ *      MIPS IRQ        Source
  *      --------        ------
- *             0	Software (ignored)
- *             1        Software (ignored)
- *             2        Combined hardware interrupt (hw0)
+ *             0        Software 0 (reschedule IPI on MT)
+ *             1        Software 1 (remote call IPI on MT)
+ *             2        Combined Atlas hardware interrupt (hw0)
  *             3        Hardware (ignored)
  *             4        Hardware (ignored)
  *             5        Hardware (ignored)
@@ -179,7 +183,7 @@ static inline unsigned int irq_ffs(unsigned int pending)
  * We handle the IRQ according to _our_ priority which is:
  *
  * Highest ----     R4k Timer
- * Lowest  ----     Combined hardware interrupt
+ * Lowest  ----     Software 0
  *
  * then we just return, if multiple IRQs are pending then we will just take
  * another exception, big deal.
@@ -193,17 +197,19 @@ asmlinkage void plat_irq_dispatch(struct pt_regs *regs)
 
 	if (irq == MIPSCPU_INT_ATLAS)
 		atlas_hw0_irqdispatch(regs);
-	else if (irq > 0)
+	else if (irq >= 0)
 		do_IRQ(MIPSCPU_INT_BASE + irq, regs);
 	else
 		spurious_interrupt(regs);
 }
 
-void __init arch_init_irq(void)
+static inline void init_atlas_irqs (int base)
 {
 	int i;
 
-	atlas_hw0_icregs = (struct atlas_ictrl_regs *)ioremap (ATLAS_ICTRL_REGS_BASE, sizeof(struct atlas_ictrl_regs *));
+	atlas_hw0_icregs = (struct atlas_ictrl_regs *)
+			   ioremap(ATLAS_ICTRL_REGS_BASE,
+				   sizeof(struct atlas_ictrl_regs *));
 
 	/*
 	 * Mask out all interrupt by writing "1" to all bit position in
@@ -211,7 +217,7 @@ void __init arch_init_irq(void)
 	 */
 	atlas_hw0_icregs->intrsten = 0xffffffff;
 
-	for (i = ATLASINT_BASE; i <= ATLASINT_END; i++) {
+	for (i = ATLAS_INT_BASE; i <= ATLAS_INT_END; i++) {
 		irq_desc[i].status	= IRQ_DISABLED;
 		irq_desc[i].action	= 0;
 		irq_desc[i].depth	= 1;
@@ -219,3 +225,62 @@ void __init arch_init_irq(void)
 		spin_lock_init(&irq_desc[i].lock);
 	}
 }
+
+static struct irqaction atlasirq = {
+	.handler = no_action,
+	.name = "Atlas cascade"
+};
+
+msc_irqmap_t __initdata msc_irqmap[] = {
+	{MSC01C_INT_TMR,		MSC01_IRQ_EDGE, 0},
+	{MSC01C_INT_PCI,		MSC01_IRQ_LEVEL, 0},
+};
+int __initdata msc_nr_irqs = sizeof(msc_irqmap) / sizeof(*msc_irqmap);
+
+msc_irqmap_t __initdata msc_eicirqmap[] = {
+	{MSC01E_INT_SW0,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_SW1,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_ATLAS,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_TMR,		MSC01_IRQ_EDGE, 0},
+	{MSC01E_INT_PCI,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_PERFCTR,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_CPUCTR,		MSC01_IRQ_LEVEL, 0}
+};
+int __initdata msc_nr_eicirqs = sizeof(msc_eicirqmap) / sizeof(*msc_eicirqmap);
+
+void __init arch_init_irq(void)
+{
+	init_atlas_irqs(ATLAS_INT_BASE);
+
+	if (!cpu_has_veic)
+		mips_cpu_irq_init(MIPSCPU_INT_BASE);
+
+	switch(mips_revision_corid) {
+	case MIPS_REVISION_CORID_CORE_MSC:
+	case MIPS_REVISION_CORID_CORE_FPGA2:
+	case MIPS_REVISION_CORID_CORE_FPGA3:
+	case MIPS_REVISION_CORID_CORE_24K:
+	case MIPS_REVISION_CORID_CORE_EMUL_MSC:
+		if (cpu_has_veic)
+			init_msc_irqs (MSC01E_INT_BASE,
+				       msc_eicirqmap, msc_nr_eicirqs);
+		else
+			init_msc_irqs (MSC01C_INT_BASE,
+				       msc_irqmap, msc_nr_irqs);
+	}
+
+
+	if (cpu_has_veic) {
+		set_vi_handler (MSC01E_INT_ATLAS, atlas_hw0_irqdispatch);
+		setup_irq (MSC01E_INT_BASE + MSC01E_INT_ATLAS, &atlasirq);
+	} else if (cpu_has_vint) {
+		set_vi_handler (MIPSCPU_INT_ATLAS, atlas_hw0_irqdispatch);
+#ifdef CONFIG_MIPS_MT_SMTC
+		setup_irq_smtc (MIPSCPU_INT_BASE + MIPSCPU_INT_ATLAS,
+				&atlasirq, (0x100 << MIPSCPU_INT_ATLAS));
+#else /* Not SMTC */
+		setup_irq(MIPSCPU_INT_BASE + MIPSCPU_INT_ATLAS, &atlasirq);
+#endif /* CONFIG_MIPS_MT_SMTC */
+	} else
+		setup_irq(MIPSCPU_INT_BASE + MIPSCPU_INT_ATLAS, &atlasirq);
+}
diff --git a/arch/mips/mips-boards/atlas/atlas_setup.c b/arch/mips/mips-boards/atlas/atlas_setup.c
index 9871a91fdb07..0c6b0ce15028 100644
--- a/arch/mips/mips-boards/atlas/atlas_setup.c
+++ b/arch/mips/mips-boards/atlas/atlas_setup.c
@@ -77,7 +77,7 @@ static void __init serial_init(void)
 #else
 	s.iobase = ATLAS_UART_REGS_BASE+3;
 #endif
-	s.irq = ATLASINT_UART;
+	s.irq = ATLAS_INT_UART;
 	s.uartclk = ATLAS_BASE_BAUD * 16;
 	s.flags = UPF_BOOT_AUTOCONF | UPF_SKIP_TEST | UPF_AUTO_IRQ;
 	s.iotype = UPIO_PORT;
diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c
index de5798e1e015..8d15861fce61 100644
--- a/arch/mips/mips-boards/generic/time.c
+++ b/arch/mips/mips-boards/generic/time.c
@@ -41,8 +41,13 @@
 
 #include <asm/mips-boards/generic.h>
 #include <asm/mips-boards/prom.h>
+
+#ifdef CONFIG_MIPS_ATLAS
+#include <asm/mips-boards/atlasint.h>
+#endif
+#ifdef CONFIG_MIPS_MALTA
 #include <asm/mips-boards/maltaint.h>
-#include <asm/mc146818-time.h>
+#endif
 
 unsigned long cpu_khz;
 
diff --git a/arch/mips/pci/fixup-atlas.c b/arch/mips/pci/fixup-atlas.c
index 439510af3037..c6cd6e9cdfbc 100644
--- a/arch/mips/pci/fixup-atlas.c
+++ b/arch/mips/pci/fixup-atlas.c
@@ -21,16 +21,16 @@
 
 #include <asm/mips-boards/atlasint.h>
 
-#define PCIA		ATLASINT_PCIA
-#define PCIB		ATLASINT_PCIB
-#define PCIC		ATLASINT_PCIC
-#define PCID		ATLASINT_PCID
-#define INTA		ATLASINT_INTA
-#define INTB		ATLASINT_INTB
-#define ETH		ATLASINT_ETH
-#define INTC		ATLASINT_INTC
-#define SCSI		ATLASINT_SCSI
-#define INTD		ATLASINT_INTD
+#define PCIA		ATLAS_INT_PCIA
+#define PCIB		ATLAS_INT_PCIB
+#define PCIC		ATLAS_INT_PCIC
+#define PCID		ATLAS_INT_PCID
+#define INTA		ATLAS_INT_INTA
+#define INTB		ATLAS_INT_INTB
+#define ETH		ATLAS_INT_ETH
+#define INTC		ATLAS_INT_INTC
+#define SCSI		ATLAS_INT_SCSI
+#define INTD		ATLAS_INT_INTD
 
 static char irq_tab[][5] __initdata = {
 	/*      INTA    INTB    INTC    INTD */
diff --git a/include/asm-mips/mach-atlas/mc146818rtc.h b/include/asm-mips/mach-atlas/mc146818rtc.h
index fc09928571d6..a73a5698420c 100644
--- a/include/asm-mips/mach-atlas/mc146818rtc.h
+++ b/include/asm-mips/mach-atlas/mc146818rtc.h
@@ -33,7 +33,7 @@
 #define RTC_PORT(x)	(ATLAS_RTC_ADR_REG + (x) * 8)
 #define RTC_IO_EXTENT	0x100
 #define RTC_IOMAPPED	0
-#define RTC_IRQ		ATLASINT_RTC
+#define RTC_IRQ		ATLAS_INT_RTC
 
 static inline unsigned char CMOS_READ(unsigned long addr)
 {
diff --git a/include/asm-mips/mips-boards/atlasint.h b/include/asm-mips/mips-boards/atlasint.h
index fd7ebc54fa90..b15e4ea0b091 100644
--- a/include/asm-mips/mips-boards/atlasint.h
+++ b/include/asm-mips/mips-boards/atlasint.h
@@ -1,6 +1,7 @@
 /*
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 1999, 2006  MIPS Technologies, Inc.  All rights reserved.
+ *	Authors: Carsten Langgaard <carstenl@mips.com>
+ *		 Maciej W. Rozycki <macro@mips.com>
  *
  * ########################################################################
  *
@@ -25,41 +26,88 @@
 #ifndef _MIPS_ATLASINT_H
 #define _MIPS_ATLASINT_H
 
-#define ATLASINT_BASE		1
-#define ATLASINT_UART		(ATLASINT_BASE+0)
-#define ATLASINT_TIM0		(ATLASINT_BASE+1)
-#define ATLASINT_RES2		(ATLASINT_BASE+2)
-#define ATLASINT_RES3		(ATLASINT_BASE+3)
-#define ATLASINT_RTC		(ATLASINT_BASE+4)
-#define ATLASINT_COREHI		(ATLASINT_BASE+5)
-#define ATLASINT_CORELO		(ATLASINT_BASE+6)
-#define ATLASINT_RES7		(ATLASINT_BASE+7)
-#define ATLASINT_PCIA		(ATLASINT_BASE+8)
-#define ATLASINT_PCIB		(ATLASINT_BASE+9)
-#define ATLASINT_PCIC		(ATLASINT_BASE+10)
-#define ATLASINT_PCID		(ATLASINT_BASE+11)
-#define ATLASINT_ENUM		(ATLASINT_BASE+12)
-#define ATLASINT_DEG		(ATLASINT_BASE+13)
-#define ATLASINT_ATXFAIL	(ATLASINT_BASE+14)
-#define ATLASINT_INTA		(ATLASINT_BASE+15)
-#define ATLASINT_INTB		(ATLASINT_BASE+16)
-#define ATLASINT_ETH		ATLASINT_INTB
-#define ATLASINT_INTC		(ATLASINT_BASE+17)
-#define ATLASINT_SCSI		ATLASINT_INTC
-#define ATLASINT_INTD		(ATLASINT_BASE+18)
-#define ATLASINT_SERR		(ATLASINT_BASE+19)
-#define ATLASINT_RES20		(ATLASINT_BASE+20)
-#define ATLASINT_RES21		(ATLASINT_BASE+21)
-#define ATLASINT_RES22		(ATLASINT_BASE+22)
-#define ATLASINT_RES23		(ATLASINT_BASE+23)
-#define ATLASINT_RES24		(ATLASINT_BASE+24)
-#define ATLASINT_RES25		(ATLASINT_BASE+25)
-#define ATLASINT_RES26		(ATLASINT_BASE+26)
-#define ATLASINT_RES27		(ATLASINT_BASE+27)
-#define ATLASINT_RES28		(ATLASINT_BASE+28)
-#define ATLASINT_RES29		(ATLASINT_BASE+29)
-#define ATLASINT_RES30		(ATLASINT_BASE+30)
-#define ATLASINT_RES31		(ATLASINT_BASE+31)
-#define ATLASINT_END		(ATLASINT_BASE+31)
+/*
+ * Interrupts 0..7 are used for Atlas CPU interrupts (nonEIC mode)
+ */
+#define MIPSCPU_INT_BASE	0
+
+/* CPU interrupt offsets */
+#define MIPSCPU_INT_SW0		0
+#define MIPSCPU_INT_SW1		1
+#define MIPSCPU_INT_MB0		2
+#define MIPSCPU_INT_ATLAS	MIPSCPU_INT_MB0
+#define MIPSCPU_INT_MB1		3
+#define MIPSCPU_INT_MB2		4
+#define MIPSCPU_INT_MB3		5
+#define MIPSCPU_INT_MB4		6
+#define MIPSCPU_INT_CPUCTR	7
+
+/*
+ * Interrupts 8..39 are used for Atlas interrupt controller interrupts
+ */
+#define ATLAS_INT_BASE		8
+#define ATLAS_INT_UART		(ATLAS_INT_BASE + 0)
+#define ATLAS_INT_TIM0		(ATLAS_INT_BASE + 1)
+#define ATLAS_INT_RES2		(ATLAS_INT_BASE + 2)
+#define ATLAS_INT_RES3		(ATLAS_INT_BASE + 3)
+#define ATLAS_INT_RTC		(ATLAS_INT_BASE + 4)
+#define ATLAS_INT_COREHI	(ATLAS_INT_BASE + 5)
+#define ATLAS_INT_CORELO	(ATLAS_INT_BASE + 6)
+#define ATLAS_INT_RES7		(ATLAS_INT_BASE + 7)
+#define ATLAS_INT_PCIA		(ATLAS_INT_BASE + 8)
+#define ATLAS_INT_PCIB		(ATLAS_INT_BASE + 9)
+#define ATLAS_INT_PCIC		(ATLAS_INT_BASE + 10)
+#define ATLAS_INT_PCID		(ATLAS_INT_BASE + 11)
+#define ATLAS_INT_ENUM		(ATLAS_INT_BASE + 12)
+#define ATLAS_INT_DEG		(ATLAS_INT_BASE + 13)
+#define ATLAS_INT_ATXFAIL	(ATLAS_INT_BASE + 14)
+#define ATLAS_INT_INTA		(ATLAS_INT_BASE + 15)
+#define ATLAS_INT_INTB		(ATLAS_INT_BASE + 16)
+#define ATLAS_INT_ETH		ATLAS_INT_INTB
+#define ATLAS_INT_INTC		(ATLAS_INT_BASE + 17)
+#define ATLAS_INT_SCSI		ATLAS_INT_INTC
+#define ATLAS_INT_INTD		(ATLAS_INT_BASE + 18)
+#define ATLAS_INT_SERR		(ATLAS_INT_BASE + 19)
+#define ATLAS_INT_RES20		(ATLAS_INT_BASE + 20)
+#define ATLAS_INT_RES21		(ATLAS_INT_BASE + 21)
+#define ATLAS_INT_RES22		(ATLAS_INT_BASE + 22)
+#define ATLAS_INT_RES23		(ATLAS_INT_BASE + 23)
+#define ATLAS_INT_RES24		(ATLAS_INT_BASE + 24)
+#define ATLAS_INT_RES25		(ATLAS_INT_BASE + 25)
+#define ATLAS_INT_RES26		(ATLAS_INT_BASE + 26)
+#define ATLAS_INT_RES27		(ATLAS_INT_BASE + 27)
+#define ATLAS_INT_RES28		(ATLAS_INT_BASE + 28)
+#define ATLAS_INT_RES29		(ATLAS_INT_BASE + 29)
+#define ATLAS_INT_RES30		(ATLAS_INT_BASE + 30)
+#define ATLAS_INT_RES31		(ATLAS_INT_BASE + 31)
+#define ATLAS_INT_END		(ATLAS_INT_BASE + 31)
+
+/*
+ * Interrupts 64..127 are used for Soc-it Classic interrupts
+ */
+#define MSC01C_INT_BASE		64
+
+/* SOC-it Classic interrupt offsets */
+#define MSC01C_INT_TMR		0
+#define MSC01C_INT_PCI		1
+
+/*
+ * Interrupts 64..127 are used for Soc-it EIC interrupts
+ */
+#define MSC01E_INT_BASE		64
+
+/* SOC-it EIC interrupt offsets */
+#define	MSC01E_INT_SW0		1
+#define	MSC01E_INT_SW1		2
+#define	MSC01E_INT_MB0		3
+#define	MSC01E_INT_ATLAS	MSC01E_INT_MB0
+#define	MSC01E_INT_MB1		4
+#define	MSC01E_INT_MB2		5
+#define	MSC01E_INT_MB3		6
+#define	MSC01E_INT_MB4		7
+#define	MSC01E_INT_TMR		8
+#define	MSC01E_INT_PCI		9
+#define	MSC01E_INT_PERFCTR	10
+#define	MSC01E_INT_CPUCTR	11
 
 #endif /* !(_MIPS_ATLASINT_H) */
-- 
cgit v1.2.3


From 6b3e5f44b56745daad8cd913ccc7bcd9a9ece5ea Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Mon, 31 Jul 2006 23:01:37 +0900
Subject: [MIPS] Use common definitions from asm-generic/signal.h

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/signal.h | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/signal.h b/include/asm-mips/signal.h
index 87a1dff95199..5fc3a599d014 100644
--- a/include/asm-mips/signal.h
+++ b/include/asm-mips/signal.h
@@ -111,14 +111,7 @@ typedef unsigned long old_sigset_t;		/* at least 32 bits */
 #define SIG_SETMASK32	256	/* Goodie from SGI for BSD compatibility:
 				   set only the low 32 bit of the sigset.  */
 
-/* Type of a signal handler.  */
-typedef void __signalfn_t(int);
-typedef __signalfn_t __user *__sighandler_t;
-
-/* Fake signal functions */
-#define SIG_DFL	((__sighandler_t)0)	/* default signal handling */
-#define SIG_IGN	((__sighandler_t)1)	/* ignore signal */
-#define SIG_ERR	((__sighandler_t)-1)	/* error return from signal */
+#include <asm-generic/signal.h>
 
 struct sigaction {
 	unsigned int	sa_flags;
-- 
cgit v1.2.3


From 633fd568c1db825b9eb563681af09260ecee87b6 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 4 Aug 2006 01:49:31 +0100
Subject: [MIPS] Move definition of IRIX compat constant into IRIX compat code.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/irixsig.c | 3 +++
 include/asm-mips/signal.h  | 2 --
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/mips/kernel/irixsig.c b/arch/mips/kernel/irixsig.c
index 719364752e77..2132485caa74 100644
--- a/arch/mips/kernel/irixsig.c
+++ b/arch/mips/kernel/irixsig.c
@@ -323,6 +323,9 @@ struct sigact_irix5 {
 	int _unused0[2];
 };
 
+#define SIG_SETMASK32	256	/* Goodie from SGI for BSD compatibility:
+				   set only the low 32 bit of the sigset.  */
+
 #ifdef DEBUG_SIG
 static inline void dump_sigact_irix5(struct sigact_irix5 *p)
 {
diff --git a/include/asm-mips/signal.h b/include/asm-mips/signal.h
index 5fc3a599d014..8b391a2f0814 100644
--- a/include/asm-mips/signal.h
+++ b/include/asm-mips/signal.h
@@ -108,8 +108,6 @@ typedef unsigned long old_sigset_t;		/* at least 32 bits */
 #define SIG_BLOCK	1	/* for blocking signals */
 #define SIG_UNBLOCK	2	/* for unblocking signals */
 #define SIG_SETMASK	3	/* for setting the signal mask */
-#define SIG_SETMASK32	256	/* Goodie from SGI for BSD compatibility:
-				   set only the low 32 bit of the sigset.  */
 
 #include <asm-generic/signal.h>
 
-- 
cgit v1.2.3


From bdb37c8d63f9455cd2ea505d62d0d6de221f3c76 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 16 Aug 2006 23:10:00 +0900
Subject: [MIPS] Remove F_SETSIG and F_GETSIG in favor of the asm-generic
 definitions.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/fcntl.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/fcntl.h b/include/asm-mips/fcntl.h
index 787220e6c1fc..00a50ec1c19f 100644
--- a/include/asm-mips/fcntl.h
+++ b/include/asm-mips/fcntl.h
@@ -25,8 +25,6 @@
 
 #define F_SETOWN	24	/*  for sockets. */
 #define F_GETOWN	23	/*  for sockets. */
-#define F_SETSIG	10	/*  for sockets. */
-#define F_GETSIG	11	/*  for sockets. */
 
 #ifndef __mips64
 #define F_GETLK64	33	/*  using 'struct flock64' */
-- 
cgit v1.2.3


From 3c70f12bfaec6e1a1c4bfb94aec0c17675bc9310 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sun, 20 Aug 2006 00:33:38 +0900
Subject: [MIPS] Qemu does not have D-cache aliases

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mach-qemu/cpu-feature-overrides.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-mips/mach-qemu/cpu-feature-overrides.h b/include/asm-mips/mach-qemu/cpu-feature-overrides.h
index f4e370e27168..529445dacedb 100644
--- a/include/asm-mips/mach-qemu/cpu-feature-overrides.h
+++ b/include/asm-mips/mach-qemu/cpu-feature-overrides.h
@@ -20,7 +20,7 @@
 
 #define cpu_has_llsc		1
 #define cpu_has_vtag_icache	0
-#define cpu_has_dc_aliases	(PAGE_SIZE < 0x4000)
+#define cpu_has_dc_aliases	0
 #define cpu_has_ic_fills_f_dc	0
 
 #define cpu_has_dsp		0
-- 
cgit v1.2.3


From 6b8aab09309c577318f8484fce401e5c3546bb3b Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 25 Aug 2006 12:34:33 +0100
Subject: [MIPS] Reformat missformated SMTC bits.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mmu_context.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/mmu_context.h b/include/asm-mips/mmu_context.h
index 18b69de87daa..fe065d6070ca 100644
--- a/include/asm-mips/mmu_context.h
+++ b/include/asm-mips/mmu_context.h
@@ -262,10 +262,10 @@ drop_mmu_context(struct mm_struct *mm, unsigned cpu)
 		/* See comments for similar code above */
 		prevvpe = dvpe();
 		oldasid = (read_c0_entryhi() & ASID_MASK);
-		if(smtc_live_asid[mytlb][oldasid]) {
-		  smtc_live_asid[mytlb][oldasid] &= ~(0x1 << cpu);
-		  if(smtc_live_asid[mytlb][oldasid] == 0)
-			smtc_flush_tlb_asid(oldasid);
+		if (smtc_live_asid[mytlb][oldasid]) {
+			smtc_live_asid[mytlb][oldasid] &= ~(0x1 << cpu);
+			if(smtc_live_asid[mytlb][oldasid] == 0)
+				smtc_flush_tlb_asid(oldasid);
 		}
 		/* See comments for similar code above */
 		write_c0_entryhi((read_c0_entryhi() & ~HW_ASID_MASK)
-- 
cgit v1.2.3


From 0c68a9b6a7da0cc9095c117bea573f9058b00fff Mon Sep 17 00:00:00 2001
From: "thomas@koeller.dyndns.org" <thomas@koeller.dyndns.org>
Date: Sun, 27 Aug 2006 13:54:31 +0200
Subject: [MIPS] Move excite_fpga.h to include/asm-mips/mach-excite

excite_fpga.h, like all platform headers, really belongs in the
platform header directory.

Signed-off-by: Thomas Koeller <thomas.koeller@baslerweb.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/basler/excite/excite_fpga.h      | 80 ------------------------------
 include/asm-mips/mach-excite/excite_fpga.h | 80 ++++++++++++++++++++++++++++++
 2 files changed, 80 insertions(+), 80 deletions(-)
 delete mode 100644 arch/mips/basler/excite/excite_fpga.h
 create mode 100644 include/asm-mips/mach-excite/excite_fpga.h

(limited to 'include')

diff --git a/arch/mips/basler/excite/excite_fpga.h b/arch/mips/basler/excite/excite_fpga.h
deleted file mode 100644
index 38fcda703a0b..000000000000
--- a/arch/mips/basler/excite/excite_fpga.h
+++ /dev/null
@@ -1,80 +0,0 @@
-#ifndef EXCITE_FPGA_H_INCLUDED
-#define EXCITE_FPGA_H_INCLUDED
-
-
-/**
- * Adress alignment of the individual FPGA bytes.
- * The address arrangement of the individual bytes of the FPGA is two
- * byte aligned at the embedded MK2 platform.
- */
-#ifdef EXCITE_CCI_FPGA_MK2
-typedef unsigned char excite_cci_fpga_align_t __attribute__ ((aligned(2)));
-#else
-typedef unsigned char excite_cci_fpga_align_t;
-#endif
-
-
-/**
- * Size of Dual Ported RAM.
- */
-#define EXCITE_DPR_SIZE 263
-
-
-/**
- * Size of Reserved Status Fields in Dual Ported RAM.
- */
-#define EXCITE_DPR_STATUS_SIZE 7
-
-
-
-/**
- * FPGA.
- * Hardware register layout of the FPGA interface. The FPGA must accessed
- * byte wise solely.
- * @see EXCITE_CCI_DPR_MK2
- */
-typedef struct excite_fpga {
-
-	/**
-	 * Dual Ported RAM.
-	 */
-	excite_cci_fpga_align_t dpr[EXCITE_DPR_SIZE];
-
-	/**
-	 * Status.
-	 */
-	excite_cci_fpga_align_t status[EXCITE_DPR_STATUS_SIZE];
-
-#ifdef EXCITE_CCI_FPGA_MK2
-	/**
-	 * RM9000 Interrupt.
-	 * Write access initiates interrupt at the RM9000 (MIPS) processor of the eXcite.
-	 */
-	excite_cci_fpga_align_t rm9k_int;
-#else
-	/**
-	 * MK2 Interrupt.
-	 * Write access initiates interrupt at the ARM processor of the MK2.
-	 */
-	excite_cci_fpga_align_t mk2_int;
-
-	excite_cci_fpga_align_t gap[0x1000-0x10f];
-
-	/**
-	 * IRQ Source/Acknowledge.
-	 */
-	excite_cci_fpga_align_t rm9k_irq_src;
-
-	/**
-	 * IRQ Mask.
-	 * Set bits enable the related interrupt.
-	 */
-	excite_cci_fpga_align_t rm9k_irq_mask;
-#endif
-
-
-} excite_fpga;
-
-
-
-#endif	/* ndef EXCITE_FPGA_H_INCLUDED */
diff --git a/include/asm-mips/mach-excite/excite_fpga.h b/include/asm-mips/mach-excite/excite_fpga.h
new file mode 100644
index 000000000000..38fcda703a0b
--- /dev/null
+++ b/include/asm-mips/mach-excite/excite_fpga.h
@@ -0,0 +1,80 @@
+#ifndef EXCITE_FPGA_H_INCLUDED
+#define EXCITE_FPGA_H_INCLUDED
+
+
+/**
+ * Adress alignment of the individual FPGA bytes.
+ * The address arrangement of the individual bytes of the FPGA is two
+ * byte aligned at the embedded MK2 platform.
+ */
+#ifdef EXCITE_CCI_FPGA_MK2
+typedef unsigned char excite_cci_fpga_align_t __attribute__ ((aligned(2)));
+#else
+typedef unsigned char excite_cci_fpga_align_t;
+#endif
+
+
+/**
+ * Size of Dual Ported RAM.
+ */
+#define EXCITE_DPR_SIZE 263
+
+
+/**
+ * Size of Reserved Status Fields in Dual Ported RAM.
+ */
+#define EXCITE_DPR_STATUS_SIZE 7
+
+
+
+/**
+ * FPGA.
+ * Hardware register layout of the FPGA interface. The FPGA must accessed
+ * byte wise solely.
+ * @see EXCITE_CCI_DPR_MK2
+ */
+typedef struct excite_fpga {
+
+	/**
+	 * Dual Ported RAM.
+	 */
+	excite_cci_fpga_align_t dpr[EXCITE_DPR_SIZE];
+
+	/**
+	 * Status.
+	 */
+	excite_cci_fpga_align_t status[EXCITE_DPR_STATUS_SIZE];
+
+#ifdef EXCITE_CCI_FPGA_MK2
+	/**
+	 * RM9000 Interrupt.
+	 * Write access initiates interrupt at the RM9000 (MIPS) processor of the eXcite.
+	 */
+	excite_cci_fpga_align_t rm9k_int;
+#else
+	/**
+	 * MK2 Interrupt.
+	 * Write access initiates interrupt at the ARM processor of the MK2.
+	 */
+	excite_cci_fpga_align_t mk2_int;
+
+	excite_cci_fpga_align_t gap[0x1000-0x10f];
+
+	/**
+	 * IRQ Source/Acknowledge.
+	 */
+	excite_cci_fpga_align_t rm9k_irq_src;
+
+	/**
+	 * IRQ Mask.
+	 * Set bits enable the related interrupt.
+	 */
+	excite_cci_fpga_align_t rm9k_irq_mask;
+#endif
+
+
+} excite_fpga;
+
+
+
+#endif	/* ndef EXCITE_FPGA_H_INCLUDED */
-- 
cgit v1.2.3


From 9dbd7b9142e95867ee8a56da5d45c72884c107d3 Mon Sep 17 00:00:00 2001
From: Peter Watkins <treestem@gmail.com>
Date: Wed, 23 Aug 2006 11:15:49 -0400
Subject: [MIPS] Fix USER_PTRS_PER_PGD for 64K page size.

The code in pgtable-64.h assumes TASK_SIZE is always bigger than a first
level PGDIR_SIZE. This is not the case for 64K pages, where task size is
40 bits (1TB) and a pgd entry can map 42 bits. This leads to
USER_PTRS_PER_PGD being zero for 64K pages.

Signed-off-by: Peter Watkins <treestem@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/pgtable-64.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-mips/pgtable-64.h b/include/asm-mips/pgtable-64.h
index c59a1e21f5b0..d05fb6f38aa7 100644
--- a/include/asm-mips/pgtable-64.h
+++ b/include/asm-mips/pgtable-64.h
@@ -93,8 +93,12 @@
 #define PTRS_PER_PMD	((PAGE_SIZE << PMD_ORDER) / sizeof(pmd_t))
 #define PTRS_PER_PTE	((PAGE_SIZE << PTE_ORDER) / sizeof(pte_t))
 
+#if PGDIR_SIZE >= TASK_SIZE
+#define USER_PTRS_PER_PGD       (1)
+#else
 #define USER_PTRS_PER_PGD	(TASK_SIZE / PGDIR_SIZE)
-#define FIRST_USER_ADDRESS	0
+#endif
+#define FIRST_USER_ADDRESS	0UL
 
 #define VMALLOC_START		MAP_BASE
 #define VMALLOC_END	\
-- 
cgit v1.2.3


From d7d86aa88a1f3922b85e39edd8a6d6c01e939842 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 8 Sep 2006 04:13:49 +0200
Subject: [MIPS] Cleanup hazard handling.

Mostly based on patch by Chris Dearman and cleanups from Yoichi.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/hazards.h | 362 ++++++++++++++++-----------------------------
 1 file changed, 126 insertions(+), 236 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/hazards.h b/include/asm-mips/hazards.h
index 25f5e8a4177d..bf0b80404946 100644
--- a/include/asm-mips/hazards.h
+++ b/include/asm-mips/hazards.h
@@ -11,103 +11,96 @@
 #define _ASM_HAZARDS_H
 
 
-#ifdef __ASSEMBLY__
-
-	.macro	_ssnop
-	sll	$0, $0, 1
-	.endm
-
-	.macro	_ehb
-	sll	$0, $0, 3
-	.endm
-
-/*
- * RM9000 hazards.  When the JTLB is updated by tlbwi or tlbwr, a subsequent
- * use of the JTLB for instructions should not occur for 4 cpu cycles and use
- * for data translations should not occur for 3 cpu cycles.
- */
-#ifdef CONFIG_CPU_RM9000
-
-	.macro	mtc0_tlbw_hazard
-	.set	push
-	.set	mips32
-	_ssnop; _ssnop; _ssnop; _ssnop
-	.set	pop
-	.endm
-
-	.macro	tlbw_eret_hazard
-	.set	push
-	.set	mips32
-	_ssnop; _ssnop; _ssnop; _ssnop
-	.set	pop
-	.endm
-
+#ifdef __ASSEMBLER__
+#define ASMMACRO(name, code...) .macro name; code; .endm
 #else
 
-/*
- * The taken branch will result in a two cycle penalty for the two killed
- * instructions on R4000 / R4400.  Other processors only have a single cycle
- * hazard so this is nice trick to have an optimal code for a range of
- * processors.
- */
-	.macro	mtc0_tlbw_hazard
-	b	. + 8
-	.endm
+#define ASMMACRO(name, code...)						\
+__asm__(".macro " #name "; " #code "; .endm");				\
+									\
+static inline void name(void)						\
+{									\
+	__asm__ __volatile__ (#name);					\
+}
 
-	.macro	tlbw_eret_hazard
-	.endm
 #endif
 
+ASMMACRO(_ssnop,
+	 sll	$0, $0, 1
+	)
+
+ASMMACRO(_ehb,
+	 sll	$0, $0, 3
+	)
+
 /*
- * mtc0->mfc0 hazard
- * The 24K has a 2 cycle mtc0/mfc0 execution hazard.
- * It is a MIPS32R2 processor so ehb will clear the hazard.
+ * TLB hazards
  */
+#if defined(CONFIG_CPU_MIPSR2)
 
-#ifdef CONFIG_CPU_MIPSR2
 /*
- * Use a macro for ehb unless explicit support for MIPSR2 is enabled
+ * MIPSR2 defines ehb for hazard avoidance
  */
 
-#define irq_enable_hazard						\
+ASMMACRO(mtc0_tlbw_hazard,
+	 _ehb
+	)
+ASMMACRO(tlbw_use_hazard,
+	 _ehb
+	)
+ASMMACRO(tlb_probe_hazard,
+	 _ehb
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
 	_ehb
-
-#define irq_disable_hazard						\
-	_ehb
-
-#elif defined(CONFIG_CPU_R10000) || defined(CONFIG_CPU_RM9000)
-
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	 _ehb
+	)
 /*
- * R10000 rocks - all hazards handled in hardware, so this becomes a nobrainer.
+ * gcc has a tradition of misscompiling the previous construct using the
+ * address of a label as argument to inline assembler.  Gas otoh has the
+ * annoying difference between la and dla which are only usable for 32-bit
+ * rsp. 64-bit code, so can't be used without conditional compilation.
+ * The alterantive is switching the assembler to 64-bit code which happens
+ * to work right even for 32-bit code ...
  */
+#define instruction_hazard()						\
+do {									\
+	unsigned long tmp;						\
+									\
+	__asm__ __volatile__(						\
+	"	.set	mips64r2				\n"	\
+	"	dla	%0, 1f					\n"	\
+	"	jr.hb	%0					\n"	\
+	"	.set	mips0					\n"	\
+	"1:							\n"	\
+	: "=r" (tmp));							\
+} while (0)
 
-#define irq_enable_hazard
-
-#define irq_disable_hazard
-
-#else
+#elif defined(CONFIG_CPU_R10000)
 
 /*
- * Classic MIPS needs 1 - 3 nops or ssnops
+ * R10000 rocks - all hazards handled in hardware, so this becomes a nobrainer.
  */
-#define irq_enable_hazard
-#define irq_disable_hazard						\
-	_ssnop; _ssnop; _ssnop
 
-#endif
-
-#else /* __ASSEMBLY__ */
-
-__asm__(
-	"	.macro	_ssnop					\n"
-	"	sll	$0, $0, 1				\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro	_ehb					\n"
-	"	sll	$0, $0, 3				\n"
-	"	.endm						\n");
+ASMMACRO(mtc0_tlbw_hazard,
+	)
+ASMMACRO(tlbw_use_hazard,
+	)
+ASMMACRO(tlb_probe_hazard,
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	)
+#define instruction_hazard() do { } while (0)
 
-#ifdef CONFIG_CPU_RM9000
+#elif defined(CONFIG_CPU_RM9000)
 
 /*
  * RM9000 hazards.  When the JTLB is updated by tlbwi or tlbwr, a subsequent
@@ -115,176 +108,73 @@ __asm__(
  * for data translations should not occur for 3 cpu cycles.
  */
 
-#define mtc0_tlbw_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	mips32					\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	.set	mips0					\n")
-
-#define tlbw_use_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	mips32					\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	.set	mips0					\n")
-
-#else
-
-/*
- * Overkill warning ...
- */
-#define mtc0_tlbw_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	noreorder				\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	.set	reorder					\n")
-
-#define tlbw_use_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	noreorder				\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	.set	reorder					\n")
-
-#endif
-
-/*
- * Interrupt enable/disable hazards
- * Some processors have hazards when modifying
- * the status register to change the interrupt state
- */
-
-#ifdef CONFIG_CPU_MIPSR2
-
-__asm__("	.macro	irq_enable_hazard			\n"
-	"	_ehb						\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro	irq_disable_hazard			\n"
-	"	_ehb						\n"
-	"	.endm						\n");
+ASMMACRO(mtc0_tlbw_hazard,
+	 _ssnop; _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(tlbw_use_hazard,
+	 _ssnop; _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(tlb_probe_hazard,
+	 _ssnop; _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	)
+#define instruction_hazard() do { } while (0)
 
-#elif defined(CONFIG_CPU_R10000) || defined(CONFIG_CPU_RM9000)
+#elif defined(CONFIG_CPU_SB1)
 
 /*
- * R10000 rocks - all hazards handled in hardware, so this becomes a nobrainer.
+ * Mostly like R4000 for historic reasons
  */
-
-__asm__(
-	"	.macro	irq_enable_hazard			\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro	irq_disable_hazard			\n"
-	"	.endm						\n");
+ASMMACRO(mtc0_tlbw_hazard,
+	)
+ASMMACRO(tlbw_use_hazard,
+	)
+ASMMACRO(tlb_probe_hazard,
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	 _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	)
+#define instruction_hazard() do { } while (0)
 
 #else
 
 /*
- * Default for classic MIPS processors.  Assume worst case hazards but don't
- * care about the irq_enable_hazard - sooner or later the hardware will
- * enable it and we don't care when exactly.
- */
-
-__asm__(
-	"	#						\n"
-	"	# There is a hazard but we do not care		\n"
-	"	#						\n"
-	"	.macro\tirq_enable_hazard			\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro\tirq_disable_hazard			\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	.endm						\n");
-
-#endif
-
-#define irq_enable_hazard()						\
-	__asm__ __volatile__("irq_enable_hazard")
-#define irq_disable_hazard()						\
-	__asm__ __volatile__("irq_disable_hazard")
-
-
-/*
- * Back-to-back hazards -
+ * Finally the catchall case for all other processors including R4000, R4400,
+ * R4600, R4700, R5000, RM7000, NEC VR41xx etc.
  *
- * What is needed to separate a move to cp0 from a subsequent read from the
- * same cp0 register?
- */
-#ifdef CONFIG_CPU_MIPSR2
-
-__asm__("	.macro	back_to_back_c0_hazard			\n"
-	"	_ehb						\n"
-	"	.endm						\n");
-
-#elif defined(CONFIG_CPU_R10000) || defined(CONFIG_CPU_RM9000) || \
-      defined(CONFIG_CPU_SB1)
-
-__asm__("	.macro	back_to_back_c0_hazard			\n"
-	"	.endm						\n");
-
-#else
-
-__asm__("	.macro	back_to_back_c0_hazard			\n"
-	"	.set	noreorder				\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	.set	reorder					\n"
-	"	.endm");
-
-#endif
-
-#define back_to_back_c0_hazard()					\
-	__asm__ __volatile__("back_to_back_c0_hazard")
-
-
-/*
- * Instruction execution hazard
- */
-#ifdef CONFIG_CPU_MIPSR2
-/*
- * gcc has a tradition of misscompiling the previous construct using the
- * address of a label as argument to inline assembler.  Gas otoh has the
- * annoying difference between la and dla which are only usable for 32-bit
- * rsp. 64-bit code, so can't be used without conditional compilation.
- * The alterantive is switching the assembler to 64-bit code which happens
- * to work right even for 32-bit code ...
+ * The taken branch will result in a two cycle penalty for the two killed
+ * instructions on R4000 / R4400.  Other processors only have a single cycle
+ * hazard so this is nice trick to have an optimal code for a range of
+ * processors.
  */
-#define instruction_hazard()						\
-do {									\
-	unsigned long tmp;						\
-									\
-	__asm__ __volatile__(						\
-	"	.set	mips64r2				\n"	\
-	"	dla	%0, 1f					\n"	\
-	"	jr.hb	%0					\n"	\
-	"	.set	mips0					\n"	\
-	"1:							\n"	\
-	: "=r" (tmp));							\
-} while (0)
-
-#else
+ASMMACRO(mtc0_tlbw_hazard,
+	nop
+	)
+ASMMACRO(tlbw_use_hazard,
+	nop; nop; nop
+	)
+ASMMACRO(tlb_probe_hazard,
+	 nop; nop; nop
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	nop; nop; nop
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	 _ssnop; _ssnop; _ssnop;
+	)
 #define instruction_hazard() do { } while (0)
-#endif
-
-extern void mips_ihb(void);
 
-#endif /* __ASSEMBLY__ */
+#endif
 
 #endif /* _ASM_HAZARDS_H */
-- 
cgit v1.2.3


From d48f1de2d8170814fb64effa320848410c466f95 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Wed, 20 Sep 2006 20:56:02 +0100
Subject: [MIPS] Remove EV96100 as previously announced.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 Documentation/feature-removal-schedule.txt   |   8 -
 Documentation/kernel-parameters.txt          |   2 -
 arch/mips/Kconfig                            |  24 -
 arch/mips/Makefile                           |   7 -
 arch/mips/configs/atlas_defconfig            |   1 -
 arch/mips/configs/bigsur_defconfig           |   1 -
 arch/mips/configs/capcella_defconfig         |   1 -
 arch/mips/configs/cobalt_defconfig           |   1 -
 arch/mips/configs/db1000_defconfig           |   1 -
 arch/mips/configs/db1100_defconfig           |   1 -
 arch/mips/configs/db1200_defconfig           |   1 -
 arch/mips/configs/db1500_defconfig           |   1 -
 arch/mips/configs/db1550_defconfig           |   1 -
 arch/mips/configs/ddb5477_defconfig          |   1 -
 arch/mips/configs/decstation_defconfig       |   1 -
 arch/mips/configs/e55_defconfig              |   1 -
 arch/mips/configs/emma2rh_defconfig          |   1 -
 arch/mips/configs/ev64120_defconfig          |   1 -
 arch/mips/configs/ev96100_defconfig          | 850 ---------------------------
 arch/mips/configs/excite_defconfig           |   1 -
 arch/mips/configs/ip22_defconfig             |   1 -
 arch/mips/configs/ip27_defconfig             |   1 -
 arch/mips/configs/ip32_defconfig             |   1 -
 arch/mips/configs/it8172_defconfig           |   1 -
 arch/mips/configs/ivr_defconfig              |   1 -
 arch/mips/configs/jaguar-atx_defconfig       |   1 -
 arch/mips/configs/jmr3927_defconfig          |   1 -
 arch/mips/configs/lasat200_defconfig         |   1 -
 arch/mips/configs/malta_defconfig            |   1 -
 arch/mips/configs/mipssim_defconfig          |   1 -
 arch/mips/configs/mpc30x_defconfig           |   1 -
 arch/mips/configs/ocelot_3_defconfig         |   1 -
 arch/mips/configs/ocelot_c_defconfig         |   1 -
 arch/mips/configs/ocelot_defconfig           |   1 -
 arch/mips/configs/ocelot_g_defconfig         |   1 -
 arch/mips/configs/pb1100_defconfig           |   1 -
 arch/mips/configs/pb1500_defconfig           |   1 -
 arch/mips/configs/pb1550_defconfig           |   1 -
 arch/mips/configs/pnx8550-jbs_defconfig      |   1 -
 arch/mips/configs/pnx8550-v2pci_defconfig    |   1 -
 arch/mips/configs/qemu_defconfig             |   1 -
 arch/mips/configs/rbhma4500_defconfig        |   1 -
 arch/mips/configs/rm200_defconfig            |   1 -
 arch/mips/configs/sb1250-swarm_defconfig     |   1 -
 arch/mips/configs/sead_defconfig             |   1 -
 arch/mips/configs/tb0226_defconfig           |   1 -
 arch/mips/configs/tb0229_defconfig           |   1 -
 arch/mips/configs/tb0287_defconfig           |   1 -
 arch/mips/configs/workpad_defconfig          |   1 -
 arch/mips/configs/wrppmc_defconfig           |   1 -
 arch/mips/configs/yosemite_defconfig         |   1 -
 arch/mips/defconfig                          |   1 -
 arch/mips/galileo-boards/ev96100/Makefile    |   9 -
 arch/mips/galileo-boards/ev96100/init.c      | 173 ------
 arch/mips/galileo-boards/ev96100/irq.c       |  77 ---
 arch/mips/galileo-boards/ev96100/puts.c      | 138 -----
 arch/mips/galileo-boards/ev96100/reset.c     |  70 ---
 arch/mips/galileo-boards/ev96100/setup.c     | 159 -----
 arch/mips/galileo-boards/ev96100/time.c      |  88 ---
 arch/mips/pci/Makefile                       |   2 -
 arch/mips/pci/fixup-ev96100.c                |  48 --
 arch/mips/pci/ops-gt96100.c                  | 169 ------
 arch/mips/pci/pci-ev96100.c                  |  63 --
 include/asm-mips/bootinfo.h                  |   3 +-
 include/asm-mips/galileo-boards/gt96100.h    | 427 --------------
 include/asm-mips/mach-ev96100/mach-gt64120.h |  46 --
 include/asm-mips/serial.h                    |   4 +-
 include/linux/pci_ids.h                      |   3 -
 68 files changed, 3 insertions(+), 2414 deletions(-)
 delete mode 100644 arch/mips/configs/ev96100_defconfig
 delete mode 100644 arch/mips/galileo-boards/ev96100/Makefile
 delete mode 100644 arch/mips/galileo-boards/ev96100/init.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/irq.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/puts.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/reset.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/setup.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/time.c
 delete mode 100644 arch/mips/pci/fixup-ev96100.c
 delete mode 100644 arch/mips/pci/ops-gt96100.c
 delete mode 100644 arch/mips/pci/pci-ev96100.c
 delete mode 100644 include/asm-mips/galileo-boards/gt96100.h
 delete mode 100644 include/asm-mips/mach-ev96100/mach-gt64120.h

(limited to 'include')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 611acc32fdf5..bf56b20652b0 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -217,14 +217,6 @@ Who:	Nick Piggin <npiggin@suse.de>
 
 ---------------------------
 
-What:	Support for the MIPS EV96100 evaluation board
-When:	September 2006
-Why:	Does no longer build since at least November 15, 2003, apparently
-	no userbase left.
-Who:	Ralf Baechle <ralf@linux-mips.org>
-
----------------------------
-
 What:	Support for the Momentum / PMC-Sierra Jaguar ATX evaluation board
 When:	September 2006
 Why:	Does no longer build since quite some time, and was never popular,
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index c918cc3f65fb..255ec535bba8 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -573,8 +573,6 @@ running once the system is up.
 	gscd=		[HW,CD]
 			Format: <io>
 
-	gt96100eth=	[NET] MIPS GT96100 Advanced Communication Controller
-
 	gus=		[HW,OSS]
 			Format: <io>,<irq>,<dma>,<dma16>
 
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 235208d658ff..30750c54bdf5 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -203,26 +203,6 @@ config MIPS_EV64120
 	  <http://www.marvell.com/>.  Say Y here if you wish to build a
 	  kernel for this platform.
 
-config MIPS_EV96100
-	bool "Galileo EV96100 Evaluation board (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
-	select DMA_NONCOHERENT
-	select HW_HAS_PCI
-	select IRQ_CPU
-	select MIPS_GT96100
-	select RM7000_CPU_SCACHE
-	select SWAP_IO_SPACE
-	select SYS_HAS_CPU_R5000
-	select SYS_HAS_CPU_RM7000
-	select SYS_SUPPORTS_32BIT_KERNEL
-	select SYS_SUPPORTS_64BIT_KERNEL if EXPERIMENTAL
-	select SYS_SUPPORTS_BIG_ENDIAN
-	help
-	  This is an evaluation board based on the Galileo GT-96100 LAN/WAN
-	  communications controllers containing a MIPS R5000 compatible core
-	  running at 83MHz. Their website is <http://www.marvell.com/>. Say Y
-	  here if you wish to build a kernel for this platform.
-
 config MIPS_IVR
 	bool "Globespan IVR board"
 	select DMA_NONCOHERENT
@@ -1069,10 +1049,6 @@ config AU1X00_USB_DEVICE
 	depends on MIPS_PB1500 || MIPS_PB1100 || MIPS_PB1000
 	default n
 
-config MIPS_GT96100
-	bool
-	select MIPS_GT64120
-
 config IT8172_CIR
 	bool
 	depends on MIPS_ITE8172 || MIPS_IVR
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index f4227d24227d..e521826b4234 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -279,13 +279,6 @@ core-$(CONFIG_MIPS_EV64120)	+= arch/mips/gt64120/common/
 cflags-$(CONFIG_MIPS_EV64120)	+= -Iinclude/asm-mips/mach-ev64120
 load-$(CONFIG_MIPS_EV64120)	+= 0xffffffff80100000
 
-#
-# Galileo EV96100 Board
-#
-core-$(CONFIG_MIPS_EV96100)	+= arch/mips/galileo-boards/ev96100/
-cflags-$(CONFIG_MIPS_EV96100)	+= -Iinclude/asm-mips/mach-ev96100
-load-$(CONFIG_MIPS_EV96100)	+= 0xffffffff80100000
-
 #
 # Wind River PPMC Board (4KC + GT64120)
 #
diff --git a/arch/mips/configs/atlas_defconfig b/arch/mips/configs/atlas_defconfig
index 54274065e9a5..2774ba9606b7 100644
--- a/arch/mips/configs/atlas_defconfig
+++ b/arch/mips/configs/atlas_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/bigsur_defconfig b/arch/mips/configs/bigsur_defconfig
index 887fd959482a..e12a475dcbf4 100644
--- a/arch/mips/configs/bigsur_defconfig
+++ b/arch/mips/configs/bigsur_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/capcella_defconfig b/arch/mips/configs/capcella_defconfig
index a01344f3a4c2..bfade9abb767 100644
--- a/arch/mips/configs/capcella_defconfig
+++ b/arch/mips/configs/capcella_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/cobalt_defconfig b/arch/mips/configs/cobalt_defconfig
index c95682445a28..78db2a8a711b 100644
--- a/arch/mips/configs/cobalt_defconfig
+++ b/arch/mips/configs/cobalt_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 CONFIG_MIPS_COBALT=y
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1000_defconfig b/arch/mips/configs/db1000_defconfig
index c2f33d3af62c..93cca1585bc3 100644
--- a/arch/mips/configs/db1000_defconfig
+++ b/arch/mips/configs/db1000_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1000=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1100_defconfig b/arch/mips/configs/db1100_defconfig
index 8c44d16ae9a2..ffd99252a837 100644
--- a/arch/mips/configs/db1100_defconfig
+++ b/arch/mips/configs/db1100_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1100=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1200_defconfig b/arch/mips/configs/db1200_defconfig
index c13768e75ac5..63eac5e89b9c 100644
--- a/arch/mips/configs/db1200_defconfig
+++ b/arch/mips/configs/db1200_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1200=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1500_defconfig b/arch/mips/configs/db1500_defconfig
index 8aea73fae7fb..25a095f7dc4e 100644
--- a/arch/mips/configs/db1500_defconfig
+++ b/arch/mips/configs/db1500_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1500=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1550_defconfig b/arch/mips/configs/db1550_defconfig
index 90ccb7359630..dda469c842b3 100644
--- a/arch/mips/configs/db1550_defconfig
+++ b/arch/mips/configs/db1550_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1550=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ddb5477_defconfig b/arch/mips/configs/ddb5477_defconfig
index b598cf08f156..fcd3dd19bc74 100644
--- a/arch/mips/configs/ddb5477_defconfig
+++ b/arch/mips/configs/ddb5477_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/decstation_defconfig b/arch/mips/configs/decstation_defconfig
index 597150b14077..8683e0df12e0 100644
--- a/arch/mips/configs/decstation_defconfig
+++ b/arch/mips/configs/decstation_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 CONFIG_MACH_DECSTATION=y
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/e55_defconfig b/arch/mips/configs/e55_defconfig
index d9a0d0eb0683..4ace61c95778 100644
--- a/arch/mips/configs/e55_defconfig
+++ b/arch/mips/configs/e55_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/emma2rh_defconfig b/arch/mips/configs/emma2rh_defconfig
index 375b2ac24a49..5847c916c130 100644
--- a/arch/mips/configs/emma2rh_defconfig
+++ b/arch/mips/configs/emma2rh_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ev64120_defconfig b/arch/mips/configs/ev64120_defconfig
index b0afc118bd5c..bc4c4f125c48 100644
--- a/arch/mips/configs/ev64120_defconfig
+++ b/arch/mips/configs/ev64120_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 CONFIG_MIPS_EV64120=y
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ev96100_defconfig b/arch/mips/configs/ev96100_defconfig
deleted file mode 100644
index 0bdc10f11610..000000000000
--- a/arch/mips/configs/ev96100_defconfig
+++ /dev/null
@@ -1,850 +0,0 @@
-#
-# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-rc1
-# Thu Jul  6 10:04:05 2006
-#
-CONFIG_MIPS=y
-
-#
-# Machine selection
-#
-# CONFIG_MIPS_MTX1 is not set
-# CONFIG_MIPS_BOSPORUS is not set
-# CONFIG_MIPS_PB1000 is not set
-# CONFIG_MIPS_PB1100 is not set
-# CONFIG_MIPS_PB1500 is not set
-# CONFIG_MIPS_PB1550 is not set
-# CONFIG_MIPS_PB1200 is not set
-# CONFIG_MIPS_DB1000 is not set
-# CONFIG_MIPS_DB1100 is not set
-# CONFIG_MIPS_DB1500 is not set
-# CONFIG_MIPS_DB1550 is not set
-# CONFIG_MIPS_DB1200 is not set
-# CONFIG_MIPS_MIRAGE is not set
-# CONFIG_BASLER_EXCITE is not set
-# CONFIG_MIPS_COBALT is not set
-# CONFIG_MACH_DECSTATION is not set
-# CONFIG_MIPS_EV64120 is not set
-CONFIG_MIPS_EV96100=y
-# CONFIG_MIPS_IVR is not set
-# CONFIG_MIPS_ITE8172 is not set
-# CONFIG_MACH_JAZZ is not set
-# CONFIG_LASAT is not set
-# CONFIG_MIPS_ATLAS is not set
-# CONFIG_MIPS_MALTA is not set
-# CONFIG_MIPS_SEAD is not set
-# CONFIG_WR_PPMC is not set
-# CONFIG_MIPS_SIM is not set
-# CONFIG_MOMENCO_JAGUAR_ATX is not set
-# CONFIG_MOMENCO_OCELOT is not set
-# CONFIG_MOMENCO_OCELOT_3 is not set
-# CONFIG_MOMENCO_OCELOT_C is not set
-# CONFIG_MOMENCO_OCELOT_G is not set
-# CONFIG_MIPS_XXS1500 is not set
-# CONFIG_PNX8550_V2PCI is not set
-# CONFIG_PNX8550_JBS is not set
-# CONFIG_DDB5477 is not set
-# CONFIG_MACH_VR41XX is not set
-# CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
-# CONFIG_MARKEINS is not set
-# CONFIG_SGI_IP22 is not set
-# CONFIG_SGI_IP27 is not set
-# CONFIG_SGI_IP32 is not set
-# CONFIG_SIBYTE_BIGSUR is not set
-# CONFIG_SIBYTE_SWARM is not set
-# CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_RHONE is not set
-# CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
-# CONFIG_SIBYTE_LITTLESUR is not set
-# CONFIG_SIBYTE_CRHINE is not set
-# CONFIG_SIBYTE_CRHONE is not set
-# CONFIG_SNI_RM200_PCI is not set
-# CONFIG_TOSHIBA_JMR3927 is not set
-# CONFIG_TOSHIBA_RBTX4927 is not set
-# CONFIG_TOSHIBA_RBTX4938 is not set
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-CONFIG_GENERIC_FIND_NEXT_BIT=y
-CONFIG_GENERIC_HWEIGHT=y
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
-CONFIG_DMA_NONCOHERENT=y
-CONFIG_DMA_NEED_PCI_MAP_STATE=y
-CONFIG_CPU_BIG_ENDIAN=y
-# CONFIG_CPU_LITTLE_ENDIAN is not set
-CONFIG_SYS_SUPPORTS_BIG_ENDIAN=y
-CONFIG_IRQ_CPU=y
-CONFIG_MIPS_GT64120=y
-CONFIG_SWAP_IO_SPACE=y
-CONFIG_MIPS_GT96100=y
-CONFIG_MIPS_L1_CACHE_SHIFT=5
-
-#
-# CPU selection
-#
-# CONFIG_CPU_MIPS32_R1 is not set
-# CONFIG_CPU_MIPS32_R2 is not set
-# CONFIG_CPU_MIPS64_R1 is not set
-# CONFIG_CPU_MIPS64_R2 is not set
-# CONFIG_CPU_R3000 is not set
-# CONFIG_CPU_TX39XX is not set
-# CONFIG_CPU_VR41XX is not set
-# CONFIG_CPU_R4300 is not set
-# CONFIG_CPU_R4X00 is not set
-# CONFIG_CPU_TX49XX is not set
-# CONFIG_CPU_R5000 is not set
-# CONFIG_CPU_R5432 is not set
-# CONFIG_CPU_R6000 is not set
-# CONFIG_CPU_NEVADA is not set
-# CONFIG_CPU_R8000 is not set
-# CONFIG_CPU_R10000 is not set
-CONFIG_CPU_RM7000=y
-# CONFIG_CPU_RM9000 is not set
-# CONFIG_CPU_SB1 is not set
-CONFIG_SYS_HAS_CPU_R5000=y
-CONFIG_SYS_HAS_CPU_RM7000=y
-CONFIG_SYS_SUPPORTS_32BIT_KERNEL=y
-CONFIG_SYS_SUPPORTS_64BIT_KERNEL=y
-CONFIG_CPU_SUPPORTS_32BIT_KERNEL=y
-CONFIG_CPU_SUPPORTS_64BIT_KERNEL=y
-
-#
-# Kernel type
-#
-CONFIG_32BIT=y
-# CONFIG_64BIT is not set
-CONFIG_PAGE_SIZE_4KB=y
-# CONFIG_PAGE_SIZE_8KB is not set
-# CONFIG_PAGE_SIZE_16KB is not set
-# CONFIG_PAGE_SIZE_64KB is not set
-CONFIG_BOARD_SCACHE=y
-CONFIG_RM7000_CPU_SCACHE=y
-CONFIG_CPU_HAS_PREFETCH=y
-CONFIG_MIPS_MT_DISABLED=y
-# CONFIG_MIPS_MT_SMTC is not set
-# CONFIG_MIPS_MT_SMP is not set
-# CONFIG_MIPS_VPE_LOADER is not set
-# CONFIG_64BIT_PHYS_ADDR is not set
-CONFIG_CPU_HAS_LLSC=y
-CONFIG_CPU_HAS_SYNC=y
-CONFIG_GENERIC_HARDIRQS=y
-CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_CPU_SUPPORTS_HIGHMEM=y
-CONFIG_ARCH_FLATMEM_ENABLE=y
-CONFIG_SELECT_MEMORY_MODEL=y
-CONFIG_FLATMEM_MANUAL=y
-# CONFIG_DISCONTIGMEM_MANUAL is not set
-# CONFIG_SPARSEMEM_MANUAL is not set
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-# CONFIG_SPARSEMEM_STATIC is not set
-CONFIG_SPLIT_PTLOCK_CPUS=4
-# CONFIG_RESOURCES_64BIT is not set
-# CONFIG_HZ_48 is not set
-# CONFIG_HZ_100 is not set
-# CONFIG_HZ_128 is not set
-# CONFIG_HZ_250 is not set
-# CONFIG_HZ_256 is not set
-CONFIG_HZ_1000=y
-# CONFIG_HZ_1024 is not set
-CONFIG_SYS_SUPPORTS_ARBIT_HZ=y
-CONFIG_HZ=1000
-CONFIG_PREEMPT_NONE=y
-# CONFIG_PREEMPT_VOLUNTARY is not set
-# CONFIG_PREEMPT is not set
-CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
-
-#
-# Code maturity level options
-#
-CONFIG_EXPERIMENTAL=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
-CONFIG_LOCALVERSION=""
-CONFIG_LOCALVERSION_AUTO=y
-CONFIG_SWAP=y
-CONFIG_SYSVIPC=y
-# CONFIG_POSIX_MQUEUE is not set
-# CONFIG_BSD_PROCESS_ACCT is not set
-CONFIG_SYSCTL=y
-# CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
-CONFIG_RELAY=y
-CONFIG_INITRAMFS_SOURCE=""
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-CONFIG_EMBEDDED=y
-CONFIG_KALLSYMS=y
-# CONFIG_KALLSYMS_EXTRA_PASS is not set
-# CONFIG_HOTPLUG is not set
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-CONFIG_ELF_CORE=y
-CONFIG_BASE_FULL=y
-CONFIG_RT_MUTEXES=y
-CONFIG_FUTEX=y
-CONFIG_EPOLL=y
-CONFIG_SHMEM=y
-CONFIG_SLAB=y
-CONFIG_VM_EVENT_COUNTERS=y
-# CONFIG_TINY_SHMEM is not set
-CONFIG_BASE_SMALL=0
-# CONFIG_SLOB is not set
-
-#
-# Loadable module support
-#
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_MODULE_FORCE_UNLOAD is not set
-CONFIG_MODVERSIONS=y
-CONFIG_MODULE_SRCVERSION_ALL=y
-# CONFIG_KMOD is not set
-
-#
-# Block layer
-#
-# CONFIG_LBD is not set
-# CONFIG_BLK_DEV_IO_TRACE is not set
-# CONFIG_LSF is not set
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-CONFIG_IOSCHED_AS=y
-CONFIG_IOSCHED_DEADLINE=y
-CONFIG_IOSCHED_CFQ=y
-CONFIG_DEFAULT_AS=y
-# CONFIG_DEFAULT_DEADLINE is not set
-# CONFIG_DEFAULT_CFQ is not set
-# CONFIG_DEFAULT_NOOP is not set
-CONFIG_DEFAULT_IOSCHED="anticipatory"
-
-#
-# Bus options (PCI, PCMCIA, EISA, ISA, TC)
-#
-CONFIG_HW_HAS_PCI=y
-# CONFIG_PCI is not set
-CONFIG_MMU=y
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
-# CONFIG_PCCARD is not set
-
-#
-# PCI Hotplug Support
-#
-
-#
-# Executable file formats
-#
-CONFIG_BINFMT_ELF=y
-# CONFIG_BINFMT_MISC is not set
-CONFIG_TRAD_SIGNALS=y
-
-#
-# Networking
-#
-CONFIG_NET=y
-
-#
-# Networking options
-#
-# CONFIG_NETDEBUG is not set
-# CONFIG_PACKET is not set
-CONFIG_UNIX=y
-CONFIG_XFRM=y
-CONFIG_XFRM_USER=m
-CONFIG_NET_KEY=y
-CONFIG_INET=y
-# CONFIG_IP_MULTICAST is not set
-# CONFIG_IP_ADVANCED_ROUTER is not set
-CONFIG_IP_FIB_HASH=y
-CONFIG_IP_PNP=y
-# CONFIG_IP_PNP_DHCP is not set
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_IP_PNP_RARP is not set
-# CONFIG_NET_IPIP is not set
-# CONFIG_NET_IPGRE is not set
-# CONFIG_ARPD is not set
-# CONFIG_SYN_COOKIES is not set
-# CONFIG_INET_AH is not set
-# CONFIG_INET_ESP is not set
-# CONFIG_INET_IPCOMP is not set
-# CONFIG_INET_XFRM_TUNNEL is not set
-# CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=m
-CONFIG_INET_XFRM_MODE_TUNNEL=m
-CONFIG_INET_DIAG=y
-CONFIG_INET_TCP_DIAG=y
-# CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_BIC=y
-# CONFIG_IPV6 is not set
-# CONFIG_INET6_XFRM_TUNNEL is not set
-# CONFIG_INET6_TUNNEL is not set
-CONFIG_NETWORK_SECMARK=y
-# CONFIG_NETFILTER is not set
-
-#
-# DCCP Configuration (EXPERIMENTAL)
-#
-# CONFIG_IP_DCCP is not set
-
-#
-# SCTP Configuration (EXPERIMENTAL)
-#
-# CONFIG_IP_SCTP is not set
-
-#
-# TIPC Configuration (EXPERIMENTAL)
-#
-# CONFIG_TIPC is not set
-# CONFIG_ATM is not set
-# CONFIG_BRIDGE is not set
-# CONFIG_VLAN_8021Q is not set
-# CONFIG_DECNET is not set
-# CONFIG_LLC2 is not set
-# CONFIG_IPX is not set
-# CONFIG_ATALK is not set
-# CONFIG_X25 is not set
-# CONFIG_LAPB is not set
-# CONFIG_NET_DIVERT is not set
-# CONFIG_ECONET is not set
-# CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
-# CONFIG_NET_SCHED is not set
-
-#
-# Network testing
-#
-# CONFIG_NET_PKTGEN is not set
-# CONFIG_HAMRADIO is not set
-# CONFIG_IRDA is not set
-# CONFIG_BT is not set
-CONFIG_IEEE80211=m
-# CONFIG_IEEE80211_DEBUG is not set
-CONFIG_IEEE80211_CRYPT_WEP=m
-CONFIG_IEEE80211_CRYPT_CCMP=m
-CONFIG_IEEE80211_SOFTMAC=m
-# CONFIG_IEEE80211_SOFTMAC_DEBUG is not set
-CONFIG_WIRELESS_EXT=y
-
-#
-# Device Drivers
-#
-
-#
-# Generic Driver Options
-#
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
-# CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
-CONFIG_CONNECTOR=m
-
-#
-# Memory Technology Devices (MTD)
-#
-# CONFIG_MTD is not set
-
-#
-# Parallel port support
-#
-# CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-
-#
-# Block devices
-#
-# CONFIG_BLK_DEV_COW_COMMON is not set
-# CONFIG_BLK_DEV_LOOP is not set
-# CONFIG_BLK_DEV_NBD is not set
-# CONFIG_BLK_DEV_RAM is not set
-# CONFIG_BLK_DEV_INITRD is not set
-CONFIG_CDROM_PKTCDVD=m
-CONFIG_CDROM_PKTCDVD_BUFFERS=8
-# CONFIG_CDROM_PKTCDVD_WCACHE is not set
-CONFIG_ATA_OVER_ETH=m
-
-#
-# ATA/ATAPI/MFM/RLL support
-#
-# CONFIG_IDE is not set
-
-#
-# SCSI device support
-#
-CONFIG_RAID_ATTRS=m
-# CONFIG_SCSI is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
-# CONFIG_MD is not set
-
-#
-# Fusion MPT device support
-#
-# CONFIG_FUSION is not set
-
-#
-# IEEE 1394 (FireWire) support
-#
-
-#
-# I2O device support
-#
-
-#
-# Network device support
-#
-CONFIG_NETDEVICES=y
-# CONFIG_DUMMY is not set
-# CONFIG_BONDING is not set
-# CONFIG_EQUALIZER is not set
-# CONFIG_TUN is not set
-
-#
-# PHY device support
-#
-CONFIG_PHYLIB=m
-
-#
-# MII PHY device drivers
-#
-CONFIG_MARVELL_PHY=m
-CONFIG_DAVICOM_PHY=m
-CONFIG_QSEMI_PHY=m
-CONFIG_LXT_PHY=m
-CONFIG_CICADA_PHY=m
-CONFIG_VITESSE_PHY=m
-CONFIG_SMSC_PHY=m
-
-#
-# Ethernet (10 or 100Mbit)
-#
-CONFIG_NET_ETHERNET=y
-# CONFIG_MII is not set
-CONFIG_MIPS_GT96100ETH=y
-# CONFIG_DM9000 is not set
-
-#
-# Ethernet (1000 Mbit)
-#
-
-#
-# Ethernet (10000 Mbit)
-#
-
-#
-# Token Ring devices
-#
-
-#
-# Wireless LAN (non-hamradio)
-#
-# CONFIG_NET_RADIO is not set
-
-#
-# Wan interfaces
-#
-# CONFIG_WAN is not set
-# CONFIG_PPP is not set
-# CONFIG_SLIP is not set
-# CONFIG_SHAPER is not set
-# CONFIG_NETCONSOLE is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
-# CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
-# CONFIG_PHONE is not set
-
-#
-# Input device support
-#
-CONFIG_INPUT=y
-
-#
-# Userland interfaces
-#
-CONFIG_INPUT_MOUSEDEV=y
-CONFIG_INPUT_MOUSEDEV_PSAUX=y
-CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
-CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
-# CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
-# CONFIG_INPUT_EVDEV is not set
-# CONFIG_INPUT_EVBUG is not set
-
-#
-# Input Device Drivers
-#
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_INPUT_JOYSTICK is not set
-# CONFIG_INPUT_TOUCHSCREEN is not set
-# CONFIG_INPUT_MISC is not set
-
-#
-# Hardware I/O ports
-#
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_LIBPS2 is not set
-CONFIG_SERIO_RAW=m
-# CONFIG_GAMEPORT is not set
-
-#
-# Character devices
-#
-CONFIG_VT=y
-CONFIG_VT_CONSOLE=y
-CONFIG_HW_CONSOLE=y
-CONFIG_VT_HW_CONSOLE_BINDING=y
-# CONFIG_SERIAL_NONSTANDARD is not set
-
-#
-# Serial drivers
-#
-CONFIG_SERIAL_8250=y
-CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
-CONFIG_SERIAL_8250_RUNTIME_UARTS=4
-# CONFIG_SERIAL_8250_EXTENDED is not set
-
-#
-# Non-8250 serial port support
-#
-CONFIG_SERIAL_CORE=y
-CONFIG_SERIAL_CORE_CONSOLE=y
-CONFIG_UNIX98_PTYS=y
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=256
-
-#
-# IPMI
-#
-# CONFIG_IPMI_HANDLER is not set
-
-#
-# Watchdog Cards
-#
-# CONFIG_WATCHDOG is not set
-# CONFIG_HW_RANDOM is not set
-# CONFIG_RTC is not set
-# CONFIG_GEN_RTC is not set
-# CONFIG_DTLK is not set
-# CONFIG_R3964 is not set
-
-#
-# Ftape, the floppy tape device driver
-#
-# CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
-# CONFIG_TCG_TPM is not set
-# CONFIG_TELCLOCK is not set
-
-#
-# I2C support
-#
-# CONFIG_I2C is not set
-
-#
-# SPI support
-#
-# CONFIG_SPI is not set
-# CONFIG_SPI_MASTER is not set
-
-#
-# Dallas's 1-wire bus
-#
-# CONFIG_W1 is not set
-
-#
-# Hardware Monitoring support
-#
-# CONFIG_HWMON is not set
-# CONFIG_HWMON_VID is not set
-
-#
-# Misc devices
-#
-
-#
-# Multimedia devices
-#
-# CONFIG_VIDEO_DEV is not set
-CONFIG_VIDEO_V4L2=y
-
-#
-# Digital Video Broadcasting Devices
-#
-# CONFIG_DVB is not set
-
-#
-# Graphics support
-#
-# CONFIG_FIRMWARE_EDID is not set
-# CONFIG_FB is not set
-
-#
-# Console display driver support
-#
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_DUMMY_CONSOLE=y
-
-#
-# Sound
-#
-# CONFIG_SOUND is not set
-
-#
-# USB support
-#
-# CONFIG_USB_ARCH_HAS_HCD is not set
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
-
-#
-# MMC/SD Card support
-#
-# CONFIG_MMC is not set
-
-#
-# LED devices
-#
-# CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
-# CONFIG_RTC_CLASS is not set
-
-#
-# DMA Engine support
-#
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# File systems
-#
-CONFIG_EXT2_FS=y
-# CONFIG_EXT2_FS_XATTR is not set
-# CONFIG_EXT2_FS_XIP is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_REISERFS_FS is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-# CONFIG_XFS_FS is not set
-# CONFIG_OCFS2_FS is not set
-# CONFIG_MINIX_FS is not set
-# CONFIG_ROMFS_FS is not set
-CONFIG_INOTIFY=y
-CONFIG_INOTIFY_USER=y
-# CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
-# CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
-CONFIG_FUSE_FS=m
-
-#
-# CD-ROM/DVD Filesystems
-#
-# CONFIG_ISO9660_FS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_PROC_KCORE=y
-CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
-# CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
-# CONFIG_CONFIGFS_FS is not set
-
-#
-# Miscellaneous filesystems
-#
-# CONFIG_ADFS_FS is not set
-# CONFIG_AFFS_FS is not set
-# CONFIG_HFS_FS is not set
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_BEFS_FS is not set
-# CONFIG_BFS_FS is not set
-# CONFIG_EFS_FS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
-CONFIG_NFS_FS=y
-# CONFIG_NFS_V3 is not set
-# CONFIG_NFS_V4 is not set
-# CONFIG_NFS_DIRECTIO is not set
-# CONFIG_NFSD is not set
-CONFIG_ROOT_NFS=y
-CONFIG_LOCKD=y
-CONFIG_NFS_COMMON=y
-CONFIG_SUNRPC=y
-# CONFIG_RPCSEC_GSS_KRB5 is not set
-# CONFIG_RPCSEC_GSS_SPKM3 is not set
-# CONFIG_SMB_FS is not set
-# CONFIG_CIFS is not set
-# CONFIG_CIFS_DEBUG2 is not set
-# CONFIG_NCP_FS is not set
-# CONFIG_CODA_FS is not set
-# CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
-
-#
-# Partition Types
-#
-# CONFIG_PARTITION_ADVANCED is not set
-CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
-# CONFIG_NLS is not set
-
-#
-# Profiling support
-#
-# CONFIG_PROFILING is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-# CONFIG_MAGIC_SYSRQ is not set
-# CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_KERNEL is not set
-CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_DEBUG_FS is not set
-CONFIG_CROSSCOMPILE=y
-CONFIG_CMDLINE=""
-
-#
-# Security options
-#
-CONFIG_KEYS=y
-CONFIG_KEYS_DEBUG_PROC_KEYS=y
-# CONFIG_SECURITY is not set
-
-#
-# Cryptographic options
-#
-CONFIG_CRYPTO=y
-CONFIG_CRYPTO_HMAC=y
-CONFIG_CRYPTO_NULL=m
-CONFIG_CRYPTO_MD4=m
-CONFIG_CRYPTO_MD5=m
-CONFIG_CRYPTO_SHA1=m
-CONFIG_CRYPTO_SHA256=m
-CONFIG_CRYPTO_SHA512=m
-CONFIG_CRYPTO_WP512=m
-CONFIG_CRYPTO_TGR192=m
-CONFIG_CRYPTO_DES=m
-CONFIG_CRYPTO_BLOWFISH=m
-CONFIG_CRYPTO_TWOFISH=m
-CONFIG_CRYPTO_SERPENT=m
-CONFIG_CRYPTO_AES=m
-CONFIG_CRYPTO_CAST5=m
-CONFIG_CRYPTO_CAST6=m
-CONFIG_CRYPTO_TEA=m
-CONFIG_CRYPTO_ARC4=m
-CONFIG_CRYPTO_KHAZAD=m
-CONFIG_CRYPTO_ANUBIS=m
-CONFIG_CRYPTO_DEFLATE=m
-CONFIG_CRYPTO_MICHAEL_MIC=m
-CONFIG_CRYPTO_CRC32C=m
-# CONFIG_CRYPTO_TEST is not set
-
-#
-# Hardware crypto devices
-#
-
-#
-# Library routines
-#
-# CONFIG_CRC_CCITT is not set
-CONFIG_CRC16=m
-CONFIG_CRC32=m
-CONFIG_LIBCRC32C=m
-CONFIG_ZLIB_INFLATE=m
-CONFIG_ZLIB_DEFLATE=m
-CONFIG_PLIST=y
diff --git a/arch/mips/configs/excite_defconfig b/arch/mips/configs/excite_defconfig
index 045ebd089893..eb87cbbfd037 100644
--- a/arch/mips/configs/excite_defconfig
+++ b/arch/mips/configs/excite_defconfig
@@ -26,7 +26,6 @@ CONFIG_BASLER_EXCITE=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ip22_defconfig b/arch/mips/configs/ip22_defconfig
index ef16d1fb5071..9f22d13e729d 100644
--- a/arch/mips/configs/ip22_defconfig
+++ b/arch/mips/configs/ip22_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index 4bf1ee7f5f00..414a649dff8a 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ip32_defconfig b/arch/mips/configs/ip32_defconfig
index f83dc09c3ca9..dec2ba6ba03f 100644
--- a/arch/mips/configs/ip32_defconfig
+++ b/arch/mips/configs/ip32_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/it8172_defconfig b/arch/mips/configs/it8172_defconfig
index a91d72a9ca86..37f9dd7187b1 100644
--- a/arch/mips/configs/it8172_defconfig
+++ b/arch/mips/configs/it8172_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 CONFIG_MIPS_ITE8172=y
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ivr_defconfig b/arch/mips/configs/ivr_defconfig
index cebc67212d06..18874a4c24fe 100644
--- a/arch/mips/configs/ivr_defconfig
+++ b/arch/mips/configs/ivr_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 CONFIG_MIPS_IVR=y
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/jaguar-atx_defconfig b/arch/mips/configs/jaguar-atx_defconfig
index 5d9eb11aba3d..9b529857f802 100644
--- a/arch/mips/configs/jaguar-atx_defconfig
+++ b/arch/mips/configs/jaguar-atx_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/jmr3927_defconfig b/arch/mips/configs/jmr3927_defconfig
index be45a9044d06..fded3f73815f 100644
--- a/arch/mips/configs/jmr3927_defconfig
+++ b/arch/mips/configs/jmr3927_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/lasat200_defconfig b/arch/mips/configs/lasat200_defconfig
index 64dc9f45a19c..0354bfa18b39 100644
--- a/arch/mips/configs/lasat200_defconfig
+++ b/arch/mips/configs/lasat200_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/malta_defconfig b/arch/mips/configs/malta_defconfig
index 2690baf15a85..b21ca470273a 100644
--- a/arch/mips/configs/malta_defconfig
+++ b/arch/mips/configs/malta_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/mipssim_defconfig b/arch/mips/configs/mipssim_defconfig
index c298979c18ae..adbeeadddb8f 100644
--- a/arch/mips/configs/mipssim_defconfig
+++ b/arch/mips/configs/mipssim_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/mpc30x_defconfig b/arch/mips/configs/mpc30x_defconfig
index 4405d127a941..79fd544fcb2a 100644
--- a/arch/mips/configs/mpc30x_defconfig
+++ b/arch/mips/configs/mpc30x_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_3_defconfig b/arch/mips/configs/ocelot_3_defconfig
index ec5758f22676..4d87da2b99fd 100644
--- a/arch/mips/configs/ocelot_3_defconfig
+++ b/arch/mips/configs/ocelot_3_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_c_defconfig b/arch/mips/configs/ocelot_c_defconfig
index 0d33d87de1a1..ce7c26475328 100644
--- a/arch/mips/configs/ocelot_c_defconfig
+++ b/arch/mips/configs/ocelot_c_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_defconfig b/arch/mips/configs/ocelot_defconfig
index 4b999102715e..3da2a43d1608 100644
--- a/arch/mips/configs/ocelot_defconfig
+++ b/arch/mips/configs/ocelot_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_g_defconfig b/arch/mips/configs/ocelot_g_defconfig
index 827b344f6010..3a3e5ff4a3ef 100644
--- a/arch/mips/configs/ocelot_g_defconfig
+++ b/arch/mips/configs/ocelot_g_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pb1100_defconfig b/arch/mips/configs/pb1100_defconfig
index 9ed60fef69e0..1a16e92900cb 100644
--- a/arch/mips/configs/pb1100_defconfig
+++ b/arch/mips/configs/pb1100_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_PB1100=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pb1500_defconfig b/arch/mips/configs/pb1500_defconfig
index 6774254b1be6..9ea8edea6f29 100644
--- a/arch/mips/configs/pb1500_defconfig
+++ b/arch/mips/configs/pb1500_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_PB1500=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pb1550_defconfig b/arch/mips/configs/pb1550_defconfig
index 1afe5bf6e765..c4a158976f8f 100644
--- a/arch/mips/configs/pb1550_defconfig
+++ b/arch/mips/configs/pb1550_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_PB1550=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pnx8550-jbs_defconfig b/arch/mips/configs/pnx8550-jbs_defconfig
index ac616c82d348..1cbf270c301c 100644
--- a/arch/mips/configs/pnx8550-jbs_defconfig
+++ b/arch/mips/configs/pnx8550-jbs_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pnx8550-v2pci_defconfig b/arch/mips/configs/pnx8550-v2pci_defconfig
index a8eb51bae3f3..bec30b15b9bd 100644
--- a/arch/mips/configs/pnx8550-v2pci_defconfig
+++ b/arch/mips/configs/pnx8550-v2pci_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/qemu_defconfig b/arch/mips/configs/qemu_defconfig
index 6a63a113b7ea..daa40c8ff694 100644
--- a/arch/mips/configs/qemu_defconfig
+++ b/arch/mips/configs/qemu_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/rbhma4500_defconfig b/arch/mips/configs/rbhma4500_defconfig
index 6779f449bd2d..2f5650227ba3 100644
--- a/arch/mips/configs/rbhma4500_defconfig
+++ b/arch/mips/configs/rbhma4500_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/rm200_defconfig b/arch/mips/configs/rm200_defconfig
index b7826d3a2b77..f26b338333ac 100644
--- a/arch/mips/configs/rm200_defconfig
+++ b/arch/mips/configs/rm200_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/sb1250-swarm_defconfig b/arch/mips/configs/sb1250-swarm_defconfig
index 625c1c619b6b..9041f095f96f 100644
--- a/arch/mips/configs/sb1250-swarm_defconfig
+++ b/arch/mips/configs/sb1250-swarm_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/sead_defconfig b/arch/mips/configs/sead_defconfig
index 4401b602118f..02abb2f1bfaf 100644
--- a/arch/mips/configs/sead_defconfig
+++ b/arch/mips/configs/sead_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/tb0226_defconfig b/arch/mips/configs/tb0226_defconfig
index 2ba4e25e8c34..ca3d0c4ba15b 100644
--- a/arch/mips/configs/tb0226_defconfig
+++ b/arch/mips/configs/tb0226_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/tb0229_defconfig b/arch/mips/configs/tb0229_defconfig
index fc8a407c1add..4e2009ace278 100644
--- a/arch/mips/configs/tb0229_defconfig
+++ b/arch/mips/configs/tb0229_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/tb0287_defconfig b/arch/mips/configs/tb0287_defconfig
index effcb63b81a3..535a813d01a9 100644
--- a/arch/mips/configs/tb0287_defconfig
+++ b/arch/mips/configs/tb0287_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/workpad_defconfig b/arch/mips/configs/workpad_defconfig
index 8f66ad71cb7e..3a3ef20b21cc 100644
--- a/arch/mips/configs/workpad_defconfig
+++ b/arch/mips/configs/workpad_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/wrppmc_defconfig b/arch/mips/configs/wrppmc_defconfig
index 3e4b16b39827..e6b1dea55842 100644
--- a/arch/mips/configs/wrppmc_defconfig
+++ b/arch/mips/configs/wrppmc_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/yosemite_defconfig b/arch/mips/configs/yosemite_defconfig
index 3a68d8a25b66..06a072b77b1c 100644
--- a/arch/mips/configs/yosemite_defconfig
+++ b/arch/mips/configs/yosemite_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/defconfig b/arch/mips/defconfig
index fff6fcc96212..d620c463a78b 100644
--- a/arch/mips/defconfig
+++ b/arch/mips/defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/galileo-boards/ev96100/Makefile b/arch/mips/galileo-boards/ev96100/Makefile
deleted file mode 100644
index cd868ec78cbc..000000000000
--- a/arch/mips/galileo-boards/ev96100/Makefile
+++ /dev/null
@@ -1,9 +0,0 @@
-#
-#  Copyright 2000 MontaVista Software Inc.
-#  Author: MontaVista Software, Inc.
-#     	ppopov@mvista.com or source@mvista.com
-#
-# Makefile for the Galileo EV96100 board.
-#
-
-obj-y		+= init.o irq.o puts.o reset.o time.o setup.o
diff --git a/arch/mips/galileo-boards/ev96100/init.c b/arch/mips/galileo-boards/ev96100/init.c
deleted file mode 100644
index a01fe9b36f2c..000000000000
--- a/arch/mips/galileo-boards/ev96100/init.c
+++ /dev/null
@@ -1,173 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/generic/generic.c
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/bootmem.h>
-#include <linux/string.h>
-#include <linux/kernel.h>
-
-#include <asm/addrspace.h>
-#include <asm/bootinfo.h>
-#include <asm/gt64120.h>
-
-
-/* Environment variable */
-
-typedef struct {
-	char *name;
-	char *val;
-} t_env_var;
-
-int prom_argc;
-char **prom_argv, **prom_envp;
-
-int init_debug = 0;
-
-char * __init prom_getcmdline(void)
-{
-	return &(arcs_cmdline[0]);
-}
-
-unsigned long __init prom_free_prom_memory(void)
-{
-	return 0;
-}
-
-void  __init prom_init_cmdline(void)
-{
-	char *cp;
-	int actr;
-
-	actr = 1; /* Always ignore argv[0] */
-
-	cp = &(arcs_cmdline[0]);
-	while(actr < prom_argc) {
-	        strcpy(cp, prom_argv[actr]);
-		cp += strlen(prom_argv[actr]);
-		*cp++ = ' ';
-		actr++;
-	}
-	if (cp != &(arcs_cmdline[0])) /* get rid of trailing space */
-		--cp;
-	*cp = '\0';
-}
-
-char *prom_getenv(char *envname)
-{
-	/*
-	 * Return a pointer to the given environment variable.
-	 */
-
-	t_env_var *env = (t_env_var *) prom_envp;
-	int i;
-
-	i = strlen(envname);
-
-	while (env->name) {
-		if (strncmp(envname, env->name, i) == 0) {
-			return (env->val);
-		}
-		env++;
-	}
-	return (NULL);
-}
-
-static inline unsigned char str2hexnum(unsigned char c)
-{
-	if (c >= '0' && c <= '9')
-		return c - '0';
-	if (c >= 'a' && c <= 'f')
-		return c - 'a' + 10;
-	return 0;		/* foo */
-}
-
-static inline void str2eaddr(unsigned char *ea, unsigned char *str)
-{
-	int i;
-
-	for (i = 0; i < 6; i++) {
-		unsigned char num;
-
-		if ((*str == '.') || (*str == ':'))
-			str++;
-		num = str2hexnum(*str++) << 4;
-		num |= (str2hexnum(*str++));
-		ea[i] = num;
-	}
-}
-
-int get_ethernet_addr(char *ethernet_addr)
-{
-	char *ethaddr_str;
-
-	ethaddr_str = prom_getenv("ethaddr");
-	if (!ethaddr_str) {
-		printk("ethaddr not set in boot prom\n");
-		return -1;
-	}
-	str2eaddr(ethernet_addr, ethaddr_str);
-
-	if (init_debug > 1) {
-		int i;
-		printk("get_ethernet_addr: ");
-		for (i = 0; i < 5; i++)
-			printk("%02x:",
-			       (unsigned char) *(ethernet_addr + i));
-		printk("%02x\n", *(ethernet_addr + i));
-	}
-
-	return 0;
-}
-
-const char *get_system_type(void)
-{
-	return "Galileo EV96100";
-}
-
-void __init prom_init(void)
-{
-	volatile unsigned char *uart;
-	char ppbuf[8];
-
-	prom_argc = fw_arg0;
-	prom_argv = (char **) fw_arg1;
-	prom_envp = (char **) fw_arg2;
-
-	mips_machgroup = MACH_GROUP_GALILEO;
-	mips_machtype = MACH_EV96100;
-
-	prom_init_cmdline();
-
-	/* 32 MB upgradable */
-	add_memory_region(0, 32 << 20, BOOT_MEM_RAM);
-}
diff --git a/arch/mips/galileo-boards/ev96100/irq.c b/arch/mips/galileo-boards/ev96100/irq.c
deleted file mode 100644
index ee5d6720f23b..000000000000
--- a/arch/mips/galileo-boards/ev96100/irq.c
+++ /dev/null
@@ -1,77 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/atlas/atlas_int.c.
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/errno.h>
-#include <linux/init.h>
-#include <linux/kernel_stat.h>
-#include <linux/irq.h>
-#include <linux/module.h>
-#include <linux/signal.h>
-#include <linux/sched.h>
-#include <linux/types.h>
-#include <linux/interrupt.h>
-#include <asm/irq_cpu.h>
-
-static inline unsigned int ffz8(unsigned int word)
-{
-	unsigned long k;
-
-	k = 7;
-	if (word & 0x0fUL) { k -= 4;  word <<= 4;  }
-	if (word & 0x30UL) { k -= 2;  word <<= 2;  }
-	if (word & 0x40UL) { k -= 1; }
-
-	return k;
-}
-
-extern void mips_timer_interrupt(struct pt_regs *regs);
-
-asmlinkage void ev96100_cpu_irq(unsigned int pending, struct pt_regs *regs)
-{
-	do_IRQ(ffz8(pending >> 8), regs);
-}
-
-asmlinkage void plat_irq_dispatch(struct pt_regs *regs)
-{
-	unsigned int pending = read_c0_cause() & read_c0_status() & ST0_IM;
-
-	if (pending & CAUSEF_IP7)
-		mips_timer_interrupt(regs);
-	else if (pending)
-		ev96100_cpu_irq(pending, regs);
-	else
-		spurious_interrupt(regs);
-}
-
-void __init arch_init_irq(void)
-{
-	mips_cpu_irq_init(0);
-}
diff --git a/arch/mips/galileo-boards/ev96100/puts.c b/arch/mips/galileo-boards/ev96100/puts.c
deleted file mode 100644
index 49dc6d137b9c..000000000000
--- a/arch/mips/galileo-boards/ev96100/puts.c
+++ /dev/null
@@ -1,138 +0,0 @@
-
-/*
- * Debug routines which directly access the uart.
- */
-
-#include <linux/types.h>
-#include <asm/gt64120.h>
-
-
-//#define SERIAL_BASE    EV96100_UART0_REGS_BASE
-#define SERIAL_BASE    0xBD000020
-#define NS16550_BASE   SERIAL_BASE
-
-#define SERA_CMD       0x0D
-#define SERA_DATA      0x08
-//#define SERB_CMD       0x05
-#define SERB_CMD       20
-#define SERB_DATA      0x00
-#define TX_BUSY        0x20
-
-#define TIMEOUT    0xffff
-#undef SLOW_DOWN
-
-static const char digits[16] = "0123456789abcdef";
-static volatile unsigned char *const com1 = (unsigned char *) SERIAL_BASE;
-
-
-#ifdef SLOW_DOWN
-static inline void slow_down()
-{
-	int k;
-	for (k = 0; k < 10000; k++);
-}
-#else
-#define slow_down()
-#endif
-
-void putch(const unsigned char c)
-{
-	unsigned char ch;
-	int i = 0;
-
-	do {
-		ch = com1[SERB_CMD];
-		slow_down();
-		i++;
-		if (i > TIMEOUT) {
-			break;
-		}
-	} while (0 == (ch & TX_BUSY));
-	com1[SERB_DATA] = c;
-}
-
-void putchar(const unsigned char c)
-{
-	unsigned char ch;
-	int i = 0;
-
-	do {
-		ch = com1[SERB_CMD];
-		slow_down();
-		i++;
-		if (i > TIMEOUT) {
-			break;
-		}
-	} while (0 == (ch & TX_BUSY));
-	com1[SERB_DATA] = c;
-}
-
-void puts(unsigned char *cp)
-{
-	unsigned char ch;
-	int i = 0;
-
-	while (*cp) {
-		do {
-			ch = com1[SERB_CMD];
-			slow_down();
-			i++;
-			if (i > TIMEOUT) {
-				break;
-			}
-		} while (0 == (ch & TX_BUSY));
-		com1[SERB_DATA] = *cp++;
-	}
-	putch('\r');
-	putch('\n');
-}
-
-void fputs(unsigned char *cp)
-{
-	unsigned char ch;
-	int i = 0;
-
-	while (*cp) {
-
-		do {
-			ch = com1[SERB_CMD];
-			slow_down();
-			i++;
-			if (i > TIMEOUT) {
-				break;
-			}
-		} while (0 == (ch & TX_BUSY));
-		com1[SERB_DATA] = *cp++;
-	}
-}
-
-
-void put64(uint64_t ul)
-{
-	int cnt;
-	unsigned ch;
-
-	cnt = 16;		/* 16 nibbles in a 64 bit long */
-	putch('0');
-	putch('x');
-	do {
-		cnt--;
-		ch = (unsigned char) (ul >> cnt * 4) & 0x0F;
-		putch(digits[ch]);
-	} while (cnt > 0);
-}
-
-void put32(unsigned u)
-{
-	int cnt;
-	unsigned ch;
-
-	cnt = 8;		/* 8 nibbles in a 32 bit long */
-	putch('0');
-	putch('x');
-	do {
-		cnt--;
-		ch = (unsigned char) (u >> cnt * 4) & 0x0F;
-		putch(digits[ch]);
-	} while (cnt > 0);
-}
diff --git a/arch/mips/galileo-boards/ev96100/reset.c b/arch/mips/galileo-boards/ev96100/reset.c
deleted file mode 100644
index 5ef9b7f896e6..000000000000
--- a/arch/mips/galileo-boards/ev96100/reset.c
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 reset routines.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/generic/reset.c
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/processor.h>
-#include <asm/reboot.h>
-#include <asm/system.h>
-#include <asm/gt64120.h>
-
-static void mips_machine_restart(char *command);
-static void mips_machine_halt(void);
-
-static void mips_machine_restart(char *command)
-{
-	set_c0_status(ST0_BEV | ST0_ERL);
-	change_c0_config(CONF_CM_CMASK, CONF_CM_UNCACHED);
-	flush_cache_all();
-	write_c0_wired(0);
-	__asm__ __volatile__("jr\t%0"::"r"(0xbfc00000));
-	while (1);
-}
-
-static void mips_machine_halt(void)
-{
-	printk(KERN_NOTICE "You can safely turn off the power\n");
-	while (1)
-		__asm__(".set\tmips3\n\t"
-	                "wait\n\t"
-			".set\tmips0");
-}
-
-void mips_reboot_setup(void)
-{
-	_machine_restart = mips_machine_restart;
-	_machine_halt = mips_machine_halt;
-}
diff --git a/arch/mips/galileo-boards/ev96100/setup.c b/arch/mips/galileo-boards/ev96100/setup.c
deleted file mode 100644
index 639ad5562c63..000000000000
--- a/arch/mips/galileo-boards/ev96100/setup.c
+++ /dev/null
@@ -1,159 +0,0 @@
-/*
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 setup.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/atlas/atlas_setup.c.
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
-#include <linux/string.h>
-#include <linux/ctype.h>
-#include <linux/pci.h>
-
-#include <asm/cpu.h>
-#include <asm/bootinfo.h>
-#include <asm/mipsregs.h>
-#include <asm/irq.h>
-#include <asm/delay.h>
-#include <asm/gt64120.h>
-#include <asm/galileo-boards/ev96100int.h>
-
-
-extern char *__init prom_getcmdline(void);
-
-extern void mips_reboot_setup(void);
-
-unsigned char mac_0_1[12];
-
-void __init plat_mem_setup(void)
-{
-	unsigned int config = read_c0_config();
-	unsigned int status = read_c0_status();
-	unsigned int info = read_c0_info();
-	u32 tmp;
-
-	char *argptr;
-
-	clear_c0_status(ST0_FR);
-
-	if (config & 0x8)
-		printk("Secondary cache is enabled\n");
-	else
-		printk("Secondary cache is disabled\n");
-
-	if (status & (1 << 27))
-		printk("User-mode cache ops enabled\n");
-	else
-		printk("User-mode cache ops disabled\n");
-
-	printk("CP0 info reg: %x\n", (unsigned) info);
-	if (info & (1 << 28))
-		printk("burst mode Scache RAMS\n");
-	else
-		printk("pipelined Scache RAMS\n");
-
-	if (info & 0x1)
-		printk("Atomic Enable is set\n");
-
-	argptr = prom_getcmdline();
-#ifdef CONFIG_SERIAL_CONSOLE
-	if (strstr(argptr, "console=") == NULL) {
-		argptr = prom_getcmdline();
-		strcat(argptr, " console=ttyS0,115200");
-	}
-#endif
-
-	mips_reboot_setup();
-
-	set_io_port_base(KSEG1);
-	ioport_resource.start = GT_PCI_IO_BASE;
-	ioport_resource.end = GT_PCI_IO_BASE + 0x01ffffff;
-
-#ifdef CONFIG_BLK_DEV_INITRD
-	ROOT_DEV = MKDEV(RAMDISK_MAJOR, 0);
-#endif
-
-
-	/*
-	 * Setup GT controller master bit so we can do config cycles
-	 */
-
-	/* Clear cause register bits */
-	GT_WRITE(GT_INTRCAUSE_OFS, ~(GT_INTRCAUSE_MASABORT0_BIT |
-				     GT_INTRCAUSE_TARABORT0_BIT));
-	/* Setup address */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_COMMAND / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-
-	udelay(2);
-	tmp = GT_READ(GT_PCI0_CFGDATA_OFS);
-
-	tmp |= (PCI_COMMAND_IO | PCI_COMMAND_MEMORY |
-		PCI_COMMAND_MASTER | PCI_COMMAND_SERR);
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_COMMAND / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-	udelay(2);
-	GT_WRITE(GT_PCI0_CFGDATA_OFS, tmp);
-
-	/* Setup address */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_COMMAND / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-
-	udelay(2);
-	tmp = GT_READ(GT_PCI0_CFGDATA_OFS);
-}
-
-unsigned short get_gt_devid(void)
-{
-	u32 gt_devid;
-
-	/* Figure out if this is a gt96100 or gt96100A */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_VENDOR_ID / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-
-	udelay(4);
-	gt_devid = GT_READ(GT_PCI0_CFGDATA_OFS);
-
-	return gt_devid >> 16;
-}
diff --git a/arch/mips/galileo-boards/ev96100/time.c b/arch/mips/galileo-boards/ev96100/time.c
deleted file mode 100644
index 8cbe8426491a..000000000000
--- a/arch/mips/galileo-boards/ev96100/time.c
+++ /dev/null
@@ -1,88 +0,0 @@
-/*
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 rtc routines.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/atlas/atlas_rtc.c.
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/kernel_stat.h>
-#include <linux/module.h>
-#include <linux/sched.h>
-#include <linux/spinlock.h>
-#include <linux/timex.h>
-
-#include <asm/mipsregs.h>
-#include <asm/ptrace.h>
-#include <asm/time.h>
-
-
-#define ALLINTS (IE_IRQ0 | IE_IRQ1 | IE_IRQ2 | IE_IRQ3 | IE_IRQ4 | IE_IRQ5)
-
-extern volatile unsigned long wall_jiffies;
-unsigned long missed_heart_beats = 0;
-
-static unsigned long r4k_offset; /* Amount to increment compare reg each time */
-static unsigned long r4k_cur;    /* What counter should be at next timer irq */
-
-static inline void ack_r4ktimer(unsigned long newval)
-{
-	write_c0_compare(newval);
-}
-
-/*
- * There are a lot of conceptually broken versions of the MIPS timer interrupt
- * handler floating around.  This one is rather different, but the algorithm
- * is probably more robust.
- */
-void mips_timer_interrupt(struct pt_regs *regs)
-{
-        int irq = 7; /* FIX ME */
-
-	if (r4k_offset == 0) {
-            goto null;
-        }
-
-	do {
-		kstat_this_cpu.irqs[irq]++;
-		do_timer(regs);
-#ifndef CONFIG_SMP
-		update_process_times(user_mode(regs));
-#endif
-		r4k_cur += r4k_offset;
-		ack_r4ktimer(r4k_cur);
-
-	} while (((unsigned long)read_c0_count()
-                    - r4k_cur) < 0x7fffffff);
-	return;
-
-null:
-	ack_r4ktimer(0);
-}
diff --git a/arch/mips/pci/Makefile b/arch/mips/pci/Makefile
index 368d3d896165..edefa97b2330 100644
--- a/arch/mips/pci/Makefile
+++ b/arch/mips/pci/Makefile
@@ -11,7 +11,6 @@ obj-$(CONFIG_ITE_BOARD_GEN)	+= ops-it8172.o
 obj-$(CONFIG_MIPS_BONITO64)	+= ops-bonito64.o
 obj-$(CONFIG_MIPS_GT64111)	+= ops-gt64111.o
 obj-$(CONFIG_MIPS_GT64120)	+= ops-gt64120.o
-obj-$(CONFIG_MIPS_GT96100)	+= ops-gt96100.o
 obj-$(CONFIG_PCI_MARVELL)	+= ops-marvell.o
 obj-$(CONFIG_MIPS_MSC)		+= ops-msc.o
 obj-$(CONFIG_MIPS_NILE4)	+= ops-nile4.o
@@ -29,7 +28,6 @@ obj-$(CONFIG_LASAT)		+= pci-lasat.o
 obj-$(CONFIG_MIPS_ATLAS)	+= fixup-atlas.o
 obj-$(CONFIG_MIPS_COBALT)	+= fixup-cobalt.o
 obj-$(CONFIG_MIPS_EV64120)	+= fixup-ev64120.o
-obj-$(CONFIG_MIPS_EV96100)	+= fixup-ev96100.o pci-ev96100.o
 obj-$(CONFIG_MIPS_ITE8172)	+= fixup-ite8172g.o
 obj-$(CONFIG_MIPS_IVR)		+= fixup-ivr.o
 obj-$(CONFIG_SOC_AU1500)	+= fixup-au1000.o ops-au1000.o
diff --git a/arch/mips/pci/fixup-ev96100.c b/arch/mips/pci/fixup-ev96100.c
deleted file mode 100644
index e2bc977b6d58..000000000000
--- a/arch/mips/pci/fixup-ev96100.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- *
- * BRIEF MODULE DESCRIPTION
- *	EV96100 Board specific pci fixups.
- *
- * Copyright 2001 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/types.h>
-#include <linux/pci.h>
-
-static char irq_tab_ev96100[][5] __initdata = {
- [8] = { 0, 5, 5, 5, 5 },
- [9] = { 0, 2, 2, 2, 2 }
-};
-
-int __init pcibios_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	return irq_tab_ev96100[slot][pin];
-}
-
-/* Do platform specific device initialization at pci_enable_device() time */
-int pcibios_plat_dev_init(struct pci_dev *dev)
-{
-	return 0;
-}
diff --git a/arch/mips/pci/ops-gt96100.c b/arch/mips/pci/ops-gt96100.c
deleted file mode 100644
index 9e4ea6627e21..000000000000
--- a/arch/mips/pci/ops-gt96100.c
+++ /dev/null
@@ -1,169 +0,0 @@
-/*
- *
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 board specific pci support.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/generic/pci.c
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-
-#include <asm/delay.h>
-#include <asm/gt64120.h>
-#include <asm/galileo-boards/ev96100.h>
-
-#define PCI_ACCESS_READ  0
-#define PCI_ACCESS_WRITE 1
-
-static int static gt96100_config_access(unsigned char access_type,
-	struct pci_bus *bus, unsigned int devfn, int where, u32 * data)
-{
-	unsigned char bus = bus->number;
-	u32 intr;
-
-	/*
-	 * Because of a bug in the galileo (for slot 31).
-	 */
-	if (bus == 0 && devfn >= PCI_DEVFN(31, 0))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	/* Clear cause register bits */
-	GT_WRITE(GT_INTRCAUSE_OFS, ~(GT_INTRCAUSE_MASABORT0_BIT |
-				     GT_INTRCAUSE_TARABORT0_BIT));
-
-	/* Setup address */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (bus << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (devfn << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((where / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-	udelay(2);
-
-
-	if (access_type == PCI_ACCESS_WRITE) {
-		if (devfn != 0)
-			*data = le32_to_cpu(*data);
-		GT_WRITE(GT_PCI0_CFGDATA_OFS, *data);
-	} else {
-		*data = GT_READ(GT_PCI0_CFGDATA_OFS);
-		if (devfn != 0)
-			*data = le32_to_cpu(*data);
-	}
-
-	udelay(2);
-
-	/* Check for master or target abort */
-	intr = GT_READ(GT_INTRCAUSE_OFS);
-
-	if (intr & (GT_INTRCAUSE_MASABORT0_BIT | GT_INTRCAUSE_TARABORT0_BIT)) {
-		/* Error occured */
-
-		/* Clear bits */
-		GT_WRITE(GT_INTRCAUSE_OFS, ~(GT_INTRCAUSE_MASABORT0_BIT |
-					     GT_INTRCAUSE_TARABORT0_BIT));
-		return -1;
-	}
-	return 0;
-}
-
-/*
- * We can't address 8 and 16 bit words directly.  Instead we have to
- * read/write a 32bit word and mask/modify the data we actually want.
- */
-static int gt96100_pcibios_read(struct pci_bus *bus, unsigned int devfn,
-	int where, int size, u32 * val)
-{
-	u32 data = 0;
-
-	if (gt96100_config_access(PCI_ACCESS_READ, bus, devfn, where, &data))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	switch (size) {
-	case 1:
-		*val = (data >> ((where & 3) << 3)) & 0xff;
-		break;
-
-	case 2:
-		*val = (data >> ((where & 3) << 3)) & 0xffff;
-		break;
-
-	case 4:
-		*val = data;
-		break;
-	}
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int gt96100_pcibios_write(struct pci_bus *bus, unsigned int devfn,
-	int where, int size, u32 val)
-{
-	u32 data = 0;
-
-	switch (size) {
-	case 1:
-		if (gt96100_config_access(PCI_ACCESS_READ, bus, devfn, where, &data))
-			return -1;
-
-		data = (data & ~(0xff << ((where & 3) << 3))) |
-		       (val << ((where & 3) << 3));
-
-		if (gt96100_config_access(PCI_ACCESS_WRITE, bus, devfn, where, &data))
-			return -1;
-
-		return PCIBIOS_SUCCESSFUL;
-
-	case 2:
-		if (gt96100_config_access(PCI_ACCESS_READ, bus, devfn, where, &data))
-			return -1;
-
-		data = (data & ~(0xffff << ((where & 3) << 3))) |
-		       (val << ((where & 3) << 3));
-
-		if (gt96100_config_access(PCI_ACCESS_WRITE, dev, where, &data))
-			return -1;
-
-
-		return PCIBIOS_SUCCESSFUL;
-
-	case 4:
-		if (gt96100_config_access(PCI_ACCESS_WRITE, dev, where, &val))
-			return -1;
-
-		return PCIBIOS_SUCCESSFUL;
-	}
-}
-
-struct pci_ops gt96100_pci_ops = {
-	.read	= gt96100_pcibios_read,
-	.write	= gt96100_pcibios_write
-};
diff --git a/arch/mips/pci/pci-ev96100.c b/arch/mips/pci/pci-ev96100.c
deleted file mode 100644
index f9457ea00def..000000000000
--- a/arch/mips/pci/pci-ev96100.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- * Copyright (C) 2004 by Ralf Baechle (ralf@linux-mips.org)
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-
-static struct resource pci_io_resource = {
-	.name	= "io pci IO space",
-	.start	= 0x10000000,
-	.end	= 0x11ffffff,
-	.flags	= IORESOURCE_IO
-};
-
-static struct resource pci_mem_resource = {
-	.name	= "ext pci memory space",
-	.start	= 0x12000000,
-	.end	= 0x13ffffff,
-	.flags	= IORESOURCE_MEM
-};
-
-extern struct pci_ops gt96100_pci_ops;
-
-struct pci_controller ev96100_controller = {
-	.pci_ops	= &gt96100_pci_ops,
-	.io_resource	= &pci_io_resource,
-	.mem_resource	= &pci_mem_resource,
-};
-
-static void ev96100_pci_init(void)
-{
-	register_pci_controller(&ev96100_controller);
-}
-
-arch_initcall(ev96100_pci_init);
diff --git a/include/asm-mips/bootinfo.h b/include/asm-mips/bootinfo.h
index 3b745e76f429..78c35ec46362 100644
--- a/include/asm-mips/bootinfo.h
+++ b/include/asm-mips/bootinfo.h
@@ -112,8 +112,7 @@
  * Valid machtype for group GALILEO
  */
 #define MACH_GROUP_GALILEO     11	/* Galileo Eval Boards		*/
-#define  MACH_EV96100		0	/* EV96100 */
-#define  MACH_EV64120A		1	/* EV64120A */
+#define  MACH_EV64120A		0	/* EV64120A */
 
 /*
  * Valid machtype for group MOMENCO
diff --git a/include/asm-mips/galileo-boards/gt96100.h b/include/asm-mips/galileo-boards/gt96100.h
deleted file mode 100644
index aabd1b629c19..000000000000
--- a/include/asm-mips/galileo-boards/gt96100.h
+++ /dev/null
@@ -1,427 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	stevel@mvista.com or source@mvista.com
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, write to the Free Software Foundation, Inc.,
- *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
- *
- * Register offsets of the MIPS GT96100 Advanced Communication Controller.
- */
-#ifndef _GT96100_H
-#define _GT96100_H
-
-/*
- * Galileo GT96100 internal register base.
- */
-#define MIPS_GT96100_BASE (KSEG1ADDR(0x14000000))
-
-#define GT96100_WRITE(ofs, data) \
-    *(volatile u32 *)(MIPS_GT96100_BASE+ofs) = cpu_to_le32(data)
-#define GT96100_READ(ofs) \
-    le32_to_cpu(*(volatile u32 *)(MIPS_GT96100_BASE+ofs))
-
-#define GT96100_ETH_IO_SIZE 0x4000
-
-/************************************************************************
- *  Register offset addresses follow
- ************************************************************************/
-
-/* CPU Interface Control Registers */
-#define GT96100_CPU_INTERF_CONFIG 0x000000
-
-/* Ethernet Ports */
-#define GT96100_ETH_PHY_ADDR_REG             0x080800
-#define GT96100_ETH_SMI_REG                  0x080810
-/*
-  These are offsets to port 0 registers. Add GT96100_ETH_IO_SIZE to
-  get offsets to port 1 registers.
-*/
-#define GT96100_ETH_PORT_CONFIG          0x084800
-#define GT96100_ETH_PORT_CONFIG_EXT      0x084808
-#define GT96100_ETH_PORT_COMM            0x084810
-#define GT96100_ETH_PORT_STATUS          0x084818
-#define GT96100_ETH_SER_PARAM            0x084820
-#define GT96100_ETH_HASH_TBL_PTR         0x084828
-#define GT96100_ETH_FLOW_CNTRL_SRC_ADDR_L    0x084830
-#define GT96100_ETH_FLOW_CNTRL_SRC_ADDR_H    0x084838
-#define GT96100_ETH_SDMA_CONFIG          0x084840
-#define GT96100_ETH_SDMA_COMM            0x084848
-#define GT96100_ETH_INT_CAUSE            0x084850
-#define GT96100_ETH_INT_MASK             0x084858
-#define GT96100_ETH_1ST_RX_DESC_PTR0         0x084880
-#define GT96100_ETH_1ST_RX_DESC_PTR1         0x084884
-#define GT96100_ETH_1ST_RX_DESC_PTR2         0x084888
-#define GT96100_ETH_1ST_RX_DESC_PTR3         0x08488C
-#define GT96100_ETH_CURR_RX_DESC_PTR0        0x0848A0
-#define GT96100_ETH_CURR_RX_DESC_PTR1        0x0848A4
-#define GT96100_ETH_CURR_RX_DESC_PTR2        0x0848A8
-#define GT96100_ETH_CURR_RX_DESC_PTR3        0x0848AC
-#define GT96100_ETH_CURR_TX_DESC_PTR0        0x0848E0
-#define GT96100_ETH_CURR_TX_DESC_PTR1        0x0848E4
-#define GT96100_ETH_MIB_COUNT_BASE           0x085800
-
-/* SDMAs */
-#define GT96100_SDMA_GROUP_CONFIG           0x101AF0
-/* SDMA Group 0 */
-#define GT96100_SDMA_G0_CHAN0_CONFIG        0x000900
-#define GT96100_SDMA_G0_CHAN0_COMM          0x000908
-#define GT96100_SDMA_G0_CHAN0_RX_DESC_BASE      0x008900
-#define GT96100_SDMA_G0_CHAN0_CURR_RX_DESC_PTR  0x008910
-#define GT96100_SDMA_G0_CHAN0_TX_DESC_BASE      0x00C900
-#define GT96100_SDMA_G0_CHAN0_CURR_TX_DESC_PTR  0x00C910
-#define GT96100_SDMA_G0_CHAN0_1ST_TX_DESC_PTR   0x00C914
-#define GT96100_SDMA_G0_CHAN1_CONFIG        0x010900
-#define GT96100_SDMA_G0_CHAN1_COMM          0x010908
-#define GT96100_SDMA_G0_CHAN1_RX_DESC_BASE      0x018900
-#define GT96100_SDMA_G0_CHAN1_CURR_RX_DESC_PTR  0x018910
-#define GT96100_SDMA_G0_CHAN1_TX_DESC_BASE      0x01C900
-#define GT96100_SDMA_G0_CHAN1_CURR_TX_DESC_PTR  0x01C910
-#define GT96100_SDMA_G0_CHAN1_1ST_TX_DESC_PTR   0x01C914
-#define GT96100_SDMA_G0_CHAN2_CONFIG        0x020900
-#define GT96100_SDMA_G0_CHAN2_COMM          0x020908
-#define GT96100_SDMA_G0_CHAN2_RX_DESC_BASE      0x028900
-#define GT96100_SDMA_G0_CHAN2_CURR_RX_DESC_PTR  0x028910
-#define GT96100_SDMA_G0_CHAN2_TX_DESC_BASE      0x02C900
-#define GT96100_SDMA_G0_CHAN2_CURR_TX_DESC_PTR  0x02C910
-#define GT96100_SDMA_G0_CHAN2_1ST_TX_DESC_PTR   0x02C914
-#define GT96100_SDMA_G0_CHAN3_CONFIG        0x030900
-#define GT96100_SDMA_G0_CHAN3_COMM          0x030908
-#define GT96100_SDMA_G0_CHAN3_RX_DESC_BASE      0x038900
-#define GT96100_SDMA_G0_CHAN3_CURR_RX_DESC_PTR  0x038910
-#define GT96100_SDMA_G0_CHAN3_TX_DESC_BASE      0x03C900
-#define GT96100_SDMA_G0_CHAN3_CURR_TX_DESC_PTR  0x03C910
-#define GT96100_SDMA_G0_CHAN3_1ST_TX_DESC_PTR   0x03C914
-#define GT96100_SDMA_G0_CHAN4_CONFIG        0x040900
-#define GT96100_SDMA_G0_CHAN4_COMM          0x040908
-#define GT96100_SDMA_G0_CHAN4_RX_DESC_BASE      0x048900
-#define GT96100_SDMA_G0_CHAN4_CURR_RX_DESC_PTR  0x048910
-#define GT96100_SDMA_G0_CHAN4_TX_DESC_BASE      0x04C900
-#define GT96100_SDMA_G0_CHAN4_CURR_TX_DESC_PTR  0x04C910
-#define GT96100_SDMA_G0_CHAN4_1ST_TX_DESC_PTR   0x04C914
-#define GT96100_SDMA_G0_CHAN5_CONFIG        0x050900
-#define GT96100_SDMA_G0_CHAN5_COMM          0x050908
-#define GT96100_SDMA_G0_CHAN5_RX_DESC_BASE      0x058900
-#define GT96100_SDMA_G0_CHAN5_CURR_RX_DESC_PTR  0x058910
-#define GT96100_SDMA_G0_CHAN5_TX_DESC_BASE      0x05C900
-#define GT96100_SDMA_G0_CHAN5_CURR_TX_DESC_PTR  0x05C910
-#define GT96100_SDMA_G0_CHAN5_1ST_TX_DESC_PTR   0x05C914
-#define GT96100_SDMA_G0_CHAN6_CONFIG        0x060900
-#define GT96100_SDMA_G0_CHAN6_COMM          0x060908
-#define GT96100_SDMA_G0_CHAN6_RX_DESC_BASE      0x068900
-#define GT96100_SDMA_G0_CHAN6_CURR_RX_DESC_PTR  0x068910
-#define GT96100_SDMA_G0_CHAN6_TX_DESC_BASE      0x06C900
-#define GT96100_SDMA_G0_CHAN6_CURR_TX_DESC_PTR  0x06C910
-#define GT96100_SDMA_G0_CHAN6_1ST_TX_DESC_PTR   0x06C914
-#define GT96100_SDMA_G0_CHAN7_CONFIG        0x070900
-#define GT96100_SDMA_G0_CHAN7_COMM          0x070908
-#define GT96100_SDMA_G0_CHAN7_RX_DESC_BASE      0x078900
-#define GT96100_SDMA_G0_CHAN7_CURR_RX_DESC_PTR  0x078910
-#define GT96100_SDMA_G0_CHAN7_TX_DESC_BASE      0x07C900
-#define GT96100_SDMA_G0_CHAN7_CURR_TX_DESC_PTR  0x07C910
-#define GT96100_SDMA_G0_CHAN7_1ST_TX_DESC_PTR   0x07C914
-/* SDMA Group 1 */
-#define GT96100_SDMA_G1_CHAN0_CONFIG        0x100900
-#define GT96100_SDMA_G1_CHAN0_COMM          0x100908
-#define GT96100_SDMA_G1_CHAN0_RX_DESC_BASE      0x108900
-#define GT96100_SDMA_G1_CHAN0_CURR_RX_DESC_PTR  0x108910
-#define GT96100_SDMA_G1_CHAN0_TX_DESC_BASE      0x10C900
-#define GT96100_SDMA_G1_CHAN0_CURR_TX_DESC_PTR  0x10C910
-#define GT96100_SDMA_G1_CHAN0_1ST_TX_DESC_PTR   0x10C914
-#define GT96100_SDMA_G1_CHAN1_CONFIG        0x110900
-#define GT96100_SDMA_G1_CHAN1_COMM          0x110908
-#define GT96100_SDMA_G1_CHAN1_RX_DESC_BASE      0x118900
-#define GT96100_SDMA_G1_CHAN1_CURR_RX_DESC_PTR  0x118910
-#define GT96100_SDMA_G1_CHAN1_TX_DESC_BASE      0x11C900
-#define GT96100_SDMA_G1_CHAN1_CURR_TX_DESC_PTR  0x11C910
-#define GT96100_SDMA_G1_CHAN1_1ST_TX_DESC_PTR   0x11C914
-#define GT96100_SDMA_G1_CHAN2_CONFIG        0x120900
-#define GT96100_SDMA_G1_CHAN2_COMM          0x120908
-#define GT96100_SDMA_G1_CHAN2_RX_DESC_BASE      0x128900
-#define GT96100_SDMA_G1_CHAN2_CURR_RX_DESC_PTR  0x128910
-#define GT96100_SDMA_G1_CHAN2_TX_DESC_BASE      0x12C900
-#define GT96100_SDMA_G1_CHAN2_CURR_TX_DESC_PTR  0x12C910
-#define GT96100_SDMA_G1_CHAN2_1ST_TX_DESC_PTR   0x12C914
-#define GT96100_SDMA_G1_CHAN3_CONFIG        0x130900
-#define GT96100_SDMA_G1_CHAN3_COMM          0x130908
-#define GT96100_SDMA_G1_CHAN3_RX_DESC_BASE      0x138900
-#define GT96100_SDMA_G1_CHAN3_CURR_RX_DESC_PTR  0x138910
-#define GT96100_SDMA_G1_CHAN3_TX_DESC_BASE      0x13C900
-#define GT96100_SDMA_G1_CHAN3_CURR_TX_DESC_PTR  0x13C910
-#define GT96100_SDMA_G1_CHAN3_1ST_TX_DESC_PTR   0x13C914
-#define GT96100_SDMA_G1_CHAN4_CONFIG        0x140900
-#define GT96100_SDMA_G1_CHAN4_COMM          0x140908
-#define GT96100_SDMA_G1_CHAN4_RX_DESC_BASE      0x148900
-#define GT96100_SDMA_G1_CHAN4_CURR_RX_DESC_PTR  0x148910
-#define GT96100_SDMA_G1_CHAN4_TX_DESC_BASE      0x14C900
-#define GT96100_SDMA_G1_CHAN4_CURR_TX_DESC_PTR  0x14C910
-#define GT96100_SDMA_G1_CHAN4_1ST_TX_DESC_PTR   0x14C914
-#define GT96100_SDMA_G1_CHAN5_CONFIG        0x150900
-#define GT96100_SDMA_G1_CHAN5_COMM          0x150908
-#define GT96100_SDMA_G1_CHAN5_RX_DESC_BASE      0x158900
-#define GT96100_SDMA_G1_CHAN5_CURR_RX_DESC_PTR  0x158910
-#define GT96100_SDMA_G1_CHAN5_TX_DESC_BASE      0x15C900
-#define GT96100_SDMA_G1_CHAN5_CURR_TX_DESC_PTR  0x15C910
-#define GT96100_SDMA_G1_CHAN5_1ST_TX_DESC_PTR   0x15C914
-#define GT96100_SDMA_G1_CHAN6_CONFIG        0x160900
-#define GT96100_SDMA_G1_CHAN6_COMM          0x160908
-#define GT96100_SDMA_G1_CHAN6_RX_DESC_BASE      0x168900
-#define GT96100_SDMA_G1_CHAN6_CURR_RX_DESC_PTR  0x168910
-#define GT96100_SDMA_G1_CHAN6_TX_DESC_BASE      0x16C900
-#define GT96100_SDMA_G1_CHAN6_CURR_TX_DESC_PTR  0x16C910
-#define GT96100_SDMA_G1_CHAN6_1ST_TX_DESC_PTR   0x16C914
-#define GT96100_SDMA_G1_CHAN7_CONFIG        0x170900
-#define GT96100_SDMA_G1_CHAN7_COMM          0x170908
-#define GT96100_SDMA_G1_CHAN7_RX_DESC_BASE      0x178900
-#define GT96100_SDMA_G1_CHAN7_CURR_RX_DESC_PTR  0x178910
-#define GT96100_SDMA_G1_CHAN7_TX_DESC_BASE      0x17C900
-#define GT96100_SDMA_G1_CHAN7_CURR_TX_DESC_PTR  0x17C910
-#define GT96100_SDMA_G1_CHAN7_1ST_TX_DESC_PTR   0x17C914
-/*  MPSCs  */
-#define GT96100_MPSC0_MAIN_CONFIG_LOW   0x000A00
-#define GT96100_MPSC0_MAIN_CONFIG_HIGH  0x000A04
-#define GT96100_MPSC0_PROTOCOL_CONFIG   0x000A08
-#define GT96100_MPSC_CHAN0_REG1         0x000A0C
-#define GT96100_MPSC_CHAN0_REG2         0x000A10
-#define GT96100_MPSC_CHAN0_REG3         0x000A14
-#define GT96100_MPSC_CHAN0_REG4         0x000A18
-#define GT96100_MPSC_CHAN0_REG5         0x000A1C
-#define GT96100_MPSC_CHAN0_REG6         0x000A20
-#define GT96100_MPSC_CHAN0_REG7         0x000A24
-#define GT96100_MPSC_CHAN0_REG8         0x000A28
-#define GT96100_MPSC_CHAN0_REG9         0x000A2C
-#define GT96100_MPSC_CHAN0_REG10        0x000A30
-#define GT96100_MPSC_CHAN0_REG11        0x000A34
-#define GT96100_MPSC1_MAIN_CONFIG_LOW   0x008A00
-#define GT96100_MPSC1_MAIN_CONFIG_HIGH  0x008A04
-#define GT96100_MPSC1_PROTOCOL_CONFIG   0x008A08
-#define GT96100_MPSC_CHAN1_REG1         0x008A0C
-#define GT96100_MPSC_CHAN1_REG2         0x008A10
-#define GT96100_MPSC_CHAN1_REG3         0x008A14
-#define GT96100_MPSC_CHAN1_REG4         0x008A18
-#define GT96100_MPSC_CHAN1_REG5         0x008A1C
-#define GT96100_MPSC_CHAN1_REG6         0x008A20
-#define GT96100_MPSC_CHAN1_REG7         0x008A24
-#define GT96100_MPSC_CHAN1_REG8         0x008A28
-#define GT96100_MPSC_CHAN1_REG9         0x008A2C
-#define GT96100_MPSC_CHAN1_REG10        0x008A30
-#define GT96100_MPSC_CHAN1_REG11        0x008A34
-#define GT96100_MPSC2_MAIN_CONFIG_LOW   0x010A00
-#define GT96100_MPSC2_MAIN_CONFIG_HIGH  0x010A04
-#define GT96100_MPSC2_PROTOCOL_CONFIG   0x010A08
-#define GT96100_MPSC_CHAN2_REG1         0x010A0C
-#define GT96100_MPSC_CHAN2_REG2         0x010A10
-#define GT96100_MPSC_CHAN2_REG3         0x010A14
-#define GT96100_MPSC_CHAN2_REG4         0x010A18
-#define GT96100_MPSC_CHAN2_REG5         0x010A1C
-#define GT96100_MPSC_CHAN2_REG6         0x010A20
-#define GT96100_MPSC_CHAN2_REG7         0x010A24
-#define GT96100_MPSC_CHAN2_REG8         0x010A28
-#define GT96100_MPSC_CHAN2_REG9         0x010A2C
-#define GT96100_MPSC_CHAN2_REG10        0x010A30
-#define GT96100_MPSC_CHAN2_REG11        0x010A34
-#define GT96100_MPSC3_MAIN_CONFIG_LOW   0x018A00
-#define GT96100_MPSC3_MAIN_CONFIG_HIGH  0x018A04
-#define GT96100_MPSC3_PROTOCOL_CONFIG   0x018A08
-#define GT96100_MPSC_CHAN3_REG1         0x018A0C
-#define GT96100_MPSC_CHAN3_REG2         0x018A10
-#define GT96100_MPSC_CHAN3_REG3         0x018A14
-#define GT96100_MPSC_CHAN3_REG4         0x018A18
-#define GT96100_MPSC_CHAN3_REG5         0x018A1C
-#define GT96100_MPSC_CHAN3_REG6         0x018A20
-#define GT96100_MPSC_CHAN3_REG7         0x018A24
-#define GT96100_MPSC_CHAN3_REG8         0x018A28
-#define GT96100_MPSC_CHAN3_REG9         0x018A2C
-#define GT96100_MPSC_CHAN3_REG10        0x018A30
-#define GT96100_MPSC_CHAN3_REG11        0x018A34
-#define GT96100_MPSC4_MAIN_CONFIG_LOW   0x020A00
-#define GT96100_MPSC4_MAIN_CONFIG_HIGH  0x020A04
-#define GT96100_MPSC4_PROTOCOL_CONFIG   0x020A08
-#define GT96100_MPSC_CHAN4_REG1         0x020A0C
-#define GT96100_MPSC_CHAN4_REG2         0x020A10
-#define GT96100_MPSC_CHAN4_REG3         0x020A14
-#define GT96100_MPSC_CHAN4_REG4         0x020A18
-#define GT96100_MPSC_CHAN4_REG5         0x020A1C
-#define GT96100_MPSC_CHAN4_REG6         0x020A20
-#define GT96100_MPSC_CHAN4_REG7         0x020A24
-#define GT96100_MPSC_CHAN4_REG8         0x020A28
-#define GT96100_MPSC_CHAN4_REG9         0x020A2C
-#define GT96100_MPSC_CHAN4_REG10        0x020A30
-#define GT96100_MPSC_CHAN4_REG11        0x020A34
-#define GT96100_MPSC5_MAIN_CONFIG_LOW   0x028A00
-#define GT96100_MPSC5_MAIN_CONFIG_HIGH  0x028A04
-#define GT96100_MPSC5_PROTOCOL_CONFIG   0x028A08
-#define GT96100_MPSC_CHAN5_REG1         0x028A0C
-#define GT96100_MPSC_CHAN5_REG2         0x028A10
-#define GT96100_MPSC_CHAN5_REG3         0x028A14
-#define GT96100_MPSC_CHAN5_REG4         0x028A18
-#define GT96100_MPSC_CHAN5_REG5         0x028A1C
-#define GT96100_MPSC_CHAN5_REG6         0x028A20
-#define GT96100_MPSC_CHAN5_REG7         0x028A24
-#define GT96100_MPSC_CHAN5_REG8         0x028A28
-#define GT96100_MPSC_CHAN5_REG9         0x028A2C
-#define GT96100_MPSC_CHAN5_REG10        0x028A30
-#define GT96100_MPSC_CHAN5_REG11        0x028A34
-#define GT96100_MPSC6_MAIN_CONFIG_LOW   0x030A00
-#define GT96100_MPSC6_MAIN_CONFIG_HIGH  0x030A04
-#define GT96100_MPSC6_PROTOCOL_CONFIG   0x030A08
-#define GT96100_MPSC_CHAN6_REG1         0x030A0C
-#define GT96100_MPSC_CHAN6_REG2         0x030A10
-#define GT96100_MPSC_CHAN6_REG3         0x030A14
-#define GT96100_MPSC_CHAN6_REG4         0x030A18
-#define GT96100_MPSC_CHAN6_REG5         0x030A1C
-#define GT96100_MPSC_CHAN6_REG6         0x030A20
-#define GT96100_MPSC_CHAN6_REG7         0x030A24
-#define GT96100_MPSC_CHAN6_REG8         0x030A28
-#define GT96100_MPSC_CHAN6_REG9         0x030A2C
-#define GT96100_MPSC_CHAN6_REG10        0x030A30
-#define GT96100_MPSC_CHAN6_REG11        0x030A34
-#define GT96100_MPSC7_MAIN_CONFIG_LOW   0x038A00
-#define GT96100_MPSC7_MAIN_CONFIG_HIGH  0x038A04
-#define GT96100_MPSC7_PROTOCOL_CONFIG   0x038A08
-#define GT96100_MPSC_CHAN7_REG1         0x038A0C
-#define GT96100_MPSC_CHAN7_REG2         0x038A10
-#define GT96100_MPSC_CHAN7_REG3         0x038A14
-#define GT96100_MPSC_CHAN7_REG4         0x038A18
-#define GT96100_MPSC_CHAN7_REG5         0x038A1C
-#define GT96100_MPSC_CHAN7_REG6         0x038A20
-#define GT96100_MPSC_CHAN7_REG7         0x038A24
-#define GT96100_MPSC_CHAN7_REG8         0x038A28
-#define GT96100_MPSC_CHAN7_REG9         0x038A2C
-#define GT96100_MPSC_CHAN7_REG10        0x038A30
-#define GT96100_MPSC_CHAN7_REG11        0x038A34
-/*  FlexTDMs  */
-/* TDPR0 - Transmit Dual Port RAM. block size 0xff */
-#define GT96100_FXTDM0_TDPR0_BLK0_BASE  0x000B00
-#define GT96100_FXTDM0_TDPR0_BLK1_BASE  0x001B00
-#define GT96100_FXTDM0_TDPR0_BLK2_BASE  0x002B00
-#define GT96100_FXTDM0_TDPR0_BLK3_BASE  0x003B00
-/* RDPR0 - Receive Dual Port RAM. block size 0xff */
-#define GT96100_FXTDM0_RDPR0_BLK0_BASE  0x004B00
-#define GT96100_FXTDM0_RDPR0_BLK1_BASE  0x005B00
-#define GT96100_FXTDM0_RDPR0_BLK2_BASE  0x006B00
-#define GT96100_FXTDM0_RDPR0_BLK3_BASE  0x007B00
-#define GT96100_FXTDM0_TX_READ_PTR      0x008B00
-#define GT96100_FXTDM0_RX_READ_PTR      0x008B04
-#define GT96100_FXTDM0_CONFIG       0x008B08
-#define GT96100_FXTDM0_AUX_CHANA_TX 0x008B0C
-#define GT96100_FXTDM0_AUX_CHANA_RX 0x008B10
-#define GT96100_FXTDM0_AUX_CHANB_TX 0x008B14
-#define GT96100_FXTDM0_AUX_CHANB_RX 0x008B18
-#define GT96100_FXTDM1_TDPR1_BLK0_BASE  0x010B00
-#define GT96100_FXTDM1_TDPR1_BLK1_BASE  0x011B00
-#define GT96100_FXTDM1_TDPR1_BLK2_BASE  0x012B00
-#define GT96100_FXTDM1_TDPR1_BLK3_BASE  0x013B00
-#define GT96100_FXTDM1_RDPR1_BLK0_BASE  0x014B00
-#define GT96100_FXTDM1_RDPR1_BLK1_BASE  0x015B00
-#define GT96100_FXTDM1_RDPR1_BLK2_BASE  0x016B00
-#define GT96100_FXTDM1_RDPR1_BLK3_BASE  0x017B00
-#define GT96100_FXTDM1_TX_READ_PTR      0x018B00
-#define GT96100_FXTDM1_RX_READ_PTR      0x018B04
-#define GT96100_FXTDM1_CONFIG       0x018B08
-#define GT96100_FXTDM1_AUX_CHANA_TX 0x018B0C
-#define GT96100_FXTDM1_AUX_CHANA_RX 0x018B10
-#define GT96100_FLTDM1_AUX_CHANB_TX 0x018B14
-#define GT96100_FLTDM1_AUX_CHANB_RX 0x018B18
-#define GT96100_FLTDM2_TDPR2_BLK0_BASE  0x020B00
-#define GT96100_FLTDM2_TDPR2_BLK1_BASE  0x021B00
-#define GT96100_FLTDM2_TDPR2_BLK2_BASE  0x022B00
-#define GT96100_FLTDM2_TDPR2_BLK3_BASE  0x023B00
-#define GT96100_FLTDM2_RDPR2_BLK0_BASE  0x024B00
-#define GT96100_FLTDM2_RDPR2_BLK1_BASE  0x025B00
-#define GT96100_FLTDM2_RDPR2_BLK2_BASE  0x026B00
-#define GT96100_FLTDM2_RDPR2_BLK3_BASE  0x027B00
-#define GT96100_FLTDM2_TX_READ_PTR      0x028B00
-#define GT96100_FLTDM2_RX_READ_PTR      0x028B04
-#define GT96100_FLTDM2_CONFIG       0x028B08
-#define GT96100_FLTDM2_AUX_CHANA_TX 0x028B0C
-#define GT96100_FLTDM2_AUX_CHANA_RX 0x028B10
-#define GT96100_FLTDM2_AUX_CHANB_TX 0x028B14
-#define GT96100_FLTDM2_AUX_CHANB_RX 0x028B18
-#define GT96100_FLTDM3_TDPR3_BLK0_BASE  0x030B00
-#define GT96100_FLTDM3_TDPR3_BLK1_BASE  0x031B00
-#define GT96100_FLTDM3_TDPR3_BLK2_BASE  0x032B00
-#define GT96100_FLTDM3_TDPR3_BLK3_BASE  0x033B00
-#define GT96100_FXTDM3_RDPR3_BLK0_BASE  0x034B00
-#define GT96100_FXTDM3_RDPR3_BLK1_BASE  0x035B00
-#define GT96100_FXTDM3_RDPR3_BLK2_BASE  0x036B00
-#define GT96100_FXTDM3_RDPR3_BLK3_BASE  0x037B00
-#define GT96100_FXTDM3_TX_READ_PTR      0x038B00
-#define GT96100_FXTDM3_RX_READ_PTR      0x038B04
-#define GT96100_FXTDM3_CONFIG       0x038B08
-#define GT96100_FXTDM3_AUX_CHANA_TX 0x038B0C
-#define GT96100_FXTDM3_AUX_CHANA_RX 0x038B10
-#define GT96100_FXTDM3_AUX_CHANB_TX 0x038B14
-#define GT96100_FXTDM3_AUX_CHANB_RX 0x038B18
-/*  Baud Rate Generators  */
-#define GT96100_BRG0_CONFIG     0x102A00
-#define GT96100_BRG0_BAUD_TUNE  0x102A04
-#define GT96100_BRG1_CONFIG     0x102A08
-#define GT96100_BRG1_BAUD_TUNE  0x102A0C
-#define GT96100_BRG2_CONFIG     0x102A10
-#define GT96100_BRG2_BAUD_TUNE  0x102A14
-#define GT96100_BRG3_CONFIG     0x102A18
-#define GT96100_BRG3_BAUD_TUNE  0x102A1C
-#define GT96100_BRG4_CONFIG     0x102A20
-#define GT96100_BRG4_BAUD_TUNE  0x102A24
-#define GT96100_BRG5_CONFIG     0x102A28
-#define GT96100_BRG5_BAUD_TUNE  0x102A2C
-#define GT96100_BRG6_CONFIG     0x102A30
-#define GT96100_BRG6_BAUD_TUNE  0x102A34
-#define GT96100_BRG7_CONFIG     0x102A38
-#define GT96100_BRG7_BAUD_TUNE  0x102A3C
-/*  Routing Registers  */
-#define GT96100_ROUTE_MAIN      0x101A00
-#define GT96100_ROUTE_RX_CLOCK  0x101A10
-#define GT96100_ROUTE_TX_CLOCK  0x101A20
-/*  General Purpose Ports  */
-#define GT96100_GPP_CONFIG0     0x100A00
-#define GT96100_GPP_CONFIG1     0x100A04
-#define GT96100_GPP_CONFIG2     0x100A08
-#define GT96100_GPP_CONFIG3     0x100A0C
-#define GT96100_GPP_IO0         0x100A20
-#define GT96100_GPP_IO1         0x100A24
-#define GT96100_GPP_IO2         0x100A28
-#define GT96100_GPP_IO3         0x100A2C
-#define GT96100_GPP_DATA0       0x100A40
-#define GT96100_GPP_DATA1       0x100A44
-#define GT96100_GPP_DATA2       0x100A48
-#define GT96100_GPP_DATA3       0x100A4C
-#define GT96100_GPP_LEVEL0      0x100A60
-#define GT96100_GPP_LEVEL1      0x100A64
-#define GT96100_GPP_LEVEL2      0x100A68
-#define GT96100_GPP_LEVEL3      0x100A6C
-/*  Watchdog  */
-#define GT96100_WD_CONFIG   0x101A80
-#define GT96100_WD_VALUE    0x101A84
-/* Communication Unit Arbiter  */
-#define GT96100_COMM_UNIT_ARBTR_CONFIG 0x101AC0
-/*  PCI Arbiters  */
-#define GT96100_PCI0_ARBTR_CONFIG 0x101AE0
-#define GT96100_PCI1_ARBTR_CONFIG 0x101AE4
-/* CIU Arbiter */
-#define GT96100_CIU_ARBITER_CONFIG 0x101AC0
-/* Interrupt Controller */
-#define GT96100_MAIN_CAUSE     0x000C18
-#define GT96100_INT0_MAIN_MASK 0x000C1C
-#define GT96100_INT1_MAIN_MASK 0x000C24
-#define GT96100_HIGH_CAUSE     0x000C98
-#define GT96100_INT0_HIGH_MASK 0x000C9C
-#define GT96100_INT1_HIGH_MASK 0x000CA4
-#define GT96100_INT0_SELECT    0x000C70
-#define GT96100_INT1_SELECT    0x000C74
-#define GT96100_SERIAL_CAUSE   0x103A00
-#define GT96100_SERINT0_MASK   0x103A80
-#define GT96100_SERINT1_MASK   0x103A88
-
-#endif /*  _GT96100_H */
diff --git a/include/asm-mips/mach-ev96100/mach-gt64120.h b/include/asm-mips/mach-ev96100/mach-gt64120.h
deleted file mode 100644
index 0ef1e6c25acf..000000000000
--- a/include/asm-mips/mach-ev96100/mach-gt64120.h
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- *  This is a direct copy of the ev96100.h file, with a global
- * search and replace.  The numbers are the same.
- *
- *  The reason I'm duplicating this is so that the 64120/96100
- * defines won't be confusing in the source code.
- */
-#ifndef _ASM_GT64120_EV96100_GT64120_DEP_H
-#define _ASM_GT64120_EV96100_GT64120_DEP_H
-
-/*
- *   GT96100 config space base address
- */
-#define GT64120_BASE	(KSEG1ADDR(0x14000000))
-
-/*
- *   PCI Bus allocation
- *
- *   (Guessing ...)
- */
-#define GT_PCI_MEM_BASE	0x12000000UL
-#define GT_PCI_MEM_SIZE	0x02000000UL
-#define GT_PCI_IO_BASE	0x10000000UL
-#define GT_PCI_IO_SIZE	0x02000000UL
-#define GT_ISA_IO_BASE	PCI_IO_BASE
-
-/*
- *   Duart I/O ports.
- */
-#define EV96100_COM1_BASE_ADDR	(0xBD000000 + 0x20)
-#define EV96100_COM2_BASE_ADDR	(0xBD000000 + 0x00)
-
-
-/*
- *   EV96100 interrupt controller register base.
- */
-#define EV96100_ICTRL_REGS_BASE	(KSEG1ADDR(0x1f000000))
-
-/*
- *   EV96100 UART register base.
- */
-#define EV96100_UART0_REGS_BASE	EV96100_COM1_BASE_ADDR
-#define EV96100_UART1_REGS_BASE	EV96100_COM2_BASE_ADDR
-#define EV96100_BASE_BAUD	( 3686400 / 16 )
-
-#endif /* _ASM_GT64120_EV96100_GT64120_DEP_H */
diff --git a/include/asm-mips/serial.h b/include/asm-mips/serial.h
index 584bd9c0ab2e..035637c67e7c 100644
--- a/include/asm-mips/serial.h
+++ b/include/asm-mips/serial.h
@@ -52,9 +52,9 @@
 #endif
 
 /*
- * Both Galileo boards have the same UART mappings.
+ * Galileo EV64120 evaluation board
  */
-#if defined (CONFIG_MIPS_EV96100) || defined (CONFIG_MIPS_EV64120)
+#ifdef CONFIG_MIPS_EV64120
 #include <asm/galileo-boards/ev96100.h>
 #include <asm/galileo-boards/ev96100int.h>
 #define EV96100_SERIAL_PORT_DEFNS                                  \
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 6a1e09834559..5c1c698a92ac 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1482,9 +1482,6 @@
 #define PCI_DEVICE_ID_MARVELL_GT64260	0x6430
 #define PCI_DEVICE_ID_MARVELL_MV64360	0x6460
 #define PCI_DEVICE_ID_MARVELL_MV64460	0x6480
-#define PCI_DEVICE_ID_MARVELL_GT96100	0x9652
-#define PCI_DEVICE_ID_MARVELL_GT96100A	0x9653
-
 
 #define PCI_VENDOR_ID_V3		0x11b0
 #define PCI_DEVICE_ID_V3_V960		0x0001
-- 
cgit v1.2.3


From ddb1199c4cd74a02d70f4c297373400893d70aa9 Mon Sep 17 00:00:00 2001
From: Richard Sandiford <richard@codesourcery.com>
Date: Sun, 17 Sep 2006 20:38:39 +0100
Subject: [MIPS] fstatat syscall names

MIPS is the only port to call its fstatat()-related syscalls
"__NR_fstatat".  Now I can see why that might be seen as every
other port being wrong, but I think for o32, it is at best confusing.
__NR_fstat provides a plain (32-bit) stat while __NR_fstatat provides a
64-bit stat.  Changing the name to __NR_fstatat64 would make things more
explicit, match x86, and make the glibc port slightly easier.

The current name is more appropriate for n32 and n64, but it would be
appropriate for other 64-bit targets too, and those targets have chosen
to call it __NR_newfstatat instead.  Using the same name for MIPS would
again be more consistent and make the glibc port slightly easier.

I'm not wedded to this idea if the current names are preferred,
but FWIW...

Signed-off-by: Richard Sandiford <richard@codesourcery.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/unistd.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h
index 558e3cba4f92..c39142920fe6 100644
--- a/include/asm-mips/unistd.h
+++ b/include/asm-mips/unistd.h
@@ -313,7 +313,7 @@
 #define __NR_mknodat			(__NR_Linux + 290)
 #define __NR_fchownat			(__NR_Linux + 291)
 #define __NR_futimesat			(__NR_Linux + 292)
-#define __NR_fstatat			(__NR_Linux + 293)
+#define __NR_fstatat64			(__NR_Linux + 293)
 #define __NR_unlinkat			(__NR_Linux + 294)
 #define __NR_renameat			(__NR_Linux + 295)
 #define __NR_linkat			(__NR_Linux + 296)
@@ -600,7 +600,7 @@
 #define __NR_mknodat			(__NR_Linux + 249)
 #define __NR_fchownat			(__NR_Linux + 250)
 #define __NR_futimesat			(__NR_Linux + 251)
-#define __NR_fstatat			(__NR_Linux + 252)
+#define __NR_newfstatat			(__NR_Linux + 252)
 #define __NR_unlinkat			(__NR_Linux + 253)
 #define __NR_renameat			(__NR_Linux + 254)
 #define __NR_linkat			(__NR_Linux + 255)
@@ -891,7 +891,7 @@
 #define __NR_mknodat			(__NR_Linux + 253)
 #define __NR_fchownat			(__NR_Linux + 254)
 #define __NR_futimesat			(__NR_Linux + 255)
-#define __NR_fstatat			(__NR_Linux + 256)
+#define __NR_newfstatat			(__NR_Linux + 256)
 #define __NR_unlinkat			(__NR_Linux + 257)
 #define __NR_renameat			(__NR_Linux + 258)
 #define __NR_linkat			(__NR_Linux + 259)
-- 
cgit v1.2.3


From e584ade1a6679bfb417620f15951796bed9805d9 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 23 Sep 2006 18:08:36 +0100
Subject: [MIPS] Have headers_install install <asm/cachectl.h> and
 <asm/sysmips.h>.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/Kbuild | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-mips/Kbuild b/include/asm-mips/Kbuild
index 01ea3e045652..7897f05e3165 100644
--- a/include/asm-mips/Kbuild
+++ b/include/asm-mips/Kbuild
@@ -1,3 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
-header-y += sgidefs.h
+header-y += cachectl.h sgidefs.h sysmips.h
-- 
cgit v1.2.3


From 36396f3c36b04f79438f87a0fccfa76aa3de6af9 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 25 Sep 2006 15:49:49 +0100
Subject: [MIPS] s/__ASSEMBLER__/__ASSEMBLY__/ for clarity sake.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/hazards.h            | 2 +-
 include/asm-mips/mach-excite/excite.h | 2 +-
 include/asm-mips/sibyte/sb1250_defs.h | 6 +++---
 include/asm-mips/sibyte/sb1250_scd.h  | 2 +-
 4 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-mips/hazards.h b/include/asm-mips/hazards.h
index bf0b80404946..0fe02945feba 100644
--- a/include/asm-mips/hazards.h
+++ b/include/asm-mips/hazards.h
@@ -11,7 +11,7 @@
 #define _ASM_HAZARDS_H
 
 
-#ifdef __ASSEMBLER__
+#ifdef __ASSEMBLY__
 #define ASMMACRO(name, code...) .macro name; code; .endm
 #else
 
diff --git a/include/asm-mips/mach-excite/excite.h b/include/asm-mips/mach-excite/excite.h
index 130bd4b8edce..4c29ba44992c 100644
--- a/include/asm-mips/mach-excite/excite.h
+++ b/include/asm-mips/mach-excite/excite.h
@@ -7,7 +7,7 @@
 
 #define EXCITE_CPU_EXT_CLOCK 100000000
 
-#if !defined(__ASSEMBLER__)
+#if !defined(__ASSEMBLY__)
 void __init excite_kgdb_init(void);
 void excite_procfs_init(void);
 extern unsigned long memsize;
diff --git a/include/asm-mips/sibyte/sb1250_defs.h b/include/asm-mips/sibyte/sb1250_defs.h
index 335dbaf1d831..a885491217c1 100644
--- a/include/asm-mips/sibyte/sb1250_defs.h
+++ b/include/asm-mips/sibyte/sb1250_defs.h
@@ -212,7 +212,7 @@
  * Note: you'll need to define uint32_t and uint64_t in your headers.
  */
 
-#if !defined(__ASSEMBLER__)
+#if !defined(__ASSEMBLY__)
 #define _SB_MAKE64(x) ((uint64_t)(x))
 #define _SB_MAKE32(x) ((uint32_t)(x))
 #else
@@ -251,9 +251,9 @@
  */
 
 
-#if defined(__mips64) && !defined(__ASSEMBLER__)
+#if defined(__mips64) && !defined(__ASSEMBLY__)
 #define SBWRITECSR(csr,val) *((volatile uint64_t *) PHYS_TO_K1(csr)) = (val)
 #define SBREADCSR(csr) (*((volatile uint64_t *) PHYS_TO_K1(csr)))
-#endif /* __ASSEMBLER__ */
+#endif /* __ASSEMBLY__ */
 
 #endif
diff --git a/include/asm-mips/sibyte/sb1250_scd.h b/include/asm-mips/sibyte/sb1250_scd.h
index f4178bdcfcb0..7ed0bb611e56 100644
--- a/include/asm-mips/sibyte/sb1250_scd.h
+++ b/include/asm-mips/sibyte/sb1250_scd.h
@@ -149,7 +149,7 @@
  * (For the assembler version, sysrev and dest may be the same register.
  * Also, it clobbers AT.)
  */
-#ifdef __ASSEMBLER__
+#ifdef __ASSEMBLY__
 #define SYS_SOC_TYPE(dest, sysrev)					\
 	.set push ;							\
 	.set reorder ;							\
-- 
cgit v1.2.3


From bbf2bef9f50eb119ffadd735eb0966ac8a04f91f Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 27 Sep 2006 01:49:25 -0700
Subject: [PATCH] fix "cpu to node relationship fixup: map cpu to node"

Fix build error introduced by 3212fe1594e577463bc8601d28aa008f520c3377

Non-NUMA case should be handled.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/topology.c | 2 --
 include/asm-ia64/numa.h     | 2 ++
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index 05bdf7affb43..5629b45e89c6 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -36,9 +36,7 @@ int arch_register_cpu(int num)
 	 */
 	if (!can_cpei_retarget() && is_cpu_cpei_target(num))
 		sysfs_cpus[num].cpu.no_control = 1;
-#ifdef CONFIG_NUMA
 	map_cpu_to_node(num, node_cpuid[num].nid);
-#endif
 #endif
 
 	return register_cpu(&sysfs_cpus[num].cpu, num);
diff --git a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h
index e0a1d173e42d..7d5e2ccc37a0 100644
--- a/include/asm-ia64/numa.h
+++ b/include/asm-ia64/numa.h
@@ -69,6 +69,8 @@ extern void unmap_cpu_from_node(int cpu, int nid);
 
 
 #else /* !CONFIG_NUMA */
+#define map_cpu_to_node(cpu, nid)	do{}while(0)
+#define unmap_cpu_from_node(cpu, nid)	do{}while(0)
 
 #define paddr_to_nid(addr)	0
 
-- 
cgit v1.2.3


From ae6ddcc5f24d6b06ae9231dc128904750a4155e0 Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Wed, 27 Sep 2006 01:49:27 -0700
Subject: [PATCH] ext3 and jbd cleanup: remove whitespace

Remove whitespace from ext3 and jbd, before we clone ext4.

Signed-off-by: Mingming Cao<cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/balloc.c         |  16 +++---
 fs/ext3/bitmap.c         |   2 +-
 fs/ext3/dir.c            |  14 +++---
 fs/ext3/file.c           |   2 +-
 fs/ext3/fsync.c          |   6 +--
 fs/ext3/hash.c           |   6 +--
 fs/ext3/ialloc.c         |  48 +++++++++---------
 fs/ext3/inode.c          |  64 ++++++++++++------------
 fs/ext3/namei.c          |  28 +++++------
 fs/ext3/super.c          |  24 ++++-----
 fs/jbd/checkpoint.c      |  30 +++++------
 fs/jbd/journal.c         |  56 ++++++++++-----------
 fs/jbd/recovery.c        |  54 ++++++++++----------
 fs/jbd/revoke.c          |  70 +++++++++++++-------------
 fs/jbd/transaction.c     | 128 +++++++++++++++++++++++------------------------
 include/linux/ext3_jbd.h |  10 ++--
 include/linux/jbd.h      |  56 ++++++++++-----------
 17 files changed, 307 insertions(+), 307 deletions(-)

(limited to 'include')

diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index 063d994bda0b..e6b983707008 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -74,7 +74,7 @@ struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb,
 }
 
 /*
- * Read the bitmap for a given block_group, reading into the specified 
+ * Read the bitmap for a given block_group, reading into the specified
  * slot in the superblock's bitmap cache.
  *
  * Return buffer_head on success or NULL in case of failure.
@@ -419,8 +419,8 @@ do_more:
 		}
 		/* @@@ This prevents newly-allocated data from being
 		 * freed and then reallocated within the same
-		 * transaction. 
-		 * 
+		 * transaction.
+		 *
 		 * Ideally we would want to allow that to happen, but to
 		 * do so requires making journal_forget() capable of
 		 * revoking the queued write of a data block, which
@@ -433,7 +433,7 @@ do_more:
 		 * safe not to set the allocation bit in the committed
 		 * bitmap, because we know that there is no outstanding
 		 * activity on the buffer any more and so it is safe to
-		 * reallocate it.  
+		 * reallocate it.
 		 */
 		BUFFER_TRACE(bitmap_bh, "set in b_committed_data");
 		J_ASSERT_BH(bitmap_bh,
@@ -518,7 +518,7 @@ void ext3_free_blocks(handle_t *handle, struct inode *inode,
  * data would allow the old block to be overwritten before the
  * transaction committed (because we force data to disk before commit).
  * This would lead to corruption if we crashed between overwriting the
- * data and committing the delete. 
+ * data and committing the delete.
  *
  * @@@ We may want to make this allocation behaviour conditional on
  * data-writes at some point, and disable it for metadata allocations or
@@ -584,7 +584,7 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 
 	if (start > 0) {
 		/*
-		 * The goal was occupied; search forward for a free 
+		 * The goal was occupied; search forward for a free
 		 * block within the next XX blocks.
 		 *
 		 * end_goal is more or less random, but it has to be
@@ -1194,7 +1194,7 @@ int ext3_should_retry_alloc(struct super_block *sb, int *retries)
 /*
  * ext3_new_block uses a goal block to assist allocation.  If the goal is
  * free, or there is a free block within 32 blocks of the goal, that block
- * is allocated.  Otherwise a forward search is made for a free block; within 
+ * is allocated.  Otherwise a forward search is made for a free block; within
  * each block group the search first looks for an entire free byte in the block
  * bitmap, and then for any free bit if that fails.
  * This function also updates quota and i_blocks field.
@@ -1303,7 +1303,7 @@ retry_alloc:
 	smp_rmb();
 
 	/*
-	 * Now search the rest of the groups.  We assume that 
+	 * Now search the rest of the groups.  We assume that
 	 * i and gdp correctly point to the last group visited.
 	 */
 	for (bgi = 0; bgi < ngroups; bgi++) {
diff --git a/fs/ext3/bitmap.c b/fs/ext3/bitmap.c
index ce4f82b9e528..b9176eed98d1 100644
--- a/fs/ext3/bitmap.c
+++ b/fs/ext3/bitmap.c
@@ -20,7 +20,7 @@ unsigned long ext3_count_free (struct buffer_head * map, unsigned int numchars)
 	unsigned int i;
 	unsigned long sum = 0;
 
-	if (!map) 
+	if (!map)
 		return (0);
 	for (i = 0; i < numchars; i++)
 		sum += nibblemap[map->b_data[i] & 0xf] +
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index fbb0d4ed07d4..6f9e5a523c87 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -59,7 +59,7 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 
 	return (ext3_filetype_table[filetype]);
 }
-			       
+
 
 int ext3_check_dir_entry (const char * function, struct inode * dir,
 			  struct ext3_dir_entry_2 * de,
@@ -162,7 +162,7 @@ revalidate:
 		 * to make sure. */
 		if (filp->f_version != inode->i_version) {
 			for (i = 0; i < sb->s_blocksize && i < offset; ) {
-				de = (struct ext3_dir_entry_2 *) 
+				de = (struct ext3_dir_entry_2 *)
 					(bh->b_data + i);
 				/* It's too expensive to do a full
 				 * dirent test each time round this
@@ -181,7 +181,7 @@ revalidate:
 			filp->f_version = inode->i_version;
 		}
 
-		while (!error && filp->f_pos < inode->i_size 
+		while (!error && filp->f_pos < inode->i_size
 		       && offset < sb->s_blocksize) {
 			de = (struct ext3_dir_entry_2 *) (bh->b_data + offset);
 			if (!ext3_check_dir_entry ("ext3_readdir", inode, de,
@@ -229,7 +229,7 @@ out:
 /*
  * These functions convert from the major/minor hash to an f_pos
  * value.
- * 
+ *
  * Currently we only use major hash numer.  This is unfortunate, but
  * on 32-bit machines, the same VFS interface is used for lseek and
  * llseek, so if we use the 64 bit offset, then the 32-bit versions of
@@ -250,7 +250,7 @@ out:
 struct fname {
 	__u32		hash;
 	__u32		minor_hash;
-	struct rb_node	rb_hash; 
+	struct rb_node	rb_hash;
 	struct fname	*next;
 	__u32		inode;
 	__u8		name_len;
@@ -410,7 +410,7 @@ static int call_filldir(struct file * filp, void * dirent,
 	curr_pos = hash2pos(fname->hash, fname->minor_hash);
 	while (fname) {
 		error = filldir(dirent, fname->name,
-				fname->name_len, curr_pos, 
+				fname->name_len, curr_pos,
 				fname->inode,
 				get_dtype(sb, fname->file_type));
 		if (error) {
@@ -465,7 +465,7 @@ static int ext3_dx_readdir(struct file * filp,
 		/*
 		 * Fill the rbtree if we have no more entries,
 		 * or the inode has changed since we last read in the
-		 * cached entries. 
+		 * cached entries.
 		 */
 		if ((!info->curr_node) ||
 		    (filp->f_version != inode->i_version)) {
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 1efefb630ea9..994efd189f4e 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -100,7 +100,7 @@ ext3_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t
 
 force_commit:
 	err = ext3_force_commit(inode->i_sb);
-	if (err) 
+	if (err)
 		return err;
 	return ret;
 }
diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c
index 49382a208e05..dd1fd3c0fc05 100644
--- a/fs/ext3/fsync.c
+++ b/fs/ext3/fsync.c
@@ -8,14 +8,14 @@
  *                      Universite Pierre et Marie Curie (Paris VI)
  *  from
  *  linux/fs/minix/truncate.c   Copyright (C) 1991, 1992  Linus Torvalds
- * 
+ *
  *  ext3fs fsync primitive
  *
  *  Big-endian to little-endian byte-swapping/bitmaps by
  *        David S. Miller (davem@caip.rutgers.edu), 1995
- * 
+ *
  *  Removed unnecessary code duplication for little endian machines
- *  and excessive __inline__s. 
+ *  and excessive __inline__s.
  *        Andi Kleen, 1997
  *
  * Major simplications and cleanup - we only need to do the metadata, because
diff --git a/fs/ext3/hash.c b/fs/ext3/hash.c
index 5a2d1235ead0..7fa637cd322a 100644
--- a/fs/ext3/hash.c
+++ b/fs/ext3/hash.c
@@ -4,7 +4,7 @@
  * Copyright (C) 2002 by Theodore Ts'o
  *
  * This file is released under the GPL v2.
- * 
+ *
  * This file may be redistributed under the terms of the GNU Public
  * License.
  */
@@ -80,11 +80,11 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num)
  * Returns the hash of a filename.  If len is 0 and name is NULL, then
  * this function can be used to test whether or not a hash version is
  * supported.
- * 
+ *
  * The seed is an 4 longword (32 bits) "secret" which can be used to
  * uniquify a hash.  If the seed is all zero's, then some default seed
  * may be used.
- * 
+ *
  * A particular hash version specifies whether or not the seed is
  * represented, and whether or not the returned hash is 32 bits or 64
  * bits.  32 bit hashes will return 0 for the minor hash.
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 36546ed36a14..5e288368499b 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -216,7 +216,7 @@ static int find_group_dir(struct super_block *sb, struct inode *parent)
 			continue;
 		if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
 			continue;
-		if (!best_desc || 
+		if (!best_desc ||
 		    (le16_to_cpu(desc->bg_free_blocks_count) >
 		     le16_to_cpu(best_desc->bg_free_blocks_count))) {
 			best_group = group;
@@ -226,30 +226,30 @@ static int find_group_dir(struct super_block *sb, struct inode *parent)
 	return best_group;
 }
 
-/* 
- * Orlov's allocator for directories. 
- * 
+/*
+ * Orlov's allocator for directories.
+ *
  * We always try to spread first-level directories.
  *
- * If there are blockgroups with both free inodes and free blocks counts 
- * not worse than average we return one with smallest directory count. 
- * Otherwise we simply return a random group. 
- * 
- * For the rest rules look so: 
- * 
- * It's OK to put directory into a group unless 
- * it has too many directories already (max_dirs) or 
- * it has too few free inodes left (min_inodes) or 
- * it has too few free blocks left (min_blocks) or 
- * it's already running too large debt (max_debt). 
- * Parent's group is prefered, if it doesn't satisfy these 
- * conditions we search cyclically through the rest. If none 
- * of the groups look good we just look for a group with more 
- * free inodes than average (starting at parent's group). 
- * 
- * Debt is incremented each time we allocate a directory and decremented 
- * when we allocate an inode, within 0--255. 
- */ 
+ * If there are blockgroups with both free inodes and free blocks counts
+ * not worse than average we return one with smallest directory count.
+ * Otherwise we simply return a random group.
+ *
+ * For the rest rules look so:
+ *
+ * It's OK to put directory into a group unless
+ * it has too many directories already (max_dirs) or
+ * it has too few free inodes left (min_inodes) or
+ * it has too few free blocks left (min_blocks) or
+ * it's already running too large debt (max_debt).
+ * Parent's group is prefered, if it doesn't satisfy these
+ * conditions we search cyclically through the rest. If none
+ * of the groups look good we just look for a group with more
+ * free inodes than average (starting at parent's group).
+ *
+ * Debt is incremented each time we allocate a directory and decremented
+ * when we allocate an inode, within 0--255.
+ */
 
 #define INODE_COST 64
 #define BLOCK_COST 256
@@ -454,7 +454,7 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode)
 			group = find_group_dir(sb, dir);
 		else
 			group = find_group_orlov(sb, dir);
-	} else 
+	} else
 		group = find_group_other(sb, dir);
 
 	err = -ENOSPC;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 84be02e93652..473d206b1d7e 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -55,7 +55,7 @@ static int ext3_inode_is_fast_symlink(struct inode *inode)
 /*
  * The ext3 forget function must perform a revoke if we are freeing data
  * which has been journaled.  Metadata (eg. indirect blocks) must be
- * revoked in all cases. 
+ * revoked in all cases.
  *
  * "bh" may be NULL: a metadata block may have been freed from memory
  * but there may still be a record of it in the journal, and that record
@@ -105,7 +105,7 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode,
  * Work out how many blocks we need to proceed with the next chunk of a
  * truncate transaction.
  */
-static unsigned long blocks_for_truncate(struct inode *inode) 
+static unsigned long blocks_for_truncate(struct inode *inode)
 {
 	unsigned long needed;
 
@@ -122,13 +122,13 @@ static unsigned long blocks_for_truncate(struct inode *inode)
 
 	/* But we need to bound the transaction so we don't overflow the
 	 * journal. */
-	if (needed > EXT3_MAX_TRANS_DATA) 
+	if (needed > EXT3_MAX_TRANS_DATA)
 		needed = EXT3_MAX_TRANS_DATA;
 
 	return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed;
 }
 
-/* 
+/*
  * Truncate transactions can be complex and absolutely huge.  So we need to
  * be able to restart the transaction at a conventient checkpoint to make
  * sure we don't overflow the journal.
@@ -136,9 +136,9 @@ static unsigned long blocks_for_truncate(struct inode *inode)
  * start_transaction gets us a new handle for a truncate transaction,
  * and extend_transaction tries to extend the existing one a bit.  If
  * extend fails, we need to propagate the failure up and restart the
- * transaction in the top-level truncate loop. --sct 
+ * transaction in the top-level truncate loop. --sct
  */
-static handle_t *start_transaction(struct inode *inode) 
+static handle_t *start_transaction(struct inode *inode)
 {
 	handle_t *result;
 
@@ -215,12 +215,12 @@ void ext3_delete_inode (struct inode * inode)
 	ext3_orphan_del(handle, inode);
 	EXT3_I(inode)->i_dtime	= get_seconds();
 
-	/* 
+	/*
 	 * One subtle ordering requirement: if anything has gone wrong
 	 * (transaction abort, IO errors, whatever), then we can still
 	 * do these next steps (the fs will already have been marked as
 	 * having errors), but we can't free the inode if the mark_dirty
-	 * fails.  
+	 * fails.
 	 */
 	if (ext3_mark_inode_dirty(handle, inode))
 		/* If that failed, just do the required in-core inode clear. */
@@ -398,7 +398,7 @@ no_block:
  *	  + if there is a block to the left of our position - allocate near it.
  *	  + if pointer will live in indirect block - allocate near that block.
  *	  + if pointer will live in inode - allocate in the same
- *	    cylinder group. 
+ *	    cylinder group.
  *
  * In the latter case we colour the starting block by the callers PID to
  * prevent it from clashing with concurrent allocations for a different inode
@@ -744,7 +744,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode,
 		jbd_debug(5, "splicing indirect only\n");
 		BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata");
 		err = ext3_journal_dirty_metadata(handle, where->bh);
-		if (err) 
+		if (err)
 			goto err_out;
 	} else {
 		/*
@@ -1137,7 +1137,7 @@ static int walk_page_buffers(	handle_t *handle,
  * So what we do is to rely on the fact that journal_stop/journal_start
  * will _not_ run commit under these circumstances because handle->h_ref
  * is elevated.  We'll still have enough credits for the tiny quotafile
- * write.  
+ * write.
  */
 static int do_journal_get_write_access(handle_t *handle,
 					struct buffer_head *bh)
@@ -1282,7 +1282,7 @@ static int ext3_journalled_commit_write(struct file *file,
 	if (inode->i_size > EXT3_I(inode)->i_disksize) {
 		EXT3_I(inode)->i_disksize = inode->i_size;
 		ret2 = ext3_mark_inode_dirty(handle, inode);
-		if (!ret) 
+		if (!ret)
 			ret = ret2;
 	}
 	ret2 = ext3_journal_stop(handle);
@@ -1291,7 +1291,7 @@ static int ext3_journalled_commit_write(struct file *file,
 	return ret;
 }
 
-/* 
+/*
  * bmap() is special.  It gets used by applications such as lilo and by
  * the swapper to find the on-disk block of a specific piece of data.
  *
@@ -1300,10 +1300,10 @@ static int ext3_journalled_commit_write(struct file *file,
  * filesystem and enables swap, then they may get a nasty shock when the
  * data getting swapped to that swapfile suddenly gets overwritten by
  * the original zero's written out previously to the journal and
- * awaiting writeback in the kernel's buffer cache. 
+ * awaiting writeback in the kernel's buffer cache.
  *
  * So, if we see any bmap calls here on a modified, data-journaled file,
- * take extra steps to flush any blocks which might be in the cache. 
+ * take extra steps to flush any blocks which might be in the cache.
  */
 static sector_t ext3_bmap(struct address_space *mapping, sector_t block)
 {
@@ -1312,16 +1312,16 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block)
 	int err;
 
 	if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) {
-		/* 
+		/*
 		 * This is a REALLY heavyweight approach, but the use of
 		 * bmap on dirty files is expected to be extremely rare:
 		 * only if we run lilo or swapon on a freshly made file
-		 * do we expect this to happen. 
+		 * do we expect this to happen.
 		 *
 		 * (bmap requires CAP_SYS_RAWIO so this does not
 		 * represent an unprivileged user DOS attack --- we'd be
 		 * in trouble if mortal users could trigger this path at
-		 * will.) 
+		 * will.)
 		 *
 		 * NB. EXT3_STATE_JDATA is not set on files other than
 		 * regular files.  If somebody wants to bmap a directory
@@ -1457,7 +1457,7 @@ static int ext3_ordered_writepage(struct page *page,
 	 */
 
 	/*
-	 * And attach them to the current transaction.  But only if 
+	 * And attach them to the current transaction.  But only if
 	 * block_write_full_page() succeeded.  Otherwise they are unmapped,
 	 * and generally junk.
 	 */
@@ -1644,7 +1644,7 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb,
 		}
 	}
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, 
+	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
 				 offset, nr_segs,
 				 ext3_get_block, NULL);
 
@@ -2025,7 +2025,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode,
 			   __le32 *first, __le32 *last)
 {
 	ext3_fsblk_t block_to_free = 0;    /* Starting block # of a run */
-	unsigned long count = 0;	    /* Number of blocks in the run */ 
+	unsigned long count = 0;	    /* Number of blocks in the run */
 	__le32 *block_to_free_p = NULL;	    /* Pointer into inode/ind
 					       corresponding to
 					       block_to_free */
@@ -2054,7 +2054,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode,
 			} else if (nr == block_to_free + count) {
 				count++;
 			} else {
-				ext3_clear_blocks(handle, inode, this_bh, 
+				ext3_clear_blocks(handle, inode, this_bh,
 						  block_to_free,
 						  count, block_to_free_p, p);
 				block_to_free = nr;
@@ -2184,7 +2184,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode,
 					*p = 0;
 					BUFFER_TRACE(parent_bh,
 					"call ext3_journal_dirty_metadata");
-					ext3_journal_dirty_metadata(handle, 
+					ext3_journal_dirty_metadata(handle,
 								    parent_bh);
 				}
 			}
@@ -2704,7 +2704,7 @@ void ext3_read_inode(struct inode * inode)
 		if (raw_inode->i_block[0])
 			init_special_inode(inode, inode->i_mode,
 			   old_decode_dev(le32_to_cpu(raw_inode->i_block[0])));
-		else 
+		else
 			init_special_inode(inode, inode->i_mode,
 			   new_decode_dev(le32_to_cpu(raw_inode->i_block[1])));
 	}
@@ -2724,8 +2724,8 @@ bad_inode:
  *
  * The caller must have write access to iloc->bh.
  */
-static int ext3_do_update_inode(handle_t *handle, 
-				struct inode *inode, 
+static int ext3_do_update_inode(handle_t *handle,
+				struct inode *inode,
 				struct ext3_iloc *iloc)
 {
 	struct ext3_inode *raw_inode = ext3_raw_inode(iloc);
@@ -2900,7 +2900,7 @@ int ext3_write_inode(struct inode *inode, int wait)
  * commit will leave the blocks being flushed in an unused state on
  * disk.  (On recovery, the inode will get truncated and the blocks will
  * be freed, so we have a strong guarantee that no future commit will
- * leave these blocks visible to the user.)  
+ * leave these blocks visible to the user.)
  *
  * Called with inode->sem down.
  */
@@ -3043,13 +3043,13 @@ int ext3_mark_iloc_dirty(handle_t *handle,
 	return err;
 }
 
-/* 
+/*
  * On success, We end up with an outstanding reference count against
- * iloc->bh.  This _must_ be cleaned up later. 
+ * iloc->bh.  This _must_ be cleaned up later.
  */
 
 int
-ext3_reserve_inode_write(handle_t *handle, struct inode *inode, 
+ext3_reserve_inode_write(handle_t *handle, struct inode *inode,
 			 struct ext3_iloc *iloc)
 {
 	int err = 0;
@@ -3139,7 +3139,7 @@ out:
 }
 
 #if 0
-/* 
+/*
  * Bind an inode's backing buffer_head into this transaction, to prevent
  * it from being flushed to disk early.  Unlike
  * ext3_reserve_inode_write, this leaves behind no bh reference and
@@ -3157,7 +3157,7 @@ static int ext3_pin_inode(handle_t *handle, struct inode *inode)
 			BUFFER_TRACE(iloc.bh, "get_write_access");
 			err = journal_get_write_access(handle, iloc.bh);
 			if (!err)
-				err = ext3_journal_dirty_metadata(handle, 
+				err = ext3_journal_dirty_metadata(handle,
 								  iloc.bh);
 			brelse(iloc.bh);
 		}
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 2aa7101b27cd..4123f5261bcd 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -76,7 +76,7 @@ static struct buffer_head *ext3_append(handle_t *handle,
 #ifdef DX_DEBUG
 #define dxtrace(command) command
 #else
-#define dxtrace(command) 
+#define dxtrace(command)
 #endif
 
 struct fake_dirent
@@ -169,7 +169,7 @@ static struct ext3_dir_entry_2* dx_pack_dirents (char *base, int size);
 static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block);
 static int ext3_htree_next_block(struct inode *dir, __u32 hash,
 				 struct dx_frame *frame,
-				 struct dx_frame *frames, 
+				 struct dx_frame *frames,
 				 __u32 *start_hash);
 static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry,
 		       struct ext3_dir_entry_2 **res_dir, int *err);
@@ -250,7 +250,7 @@ static void dx_show_index (char * label, struct dx_entry *entries)
 }
 
 struct stats
-{ 
+{
 	unsigned names;
 	unsigned space;
 	unsigned bcount;
@@ -464,7 +464,7 @@ static void dx_release (struct dx_frame *frames)
  */
 static int ext3_htree_next_block(struct inode *dir, __u32 hash,
 				 struct dx_frame *frame,
-				 struct dx_frame *frames, 
+				 struct dx_frame *frames,
 				 __u32 *start_hash)
 {
 	struct dx_frame *p;
@@ -632,7 +632,7 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 		}
 		count += ret;
 		hashval = ~0;
-		ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS, 
+		ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS,
 					    frame, frames, &hashval);
 		*next_hash = hashval;
 		if (ret < 0) {
@@ -649,7 +649,7 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 			break;
 	}
 	dx_release(frames);
-	dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", 
+	dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n",
 		       count, *next_hash));
 	return count;
 errout:
@@ -1050,7 +1050,7 @@ struct dentry *ext3_get_parent(struct dentry *child)
 		parent = ERR_PTR(-ENOMEM);
 	}
 	return parent;
-} 
+}
 
 #define S_SHIFT 12
 static unsigned char ext3_type_by_mode[S_IFMT >> S_SHIFT] = {
@@ -1198,7 +1198,7 @@ errout:
  * add_dirent_to_buf will attempt search the directory block for
  * space.  It will return -ENOSPC if no space is available, and -EIO
  * and -EEXIST if directory entry already exists.
- * 
+ *
  * NOTE!  bh is NOT released in the case where ENOSPC is returned.  In
  * all other cases bh is released.
  */
@@ -1572,7 +1572,7 @@ cleanup:
  * ext3_delete_entry deletes a directory entry by merging it with the
  * previous entry
  */
-static int ext3_delete_entry (handle_t *handle, 
+static int ext3_delete_entry (handle_t *handle,
 			      struct inode * dir,
 			      struct ext3_dir_entry_2 * de_del,
 			      struct buffer_head * bh)
@@ -1643,12 +1643,12 @@ static int ext3_add_nondir(handle_t *handle,
  * is so far negative - it has no inode.
  *
  * If the create succeeds, we fill in the inode information
- * with d_instantiate(). 
+ * with d_instantiate().
  */
 static int ext3_create (struct inode * dir, struct dentry * dentry, int mode,
 		struct nameidata *nd)
 {
-	handle_t *handle; 
+	handle_t *handle;
 	struct inode * inode;
 	int err, retries = 0;
 
@@ -1813,7 +1813,7 @@ static int empty_dir (struct inode * inode)
 	de1 = (struct ext3_dir_entry_2 *)
 			((char *) de + le16_to_cpu(de->rec_len));
 	if (le32_to_cpu(de->inode) != inode->i_ino ||
-			!le32_to_cpu(de1->inode) || 
+			!le32_to_cpu(de1->inode) ||
 			strcmp (".", de->name) ||
 			strcmp ("..", de1->name)) {
 	    	ext3_warning (inode->i_sb, "empty_dir",
@@ -1883,7 +1883,7 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode)
 	 * being truncated, or files being unlinked. */
 
 	/* @@@ FIXME: Observation from aviro:
-	 * I think I can trigger J_ASSERT in ext3_orphan_add().  We block 
+	 * I think I can trigger J_ASSERT in ext3_orphan_add().  We block
 	 * here (on lock_super()), so race with ext3_link() which might bump
 	 * ->i_nlink. For, say it, character device. Not a regular file,
 	 * not a directory, not a symlink and ->i_nlink > 0.
@@ -2393,4 +2393,4 @@ struct inode_operations ext3_special_inode_operations = {
 	.removexattr	= generic_removexattr,
 #endif
 	.permission	= ext3_permission,
-}; 
+};
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 3559086eee5f..4b95bfe4c8f7 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -62,13 +62,13 @@ static void ext3_unlockfs(struct super_block *sb);
 static void ext3_write_super (struct super_block * sb);
 static void ext3_write_super_lockfs(struct super_block *sb);
 
-/* 
+/*
  * Wrappers for journal_start/end.
  *
  * The only special thing we need to do here is to make sure that all
  * journal_end calls result in the superblock being marked dirty, so
  * that sync() will call the filesystem's write_super callback if
- * appropriate. 
+ * appropriate.
  */
 handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks)
 {
@@ -90,11 +90,11 @@ handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks)
 	return journal_start(journal, nblocks);
 }
 
-/* 
+/*
  * The only special thing we need to do here is to make sure that all
  * journal_stop calls result in the superblock being marked dirty, so
  * that sync() will call the filesystem's write_super callback if
- * appropriate. 
+ * appropriate.
  */
 int __ext3_journal_stop(const char *where, handle_t *handle)
 {
@@ -369,7 +369,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 {
 	struct list_head *l;
 
-	printk(KERN_ERR "sb orphan head is %d\n", 
+	printk(KERN_ERR "sb orphan head is %d\n",
 	       le32_to_cpu(sbi->s_es->s_last_orphan));
 
 	printk(KERN_ERR "sb_info orphan list:\n");
@@ -378,7 +378,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 		printk(KERN_ERR "  "
 		       "inode %s:%ld at %p: mode %o, nlink %d, next %d\n",
 		       inode->i_sb->s_id, inode->i_ino, inode,
-		       inode->i_mode, inode->i_nlink, 
+		       inode->i_mode, inode->i_nlink,
 		       NEXT_ORPHAN(inode));
 	}
 }
@@ -475,7 +475,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 		inode_init_once(&ei->vfs_inode);
 	}
 }
- 
+
 static int init_inodecache(void)
 {
 	ext3_inode_cachep = kmem_cache_create("ext3_inode_cache",
@@ -1483,7 +1483,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	    (EXT3_HAS_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT3_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT3_HAS_INCOMPAT_FEATURE(sb, ~0U)))
-		printk(KERN_WARNING 
+		printk(KERN_WARNING
 		       "EXT3-fs warning: feature flags set on rev 0 fs, "
 		       "running e2fsck is recommended\n");
 	/*
@@ -1509,7 +1509,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 	if (blocksize < EXT3_MIN_BLOCK_SIZE ||
 	    blocksize > EXT3_MAX_BLOCK_SIZE) {
-		printk(KERN_ERR 
+		printk(KERN_ERR
 		       "EXT3-fs: Unsupported filesystem blocksize %d on %s.\n",
 		       blocksize, sb->s_id);
 		goto failed_mount;
@@ -1533,14 +1533,14 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize;
 		bh = sb_bread(sb, logic_sb_block);
 		if (!bh) {
-			printk(KERN_ERR 
+			printk(KERN_ERR
 			       "EXT3-fs: Can't read superblock on 2nd try.\n");
 			goto failed_mount;
 		}
 		es = (struct ext3_super_block *)(((char *)bh->b_data) + offset);
 		sbi->s_es = es;
 		if (es->s_magic != cpu_to_le16(EXT3_SUPER_MAGIC)) {
-			printk (KERN_ERR 
+			printk (KERN_ERR
 				"EXT3-fs: Magic mismatch, very weird !\n");
 			goto failed_mount;
 		}
@@ -1820,7 +1820,7 @@ out_fail:
 /*
  * Setup any per-fs journal parameters now.  We'll do this both on
  * initial mount, once the journal has been initialised but before we've
- * done any recovery; and again on any subsequent remount. 
+ * done any recovery; and again on any subsequent remount.
  */
 static void ext3_init_journal_params(struct super_block *sb, journal_t *journal)
 {
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index d0685596e5a6..961ada28db5e 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/checkpoint.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
  *
  * Copyright 1999 Red Hat Software --- All Rights Reserved
@@ -9,8 +9,8 @@
  * the terms of the GNU General Public License, version 2, or at your
  * option, any later version, incorporated herein by reference.
  *
- * Checkpoint routines for the generic filesystem journaling code.  
- * Part of the ext2fs journaling system.  
+ * Checkpoint routines for the generic filesystem journaling code.
+ * Part of the ext2fs journaling system.
  *
  * Checkpointing is the process of ensuring that a section of the log is
  * committed fully to disk, so that that portion of the log can be
@@ -226,7 +226,7 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
  * Try to flush one buffer from the checkpoint list to disk.
  *
  * Return 1 if something happened which requires us to abort the current
- * scan of the checkpoint list.  
+ * scan of the checkpoint list.
  *
  * Called with j_list_lock held and drops it if 1 is returned
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
@@ -270,7 +270,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		 * possibly block, while still holding the journal lock.
 		 * We cannot afford to let the transaction logic start
 		 * messing around with this buffer before we write it to
-		 * disk, as that would break recoverability.  
+		 * disk, as that would break recoverability.
 		 */
 		BUFFER_TRACE(bh, "queue");
 		get_bh(bh);
@@ -293,7 +293,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
  * Perform an actual checkpoint. We take the first transaction on the
  * list of transactions to be checkpointed and send all its buffers
  * to disk. We submit larger chunks of data at once.
- * 
+ *
  * The journal should be locked before calling this function.
  */
 int log_do_checkpoint(journal_t *journal)
@@ -304,10 +304,10 @@ int log_do_checkpoint(journal_t *journal)
 
 	jbd_debug(1, "Start checkpoint\n");
 
-	/* 
+	/*
 	 * First thing: if there are any transactions in the log which
 	 * don't need checkpointing, just eliminate them from the
-	 * journal straight away.  
+	 * journal straight away.
 	 */
 	result = cleanup_journal_tail(journal);
 	jbd_debug(1, "cleanup_journal_tail returned %d\n", result);
@@ -385,9 +385,9 @@ out:
  * we have already got rid of any since the last update of the log tail
  * in the journal superblock.  If so, we can instantly roll the
  * superblock forward to remove those transactions from the log.
- * 
+ *
  * Return <0 on error, 0 on success, 1 if there was nothing to clean up.
- * 
+ *
  * Called with the journal lock held.
  *
  * This is the only part of the journaling code which really needs to be
@@ -404,8 +404,8 @@ int cleanup_journal_tail(journal_t *journal)
 	unsigned long	blocknr, freed;
 
 	/* OK, work out the oldest transaction remaining in the log, and
-	 * the log block it starts at. 
-	 * 
+	 * the log block it starts at.
+	 *
 	 * If the log is now empty, we need to work out which is the
 	 * next transaction ID we will write, and where it will
 	 * start. */
@@ -558,7 +558,7 @@ out:
 	return ret;
 }
 
-/* 
+/*
  * journal_remove_checkpoint: called after a buffer has been committed
  * to disk (either by being write-back flushed to disk, or being
  * committed to the log).
@@ -636,7 +636,7 @@ out:
  * Called with the journal locked.
  * Called with j_list_lock held.
  */
-void __journal_insert_checkpoint(struct journal_head *jh, 
+void __journal_insert_checkpoint(struct journal_head *jh,
 			       transaction_t *transaction)
 {
 	JBUFFER_TRACE(jh, "entry");
@@ -658,7 +658,7 @@ void __journal_insert_checkpoint(struct journal_head *jh,
 
 /*
  * We've finished with this transaction structure: adios...
- * 
+ *
  * The transaction must have no links except for the checkpoint by this
  * point.
  *
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index f66724ce443a..87c5a6d00805 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -578,7 +578,7 @@ int journal_next_log_block(journal_t *journal, unsigned long *retp)
  * this is a no-op.  If needed, we can use j_blk_offset - everything is
  * ready.
  */
-int journal_bmap(journal_t *journal, unsigned long blocknr, 
+int journal_bmap(journal_t *journal, unsigned long blocknr,
 		 unsigned long *retp)
 {
 	int err = 0;
@@ -699,10 +699,10 @@ fail:
  *  @len:  Lenght of the journal in blocks.
  *  @blocksize: blocksize of journalling device
  *  @returns: a newly created journal_t *
- *  
+ *
  *  journal_init_dev creates a journal which maps a fixed contiguous
  *  range of blocks on an arbitrary block device.
- * 
+ *
  */
 journal_t * journal_init_dev(struct block_device *bdev,
 			struct block_device *fs_dev,
@@ -739,11 +739,11 @@ journal_t * journal_init_dev(struct block_device *bdev,
 
 	return journal;
 }
- 
-/** 
+
+/**
  *  journal_t * journal_init_inode () - creates a journal which maps to a inode.
  *  @inode: An inode to create the journal in
- *  
+ *
  * journal_init_inode creates a journal which maps an on-disk inode as
  * the journal.  The inode must exist already, must support bmap() and
  * must have all data blocks preallocated.
@@ -763,7 +763,7 @@ journal_t * journal_init_inode (struct inode *inode)
 	journal->j_inode = inode;
 	jbd_debug(1,
 		  "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n",
-		  journal, inode->i_sb->s_id, inode->i_ino, 
+		  journal, inode->i_sb->s_id, inode->i_ino,
 		  (long long) inode->i_size,
 		  inode->i_sb->s_blocksize_bits, inode->i_sb->s_blocksize);
 
@@ -798,10 +798,10 @@ journal_t * journal_init_inode (struct inode *inode)
 	return journal;
 }
 
-/* 
+/*
  * If the journal init or create aborts, we need to mark the journal
  * superblock as being NULL to prevent the journal destroy from writing
- * back a bogus superblock. 
+ * back a bogus superblock.
  */
 static void journal_fail_superblock (journal_t *journal)
 {
@@ -844,13 +844,13 @@ static int journal_reset(journal_t *journal)
 	return 0;
 }
 
-/** 
+/**
  * int journal_create() - Initialise the new journal file
  * @journal: Journal to create. This structure must have been initialised
- * 
+ *
  * Given a journal_t structure which tells us which disk blocks we can
  * use, create a new journal superblock and initialise all of the
- * journal fields from scratch.  
+ * journal fields from scratch.
  **/
 int journal_create(journal_t *journal)
 {
@@ -915,7 +915,7 @@ int journal_create(journal_t *journal)
 	return journal_reset(journal);
 }
 
-/** 
+/**
  * void journal_update_superblock() - Update journal sb on disk.
  * @journal: The journal to update.
  * @wait: Set to '0' if you don't want to wait for IO completion.
@@ -939,7 +939,7 @@ void journal_update_superblock(journal_t *journal, int wait)
 				journal->j_transaction_sequence) {
 		jbd_debug(1,"JBD: Skipping superblock update on recovered sb "
 			"(start %ld, seq %d, errno %d)\n",
-			journal->j_tail, journal->j_tail_sequence, 
+			journal->j_tail, journal->j_tail_sequence,
 			journal->j_errno);
 		goto out;
 	}
@@ -1062,7 +1062,7 @@ static int load_superblock(journal_t *journal)
 /**
  * int journal_load() - Read journal from disk.
  * @journal: Journal to act on.
- * 
+ *
  * Given a journal_t structure which tells us which disk blocks contain
  * a journal, read the journal from disk to initialise the in-memory
  * structures.
@@ -1172,9 +1172,9 @@ void journal_destroy(journal_t *journal)
  * @compat: bitmask of compatible features
  * @ro: bitmask of features that force read-only mount
  * @incompat: bitmask of incompatible features
- * 
+ *
  * Check whether the journal uses all of a given set of
- * features.  Return true (non-zero) if it does. 
+ * features.  Return true (non-zero) if it does.
  **/
 
 int journal_check_used_features (journal_t *journal, unsigned long compat,
@@ -1203,7 +1203,7 @@ int journal_check_used_features (journal_t *journal, unsigned long compat,
  * @compat: bitmask of compatible features
  * @ro: bitmask of features that force read-only mount
  * @incompat: bitmask of incompatible features
- * 
+ *
  * Check whether the journaling code supports the use of
  * all of a given set of features on this journal.  Return true
  * (non-zero) if it can. */
@@ -1241,7 +1241,7 @@ int journal_check_available_features (journal_t *journal, unsigned long compat,
  * @incompat: bitmask of incompatible features
  *
  * Mark a given journal feature as present on the
- * superblock.  Returns true if the requested features could be set. 
+ * superblock.  Returns true if the requested features could be set.
  *
  */
 
@@ -1327,7 +1327,7 @@ static int journal_convert_superblock_v1(journal_t *journal,
 /**
  * int journal_flush () - Flush journal
  * @journal: Journal to act on.
- * 
+ *
  * Flush all data for a given journal to disk and empty the journal.
  * Filesystems can use this when remounting readonly to ensure that
  * recovery does not need to happen on remount.
@@ -1394,7 +1394,7 @@ int journal_flush(journal_t *journal)
  * int journal_wipe() - Wipe journal contents
  * @journal: Journal to act on.
  * @write: flag (see below)
- * 
+ *
  * Wipe out all of the contents of a journal, safely.  This will produce
  * a warning if the journal contains any valid recovery information.
  * Must be called between journal_init_*() and journal_load().
@@ -1449,7 +1449,7 @@ static const char *journal_dev_name(journal_t *journal, char *buffer)
 
 /*
  * Journal abort has very specific semantics, which we describe
- * for journal abort. 
+ * for journal abort.
  *
  * Two internal function, which provide abort to te jbd layer
  * itself are here.
@@ -1504,7 +1504,7 @@ static void __journal_abort_soft (journal_t *journal, int errno)
  * Perform a complete, immediate shutdown of the ENTIRE
  * journal (not of a single transaction).  This operation cannot be
  * undone without closing and reopening the journal.
- *           
+ *
  * The journal_abort function is intended to support higher level error
  * recovery mechanisms such as the ext2/ext3 remount-readonly error
  * mode.
@@ -1538,7 +1538,7 @@ static void __journal_abort_soft (journal_t *journal, int errno)
  * supply an errno; a null errno implies that absolutely no further
  * writes are done to the journal (unless there are any already in
  * progress).
- * 
+ *
  */
 
 void journal_abort(journal_t *journal, int errno)
@@ -1546,7 +1546,7 @@ void journal_abort(journal_t *journal, int errno)
 	__journal_abort_soft(journal, errno);
 }
 
-/** 
+/**
  * int journal_errno () - returns the journal's error state.
  * @journal: journal to examine.
  *
@@ -1570,7 +1570,7 @@ int journal_errno(journal_t *journal)
 	return err;
 }
 
-/** 
+/**
  * int journal_clear_err () - clears the journal's error state
  * @journal: journal to act on.
  *
@@ -1590,7 +1590,7 @@ int journal_clear_err(journal_t *journal)
 	return err;
 }
 
-/** 
+/**
  * void journal_ack_err() - Ack journal err.
  * @journal: journal to act on.
  *
@@ -1612,7 +1612,7 @@ int journal_blocks_per_page(struct inode *inode)
 
 /*
  * Simple support for retrying memory allocations.  Introduced to help to
- * debug different VM deadlock avoidance strategies. 
+ * debug different VM deadlock avoidance strategies.
  */
 void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry)
 {
diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
index de5bafb4e853..73bb64806ed3 100644
--- a/fs/jbd/recovery.c
+++ b/fs/jbd/recovery.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/recovery.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
  *
  * Copyright 1999-2000 Red Hat Software --- All Rights Reserved
@@ -10,7 +10,7 @@
  * option, any later version, incorporated herein by reference.
  *
  * Journal recovery routines for the generic filesystem journaling code;
- * part of the ext2fs journaling system.  
+ * part of the ext2fs journaling system.
  */
 
 #ifndef __KERNEL__
@@ -25,9 +25,9 @@
 
 /*
  * Maintain information about the progress of the recovery job, so that
- * the different passes can carry information between them. 
+ * the different passes can carry information between them.
  */
-struct recovery_info 
+struct recovery_info
 {
 	tid_t		start_transaction;
 	tid_t		end_transaction;
@@ -116,7 +116,7 @@ static int do_readahead(journal_t *journal, unsigned int start)
 	err = 0;
 
 failed:
-	if (nbufs) 
+	if (nbufs)
 		journal_brelse_array(bufs, nbufs);
 	return err;
 }
@@ -128,7 +128,7 @@ failed:
  * Read a block from the journal
  */
 
-static int jread(struct buffer_head **bhp, journal_t *journal, 
+static int jread(struct buffer_head **bhp, journal_t *journal,
 		 unsigned int offset)
 {
 	int err;
@@ -212,14 +212,14 @@ do {									\
 /**
  * journal_recover - recovers a on-disk journal
  * @journal: the journal to recover
- * 
+ *
  * The primary function for recovering the log contents when mounting a
- * journaled device.  
+ * journaled device.
  *
  * Recovery is done in three passes.  In the first pass, we look for the
  * end of the log.  In the second, we assemble the list of revoke
  * blocks.  In the third and final pass, we replay any un-revoked blocks
- * in the log.  
+ * in the log.
  */
 int journal_recover(journal_t *journal)
 {
@@ -231,10 +231,10 @@ int journal_recover(journal_t *journal)
 	memset(&info, 0, sizeof(info));
 	sb = journal->j_superblock;
 
-	/* 
+	/*
 	 * The journal superblock's s_start field (the current log head)
 	 * is always zero if, and only if, the journal was cleanly
-	 * unmounted.  
+	 * unmounted.
 	 */
 
 	if (!sb->s_start) {
@@ -253,7 +253,7 @@ int journal_recover(journal_t *journal)
 	jbd_debug(0, "JBD: recovery, exit status %d, "
 		  "recovered transactions %u to %u\n",
 		  err, info.start_transaction, info.end_transaction);
-	jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n", 
+	jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n",
 		  info.nr_replays, info.nr_revoke_hits, info.nr_revokes);
 
 	/* Restart the log at the next transaction ID, thus invalidating
@@ -268,15 +268,15 @@ int journal_recover(journal_t *journal)
 /**
  * journal_skip_recovery - Start journal and wipe exiting records
  * @journal: journal to startup
- * 
+ *
  * Locate any valid recovery information from the journal and set up the
  * journal structures in memory to ignore it (presumably because the
- * caller has evidence that it is out of date).  
+ * caller has evidence that it is out of date).
  * This function does'nt appear to be exorted..
  *
  * We perform one pass over the journal to allow us to tell the user how
  * much recovery information is being erased, and to let us initialise
- * the journal transaction sequence numbers to the next unused ID. 
+ * the journal transaction sequence numbers to the next unused ID.
  */
 int journal_skip_recovery(journal_t *journal)
 {
@@ -297,7 +297,7 @@ int journal_skip_recovery(journal_t *journal)
 #ifdef CONFIG_JBD_DEBUG
 		int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence);
 #endif
-		jbd_debug(0, 
+		jbd_debug(0,
 			  "JBD: ignoring %d transaction%s from the journal.\n",
 			  dropped, (dropped == 1) ? "" : "s");
 		journal->j_transaction_sequence = ++info.end_transaction;
@@ -324,10 +324,10 @@ static int do_one_pass(journal_t *journal,
 	MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t))
 			       / sizeof(journal_block_tag_t));
 
-	/* 
+	/*
 	 * First thing is to establish what we expect to find in the log
 	 * (in terms of transaction IDs), and where (in terms of log
-	 * block offsets): query the superblock.  
+	 * block offsets): query the superblock.
 	 */
 
 	sb = journal->j_superblock;
@@ -344,7 +344,7 @@ static int do_one_pass(journal_t *journal,
 	 * Now we walk through the log, transaction by transaction,
 	 * making sure that each transaction has a commit block in the
 	 * expected place.  Each complete transaction gets replayed back
-	 * into the main filesystem. 
+	 * into the main filesystem.
 	 */
 
 	while (1) {
@@ -379,8 +379,8 @@ static int do_one_pass(journal_t *journal,
 		next_log_block++;
 		wrap(journal, next_log_block);
 
-		/* What kind of buffer is it? 
-		 * 
+		/* What kind of buffer is it?
+		 *
 		 * If it is a descriptor block, check that it has the
 		 * expected sequence number.  Otherwise, we're all done
 		 * here. */
@@ -394,7 +394,7 @@ static int do_one_pass(journal_t *journal,
 
 		blocktype = be32_to_cpu(tmp->h_blocktype);
 		sequence = be32_to_cpu(tmp->h_sequence);
-		jbd_debug(3, "Found magic %d, sequence %d\n", 
+		jbd_debug(3, "Found magic %d, sequence %d\n",
 			  blocktype, sequence);
 
 		if (sequence != next_commit_ID) {
@@ -438,7 +438,7 @@ static int do_one_pass(journal_t *journal,
 					/* Recover what we can, but
 					 * report failure at the end. */
 					success = err;
-					printk (KERN_ERR 
+					printk (KERN_ERR
 						"JBD: IO error %d recovering "
 						"block %ld in log\n",
 						err, io_block);
@@ -452,7 +452,7 @@ static int do_one_pass(journal_t *journal,
 					 * revoked, then we're all done
 					 * here. */
 					if (journal_test_revoke
-					    (journal, blocknr, 
+					    (journal, blocknr,
 					     next_commit_ID)) {
 						brelse(obh);
 						++info->nr_revoke_hits;
@@ -465,7 +465,7 @@ static int do_one_pass(journal_t *journal,
 							blocknr,
 							journal->j_blocksize);
 					if (nbh == NULL) {
-						printk(KERN_ERR 
+						printk(KERN_ERR
 						       "JBD: Out of memory "
 						       "during recovery.\n");
 						err = -ENOMEM;
@@ -537,7 +537,7 @@ static int do_one_pass(journal_t *journal,
 	}
 
  done:
-	/* 
+	/*
 	 * We broke out of the log scan loop: either we came to the
 	 * known end of the log or we found an unexpected block in the
 	 * log.  If the latter happened, then we know that the "current"
@@ -567,7 +567,7 @@ static int do_one_pass(journal_t *journal,
 
 /* Scan a revoke record, marking all blocks mentioned as revoked. */
 
-static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, 
+static int scan_revoke_records(journal_t *journal, struct buffer_head *bh,
 			       tid_t sequence, struct recovery_info *info)
 {
 	journal_revoke_header_t *header;
diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c
index a56144183462..c532429d8d9b 100644
--- a/fs/jbd/revoke.c
+++ b/fs/jbd/revoke.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/revoke.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 2000
  *
  * Copyright 2000 Red Hat corp --- All Rights Reserved
@@ -15,10 +15,10 @@
  * Revoke is the mechanism used to prevent old log records for deleted
  * metadata from being replayed on top of newer data using the same
  * blocks.  The revoke mechanism is used in two separate places:
- * 
+ *
  * + Commit: during commit we write the entire list of the current
  *   transaction's revoked blocks to the journal
- * 
+ *
  * + Recovery: during recovery we record the transaction ID of all
  *   revoked blocks.  If there are multiple revoke records in the log
  *   for a single block, only the last one counts, and if there is a log
@@ -29,7 +29,7 @@
  * single transaction:
  *
  * Block is revoked and then journaled:
- *   The desired end result is the journaling of the new block, so we 
+ *   The desired end result is the journaling of the new block, so we
  *   cancel the revoke before the transaction commits.
  *
  * Block is journaled and then revoked:
@@ -41,7 +41,7 @@
  *   transaction must have happened after the block was journaled and so
  *   the revoke must take precedence.
  *
- * Block is revoked and then written as data: 
+ * Block is revoked and then written as data:
  *   The data write is allowed to succeed, but the revoke is _not_
  *   cancelled.  We still need to prevent old log records from
  *   overwriting the new data.  We don't even need to clear the revoke
@@ -54,7 +54,7 @@
  *			buffer has not been revoked, and cancel_revoke
  *			need do nothing.
  * RevokeValid set, Revoked set:
- *			buffer has been revoked.  
+ *			buffer has been revoked.
  */
 
 #ifndef __KERNEL__
@@ -77,7 +77,7 @@ static kmem_cache_t *revoke_table_cache;
    journal replay, this involves recording the transaction ID of the
    last transaction to revoke this block. */
 
-struct jbd_revoke_record_s 
+struct jbd_revoke_record_s
 {
 	struct list_head  hash;
 	tid_t		  sequence;	/* Used for recovery only */
@@ -90,8 +90,8 @@ struct jbd_revoke_table_s
 {
 	/* It is conceivable that we might want a larger hash table
 	 * for recovery.  Must be a power of two. */
-	int		  hash_size; 
-	int		  hash_shift; 
+	int		  hash_size;
+	int		  hash_shift;
 	struct list_head *hash_table;
 };
 
@@ -301,22 +301,22 @@ void journal_destroy_revoke(journal_t *journal)
 
 #ifdef __KERNEL__
 
-/* 
+/*
  * journal_revoke: revoke a given buffer_head from the journal.  This
  * prevents the block from being replayed during recovery if we take a
  * crash after this current transaction commits.  Any subsequent
  * metadata writes of the buffer in this transaction cancel the
- * revoke.  
+ * revoke.
  *
  * Note that this call may block --- it is up to the caller to make
  * sure that there are no further calls to journal_write_metadata
  * before the revoke is complete.  In ext3, this implies calling the
  * revoke before clearing the block bitmap when we are deleting
- * metadata. 
+ * metadata.
  *
  * Revoke performs a journal_forget on any buffer_head passed in as a
  * parameter, but does _not_ forget the buffer_head if the bh was only
- * found implicitly. 
+ * found implicitly.
  *
  * bh_in may not be a journalled buffer - it may have come off
  * the hash tables without an attached journal_head.
@@ -325,7 +325,7 @@ void journal_destroy_revoke(journal_t *journal)
  * by one.
  */
 
-int journal_revoke(handle_t *handle, unsigned long blocknr, 
+int journal_revoke(handle_t *handle, unsigned long blocknr,
 		   struct buffer_head *bh_in)
 {
 	struct buffer_head *bh = NULL;
@@ -487,7 +487,7 @@ void journal_switch_revoke_table(journal_t *journal)
 	else
 		journal->j_revoke = journal->j_revoke_table[0];
 
-	for (i = 0; i < journal->j_revoke->hash_size; i++) 
+	for (i = 0; i < journal->j_revoke->hash_size; i++)
 		INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]);
 }
 
@@ -498,7 +498,7 @@ void journal_switch_revoke_table(journal_t *journal)
  * Called with the journal lock held.
  */
 
-void journal_write_revoke_records(journal_t *journal, 
+void journal_write_revoke_records(journal_t *journal,
 				  transaction_t *transaction)
 {
 	struct journal_head *descriptor;
@@ -507,7 +507,7 @@ void journal_write_revoke_records(journal_t *journal,
 	struct list_head *hash_list;
 	int i, offset, count;
 
-	descriptor = NULL; 
+	descriptor = NULL;
 	offset = 0;
 	count = 0;
 
@@ -519,10 +519,10 @@ void journal_write_revoke_records(journal_t *journal,
 		hash_list = &revoke->hash_table[i];
 
 		while (!list_empty(hash_list)) {
-			record = (struct jbd_revoke_record_s *) 
+			record = (struct jbd_revoke_record_s *)
 				hash_list->next;
 			write_one_revoke_record(journal, transaction,
-						&descriptor, &offset, 
+						&descriptor, &offset,
 						record);
 			count++;
 			list_del(&record->hash);
@@ -534,14 +534,14 @@ void journal_write_revoke_records(journal_t *journal,
 	jbd_debug(1, "Wrote %d revoke records\n", count);
 }
 
-/* 
+/*
  * Write out one revoke record.  We need to create a new descriptor
- * block if the old one is full or if we have not already created one.  
+ * block if the old one is full or if we have not already created one.
  */
 
-static void write_one_revoke_record(journal_t *journal, 
+static void write_one_revoke_record(journal_t *journal,
 				    transaction_t *transaction,
-				    struct journal_head **descriptorp, 
+				    struct journal_head **descriptorp,
 				    int *offsetp,
 				    struct jbd_revoke_record_s *record)
 {
@@ -584,21 +584,21 @@ static void write_one_revoke_record(journal_t *journal,
 		*descriptorp = descriptor;
 	}
 
-	* ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = 
+	* ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) =
 		cpu_to_be32(record->blocknr);
 	offset += 4;
 	*offsetp = offset;
 }
 
-/* 
+/*
  * Flush a revoke descriptor out to the journal.  If we are aborting,
  * this is a noop; otherwise we are generating a buffer which needs to
  * be waited for during commit, so it has to go onto the appropriate
  * journal buffer list.
  */
 
-static void flush_descriptor(journal_t *journal, 
-			     struct journal_head *descriptor, 
+static void flush_descriptor(journal_t *journal,
+			     struct journal_head *descriptor,
 			     int offset)
 {
 	journal_revoke_header_t *header;
@@ -618,7 +618,7 @@ static void flush_descriptor(journal_t *journal,
 }
 #endif
 
-/* 
+/*
  * Revoke support for recovery.
  *
  * Recovery needs to be able to:
@@ -629,7 +629,7 @@ static void flush_descriptor(journal_t *journal,
  *  check whether a given block in a given transaction should be replayed
  *  (ie. has not been revoked by a revoke record in that or a subsequent
  *  transaction)
- * 
+ *
  *  empty the revoke table after recovery.
  */
 
@@ -637,11 +637,11 @@ static void flush_descriptor(journal_t *journal,
  * First, setting revoke records.  We create a new revoke record for
  * every block ever revoked in the log as we scan it for recovery, and
  * we update the existing records if we find multiple revokes for a
- * single block. 
+ * single block.
  */
 
-int journal_set_revoke(journal_t *journal, 
-		       unsigned long blocknr, 
+int journal_set_revoke(journal_t *journal,
+		       unsigned long blocknr,
 		       tid_t sequence)
 {
 	struct jbd_revoke_record_s *record;
@@ -653,18 +653,18 @@ int journal_set_revoke(journal_t *journal,
 		if (tid_gt(sequence, record->sequence))
 			record->sequence = sequence;
 		return 0;
-	} 
+	}
 	return insert_revoke_hash(journal, blocknr, sequence);
 }
 
-/* 
+/*
  * Test revoke records.  For a given block referenced in the log, has
  * that block been revoked?  A revoke record with a given transaction
  * sequence number revokes all blocks in that transaction and earlier
  * ones, but later transactions still need replayed.
  */
 
-int journal_test_revoke(journal_t *journal, 
+int journal_test_revoke(journal_t *journal,
 			unsigned long blocknr,
 			tid_t sequence)
 {
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index f5169a96260e..bf7fd7117817 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/transaction.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
  *
  * Copyright 1998 Red Hat corp --- All Rights Reserved
@@ -10,7 +10,7 @@
  * option, any later version, incorporated herein by reference.
  *
  * Generic filesystem transaction handling code; part of the ext2fs
- * journaling system.  
+ * journaling system.
  *
  * This file manages transactions (compound commits managed by the
  * journaling code) and handles (individual atomic operations by the
@@ -74,7 +74,7 @@ get_transaction(journal_t *journal, transaction_t *transaction)
  * start_this_handle: Given a handle, deal with any locking or stalling
  * needed to make sure that there is enough journal space for the handle
  * to begin.  Attach the handle to a transaction and set up the
- * transaction's buffer credits.  
+ * transaction's buffer credits.
  */
 
 static int start_this_handle(journal_t *journal, handle_t *handle)
@@ -117,7 +117,7 @@ repeat_locked:
 	if (is_journal_aborted(journal) ||
 	    (journal->j_errno != 0 && !(journal->j_flags & JFS_ACK_ERR))) {
 		spin_unlock(&journal->j_state_lock);
-		ret = -EROFS; 
+		ret = -EROFS;
 		goto out;
 	}
 
@@ -182,7 +182,7 @@ repeat_locked:
 		goto repeat;
 	}
 
-	/* 
+	/*
 	 * The commit code assumes that it can get enough log space
 	 * without forcing a checkpoint.  This is *critical* for
 	 * correctness: a checkpoint of a buffer which is also
@@ -191,7 +191,7 @@ repeat_locked:
 	 *
 	 * We must therefore ensure the necessary space in the journal
 	 * *before* starting to dirty potentially checkpointed buffers
-	 * in the new transaction. 
+	 * in the new transaction.
 	 *
 	 * The worst part is, any transaction currently committing can
 	 * reduce the free space arbitrarily.  Be careful to account for
@@ -246,13 +246,13 @@ static handle_t *new_handle(int nblocks)
 }
 
 /**
- * handle_t *journal_start() - Obtain a new handle.  
+ * handle_t *journal_start() - Obtain a new handle.
  * @journal: Journal to start transaction on.
  * @nblocks: number of block buffer we might modify
  *
  * We make sure that the transaction can guarantee at least nblocks of
  * modified buffers in the log.  We block until the log can guarantee
- * that much space.  
+ * that much space.
  *
  * This function is visible to journal users (like ext3fs), so is not
  * called with the journal already locked.
@@ -292,11 +292,11 @@ handle_t *journal_start(journal_t *journal, int nblocks)
  * int journal_extend() - extend buffer credits.
  * @handle:  handle to 'extend'
  * @nblocks: nr blocks to try to extend by.
- * 
+ *
  * Some transactions, such as large extends and truncates, can be done
  * atomically all at once or in several stages.  The operation requests
  * a credit for a number of buffer modications in advance, but can
- * extend its credit if it needs more.  
+ * extend its credit if it needs more.
  *
  * journal_extend tries to give the running handle more buffer credits.
  * It does not guarantee that allocation - this is a best-effort only.
@@ -363,7 +363,7 @@ out:
  * int journal_restart() - restart a handle .
  * @handle:  handle to restart
  * @nblocks: nr credits requested
- * 
+ *
  * Restart a handle for a multi-transaction filesystem
  * operation.
  *
@@ -462,7 +462,7 @@ void journal_lock_updates(journal_t *journal)
 /**
  * void journal_unlock_updates (journal_t* journal) - release barrier
  * @journal:  Journal to release the barrier on.
- * 
+ *
  * Release a transaction barrier obtained with journal_lock_updates().
  *
  * Should be called without the journal lock held.
@@ -547,8 +547,8 @@ repeat:
 	jbd_lock_bh_state(bh);
 
 	/* We now hold the buffer lock so it is safe to query the buffer
-	 * state.  Is the buffer dirty? 
-	 * 
+	 * state.  Is the buffer dirty?
+	 *
 	 * If so, there are two possibilities.  The buffer may be
 	 * non-journaled, and undergoing a quite legitimate writeback.
 	 * Otherwise, it is journaled, and we don't expect dirty buffers
@@ -566,7 +566,7 @@ repeat:
 		 */
 		if (jh->b_transaction) {
 			J_ASSERT_JH(jh,
-				jh->b_transaction == transaction || 
+				jh->b_transaction == transaction ||
 				jh->b_transaction ==
 					journal->j_committing_transaction);
 			if (jh->b_next_transaction)
@@ -653,7 +653,7 @@ repeat:
 		 * buffer had better remain locked during the kmalloc,
 		 * but that should be true --- we hold the journal lock
 		 * still and the buffer is already on the BUF_JOURNAL
-		 * list so won't be flushed. 
+		 * list so won't be flushed.
 		 *
 		 * Subtle point, though: if this is a get_undo_access,
 		 * then we will be relying on the frozen_data to contain
@@ -765,8 +765,8 @@ int journal_get_write_access(handle_t *handle, struct buffer_head *bh)
  * manually rather than reading off disk), then we need to keep the
  * buffer_head locked until it has been completely filled with new
  * data.  In this case, we should be able to make the assertion that
- * the bh is not already part of an existing transaction.  
- * 
+ * the bh is not already part of an existing transaction.
+ *
  * The buffer should already be locked by the caller by this point.
  * There is no lock ranking violation: it was a newly created,
  * unlocked buffer beforehand. */
@@ -778,7 +778,7 @@ int journal_get_write_access(handle_t *handle, struct buffer_head *bh)
  *
  * Call this if you create a new bh.
  */
-int journal_get_create_access(handle_t *handle, struct buffer_head *bh) 
+int journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
@@ -847,13 +847,13 @@ out:
  * do not reuse freed space until the deallocation has been committed,
  * since if we overwrote that space we would make the delete
  * un-rewindable in case of a crash.
- * 
+ *
  * To deal with that, journal_get_undo_access requests write access to a
  * buffer for parts of non-rewindable operations such as delete
  * operations on the bitmaps.  The journaling code must keep a copy of
  * the buffer's contents prior to the undo_access call until such time
  * as we know that the buffer has definitely been committed to disk.
- * 
+ *
  * We never need to know which transaction the committed data is part
  * of, buffers touched here are guaranteed to be dirtied later and so
  * will be committed to a new transaction in due course, at which point
@@ -911,13 +911,13 @@ out:
 	return err;
 }
 
-/** 
+/**
  * int journal_dirty_data() -  mark a buffer as containing dirty data which
  *                             needs to be flushed before we can commit the
- *                             current transaction.  
+ *                             current transaction.
  * @handle: transaction
  * @bh: bufferhead to mark
- * 
+ *
  * The buffer is placed on the transaction's data list and is marked as
  * belonging to the transaction.
  *
@@ -946,15 +946,15 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh)
 
 	/*
 	 * What if the buffer is already part of a running transaction?
-	 * 
+	 *
 	 * There are two cases:
 	 * 1) It is part of the current running transaction.  Refile it,
 	 *    just in case we have allocated it as metadata, deallocated
-	 *    it, then reallocated it as data. 
+	 *    it, then reallocated it as data.
 	 * 2) It is part of the previous, still-committing transaction.
 	 *    If all we want to do is to guarantee that the buffer will be
 	 *    written to disk before this new transaction commits, then
-	 *    being sure that the *previous* transaction has this same 
+	 *    being sure that the *previous* transaction has this same
 	 *    property is sufficient for us!  Just leave it on its old
 	 *    transaction.
 	 *
@@ -1076,18 +1076,18 @@ no_journal:
 	return 0;
 }
 
-/** 
+/**
  * int journal_dirty_metadata() -  mark a buffer as containing dirty metadata
  * @handle: transaction to add buffer to.
- * @bh: buffer to mark 
- * 
+ * @bh: buffer to mark
+ *
  * mark dirty metadata which needs to be journaled as part of the current
  * transaction.
  *
  * The buffer is placed on the transaction's metadata list and is marked
- * as belonging to the transaction.  
+ * as belonging to the transaction.
  *
- * Returns error number or 0 on success.  
+ * Returns error number or 0 on success.
  *
  * Special care needs to be taken if the buffer already belongs to the
  * current committing transaction (in which case we should have frozen
@@ -1135,11 +1135,11 @@ int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 
 	set_buffer_jbddirty(bh);
 
-	/* 
+	/*
 	 * Metadata already on the current transaction list doesn't
 	 * need to be filed.  Metadata on another transaction's list must
 	 * be committing, and will be refiled once the commit completes:
-	 * leave it alone for now. 
+	 * leave it alone for now.
 	 */
 	if (jh->b_transaction != transaction) {
 		JBUFFER_TRACE(jh, "already on other transaction");
@@ -1165,7 +1165,7 @@ out:
 	return 0;
 }
 
-/* 
+/*
  * journal_release_buffer: undo a get_write_access without any buffer
  * updates, if the update decided in the end that it didn't need access.
  *
@@ -1176,20 +1176,20 @@ journal_release_buffer(handle_t *handle, struct buffer_head *bh)
 	BUFFER_TRACE(bh, "entry");
 }
 
-/** 
+/**
  * void journal_forget() - bforget() for potentially-journaled buffers.
  * @handle: transaction handle
  * @bh:     bh to 'forget'
  *
  * We can only do the bforget if there are no commits pending against the
  * buffer.  If the buffer is dirty in the current running transaction we
- * can safely unlink it. 
+ * can safely unlink it.
  *
  * bh may not be a journalled buffer at all - it may be a non-JBD
  * buffer which came off the hashtable.  Check for this.
  *
  * Decrements bh->b_count by one.
- * 
+ *
  * Allow this call even if the handle has aborted --- it may be part of
  * the caller's cleanup after an abort.
  */
@@ -1237,7 +1237,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh)
 
 		drop_reserve = 1;
 
-		/* 
+		/*
 		 * We are no longer going to journal this buffer.
 		 * However, the commit of this transaction is still
 		 * important to the buffer: the delete that we are now
@@ -1246,7 +1246,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh)
 		 *
 		 * So, if we have a checkpoint on the buffer, we should
 		 * now refile the buffer on our BJ_Forget list so that
-		 * we know to remove the checkpoint after we commit. 
+		 * we know to remove the checkpoint after we commit.
 		 */
 
 		if (jh->b_cp_transaction) {
@@ -1264,7 +1264,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh)
 			}
 		}
 	} else if (jh->b_transaction) {
-		J_ASSERT_JH(jh, (jh->b_transaction == 
+		J_ASSERT_JH(jh, (jh->b_transaction ==
 				 journal->j_committing_transaction));
 		/* However, if the buffer is still owned by a prior
 		 * (committing) transaction, we can't drop it yet... */
@@ -1294,7 +1294,7 @@ drop:
 /**
  * int journal_stop() - complete a transaction
  * @handle: tranaction to complete.
- * 
+ *
  * All done for a particular handle.
  *
  * There is not much action needed here.  We just return any remaining
@@ -1303,7 +1303,7 @@ drop:
  * filesystem is marked for synchronous update.
  *
  * journal_stop itself will not usually return an error, but it may
- * do so in unusual circumstances.  In particular, expect it to 
+ * do so in unusual circumstances.  In particular, expect it to
  * return -EIO if a journal_abort has been executed since the
  * transaction began.
  */
@@ -1388,7 +1388,7 @@ int journal_stop(handle_t *handle)
 
 		/*
 		 * Special case: JFS_SYNC synchronous updates require us
-		 * to wait for the commit to complete.  
+		 * to wait for the commit to complete.
 		 */
 		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
 			err = log_wait_commit(journal, tid);
@@ -1439,7 +1439,7 @@ int journal_force_commit(journal_t *journal)
  * jbd_lock_bh_state(jh2bh(jh)) is held.
  */
 
-static inline void 
+static inline void
 __blist_add_buffer(struct journal_head **list, struct journal_head *jh)
 {
 	if (!*list) {
@@ -1454,7 +1454,7 @@ __blist_add_buffer(struct journal_head **list, struct journal_head *jh)
 	}
 }
 
-/* 
+/*
  * Remove a buffer from a transaction list, given the transaction's list
  * head pointer.
  *
@@ -1475,7 +1475,7 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh)
 	jh->b_tnext->b_tprev = jh->b_tprev;
 }
 
-/* 
+/*
  * Remove a buffer from the appropriate transaction list.
  *
  * Note that this function can *change* the value of
@@ -1595,17 +1595,17 @@ out:
 }
 
 
-/** 
+/**
  * int journal_try_to_free_buffers() - try to free page buffers.
  * @journal: journal for operation
  * @page: to try and free
  * @unused_gfp_mask: unused
  *
- * 
+ *
  * For all the buffers on this page,
  * if they are fully written out ordered data, move them onto BUF_CLEAN
  * so try_to_free_buffers() can reap them.
- * 
+ *
  * This function returns non-zero if we wish try_to_free_buffers()
  * to be called. We do this if the page is releasable by try_to_free_buffers().
  * We also do it if the page has locked or dirty buffers and the caller wants
@@ -1629,7 +1629,7 @@ out:
  * cannot happen because we never reallocate freed data as metadata
  * while the data is part of a transaction.  Yes?
  */
-int journal_try_to_free_buffers(journal_t *journal, 
+int journal_try_to_free_buffers(journal_t *journal,
 				struct page *page, gfp_t unused_gfp_mask)
 {
 	struct buffer_head *head;
@@ -1697,7 +1697,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
 }
 
 /*
- * journal_invalidatepage 
+ * journal_invalidatepage
  *
  * This code is tricky.  It has a number of cases to deal with.
  *
@@ -1705,15 +1705,15 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
  *
  * i_size must be updated on disk before we start calling invalidatepage on the
  * data.
- * 
+ *
  *  This is done in ext3 by defining an ext3_setattr method which
  *  updates i_size before truncate gets going.  By maintaining this
  *  invariant, we can be sure that it is safe to throw away any buffers
  *  attached to the current transaction: once the transaction commits,
  *  we know that the data will not be needed.
- * 
+ *
  *  Note however that we can *not* throw away data belonging to the
- *  previous, committing transaction!  
+ *  previous, committing transaction!
  *
  * Any disk blocks which *are* part of the previous, committing
  * transaction (and which therefore cannot be discarded immediately) are
@@ -1732,7 +1732,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
  * don't make guarantees about the order in which data hits disk --- in
  * particular we don't guarantee that new dirty data is flushed before
  * transaction commit --- so it is always safe just to discard data
- * immediately in that mode.  --sct 
+ * immediately in that mode.  --sct
  */
 
 /*
@@ -1876,9 +1876,9 @@ zap_buffer_unlocked:
 	return may_free;
 }
 
-/** 
+/**
  * void journal_invalidatepage()
- * @journal: journal to use for flush... 
+ * @journal: journal to use for flush...
  * @page:    page to flush
  * @offset:  length of page to invalidate.
  *
@@ -1886,7 +1886,7 @@ zap_buffer_unlocked:
  *
  */
 void journal_invalidatepage(journal_t *journal,
-		      struct page *page, 
+		      struct page *page,
 		      unsigned long offset)
 {
 	struct buffer_head *head, *bh, *next;
@@ -1924,8 +1924,8 @@ void journal_invalidatepage(journal_t *journal,
 	}
 }
 
-/* 
- * File a buffer on the given transaction list. 
+/*
+ * File a buffer on the given transaction list.
  */
 void __journal_file_buffer(struct journal_head *jh,
 			transaction_t *transaction, int jlist)
@@ -1948,7 +1948,7 @@ void __journal_file_buffer(struct journal_head *jh,
 	 * with __jbd_unexpected_dirty_buffer()'s handling of dirty
 	 * state. */
 
-	if (jlist == BJ_Metadata || jlist == BJ_Reserved || 
+	if (jlist == BJ_Metadata || jlist == BJ_Reserved ||
 	    jlist == BJ_Shadow || jlist == BJ_Forget) {
 		if (test_clear_buffer_dirty(bh) ||
 		    test_clear_buffer_jbddirty(bh))
@@ -2008,7 +2008,7 @@ void journal_file_buffer(struct journal_head *jh,
 	jbd_unlock_bh_state(jh2bh(jh));
 }
 
-/* 
+/*
  * Remove a buffer from its current buffer list in preparation for
  * dropping it from its current transaction entirely.  If the buffer has
  * already started to be used by a subsequent transaction, refile the
@@ -2060,7 +2060,7 @@ void __journal_refile_buffer(struct journal_head *jh)
  * to the caller to remove the journal_head if necessary.  For the
  * unlocked journal_refile_buffer call, the caller isn't going to be
  * doing anything else to the buffer so we need to do the cleanup
- * ourselves to avoid a jh leak. 
+ * ourselves to avoid a jh leak.
  *
  * *** The journal_head may be freed by this call! ***
  */
diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h
index c8307c02dd07..ce0e6109aff0 100644
--- a/include/linux/ext3_jbd.h
+++ b/include/linux/ext3_jbd.h
@@ -23,7 +23,7 @@
 
 /* Define the number of blocks we need to account to a transaction to
  * modify one block of data.
- * 
+ *
  * We may have to touch one inode, one bitmap buffer, up to three
  * indirection blocks, the group and superblock summaries, and the data
  * block to complete the transaction.  */
@@ -88,16 +88,16 @@
 #endif
 
 int
-ext3_mark_iloc_dirty(handle_t *handle, 
+ext3_mark_iloc_dirty(handle_t *handle,
 		     struct inode *inode,
 		     struct ext3_iloc *iloc);
 
-/* 
+/*
  * On success, We end up with an outstanding reference count against
- * iloc->bh.  This _must_ be cleaned up later. 
+ * iloc->bh.  This _must_ be cleaned up later.
  */
 
-int ext3_reserve_inode_write(handle_t *handle, struct inode *inode, 
+int ext3_reserve_inode_write(handle_t *handle, struct inode *inode,
 			struct ext3_iloc *iloc);
 
 int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode);
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index a04c154c5207..7d847931ee5a 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -1,6 +1,6 @@
 /*
  * linux/include/linux/jbd.h
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>
  *
  * Copyright 1998-2000 Red Hat, Inc --- All Rights Reserved
@@ -97,8 +97,8 @@ extern void jbd_slab_free(void *ptr, size_t size);
  * number of outstanding buffers possible at any time.  When the
  * operation completes, any buffer credits not used are credited back to
  * the transaction, so that at all times we know how many buffers the
- * outstanding updates on a transaction might possibly touch. 
- * 
+ * outstanding updates on a transaction might possibly touch.
+ *
  * This is an opaque datatype.
  **/
 typedef struct handle_s		handle_t;	/* Atomic operation type */
@@ -108,7 +108,7 @@ typedef struct handle_s		handle_t;	/* Atomic operation type */
  * typedef journal_t - The journal_t maintains all of the journaling state information for a single filesystem.
  *
  * journal_t is linked to from the fs superblock structure.
- * 
+ *
  * We use the journal_t to keep track of all outstanding transaction
  * activity on the filesystem, and to manage the state of the log
  * writing process.
@@ -128,7 +128,7 @@ typedef struct journal_s	journal_t;	/* Journal control structure */
  * On-disk structures
  */
 
-/* 
+/*
  * Descriptor block types:
  */
 
@@ -149,8 +149,8 @@ typedef struct journal_header_s
 } journal_header_t;
 
 
-/* 
- * The block tag: used to describe a single buffer in the journal 
+/*
+ * The block tag: used to describe a single buffer in the journal
  */
 typedef struct journal_block_tag_s
 {
@@ -158,9 +158,9 @@ typedef struct journal_block_tag_s
 	__be32		t_flags;	/* See below */
 } journal_block_tag_t;
 
-/* 
+/*
  * The revoke descriptor: used on disk to describe a series of blocks to
- * be revoked from the log 
+ * be revoked from the log
  */
 typedef struct journal_revoke_header_s
 {
@@ -374,10 +374,10 @@ struct jbd_revoke_table_s;
  **/
 
 /* Docbook can't yet cope with the bit fields, but will leave the documentation
- * in so it can be fixed later. 
+ * in so it can be fixed later.
  */
 
-struct handle_s 
+struct handle_s
 {
 	/* Which compound transaction is this update a part of? */
 	transaction_t		*h_transaction;
@@ -435,7 +435,7 @@ struct handle_s
  *
  */
 
-struct transaction_s 
+struct transaction_s
 {
 	/* Pointer to the journal for this transaction. [no locking] */
 	journal_t		*t_journal;
@@ -455,7 +455,7 @@ struct transaction_s
 		T_RUNDOWN,
 		T_FLUSH,
 		T_COMMIT,
-		T_FINISHED 
+		T_FINISHED
 	}			t_state;
 
 	/*
@@ -569,7 +569,7 @@ struct transaction_s
  *     journal_t.
  * @j_flags:  General journaling state flags
  * @j_errno:  Is there an outstanding uncleared error on the journal (from a
- *     prior abort)? 
+ *     prior abort)?
  * @j_sb_buffer: First part of superblock buffer
  * @j_superblock: Second part of superblock buffer
  * @j_format_version: Version of the superblock format
@@ -583,7 +583,7 @@ struct transaction_s
  * @j_wait_transaction_locked: Wait queue for waiting for a locked transaction
  *  to start committing, or for a barrier lock to be released
  * @j_wait_logspace: Wait queue for waiting for checkpointing to complete
- * @j_wait_done_commit: Wait queue for waiting for commit to complete 
+ * @j_wait_done_commit: Wait queue for waiting for commit to complete
  * @j_wait_checkpoint:  Wait queue to trigger checkpointing
  * @j_wait_commit: Wait queue to trigger commit
  * @j_wait_updates: Wait queue to wait for updates to complete
@@ -592,7 +592,7 @@ struct transaction_s
  * @j_tail: Journal tail - identifies the oldest still-used block in the
  *  journal.
  * @j_free: Journal free - how many free blocks are there in the journal?
- * @j_first: The block number of the first usable block 
+ * @j_first: The block number of the first usable block
  * @j_last: The block number one beyond the last usable block
  * @j_dev: Device where we store the journal
  * @j_blocksize: blocksize for the location where we store the journal.
@@ -604,12 +604,12 @@ struct transaction_s
  * @j_list_lock: Protects the buffer lists and internal buffer state.
  * @j_inode: Optional inode where we store the journal.  If present, all journal
  *     block numbers are mapped into this inode via bmap().
- * @j_tail_sequence:  Sequence number of the oldest transaction in the log 
+ * @j_tail_sequence:  Sequence number of the oldest transaction in the log
  * @j_transaction_sequence: Sequence number of the next transaction to grant
  * @j_commit_sequence: Sequence number of the most recently committed
  *  transaction
  * @j_commit_request: Sequence number of the most recent transaction wanting
- *     commit 
+ *     commit
  * @j_uuid: Uuid of client object.
  * @j_task: Pointer to the current commit thread for this journal
  * @j_max_transaction_buffers:  Maximum number of metadata buffers to allow in a
@@ -823,8 +823,8 @@ struct journal_s
 	void *j_private;
 };
 
-/* 
- * Journal flag definitions 
+/*
+ * Journal flag definitions
  */
 #define JFS_UNMOUNT	0x001	/* Journal thread is being destroyed */
 #define JFS_ABORT	0x002	/* Journaling has been aborted for errors. */
@@ -833,7 +833,7 @@ struct journal_s
 #define JFS_LOADED	0x010	/* The journal superblock has been loaded */
 #define JFS_BARRIER	0x020	/* Use IDE barriers */
 
-/* 
+/*
  * Function declarations for the journaling transaction and buffer
  * management
  */
@@ -862,7 +862,7 @@ int __journal_remove_checkpoint(struct journal_head *);
 void __journal_insert_checkpoint(struct journal_head *, transaction_t *);
 
 /* Buffer IO */
-extern int 
+extern int
 journal_write_metadata_buffer(transaction_t	  *transaction,
 			      struct journal_head  *jh_in,
 			      struct journal_head **jh_out,
@@ -890,7 +890,7 @@ static inline handle_t *journal_current_handle(void)
 /* The journaling code user interface:
  *
  * Create and destroy handles
- * Register buffer modifications against the current transaction. 
+ * Register buffer modifications against the current transaction.
  */
 
 extern handle_t *journal_start(journal_t *, int nblocks);
@@ -917,11 +917,11 @@ extern journal_t * journal_init_dev(struct block_device *bdev,
 				int start, int len, int bsize);
 extern journal_t * journal_init_inode (struct inode *);
 extern int	   journal_update_format (journal_t *);
-extern int	   journal_check_used_features 
+extern int	   journal_check_used_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
-extern int	   journal_check_available_features 
+extern int	   journal_check_available_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
-extern int	   journal_set_features 
+extern int	   journal_set_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   journal_create     (journal_t *);
 extern int	   journal_load       (journal_t *journal);
@@ -1015,7 +1015,7 @@ do {								           \
  * bit, when set, indicates that we have had a fatal error somewhere,
  * either inside the journaling layer or indicated to us by the client
  * (eg. ext3), and that we and should not commit any further
- * transactions.  
+ * transactions.
  */
 
 static inline int is_journal_aborted(journal_t *journal)
@@ -1082,7 +1082,7 @@ static inline int jbd_space_needed(journal_t *journal)
 #define BJ_Reserved	7	/* Buffer is reserved for access by journal */
 #define BJ_Locked	8	/* Locked for I/O during commit */
 #define BJ_Types	9
- 
+
 extern int jbd_blocks_per_page(struct inode *inode);
 
 #ifdef __KERNEL__
-- 
cgit v1.2.3


From 37ed322290eb6d5cf2ab33915793ed4219eae1d6 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <esandeen@redhat.com>
Date: Wed, 27 Sep 2006 01:49:31 -0700
Subject: [PATCH] JBD: 16T fixes

These are a few places I've found in jbd that look like they may not be
16T-safe, or consistent with the use of unsigned longs for block
containers.  Problems here would be somewhat hard to hit, would require
journal blocks past the 8T boundary, which would not be terribly common.
Still, should fix.

(some of these have come from the ext4 work on jbd as well).

I think there's one more possibility that the wrap() function may not be
safe IF your last block in the journal butts right up against the 232 block
boundary, but that seems like a VERY remote possibility, and I'm not
worrying about it at this point.

Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/journal.c    | 6 +++---
 include/linux/jbd.h | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 2613fca92740..88d970e09ce6 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -271,7 +271,7 @@ static void journal_kill_thread(journal_t *journal)
 int journal_write_metadata_buffer(transaction_t *transaction,
 				  struct journal_head  *jh_in,
 				  struct journal_head **jh_out,
-				  int blocknr)
+				  unsigned long blocknr)
 {
 	int need_copy_out = 0;
 	int done_copy_out = 0;
@@ -696,7 +696,7 @@ fail:
  *  @bdev: Block device on which to create the journal
  *  @fs_dev: Device which hold journalled filesystem for this journal.
  *  @start: Block nr Start of journal.
- *  @len:  Lenght of the journal in blocks.
+ *  @len:  Length of the journal in blocks.
  *  @blocksize: blocksize of journalling device
  *  @returns: a newly created journal_t *
  *
@@ -820,7 +820,7 @@ static void journal_fail_superblock (journal_t *journal)
 static int journal_reset(journal_t *journal)
 {
 	journal_superblock_t *sb = journal->j_superblock;
-	unsigned int first, last;
+	unsigned long first, last;
 
 	first = be32_to_cpu(sb->s_first);
 	last = be32_to_cpu(sb->s_maxlen);
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 7d847931ee5a..dc262624efa5 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -732,7 +732,7 @@ struct journal_s
 	 */
 	struct block_device	*j_dev;
 	int			j_blocksize;
-	unsigned int		j_blk_offset;
+	unsigned long		j_blk_offset;
 
 	/*
 	 * Device which holds the client fs.  For internal journal this will be
@@ -866,7 +866,7 @@ extern int
 journal_write_metadata_buffer(transaction_t	  *transaction,
 			      struct journal_head  *jh_in,
 			      struct journal_head **jh_out,
-			      int		   blocknr);
+			      unsigned long	   blocknr);
 
 /* Transaction locking */
 extern void		__wait_on_journal (journal_t *);
-- 
cgit v1.2.3


From e9ad5620bfb901df8a7a2603c88689ededeecaf1 Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@austin.ibm.com>
Date: Wed, 27 Sep 2006 01:49:35 -0700
Subject: [PATCH] ext3: More whitespace cleanups

More white space cleanups in preparation of cloning ext4 from ext3.
Removing spaces that precede a tab.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/acl.c             |  2 +-
 fs/ext3/balloc.c          | 32 ++++++++++++++++----------------
 fs/ext3/dir.c             |  2 +-
 fs/ext3/hash.c            |  2 +-
 fs/ext3/inode.c           |  8 ++++----
 fs/ext3/namei.c           | 18 +++++++++---------
 fs/ext3/super.c           |  6 +++---
 fs/jbd/checkpoint.c       |  2 +-
 fs/jbd/journal.c          |  2 +-
 fs/jbd/recovery.c         |  2 +-
 fs/jbd/transaction.c      |  6 +++---
 include/linux/ext3_fs.h   |  2 +-
 include/linux/ext3_fs_i.h |  2 +-
 include/linux/jbd.h       | 10 +++++-----
 14 files changed, 48 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 0d21d558b87a..92bf78221429 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -258,7 +258,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type,
 		default:
 			return -EINVAL;
 	}
- 	if (acl) {
+	if (acl) {
 		value = ext3_acl_to_disk(acl, &size);
 		if (IS_ERR(value))
 			return (int)PTR_ERR(value);
diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index f1897feb18f4..c19be8fc3e51 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -40,7 +40,7 @@
 
 /**
  * ext3_get_group_desc() -- load group descriptor from disk
- * @sb: 		super block
+ * @sb:			super block
  * @block_group:	given block group
  * @bh:			pointer to the buffer head to store the block
  *			group descriptor
@@ -355,7 +355,7 @@ void ext3_init_block_alloc_info(struct inode *inode)
 		rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
 		rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
 
-	 	/*
+		/*
 		 * if filesystem is mounted with NORESERVATION, the goal
 		 * reservation window size is set to zero to indicate
 		 * block reservation is off
@@ -681,7 +681,7 @@ bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 		jbd_lock_bh_state(bh);
 		if (jh->b_committed_data)
 			start = ext3_find_next_zero_bit(jh->b_committed_data,
-						 	maxblocks, next);
+							maxblocks, next);
 		jbd_unlock_bh_state(bh);
 	}
 	return -1;
@@ -790,10 +790,10 @@ claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh)
  * and at last, allocate the blocks by claiming the found free bit as allocated.
  *
  * To set the range of this allocation:
- * 	if there is a reservation window, only try to allocate block(s) from the
+ *	if there is a reservation window, only try to allocate block(s) from the
  *	file's own reservation window;
  *	Otherwise, the allocation range starts from the give goal block, ends at
- * 	the block group's last block.
+ *	the block group's last block.
  *
  * If we failed to allocate the desired block then we may end up crossing to a
  * new bitmap.  In that case we must release write access to the old one via
@@ -880,12 +880,12 @@ fail_access:
 }
 
 /**
- * 	find_next_reservable_window():
+ *	find_next_reservable_window():
  *		find a reservable space within the given range.
  *		It does not allocate the reservation window for now:
  *		alloc_new_reservation() will do the work later.
  *
- * 	@search_head: the head of the searching list;
+ *	@search_head: the head of the searching list;
  *		This is not necessarily the list head of the whole filesystem
  *
  *		We have both head and start_block to assist the search
@@ -893,12 +893,12 @@ fail_access:
  *		but we will shift to the place where start_block is,
  *		then start from there, when looking for a reservable space.
  *
- * 	@size: the target new reservation window size
+ *	@size: the target new reservation window size
  *
- * 	@group_first_block: the first block we consider to start
+ *	@group_first_block: the first block we consider to start
  *			the real search from
  *
- * 	@last_block:
+ *	@last_block:
  *		the maximum block number that our goal reservable space
  *		could start from. This is normally the last block in this
  *		group. The search will end when we found the start of next
@@ -906,10 +906,10 @@ fail_access:
  *		This could handle the cross boundary reservation window
  *		request.
  *
- * 	basically we search from the given range, rather than the whole
- * 	reservation double linked list, (start_block, last_block)
- * 	to find a free region that is of my size and has not
- * 	been reserved.
+ *	basically we search from the given range, rather than the whole
+ *	reservation double linked list, (start_block, last_block)
+ *	to find a free region that is of my size and has not
+ *	been reserved.
  *
  */
 static int find_next_reservable_window(
@@ -962,7 +962,7 @@ static int find_next_reservable_window(
 			/*
 			 * Found a reserveable space big enough.  We could
 			 * have a reservation across the group boundary here
-		 	 */
+			 */
 			break;
 		}
 	}
@@ -998,7 +998,7 @@ static int find_next_reservable_window(
 }
 
 /**
- * 	alloc_new_reservation()--allocate a new reservation window
+ *	alloc_new_reservation()--allocate a new reservation window
  *
  *		To make a new reservation, we search part of the filesystem
  *		reservation list (the list that inside the group). We try to
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 6f9e5a523c87..4d97f437cd48 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -67,7 +67,7 @@ int ext3_check_dir_entry (const char * function, struct inode * dir,
 			  unsigned long offset)
 {
 	const char * error_msg = NULL;
- 	const int rlen = le16_to_cpu(de->rec_len);
+	const int rlen = le16_to_cpu(de->rec_len);
 
 	if (rlen < EXT3_DIR_REC_LEN(1))
 		error_msg = "rec_len is smaller than minimal";
diff --git a/fs/ext3/hash.c b/fs/ext3/hash.c
index 7fa637cd322a..deeb27b5ba83 100644
--- a/fs/ext3/hash.c
+++ b/fs/ext3/hash.c
@@ -95,7 +95,7 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 	__u32	minor_hash = 0;
 	const char	*p;
 	int		i;
-	__u32 		in[8], buf[4];
+	__u32		in[8], buf[4];
 
 	/* Initialize the default seed for the hash checksum functions */
 	buf[0] = 0x67452301;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 56665fab027d..c9fc15f8b609 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -13,11 +13,11 @@
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *
  *  Goal-directed block allocation by Stephen Tweedie
- * 	(sct@redhat.com), 1993, 1998
+ *	(sct@redhat.com), 1993, 1998
  *  Big-endian to little-endian byte-swapping/bitmaps by
  *        David S. Miller (davem@caip.rutgers.edu), 1995
  *  64-bit file support on 64-bit platforms by Jakub Jelinek
- * 	(jj@sunsite.ms.mff.cuni.cz)
+ *	(jj@sunsite.ms.mff.cuni.cz)
  *
  *  Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000
  */
@@ -470,7 +470,7 @@ static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block,
  *	ext3_blks_to_allocate: Look up the block map and count the number
  *	of direct blocks need to be allocated for the given branch.
  *
- * 	@branch: chain of indirect blocks
+ *	@branch: chain of indirect blocks
  *	@k: number of blocks need for indirect blocks
  *	@blks: number of data blocks to be mapped.
  *	@blocks_to_boundary:  the offset in the indirect block
@@ -1098,7 +1098,7 @@ static int walk_page_buffers(	handle_t *handle,
 
 	for (	bh = head, block_start = 0;
 		ret == 0 && (bh != head || !block_start);
-	    	block_start = block_end, bh = next)
+		block_start = block_end, bh = next)
 	{
 		next = bh->b_this_page;
 		block_end = block_start + blocksize;
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 70dc5206ebfa..85d132c37ee0 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -15,13 +15,13 @@
  *  Big-endian to little-endian byte-swapping/bitmaps by
  *        David S. Miller (davem@caip.rutgers.edu), 1995
  *  Directory entry file type support and forward compatibility hooks
- *  	for B-tree directories by Theodore Ts'o (tytso@mit.edu), 1998
+ *	for B-tree directories by Theodore Ts'o (tytso@mit.edu), 1998
  *  Hash Tree Directory indexing (c)
- *  	Daniel Phillips, 2001
+ *	Daniel Phillips, 2001
  *  Hash Tree Directory indexing porting
- *  	Christopher Li, 2002
+ *	Christopher Li, 2002
  *  Hash Tree Directory indexing cleanup
- * 	Theodore Ts'o, 2002
+ *	Theodore Ts'o, 2002
  */
 
 #include <linux/fs.h>
@@ -278,7 +278,7 @@ static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext3_dir_ent
 				       ((char *) de - base));
 			}
 			space += EXT3_DIR_REC_LEN(de->name_len);
-	 		names++;
+			names++;
 		}
 		de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len));
 	}
@@ -1688,7 +1688,7 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry,
 
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
-			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
+					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
 					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
@@ -1816,7 +1816,7 @@ static int empty_dir (struct inode * inode)
 			!le32_to_cpu(de1->inode) ||
 			strcmp (".", de->name) ||
 			strcmp ("..", de1->name)) {
-	    	ext3_warning (inode->i_sb, "empty_dir",
+		ext3_warning (inode->i_sb, "empty_dir",
 			      "bad directory (dir #%lu) - no `.' or `..'",
 			      inode->i_ino);
 		brelse (bh);
@@ -2129,7 +2129,7 @@ static int ext3_symlink (struct inode * dir,
 
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
-			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
+					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
 					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
@@ -2227,7 +2227,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry,
 		DQUOT_INIT(new_dentry->d_inode);
 	handle = ext3_journal_start(old_dir, 2 *
 					EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) +
-			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2);
+					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 4b526b496102..0544325d9d7e 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -734,8 +734,8 @@ static match_table_t tokens = {
 
 static ext3_fsblk_t get_sb_block(void **data)
 {
-	ext3_fsblk_t 	sb_block;
-	char 		*options = (char *) *data;
+	ext3_fsblk_t	sb_block;
+	char		*options = (char *) *data;
 
 	if (!options || strncmp(options, "sb=", 3) != 0)
 		return 1;	/* Default location */
@@ -2740,7 +2740,7 @@ static int __init init_ext3_fs(void)
 out:
 	destroy_inodecache();
 out1:
- 	exit_ext3_xattr();
+	exit_ext3_xattr();
 	return err;
 }
 
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 961ada28db5e..0208cc7ac5d0 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -480,7 +480,7 @@ static int journal_clean_one_cp_list(struct journal_head *jh, int *released)
 	if (!jh)
 		return 0;
 
- 	last_jh = jh->b_cpprev;
+	last_jh = jh->b_cpprev;
 	do {
 		jh = next_jh;
 		next_jh = jh->b_cpnext;
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 88d970e09ce6..b571209d6a1e 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -181,7 +181,7 @@ loop:
 						transaction->t_expires))
 			should_sleep = 0;
 		if (journal->j_flags & JFS_UNMOUNT)
- 			should_sleep = 0;
+			should_sleep = 0;
 		if (should_sleep) {
 			spin_unlock(&journal->j_state_lock);
 			schedule();
diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
index 73bb64806ed3..445eed6ce5dc 100644
--- a/fs/jbd/recovery.c
+++ b/fs/jbd/recovery.c
@@ -314,7 +314,7 @@ static int do_one_pass(journal_t *journal,
 	unsigned long		next_log_block;
 	int			err, success = 0;
 	journal_superblock_t *	sb;
-	journal_header_t * 	tmp;
+	journal_header_t *	tmp;
 	struct buffer_head *	bh;
 	unsigned int		sequence;
 	int			blocktype;
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index bf7fd7117817..e1b3c8af4d17 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -580,7 +580,7 @@ repeat:
 		 */
 		JBUFFER_TRACE(jh, "Unexpected dirty buffer");
 		jbd_unexpected_dirty_buffer(jh);
- 	}
+	}
 
 	unlock_buffer(bh);
 
@@ -1373,7 +1373,7 @@ int journal_stop(handle_t *handle)
 	if (handle->h_sync ||
 			transaction->t_outstanding_credits >
 				journal->j_max_transaction_buffers ||
-	    		time_after_eq(jiffies, transaction->t_expires)) {
+			time_after_eq(jiffies, transaction->t_expires)) {
 		/* Do this even for aborted journals: an abort still
 		 * completes the commit thread, it just doesn't write
 		 * anything to disk. */
@@ -1908,7 +1908,7 @@ void journal_invalidatepage(journal_t *journal,
 		next = bh->b_this_page;
 
 		if (offset <= curr_off) {
-		 	/* This block is wholly outside the truncation point */
+			/* This block is wholly outside the truncation point */
 			lock_buffer(bh);
 			may_free &= journal_unmap_buffer(journal, bh);
 			unlock_buffer(bh);
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 0eed918b3816..369c67502346 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -473,7 +473,7 @@ struct ext3_super_block {
 	__u8	s_reserved_char_pad;
 	__u16	s_reserved_word_pad;
 	__le32	s_default_mount_opts;
-	__le32	s_first_meta_bg; 	/* First metablock block group */
+	__le32	s_first_meta_bg;	/* First metablock block group */
 	__u32	s_reserved[190];	/* Padding to the end of the block */
 };
 
diff --git a/include/linux/ext3_fs_i.h b/include/linux/ext3_fs_i.h
index 2f18b9511f21..4395e5206746 100644
--- a/include/linux/ext3_fs_i.h
+++ b/include/linux/ext3_fs_i.h
@@ -35,7 +35,7 @@ struct ext3_reserve_window {
 };
 
 struct ext3_reserve_window_node {
-	struct rb_node	 	rsv_node;
+	struct rb_node		rsv_node;
 	__u32			rsv_goal_size;
 	__u32			rsv_alloc_hit;
 	struct ext3_reserve_window	rsv_window;
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index dc262624efa5..a6d9daa38c6d 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -64,7 +64,7 @@ extern int journal_enable_debug;
 		if ((n) <= journal_enable_debug) {			\
 			printk (KERN_DEBUG "(%s, %d): %s: ",		\
 				__FILE__, __LINE__, __FUNCTION__);	\
-		  	printk (f, ## a);				\
+			printk (f, ## a);				\
 		}							\
 	} while (0)
 #else
@@ -201,9 +201,9 @@ typedef struct journal_superblock_s
 
 /* 0x0024 */
 	/* Remaining fields are only valid in a version-2 superblock */
-	__be32	s_feature_compat; 	/* compatible feature set */
-	__be32	s_feature_incompat; 	/* incompatible feature set */
-	__be32	s_feature_ro_compat; 	/* readonly-compatible feature set */
+	__be32	s_feature_compat;	/* compatible feature set */
+	__be32	s_feature_incompat;	/* incompatible feature set */
+	__be32	s_feature_ro_compat;	/* readonly-compatible feature set */
 /* 0x0030 */
 	__u8	s_uuid[16];		/* 128-bit uuid for journal */
 
@@ -699,7 +699,7 @@ struct journal_s
 	wait_queue_head_t	j_wait_updates;
 
 	/* Semaphore for locking against concurrent checkpoints */
-	struct mutex	 	j_checkpoint_mutex;
+	struct mutex		j_checkpoint_mutex;
 
 	/*
 	 * Journal head: identifies the first unused block in the journal.
-- 
cgit v1.2.3


From a4e4de36dc446b2193bdc8ebb96a96e44b69dd94 Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@austin.ibm.com>
Date: Wed, 27 Sep 2006 01:49:36 -0700
Subject: [PATCH] ext3: Fix sparse warnings

Fixing up some endian-ness warnings in preparation to clone ext4 from ext3.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/resize.c        | 21 +++++++++++----------
 fs/ext3/super.c         |  4 +---
 fs/jbd/journal.c        |  2 +-
 include/linux/ext3_fs.h |  2 +-
 4 files changed, 14 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index 5e1337fd878a..e186f7fb698b 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -336,7 +336,7 @@ static int verify_reserved_gdb(struct super_block *sb,
 	unsigned five = 5;
 	unsigned seven = 7;
 	unsigned grp;
-	__u32 *p = (__u32 *)primary->b_data;
+	__le32 *p = (__le32 *)primary->b_data;
 	int gdbackups = 0;
 
 	while ((grp = ext3_list_backups(sb, &three, &five, &seven)) < end) {
@@ -380,7 +380,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	struct buffer_head *dind;
 	int gdbackups;
 	struct ext3_iloc iloc;
-	__u32 *data;
+	__le32 *data;
 	int err;
 
 	if (test_opt(sb, DEBUG))
@@ -417,7 +417,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 		goto exit_bh;
 	}
 
-	data = (__u32 *)dind->b_data;
+	data = (__le32 *)dind->b_data;
 	if (le32_to_cpu(data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)]) != gdblock) {
 		ext3_warning(sb, __FUNCTION__,
 			     "new group %u GDT block "E3FSBLK" not reserved",
@@ -519,7 +519,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	struct buffer_head *dind;
 	struct ext3_iloc iloc;
 	ext3_fsblk_t blk;
-	__u32 *data, *end;
+	__le32 *data, *end;
 	int gdbackups = 0;
 	int res, i;
 	int err;
@@ -536,8 +536,8 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	}
 
 	blk = EXT3_SB(sb)->s_sbh->b_blocknr + 1 + EXT3_SB(sb)->s_gdb_count;
-	data = (__u32 *)dind->b_data + EXT3_SB(sb)->s_gdb_count;
-	end = (__u32 *)dind->b_data + EXT3_ADDR_PER_BLOCK(sb);
+	data = (__le32 *)dind->b_data + EXT3_SB(sb)->s_gdb_count;
+	end = (__le32 *)dind->b_data + EXT3_ADDR_PER_BLOCK(sb);
 
 	/* Get each reserved primary GDT block and verify it holds backups */
 	for (res = 0; res < reserved_gdb; res++, blk++) {
@@ -545,7 +545,8 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 			ext3_warning(sb, __FUNCTION__,
 				     "reserved block "E3FSBLK
 				     " not at offset %ld",
-				     blk, (long)(data - (__u32 *)dind->b_data));
+				     blk,
+				     (long)(data - (__le32 *)dind->b_data));
 			err = -EINVAL;
 			goto exit_bh;
 		}
@@ -560,7 +561,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 			goto exit_bh;
 		}
 		if (++data >= end)
-			data = (__u32 *)dind->b_data;
+			data = (__le32 *)dind->b_data;
 	}
 
 	for (i = 0; i < reserved_gdb; i++) {
@@ -584,7 +585,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	blk = input->group * EXT3_BLOCKS_PER_GROUP(sb);
 	for (i = 0; i < reserved_gdb; i++) {
 		int err2;
-		data = (__u32 *)primary[i]->b_data;
+		data = (__le32 *)primary[i]->b_data;
 		/* printk("reserving backup %lu[%u] = %lu\n",
 		       primary[i]->b_blocknr, gdbackups,
 		       blk + primary[i]->b_blocknr); */
@@ -689,7 +690,7 @@ exit_err:
 			     "can't update backup for group %d (err %d), "
 			     "forcing fsck on next reboot", group, err);
 		sbi->s_mount_state &= ~EXT3_VALID_FS;
-		sbi->s_es->s_state &= ~cpu_to_le16(EXT3_VALID_FS);
+		sbi->s_es->s_state &= cpu_to_le16(~EXT3_VALID_FS);
 		mark_buffer_dirty(sbi->s_sbh);
 	}
 }
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 0544325d9d7e..10985017765b 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -2348,10 +2348,8 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data)
 			 */
 			ext3_clear_journal_err(sb, es);
 			sbi->s_mount_state = le16_to_cpu(es->s_state);
-			if ((ret = ext3_group_extend(sb, es, n_blocks_count))) {
-				err = ret;
+			if ((err = ext3_group_extend(sb, es, n_blocks_count)))
 				goto restore_opts;
-			}
 			if (!ext3_setup_super (sb, es, 0))
 				sb->s_flags &= ~MS_RDONLY;
 		}
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index b571209d6a1e..2fc66c3e6681 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -1094,7 +1094,7 @@ int journal_load(journal_t *journal)
 	/*
 	 * Create a slab for this blocksize
 	 */
-	err = journal_create_jbd_slab(cpu_to_be32(sb->s_blocksize));
+	err = journal_create_jbd_slab(be32_to_cpu(sb->s_blocksize));
 	if (err)
 		return err;
 
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 369c67502346..cc08f56750da 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -460,7 +460,7 @@ struct ext3_super_block {
 	 */
 	__u8	s_prealloc_blocks;	/* Nr of blocks to try to preallocate*/
 	__u8	s_prealloc_dir_blocks;	/* Nr to preallocate for dirs */
-	__u16	s_reserved_gdt_blocks;	/* Per group desc for online growth */
+	__le16	s_reserved_gdt_blocks;	/* Per group desc for online growth */
 	/*
 	 * Journaling support valid if EXT3_FEATURE_COMPAT_HAS_JOURNAL set.
 	 */
-- 
cgit v1.2.3


From 133d205a18b7a4d8cb52959c5310f6664277cf61 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Wed, 27 Sep 2006 01:49:41 -0700
Subject: [PATCH] Make kmem_cache_destroy() return void

un-, de-, -free, -destroy, -exit, etc functions should in general return
void.  Also,

There is very little, say, filesystem driver code can do upon failed
kmem_cache_destroy().  If it will be decided to BUG in this case, BUG
should be put in generic code, instead.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/slab.h | 4 ++--
 mm/slab.c            | 6 ++----
 mm/slob.c            | 3 +--
 3 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 66d6eb78d1c6..a96fd9310d55 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -60,7 +60,7 @@ extern void __init kmem_cache_init(void);
 extern kmem_cache_t *kmem_cache_create(const char *, size_t, size_t, unsigned long,
 				       void (*)(void *, kmem_cache_t *, unsigned long),
 				       void (*)(void *, kmem_cache_t *, unsigned long));
-extern int kmem_cache_destroy(kmem_cache_t *);
+extern void kmem_cache_destroy(kmem_cache_t *);
 extern int kmem_cache_shrink(kmem_cache_t *);
 extern void *kmem_cache_alloc(kmem_cache_t *, gfp_t);
 extern void *kmem_cache_zalloc(struct kmem_cache *, gfp_t);
@@ -249,7 +249,7 @@ struct kmem_cache *kmem_cache_create(const char *c, size_t, size_t,
 	unsigned long,
 	void (*)(void *, struct kmem_cache *, unsigned long),
 	void (*)(void *, struct kmem_cache *, unsigned long));
-int kmem_cache_destroy(struct kmem_cache *c);
+void kmem_cache_destroy(struct kmem_cache *c);
 void *kmem_cache_alloc(struct kmem_cache *c, gfp_t flags);
 void *kmem_cache_zalloc(struct kmem_cache *, gfp_t);
 void kmem_cache_free(struct kmem_cache *c, void *b);
diff --git a/mm/slab.c b/mm/slab.c
index 7a48eb1a60c8..c52ebf9c4462 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2442,7 +2442,6 @@ EXPORT_SYMBOL(kmem_cache_shrink);
  * @cachep: the cache to destroy
  *
  * Remove a struct kmem_cache object from the slab cache.
- * Returns 0 on success.
  *
  * It is expected this function will be called by a module when it is
  * unloaded.  This will remove the cache completely, and avoid a duplicate
@@ -2454,7 +2453,7 @@ EXPORT_SYMBOL(kmem_cache_shrink);
  * The caller must guarantee that noone will allocate memory from the cache
  * during the kmem_cache_destroy().
  */
-int kmem_cache_destroy(struct kmem_cache *cachep)
+void kmem_cache_destroy(struct kmem_cache *cachep)
 {
 	BUG_ON(!cachep || in_interrupt());
 
@@ -2475,7 +2474,7 @@ int kmem_cache_destroy(struct kmem_cache *cachep)
 		list_add(&cachep->next, &cache_chain);
 		mutex_unlock(&cache_chain_mutex);
 		unlock_cpu_hotplug();
-		return 1;
+		return;
 	}
 
 	if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU))
@@ -2483,7 +2482,6 @@ int kmem_cache_destroy(struct kmem_cache *cachep)
 
 	__kmem_cache_destroy(cachep);
 	unlock_cpu_hotplug();
-	return 0;
 }
 EXPORT_SYMBOL(kmem_cache_destroy);
 
diff --git a/mm/slob.c b/mm/slob.c
index 20188627347c..542394184a58 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -270,10 +270,9 @@ struct kmem_cache *kmem_cache_create(const char *name, size_t size,
 }
 EXPORT_SYMBOL(kmem_cache_create);
 
-int kmem_cache_destroy(struct kmem_cache *c)
+void kmem_cache_destroy(struct kmem_cache *c)
 {
 	slob_free(c, sizeof(struct kmem_cache));
-	return 0;
 }
 EXPORT_SYMBOL(kmem_cache_destroy);
 
-- 
cgit v1.2.3


From 2bd0cfbde2c0a74e209acbf045f1298cc2f61e01 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:49:42 -0700
Subject: [PATCH] fix x86_64-mm-spinlock-cleanup

We need processor.h for cpu_relax().

Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-x86_64/spinlock.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include')

diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index be7a9e629fb2..3daf5b005905 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -4,6 +4,7 @@
 #include <asm/atomic.h>
 #include <asm/rwlock.h>
 #include <asm/page.h>
+#include <asm/processor.h>
 
 /*
  * Your basic SMP spinlocks, allowing only a single CPU anywhere
-- 
cgit v1.2.3


From c713216deebd95d2b0ab38fef8bb2361c0180c2d Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:43 -0700
Subject: [PATCH] Introduce mechanism for registering active regions of memory

At a basic level, architectures define structures to record where active
ranges of page frames are located.  Once located, the code to calculate zone
sizes and holes in each architecture is very similar.  Some of this zone and
hole sizing code is difficult to read for no good reason.  This set of patches
eliminates the similar-looking architecture-specific code.

The patches introduce a mechanism where architectures register where the
active ranges of page frames are with add_active_range().  When all areas have
been discovered, free_area_init_nodes() is called to initialise the pgdat and
zones.  The zone sizes and holes are then calculated in an architecture
independent manner.

Patch 1 introduces the mechanism for registering and initialising PFN ranges
Patch 2 changes ppc to use the mechanism - 139 arch-specific LOC removed
Patch 3 changes x86 to use the mechanism - 136 arch-specific LOC removed
Patch 4 changes x86_64 to use the mechanism - 74 arch-specific LOC removed
Patch 5 changes ia64 to use the mechanism - 52 arch-specific LOC removed
Patch 6 accounts for mem_map as a memory hole as the pages are not reclaimable.
	It adjusts the watermarks slightly

Tony Luck has successfully tested for ia64 on Itanium with tiger_defconfig,
gensparse_defconfig and defconfig.  Bob Picco has also tested and debugged on
IA64.  Jack Steiner successfully boot tested on a mammoth SGI IA64-based
machine.  These were on patches against 2.6.17-rc1 and release 3 of these
patches but there have been no ia64-changes since release 3.

There are differences in the zone sizes for x86_64 as the arch-specific code
for x86_64 accounts the kernel image and the starting mem_maps as memory holes
but the architecture-independent code accounts the memory as present.

The big benefit of this set of patches is a sizable reduction of
architecture-specific code, some of which is very hairy.  There should be a
greater reduction when other architectures use the same mechanisms for zone
and hole sizing but I lack the hardware to test on.

Additional credit;
	Dave Hansen for the initial suggestion and comments on early patches
	Andy Whitcroft for reviewing early versions and catching numerous
		errors
	Tony Luck for testing and debugging on IA64
	Bob Picco for fixing bugs related to pfn registration, reviewing a
		number of patch revisions, providing a number of suggestions
		on future direction and testing heavily
	Jack Steiner and Robin Holt for testing on IA64 and clarifying
		issues related to memory holes
	Yasunori for testing on IA64
	Andi Kleen for reviewing and feeding back about x86_64
	Christian Kujau for providing valuable information related to ACPI
		problems on x86_64 and testing potential fixes

This patch:

Define the structure to represent an active range of page frames within a node
in an architecture independent manner.  Architectures are expected to register
active ranges of PFNs using add_active_range(nid, start_pfn, end_pfn) and call
free_area_init_nodes() passing the PFNs of the end of each zone.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     |  47 +++++
 include/linux/mmzone.h |  10 +-
 mm/page_alloc.c        | 552 ++++++++++++++++++++++++++++++++++++++++++++++---
 3 files changed, 584 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 856f0ee7e84a..c0402da7cce0 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -937,6 +937,53 @@ extern void free_area_init(unsigned long * zones_size);
 extern void free_area_init_node(int nid, pg_data_t *pgdat,
 	unsigned long * zones_size, unsigned long zone_start_pfn, 
 	unsigned long *zholes_size);
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+/*
+ * With CONFIG_ARCH_POPULATES_NODE_MAP set, an architecture may initialise its
+ * zones, allocate the backing mem_map and account for memory holes in a more
+ * architecture independent manner. This is a substitute for creating the
+ * zone_sizes[] and zholes_size[] arrays and passing them to
+ * free_area_init_node()
+ *
+ * An architecture is expected to register range of page frames backed by
+ * physical memory with add_active_range() before calling
+ * free_area_init_nodes() passing in the PFN each zone ends at. At a basic
+ * usage, an architecture is expected to do something like
+ *
+ * unsigned long max_zone_pfns[MAX_NR_ZONES] = {max_dma, max_normal_pfn,
+ * 							 max_highmem_pfn};
+ * for_each_valid_physical_page_range()
+ * 	add_active_range(node_id, start_pfn, end_pfn)
+ * free_area_init_nodes(max_zone_pfns);
+ *
+ * If the architecture guarantees that there are no holes in the ranges
+ * registered with add_active_range(), free_bootmem_active_regions()
+ * will call free_bootmem_node() for each registered physical page range.
+ * Similarly sparse_memory_present_with_active_regions() calls
+ * memory_present() for each range when SPARSEMEM is enabled.
+ *
+ * See mm/page_alloc.c for more information on each function exposed by
+ * CONFIG_ARCH_POPULATES_NODE_MAP
+ */
+extern void free_area_init_nodes(unsigned long *max_zone_pfn);
+extern void add_active_range(unsigned int nid, unsigned long start_pfn,
+					unsigned long end_pfn);
+extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
+						unsigned long new_end_pfn);
+extern void remove_all_active_ranges(void);
+extern unsigned long absent_pages_in_range(unsigned long start_pfn,
+						unsigned long end_pfn);
+extern void get_pfn_range_for_nid(unsigned int nid,
+			unsigned long *start_pfn, unsigned long *end_pfn);
+extern unsigned long find_min_pfn_with_active_regions(void);
+extern unsigned long find_max_pfn_with_active_regions(void);
+extern void free_bootmem_with_active_regions(int nid,
+						unsigned long max_low_pfn);
+extern void sparse_memory_present_with_active_regions(int nid);
+#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+extern int early_pfn_to_nid(unsigned long pfn);
+#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long);
 extern void setup_per_zone_pages_min(void);
 extern void mem_init(void);
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 3693f1a52788..7fa1cbe9fa7a 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -305,6 +305,13 @@ struct zonelist {
 	struct zone *zones[MAX_NUMNODES * MAX_NR_ZONES + 1]; // NULL delimited
 };
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+struct node_active_region {
+	unsigned long start_pfn;
+	unsigned long end_pfn;
+	int nid;
+};
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
 /*
  * The pg_data_t structure is used in machines with CONFIG_DISCONTIGMEM
@@ -518,7 +525,8 @@ extern struct zone *next_zone(struct zone *zone);
 
 #endif
 
-#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+#if !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID) && \
+	!defined(CONFIG_ARCH_POPULATES_NODE_MAP)
 #define early_pfn_to_nid(nid)  (0UL)
 #endif
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9810f0a60db7..26c9939857fa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -37,6 +37,8 @@
 #include <linux/vmalloc.h>
 #include <linux/mempolicy.h>
 #include <linux/stop_machine.h>
+#include <linux/sort.h>
+#include <linux/pfn.h>
 
 #include <asm/tlbflush.h>
 #include <asm/div64.h>
@@ -103,6 +105,33 @@ int min_free_kbytes = 1024;
 unsigned long __meminitdata nr_kernel_pages;
 unsigned long __meminitdata nr_all_pages;
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+  /*
+   * MAX_ACTIVE_REGIONS determines the maxmimum number of distinct
+   * ranges of memory (RAM) that may be registered with add_active_range().
+   * Ranges passed to add_active_range() will be merged if possible
+   * so the number of times add_active_range() can be called is
+   * related to the number of nodes and the number of holes
+   */
+  #ifdef CONFIG_MAX_ACTIVE_REGIONS
+    /* Allow an architecture to set MAX_ACTIVE_REGIONS to save memory */
+    #define MAX_ACTIVE_REGIONS CONFIG_MAX_ACTIVE_REGIONS
+  #else
+    #if MAX_NUMNODES >= 32
+      /* If there can be many nodes, allow up to 50 holes per node */
+      #define MAX_ACTIVE_REGIONS (MAX_NUMNODES*50)
+    #else
+      /* By default, allow up to 256 distinct regions */
+      #define MAX_ACTIVE_REGIONS 256
+    #endif
+  #endif
+
+  struct node_active_region __initdata early_node_map[MAX_ACTIVE_REGIONS];
+  int __initdata nr_nodemap_entries;
+  unsigned long __initdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES];
+  unsigned long __initdata arch_zone_highest_possible_pfn[MAX_NR_ZONES];
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+
 #ifdef CONFIG_DEBUG_VM
 static int page_outside_zone_boundaries(struct zone *zone, struct page *page)
 {
@@ -1642,25 +1671,6 @@ static inline unsigned long wait_table_bits(unsigned long size)
 
 #define LONG_ALIGN(x) (((x)+(sizeof(long))-1)&~((sizeof(long))-1))
 
-static void __init calculate_zone_totalpages(struct pglist_data *pgdat,
-		unsigned long *zones_size, unsigned long *zholes_size)
-{
-	unsigned long realtotalpages, totalpages = 0;
-	enum zone_type i;
-
-	for (i = 0; i < MAX_NR_ZONES; i++)
-		totalpages += zones_size[i];
-	pgdat->node_spanned_pages = totalpages;
-
-	realtotalpages = totalpages;
-	if (zholes_size)
-		for (i = 0; i < MAX_NR_ZONES; i++)
-			realtotalpages -= zholes_size[i];
-	pgdat->node_present_pages = realtotalpages;
-	printk(KERN_DEBUG "On node %d totalpages: %lu\n", pgdat->node_id, realtotalpages);
-}
-
-
 /*
  * Initially all pages are reserved - free ones are freed
  * up by free_all_bootmem() once the early boot process is
@@ -1977,6 +1987,272 @@ __meminit int init_currently_empty_zone(struct zone *zone,
 	return 0;
 }
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+/*
+ * Basic iterator support. Return the first range of PFNs for a node
+ * Note: nid == MAX_NUMNODES returns first region regardless of node
+ */
+static int __init first_active_region_index_in_nid(int nid)
+{
+	int i;
+
+	for (i = 0; i < nr_nodemap_entries; i++)
+		if (nid == MAX_NUMNODES || early_node_map[i].nid == nid)
+			return i;
+
+	return -1;
+}
+
+/*
+ * Basic iterator support. Return the next active range of PFNs for a node
+ * Note: nid == MAX_NUMNODES returns next region regardles of node
+ */
+static int __init next_active_region_index_in_nid(int index, int nid)
+{
+	for (index = index + 1; index < nr_nodemap_entries; index++)
+		if (nid == MAX_NUMNODES || early_node_map[index].nid == nid)
+			return index;
+
+	return -1;
+}
+
+#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+/*
+ * Required by SPARSEMEM. Given a PFN, return what node the PFN is on.
+ * Architectures may implement their own version but if add_active_range()
+ * was used and there are no special requirements, this is a convenient
+ * alternative
+ */
+int __init early_pfn_to_nid(unsigned long pfn)
+{
+	int i;
+
+	for (i = 0; i < nr_nodemap_entries; i++) {
+		unsigned long start_pfn = early_node_map[i].start_pfn;
+		unsigned long end_pfn = early_node_map[i].end_pfn;
+
+		if (start_pfn <= pfn && pfn < end_pfn)
+			return early_node_map[i].nid;
+	}
+
+	return 0;
+}
+#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
+
+/* Basic iterator support to walk early_node_map[] */
+#define for_each_active_range_index_in_nid(i, nid) \
+	for (i = first_active_region_index_in_nid(nid); i != -1; \
+				i = next_active_region_index_in_nid(i, nid))
+
+/**
+ * free_bootmem_with_active_regions - Call free_bootmem_node for each active range
+ * @nid: The node to free memory on. If MAX_NUMNODES, all nodes are freed
+ * @max_low_pfn: The highest PFN that till be passed to free_bootmem_node
+ *
+ * If an architecture guarantees that all ranges registered with
+ * add_active_ranges() contain no holes and may be freed, this
+ * this function may be used instead of calling free_bootmem() manually.
+ */
+void __init free_bootmem_with_active_regions(int nid,
+						unsigned long max_low_pfn)
+{
+	int i;
+
+	for_each_active_range_index_in_nid(i, nid) {
+		unsigned long size_pages = 0;
+		unsigned long end_pfn = early_node_map[i].end_pfn;
+
+		if (early_node_map[i].start_pfn >= max_low_pfn)
+			continue;
+
+		if (end_pfn > max_low_pfn)
+			end_pfn = max_low_pfn;
+
+		size_pages = end_pfn - early_node_map[i].start_pfn;
+		free_bootmem_node(NODE_DATA(early_node_map[i].nid),
+				PFN_PHYS(early_node_map[i].start_pfn),
+				size_pages << PAGE_SHIFT);
+	}
+}
+
+/**
+ * sparse_memory_present_with_active_regions - Call memory_present for each active range
+ * @nid: The node to call memory_present for. If MAX_NUMNODES, all nodes will be used
+ *
+ * If an architecture guarantees that all ranges registered with
+ * add_active_ranges() contain no holes and may be freed, this
+ * this function may be used instead of calling memory_present() manually.
+ */
+void __init sparse_memory_present_with_active_regions(int nid)
+{
+	int i;
+
+	for_each_active_range_index_in_nid(i, nid)
+		memory_present(early_node_map[i].nid,
+				early_node_map[i].start_pfn,
+				early_node_map[i].end_pfn);
+}
+
+/**
+ * get_pfn_range_for_nid - Return the start and end page frames for a node
+ * @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned
+ * @start_pfn: Passed by reference. On return, it will have the node start_pfn
+ * @end_pfn: Passed by reference. On return, it will have the node end_pfn
+ *
+ * It returns the start and end page frame of a node based on information
+ * provided by an arch calling add_active_range(). If called for a node
+ * with no available memory, a warning is printed and the start and end
+ * PFNs will be 0
+ */
+void __init get_pfn_range_for_nid(unsigned int nid,
+			unsigned long *start_pfn, unsigned long *end_pfn)
+{
+	int i;
+	*start_pfn = -1UL;
+	*end_pfn = 0;
+
+	for_each_active_range_index_in_nid(i, nid) {
+		*start_pfn = min(*start_pfn, early_node_map[i].start_pfn);
+		*end_pfn = max(*end_pfn, early_node_map[i].end_pfn);
+	}
+
+	if (*start_pfn == -1UL) {
+		printk(KERN_WARNING "Node %u active with no memory\n", nid);
+		*start_pfn = 0;
+	}
+}
+
+/*
+ * Return the number of pages a zone spans in a node, including holes
+ * present_pages = zone_spanned_pages_in_node() - zone_absent_pages_in_node()
+ */
+unsigned long __init zone_spanned_pages_in_node(int nid,
+					unsigned long zone_type,
+					unsigned long *ignored)
+{
+	unsigned long node_start_pfn, node_end_pfn;
+	unsigned long zone_start_pfn, zone_end_pfn;
+
+	/* Get the start and end of the node and zone */
+	get_pfn_range_for_nid(nid, &node_start_pfn, &node_end_pfn);
+	zone_start_pfn = arch_zone_lowest_possible_pfn[zone_type];
+	zone_end_pfn = arch_zone_highest_possible_pfn[zone_type];
+
+	/* Check that this node has pages within the zone's required range */
+	if (zone_end_pfn < node_start_pfn || zone_start_pfn > node_end_pfn)
+		return 0;
+
+	/* Move the zone boundaries inside the node if necessary */
+	zone_end_pfn = min(zone_end_pfn, node_end_pfn);
+	zone_start_pfn = max(zone_start_pfn, node_start_pfn);
+
+	/* Return the spanned pages */
+	return zone_end_pfn - zone_start_pfn;
+}
+
+/*
+ * Return the number of holes in a range on a node. If nid is MAX_NUMNODES,
+ * then all holes in the requested range will be accounted for
+ */
+unsigned long __init __absent_pages_in_range(int nid,
+				unsigned long range_start_pfn,
+				unsigned long range_end_pfn)
+{
+	int i = 0;
+	unsigned long prev_end_pfn = 0, hole_pages = 0;
+	unsigned long start_pfn;
+
+	/* Find the end_pfn of the first active range of pfns in the node */
+	i = first_active_region_index_in_nid(nid);
+	if (i == -1)
+		return 0;
+
+	prev_end_pfn = early_node_map[i].start_pfn;
+
+	/* Find all holes for the zone within the node */
+	for (; i != -1; i = next_active_region_index_in_nid(i, nid)) {
+
+		/* No need to continue if prev_end_pfn is outside the zone */
+		if (prev_end_pfn >= range_end_pfn)
+			break;
+
+		/* Make sure the end of the zone is not within the hole */
+		start_pfn = min(early_node_map[i].start_pfn, range_end_pfn);
+		prev_end_pfn = max(prev_end_pfn, range_start_pfn);
+
+		/* Update the hole size cound and move on */
+		if (start_pfn > range_start_pfn) {
+			BUG_ON(prev_end_pfn > start_pfn);
+			hole_pages += start_pfn - prev_end_pfn;
+		}
+		prev_end_pfn = early_node_map[i].end_pfn;
+	}
+
+	return hole_pages;
+}
+
+/**
+ * absent_pages_in_range - Return number of page frames in holes within a range
+ * @start_pfn: The start PFN to start searching for holes
+ * @end_pfn: The end PFN to stop searching for holes
+ *
+ * It returns the number of pages frames in memory holes within a range
+ */
+unsigned long __init absent_pages_in_range(unsigned long start_pfn,
+							unsigned long end_pfn)
+{
+	return __absent_pages_in_range(MAX_NUMNODES, start_pfn, end_pfn);
+}
+
+/* Return the number of page frames in holes in a zone on a node */
+unsigned long __init zone_absent_pages_in_node(int nid,
+					unsigned long zone_type,
+					unsigned long *ignored)
+{
+	return __absent_pages_in_range(nid,
+				arch_zone_lowest_possible_pfn[zone_type],
+				arch_zone_highest_possible_pfn[zone_type]);
+}
+#else
+static inline unsigned long zone_spanned_pages_in_node(int nid,
+					unsigned long zone_type,
+					unsigned long *zones_size)
+{
+	return zones_size[zone_type];
+}
+
+static inline unsigned long zone_absent_pages_in_node(int nid,
+						unsigned long zone_type,
+						unsigned long *zholes_size)
+{
+	if (!zholes_size)
+		return 0;
+
+	return zholes_size[zone_type];
+}
+#endif
+
+static void __init calculate_node_totalpages(struct pglist_data *pgdat,
+		unsigned long *zones_size, unsigned long *zholes_size)
+{
+	unsigned long realtotalpages, totalpages = 0;
+	enum zone_type i;
+
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		totalpages += zone_spanned_pages_in_node(pgdat->node_id, i,
+								zones_size);
+	pgdat->node_spanned_pages = totalpages;
+
+	realtotalpages = totalpages;
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		realtotalpages -=
+			zone_absent_pages_in_node(pgdat->node_id, i,
+								zholes_size);
+	pgdat->node_present_pages = realtotalpages;
+	printk(KERN_DEBUG "On node %d totalpages: %lu\n", pgdat->node_id,
+							realtotalpages);
+}
+
 /*
  * Set up the zone data structures:
  *   - mark all pages reserved
@@ -2000,9 +2276,9 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		struct zone *zone = pgdat->node_zones + j;
 		unsigned long size, realsize;
 
-		realsize = size = zones_size[j];
-		if (zholes_size)
-			realsize -= zholes_size[j];
+		size = zone_spanned_pages_in_node(nid, j, zones_size);
+		realsize = size - zone_absent_pages_in_node(nid, j,
+								zholes_size);
 
 		if (!is_highmem_idx(j))
 			nr_kernel_pages += realsize;
@@ -2073,8 +2349,13 @@ static void __init alloc_node_mem_map(struct pglist_data *pgdat)
 	/*
 	 * With no DISCONTIG, the global mem_map is just set as node 0's
 	 */
-	if (pgdat == NODE_DATA(0))
+	if (pgdat == NODE_DATA(0)) {
 		mem_map = NODE_DATA(0)->node_mem_map;
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+		if (page_to_pfn(mem_map) != pgdat->node_start_pfn)
+			mem_map -= pgdat->node_start_pfn;
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+	}
 #endif
 #endif /* CONFIG_FLAT_NODE_MEM_MAP */
 }
@@ -2085,13 +2366,236 @@ void __meminit free_area_init_node(int nid, struct pglist_data *pgdat,
 {
 	pgdat->node_id = nid;
 	pgdat->node_start_pfn = node_start_pfn;
-	calculate_zone_totalpages(pgdat, zones_size, zholes_size);
+	calculate_node_totalpages(pgdat, zones_size, zholes_size);
 
 	alloc_node_mem_map(pgdat);
 
 	free_area_init_core(pgdat, zones_size, zholes_size);
 }
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+/**
+ * add_active_range - Register a range of PFNs backed by physical memory
+ * @nid: The node ID the range resides on
+ * @start_pfn: The start PFN of the available physical memory
+ * @end_pfn: The end PFN of the available physical memory
+ *
+ * These ranges are stored in an early_node_map[] and later used by
+ * free_area_init_nodes() to calculate zone sizes and holes. If the
+ * range spans a memory hole, it is up to the architecture to ensure
+ * the memory is not freed by the bootmem allocator. If possible
+ * the range being registered will be merged with existing ranges.
+ */
+void __init add_active_range(unsigned int nid, unsigned long start_pfn,
+						unsigned long end_pfn)
+{
+	int i;
+
+	printk(KERN_DEBUG "Entering add_active_range(%d, %lu, %lu) "
+			  "%d entries of %d used\n",
+			  nid, start_pfn, end_pfn,
+			  nr_nodemap_entries, MAX_ACTIVE_REGIONS);
+
+	/* Merge with existing active regions if possible */
+	for (i = 0; i < nr_nodemap_entries; i++) {
+		if (early_node_map[i].nid != nid)
+			continue;
+
+		/* Skip if an existing region covers this new one */
+		if (start_pfn >= early_node_map[i].start_pfn &&
+				end_pfn <= early_node_map[i].end_pfn)
+			return;
+
+		/* Merge forward if suitable */
+		if (start_pfn <= early_node_map[i].end_pfn &&
+				end_pfn > early_node_map[i].end_pfn) {
+			early_node_map[i].end_pfn = end_pfn;
+			return;
+		}
+
+		/* Merge backward if suitable */
+		if (start_pfn < early_node_map[i].end_pfn &&
+				end_pfn >= early_node_map[i].start_pfn) {
+			early_node_map[i].start_pfn = start_pfn;
+			return;
+		}
+	}
+
+	/* Check that early_node_map is large enough */
+	if (i >= MAX_ACTIVE_REGIONS) {
+		printk(KERN_CRIT "More than %d memory regions, truncating\n",
+							MAX_ACTIVE_REGIONS);
+		return;
+	}
+
+	early_node_map[i].nid = nid;
+	early_node_map[i].start_pfn = start_pfn;
+	early_node_map[i].end_pfn = end_pfn;
+	nr_nodemap_entries = i + 1;
+}
+
+/**
+ * shrink_active_range - Shrink an existing registered range of PFNs
+ * @nid: The node id the range is on that should be shrunk
+ * @old_end_pfn: The old end PFN of the range
+ * @new_end_pfn: The new PFN of the range
+ *
+ * i386 with NUMA use alloc_remap() to store a node_mem_map on a local node.
+ * The map is kept at the end physical page range that has already been
+ * registered with add_active_range(). This function allows an arch to shrink
+ * an existing registered range.
+ */
+void __init shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
+						unsigned long new_end_pfn)
+{
+	int i;
+
+	/* Find the old active region end and shrink */
+	for_each_active_range_index_in_nid(i, nid)
+		if (early_node_map[i].end_pfn == old_end_pfn) {
+			early_node_map[i].end_pfn = new_end_pfn;
+			break;
+		}
+}
+
+/**
+ * remove_all_active_ranges - Remove all currently registered regions
+ * During discovery, it may be found that a table like SRAT is invalid
+ * and an alternative discovery method must be used. This function removes
+ * all currently registered regions.
+ */
+void __init remove_all_active_ranges()
+{
+	memset(early_node_map, 0, sizeof(early_node_map));
+	nr_nodemap_entries = 0;
+}
+
+/* Compare two active node_active_regions */
+static int __init cmp_node_active_region(const void *a, const void *b)
+{
+	struct node_active_region *arange = (struct node_active_region *)a;
+	struct node_active_region *brange = (struct node_active_region *)b;
+
+	/* Done this way to avoid overflows */
+	if (arange->start_pfn > brange->start_pfn)
+		return 1;
+	if (arange->start_pfn < brange->start_pfn)
+		return -1;
+
+	return 0;
+}
+
+/* sort the node_map by start_pfn */
+static void __init sort_node_map(void)
+{
+	sort(early_node_map, (size_t)nr_nodemap_entries,
+			sizeof(struct node_active_region),
+			cmp_node_active_region, NULL);
+}
+
+/* Find the lowest pfn for a node. This depends on a sorted early_node_map */
+unsigned long __init find_min_pfn_for_node(unsigned long nid)
+{
+	int i;
+
+	/* Assuming a sorted map, the first range found has the starting pfn */
+	for_each_active_range_index_in_nid(i, nid)
+		return early_node_map[i].start_pfn;
+
+	printk(KERN_WARNING "Could not find start_pfn for node %lu\n", nid);
+	return 0;
+}
+
+/**
+ * find_min_pfn_with_active_regions - Find the minimum PFN registered
+ *
+ * It returns the minimum PFN based on information provided via
+ * add_active_range()
+ */
+unsigned long __init find_min_pfn_with_active_regions(void)
+{
+	return find_min_pfn_for_node(MAX_NUMNODES);
+}
+
+/**
+ * find_max_pfn_with_active_regions - Find the maximum PFN registered
+ *
+ * It returns the maximum PFN based on information provided via
+ * add_active_range()
+ */
+unsigned long __init find_max_pfn_with_active_regions(void)
+{
+	int i;
+	unsigned long max_pfn = 0;
+
+	for (i = 0; i < nr_nodemap_entries; i++)
+		max_pfn = max(max_pfn, early_node_map[i].end_pfn);
+
+	return max_pfn;
+}
+
+/**
+ * free_area_init_nodes - Initialise all pg_data_t and zone data
+ * @arch_max_dma_pfn: The maximum PFN usable for ZONE_DMA
+ * @arch_max_dma32_pfn: The maximum PFN usable for ZONE_DMA32
+ * @arch_max_low_pfn: The maximum PFN usable for ZONE_NORMAL
+ * @arch_max_high_pfn: The maximum PFN usable for ZONE_HIGHMEM
+ *
+ * This will call free_area_init_node() for each active node in the system.
+ * Using the page ranges provided by add_active_range(), the size of each
+ * zone in each node and their holes is calculated. If the maximum PFN
+ * between two adjacent zones match, it is assumed that the zone is empty.
+ * For example, if arch_max_dma_pfn == arch_max_dma32_pfn, it is assumed
+ * that arch_max_dma32_pfn has no pages. It is also assumed that a zone
+ * starts where the previous one ended. For example, ZONE_DMA32 starts
+ * at arch_max_dma_pfn.
+ */
+void __init free_area_init_nodes(unsigned long *max_zone_pfn)
+{
+	unsigned long nid;
+	enum zone_type i;
+
+	/* Record where the zone boundaries are */
+	memset(arch_zone_lowest_possible_pfn, 0,
+				sizeof(arch_zone_lowest_possible_pfn));
+	memset(arch_zone_highest_possible_pfn, 0,
+				sizeof(arch_zone_highest_possible_pfn));
+	arch_zone_lowest_possible_pfn[0] = find_min_pfn_with_active_regions();
+	arch_zone_highest_possible_pfn[0] = max_zone_pfn[0];
+	for (i = 1; i < MAX_NR_ZONES; i++) {
+		arch_zone_lowest_possible_pfn[i] =
+			arch_zone_highest_possible_pfn[i-1];
+		arch_zone_highest_possible_pfn[i] =
+			max(max_zone_pfn[i], arch_zone_lowest_possible_pfn[i]);
+	}
+
+	/* Regions in the early_node_map can be in any order */
+	sort_node_map();
+
+	/* Print out the zone ranges */
+	printk("Zone PFN ranges:\n");
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		printk("  %-8s %8lu -> %8lu\n",
+				zone_names[i],
+				arch_zone_lowest_possible_pfn[i],
+				arch_zone_highest_possible_pfn[i]);
+
+	/* Print out the early_node_map[] */
+	printk("early_node_map[%d] active PFN ranges\n", nr_nodemap_entries);
+	for (i = 0; i < nr_nodemap_entries; i++)
+		printk("  %3d: %8lu -> %8lu\n", early_node_map[i].nid,
+						early_node_map[i].start_pfn,
+						early_node_map[i].end_pfn);
+
+	/* Initialise every node */
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		free_area_init_node(nid, pgdat, NULL,
+				find_min_pfn_for_node(nid), NULL);
+	}
+}
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 static bootmem_data_t contig_bootmem_data;
 struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data };
-- 
cgit v1.2.3


From 5cb248abf5ab65ab543b2d5fc16c738b28031fc0 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:52 -0700
Subject: [PATCH] Have x86_64 use add_active_range() and free_area_init_nodes

Size zones and holes in an architecture independent manner for x86_64.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/Kconfig         |   3 ++
 arch/x86_64/kernel/e820.c   | 125 ++++++++++++++++----------------------------
 arch/x86_64/kernel/setup.c  |   7 ++-
 arch/x86_64/mm/init.c       |  65 +++--------------------
 arch/x86_64/mm/k8topology.c |   3 ++
 arch/x86_64/mm/numa.c       |  21 ++++----
 arch/x86_64/mm/srat.c       |  11 ++--
 include/asm-x86_64/e820.h   |   5 +-
 include/asm-x86_64/proto.h  |   2 -
 9 files changed, 84 insertions(+), 158 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index efe249e7d6b3..326aff7a87ea 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -85,6 +85,9 @@ config ARCH_MAY_HAVE_PC_FDC
 	bool
 	default y
 
+config ARCH_POPULATES_NODE_MAP
+	def_bool y
+
 config DMI
 	bool
 	default y
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index c0af3828df45..b3f0908668ec 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -162,59 +162,14 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end, unsi
 	return -1UL;		
 } 
 
-/* 
- * Free bootmem based on the e820 table for a node.
- */
-void __init e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end)
-{
-	int i;
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i]; 
-		unsigned long last, addr;
-
-		if (ei->type != E820_RAM || 
-		    ei->addr+ei->size <= start || 
-		    ei->addr >= end)
-			continue;
-
-		addr = round_up(ei->addr, PAGE_SIZE);
-		if (addr < start) 
-			addr = start;
-
-		last = round_down(ei->addr + ei->size, PAGE_SIZE); 
-		if (last >= end)
-			last = end; 
-
-		if (last > addr && last-addr >= PAGE_SIZE)
-			free_bootmem_node(pgdat, addr, last-addr);
-	}
-}
-
 /*
  * Find the highest page frame number we have available
  */
 unsigned long __init e820_end_of_ram(void)
 {
-	int i;
 	unsigned long end_pfn = 0;
+	end_pfn = find_max_pfn_with_active_regions();
 	
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i]; 
-		unsigned long start, end;
-
-		start = round_up(ei->addr, PAGE_SIZE); 
-		end = round_down(ei->addr + ei->size, PAGE_SIZE); 
-		if (start >= end)
-			continue;
-		if (ei->type == E820_RAM) { 
-		if (end > end_pfn<<PAGE_SHIFT)
-			end_pfn = end>>PAGE_SHIFT;
-		} else { 
-			if (end > end_pfn_map<<PAGE_SHIFT) 
-				end_pfn_map = end>>PAGE_SHIFT;
-		} 
-	}
-
 	if (end_pfn > end_pfn_map) 
 		end_pfn_map = end_pfn;
 	if (end_pfn_map > MAXMEM>>PAGE_SHIFT)
@@ -224,43 +179,10 @@ unsigned long __init e820_end_of_ram(void)
 	if (end_pfn > end_pfn_map) 
 		end_pfn = end_pfn_map; 
 
+	printk("end_pfn_map = %lu\n", end_pfn_map);
 	return end_pfn;	
 }
 
-/* 
- * Compute how much memory is missing in a range.
- * Unlike the other functions in this file the arguments are in page numbers.
- */
-unsigned long __init
-e820_hole_size(unsigned long start_pfn, unsigned long end_pfn)
-{
-	unsigned long ram = 0;
-	unsigned long start = start_pfn << PAGE_SHIFT;
-	unsigned long end = end_pfn << PAGE_SHIFT;
-	int i;
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
-		unsigned long last, addr;
-
-		if (ei->type != E820_RAM ||
-		    ei->addr+ei->size <= start ||
-		    ei->addr >= end)
-			continue;
-
-		addr = round_up(ei->addr, PAGE_SIZE);
-		if (addr < start)
-			addr = start;
-
-		last = round_down(ei->addr + ei->size, PAGE_SIZE);
-		if (last >= end)
-			last = end;
-
-		if (last > addr)
-			ram += last - addr;
-	}
-	return ((end - start) - ram) >> PAGE_SHIFT;
-}
-
 /*
  * Mark e820 reserved areas as busy for the resource manager.
  */
@@ -342,6 +264,49 @@ void __init e820_mark_nosave_regions(void)
 	}
 }
 
+/* Walk the e820 map and register active regions within a node */
+void __init
+e820_register_active_regions(int nid, unsigned long start_pfn,
+							unsigned long end_pfn)
+{
+	int i;
+	unsigned long ei_startpfn, ei_endpfn;
+	for (i = 0; i < e820.nr_map; i++) {
+		struct e820entry *ei = &e820.map[i];
+		ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT;
+		ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE)
+								>> PAGE_SHIFT;
+
+		/* Skip map entries smaller than a page */
+		if (ei_startpfn > ei_endpfn)
+			continue;
+
+		/* Check if end_pfn_map should be updated */
+		if (ei->type != E820_RAM && ei_endpfn > end_pfn_map)
+			end_pfn_map = ei_endpfn;
+
+		/* Skip if map is outside the node */
+		if (ei->type != E820_RAM ||
+				ei_endpfn <= start_pfn ||
+				ei_startpfn >= end_pfn)
+			continue;
+
+		/* Check for overlaps */
+		if (ei_startpfn < start_pfn)
+			ei_startpfn = start_pfn;
+		if (ei_endpfn > end_pfn)
+			ei_endpfn = end_pfn;
+
+		/* Obey end_user_pfn to save on memmap */
+		if (ei_startpfn >= end_user_pfn)
+			continue;
+		if (ei_endpfn > end_user_pfn)
+			ei_endpfn = end_user_pfn;
+
+		add_active_range(nid, ei_startpfn, ei_endpfn);
+	}
+}
+
 /* 
  * Add a memory region to the kernel e820 map.
  */ 
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index f98e48cae6da..0b00bb2ea576 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -292,7 +292,8 @@ contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	if (bootmap == -1L)
 		panic("Cannot find bootmem map of size %ld\n",bootmap_size);
 	bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
-	e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
+	e820_register_active_regions(0, start_pfn, end_pfn);
+	free_bootmem_with_active_regions(0, end_pfn);
 	reserve_bootmem(bootmap, bootmap_size);
 } 
 #endif
@@ -384,6 +385,7 @@ void __init setup_arch(char **cmdline_p)
 
 	finish_e820_parsing();
 
+	e820_register_active_regions(0, 0, -1UL);
 	/*
 	 * partially used pages are not usable - thus
 	 * we are rounding upwards:
@@ -414,6 +416,9 @@ void __init setup_arch(char **cmdline_p)
 	max_pfn = end_pfn;
 	high_memory = (void *)__va(end_pfn * PAGE_SIZE - 1) + 1;
 
+	/* Remove active ranges so rediscovery with NUMA-awareness happens */
+	remove_all_active_ranges();
+
 #ifdef CONFIG_ACPI_NUMA
 	/*
 	 * Parse SRAT to discover nodes.
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 1e4669fa5734..47928399e38a 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -403,69 +403,15 @@ void __cpuinit zap_low_mappings(int cpu)
 	__flush_tlb_all();
 }
 
-/* Compute zone sizes for the DMA and DMA32 zones in a node. */
-__init void
-size_zones(unsigned long *z, unsigned long *h,
-	   unsigned long start_pfn, unsigned long end_pfn)
-{
- 	int i;
- 	unsigned long w;
-
- 	for (i = 0; i < MAX_NR_ZONES; i++)
- 		z[i] = 0;
-
- 	if (start_pfn < MAX_DMA_PFN)
- 		z[ZONE_DMA] = MAX_DMA_PFN - start_pfn;
- 	if (start_pfn < MAX_DMA32_PFN) {
- 		unsigned long dma32_pfn = MAX_DMA32_PFN;
- 		if (dma32_pfn > end_pfn)
- 			dma32_pfn = end_pfn;
- 		z[ZONE_DMA32] = dma32_pfn - start_pfn;
- 	}
- 	z[ZONE_NORMAL] = end_pfn - start_pfn;
-
- 	/* Remove lower zones from higher ones. */
- 	w = 0;
- 	for (i = 0; i < MAX_NR_ZONES; i++) {
- 		if (z[i])
- 			z[i] -= w;
- 	        w += z[i];
-	}
-
-	/* Compute holes */
-	w = start_pfn;
-	for (i = 0; i < MAX_NR_ZONES; i++) {
-		unsigned long s = w;
-		w += z[i];
-		h[i] = e820_hole_size(s, w);
-	}
-
-	/* Add the space pace needed for mem_map to the holes too. */
-	for (i = 0; i < MAX_NR_ZONES; i++)
-		h[i] += (z[i] * sizeof(struct page)) / PAGE_SIZE;
-
-	/* The 16MB DMA zone has the kernel and other misc mappings.
- 	   Account them too */
-	if (h[ZONE_DMA]) {
-		h[ZONE_DMA] += dma_reserve;
-		if (h[ZONE_DMA] >= z[ZONE_DMA]) {
-			printk(KERN_WARNING
-				"Kernel too large and filling up ZONE_DMA?\n");
-			h[ZONE_DMA] = z[ZONE_DMA];
-		}
-	}
-}
-
 #ifndef CONFIG_NUMA
 void __init paging_init(void)
 {
-	unsigned long zones[MAX_NR_ZONES], holes[MAX_NR_ZONES];
-
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = {MAX_DMA_PFN,
+							MAX_DMA32_PFN,
+							end_pfn};
 	memory_present(0, 0, end_pfn);
 	sparse_init();
-	size_zones(zones, holes, 0, end_pfn);
-	free_area_init_node(0, NODE_DATA(0), zones,
-			    __pa(PAGE_OFFSET) >> PAGE_SHIFT, holes);
+	free_area_init_nodes(max_zone_pfns);
 }
 #endif
 
@@ -608,7 +554,8 @@ void __init mem_init(void)
 #else
 	totalram_pages = free_all_bootmem();
 #endif
-	reservedpages = end_pfn - totalram_pages - e820_hole_size(0, end_pfn);
+	reservedpages = end_pfn - totalram_pages -
+					absent_pages_in_range(0, end_pfn);
 
 	after_bootmem = 1;
 
diff --git a/arch/x86_64/mm/k8topology.c b/arch/x86_64/mm/k8topology.c
index 5cf594f9230d..b5b8dba28b4e 100644
--- a/arch/x86_64/mm/k8topology.c
+++ b/arch/x86_64/mm/k8topology.c
@@ -149,6 +149,9 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 		
 		nodes[nodeid].start = base; 
 		nodes[nodeid].end = limit;
+		e820_register_active_regions(nodeid,
+				nodes[nodeid].start >> PAGE_SHIFT,
+				nodes[nodeid].end >> PAGE_SHIFT);
 
 		prevbase = base;
 
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 322bf45fc36a..829a008bd39b 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -161,7 +161,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 					 bootmap_start >> PAGE_SHIFT, 
 					 start_pfn, end_pfn); 
 
-	e820_bootmem_free(NODE_DATA(nodeid), start, end);
+	free_bootmem_with_active_regions(nodeid, end);
 
 	reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys, pgdat_size); 
 	reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<<PAGE_SHIFT);
@@ -175,13 +175,11 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 void __init setup_node_zones(int nodeid)
 { 
 	unsigned long start_pfn, end_pfn, memmapsize, limit;
-	unsigned long zones[MAX_NR_ZONES];
-	unsigned long holes[MAX_NR_ZONES];
 
  	start_pfn = node_start_pfn(nodeid);
  	end_pfn = node_end_pfn(nodeid);
 
-	Dprintk(KERN_INFO "Setting up node %d %lx-%lx\n",
+	Dprintk(KERN_INFO "Setting up memmap for node %d %lx-%lx\n",
 		nodeid, start_pfn, end_pfn);
 
 	/* Try to allocate mem_map at end to not fill up precious <4GB
@@ -195,10 +193,6 @@ void __init setup_node_zones(int nodeid)
 				round_down(limit - memmapsize, PAGE_SIZE), 
 				limit);
 #endif
-
-	size_zones(zones, holes, start_pfn, end_pfn);
-	free_area_init_node(nodeid, NODE_DATA(nodeid), zones,
-			    start_pfn, holes);
 } 
 
 void __init numa_init_array(void)
@@ -259,8 +253,11 @@ static int __init numa_emulation(unsigned long start_pfn, unsigned long end_pfn)
  		printk(KERN_ERR "No NUMA hash function found. Emulation disabled.\n");
  		return -1;
  	}
- 	for_each_online_node(i)
+ 	for_each_online_node(i) {
+		e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT,
+						nodes[i].end >> PAGE_SHIFT);
  		setup_node_bootmem(i, nodes[i].start, nodes[i].end);
+	}
  	numa_init_array();
  	return 0;
 }
@@ -299,6 +296,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	for (i = 0; i < NR_CPUS; i++)
 		numa_set_node(i, 0);
 	node_to_cpumask[0] = cpumask_of_cpu(0);
+	e820_register_active_regions(0, start_pfn, end_pfn);
 	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
 }
 
@@ -340,12 +338,17 @@ static void __init arch_sparse_init(void)
 void __init paging_init(void)
 { 
 	int i;
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = { MAX_DMA_PFN,
+		MAX_DMA32_PFN,
+		end_pfn};
 
 	arch_sparse_init();
 
 	for_each_online_node(i) {
 		setup_node_zones(i); 
 	}
+
+	free_area_init_nodes(max_zone_pfns);
 } 
 
 static __init int numa_setup(char *opt)
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index ca10701e7a90..7b50bb1caabe 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -93,6 +93,7 @@ static __init void bad_srat(void)
 		apicid_to_node[i] = NUMA_NO_NODE;
 	for (i = 0; i < MAX_NUMNODES; i++)
 		nodes_add[i].start = nodes[i].end = 0;
+	remove_all_active_ranges();
 }
 
 static __init inline int srat_disabled(void)
@@ -175,7 +176,7 @@ static int hotadd_enough_memory(struct bootnode *nd)
 
 	if (mem < 0)
 		return 0;
-	allowed = (end_pfn - e820_hole_size(0, end_pfn)) * PAGE_SIZE;
+	allowed = (end_pfn - absent_pages_in_range(0, end_pfn)) * PAGE_SIZE;
 	allowed = (allowed / 100) * hotadd_percent;
 	if (allocated + mem > allowed) {
 		unsigned long range;
@@ -225,7 +226,7 @@ static int reserve_hotadd(int node, unsigned long start, unsigned long end)
 	}
 
 	/* This check might be a bit too strict, but I'm keeping it for now. */
-	if (e820_hole_size(s_pfn, e_pfn) != e_pfn - s_pfn) {
+	if (absent_pages_in_range(s_pfn, e_pfn) != e_pfn - s_pfn) {
 		printk(KERN_ERR "SRAT: Hotplug area has existing memory\n");
 		return -1;
 	}
@@ -319,6 +320,8 @@ acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma)
 
 	printk(KERN_INFO "SRAT: Node %u PXM %u %Lx-%Lx\n", node, pxm,
 	       nd->start, nd->end);
+	e820_register_active_regions(node, nd->start >> PAGE_SHIFT,
+						nd->end >> PAGE_SHIFT);
 
 #ifdef RESERVE_HOTADD
  	if (ma->flags.hot_pluggable && reserve_hotadd(node, start, end) < 0) {
@@ -343,13 +346,13 @@ static int nodes_cover_memory(void)
 		unsigned long s = nodes[i].start >> PAGE_SHIFT;
 		unsigned long e = nodes[i].end >> PAGE_SHIFT;
 		pxmram += e - s;
-		pxmram -= e820_hole_size(s, e);
+		pxmram -= absent_pages_in_range(s, e);
 		pxmram -= nodes_add[i].end - nodes_add[i].start;
 		if ((long)pxmram < 0)
 			pxmram = 0;
 	}
 
-	e820ram = end_pfn - e820_hole_size(0, end_pfn);
+	e820ram = end_pfn - absent_pages_in_range(0, end_pfn);
 	/* We seem to lose 3 pages somewhere. Allow a bit of slack. */
 	if ((long)(e820ram - pxmram) >= 1*1024*1024) {
 		printk(KERN_ERR
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index e15d3c8628f3..fa2086774105 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -47,10 +47,9 @@ extern void e820_print_map(char *who);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
 
-extern void e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end);
 extern void e820_setup_gap(void);
-extern unsigned long e820_hole_size(unsigned long start_pfn,
-				    unsigned long end_pfn);
+extern void e820_register_active_regions(int nid,
+				unsigned long start_pfn, unsigned long end_pfn);
 
 extern void finish_e820_parsing(void);
 
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index b73d0c76613c..c28fc2db2171 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -24,8 +24,6 @@ extern void mtrr_bp_init(void);
 #define mtrr_bp_init() do {} while (0)
 #endif
 extern void init_memory_mapping(unsigned long start, unsigned long end);
-extern void size_zones(unsigned long *z, unsigned long *h,
-			unsigned long start_pfn, unsigned long end_pfn);
 
 extern void system_call(void); 
 extern int kernel_syscall(void);
-- 
cgit v1.2.3


From 05e0caad3b7bd0d0fbeff980bca22f186241a501 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:54 -0700
Subject: [PATCH] Have ia64 use add_active_range() and free_area_init_nodes

Size zones and holes in an architecture independent manner for ia64.

[bob.picco@hp.com: fix ia64 FLATMEM+VIRTUAL_MEM_MAP]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig          |  3 +++
 arch/ia64/mm/contig.c      | 67 +++++++++++-----------------------------------
 arch/ia64/mm/discontig.c   | 44 +++++++-----------------------
 arch/ia64/mm/init.c        | 12 +++++++++
 include/asm-ia64/meminit.h |  1 +
 5 files changed, 40 insertions(+), 87 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index f521f2f60a78..d5ee4fc8fe66 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -356,6 +356,9 @@ config NODES_SHIFT
 	  MAX_NUMNODES will be 2^(This value).
 	  If in doubt, use the default.
 
+config ARCH_POPULATES_NODE_MAP
+	def_bool y
+
 # VIRTUAL_MEM_MAP and FLAT_NODE_MEM_MAP are functionally equivalent.
 # VIRTUAL_MEM_MAP has been retained for historical reasons.
 config VIRTUAL_MEM_MAP
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index e004143ba86b..719d476e71ba 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -26,7 +26,6 @@
 #include <asm/mca.h>
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
-static unsigned long num_dma_physpages;
 static unsigned long max_gap;
 #endif
 
@@ -218,18 +217,6 @@ count_pages (u64 start, u64 end, void *arg)
 	return 0;
 }
 
-#ifdef CONFIG_VIRTUAL_MEM_MAP
-static int
-count_dma_pages (u64 start, u64 end, void *arg)
-{
-	unsigned long *count = arg;
-
-	if (start < MAX_DMA_ADDRESS)
-		*count += (min(end, MAX_DMA_ADDRESS) - start) >> PAGE_SHIFT;
-	return 0;
-}
-#endif
-
 /*
  * Set up the page tables.
  */
@@ -238,45 +225,22 @@ void __init
 paging_init (void)
 {
 	unsigned long max_dma;
-	unsigned long zones_size[MAX_NR_ZONES];
-#ifdef CONFIG_VIRTUAL_MEM_MAP
-	unsigned long zholes_size[MAX_NR_ZONES];
-#endif
-
-	/* initialize mem_map[] */
-
-	memset(zones_size, 0, sizeof(zones_size));
+	unsigned long nid = 0;
+	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
 	num_physpages = 0;
 	efi_memmap_walk(count_pages, &num_physpages);
 
 	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
+	max_zone_pfns[ZONE_DMA] = max_dma;
+	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
-	memset(zholes_size, 0, sizeof(zholes_size));
-
-	num_dma_physpages = 0;
-	efi_memmap_walk(count_dma_pages, &num_dma_physpages);
-
-	if (max_low_pfn < max_dma) {
-		zones_size[ZONE_DMA] = max_low_pfn;
-		zholes_size[ZONE_DMA] = max_low_pfn - num_dma_physpages;
-	} else {
-		zones_size[ZONE_DMA] = max_dma;
-		zholes_size[ZONE_DMA] = max_dma - num_dma_physpages;
-		if (num_physpages > num_dma_physpages) {
-			zones_size[ZONE_NORMAL] = max_low_pfn - max_dma;
-			zholes_size[ZONE_NORMAL] =
-				((max_low_pfn - max_dma) -
-				 (num_physpages - num_dma_physpages));
-		}
-	}
-
+	efi_memmap_walk(register_active_ranges, &nid);
 	efi_memmap_walk(find_largest_hole, (u64 *)&max_gap);
 	if (max_gap < LARGE_GAP) {
 		vmem_map = (struct page *) 0;
-		free_area_init_node(0, NODE_DATA(0), zones_size, 0,
-				    zholes_size);
+		free_area_init_nodes(max_zone_pfns);
 	} else {
 		unsigned long map_size;
 
@@ -288,20 +252,19 @@ paging_init (void)
 		vmem_map = (struct page *) vmalloc_end;
 		efi_memmap_walk(create_mem_map_page_table, NULL);
 
-		NODE_DATA(0)->node_mem_map = vmem_map;
-		free_area_init_node(0, NODE_DATA(0), zones_size,
-				    0, zholes_size);
+		/*
+		 * alloc_node_mem_map makes an adjustment for mem_map
+		 * which isn't compatible with vmem_map.
+		 */
+		NODE_DATA(0)->node_mem_map = vmem_map +
+			find_min_pfn_with_active_regions();
+		free_area_init_nodes(max_zone_pfns);
 
 		printk("Virtual mem_map starts at 0x%p\n", mem_map);
 	}
 #else /* !CONFIG_VIRTUAL_MEM_MAP */
-	if (max_low_pfn < max_dma)
-		zones_size[ZONE_DMA] = max_low_pfn;
-	else {
-		zones_size[ZONE_DMA] = max_dma;
-		zones_size[ZONE_NORMAL] = max_low_pfn - max_dma;
-	}
-	free_area_init(zones_size);
+	add_active_range(0, 0, max_low_pfn);
+	free_area_init_nodes(max_zone_pfns);
 #endif /* !CONFIG_VIRTUAL_MEM_MAP */
 	zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page));
 }
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index d260bffa01ab..7bd28079dcc4 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -654,6 +654,7 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
 {
 	unsigned long end = start + len;
 
+	add_active_range(node, start >> PAGE_SHIFT, end >> PAGE_SHIFT);
 	mem_data[node].num_physpages += len >> PAGE_SHIFT;
 	if (start <= __pa(MAX_DMA_ADDRESS))
 		mem_data[node].num_dma_physpages +=
@@ -678,10 +679,10 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
 void __init paging_init(void)
 {
 	unsigned long max_dma;
-	unsigned long zones_size[MAX_NR_ZONES];
-	unsigned long zholes_size[MAX_NR_ZONES];
 	unsigned long pfn_offset = 0;
+	unsigned long max_pfn = 0;
 	int node;
+	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
 	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
 
@@ -698,47 +699,20 @@ void __init paging_init(void)
 #endif
 
 	for_each_online_node(node) {
-		memset(zones_size, 0, sizeof(zones_size));
-		memset(zholes_size, 0, sizeof(zholes_size));
-
 		num_physpages += mem_data[node].num_physpages;
-
-		if (mem_data[node].min_pfn >= max_dma) {
-			/* All of this node's memory is above ZONE_DMA */
-			zones_size[ZONE_NORMAL] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn;
-			zholes_size[ZONE_NORMAL] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn -
-				mem_data[node].num_physpages;
-		} else if (mem_data[node].max_pfn < max_dma) {
-			/* All of this node's memory is in ZONE_DMA */
-			zones_size[ZONE_DMA] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn;
-			zholes_size[ZONE_DMA] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn -
-				mem_data[node].num_dma_physpages;
-		} else {
-			/* This node has memory in both zones */
-			zones_size[ZONE_DMA] = max_dma -
-				mem_data[node].min_pfn;
-			zholes_size[ZONE_DMA] = zones_size[ZONE_DMA] -
-				mem_data[node].num_dma_physpages;
-			zones_size[ZONE_NORMAL] = mem_data[node].max_pfn -
-				max_dma;
-			zholes_size[ZONE_NORMAL] = zones_size[ZONE_NORMAL] -
-				(mem_data[node].num_physpages -
-				 mem_data[node].num_dma_physpages);
-		}
-
 		pfn_offset = mem_data[node].min_pfn;
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
 		NODE_DATA(node)->node_mem_map = vmem_map + pfn_offset;
 #endif
-		free_area_init_node(node, NODE_DATA(node), zones_size,
-				    pfn_offset, zholes_size);
+		if (mem_data[node].max_pfn > max_pfn)
+			max_pfn = mem_data[node].max_pfn;
 	}
 
+	max_zone_pfns[ZONE_DMA] = max_dma;
+	max_zone_pfns[ZONE_NORMAL] = max_pfn;
+	free_area_init_nodes(max_zone_pfns);
+
 	zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page));
 }
 
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index 30617ccb4f7e..ff87a5cba399 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -593,6 +593,18 @@ find_largest_hole (u64 start, u64 end, void *arg)
 	last_end = end;
 	return 0;
 }
+
+int __init
+register_active_ranges(u64 start, u64 end, void *nid)
+{
+	BUG_ON(nid == NULL);
+	BUG_ON(*(unsigned long *)nid >= MAX_NUMNODES);
+
+	add_active_range(*(unsigned long *)nid,
+				__pa(start) >> PAGE_SHIFT,
+				__pa(end) >> PAGE_SHIFT);
+	return 0;
+}
 #endif /* CONFIG_VIRTUAL_MEM_MAP */
 
 static int __init
diff --git a/include/asm-ia64/meminit.h b/include/asm-ia64/meminit.h
index 6a33a07b3f1d..c3b1f862e6e7 100644
--- a/include/asm-ia64/meminit.h
+++ b/include/asm-ia64/meminit.h
@@ -55,6 +55,7 @@ extern void efi_memmap_init(unsigned long *, unsigned long *);
   extern unsigned long vmalloc_end;
   extern struct page *vmem_map;
   extern int find_largest_hole (u64 start, u64 end, void *arg);
+  extern int register_active_ranges (u64 start, u64 end, void *arg);
   extern int create_mem_map_page_table (u64 start, u64 end, void *arg);
   extern int vmemmap_find_next_valid_pfn(int, int);
 #else
-- 
cgit v1.2.3


From 0e0b864e069c52a7b3e4a7da56e29b03a012fd75 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:56 -0700
Subject: [PATCH] Account for memmap and optionally the kernel image as holes

The x86_64 code accounted for memmap and some portions of the the DMA zone as
holes.  This was because those areas would never be reclaimed and accounting
for them as memory affects min watermarks.  This patch will account for the
memmap as a memory hole.  Architectures may optionally use set_dma_reserve()
if they wish to account for a portion of memory in ZONE_DMA as a hole.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/init.c |  4 +++-
 include/linux/mm.h    |  1 +
 mm/page_alloc.c       | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 63 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 47928399e38a..3e16fe08150e 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -655,8 +655,10 @@ void __init reserve_bootmem_generic(unsigned long phys, unsigned len)
 #else       		
 	reserve_bootmem(phys, len);    
 #endif
-	if (phys+len <= MAX_DMA_PFN*PAGE_SIZE)
+	if (phys+len <= MAX_DMA_PFN*PAGE_SIZE) {
 		dma_reserve += len / PAGE_SIZE;
+		set_dma_reserve(dma_reserve);
+	}
 }
 
 int kern_addr_valid(unsigned long addr) 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index c0402da7cce0..22936e1fcdf2 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -984,6 +984,7 @@ extern void sparse_memory_present_with_active_regions(int nid);
 extern int early_pfn_to_nid(unsigned long pfn);
 #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+extern void set_dma_reserve(unsigned long new_dma_reserve);
 extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long);
 extern void setup_per_zone_pages_min(void);
 extern void mem_init(void);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 26c9939857fa..8d9a1eb9fbba 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -104,6 +104,7 @@ int min_free_kbytes = 1024;
 
 unsigned long __meminitdata nr_kernel_pages;
 unsigned long __meminitdata nr_all_pages;
+static unsigned long __initdata dma_reserve;
 
 #ifdef CONFIG_ARCH_POPULATES_NODE_MAP
   /*
@@ -2213,6 +2214,20 @@ unsigned long __init zone_absent_pages_in_node(int nid,
 				arch_zone_lowest_possible_pfn[zone_type],
 				arch_zone_highest_possible_pfn[zone_type]);
 }
+
+/* Return the zone index a PFN is in */
+int memmap_zone_idx(struct page *lmem_map)
+{
+	int i;
+	unsigned long phys_addr = virt_to_phys(lmem_map);
+	unsigned long pfn = phys_addr >> PAGE_SHIFT;
+
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		if (pfn < arch_zone_highest_possible_pfn[i])
+			break;
+
+	return i;
+}
 #else
 static inline unsigned long zone_spanned_pages_in_node(int nid,
 					unsigned long zone_type,
@@ -2230,6 +2245,11 @@ static inline unsigned long zone_absent_pages_in_node(int nid,
 
 	return zholes_size[zone_type];
 }
+
+static inline int memmap_zone_idx(struct page *lmem_map)
+{
+	return MAX_NR_ZONES;
+}
 #endif
 
 static void __init calculate_node_totalpages(struct pglist_data *pgdat,
@@ -2274,12 +2294,35 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 	
 	for (j = 0; j < MAX_NR_ZONES; j++) {
 		struct zone *zone = pgdat->node_zones + j;
-		unsigned long size, realsize;
+		unsigned long size, realsize, memmap_pages;
 
 		size = zone_spanned_pages_in_node(nid, j, zones_size);
 		realsize = size - zone_absent_pages_in_node(nid, j,
 								zholes_size);
 
+		/*
+		 * Adjust realsize so that it accounts for how much memory
+		 * is used by this zone for memmap. This affects the watermark
+		 * and per-cpu initialisations
+		 */
+		memmap_pages = (size * sizeof(struct page)) >> PAGE_SHIFT;
+		if (realsize >= memmap_pages) {
+			realsize -= memmap_pages;
+			printk(KERN_DEBUG
+				"  %s zone: %lu pages used for memmap\n",
+				zone_names[j], memmap_pages);
+		} else
+			printk(KERN_WARNING
+				"  %s zone: %lu pages exceeds realsize %lu\n",
+				zone_names[j], memmap_pages, realsize);
+
+		/* Account for reserved DMA pages */
+		if (j == ZONE_DMA && realsize > dma_reserve) {
+			realsize -= dma_reserve;
+			printk(KERN_DEBUG "  DMA zone: %lu pages reserved\n",
+								dma_reserve);
+		}
+
 		if (!is_highmem_idx(j))
 			nr_kernel_pages += realsize;
 		nr_all_pages += realsize;
@@ -2596,6 +2639,21 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn)
 }
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
+/**
+ * set_dma_reserve - Account the specified number of pages reserved in ZONE_DMA
+ * @new_dma_reserve - The number of pages to mark reserved
+ *
+ * The per-cpu batchsize and zone watermarks are determined by present_pages.
+ * In the DMA zone, a significant percentage may be consumed by kernel image
+ * and other unfreeable allocations which can skew the watermarks badly. This
+ * function may optionally be used to account for unfreeable pages in
+ * ZONE_DMA. The effect will be lower watermarks and smaller per-cpu batchsize
+ */
+void __init set_dma_reserve(unsigned long new_dma_reserve)
+{
+	dma_reserve = new_dma_reserve;
+}
+
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 static bootmem_data_t contig_bootmem_data;
 struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data };
-- 
cgit v1.2.3


From fb01439c5b778d5974a488c5d4fe85e6d0e18a68 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@skynet.ie>
Date: Wed, 27 Sep 2006 01:49:59 -0700
Subject: [PATCH] Allow an arch to expand node boundaries

Arch-independent zone-sizing determines the size of a node
(pgdat->node_spanned_pages) based on the physical memory that was
registered by the architecture.  However, when
CONFIG_MEMORY_HOTPLUG_RESERVE is set, the architecture expects that the
spanned_pages will be much larger and that mem_map will be allocated that
is used lated on memory hot-add.

This patch allows an architecture that sets CONFIG_MEMORY_HOTPLUG_RESERVE
to call push_node_boundaries() which will set the node beginning and end to
at *least* the requested boundary.

Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/srat.c |  2 ++
 include/linux/mm.h    |  2 ++
 mm/page_alloc.c       | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 71 insertions(+)

(limited to 'include')

diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index db1b2e11cf8f..f8c04d6935c9 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -324,6 +324,8 @@ acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma)
 	       nd->start, nd->end);
 	e820_register_active_regions(node, nd->start >> PAGE_SHIFT,
 						nd->end >> PAGE_SHIFT);
+	push_node_boundaries(node, nd->start >> PAGE_SHIFT,
+						nd->end >> PAGE_SHIFT);
 
 #ifdef RESERVE_HOTADD
  	if (ma->flags.hot_pluggable && reserve_hotadd(node, start, end) < 0) {
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 22936e1fcdf2..9d046db31e76 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -970,6 +970,8 @@ extern void add_active_range(unsigned int nid, unsigned long start_pfn,
 					unsigned long end_pfn);
 extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
 						unsigned long new_end_pfn);
+extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
+					unsigned long end_pfn);
 extern void remove_all_active_ranges(void);
 extern unsigned long absent_pages_in_range(unsigned long start_pfn,
 						unsigned long end_pfn);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 75133e1dc4b1..acbf58f8a1b7 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -131,6 +131,10 @@ static unsigned long __initdata dma_reserve;
   int __initdata nr_nodemap_entries;
   unsigned long __initdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES];
   unsigned long __initdata arch_zone_highest_possible_pfn[MAX_NR_ZONES];
+#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
+  unsigned long __initdata node_boundary_start_pfn[MAX_NUMNODES];
+  unsigned long __initdata node_boundary_end_pfn[MAX_NUMNODES];
+#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
 #ifdef CONFIG_DEBUG_VM
@@ -2094,6 +2098,62 @@ void __init sparse_memory_present_with_active_regions(int nid)
 				early_node_map[i].end_pfn);
 }
 
+/**
+ * push_node_boundaries - Push node boundaries to at least the requested boundary
+ * @nid: The nid of the node to push the boundary for
+ * @start_pfn: The start pfn of the node
+ * @end_pfn: The end pfn of the node
+ *
+ * In reserve-based hot-add, mem_map is allocated that is unused until hotadd
+ * time. Specifically, on x86_64, SRAT will report ranges that can potentially
+ * be hotplugged even though no physical memory exists. This function allows
+ * an arch to push out the node boundaries so mem_map is allocated that can
+ * be used later.
+ */
+#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
+void __init push_node_boundaries(unsigned int nid,
+		unsigned long start_pfn, unsigned long end_pfn)
+{
+	printk(KERN_DEBUG "Entering push_node_boundaries(%u, %lu, %lu)\n",
+			nid, start_pfn, end_pfn);
+
+	/* Initialise the boundary for this node if necessary */
+	if (node_boundary_end_pfn[nid] == 0)
+		node_boundary_start_pfn[nid] = -1UL;
+
+	/* Update the boundaries */
+	if (node_boundary_start_pfn[nid] > start_pfn)
+		node_boundary_start_pfn[nid] = start_pfn;
+	if (node_boundary_end_pfn[nid] < end_pfn)
+		node_boundary_end_pfn[nid] = end_pfn;
+}
+
+/* If necessary, push the node boundary out for reserve hotadd */
+static void __init account_node_boundary(unsigned int nid,
+		unsigned long *start_pfn, unsigned long *end_pfn)
+{
+	printk(KERN_DEBUG "Entering account_node_boundary(%u, %lu, %lu)\n",
+			nid, *start_pfn, *end_pfn);
+
+	/* Return if boundary information has not been provided */
+	if (node_boundary_end_pfn[nid] == 0)
+		return;
+
+	/* Check the boundaries and update if necessary */
+	if (node_boundary_start_pfn[nid] < *start_pfn)
+		*start_pfn = node_boundary_start_pfn[nid];
+	if (node_boundary_end_pfn[nid] > *end_pfn)
+		*end_pfn = node_boundary_end_pfn[nid];
+}
+#else
+void __init push_node_boundaries(unsigned int nid,
+		unsigned long start_pfn, unsigned long end_pfn) {}
+
+static void __init account_node_boundary(unsigned int nid,
+		unsigned long *start_pfn, unsigned long *end_pfn) {}
+#endif
+
+
 /**
  * get_pfn_range_for_nid - Return the start and end page frames for a node
  * @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned
@@ -2121,6 +2181,9 @@ void __init get_pfn_range_for_nid(unsigned int nid,
 		printk(KERN_WARNING "Node %u active with no memory\n", nid);
 		*start_pfn = 0;
 	}
+
+	/* Push the node boundaries out if requested */
+	account_node_boundary(nid, start_pfn, end_pfn);
 }
 
 /*
@@ -2527,6 +2590,10 @@ void __init remove_all_active_ranges()
 {
 	memset(early_node_map, 0, sizeof(early_node_map));
 	nr_nodemap_entries = 0;
+#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
+	memset(node_boundary_start_pfn, 0, sizeof(node_boundary_start_pfn));
+	memset(node_boundary_end_pfn, 0, sizeof(node_boundary_end_pfn));
+#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
 }
 
 /* Compare two active node_active_regions */
-- 
cgit v1.2.3


From e129b5c23c2b471d47f1c5d2b8b193fc2034af43 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:50:00 -0700
Subject: [PATCH] vm: add per-zone writeout counter

The VM is supposed to minimise the number of pages which get written off the
LRU (for IO scheduling efficiency, and for high reclaim-success rates).  But
we don't actually have a clear way of showing how true this is.

So add `nr_vmscan_write' to /proc/vmstat and /proc/zoneinfo - the number of
pages which have been written by the vm scanner in this zone and globally.

Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h | 1 +
 mm/vmscan.c            | 3 ++-
 mm/vmstat.c            | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 7fa1cbe9fa7a..1b0680cd84d2 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -58,6 +58,7 @@ enum zone_stat_item {
 	NR_WRITEBACK,
 	NR_UNSTABLE_NFS,	/* NFS unstable pages */
 	NR_BOUNCE,
+	NR_VMSCAN_WRITE,
 #ifdef CONFIG_NUMA
 	NUMA_HIT,		/* allocated in intended node */
 	NUMA_MISS,		/* allocated in non intended node */
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 87779dda4ec6..87d340999ce8 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -19,6 +19,7 @@
 #include <linux/pagemap.h>
 #include <linux/init.h>
 #include <linux/highmem.h>
+#include <linux/vmstat.h>
 #include <linux/file.h>
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
@@ -370,7 +371,7 @@ static pageout_t pageout(struct page *page, struct address_space *mapping)
 			/* synchronous write or broken a_ops? */
 			ClearPageReclaim(page);
 		}
-
+		inc_zone_page_state(page, NR_VMSCAN_WRITE);
 		return PAGE_SUCCESS;
 	}
 
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 490d8c1a0ded..69c657132e1f 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -465,6 +465,7 @@ static char *vmstat_text[] = {
 	"nr_writeback",
 	"nr_unstable",
 	"nr_bounce",
+	"nr_vmscan_write",
 
 #ifdef CONFIG_NUMA
 	"numa_hit",
-- 
cgit v1.2.3


From 5b99cd0effaf846240a15441aec459a592577eaf Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 27 Sep 2006 01:50:01 -0700
Subject: [PATCH] own header file for struct page

This moves the definition of struct page from mm.h to its own header file
page-struct.h.  This is a prereq to fix SetPageUptodate which is broken on
s390:

#define SetPageUptodate(_page)
       do {
               struct page *__page = (_page);
               if (!test_and_set_bit(PG_uptodate, &__page->flags))
                       page_test_and_clear_dirty(_page);
       } while (0)

_page gets used twice in this macro which can cause subtle bugs.  Using
__page for the page_test_and_clear_dirty call doesn't work since it causes
yet another problem with the page_test_and_clear_dirty macro as well.

In order to avoid all these problems caused by macros it seems to be a good
idea to get rid of them and convert them to static inline functions.
Because of header file include order it's necessary to have a seperate
header file for the struct page definition.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h       | 62 +-------------------------------------------
 include/linux/mm_types.h | 67 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/mmzone.h   |  5 ++++
 3 files changed, 73 insertions(+), 61 deletions(-)
 create mode 100644 include/linux/mm_types.h

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 9d046db31e76..7477fb59c4f2 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -16,6 +16,7 @@
 #include <linux/mutex.h>
 #include <linux/debug_locks.h>
 #include <linux/backing-dev.h>
+#include <linux/mm_types.h>
 
 struct mempolicy;
 struct anon_vma;
@@ -215,62 +216,6 @@ struct vm_operations_struct {
 struct mmu_gather;
 struct inode;
 
-/*
- * Each physical page in the system has a struct page associated with
- * it to keep track of whatever it is we are using the page for at the
- * moment. Note that we have no way to track which tasks are using
- * a page, though if it is a pagecache page, rmap structures can tell us
- * who is mapping it.
- */
-struct page {
-	unsigned long flags;		/* Atomic flags, some possibly
-					 * updated asynchronously */
-	atomic_t _count;		/* Usage count, see below. */
-	atomic_t _mapcount;		/* Count of ptes mapped in mms,
-					 * to show when page is mapped
-					 * & limit reverse map searches.
-					 */
-	union {
-	    struct {
-		unsigned long private;		/* Mapping-private opaque data:
-					 	 * usually used for buffer_heads
-						 * if PagePrivate set; used for
-						 * swp_entry_t if PageSwapCache;
-						 * indicates order in the buddy
-						 * system if PG_buddy is set.
-						 */
-		struct address_space *mapping;	/* If low bit clear, points to
-						 * inode address_space, or NULL.
-						 * If page mapped as anonymous
-						 * memory, low bit is set, and
-						 * it points to anon_vma object:
-						 * see PAGE_MAPPING_ANON below.
-						 */
-	    };
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
-	    spinlock_t ptl;
-#endif
-	};
-	pgoff_t index;			/* Our offset within mapping. */
-	struct list_head lru;		/* Pageout list, eg. active_list
-					 * protected by zone->lru_lock !
-					 */
-	/*
-	 * On machines where all RAM is mapped into kernel address space,
-	 * we can simply calculate the virtual address. On machines with
-	 * highmem some memory is mapped into kernel virtual memory
-	 * dynamically, so we need a place to store that address.
-	 * Note that this field could be 16 bits on x86 ... ;)
-	 *
-	 * Architectures with slow multiplication can define
-	 * WANT_PAGE_VIRTUAL in asm/page.h
-	 */
-#if defined(WANT_PAGE_VIRTUAL)
-	void *virtual;			/* Kernel virtual address (NULL if
-					   not kmapped, ie. highmem) */
-#endif /* WANT_PAGE_VIRTUAL */
-};
-
 #define page_private(page)		((page)->private)
 #define set_page_private(page, v)	((page)->private = (v))
 
@@ -546,11 +491,6 @@ static inline void set_page_links(struct page *page, enum zone_type zone,
  */
 #include <linux/vmstat.h>
 
-#ifndef CONFIG_DISCONTIGMEM
-/* The array of struct pages - for discontigmem use pgdat->lmem_map */
-extern struct page *mem_map;
-#endif
-
 static __always_inline void *lowmem_page_address(struct page *page)
 {
 	return __va(page_to_pfn(page) << PAGE_SHIFT);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
new file mode 100644
index 000000000000..c3852fd4a1cc
--- /dev/null
+++ b/include/linux/mm_types.h
@@ -0,0 +1,67 @@
+#ifndef _LINUX_MM_TYPES_H
+#define _LINUX_MM_TYPES_H
+
+#include <linux/types.h>
+#include <linux/threads.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+struct address_space;
+
+/*
+ * Each physical page in the system has a struct page associated with
+ * it to keep track of whatever it is we are using the page for at the
+ * moment. Note that we have no way to track which tasks are using
+ * a page, though if it is a pagecache page, rmap structures can tell us
+ * who is mapping it.
+ */
+struct page {
+	unsigned long flags;		/* Atomic flags, some possibly
+					 * updated asynchronously */
+	atomic_t _count;		/* Usage count, see below. */
+	atomic_t _mapcount;		/* Count of ptes mapped in mms,
+					 * to show when page is mapped
+					 * & limit reverse map searches.
+					 */
+	union {
+	    struct {
+		unsigned long private;		/* Mapping-private opaque data:
+					 	 * usually used for buffer_heads
+						 * if PagePrivate set; used for
+						 * swp_entry_t if PageSwapCache;
+						 * indicates order in the buddy
+						 * system if PG_buddy is set.
+						 */
+		struct address_space *mapping;	/* If low bit clear, points to
+						 * inode address_space, or NULL.
+						 * If page mapped as anonymous
+						 * memory, low bit is set, and
+						 * it points to anon_vma object:
+						 * see PAGE_MAPPING_ANON below.
+						 */
+	    };
+#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+	    spinlock_t ptl;
+#endif
+	};
+	pgoff_t index;			/* Our offset within mapping. */
+	struct list_head lru;		/* Pageout list, eg. active_list
+					 * protected by zone->lru_lock !
+					 */
+	/*
+	 * On machines where all RAM is mapped into kernel address space,
+	 * we can simply calculate the virtual address. On machines with
+	 * highmem some memory is mapped into kernel virtual memory
+	 * dynamically, so we need a place to store that address.
+	 * Note that this field could be 16 bits on x86 ... ;)
+	 *
+	 * Architectures with slow multiplication can define
+	 * WANT_PAGE_VIRTUAL in asm/page.h
+	 */
+#if defined(WANT_PAGE_VIRTUAL)
+	void *virtual;			/* Kernel virtual address (NULL if
+					   not kmapped, ie. highmem) */
+#endif /* WANT_PAGE_VIRTUAL */
+};
+
+#endif /* _LINUX_MM_TYPES_H */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 1b0680cd84d2..562cf7a8f3ee 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -314,6 +314,11 @@ struct node_active_region {
 };
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
+#ifndef CONFIG_DISCONTIGMEM
+/* The array of struct pages - for discontigmem use pgdat->lmem_map */
+extern struct page *mem_map;
+#endif
+
 /*
  * The pg_data_t structure is used in machines with CONFIG_DISCONTIGMEM
  * (mostly NUMA machines?) to denote a higher-level memory zone than the
-- 
cgit v1.2.3


From 08e0f6a9705376732fd3bc9bf8ba97a6b5211eb1 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:06 -0700
Subject: [PATCH] Add NUMA_BUILD definition in kernel.h to avoid #ifdef
 CONFIG_NUMA

The NUMA_BUILD constant is always available and will be set to 1 on
NUMA_BUILDs.  That way checks valid only under CONFIG_NUMA can easily be done
without #ifdef CONFIG_NUMA

F.e.

if (NUMA_BUILD && <numa_condition>) {
...
}

[akpm: not a thing we'd normally do, but CONFIG_NUMA is special: it is
 causing ifdef explosion in core kernel, so let's see if this is a comfortable
 way in whcih to control that]

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kernel.h |  7 +++++++
 mm/page_alloc.c        | 12 +++++-------
 2 files changed, 12 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 4fa373bb18ac..4d00988dad03 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -350,4 +350,11 @@ struct sysinfo {
 /* Trap pasters of __FUNCTION__ at compile-time */
 #define __FUNCTION__ (__func__)
 
+/* This helps us to avoid #ifdef CONFIG_NUMA */
+#ifdef CONFIG_NUMA
+#define NUMA_BUILD 1
+#else
+#define NUMA_BUILD 0
+#endif
+
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8a6418b0d2c5..4c76188b1681 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -942,7 +942,7 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order,
 	 */
 	do {
 		zone = *z;
-		if (unlikely((gfp_mask & __GFP_THISNODE) &&
+		if (unlikely(NUMA_BUILD && (gfp_mask & __GFP_THISNODE) &&
 			zone->zone_pgdat != zonelist->zones[0]->zone_pgdat))
 				break;
 		if ((alloc_flags & ALLOC_CPUSET) &&
@@ -1256,14 +1256,12 @@ unsigned int nr_free_pagecache_pages(void)
 {
 	return nr_free_zone_pages(gfp_zone(GFP_HIGHUSER));
 }
-#ifdef CONFIG_NUMA
-static void show_node(struct zone *zone)
+
+static inline void show_node(struct zone *zone)
 {
-	printk("Node %ld ", zone_to_nid(zone));
+	if (NUMA_BUILD)
+		printk("Node %ld ", zone_to_nid(zone));
 }
-#else
-#define show_node(zone)	do { } while (0)
-#endif
 
 void si_meminfo(struct sysinfo *val)
 {
-- 
cgit v1.2.3


From 77f700dab4c05f8ee17584ec869672796d7bcb87 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:07 -0700
Subject: [PATCH] Disable GFP_THISNODE in the non-NUMA case

GFP_THISNODE must be set to 0 in the non numa case otherwise we disable retry
and warnings for failing allocations in the SMP and UP case.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include')

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 8b34aabfe4c6..bf2b6bc3f6fd 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -67,7 +67,12 @@ struct vm_area_struct;
 #define GFP_HIGHUSER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
 			 __GFP_HIGHMEM)
 
+#ifdef CONFIG_NUMA
 #define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
+#else
+#define GFP_THISNODE	0
+#endif
+
 
 /* Flag - indicates that the buffer will be suitable for DMA.  Ignored on some
    platforms, used as appropriate on others */
-- 
cgit v1.2.3


From d5f541ed6e31518508c688912e7464facf253c87 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:08 -0700
Subject: [PATCH] Add node to zone for the NUMA case

Add the node in order to optimize zone_to_nid.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     | 6 +++++-
 include/linux/mmzone.h | 1 +
 mm/page_alloc.c        | 1 +
 3 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7477fb59c4f2..8e433bbc6e7e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -446,7 +446,11 @@ static inline struct zone *page_zone(struct page *page)
 
 static inline unsigned long zone_to_nid(struct zone *zone)
 {
-	return zone->zone_pgdat->node_id;
+#ifdef CONFIG_NUMA
+	return zone->node;
+#else
+	return 0;
+#endif
 }
 
 static inline unsigned long page_to_nid(struct page *page)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 562cf7a8f3ee..59855b8718a0 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -168,6 +168,7 @@ struct zone {
 	unsigned long		lowmem_reserve[MAX_NR_ZONES];
 
 #ifdef CONFIG_NUMA
+	int node;
 	/*
 	 * zone reclaim becomes active if more unmapped pages exist.
 	 */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4c76188b1681..d0432e44f77d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2405,6 +2405,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		zone->spanned_pages = size;
 		zone->present_pages = realsize;
 #ifdef CONFIG_NUMA
+		zone->node = nid;
 		zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio)
 						/ 100;
 		zone->min_slab_pages = (realsize * sysctl_min_slab_ratio) / 100;
-- 
cgit v1.2.3


From f4b81804a2d1ab341a4613089dc31ecce0800ed8 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Wed, 27 Sep 2006 01:50:10 -0700
Subject: [PATCH] do_no_pfn()

Implement do_no_pfn() for handling mapping of memory without a struct page
backing it.  This avoids creating fake page table entries for regions which
are not backed by real memory.

This feature is used by the MSPEC driver and other users, where it is
highly undesirable to have a struct page sitting behind the page (for
instance if the page is accessed in cached mode via the struct page in
parallel to the the driver accessing it uncached, which can result in data
corruption on some architectures, such as ia64).

This version uses specific NOPFN_{SIGBUS,OOM} return values, rather than
expect all negative pfn values would be an error.  It also bugs on cow
mappings as this would not work with the VM.

[akpm@osdl.org: micro-optimise]
Signed-off-by: Jes Sorensen <jes@sgi.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h |  7 ++++++
 mm/memory.c        | 64 +++++++++++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 66 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 8e433bbc6e7e..22165cb18906 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -199,6 +199,7 @@ struct vm_operations_struct {
 	void (*open)(struct vm_area_struct * area);
 	void (*close)(struct vm_area_struct * area);
 	struct page * (*nopage)(struct vm_area_struct * area, unsigned long address, int *type);
+	unsigned long (*nopfn)(struct vm_area_struct * area, unsigned long address);
 	int (*populate)(struct vm_area_struct * area, unsigned long address, unsigned long len, pgprot_t prot, unsigned long pgoff, int nonblock);
 
 	/* notification that a previously read-only page is about to become
@@ -593,6 +594,12 @@ static inline int page_mapped(struct page *page)
 #define NOPAGE_SIGBUS	(NULL)
 #define NOPAGE_OOM	((struct page *) (-1))
 
+/*
+ * Error return values for the *_nopfn functions
+ */
+#define NOPFN_SIGBUS	((unsigned long) -1)
+#define NOPFN_OOM	((unsigned long) -2)
+
 /*
  * Different kinds of faults, as returned by handle_mm_fault().
  * Used to decide whether a process gets delivered SIGBUS or
diff --git a/mm/memory.c b/mm/memory.c
index 92a3ebd8d795..f2ef1dcfff77 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2255,6 +2255,54 @@ oom:
 	return VM_FAULT_OOM;
 }
 
+/*
+ * do_no_pfn() tries to create a new page mapping for a page without
+ * a struct_page backing it
+ *
+ * As this is called only for pages that do not currently exist, we
+ * do not need to flush old virtual caches or the TLB.
+ *
+ * We enter with non-exclusive mmap_sem (to exclude vma changes,
+ * but allow concurrent faults), and pte mapped but not yet locked.
+ * We return with mmap_sem still held, but pte unmapped and unlocked.
+ *
+ * It is expected that the ->nopfn handler always returns the same pfn
+ * for a given virtual mapping.
+ *
+ * Mark this `noinline' to prevent it from bloating the main pagefault code.
+ */
+static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma,
+		     unsigned long address, pte_t *page_table, pmd_t *pmd,
+		     int write_access)
+{
+	spinlock_t *ptl;
+	pte_t entry;
+	unsigned long pfn;
+	int ret = VM_FAULT_MINOR;
+
+	pte_unmap(page_table);
+	BUG_ON(!(vma->vm_flags & VM_PFNMAP));
+	BUG_ON(is_cow_mapping(vma->vm_flags));
+
+	pfn = vma->vm_ops->nopfn(vma, address & PAGE_MASK);
+	if (pfn == NOPFN_OOM)
+		return VM_FAULT_OOM;
+	if (pfn == NOPFN_SIGBUS)
+		return VM_FAULT_SIGBUS;
+
+	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
+
+	/* Only go through if we didn't race with anybody else... */
+	if (pte_none(*page_table)) {
+		entry = pfn_pte(pfn, vma->vm_page_prot);
+		if (write_access)
+			entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+		set_pte_at(mm, address, page_table, entry);
+	}
+	pte_unmap_unlock(page_table, ptl);
+	return ret;
+}
+
 /*
  * Fault of a previously existing named mapping. Repopulate the pte
  * from the encoded file_pte if possible. This enables swappable
@@ -2317,11 +2365,17 @@ static inline int handle_pte_fault(struct mm_struct *mm,
 	old_entry = entry = *pte;
 	if (!pte_present(entry)) {
 		if (pte_none(entry)) {
-			if (!vma->vm_ops || !vma->vm_ops->nopage)
-				return do_anonymous_page(mm, vma, address,
-					pte, pmd, write_access);
-			return do_no_page(mm, vma, address,
-					pte, pmd, write_access);
+			if (vma->vm_ops) {
+				if (vma->vm_ops->nopage)
+					return do_no_page(mm, vma, address,
+							  pte, pmd,
+							  write_access);
+				if (unlikely(vma->vm_ops->nopfn))
+					return do_no_pfn(mm, vma, address, pte,
+							 pmd, write_access);
+			}
+			return do_anonymous_page(mm, vma, address,
+						 pte, pmd, write_access);
 		}
 		if (pte_file(entry))
 			return do_file_page(mm, vma, address,
-- 
cgit v1.2.3


From d24afc57d51b1be41f95521e81399061fa5875a6 Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Wed, 27 Sep 2006 01:50:13 -0700
Subject: [PATCH] Mark __remove_vm_area() static

The function is exported but not used from anywhere else.  It's also marked as
"not for driver use" so noone out there should really care.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/vmalloc.h | 1 -
 mm/vmalloc.c            | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index dee88c6b6fa7..ce5f1482e6be 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -62,7 +62,6 @@ extern struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags,
 extern struct vm_struct *get_vm_area_node(unsigned long size,
 					unsigned long flags, int node);
 extern struct vm_struct *remove_vm_area(void *addr);
-extern struct vm_struct *__remove_vm_area(void *addr);
 extern int map_vm_area(struct vm_struct *area, pgprot_t prot,
 			struct page ***pages);
 extern void unmap_vm_area(struct vm_struct *area);
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 659ec634856a..1ac191ce5641 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -272,7 +272,7 @@ static struct vm_struct *__find_vm_area(void *addr)
 }
 
 /* Caller must hold vmlist_lock */
-struct vm_struct *__remove_vm_area(void *addr)
+static struct vm_struct *__remove_vm_area(void *addr)
 {
 	struct vm_struct **p, *tmp;
 
-- 
cgit v1.2.3


From 5da6185bca064e35aa73a7c1f27488d2b96434f4 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:16 -0700
Subject: [PATCH] NOMMU: Set BDI capabilities for /dev/mem and /dev/kmem

Set the backing device info capabilities for /dev/mem and /dev/kmem to
permit direct sharing under no-MMU conditions and full mapping capabilities
under MMU conditions.  Make the BDI used by these available to all directly
mappable character devices.

Also comment the capabilities for /dev/zero.

[akpm@osdl.org: ifdef reductions]
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/mem.c   | 39 +++++++++++++++++++++++++++++++++++++++
 fs/char_dev.c        | 20 ++++++++++++++++++++
 include/linux/cdev.h |  2 ++
 3 files changed, 61 insertions(+)

(limited to 'include')

diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index 917b20402664..4ac70ec697f0 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -238,6 +238,32 @@ static pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
 }
 #endif
 
+#ifndef CONFIG_MMU
+static unsigned long get_unmapped_area_mem(struct file *file,
+					   unsigned long addr,
+					   unsigned long len,
+					   unsigned long pgoff,
+					   unsigned long flags)
+{
+	if (!valid_mmap_phys_addr_range(pgoff, len))
+		return (unsigned long) -EINVAL;
+	return pgoff;
+}
+
+/* can't do an in-place private mapping if there's no MMU */
+static inline int private_mapping_ok(struct vm_area_struct *vma)
+{
+	return vma->vm_flags & VM_MAYSHARE;
+}
+#else
+#define get_unmapped_area_mem	NULL
+
+static inline int private_mapping_ok(struct vm_area_struct *vma)
+{
+	return 1;
+}
+#endif
+
 static int mmap_mem(struct file * file, struct vm_area_struct * vma)
 {
 	size_t size = vma->vm_end - vma->vm_start;
@@ -245,6 +271,9 @@ static int mmap_mem(struct file * file, struct vm_area_struct * vma)
 	if (!valid_mmap_phys_addr_range(vma->vm_pgoff, size))
 		return -EINVAL;
 
+	if (!private_mapping_ok(vma))
+		return -ENOSYS;
+
 	vma->vm_page_prot = phys_mem_access_prot(file, vma->vm_pgoff,
 						 size,
 						 vma->vm_page_prot);
@@ -782,6 +811,7 @@ static const struct file_operations mem_fops = {
 	.write		= write_mem,
 	.mmap		= mmap_mem,
 	.open		= open_mem,
+	.get_unmapped_area = get_unmapped_area_mem,
 };
 
 static const struct file_operations kmem_fops = {
@@ -790,6 +820,7 @@ static const struct file_operations kmem_fops = {
 	.write		= write_kmem,
 	.mmap		= mmap_kmem,
 	.open		= open_kmem,
+	.get_unmapped_area = get_unmapped_area_mem,
 };
 
 static const struct file_operations null_fops = {
@@ -815,6 +846,10 @@ static const struct file_operations zero_fops = {
 	.mmap		= mmap_zero,
 };
 
+/*
+ * capabilities for /dev/zero
+ * - permits private mappings, "copies" are taken of the source of zeros
+ */
 static struct backing_dev_info zero_bdi = {
 	.capabilities	= BDI_CAP_MAP_COPY,
 };
@@ -862,9 +897,13 @@ static int memory_open(struct inode * inode, struct file * filp)
 	switch (iminor(inode)) {
 		case 1:
 			filp->f_op = &mem_fops;
+			filp->f_mapping->backing_dev_info =
+				&directly_mappable_cdev_bdi;
 			break;
 		case 2:
 			filp->f_op = &kmem_fops;
+			filp->f_mapping->backing_dev_info =
+				&directly_mappable_cdev_bdi;
 			break;
 		case 3:
 			filp->f_op = &null_fops;
diff --git a/fs/char_dev.c b/fs/char_dev.c
index 3483d3cf8087..0009346d827f 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -19,11 +19,30 @@
 #include <linux/kobj_map.h>
 #include <linux/cdev.h>
 #include <linux/mutex.h>
+#include <linux/backing-dev.h>
 
 #ifdef CONFIG_KMOD
 #include <linux/kmod.h>
 #endif
 
+/*
+ * capabilities for /dev/mem, /dev/kmem and similar directly mappable character
+ * devices
+ * - permits shared-mmap for read, write and/or exec
+ * - does not permit private mmap in NOMMU mode (can't do COW)
+ * - no readahead or I/O queue unplugging required
+ */
+struct backing_dev_info directly_mappable_cdev_bdi = {
+	.capabilities	= (
+#ifdef CONFIG_MMU
+		/* permit private copies of the data to be taken */
+		BDI_CAP_MAP_COPY |
+#endif
+		/* permit direct mmap, for read, write or exec */
+		BDI_CAP_MAP_DIRECT |
+		BDI_CAP_READ_MAP | BDI_CAP_WRITE_MAP | BDI_CAP_EXEC_MAP),
+};
+
 static struct kobj_map *cdev_map;
 
 static DEFINE_MUTEX(chrdevs_lock);
@@ -461,3 +480,4 @@ EXPORT_SYMBOL(cdev_del);
 EXPORT_SYMBOL(cdev_add);
 EXPORT_SYMBOL(register_chrdev);
 EXPORT_SYMBOL(unregister_chrdev);
+EXPORT_SYMBOL(directly_mappable_cdev_bdi);
diff --git a/include/linux/cdev.h b/include/linux/cdev.h
index 2216638962d2..ee5f53f2ca15 100644
--- a/include/linux/cdev.h
+++ b/include/linux/cdev.h
@@ -23,5 +23,7 @@ void cdev_del(struct cdev *);
 
 void cd_forget(struct inode *);
 
+extern struct backing_dev_info directly_mappable_cdev_bdi;
+
 #endif
 #endif
-- 
cgit v1.2.3


From dbf8685c8e21404e3a8ed244bd0219d3c4b89101 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:19 -0700
Subject: [PATCH] NOMMU: Implement /proc/pid/maps for NOMMU

Implement /proc/pid/maps for NOMMU by reading the vm_area_list attached to
current->mm->context.vmlist.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/nommu-mmap.txt |  3 ++
 fs/proc/internal.h           |  1 +
 fs/proc/nommu.c              | 20 ++++++++----
 fs/proc/task_nommu.c         | 74 +++++++++++++++++++++++++++++++++++---------
 include/linux/proc_fs.h      |  2 ++
 5 files changed, 80 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index b88ebe4d808c..83f4b083d57e 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -116,6 +116,9 @@ FURTHER NOTES ON NO-MMU MMAP
  (*) A list of all the mappings on the system is visible through /proc/maps in
      no-MMU mode.
 
+ (*) A list of all the mappings in use by a process is visible through
+     /proc/<pid>/maps in no-MMU mode.
+
  (*) Supplying MAP_FIXED or a requesting a particular mapping address will
      result in an error.
 
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 146a434ba944..987c773dbb20 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -28,6 +28,7 @@ do {						\
 	(vmi)->largest_chunk = 0;		\
 } while(0)
 
+extern int nommu_vma_show(struct seq_file *, struct vm_area_struct *);
 #endif
 
 extern void create_seq_entry(char *name, mode_t mode, const struct file_operations *f);
diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
index cff10ab1af63..d7dbdf9e0f49 100644
--- a/fs/proc/nommu.c
+++ b/fs/proc/nommu.c
@@ -33,19 +33,15 @@
 #include "internal.h"
 
 /*
- * display a list of all the VMAs the kernel knows about
- * - nommu kernals have a single flat list
+ * display a single VMA to a sequenced file
  */
-static int nommu_vma_list_show(struct seq_file *m, void *v)
+int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
 {
-	struct vm_area_struct *vma;
 	unsigned long ino = 0;
 	struct file *file;
 	dev_t dev = 0;
 	int flags, len;
 
-	vma = rb_entry((struct rb_node *) v, struct vm_area_struct, vm_rb);
-
 	flags = vma->vm_flags;
 	file = vma->vm_file;
 
@@ -78,6 +74,18 @@ static int nommu_vma_list_show(struct seq_file *m, void *v)
 	return 0;
 }
 
+/*
+ * display a list of all the VMAs the kernel knows about
+ * - nommu kernals have a single flat list
+ */
+static int nommu_vma_list_show(struct seq_file *m, void *v)
+{
+	struct vm_area_struct *vma;
+
+	vma = rb_entry((struct rb_node *) v, struct vm_area_struct, vm_rb);
+	return nommu_vma_show(m, vma);
+}
+
 static void *nommu_vma_list_start(struct seq_file *m, loff_t *_pos)
 {
 	struct rb_node *_rb;
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 4616ed50ffcd..091aa8e48e02 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -138,25 +138,63 @@ out:
 }
 
 /*
- * Albert D. Cahalan suggested to fake entries for the traditional
- * sections here.  This might be worth investigating.
+ * display mapping lines for a particular process's /proc/pid/maps
  */
-static int show_map(struct seq_file *m, void *v)
+static int show_map(struct seq_file *m, void *_vml)
 {
-	return 0;
+	struct vm_list_struct *vml = _vml;
+	return nommu_vma_show(m, vml->vma);
 }
+
 static void *m_start(struct seq_file *m, loff_t *pos)
 {
+	struct proc_maps_private *priv = m->private;
+	struct vm_list_struct *vml;
+	struct mm_struct *mm;
+	loff_t n = *pos;
+
+	/* pin the task and mm whilst we play with them */
+	priv->task = get_pid_task(priv->pid, PIDTYPE_PID);
+	if (!priv->task)
+		return NULL;
+
+	mm = get_task_mm(priv->task);
+	if (!mm) {
+		put_task_struct(priv->task);
+		priv->task = NULL;
+		return NULL;
+	}
+
+	down_read(&mm->mmap_sem);
+
+	/* start from the Nth VMA */
+	for (vml = mm->context.vmlist; vml; vml = vml->next)
+		if (n-- == 0)
+			return vml;
 	return NULL;
 }
-static void m_stop(struct seq_file *m, void *v)
+
+static void m_stop(struct seq_file *m, void *_vml)
 {
+	struct proc_maps_private *priv = m->private;
+
+	if (priv->task) {
+		struct mm_struct *mm = priv->task->mm;
+		up_read(&mm->mmap_sem);
+		mmput(mm);
+		put_task_struct(priv->task);
+	}
 }
-static void *m_next(struct seq_file *m, void *v, loff_t *pos)
+
+static void *m_next(struct seq_file *m, void *_vml, loff_t *pos)
 {
-	return NULL;
+	struct vm_list_struct *vml = _vml;
+
+	(*pos)++;
+	return vml ? vml->next : NULL;
 }
-static struct seq_operations proc_pid_maps_op = {
+
+static struct seq_operations proc_pid_maps_ops = {
 	.start	= m_start,
 	.next	= m_next,
 	.stop	= m_stop,
@@ -165,11 +203,19 @@ static struct seq_operations proc_pid_maps_op = {
 
 static int maps_open(struct inode *inode, struct file *file)
 {
-	int ret;
-	ret = seq_open(file, &proc_pid_maps_op);
-	if (!ret) {
-		struct seq_file *m = file->private_data;
-		m->private = NULL;
+	struct proc_maps_private *priv;
+	int ret = -ENOMEM;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (priv) {
+		priv->pid = proc_pid(inode);
+		ret = seq_open(file, &proc_pid_maps_ops);
+		if (!ret) {
+			struct seq_file *m = file->private_data;
+			m->private = priv;
+		} else {
+			kfree(priv);
+		}
 	}
 	return ret;
 }
@@ -178,6 +224,6 @@ struct file_operations proc_maps_operations = {
 	.open		= maps_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= seq_release,
+	.release	= seq_release_private,
 };
 
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index 3435ca38dd14..57f70bc8b24b 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -268,7 +268,9 @@ static inline struct proc_dir_entry *PDE(const struct inode *inode)
 struct proc_maps_private {
 	struct pid *pid;
 	struct task_struct *task;
+#ifdef CONFIG_MMU
 	struct vm_area_struct *tail_vma;
+#endif
 };
 
 #endif /* _LINUX_PROC_FS_H */
-- 
cgit v1.2.3


From f269fdd1829acc5e53bf57b145003e5733133f2b Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:23 -0700
Subject: [PATCH] NOMMU: move the fallback arch_vma_name() to a sensible place

Move the fallback arch_vma_name() to a sensible place (kernel/signal.c).

Currently it's in fs/proc/task_mmu.c, a file that is dependent on both
CONFIG_PROC_FS and CONFIG_MMU being enabled, but it's used from
kernel/signal.c from where it is called unconditionally.

[akpm@osdl.org: build fix]
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/proc/task_mmu.c | 5 -----
 include/linux/mm.h | 2 +-
 kernel/signal.c    | 5 +++++
 3 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 0a163a4f7764..6b769afac55a 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -122,11 +122,6 @@ struct mem_size_stats
 	unsigned long private_dirty;
 };
 
-__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma)
-{
-	return NULL;
-}
-
 static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats *mss)
 {
 	struct proc_maps_private *priv = m->private;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 22165cb18906..7b703b6d4358 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1131,7 +1131,7 @@ void drop_slab(void);
 extern int randomize_va_space;
 #endif
 
-const char *arch_vma_name(struct vm_area_struct *vma);
+__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma);
 
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */
diff --git a/kernel/signal.c b/kernel/signal.c
index bfdb5686fa3e..05853a7337e3 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2577,6 +2577,11 @@ asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize)
 }
 #endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */
 
+__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma)
+{
+	return NULL;
+}
+
 void __init signals_init(void)
 {
 	sigqueue_cachep =
-- 
cgit v1.2.3


From a27f3113322edff36743014cc9e752a21ffc0324 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Wed, 27 Sep 2006 01:50:24 -0700
Subject: [PATCH] m32r: Fix "value computed not used" warnings

Fix to remove annoying gcc-4.1 warnings "value computed not used" for m32r;
Modify set_mb to cast to void for SMP.

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/system.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-m32r/system.h b/include/asm-m32r/system.h
index 9e618afec6ed..4ce0619f6989 100644
--- a/include/asm-m32r/system.h
+++ b/include/asm-m32r/system.h
@@ -328,15 +328,15 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 #define smp_rmb()	rmb()
 #define smp_wmb()	wmb()
 #define smp_read_barrier_depends()	read_barrier_depends()
+#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
 #else
 #define smp_mb()	barrier()
 #define smp_rmb()	barrier()
 #define smp_wmb()	barrier()
 #define smp_read_barrier_depends()	do { } while (0)
+#define set_mb(var, value) do { var = value; barrier(); } while (0)
 #endif
 
-#define set_mb(var, value) do { xchg(&var, value); } while (0)
-
 #define arch_align_stack(x) (x)
 
 #endif  /* _ASM_M32R_SYSTEM_H */
-- 
cgit v1.2.3


From 85f651794c46e8e3faf204a767d1caa7f9f278f0 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata.hirokazu@renesas.com>
Date: Wed, 27 Sep 2006 01:50:24 -0700
Subject: [PATCH] m32r: revise __raw_read_trylock()

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Cc: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/spinlock.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-m32r/spinlock.h b/include/asm-m32r/spinlock.h
index f94c1a673569..f9f90727a4a1 100644
--- a/include/asm-m32r/spinlock.h
+++ b/include/asm-m32r/spinlock.h
@@ -298,7 +298,14 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw)
 	);
 }
 
-#define __raw_read_trylock(lock) generic__raw_read_trylock(lock)
+static inline int __raw_read_trylock(raw_rwlock_t *lock)
+{
+	atomic_t *count = (atomic_t*)lock;
+	if (atomic_dec_return(count) >= 0)
+		return 1;
+	atomic_inc(count);
+	return 0;
+}
 
 static inline int __raw_write_trylock(raw_rwlock_t *lock)
 {
-- 
cgit v1.2.3


From a8b4fc4d7c3ccf80d4fa1805cee85c06c2aa653e Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:35 -0700
Subject: [PATCH] uml: fix missing x86_64 register definitions

The UML/x86_64 headers were missing ptrace support for some segment registers.
 The underlying problem was that the x86_64 kernel uses user_regs_struct
rather than the ptrace register definitions in ptrace.  This patch switches
UML/x86_64 to using user_regs_struct for its definitions of the host's
registers.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/sysdep-x86_64/ptrace.h | 43 +++++++++++++++++++++++++++++++---
 arch/um/include/sysdep-x86_64/sc.h     |  2 +-
 include/asm-um/ptrace-x86_64.h         |  3 +++
 3 files changed, 44 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/arch/um/include/sysdep-x86_64/ptrace.h b/arch/um/include/sysdep-x86_64/ptrace.h
index 8d353f0feec1..617bb9efc934 100644
--- a/arch/um/include/sysdep-x86_64/ptrace.h
+++ b/arch/um/include/sysdep-x86_64/ptrace.h
@@ -50,6 +50,21 @@
 #define HOST_FS 25
 #define HOST_GS 26
 
+/* Also defined in asm/ptrace-x86_64.h, but not in libc headers.  So, these
+ * are already defined for kernel code, but not for userspace code.
+ */
+#ifndef FS_BASE
+/* These aren't defined in ptrace.h, but exist in struct user_regs_struct,
+ * which is what x86_64 ptrace actually uses.
+ */
+#define FS_BASE (HOST_FS_BASE * sizeof(long))
+#define GS_BASE (HOST_GS_BASE * sizeof(long))
+#define DS (HOST_DS * sizeof(long))
+#define ES (HOST_ES * sizeof(long))
+#define FS (HOST_FS * sizeof(long))
+#define GS (HOST_GS * sizeof(long))
+#endif
+
 #define REGS_FS_BASE(r) ((r)[HOST_FS_BASE])
 #define REGS_GS_BASE(r) ((r)[HOST_GS_BASE])
 #define REGS_DS(r) ((r)[HOST_DS])
@@ -89,9 +104,12 @@ union uml_pt_regs {
 #endif
 #ifdef UML_CONFIG_MODE_SKAS
 	struct skas_regs {
-		/* XXX */
-		unsigned long regs[27];
-		unsigned long fp[65];
+		/* x86_64 ptrace uses sizeof(user_regs_struct) as its register
+		 * file size, while i386 uses FRAME_SIZE.  Therefore, we need
+		 * to use UM_FRAME_SIZE here instead of HOST_FRAME_SIZE.
+		 */
+		unsigned long regs[UM_FRAME_SIZE];
+		unsigned long fp[HOST_FP_SIZE];
                 struct faultinfo faultinfo;
 		long syscall;
 		int is_user;
@@ -120,11 +138,16 @@ extern int mode_tt;
 #define UPT_R14(r) __CHOOSE_MODE(SC_R14(UPT_SC(r)), REGS_R14((r)->skas.regs))
 #define UPT_R15(r) __CHOOSE_MODE(SC_R15(UPT_SC(r)), REGS_R15((r)->skas.regs))
 #define UPT_CS(r) __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs))
+#define UPT_FS_BASE(r) \
+	__CHOOSE_MODE(SC_FS_BASE(UPT_SC(r)), REGS_FS_BASE((r)->skas.regs))
 #define UPT_FS(r) __CHOOSE_MODE(SC_FS(UPT_SC(r)), REGS_FS((r)->skas.regs))
+#define UPT_GS_BASE(r) \
+	__CHOOSE_MODE(SC_GS_BASE(UPT_SC(r)), REGS_GS_BASE((r)->skas.regs))
 #define UPT_GS(r) __CHOOSE_MODE(SC_GS(UPT_SC(r)), REGS_GS((r)->skas.regs))
 #define UPT_DS(r) __CHOOSE_MODE(SC_DS(UPT_SC(r)), REGS_DS((r)->skas.regs))
 #define UPT_ES(r) __CHOOSE_MODE(SC_ES(UPT_SC(r)), REGS_ES((r)->skas.regs))
 #define UPT_CS(r) __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs))
+#define UPT_SS(r) __CHOOSE_MODE(SC_SS(UPT_SC(r)), REGS_SS((r)->skas.regs))
 #define UPT_ORIG_RAX(r) \
 	__CHOOSE_MODE((r)->tt.orig_rax, REGS_ORIG_RAX((r)->skas.regs))
 
@@ -183,6 +206,13 @@ struct syscall_args {
                 case RBP: val = UPT_RBP(regs); break; \
                 case ORIG_RAX: val = UPT_ORIG_RAX(regs); break; \
                 case CS: val = UPT_CS(regs); break; \
+                case SS: val = UPT_SS(regs); break; \
+		case FS_BASE: val = UPT_FS_BASE(regs); break; \
+                case GS_BASE: val = UPT_GS_BASE(regs); break; \
+                case DS: val = UPT_DS(regs); break; \
+                case ES: val = UPT_ES(regs); break; \
+                case FS : val = UPT_FS (regs); break; \
+		case GS: val = UPT_GS(regs); break;	    \
                 case EFLAGS: val = UPT_EFLAGS(regs); break; \
                 default :  \
                         panic("Bad register in UPT_REG : %d\n", reg);  \
@@ -214,6 +244,13 @@ struct syscall_args {
                 case RBP: UPT_RBP(regs) = __upt_val; break; \
                 case ORIG_RAX: UPT_ORIG_RAX(regs) = __upt_val; break; \
                 case CS: UPT_CS(regs) = __upt_val; break; \
+                case SS: UPT_SS(regs) = __upt_val; break; \
+                case FS_BASE: UPT_FS_BASE(regs) = __upt_val; break; \
+                case GS_BASE: UPT_GS_BASE(regs) = __upt_val; break; \
+                case DS: UPT_DS(regs) = __upt_val; break; \
+                case ES: UPT_ES(regs) = __upt_val; break; \
+                case FS: UPT_FS(regs) = __upt_val; break; \
+                case GS: UPT_GS(regs) = __upt_val; break; \
                 case EFLAGS: UPT_EFLAGS(regs) = __upt_val; break; \
                 default :  \
                         panic("Bad register in UPT_SET : %d\n", reg);  \
diff --git a/arch/um/include/sysdep-x86_64/sc.h b/arch/um/include/sysdep-x86_64/sc.h
index a160d9fcc596..8aee45b07434 100644
--- a/arch/um/include/sysdep-x86_64/sc.h
+++ b/arch/um/include/sysdep-x86_64/sc.h
@@ -35,11 +35,11 @@
 #define SC_GS(sc) SC_OFFSET(sc, SC_GS)
 #define SC_EFLAGS(sc) SC_OFFSET(sc, SC_EFLAGS)
 #define SC_SIGMASK(sc) SC_OFFSET(sc, SC_SIGMASK)
+#define SC_SS(sc) SC_OFFSET(sc, SC_SS)
 #if 0
 #define SC_ORIG_RAX(sc) SC_OFFSET(sc, SC_ORIG_RAX)
 #define SC_DS(sc) SC_OFFSET(sc, SC_DS)
 #define SC_ES(sc) SC_OFFSET(sc, SC_ES)
-#define SC_SS(sc) SC_OFFSET(sc, SC_SS)
 #endif
 
 #endif
diff --git a/include/asm-um/ptrace-x86_64.h b/include/asm-um/ptrace-x86_64.h
index 2074483e6ca4..03b4af4ac09a 100644
--- a/include/asm-um/ptrace-x86_64.h
+++ b/include/asm-um/ptrace-x86_64.h
@@ -16,12 +16,15 @@
 
 #define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64
 
+/* Also defined in sysdep/ptrace.h, so may already be defined. */
+#ifndef FS_BASE
 #define FS_BASE (21 * sizeof(unsigned long))
 #define GS_BASE (22 * sizeof(unsigned long))
 #define DS (23 * sizeof(unsigned long))
 #define ES (24 * sizeof(unsigned long))
 #define FS (25 * sizeof(unsigned long))
 #define GS (26 * sizeof(unsigned long))
+#endif
 
 #define PT_REGS_RBX(r) UPT_RBX(&(r)->regs)
 #define PT_REGS_RCX(r) UPT_RCX(&(r)->regs)
-- 
cgit v1.2.3


From 3c9173509985b957bea692ea887a8a0e5055cfe8 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:40 -0700
Subject: [PATCH] uml: thread creation tidying

fork on UML has always somewhat subtle.  The underlying cause has been the
need to initialize a stack for the new process.  The only portable way to
initialize a new stack is to set it as the alternate signal stack and take a
signal.  The signal handler does whatever initialization is needed and jumps
back to the original stack, where the fork processing is finished.  The basic
context switching mechanism is a jmp_buf for each process.  You switch to a
new process by longjmping to its jmp_buf.

Now that UML has its own implementation of setjmp and longjmp, and I can poke
around inside a jmp_buf without fear that libc will change the structure, a
much simpler mechanism is possible.  The jmpbuf can simply be initialized by
hand.

This eliminates -
	the need to set up and remove the alternate signal stack
	sending and handling a signal
	the signal blocking needed around the stack switching, since
there is no stack switching
	setting up the jmp_buf needed to jump back to the original
stack after the new one is set up

In addition, since jmp_buf is now defined by UML, and not by libc, it can be
embedded in the thread struct.  This makes it unnecessary to have it exist on
the stack, where it used to be.  It also simplifies interfaces, since the
switch jmp_buf used to be a void * inside the thread struct, and functions
which took it as an argument needed to define a jmp_buf variable and assign it
from the void *.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/os.h                       | 10 ++---
 arch/um/include/skas/skas.h                |  3 +-
 arch/um/include/sysdep-i386/archsetjmp.h   |  3 ++
 arch/um/include/sysdep-x86_64/archsetjmp.h |  3 ++
 arch/um/kernel/skas/process_kern.c         | 36 ++++++----------
 arch/um/os-Linux/skas/process.c            | 68 +++++++-----------------------
 include/asm-um/processor-generic.h         |  6 +--
 7 files changed, 43 insertions(+), 86 deletions(-)

(limited to 'include')

diff --git a/arch/um/include/os.h b/arch/um/include/os.h
index be71b3d4c961..120ca21a513a 100644
--- a/arch/um/include/os.h
+++ b/arch/um/include/os.h
@@ -14,6 +14,7 @@
 #include "skas/mm_id.h"
 #include "irq_user.h"
 #include "sysdep/tls.h"
+#include "sysdep/archsetjmp.h"
 
 #define OS_TYPE_FILE 1
 #define OS_TYPE_DIR 2
@@ -308,12 +309,9 @@ extern int copy_context_skas0(unsigned long stack, int pid);
 extern void userspace(union uml_pt_regs *regs);
 extern void map_stub_pages(int fd, unsigned long code,
 			   unsigned long data, unsigned long stack);
-extern void new_thread(void *stack, void **switch_buf_ptr,
-			 void **fork_buf_ptr, void (*handler)(int));
-extern void thread_wait(void *sw, void *fb);
-extern void switch_threads(void *me, void *next);
-extern int start_idle_thread(void *stack, void *switch_buf_ptr,
-			     void **fork_buf_ptr);
+extern void new_thread(void *stack, jmp_buf *buf, void (*handler)(void));
+extern void switch_threads(jmp_buf *me, jmp_buf *you);
+extern int start_idle_thread(void *stack, jmp_buf *switch_buf);
 extern void initial_thread_cb_skas(void (*proc)(void *),
 				 void *arg);
 extern void halt_skas(void);
diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h
index 853b26f148c5..e88926b16072 100644
--- a/arch/um/include/skas/skas.h
+++ b/arch/um/include/skas/skas.h
@@ -14,8 +14,7 @@ extern int proc_mm, ptrace_faultinfo, ptrace_ldt;
 extern int skas_needs_stub;
 
 extern int user_thread(unsigned long stack, int flags);
-extern void new_thread_proc(void *stack, void (*handler)(int sig));
-extern void new_thread_handler(int sig);
+extern void new_thread_handler(void);
 extern void handle_syscall(union uml_pt_regs *regs);
 extern int new_mm(unsigned long stack);
 extern void get_skas_faultinfo(int pid, struct faultinfo * fi);
diff --git a/arch/um/include/sysdep-i386/archsetjmp.h b/arch/um/include/sysdep-i386/archsetjmp.h
index ea1ba3d42aee..11bafab669e9 100644
--- a/arch/um/include/sysdep-i386/archsetjmp.h
+++ b/arch/um/include/sysdep-i386/archsetjmp.h
@@ -16,4 +16,7 @@ struct __jmp_buf {
 
 typedef struct __jmp_buf jmp_buf[1];
 
+#define JB_IP __eip
+#define JB_SP __esp
+
 #endif				/* _SETJMP_H */
diff --git a/arch/um/include/sysdep-x86_64/archsetjmp.h b/arch/um/include/sysdep-x86_64/archsetjmp.h
index 454fc60aff6d..9a5e1a6ec800 100644
--- a/arch/um/include/sysdep-x86_64/archsetjmp.h
+++ b/arch/um/include/sysdep-x86_64/archsetjmp.h
@@ -18,4 +18,7 @@ struct __jmp_buf {
 
 typedef struct __jmp_buf jmp_buf[1];
 
+#define JB_IP __rip
+#define JB_SP __rsp
+
 #endif				/* _SETJMP_H */
diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c
index 55caeec8b257..c73d47c57abe 100644
--- a/arch/um/kernel/skas/process_kern.c
+++ b/arch/um/kernel/skas/process_kern.c
@@ -33,7 +33,7 @@ void switch_to_skas(void *prev, void *next)
 		switch_timers(0);
 
 	switch_threads(&from->thread.mode.skas.switch_buf,
-		       to->thread.mode.skas.switch_buf);
+		       &to->thread.mode.skas.switch_buf);
 
 	arch_switch_to_skas(current->thread.prev_sched, current);
 
@@ -43,21 +43,21 @@ void switch_to_skas(void *prev, void *next)
 
 extern void schedule_tail(struct task_struct *prev);
 
-void new_thread_handler(int sig)
+/* This is called magically, by its address being stuffed in a jmp_buf
+ * and being longjmp-d to.
+ */
+void new_thread_handler(void)
 {
 	int (*fn)(void *), n;
 	void *arg;
 
-	fn = current->thread.request.u.thread.proc;
-	arg = current->thread.request.u.thread.arg;
-	os_usr1_signal(1);
-	thread_wait(&current->thread.mode.skas.switch_buf,
-		    current->thread.mode.skas.fork_buf);
-
 	if(current->thread.prev_sched != NULL)
 		schedule_tail(current->thread.prev_sched);
 	current->thread.prev_sched = NULL;
 
+	fn = current->thread.request.u.thread.proc;
+	arg = current->thread.request.u.thread.arg;
+
 	/* The return value is 1 if the kernel thread execs a process,
 	 * 0 if it just exits
 	 */
@@ -70,22 +70,13 @@ void new_thread_handler(int sig)
 	else do_exit(0);
 }
 
-void new_thread_proc(void *stack, void (*handler)(int sig))
-{
-	init_new_thread_stack(stack, handler);
-	os_usr1_process(os_getpid());
-}
-
 void release_thread_skas(struct task_struct *task)
 {
 }
 
-void fork_handler(int sig)
+/* Called magically, see new_thread_handler above */
+void fork_handler(void)
 {
-	os_usr1_signal(1);
-	thread_wait(&current->thread.mode.skas.switch_buf,
-		    current->thread.mode.skas.fork_buf);
-  	
 	force_flush_all();
 	if(current->thread.prev_sched == NULL)
 		panic("blech");
@@ -109,7 +100,7 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
 		     unsigned long stack_top, struct task_struct * p,
 		     struct pt_regs *regs)
 {
-  	void (*handler)(int);
+  	void (*handler)(void);
 
 	if(current->thread.forking){
 	  	memcpy(&p->thread.regs.regs.skas, &regs->regs.skas,
@@ -128,7 +119,7 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
 	}
 
 	new_thread(task_stack_page(p), &p->thread.mode.skas.switch_buf,
-		   &p->thread.mode.skas.fork_buf, handler);
+		   handler);
 	return(0);
 }
 
@@ -182,8 +173,7 @@ int start_uml_skas(void)
 	init_task.thread.request.u.thread.proc = start_kernel_proc;
 	init_task.thread.request.u.thread.arg = NULL;
 	return(start_idle_thread(task_stack_page(&init_task),
-				 &init_task.thread.mode.skas.switch_buf,
-				 &init_task.thread.mode.skas.fork_buf));
+				 &init_task.thread.mode.skas.switch_buf));
 }
 
 int external_pid_skas(struct task_struct *task)
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 42e3d1ed802c..cb9ab54146cc 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -444,56 +444,22 @@ void map_stub_pages(int fd, unsigned long code,
 	}
 }
 
-void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr,
-		void (*handler)(int))
+void new_thread(void *stack, jmp_buf *buf, void (*handler)(void))
 {
-	unsigned long flags;
-	jmp_buf switch_buf, fork_buf;
-
-	*switch_buf_ptr = &switch_buf;
-	*fork_buf_ptr = &fork_buf;
-
-	/* Somewhat subtle - siglongjmp restores the signal mask before doing
-	 * the longjmp.  This means that when jumping from one stack to another
-	 * when the target stack has interrupts enabled, an interrupt may occur
-	 * on the source stack.  This is bad when starting up a process because
-	 * it's not supposed to get timer ticks until it has been scheduled.
-	 * So, we disable interrupts around the sigsetjmp to ensure that
-	 * they can't happen until we get back here where they are safe.
-	 */
-	flags = get_signals();
-	block_signals();
-	if(UML_SETJMP(&fork_buf) == 0)
-		new_thread_proc(stack, handler);
-
-	remove_sigstack();
-
-	set_signals(flags);
+	(*buf)[0].JB_IP = (unsigned long) handler;
+	(*buf)[0].JB_SP = (unsigned long) stack +
+		(PAGE_SIZE << UML_CONFIG_KERNEL_STACK_ORDER) - sizeof(void *);
 }
 
 #define INIT_JMP_NEW_THREAD 0
-#define INIT_JMP_REMOVE_SIGSTACK 1
-#define INIT_JMP_CALLBACK 2
-#define INIT_JMP_HALT 3
-#define INIT_JMP_REBOOT 4
-
-void thread_wait(void *sw, void *fb)
-{
-	jmp_buf buf, **switch_buf = sw, *fork_buf;
-
-	*switch_buf = &buf;
-	fork_buf = fb;
-	if(UML_SETJMP(&buf) == 0)
-		UML_LONGJMP(fork_buf, INIT_JMP_REMOVE_SIGSTACK);
-}
+#define INIT_JMP_CALLBACK 1
+#define INIT_JMP_HALT 2
+#define INIT_JMP_REBOOT 3
 
-void switch_threads(void *me, void *next)
+void switch_threads(jmp_buf *me, jmp_buf *you)
 {
-	jmp_buf my_buf, **me_ptr = me, *next_buf = next;
-
-	*me_ptr = &my_buf;
-	if(UML_SETJMP(&my_buf) == 0)
-		UML_LONGJMP(next_buf, 1);
+	if(UML_SETJMP(me) == 0)
+		UML_LONGJMP(you, 1);
 }
 
 static jmp_buf initial_jmpbuf;
@@ -503,23 +469,21 @@ static void (*cb_proc)(void *arg);
 static void *cb_arg;
 static jmp_buf *cb_back;
 
-int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr)
+int start_idle_thread(void *stack, jmp_buf *switch_buf)
 {
-	jmp_buf **switch_buf = switch_buf_ptr;
 	int n;
 
 	set_handler(SIGWINCH, (__sighandler_t) sig_handler,
 		    SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGALRM,
 		    SIGVTALRM, -1);
 
-	*fork_buf_ptr = &initial_jmpbuf;
 	n = UML_SETJMP(&initial_jmpbuf);
 	switch(n){
 	case INIT_JMP_NEW_THREAD:
-		new_thread_proc((void *) stack, new_thread_handler);
-		break;
-	case INIT_JMP_REMOVE_SIGSTACK:
-		remove_sigstack();
+		(*switch_buf)[0].JB_IP = (unsigned long) new_thread_handler;
+		(*switch_buf)[0].JB_SP = (unsigned long) stack +
+			(PAGE_SIZE << UML_CONFIG_KERNEL_STACK_ORDER) -
+			sizeof(void *);
 		break;
 	case INIT_JMP_CALLBACK:
 		(*cb_proc)(cb_arg);
@@ -534,7 +498,7 @@ int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr)
 	default:
 		panic("Bad sigsetjmp return in start_idle_thread - %d\n", n);
 	}
-	UML_LONGJMP(*switch_buf, 1);
+	UML_LONGJMP(switch_buf, 1);
 }
 
 void initial_thread_cb_skas(void (*proc)(void *), void *arg)
diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h
index afa4fe1ca9f1..d99bbddffdb9 100644
--- a/include/asm-um/processor-generic.h
+++ b/include/asm-um/processor-generic.h
@@ -13,6 +13,7 @@ struct task_struct;
 #include "asm/ptrace.h"
 #include "choose-mode.h"
 #include "registers.h"
+#include "sysdep/archsetjmp.h"
 
 struct mm_struct;
 
@@ -43,8 +44,7 @@ struct thread_struct {
 #endif
 #ifdef CONFIG_MODE_SKAS
 		struct {
-			void *switch_buf;
-			void *fork_buf;
+			jmp_buf switch_buf;
 			int mm_count;
 		} skas;
 #endif
@@ -138,7 +138,7 @@ extern struct cpuinfo_um cpu_data[];
 
 #ifdef CONFIG_MODE_SKAS
 #define KSTK_REG(tsk, reg) \
-	get_thread_reg(reg, tsk->thread.mode.skas.switch_buf)
+	get_thread_reg(reg, &tsk->thread.mode.skas.switch_buf)
 #else
 #define KSTK_REG(tsk, reg) (0xbadbabe)
 #endif
-- 
cgit v1.2.3


From 7e96287ddc4f42081e18248b6167041c0908004c Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Wed, 27 Sep 2006 01:50:44 -0700
Subject: [PATCH] kdump: introduce "reset_devices" command line option

Resetting the devices during driver initialization can be a costly
operation in terms of time (especially scsi devices).  This option can be
used by drivers to know that user forcibly wants the devices to be reset
during initialization.

This option can be useful while kernel is booting in unreliable
environment.  For ex.  during kdump boot where devices are in unknown
random state and BIOS execution has been skipped.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |  3 +++
 include/linux/init.h                |  1 +
 init/main.c                         | 20 ++++++++++++++++++++
 3 files changed, 24 insertions(+)

(limited to 'include')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 255ec535bba8..54983246930d 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1370,6 +1370,9 @@ running once the system is up.
 			Reserves a hole at the top of the kernel virtual
 			address space.
 
+	reset_devices	[KNL] Force drivers to reset the underlying device
+			during initialization.
+
 	resume=		[SWSUSP]
 			Specify the partition device for software suspend
 
diff --git a/include/linux/init.h b/include/linux/init.h
index 6667785dd1ff..e92b1455d7af 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -68,6 +68,7 @@ extern initcall_t __security_initcall_start[], __security_initcall_end[];
 
 /* Defined in init/main.c */
 extern char saved_command_line[];
+extern unsigned int reset_devices;
 
 /* used by init/main.c */
 extern void setup_arch(char **);
diff --git a/init/main.c b/init/main.c
index 913e48d658ee..0766e69712b2 100644
--- a/init/main.c
+++ b/init/main.c
@@ -127,6 +127,18 @@ static char *ramdisk_execute_command;
 /* Setup configured maximum number of CPUs to activate */
 static unsigned int max_cpus = NR_CPUS;
 
+/*
+ * If set, this is an indication to the drivers that reset the underlying
+ * device before going ahead with the initialization otherwise driver might
+ * rely on the BIOS and skip the reset operation.
+ *
+ * This is useful if kernel is booting in an unreliable environment.
+ * For ex. kdump situaiton where previous kernel has crashed, BIOS has been
+ * skipped and devices will be in unknown state.
+ */
+unsigned int reset_devices;
+EXPORT_SYMBOL(reset_devices);
+
 /*
  * Setup routine for controlling SMP activation
  *
@@ -153,6 +165,14 @@ static int __init maxcpus(char *str)
 
 __setup("maxcpus=", maxcpus);
 
+static int __init set_reset_devices(char *str)
+{
+	reset_devices = 1;
+	return 1;
+}
+
+__setup("reset_devices", set_reset_devices);
+
 static char * argv_init[MAX_INIT_ARGS+2] = { "init", NULL, };
 char * envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, };
 static const char *panic_later, *panic_param;
-- 
cgit v1.2.3


From 8e18e2941c53416aa219708e7dcad21fb4bd6794 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:46 -0700
Subject: [PATCH] inode_diet: Replace inode.u.generic_ip with inode.i_private

The following patches reduce the size of the VFS inode structure by 28 bytes
on a UP x86.  (It would be more on an x86_64 system).  This is a 10% reduction
in the inode size on a UP kernel that is configured in a production mode
(i.e., with no spinlock or other debugging functions enabled; if you want to
save memory taken up by in-core inodes, the first thing you should do is
disable the debugging options; they are responsible for a huge amount of bloat
in the VFS inode structure).

This patch:

The filesystem or device-specific pointer in the inode is inside a union,
which is pretty pointless given that all 30+ users of this field have been
using the void pointer.  Get rid of the union and rename it to i_private, with
a comment to explain who is allowed to use the void pointer.  This is just a
cleanup, but it allows us to reuse the union 'u' for something something where
the union will actually be used.

[judith@osdl.org: powerpc build fix]
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Judith Lebzelter <judith@osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/platforms/cell/spufs/inode.c      |  2 +-
 arch/powerpc/platforms/pseries/hvCall_inst.c   |  2 +-
 arch/s390/hypfs/inode.c                        |  6 ++---
 arch/s390/kernel/debug.c                       |  2 +-
 block/blktrace.c                               |  2 +-
 drivers/i2c/chips/tps65010.c                   |  2 +-
 drivers/infiniband/hw/ipath/ipath_fs.c         | 12 ++++-----
 drivers/infiniband/ulp/ipoib/ipoib_fs.c        |  4 +--
 drivers/misc/ibmasm/ibmasmfs.c                 | 16 ++++++------
 drivers/net/irda/vlsi_ir.h                     |  2 +-
 drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c |  2 +-
 drivers/oprofile/oprofilefs.c                  | 10 ++++----
 drivers/pci/hotplug/cpqphp_sysfs.c             |  2 +-
 drivers/usb/core/devio.c                       |  2 +-
 drivers/usb/core/inode.c                       |  6 ++---
 drivers/usb/gadget/inode.c                     |  6 ++---
 drivers/usb/host/isp116x-hcd.c                 |  2 +-
 drivers/usb/host/uhci-debug.c                  |  2 +-
 drivers/usb/mon/mon_stat.c                     |  2 +-
 drivers/usb/mon/mon_text.c                     |  4 +--
 fs/autofs/inode.c                              |  2 +-
 fs/autofs/symlink.c                            |  2 +-
 fs/binfmt_misc.c                               |  8 +++---
 fs/debugfs/file.c                              |  4 +--
 fs/debugfs/inode.c                             |  4 +--
 fs/devpts/inode.c                              |  4 +--
 fs/freevxfs/vxfs.h                             |  2 +-
 fs/freevxfs/vxfs_inode.c                       |  4 +--
 fs/fuse/control.c                              |  6 ++---
 fs/inode.c                                     |  2 +-
 fs/jffs/inode-v23.c                            | 34 +++++++++++++-------------
 fs/libfs.c                                     |  2 +-
 fs/ocfs2/dlmglue.c                             |  2 +-
 include/linux/fs.h                             |  4 +--
 kernel/relay.c                                 |  2 +-
 security/inode.c                               |  8 +++---
 36 files changed, 88 insertions(+), 90 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index 7b4572805db9..b837f12a84ae 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -120,7 +120,7 @@ spufs_new_file(struct super_block *sb, struct dentry *dentry,
 	ret = 0;
 	inode->i_op = &spufs_file_iops;
 	inode->i_fop = fops;
-	inode->u.generic_ip = SPUFS_I(inode)->i_ctx = get_spu_context(ctx);
+	inode->i_private = SPUFS_I(inode)->i_ctx = get_spu_context(ctx);
 	d_add(dentry, inode);
 out:
 	return ret;
diff --git a/arch/powerpc/platforms/pseries/hvCall_inst.c b/arch/powerpc/platforms/pseries/hvCall_inst.c
index 641e6511cf06..446e17d162a5 100644
--- a/arch/powerpc/platforms/pseries/hvCall_inst.c
+++ b/arch/powerpc/platforms/pseries/hvCall_inst.c
@@ -85,7 +85,7 @@ static int hcall_inst_seq_open(struct inode *inode, struct file *file)
 
 	rc = seq_open(file, &hcall_inst_seq_ops);
 	seq = file->private_data;
-	seq->private = file->f_dentry->d_inode->u.generic_ip;
+	seq->private = file->f_dentry->d_inode->i_private;
 
 	return rc;
 }
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index bdade5f2e325..8735024d235b 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -104,13 +104,13 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode)
 
 static void hypfs_drop_inode(struct inode *inode)
 {
-	kfree(inode->u.generic_ip);
+	kfree(inode->i_private);
 	generic_delete_inode(inode);
 }
 
 static int hypfs_open(struct inode *inode, struct file *filp)
 {
-	char *data = filp->f_dentry->d_inode->u.generic_ip;
+	char *data = filp->f_dentry->d_inode->i_private;
 	struct hypfs_sb_info *fs_info;
 
 	if (filp->f_mode & FMODE_WRITE) {
@@ -352,7 +352,7 @@ static struct dentry *hypfs_create_file(struct super_block *sb,
 		parent->d_inode->i_nlink++;
 	} else
 		BUG();
-	inode->u.generic_ip = data;
+	inode->i_private = data;
 	d_instantiate(dentry, inode);
 	dget(dentry);
 	return dentry;
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 7ba20922a535..43f3d0c7e132 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -603,7 +603,7 @@ debug_open(struct inode *inode, struct file *file)
 	debug_info_t *debug_info, *debug_info_snapshot;
 
 	down(&debug_lock);
-	debug_info = (struct debug_info*)file->f_dentry->d_inode->u.generic_ip;
+	debug_info = file->f_dentry->d_inode->i_private;
 	/* find debug view */
 	for (i = 0; i < DEBUG_MAX_VIEWS; i++) {
 		if (!debug_info->views[i])
diff --git a/block/blktrace.c b/block/blktrace.c
index 265f7a830619..2b4ef2b89b8d 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -217,7 +217,7 @@ static int blk_trace_remove(request_queue_t *q)
 
 static int blk_dropped_open(struct inode *inode, struct file *filp)
 {
-	filp->private_data = inode->u.generic_ip;
+	filp->private_data = inode->i_private;
 
 	return 0;
 }
diff --git a/drivers/i2c/chips/tps65010.c b/drivers/i2c/chips/tps65010.c
index 0be6fd6a267d..6a7578217177 100644
--- a/drivers/i2c/chips/tps65010.c
+++ b/drivers/i2c/chips/tps65010.c
@@ -305,7 +305,7 @@ static int dbg_show(struct seq_file *s, void *_)
 
 static int dbg_tps_open(struct inode *inode, struct file *file)
 {
-	return single_open(file, dbg_show, inode->u.generic_ip);
+	return single_open(file, dbg_show, inode->i_private);
 }
 
 static struct file_operations debug_fops = {
diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c
index a5eb30a06a5c..055cdd089b28 100644
--- a/drivers/infiniband/hw/ipath/ipath_fs.c
+++ b/drivers/infiniband/hw/ipath/ipath_fs.c
@@ -64,7 +64,7 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry,
 	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	inode->u.generic_ip = data;
+	inode->i_private = data;
 	if ((mode & S_IFMT) == S_IFDIR) {
 		inode->i_op = &simple_dir_inode_operations;
 		inode->i_nlink++;
@@ -119,7 +119,7 @@ static ssize_t atomic_counters_read(struct file *file, char __user *buf,
 	u16 i;
 	struct ipath_devdata *dd;
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 
 	for (i = 0; i < NUM_COUNTERS; i++)
 		counters[i] = ipath_snap_cntr(dd, i);
@@ -139,7 +139,7 @@ static ssize_t atomic_node_info_read(struct file *file, char __user *buf,
 	struct ipath_devdata *dd;
 	u64 guid;
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 
 	guid = be64_to_cpu(dd->ipath_guid);
 
@@ -178,7 +178,7 @@ static ssize_t atomic_port_info_read(struct file *file, char __user *buf,
 	u32 tmp, tmp2;
 	struct ipath_devdata *dd;
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 
 	/* so we only initialize non-zero fields. */
 	memset(portinfo, 0, sizeof portinfo);
@@ -325,7 +325,7 @@ static ssize_t flash_read(struct file *file, char __user *buf,
 		goto bail;
 	}
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 	if (ipath_eeprom_read(dd, pos, tmp, count)) {
 		ipath_dev_err(dd, "failed to read from flash\n");
 		ret = -ENXIO;
@@ -381,7 +381,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf,
 		goto bail_tmp;
 	}
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 	if (ipath_eeprom_write(dd, pos, tmp, count)) {
 		ret = -ENXIO;
 		ipath_dev_err(dd, "failed to write to flash\n");
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c
index 5dde380e8dbe..f1cb83688b31 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c
@@ -141,7 +141,7 @@ static int ipoib_mcg_open(struct inode *inode, struct file *file)
 		return ret;
 
 	seq = file->private_data;
-	seq->private = inode->u.generic_ip;
+	seq->private = inode->i_private;
 
 	return 0;
 }
@@ -247,7 +247,7 @@ static int ipoib_path_open(struct inode *inode, struct file *file)
 		return ret;
 
 	seq = file->private_data;
-	seq->private = inode->u.generic_ip;
+	seq->private = inode->i_private;
 
 	return 0;
 }
diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c
index 4a35caff5d02..0e909b617226 100644
--- a/drivers/misc/ibmasm/ibmasmfs.c
+++ b/drivers/misc/ibmasm/ibmasmfs.c
@@ -175,7 +175,7 @@ static struct dentry *ibmasmfs_create_file (struct super_block *sb,
 	}
 
 	inode->i_fop = fops;
-	inode->u.generic_ip = data;
+	inode->i_private = data;
 
 	d_add(dentry, inode);
 	return dentry;
@@ -244,7 +244,7 @@ static int command_file_open(struct inode *inode, struct file *file)
 {
 	struct ibmasmfs_command_data *command_data;
 
-	if (!inode->u.generic_ip)
+	if (!inode->i_private)
 		return -ENODEV;
 
 	command_data = kmalloc(sizeof(struct ibmasmfs_command_data), GFP_KERNEL);
@@ -252,7 +252,7 @@ static int command_file_open(struct inode *inode, struct file *file)
 		return -ENOMEM;
 
 	command_data->command = NULL;
-	command_data->sp = inode->u.generic_ip;
+	command_data->sp = inode->i_private;
 	file->private_data = command_data;
 	return 0;
 }
@@ -351,10 +351,10 @@ static int event_file_open(struct inode *inode, struct file *file)
 	struct ibmasmfs_event_data *event_data;
 	struct service_processor *sp; 
 
-	if (!inode->u.generic_ip)
+	if (!inode->i_private)
 		return -ENODEV;
 
-	sp = inode->u.generic_ip;
+	sp = inode->i_private;
 
 	event_data = kmalloc(sizeof(struct ibmasmfs_event_data), GFP_KERNEL);
 	if (!event_data)
@@ -439,14 +439,14 @@ static int r_heartbeat_file_open(struct inode *inode, struct file *file)
 {
 	struct ibmasmfs_heartbeat_data *rhbeat;
 
-	if (!inode->u.generic_ip)
+	if (!inode->i_private)
 		return -ENODEV;
 
 	rhbeat = kmalloc(sizeof(struct ibmasmfs_heartbeat_data), GFP_KERNEL);
 	if (!rhbeat)
 		return -ENOMEM;
 
-	rhbeat->sp = (struct service_processor *)inode->u.generic_ip;
+	rhbeat->sp = inode->i_private;
 	rhbeat->active = 0;
 	ibmasm_init_reverse_heartbeat(rhbeat->sp, &rhbeat->heartbeat);
 	file->private_data = rhbeat;
@@ -508,7 +508,7 @@ static ssize_t r_heartbeat_file_write(struct file *file, const char __user *buf,
 
 static int remote_settings_file_open(struct inode *inode, struct file *file)
 {
-	file->private_data = inode->u.generic_ip;
+	file->private_data = inode->i_private;
 	return 0;
 }
 
diff --git a/drivers/net/irda/vlsi_ir.h b/drivers/net/irda/vlsi_ir.h
index a82a4ba8de4f..c37f0bc4c7f9 100644
--- a/drivers/net/irda/vlsi_ir.h
+++ b/drivers/net/irda/vlsi_ir.h
@@ -58,7 +58,7 @@ typedef void irqreturn_t;
 
 /* PDE() introduced in 2.5.4 */
 #ifdef CONFIG_PROC_FS
-#define PDE(inode) ((inode)->u.generic_ip)
+#define PDE(inode) ((inode)->i_private)
 #endif
 
 /* irda crc16 calculation exported in 2.5.42 */
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c b/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c
index 923275ea0789..b9df06a06ea9 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c
@@ -54,7 +54,7 @@ static ssize_t write_file_dummy(struct file *file, const char __user *buf,
 
 static int open_file_generic(struct inode *inode, struct file *file)
 {
-	file->private_data = inode->u.generic_ip;
+	file->private_data = inode->i_private;
 	return 0;
 }
 
diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c
index 71c2da277d6e..deb37354785b 100644
--- a/drivers/oprofile/oprofilefs.c
+++ b/drivers/oprofile/oprofilefs.c
@@ -110,8 +110,8 @@ static ssize_t ulong_write_file(struct file * file, char const __user * buf, siz
 
 static int default_open(struct inode * inode, struct file * filp)
 {
-	if (inode->u.generic_ip)
-		filp->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		filp->private_data = inode->i_private;
 	return 0;
 }
 
@@ -158,7 +158,7 @@ int oprofilefs_create_ulong(struct super_block * sb, struct dentry * root,
 	if (!d)
 		return -EFAULT;
 
-	d->d_inode->u.generic_ip = val;
+	d->d_inode->i_private = val;
 	return 0;
 }
 
@@ -171,7 +171,7 @@ int oprofilefs_create_ro_ulong(struct super_block * sb, struct dentry * root,
 	if (!d)
 		return -EFAULT;
 
-	d->d_inode->u.generic_ip = val;
+	d->d_inode->i_private = val;
 	return 0;
 }
 
@@ -197,7 +197,7 @@ int oprofilefs_create_ro_atomic(struct super_block * sb, struct dentry * root,
 	if (!d)
 		return -EFAULT;
 
-	d->d_inode->u.generic_ip = val;
+	d->d_inode->i_private = val;
 	return 0;
 }
 
diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c
index 8b3da007e859..5bab666cd67e 100644
--- a/drivers/pci/hotplug/cpqphp_sysfs.c
+++ b/drivers/pci/hotplug/cpqphp_sysfs.c
@@ -140,7 +140,7 @@ struct ctrl_dbg {
 
 static int open(struct inode *inode, struct file *file)
 {
-	struct controller *ctrl = inode->u.generic_ip;
+	struct controller *ctrl = inode->i_private;
 	struct ctrl_dbg *dbg;
 	int retval = -ENOMEM;
 
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 218621b9958e..32e03000420c 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -555,7 +555,7 @@ static int usbdev_open(struct inode *inode, struct file *file)
 	if (imajor(inode) == USB_DEVICE_MAJOR)
 		dev = usbdev_lookup_minor(iminor(inode));
 	if (!dev)
-		dev = inode->u.generic_ip;
+		dev = inode->i_private;
 	if (!dev) {
 		kfree(ps);
 		goto out;
diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index 3182c2224ba2..482f253085e5 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -402,8 +402,8 @@ static loff_t default_file_lseek (struct file *file, loff_t offset, int orig)
 
 static int default_open (struct inode *inode, struct file *file)
 {
-	if (inode->u.generic_ip)
-		file->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		file->private_data = inode->i_private;
 
 	return 0;
 }
@@ -509,7 +509,7 @@ static struct dentry *fs_create_file (const char *name, mode_t mode,
 	} else {
 		if (dentry->d_inode) {
 			if (data)
-				dentry->d_inode->u.generic_ip = data;
+				dentry->d_inode->i_private = data;
 			if (fops)
 				dentry->d_inode->i_fop = fops;
 			dentry->d_inode->i_uid = uid;
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 3bdc5e3ba234..ffaa8c1afad8 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -844,7 +844,7 @@ fail1:
 static int
 ep_open (struct inode *inode, struct file *fd)
 {
-	struct ep_data		*data = inode->u.generic_ip;
+	struct ep_data		*data = inode->i_private;
 	int			value = -EBUSY;
 
 	if (down_interruptible (&data->lock) != 0)
@@ -1909,7 +1909,7 @@ fail:
 static int
 dev_open (struct inode *inode, struct file *fd)
 {
-	struct dev_data		*dev = inode->u.generic_ip;
+	struct dev_data		*dev = inode->i_private;
 	int			value = -EBUSY;
 
 	if (dev->state == STATE_DEV_DISABLED) {
@@ -1970,7 +1970,7 @@ gadgetfs_make_inode (struct super_block *sb,
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime
 				= CURRENT_TIME;
-		inode->u.generic_ip = data;
+		inode->i_private = data;
 		inode->i_fop = fops;
 	}
 	return inode;
diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c
index 5147ed4a6662..8c6b38a0b5bb 100644
--- a/drivers/usb/host/isp116x-hcd.c
+++ b/drivers/usb/host/isp116x-hcd.c
@@ -1204,7 +1204,7 @@ static int isp116x_show_dbg(struct seq_file *s, void *unused)
 
 static int isp116x_open_seq(struct inode *inode, struct file *file)
 {
-	return single_open(file, isp116x_show_dbg, inode->u.generic_ip);
+	return single_open(file, isp116x_show_dbg, inode->i_private);
 }
 
 static struct file_operations isp116x_debug_fops = {
diff --git a/drivers/usb/host/uhci-debug.c b/drivers/usb/host/uhci-debug.c
index dc286a48cafd..d1372cb27f33 100644
--- a/drivers/usb/host/uhci-debug.c
+++ b/drivers/usb/host/uhci-debug.c
@@ -428,7 +428,7 @@ struct uhci_debug {
 
 static int uhci_debug_open(struct inode *inode, struct file *file)
 {
-	struct uhci_hcd *uhci = inode->u.generic_ip;
+	struct uhci_hcd *uhci = inode->i_private;
 	struct uhci_debug *up;
 	int ret = -ENOMEM;
 	unsigned long flags;
diff --git a/drivers/usb/mon/mon_stat.c b/drivers/usb/mon/mon_stat.c
index 1fe01d994a79..86ad2b381c4b 100644
--- a/drivers/usb/mon/mon_stat.c
+++ b/drivers/usb/mon/mon_stat.c
@@ -28,7 +28,7 @@ static int mon_stat_open(struct inode *inode, struct file *file)
 	if ((sp = kmalloc(sizeof(struct snap), GFP_KERNEL)) == NULL)
 		return -ENOMEM;
 
-	mbus = inode->u.generic_ip;
+	mbus = inode->i_private;
 
 	sp->slen = snprintf(sp->str, STAT_BUF_SIZE,
 	    "nreaders %d events %u text_lost %u\n",
diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c
index f961a770cee2..2fd39b4fa166 100644
--- a/drivers/usb/mon/mon_text.c
+++ b/drivers/usb/mon/mon_text.c
@@ -238,7 +238,7 @@ static int mon_text_open(struct inode *inode, struct file *file)
 	int rc;
 
 	mutex_lock(&mon_lock);
-	mbus = inode->u.generic_ip;
+	mbus = inode->i_private;
 	ubus = mbus->u_bus;
 
 	rp = kzalloc(sizeof(struct mon_reader_text), GFP_KERNEL);
@@ -401,7 +401,7 @@ static int mon_text_release(struct inode *inode, struct file *file)
 	struct mon_event_text *ep;
 
 	mutex_lock(&mon_lock);
-	mbus = inode->u.generic_ip;
+	mbus = inode->i_private;
 
 	if (mbus->nreaders <= 0) {
 		printk(KERN_ERR TAG ": consistency error on close\n");
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index c81d6b8c2828..d39d2acd9b38 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -241,7 +241,7 @@ static void autofs_read_inode(struct inode *inode)
 		
 		inode->i_op = &autofs_symlink_inode_operations;
 		sl = &sbi->symlink[n];
-		inode->u.generic_ip = sl;
+		inode->i_private = sl;
 		inode->i_mode = S_IFLNK | S_IRWXUGO;
 		inode->i_mtime.tv_sec = inode->i_ctime.tv_sec = sl->mtime;
 		inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = 0;
diff --git a/fs/autofs/symlink.c b/fs/autofs/symlink.c
index 52e8772b066e..c74f2eb65775 100644
--- a/fs/autofs/symlink.c
+++ b/fs/autofs/symlink.c
@@ -15,7 +15,7 @@
 /* Nothing to release.. */
 static void *autofs_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
-	char *s=((struct autofs_symlink *)dentry->d_inode->u.generic_ip)->data;
+	char *s=((struct autofs_symlink *)dentry->d_inode->i_private)->data;
 	nd_set_link(nd, s);
 	return NULL;
 }
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 34ebbc191e46..6759b9839ce8 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -517,7 +517,7 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
 
 static void bm_clear_inode(struct inode *inode)
 {
-	kfree(inode->u.generic_ip);
+	kfree(inode->i_private);
 }
 
 static void kill_node(Node *e)
@@ -545,7 +545,7 @@ static void kill_node(Node *e)
 static ssize_t
 bm_entry_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos)
 {
-	Node *e = file->f_dentry->d_inode->u.generic_ip;
+	Node *e = file->f_dentry->d_inode->i_private;
 	loff_t pos = *ppos;
 	ssize_t res;
 	char *page;
@@ -579,7 +579,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer,
 				size_t count, loff_t *ppos)
 {
 	struct dentry *root;
-	Node *e = file->f_dentry->d_inode->u.generic_ip;
+	Node *e = file->f_dentry->d_inode->i_private;
 	int res = parse_command(buffer, count);
 
 	switch (res) {
@@ -646,7 +646,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer,
 	}
 
 	e->dentry = dget(dentry);
-	inode->u.generic_ip = e;
+	inode->i_private = e;
 	inode->i_fop = &bm_entry_operations;
 
 	d_instantiate(dentry, inode);
diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index e4b430552c88..bf3901ab1744 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -32,8 +32,8 @@ static ssize_t default_write_file(struct file *file, const char __user *buf,
 
 static int default_open(struct inode *inode, struct file *file)
 {
-	if (inode->u.generic_ip)
-		file->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		file->private_data = inode->i_private;
 
 	return 0;
 }
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 3ca268d2e5a2..717f4821ed02 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -168,7 +168,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
  *          directory dentry if set.  If this paramater is NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @data: a pointer to something that the caller will want to get to later
- *        on.  The inode.u.generic_ip pointer will point to this value on
+ *        on.  The inode.i_private pointer will point to this value on
  *        the open() call.
  * @fops: a pointer to a struct file_operations that should be used for
  *        this file.
@@ -209,7 +209,7 @@ struct dentry *debugfs_create_file(const char *name, mode_t mode,
 
 	if (dentry->d_inode) {
 		if (data)
-			dentry->d_inode->u.generic_ip = data;
+			dentry->d_inode->i_private = data;
 		if (fops)
 			dentry->d_inode->i_fop = fops;
 	}
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index f7aef5bb584a..5bf06a10dddf 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -177,7 +177,7 @@ int devpts_pty_new(struct tty_struct *tty)
 	inode->i_gid = config.setgid ? config.gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	init_special_inode(inode, S_IFCHR|config.mode, device);
-	inode->u.generic_ip = tty;
+	inode->i_private = tty;
 
 	dentry = get_node(number);
 	if (!IS_ERR(dentry) && !dentry->d_inode)
@@ -196,7 +196,7 @@ struct tty_struct *devpts_get_tty(int number)
 	tty = NULL;
 	if (!IS_ERR(dentry)) {
 		if (dentry->d_inode)
-			tty = dentry->d_inode->u.generic_ip;
+			tty = dentry->d_inode->i_private;
 		dput(dentry);
 	}
 
diff --git a/fs/freevxfs/vxfs.h b/fs/freevxfs/vxfs.h
index d35979a58743..c8a92652612a 100644
--- a/fs/freevxfs/vxfs.h
+++ b/fs/freevxfs/vxfs.h
@@ -252,7 +252,7 @@ enum {
  * Get filesystem private data from VFS inode.
  */
 #define VXFS_INO(ip) \
-	((struct vxfs_inode_info *)(ip)->u.generic_ip)
+	((struct vxfs_inode_info *)(ip)->i_private)
 
 /*
  * Get filesystem private data from VFS superblock.
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index ca6a39714771..32a82ed108e4 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -243,7 +243,7 @@ vxfs_iinit(struct inode *ip, struct vxfs_inode_info *vip)
 	ip->i_blocks = vip->vii_blocks;
 	ip->i_generation = vip->vii_gen;
 
-	ip->u.generic_ip = (void *)vip;
+	ip->i_private = vip;
 	
 }
 
@@ -338,5 +338,5 @@ vxfs_read_inode(struct inode *ip)
 void
 vxfs_clear_inode(struct inode *ip)
 {
-	kmem_cache_free(vxfs_inode_cachep, ip->u.generic_ip);
+	kmem_cache_free(vxfs_inode_cachep, ip->i_private);
 }
diff --git a/fs/fuse/control.c b/fs/fuse/control.c
index 46fe60b2da23..79ec1f23d4d2 100644
--- a/fs/fuse/control.c
+++ b/fs/fuse/control.c
@@ -23,7 +23,7 @@ static struct fuse_conn *fuse_ctl_file_conn_get(struct file *file)
 {
 	struct fuse_conn *fc;
 	mutex_lock(&fuse_mutex);
-	fc = file->f_dentry->d_inode->u.generic_ip;
+	fc = file->f_dentry->d_inode->i_private;
 	if (fc)
 		fc = fuse_conn_get(fc);
 	mutex_unlock(&fuse_mutex);
@@ -98,7 +98,7 @@ static struct dentry *fuse_ctl_add_dentry(struct dentry *parent,
 		inode->i_op = iop;
 	inode->i_fop = fop;
 	inode->i_nlink = nlink;
-	inode->u.generic_ip = fc;
+	inode->i_private = fc;
 	d_add(dentry, inode);
 	return dentry;
 }
@@ -150,7 +150,7 @@ void fuse_ctl_remove_conn(struct fuse_conn *fc)
 
 	for (i = fc->ctl_ndents - 1; i >= 0; i--) {
 		struct dentry *dentry = fc->ctl_dentry[i];
-		dentry->d_inode->u.generic_ip = NULL;
+		dentry->d_inode->i_private = NULL;
 		d_drop(dentry);
 		dput(dentry);
 	}
diff --git a/fs/inode.c b/fs/inode.c
index 0bf9f0444a96..77e254792025 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -163,7 +163,7 @@ static struct inode *alloc_inode(struct super_block *sb)
 				bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info;
 			mapping->backing_dev_info = bdi;
 		}
-		memset(&inode->u, 0, sizeof(inode->u));
+		inode->i_private = 0;
 		inode->i_mapping = mapping;
 	}
 	return inode;
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index b59553d28d13..7358ef87f16b 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -369,7 +369,7 @@ jffs_new_inode(const struct inode * dir, struct jffs_raw_inode *raw_inode,
 
 	f = jffs_find_file(c, raw_inode->ino);
 
-	inode->u.generic_ip = (void *)f;
+	inode->i_private = (void *)f;
 	insert_inode_hash(inode);
 
 	return inode;
@@ -442,7 +442,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	});
 
 	result = -ENOTDIR;
-	if (!(old_dir_f = (struct jffs_file *)old_dir->u.generic_ip)) {
+	if (!(old_dir_f = old_dir->i_private)) {
 		D(printk("jffs_rename(): Old dir invalid.\n"));
 		goto jffs_rename_end;
 	}
@@ -456,7 +456,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	/* Find the new directory.  */
 	result = -ENOTDIR;
-	if (!(new_dir_f = (struct jffs_file *)new_dir->u.generic_ip)) {
+	if (!(new_dir_f = new_dir->i_private)) {
 		D(printk("jffs_rename(): New dir invalid.\n"));
 		goto jffs_rename_end;
 	}
@@ -593,7 +593,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		}
 		else {
 			ddino = ((struct jffs_file *)
-				 inode->u.generic_ip)->pino;
+				 inode->i_private)->pino;
 		}
 		D3(printk("jffs_readdir(): \"..\" %u\n", ddino));
 		if (filldir(dirent, "..", 2, filp->f_pos, ddino, DT_DIR) < 0) {
@@ -604,7 +604,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		}
 		filp->f_pos++;
 	}
-	f = ((struct jffs_file *)inode->u.generic_ip)->children;
+	f = ((struct jffs_file *)inode->i_private)->children;
 
 	j = 2;
 	while(f && (f->deleted || j++ < filp->f_pos )) {
@@ -668,7 +668,7 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	}
 
 	r = -EACCES;
-	if (!(d = (struct jffs_file *)dir->u.generic_ip)) {
+	if (!(d = (struct jffs_file *)dir->i_private)) {
 		D(printk("jffs_lookup(): No such inode! (%lu)\n",
 			 dir->i_ino));
 		goto jffs_lookup_end;
@@ -739,7 +739,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page)
 	unsigned long read_len;
 	int result;
 	struct inode *inode = (struct inode*)page->mapping->host;
-	struct jffs_file *f = (struct jffs_file *)inode->u.generic_ip;
+	struct jffs_file *f = (struct jffs_file *)inode->i_private;
 	struct jffs_control *c = (struct jffs_control *)inode->i_sb->s_fs_info;
 	int r;
 	loff_t offset;
@@ -828,7 +828,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	});
 
 	lock_kernel();
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 
 	ASSERT(if (!dir_f) {
 		printk(KERN_ERR "jffs_mkdir(): No reference to a "
@@ -972,7 +972,7 @@ jffs_remove(struct inode *dir, struct dentry *dentry, int type)
 		kfree(_name);
 	});
 
-	dir_f = (struct jffs_file *) dir->u.generic_ip;
+	dir_f = dir->i_private;
 	c = dir_f->c;
 
 	result = -ENOENT;
@@ -1082,7 +1082,7 @@ jffs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
 	if (!old_valid_dev(rdev))
 		return -EINVAL;
 	lock_kernel();
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 	c = dir_f->c;
 
 	D3(printk (KERN_NOTICE "mknod(): down biglock\n"));
@@ -1186,7 +1186,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 		kfree(_symname);
 	});
 
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 	ASSERT(if (!dir_f) {
 		printk(KERN_ERR "jffs_symlink(): No reference to a "
 		       "jffs_file struct in inode.\n");
@@ -1289,7 +1289,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
 		kfree(s);
 	});
 
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 	ASSERT(if (!dir_f) {
 		printk(KERN_ERR "jffs_create(): No reference to a "
 		       "jffs_file struct in inode.\n");
@@ -1403,9 +1403,9 @@ jffs_file_write(struct file *filp, const char *buf, size_t count,
 		goto out_isem;
 	}
 
-	if (!(f = (struct jffs_file *)inode->u.generic_ip)) {
-		D(printk("jffs_file_write(): inode->u.generic_ip = 0x%p\n",
-				inode->u.generic_ip));
+	if (!(f = inode->i_private)) {
+		D(printk("jffs_file_write(): inode->i_private = 0x%p\n",
+				inode->i_private));
 		goto out_isem;
 	}
 
@@ -1693,7 +1693,7 @@ jffs_read_inode(struct inode *inode)
 		mutex_unlock(&c->fmc->biglock);
 		return;
 	}
-	inode->u.generic_ip = (void *)f;
+	inode->i_private = f;
 	inode->i_mode = f->mode;
 	inode->i_nlink = f->nlink;
 	inode->i_uid = f->uid;
@@ -1748,7 +1748,7 @@ jffs_delete_inode(struct inode *inode)
 	lock_kernel();
 	inode->i_size = 0;
 	inode->i_blocks = 0;
-	inode->u.generic_ip = NULL;
+	inode->i_private = NULL;
 	clear_inode(inode);
 	if (inode->i_nlink == 0) {
 		c = (struct jffs_control *) inode->i_sb->s_fs_info;
diff --git a/fs/libfs.c b/fs/libfs.c
index ac02ea602c3d..2751793beeaa 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -547,7 +547,7 @@ int simple_attr_open(struct inode *inode, struct file *file,
 
 	attr->get = get;
 	attr->set = set;
-	attr->data = inode->u.generic_ip;
+	attr->data = inode->i_private;
 	attr->fmt = fmt;
 	mutex_init(&attr->mutex);
 
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index de887063dcfc..8801e41afe80 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -2052,7 +2052,7 @@ static int ocfs2_dlm_debug_open(struct inode *inode, struct file *file)
 		mlog_errno(ret);
 		goto out;
 	}
-	osb = (struct ocfs2_super *) inode->u.generic_ip;
+	osb = inode->i_private;
 	ocfs2_get_dlm_debug(osb->osb_dlm_debug);
 	priv->p_dlm_debug = osb->osb_dlm_debug;
 	INIT_LIST_HEAD(&priv->p_iter_res.l_debug_list);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1d3e601ece73..4f77ec9c3353 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -554,9 +554,7 @@ struct inode {
 
 	atomic_t		i_writecount;
 	void			*i_security;
-	union {
-		void		*generic_ip;
-	} u;
+	void			*i_private; /* fs or device private pointer */
 #ifdef __NEED_I_SIZE_ORDERED
 	seqcount_t		i_size_seqcount;
 #endif
diff --git a/kernel/relay.c b/kernel/relay.c
index 33345e73485c..85786ff2a4f9 100644
--- a/kernel/relay.c
+++ b/kernel/relay.c
@@ -669,7 +669,7 @@ EXPORT_SYMBOL_GPL(relay_flush);
  */
 static int relay_file_open(struct inode *inode, struct file *filp)
 {
-	struct rchan_buf *buf = inode->u.generic_ip;
+	struct rchan_buf *buf = inode->i_private;
 	kref_get(&buf->kref);
 	filp->private_data = buf;
 
diff --git a/security/inode.c b/security/inode.c
index 47eb63480dac..176aacea8ca4 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -44,8 +44,8 @@ static ssize_t default_write_file(struct file *file, const char __user *buf,
 
 static int default_open(struct inode *inode, struct file *file)
 {
-	if (inode->u.generic_ip)
-		file->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		file->private_data = inode->i_private;
 
 	return 0;
 }
@@ -194,7 +194,7 @@ static int create_by_name(const char *name, mode_t mode,
  *          directory dentry if set.  If this paramater is NULL, then the
  *          file will be created in the root of the securityfs filesystem.
  * @data: a pointer to something that the caller will want to get to later
- *        on.  The inode.u.generic_ip pointer will point to this value on
+ *        on.  The inode.i_private pointer will point to this value on
  *        the open() call.
  * @fops: a pointer to a struct file_operations that should be used for
  *        this file.
@@ -240,7 +240,7 @@ struct dentry *securityfs_create_file(const char *name, mode_t mode,
 		if (fops)
 			dentry->d_inode->i_fop = fops;
 		if (data)
-			dentry->d_inode->u.generic_ip = data;
+			dentry->d_inode->i_private = data;
 	}
 exit:
 	return dentry;
-- 
cgit v1.2.3


From 4c1541680f8d189d21dd07b053bc12996574646e Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:47 -0700
Subject: [PATCH] inode-diet: Move i_pipe into a union

Move the i_pipe pointer into a union that will be shared with i_bdev and
i_cdev.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/fs.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4f77ec9c3353..ca695fc8d69c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -528,9 +528,10 @@ struct inode {
 #ifdef CONFIG_QUOTA
 	struct dquot		*i_dquot[MAXQUOTAS];
 #endif
-	/* These three should probably be a union */
 	struct list_head	i_devices;
-	struct pipe_inode_info	*i_pipe;
+	union {
+		struct pipe_inode_info	*i_pipe;
+	};
 	struct block_device	*i_bdev;
 	struct cdev		*i_cdev;
 	int			i_cindex;
-- 
cgit v1.2.3


From eaf796e7ef6014f208c409b2b14fddcfaafe7e3a Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:48 -0700
Subject: [PATCH] inode-diet: Move i_bdev into a union

Move the i_bdev pointer in struct inode into a union.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/inode.c         | 2 +-
 include/linux/fs.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/inode.c b/fs/inode.c
index 77e254792025..41b462232f7f 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -254,7 +254,7 @@ void clear_inode(struct inode *inode)
 	DQUOT_DROP(inode);
 	if (inode->i_sb && inode->i_sb->s_op->clear_inode)
 		inode->i_sb->s_op->clear_inode(inode);
-	if (inode->i_bdev)
+	if (S_ISBLK(inode->i_mode) && inode->i_bdev)
 		bd_forget(inode);
 	if (inode->i_cdev)
 		cd_forget(inode);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index ca695fc8d69c..98ff684a5b1c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -531,8 +531,8 @@ struct inode {
 	struct list_head	i_devices;
 	union {
 		struct pipe_inode_info	*i_pipe;
+		struct block_device	*i_bdev;
 	};
-	struct block_device	*i_bdev;
 	struct cdev		*i_cdev;
 	int			i_cindex;
 
-- 
cgit v1.2.3


From 577c4eb09d1034d0739e3135fd2cff50588024be Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:49 -0700
Subject: [PATCH] inode-diet: Move i_cdev into a union

Move the i_cdev pointer in struct inode into a union.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/file_table.c    | 2 +-
 fs/inode.c         | 2 +-
 include/linux/fs.h | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/file_table.c b/fs/file_table.c
index 0131ba06e1ee..bc35a40417d7 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -169,7 +169,7 @@ void fastcall __fput(struct file *file)
 	if (file->f_op && file->f_op->release)
 		file->f_op->release(inode, file);
 	security_file_free(file);
-	if (unlikely(inode->i_cdev != NULL))
+	if (unlikely(S_ISCHR(inode->i_mode) && inode->i_cdev != NULL))
 		cdev_put(inode->i_cdev);
 	fops_put(file->f_op);
 	if (file->f_mode & FMODE_WRITE)
diff --git a/fs/inode.c b/fs/inode.c
index 41b462232f7f..f5c04dd9ae8a 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -256,7 +256,7 @@ void clear_inode(struct inode *inode)
 		inode->i_sb->s_op->clear_inode(inode);
 	if (S_ISBLK(inode->i_mode) && inode->i_bdev)
 		bd_forget(inode);
-	if (inode->i_cdev)
+	if (S_ISCHR(inode->i_mode) && inode->i_cdev)
 		cd_forget(inode);
 	inode->i_state = I_CLEAR;
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 98ff684a5b1c..192e69bb55b5 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -532,8 +532,8 @@ struct inode {
 	union {
 		struct pipe_inode_info	*i_pipe;
 		struct block_device	*i_bdev;
+		struct cdev		*i_cdev;
 	};
-	struct cdev		*i_cdev;
 	int			i_cindex;
 
 	__u32			i_generation;
-- 
cgit v1.2.3


From ba52de123d454b57369f291348266d86f4b35070 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:49 -0700
Subject: [PATCH] inode-diet: Eliminate i_blksize from the inode structure

This eliminates the i_blksize field from struct inode.  Filesystems that want
to provide a per-inode st_blksize can do so by providing their own getattr
routine instead of using the generic_fillattr() function.

Note that some filesystems were providing pretty much random (and incorrect)
values for i_blksize.

[bunk@stusta.de: cleanup]
[akpm@osdl.org: generic_fillattr() fix]
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/platforms/cell/spufs/inode.c |  1 -
 arch/s390/hypfs/inode.c                   |  1 -
 drivers/block/loop.c                      |  7 +++++--
 drivers/infiniband/hw/ipath/ipath_fs.c    |  1 -
 drivers/isdn/capi/capifs.c                |  2 --
 drivers/misc/ibmasm/ibmasmfs.c            |  1 -
 drivers/oprofile/oprofilefs.c             |  1 -
 drivers/usb/core/inode.c                  |  1 -
 drivers/usb/gadget/inode.c                |  1 -
 fs/9p/vfs_inode.c                         |  4 +---
 fs/adfs/inode.c                           |  1 -
 fs/afs/inode.c                            |  1 -
 fs/autofs/inode.c                         |  1 -
 fs/autofs4/inode.c                        |  1 -
 fs/befs/linuxvfs.c                        |  1 -
 fs/bfs/dir.c                              |  2 +-
 fs/bfs/inode.c                            |  1 -
 fs/binfmt_misc.c                          |  1 -
 fs/cifs/cifsfs.c                          |  1 -
 fs/cifs/readdir.c                         |  5 ++---
 fs/coda/coda_linux.c                      |  2 --
 fs/configfs/inode.c                       |  1 -
 fs/cramfs/inode.c                         |  1 -
 fs/debugfs/inode.c                        |  1 -
 fs/devpts/inode.c                         |  2 --
 fs/eventpoll.c                            |  1 -
 fs/ext2/ialloc.c                          |  1 -
 fs/ext2/inode.c                           |  1 -
 fs/ext3/ialloc.c                          |  1 -
 fs/ext3/inode.c                           |  3 ---
 fs/fat/inode.c                            |  3 ---
 fs/freevxfs/vxfs_inode.c                  |  1 -
 fs/fuse/inode.c                           |  1 -
 fs/hfs/inode.c                            |  2 --
 fs/hfsplus/inode.c                        |  2 --
 fs/hostfs/hostfs_kern.c                   |  1 -
 fs/hpfs/inode.c                           |  1 -
 fs/hppfs/hppfs_kern.c                     |  1 -
 fs/hugetlbfs/inode.c                      |  1 -
 fs/isofs/inode.c                          |  3 +--
 fs/jffs/inode-v23.c                       |  2 --
 fs/jffs2/fs.c                             |  2 --
 fs/jfs/jfs_extent.c                       |  2 +-
 fs/jfs/jfs_imap.c                         |  1 -
 fs/jfs/jfs_inode.c                        |  1 -
 fs/jfs/jfs_metapage.c                     |  2 +-
 fs/libfs.c                                |  2 --
 fs/minix/bitmap.c                         |  2 +-
 fs/minix/inode.c                          |  4 ++--
 fs/ncpfs/inode.c                          |  1 -
 fs/nfs/inode.c                            |  4 ----
 fs/ntfs/inode.c                           |  4 ----
 fs/ntfs/mft.c                             |  5 -----
 fs/ocfs2/dlm/dlmfs.c                      |  2 --
 fs/ocfs2/inode.c                          |  4 ----
 fs/pipe.c                                 |  1 -
 fs/qnx4/inode.c                           |  1 -
 fs/ramfs/inode.c                          |  1 -
 fs/reiserfs/inode.c                       |  4 ----
 fs/smbfs/inode.c                          |  2 --
 fs/smbfs/proc.c                           |  1 -
 fs/stat.c                                 |  3 ++-
 fs/sysfs/inode.c                          |  1 -
 fs/sysv/ialloc.c                          |  2 +-
 fs/sysv/inode.c                           |  2 +-
 fs/udf/ialloc.c                           |  1 -
 fs/udf/inode.c                            |  2 --
 fs/ufs/ialloc.c                           |  1 -
 fs/ufs/inode.c                            |  1 -
 fs/xfs/linux-2.6/xfs_super.c              |  1 -
 fs/xfs/linux-2.6/xfs_vnode.c              |  1 -
 include/linux/fs.h                        |  1 -
 include/linux/nfsd/nfsfh.h                | 10 ++--------
 include/linux/smb.h                       |  1 -
 ipc/mqueue.c                              |  1 -
 kernel/cpuset.c                           |  1 -
 mm/shmem.c                                |  1 -
 net/sunrpc/rpc_pipe.c                     |  1 -
 security/inode.c                          |  1 -
 security/selinux/selinuxfs.c              |  1 -
 80 files changed, 21 insertions(+), 125 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index b837f12a84ae..3950ddccb2c8 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -82,7 +82,6 @@ spufs_new_inode(struct super_block *sb, int mode)
 	inode->i_mode = mode;
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 out:
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index 8735024d235b..813fc21358f9 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -91,7 +91,6 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode)
 		ret->i_mode = mode;
 		ret->i_uid = hypfs_info->uid;
 		ret->i_gid = hypfs_info->gid;
-		ret->i_blksize = PAGE_CACHE_SIZE;
 		ret->i_blocks = 0;
 		ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
 		if (mode & S_IFDIR)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 7b3b94ddddcc..c774121684d7 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -662,7 +662,8 @@ static void do_loop_switch(struct loop_device *lo, struct switch_request *p)
 
 	mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask);
 	lo->lo_backing_file = file;
-	lo->lo_blocksize = mapping->host->i_blksize;
+	lo->lo_blocksize = S_ISBLK(mapping->host->i_mode) ?
+		mapping->host->i_bdev->bd_block_size : PAGE_SIZE;
 	lo->old_gfp_mask = mapping_gfp_mask(mapping);
 	mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
 	complete(&p->wait);
@@ -794,7 +795,9 @@ static int loop_set_fd(struct loop_device *lo, struct file *lo_file,
 		if (!(lo_flags & LO_FLAGS_USE_AOPS) && !file->f_op->write)
 			lo_flags |= LO_FLAGS_READ_ONLY;
 
-		lo_blocksize = inode->i_blksize;
+		lo_blocksize = S_ISBLK(inode->i_mode) ?
+			inode->i_bdev->bd_block_size : PAGE_SIZE;
+
 		error = 0;
 	} else {
 		goto out_putf;
diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c
index 055cdd089b28..c8a8af0fe471 100644
--- a/drivers/infiniband/hw/ipath/ipath_fs.c
+++ b/drivers/infiniband/hw/ipath/ipath_fs.c
@@ -61,7 +61,6 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry,
 	inode->i_mode = mode;
 	inode->i_uid = 0;
 	inode->i_gid = 0;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	inode->i_private = data;
diff --git a/drivers/isdn/capi/capifs.c b/drivers/isdn/capi/capifs.c
index 9ea6bd0ddc35..2dd1b57b0ba4 100644
--- a/drivers/isdn/capi/capifs.c
+++ b/drivers/isdn/capi/capifs.c
@@ -104,7 +104,6 @@ capifs_fill_super(struct super_block *s, void *data, int silent)
 	inode->i_ino = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
-	inode->i_blksize = 1024;
 	inode->i_uid = inode->i_gid = 0;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
 	inode->i_op = &simple_dir_inode_operations;
@@ -149,7 +148,6 @@ void capifs_new_ncci(unsigned int number, dev_t device)
 	if (!inode)
 		return;
 	inode->i_ino = number+2;
-	inode->i_blksize = 1024;
 	inode->i_uid = config.setuid ? config.uid : current->fsuid;
 	inode->i_gid = config.setgid ? config.gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c
index 0e909b617226..b99dc507de2e 100644
--- a/drivers/misc/ibmasm/ibmasmfs.c
+++ b/drivers/misc/ibmasm/ibmasmfs.c
@@ -147,7 +147,6 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode)
 	if (ret) {
 		ret->i_mode = mode;
 		ret->i_uid = ret->i_gid = 0;
-		ret->i_blksize = PAGE_CACHE_SIZE;
 		ret->i_blocks = 0;
 		ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
 	}
diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c
index deb37354785b..5756401fb15b 100644
--- a/drivers/oprofile/oprofilefs.c
+++ b/drivers/oprofile/oprofilefs.c
@@ -31,7 +31,6 @@ static struct inode * oprofilefs_get_inode(struct super_block * sb, int mode)
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	}
diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index 482f253085e5..58b4b1012120 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -249,7 +249,6 @@ static struct inode *usbfs_get_inode (struct super_block *sb, int mode, dev_t de
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index ffaa8c1afad8..2a7162d89799 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1966,7 +1966,6 @@ gadgetfs_make_inode (struct super_block *sb,
 		inode->i_mode = mode;
 		inode->i_uid = default_uid;
 		inode->i_gid = default_gid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime
 				= CURRENT_TIME;
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index eae50c9d6dc4..7a7ec2d1d2f4 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -204,7 +204,6 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = sb->s_blocksize;
 		inode->i_blocks = 0;
 		inode->i_rdev = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
@@ -950,9 +949,8 @@ v9fs_stat2inode(struct v9fs_stat *stat, struct inode *inode,
 
 	inode->i_size = stat->length;
 
-	inode->i_blksize = sb->s_blocksize;
 	inode->i_blocks =
-	    (inode->i_size + inode->i_blksize - 1) >> sb->s_blocksize_bits;
+	    (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
 }
 
 /**
diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c
index 534f3eecc985..7e7a04be1278 100644
--- a/fs/adfs/inode.c
+++ b/fs/adfs/inode.c
@@ -269,7 +269,6 @@ adfs_iget(struct super_block *sb, struct object_info *obj)
 	inode->i_ino	 = obj->file_id;
 	inode->i_size	 = obj->size;
 	inode->i_nlink	 = 2;
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks	 = (inode->i_size + sb->s_blocksize - 1) >>
 			    sb->s_blocksize_bits;
 
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 4ebb30a50ed5..6f37754906c2 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -72,7 +72,6 @@ static int afs_inode_map_status(struct afs_vnode *vnode)
 	inode->i_ctime.tv_sec	= vnode->status.mtime_server;
 	inode->i_ctime.tv_nsec	= 0;
 	inode->i_atime		= inode->i_mtime = inode->i_ctime;
-	inode->i_blksize	= PAGE_CACHE_SIZE;
 	inode->i_blocks		= 0;
 	inode->i_version	= vnode->fid.unique;
 	inode->i_mapping->a_ops	= &afs_fs_aops;
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index d39d2acd9b38..2c9759baad61 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -216,7 +216,6 @@ static void autofs_read_inode(struct inode *inode)
 	inode->i_nlink = 2;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
-	inode->i_blksize = 1024;
 
 	if ( ino == AUTOFS_ROOT_INO ) {
 		inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 11a6a9ae51b7..800ce876caec 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -447,7 +447,6 @@ struct inode *autofs4_get_inode(struct super_block *sb,
 		inode->i_uid = 0;
 		inode->i_gid = 0;
 	}
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index f6676fbe9484..57020c7a7e65 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -365,7 +365,6 @@ befs_read_inode(struct inode *inode)
 	inode->i_mtime.tv_nsec = 0;   /* lower 16 bits are not a time */	
 	inode->i_ctime = inode->i_mtime;
 	inode->i_atime = inode->i_mtime;
-	inode->i_blksize = befs_sb->block_size;
 
 	befs_ino->i_inode_num = fsrun_to_cpu(sb, raw_inode->inode_num);
 	befs_ino->i_parent = fsrun_to_cpu(sb, raw_inode->parent);
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index 26fad9621738..dcf04cb13283 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -102,7 +102,7 @@ static int bfs_create(struct inode * dir, struct dentry * dentry, int mode,
 	inode->i_uid = current->fsuid;
 	inode->i_gid = (dir->i_mode & S_ISGID) ? dir->i_gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	inode->i_op = &bfs_file_inops;
 	inode->i_fop = &bfs_file_operations;
 	inode->i_mapping->a_ops = &bfs_aops;
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 8fc2e8e49dbe..ed27ffb3459e 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -76,7 +76,6 @@ static void bfs_read_inode(struct inode * inode)
 	inode->i_size = BFS_FILESIZE(di);
 	inode->i_blocks = BFS_FILEBLOCKS(di);
         if (inode->i_size || inode->i_blocks) dprintf("Registered inode with %lld size, %ld blocks\n", inode->i_size, inode->i_blocks);
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_atime.tv_sec =  le32_to_cpu(di->i_atime);
 	inode->i_mtime.tv_sec =  le32_to_cpu(di->i_mtime);
 	inode->i_ctime.tv_sec =  le32_to_cpu(di->i_ctime);
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 6759b9839ce8..66ba137f8661 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -507,7 +507,6 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime =
 			current_fs_time(inode->i_sb);
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 4197a5043f13..22bcf4d7e7ae 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -253,7 +253,6 @@ cifs_alloc_inode(struct super_block *sb)
 	file data or metadata */
 	cifs_inode->clientCanCacheRead = FALSE;
 	cifs_inode->clientCanCacheAll = FALSE;
-	cifs_inode->vfs_inode.i_blksize = CIFS_MAX_MSGSIZE;
 	cifs_inode->vfs_inode.i_blkbits = 14;  /* 2**14 = CIFS_MAX_MSGSIZE */
 	cifs_inode->vfs_inode.i_flags = S_NOATIME | S_NOCMTIME;
 	INIT_LIST_HEAD(&cifs_inode->openFileList);
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 9aeb58a7d369..b27b34537bf2 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -216,10 +216,9 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type,
 
 	if (allocation_size < end_of_file)
 		cFYI(1, ("May be sparse file, allocation less than file size"));
-	cFYI(1, ("File Size %ld and blocks %llu and blocksize %ld",
+	cFYI(1, ("File Size %ld and blocks %llu",
 		(unsigned long)tmp_inode->i_size,
-		(unsigned long long)tmp_inode->i_blocks,
-		tmp_inode->i_blksize));
+		(unsigned long long)tmp_inode->i_blocks));
 	if (S_ISREG(tmp_inode->i_mode)) {
 		cFYI(1, ("File inode"));
 		tmp_inode->i_op = &cifs_file_inode_ops;
diff --git a/fs/coda/coda_linux.c b/fs/coda/coda_linux.c
index 5597080cb811..95a54253c047 100644
--- a/fs/coda/coda_linux.c
+++ b/fs/coda/coda_linux.c
@@ -110,8 +110,6 @@ void coda_vattr_to_iattr(struct inode *inode, struct coda_vattr *attr)
 	        inode->i_nlink = attr->va_nlink;
 	if (attr->va_size != -1)
 	        inode->i_size = attr->va_size;
-	if (attr->va_blocksize != -1)
-		inode->i_blksize = attr->va_blocksize;
 	if (attr->va_size != -1)
 		inode->i_blocks = (attr->va_size + 511) >> 9;
 	if (attr->va_atime.tv_sec != -1) 
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index 5047e6a7581e..fb18917954a9 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -135,7 +135,6 @@ struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd)
 {
 	struct inode * inode = new_inode(configfs_sb);
 	if (inode) {
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &configfs_aops;
 		inode->i_mapping->backing_dev_info = &configfs_backing_dev_info;
diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index d09b6777c41a..ad96b6990715 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -73,7 +73,6 @@ static int cramfs_iget5_set(struct inode *inode, void *opaque)
 	inode->i_uid = cramfs_inode->uid;
 	inode->i_size = cramfs_inode->size;
 	inode->i_blocks = (cramfs_inode->size - 1) / 512 + 1;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_gid = cramfs_inode->gid;
 	/* Struct copy intentional */
 	inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime;
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 717f4821ed02..269e649e6dc6 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -40,7 +40,6 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 5bf06a10dddf..5f7b5a6025bf 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -113,7 +113,6 @@ devpts_fill_super(struct super_block *s, void *data, int silent)
 	inode->i_ino = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
-	inode->i_blksize = 1024;
 	inode->i_uid = inode->i_gid = 0;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
 	inode->i_op = &simple_dir_inode_operations;
@@ -172,7 +171,6 @@ int devpts_pty_new(struct tty_struct *tty)
 		return -ENOMEM;
 
 	inode->i_ino = number+2;
-	inode->i_blksize = 1024;
 	inode->i_uid = config.setuid ? config.uid : current->fsuid;
 	inode->i_gid = config.setgid ? config.gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 3a3567433b92..8d544334bcd2 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1590,7 +1590,6 @@ static struct inode *ep_eventpoll_inode(void)
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	inode->i_blksize = PAGE_SIZE;
 	return inode;
 
 eexit_1:
diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index 695f69ccf908..2cb545bf0f3c 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -574,7 +574,6 @@ got:
 	inode->i_mode = mode;
 
 	inode->i_ino = ino;
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size (for stat), not the fs block size */
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	memset(ei->i_data, 0, sizeof(ei->i_data));
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index fb4d3220eb8d..dd4e14c221e0 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -1094,7 +1094,6 @@ void ext2_read_inode (struct inode * inode)
 		brelse (bh);
 		goto bad_inode;
 	}
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size (for stat), not the fs block size */
 	inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
 	ei->i_faddr = le32_to_cpu(raw_inode->i_faddr);
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 1e4ded8aa3cd..e45dbd651736 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -559,7 +559,6 @@ got:
 
 	inode->i_ino = ino;
 	/* This is the optimal IO size (for stat), not the fs block size */
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index c9fc15f8b609..dcf4f1dd108b 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2632,9 +2632,6 @@ void ext3_read_inode(struct inode * inode)
 		 * recovery code: that's fine, we're about to complete
 		 * the process of deleting those. */
 	}
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size
-					 * (for stat), not the fs block
-					 * size */  
 	inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
 #ifdef EXT3_FRAGMENTS
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index e1035a590664..ab96ae823753 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -370,8 +370,6 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 			inode->i_flags |= S_IMMUTABLE;
 	}
 	MSDOS_I(inode)->i_attrs = de->attr & ATTR_UNUSED;
-	/* this is as close to the truth as we can get ... */
-	inode->i_blksize = sbi->cluster_size;
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
 	inode->i_mtime.tv_sec =
@@ -1131,7 +1129,6 @@ static int fat_read_root(struct inode *inode)
 		MSDOS_I(inode)->i_start = 0;
 		inode->i_size = sbi->dir_entries * sizeof(struct msdos_dir_entry);
 	}
-	inode->i_blksize = sbi->cluster_size;
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
 	MSDOS_I(inode)->i_logstart = 0;
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index 32a82ed108e4..4786d51ad3bd 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -239,7 +239,6 @@ vxfs_iinit(struct inode *ip, struct vxfs_inode_info *vip)
 	ip->i_ctime.tv_nsec = 0;
 	ip->i_mtime.tv_nsec = 0;
 
-	ip->i_blksize = PAGE_SIZE;
 	ip->i_blocks = vip->vii_blocks;
 	ip->i_generation = vip->vii_gen;
 
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 7d25092262ae..cb7cadb0b790 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -118,7 +118,6 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr)
 	inode->i_uid     = attr->uid;
 	inode->i_gid     = attr->gid;
 	i_size_write(inode, attr->size);
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks  = attr->blocks;
 	inode->i_atime.tv_sec   = attr->atime;
 	inode->i_atime.tv_nsec  = attr->atimensec;
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 315cf44a90b2..d05641c35fc9 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -154,7 +154,6 @@ struct inode *hfs_new_inode(struct inode *dir, struct qstr *name, int mode)
 	inode->i_gid = current->fsgid;
 	inode->i_nlink = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blksize = HFS_SB(sb)->alloc_blksz;
 	HFS_I(inode)->flags = 0;
 	HFS_I(inode)->rsrc_inode = NULL;
 	HFS_I(inode)->fs_blocks = 0;
@@ -284,7 +283,6 @@ static int hfs_read_inode(struct inode *inode, void *data)
 	inode->i_uid = hsb->s_uid;
 	inode->i_gid = hsb->s_gid;
 	inode->i_nlink = 1;
-	inode->i_blksize = HFS_SB(inode->i_sb)->alloc_blksz;
 
 	if (idata->key)
 		HFS_I(inode)->cat_key = *idata->key;
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 924ecdef8091..0eb1a6092668 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -304,7 +304,6 @@ struct inode *hfsplus_new_inode(struct super_block *sb, int mode)
 	inode->i_gid = current->fsgid;
 	inode->i_nlink = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blksize = HFSPLUS_SB(sb).alloc_blksz;
 	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
 	init_MUTEX(&HFSPLUS_I(inode).extents_lock);
 	atomic_set(&HFSPLUS_I(inode).opencnt, 0);
@@ -407,7 +406,6 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
 	type = hfs_bnode_read_u16(fd->bnode, fd->entryoffset);
 
 	HFSPLUS_I(inode).dev = 0;
-	inode->i_blksize = HFSPLUS_SB(inode->i_sb).alloc_blksz;
 	if (type == HFSPLUS_FOLDER) {
 		struct hfsplus_cat_folder *folder = &entry.folder;
 
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index b82e3d9c8790..322e876c35ed 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -156,7 +156,6 @@ static int read_name(struct inode *ino, char *name)
 	ino->i_mode = i_mode;
 	ino->i_nlink = i_nlink;
 	ino->i_size = i_size;
-	ino->i_blksize = i_blksize;
 	ino->i_blocks = i_blocks;
 	return(0);
 }
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index 56f2c338c4d9..bcf6ee36e065 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -17,7 +17,6 @@ void hpfs_init_inode(struct inode *i)
 	i->i_gid = hpfs_sb(sb)->sb_gid;
 	i->i_mode = hpfs_sb(sb)->sb_mode;
 	hpfs_inode->i_conv = hpfs_sb(sb)->sb_conv;
-	i->i_blksize = 512;
 	i->i_size = -1;
 	i->i_blocks = -1;
 	
diff --git a/fs/hppfs/hppfs_kern.c b/fs/hppfs/hppfs_kern.c
index 3a9bdf58166f..dcb6d2e988b8 100644
--- a/fs/hppfs/hppfs_kern.c
+++ b/fs/hppfs/hppfs_kern.c
@@ -152,7 +152,6 @@ static void hppfs_read_inode(struct inode *ino)
 	ino->i_mode = proc_ino->i_mode;
 	ino->i_nlink = proc_ino->i_nlink;
 	ino->i_size = proc_ino->i_size;
-	ino->i_blksize = proc_ino->i_blksize;
 	ino->i_blocks = proc_ino->i_blocks;
 }
 
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index c3920c96dadf..e025a31b4c64 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -357,7 +357,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid,
 		inode->i_mode = mode;
 		inode->i_uid = uid;
 		inode->i_gid = gid;
-		inode->i_blksize = HPAGE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &hugetlbfs_aops;
 		inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info;
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 10e47897bac7..4527692f432b 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -1235,7 +1235,7 @@ static void isofs_read_inode(struct inode *inode)
 	}
 	inode->i_uid = sbi->s_uid;
 	inode->i_gid = sbi->s_gid;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 
 	ei->i_format_parm[0] = 0;
 	ei->i_format_parm[1] = 0;
@@ -1291,7 +1291,6 @@ static void isofs_read_inode(struct inode *inode)
 			      isonum_711 (de->ext_attr_length));
 
 	/* Set the number of blocks for stat() - should be done before RR */
-	inode->i_blksize = PAGE_CACHE_SIZE; /* For stat() only */
 	inode->i_blocks  = (inode->i_size + 511) >> 9;
 
 	/*
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index 7358ef87f16b..f5cf9c93e243 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -364,7 +364,6 @@ jffs_new_inode(const struct inode * dir, struct jffs_raw_inode *raw_inode,
 	inode->i_ctime.tv_nsec = 0;
 	inode->i_mtime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = (inode->i_size + 511) >> 9;
 
 	f = jffs_find_file(c, raw_inode->ino);
@@ -1706,7 +1705,6 @@ jffs_read_inode(struct inode *inode)
 	inode->i_mtime.tv_nsec = 
 	inode->i_ctime.tv_nsec = 0;
 
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = (inode->i_size + 511) >> 9;
 	if (S_ISREG(inode->i_mode)) {
 		inode->i_op = &jffs_file_inode_operations;
diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index 4780f82825d6..72d9909d95ff 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -263,7 +263,6 @@ void jffs2_read_inode (struct inode *inode)
 
 	inode->i_nlink = f->inocache->nlink;
 
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = (inode->i_size + 511) >> 9;
 
 	switch (inode->i_mode & S_IFMT) {
@@ -449,7 +448,6 @@ struct inode *jffs2_new_inode (struct inode *dir_i, int mode, struct jffs2_raw_i
 	inode->i_atime = inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC;
 	ri->atime = ri->mtime = ri->ctime = cpu_to_je32(I_SEC(inode->i_mtime));
 
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_size = 0;
 
diff --git a/fs/jfs/jfs_extent.c b/fs/jfs/jfs_extent.c
index 4d52593a5fc6..4c74f0944f7e 100644
--- a/fs/jfs/jfs_extent.c
+++ b/fs/jfs/jfs_extent.c
@@ -468,7 +468,7 @@ int extRecord(struct inode *ip, xad_t * xp)
 int extFill(struct inode *ip, xad_t * xp)
 {
 	int rc, nbperpage = JFS_SBI(ip->i_sb)->nbperpage;
-	s64 blkno = offsetXAD(xp) >> ip->i_blksize;
+	s64 blkno = offsetXAD(xp) >> ip->i_blkbits;
 
 //      assert(ISSPARSE(ip));
 
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index ccbe60aff83d..369d7f39c040 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -3115,7 +3115,6 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip)
 	ip->i_mtime.tv_nsec = le32_to_cpu(dip->di_mtime.tv_nsec);
 	ip->i_ctime.tv_sec = le32_to_cpu(dip->di_ctime.tv_sec);
 	ip->i_ctime.tv_nsec = le32_to_cpu(dip->di_ctime.tv_nsec);
-	ip->i_blksize = ip->i_sb->s_blocksize;
 	ip->i_blocks = LBLK2PBLK(ip->i_sb, le64_to_cpu(dip->di_nblocks));
 	ip->i_generation = le32_to_cpu(dip->di_gen);
 
diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c
index 495df402916d..bffaca9ae3a2 100644
--- a/fs/jfs/jfs_inode.c
+++ b/fs/jfs/jfs_inode.c
@@ -115,7 +115,6 @@ struct inode *ialloc(struct inode *parent, umode_t mode)
 	}
 	jfs_inode->mode2 |= mode;
 
-	inode->i_blksize = sb->s_blocksize;
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	jfs_inode->otime = inode->i_ctime.tv_sec;
diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c
index e1e0a6e6ebdf..f5afc129d6b1 100644
--- a/fs/jfs/jfs_metapage.c
+++ b/fs/jfs/jfs_metapage.c
@@ -257,7 +257,7 @@ static sector_t metapage_get_blocks(struct inode *inode, sector_t lblock,
 	int rc = 0;
 	int xflag;
 	s64 xaddr;
-	sector_t file_blocks = (inode->i_size + inode->i_blksize - 1) >>
+	sector_t file_blocks = (inode->i_size + inode->i_sb->s_blocksize - 1) >>
 			       inode->i_blkbits;
 
 	if (lblock >= file_blocks)
diff --git a/fs/libfs.c b/fs/libfs.c
index 2751793beeaa..8db5afb7b0a7 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -383,7 +383,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files
 		return -ENOMEM;
 	inode->i_mode = S_IFDIR | 0755;
 	inode->i_uid = inode->i_gid = 0;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	inode->i_op = &simple_dir_inode_operations;
@@ -405,7 +404,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files
 			goto out;
 		inode->i_mode = S_IFREG | files->mode;
 		inode->i_uid = inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		inode->i_fop = files->ops;
diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c
index 4a6abc49418e..df6b1075b549 100644
--- a/fs/minix/bitmap.c
+++ b/fs/minix/bitmap.c
@@ -254,7 +254,7 @@ struct inode * minix_new_inode(const struct inode * dir, int * error)
 	inode->i_gid = (dir->i_mode & S_ISGID) ? dir->i_gid : current->fsgid;
 	inode->i_ino = j;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	memset(&minix_i(inode)->u, 0, sizeof(minix_i(inode)->u));
 	insert_inode_hash(inode);
 	mark_inode_dirty(inode);
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 826b9d830650..c11a4b9fb863 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -396,7 +396,7 @@ static void V1_minix_read_inode(struct inode * inode)
 	inode->i_mtime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
 	inode->i_ctime.tv_nsec = 0;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	for (i = 0; i < 9; i++)
 		minix_inode->u.i1_data[i] = raw_inode->i_zone[i];
 	minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0]));
@@ -429,7 +429,7 @@ static void V2_minix_read_inode(struct inode * inode)
 	inode->i_mtime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
 	inode->i_ctime.tv_nsec = 0;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	for (i = 0; i < 10; i++)
 		minix_inode->u.i2_data[i] = raw_inode->i_zone[i];
 	minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0]));
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 8244710e97dd..42e3bef270c9 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -223,7 +223,6 @@ static void ncp_set_attr(struct inode *inode, struct ncp_entry_info *nwinfo)
 	inode->i_nlink = 1;
 	inode->i_uid = server->m.uid;
 	inode->i_gid = server->m.gid;
-	inode->i_blksize = NCP_BLOCK_SIZE;
 
 	ncp_update_dates(inode, &nwinfo->i);
 	ncp_update_inode(inode, nwinfo);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 931f52a19579..bc9376ca86cd 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -277,10 +277,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 			 * report the blocks in 512byte units
 			 */
 			inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
-			inode->i_blksize = inode->i_sb->s_blocksize;
 		} else {
 			inode->i_blocks = fattr->du.nfs2.blocks;
-			inode->i_blksize = fattr->du.nfs2.blocksize;
 		}
 		nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
 		nfsi->attrtimeo_timestamp = jiffies;
@@ -969,10 +967,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 		 * report the blocks in 512byte units
 		 */
 		inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
-		inode->i_blksize = inode->i_sb->s_blocksize;
  	} else {
  		inode->i_blocks = fattr->du.nfs2.blocks;
- 		inode->i_blksize = fattr->du.nfs2.blocksize;
  	}
 
 	if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 &&
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 31852121b3f5..933dbd89c2a4 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -556,8 +556,6 @@ static int ntfs_read_locked_inode(struct inode *vi)
 
 	/* Setup the generic vfs inode parts now. */
 
-	/* This is the optimal IO size (for stat), not the fs block size. */
-	vi->i_blksize = PAGE_CACHE_SIZE;
 	/*
 	 * This is for checking whether an inode has changed w.r.t. a file so
 	 * that the file can be updated if necessary (compare with f_version).
@@ -1234,7 +1232,6 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
 	base_ni = NTFS_I(base_vi);
 
 	/* Just mirror the values from the base inode. */
-	vi->i_blksize	= base_vi->i_blksize;
 	vi->i_version	= base_vi->i_version;
 	vi->i_uid	= base_vi->i_uid;
 	vi->i_gid	= base_vi->i_gid;
@@ -1504,7 +1501,6 @@ static int ntfs_read_locked_index_inode(struct inode *base_vi, struct inode *vi)
 	ni	= NTFS_I(vi);
 	base_ni = NTFS_I(base_vi);
 	/* Just mirror the values from the base inode. */
-	vi->i_blksize	= base_vi->i_blksize;
 	vi->i_version	= base_vi->i_version;
 	vi->i_uid	= base_vi->i_uid;
 	vi->i_gid	= base_vi->i_gid;
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 578fb3d5e803..584260fd6848 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -2637,11 +2637,6 @@ mft_rec_already_initialized:
 			goto undo_mftbmp_alloc;
 		}
 		vi->i_ino = bit;
-		/*
-		 * This is the optimal IO size (for stat), not the fs block
-		 * size.
-		 */
-		vi->i_blksize = PAGE_CACHE_SIZE;
 		/*
 		 * This is for checking whether an inode has changed w.r.t. a
 		 * file so that the file can be updated if necessary (compare
diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c
index 0ff0898a0b9c..0368c6402182 100644
--- a/fs/ocfs2/dlm/dlmfs.c
+++ b/fs/ocfs2/dlm/dlmfs.c
@@ -335,7 +335,6 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
@@ -362,7 +361,6 @@ static struct inode *dlmfs_get_inode(struct inode *parent,
 	inode->i_mode = mode;
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 69d3db569166..16e8e74dc966 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -269,7 +269,6 @@ int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
 	inode->i_mode = le16_to_cpu(fe->i_mode);
 	inode->i_uid = le32_to_cpu(fe->i_uid);
 	inode->i_gid = le32_to_cpu(fe->i_gid);
-	inode->i_blksize = (u32)osb->s_clustersize;
 
 	/* Fast symlinks will have i_size but no allocated clusters. */
 	if (S_ISLNK(inode->i_mode) && !fe->i_clusters)
@@ -1258,8 +1257,6 @@ leave:
 void ocfs2_refresh_inode(struct inode *inode,
 			 struct ocfs2_dinode *fe)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
-
 	spin_lock(&OCFS2_I(inode)->ip_lock);
 
 	OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
@@ -1270,7 +1267,6 @@ void ocfs2_refresh_inode(struct inode *inode,
 	inode->i_uid = le32_to_cpu(fe->i_uid);
 	inode->i_gid = le32_to_cpu(fe->i_gid);
 	inode->i_mode = le16_to_cpu(fe->i_mode);
-	inode->i_blksize = (u32) osb->s_clustersize;
 	if (S_ISLNK(inode->i_mode) && le32_to_cpu(fe->i_clusters) == 0)
 		inode->i_blocks = 0;
 	else
diff --git a/fs/pipe.c b/fs/pipe.c
index 20352573e025..f3b6f71e9d0b 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -879,7 +879,6 @@ static struct inode * get_pipe_inode(void)
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	inode->i_blksize = PAGE_SIZE;
 
 	return inode;
 
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index fddbd61c68d0..5a41db2a218d 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -496,7 +496,6 @@ static void qnx4_read_inode(struct inode *inode)
 	inode->i_ctime.tv_sec   = le32_to_cpu(raw_inode->di_ctime);
 	inode->i_ctime.tv_nsec = 0;
 	inode->i_blocks  = le32_to_cpu(raw_inode->di_first_xtnt.xtnt_size);
-	inode->i_blksize = QNX4_DIR_ENTRY_SIZE;
 
 	memcpy(qnx4_inode, raw_inode, QNX4_DIR_ENTRY_SIZE);
 	if (S_ISREG(inode->i_mode)) {
diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c
index b9677335cc8d..bc0e51662424 100644
--- a/fs/ramfs/inode.c
+++ b/fs/ramfs/inode.c
@@ -58,7 +58,6 @@ struct inode *ramfs_get_inode(struct super_block *sb, int mode, dev_t dev)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &ramfs_aops;
 		inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info;
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 52f1e2136546..8810fda0da46 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -17,8 +17,6 @@
 #include <linux/writeback.h>
 #include <linux/quotaops.h>
 
-extern int reiserfs_default_io_size;	/* default io size devuned in super.c */
-
 static int reiserfs_commit_write(struct file *f, struct page *page,
 				 unsigned from, unsigned to);
 static int reiserfs_prepare_write(struct file *f, struct page *page,
@@ -1122,7 +1120,6 @@ static void init_inode(struct inode *inode, struct path *path)
 	ih = PATH_PITEM_HEAD(path);
 
 	copy_key(INODE_PKEY(inode), &(ih->ih_key));
-	inode->i_blksize = reiserfs_default_io_size;
 
 	INIT_LIST_HEAD(&(REISERFS_I(inode)->i_prealloc_list));
 	REISERFS_I(inode)->i_flags = 0;
@@ -1877,7 +1874,6 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	}
 	// these do not go to on-disk stat data
 	inode->i_ino = le32_to_cpu(ih.ih_key.k_objectid);
-	inode->i_blksize = reiserfs_default_io_size;
 
 	// store in in-core inode the key of stat data and version all
 	// object items will have (directory items will have old offset
diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c
index 92cf60aa6121..2c122ee83adb 100644
--- a/fs/smbfs/inode.c
+++ b/fs/smbfs/inode.c
@@ -166,7 +166,6 @@ smb_get_inode_attr(struct inode *inode, struct smb_fattr *fattr)
 	fattr->f_mtime	= inode->i_mtime;
 	fattr->f_ctime	= inode->i_ctime;
 	fattr->f_atime	= inode->i_atime;
-	fattr->f_blksize= inode->i_blksize;
 	fattr->f_blocks	= inode->i_blocks;
 
 	fattr->attr	= SMB_I(inode)->attr;
@@ -200,7 +199,6 @@ smb_set_inode_attr(struct inode *inode, struct smb_fattr *fattr)
 	inode->i_uid	= fattr->f_uid;
 	inode->i_gid	= fattr->f_gid;
 	inode->i_ctime	= fattr->f_ctime;
-	inode->i_blksize= fattr->f_blksize;
 	inode->i_blocks = fattr->f_blocks;
 	inode->i_size	= fattr->f_size;
 	inode->i_mtime	= fattr->f_mtime;
diff --git a/fs/smbfs/proc.c b/fs/smbfs/proc.c
index c3495059889d..40e174db9872 100644
--- a/fs/smbfs/proc.c
+++ b/fs/smbfs/proc.c
@@ -1826,7 +1826,6 @@ smb_init_dirent(struct smb_sb_info *server, struct smb_fattr *fattr)
 	fattr->f_nlink = 1;
 	fattr->f_uid = server->mnt->uid;
 	fattr->f_gid = server->mnt->gid;
-	fattr->f_blksize = SMB_ST_BLKSIZE;
 	fattr->f_unix = 0;
 }
 
diff --git a/fs/stat.c b/fs/stat.c
index 3a44dcf97da2..60a31d5e5966 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -14,6 +14,7 @@
 #include <linux/namei.h>
 #include <linux/security.h>
 #include <linux/syscalls.h>
+#include <linux/pagemap.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -32,7 +33,7 @@ void generic_fillattr(struct inode *inode, struct kstat *stat)
 	stat->ctime = inode->i_ctime;
 	stat->size = i_size_read(inode);
 	stat->blocks = inode->i_blocks;
-	stat->blksize = inode->i_blksize;
+	stat->blksize = (1 << inode->i_blkbits);
 }
 
 EXPORT_SYMBOL(generic_fillattr);
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index fd7cd5f843d2..e79e38d52c00 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -125,7 +125,6 @@ struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent * sd)
 {
 	struct inode * inode = new_inode(sysfs_sb);
 	if (inode) {
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &sysfs_aops;
 		inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info;
diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c
index 9b585d1081c0..115ab0d6f4bc 100644
--- a/fs/sysv/ialloc.c
+++ b/fs/sysv/ialloc.c
@@ -170,7 +170,7 @@ struct inode * sysv_new_inode(const struct inode * dir, mode_t mode)
 	inode->i_uid = current->fsuid;
 	inode->i_ino = fs16_to_cpu(sbi, ino);
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	memset(SYSV_I(inode)->i_data, 0, sizeof(SYSV_I(inode)->i_data));
 	SYSV_I(inode)->i_dir_start_lookup = 0;
 	insert_inode_hash(inode);
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index 58b2d22142ba..d63c5e48b050 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -201,7 +201,7 @@ static void sysv_read_inode(struct inode *inode)
 	inode->i_ctime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
 	inode->i_mtime.tv_nsec = 0;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 
 	si = SYSV_I(inode);
 	for (block = 0; block < 10+1+1+1; block++)
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index d1d38238ce65..8206983f2ebf 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -121,7 +121,6 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
 	UDF_I_LOCATION(inode).logicalBlockNum = block;
 	UDF_I_LOCATION(inode).partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum;
 	inode->i_ino = udf_get_lb_pblock(sb, UDF_I_LOCATION(inode), 0);
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = 0;
 	UDF_I_LENEATTR(inode) = 0;
 	UDF_I_LENALLOC(inode) = 0;
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 605f5111b6d8..b223b32db991 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -916,8 +916,6 @@ __udf_read_inode(struct inode *inode)
 	 *      i_nlink = 1
 	 *      i_op = NULL;
 	 */
-	inode->i_blksize = PAGE_SIZE;
-
 	bh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 0, &ident);
 
 	if (!bh)
diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c
index 9501dcd3b213..2ad1259c6eca 100644
--- a/fs/ufs/ialloc.c
+++ b/fs/ufs/ialloc.c
@@ -255,7 +255,6 @@ cg_found:
 		inode->i_gid = current->fsgid;
 
 	inode->i_ino = cg * uspi->s_ipg + bit;
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size (for stat), not the fs block size */
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	ufsi->i_flags = UFS_I(dir)->i_flags;
diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index 30c6e8a9446c..ee1eaa6f4ec2 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -741,7 +741,6 @@ void ufs_read_inode(struct inode * inode)
 		ufs1_read_inode(inode, ufs_inode + ufs_inotofsbo(inode->i_ino));
 	}
 
-	inode->i_blksize = PAGE_SIZE;/*This is the optimal IO size (for stat)*/
 	inode->i_version++;
 	ufsi->i_lastfrag =
 		(inode->i_size + uspi->s_fsize - 1) >> uspi->s_fshift;
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 4754f342a5d3..9df9ed37d219 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -171,7 +171,6 @@ xfs_revalidate_inode(
 		break;
 	}
 
-	inode->i_blksize = xfs_preferred_iosize(mp);
 	inode->i_generation = ip->i_d.di_gen;
 	i_size_write(inode, ip->i_d.di_size);
 	inode->i_blocks =
diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c
index 6628d96b6fd6..553fa731ade5 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.c
+++ b/fs/xfs/linux-2.6/xfs_vnode.c
@@ -122,7 +122,6 @@ vn_revalidate_core(
 	inode->i_blocks	    = vap->va_nblocks;
 	inode->i_mtime	    = vap->va_mtime;
 	inode->i_ctime	    = vap->va_ctime;
-	inode->i_blksize    = vap->va_blocksize;
 	if (vap->va_xflags & XFS_XFLAG_IMMUTABLE)
 		inode->i_flags |= S_IMMUTABLE;
 	else
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 192e69bb55b5..8f74dfbb2edd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -512,7 +512,6 @@ struct inode {
 	struct timespec		i_mtime;
 	struct timespec		i_ctime;
 	unsigned int		i_blkbits;
-	unsigned long		i_blksize;
 	unsigned long		i_version;
 	blkcnt_t		i_blocks;
 	unsigned short          i_bytes;
diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h
index f9edcd2ff3c8..31a3cb617ce0 100644
--- a/include/linux/nfsd/nfsfh.h
+++ b/include/linux/nfsd/nfsfh.h
@@ -269,14 +269,8 @@ fill_post_wcc(struct svc_fh *fhp)
 	fhp->fh_post_uid	= inode->i_uid;
 	fhp->fh_post_gid	= inode->i_gid;
 	fhp->fh_post_size       = inode->i_size;
-	if (inode->i_blksize) {
-		fhp->fh_post_blksize    = inode->i_blksize;
-		fhp->fh_post_blocks     = inode->i_blocks;
-	} else {
-		fhp->fh_post_blksize    = BLOCK_SIZE;
-		/* how much do we care for accuracy with MinixFS? */
-		fhp->fh_post_blocks     = (inode->i_size+511) >> 9;
-	}
+	fhp->fh_post_blksize    = BLOCK_SIZE;
+	fhp->fh_post_blocks     = inode->i_blocks;
 	fhp->fh_post_rdev[0]    = htonl((u32)imajor(inode));
 	fhp->fh_post_rdev[1]    = htonl((u32)iminor(inode));
 	fhp->fh_post_atime      = inode->i_atime;
diff --git a/include/linux/smb.h b/include/linux/smb.h
index 6df3b1501559..f098dff93f6b 100644
--- a/include/linux/smb.h
+++ b/include/linux/smb.h
@@ -89,7 +89,6 @@ struct smb_fattr {
 	struct timespec	f_atime;
 	struct timespec f_mtime;
 	struct timespec f_ctime;
-	unsigned long	f_blksize;
 	unsigned long	f_blocks;
 	int		f_unix;
 };
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index f8381f0660b2..840f8a6fb85f 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -115,7 +115,6 @@ static struct inode *mqueue_get_inode(struct super_block *sb, int mode,
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mtime = inode->i_ctime = inode->i_atime =
 				CURRENT_TIME;
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index cff41511269f..1b32c2c04c15 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -289,7 +289,6 @@ static struct inode *cpuset_new_inode(mode_t mode)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		inode->i_mapping->backing_dev_info = &cpuset_backing_dev_info;
diff --git a/mm/shmem.c b/mm/shmem.c
index 0a8e29cf87e0..eda907c3a86a 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1351,7 +1351,6 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &shmem_aops;
 		inode->i_mapping->backing_dev_info = &shmem_backing_dev_info;
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index f65bea74734d..700c6e061a04 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -488,7 +488,6 @@ rpc_get_inode(struct super_block *sb, int mode)
 		return NULL;
 	inode->i_mode = mode;
 	inode->i_uid = inode->i_gid = 0;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	switch(mode & S_IFMT) {
diff --git a/security/inode.c b/security/inode.c
index 176aacea8ca4..49ee51529396 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -64,7 +64,6 @@ static struct inode *get_inode(struct super_block *sb, int mode, dev_t dev)
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index 00534c302ba2..bab7b386cb8d 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -771,7 +771,6 @@ static struct inode *sel_make_inode(struct super_block *sb, int mode)
 	if (ret) {
 		ret->i_mode = mode;
 		ret->i_uid = ret->i_gid = 0;
-		ret->i_blksize = PAGE_CACHE_SIZE;
 		ret->i_blocks = 0;
 		ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
 	}
-- 
cgit v1.2.3


From ebba5f9fcb882306bef7175dee987342ec6fcf2f Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Wed, 27 Sep 2006 01:50:55 -0700
Subject: [PATCH] consistently use MAX_ERRNO in __syscall_return

Consistently use MAX_ERRNO when checking for errors in __syscall_return().

[ralf@linux-mips.org: build fix]
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-arm/unistd.h       | 3 ++-
 include/asm-arm26/unistd.h     | 3 ++-
 include/asm-frv/unistd.h       | 3 ++-
 include/asm-h8300/unistd.h     | 6 +++---
 include/asm-i386/unistd.h      | 5 +++--
 include/asm-m32r/unistd.h      | 5 +++--
 include/asm-m68k/unistd.h      | 5 +++--
 include/asm-m68knommu/unistd.h | 5 +++--
 include/asm-s390/unistd.h      | 4 +++-
 include/asm-sh/unistd.h        | 7 +++++--
 include/asm-sh64/unistd.h      | 6 ++++--
 include/asm-v850/unistd.h      | 5 +++--
 include/asm-x86_64/unistd.h    | 5 +++--
 include/linux/err.h            | 4 ++++
 14 files changed, 43 insertions(+), 23 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/unistd.h b/include/asm-arm/unistd.h
index 1e891f860ef3..2ab4078334bf 100644
--- a/include/asm-arm/unistd.h
+++ b/include/asm-arm/unistd.h
@@ -377,6 +377,7 @@
 #endif
 
 #ifdef __KERNEL__
+#include <linux/err.h>
 #include <linux/linkage.h>
 
 #define __sys2(x) #x
@@ -396,7 +397,7 @@
 
 #define __syscall_return(type, res)					\
 do {									\
-	if ((unsigned long)(res) >= (unsigned long)(-129)) {		\
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) {	\
 		errno = -(res);						\
 		res = -1;						\
 	}								\
diff --git a/include/asm-arm26/unistd.h b/include/asm-arm26/unistd.h
index 70eb6d91cfd0..c6d2436c9d34 100644
--- a/include/asm-arm26/unistd.h
+++ b/include/asm-arm26/unistd.h
@@ -311,6 +311,7 @@
 #define __ARM_NR_usr26			(__ARM_NR_BASE+3)
 
 #ifdef __KERNEL__
+#include <linux/err.h>
 #include <linux/linkage.h>
 
 #define __sys2(x) #x
@@ -322,7 +323,7 @@
 
 #define __syscall_return(type, res)					\
 do {									\
-	if ((unsigned long)(res) >= (unsigned long)(-125)) {		\
+	if ((unsigned long)(res) >= (unsigned long)-MAX_ERRNO) {	\
 		errno = -(res);						\
 		res = -1;						\
 	}								\
diff --git a/include/asm-frv/unistd.h b/include/asm-frv/unistd.h
index b80dbd839475..d104d1b91d39 100644
--- a/include/asm-frv/unistd.h
+++ b/include/asm-frv/unistd.h
@@ -320,6 +320,7 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 310
+#include <linux/err.h>
 
 /*
  * process the return value of a syscall, consigning it to one of two possible fates
@@ -329,7 +330,7 @@
 #define __syscall_return(type, res)					\
 do {									\
         unsigned long __sr2 = (res);					\
-	if (__builtin_expect(__sr2 >= (unsigned long)(-4095), 0)) {	\
+	if (__builtin_expect(__sr2 >= (unsigned long)(-MAX_ERRNO), 0)) { \
 		errno = (-__sr2);					\
 		__sr2 = ~0UL;						\
 	}								\
diff --git a/include/asm-h8300/unistd.h b/include/asm-h8300/unistd.h
index 226dd596c2da..a2dd90462d80 100644
--- a/include/asm-h8300/unistd.h
+++ b/include/asm-h8300/unistd.h
@@ -295,14 +295,14 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 289
+#include <linux/err.h>
 
-
-/* user-visible error numbers are in the range -1 - -122: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
    <asm-m68k/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* avoid using res which is declared to be in register d0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index 565d0897b205..bd9987087adc 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -328,14 +328,15 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 319
+#include <linux/err.h>
 
 /*
- * user-visible error numbers are in the range -1 - -128: see
+ * user-visible error numbers are in the range -1 - -MAX_ERRNO: see
  * <asm-i386/errno.h>
  */
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-(128 + 1))) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 		errno = -(res); \
 		res = -1; \
 	} \
diff --git a/include/asm-m32r/unistd.h b/include/asm-m32r/unistd.h
index 89f376e6229f..5c6a9ac6cf1a 100644
--- a/include/asm-m32r/unistd.h
+++ b/include/asm-m32r/unistd.h
@@ -296,8 +296,9 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 285
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -124: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
  * <asm-m32r/errno.h>
  */
 
@@ -305,7 +306,7 @@
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-(124 + 1))) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* Avoid using "res" which is declared to be in register r0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-m68k/unistd.h b/include/asm-m68k/unistd.h
index 7c0b6296b45c..751632b904db 100644
--- a/include/asm-m68k/unistd.h
+++ b/include/asm-m68k/unistd.h
@@ -288,13 +288,14 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls		282
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -124: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
    <asm-m68k/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* avoid using res which is declared to be in register d0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-m68knommu/unistd.h b/include/asm-m68knommu/unistd.h
index 1b2abdf281e1..21fdc37c5c2c 100644
--- a/include/asm-m68knommu/unistd.h
+++ b/include/asm-m68knommu/unistd.h
@@ -289,13 +289,14 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls		282
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -122: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
    <asm-m68k/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* avoid using res which is declared to be in register d0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index 02b942d85c37..d49c54cb5505 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -342,9 +342,11 @@
 
 #ifdef __KERNEL__
 
+#include <linux/err.h>
+
 #define __syscall_return(type, res)			     \
 do {							     \
-	if ((unsigned long)(res) >= (unsigned long)(-4095)) {\
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 		errno = -(res);				     \
 		res = -1;				     \
 	}						     \
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index 76b5430cb458..da127d7901af 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -306,11 +306,14 @@
 
 #ifdef __KERNEL__
 
-/* user-visible error numbers are in the range -1 - -124: see <asm-sh/errno.h> */
+#include <linux/err.h>
+
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO:
+ * see <asm-sh/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-124)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* Avoid using "res" which is declared to be in register r0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-sh64/unistd.h b/include/asm-sh64/unistd.h
index 9a1590fffc15..c113566bef33 100644
--- a/include/asm-sh64/unistd.h
+++ b/include/asm-sh64/unistd.h
@@ -347,8 +347,10 @@
 #ifdef __KERNEL__ 
 
 #define NR_syscalls 321
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -125: see <asm-sh64/errno.h> */
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO:
+ * see <asm-sh64/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
@@ -358,7 +360,7 @@ do { \
 	**       life easier in the system call epilogue (see entry.S)      \
 	*/								    \
         register unsigned long __sr2 __asm__ ("r2") = res;		    \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) {	    \
 		errno = -(res);						    \
 		__sr2 = -1; 						    \
 	} \
diff --git a/include/asm-v850/unistd.h b/include/asm-v850/unistd.h
index bcb44bfe577a..552b7c873a57 100644
--- a/include/asm-v850/unistd.h
+++ b/include/asm-v850/unistd.h
@@ -238,12 +238,13 @@
 #ifdef __KERNEL__
 
 #include <asm/clinkage.h>
+#include <linux/err.h>
 
 #define __syscall_return(type, res)					      \
   do {									      \
-	  /* user-visible error numbers are in the range -1 - -124:	      \
+	  /* user-visible error numbers are in the range -1 - -MAX_ERRNO:      \
 	     see <asm-v850/errno.h> */					      \
-	  if (__builtin_expect ((unsigned long)(res) >= (unsigned long)(-125), 0)) { \
+	  if (__builtin_expect ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO), 0)) { \
 		  errno = -(res);					      \
 		  res = -1;						      \
 	  }								      \
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index eeb98c168e98..6137146516d3 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -623,16 +623,17 @@ __SYSCALL(__NR_move_pages, sys_move_pages)
 #ifdef __KERNEL__
 
 #define __NR_syscall_max __NR_move_pages
+#include <linux/err.h>
 
 #ifndef __NO_STUBS
 
-/* user-visible error numbers are in the range -1 - -4095 */
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO */
 
 #define __syscall_clobber "r11","rcx","memory" 
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-127)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 		errno = -(res); \
 		res = -1; \
 	} \
diff --git a/include/linux/err.h b/include/linux/err.h
index cd3b367f7445..1ab1d44f8d3b 100644
--- a/include/linux/err.h
+++ b/include/linux/err.h
@@ -15,6 +15,8 @@
  */
 #define MAX_ERRNO	4095
 
+#ifndef __ASSEMBLY__
+
 #define IS_ERR_VALUE(x) unlikely((x) >= (unsigned long)-MAX_ERRNO)
 
 static inline void *ERR_PTR(long error)
@@ -32,4 +34,6 @@ static inline long IS_ERR(const void *ptr)
 	return IS_ERR_VALUE((unsigned long)ptr);
 }
 
+#endif
+
 #endif /* _LINUX_ERR_H */
-- 
cgit v1.2.3


From 07563c711fbc25389e58ab9c9f0b9de2fce56760 Mon Sep 17 00:00:00 2001
From: Michael Tokarev <mjt@tls.mks.ru>
Date: Wed, 27 Sep 2006 01:50:56 -0700
Subject: [PATCH] EISA bus MODALIAS attributes support

Add modalias attribute support for the almost forgotten now EISA bus and
(at least some) EISA-aware modules.

The modalias entry looks like (for an 3c509 NIC):

 eisa:sTCM5093

and the in-module alias like:

 eisa:sTCM5093*

The patch moves struct eisa_device_id declaration from include/linux/eisa.h
to include/linux/mod_devicetable.h (so that the former now #includes the
latter), adds proper MODULE_DEVICE_TABLE(eisa, ...) statements for all
drivers with EISA IDs I found (some drivers already have that DEVICE_TABLE
declared), and adds recognision of __mod_eisa_device_table to
scripts/mod/file2alias.c so that proper modules.alias will be generated.

There's no support for /lib/modules/$kver/modules.eisamap, as it's not used
by any existing tools, and because with in-kernel modalias mechanism those
maps are obsolete anyway.

The rationale for this patch is:

 a) to make EISA bus to act as other busses with modalias
    support, to unify driver loading

 b) to foget about EISA finally - with this patch, kernel
    (who still supports EISA) will be the only one who knows
    how to choose the necessary drivers for this bus ;)

[akpm@osdl.org: fix the kbuild bit]
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Acked-the-net-bits-by: Jeff Garzik <jeff@garzik.org>
Acked-the-tulip-bit-by: Valerie Henson <val_henson@linux.intel.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/eisa/eisa-bus.c            | 23 +++++++++++++++++++++++
 drivers/net/3c509.c                |  1 +
 drivers/net/3c59x.c                |  1 +
 drivers/net/ne3210.c               |  1 +
 drivers/net/tulip/de4x5.c          |  1 +
 drivers/scsi/aha1740.c             |  1 +
 drivers/scsi/aic7xxx/aic7770_osm.c |  3 ++-
 drivers/scsi/sim710.c              |  1 +
 include/linux/eisa.h               |  8 +-------
 include/linux/mod_devicetable.h    | 12 ++++++++++++
 scripts/mod/file2alias.c           | 12 ++++++++++++
 11 files changed, 56 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c
index 6078e2f58817..3a365e159d89 100644
--- a/drivers/eisa/eisa-bus.c
+++ b/drivers/eisa/eisa-bus.c
@@ -128,9 +128,23 @@ static int eisa_bus_match (struct device *dev, struct device_driver *drv)
 	return 0;
 }
 
+static int eisa_bus_uevent(struct device *dev, char **envp, int num_envp,
+			   char *buffer, int buffer_size)
+{
+	struct eisa_device *edev = to_eisa_device(dev);
+	int i = 0;
+	int length = 0;
+
+	add_uevent_var(envp, num_envp, &i, buffer, buffer_size, &length,
+		       "MODALIAS=" EISA_DEVICE_MODALIAS_FMT, edev->id.sig);
+	envp[i] = NULL;
+	return 0;
+}
+
 struct bus_type eisa_bus_type = {
 	.name  = "eisa",
 	.match = eisa_bus_match,
+	.uevent = eisa_bus_uevent,
 };
 
 int eisa_driver_register (struct eisa_driver *edrv)
@@ -160,6 +174,14 @@ static ssize_t eisa_show_state (struct device *dev, struct device_attribute *att
 
 static DEVICE_ATTR(enabled, S_IRUGO, eisa_show_state, NULL);
 
+static ssize_t eisa_show_modalias (struct device *dev, struct device_attribute *attr, char *buf)
+{
+        struct eisa_device *edev = to_eisa_device (dev);
+        return sprintf (buf, EISA_DEVICE_MODALIAS_FMT "\n", edev->id.sig);
+}
+
+static DEVICE_ATTR(modalias, S_IRUGO, eisa_show_modalias, NULL);
+
 static int __init eisa_init_device (struct eisa_root_device *root,
 				    struct eisa_device *edev,
 				    int slot)
@@ -209,6 +231,7 @@ static int __init eisa_register_device (struct eisa_device *edev)
 
 	device_create_file (&edev->dev, &dev_attr_signature);
 	device_create_file (&edev->dev, &dev_attr_enabled);
+	device_create_file (&edev->dev, &dev_attr_modalias);
 
 	return 0;
 }
diff --git a/drivers/net/3c509.c b/drivers/net/3c509.c
index 59c33925be62..b936373ab2a5 100644
--- a/drivers/net/3c509.c
+++ b/drivers/net/3c509.c
@@ -225,6 +225,7 @@ static struct eisa_device_id el3_eisa_ids[] = {
 		{ "TCM5095" },
 		{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, el3_eisa_ids);
 
 static int el3_eisa_probe (struct device *device);
 
diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c
index af301f09d674..df42e28cc80f 100644
--- a/drivers/net/3c59x.c
+++ b/drivers/net/3c59x.c
@@ -851,6 +851,7 @@ static struct eisa_device_id vortex_eisa_ids[] = {
 	{ "TCM5970", CH_3C597 },
 	{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, vortex_eisa_ids);
 
 static int vortex_eisa_probe(struct device *device);
 static int vortex_eisa_remove(struct device *device);
diff --git a/drivers/net/ne3210.c b/drivers/net/ne3210.c
index 0fa8e4d22769..d66328975425 100644
--- a/drivers/net/ne3210.c
+++ b/drivers/net/ne3210.c
@@ -343,6 +343,7 @@ static struct eisa_device_id ne3210_ids[] = {
 	{ "NVL1801" },
 	{ "" },
 };
+MODULE_DEVICE_TABLE(eisa, ne3210_ids);
 
 static struct eisa_driver ne3210_eisa_driver = {
 	.id_table = ne3210_ids,
diff --git a/drivers/net/tulip/de4x5.c b/drivers/net/tulip/de4x5.c
index e661d0a9cc64..fb5fa7d68888 100644
--- a/drivers/net/tulip/de4x5.c
+++ b/drivers/net/tulip/de4x5.c
@@ -2114,6 +2114,7 @@ static struct eisa_device_id de4x5_eisa_ids[] = {
         { "DEC4250", 0 },	/* 0 is the board name index... */
         { "" }
 };
+MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
 
 static struct eisa_driver de4x5_eisa_driver = {
         .id_table = de4x5_eisa_ids,
diff --git a/drivers/scsi/aha1740.c b/drivers/scsi/aha1740.c
index 0e4a7ebe300a..6b35ed8301e0 100644
--- a/drivers/scsi/aha1740.c
+++ b/drivers/scsi/aha1740.c
@@ -681,6 +681,7 @@ static struct eisa_device_id aha1740_ids[] = {
 	{ "ADP0400" },		/* 1744  */
 	{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, aha1740_ids);
 
 static struct eisa_driver aha1740_driver = {
 	.id_table = aha1740_ids,
diff --git a/drivers/scsi/aic7xxx/aic7770_osm.c b/drivers/scsi/aic7xxx/aic7770_osm.c
index 867cbe23579b..1ac119733bac 100644
--- a/drivers/scsi/aic7xxx/aic7770_osm.c
+++ b/drivers/scsi/aic7xxx/aic7770_osm.c
@@ -132,7 +132,8 @@ static struct eisa_device_id aic7770_ids[] = {
 	{ "ADP7770", 5 }, /* AIC7770 generic */
 	{ "" }
 };
-  
+MODULE_DEVICE_TABLE(eisa, aic7770_ids);
+
 static struct eisa_driver aic7770_driver = {
 	.id_table	= aic7770_ids,
 	.driver = {
diff --git a/drivers/scsi/sim710.c b/drivers/scsi/sim710.c
index b27e85428daa..551baccec523 100644
--- a/drivers/scsi/sim710.c
+++ b/drivers/scsi/sim710.c
@@ -282,6 +282,7 @@ static struct eisa_device_id sim710_eisa_ids[] = {
 	{ "HWP0C80" },
 	{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, sim710_eisa_ids);
 
 static __init int
 sim710_eisa_probe(struct device *dev)
diff --git a/include/linux/eisa.h b/include/linux/eisa.h
index 4079242dced8..1ff7c1392525 100644
--- a/include/linux/eisa.h
+++ b/include/linux/eisa.h
@@ -3,8 +3,8 @@
 
 #include <linux/ioport.h>
 #include <linux/device.h>
+#include <linux/mod_devicetable.h>
 
-#define EISA_SIG_LEN   8
 #define EISA_MAX_SLOTS 8
 
 #define EISA_MAX_RESOURCES 4
@@ -27,12 +27,6 @@
 #define EISA_CONFIG_ENABLED         1
 #define EISA_CONFIG_FORCED          2
 
-/* The EISA signature, in ASCII form, null terminated */
-struct eisa_device_id {
-	char          sig[EISA_SIG_LEN];
-	unsigned long driver_data;
-};
-
 /* There is not much we can say about an EISA device, apart from
  * signature, slot number, and base address. dma_mask is set by
  * default to parent device mask..*/
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index f7ca0b09075d..e0c393cc7240 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -308,4 +308,16 @@ struct input_device_id {
 	kernel_ulong_t driver_info;
 };
 
+/* EISA */
+
+#define EISA_SIG_LEN   8
+
+/* The EISA signature, in ASCII form, null terminated */
+struct eisa_device_id {
+	char          sig[EISA_SIG_LEN];
+	kernel_ulong_t driver_data;
+};
+
+#define EISA_DEVICE_MODALIAS_FMT "eisa:s%s"
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index de76da80443f..f61c9ccef6aa 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -444,6 +444,14 @@ static int do_input_entry(const char *filename, struct input_device_id *id,
 	return 1;
 }
 
+static int do_eisa_entry(const char *filename, struct eisa_device_id *eisa,
+		char *alias)
+{
+	if (eisa->sig[0])
+		sprintf(alias, EISA_DEVICE_MODALIAS_FMT "*", eisa->sig);
+	return 1;
+}
+
 /* Ignore any prefix, eg. v850 prepends _ */
 static inline int sym_is(const char *symbol, const char *name)
 {
@@ -547,6 +555,10 @@ void handle_moddevtable(struct module *mod, struct elf_info *info,
 		do_table(symval, sym->st_size,
 			 sizeof(struct input_device_id), "input",
 			 do_input_entry, mod);
+	else if (sym_is(symname, "__mod_eisa_device_table"))
+		do_table(symval, sym->st_size,
+			 sizeof(struct eisa_device_id), "eisa",
+			 do_eisa_entry, mod);
 }
 
 /* Now add out buffered information to the generated C source */
-- 
cgit v1.2.3


From 7583ddfd3aae1007bc4fc67ea4c07d573d376e9e Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <marcelo@kvack.org>
Date: Wed, 27 Sep 2006 01:51:02 -0700
Subject: [PATCH] Include __param section in read-only data range

The param section is an array of "kernel_param" structures, storing only
constant data: pointer to name, permission of the variable pointed to by
(void *)arg and pointers to set/get methods.

Move end_rodata down to include __param section in the read-only range used
by CONFIG_DEBUG_RODATA.

Signed-off-by: Marcelo Tosatti <marcelo@kvack.org>
Acked-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/vmlinux.lds.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 253ae1328271..69240b52f8e1 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -118,15 +118,15 @@
         __ksymtab_strings : AT(ADDR(__ksymtab_strings) - LOAD_OFFSET) {	\
 		*(__ksymtab_strings)					\
 	}								\
-	__end_rodata = .;						\
-	. = ALIGN(4096);						\
 									\
 	/* Built-in module parameters. */				\
 	__param : AT(ADDR(__param) - LOAD_OFFSET) {			\
 		VMLINUX_SYMBOL(__start___param) = .;			\
 		*(__param)						\
 		VMLINUX_SYMBOL(__stop___param) = .;			\
-	}
+	}								\
+	__end_rodata = .;						\
+	. = ALIGN(4096);
 
 #define SECURITY_INIT							\
 	.security_initcall.init : AT(ADDR(.security_initcall.init) - LOAD_OFFSET) { \
-- 
cgit v1.2.3


From c18258c6f0848f97e85287f6271c511a092bb784 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 27 Sep 2006 01:51:06 -0700
Subject: [PATCH] pid: Implement transfer_pid and use it to simplify de_thread

In de_thread we move pids from one process to another, a rather ugly case.
The function transfer_pid makes it clear what we are doing, and makes the
action atomic.  This is useful we ever want to atomically traverse the
process group and session lists, in a rcu safe manner.

Even if the atomic properties this change should be a win as transfer_pid
should be less code to execute than executing both attach_pid and
detach_pid, and this should make de_thread slightly smaller as only a
single function call needs to be emitted.  The only downside is that the
code might be slower to execute as the odds are against transfer_pid being
in cache.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/exec.c           | 11 ++++-------
 include/linux/pid.h |  2 ++
 kernel/pid.c        |  9 +++++++++
 3 files changed, 15 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/fs/exec.c b/fs/exec.c
index 54135df2a966..b7aa3d6422d6 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -696,23 +696,20 @@ static int de_thread(struct task_struct *tsk)
 		 */
 
 		/* Become a process group leader with the old leader's pid.
-		 * Note: The old leader also uses thispid until release_task
+		 * The old leader becomes a thread of the this thread group.
+		 * Note: The old leader also uses this pid until release_task
 		 *       is called.  Odd but simple and correct.
 		 */
 		detach_pid(current, PIDTYPE_PID);
 		current->pid = leader->pid;
 		attach_pid(current, PIDTYPE_PID,  current->pid);
-		attach_pid(current, PIDTYPE_PGID, current->signal->pgrp);
-		attach_pid(current, PIDTYPE_SID,  current->signal->session);
+		transfer_pid(leader, current, PIDTYPE_PGID);
+		transfer_pid(leader, current, PIDTYPE_SID);
 		list_replace_rcu(&leader->tasks, &current->tasks);
 
 		current->group_leader = current;
 		leader->group_leader = current;
 
-		/* Reduce leader to a thread */
-		detach_pid(leader, PIDTYPE_PGID);
-		detach_pid(leader, PIDTYPE_SID);
-
 		current->exit_signal = SIGCHLD;
 
 		BUG_ON(leader->exit_state != EXIT_ZOMBIE);
diff --git a/include/linux/pid.h b/include/linux/pid.h
index 29960b03bef7..93da7e2d9f30 100644
--- a/include/linux/pid.h
+++ b/include/linux/pid.h
@@ -76,6 +76,8 @@ extern int FASTCALL(attach_pid(struct task_struct *task,
 				enum pid_type type, int nr));
 
 extern void FASTCALL(detach_pid(struct task_struct *task, enum pid_type));
+extern void FASTCALL(transfer_pid(struct task_struct *old,
+				  struct task_struct *new, enum pid_type));
 
 /*
  * look up a PID in the hash table. Must be called with the tasklist_lock
diff --git a/kernel/pid.c b/kernel/pid.c
index 93e212f20671..6db82b68e2f8 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -252,6 +252,15 @@ void fastcall detach_pid(struct task_struct *task, enum pid_type type)
 	free_pid(pid);
 }
 
+/* transfer_pid is an optimization of attach_pid(new), detach_pid(old) */
+void fastcall transfer_pid(struct task_struct *old, struct task_struct *new,
+			   enum pid_type type)
+{
+	new->pids[type].pid = old->pids[type].pid;
+	hlist_replace_rcu(&old->pids[type].node, &new->pids[type].node);
+	old->pids[type].pid = NULL;
+}
+
 struct task_struct * fastcall pid_task(struct pid *pid, enum pid_type type)
 {
 	struct task_struct *result = NULL;
-- 
cgit v1.2.3


From 66f37509fc7191df468a8d183374f48b13bacb73 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Wed, 27 Sep 2006 01:51:13 -0700
Subject: [PATCH] fs/nfs/: make code static

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfs/namespace.c     | 12 +++++++++---
 include/linux/nfs_fs.h |  4 ----
 2 files changed, 9 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 77b00684894d..60408646176b 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -26,6 +26,11 @@ LIST_HEAD(nfs_automount_list);
 static DECLARE_WORK(nfs_automount_task, nfs_expire_automounts, &nfs_automount_list);
 int nfs_mountpoint_expiry_timeout = 500 * HZ;
 
+static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
+					const struct dentry *dentry,
+					struct nfs_fh *fh,
+					struct nfs_fattr *fattr);
+
 /*
  * nfs_path - reconstruct the path given an arbitrary dentry
  * @base - arbitrary string to prepend to the path
@@ -209,9 +214,10 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
  * @fattr - attributes for new root inode
  *
  */
-struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
-		const struct dentry *dentry, struct nfs_fh *fh,
-		struct nfs_fattr *fattr)
+static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
+					const struct dentry *dentry,
+					struct nfs_fh *fh,
+					struct nfs_fattr *fattr)
 {
 	struct nfs_clone_mount mountdata = {
 		.sb = mnt_parent->mnt_sb,
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 36f5bcf513b0..98c9b9f667a5 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -315,10 +315,6 @@ extern void nfs_end_data_update(struct inode *);
 extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx);
 extern void put_nfs_open_context(struct nfs_open_context *ctx);
 extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, int mode);
-extern struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
-					const struct dentry *dentry,
-					struct nfs_fh *fh,
-					struct nfs_fattr *fattr);
 
 /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */
 extern u32 root_nfs_parse_addr(char *name); /*__init*/
-- 
cgit v1.2.3


From 1b79e5513d52e8533a08af35a3595dad80c74d1f Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:51:14 -0700
Subject: [PATCH] add probe_kernel_address()

Add a version of __get_user() which is safe to call inside mmap_sem.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/uaccess.h | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

(limited to 'include')

diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 391e7ed1eb3f..a48d7f11c7be 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -19,4 +19,26 @@ static inline unsigned long __copy_from_user_nocache(void *to,
 
 #endif		/* ARCH_HAS_NOCACHE_UACCESS */
 
+/**
+ * probe_kernel_address(): safely attempt to read from a location
+ * @addr: address to read from - its type is type typeof(retval)*
+ * @retval: read into this variable
+ *
+ * Safely read from address @addr into variable @revtal.  If a kernel fault
+ * happens, handle that and return -EFAULT.
+ * We ensure that the __get_user() is executed in atomic context so that
+ * do_page_fault() doesn't attempt to take mmap_sem.  This makes
+ * probe_kernel_address() suitable for use within regions where the caller
+ * already holds mmap_sem, or other locks which nest inside mmap_sem.
+ */
+#define probe_kernel_address(addr, retval)		\
+	({						\
+		long ret;				\
+							\
+		inc_preempt_count();			\
+		ret = __get_user(retval, addr);		\
+		dec_preempt_count();			\
+		ret;					\
+	})
+
 #endif		/* __LINUX_UACCESS_H__ */
-- 
cgit v1.2.3


From f12d0d7c7786af39435ef6ae9defe47fb58f6091 Mon Sep 17 00:00:00 2001
From: "Hyok S. Choi" <hyok.choi@samsung.com>
Date: Tue, 26 Sep 2006 17:36:37 +0900
Subject: [ARM] nommu: manage the CP15 things

All the current CP15 access codes in ARM arch can be categorized and
conditioned by the defines as follows:

     Related operation	Safe condition
  a. any CP15 access	!CPU_CP15
  b. alignment trap	CPU_CP15_MMU
  c. D-cache(C-bit)	CPU_CP15
  d. I-cache		CPU_CP15 && !( CPU_ARM610 || CPU_ARM710 ||
				CPU_ARM720 || CPU_ARM740 ||
				CPU_XSCALE || CPU_XSC3 )
  e. alternate vector	CPU_CP15 && !CPU_ARM740
  f. TTB		CPU_CP15_MMU
  g. Domain		CPU_CP15_MMU
  h. FSR/FAR		CPU_CP15_MMU

For example, alternate vector is supported if and only if
"CPU_CP15 && !CPU_ARM740" is satisfied.

Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                  |  3 ++-
 arch/arm/Kconfig-nommu            |  8 ++++++++
 arch/arm/boot/compressed/Makefile |  4 ++++
 arch/arm/boot/compressed/head.S   |  6 ++++++
 arch/arm/kernel/head-nommu.S      |  7 ++++++-
 arch/arm/kernel/process.c         | 22 ++++++++++++++++------
 arch/arm/mm/Kconfig               |  8 ++++----
 arch/arm/mm/cache-v4.S            | 10 ++++++++++
 include/asm-arm/system.h          |  4 ++++
 9 files changed, 60 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 0810d27c039e..2673fee3281c 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -621,6 +621,7 @@ config LEDS_CPU
 
 config ALIGNMENT_TRAP
 	bool
+	depends on CPU_CP15_MMU
 	default y if !ARCH_EBSA110
 	help
 	  ARM processors can not fetch/store information which is not
@@ -852,7 +853,7 @@ source "drivers/base/Kconfig"
 
 source "drivers/connector/Kconfig"
 
-if ALIGNMENT_TRAP
+if ALIGNMENT_TRAP || !CPU_CP15_MMU
 source "drivers/mtd/Kconfig"
 endif
 
diff --git a/arch/arm/Kconfig-nommu b/arch/arm/Kconfig-nommu
index e1574be2ded6..f087376748d1 100644
--- a/arch/arm/Kconfig-nommu
+++ b/arch/arm/Kconfig-nommu
@@ -25,6 +25,14 @@ config FLASH_SIZE
 	hex 'FLASH Size' if SET_MEM_PARAM
 	default 0x00400000
 
+config PROCESSOR_ID
+	hex
+	default 0x00007700
+	depends on !CPU_CP15
+	help
+	  If processor has no CP15 register, this processor ID is
+	  used instead of the auto-probing which utilizes the register.
+
 config REMAP_VECTORS_TO_RAM
 	bool 'Install vectors to the begining of RAM' if DRAM_BASE
 	depends on DRAM_BASE
diff --git a/arch/arm/boot/compressed/Makefile b/arch/arm/boot/compressed/Makefile
index 2adc1527e0eb..adddc7131685 100644
--- a/arch/arm/boot/compressed/Makefile
+++ b/arch/arm/boot/compressed/Makefile
@@ -51,7 +51,11 @@ OBJS		+= head-at91rm9200.o
 endif
 
 ifeq ($(CONFIG_CPU_BIG_ENDIAN),y)
+ifeq ($(CONFIG_CPU_CP15),y)
 OBJS		+= big-endian.o
+else
+# The endian should be set by h/w design.
+endif
 endif
 
 #
diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
index 75df1f764a10..e5ab51b9cceb 100644
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -82,9 +82,11 @@
 		kphex	r6, 8		/* processor id */
 		kputc	#':'
 		kphex	r7, 8		/* architecture id */
+#ifdef CONFIG_CPU_CP15
 		kputc	#':'
 		mrc	p15, 0, r0, c1, c0
 		kphex	r0, 8		/* control reg */
+#endif
 		kputc	#'\n'
 		kphex	r5, 8		/* decompressed kernel start */
 		kputc	#'-'
@@ -507,7 +509,11 @@ call_kernel:	bl	cache_clean_flush
  */
 
 call_cache_fn:	adr	r12, proc_types
+#ifdef CONFIG_CPU_CP15
 		mrc	p15, 0, r6, c0, c0	@ get processor ID
+#else
+		ldr	r6, =CONFIG_PROCESSOR_ID
+#endif
 1:		ldr	r1, [r12, #0]		@ get value
 		ldr	r2, [r12, #4]		@ get mask
 		eor	r1, r1, r6		@ (real ^ match)
diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S
index ac9eb3d30518..698a537915dd 100644
--- a/arch/arm/kernel/head-nommu.S
+++ b/arch/arm/kernel/head-nommu.S
@@ -9,7 +9,6 @@
  * published by the Free Software Foundation.
  *
  *  Common kernel startup code (non-paged MM)
- *    for 32-bit CPUs which has a process ID register(CP15).
  *
  */
 #include <linux/linkage.h>
@@ -40,7 +39,11 @@
 ENTRY(stext)
 	msr	cpsr_c, #PSR_F_BIT | PSR_I_BIT | SVC_MODE @ ensure svc mode
 						@ and irqs disabled
+#ifndef CONFIG_CPU_CP15
+	ldr	r9, =CONFIG_PROCESSOR_ID
+#else
 	mrc	p15, 0, r9, c0, c0		@ get processor id
+#endif
 	bl	__lookup_processor_type		@ r5=procinfo r9=cpuid
 	movs	r10, r5				@ invalid processor (r5=0)?
 	beq	__error_p				@ yes, error 'p'
@@ -58,6 +61,7 @@ ENTRY(stext)
  */
 	.type	__after_proc_init, %function
 __after_proc_init:
+#ifdef CONFIG_CPU_CP15
 	mrc	p15, 0, r0, c1, c0, 0		@ read control reg
 #ifdef CONFIG_ALIGNMENT_TRAP
 	orr	r0, r0, #CR_A
@@ -74,6 +78,7 @@ __after_proc_init:
 	bic	r0, r0, #CR_I
 #endif
 	mcr	p15, 0, r0, c1, c0, 0		@ write control reg
+#endif /* CONFIG_CPU_CP15 */
 
 	mov	pc, r13				@ clear the BSS and jump
 						@ to start_kernel
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 3079535afccd..bf35c178a877 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -221,16 +221,26 @@ void __show_regs(struct pt_regs *regs)
 		processor_modes[processor_mode(regs)],
 		thumb_mode(regs) ? " (T)" : "",
 		get_fs() == get_ds() ? "kernel" : "user");
+#if CONFIG_CPU_CP15
 	{
-		unsigned int ctrl, transbase, dac;
+		unsigned int ctrl;
 		  __asm__ (
 		"	mrc p15, 0, %0, c1, c0\n"
-		"	mrc p15, 0, %1, c2, c0\n"
-		"	mrc p15, 0, %2, c3, c0\n"
-		: "=r" (ctrl), "=r" (transbase), "=r" (dac));
-		printk("Control: %04X  Table: %08X  DAC: %08X\n",
-		  	ctrl, transbase, dac);
+		: "=r" (ctrl));
+		printk("Control: %04X\n", ctrl);
 	}
+#ifdef CONFIG_CPU_CP15_MMU
+	{
+		unsigned int transbase, dac;
+		  __asm__ (
+		"	mrc p15, 0, %0, c2, c0\n"
+		"	mrc p15, 0, %1, c3, c0\n"
+		: "=r" (transbase), "=r" (dac));
+		printk("Table: %08X  DAC: %08X\n",
+		  	transbase, dac);
+	}
+#endif
+#endif
 }
 
 void show_regs(struct pt_regs * regs)
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index c7fb835c148c..0ac11ea84508 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -445,15 +445,15 @@ config CPU_BIG_ENDIAN
 	  of your chipset/board/processor.
 
 config CPU_ICACHE_DISABLE
-	bool "Disable I-Cache"
-	depends on CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_V6
+	bool "Disable I-Cache (I-bit)"
+	depends on CPU_CP15 && !(CPU_ARM610 || CPU_ARM710 || CPU_ARM720T || CPU_ARM740T || CPU_XSCALE || CPU_XSC3)
 	help
 	  Say Y here to disable the processor instruction cache. Unless
 	  you have a reason not to or are unsure, say N.
 
 config CPU_DCACHE_DISABLE
-	bool "Disable D-Cache"
-	depends on CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_V6
+	bool "Disable D-Cache (C-bit)"
+	depends on CPU_CP15
 	help
 	  Say Y here to disable the processor data cache. Unless
 	  you have a reason not to or are unsure, say N.
diff --git a/arch/arm/mm/cache-v4.S b/arch/arm/mm/cache-v4.S
index b8ad5d58ebe2..b2908063ed6a 100644
--- a/arch/arm/mm/cache-v4.S
+++ b/arch/arm/mm/cache-v4.S
@@ -29,9 +29,13 @@ ENTRY(v4_flush_user_cache_all)
  *	Clean and invalidate the entire cache.
  */
 ENTRY(v4_flush_kern_cache_all)
+#ifdef CPU_CP15
 	mov	r0, #0
 	mcr	p15, 0, r0, c7, c7, 0		@ flush ID cache
 	mov	pc, lr
+#else
+	/* FALLTHROUGH */
+#endif
 
 /*
  *	flush_user_cache_range(start, end, flags)
@@ -44,9 +48,13 @@ ENTRY(v4_flush_kern_cache_all)
  *	- flags	- vma_area_struct flags describing address space
  */
 ENTRY(v4_flush_user_cache_range)
+#ifdef CPU_CP15
 	mov	ip, #0
 	mcreq	p15, 0, ip, c7, c7, 0		@ flush ID cache
 	mov	pc, lr
+#else
+	/* FALLTHROUGH */
+#endif
 
 /*
  *	coherent_kern_range(start, end)
@@ -108,8 +116,10 @@ ENTRY(v4_dma_inv_range)
  *	- end	 - virtual end address
  */
 ENTRY(v4_dma_flush_range)
+#ifdef CPU_CP15
 	mov	r0, #0
 	mcr	p15, 0, r0, c7, c7, 0		@ flush ID cache
+#endif
 	/* FALLTHROUGH */
 
 /*
diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h
index c19c5b009f71..f05fbe31576c 100644
--- a/include/asm-arm/system.h
+++ b/include/asm-arm/system.h
@@ -46,6 +46,7 @@
 #define CPUID_TCM	2
 #define CPUID_TLBTYPE	3
 
+#ifdef CONFIG_CPU_CP15
 #define read_cpuid(reg)							\
 	({								\
 		unsigned int __val;					\
@@ -55,6 +56,9 @@
 		    : "cc");						\
 		__val;							\
 	})
+#else
+#define read_cpuid(reg) (processor_id)
+#endif
 
 /*
  * This is used to ensure the compiler did actually allocate the register we
-- 
cgit v1.2.3


From 07e0da78abdc679714a12e7a60137d950c346681 Mon Sep 17 00:00:00 2001
From: "Hyok S. Choi" <hyok.choi@samsung.com>
Date: Tue, 26 Sep 2006 17:37:36 +0900
Subject: [ARM] nommu: add ARM7TDMI core support

This patch adds ARM7TDMI core support which has no cache and no CP15
register(no memory control unit).

Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Makefile            |   1 +
 arch/arm/mm/Kconfig          |  13 +++
 arch/arm/mm/Makefile         |   1 +
 arch/arm/mm/proc-arm7tdmi.S  | 249 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/cacheflush.h |   2 +-
 include/asm-arm/proc-fns.h   |   8 ++
 6 files changed, 273 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mm/proc-arm7tdmi.S

(limited to 'include')

diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 80cee786500c..282333fe0a74 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -55,6 +55,7 @@ arch-$(CONFIG_CPU_32v3)		:=-D__LINUX_ARM_ARCH__=3 -march=armv3
 # This selects how we optimise for the processor.
 tune-$(CONFIG_CPU_ARM610)	:=-mtune=arm610
 tune-$(CONFIG_CPU_ARM710)	:=-mtune=arm710
+tune-$(CONFIG_CPU_ARM7TDMI)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM720T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM920T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM922T)	:=-mtune=arm9tdmi
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 0ac11ea84508..9f860aa9c908 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -25,6 +25,19 @@ config CPU_ARM610
 	  Say Y if you want support for the ARM610 processor.
 	  Otherwise, say N.
 
+# ARM7TDMI
+config CPU_ARM7TDMI
+	bool "Support ARM7TDMI processor"
+	select CPU_32v4T
+	select CPU_ABRT_LV4T
+	select CPU_CACHE_V4
+	help
+	  A 32-bit RISC microprocessor based on the ARM7 processor core
+	  which has no memory control unit and cache.
+
+	  Say Y if you want support for the ARM7TDMI processor.
+	  Otherwise, say N.
+
 # ARM710
 config CPU_ARM710
 	bool "Support ARM710 processor" if !ARCH_CLPS7500 && ARCH_RPC
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 2d0ab19db496..908f6d437174 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -46,6 +46,7 @@ obj-$(CONFIG_CPU_TLB_V6)	+= tlb-v6.o
 
 obj-$(CONFIG_CPU_ARM610)	+= proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM710)	+= proc-arm6_7.o
+obj-$(CONFIG_CPU_ARM7TDMI)	+= proc-arm7tdmi.o
 obj-$(CONFIG_CPU_ARM720T)	+= proc-arm720.o
 obj-$(CONFIG_CPU_ARM920T)	+= proc-arm920.o
 obj-$(CONFIG_CPU_ARM922T)	+= proc-arm922.o
diff --git a/arch/arm/mm/proc-arm7tdmi.S b/arch/arm/mm/proc-arm7tdmi.S
new file mode 100644
index 000000000000..22d7e3100ea6
--- /dev/null
+++ b/arch/arm/mm/proc-arm7tdmi.S
@@ -0,0 +1,249 @@
+/*
+ *  linux/arch/arm/mm/proc-arm7tdmi.S: utility functions for ARM7TDMI
+ *
+ *  Copyright (C) 2003-2006 Hyok S. Choi <hyok.choi@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/procinfo.h>
+#include <asm/ptrace.h>
+
+	.text
+/*
+ * cpu_arm7tdmi_proc_init()
+ * cpu_arm7tdmi_do_idle()
+ * cpu_arm7tdmi_dcache_clean_area()
+ * cpu_arm7tdmi_switch_mm()
+ *
+ * These are not required.
+ */
+ENTRY(cpu_arm7tdmi_proc_init)
+ENTRY(cpu_arm7tdmi_do_idle)
+ENTRY(cpu_arm7tdmi_dcache_clean_area)
+ENTRY(cpu_arm7tdmi_switch_mm)
+		mov	pc, lr
+
+/*
+ * cpu_arm7tdmi_proc_fin()
+ */
+ENTRY(cpu_arm7tdmi_proc_fin)
+		mov	r0, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+		msr	cpsr_c, r0
+		mov	pc, lr
+
+/*
+ * Function: cpu_arm7tdmi_reset(loc)
+ * Params  : loc(r0)	address to jump to
+ * Purpose : Sets up everything for a reset and jump to the location for soft reset.
+ */
+ENTRY(cpu_arm7tdmi_reset)
+		mov	pc, r0
+
+		__INIT
+
+		.type	__arm7tdmi_setup, #function
+__arm7tdmi_setup:
+		mov	pc, lr
+		.size	__arm7tdmi_setup, . - __arm7tdmi_setup
+
+		__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+		.type	arm7tdmi_processor_functions, #object
+ENTRY(arm7tdmi_processor_functions)
+		.word	v4t_late_abort
+		.word	cpu_arm7tdmi_proc_init
+		.word	cpu_arm7tdmi_proc_fin
+		.word	cpu_arm7tdmi_reset
+		.word	cpu_arm7tdmi_do_idle
+		.word	cpu_arm7tdmi_dcache_clean_area
+		.word	cpu_arm7tdmi_switch_mm
+		.word	0		@ cpu_*_set_pte
+		.size	arm7tdmi_processor_functions, . - arm7tdmi_processor_functions
+
+		.section ".rodata"
+
+		.type	cpu_arch_name, #object
+cpu_arch_name:
+		.asciz	"armv4t"
+		.size	cpu_arch_name, . - cpu_arch_name
+
+		.type	cpu_elf_name, #object
+cpu_elf_name:
+		.asciz	"v4"
+		.size	cpu_elf_name, . - cpu_elf_name
+
+		.type	cpu_arm7tdmi_name, #object
+cpu_arm7tdmi_name:
+		.asciz	"ARM7TDMI"
+		.size	cpu_arm7tdmi_name, . - cpu_arm7tdmi_name
+
+		.type	cpu_triscenda7_name, #object
+cpu_triscenda7_name:
+		.asciz	"Triscend-A7x"
+		.size	cpu_triscenda7_name, . - cpu_triscenda7_name
+
+		.type	cpu_at91_name, #object
+cpu_at91_name:
+		.asciz	"Atmel-AT91M40xxx"
+		.size	cpu_at91_name, . - cpu_at91_name
+
+		.type	cpu_s3c3410_name, #object
+cpu_s3c3410_name:
+		.asciz	"Samsung-S3C3410"
+		.size	cpu_s3c3410_name, . - cpu_s3c3410_name
+
+		.type	cpu_s3c44b0x_name, #object
+cpu_s3c44b0x_name:
+		.asciz	"Samsung-S3C44B0x"
+		.size	cpu_s3c44b0x_name, . - cpu_s3c44b0x_name
+
+		.type	cpu_s3c4510b, #object
+cpu_s3c4510b_name:
+		.asciz	"Samsung-S3C4510B"
+		.size	cpu_s3c4510b_name, . - cpu_s3c4510b_name
+
+		.type	cpu_s3c4530_name, #object
+cpu_s3c4530_name:
+		.asciz	"Samsung-S3C4530"
+		.size	cpu_s3c4530_name, . - cpu_s3c4530_name
+
+		.type	cpu_netarm_name, #object
+cpu_netarm_name:
+		.asciz	"NETARM"
+		.size	cpu_netarm_name, . - cpu_netarm_name
+
+		.align
+
+		.section ".proc.info.init", #alloc, #execinstr
+
+		.type	__arm7tdmi_proc_info, #object
+__arm7tdmi_proc_info:
+		.long	0x41007700
+		.long	0xfff8ff00
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_26BIT
+		.long	cpu_arm7tdmi_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__arm7tdmi_proc_info, . - __arm7dmi_proc_info
+
+		.type	__triscenda7_proc_info, #object
+__triscenda7_proc_info:
+		.long	0x0001d2ff
+		.long	0x0001ffff
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_triscenda7_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__triscenda7_proc_info, . - __triscenda7_proc_info
+
+		.type	__at91_proc_info, #object
+__at91_proc_info:
+		.long	0x14000040
+		.long	0xfff000e0
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_at91_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__at91_proc_info, . - __at91_proc_info
+
+		.type	__s3c4510b_proc_info, #object
+__s3c4510b_proc_info:
+		.long	0x36365000
+		.long	0xfffff000
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_s3c4510b_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__s3c4510b_proc_info, . - __s3c4510b_proc_info
+
+		.type	__s3c4530_proc_info, #object
+__s3c4530_proc_info:
+		.long	0x4c000000
+		.long	0xfff000e0
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_s3c4530_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__s3c4530_proc_info, . - __s3c4530_proc_info
+
+		.type	__s3c3410_proc_info, #object
+__s3c3410_proc_info:
+		.long	0x34100000
+		.long	0xffff0000
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_s3c3410_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__s3c3410_proc_info, . - __s3c3410_proc_info
+
+		.type	__s3c44b0x_proc_info, #object
+__s3c44b0x_proc_info:
+		.long	0x44b00000
+		.long	0xffff0000
+		.long	0
+		.long	0
+		b	__arm7tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_s3c44b0x_name
+		.long	arm7tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__s3c44b0x_proc_info, . - __s3c44b0x_proc_info
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index e4a2569c636c..e7bfff298e46 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -33,7 +33,7 @@
 # endif
 #endif
 
-#if defined(CONFIG_CPU_ARM720T)
+#if defined(CONFIG_CPU_ARM720T) || defined(CONFIG_CPU_ARM7TDMI)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
 # else
diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h
index 1bde92cdaebd..3e8c057e66b5 100644
--- a/include/asm-arm/proc-fns.h
+++ b/include/asm-arm/proc-fns.h
@@ -33,6 +33,14 @@
 #   define CPU_NAME cpu_arm6
 #  endif
 # endif
+# ifdef CONFIG_CPU_ARM7TDMI
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_arm7tdmi
+#  endif
+# endif
 # ifdef CONFIG_CPU_ARM710
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU
-- 
cgit v1.2.3


From b731c3118d87f26c8bf3f358ffbbc24450af50a6 Mon Sep 17 00:00:00 2001
From: "Hyok S. Choi" <hyok.choi@samsung.com>
Date: Tue, 26 Sep 2006 17:37:50 +0900
Subject: [ARM] nommu: add ARM740T core support

This patch adds ARM740T core support which has a MPU and 4KB or 8KB cache.

Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Makefile            |   1 +
 arch/arm/mm/Kconfig          |  19 ++++-
 arch/arm/mm/Makefile         |   1 +
 arch/arm/mm/proc-arm740.S    | 174 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/cacheflush.h |   3 +-
 include/asm-arm/proc-fns.h   |   8 ++
 6 files changed, 203 insertions(+), 3 deletions(-)
 create mode 100644 arch/arm/mm/proc-arm740.S

(limited to 'include')

diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 282333fe0a74..8ec4b46d2cbb 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -57,6 +57,7 @@ tune-$(CONFIG_CPU_ARM610)	:=-mtune=arm610
 tune-$(CONFIG_CPU_ARM710)	:=-mtune=arm710
 tune-$(CONFIG_CPU_ARM7TDMI)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM720T)	:=-mtune=arm7tdmi
+tune-$(CONFIG_CPU_ARM740T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM920T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM922T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM925T)	:=-mtune=arm9tdmi
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 9f860aa9c908..87f9fece9606 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -75,6 +75,21 @@ config CPU_ARM720T
 	  Say Y if you want support for the ARM720T processor.
 	  Otherwise, say N.
 
+# ARM740T
+config CPU_ARM740T
+	bool "Support ARM740T processor" if ARCH_INTEGRATOR
+	select CPU_32v4T
+	select CPU_ABRT_LV4T
+	select CPU_CACHE_V3	# although the core is v4t
+	select CPU_CP15_MPU
+	help
+	  A 32-bit RISC processor with 8KB cache or 4KB variants,
+	  write buffer and MPU(Protection Unit) built around
+	  an ARM7TDMI core.
+
+	  Say Y if you want support for the ARM740T processor.
+	  Otherwise, say N.
+
 # ARM920T
 config CPU_ARM920T
 	bool "Support ARM920T processor"
@@ -436,7 +451,7 @@ comment "Processor Features"
 
 config ARM_THUMB
 	bool "Support Thumb user binaries"
-	depends on CPU_ARM720T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6
+	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6
 	default y
 	help
 	  Say Y if you want to include kernel support for running user space
@@ -473,7 +488,7 @@ config CPU_DCACHE_DISABLE
 
 config CPU_DCACHE_WRITETHROUGH
 	bool "Force write through D-cache"
-	depends on (CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_V6) && !CPU_DCACHE_DISABLE
+	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_V6) && !CPU_DCACHE_DISABLE
 	default y if CPU_ARM925T
 	help
 	  Say Y here to use the data cache in writethrough mode. Unless you
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 908f6d437174..d10db822d3b8 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_CPU_ARM610)	+= proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM710)	+= proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM7TDMI)	+= proc-arm7tdmi.o
 obj-$(CONFIG_CPU_ARM720T)	+= proc-arm720.o
+obj-$(CONFIG_CPU_ARM740T)	+= proc-arm740.o
 obj-$(CONFIG_CPU_ARM920T)	+= proc-arm920.o
 obj-$(CONFIG_CPU_ARM922T)	+= proc-arm922.o
 obj-$(CONFIG_CPU_ARM925T)	+= proc-arm925.o
diff --git a/arch/arm/mm/proc-arm740.S b/arch/arm/mm/proc-arm740.S
new file mode 100644
index 000000000000..40713818a87b
--- /dev/null
+++ b/arch/arm/mm/proc-arm740.S
@@ -0,0 +1,174 @@
+/*
+ *  linux/arch/arm/mm/arm740.S: utility functions for ARM740
+ *
+ *  Copyright (C) 2004-2006 Hyok S. Choi (hyok.choi@samsung.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/procinfo.h>
+#include <asm/ptrace.h>
+
+	.text
+/*
+ * cpu_arm740_proc_init()
+ * cpu_arm740_do_idle()
+ * cpu_arm740_dcache_clean_area()
+ * cpu_arm740_switch_mm()
+ *
+ * These are not required.
+ */
+ENTRY(cpu_arm740_proc_init)
+ENTRY(cpu_arm740_do_idle)
+ENTRY(cpu_arm740_dcache_clean_area)
+ENTRY(cpu_arm740_switch_mm)
+	mov	pc, lr
+
+/*
+ * cpu_arm740_proc_fin()
+ */
+ENTRY(cpu_arm740_proc_fin)
+	stmfd	sp!, {lr}
+	mov	ip, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+	msr	cpsr_c, ip
+	mrc	p15, 0, r0, c1, c0, 0
+	bic	r0, r0, #0x3f000000		@ bank/f/lock/s
+	bic	r0, r0, #0x0000000c		@ w-buffer/cache
+	mcr	p15, 0, r0, c1, c0, 0		@ disable caches
+	mcr	p15, 0, r0, c7, c0, 0		@ invalidate cache
+	ldmfd	sp!, {pc}
+
+/*
+ * cpu_arm740_reset(loc)
+ * Params  : r0 = address to jump to
+ * Notes   : This sets up everything for a reset
+ */
+ENTRY(cpu_arm740_reset)
+	mov	ip, #0
+	mcr	p15, 0, ip, c7, c0, 0		@ invalidate cache
+	mrc	p15, 0, ip, c1, c0, 0		@ get ctrl register
+	bic	ip, ip, #0x0000000c		@ ............wc..
+	mcr	p15, 0, ip, c1, c0, 0		@ ctrl register
+	mov	pc, r0
+
+	__INIT
+
+	.type	__arm740_setup, #function
+__arm740_setup:
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c0, 0		@ invalidate caches
+
+	mcr	p15, 0, r0, c6, c3		@ disable area 3~7
+	mcr	p15, 0, r0, c6, c4
+	mcr	p15, 0, r0, c6, c5
+	mcr	p15, 0, r0, c6, c6
+	mcr	p15, 0, r0, c6, c7
+
+	mov	r0, #0x0000003F			@ base = 0, size = 4GB
+	mcr	p15, 0, r0, c6,	c0		@ set area 0, default
+
+	ldr	r0, =(CONFIG_DRAM_BASE & 0xFFFFF000) @ base[31:12] of RAM
+	ldr	r1, =(CONFIG_DRAM_SIZE >> 12)	@ size of RAM (must be >= 4KB)
+	mov	r2, #10				@ 11 is the minimum (4KB)
+1:	add	r2, r2, #1			@ area size *= 2
+	mov	r1, r1, lsr #1
+	bne	1b				@ count not zero r-shift
+	orr	r0, r0, r2, lsl #1		@ the area register value
+	orr	r0, r0, #1			@ set enable bit
+	mcr	p15, 0, r0, c6,	c1		@ set area 1, RAM
+
+	ldr	r0, =(CONFIG_FLASH_MEM_BASE & 0xFFFFF000) @ base[31:12] of FLASH
+	ldr	r1, =(CONFIG_FLASH_SIZE >> 12)	@ size of FLASH (must be >= 4KB)
+	mov	r2, #10				@ 11 is the minimum (4KB)
+1:	add	r2, r2, #1			@ area size *= 2
+	mov	r1, r1, lsr #1
+	bne	1b				@ count not zero r-shift
+	orr	r0, r0, r2, lsl #1		@ the area register value
+	orr	r0, r0, #1			@ set enable bit
+	mcr	p15, 0, r0, c6,	c2		@ set area 2, ROM/FLASH
+
+	mov	r0, #0x06
+	mcr	p15, 0, r0, c2, c0		@ Region 1&2 cacheable
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mov	r0, #0x00			@ disable whole write buffer
+#else
+	mov	r0, #0x02			@ Region 1 write bufferred
+#endif
+	mcr	p15, 0, r0, c3, c0
+
+	mov	r0, #0x10000
+	sub	r0, r0, #1			@ r0 = 0xffff
+	mcr	p15, 0, r0, c5, c0		@ all read/write access
+
+	mrc	p15, 0, r0, c1, c0		@ get control register
+	bic	r0, r0, #0x3F000000		@ set to standard caching mode
+						@ need some benchmark
+	orr	r0, r0, #0x0000000d		@ MPU/Cache/WB
+
+	mov	pc, lr
+
+	.size	__arm740_setup, . - __arm740_setup
+
+	__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+	.type	arm740_processor_functions, #object
+ENTRY(arm740_processor_functions)
+	.word	v4t_late_abort
+	.word	cpu_arm740_proc_init
+	.word	cpu_arm740_proc_fin
+	.word	cpu_arm740_reset
+	.word   cpu_arm740_do_idle
+	.word	cpu_arm740_dcache_clean_area
+	.word	cpu_arm740_switch_mm
+	.word	0			@ cpu_*_set_pte
+	.size	arm740_processor_functions, . - arm740_processor_functions
+
+	.section ".rodata"
+
+	.type	cpu_arch_name, #object
+cpu_arch_name:
+	.asciz	"armv4"
+	.size	cpu_arch_name, . - cpu_arch_name
+
+	.type	cpu_elf_name, #object
+cpu_elf_name:
+	.asciz	"v4"
+	.size	cpu_elf_name, . - cpu_elf_name
+
+	.type	cpu_arm740_name, #object
+cpu_arm740_name:
+	.ascii	"ARM740T"
+	.size	cpu_arm740_name, . - cpu_arm740_name
+
+	.align
+
+	.section ".proc.info.init", #alloc, #execinstr
+	.type	__arm740_proc_info,#object
+__arm740_proc_info:
+	.long	0x41807400
+	.long	0xfffffff0
+	.long	0
+	b	__arm740_setup
+	.long	cpu_arch_name
+	.long	cpu_elf_name
+	.long	HWCAP_SWP | HWCAP_HALF | HWCAP_26BIT
+	.long	cpu_arm740_name
+	.long	arm740_processor_functions
+	.long	0
+	.long	0
+	.long	v3_cache_fns			@ cache model
+	.size	__arm740_proc_info, . - __arm740_proc_info
+
+
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index e7bfff298e46..24924e64f883 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -25,7 +25,8 @@
 #undef _CACHE
 #undef MULTI_CACHE
 
-#if defined(CONFIG_CPU_ARM610) || defined(CONFIG_CPU_ARM710)
+#if defined(CONFIG_CPU_ARM610) || defined(CONFIG_CPU_ARM710) || \
+    defined(CONFIG_CPU_ARM740T)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
 # else
diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h
index 3e8c057e66b5..17dfc0de9658 100644
--- a/include/asm-arm/proc-fns.h
+++ b/include/asm-arm/proc-fns.h
@@ -57,6 +57,14 @@
 #   define CPU_NAME cpu_arm720
 #  endif
 # endif
+# ifdef CONFIG_CPU_ARM740T
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_arm740
+#  endif
+# endif
 # ifdef CONFIG_CPU_ARM920T
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU
-- 
cgit v1.2.3


From 43f5f0146ef5c3a3421ea53a0708fd37edcb8905 Mon Sep 17 00:00:00 2001
From: "Hyok S. Choi" <hyok.choi@samsung.com>
Date: Tue, 26 Sep 2006 17:38:05 +0900
Subject: [ARM] nommu: add ARM9TDMI core support

This patch adds ARM9TDMI core support which has no cache and no CP15
register(no memory control unit).

Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Makefile            |   1 +
 arch/arm/mm/Kconfig          |  13 +++++
 arch/arm/mm/Makefile         |   1 +
 arch/arm/mm/proc-arm9tdmi.S  | 134 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/cacheflush.h |   3 +-
 include/asm-arm/proc-fns.h   |   8 +++
 6 files changed, 159 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/mm/proc-arm9tdmi.S

(limited to 'include')

diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 8ec4b46d2cbb..a29e64175f8f 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -58,6 +58,7 @@ tune-$(CONFIG_CPU_ARM710)	:=-mtune=arm710
 tune-$(CONFIG_CPU_ARM7TDMI)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM720T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM740T)	:=-mtune=arm7tdmi
+tune-$(CONFIG_CPU_ARM9TDMI)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM920T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM922T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM925T)	:=-mtune=arm9tdmi
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 87f9fece9606..ed9d32c2469e 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -90,6 +90,19 @@ config CPU_ARM740T
 	  Say Y if you want support for the ARM740T processor.
 	  Otherwise, say N.
 
+# ARM9TDMI
+config CPU_ARM9TDMI
+	bool "Support ARM9TDMI processor"
+	select CPU_32v4T
+	select CPU_ABRT_EV4T
+	select CPU_CACHE_V4
+	help
+	  A 32-bit RISC microprocessor based on the ARM9 processor core
+	  which has no memory control unit and cache.
+
+	  Say Y if you want support for the ARM9TDMI processor.
+	  Otherwise, say N.
+
 # ARM920T
 config CPU_ARM920T
 	bool "Support ARM920T processor"
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index d10db822d3b8..053957eebe24 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -49,6 +49,7 @@ obj-$(CONFIG_CPU_ARM710)	+= proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM7TDMI)	+= proc-arm7tdmi.o
 obj-$(CONFIG_CPU_ARM720T)	+= proc-arm720.o
 obj-$(CONFIG_CPU_ARM740T)	+= proc-arm740.o
+obj-$(CONFIG_CPU_ARM9TDMI)	+= proc-arm9tdmi.o
 obj-$(CONFIG_CPU_ARM920T)	+= proc-arm920.o
 obj-$(CONFIG_CPU_ARM922T)	+= proc-arm922.o
 obj-$(CONFIG_CPU_ARM925T)	+= proc-arm925.o
diff --git a/arch/arm/mm/proc-arm9tdmi.S b/arch/arm/mm/proc-arm9tdmi.S
new file mode 100644
index 000000000000..95c90f13caa8
--- /dev/null
+++ b/arch/arm/mm/proc-arm9tdmi.S
@@ -0,0 +1,134 @@
+/*
+ *  linux/arch/arm/mm/proc-arm9tdmi.S: utility functions for ARM9TDMI
+ *
+ *  Copyright (C) 2003-2006 Hyok S. Choi <hyok.choi@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/procinfo.h>
+#include <asm/ptrace.h>
+
+	.text
+/*
+ * cpu_arm9tdmi_proc_init()
+ * cpu_arm9tdmi_do_idle()
+ * cpu_arm9tdmi_dcache_clean_area()
+ * cpu_arm9tdmi_switch_mm()
+ *
+ * These are not required.
+ */
+ENTRY(cpu_arm9tdmi_proc_init)
+ENTRY(cpu_arm9tdmi_do_idle)
+ENTRY(cpu_arm9tdmi_dcache_clean_area)
+ENTRY(cpu_arm9tdmi_switch_mm)
+		mov	pc, lr
+
+/*
+ * cpu_arm9tdmi_proc_fin()
+ */
+ENTRY(cpu_arm9tdmi_proc_fin)
+		mov	r0, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+		msr	cpsr_c, r0
+		mov	pc, lr
+
+/*
+ * Function: cpu_arm9tdmi_reset(loc)
+ * Params  : loc(r0)	address to jump to
+ * Purpose : Sets up everything for a reset and jump to the location for soft reset.
+ */
+ENTRY(cpu_arm9tdmi_reset)
+		mov	pc, r0
+
+		__INIT
+
+		.type	__arm9tdmi_setup, #function
+__arm9tdmi_setup:
+		mov	pc, lr
+		.size	__arm9tdmi_setup, . - __arm9tdmi_setup
+
+		__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+		.type	arm9tdmi_processor_functions, #object
+ENTRY(arm9tdmi_processor_functions)
+		.word	v4t_early_abort
+		.word	cpu_arm9tdmi_proc_init
+		.word	cpu_arm9tdmi_proc_fin
+		.word	cpu_arm9tdmi_reset
+		.word	cpu_arm9tdmi_do_idle
+		.word	cpu_arm9tdmi_dcache_clean_area
+		.word	cpu_arm9tdmi_switch_mm
+		.word	0		@ cpu_*_set_pte
+		.size	arm9tdmi_processor_functions, . - arm9tdmi_processor_functions
+
+		.section ".rodata"
+
+		.type	cpu_arch_name, #object
+cpu_arch_name:
+		.asciz	"armv4t"
+		.size	cpu_arch_name, . - cpu_arch_name
+
+		.type	cpu_elf_name, #object
+cpu_elf_name:
+		.asciz	"v4"
+		.size	cpu_elf_name, . - cpu_elf_name
+
+		.type	cpu_arm9tdmi_name, #object
+cpu_arm9tdmi_name:
+		.asciz	"ARM9TDMI"
+		.size	cpu_arm9tdmi_name, . - cpu_arm9tdmi_name
+
+		.type	cpu_p2001_name, #object
+cpu_p2001_name:
+		.asciz	"P2001"
+		.size	cpu_p2001_name, . - cpu_p2001_name
+
+		.align
+
+		.section ".proc.info.init", #alloc, #execinstr
+
+		.type	__arm9tdmi_proc_info, #object
+__arm9tdmi_proc_info:
+		.long	0x41009900
+		.long	0xfff8ff00
+		.long	0
+		.long	0
+		b	__arm9tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_arm9tdmi_name
+		.long	arm9tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__arm9tdmi_proc_info, . - __arm9dmi_proc_info
+
+		.type	__p2001_proc_info, #object
+__p2001_proc_info:
+		.long	0x41029000
+		.long	0xffffffff
+		.long	0
+		.long	0
+		b	__arm9tdmi_setup
+		.long	cpu_arch_name
+		.long	cpu_elf_name
+		.long	HWCAP_SWP | HWCAP_THUMB | HWCAP_26BIT
+		.long	cpu_p2001_name
+		.long	arm9tdmi_processor_functions
+		.long	0
+		.long	0
+		.long	v4_cache_fns
+		.size	__p2001_proc_info, . - __p2001_proc_info
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index 24924e64f883..77ff6fe86dbd 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -34,7 +34,8 @@
 # endif
 #endif
 
-#if defined(CONFIG_CPU_ARM720T) || defined(CONFIG_CPU_ARM7TDMI)
+#if defined(CONFIG_CPU_ARM720T) || defined(CONFIG_CPU_ARM7TDMI) || \
+    defined(CONFIG_CPU_ARM9TDMI)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
 # else
diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h
index 17dfc0de9658..7bb9dab310f5 100644
--- a/include/asm-arm/proc-fns.h
+++ b/include/asm-arm/proc-fns.h
@@ -65,6 +65,14 @@
 #   define CPU_NAME cpu_arm740
 #  endif
 # endif
+# ifdef CONFIG_CPU_ARM9TDMI
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_arm9tdmi
+#  endif
+# endif
 # ifdef CONFIG_CPU_ARM920T
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU
-- 
cgit v1.2.3


From d60674eb5d961b2421db16cc373dc163f38cc105 Mon Sep 17 00:00:00 2001
From: "Hyok S. Choi" <hyok.choi@samsung.com>
Date: Tue, 26 Sep 2006 17:38:18 +0900
Subject: [ARM] nommu: add ARM940T core support

This patch adds ARM940T core support which has 4KB D-cache, 4KB I-cache
and a MPU.

Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Makefile            |   1 +
 arch/arm/mm/Kconfig          |  20 ++-
 arch/arm/mm/Makefile         |   1 +
 arch/arm/mm/proc-arm940.S    | 369 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/cacheflush.h |   8 +
 include/asm-arm/proc-fns.h   |   8 +
 6 files changed, 405 insertions(+), 2 deletions(-)
 create mode 100644 arch/arm/mm/proc-arm940.S

(limited to 'include')

diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index a29e64175f8f..4c424c95dd3a 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -59,6 +59,7 @@ tune-$(CONFIG_CPU_ARM7TDMI)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM720T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM740T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM9TDMI)	:=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM940T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM920T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM922T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM925T)	:=-mtune=arm9tdmi
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index ed9d32c2469e..34d00738293f 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -184,6 +184,22 @@ config CPU_ARM926T
 	  Say Y if you want support for the ARM926T processor.
 	  Otherwise, say N.
 
+# ARM940T
+config CPU_ARM940T
+	bool "Support ARM940T processor" if ARCH_INTEGRATOR
+	select CPU_32v4T
+	select CPU_ABRT_EV4T
+	select CPU_CACHE_VIVT
+	select CPU_CP15_MPU
+	help
+	  ARM940T is a member of the ARM9TDMI family of general-
+	  purpose microprocessors with MPU and seperate 4KB
+	  instruction and 4KB data cases, each with a 4-word line
+	  length.
+
+	  Say Y if you want support for the ARM940T processor.
+	  Otherwise, say N.
+
 # ARM1020 - needs validating
 config CPU_ARM1020
 	bool "Support ARM1020T (rev 0) processor"
@@ -464,7 +480,7 @@ comment "Processor Features"
 
 config ARM_THUMB
 	bool "Support Thumb user binaries"
-	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6
+	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6
 	default y
 	help
 	  Say Y if you want to include kernel support for running user space
@@ -501,7 +517,7 @@ config CPU_DCACHE_DISABLE
 
 config CPU_DCACHE_WRITETHROUGH
 	bool "Force write through D-cache"
-	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM1020 || CPU_V6) && !CPU_DCACHE_DISABLE
+	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM1020 || CPU_V6) && !CPU_DCACHE_DISABLE
 	default y if CPU_ARM925T
 	help
 	  Say Y here to use the data cache in writethrough mode. Unless you
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 053957eebe24..f89fe1b2a4ad 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -54,6 +54,7 @@ obj-$(CONFIG_CPU_ARM920T)	+= proc-arm920.o
 obj-$(CONFIG_CPU_ARM922T)	+= proc-arm922.o
 obj-$(CONFIG_CPU_ARM925T)	+= proc-arm925.o
 obj-$(CONFIG_CPU_ARM926T)	+= proc-arm926.o
+obj-$(CONFIG_CPU_ARM940T)	+= proc-arm940.o
 obj-$(CONFIG_CPU_ARM1020)	+= proc-arm1020.o
 obj-$(CONFIG_CPU_ARM1020E)	+= proc-arm1020e.o
 obj-$(CONFIG_CPU_ARM1022)	+= proc-arm1022.o
diff --git a/arch/arm/mm/proc-arm940.S b/arch/arm/mm/proc-arm940.S
new file mode 100644
index 000000000000..14846ac2d0e6
--- /dev/null
+++ b/arch/arm/mm/proc-arm940.S
@@ -0,0 +1,369 @@
+/*
+ *  linux/arch/arm/mm/arm940.S: utility functions for ARM940T
+ *
+ *  Copyright (C) 2004-2006 Hyok S. Choi (hyok.choi@samsung.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/procinfo.h>
+#include <asm/ptrace.h>
+
+/* ARM940T has a 4KB DCache comprising 256 lines of 4 words */
+#define CACHE_DLINESIZE	16
+#define CACHE_DSEGMENTS	4
+#define CACHE_DENTRIES	64
+
+	.text
+/*
+ * cpu_arm940_proc_init()
+ * cpu_arm940_switch_mm()
+ *
+ * These are not required.
+ */
+ENTRY(cpu_arm940_proc_init)
+ENTRY(cpu_arm940_switch_mm)
+	mov	pc, lr
+
+/*
+ * cpu_arm940_proc_fin()
+ */
+ENTRY(cpu_arm940_proc_fin)
+	stmfd	sp!, {lr}
+	mov	ip, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+	msr	cpsr_c, ip
+	bl	arm940_flush_kern_cache_all
+	mrc	p15, 0, r0, c1, c0, 0		@ ctrl register
+	bic	r0, r0, #0x00001000		@ i-cache
+	bic	r0, r0, #0x00000004		@ d-cache
+	mcr	p15, 0, r0, c1, c0, 0		@ disable caches
+	ldmfd	sp!, {pc}
+
+/*
+ * cpu_arm940_reset(loc)
+ * Params  : r0 = address to jump to
+ * Notes   : This sets up everything for a reset
+ */
+ENTRY(cpu_arm940_reset)
+	mov	ip, #0
+	mcr	p15, 0, ip, c7, c5, 0		@ flush I cache
+	mcr	p15, 0, ip, c7, c6, 0		@ flush D cache
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mrc	p15, 0, ip, c1, c0, 0		@ ctrl register
+	bic	ip, ip, #0x00000005		@ .............c.p
+	bic	ip, ip, #0x00001000		@ i-cache
+	mcr	p15, 0, ip, c1, c0, 0		@ ctrl register
+	mov	pc, r0
+
+/*
+ * cpu_arm940_do_idle()
+ */
+	.align	5
+ENTRY(cpu_arm940_do_idle)
+	mcr	p15, 0, r0, c7, c0, 4		@ Wait for interrupt
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_all()
+ */
+ENTRY(arm940_flush_user_cache_all)
+	/* FALLTHROUGH */
+
+/*
+ *	flush_kern_cache_all()
+ *
+ *	Clean and invalidate the entire cache.
+ */
+ENTRY(arm940_flush_kern_cache_all)
+	mov	r2, #VM_EXEC
+	/* FALLTHROUGH */
+
+/*
+ *	flush_user_cache_range(start, end, flags)
+ *
+ *	There is no efficient way to flush a range of cache entries
+ *	in the specified address range. Thus, flushes all.
+ *
+ *	- start	- start address (inclusive)
+ *	- end	- end address (exclusive)
+ *	- flags	- vm_flags describing address space
+ */
+ENTRY(arm940_flush_user_cache_range)
+	mov	ip, #0
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, ip, c7, c6, 0		@ flush D cache
+#else
+	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
+1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
+2:	mcr	p15, 0, r3, c7, c14, 2		@ clean/flush D index
+	subs	r3, r3, #1 << 26
+	bcs	2b				@ entries 63 to 0
+	subs	r1, r1, #1 << 4
+	bcs	1b				@ segments 3 to 0
+#endif
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c5, 0		@ invalidate I cache
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	coherent_kern_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(arm940_coherent_kern_range)
+	/* FALLTHROUGH */
+
+/*
+ *	coherent_user_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(arm940_coherent_user_range)
+	/* FALLTHROUGH */
+
+/*
+ *	flush_kern_dcache_page(void *page)
+ *
+ *	Ensure no D cache aliasing occurs, either with itself or
+ *	the I cache
+ *
+ *	- addr	- page aligned address
+ */
+ENTRY(arm940_flush_kern_dcache_page)
+	mov	ip, #0
+	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
+1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
+2:	mcr	p15, 0, r3, c7, c14, 2		@ clean/flush D index
+	subs	r3, r3, #1 << 26
+	bcs	2b				@ entries 63 to 0
+	subs	r1, r1, #1 << 4
+	bcs	1b				@ segments 7 to 0
+	mcr	p15, 0, ip, c7, c5, 0		@ invalidate I cache
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_inv_range(start, end)
+ *
+ *	There is no efficient way to invalidate a specifid virtual
+ *	address range. Thus, invalidates all.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(arm940_dma_inv_range)
+	mov	ip, #0
+	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
+1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
+2:	mcr	p15, 0, r3, c7, c6, 2		@ flush D entry
+	subs	r3, r3, #1 << 26
+	bcs	2b				@ entries 63 to 0
+	subs	r1, r1, #1 << 4
+	bcs	1b				@ segments 7 to 0
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_clean_range(start, end)
+ *
+ *	There is no efficient way to clean a specifid virtual
+ *	address range. Thus, cleans all.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(arm940_dma_clean_range)
+ENTRY(cpu_arm940_dcache_clean_area)
+	mov	ip, #0
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
+1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
+2:	mcr	p15, 0, r3, c7, c10, 2		@ clean D entry
+	subs	r3, r3, #1 << 26
+	bcs	2b				@ entries 63 to 0
+	subs	r1, r1, #1 << 4
+	bcs	1b				@ segments 7 to 0
+#endif
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_flush_range(start, end)
+ *
+ *	There is no efficient way to clean and invalidate a specifid
+ *	virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(arm940_dma_flush_range)
+	mov	ip, #0
+	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
+1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
+2:
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r3, c7, c14, 2		@ clean/flush D entry
+#else
+	mcr	p15, 0, r3, c7, c10, 2		@ clean D entry
+#endif
+	subs	r3, r3, #1 << 26
+	bcs	2b				@ entries 63 to 0
+	subs	r1, r1, #1 << 4
+	bcs	1b				@ segments 7 to 0
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+ENTRY(arm940_cache_fns)
+	.long	arm940_flush_kern_cache_all
+	.long	arm940_flush_user_cache_all
+	.long	arm940_flush_user_cache_range
+	.long	arm940_coherent_kern_range
+	.long	arm940_coherent_user_range
+	.long	arm940_flush_kern_dcache_page
+	.long	arm940_dma_inv_range
+	.long	arm940_dma_clean_range
+	.long	arm940_dma_flush_range
+
+	__INIT
+
+	.type	__arm940_setup, #function
+__arm940_setup:
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c5, 0		@ invalidate I cache
+	mcr	p15, 0, r0, c7, c6, 0		@ invalidate D cache
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+
+	mcr	p15, 0, r0, c6, c3, 0		@ disable data area 3~7
+	mcr	p15, 0, r0, c6, c4, 0
+	mcr	p15, 0, r0, c6, c5, 0
+	mcr	p15, 0, r0, c6, c6, 0
+	mcr	p15, 0, r0, c6, c7, 0
+
+	mcr	p15, 0, r0, c6, c3, 1		@ disable instruction area 3~7
+	mcr	p15, 0, r0, c6, c4, 1
+	mcr	p15, 0, r0, c6, c5, 1
+	mcr	p15, 0, r0, c6, c6, 1
+	mcr	p15, 0, r0, c6, c7, 1
+
+	mov	r0, #0x0000003F			@ base = 0, size = 4GB
+	mcr	p15, 0, r0, c6,	c0, 0		@ set area 0, default
+	mcr	p15, 0, r0, c6,	c0, 1
+
+	ldr	r0, =(CONFIG_DRAM_BASE & 0xFFFFF000) @ base[31:12] of RAM
+	ldr	r1, =(CONFIG_DRAM_SIZE >> 12)	@ size of RAM (must be >= 4KB)
+	mov	r2, #10				@ 11 is the minimum (4KB)
+1:	add	r2, r2, #1			@ area size *= 2
+	mov	r1, r1, lsr #1
+	bne	1b				@ count not zero r-shift
+	orr	r0, r0, r2, lsl #1		@ the area register value
+	orr	r0, r0, #1			@ set enable bit
+	mcr	p15, 0, r0, c6,	c1, 0		@ set area 1, RAM
+	mcr	p15, 0, r0, c6,	c1, 1
+
+	ldr	r0, =(CONFIG_FLASH_MEM_BASE & 0xFFFFF000) @ base[31:12] of FLASH
+	ldr	r1, =(CONFIG_FLASH_SIZE >> 12)	@ size of FLASH (must be >= 4KB)
+	mov	r2, #10				@ 11 is the minimum (4KB)
+1:	add	r2, r2, #1			@ area size *= 2
+	mov	r1, r1, lsr #1
+	bne	1b				@ count not zero r-shift
+	orr	r0, r0, r2, lsl #1		@ the area register value
+	orr	r0, r0, #1			@ set enable bit
+	mcr	p15, 0, r0, c6,	c2, 0		@ set area 2, ROM/FLASH
+	mcr	p15, 0, r0, c6,	c2, 1
+
+	mov	r0, #0x06
+	mcr	p15, 0, r0, c2, c0, 0		@ Region 1&2 cacheable
+	mcr	p15, 0, r0, c2, c0, 1
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mov	r0, #0x00			@ disable whole write buffer
+#else
+	mov	r0, #0x02			@ Region 1 write bufferred
+#endif
+	mcr	p15, 0, r0, c3, c0, 0
+
+	mov	r0, #0x10000
+	sub	r0, r0, #1			@ r0 = 0xffff
+	mcr	p15, 0, r0, c5, c0, 0		@ all read/write access
+	mcr	p15, 0, r0, c5, c0, 1
+
+	mrc	p15, 0, r0, c1, c0		@ get control register
+	orr	r0, r0, #0x00001000		@ I-cache
+	orr	r0, r0, #0x00000005		@ MPU/D-cache
+
+	mov	pc, lr
+
+	.size	__arm940_setup, . - __arm940_setup
+
+	__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+	.type	arm940_processor_functions, #object
+ENTRY(arm940_processor_functions)
+	.word	v4t_early_abort
+	.word	cpu_arm940_proc_init
+	.word	cpu_arm940_proc_fin
+	.word	cpu_arm940_reset
+	.word   cpu_arm940_do_idle
+	.word	cpu_arm940_dcache_clean_area
+	.word	cpu_arm940_switch_mm
+	.word	0		@ cpu_*_set_pte
+	.size	arm940_processor_functions, . - arm940_processor_functions
+
+	.section ".rodata"
+
+.type	cpu_arch_name, #object
+cpu_arch_name:
+	.asciz	"armv4t"
+	.size	cpu_arch_name, . - cpu_arch_name
+
+	.type	cpu_elf_name, #object
+cpu_elf_name:
+	.asciz	"v4"
+	.size	cpu_elf_name, . - cpu_elf_name
+
+	.type	cpu_arm940_name, #object
+cpu_arm940_name:
+	.ascii	"ARM940T"
+	.size	cpu_arm940_name, . - cpu_arm940_name
+
+	.align
+
+	.section ".proc.info.init", #alloc, #execinstr
+
+	.type	__arm940_proc_info,#object
+__arm940_proc_info:
+	.long	0x41009400
+	.long	0xff00fff0
+	.long	0
+	b	__arm940_setup
+	.long	cpu_arch_name
+	.long	cpu_elf_name
+	.long	HWCAP_SWP | HWCAP_HALF | HWCAP_THUMB
+	.long	cpu_arm940_name
+	.long	arm940_processor_functions
+	.long	0
+	.long	0
+	.long	arm940_cache_fns
+	.size	__arm940_proc_info, . - __arm940_proc_info
+
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index 77ff6fe86dbd..b0a8603400be 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -56,6 +56,14 @@
 # endif
 #endif
 
+#if defined(CONFIG_CPU_ARM940T)
+# ifdef _CACHE
+#  define MULTI_CACHE 1
+# else
+#  define _CACHE arm940
+# endif
+#endif
+
 #if defined(CONFIG_CPU_SA110) || defined(CONFIG_CPU_SA1100)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h
index 7bb9dab310f5..87f3ea97f48c 100644
--- a/include/asm-arm/proc-fns.h
+++ b/include/asm-arm/proc-fns.h
@@ -105,6 +105,14 @@
 #   define CPU_NAME cpu_arm926
 #  endif
 # endif
+# ifdef CONFIG_CPU_ARM940T
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_arm940
+#  endif
+# endif
 # ifdef CONFIG_CPU_SA110
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU
-- 
cgit v1.2.3


From f37f46eb1c0bd0b11c34ef06c7365658be989d80 Mon Sep 17 00:00:00 2001
From: "Hyok S. Choi" <hyok.choi@samsung.com>
Date: Tue, 26 Sep 2006 17:38:32 +0900
Subject: [ARM] nommu: add ARM946E-S core support

This patch adds ARM946E-S core support which has typically 8KB I&D cache.
It has a MPU and supports ARMv5TE instruction set.

Because the ARM946E-S core can be synthesizable with various cache size,
CONFIG_CPU_DCACHE_SIZE is defined for vendor specific configurations.

Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Makefile            |   1 +
 arch/arm/mm/Kconfig          |  34 +++-
 arch/arm/mm/Makefile         |   1 +
 arch/arm/mm/proc-arm946.S    | 424 +++++++++++++++++++++++++++++++++++++++++++
 include/asm-arm/cacheflush.h |   8 +
 include/asm-arm/proc-fns.h   |   8 +
 6 files changed, 473 insertions(+), 3 deletions(-)
 create mode 100644 arch/arm/mm/proc-arm946.S

(limited to 'include')

diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 4c424c95dd3a..2a0b2c8a1fe0 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -60,6 +60,7 @@ tune-$(CONFIG_CPU_ARM720T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM740T)	:=-mtune=arm7tdmi
 tune-$(CONFIG_CPU_ARM9TDMI)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM940T)	:=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM946T)	:=$(call cc-option,-mtune=arm9e,-mtune=arm9tdmi)
 tune-$(CONFIG_CPU_ARM920T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM922T)	:=-mtune=arm9tdmi
 tune-$(CONFIG_CPU_ARM925T)	:=-mtune=arm9tdmi
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 34d00738293f..893d3fc8078f 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -200,6 +200,21 @@ config CPU_ARM940T
 	  Say Y if you want support for the ARM940T processor.
 	  Otherwise, say N.
 
+# ARM946E-S
+config CPU_ARM946E
+	bool "Support ARM946E-S processor" if ARCH_INTEGRATOR
+	select CPU_32v5
+	select CPU_ABRT_EV5T
+	select CPU_CACHE_VIVT
+	select CPU_CP15_MPU
+	help
+	  ARM946E-S is a member of the ARM9E-S family of high-
+	  performance, 32-bit system-on-chip processor solutions.
+	  The TCM and ARMv5TE 32-bit instruction set is supported.
+
+	  Say Y if you want support for the ARM946E-S processor.
+	  Otherwise, say N.
+
 # ARM1020 - needs validating
 config CPU_ARM1020
 	bool "Support ARM1020T (rev 0) processor"
@@ -480,7 +495,7 @@ comment "Processor Features"
 
 config ARM_THUMB
 	bool "Support Thumb user binaries"
-	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6
+	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6
 	default y
 	help
 	  Say Y if you want to include kernel support for running user space
@@ -515,9 +530,22 @@ config CPU_DCACHE_DISABLE
 	  Say Y here to disable the processor data cache. Unless
 	  you have a reason not to or are unsure, say N.
 
+config CPU_DCACHE_SIZE
+	hex
+	depends on CPU_ARM740T || CPU_ARM946E
+	default 0x00001000 if CPU_ARM740T
+	default 0x00002000 # default size for ARM946E-S
+	help
+	  Some cores are synthesizable to have various sized cache. For
+	  ARM946E-S case, it can vary from 0KB to 1MB.
+	  To support such cache operations, it is efficient to know the size
+	  before compile time.
+	  If your SoC is configured to have a different size, define the value
+	  here with proper conditions.
+
 config CPU_DCACHE_WRITETHROUGH
 	bool "Force write through D-cache"
-	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM1020 || CPU_V6) && !CPU_DCACHE_DISABLE
+	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_V6) && !CPU_DCACHE_DISABLE
 	default y if CPU_ARM925T
 	help
 	  Say Y here to use the data cache in writethrough mode. Unless you
@@ -525,7 +553,7 @@ config CPU_DCACHE_WRITETHROUGH
 
 config CPU_CACHE_ROUND_ROBIN
 	bool "Round robin I and D cache replacement algorithm"
-	depends on (CPU_ARM926T || CPU_ARM1020) && (!CPU_ICACHE_DISABLE || !CPU_DCACHE_DISABLE)
+	depends on (CPU_ARM926T || CPU_ARM946E || CPU_ARM1020) && (!CPU_ICACHE_DISABLE || !CPU_DCACHE_DISABLE)
 	help
 	  Say Y here to use the predictable round-robin cache replacement
 	  policy.  Unless you specifically require this or are unsure, say N.
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index f89fe1b2a4ad..ed81b9ef10cb 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -55,6 +55,7 @@ obj-$(CONFIG_CPU_ARM922T)	+= proc-arm922.o
 obj-$(CONFIG_CPU_ARM925T)	+= proc-arm925.o
 obj-$(CONFIG_CPU_ARM926T)	+= proc-arm926.o
 obj-$(CONFIG_CPU_ARM940T)	+= proc-arm940.o
+obj-$(CONFIG_CPU_ARM946E)	+= proc-arm946.o
 obj-$(CONFIG_CPU_ARM1020)	+= proc-arm1020.o
 obj-$(CONFIG_CPU_ARM1020E)	+= proc-arm1020e.o
 obj-$(CONFIG_CPU_ARM1022)	+= proc-arm1022.o
diff --git a/arch/arm/mm/proc-arm946.S b/arch/arm/mm/proc-arm946.S
new file mode 100644
index 000000000000..6dc7942c4cfe
--- /dev/null
+++ b/arch/arm/mm/proc-arm946.S
@@ -0,0 +1,424 @@
+/*
+ *  linux/arch/arm/mm/arm946.S: utility functions for ARM946E-S
+ *
+ *  Copyright (C) 2004-2006 Hyok S. Choi (hyok.choi@samsung.com)
+ *
+ *  (Many of cache codes are from proc-arm926.S)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/procinfo.h>
+#include <asm/ptrace.h>
+
+/*
+ * ARM946E-S is synthesizable to have 0KB to 1MB sized D-Cache,
+ * comprising 256 lines of 32 bytes (8 words).
+ */
+#define CACHE_DSIZE	(CONFIG_CPU_DCACHE_SIZE) /* typically 8KB. */
+#define CACHE_DLINESIZE	32			/* fixed */
+#define CACHE_DSEGMENTS	4			/* fixed */
+#define CACHE_DENTRIES	(CACHE_DSIZE / CACHE_DSEGMENTS / CACHE_DLINESIZE)
+#define CACHE_DLIMIT	(CACHE_DSIZE * 4)	/* benchmark needed */
+
+	.text
+/*
+ * cpu_arm946_proc_init()
+ * cpu_arm946_switch_mm()
+ *
+ * These are not required.
+ */
+ENTRY(cpu_arm946_proc_init)
+ENTRY(cpu_arm946_switch_mm)
+	mov	pc, lr
+
+/*
+ * cpu_arm946_proc_fin()
+ */
+ENTRY(cpu_arm946_proc_fin)
+	stmfd	sp!, {lr}
+	mov	ip, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+	msr	cpsr_c, ip
+	bl	arm946_flush_kern_cache_all
+	mrc	p15, 0, r0, c1, c0, 0		@ ctrl register
+	bic	r0, r0, #0x00001000		@ i-cache
+	bic	r0, r0, #0x00000004		@ d-cache
+	mcr	p15, 0, r0, c1, c0, 0		@ disable caches
+	ldmfd	sp!, {pc}
+
+/*
+ * cpu_arm946_reset(loc)
+ * Params  : r0 = address to jump to
+ * Notes   : This sets up everything for a reset
+ */
+ENTRY(cpu_arm946_reset)
+	mov	ip, #0
+	mcr	p15, 0, ip, c7, c5, 0		@ flush I cache
+	mcr	p15, 0, ip, c7, c6, 0		@ flush D cache
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+	mrc	p15, 0, ip, c1, c0, 0		@ ctrl register
+	bic	ip, ip, #0x00000005		@ .............c.p
+	bic	ip, ip, #0x00001000		@ i-cache
+	mcr	p15, 0, ip, c1, c0, 0		@ ctrl register
+	mov	pc, r0
+
+/*
+ * cpu_arm946_do_idle()
+ */
+	.align	5
+ENTRY(cpu_arm946_do_idle)
+	mcr	p15, 0, r0, c7, c0, 4		@ Wait for interrupt
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_all()
+ */
+ENTRY(arm946_flush_user_cache_all)
+	/* FALLTHROUGH */
+
+/*
+ *	flush_kern_cache_all()
+ *
+ *	Clean and invalidate the entire cache.
+ */
+ENTRY(arm946_flush_kern_cache_all)
+	mov	r2, #VM_EXEC
+	mov	ip, #0
+__flush_whole_cache:
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, ip, c7, c6, 0		@ flush D cache
+#else
+	mov	r1, #(CACHE_DSEGMENTS - 1) << 29 @ 4 segments
+1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 4 @ n entries
+2:	mcr	p15, 0, r3, c7, c14, 2		@ clean/flush D index
+	subs	r3, r3, #1 << 4
+	bcs	2b				@ entries n to 0
+	subs	r1, r1, #1 << 29
+	bcs	1b				@ segments 3 to 0
+#endif
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c5, 0		@ flush I cache
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_range(start, end, flags)
+ *
+ *	Clean and invalidate a range of cache entries in the
+ *	specified address range.
+ *
+ *	- start	- start address (inclusive)
+ *	- end	- end address (exclusive)
+ *	- flags	- vm_flags describing address space
+ * (same as arm926)
+ */
+ENTRY(arm946_flush_user_cache_range)
+	mov	ip, #0
+	sub	r3, r1, r0			@ calculate total size
+	cmp	r3, #CACHE_DLIMIT
+	bhs	__flush_whole_cache
+
+1:	tst	r2, #VM_EXEC
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+#else
+	mcr	p15, 0, r0, c7, c14, 1		@ clean and invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	mcr	p15, 0, r0, c7, c14, 1		@ clean and invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+#endif
+	cmp	r0, r1
+	blo	1b
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	coherent_kern_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(arm946_coherent_kern_range)
+	/* FALLTHROUGH */
+
+/*
+ *	coherent_user_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ * (same as arm926)
+ */
+ENTRY(arm946_coherent_user_range)
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	mcr	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	flush_kern_dcache_page(void *page)
+ *
+ *	Ensure no D cache aliasing occurs, either with itself or
+ *	the I cache
+ *
+ *	- addr	- page aligned address
+ * (same as arm926)
+ */
+ENTRY(arm946_flush_kern_dcache_page)
+	add	r1, r0, #PAGE_SZ
+1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c5, 0		@ invalidate I cache
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_inv_range(start, end)
+ *
+ *	Invalidate (discard) the specified virtual address range.
+ *	May not write back any entries.  If 'start' or 'end'
+ *	are not cache line aligned, those lines must be written
+ *	back.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ * (same as arm926)
+ */
+ENTRY(arm946_dma_inv_range)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	tst	r0, #CACHE_DLINESIZE - 1
+	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
+	tst	r1, #CACHE_DLINESIZE - 1
+	mcrne	p15, 0, r1, c7, c10, 1		@ clean D entry
+#endif
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_clean_range(start, end)
+ *
+ *	Clean the specified virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ *
+ * (same as arm926)
+ */
+ENTRY(arm946_dma_clean_range)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_flush_range(start, end)
+ *
+ *	Clean and invalidate the specified virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ *
+ * (same as arm926)
+ */
+ENTRY(arm946_dma_flush_range)
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
+#else
+	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+#endif
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+ENTRY(arm946_cache_fns)
+	.long	arm946_flush_kern_cache_all
+	.long	arm946_flush_user_cache_all
+	.long	arm946_flush_user_cache_range
+	.long	arm946_coherent_kern_range
+	.long	arm946_coherent_user_range
+	.long	arm946_flush_kern_dcache_page
+	.long	arm946_dma_inv_range
+	.long	arm946_dma_clean_range
+	.long	arm946_dma_flush_range
+
+
+ENTRY(cpu_arm946_dcache_clean_area)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	subs	r1, r1, #CACHE_DLINESIZE
+	bhi	1b
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+	__INIT
+
+	.type	__arm946_setup, #function
+__arm946_setup:
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c5, 0		@ invalidate I cache
+	mcr	p15, 0, r0, c7, c6, 0		@ invalidate D cache
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+
+	mcr	p15, 0, r0, c6, c3, 0		@ disable memory region 3~7
+	mcr	p15, 0, r0, c6, c4, 0
+	mcr	p15, 0, r0, c6, c5, 0
+	mcr	p15, 0, r0, c6, c6, 0
+	mcr	p15, 0, r0, c6, c7, 0
+
+	mov	r0, #0x0000003F			@ base = 0, size = 4GB
+	mcr	p15, 0, r0, c6,	c0, 0		@ set region 0, default
+
+	ldr	r0, =(CONFIG_DRAM_BASE & 0xFFFFF000) @ base[31:12] of RAM
+	ldr	r1, =(CONFIG_DRAM_SIZE >> 12)	@ size of RAM (must be >= 4KB)
+	mov	r2, #10				@ 11 is the minimum (4KB)
+1:	add	r2, r2, #1			@ area size *= 2
+	mov	r1, r1, lsr #1
+	bne	1b				@ count not zero r-shift
+	orr	r0, r0, r2, lsl #1		@ the region register value
+	orr	r0, r0, #1			@ set enable bit
+	mcr	p15, 0, r0, c6,	c1, 0		@ set region 1, RAM
+
+	ldr	r0, =(CONFIG_FLASH_MEM_BASE & 0xFFFFF000) @ base[31:12] of FLASH
+	ldr	r1, =(CONFIG_FLASH_SIZE >> 12)	@ size of FLASH (must be >= 4KB)
+	mov	r2, #10				@ 11 is the minimum (4KB)
+1:	add	r2, r2, #1			@ area size *= 2
+	mov	r1, r1, lsr #1
+	bne	1b				@ count not zero r-shift
+	orr	r0, r0, r2, lsl #1		@ the region register value
+	orr	r0, r0, #1			@ set enable bit
+	mcr	p15, 0, r0, c6,	c2, 0		@ set region 2, ROM/FLASH
+
+	mov	r0, #0x06
+	mcr	p15, 0, r0, c2, c0, 0		@ region 1,2 d-cacheable
+	mcr	p15, 0, r0, c2, c0, 1		@ region 1,2 i-cacheable
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mov	r0, #0x00			@ disable whole write buffer
+#else
+	mov	r0, #0x02			@ region 1 write bufferred
+#endif
+	mcr	p15, 0, r0, c3, c0, 0
+
+/*
+ *  Access Permission Settings for future permission control by PU.
+ *
+ *				priv.	user
+ * 	region 0 (whole)	rw	--	: b0001
+ * 	region 1 (RAM)		rw	rw	: b0011
+ * 	region 2 (FLASH)	rw	r-	: b0010
+ *	region 3~7 (none)	--	--	: b0000
+ */
+	mov	r0, #0x00000031
+	orr	r0, r0, #0x00000200
+	mcr	p15, 0, r0, c5, c0, 2		@ set data access permission
+	mcr	p15, 0, r0, c5, c0, 3		@ set inst. access permission
+
+	mrc	p15, 0, r0, c1, c0		@ get control register
+	orr	r0, r0, #0x00001000		@ I-cache
+	orr	r0, r0, #0x00000005		@ MPU/D-cache
+#ifdef CONFIG_CPU_CACHE_ROUND_ROBIN
+	orr	r0, r0, #0x00004000		@ .1.. .... .... ....
+#endif
+	mov	pc, lr
+
+	.size	__arm946_setup, . - __arm946_setup
+
+	__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+	.type	arm946_processor_functions, #object
+ENTRY(arm946_processor_functions)
+	.word	v5t_early_abort
+	.word	cpu_arm946_proc_init
+	.word	cpu_arm946_proc_fin
+	.word	cpu_arm946_reset
+	.word   cpu_arm946_do_idle
+
+	.word	cpu_arm946_dcache_clean_area
+	.word	cpu_arm946_switch_mm
+	.word	0		@ cpu_*_set_pte
+	.size	arm946_processor_functions, . - arm946_processor_functions
+
+	.section ".rodata"
+
+	.type	cpu_arch_name, #object
+cpu_arch_name:
+	.asciz	"armv5te"
+	.size	cpu_arch_name, . - cpu_arch_name
+
+	.type	cpu_elf_name, #object
+cpu_elf_name:
+	.asciz	"v5t"
+	.size	cpu_elf_name, . - cpu_elf_name
+
+	.type	cpu_arm946_name, #object
+cpu_arm946_name:
+	.ascii	"ARM946E-S"
+	.size	cpu_arm946_name, . - cpu_arm946_name
+
+	.align
+
+	.section ".proc.info.init", #alloc, #execinstr
+	.type	__arm946_proc_info,#object
+__arm946_proc_info:
+	.long	0x41009460
+	.long	0xff00fff0
+	.long	0
+	b	__arm946_setup
+	.long	cpu_arch_name
+	.long	cpu_elf_name
+	.long	HWCAP_SWP | HWCAP_HALF | HWCAP_THUMB
+	.long	cpu_arm946_name
+	.long	arm946_processor_functions
+	.long	0
+	.long	0
+	.long	arm940_cache_fns
+	.size	__arm946_proc_info, . - __arm946_proc_info
+
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index b0a8603400be..b611a8ea0bb2 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -64,6 +64,14 @@
 # endif
 #endif
 
+#if defined(CONFIG_CPU_ARM946E)
+# ifdef _CACHE
+#  define MULTI_CACHE 1
+# else
+#  define _CACHE arm946
+# endif
+#endif
+
 #if defined(CONFIG_CPU_SA110) || defined(CONFIG_CPU_SA1100)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
diff --git a/include/asm-arm/proc-fns.h b/include/asm-arm/proc-fns.h
index 87f3ea97f48c..ea7e54c319be 100644
--- a/include/asm-arm/proc-fns.h
+++ b/include/asm-arm/proc-fns.h
@@ -113,6 +113,14 @@
 #   define CPU_NAME cpu_arm940
 #  endif
 # endif
+# ifdef CONFIG_CPU_ARM946E
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_arm946
+#  endif
+# endif
 # ifdef CONFIG_CPU_SA110
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU
-- 
cgit v1.2.3


From 6cc7cbef948ea2660cc40d7aab090a479f7db6a2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 27 Sep 2006 18:00:35 +0100
Subject: [ARM] Use CPU_CACHE_* where possible in asm/cacheflush.h

Three of the generic cache method options were using explicit CPU
types, whereas they could use the CPU_CACHE_* definitions instead.
Switch them over to use the CPU_CACHE_* definitions.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/cacheflush.h | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index b611a8ea0bb2..f0845646aacb 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -25,8 +25,7 @@
 #undef _CACHE
 #undef MULTI_CACHE
 
-#if defined(CONFIG_CPU_ARM610) || defined(CONFIG_CPU_ARM710) || \
-    defined(CONFIG_CPU_ARM740T)
+#if defined(CONFIG_CPU_CACHE_V3)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
 # else
@@ -34,8 +33,7 @@
 # endif
 #endif
 
-#if defined(CONFIG_CPU_ARM720T) || defined(CONFIG_CPU_ARM7TDMI) || \
-    defined(CONFIG_CPU_ARM9TDMI)
+#if defined(CONFIG_CPU_CACHE_V4)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
 # else
@@ -72,7 +70,7 @@
 # endif
 #endif
 
-#if defined(CONFIG_CPU_SA110) || defined(CONFIG_CPU_SA1100)
+#if defined(CONFIG_CPU_CACHE_V4WB)
 # ifdef _CACHE
 #  define MULTI_CACHE 1
 # else
-- 
cgit v1.2.3


From 3a16f7b4a75d68364c3278523f51ac141a12758a Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 29 Jun 2006 12:27:23 -0700
Subject: USB: move <linux/usb_otg.h> to <linux/usb/otg.h>

Move <linux/usb_otg.h> to <linux/usb/otg.h>.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/plat-omap/usb.c         |   2 +-
 drivers/i2c/chips/isp1301_omap.c |   2 +-
 drivers/usb/gadget/omap_udc.c    |   2 +-
 drivers/usb/host/ohci-hcd.c      |   2 +-
 include/linux/usb/otg.h          | 131 +++++++++++++++++++++++++++++++++++++++
 include/linux/usb_otg.h          | 131 ---------------------------------------
 6 files changed, 135 insertions(+), 135 deletions(-)
 create mode 100644 include/linux/usb/otg.h
 delete mode 100644 include/linux/usb_otg.h

(limited to 'include')

diff --git a/arch/arm/plat-omap/usb.c b/arch/arm/plat-omap/usb.c
index 9b815327b6a5..7e8096809be2 100644
--- a/arch/arm/plat-omap/usb.c
+++ b/arch/arm/plat-omap/usb.c
@@ -26,7 +26,7 @@
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/platform_device.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
diff --git a/drivers/i2c/chips/isp1301_omap.c b/drivers/i2c/chips/isp1301_omap.c
index f92505b94c61..182f04953466 100644
--- a/drivers/i2c/chips/isp1301_omap.c
+++ b/drivers/i2c/chips/isp1301_omap.c
@@ -30,7 +30,7 @@
 #include <linux/usb_ch9.h>
 #include <linux/usb_gadget.h>
 #include <linux/usb.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 #include <linux/i2c.h>
 #include <linux/workqueue.h>
 
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 2de9748ee673..81f0389fcc94 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -40,7 +40,7 @@
 #include <linux/platform_device.h>
 #include <linux/usb_ch9.h>
 #include <linux/usb_gadget.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 #include <linux/dma-mapping.h>
 
 #include <asm/byteorder.h>
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index cbf38ae5ae76..7c3d8c60a60f 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -88,7 +88,7 @@
 #include <linux/timer.h>
 #include <linux/list.h>
 #include <linux/usb.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 #include <linux/dma-mapping.h> 
 #include <linux/dmapool.h>
 #include <linux/reboot.h>
diff --git a/include/linux/usb/otg.h b/include/linux/usb/otg.h
new file mode 100644
index 000000000000..9897f7a818c5
--- /dev/null
+++ b/include/linux/usb/otg.h
@@ -0,0 +1,131 @@
+// include/linux/usb/otg.h
+
+/*
+ * These APIs may be used between USB controllers.  USB device drivers
+ * (for either host or peripheral roles) don't use these calls; they
+ * continue to use just usb_device and usb_gadget.
+ */
+
+
+/* OTG defines lots of enumeration states before device reset */
+enum usb_otg_state {
+	OTG_STATE_UNDEFINED = 0,
+
+	/* single-role peripheral, and dual-role default-b */
+	OTG_STATE_B_IDLE,
+	OTG_STATE_B_SRP_INIT,
+	OTG_STATE_B_PERIPHERAL,
+
+	/* extra dual-role default-b states */
+	OTG_STATE_B_WAIT_ACON,
+	OTG_STATE_B_HOST,
+
+	/* dual-role default-a */
+	OTG_STATE_A_IDLE,
+	OTG_STATE_A_WAIT_VRISE,
+	OTG_STATE_A_WAIT_BCON,
+	OTG_STATE_A_HOST,
+	OTG_STATE_A_SUSPEND,
+	OTG_STATE_A_PERIPHERAL,
+	OTG_STATE_A_WAIT_VFALL,
+	OTG_STATE_A_VBUS_ERR,
+};
+
+/*
+ * the otg driver needs to interact with both device side and host side
+ * usb controllers.  it decides which controller is active at a given
+ * moment, using the transceiver, ID signal, HNP and sometimes static
+ * configuration information (including "board isn't wired for otg").
+ */
+struct otg_transceiver {
+	struct device		*dev;
+	const char		*label;
+
+	u8			default_a;
+	enum usb_otg_state	state;
+
+	struct usb_bus		*host;
+	struct usb_gadget	*gadget;
+
+	/* to pass extra port status to the root hub */
+	u16			port_status;
+	u16			port_change;
+
+	/* bind/unbind the host controller */
+	int	(*set_host)(struct otg_transceiver *otg,
+				struct usb_bus *host);
+
+	/* bind/unbind the peripheral controller */
+	int	(*set_peripheral)(struct otg_transceiver *otg,
+				struct usb_gadget *gadget);
+
+	/* effective for B devices, ignored for A-peripheral */
+	int	(*set_power)(struct otg_transceiver *otg,
+				unsigned mA);
+
+	/* for non-OTG B devices: set transceiver into suspend mode */
+	int	(*set_suspend)(struct otg_transceiver *otg,
+				int suspend);
+
+	/* for B devices only:  start session with A-Host */
+	int	(*start_srp)(struct otg_transceiver *otg);
+
+	/* start or continue HNP role switch */
+	int	(*start_hnp)(struct otg_transceiver *otg);
+
+};
+
+
+/* for board-specific init logic */
+extern int otg_set_transceiver(struct otg_transceiver *);
+
+
+/* for usb host and peripheral controller drivers */
+extern struct otg_transceiver *otg_get_transceiver(void);
+
+static inline int
+otg_start_hnp(struct otg_transceiver *otg)
+{
+	return otg->start_hnp(otg);
+}
+
+
+/* for HCDs */
+static inline int
+otg_set_host(struct otg_transceiver *otg, struct usb_bus *host)
+{
+	return otg->set_host(otg, host);
+}
+
+
+/* for usb peripheral controller drivers */
+static inline int
+otg_set_peripheral(struct otg_transceiver *otg, struct usb_gadget *periph)
+{
+	return otg->set_peripheral(otg, periph);
+}
+
+static inline int
+otg_set_power(struct otg_transceiver *otg, unsigned mA)
+{
+	return otg->set_power(otg, mA);
+}
+
+static inline int
+otg_set_suspend(struct otg_transceiver *otg, int suspend)
+{
+	if (otg->set_suspend != NULL)
+		return otg->set_suspend(otg, suspend);
+	else
+		return 0;
+}
+
+static inline int
+otg_start_srp(struct otg_transceiver *otg)
+{
+	return otg->start_srp(otg);
+}
+
+
+/* for OTG controller drivers (and maybe other stuff) */
+extern int usb_bus_start_enum(struct usb_bus *bus, unsigned port_num);
diff --git a/include/linux/usb_otg.h b/include/linux/usb_otg.h
deleted file mode 100644
index f827f6e203c2..000000000000
--- a/include/linux/usb_otg.h
+++ /dev/null
@@ -1,131 +0,0 @@
-// include/linux/usb_otg.h 
-
-/*
- * These APIs may be used between USB controllers.  USB device drivers
- * (for either host or peripheral roles) don't use these calls; they
- * continue to use just usb_device and usb_gadget.
- */
-
-
-/* OTG defines lots of enumeration states before device reset */
-enum usb_otg_state {
-	OTG_STATE_UNDEFINED = 0,
-
-	/* single-role peripheral, and dual-role default-b */
-	OTG_STATE_B_IDLE,
-	OTG_STATE_B_SRP_INIT,
-	OTG_STATE_B_PERIPHERAL,
-
-	/* extra dual-role default-b states */
-	OTG_STATE_B_WAIT_ACON,
-	OTG_STATE_B_HOST,
-
-	/* dual-role default-a */
-	OTG_STATE_A_IDLE,
-	OTG_STATE_A_WAIT_VRISE,
-	OTG_STATE_A_WAIT_BCON,
-	OTG_STATE_A_HOST,
-	OTG_STATE_A_SUSPEND,
-	OTG_STATE_A_PERIPHERAL,
-	OTG_STATE_A_WAIT_VFALL,
-	OTG_STATE_A_VBUS_ERR,
-};
-
-/*
- * the otg driver needs to interact with both device side and host side
- * usb controllers.  it decides which controller is active at a given
- * moment, using the transceiver, ID signal, HNP and sometimes static
- * configuration information (including "board isn't wired for otg").
- */
-struct otg_transceiver {
-	struct device		*dev;
-	const char		*label;
-
-	u8			default_a;
-	enum usb_otg_state	state;
-
-	struct usb_bus		*host;
-	struct usb_gadget	*gadget;
-
-	/* to pass extra port status to the root hub */
-	u16			port_status;
-	u16			port_change;
-
-	/* bind/unbind the host controller */
-	int 	(*set_host)(struct otg_transceiver *otg,
-				struct usb_bus *host);
-
-	/* bind/unbind the peripheral controller */
-	int	(*set_peripheral)(struct otg_transceiver *otg,
-				struct usb_gadget *gadget);
-
-	/* effective for B devices, ignored for A-peripheral */
-	int	(*set_power)(struct otg_transceiver *otg,
-				unsigned mA);
-
-	/* for non-OTG B devices: set transceiver into suspend mode */
-	int	(*set_suspend)(struct otg_transceiver *otg,
-				int suspend);
-
-	/* for B devices only:  start session with A-Host */
-	int	(*start_srp)(struct otg_transceiver *otg);
-
-	/* start or continue HNP role switch */
-	int	(*start_hnp)(struct otg_transceiver *otg);
-
-};
-
-
-/* for board-specific init logic */
-extern int otg_set_transceiver(struct otg_transceiver *);
-
-
-/* for usb host and peripheral controller drivers */
-extern struct otg_transceiver *otg_get_transceiver(void);
-
-static inline int
-otg_start_hnp(struct otg_transceiver *otg)
-{
-	return otg->start_hnp(otg);
-}
-
-
-/* for HCDs */
-static inline int
-otg_set_host(struct otg_transceiver *otg, struct usb_bus *host)
-{
-	return otg->set_host(otg, host);
-}
-
-
-/* for usb peripheral controller drivers */
-static inline int
-otg_set_peripheral(struct otg_transceiver *otg, struct usb_gadget *periph)
-{
-	return otg->set_peripheral(otg, periph);
-}
-
-static inline int
-otg_set_power(struct otg_transceiver *otg, unsigned mA)
-{
-	return otg->set_power(otg, mA);
-}
-
-static inline int
-otg_set_suspend(struct otg_transceiver *otg, int suspend)
-{
-	if (otg->set_suspend != NULL)
-		return otg->set_suspend(otg, suspend);
-	else
-		return 0;
-}
-
-static inline int
-otg_start_srp(struct otg_transceiver *otg)
-{
-	return otg->start_srp(otg);
-}
-
-
-/* for OTG controller drivers (and maybe other stuff) */
-extern int usb_bus_start_enum(struct usb_bus *bus, unsigned port_num);
-- 
cgit v1.2.3


From b2bbb20b37d734443d1c279d0033a64f6095db54 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 29 Jun 2006 12:25:39 -0700
Subject: USB: pxa2xx_udc understands GPIO based VBUS sensing

This updates the PXA 25x UDC board-independent infrastructure for VBUS sensing
and the D+ pullup.  The original code evolved from rather bizarre support on
Intel's "Lubbock" reference hardware, so that on more sensible hardware it
doesn't work as well as it could/should.

The change is just to teach the UDC driver how to use built-in PXA GPIO pins
directly.  This reduces the amount of board-specfic object code needed, and
enables the use of a VBUS sensing IRQ on boards (like Gumstix) that have one.
With VBUS sensing, the UDC is unclocked until a host is actually connected.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/mach-pxa/corgi.c       | 15 +--------
 drivers/usb/gadget/pxa2xx_udc.c | 70 +++++++++++++++++++++++++++++++++++++++--
 drivers/usb/gadget/pxa2xx_udc.h | 24 +++++---------
 include/asm-arm/arch-pxa/udc.h  |  8 +++++
 4 files changed, 85 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-pxa/corgi.c b/arch/arm/mach-pxa/corgi.c
index cce26576999e..337c01c4ac37 100644
--- a/arch/arm/mach-pxa/corgi.c
+++ b/arch/arm/mach-pxa/corgi.c
@@ -284,21 +284,9 @@ static struct pxaficp_platform_data corgi_ficp_platform_data = {
 /*
  * USB Device Controller
  */
-static void corgi_udc_command(int cmd)
-{
-	switch(cmd)	{
-	case PXA2XX_UDC_CMD_CONNECT:
-		GPSR(CORGI_GPIO_USB_PULLUP) = GPIO_bit(CORGI_GPIO_USB_PULLUP);
-		break;
-	case PXA2XX_UDC_CMD_DISCONNECT:
-		GPCR(CORGI_GPIO_USB_PULLUP) = GPIO_bit(CORGI_GPIO_USB_PULLUP);
-		break;
-	}
-}
-
 static struct pxa2xx_udc_mach_info udc_info __initdata = {
 	/* no connect GPIO; corgi can't tell connection status */
-	.udc_command		= corgi_udc_command,
+	.gpio_pullup		= CORGI_GPIO_USB_PULLUP,
 };
 
 
@@ -350,7 +338,6 @@ static void __init corgi_init(void)
 	corgi_ssp_set_machinfo(&corgi_ssp_machinfo);
 
 	pxa_gpio_mode(CORGI_GPIO_IR_ON | GPIO_OUT);
-	pxa_gpio_mode(CORGI_GPIO_USB_PULLUP | GPIO_OUT);
 	pxa_gpio_mode(CORGI_GPIO_HSYNC | GPIO_IN);
 
  	pxa_set_udc_info(&udc_info);
diff --git a/drivers/usb/gadget/pxa2xx_udc.c b/drivers/usb/gadget/pxa2xx_udc.c
index fff027d30a09..f1adcf8b2023 100644
--- a/drivers/usb/gadget/pxa2xx_udc.c
+++ b/drivers/usb/gadget/pxa2xx_udc.c
@@ -150,6 +150,39 @@ MODULE_PARM_DESC (fifo_mode, "pxa2xx udc fifo mode");
 static void pxa2xx_ep_fifo_flush (struct usb_ep *ep);
 static void nuke (struct pxa2xx_ep *, int status);
 
+/* one GPIO should be used to detect VBUS from the host */
+static int is_vbus_present(void)
+{
+	struct pxa2xx_udc_mach_info		*mach = the_controller->mach;
+
+	if (mach->gpio_vbus)
+		return pxa_gpio_get(mach->gpio_vbus);
+	if (mach->udc_is_connected)
+		return mach->udc_is_connected();
+	return 1;
+}
+
+/* one GPIO should control a D+ pullup, so host sees this device (or not) */
+static void pullup_off(void)
+{
+	struct pxa2xx_udc_mach_info		*mach = the_controller->mach;
+
+	if (mach->gpio_pullup)
+		pxa_gpio_set(mach->gpio_pullup, 0);
+	else if (mach->udc_command)
+		mach->udc_command(PXA2XX_UDC_CMD_DISCONNECT);
+}
+
+static void pullup_on(void)
+{
+	struct pxa2xx_udc_mach_info		*mach = the_controller->mach;
+
+	if (mach->gpio_pullup)
+		pxa_gpio_set(mach->gpio_pullup, 1);
+	else if (mach->udc_command)
+		mach->udc_command(PXA2XX_UDC_CMD_CONNECT);
+}
+
 static void pio_irq_enable(int bEndpointAddress)
 {
         bEndpointAddress &= 0xf;
@@ -1721,6 +1754,16 @@ lubbock_vbus_irq(int irq, void *_dev, struct pt_regs *r)
 
 #endif
 
+static irqreturn_t
+udc_vbus_irq(int irq, void *_dev, struct pt_regs *r)
+{
+	struct pxa2xx_udc	*dev = _dev;
+	int			vbus = pxa_gpio_get(dev->mach->gpio_vbus);
+
+	pxa2xx_udc_vbus_session(&dev->gadget, vbus);
+	return IRQ_HANDLED;
+}
+
 
 /*-------------------------------------------------------------------------*/
 
@@ -2438,7 +2481,7 @@ static struct pxa2xx_udc memory = {
 static int __init pxa2xx_udc_probe(struct platform_device *pdev)
 {
 	struct pxa2xx_udc *dev = &memory;
-	int retval, out_dma = 1;
+	int retval, out_dma = 1, vbus_irq;
 	u32 chiprev;
 
 	/* insist on Intel/ARM/XScale */
@@ -2502,6 +2545,16 @@ static int __init pxa2xx_udc_probe(struct platform_device *pdev)
 	/* other non-static parts of init */
 	dev->dev = &pdev->dev;
 	dev->mach = pdev->dev.platform_data;
+	if (dev->mach->gpio_vbus) {
+		vbus_irq = IRQ_GPIO(dev->mach->gpio_vbus & GPIO_MD_MASK_NR);
+		pxa_gpio_mode((dev->mach->gpio_vbus & GPIO_MD_MASK_NR)
+				| GPIO_IN);
+		set_irq_type(vbus_irq, IRQT_BOTHEDGE);
+	} else
+		vbus_irq = 0;
+	if (dev->mach->gpio_pullup)
+		pxa_gpio_mode((dev->mach->gpio_pullup & GPIO_MD_MASK_NR)
+				| GPIO_OUT | GPIO_DFLT_LOW);
 
 	init_timer(&dev->timer);
 	dev->timer.function = udc_watchdog;
@@ -2557,8 +2610,19 @@ lubbock_fail0:
 		HEX_DISPLAY(dev->stats.irqs);
 		LUB_DISC_BLNK_LED &= 0xff;
 #endif
-	}
+	} else
 #endif
+	if (vbus_irq) {
+		retval = request_irq(vbus_irq, udc_vbus_irq,
+				SA_INTERRUPT | SA_SAMPLE_RANDOM,
+				driver_name, dev);
+		if (retval != 0) {
+			printk(KERN_ERR "%s: can't get irq %i, err %d\n",
+				driver_name, vbus_irq, retval);
+			free_irq(IRQ_USB, dev);
+			return -EBUSY;
+		}
+	}
 	create_proc_files();
 
 	return 0;
@@ -2587,6 +2651,8 @@ static int __exit pxa2xx_udc_remove(struct platform_device *pdev)
 		free_irq(LUBBOCK_USB_IRQ, dev);
 	}
 #endif
+	if (dev->mach->gpio_vbus)
+		free_irq(IRQ_GPIO(dev->mach->gpio_vbus), dev);
 	platform_set_drvdata(pdev, NULL);
 	the_controller = NULL;
 	return 0;
diff --git a/drivers/usb/gadget/pxa2xx_udc.h b/drivers/usb/gadget/pxa2xx_udc.h
index 19a883f7d1b8..8e598c8bf4e3 100644
--- a/drivers/usb/gadget/pxa2xx_udc.h
+++ b/drivers/usb/gadget/pxa2xx_udc.h
@@ -177,27 +177,19 @@ struct pxa2xx_udc {
 
 static struct pxa2xx_udc *the_controller;
 
-/* one GPIO should be used to detect VBUS from the host */
-static inline int is_vbus_present(void)
+static inline int pxa_gpio_get(unsigned gpio)
 {
-	if (!the_controller->mach->udc_is_connected)
-		return 1;
-	return the_controller->mach->udc_is_connected();
+	return (GPLR(gpio) & GPIO_bit(gpio)) != 0;
 }
 
-/* one GPIO should control a D+ pullup, so host sees this device (or not) */
-static inline void pullup_off(void)
+static inline void pxa_gpio_set(unsigned gpio, int is_on)
 {
-	if (!the_controller->mach->udc_command)
-		return;
-	the_controller->mach->udc_command(PXA2XX_UDC_CMD_DISCONNECT);
-}
+	int mask = GPIO_bit(gpio);
 
-static inline void pullup_on(void)
-{
-	if (!the_controller->mach->udc_command)
-		return;
-	the_controller->mach->udc_command(PXA2XX_UDC_CMD_CONNECT);
+	if (is_on)
+		GPSR(gpio) = mask;
+	else
+		GPCR(gpio) = mask;
 }
 
 /*-------------------------------------------------------------------------*/
diff --git a/include/asm-arm/arch-pxa/udc.h b/include/asm-arm/arch-pxa/udc.h
index 30548a30c773..121cd241115d 100644
--- a/include/asm-arm/arch-pxa/udc.h
+++ b/include/asm-arm/arch-pxa/udc.h
@@ -12,6 +12,14 @@ struct pxa2xx_udc_mach_info {
         void (*udc_command)(int cmd);
 #define	PXA2XX_UDC_CMD_CONNECT		0	/* let host see us */
 #define	PXA2XX_UDC_CMD_DISCONNECT	1	/* so host won't see us */
+
+	/* Boards following the design guidelines in the developer's manual,
+	 * with on-chip GPIOs not Lubbock's wierd hardware, can have a sane
+	 * VBUS IRQ and omit the methods above.  Store the GPIO number
+	 * here; for GPIO 0, also mask in one of the pxa_gpio_mode() bits.
+	 */
+	u16	gpio_vbus;			/* high == vbus present */
+	u16	gpio_pullup;			/* high == pullup activated */
 };
 
 extern void pxa_set_udc_info(struct pxa2xx_udc_mach_info *info);
-- 
cgit v1.2.3


From 8bb54ab573ecd1b4fe2ed66416a8d99a86e65316 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:08:49 -0400
Subject: usbcore: add usb_device_driver definition

This patch (as732) adds a usb_device_driver structure, for representing
drivers that manage an entire USB device as opposed to just an
interface.  Support routines like usb_register_device_driver,
usb_deregister_device_driver, usb_probe_device, and usb_unbind_device
are also added.

Unlike an earlier version of this patch, the new code is type-safe.  To
accomplish this, the existing struct driver embedded in struct
usb_driver had to be wrapped in an intermediate wrapper.  This enables
the core to tell at runtime whether a particular struct driver belongs
to a device driver or to an interface driver.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c  | 219 +++++++++++++++++++++++++++++++++------------
 drivers/usb/core/generic.c |  17 +---
 drivers/usb/core/usb.c     |  15 ++--
 drivers/usb/core/usb.h     |  20 ++++-
 include/linux/usb.h        |  58 ++++++++++--
 5 files changed, 245 insertions(+), 84 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 8dcf2cd0c569..0d4b5dcee3ab 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -84,7 +84,7 @@ static int usb_create_newid_file(struct usb_driver *usb_drv)
 		goto exit;
 
 	if (usb_drv->probe != NULL)
-		error = sysfs_create_file(&usb_drv->driver.kobj,
+		error = sysfs_create_file(&usb_drv->drvwrap.driver.kobj,
 					  &driver_attr_new_id.attr);
 exit:
 	return error;
@@ -96,7 +96,7 @@ static void usb_remove_newid_file(struct usb_driver *usb_drv)
 		return;
 
 	if (usb_drv->probe != NULL)
-		sysfs_remove_file(&usb_drv->driver.kobj,
+		sysfs_remove_file(&usb_drv->drvwrap.driver.kobj,
 				  &driver_attr_new_id.attr);
 }
 
@@ -143,18 +143,55 @@ static const struct usb_device_id *usb_match_dynamic_id(struct usb_interface *in
 }
 
 
-/* called from driver core with usb_bus_type.subsys writelock */
+/* called from driver core with dev locked */
+static int usb_probe_device(struct device *dev)
+{
+	struct usb_device_driver *udriver = to_usb_device_driver(dev->driver);
+	struct usb_device *udev;
+	int error = -ENODEV;
+
+	dev_dbg(dev, "%s\n", __FUNCTION__);
+
+	if (!is_usb_device(dev))	/* Sanity check */
+		return error;
+
+	udev = to_usb_device(dev);
+
+	/* FIXME: resume a suspended device */
+	if (udev->state == USB_STATE_SUSPENDED)
+		return -EHOSTUNREACH;
+
+	/* TODO: Add real matching code */
+
+	error = udriver->probe(udev);
+	return error;
+}
+
+/* called from driver core with dev locked */
+static int usb_unbind_device(struct device *dev)
+{
+	struct usb_device_driver *udriver = to_usb_device_driver(dev->driver);
+
+	udriver->disconnect(to_usb_device(dev));
+	return 0;
+}
+
+
+/* called from driver core with dev locked */
 static int usb_probe_interface(struct device *dev)
 {
-	struct usb_interface * intf = to_usb_interface(dev);
-	struct usb_driver * driver = to_usb_driver(dev->driver);
+	struct usb_driver *driver = to_usb_driver(dev->driver);
+	struct usb_interface *intf;
 	const struct usb_device_id *id;
 	int error = -ENODEV;
 
 	dev_dbg(dev, "%s\n", __FUNCTION__);
 
-	if (!driver->probe)
+	if (is_usb_device(dev))		/* Sanity check */
 		return error;
+
+	intf = to_usb_interface(dev);
+
 	/* FIXME we'd much prefer to just resume it ... */
 	if (interface_to_usbdev(intf)->state == USB_STATE_SUSPENDED)
 		return -EHOSTUNREACH;
@@ -182,19 +219,18 @@ static int usb_probe_interface(struct device *dev)
 	return error;
 }
 
-/* called from driver core with usb_bus_type.subsys writelock */
+/* called from driver core with dev locked */
 static int usb_unbind_interface(struct device *dev)
 {
+	struct usb_driver *driver = to_usb_driver(dev->driver);
 	struct usb_interface *intf = to_usb_interface(dev);
-	struct usb_driver *driver = to_usb_driver(intf->dev.driver);
 
 	intf->condition = USB_INTERFACE_UNBINDING;
 
 	/* release all urbs for this interface */
 	usb_disable_interface(interface_to_usbdev(intf), intf);
 
-	if (driver && driver->disconnect)
-		driver->disconnect(intf);
+	driver->disconnect(intf);
 
 	/* reset other interface state */
 	usb_set_interface(interface_to_usbdev(intf),
@@ -235,7 +271,7 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 	if (dev->driver)
 		return -EBUSY;
 
-	dev->driver = &driver->driver;
+	dev->driver = &driver->drvwrap.driver;
 	usb_set_intfdata(iface, priv);
 	iface->condition = USB_INTERFACE_BOUND;
 	mark_active(iface);
@@ -270,7 +306,7 @@ void usb_driver_release_interface(struct usb_driver *driver,
 	struct device *dev = &iface->dev;
 
 	/* this should never happen, don't release something that's not ours */
-	if (!dev->driver || dev->driver != &driver->driver)
+	if (!dev->driver || dev->driver != &driver->drvwrap.driver)
 		return;
 
 	/* don't release from within disconnect() */
@@ -433,24 +469,37 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_match_id);
 
 int usb_device_match(struct device *dev, struct device_driver *drv)
 {
-	struct usb_interface *intf;
-	struct usb_driver *usb_drv;
-	const struct usb_device_id *id;
-
-	/* check for generic driver, which we don't match any device with */
-	if (drv == &usb_generic_driver)
-		return 0;
+	/* devices and interfaces are handled separately */
+	if (is_usb_device(dev)) {
 
-	intf = to_usb_interface(dev);
-	usb_drv = to_usb_driver(drv);
+		/* interface drivers never match devices */
+		if (!is_usb_device_driver(drv))
+			return 0;
 
-	id = usb_match_id(intf, usb_drv->id_table);
-	if (id)
+		/* TODO: Add real matching code */
 		return 1;
 
-	id = usb_match_dynamic_id(intf, usb_drv);
-	if (id)
-		return 1;
+	} else {
+		struct usb_interface *intf;
+		struct usb_driver *usb_drv;
+		const struct usb_device_id *id;
+
+		/* device drivers never match interfaces */
+		if (is_usb_device_driver(drv))
+			return 0;
+
+		intf = to_usb_interface(dev);
+		usb_drv = to_usb_driver(drv);
+
+		id = usb_match_id(intf, usb_drv->id_table);
+		if (id)
+			return 1;
+
+		id = usb_match_dynamic_id(intf, usb_drv);
+		if (id)
+			return 1;
+	}
+
 	return 0;
 }
 
@@ -481,14 +530,13 @@ static int usb_uevent(struct device *dev, char **envp, int num_envp,
 	/* driver is often null here; dev_dbg() would oops */
 	pr_debug ("usb %s: uevent\n", dev->bus_id);
 
-	/* Must check driver_data here, as on remove driver is always NULL */
-	if ((dev->driver == &usb_generic_driver) ||
-	    (dev->driver_data == &usb_generic_driver_data))
+	if (is_usb_device(dev))
 		return 0;
-
-	intf = to_usb_interface(dev);
-	usb_dev = interface_to_usbdev (intf);
-	alt = intf->cur_altsetting;
+	else {
+		intf = to_usb_interface(dev);
+		usb_dev = interface_to_usbdev(intf);
+		alt = intf->cur_altsetting;
+	}
 
 	if (usb_dev->devnum < 0) {
 		pr_debug ("usb %s: already deleted?\n", dev->bus_id);
@@ -569,13 +617,71 @@ static int usb_uevent(struct device *dev, char **envp,
 #endif	/* CONFIG_HOTPLUG */
 
 /**
- * usb_register_driver - register a USB driver
- * @new_driver: USB operations for the driver
+ * usb_register_device_driver - register a USB device (not interface) driver
+ * @new_udriver: USB operations for the device driver
  * @owner: module owner of this driver.
  *
- * Registers a USB driver with the USB core.  The list of unattached
- * interfaces will be rescanned whenever a new driver is added, allowing
- * the new driver to attach to any recognized devices.
+ * Registers a USB device driver with the USB core.  The list of
+ * unattached devices will be rescanned whenever a new driver is
+ * added, allowing the new driver to attach to any recognized devices.
+ * Returns a negative error code on failure and 0 on success.
+ */
+int usb_register_device_driver(struct usb_device_driver *new_udriver,
+		struct module *owner)
+{
+	int retval = 0;
+
+	if (usb_disabled())
+		return -ENODEV;
+
+	new_udriver->drvwrap.for_devices = 1;
+	new_udriver->drvwrap.driver.name = (char *) new_udriver->name;
+	new_udriver->drvwrap.driver.bus = &usb_bus_type;
+	new_udriver->drvwrap.driver.probe = usb_probe_device;
+	new_udriver->drvwrap.driver.remove = usb_unbind_device;
+	new_udriver->drvwrap.driver.owner = owner;
+
+	retval = driver_register(&new_udriver->drvwrap.driver);
+
+	if (!retval) {
+		pr_info("%s: registered new device driver %s\n",
+			usbcore_name, new_udriver->name);
+		usbfs_update_special();
+	} else {
+		printk(KERN_ERR "%s: error %d registering device "
+			"	driver %s\n",
+			usbcore_name, retval, new_udriver->name);
+	}
+
+	return retval;
+}
+EXPORT_SYMBOL_GPL(usb_register_device_driver);
+
+/**
+ * usb_deregister_device_driver - unregister a USB device (not interface) driver
+ * @udriver: USB operations of the device driver to unregister
+ * Context: must be able to sleep
+ *
+ * Unlinks the specified driver from the internal USB driver list.
+ */
+void usb_deregister_device_driver(struct usb_device_driver *udriver)
+{
+	pr_info("%s: deregistering device driver %s\n",
+			usbcore_name, udriver->name);
+
+	driver_unregister(&udriver->drvwrap.driver);
+	usbfs_update_special();
+}
+EXPORT_SYMBOL_GPL(usb_deregister_device_driver);
+
+/**
+ * usb_register_driver - register a USB interface driver
+ * @new_driver: USB operations for the interface driver
+ * @owner: module owner of this driver.
+ *
+ * Registers a USB interface driver with the USB core.  The list of
+ * unattached interfaces will be rescanned whenever a new driver is
+ * added, allowing the new driver to attach to any recognized interfaces.
  * Returns a negative error code on failure and 0 on success.
  *
  * NOTE: if you want your driver to use the USB major number, you must call
@@ -589,23 +695,25 @@ int usb_register_driver(struct usb_driver *new_driver, struct module *owner)
 	if (usb_disabled())
 		return -ENODEV;
 
-	new_driver->driver.name = (char *)new_driver->name;
-	new_driver->driver.bus = &usb_bus_type;
-	new_driver->driver.probe = usb_probe_interface;
-	new_driver->driver.remove = usb_unbind_interface;
-	new_driver->driver.owner = owner;
+	new_driver->drvwrap.for_devices = 0;
+	new_driver->drvwrap.driver.name = (char *) new_driver->name;
+	new_driver->drvwrap.driver.bus = &usb_bus_type;
+	new_driver->drvwrap.driver.probe = usb_probe_interface;
+	new_driver->drvwrap.driver.remove = usb_unbind_interface;
+	new_driver->drvwrap.driver.owner = owner;
 	spin_lock_init(&new_driver->dynids.lock);
 	INIT_LIST_HEAD(&new_driver->dynids.list);
 
-	retval = driver_register(&new_driver->driver);
+	retval = driver_register(&new_driver->drvwrap.driver);
 
 	if (!retval) {
-		pr_info("%s: registered new driver %s\n",
+		pr_info("%s: registered new interface driver %s\n",
 			usbcore_name, new_driver->name);
 		usbfs_update_special();
 		usb_create_newid_file(new_driver);
 	} else {
-		printk(KERN_ERR "%s: error %d registering driver %s\n",
+		printk(KERN_ERR "%s: error %d registering interface "
+			"	driver %s\n",
 			usbcore_name, retval, new_driver->name);
 	}
 
@@ -614,8 +722,8 @@ int usb_register_driver(struct usb_driver *new_driver, struct module *owner)
 EXPORT_SYMBOL_GPL_FUTURE(usb_register_driver);
 
 /**
- * usb_deregister - unregister a USB driver
- * @driver: USB operations of the driver to unregister
+ * usb_deregister - unregister a USB interface driver
+ * @driver: USB operations of the interface driver to unregister
  * Context: must be able to sleep
  *
  * Unlinks the specified driver from the internal USB driver list.
@@ -626,11 +734,12 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_register_driver);
  */
 void usb_deregister(struct usb_driver *driver)
 {
-	pr_info("%s: deregistering driver %s\n", usbcore_name, driver->name);
+	pr_info("%s: deregistering interface driver %s\n",
+			usbcore_name, driver->name);
 
 	usb_remove_newid_file(driver);
 	usb_free_dynids(driver);
-	driver_unregister(&driver->driver);
+	driver_unregister(&driver->drvwrap.driver);
 
 	usbfs_update_special();
 }
@@ -655,7 +764,7 @@ static int usb_generic_suspend(struct device *dev, pm_message_t message)
 	 * marked for FREEZE as soon as their children are already idled.
 	 * But those semantics are useless, so we equate the two (sigh).
 	 */
-	if (dev->driver == &usb_generic_driver) {
+	if (is_usb_device(dev)) {
 		if (dev->power.power_state.event == message.event)
 			return 0;
 		/* we need to rule out bogus requests through sysfs */
@@ -665,8 +774,7 @@ static int usb_generic_suspend(struct device *dev, pm_message_t message)
  		return usb_port_suspend(to_usb_device(dev));
 	}
 
-	if ((dev->driver == NULL) ||
-	    (dev->driver_data == &usb_generic_driver_data))
+	if (dev->driver == NULL)
 		return 0;
 
 	intf = to_usb_interface(dev);
@@ -705,15 +813,14 @@ static int usb_generic_resume(struct device *dev)
 	dev->power.power_state.event = PM_EVENT_ON;
 
 	/* devices resume through their hubs */
-	if (dev->driver == &usb_generic_driver) {
+	if (is_usb_device(dev)) {
 		udev = to_usb_device(dev);
 		if (udev->state == USB_STATE_NOTATTACHED)
 			return 0;
 		return usb_port_resume(udev);
 	}
 
-	if ((dev->driver == NULL) ||
-	    (dev->driver_data == &usb_generic_driver_data)) {
+	if (dev->driver == NULL) {
 		dev->power.power_state.event = PM_EVENT_FREEZE;
 		return 0;
 	}
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index 7bab9769b34f..fa6f34a12b4b 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -21,14 +21,12 @@
 #include <linux/usb.h>
 #include "usb.h"
 
-static int generic_probe(struct device *dev)
+static int generic_probe(struct usb_device *udev)
 {
 	return 0;
 }
-static int generic_remove(struct device *dev)
+static void generic_disconnect(struct usb_device *udev)
 {
-	struct usb_device *udev = to_usb_device(dev);
-
 	/* if this is only an unbind, not a physical disconnect, then
 	 * unconfigure the device */
 	if (udev->state == USB_STATE_CONFIGURED)
@@ -37,17 +35,10 @@ static int generic_remove(struct device *dev)
 	/* in case the call failed or the device was suspended */
 	if (udev->state >= USB_STATE_CONFIGURED)
 		usb_disable_device(udev, 0);
-	return 0;
 }
 
-struct device_driver usb_generic_driver = {
-	.owner = THIS_MODULE,
+struct usb_device_driver usb_generic_driver = {
 	.name =	"usb",
-	.bus = &usb_bus_type,
 	.probe = generic_probe,
-	.remove = generic_remove,
+	.disconnect = generic_disconnect,
 };
-
-/* Fun hack to determine if the struct device is a
- * usb device or a usb interface. */
-int usb_generic_driver_data;
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 0b8c67bcde60..6dfbc284369b 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -123,7 +123,7 @@ static int __find_interface(struct device * dev, void * data)
 	struct usb_interface *intf;
 
 	/* can't look at usb devices, only interfaces */
-	if (dev->driver == &usb_generic_driver)
+	if (is_usb_device(dev))
 		return 0;
 
 	intf = to_usb_interface(dev);
@@ -149,7 +149,8 @@ struct usb_interface *usb_find_interface(struct usb_driver *drv, int minor)
 
 	argb.minor = minor;
 	argb.interface = NULL;
-	driver_for_each_device(&drv->driver, NULL, &argb, __find_interface);
+	driver_for_each_device(&drv->drvwrap.driver, NULL, &argb,
+			__find_interface);
 	return argb.interface;
 }
 
@@ -204,11 +205,13 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	device_initialize(&dev->dev);
 	dev->dev.bus = &usb_bus_type;
 	dev->dev.dma_mask = bus->controller->dma_mask;
-	dev->dev.driver_data = &usb_generic_driver_data;
-	dev->dev.driver = &usb_generic_driver;
+	dev->dev.driver = &usb_generic_driver.drvwrap.driver;
 	dev->dev.release = usb_release_dev;
 	dev->state = USB_STATE_ATTACHED;
 
+	/* This magic assignment distinguishes devices from interfaces */
+	dev->dev.platform_data = &usb_generic_driver;
+
 	INIT_LIST_HEAD(&dev->ep0.urb_list);
 	dev->ep0.desc.bLength = USB_DT_ENDPOINT_SIZE;
 	dev->ep0.desc.bDescriptorType = USB_DT_ENDPOINT;
@@ -838,7 +841,7 @@ static int __init usb_init(void)
 	retval = usb_hub_init();
 	if (retval)
 		goto hub_init_failed;
-	retval = driver_register(&usb_generic_driver);
+	retval = usb_register_device_driver(&usb_generic_driver, THIS_MODULE);
 	if (!retval)
 		goto out;
 
@@ -868,7 +871,7 @@ static void __exit usb_exit(void)
 	if (nousb)
 		return;
 
-	driver_unregister(&usb_generic_driver);
+	usb_deregister_device_driver(&usb_generic_driver);
 	usb_major_cleanup();
 	usbfs_cleanup();
 	usb_deregister(&usbfs_driver);
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 82d397a6f773..1d25ccac7832 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -34,8 +34,24 @@ extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 
 extern struct bus_type usb_bus_type;
-extern struct device_driver usb_generic_driver;
-extern int usb_generic_driver_data;
+extern struct usb_device_driver usb_generic_driver;
+
+/* Here's how we tell apart devices and interfaces.  Luckily there's
+ * no such thing as a platform USB device, so we can steal the use
+ * of the platform_data field. */
+
+static inline int is_usb_device(struct device *dev)
+{
+	return dev->platform_data == &usb_generic_driver;
+}
+
+/* Do the same for device drivers and interface drivers. */
+
+static inline int is_usb_device_driver(struct device_driver *drv)
+{
+	return container_of(drv, struct usbdrv_wrap, driver)->
+			for_devices;
+}
 
 /* Interfaces and their "power state" are owned by usbcore */
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index d2bd0c8e0154..b4ccce6d0982 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -540,7 +540,17 @@ struct usb_dynids {
 };
 
 /**
- * struct usb_driver - identifies USB driver to usbcore
+ * struct usbdrv_wrap - wrapper for driver-model structure
+ * @driver: The driver-model core driver structure.
+ * @for_devices: Non-zero for device drivers, 0 for interface drivers.
+ */
+struct usbdrv_wrap {
+	struct device_driver driver;
+	int for_devices;
+};
+
+/**
+ * struct usb_driver - identifies USB interface driver to usbcore
  * @name: The driver name should be unique among USB drivers,
  *	and should normally be the same as the module name.
  * @probe: Called to see if the driver is willing to manage a particular
@@ -567,12 +577,12 @@ struct usb_dynids {
  *	or your driver's probe function will never get called.
  * @dynids: used internally to hold the list of dynamically added device
  *	ids for this driver.
- * @driver: the driver model core driver structure.
+ * @drvwrap: Driver-model core structure wrapper.
  * @no_dynamic_id: if set to 1, the USB core will not allow dynamic ids to be
  *	added to this driver by preventing the sysfs file from being created.
  *
- * USB drivers must provide a name, probe() and disconnect() methods,
- * and an id_table.  Other driver fields are optional.
+ * USB interface drivers must provide a name, probe() and disconnect()
+ * methods, and an id_table.  Other driver fields are optional.
  *
  * The id_table is used in hotplugging.  It holds a set of descriptors,
  * and specialized data may be associated with each entry.  That table
@@ -606,10 +616,40 @@ struct usb_driver {
 	const struct usb_device_id *id_table;
 
 	struct usb_dynids dynids;
-	struct device_driver driver;
+	struct usbdrv_wrap drvwrap;
 	unsigned int no_dynamic_id:1;
 };
-#define	to_usb_driver(d) container_of(d, struct usb_driver, driver)
+#define	to_usb_driver(d) container_of(d, struct usb_driver, drvwrap.driver)
+
+/**
+ * struct usb_device_driver - identifies USB device driver to usbcore
+ * @name: The driver name should be unique among USB drivers,
+ *	and should normally be the same as the module name.
+ * @probe: Called to see if the driver is willing to manage a particular
+ *	device.  If it is, probe returns zero and uses dev_set_drvdata()
+ *	to associate driver-specific data with the device.  If unwilling
+ *	to manage the device, return a negative errno value.
+ * @disconnect: Called when the device is no longer accessible, usually
+ *	because it has been (or is being) disconnected or the driver's
+ *	module is being unloaded.
+ * @suspend: Called when the device is going to be suspended by the system.
+ * @resume: Called when the device is being resumed by the system.
+ * @drvwrap: Driver-model core structure wrapper.
+ *
+ * USB drivers must provide all the fields listed above except drvwrap.
+ */
+struct usb_device_driver {
+	const char *name;
+
+	int (*probe) (struct usb_device *udev);
+	void (*disconnect) (struct usb_device *udev);
+
+	int (*suspend) (struct usb_device *udev, pm_message_t message);
+	int (*resume) (struct usb_device *udev);
+	struct usbdrv_wrap drvwrap;
+};
+#define	to_usb_device_driver(d) container_of(d, struct usb_device_driver, \
+		drvwrap.driver)
 
 extern struct bus_type usb_bus_type;
 
@@ -633,13 +673,17 @@ struct usb_class_driver {
  * use these in module_init()/module_exit()
  * and don't forget MODULE_DEVICE_TABLE(usb, ...)
  */
-int usb_register_driver(struct usb_driver *, struct module *);
+extern int usb_register_driver(struct usb_driver *, struct module *);
 static inline int usb_register(struct usb_driver *driver)
 {
 	return usb_register_driver(driver, THIS_MODULE);
 }
 extern void usb_deregister(struct usb_driver *);
 
+extern int usb_register_device_driver(struct usb_device_driver *,
+			struct module *);
+extern void usb_deregister_device_driver(struct usb_device_driver *);
+
 extern int usb_register_dev(struct usb_interface *intf,
 			    struct usb_class_driver *class_driver);
 extern void usb_deregister_dev(struct usb_interface *intf,
-- 
cgit v1.2.3


From 4d064c080265a41324d108fccc26b72106d43db3 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:11:44 -0400
Subject: usbcore: track whether interfaces are suspended

Currently we rely on intf->dev.power.power_state.event for tracking
whether intf is suspended.  This is not a reliable technique because
that value is owned by the PM core, not by usbcore.  This patch (as718b)
adds a new flag so that we can accurately tell which interfaces are
suspended and which aren't.

At first one might think these flags aren't needed, since interfaces
will be suspended along with their devices.  It turns out there are a
couple of intermediate situations where that's not quite true, such as
while processing a remote-wakeup request.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.h | 6 +++---
 include/linux/usb.h    | 3 +++
 2 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index cc42972b6bb0..74df0db954c9 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -59,17 +59,17 @@ static inline int is_usb_device_driver(struct device_driver *drv)
 
 static inline void mark_active(struct usb_interface *f)
 {
-	f->dev.power.power_state.event = PM_EVENT_ON;
+	f->is_active = 1;
 }
 
 static inline void mark_quiesced(struct usb_interface *f)
 {
-	f->dev.power.power_state.event = PM_EVENT_FREEZE;
+	f->is_active = 0;
 }
 
 static inline int is_active(struct usb_interface *f)
 {
-	return f->dev.power.power_state.event == PM_EVENT_ON;
+	return f->is_active;
 }
 
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index b4ccce6d0982..e22f4b386605 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -102,6 +102,7 @@ enum usb_interface_condition {
  *	number from the USB core by calling usb_register_dev().
  * @condition: binding state of the interface: not bound, binding
  *	(in probe()), bound to a driver, or unbinding (in disconnect())
+ * @is_active: flag set when the interface is bound and not suspended.
  * @dev: driver model's view of this device
  * @class_dev: driver model's class view of this device.
  *
@@ -142,6 +143,8 @@ struct usb_interface {
 	int minor;			/* minor number this interface is
 					 * bound to */
 	enum usb_interface_condition condition;		/* state of binding */
+	unsigned is_active:1;		/* the interface is not suspended */
+
 	struct device dev;		/* interface specific device info */
 	struct class_device *class_dev;
 };
-- 
cgit v1.2.3


From f2ebf92c9e1930a8f79b7eb49a32122931929014 Mon Sep 17 00:00:00 2001
From: Ben Williamson <ben.williamson@greyinnovation.com>
Date: Tue, 1 Aug 2006 11:28:16 +1000
Subject: USB: gmidi: New USB MIDI Gadget class driver.

This driver is glue between the USB gadget interface
and the ALSA MIDI interface. It allows us to appear
as a MIDI Streaming device to a host system on the
other end of a USB cable.

This includes linux/usb/audio.h and linux/usb/midi.h
containing definitions from the relevant USB specifications
for USB audio and USB MIDI devices.

The following changes have been made since the first RFC
posting:

* Bug fixes to endpoint handling.
* Workaround for USB_REQ_SET_CONFIGURATION handling,
  not understood yet.
* Added SND and SND_RAWMIDI dependencies in Kconfig.
* Moved usb_audio.h and usb_midi.h to usb/*.h
* Added module parameters for ALSA card index and id.
* Added module parameters for USB descriptor IDs and strings.
* Removed some unneeded stuff inherited from zero.c, more to go.
* Provide DECLARE_* macros for the variable-length structs.
* Use kmalloc instead of usb_ep_alloc_buffer.
* Limit source to 80 columns.
* Return actual error code instead of -ENOMEM in a few places.

Signed-off-by: Ben Williamson <ben.williamson@greyinnovation.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/Kconfig  |   14 +
 drivers/usb/gadget/Makefile |    2 +
 drivers/usb/gadget/gmidi.c  | 1337 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb/audio.h   |   53 ++
 include/linux/usb/midi.h    |  112 ++++
 5 files changed, 1518 insertions(+)
 create mode 100644 drivers/usb/gadget/gmidi.c
 create mode 100644 include/linux/usb/audio.h
 create mode 100644 include/linux/usb/midi.h

(limited to 'include')

diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
index 1a32d96774b4..4301e96c417b 100644
--- a/drivers/usb/gadget/Kconfig
+++ b/drivers/usb/gadget/Kconfig
@@ -404,6 +404,20 @@ config USB_G_SERIAL
 	  which includes instructions and a "driver info file" needed to
 	  make MS-Windows work with this driver.
 
+config USB_MIDI_GADGET
+	tristate "MIDI Gadget (EXPERIMENTAL)"
+	depends on SND && EXPERIMENTAL
+	select SND_RAWMIDI
+	help
+	  The MIDI Gadget acts as a USB Audio device, with one MIDI
+	  input and one MIDI output. These MIDI jacks appear as
+	  a sound "card" in the ALSA sound system. Other MIDI
+	  connections can then be made on the gadget system, using
+	  ALSA's aconnect utility etc.
+
+	  Say "y" to link the driver statically, or "m" to build a
+	  dynamically linked module called "g_midi".
+
 
 # put drivers that need isochronous transfer support (for audio
 # or video class gadget drivers), or specific hardware, here.
diff --git a/drivers/usb/gadget/Makefile b/drivers/usb/gadget/Makefile
index 5a28e61392ec..e71e086a1cfa 100644
--- a/drivers/usb/gadget/Makefile
+++ b/drivers/usb/gadget/Makefile
@@ -15,6 +15,7 @@ obj-$(CONFIG_USB_AT91)		+= at91_udc.o
 g_zero-objs			:= zero.o usbstring.o config.o epautoconf.o
 g_ether-objs			:= ether.o usbstring.o config.o epautoconf.o
 g_serial-objs			:= serial.o usbstring.o config.o epautoconf.o
+g_midi-objs			:= gmidi.o usbstring.o config.o epautoconf.o
 gadgetfs-objs			:= inode.o
 g_file_storage-objs		:= file_storage.o usbstring.o config.o \
 					epautoconf.o
@@ -28,4 +29,5 @@ obj-$(CONFIG_USB_ETH)		+= g_ether.o
 obj-$(CONFIG_USB_GADGETFS)	+= gadgetfs.o
 obj-$(CONFIG_USB_FILE_STORAGE)	+= g_file_storage.o
 obj-$(CONFIG_USB_G_SERIAL)	+= g_serial.o
+obj-$(CONFIG_USB_MIDI_GADGET)	+= g_midi.o
 
diff --git a/drivers/usb/gadget/gmidi.c b/drivers/usb/gadget/gmidi.c
new file mode 100644
index 000000000000..b68cecd57411
--- /dev/null
+++ b/drivers/usb/gadget/gmidi.c
@@ -0,0 +1,1337 @@
+/*
+ * gmidi.c -- USB MIDI Gadget Driver
+ *
+ * Copyright (C) 2006 Thumtronics Pty Ltd.
+ * Developed for Thumtronics by Grey Innovation
+ * Ben Williamson <ben.williamson@greyinnovation.com>
+ *
+ * This software is distributed under the terms of the GNU General Public
+ * License ("GPL") version 2, as published by the Free Software Foundation.
+ *
+ * This code is based in part on:
+ *
+ * Gadget Zero driver, Copyright (C) 2003-2004 David Brownell.
+ * USB Audio driver, Copyright (C) 2002 by Takashi Iwai.
+ * USB MIDI driver, Copyright (C) 2002-2005 Clemens Ladisch.
+ *
+ * Refer to the USB Device Class Definition for MIDI Devices:
+ * http://www.usb.org/developers/devclass_docs/midi10.pdf
+ */
+
+#define DEBUG 1
+// #define VERBOSE
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/utsname.h>
+#include <linux/device.h>
+#include <linux/moduleparam.h>
+
+#include <sound/driver.h>
+#include <sound/core.h>
+#include <sound/initval.h>
+#include <sound/rawmidi.h>
+
+#include <linux/usb_ch9.h>
+#include <linux/usb_gadget.h>
+#include <linux/usb/audio.h>
+#include <linux/usb/midi.h>
+
+#include "gadget_chips.h"
+
+MODULE_AUTHOR("Ben Williamson");
+MODULE_LICENSE("GPL v2");
+
+#define DRIVER_VERSION "25 Jul 2006"
+
+static const char shortname[] = "g_midi";
+static const char longname[] = "MIDI Gadget";
+
+static int index = SNDRV_DEFAULT_IDX1;
+static char *id = SNDRV_DEFAULT_STR1;
+
+module_param(index, int, 0444);
+MODULE_PARM_DESC(index, "Index value for the USB MIDI Gadget adapter.");
+module_param(id, charp, 0444);
+MODULE_PARM_DESC(id, "ID string for the USB MIDI Gadget adapter.");
+
+/* Some systems will want different product identifers published in the
+ * device descriptor, either numbers or strings or both.  These string
+ * parameters are in UTF-8 (superset of ASCII's 7 bit characters).
+ */
+
+static ushort idVendor;
+module_param(idVendor, ushort, S_IRUGO);
+MODULE_PARM_DESC(idVendor, "USB Vendor ID");
+
+static ushort idProduct;
+module_param(idProduct, ushort, S_IRUGO);
+MODULE_PARM_DESC(idProduct, "USB Product ID");
+
+static ushort bcdDevice;
+module_param(bcdDevice, ushort, S_IRUGO);
+MODULE_PARM_DESC(bcdDevice, "USB Device version (BCD)");
+
+static char *iManufacturer;
+module_param(iManufacturer, charp, S_IRUGO);
+MODULE_PARM_DESC(iManufacturer, "USB Manufacturer string");
+
+static char *iProduct;
+module_param(iProduct, charp, S_IRUGO);
+MODULE_PARM_DESC(iProduct, "USB Product string");
+
+static char *iSerialNumber;
+module_param(iSerialNumber, charp, S_IRUGO);
+MODULE_PARM_DESC(iSerialNumber, "SerialNumber");
+
+/*
+ * this version autoconfigures as much as possible,
+ * which is reasonable for most "bulk-only" drivers.
+ */
+static const char *EP_IN_NAME;
+static const char *EP_OUT_NAME;
+
+
+/* big enough to hold our biggest descriptor */
+#define USB_BUFSIZ 256
+
+
+/* This is a gadget, and the IN/OUT naming is from the host's perspective.
+   USB -> OUT endpoint -> rawmidi
+   USB <- IN endpoint  <- rawmidi */
+struct gmidi_in_port {
+	struct gmidi_device* dev;
+	int active;
+	uint8_t cable;		/* cable number << 4 */
+	uint8_t state;
+#define STATE_UNKNOWN	0
+#define STATE_1PARAM	1
+#define STATE_2PARAM_1	2
+#define STATE_2PARAM_2	3
+#define STATE_SYSEX_0	4
+#define STATE_SYSEX_1	5
+#define STATE_SYSEX_2	6
+	uint8_t data[2];
+};
+
+struct gmidi_device {
+	spinlock_t		lock;
+	struct usb_gadget	*gadget;
+	struct usb_request	*req;		/* for control responses */
+	u8			config;
+	struct usb_ep		*in_ep, *out_ep;
+	struct snd_card 	*card;
+	struct snd_rawmidi	*rmidi;
+	struct snd_rawmidi_substream *in_substream;
+	struct snd_rawmidi_substream *out_substream;
+
+	/* For the moment we only support one port in
+	   each direction, but in_port is kept as a
+	   separate struct so we can have more later. */
+	struct gmidi_in_port	in_port;
+	unsigned long		out_triggered;
+	struct tasklet_struct	tasklet;
+};
+
+static void gmidi_transmit(struct gmidi_device* dev, struct usb_request* req);
+
+
+#define xprintk(d,level,fmt,args...) \
+	dev_printk(level , &(d)->gadget->dev , fmt , ## args)
+
+#ifdef DEBUG
+#define DBG(dev,fmt,args...) \
+	xprintk(dev , KERN_DEBUG , fmt , ## args)
+#else
+#define DBG(dev,fmt,args...) \
+	do { } while (0)
+#endif /* DEBUG */
+
+#ifdef VERBOSE
+#define VDBG	DBG
+#else
+#define VDBG(dev,fmt,args...) \
+	do { } while (0)
+#endif /* VERBOSE */
+
+#define ERROR(dev,fmt,args...) \
+	xprintk(dev , KERN_ERR , fmt , ## args)
+#define WARN(dev,fmt,args...) \
+	xprintk(dev , KERN_WARNING , fmt , ## args)
+#define INFO(dev,fmt,args...) \
+	xprintk(dev , KERN_INFO , fmt , ## args)
+
+
+static unsigned buflen = 256;
+static unsigned qlen = 32;
+
+module_param(buflen, uint, S_IRUGO);
+module_param(qlen, uint, S_IRUGO);
+
+
+/* Thanks to Grey Innovation for donating this product ID.
+ *
+ * DO NOT REUSE THESE IDs with a protocol-incompatible driver!!  Ever!!
+ * Instead:  allocate your own, using normal USB-IF procedures.
+ */
+#define DRIVER_VENDOR_NUM	0x17b3		/* Grey Innovation */
+#define DRIVER_PRODUCT_NUM	0x0004		/* Linux-USB "MIDI Gadget" */
+
+
+/*
+ * DESCRIPTORS ... most are static, but strings and (full)
+ * configuration descriptors are built on demand.
+ */
+
+#define STRING_MANUFACTURER	25
+#define STRING_PRODUCT		42
+#define STRING_SERIAL		101
+#define STRING_MIDI_GADGET	250
+
+/* We only have the one configuration, it's number 1. */
+#define	GMIDI_CONFIG		1
+
+/* We have two interfaces- AudioControl and MIDIStreaming */
+#define GMIDI_AC_INTERFACE	0
+#define GMIDI_MS_INTERFACE	1
+#define GMIDI_NUM_INTERFACES	2
+
+DECLARE_USB_AC_HEADER_DESCRIPTOR(1);
+DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(1);
+DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(1);
+
+/* B.1  Device Descriptor */
+static struct usb_device_descriptor device_desc = {
+	.bLength =		USB_DT_DEVICE_SIZE,
+	.bDescriptorType =	USB_DT_DEVICE,
+	.bcdUSB =		__constant_cpu_to_le16(0x0200),
+	.bDeviceClass =		USB_CLASS_PER_INTERFACE,
+	.idVendor =		__constant_cpu_to_le16(DRIVER_VENDOR_NUM),
+	.idProduct =		__constant_cpu_to_le16(DRIVER_PRODUCT_NUM),
+	.iManufacturer =	STRING_MANUFACTURER,
+	.iProduct =		STRING_PRODUCT,
+	.bNumConfigurations =	1,
+};
+
+/* B.2  Configuration Descriptor */
+static struct usb_config_descriptor config_desc = {
+	.bLength =		USB_DT_CONFIG_SIZE,
+	.bDescriptorType =	USB_DT_CONFIG,
+	/* compute wTotalLength on the fly */
+	.bNumInterfaces =	GMIDI_NUM_INTERFACES,
+	.bConfigurationValue =	GMIDI_CONFIG,
+	.iConfiguration =	STRING_MIDI_GADGET,
+	/*
+	 * FIXME: When embedding this driver in a device,
+	 * these need to be set to reflect the actual
+	 * power properties of the device. Is it selfpowered?
+	 */
+	.bmAttributes =		USB_CONFIG_ATT_ONE,
+	.bMaxPower =		1,
+};
+
+/* B.3.1  Standard AC Interface Descriptor */
+static const struct usb_interface_descriptor ac_interface_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bInterfaceNumber =	GMIDI_AC_INTERFACE,
+	.bNumEndpoints =	0,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_AUDIOCONTROL,
+	.iInterface =		STRING_MIDI_GADGET,
+};
+
+/* B.3.2  Class-Specific AC Interface Descriptor */
+static const struct usb_ac_header_descriptor_1 ac_header_desc = {
+	.bLength =		USB_DT_AC_HEADER_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_HEADER,
+	.bcdADC =		__constant_cpu_to_le16(0x0100),
+	.wTotalLength =		USB_DT_AC_HEADER_SIZE(1),
+	.bInCollection =	1,
+	.baInterfaceNr = {
+		[0] =		GMIDI_MS_INTERFACE,
+	}
+};
+
+/* B.4.1  Standard MS Interface Descriptor */
+static const struct usb_interface_descriptor ms_interface_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bInterfaceNumber =	GMIDI_MS_INTERFACE,
+	.bNumEndpoints =	2,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_MIDISTREAMING,
+	.iInterface =		STRING_MIDI_GADGET,
+};
+
+/* B.4.2  Class-Specific MS Interface Descriptor */
+static const struct usb_ms_header_descriptor ms_header_desc = {
+	.bLength =		USB_DT_MS_HEADER_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_HEADER,
+	.bcdMSC =		__constant_cpu_to_le16(0x0100),
+	.wTotalLength =		USB_DT_MS_HEADER_SIZE
+				+ 2*USB_DT_MIDI_IN_SIZE
+				+ 2*USB_DT_MIDI_OUT_SIZE(1),
+};
+
+#define JACK_IN_EMB	1
+#define JACK_IN_EXT	2
+#define JACK_OUT_EMB	3
+#define JACK_OUT_EXT	4
+
+/* B.4.3  MIDI IN Jack Descriptors */
+static const struct usb_midi_in_jack_descriptor jack_in_emb_desc = {
+	.bLength =		USB_DT_MIDI_IN_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_IN_JACK,
+	.bJackType =		USB_MS_EMBEDDED,
+	.bJackID =		JACK_IN_EMB,
+};
+
+static const struct usb_midi_in_jack_descriptor jack_in_ext_desc = {
+	.bLength =		USB_DT_MIDI_IN_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_IN_JACK,
+	.bJackType =		USB_MS_EXTERNAL,
+	.bJackID =		JACK_IN_EXT,
+};
+
+/* B.4.4  MIDI OUT Jack Descriptors */
+static const struct usb_midi_out_jack_descriptor_1 jack_out_emb_desc = {
+	.bLength =		USB_DT_MIDI_OUT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK,
+	.bJackType =		USB_MS_EMBEDDED,
+	.bJackID =		JACK_OUT_EMB,
+	.bNrInputPins =		1,
+	.pins = {
+		[0] = {
+			.baSourceID =	JACK_IN_EXT,
+			.baSourcePin =	1,
+		}
+	}
+};
+
+static const struct usb_midi_out_jack_descriptor_1 jack_out_ext_desc = {
+	.bLength =		USB_DT_MIDI_OUT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK,
+	.bJackType =		USB_MS_EXTERNAL,
+	.bJackID =		JACK_OUT_EXT,
+	.bNrInputPins =		1,
+	.pins = {
+		[0] = {
+			.baSourceID =	JACK_IN_EMB,
+			.baSourcePin =	1,
+		}
+	}
+};
+
+/* B.5.1  Standard Bulk OUT Endpoint Descriptor */
+static struct usb_endpoint_descriptor bulk_out_desc = {
+	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
+	.bDescriptorType =	USB_DT_ENDPOINT,
+	.bEndpointAddress =	USB_DIR_OUT,
+	.bmAttributes =		USB_ENDPOINT_XFER_BULK,
+};
+
+/* B.5.2  Class-specific MS Bulk OUT Endpoint Descriptor */
+static const struct usb_ms_endpoint_descriptor_1 ms_out_desc = {
+	.bLength =		USB_DT_MS_ENDPOINT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_ENDPOINT,
+	.bDescriptorSubtype =	USB_MS_GENERAL,
+	.bNumEmbMIDIJack =	1,
+	.baAssocJackID = {
+		[0] =		JACK_IN_EMB,
+	}
+};
+
+/* B.6.1  Standard Bulk IN Endpoint Descriptor */
+static struct usb_endpoint_descriptor bulk_in_desc = {
+	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
+	.bDescriptorType =	USB_DT_ENDPOINT,
+	.bEndpointAddress =	USB_DIR_IN,
+	.bmAttributes =		USB_ENDPOINT_XFER_BULK,
+};
+
+/* B.6.2  Class-specific MS Bulk IN Endpoint Descriptor */
+static const struct usb_ms_endpoint_descriptor_1 ms_in_desc = {
+	.bLength =		USB_DT_MS_ENDPOINT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_ENDPOINT,
+	.bDescriptorSubtype =	USB_MS_GENERAL,
+	.bNumEmbMIDIJack =	1,
+	.baAssocJackID = {
+		[0] =		JACK_OUT_EMB,
+	}
+};
+
+static const struct usb_descriptor_header *gmidi_function [] = {
+	(struct usb_descriptor_header *)&ac_interface_desc,
+	(struct usb_descriptor_header *)&ac_header_desc,
+	(struct usb_descriptor_header *)&ms_interface_desc,
+
+	(struct usb_descriptor_header *)&ms_header_desc,
+	(struct usb_descriptor_header *)&jack_in_emb_desc,
+	(struct usb_descriptor_header *)&jack_in_ext_desc,
+	(struct usb_descriptor_header *)&jack_out_emb_desc,
+	(struct usb_descriptor_header *)&jack_out_ext_desc,
+	/* If you add more jacks, update ms_header_desc.wTotalLength */
+
+	(struct usb_descriptor_header *)&bulk_out_desc,
+	(struct usb_descriptor_header *)&ms_out_desc,
+	(struct usb_descriptor_header *)&bulk_in_desc,
+	(struct usb_descriptor_header *)&ms_in_desc,
+	NULL,
+};
+
+static char manufacturer[50];
+static char product_desc[40] = "MIDI Gadget";
+static char serial_number[20];
+
+/* static strings, in UTF-8 */
+static struct usb_string strings [] = {
+	{ STRING_MANUFACTURER, manufacturer, },
+	{ STRING_PRODUCT, product_desc, },
+	{ STRING_SERIAL, serial_number, },
+	{ STRING_MIDI_GADGET, longname, },
+	{  }			/* end of list */
+};
+
+static struct usb_gadget_strings stringtab = {
+	.language	= 0x0409,	/* en-us */
+	.strings	= strings,
+};
+
+static int config_buf(struct usb_gadget *gadget,
+		u8 *buf, u8 type, unsigned index)
+{
+	int len;
+
+	/* only one configuration */
+	if (index != 0) {
+		return -EINVAL;
+	}
+	len = usb_gadget_config_buf(&config_desc,
+			buf, USB_BUFSIZ, gmidi_function);
+	if (len < 0) {
+		return len;
+	}
+	((struct usb_config_descriptor *)buf)->bDescriptorType = type;
+	return len;
+}
+
+static struct usb_request* alloc_ep_req(struct usb_ep *ep, unsigned length)
+{
+	struct usb_request	*req;
+
+	req = usb_ep_alloc_request(ep, GFP_ATOMIC);
+	if (req) {
+		req->length = length;
+		req->buf = kmalloc(length, GFP_ATOMIC);
+		if (!req->buf) {
+			usb_ep_free_request(ep, req);
+			req = NULL;
+		}
+	}
+	return req;
+}
+
+static void free_ep_req(struct usb_ep *ep, struct usb_request *req)
+{
+	kfree(req->buf);
+	usb_ep_free_request(ep, req);
+}
+
+static const uint8_t gmidi_cin_length[] = {
+	0, 0, 2, 3, 3, 1, 2, 3, 3, 3, 3, 3, 2, 2, 3, 1
+};
+
+/*
+ * Receives a chunk of MIDI data.
+ */
+static void gmidi_read_data(struct usb_ep *ep, int cable,
+				   uint8_t* data, int length)
+{
+	struct gmidi_device *dev = ep->driver_data;
+	/* cable is ignored, because for now we only have one. */
+
+	if (!dev->out_substream) {
+		/* Nobody is listening - throw it on the floor. */
+		return;
+	}
+	if (!test_bit(dev->out_substream->number, &dev->out_triggered)) {
+		return;
+	}
+	snd_rawmidi_receive(dev->out_substream, data, length);
+}
+
+static void gmidi_handle_out_data(struct usb_ep *ep, struct usb_request *req)
+{
+	unsigned i;
+	u8 *buf = req->buf;
+
+	for (i = 0; i + 3 < req->actual; i += 4) {
+		if (buf[i] != 0) {
+			int cable = buf[i] >> 4;
+			int length = gmidi_cin_length[buf[i] & 0x0f];
+			gmidi_read_data(ep, cable, &buf[i + 1], length);
+		}
+	}
+}
+
+static void gmidi_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	struct gmidi_device *dev = ep->driver_data;
+	int status = req->status;
+
+	switch (status) {
+	case 0: 			/* normal completion */
+		if (ep == dev->out_ep) {
+			/* we received stuff.
+			   req is queued again, below */
+			gmidi_handle_out_data(ep, req);
+		} else if (ep == dev->in_ep) {
+			/* our transmit completed.
+			   see if there's more to go.
+			   gmidi_transmit eats req, don't queue it again. */
+			gmidi_transmit(dev, req);
+			return;
+		}
+		break;
+
+	/* this endpoint is normally active while we're configured */
+	case -ECONNABORTED: 		/* hardware forced ep reset */
+	case -ECONNRESET:		/* request dequeued */
+	case -ESHUTDOWN:		/* disconnect from host */
+		VDBG(dev, "%s gone (%d), %d/%d\n", ep->name, status,
+				req->actual, req->length);
+		if (ep == dev->out_ep) {
+			gmidi_handle_out_data(ep, req);
+		}
+		free_ep_req(ep, req);
+		return;
+
+	case -EOVERFLOW:		/* buffer overrun on read means that
+					 * we didn't provide a big enough
+					 * buffer.
+					 */
+	default:
+		DBG(dev, "%s complete --> %d, %d/%d\n", ep->name,
+				status, req->actual, req->length);
+		break;
+	case -EREMOTEIO:		/* short read */
+		break;
+	}
+
+	status = usb_ep_queue(ep, req, GFP_ATOMIC);
+	if (status) {
+		ERROR(dev, "kill %s:  resubmit %d bytes --> %d\n",
+				ep->name, req->length, status);
+		usb_ep_set_halt(ep);
+		/* FIXME recover later ... somehow */
+	}
+}
+
+static int set_gmidi_config(struct gmidi_device *dev, gfp_t gfp_flags)
+{
+	int err = 0;
+	struct usb_request *req;
+	struct usb_ep* ep;
+	unsigned i;
+
+	err = usb_ep_enable(dev->in_ep, &bulk_in_desc);
+	if (err) {
+		ERROR(dev, "can't start %s: %d\n", dev->in_ep->name, err);
+		goto fail;
+	}
+	dev->in_ep->driver_data = dev;
+
+	err = usb_ep_enable(dev->out_ep, &bulk_out_desc);
+	if (err) {
+		ERROR(dev, "can't start %s: %d\n", dev->out_ep->name, err);
+		goto fail;
+	}
+	dev->out_ep->driver_data = dev;
+
+	/* allocate a bunch of read buffers and queue them all at once. */
+	ep = dev->out_ep;
+	for (i = 0; i < qlen && err == 0; i++) {
+		req = alloc_ep_req(ep, buflen);
+		if (req) {
+			req->complete = gmidi_complete;
+			err = usb_ep_queue(ep, req, GFP_ATOMIC);
+			if (err) {
+				DBG(dev, "%s queue req: %d\n", ep->name, err);
+			}
+		} else {
+			err = -ENOMEM;
+		}
+	}
+fail:
+	/* caller is responsible for cleanup on error */
+	return err;
+}
+
+
+static void gmidi_reset_config(struct gmidi_device *dev)
+{
+	if (dev->config == 0) {
+		return;
+	}
+
+	DBG(dev, "reset config\n");
+
+	/* just disable endpoints, forcing completion of pending i/o.
+	 * all our completion handlers free their requests in this case.
+	 */
+	usb_ep_disable(dev->in_ep);
+	usb_ep_disable(dev->out_ep);
+	dev->config = 0;
+}
+
+/* change our operational config.  this code must agree with the code
+ * that returns config descriptors, and altsetting code.
+ *
+ * it's also responsible for power management interactions. some
+ * configurations might not work with our current power sources.
+ *
+ * note that some device controller hardware will constrain what this
+ * code can do, perhaps by disallowing more than one configuration or
+ * by limiting configuration choices (like the pxa2xx).
+ */
+static int
+gmidi_set_config(struct gmidi_device *dev, unsigned number, gfp_t gfp_flags)
+{
+	int result = 0;
+	struct usb_gadget *gadget = dev->gadget;
+
+#if 0
+	/* FIXME */
+	/* Hacking this bit out fixes a bug where on receipt of two
+	   USB_REQ_SET_CONFIGURATION messages, we end up with no
+	   buffered OUT requests waiting for data. This is clearly
+	   hiding a bug elsewhere, because if the config didn't
+	   change then we really shouldn't do anything. */
+	/* Having said that, when we do "change" from config 1
+	   to config 1, we at least gmidi_reset_config() which
+	   clears out any requests on endpoints, so it's not like
+	   we leak or anything. */
+	if (number == dev->config) {
+		return 0;
+	}
+#endif
+
+	if (gadget_is_sa1100(gadget) && dev->config) {
+		/* tx fifo is full, but we can't clear it...*/
+		INFO(dev, "can't change configurations\n");
+		return -ESPIPE;
+	}
+	gmidi_reset_config(dev);
+
+	switch (number) {
+	case GMIDI_CONFIG:
+		result = set_gmidi_config(dev, gfp_flags);
+		break;
+	default:
+		result = -EINVAL;
+		/* FALL THROUGH */
+	case 0:
+		return result;
+	}
+
+	if (!result && (!dev->in_ep || !dev->out_ep)) {
+		result = -ENODEV;
+	}
+	if (result) {
+		gmidi_reset_config(dev);
+	} else {
+		char *speed;
+
+		switch (gadget->speed) {
+		case USB_SPEED_LOW:	speed = "low"; break;
+		case USB_SPEED_FULL:	speed = "full"; break;
+		case USB_SPEED_HIGH:	speed = "high"; break;
+		default: 		speed = "?"; break;
+		}
+
+		dev->config = number;
+		INFO(dev, "%s speed\n", speed);
+	}
+	return result;
+}
+
+
+static void gmidi_setup_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	if (req->status || req->actual != req->length) {
+		DBG((struct gmidi_device *) ep->driver_data,
+				"setup complete --> %d, %d/%d\n",
+				req->status, req->actual, req->length);
+	}
+}
+
+/*
+ * The setup() callback implements all the ep0 functionality that's
+ * not handled lower down, in hardware or the hardware driver (like
+ * device and endpoint feature flags, and their status).  It's all
+ * housekeeping for the gadget function we're implementing.  Most of
+ * the work is in config-specific setup.
+ */
+static int gmidi_setup(struct usb_gadget *gadget,
+			const struct usb_ctrlrequest *ctrl)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+	struct usb_request *req = dev->req;
+	int value = -EOPNOTSUPP;
+	u16 w_index = le16_to_cpu(ctrl->wIndex);
+	u16 w_value = le16_to_cpu(ctrl->wValue);
+	u16 w_length = le16_to_cpu(ctrl->wLength);
+
+	/* usually this stores reply data in the pre-allocated ep0 buffer,
+	 * but config change events will reconfigure hardware.
+	 */
+	req->zero = 0;
+	switch (ctrl->bRequest) {
+
+	case USB_REQ_GET_DESCRIPTOR:
+		if (ctrl->bRequestType != USB_DIR_IN) {
+			goto unknown;
+		}
+		switch (w_value >> 8) {
+
+		case USB_DT_DEVICE:
+			value = min(w_length, (u16) sizeof(device_desc));
+			memcpy(req->buf, &device_desc, value);
+			break;
+		case USB_DT_CONFIG:
+			value = config_buf(gadget, req->buf,
+					w_value >> 8,
+					w_value & 0xff);
+			if (value >= 0) {
+				value = min(w_length, (u16)value);
+			}
+			break;
+
+		case USB_DT_STRING:
+			/* wIndex == language code.
+			 * this driver only handles one language, you can
+			 * add string tables for other languages, using
+			 * any UTF-8 characters
+			 */
+			value = usb_gadget_get_string(&stringtab,
+					w_value & 0xff, req->buf);
+			if (value >= 0) {
+				value = min(w_length, (u16)value);
+			}
+			break;
+		}
+		break;
+
+	/* currently two configs, two speeds */
+	case USB_REQ_SET_CONFIGURATION:
+		if (ctrl->bRequestType != 0) {
+			goto unknown;
+		}
+		if (gadget->a_hnp_support) {
+			DBG(dev, "HNP available\n");
+		} else if (gadget->a_alt_hnp_support) {
+			DBG(dev, "HNP needs a different root port\n");
+		} else {
+			VDBG(dev, "HNP inactive\n");
+		}
+		spin_lock(&dev->lock);
+		value = gmidi_set_config(dev, w_value, GFP_ATOMIC);
+		spin_unlock(&dev->lock);
+		break;
+	case USB_REQ_GET_CONFIGURATION:
+		if (ctrl->bRequestType != USB_DIR_IN) {
+			goto unknown;
+		}
+		*(u8 *)req->buf = dev->config;
+		value = min(w_length, (u16)1);
+		break;
+
+	/* until we add altsetting support, or other interfaces,
+	 * only 0/0 are possible.  pxa2xx only supports 0/0 (poorly)
+	 * and already killed pending endpoint I/O.
+	 */
+	case USB_REQ_SET_INTERFACE:
+		if (ctrl->bRequestType != USB_RECIP_INTERFACE) {
+			goto unknown;
+		}
+		spin_lock(&dev->lock);
+		if (dev->config && w_index < GMIDI_NUM_INTERFACES
+			&& w_value == 0)
+		{
+			u8 config = dev->config;
+
+			/* resets interface configuration, forgets about
+			 * previous transaction state (queued bufs, etc)
+			 * and re-inits endpoint state (toggle etc)
+			 * no response queued, just zero status == success.
+			 * if we had more than one interface we couldn't
+			 * use this "reset the config" shortcut.
+			 */
+			gmidi_reset_config(dev);
+			gmidi_set_config(dev, config, GFP_ATOMIC);
+			value = 0;
+		}
+		spin_unlock(&dev->lock);
+		break;
+	case USB_REQ_GET_INTERFACE:
+		if (ctrl->bRequestType != (USB_DIR_IN|USB_RECIP_INTERFACE)) {
+			goto unknown;
+		}
+		if (!dev->config) {
+			break;
+		}
+		if (w_index >= GMIDI_NUM_INTERFACES) {
+			value = -EDOM;
+			break;
+		}
+		*(u8 *)req->buf = 0;
+		value = min(w_length, (u16)1);
+		break;
+
+	default:
+unknown:
+		VDBG(dev, "unknown control req%02x.%02x v%04x i%04x l%d\n",
+			ctrl->bRequestType, ctrl->bRequest,
+			w_value, w_index, w_length);
+	}
+
+	/* respond with data transfer before status phase? */
+	if (value >= 0) {
+		req->length = value;
+		req->zero = value < w_length;
+		value = usb_ep_queue(gadget->ep0, req, GFP_ATOMIC);
+		if (value < 0) {
+			DBG(dev, "ep_queue --> %d\n", value);
+			req->status = 0;
+			gmidi_setup_complete(gadget->ep0, req);
+		}
+	}
+
+	/* device either stalls (value < 0) or reports success */
+	return value;
+}
+
+static void gmidi_disconnect(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	gmidi_reset_config(dev);
+
+	/* a more significant application might have some non-usb
+	 * activities to quiesce here, saving resources like power
+	 * or pushing the notification up a network stack.
+	 */
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	/* next we may get setup() calls to enumerate new connections;
+	 * or an unbind() during shutdown (including removing module).
+	 */
+}
+
+static void /* __init_or_exit */ gmidi_unbind(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+	struct snd_card* card;
+
+	DBG(dev, "unbind\n");
+
+	card = dev->card;
+	dev->card = NULL;
+	if (card) {
+		snd_card_free(card);
+	}
+
+	/* we've already been disconnected ... no i/o is active */
+	if (dev->req) {
+		dev->req->length = USB_BUFSIZ;
+		free_ep_req(gadget->ep0, dev->req);
+	}
+	kfree(dev);
+	set_gadget_data(gadget, NULL);
+}
+
+static int gmidi_snd_free(struct snd_device *device)
+{
+	return 0;
+}
+
+static void gmidi_transmit_packet(struct usb_request* req, uint8_t p0,
+					uint8_t p1, uint8_t p2, uint8_t p3)
+{
+	unsigned length = req->length;
+
+	uint8_t* buf = (uint8_t*)req->buf + length;
+	buf[0] = p0;
+	buf[1] = p1;
+	buf[2] = p2;
+	buf[3] = p3;
+	req->length = length + 4;
+}
+
+/*
+ * Converts MIDI commands to USB MIDI packets.
+ */
+static void gmidi_transmit_byte(struct usb_request* req,
+				struct gmidi_in_port* port, uint8_t b)
+{
+	uint8_t p0 = port->cable;
+
+	if (b >= 0xf8) {
+		gmidi_transmit_packet(req, p0 | 0x0f, b, 0, 0);
+	} else if (b >= 0xf0) {
+		switch (b) {
+		case 0xf0:
+			port->data[0] = b;
+			port->state = STATE_SYSEX_1;
+			break;
+		case 0xf1:
+		case 0xf3:
+			port->data[0] = b;
+			port->state = STATE_1PARAM;
+			break;
+		case 0xf2:
+			port->data[0] = b;
+			port->state = STATE_2PARAM_1;
+			break;
+		case 0xf4:
+		case 0xf5:
+			port->state = STATE_UNKNOWN;
+			break;
+		case 0xf6:
+			gmidi_transmit_packet(req, p0 | 0x05, 0xf6, 0, 0);
+			port->state = STATE_UNKNOWN;
+			break;
+		case 0xf7:
+			switch (port->state) {
+			case STATE_SYSEX_0:
+				gmidi_transmit_packet(req,
+					p0 | 0x05, 0xf7, 0, 0);
+				break;
+			case STATE_SYSEX_1:
+				gmidi_transmit_packet(req,
+					p0 | 0x06, port->data[0], 0xf7, 0);
+				break;
+			case STATE_SYSEX_2:
+				gmidi_transmit_packet(req,
+					p0 | 0x07, port->data[0],
+					port->data[1], 0xf7);
+				break;
+			}
+			port->state = STATE_UNKNOWN;
+			break;
+		}
+	} else if (b >= 0x80) {
+		port->data[0] = b;
+		if (b >= 0xc0 && b <= 0xdf)
+			port->state = STATE_1PARAM;
+		else
+			port->state = STATE_2PARAM_1;
+	} else { /* b < 0x80 */
+		switch (port->state) {
+		case STATE_1PARAM:
+			if (port->data[0] < 0xf0) {
+				p0 |= port->data[0] >> 4;
+			} else {
+				p0 |= 0x02;
+				port->state = STATE_UNKNOWN;
+			}
+			gmidi_transmit_packet(req, p0, port->data[0], b, 0);
+			break;
+		case STATE_2PARAM_1:
+			port->data[1] = b;
+			port->state = STATE_2PARAM_2;
+			break;
+		case STATE_2PARAM_2:
+			if (port->data[0] < 0xf0) {
+				p0 |= port->data[0] >> 4;
+				port->state = STATE_2PARAM_1;
+			} else {
+				p0 |= 0x03;
+				port->state = STATE_UNKNOWN;
+			}
+			gmidi_transmit_packet(req,
+				p0, port->data[0], port->data[1], b);
+			break;
+		case STATE_SYSEX_0:
+			port->data[0] = b;
+			port->state = STATE_SYSEX_1;
+			break;
+		case STATE_SYSEX_1:
+			port->data[1] = b;
+			port->state = STATE_SYSEX_2;
+			break;
+		case STATE_SYSEX_2:
+			gmidi_transmit_packet(req,
+				p0 | 0x04, port->data[0], port->data[1], b);
+			port->state = STATE_SYSEX_0;
+			break;
+		}
+	}
+}
+
+static void gmidi_transmit(struct gmidi_device* dev, struct usb_request* req)
+{
+	struct usb_ep* ep = dev->in_ep;
+	struct gmidi_in_port* port = &dev->in_port;
+
+	if (!ep) {
+		return;
+	}
+	if (!req) {
+		req = alloc_ep_req(ep, buflen);
+	}
+	if (!req) {
+		ERROR(dev, "gmidi_transmit: alloc_ep_request failed\n");
+		return;
+	}
+	req->length = 0;
+	req->complete = gmidi_complete;
+
+	if (port->active) {
+		while (req->length + 3 < buflen) {
+			uint8_t b;
+			if (snd_rawmidi_transmit(dev->in_substream, &b, 1)
+				!= 1)
+			{
+				port->active = 0;
+				break;
+			}
+			gmidi_transmit_byte(req, port, b);
+		}
+	}
+	if (req->length > 0) {
+		usb_ep_queue(ep, req, GFP_ATOMIC);
+	} else {
+		free_ep_req(ep, req);
+	}
+}
+
+static void gmidi_in_tasklet(unsigned long data)
+{
+	struct gmidi_device* dev = (struct gmidi_device*)data;
+
+	gmidi_transmit(dev, NULL);
+}
+
+static int gmidi_in_open(struct snd_rawmidi_substream *substream)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_in_open\n");
+	dev->in_substream = substream;
+	dev->in_port.state = STATE_UNKNOWN;
+	return 0;
+}
+
+static int gmidi_in_close(struct snd_rawmidi_substream *substream)
+{
+	VDBG(dev, "gmidi_in_close\n");
+	return 0;
+}
+
+static void gmidi_in_trigger(struct snd_rawmidi_substream *substream, int up)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_in_trigger %d\n", up);
+	dev->in_port.active = up;
+	if (up) {
+		tasklet_hi_schedule(&dev->tasklet);
+	}
+}
+
+static int gmidi_out_open(struct snd_rawmidi_substream *substream)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_out_open\n");
+	dev->out_substream = substream;
+	return 0;
+}
+
+static int gmidi_out_close(struct snd_rawmidi_substream *substream)
+{
+	VDBG(dev, "gmidi_out_close\n");
+	return 0;
+}
+
+static void gmidi_out_trigger(struct snd_rawmidi_substream *substream, int up)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_out_trigger %d\n", up);
+	if (up) {
+		set_bit(substream->number, &dev->out_triggered);
+	} else {
+		clear_bit(substream->number, &dev->out_triggered);
+	}
+}
+
+static struct snd_rawmidi_ops gmidi_in_ops = {
+	.open = gmidi_in_open,
+	.close = gmidi_in_close,
+	.trigger = gmidi_in_trigger,
+};
+
+static struct snd_rawmidi_ops gmidi_out_ops = {
+	.open = gmidi_out_open,
+	.close = gmidi_out_close,
+	.trigger = gmidi_out_trigger
+};
+
+/* register as a sound "card" */
+static int gmidi_register_card(struct gmidi_device *dev)
+{
+	struct snd_card *card;
+	struct snd_rawmidi *rmidi;
+	int err;
+	int out_ports = 1;
+	int in_ports = 1;
+	static struct snd_device_ops ops = {
+		.dev_free = gmidi_snd_free,
+	};
+
+	card = snd_card_new(index, id, THIS_MODULE, 0);
+	if (!card) {
+		ERROR(dev, "snd_card_new failed\n");
+		err = -ENOMEM;
+		goto fail;
+	}
+	dev->card = card;
+
+	err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, dev, &ops);
+	if (err < 0) {
+		ERROR(dev, "snd_device_new failed: error %d\n", err);
+		goto fail;
+	}
+
+	strcpy(card->driver, longname);
+	strcpy(card->longname, longname);
+	strcpy(card->shortname, shortname);
+
+	/* Set up rawmidi */
+	dev->in_port.dev = dev;
+	dev->in_port.active = 0;
+	snd_component_add(card, "MIDI");
+	err = snd_rawmidi_new(card, "USB MIDI Gadget", 0,
+			      out_ports, in_ports, &rmidi);
+	if (err < 0) {
+		ERROR(dev, "snd_rawmidi_new failed: error %d\n", err);
+		goto fail;
+	}
+	dev->rmidi = rmidi;
+	strcpy(rmidi->name, card->shortname);
+	rmidi->info_flags = SNDRV_RAWMIDI_INFO_OUTPUT |
+			    SNDRV_RAWMIDI_INFO_INPUT |
+			    SNDRV_RAWMIDI_INFO_DUPLEX;
+	rmidi->private_data = dev;
+
+	/* Yes, rawmidi OUTPUT = USB IN, and rawmidi INPUT = USB OUT.
+	   It's an upside-down world being a gadget. */
+	snd_rawmidi_set_ops(rmidi, SNDRV_RAWMIDI_STREAM_OUTPUT, &gmidi_in_ops);
+	snd_rawmidi_set_ops(rmidi, SNDRV_RAWMIDI_STREAM_INPUT, &gmidi_out_ops);
+
+	snd_card_set_dev(card, &dev->gadget->dev);
+
+	/* register it - we're ready to go */
+	err = snd_card_register(card);
+	if (err < 0) {
+		ERROR(dev, "snd_card_register failed\n");
+		goto fail;
+	}
+
+	VDBG(dev, "gmidi_register_card finished ok\n");
+	return 0;
+
+fail:
+	if (dev->card) {
+		snd_card_free(dev->card);
+		dev->card = NULL;
+	}
+	return err;
+}
+
+/*
+ * Creates an output endpoint, and initializes output ports.
+ */
+static int __devinit gmidi_bind(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev;
+	struct usb_ep *in_ep, *out_ep;
+	int gcnum, err = 0;
+
+	/* support optional vendor/distro customization */
+	if (idVendor) {
+		if (!idProduct) {
+			printk(KERN_ERR "idVendor needs idProduct!\n");
+			return -ENODEV;
+		}
+		device_desc.idVendor = cpu_to_le16(idVendor);
+		device_desc.idProduct = cpu_to_le16(idProduct);
+		if (bcdDevice) {
+			device_desc.bcdDevice = cpu_to_le16(bcdDevice);
+		}
+	}
+	if (iManufacturer) {
+		strlcpy(manufacturer, iManufacturer, sizeof(manufacturer));
+	} else {
+		snprintf(manufacturer, sizeof(manufacturer), "%s %s with %s",
+			system_utsname.sysname, system_utsname.release,
+			gadget->name);
+	}
+	if (iProduct) {
+		strlcpy(product_desc, iProduct, sizeof(product_desc));
+	}
+	if (iSerialNumber) {
+		device_desc.iSerialNumber = STRING_SERIAL,
+		strlcpy(serial_number, iSerialNumber, sizeof(serial_number));
+	}
+
+	/* Bulk-only drivers like this one SHOULD be able to
+	 * autoconfigure on any sane usb controller driver,
+	 * but there may also be important quirks to address.
+	 */
+	usb_ep_autoconfig_reset(gadget);
+	in_ep = usb_ep_autoconfig(gadget, &bulk_in_desc);
+	if (!in_ep) {
+autoconf_fail:
+		printk(KERN_ERR "%s: can't autoconfigure on %s\n",
+			shortname, gadget->name);
+		return -ENODEV;
+	}
+	EP_IN_NAME = in_ep->name;
+	in_ep->driver_data = in_ep;	/* claim */
+
+	out_ep = usb_ep_autoconfig(gadget, &bulk_out_desc);
+	if (!out_ep) {
+		goto autoconf_fail;
+	}
+	EP_OUT_NAME = out_ep->name;
+	out_ep->driver_data = out_ep;	/* claim */
+
+	gcnum = usb_gadget_controller_number(gadget);
+	if (gcnum >= 0) {
+		device_desc.bcdDevice = cpu_to_le16(0x0200 + gcnum);
+	} else {
+		/* gmidi is so simple (no altsettings) that
+		 * it SHOULD NOT have problems with bulk-capable hardware.
+		 * so warn about unrecognized controllers, don't panic.
+		 */
+		printk(KERN_WARNING "%s: controller '%s' not recognized\n",
+			shortname, gadget->name);
+		device_desc.bcdDevice = __constant_cpu_to_le16(0x9999);
+	}
+
+
+	/* ok, we made sense of the hardware ... */
+	dev = kzalloc(sizeof(*dev), SLAB_KERNEL);
+	if (!dev) {
+		return -ENOMEM;
+	}
+	spin_lock_init(&dev->lock);
+	dev->gadget = gadget;
+	dev->in_ep = in_ep;
+	dev->out_ep = out_ep;
+	set_gadget_data(gadget, dev);
+	tasklet_init(&dev->tasklet, gmidi_in_tasklet, (unsigned long)dev);
+
+	/* preallocate control response and buffer */
+	dev->req = usb_ep_alloc_request(gadget->ep0, GFP_KERNEL);
+	if (!dev->req) {
+		err = -ENOMEM;
+		goto fail;
+	}
+	dev->req->buf = usb_ep_alloc_buffer(gadget->ep0, USB_BUFSIZ,
+				&dev->req->dma, GFP_KERNEL);
+	if (!dev->req->buf) {
+		err = -ENOMEM;
+		goto fail;
+	}
+
+	dev->req->complete = gmidi_setup_complete;
+
+	device_desc.bMaxPacketSize0 = gadget->ep0->maxpacket;
+
+	gadget->ep0->driver_data = dev;
+
+	INFO(dev, "%s, version: " DRIVER_VERSION "\n", longname);
+	INFO(dev, "using %s, OUT %s IN %s\n", gadget->name,
+		EP_OUT_NAME, EP_IN_NAME);
+
+	/* register as an ALSA sound card */
+	err = gmidi_register_card(dev);
+	if (err < 0) {
+		goto fail;
+	}
+
+	VDBG(dev, "gmidi_bind finished ok\n");
+	return 0;
+
+fail:
+	gmidi_unbind(gadget);
+	return err;
+}
+
+
+static void gmidi_suspend(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+
+	if (gadget->speed == USB_SPEED_UNKNOWN) {
+		return;
+	}
+
+	DBG(dev, "suspend\n");
+}
+
+static void gmidi_resume(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+
+	DBG(dev, "resume\n");
+}
+
+
+static struct usb_gadget_driver gmidi_driver = {
+	.speed		= USB_SPEED_FULL,
+	.function	= (char *)longname,
+	.bind		= gmidi_bind,
+	.unbind		= __exit_p(gmidi_unbind),
+
+	.setup		= gmidi_setup,
+	.disconnect	= gmidi_disconnect,
+
+	.suspend	= gmidi_suspend,
+	.resume		= gmidi_resume,
+
+	.driver 	= {
+		.name		= (char *)shortname,
+		.owner		= THIS_MODULE,
+	},
+};
+
+static int __init gmidi_init(void)
+{
+	return usb_gadget_register_driver(&gmidi_driver);
+}
+module_init(gmidi_init);
+
+static void __exit gmidi_cleanup(void)
+{
+	usb_gadget_unregister_driver(&gmidi_driver);
+}
+module_exit(gmidi_cleanup);
+
diff --git a/include/linux/usb/audio.h b/include/linux/usb/audio.h
new file mode 100644
index 000000000000..6bd235994dc2
--- /dev/null
+++ b/include/linux/usb/audio.h
@@ -0,0 +1,53 @@
+/*
+ * <linux/usb/audio.h> -- USB Audio definitions.
+ *
+ * Copyright (C) 2006 Thumtronics Pty Ltd.
+ * Developed for Thumtronics by Grey Innovation
+ * Ben Williamson <ben.williamson@greyinnovation.com>
+ *
+ * This software is distributed under the terms of the GNU General Public
+ * License ("GPL") version 2, as published by the Free Software Foundation.
+ *
+ * This file holds USB constants and structures defined
+ * by the USB Device Class Definition for Audio Devices.
+ * Comments below reference relevant sections of that document:
+ *
+ * http://www.usb.org/developers/devclass_docs/audio10.pdf
+ */
+
+#ifndef __LINUX_USB_AUDIO_H
+#define __LINUX_USB_AUDIO_H
+
+#include <linux/types.h>
+
+/* A.2 Audio Interface Subclass Codes */
+#define USB_SUBCLASS_AUDIOCONTROL	0x01
+#define USB_SUBCLASS_AUDIOSTREAMING	0x02
+#define USB_SUBCLASS_MIDISTREAMING	0x03
+
+/* 4.3.2  Class-Specific AC Interface Descriptor */
+struct usb_ac_header_descriptor {
+	__u8  bLength;			// 8+n
+	__u8  bDescriptorType;		// USB_DT_CS_INTERFACE
+	__u8  bDescriptorSubtype;	// USB_MS_HEADER
+	__le16 bcdADC;			// 0x0100
+	__le16 wTotalLength;		// includes Unit and Terminal desc.
+	__u8  bInCollection;		// n
+	__u8  baInterfaceNr[];		// [n]
+} __attribute__ ((packed));
+
+#define USB_DT_AC_HEADER_SIZE(n)	(8+(n))
+
+/* As above, but more useful for defining your own descriptors: */
+#define DECLARE_USB_AC_HEADER_DESCRIPTOR(n) 			\
+struct usb_ac_header_descriptor_##n {				\
+	__u8  bLength;						\
+	__u8  bDescriptorType;					\
+	__u8  bDescriptorSubtype;				\
+	__le16 bcdADC;						\
+	__le16 wTotalLength;					\
+	__u8  bInCollection;					\
+	__u8  baInterfaceNr[n];					\
+} __attribute__ ((packed))
+
+#endif
diff --git a/include/linux/usb/midi.h b/include/linux/usb/midi.h
new file mode 100644
index 000000000000..11a97d5ffd34
--- /dev/null
+++ b/include/linux/usb/midi.h
@@ -0,0 +1,112 @@
+/*
+ * <linux/usb/midi.h> -- USB MIDI definitions.
+ *
+ * Copyright (C) 2006 Thumtronics Pty Ltd.
+ * Developed for Thumtronics by Grey Innovation
+ * Ben Williamson <ben.williamson@greyinnovation.com>
+ *
+ * This software is distributed under the terms of the GNU General Public
+ * License ("GPL") version 2, as published by the Free Software Foundation.
+ *
+ * This file holds USB constants and structures defined
+ * by the USB Device Class Definition for MIDI Devices.
+ * Comments below reference relevant sections of that document:
+ *
+ * http://www.usb.org/developers/devclass_docs/midi10.pdf
+ */
+
+#ifndef __LINUX_USB_MIDI_H
+#define __LINUX_USB_MIDI_H
+
+#include <linux/types.h>
+
+/* A.1  MS Class-Specific Interface Descriptor Subtypes */
+#define USB_MS_HEADER		0x01
+#define USB_MS_MIDI_IN_JACK	0x02
+#define USB_MS_MIDI_OUT_JACK	0x03
+#define USB_MS_ELEMENT		0x04
+
+/* A.2  MS Class-Specific Endpoint Descriptor Subtypes */
+#define USB_MS_GENERAL		0x01
+
+/* A.3  MS MIDI IN and OUT Jack Types */
+#define USB_MS_EMBEDDED		0x01
+#define USB_MS_EXTERNAL		0x02
+
+/* 6.1.2.1  Class-Specific MS Interface Header Descriptor */
+struct usb_ms_header_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+	__u8  bDescriptorSubtype;
+	__le16 bcdMSC;
+	__le16 wTotalLength;
+} __attribute__ ((packed));
+
+#define USB_DT_MS_HEADER_SIZE	7
+
+/* 6.1.2.2  MIDI IN Jack Descriptor */
+struct usb_midi_in_jack_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;		// USB_DT_CS_INTERFACE
+	__u8  bDescriptorSubtype;	// USB_MS_MIDI_IN_JACK
+	__u8  bJackType;		// USB_MS_EMBEDDED/EXTERNAL
+	__u8  bJackID;
+	__u8  iJack;
+} __attribute__ ((packed));
+
+#define USB_DT_MIDI_IN_SIZE	6
+
+struct usb_midi_source_pin {
+	__u8  baSourceID;
+	__u8  baSourcePin;
+} __attribute__ ((packed));
+
+/* 6.1.2.3  MIDI OUT Jack Descriptor */
+struct usb_midi_out_jack_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;		// USB_DT_CS_INTERFACE
+	__u8  bDescriptorSubtype;	// USB_MS_MIDI_OUT_JACK
+	__u8  bJackType;		// USB_MS_EMBEDDED/EXTERNAL
+	__u8  bJackID;
+	__u8  bNrInputPins;		// p
+	struct usb_midi_source_pin pins[]; // [p]
+	/*__u8  iJack;  -- ommitted due to variable-sized pins[] */
+} __attribute__ ((packed));
+
+#define USB_DT_MIDI_OUT_SIZE(p)	(7 + 2 * (p))
+
+/* As above, but more useful for defining your own descriptors: */
+#define DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(p)			\
+struct usb_midi_out_jack_descriptor_##p {			\
+	__u8  bLength;						\
+	__u8  bDescriptorType;					\
+	__u8  bDescriptorSubtype;				\
+	__u8  bJackType;					\
+	__u8  bJackID;						\
+	__u8  bNrInputPins;					\
+	struct usb_midi_source_pin pins[p];			\
+	__u8  iJack;						\
+} __attribute__ ((packed))
+
+/* 6.2.2  Class-Specific MS Bulk Data Endpoint Descriptor */
+struct usb_ms_endpoint_descriptor {
+	__u8  bLength;			// 4+n
+	__u8  bDescriptorType;		// USB_DT_CS_ENDPOINT
+	__u8  bDescriptorSubtype;	// USB_MS_GENERAL
+	__u8  bNumEmbMIDIJack;		// n
+	__u8  baAssocJackID[];		// [n]
+} __attribute__ ((packed));
+
+#define USB_DT_MS_ENDPOINT_SIZE(n)	(4 + (n))
+
+/* As above, but more useful for defining your own descriptors: */
+#define DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(n)			\
+struct usb_ms_endpoint_descriptor_##n {				\
+	__u8  bLength;						\
+	__u8  bDescriptorType;					\
+	__u8  bDescriptorSubtype;				\
+	__u8  bNumEmbMIDIJack;					\
+	__u8  baAssocJackID[n];					\
+} __attribute__ ((packed))
+
+#endif
-- 
cgit v1.2.3


From 3d5b2510f6e361e2203e163c03b93d0026de5629 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jesper.juhl@gmail.com>
Date: Sun, 30 Jul 2006 18:43:43 +0200
Subject: USB: making the kernel -Wshadow clean - USB & completion

include/linux/usb.h causes a lot of -Wshadow warnings - fix them.

  include/linux/usb.h:901: warning: declaration of 'complete' shadows a global declaration
  include/linux/completion.h:52: warning: shadowed declaration is here
  include/linux/usb.h:932: warning: declaration of 'complete' shadows a global declaration
  include/linux/completion.h:52: warning: shadowed declaration is here
  include/linux/usb.h:967: warning: declaration of 'complete' shadows a global declaration
  include/linux/completion.h:52: warning: shadowed declaration is here


Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb.h | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/include/linux/usb.h b/include/linux/usb.h
index e22f4b386605..3d5cfa731680 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -932,7 +932,7 @@ struct urb
  * @setup_packet: pointer to the setup_packet buffer
  * @transfer_buffer: pointer to the transfer buffer
  * @buffer_length: length of the transfer buffer
- * @complete: pointer to the usb_complete_t function
+ * @complete_fn: pointer to the usb_complete_t function
  * @context: what to set the urb context to.
  *
  * Initializes a control urb with the proper information needed to submit
@@ -944,7 +944,7 @@ static inline void usb_fill_control_urb (struct urb *urb,
 					 unsigned char *setup_packet,
 					 void *transfer_buffer,
 					 int buffer_length,
-					 usb_complete_t complete,
+					 usb_complete_t complete_fn,
 					 void *context)
 {
 	spin_lock_init(&urb->lock);
@@ -953,7 +953,7 @@ static inline void usb_fill_control_urb (struct urb *urb,
 	urb->setup_packet = setup_packet;
 	urb->transfer_buffer = transfer_buffer;
 	urb->transfer_buffer_length = buffer_length;
-	urb->complete = complete;
+	urb->complete = complete_fn;
 	urb->context = context;
 }
 
@@ -964,7 +964,7 @@ static inline void usb_fill_control_urb (struct urb *urb,
  * @pipe: the endpoint pipe
  * @transfer_buffer: pointer to the transfer buffer
  * @buffer_length: length of the transfer buffer
- * @complete: pointer to the usb_complete_t function
+ * @complete_fn: pointer to the usb_complete_t function
  * @context: what to set the urb context to.
  *
  * Initializes a bulk urb with the proper information needed to submit it
@@ -975,7 +975,7 @@ static inline void usb_fill_bulk_urb (struct urb *urb,
 				      unsigned int pipe,
 				      void *transfer_buffer,
 				      int buffer_length,
-				      usb_complete_t complete,
+				      usb_complete_t complete_fn,
 				      void *context)
 {
 	spin_lock_init(&urb->lock);
@@ -983,7 +983,7 @@ static inline void usb_fill_bulk_urb (struct urb *urb,
 	urb->pipe = pipe;
 	urb->transfer_buffer = transfer_buffer;
 	urb->transfer_buffer_length = buffer_length;
-	urb->complete = complete;
+	urb->complete = complete_fn;
 	urb->context = context;
 }
 
@@ -994,7 +994,7 @@ static inline void usb_fill_bulk_urb (struct urb *urb,
  * @pipe: the endpoint pipe
  * @transfer_buffer: pointer to the transfer buffer
  * @buffer_length: length of the transfer buffer
- * @complete: pointer to the usb_complete_t function
+ * @complete_fn: pointer to the usb_complete_t function
  * @context: what to set the urb context to.
  * @interval: what to set the urb interval to, encoded like
  *	the endpoint descriptor's bInterval value.
@@ -1010,7 +1010,7 @@ static inline void usb_fill_int_urb (struct urb *urb,
 				     unsigned int pipe,
 				     void *transfer_buffer,
 				     int buffer_length,
-				     usb_complete_t complete,
+				     usb_complete_t complete_fn,
 				     void *context,
 				     int interval)
 {
@@ -1019,7 +1019,7 @@ static inline void usb_fill_int_urb (struct urb *urb,
 	urb->pipe = pipe;
 	urb->transfer_buffer = transfer_buffer;
 	urb->transfer_buffer_length = buffer_length;
-	urb->complete = complete;
+	urb->complete = complete_fn;
 	urb->context = context;
 	if (dev->speed == USB_SPEED_HIGH)
 		urb->interval = 1 << (interval - 1);
-- 
cgit v1.2.3


From b7cfaaaf86571732c7728e95a2231a860385463c Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: New functions to check endpoints info.

These functions makes USB driver's code simpler when dealing with endpoints
by avoiding them from accessing the endpoint's descriptor structure directly
when they only need to know the endpoint's transfer type and/or
direction.

Please, read each functions' documentation in order to know how to use
them.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.c | 144 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb.h    |  14 +++++
 2 files changed, 158 insertions(+)

(limited to 'include')

diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 9ebfc0fe819d..82837d45b484 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -482,6 +482,138 @@ int usb_get_current_frame_number(struct usb_device *dev)
 	return dev->bus->op->get_frame_number (dev);
 }
 
+/**
+ * usb_endpoint_dir_in - check if the endpoint has IN direction
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type IN, otherwise it returns false.
+ */
+int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN);
+}
+
+/**
+ * usb_endpoint_dir_out - check if the endpoint has OUT direction
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type OUT, otherwise it returns false.
+ */
+int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT);
+}
+
+/**
+ * usb_endpoint_xfer_bulk - check if the endpoint has bulk transfer type
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type bulk, otherwise it returns false.
+ */
+int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) ==
+		USB_ENDPOINT_XFER_BULK);
+}
+
+/**
+ * usb_endpoint_xfer_int - check if the endpoint has interrupt transfer type
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type interrupt, otherwise it returns
+ * false.
+ */
+int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) ==
+		USB_ENDPOINT_XFER_INT);
+}
+
+/**
+ * usb_endpoint_xfer_isoc - check if the endpoint has isochronous transfer type
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type isochronous, otherwise it returns
+ * false.
+ */
+int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) ==
+		USB_ENDPOINT_XFER_ISOC);
+}
+
+/**
+ * usb_endpoint_is_bulk_in - check if the endpoint is bulk IN
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has bulk transfer type and IN direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_in(epd));
+}
+
+/**
+ * usb_endpoint_is_bulk_out - check if the endpoint is bulk OUT
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has bulk transfer type and OUT direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_out(epd));
+}
+
+/**
+ * usb_endpoint_is_int_in - check if the endpoint is interrupt IN
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has interrupt transfer type and IN direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_in(epd));
+}
+
+/**
+ * usb_endpoint_is_int_out - check if the endpoint is interrupt OUT
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has interrupt transfer type and OUT direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_out(epd));
+}
+
+/**
+ * usb_endpoint_is_isoc_in - check if the endpoint is isochronous IN
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has isochronous transfer type and IN direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_in(epd));
+}
+
+/**
+ * usb_endpoint_is_isoc_out - check if the endpoint is isochronous OUT
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has isochronous transfer type and OUT direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_out(epd));
+}
+
 /*-------------------------------------------------------------------*/
 /*
  * __usb_get_extra_descriptor() finds a descriptor of specific type in the
@@ -909,6 +1041,18 @@ EXPORT_SYMBOL(__usb_get_extra_descriptor);
 EXPORT_SYMBOL(usb_find_device);
 EXPORT_SYMBOL(usb_get_current_frame_number);
 
+EXPORT_SYMBOL_GPL(usb_endpoint_dir_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_dir_out);
+EXPORT_SYMBOL_GPL(usb_endpoint_xfer_bulk);
+EXPORT_SYMBOL_GPL(usb_endpoint_xfer_int);
+EXPORT_SYMBOL_GPL(usb_endpoint_xfer_isoc);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_out);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_int_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_int_out);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_out);
+
 EXPORT_SYMBOL (usb_buffer_alloc);
 EXPORT_SYMBOL (usb_buffer_free);
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 3d5cfa731680..f807479ef65b 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -467,6 +467,20 @@ static inline int usb_make_path (struct usb_device *dev, char *buf,
 
 /*-------------------------------------------------------------------------*/
 
+extern int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd);
+
+/*-------------------------------------------------------------------------*/
+
 #define USB_DEVICE_ID_MATCH_DEVICE \
 		(USB_DEVICE_ID_MATCH_VENDOR | USB_DEVICE_ID_MATCH_PRODUCT)
 #define USB_DEVICE_ID_MATCH_DEV_RANGE \
-- 
cgit v1.2.3


From dfe0d3ba20e860d0b9a16c4c6524180b8f93be05 Mon Sep 17 00:00:00 2001
From: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Date: Sun, 13 Aug 2006 17:30:14 -0700
Subject: USB Storage: add rio karma eject support

This changeset from Keith Bennett (via Bob Copeland) moves the Karma
initializer to its own file and adds trapping of the START_STOP command to
enable eject of the device.

Signed-off-by: Keith Bennett <keith@mcs.st-and.ac.uk>
Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/Kconfig        |  12 +++
 drivers/usb/storage/Makefile       |   1 +
 drivers/usb/storage/initializers.c |  73 -----------------
 drivers/usb/storage/initializers.h |   1 -
 drivers/usb/storage/karma.c        | 155 +++++++++++++++++++++++++++++++++++++
 drivers/usb/storage/karma.h        |   7 ++
 drivers/usb/storage/unusual_devs.h |   2 +-
 drivers/usb/storage/usb.c          |  11 +++
 include/linux/usb_usual.h          |   3 +
 9 files changed, 190 insertions(+), 75 deletions(-)
 create mode 100644 drivers/usb/storage/karma.c
 create mode 100644 drivers/usb/storage/karma.h

(limited to 'include')

diff --git a/drivers/usb/storage/Kconfig b/drivers/usb/storage/Kconfig
index be9eec225743..86e48c42d6af 100644
--- a/drivers/usb/storage/Kconfig
+++ b/drivers/usb/storage/Kconfig
@@ -135,6 +135,18 @@ config USB_STORAGE_ONETOUCH
 	  this input in any keybinding software. (e.g. gnome's keyboard short-
 	  cuts)
 
+config USB_STORAGE_KARMA
+	bool "Support for Rio Karma music player"
+	depends on USB_STORAGE
+	help
+	  Say Y here to include additional code to support the Rio Karma
+	  USB interface.
+
+	  This code places the Rio Karma into mass storage mode, enabling
+	  it to be mounted as an ordinary filesystem. Performing an eject
+	  on the resulting scsi device node returns the Karma to normal
+	  operation.
+
 config USB_LIBUSUAL
 	bool "The shared table of common (or usual) storage devices"
 	depends on USB
diff --git a/drivers/usb/storage/Makefile b/drivers/usb/storage/Makefile
index 8cbba22508a4..023969b4385b 100644
--- a/drivers/usb/storage/Makefile
+++ b/drivers/usb/storage/Makefile
@@ -20,6 +20,7 @@ usb-storage-obj-$(CONFIG_USB_STORAGE_DATAFAB)	+= datafab.o
 usb-storage-obj-$(CONFIG_USB_STORAGE_JUMPSHOT)	+= jumpshot.o
 usb-storage-obj-$(CONFIG_USB_STORAGE_ALAUDA)	+= alauda.o
 usb-storage-obj-$(CONFIG_USB_STORAGE_ONETOUCH)	+= onetouch.o
+usb-storage-obj-$(CONFIG_USB_STORAGE_KARMA)	+= karma.o
 
 usb-storage-objs :=	scsiglue.o protocol.o transport.o usb.o \
 			initializers.o $(usb-storage-obj-y)
diff --git a/drivers/usb/storage/initializers.c b/drivers/usb/storage/initializers.c
index ab173b30076e..5b06f9240d05 100644
--- a/drivers/usb/storage/initializers.c
+++ b/drivers/usb/storage/initializers.c
@@ -45,12 +45,6 @@
 #include "debug.h"
 #include "transport.h"
 
-#define RIO_MSC 0x08
-#define RIOP_INIT "RIOP\x00\x01\x08"
-#define RIOP_INIT_LEN 7
-#define RIO_SEND_LEN 40
-#define RIO_RECV_LEN 0x200
-
 /* This places the Shuttle/SCM USB<->SCSI bridge devices in multi-target
  * mode */
 int usb_stor_euscsi_init(struct us_data *us)
@@ -97,70 +91,3 @@ int usb_stor_ucr61s2b_init(struct us_data *us)
 
 	return (res ? -1 : 0);
 }
-
-/* Place the Rio Karma into mass storage mode.
- *
- * The initialization begins by sending 40 bytes starting
- * RIOP\x00\x01\x08\x00, which the device will ack with a 512-byte
- * packet with the high four bits set and everything else null.
- *
- * Next, we send RIOP\x80\x00\x08\x00.  Each time, a 512 byte response
- * must be read, but we must loop until byte 5 in the response is 0x08,
- * indicating success.  */
-int rio_karma_init(struct us_data *us)
-{
-	int result, partial;
-	char *recv;
-	unsigned long timeout;
-
-	// us->iobuf is big enough to hold cmd but not receive
-	if (!(recv = kmalloc(RIO_RECV_LEN, GFP_KERNEL)))
-		goto die_nomem;
-
-	US_DEBUGP("Initializing Karma...\n");
-
-	memset(us->iobuf, 0, RIO_SEND_LEN);
-	memcpy(us->iobuf, RIOP_INIT, RIOP_INIT_LEN);
-
-	result = usb_stor_bulk_transfer_buf(us, us->send_bulk_pipe,
-		us->iobuf, RIO_SEND_LEN, &partial);
-	if (result != USB_STOR_XFER_GOOD)
-		goto die;
-
-	result = usb_stor_bulk_transfer_buf(us, us->recv_bulk_pipe,
-		recv, RIO_RECV_LEN, &partial);
-	if (result != USB_STOR_XFER_GOOD)
-		goto die;
-
-	us->iobuf[4] = 0x80;
-	us->iobuf[5] = 0;
-	timeout = jiffies + msecs_to_jiffies(3000);
-	for (;;) {
-		US_DEBUGP("Sending init command\n");
-		result = usb_stor_bulk_transfer_buf(us, us->send_bulk_pipe,
-			us->iobuf, RIO_SEND_LEN, &partial);
-		if (result != USB_STOR_XFER_GOOD)
-			goto die;
-
-		result = usb_stor_bulk_transfer_buf(us, us->recv_bulk_pipe,
-			recv, RIO_RECV_LEN, &partial);
-		if (result != USB_STOR_XFER_GOOD)
-			goto die;
-
-		if (recv[5] == RIO_MSC)
-			break;
-		if (time_after(jiffies, timeout))
-			goto die;
-		msleep(10);
-	}
-	US_DEBUGP("Karma initialized.\n");
-	kfree(recv);
-	return 0;
-
-die:
-	kfree(recv);
-die_nomem:
-	US_DEBUGP("Could not initialize karma.\n");
-	return USB_STOR_TRANSPORT_FAILED;
-}
-
diff --git a/drivers/usb/storage/initializers.h b/drivers/usb/storage/initializers.h
index 927f7781080f..e2967a4d48a2 100644
--- a/drivers/usb/storage/initializers.h
+++ b/drivers/usb/storage/initializers.h
@@ -47,4 +47,3 @@ int usb_stor_euscsi_init(struct us_data *us);
 /* This function is required to activate all four slots on the UCR-61S2B
  * flash reader */
 int usb_stor_ucr61s2b_init(struct us_data *us);
-int rio_karma_init(struct us_data *us);
diff --git a/drivers/usb/storage/karma.c b/drivers/usb/storage/karma.c
new file mode 100644
index 000000000000..0d79ae5683f7
--- /dev/null
+++ b/drivers/usb/storage/karma.c
@@ -0,0 +1,155 @@
+/* Driver for Rio Karma
+ *
+ *   (c) 2006 Bob Copeland <me@bobcopeland.com>
+ *   (c) 2006 Keith Bennett <keith@mcs.st-and.ac.uk>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_device.h>
+
+#include "usb.h"
+#include "transport.h"
+#include "debug.h"
+#include "karma.h"
+
+#define RIO_PREFIX "RIOP\x00"
+#define RIO_PREFIX_LEN 5
+#define RIO_SEND_LEN 40
+#define RIO_RECV_LEN 0x200
+
+#define RIO_ENTER_STORAGE 0x1
+#define RIO_LEAVE_STORAGE 0x2
+#define RIO_RESET 0xC
+
+extern int usb_stor_Bulk_transport(struct scsi_cmnd *, struct us_data *);
+
+struct karma_data {
+	int in_storage;
+	char *recv;
+};
+
+/*
+ * Send commands to Rio Karma.
+ *
+ * For each command we send 40 bytes starting 'RIOP\0' followed by
+ * the command number and a sequence number, which the device will ack
+ * with a 512-byte packet with the high four bits set and everything
+ * else null.  Then we send 'RIOP\x80' followed by a zero and the
+ * sequence number, until byte 5 in the response repeats the sequence
+ * number.
+ */
+static int rio_karma_send_command(char cmd, struct us_data *us)
+{
+	int result, partial;
+	unsigned long timeout;
+	static unsigned char seq = 1;
+	struct karma_data *data = (struct karma_data *) us->extra;
+
+	US_DEBUGP("karma: sending command %04x\n", cmd);
+	memset(us->iobuf, 0, RIO_SEND_LEN);
+	memcpy(us->iobuf, RIO_PREFIX, RIO_PREFIX_LEN);
+	us->iobuf[5] = cmd;
+	us->iobuf[6] = seq;
+
+	timeout = jiffies + msecs_to_jiffies(6000);
+	for (;;) {
+		result = usb_stor_bulk_transfer_buf(us, us->send_bulk_pipe,
+			us->iobuf, RIO_SEND_LEN, &partial);
+		if (result != USB_STOR_XFER_GOOD)
+			goto err;
+
+		result = usb_stor_bulk_transfer_buf(us, us->recv_bulk_pipe,
+			data->recv, RIO_RECV_LEN, &partial);
+		if (result != USB_STOR_XFER_GOOD)
+			goto err;
+
+		if (data->recv[5] == seq)
+			break;
+
+		if (time_after(jiffies, timeout))
+			goto err;
+
+		us->iobuf[4] = 0x80;
+		us->iobuf[5] = 0;
+		msleep(50);
+	}
+
+	seq++;
+	if (seq == 0)
+		seq = 1;
+
+	US_DEBUGP("karma: sent command %04x\n", cmd);
+	return 0;
+err:
+	US_DEBUGP("karma: command %04x failed\n", cmd);
+	return USB_STOR_TRANSPORT_FAILED;
+}
+
+/*
+ * Trap START_STOP and READ_10 to leave/re-enter storage mode.
+ * Everything else is propagated to the normal bulk layer.
+ */
+int rio_karma_transport(struct scsi_cmnd *srb, struct us_data *us)
+{
+	int ret;
+	struct karma_data *data = (struct karma_data *) us->extra;
+
+	if (srb->cmnd[0] == READ_10 && !data->in_storage) {
+		ret = rio_karma_send_command(RIO_ENTER_STORAGE, us);
+		if (ret)
+			return ret;
+
+		data->in_storage = 1;
+		return usb_stor_Bulk_transport(srb, us);
+	} else if (srb->cmnd[0] == START_STOP) {
+		ret = rio_karma_send_command(RIO_LEAVE_STORAGE, us);
+		if (ret)
+			return ret;
+
+		data->in_storage = 0;
+		return rio_karma_send_command(RIO_RESET, us);
+	}
+	return usb_stor_Bulk_transport(srb, us);
+}
+
+static void rio_karma_destructor(void *extra)
+{
+	struct karma_data *data = (struct karma_data *) extra;
+	kfree(data->recv);
+}
+
+int rio_karma_init(struct us_data *us)
+{
+	int ret = 0;
+	struct karma_data *data = kzalloc(sizeof(struct karma_data), GFP_NOIO);
+	if (!data)
+		goto out;
+
+	data->recv = kmalloc(RIO_RECV_LEN, GFP_NOIO);
+	if (!data->recv) {
+		kfree(data);
+		goto out;
+	}
+
+	us->extra = data;
+	us->extra_destructor = rio_karma_destructor;
+	ret = rio_karma_send_command(RIO_ENTER_STORAGE, us);
+	data->in_storage = (ret == 0);
+out:
+	return ret;
+}
diff --git a/drivers/usb/storage/karma.h b/drivers/usb/storage/karma.h
new file mode 100644
index 000000000000..8a60972af8c5
--- /dev/null
+++ b/drivers/usb/storage/karma.h
@@ -0,0 +1,7 @@
+#ifndef _KARMA_USB_H
+#define _KARMA_USB_H
+
+extern int rio_karma_init(struct us_data *us);
+extern int rio_karma_transport(struct scsi_cmnd *srb, struct us_data *us);
+
+#endif
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index fa49357289c4..1f11c9d44eaa 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -221,7 +221,7 @@ UNUSUAL_DEV(  0x0457, 0x0151, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x045a, 0x5210, 0x0101, 0x0101,
 		"Rio",
 		"Rio Karma",
-		US_SC_SCSI, US_PR_BULK, rio_karma_init, 0),
+		US_SC_SCSI, US_PR_KARMA, rio_karma_init, 0),
 
 /* Patch submitted by Philipp Friedrich <philipp@void.at> */
 UNUSUAL_DEV(  0x0482, 0x0100, 0x0100, 0x0100,
diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c
index 8d7bdcb5924d..b8d6031b0975 100644
--- a/drivers/usb/storage/usb.c
+++ b/drivers/usb/storage/usb.c
@@ -98,6 +98,9 @@
 #ifdef CONFIG_USB_STORAGE_ALAUDA
 #include "alauda.h"
 #endif
+#ifdef CONFIG_USB_STORAGE_KARMA
+#include "karma.h"
+#endif
 
 /* Some informational data */
 MODULE_AUTHOR("Matthew Dharm <mdharm-usb@one-eyed-alien.net>");
@@ -646,6 +649,14 @@ static int get_transport(struct us_data *us)
 		break;
 #endif
 
+#ifdef CONFIG_USB_STORAGE_KARMA
+	case US_PR_KARMA:
+		us->transport_name = "Rio Karma/Bulk";
+		us->transport = rio_karma_transport;
+		us->transport_reset = usb_stor_Bulk_reset;
+		break;
+#endif
+
 	default:
 		return -EIO;
 	}
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index e7fc5fed5b98..2ae76fe52ff7 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -108,6 +108,9 @@ enum { US_DO_ALL_FLAGS };
 #ifdef CONFIG_USB_STORAGE_ALAUDA
 #define US_PR_ALAUDA    0xf4		/* Alauda chipsets */
 #endif
+#ifdef CONFIG_USB_STORAGE_KARMA
+#define US_PR_KARMA     0xf5		/* Rio Karma */
+#endif
 
 #define US_PR_DEVICE	0xff		/* Use device's value */
 
-- 
cgit v1.2.3


From 095bc335360a51623dd8571839bbf465851a7f4b Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Sat, 26 Aug 2006 23:48:11 -0300
Subject: USB core: Use const where possible.

This patch marks some USB core's functions parameters as const. This
improves the design (we're saying to the caller that its parameter is
not going to be modified) and may help in compiler's optimisation work.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.c | 21 +++++++++++----------
 drivers/usb/core/usb.h |  4 ++--
 include/linux/usb.h    | 18 +++++++++---------
 3 files changed, 22 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 82837d45b484..4eb98eb3804f 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -67,7 +67,8 @@ static int nousb;	/* Disable USB when built into kernel image */
  * Don't call this function unless you are bound to one of the interfaces
  * on this device or you have locked the device!
  */
-struct usb_interface *usb_ifnum_to_if(struct usb_device *dev, unsigned ifnum)
+struct usb_interface *usb_ifnum_to_if(const struct usb_device *dev,
+				      unsigned ifnum)
 {
 	struct usb_host_config *config = dev->actconfig;
 	int i;
@@ -100,8 +101,8 @@ struct usb_interface *usb_ifnum_to_if(struct usb_device *dev, unsigned ifnum)
  * Don't call this function unless you are bound to the intf interface
  * or you have locked the device!
  */
-struct usb_host_interface *usb_altnum_to_altsetting(struct usb_interface *intf,
-		unsigned int altnum)
+struct usb_host_interface *usb_altnum_to_altsetting(const struct usb_interface *intf,
+						    unsigned int altnum)
 {
 	int i;
 
@@ -356,7 +357,7 @@ void usb_put_intf(struct usb_interface *intf)
  * case the driver already owns the device lock.)
  */
 int usb_lock_device_for_reset(struct usb_device *udev,
-		struct usb_interface *iface)
+			      const struct usb_interface *iface)
 {
 	unsigned long jiffies_expire = jiffies + HZ;
 
@@ -852,8 +853,8 @@ void usb_buffer_unmap (struct urb *urb)
  *
  * Reverse the effect of this call with usb_buffer_unmap_sg().
  */
-int usb_buffer_map_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int nents)
+int usb_buffer_map_sg(const struct usb_device *dev, unsigned pipe,
+		      struct scatterlist *sg, int nents)
 {
 	struct usb_bus		*bus;
 	struct device		*controller;
@@ -887,8 +888,8 @@ int usb_buffer_map_sg (struct usb_device *dev, unsigned pipe,
  * Use this when you are re-using a scatterlist's data buffers for
  * another USB request.
  */
-void usb_buffer_dmasync_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents)
+void usb_buffer_dmasync_sg(const struct usb_device *dev, unsigned pipe,
+			   struct scatterlist *sg, int n_hw_ents)
 {
 	struct usb_bus		*bus;
 	struct device		*controller;
@@ -913,8 +914,8 @@ void usb_buffer_dmasync_sg (struct usb_device *dev, unsigned pipe,
  *
  * Reverses the effect of usb_buffer_map_sg().
  */
-void usb_buffer_unmap_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents)
+void usb_buffer_unmap_sg(const struct usb_device *dev, unsigned pipe,
+			 struct scatterlist *sg, int n_hw_ents)
 {
 	struct usb_bus		*bus;
 	struct device		*controller;
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 6096ead2758c..67da6d0b316f 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -53,7 +53,7 @@ extern struct usb_device_driver usb_generic_driver;
  * no such thing as a platform USB device, so we can steal the use
  * of the platform_data field. */
 
-static inline int is_usb_device(struct device *dev)
+static inline int is_usb_device(const struct device *dev)
 {
 	return dev->platform_data == &usb_generic_driver;
 }
@@ -78,7 +78,7 @@ static inline void mark_quiesced(struct usb_interface *f)
 	f->is_active = 0;
 }
 
-static inline int is_active(struct usb_interface *f)
+static inline int is_active(const struct usb_interface *f)
 {
 	return f->is_active;
 }
diff --git a/include/linux/usb.h b/include/linux/usb.h
index f807479ef65b..26d8a5f36896 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -387,7 +387,7 @@ extern void usb_put_dev(struct usb_device *dev);
 #define usb_unlock_device(udev)		up(&(udev)->dev.sem)
 #define usb_trylock_device(udev)	down_trylock(&(udev)->dev.sem)
 extern int usb_lock_device_for_reset(struct usb_device *udev,
-		struct usb_interface *iface);
+				     const struct usb_interface *iface);
 
 /* USB port reset for device reinitialization */
 extern int usb_reset_device(struct usb_device *dev);
@@ -426,10 +426,10 @@ const struct usb_device_id *usb_match_id(struct usb_interface *interface,
 
 extern struct usb_interface *usb_find_interface(struct usb_driver *drv,
 		int minor);
-extern struct usb_interface *usb_ifnum_to_if(struct usb_device *dev,
+extern struct usb_interface *usb_ifnum_to_if(const struct usb_device *dev,
 		unsigned ifnum);
 extern struct usb_host_interface *usb_altnum_to_altsetting(
-		struct usb_interface *intf, unsigned int altnum);
+		const struct usb_interface *intf, unsigned int altnum);
 
 
 /**
@@ -1064,14 +1064,14 @@ void usb_buffer_unmap (struct urb *urb);
 #endif
 
 struct scatterlist;
-int usb_buffer_map_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int nents);
+int usb_buffer_map_sg(const struct usb_device *dev, unsigned pipe,
+		      struct scatterlist *sg, int nents);
 #if 0
-void usb_buffer_dmasync_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents);
+void usb_buffer_dmasync_sg(const struct usb_device *dev, unsigned pipe,
+			   struct scatterlist *sg, int n_hw_ents);
 #endif
-void usb_buffer_unmap_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents);
+void usb_buffer_unmap_sg(const struct usb_device *dev, unsigned pipe,
+			 struct scatterlist *sg, int n_hw_ents);
 
 /*-------------------------------------------------------------------*
  *                         SYNCHRONOUS CALL SUPPORT                  *
-- 
cgit v1.2.3


From 088dc270e1da03744d977cbd9edd4311af142348 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 21 Aug 2006 12:08:19 -0400
Subject: usbcore: help drivers to change device configs

It's generally a bad idea for USB interface drivers to try to change a
device's configuration, and usbcore doesn't provide any way for them
to do it.  However in a few exceptional circumstances it can make
sense.  This patch (as767) adds a roundabout mechanism to help drivers
that may need it.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/message.c | 59 ++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb.h        |  3 +++
 2 files changed, 62 insertions(+)

(limited to 'include')

diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 49cfd7928a1c..333b22c68aa4 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -1493,6 +1493,65 @@ free_interfaces:
 	return 0;
 }
 
+struct set_config_request {
+	struct usb_device	*udev;
+	int			config;
+	struct work_struct	work;
+};
+
+/* Worker routine for usb_driver_set_configuration() */
+static void driver_set_config_work(void *_req)
+{
+	struct set_config_request *req = _req;
+
+	usb_lock_device(req->udev);
+	usb_set_configuration(req->udev, req->config);
+	usb_unlock_device(req->udev);
+	usb_put_dev(req->udev);
+	kfree(req);
+}
+
+/**
+ * usb_driver_set_configuration - Provide a way for drivers to change device configurations
+ * @udev: the device whose configuration is being updated
+ * @config: the configuration being chosen.
+ * Context: In process context, must be able to sleep
+ *
+ * Device interface drivers are not allowed to change device configurations.
+ * This is because changing configurations will destroy the interface the
+ * driver is bound to and create new ones; it would be like a floppy-disk
+ * driver telling the computer to replace the floppy-disk drive with a
+ * tape drive!
+ *
+ * Still, in certain specialized circumstances the need may arise.  This
+ * routine gets around the normal restrictions by using a work thread to
+ * submit the change-config request.
+ *
+ * Returns 0 if the request was succesfully queued, error code otherwise.
+ * The caller has no way to know whether the queued request will eventually
+ * succeed.
+ */
+int usb_driver_set_configuration(struct usb_device *udev, int config)
+{
+	struct set_config_request *req;
+
+	req = kmalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+	req->udev = udev;
+	req->config = config;
+	INIT_WORK(&req->work, driver_set_config_work, req);
+
+	usb_get_dev(udev);
+	if (!schedule_work(&req->work)) {
+		usb_put_dev(udev);
+		kfree(req);
+		return -EINVAL;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(usb_driver_set_configuration);
+
 // synchronous request completion model
 EXPORT_SYMBOL(usb_control_msg);
 EXPORT_SYMBOL(usb_bulk_msg);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 26d8a5f36896..f104efa04d79 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1099,6 +1099,9 @@ extern int usb_clear_halt(struct usb_device *dev, int pipe);
 extern int usb_reset_configuration(struct usb_device *dev);
 extern int usb_set_interface(struct usb_device *dev, int ifnum, int alternate);
 
+/* this request isn't really synchronous, but it belongs with the others */
+extern int usb_driver_set_configuration(struct usb_device *udev, int config);
+
 /*
  * timeouts, in milliseconds, used for sending/receiving control messages
  * they typically complete within a few frames (msec) after they're issued
-- 
cgit v1.2.3


From a6d2bb9ff919b4685bd684620ec7a1ffa8bf2349 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:27:36 -0400
Subject: USB: remove struct usb_operations

All of the currently-supported USB host controller drivers use the HCD
bus-glue framework.  As part of the program for flattening out the glue
layer, this patch (as769) removes the usb_operations structure.  All
function calls now go directly to the HCD routines (slightly renamed
to remain within the "usb_" namespace).

The patch also removes usb_alloc_bus(), because it's not useful in the
HCD framework and it wasn't referenced anywhere.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c     | 49 +++++-----------------------------------------
 drivers/usb/core/hcd.h     | 32 +++++++-----------------------
 drivers/usb/core/message.c |  4 ++--
 drivers/usb/core/urb.c     | 13 +++++-------
 drivers/usb/core/usb.c     | 10 +++++-----
 include/linux/usb.h        |  4 ----
 6 files changed, 24 insertions(+), 88 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index ea20a3a5a9b9..2102c4deec1e 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -731,30 +731,6 @@ static void usb_bus_init (struct usb_bus *bus)
 	kref_init(&bus->kref);
 }
 
-/**
- * usb_alloc_bus - creates a new USB host controller structure
- * @op: pointer to a struct usb_operations that this bus structure should use
- * Context: !in_interrupt()
- *
- * Creates a USB host controller bus structure with the specified 
- * usb_operations and initializes all the necessary internal objects.
- *
- * If no memory is available, NULL is returned.
- *
- * The caller should call usb_put_bus() when it is finished with the structure.
- */
-struct usb_bus *usb_alloc_bus (struct usb_operations *op)
-{
-	struct usb_bus *bus;
-
-	bus = kzalloc (sizeof *bus, GFP_KERNEL);
-	if (!bus)
-		return NULL;
-	usb_bus_init (bus);
-	bus->op = op;
-	return bus;
-}
-
 /*-------------------------------------------------------------------------*/
 
 /**
@@ -1102,7 +1078,7 @@ static void urb_unlink (struct urb *urb)
  * expects usb_submit_urb() to have sanity checked and conditioned all
  * inputs in the urb
  */
-static int hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
+int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
 {
 	int			status;
 	struct usb_hcd		*hcd = urb->dev->bus->hcpriv;
@@ -1211,7 +1187,7 @@ done:
 /*-------------------------------------------------------------------------*/
 
 /* called in any context */
-static int hcd_get_frame_number (struct usb_device *udev)
+int usb_hcd_get_frame_number (struct usb_device *udev)
 {
 	struct usb_hcd	*hcd = (struct usb_hcd *)udev->bus->hcpriv;
 	if (!HC_IS_RUNNING (hcd->state))
@@ -1253,7 +1229,7 @@ unlink1 (struct usb_hcd *hcd, struct urb *urb)
  * caller guarantees urb won't be recycled till both unlink()
  * and the urb's completion function return
  */
-static int hcd_unlink_urb (struct urb *urb, int status)
+int usb_hcd_unlink_urb (struct urb *urb, int status)
 {
 	struct usb_host_endpoint	*ep;
 	struct usb_hcd			*hcd = NULL;
@@ -1351,8 +1327,8 @@ done:
  * example:  a qh stored in ep->hcpriv, holding state related to endpoint
  * type, maxpacket size, toggle, halt status, and scheduling.
  */
-static void
-hcd_endpoint_disable (struct usb_device *udev, struct usb_host_endpoint *ep)
+void usb_hcd_endpoint_disable (struct usb_device *udev,
+		struct usb_host_endpoint *ep)
 {
 	struct usb_hcd		*hcd;
 	struct urb		*urb;
@@ -1589,20 +1565,6 @@ EXPORT_SYMBOL (usb_bus_start_enum);
 
 /*-------------------------------------------------------------------------*/
 
-/*
- * usb_hcd_operations - adapts usb_bus framework to HCD framework (bus glue)
- */
-static struct usb_operations usb_hcd_operations = {
-	.get_frame_number =	hcd_get_frame_number,
-	.submit_urb =		hcd_submit_urb,
-	.unlink_urb =		hcd_unlink_urb,
-	.buffer_alloc =		hcd_buffer_alloc,
-	.buffer_free =		hcd_buffer_free,
-	.disable =		hcd_endpoint_disable,
-};
-
-/*-------------------------------------------------------------------------*/
-
 /**
  * usb_hcd_giveback_urb - return URB from HCD to device driver
  * @hcd: host controller returning the URB
@@ -1744,7 +1706,6 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 	dev_set_drvdata(dev, hcd);
 
 	usb_bus_init(&hcd->self);
-	hcd->self.op = &usb_hcd_operations;
 	hcd->self.hcpriv = hcd;
 	hcd->self.release = &hcd_release;
 	hcd->self.controller = dev;
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index fc71a08a1af4..83e229914797 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -139,28 +139,6 @@ struct hcd_timeout {	/* timeouts we allocate */
 
 /*-------------------------------------------------------------------------*/
 
-/*
- * FIXME usb_operations should vanish or become hc_driver,
- * when usb_bus and usb_hcd become the same thing.
- */
-
-struct usb_operations {
-	int (*get_frame_number) (struct usb_device *usb_dev);
-	int (*submit_urb) (struct urb *urb, gfp_t mem_flags);
-	int (*unlink_urb) (struct urb *urb, int status);
-
-	/* allocate dma-consistent buffer for URB_DMA_NOMAPPING */
-	void *(*buffer_alloc)(struct usb_bus *bus, size_t size,
-			gfp_t mem_flags,
-			dma_addr_t *dma);
-	void (*buffer_free)(struct usb_bus *bus, size_t size,
-			void *addr, dma_addr_t dma);
-
-	void (*disable)(struct usb_device *udev,
-			struct usb_host_endpoint *ep);
-};
-
-/* each driver provides one of these, and hardware init support */
 
 struct pt_regs;
 
@@ -222,7 +200,13 @@ struct hc_driver {
 		/* Needed only if port-change IRQs are level-triggered */
 };
 
-extern void usb_hcd_giveback_urb (struct usb_hcd *hcd, struct urb *urb, struct pt_regs *regs);
+extern int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags);
+extern int usb_hcd_unlink_urb (struct urb *urb, int status);
+extern void usb_hcd_giveback_urb (struct usb_hcd *hcd, struct urb *urb,
+		struct pt_regs *regs);
+extern void usb_hcd_endpoint_disable (struct usb_device *udev,
+		struct usb_host_endpoint *ep);
+extern int usb_hcd_get_frame_number (struct usb_device *udev);
 
 extern struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 		struct device *dev, char *bus_name);
@@ -361,8 +345,6 @@ extern long usb_calc_bus_time (int speed, int is_input,
 
 /*-------------------------------------------------------------------------*/
 
-extern struct usb_bus *usb_alloc_bus (struct usb_operations *);
-
 extern void usb_set_device_state(struct usb_device *udev,
 		enum usb_device_state new_state);
 
diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 333b22c68aa4..1580c81a0db7 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -984,8 +984,8 @@ void usb_disable_endpoint(struct usb_device *dev, unsigned int epaddr)
 		ep = dev->ep_in[epnum];
 		dev->ep_in[epnum] = NULL;
 	}
-	if (ep && dev->bus && dev->bus->op && dev->bus->op->disable)
-		dev->bus->op->disable(dev, ep);
+	if (ep && dev->bus)
+		usb_hcd_endpoint_disable(dev, ep);
 }
 
 /**
diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c
index 9864988377c7..576919927f53 100644
--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -221,7 +221,6 @@ int usb_submit_urb(struct urb *urb, gfp_t mem_flags)
 {
 	int			pipe, temp, max;
 	struct usb_device	*dev;
-	struct usb_operations	*op;
 	int			is_out;
 
 	if (!urb || urb->hcpriv || !urb->complete)
@@ -233,8 +232,6 @@ int usb_submit_urb(struct urb *urb, gfp_t mem_flags)
 	if (dev->bus->controller->power.power_state.event != PM_EVENT_ON
 			|| dev->state == USB_STATE_SUSPENDED)
 		return -EHOSTUNREACH;
-	if (!(op = dev->bus->op) || !op->submit_urb)
-		return -ENODEV;
 
 	urb->status = -EINPROGRESS;
 	urb->actual_length = 0;
@@ -376,7 +373,7 @@ int usb_submit_urb(struct urb *urb, gfp_t mem_flags)
 		urb->interval = temp;
 	}
 
-	return op->submit_urb (urb, mem_flags);
+	return usb_hcd_submit_urb (urb, mem_flags);
 }
 
 /*-------------------------------------------------------------------*/
@@ -440,9 +437,9 @@ int usb_unlink_urb(struct urb *urb)
 {
 	if (!urb)
 		return -EINVAL;
-	if (!(urb->dev && urb->dev->bus && urb->dev->bus->op))
+	if (!(urb->dev && urb->dev->bus))
 		return -ENODEV;
-	return urb->dev->bus->op->unlink_urb(urb, -ECONNRESET);
+	return usb_hcd_unlink_urb(urb, -ECONNRESET);
 }
 
 /**
@@ -468,13 +465,13 @@ int usb_unlink_urb(struct urb *urb)
 void usb_kill_urb(struct urb *urb)
 {
 	might_sleep();
-	if (!(urb && urb->dev && urb->dev->bus && urb->dev->bus->op))
+	if (!(urb && urb->dev && urb->dev->bus))
 		return;
 	spin_lock_irq(&urb->lock);
 	++urb->reject;
 	spin_unlock_irq(&urb->lock);
 
-	urb->dev->bus->op->unlink_urb(urb, -ENOENT);
+	usb_hcd_unlink_urb(urb, -ENOENT);
 	wait_event(usb_kill_urb_queue, atomic_read(&urb->use_count) == 0);
 
 	spin_lock_irq(&urb->lock);
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 4eb98eb3804f..7ab9d29215f8 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -480,7 +480,7 @@ exit:
  */
 int usb_get_current_frame_number(struct usb_device *dev)
 {
-	return dev->bus->op->get_frame_number (dev);
+	return usb_hcd_get_frame_number (dev);
 }
 
 /**
@@ -677,9 +677,9 @@ void *usb_buffer_alloc (
 	dma_addr_t *dma
 )
 {
-	if (!dev || !dev->bus || !dev->bus->op || !dev->bus->op->buffer_alloc)
+	if (!dev || !dev->bus)
 		return NULL;
-	return dev->bus->op->buffer_alloc (dev->bus, size, mem_flags, dma);
+	return hcd_buffer_alloc (dev->bus, size, mem_flags, dma);
 }
 
 /**
@@ -700,11 +700,11 @@ void usb_buffer_free (
 	dma_addr_t dma
 )
 {
-	if (!dev || !dev->bus || !dev->bus->op || !dev->bus->op->buffer_free)
+	if (!dev || !dev->bus)
 		return;
 	if (!addr)
 		return;
-	dev->bus->op->buffer_free (dev->bus, size, addr, dma);
+	hcd_buffer_free (dev->bus, size, addr, dma);
 }
 
 /**
diff --git a/include/linux/usb.h b/include/linux/usb.h
index f104efa04d79..4709033f8fa7 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -257,8 +257,6 @@ int __usb_get_extra_descriptor(char *buffer, unsigned size,
 
 /* ----------------------------------------------------------------------- */
 
-struct usb_operations;
-
 /* USB device number allocation bitmap */
 struct usb_devmap {
 	unsigned long devicemap[128 / (8*sizeof(unsigned long))];
@@ -279,7 +277,6 @@ struct usb_bus {
 					 * round-robin allocation */
 
 	struct usb_devmap devmap;	/* device address allocation map */
-	struct usb_operations *op;	/* Operations (specific to the HC) */
 	struct usb_device *root_hub;	/* Root hub */
 	struct list_head bus_list;	/* list of busses */
 	void *hcpriv;                   /* Host Controller private data */
@@ -1051,7 +1048,6 @@ extern int usb_submit_urb(struct urb *urb, gfp_t mem_flags);
 extern int usb_unlink_urb(struct urb *urb);
 extern void usb_kill_urb(struct urb *urb);
 
-#define HAVE_USB_BUFFERS
 void *usb_buffer_alloc (struct usb_device *dev, size_t size,
 	gfp_t mem_flags, dma_addr_t *dma);
 void usb_buffer_free (struct usb_device *dev, size_t size,
-- 
cgit v1.2.3


From dd990f16a39d4e615c0b70a0ab50b79b32bfb16d Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:29:56 -0400
Subject: usbcore: Add flag for whether a host controller uses DMA

This patch (as770b) introduces a new field to usb_bus: a flag
indicating whether or not the host controller uses DMA.  This serves
to encapsulate the computation.  It also means we will have only one
spot to update if the DMA API changes.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c | 8 +++++---
 include/linux/usb.h    | 1 +
 2 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 2102c4deec1e..0cc14206920a 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1152,7 +1152,7 @@ doit:
 	/* lower level hcd code should use *_dma exclusively,
 	 * unless it uses pio or talks to another transport.
 	 */
-	if (hcd->self.controller->dma_mask) {
+	if (hcd->self.uses_dma) {
 		if (usb_pipecontrol (urb->pipe)
 			&& !(urb->transfer_flags & URB_NO_SETUP_DMA_MAP))
 			urb->setup_dma = dma_map_single (
@@ -1585,8 +1585,9 @@ void usb_hcd_giveback_urb (struct usb_hcd *hcd, struct urb *urb, struct pt_regs
 	at_root_hub = (urb->dev == hcd->self.root_hub);
 	urb_unlink (urb);
 
-	/* lower level hcd code should use *_dma exclusively */
-	if (hcd->self.controller->dma_mask && !at_root_hub) {
+	/* lower level hcd code should use *_dma exclusively if the
+	 * host controller does DMA */
+	if (hcd->self.uses_dma && !at_root_hub) {
 		if (usb_pipecontrol (urb->pipe)
 			&& !(urb->transfer_flags & URB_NO_SETUP_DMA_MAP))
 			dma_unmap_single (hcd->self.controller, urb->setup_dma,
@@ -1710,6 +1711,7 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 	hcd->self.release = &hcd_release;
 	hcd->self.controller = dev;
 	hcd->self.bus_name = bus_name;
+	hcd->self.uses_dma = (dev->dma_mask != NULL);
 
 	init_timer(&hcd->rh_timer);
 	hcd->rh_timer.function = rh_timer_func;
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 4709033f8fa7..09661759621f 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -269,6 +269,7 @@ struct usb_bus {
 	struct device *controller;	/* host/master side hardware */
 	int busnum;			/* Bus number (in order of reg) */
 	char *bus_name;			/* stable id (PCI slot_name etc) */
+	u8 uses_dma;			/* Does the host controller use DMA? */
 	u8 otg_port;			/* 0, or number of OTG/HNP port */
 	unsigned is_b_host:1;		/* true during some HNP roleswitches */
 	unsigned b_hnp_enable:1;	/* OTG: did A-Host enable HNP? */
-- 
cgit v1.2.3


From 1720058343fa43a1a25bfad9e62ea06e7e9743b6 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:32:52 -0400
Subject: usbcore: trim down usb_bus structure

As part of the ongoing program to flatten out the HCD bus-glue layer,
this patch (as771b) eliminates the hcpriv, release, and kref fields
from struct usb_bus.  hcpriv and release were not being used for
anything worthwhile, and kref has been moved into the enclosing
usb_hcd structure.

Along with those changes, the patch gets rid of usb_bus_get and
usb_bus_put, replacing them with usb_get_hcd and usb_put_hcd.

The one interesting aspect is that the dev_set_drvdata call was
removed from usb_put_hcd, where it clearly doesn't belong.  This means
the driver private data won't get reset to NULL.  It shouldn't cause
any problems, since the private data is undefined when no driver is
bound.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/buffer.c      |  4 +--
 drivers/usb/core/hcd.c         | 67 +++++++++++++++---------------------------
 drivers/usb/core/hcd.h         | 13 ++++----
 drivers/usb/core/usb.c         |  5 ++--
 drivers/usb/gadget/dummy_hcd.c |  8 ++---
 drivers/usb/host/ehci-dbg.c    |  6 ++--
 drivers/usb/host/ohci-dbg.c    |  6 ++--
 drivers/usb/mon/mon_main.c     |  6 ++--
 include/linux/usb.h            |  3 --
 9 files changed, 44 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/buffer.c b/drivers/usb/core/buffer.c
index f4f4ef0f377a..840442a25b61 100644
--- a/drivers/usb/core/buffer.c
+++ b/drivers/usb/core/buffer.c
@@ -104,7 +104,7 @@ void *hcd_buffer_alloc (
 	dma_addr_t		*dma
 )
 {
-	struct usb_hcd		*hcd = bus->hcpriv;
+	struct usb_hcd		*hcd = bus_to_hcd(bus);
 	int 			i;
 
 	/* some USB hosts just use PIO */
@@ -127,7 +127,7 @@ void hcd_buffer_free (
 	dma_addr_t		dma
 )
 {
-	struct usb_hcd		*hcd = bus->hcpriv;
+	struct usb_hcd		*hcd = bus_to_hcd(bus);
 	int 			i;
 
 	if (!addr)
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 0cc14206920a..9dfc812de034 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -664,31 +664,6 @@ static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 
 /*-------------------------------------------------------------------------*/
 
-/* exported only within usbcore */
-struct usb_bus *usb_bus_get(struct usb_bus *bus)
-{
-	if (bus)
-		kref_get(&bus->kref);
-	return bus;
-}
-
-static void usb_host_release(struct kref *kref)
-{
-	struct usb_bus *bus = container_of(kref, struct usb_bus, kref);
-
-	if (bus->release)
-		bus->release(bus);
-}
-
-/* exported only within usbcore */
-void usb_bus_put(struct usb_bus *bus)
-{
-	if (bus)
-		kref_put(&bus->kref, usb_host_release);
-}
-
-/*-------------------------------------------------------------------------*/
-
 static struct class *usb_host_class;
 
 int usb_host_init(void)
@@ -720,15 +695,12 @@ static void usb_bus_init (struct usb_bus *bus)
 	bus->devnum_next = 1;
 
 	bus->root_hub = NULL;
-	bus->hcpriv = NULL;
 	bus->busnum = -1;
 	bus->bandwidth_allocated = 0;
 	bus->bandwidth_int_reqs  = 0;
 	bus->bandwidth_isoc_reqs = 0;
 
 	INIT_LIST_HEAD (&bus->bus_list);
-
-	kref_init(&bus->kref);
 }
 
 /*-------------------------------------------------------------------------*/
@@ -1081,7 +1053,7 @@ static void urb_unlink (struct urb *urb)
 int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
 {
 	int			status;
-	struct usb_hcd		*hcd = urb->dev->bus->hcpriv;
+	struct usb_hcd		*hcd = bus_to_hcd(urb->dev->bus);
 	struct usb_host_endpoint *ep;
 	unsigned long		flags;
 
@@ -1189,7 +1161,8 @@ done:
 /* called in any context */
 int usb_hcd_get_frame_number (struct usb_device *udev)
 {
-	struct usb_hcd	*hcd = (struct usb_hcd *)udev->bus->hcpriv;
+	struct usb_hcd	*hcd = bus_to_hcd(udev->bus);
+
 	if (!HC_IS_RUNNING (hcd->state))
 		return -ESHUTDOWN;
 	return hcd->driver->get_frame_number (hcd);
@@ -1262,7 +1235,7 @@ int usb_hcd_unlink_urb (struct urb *urb, int status)
 	spin_lock (&hcd_data_lock);
 
 	sys = &urb->dev->dev;
-	hcd = urb->dev->bus->hcpriv;
+	hcd = bus_to_hcd(urb->dev->bus);
 	if (hcd == NULL) {
 		retval = -ENODEV;
 		goto done;
@@ -1333,7 +1306,7 @@ void usb_hcd_endpoint_disable (struct usb_device *udev,
 	struct usb_hcd		*hcd;
 	struct urb		*urb;
 
-	hcd = udev->bus->hcpriv;
+	hcd = bus_to_hcd(udev->bus);
 
 	WARN_ON (!HC_IS_RUNNING (hcd->state) && hcd->state != HC_STATE_HALT &&
 			udev->state != USB_STATE_NOTATTACHED);
@@ -1673,14 +1646,6 @@ EXPORT_SYMBOL_GPL (usb_hc_died);
 
 /*-------------------------------------------------------------------------*/
 
-static void hcd_release (struct usb_bus *bus)
-{
-	struct usb_hcd *hcd;
-
-	hcd = container_of(bus, struct usb_hcd, self);
-	kfree(hcd);
-}
-
 /**
  * usb_create_hcd - create and initialize an HCD structure
  * @driver: HC driver that will use this hcd
@@ -1705,10 +1670,9 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 		return NULL;
 	}
 	dev_set_drvdata(dev, hcd);
+	kref_init(&hcd->kref);
 
 	usb_bus_init(&hcd->self);
-	hcd->self.hcpriv = hcd;
-	hcd->self.release = &hcd_release;
 	hcd->self.controller = dev;
 	hcd->self.bus_name = bus_name;
 	hcd->self.uses_dma = (dev->dma_mask != NULL);
@@ -1725,10 +1689,25 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 }
 EXPORT_SYMBOL (usb_create_hcd);
 
+static void hcd_release (struct kref *kref)
+{
+	struct usb_hcd *hcd = container_of (kref, struct usb_hcd, kref);
+
+	kfree(hcd);
+}
+
+struct usb_hcd *usb_get_hcd (struct usb_hcd *hcd)
+{
+	if (hcd)
+		kref_get (&hcd->kref);
+	return hcd;
+}
+EXPORT_SYMBOL (usb_get_hcd);
+
 void usb_put_hcd (struct usb_hcd *hcd)
 {
-	dev_set_drvdata(hcd->self.controller, NULL);
-	usb_bus_put(&hcd->self);
+	if (hcd)
+		kref_put (&hcd->kref, hcd_release);
 }
 EXPORT_SYMBOL (usb_put_hcd);
 
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index 83e229914797..7a2bcba2ae61 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -55,12 +55,13 @@
 
 /*-------------------------------------------------------------------------*/
 
-struct usb_hcd {	/* usb_bus.hcpriv points to this */
+struct usb_hcd {
 
 	/*
 	 * housekeeping
 	 */
 	struct usb_bus		self;		/* hcd is-a bus */
+	struct kref		kref;		/* reference counter */
 
 	const char		*product_desc;	/* product/vendor string */
 	char			irq_descr[24];	/* driver + bus # */
@@ -129,8 +130,10 @@ static inline struct usb_bus *hcd_to_bus (struct usb_hcd *hcd)
 	return &hcd->self;
 }
 
-
-// urb.hcpriv is really hardware-specific
+static inline struct usb_hcd *bus_to_hcd (struct usb_bus *bus)
+{
+	return container_of(bus, struct usb_hcd, self);
+}
 
 struct hcd_timeout {	/* timeouts we allocate */
 	struct list_head	timeout_list;
@@ -210,6 +213,7 @@ extern int usb_hcd_get_frame_number (struct usb_device *udev);
 
 extern struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 		struct device *dev, char *bus_name);
+extern struct usb_hcd *usb_get_hcd (struct usb_hcd *hcd);
 extern void usb_put_hcd (struct usb_hcd *hcd);
 extern int usb_add_hcd(struct usb_hcd *hcd,
 		unsigned int irqnum, unsigned long irqflags);
@@ -356,9 +360,6 @@ extern struct list_head usb_bus_list;
 extern struct mutex usb_bus_list_lock;
 extern wait_queue_head_t usb_kill_urb_queue;
 
-extern struct usb_bus *usb_bus_get (struct usb_bus *bus);
-extern void usb_bus_put (struct usb_bus *bus);
-
 extern void usb_enable_root_hub_irq (struct usb_bus *bus);
 
 extern int usb_find_interface_driver (struct usb_device *dev,
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 7ab9d29215f8..b0c0a993338f 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -169,7 +169,7 @@ static void usb_release_dev(struct device *dev)
 	udev = to_usb_device(dev);
 
 	usb_destroy_configuration(udev);
-	usb_bus_put(udev->bus);
+	usb_put_hcd(bus_to_hcd(udev->bus));
 	kfree(udev->product);
 	kfree(udev->manufacturer);
 	kfree(udev->serial);
@@ -197,8 +197,7 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	if (!dev)
 		return NULL;
 
-	bus = usb_bus_get(bus);
-	if (!bus) {
+	if (!usb_get_hcd(bus_to_hcd(bus))) {
 		kfree(dev);
 		return NULL;
 	}
diff --git a/drivers/usb/gadget/dummy_hcd.c b/drivers/usb/gadget/dummy_hcd.c
index 7d1c22c34957..fdab97a27c08 100644
--- a/drivers/usb/gadget/dummy_hcd.c
+++ b/drivers/usb/gadget/dummy_hcd.c
@@ -889,11 +889,9 @@ EXPORT_SYMBOL (net2280_set_fifo_mode);
 static void
 dummy_gadget_release (struct device *dev)
 {
-#if 0		/* usb_bus_put isn't EXPORTed! */
 	struct dummy	*dum = gadget_dev_to_dummy (dev);
 
-	usb_bus_put (&dummy_to_hcd (dum)->self);
-#endif
+	usb_put_hcd (dummy_to_hcd (dum));
 }
 
 static int dummy_udc_probe (struct platform_device *pdev)
@@ -915,9 +913,7 @@ static int dummy_udc_probe (struct platform_device *pdev)
 	if (rc < 0)
 		return rc;
 
-#if 0		/* usb_bus_get isn't EXPORTed! */
-	usb_bus_get (&dummy_to_hcd (dum)->self);
-#endif
+	usb_get_hcd (dummy_to_hcd (dum));
 
 	platform_set_drvdata (pdev, dum);
 	device_create_file (&dum->gadget.dev, &dev_attr_function);
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index 65ac9fef3a7c..215ce6d06394 100644
--- a/drivers/usb/host/ehci-dbg.c
+++ b/drivers/usb/host/ehci-dbg.c
@@ -451,7 +451,7 @@ show_async (struct class_device *class_dev, char *buf)
 	*buf = 0;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
 	size = PAGE_SIZE;
@@ -497,7 +497,7 @@ show_periodic (struct class_device *class_dev, char *buf)
 	seen_count = 0;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
 	size = PAGE_SIZE;
@@ -634,7 +634,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	static char		label [] = "";
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
 	size = PAGE_SIZE;
diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c
index da52609a9290..534d07dcb824 100644
--- a/drivers/usb/host/ohci-dbg.c
+++ b/drivers/usb/host/ohci-dbg.c
@@ -477,7 +477,7 @@ show_async (struct class_device *class_dev, char *buf)
 	unsigned long		flags;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ohci = hcd_to_ohci(hcd);
 
 	/* display control and bulk lists together, for simplicity */
@@ -510,7 +510,7 @@ show_periodic (struct class_device *class_dev, char *buf)
 	seen_count = 0;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ohci = hcd_to_ohci(hcd);
 	next = buf;
 	size = PAGE_SIZE;
@@ -607,7 +607,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	u32			rdata;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ohci = hcd_to_ohci(hcd);
 	regs = ohci->regs;
 	next = buf;
diff --git a/drivers/usb/mon/mon_main.c b/drivers/usb/mon/mon_main.c
index 275a66f83058..e0ed36cdfd8b 100644
--- a/drivers/usb/mon/mon_main.c
+++ b/drivers/usb/mon/mon_main.c
@@ -265,7 +265,6 @@ static void mon_dissolve(struct mon_bus *mbus, struct usb_bus *ubus)
 	ubus->mon_bus = NULL;
 	mbus->u_bus = NULL;
 	mb();
-	// usb_bus_put(ubus);
 }
 
 /*
@@ -297,10 +296,9 @@ static void mon_bus_init(struct dentry *mondir, struct usb_bus *ubus)
 	INIT_LIST_HEAD(&mbus->r_list);
 
 	/*
-	 * This usb_bus_get here is superfluous, because we receive
-	 * a notification if usb_bus is about to be removed.
+	 * We don't need to take a reference to ubus, because we receive
+	 * a notification if the bus is about to be removed.
 	 */
-	// usb_bus_get(ubus);
 	mbus->u_bus = ubus;
 	ubus->mon_bus = mbus;
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 09661759621f..c66303285a45 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -280,7 +280,6 @@ struct usb_bus {
 	struct usb_devmap devmap;	/* device address allocation map */
 	struct usb_device *root_hub;	/* Root hub */
 	struct list_head bus_list;	/* list of busses */
-	void *hcpriv;                   /* Host Controller private data */
 
 	int bandwidth_allocated;	/* on this bus: how much of the time
 					 * reserved for periodic (intr/iso)
@@ -295,8 +294,6 @@ struct usb_bus {
 	struct dentry *usbfs_dentry;	/* usbfs dentry entry for the bus */
 
 	struct class_device *class_dev;	/* class device for this bus */
-	struct kref kref;		/* reference counting for this bus */
-	void (*release)(struct usb_bus *bus);
 
 #if defined(CONFIG_USB_MON)
 	struct mon_bus *mon_bus;	/* non-null when associated */
-- 
cgit v1.2.3


From b6956ffa595db97656d5901ca8fec77ef272d41a Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 15:46:48 -0400
Subject: usbcore: store each usb_device's level in the tree

This patch (as778) adds a field to struct usb_device to store the
device's level in the USB tree.  In itself this number isn't really
important.  But the overhead is very low, and in a later patch it will
be used for preventing bogus warnings from the lockdep checker.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 1 +
 include/linux/usb.h    | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include')

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index f5adce049b35..78e910b2046c 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2414,6 +2414,7 @@ static void hub_port_connect_change(struct usb_hub *hub, int port1,
 		usb_set_device_state(udev, USB_STATE_POWERED);
 		udev->speed = USB_SPEED_UNKNOWN;
  		udev->bus_mA = hub->mA_per_port;
+		udev->level = hdev->level + 1;
 
 		/* set the address */
 		choose_address(udev);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index c66303285a45..df5c93eb3ce9 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -348,6 +348,7 @@ struct usb_device {
 
 	unsigned short bus_mA;		/* Current available from the bus */
 	u8 portnum;			/* Parent port number (origin 1) */
+	u8 level;			/* Number of USB hub ancestors */
 
 	int have_langid;		/* whether string_langid is valid */
 	int string_langid;		/* language ID for strings */
-- 
cgit v1.2.3


From 645daaab0b6adc35c1838df2a82f9d729fdb1767 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 15:47:02 -0400
Subject: usbcore: add autosuspend/autoresume infrastructure

This patch (as739) adds the basic infrastructure for USB autosuspend
and autoresume.  The main features are:

	PM usage counters added to struct usb_device and struct
	usb_interface, indicating whether it's okay to autosuspend
	them or they are currently in use.

	Flag added to usb_device indicating whether the current
	suspend/resume operation originated from outside or as an
	autosuspend/autoresume.

	Flag added to usb_driver indicating whether the driver
	supports autosuspend.  If not, no device bound to the driver
	will be autosuspended.

	Mutex added to usb_device for protecting PM operations.
	Unlike the device semaphore, the locking rule for the pm_mutex
	is that you must acquire the locks going _up_ the device tree.

	New routines handling autosuspend/autoresume requests for
	interfaces and devices.

	Suspend and resume requests are propagated up the device tree
	(but not outside the USB subsystem).

	work_struct added to usb_device, for carrying out delayed
	autosuspend requests.

	Autoresume added (and autosuspend prevented) during probe and
	disconnect.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 419 ++++++++++++++++++++++++++++++++++++++++++----
 drivers/usb/core/hub.c    |  49 ++++--
 drivers/usb/core/usb.c    |  23 +++
 drivers/usb/core/usb.h    |  14 ++
 include/linux/usb.h       |  33 ++++
 5 files changed, 493 insertions(+), 45 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index a5d11461f5a9..2b2000ac05ab 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -157,12 +157,13 @@ static int usb_probe_device(struct device *dev)
 
 	udev = to_usb_device(dev);
 
-	/* FIXME: resume a suspended device */
-	if (udev->state == USB_STATE_SUSPENDED)
-		return -EHOSTUNREACH;
-
 	/* TODO: Add real matching code */
 
+	/* The device should always appear to be in use
+	 * unless the driver suports autosuspend.
+	 */
+	udev->pm_usage_cnt = !(udriver->supports_autosuspend);
+
 	error = udriver->probe(udev);
 	return error;
 }
@@ -182,6 +183,7 @@ static int usb_probe_interface(struct device *dev)
 {
 	struct usb_driver *driver = to_usb_driver(dev->driver);
 	struct usb_interface *intf;
+	struct usb_device *udev;
 	const struct usb_device_id *id;
 	int error = -ENODEV;
 
@@ -191,10 +193,7 @@ static int usb_probe_interface(struct device *dev)
 		return error;
 
 	intf = to_usb_interface(dev);
-
-	/* FIXME we'd much prefer to just resume it ... */
-	if (interface_to_usbdev(intf)->state == USB_STATE_SUSPENDED)
-		return -EHOSTUNREACH;
+	udev = interface_to_usbdev(intf);
 
 	id = usb_match_id(intf, driver->id_table);
 	if (!id)
@@ -202,18 +201,31 @@ static int usb_probe_interface(struct device *dev)
 	if (id) {
 		dev_dbg(dev, "%s - got id\n", __FUNCTION__);
 
+		error = usb_autoresume_device(udev, 1);
+		if (error)
+			return error;
+
 		/* Interface "power state" doesn't correspond to any hardware
 		 * state whatsoever.  We use it to record when it's bound to
 		 * a driver that may start I/0:  it's not frozen/quiesced.
 		 */
 		mark_active(intf);
 		intf->condition = USB_INTERFACE_BINDING;
+
+		/* The interface should always appear to be in use
+		 * unless the driver suports autosuspend.
+		 */
+		intf->pm_usage_cnt = !(driver->supports_autosuspend);
+
 		error = driver->probe(intf, id);
 		if (error) {
 			mark_quiesced(intf);
+			intf->needs_remote_wakeup = 0;
 			intf->condition = USB_INTERFACE_UNBOUND;
 		} else
 			intf->condition = USB_INTERFACE_BOUND;
+
+		usb_autosuspend_device(udev, 1);
 	}
 
 	return error;
@@ -224,9 +236,15 @@ static int usb_unbind_interface(struct device *dev)
 {
 	struct usb_driver *driver = to_usb_driver(dev->driver);
 	struct usb_interface *intf = to_usb_interface(dev);
+	struct usb_device *udev;
+	int error;
 
 	intf->condition = USB_INTERFACE_UNBINDING;
 
+	/* Autoresume for set_interface call below */
+	udev = interface_to_usbdev(intf);
+	error = usb_autoresume_device(udev, 1);
+
 	/* release all urbs for this interface */
 	usb_disable_interface(interface_to_usbdev(intf), intf);
 
@@ -237,8 +255,13 @@ static int usb_unbind_interface(struct device *dev)
 			intf->altsetting[0].desc.bInterfaceNumber,
 			0);
 	usb_set_intfdata(intf, NULL);
+
 	intf->condition = USB_INTERFACE_UNBOUND;
 	mark_quiesced(intf);
+	intf->needs_remote_wakeup = 0;
+
+	if (!error)
+		usb_autosuspend_device(udev, 1);
 
 	return 0;
 }
@@ -267,14 +290,19 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 				struct usb_interface *iface, void* priv)
 {
 	struct device *dev = &iface->dev;
+	struct usb_device *udev = interface_to_usbdev(iface);
 
 	if (dev->driver)
 		return -EBUSY;
 
 	dev->driver = &driver->drvwrap.driver;
 	usb_set_intfdata(iface, priv);
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
 	iface->condition = USB_INTERFACE_BOUND;
 	mark_active(iface);
+	iface->pm_usage_cnt = !(driver->supports_autosuspend);
+	mutex_unlock(&udev->pm_mutex);
 
 	/* if interface was already added, bind now; else let
 	 * the future device_add() bind it, bypassing probe()
@@ -304,6 +332,7 @@ void usb_driver_release_interface(struct usb_driver *driver,
 					struct usb_interface *iface)
 {
 	struct device *dev = &iface->dev;
+	struct usb_device *udev = interface_to_usbdev(iface);
 
 	/* this should never happen, don't release something that's not ours */
 	if (!dev->driver || dev->driver != &driver->drvwrap.driver)
@@ -321,8 +350,12 @@ void usb_driver_release_interface(struct usb_driver *driver,
 
 	dev->driver = NULL;
 	usb_set_intfdata(iface, NULL);
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
 	iface->condition = USB_INTERFACE_UNBOUND;
 	mark_quiesced(iface);
+	iface->needs_remote_wakeup = 0;
+	mutex_unlock(&udev->pm_mutex);
 }
 EXPORT_SYMBOL(usb_driver_release_interface);
 
@@ -751,7 +784,7 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
 
 #ifdef CONFIG_PM
 
-/* Caller has locked udev */
+/* Caller has locked udev->pm_mutex */
 static int suspend_device(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_device_driver	*udriver;
@@ -763,6 +796,7 @@ static int suspend_device(struct usb_device *udev, pm_message_t msg)
 
 	/* For devices that don't have a driver, we do a standard suspend. */
 	if (udev->dev.driver == NULL) {
+		udev->do_remote_wakeup = 0;
 		status = usb_port_suspend(udev);
 		goto done;
 	}
@@ -771,12 +805,13 @@ static int suspend_device(struct usb_device *udev, pm_message_t msg)
 	status = udriver->suspend(udev, msg);
 
 done:
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		udev->dev.power.power_state.event = msg.event;
 	return status;
 }
 
-/* Caller has locked udev */
+/* Caller has locked udev->pm_mutex */
 static int resume_device(struct usb_device *udev)
 {
 	struct usb_device_driver	*udriver;
@@ -796,12 +831,13 @@ static int resume_device(struct usb_device *udev)
 	status = udriver->resume(udev);
 
 done:
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		udev->dev.power.power_state.event = PM_EVENT_ON;
 	return status;
 }
 
-/* Caller has locked intf's usb_device */
+/* Caller has locked intf's usb_device's pm_mutex */
 static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 {
 	struct usb_driver	*driver;
@@ -812,31 +848,33 @@ static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 			!is_active(intf))
 		goto done;
 
-	if (intf->dev.driver == NULL)		/* This can't happen */
+	if (intf->condition == USB_INTERFACE_UNBOUND)	/* This can't happen */
 		goto done;
 	driver = to_usb_driver(intf->dev.driver);
 
 	if (driver->suspend && driver->resume) {
 		status = driver->suspend(intf, msg);
-		if (status)
+		if (status == 0)
+			mark_quiesced(intf);
+		else if (!interface_to_usbdev(intf)->auto_pm)
 			dev_err(&intf->dev, "%s error %d\n",
 					"suspend", status);
-		else
-			mark_quiesced(intf);
 	} else {
 		// FIXME else if there's no suspend method, disconnect...
+		// Not possible if auto_pm is set...
 		dev_warn(&intf->dev, "no suspend for driver %s?\n",
 				driver->name);
 		mark_quiesced(intf);
 	}
 
 done:
+	// dev_dbg(&intf->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		intf->dev.power.power_state.event = msg.event;
 	return status;
 }
 
-/* Caller has locked intf's usb_device */
+/* Caller has locked intf's usb_device's pm_mutex */
 static int resume_interface(struct usb_interface *intf)
 {
 	struct usb_driver	*driver;
@@ -846,8 +884,12 @@ static int resume_interface(struct usb_interface *intf)
 			is_active(intf))
 		goto done;
 
+	/* Don't let autoresume interfere with unbinding */
+	if (intf->condition == USB_INTERFACE_UNBINDING)
+		goto done;
+
 	/* Can't resume it if it doesn't have a driver. */
-	if (intf->dev.driver == NULL) {
+	if (intf->condition == USB_INTERFACE_UNBOUND) {
 		status = -ENOTCONN;
 		goto done;
 	}
@@ -867,18 +909,88 @@ static int resume_interface(struct usb_interface *intf)
 	}
 
 done:
+	// dev_dbg(&intf->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		intf->dev.power.power_state.event = PM_EVENT_ON;
 	return status;
 }
 
-/* Caller has locked udev */
+/**
+ * usb_suspend_both - suspend a USB device and its interfaces
+ * @udev: the usb_device to suspend
+ * @msg: Power Management message describing this state transition
+ *
+ * This is the central routine for suspending USB devices.  It calls the
+ * suspend methods for all the interface drivers in @udev and then calls
+ * the suspend method for @udev itself.  If an error occurs at any stage,
+ * all the interfaces which were suspended are resumed so that they remain
+ * in the same state as the device.
+ *
+ * If an autosuspend is in progress (@udev->auto_pm is set), the routine
+ * checks first to make sure that neither the device itself or any of its
+ * active interfaces is in use (pm_usage_cnt is greater than 0).  If they
+ * are, the autosuspend fails.
+ *
+ * If the suspend succeeds, the routine recursively queues an autosuspend
+ * request for @udev's parent device, thereby propagating the change up
+ * the device tree.  If all of the parent's children are now suspended,
+ * the parent will autosuspend in turn.
+ *
+ * The suspend method calls are subject to mutual exclusion under control
+ * of @udev's pm_mutex.  Many of these calls are also under the protection
+ * of @udev's device lock (including all requests originating outside the
+ * USB subsystem), but autosuspend requests generated by a child device or
+ * interface driver may not be.  Usbcore will insure that the method calls
+ * do not arrive during bind, unbind, or reset operations.  However, drivers
+ * must be prepared to handle suspend calls arriving at unpredictable times.
+ * The only way to block such calls is to do an autoresume (preventing
+ * autosuspends) while holding @udev's device lock (preventing outside
+ * suspends).
+ *
+ * The caller must hold @udev->pm_mutex.
+ *
+ * This routine can run only in process context.
+ */
 int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 {
 	int			status = 0;
 	int			i = 0;
 	struct usb_interface	*intf;
+	struct usb_device	*parent = udev->parent;
+
+	cancel_delayed_work(&udev->autosuspend);
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return 0;
+	if (udev->state == USB_STATE_SUSPENDED)
+		return 0;
 
+	udev->do_remote_wakeup = device_may_wakeup(&udev->dev);
+
+	/* For autosuspend, fail fast if anything is in use.
+	 * Also fail if any interfaces require remote wakeup but it
+	 * isn't available. */
+	if (udev->auto_pm) {
+		if (udev->pm_usage_cnt > 0)
+			return -EBUSY;
+		if (udev->actconfig) {
+			for (; i < udev->actconfig->desc.bNumInterfaces; i++) {
+				intf = udev->actconfig->interface[i];
+				if (!is_active(intf))
+					continue;
+				if (intf->pm_usage_cnt > 0)
+					return -EBUSY;
+				if (intf->needs_remote_wakeup &&
+						!udev->do_remote_wakeup) {
+					dev_dbg(&udev->dev,
+	"remote wakeup needed for autosuspend\n");
+					return -EOPNOTSUPP;
+				}
+			}
+			i = 0;
+		}
+	}
+
+	/* Suspend all the interfaces and then udev itself */
 	if (udev->actconfig) {
 		for (; i < udev->actconfig->desc.bNumInterfaces; i++) {
 			intf = udev->actconfig->interface[i];
@@ -896,40 +1008,282 @@ int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 			intf = udev->actconfig->interface[i];
 			resume_interface(intf);
 		}
-	}
+
+	/* If the suspend succeeded, propagate it up the tree */
+	} else if (parent)
+		usb_autosuspend_device(parent, 0);
+
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
 	return status;
 }
 
-/* Caller has locked udev */
+/**
+ * usb_resume_both - resume a USB device and its interfaces
+ * @udev: the usb_device to resume
+ *
+ * This is the central routine for resuming USB devices.  It calls the
+ * the resume method for @udev and then calls the resume methods for all
+ * the interface drivers in @udev.
+ *
+ * Before starting the resume, the routine calls itself recursively for
+ * the parent device of @udev, thereby propagating the change up the device
+ * tree and assuring that @udev will be able to resume.  If the parent is
+ * unable to resume successfully, the routine fails.
+ *
+ * The resume method calls are subject to mutual exclusion under control
+ * of @udev's pm_mutex.  Many of these calls are also under the protection
+ * of @udev's device lock (including all requests originating outside the
+ * USB subsystem), but autoresume requests generated by a child device or
+ * interface driver may not be.  Usbcore will insure that the method calls
+ * do not arrive during bind, unbind, or reset operations.  However, drivers
+ * must be prepared to handle resume calls arriving at unpredictable times.
+ * The only way to block such calls is to do an autoresume (preventing
+ * other autoresumes) while holding @udev's device lock (preventing outside
+ * resumes).
+ *
+ * The caller must hold @udev->pm_mutex.
+ *
+ * This routine can run only in process context.
+ */
 int usb_resume_both(struct usb_device *udev)
 {
-	int			status;
+	int			status = 0;
 	int			i;
 	struct usb_interface	*intf;
+	struct usb_device	*parent = udev->parent;
+
+	cancel_delayed_work(&udev->autosuspend);
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return -ENODEV;
 
-	/* Can't resume if the parent is suspended */
-	if (udev->parent && udev->parent->state == USB_STATE_SUSPENDED) {
-		dev_warn(&udev->dev, "can't resume; parent is suspended\n");
-		return -EHOSTUNREACH;
+	/* Propagate the resume up the tree, if necessary */
+	if (udev->state == USB_STATE_SUSPENDED) {
+		if (parent) {
+			mutex_lock_nested(&parent->pm_mutex, parent->level);
+			parent->auto_pm = 1;
+			status = usb_resume_both(parent);
+		} else {
+
+			/* We can't progagate beyond the USB subsystem,
+			 * so if a root hub's controller is suspended
+			 * then we're stuck. */
+			if (udev->dev.parent->power.power_state.event !=
+					PM_EVENT_ON)
+				status = -EHOSTUNREACH;
+		}
+		if (status == 0 && udev->state == USB_STATE_SUSPENDED)
+			status = resume_device(udev);
+		if (parent)
+			mutex_unlock(&parent->pm_mutex);
 	}
 
-	status = resume_device(udev);
+	/* Now the parent won't suspend until we are finished */
+
 	if (status == 0 && udev->actconfig) {
 		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
 			intf = udev->actconfig->interface[i];
 			resume_interface(intf);
 		}
 	}
+
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
+	return status;
+}
+
+#ifdef CONFIG_USB_SUSPEND
+
+/**
+ * usb_autosuspend_device - delayed autosuspend of a USB device and its interfaces
+ * @udev - the usb_device to autosuspend
+ * @dec_usage_cnt - flag to decrement @udev's PM-usage counter
+ *
+ * This routine should be called when a core subsystem is finished using
+ * @udev and wants to allow it to autosuspend.  Examples would be when
+ * @udev's device file in usbfs is closed or after a configuration change.
+ *
+ * @dec_usage_cnt should be 1 if the subsystem previously incremented
+ * @udev's usage counter (such as by passing 1 to usb_autoresume_device);
+ * otherwise it should be 0.
+ *
+ * If the usage counter for @udev or any of its active interfaces is greater
+ * than 0, the autosuspend request will not be queued.  (If an interface
+ * driver does not support autosuspend then its usage counter is permanently
+ * positive.)  Likewise, if an interface driver requires remote-wakeup
+ * capability during autosuspend but remote wakeup is disabled, the
+ * autosuspend will fail.
+ *
+ * Often the caller will hold @udev's device lock, but this is not
+ * necessary.
+ *
+ * This routine can run only in process context.
+ */
+void usb_autosuspend_device(struct usb_device *udev, int dec_usage_cnt)
+{
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	udev->pm_usage_cnt -= dec_usage_cnt;
+	if (udev->pm_usage_cnt <= 0)
+		schedule_delayed_work(&udev->autosuspend,
+				USB_AUTOSUSPEND_DELAY);
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&udev->dev, "%s: cnt %d\n",
+	//		__FUNCTION__, udev->pm_usage_cnt);
+}
+
+/**
+ * usb_autoresume_device - immediately autoresume a USB device and its interfaces
+ * @udev - the usb_device to autoresume
+ * @inc_usage_cnt - flag to increment @udev's PM-usage counter
+ *
+ * This routine should be called when a core subsystem wants to use @udev
+ * and needs to guarantee that it is not suspended.  In addition, the
+ * caller can prevent @udev from being autosuspended subsequently.  (Note
+ * that this will not prevent suspend events originating in the PM core.)
+ * Examples would be when @udev's device file in usbfs is opened (autosuspend
+ * should be prevented until the file is closed) or when a remote-wakeup
+ * request is received (later autosuspends should not be prevented).
+ *
+ * @inc_usage_cnt should be 1 to increment @udev's usage counter and prevent
+ * autosuspends.  This prevention will persist until the usage counter is
+ * decremented again (such as by passing 1 to usb_autosuspend_device).
+ * Otherwise @inc_usage_cnt should be 0 to leave the usage counter unchanged.
+ * Regardless, if the autoresume fails then the usage counter is not
+ * incremented.
+ *
+ * Often the caller will hold @udev's device lock, but this is not
+ * necessary (and attempting it might cause deadlock).
+ *
+ * This routine can run only in process context.
+ */
+int usb_autoresume_device(struct usb_device *udev, int inc_usage_cnt)
+{
+	int	status;
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	udev->pm_usage_cnt += inc_usage_cnt;
+	udev->auto_pm = 1;
+	status = usb_resume_both(udev);
+	if (status != 0)
+		udev->pm_usage_cnt -= inc_usage_cnt;
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&udev->dev, "%s: status %d cnt %d\n",
+	//		__FUNCTION__, status, udev->pm_usage_cnt);
+	return status;
+}
+
+/**
+ * usb_autopm_put_interface - decrement a USB interface's PM-usage counter
+ * @intf - the usb_interface whose counter should be decremented
+ *
+ * This routine should be called by an interface driver when it is
+ * finished using @intf and wants to allow it to autosuspend.  A typical
+ * example would be a character-device driver when its device file is
+ * closed.
+ *
+ * The routine decrements @intf's usage counter.  When the counter reaches
+ * 0, a delayed autosuspend request for @intf's device is queued.  When
+ * the delay expires, if @intf->pm_usage_cnt is still <= 0 along with all
+ * the other usage counters for the sibling interfaces and @intf's
+ * usb_device, the device and all its interfaces will be autosuspended.
+ *
+ * Note that @intf->pm_usage_cnt is owned by the interface driver.  The
+ * core will not change its value other than the increment and decrement
+ * in usb_autopm_get_interface and usb_autopm_put_interface.  The driver
+ * may use this simple counter-oriented discipline or may set the value
+ * any way it likes.
+ *
+ * If the driver has set @intf->needs_remote_wakeup then autosuspend will
+ * take place only if the device's remote-wakeup facility is enabled.
+ *
+ * Suspend method calls queued by this routine can arrive at any time
+ * while @intf is resumed and its usage counter is equal to 0.  They are
+ * not protected by the usb_device's lock but only by its pm_mutex.
+ * Drivers must provide their own synchronization.
+ *
+ * This routine can run only in process context.
+ */
+void usb_autopm_put_interface(struct usb_interface *intf)
+{
+	struct usb_device	*udev = interface_to_usbdev(intf);
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	if (intf->condition != USB_INTERFACE_UNBOUND) {
+		if (--intf->pm_usage_cnt <= 0)
+			schedule_delayed_work(&udev->autosuspend,
+					USB_AUTOSUSPEND_DELAY);
+	}
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&intf->dev, "%s: cnt %d\n",
+	//		__FUNCTION__, intf->pm_usage_cnt);
+}
+EXPORT_SYMBOL_GPL(usb_autopm_put_interface);
+
+/**
+ * usb_autopm_get_interface - increment a USB interface's PM-usage counter
+ * @intf - the usb_interface whose counter should be incremented
+ *
+ * This routine should be called by an interface driver when it wants to
+ * use @intf and needs to guarantee that it is not suspended.  In addition,
+ * the routine prevents @intf from being autosuspended subsequently.  (Note
+ * that this will not prevent suspend events originating in the PM core.)
+ * This prevention will persist until usb_autopm_put_interface() is called
+ * or @intf is unbound.  A typical example would be a character-device
+ * driver when its device file is opened.
+ *
+ * The routine increments @intf's usage counter.  So long as the counter
+ * is greater than 0, autosuspend will not be allowed for @intf or its
+ * usb_device.  When the driver is finished using @intf it should call
+ * usb_autopm_put_interface() to decrement the usage counter and queue
+ * a delayed autosuspend request (if the counter is <= 0).
+ *
+ * Note that @intf->pm_usage_cnt is owned by the interface driver.  The
+ * core will not change its value other than the increment and decrement
+ * in usb_autopm_get_interface and usb_autopm_put_interface.  The driver
+ * may use this simple counter-oriented discipline or may set the value
+ * any way it likes.
+ *
+ * Resume method calls generated by this routine can arrive at any time
+ * while @intf is suspended.  They are not protected by the usb_device's
+ * lock but only by its pm_mutex.  Drivers must provide their own
+ * synchronization.
+ *
+ * This routine can run only in process context.
+ */
+int usb_autopm_get_interface(struct usb_interface *intf)
+{
+	struct usb_device	*udev = interface_to_usbdev(intf);
+	int			status;
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	if (intf->condition == USB_INTERFACE_UNBOUND)
+		status = -ENODEV;
+	else {
+		++intf->pm_usage_cnt;
+		udev->auto_pm = 1;
+		status = usb_resume_both(udev);
+		if (status != 0)
+			--intf->pm_usage_cnt;
+	}
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&intf->dev, "%s: status %d cnt %d\n",
+	//		__FUNCTION__, status, intf->pm_usage_cnt);
 	return status;
 }
+EXPORT_SYMBOL_GPL(usb_autopm_get_interface);
+
+#endif /* CONFIG_USB_SUSPEND */
 
 static int usb_suspend(struct device *dev, pm_message_t message)
 {
 	int	status;
 
-	if (is_usb_device(dev))
-		status = usb_suspend_both(to_usb_device(dev), message);
-	else
+	if (is_usb_device(dev)) {
+		struct usb_device *udev = to_usb_device(dev);
+
+		mutex_lock_nested(&udev->pm_mutex, udev->level);
+		udev->auto_pm = 0;
+		status = usb_suspend_both(udev, message);
+		mutex_unlock(&udev->pm_mutex);
+	} else
 		status = 0;
 	return status;
 }
@@ -939,7 +1293,12 @@ static int usb_resume(struct device *dev)
 	int	status;
 
 	if (is_usb_device(dev)) {
-		status = usb_resume_both(to_usb_device(dev));
+		struct usb_device *udev = to_usb_device(dev);
+
+		mutex_lock_nested(&udev->pm_mutex, udev->level);
+		udev->auto_pm = 0;
+		status = usb_resume_both(udev);
+		mutex_unlock(&udev->pm_mutex);
 
 		/* Rebind drivers that had no suspend method? */
 	} else
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 78e910b2046c..dee812bc6c43 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1017,19 +1017,22 @@ void usb_set_device_state(struct usb_device *udev,
 	if (udev->state == USB_STATE_NOTATTACHED)
 		;	/* do nothing */
 	else if (new_state != USB_STATE_NOTATTACHED) {
-		udev->state = new_state;
 
 		/* root hub wakeup capabilities are managed out-of-band
 		 * and may involve silicon errata ... ignore them here.
 		 */
 		if (udev->parent) {
-			if (new_state == USB_STATE_CONFIGURED)
+			if (udev->state == USB_STATE_SUSPENDED
+					|| new_state == USB_STATE_SUSPENDED)
+				;	/* No change to wakeup settings */
+			else if (new_state == USB_STATE_CONFIGURED)
 				device_init_wakeup(&udev->dev,
 					(udev->actconfig->desc.bmAttributes
 					 & USB_CONFIG_ATT_WAKEUP));
-			else if (new_state != USB_STATE_SUSPENDED)
+			else
 				device_init_wakeup(&udev->dev, 0);
 		}
+		udev->state = new_state;
 	} else
 		recursively_mark_NOTATTACHED(udev);
 	spin_unlock_irqrestore(&device_state_lock, flags);
@@ -1507,7 +1510,7 @@ static int hub_port_suspend(struct usb_hub *hub, int port1,
 	 * NOTE:  OTG devices may issue remote wakeup (or SRP) even when
 	 * we don't explicitly enable it here.
 	 */
-	if (device_may_wakeup(&udev->dev)) {
+	if (udev->do_remote_wakeup) {
 		status = usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
 				USB_REQ_SET_FEATURE, USB_RECIP_DEVICE,
 				USB_DEVICE_REMOTE_WAKEUP, 0,
@@ -1533,7 +1536,8 @@ static int hub_port_suspend(struct usb_hub *hub, int port1,
 				USB_CTRL_SET_TIMEOUT);
 	} else {
 		/* device has up to 10 msec to fully suspend */
-		dev_dbg(&udev->dev, "usb suspend\n");
+		dev_dbg(&udev->dev, "usb %ssuspend\n",
+				udev->auto_pm ? "auto-" : "");
 		usb_set_device_state(udev, USB_STATE_SUSPENDED);
 		msleep(10);
 	}
@@ -1573,7 +1577,8 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
 		status = hub_port_suspend(hdev_to_hub(udev->parent), port1,
 				udev);
 	else {
-		dev_dbg(&udev->dev, "usb suspend\n");
+		dev_dbg(&udev->dev, "usb %ssuspend\n",
+				udev->auto_pm ? "auto-" : "");
 		usb_set_device_state(udev, USB_STATE_SUSPENDED);
 	}
 	return status;
@@ -1687,7 +1692,8 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 
 		/* drive resume for at least 20 msec */
 		if (udev)
-			dev_dbg(&udev->dev, "RESUME\n");
+			dev_dbg(&udev->dev, "usb %sresume\n",
+					udev->auto_pm ? "auto-" : "");
 		msleep(25);
 
 #define LIVE_FLAGS	( USB_PORT_STAT_POWER \
@@ -1754,8 +1760,11 @@ int usb_port_resume(struct usb_device *udev)
 		// NOTE this fails if parent is also suspended...
 		status = hub_port_resume(hdev_to_hub(udev->parent),
 				udev->portnum, udev);
-	} else
+	} else {
+		dev_dbg(&udev->dev, "usb %sresume\n",
+				udev->auto_pm ? "auto-" : "");
 		status = finish_port_resume(udev);
+	}
 	if (status < 0)
 		dev_dbg(&udev->dev, "can't resume, status %d\n", status);
 	return status;
@@ -1765,19 +1774,23 @@ static int remote_wakeup(struct usb_device *udev)
 {
 	int	status = 0;
 
-	/* don't repeat RESUME sequence if this device
-	 * was already woken up by some other task
-	 */
+	/* All this just to avoid sending a port-resume message
+	 * to the parent hub! */
+
 	usb_lock_device(udev);
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
 	if (udev->state == USB_STATE_SUSPENDED) {
-		dev_dbg(&udev->dev, "RESUME (wakeup)\n");
+		dev_dbg(&udev->dev, "usb %sresume\n", "wakeup-");
 		/* TRSMRCY = 10 msec */
 		msleep(10);
 		status = finish_port_resume(udev);
+		if (status == 0)
+			udev->dev.power.power_state.event = PM_EVENT_ON;
 	}
+	mutex_unlock(&udev->pm_mutex);
 
 	if (status == 0)
-		usb_resume_both(udev);
+		usb_autoresume_device(udev, 0);
 	usb_unlock_device(udev);
 	return status;
 }
@@ -1834,7 +1847,9 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
 					== PM_EVENT_ON
 #endif
 				) {
-			dev_dbg(&intf->dev, "port %d nyet suspended\n", port1);
+			if (!hdev->auto_pm)
+				dev_dbg(&intf->dev, "port %d nyet suspended\n",
+						port1);
 			return -EBUSY;
 		}
 	}
@@ -2587,7 +2602,7 @@ static void hub_events(void)
 		 * stub "device" node was never suspended.
 		 */
 		if (i)
-			usb_resume_both(hdev);
+			usb_autoresume_device(hdev, 0);
 
 		/* If this is an inactive or suspended hub, do nothing */
 		if (hub->quiescing)
@@ -2993,6 +3008,9 @@ int usb_reset_composite_device(struct usb_device *udev,
 		return -EINVAL;
 	}
 
+	/* Prevent autosuspend during the reset */
+	usb_autoresume_device(udev, 1);
+
 	if (iface && iface->condition != USB_INTERFACE_BINDING)
 		iface = NULL;
 
@@ -3034,6 +3052,7 @@ int usb_reset_composite_device(struct usb_device *udev,
 		}
 	}
 
+	usb_autosuspend_device(udev, 1);
 	return ret;
 }
 EXPORT_SYMBOL(usb_reset_composite_device);
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index b0c0a993338f..6b029cdb8671 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -168,6 +168,10 @@ static void usb_release_dev(struct device *dev)
 
 	udev = to_usb_device(dev);
 
+#ifdef	CONFIG_PM
+	cancel_delayed_work(&udev->autosuspend);
+	flush_scheduled_work();
+#endif
 	usb_destroy_configuration(udev);
 	usb_put_hcd(bus_to_hcd(udev->bus));
 	kfree(udev->product);
@@ -176,6 +180,21 @@ static void usb_release_dev(struct device *dev)
 	kfree(udev);
 }
 
+#ifdef	CONFIG_PM
+
+/* usb_autosuspend_work - callback routine to autosuspend a USB device */
+static void usb_autosuspend_work(void *_udev)
+{
+	struct usb_device	*udev = _udev;
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	udev->auto_pm = 1;
+	usb_suspend_both(udev, PMSG_SUSPEND);
+	mutex_unlock(&udev->pm_mutex);
+}
+
+#endif
+
 /**
  * usb_alloc_dev - usb device constructor (usbcore-internal)
  * @parent: hub to which device is connected; null to allocate a root hub
@@ -251,6 +270,10 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	dev->parent = parent;
 	INIT_LIST_HEAD(&dev->filelist);
 
+#ifdef	CONFIG_PM
+	mutex_init(&dev->pm_mutex);
+	INIT_WORK(&dev->autosuspend, usb_autosuspend_work, dev);
+#endif
 	return dev;
 }
 
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 5162cb370215..10688ad73c6d 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -49,6 +49,20 @@ static inline int usb_resume_both(struct usb_device *udev)
 
 #endif
 
+#ifdef CONFIG_USB_SUSPEND
+
+#define USB_AUTOSUSPEND_DELAY	(HZ*2)
+
+extern void usb_autosuspend_device(struct usb_device *udev, int dec_busy_cnt);
+extern int usb_autoresume_device(struct usb_device *udev, int inc_busy_cnt);
+
+#else
+
+#define usb_autosuspend_device(udev, dec_busy_cnt)	do {} while (0)
+#define usb_autoresume_device(udev, inc_busy_cnt)	0
+
+#endif
+
 extern struct bus_type usb_bus_type;
 extern struct usb_device_driver usb_generic_driver;
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index df5c93eb3ce9..0da15b0b02be 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -19,6 +19,7 @@
 #include <linux/fs.h>		/* for struct file_operations */
 #include <linux/completion.h>	/* for struct completion */
 #include <linux/sched.h>	/* for current && schedule_timeout */
+#include <linux/mutex.h>	/* for struct mutex */
 
 struct usb_device;
 struct usb_driver;
@@ -103,8 +104,12 @@ enum usb_interface_condition {
  * @condition: binding state of the interface: not bound, binding
  *	(in probe()), bound to a driver, or unbinding (in disconnect())
  * @is_active: flag set when the interface is bound and not suspended.
+ * @needs_remote_wakeup: flag set when the driver requires remote-wakeup
+ *	capability during autosuspend.
  * @dev: driver model's view of this device
  * @class_dev: driver model's class view of this device.
+ * @pm_usage_cnt: PM usage counter for this interface; autosuspend is not
+ *	allowed unless the counter is 0.
  *
  * USB device drivers attach to interfaces on a physical device.  Each
  * interface encapsulates a single high level function, such as feeding
@@ -144,9 +149,11 @@ struct usb_interface {
 					 * bound to */
 	enum usb_interface_condition condition;		/* state of binding */
 	unsigned is_active:1;		/* the interface is not suspended */
+	unsigned needs_remote_wakeup:1;	/* driver requires remote wakeup */
 
 	struct device dev;		/* interface specific device info */
 	struct class_device *class_dev;
+	int pm_usage_cnt;		/* usage counter for autosuspend */
 };
 #define	to_usb_interface(d) container_of(d, struct usb_interface, dev)
 #define	interface_to_usbdev(intf) \
@@ -372,6 +379,15 @@ struct usb_device {
 
 	int maxchild;			/* Number of ports if hub */
 	struct usb_device *children[USB_MAXCHILDREN];
+
+#ifdef CONFIG_PM
+	struct work_struct autosuspend;	/* for delayed autosuspends */
+	struct mutex pm_mutex;		/* protects PM operations */
+	int pm_usage_cnt;		/* usage counter for autosuspend */
+
+	unsigned auto_pm:1;		/* autosuspend/resume in progress */
+	unsigned do_remote_wakeup:1;	/* remote wakeup should be enabled */
+#endif
 };
 #define	to_usb_device(d) container_of(d, struct usb_device, dev)
 
@@ -392,6 +408,17 @@ extern int usb_reset_composite_device(struct usb_device *dev,
 
 extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id);
 
+/* USB autosuspend and autoresume */
+#ifdef CONFIG_USB_SUSPEND
+extern int usb_autopm_get_interface(struct usb_interface *intf);
+extern void usb_autopm_put_interface(struct usb_interface *intf);
+
+#else
+#define usb_autopm_get_interface(intf)		0
+#define usb_autopm_put_interface(intf)		do {} while (0)
+#endif
+
+
 /*-------------------------------------------------------------------------*/
 
 /* for drivers using iso endpoints */
@@ -593,6 +620,8 @@ struct usbdrv_wrap {
  * @drvwrap: Driver-model core structure wrapper.
  * @no_dynamic_id: if set to 1, the USB core will not allow dynamic ids to be
  *	added to this driver by preventing the sysfs file from being created.
+ * @supports_autosuspend: if set to 0, the USB core will not allow autosuspend
+ *	for interfaces bound to this driver.
  *
  * USB interface drivers must provide a name, probe() and disconnect()
  * methods, and an id_table.  Other driver fields are optional.
@@ -631,6 +660,7 @@ struct usb_driver {
 	struct usb_dynids dynids;
 	struct usbdrv_wrap drvwrap;
 	unsigned int no_dynamic_id:1;
+	unsigned int supports_autosuspend:1;
 };
 #define	to_usb_driver(d) container_of(d, struct usb_driver, drvwrap.driver)
 
@@ -648,6 +678,8 @@ struct usb_driver {
  * @suspend: Called when the device is going to be suspended by the system.
  * @resume: Called when the device is being resumed by the system.
  * @drvwrap: Driver-model core structure wrapper.
+ * @supports_autosuspend: if set to 0, the USB core will not allow autosuspend
+ *	for devices bound to this driver.
  *
  * USB drivers must provide all the fields listed above except drvwrap.
  */
@@ -660,6 +692,7 @@ struct usb_device_driver {
 	int (*suspend) (struct usb_device *udev, pm_message_t message);
 	int (*resume) (struct usb_device *udev);
 	struct usbdrv_wrap drvwrap;
+	unsigned int supports_autosuspend:1;
 };
 #define	to_usb_device_driver(d) container_of(d, struct usb_device_driver, \
 		drvwrap.driver)
-- 
cgit v1.2.3


From 8d48427ecb0639593ccf14e807479b7873254ccb Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@rpsys.net>
Date: Mon, 25 Sep 2006 20:11:48 +0100
Subject: [ARM] 3847/2:  Convert LOMOMO to use struct device for GPIOs

Convert LOMOMO to use struct device * for GPIOs instead of struct
locomo_dev. This enables access to the GPIOs from code which is not
a locomo device itself (such as audio). Access for gpio 31 is removed
for error handling (no such hardware exists).

Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/locomo.c            | 32 ++++++++++++++++++++++----------
 arch/arm/mach-sa1100/collie.c       | 23 ++++++++++-------------
 drivers/video/backlight/locomolcd.c | 30 +++++++++++++++---------------
 include/asm-arm/hardware/locomo.h   |  9 +++++----
 4 files changed, 52 insertions(+), 42 deletions(-)

(limited to 'include')

diff --git a/arch/arm/common/locomo.c b/arch/arm/common/locomo.c
index 4e0dcaef6eb2..c46dc65ec79a 100644
--- a/arch/arm/common/locomo.c
+++ b/arch/arm/common/locomo.c
@@ -814,12 +814,15 @@ static inline struct locomo *locomo_chip_driver(struct locomo_dev *ldev)
 	return (struct locomo *)dev_get_drvdata(ldev->dev.parent);
 }
 
-void locomo_gpio_set_dir(struct locomo_dev *ldev, unsigned int bits, unsigned int dir)
+void locomo_gpio_set_dir(struct device *dev, unsigned int bits, unsigned int dir)
 {
-	struct locomo *lchip = locomo_chip_driver(ldev);
+	struct locomo *lchip = dev_get_drvdata(dev);
 	unsigned long flags;
 	unsigned int r;
 
+	if (!lchip)
+		return;
+
 	spin_lock_irqsave(&lchip->lock, flags);
 
 	r = locomo_readl(lchip->base + LOCOMO_GPD);
@@ -836,12 +839,15 @@ void locomo_gpio_set_dir(struct locomo_dev *ldev, unsigned int bits, unsigned in
 	spin_unlock_irqrestore(&lchip->lock, flags);
 }
 
-unsigned int locomo_gpio_read_level(struct locomo_dev *ldev, unsigned int bits)
+int locomo_gpio_read_level(struct device *dev, unsigned int bits)
 {
-	struct locomo *lchip = locomo_chip_driver(ldev);
+	struct locomo *lchip = dev_get_drvdata(dev);
 	unsigned long flags;
 	unsigned int ret;
 
+	if (!lchip)
+		return -ENODEV;
+
 	spin_lock_irqsave(&lchip->lock, flags);
 	ret = locomo_readl(lchip->base + LOCOMO_GPL);
 	spin_unlock_irqrestore(&lchip->lock, flags);
@@ -850,12 +856,15 @@ unsigned int locomo_gpio_read_level(struct locomo_dev *ldev, unsigned int bits)
 	return ret;
 }
 
-unsigned int locomo_gpio_read_output(struct locomo_dev *ldev, unsigned int bits)
+int locomo_gpio_read_output(struct device *dev, unsigned int bits)
 {
-	struct locomo *lchip = locomo_chip_driver(ldev);
+	struct locomo *lchip = dev_get_drvdata(dev);
 	unsigned long flags;
 	unsigned int ret;
 
+	if (!lchip)
+		return -ENODEV;
+
 	spin_lock_irqsave(&lchip->lock, flags);
 	ret = locomo_readl(lchip->base + LOCOMO_GPO);
 	spin_unlock_irqrestore(&lchip->lock, flags);
@@ -864,12 +873,15 @@ unsigned int locomo_gpio_read_output(struct locomo_dev *ldev, unsigned int bits)
 	return ret;
 }
 
-void locomo_gpio_write(struct locomo_dev *ldev, unsigned int bits, unsigned int set)
+void locomo_gpio_write(struct device *dev, unsigned int bits, unsigned int set)
 {
-	struct locomo *lchip = locomo_chip_driver(ldev);
+	struct locomo *lchip = dev_get_drvdata(dev);
 	unsigned long flags;
 	unsigned int r;
 
+	if (!lchip)
+		return;
+
 	spin_lock_irqsave(&lchip->lock, flags);
 
 	r = locomo_readl(lchip->base + LOCOMO_GPO);
@@ -1058,9 +1070,9 @@ void locomo_frontlight_set(struct locomo_dev *dev, int duty, int vr, int bpwf)
 	struct locomo *lchip = locomo_chip_driver(dev);
 
 	if (vr)
-		locomo_gpio_write(dev, LOCOMO_GPIO_FL_VR, 1);
+		locomo_gpio_write(dev->dev.parent, LOCOMO_GPIO_FL_VR, 1);
 	else
-		locomo_gpio_write(dev, LOCOMO_GPIO_FL_VR, 0);
+		locomo_gpio_write(dev->dev.parent, LOCOMO_GPIO_FL_VR, 0);
 
 	spin_lock_irqsave(&lchip->lock, flags);
 	locomo_writel(bpwf, lchip->base + LOCOMO_FRONTLIGHT + LOCOMO_ALS);
diff --git a/arch/arm/mach-sa1100/collie.c b/arch/arm/mach-sa1100/collie.c
index a0dfa390e34b..6496eb645cee 100644
--- a/arch/arm/mach-sa1100/collie.c
+++ b/arch/arm/mach-sa1100/collie.c
@@ -91,30 +91,29 @@ static struct mcp_plat_data collie_mcp_data = {
 /*
  * low-level UART features.
  */
-static struct locomo_dev *uart_dev = NULL;
+struct platform_device collie_locomo_device;
 
 static void collie_uart_set_mctrl(struct uart_port *port, u_int mctrl)
 {
- 	if (!uart_dev) return;
-
  	if (mctrl & TIOCM_RTS)
-		locomo_gpio_write(uart_dev, LOCOMO_GPIO_RTS, 0);
+		locomo_gpio_write(&collie_locomo_device.dev, LOCOMO_GPIO_RTS, 0);
  	else
-		locomo_gpio_write(uart_dev, LOCOMO_GPIO_RTS, 1);
+		locomo_gpio_write(&collie_locomo_device.dev, LOCOMO_GPIO_RTS, 1);
 
  	if (mctrl & TIOCM_DTR)
-		locomo_gpio_write(uart_dev, LOCOMO_GPIO_DTR, 0);
+		locomo_gpio_write(&collie_locomo_device.dev, LOCOMO_GPIO_DTR, 0);
  	else
-		locomo_gpio_write(uart_dev, LOCOMO_GPIO_DTR, 1);
+		locomo_gpio_write(&collie_locomo_device.dev, LOCOMO_GPIO_DTR, 1);
 }
 
 static u_int collie_uart_get_mctrl(struct uart_port *port)
 {
 	int ret = TIOCM_CD;
 	unsigned int r;
-	if (!uart_dev) return ret;
 
-	r = locomo_gpio_read_output(uart_dev, LOCOMO_GPIO_CTS & LOCOMO_GPIO_DSR);
+	r = locomo_gpio_read_output(&collie_locomo_device.dev, LOCOMO_GPIO_CTS & LOCOMO_GPIO_DSR);
+	if (r == -ENODEV)
+		return ret;
 	if (r & LOCOMO_GPIO_CTS)
 		ret |= TIOCM_CTS;
 	if (r & LOCOMO_GPIO_DSR)
@@ -130,13 +129,11 @@ static struct sa1100_port_fns collie_port_fns __initdata = {
 
 static int collie_uart_probe(struct locomo_dev *dev)
 {
-	uart_dev = dev;
 	return 0;
 }
 
 static int collie_uart_remove(struct locomo_dev *dev)
 {
-	uart_dev = NULL;
 	return 0;
 }
 
@@ -170,7 +167,7 @@ static struct resource locomo_resources[] = {
 	},
 };
 
-static struct platform_device locomo_device = {
+struct platform_device collie_locomo_device = {
 	.name		= "locomo",
 	.id		= 0,
 	.num_resources	= ARRAY_SIZE(locomo_resources),
@@ -178,7 +175,7 @@ static struct platform_device locomo_device = {
 };
 
 static struct platform_device *devices[] __initdata = {
-	&locomo_device,
+	&collie_locomo_device,
 	&colliescoop_device,
 };
 
diff --git a/drivers/video/backlight/locomolcd.c b/drivers/video/backlight/locomolcd.c
index caf1eca199b0..628571c63bac 100644
--- a/drivers/video/backlight/locomolcd.c
+++ b/drivers/video/backlight/locomolcd.c
@@ -33,19 +33,19 @@ static unsigned long locomolcd_flags;
 
 static void locomolcd_on(int comadj)
 {
-	locomo_gpio_set_dir(locomolcd_dev, LOCOMO_GPIO_LCD_VSHA_ON, 0);
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_VSHA_ON, 1);
+	locomo_gpio_set_dir(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VSHA_ON, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VSHA_ON, 1);
 	mdelay(2);
 
-	locomo_gpio_set_dir(locomolcd_dev, LOCOMO_GPIO_LCD_VSHD_ON, 0);
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_VSHD_ON, 1);
+	locomo_gpio_set_dir(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VSHD_ON, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VSHD_ON, 1);
 	mdelay(2);
 
 	locomo_m62332_senddata(locomolcd_dev, comadj, 0);
 	mdelay(5);
 
-	locomo_gpio_set_dir(locomolcd_dev, LOCOMO_GPIO_LCD_VEE_ON, 0);
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_VEE_ON, 1);
+	locomo_gpio_set_dir(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VEE_ON, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VEE_ON, 1);
 	mdelay(10);
 
 	/* TFTCRST | CPSOUT=0 | CPSEN */
@@ -58,8 +58,8 @@ static void locomolcd_on(int comadj)
 	locomo_writel((0x04 | 0x01), locomolcd_dev->mapbase + LOCOMO_TC);
 	mdelay(10);
 
-	locomo_gpio_set_dir(locomolcd_dev, LOCOMO_GPIO_LCD_MOD, 0);
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_MOD, 1);
+	locomo_gpio_set_dir(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_MOD, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_MOD, 1);
 }
 
 static void locomolcd_off(int comadj)
@@ -68,16 +68,16 @@ static void locomolcd_off(int comadj)
 	locomo_writel(0x06, locomolcd_dev->mapbase + LOCOMO_TC);
 	mdelay(1);
 
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_VSHA_ON, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VSHA_ON, 0);
 	mdelay(110);
 
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_VEE_ON, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VEE_ON, 0);
 	mdelay(700);
 
 	/* TFTCRST=0 | CPSOUT=0 | CPSEN = 0 */
 	locomo_writel(0, locomolcd_dev->mapbase + LOCOMO_TC);
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_MOD, 0);
-	locomo_gpio_write(locomolcd_dev, LOCOMO_GPIO_LCD_VSHD_ON, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_MOD, 0);
+	locomo_gpio_write(locomolcd_dev->dev.parent, LOCOMO_GPIO_LCD_VSHD_ON, 0);
 }
 
 void locomolcd_power(int on)
@@ -167,14 +167,14 @@ static int locomolcd_resume(struct locomo_dev *dev)
 #define locomolcd_resume	NULL
 #endif
 
-static int locomolcd_probe(struct locomo_dev *dev)
+static int locomolcd_probe(struct locomo_dev *ldev)
 {
 	unsigned long flags;
 
 	local_irq_save(flags);
-	locomolcd_dev = dev;
+	locomolcd_dev = ldev;
 
-	locomo_gpio_set_dir(dev, LOCOMO_GPIO_FL_VR, 0);
+	locomo_gpio_set_dir(ldev->dev.parent, LOCOMO_GPIO_FL_VR, 0);
 
 	/* the poodle_lcd_power function is called for the first time
 	 * from fs_initcall, which is before locomo is activated.
diff --git a/include/asm-arm/hardware/locomo.h b/include/asm-arm/hardware/locomo.h
index 22dfb1737768..2599a6bc70e4 100644
--- a/include/asm-arm/hardware/locomo.h
+++ b/include/asm-arm/hardware/locomo.h
@@ -197,10 +197,11 @@ int locomo_driver_register(struct locomo_driver *);
 void locomo_driver_unregister(struct locomo_driver *);
 
 /* GPIO control functions */
-void locomo_gpio_set_dir(struct locomo_dev *ldev, unsigned int bits, unsigned int dir);
-unsigned int locomo_gpio_read_level(struct locomo_dev *ldev, unsigned int bits);
-unsigned int locomo_gpio_read_output(struct locomo_dev *ldev, unsigned int bits);
-void locomo_gpio_write(struct locomo_dev *ldev, unsigned int bits, unsigned int set);
+void locomo_gpio_set_dir(struct device *dev, unsigned int bits, unsigned int dir);
+int locomo_gpio_read_level(struct device *dev, unsigned int bits);
+int locomo_gpio_read_output(struct device *dev, unsigned int bits);
+void locomo_gpio_write(struct device *dev, unsigned int bits, unsigned int set);
+
 
 /* M62332 control function */
 void locomo_m62332_senddata(struct locomo_dev *ldev, unsigned int dac_data, int channel);
-- 
cgit v1.2.3


From a2025e7f73ae5eab0a25dad88c60aba67e3ae690 Mon Sep 17 00:00:00 2001
From: Dirk Opfer <dirk@opfer-online.de>
Date: Mon, 25 Sep 2006 22:41:47 +0100
Subject: [ARM] 3863/1: Add Locomo SPI Device

The Locomo chip has a SPI interface which is used for SD/MMC cards (only collie).
This patch adds the definition for the SPI device inside the Locomo chip.

Signed-off-by: Dirk Opfer <Dirk@Opfer-Online.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/locomo.c          | 29 ++++++++++++++++++-----------
 include/asm-arm/hardware/locomo.h | 24 +++++++++++++-----------
 2 files changed, 31 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/arch/arm/common/locomo.c b/arch/arm/common/locomo.c
index c46dc65ec79a..181ef1ead5b8 100644
--- a/arch/arm/common/locomo.c
+++ b/arch/arm/common/locomo.c
@@ -121,6 +121,13 @@ static struct locomo_dev_info locomo_devices[] = {
 		.offset		= 0,
 		.length		= 0,
 	},
+	{
+		.devid		= LOCOMO_DEVID_SPI,
+		.irq		= {},
+		.name		= "locomo-spi",
+		.offset		= LOCOMO_SPI,
+		.length		= 0x30,
+	},
 };
 
 
@@ -374,7 +381,7 @@ static void locomo_spi_handler(unsigned int irq, struct irqdesc *desc,
 	struct irqdesc *d;
 	void __iomem *mapbase = get_irq_chipdata(irq);
 
-	req = locomo_readl(mapbase + LOCOMO_SPIIR) & 0x000F;
+	req = locomo_readl(mapbase + LOCOMO_SPI + LOCOMO_SPIIR) & 0x000F;
 	if (req) {
 		irq = LOCOMO_IRQ_SPI_START;
 		d = irq_desc + irq;
@@ -391,35 +398,35 @@ static void locomo_spi_ack_irq(unsigned int irq)
 {
 	void __iomem *mapbase = get_irq_chipdata(irq);
 	unsigned int r;
-	r = locomo_readl(mapbase + LOCOMO_SPIWE);
+	r = locomo_readl(mapbase + LOCOMO_SPI + LOCOMO_SPIWE);
 	r |= (0x0001 << (irq - LOCOMO_IRQ_SPI_START));
-	locomo_writel(r, mapbase + LOCOMO_SPIWE);
+	locomo_writel(r, mapbase + LOCOMO_SPI + LOCOMO_SPIWE);
 
-	r = locomo_readl(mapbase + LOCOMO_SPIIS);
+	r = locomo_readl(mapbase + LOCOMO_SPI + LOCOMO_SPIIS);
 	r &= ~(0x0001 << (irq - LOCOMO_IRQ_SPI_START));
-	locomo_writel(r, mapbase + LOCOMO_SPIIS);
+	locomo_writel(r, mapbase + LOCOMO_SPI + LOCOMO_SPIIS);
 
-	r = locomo_readl(mapbase + LOCOMO_SPIWE);
+	r = locomo_readl(mapbase + LOCOMO_SPI + LOCOMO_SPIWE);
 	r &= ~(0x0001 << (irq - LOCOMO_IRQ_SPI_START));
-	locomo_writel(r, mapbase + LOCOMO_SPIWE);
+	locomo_writel(r, mapbase + LOCOMO_SPI + LOCOMO_SPIWE);
 }
 
 static void locomo_spi_mask_irq(unsigned int irq)
 {
 	void __iomem *mapbase = get_irq_chipdata(irq);
 	unsigned int r;
-	r = locomo_readl(mapbase + LOCOMO_SPIIE);
+	r = locomo_readl(mapbase + LOCOMO_SPI + LOCOMO_SPIIE);
 	r &= ~(0x0001 << (irq - LOCOMO_IRQ_SPI_START));
-	locomo_writel(r, mapbase + LOCOMO_SPIIE);
+	locomo_writel(r, mapbase + LOCOMO_SPI + LOCOMO_SPIIE);
 }
 
 static void locomo_spi_unmask_irq(unsigned int irq)
 {
 	void __iomem *mapbase = get_irq_chipdata(irq);
 	unsigned int r;
-	r = locomo_readl(mapbase + LOCOMO_SPIIE);
+	r = locomo_readl(mapbase + LOCOMO_SPI + LOCOMO_SPIIE);
 	r |= (0x0001 << (irq - LOCOMO_IRQ_SPI_START));
-	locomo_writel(r, mapbase + LOCOMO_SPIIE);
+	locomo_writel(r, mapbase + LOCOMO_SPI + LOCOMO_SPIIE);
 }
 
 static struct irq_chip locomo_spi_chip = {
diff --git a/include/asm-arm/hardware/locomo.h b/include/asm-arm/hardware/locomo.h
index 2599a6bc70e4..adab77780ed3 100644
--- a/include/asm-arm/hardware/locomo.h
+++ b/include/asm-arm/hardware/locomo.h
@@ -54,17 +54,18 @@
 #define	LOCOMO_DAC_SDAOEB	0x01	/* SDA pin output data       */
 
 /* SPI interface */
-#define LOCOMO_SPIMD	0x60		/* SPI mode setting */
-#define LOCOMO_SPICT	0x64		/* SPI mode control */
-#define LOCOMO_SPIST	0x68		/* SPI status */
-#define LOCOMO_SPIIS	0x70		/* SPI interrupt status */
-#define LOCOMO_SPIWE	0x74		/* SPI interrupt status write enable */
-#define LOCOMO_SPIIE	0x78		/* SPI interrupt enable */
-#define LOCOMO_SPIIR	0x7c		/* SPI interrupt request */
-#define LOCOMO_SPITD	0x80		/* SPI transfer data write */
-#define LOCOMO_SPIRD	0x84		/* SPI receive data read */
-#define LOCOMO_SPITS	0x88		/* SPI transfer data shift */
-#define LOCOMO_SPIRS	0x8C		/* SPI receive data shift */
+#define LOCOMO_SPI	0x60
+#define LOCOMO_SPIMD	0x00		/* SPI mode setting */
+#define LOCOMO_SPICT	0x04		/* SPI mode control */
+#define LOCOMO_SPIST	0x08		/* SPI status */
+#define LOCOMO_SPIIS	0x10		/* SPI interrupt status */
+#define LOCOMO_SPIWE	0x14		/* SPI interrupt status write enable */
+#define LOCOMO_SPIIE	0x18		/* SPI interrupt enable */
+#define LOCOMO_SPIIR	0x1c		/* SPI interrupt request */
+#define LOCOMO_SPITD	0x20		/* SPI transfer data write */
+#define LOCOMO_SPIRD	0x24		/* SPI receive data read */
+#define LOCOMO_SPITS	0x28		/* SPI transfer data shift */
+#define LOCOMO_SPIRS	0x2C		/* SPI receive data shift */
 #define	LOCOMO_SPI_TEND	(1 << 3)	/* Transfer end bit */
 #define	LOCOMO_SPI_OVRN	(1 << 2)	/* Over Run bit */
 #define	LOCOMO_SPI_RFW	(1 << 1)	/* write buffer bit */
@@ -161,6 +162,7 @@ extern struct bus_type locomo_bus_type;
 #define LOCOMO_DEVID_AUDIO	3
 #define LOCOMO_DEVID_LED	4
 #define LOCOMO_DEVID_UART	5
+#define LOCOMO_DEVID_SPI		6
 
 struct locomo_dev {
 	struct device	dev;
-- 
cgit v1.2.3


From 576b3ef2495c732a56509febd5de5144f3ebccf6 Mon Sep 17 00:00:00 2001
From: Dirk Opfer <dirk@opfer-online.de>
Date: Mon, 25 Sep 2006 22:51:02 +0100
Subject: [ARM] 3864/1: Refactore sharpsl_pm

This patch adds another hook into sharpsl_pm to notify the machine
specific driver immediately after resume. This is needed to support the Sharp SL-6000 (Tosa).

Signed-off-by: Dirk Opfer <Dirk@Opfer-Online.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/sharpsl_pm.c          | 4 ++++
 include/asm-arm/hardware/sharpsl_pm.h | 1 +
 2 files changed, 5 insertions(+)

(limited to 'include')

diff --git a/arch/arm/common/sharpsl_pm.c b/arch/arm/common/sharpsl_pm.c
index 59b5ddec480f..f412dedda684 100644
--- a/arch/arm/common/sharpsl_pm.c
+++ b/arch/arm/common/sharpsl_pm.c
@@ -40,6 +40,7 @@
 #define SHARPSL_CHARGE_FINISH_TIME             (msecs_to_jiffies(10*60*1000)) /* 10 min */
 #define SHARPSL_BATCHK_TIME                    (msecs_to_jiffies(15*1000))    /* 15 sec */
 #define SHARPSL_BATCHK_TIME_SUSPEND            (60*10)                        /* 10 min */
+
 #define SHARPSL_WAIT_CO_TIME                   15  /* 15 sec */
 #define SHARPSL_WAIT_DISCHARGE_ON              100 /* 100 msec */
 #define SHARPSL_CHECK_BATTERY_WAIT_TIME_TEMP   10  /* 10 msec */
@@ -575,6 +576,9 @@ static int corgi_pxa_pm_enter(suspend_state_t state)
 	while (corgi_enter_suspend(alarm_time,alarm_status,state))
 		{}
 
+	if (sharpsl_pm.machinfo->earlyresume)
+		sharpsl_pm.machinfo->earlyresume();
+
 	dev_dbg(sharpsl_pm.dev, "SharpSL resuming...\n");
 
 	return 0;
diff --git a/include/asm-arm/hardware/sharpsl_pm.h b/include/asm-arm/hardware/sharpsl_pm.h
index ecf15b83956f..a836e76a14f7 100644
--- a/include/asm-arm/hardware/sharpsl_pm.h
+++ b/include/asm-arm/hardware/sharpsl_pm.h
@@ -25,6 +25,7 @@ struct sharpsl_charger_machinfo {
 	void (*measure_temp)(int);
 	void (*presuspend)(void);
 	void (*postsuspend)(void);
+	void (*earlyresume)(void);
 	unsigned long (*read_devdata)(int);
 #define SHARPSL_BATT_VOLT       1
 #define SHARPSL_BATT_TEMP       2
-- 
cgit v1.2.3


From 00463c1633b6d6a2178d2dc794c0a70ac2f9ce6b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 27 Sep 2006 21:38:56 +0200
Subject: [PATCH] i386: Use early clobbers for semaphores now

The new code does clobber the result early, so make sure to tell
gcc to not put it into the same register as a input argument

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andrew Morton <akpm@osdl.org>
Acked-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/semaphore.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/semaphore.h b/include/asm-i386/semaphore.h
index e63b6a68f04c..4e34a468c383 100644
--- a/include/asm-i386/semaphore.h
+++ b/include/asm-i386/semaphore.h
@@ -126,7 +126,7 @@ static inline int down_interruptible(struct semaphore * sem)
 		"lea %1,%%eax\n\t"
 		"call __down_failed_interruptible\n"
 		"2:"
-		:"=a" (result), "+m" (sem->count)
+		:"=&a" (result), "+m" (sem->count)
 		:
 		:"memory");
 	return result;
@@ -148,7 +148,7 @@ static inline int down_trylock(struct semaphore * sem)
 		"lea %1,%%eax\n\t"
 		"call __down_failed_trylock\n\t"
 		"2:\n"
-		:"=a" (result), "+m" (sem->count)
+		:"=&a" (result), "+m" (sem->count)
 		:
 		:"memory");
 	return result;
-- 
cgit v1.2.3


From 72729910c38ca5b4736032c15dc3f9d48fe4f68a Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Wed, 27 Sep 2006 09:44:11 +0100
Subject: [ARM] 3865/1: AT91RM9200 header updates

This is more preparation for adding support for the new Atmel AT91SAM9
processors.

Changes include:
- Replace AT91_BASE_* with AT91RM9200_BASE_*
- Replace AT91_ID_* with AT91RM9200_ID_*
- ROM, SRAM and UHP address definitions moved to at91rm9200.h.
- The raw AT91_P[ABCD]_* definitions are now depreciated in favour of
the GPIO API.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91rm9200/at91rm9200.c        |  32 ++++----
 arch/arm/mach-at91rm9200/clock.c             |  28 +++----
 arch/arm/mach-at91rm9200/devices.c           |  72 ++++++++--------
 arch/arm/mach-at91rm9200/gpio.c              |  16 ++--
 arch/arm/mach-at91rm9200/irq.c               |   4 +-
 arch/arm/mach-at91rm9200/pm.c                |  14 ++--
 drivers/mmc/at91_mci.c                       |   4 +-
 drivers/net/arm/at91_ether.c                 |   2 +-
 drivers/serial/at91_serial.c                 |   2 +-
 drivers/usb/gadget/at91_udc.c                |   6 +-
 include/asm-arm/arch-at91rm9200/at91rm9200.h | 118 +++++++++++++++------------
 include/asm-arm/arch-at91rm9200/gpio.h       |   2 +-
 include/asm-arm/arch-at91rm9200/hardware.h   |  37 ++++-----
 include/asm-arm/arch-at91rm9200/irqs.h       |   2 +-
 14 files changed, 173 insertions(+), 166 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-at91rm9200/at91rm9200.c b/arch/arm/mach-at91rm9200/at91rm9200.c
index 0985b1c42c7c..e21cb845d112 100644
--- a/arch/arm/mach-at91rm9200/at91rm9200.c
+++ b/arch/arm/mach-at91rm9200/at91rm9200.c
@@ -26,78 +26,78 @@ static struct map_desc at91rm9200_io_desc[] __initdata = {
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_SPI,
-		.pfn		= __phys_to_pfn(AT91_BASE_SPI),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SPI),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_SSC2,
-		.pfn		= __phys_to_pfn(AT91_BASE_SSC2),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SSC2),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_SSC1,
-		.pfn		= __phys_to_pfn(AT91_BASE_SSC1),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SSC1),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_SSC0,
-		.pfn		= __phys_to_pfn(AT91_BASE_SSC0),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SSC0),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_US3,
-		.pfn		= __phys_to_pfn(AT91_BASE_US3),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US3),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_US2,
-		.pfn		= __phys_to_pfn(AT91_BASE_US2),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US2),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_US1,
-		.pfn		= __phys_to_pfn(AT91_BASE_US1),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US1),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_US0,
-		.pfn		= __phys_to_pfn(AT91_BASE_US0),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US0),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_EMAC,
-		.pfn		= __phys_to_pfn(AT91_BASE_EMAC),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_EMAC),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_TWI,
-		.pfn		= __phys_to_pfn(AT91_BASE_TWI),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_TWI),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_MCI,
-		.pfn		= __phys_to_pfn(AT91_BASE_MCI),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_MCI),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_UDP,
-		.pfn		= __phys_to_pfn(AT91_BASE_UDP),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_UDP),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_TCB1,
-		.pfn		= __phys_to_pfn(AT91_BASE_TCB1),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_TCB1),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_TCB0,
-		.pfn		= __phys_to_pfn(AT91_BASE_TCB0),
+		.pfn		= __phys_to_pfn(AT91RM9200_BASE_TCB0),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_SRAM_VIRT_BASE,
-		.pfn		= __phys_to_pfn(AT91_SRAM_BASE),
-		.length		= AT91_SRAM_SIZE,
+		.pfn		= __phys_to_pfn(AT91RM9200_SRAM_BASE),
+		.length		= AT91RM9200_SRAM_SIZE,
 		.type		= MT_DEVICE,
 	},
 };
diff --git a/arch/arm/mach-at91rm9200/clock.c b/arch/arm/mach-at91rm9200/clock.c
index edc2cc837ae6..5b7892277bee 100644
--- a/arch/arm/mach-at91rm9200/clock.c
+++ b/arch/arm/mach-at91rm9200/clock.c
@@ -190,85 +190,85 @@ static void pmc_periph_mode(struct clk *clk, int is_on)
 static struct clk udc_clk = {
 	.name		= "udc_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_UDP,
+	.pmc_mask	= 1 << AT91RM9200_ID_UDP,
 	.mode		= pmc_periph_mode,
 };
 static struct clk ohci_clk = {
 	.name		= "ohci_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_UHP,
+	.pmc_mask	= 1 << AT91RM9200_ID_UHP,
 	.mode		= pmc_periph_mode,
 };
 static struct clk ether_clk = {
 	.name		= "ether_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_EMAC,
+	.pmc_mask	= 1 << AT91RM9200_ID_EMAC,
 	.mode		= pmc_periph_mode,
 };
 static struct clk mmc_clk = {
 	.name		= "mci_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_MCI,
+	.pmc_mask	= 1 << AT91RM9200_ID_MCI,
 	.mode		= pmc_periph_mode,
 };
 static struct clk twi_clk = {
 	.name		= "twi_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_TWI,
+	.pmc_mask	= 1 << AT91RM9200_ID_TWI,
 	.mode		= pmc_periph_mode,
 };
 static struct clk usart0_clk = {
 	.name		= "usart0_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_US0,
+	.pmc_mask	= 1 << AT91RM9200_ID_US0,
 	.mode		= pmc_periph_mode,
 };
 static struct clk usart1_clk = {
 	.name		= "usart1_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_US1,
+	.pmc_mask	= 1 << AT91RM9200_ID_US1,
 	.mode		= pmc_periph_mode,
 };
 static struct clk usart2_clk = {
 	.name		= "usart2_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_US2,
+	.pmc_mask	= 1 << AT91RM9200_ID_US2,
 	.mode		= pmc_periph_mode,
 };
 static struct clk usart3_clk = {
 	.name		= "usart3_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_US3,
+	.pmc_mask	= 1 << AT91RM9200_ID_US3,
 	.mode		= pmc_periph_mode,
 };
 static struct clk spi_clk = {
 	.name		= "spi0_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_SPI,
+	.pmc_mask	= 1 << AT91RM9200_ID_SPI,
 	.mode		= pmc_periph_mode,
 };
 static struct clk pioA_clk = {
 	.name		= "pioA_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_PIOA,
+	.pmc_mask	= 1 << AT91RM9200_ID_PIOA,
 	.mode		= pmc_periph_mode,
 };
 static struct clk pioB_clk = {
 	.name		= "pioB_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_PIOB,
+	.pmc_mask	= 1 << AT91RM9200_ID_PIOB,
 	.mode		= pmc_periph_mode,
 };
 static struct clk pioC_clk = {
 	.name		= "pioC_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_PIOC,
+	.pmc_mask	= 1 << AT91RM9200_ID_PIOC,
 	.mode		= pmc_periph_mode,
 };
 static struct clk pioD_clk = {
 	.name		= "pioD_clk",
 	.parent		= &mck,
-	.pmc_mask	= 1 << AT91_ID_PIOD,
+	.pmc_mask	= 1 << AT91RM9200_ID_PIOD,
 	.mode		= pmc_periph_mode,
 };
 
diff --git a/arch/arm/mach-at91rm9200/devices.c b/arch/arm/mach-at91rm9200/devices.c
index 4352acb88178..db7c684cdb63 100644
--- a/arch/arm/mach-at91rm9200/devices.c
+++ b/arch/arm/mach-at91rm9200/devices.c
@@ -35,13 +35,13 @@ static struct at91_usbh_data usbh_data;
 
 static struct resource at91_usbh_resources[] = {
 	[0] = {
-		.start	= AT91_UHP_BASE,
-		.end	= AT91_UHP_BASE + SZ_1M - 1,
+		.start	= AT91RM9200_UHP_BASE,
+		.end	= AT91RM9200_UHP_BASE + SZ_1M - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_UHP,
-		.end	= AT91_ID_UHP,
+		.start	= AT91RM9200_ID_UHP,
+		.end	= AT91RM9200_ID_UHP,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -80,13 +80,13 @@ static struct at91_udc_data udc_data;
 
 static struct resource at91_udc_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_UDP,
-		.end	= AT91_BASE_UDP + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_UDP,
+		.end	= AT91RM9200_BASE_UDP + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_UDP,
-		.end	= AT91_ID_UDP,
+		.start	= AT91RM9200_ID_UDP,
+		.end	= AT91RM9200_ID_UDP,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -131,13 +131,13 @@ static struct at91_eth_data eth_data;
 
 static struct resource at91_eth_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_EMAC,
-		.end	= AT91_BASE_EMAC + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_EMAC,
+		.end	= AT91RM9200_BASE_EMAC + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_EMAC,
-		.end	= AT91_ID_EMAC,
+		.start	= AT91RM9200_ID_EMAC,
+		.end	= AT91RM9200_ID_EMAC,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -263,13 +263,13 @@ static struct at91_mmc_data mmc_data;
 
 static struct resource at91_mmc_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_MCI,
-		.end	= AT91_BASE_MCI + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_MCI,
+		.end	= AT91RM9200_BASE_MCI + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_MCI,
-		.end	= AT91_ID_MCI,
+		.start	= AT91RM9200_ID_MCI,
+		.end	= AT91RM9200_ID_MCI,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -423,13 +423,13 @@ static u64 spi_dmamask = 0xffffffffUL;
 
 static struct resource at91_spi_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_SPI,
-		.end	= AT91_BASE_SPI + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_SPI,
+		.end	= AT91RM9200_BASE_SPI + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_SPI,
-		.end	= AT91_ID_SPI,
+		.start	= AT91RM9200_ID_SPI,
+		.end	= AT91RM9200_ID_SPI,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -582,13 +582,13 @@ static inline void configure_dbgu_pins(void)
 
 static struct resource uart0_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_US0,
-		.end	= AT91_BASE_US0 + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_US0,
+		.end	= AT91RM9200_BASE_US0 + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_US0,
-		.end	= AT91_ID_US0,
+		.start	= AT91RM9200_ID_US0,
+		.end	= AT91RM9200_ID_US0,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -624,13 +624,13 @@ static inline void configure_usart0_pins(void)
 
 static struct resource uart1_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_US1,
-		.end	= AT91_BASE_US1 + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_US1,
+		.end	= AT91RM9200_BASE_US1 + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_US1,
-		.end	= AT91_ID_US1,
+		.start	= AT91RM9200_ID_US1,
+		.end	= AT91RM9200_ID_US1,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -665,13 +665,13 @@ static inline void configure_usart1_pins(void)
 
 static struct resource uart2_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_US2,
-		.end	= AT91_BASE_US2 + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_US2,
+		.end	= AT91RM9200_BASE_US2 + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_US2,
-		.end	= AT91_ID_US2,
+		.start	= AT91RM9200_ID_US2,
+		.end	= AT91RM9200_ID_US2,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
@@ -700,13 +700,13 @@ static inline void configure_usart2_pins(void)
 
 static struct resource uart3_resources[] = {
 	[0] = {
-		.start	= AT91_BASE_US3,
-		.end	= AT91_BASE_US3 + SZ_16K - 1,
+		.start	= AT91RM9200_BASE_US3,
+		.end	= AT91RM9200_BASE_US3 + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= AT91_ID_US3,
-		.end	= AT91_ID_US3,
+		.start	= AT91RM9200_ID_US3,
+		.end	= AT91RM9200_ID_US3,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
diff --git a/arch/arm/mach-at91rm9200/gpio.c b/arch/arm/mach-at91rm9200/gpio.c
index cec199fd6721..8476cb8856c9 100644
--- a/arch/arm/mach-at91rm9200/gpio.c
+++ b/arch/arm/mach-at91rm9200/gpio.c
@@ -261,10 +261,10 @@ void at91_gpio_suspend(void)
 		at91_sys_write(pio_controller_offset[i] + PIO_IER, wakeups[i]);
 
 		if (!wakeups[i]) {
-			disable_irq_wake(AT91_ID_PIOA + i);
-			at91_sys_write(AT91_PMC_PCDR, 1 << (AT91_ID_PIOA + i));
+			disable_irq_wake(AT91RM9200_ID_PIOA + i);
+			at91_sys_write(AT91_PMC_PCDR, 1 << (AT91RM9200_ID_PIOA + i));
 		} else {
-			enable_irq_wake(AT91_ID_PIOA + i);
+			enable_irq_wake(AT91RM9200_ID_PIOA + i);
 #ifdef CONFIG_PM_DEBUG
 			printk(KERN_DEBUG "GPIO-%c may wake for %08x\n", "ABCD"[i], wakeups[i]);
 #endif
@@ -282,10 +282,10 @@ void at91_gpio_resume(void)
 	}
 
 	at91_sys_write(AT91_PMC_PCER,
-			  (1 << AT91_ID_PIOA)
-			| (1 << AT91_ID_PIOB)
-			| (1 << AT91_ID_PIOC)
-			| (1 << AT91_ID_PIOD));
+			  (1 << AT91RM9200_ID_PIOA)
+			| (1 << AT91RM9200_ID_PIOB)
+			| (1 << AT91RM9200_ID_PIOC)
+			| (1 << AT91RM9200_ID_PIOD));
 }
 
 #else
@@ -384,7 +384,7 @@ void __init at91_gpio_irq_setup(unsigned banks)
 
 	if (banks > 4)
 		banks = 4;
-	for (pioc = 0, pin = PIN_BASE, id = AT91_ID_PIOA;
+	for (pioc = 0, pin = PIN_BASE, id = AT91RM9200_ID_PIOA;
 			pioc < banks;
 			pioc++, id++) {
 		void __iomem	*controller;
diff --git a/arch/arm/mach-at91rm9200/irq.c b/arch/arm/mach-at91rm9200/irq.c
index c3a5e777f9f8..8cd60a99c5b8 100644
--- a/arch/arm/mach-at91rm9200/irq.c
+++ b/arch/arm/mach-at91rm9200/irq.c
@@ -61,12 +61,12 @@ static int at91_aic_set_type(unsigned irq, unsigned type)
 		srctype = AT91_AIC_SRCTYPE_RISING;
 		break;
 	case IRQT_LOW:
-		if ((irq > AT91_ID_FIQ) && (irq < AT91_ID_IRQ0))	/* only supported on external interrupts */
+		if ((irq > AT91_ID_FIQ) && (irq < AT91RM9200_ID_IRQ0))	/* only supported on external interrupts */
 			return -EINVAL;
 		srctype = AT91_AIC_SRCTYPE_LOW;
 		break;
 	case IRQT_FALLING:
-		if ((irq > AT91_ID_FIQ) && (irq < AT91_ID_IRQ0))	/* only supported on external interrupts */
+		if ((irq > AT91_ID_FIQ) && (irq < AT91RM9200_ID_IRQ0))	/* only supported on external interrupts */
 			return -EINVAL;
 		srctype = AT91_AIC_SRCTYPE_FALLING;
 		break;
diff --git a/arch/arm/mach-at91rm9200/pm.c b/arch/arm/mach-at91rm9200/pm.c
index 47e5480feb7e..32c95d8eaacf 100644
--- a/arch/arm/mach-at91rm9200/pm.c
+++ b/arch/arm/mach-at91rm9200/pm.c
@@ -123,13 +123,13 @@ static int at91_pm_enter(suspend_state_t state)
 			(at91_sys_read(AT91_PMC_PCSR)
 					| (1 << AT91_ID_FIQ)
 					| (1 << AT91_ID_SYS)
-					| (1 << AT91_ID_IRQ0)
-					| (1 << AT91_ID_IRQ1)
-					| (1 << AT91_ID_IRQ2)
-					| (1 << AT91_ID_IRQ3)
-					| (1 << AT91_ID_IRQ4)
-					| (1 << AT91_ID_IRQ5)
-					| (1 << AT91_ID_IRQ6))
+					| (1 << AT91RM9200_ID_IRQ0)
+					| (1 << AT91RM9200_ID_IRQ1)
+					| (1 << AT91RM9200_ID_IRQ2)
+					| (1 << AT91RM9200_ID_IRQ3)
+					| (1 << AT91RM9200_ID_IRQ4)
+					| (1 << AT91RM9200_ID_IRQ5)
+					| (1 << AT91RM9200_ID_IRQ6))
 				& at91_sys_read(AT91_AIC_IMR),
 			state);
 
diff --git a/drivers/mmc/at91_mci.c b/drivers/mmc/at91_mci.c
index 6b7638b84290..4b3d4d637030 100644
--- a/drivers/mmc/at91_mci.c
+++ b/drivers/mmc/at91_mci.c
@@ -850,7 +850,7 @@ static int at91_mci_probe(struct platform_device *pdev)
 	/*
 	 * Allocate the MCI interrupt
 	 */
-	ret = request_irq(AT91_ID_MCI, at91_mci_irq, IRQF_SHARED, DRIVER_NAME, host);
+	ret = request_irq(AT91RM9200_ID_MCI, at91_mci_irq, IRQF_SHARED, DRIVER_NAME, host);
 	if (ret) {
 		printk(KERN_ERR "Failed to request MCI interrupt\n");
 		clk_disable(mci_clk);
@@ -906,7 +906,7 @@ static int at91_mci_remove(struct platform_device *pdev)
 
 	mmc_remove_host(mmc);
 	at91_mci_disable();
-	free_irq(AT91_ID_MCI, host);
+	free_irq(AT91RM9200_ID_MCI, host);
 	mmc_free_host(mmc);
 
 	clk_disable(mci_clk);				/* Disable the peripheral clock */
diff --git a/drivers/net/arm/at91_ether.c b/drivers/net/arm/at91_ether.c
index 95b28aa01f4f..3ecf2cc53a7c 100644
--- a/drivers/net/arm/at91_ether.c
+++ b/drivers/net/arm/at91_ether.c
@@ -947,7 +947,7 @@ static int __init at91ether_setup(unsigned long phy_type, unsigned short phy_add
 		return -ENOMEM;
 
 	dev->base_addr = AT91_VA_BASE_EMAC;
-	dev->irq = AT91_ID_EMAC;
+	dev->irq = AT91RM9200_ID_EMAC;
 	SET_MODULE_OWNER(dev);
 
 	/* Install the interrupt handler */
diff --git a/drivers/serial/at91_serial.c b/drivers/serial/at91_serial.c
index 54c6b2adf7b7..bf4bf103e5a0 100644
--- a/drivers/serial/at91_serial.c
+++ b/drivers/serial/at91_serial.c
@@ -139,7 +139,7 @@ static void at91_set_mctrl(struct uart_port *port, u_int mctrl)
 		 * AT91RM9200 Errata #39: RTS0 is not internally connected to PA21.
 		 *  We need to drive the pin manually.
 		 */
-		if (port->mapbase == AT91_BASE_US0) {
+		if (port->mapbase == AT91RM9200_BASE_US0) {
 			if (mctrl & TIOCM_RTS)
 				at91_set_gpio_value(AT91_PIN_PA21, 0);
 			else
diff --git a/drivers/usb/gadget/at91_udc.c b/drivers/usb/gadget/at91_udc.c
index cfebca05ead5..043744f1ab72 100644
--- a/drivers/usb/gadget/at91_udc.c
+++ b/drivers/usb/gadget/at91_udc.c
@@ -1658,7 +1658,7 @@ static int __devinit at91udc_probe(struct platform_device *pdev)
 		return -ENODEV;
 	}
 
-	if (!request_mem_region(AT91_BASE_UDP, SZ_16K, driver_name)) {
+	if (!request_mem_region(AT91RM9200_BASE_UDP, SZ_16K, driver_name)) {
 		DBG("someone's using UDC memory\n");
 		return -EBUSY;
 	}
@@ -1720,7 +1720,7 @@ static int __devinit at91udc_probe(struct platform_device *pdev)
 fail1:
 	device_unregister(&udc->gadget.dev);
 fail0:
-	release_mem_region(AT91_BASE_UDP, SZ_16K);
+	release_mem_region(AT91RM9200_BASE_UDP, SZ_16K);
 	DBG("%s probe failed, %d\n", driver_name, retval);
 	return retval;
 }
@@ -1742,7 +1742,7 @@ static int __devexit at91udc_remove(struct platform_device *pdev)
 		free_irq(udc->board.vbus_pin, udc);
 	free_irq(udc->udp_irq, udc);
 	device_unregister(&udc->gadget.dev);
-	release_mem_region(AT91_BASE_UDP, SZ_16K);
+	release_mem_region(AT91RM9200_BASE_UDP, SZ_16K);
 
 	clk_put(udc->iclk);
 	clk_put(udc->fclk);
diff --git a/include/asm-arm/arch-at91rm9200/at91rm9200.h b/include/asm-arm/arch-at91rm9200/at91rm9200.h
index 58f40931a5c1..a5a86b1ff886 100644
--- a/include/asm-arm/arch-at91rm9200/at91rm9200.h
+++ b/include/asm-arm/arch-at91rm9200/at91rm9200.h
@@ -19,66 +19,79 @@
 /*
  * Peripheral identifiers/interrupts.
  */
-#define AT91_ID_FIQ	0	/* Advanced Interrupt Controller (FIQ) */
-#define AT91_ID_SYS	1	/* System Peripheral */
-#define AT91_ID_PIOA	2	/* Parallel IO Controller A */
-#define AT91_ID_PIOB	3	/* Parallel IO Controller B */
-#define AT91_ID_PIOC	4	/* Parallel IO Controller C */
-#define AT91_ID_PIOD	5	/* Parallel IO Controller D */
-#define AT91_ID_US0	6	/* USART 0 */
-#define AT91_ID_US1	7	/* USART 1 */
-#define AT91_ID_US2	8	/* USART 2 */
-#define AT91_ID_US3	9	/* USART 3 */
-#define AT91_ID_MCI	10	/* Multimedia Card Interface */
-#define AT91_ID_UDP	11	/* USB Device Port */
-#define AT91_ID_TWI	12	/* Two-Wire Interface */
-#define AT91_ID_SPI	13	/* Serial Peripheral Interface */
-#define AT91_ID_SSC0	14	/* Serial Synchronous Controller 0 */
-#define AT91_ID_SSC1	15	/* Serial Synchronous Controller 1 */
-#define AT91_ID_SSC2	16	/* Serial Synchronous Controller 2 */
-#define AT91_ID_TC0	17	/* Timer Counter 0 */
-#define AT91_ID_TC1	18	/* Timer Counter 1 */
-#define AT91_ID_TC2	19	/* Timer Counter 2 */
-#define AT91_ID_TC3	20	/* Timer Counter 3 */
-#define AT91_ID_TC4	21	/* Timer Counter 4 */
-#define AT91_ID_TC5	22	/* Timer Counter 5 */
-#define AT91_ID_UHP	23	/* USB Host port */
-#define AT91_ID_EMAC	24	/* Ethernet MAC */
-#define AT91_ID_IRQ0	25	/* Advanced Interrupt Controller (IRQ0) */
-#define AT91_ID_IRQ1	26	/* Advanced Interrupt Controller (IRQ1) */
-#define AT91_ID_IRQ2	27	/* Advanced Interrupt Controller (IRQ2) */
-#define AT91_ID_IRQ3	28	/* Advanced Interrupt Controller (IRQ3) */
-#define AT91_ID_IRQ4	29	/* Advanced Interrupt Controller (IRQ4) */
-#define AT91_ID_IRQ5	30	/* Advanced Interrupt Controller (IRQ5) */
-#define AT91_ID_IRQ6	31	/* Advanced Interrupt Controller (IRQ6) */
+#define AT91_ID_FIQ		0	/* Advanced Interrupt Controller (FIQ) */
+#define AT91_ID_SYS		1	/* System Peripheral */
+#define AT91RM9200_ID_PIOA	2	/* Parallel IO Controller A */
+#define AT91RM9200_ID_PIOB	3	/* Parallel IO Controller B */
+#define AT91RM9200_ID_PIOC	4	/* Parallel IO Controller C */
+#define AT91RM9200_ID_PIOD	5	/* Parallel IO Controller D */
+#define AT91RM9200_ID_US0	6	/* USART 0 */
+#define AT91RM9200_ID_US1	7	/* USART 1 */
+#define AT91RM9200_ID_US2	8	/* USART 2 */
+#define AT91RM9200_ID_US3	9	/* USART 3 */
+#define AT91RM9200_ID_MCI	10	/* Multimedia Card Interface */
+#define AT91RM9200_ID_UDP	11	/* USB Device Port */
+#define AT91RM9200_ID_TWI	12	/* Two-Wire Interface */
+#define AT91RM9200_ID_SPI	13	/* Serial Peripheral Interface */
+#define AT91RM9200_ID_SSC0	14	/* Serial Synchronous Controller 0 */
+#define AT91RM9200_ID_SSC1	15	/* Serial Synchronous Controller 1 */
+#define AT91RM9200_ID_SSC2	16	/* Serial Synchronous Controller 2 */
+#define AT91RM9200_ID_TC0	17	/* Timer Counter 0 */
+#define AT91RM9200_ID_TC1	18	/* Timer Counter 1 */
+#define AT91RM9200_ID_TC2	19	/* Timer Counter 2 */
+#define AT91RM9200_ID_TC3	20	/* Timer Counter 3 */
+#define AT91RM9200_ID_TC4	21	/* Timer Counter 4 */
+#define AT91RM9200_ID_TC5	22	/* Timer Counter 5 */
+#define AT91RM9200_ID_UHP	23	/* USB Host port */
+#define AT91RM9200_ID_EMAC	24	/* Ethernet MAC */
+#define AT91RM9200_ID_IRQ0	25	/* Advanced Interrupt Controller (IRQ0) */
+#define AT91RM9200_ID_IRQ1	26	/* Advanced Interrupt Controller (IRQ1) */
+#define AT91RM9200_ID_IRQ2	27	/* Advanced Interrupt Controller (IRQ2) */
+#define AT91RM9200_ID_IRQ3	28	/* Advanced Interrupt Controller (IRQ3) */
+#define AT91RM9200_ID_IRQ4	29	/* Advanced Interrupt Controller (IRQ4) */
+#define AT91RM9200_ID_IRQ5	30	/* Advanced Interrupt Controller (IRQ5) */
+#define AT91RM9200_ID_IRQ6	31	/* Advanced Interrupt Controller (IRQ6) */
 
 
 /*
  * Peripheral physical base addresses.
  */
-#define AT91_BASE_TCB0		0xfffa0000
-#define AT91_BASE_TC0		0xfffa0000
-#define AT91_BASE_TC1		0xfffa0040
-#define AT91_BASE_TC2		0xfffa0080
-#define AT91_BASE_TCB1		0xfffa4000
-#define AT91_BASE_TC3		0xfffa4000
-#define AT91_BASE_TC4		0xfffa4040
-#define AT91_BASE_TC5		0xfffa4080
-#define AT91_BASE_UDP		0xfffb0000
-#define AT91_BASE_MCI		0xfffb4000
-#define AT91_BASE_TWI		0xfffb8000
-#define AT91_BASE_EMAC		0xfffbc000
-#define AT91_BASE_US0		0xfffc0000
-#define AT91_BASE_US1		0xfffc4000
-#define AT91_BASE_US2		0xfffc8000
-#define AT91_BASE_US3		0xfffcc000
-#define AT91_BASE_SSC0		0xfffd0000
-#define AT91_BASE_SSC1		0xfffd4000
-#define AT91_BASE_SSC2		0xfffd8000
-#define AT91_BASE_SPI		0xfffe0000
+#define AT91RM9200_BASE_TCB0	0xfffa0000
+#define AT91RM9200_BASE_TC0	0xfffa0000
+#define AT91RM9200_BASE_TC1	0xfffa0040
+#define AT91RM9200_BASE_TC2	0xfffa0080
+#define AT91RM9200_BASE_TCB1	0xfffa4000
+#define AT91RM9200_BASE_TC3	0xfffa4000
+#define AT91RM9200_BASE_TC4	0xfffa4040
+#define AT91RM9200_BASE_TC5	0xfffa4080
+#define AT91RM9200_BASE_UDP	0xfffb0000
+#define AT91RM9200_BASE_MCI	0xfffb4000
+#define AT91RM9200_BASE_TWI	0xfffb8000
+#define AT91RM9200_BASE_EMAC	0xfffbc000
+#define AT91RM9200_BASE_US0	0xfffc0000
+#define AT91RM9200_BASE_US1	0xfffc4000
+#define AT91RM9200_BASE_US2	0xfffc8000
+#define AT91RM9200_BASE_US3	0xfffcc000
+#define AT91RM9200_BASE_SSC0	0xfffd0000
+#define AT91RM9200_BASE_SSC1	0xfffd4000
+#define AT91RM9200_BASE_SSC2	0xfffd8000
+#define AT91RM9200_BASE_SPI	0xfffe0000
 #define AT91_BASE_SYS		0xfffff000
 
 
+/*
+ * Internal Memory.
+ */
+#define AT91RM9200_ROM_BASE	0x00100000	/* Internal ROM base address */
+#define AT91RM9200_ROM_SIZE	SZ_128K		/* Internal ROM size (128Kb) */
+
+#define AT91RM9200_SRAM_BASE	0x00200000	/* Internal SRAM base address */
+#define AT91RM9200_SRAM_SIZE	SZ_16K		/* Internal SRAM size (16Kb) */
+
+#define AT91RM9200_UHP_BASE	0x00300000	/* USB Host controller */
+
+
+#if 0
 /*
  * PIO pin definitions (peripheral A/B multiplexing).
  */
@@ -257,5 +270,6 @@
 #define AT91_PD25_TPK13		(1 << 25)	/* B: ETM Trace Packet Port 13 */
 #define AT91_PD26_TPK14		(1 << 26)	/* B: ETM Trace Packet Port 14 */
 #define AT91_PD27_TPK15		(1 << 27)	/* B: ETM Trace Packet Port 15 */
+#endif
 
 #endif
diff --git a/include/asm-arm/arch-at91rm9200/gpio.h b/include/asm-arm/arch-at91rm9200/gpio.h
index dbde1baaf251..6243f28a0b81 100644
--- a/include/asm-arm/arch-at91rm9200/gpio.h
+++ b/include/asm-arm/arch-at91rm9200/gpio.h
@@ -20,7 +20,7 @@
 #define PQFP_GPIO_BANKS		3	/* PQFP package has 3 banks */
 #define BGA_GPIO_BANKS		4	/* BGA package has 4 banks */
 
-/* these pin numbers double as IRQ numbers, like AT91_ID_* values */
+/* these pin numbers double as IRQ numbers, like AT91xxx_ID_* values */
 
 #define	AT91_PIN_PA0	(PIN_BASE + 0x00 + 0)
 #define	AT91_PIN_PA1	(PIN_BASE + 0x00 + 1)
diff --git a/include/asm-arm/arch-at91rm9200/hardware.h b/include/asm-arm/arch-at91rm9200/hardware.h
index 235d39d91107..878e65f369bf 100644
--- a/include/asm-arm/arch-at91rm9200/hardware.h
+++ b/include/asm-arm/arch-at91rm9200/hardware.h
@@ -34,27 +34,23 @@
  * Virtual to Physical Address mapping for IO devices.
  */
 #define AT91_VA_BASE_SYS	AT91_IO_P2V(AT91_BASE_SYS)
-#define AT91_VA_BASE_SPI	AT91_IO_P2V(AT91_BASE_SPI)
-#define AT91_VA_BASE_SSC2	AT91_IO_P2V(AT91_BASE_SSC2)
-#define AT91_VA_BASE_SSC1	AT91_IO_P2V(AT91_BASE_SSC1)
-#define AT91_VA_BASE_SSC0	AT91_IO_P2V(AT91_BASE_SSC0)
-#define AT91_VA_BASE_US3	AT91_IO_P2V(AT91_BASE_US3)
-#define AT91_VA_BASE_US2	AT91_IO_P2V(AT91_BASE_US2)
-#define AT91_VA_BASE_US1	AT91_IO_P2V(AT91_BASE_US1)
-#define AT91_VA_BASE_US0	AT91_IO_P2V(AT91_BASE_US0)
-#define AT91_VA_BASE_EMAC	AT91_IO_P2V(AT91_BASE_EMAC)
-#define AT91_VA_BASE_TWI	AT91_IO_P2V(AT91_BASE_TWI)
-#define AT91_VA_BASE_MCI	AT91_IO_P2V(AT91_BASE_MCI)
-#define AT91_VA_BASE_UDP	AT91_IO_P2V(AT91_BASE_UDP)
-#define AT91_VA_BASE_TCB1	AT91_IO_P2V(AT91_BASE_TCB1)
-#define AT91_VA_BASE_TCB0	AT91_IO_P2V(AT91_BASE_TCB0)
-
-/* Internal SRAM */
-#define AT91_SRAM_BASE		0x00200000	/* Internal SRAM base address */
-#define AT91_SRAM_SIZE		0x00004000	/* Internal SRAM SIZE (16Kb) */
+#define AT91_VA_BASE_SPI	AT91_IO_P2V(AT91RM9200_BASE_SPI)
+#define AT91_VA_BASE_SSC2	AT91_IO_P2V(AT91RM9200_BASE_SSC2)
+#define AT91_VA_BASE_SSC1	AT91_IO_P2V(AT91RM9200_BASE_SSC1)
+#define AT91_VA_BASE_SSC0	AT91_IO_P2V(AT91RM9200_BASE_SSC0)
+#define AT91_VA_BASE_US3	AT91_IO_P2V(AT91RM9200_BASE_US3)
+#define AT91_VA_BASE_US2	AT91_IO_P2V(AT91RM9200_BASE_US2)
+#define AT91_VA_BASE_US1	AT91_IO_P2V(AT91RM9200_BASE_US1)
+#define AT91_VA_BASE_US0	AT91_IO_P2V(AT91RM9200_BASE_US0)
+#define AT91_VA_BASE_EMAC	AT91_IO_P2V(AT91RM9200_BASE_EMAC)
+#define AT91_VA_BASE_TWI	AT91_IO_P2V(AT91RM9200_BASE_TWI)
+#define AT91_VA_BASE_MCI	AT91_IO_P2V(AT91RM9200_BASE_MCI)
+#define AT91_VA_BASE_UDP	AT91_IO_P2V(AT91RM9200_BASE_UDP)
+#define AT91_VA_BASE_TCB1	AT91_IO_P2V(AT91RM9200_BASE_TCB1)
+#define AT91_VA_BASE_TCB0	AT91_IO_P2V(AT91RM9200_BASE_TCB0)
 
  /* Internal SRAM is mapped below the IO devices */
-#define AT91_SRAM_VIRT_BASE	(AT91_IO_VIRT_BASE - AT91_SRAM_SIZE)
+#define AT91_SRAM_VIRT_BASE	(AT91_IO_VIRT_BASE - AT91RM9200_SRAM_SIZE)
 
 /* Serial ports */
 #define AT91_NR_UART		5		/* 4 USART3's and one DBGU port */
@@ -71,9 +67,6 @@
 /* Compact Flash */
 #define AT91_CF_BASE		0x50000000	/* NCS4-NCS6: Compact Flash physical base address */
 
-/* Multi-Master Memory controller */
-#define AT91_UHP_BASE		0x00300000	/* USB Host controller */
-
 /* Clocks */
 #define AT91_SLOW_CLOCK		32768		/* slow clock */
 
diff --git a/include/asm-arm/arch-at91rm9200/irqs.h b/include/asm-arm/arch-at91rm9200/irqs.h
index f63842c2c093..763cb96c418b 100644
--- a/include/asm-arm/arch-at91rm9200/irqs.h
+++ b/include/asm-arm/arch-at91rm9200/irqs.h
@@ -32,7 +32,7 @@
 
 
 /*
- * IRQ interrupt symbols are the AT91_ID_* symbols in at91rm9200.h
+ * IRQ interrupt symbols are the AT91xxx_ID_* symbols
  * for IRQs handled directly through the AIC, or else the AT91_PIN_*
  * symbols in gpio.h for ones handled indirectly as GPIOs.
  * We make provision for 4 banks of GPIO.
-- 
cgit v1.2.3


From f21738341ca330ec83ef978ee63ffa5ecf13f082 Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Wed, 27 Sep 2006 13:23:00 +0100
Subject: [ARM] 3867/1: AT91 GPIO update

This patch makes the AT91 gpio.c support processor-generic (AT91RM9200
and AT91SAM9xxx).  The GPIO controllers supported by a particular AT91
processor are defined in the processor-specific file and are registered
with gpio.c at startup.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91rm9200/at91rm9200.c    | 39 +++++++++++++-
 arch/arm/mach-at91rm9200/board-1arm.c    | 15 +++---
 arch/arm/mach-at91rm9200/board-carmeva.c | 16 +++---
 arch/arm/mach-at91rm9200/board-csb337.c  | 15 +++---
 arch/arm/mach-at91rm9200/board-csb637.c  | 15 +++---
 arch/arm/mach-at91rm9200/board-dk.c      | 15 +++---
 arch/arm/mach-at91rm9200/board-eb9200.c  | 15 +++---
 arch/arm/mach-at91rm9200/board-ek.c      | 15 +++---
 arch/arm/mach-at91rm9200/board-kafa.c    | 15 +++---
 arch/arm/mach-at91rm9200/board-kb9202.c  | 15 +++---
 arch/arm/mach-at91rm9200/generic.h       | 16 ++++--
 arch/arm/mach-at91rm9200/gpio.c          | 89 ++++++++++++++++++--------------
 arch/arm/mach-at91rm9200/irq.c           |  2 -
 include/asm-arm/arch-at91rm9200/gpio.h   | 16 +++---
 14 files changed, 164 insertions(+), 134 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-at91rm9200/at91rm9200.c b/arch/arm/mach-at91rm9200/at91rm9200.c
index ae04cbdf0bff..c32d0c996f90 100644
--- a/arch/arm/mach-at91rm9200/at91rm9200.c
+++ b/arch/arm/mach-at91rm9200/at91rm9200.c
@@ -243,11 +243,34 @@ static void __init at91rm9200_register_clocks(void)
 	clk_register(&pck3);
 }
 
+/* --------------------------------------------------------------------
+ *  GPIO
+ * -------------------------------------------------------------------- */
+
+static struct at91_gpio_bank at91rm9200_gpio[] = {
+	{
+		.id		= AT91RM9200_ID_PIOA,
+		.offset		= AT91_PIOA,
+		.clock		= &pioA_clk,
+	}, {
+		.id		= AT91RM9200_ID_PIOB,
+		.offset		= AT91_PIOB,
+		.clock		= &pioB_clk,
+	}, {
+		.id		= AT91RM9200_ID_PIOC,
+		.offset		= AT91_PIOC,
+		.clock		= &pioC_clk,
+	}, {
+		.id		= AT91RM9200_ID_PIOD,
+		.offset		= AT91_PIOD,
+		.clock		= &pioD_clk,
+	}
+};
 
 /* --------------------------------------------------------------------
  *  AT91RM9200 processor initialization
  * -------------------------------------------------------------------- */
-void __init at91rm9200_initialize(unsigned long main_clock)
+void __init at91rm9200_initialize(unsigned long main_clock, unsigned short banks)
 {
 	/* Map peripherals */
 	iotable_init(at91rm9200_io_desc, ARRAY_SIZE(at91rm9200_io_desc));
@@ -257,8 +280,16 @@ void __init at91rm9200_initialize(unsigned long main_clock)
 
 	/* Register the processor-specific clocks */
 	at91rm9200_register_clocks();
+
+	/* Initialize GPIO subsystem */
+	at91_gpio_init(at91rm9200_gpio, banks);
 }
 
+
+/* --------------------------------------------------------------------
+ *  Interrupt initialization
+ * -------------------------------------------------------------------- */
+
 /*
  * The default interrupt priority levels (0 = lowest, 7 = highest).
  */
@@ -297,10 +328,14 @@ static unsigned int at91rm9200_default_irq_priority[NR_AIC_IRQS] __initdata = {
 	0	/* Advanced Interrupt Controller (IRQ6) */
 };
 
-void __init at91rm9200_init_irq(unsigned int priority[NR_AIC_IRQS])
+void __init at91rm9200_init_interrupts(unsigned int priority[NR_AIC_IRQS])
 {
 	if (!priority)
 		priority = at91rm9200_default_irq_priority;
 
+	/* Initialize the AIC interrupt controller */
 	at91_aic_init(priority);
+
+	/* Enable GPIO interrupts */
+	at91_gpio_irq_setup();
 }
diff --git a/arch/arm/mach-at91rm9200/board-1arm.c b/arch/arm/mach-at91rm9200/board-1arm.c
index d2aa9c4a8b24..36eecd7161f5 100644
--- a/arch/arm/mach-at91rm9200/board-1arm.c
+++ b/arch/arm/mach-at91rm9200/board-1arm.c
@@ -39,14 +39,6 @@
 
 #include "generic.h"
 
-static void __init onearm_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(PQFP_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -62,12 +54,17 @@ static struct at91_uart_config __initdata onearm_uart_config = {
 static void __init onearm_map_io(void)
 {
 	/* Initialize processor: 18.432 MHz crystal */
-	at91rm9200_initialize(18432000);
+	at91rm9200_initialize(18432000, AT91RM9200_PQFP);
 
 	/* Setup the serial ports and console */
 	at91_init_serial(&onearm_uart_config);
 }
 
+static void __init onearm_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata onearm_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC4,
 	.is_rmii	= 1,
diff --git a/arch/arm/mach-at91rm9200/board-carmeva.c b/arch/arm/mach-at91rm9200/board-carmeva.c
index 6af252bcb083..50e513681ae6 100644
--- a/arch/arm/mach-at91rm9200/board-carmeva.c
+++ b/arch/arm/mach-at91rm9200/board-carmeva.c
@@ -40,14 +40,6 @@
 
 #include "generic.h"
 
-static void __init carmeva_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(BGA_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -63,12 +55,18 @@ static struct at91_uart_config __initdata carmeva_uart_config = {
 static void __init carmeva_map_io(void)
 {
 	/* Initialize processor: 20.000 MHz crystal */
-	at91rm9200_initialize(20000000);
+	at91rm9200_initialize(20000000, AT91RM9200_BGA);
 
 	/* Setup the serial ports and console */
 	at91_init_serial(&carmeva_uart_config);
 }
 
+static void __init carmeva_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
+
 static struct at91_eth_data __initdata carmeva_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC4,
 	.is_rmii	= 1,
diff --git a/arch/arm/mach-at91rm9200/board-csb337.c b/arch/arm/mach-at91rm9200/board-csb337.c
index f74b483329c0..8eeae491ce71 100644
--- a/arch/arm/mach-at91rm9200/board-csb337.c
+++ b/arch/arm/mach-at91rm9200/board-csb337.c
@@ -39,14 +39,6 @@
 
 #include "generic.h"
 
-static void __init csb337_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(BGA_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -62,7 +54,7 @@ static struct at91_uart_config __initdata csb337_uart_config = {
 static void __init csb337_map_io(void)
 {
 	/* Initialize processor: 3.6864 MHz crystal */
-	at91rm9200_initialize(3686400);
+	at91rm9200_initialize(3686400, AT91RM9200_BGA);
 
 	/* Setup the LEDs */
 	at91_init_leds(AT91_PIN_PB0, AT91_PIN_PB1);
@@ -71,6 +63,11 @@ static void __init csb337_map_io(void)
 	at91_init_serial(&csb337_uart_config);
 }
 
+static void __init csb337_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata csb337_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC2,
 	.is_rmii	= 0,
diff --git a/arch/arm/mach-at91rm9200/board-csb637.c b/arch/arm/mach-at91rm9200/board-csb637.c
index 880350997a3b..a29fa0e822ce 100644
--- a/arch/arm/mach-at91rm9200/board-csb637.c
+++ b/arch/arm/mach-at91rm9200/board-csb637.c
@@ -38,14 +38,6 @@
 
 #include "generic.h"
 
-static void __init csb637_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(BGA_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -61,7 +53,7 @@ static struct at91_uart_config __initdata csb637_uart_config = {
 static void __init csb637_map_io(void)
 {
 	/* Initialize processor: 3.6864 MHz crystal */
-	at91rm9200_initialize(3686400);
+	at91rm9200_initialize(3686400, AT91RM9200_BGA);
 
 	/* Setup the LEDs */
 	at91_init_leds(AT91_PIN_PB2, AT91_PIN_PB2);
@@ -70,6 +62,11 @@ static void __init csb637_map_io(void)
 	at91_init_serial(&csb637_uart_config);
 }
 
+static void __init csb637_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata csb637_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC0,
 	.is_rmii	= 0,
diff --git a/arch/arm/mach-at91rm9200/board-dk.c b/arch/arm/mach-at91rm9200/board-dk.c
index 23ea9d63b986..54c32d00b610 100644
--- a/arch/arm/mach-at91rm9200/board-dk.c
+++ b/arch/arm/mach-at91rm9200/board-dk.c
@@ -42,14 +42,6 @@
 
 #include "generic.h"
 
-static void __init dk_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(BGA_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -65,7 +57,7 @@ static struct at91_uart_config __initdata dk_uart_config = {
 static void __init dk_map_io(void)
 {
 	/* Initialize processor: 18.432 MHz crystal */
-	at91rm9200_initialize(18432000);
+	at91rm9200_initialize(18432000, AT91RM9200_BGA);
 
 	/* Setup the LEDs */
 	at91_init_leds(AT91_PIN_PB2, AT91_PIN_PB2);
@@ -74,6 +66,11 @@ static void __init dk_map_io(void)
 	at91_init_serial(&dk_uart_config);
 }
 
+static void __init dk_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata dk_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC4,
 	.is_rmii	= 1,
diff --git a/arch/arm/mach-at91rm9200/board-eb9200.c b/arch/arm/mach-at91rm9200/board-eb9200.c
index 2ceb267585ad..c6e0d51fbea0 100644
--- a/arch/arm/mach-at91rm9200/board-eb9200.c
+++ b/arch/arm/mach-at91rm9200/board-eb9200.c
@@ -40,14 +40,6 @@
 
 #include "generic.h"
 
-static void __init eb9200_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(BGA_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -63,12 +55,17 @@ static struct at91_uart_config __initdata eb9200_uart_config = {
 static void __init eb9200_map_io(void)
 {
 	/* Initialize processor: 18.432 MHz crystal */
-	at91rm9200_initialize(18432000);
+	at91rm9200_initialize(18432000, AT91RM9200_BGA);
 
 	/* Setup the serial ports and console */
 	at91_init_serial(&eb9200_uart_config);
 }
 
+static void __init eb9200_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata eb9200_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC4,
 	.is_rmii	= 1,
diff --git a/arch/arm/mach-at91rm9200/board-ek.c b/arch/arm/mach-at91rm9200/board-ek.c
index 8cd83364cd80..830eb7932178 100644
--- a/arch/arm/mach-at91rm9200/board-ek.c
+++ b/arch/arm/mach-at91rm9200/board-ek.c
@@ -42,14 +42,6 @@
 
 #include "generic.h"
 
-static void __init ek_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(BGA_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -65,7 +57,7 @@ static struct at91_uart_config __initdata ek_uart_config = {
 static void __init ek_map_io(void)
 {
 	/* Initialize processor: 18.432 MHz crystal */
-	at91rm9200_initialize(18432000);
+	at91rm9200_initialize(18432000, AT91RM9200_BGA);
 
 	/* Setup the LEDs */
 	at91_init_leds(AT91_PIN_PB1, AT91_PIN_PB2);
@@ -74,6 +66,11 @@ static void __init ek_map_io(void)
 	at91_init_serial(&ek_uart_config);
 }
 
+static void __init ek_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata ek_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC4,
 	.is_rmii	= 1,
diff --git a/arch/arm/mach-at91rm9200/board-kafa.c b/arch/arm/mach-at91rm9200/board-kafa.c
index d9a6b7e7d63a..91e301924f2c 100644
--- a/arch/arm/mach-at91rm9200/board-kafa.c
+++ b/arch/arm/mach-at91rm9200/board-kafa.c
@@ -39,14 +39,6 @@
 
 #include "generic.h"
 
-static void __init kafa_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(PQFP_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -62,7 +54,7 @@ static struct at91_uart_config __initdata kafa_uart_config = {
 static void __init kafa_map_io(void)
 {
 	/* Initialize processor: 18.432 MHz crystal */
-	at91rm9200_initialize(18432000);
+	at91rm9200_initialize(18432000, AT91RM9200_PQFP);
 
 	/* Set up the LEDs */
 	at91_init_leds(AT91_PIN_PB4, AT91_PIN_PB4);
@@ -71,6 +63,11 @@ static void __init kafa_map_io(void)
 	at91_init_serial(&kafa_uart_config);
 }
 
+static void __init kafa_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata kafa_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PC4,
 	.is_rmii	= 0,
diff --git a/arch/arm/mach-at91rm9200/board-kb9202.c b/arch/arm/mach-at91rm9200/board-kb9202.c
index 935238ffa330..55b08f9b9537 100644
--- a/arch/arm/mach-at91rm9200/board-kb9202.c
+++ b/arch/arm/mach-at91rm9200/board-kb9202.c
@@ -40,14 +40,6 @@
 
 #include "generic.h"
 
-static void __init kb9202_init_irq(void)
-{
-	/* Initialize AIC controller */
-	at91rm9200_init_irq(NULL);
-
-	/* Set up the GPIO interrupts */
-	at91_gpio_irq_setup(PQFP_GPIO_BANKS);
-}
 
 /*
  * Serial port configuration.
@@ -63,7 +55,7 @@ static struct at91_uart_config __initdata kb9202_uart_config = {
 static void __init kb9202_map_io(void)
 {
 	/* Initialize processor: 10 MHz crystal */
-	at91rm9200_initialize(10000000);
+	at91rm9200_initialize(10000000, AT91RM9200_PQFP);
 
 	/* Set up the LEDs */
 	at91_init_leds(AT91_PIN_PC19, AT91_PIN_PC18);
@@ -72,6 +64,11 @@ static void __init kb9202_map_io(void)
 	at91_init_serial(&kb9202_uart_config);
 }
 
+static void __init kb9202_init_irq(void)
+{
+	at91rm9200_init_interrupts(NULL);
+}
+
 static struct at91_eth_data __initdata kb9202_eth_data = {
 	.phy_irq_pin	= AT91_PIN_PB29,
 	.is_rmii	= 0,
diff --git a/arch/arm/mach-at91rm9200/generic.h b/arch/arm/mach-at91rm9200/generic.h
index c44e0e074682..694e411e285f 100644
--- a/arch/arm/mach-at91rm9200/generic.h
+++ b/arch/arm/mach-at91rm9200/generic.h
@@ -9,12 +9,11 @@
  */
 
  /* Processors */
-extern void __init at91rm9200_initialize(unsigned long main_clock);
+extern void __init at91rm9200_initialize(unsigned long main_clock, unsigned short banks);
 
  /* Interrupts */
-extern void __init at91rm9200_init_irq(unsigned int priority[]);
+extern void __init at91rm9200_init_interrupts(unsigned int priority[]);
 extern void __init at91_aic_init(unsigned int priority[]);
-extern void __init at91_gpio_irq_setup(unsigned banks);
 
  /* Timer */
 struct sys_timer;
@@ -29,3 +28,14 @@ extern void __init at91_clock_associate(const char *id, struct device *dev, cons
 extern void at91_irq_suspend(void);
 extern void at91_irq_resume(void);
 
+ /* GPIO */
+#define AT91RM9200_PQFP		3	/* AT91RM9200 PQFP package has 3 banks */
+#define AT91RM9200_BGA		4	/* AT91RM9200 BGA package has 4 banks */
+
+struct at91_gpio_bank {
+	unsigned short id;		/* peripheral ID */
+	unsigned long offset;		/* offset from system peripheral base */
+	struct clk *clock;		/* associated clock */
+};
+extern void __init at91_gpio_init(struct at91_gpio_bank *, int nr_banks);
+extern void __init at91_gpio_irq_setup(void);
diff --git a/arch/arm/mach-at91rm9200/gpio.c b/arch/arm/mach-at91rm9200/gpio.c
index 8476cb8856c9..58c9bf5e9520 100644
--- a/arch/arm/mach-at91rm9200/gpio.c
+++ b/arch/arm/mach-at91rm9200/gpio.c
@@ -9,6 +9,7 @@
  * (at your option) any later version.
  */
 
+#include <linux/clk.h>
 #include <linux/errno.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>
@@ -20,12 +21,12 @@
 #include <asm/hardware.h>
 #include <asm/arch/gpio.h>
 
-static const u32 pio_controller_offset[4] = {
-	AT91_PIOA,
-	AT91_PIOB,
-	AT91_PIOC,
-	AT91_PIOD,
-};
+#include "generic.h"
+
+
+static struct at91_gpio_bank *gpio;
+static int gpio_banks;
+
 
 static inline void __iomem *pin_to_controller(unsigned pin)
 {
@@ -33,8 +34,8 @@ static inline void __iomem *pin_to_controller(unsigned pin)
 
 	pin -= PIN_BASE;
 	pin /= 32;
-	if (likely(pin < BGA_GPIO_BANKS))
-		return sys_base + pio_controller_offset[pin];
+	if (likely(pin < gpio_banks))
+		return sys_base + gpio[pin].offset;
 
 	return NULL;
 }
@@ -179,7 +180,6 @@ EXPORT_SYMBOL(at91_set_multi_drive);
 
 /*--------------------------------------------------------------------------*/
 
-
 /*
  * assuming the pin is muxed as a gpio output, set its value.
  */
@@ -216,8 +216,8 @@ EXPORT_SYMBOL(at91_get_gpio_value);
 
 #ifdef CONFIG_PM
 
-static u32 wakeups[BGA_GPIO_BANKS];
-static u32 backups[BGA_GPIO_BANKS];
+static u32 wakeups[MAX_GPIO_BANKS];
+static u32 backups[MAX_GPIO_BANKS];
 
 static int gpio_irq_set_wake(unsigned pin, unsigned state)
 {
@@ -226,7 +226,7 @@ static int gpio_irq_set_wake(unsigned pin, unsigned state)
 	pin -= PIN_BASE;
 	pin /= 32;
 
-	if (unlikely(pin >= BGA_GPIO_BANKS))
+	if (unlikely(pin >= MAX_GPIO_BANKS))
 		return -EINVAL;
 
 	if (state)
@@ -241,8 +241,8 @@ void at91_gpio_suspend(void)
 {
 	int i;
 
-	for (i = 0; i < BGA_GPIO_BANKS; i++) {
-		u32 pio = pio_controller_offset[i];
+	for (i = 0; i < gpio_banks; i++) {
+		u32 pio = gpio[i].offset;
 
 		/*
 		 * Note: drivers should have disabled GPIO interrupts that
@@ -257,14 +257,14 @@ void at91_gpio_suspend(void)
 		 * first place!
 		 */
 		backups[i] = at91_sys_read(pio + PIO_IMR);
-		at91_sys_write(pio_controller_offset[i] + PIO_IDR, backups[i]);
-		at91_sys_write(pio_controller_offset[i] + PIO_IER, wakeups[i]);
+		at91_sys_write(pio + PIO_IDR, backups[i]);
+		at91_sys_write(pio + PIO_IER, wakeups[i]);
 
 		if (!wakeups[i]) {
-			disable_irq_wake(AT91RM9200_ID_PIOA + i);
-			at91_sys_write(AT91_PMC_PCDR, 1 << (AT91RM9200_ID_PIOA + i));
+			disable_irq_wake(gpio[i].id);
+			at91_sys_write(AT91_PMC_PCDR, 1 << gpio[i].id);
 		} else {
-			enable_irq_wake(AT91RM9200_ID_PIOA + i);
+			enable_irq_wake(gpio[i].id);
 #ifdef CONFIG_PM_DEBUG
 			printk(KERN_DEBUG "GPIO-%c may wake for %08x\n", "ABCD"[i], wakeups[i]);
 #endif
@@ -276,16 +276,13 @@ void at91_gpio_resume(void)
 {
 	int i;
 
-	for (i = 0; i < BGA_GPIO_BANKS; i++) {
-		at91_sys_write(pio_controller_offset[i] + PIO_IDR, wakeups[i]);
-		at91_sys_write(pio_controller_offset[i] + PIO_IER, backups[i]);
-	}
+	for (i = 0; i < gpio_banks; i++) {
+		u32 pio = gpio[i].offset;
 
-	at91_sys_write(AT91_PMC_PCER,
-			  (1 << AT91RM9200_ID_PIOA)
-			| (1 << AT91RM9200_ID_PIOB)
-			| (1 << AT91RM9200_ID_PIOC)
-			| (1 << AT91RM9200_ID_PIOD));
+		at91_sys_write(pio + PIO_IDR, wakeups[i]);
+		at91_sys_write(pio + PIO_IER, backups[i]);
+		at91_sys_write(AT91_PMC_PCER, 1 << gpio[i].id);
+	}
 }
 
 #else
@@ -377,20 +374,25 @@ static void gpio_irq_handler(unsigned irq, struct irqdesc *desc, struct pt_regs
 	/* now it may re-trigger */
 }
 
-/* call this from board-specific init_irq */
-void __init at91_gpio_irq_setup(unsigned banks)
+/*--------------------------------------------------------------------------*/
+
+/*
+ * Called from the processor-specific init to enable GPIO interrupt support.
+ */
+void __init at91_gpio_irq_setup(void)
 {
-	unsigned	pioc, pin, id;
+	unsigned	pioc, pin;
 
-	if (banks > 4)
-		banks = 4;
-	for (pioc = 0, pin = PIN_BASE, id = AT91RM9200_ID_PIOA;
-			pioc < banks;
-			pioc++, id++) {
+	for (pioc = 0, pin = PIN_BASE;
+			pioc < gpio_banks;
+			pioc++) {
 		void __iomem	*controller;
+		unsigned	id = gpio[pioc].id;
 		unsigned	i;
 
-		controller = (void __iomem *) AT91_VA_BASE_SYS + pio_controller_offset[pioc];
+		clk_enable(gpio[pioc].clock);	/* enable PIO controller's clock */
+
+		controller = (void __iomem *) AT91_VA_BASE_SYS + gpio[pioc].offset;
 		__raw_writel(~0, controller + PIO_IDR);
 
 		set_irq_data(id, (void *) pin);
@@ -408,5 +410,16 @@ void __init at91_gpio_irq_setup(unsigned banks)
 
 		set_irq_chained_handler(id, gpio_irq_handler);
 	}
-	pr_info("AT91: %d gpio irqs in %d banks\n", pin - PIN_BASE, banks);
+	pr_info("AT91: %d gpio irqs in %d banks\n", pin - PIN_BASE, gpio_banks);
+}
+
+/*
+ * Called from the processor-specific init to enable GPIO pin support.
+ */
+void __init at91_gpio_init(struct at91_gpio_bank *data, int nr_banks)
+{
+	BUG_ON(nr_banks > MAX_GPIO_BANKS);
+
+	gpio = data;
+	gpio_banks = nr_banks;
 }
diff --git a/arch/arm/mach-at91rm9200/irq.c b/arch/arm/mach-at91rm9200/irq.c
index 8cd60a99c5b8..3e488117ca91 100644
--- a/arch/arm/mach-at91rm9200/irq.c
+++ b/arch/arm/mach-at91rm9200/irq.c
@@ -34,8 +34,6 @@
 #include <asm/mach/irq.h>
 #include <asm/mach/map.h>
 
-#include "generic.h"
-
 
 static void at91_aic_mask_irq(unsigned int irq)
 {
diff --git a/include/asm-arm/arch-at91rm9200/gpio.h b/include/asm-arm/arch-at91rm9200/gpio.h
index 6243f28a0b81..a011d27876a2 100644
--- a/include/asm-arm/arch-at91rm9200/gpio.h
+++ b/include/asm-arm/arch-at91rm9200/gpio.h
@@ -17,8 +17,7 @@
 
 #define PIN_BASE		NR_AIC_IRQS
 
-#define PQFP_GPIO_BANKS		3	/* PQFP package has 3 banks */
-#define BGA_GPIO_BANKS		4	/* BGA package has 4 banks */
+#define MAX_GPIO_BANKS		4
 
 /* these pin numbers double as IRQ numbers, like AT91xxx_ID_* values */
 
@@ -180,17 +179,18 @@
 
 #ifndef __ASSEMBLY__
 /* setup setup routines, called from board init or driver probe() */
-extern int at91_set_A_periph(unsigned pin, int use_pullup);
-extern int at91_set_B_periph(unsigned pin, int use_pullup);
-extern int at91_set_gpio_input(unsigned pin, int use_pullup);
-extern int at91_set_gpio_output(unsigned pin, int value);
-extern int at91_set_deglitch(unsigned pin, int is_on);
-extern int at91_set_multi_drive(unsigned pin, int is_on);
+extern int __init_or_module at91_set_A_periph(unsigned pin, int use_pullup);
+extern int __init_or_module at91_set_B_periph(unsigned pin, int use_pullup);
+extern int __init_or_module at91_set_gpio_input(unsigned pin, int use_pullup);
+extern int __init_or_module at91_set_gpio_output(unsigned pin, int value);
+extern int __init_or_module at91_set_deglitch(unsigned pin, int is_on);
+extern int __init_or_module at91_set_multi_drive(unsigned pin, int is_on);
 
 /* callable at any time */
 extern int at91_set_gpio_value(unsigned pin, int value);
 extern int at91_get_gpio_value(unsigned pin);
 
+/* callable only from core power-management code */
 extern void at91_gpio_suspend(void);
 extern void at91_gpio_resume(void);
 #endif
-- 
cgit v1.2.3


From 97f0fb68f142b477773c05140da27c1dbd31a2ab Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Wed, 27 Sep 2006 16:18:18 +0100
Subject: [ARM] 3868/1: AT91 hardware header update

This patch adds the hardware register definitions for the TWI (I2C)
controller found on the AT91RM9200 and AT91SAM9xx processors.
It also defines the AIC Fast-Forcing registers added to the AT91SAM9's.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-at91rm9200/at91rm9200_sys.h |  3 ++
 include/asm-arm/arch-at91rm9200/at91rm9200_twi.h | 57 ++++++++++++++++++++++++
 2 files changed, 60 insertions(+)
 create mode 100644 include/asm-arm/arch-at91rm9200/at91rm9200_twi.h

(limited to 'include')

diff --git a/include/asm-arm/arch-at91rm9200/at91rm9200_sys.h b/include/asm-arm/arch-at91rm9200/at91rm9200_sys.h
index 0f4c12d5f0cd..73693fea76a2 100644
--- a/include/asm-arm/arch-at91rm9200/at91rm9200_sys.h
+++ b/include/asm-arm/arch-at91rm9200/at91rm9200_sys.h
@@ -80,6 +80,9 @@
 #define		AT91_CIDR_NVPTYP	(7    << 28)		/* Nonvolatile Program Memory Type */
 #define		AT91_CIDR_EXT		(1    << 31)		/* Extension Flag */
 
+#define AT91_AIC_FFER		(AT91_AIC + 0x140)	/* Fast Forcing Enable Register [SAM9 only] */
+#define AT91_AIC_FFDR		(AT91_AIC + 0x144)	/* Fast Forcing Disable Register [SAM9 only] */
+#define AT91_AIC_FFSR		(AT91_AIC + 0x148)	/* Fast Forcing Status Register [SAM9 only] */
 
 /*
  * PIO Controllers.
diff --git a/include/asm-arm/arch-at91rm9200/at91rm9200_twi.h b/include/asm-arm/arch-at91rm9200/at91rm9200_twi.h
new file mode 100644
index 000000000000..93547d7482bd
--- /dev/null
+++ b/include/asm-arm/arch-at91rm9200/at91rm9200_twi.h
@@ -0,0 +1,57 @@
+/*
+ * include/asm-arm/arch-at91rm9200/at91rm9200_twi.h
+ *
+ * Copyright (C) 2005 Ivan Kokshaysky
+ * Copyright (C) SAN People
+ *
+ * Two-wire Interface (TWI) registers.
+ * Based on AT91RM9200 datasheet revision E.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91RM9200_TWI_H
+#define AT91RM9200_TWI_H
+
+#define	AT91_TWI_CR		0x00		/* Control Register */
+#define		AT91_TWI_START		(1 <<  0)	/* Send a Start Condition */
+#define		AT91_TWI_STOP		(1 <<  1)	/* Send a Stop Condition */
+#define		AT91_TWI_MSEN		(1 <<  2)	/* Master Transfer Enable */
+#define		AT91_TWI_MSDIS		(1 <<  3)	/* Master Transfer Disable */
+#define		AT91_TWI_SWRST		(1 <<  7)	/* Software Reset */
+
+#define	AT91_TWI_MMR		0x04		/* Master Mode Register */
+#define		AT91_TWI_IADRSZ		(3    <<  8)	/* Internal Device Address Size */
+#define			AT91_TWI_IADRSZ_NO		(0 << 8)
+#define			AT91_TWI_IADRSZ_1		(1 << 8)
+#define			AT91_TWI_IADRSZ_2		(2 << 8)
+#define			AT91_TWI_IADRSZ_3		(3 << 8)
+#define		AT91_TWI_MREAD		(1    << 12)	/* Master Read Direction */
+#define		AT91_TWI_DADR		(0x7f << 16)	/* Device Address */
+
+#define	AT91_TWI_IADR		0x0c		/* Internal Address Register */
+
+#define	AT91_TWI_CWGR		0x10		/* Clock Waveform Generator Register */
+#define		AT91_TWI_CLDIV		(0xff <<  0)	/* Clock Low Divisor */
+#define		AT91_TWI_CHDIV		(0xff <<  8)	/* Clock High Divisor */
+#define		AT91_TWI_CKDIV		(7    << 16)	/* Clock Divider */
+
+#define	AT91_TWI_SR		0x20		/* Status Register */
+#define		AT91_TWI_TXCOMP		(1 <<  0)	/* Transmission Complete */
+#define		AT91_TWI_RXRDY		(1 <<  1)	/* Receive Holding Register Ready */
+#define		AT91_TWI_TXRDY		(1 <<  2)	/* Transmit Holding Register Ready */
+#define		AT91_TWI_OVRE		(1 <<  6)	/* Overrun Error */
+#define		AT91_TWI_UNRE		(1 <<  7)	/* Underrun Error */
+#define		AT91_TWI_NACK		(1 <<  8)	/* Not Acknowledged */
+
+#define	AT91_TWI_IER		0x24		/* Interrupt Enable Register */
+#define	AT91_TWI_IDR		0x28		/* Interrupt Disable Register */
+#define	AT91_TWI_IMR		0x2c		/* Interrupt Mask Register */
+#define	AT91_TWI_RHR		0x30		/* Receive Holding Register */
+#define	AT91_TWI_THR		0x34		/* Transmit Holding Register */
+
+#endif
+
-- 
cgit v1.2.3


From d9f7a745d55527d0d41684b22506a86c4381f7f1 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Thu, 28 Sep 2006 16:55:39 +0200
Subject: [S390] __div64_32 for 31 bit.

The clocksource infrastructure introduced with commit
ad596171ed635c51a9eef829187af100cbf8dcf7 broke 31 bit s390.
The reason is that the do_div() primitive for 31 bit always
had a restriction: it could only divide an unsigned 64 bit
integer by an unsigned 31 bit integer. The clocksource code
now uses do_div() with a base value that has the most
significant bit set. The result is that clock->cycle_interval
has a funny value which causes the linux time to jump around
like mad.
The solution is "obvious": implement a proper __div64_32
function for 31 bit s390.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig        |   4 ++
 arch/s390/lib/Makefile   |   1 +
 arch/s390/lib/div64.c    | 151 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-s390/div64.h |  48 ---------------
 4 files changed, 156 insertions(+), 48 deletions(-)
 create mode 100644 arch/s390/lib/div64.c

(limited to 'include')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index b216ca659cdf..b6b42f9f0d52 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -51,6 +51,10 @@ config 64BIT
 	  Select this option if you have a 64 bit IBM zSeries machine
 	  and want to use the 64 bit addressing mode.
 
+config 32BIT
+	bool
+	default y if !64BIT
+
 config SMP
 	bool "Symmetric multi-processing support"
 	---help---
diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile
index c42ffedfdb49..b0cfa6c4883d 100644
--- a/arch/s390/lib/Makefile
+++ b/arch/s390/lib/Makefile
@@ -5,5 +5,6 @@
 EXTRA_AFLAGS := -traditional
 
 lib-y += delay.o string.o uaccess_std.o
+lib-$(CONFIG_32BIT) += div64.o
 lib-$(CONFIG_64BIT) += uaccess_mvcos.o
 lib-$(CONFIG_SMP) += spinlock.o
diff --git a/arch/s390/lib/div64.c b/arch/s390/lib/div64.c
new file mode 100644
index 000000000000..0481f3424a13
--- /dev/null
+++ b/arch/s390/lib/div64.c
@@ -0,0 +1,151 @@
+/*
+ *  arch/s390/lib/div64.c
+ *
+ *  __div64_32 implementation for 31 bit.
+ *
+ *    Copyright (C) IBM Corp. 2006
+ *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
+ */
+
+#include <linux/types.h>
+#include <linux/module.h>
+
+#ifdef CONFIG_MARCH_G5
+
+/*
+ * Function to divide an unsigned 64 bit integer by an unsigned
+ * 31 bit integer using signed 64/32 bit division.
+ */
+static uint32_t __div64_31(uint64_t *n, uint32_t base)
+{
+	register uint32_t reg2 asm("2");
+	register uint32_t reg3 asm("3");
+	uint32_t *words = (uint32_t *) n;
+	uint32_t tmp;
+
+	/* Special case base==1, remainder = 0, quotient = n */
+	if (base == 1)
+		return 0;
+	/*
+	 * Special case base==0 will cause a fixed point divide exception
+	 * on the dr instruction and may not happen anyway. For the
+	 * following calculation we can assume base > 1. The first
+	 * signed 64 / 32 bit division with an upper half of 0 will
+	 * give the correct upper half of the 64 bit quotient.
+	 */
+	reg2 = 0UL;
+	reg3 = words[0];
+	asm volatile(
+		"	dr	%0,%2\n"
+		: "+d" (reg2), "+d" (reg3) : "d" (base) : "cc" );
+	words[0] = reg3;
+	reg3 = words[1];
+	/*
+	 * To get the lower half of the 64 bit quotient and the 32 bit
+	 * remainder we have to use a little trick. Since we only have
+	 * a signed division the quotient can get too big. To avoid this
+	 * the 64 bit dividend is halved, then the signed division will
+	 * work. Afterwards the quotient and the remainder are doubled.
+	 * If the last bit of the dividend has been one the remainder
+	 * is increased by one then checked against the base. If the
+	 * remainder has overflown subtract base and increase the
+	 * quotient. Simple, no ?
+	 */
+	asm volatile(
+		"	nr	%2,%1\n"
+		"	srdl	%0,1\n"
+		"	dr	%0,%3\n"
+		"	alr	%0,%0\n"
+		"	alr	%1,%1\n"
+		"	alr	%0,%2\n"
+		"	clr	%0,%3\n"
+		"	jl	0f\n"
+		"	slr	%0,%3\n"
+		"	alr	%1,%2\n"
+		"0:\n"
+		: "+d" (reg2), "+d" (reg3), "=d" (tmp)
+		: "d" (base), "2" (1UL) : "cc" );
+	words[1] = reg3;
+	return reg2;
+}
+
+/*
+ * Function to divide an unsigned 64 bit integer by an unsigned
+ * 32 bit integer using the unsigned 64/31 bit division.
+ */
+uint32_t __div64_32(uint64_t *n, uint32_t base)
+{
+	uint32_t r;
+
+	/*
+	 * If the most significant bit of base is set, divide n by
+	 * (base/2). That allows to use 64/31 bit division and gives a
+	 * good approximation of the result: n = (base/2)*q + r. The
+	 * result needs to be corrected with two simple transformations.
+	 * If base is already < 2^31-1 __div64_31 can be used directly.
+	 */
+	r = __div64_31(n, ((signed) base < 0) ? (base/2) : base);
+	if ((signed) base < 0) {
+		uint64_t q = *n;
+		/*
+		 * First transformation:
+		 * n = (base/2)*q + r
+		 *   = ((base/2)*2)*(q/2) + ((q&1) ? (base/2) : 0) + r
+		 * Since r < (base/2), r + (base/2) < base.
+		 * With q1 = (q/2) and r1 = r + ((q&1) ? (base/2) : 0)
+		 * n = ((base/2)*2)*q1 + r1 with r1 < base.
+		 */
+		if (q & 1)
+			r += base/2;
+		q >>= 1;
+		/*
+		 * Second transformation. ((base/2)*2) could have lost the
+		 * last bit.
+		 * n = ((base/2)*2)*q1 + r1
+		 *   = base*q1 - ((base&1) ? q1 : 0) + r1
+		 */
+		if (base & 1) {
+			int64_t rx = r - q;
+			/*
+			 * base is >= 2^31. The worst case for the while
+			 * loop is n=2^64-1 base=2^31+1. That gives a
+			 * maximum for q=(2^64-1)/2^31 = 0x1ffffffff. Since
+			 * base >= 2^31 the loop is finished after a maximum
+			 * of three iterations.
+			 */
+			while (rx < 0) {
+				rx += base;
+				q--;
+			}
+			r = rx;
+		}
+		*n = q;
+	}
+	return r;
+}
+
+#else /* MARCH_G5 */
+
+uint32_t __div64_32(uint64_t *n, uint32_t base)
+{
+	register uint32_t reg2 asm("2");
+	register uint32_t reg3 asm("3");
+	uint32_t *words = (uint32_t *) n;
+
+	reg2 = 0UL;
+	reg3 = words[0];
+	asm volatile(
+		"	dlr	%0,%2\n"
+		: "+d" (reg2), "+d" (reg3) : "d" (base) : "cc" );
+	words[0] = reg3;
+	reg3 = words[1];
+	asm volatile(
+		"	dlr	%0,%2\n"
+		: "+d" (reg2), "+d" (reg3) : "d" (base) : "cc" );
+	words[1] = reg3;
+	return reg2;
+}
+
+#endif /* MARCH_G5 */
+
+EXPORT_SYMBOL(__div64_32);
diff --git a/include/asm-s390/div64.h b/include/asm-s390/div64.h
index af098dc3cf59..6cd978cefb28 100644
--- a/include/asm-s390/div64.h
+++ b/include/asm-s390/div64.h
@@ -1,49 +1 @@
-#ifndef __S390_DIV64
-#define __S390_DIV64
-
-#ifndef __s390x__
-
-/* for do_div "base" needs to be smaller than 2^31-1 */
-#define do_div(n, base) ({                                      \
-	unsigned long long __n = (n);				\
-	unsigned long __r;					\
-								\
-	asm ("   slr  0,0\n"					\
-	     "   l    1,%1\n"					\
-	     "   srdl 0,1\n"					\
-	     "   dr   0,%2\n"					\
-	     "   alr  1,1\n"					\
-	     "   alr  0,0\n"					\
-	     "   lhi  2,1\n"					\
-	     "   n    2,%1\n"					\
-	     "   alr  0,2\n"					\
-	     "   clr  0,%2\n"					\
-	     "   jl   0f\n"					\
-	     "   slr  0,%2\n"					\
-             "   ahi  1,1\n"					\
-	     "0: st   1,%1\n"					\
-	     "   l    1,4+%1\n"					\
-	     "   srdl 0,1\n"					\
-             "   dr   0,%2\n"					\
-	     "   alr  1,1\n"					\
-	     "   alr  0,0\n"					\
-	     "   lhi  2,1\n"					\
-	     "   n    2,4+%1\n"					\
-	     "   alr  0,2\n"					\
-	     "   clr  0,%2\n"					\
-             "   jl   1f\n"					\
-	     "   slr  0,%2\n"					\
-	     "   ahi  1,1\n"					\
-	     "1: st   1,4+%1\n"					\
-             "   lr   %0,0"					\
-	     : "=d" (__r), "=m" (__n)				\
-	     : "d" (base), "m" (__n) : "0", "1", "2", "cc" );	\
-	(n) = (__n);						\
-        __r;                                                    \
-})
-
-#else /* __s390x__ */
 #include <asm-generic/div64.h>
-#endif /* __s390x__ */
-
-#endif
-- 
cgit v1.2.3


From 94c12cc7d196bab34aaa98d38521549fa1e5ef76 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Thu, 28 Sep 2006 16:56:43 +0200
Subject: [S390] Inline assembly cleanup.

Major cleanup of all s390 inline assemblies. They now have a common
coding style. Quite a few have been shortened, mainly by using register
asm variables. Use of the EX_TABLE macro helps  as well. The atomic ops,
bit ops and locking inlines new use the Q-constraint if a newer gcc
is used.  That results in slightly better code.

Thanks to Christian Borntraeger for proof reading the changes.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/crypto/crypt_s390.h   | 204 +++++--------
 arch/s390/hypfs/hypfs_diag.c    |  23 +-
 arch/s390/kernel/compat_linux.c |   5 +-
 arch/s390/kernel/cpcmd.c        |  83 ++----
 arch/s390/kernel/ipl.c          |  21 +-
 arch/s390/kernel/process.c      |   5 +-
 arch/s390/kernel/semaphore.c    |  22 +-
 arch/s390/kernel/setup.c        |   2 +-
 arch/s390/kernel/smp.c          |  73 ++---
 arch/s390/kernel/time.c         |  10 +-
 arch/s390/kernel/traps.c        |   3 +-
 arch/s390/lib/delay.c           |  11 +-
 arch/s390/math-emu/math.c       | 126 ++++----
 arch/s390/math-emu/sfp-util.h   |  73 ++---
 arch/s390/mm/extmem.c           |  16 +-
 arch/s390/mm/fault.c            |  34 +--
 arch/s390/mm/init.c             |  41 +--
 drivers/s390/block/dasd_diag.c  |  34 +--
 drivers/s390/block/xpram.c      |  54 +---
 drivers/s390/char/sclp.c        |  31 +-
 drivers/s390/char/vmwatchdog.c  |  52 +---
 drivers/s390/cio/device_id.c    |  38 +--
 drivers/s390/cio/ioasm.h        | 220 ++++++--------
 drivers/s390/cio/qdio.h         | 192 ++++--------
 drivers/s390/net/iucv.c         |  39 ++-
 drivers/s390/s390mach.c         |  93 +++---
 include/asm-s390/appldata.h     |   2 +-
 include/asm-s390/atomic.h       | 120 ++++++--
 include/asm-s390/bitops.h       | 626 +++++++++++++++++++---------------------
 include/asm-s390/byteorder.h    |  50 ++--
 include/asm-s390/checksum.h     | 176 ++++-------
 include/asm-s390/ebcdic.h       |  20 +-
 include/asm-s390/io.h           |  14 +-
 include/asm-s390/irqflags.h     | 110 +++++--
 include/asm-s390/lowcore.h      |   2 +-
 include/asm-s390/page.h         | 111 +++----
 include/asm-s390/pgtable.h      |  28 +-
 include/asm-s390/processor.h    | 130 ++++-----
 include/asm-s390/ptrace.h       |   2 +-
 include/asm-s390/rwsem.h        | 238 +++++++--------
 include/asm-s390/semaphore.h    |  16 +-
 include/asm-s390/sfp-machine.h  |  64 ++--
 include/asm-s390/sigp.h         |  65 ++---
 include/asm-s390/smp.h          |   2 +-
 include/asm-s390/spinlock.h     |  27 +-
 include/asm-s390/string.h       |  56 ++--
 include/asm-s390/system.h       | 342 +++++++++-------------
 include/asm-s390/timex.h        |  19 +-
 include/asm-s390/tlbflush.h     |  32 +-
 include/asm-s390/uaccess.h      |  13 +-
 include/asm-s390/unistd.h       | 258 ++++++++---------
 51 files changed, 1759 insertions(+), 2269 deletions(-)

(limited to 'include')

diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h
index efd836c2e4a6..2b137089f625 100644
--- a/arch/s390/crypto/crypt_s390.h
+++ b/arch/s390/crypto/crypt_s390.h
@@ -104,63 +104,6 @@ struct crypt_s390_query_status {
 	u64 low;
 };
 
-/*
- * Standard fixup and ex_table sections for crypt_s390 inline functions.
- * label 0: the s390 crypto operation
- * label 1: just after 1 to catch illegal operation exception
- *          (unsupported model)
- * label 6: the return point after fixup
- * label 7: set error value if exception _in_ crypto operation
- * label 8: set error value if illegal operation exception
- * [ret] is the variable to receive the error code
- * [ERR] is the error code value
- */
-#ifndef CONFIG_64BIT
-#define __crypt_s390_fixup \
-	".section .fixup,\"ax\" \n"	\
-	"7:	lhi	%0,%h[e1] \n"	\
-	"	bras	1,9f \n"	\
-	"	.long	6b \n"		\
-	"8:	lhi	%0,%h[e2] \n"	\
-	"	bras	1,9f \n"	\
-	"	.long	6b \n"		\
-	"9:	l	1,0(1) \n"	\
-	"	br	1 \n"		\
-	".previous \n"			\
-	".section __ex_table,\"a\" \n"	\
-	"	.align	4 \n"		\
-	"	.long	0b,7b \n"	\
-	"	.long	1b,8b \n"	\
-	".previous"
-#else /* CONFIG_64BIT */
-#define __crypt_s390_fixup \
-	".section .fixup,\"ax\" \n"	\
-	"7:	lhi	%0,%h[e1] \n"	\
-	"	jg	6b \n"		\
-	"8:	lhi	%0,%h[e2] \n"	\
-	"	jg	6b \n"		\
-	".previous\n"			\
-	".section __ex_table,\"a\" \n"	\
-	"	.align	8 \n"		\
-	"	.quad	0b,7b \n"	\
-	"	.quad	1b,8b \n"	\
-	".previous"
-#endif /* CONFIG_64BIT */
-
-/*
- * Standard code for setting the result of s390 crypto instructions.
- * %0: the register which will receive the result
- * [result]: the register containing the result (e.g. second operand length
- * to compute number of processed bytes].
- */
-#ifndef CONFIG_64BIT
-#define __crypt_s390_set_result \
-	"	lr	%0,%[result] \n"
-#else /* CONFIG_64BIT */
-#define __crypt_s390_set_result \
-	"	lgr	%0,%[result] \n"
-#endif
-
 /*
  * Executes the KM (CIPHER MESSAGE) operation of the CPU.
  * @param func: the function code passed to KM; see crypt_s390_km_func
@@ -176,28 +119,24 @@ crypt_s390_km(long func, void* param, u8* dest, const u8* src, long src_len)
 {
 	register long __func asm("0") = func & CRYPT_S390_FUNC_MASK;
 	register void* __param asm("1") = param;
-	register u8* __dest asm("4") = dest;
 	register const u8* __src asm("2") = src;
 	register long __src_len asm("3") = src_len;
+	register u8* __dest asm("4") = dest;
 	int ret;
 
-	ret = 0;
-	__asm__ __volatile__ (
-		"0:	.insn	rre,0xB92E0000,%1,%2 \n" /* KM opcode */
+	asm volatile(
+		"0:	.insn	rre,0xb92e0000,%3,%1 \n" /* KM opcode */
 		"1:	brc	1,0b \n" /* handle partial completion */
-		__crypt_s390_set_result
-		"6:	\n"
-		__crypt_s390_fixup
-		: "+d" (ret), "+a" (__dest), "+a" (__src),
-		  [result] "+d" (__src_len)
-		: [e1] "K" (-EFAULT), [e2] "K" (-ENOSYS), "d" (__func),
-		  "a" (__param)
-		: "cc", "memory"
-	);
-	if (ret >= 0 && func & CRYPT_S390_FUNC_MASK){
-		ret = src_len - ret;
-	}
-	return ret;
+		"	ahi	%0,%h7\n"
+		"2:	ahi	%0,%h8\n"
+		"3:\n"
+		EX_TABLE(0b,3b) EX_TABLE(1b,2b)
+		: "=d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest)
+		: "d" (__func), "a" (__param), "0" (-EFAULT),
+		  "K" (ENOSYS), "K" (-ENOSYS + EFAULT) : "cc", "memory");
+	if (ret < 0)
+		return ret;
+	return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len;
 }
 
 /*
@@ -215,28 +154,24 @@ crypt_s390_kmc(long func, void* param, u8* dest, const u8* src, long src_len)
 {
 	register long __func asm("0") = func & CRYPT_S390_FUNC_MASK;
 	register void* __param asm("1") = param;
-	register u8* __dest asm("4") = dest;
 	register const u8* __src asm("2") = src;
 	register long __src_len asm("3") = src_len;
+	register u8* __dest asm("4") = dest;
 	int ret;
 
-	ret = 0;
-	__asm__ __volatile__ (
-		"0:	.insn	rre,0xB92F0000,%1,%2 \n" /* KMC opcode */
+	asm volatile(
+		"0:	.insn	rre,0xb92f0000,%3,%1 \n" /* KMC opcode */
 		"1:	brc	1,0b \n" /* handle partial completion */
-		__crypt_s390_set_result
-		"6:	\n"
-		__crypt_s390_fixup
-		: "+d" (ret), "+a" (__dest), "+a" (__src),
-		  [result] "+d" (__src_len)
-		: [e1] "K" (-EFAULT), [e2] "K" (-ENOSYS), "d" (__func),
-		  "a" (__param)
-		: "cc", "memory"
-	);
-	if (ret >= 0 && func & CRYPT_S390_FUNC_MASK){
-		ret = src_len - ret;
-	}
-	return ret;
+		"	ahi	%0,%h7\n"
+		"2:	ahi	%0,%h8\n"
+		"3:\n"
+		EX_TABLE(0b,3b) EX_TABLE(1b,2b)
+		: "=d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest)
+		: "d" (__func), "a" (__param), "0" (-EFAULT),
+		  "K" (ENOSYS), "K" (-ENOSYS + EFAULT) : "cc", "memory");
+	if (ret < 0)
+		return ret;
+	return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len;
 }
 
 /*
@@ -258,22 +193,19 @@ crypt_s390_kimd(long func, void* param, const u8* src, long src_len)
 	register long __src_len asm("3") = src_len;
 	int ret;
 
-	ret = 0;
-	__asm__ __volatile__ (
-		"0:	.insn	rre,0xB93E0000,%1,%1 \n" /* KIMD opcode */
-		"1:	brc	1,0b \n" /* handle partical completion */
-		__crypt_s390_set_result
-		"6:	\n"
-		__crypt_s390_fixup
-		: "+d" (ret), "+a" (__src), [result] "+d" (__src_len)
-		: [e1] "K" (-EFAULT), [e2] "K" (-ENOSYS), "d" (__func),
-		  "a" (__param)
-		: "cc", "memory"
-	);
-	if (ret >= 0 && (func & CRYPT_S390_FUNC_MASK)){
-		ret = src_len - ret;
-	}
-	return ret;
+	asm volatile(
+		"0:	.insn	rre,0xb93e0000,%1,%1 \n" /* KIMD opcode */
+		"1:	brc	1,0b \n" /* handle partial completion */
+		"	ahi	%0,%h6\n"
+		"2:	ahi	%0,%h7\n"
+		"3:\n"
+		EX_TABLE(0b,3b) EX_TABLE(1b,2b)
+		: "=d" (ret), "+a" (__src), "+d" (__src_len)
+		: "d" (__func), "a" (__param), "0" (-EFAULT),
+		  "K" (ENOSYS), "K" (-ENOSYS + EFAULT) : "cc", "memory");
+	if (ret < 0)
+		return ret;
+	return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len;
 }
 
 /*
@@ -294,22 +226,19 @@ crypt_s390_klmd(long func, void* param, const u8* src, long src_len)
 	register long __src_len asm("3") = src_len;
 	int ret;
 
-	ret = 0;
-	__asm__ __volatile__ (
-		"0:	.insn	rre,0xB93F0000,%1,%1 \n" /* KLMD opcode */
-		"1:	brc	1,0b \n" /* handle partical completion */
-		__crypt_s390_set_result
-		"6:	\n"
-		__crypt_s390_fixup
-		: "+d" (ret), "+a" (__src), [result] "+d" (__src_len)
-		: [e1] "K" (-EFAULT), [e2] "K" (-ENOSYS), "d" (__func),
-		  "a" (__param)
-		: "cc", "memory"
-	);
-	if (ret >= 0 && func & CRYPT_S390_FUNC_MASK){
-		ret = src_len - ret;
-	}
-	return ret;
+	asm volatile(
+		"0:	.insn	rre,0xb93f0000,%1,%1 \n" /* KLMD opcode */
+		"1:	brc	1,0b \n" /* handle partial completion */
+		"	ahi	%0,%h6\n"
+		"2:	ahi	%0,%h7\n"
+		"3:\n"
+		EX_TABLE(0b,3b) EX_TABLE(1b,2b)
+		: "=d" (ret), "+a" (__src), "+d" (__src_len)
+		: "d" (__func), "a" (__param), "0" (-EFAULT),
+		  "K" (ENOSYS), "K" (-ENOSYS + EFAULT) : "cc", "memory");
+	if (ret < 0)
+		return ret;
+	return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len;
 }
 
 /*
@@ -331,22 +260,19 @@ crypt_s390_kmac(long func, void* param, const u8* src, long src_len)
 	register long __src_len asm("3") = src_len;
 	int ret;
 
-	ret = 0;
-	__asm__ __volatile__ (
-		"0:	.insn	rre,0xB91E0000,%5,%5 \n" /* KMAC opcode */
-		"1:	brc	1,0b \n" /* handle partical completion */
-		__crypt_s390_set_result
-		"6:	\n"
-		__crypt_s390_fixup
-		: "+d" (ret), "+a" (__src), [result] "+d" (__src_len)
-		: [e1] "K" (-EFAULT), [e2] "K" (-ENOSYS), "d" (__func),
-		  "a" (__param)
-		: "cc", "memory"
-	);
-	if (ret >= 0 && func & CRYPT_S390_FUNC_MASK){
-		ret = src_len - ret;
-	}
-	return ret;
+	asm volatile(
+		"0:	.insn	rre,0xb91e0000,%1,%1 \n" /* KLAC opcode */
+		"1:	brc	1,0b \n" /* handle partial completion */
+		"	ahi	%0,%h6\n"
+		"2:	ahi	%0,%h7\n"
+		"3:\n"
+		EX_TABLE(0b,3b) EX_TABLE(1b,2b)
+		: "=d" (ret), "+a" (__src), "+d" (__src_len)
+		: "d" (__func), "a" (__param), "0" (-EFAULT),
+		  "K" (ENOSYS), "K" (-ENOSYS + EFAULT) : "cc", "memory");
+	if (ret < 0)
+		return ret;
+	return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len;
 }
 
 /**
diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c
index 684384f2b364..443fa377d9ff 100644
--- a/arch/s390/hypfs/hypfs_diag.c
+++ b/arch/s390/hypfs/hypfs_diag.c
@@ -333,22 +333,14 @@ static int diag204(unsigned long subcode, unsigned long size, void *addr)
 	register unsigned long _subcode asm("0") = subcode;
 	register unsigned long _size asm("1") = size;
 
-	asm volatile ("   diag    %2,%0,0x204\n"
-		      "0: \n" ".section __ex_table,\"a\"\n"
-#ifndef __s390x__
-		      "    .align 4\n"
-		      "    .long  0b,0b\n"
-#else
-		      "    .align 8\n"
-		      "    .quad  0b,0b\n"
-#endif
-		      ".previous":"+d" (_subcode), "+d"(_size)
-		      :"d"(addr)
-		      :"memory");
+	asm volatile(
+		"	diag	%2,%0,0x204\n"
+		"0:\n"
+		EX_TABLE(0b,0b)
+		: "+d" (_subcode), "+d" (_size) : "d" (addr) : "memory");
 	if (_subcode)
 		return -1;
-	else
-		return _size;
+	return _size;
 }
 
 /*
@@ -491,8 +483,7 @@ out:
 
 static void diag224(void *ptr)
 {
-	asm volatile("   diag    %0,%1,0x224\n"
-		     : :"d" (0), "d"(ptr) : "memory");
+	asm volatile("diag %0,%1,0x224" : :"d" (0), "d"(ptr) : "memory");
 }
 
 static int diag224_get_name_table(void)
diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c
index 91b2884fa5c4..c46e3d48e410 100644
--- a/arch/s390/kernel/compat_linux.c
+++ b/arch/s390/kernel/compat_linux.c
@@ -544,10 +544,7 @@ sys32_execve(struct pt_regs regs)
 		current->ptrace &= ~PT_DTRACE;
 		task_unlock(current);
 		current->thread.fp_regs.fpc=0;
-		__asm__ __volatile__
-		        ("sr  0,0\n\t"
-		         "sfpc 0,0\n\t"
-			 : : :"0");
+		asm volatile("sfpc %0,0" : : "d" (0));
 	}
         putname(filename);
 out:
diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c
index 4ef44e536b2c..1eae74e72f95 100644
--- a/arch/s390/kernel/cpcmd.c
+++ b/arch/s390/kernel/cpcmd.c
@@ -25,11 +25,8 @@ static char cpcmd_buf[241];
  */
 int  __cpcmd(const char *cmd, char *response, int rlen, int *response_code)
 {
-	const int mask = 0x40000000L;
-	unsigned long flags;
-	int return_code;
-	int return_len;
-	int cmdlen;
+	unsigned long flags, cmdlen;
+	int return_code, return_len;
 
 	spin_lock_irqsave(&cpcmd_lock, flags);
 	cmdlen = strlen(cmd);
@@ -38,64 +35,44 @@ int  __cpcmd(const char *cmd, char *response, int rlen, int *response_code)
 	ASCEBC(cpcmd_buf, cmdlen);
 
 	if (response != NULL && rlen > 0) {
+		register unsigned long reg2 asm ("2") = (addr_t) cpcmd_buf;
+		register unsigned long reg3 asm ("3") = (addr_t) response;
+		register unsigned long reg4 asm ("4") = cmdlen | 0x40000000L;
+		register unsigned long reg5 asm ("5") = rlen;
+
 		memset(response, 0, rlen);
+		asm volatile(
 #ifndef CONFIG_64BIT
-		asm volatile (	"lra	2,0(%2)\n"
-				"lr	4,%3\n"
-				"o	4,%6\n"
-				"lra	3,0(%4)\n"
-				"lr	5,%5\n"
-				"diag	2,4,0x8\n"
-				"brc	8, 1f\n"
-				"ar	5, %5\n"
-				"1: \n"
-				"lr	%0,4\n"
-				"lr	%1,5\n"
-				: "=d" (return_code), "=d" (return_len)
-				: "a" (cpcmd_buf), "d" (cmdlen),
-				"a" (response), "d" (rlen), "m" (mask)
-				: "cc", "2", "3", "4", "5" );
+			"	diag	%2,%0,0x8\n"
+			"	brc	8,1f\n"
+			"	ar	%1,%4\n"
 #else /* CONFIG_64BIT */
-                asm volatile (	"lrag	2,0(%2)\n"
-				"lgr	4,%3\n"
-				"o	4,%6\n"
-				"lrag	3,0(%4)\n"
-				"lgr	5,%5\n"
-				"sam31\n"
-				"diag	2,4,0x8\n"
-				"sam64\n"
-				"brc	8, 1f\n"
-				"agr	5, %5\n"
-				"1: \n"
-				"lgr	%0,4\n"
-				"lgr	%1,5\n"
-				: "=d" (return_code), "=d" (return_len)
-				: "a" (cpcmd_buf), "d" (cmdlen),
-				"a" (response), "d" (rlen), "m" (mask)
-				: "cc", "2", "3", "4", "5" );
+			"	sam31\n"
+			"	diag	%2,%0,0x8\n"
+			"	sam64\n"
+			"	brc	8,1f\n"
+			"	agr	%1,%4\n"
 #endif /* CONFIG_64BIT */
+			"1:\n"
+			: "+d" (reg4), "+d" (reg5)
+			: "d" (reg2), "d" (reg3), "d" (rlen) : "cc");
+		return_code = (int) reg4;
+		return_len = (int) reg5;
                 EBCASC(response, rlen);
         } else {
+		register unsigned long reg2 asm ("2") = (addr_t) cpcmd_buf;
+		register unsigned long reg3 asm ("3") = cmdlen;
 		return_len = 0;
+		asm volatile(
 #ifndef CONFIG_64BIT
-                asm volatile (	"lra	2,0(%1)\n"
-				"lr	3,%2\n"
-				"diag	2,3,0x8\n"
-				"lr	%0,3\n"
-				: "=d" (return_code)
-				: "a" (cpcmd_buf), "d" (cmdlen)
-				: "2", "3"  );
+			"	diag	%1,%0,0x8\n"
 #else /* CONFIG_64BIT */
-                asm volatile (	"lrag	2,0(%1)\n"
-				"lgr	3,%2\n"
-				"sam31\n"
-				"diag	2,3,0x8\n"
-				"sam64\n"
-				"lgr	%0,3\n"
-				: "=d" (return_code)
-				: "a" (cpcmd_buf), "d" (cmdlen)
-				: "2", "3" );
+			"	sam31\n"
+			"	diag	%1,%0,0x8\n"
+			"	sam64\n"
 #endif /* CONFIG_64BIT */
+			: "+d" (reg3) : "d" (reg2) : "cc");
+		return_code = (int) reg3;
         }
 	spin_unlock_irqrestore(&cpcmd_lock, flags);
 	if (response_code != NULL)
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 6555cc48e28f..1f5e782b3d05 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -120,24 +120,15 @@ static enum shutdown_action on_panic_action = SHUTDOWN_STOP;
 
 static int diag308(unsigned long subcode, void *addr)
 {
-	register unsigned long _addr asm("0") = (unsigned long)addr;
+	register unsigned long _addr asm("0") = (unsigned long) addr;
 	register unsigned long _rc asm("1") = 0;
 
-	asm volatile (
-		"   diag %0,%2,0x308\n"
-		"0: \n"
-		".section __ex_table,\"a\"\n"
-#ifdef CONFIG_64BIT
-		"   .align 8\n"
-		"   .quad 0b, 0b\n"
-#else
-		"   .align 4\n"
-		"   .long 0b, 0b\n"
-#endif
-		".previous\n"
+	asm volatile(
+		"	diag	%0,%2,0x308\n"
+		"0:\n"
+		EX_TABLE(0b,0b)
 		: "+d" (_addr), "+d" (_rc)
-		: "d" (subcode) : "cc", "memory" );
-
+		: "d" (subcode) : "cc", "memory");
 	return _rc;
 }
 
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index d3cbfa3005ec..6603fbb41d07 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -45,7 +45,7 @@
 #include <asm/irq.h>
 #include <asm/timer.h>
 
-asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
+asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
 
 /*
  * Return saved PC of a blocked thread. used in kernel/sched.
@@ -177,7 +177,8 @@ void show_regs(struct pt_regs *regs)
 
 extern void kernel_thread_starter(void);
 
-__asm__(".align 4\n"
+asm(
+	".align 4\n"
 	"kernel_thread_starter:\n"
 	"    la    2,0(10)\n"
 	"    basr  14,9\n"
diff --git a/arch/s390/kernel/semaphore.c b/arch/s390/kernel/semaphore.c
index 8dfb690c159f..191303f6c1d8 100644
--- a/arch/s390/kernel/semaphore.c
+++ b/arch/s390/kernel/semaphore.c
@@ -26,17 +26,17 @@ static inline int __sem_update_count(struct semaphore *sem, int incr)
 {
 	int old_val, new_val;
 
-        __asm__ __volatile__("   l     %0,0(%3)\n"
-                             "0: ltr   %1,%0\n"
-			     "   jhe   1f\n"
-			     "   lhi   %1,0\n"
-			     "1: ar    %1,%4\n"
-                             "   cs    %0,%1,0(%3)\n"
-                             "   jl    0b\n"
-                             : "=&d" (old_val), "=&d" (new_val),
-			       "=m" (sem->count)
-			     : "a" (&sem->count), "d" (incr), "m" (sem->count)
-			     : "cc" );
+	asm volatile(
+		"	l	%0,0(%3)\n"
+		"0:	ltr	%1,%0\n"
+		"	jhe	1f\n"
+		"	lhi	%1,0\n"
+		"1:	ar	%1,%4\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b\n"
+		: "=&d" (old_val), "=&d" (new_val), "=m" (sem->count)
+		: "a" (&sem->count), "d" (incr), "m" (sem->count)
+		: "cc");
 	return old_val;
 }
 
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index e3d9325f6022..a21cfbb9d97e 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -101,7 +101,7 @@ void __devinit cpu_init (void)
         /*
          * Store processor id in lowcore (used e.g. in timer_interrupt)
          */
-        asm volatile ("stidp %0": "=m" (S390_lowcore.cpu_data.cpu_id));
+	asm volatile("stidp %0": "=m" (S390_lowcore.cpu_data.cpu_id));
         S390_lowcore.cpu_data.cpu_addr = addr;
 
         /*
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index b2e6f4c8d382..a8e6199755d4 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -63,7 +63,7 @@ static void smp_ext_bitcall(int, ec_bit_sig);
 static void smp_ext_bitcall_others(ec_bit_sig);
 
 /*
- * Structure and data for smp_call_function(). This is designed to minimise
+5B * Structure and data for smp_call_function(). This is designed to minimise
  * static memory requirements. It also looks cleaner.
  */
 static DEFINE_SPINLOCK(call_lock);
@@ -418,59 +418,49 @@ void smp_send_reschedule(int cpu)
 /*
  * parameter area for the set/clear control bit callbacks
  */
-typedef struct
-{
-	__u16 start_ctl;
-	__u16 end_ctl;
+struct ec_creg_mask_parms {
 	unsigned long orvals[16];
 	unsigned long andvals[16];
-} ec_creg_mask_parms;
+};
 
 /*
  * callback for setting/clearing control bits
  */
 void smp_ctl_bit_callback(void *info) {
-	ec_creg_mask_parms *pp;
+	struct ec_creg_mask_parms *pp = info;
 	unsigned long cregs[16];
 	int i;
 	
-	pp = (ec_creg_mask_parms *) info;
-	__ctl_store(cregs[pp->start_ctl], pp->start_ctl, pp->end_ctl);
-	for (i = pp->start_ctl; i <= pp->end_ctl; i++)
+	__ctl_store(cregs, 0, 15);
+	for (i = 0; i <= 15; i++)
 		cregs[i] = (cregs[i] & pp->andvals[i]) | pp->orvals[i];
-	__ctl_load(cregs[pp->start_ctl], pp->start_ctl, pp->end_ctl);
+	__ctl_load(cregs, 0, 15);
 }
 
 /*
  * Set a bit in a control register of all cpus
  */
-void smp_ctl_set_bit(int cr, int bit) {
-        ec_creg_mask_parms parms;
+void smp_ctl_set_bit(int cr, int bit)
+{
+	struct ec_creg_mask_parms parms;
 
-	parms.start_ctl = cr;
-	parms.end_ctl = cr;
+	memset(&parms.orvals, 0, sizeof(parms.orvals));
+	memset(&parms.andvals, 0xff, sizeof(parms.andvals));
 	parms.orvals[cr] = 1 << bit;
-	parms.andvals[cr] = -1L;
-	preempt_disable();
-	smp_call_function(smp_ctl_bit_callback, &parms, 0, 1);
-        __ctl_set_bit(cr, bit);
-	preempt_enable();
+	on_each_cpu(smp_ctl_bit_callback, &parms, 0, 1);
 }
 
 /*
  * Clear a bit in a control register of all cpus
  */
-void smp_ctl_clear_bit(int cr, int bit) {
-        ec_creg_mask_parms parms;
+void smp_ctl_clear_bit(int cr, int bit)
+{
+	struct ec_creg_mask_parms parms;
 
-	parms.start_ctl = cr;
-	parms.end_ctl = cr;
-	parms.orvals[cr] = 0;
+	memset(&parms.orvals, 0, sizeof(parms.orvals));
+	memset(&parms.andvals, 0xff, sizeof(parms.andvals));
 	parms.andvals[cr] = ~(1L << bit);
-	preempt_disable();
-	smp_call_function(smp_ctl_bit_callback, &parms, 0, 1);
-        __ctl_clear_bit(cr, bit);
-	preempt_enable();
+	on_each_cpu(smp_ctl_bit_callback, &parms, 0, 1);
 }
 
 /*
@@ -650,9 +640,9 @@ __cpu_up(unsigned int cpu)
 	sf->gprs[9] = (unsigned long) sf;
 	cpu_lowcore->save_area[15] = (unsigned long) sf;
 	__ctl_store(cpu_lowcore->cregs_save_area[0], 0, 15);
-	__asm__ __volatile__("stam  0,15,0(%0)"
-			     : : "a" (&cpu_lowcore->access_regs_save_area)
-			     : "memory");
+	asm volatile(
+		"	stam	0,15,0(%0)"
+		: : "a" (&cpu_lowcore->access_regs_save_area) : "memory");
 	cpu_lowcore->percpu_offset = __per_cpu_offset[cpu];
         cpu_lowcore->current_task = (unsigned long) idle;
         cpu_lowcore->cpu_data.cpu_nr = cpu;
@@ -708,7 +698,7 @@ int
 __cpu_disable(void)
 {
 	unsigned long flags;
-	ec_creg_mask_parms cr_parms;
+	struct ec_creg_mask_parms cr_parms;
 	int cpu = smp_processor_id();
 
 	spin_lock_irqsave(&smp_reserve_lock, flags);
@@ -724,30 +714,21 @@ __cpu_disable(void)
 		pfault_fini();
 #endif
 
-	/* disable all external interrupts */
+	memset(&cr_parms.orvals, 0, sizeof(cr_parms.orvals));
+	memset(&cr_parms.andvals, 0xff, sizeof(cr_parms.andvals));
 
-	cr_parms.start_ctl = 0;
-	cr_parms.end_ctl = 0;
+	/* disable all external interrupts */
 	cr_parms.orvals[0] = 0;
 	cr_parms.andvals[0] = ~(1<<15 | 1<<14 | 1<<13 | 1<<12 |
 				1<<11 | 1<<10 | 1<< 6 | 1<< 4);
-	smp_ctl_bit_callback(&cr_parms);
-
 	/* disable all I/O interrupts */
-
-	cr_parms.start_ctl = 6;
-	cr_parms.end_ctl = 6;
 	cr_parms.orvals[6] = 0;
 	cr_parms.andvals[6] = ~(1<<31 | 1<<30 | 1<<29 | 1<<28 |
 				1<<27 | 1<<26 | 1<<25 | 1<<24);
-	smp_ctl_bit_callback(&cr_parms);
-
 	/* disable most machine checks */
-
-	cr_parms.start_ctl = 14;
-	cr_parms.end_ctl = 14;
 	cr_parms.orvals[14] = 0;
 	cr_parms.andvals[14] = ~(1<<28 | 1<<27 | 1<<26 | 1<<25 | 1<<24);
+
 	smp_ctl_bit_callback(&cr_parms);
 
 	spin_unlock_irqrestore(&smp_reserve_lock, flags);
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 74e6178fbaf2..1981c6199fa2 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -351,10 +351,12 @@ void __init time_init(void)
 	int cc;
 
         /* kick the TOD clock */
-        asm volatile ("STCK 0(%1)\n\t"
-                      "IPM  %0\n\t"
-                      "SRL  %0,28" : "=r" (cc) : "a" (&init_timer_cc) 
-				   : "memory", "cc");
+	asm volatile(
+		"	stck	0(%2)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (cc), "=m" (init_timer_cc)
+		: "a" (&init_timer_cc) : "cc");
         switch (cc) {
         case 0: /* clock in set state: all is fine */
                 break;
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index c4982c963424..3eb4fab048b8 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -597,8 +597,7 @@ asmlinkage void data_exception(struct pt_regs * regs, long interruption_code)
 		local_irq_enable();
 
 	if (MACHINE_HAS_IEEE)
-		__asm__ volatile ("stfpc %0\n\t" 
-				  : "=m" (current->thread.fp_regs.fpc));
+		asm volatile("stfpc %0" : "=m" (current->thread.fp_regs.fpc));
 
 #ifdef CONFIG_MATHEMU
         else if (regs->psw.mask & PSW_MASK_PSTATE) {
diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index 468f4ea33f99..027c4742a001 100644
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -27,9 +27,7 @@ void __delay(unsigned long loops)
          * yield the megahertz number of the cpu. The important function
          * is udelay and that is done using the tod clock. -- martin.
          */
-        __asm__ __volatile__(
-                "0: brct %0,0b"
-                : /* no outputs */ : "r" ((loops/2) + 1));
+	asm volatile("0: brct %0,0b" : : "d" ((loops/2) + 1));
 }
 
 /*
@@ -38,13 +36,12 @@ void __delay(unsigned long loops)
  */
 void __udelay(unsigned long usecs)
 {
-        uint64_t start_cc, end_cc;
+	uint64_t start_cc;
 
         if (usecs == 0)
                 return;
-        asm volatile ("STCK %0" : "=m" (start_cc));
+	start_cc = get_clock();
         do {
 		cpu_relax();
-                asm volatile ("STCK %0" : "=m" (end_cc));
-        } while (((end_cc - start_cc)/4096) < usecs);
+	} while (((get_clock() - start_cc)/4096) < usecs);
 }
diff --git a/arch/s390/math-emu/math.c b/arch/s390/math-emu/math.c
index b4957c84e4d6..6b9aec5a2c18 100644
--- a/arch/s390/math-emu/math.c
+++ b/arch/s390/math-emu/math.c
@@ -1564,52 +1564,52 @@ static int emu_tceb (struct pt_regs *regs, int rx, long val) {
 }
 
 static inline void emu_load_regd(int reg) {
-        if ((reg&9) != 0)         /* test if reg in {0,2,4,6} */
+	if ((reg&9) != 0)	/* test if reg in {0,2,4,6} */
                 return;
-        asm volatile (            /* load reg from fp_regs.fprs[reg] */
-                "     bras  1,0f\n"
-                "     ld    0,0(%1)\n"
-                "0:   ex    %0,0(1)"
-                : /* no output */
-                : "a" (reg<<4),"a" (&current->thread.fp_regs.fprs[reg].d)
-                : "1" );
+	asm volatile(		/* load reg from fp_regs.fprs[reg] */
+		"	bras	1,0f\n"
+		"	ld	0,0(%1)\n"
+		"0:	ex	%0,0(1)"
+		: /* no output */
+		: "a" (reg<<4),"a" (&current->thread.fp_regs.fprs[reg].d)
+		: "1");
 }
 
 static inline void emu_load_rege(int reg) {
-        if ((reg&9) != 0)         /* test if reg in {0,2,4,6} */
+	if ((reg&9) != 0)	/* test if reg in {0,2,4,6} */
                 return;
-        asm volatile (            /* load reg from fp_regs.fprs[reg] */
-                "     bras  1,0f\n"
-                "     le    0,0(%1)\n"
-                "0:   ex    %0,0(1)"
-                : /* no output */
-                : "a" (reg<<4), "a" (&current->thread.fp_regs.fprs[reg].f)
-                : "1" );
+	asm volatile(		/* load reg from fp_regs.fprs[reg] */
+		"	bras	1,0f\n"
+		"	le	0,0(%1)\n"
+		"0:	ex	%0,0(1)"
+		: /* no output */
+		: "a" (reg<<4), "a" (&current->thread.fp_regs.fprs[reg].f)
+		: "1");
 }
 
 static inline void emu_store_regd(int reg) {
-        if ((reg&9) != 0)         /* test if reg in {0,2,4,6} */
+	if ((reg&9) != 0)	/* test if reg in {0,2,4,6} */
                 return;
-        asm volatile (            /* store reg to fp_regs.fprs[reg] */
-                "     bras  1,0f\n"
-                "     std   0,0(%1)\n"
-                "0:   ex    %0,0(1)"
-                : /* no output */
-                : "a" (reg<<4), "a" (&current->thread.fp_regs.fprs[reg].d)
-                : "1" );
+	asm volatile(		/* store reg to fp_regs.fprs[reg] */
+		"	bras	1,0f\n"
+		"	std	0,0(%1)\n"
+		"0:	ex	%0,0(1)"
+		: /* no output */
+		: "a" (reg<<4), "a" (&current->thread.fp_regs.fprs[reg].d)
+		: "1");
 }
 
 
 static inline void emu_store_rege(int reg) {
-        if ((reg&9) != 0)         /* test if reg in {0,2,4,6} */
+	if ((reg&9) != 0)	/* test if reg in {0,2,4,6} */
                 return;
-        asm volatile (            /* store reg to fp_regs.fprs[reg] */
-                "     bras  1,0f\n"
-                "     ste   0,0(%1)\n"
-                "0:   ex    %0,0(1)"
-                : /* no output */
-                : "a" (reg<<4), "a" (&current->thread.fp_regs.fprs[reg].f)
-                : "1" );
+	asm volatile(		/* store reg to fp_regs.fprs[reg] */
+		"	bras	1,0f\n"
+		"	ste	0,0(%1)\n"
+		"0:	ex	%0,0(1)"
+		: /* no output */
+		: "a" (reg<<4), "a" (&current->thread.fp_regs.fprs[reg].f)
+		: "1");
 }
 
 int math_emu_b3(__u8 *opcode, struct pt_regs * regs) {
@@ -2089,23 +2089,22 @@ int math_emu_ldr(__u8 *opcode) {
 
         if ((opc & 0x90) == 0) {           /* test if rx in {0,2,4,6} */
                 /* we got an exception therfore ry can't be in {0,2,4,6} */
-                __asm__ __volatile (       /* load rx from fp_regs.fprs[ry] */
-                        "     bras  1,0f\n"
-                        "     ld    0,0(%1)\n"
-                        "0:   ex    %0,0(1)"
-                        : /* no output */
-                        : "a" (opc & 0xf0),
-                          "a" (&fp_regs->fprs[opc & 0xf].d)
-                        : "1" );
+		asm volatile(		/* load rx from fp_regs.fprs[ry] */
+			"	bras	1,0f\n"
+			"	ld	0,0(%1)\n"
+			"0:	ex	%0,0(1)"
+			: /* no output */
+			: "a" (opc & 0xf0), "a" (&fp_regs->fprs[opc & 0xf].d)
+			: "1");
         } else if ((opc & 0x9) == 0) {     /* test if ry in {0,2,4,6} */
-                __asm__ __volatile (       /* store ry to fp_regs.fprs[rx] */
-                        "     bras  1,0f\n"
-                        "     std   0,0(%1)\n"
-                        "0:   ex    %0,0(1)"
-                        : /* no output */
-                        : "a" ((opc & 0xf) << 4),
-                          "a" (&fp_regs->fprs[(opc & 0xf0)>>4].d)
-                        : "1" );
+		asm volatile (		/* store ry to fp_regs.fprs[rx] */
+			"	bras	1,0f\n"
+			"	std	0,0(%1)\n"
+			"0:	ex	%0,0(1)"
+			: /* no output */
+			: "a" ((opc & 0xf) << 4),
+			  "a" (&fp_regs->fprs[(opc & 0xf0)>>4].d)
+			: "1");
         } else  /* move fp_regs.fprs[ry] to fp_regs.fprs[rx] */
                 fp_regs->fprs[(opc & 0xf0) >> 4] = fp_regs->fprs[opc & 0xf];
 	return 0;
@@ -2120,23 +2119,22 @@ int math_emu_ler(__u8 *opcode) {
 
         if ((opc & 0x90) == 0) {           /* test if rx in {0,2,4,6} */
                 /* we got an exception therfore ry can't be in {0,2,4,6} */
-                __asm__ __volatile (       /* load rx from fp_regs.fprs[ry] */
-                        "     bras  1,0f\n"
-                        "     le    0,0(%1)\n"
-                        "0:   ex    %0,0(1)"
-                        : /* no output */
-                        : "a" (opc & 0xf0),
-                          "a" (&fp_regs->fprs[opc & 0xf].f)
-                        : "1" );
+		asm volatile(		/* load rx from fp_regs.fprs[ry] */
+			"	bras	1,0f\n"
+			"	le	0,0(%1)\n"
+			"0:	ex	%0,0(1)"
+			: /* no output */
+			: "a" (opc & 0xf0), "a" (&fp_regs->fprs[opc & 0xf].f)
+			: "1");
         } else if ((opc & 0x9) == 0) {     /* test if ry in {0,2,4,6} */
-                __asm__ __volatile (       /* store ry to fp_regs.fprs[rx] */
-                        "     bras  1,0f\n"
-                        "     ste   0,0(%1)\n"
-                        "0:   ex    %0,0(1)"
-                        : /* no output */
-                        : "a" ((opc & 0xf) << 4),
-                          "a" (&fp_regs->fprs[(opc & 0xf0) >> 4].f)
-                        : "1" );
+		asm volatile(		/* store ry to fp_regs.fprs[rx] */
+			"	bras	1,0f\n"
+			"	ste	0,0(%1)\n"
+			"0:	ex	%0,0(1)"
+			: /* no output */
+			: "a" ((opc & 0xf) << 4),
+			  "a" (&fp_regs->fprs[(opc & 0xf0) >> 4].f)
+			: "1");
         } else  /* move fp_regs.fprs[ry] to fp_regs.fprs[rx] */
                 fp_regs->fprs[(opc & 0xf0) >> 4] = fp_regs->fprs[opc & 0xf];
 	return 0;
diff --git a/arch/s390/math-emu/sfp-util.h b/arch/s390/math-emu/sfp-util.h
index ab556b600f73..5b6ca4570ea4 100644
--- a/arch/s390/math-emu/sfp-util.h
+++ b/arch/s390/math-emu/sfp-util.h
@@ -4,48 +4,51 @@
 #include <asm/byteorder.h>
 
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) ({		\
-        unsigned int __sh = (ah);			\
-        unsigned int __sl = (al);			\
-        __asm__ ("   alr  %1,%3\n"			\
-                 "   brc  12,0f\n"			\
-                 "   ahi  %0,1\n"			\
-                 "0: alr  %0,%2"			\
-                 : "+&d" (__sh), "+d" (__sl)		\
-                 : "d" (bh), "d" (bl) : "cc" );		\
-        (sh) = __sh;					\
-        (sl) = __sl;					\
+	unsigned int __sh = (ah);			\
+	unsigned int __sl = (al);			\
+	asm volatile(					\
+		"	alr	%1,%3\n"		\
+		"	brc	12,0f\n"		\
+		"	ahi	%0,1\n"			\
+		"0:	alr  %0,%2"			\
+		: "+&d" (__sh), "+d" (__sl)		\
+		: "d" (bh), "d" (bl) : "cc");		\
+	(sh) = __sh;					\
+	(sl) = __sl;					\
 })
 
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) ({		\
-       unsigned int __sh = (ah);			\
-       unsigned int __sl = (al);			\
-       __asm__ ("   slr  %1,%3\n"			\
-                "   brc  3,0f\n"			\
-                "   ahi  %0,-1\n"			\
-                "0: slr  %0,%2"				\
-                : "+&d" (__sh), "+d" (__sl)		\
-                : "d" (bh), "d" (bl) : "cc" );		\
-       (sh) = __sh;					\
-       (sl) = __sl;					\
+	unsigned int __sh = (ah);			\
+	unsigned int __sl = (al);			\
+	asm volatile(					\
+		"	slr	%1,%3\n"		\
+		"	brc	3,0f\n"			\
+		"	ahi	%0,-1\n"		\
+		"0:	slr	%0,%2"			\
+		: "+&d" (__sh), "+d" (__sl)		\
+		: "d" (bh), "d" (bl) : "cc");		\
+	(sh) = __sh;					\
+	(sl) = __sl;					\
 })
 
 /* a umul b = a mul b + (a>=2<<31) ? b<<32:0 + (b>=2<<31) ? a<<32:0 */
 #define umul_ppmm(wh, wl, u, v) ({			\
-        unsigned int __wh = u;				\
-        unsigned int __wl = v;				\
-        __asm__ ("   ltr  1,%0\n"			\
-                 "   mr   0,%1\n"			\
-                 "   jnm  0f\n"				\
-                 "   alr  0,%1\n"			\
-                 "0: ltr  %1,%1\n"			\
-                 "   jnm  1f\n"				\
-                 "   alr  0,%0\n"			\
-                 "1: lr   %0,0\n"			\
-                 "   lr   %1,1\n"			\
-                 : "+d" (__wh), "+d" (__wl)		\
-                 : : "0", "1", "cc" );			\
-        wh = __wh;					\
-        wl = __wl;					\
+	unsigned int __wh = u;				\
+	unsigned int __wl = v;				\
+	asm volatile(					\
+		"	ltr	1,%0\n"			\
+		"	mr	0,%1\n"			\
+		"	jnm	0f\n"				\
+		"	alr	0,%1\n"			\
+		"0:	ltr	%1,%1\n"			\
+		"	jnm	1f\n"				\
+		"	alr	0,%0\n"			\
+		"1:	lr	%0,0\n"			\
+		"	lr	%1,1\n"			\
+		: "+d" (__wh), "+d" (__wl)		\
+		: : "0", "1", "cc");			\
+	wh = __wh;					\
+	wl = __wl;					\
 })
 
 #define udiv_qrnnd(q, r, n1, n0, d)			\
diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index 9b11e3e20903..226275d5c4f6 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c
@@ -142,17 +142,17 @@ dcss_diag (__u8 func, void *parameter,
 
 	rx = (unsigned long) parameter;
 	ry = (unsigned long) func;
-	__asm__ __volatile__(
+	asm volatile(
 #ifdef CONFIG_64BIT
-		"   sam31\n" // switch to 31 bit
-		"   diag    %0,%1,0x64\n"
-		"   sam64\n" // switch back to 64 bit
+		"	sam31\n"
+		"	diag	%0,%1,0x64\n"
+		"	sam64\n"
 #else
-		"   diag    %0,%1,0x64\n"
+		"	diag	%0,%1,0x64\n"
 #endif
-		"   ipm     %2\n"
-		"   srl     %2,28\n"
-		: "+d" (rx), "+d" (ry), "=d" (rc) : : "cc" );
+		"	ipm	%2\n"
+		"	srl	%2,28\n"
+		: "+d" (rx), "+d" (ry), "=d" (rc) : : "cc");
 	*ret1 = rx;
 	*ret2 = ry;
 	return rc;
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index a393c308bb29..f2b9a84dc2bf 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -424,20 +424,13 @@ int pfault_init(void)
 
 	if (pfault_disable)
 		return -1;
-        __asm__ __volatile__(
-                "    diag  %1,%0,0x258\n"
-		"0:  j     2f\n"
-		"1:  la    %0,8\n"
+	asm volatile(
+		"	diag	%1,%0,0x258\n"
+		"0:	j	2f\n"
+		"1:	la	%0,8\n"
 		"2:\n"
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-#ifndef CONFIG_64BIT
-		"   .long  0b,1b\n"
-#else /* CONFIG_64BIT */
-		"   .quad  0b,1b\n"
-#endif /* CONFIG_64BIT */
-		".previous"
-                : "=d" (rc) : "a" (&refbk), "m" (refbk) : "cc" );
+		EX_TABLE(0b,1b)
+		: "=d" (rc) : "a" (&refbk), "m" (refbk) : "cc");
         __ctl_set_bit(0, 9);
         return rc;
 }
@@ -450,18 +443,11 @@ void pfault_fini(void)
 	if (pfault_disable)
 		return;
 	__ctl_clear_bit(0,9);
-        __asm__ __volatile__(
-                "    diag  %0,0,0x258\n"
+	asm volatile(
+		"	diag	%0,0,0x258\n"
 		"0:\n"
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-#ifndef CONFIG_64BIT
-		"   .long  0b,0b\n"
-#else /* CONFIG_64BIT */
-		"   .quad  0b,0b\n"
-#endif /* CONFIG_64BIT */
-		".previous"
-		: : "a" (&refbk), "m" (refbk) : "cc" );
+		EX_TABLE(0b,0b)
+		: : "a" (&refbk), "m" (refbk) : "cc");
 }
 
 asmlinkage void
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index cfd9b8f7a523..127044e1707c 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -45,26 +45,17 @@ void diag10(unsigned long addr)
 {
         if (addr >= 0x7ff00000)
                 return;
+	asm volatile(
 #ifdef CONFIG_64BIT
-        asm volatile (
-		"   sam31\n"
-		"   diag %0,%0,0x10\n"
-		"0: sam64\n"
-		".section __ex_table,\"a\"\n"
-		"   .align 8\n"
-		"   .quad 0b, 0b\n"
-		".previous\n"
-		: : "a" (addr));
+		"	sam31\n"
+		"	diag	%0,%0,0x10\n"
+		"0:	sam64\n"
 #else
-        asm volatile (
-		"   diag %0,%0,0x10\n"
+		"	diag	%0,%0,0x10\n"
 		"0:\n"
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-		"   .long 0b, 0b\n"
-		".previous\n"
-		: : "a" (addr));
 #endif
+		EX_TABLE(0b,0b)
+		: : "a" (addr));
 }
 
 void show_mem(void)
@@ -156,11 +147,10 @@ void __init paging_init(void)
 	S390_lowcore.kernel_asce = pgdir_k;
 
         /* enable virtual mapping in kernel mode */
-        __asm__ __volatile__("    LCTL  1,1,%0\n"
-                             "    LCTL  7,7,%0\n"
-                             "    LCTL  13,13,%0\n"
-                             "    SSM   %1" 
-			     : : "m" (pgdir_k), "m" (ssm_mask));
+	__ctl_load(pgdir_k, 1, 1);
+	__ctl_load(pgdir_k, 7, 7);
+	__ctl_load(pgdir_k, 13, 13);
+	__raw_local_irq_ssm(ssm_mask);
 
         local_flush_tlb();
         return;
@@ -241,11 +231,10 @@ void __init paging_init(void)
 	S390_lowcore.kernel_asce = pgdir_k;
 
         /* enable virtual mapping in kernel mode */
-        __asm__ __volatile__("lctlg 1,1,%0\n\t"
-                             "lctlg 7,7,%0\n\t"
-                             "lctlg 13,13,%0\n\t"
-                             "ssm   %1"
-			     : :"m" (pgdir_k), "m" (ssm_mask));
+	__ctl_load(pgdir_k, 1, 1);
+	__ctl_load(pgdir_k, 7, 7);
+	__ctl_load(pgdir_k, 13, 13);
+	__raw_local_irq_ssm(ssm_mask);
 
         local_flush_tlb();
 
diff --git a/drivers/s390/block/dasd_diag.c b/drivers/s390/block/dasd_diag.c
index 23fa0b289173..9d051e5687ea 100644
--- a/drivers/s390/block/dasd_diag.c
+++ b/drivers/s390/block/dasd_diag.c
@@ -63,44 +63,26 @@ static const u8 DASD_DIAG_CMS1[] = { 0xc3, 0xd4, 0xe2, 0xf1 };/* EBCDIC CMS1 */
  * and function code cmd.
  * In case of an exception return 3. Otherwise return result of bitwise OR of
  * resulting condition code and DIAG return code. */
-static __inline__ int
-dia250(void *iob, int cmd)
+static inline int dia250(void *iob, int cmd)
 {
+	register unsigned long reg0 asm ("0") = (unsigned long) iob;
 	typedef union {
 		struct dasd_diag_init_io init_io;
 		struct dasd_diag_rw_io rw_io;
 	} addr_type;
 	int rc;
 
-	__asm__ __volatile__(
-#ifdef CONFIG_64BIT
-		"	lghi	%0,3\n"
-		"	lgr	0,%3\n"
-		"	diag	0,%2,0x250\n"
-		"0:	ipm	%0\n"
-		"	srl	%0,28\n"
-		"	or	%0,1\n"
-		"1:\n"
-		".section __ex_table,\"a\"\n"
-		"	.align 8\n"
-		"	.quad  0b,1b\n"
-		".previous\n"
-#else
-		"	lhi	%0,3\n"
-		"	lr	0,%3\n"
+	rc = 3;
+	asm volatile(
 		"	diag	0,%2,0x250\n"
 		"0:	ipm	%0\n"
 		"	srl	%0,28\n"
 		"	or	%0,1\n"
 		"1:\n"
-		".section __ex_table,\"a\"\n"
-		"	.align 4\n"
-		"	.long 0b,1b\n"
-		".previous\n"
-#endif
-		: "=&d" (rc), "=m" (*(addr_type *) iob)
-		: "d" (cmd), "d" (iob), "m" (*(addr_type *) iob)
-		: "0", "1", "cc");
+		EX_TABLE(0b,1b)
+		: "+d" (rc), "=m" (*(addr_type *) iob)
+		: "d" (cmd), "d" (reg0), "m" (*(addr_type *) iob)
+		: "1", "cc");
 	return rc;
 }
 
diff --git a/drivers/s390/block/xpram.c b/drivers/s390/block/xpram.c
index cab2c736683a..a04d9120cef0 100644
--- a/drivers/s390/block/xpram.c
+++ b/drivers/s390/block/xpram.c
@@ -89,28 +89,15 @@ MODULE_LICENSE("GPL");
  */
 static int xpram_page_in (unsigned long page_addr, unsigned int xpage_index)
 {
-	int cc;
+	int cc = 2;	/* return unused cc 2 if pgin traps */
 
-	__asm__ __volatile__ (
-		"   lhi   %0,2\n"  /* return unused cc 2 if pgin traps */
-		"   .insn rre,0xb22e0000,%1,%2\n"  /* pgin %1,%2 */
-                "0: ipm   %0\n"
-		"   srl   %0,28\n"
+	asm volatile(
+		"	.insn	rre,0xb22e0000,%1,%2\n"  /* pgin %1,%2 */
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
 		"1:\n"
-#ifndef CONFIG_64BIT
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-		"   .long  0b,1b\n"
-		".previous"
-#else
-                ".section __ex_table,\"a\"\n"
-                "   .align 8\n"
-                "   .quad 0b,1b\n"
-                ".previous"
-#endif
-		: "=&d" (cc) 
-		: "a" (__pa(page_addr)), "a" (xpage_index) 
-		: "cc" );
+		EX_TABLE(0b,1b)
+		: "+d" (cc) : "a" (__pa(page_addr)), "d" (xpage_index) : "cc");
 	if (cc == 3)
 		return -ENXIO;
 	if (cc == 2) {
@@ -137,28 +124,15 @@ static int xpram_page_in (unsigned long page_addr, unsigned int xpage_index)
  */
 static long xpram_page_out (unsigned long page_addr, unsigned int xpage_index)
 {
-	int cc;
+	int cc = 2;	/* return unused cc 2 if pgin traps */
 
-	__asm__ __volatile__ (
-		"   lhi   %0,2\n"  /* return unused cc 2 if pgout traps */
-		"   .insn rre,0xb22f0000,%1,%2\n"  /* pgout %1,%2 */
-                "0: ipm   %0\n"
-		"   srl   %0,28\n"
+	asm volatile(
+		"	.insn	rre,0xb22f0000,%1,%2\n"  /* pgout %1,%2 */
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
 		"1:\n"
-#ifndef CONFIG_64BIT
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-		"   .long  0b,1b\n"
-		".previous"
-#else
-                ".section __ex_table,\"a\"\n"
-                "   .align 8\n"
-                "   .quad 0b,1b\n"
-                ".previous"
-#endif
-		: "=&d" (cc) 
-		: "a" (__pa(page_addr)), "a" (xpage_index) 
-		: "cc" );
+		EX_TABLE(0b,1b)
+		: "+d" (cc) : "a" (__pa(page_addr)), "d" (xpage_index) : "cc");
 	if (cc == 3)
 		return -ENXIO;
 	if (cc == 2) {
diff --git a/drivers/s390/char/sclp.c b/drivers/s390/char/sclp.c
index 985d1613baaa..31e335751d6d 100644
--- a/drivers/s390/char/sclp.c
+++ b/drivers/s390/char/sclp.c
@@ -100,13 +100,12 @@ service_call(sclp_cmdw_t command, void *sccb)
 {
 	int cc;
 
-	__asm__ __volatile__(
-		"   .insn rre,0xb2200000,%1,%2\n"  /* servc %1,%2 */
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=&d" (cc)
-		: "d" (command), "a" (__pa(sccb))
-		: "cc", "memory" );
+	asm volatile(
+		"	.insn	rre,0xb2200000,%1,%2\n"  /* servc %1,%2 */
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=&d" (cc) : "d" (command), "a" (__pa(sccb))
+		: "cc", "memory");
 	if (cc == 3)
 		return -EIO;
 	if (cc == 2)
@@ -360,16 +359,6 @@ sclp_interrupt_handler(struct pt_regs *regs, __u16 code)
 	sclp_process_queue();
 }
 
-/* Return current Time-Of-Day clock. */
-static inline u64
-sclp_get_clock(void)
-{
-	u64 result;
-
-	asm volatile ("STCK 0(%1)" : "=m" (result) : "a" (&(result)) : "cc");
-	return result;
-}
-
 /* Convert interval in jiffies to TOD ticks. */
 static inline u64
 sclp_tod_from_jiffies(unsigned long jiffies)
@@ -382,7 +371,6 @@ sclp_tod_from_jiffies(unsigned long jiffies)
 void
 sclp_sync_wait(void)
 {
-	unsigned long psw_mask;
 	unsigned long flags;
 	unsigned long cr0, cr0_sync;
 	u64 timeout;
@@ -392,7 +380,7 @@ sclp_sync_wait(void)
 	timeout = 0;
 	if (timer_pending(&sclp_request_timer)) {
 		/* Get timeout TOD value */
-		timeout = sclp_get_clock() +
+		timeout = get_clock() +
 			  sclp_tod_from_jiffies(sclp_request_timer.expires -
 						jiffies);
 	}
@@ -406,13 +394,12 @@ sclp_sync_wait(void)
 	cr0_sync |= 0x00000200;
 	cr0_sync &= 0xFFFFF3AC;
 	__ctl_load(cr0_sync, 0, 0);
-	asm volatile ("STOSM 0(%1),0x01"
-		      : "=m" (psw_mask) : "a" (&psw_mask) : "memory");
+	__raw_local_irq_stosm(0x01);
 	/* Loop until driver state indicates finished request */
 	while (sclp_running_state != sclp_running_state_idle) {
 		/* Check for expired request timer */
 		if (timer_pending(&sclp_request_timer) &&
-		    sclp_get_clock() > timeout &&
+		    get_clock() > timeout &&
 		    del_timer(&sclp_request_timer))
 			sclp_request_timer.function(sclp_request_timer.data);
 		barrier();
diff --git a/drivers/s390/char/vmwatchdog.c b/drivers/s390/char/vmwatchdog.c
index 807320a41fa4..4b868f72fe89 100644
--- a/drivers/s390/char/vmwatchdog.c
+++ b/drivers/s390/char/vmwatchdog.c
@@ -54,48 +54,20 @@ enum vmwdt_func {
 static int __diag288(enum vmwdt_func func, unsigned int timeout,
 			    char *cmd, size_t len)
 {
-	register unsigned long __func asm("2");
-	register unsigned long __timeout asm("3");
-	register unsigned long __cmdp asm("4");
-	register unsigned long __cmdl asm("5");
+	register unsigned long __func asm("2") = func;
+	register unsigned long __timeout asm("3") = timeout;
+	register unsigned long __cmdp asm("4") = virt_to_phys(cmd);
+	register unsigned long __cmdl asm("5") = len;
 	int err;
 
-	__func = func;
-	__timeout = timeout;
-	__cmdp = virt_to_phys(cmd);
-	__cmdl = len;
-	err = 0;
-	asm volatile (
-#ifdef CONFIG_64BIT
-		       "diag %2,%4,0x288\n"
-		"1:	\n"
-		".section .fixup,\"ax\"\n"
-		"2:	lghi %0,%1\n"
-		"	jg 1b\n"
-		".previous\n"
-		".section __ex_table,\"a\"\n"
-		"	.align 8\n"
-		"	.quad 1b,2b\n"
-		".previous\n"
-#else
-		       "diag %2,%4,0x288\n"
-		"1:	\n"
-		".section .fixup,\"ax\"\n"
-		"2:	lhi %0,%1\n"
-		"	bras 1,3f\n"
-		"	.long 1b\n"
-		"3:	l 1,0(1)\n"
-		"	br 1\n"
-		".previous\n"
-		".section __ex_table,\"a\"\n"
-		"	.align 4\n"
-		"	.long 1b,2b\n"
-		".previous\n"
-#endif
-		: "+&d"(err)
-		: "i"(-EINVAL), "d"(__func), "d"(__timeout),
-		  "d"(__cmdp), "d"(__cmdl)
-		: "1", "cc");
+	err = -EINVAL;
+	asm volatile(
+		"	diag	%1,%3,0x288\n"
+		"0:	la	%0,0\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "=d" (err) : "d"(__func), "d"(__timeout),
+		  "d"(__cmdp), "d"(__cmdl), "0" (-EINVAL) : "1", "cc");
 	return err;
 }
 
diff --git a/drivers/s390/cio/device_id.c b/drivers/s390/cio/device_id.c
index 438db483035d..1398367b5f68 100644
--- a/drivers/s390/cio/device_id.c
+++ b/drivers/s390/cio/device_id.c
@@ -42,18 +42,15 @@ diag210(struct diag210 * addr)
 	spin_lock_irqsave(&diag210_lock, flags);
 	diag210_tmp = *addr;
 
-	asm volatile (
-		"   lhi	  %0,-1\n"
-		"   sam31\n"
-		"   diag  %1,0,0x210\n"
-		"0: ipm	  %0\n"
-		"   srl	  %0,28\n"
-		"1: sam64\n"
-		".section __ex_table,\"a\"\n"
-		"    .align 8\n"
-		"    .quad 0b,1b\n"
-		".previous"
-		: "=&d" (ccode) : "a" (__pa(&diag210_tmp)) : "cc", "memory" );
+	asm volatile(
+		"	lhi	%0,-1\n"
+		"	sam31\n"
+		"	diag	%1,0,0x210\n"
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
+		"1:	sam64\n"
+		EX_TABLE(0b,1b)
+		: "=&d" (ccode) : "a" (__pa(&diag210_tmp)) : "cc", "memory");
 
 	*addr = diag210_tmp;
 	spin_unlock_irqrestore(&diag210_lock, flags);
@@ -66,17 +63,14 @@ diag210(struct diag210 * addr)
 {
 	int ccode;
 
-	asm volatile (
-		"   lhi	  %0,-1\n"
-		"   diag  %1,0,0x210\n"
-		"0: ipm	  %0\n"
-		"   srl	  %0,28\n"
+	asm volatile(
+		"	lhi	%0,-1\n"
+		"	diag	%1,0,0x210\n"
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
 		"1:\n"
-		".section __ex_table,\"a\"\n"
-		"    .align 4\n"
-		"    .long 0b,1b\n"
-		".previous"
-		: "=&d" (ccode) : "a" (__pa(addr)) : "cc", "memory" );
+		EX_TABLE(0b,1b)
+		: "=&d" (ccode) : "a" (__pa(addr)) : "cc", "memory");
 
 	return ccode;
 }
diff --git a/drivers/s390/cio/ioasm.h b/drivers/s390/cio/ioasm.h
index 95a9462f9a91..ad6d82940069 100644
--- a/drivers/s390/cio/ioasm.h
+++ b/drivers/s390/cio/ioasm.h
@@ -25,106 +25,74 @@ struct tpi_info {
 static inline int stsch(struct subchannel_id schid,
 			    volatile struct schib *addr)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   stsch 0(%2)\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid), "a" (addr), "m" (*addr)
-		: "cc", "1" );
+	asm volatile(
+		"	stsch	0(%2)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1), "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
 static inline int stsch_err(struct subchannel_id schid,
 				volatile struct schib *addr)
 {
-	int ccode;
+	register struct subchannel_id reg1 asm ("1") = schid;
+	int ccode = -EIO;
 
-	__asm__ __volatile__(
-		"    lhi  %0,%3\n"
-		"    lr	  1,%1\n"
-		"    stsch 0(%2)\n"
-		"0:  ipm  %0\n"
-		"    srl  %0,28\n"
+	asm volatile(
+		"	stsch	0(%2)\n"
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
 		"1:\n"
-#ifdef CONFIG_64BIT
-		".section __ex_table,\"a\"\n"
-		"   .align 8\n"
-		"   .quad 0b,1b\n"
-		".previous"
-#else
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-		"   .long 0b,1b\n"
-		".previous"
-#endif
-		: "=&d" (ccode)
-		: "d" (schid), "a" (addr), "K" (-EIO), "m" (*addr)
-		: "cc", "1" );
+		EX_TABLE(0b,1b)
+		: "+d" (ccode) : "d" (reg1), "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
 static inline int msch(struct subchannel_id schid,
 			   volatile struct schib *addr)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   msch  0(%2)\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid), "a" (addr), "m" (*addr)
-		: "cc", "1" );
+	asm volatile(
+		"	msch	0(%2)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1), "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
 static inline int msch_err(struct subchannel_id schid,
 			       volatile struct schib *addr)
 {
-	int ccode;
+	register struct subchannel_id reg1 asm ("1") = schid;
+	int ccode = -EIO;
 
-	__asm__ __volatile__(
-		"    lhi  %0,%3\n"
-		"    lr	  1,%1\n"
-		"    msch 0(%2)\n"
-		"0:  ipm  %0\n"
-		"    srl  %0,28\n"
+	asm volatile(
+		"	msch	0(%2)\n"
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
 		"1:\n"
-#ifdef CONFIG_64BIT
-		".section __ex_table,\"a\"\n"
-		"   .align 8\n"
-		"   .quad 0b,1b\n"
-		".previous"
-#else
-		".section __ex_table,\"a\"\n"
-		"   .align 4\n"
-		"   .long 0b,1b\n"
-		".previous"
-#endif
-		: "=&d" (ccode)
-		: "d" (schid), "a" (addr), "K" (-EIO), "m" (*addr)
-		: "cc", "1" );
+		EX_TABLE(0b,1b)
+		: "+d" (ccode) : "d" (reg1), "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
 static inline int tsch(struct subchannel_id schid,
 			   volatile struct irb *addr)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   tsch  0(%2)\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid), "a" (addr), "m" (*addr)
-		: "cc", "1" );
+	asm volatile(
+		"	tsch	0(%2)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1), "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
@@ -132,89 +100,77 @@ static inline int tpi( volatile struct tpi_info *addr)
 {
 	int ccode;
 
-	__asm__ __volatile__(
-		"   tpi	  0(%1)\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "a" (addr), "m" (*addr)
-		: "cc", "1" );
+	asm volatile(
+		"	tpi	0(%1)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
 static inline int ssch(struct subchannel_id schid,
 			   volatile struct orb *addr)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   ssch  0(%2)\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid), "a" (addr), "m" (*addr)
-		: "cc", "1" );
+	asm volatile(
+		"	ssch	0(%2)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1), "a" (addr), "m" (*addr) : "cc");
 	return ccode;
 }
 
 static inline int rsch(struct subchannel_id schid)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   rsch\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid)
-		: "cc", "1" );
+	asm volatile(
+		"	rsch\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1) : "cc");
 	return ccode;
 }
 
 static inline int csch(struct subchannel_id schid)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   csch\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid)
-		: "cc", "1" );
+	asm volatile(
+		"	csch\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1) : "cc");
 	return ccode;
 }
 
 static inline int hsch(struct subchannel_id schid)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   hsch\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid)
-		: "cc", "1" );
+	asm volatile(
+		"	hsch\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1) : "cc");
 	return ccode;
 }
 
 static inline int xsch(struct subchannel_id schid)
 {
+	register struct subchannel_id reg1 asm ("1") = schid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   .insn rre,0xb2760000,%1,0\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (schid)
-		: "cc", "1" );
+	asm volatile(
+		"	.insn	rre,0xb2760000,%1,0\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1) : "cc");
 	return ccode;
 }
 
@@ -223,41 +179,27 @@ static inline int chsc(void *chsc_area)
 	typedef struct { char _[4096]; } addr_type;
 	int cc;
 
-	__asm__ __volatile__ (
-		".insn	rre,0xb25f0000,%2,0	\n\t"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
+	asm volatile(
+		"	.insn	rre,0xb25f0000,%2,0\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
 		: "=d" (cc), "=m" (*(addr_type *) chsc_area)
 		: "d" (chsc_area), "m" (*(addr_type *) chsc_area)
-		: "cc" );
-
+		: "cc");
 	return cc;
 }
 
-static inline int iac( void)
-{
-	int ccode;
-
-	__asm__ __volatile__(
-		"   iac	  1\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode) : : "cc", "1" );
-	return ccode;
-}
-
 static inline int rchp(int chpid)
 {
+	register unsigned int reg1 asm ("1") = chpid;
 	int ccode;
 
-	__asm__ __volatile__(
-		"   lr	  1,%1\n"
-		"   rchp\n"
-		"   ipm	  %0\n"
-		"   srl	  %0,28"
-		: "=d" (ccode)
-		: "d" (chpid)
-		: "cc", "1" );
+	asm volatile(
+		"	lr	1,%1\n"
+		"	rchp\n"
+		"	ipm	%0\n"
+		"	srl	%0,28"
+		: "=d" (ccode) : "d" (reg1) : "cc");
 	return ccode;
 }
 
diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h
index 124569362f02..49bb9e371c32 100644
--- a/drivers/s390/cio/qdio.h
+++ b/drivers/s390/cio/qdio.h
@@ -274,12 +274,11 @@ do_sqbs(unsigned long sch, unsigned char state, int queue,
        register unsigned long _sch asm ("1") = sch;
        unsigned long _queuestart = ((unsigned long)queue << 32) | *start;
 
-       asm volatile (
-              " .insn rsy,0xeb000000008A,%1,0,0(%2)\n\t"
-              : "+d" (_ccq), "+d" (_queuestart)
-              : "d" ((unsigned long)state), "d" (_sch)
-              : "memory", "cc"
-       );
+       asm volatile(
+	       "	.insn	rsy,0xeb000000008A,%1,0,0(%2)"
+	       : "+d" (_ccq), "+d" (_queuestart)
+	       : "d" ((unsigned long)state), "d" (_sch)
+	       : "memory", "cc");
        *count = _ccq & 0xff;
        *start = _queuestart & 0xff;
 
@@ -299,12 +298,11 @@ do_eqbs(unsigned long sch, unsigned char *state, int queue,
 	unsigned long _queuestart = ((unsigned long)queue << 32) | *start;
 	unsigned long _state = 0;
 
-	asm volatile (
-	      " .insn rrf,0xB99c0000,%1,%2,0,0  \n\t"
-	      : "+d" (_ccq), "+d" (_queuestart), "+d" (_state)
-	      : "d" (_sch)
-	      : "memory", "cc"
-	);
+	asm volatile(
+		"	.insn	rrf,0xB99c0000,%1,%2,0,0"
+		: "+d" (_ccq), "+d" (_queuestart), "+d" (_state)
+		: "d" (_sch)
+		: "memory", "cc" );
 	*count = _ccq & 0xff;
 	*start = _queuestart & 0xff;
 	*state = _state & 0xff;
@@ -319,69 +317,35 @@ do_eqbs(unsigned long sch, unsigned char *state, int queue,
 static inline int
 do_siga_sync(struct subchannel_id schid, unsigned int mask1, unsigned int mask2)
 {
+	register unsigned long reg0 asm ("0") = 2;
+	register struct subchannel_id reg1 asm ("1") = schid;
+	register unsigned long reg2 asm ("2") = mask1;
+	register unsigned long reg3 asm ("3") = mask2;
 	int cc;
 
-#ifndef CONFIG_64BIT
-	asm volatile (
-		"lhi	0,2	\n\t"
-		"lr	1,%1	\n\t"
-		"lr	2,%2	\n\t"
-		"lr	3,%3	\n\t"
-		"siga   0	\n\t"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
+	asm volatile(
+		"	siga	0\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
 		: "=d" (cc)
-		: "d" (schid), "d" (mask1), "d" (mask2)
-		: "cc", "0", "1", "2", "3"
-		);
-#else /* CONFIG_64BIT */
-	asm volatile (
-		"lghi	0,2	\n\t"
-		"llgfr	1,%1	\n\t"
-		"llgfr	2,%2	\n\t"
-		"llgfr	3,%3	\n\t"
-		"siga   0	\n\t"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
-		: "=d" (cc)
-		: "d" (schid), "d" (mask1), "d" (mask2)
-		: "cc", "0", "1", "2", "3"
-		);
-#endif /* CONFIG_64BIT */
+		: "d" (reg0), "d" (reg1), "d" (reg2), "d" (reg3) : "cc");
 	return cc;
 }
 
 static inline int
 do_siga_input(struct subchannel_id schid, unsigned int mask)
 {
+	register unsigned long reg0 asm ("0") = 1;
+	register struct subchannel_id reg1 asm ("1") = schid;
+	register unsigned long reg2 asm ("2") = mask;
 	int cc;
 
-#ifndef CONFIG_64BIT
-	asm volatile (
-		"lhi	0,1	\n\t"
-		"lr	1,%1	\n\t"
-		"lr	2,%2	\n\t"
-		"siga   0	\n\t"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
-		: "=d" (cc)
-		: "d" (schid), "d" (mask)
-		: "cc", "0", "1", "2", "memory"
-		);
-#else /* CONFIG_64BIT */
-	asm volatile (
-		"lghi	0,1	\n\t"
-		"llgfr	1,%1	\n\t"
-		"llgfr	2,%2	\n\t"
-		"siga   0	\n\t"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
+	asm volatile(
+		"	siga	0\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
 		: "=d" (cc)
-		: "d" (schid), "d" (mask)
-		: "cc", "0", "1", "2", "memory"
-		);
-#endif /* CONFIG_64BIT */
-	
+		: "d" (reg0), "d" (reg1), "d" (reg2) : "cc", "memory");
 	return cc;
 }
 
@@ -389,93 +353,35 @@ static inline int
 do_siga_output(unsigned long schid, unsigned long mask, __u32 *bb,
 	       unsigned int fc)
 {
+	register unsigned long __fc asm("0") = fc;
+	register unsigned long __schid asm("1") = schid;
+	register unsigned long __mask asm("2") = mask;
 	int cc;
-	__u32 busy_bit;
-
-#ifndef CONFIG_64BIT
-	asm volatile (
-		"lhi	0,0	\n\t"
-		"lr	1,%2	\n\t"
-		"lr	2,%3	\n\t"
-		"siga	0	\n\t"
-		"0:"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
-		"srl	0,31	\n\t"
-		"lr	%1,0	\n\t"
-		"1:	\n\t"
-		".section .fixup,\"ax\"\n\t"
-		"2:	\n\t"
-		"lhi	%0,%4	\n\t"
-		"bras	1,3f	\n\t"
-		".long 1b	\n\t"
-		"3:	\n\t"
-		"l	1,0(1)	\n\t"
-		"br	1	\n\t"
-		".previous	\n\t"
-		".section __ex_table,\"a\"\n\t"
-		".align 4	\n\t"
-		".long	0b,2b	\n\t"
-		".previous	\n\t"
-		: "=d" (cc), "=d" (busy_bit)
-		: "d" (schid), "d" (mask),
-		"i" (QDIO_SIGA_ERROR_ACCESS_EXCEPTION)
-		: "cc", "0", "1", "2", "memory"
-		);
-#else /* CONFIG_64BIT */
-	asm volatile (
-        	"llgfr  0,%5    \n\t"
-                "lgr    1,%2    \n\t"
-		"llgfr	2,%3	\n\t"
-		"siga	0	\n\t"
-		"0:"
-		"ipm	%0	\n\t"
-		"srl	%0,28	\n\t"
-		"srl	0,31	\n\t"
-		"llgfr	%1,0	\n\t"
-		"1:	\n\t"
-		".section .fixup,\"ax\"\n\t"
-		"lghi	%0,%4	\n\t"
-		"jg	1b	\n\t"
-		".previous\n\t"
-		".section __ex_table,\"a\"\n\t"
-		".align 8	\n\t"
-		".quad	0b,1b	\n\t"
-		".previous	\n\t"
-		: "=d" (cc), "=d" (busy_bit)
-		: "d" (schid), "d" (mask),
-		"i" (QDIO_SIGA_ERROR_ACCESS_EXCEPTION), "d" (fc)
-		: "cc", "0", "1", "2", "memory"
-		);
-#endif /* CONFIG_64BIT */
-	
-	(*bb) = busy_bit;
+
+	asm volatile(
+		"	siga	0\n"
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "=d" (cc), "+d" (__fc), "+d" (__schid), "+d" (__mask)
+		: "0" (QDIO_SIGA_ERROR_ACCESS_EXCEPTION)
+		: "cc", "memory");
+	(*bb) = ((unsigned int) __fc) >> 31;
 	return cc;
 }
 
 static inline unsigned long
 do_clear_global_summary(void)
 {
-
-	unsigned long time;
-
-#ifndef CONFIG_64BIT
-	asm volatile (
-		"lhi	1,3	\n\t"
-		".insn	rre,0xb2650000,2,0	\n\t"
-		"lr	%0,3	\n\t"
-		: "=d" (time) : : "cc", "1", "2", "3"
-		);
-#else /* CONFIG_64BIT */
-	asm volatile (
-		"lghi	1,3	\n\t"
-		".insn	rre,0xb2650000,2,0	\n\t"
-		"lgr	%0,3	\n\t"
-		: "=d" (time) : : "cc", "1", "2", "3"
-		);
-#endif /* CONFIG_64BIT */
-	
-	return time;
+	register unsigned long __fn asm("1") = 3;
+	register unsigned long __tmp asm("2");
+	register unsigned long __time asm("3");
+
+	asm volatile(
+		"	.insn	rre,0xb2650000,2,0"
+		: "+d" (__fn), "=d" (__tmp), "=d" (__time));
+	return __time;
 }
 	
 /*
diff --git a/drivers/s390/net/iucv.c b/drivers/s390/net/iucv.c
index 821dde86e240..809dd8d7f47a 100644
--- a/drivers/s390/net/iucv.c
+++ b/drivers/s390/net/iucv.c
@@ -534,19 +534,15 @@ iucv_add_handler (handler *new)
  *
  * Returns: return code from CP's IUCV call
  */
-static __inline__ ulong
-b2f0(__u32 code, void *parm)
+static inline ulong b2f0(__u32 code, void *parm)
 {
+	register unsigned long reg0 asm ("0");
+	register unsigned long reg1 asm ("1");
 	iucv_dumpit("iparml before b2f0 call:", parm, sizeof(iucv_param));
 
-	asm volatile (
-		"LRA   1,0(%1)\n\t"
-		"LR    0,%0\n\t"
-		".long 0xb2f01000"
-		:
-		: "d" (code), "a" (parm)
-		: "0", "1"
-		);
+	reg0 = code;
+	reg1 = virt_to_phys(parm);
+	asm volatile(".long 0xb2f01000" : : "d" (reg0), "a" (reg1));
 
 	iucv_dumpit("iparml after b2f0 call:", parm, sizeof(iucv_param));
 
@@ -1248,6 +1244,8 @@ iucv_purge (__u16 pathid, __u32 msgid, __u32 srccls, __u32 *audit)
 static int
 iucv_query_generic(int want_maxconn)
 {
+	register unsigned long reg0 asm ("0");
+	register unsigned long reg1 asm ("1");
 	iparml_purge *parm = (iparml_purge *)grab_param();
 	int bufsize, maxconn;
 	int ccode;
@@ -1256,18 +1254,15 @@ iucv_query_generic(int want_maxconn)
 	 * Call b2f0 and store R0 (max buffer size),
 	 * R1 (max connections) and CC.
 	 */
-	asm volatile (
-		"LRA   1,0(%4)\n\t"
-		"LR    0,%3\n\t"
-		".long 0xb2f01000\n\t"
-		"IPM   %0\n\t"
-		"SRL   %0,28\n\t"
-		"ST    0,%1\n\t"
-		"ST    1,%2\n\t"
-		: "=d" (ccode), "=m" (bufsize), "=m" (maxconn)
-		: "d" (QUERY), "a" (parm)
-		: "0", "1", "cc"
-		);
+	reg0 = QUERY;
+	reg1 = virt_to_phys(parm);
+	asm volatile(
+		"	.long	0xb2f01000\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (ccode), "+d" (reg0), "+d" (reg1) : : "cc");
+	bufsize = reg0;
+	maxconn = reg1;
 	release_param(parm);
 
 	if (ccode)
diff --git a/drivers/s390/s390mach.c b/drivers/s390/s390mach.c
index a914129a4da9..479364d0332a 100644
--- a/drivers/s390/s390mach.c
+++ b/drivers/s390/s390mach.c
@@ -253,11 +253,12 @@ s390_revalidate_registers(struct mci *mci)
 		kill_task = 1;
 
 #ifndef CONFIG_64BIT
-	asm volatile("ld 0,0(%0)\n"
-		     "ld 2,8(%0)\n"
-		     "ld 4,16(%0)\n"
-		     "ld 6,24(%0)"
-		     : : "a" (&S390_lowcore.floating_pt_save_area));
+	asm volatile(
+		"	ld	0,0(%0)\n"
+		"	ld	2,8(%0)\n"
+		"	ld	4,16(%0)\n"
+		"	ld	6,24(%0)"
+		: : "a" (&S390_lowcore.floating_pt_save_area));
 #endif
 
 	if (MACHINE_HAS_IEEE) {
@@ -274,37 +275,36 @@ s390_revalidate_registers(struct mci *mci)
 			 * Floating point control register can't be restored.
 			 * Task will be terminated.
 			 */
-			asm volatile ("lfpc 0(%0)" : : "a" (&zero), "m" (zero));
+			asm volatile("lfpc 0(%0)" : : "a" (&zero), "m" (zero));
 			kill_task = 1;
 
-		}
-		else
-			asm volatile (
-				"lfpc 0(%0)"
-				: : "a" (fpt_creg_save_area));
-
-		asm volatile("ld  0,0(%0)\n"
-			     "ld  1,8(%0)\n"
-			     "ld  2,16(%0)\n"
-			     "ld  3,24(%0)\n"
-			     "ld  4,32(%0)\n"
-			     "ld  5,40(%0)\n"
-			     "ld  6,48(%0)\n"
-			     "ld  7,56(%0)\n"
-			     "ld  8,64(%0)\n"
-			     "ld  9,72(%0)\n"
-			     "ld 10,80(%0)\n"
-			     "ld 11,88(%0)\n"
-			     "ld 12,96(%0)\n"
-			     "ld 13,104(%0)\n"
-			     "ld 14,112(%0)\n"
-			     "ld 15,120(%0)\n"
-			     : : "a" (fpt_save_area));
+		} else
+			asm volatile("lfpc 0(%0)" : : "a" (fpt_creg_save_area));
+
+		asm volatile(
+			"	ld	0,0(%0)\n"
+			"	ld	1,8(%0)\n"
+			"	ld	2,16(%0)\n"
+			"	ld	3,24(%0)\n"
+			"	ld	4,32(%0)\n"
+			"	ld	5,40(%0)\n"
+			"	ld	6,48(%0)\n"
+			"	ld	7,56(%0)\n"
+			"	ld	8,64(%0)\n"
+			"	ld	9,72(%0)\n"
+			"	ld	10,80(%0)\n"
+			"	ld	11,88(%0)\n"
+			"	ld	12,96(%0)\n"
+			"	ld	13,104(%0)\n"
+			"	ld	14,112(%0)\n"
+			"	ld	15,120(%0)\n"
+			: : "a" (fpt_save_area));
 	}
 
 	/* Revalidate access registers */
-	asm volatile("lam 0,15,0(%0)"
-		     : : "a" (&S390_lowcore.access_regs_save_area));
+	asm volatile(
+		"	lam	0,15,0(%0)"
+		: : "a" (&S390_lowcore.access_regs_save_area));
 	if (!mci->ar)
 		/*
 		 * Access registers have unknown contents.
@@ -321,11 +321,13 @@ s390_revalidate_registers(struct mci *mci)
 		s390_handle_damage("invalid control registers.");
 	else
 #ifdef CONFIG_64BIT
-		asm volatile("lctlg 0,15,0(%0)"
-			     : : "a" (&S390_lowcore.cregs_save_area));
+		asm volatile(
+			"	lctlg	0,15,0(%0)"
+			: : "a" (&S390_lowcore.cregs_save_area));
 #else
-		asm volatile("lctl 0,15,0(%0)"
-			     : : "a" (&S390_lowcore.cregs_save_area));
+		asm volatile(
+			"	lctl	0,15,0(%0)"
+			: : "a" (&S390_lowcore.cregs_save_area));
 #endif
 
 	/*
@@ -339,20 +341,23 @@ s390_revalidate_registers(struct mci *mci)
 	 * old contents (should be zero) otherwise set it to zero.
 	 */
 	if (!mci->pr)
-		asm volatile("sr 0,0\n"
-			     "sckpf"
-			     : : : "0", "cc");
+		asm volatile(
+			"	sr	0,0\n"
+			"	sckpf"
+			: : : "0", "cc");
 	else
 		asm volatile(
-			"l 0,0(%0)\n"
-			"sckpf"
-			: : "a" (&S390_lowcore.tod_progreg_save_area) : "0", "cc");
+			"	l	0,0(%0)\n"
+			"	sckpf"
+			: : "a" (&S390_lowcore.tod_progreg_save_area)
+			: "0", "cc");
 #endif
 
 	/* Revalidate clock comparator register */
-	asm volatile ("stck 0(%1)\n"
-		      "sckc 0(%1)"
-		      : "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory");
+	asm volatile(
+		"	stck	0(%1)\n"
+		"	sckc	0(%1)"
+		: "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory");
 
 	/* Check if old PSW is valid */
 	if (!mci->wp)
diff --git a/include/asm-s390/appldata.h b/include/asm-s390/appldata.h
index b1770703b706..79283dac8281 100644
--- a/include/asm-s390/appldata.h
+++ b/include/asm-s390/appldata.h
@@ -80,7 +80,7 @@ static inline int appldata_asm(struct appldata_product_id *id,
 	parm_list.product_id_addr = (unsigned long) id;
 	parm_list.buffer_addr = virt_to_phys(buffer);
 	asm volatile(
-		"diag %1,%0,0xdc"
+		"	diag	%1,%0,0xdc"
 		: "=d" (ry)
 		: "d" (&parm_list), "m" (parm_list), "m" (*id)
 		: "cc");
diff --git a/include/asm-s390/atomic.h b/include/asm-s390/atomic.h
index 399bf02894dd..af20c7462485 100644
--- a/include/asm-s390/atomic.h
+++ b/include/asm-s390/atomic.h
@@ -30,20 +30,43 @@ typedef struct {
 
 #ifdef __KERNEL__
 
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+
 #define __CS_LOOP(ptr, op_val, op_string) ({				\
 	typeof(ptr->counter) old_val, new_val;				\
-        __asm__ __volatile__("   l     %0,0(%3)\n"			\
-                             "0: lr    %1,%0\n"				\
-                             op_string "  %1,%4\n"			\
-                             "   cs    %0,%1,0(%3)\n"			\
-                             "   jl    0b"				\
-                             : "=&d" (old_val), "=&d" (new_val),	\
-			       "=m" (((atomic_t *)(ptr))->counter)	\
-			     : "a" (ptr), "d" (op_val),			\
-			       "m" (((atomic_t *)(ptr))->counter)	\
-			     : "cc", "memory" );			\
+	asm volatile(							\
+		"	l	%0,%2\n"				\
+		"0:	lr	%1,%0\n"				\
+		op_string "	%1,%3\n"				\
+		"	cs	%0,%1,%2\n"				\
+		"	jl	0b"					\
+		: "=&d" (old_val), "=&d" (new_val),			\
+		  "=Q" (((atomic_t *)(ptr))->counter)			\
+		: "d" (op_val),	 "Q" (((atomic_t *)(ptr))->counter)	\
+		: "cc", "memory");					\
 	new_val;							\
 })
+
+#else /* __GNUC__ */
+
+#define __CS_LOOP(ptr, op_val, op_string) ({				\
+	typeof(ptr->counter) old_val, new_val;				\
+	asm volatile(							\
+		"	l	%0,0(%3)\n"				\
+		"0:	lr	%1,%0\n"				\
+		op_string "	%1,%4\n"				\
+		"	cs	%0,%1,0(%3)\n"				\
+		"	jl	0b"					\
+		: "=&d" (old_val), "=&d" (new_val),			\
+		  "=m" (((atomic_t *)(ptr))->counter)			\
+		: "a" (ptr), "d" (op_val),				\
+		  "m" (((atomic_t *)(ptr))->counter)			\
+		: "cc", "memory");					\
+	new_val;							\
+})
+
+#endif /* __GNUC__ */
+
 #define atomic_read(v)          ((v)->counter)
 #define atomic_set(v,i)         (((v)->counter) = (i))
 
@@ -81,10 +104,19 @@ static __inline__ void atomic_set_mask(unsigned long mask, atomic_t * v)
 
 static __inline__ int atomic_cmpxchg(atomic_t *v, int old, int new)
 {
-	__asm__ __volatile__("  cs   %0,%3,0(%2)\n"
-			     : "+d" (old), "=m" (v->counter)
-			     : "a" (v), "d" (new), "m" (v->counter)
-			     : "cc", "memory" );
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+	asm volatile(
+		"	cs	%0,%2,%1"
+		: "+d" (old), "=Q" (v->counter)
+		: "d" (new), "Q" (v->counter)
+		: "cc", "memory");
+#else /* __GNUC__ */
+	asm volatile(
+		"	cs	%0,%3,0(%2)"
+		: "+d" (old), "=m" (v->counter)
+		: "a" (v), "d" (new), "m" (v->counter)
+		: "cc", "memory");
+#endif /* __GNUC__ */
 	return old;
 }
 
@@ -113,20 +145,43 @@ typedef struct {
 } __attribute__ ((aligned (8))) atomic64_t;
 #define ATOMIC64_INIT(i)  { (i) }
 
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+
 #define __CSG_LOOP(ptr, op_val, op_string) ({				\
 	typeof(ptr->counter) old_val, new_val;				\
-        __asm__ __volatile__("   lg    %0,0(%3)\n"			\
-                             "0: lgr   %1,%0\n"				\
-                             op_string "  %1,%4\n"			\
-                             "   csg   %0,%1,0(%3)\n"			\
-                             "   jl    0b"				\
-                             : "=&d" (old_val), "=&d" (new_val),	\
-			       "=m" (((atomic_t *)(ptr))->counter)	\
-			     : "a" (ptr), "d" (op_val),			\
-			       "m" (((atomic_t *)(ptr))->counter)	\
-			     : "cc", "memory" );			\
+	asm volatile(							\
+		"	lg	%0,%2\n"				\
+		"0:	lgr	%1,%0\n"				\
+		op_string "	%1,%3\n"				\
+		"	csg	%0,%1,%2\n"				\
+		"	jl	0b"					\
+		: "=&d" (old_val), "=&d" (new_val),			\
+		  "=Q" (((atomic_t *)(ptr))->counter)			\
+		: "d" (op_val),	"Q" (((atomic_t *)(ptr))->counter)	\
+		: "cc", "memory" );					\
 	new_val;							\
 })
+
+#else /* __GNUC__ */
+
+#define __CSG_LOOP(ptr, op_val, op_string) ({				\
+	typeof(ptr->counter) old_val, new_val;				\
+	asm volatile(							\
+		"	lg	%0,0(%3)\n"				\
+		"0:	lgr	%1,%0\n"				\
+		op_string "	%1,%4\n"				\
+		"	csg	%0,%1,0(%3)\n"				\
+		"	jl	0b"					\
+		: "=&d" (old_val), "=&d" (new_val),			\
+		  "=m" (((atomic_t *)(ptr))->counter)			\
+		: "a" (ptr), "d" (op_val),				\
+		  "m" (((atomic_t *)(ptr))->counter)			\
+		: "cc", "memory" );					\
+	new_val;							\
+})
+
+#endif /* __GNUC__ */
+
 #define atomic64_read(v)          ((v)->counter)
 #define atomic64_set(v,i)         (((v)->counter) = (i))
 
@@ -163,10 +218,19 @@ static __inline__ void atomic64_set_mask(unsigned long mask, atomic64_t * v)
 static __inline__ long long atomic64_cmpxchg(atomic64_t *v,
 					     long long old, long long new)
 {
-	__asm__ __volatile__("  csg  %0,%3,0(%2)\n"
-			     : "+d" (old), "=m" (v->counter)
-			     : "a" (v), "d" (new), "m" (v->counter)
-			     : "cc", "memory" );
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+	asm volatile(
+		"	csg	%0,%2,%1"
+		: "+d" (old), "=Q" (v->counter)
+		: "d" (new), "Q" (v->counter)
+		: "cc", "memory");
+#else /* __GNUC__ */
+	asm volatile(
+		"	csg	%0,%3,0(%2)"
+		: "+d" (old), "=m" (v->counter)
+		: "a" (v), "d" (new), "m" (v->counter)
+		: "cc", "memory");
+#endif /* __GNUC__ */
 	return old;
 }
 
diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h
index 0ddcdba79e4a..f79c9b792af1 100644
--- a/include/asm-s390/bitops.h
+++ b/include/asm-s390/bitops.h
@@ -67,16 +67,35 @@ extern const char _sb_findmap[];
 #define __BITOPS_AND		"nr"
 #define __BITOPS_XOR		"xr"
 
-#define __BITOPS_LOOP(__old, __new, __addr, __val, __op_string)		\
-	__asm__ __volatile__("   l   %0,0(%4)\n"			\
-			     "0: lr  %1,%0\n"				\
-			     __op_string "  %1,%3\n"			\
-			     "   cs  %0,%1,0(%4)\n"			\
-			     "   jl  0b"				\
-			     : "=&d" (__old), "=&d" (__new),	       	\
-			       "=m" (*(unsigned long *) __addr)		\
-			     : "d" (__val), "a" (__addr),		\
-			       "m" (*(unsigned long *) __addr) : "cc" );
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+
+#define __BITOPS_LOOP(__old, __new, __addr, __val, __op_string)	\
+	asm volatile(						\
+		"	l	%0,%2\n"			\
+		"0:	lr	%1,%0\n"			\
+		__op_string "	%1,%3\n"			\
+		"	cs	%0,%1,%2\n"			\
+		"	jl	0b"				\
+		: "=&d" (__old), "=&d" (__new),			\
+		  "=Q" (*(unsigned long *) __addr)		\
+		: "d" (__val), "Q" (*(unsigned long *) __addr)	\
+		: "cc");
+
+#else /* __GNUC__ */
+
+#define __BITOPS_LOOP(__old, __new, __addr, __val, __op_string)	\
+	asm volatile(						\
+		"	l	%0,0(%4)\n"			\
+		"0:	lr	%1,%0\n"			\
+		__op_string "	%1,%3\n"			\
+		"	cs	%0,%1,0(%4)\n"			\
+		"	jl	0b"				\
+		: "=&d" (__old), "=&d" (__new),			\
+		  "=m" (*(unsigned long *) __addr)		\
+		: "d" (__val), "a" (__addr),			\
+		  "m" (*(unsigned long *) __addr) : "cc");
+
+#endif /* __GNUC__ */
 
 #else /* __s390x__ */
 
@@ -86,21 +105,41 @@ extern const char _sb_findmap[];
 #define __BITOPS_AND		"ngr"
 #define __BITOPS_XOR		"xgr"
 
-#define __BITOPS_LOOP(__old, __new, __addr, __val, __op_string)		\
-	__asm__ __volatile__("   lg  %0,0(%4)\n"			\
-			     "0: lgr %1,%0\n"				\
-			     __op_string "  %1,%3\n"			\
-			     "   csg %0,%1,0(%4)\n"			\
-			     "   jl  0b"				\
-			     : "=&d" (__old), "=&d" (__new),	       	\
-			       "=m" (*(unsigned long *) __addr)		\
-			     : "d" (__val), "a" (__addr),		\
-			       "m" (*(unsigned long *) __addr) : "cc" );
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+
+#define __BITOPS_LOOP(__old, __new, __addr, __val, __op_string)	\
+	asm volatile(						\
+		"	lg	%0,%2\n"			\
+		"0:	lgr	%1,%0\n"			\
+		__op_string "	%1,%3\n"			\
+		"	csg	%0,%1,%2\n"			\
+		"	jl	0b"				\
+		: "=&d" (__old), "=&d" (__new),			\
+		  "=Q" (*(unsigned long *) __addr)		\
+		: "d" (__val), "Q" (*(unsigned long *) __addr)	\
+		: "cc");
+
+#else /* __GNUC__ */
+
+#define __BITOPS_LOOP(__old, __new, __addr, __val, __op_string)	\
+	asm volatile(						\
+		"	lg	%0,0(%4)\n"			\
+		"0:	lgr	%1,%0\n"			\
+		__op_string "	%1,%3\n"			\
+		"	csg	%0,%1,0(%4)\n"			\
+		"	jl	0b"				\
+		: "=&d" (__old), "=&d" (__new),			\
+		  "=m" (*(unsigned long *) __addr)		\
+		: "d" (__val), "a" (__addr),			\
+		  "m" (*(unsigned long *) __addr) : "cc");
+
+
+#endif /* __GNUC__ */
 
 #endif /* __s390x__ */
 
 #define __BITOPS_WORDS(bits) (((bits)+__BITOPS_WORDSIZE-1)/__BITOPS_WORDSIZE)
-#define __BITOPS_BARRIER() __asm__ __volatile__ ( "" : : : "memory" )
+#define __BITOPS_BARRIER() asm volatile("" : : : "memory")
 
 #ifdef CONFIG_SMP
 /*
@@ -217,10 +256,10 @@ static inline void __set_bit(unsigned long nr, volatile unsigned long *ptr)
 	unsigned long addr;
 
 	addr = (unsigned long) ptr + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
-        asm volatile("oc 0(1,%1),0(%2)"
-		     : "=m" (*(char *) addr)
-		     : "a" (addr), "a" (_oi_bitmap + (nr & 7)),
-		       "m" (*(char *) addr) : "cc" );
+	asm volatile(
+		"	oc	0(1,%1),0(%2)"
+		: "=m" (*(char *) addr) : "a" (addr),
+		  "a" (_oi_bitmap + (nr & 7)), "m" (*(char *) addr) : "cc" );
 }
 
 static inline void 
@@ -229,40 +268,7 @@ __constant_set_bit(const unsigned long nr, volatile unsigned long *ptr)
 	unsigned long addr;
 
 	addr = ((unsigned long) ptr) + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
-	switch (nr&7) {
-	case 0:
-		asm volatile ("oi 0(%1),0x01" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 1:
-		asm volatile ("oi 0(%1),0x02" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 2:
-		asm volatile ("oi 0(%1),0x04" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 3:
-		asm volatile ("oi 0(%1),0x08" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 4:
-		asm volatile ("oi 0(%1),0x10" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 5:
-		asm volatile ("oi 0(%1),0x20" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 6:
-		asm volatile ("oi 0(%1),0x40" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 7:
-		asm volatile ("oi 0(%1),0x80" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	}
+	*(unsigned char *) addr |= 1 << (nr & 7);
 }
 
 #define set_bit_simple(nr,addr) \
@@ -279,10 +285,10 @@ __clear_bit(unsigned long nr, volatile unsigned long *ptr)
 	unsigned long addr;
 
 	addr = (unsigned long) ptr + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
-        asm volatile("nc 0(1,%1),0(%2)"
-		     : "=m" (*(char *) addr)
-		     : "a" (addr), "a" (_ni_bitmap + (nr & 7)),
-		       "m" (*(char *) addr) : "cc" );
+	asm volatile(
+		"	nc	0(1,%1),0(%2)"
+		: "=m" (*(char *) addr)	: "a" (addr),
+		  "a" (_ni_bitmap + (nr & 7)), "m" (*(char *) addr) : "cc");
 }
 
 static inline void 
@@ -291,40 +297,7 @@ __constant_clear_bit(const unsigned long nr, volatile unsigned long *ptr)
 	unsigned long addr;
 
 	addr = ((unsigned long) ptr) + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
-	switch (nr&7) {
-	case 0:
-		asm volatile ("ni 0(%1),0xFE" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 1:
-		asm volatile ("ni 0(%1),0xFD": "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 2:
-		asm volatile ("ni 0(%1),0xFB" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 3:
-		asm volatile ("ni 0(%1),0xF7" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 4:
-		asm volatile ("ni 0(%1),0xEF" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 5:
-		asm volatile ("ni 0(%1),0xDF" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 6:
-		asm volatile ("ni 0(%1),0xBF" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 7:
-		asm volatile ("ni 0(%1),0x7F" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	}
+	*(unsigned char *) addr &= ~(1 << (nr & 7));
 }
 
 #define clear_bit_simple(nr,addr) \
@@ -340,10 +313,10 @@ static inline void __change_bit(unsigned long nr, volatile unsigned long *ptr)
 	unsigned long addr;
 
 	addr = (unsigned long) ptr + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
-        asm volatile("xc 0(1,%1),0(%2)"
-		     :  "=m" (*(char *) addr)
-		     : "a" (addr), "a" (_oi_bitmap + (nr & 7)),
-		       "m" (*(char *) addr) : "cc" );
+	asm volatile(
+		"	xc	0(1,%1),0(%2)"
+		:  "=m" (*(char *) addr) : "a" (addr),
+		   "a" (_oi_bitmap + (nr & 7)), "m" (*(char *) addr) : "cc" );
 }
 
 static inline void 
@@ -352,40 +325,7 @@ __constant_change_bit(const unsigned long nr, volatile unsigned long *ptr)
 	unsigned long addr;
 
 	addr = ((unsigned long) ptr) + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
-	switch (nr&7) {
-	case 0:
-		asm volatile ("xi 0(%1),0x01" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 1:
-		asm volatile ("xi 0(%1),0x02" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 2:
-		asm volatile ("xi 0(%1),0x04" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 3:
-		asm volatile ("xi 0(%1),0x08" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 4:
-		asm volatile ("xi 0(%1),0x10" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 5:
-		asm volatile ("xi 0(%1),0x20" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 6:
-		asm volatile ("xi 0(%1),0x40" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	case 7:
-		asm volatile ("xi 0(%1),0x80" : "=m" (*(char *) addr)
-			      : "a" (addr), "m" (*(char *) addr) : "cc" );
-		break;
-	}
+	*(unsigned char *) addr ^= 1 << (nr & 7);
 }
 
 #define change_bit_simple(nr,addr) \
@@ -404,10 +344,11 @@ test_and_set_bit_simple(unsigned long nr, volatile unsigned long *ptr)
 
 	addr = (unsigned long) ptr + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
 	ch = *(unsigned char *) addr;
-        asm volatile("oc 0(1,%1),0(%2)"
-		     : "=m" (*(char *) addr)
-		     : "a" (addr), "a" (_oi_bitmap + (nr & 7)),
-		       "m" (*(char *) addr) : "cc", "memory" );
+	asm volatile(
+		"	oc	0(1,%1),0(%2)"
+		: "=m" (*(char *) addr)
+		: "a" (addr), "a" (_oi_bitmap + (nr & 7)),
+		  "m" (*(char *) addr) : "cc", "memory");
 	return (ch >> (nr & 7)) & 1;
 }
 #define __test_and_set_bit(X,Y)		test_and_set_bit_simple(X,Y)
@@ -423,10 +364,11 @@ test_and_clear_bit_simple(unsigned long nr, volatile unsigned long *ptr)
 
 	addr = (unsigned long) ptr + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
 	ch = *(unsigned char *) addr;
-        asm volatile("nc 0(1,%1),0(%2)"
-		     : "=m" (*(char *) addr)
-		     : "a" (addr), "a" (_ni_bitmap + (nr & 7)),
-		       "m" (*(char *) addr) : "cc", "memory" );
+	asm volatile(
+		"	nc	0(1,%1),0(%2)"
+		: "=m" (*(char *) addr)
+		: "a" (addr), "a" (_ni_bitmap + (nr & 7)),
+		  "m" (*(char *) addr) : "cc", "memory");
 	return (ch >> (nr & 7)) & 1;
 }
 #define __test_and_clear_bit(X,Y)	test_and_clear_bit_simple(X,Y)
@@ -442,10 +384,11 @@ test_and_change_bit_simple(unsigned long nr, volatile unsigned long *ptr)
 
 	addr = (unsigned long) ptr + ((nr ^ (__BITOPS_WORDSIZE - 8)) >> 3);
 	ch = *(unsigned char *) addr;
-        asm volatile("xc 0(1,%1),0(%2)"
-		     : "=m" (*(char *) addr)
-		     : "a" (addr), "a" (_oi_bitmap + (nr & 7)),
-		       "m" (*(char *) addr) : "cc", "memory" );
+	asm volatile(
+		"	xc	0(1,%1),0(%2)"
+		: "=m" (*(char *) addr)
+		: "a" (addr), "a" (_oi_bitmap + (nr & 7)),
+		  "m" (*(char *) addr) : "cc", "memory");
 	return (ch >> (nr & 7)) & 1;
 }
 #define __test_and_change_bit(X,Y)	test_and_change_bit_simple(X,Y)
@@ -557,35 +500,36 @@ find_first_zero_bit(const unsigned long * addr, unsigned long size)
 
         if (!size)
                 return 0;
-        __asm__("   lhi  %1,-1\n"
-                "   lr   %2,%3\n"
-                "   slr  %0,%0\n"
-                "   ahi  %2,31\n"
-                "   srl  %2,5\n"
-                "0: c    %1,0(%0,%4)\n"
-                "   jne  1f\n"
-                "   la   %0,4(%0)\n"
-                "   brct %2,0b\n"
-                "   lr   %0,%3\n"
-                "   j    4f\n"
-                "1: l    %2,0(%0,%4)\n"
-                "   sll  %0,3\n"
-                "   lhi  %1,0xff\n"
-                "   tml  %2,0xffff\n"
-                "   jno  2f\n"
-                "   ahi  %0,16\n"
-                "   srl  %2,16\n"
-                "2: tml  %2,0x00ff\n"
-                "   jno  3f\n"
-                "   ahi  %0,8\n"
-                "   srl  %2,8\n"
-                "3: nr   %2,%1\n"
-                "   ic   %2,0(%2,%5)\n"
-                "   alr  %0,%2\n"
-                "4:"
-                : "=&a" (res), "=&d" (cmp), "=&a" (count)
-                : "a" (size), "a" (addr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) addr) : "cc" );
+	asm volatile(
+		"	lhi	%1,-1\n"
+		"	lr	%2,%3\n"
+		"	slr	%0,%0\n"
+		"	ahi	%2,31\n"
+		"	srl	%2,5\n"
+		"0:	c	%1,0(%0,%4)\n"
+		"	jne	1f\n"
+		"	la	%0,4(%0)\n"
+		"	brct	%2,0b\n"
+		"	lr	%0,%3\n"
+		"	j	4f\n"
+		"1:	l	%2,0(%0,%4)\n"
+		"	sll	%0,3\n"
+		"	lhi	%1,0xff\n"
+		"	tml	%2,0xffff\n"
+		"	jno	2f\n"
+		"	ahi	%0,16\n"
+		"	srl	%2,16\n"
+		"2:	tml	%2,0x00ff\n"
+		"	jno	3f\n"
+		"	ahi	%0,8\n"
+		"	srl	%2,8\n"
+		"3:	nr	%2,%1\n"
+		"	ic	%2,0(%2,%5)\n"
+		"	alr	%0,%2\n"
+		"4:"
+		: "=&a" (res), "=&d" (cmp), "=&a" (count)
+		: "a" (size), "a" (addr), "a" (&_zb_findmap),
+		  "m" (*(addrtype *) addr) : "cc");
         return (res < size) ? res : size;
 }
 
@@ -598,35 +542,36 @@ find_first_bit(const unsigned long * addr, unsigned long size)
 
         if (!size)
                 return 0;
-        __asm__("   slr  %1,%1\n"
-                "   lr   %2,%3\n"
-                "   slr  %0,%0\n"
-                "   ahi  %2,31\n"
-                "   srl  %2,5\n"
-                "0: c    %1,0(%0,%4)\n"
-                "   jne  1f\n"
-                "   la   %0,4(%0)\n"
-                "   brct %2,0b\n"
-                "   lr   %0,%3\n"
-                "   j    4f\n"
-                "1: l    %2,0(%0,%4)\n"
-                "   sll  %0,3\n"
-                "   lhi  %1,0xff\n"
-                "   tml  %2,0xffff\n"
-                "   jnz  2f\n"
-                "   ahi  %0,16\n"
-                "   srl  %2,16\n"
-                "2: tml  %2,0x00ff\n"
-                "   jnz  3f\n"
-                "   ahi  %0,8\n"
-                "   srl  %2,8\n"
-                "3: nr   %2,%1\n"
-                "   ic   %2,0(%2,%5)\n"
-                "   alr  %0,%2\n"
-                "4:"
-                : "=&a" (res), "=&d" (cmp), "=&a" (count)
-                : "a" (size), "a" (addr), "a" (&_sb_findmap),
-		  "m" (*(addrtype *) addr) : "cc" );
+	asm volatile(
+		"	slr	%1,%1\n"
+		"	lr	%2,%3\n"
+		"	slr	%0,%0\n"
+		"	ahi	%2,31\n"
+		"	srl	%2,5\n"
+		"0:	c	%1,0(%0,%4)\n"
+		"	jne	1f\n"
+		"	la	%0,4(%0)\n"
+		"	brct	%2,0b\n"
+		"	lr	%0,%3\n"
+		"	j	4f\n"
+		"1:	l	%2,0(%0,%4)\n"
+		"	sll	%0,3\n"
+		"	lhi	%1,0xff\n"
+		"	tml	%2,0xffff\n"
+		"	jnz	2f\n"
+		"	ahi	%0,16\n"
+		"	srl	%2,16\n"
+		"2:	tml	%2,0x00ff\n"
+		"	jnz	3f\n"
+		"	ahi	%0,8\n"
+		"	srl	%2,8\n"
+		"3:	nr	%2,%1\n"
+		"	ic	%2,0(%2,%5)\n"
+		"	alr	%0,%2\n"
+		"4:"
+		: "=&a" (res), "=&d" (cmp), "=&a" (count)
+		: "a" (size), "a" (addr), "a" (&_sb_findmap),
+		  "m" (*(addrtype *) addr) : "cc");
         return (res < size) ? res : size;
 }
 
@@ -640,39 +585,40 @@ find_first_zero_bit(const unsigned long * addr, unsigned long size)
 
         if (!size)
                 return 0;
-        __asm__("   lghi  %1,-1\n"
-                "   lgr   %2,%3\n"
-                "   slgr  %0,%0\n"
-                "   aghi  %2,63\n"
-                "   srlg  %2,%2,6\n"
-                "0: cg    %1,0(%0,%4)\n"
-                "   jne   1f\n"
-                "   la    %0,8(%0)\n"
-                "   brct  %2,0b\n"
-                "   lgr   %0,%3\n"
-                "   j     5f\n"
-                "1: lg    %2,0(%0,%4)\n"
-                "   sllg  %0,%0,3\n"
-                "   clr   %2,%1\n"
-		"   jne   2f\n"
-		"   aghi  %0,32\n"
-                "   srlg  %2,%2,32\n"
-		"2: lghi  %1,0xff\n"
-                "   tmll  %2,0xffff\n"
-                "   jno   3f\n"
-                "   aghi  %0,16\n"
-                "   srl   %2,16\n"
-                "3: tmll  %2,0x00ff\n"
-                "   jno   4f\n"
-                "   aghi  %0,8\n"
-                "   srl   %2,8\n"
-                "4: ngr   %2,%1\n"
-                "   ic    %2,0(%2,%5)\n"
-                "   algr  %0,%2\n"
-                "5:"
-                : "=&a" (res), "=&d" (cmp), "=&a" (count)
+	asm volatile(
+		"	lghi	%1,-1\n"
+		"	lgr	%2,%3\n"
+		"	slgr	%0,%0\n"
+		"	aghi	%2,63\n"
+		"	srlg	%2,%2,6\n"
+		"0:	cg	%1,0(%0,%4)\n"
+		"	jne	1f\n"
+		"	la	%0,8(%0)\n"
+		"	brct	%2,0b\n"
+		"	lgr	%0,%3\n"
+		"	j	5f\n"
+		"1:	lg	%2,0(%0,%4)\n"
+		"	sllg	%0,%0,3\n"
+		"	clr	%2,%1\n"
+		"	jne	2f\n"
+		"	aghi	%0,32\n"
+		"	srlg	%2,%2,32\n"
+		"2:	lghi	%1,0xff\n"
+		"	tmll	%2,0xffff\n"
+		"	jno	3f\n"
+		"	aghi	%0,16\n"
+		"	srl	%2,16\n"
+		"3:	tmll	%2,0x00ff\n"
+		"	jno	4f\n"
+		"	aghi	%0,8\n"
+		"	srl	%2,8\n"
+		"4:	ngr	%2,%1\n"
+		"	ic	%2,0(%2,%5)\n"
+		"	algr	%0,%2\n"
+		"5:"
+		: "=&a" (res), "=&d" (cmp), "=&a" (count)
 		: "a" (size), "a" (addr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) addr) : "cc" );
+		  "m" (*(addrtype *) addr) : "cc");
         return (res < size) ? res : size;
 }
 
@@ -684,39 +630,40 @@ find_first_bit(const unsigned long * addr, unsigned long size)
 
         if (!size)
                 return 0;
-        __asm__("   slgr  %1,%1\n"
-                "   lgr   %2,%3\n"
-                "   slgr  %0,%0\n"
-                "   aghi  %2,63\n"
-                "   srlg  %2,%2,6\n"
-                "0: cg    %1,0(%0,%4)\n"
-                "   jne   1f\n"
-                "   aghi  %0,8\n"
-                "   brct  %2,0b\n"
-                "   lgr   %0,%3\n"
-                "   j     5f\n"
-                "1: lg    %2,0(%0,%4)\n"
-                "   sllg  %0,%0,3\n"
-                "   clr   %2,%1\n"
-		"   jne   2f\n"
-		"   aghi  %0,32\n"
-                "   srlg  %2,%2,32\n"
-		"2: lghi  %1,0xff\n"
-                "   tmll  %2,0xffff\n"
-                "   jnz   3f\n"
-                "   aghi  %0,16\n"
-                "   srl   %2,16\n"
-                "3: tmll  %2,0x00ff\n"
-                "   jnz   4f\n"
-                "   aghi  %0,8\n"
-                "   srl   %2,8\n"
-                "4: ngr   %2,%1\n"
-                "   ic    %2,0(%2,%5)\n"
-                "   algr  %0,%2\n"
-                "5:"
-                : "=&a" (res), "=&d" (cmp), "=&a" (count)
+	asm volatile(
+		"	slgr	%1,%1\n"
+		"	lgr	%2,%3\n"
+		"	slgr	%0,%0\n"
+		"	aghi	%2,63\n"
+		"	srlg	%2,%2,6\n"
+		"0:	cg	%1,0(%0,%4)\n"
+		"	jne	1f\n"
+		"	aghi	%0,8\n"
+		"	brct	%2,0b\n"
+		"	lgr	%0,%3\n"
+		"	j	5f\n"
+		"1:	lg	%2,0(%0,%4)\n"
+		"	sllg	%0,%0,3\n"
+		"	clr	%2,%1\n"
+		"	jne	2f\n"
+		"	aghi	%0,32\n"
+		"	srlg	%2,%2,32\n"
+		"2:	lghi	%1,0xff\n"
+		"	tmll	%2,0xffff\n"
+		"	jnz	3f\n"
+		"	aghi	%0,16\n"
+		"	srl	%2,16\n"
+		"3:	tmll	%2,0x00ff\n"
+		"	jnz	4f\n"
+		"	aghi	%0,8\n"
+		"	srl	%2,8\n"
+		"4:	ngr	%2,%1\n"
+		"	ic	%2,0(%2,%5)\n"
+		"	algr	%0,%2\n"
+		"5:"
+		: "=&a" (res), "=&d" (cmp), "=&a" (count)
 		: "a" (size), "a" (addr), "a" (&_sb_findmap),
-		  "m" (*(addrtype *) addr) : "cc" );
+		  "m" (*(addrtype *) addr) : "cc");
         return (res < size) ? res : size;
 }
 
@@ -832,36 +779,37 @@ ext2_find_first_zero_bit(void *vaddr, unsigned int size)
 
         if (!size)
                 return 0;
-        __asm__("   lhi  %1,-1\n"
-                "   lr   %2,%3\n"
-                "   ahi  %2,31\n"
-                "   srl  %2,5\n"
-                "   slr  %0,%0\n"
-                "0: cl   %1,0(%0,%4)\n"
-                "   jne  1f\n"
-                "   ahi  %0,4\n"
-                "   brct %2,0b\n"
-                "   lr   %0,%3\n"
-                "   j    4f\n"
-                "1: l    %2,0(%0,%4)\n"
-                "   sll  %0,3\n"
-                "   ahi  %0,24\n"
-                "   lhi  %1,0xff\n"
-                "   tmh  %2,0xffff\n"
-                "   jo   2f\n"
-                "   ahi  %0,-16\n"
-                "   srl  %2,16\n"
-                "2: tml  %2,0xff00\n"
-                "   jo   3f\n"
-                "   ahi  %0,-8\n"
-                "   srl  %2,8\n"
-                "3: nr   %2,%1\n"
-                "   ic   %2,0(%2,%5)\n"
-                "   alr  %0,%2\n"
-                "4:"
-                : "=&a" (res), "=&d" (cmp), "=&a" (count)
-                : "a" (size), "a" (vaddr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) vaddr) : "cc" );
+	asm volatile(
+		"	lhi	%1,-1\n"
+		"	lr	%2,%3\n"
+		"	ahi	%2,31\n"
+		"	srl	%2,5\n"
+		"	slr	%0,%0\n"
+		"0:	cl	%1,0(%0,%4)\n"
+		"	jne	1f\n"
+		"	ahi	%0,4\n"
+		"	brct	%2,0b\n"
+		"	lr	%0,%3\n"
+		"	j	4f\n"
+		"1:	l	%2,0(%0,%4)\n"
+		"	sll	%0,3\n"
+		"	ahi	%0,24\n"
+		"	lhi	%1,0xff\n"
+		"	tmh	%2,0xffff\n"
+		"	jo	2f\n"
+		"	ahi	%0,-16\n"
+		"	srl	%2,16\n"
+		"2:	tml	%2,0xff00\n"
+		"	jo	3f\n"
+		"	ahi	%0,-8\n"
+		"	srl	%2,8\n"
+		"3:	nr	%2,%1\n"
+		"	ic	%2,0(%2,%5)\n"
+		"	alr	%0,%2\n"
+		"4:"
+		: "=&a" (res), "=&d" (cmp), "=&a" (count)
+		: "a" (size), "a" (vaddr), "a" (&_zb_findmap),
+		  "m" (*(addrtype *) vaddr) : "cc");
         return (res < size) ? res : size;
 }
 
@@ -875,39 +823,40 @@ ext2_find_first_zero_bit(void *vaddr, unsigned long size)
 
         if (!size)
                 return 0;
-        __asm__("   lghi  %1,-1\n"
-                "   lgr   %2,%3\n"
-                "   aghi  %2,63\n"
-                "   srlg  %2,%2,6\n"
-                "   slgr  %0,%0\n"
-                "0: clg   %1,0(%0,%4)\n"
-                "   jne   1f\n"
-                "   aghi  %0,8\n"
-                "   brct  %2,0b\n"
-                "   lgr   %0,%3\n"
-                "   j     5f\n"
-                "1: cl    %1,0(%0,%4)\n"
-		"   jne   2f\n"
-		"   aghi  %0,4\n"
-		"2: l     %2,0(%0,%4)\n"
-                "   sllg  %0,%0,3\n"
-                "   aghi  %0,24\n"
-                "   lghi  %1,0xff\n"
-                "   tmlh  %2,0xffff\n"
-                "   jo    3f\n"
-                "   aghi  %0,-16\n"
-                "   srl   %2,16\n"
-                "3: tmll  %2,0xff00\n"
-                "   jo    4f\n"
-                "   aghi  %0,-8\n"
-                "   srl   %2,8\n"
-                "4: ngr   %2,%1\n"
-                "   ic    %2,0(%2,%5)\n"
-                "   algr  %0,%2\n"
-                "5:"
-                : "=&a" (res), "=&d" (cmp), "=&a" (count)
+	asm volatile(
+		"	lghi	%1,-1\n"
+		"	lgr	%2,%3\n"
+		"	aghi	%2,63\n"
+		"	srlg	%2,%2,6\n"
+		"	slgr	%0,%0\n"
+		"0:	clg	%1,0(%0,%4)\n"
+		"	jne	1f\n"
+		"	aghi	%0,8\n"
+		"	brct	%2,0b\n"
+		"	lgr	%0,%3\n"
+		"	j	5f\n"
+		"1:	cl	%1,0(%0,%4)\n"
+		"	jne	2f\n"
+		"	aghi	%0,4\n"
+		"2:	l	%2,0(%0,%4)\n"
+		"	sllg	%0,%0,3\n"
+		"	aghi	%0,24\n"
+		"	lghi	%1,0xff\n"
+		"	tmlh	%2,0xffff\n"
+		"	jo	3f\n"
+		"	aghi	%0,-16\n"
+		"	srl	%2,16\n"
+		"3:	tmll	%2,0xff00\n"
+		"	jo	4f\n"
+		"	aghi	%0,-8\n"
+		"	srl	%2,8\n"
+		"4:	ngr	%2,%1\n"
+		"	ic	%2,0(%2,%5)\n"
+		"	algr	%0,%2\n"
+		"5:"
+		: "=&a" (res), "=&d" (cmp), "=&a" (count)
 		: "a" (size), "a" (vaddr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) vaddr) : "cc" );
+		  "m" (*(addrtype *) vaddr) : "cc");
         return (res < size) ? res : size;
 }
 
@@ -927,13 +876,16 @@ ext2_find_next_zero_bit(void *vaddr, unsigned long size, unsigned long offset)
 	p = addr + offset / __BITOPS_WORDSIZE;
         if (bit) {
 #ifndef __s390x__
-                asm("   ic   %0,0(%1)\n"
-		    "   icm  %0,2,1(%1)\n"
-		    "   icm  %0,4,2(%1)\n"
-		    "   icm  %0,8,3(%1)"
-		    : "=&a" (word) : "a" (p), "m" (*p) : "cc" );
+		asm volatile(
+			"	ic	%0,0(%1)\n"
+			"	icm	%0,2,1(%1)\n"
+			"	icm	%0,4,2(%1)\n"
+			"	icm	%0,8,3(%1)"
+			: "=&a" (word) : "a" (p), "m" (*p) : "cc");
 #else
-                asm("   lrvg %0,%1" : "=a" (word) : "m" (*p) );
+		asm volatile(
+			"	lrvg	%0,%1"
+			: "=a" (word) : "m" (*p) );
 #endif
 		/*
 		 * s390 version of ffz returns __BITOPS_WORDSIZE
diff --git a/include/asm-s390/byteorder.h b/include/asm-s390/byteorder.h
index 2cc35a0e188e..1fe2492baa8d 100644
--- a/include/asm-s390/byteorder.h
+++ b/include/asm-s390/byteorder.h
@@ -14,60 +14,54 @@
 #ifdef __GNUC__
 
 #ifdef __s390x__
-static __inline__ __u64 ___arch__swab64p(const __u64 *x)
+static inline __u64 ___arch__swab64p(const __u64 *x)
 {
 	__u64 result;
 
-	__asm__ __volatile__ (
-		"   lrvg %0,%1"
-		: "=d" (result) : "m" (*x) );
+	asm volatile("lrvg %0,%1" : "=d" (result) : "m" (*x));
 	return result;
 }
 
-static __inline__ __u64 ___arch__swab64(__u64 x)
+static inline __u64 ___arch__swab64(__u64 x)
 {
 	__u64 result;
 
-	__asm__ __volatile__ (
-		"   lrvgr %0,%1"
-		: "=d" (result) : "d" (x) );
+	asm volatile("lrvgr %0,%1" : "=d" (result) : "d" (x));
 	return result;
 }
 
-static __inline__ void ___arch__swab64s(__u64 *x)
+static inline void ___arch__swab64s(__u64 *x)
 {
 	*x = ___arch__swab64p(x);
 }
 #endif /* __s390x__ */
 
-static __inline__ __u32 ___arch__swab32p(const __u32 *x)
+static inline __u32 ___arch__swab32p(const __u32 *x)
 {
 	__u32 result;
 	
-	__asm__ __volatile__ (
+	asm volatile(
 #ifndef __s390x__
-		"        icm   %0,8,3(%1)\n"
-		"        icm   %0,4,2(%1)\n"
-		"        icm   %0,2,1(%1)\n"
-		"        ic    %0,0(%1)"
-		: "=&d" (result) : "a" (x), "m" (*x) : "cc" );
+		"	icm	%0,8,3(%1)\n"
+		"	icm	%0,4,2(%1)\n"
+		"	icm	%0,2,1(%1)\n"
+		"	ic	%0,0(%1)"
+		: "=&d" (result) : "a" (x), "m" (*x) : "cc");
 #else /* __s390x__ */
-		"   lrv  %0,%1"
-		: "=d" (result) : "m" (*x) );
+		"	lrv	%0,%1"
+		: "=d" (result) : "m" (*x));
 #endif /* __s390x__ */
 	return result;
 }
 
-static __inline__ __u32 ___arch__swab32(__u32 x)
+static inline __u32 ___arch__swab32(__u32 x)
 {
 #ifndef __s390x__
 	return ___arch__swab32p(&x);
 #else /* __s390x__ */
 	__u32 result;
 	
-	__asm__ __volatile__ (
-		"   lrvr  %0,%1"
-		: "=d" (result) : "d" (x) );
+	asm volatile("lrvr  %0,%1" : "=d" (result) : "d" (x));
 	return result;
 #endif /* __s390x__ */
 }
@@ -81,14 +75,14 @@ static __inline__ __u16 ___arch__swab16p(const __u16 *x)
 {
 	__u16 result;
 	
-	__asm__ __volatile__ (
+	asm volatile(
 #ifndef __s390x__
-		"        icm   %0,2,1(%1)\n"
-		"        ic    %0,0(%1)\n"
-		: "=&d" (result) : "a" (x), "m" (*x) : "cc" );
+		"	icm	%0,2,1(%1)\n"
+		"	ic	%0,0(%1)\n"
+		: "=&d" (result) : "a" (x), "m" (*x) : "cc");
 #else /* __s390x__ */
-		"   lrvh %0,%1"
-		: "=d" (result) : "m" (*x) );
+		"	lrvh	%0,%1"
+		: "=d" (result) : "m" (*x));
 #endif /* __s390x__ */
 	return result;
 }
diff --git a/include/asm-s390/checksum.h b/include/asm-s390/checksum.h
index 471f2af2b16a..37c362d89fad 100644
--- a/include/asm-s390/checksum.h
+++ b/include/asm-s390/checksum.h
@@ -30,57 +30,13 @@
 static inline unsigned int
 csum_partial(const unsigned char * buff, int len, unsigned int sum)
 {
-	/*
-	 * Experiments with ethernet and slip connections show that buf
-	 * is aligned on either a 2-byte or 4-byte boundary.
-	 */
-#ifndef __s390x__
-	register_pair rp;
-
-	rp.subreg.even = (unsigned long) buff;
-	rp.subreg.odd = (unsigned long) len;
-	__asm__ __volatile__ (
-		"0:  cksm %0,%1\n"	/* do checksum on longs */
-		"    jo   0b\n"
-		: "+&d" (sum), "+&a" (rp) : : "cc", "memory" );
-#else /* __s390x__ */
-        __asm__ __volatile__ (
-                "    lgr  2,%1\n"    /* address in gpr 2 */
-                "    lgfr 3,%2\n"    /* length in gpr 3 */
-                "0:  cksm %0,2\n"    /* do checksum on longs */
-                "    jo   0b\n"
-                : "+&d" (sum)
-                : "d" (buff), "d" (len)
-                : "cc", "memory", "2", "3" );
-#endif /* __s390x__ */
-	return sum;
-}
-
-/*
- * csum_partial as an inline function
- */
-static inline unsigned int 
-csum_partial_inline(const unsigned char * buff, int len, unsigned int sum)
-{
-#ifndef __s390x__
-	register_pair rp;
+	register unsigned long reg2 asm("2") = (unsigned long) buff;
+	register unsigned long reg3 asm("3") = (unsigned long) len;
 
-	rp.subreg.even = (unsigned long) buff;
-	rp.subreg.odd = (unsigned long) len;
-	__asm__ __volatile__ (
-		"0:  cksm %0,%1\n"    /* do checksum on longs */
-		"    jo   0b\n"
-                : "+&d" (sum), "+&a" (rp) : : "cc", "memory" );
-#else /* __s390x__ */
-	__asm__ __volatile__ (
-		"    lgr  2,%1\n"    /* address in gpr 2 */
-		"    lgfr 3,%2\n"    /* length in gpr 3 */
-		"0:  cksm %0,2\n"    /* do checksum on longs */
-		"    jo   0b\n"
-                : "+&d" (sum)
-		: "d" (buff), "d" (len)
-                : "cc", "memory", "2", "3" );
-#endif /* __s390x__ */
+	asm volatile(
+		"0:	cksm	%0,%1\n"	/* do checksum on longs */
+		"	jo	0b\n"
+		: "+d" (sum), "+d" (reg2), "+d" (reg3) : : "cc", "memory");
 	return sum;
 }
 
@@ -114,7 +70,7 @@ static inline unsigned int
 csum_partial_copy_nocheck (const char *src, char *dst, int len, unsigned int sum)
 {
         memcpy(dst,src,len);
-        return csum_partial_inline(dst, len, sum);
+	return csum_partial(dst, len, sum);
 }
 
 /*
@@ -126,22 +82,22 @@ csum_fold(unsigned int sum)
 #ifndef __s390x__
 	register_pair rp;
 
-	__asm__ __volatile__ (
-		"    slr  %N1,%N1\n" /* %0 = H L */
-		"    lr   %1,%0\n"   /* %0 = H L, %1 = H L 0 0 */
-		"    srdl %1,16\n"   /* %0 = H L, %1 = 0 H L 0 */
-		"    alr  %1,%N1\n"  /* %0 = H L, %1 = L H L 0 */
-		"    alr  %0,%1\n"   /* %0 = H+L+C L+H */
-		"    srl  %0,16\n"   /* %0 = H+L+C */
-		: "+&d" (sum), "=d" (rp) : : "cc" );
+	asm volatile(
+		"	slr	%N1,%N1\n"	/* %0 = H L */
+		"	lr	%1,%0\n"	/* %0 = H L, %1 = H L 0 0 */
+		"	srdl	%1,16\n"	/* %0 = H L, %1 = 0 H L 0 */
+		"	alr	%1,%N1\n"	/* %0 = H L, %1 = L H L 0 */
+		"	alr	%0,%1\n"	/* %0 = H+L+C L+H */
+		"	srl	%0,16\n"	/* %0 = H+L+C */
+		: "+&d" (sum), "=d" (rp) : : "cc");
 #else /* __s390x__ */
-	__asm__ __volatile__ (
-		"    sr   3,3\n"   /* %0 = H*65536 + L */
-		"    lr   2,%0\n"  /* %0 = H L, R2/R3 = H L / 0 0 */
-		"    srdl 2,16\n"  /* %0 = H L, R2/R3 = 0 H / L 0 */
-		"    alr  2,3\n"   /* %0 = H L, R2/R3 = L H / L 0 */
-		"    alr  %0,2\n"  /* %0 = H+L+C L+H */
-                "    srl  %0,16\n" /* %0 = H+L+C */
+	asm volatile(
+		"	sr	3,3\n"		/* %0 = H*65536 + L */
+		"	lr	2,%0\n"		/* %0 = H L, 2/3 = H L / 0 0 */
+		"	srdl	2,16\n"		/* %0 = H L, 2/3 = 0 H / L 0 */
+		"	alr	2,3\n"		/* %0 = H L, 2/3 = L H / L 0 */
+		"	alr	%0,2\n"		/* %0 = H+L+C L+H */
+		"	srl	%0,16\n"	/* %0 = H+L+C */
 		: "+&d" (sum) : : "cc", "2", "3");
 #endif /* __s390x__ */
 	return ((unsigned short) ~sum);
@@ -155,29 +111,7 @@ csum_fold(unsigned int sum)
 static inline unsigned short
 ip_fast_csum(unsigned char *iph, unsigned int ihl)
 {
-	unsigned long sum;
-#ifndef __s390x__
-	register_pair rp;
-
-	rp.subreg.even = (unsigned long) iph;
-	rp.subreg.odd = (unsigned long) ihl*4;
-        __asm__ __volatile__ (
-		"    sr   %0,%0\n"   /* set sum to zero */
-                "0:  cksm %0,%1\n"   /* do checksum on longs */
-                "    jo   0b\n"
-                : "=&d" (sum), "+&a" (rp) : : "cc", "memory" );
-#else /* __s390x__ */
-        __asm__ __volatile__ (
-		"    slgr %0,%0\n"   /* set sum to zero */
-                "    lgr  2,%1\n"    /* address in gpr 2 */
-                "    lgfr 3,%2\n"    /* length in gpr 3 */
-                "0:  cksm %0,2\n"    /* do checksum on ints */
-                "    jo   0b\n"
-                : "=&d" (sum)
-                : "d" (iph), "d" (ihl*4)
-                : "cc", "memory", "2", "3" );
-#endif /* __s390x__ */
-        return csum_fold(sum);
+	return csum_fold(csum_partial(iph, ihl*4, 0));
 }
 
 /*
@@ -190,47 +124,47 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
                    unsigned int sum)
 {
 #ifndef __s390x__
-	__asm__ __volatile__ (
-                "    alr   %0,%1\n"  /* sum += saddr */
-                "    brc   12,0f\n"
-		"    ahi   %0,1\n"   /* add carry */
+	asm volatile(
+		"	alr	%0,%1\n" /* sum += saddr */
+		"	brc	12,0f\n"
+		"	ahi	%0,1\n"  /* add carry */
 		"0:"
-		: "+&d" (sum) : "d" (saddr) : "cc" );
-	__asm__ __volatile__ (
-                "    alr   %0,%1\n"  /* sum += daddr */
-                "    brc   12,1f\n"
-                "    ahi   %0,1\n"   /* add carry */
+		: "+&d" (sum) : "d" (saddr) : "cc");
+	asm volatile(
+		"	alr	%0,%1\n" /* sum += daddr */
+		"	brc	12,1f\n"
+		"	ahi	%0,1\n"  /* add carry */
 		"1:"
-		: "+&d" (sum) : "d" (daddr) : "cc" );
-	__asm__ __volatile__ (
-                "    alr   %0,%1\n"  /* sum += (len<<16) + (proto<<8) */
-		"    brc   12,2f\n"
-		"    ahi   %0,1\n"   /* add carry */
+		: "+&d" (sum) : "d" (daddr) : "cc");
+	asm volatile(
+		"	alr	%0,%1\n" /* sum += (len<<16) + (proto<<8) */
+		"	brc	12,2f\n"
+		"	ahi	%0,1\n"  /* add carry */
 		"2:"
 		: "+&d" (sum)
 		: "d" (((unsigned int) len<<16) + (unsigned int) proto)
-		: "cc" );
+		: "cc");
 #else /* __s390x__ */
-	__asm__ __volatile__ (
-                "    lgfr  %0,%0\n"
-                "    algr  %0,%1\n"  /* sum += saddr */
-                "    brc   12,0f\n"
-		"    aghi  %0,1\n"   /* add carry */
-		"0:  algr  %0,%2\n"  /* sum += daddr */
-                "    brc   12,1f\n"
-                "    aghi  %0,1\n"   /* add carry */
-		"1:  algfr %0,%3\n"  /* sum += (len<<16) + proto */
-		"    brc   12,2f\n"
-		"    aghi  %0,1\n"   /* add carry */
-		"2:  srlg  0,%0,32\n"
-                "    alr   %0,0\n"   /* fold to 32 bits */
-                "    brc   12,3f\n"
-                "    ahi   %0,1\n"   /* add carry */
-                "3:  llgfr %0,%0"
+	asm volatile(
+		"	lgfr	%0,%0\n"
+		"	algr	%0,%1\n"  /* sum += saddr */
+		"	brc	12,0f\n"
+		"	aghi	%0,1\n"   /* add carry */
+		"0:	algr	%0,%2\n"  /* sum += daddr */
+		"	brc	12,1f\n"
+		"	aghi	%0,1\n"   /* add carry */
+		"1:	algfr	%0,%3\n"  /* sum += (len<<16) + proto */
+		"	brc	12,2f\n"
+		"	aghi	%0,1\n"   /* add carry */
+		"2:	srlg	0,%0,32\n"
+		"	alr	%0,0\n"   /* fold to 32 bits */
+		"	brc	12,3f\n"
+		"	ahi	%0,1\n"   /* add carry */
+		"3:	llgfr	%0,%0"
 		: "+&d" (sum)
 		: "d" (saddr), "d" (daddr),
 		  "d" (((unsigned int) len<<16) + (unsigned int) proto)
-		: "cc", "0" );
+		: "cc", "0");
 #endif /* __s390x__ */
 	return sum;
 }
diff --git a/include/asm-s390/ebcdic.h b/include/asm-s390/ebcdic.h
index 15fd2eda6c90..7f6f641d32f4 100644
--- a/include/asm-s390/ebcdic.h
+++ b/include/asm-s390/ebcdic.h
@@ -26,16 +26,16 @@ codepage_convert(const __u8 *codepage, volatile __u8 * addr, unsigned long nr)
 {
 	if (nr-- <= 0)
 		return;
-        __asm__ __volatile__(
-		"   bras 1,1f\n"
-		"   tr   0(1,%0),0(%2)\n"
-                "0: tr   0(256,%0),0(%2)\n"
-		"   la   %0,256(%0)\n"
-		"1: ahi  %1,-256\n"
-		"   jnm  0b\n"
-		"   ex   %1,0(1)"
-                : "+&a" (addr), "+&a" (nr)
-                : "a" (codepage) : "cc", "memory", "1" );
+	asm volatile(
+		"	bras	1,1f\n"
+		"	tr	0(1,%0),0(%2)\n"
+		"0:	tr	0(256,%0),0(%2)\n"
+		"	la	%0,256(%0)\n"
+		"1:	ahi	%1,-256\n"
+		"	jnm	0b\n"
+		"	ex	%1,0(1)"
+		: "+&a" (addr), "+&a" (nr)
+		: "a" (codepage) : "cc", "memory", "1");
 }
 
 #define ASCEBC(addr,nr) codepage_convert(_ascebc, addr, nr)
diff --git a/include/asm-s390/io.h b/include/asm-s390/io.h
index a6cc27e77007..63c78b9399c4 100644
--- a/include/asm-s390/io.h
+++ b/include/asm-s390/io.h
@@ -27,18 +27,16 @@
 static inline unsigned long virt_to_phys(volatile void * address)
 {
 	unsigned long real_address;
-	__asm__ (
+	asm volatile(
 #ifndef __s390x__
-		 "   lra    %0,0(%1)\n"
-                 "   jz     0f\n"
-                 "   sr     %0,%0\n"
+		 "	lra	%0,0(%1)\n"
 #else /* __s390x__ */
-		 "   lrag   %0,0(%1)\n"
-                 "   jz     0f\n"
-                 "   slgr   %0,%0\n"
+		 "	lrag	%0,0(%1)\n"
 #endif /* __s390x__ */
+		 "	jz	0f\n"
+		 "	la	%0,0\n"
                  "0:"
-                 : "=a" (real_address) : "a" (address) : "cc" );
+		 : "=a" (real_address) : "a" (address) : "cc");
         return real_address;
 }
 
diff --git a/include/asm-s390/irqflags.h b/include/asm-s390/irqflags.h
index 3b566a5b3cc7..3f26131120b7 100644
--- a/include/asm-s390/irqflags.h
+++ b/include/asm-s390/irqflags.h
@@ -10,43 +10,93 @@
 
 #ifdef __KERNEL__
 
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+
+/* store then or system mask. */
+#define __raw_local_irq_stosm(__or)					\
+({									\
+	unsigned long __mask;						\
+	asm volatile(							\
+		"	stosm	%0,%1"					\
+		: "=Q" (__mask) : "i" (__or) : "memory");		\
+	__mask;								\
+})
+
+/* store then and system mask. */
+#define __raw_local_irq_stnsm(__and)					\
+({									\
+	unsigned long __mask;						\
+	asm volatile(							\
+		"	stnsm	%0,%1"					\
+		: "=Q" (__mask) : "i" (__and) : "memory");		\
+	__mask;								\
+})
+
+/* set system mask. */
+#define __raw_local_irq_ssm(__mask)					\
+({									\
+	asm volatile("ssm   %0" : : "Q" (__mask) : "memory");		\
+})
+
+#else /* __GNUC__ */
+
+/* store then or system mask. */
+#define __raw_local_irq_stosm(__or)					\
+({									\
+	unsigned long __mask;						\
+	asm volatile(							\
+		"	stosm	0(%1),%2"				\
+		: "=m" (__mask)						\
+		: "a" (&__mask), "i" (__or) : "memory");		\
+	__mask;								\
+})
+
+/* store then and system mask. */
+#define __raw_local_irq_stnsm(__and)					\
+({									\
+	unsigned long __mask;						\
+	asm volatile(							\
+		"	stnsm	0(%1),%2"				\
+		: "=m" (__mask)						\
+		: "a" (&__mask), "i" (__and) : "memory");		\
+	__mask;								\
+})
+
+/* set system mask. */
+#define __raw_local_irq_ssm(__mask)					\
+({									\
+	asm volatile(							\
+		"	ssm	0(%0)"					\
+		: : "a" (&__mask), "m" (__mask) : "memory");		\
+})
+
+#endif /* __GNUC__ */
+
 /* interrupt control.. */
-#define raw_local_irq_enable() ({ \
-	unsigned long  __dummy; \
-	__asm__ __volatile__ ( \
-		"stosm 0(%1),0x03" \
-		: "=m" (__dummy) : "a" (&__dummy) : "memory" ); \
-	})
-
-#define raw_local_irq_disable() ({ \
-	unsigned long __flags; \
-	__asm__ __volatile__ ( \
-		"stnsm 0(%1),0xfc" : "=m" (__flags) : "a" (&__flags) ); \
-	__flags; \
-	})
-
-#define raw_local_save_flags(x)							\
-do {										\
-	typecheck(unsigned long, x);						\
-	__asm__ __volatile__("stosm 0(%1),0" : "=m" (x) : "a" (&x), "m" (x) );	\
-} while (0)
+static inline unsigned long raw_local_irq_enable(void)
+{
+	return __raw_local_irq_stosm(0x03);
+}
 
-#define raw_local_irq_restore(x)						\
-do {										\
-	typecheck(unsigned long, x);						\
-	__asm__ __volatile__("ssm   0(%0)" : : "a" (&x), "m" (x) : "memory");	\
+static inline unsigned long raw_local_irq_disable(void)
+{
+	return __raw_local_irq_stnsm(0xfc);
+}
+
+#define raw_local_save_flags(x)						\
+do {									\
+	typecheck(unsigned long, x);					\
+	(x) = __raw_local_irq_stosm(0x00);				\
 } while (0)
 
-#define raw_irqs_disabled()		\
-({					\
-	unsigned long flags;		\
-	raw_local_save_flags(flags);	\
-	!((flags >> __FLAG_SHIFT) & 3);	\
-})
+static inline void raw_local_irq_restore(unsigned long flags)
+{
+	__raw_local_irq_ssm(flags);
+}
 
 static inline int raw_irqs_disabled_flags(unsigned long flags)
 {
-	return !((flags >> __FLAG_SHIFT) & 3);
+	return !(flags & (3UL << (BITS_PER_LONG - 8)));
 }
 
 /* For spinlocks etc */
diff --git a/include/asm-s390/lowcore.h b/include/asm-s390/lowcore.h
index 18695d10dedf..06583ed0bde7 100644
--- a/include/asm-s390/lowcore.h
+++ b/include/asm-s390/lowcore.h
@@ -359,7 +359,7 @@ extern struct _lowcore *lowcore_ptr[];
 
 static inline void set_prefix(__u32 address)
 {
-        __asm__ __volatile__ ("spx %0" : : "m" (address) : "memory" );
+	asm volatile("spx %0" : : "m" (address) : "memory");
 }
 
 #define __PANIC_MAGIC           0xDEADC0DE
diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h
index b2628dc5c490..796c400f2b79 100644
--- a/include/asm-s390/page.h
+++ b/include/asm-s390/page.h
@@ -22,89 +22,45 @@
 #include <asm/setup.h>
 #ifndef __ASSEMBLY__
 
-#ifndef __s390x__
-
-static inline void clear_page(void *page)
-{
-	register_pair rp;
-
-	rp.subreg.even = (unsigned long) page;
-	rp.subreg.odd = (unsigned long) 4096;
-        asm volatile ("   slr  1,1\n"
-		      "   mvcl %0,0"
-		      : "+&a" (rp) : : "memory", "cc", "1" );
-}
-
-static inline void copy_page(void *to, void *from)
-{
-        if (MACHINE_HAS_MVPG)
-		asm volatile ("   sr   0,0\n"
-			      "   mvpg %0,%1"
-			      : : "a" ((void *)(to)), "a" ((void *)(from))
-			      : "memory", "cc", "0" );
-	else
-		asm volatile ("   mvc  0(256,%0),0(%1)\n"
-			      "   mvc  256(256,%0),256(%1)\n"
-			      "   mvc  512(256,%0),512(%1)\n"
-			      "   mvc  768(256,%0),768(%1)\n"
-			      "   mvc  1024(256,%0),1024(%1)\n"
-			      "   mvc  1280(256,%0),1280(%1)\n"
-			      "   mvc  1536(256,%0),1536(%1)\n"
-			      "   mvc  1792(256,%0),1792(%1)\n"
-			      "   mvc  2048(256,%0),2048(%1)\n"
-			      "   mvc  2304(256,%0),2304(%1)\n"
-			      "   mvc  2560(256,%0),2560(%1)\n"
-			      "   mvc  2816(256,%0),2816(%1)\n"
-			      "   mvc  3072(256,%0),3072(%1)\n"
-			      "   mvc  3328(256,%0),3328(%1)\n"
-			      "   mvc  3584(256,%0),3584(%1)\n"
-			      "   mvc  3840(256,%0),3840(%1)\n"
-			      : : "a"((void *)(to)),"a"((void *)(from)) 
-			      : "memory" );
-}
-
-#else /* __s390x__ */
-
 static inline void clear_page(void *page)
 {
-        asm volatile ("   lgr  2,%0\n"
-                      "   lghi 3,4096\n"
-                      "   slgr 1,1\n"
-                      "   mvcl 2,0"
-                      : : "a" ((void *) (page))
-		      : "memory", "cc", "1", "2", "3" );
+	register unsigned long reg1 asm ("1") = 0;
+	register void *reg2 asm ("2") = page;
+	register unsigned long reg3 asm ("3") = 4096;
+	asm volatile(
+		"	mvcl	2,0"
+		: "+d" (reg2), "+d" (reg3) : "d" (reg1) : "memory", "cc");
 }
 
 static inline void copy_page(void *to, void *from)
 {
-        if (MACHINE_HAS_MVPG)
-		asm volatile ("   sgr  0,0\n"
-			      "   mvpg %0,%1"
-			      : : "a" ((void *)(to)), "a" ((void *)(from))
-			      : "memory", "cc", "0" );
-	else
-		asm volatile ("   mvc  0(256,%0),0(%1)\n"
-			      "   mvc  256(256,%0),256(%1)\n"
-			      "   mvc  512(256,%0),512(%1)\n"
-			      "   mvc  768(256,%0),768(%1)\n"
-			      "   mvc  1024(256,%0),1024(%1)\n"
-			      "   mvc  1280(256,%0),1280(%1)\n"
-			      "   mvc  1536(256,%0),1536(%1)\n"
-			      "   mvc  1792(256,%0),1792(%1)\n"
-			      "   mvc  2048(256,%0),2048(%1)\n"
-			      "   mvc  2304(256,%0),2304(%1)\n"
-			      "   mvc  2560(256,%0),2560(%1)\n"
-			      "   mvc  2816(256,%0),2816(%1)\n"
-			      "   mvc  3072(256,%0),3072(%1)\n"
-			      "   mvc  3328(256,%0),3328(%1)\n"
-			      "   mvc  3584(256,%0),3584(%1)\n"
-			      "   mvc  3840(256,%0),3840(%1)\n"
-			      : : "a"((void *)(to)),"a"((void *)(from)) 
-			      : "memory" );
+	if (MACHINE_HAS_MVPG) {
+		register unsigned long reg0 asm ("0") = 0;
+		asm volatile(
+			"	mvpg	%0,%1"
+			: : "a" (to), "a" (from), "d" (reg0)
+			: "memory", "cc");
+	} else
+		asm volatile(
+			"	mvc	0(256,%0),0(%1)\n"
+			"	mvc	256(256,%0),256(%1)\n"
+			"	mvc	512(256,%0),512(%1)\n"
+			"	mvc	768(256,%0),768(%1)\n"
+			"	mvc	1024(256,%0),1024(%1)\n"
+			"	mvc	1280(256,%0),1280(%1)\n"
+			"	mvc	1536(256,%0),1536(%1)\n"
+			"	mvc	1792(256,%0),1792(%1)\n"
+			"	mvc	2048(256,%0),2048(%1)\n"
+			"	mvc	2304(256,%0),2304(%1)\n"
+			"	mvc	2560(256,%0),2560(%1)\n"
+			"	mvc	2816(256,%0),2816(%1)\n"
+			"	mvc	3072(256,%0),3072(%1)\n"
+			"	mvc	3328(256,%0),3328(%1)\n"
+			"	mvc	3584(256,%0),3584(%1)\n"
+			"	mvc	3840(256,%0),3840(%1)\n"
+			: : "a" (to), "a" (from) : "memory");
 }
 
-#endif /* __s390x__ */
-
 #define clear_user_page(page, vaddr, pg)	clear_page(page)
 #define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
 
@@ -159,7 +115,7 @@ extern unsigned int default_storage_key;
 static inline void
 page_set_storage_key(unsigned long addr, unsigned int skey)
 {
-	asm volatile ( "sske %0,%1" : : "d" (skey), "a" (addr) );
+	asm volatile("sske %0,%1" : : "d" (skey), "a" (addr));
 }
 
 static inline unsigned int
@@ -167,8 +123,7 @@ page_get_storage_key(unsigned long addr)
 {
 	unsigned int skey;
 
-	asm volatile ( "iske %0,%1" : "=d" (skey) : "a" (addr), "0" (0) );
-
+	asm volatile("iske %0,%1" : "=d" (skey) : "a" (addr), "0" (0));
 	return skey;
 }
 
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index e965309fedac..83425cdefc91 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -554,9 +554,10 @@ static inline void __ptep_ipte(unsigned long address, pte_t *ptep)
 		/* ipte in zarch mode can do the math */
 		pte_t *pto = ptep;
 #endif
-		asm volatile ("ipte %2,%3"
-			      : "=m" (*ptep) : "m" (*ptep),
-				"a" (pto), "a" (address) );
+		asm volatile(
+			"	ipte	%2,%3"
+			: "=m" (*ptep) : "m" (*ptep),
+			  "a" (pto), "a" (address));
 	}
 	pte_val(*ptep) = _PAGE_TYPE_EMPTY;
 }
@@ -609,16 +610,17 @@ ptep_establish(struct vm_area_struct *vma,
 /*
  * Test and clear referenced bit in storage key.
  */
-#define page_test_and_clear_young(page)					  \
-({									  \
-	struct page *__page = (page);					  \
-	unsigned long __physpage = __pa((__page-mem_map) << PAGE_SHIFT);  \
-	int __ccode;							  \
-	asm volatile ("rrbe 0,%1\n\t"					  \
-		      "ipm  %0\n\t"					  \
-		      "srl  %0,28\n\t" 					  \
-                      : "=d" (__ccode) : "a" (__physpage) : "cc" );	  \
-	(__ccode & 2);							  \
+#define page_test_and_clear_young(page)					\
+({									\
+	struct page *__page = (page);					\
+	unsigned long __physpage = __pa((__page-mem_map) << PAGE_SHIFT);\
+	int __ccode;							\
+	asm volatile(							\
+		"	rrbe	0,%1\n"					\
+		"	ipm	%0\n"					\
+		"	srl	%0,28\n"				\
+		: "=d" (__ccode) : "a" (__physpage) : "cc");		\
+	(__ccode & 2);							\
 })
 
 /*
diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index 578c2209fa76..cbbedc63ba25 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -13,7 +13,6 @@
 #ifndef __ASM_S390_PROCESSOR_H
 #define __ASM_S390_PROCESSOR_H
 
-#include <asm/page.h>
 #include <asm/ptrace.h>
 
 #ifdef __KERNEL__
@@ -21,7 +20,7 @@
  * Default implementation of macro that returns current
  * instruction pointer ("program counter").
  */
-#define current_text_addr() ({ void *pc; __asm__("basr %0,0":"=a"(pc)); pc; })
+#define current_text_addr() ({ void *pc; asm("basr %0,0" : "=a" (pc)); pc; })
 
 /*
  *  CPU type and hardware bug flags. Kept separately for each CPU.
@@ -202,7 +201,7 @@ unsigned long get_wchan(struct task_struct *p);
 static inline void cpu_relax(void)
 {
 	if (MACHINE_HAS_DIAG44)
-		asm volatile ("diag 0,0,68" : : : "memory");
+		asm volatile("diag 0,0,68" : : : "memory");
 	else
 		barrier();
 }
@@ -213,9 +212,9 @@ static inline void cpu_relax(void)
 static inline void __load_psw(psw_t psw)
 {
 #ifndef __s390x__
-	asm volatile ("lpsw  0(%0)" : : "a" (&psw), "m" (psw) : "cc" );
+	asm volatile("lpsw  0(%0)" : : "a" (&psw), "m" (psw) : "cc");
 #else
-	asm volatile ("lpswe 0(%0)" : : "a" (&psw), "m" (psw) : "cc" );
+	asm volatile("lpswe 0(%0)" : : "a" (&psw), "m" (psw) : "cc");
 #endif
 }
 
@@ -232,20 +231,20 @@ static inline void __load_psw_mask (unsigned long mask)
 	psw.mask = mask;
 
 #ifndef __s390x__
-	asm volatile (
-		"    basr %0,0\n"
-		"0:  ahi  %0,1f-0b\n"
-		"    st	  %0,4(%1)\n"
-		"    lpsw 0(%1)\n"
+	asm volatile(
+		"	basr	%0,0\n"
+		"0:	ahi	%0,1f-0b\n"
+		"	st	%0,4(%1)\n"
+		"	lpsw	0(%1)\n"
 		"1:"
-		: "=&d" (addr) : "a" (&psw), "m" (psw) : "memory", "cc" );
+		: "=&d" (addr) : "a" (&psw), "m" (psw) : "memory", "cc");
 #else /* __s390x__ */
-	asm volatile (
-		"    larl  %0,1f\n"
-		"    stg   %0,8(%1)\n"
-		"    lpswe 0(%1)\n"
+	asm volatile(
+		"	larl	%0,1f\n"
+		"	stg	%0,8(%1)\n"
+		"	lpswe	0(%1)\n"
 		"1:"
-		: "=&d" (addr) : "a" (&psw), "m" (psw) : "memory", "cc" );
+		: "=&d" (addr) : "a" (&psw), "m" (psw) : "memory", "cc");
 #endif /* __s390x__ */
 }
  
@@ -274,56 +273,57 @@ static inline void disabled_wait(unsigned long code)
          * the processor is dead afterwards
          */
 #ifndef __s390x__
-        asm volatile ("    stctl 0,0,0(%2)\n"
-                      "    ni    0(%2),0xef\n" /* switch off protection */
-                      "    lctl  0,0,0(%2)\n"
-                      "    stpt  0xd8\n"       /* store timer */
-                      "    stckc 0xe0\n"       /* store clock comparator */
-                      "    stpx  0x108\n"      /* store prefix register */
-                      "    stam  0,15,0x120\n" /* store access registers */
-                      "    std   0,0x160\n"    /* store f0 */
-                      "    std   2,0x168\n"    /* store f2 */
-                      "    std   4,0x170\n"    /* store f4 */
-                      "    std   6,0x178\n"    /* store f6 */
-                      "    stm   0,15,0x180\n" /* store general registers */
-                      "    stctl 0,15,0x1c0\n" /* store control registers */
-                      "    oi    0x1c0,0x10\n" /* fake protection bit */
-                      "    lpsw 0(%1)"
-                      : "=m" (ctl_buf)
-		      : "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc" );
+	asm volatile(
+		"	stctl	0,0,0(%2)\n"
+		"	ni	0(%2),0xef\n"	/* switch off protection */
+		"	lctl	0,0,0(%2)\n"
+		"	stpt	0xd8\n"		/* store timer */
+		"	stckc	0xe0\n"		/* store clock comparator */
+		"	stpx	0x108\n"	/* store prefix register */
+		"	stam	0,15,0x120\n"	/* store access registers */
+		"	std	0,0x160\n"	/* store f0 */
+		"	std	2,0x168\n"	/* store f2 */
+		"	std	4,0x170\n"	/* store f4 */
+		"	std	6,0x178\n"	/* store f6 */
+		"	stm	0,15,0x180\n"	/* store general registers */
+		"	stctl	0,15,0x1c0\n"	/* store control registers */
+		"	oi	0x1c0,0x10\n"	/* fake protection bit */
+		"	lpsw	0(%1)"
+		: "=m" (ctl_buf)
+		: "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc");
 #else /* __s390x__ */
-        asm volatile ("    stctg 0,0,0(%2)\n"
-                      "    ni    4(%2),0xef\n" /* switch off protection */
-                      "    lctlg 0,0,0(%2)\n"
-                      "    lghi  1,0x1000\n"
-                      "    stpt  0x328(1)\n"      /* store timer */
-                      "    stckc 0x330(1)\n"      /* store clock comparator */
-                      "    stpx  0x318(1)\n"      /* store prefix register */
-                      "    stam  0,15,0x340(1)\n" /* store access registers */
-                      "    stfpc 0x31c(1)\n"      /* store fpu control */
-                      "    std   0,0x200(1)\n"    /* store f0 */
-                      "    std   1,0x208(1)\n"    /* store f1 */
-                      "    std   2,0x210(1)\n"    /* store f2 */
-                      "    std   3,0x218(1)\n"    /* store f3 */
-                      "    std   4,0x220(1)\n"    /* store f4 */
-                      "    std   5,0x228(1)\n"    /* store f5 */
-                      "    std   6,0x230(1)\n"    /* store f6 */
-                      "    std   7,0x238(1)\n"    /* store f7 */
-                      "    std   8,0x240(1)\n"    /* store f8 */
-                      "    std   9,0x248(1)\n"    /* store f9 */
-                      "    std   10,0x250(1)\n"   /* store f10 */
-                      "    std   11,0x258(1)\n"   /* store f11 */
-                      "    std   12,0x260(1)\n"   /* store f12 */
-                      "    std   13,0x268(1)\n"   /* store f13 */
-                      "    std   14,0x270(1)\n"   /* store f14 */
-                      "    std   15,0x278(1)\n"   /* store f15 */
-                      "    stmg  0,15,0x280(1)\n" /* store general registers */
-                      "    stctg 0,15,0x380(1)\n" /* store control registers */
-                      "    oi    0x384(1),0x10\n" /* fake protection bit */
-                      "    lpswe 0(%1)"
-                      : "=m" (ctl_buf)
-		      : "a" (&dw_psw), "a" (&ctl_buf),
-		        "m" (dw_psw) : "cc", "0", "1");
+	asm volatile(
+		"	stctg	0,0,0(%2)\n"
+		"	ni	4(%2),0xef\n"	/* switch off protection */
+		"	lctlg	0,0,0(%2)\n"
+		"	lghi	1,0x1000\n"
+		"	stpt	0x328(1)\n"	/* store timer */
+		"	stckc	0x330(1)\n"	/* store clock comparator */
+		"	stpx	0x318(1)\n"	/* store prefix register */
+		"	stam	0,15,0x340(1)\n"/* store access registers */
+		"	stfpc	0x31c(1)\n"	/* store fpu control */
+		"	std	0,0x200(1)\n"	/* store f0 */
+		"	std	1,0x208(1)\n"	/* store f1 */
+		"	std	2,0x210(1)\n"	/* store f2 */
+		"	std	3,0x218(1)\n"	/* store f3 */
+		"	std	4,0x220(1)\n"	/* store f4 */
+		"	std	5,0x228(1)\n"	/* store f5 */
+		"	std	6,0x230(1)\n"	/* store f6 */
+		"	std	7,0x238(1)\n"	/* store f7 */
+		"	std	8,0x240(1)\n"	/* store f8 */
+		"	std	9,0x248(1)\n"	/* store f9 */
+		"	std	10,0x250(1)\n"	/* store f10 */
+		"	std	11,0x258(1)\n"	/* store f11 */
+		"	std	12,0x260(1)\n"	/* store f12 */
+		"	std	13,0x268(1)\n"	/* store f13 */
+		"	std	14,0x270(1)\n"	/* store f14 */
+		"	std	15,0x278(1)\n"	/* store f15 */
+		"	stmg	0,15,0x280(1)\n"/* store general registers */
+		"	stctg	0,15,0x380(1)\n"/* store control registers */
+		"	oi	0x384(1),0x10\n"/* fake protection bit */
+		"	lpswe	0(%1)"
+		: "=m" (ctl_buf)
+		: "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc", "0");
 #endif /* __s390x__ */
 }
 
diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h
index 4d75d77b0f99..8d2bf65b0b64 100644
--- a/include/asm-s390/ptrace.h
+++ b/include/asm-s390/ptrace.h
@@ -479,7 +479,7 @@ extern void show_regs(struct pt_regs * regs);
 static inline void
 psw_set_key(unsigned int key)
 {
-	asm volatile ( "spka 0(%0)" : : "d" (key) );
+	asm volatile("spka 0(%0)" : : "d" (key));
 }
 
 #endif /* __ASSEMBLY__ */
diff --git a/include/asm-s390/rwsem.h b/include/asm-s390/rwsem.h
index 13ec16965150..90f4eccaa290 100644
--- a/include/asm-s390/rwsem.h
+++ b/include/asm-s390/rwsem.h
@@ -122,23 +122,23 @@ static inline void __down_read(struct rw_semaphore *sem)
 {
 	signed long old, new;
 
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   ahi  %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	ahi	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   aghi %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	aghi	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count),
-		  "i" (RWSEM_ACTIVE_READ_BIAS) : "cc", "memory" );
+		  "i" (RWSEM_ACTIVE_READ_BIAS) : "cc", "memory");
 	if (old < 0)
 		rwsem_down_read_failed(sem);
 }
@@ -150,27 +150,27 @@ static inline int __down_read_trylock(struct rw_semaphore *sem)
 {
 	signed long old, new;
 
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: ltr  %1,%0\n"
-		"   jm   1f\n"
-		"   ahi  %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b\n"
+		"	l	%0,0(%3)\n"
+		"0:	ltr	%1,%0\n"
+		"	jm	1f\n"
+		"	ahi	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b\n"
 		"1:"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: ltgr %1,%0\n"
-		"   jm   1f\n"
-		"   aghi %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b\n"
+		"	lg	%0,0(%3)\n"
+		"0:	ltgr	%1,%0\n"
+		"	jm	1f\n"
+		"	aghi	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b\n"
 		"1:"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count),
-		  "i" (RWSEM_ACTIVE_READ_BIAS) : "cc", "memory" );
+		  "i" (RWSEM_ACTIVE_READ_BIAS) : "cc", "memory");
 	return old >= 0 ? 1 : 0;
 }
 
@@ -182,23 +182,23 @@ static inline void __down_write_nested(struct rw_semaphore *sem, int subclass)
 	signed long old, new, tmp;
 
 	tmp = RWSEM_ACTIVE_WRITE_BIAS;
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   a    %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	a	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   ag   %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	ag	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count), "m" (tmp)
-		: "cc", "memory" );
+		: "cc", "memory");
 	if (old != 0)
 		rwsem_down_write_failed(sem);
 }
@@ -215,24 +215,24 @@ static inline int __down_write_trylock(struct rw_semaphore *sem)
 {
 	signed long old;
 
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%2)\n"
-		"0: ltr  %0,%0\n"
-		"   jnz  1f\n"
-		"   cs   %0,%4,0(%2)\n"
-		"   jl   0b\n"
+		"	l	%0,0(%2)\n"
+		"0:	ltr	%0,%0\n"
+		"	jnz	1f\n"
+		"	cs	%0,%4,0(%2)\n"
+		"	jl	0b\n"
 #else /* __s390x__ */
-		"   lg   %0,0(%2)\n"
-		"0: ltgr %0,%0\n"
-		"   jnz  1f\n"
-		"   csg  %0,%4,0(%2)\n"
-		"   jl   0b\n"
+		"	lg	%0,0(%2)\n"
+		"0:	ltgr	%0,%0\n"
+		"	jnz	1f\n"
+		"	csg	%0,%4,0(%2)\n"
+		"	jl	0b\n"
 #endif /* __s390x__ */
 		"1:"
-                : "=&d" (old), "=m" (sem->count)
+		: "=&d" (old), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count),
-		  "d" (RWSEM_ACTIVE_WRITE_BIAS) : "cc", "memory" );
+		  "d" (RWSEM_ACTIVE_WRITE_BIAS) : "cc", "memory");
 	return (old == RWSEM_UNLOCKED_VALUE) ? 1 : 0;
 }
 
@@ -243,24 +243,24 @@ static inline void __up_read(struct rw_semaphore *sem)
 {
 	signed long old, new;
 
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   ahi  %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	ahi	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   aghi %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	aghi	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count),
 		  "i" (-RWSEM_ACTIVE_READ_BIAS)
-		: "cc", "memory" );
+		: "cc", "memory");
 	if (new < 0)
 		if ((new & RWSEM_ACTIVE_MASK) == 0)
 			rwsem_wake(sem);
@@ -274,23 +274,23 @@ static inline void __up_write(struct rw_semaphore *sem)
 	signed long old, new, tmp;
 
 	tmp = -RWSEM_ACTIVE_WRITE_BIAS;
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   a    %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	a	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   ag   %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	ag	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count), "m" (tmp)
-		: "cc", "memory" );
+		: "cc", "memory");
 	if (new < 0)
 		if ((new & RWSEM_ACTIVE_MASK) == 0)
 			rwsem_wake(sem);
@@ -304,23 +304,23 @@ static inline void __downgrade_write(struct rw_semaphore *sem)
 	signed long old, new, tmp;
 
 	tmp = -RWSEM_WAITING_BIAS;
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   a    %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	a	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   ag   %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	ag	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count), "m" (tmp)
-		: "cc", "memory" );
+		: "cc", "memory");
 	if (new > 1)
 		rwsem_downgrade_wake(sem);
 }
@@ -332,23 +332,23 @@ static inline void rwsem_atomic_add(long delta, struct rw_semaphore *sem)
 {
 	signed long old, new;
 
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   ar   %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	ar	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   agr  %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	agr	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count), "d" (delta)
-		: "cc", "memory" );
+		: "cc", "memory");
 }
 
 /*
@@ -358,23 +358,23 @@ static inline long rwsem_atomic_update(long delta, struct rw_semaphore *sem)
 {
 	signed long old, new;
 
-	__asm__ __volatile__(
+	asm volatile(
 #ifndef __s390x__
-		"   l    %0,0(%3)\n"
-		"0: lr   %1,%0\n"
-		"   ar   %1,%5\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	l	%0,0(%3)\n"
+		"0:	lr	%1,%0\n"
+		"	ar	%1,%5\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b"
 #else /* __s390x__ */
-		"   lg   %0,0(%3)\n"
-		"0: lgr  %1,%0\n"
-		"   agr  %1,%5\n"
-		"   csg  %0,%1,0(%3)\n"
-		"   jl   0b"
+		"	lg	%0,0(%3)\n"
+		"0:	lgr	%1,%0\n"
+		"	agr	%1,%5\n"
+		"	csg	%0,%1,0(%3)\n"
+		"	jl	0b"
 #endif /* __s390x__ */
-                : "=&d" (old), "=&d" (new), "=m" (sem->count)
+		: "=&d" (old), "=&d" (new), "=m" (sem->count)
 		: "a" (&sem->count), "m" (sem->count), "d" (delta)
-		: "cc", "memory" );
+		: "cc", "memory");
 	return new;
 }
 
diff --git a/include/asm-s390/semaphore.h b/include/asm-s390/semaphore.h
index 32cdc69f39f4..dbce058aefa9 100644
--- a/include/asm-s390/semaphore.h
+++ b/include/asm-s390/semaphore.h
@@ -85,17 +85,17 @@ static inline int down_trylock(struct semaphore * sem)
 	 *       sem->count.counter = --new_val;
 	 * In the ppc code this is called atomic_dec_if_positive.
 	 */
-	__asm__ __volatile__ (
-		"   l    %0,0(%3)\n"
-		"0: ltr  %1,%0\n"
-		"   jle  1f\n"
-		"   ahi  %1,-1\n"
-		"   cs   %0,%1,0(%3)\n"
-		"   jl   0b\n"
+	asm volatile(
+		"	l	%0,0(%3)\n"
+		"0:	ltr	%1,%0\n"
+		"	jle	1f\n"
+		"	ahi	%1,-1\n"
+		"	cs	%0,%1,0(%3)\n"
+		"	jl	0b\n"
 		"1:"
 		: "=&d" (old_val), "=&d" (new_val), "=m" (sem->count.counter)
 		: "a" (&sem->count.counter), "m" (sem->count.counter)
-		: "cc", "memory" );
+		: "cc", "memory");
 	return old_val <= 0;
 }
 
diff --git a/include/asm-s390/sfp-machine.h b/include/asm-s390/sfp-machine.h
index de69dfa46fbb..8ca8c77b2d04 100644
--- a/include/asm-s390/sfp-machine.h
+++ b/include/asm-s390/sfp-machine.h
@@ -76,21 +76,23 @@
 	unsigned int __r2 = (x2) + (y2);			\
 	unsigned int __r1 = (x1);				\
 	unsigned int __r0 = (x0);				\
-	__asm__ ("   alr %2,%3\n"				\
-		 "   brc 12,0f\n"				\
-		 "   lhi 0,1\n"					\
-		 "   alr %1,0\n"				\
-		 "   brc 12,0f\n"				\
-		 "   alr %0,0\n"				\
-		 "0:"						\
-		 : "+&d" (__r2), "+&d" (__r1), "+&d" (__r0)	\
-		 : "d" (y0), "i" (1) : "cc", "0" );		\
-	__asm__ ("   alr %1,%2\n"				\
-		 "   brc 12,0f\n"				\
-		 "   ahi %0,1\n"				\
-		 "0:"						\
-		 : "+&d" (__r2), "+&d" (__r1)			\
-		 : "d" (y1) : "cc" );				\
+	asm volatile(						\
+		"	alr	%2,%3\n"			\
+		"	brc	12,0f\n"			\
+		"	lhi	0,1\n"				\
+		"	alr	%1,0\n"				\
+		"	brc	12,0f\n"			\
+		"	alr	%0,0\n"				\
+		"0:"						\
+		: "+&d" (__r2), "+&d" (__r1), "+&d" (__r0)	\
+		: "d" (y0), "i" (1) : "cc", "0" );		\
+	asm volatile(						\
+		"	alr	%1,%2\n"			\
+		"	brc	12,0f\n"			\
+		"	ahi	%0,1\n"				\
+		"0:"						\
+		: "+&d" (__r2), "+&d" (__r1)			\
+		: "d" (y1) : "cc");				\
 	(r2) = __r2;						\
 	(r1) = __r1;						\
 	(r0) = __r0;						\
@@ -100,21 +102,23 @@
 	unsigned int __r2 = (x2) - (y2);			\
 	unsigned int __r1 = (x1);				\
 	unsigned int __r0 = (x0);				\
-	__asm__ ("   slr %2,%3\n"				\
-		 "   brc 3,0f\n"				\
-		 "   lhi 0,1\n"					\
-		 "   slr %1,0\n"				\
-		 "   brc 3,0f\n"				\
-		 "   slr %0,0\n"				\
-		 "0:"						\
-		 : "+&d" (__r2), "+&d" (__r1), "+&d" (__r0)	\
-		 : "d" (y0) : "cc", "0" );			\
-	__asm__ ("   slr %1,%2\n"				\
-		 "   brc 3,0f\n"				\
-		 "   ahi %0,-1\n"				\
-		 "0:"						\
-		 : "+&d" (__r2), "+&d" (__r1)			\
-		 : "d" (y1) : "cc" );				\
+	asm volatile(						\
+		"	slr   %2,%3\n"				\
+		"	brc	3,0f\n"				\
+		"	lhi	0,1\n"				\
+		"	slr	%1,0\n"				\
+		"	brc	3,0f\n"				\
+		"	slr	%0,0\n"				\
+		"0:"						\
+		: "+&d" (__r2), "+&d" (__r1), "+&d" (__r0)	\
+		: "d" (y0) : "cc", "0");			\
+	asm volatile(						\
+		"	slr	%1,%2\n"			\
+		"	brc	3,0f\n"				\
+		"	ahi	%0,-1\n"			\
+		"0:"						\
+		: "+&d" (__r2), "+&d" (__r1)			\
+		: "d" (y1) : "cc");				\
 	(r2) = __r2;						\
 	(r1) = __r1;						\
 	(r0) = __r0;						\
diff --git a/include/asm-s390/sigp.h b/include/asm-s390/sigp.h
index fc56458aff66..e16d56f8dfe1 100644
--- a/include/asm-s390/sigp.h
+++ b/include/asm-s390/sigp.h
@@ -70,16 +70,16 @@ typedef enum
 static inline sigp_ccode
 signal_processor(__u16 cpu_addr, sigp_order_code order_code)
 {
+	register unsigned long reg1 asm ("1") = 0;
 	sigp_ccode ccode;
 
-	__asm__ __volatile__(
-		"    sr     1,1\n"        /* parameter=0 in gpr 1 */
-		"    sigp   1,%1,0(%2)\n"
-		"    ipm    %0\n"
-		"    srl    %0,28\n"
-		: "=d" (ccode)
-		: "d" (__cpu_logical_map[cpu_addr]), "a" (order_code)
-		: "cc" , "memory", "1" );
+	asm volatile(
+		"	sigp	%1,%2,0(%3)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		:	"=d"	(ccode)
+		: "d" (reg1), "d" (__cpu_logical_map[cpu_addr]),
+		  "a" (order_code) : "cc" , "memory");
 	return ccode;
 }
 
@@ -87,20 +87,18 @@ signal_processor(__u16 cpu_addr, sigp_order_code order_code)
  * Signal processor with parameter
  */
 static inline sigp_ccode
-signal_processor_p(__u32 parameter, __u16 cpu_addr,
-		   sigp_order_code order_code)
+signal_processor_p(__u32 parameter, __u16 cpu_addr, sigp_order_code order_code)
 {
+	register unsigned int reg1 asm ("1") = parameter;
 	sigp_ccode ccode;
-	
-	__asm__ __volatile__(
-		"    lr     1,%1\n"       /* parameter in gpr 1 */
-		"    sigp   1,%2,0(%3)\n"
-		"    ipm    %0\n"
-		"    srl    %0,28\n"
+
+	asm volatile(
+		"	sigp	%1,%2,0(%3)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
 		: "=d" (ccode)
-		: "d" (parameter), "d" (__cpu_logical_map[cpu_addr]),
-                  "a" (order_code)
-		: "cc" , "memory", "1" );
+		: "d" (reg1), "d" (__cpu_logical_map[cpu_addr]),
+		  "a" (order_code) : "cc" , "memory");
 	return ccode;
 }
 
@@ -108,24 +106,21 @@ signal_processor_p(__u32 parameter, __u16 cpu_addr,
  * Signal processor with parameter and return status
  */
 static inline sigp_ccode
-signal_processor_ps(__u32 *statusptr, __u32 parameter,
-		    __u16 cpu_addr, sigp_order_code order_code)
+signal_processor_ps(__u32 *statusptr, __u32 parameter, __u16 cpu_addr,
+		    sigp_order_code order_code)
 {
+	register unsigned int reg1 asm ("1") = parameter;
 	sigp_ccode ccode;
-	
-	__asm__ __volatile__(
-		"    sr     2,2\n"        /* clear status */
-		"    lr     3,%2\n"       /* parameter in gpr 3 */
-		"    sigp   2,%3,0(%4)\n"
-		"    st     2,%1\n"
-		"    ipm    %0\n"
-		"    srl    %0,28\n"
-		: "=d" (ccode), "=m" (*statusptr)
-		: "d" (parameter), "d" (__cpu_logical_map[cpu_addr]),
-                  "a" (order_code)
-		: "cc" , "memory", "2" , "3"
-		);
-   return ccode;
+
+	asm volatile(
+		"	sigp	%1,%2,0(%3)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (ccode), "+d" (reg1)
+		: "d" (__cpu_logical_map[cpu_addr]), "a" (order_code)
+		: "cc" , "memory");
+	*statusptr = reg1;
+	return ccode;
 }
 
 #endif /* __SIGP__ */
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 9fb02e9779c9..c3cf030ada4d 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -56,7 +56,7 @@ static inline __u16 hard_smp_processor_id(void)
 {
         __u16 cpu_address;
  
-        __asm__ ("stap %0\n" : "=m" (cpu_address));
+	asm volatile("stap %0" : "=m" (cpu_address));
         return cpu_address;
 }
 
diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h
index 273dbecf8ace..ce3edf6d63b3 100644
--- a/include/asm-s390/spinlock.h
+++ b/include/asm-s390/spinlock.h
@@ -11,17 +11,36 @@
 #ifndef __ASM_SPINLOCK_H
 #define __ASM_SPINLOCK_H
 
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+
+static inline int
+_raw_compare_and_swap(volatile unsigned int *lock,
+		      unsigned int old, unsigned int new)
+{
+	asm volatile(
+		"	cs	%0,%3,%1"
+		: "=d" (old), "=Q" (*lock)
+		: "0" (old), "d" (new), "Q" (*lock)
+		: "cc", "memory" );
+	return old;
+}
+
+#else /* __GNUC__ */
+
 static inline int
 _raw_compare_and_swap(volatile unsigned int *lock,
 		      unsigned int old, unsigned int new)
 {
-	asm volatile ("cs %0,%3,0(%4)"
-		      : "=d" (old), "=m" (*lock)
-		      : "0" (old), "d" (new), "a" (lock), "m" (*lock)
-		      : "cc", "memory" );
+	asm volatile(
+		"	cs	%0,%3,0(%4)"
+		: "=d" (old), "=m" (*lock)
+		: "0" (old), "d" (new), "a" (lock), "m" (*lock)
+		: "cc", "memory" );
 	return old;
 }
 
+#endif /* __GNUC__ */
+
 /*
  * Simple spin lock operations.  There are two variants, one clears IRQ's
  * on the local processor, one does not.
diff --git a/include/asm-s390/string.h b/include/asm-s390/string.h
index 23a4c390489f..d074673a6d9b 100644
--- a/include/asm-s390/string.h
+++ b/include/asm-s390/string.h
@@ -60,12 +60,13 @@ static inline void *memchr(const void * s, int c, size_t n)
 	register int r0 asm("0") = (char) c;
 	const void *ret = s + n;
 
-	asm volatile ("0: srst  %0,%1\n"
-		      "   jo    0b\n"
-		      "   jl	1f\n"
-		      "   la    %0,0\n"
-		      "1:"
-		      : "+a" (ret), "+&a" (s) : "d" (r0) : "cc" );
+	asm volatile(
+		"0:	srst	%0,%1\n"
+		"	jo	0b\n"
+		"	jl	1f\n"
+		"	la	%0,0\n"
+		"1:"
+		: "+a" (ret), "+&a" (s) : "d" (r0) : "cc");
 	return (void *) ret;
 }
 
@@ -74,9 +75,10 @@ static inline void *memscan(void *s, int c, size_t n)
 	register int r0 asm("0") = (char) c;
 	const void *ret = s + n;
 
-	asm volatile ("0: srst  %0,%1\n"
-		      "   jo    0b\n"
-		      : "+a" (ret), "+&a" (s) : "d" (r0) : "cc" );
+	asm volatile(
+		"0:	srst	%0,%1\n"
+		"	jo	0b\n"
+		: "+a" (ret), "+&a" (s) : "d" (r0) : "cc");
 	return (void *) ret;
 }
 
@@ -86,12 +88,13 @@ static inline char *strcat(char *dst, const char *src)
 	unsigned long dummy;
 	char *ret = dst;
 
-	asm volatile ("0: srst  %0,%1\n"
-		      "   jo    0b\n"
-		      "1: mvst  %0,%2\n"
-		      "   jo    1b"
-		      : "=&a" (dummy), "+a" (dst), "+a" (src)
-		      : "d" (r0), "0" (0) : "cc", "memory" );
+	asm volatile(
+		"0:	srst	%0,%1\n"
+		"	jo	0b\n"
+		"1:	mvst	%0,%2\n"
+		"	jo	1b"
+		: "=&a" (dummy), "+a" (dst), "+a" (src)
+		: "d" (r0), "0" (0) : "cc", "memory" );
 	return ret;
 }
 
@@ -100,10 +103,11 @@ static inline char *strcpy(char *dst, const char *src)
 	register int r0 asm("0") = 0;
 	char *ret = dst;
 
-	asm volatile ("0: mvst  %0,%1\n"
-		      "   jo    0b"
-		      : "+&a" (dst), "+&a" (src) : "d" (r0)
-		      : "cc", "memory" );
+	asm volatile(
+		"0:	mvst	%0,%1\n"
+		"	jo	0b"
+		: "+&a" (dst), "+&a" (src) : "d" (r0)
+		: "cc", "memory");
 	return ret;
 }
 
@@ -112,9 +116,10 @@ static inline size_t strlen(const char *s)
 	register unsigned long r0 asm("0") = 0;
 	const char *tmp = s;
 
-	asm volatile ("0: srst  %0,%1\n"
-		      "   jo    0b"
-		      : "+d" (r0), "+a" (tmp) :  : "cc" );
+	asm volatile(
+		"0:	srst	%0,%1\n"
+		"	jo	0b"
+		: "+d" (r0), "+a" (tmp) :  : "cc");
 	return r0 - (unsigned long) s;
 }
 
@@ -124,9 +129,10 @@ static inline size_t strnlen(const char * s, size_t n)
 	const char *tmp = s;
 	const char *end = s + n;
 
-	asm volatile ("0: srst  %0,%1\n"
-		      "   jo    0b"
-		      : "+a" (end), "+a" (tmp) : "d" (r0)  : "cc" );
+	asm volatile(
+		"0:	srst	%0,%1\n"
+		"	jo	0b"
+		: "+a" (end), "+a" (tmp) : "d" (r0)  : "cc");
 	return end - s;
 }
 
diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h
index 16040048cd1b..ccbafe4bf2cb 100644
--- a/include/asm-s390/system.h
+++ b/include/asm-s390/system.h
@@ -23,74 +23,68 @@ struct task_struct;
 
 extern struct task_struct *__switch_to(void *, void *);
 
-#ifdef __s390x__
-#define __FLAG_SHIFT 56
-#else /* ! __s390x__ */
-#define __FLAG_SHIFT 24
-#endif /* ! __s390x__ */
-
 static inline void save_fp_regs(s390_fp_regs *fpregs)
 {
-	asm volatile (
-		"   std   0,8(%1)\n"
-		"   std   2,24(%1)\n"
-		"   std   4,40(%1)\n"
-		"   std   6,56(%1)"
-		: "=m" (*fpregs) : "a" (fpregs), "m" (*fpregs) : "memory" );
+	asm volatile(
+		"	std	0,8(%1)\n"
+		"	std	2,24(%1)\n"
+		"	std	4,40(%1)\n"
+		"	std	6,56(%1)"
+		: "=m" (*fpregs) : "a" (fpregs), "m" (*fpregs) : "memory");
 	if (!MACHINE_HAS_IEEE)
 		return;
 	asm volatile(
-		"   stfpc 0(%1)\n"
-		"   std   1,16(%1)\n"
-		"   std   3,32(%1)\n"
-		"   std   5,48(%1)\n"
-		"   std   7,64(%1)\n"
-		"   std   8,72(%1)\n"
-		"   std   9,80(%1)\n"
-		"   std   10,88(%1)\n"
-		"   std   11,96(%1)\n"
-		"   std   12,104(%1)\n"
-		"   std   13,112(%1)\n"
-		"   std   14,120(%1)\n"
-		"   std   15,128(%1)\n"
-		: "=m" (*fpregs) : "a" (fpregs), "m" (*fpregs) : "memory" );
+		"	stfpc	0(%1)\n"
+		"	std	1,16(%1)\n"
+		"	std	3,32(%1)\n"
+		"	std	5,48(%1)\n"
+		"	std	7,64(%1)\n"
+		"	std	8,72(%1)\n"
+		"	std	9,80(%1)\n"
+		"	std	10,88(%1)\n"
+		"	std	11,96(%1)\n"
+		"	std	12,104(%1)\n"
+		"	std	13,112(%1)\n"
+		"	std	14,120(%1)\n"
+		"	std	15,128(%1)\n"
+		: "=m" (*fpregs) : "a" (fpregs), "m" (*fpregs) : "memory");
 }
 
 static inline void restore_fp_regs(s390_fp_regs *fpregs)
 {
-	asm volatile (
-		"   ld    0,8(%0)\n"
-		"   ld    2,24(%0)\n"
-		"   ld    4,40(%0)\n"
-		"   ld    6,56(%0)"
-		: : "a" (fpregs), "m" (*fpregs) );
+	asm volatile(
+		"	ld	0,8(%0)\n"
+		"	ld	2,24(%0)\n"
+		"	ld	4,40(%0)\n"
+		"	ld	6,56(%0)"
+		: : "a" (fpregs), "m" (*fpregs));
 	if (!MACHINE_HAS_IEEE)
 		return;
 	asm volatile(
-		"   lfpc  0(%0)\n"
-		"   ld    1,16(%0)\n"
-		"   ld    3,32(%0)\n"
-		"   ld    5,48(%0)\n"
-		"   ld    7,64(%0)\n"
-		"   ld    8,72(%0)\n"
-		"   ld    9,80(%0)\n"
-		"   ld    10,88(%0)\n"
-		"   ld    11,96(%0)\n"
-		"   ld    12,104(%0)\n"
-		"   ld    13,112(%0)\n"
-		"   ld    14,120(%0)\n"
-		"   ld    15,128(%0)\n"
-		: : "a" (fpregs), "m" (*fpregs) );
+		"	lfpc	0(%0)\n"
+		"	ld	1,16(%0)\n"
+		"	ld	3,32(%0)\n"
+		"	ld	5,48(%0)\n"
+		"	ld	7,64(%0)\n"
+		"	ld	8,72(%0)\n"
+		"	ld	9,80(%0)\n"
+		"	ld	10,88(%0)\n"
+		"	ld	11,96(%0)\n"
+		"	ld	12,104(%0)\n"
+		"	ld	13,112(%0)\n"
+		"	ld	14,120(%0)\n"
+		"	ld	15,128(%0)\n"
+		: : "a" (fpregs), "m" (*fpregs));
 }
 
 static inline void save_access_regs(unsigned int *acrs)
 {
-	asm volatile ("stam 0,15,0(%0)" : : "a" (acrs) : "memory" );
+	asm volatile("stam 0,15,0(%0)" : : "a" (acrs) : "memory");
 }
 
 static inline void restore_access_regs(unsigned int *acrs)
 {
-	asm volatile ("lam 0,15,0(%0)" : : "a" (acrs) );
+	asm volatile("lam 0,15,0(%0)" : : "a" (acrs));
 }
 
 #define switch_to(prev,next,last) do {					     \
@@ -126,7 +120,7 @@ extern void account_system_vtime(struct task_struct *);
 	account_vtime(prev);						     \
 } while (0)
 
-#define nop() __asm__ __volatile__ ("nop")
+#define nop() asm volatile("nop")
 
 #define xchg(ptr,x)							  \
 ({									  \
@@ -147,15 +141,15 @@ static inline unsigned long __xchg(unsigned long x, void * ptr, int size)
 		shift = (3 ^ (addr & 3)) << 3;
 		addr ^= addr & 3;
 		asm volatile(
-			"    l   %0,0(%4)\n"
-			"0:  lr  0,%0\n"
-			"    nr  0,%3\n"
-			"    or  0,%2\n"
-			"    cs  %0,0,0(%4)\n"
-			"    jl  0b\n"
+			"	l	%0,0(%4)\n"
+			"0:	lr	0,%0\n"
+			"	nr	0,%3\n"
+			"	or	0,%2\n"
+			"	cs	%0,0,0(%4)\n"
+			"	jl	0b\n"
 			: "=&d" (old), "=m" (*(int *) addr)
 			: "d" (x << shift), "d" (~(255 << shift)), "a" (addr),
-			  "m" (*(int *) addr) : "memory", "cc", "0" );
+			  "m" (*(int *) addr) : "memory", "cc", "0");
 		x = old >> shift;
 		break;
 	case 2:
@@ -163,36 +157,36 @@ static inline unsigned long __xchg(unsigned long x, void * ptr, int size)
 		shift = (2 ^ (addr & 2)) << 3;
 		addr ^= addr & 2;
 		asm volatile(
-			"    l   %0,0(%4)\n"
-			"0:  lr  0,%0\n"
-			"    nr  0,%3\n"
-			"    or  0,%2\n"
-			"    cs  %0,0,0(%4)\n"
-			"    jl  0b\n"
+			"	l	%0,0(%4)\n"
+			"0:	lr	0,%0\n"
+			"	nr	0,%3\n"
+			"	or	0,%2\n"
+			"	cs	%0,0,0(%4)\n"
+			"	jl	0b\n"
 			: "=&d" (old), "=m" (*(int *) addr)
 			: "d" (x << shift), "d" (~(65535 << shift)), "a" (addr),
-			  "m" (*(int *) addr) : "memory", "cc", "0" );
+			  "m" (*(int *) addr) : "memory", "cc", "0");
 		x = old >> shift;
 		break;
 	case 4:
-		asm volatile (
-			"    l   %0,0(%3)\n"
-			"0:  cs  %0,%2,0(%3)\n"
-			"    jl  0b\n"
+		asm volatile(
+			"	l	%0,0(%3)\n"
+			"0:	cs	%0,%2,0(%3)\n"
+			"	jl	0b\n"
 			: "=&d" (old), "=m" (*(int *) ptr)
 			: "d" (x), "a" (ptr), "m" (*(int *) ptr)
-			: "memory", "cc" );
+			: "memory", "cc");
 		x = old;
 		break;
 #ifdef __s390x__
 	case 8:
-		asm volatile (
-			"    lg  %0,0(%3)\n"
-			"0:  csg %0,%2,0(%3)\n"
-			"    jl  0b\n"
+		asm volatile(
+			"	lg	%0,0(%3)\n"
+			"0:	csg	%0,%2,0(%3)\n"
+			"	jl	0b\n"
 			: "=&d" (old), "=m" (*(long *) ptr)
 			: "d" (x), "a" (ptr), "m" (*(long *) ptr)
-			: "memory", "cc" );
+			: "memory", "cc");
 		x = old;
 		break;
 #endif /* __s390x__ */
@@ -224,55 +218,55 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 		shift = (3 ^ (addr & 3)) << 3;
 		addr ^= addr & 3;
 		asm volatile(
-			"    l   %0,0(%4)\n"
-			"0:  nr  %0,%5\n"
-                        "    lr  %1,%0\n"
-			"    or  %0,%2\n"
-			"    or  %1,%3\n"
-			"    cs  %0,%1,0(%4)\n"
-			"    jnl 1f\n"
-			"    xr  %1,%0\n"
-			"    nr  %1,%5\n"
-			"    jnz 0b\n"
+			"	l	%0,0(%4)\n"
+			"0:	nr	%0,%5\n"
+			"	lr	%1,%0\n"
+			"	or	%0,%2\n"
+			"	or	%1,%3\n"
+			"	cs	%0,%1,0(%4)\n"
+			"	jnl	1f\n"
+			"	xr	%1,%0\n"
+			"	nr	%1,%5\n"
+			"	jnz	0b\n"
 			"1:"
 			: "=&d" (prev), "=&d" (tmp)
 			: "d" (old << shift), "d" (new << shift), "a" (ptr),
 			  "d" (~(255 << shift))
-			: "memory", "cc" );
+			: "memory", "cc");
 		return prev >> shift;
 	case 2:
 		addr = (unsigned long) ptr;
 		shift = (2 ^ (addr & 2)) << 3;
 		addr ^= addr & 2;
 		asm volatile(
-			"    l   %0,0(%4)\n"
-			"0:  nr  %0,%5\n"
-                        "    lr  %1,%0\n"
-			"    or  %0,%2\n"
-			"    or  %1,%3\n"
-			"    cs  %0,%1,0(%4)\n"
-			"    jnl 1f\n"
-			"    xr  %1,%0\n"
-			"    nr  %1,%5\n"
-			"    jnz 0b\n"
+			"	l	%0,0(%4)\n"
+			"0:	nr	%0,%5\n"
+			"	lr	%1,%0\n"
+			"	or	%0,%2\n"
+			"	or	%1,%3\n"
+			"	cs	%0,%1,0(%4)\n"
+			"	jnl	1f\n"
+			"	xr	%1,%0\n"
+			"	nr	%1,%5\n"
+			"	jnz	0b\n"
 			"1:"
 			: "=&d" (prev), "=&d" (tmp)
 			: "d" (old << shift), "d" (new << shift), "a" (ptr),
 			  "d" (~(65535 << shift))
-			: "memory", "cc" );
+			: "memory", "cc");
 		return prev >> shift;
 	case 4:
-		asm volatile (
-			"    cs  %0,%2,0(%3)\n"
+		asm volatile(
+			"	cs	%0,%2,0(%3)\n"
 			: "=&d" (prev) : "0" (old), "d" (new), "a" (ptr)
-			: "memory", "cc" );
+			: "memory", "cc");
 		return prev;
 #ifdef __s390x__
 	case 8:
-		asm volatile (
-			"    csg %0,%2,0(%3)\n"
+		asm volatile(
+			"	csg	%0,%2,0(%3)\n"
 			: "=&d" (prev) : "0" (old), "d" (new), "a" (ptr)
-			: "memory", "cc" );
+			: "memory", "cc");
 		return prev;
 #endif /* __s390x__ */
         }
@@ -289,8 +283,8 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
  * all memory ops have completed wrt other CPU's ( see 7-15 POP  DJB ).
  */
 
-#define eieio()  __asm__ __volatile__ ( "bcr 15,0" : : : "memory" ) 
-# define SYNC_OTHER_CORES(x)   eieio() 
+#define eieio()	asm volatile("bcr 15,0" : : : "memory")
+#define SYNC_OTHER_CORES(x)   eieio()
 #define mb()    eieio()
 #define rmb()   eieio()
 #define wmb()   eieio()
@@ -307,117 +301,56 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 
 #ifdef __s390x__
 
-#define __ctl_load(array, low, high) ({ \
-	typedef struct { char _[sizeof(array)]; } addrtype; \
-	__asm__ __volatile__ ( \
-		"   bras  1,0f\n" \
-                "   lctlg 0,0,0(%0)\n" \
-		"0: ex    %1,0(1)" \
-		: : "a" (&array), "a" (((low)<<4)+(high)), \
-		    "m" (*(addrtype *)(array)) : "1" ); \
+#define __ctl_load(array, low, high) ({				\
+	typedef struct { char _[sizeof(array)]; } addrtype;	\
+	asm volatile(						\
+		"	lctlg	%1,%2,0(%0)\n"			\
+		: : "a" (&array), "i" (low), "i" (high),	\
+		    "m" (*(addrtype *)(array)));		\
 	})
 
-#define __ctl_store(array, low, high) ({ \
-	typedef struct { char _[sizeof(array)]; } addrtype; \
-	__asm__ __volatile__ ( \
-		"   bras  1,0f\n" \
-		"   stctg 0,0,0(%1)\n" \
-		"0: ex    %2,0(1)" \
-		: "=m" (*(addrtype *)(array)) \
-		: "a" (&array), "a" (((low)<<4)+(high)) : "1" ); \
+#define __ctl_store(array, low, high) ({			\
+	typedef struct { char _[sizeof(array)]; } addrtype;	\
+	asm volatile(						\
+		"	stctg	%2,%3,0(%1)\n"			\
+		: "=m" (*(addrtype *)(array))			\
+		: "a" (&array), "i" (low), "i" (high));		\
 	})
 
-#define __ctl_set_bit(cr, bit) ({ \
-        __u8 __dummy[24]; \
-        __asm__ __volatile__ ( \
-                "    bras  1,0f\n"       /* skip indirect insns */ \
-                "    stctg 0,0,0(%1)\n" \
-                "    lctlg 0,0,0(%1)\n" \
-                "0:  ex    %2,0(1)\n"    /* execute stctl */ \
-                "    lg    0,0(%1)\n" \
-                "    ogr   0,%3\n"       /* set the bit */ \
-                "    stg   0,0(%1)\n" \
-                "1:  ex    %2,6(1)"      /* execute lctl */ \
-                : "=m" (__dummy) \
-		: "a" ((((unsigned long) &__dummy) + 7) & ~7UL), \
-		  "a" (cr*17), "a" (1L<<(bit)) \
-                : "cc", "0", "1" ); \
-        })
-
-#define __ctl_clear_bit(cr, bit) ({ \
-        __u8 __dummy[16]; \
-        __asm__ __volatile__ ( \
-                "    bras  1,0f\n"       /* skip indirect insns */ \
-                "    stctg 0,0,0(%1)\n" \
-                "    lctlg 0,0,0(%1)\n" \
-                "0:  ex    %2,0(1)\n"    /* execute stctl */ \
-                "    lg    0,0(%1)\n" \
-                "    ngr   0,%3\n"       /* set the bit */ \
-                "    stg   0,0(%1)\n" \
-                "1:  ex    %2,6(1)"      /* execute lctl */ \
-                : "=m" (__dummy) \
-		: "a" ((((unsigned long) &__dummy) + 7) & ~7UL), \
-		  "a" (cr*17), "a" (~(1L<<(bit))) \
-                : "cc", "0", "1" ); \
-        })
-
 #else /* __s390x__ */
 
-#define __ctl_load(array, low, high) ({ \
-	typedef struct { char _[sizeof(array)]; } addrtype; \
-	__asm__ __volatile__ ( \
-		"   bras  1,0f\n" \
-                "   lctl 0,0,0(%0)\n" \
-		"0: ex    %1,0(1)" \
-		: : "a" (&array), "a" (((low)<<4)+(high)), \
-		    "m" (*(addrtype *)(array)) : "1" ); \
-	})
+#define __ctl_load(array, low, high) ({				\
+	typedef struct { char _[sizeof(array)]; } addrtype;	\
+	asm volatile(						\
+		"	lctl	%1,%2,0(%0)\n"			\
+		: : "a" (&array), "i" (low), "i" (high),	\
+		    "m" (*(addrtype *)(array)));		\
+})
 
-#define __ctl_store(array, low, high) ({ \
-	typedef struct { char _[sizeof(array)]; } addrtype; \
-	__asm__ __volatile__ ( \
-		"   bras  1,0f\n" \
-		"   stctl 0,0,0(%1)\n" \
-		"0: ex    %2,0(1)" \
-		: "=m" (*(addrtype *)(array)) \
-		: "a" (&array), "a" (((low)<<4)+(high)): "1" ); \
+#define __ctl_store(array, low, high) ({			\
+	typedef struct { char _[sizeof(array)]; } addrtype;	\
+	asm volatile(						\
+		"	stctl	%2,%3,0(%1)\n"			\
+		: "=m" (*(addrtype *)(array))			\
+		: "a" (&array), "i" (low), "i" (high));		\
 	})
 
-#define __ctl_set_bit(cr, bit) ({ \
-        __u8 __dummy[16]; \
-        __asm__ __volatile__ ( \
-                "    bras  1,0f\n"       /* skip indirect insns */ \
-                "    stctl 0,0,0(%1)\n" \
-                "    lctl  0,0,0(%1)\n" \
-                "0:  ex    %2,0(1)\n"    /* execute stctl */ \
-                "    l     0,0(%1)\n" \
-                "    or    0,%3\n"       /* set the bit */ \
-                "    st    0,0(%1)\n" \
-                "1:  ex    %2,4(1)"      /* execute lctl */ \
-                : "=m" (__dummy) \
-		: "a" ((((unsigned long) &__dummy) + 7) & ~7UL), \
-		  "a" (cr*17), "a" (1<<(bit)) \
-                : "cc", "0", "1" ); \
-        })
-
-#define __ctl_clear_bit(cr, bit) ({ \
-        __u8 __dummy[16]; \
-        __asm__ __volatile__ ( \
-                "    bras  1,0f\n"       /* skip indirect insns */ \
-                "    stctl 0,0,0(%1)\n" \
-                "    lctl  0,0,0(%1)\n" \
-                "0:  ex    %2,0(1)\n"    /* execute stctl */ \
-                "    l     0,0(%1)\n" \
-                "    nr    0,%3\n"       /* set the bit */ \
-                "    st    0,0(%1)\n" \
-                "1:  ex    %2,4(1)"      /* execute lctl */ \
-                : "=m" (__dummy) \
-		: "a" ((((unsigned long) &__dummy) + 7) & ~7UL), \
-		  "a" (cr*17), "a" (~(1<<(bit))) \
-                : "cc", "0", "1" ); \
-        })
 #endif /* __s390x__ */
 
+#define __ctl_set_bit(cr, bit) ({	\
+	unsigned long __dummy;		\
+	__ctl_store(__dummy, cr, cr);	\
+	__dummy |= 1UL << (bit);	\
+	__ctl_load(__dummy, cr, cr);	\
+})
+
+#define __ctl_clear_bit(cr, bit) ({	\
+	unsigned long __dummy;		\
+	__ctl_store(__dummy, cr, cr);	\
+	__dummy &= ~(1UL << (bit));	\
+	__ctl_load(__dummy, cr, cr);	\
+})
+
 #include <linux/irqflags.h>
 
 /*
@@ -427,8 +360,7 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 static inline void
 __set_psw_mask(unsigned long mask)
 {
-	local_save_flags(mask);
-	__load_psw_mask(mask);
+	__load_psw_mask(mask | (__raw_local_irq_stosm(0x00) & ~(-1UL >> 8)));
 }
 
 #define local_mcck_enable()  __set_psw_mask(PSW_KERNEL_BITS)
diff --git a/include/asm-s390/timex.h b/include/asm-s390/timex.h
index 5d0332a4c2bd..4df4a41029a3 100644
--- a/include/asm-s390/timex.h
+++ b/include/asm-s390/timex.h
@@ -15,20 +15,21 @@
 
 typedef unsigned long long cycles_t;
 
-static inline cycles_t get_cycles(void)
-{
-	cycles_t cycles;
-
-	__asm__ __volatile__ ("stck 0(%1)" : "=m" (cycles) : "a" (&cycles) : "cc");
-	return cycles >> 2;
-}
-
 static inline unsigned long long get_clock (void)
 {
 	unsigned long long clk;
 
-	__asm__ __volatile__ ("stck 0(%1)" : "=m" (clk) : "a" (&clk) : "cc");
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+	asm volatile("stck %0" : "=Q" (clk) : : "cc");
+#else /* __GNUC__ */
+	asm volatile("stck 0(%1)" : "=m" (clk) : "a" (&clk) : "cc");
+#endif /* __GNUC__ */
 	return clk;
 }
 
+static inline cycles_t get_cycles(void)
+{
+	return (cycles_t) get_clock() >> 2;
+}
+
 #endif
diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h
index 73cd85bebfb2..fa4dc916a9bf 100644
--- a/include/asm-s390/tlbflush.h
+++ b/include/asm-s390/tlbflush.h
@@ -25,7 +25,7 @@
  */
 
 #define local_flush_tlb() \
-do {  __asm__ __volatile__("ptlb": : :"memory"); } while (0)
+do {  asm volatile("ptlb": : :"memory"); } while (0)
 
 #ifndef CONFIG_SMP
 
@@ -68,24 +68,24 @@ extern void smp_ptlb_all(void);
 
 static inline void global_flush_tlb(void)
 {
+	register unsigned long reg2 asm("2");
+	register unsigned long reg3 asm("3");
+	register unsigned long reg4 asm("4");
+	long dummy;
+
 #ifndef __s390x__
 	if (!MACHINE_HAS_CSP) {
 		smp_ptlb_all();
 		return;
 	}
 #endif /* __s390x__ */
-	{
-		register unsigned long addr asm("4");
-		long dummy;
-
-		dummy = 0;
-		addr = ((unsigned long) &dummy) + 1;
-		__asm__ __volatile__ (
-			"    slr  2,2\n"
-			"    slr  3,3\n"
-			"    csp  2,%0"
-			: : "a" (addr), "m" (dummy) : "cc", "2", "3" );
-	}
+
+	dummy = 0;
+	reg2 = reg3 = 0;
+	reg4 = ((unsigned long) &dummy) + 1;
+	asm volatile(
+		"	csp	%0,%2"
+		: : "d" (reg2), "d" (reg3), "d" (reg4), "m" (dummy) : "cc" );
 }
 
 /*
@@ -102,9 +102,9 @@ static inline void __flush_tlb_mm(struct mm_struct * mm)
 	if (unlikely(cpus_empty(mm->cpu_vm_mask)))
 		return;
 	if (MACHINE_HAS_IDTE) {
-		asm volatile (".insn rrf,0xb98e0000,0,%0,%1,0"
-			      : : "a" (2048),
-			      "a" (__pa(mm->pgd)&PAGE_MASK) : "cc" );
+		asm volatile(
+			"	.insn	rrf,0xb98e0000,0,%0,%1,0"
+			: : "a" (2048), "a" (__pa(mm->pgd)&PAGE_MASK) : "cc");
 		return;
 	}
 	preempt_disable();
diff --git a/include/asm-s390/uaccess.h b/include/asm-s390/uaccess.h
index e2047b0c9092..72ae4efddb49 100644
--- a/include/asm-s390/uaccess.h
+++ b/include/asm-s390/uaccess.h
@@ -38,25 +38,14 @@
 #define get_ds()        (KERNEL_DS)
 #define get_fs()        (current->thread.mm_segment)
 
-#ifdef __s390x__
 #define set_fs(x) \
 ({									\
 	unsigned long __pto;						\
 	current->thread.mm_segment = (x);				\
 	__pto = current->thread.mm_segment.ar4 ?			\
 		S390_lowcore.user_asce : S390_lowcore.kernel_asce;	\
-	asm volatile ("lctlg 7,7,%0" : : "m" (__pto) );			\
+	__ctl_load(__pto, 7, 7);					\
 })
-#else /* __s390x__ */
-#define set_fs(x) \
-({									\
-	unsigned long __pto;						\
-	current->thread.mm_segment = (x);				\
-	__pto = current->thread.mm_segment.ar4 ?			\
-		S390_lowcore.user_asce : S390_lowcore.kernel_asce;	\
-	asm volatile ("lctl  7,7,%0" : : "m" (__pto) );			\
-})
-#endif /* __s390x__ */
 
 #define segment_eq(a,b) ((a).ar4 == (b).ar4)
 
diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index d49c54cb5505..0361ac5dcde3 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -355,145 +355,145 @@ do {							     \
 
 #define _svc_clobber "1", "cc", "memory"
 
-#define _syscall0(type,name)				     \
-type name(void) {					     \
-	register long __svcres asm("2");		     \
-	long __res;					     \
-	__asm__ __volatile__ (				     \
-		"    .if %1 < 256\n"			     \
-		"    svc %b1\n"				     \
-		"    .else\n"				     \
-		"    la  %%r1,%1\n"			     \
-		"    svc 0\n"				     \
-		"    .endif"				     \
-		: "=d" (__svcres)			     \
-		: "i" (__NR_##name)			     \
-		: _svc_clobber );			     \
-	__res = __svcres;				     \
-	__syscall_return(type,__res);			     \
+#define _syscall0(type,name)					\
+type name(void) {						\
+	register long __svcres asm("2");			\
+	long __res;						\
+	asm volatile(						\
+		"	.if	%1 < 256\n"			\
+		"	svc	%b1\n"				\
+		"	.else\n"				\
+		"	la	%%r1,%1\n"			\
+		"	svc	0\n"				\
+		"	.endif"					\
+		: "=d" (__svcres)				\
+		: "i" (__NR_##name)				\
+		: _svc_clobber);				\
+	__res = __svcres;					\
+	__syscall_return(type,__res);				\
 }
 
-#define _syscall1(type,name,type1,arg1)			     \
-type name(type1 arg1) {					     \
-	register type1 __arg1 asm("2") = arg1;		     \
-	register long __svcres asm("2");		     \
-	long __res;					     \
-	__asm__ __volatile__ (				     \
-		"    .if %1 < 256\n"			     \
-		"    svc %b1\n"				     \
-		"    .else\n"				     \
-		"    la  %%r1,%1\n"			     \
-		"    svc 0\n"				     \
-		"    .endif"				     \
-		: "=d" (__svcres)			     \
-		: "i" (__NR_##name),			     \
-		  "0" (__arg1)				     \
-		: _svc_clobber );			     \
-	__res = __svcres;				     \
-	__syscall_return(type,__res);			     \
+#define _syscall1(type,name,type1,arg1)				\
+type name(type1 arg1) {						\
+	register type1 __arg1 asm("2") = arg1;			\
+	register long __svcres asm("2");			\
+	long __res;						\
+	asm volatile(						\
+		"	.if	%1 < 256\n"			\
+		"	svc	%b1\n"				\
+		"	.else\n"				\
+		"	la	%%r1,%1\n"			\
+		"	svc	0\n"				\
+		"	.endif"					\
+		: "=d" (__svcres)				\
+		: "i" (__NR_##name),				\
+		  "0" (__arg1)					\
+		: _svc_clobber);				\
+	__res = __svcres;					\
+	__syscall_return(type,__res);				\
 }
 
-#define _syscall2(type,name,type1,arg1,type2,arg2)	     \
-type name(type1 arg1, type2 arg2) {			     \
-	register type1 __arg1 asm("2") = arg1;		     \
-	register type2 __arg2 asm("3") = arg2;		     \
-	register long __svcres asm("2");		     \
-	long __res;					     \
-	__asm__ __volatile__ (				     \
-		"    .if %1 < 256\n"			     \
-		"    svc %b1\n"				     \
-		"    .else\n"				     \
-		"    la %%r1,%1\n"			     \
-		"    svc 0\n"				     \
-		"    .endif"				     \
-		: "=d" (__svcres)			     \
-		: "i" (__NR_##name),			     \
-		  "0" (__arg1),				     \
-		  "d" (__arg2)				     \
-		: _svc_clobber );			     \
-	__res = __svcres;				     \
-	__syscall_return(type,__res);			     \
+#define _syscall2(type,name,type1,arg1,type2,arg2)		\
+type name(type1 arg1, type2 arg2) {				\
+	register type1 __arg1 asm("2") = arg1;			\
+	register type2 __arg2 asm("3") = arg2;			\
+	register long __svcres asm("2");			\
+	long __res;						\
+	asm volatile(						\
+		"	.if	%1 < 256\n"			\
+		"	svc	%b1\n"				\
+		"	.else\n"				\
+		"	la	%%r1,%1\n"			\
+		"	svc	0\n"				\
+		"	.endif"					\
+		: "=d" (__svcres)				\
+		: "i" (__NR_##name),				\
+		  "0" (__arg1),					\
+		  "d" (__arg2)					\
+		: _svc_clobber );				\
+	__res = __svcres;					\
+	__syscall_return(type,__res);				\
 }
 
-#define _syscall3(type,name,type1,arg1,type2,arg2,type3,arg3)\
-type name(type1 arg1, type2 arg2, type3 arg3) {		     \
-	register type1 __arg1 asm("2") = arg1;		     \
-	register type2 __arg2 asm("3") = arg2;		     \
-	register type3 __arg3 asm("4") = arg3;		     \
-	register long __svcres asm("2");		     \
-	long __res;					     \
-	__asm__ __volatile__ (				     \
-		"    .if %1 < 256\n"			     \
-		"    svc %b1\n"				     \
-		"    .else\n"				     \
-		"    la  %%r1,%1\n"			     \
-		"    svc 0\n"				     \
-		"    .endif"				     \
-		: "=d" (__svcres)			     \
-		: "i" (__NR_##name),			     \
-		  "0" (__arg1),				     \
-		  "d" (__arg2),				     \
-		  "d" (__arg3)				     \
-		: _svc_clobber );			     \
-	__res = __svcres;				     \
-	__syscall_return(type,__res);			     \
+#define _syscall3(type,name,type1,arg1,type2,arg2,type3,arg3)	\
+type name(type1 arg1, type2 arg2, type3 arg3) {			\
+	register type1 __arg1 asm("2") = arg1;			\
+	register type2 __arg2 asm("3") = arg2;			\
+	register type3 __arg3 asm("4") = arg3;			\
+	register long __svcres asm("2");			\
+	long __res;						\
+	asm volatile(						\
+		"	.if	%1 < 256\n"			\
+		"	svc	%b1\n"				\
+		"	.else\n"				\
+		"	la	%%r1,%1\n"			\
+		"	svc	0\n"				\
+		"	.endif"					\
+		: "=d" (__svcres)				\
+		: "i" (__NR_##name),				\
+		  "0" (__arg1),					\
+		  "d" (__arg2),					\
+		  "d" (__arg3)					\
+		: _svc_clobber);				\
+	__res = __svcres;					\
+	__syscall_return(type,__res);				\
 }
 
-#define _syscall4(type,name,type1,arg1,type2,arg2,type3,arg3,\
-		  type4,name4)				     \
-type name(type1 arg1, type2 arg2, type3 arg3, type4 arg4) {  \
-	register type1 __arg1 asm("2") = arg1;		     \
-	register type2 __arg2 asm("3") = arg2;		     \
-	register type3 __arg3 asm("4") = arg3;		     \
-	register type4 __arg4 asm("5") = arg4;		     \
-	register long __svcres asm("2");		     \
-	long __res;					     \
-	__asm__ __volatile__ (				     \
-		"    .if %1 < 256\n"			     \
-		"    svc %b1\n"				     \
-		"    .else\n"				     \
-		"    la  %%r1,%1\n"			     \
-		"    svc 0\n"				     \
-		"    .endif"				     \
-		: "=d" (__svcres)			     \
-		: "i" (__NR_##name),			     \
-		  "0" (__arg1),				     \
-		  "d" (__arg2),				     \
-		  "d" (__arg3),				     \
-		  "d" (__arg4)				     \
-		: _svc_clobber );			     \
-	__res = __svcres;				     \
-	__syscall_return(type,__res);			     \
+#define _syscall4(type,name,type1,arg1,type2,arg2,type3,arg3,	\
+		  type4,name4)					\
+type name(type1 arg1, type2 arg2, type3 arg3, type4 arg4) {	\
+	register type1 __arg1 asm("2") = arg1;			\
+	register type2 __arg2 asm("3") = arg2;			\
+	register type3 __arg3 asm("4") = arg3;			\
+	register type4 __arg4 asm("5") = arg4;			\
+	register long __svcres asm("2");			\
+	long __res;						\
+	asm volatile(						\
+		"	.if	%1 < 256\n"			\
+		"	svc	%b1\n"				\
+		"	.else\n"				\
+		"	la	%%r1,%1\n"			\
+		"	svc	0\n"				\
+		"	.endif"					\
+		: "=d" (__svcres)				\
+		: "i" (__NR_##name),				\
+		  "0" (__arg1),					\
+		  "d" (__arg2),					\
+		  "d" (__arg3),					\
+		  "d" (__arg4)					\
+		: _svc_clobber);				\
+	__res = __svcres;					\
+	__syscall_return(type,__res);				\
 }
 
-#define _syscall5(type,name,type1,arg1,type2,arg2,type3,arg3,\
-		  type4,name4,type5,name5)		     \
-type name(type1 arg1, type2 arg2, type3 arg3, type4 arg4,    \
-	  type5 arg5) {					     \
-	register type1 __arg1 asm("2") = arg1;		     \
-	register type2 __arg2 asm("3") = arg2;		     \
-	register type3 __arg3 asm("4") = arg3;		     \
-	register type4 __arg4 asm("5") = arg4;		     \
-	register type5 __arg5 asm("6") = arg5;		     \
-	register long __svcres asm("2");		     \
-	long __res;					     \
-	__asm__ __volatile__ (				     \
-		"    .if %1 < 256\n"			     \
-		"    svc %b1\n"				     \
-		"    .else\n"				     \
-		"    la  %%r1,%1\n"			     \
-		"    svc 0\n"				     \
-		"    .endif"				     \
-		: "=d" (__svcres)			     \
-		: "i" (__NR_##name),			     \
-		  "0" (__arg1),				     \
-		  "d" (__arg2),				     \
-		  "d" (__arg3),				     \
-		  "d" (__arg4),				     \
-		  "d" (__arg5)				     \
-		: _svc_clobber );			     \
-	__res = __svcres;				     \
-	__syscall_return(type,__res);			     \
+#define _syscall5(type,name,type1,arg1,type2,arg2,type3,arg3,	\
+		  type4,name4,type5,name5)			\
+type name(type1 arg1, type2 arg2, type3 arg3, type4 arg4,	\
+	  type5 arg5) {						\
+	register type1 __arg1 asm("2") = arg1;			\
+	register type2 __arg2 asm("3") = arg2;			\
+	register type3 __arg3 asm("4") = arg3;			\
+	register type4 __arg4 asm("5") = arg4;			\
+	register type5 __arg5 asm("6") = arg5;			\
+	register long __svcres asm("2");			\
+	long __res;						\
+	asm volatile(						\
+		"	.if	%1 < 256\n"			\
+		"	svc	%b1\n"				\
+		"	.else\n"				\
+		"	la	%%r1,%1\n"			\
+		"	svc	0\n"				\
+		"	.endif"					\
+		: "=d" (__svcres)				\
+		: "i" (__NR_##name),				\
+		  "0" (__arg1),					\
+		  "d" (__arg2),					\
+		  "d" (__arg3),					\
+		  "d" (__arg4),					\
+		  "d" (__arg5)					\
+		: _svc_clobber);				\
+	__res = __svcres;					\
+	__syscall_return(type,__res);				\
 }
 
 #define __ARCH_WANT_IPC_PARSE_VERSION
-- 
cgit v1.2.3


From 1f51c10c5e85050506663bce1d69513eb901db87 Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Thu, 28 Sep 2006 16:26:47 +0100
Subject: [ARM] 3870/1: AT91: Start removing static memory mappings

This patch removes the static memory mapping for the currently-unused
peripherals [Synchronous Serial, Timer/Counter unit], and for those
drivers that already ioremap() their registers [UART].

Also, the Ethernet driver now uses the platform_device resources but
doesn't yet use ioremap() so we need to pass it the virtual address
instead of the physical address.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91rm9200/at91rm9200.c      | 45 ------------------------------
 arch/arm/mach-at91rm9200/devices.c         |  4 +--
 include/asm-arm/arch-at91rm9200/hardware.h |  9 ------
 3 files changed, 2 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/arch/arm/mach-at91rm9200/at91rm9200.c b/arch/arm/mach-at91rm9200/at91rm9200.c
index c32d0c996f90..dcf6136fedf9 100644
--- a/arch/arm/mach-at91rm9200/at91rm9200.c
+++ b/arch/arm/mach-at91rm9200/at91rm9200.c
@@ -30,41 +30,6 @@ static struct map_desc at91rm9200_io_desc[] __initdata = {
 		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SPI),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_SSC2,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SSC2),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_SSC1,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SSC1),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_SSC0,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_SSC0),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_US3,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US3),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_US2,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US2),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_US1,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US1),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_US0,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_US0),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_VA_BASE_EMAC,
 		.pfn		= __phys_to_pfn(AT91RM9200_BASE_EMAC),
@@ -85,16 +50,6 @@ static struct map_desc at91rm9200_io_desc[] __initdata = {
 		.pfn		= __phys_to_pfn(AT91RM9200_BASE_UDP),
 		.length		= SZ_16K,
 		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_TCB1,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_TCB1),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
-	}, {
-		.virtual	= AT91_VA_BASE_TCB0,
-		.pfn		= __phys_to_pfn(AT91RM9200_BASE_TCB0),
-		.length		= SZ_16K,
-		.type		= MT_DEVICE,
 	}, {
 		.virtual	= AT91_SRAM_VIRT_BASE,
 		.pfn		= __phys_to_pfn(AT91RM9200_SRAM_BASE),
diff --git a/arch/arm/mach-at91rm9200/devices.c b/arch/arm/mach-at91rm9200/devices.c
index db7c684cdb63..01525530c287 100644
--- a/arch/arm/mach-at91rm9200/devices.c
+++ b/arch/arm/mach-at91rm9200/devices.c
@@ -131,8 +131,8 @@ static struct at91_eth_data eth_data;
 
 static struct resource at91_eth_resources[] = {
 	[0] = {
-		.start	= AT91RM9200_BASE_EMAC,
-		.end	= AT91RM9200_BASE_EMAC + SZ_16K - 1,
+		.start	= AT91_VA_BASE_EMAC,
+		.end	= AT91_VA_BASE_EMAC + SZ_16K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
diff --git a/include/asm-arm/arch-at91rm9200/hardware.h b/include/asm-arm/arch-at91rm9200/hardware.h
index 878e65f369bf..6551b4d1ff7b 100644
--- a/include/asm-arm/arch-at91rm9200/hardware.h
+++ b/include/asm-arm/arch-at91rm9200/hardware.h
@@ -35,19 +35,10 @@
  */
 #define AT91_VA_BASE_SYS	AT91_IO_P2V(AT91_BASE_SYS)
 #define AT91_VA_BASE_SPI	AT91_IO_P2V(AT91RM9200_BASE_SPI)
-#define AT91_VA_BASE_SSC2	AT91_IO_P2V(AT91RM9200_BASE_SSC2)
-#define AT91_VA_BASE_SSC1	AT91_IO_P2V(AT91RM9200_BASE_SSC1)
-#define AT91_VA_BASE_SSC0	AT91_IO_P2V(AT91RM9200_BASE_SSC0)
-#define AT91_VA_BASE_US3	AT91_IO_P2V(AT91RM9200_BASE_US3)
-#define AT91_VA_BASE_US2	AT91_IO_P2V(AT91RM9200_BASE_US2)
-#define AT91_VA_BASE_US1	AT91_IO_P2V(AT91RM9200_BASE_US1)
-#define AT91_VA_BASE_US0	AT91_IO_P2V(AT91RM9200_BASE_US0)
 #define AT91_VA_BASE_EMAC	AT91_IO_P2V(AT91RM9200_BASE_EMAC)
 #define AT91_VA_BASE_TWI	AT91_IO_P2V(AT91RM9200_BASE_TWI)
 #define AT91_VA_BASE_MCI	AT91_IO_P2V(AT91RM9200_BASE_MCI)
 #define AT91_VA_BASE_UDP	AT91_IO_P2V(AT91RM9200_BASE_UDP)
-#define AT91_VA_BASE_TCB1	AT91_IO_P2V(AT91RM9200_BASE_TCB1)
-#define AT91_VA_BASE_TCB0	AT91_IO_P2V(AT91RM9200_BASE_TCB0)
 
  /* Internal SRAM is mapped below the IO devices */
 #define AT91_SRAM_VIRT_BASE	(AT91_IO_VIRT_BASE - AT91RM9200_SRAM_SIZE)
-- 
cgit v1.2.3


From 746140c71d537560bbd22c1b148fb21031c30e71 Mon Sep 17 00:00:00 2001
From: Kevin Hilman <khilman@mvista.com>
Date: Fri, 22 Sep 2006 00:16:30 +0100
Subject: [ARM] 3855/1: Add generic time support

This patch adds Generic time-of-day support for the ARM architecture.

The support is currently added using #ifdef's so that it can support
sub-arches that do not (yet) have a clocksource added.  As sub-arches
add clocksource support, they should 'select GENERIC_TIME'

Signed-off-by: Deepak Saxena <dsaxena@mvista.com>
Signed-off-by: Daniel Walker <dwalker@mvista.com>

Signed-off-by: Kevin Hilman <khilman@mvista.com>

Acked-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig            | 4 ++++
 arch/arm/kernel/time.c      | 6 ++++++
 include/asm-arm/mach/time.h | 2 ++
 include/asm-arm/timeofday.h | 4 ++++
 4 files changed, 16 insertions(+)
 create mode 100644 include/asm-arm/timeofday.h

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 0810d27c039e..314ac6ba0ec4 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -17,6 +17,10 @@ config ARM
 	  Europe.  There is an ARM Linux project with a web page at
 	  <http://www.arm.linux.org.uk/>.
 
+config GENERIC_TIME
+	bool
+	default n
+
 config MMU
 	bool
 	default y
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index 09a67d771857..d4dceb5f06e9 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -69,10 +69,12 @@ EXPORT_SYMBOL(profile_pc);
  */
 int (*set_rtc)(void);
 
+#ifndef CONFIG_GENERIC_TIME
 static unsigned long dummy_gettimeoffset(void)
 {
 	return 0;
 }
+#endif
 
 /*
  * Scheduler clock - returns current time in nanosec units.
@@ -230,6 +232,7 @@ static inline void do_leds(void)
 #define	do_leds()
 #endif
 
+#ifndef CONFIG_GENERIC_TIME
 void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
@@ -291,6 +294,7 @@ int do_settimeofday(struct timespec *tv)
 }
 
 EXPORT_SYMBOL(do_settimeofday);
+#endif /* !CONFIG_GENERIC_TIME */
 
 /**
  * save_time_delta - Save the offset between system time and RTC time
@@ -500,8 +504,10 @@ device_initcall(timer_init_sysfs);
 
 void __init time_init(void)
 {
+#ifndef CONFIG_GENERIC_TIME
 	if (system_timer->offset == NULL)
 		system_timer->offset = dummy_gettimeoffset;
+#endif
 	system_timer->init();
 
 #ifdef CONFIG_NO_IDLE_HZ
diff --git a/include/asm-arm/mach/time.h b/include/asm-arm/mach/time.h
index dee0bc336fe8..1eb93f5c0d6c 100644
--- a/include/asm-arm/mach/time.h
+++ b/include/asm-arm/mach/time.h
@@ -38,7 +38,9 @@ struct sys_timer {
 	void			(*init)(void);
 	void			(*suspend)(void);
 	void			(*resume)(void);
+#ifndef CONFIG_GENERIC_TIME
 	unsigned long		(*offset)(void);
+#endif
 
 #ifdef CONFIG_NO_IDLE_HZ
 	struct dyn_tick_timer	*dyn_tick;
diff --git a/include/asm-arm/timeofday.h b/include/asm-arm/timeofday.h
new file mode 100644
index 000000000000..27254bd5b94f
--- /dev/null
+++ b/include/asm-arm/timeofday.h
@@ -0,0 +1,4 @@
+#ifndef _ASM_ARM_TIMEOFDAY_H
+#define _ASM_ARM_TIMEOFDAY_H
+#include <asm-generic/timeofday.h>
+#endif
-- 
cgit v1.2.3


From 84904d0ead0a8c419abd45c7b2ac8d76d50a0d48 Mon Sep 17 00:00:00 2001
From: Kevin Hilman <khilman@mvista.com>
Date: Fri, 22 Sep 2006 00:58:57 +0100
Subject: [ARM] 3856/1: Add clocksource for Intel IXP4xx platforms

Enables the ixp4xx platforms to use Generic time-of-day.

Signed-off-by: Kevin Hilman <khilman@mvista.com>
Acked-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig                       |  1 +
 arch/arm/mach-ixp4xx/common.c          | 38 ++++++++++++++++++++++++----------
 arch/arm/mach-ixp4xx/nslu2-setup.c     |  2 ++
 include/asm-arm/arch-ixp4xx/platform.h |  5 +++++
 4 files changed, 35 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 314ac6ba0ec4..147abe0383d8 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -226,6 +226,7 @@ config ARCH_IOP33X
 config ARCH_IXP4XX
 	bool "IXP4xx-based"
 	depends on MMU
+	select GENERIC_TIME
 	help
 	  Support for Intel's IXP4XX (XScale) family of processors.
 
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 7c25dbd5a181..35dd8b3824b0 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -26,6 +26,7 @@
 #include <linux/bitops.h>
 #include <linux/time.h>
 #include <linux/timex.h>
+#include <linux/clocksource.h>
 
 #include <asm/hardware.h>
 #include <asm/uaccess.h>
@@ -255,16 +256,6 @@ static unsigned volatile last_jiffy_time;
 
 #define CLOCK_TICKS_PER_USEC	((CLOCK_TICK_RATE + USEC_PER_SEC/2) / USEC_PER_SEC)
 
-/* IRQs are disabled before entering here from do_gettimeofday() */
-static unsigned long ixp4xx_gettimeoffset(void)
-{
-	u32 elapsed;
-
-	elapsed = *IXP4XX_OSTS - last_jiffy_time;
-
-	return elapsed / CLOCK_TICKS_PER_USEC;
-}
-
 static irqreturn_t ixp4xx_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
 	write_seqlock(&xtime_lock);
@@ -309,7 +300,6 @@ static void __init ixp4xx_timer_init(void)
 
 struct sys_timer ixp4xx_timer = {
 	.init		= ixp4xx_timer_init,
-	.offset		= ixp4xx_gettimeoffset,
 };
 
 static struct resource ixp46x_i2c_resources[] = {
@@ -365,3 +355,29 @@ void __init ixp4xx_sys_init(void)
 			ixp4xx_exp_bus_size >> 20);
 }
 
+cycle_t ixp4xx_get_cycles(void)
+{
+	return *IXP4XX_OSTS;
+}
+
+static struct clocksource clocksource_ixp4xx = {
+	.name 		= "OSTS",
+	.rating		= 200,
+	.read		= ixp4xx_get_cycles,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.shift 		= 20,
+	.is_continuous 	= 1,
+};
+
+unsigned long ixp4xx_timer_freq = FREQ;
+static int __init ixp4xx_clocksource_init(void)
+{
+	clocksource_ixp4xx.mult =
+		clocksource_hz2mult(ixp4xx_timer_freq,
+				    clocksource_ixp4xx.shift);
+	clocksource_register(&clocksource_ixp4xx);
+
+	return 0;
+}
+
+device_initcall(ixp4xx_clocksource_init);
diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c
index 749a337494d3..162c266e5f8f 100644
--- a/arch/arm/mach-ixp4xx/nslu2-setup.c
+++ b/arch/arm/mach-ixp4xx/nslu2-setup.c
@@ -159,6 +159,8 @@ static void nslu2_power_off(void)
 
 static void __init nslu2_init(void)
 {
+	ixp4xx_timer_freq = NSLU2_FREQ;
+
 	ixp4xx_sys_init();
 
 	nslu2_flash_resource.start = IXP4XX_EXP_BUS_BASE(0);
diff --git a/include/asm-arm/arch-ixp4xx/platform.h b/include/asm-arm/arch-ixp4xx/platform.h
index 13aee17b0475..8d10a9187693 100644
--- a/include/asm-arm/arch-ixp4xx/platform.h
+++ b/include/asm-arm/arch-ixp4xx/platform.h
@@ -89,6 +89,11 @@ struct ixp4xx_i2c_pins {
 
 struct sys_timer;
 
+/*
+ * Frequency of clock used for primary clocksource
+ */
+extern unsigned long ixp4xx_timer_freq;
+
 /*
  * Functions used by platform-level setup code
  */
-- 
cgit v1.2.3


From 29fa06c1292f473ae51a84f55c8fe22179bc1080 Mon Sep 17 00:00:00 2001
From: Rudolf Marek <r.marek@sh.cvut.cz>
Date: Mon, 28 Aug 2006 14:40:17 +0200
Subject: hwmon: New driver k8temp

Add support for the temperature sensor(s) found in AMD K8 CPUs.

Signed-off-by: Rudolf Marek <r.marek@sh.cvut.cz>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/hwmon/Kconfig   |  10 ++
 drivers/hwmon/Makefile  |   1 +
 drivers/hwmon/k8temp.c  | 292 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h |   1 +
 4 files changed, 304 insertions(+)
 create mode 100644 drivers/hwmon/k8temp.c

(limited to 'include')

diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 78c237f8fa0a..d9f86e9d405b 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -94,6 +94,16 @@ config SENSORS_ADM9240
 	  This driver can also be built as a module.  If so, the module
 	  will be called adm9240.
 
+config SENSORS_K8TEMP
+	tristate "AMD K8 processor sensor"
+	depends on HWMON && X86 && PCI && EXPERIMENTAL
+	help
+	  If you say yes here you get support for the temperature
+	  sensor(s) inside your AMD K8 CPU.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called k8temp.
+
 config SENSORS_ASB100
 	tristate "Asus ASB100 Bach"
 	depends on HWMON && I2C && EXPERIMENTAL
diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
index 31415843a91a..aab4c1063059 100644
--- a/drivers/hwmon/Makefile
+++ b/drivers/hwmon/Makefile
@@ -27,6 +27,7 @@ obj-$(CONFIG_SENSORS_GL518SM)	+= gl518sm.o
 obj-$(CONFIG_SENSORS_GL520SM)	+= gl520sm.o
 obj-$(CONFIG_SENSORS_HDAPS)	+= hdaps.o
 obj-$(CONFIG_SENSORS_IT87)	+= it87.o
+obj-$(CONFIG_SENSORS_K8TEMP)	+= k8temp.o
 obj-$(CONFIG_SENSORS_LM63)	+= lm63.o
 obj-$(CONFIG_SENSORS_LM70)	+= lm70.o
 obj-$(CONFIG_SENSORS_LM75)	+= lm75.o
diff --git a/drivers/hwmon/k8temp.c b/drivers/hwmon/k8temp.c
new file mode 100644
index 000000000000..50162ffa8832
--- /dev/null
+++ b/drivers/hwmon/k8temp.c
@@ -0,0 +1,292 @@
+/*
+ * k8temp.c - Linux kernel module for hardware monitoring
+ *
+ * Copyright (C) 2006 Rudolf Marek <r.marek@sh.cvut.cz>
+ *
+ * Inspired from the w83785 and amd756 drivers.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301 USA.
+ */
+
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/jiffies.h>
+#include <linux/pci.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#include <linux/err.h>
+#include <linux/mutex.h>
+
+#define TEMP_FROM_REG(val)	(((((val) >> 16) & 0xff) - 49) * 1000)
+#define REG_TEMP	0xe4
+#define SEL_PLACE	0x40
+#define SEL_CORE	0x04
+
+struct k8temp_data {
+	struct class_device *class_dev;
+	struct mutex update_lock;
+	const char *name;
+	char valid;		/* zero until following fields are valid */
+	unsigned long last_updated;	/* in jiffies */
+
+	/* registers values */
+	u8 sensorsp;		/* sensor presence bits - SEL_CORE & SEL_PLACE */
+	u32 temp[2][2];		/* core, place */
+};
+
+static struct k8temp_data *k8temp_update_device(struct device *dev)
+{
+	struct k8temp_data *data = dev_get_drvdata(dev);
+	struct pci_dev *pdev = to_pci_dev(dev);
+	u8 tmp;
+
+	mutex_lock(&data->update_lock);
+
+	if (!data->valid
+	    || time_after(jiffies, data->last_updated + HZ)) {
+		pci_read_config_byte(pdev, REG_TEMP, &tmp);
+		tmp &= ~(SEL_PLACE | SEL_CORE);		/* Select sensor 0, core0 */
+		pci_write_config_byte(pdev, REG_TEMP, tmp);
+		pci_read_config_dword(pdev, REG_TEMP, &data->temp[0][0]);
+
+		if (data->sensorsp & SEL_PLACE) {
+			tmp |= SEL_PLACE;	/* Select sensor 1, core0 */
+			pci_write_config_byte(pdev, REG_TEMP, tmp);
+			pci_read_config_dword(pdev, REG_TEMP,
+					      &data->temp[0][1]);
+		}
+
+		if (data->sensorsp & SEL_CORE) {
+			tmp &= ~SEL_PLACE;	/* Select sensor 0, core1 */
+			tmp |= SEL_CORE;
+			pci_write_config_byte(pdev, REG_TEMP, tmp);
+			pci_read_config_dword(pdev, REG_TEMP,
+					      &data->temp[1][0]);
+
+			if (data->sensorsp & SEL_PLACE) {
+				tmp |= SEL_PLACE;	/* Select sensor 1, core1 */
+				pci_write_config_byte(pdev, REG_TEMP, tmp);
+				pci_read_config_dword(pdev, REG_TEMP,
+						      &data->temp[1][1]);
+			}
+		}
+
+		data->last_updated = jiffies;
+		data->valid = 1;
+	}
+
+	mutex_unlock(&data->update_lock);
+	return data;
+}
+
+/*
+ * Sysfs stuff
+ */
+
+static ssize_t show_name(struct device *dev, struct device_attribute
+			 *devattr, char *buf)
+{
+	struct k8temp_data *data = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%s\n", data->name);
+}
+
+
+static ssize_t show_temp(struct device *dev,
+			 struct device_attribute *devattr, char *buf)
+{
+	struct sensor_device_attribute_2 *attr =
+	    to_sensor_dev_attr_2(devattr);
+	int core = attr->nr;
+	int place = attr->index;
+	struct k8temp_data *data = k8temp_update_device(dev);
+
+	return sprintf(buf, "%d\n",
+		       TEMP_FROM_REG(data->temp[core][place]));
+}
+
+/* core, place */
+
+static SENSOR_DEVICE_ATTR_2(temp1_input, S_IRUGO, show_temp, NULL, 0, 0);
+static SENSOR_DEVICE_ATTR_2(temp2_input, S_IRUGO, show_temp, NULL, 0, 1);
+static SENSOR_DEVICE_ATTR_2(temp3_input, S_IRUGO, show_temp, NULL, 1, 0);
+static SENSOR_DEVICE_ATTR_2(temp4_input, S_IRUGO, show_temp, NULL, 1, 1);
+static DEVICE_ATTR(name, S_IRUGO, show_name, NULL);
+
+static struct pci_device_id k8temp_ids[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) },
+	{ 0 },
+};
+
+static int __devinit k8temp_probe(struct pci_dev *pdev,
+				  const struct pci_device_id *id)
+{
+	int err;
+	u8 scfg;
+	u32 temp;
+	struct k8temp_data *data;
+	u32 cpuid = cpuid_eax(1);
+
+	/* this feature should be available since SH-C0 core */
+	if ((cpuid == 0xf40) || (cpuid == 0xf50) || (cpuid == 0xf51)) {
+		err = -ENODEV;
+		goto exit;
+	}
+
+	if (!(data = kzalloc(sizeof(struct k8temp_data), GFP_KERNEL))) {
+		err = -ENOMEM;
+		goto exit;
+	}
+
+	pci_read_config_byte(pdev, REG_TEMP, &scfg);
+	scfg &= ~(SEL_PLACE | SEL_CORE);		/* Select sensor 0, core0 */
+	pci_write_config_byte(pdev, REG_TEMP, scfg);
+	pci_read_config_byte(pdev, REG_TEMP, &scfg);
+
+	if (scfg & (SEL_PLACE | SEL_CORE)) {
+		dev_err(&pdev->dev, "Configuration bit(s) stuck at 1!\n");
+		err = -ENODEV;
+		goto exit_free;
+	}
+
+	scfg |= (SEL_PLACE | SEL_CORE);
+	pci_write_config_byte(pdev, REG_TEMP, scfg);
+
+	/* now we know if we can change core and/or sensor */
+	pci_read_config_byte(pdev, REG_TEMP, &data->sensorsp);
+
+	if (data->sensorsp & SEL_PLACE) {
+		scfg &= ~SEL_CORE;	/* Select sensor 1, core0 */
+		pci_write_config_byte(pdev, REG_TEMP, scfg);
+		pci_read_config_dword(pdev, REG_TEMP, &temp);
+		scfg |= SEL_CORE;	/* prepare for next selection */
+		if (!((temp >> 16) & 0xff))	/* if temp is 0 -49C is not likely */
+			data->sensorsp &= ~SEL_PLACE;
+	}
+
+	if (data->sensorsp & SEL_CORE) {
+		scfg &= ~SEL_PLACE;	/* Select sensor 0, core1 */
+		pci_write_config_byte(pdev, REG_TEMP, scfg);
+		pci_read_config_dword(pdev, REG_TEMP, &temp);
+		if (!((temp >> 16) & 0xff))	/* if temp is 0 -49C is not likely */
+			data->sensorsp &= ~SEL_CORE;
+	}
+
+	data->name = "k8temp";
+	mutex_init(&data->update_lock);
+	dev_set_drvdata(&pdev->dev, data);
+
+	/* Register sysfs hooks */
+	err = device_create_file(&pdev->dev,
+			   &sensor_dev_attr_temp1_input.dev_attr);
+	if (err)
+		goto exit_remove;
+
+	/* sensor can be changed and reports something */
+	if (data->sensorsp & SEL_PLACE) {
+		err = device_create_file(&pdev->dev,
+				   &sensor_dev_attr_temp2_input.dev_attr);
+		if (err)
+			goto exit_remove;
+	}
+
+	/* core can be changed and reports something */
+	if (data->sensorsp & SEL_CORE) {
+		err = device_create_file(&pdev->dev,
+				   &sensor_dev_attr_temp3_input.dev_attr);
+		if (err)
+			goto exit_remove;
+		if (data->sensorsp & SEL_PLACE)
+			err = device_create_file(&pdev->dev,
+					   &sensor_dev_attr_temp4_input.
+					   dev_attr);
+			if (err)
+				goto exit_remove;
+	}
+
+	err = device_create_file(&pdev->dev, &dev_attr_name);
+	if (err)
+		goto exit_remove;
+
+	data->class_dev = hwmon_device_register(&pdev->dev);
+
+	if (IS_ERR(data->class_dev)) {
+		err = PTR_ERR(data->class_dev);
+		goto exit_remove;
+	}
+
+	return 0;
+
+exit_remove:
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp1_input.dev_attr);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp2_input.dev_attr);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp3_input.dev_attr);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp4_input.dev_attr);
+	device_remove_file(&pdev->dev, &dev_attr_name);
+exit_free:
+	dev_set_drvdata(&pdev->dev, NULL);
+	kfree(data);
+exit:
+	return err;
+}
+
+static void __devexit k8temp_remove(struct pci_dev *pdev)
+{
+	struct k8temp_data *data = dev_get_drvdata(&pdev->dev);
+
+	hwmon_device_unregister(data->class_dev);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp1_input.dev_attr);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp2_input.dev_attr);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp3_input.dev_attr);
+	device_remove_file(&pdev->dev,
+			   &sensor_dev_attr_temp4_input.dev_attr);
+	device_remove_file(&pdev->dev, &dev_attr_name);
+	dev_set_drvdata(&pdev->dev, NULL);
+	kfree(data);
+}
+
+static struct pci_driver k8temp_driver = {
+	.name = "k8temp",
+	.id_table = k8temp_ids,
+	.probe = k8temp_probe,
+	.remove = __devexit_p(k8temp_remove),
+};
+
+static int __init k8temp_init(void)
+{
+	return pci_register_driver(&k8temp_driver);
+}
+
+static void __exit k8temp_exit(void)
+{
+	pci_unregister_driver(&k8temp_driver);
+}
+
+MODULE_AUTHOR("Rudolf Marek <r.marek@sh.cvut.cz>");
+MODULE_DESCRIPTION("AMD K8 core temperature monitor");
+MODULE_LICENSE("GPL");
+
+module_init(k8temp_init)
+module_exit(k8temp_exit)
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index ab032ceafa84..61db1907f06f 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -479,6 +479,7 @@
 
 #define PCI_VENDOR_ID_AMD		0x1022
 #define PCI_DEVICE_ID_AMD_K8_NB		0x1100
+#define PCI_DEVICE_ID_AMD_K8_NB_MISC	0x1103
 #define PCI_DEVICE_ID_AMD_LANCE		0x2000
 #define PCI_DEVICE_ID_AMD_LANCE_HOME	0x2001
 #define PCI_DEVICE_ID_AMD_SCSI		0x2020
-- 
cgit v1.2.3


From e0318ebff4d96131bb3524308b845f642e64df81 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Tue, 26 Sep 2006 14:50:20 -0400
Subject: USB: fix autosuspend when CONFIG_PM isn't set

This patch (as791b) fixes things up to avoid compiler warnings or
errors when CONFIG_USB_SUSPEND or CONFIG_PM isn't set.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 42 +++++++++++++++++++++---------------------
 drivers/usb/core/hub.c    |  4 ++--
 drivers/usb/core/usb.c    |  4 ++--
 drivers/usb/core/usb.h    | 18 +++++++++++++++++-
 include/linux/usb.h       |  2 +-
 5 files changed, 43 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index ee18d187ca17..113e484c763e 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -303,11 +303,11 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 	dev->driver = &driver->drvwrap.driver;
 	usb_set_intfdata(iface, priv);
 
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	iface->condition = USB_INTERFACE_BOUND;
 	mark_active(iface);
 	iface->pm_usage_cnt = !(driver->supports_autosuspend);
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 
 	/* if interface was already added, bind now; else let
 	 * the future device_add() bind it, bypassing probe()
@@ -356,11 +356,11 @@ void usb_driver_release_interface(struct usb_driver *driver,
 	dev->driver = NULL;
 	usb_set_intfdata(iface, NULL);
 
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	iface->condition = USB_INTERFACE_UNBOUND;
 	mark_quiesced(iface);
 	iface->needs_remote_wakeup = 0;
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 }
 EXPORT_SYMBOL(usb_driver_release_interface);
 
@@ -789,7 +789,7 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
 
 #ifdef CONFIG_PM
 
-/* Caller has locked udev->pm_mutex */
+/* Caller has locked udev's pm_mutex */
 static int suspend_device(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_device_driver	*udriver;
@@ -816,7 +816,7 @@ done:
 	return status;
 }
 
-/* Caller has locked udev->pm_mutex */
+/* Caller has locked udev's pm_mutex */
 static int resume_device(struct usb_device *udev)
 {
 	struct usb_device_driver	*udriver;
@@ -842,7 +842,7 @@ done:
 	return status;
 }
 
-/* Caller has locked intf's usb_device's pm_mutex */
+/* Caller has locked intf's usb_device's pm mutex */
 static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 {
 	struct usb_driver	*driver;
@@ -1064,7 +1064,7 @@ int usb_resume_both(struct usb_device *udev)
 	/* Propagate the resume up the tree, if necessary */
 	if (udev->state == USB_STATE_SUSPENDED) {
 		if (parent) {
-			mutex_lock_nested(&parent->pm_mutex, parent->level);
+			usb_pm_lock(parent);
 			parent->auto_pm = 1;
 			status = usb_resume_both(parent);
 		} else {
@@ -1079,7 +1079,7 @@ int usb_resume_both(struct usb_device *udev)
 		if (status == 0)
 			status = resume_device(udev);
 		if (parent)
-			mutex_unlock(&parent->pm_mutex);
+			usb_pm_unlock(parent);
 	} else {
 
 		/* Needed only for setting udev->dev.power.power_state.event
@@ -1129,12 +1129,12 @@ int usb_resume_both(struct usb_device *udev)
  */
 void usb_autosuspend_device(struct usb_device *udev, int dec_usage_cnt)
 {
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	udev->pm_usage_cnt -= dec_usage_cnt;
 	if (udev->pm_usage_cnt <= 0)
 		queue_delayed_work(ksuspend_usb_wq, &udev->autosuspend,
 				USB_AUTOSUSPEND_DELAY);
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 	// dev_dbg(&udev->dev, "%s: cnt %d\n",
 	//		__FUNCTION__, udev->pm_usage_cnt);
 }
@@ -1168,13 +1168,13 @@ int usb_autoresume_device(struct usb_device *udev, int inc_usage_cnt)
 {
 	int	status;
 
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	udev->pm_usage_cnt += inc_usage_cnt;
 	udev->auto_pm = 1;
 	status = usb_resume_both(udev);
 	if (status != 0)
 		udev->pm_usage_cnt -= inc_usage_cnt;
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 	// dev_dbg(&udev->dev, "%s: status %d cnt %d\n",
 	//		__FUNCTION__, status, udev->pm_usage_cnt);
 	return status;
@@ -1215,13 +1215,13 @@ void usb_autopm_put_interface(struct usb_interface *intf)
 {
 	struct usb_device	*udev = interface_to_usbdev(intf);
 
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	if (intf->condition != USB_INTERFACE_UNBOUND &&
 			--intf->pm_usage_cnt <= 0) {
 		queue_delayed_work(ksuspend_usb_wq, &udev->autosuspend,
 				USB_AUTOSUSPEND_DELAY);
 	}
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 	// dev_dbg(&intf->dev, "%s: cnt %d\n",
 	//		__FUNCTION__, intf->pm_usage_cnt);
 }
@@ -1263,7 +1263,7 @@ int usb_autopm_get_interface(struct usb_interface *intf)
 	struct usb_device	*udev = interface_to_usbdev(intf);
 	int			status;
 
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	if (intf->condition == USB_INTERFACE_UNBOUND)
 		status = -ENODEV;
 	else {
@@ -1273,7 +1273,7 @@ int usb_autopm_get_interface(struct usb_interface *intf)
 		if (status != 0)
 			--intf->pm_usage_cnt;
 	}
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 	// dev_dbg(&intf->dev, "%s: status %d cnt %d\n",
 	//		__FUNCTION__, status, intf->pm_usage_cnt);
 	return status;
@@ -1289,10 +1289,10 @@ static int usb_suspend(struct device *dev, pm_message_t message)
 	if (is_usb_device(dev)) {
 		struct usb_device *udev = to_usb_device(dev);
 
-		mutex_lock_nested(&udev->pm_mutex, udev->level);
+		usb_pm_lock(udev);
 		udev->auto_pm = 0;
 		status = usb_suspend_both(udev, message);
-		mutex_unlock(&udev->pm_mutex);
+		usb_pm_unlock(udev);
 	} else
 		status = 0;
 	return status;
@@ -1305,10 +1305,10 @@ static int usb_resume(struct device *dev)
 	if (is_usb_device(dev)) {
 		struct usb_device *udev = to_usb_device(dev);
 
-		mutex_lock_nested(&udev->pm_mutex, udev->level);
+		usb_pm_lock(udev);
 		udev->auto_pm = 0;
 		status = usb_resume_both(udev);
-		mutex_unlock(&udev->pm_mutex);
+		usb_pm_unlock(udev);
 
 		/* Rebind drivers that had no suspend method? */
 	} else
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 2a8cb3c2b19c..7676690a0386 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1779,7 +1779,7 @@ static int remote_wakeup(struct usb_device *udev)
 	 * to the parent hub! */
 
 	usb_lock_device(udev);
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	if (udev->state == USB_STATE_SUSPENDED) {
 		dev_dbg(&udev->dev, "usb %sresume\n", "wakeup-");
 		/* TRSMRCY = 10 msec */
@@ -1788,7 +1788,7 @@ static int remote_wakeup(struct usb_device *udev)
 		if (status == 0)
 			udev->dev.power.power_state.event = PM_EVENT_ON;
 	}
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 
 	if (status == 0)
 		usb_autoresume_device(udev, 0);
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 239f8e5d247f..e4df9edf1bc0 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -214,10 +214,10 @@ static void usb_autosuspend_work(void *_udev)
 {
 	struct usb_device	*udev = _udev;
 
-	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	usb_pm_lock(udev);
 	udev->auto_pm = 1;
 	usb_suspend_both(udev, PMSG_SUSPEND);
-	mutex_unlock(&udev->pm_mutex);
+	usb_pm_unlock(udev);
 }
 
 #else
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index fb6eb41c374f..f69df137ec0e 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -36,6 +36,16 @@ extern int usb_resume_both(struct usb_device *udev);
 extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 
+static inline void usb_pm_lock(struct usb_device *udev)
+{
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+}
+
+static inline void usb_pm_unlock(struct usb_device *udev)
+{
+	mutex_unlock(&udev->pm_mutex);
+}
+
 #else
 
 #define usb_suspend_both(udev, msg)	0
@@ -45,6 +55,8 @@ static inline int usb_resume_both(struct usb_device *udev)
 }
 #define usb_port_suspend(dev)		0
 #define usb_port_resume(dev)		0
+static inline void usb_pm_lock(struct usb_device *udev) {}
+static inline void usb_pm_unlock(struct usb_device *udev) {}
 
 #endif
 
@@ -58,7 +70,11 @@ extern int usb_autoresume_device(struct usb_device *udev, int inc_busy_cnt);
 #else
 
 #define usb_autosuspend_device(udev, dec_busy_cnt)	do {} while (0)
-#define usb_autoresume_device(udev, inc_busy_cnt)	0
+static inline int usb_autoresume_device(struct usb_device *udev,
+		int inc_busy_cnt)
+{
+	return 0;
+}
 
 #endif
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 0da15b0b02be..190cc1b78fe2 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -380,10 +380,10 @@ struct usb_device {
 	int maxchild;			/* Number of ports if hub */
 	struct usb_device *children[USB_MAXCHILDREN];
 
+	int pm_usage_cnt;		/* usage counter for autosuspend */
 #ifdef CONFIG_PM
 	struct work_struct autosuspend;	/* for delayed autosuspends */
 	struct mutex pm_mutex;		/* protects PM operations */
-	int pm_usage_cnt;		/* usage counter for autosuspend */
 
 	unsigned auto_pm:1;		/* autosuspend/resume in progress */
 	unsigned do_remote_wakeup:1;	/* remote wakeup should be enabled */
-- 
cgit v1.2.3


From 2a50f28c326d20ab4556be1b867ecddf6aefbb88 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:22:08 -0700
Subject: [ATALK]: endianness annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/appletalk/ipddp.c |  5 +--
 include/linux/atalk.h         | 40 +---------------------
 net/appletalk/ddp.c           | 79 +++++++++++++++----------------------------
 3 files changed, 30 insertions(+), 94 deletions(-)

(limited to 'include')

diff --git a/drivers/net/appletalk/ipddp.c b/drivers/net/appletalk/ipddp.c
index 7f7dd450226a..b98592a8bac8 100644
--- a/drivers/net/appletalk/ipddp.c
+++ b/drivers/net/appletalk/ipddp.c
@@ -145,9 +145,7 @@ static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/* Create the Extended DDP header */
 	ddp = (struct ddpehdr *)skb->data;
-        ddp->deh_len = skb->len;
-        ddp->deh_hops = 1;
-        ddp->deh_pad = 0;
+        ddp->deh_len_hops = htons(skb->len + (1<<10));
         ddp->deh_sum = 0;
 
 	/*
@@ -170,7 +168,6 @@ static int ipddp_xmit(struct sk_buff *skb, struct net_device *dev)
         ddp->deh_sport = 72;
 
         *((__u8 *)(ddp+1)) = 22;        	/* ddp type = IP */
-        *((__u16 *)ddp)=ntohs(*((__u16 *)ddp));	/* fix up length field */
 
         skb->protocol = htons(ETH_P_ATALK);     /* Protocol has changed */
 
diff --git a/include/linux/atalk.h b/include/linux/atalk.h
index 6ba3aa8a81f4..75b8baca08f3 100644
--- a/include/linux/atalk.h
+++ b/include/linux/atalk.h
@@ -88,15 +88,7 @@ static inline struct atalk_sock *at_sk(struct sock *sk)
 #include <asm/byteorder.h>
 
 struct ddpehdr {
-#ifdef __LITTLE_ENDIAN_BITFIELD
-	__u16	deh_len:10,
-		deh_hops:4,
-		deh_pad:2;
-#else
-	__u16	deh_pad:2,
-		deh_hops:4,
-		deh_len:10;
-#endif
+	__be16	deh_len_hops;	/* lower 10 bits are length, next 4 - hops */
 	__be16	deh_sum;
 	__be16	deh_dnet;
 	__be16	deh_snet;
@@ -112,36 +104,6 @@ static __inline__ struct ddpehdr *ddp_hdr(struct sk_buff *skb)
 	return (struct ddpehdr *)skb->h.raw;
 }
 
-/*
- *	Don't drop the struct into the struct above.  You'll get some
- *	surprise padding.
- */
-struct ddpebits {
-#ifdef __LITTLE_ENDIAN_BITFIELD
-	__u16	deh_len:10,
-		deh_hops:4,
-		deh_pad:2;
-#else
-	__u16	deh_pad:2,
-		deh_hops:4,
-		deh_len:10;
-#endif
-};
-
-/* Short form header */
-struct ddpshdr {
-#ifdef __LITTLE_ENDIAN_BITFIELD
-	__u16	dsh_len:10,
-		dsh_pad:6;
-#else
-	__u16	dsh_pad:6,
-		dsh_len:10;
-#endif
-	__u8	dsh_dport;
-	__u8	dsh_sport;
-	/* And netatalk apps expect to stick the type in themselves */
-};
-
 /* AppleTalk AARP headers */
 struct elapaarp {
 	__be16	hw_type;
diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c
index 96dc6bb52d14..708e2e0371af 100644
--- a/net/appletalk/ddp.c
+++ b/net/appletalk/ddp.c
@@ -1002,7 +1002,7 @@ static unsigned long atalk_sum_skb(const struct sk_buff *skb, int offset,
 	return sum;
 }
 
-static unsigned short atalk_checksum(const struct sk_buff *skb, int len)
+static __be16 atalk_checksum(const struct sk_buff *skb, int len)
 {
 	unsigned long sum;
 
@@ -1010,7 +1010,7 @@ static unsigned short atalk_checksum(const struct sk_buff *skb, int len)
 	sum = atalk_sum_skb(skb, 4, len-4, 0);
 
 	/* Use 0xFFFF for 0. 0 itself means none */
-	return sum ? htons((unsigned short)sum) : 0xFFFF;
+	return sum ? htons((unsigned short)sum) : htons(0xFFFF);
 }
 
 static struct proto ddp_proto = {
@@ -1289,7 +1289,7 @@ static int handle_ip_over_ddp(struct sk_buff *skb)
 #endif
 
 static void atalk_route_packet(struct sk_buff *skb, struct net_device *dev,
-			       struct ddpehdr *ddp, struct ddpebits *ddphv,
+			       struct ddpehdr *ddp, __u16 len_hops,
 			       int origlen)
 {
 	struct atalk_route *rt;
@@ -1317,10 +1317,12 @@ static void atalk_route_packet(struct sk_buff *skb, struct net_device *dev,
 
 	/* Route the packet */
 	rt = atrtr_find(&ta);
-	if (!rt || ddphv->deh_hops == DDP_MAXHOPS)
+	/* increment hops count */
+	len_hops += 1 << 10;
+	if (!rt || !(len_hops & (15 << 10)))
 		goto free_it;
+
 	/* FIXME: use skb->cb to be able to use shared skbs */
-	ddphv->deh_hops++;
 
 	/*
 	 * Route goes through another gateway, so set the target to the
@@ -1335,11 +1337,10 @@ static void atalk_route_packet(struct sk_buff *skb, struct net_device *dev,
         /* Fix up skb->len field */
         skb_trim(skb, min_t(unsigned int, origlen,
 			    (rt->dev->hard_header_len +
-			     ddp_dl->header_length + ddphv->deh_len)));
+			     ddp_dl->header_length + (len_hops & 1023))));
 
-	/* Mend the byte order */
 	/* FIXME: use skb->cb to be able to use shared skbs */
-	*((__u16 *)ddp) = ntohs(*((__u16 *)ddphv));
+	ddp->deh_len_hops = htons(len_hops);
 
 	/*
 	 * Send the buffer onwards
@@ -1394,7 +1395,7 @@ static int atalk_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct atalk_iface *atif;
 	struct sockaddr_at tosat;
         int origlen;
-        struct ddpebits ddphv;
+	__u16 len_hops;
 
 	/* Don't mangle buffer if shared */
 	if (!(skb = skb_share_check(skb, GFP_ATOMIC))) 
@@ -1406,16 +1407,11 @@ static int atalk_rcv(struct sk_buff *skb, struct net_device *dev,
 
 	ddp = ddp_hdr(skb);
 
-	/*
-	 *	Fix up the length field	[Ok this is horrible but otherwise
-	 *	I end up with unions of bit fields and messy bit field order
-	 *	compiler/endian dependencies..]
-	 */
-	*((__u16 *)&ddphv) = ntohs(*((__u16 *)ddp));
+	len_hops = ntohs(ddp->deh_len_hops);
 
 	/* Trim buffer in case of stray trailing data */
 	origlen = skb->len;
-	skb_trim(skb, min_t(unsigned int, skb->len, ddphv.deh_len));
+	skb_trim(skb, min_t(unsigned int, skb->len, len_hops & 1023));
 
 	/*
 	 * Size check to see if ddp->deh_len was crap
@@ -1430,7 +1426,7 @@ static int atalk_rcv(struct sk_buff *skb, struct net_device *dev,
 	 * valid for net byte orders all over the networking code...
 	 */
 	if (ddp->deh_sum &&
-	    atalk_checksum(skb, ddphv.deh_len) != ddp->deh_sum)
+	    atalk_checksum(skb, len_hops & 1023) != ddp->deh_sum)
 		/* Not a valid AppleTalk frame - dustbin time */
 		goto freeit;
 
@@ -1444,7 +1440,7 @@ static int atalk_rcv(struct sk_buff *skb, struct net_device *dev,
 		/* Not ours, so we route the packet via the correct
 		 * AppleTalk iface
 		 */
-		atalk_route_packet(skb, dev, ddp, &ddphv, origlen);
+		atalk_route_packet(skb, dev, ddp, len_hops, origlen);
 		goto out;
 	}
 
@@ -1489,7 +1485,7 @@ static int ltalk_rcv(struct sk_buff *skb, struct net_device *dev,
 		/* Find our address */
 		struct atalk_addr *ap = atalk_find_dev_addr(dev);
 
-		if (!ap || skb->len < sizeof(struct ddpshdr))
+		if (!ap || skb->len < sizeof(__be16) || skb->len > 1023)
 			goto freeit;
 
 		/* Don't mangle buffer if shared */
@@ -1519,11 +1515,8 @@ static int ltalk_rcv(struct sk_buff *skb, struct net_device *dev,
 		/*
 		 * Not sure about this bit...
 		 */
-		ddp->deh_len   = skb->len;
-		ddp->deh_hops  = DDP_MAXHOPS;	/* Non routable, so force a drop
-						   if we slip up later */
-		/* Mend the byte order */
-		*((__u16 *)ddp) = htons(*((__u16 *)ddp));
+		/* Non routable, so force a drop if we slip up later */
+		ddp->deh_len_hops = htons(skb->len + (DDP_MAXHOPS << 10));
 	}
 	skb->h.raw = skb->data;
 
@@ -1622,16 +1615,7 @@ static int atalk_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr
 	SOCK_DEBUG(sk, "SK %p: Begin build.\n", sk);
 
 	ddp = (struct ddpehdr *)skb_put(skb, sizeof(struct ddpehdr));
-	ddp->deh_pad  = 0;
-	ddp->deh_hops = 0;
-	ddp->deh_len  = len + sizeof(*ddp);
-	/*
-	 * Fix up the length field [Ok this is horrible but otherwise
-	 * I end up with unions of bit fields and messy bit field order
-	 * compiler/endian dependencies..
-	 */
-	*((__u16 *)ddp) = ntohs(*((__u16 *)ddp));
-
+	ddp->deh_len_hops  = htons(len + sizeof(*ddp));
 	ddp->deh_dnet  = usat->sat_addr.s_net;
 	ddp->deh_snet  = at->src_net;
 	ddp->deh_dnode = usat->sat_addr.s_node;
@@ -1712,8 +1696,8 @@ static int atalk_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr
 	struct sockaddr_at *sat = (struct sockaddr_at *)msg->msg_name;
 	struct ddpehdr *ddp;
 	int copied = 0;
+	int offset = 0;
 	int err = 0;
-        struct ddpebits ddphv;
 	struct sk_buff *skb = skb_recv_datagram(sk, flags & ~MSG_DONTWAIT,
 						flags & MSG_DONTWAIT, &err);
 	if (!skb)
@@ -1721,25 +1705,18 @@ static int atalk_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr
 
 	/* FIXME: use skb->cb to be able to use shared skbs */
 	ddp = ddp_hdr(skb);
-	*((__u16 *)&ddphv) = ntohs(*((__u16 *)ddp));
+	copied = ntohs(ddp->deh_len_hops) & 1023;
 
-	if (sk->sk_type == SOCK_RAW) {
-		copied = ddphv.deh_len;
-		if (copied > size) {
-			copied = size;
-			msg->msg_flags |= MSG_TRUNC;
-		}
+	if (sk->sk_type != SOCK_RAW) {
+		offset = sizeof(*ddp);
+		copied -= offset;
+	}
 
-		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
-	} else {
-		copied = ddphv.deh_len - sizeof(*ddp);
-		if (copied > size) {
-			copied = size;
-			msg->msg_flags |= MSG_TRUNC;
-		}
-		err = skb_copy_datagram_iovec(skb, sizeof(*ddp),
-					      msg->msg_iov, copied);
+	if (copied > size) {
+		copied = size;
+		msg->msg_flags |= MSG_TRUNC;
 	}
+	err = skb_copy_datagram_iovec(skb, offset, msg->msg_iov, copied);
 
 	if (!err) {
 		if (sat) {
-- 
cgit v1.2.3


From 0ac0760a57a6b1eb75c21a590e578be5dfc2f88b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:23:16 -0700
Subject: [TR]: endiannness annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/trdevice.h |  2 +-
 net/802/tr.c             | 12 ++++++------
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/trdevice.h b/include/linux/trdevice.h
index 99e02ef54c47..bfc84a7aecc5 100644
--- a/include/linux/trdevice.h
+++ b/include/linux/trdevice.h
@@ -28,7 +28,7 @@
 #include <linux/if_tr.h>
 
 #ifdef __KERNEL__
-extern unsigned short	tr_type_trans(struct sk_buff *skb, struct net_device *dev);
+extern __be16 tr_type_trans(struct sk_buff *skb, struct net_device *dev);
 extern void tr_source_route(struct sk_buff *skb, struct trh_hdr *trh, struct net_device *dev);
 extern struct net_device *alloc_trdev(int sizeof_priv);
 
diff --git a/net/802/tr.c b/net/802/tr.c
index d7d8f40c4fed..829deb41ce81 100644
--- a/net/802/tr.c
+++ b/net/802/tr.c
@@ -164,7 +164,7 @@ static int tr_rebuild_header(struct sk_buff *skb)
 	 */
 	 
 	if(trllc->ethertype != htons(ETH_P_IP)) {
-		printk("tr_rebuild_header: Don't know how to resolve type %04X addresses ?\n",(unsigned int)htons(trllc->ethertype));
+		printk("tr_rebuild_header: Don't know how to resolve type %04X addresses ?\n", ntohs(trllc->ethertype));
 		return 0;
 	}
 
@@ -186,7 +186,7 @@ static int tr_rebuild_header(struct sk_buff *skb)
  *	it via SNAP.
  */
  
-unsigned short tr_type_trans(struct sk_buff *skb, struct net_device *dev) 
+__be16 tr_type_trans(struct sk_buff *skb, struct net_device *dev)
 {
 
 	struct trh_hdr *trh=(struct trh_hdr *)skb->data;
@@ -229,15 +229,15 @@ unsigned short tr_type_trans(struct sk_buff *skb, struct net_device *dev)
 	 */
 
 	if (trllc->dsap == EXTENDED_SAP &&
-	    (trllc->ethertype == ntohs(ETH_P_IP) ||
-	     trllc->ethertype == ntohs(ETH_P_IPV6) ||
-	     trllc->ethertype == ntohs(ETH_P_ARP)))
+	    (trllc->ethertype == htons(ETH_P_IP) ||
+	     trllc->ethertype == htons(ETH_P_IPV6) ||
+	     trllc->ethertype == htons(ETH_P_ARP)))
 	{
 		skb_pull(skb, sizeof(struct trllc));
 		return trllc->ethertype;
 	}
 
-	return ntohs(ETH_P_TR_802_2);
+	return htons(ETH_P_TR_802_2);
 }
 
 /*
-- 
cgit v1.2.3


From 046d033148e6936ee2466d38214cf0743a210f39 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:24:24 -0700
Subject: [IPV4]: headers endianness

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ip.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/ip.h b/include/linux/ip.h
index 2f4600146f83..6b25d36fc54c 100644
--- a/include/linux/ip.h
+++ b/include/linux/ip.h
@@ -96,7 +96,7 @@ struct iphdr {
 	__be16	frag_off;
 	__u8	ttl;
 	__u8	protocol;
-	__u16	check;
+	__be16	check;
 	__be32	saddr;
 	__be32	daddr;
 	/*The options start here. */
@@ -105,22 +105,22 @@ struct iphdr {
 struct ip_auth_hdr {
 	__u8  nexthdr;
 	__u8  hdrlen;		/* This one is measured in 32 bit units! */
-	__u16 reserved;
-	__u32 spi;
-	__u32 seq_no;		/* Sequence number */
+	__be16 reserved;
+	__be32 spi;
+	__be32 seq_no;		/* Sequence number */
 	__u8  auth_data[0];	/* Variable len but >=4. Mind the 64 bit alignment! */
 };
 
 struct ip_esp_hdr {
-	__u32 spi;
-	__u32 seq_no;		/* Sequence number */
+	__be32 spi;
+	__be32 seq_no;		/* Sequence number */
 	__u8  enc_data[0];	/* Variable len but >=8. Mind the 64 bit alignment! */
 };
 
 struct ip_comp_hdr {
 	__u8 nexthdr;
 	__u8 flags;
-	__u16 cpi;
+	__be16 cpi;
 };
 
 #endif	/* _LINUX_IP_H */
-- 
cgit v1.2.3


From 9e12bb22e32389b41222c9d9fb55724fed83a038 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:25:20 -0700
Subject: [IPV4]: ip_route_input() annotations

ip_route_input() takes net-endian source and destination address.
* Annotated as such.
* arguments of its invocations annotated where needed.
* local helpers getting the same values passed to by it (ip_route_input_mc(),
ip_route_input_slow(), ip_handle_martian_source(), ip_mkroute_input(),
ip_mkroute_input_def(), __mkroute_input()) annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h   |  2 +-
 net/ipv4/arp.c        |  2 +-
 net/ipv4/ip_options.c |  2 +-
 net/ipv4/route.c      | 22 ++++++++++++----------
 4 files changed, 15 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 7f93ac0e0899..9cfa67089f7f 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -116,7 +116,7 @@ extern void		rt_cache_flush(int how);
 extern int		__ip_route_output_key(struct rtable **, const struct flowi *flp);
 extern int		ip_route_output_key(struct rtable **, struct flowi *flp);
 extern int		ip_route_output_flow(struct rtable **rp, struct flowi *flp, struct sock *sk, int flags);
-extern int		ip_route_input(struct sk_buff*, u32 dst, u32 src, u8 tos, struct net_device *devin);
+extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, struct net_device *devin);
 extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu);
 extern void		ip_rt_send_redirect(struct sk_buff *skb);
 
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index c8a3723bc001..48e1ccb2ff55 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -710,7 +710,7 @@ static int arp_process(struct sk_buff *skb)
 	unsigned char *arp_ptr;
 	struct rtable *rt;
 	unsigned char *sha, *tha;
-	u32 sip, tip;
+	__be32 sip, tip;
 	u16 dev_type = dev->type;
 	int addr_type;
 	struct neighbour *n;
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index e7437c091326..87f25486fd64 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -607,7 +607,7 @@ int ip_options_rcv_srr(struct sk_buff *skb)
 {
 	struct ip_options *opt = &(IPCB(skb)->opt);
 	int srrspace, srrptr;
-	u32 nexthop;
+	__be32 nexthop;
 	struct iphdr *iph = skb->nh.iph;
 	unsigned char * optptr = skb->nh.raw + opt->srr;
 	struct rtable *rt = (struct rtable*)skb->dst;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 20ffe8e88c0f..bbe529774fd9 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1596,7 +1596,7 @@ static void rt_set_nexthop(struct rtable *rt, struct fib_result *res, u32 itag)
         rt->rt_type = res->type;
 }
 
-static int ip_route_input_mc(struct sk_buff *skb, u32 daddr, u32 saddr,
+static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 				u8 tos, struct net_device *dev, int our)
 {
 	unsigned hash;
@@ -1681,8 +1681,8 @@ e_inval:
 static void ip_handle_martian_source(struct net_device *dev,
 				     struct in_device *in_dev,
 				     struct sk_buff *skb,
-				     u32 daddr,
-				     u32 saddr) 
+				     __be32 daddr,
+				     __be32 saddr)
 {
 	RT_CACHE_STAT_INC(in_martian_src);
 #ifdef CONFIG_IP_ROUTE_VERBOSE
@@ -1712,7 +1712,7 @@ static void ip_handle_martian_source(struct net_device *dev,
 static inline int __mkroute_input(struct sk_buff *skb, 
 				  struct fib_result* res, 
 				  struct in_device *in_dev, 
-				  u32 daddr, u32 saddr, u32 tos, 
+				  __be32 daddr, __be32 saddr, u32 tos,
 				  struct rtable **result) 
 {
 
@@ -1813,7 +1813,7 @@ static inline int ip_mkroute_input_def(struct sk_buff *skb,
 				       struct fib_result* res, 
 				       const struct flowi *fl,
 				       struct in_device *in_dev,
-				       u32 daddr, u32 saddr, u32 tos)
+				       __be32 daddr, __be32 saddr, u32 tos)
 {
 	struct rtable* rth = NULL;
 	int err;
@@ -1838,7 +1838,7 @@ static inline int ip_mkroute_input(struct sk_buff *skb,
 				   struct fib_result* res, 
 				   const struct flowi *fl,
 				   struct in_device *in_dev,
-				   u32 daddr, u32 saddr, u32 tos)
+				   __be32 daddr, __be32 saddr, u32 tos)
 {
 #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
 	struct rtable* rth = NULL, *rtres;
@@ -1901,7 +1901,7 @@ static inline int ip_mkroute_input(struct sk_buff *skb,
  *	2. IP spoofing attempts are filtered with 100% of guarantee.
  */
 
-static int ip_route_input_slow(struct sk_buff *skb, u32 daddr, u32 saddr,
+static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 			       u8 tos, struct net_device *dev)
 {
 	struct fib_result res;
@@ -1920,7 +1920,7 @@ static int ip_route_input_slow(struct sk_buff *skb, u32 daddr, u32 saddr,
 	u32		itag = 0;
 	struct rtable * rth;
 	unsigned	hash;
-	u32		spec_dst;
+	__be32		spec_dst;
 	int		err = -EINVAL;
 	int		free_res = 0;
 
@@ -2087,7 +2087,7 @@ martian_source:
 	goto e_inval;
 }
 
-int ip_route_input(struct sk_buff *skb, u32 daddr, u32 saddr,
+int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 		   u8 tos, struct net_device *dev)
 {
 	struct rtable * rth;
@@ -2740,7 +2740,9 @@ int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
 	struct rtmsg *rtm;
 	struct nlattr *tb[RTA_MAX+1];
 	struct rtable *rt = NULL;
-	u32 dst, src, iif;
+	__be32 dst = 0;
+	__be32 src = 0;
+	u32 iif;
 	int err;
 	struct sk_buff *skb;
 
-- 
cgit v1.2.3


From f7655229c06d041323b40bd6eb9f95ca0ce95506 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:25:43 -0700
Subject: [IPV4]: ip_rt_redirect() annotations

The first 4 arguments of ip_rt_redirect() are net-endian.  Annotated.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h | 4 ++--
 net/ipv4/route.c    | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 9cfa67089f7f..64a18811526d 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -109,8 +109,8 @@ extern struct ip_rt_acct *ip_rt_acct;
 
 struct in_device;
 extern int		ip_rt_init(void);
-extern void		ip_rt_redirect(u32 old_gw, u32 dst, u32 new_gw,
-				       u32 src, struct net_device *dev);
+extern void		ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
+				       __be32 src, struct net_device *dev);
 extern void		ip_rt_advice(struct rtable **rp, int advice);
 extern void		rt_cache_flush(int how);
 extern int		__ip_route_output_key(struct rtable **, const struct flowi *flp);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index bbe529774fd9..e14df8ec5082 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1118,13 +1118,13 @@ static void rt_del(unsigned hash, struct rtable *rt)
 	spin_unlock_bh(rt_hash_lock_addr(hash));
 }
 
-void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
-		    u32 saddr, struct net_device *dev)
+void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
+		    __be32 saddr, struct net_device *dev)
 {
 	int i, k;
 	struct in_device *in_dev = in_dev_get(dev);
 	struct rtable *rth, **rthp;
-	u32  skeys[2] = { saddr, 0 };
+	__be32  skeys[2] = { saddr, 0 };
 	int  ikeys[2] = { dev->ifindex, 0 };
 	struct netevent_redirect netevent;
 
-- 
cgit v1.2.3


From f2c3fe24119ee4f8faca08699f0488f500014a27 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:26:42 -0700
Subject: [IPV4]: annotate ipv4 addresses in struct rtable and struct flowi

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/flow.h  | 4 ++--
 include/net/route.h | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/flow.h b/include/net/flow.h
index 3ca210ec1379..97569af9bda7 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -16,8 +16,8 @@ struct flowi {
 
 	union {
 		struct {
-			__u32			daddr;
-			__u32			saddr;
+			__be32			daddr;
+			__be32			saddr;
 			__u32			fwmark;
 			__u8			tos;
 			__u8			scope;
diff --git a/include/net/route.h b/include/net/route.h
index 64a18811526d..5bb2b15b4391 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -62,18 +62,18 @@ struct rtable
 	__u16			rt_type;
 	__u16			rt_multipath_alg;
 
-	__u32			rt_dst;	/* Path destination	*/
-	__u32			rt_src;	/* Path source		*/
+	__be32			rt_dst;	/* Path destination	*/
+	__be32			rt_src;	/* Path source		*/
 	int			rt_iif;
 
 	/* Info on neighbour */
-	__u32			rt_gateway;
+	__be32			rt_gateway;
 
 	/* Cache lookup keys */
 	struct flowi		fl;
 
 	/* Miscellaneous cached information */
-	__u32			rt_spec_dst; /* RFC1122 specific destination */
+	__be32			rt_spec_dst; /* RFC1122 specific destination */
 	struct inet_peer	*peer; /* long-living peer info */
 };
 
-- 
cgit v1.2.3


From bada8adc4e6622764205921e6ba3f717aa03c882 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:27:15 -0700
Subject: [IPV4]: ip_route_connect() ipv4 address arguments annotated

annotated address arguments (port number left alone for now); ditto
for inferred net-endian variables in callers.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h | 4 ++--
 net/dccp/ipv4.c     | 2 +-
 net/ipv4/af_inet.c  | 2 +-
 net/ipv4/datagram.c | 2 +-
 net/ipv4/tcp_ipv4.c | 2 +-
 5 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 5bb2b15b4391..63524843f6d9 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -144,8 +144,8 @@ static inline char rt_tos2priority(u8 tos)
 	return ip_tos2prio[IPTOS_TOS(tos)>>1];
 }
 
-static inline int ip_route_connect(struct rtable **rp, u32 dst,
-				   u32 src, u32 tos, int oif, u8 protocol,
+static inline int ip_route_connect(struct rtable **rp, __be32 dst,
+				   __be32 src, u32 tos, int oif, u8 protocol,
 				   u16 sport, u16 dport, struct sock *sk)
 {
 	struct flowi fl = { .oif = oif,
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 66be29b6f508..bf692c1c116f 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -50,7 +50,7 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	struct dccp_sock *dp = dccp_sk(sk);
 	const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
 	struct rtable *rt;
-	u32 daddr, nexthop;
+	__be32 daddr, nexthop;
 	int tmp;
 	int err;
 
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index fdd89e37b9aa..250a02b23e4e 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -996,7 +996,7 @@ static int inet_sk_reselect_saddr(struct sock *sk)
 	struct rtable *rt;
 	__u32 old_saddr = inet->saddr;
 	__u32 new_saddr;
-	__u32 daddr = inet->daddr;
+	__be32 daddr = inet->daddr;
 
 	if (inet->opt && inet->opt->srr)
 		daddr = inet->opt->faddr;
diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c
index ec5da4fbd9f4..7b068a891953 100644
--- a/net/ipv4/datagram.c
+++ b/net/ipv4/datagram.c
@@ -25,7 +25,7 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	struct inet_sock *inet = inet_sk(sk);
 	struct sockaddr_in *usin = (struct sockaddr_in *) uaddr;
 	struct rtable *rt;
-	u32 saddr;
+	__be32 saddr;
 	int oif;
 	int err;
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 39b179856082..9018aa41cb55 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -159,7 +159,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
 	struct rtable *rt;
-	u32 daddr, nexthop;
+	__be32 daddr, nexthop;
 	int tmp;
 	int err;
 
-- 
cgit v1.2.3


From 011a92610826bdeec4b80ab423958d4c512639f6 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:27:35 -0700
Subject: [IPV4]: annotated ipv4 addresses in struct inet_sock

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index f6242710f2ff..fc0b9e157b61 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -110,11 +110,11 @@ struct inet_sock {
 	struct ipv6_pinfo	*pinet6;
 #endif
 	/* Socket demultiplex comparisons on incoming packets. */
-	__u32			daddr;
-	__u32			rcv_saddr;
+	__be32			daddr;
+	__be32			rcv_saddr;
 	__u16			dport;
 	__u16			num;
-	__u32			saddr;
+	__be32			saddr;
 	__s16			uc_ttl;
 	__u16			cmsg_flags;
 	struct ip_options	*opt;
@@ -129,7 +129,7 @@ struct inet_sock {
 				hdrincl:1,
 				mc_loop:1;
 	int			mc_index;
-	__u32			mc_addr;
+	__be32			mc_addr;
 	struct ip_mc_socklist	*mc_list;
 	struct {
 		unsigned int		flags;
@@ -137,7 +137,7 @@ struct inet_sock {
 		struct ip_options	*opt;
 		struct rtable		*rt;
 		int			length; /* Total length of all frames */
-		u32			addr;
+		__be32			addr;
 		struct flowi		fl;
 	} cork;
 };
-- 
cgit v1.2.3


From a61ced5d1c2e773620d7855ea2009d770c10a6e6 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:27:54 -0700
Subject: [IPV4]: inet_select_addr() annotations

argument and return value are net-endian.  Annotated function and inferred
net-endian variables in callers.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h          | 2 +-
 net/ipv4/arp.c                      | 4 ++--
 net/ipv4/devinet.c                  | 4 ++--
 net/ipv4/icmp.c                     | 2 +-
 net/ipv4/ipvs/ip_vs_sync.c          | 2 +-
 net/ipv4/netfilter/ipt_MASQUERADE.c | 2 +-
 net/ipv4/route.c                    | 4 ++--
 7 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 92297ff24e85..40d07d0b896b 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -110,7 +110,7 @@ extern int		devinet_ioctl(unsigned int cmd, void __user *);
 extern void		devinet_init(void);
 extern struct in_device *inetdev_init(struct net_device *dev);
 extern struct in_device	*inetdev_by_index(int);
-extern u32		inet_select_addr(const struct net_device *dev, u32 dst, int scope);
+extern __be32		inet_select_addr(const struct net_device *dev, __be32 dst, int scope);
 extern u32		inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scope);
 extern struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, u32 prefix, u32 mask);
 extern void		inet_forward_change(void);
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 48e1ccb2ff55..db72339c21e1 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -330,10 +330,10 @@ static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb)
 
 static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
 {
-	u32 saddr = 0;
+	__be32 saddr = 0;
 	u8  *dst_ha = NULL;
 	struct net_device *dev = neigh->dev;
-	u32 target = *(u32*)neigh->primary_key;
+	__be32 target = *(__be32*)neigh->primary_key;
 	int probes = atomic_read(&neigh->probes);
 	struct in_device *in_dev = in_dev_get(dev);
 
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 8e8d1f17d77a..bffbbecef455 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -876,9 +876,9 @@ out:
 	return done;
 }
 
-u32 inet_select_addr(const struct net_device *dev, u32 dst, int scope)
+__be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope)
 {
-	u32 addr = 0;
+	__be32 addr = 0;
 	struct in_device *in_dev;
 
 	rcu_read_lock();
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index c2ad07e48ab4..fd39685241c1 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -437,7 +437,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, u32 info)
 	struct icmp_bxm icmp_param;
 	struct rtable *rt = (struct rtable *)skb_in->dst;
 	struct ipcm_cookie ipc;
-	u32 saddr;
+	__be32 saddr;
 	u8  tos;
 
 	if (!rt)
diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c
index 1bca714bda3d..0c8d20da6139 100644
--- a/net/ipv4/ipvs/ip_vs_sync.c
+++ b/net/ipv4/ipvs/ip_vs_sync.c
@@ -464,7 +464,7 @@ join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname)
 static int bind_mcastif_addr(struct socket *sock, char *ifname)
 {
 	struct net_device *dev;
-	u32 addr;
+	__be32 addr;
 	struct sockaddr_in sin;
 
 	if ((dev = __dev_get_by_name(ifname)) == NULL)
diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c
index bc65168a3437..3dbfcfac8a84 100644
--- a/net/ipv4/netfilter/ipt_MASQUERADE.c
+++ b/net/ipv4/netfilter/ipt_MASQUERADE.c
@@ -70,7 +70,7 @@ masquerade_target(struct sk_buff **pskb,
 	const struct ip_nat_multi_range_compat *mr;
 	struct ip_nat_range newrange;
 	struct rtable *rt;
-	u_int32_t newsrc;
+	__be32 newsrc;
 
 	IP_NF_ASSERT(hooknum == NF_IP_POST_ROUTING);
 
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 32fcb77295f0..9d8fbf1e5bc3 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1532,7 +1532,7 @@ static int ip_rt_bug(struct sk_buff *skb)
 
 void ip_rt_get_source(u8 *addr, struct rtable *rt)
 {
-	u32 src;
+	__be32 src;
 	struct fib_result res;
 
 	if (rt->fl.iif == 0)
@@ -1603,7 +1603,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 {
 	unsigned hash;
 	struct rtable *rth;
-	u32 spec_dst;
+	__be32 spec_dst;
 	struct in_device *in_dev = in_dev_get(dev);
 	u32 itag = 0;
 
-- 
cgit v1.2.3


From d9c9df8c9368f4102324e8c3923edae83974602b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:28:14 -0700
Subject: [IPV4]: fib_validate_source() annotations

annotated arguments and inferred net-endian variables in callers

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    | 4 ++--
 net/ipv4/fib_frontend.c | 4 ++--
 net/ipv4/route.c        | 3 ++-
 3 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index fcc159a4ac17..c4eca2575b84 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -222,8 +222,8 @@ extern int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *ar
 extern int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
 extern int inet_rtm_getroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
 extern int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb);
-extern int fib_validate_source(u32 src, u32 dst, u8 tos, int oif,
-			       struct net_device *dev, u32 *spec_dst, u32 *itag);
+extern int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
+			       struct net_device *dev, __be32 *spec_dst, u32 *itag);
 extern void fib_select_multipath(const struct flowi *flp, struct fib_result *res);
 
 struct rtentry;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index cfb527c060e4..684b91f7c2ae 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -180,8 +180,8 @@ unsigned inet_addr_type(u32 addr)
    - check, that packet arrived from expected physical interface.
  */
 
-int fib_validate_source(u32 src, u32 dst, u8 tos, int oif,
-			struct net_device *dev, u32 *spec_dst, u32 *itag)
+int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
+			struct net_device *dev, __be32 *spec_dst, u32 *itag)
 {
 	struct in_device *in_dev;
 	struct flowi fl = { .nl_u = { .ip4_u =
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 9d8fbf1e5bc3..4bb66fee4543 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1722,7 +1722,8 @@ static inline int __mkroute_input(struct sk_buff *skb,
 	int err;
 	struct in_device *out_dev;
 	unsigned flags = 0;
-	u32 spec_dst, itag;
+	__be32 spec_dst;
+	u32 itag;
 
 	/* get a working reference to the output device */
 	out_dev = in_dev_get(FIB_RES_DEV(*res));
-- 
cgit v1.2.3


From a60c4923da795c74db9ff61a60e2f1df5754e4ce Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 21:28:34 -0700
Subject: [IPV4]: ip_check_mc() annotations

annotated arguments

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h | 2 +-
 net/ipv4/igmp.c      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 899c3d4776f3..8e7eedb3a574 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -197,7 +197,7 @@ struct ip_mc_list
 #define IGMPV3_QQIC(value) IGMPV3_EXP(0x80, 4, 3, value)
 #define IGMPV3_MRC(value) IGMPV3_EXP(0x80, 4, 3, value)
 
-extern int ip_check_mc(struct in_device *dev, u32 mc_addr, u32 src_addr, u16 proto);
+extern int ip_check_mc(struct in_device *dev, __be32 mc_addr, __be32 src_addr, u16 proto);
 extern int igmp_rcv(struct sk_buff *);
 extern int ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr);
 extern int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr);
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 58be8227b0cb..d8068c483781 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -2216,7 +2216,7 @@ void ip_mc_drop_socket(struct sock *sk)
 	rtnl_unlock();
 }
 
-int ip_check_mc(struct in_device *in_dev, u32 mc_addr, u32 src_addr, u16 proto)
+int ip_check_mc(struct in_device *in_dev, __be32 mc_addr, __be32 src_addr, u16 proto)
 {
 	struct ip_mc_list *im;
 	struct ip_sf_list *psf;
-- 
cgit v1.2.3


From ff428d72c59b35e4ba34bc1b487e707648010fe3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:13:35 -0700
Subject: [IPV4]: inet_addr_onlink() annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h | 2 +-
 net/ipv4/devinet.c         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 40d07d0b896b..5ae09372c144 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -105,7 +105,7 @@ extern int register_inetaddr_notifier(struct notifier_block *nb);
 extern int unregister_inetaddr_notifier(struct notifier_block *nb);
 
 extern struct net_device 	*ip_dev_find(u32 addr);
-extern int		inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b);
+extern int		inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b);
 extern int		devinet_ioctl(unsigned int cmd, void __user *);
 extern void		devinet_init(void);
 extern struct in_device *inetdev_init(struct net_device *dev);
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index bffbbecef455..5988584f6a6d 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -224,7 +224,7 @@ static void inetdev_destroy(struct in_device *in_dev)
 	call_rcu(&in_dev->rcu_head, in_dev_rcu_put);
 }
 
-int inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b)
+int inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b)
 {
 	rcu_read_lock();
 	for_primary_ifa(in_dev) {
-- 
cgit v1.2.3


From ed49e3caaa6126f8e29f08e8b4fdcafcae431b57 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:13:54 -0700
Subject: [IPV4]: fib_hn ->nh_gw is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index c4eca2575b84..9f84e9fc1b1f 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -63,7 +63,7 @@ struct fib_nh {
 	__u32			nh_tclassid;
 #endif
 	int			nh_oif;
-	u32			nh_gw;
+	__be32			nh_gw;
 };
 
 /*
-- 
cgit v1.2.3


From b83738ae003dde613712ddb1e90a8a01f5587b51 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:14:15 -0700
Subject: [IPV4]: FIB_RES_PREFSRC() annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     | 4 ++--
 net/ipv4/fib_semantics.c | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 9f84e9fc1b1f..ece746ff4ef8 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -78,7 +78,7 @@ struct fib_info {
 	int			fib_dead;
 	unsigned		fib_flags;
 	int			fib_protocol;
-	u32			fib_prefsrc;
+	__be32			fib_prefsrc;
 	u32			fib_priority;
 	u32			fib_metrics[RTAX_MAX];
 #define fib_mtu fib_metrics[RTAX_MTU-1]
@@ -232,7 +232,7 @@ struct rtentry;
 extern int ip_fib_check_default(u32 gw, struct net_device *dev);
 extern int fib_sync_down(u32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
-extern u32  __fib_res_prefsrc(struct fib_result *res);
+extern __be32  __fib_res_prefsrc(struct fib_result *res);
 
 /* Exported by fib_hash.c */
 extern struct fib_table *fib_hash_init(u32 id);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 2ead09543f68..152d9a264fab 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -923,7 +923,7 @@ out_fill_res:
 
 /* Find appropriate source address to this destination */
 
-u32 __fib_res_prefsrc(struct fib_result *res)
+__be32 __fib_res_prefsrc(struct fib_result *res)
 {
 	return inet_select_addr(FIB_RES_DEV(*res), FIB_RES_GW(*res), res->scope);
 }
-- 
cgit v1.2.3


From 00012e5bb9527022cbc843c5d372b282dbe6c4af Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:14:41 -0700
Subject: [IPV4]: introduce nla_get_be32()/NLA_PUT_BE32()

net-endian counterparts of nla_get_u32()/NLA_PUT_U32()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 4ab68a7a636a..ce5cba19c393 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -831,6 +831,9 @@ static inline int nla_put_msecs(struct sk_buff *skb, int attrtype,
 #define NLA_PUT_U32(skb, attrtype, value) \
 	NLA_PUT_TYPE(skb, u32, attrtype, value)
 
+#define NLA_PUT_BE32(skb, attrtype, value) \
+	NLA_PUT_TYPE(skb, __be32, attrtype, value)
+
 #define NLA_PUT_U64(skb, attrtype, value) \
 	NLA_PUT_TYPE(skb, u64, attrtype, value)
 
@@ -852,6 +855,15 @@ static inline u32 nla_get_u32(struct nlattr *nla)
 	return *(u32 *) nla_data(nla);
 }
 
+/**
+ * nla_get_be32 - return payload of __be32 attribute
+ * @nla: __be32 netlink attribute
+ */
+static inline __be32 nla_get_be32(struct nlattr *nla)
+{
+	return *(__be32 *) nla_data(nla);
+}
+
 /**
  * nla_get_u16 - return payload of u16 attribute
  * @nla: u16 netlink attribute
-- 
cgit v1.2.3


From 6d85c10abe840e98cbac673202fe7cc9ada2180c Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:15:46 -0700
Subject: [IPV4]: struct fib_config IPv4 address fields annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h    | 8 ++++----
 net/ipv4/fib_frontend.c | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index ece746ff4ef8..cacf4c52a8c0 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -30,13 +30,13 @@ struct fib_config {
 	u8			fc_type;
 	/* 1 byte unused */
 	u32			fc_table;
-	u32			fc_dst;
-	u32			fc_src;
-	u32			fc_gw;
+	__be32			fc_dst;
+	__be32			fc_src;
+	__be32			fc_gw;
 	int			fc_oif;
 	u32			fc_flags;
 	u32			fc_priority;
-	u32			fc_prefsrc;
+	__be32			fc_prefsrc;
 	struct nlattr		*fc_mx;
 	struct rtnexthop	*fc_mp;
 	int			fc_mx_len;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 319aaafd3545..7f5217907e5a 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -273,7 +273,7 @@ static int put_rtax(struct nlattr *mx, int len, int type, u32 value)
 static int rtentry_to_fib_config(int cmd, struct rtentry *rt,
 				 struct fib_config *cfg)
 {
-	u32 addr;
+	__be32 addr;
 	int plen;
 
 	memset(cfg, 0, sizeof(*cfg));
-- 
cgit v1.2.3


From a144ea4b7a13087081ab5402fa9ad0bcfd249e67 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 18:00:55 -0700
Subject: [IPV4]: annotate struct in_ifaddr

ifa_local, ifa_address, ifa_mask, ifa_broadcast and ifa_anycast are
net-endian.  Annotated them and variables that are inferred to be
net-endian.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/ia64/hp/sim/simeth.c                    |  4 ++--
 arch/um/drivers/net_kern.c                   |  2 +-
 arch/xtensa/platform-iss/network.c           |  2 +-
 drivers/isdn/i4l/isdn_net.c                  |  4 ++--
 drivers/net/bonding/bond_main.c              |  2 +-
 drivers/net/wan/hdlc_cisco.c                 |  2 +-
 drivers/net/wan/syncppp.c                    |  2 +-
 drivers/net/wireless/strip.c                 |  4 ++--
 include/linux/inetdevice.h                   | 10 +++++-----
 net/ipv4/devinet.c                           |  4 ++--
 net/ipv4/fib_frontend.c                      | 12 ++++++------
 net/ipv4/icmp.c                              |  2 +-
 net/ipv4/netfilter/ip_conntrack_netbios_ns.c |  2 +-
 net/ipv4/netfilter/ipt_REDIRECT.c            |  2 +-
 14 files changed, 27 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/hp/sim/simeth.c b/arch/ia64/hp/sim/simeth.c
index b5195be62818..e1a1b11473e2 100644
--- a/arch/ia64/hp/sim/simeth.c
+++ b/arch/ia64/hp/sim/simeth.c
@@ -320,7 +320,7 @@ simeth_device_event(struct notifier_block *this,unsigned long event, void *ptr)
 	}
 
 	printk(KERN_INFO "simeth_device_event: %s ipaddr=0x%x\n",
-	       dev->name, htonl(ifa->ifa_local));
+	       dev->name, ntohl(ifa->ifa_local));
 
 	/*
 	 * XXX Fix me
@@ -331,7 +331,7 @@ simeth_device_event(struct notifier_block *this,unsigned long event, void *ptr)
 	local = dev->priv;
 	/* now do it for real */
 	r = event == NETDEV_UP ?
-		netdev_attach(local->simfd, dev->irq, htonl(ifa->ifa_local)):
+		netdev_attach(local->simfd, dev->irq, ntohl(ifa->ifa_local)):
 		netdev_detach(local->simfd);
 
 	printk(KERN_INFO "simeth: netdev_attach/detach: event=%s ->%d\n",
diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 664c2e2fb820..bd1178fa4e9a 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -825,7 +825,7 @@ int dev_netmask(void *d, void *m)
 	struct net_device *dev = d;
 	struct in_device *ip = dev->ip_ptr;
 	struct in_ifaddr *in;
-	__u32 *mask_out = m;
+	__be32 *mask_out = m;
 
 	if(ip == NULL) 
 		return(1);
diff --git a/arch/xtensa/platform-iss/network.c b/arch/xtensa/platform-iss/network.c
index d96164e602fe..15d64414bd60 100644
--- a/arch/xtensa/platform-iss/network.c
+++ b/arch/xtensa/platform-iss/network.c
@@ -201,7 +201,7 @@ static void dev_ip_addr(void *d, char *buf, char *bin_buf)
 	struct net_device *dev = d;
 	struct in_device *ip = dev->ip_ptr;
 	struct in_ifaddr *in;
-	u32 addr;
+	__be32 addr;
 
 	if ((ip == NULL) || ((in = ip->ifa_list) == NULL)) {
 		printk(KERN_WARNING "Device not assigned an IP address!\n");
diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c
index 43da8ae1b2ad..1f8d6ae66b41 100644
--- a/drivers/isdn/i4l/isdn_net.c
+++ b/drivers/isdn/i4l/isdn_net.c
@@ -1614,8 +1614,8 @@ isdn_net_ciscohdlck_slarp_send_reply(isdn_net_local *lp)
 	struct sk_buff *skb;
 	unsigned char *p;
 	struct in_device *in_dev = NULL;
-	u32 addr = 0;		/* local ipv4 address */
-	u32 mask = 0;		/* local netmask */
+	__be32 addr = 0;		/* local ipv4 address */
+	__be32 mask = 0;		/* local netmask */
 
 	if ((in_dev = lp->netdev->dev.ip_ptr) != NULL) {
 		/* take primary(first) address of interface */
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 0fb5f653d3ce..c0bbddae4ec4 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2252,7 +2252,7 @@ static u32 bond_glean_dev_ip(struct net_device *dev)
 {
 	struct in_device *idev;
 	struct in_ifaddr *ifa;
-	u32 addr = 0;
+	__be32 addr = 0;
 
 	if (!dev)
 		return 0;
diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index 7ec2b2f9b7ee..b0bc5ddcf1b1 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -161,7 +161,7 @@ static int cisco_rx(struct sk_buff *skb)
 	struct hdlc_header *data = (struct hdlc_header*)skb->data;
 	struct cisco_packet *cisco_data;
 	struct in_device *in_dev;
-	u32 addr, mask;
+	__be32 addr, mask;
 
 	if (skb->len < sizeof(struct hdlc_header))
 		goto rx_error;
diff --git a/drivers/net/wan/syncppp.c b/drivers/net/wan/syncppp.c
index c13b459a0137..d1173089f334 100644
--- a/drivers/net/wan/syncppp.c
+++ b/drivers/net/wan/syncppp.c
@@ -763,7 +763,7 @@ static void sppp_cisco_input (struct sppp *sp, struct sk_buff *skb)
 		{
 		struct in_device *in_dev;
 		struct in_ifaddr *ifa;
-		u32 addr = 0, mask = ~0; /* FIXME: is the mask correct? */
+		__be32 addr = 0, mask = ~0; /* FIXME: is the mask correct? */
 #ifdef CONFIG_INET
 		rcu_read_lock();
 		if ((in_dev = __in_dev_get_rcu(dev)) != NULL)
diff --git a/drivers/net/wireless/strip.c b/drivers/net/wireless/strip.c
index ccaf28e8db0a..337c692f6fd6 100644
--- a/drivers/net/wireless/strip.c
+++ b/drivers/net/wireless/strip.c
@@ -1342,7 +1342,7 @@ static unsigned char *strip_make_packet(unsigned char *buffer,
 	 * 'broadcast hub' radio (First byte of address being 0xFF means broadcast)
 	 */
 	if (haddr.c[0] == 0xFF) {
-		u32 brd = 0;
+		__be32 brd = 0;
 		struct in_device *in_dev;
 
 		rcu_read_lock();
@@ -1406,7 +1406,7 @@ static void strip_send(struct strip *strip_info, struct sk_buff *skb)
 	int doreset = (long) jiffies - strip_info->watchdog_doreset >= 0;
 	int doprobe = (long) jiffies - strip_info->watchdog_doprobe >= 0
 	    && !doreset;
-	u32 addr, brd;
+	__be32 addr, brd;
 
 	/*
 	 * 1. If we have a packet, encapsulate it and put it in the buffer
diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 5ae09372c144..54b32e8b8f65 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -90,11 +90,11 @@ struct in_ifaddr
 	struct in_ifaddr	*ifa_next;
 	struct in_device	*ifa_dev;
 	struct rcu_head		rcu_head;
-	u32			ifa_local;
-	u32			ifa_address;
-	u32			ifa_mask;
-	u32			ifa_broadcast;
-	u32			ifa_anycast;
+	__be32			ifa_local;
+	__be32			ifa_address;
+	__be32			ifa_mask;
+	__be32			ifa_broadcast;
+	__be32			ifa_anycast;
 	unsigned char		ifa_scope;
 	unsigned char		ifa_flags;
 	unsigned char		ifa_prefixlen;
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 5988584f6a6d..a0a7780e7515 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -805,7 +805,7 @@ int devinet_ioctl(unsigned int cmd, void __user *arg)
 			break;
 		ret = 0;
 		if (ifa->ifa_mask != sin->sin_addr.s_addr) {
-			u32 old_mask = ifa->ifa_mask;
+			__be32 old_mask = ifa->ifa_mask;
 			inet_del_ifa(in_dev, ifap, 0);
 			ifa->ifa_mask = sin->sin_addr.s_addr;
 			ifa->ifa_prefixlen = inet_mask_len(ifa->ifa_mask);
@@ -931,7 +931,7 @@ static u32 confirm_addr_indev(struct in_device *in_dev, u32 dst,
 			      u32 local, int scope)
 {
 	int same = 0;
-	u32 addr = 0;
+	__be32 addr = 0;
 
 	for_ifa(in_dev) {
 		if (!addr &&
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 7f5217907e5a..62ee71ee6bc9 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -667,9 +667,9 @@ void fib_add_ifaddr(struct in_ifaddr *ifa)
 	struct in_device *in_dev = ifa->ifa_dev;
 	struct net_device *dev = in_dev->dev;
 	struct in_ifaddr *prim = ifa;
-	u32 mask = ifa->ifa_mask;
-	u32 addr = ifa->ifa_local;
-	u32 prefix = ifa->ifa_address&mask;
+	__be32 mask = ifa->ifa_mask;
+	__be32 addr = ifa->ifa_local;
+	__be32 prefix = ifa->ifa_address&mask;
 
 	if (ifa->ifa_flags&IFA_F_SECONDARY) {
 		prim = inet_ifa_byprefix(in_dev, prefix, mask);
@@ -685,7 +685,7 @@ void fib_add_ifaddr(struct in_ifaddr *ifa)
 		return;
 
 	/* Add broadcast address, if it is explicitly assigned. */
-	if (ifa->ifa_broadcast && ifa->ifa_broadcast != 0xFFFFFFFF)
+	if (ifa->ifa_broadcast && ifa->ifa_broadcast != htonl(0xFFFFFFFF))
 		fib_magic(RTM_NEWROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim);
 
 	if (!ZERONET(prefix) && !(ifa->ifa_flags&IFA_F_SECONDARY) &&
@@ -707,8 +707,8 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
 	struct net_device *dev = in_dev->dev;
 	struct in_ifaddr *ifa1;
 	struct in_ifaddr *prim = ifa;
-	u32 brd = ifa->ifa_address|~ifa->ifa_mask;
-	u32 any = ifa->ifa_address&ifa->ifa_mask;
+	__be32 brd = ifa->ifa_address|~ifa->ifa_mask;
+	__be32 any = ifa->ifa_address&ifa->ifa_mask;
 #define LOCAL_OK	1
 #define BRD_OK		2
 #define BRD0_OK		4
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index fd39685241c1..428f1c91ec45 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -895,7 +895,7 @@ static void icmp_address_reply(struct sk_buff *skb)
 	if (in_dev->ifa_list &&
 	    IN_DEV_LOG_MARTIANS(in_dev) &&
 	    IN_DEV_FORWARD(in_dev)) {
-		u32 _mask, *mp;
+		__be32 _mask, *mp;
 
 		mp = skb_header_pointer(skb, 0, sizeof(_mask), &_mask);
 		BUG_ON(mp == NULL);
diff --git a/net/ipv4/netfilter/ip_conntrack_netbios_ns.c b/net/ipv4/netfilter/ip_conntrack_netbios_ns.c
index 3d0b438783db..4adec47aae32 100644
--- a/net/ipv4/netfilter/ip_conntrack_netbios_ns.c
+++ b/net/ipv4/netfilter/ip_conntrack_netbios_ns.c
@@ -48,7 +48,7 @@ static int help(struct sk_buff **pskb,
 	struct iphdr *iph = (*pskb)->nh.iph;
 	struct rtable *rt = (struct rtable *)(*pskb)->dst;
 	struct in_device *in_dev;
-	u_int32_t mask = 0;
+	__be32 mask = 0;
 
 	/* we're only interested in locally generated packets */
 	if ((*pskb)->sk == NULL)
diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c
index f03d43671c6d..c0dcfe9d610c 100644
--- a/net/ipv4/netfilter/ipt_REDIRECT.c
+++ b/net/ipv4/netfilter/ipt_REDIRECT.c
@@ -61,7 +61,7 @@ redirect_target(struct sk_buff **pskb,
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
-	u_int32_t newdst;
+	__be32 newdst;
 	const struct ip_nat_multi_range_compat *mr = targinfo;
 	struct ip_nat_range newrange;
 
-- 
cgit v1.2.3


From 60cad5da5791ceb0beefe9a79b570cca45791f50 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:17:09 -0700
Subject: [IPV4]: annotate inetdev.h helpers

inet_confirm_addr(), inet_ifa_byprefix(), ip_dev_find(), inet_make_mask() and
inet_ifa_match() annotated, along with inferred net-endian variables

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/core/addr.c |  4 ++--
 include/linux/inetdevice.h     | 10 +++++-----
 net/ipv4/devinet.c             | 12 ++++++------
 net/ipv4/fib_frontend.c        |  2 +-
 4 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 9cbf09e2052f..60d3fbdd216c 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -86,7 +86,7 @@ EXPORT_SYMBOL(rdma_copy_addr);
 int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr)
 {
 	struct net_device *dev;
-	u32 ip = ((struct sockaddr_in *) addr)->sin_addr.s_addr;
+	__be32 ip = ((struct sockaddr_in *) addr)->sin_addr.s_addr;
 	int ret;
 
 	dev = ip_dev_find(ip);
@@ -239,7 +239,7 @@ static int addr_resolve_local(struct sockaddr_in *src_in,
 {
 	struct net_device *dev;
 	u32 src_ip = src_in->sin_addr.s_addr;
-	u32 dst_ip = dst_in->sin_addr.s_addr;
+	__be32 dst_ip = dst_in->sin_addr.s_addr;
 	int ret;
 
 	dev = ip_dev_find(dst_ip);
diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 54b32e8b8f65..5a0ab04627bc 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -104,18 +104,18 @@ struct in_ifaddr
 extern int register_inetaddr_notifier(struct notifier_block *nb);
 extern int unregister_inetaddr_notifier(struct notifier_block *nb);
 
-extern struct net_device 	*ip_dev_find(u32 addr);
+extern struct net_device 	*ip_dev_find(__be32 addr);
 extern int		inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b);
 extern int		devinet_ioctl(unsigned int cmd, void __user *);
 extern void		devinet_init(void);
 extern struct in_device *inetdev_init(struct net_device *dev);
 extern struct in_device	*inetdev_by_index(int);
 extern __be32		inet_select_addr(const struct net_device *dev, __be32 dst, int scope);
-extern u32		inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scope);
-extern struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, u32 prefix, u32 mask);
+extern __be32		inet_confirm_addr(const struct net_device *dev, __be32 dst, __be32 local, int scope);
+extern struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, __be32 mask);
 extern void		inet_forward_change(void);
 
-static __inline__ int inet_ifa_match(u32 addr, struct in_ifaddr *ifa)
+static __inline__ int inet_ifa_match(__be32 addr, struct in_ifaddr *ifa)
 {
 	return !((addr^ifa->ifa_address)&ifa->ifa_mask);
 }
@@ -183,7 +183,7 @@ static inline void in_dev_put(struct in_device *idev)
 
 #endif /* __KERNEL__ */
 
-static __inline__ __u32 inet_make_mask(int logmask)
+static __inline__ __be32 inet_make_mask(int logmask)
 {
 	if (logmask)
 		return htonl(~((1<<(32-logmask))-1));
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index afba56222fb5..7602c79a389b 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -429,8 +429,8 @@ struct in_device *inetdev_by_index(int ifindex)
 
 /* Called only from RTNL semaphored context. No locks. */
 
-struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, u32 prefix,
-				    u32 mask)
+struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix,
+				    __be32 mask)
 {
 	ASSERT_RTNL();
 
@@ -927,8 +927,8 @@ out:
 	return addr;
 }
 
-static u32 confirm_addr_indev(struct in_device *in_dev, u32 dst,
-			      u32 local, int scope)
+static __be32 confirm_addr_indev(struct in_device *in_dev, __be32 dst,
+			      __be32 local, int scope)
 {
 	int same = 0;
 	__be32 addr = 0;
@@ -971,9 +971,9 @@ static u32 confirm_addr_indev(struct in_device *in_dev, u32 dst,
  * - local: address, 0=autoselect the local address
  * - scope: maximum allowed scope value for the local address
  */
-u32 inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scope)
+__be32 inet_confirm_addr(const struct net_device *dev, __be32 dst, __be32 local, int scope)
 {
-	u32 addr = 0;
+	__be32 addr = 0;
 	struct in_device *in_dev;
 
 	if (dev) {
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 62ee71ee6bc9..c0bd2f122da2 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -122,7 +122,7 @@ static void fib_flush(void)
  *	Find the first device with a given source address.
  */
 
-struct net_device * ip_dev_find(u32 addr)
+struct net_device * ip_dev_find(__be32 addr)
 {
 	struct flowi fl = { .nl_u = { .ip4_u = { .daddr = addr } } };
 	struct fib_result res;
-- 
cgit v1.2.3


From e4883014f48f8c17c17a2526cb5cb6e17c5f94e7 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:17:28 -0700
Subject: [IPV4]: icmp_send() annotation

The last argument is network-endian (it will go straight into the packet).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/icmp.h | 2 +-
 net/ipv4/icmp.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/icmp.h b/include/net/icmp.h
index 05f8ff7d9316..dc09474efcf3 100644
--- a/include/net/icmp.h
+++ b/include/net/icmp.h
@@ -38,7 +38,7 @@ struct dst_entry;
 struct net_proto_family;
 struct sk_buff;
 
-extern void	icmp_send(struct sk_buff *skb_in,  int type, int code, u32 info);
+extern void	icmp_send(struct sk_buff *skb_in,  int type, int code, __be32 info);
 extern int	icmp_rcv(struct sk_buff *skb);
 extern int	icmp_ioctl(struct sock *sk, int cmd, unsigned long arg);
 extern void	icmp_init(struct net_proto_family *ops);
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 428f1c91ec45..45396c53560d 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -430,7 +430,7 @@ out_unlock:
  *			MUST reply to only the first fragment.
  */
 
-void icmp_send(struct sk_buff *skb_in, int type, int code, u32 info)
+void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 {
 	struct iphdr *iph;
 	int room;
-- 
cgit v1.2.3


From fd6832220974809141b3981e380b78690bba8911 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:17:51 -0700
Subject: [IPV4]: inet_addr_type() annotations

argument and inferred net-endian variables in callers annotated.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_main.c      | 2 +-
 include/net/route.h               | 2 +-
 net/ipv4/arp.c                    | 4 ++--
 net/ipv4/fib_frontend.c           | 2 +-
 net/ipv4/ip_options.c             | 4 ++--
 net/ipv4/netfilter/ipt_addrtype.c | 2 +-
 net/ipv6/af_inet6.c               | 2 +-
 7 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c
index 5613b4564fa2..8364d5475ac7 100644
--- a/drivers/s390/net/qeth_main.c
+++ b/drivers/s390/net/qeth_main.c
@@ -8067,7 +8067,7 @@ qeth_arp_constructor(struct neighbour *neigh)
 	neigh->parms = neigh_parms_clone(parms);
 	rcu_read_unlock();
 
-	neigh->type = inet_addr_type(*(u32 *) neigh->primary_key);
+	neigh->type = inet_addr_type(*(__be32 *) neigh->primary_key);
 	neigh->nud_state = NUD_NOARP;
 	neigh->ops = arp_direct_ops;
 	neigh->output = neigh->ops->queue_xmit;
diff --git a/include/net/route.h b/include/net/route.h
index 63524843f6d9..58752722c968 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -120,7 +120,7 @@ extern int		ip_route_input(struct sk_buff*, __be32 dst, __be32 src, u8 tos, stru
 extern unsigned short	ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu);
 extern void		ip_rt_send_redirect(struct sk_buff *skb);
 
-extern unsigned		inet_addr_type(u32 addr);
+extern unsigned		inet_addr_type(__be32 addr);
 extern void		ip_rt_multicast_event(struct in_device *);
 extern int		ip_rt_ioctl(unsigned int cmd, void __user *arg);
 extern void		ip_rt_get_source(u8 *src, struct rtable *rt);
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index db72339c21e1..2a29a5ebc978 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -234,7 +234,7 @@ static u32 arp_hash(const void *pkey, const struct net_device *dev)
 
 static int arp_constructor(struct neighbour *neigh)
 {
-	u32 addr = *(u32*)neigh->primary_key;
+	__be32 addr = *(__be32*)neigh->primary_key;
 	struct net_device *dev = neigh->dev;
 	struct in_device *in_dev;
 	struct neigh_parms *parms;
@@ -470,7 +470,7 @@ static int arp_set_predefined(int addr_hint, unsigned char * haddr, u32 paddr, s
 int arp_find(unsigned char *haddr, struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
-	u32 paddr;
+	__be32 paddr;
 	struct neighbour *n;
 
 	if (!skb->dst) {
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index c0bd2f122da2..34dc640478a0 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -146,7 +146,7 @@ out:
 	return dev;
 }
 
-unsigned inet_addr_type(u32 addr)
+unsigned inet_addr_type(__be32 addr)
 {
 	struct flowi		fl = { .nl_u = { .ip4_u = { .daddr = addr } } };
 	struct fib_result	res;
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 87f25486fd64..3245bec23f4d 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -148,7 +148,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 					dopt->ts_needtime = 0;
 
 					if (soffset + 8 <= optlen) {
-						__u32 addr;
+						__be32 addr;
 
 						memcpy(&addr, sptr+soffset-1, 4);
 						if (inet_addr_type(addr) != RTN_LOCAL) {
@@ -396,7 +396,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb)
 					}
 					opt->ts = optptr - iph;
 					{
-						u32 addr;
+						__be32 addr;
 						memcpy(&addr, &optptr[optptr[2]-1], 4);
 						if (inet_addr_type(addr) == RTN_UNICAST)
 							break;
diff --git a/net/ipv4/netfilter/ipt_addrtype.c b/net/ipv4/netfilter/ipt_addrtype.c
index 893dae210b04..7b60eb74788b 100644
--- a/net/ipv4/netfilter/ipt_addrtype.c
+++ b/net/ipv4/netfilter/ipt_addrtype.c
@@ -22,7 +22,7 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>");
 MODULE_DESCRIPTION("iptables addrtype match");
 
-static inline int match_type(u_int32_t addr, u_int16_t mask)
+static inline int match_type(__be32 addr, u_int16_t mask)
 {
 	return !!(mask & (1 << inet_addr_type(addr)));
 }
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index bf6e8aff19d4..e94eccb99707 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -246,7 +246,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	struct sock *sk = sock->sk;
 	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
-	__u32 v4addr = 0;
+	__be32 v4addr = 0;
 	unsigned short snum;
 	int addr_type = 0;
 	int err = 0;
-- 
cgit v1.2.3


From d878e72e419db9ff4c66848375ee30a19820e4de Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:18:13 -0700
Subject: [IPV4]: ip_fib_check_default() annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     | 2 +-
 net/ipv4/fib_semantics.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index cacf4c52a8c0..9cd0b56cd758 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -229,7 +229,7 @@ extern void fib_select_multipath(const struct flowi *flp, struct fib_result *res
 struct rtentry;
 
 /* Exported by fib_semantics.c */
-extern int ip_fib_check_default(u32 gw, struct net_device *dev);
+extern int ip_fib_check_default(__be32 gw, struct net_device *dev);
 extern int fib_sync_down(u32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
 extern __be32  __fib_res_prefsrc(struct fib_result *res);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index d91244782216..d853bd2bd0d2 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -248,7 +248,7 @@ static inline unsigned int fib_devindex_hashfn(unsigned int val)
    Used only by redirect accept routine.
  */
 
-int ip_fib_check_default(u32 gw, struct net_device *dev)
+int ip_fib_check_default(__be32 gw, struct net_device *dev)
 {
 	struct hlist_head *head;
 	struct hlist_node *node;
-- 
cgit v1.2.3


From 53576d9b995605a9edc7414b900a9218c8f23b1f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:18:43 -0700
Subject: [IPV4]: inetpeer annotations

This one is interesting - we use net-endian value as search key, but
order the tree by *host-endian* comparisons of keys.  OK since we only
care about lookups.  Annotated inet_getpeer() and friends.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inetpeer.h | 4 ++--
 net/ipv4/inetpeer.c    | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h
index 0965515f40cf..925573fd2aed 100644
--- a/include/net/inetpeer.h
+++ b/include/net/inetpeer.h
@@ -22,7 +22,7 @@ struct inet_peer
 	unsigned long		dtime;		/* the time of last use of not
 						 * referenced entries */
 	atomic_t		refcnt;
-	__u32			v4daddr;	/* peer's address */
+	__be32			v4daddr;	/* peer's address */
 	__u16			avl_height;
 	__u16			ip_id_count;	/* IP ID for the next packet */
 	atomic_t		rid;		/* Frag reception counter */
@@ -33,7 +33,7 @@ struct inet_peer
 void			inet_initpeers(void) __init;
 
 /* can be called with or without local BH being disabled */
-struct inet_peer	*inet_getpeer(__u32 daddr, int create);
+struct inet_peer	*inet_getpeer(__be32 daddr, int create);
 
 extern spinlock_t inet_peer_unused_lock;
 extern struct inet_peer **inet_peer_unused_tailp;
diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c
index a675602ef295..2b1a54b59c48 100644
--- a/net/ipv4/inetpeer.c
+++ b/net/ipv4/inetpeer.c
@@ -163,7 +163,7 @@ static void unlink_from_unused(struct inet_peer *p)
 	for (u = peer_root; u != peer_avl_empty; ) {		\
 		if (daddr == u->v4daddr)			\
 			break;					\
-		if (daddr < u->v4daddr)				\
+		if ((__force __u32)daddr < (__force __u32)u->v4daddr)	\
 			v = &u->avl_left;			\
 		else						\
 			v = &u->avl_right;			\
@@ -368,7 +368,7 @@ static int cleanup_once(unsigned long ttl)
 }
 
 /* Called with or without local BH being disabled. */
-struct inet_peer *inet_getpeer(__u32 daddr, int create)
+struct inet_peer *inet_getpeer(__be32 daddr, int create)
 {
 	struct inet_peer *p, *n;
 	struct inet_peer **stack[PEER_MAXDEPTH], ***stackptr;
-- 
cgit v1.2.3


From 80e856e16a145d7f44f613d9f3d903bf459510ca Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:19:36 -0700
Subject: [IPV4]: annotate addresses in fib_result and fib_result_nl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 9cd0b56cd758..19f70896ea98 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -107,8 +107,8 @@ struct fib_result {
 	unsigned char	type;
 	unsigned char	scope;
 #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
-	__u32           network;
-	__u32           netmask;
+	__be32          network;
+	__be32          netmask;
 #endif
 	struct fib_info *fi;
 #ifdef CONFIG_IP_MULTIPLE_TABLES
@@ -117,7 +117,7 @@ struct fib_result {
 };
 
 struct fib_result_nl {
-	u32		fl_addr;   /* To be looked up*/ 
+	__be32		fl_addr;   /* To be looked up*/
 	u32		fl_fwmark; 
 	unsigned char	fl_tos;
 	unsigned char   fl_scope;
-- 
cgit v1.2.3


From d9cd66e0e593929077b5ecf87384e23db7271c6e Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:22:50 -0700
Subject: [IPV4]: multipath_set_nhinfo() annotations

multipath_set_nhinfo() (and underlying callback) take net-endian
network and netmask.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_mp_alg.h      | 4 ++--
 net/ipv4/multipath_wrandom.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_mp_alg.h b/include/net/ip_mp_alg.h
index ac747b64734c..beffdd66ad74 100644
--- a/include/net/ip_mp_alg.h
+++ b/include/net/ip_mp_alg.h
@@ -17,7 +17,7 @@ struct ip_mp_alg_ops {
 	void	(*mp_alg_select_route)(const struct flowi *flp,
 				       struct rtable *rth, struct rtable **rp);
 	void	(*mp_alg_flush)(void);
-	void	(*mp_alg_set_nhinfo)(__u32 network, __u32 netmask,
+	void	(*mp_alg_set_nhinfo)(__be32 network, __be32 netmask,
 				     unsigned char prefixlen,
 				     const struct fib_nh *nh);
 	void	(*mp_alg_remove)(struct rtable *rth);
@@ -59,7 +59,7 @@ static inline void multipath_flush(void)
 }
 
 static inline void multipath_set_nhinfo(struct rtable *rth,
-					__u32 network, __u32 netmask,
+					__be32 network, __be32 netmask,
 					unsigned char prefixlen,
 					const struct fib_nh *nh)
 {
diff --git a/net/ipv4/multipath_wrandom.c b/net/ipv4/multipath_wrandom.c
index d25ec4ae09e5..773acc9ccd7c 100644
--- a/net/ipv4/multipath_wrandom.c
+++ b/net/ipv4/multipath_wrandom.c
@@ -217,8 +217,8 @@ static void wrandom_select_route(const struct flowi *flp,
 	*rp = decision;
 }
 
-static void wrandom_set_nhinfo(__u32 network,
-			       __u32 netmask,
+static void wrandom_set_nhinfo(__be32 network,
+			       __be32 netmask,
 			       unsigned char prefixlen,
 			       const struct fib_nh *nh)
 {
-- 
cgit v1.2.3


From 8712f774dc47ec6353c9b75317d6db62e58d9367 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:27:05 -0700
Subject: [IPV4]: ip_options_build() annotations

daddr is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h      | 2 +-
 net/ipv4/ip_options.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 98f908400771..b9a5bc9487e3 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -335,7 +335,7 @@ extern int ip_net_unreachable(struct sk_buff *skb);
  *	Functions provided by ip_options.c
  */
  
-extern void ip_options_build(struct sk_buff *skb, struct ip_options *opt, u32 daddr, struct rtable *rt, int is_frag);
+extern void ip_options_build(struct sk_buff *skb, struct ip_options *opt, __be32 daddr, struct rtable *rt, int is_frag);
 extern int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb);
 extern void ip_options_fragment(struct sk_buff *skb);
 extern int ip_options_compile(struct ip_options *opt, struct sk_buff *skb);
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 3245bec23f4d..889e14180b04 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -38,7 +38,7 @@
  */
 
 void ip_options_build(struct sk_buff * skb, struct ip_options * opt,
-			    u32 daddr, struct rtable *rt, int is_frag) 
+			    __be32 daddr, struct rtable *rt, int is_frag)
 {
 	unsigned char * iph = skb->nh.raw;
 
-- 
cgit v1.2.3


From 13d8eaa06abfeb708b60fa64203a20db033088b3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 26 Sep 2006 22:27:30 -0700
Subject: [IPV4]: ip_build_and_send_pkt() annotations

saddr and daddr are net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h     | 2 +-
 net/ipv4/ip_output.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index b9a5bc9487e3..6da1229c041a 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -86,7 +86,7 @@ extern int		igmp_mc_proc_init(void);
  */
 
 extern int		ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk,
-					      u32 saddr, u32 daddr,
+					      __be32 saddr, __be32 daddr,
 					      struct ip_options *opt);
 extern int		ip_rcv(struct sk_buff *skb, struct net_device *dev,
 			       struct packet_type *pt, struct net_device *orig_dev);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 97aee76fb746..5bf2f094e317 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -118,7 +118,7 @@ static inline int ip_select_ttl(struct inet_sock *inet, struct dst_entry *dst)
  *
  */
 int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk,
-			  u32 saddr, u32 daddr, struct ip_options *opt)
+			  __be32 saddr, __be32 daddr, struct ip_options *opt)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct rtable *rt = (struct rtable *)skb->dst;
-- 
cgit v1.2.3


From 7699431301b189fca7ccbb64fe54e5a5170f8497 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Tue, 26 Sep 2006 22:28:46 -0700
Subject: [SUNRPC]: svc_{get,put}nl()

* add svc_getnl():
	Take network-endian value from buffer, convert to host-endian
	and return it.
* add svc_putnl():
	Take host-endian value, convert to network-endian and put it
	into a buffer.
* annotate svc_getu32()/svc_putu32() as dealing with network-endian.
* convert to svc_getnl(), svc_putnl().

[AV: in large part it's a carved-up Alexey's patch]

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sunrpc/svc.h        | 31 +++++++++++++++++------
 net/sunrpc/auth_gss/svcauth_gss.c | 52 +++++++++++++++++++--------------------
 net/sunrpc/svc.c                  | 44 ++++++++++++++++-----------------
 net/sunrpc/svcauth.c              |  2 +-
 net/sunrpc/svcauth_unix.c         | 22 ++++++++---------
 5 files changed, 84 insertions(+), 67 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 7b27c09b5604..5df1d319f5d5 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -78,28 +78,45 @@ struct svc_serv {
  */
 #define RPCSVC_MAXPAGES		((RPCSVC_MAXPAYLOAD+PAGE_SIZE-1)/PAGE_SIZE + 2)
 
-static inline u32 svc_getu32(struct kvec *iov)
+static inline u32 svc_getnl(struct kvec *iov)
 {
-	u32 val, *vp;
+	__be32 val, *vp;
 	vp = iov->iov_base;
 	val = *vp++;
 	iov->iov_base = (void*)vp;
-	iov->iov_len -= sizeof(u32);
+	iov->iov_len -= sizeof(__be32);
+	return ntohl(val);
+}
+
+static inline void svc_putnl(struct kvec *iov, u32 val)
+{
+	__be32 *vp = iov->iov_base + iov->iov_len;
+	*vp = htonl(val);
+	iov->iov_len += sizeof(__be32);
+}
+
+static inline __be32 svc_getu32(struct kvec *iov)
+{
+	__be32 val, *vp;
+	vp = iov->iov_base;
+	val = *vp++;
+	iov->iov_base = (void*)vp;
+	iov->iov_len -= sizeof(__be32);
 	return val;
 }
 
 static inline void svc_ungetu32(struct kvec *iov)
 {
-	u32 *vp = (u32 *)iov->iov_base;
+	__be32 *vp = (__be32 *)iov->iov_base;
 	iov->iov_base = (void *)(vp - 1);
 	iov->iov_len += sizeof(*vp);
 }
 
-static inline void svc_putu32(struct kvec *iov, u32 val)
+static inline void svc_putu32(struct kvec *iov, __be32 val)
 {
-	u32 *vp = iov->iov_base + iov->iov_len;
+	__be32 *vp = iov->iov_base + iov->iov_len;
 	*vp = val;
-	iov->iov_len += sizeof(u32);
+	iov->iov_len += sizeof(__be32);
 }
 
 	
diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index 94217ec9e2dd..cd3d77ca3a31 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -607,7 +607,7 @@ svc_safe_getnetobj(struct kvec *argv, struct xdr_netobj *o)
 
 	if (argv->iov_len < 4)
 		return -1;
-	o->len = ntohl(svc_getu32(argv));
+	o->len = svc_getnl(argv);
 	l = round_up_to_quad(o->len);
 	if (argv->iov_len < l)
 		return -1;
@@ -624,7 +624,7 @@ svc_safe_putnetobj(struct kvec *resv, struct xdr_netobj *o)
 
 	if (resv->iov_len + 4 > PAGE_SIZE)
 		return -1;
-	svc_putu32(resv, htonl(o->len));
+	svc_putnl(resv, o->len);
 	p = resv->iov_base + resv->iov_len;
 	resv->iov_len += round_up_to_quad(o->len);
 	if (resv->iov_len > PAGE_SIZE)
@@ -657,7 +657,7 @@ gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci,
 	*authp = rpc_autherr_badverf;
 	if (argv->iov_len < 4)
 		return SVC_DENIED;
-	flavor = ntohl(svc_getu32(argv));
+	flavor = svc_getnl(argv);
 	if (flavor != RPC_AUTH_GSS)
 		return SVC_DENIED;
 	if (svc_safe_getnetobj(argv, &checksum))
@@ -689,7 +689,7 @@ gss_write_null_verf(struct svc_rqst *rqstp)
 {
 	u32     *p;
 
-	svc_putu32(rqstp->rq_res.head, htonl(RPC_AUTH_NULL));
+	svc_putnl(rqstp->rq_res.head, RPC_AUTH_NULL);
 	p = rqstp->rq_res.head->iov_base + rqstp->rq_res.head->iov_len;
 	/* don't really need to check if head->iov_len > PAGE_SIZE ... */
 	*p++ = 0;
@@ -708,7 +708,7 @@ gss_write_verf(struct svc_rqst *rqstp, struct gss_ctx *ctx_id, u32 seq)
 	u32			*p;
 	struct kvec		iov;
 
-	svc_putu32(rqstp->rq_res.head, htonl(RPC_AUTH_GSS));
+	svc_putnl(rqstp->rq_res.head, RPC_AUTH_GSS);
 	xdr_seq = htonl(seq);
 
 	iov.iov_base = &xdr_seq;
@@ -805,7 +805,7 @@ unwrap_integ_data(struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx)
 	struct xdr_netobj mic;
 	struct xdr_buf integ_buf;
 
-	integ_len = ntohl(svc_getu32(&buf->head[0]));
+	integ_len = svc_getnl(&buf->head[0]);
 	if (integ_len & 3)
 		goto out;
 	if (integ_len > buf->len)
@@ -825,7 +825,7 @@ unwrap_integ_data(struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx)
 	maj_stat = gss_verify_mic(ctx, &integ_buf, &mic);
 	if (maj_stat != GSS_S_COMPLETE)
 		goto out;
-	if (ntohl(svc_getu32(&buf->head[0])) != seq)
+	if (svc_getnl(&buf->head[0]) != seq)
 		goto out;
 	stat = 0;
 out:
@@ -857,7 +857,7 @@ unwrap_priv_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gs
 
 	rqstp->rq_sendfile_ok = 0;
 
-	priv_len = ntohl(svc_getu32(&buf->head[0]));
+	priv_len = svc_getnl(&buf->head[0]);
 	if (rqstp->rq_deferred) {
 		/* Already decrypted last time through! The sequence number
 		 * check at out_seq is unnecessary but harmless: */
@@ -895,7 +895,7 @@ unwrap_priv_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gs
 	if (maj_stat != GSS_S_COMPLETE)
 		return -EINVAL;
 out_seq:
-	if (ntohl(svc_getu32(&buf->head[0])) != seq)
+	if (svc_getnl(&buf->head[0]) != seq)
 		return -EINVAL;
 	return 0;
 }
@@ -985,12 +985,12 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 
 	if (argv->iov_len < 5 * 4)
 		goto auth_err;
-	crlen = ntohl(svc_getu32(argv));
-	if (ntohl(svc_getu32(argv)) != RPC_GSS_VERSION)
+	crlen = svc_getnl(argv);
+	if (svc_getnl(argv) != RPC_GSS_VERSION)
 		goto auth_err;
-	gc->gc_proc = ntohl(svc_getu32(argv));
-	gc->gc_seq = ntohl(svc_getu32(argv));
-	gc->gc_svc = ntohl(svc_getu32(argv));
+	gc->gc_proc = svc_getnl(argv);
+	gc->gc_seq = svc_getnl(argv);
+	gc->gc_svc = svc_getnl(argv);
 	if (svc_safe_getnetobj(argv, &gc->gc_ctx))
 		goto auth_err;
 	if (crlen != round_up_to_quad(gc->gc_ctx.len) + 5 * 4)
@@ -1016,9 +1016,9 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 	case RPC_GSS_PROC_CONTINUE_INIT:
 		if (argv->iov_len < 2 * 4)
 			goto auth_err;
-		if (ntohl(svc_getu32(argv)) != RPC_AUTH_NULL)
+		if (svc_getnl(argv) != RPC_AUTH_NULL)
 			goto auth_err;
-		if (ntohl(svc_getu32(argv)) != 0)
+		if (svc_getnl(argv) != 0)
 			goto auth_err;
 		break;
 	case RPC_GSS_PROC_DATA:
@@ -1076,14 +1076,14 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 				goto drop;
 			if (resv->iov_len + 4 > PAGE_SIZE)
 				goto drop;
-			svc_putu32(resv, rpc_success);
+			svc_putnl(resv, RPC_SUCCESS);
 			if (svc_safe_putnetobj(resv, &rsip->out_handle))
 				goto drop;
 			if (resv->iov_len + 3 * 4 > PAGE_SIZE)
 				goto drop;
-			svc_putu32(resv, htonl(rsip->major_status));
-			svc_putu32(resv, htonl(rsip->minor_status));
-			svc_putu32(resv, htonl(GSS_SEQ_WIN));
+			svc_putnl(resv, rsip->major_status);
+			svc_putnl(resv, rsip->minor_status);
+			svc_putnl(resv, GSS_SEQ_WIN);
 			if (svc_safe_putnetobj(resv, &rsip->out_token))
 				goto drop;
 			rqstp->rq_client = NULL;
@@ -1093,7 +1093,7 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 		set_bit(CACHE_NEGATIVE, &rsci->h.flags);
 		if (resv->iov_len + 4 > PAGE_SIZE)
 			goto drop;
-		svc_putu32(resv, rpc_success);
+		svc_putnl(resv, RPC_SUCCESS);
 		goto complete;
 	case RPC_GSS_PROC_DATA:
 		*authp = rpcsec_gsserr_ctxproblem;
@@ -1111,8 +1111,8 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 				goto auth_err;
 			/* placeholders for length and seq. number: */
 			svcdata->body_start = resv->iov_base + resv->iov_len;
-			svc_putu32(resv, 0);
-			svc_putu32(resv, 0);
+			svc_putnl(resv, 0);
+			svc_putnl(resv, 0);
 			break;
 		case RPC_GSS_SVC_PRIVACY:
 			if (unwrap_priv_data(rqstp, &rqstp->rq_arg,
@@ -1120,8 +1120,8 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 				goto auth_err;
 			/* placeholders for length and seq. number: */
 			svcdata->body_start = resv->iov_base + resv->iov_len;
-			svc_putu32(resv, 0);
-			svc_putu32(resv, 0);
+			svc_putnl(resv, 0);
+			svc_putnl(resv, 0);
 			break;
 		default:
 			goto auth_err;
@@ -1199,7 +1199,7 @@ svcauth_gss_wrap_resp_integ(struct svc_rqst *rqstp)
 	mic.data = (u8 *)resv->iov_base + resv->iov_len + 4;
 	if (gss_get_mic(gsd->rsci->mechctx, &integ_buf, &mic))
 		goto out_err;
-	svc_putu32(resv, htonl(mic.len));
+	svc_putnl(resv, mic.len);
 	memset(mic.data + mic.len, 0,
 			round_up_to_quad(mic.len) - mic.len);
 	resv->iov_len += XDR_QUADLEN(mic.len) << 2;
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index b76a227dd3ad..6589e1ad47fa 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -284,16 +284,16 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
 	rqstp->rq_sendfile_ok = 1;
 	/* tcp needs a space for the record length... */
 	if (rqstp->rq_prot == IPPROTO_TCP)
-		svc_putu32(resv, 0);
+		svc_putnl(resv, 0);
 
 	rqstp->rq_xid = svc_getu32(argv);
 	svc_putu32(resv, rqstp->rq_xid);
 
-	dir  = ntohl(svc_getu32(argv));
-	vers = ntohl(svc_getu32(argv));
+	dir  = svc_getnl(argv);
+	vers = svc_getnl(argv);
 
 	/* First words of reply: */
-	svc_putu32(resv, xdr_one);		/* REPLY */
+	svc_putnl(resv, 1);		/* REPLY */
 
 	if (dir != 0)		/* direction != CALL */
 		goto err_bad_dir;
@@ -303,11 +303,11 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
 	/* Save position in case we later decide to reject: */
 	accept_statp = resv->iov_base + resv->iov_len;
 
-	svc_putu32(resv, xdr_zero);		/* ACCEPT */
+	svc_putnl(resv, 0);		/* ACCEPT */
 
-	rqstp->rq_prog = prog = ntohl(svc_getu32(argv));	/* program number */
-	rqstp->rq_vers = vers = ntohl(svc_getu32(argv));	/* version number */
-	rqstp->rq_proc = proc = ntohl(svc_getu32(argv));	/* procedure number */
+	rqstp->rq_prog = prog = svc_getnl(argv);	/* program number */
+	rqstp->rq_vers = vers = svc_getnl(argv);	/* version number */
+	rqstp->rq_proc = proc = svc_getnl(argv);	/* procedure number */
 
 	progp = serv->sv_program;
 
@@ -361,7 +361,7 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
 
 	/* Build the reply header. */
 	statp = resv->iov_base +resv->iov_len;
-	svc_putu32(resv, rpc_success);		/* RPC_SUCCESS */
+	svc_putnl(resv, RPC_SUCCESS);
 
 	/* Bump per-procedure stats counter */
 	procp->pc_count++;
@@ -439,10 +439,10 @@ err_bad_dir:
 
 err_bad_rpc:
 	serv->sv_stats->rpcbadfmt++;
-	svc_putu32(resv, xdr_one);	/* REJECT */
-	svc_putu32(resv, xdr_zero);	/* RPC_MISMATCH */
-	svc_putu32(resv, xdr_two);	/* Only RPCv2 supported */
-	svc_putu32(resv, xdr_two);
+	svc_putnl(resv, 1);	/* REJECT */
+	svc_putnl(resv, 0);	/* RPC_MISMATCH */
+	svc_putnl(resv, 2);	/* Only RPCv2 supported */
+	svc_putnl(resv, 2);
 	goto sendit;
 
 err_bad_auth:
@@ -450,15 +450,15 @@ err_bad_auth:
 	serv->sv_stats->rpcbadauth++;
 	/* Restore write pointer to location of accept status: */
 	xdr_ressize_check(rqstp, accept_statp);
-	svc_putu32(resv, xdr_one);	/* REJECT */
-	svc_putu32(resv, xdr_one);	/* AUTH_ERROR */
-	svc_putu32(resv, auth_stat);	/* status */
+	svc_putnl(resv, 1);	/* REJECT */
+	svc_putnl(resv, 1);	/* AUTH_ERROR */
+	svc_putnl(resv, ntohl(auth_stat));	/* status */
 	goto sendit;
 
 err_bad_prog:
 	dprintk("svc: unknown program %d\n", prog);
 	serv->sv_stats->rpcbadfmt++;
-	svc_putu32(resv, rpc_prog_unavail);
+	svc_putnl(resv, RPC_PROG_UNAVAIL);
 	goto sendit;
 
 err_bad_vers:
@@ -466,9 +466,9 @@ err_bad_vers:
 	printk("svc: unknown version (%d)\n", vers);
 #endif
 	serv->sv_stats->rpcbadfmt++;
-	svc_putu32(resv, rpc_prog_mismatch);
-	svc_putu32(resv, htonl(progp->pg_lovers));
-	svc_putu32(resv, htonl(progp->pg_hivers));
+	svc_putnl(resv, RPC_PROG_MISMATCH);
+	svc_putnl(resv, progp->pg_lovers);
+	svc_putnl(resv, progp->pg_hivers);
 	goto sendit;
 
 err_bad_proc:
@@ -476,7 +476,7 @@ err_bad_proc:
 	printk("svc: unknown procedure (%d)\n", proc);
 #endif
 	serv->sv_stats->rpcbadfmt++;
-	svc_putu32(resv, rpc_proc_unavail);
+	svc_putnl(resv, RPC_PROC_UNAVAIL);
 	goto sendit;
 
 err_garbage:
@@ -486,6 +486,6 @@ err_garbage:
 	rpc_stat = rpc_garbage_args;
 err_bad:
 	serv->sv_stats->rpcbadfmt++;
-	svc_putu32(resv, rpc_stat);
+	svc_putnl(resv, ntohl(rpc_stat));
 	goto sendit;
 }
diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c
index 5b28c6176806..3f4d1f622de8 100644
--- a/net/sunrpc/svcauth.c
+++ b/net/sunrpc/svcauth.c
@@ -42,7 +42,7 @@ svc_authenticate(struct svc_rqst *rqstp, u32 *authp)
 
 	*authp = rpc_auth_ok;
 
-	flavor = ntohl(svc_getu32(&rqstp->rq_arg.head[0]));
+	flavor = svc_getnl(&rqstp->rq_arg.head[0]);
 
 	dprintk("svc: svc_authenticate (%d)\n", flavor);
 
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index 7e5707e2d6b6..27f443b44af8 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -427,7 +427,7 @@ svcauth_null_accept(struct svc_rqst *rqstp, u32 *authp)
 		*authp = rpc_autherr_badcred;
 		return SVC_DENIED;
 	}
-	if (svc_getu32(argv) != RPC_AUTH_NULL || svc_getu32(argv) != 0) {
+	if (svc_getu32(argv) != htonl(RPC_AUTH_NULL) || svc_getu32(argv) != 0) {
 		dprintk("svc: bad null verf\n");
 		*authp = rpc_autherr_badverf;
 		return SVC_DENIED;
@@ -441,8 +441,8 @@ svcauth_null_accept(struct svc_rqst *rqstp, u32 *authp)
 		return SVC_DROP; /* kmalloc failure - client must retry */
 
 	/* Put NULL verifier */
-	svc_putu32(resv, RPC_AUTH_NULL);
-	svc_putu32(resv, 0);
+	svc_putnl(resv, RPC_AUTH_NULL);
+	svc_putnl(resv, 0);
 
 	return SVC_OK;
 }
@@ -488,31 +488,31 @@ svcauth_unix_accept(struct svc_rqst *rqstp, u32 *authp)
 
 	svc_getu32(argv);			/* length */
 	svc_getu32(argv);			/* time stamp */
-	slen = XDR_QUADLEN(ntohl(svc_getu32(argv)));	/* machname length */
+	slen = XDR_QUADLEN(svc_getnl(argv));	/* machname length */
 	if (slen > 64 || (len -= (slen + 3)*4) < 0)
 		goto badcred;
 	argv->iov_base = (void*)((u32*)argv->iov_base + slen);	/* skip machname */
 	argv->iov_len -= slen*4;
 
-	cred->cr_uid = ntohl(svc_getu32(argv));		/* uid */
-	cred->cr_gid = ntohl(svc_getu32(argv));		/* gid */
-	slen = ntohl(svc_getu32(argv));			/* gids length */
+	cred->cr_uid = svc_getnl(argv);		/* uid */
+	cred->cr_gid = svc_getnl(argv);		/* gid */
+	slen = svc_getnl(argv);			/* gids length */
 	if (slen > 16 || (len -= (slen + 2)*4) < 0)
 		goto badcred;
 	cred->cr_group_info = groups_alloc(slen);
 	if (cred->cr_group_info == NULL)
 		return SVC_DROP;
 	for (i = 0; i < slen; i++)
-		GROUP_AT(cred->cr_group_info, i) = ntohl(svc_getu32(argv));
+		GROUP_AT(cred->cr_group_info, i) = svc_getnl(argv);
 
-	if (svc_getu32(argv) != RPC_AUTH_NULL || svc_getu32(argv) != 0) {
+	if (svc_getu32(argv) != htonl(RPC_AUTH_NULL) || svc_getu32(argv) != 0) {
 		*authp = rpc_autherr_badverf;
 		return SVC_DENIED;
 	}
 
 	/* Put NULL verifier */
-	svc_putu32(resv, RPC_AUTH_NULL);
-	svc_putu32(resv, 0);
+	svc_putnl(resv, RPC_AUTH_NULL);
+	svc_putnl(resv, 0);
 
 	return SVC_OK;
 
-- 
cgit v1.2.3


From d8ed029d6000ba2e2908d9286409e4833c091b4c Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Tue, 26 Sep 2006 22:29:38 -0700
Subject: [SUNRPC]: trivial endianness annotations

pure s/u32/__be32/

[AV: large part based on Alexey's patches]

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sunrpc/auth.h         | 16 +++++------
 include/linux/sunrpc/msg_prot.h     |  2 +-
 include/linux/sunrpc/svc.h          | 14 +++++-----
 include/linux/sunrpc/svcauth.h      |  4 +--
 include/linux/sunrpc/xdr.h          | 38 +++++++++++++-------------
 include/linux/sunrpc/xprt.h         | 12 ++++-----
 net/sunrpc/auth.c                   | 12 ++++-----
 net/sunrpc/auth_gss/auth_gss.c      | 33 ++++++++++++-----------
 net/sunrpc/auth_gss/gss_krb5_seal.c |  2 +-
 net/sunrpc/auth_gss/gss_krb5_wrap.c |  4 +--
 net/sunrpc/auth_gss/svcauth_gss.c   | 24 ++++++++---------
 net/sunrpc/auth_null.c              |  8 +++---
 net/sunrpc/auth_unix.c              | 10 +++----
 net/sunrpc/clnt.c                   | 23 ++++++++--------
 net/sunrpc/pmap_clnt.c              |  6 ++---
 net/sunrpc/svc.c                    |  8 +++---
 net/sunrpc/svcauth.c                |  2 +-
 net/sunrpc/svcauth_unix.c           |  8 +++---
 net/sunrpc/svcsock.c                |  2 +-
 net/sunrpc/xdr.c                    | 54 ++++++++++++++++++-------------------
 net/sunrpc/xprt.c                   |  4 +--
 net/sunrpc/xprtsock.c               |  3 ++-
 22 files changed, 146 insertions(+), 143 deletions(-)

(limited to 'include')

diff --git a/include/linux/sunrpc/auth.h b/include/linux/sunrpc/auth.h
index a6de332e57d4..862c0d8c8381 100644
--- a/include/linux/sunrpc/auth.h
+++ b/include/linux/sunrpc/auth.h
@@ -109,13 +109,13 @@ struct rpc_credops {
 	void			(*crdestroy)(struct rpc_cred *);
 
 	int			(*crmatch)(struct auth_cred *, struct rpc_cred *, int);
-	u32 *			(*crmarshal)(struct rpc_task *, u32 *);
+	__be32 *		(*crmarshal)(struct rpc_task *, __be32 *);
 	int			(*crrefresh)(struct rpc_task *);
-	u32 *			(*crvalidate)(struct rpc_task *, u32 *);
+	__be32 *		(*crvalidate)(struct rpc_task *, __be32 *);
 	int			(*crwrap_req)(struct rpc_task *, kxdrproc_t,
-						void *, u32 *, void *);
+						void *, __be32 *, void *);
 	int			(*crunwrap_resp)(struct rpc_task *, kxdrproc_t,
-						void *, u32 *, void *);
+						void *, __be32 *, void *);
 };
 
 extern struct rpc_authops	authunix_ops;
@@ -134,10 +134,10 @@ struct rpc_cred *	rpcauth_bindcred(struct rpc_task *);
 void			rpcauth_holdcred(struct rpc_task *);
 void			put_rpccred(struct rpc_cred *);
 void			rpcauth_unbindcred(struct rpc_task *);
-u32 *			rpcauth_marshcred(struct rpc_task *, u32 *);
-u32 *			rpcauth_checkverf(struct rpc_task *, u32 *);
-int			rpcauth_wrap_req(struct rpc_task *task, kxdrproc_t encode, void *rqstp, u32 *data, void *obj);
-int			rpcauth_unwrap_resp(struct rpc_task *task, kxdrproc_t decode, void *rqstp, u32 *data, void *obj);
+__be32 *		rpcauth_marshcred(struct rpc_task *, __be32 *);
+__be32 *		rpcauth_checkverf(struct rpc_task *, __be32 *);
+int			rpcauth_wrap_req(struct rpc_task *task, kxdrproc_t encode, void *rqstp, __be32 *data, void *obj);
+int			rpcauth_unwrap_resp(struct rpc_task *task, kxdrproc_t decode, void *rqstp, __be32 *data, void *obj);
 int			rpcauth_refreshcred(struct rpc_task *);
 void			rpcauth_invalcred(struct rpc_task *);
 int			rpcauth_uptodatecred(struct rpc_task *);
diff --git a/include/linux/sunrpc/msg_prot.h b/include/linux/sunrpc/msg_prot.h
index f43f237360ae..d9f5934ac9fe 100644
--- a/include/linux/sunrpc/msg_prot.h
+++ b/include/linux/sunrpc/msg_prot.h
@@ -95,7 +95,7 @@ enum rpc_auth_stat {
  * 2GB.
  */
 
-typedef u32	rpc_fraghdr;
+typedef __be32	rpc_fraghdr;
 
 #define	RPC_LAST_STREAM_FRAGMENT	(1U << 31)
 #define	RPC_FRAGMENT_SIZE_MASK		(~RPC_LAST_STREAM_FRAGMENT)
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 5df1d319f5d5..73140ee5c638 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -147,7 +147,7 @@ struct svc_rqst {
 	short			rq_arghi;	/* pages available in argument page list */
 	short			rq_resused;	/* pages used for result */
 
-	u32			rq_xid;		/* transmission id */
+	__be32			rq_xid;		/* transmission id */
 	u32			rq_prog;	/* program number */
 	u32			rq_vers;	/* program version */
 	u32			rq_proc;	/* procedure number */
@@ -156,7 +156,7 @@ struct svc_rqst {
 				rq_secure  : 1;	/* secure port */
 
 
-	__u32			rq_daddr;	/* dest addr of request - reply from here */
+	__be32			rq_daddr;	/* dest addr of request - reply from here */
 
 	void *			rq_argp;	/* decoded arguments */
 	void *			rq_resp;	/* xdr'd results */
@@ -186,7 +186,7 @@ struct svc_rqst {
  * Check buffer bounds after decoding arguments
  */
 static inline int
-xdr_argsize_check(struct svc_rqst *rqstp, u32 *p)
+xdr_argsize_check(struct svc_rqst *rqstp, __be32 *p)
 {
 	char *cp = (char *)p;
 	struct kvec *vec = &rqstp->rq_arg.head[0];
@@ -195,7 +195,7 @@ xdr_argsize_check(struct svc_rqst *rqstp, u32 *p)
 }
 
 static inline int
-xdr_ressize_check(struct svc_rqst *rqstp, u32 *p)
+xdr_ressize_check(struct svc_rqst *rqstp, __be32 *p)
 {
 	struct kvec *vec = &rqstp->rq_res.head[0];
 	char *cp = (char*)p;
@@ -266,10 +266,10 @@ struct svc_deferred_req {
 	u32			prot;	/* protocol (UDP or TCP) */
 	struct sockaddr_in	addr;
 	struct svc_sock		*svsk;	/* where reply must go */
-	u32			daddr;	/* where reply must come from */
+	__be32			daddr;	/* where reply must come from */
 	struct cache_deferred_req handle;
 	int			argslen;
-	u32			args[0];
+	__be32			args[0];
 };
 
 /*
@@ -301,7 +301,7 @@ struct svc_version {
 	 * A return value of 0 means drop the request. 
 	 * vs_dispatch == NULL means use default dispatcher.
 	 */
-	int			(*vs_dispatch)(struct svc_rqst *, u32 *);
+	int			(*vs_dispatch)(struct svc_rqst *, __be32 *);
 };
 
 /*
diff --git a/include/linux/sunrpc/svcauth.h b/include/linux/sunrpc/svcauth.h
index 2fe2087edd66..a6601650deeb 100644
--- a/include/linux/sunrpc/svcauth.h
+++ b/include/linux/sunrpc/svcauth.h
@@ -95,7 +95,7 @@ struct auth_ops {
 	char *	name;
 	struct module *owner;
 	int	flavour;
-	int	(*accept)(struct svc_rqst *rq, u32 *authp);
+	int	(*accept)(struct svc_rqst *rq, __be32 *authp);
 	int	(*release)(struct svc_rqst *rq);
 	void	(*domain_release)(struct auth_domain *);
 	int	(*set_client)(struct svc_rqst *rq);
@@ -112,7 +112,7 @@ struct auth_ops {
 #define	SVC_COMPLETE	9
 
 
-extern int	svc_authenticate(struct svc_rqst *rqstp, u32 *authp);
+extern int	svc_authenticate(struct svc_rqst *rqstp, __be32 *authp);
 extern int	svc_authorise(struct svc_rqst *rqstp);
 extern int	svc_set_client(struct svc_rqst *rqstp);
 extern int	svc_auth_register(rpc_authflavor_t flavor, struct auth_ops *aops);
diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h
index e6d3d349506c..953723b09bc6 100644
--- a/include/linux/sunrpc/xdr.h
+++ b/include/linux/sunrpc/xdr.h
@@ -32,7 +32,7 @@ struct xdr_netobj {
  * side) or svc_rqst pointer (server side).
  * Encode functions always assume there's enough room in the buffer.
  */
-typedef int	(*kxdrproc_t)(void *rqstp, u32 *data, void *obj);
+typedef int	(*kxdrproc_t)(void *rqstp, __be32 *data, void *obj);
 
 /*
  * Basic structure for transmission/reception of a client XDR message.
@@ -88,19 +88,19 @@ struct xdr_buf {
 /*
  * Miscellaneous XDR helper functions
  */
-u32 *	xdr_encode_opaque_fixed(u32 *p, const void *ptr, unsigned int len);
-u32 *	xdr_encode_opaque(u32 *p, const void *ptr, unsigned int len);
-u32 *	xdr_encode_string(u32 *p, const char *s);
-u32 *	xdr_decode_string_inplace(u32 *p, char **sp, int *lenp, int maxlen);
-u32 *	xdr_encode_netobj(u32 *p, const struct xdr_netobj *);
-u32 *	xdr_decode_netobj(u32 *p, struct xdr_netobj *);
+__be32 *xdr_encode_opaque_fixed(__be32 *p, const void *ptr, unsigned int len);
+__be32 *xdr_encode_opaque(__be32 *p, const void *ptr, unsigned int len);
+__be32 *xdr_encode_string(__be32 *p, const char *s);
+__be32 *xdr_decode_string_inplace(__be32 *p, char **sp, int *lenp, int maxlen);
+__be32 *xdr_encode_netobj(__be32 *p, const struct xdr_netobj *);
+__be32 *xdr_decode_netobj(__be32 *p, struct xdr_netobj *);
 
 void	xdr_encode_pages(struct xdr_buf *, struct page **, unsigned int,
 			 unsigned int);
 void	xdr_inline_pages(struct xdr_buf *, unsigned int,
 			 struct page **, unsigned int, unsigned int);
 
-static inline u32 *xdr_encode_array(u32 *p, const void *s, unsigned int len)
+static inline __be32 *xdr_encode_array(__be32 *p, const void *s, unsigned int len)
 {
 	return xdr_encode_opaque(p, s, len);
 }
@@ -108,16 +108,16 @@ static inline u32 *xdr_encode_array(u32 *p, const void *s, unsigned int len)
 /*
  * Decode 64bit quantities (NFSv3 support)
  */
-static inline u32 *
-xdr_encode_hyper(u32 *p, __u64 val)
+static inline __be32 *
+xdr_encode_hyper(__be32 *p, __u64 val)
 {
 	*p++ = htonl(val >> 32);
 	*p++ = htonl(val & 0xFFFFFFFF);
 	return p;
 }
 
-static inline u32 *
-xdr_decode_hyper(u32 *p, __u64 *valp)
+static inline __be32 *
+xdr_decode_hyper(__be32 *p, __u64 *valp)
 {
 	*valp  = ((__u64) ntohl(*p++)) << 32;
 	*valp |= ntohl(*p++);
@@ -128,7 +128,7 @@ xdr_decode_hyper(u32 *p, __u64 *valp)
  * Adjust kvec to reflect end of xdr'ed data (RPC client XDR)
  */
 static inline int
-xdr_adjust_iovec(struct kvec *iov, u32 *p)
+xdr_adjust_iovec(struct kvec *iov, __be32 *p)
 {
 	return iov->iov_len = ((u8 *) p - (u8 *) iov->iov_base);
 }
@@ -180,19 +180,19 @@ extern int xdr_encode_array2(struct xdr_buf *buf, unsigned int base,
  * Provide some simple tools for XDR buffer overflow-checking etc.
  */
 struct xdr_stream {
-	uint32_t *p;		/* start of available buffer */
+	__be32 *p;		/* start of available buffer */
 	struct xdr_buf *buf;	/* XDR buffer to read/write */
 
-	uint32_t *end;		/* end of available buffer space */
+	__be32 *end;		/* end of available buffer space */
 	struct kvec *iov;	/* pointer to the current kvec */
 };
 
-extern void xdr_init_encode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p);
-extern uint32_t *xdr_reserve_space(struct xdr_stream *xdr, size_t nbytes);
+extern void xdr_init_encode(struct xdr_stream *xdr, struct xdr_buf *buf, __be32 *p);
+extern __be32 *xdr_reserve_space(struct xdr_stream *xdr, size_t nbytes);
 extern void xdr_write_pages(struct xdr_stream *xdr, struct page **pages,
 		unsigned int base, unsigned int len);
-extern void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p);
-extern uint32_t *xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes);
+extern void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, __be32 *p);
+extern __be32 *xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes);
 extern void xdr_read_pages(struct xdr_stream *xdr, unsigned int len);
 extern void xdr_enter_page(struct xdr_stream *xdr, unsigned int len);
 
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index bdeba8538c71..6cf626580752 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -79,7 +79,7 @@ struct rpc_rqst {
 	 * This is the private part
 	 */
 	struct rpc_task *	rq_task;	/* RPC task data */
-	__u32			rq_xid;		/* request XID */
+	__be32			rq_xid;		/* request XID */
 	int			rq_cong;	/* has incremented xprt->cong */
 	int			rq_received;	/* receive completed */
 	u32			rq_seqno;	/* gss seq no. used on req. */
@@ -171,9 +171,9 @@ struct rpc_xprt {
 	/*
 	 * State of TCP reply receive stuff
 	 */
-	u32			tcp_recm,	/* Fragment header */
-				tcp_xid,	/* Current XID */
-				tcp_reclen,	/* fragment length */
+	__be32			tcp_recm,	/* Fragment header */
+				tcp_xid;	/* Current XID */
+	u32			tcp_reclen,	/* fragment length */
 				tcp_offset;	/* fragment offset */
 	unsigned long		tcp_copied,	/* copied to request */
 				tcp_flags;
@@ -253,7 +253,7 @@ void			xprt_release(struct rpc_task *task);
 struct rpc_xprt *	xprt_get(struct rpc_xprt *xprt);
 void			xprt_put(struct rpc_xprt *xprt);
 
-static inline u32 *xprt_skip_transport_header(struct rpc_xprt *xprt, u32 *p)
+static inline __be32 *xprt_skip_transport_header(struct rpc_xprt *xprt, __be32 *p)
 {
 	return p + xprt->tsh_size;
 }
@@ -268,7 +268,7 @@ void			xprt_wait_for_buffer_space(struct rpc_task *task);
 void			xprt_write_space(struct rpc_xprt *xprt);
 void			xprt_update_rtt(struct rpc_task *task);
 void			xprt_adjust_cwnd(struct rpc_task *task, int result);
-struct rpc_rqst *	xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid);
+struct rpc_rqst *	xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid);
 void			xprt_complete_rqst(struct rpc_task *task, int copied);
 void			xprt_release_rqst_cong(struct rpc_task *task);
 void			xprt_disconnect(struct rpc_xprt *xprt);
diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c
index 55163af3dcaf..993ff1a5d945 100644
--- a/net/sunrpc/auth.c
+++ b/net/sunrpc/auth.c
@@ -331,8 +331,8 @@ rpcauth_unbindcred(struct rpc_task *task)
 	task->tk_msg.rpc_cred = NULL;
 }
 
-u32 *
-rpcauth_marshcred(struct rpc_task *task, u32 *p)
+__be32 *
+rpcauth_marshcred(struct rpc_task *task, __be32 *p)
 {
 	struct rpc_cred	*cred = task->tk_msg.rpc_cred;
 
@@ -342,8 +342,8 @@ rpcauth_marshcred(struct rpc_task *task, u32 *p)
 	return cred->cr_ops->crmarshal(task, p);
 }
 
-u32 *
-rpcauth_checkverf(struct rpc_task *task, u32 *p)
+__be32 *
+rpcauth_checkverf(struct rpc_task *task, __be32 *p)
 {
 	struct rpc_cred	*cred = task->tk_msg.rpc_cred;
 
@@ -355,7 +355,7 @@ rpcauth_checkverf(struct rpc_task *task, u32 *p)
 
 int
 rpcauth_wrap_req(struct rpc_task *task, kxdrproc_t encode, void *rqstp,
-		u32 *data, void *obj)
+		__be32 *data, void *obj)
 {
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 
@@ -369,7 +369,7 @@ rpcauth_wrap_req(struct rpc_task *task, kxdrproc_t encode, void *rqstp,
 
 int
 rpcauth_unwrap_resp(struct rpc_task *task, kxdrproc_t decode, void *rqstp,
-		u32 *data, void *obj)
+		__be32 *data, void *obj)
 {
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 6eed3e166ba3..a6ed2d22a6e6 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -826,14 +826,14 @@ out:
 * Marshal credentials.
 * Maybe we should keep a cached credential for performance reasons.
 */
-static u32 *
-gss_marshal(struct rpc_task *task, u32 *p)
+static __be32 *
+gss_marshal(struct rpc_task *task, __be32 *p)
 {
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 	struct gss_cred	*gss_cred = container_of(cred, struct gss_cred,
 						 gc_base);
 	struct gss_cl_ctx	*ctx = gss_cred_get_ctx(cred);
-	u32		*cred_len;
+	__be32		*cred_len;
 	struct rpc_rqst *req = task->tk_rqstp;
 	u32             maj_stat = 0;
 	struct xdr_netobj mic;
@@ -894,12 +894,12 @@ gss_refresh(struct rpc_task *task)
 	return 0;
 }
 
-static u32 *
-gss_validate(struct rpc_task *task, u32 *p)
+static __be32 *
+gss_validate(struct rpc_task *task, __be32 *p)
 {
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 	struct gss_cl_ctx *ctx = gss_cred_get_ctx(cred);
-	u32		seq;
+	__be32		seq;
 	struct kvec	iov;
 	struct xdr_buf	verf_buf;
 	struct xdr_netobj mic;
@@ -940,13 +940,14 @@ out_bad:
 
 static inline int
 gss_wrap_req_integ(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
-		kxdrproc_t encode, struct rpc_rqst *rqstp, u32 *p, void *obj)
+		kxdrproc_t encode, struct rpc_rqst *rqstp, __be32 *p, void *obj)
 {
 	struct xdr_buf	*snd_buf = &rqstp->rq_snd_buf;
 	struct xdr_buf	integ_buf;
-	u32             *integ_len = NULL;
+	__be32          *integ_len = NULL;
 	struct xdr_netobj mic;
-	u32		offset, *q;
+	u32		offset;
+	__be32		*q;
 	struct kvec	*iov;
 	u32             maj_stat = 0;
 	int		status = -EIO;
@@ -1032,13 +1033,13 @@ out:
 
 static inline int
 gss_wrap_req_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
-		kxdrproc_t encode, struct rpc_rqst *rqstp, u32 *p, void *obj)
+		kxdrproc_t encode, struct rpc_rqst *rqstp, __be32 *p, void *obj)
 {
 	struct xdr_buf	*snd_buf = &rqstp->rq_snd_buf;
 	u32		offset;
 	u32             maj_stat;
 	int		status;
-	u32		*opaque_len;
+	__be32		*opaque_len;
 	struct page	**inpages;
 	int		first;
 	int		pad;
@@ -1095,7 +1096,7 @@ gss_wrap_req_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
 
 static int
 gss_wrap_req(struct rpc_task *task,
-	     kxdrproc_t encode, void *rqstp, u32 *p, void *obj)
+	     kxdrproc_t encode, void *rqstp, __be32 *p, void *obj)
 {
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 	struct gss_cred	*gss_cred = container_of(cred, struct gss_cred,
@@ -1132,7 +1133,7 @@ out:
 
 static inline int
 gss_unwrap_resp_integ(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
-		struct rpc_rqst *rqstp, u32 **p)
+		struct rpc_rqst *rqstp, __be32 **p)
 {
 	struct xdr_buf	*rcv_buf = &rqstp->rq_rcv_buf;
 	struct xdr_buf integ_buf;
@@ -1169,7 +1170,7 @@ gss_unwrap_resp_integ(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
 
 static inline int
 gss_unwrap_resp_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
-		struct rpc_rqst *rqstp, u32 **p)
+		struct rpc_rqst *rqstp, __be32 **p)
 {
 	struct xdr_buf  *rcv_buf = &rqstp->rq_rcv_buf;
 	u32 offset;
@@ -1198,13 +1199,13 @@ gss_unwrap_resp_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
 
 static int
 gss_unwrap_resp(struct rpc_task *task,
-		kxdrproc_t decode, void *rqstp, u32 *p, void *obj)
+		kxdrproc_t decode, void *rqstp, __be32 *p, void *obj)
 {
 	struct rpc_cred *cred = task->tk_msg.rpc_cred;
 	struct gss_cred *gss_cred = container_of(cred, struct gss_cred,
 			gc_base);
 	struct gss_cl_ctx *ctx = gss_cred_get_ctx(cred);
-	u32		*savedp = p;
+	__be32		*savedp = p;
 	struct kvec	*head = ((struct rpc_rqst *)rqstp)->rq_rcv_buf.head;
 	int		savedlen = head->iov_len;
 	int             status = -EIO;
diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c
index 2f312164d6d5..08601ee4cd73 100644
--- a/net/sunrpc/auth_gss/gss_krb5_seal.c
+++ b/net/sunrpc/auth_gss/gss_krb5_seal.c
@@ -115,7 +115,7 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text,
 	krb5_hdr = ptr - 2;
 	msg_start = krb5_hdr + 24;
 
-	*(u16 *)(krb5_hdr + 2) = htons(ctx->signalg);
+	*(__be16 *)(krb5_hdr + 2) = htons(ctx->signalg);
 	memset(krb5_hdr + 4, 0xff, 4);
 
 	if (make_checksum(checksum_type, krb5_hdr, 8, text, 0, &md5cksum))
diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c
index f179415d0c38..cc45c1605f80 100644
--- a/net/sunrpc/auth_gss/gss_krb5_wrap.c
+++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c
@@ -177,9 +177,9 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 	msg_start = krb5_hdr + 24;
 	/* XXXJBF: */ BUG_ON(buf->head[0].iov_base + offset + headlen != msg_start + blocksize);
 
-	*(u16 *)(krb5_hdr + 2) = htons(kctx->signalg);
+	*(__be16 *)(krb5_hdr + 2) = htons(kctx->signalg);
 	memset(krb5_hdr + 4, 0xff, 4);
-	*(u16 *)(krb5_hdr + 4) = htons(kctx->sealalg);
+	*(__be16 *)(krb5_hdr + 4) = htons(kctx->sealalg);
 
 	make_confounder(msg_start, blocksize);
 
diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index cd3d77ca3a31..a2587e8eef7b 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -640,7 +640,7 @@ svc_safe_putnetobj(struct kvec *resv, struct xdr_netobj *o)
  */
 static int
 gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci,
-		  u32 *rpcstart, struct rpc_gss_wire_cred *gc, u32 *authp)
+		  __be32 *rpcstart, struct rpc_gss_wire_cred *gc, __be32 *authp)
 {
 	struct gss_ctx		*ctx_id = rsci->mechctx;
 	struct xdr_buf		rpchdr;
@@ -687,7 +687,7 @@ gss_verify_header(struct svc_rqst *rqstp, struct rsc *rsci,
 static int
 gss_write_null_verf(struct svc_rqst *rqstp)
 {
-	u32     *p;
+	__be32     *p;
 
 	svc_putnl(rqstp->rq_res.head, RPC_AUTH_NULL);
 	p = rqstp->rq_res.head->iov_base + rqstp->rq_res.head->iov_len;
@@ -701,11 +701,11 @@ gss_write_null_verf(struct svc_rqst *rqstp)
 static int
 gss_write_verf(struct svc_rqst *rqstp, struct gss_ctx *ctx_id, u32 seq)
 {
-	u32			xdr_seq;
+	__be32			xdr_seq;
 	u32			maj_stat;
 	struct xdr_buf		verf_data;
 	struct xdr_netobj	mic;
-	u32			*p;
+	__be32			*p;
 	struct kvec		iov;
 
 	svc_putnl(rqstp->rq_res.head, RPC_AUTH_GSS);
@@ -782,7 +782,7 @@ EXPORT_SYMBOL(svcauth_gss_register_pseudoflavor);
 static inline int
 read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj)
 {
-	u32     raw;
+	__be32  raw;
 	int     status;
 
 	status = read_bytes_from_xdr_buf(buf, base, &raw, sizeof(*obj));
@@ -905,7 +905,7 @@ struct gss_svc_data {
 	struct rpc_gss_wire_cred	clcred;
 	/* pointer to the beginning of the procedure-specific results,
 	 * which may be encrypted/checksummed in svcauth_gss_release: */
-	u32				*body_start;
+	__be32				*body_start;
 	struct rsc			*rsci;
 };
 
@@ -946,7 +946,7 @@ gss_write_init_verf(struct svc_rqst *rqstp, struct rsi *rsip)
  * response here and return SVC_COMPLETE.
  */
 static int
-svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
+svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp)
 {
 	struct kvec	*argv = &rqstp->rq_arg.head[0];
 	struct kvec	*resv = &rqstp->rq_res.head[0];
@@ -956,8 +956,8 @@ svcauth_gss_accept(struct svc_rqst *rqstp, u32 *authp)
 	struct rpc_gss_wire_cred *gc;
 	struct rsc	*rsci = NULL;
 	struct rsi	*rsip, rsikey;
-	u32		*rpcstart;
-	u32		*reject_stat = resv->iov_base + resv->iov_len;
+	__be32		*rpcstart;
+	__be32		*reject_stat = resv->iov_base + resv->iov_len;
 	int		ret;
 
 	dprintk("RPC:      svcauth_gss: argv->iov_len = %zd\n",argv->iov_len);
@@ -1156,7 +1156,7 @@ svcauth_gss_wrap_resp_integ(struct svc_rqst *rqstp)
 	struct xdr_buf integ_buf;
 	struct xdr_netobj mic;
 	struct kvec *resv;
-	u32 *p;
+	__be32 *p;
 	int integ_offset, integ_len;
 	int stat = -EINVAL;
 
@@ -1219,7 +1219,7 @@ svcauth_gss_wrap_resp_priv(struct svc_rqst *rqstp)
 	struct rpc_gss_wire_cred *gc = &gsd->clcred;
 	struct xdr_buf *resbuf = &rqstp->rq_res;
 	struct page **inpages = NULL;
-	u32 *p;
+	__be32 *p;
 	int offset, *len;
 	int pad;
 
@@ -1264,7 +1264,7 @@ svcauth_gss_wrap_resp_priv(struct svc_rqst *rqstp)
 		return -ENOMEM;
 	*len = htonl(resbuf->len - offset);
 	pad = 3 - ((resbuf->len - offset - 1)&3);
-	p = (u32 *)(resbuf->tail[0].iov_base + resbuf->tail[0].iov_len);
+	p = (__be32 *)(resbuf->tail[0].iov_base + resbuf->tail[0].iov_len);
 	memset(p, 0, pad);
 	resbuf->tail[0].iov_len += pad;
 	resbuf->len += pad;
diff --git a/net/sunrpc/auth_null.c b/net/sunrpc/auth_null.c
index 2eccffa96ba1..3be257dc32b2 100644
--- a/net/sunrpc/auth_null.c
+++ b/net/sunrpc/auth_null.c
@@ -60,8 +60,8 @@ nul_match(struct auth_cred *acred, struct rpc_cred *cred, int taskflags)
 /*
  * Marshal credential.
  */
-static u32 *
-nul_marshal(struct rpc_task *task, u32 *p)
+static __be32 *
+nul_marshal(struct rpc_task *task, __be32 *p)
 {
 	*p++ = htonl(RPC_AUTH_NULL);
 	*p++ = 0;
@@ -81,8 +81,8 @@ nul_refresh(struct rpc_task *task)
 	return 0;
 }
 
-static u32 *
-nul_validate(struct rpc_task *task, u32 *p)
+static __be32 *
+nul_validate(struct rpc_task *task, __be32 *p)
 {
 	rpc_authflavor_t	flavor;
 	u32			size;
diff --git a/net/sunrpc/auth_unix.c b/net/sunrpc/auth_unix.c
index 74c7406a1054..f7f990c9afe2 100644
--- a/net/sunrpc/auth_unix.c
+++ b/net/sunrpc/auth_unix.c
@@ -137,12 +137,12 @@ unx_match(struct auth_cred *acred, struct rpc_cred *rcred, int flags)
  * Marshal credentials.
  * Maybe we should keep a cached credential for performance reasons.
  */
-static u32 *
-unx_marshal(struct rpc_task *task, u32 *p)
+static __be32 *
+unx_marshal(struct rpc_task *task, __be32 *p)
 {
 	struct rpc_clnt	*clnt = task->tk_client;
 	struct unx_cred	*cred = (struct unx_cred *) task->tk_msg.rpc_cred;
-	u32		*base, *hold;
+	__be32		*base, *hold;
 	int		i;
 
 	*p++ = htonl(RPC_AUTH_UNIX);
@@ -178,8 +178,8 @@ unx_refresh(struct rpc_task *task)
 	return 0;
 }
 
-static u32 *
-unx_validate(struct rpc_task *task, u32 *p)
+static __be32 *
+unx_validate(struct rpc_task *task, __be32 *p)
 {
 	rpc_authflavor_t	flavor;
 	u32			size;
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 084a0ad5c64e..124ff0ceb55b 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -60,8 +60,8 @@ static void	call_refreshresult(struct rpc_task *task);
 static void	call_timeout(struct rpc_task *task);
 static void	call_connect(struct rpc_task *task);
 static void	call_connect_status(struct rpc_task *task);
-static u32 *	call_header(struct rpc_task *task);
-static u32 *	call_verify(struct rpc_task *task);
+static __be32 *	call_header(struct rpc_task *task);
+static __be32 *	call_verify(struct rpc_task *task);
 
 
 static int
@@ -782,7 +782,7 @@ call_encode(struct rpc_task *task)
 	struct xdr_buf *rcvbuf = &req->rq_rcv_buf;
 	unsigned int	bufsiz;
 	kxdrproc_t	encode;
-	u32		*p;
+	__be32		*p;
 
 	dprintk("RPC: %4d call_encode (status %d)\n", 
 				task->tk_pid, task->tk_status);
@@ -1100,7 +1100,7 @@ call_decode(struct rpc_task *task)
 	struct rpc_clnt	*clnt = task->tk_client;
 	struct rpc_rqst	*req = task->tk_rqstp;
 	kxdrproc_t	decode = task->tk_msg.rpc_proc->p_decode;
-	u32		*p;
+	__be32		*p;
 
 	dprintk("RPC: %4d call_decode (status %d)\n", 
 				task->tk_pid, task->tk_status);
@@ -1197,12 +1197,12 @@ call_refreshresult(struct rpc_task *task)
 /*
  * Call header serialization
  */
-static u32 *
+static __be32 *
 call_header(struct rpc_task *task)
 {
 	struct rpc_clnt *clnt = task->tk_client;
 	struct rpc_rqst	*req = task->tk_rqstp;
-	u32		*p = req->rq_svec[0].iov_base;
+	__be32		*p = req->rq_svec[0].iov_base;
 
 	/* FIXME: check buffer size? */
 
@@ -1221,12 +1221,13 @@ call_header(struct rpc_task *task)
 /*
  * Reply header verification
  */
-static u32 *
+static __be32 *
 call_verify(struct rpc_task *task)
 {
 	struct kvec *iov = &task->tk_rqstp->rq_rcv_buf.head[0];
 	int len = task->tk_rqstp->rq_rcv_buf.len >> 2;
-	u32	*p = iov->iov_base, n;
+	__be32	*p = iov->iov_base;
+	u32 n;
 	int error = -EACCES;
 
 	if ((task->tk_rqstp->rq_rcv_buf.len & 3) != 0) {
@@ -1303,7 +1304,7 @@ call_verify(struct rpc_task *task)
 		printk(KERN_WARNING "call_verify: auth check failed\n");
 		goto out_garbage;		/* bad verifier, retry */
 	}
-	len = p - (u32 *)iov->iov_base - 1;
+	len = p - (__be32 *)iov->iov_base - 1;
 	if (len < 0)
 		goto out_overflow;
 	switch ((n = ntohl(*p++))) {
@@ -1358,12 +1359,12 @@ out_overflow:
 	goto out_garbage;
 }
 
-static int rpcproc_encode_null(void *rqstp, u32 *data, void *obj)
+static int rpcproc_encode_null(void *rqstp, __be32 *data, void *obj)
 {
 	return 0;
 }
 
-static int rpcproc_decode_null(void *rqstp, u32 *data, void *obj)
+static int rpcproc_decode_null(void *rqstp, __be32 *data, void *obj)
 {
 	return 0;
 }
diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c
index c04609d3476a..919d5ba7ca0a 100644
--- a/net/sunrpc/pmap_clnt.c
+++ b/net/sunrpc/pmap_clnt.c
@@ -300,7 +300,7 @@ static struct rpc_clnt *pmap_create(char *hostname, struct sockaddr_in *srvaddr,
 /*
  * XDR encode/decode functions for PMAP
  */
-static int xdr_encode_mapping(struct rpc_rqst *req, u32 *p, struct portmap_args *map)
+static int xdr_encode_mapping(struct rpc_rqst *req, __be32 *p, struct portmap_args *map)
 {
 	dprintk("RPC: xdr_encode_mapping(%u, %u, %u, %u)\n",
 		map->pm_prog, map->pm_vers, map->pm_prot, map->pm_port);
@@ -313,13 +313,13 @@ static int xdr_encode_mapping(struct rpc_rqst *req, u32 *p, struct portmap_args
 	return 0;
 }
 
-static int xdr_decode_port(struct rpc_rqst *req, u32 *p, unsigned short *portp)
+static int xdr_decode_port(struct rpc_rqst *req, __be32 *p, unsigned short *portp)
 {
 	*portp = (unsigned short) ntohl(*p++);
 	return 0;
 }
 
-static int xdr_decode_bool(struct rpc_rqst *req, u32 *p, unsigned int *boolp)
+static int xdr_decode_bool(struct rpc_rqst *req, __be32 *p, unsigned int *boolp)
 {
 	*boolp = (unsigned int) ntohl(*p++);
 	return 0;
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 6589e1ad47fa..44b8d9d4c18a 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -256,11 +256,11 @@ svc_process(struct svc_serv *serv, struct svc_rqst *rqstp)
 	struct kvec *		argv = &rqstp->rq_arg.head[0];
 	struct kvec *		resv = &rqstp->rq_res.head[0];
 	kxdrproc_t		xdr;
-	u32			*statp;
-	u32			dir, prog, vers, proc,
-				auth_stat, rpc_stat;
+	__be32			*statp;
+	u32			dir, prog, vers, proc;
+	__be32			auth_stat, rpc_stat;
 	int			auth_res;
-	u32			*accept_statp;
+	__be32			*accept_statp;
 
 	rpc_stat = rpc_success;
 
diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c
index 3f4d1f622de8..8f2320aded5c 100644
--- a/net/sunrpc/svcauth.c
+++ b/net/sunrpc/svcauth.c
@@ -35,7 +35,7 @@ static struct auth_ops	*authtab[RPC_AUTH_MAXFLAVOR] = {
 };
 
 int
-svc_authenticate(struct svc_rqst *rqstp, u32 *authp)
+svc_authenticate(struct svc_rqst *rqstp, __be32 *authp)
 {
 	rpc_authflavor_t	flavor;
 	struct auth_ops		*aops;
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index 27f443b44af8..f98229fb5e10 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -145,7 +145,7 @@ static void ip_map_request(struct cache_detail *cd,
 {
 	char text_addr[20];
 	struct ip_map *im = container_of(h, struct ip_map, h);
-	__u32 addr = im->m_addr.s_addr;
+	__be32 addr = im->m_addr.s_addr;
 	
 	snprintf(text_addr, 20, "%u.%u.%u.%u",
 		 ntohl(addr) >> 24 & 0xff,
@@ -410,7 +410,7 @@ svcauth_unix_set_client(struct svc_rqst *rqstp)
 }
 
 static int
-svcauth_null_accept(struct svc_rqst *rqstp, u32 *authp)
+svcauth_null_accept(struct svc_rqst *rqstp, __be32 *authp)
 {
 	struct kvec	*argv = &rqstp->rq_arg.head[0];
 	struct kvec	*resv = &rqstp->rq_res.head[0];
@@ -472,7 +472,7 @@ struct auth_ops svcauth_null = {
 
 
 static int
-svcauth_unix_accept(struct svc_rqst *rqstp, u32 *authp)
+svcauth_unix_accept(struct svc_rqst *rqstp, __be32 *authp)
 {
 	struct kvec	*argv = &rqstp->rq_arg.head[0];
 	struct kvec	*resv = &rqstp->rq_res.head[0];
@@ -491,7 +491,7 @@ svcauth_unix_accept(struct svc_rqst *rqstp, u32 *authp)
 	slen = XDR_QUADLEN(svc_getnl(argv));	/* machname length */
 	if (slen > 64 || (len -= (slen + 3)*4) < 0)
 		goto badcred;
-	argv->iov_base = (void*)((u32*)argv->iov_base + slen);	/* skip machname */
+	argv->iov_base = (void*)((__be32*)argv->iov_base + slen);	/* skip machname */
 	argv->iov_len -= slen*4;
 
 	cred->cr_uid = svc_getnl(argv);		/* uid */
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 953aff89bcac..f09f7487051f 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1030,7 +1030,7 @@ svc_tcp_sendto(struct svc_rqst *rqstp)
 {
 	struct xdr_buf	*xbufp = &rqstp->rq_res;
 	int sent;
-	u32 reclen;
+	__be32 reclen;
 
 	/* Set up the first element of the reply kvec.
 	 * Any other kvecs that may be in use have been taken
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 6ac45103a272..9022eb8b37ed 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -18,8 +18,8 @@
 /*
  * XDR functions for basic NFS types
  */
-u32 *
-xdr_encode_netobj(u32 *p, const struct xdr_netobj *obj)
+__be32 *
+xdr_encode_netobj(__be32 *p, const struct xdr_netobj *obj)
 {
 	unsigned int	quadlen = XDR_QUADLEN(obj->len);
 
@@ -29,8 +29,8 @@ xdr_encode_netobj(u32 *p, const struct xdr_netobj *obj)
 	return p + XDR_QUADLEN(obj->len);
 }
 
-u32 *
-xdr_decode_netobj(u32 *p, struct xdr_netobj *obj)
+__be32 *
+xdr_decode_netobj(__be32 *p, struct xdr_netobj *obj)
 {
 	unsigned int	len;
 
@@ -55,7 +55,7 @@ xdr_decode_netobj(u32 *p, struct xdr_netobj *obj)
  * Returns the updated current XDR buffer position
  *
  */
-u32 *xdr_encode_opaque_fixed(u32 *p, const void *ptr, unsigned int nbytes)
+__be32 *xdr_encode_opaque_fixed(__be32 *p, const void *ptr, unsigned int nbytes)
 {
 	if (likely(nbytes != 0)) {
 		unsigned int quadlen = XDR_QUADLEN(nbytes);
@@ -79,21 +79,21 @@ EXPORT_SYMBOL(xdr_encode_opaque_fixed);
  *
  * Returns the updated current XDR buffer position
  */
-u32 *xdr_encode_opaque(u32 *p, const void *ptr, unsigned int nbytes)
+__be32 *xdr_encode_opaque(__be32 *p, const void *ptr, unsigned int nbytes)
 {
 	*p++ = htonl(nbytes);
 	return xdr_encode_opaque_fixed(p, ptr, nbytes);
 }
 EXPORT_SYMBOL(xdr_encode_opaque);
 
-u32 *
-xdr_encode_string(u32 *p, const char *string)
+__be32 *
+xdr_encode_string(__be32 *p, const char *string)
 {
 	return xdr_encode_array(p, string, strlen(string));
 }
 
-u32 *
-xdr_decode_string_inplace(u32 *p, char **sp, int *lenp, int maxlen)
+__be32 *
+xdr_decode_string_inplace(__be32 *p, char **sp, int *lenp, int maxlen)
 {
 	unsigned int	len;
 
@@ -432,7 +432,7 @@ xdr_shift_buf(struct xdr_buf *buf, size_t len)
  *	 of the buffer length, and takes care of adjusting the kvec
  *	 length for us.
  */
-void xdr_init_encode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p)
+void xdr_init_encode(struct xdr_stream *xdr, struct xdr_buf *buf, __be32 *p)
 {
 	struct kvec *iov = buf->head;
 	int scratch_len = buf->buflen - buf->page_len - buf->tail[0].iov_len;
@@ -440,8 +440,8 @@ void xdr_init_encode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p)
 	BUG_ON(scratch_len < 0);
 	xdr->buf = buf;
 	xdr->iov = iov;
-	xdr->p = (uint32_t *)((char *)iov->iov_base + iov->iov_len);
-	xdr->end = (uint32_t *)((char *)iov->iov_base + scratch_len);
+	xdr->p = (__be32 *)((char *)iov->iov_base + iov->iov_len);
+	xdr->end = (__be32 *)((char *)iov->iov_base + scratch_len);
 	BUG_ON(iov->iov_len > scratch_len);
 
 	if (p != xdr->p && p != NULL) {
@@ -465,10 +465,10 @@ EXPORT_SYMBOL(xdr_init_encode);
  * bytes of data. If so, update the total xdr_buf length, and
  * adjust the length of the current kvec.
  */
-uint32_t * xdr_reserve_space(struct xdr_stream *xdr, size_t nbytes)
+__be32 * xdr_reserve_space(struct xdr_stream *xdr, size_t nbytes)
 {
-	uint32_t *p = xdr->p;
-	uint32_t *q;
+	__be32 *p = xdr->p;
+	__be32 *q;
 
 	/* align nbytes on the next 32-bit boundary */
 	nbytes += 3;
@@ -524,7 +524,7 @@ EXPORT_SYMBOL(xdr_write_pages);
  * @buf: pointer to XDR buffer from which to decode data
  * @p: current pointer inside XDR buffer
  */
-void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p)
+void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, __be32 *p)
 {
 	struct kvec *iov = buf->head;
 	unsigned int len = iov->iov_len;
@@ -534,7 +534,7 @@ void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, uint32_t *p)
 	xdr->buf = buf;
 	xdr->iov = iov;
 	xdr->p = p;
-	xdr->end = (uint32_t *)((char *)iov->iov_base + len);
+	xdr->end = (__be32 *)((char *)iov->iov_base + len);
 }
 EXPORT_SYMBOL(xdr_init_decode);
 
@@ -548,10 +548,10 @@ EXPORT_SYMBOL(xdr_init_decode);
  * If so return the current pointer, then update the current
  * pointer position.
  */
-uint32_t * xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes)
+__be32 * xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes)
 {
-	uint32_t *p = xdr->p;
-	uint32_t *q = p + XDR_QUADLEN(nbytes);
+	__be32 *p = xdr->p;
+	__be32 *q = p + XDR_QUADLEN(nbytes);
 
 	if (unlikely(q > xdr->end || q < p))
 		return NULL;
@@ -599,8 +599,8 @@ void xdr_read_pages(struct xdr_stream *xdr, unsigned int len)
 	 * Position current pointer at beginning of tail, and
 	 * set remaining message length.
 	 */
-	xdr->p = (uint32_t *)((char *)iov->iov_base + padding);
-	xdr->end = (uint32_t *)((char *)iov->iov_base + end);
+	xdr->p = (__be32 *)((char *)iov->iov_base + padding);
+	xdr->end = (__be32 *)((char *)iov->iov_base + end);
 }
 EXPORT_SYMBOL(xdr_read_pages);
 
@@ -624,8 +624,8 @@ void xdr_enter_page(struct xdr_stream *xdr, unsigned int len)
 	 */
 	if (len > PAGE_CACHE_SIZE - xdr->buf->page_base)
 		len = PAGE_CACHE_SIZE - xdr->buf->page_base;
-	xdr->p = (uint32_t *)(kaddr + xdr->buf->page_base);
-	xdr->end = (uint32_t *)((char *)xdr->p + len);
+	xdr->p = (__be32 *)(kaddr + xdr->buf->page_base);
+	xdr->end = (__be32 *)((char *)xdr->p + len);
 }
 EXPORT_SYMBOL(xdr_enter_page);
 
@@ -743,7 +743,7 @@ out:
 int
 xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj)
 {
-	u32	raw;
+	__be32	raw;
 	int	status;
 
 	status = read_bytes_from_xdr_buf(buf, base, &raw, sizeof(*obj));
@@ -756,7 +756,7 @@ xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj)
 int
 xdr_encode_word(struct xdr_buf *buf, int base, u32 obj)
 {
-	u32	raw = htonl(obj);
+	__be32	raw = htonl(obj);
 
 	return write_bytes_to_xdr_buf(buf, base, &raw, sizeof(obj));
 }
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 1f786f68729d..80857470dc11 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -594,7 +594,7 @@ static void xprt_connect_status(struct rpc_task *task)
  * @xid: RPC XID of incoming reply
  *
  */
-struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, u32 xid)
+struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
 {
 	struct list_head *pos;
 
@@ -801,7 +801,7 @@ void xprt_reserve(struct rpc_task *task)
 	spin_unlock(&xprt->reserve_lock);
 }
 
-static inline u32 xprt_alloc_xid(struct rpc_xprt *xprt)
+static inline __be32 xprt_alloc_xid(struct rpc_xprt *xprt)
 {
 	return xprt->xid++;
 }
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 9b62923a9c06..28100e019225 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -548,7 +548,8 @@ static void xs_udp_data_ready(struct sock *sk, int len)
 	struct rpc_rqst *rovr;
 	struct sk_buff *skb;
 	int err, repsize, copied;
-	u32 _xid, *xp;
+	u32 _xid;
+	__be32 *xp;
 
 	read_lock(&sk->sk_callback_lock);
 	dprintk("RPC:      xs_udp_data_ready...\n");
-- 
cgit v1.2.3


From b219e3ac66183fc9771b94af931fb5fd41d586ec Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Thu, 6 Jul 2006 12:38:46 +0200
Subject: [Bluetooth] Integrate low-level connections into the driver model

This patch integrates the low-level connections (ACL and SCO) into the
driver model. Every connection is presented as device with the parent
set to its host controller device.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci_core.h |  6 +++
 net/bluetooth/hci_conn.c         |  8 +++-
 net/bluetooth/hci_sysfs.c        | 79 +++++++++++++++++++++++++++++++++++++++-
 3 files changed, 90 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index d84855fe7336..263e42b68e8d 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -165,6 +165,10 @@ struct hci_conn {
 	struct timer_list disc_timer;
 	struct timer_list idle_timer;
 
+	struct work_struct work;
+
+	struct device	dev;
+
 	struct hci_dev	*hdev;
 	void		*l2cap_data;
 	void		*sco_data;
@@ -412,6 +416,8 @@ static inline int hci_recv_frame(struct sk_buff *skb)
 
 int hci_register_sysfs(struct hci_dev *hdev);
 void hci_unregister_sysfs(struct hci_dev *hdev);
+void hci_conn_add_sysfs(struct hci_conn *conn);
+void hci_conn_del_sysfs(struct hci_conn *conn);
 
 #define SET_HCIDEV_DEV(hdev, pdev) ((hdev)->parent = (pdev))
 
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 420ed4d7e57e..7e9515b41cc0 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -179,6 +179,8 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst)
 	if (hdev->notify)
 		hdev->notify(hdev, HCI_NOTIFY_CONN_ADD);
 
+	hci_conn_add_sysfs(conn);
+
 	tasklet_enable(&hdev->tx_task);
 
 	return conn;
@@ -211,6 +213,8 @@ int hci_conn_del(struct hci_conn *conn)
 
 	tasklet_disable(&hdev->tx_task);
 
+	hci_conn_del_sysfs(conn);
+
 	hci_conn_hash_del(hdev, conn);
 	if (hdev->notify)
 		hdev->notify(hdev, HCI_NOTIFY_CONN_DEL);
@@ -221,7 +225,9 @@ int hci_conn_del(struct hci_conn *conn)
 
 	hci_dev_put(hdev);
 
-	kfree(conn);
+	/* will free via device release */
+	put_device(&conn->dev);
+
 	return 0;
 }
 
diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c
index 3ccb4b0b8fc2..58df4360d242 100644
--- a/net/bluetooth/hci_sysfs.c
+++ b/net/bluetooth/hci_sysfs.c
@@ -170,6 +170,32 @@ static struct device_attribute *bt_attrs[] = {
 	NULL
 };
 
+static ssize_t show_conn_type(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hci_conn *conn = dev_get_drvdata(dev);
+	return sprintf(buf, "%s\n", conn->type == ACL_LINK ? "ACL" : "SCO");
+}
+
+static ssize_t show_conn_address(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hci_conn *conn = dev_get_drvdata(dev);
+	bdaddr_t bdaddr;
+	baswap(&bdaddr, &conn->dst);
+	return sprintf(buf, "%s\n", batostr(&bdaddr));
+}
+
+#define CONN_ATTR(_name,_mode,_show,_store) \
+struct device_attribute conn_attr_##_name = __ATTR(_name,_mode,_show,_store)
+
+static CONN_ATTR(type, S_IRUGO, show_conn_type, NULL);
+static CONN_ATTR(address, S_IRUGO, show_conn_address, NULL);
+
+static struct device_attribute *conn_attrs[] = {
+	&conn_attr_type,
+	&conn_attr_address,
+	NULL
+};
+
 struct class *bt_class = NULL;
 EXPORT_SYMBOL_GPL(bt_class);
 
@@ -181,8 +207,57 @@ static struct platform_device *bt_platform;
 
 static void bt_release(struct device *dev)
 {
-	struct hci_dev *hdev = dev_get_drvdata(dev);
-	kfree(hdev);
+	void *data = dev_get_drvdata(dev);
+	kfree(data);
+}
+
+static void add_conn(void *data)
+{
+	struct hci_conn *conn = data;
+	int i;
+
+	device_register(&conn->dev);
+
+	for (i = 0; conn_attrs[i]; i++)
+		device_create_file(&conn->dev, conn_attrs[i]);
+}
+
+void hci_conn_add_sysfs(struct hci_conn *conn)
+{
+	struct hci_dev *hdev = conn->hdev;
+	bdaddr_t *ba = &conn->dst;
+
+	BT_DBG("conn %p", conn);
+
+	conn->dev.parent  = &hdev->dev;
+	conn->dev.release = bt_release;
+
+	snprintf(conn->dev.bus_id, BUS_ID_SIZE,
+			"%s%2.2X%2.2X%2.2X%2.2X%2.2X%2.2X",
+			conn->type == ACL_LINK ? "acl" : "sco",
+			ba->b[5], ba->b[4], ba->b[3],
+			ba->b[2], ba->b[1], ba->b[0]);
+
+	dev_set_drvdata(&conn->dev, conn);
+
+	INIT_WORK(&conn->work, add_conn, (void *) conn);
+
+	schedule_work(&conn->work);
+}
+
+static void del_conn(void *data)
+{
+	struct hci_conn *conn = data;
+	device_del(&conn->dev);
+}
+
+void hci_conn_del_sysfs(struct hci_conn *conn)
+{
+	BT_DBG("conn %p", conn);
+
+	INIT_WORK(&conn->work, del_conn, (void *) conn);
+
+	schedule_work(&conn->work);
 }
 
 int hci_register_sysfs(struct hci_dev *hdev)
-- 
cgit v1.2.3


From 0ac53939a06c610b394aeb0211b985804f2d2da3 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Sat, 8 Jul 2006 13:57:15 +0200
Subject: [Bluetooth] Add HCI device identifier for SDIO cards

This patch assigns the next free HCI device identifier to Bluetooth
devices based on the SDIO interface.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/hci_vhci.c | 2 +-
 include/net/bluetooth/hci.h  | 3 ++-
 net/bluetooth/hci_sysfs.c    | 4 +++-
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c
index fc2130f1776a..a278d98a9151 100644
--- a/drivers/bluetooth/hci_vhci.c
+++ b/drivers/bluetooth/hci_vhci.c
@@ -276,7 +276,7 @@ static int vhci_open(struct inode *inode, struct file *file)
 
 	data->hdev = hdev;
 
-	hdev->type = HCI_VHCI;
+	hdev->type = HCI_VIRTUAL;
 	hdev->driver_data = data;
 
 	hdev->open     = vhci_open_dev;
diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index b2bdb1aa0429..fde08f452b59 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -44,12 +44,13 @@
 #define HCI_NOTIFY_VOICE_SETTING	3
 
 /* HCI device types */
-#define HCI_VHCI	0
+#define HCI_VIRTUAL	0
 #define HCI_USB		1
 #define HCI_PCCARD	2
 #define HCI_UART	3
 #define HCI_RS232	4
 #define HCI_PCI		5
+#define HCI_SDIO	6
 
 /* HCI device quirks */
 enum {
diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c
index 58df4360d242..09c61615e961 100644
--- a/net/bluetooth/hci_sysfs.c
+++ b/net/bluetooth/hci_sysfs.c
@@ -16,7 +16,7 @@
 static inline char *typetostr(int type)
 {
 	switch (type) {
-	case HCI_VHCI:
+	case HCI_VIRTUAL:
 		return "VIRTUAL";
 	case HCI_USB:
 		return "USB";
@@ -28,6 +28,8 @@ static inline char *typetostr(int type)
 		return "RS232";
 	case HCI_PCI:
 		return "PCI";
+	case HCI_SDIO:
+		return "SDIO";
 	default:
 		return "UNKNOWN";
 	}
-- 
cgit v1.2.3


From defc761bc25643eeedee3abd6af0079ef214b55d Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Thu, 21 Sep 2006 16:04:00 +0200
Subject: [Bluetooth] Handle command complete event for exit periodic inquiry

The command complete event of the exit periodic inquiry command must
clear the HCI_INQUIRY flag and finish the HCI request.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h | 2 ++
 net/bluetooth/hci_event.c   | 1 +
 2 files changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index fde08f452b59..5f04181b8109 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -340,6 +340,8 @@ struct hci_cp_inquiry {
 
 #define OCF_INQUIRY_CANCEL	0x0002
 
+#define OCF_EXIT_PERIODIC_INQ	0x0004
+
 #define OCF_LINK_KEY_REPLY	0x000B
 struct hci_cp_link_key_reply {
 	bdaddr_t bdaddr;
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 3896dabab11d..c6cd243939e6 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -62,6 +62,7 @@ static void hci_cc_link_ctl(struct hci_dev *hdev, __u16 ocf, struct sk_buff *skb
 
 	switch (ocf) {
 	case OCF_INQUIRY_CANCEL:
+	case OCF_EXIT_PERIODIC_INQ:
 		status = *((__u8 *) skb->data);
 
 		if (status) {
-- 
cgit v1.2.3


From 1143e5a6d4d69cd36d44e0184769aa2b17041a10 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Sat, 23 Sep 2006 09:57:20 +0200
Subject: [Bluetooth] Read local version information on device init

The local version information are needed to identify certain feature
sets of devices. They must be read on device init and stored for later
use. It is also possible to access them through the device model.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci_core.h |  3 +++
 net/bluetooth/hci_core.c         |  3 +++
 net/bluetooth/hci_event.c        | 21 ++++++++++++++++++++-
 net/bluetooth/hci_sysfs.c        | 24 ++++++++++++++++++++++++
 4 files changed, 50 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 263e42b68e8d..7451a9c92d9d 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -72,6 +72,9 @@ struct hci_dev {
 	__u8		type;
 	bdaddr_t	bdaddr;
 	__u8		features[8];
+	__u8		hci_ver;
+	__u16		hci_rev;
+	__u16		manufacturer;
 	__u16		voice_setting;
 
 	__u16		pkt_type;
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 5ed474277903..338ae977a31b 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -206,6 +206,9 @@ static void hci_init_req(struct hci_dev *hdev, unsigned long opt)
 	/* Read Local Supported Features */
 	hci_send_cmd(hdev, OGF_INFO_PARAM, OCF_READ_LOCAL_FEATURES, 0, NULL);
 
+	/* Read Local Version */
+	hci_send_cmd(hdev, OGF_INFO_PARAM, OCF_READ_LOCAL_VERSION, 0, NULL);
+
 	/* Read Buffer Size (ACL mtu, max pkt, etc.) */
 	hci_send_cmd(hdev, OGF_INFO_PARAM, OCF_READ_BUFFER_SIZE, 0, NULL);
 
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index c6cd243939e6..7518bdbf34cd 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -298,6 +298,7 @@ static void hci_cc_host_ctl(struct hci_dev *hdev, __u16 ocf, struct sk_buff *skb
 /* Command Complete OGF INFO_PARAM  */
 static void hci_cc_info_param(struct hci_dev *hdev, __u16 ocf, struct sk_buff *skb)
 {
+	struct hci_rp_read_loc_version *lv;
 	struct hci_rp_read_local_features *lf;
 	struct hci_rp_read_buffer_size *bs;
 	struct hci_rp_read_bd_addr *ba;
@@ -305,6 +306,23 @@ static void hci_cc_info_param(struct hci_dev *hdev, __u16 ocf, struct sk_buff *s
 	BT_DBG("%s ocf 0x%x", hdev->name, ocf);
 
 	switch (ocf) {
+	case OCF_READ_LOCAL_VERSION:
+		lv = (struct hci_rp_read_loc_version *) skb->data;
+
+		if (lv->status) {
+			BT_DBG("%s READ_LOCAL_VERSION failed %d", hdev->name, lf->status);
+			break;
+		}
+
+		hdev->hci_ver = lv->hci_ver;
+		hdev->hci_rev = btohs(lv->hci_rev);
+		hdev->manufacturer = btohs(lv->manufacturer);
+
+		BT_DBG("%s: manufacturer %d hci_ver %d hci_rev %d", hdev->name,
+				hdev->manufacturer, hdev->hci_ver, hdev->hci_rev);
+
+		break;
+
 	case OCF_READ_LOCAL_FEATURES:
 		lf = (struct hci_rp_read_local_features *) skb->data;
 
@@ -329,7 +347,8 @@ static void hci_cc_info_param(struct hci_dev *hdev, __u16 ocf, struct sk_buff *s
 		if (hdev->features[1] & LMP_HV3)
 			hdev->pkt_type |= (HCI_HV3);
 
-		BT_DBG("%s: features 0x%x 0x%x 0x%x", hdev->name, lf->features[0], lf->features[1], lf->features[2]);
+		BT_DBG("%s: features 0x%x 0x%x 0x%x", hdev->name,
+				lf->features[0], lf->features[1], lf->features[2]);
 
 		break;
 
diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c
index 09c61615e961..a5c4804b77f4 100644
--- a/net/bluetooth/hci_sysfs.c
+++ b/net/bluetooth/hci_sysfs.c
@@ -49,6 +49,24 @@ static ssize_t show_address(struct device *dev, struct device_attribute *attr, c
 	return sprintf(buf, "%s\n", batostr(&bdaddr));
 }
 
+static ssize_t show_manufacturer(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hci_dev *hdev = dev_get_drvdata(dev);
+	return sprintf(buf, "%d\n", hdev->manufacturer);
+}
+
+static ssize_t show_hci_version(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hci_dev *hdev = dev_get_drvdata(dev);
+	return sprintf(buf, "%d\n", hdev->hci_ver);
+}
+
+static ssize_t show_hci_revision(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hci_dev *hdev = dev_get_drvdata(dev);
+	return sprintf(buf, "%d\n", hdev->hci_rev);
+}
+
 static ssize_t show_inquiry_cache(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct hci_dev *hdev = dev_get_drvdata(dev);
@@ -153,6 +171,9 @@ static ssize_t store_sniff_min_interval(struct device *dev, struct device_attrib
 
 static DEVICE_ATTR(type, S_IRUGO, show_type, NULL);
 static DEVICE_ATTR(address, S_IRUGO, show_address, NULL);
+static DEVICE_ATTR(manufacturer, S_IRUGO, show_manufacturer, NULL);
+static DEVICE_ATTR(hci_version, S_IRUGO, show_hci_version, NULL);
+static DEVICE_ATTR(hci_revision, S_IRUGO, show_hci_revision, NULL);
 static DEVICE_ATTR(inquiry_cache, S_IRUGO, show_inquiry_cache, NULL);
 
 static DEVICE_ATTR(idle_timeout, S_IRUGO | S_IWUSR,
@@ -165,6 +186,9 @@ static DEVICE_ATTR(sniff_min_interval, S_IRUGO | S_IWUSR,
 static struct device_attribute *bt_attrs[] = {
 	&dev_attr_type,
 	&dev_attr_address,
+	&dev_attr_manufacturer,
+	&dev_attr_hci_version,
+	&dev_attr_hci_revision,
 	&dev_attr_inquiry_cache,
 	&dev_attr_idle_timeout,
 	&dev_attr_sniff_max_interval,
-- 
cgit v1.2.3


From 6ac59344ef25d5f0ebadb5663cf700d25d2a3886 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Tue, 26 Sep 2006 09:43:48 +0200
Subject: [Bluetooth] Support create connection cancel command

In case of non-blocking connects it is possible that the last user
of an ACL link quits before the connection has been fully established.
This will lead to a race condition where the internal state of a
connection is closed, but the actual link has been established and is
active. In case of Bluetooth 1.2 and later devices it is possible to
call create connection cancel to abort the connect. For older devices
the disconnect timer will be used to trigger the needed disconnect.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/hci.h      |  6 ++++++
 include/net/bluetooth/hci_core.h |  9 ++++++---
 net/bluetooth/hci_conn.c         | 31 +++++++++++++++++++++++++++----
 net/bluetooth/hci_event.c        |  4 ++++
 4 files changed, 43 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 5f04181b8109..10a3eec191fd 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -297,6 +297,7 @@ struct hci_cp_host_buffer_size {
 
 /* Link Control */
 #define OGF_LINK_CTL	0x01 
+
 #define OCF_CREATE_CONN		0x0005
 struct hci_cp_create_conn {
 	bdaddr_t bdaddr;
@@ -307,6 +308,11 @@ struct hci_cp_create_conn {
 	__u8     role_switch;
 } __attribute__ ((packed));
 
+#define OCF_CREATE_CONN_CANCEL	0x0008
+struct hci_cp_create_conn_cancel {
+	bdaddr_t bdaddr;
+} __attribute__ ((packed));
+
 #define OCF_ACCEPT_CONN_REQ	0x0009
 struct hci_cp_accept_conn_req {
 	bdaddr_t bdaddr;
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 7451a9c92d9d..df22efcfcc0b 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -316,10 +316,13 @@ static inline void hci_conn_put(struct hci_conn *conn)
 	if (atomic_dec_and_test(&conn->refcnt)) {
 		unsigned long timeo;
 		if (conn->type == ACL_LINK) {
-			timeo = msecs_to_jiffies(HCI_DISCONN_TIMEOUT);
-			if (!conn->out)
-				timeo *= 2;
 			del_timer(&conn->idle_timer);
+			if (conn->state == BT_CONNECTED) {
+				timeo = msecs_to_jiffies(HCI_DISCONN_TIMEOUT);
+				if (!conn->out)
+					timeo *= 2;
+			} else
+				timeo = msecs_to_jiffies(10);
 		} else
 			timeo = msecs_to_jiffies(10);
 		mod_timer(&conn->disc_timer, jiffies + timeo);
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 7e9515b41cc0..90e3a285a17e 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -84,6 +84,20 @@ static void hci_acl_connect(struct hci_conn *conn)
 	hci_send_cmd(hdev, OGF_LINK_CTL, OCF_CREATE_CONN, sizeof(cp), &cp);
 }
 
+static void hci_acl_connect_cancel(struct hci_conn *conn)
+{
+	struct hci_cp_create_conn_cancel cp;
+
+	BT_DBG("%p", conn);
+
+	if (conn->hdev->hci_ver < 2)
+		return;
+
+	bacpy(&cp.bdaddr, &conn->dst);
+	hci_send_cmd(conn->hdev, OGF_LINK_CTL,
+				OCF_CREATE_CONN_CANCEL, sizeof(cp), &cp);
+}
+
 void hci_acl_disconn(struct hci_conn *conn, __u8 reason)
 {
 	struct hci_cp_disconnect cp;
@@ -94,7 +108,8 @@ void hci_acl_disconn(struct hci_conn *conn, __u8 reason)
 
 	cp.handle = __cpu_to_le16(conn->handle);
 	cp.reason = reason;
-	hci_send_cmd(conn->hdev, OGF_LINK_CTL, OCF_DISCONNECT, sizeof(cp), &cp);
+	hci_send_cmd(conn->hdev, OGF_LINK_CTL,
+				OCF_DISCONNECT, sizeof(cp), &cp);
 }
 
 void hci_add_sco(struct hci_conn *conn, __u16 handle)
@@ -124,12 +139,20 @@ static void hci_conn_timeout(unsigned long arg)
 		return;
 
 	hci_dev_lock(hdev);
- 	if (conn->state == BT_CONNECTED)
+
+	switch (conn->state) {
+	case BT_CONNECT:
+		hci_acl_connect_cancel(conn);
+		break;
+ 	case BT_CONNECTED:
 		hci_acl_disconn(conn, 0x13);
-	else
+		break;
+	default:
 		conn->state = BT_CLOSED;
+		break;
+	}
+
 	hci_dev_unlock(hdev);
-	return;
 }
 
 static void hci_conn_idle(unsigned long arg)
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 7518bdbf34cd..bb25484b8747 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -750,6 +750,8 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 		if (test_bit(HCI_ENCRYPT, &hdev->flags))
 			conn->link_mode |= HCI_LM_ENCRYPT;
 
+		hci_conn_hold(conn);
+
 		/* Get remote features */
 		if (conn->type == ACL_LINK) {
 			struct hci_cp_read_remote_features cp;
@@ -778,6 +780,8 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s
 			hci_send_cmd(hdev, OGF_LINK_CTL,
 				OCF_CHANGE_CONN_PTYPE, sizeof(cp), &cp);
 		}
+
+		hci_conn_put(conn);
 	} else
 		conn->state = BT_CLOSED;
 
-- 
cgit v1.2.3


From 126a336822a6594662f5898f1ddf33e6d048fcc7 Mon Sep 17 00:00:00 2001
From: Michael Chan <mchan@broadcom.com>
Date: Wed, 27 Sep 2006 16:03:07 -0700
Subject: [TG3]: Add 5722 and 5756 support.

Add IDs to support 5722 and 5756.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c       | 3 +++
 drivers/net/tg3.h       | 4 +++-
 include/linux/pci_ids.h | 2 ++
 3 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index 14e964524969..d443b7372325 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -173,6 +173,7 @@ static struct pci_device_id tg3_pci_tbl[] = {
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5705F)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5720)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5721)},
+	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5722)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5750)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5751)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5750M)},
@@ -187,6 +188,7 @@ static struct pci_device_id tg3_pci_tbl[] = {
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5754M)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5755)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5755M)},
+	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5756)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5786)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787M)},
@@ -11273,6 +11275,7 @@ static char * __devinit tg3_phy_string(struct tg3 *tp)
 	case PHY_ID_BCM5780:	return "5780";
 	case PHY_ID_BCM5755:	return "5755";
 	case PHY_ID_BCM5787:	return "5787";
+	case PHY_ID_BCM5756:	return "5722/5756";
 	case PHY_ID_BCM8002:	return "8002/serdes";
 	case 0:			return "serdes";
 	default:		return "unknown";
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index f7462c2ccc0a..feed13dc8719 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -2282,6 +2282,7 @@ struct tg3 {
 #define PHY_ID_BCM5780			0x60008350
 #define PHY_ID_BCM5755			0xbc050cc0
 #define PHY_ID_BCM5787			0xbc050ce0
+#define PHY_ID_BCM5756			0xbc050ed0
 #define PHY_ID_BCM8002			0x60010140
 #define PHY_ID_INVALID			0xffffffff
 #define PHY_ID_REV_MASK			0x0000000f
@@ -2308,7 +2309,8 @@ struct tg3 {
 	 (X) == PHY_ID_BCM5705 || (X) == PHY_ID_BCM5750 || \
 	 (X) == PHY_ID_BCM5752 || (X) == PHY_ID_BCM5714 || \
 	 (X) == PHY_ID_BCM5780 || (X) == PHY_ID_BCM5787 || \
-	 (X) == PHY_ID_BCM5755 || (X) == PHY_ID_BCM8002)
+	 (X) == PHY_ID_BCM5755 || (X) == PHY_ID_BCM5756 || \
+	 (X) == PHY_ID_BCM8002)
 
 	struct tg3_hw_stats		*hw_stats;
 	dma_addr_t			stats_mapping;
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 61db1907f06f..ea3140d226e6 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1904,6 +1904,7 @@
 #define PCI_DEVICE_ID_TIGON3_5705_2	0x1654
 #define PCI_DEVICE_ID_TIGON3_5720	0x1658
 #define PCI_DEVICE_ID_TIGON3_5721	0x1659
+#define PCI_DEVICE_ID_TIGON3_5722	0x165a
 #define PCI_DEVICE_ID_TIGON3_5705M	0x165d
 #define PCI_DEVICE_ID_TIGON3_5705M_2	0x165e
 #define PCI_DEVICE_ID_TIGON3_5714	0x1668
@@ -1913,6 +1914,7 @@
 #define PCI_DEVICE_ID_TIGON3_5705F	0x166e
 #define PCI_DEVICE_ID_TIGON3_5754M	0x1672
 #define PCI_DEVICE_ID_TIGON3_5755M	0x1673
+#define PCI_DEVICE_ID_TIGON3_5756	0x1674
 #define PCI_DEVICE_ID_TIGON3_5750	0x1676
 #define PCI_DEVICE_ID_TIGON3_5751	0x1677
 #define PCI_DEVICE_ID_TIGON3_5715	0x1678
-- 
cgit v1.2.3


From b5d3772ccbe0bc5ac8ffbb5356b74ca698aee28c Mon Sep 17 00:00:00 2001
From: Michael Chan <mchan@broadcom.com>
Date: Wed, 27 Sep 2006 16:06:21 -0700
Subject: [TG3]: Add basic 5906 support.

Add support for the new 5709 device.  This is a new 10/100 Mbps chip.
The mailbox access and firmware interface are quite different from
all other tg3 chips.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c       | 126 +++++++++++++++++++++++++++++++++++++++++++-----
 drivers/net/tg3.h       |  28 +++++++++--
 include/linux/pci_ids.h |   2 +
 3 files changed, 141 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index eafca2a0dd00..2b062d776511 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -199,6 +199,8 @@ static struct pci_device_id tg3_pci_tbl[] = {
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5780)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5780S)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5781)},
+	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5906)},
+	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5906M)},
 	{PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, PCI_DEVICE_ID_SYSKONNECT_9DXX)},
 	{PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, PCI_DEVICE_ID_SYSKONNECT_9MXX)},
 	{PCI_DEVICE(PCI_VENDOR_ID_ALTIMA, PCI_DEVICE_ID_ALTIMA_AC1000)},
@@ -426,6 +428,16 @@ static void tg3_write32_tx_mbox(struct tg3 *tp, u32 off, u32 val)
 		readl(mbox);
 }
 
+static u32 tg3_read32_mbox_5906(struct tg3 *tp, u32 off)
+{
+	return (readl(tp->regs + off + GRCMBOX_BASE));
+}
+
+static void tg3_write32_mbox_5906(struct tg3 *tp, u32 off, u32 val)
+{
+	writel(val, tp->regs + off + GRCMBOX_BASE);
+}
+
 #define tw32_mailbox(reg, val)	tp->write32_mbox(tp, reg, val)
 #define tw32_mailbox_f(reg, val)	tw32_mailbox_flush(tp, (reg), (val))
 #define tw32_rx_mbox(reg, val)	tp->write32_rx_mbox(tp, reg, val)
@@ -441,6 +453,10 @@ static void tg3_write_mem(struct tg3 *tp, u32 off, u32 val)
 {
 	unsigned long flags;
 
+	if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) &&
+	    (off >= NIC_SRAM_STATS_BLK) && (off < NIC_SRAM_TX_BUFFER_DESC))
+		return;
+
 	spin_lock_irqsave(&tp->indirect_lock, flags);
 	if (tp->tg3_flags & TG3_FLAG_SRAM_USE_CONFIG) {
 		pci_write_config_dword(tp->pdev, TG3PCI_MEM_WIN_BASE_ADDR, off);
@@ -462,6 +478,12 @@ static void tg3_read_mem(struct tg3 *tp, u32 off, u32 *val)
 {
 	unsigned long flags;
 
+	if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) &&
+	    (off >= NIC_SRAM_STATS_BLK) && (off < NIC_SRAM_TX_BUFFER_DESC)) {
+		*val = 0;
+		return;
+	}
+
 	spin_lock_irqsave(&tp->indirect_lock, flags);
 	if (tp->tg3_flags & TG3_FLAG_SRAM_USE_CONFIG) {
 		pci_write_config_dword(tp->pdev, TG3PCI_MEM_WIN_BASE_ADDR, off);
@@ -491,6 +513,9 @@ static inline void tg3_cond_int(struct tg3 *tp)
 	if (!(tp->tg3_flags & TG3_FLAG_TAGGED_STATUS) &&
 	    (tp->hw_status->status & SD_STATUS_UPDATED))
 		tw32(GRC_LOCAL_CTRL, tp->grc_local_ctrl | GRC_LCLCTRL_SETINT);
+	else
+		tw32(HOSTCC_MODE, tp->coalesce_mode |
+		     (HOSTCC_MODE_ENABLE | HOSTCC_MODE_NOW));
 }
 
 static void tg3_enable_ints(struct tg3 *tp)
@@ -656,6 +681,10 @@ static int tg3_writephy(struct tg3 *tp, int reg, u32 val)
 	unsigned int loops;
 	int ret;
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906 &&
+	    (reg == MII_TG3_CTRL || reg == MII_TG3_AUX_CTRL))
+		return 0;
+
 	if ((tp->mi_mode & MAC_MI_MODE_AUTO_POLL) != 0) {
 		tw32_f(MAC_MI_MODE,
 		     (tp->mi_mode & ~MAC_MI_MODE_AUTO_POLL));
@@ -1207,7 +1236,12 @@ static int tg3_set_power_state(struct tg3 *tp, pci_power_t state)
 		tg3_setup_phy(tp, 0);
 	}
 
-	if (!(tp->tg3_flags & TG3_FLAG_ENABLE_ASF)) {
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		u32 val;
+
+		val = tr32(GRC_VCPU_EXT_CTRL);
+		tw32(GRC_VCPU_EXT_CTRL, val | GRC_VCPU_EXT_CTRL_DISABLE_WOL);
+	} else if (!(tp->tg3_flags & TG3_FLAG_ENABLE_ASF)) {
 		int i;
 		u32 val;
 
@@ -4667,6 +4701,15 @@ static int tg3_poll_fw(struct tg3 *tp)
 	int i;
 	u32 val;
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		for (i = 0; i < 400; i++) {
+			if (tr32(VCPU_STATUS) & VCPU_STATUS_INIT_DONE)
+				return 0;
+			udelay(10);
+		}
+		return -ENODEV;
+	}
+
 	/* Wait for firmware initialization to complete. */
 	for (i = 0; i < 100000; i++) {
 		tg3_read_mem(tp, NIC_SRAM_FIRMWARE_MBOX, &val);
@@ -4735,6 +4778,12 @@ static int tg3_chip_reset(struct tg3 *tp)
 		}
 	}
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		tw32(VCPU_STATUS, tr32(VCPU_STATUS) | VCPU_STATUS_DRV_RESET);
+		tw32(GRC_VCPU_EXT_CTRL,
+		     tr32(GRC_VCPU_EXT_CTRL) & ~GRC_VCPU_EXT_CTRL_HALT_CPU);
+	}
+
 	if (tp->tg3_flags2 & TG3_FLG2_5705_PLUS)
 		val |= GRC_MISC_CFG_KEEP_GPHY_POWER;
 	tw32(GRC_MISC_CFG, val);
@@ -5066,6 +5115,12 @@ static int tg3_halt_cpu(struct tg3 *tp, u32 offset)
 	BUG_ON(offset == TX_CPU_BASE &&
 	    (tp->tg3_flags2 & TG3_FLG2_5705_PLUS));
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		u32 val = tr32(GRC_VCPU_EXT_CTRL);
+
+		tw32(GRC_VCPU_EXT_CTRL, val | GRC_VCPU_EXT_CTRL_HALT_CPU);
+		return 0;
+	}
 	if (offset == RX_CPU_BASE) {
 		for (i = 0; i < 10000; i++) {
 			tw32(offset + CPU_STATE, 0xffffffff);
@@ -6070,6 +6125,13 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 		val = 1;
 	else if (val > tp->rx_std_max_post)
 		val = tp->rx_std_max_post;
+	else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		if (tp->pci_chip_rev_id == CHIPREV_ID_5906_A1)
+			tw32(ISO_PKT_TX, (tr32(ISO_PKT_TX) & ~0x3) | 0x2);
+
+		if (val > (TG3_RX_INTERNAL_RING_SZ_5906 / 2))
+			val = TG3_RX_INTERNAL_RING_SZ_5906 / 2;
+	}
 
 	tw32(RCVBDI_STD_THRESH, val);
 
@@ -6984,9 +7046,10 @@ static int tg3_open(struct net_device *dev)
 
 		if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) {
 			if (tp->tg3_flags2 & TG3_FLG2_1SHOT_MSI) {
-				u32 val = tr32(0x7c04);
+				u32 val = tr32(PCIE_TRANSACTION_CFG);
 
-				tw32(0x7c04, val | (1 << 29));
+				tw32(PCIE_TRANSACTION_CFG,
+				     val | PCIE_TRANS_CFG_1SHOT_MSI);
 			}
 		}
 	}
@@ -7941,7 +8004,8 @@ static int tg3_set_tso(struct net_device *dev, u32 value)
 			return -EINVAL;
 		return 0;
 	}
-	if (tp->tg3_flags2 & TG3_FLG2_HW_TSO_2) {
+	if ((tp->tg3_flags2 & TG3_FLG2_HW_TSO_2) &&
+	    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)) {
 		if (value)
 			dev->features |= NETIF_F_TSO6;
 		else
@@ -9257,6 +9321,13 @@ static void __devinit tg3_get_5787_nvram_info(struct tg3 *tp)
 	}
 }
 
+static void __devinit tg3_get_5906_nvram_info(struct tg3 *tp)
+{
+	tp->nvram_jedecnum = JEDEC_ATMEL;
+	tp->tg3_flags |= TG3_FLAG_NVRAM_BUFFERED;
+	tp->nvram_pagesize = ATMEL_AT24C512_CHIP_SIZE;
+}
+
 /* Chips other than 5700/5701 use the NVRAM for fetching info. */
 static void __devinit tg3_nvram_init(struct tg3 *tp)
 {
@@ -9293,6 +9364,8 @@ static void __devinit tg3_nvram_init(struct tg3 *tp)
 			tg3_get_5755_nvram_info(tp);
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787)
 			tg3_get_5787_nvram_info(tp);
+		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
+			tg3_get_5906_nvram_info(tp);
 		else
 			tg3_get_nvram_info(tp);
 
@@ -9766,6 +9839,12 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 	/* Assume an onboard device by default.  */
 	tp->tg3_flags |= TG3_FLAG_EEPROM_WRITE_PROT;
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		if (!(tr32(PCIE_TRANSACTION_CFG) & PCIE_TRANS_CFG_LOM))
+			tp->tg3_flags &= ~TG3_FLAG_EEPROM_WRITE_PROT;
+		return;
+	}
+
 	tg3_read_mem(tp, NIC_SRAM_DATA_SIG, &val);
 	if (val == NIC_SRAM_DATA_SIG_MAGIC) {
 		u32 nic_cfg, led_cfg;
@@ -10097,7 +10176,10 @@ static void __devinit tg3_read_partno(struct tg3 *tp)
 	}
 
 out_not_found:
-	strcpy(tp->board_part_number, "none");
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
+		strcpy(tp->board_part_number, "BCM95906");
+	else
+		strcpy(tp->board_part_number, "none");
 }
 
 static void __devinit tg3_read_fw_ver(struct tg3 *tp)
@@ -10299,6 +10381,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906 ||
 	    (tp->tg3_flags2 & TG3_FLG2_5780_CLASS))
 		tp->tg3_flags2 |= TG3_FLG2_5750_PLUS;
 
@@ -10308,7 +10391,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 
 	if (tp->tg3_flags2 & TG3_FLG2_5750_PLUS) {
 		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
-		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787) {
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
 			tp->tg3_flags2 |= TG3_FLG2_HW_TSO_2;
 			tp->tg3_flags2 |= TG3_FLG2_1SHOT_MSI;
 		} else {
@@ -10325,7 +10409,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5750 &&
 	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5752 &&
 	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5755 &&
-	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5787)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5787 &&
+	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)
 		tp->tg3_flags2 |= TG3_FLG2_JUMBO_CAPABLE;
 
 	if (pci_find_capability(tp->pdev, PCI_CAP_ID_EXP) != 0)
@@ -10455,6 +10540,12 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 		pci_cmd &= ~PCI_COMMAND_MEMORY;
 		pci_write_config_word(tp->pdev, PCI_COMMAND, pci_cmd);
 	}
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+		tp->read32_mbox = tg3_read32_mbox_5906;
+		tp->write32_mbox = tg3_write32_mbox_5906;
+		tp->write32_tx_mbox = tg3_write32_mbox_5906;
+		tp->write32_rx_mbox = tg3_write32_mbox_5906;
+	}
 
 	if (tp->write32 == tg3_write_indirect_reg32 ||
 	    ((tp->tg3_flags & TG3_FLAG_PCIX_MODE) &&
@@ -10526,6 +10617,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5705) &&
 	     (tp->pci_chip_rev_id != CHIPREV_ID_5705_A0) &&
 	     (tp->pci_chip_rev_id != CHIPREV_ID_5705_A1)) ||
+	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) ||
 	    (tp->tg3_flags2 & TG3_FLG2_ANY_SERDES))
 		tp->tg3_flags2 |= TG3_FLG2_NO_ETH_WIRE_SPEED;
 
@@ -10539,7 +10631,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787)
 			tp->tg3_flags2 |= TG3_FLG2_PHY_JITTER_BUG;
-		else
+		else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)
 			tp->tg3_flags2 |= TG3_FLG2_PHY_BER_BUG;
 	}
 
@@ -10629,7 +10721,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	      tp->pdev->device == PCI_DEVICE_ID_TIGON3_5705F)) ||
 	    (tp->pdev->vendor == PCI_VENDOR_ID_BROADCOM &&
 	     (tp->pdev->device == PCI_DEVICE_ID_TIGON3_5751F ||
-	      tp->pdev->device == PCI_DEVICE_ID_TIGON3_5753F)))
+	      tp->pdev->device == PCI_DEVICE_ID_TIGON3_5753F)) ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
 		tp->tg3_flags |= TG3_FLAG_10_100_ONLY;
 
 	err = tg3_phy_probe(tp);
@@ -10680,7 +10773,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	 * straddle the 4GB address boundary in some cases.
 	 */
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
 		tp->dev->hard_start_xmit = tg3_start_xmit;
 	else
 		tp->dev->hard_start_xmit = tg3_start_xmit_dma_bug;
@@ -10761,6 +10855,8 @@ static int __devinit tg3_get_device_address(struct tg3 *tp)
 		else
 			tg3_nvram_unlock(tp);
 	}
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
+		mac_offset = 0x10;
 
 	/* First try to get it from MAC address mailbox. */
 	tg3_read_mem(tp, NIC_SRAM_MAC_ADDR_HIGH_MBOX, &hi);
@@ -11244,6 +11340,12 @@ static void __devinit tg3_init_bufmgr_config(struct tg3 *tp)
 			DEFAULT_MB_MACRX_LOW_WATER_5705;
 		tp->bufmgr_config.mbuf_high_water =
 			DEFAULT_MB_HIGH_WATER_5705;
+		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
+			tp->bufmgr_config.mbuf_mac_rx_low_water =
+				DEFAULT_MB_MACRX_LOW_WATER_5906;
+			tp->bufmgr_config.mbuf_high_water =
+				DEFAULT_MB_HIGH_WATER_5906;
+		}
 
 		tp->bufmgr_config.mbuf_read_dma_low_water_jumbo =
 			DEFAULT_MB_RDMA_LOW_WATER_JUMBO_5780;
@@ -11288,6 +11390,7 @@ static char * __devinit tg3_phy_string(struct tg3 *tp)
 	case PHY_ID_BCM5755:	return "5755";
 	case PHY_ID_BCM5787:	return "5787";
 	case PHY_ID_BCM5756:	return "5722/5756";
+	case PHY_ID_BCM5906:	return "5906";
 	case PHY_ID_BCM8002:	return "8002/serdes";
 	case 0:			return "serdes";
 	default:		return "unknown";
@@ -11590,7 +11693,8 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 	 */
 	if (tp->tg3_flags2 & TG3_FLG2_HW_TSO) {
 		dev->features |= NETIF_F_TSO;
-		if (tp->tg3_flags2 & TG3_FLG2_HW_TSO_2)
+		if ((tp->tg3_flags2 & TG3_FLG2_HW_TSO_2) &&
+		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906))
 			dev->features |= NETIF_F_TSO6;
 	}
 
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index feed13dc8719..2f5e00c96016 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -24,6 +24,8 @@
 
 #define RX_COPY_THRESHOLD  		256
 
+#define TG3_RX_INTERNAL_RING_SZ_5906	32
+
 #define RX_STD_MAX_SIZE			1536
 #define RX_STD_MAX_SIZE_5705		512
 #define RX_JUMBO_MAX_SIZE		0xdeadbeef /* XXX */
@@ -129,6 +131,7 @@
 #define  CHIPREV_ID_5752_A0_HW		 0x5000
 #define  CHIPREV_ID_5752_A0		 0x6000
 #define  CHIPREV_ID_5752_A1		 0x6001
+#define  CHIPREV_ID_5906_A1		 0xc001
 #define  GET_ASIC_REV(CHIP_REV_ID)	((CHIP_REV_ID) >> 12)
 #define   ASIC_REV_5700			 0x07
 #define   ASIC_REV_5701			 0x00
@@ -141,6 +144,7 @@
 #define   ASIC_REV_5714			 0x09
 #define   ASIC_REV_5755			 0x0a
 #define   ASIC_REV_5787			 0x0b
+#define   ASIC_REV_5906			 0x0c
 #define  GET_CHIP_REV(CHIP_REV_ID)	((CHIP_REV_ID) >> 8)
 #define   CHIPREV_5700_AX		 0x70
 #define   CHIPREV_5700_BX		 0x71
@@ -646,7 +650,8 @@
 #define  SNDDATAI_SCTRL_FORCE_ZERO	 0x00000010
 #define SNDDATAI_STATSENAB		0x00000c0c
 #define SNDDATAI_STATSINCMASK		0x00000c10
-/* 0xc14 --> 0xc80 unused */
+#define ISO_PKT_TX			0x00000c20
+/* 0xc24 --> 0xc80 unused */
 #define SNDDATAI_COS_CNT_0		0x00000c80
 #define SNDDATAI_COS_CNT_1		0x00000c84
 #define SNDDATAI_COS_CNT_2		0x00000c88
@@ -997,11 +1002,13 @@
 #define BUFMGR_MB_MACRX_LOW_WATER	0x00004414
 #define  DEFAULT_MB_MACRX_LOW_WATER	  0x00000020
 #define  DEFAULT_MB_MACRX_LOW_WATER_5705  0x00000010
+#define  DEFAULT_MB_MACRX_LOW_WATER_5906  0x00000004
 #define  DEFAULT_MB_MACRX_LOW_WATER_JUMBO 0x00000098
 #define  DEFAULT_MB_MACRX_LOW_WATER_JUMBO_5780 0x0000004b
 #define BUFMGR_MB_HIGH_WATER		0x00004418
 #define  DEFAULT_MB_HIGH_WATER		 0x00000060
 #define  DEFAULT_MB_HIGH_WATER_5705	 0x00000060
+#define  DEFAULT_MB_HIGH_WATER_5906	 0x00000010
 #define  DEFAULT_MB_HIGH_WATER_JUMBO	 0x0000017c
 #define  DEFAULT_MB_HIGH_WATER_JUMBO_5780 0x00000096
 #define BUFMGR_RX_MB_ALLOC_REQ		0x0000441c
@@ -1138,7 +1145,12 @@
 #define TX_CPU_STATE			0x00005404
 #define TX_CPU_PGMCTR			0x0000541c
 
+#define VCPU_STATUS			0x00005100
+#define  VCPU_STATUS_INIT_DONE		 0x04000000
+#define  VCPU_STATUS_DRV_RESET		 0x08000000
+
 /* Mailboxes */
+#define GRCMBOX_BASE			0x00005600
 #define GRCMBOX_INTERRUPT_0		0x00005800 /* 64-bit */
 #define GRCMBOX_INTERRUPT_1		0x00005808 /* 64-bit */
 #define GRCMBOX_INTERRUPT_2		0x00005810 /* 64-bit */
@@ -1398,7 +1410,10 @@
 #define GRC_EEPROM_CTRL			0x00006840
 #define GRC_MDI_CTRL			0x00006844
 #define GRC_SEEPROM_DELAY		0x00006848
-/* 0x684c --> 0x6c00 unused */
+/* 0x684c --> 0x6890 unused */
+#define GRC_VCPU_EXT_CTRL		0x00006890
+#define GRC_VCPU_EXT_CTRL_HALT_CPU	 0x00400000
+#define GRC_VCPU_EXT_CTRL_DISABLE_WOL	 0x20000000
 #define GRC_FASTBOOT_PC			0x00006894	/* 5752, 5755, 5787 */
 
 /* 0x6c00 --> 0x7000 unused */
@@ -1485,7 +1500,11 @@
 #define NVRAM_WRITE1			0x00007028
 /* 0x702c --> 0x7400 unused */
 
-/* 0x7400 --> 0x8000 unused */
+/* 0x7400 --> 0x7c00 unused */
+#define PCIE_TRANSACTION_CFG		0x00007c04
+#define PCIE_TRANS_CFG_1SHOT_MSI	 0x20000000
+#define PCIE_TRANS_CFG_LOM		 0x00000020
+
 
 #define TG3_EEPROM_MAGIC		0x669955aa
 
@@ -2283,6 +2302,7 @@ struct tg3 {
 #define PHY_ID_BCM5755			0xbc050cc0
 #define PHY_ID_BCM5787			0xbc050ce0
 #define PHY_ID_BCM5756			0xbc050ed0
+#define PHY_ID_BCM5906			0xdc00ac40
 #define PHY_ID_BCM8002			0x60010140
 #define PHY_ID_INVALID			0xffffffff
 #define PHY_ID_REV_MASK			0x0000000f
@@ -2310,7 +2330,7 @@ struct tg3 {
 	 (X) == PHY_ID_BCM5752 || (X) == PHY_ID_BCM5714 || \
 	 (X) == PHY_ID_BCM5780 || (X) == PHY_ID_BCM5787 || \
 	 (X) == PHY_ID_BCM5755 || (X) == PHY_ID_BCM5756 || \
-	 (X) == PHY_ID_BCM8002)
+	 (X) == PHY_ID_BCM5906 || (X) == PHY_ID_BCM8002)
 
 	struct tg3_hw_stats		*hw_stats;
 	dma_addr_t			stats_mapping;
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index ea3140d226e6..b7e85ff045ea 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1944,6 +1944,8 @@
 #define PCI_DEVICE_ID_TIGON3_5901	0x170d
 #define PCI_DEVICE_ID_BCM4401B1		0x170c
 #define PCI_DEVICE_ID_TIGON3_5901_2	0x170e
+#define PCI_DEVICE_ID_TIGON3_5906	0x1712
+#define PCI_DEVICE_ID_TIGON3_5906M	0x1713
 #define PCI_DEVICE_ID_BCM4401		0x4401
 #define PCI_DEVICE_ID_BCM4401B0		0x4402
 
-- 
cgit v1.2.3


From adaf345b537681c6ed3657941904d976fe72f342 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:27:13 -0700
Subject: [IPV4]: annotate address in inet_request_sock

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 4 ++--
 net/ipv4/tcp_ipv4.c     | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index fc0b9e157b61..8130a375b867 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -62,8 +62,8 @@ struct inet_request_sock {
 	u16			inet6_rsk_offset;
 	/* 2 bytes hole, try to pack */
 #endif
-	u32			loc_addr;
-	u32			rmt_addr;
+	__be32			loc_addr;
+	__be32			rmt_addr;
 	u16			rmt_port;
 	u16			snd_wscale : 4, 
 				rcv_wscale : 4, 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 9018aa41cb55..89f3cf94685a 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -734,8 +734,8 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 	struct inet_request_sock *ireq;
 	struct tcp_options_received tmp_opt;
 	struct request_sock *req;
-	__u32 saddr = skb->nh.iph->saddr;
-	__u32 daddr = skb->nh.iph->daddr;
+	__be32 saddr = skb->nh.iph->saddr;
+	__be32 daddr = skb->nh.iph->daddr;
 	__u32 isn = TCP_SKB_CB(skb)->when;
 	struct dst_entry *dst = NULL;
 #ifdef CONFIG_SYN_COOKIES
-- 
cgit v1.2.3


From 7f25afbbefb266520a237df0e9b59112704a7a42 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:27:47 -0700
Subject: [IPV4]: inet_csk_search_req() (partial) annotations

raddr is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_connection_sock.h | 4 ++--
 net/ipv4/inet_connection_sock.c    | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index de4e83b6da4b..8122bcc83c1d 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -239,8 +239,8 @@ extern struct sock *inet_csk_accept(struct sock *sk, int flags, int *err);
 extern struct request_sock *inet_csk_search_req(const struct sock *sk,
 						struct request_sock ***prevp,
 						const __u16 rport,
-						const __u32 raddr,
-						const __u32 laddr);
+						const __be32 raddr,
+						const __be32 laddr);
 extern int inet_csk_bind_conflict(const struct sock *sk,
 				  const struct inet_bind_bucket *tb);
 extern int inet_csk_get_port(struct inet_hashinfo *hashinfo,
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 07204391d083..b9f4b7d5726b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -342,10 +342,10 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 
 EXPORT_SYMBOL_GPL(inet_csk_route_req);
 
-static inline u32 inet_synq_hash(const u32 raddr, const u16 rport,
+static inline u32 inet_synq_hash(const __be32 raddr, const u16 rport,
 				 const u32 rnd, const u16 synq_hsize)
 {
-	return jhash_2words(raddr, (u32)rport, rnd) & (synq_hsize - 1);
+	return jhash_2words((__force u32)raddr, (u32)rport, rnd) & (synq_hsize - 1);
 }
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
@@ -356,8 +356,8 @@ static inline u32 inet_synq_hash(const u32 raddr, const u16 rport,
 
 struct request_sock *inet_csk_search_req(const struct sock *sk,
 					 struct request_sock ***prevp,
-					 const __u16 rport, const __u32 raddr,
-					 const __u32 laddr)
+					 const __u16 rport, const __be32 raddr,
+					 const __be32 laddr)
 {
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	struct listen_sock *lopt = icsk->icsk_accept_queue.listen_opt;
-- 
cgit v1.2.3


From 3ca3c68e76686bee058937ade2b96f4de58ee434 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:28:07 -0700
Subject: [IPV4]: struct ip_options annotations

->faddr is net-endian; annotated as such, variables inferred to be net-endian
annotated.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 2 +-
 net/ipv4/af_inet.c      | 2 +-
 net/ipv4/icmp.c         | 2 +-
 net/ipv4/ip_options.c   | 2 +-
 net/ipv4/ip_output.c    | 4 ++--
 net/ipv4/raw.c          | 2 +-
 net/ipv4/udp.c          | 2 +-
 7 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 8130a375b867..3c19dbf30211 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -36,7 +36,7 @@
  * @ts_needaddr - Need to record addr of outgoing dev
  */
 struct ip_options {
-	__u32		faddr;
+	__be32		faddr;
 	unsigned char	optlen;
 	unsigned char	srr;
 	unsigned char	rr;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index ea5844832683..edcf0932ac6d 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1043,7 +1043,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct rtable *rt = (struct rtable *)__sk_dst_check(sk, 0);
-	u32 daddr;
+	__be32 daddr;
 	int err;
 
 	/* Route is OK, nothing to do. */
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 45396c53560d..2afa619d17eb 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -381,7 +381,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 	struct inet_sock *inet = inet_sk(sk);
 	struct ipcm_cookie ipc;
 	struct rtable *rt = (struct rtable *)skb->dst;
-	u32 daddr;
+	__be32 daddr;
 
 	if (ip_options_echo(&icmp_param->replyopts, skb))
 		return;
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 889e14180b04..3e6d3d4e50d4 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -165,7 +165,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb)
 	}
 	if (sopt->srr) {
 		unsigned char * start = sptr+sopt->srr;
-		u32 faddr;
+		__be32 faddr;
 
 		optlen  = start[1];
 		soffset = start[2];
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 5bf2f094e317..fc195a44fc2e 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -306,7 +306,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok)
 	/* Make sure we can route this packet. */
 	rt = (struct rtable *)__sk_dst_check(sk, 0);
 	if (rt == NULL) {
-		u32 daddr;
+		__be32 daddr;
 
 		/* Use correct destination address if we have options. */
 		daddr = inet->daddr;
@@ -1340,7 +1340,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 		char			data[40];
 	} replyopts;
 	struct ipcm_cookie ipc;
-	u32 daddr;
+	__be32 daddr;
 	struct rtable *rt = (struct rtable*)skb->dst;
 
 	if (ip_options_echo(&replyopts.opt, skb))
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 0e935b4c8741..49e5b4b55b93 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -381,7 +381,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	struct ipcm_cookie ipc;
 	struct rtable *rt = NULL;
 	int free = 0;
-	u32 daddr;
+	__be32 daddr;
 	u32 saddr;
 	u8  tos;
 	int err;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 77e265d7bb8f..b63fd3d39024 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -513,7 +513,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	struct rtable *rt = NULL;
 	int free = 0;
 	int connected = 0;
-	u32 daddr, faddr, saddr;
+	__be32 daddr, faddr, saddr;
 	u16 dport;
 	u8  tos;
 	int err;
-- 
cgit v1.2.3


From c1d18f9fa09489635a451ee13c1727e1683c2333 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:28:28 -0700
Subject: [IPV4]: struct ipcm_cookie annotation

->addr is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h | 2 +-
 net/ipv4/raw.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 6da1229c041a..b40bd2f9ed79 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -45,7 +45,7 @@ struct inet_skb_parm
 
 struct ipcm_cookie
 {
-	u32			addr;
+	__be32			addr;
 	int			oif;
 	struct ip_options	*opt;
 };
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 49e5b4b55b93..b430cf2a4f66 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -382,7 +382,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	struct rtable *rt = NULL;
 	int free = 0;
 	__be32 daddr;
-	u32 saddr;
+	__be32 saddr;
 	u8  tos;
 	int err;
 
-- 
cgit v1.2.3


From 4b06a7cf2f3c053e7fc47ca6a4c74553e2291e24 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:29:07 -0700
Subject: [IPV4]: ip_local_error() ipv4 address argument annotated

daddr is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h       | 2 +-
 net/ipv4/ip_sockglue.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index b40bd2f9ed79..96030314f70b 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -364,7 +364,7 @@ extern int	ip_ra_control(struct sock *sk, unsigned char on, void (*destructor)(s
 extern int 	ip_recv_error(struct sock *sk, struct msghdr *msg, int len);
 extern void	ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, 
 			      u16 port, u32 info, u8 *payload);
-extern void	ip_local_error(struct sock *sk, int err, u32 daddr, u16 dport,
+extern void	ip_local_error(struct sock *sk, int err, __be32 daddr, u16 dport,
 			       u32 info);
 
 /* sysctl helpers - any sysctl which holds a value that ends up being
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 2d05c4133d3e..d5ca43c0cd73 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -283,7 +283,7 @@ void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err,
 		kfree_skb(skb);
 }
 
-void ip_local_error(struct sock *sk, int err, u32 daddr, u16 port, u32 info)
+void ip_local_error(struct sock *sk, int err, __be32 daddr, u16 port, u32 info)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct sock_exterr_skb *serr;
-- 
cgit v1.2.3


From 00a5020cd51febbb3166ff7a09a2901c47ba251a Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:29:47 -0700
Subject: [IPV4]: annotate ipv4 address fields in struct ip_msfilter and struct
 ip_mreq_source

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/in.h b/include/linux/in.h
index bcaca8399aed..d79fc75fa7c2 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -123,17 +123,17 @@ struct ip_mreqn
 };
 
 struct ip_mreq_source {
-	__u32		imr_multiaddr;
-	__u32		imr_interface;
-	__u32		imr_sourceaddr;
+	__be32		imr_multiaddr;
+	__be32		imr_interface;
+	__be32		imr_sourceaddr;
 };
 
 struct ip_msfilter {
-	__u32		imsf_multiaddr;
-	__u32		imsf_interface;
+	__be32		imsf_multiaddr;
+	__be32		imsf_interface;
 	__u32		imsf_fmode;
 	__u32		imsf_numsrc;
-	__u32		imsf_slist[1];
+	__be32		imsf_slist[1];
 };
 
 #define IP_MSFILTER_SIZE(numsrc) \
-- 
cgit v1.2.3


From 8f935bbd7c6c66796c2403aefdab74bb48045bf6 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:30:07 -0700
Subject: [IPV4]: ip_mc_{inc,dec}_group() annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h |  4 ++--
 net/ipv4/igmp.c      | 24 ++++++++++++------------
 2 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 8e7eedb3a574..dd49ba9065ae 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -215,7 +215,7 @@ extern void ip_mc_init_dev(struct in_device *);
 extern void ip_mc_destroy_dev(struct in_device *);
 extern void ip_mc_up(struct in_device *);
 extern void ip_mc_down(struct in_device *);
-extern void ip_mc_dec_group(struct in_device *in_dev, u32 addr);
-extern void ip_mc_inc_group(struct in_device *in_dev, u32 addr);
+extern void ip_mc_dec_group(struct in_device *in_dev, __be32 addr);
+extern void ip_mc_inc_group(struct in_device *in_dev, __be32 addr);
 #endif
 #endif
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index d8068c483781..a2e733a82de2 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -144,8 +144,8 @@ static int sf_setstate(struct ip_mc_list *pmc);
 static void sf_markstate(struct ip_mc_list *pmc);
 #endif
 static void ip_mc_clear_src(struct ip_mc_list *pmc);
-static int ip_mc_add_src(struct in_device *in_dev, __u32 *pmca, int sfmode,
-			 int sfcount, __u32 *psfsrc, int delta);
+static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
+			 int sfcount, __be32 *psfsrc, int delta);
 
 static void ip_ma_put(struct ip_mc_list *im)
 {
@@ -1193,7 +1193,7 @@ static void igmp_group_added(struct ip_mc_list *im)
  *	A socket has joined a multicast group on device dev.
  */
 
-void ip_mc_inc_group(struct in_device *in_dev, u32 addr)
+void ip_mc_inc_group(struct in_device *in_dev, __be32 addr)
 {
 	struct ip_mc_list *im;
 
@@ -1252,7 +1252,7 @@ out:
  *	A socket has left a multicast group on device dev
  */
 
-void ip_mc_dec_group(struct in_device *in_dev, u32 addr)
+void ip_mc_dec_group(struct in_device *in_dev, __be32 addr)
 {
 	struct ip_mc_list *i, **ip;
 	
@@ -1402,7 +1402,7 @@ int sysctl_igmp_max_msf __read_mostly = IP_MAX_MSF;
 
 
 static int ip_mc_del1_src(struct ip_mc_list *pmc, int sfmode,
-	__u32 *psfsrc)
+	__be32 *psfsrc)
 {
 	struct ip_sf_list *psf, *psf_prev;
 	int rv = 0;
@@ -1450,8 +1450,8 @@ static int ip_mc_del1_src(struct ip_mc_list *pmc, int sfmode,
 #define igmp_ifc_event(x)	do { } while (0)
 #endif
 
-static int ip_mc_del_src(struct in_device *in_dev, __u32 *pmca, int sfmode,
-			 int sfcount, __u32 *psfsrc, int delta)
+static int ip_mc_del_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
+			 int sfcount, __be32 *psfsrc, int delta)
 {
 	struct ip_mc_list *pmc;
 	int	changerec = 0;
@@ -1517,7 +1517,7 @@ out_unlock:
  * Add multicast single-source filter to the interface list
  */
 static int ip_mc_add1_src(struct ip_mc_list *pmc, int sfmode,
-	__u32 *psfsrc, int delta)
+	__be32 *psfsrc, int delta)
 {
 	struct ip_sf_list *psf, *psf_prev;
 
@@ -1623,8 +1623,8 @@ static int sf_setstate(struct ip_mc_list *pmc)
 /*
  * Add multicast source filter list to the interface list
  */
-static int ip_mc_add_src(struct in_device *in_dev, __u32 *pmca, int sfmode,
-			 int sfcount, __u32 *psfsrc, int delta)
+static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
+			 int sfcount, __be32 *psfsrc, int delta)
 {
 	struct ip_mc_list *pmc;
 	int	isexclude;
@@ -1717,7 +1717,7 @@ static void ip_mc_clear_src(struct ip_mc_list *pmc)
 int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr)
 {
 	int err;
-	u32 addr = imr->imr_multiaddr.s_addr;
+	__be32 addr = imr->imr_multiaddr.s_addr;
 	struct ip_mc_socklist *iml=NULL, *i;
 	struct in_device *in_dev;
 	struct inet_sock *inet = inet_sk(sk);
@@ -1791,7 +1791,7 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
 	struct inet_sock *inet = inet_sk(sk);
 	struct ip_mc_socklist *iml, **imlp;
 	struct in_device *in_dev;
-	u32 group = imr->imr_multiaddr.s_addr;
+	__be32 group = imr->imr_multiaddr.s_addr;
 	u32 ifindex;
 	int ret = -EADDRNOTAVAIL;
 
-- 
cgit v1.2.3


From 942bf921e922560c05fde6afb00ddedf6224c608 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:30:28 -0700
Subject: [IPV4]: IGMP on-the-wire data is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index dd49ba9065ae..fd207d9b2733 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -30,8 +30,8 @@ struct igmphdr
 {
 	__u8 type;
 	__u8 code;		/* For newer IGMP */
-	__u16 csum;
-	__u32 group;
+	__be16 csum;
+	__be32 group;
 };
 
 /* V3 group record types [grec_type] */
@@ -45,25 +45,25 @@ struct igmphdr
 struct igmpv3_grec {
 	__u8	grec_type;
 	__u8	grec_auxwords;
-	__u16	grec_nsrcs;
-	__u32	grec_mca;
-	__u32	grec_src[0];
+	__be16	grec_nsrcs;
+	__be32	grec_mca;
+	__be32	grec_src[0];
 };
 
 struct igmpv3_report {
 	__u8 type;
 	__u8 resv1;
-	__u16 csum;
-	__u16 resv2;
-	__u16 ngrec;
+	__be16 csum;
+	__be16 resv2;
+	__be16 ngrec;
 	struct igmpv3_grec grec[0];
 };
 
 struct igmpv3_query {
 	__u8 type;
 	__u8 code;
-	__u16 csum;
-	__u32 group;
+	__be16 csum;
+	__be32 group;
 #if defined(__LITTLE_ENDIAN_BITFIELD)
 	__u8 qrv:3,
 	     suppress:1,
@@ -76,8 +76,8 @@ struct igmpv3_query {
 #error "Please fix <asm/byteorder.h>"
 #endif
 	__u8 qqic;
-	__u16 nsrcs;
-	__u32 srcs[0];
+	__be16 nsrcs;
+	__be32 srcs[0];
 };
 
 #define IGMP_HOST_MEMBERSHIP_QUERY	0x11	/* From RFC1112 */
-- 
cgit v1.2.3


From ea4d9e7220d32348cc9742ba6d27de5165262664 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:30:52 -0700
Subject: [IPV4]: struct ip_sf_list and struct ip_sf_socklist annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h | 4 ++--
 net/ipv4/igmp.c      | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index fd207d9b2733..4e9f3fe77cf9 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -136,7 +136,7 @@ struct ip_sf_socklist
 {
 	unsigned int		sl_max;
 	unsigned int		sl_count;
-	__u32			sl_addr[0];
+	__be32			sl_addr[0];
 };
 
 #define IP_SFLSIZE(count)	(sizeof(struct ip_sf_socklist) + \
@@ -159,7 +159,7 @@ struct ip_mc_socklist
 struct ip_sf_list
 {
 	struct ip_sf_list	*sf_next;
-	__u32			sf_inaddr;
+	__be32			sf_inaddr;
 	unsigned long		sf_count[2];	/* include/exclude counts */
 	unsigned char		sf_gsresp;	/* include in g & s response? */
 	unsigned char		sf_oldin;	/* change state */
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index a2e733a82de2..f6ca51a2dcc2 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -426,7 +426,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc,
 	first = 1;
 	psf_prev = NULL;
 	for (psf=*psf_list; psf; psf=psf_next) {
-		u32 *psrc;
+		__be32 *psrc;
 
 		psf_next = psf->sf_next;
 
@@ -455,7 +455,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc,
 			skb = add_grhead(skb, pmc, type, &pgr);
 			first = 0;
 		}
-		psrc = (u32 *)skb_put(skb, sizeof(u32));
+		psrc = (__be32 *)skb_put(skb, sizeof(u32));
 		*psrc = psf->sf_inaddr;
 		scount++; stotal++;
 		if ((type == IGMPV3_ALLOW_NEW_SOURCES ||
@@ -748,7 +748,7 @@ static void igmp_timer_expire(unsigned long data)
 }
 
 /* mark EXCLUDE-mode sources */
-static int igmp_xmarksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs)
+static int igmp_xmarksources(struct ip_mc_list *pmc, int nsrcs, __be32 *srcs)
 {
 	struct ip_sf_list *psf;
 	int i, scount;
@@ -775,7 +775,7 @@ static int igmp_xmarksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs)
 	return 1;
 }
 
-static int igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __u32 *srcs)
+static int igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __be32 *srcs)
 {
 	struct ip_sf_list *psf;
 	int i, scount;
-- 
cgit v1.2.3


From c0cda068aac3481d40795b115e4fd36f7d386e3a Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:31:10 -0700
Subject: [IPV4]: ip_mc_sf_allow() annotated

ip_mc_sf_allow() expects addresses to be passed net-endian.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h | 2 +-
 net/ipv4/igmp.c      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 4e9f3fe77cf9..7514cceb4fe3 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -209,7 +209,7 @@ extern int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf,
 		struct ip_msfilter __user *optval, int __user *optlen);
 extern int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 		struct group_filter __user *optval, int __user *optlen);
-extern int ip_mc_sf_allow(struct sock *sk, u32 local, u32 rmt, int dif);
+extern int ip_mc_sf_allow(struct sock *sk, __be32 local, __be32 rmt, int dif);
 extern void ip_mr_init(void);
 extern void ip_mc_init_dev(struct in_device *);
 extern void ip_mc_destroy_dev(struct in_device *);
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index f6ca51a2dcc2..a2c7b2b1b3fb 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -2156,7 +2156,7 @@ done:
 /*
  * check if a multicast source filter allows delivery for a given <src,dst,intf>
  */
-int ip_mc_sf_allow(struct sock *sk, u32 loc_addr, u32 rmt_addr, int dif)
+int ip_mc_sf_allow(struct sock *sk, __be32 loc_addr, __be32 rmt_addr, int dif)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct ip_mc_socklist *pmc;
-- 
cgit v1.2.3


From 63007727e0bb09e8d906f73d36a09b9fac0d5893 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:31:32 -0700
Subject: [IPV4]: trivial igmp annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/igmp.h |  2 +-
 net/ipv4/igmp.c      | 34 +++++++++++++++++-----------------
 2 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 7514cceb4fe3..03f43e2893a4 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -140,7 +140,7 @@ struct ip_sf_socklist
 };
 
 #define IP_SFLSIZE(count)	(sizeof(struct ip_sf_socklist) + \
-	(count) * sizeof(__u32))
+	(count) * sizeof(__be32))
 
 #define IP_SFBLOCK	10	/* allocate this many at once */
 
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index a2c7b2b1b3fb..6eee71647b7c 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -138,7 +138,7 @@
 		time_before(jiffies, (in_dev)->mr_v2_seen)))
 
 static void igmpv3_add_delrec(struct in_device *in_dev, struct ip_mc_list *im);
-static void igmpv3_del_delrec(struct in_device *in_dev, __u32 multiaddr);
+static void igmpv3_del_delrec(struct in_device *in_dev, __be32 multiaddr);
 static void igmpv3_clear_delrec(struct in_device *in_dev);
 static int sf_setstate(struct ip_mc_list *pmc);
 static void sf_markstate(struct ip_mc_list *pmc);
@@ -439,7 +439,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc,
 		if (isquery)
 			psf->sf_gsresp = 0;
 
-		if (AVAILABLE(skb) < sizeof(u32) +
+		if (AVAILABLE(skb) < sizeof(__be32) +
 		    first*sizeof(struct igmpv3_grec)) {
 			if (truncate && !first)
 				break;	 /* truncate these */
@@ -455,7 +455,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc,
 			skb = add_grhead(skb, pmc, type, &pgr);
 			first = 0;
 		}
-		psrc = (__be32 *)skb_put(skb, sizeof(u32));
+		psrc = (__be32 *)skb_put(skb, sizeof(__be32));
 		*psrc = psf->sf_inaddr;
 		scount++; stotal++;
 		if ((type == IGMPV3_ALLOW_NEW_SOURCES ||
@@ -630,8 +630,8 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
 	struct igmphdr *ih;
 	struct rtable *rt;
 	struct net_device *dev = in_dev->dev;
-	u32	group = pmc ? pmc->multiaddr : 0;
-	u32	dst;
+	__be32	group = pmc ? pmc->multiaddr : 0;
+	__be32	dst;
 
 	if (type == IGMPV3_HOST_MEMBERSHIP_REPORT)
 		return igmpv3_send_report(in_dev, pmc);
@@ -803,7 +803,7 @@ static int igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __be32 *srcs)
 	return 1;
 }
 
-static void igmp_heard_report(struct in_device *in_dev, u32 group)
+static void igmp_heard_report(struct in_device *in_dev, __be32 group)
 {
 	struct ip_mc_list *im;
 
@@ -828,7 +828,7 @@ static void igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb,
 	struct igmphdr 		*ih = skb->h.igmph;
 	struct igmpv3_query *ih3 = (struct igmpv3_query *)ih;
 	struct ip_mc_list	*im;
-	u32			group = ih->group;
+	__be32			group = ih->group;
 	int			max_delay;
 	int			mark = 0;
 
@@ -862,7 +862,7 @@ static void igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb,
 		ih3 = (struct igmpv3_query *) skb->h.raw;
 		if (ih3->nsrcs) {
 			if (!pskb_may_pull(skb, sizeof(struct igmpv3_query) 
-					   + ntohs(ih3->nsrcs)*sizeof(__u32)))
+					   + ntohs(ih3->nsrcs)*sizeof(__be32)))
 				return;
 			ih3 = (struct igmpv3_query *) skb->h.raw;
 		}
@@ -985,7 +985,7 @@ drop:
  *	Add a filter to a device
  */
 
-static void ip_mc_filter_add(struct in_device *in_dev, u32 addr)
+static void ip_mc_filter_add(struct in_device *in_dev, __be32 addr)
 {
 	char buf[MAX_ADDR_LEN];
 	struct net_device *dev = in_dev->dev;
@@ -1005,7 +1005,7 @@ static void ip_mc_filter_add(struct in_device *in_dev, u32 addr)
  *	Remove a filter from a device
  */
 
-static void ip_mc_filter_del(struct in_device *in_dev, u32 addr)
+static void ip_mc_filter_del(struct in_device *in_dev, __be32 addr)
 {
 	char buf[MAX_ADDR_LEN];
 	struct net_device *dev = in_dev->dev;
@@ -1055,7 +1055,7 @@ static void igmpv3_add_delrec(struct in_device *in_dev, struct ip_mc_list *im)
 	spin_unlock_bh(&in_dev->mc_tomb_lock);
 }
 
-static void igmpv3_del_delrec(struct in_device *in_dev, __u32 multiaddr)
+static void igmpv3_del_delrec(struct in_device *in_dev, __be32 multiaddr)
 {
 	struct ip_mc_list *pmc, *pmc_prev;
 	struct ip_sf_list *psf, *psf_next;
@@ -1829,7 +1829,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct
 {
 	int err;
 	struct ip_mreqn imr;
-	u32 addr = mreqs->imr_multiaddr;
+	__be32 addr = mreqs->imr_multiaddr;
 	struct ip_mc_socklist *pmc;
 	struct in_device *in_dev = NULL;
 	struct inet_sock *inet = inet_sk(sk);
@@ -1883,7 +1883,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct
 		rv = !0;
 		for (i=0; i<psl->sl_count; i++) {
 			rv = memcmp(&psl->sl_addr[i], &mreqs->imr_sourceaddr,
-				sizeof(__u32));
+				sizeof(__be32));
 			if (rv == 0)
 				break;
 		}
@@ -1935,7 +1935,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct
 	rv = 1;	/* > 0 for insert logic below if sl_count is 0 */
 	for (i=0; i<psl->sl_count; i++) {
 		rv = memcmp(&psl->sl_addr[i], &mreqs->imr_sourceaddr,
-			sizeof(__u32));
+			sizeof(__be32));
 		if (rv == 0)
 			break;
 	}
@@ -1960,7 +1960,7 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex)
 {
 	int err = 0;
 	struct ip_mreqn	imr;
-	u32 addr = msf->imsf_multiaddr;
+	__be32 addr = msf->imsf_multiaddr;
 	struct ip_mc_socklist *pmc;
 	struct in_device *in_dev;
 	struct inet_sock *inet = inet_sk(sk);
@@ -2044,7 +2044,7 @@ int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf,
 {
 	int err, len, count, copycount;
 	struct ip_mreqn	imr;
-	u32 addr = msf->imsf_multiaddr;
+	__be32 addr = msf->imsf_multiaddr;
 	struct ip_mc_socklist *pmc;
 	struct in_device *in_dev;
 	struct inet_sock *inet = inet_sk(sk);
@@ -2103,7 +2103,7 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 {
 	int err, i, count, copycount;
 	struct sockaddr_in *psin;
-	u32 addr;
+	__be32 addr;
 	struct ip_mc_socklist *pmc;
 	struct inet_sock *inet = inet_sk(sk);
 	struct ip_sf_socklist *psl;
-- 
cgit v1.2.3


From 46a97324a5ebdc1e343a0223d993e79551adab0f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:31:51 -0700
Subject: [IPV4]: TCP headers annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 8ebf497907f8..543f06371840 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -21,10 +21,10 @@
 #include <asm/byteorder.h>
 
 struct tcphdr {
-	__u16	source;
-	__u16	dest;
-	__u32	seq;
-	__u32	ack_seq;
+	__be16	source;
+	__be16	dest;
+	__be32	seq;
+	__be32	ack_seq;
 #if defined(__LITTLE_ENDIAN_BITFIELD)
 	__u16	res1:4,
 		doff:4,
@@ -50,9 +50,9 @@ struct tcphdr {
 #else
 #error	"Adjust your <asm/byteorder.h> defines"
 #endif	
-	__u16	window;
-	__u16	check;
-	__u16	urg_ptr;
+	__be16	window;
+	__be16	check;
+	__be16	urg_ptr;
 };
 
 /*
@@ -62,7 +62,7 @@ struct tcphdr {
  */
 union tcp_word_hdr { 
 	struct tcphdr hdr;
-	__u32 		  words[5];
+	__be32 		  words[5];
 }; 
 
 #define tcp_flag_word(tp) ( ((union tcp_word_hdr *)(tp))->words [3]) 
-- 
cgit v1.2.3


From 269bd27e66037a7932cee6d6aa7ef7defd0bfe38 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:32:28 -0700
Subject: [TCP]: struct tcp_sack_block annotations

Some of the instances of tcp_sack_block are host-endian, some - net-endian.
Define struct tcp_sack_block_wire identical to struct tcp_sack_block
with u32 replaced with __be32; annotate uses of tcp_sack_block replacing
net-endian ones with tcp_sack_block_wire.  Change is obviously safe since
for cc(1) __be32 is typedefed to u32.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h                | 5 +++++
 net/ipv4/netfilter/ip_nat_helper.c | 2 +-
 net/ipv4/tcp_input.c               | 2 +-
 3 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 543f06371840..9632aa866de4 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -166,6 +166,11 @@ struct tcp_info
 #include <net/inet_timewait_sock.h>
 
 /* This defines a selective acknowledgement block. */
+struct tcp_sack_block_wire {
+	__be32	start_seq;
+	__be32	end_seq;
+};
+
 struct tcp_sack_block {
 	__u32	start_seq;
 	__u32	end_seq;
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index 7f6a75984f6c..e9c5187ea5b2 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -283,7 +283,7 @@ sack_adjust(struct sk_buff *skb,
 	    struct ip_nat_seq *natseq)
 {
 	while (sackoff < sackend) {
-		struct tcp_sack_block *sack;
+		struct tcp_sack_block_wire *sack;
 		u_int32_t new_start_seq, new_end_seq;
 
 		sack = (void *)skb->data + sackoff;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index b3def0df14fb..46984ffd73cd 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -935,7 +935,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
 	unsigned char *ptr = ack_skb->h.raw + TCP_SKB_CB(ack_skb)->sacked;
-	struct tcp_sack_block *sp = (struct tcp_sack_block *)(ptr+2);
+	struct tcp_sack_block_wire *sp = (struct tcp_sack_block_wire *)(ptr+2);
 	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
 	int reord = tp->packets_out;
 	int prior_fackets;
-- 
cgit v1.2.3


From dddc93c05d7dba60b44866486502c155e96ab915 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:32:46 -0700
Subject: [TCP]: struct tcp_sock .pred_flags is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 9632aa866de4..0e058a2d1c6d 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -216,7 +216,7 @@ struct tcp_sock {
  *	Header prediction flags
  *	0x5?10 << 16 + snd_wnd in net byte order
  */
-	__u32	pred_flags;
+	__be32	pred_flags;
 
 /*
  *	RFC793 variables by their proper names. This means you can
-- 
cgit v1.2.3


From 2816e1284a2db03ad5e205bab4eacbc5f7d4f991 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:33:05 -0700
Subject: [IPV4]: ports in struct inet_sock are net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 3c19dbf30211..7a1f69f01269 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -112,13 +112,13 @@ struct inet_sock {
 	/* Socket demultiplex comparisons on incoming packets. */
 	__be32			daddr;
 	__be32			rcv_saddr;
-	__u16			dport;
+	__be16			dport;
 	__u16			num;
 	__be32			saddr;
 	__s16			uc_ttl;
 	__u16			cmsg_flags;
 	struct ip_options	*opt;
-	__u16			sport;
+	__be16			sport;
 	__u16			id;
 	__u8			tos;
 	__u8			mc_ttl;
-- 
cgit v1.2.3


From cc939d37349bf82891bd1f4558284d7fafe7acb2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:33:22 -0700
Subject: [NET]: ip ports in struct flowi are net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/flow.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/flow.h b/include/net/flow.h
index 97569af9bda7..5700b0ad63c0 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -56,8 +56,8 @@ struct flowi {
 #define FLOWI_FLAG_MULTIPATHOLDROUTE 0x01
 	union {
 		struct {
-			__u16	sport;
-			__u16	dport;
+			__be16	sport;
+			__be16	dport;
 		} ports;
 
 		struct {
-- 
cgit v1.2.3


From 0579016ec4691116f6322ec6ed7fb7ce746948e9 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:33:40 -0700
Subject: [IPV4]: ip_local_error() annotations

port argument is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h       | 2 +-
 net/ipv4/ip_sockglue.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 96030314f70b..7e7e9731ada8 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -364,7 +364,7 @@ extern int	ip_ra_control(struct sock *sk, unsigned char on, void (*destructor)(s
 extern int 	ip_recv_error(struct sock *sk, struct msghdr *msg, int len);
 extern void	ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, 
 			      u16 port, u32 info, u8 *payload);
-extern void	ip_local_error(struct sock *sk, int err, __be32 daddr, u16 dport,
+extern void	ip_local_error(struct sock *sk, int err, __be32 daddr, __be16 dport,
 			       u32 info);
 
 /* sysctl helpers - any sysctl which holds a value that ends up being
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index d5ca43c0cd73..0660494196e6 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -283,7 +283,7 @@ void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err,
 		kfree_skb(skb);
 }
 
-void ip_local_error(struct sock *sk, int err, __be32 daddr, u16 port, u32 info)
+void ip_local_error(struct sock *sk, int err, __be32 daddr, __be16 port, u32 info)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct sock_exterr_skb *serr;
-- 
cgit v1.2.3


From b406313c733156c8eea7d9c1891476f400914367 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:34:02 -0700
Subject: [NET]: struct sock_exterr_skb annotations

->port is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/errqueue.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/errqueue.h b/include/linux/errqueue.h
index 408118a07763..92f8d4fab32b 100644
--- a/include/linux/errqueue.h
+++ b/include/linux/errqueue.h
@@ -38,7 +38,7 @@ struct sock_exterr_skb
 	} header;
 	struct sock_extended_err	ee;
 	u16				addr_offset;
-	u16				port;
+	__be16				port;
 };
 
 #endif
-- 
cgit v1.2.3


From 35986b329f5476630cef00cc7a164ff336ec1a21 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:34:21 -0700
Subject: [IPV4]: ip_icmp_error() annotations

port is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h       | 2 +-
 net/ipv4/ip_sockglue.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/ip.h b/include/net/ip.h
index 7e7e9731ada8..b6d95e553401 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -363,7 +363,7 @@ extern int	ip_ra_control(struct sock *sk, unsigned char on, void (*destructor)(s
 
 extern int 	ip_recv_error(struct sock *sk, struct msghdr *msg, int len);
 extern void	ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, 
-			      u16 port, u32 info, u8 *payload);
+			      __be16 port, u32 info, u8 *payload);
 extern void	ip_local_error(struct sock *sk, int err, __be32 daddr, __be16 dport,
 			       u32 info);
 
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 0660494196e6..4b132953bcc2 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -254,7 +254,7 @@ int ip_ra_control(struct sock *sk, unsigned char on, void (*destructor)(struct s
 }
 
 void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, 
-		   u16 port, u32 info, u8 *payload)
+		   __be16 port, u32 info, u8 *payload)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct sock_exterr_skb *serr;
-- 
cgit v1.2.3


From 39dccd9d922b595301e5d43ca7a30823d81393b6 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:34:41 -0700
Subject: [IPV4]: route.h annotations

ip_route_connect(), ip_route_newports() get port numbers net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/route.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/net/route.h b/include/net/route.h
index 58752722c968..486e37aff06c 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -146,7 +146,7 @@ static inline char rt_tos2priority(u8 tos)
 
 static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 				   __be32 src, u32 tos, int oif, u8 protocol,
-				   u16 sport, u16 dport, struct sock *sk)
+				   __be16 sport, __be16 dport, struct sock *sk)
 {
 	struct flowi fl = { .oif = oif,
 			    .nl_u = { .ip4_u = { .daddr = dst,
@@ -172,7 +172,7 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
 }
 
 static inline int ip_route_newports(struct rtable **rp, u8 protocol,
-				    u16 sport, u16 dport, struct sock *sk)
+				    __be16 sport, __be16 dport, struct sock *sk)
 {
 	if (sport != (*rp)->fl.fl_ip_sport ||
 	    dport != (*rp)->fl.fl_ip_dport) {
-- 
cgit v1.2.3


From e11be94bf6a3bfc90816d37847e6b5b179ca2ff6 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:35:29 -0700
Subject: [IPV4]: struct inet_request_sock annotations

->port is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_sock.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 7a1f69f01269..a448bb226bb3 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -64,7 +64,7 @@ struct inet_request_sock {
 #endif
 	__be32			loc_addr;
 	__be32			rmt_addr;
-	u16			rmt_port;
+	__be16			rmt_port;
 	u16			snd_wscale : 4, 
 				rcv_wscale : 4, 
 				tstamp_ok  : 1,
-- 
cgit v1.2.3


From bd6d610a14f2ed896b76dfb61fbdec829e44b8d3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:35:47 -0700
Subject: [IPV4]: ARP header annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_arp.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h
index a8b1a2071838..7f5714214ee3 100644
--- a/include/linux/if_arp.h
+++ b/include/linux/if_arp.h
@@ -130,11 +130,11 @@ struct arpreq_old {
 
 struct arphdr
 {
-	unsigned short	ar_hrd;		/* format of hardware address	*/
-	unsigned short	ar_pro;		/* format of protocol address	*/
+	__be16		ar_hrd;		/* format of hardware address	*/
+	__be16		ar_pro;		/* format of protocol address	*/
 	unsigned char	ar_hln;		/* length of hardware address	*/
 	unsigned char	ar_pln;		/* length of protocol address	*/
-	unsigned short	ar_op;		/* ARP opcode (command)		*/
+	__be16		ar_op;		/* ARP opcode (command)		*/
 
 #if 0
 	 /*
-- 
cgit v1.2.3


From ed9bad06eec5ee7842851f9abeb406e9a73084e8 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:36:36 -0700
Subject: [IPV4] net/ipv4/arp.c: trivial annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/arp.h                   |  8 ++++----
 net/bridge/netfilter/ebt_arpreply.c |  2 +-
 net/ipv4/arp.c                      | 30 +++++++++++++++---------------
 3 files changed, 20 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/include/net/arp.h b/include/net/arp.h
index 643bded9f557..6a3d9a7d302b 100644
--- a/include/net/arp.h
+++ b/include/net/arp.h
@@ -12,15 +12,15 @@ extern struct neigh_table arp_tbl;
 extern void	arp_init(void);
 extern int	arp_find(unsigned char *haddr, struct sk_buff *skb);
 extern int	arp_ioctl(unsigned int cmd, void __user *arg);
-extern void     arp_send(int type, int ptype, u32 dest_ip, 
-			 struct net_device *dev, u32 src_ip, 
+extern void     arp_send(int type, int ptype, __be32 dest_ip,
+			 struct net_device *dev, __be32 src_ip,
 			 unsigned char *dest_hw, unsigned char *src_hw, unsigned char *th);
 extern int	arp_bind_neighbour(struct dst_entry *dst);
 extern int	arp_mc_map(u32 addr, u8 *haddr, struct net_device *dev, int dir);
 extern void	arp_ifdown(struct net_device *dev);
 
-extern struct sk_buff *arp_create(int type, int ptype, u32 dest_ip,
-				  struct net_device *dev, u32 src_ip,
+extern struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip,
+				  struct net_device *dev, __be32 src_ip,
 				  unsigned char *dest_hw, unsigned char *src_hw,
 				  unsigned char *target_hw);
 extern void arp_xmit(struct sk_buff *skb);
diff --git a/net/bridge/netfilter/ebt_arpreply.c b/net/bridge/netfilter/ebt_arpreply.c
index d19fc4b328dc..0aa7b9910a86 100644
--- a/net/bridge/netfilter/ebt_arpreply.c
+++ b/net/bridge/netfilter/ebt_arpreply.c
@@ -20,7 +20,7 @@ static int ebt_target_reply(struct sk_buff **pskb, unsigned int hooknr,
    const void *data, unsigned int datalen)
 {
 	struct ebt_arpreply_info *info = (struct ebt_arpreply_info *)data;
-	u32 _sip, *siptr, _dip, *diptr;
+	__be32 _sip, *siptr, _dip, *diptr;
 	struct arphdr _ah, *ap;
 	unsigned char _sha[ETH_ALEN], *shp;
 	struct sk_buff *skb = *pskb;
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 2a29a5ebc978..cfe5c8474286 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -385,7 +385,7 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
 }
 
 static int arp_ignore(struct in_device *in_dev, struct net_device *dev,
-		      u32 sip, u32 tip)
+		      __be32 sip, __be32 tip)
 {
 	int scope;
 
@@ -420,7 +420,7 @@ static int arp_ignore(struct in_device *in_dev, struct net_device *dev,
 	return !inet_confirm_addr(dev, sip, tip, scope);
 }
 
-static int arp_filter(__u32 sip, __u32 tip, struct net_device *dev)
+static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev)
 {
 	struct flowi fl = { .nl_u = { .ip4_u = { .daddr = sip,
 						 .saddr = tip } } };
@@ -449,7 +449,7 @@ static int arp_filter(__u32 sip, __u32 tip, struct net_device *dev)
  *	is allowed to use this function, it is scheduled to be removed. --ANK
  */
 
-static int arp_set_predefined(int addr_hint, unsigned char * haddr, u32 paddr, struct net_device * dev)
+static int arp_set_predefined(int addr_hint, unsigned char * haddr, __be32 paddr, struct net_device * dev)
 {
 	switch (addr_hint) {
 	case RTN_LOCAL:
@@ -511,7 +511,7 @@ int arp_bind_neighbour(struct dst_entry *dst)
 	if (dev == NULL)
 		return -EINVAL;
 	if (n == NULL) {
-		u32 nexthop = ((struct rtable*)dst)->rt_gateway;
+		__be32 nexthop = ((struct rtable*)dst)->rt_gateway;
 		if (dev->flags&(IFF_LOOPBACK|IFF_POINTOPOINT))
 			nexthop = 0;
 		n = __neigh_lookup_errno(
@@ -560,8 +560,8 @@ static inline int arp_fwd_proxy(struct in_device *in_dev, struct rtable *rt)
  *	Create an arp packet. If (dest_hw == NULL), we create a broadcast
  *	message.
  */
-struct sk_buff *arp_create(int type, int ptype, u32 dest_ip,
-			   struct net_device *dev, u32 src_ip,
+struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip,
+			   struct net_device *dev, __be32 src_ip,
 			   unsigned char *dest_hw, unsigned char *src_hw,
 			   unsigned char *target_hw)
 {
@@ -675,8 +675,8 @@ void arp_xmit(struct sk_buff *skb)
 /*
  *	Create and send an arp packet.
  */
-void arp_send(int type, int ptype, u32 dest_ip, 
-	      struct net_device *dev, u32 src_ip, 
+void arp_send(int type, int ptype, __be32 dest_ip,
+	      struct net_device *dev, __be32 src_ip,
 	      unsigned char *dest_hw, unsigned char *src_hw,
 	      unsigned char *target_hw)
 {
@@ -969,13 +969,13 @@ out_of_mem:
 
 static int arp_req_set(struct arpreq *r, struct net_device * dev)
 {
-	u32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr;
+	__be32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr;
 	struct neighbour *neigh;
 	int err;
 
 	if (r->arp_flags&ATF_PUBL) {
-		u32 mask = ((struct sockaddr_in *) &r->arp_netmask)->sin_addr.s_addr;
-		if (mask && mask != 0xFFFFFFFF)
+		__be32 mask = ((struct sockaddr_in *) &r->arp_netmask)->sin_addr.s_addr;
+		if (mask && mask != htonl(0xFFFFFFFF))
 			return -EINVAL;
 		if (!dev && (r->arp_flags & ATF_COM)) {
 			dev = dev_getbyhwaddr(r->arp_ha.sa_family, r->arp_ha.sa_data);
@@ -1063,7 +1063,7 @@ static unsigned arp_state_to_flags(struct neighbour *neigh)
 
 static int arp_req_get(struct arpreq *r, struct net_device *dev)
 {
-	u32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr;
+	__be32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr;
 	struct neighbour *neigh;
 	int err = -ENXIO;
 
@@ -1084,13 +1084,13 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev)
 static int arp_req_delete(struct arpreq *r, struct net_device * dev)
 {
 	int err;
-	u32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr;
+	__be32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr;
 	struct neighbour *neigh;
 
 	if (r->arp_flags & ATF_PUBL) {
-		u32 mask =
+		__be32 mask =
 		       ((struct sockaddr_in *)&r->arp_netmask)->sin_addr.s_addr;
-		if (mask == 0xFFFFFFFF)
+		if (mask == htonl(0xFFFFFFFF))
 			return pneigh_delete(&arp_tbl, &ip, dev);
 		if (mask == 0) {
 			if (dev == NULL) {
-- 
cgit v1.2.3


From 6b72977bd6c6fefc6497d4f0275079f539eaf0ac Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:36:59 -0700
Subject: [IPV4]: inet_csk_search_req() annotations

rport argument is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_connection_sock.h | 2 +-
 net/ipv4/inet_connection_sock.c    | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 8122bcc83c1d..0bcf9f237e1f 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -238,7 +238,7 @@ extern struct sock *inet_csk_accept(struct sock *sk, int flags, int *err);
 
 extern struct request_sock *inet_csk_search_req(const struct sock *sk,
 						struct request_sock ***prevp,
-						const __u16 rport,
+						const __be16 rport,
 						const __be32 raddr,
 						const __be32 laddr);
 extern int inet_csk_bind_conflict(const struct sock *sk,
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index b9f4b7d5726b..274b0b846c25 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -342,10 +342,10 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 
 EXPORT_SYMBOL_GPL(inet_csk_route_req);
 
-static inline u32 inet_synq_hash(const __be32 raddr, const u16 rport,
+static inline u32 inet_synq_hash(const __be32 raddr, const __be16 rport,
 				 const u32 rnd, const u16 synq_hsize)
 {
-	return jhash_2words((__force u32)raddr, (u32)rport, rnd) & (synq_hsize - 1);
+	return jhash_2words((__force u32)raddr, (__force u32)rport, rnd) & (synq_hsize - 1);
 }
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
@@ -356,7 +356,7 @@ static inline u32 inet_synq_hash(const __be32 raddr, const u16 rport,
 
 struct request_sock *inet_csk_search_req(const struct sock *sk,
 					 struct request_sock ***prevp,
-					 const __u16 rport, const __be32 raddr,
+					 const __be16 rport, const __be32 raddr,
 					 const __be32 laddr)
 {
 	const struct inet_connection_sock *icsk = inet_csk(sk);
-- 
cgit v1.2.3


From 4e7e0c7592cafe5453e5b2f115fc0065d11b3d44 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:37:19 -0700
Subject: [IPV4]: UDP header annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 90223f057d50..014b41d1e308 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -20,10 +20,10 @@
 #include <linux/types.h>
 
 struct udphdr {
-	__u16	source;
-	__u16	dest;
-	__u16	len;
-	__u16	check;
+	__be16	source;
+	__be16	dest;
+	__be16	len;
+	__be16	check;
 };
 
 /* UDP socket options */
-- 
cgit v1.2.3


From b1dd39ac963040c2d282ab8026b9c9aa9306ea06 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:38:13 -0700
Subject: [IPV4]: ICMP header annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/icmp.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/icmp.h b/include/linux/icmp.h
index f0b571f1060b..878cfe4e587f 100644
--- a/include/linux/icmp.h
+++ b/include/linux/icmp.h
@@ -68,16 +68,16 @@
 struct icmphdr {
   __u8		type;
   __u8		code;
-  __u16		checksum;
+  __be16	checksum;
   union {
 	struct {
-		__u16	id;
-		__u16	sequence;
+		__be16	id;
+		__be16	sequence;
 	} echo;
-	__u32	gateway;
+	__be32	gateway;
 	struct {
-		__u16	__unused;
-		__u16	mtu;
+		__be16	__unused;
+		__be16	mtu;
 	} frag;
   } un;
 };
-- 
cgit v1.2.3


From 1b620154273d5cc57690e0d199282c6bb9e56974 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:39:09 -0700
Subject: [IPV4]: PIMv2 header annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/mroute.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index e05d54a90743..c7dd4c11f667 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -213,8 +213,8 @@ struct pimreghdr
 {
 	__u8	type;
 	__u8	reserved;
-	__u16	csum;
-	__u32	flags;
+	__be16	csum;
+	__be32	flags;
 };
 
 extern int pim_rcv_v1(struct sk_buff *);
-- 
cgit v1.2.3


From 114c7844f34c1608aec20ae7ff85cec471ac90ae Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:39:29 -0700
Subject: [IPV4]: mroute annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/mroute.h | 10 +++++-----
 net/ipv4/ipmr.c        |  4 ++--
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index c7dd4c11f667..7da2cee8e132 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -142,7 +142,7 @@ struct vif_device
 	unsigned long	rate_limit;		/* Traffic shaping (NI) 	*/
 	unsigned char	threshold;		/* TTL threshold 		*/
 	unsigned short	flags;			/* Control flags 		*/
-	__u32		local,remote;		/* Addresses(remote for tunnels)*/
+	__be32		local,remote;		/* Addresses(remote for tunnels)*/
 	int		link;			/* Physical interface index	*/
 };
 
@@ -151,8 +151,8 @@ struct vif_device
 struct mfc_cache 
 {
 	struct mfc_cache *next;			/* Next entry on cache line 	*/
-	__u32 mfc_mcastgrp;			/* Group the entry belongs to 	*/
-	__u32 mfc_origin;			/* Source of packet 		*/
+	__be32 mfc_mcastgrp;			/* Group the entry belongs to 	*/
+	__be32 mfc_origin;			/* Source of packet 		*/
 	vifi_t mfc_parent;			/* Source interface		*/
 	int mfc_flags;				/* Flags on line		*/
 
@@ -179,9 +179,9 @@ struct mfc_cache
 #define MFC_LINES		64
 
 #ifdef __BIG_ENDIAN
-#define MFC_HASH(a,b)	((((a)>>24)^((b)>>26))&(MFC_LINES-1))
+#define MFC_HASH(a,b)	(((((__force u32)(__be32)a)>>24)^(((__force u32)(__be32)b)>>26))&(MFC_LINES-1))
 #else
-#define MFC_HASH(a,b)	(((a)^((b)>>2))&(MFC_LINES-1))
+#define MFC_HASH(a,b)	((((__force u32)(__be32)a)^(((__force u32)(__be32)b)>>2))&(MFC_LINES-1))
 #endif		
 
 #endif
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index ba49588da242..97cfa97c8abb 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -462,7 +462,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock)
 	return 0;
 }
 
-static struct mfc_cache *ipmr_cache_find(__u32 origin, __u32 mcastgrp)
+static struct mfc_cache *ipmr_cache_find(__be32 origin, __be32 mcastgrp)
 {
 	int line=MFC_HASH(mcastgrp,origin);
 	struct mfc_cache *c;
@@ -1097,7 +1097,7 @@ static struct notifier_block ip_mr_notifier={
  *	important for multicast video.
  */
  
-static void ip_encap(struct sk_buff *skb, u32 saddr, u32 daddr)
+static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr)
 {
 	struct iphdr *iph = (struct iphdr *)skb_push(skb,sizeof(struct iphdr));
 
-- 
cgit v1.2.3


From 81f7bf6cbaca02c034b0393c51fc22b29cba20f7 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:40:00 -0700
Subject: [IPV4]: net/ipv4/fib annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_fib.h     |  2 +-
 net/ipv4/fib_frontend.c  |  7 +++----
 net/ipv4/fib_lookup.h    |  4 ++--
 net/ipv4/fib_rules.c     | 12 ++++++------
 net/ipv4/fib_semantics.c | 12 ++++++------
 5 files changed, 18 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 19f70896ea98..82229146bac7 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -230,7 +230,7 @@ struct rtentry;
 
 /* Exported by fib_semantics.c */
 extern int ip_fib_check_default(__be32 gw, struct net_device *dev);
-extern int fib_sync_down(u32 local, struct net_device *dev, int force);
+extern int fib_sync_down(__be32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
 extern __be32  __fib_res_prefsrc(struct fib_result *res);
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 34dc640478a0..9c399a70dd5d 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -253,7 +253,7 @@ e_inval:
 
 #ifndef CONFIG_IP_NOSIOCRT
 
-static inline u32 sk_extract_addr(struct sockaddr *addr)
+static inline __be32 sk_extract_addr(struct sockaddr *addr)
 {
 	return ((struct sockaddr_in *) addr)->sin_addr.s_addr;
 }
@@ -292,7 +292,7 @@ static int rtentry_to_fib_config(int cmd, struct rtentry *rt,
 	plen = 32;
 	addr = sk_extract_addr(&rt->rt_dst);
 	if (!(rt->rt_flags & RTF_HOST)) {
-		u32 mask = sk_extract_addr(&rt->rt_genmask);
+		__be32 mask = sk_extract_addr(&rt->rt_genmask);
 
 		if (rt->rt_genmask.sa_family != AF_INET) {
 			if (mask || rt->rt_genmask.sa_family)
@@ -627,8 +627,7 @@ out:
    only when netlink is already locked.
  */
 
-static void fib_magic(int cmd, int type, u32 dst, int dst_len,
-		      struct in_ifaddr *ifa)
+static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifaddr *ifa)
 {
 	struct fib_table *tb;
 	struct fib_config cfg = {
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index 6e9b3e86ae84..0e8b70bad4e1 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -26,10 +26,10 @@ extern void fib_release_info(struct fib_info *);
 extern struct fib_info *fib_create_info(struct fib_config *cfg);
 extern int fib_nh_match(struct fib_config *cfg, struct fib_info *fi);
 extern int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-			 u32 tb_id, u8 type, u8 scope, u32 dst,
+			 u32 tb_id, u8 type, u8 scope, __be32 dst,
 			 int dst_len, u8 tos, struct fib_info *fi,
 			 unsigned int);
-extern void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
+extern void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 		      int dst_len, u32 tb_id, struct nl_info *info);
 extern struct fib_alias *fib_find_alias(struct list_head *fah,
 					u8 tos, u32 prio);
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 52b2adae4f22..f2d4070aaf01 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -40,10 +40,10 @@ struct fib4_rule
 	u8			dst_len;
 	u8			src_len;
 	u8			tos;
-	u32			src;
-	u32			srcmask;
-	u32			dst;
-	u32			dstmask;
+	__be32			src;
+	__be32			srcmask;
+	__be32			dst;
+	__be32			dstmask;
 #ifdef CONFIG_IP_ROUTE_FWMARK
 	u32			fwmark;
 	u32			fwmask;
@@ -150,8 +150,8 @@ void fib_select_default(const struct flowi *flp, struct fib_result *res)
 static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 {
 	struct fib4_rule *r = (struct fib4_rule *) rule;
-	u32 daddr = fl->fl4_dst;
-	u32 saddr = fl->fl4_src;
+	__be32 daddr = fl->fl4_dst;
+	__be32 saddr = fl->fl4_src;
 
 	if (((saddr ^ r->src) & r->srcmask) ||
 	    ((daddr ^ r->dst) & r->dstmask))
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index f7567e92cd9e..884d176e0082 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -203,7 +203,7 @@ static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
 	unsigned int val = fi->fib_nhs;
 
 	val ^= fi->fib_protocol;
-	val ^= fi->fib_prefsrc;
+	val ^= (__force u32)fi->fib_prefsrc;
 	val ^= fi->fib_priority;
 
 	return (val ^ (val >> 7) ^ (val >> 12)) & mask;
@@ -273,7 +273,7 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev)
 	return -1;
 }
 
-void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
+void rtmsg_fib(int event, __be32 key, struct fib_alias *fa,
 	       int dst_len, u32 tb_id, struct nl_info *info)
 {
 	struct sk_buff *skb;
@@ -568,11 +568,11 @@ out:
 	return 0;
 }
 
-static inline unsigned int fib_laddr_hashfn(u32 val)
+static inline unsigned int fib_laddr_hashfn(__be32 val)
 {
 	unsigned int mask = (fib_hash_size - 1);
 
-	return (val ^ (val >> 7) ^ (val >> 14)) & mask;
+	return ((__force u32)val ^ ((__force u32)val >> 7) ^ ((__force u32)val >> 14)) & mask;
 }
 
 static struct hlist_head *fib_hash_alloc(int bytes)
@@ -928,7 +928,7 @@ __be32 __fib_res_prefsrc(struct fib_result *res)
 }
 
 int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-		  u32 tb_id, u8 type, u8 scope, u32 dst, int dst_len, u8 tos,
+		  u32 tb_id, u8 type, u8 scope, __be32 dst, int dst_len, u8 tos,
 		  struct fib_info *fi, unsigned int flags)
 {
 	struct nlmsghdr *nlh;
@@ -1017,7 +1017,7 @@ nla_put_failure:
    - device went down -> we must shutdown all nexthops going via it.
  */
 
-int fib_sync_down(u32 local, struct net_device *dev, int force)
+int fib_sync_down(__be32 local, struct net_device *dev, int force)
 {
 	int ret = 0;
 	int scope = RT_SCOPE_NOWHERE;
-- 
cgit v1.2.3


From 4f765d842fa6e6fe15d555b247b640118d65b4dd Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:43:07 -0700
Subject: [IPV4]: INET_MATCH() annotations

INET_MATCH() and friends depend on an interesting set of kludges:
	* there's a pair of adjacent fields in struct inet_sock - __be16 dport
followed by __u16 num.  We want to search by pair, so we combine the keys into
a single 32bit value and compare with 32bit value read from &...->dport.
	* on 64bit targets we combine comparisons with pair of adjacent __be32
fields in the same way.

Make sure that we don't mix those values with anything else and that pairs
we form them from have correct types.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h          |  2 +-
 include/net/inet_hashtables.h | 36 +++++++++++++++++++++++++-----------
 net/ipv4/inet_hashtables.c    |  2 +-
 net/ipv6/inet6_hashtables.c   |  8 ++++----
 4 files changed, 31 insertions(+), 17 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index caca57df0d7d..6dc07ee7702e 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -461,7 +461,7 @@ static inline struct raw6_sock *raw6_sk(const struct sock *sk)
 
 #define INET6_MATCH(__sk, __hash, __saddr, __daddr, __ports, __dif)\
 	(((__sk)->sk_hash == (__hash))				&& \
-	 ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports))  	&& \
+	 ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports))  	&& \
 	 ((__sk)->sk_family		== AF_INET6)		&& \
 	 ipv6_addr_equal(&inet6_sk(__sk)->daddr, (__saddr))	&& \
 	 ipv6_addr_equal(&inet6_sk(__sk)->rcv_saddr, (__daddr))	&& \
diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index b4491c9e2a5a..fb0c09c7090c 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -283,31 +283,45 @@ static inline struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
 }
 
 /* Socket demux engine toys. */
+/* What happens here is ugly; there's a pair of adjacent fields in
+   struct inet_sock; __be16 dport followed by __u16 num.  We want to
+   search by pair, so we combine the keys into a single 32bit value
+   and compare with 32bit value read from &...->dport.  Let's at least
+   make sure that it's not mixed with anything else...
+   On 64bit targets we combine comparisons with pair of adjacent __be32
+   fields in the same way.
+*/
+typedef __u32 __bitwise __portpair;
 #ifdef __BIG_ENDIAN
 #define INET_COMBINED_PORTS(__sport, __dport) \
-	(((__u32)(__sport) << 16) | (__u32)(__dport))
+	((__force __portpair)(((__force __u32)(__be16)(__sport) << 16) | (__u32)(__dport)))
 #else /* __LITTLE_ENDIAN */
 #define INET_COMBINED_PORTS(__sport, __dport) \
-	(((__u32)(__dport) << 16) | (__u32)(__sport))
+	((__force __portpair)(((__u32)(__dport) << 16) | (__force __u32)(__be16)(__sport)))
 #endif
 
 #if (BITS_PER_LONG == 64)
+typedef __u64 __bitwise __addrpair;
 #ifdef __BIG_ENDIAN
 #define INET_ADDR_COOKIE(__name, __saddr, __daddr) \
-	const __u64 __name = (((__u64)(__saddr)) << 32) | ((__u64)(__daddr));
+	const __addrpair __name = (__force __addrpair) ( \
+				   (((__force __u64)(__be32)(__saddr)) << 32) | \
+				   ((__force __u64)(__be32)(__daddr)));
 #else /* __LITTLE_ENDIAN */
 #define INET_ADDR_COOKIE(__name, __saddr, __daddr) \
-	const __u64 __name = (((__u64)(__daddr)) << 32) | ((__u64)(__saddr));
+	const __addrpair __name = (__force __addrpair) ( \
+				   (((__force __u64)(__be32)(__daddr)) << 32) | \
+				   ((__force __u64)(__be32)(__saddr)));
 #endif /* __BIG_ENDIAN */
 #define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
 	(((__sk)->sk_hash == (__hash))				&&	\
-	 ((*((__u64 *)&(inet_sk(__sk)->daddr))) == (__cookie))	&&	\
-	 ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports))	&&	\
+	 ((*((__addrpair *)&(inet_sk(__sk)->daddr))) == (__cookie))	&&	\
+	 ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports))	&&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #define INET_TW_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
 	(((__sk)->sk_hash == (__hash))				&&	\
-	 ((*((__u64 *)&(inet_twsk(__sk)->tw_daddr))) == (__cookie)) &&	\
-	 ((*((__u32 *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) &&	\
+	 ((*((__addrpair *)&(inet_twsk(__sk)->tw_daddr))) == (__cookie)) &&	\
+	 ((*((__portpair *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) &&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #else /* 32-bit arch */
 #define INET_ADDR_COOKIE(__name, __saddr, __daddr)
@@ -315,13 +329,13 @@ static inline struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
 	(((__sk)->sk_hash == (__hash))				&&	\
 	 (inet_sk(__sk)->daddr		== (__saddr))		&&	\
 	 (inet_sk(__sk)->rcv_saddr	== (__daddr))		&&	\
-	 ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports))	&&	\
+	 ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports))	&&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #define INET_TW_MATCH(__sk, __hash,__cookie, __saddr, __daddr, __ports, __dif)	\
 	(((__sk)->sk_hash == (__hash))				&&	\
 	 (inet_twsk(__sk)->tw_daddr	== (__saddr))		&&	\
 	 (inet_twsk(__sk)->tw_rcv_saddr	== (__daddr))		&&	\
-	 ((*((__u32 *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) &&	\
+	 ((*((__portpair *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) &&	\
 	 (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
 #endif /* 64-bit arch */
 
@@ -338,7 +352,7 @@ static inline struct sock *
 				  const int dif)
 {
 	INET_ADDR_COOKIE(acookie, saddr, daddr)
-	const __u32 ports = INET_COMBINED_PORTS(sport, hnum);
+	const __portpair ports = INET_COMBINED_PORTS(sport, hnum);
 	struct sock *sk;
 	const struct hlist_node *node;
 	/* Optimize here for direct hit, only listening connections can
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index fb296c9a7f3f..729d1eb39483 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -201,7 +201,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row,
 	u32 saddr = inet->daddr;
 	int dif = sk->sk_bound_dev_if;
 	INET_ADDR_COOKIE(acookie, saddr, daddr)
-	const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
+	const __portpair ports = INET_COMBINED_PORTS(inet->dport, lport);
 	unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport);
 	struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash);
 	struct sock *sk2;
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index d2f3fc990bfa..8accd1fbeeda 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -64,7 +64,7 @@ struct sock *__inet6_lookup_established(struct inet_hashinfo *hashinfo,
 {
 	struct sock *sk;
 	const struct hlist_node *node;
-	const __u32 ports = INET_COMBINED_PORTS(sport, hnum);
+	const __portpair ports = INET_COMBINED_PORTS(sport, hnum);
 	/* Optimize here for direct hit, only listening connections can
 	 * have wildcards anyways.
 	 */
@@ -82,7 +82,7 @@ struct sock *__inet6_lookup_established(struct inet_hashinfo *hashinfo,
 	sk_for_each(sk, node, &(head + hashinfo->ehash_size)->chain) {
 		const struct inet_timewait_sock *tw = inet_twsk(sk);
 
-		if(*((__u32 *)&(tw->tw_dport))	== ports	&&
+		if(*((__portpair *)&(tw->tw_dport))	== ports	&&
 		   sk->sk_family		== PF_INET6) {
 			const struct inet6_timewait_sock *tw6 = inet6_twsk(sk);
 
@@ -171,7 +171,7 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row,
 	const struct in6_addr *daddr = &np->rcv_saddr;
 	const struct in6_addr *saddr = &np->daddr;
 	const int dif = sk->sk_bound_dev_if;
-	const u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
+	const __portpair ports = INET_COMBINED_PORTS(inet->dport, lport);
 	const unsigned int hash = inet6_ehashfn(daddr, inet->num, saddr,
 						inet->dport);
 	struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash);
@@ -188,7 +188,7 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row,
 
 		tw = inet_twsk(sk2);
 
-		if(*((__u32 *)&(tw->tw_dport)) == ports		 &&
+		if(*((__portpair *)&(tw->tw_dport)) == ports		 &&
 		   sk2->sk_family	       == PF_INET6	 &&
 		   ipv6_addr_equal(&tw6->tw_v6_daddr, saddr)	 &&
 		   ipv6_addr_equal(&tw6->tw_v6_rcv_saddr, daddr) &&
-- 
cgit v1.2.3


From fb99c848e5ae6b8b2bc11f0f90c9e2bb3d702c0d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:43:33 -0700
Subject: [IPV4]: annotate inet_lookup() and friends

inet_lookup() annotated along with helper functions (__inet_lookup(),
__inet_lookup_established(), inet_lookup_established(),
inet_lookup_listener(), __inet_lookup_listener() and inet_ehashfn())

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_hashtables.h | 20 ++++++++++----------
 include/net/inet_sock.h       | 12 ++++++------
 net/ipv4/inet_hashtables.c    | 10 +++++-----
 3 files changed, 21 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index fb0c09c7090c..a9eb2eaf094e 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -272,12 +272,12 @@ static inline int inet_iif(const struct sk_buff *skb)
 }
 
 extern struct sock *__inet_lookup_listener(struct inet_hashinfo *hashinfo,
-					   const u32 daddr,
+					   const __be32 daddr,
 					   const unsigned short hnum,
 					   const int dif);
 
 static inline struct sock *inet_lookup_listener(struct inet_hashinfo *hashinfo,
-						u32 daddr, u16 dport, int dif)
+						__be32 daddr, __be16 dport, int dif)
 {
 	return __inet_lookup_listener(hashinfo, daddr, ntohs(dport), dif);
 }
@@ -347,8 +347,8 @@ typedef __u64 __bitwise __addrpair;
  */
 static inline struct sock *
 	__inet_lookup_established(struct inet_hashinfo *hashinfo,
-				  const u32 saddr, const u16 sport,
-				  const u32 daddr, const u16 hnum,
+				  const __be32 saddr, const __be16 sport,
+				  const __be32 daddr, const u16 hnum,
 				  const int dif)
 {
 	INET_ADDR_COOKIE(acookie, saddr, daddr)
@@ -384,8 +384,8 @@ hit:
 
 static inline struct sock *
 	inet_lookup_established(struct inet_hashinfo *hashinfo,
-				const u32 saddr, const u16 sport,
-				const u32 daddr, const u16 dport,
+				const __be32 saddr, const __be16 sport,
+				const __be32 daddr, const __be16 dport,
 				const int dif)
 {
 	return __inet_lookup_established(hashinfo, saddr, sport, daddr,
@@ -393,8 +393,8 @@ static inline struct sock *
 }
 
 static inline struct sock *__inet_lookup(struct inet_hashinfo *hashinfo,
-					 const u32 saddr, const u16 sport,
-					 const u32 daddr, const u16 dport,
+					 const __be32 saddr, const __be16 sport,
+					 const __be32 daddr, const __be16 dport,
 					 const int dif)
 {
 	u16 hnum = ntohs(dport);
@@ -404,8 +404,8 @@ static inline struct sock *__inet_lookup(struct inet_hashinfo *hashinfo,
 }
 
 static inline struct sock *inet_lookup(struct inet_hashinfo *hashinfo,
-				       const u32 saddr, const u16 sport,
-				       const u32 daddr, const u16 dport,
+				       const __be32 saddr, const __be16 sport,
+				       const __be32 daddr, const __be16 dport,
 				       const int dif)
 {
 	struct sock *sk;
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index a448bb226bb3..ce6da97bc848 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -167,10 +167,10 @@ static inline void inet_sk_copy_descendant(struct sock *sk_to,
 
 extern int inet_sk_rebuild_header(struct sock *sk);
 
-static inline unsigned int inet_ehashfn(const __u32 laddr, const __u16 lport,
-					const __u32 faddr, const __u16 fport)
+static inline unsigned int inet_ehashfn(const __be32 laddr, const __u16 lport,
+					const __be32 faddr, const __be16 fport)
 {
-	unsigned int h = (laddr ^ lport) ^ (faddr ^ fport);
+	unsigned int h = ((__force __u32)laddr ^ lport) ^ ((__force __u32)faddr ^ (__force __u32)fport);
 	h ^= h >> 16;
 	h ^= h >> 8;
 	return h;
@@ -179,10 +179,10 @@ static inline unsigned int inet_ehashfn(const __u32 laddr, const __u16 lport,
 static inline int inet_sk_ehashfn(const struct sock *sk)
 {
 	const struct inet_sock *inet = inet_sk(sk);
-	const __u32 laddr = inet->rcv_saddr;
+	const __be32 laddr = inet->rcv_saddr;
 	const __u16 lport = inet->num;
-	const __u32 faddr = inet->daddr;
-	const __u16 fport = inet->dport;
+	const __be32 faddr = inet->daddr;
+	const __be16 fport = inet->dport;
 
 	return inet_ehashfn(laddr, lport, faddr, fport);
 }
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 729d1eb39483..244c4f445c7d 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -125,7 +125,7 @@ EXPORT_SYMBOL(inet_listen_wlock);
  * wildcarded during the search since they can never be otherwise.
  */
 static struct sock *inet_lookup_listener_slow(const struct hlist_head *head,
-					      const u32 daddr,
+					      const __be32 daddr,
 					      const unsigned short hnum,
 					      const int dif)
 {
@@ -137,7 +137,7 @@ static struct sock *inet_lookup_listener_slow(const struct hlist_head *head,
 		const struct inet_sock *inet = inet_sk(sk);
 
 		if (inet->num == hnum && !ipv6_only_sock(sk)) {
-			const __u32 rcv_saddr = inet->rcv_saddr;
+			const __be32 rcv_saddr = inet->rcv_saddr;
 			int score = sk->sk_family == PF_INET ? 1 : 0;
 
 			if (rcv_saddr) {
@@ -163,7 +163,7 @@ static struct sock *inet_lookup_listener_slow(const struct hlist_head *head,
 
 /* Optimize the common listener case. */
 struct sock *__inet_lookup_listener(struct inet_hashinfo *hashinfo,
-				    const u32 daddr, const unsigned short hnum,
+				    const __be32 daddr, const unsigned short hnum,
 				    const int dif)
 {
 	struct sock *sk = NULL;
@@ -197,8 +197,8 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row,
 {
 	struct inet_hashinfo *hinfo = death_row->hashinfo;
 	struct inet_sock *inet = inet_sk(sk);
-	u32 daddr = inet->rcv_saddr;
-	u32 saddr = inet->daddr;
+	__be32 daddr = inet->rcv_saddr;
+	__be32 saddr = inet->daddr;
 	int dif = sk->sk_bound_dev_if;
 	INET_ADDR_COOKIE(acookie, saddr, daddr)
 	const __portpair ports = INET_COMBINED_PORTS(inet->dport, lport);
-- 
cgit v1.2.3


From 23f33c2d4fd5986243b67a2bf5e63ebae1a76ffa Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:43:50 -0700
Subject: [IPV4]: struct inet_timewait_sock annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_timewait_sock.h | 8 ++++----
 net/ipv4/tcp_ipv4.c              | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 600cb543550d..5794bc6bf0bf 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -120,10 +120,10 @@ struct inet_timewait_sock {
 	unsigned char		tw_rcv_wscale;
 	/* Socket demultiplex comparisons on incoming packets. */
 	/* these five are in inet_sock */
-	__u16			tw_sport;
-	__u32			tw_daddr __attribute__((aligned(INET_TIMEWAIT_ADDRCMP_ALIGN_BYTES)));
-	__u32			tw_rcv_saddr;
-	__u16			tw_dport;
+	__be16			tw_sport;
+	__be32			tw_daddr __attribute__((aligned(INET_TIMEWAIT_ADDRCMP_ALIGN_BYTES)));
+	__be32			tw_rcv_saddr;
+	__be16			tw_dport;
 	__u16			tw_num;
 	/* And these are ours. */
 	__u8			tw_ipv6only:1;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 89f3cf94685a..c83938b8fcb1 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1763,7 +1763,7 @@ static void get_tcp4_sock(struct sock *sp, char *tmpbuf, int i)
 
 static void get_timewait4_sock(struct inet_timewait_sock *tw, char *tmpbuf, int i)
 {
-	unsigned int dest, src;
+	__be32 dest, src;
 	__u16 destp, srcp;
 	int ttd = tw->tw_ttd - jiffies;
 
-- 
cgit v1.2.3


From 82103232edc4b4ed48949a195aca93cfa3fe3fa8 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:44:10 -0700
Subject: [IPV4]: inet_rcv_saddr() annotations

inet_rcv_saddr() returns net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_timewait_sock.h | 2 +-
 net/ipv4/inet_connection_sock.c  | 4 ++--
 net/ipv6/addrconf.c              | 4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 5794bc6bf0bf..6d14c22a00c5 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -186,7 +186,7 @@ static inline struct inet_timewait_sock *inet_twsk(const struct sock *sk)
 	return (struct inet_timewait_sock *)sk;
 }
 
-static inline u32 inet_rcv_saddr(const struct sock *sk)
+static inline __be32 inet_rcv_saddr(const struct sock *sk)
 {
 	return likely(sk->sk_state != TCP_TIME_WAIT) ?
 		inet_sk(sk)->rcv_saddr : inet_twsk(sk)->tw_rcv_saddr;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 274b0b846c25..96bbe2a0aa1b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -39,7 +39,7 @@ int sysctl_local_port_range[2] = { 1024, 4999 };
 int inet_csk_bind_conflict(const struct sock *sk,
 			   const struct inet_bind_bucket *tb)
 {
-	const u32 sk_rcv_saddr = inet_rcv_saddr(sk);
+	const __be32 sk_rcv_saddr = inet_rcv_saddr(sk);
 	struct sock *sk2;
 	struct hlist_node *node;
 	int reuse = sk->sk_reuse;
@@ -52,7 +52,7 @@ int inet_csk_bind_conflict(const struct sock *sk,
 		     sk->sk_bound_dev_if == sk2->sk_bound_dev_if)) {
 			if (!reuse || !sk2->sk_reuse ||
 			    sk2->sk_state == TCP_LISTEN) {
-				const u32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
+				const __be32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
 				if (!sk2_rcv_saddr || !sk_rcv_saddr ||
 				    sk2_rcv_saddr == sk_rcv_saddr)
 					break;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index c18676352397..e03c33b2465b 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1258,8 +1258,8 @@ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2)
 {
 	const struct in6_addr *sk_rcv_saddr6 = &inet6_sk(sk)->rcv_saddr;
 	const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2);
-	u32 sk_rcv_saddr = inet_sk(sk)->rcv_saddr;
-	u32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
+	__be32 sk_rcv_saddr = inet_sk(sk)->rcv_saddr;
+	__be32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
 	int sk_ipv6only = ipv6_only_sock(sk);
 	int sk2_ipv6only = inet_v6_ipv6only(sk2);
 	int addr_type = ipv6_addr_type(sk_rcv_saddr6);
-- 
cgit v1.2.3


From 9f8552996d969f56039ec88128cf5ad35b12f141 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:44:30 -0700
Subject: [IPV4]: inet_diag annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inet_diag.h | 10 +++++-----
 net/ipv4/inet_diag.c      | 12 ++++++------
 2 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h
index a4606e5810e5..6e8bc548635a 100644
--- a/include/linux/inet_diag.h
+++ b/include/linux/inet_diag.h
@@ -9,10 +9,10 @@
 
 /* Socket identity */
 struct inet_diag_sockid {
-	__u16	idiag_sport;
-	__u16	idiag_dport;
-	__u32	idiag_src[4];
-	__u32	idiag_dst[4];
+	__be16	idiag_sport;
+	__be16	idiag_dport;
+	__be32	idiag_src[4];
+	__be32	idiag_dst[4];
 	__u32	idiag_if;
 	__u32	idiag_cookie[2];
 #define INET_DIAG_NOCOOKIE (~0U)
@@ -67,7 +67,7 @@ struct inet_diag_hostcond {
 	__u8	family;
 	__u8	prefix_len;
 	int	port;
-	__u32	addr[0];
+	__be32	addr[0];
 };
 
 /* Base info structure. It contains socket identity (addrs/ports/cookie)
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 492858e6faf0..77761ac4f7bb 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -36,8 +36,8 @@
 static const struct inet_diag_handler **inet_diag_table;
 
 struct inet_diag_entry {
-	u32 *saddr;
-	u32 *daddr;
+	__be32 *saddr;
+	__be32 *daddr;
 	u16 sport;
 	u16 dport;
 	u16 family;
@@ -294,7 +294,7 @@ out:
 	return err;
 }
 
-static int bitstring_match(const u32 *a1, const u32 *a2, int bits)
+static int bitstring_match(const __be32 *a1, const __be32 *a2, int bits)
 {
 	int words = bits >> 5;
 
@@ -305,8 +305,8 @@ static int bitstring_match(const u32 *a1, const u32 *a2, int bits)
 			return 0;
 	}
 	if (bits) {
-		__u32 w1, w2;
-		__u32 mask;
+		__be32 w1, w2;
+		__be32 mask;
 
 		w1 = a1[words];
 		w2 = a2[words];
@@ -352,7 +352,7 @@ static int inet_diag_bc_run(const void *bc, int len,
 		case INET_DIAG_BC_S_COND:
 		case INET_DIAG_BC_D_COND: {
 			struct inet_diag_hostcond *cond;
-			u32 *addr;
+			__be32 *addr;
 
 			cond = (struct inet_diag_hostcond *)(op + 1);
 			if (cond->port != -1 &&
-- 
cgit v1.2.3


From 48818f822d2b2e16f4bf4d1ed1185e7d2146dc34 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:44:54 -0700
Subject: [IPV6]: struct in6_addr annotations

in6_addr elements are net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in6.h | 4 ++--
 include/net/ipv6.h  | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index d776829b443f..348ecd4a36fe 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -32,8 +32,8 @@ struct in6_addr
 	union 
 	{
 		__u8		u6_addr8[16];
-		__u16		u6_addr16[8];
-		__u32		u6_addr32[4];
+		__be16		u6_addr16[8];
+		__be32		u6_addr32[4];
 	} in6_u;
 #define s6_addr			in6_u.u6_addr8
 #define s6_addr16		in6_u.u6_addr16
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 72bf47b2a4e0..8223c4410b4b 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -318,8 +318,8 @@ static inline void ipv6_addr_prefix(struct in6_addr *pfx,
 
 #ifndef __HAVE_ARCH_ADDR_SET
 static inline void ipv6_addr_set(struct in6_addr *addr, 
-				     __u32 w1, __u32 w2,
-				     __u32 w3, __u32 w4)
+				     __be32 w1, __be32 w2,
+				     __be32 w3, __be32 w4)
 {
 	addr->s6_addr32[0] = w1;
 	addr->s6_addr32[1] = w2;
@@ -337,7 +337,7 @@ static inline int ipv6_addr_equal(const struct in6_addr *a1,
 		a1->s6_addr32[3] == a2->s6_addr32[3]);
 }
 
-static inline int __ipv6_prefix_equal(const u32 *a1, const u32 *a2,
+static inline int __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2,
 				      unsigned int prefixlen)
 {
 	unsigned pdw, pbi;
-- 
cgit v1.2.3


From 43505077df075545e9b28b7c6ea12d82b3caf036 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:45:11 -0700
Subject: [IPV6]: IPv6 headers annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 6dc07ee7702e..4f435c59de06 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -99,22 +99,22 @@ struct ipv6_destopt_hao {
 struct ipv6_auth_hdr {
 	__u8  nexthdr;
 	__u8  hdrlen;           /* This one is measured in 32 bit units! */
-	__u16 reserved;
-	__u32 spi;
-	__u32 seq_no;           /* Sequence number */
+	__be16 reserved;
+	__be32 spi;
+	__be32 seq_no;           /* Sequence number */
 	__u8  auth_data[0];     /* Length variable but >=4. Mind the 64 bit alignment! */
 };
 
 struct ipv6_esp_hdr {
-	__u32 spi;
-	__u32 seq_no;           /* Sequence number */
+	__be32 spi;
+	__be32 seq_no;           /* Sequence number */
 	__u8  enc_data[0];      /* Length variable but >=8. Mind the 64 bit alignment! */
 };
 
 struct ipv6_comp_hdr {
 	__u8 nexthdr;
 	__u8 flags;
-	__u16 cpi;
+	__be16 cpi;
 };
 
 /*
@@ -136,7 +136,7 @@ struct ipv6hdr {
 #endif
 	__u8			flow_lbl[3];
 
-	__u16			payload_len;
+	__be16			payload_len;
 	__u8			nexthdr;
 	__u8			hop_limit;
 
-- 
cgit v1.2.3


From e2e38e819bd03e9674c102aaa2c9dc8151a3c3d0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:45:27 -0700
Subject: [IPV6]: sin6_port is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in6.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index 348ecd4a36fe..9be6a4756f0b 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -53,7 +53,7 @@ extern const struct in6_addr in6addr_loopback;
 
 struct sockaddr_in6 {
 	unsigned short int	sin6_family;    /* AF_INET6 */
-	__u16			sin6_port;      /* Transport layer port # */
+	__be16			sin6_port;      /* Transport layer port # */
 	__u32			sin6_flowinfo;  /* IPv6 flow information */
 	struct in6_addr		sin6_addr;      /* IPv6 address */
 	__u32			sin6_scope_id;  /* scope id (new in RFC2553) */
-- 
cgit v1.2.3


From f9d07e41f89e7305eb2c0475c170c51d21425581 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:45:50 -0700
Subject: [XFRM]: xfrm_flowi_[sd]port() annotations

both return net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 11e0b1d6bd47..17e98c4bd877 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -462,9 +462,9 @@ static __inline__ int addr_match(void *token1, void *token2, int prefixlen)
 }
 
 static __inline__
-u16 xfrm_flowi_sport(struct flowi *fl)
+__be16 xfrm_flowi_sport(struct flowi *fl)
 {
-	u16 port;
+	__be16 port;
 	switch(fl->proto) {
 	case IPPROTO_TCP:
 	case IPPROTO_UDP:
@@ -487,9 +487,9 @@ u16 xfrm_flowi_sport(struct flowi *fl)
 }
 
 static __inline__
-u16 xfrm_flowi_dport(struct flowi *fl)
+__be16 xfrm_flowi_dport(struct flowi *fl)
 {
-	u16 port;
+	__be16 port;
 	switch(fl->proto) {
 	case IPPROTO_TCP:
 	case IPPROTO_UDP:
-- 
cgit v1.2.3


From 8f83f23e6db8b9a9fe787d02f73489224668c4e2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:46:11 -0700
Subject: [XFRM]: ports in struct xfrm_selector annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h   | 8 ++++----
 net/ipv4/xfrm4_state.c | 4 ++--
 net/ipv6/xfrm6_state.c | 4 ++--
 net/key/af_key.c       | 8 ++++----
 4 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 14ecd19f4cdc..3aae9b9ce79b 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -49,10 +49,10 @@ struct xfrm_selector
 {
 	xfrm_address_t	daddr;
 	xfrm_address_t	saddr;
-	__u16	dport;
-	__u16	dport_mask;
-	__u16	sport;
-	__u16	sport_mask;
+	__be16	dport;
+	__be16	dport_mask;
+	__be16	sport;
+	__be16	sport_mask;
 	__u16	family;
 	__u8	prefixlen_d;
 	__u8	prefixlen_s;
diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c
index fe2034494d08..3cc3df0c6ece 100644
--- a/net/ipv4/xfrm4_state.c
+++ b/net/ipv4/xfrm4_state.c
@@ -29,9 +29,9 @@ __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	x->sel.daddr.a4 = fl->fl4_dst;
 	x->sel.saddr.a4 = fl->fl4_src;
 	x->sel.dport = xfrm_flowi_dport(fl);
-	x->sel.dport_mask = ~0;
+	x->sel.dport_mask = htons(0xffff);
 	x->sel.sport = xfrm_flowi_sport(fl);
-	x->sel.sport_mask = ~0;
+	x->sel.sport_mask = htons(0xffff);
 	x->sel.prefixlen_d = 32;
 	x->sel.prefixlen_s = 32;
 	x->sel.proto = fl->proto;
diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c
index 711bfafb2472..9ddaa9d41539 100644
--- a/net/ipv6/xfrm6_state.c
+++ b/net/ipv6/xfrm6_state.c
@@ -29,9 +29,9 @@ __xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	ipv6_addr_copy((struct in6_addr *)&x->sel.daddr, &fl->fl6_dst);
 	ipv6_addr_copy((struct in6_addr *)&x->sel.saddr, &fl->fl6_src);
 	x->sel.dport = xfrm_flowi_dport(fl);
-	x->sel.dport_mask = ~0;
+	x->sel.dport_mask = htons(0xffff);
 	x->sel.sport = xfrm_flowi_sport(fl);
-	x->sel.sport_mask = ~0;
+	x->sel.sport_mask = htons(0xffff);
 	x->sel.prefixlen_d = 128;
 	x->sel.prefixlen_s = 128;
 	x->sel.proto = fl->proto;
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 83b443ddc72f..ff98e70b0931 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -2140,7 +2140,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h
 	xp->selector.proto = pfkey_proto_to_xfrm(sa->sadb_address_proto);
 	xp->selector.sport = ((struct sockaddr_in *)(sa+1))->sin_port;
 	if (xp->selector.sport)
-		xp->selector.sport_mask = ~0;
+		xp->selector.sport_mask = htons(0xffff);
 
 	sa = ext_hdrs[SADB_EXT_ADDRESS_DST-1], 
 	pfkey_sadb_addr2xfrm_addr(sa, &xp->selector.daddr);
@@ -2153,7 +2153,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h
 
 	xp->selector.dport = ((struct sockaddr_in *)(sa+1))->sin_port;
 	if (xp->selector.dport)
-		xp->selector.dport_mask = ~0;
+		xp->selector.dport_mask = htons(0xffff);
 
 	sec_ctx = (struct sadb_x_sec_ctx *) ext_hdrs[SADB_X_EXT_SEC_CTX-1];
 	if (sec_ctx != NULL) {
@@ -2243,7 +2243,7 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg
 	sel.proto = pfkey_proto_to_xfrm(sa->sadb_address_proto);
 	sel.sport = ((struct sockaddr_in *)(sa+1))->sin_port;
 	if (sel.sport)
-		sel.sport_mask = ~0;
+		sel.sport_mask = htons(0xffff);
 
 	sa = ext_hdrs[SADB_EXT_ADDRESS_DST-1], 
 	pfkey_sadb_addr2xfrm_addr(sa, &sel.daddr);
@@ -2251,7 +2251,7 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg
 	sel.proto = pfkey_proto_to_xfrm(sa->sadb_address_proto);
 	sel.dport = ((struct sockaddr_in *)(sa+1))->sin_port;
 	if (sel.dport)
-		sel.dport_mask = ~0;
+		sel.dport_mask = htons(0xffff);
 
 	sec_ctx = (struct sadb_x_sec_ctx *) ext_hdrs[SADB_X_EXT_SEC_CTX-1];
 	memset(&tmp, 0, sizeof(struct xfrm_policy));
-- 
cgit v1.2.3


From 5f19343fb19613539355296b23cbc08d1336b52d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:46:32 -0700
Subject: [XFRM]: addr_match() annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 17e98c4bd877..b096e9058f3f 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -437,8 +437,8 @@ static inline void xfrm_state_hold(struct xfrm_state *x)
 
 static __inline__ int addr_match(void *token1, void *token2, int prefixlen)
 {
-	__u32 *a1 = token1;
-	__u32 *a2 = token2;
+	__be32 *a1 = token1;
+	__be32 *a2 = token2;
 	int pdw;
 	int pbi;
 
@@ -450,7 +450,7 @@ static __inline__ int addr_match(void *token1, void *token2, int prefixlen)
 			return 0;
 
 	if (pbi) {
-		__u32 mask;
+		__be32 mask;
 
 		mask = htonl((0xffffffff) << (32 - pbi));
 
-- 
cgit v1.2.3


From 737b5761dfc609b5be4163deb2cf09226f56bcbc Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:46:48 -0700
Subject: [XFRM]: xfrm_address_t annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 3aae9b9ce79b..4b321842d65f 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -12,8 +12,8 @@
  */
 typedef union
 {
-	__u32		a4;
-	__u32		a6[4];
+	__be32		a4;
+	__be32		a6[4];
 } xfrm_address_t;
 
 /* Ident of a specific xfrm_state. It is used on input to lookup
-- 
cgit v1.2.3


From 26977b4ed728ae911a162b16dbfe1a165b7cf9a1 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:47:05 -0700
Subject: [XFRM]: xfrm_alloc_spi() annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    |  2 +-
 net/xfrm/xfrm_state.c | 10 +++++-----
 2 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index b096e9058f3f..80a19748e96a 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -989,7 +989,7 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir,
 struct xfrm_policy *xfrm_policy_byid(u8, int dir, u32 id, int delete);
 void xfrm_policy_flush(u8 type);
 u32 xfrm_get_acqseq(void);
-void xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi);
+void xfrm_alloc_spi(struct xfrm_state *x, __be32 minspi, __be32 maxspi);
 struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto, 
 				  xfrm_address_t *daddr, xfrm_address_t *saddr, 
 				  int create, unsigned short family);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 9f63edd39346..e40a8862db5d 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1040,7 +1040,7 @@ u32 xfrm_get_acqseq(void)
 EXPORT_SYMBOL(xfrm_get_acqseq);
 
 void
-xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi)
+xfrm_alloc_spi(struct xfrm_state *x, __be32 minspi, __be32 maxspi)
 {
 	unsigned int h;
 	struct xfrm_state *x0;
@@ -1057,10 +1057,10 @@ xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi)
 		x->id.spi = minspi;
 	} else {
 		u32 spi = 0;
-		minspi = ntohl(minspi);
-		maxspi = ntohl(maxspi);
-		for (h=0; h<maxspi-minspi+1; h++) {
-			spi = minspi + net_random()%(maxspi-minspi+1);
+		u32 low = ntohl(minspi);
+		u32 high = ntohl(maxspi);
+		for (h=0; h<high-low+1; h++) {
+			spi = low + net_random()%(high-low+1);
 			x0 = xfrm_state_lookup(&x->id.daddr, htonl(spi), x->id.proto, x->props.family);
 			if (x0 == NULL) {
 				x->id.spi = htonl(spi);
-- 
cgit v1.2.3


From a94cfd19744a568d97b14bbaa500b2a0c3684f34 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:47:24 -0700
Subject: [XFRM]: xfrm_state_lookup() annotations

spi argument of xfrm_state_lookup() is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    | 2 +-
 net/ipv4/ipcomp.c     | 2 +-
 net/ipv6/ipcomp6.c    | 4 ++--
 net/xfrm/xfrm_state.c | 4 ++--
 4 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 80a19748e96a..6da1c7c72b47 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -912,7 +912,7 @@ extern int xfrm_state_check_expire(struct xfrm_state *x);
 extern void xfrm_state_insert(struct xfrm_state *x);
 extern int xfrm_state_add(struct xfrm_state *x);
 extern int xfrm_state_update(struct xfrm_state *x);
-extern struct xfrm_state *xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family);
+extern struct xfrm_state *xfrm_state_lookup(xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family);
 extern struct xfrm_state *xfrm_state_lookup_byaddr(xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family);
 #ifdef CONFIG_XFRM_SUB_POLICY
 extern int xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src,
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index 17342430a843..2017d36024d4 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -183,7 +183,7 @@ out_ok:
 
 static void ipcomp4_err(struct sk_buff *skb, u32 info)
 {
-	u32 spi;
+	__be32 spi;
 	struct iphdr *iph = (struct iphdr *)skb->data;
 	struct ip_comp_hdr *ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
 	struct xfrm_state *x;
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index ad9c6e824e62..a2860e35efd7 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -178,7 +178,7 @@ out_ok:
 static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		                int type, int code, int offset, __u32 info)
 {
-	u32 spi;
+	__be32 spi;
 	struct ipv6hdr *iph = (struct ipv6hdr*)skb->data;
 	struct ipv6_comp_hdr *ipcomph = (struct ipv6_comp_hdr*)(skb->data+offset);
 	struct xfrm_state *x;
@@ -234,7 +234,7 @@ static int ipcomp6_tunnel_attach(struct xfrm_state *x)
 {
 	int err = 0;
 	struct xfrm_state *t = NULL;
-	u32 spi;
+	__be32 spi;
 
 	spi = xfrm6_tunnel_spi_lookup((xfrm_address_t *)&x->props.saddr);
 	if (spi)
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index e40a8862db5d..3692a4783a75 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -421,7 +421,7 @@ xfrm_init_tempsel(struct xfrm_state *x, struct flowi *fl,
 	return 0;
 }
 
-static struct xfrm_state *__xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family)
+static struct xfrm_state *__xfrm_state_lookup(xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family)
 {
 	unsigned int h = xfrm_spi_hash(daddr, spi, proto, family);
 	struct xfrm_state *x;
@@ -916,7 +916,7 @@ err:
 EXPORT_SYMBOL(xfrm_state_check);
 
 struct xfrm_state *
-xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto,
+xfrm_state_lookup(xfrm_address_t *daddr, __be32 spi, u8 proto,
 		  unsigned short family)
 {
 	struct xfrm_state *x;
-- 
cgit v1.2.3


From e037c39bf965ca66fde902e191d849a90de278fe Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:47:40 -0700
Subject: [XFRM]: struct xfrm_id annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index 4b321842d65f..c894267ff5dd 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -23,7 +23,7 @@ typedef union
 struct xfrm_id
 {
 	xfrm_address_t	daddr;
-	__u32		spi;
+	__be32		spi;
 	__u8		proto;
 };
 
-- 
cgit v1.2.3


From 6067b2baba32211e84d1ef2dba863422281bd6c7 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:47:59 -0700
Subject: [XFRM]: xfrm_parse_spi() annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     | 2 +-
 net/ipv4/xfrm4_input.c | 4 ++--
 net/ipv6/xfrm6_input.c | 2 +-
 net/xfrm/xfrm_input.c  | 8 ++++----
 4 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 6da1c7c72b47..460551968560 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1004,7 +1004,7 @@ extern void km_policy_expired(struct xfrm_policy *pol, int dir, int hard, u32 pi
 extern int km_report(u8 proto, struct xfrm_selector *sel, xfrm_address_t *addr);
 
 extern void xfrm_input_init(void);
-extern int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, u32 *spi, u32 *seq);
+extern int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, __be32 *spi, __be32 *seq);
 
 extern void xfrm_probe_algs(void);
 extern int xfrm_count_auth_supported(void);
diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c
index 040e8475f295..8655d038364c 100644
--- a/net/ipv4/xfrm4_input.c
+++ b/net/ipv4/xfrm4_input.c
@@ -23,7 +23,7 @@ int xfrm4_rcv(struct sk_buff *skb)
 
 EXPORT_SYMBOL(xfrm4_rcv);
 
-static int xfrm4_parse_spi(struct sk_buff *skb, u8 nexthdr, u32 *spi, u32 *seq)
+static int xfrm4_parse_spi(struct sk_buff *skb, u8 nexthdr, __be32 *spi, __be32 *seq)
 {
 	switch (nexthdr) {
 	case IPPROTO_IPIP:
@@ -55,7 +55,7 @@ drop:
 int xfrm4_rcv_encap(struct sk_buff *skb, __u16 encap_type)
 {
 	int err;
-	u32 spi, seq;
+	__be32 spi, seq;
 	struct xfrm_state *xfrm_vec[XFRM_MAX_DEPTH];
 	struct xfrm_state *x;
 	int xfrm_nr = 0;
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index a40a05789013..140aa8769acc 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -19,7 +19,7 @@
 int xfrm6_rcv_spi(struct sk_buff *skb, u32 spi)
 {
 	int err;
-	u32 seq;
+	__be32 seq;
 	struct xfrm_state *xfrm_vec[XFRM_MAX_DEPTH];
 	struct xfrm_state *x;
 	int xfrm_nr = 0;
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index dfc90bb1cf1f..e8198a2c785d 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -46,7 +46,7 @@ EXPORT_SYMBOL(secpath_dup);
 
 /* Fetch spi and seq from ipsec header */
 
-int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, u32 *spi, u32 *seq)
+int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, __be32 *spi, __be32 *seq)
 {
 	int offset, offset_seq;
 
@@ -62,7 +62,7 @@ int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, u32 *spi, u32 *seq)
 	case IPPROTO_COMP:
 		if (!pskb_may_pull(skb, sizeof(struct ip_comp_hdr)))
 			return -EINVAL;
-		*spi = htonl(ntohs(*(u16*)(skb->h.raw + 2)));
+		*spi = htonl(ntohs(*(__be16*)(skb->h.raw + 2)));
 		*seq = 0;
 		return 0;
 	default:
@@ -72,8 +72,8 @@ int xfrm_parse_spi(struct sk_buff *skb, u8 nexthdr, u32 *spi, u32 *seq)
 	if (!pskb_may_pull(skb, 16))
 		return -EINVAL;
 
-	*spi = *(u32*)(skb->h.raw + offset);
-	*seq = *(u32*)(skb->h.raw + offset_seq);
+	*spi = *(__be32*)(skb->h.raw + offset);
+	*seq = *(__be32*)(skb->h.raw + offset_seq);
 	return 0;
 }
 EXPORT_SYMBOL(xfrm_parse_spi);
-- 
cgit v1.2.3


From a252cc2371930debe3162f1ac91467b9791324cb Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:48:18 -0700
Subject: [XFRM]: xrfm_replay_check() annotations

seq argument is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h      | 4 ++--
 net/ipv6/xfrm6_input.c  | 2 +-
 net/ipv6/xfrm6_tunnel.c | 2 +-
 net/xfrm/xfrm_state.c   | 5 ++---
 4 files changed, 6 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 460551968560..e6110559e505 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -935,7 +935,7 @@ static inline int xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **s
 extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq);
 extern int xfrm_state_delete(struct xfrm_state *x);
 extern void xfrm_state_flush(u8 proto);
-extern int xfrm_replay_check(struct xfrm_state *x, u32 seq);
+extern int xfrm_replay_check(struct xfrm_state *x, __be32 seq);
 extern void xfrm_replay_advance(struct xfrm_state *x, u32 seq);
 extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_check(struct xfrm_state *x, struct sk_buff *skb);
@@ -945,7 +945,7 @@ extern int xfrm4_rcv(struct sk_buff *skb);
 extern int xfrm4_output(struct sk_buff *skb);
 extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler);
 extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler);
-extern int xfrm6_rcv_spi(struct sk_buff *skb, u32 spi);
+extern int xfrm6_rcv_spi(struct sk_buff *skb, __be32 spi);
 extern int xfrm6_rcv(struct sk_buff **pskb);
 extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr,
 			    xfrm_address_t *saddr, u8 proto);
diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c
index 140aa8769acc..5c8b7a568800 100644
--- a/net/ipv6/xfrm6_input.c
+++ b/net/ipv6/xfrm6_input.c
@@ -16,7 +16,7 @@
 #include <net/ipv6.h>
 #include <net/xfrm.h>
 
-int xfrm6_rcv_spi(struct sk_buff *skb, u32 spi)
+int xfrm6_rcv_spi(struct sk_buff *skb, __be32 spi)
 {
 	int err;
 	__be32 seq;
diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
index 59685ee8f700..7af227bb1551 100644
--- a/net/ipv6/xfrm6_tunnel.c
+++ b/net/ipv6/xfrm6_tunnel.c
@@ -258,7 +258,7 @@ static int xfrm6_tunnel_input(struct xfrm_state *x, struct sk_buff *skb)
 static int xfrm6_tunnel_rcv(struct sk_buff *skb)
 {
 	struct ipv6hdr *iph = skb->nh.ipv6h;
-	u32 spi;
+	__be32 spi;
 
 	spi = xfrm6_tunnel_spi_lookup((xfrm_address_t *)&iph->saddr);
 	return xfrm6_rcv_spi(skb, spi);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 3692a4783a75..33e982961db0 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1180,11 +1180,10 @@ static void xfrm_replay_timer_handler(unsigned long data)
 	spin_unlock(&x->lock);
 }
 
-int xfrm_replay_check(struct xfrm_state *x, u32 seq)
+int xfrm_replay_check(struct xfrm_state *x, __be32 net_seq)
 {
 	u32 diff;
-
-	seq = ntohl(seq);
+	u32 seq = ntohl(net_seq);
 
 	if (unlikely(seq == 0))
 		return -EINVAL;
-- 
cgit v1.2.3


From 61f4627b2fecce9d5c9645e4b47e75a0c29ad8c3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:48:33 -0700
Subject: [XFRM]: xfrm_replay_advance() annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    | 2 +-
 net/xfrm/xfrm_state.c | 5 ++---
 2 files changed, 3 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index e6110559e505..1e2a4ddec96e 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -936,7 +936,7 @@ extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq);
 extern int xfrm_state_delete(struct xfrm_state *x);
 extern void xfrm_state_flush(u8 proto);
 extern int xfrm_replay_check(struct xfrm_state *x, __be32 seq);
-extern void xfrm_replay_advance(struct xfrm_state *x, u32 seq);
+extern void xfrm_replay_advance(struct xfrm_state *x, __be32 seq);
 extern void xfrm_replay_notify(struct xfrm_state *x, int event);
 extern int xfrm_state_check(struct xfrm_state *x, struct sk_buff *skb);
 extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 33e982961db0..25657f57fbb0 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1205,11 +1205,10 @@ int xfrm_replay_check(struct xfrm_state *x, __be32 net_seq)
 }
 EXPORT_SYMBOL(xfrm_replay_check);
 
-void xfrm_replay_advance(struct xfrm_state *x, u32 seq)
+void xfrm_replay_advance(struct xfrm_state *x, __be32 net_seq)
 {
 	u32 diff;
-
-	seq = ntohl(seq);
+	u32 seq = ntohl(net_seq);
 
 	if (seq > x->replay.seq) {
 		diff = seq - x->replay.seq;
-- 
cgit v1.2.3


From 9916ecb0a6f52f1475fa2f71845227d3c75fb40c Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:48:48 -0700
Subject: [XFRM]: struct xfrm_usersa_id annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/xfrm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h
index c894267ff5dd..430afd058269 100644
--- a/include/linux/xfrm.h
+++ b/include/linux/xfrm.h
@@ -281,7 +281,7 @@ struct xfrm_usersa_info {
 
 struct xfrm_usersa_id {
 	xfrm_address_t			daddr;
-	__u32				spi;
+	__be32				spi;
 	__u16				family;
 	__u8				proto;
 };
-- 
cgit v1.2.3


From 4324a174304d1d826582be5422a976e09d2b1e63 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Sep 2006 18:49:07 -0700
Subject: [XFRM]: fl_ipsec_spi is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/flow.h      | 2 +-
 net/ipv4/xfrm4_policy.c | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/net/flow.h b/include/net/flow.h
index 5700b0ad63c0..ddf5f3ca1720 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -73,7 +73,7 @@ struct flowi {
 			__u8	objname[16]; /* Not zero terminated */
 		} dnports;
 
-		__u32		spi;
+		__be32		spi;
 
 #ifdef CONFIG_IPV6_MIP6
 		struct {
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index eabcd27b1767..7a7a00147e55 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -221,7 +221,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl)
 
 		case IPPROTO_ESP:
 			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
-				u32 *ehdr = (u32 *)xprth;
+				__be32 *ehdr = (__be32 *)xprth;
 
 				fl->fl_ipsec_spi = ehdr[0];
 			}
@@ -229,7 +229,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl)
 
 		case IPPROTO_AH:
 			if (pskb_may_pull(skb, xprth + 8 - skb->data)) {
-				u32 *ah_hdr = (u32*)xprth;
+				__be32 *ah_hdr = (__be32*)xprth;
 
 				fl->fl_ipsec_spi = ah_hdr[1];
 			}
@@ -237,7 +237,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl)
 
 		case IPPROTO_COMP:
 			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
-				u16 *ipcomp_hdr = (u16 *)xprth;
+				__be16 *ipcomp_hdr = (__be16 *)xprth;
 
 				fl->fl_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
 			}
-- 
cgit v1.2.3


From 1b0fee7d68f234be6b270cda51d9fcb71bebd780 Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <samuel@sortiz.org>
Date: Wed, 27 Sep 2006 20:06:44 -0700
Subject: [IrDA]: Memory allocations cleanups

This patch replaces the bunch of arbitrary 64 and 128 bytes alloc_skb() calls
with more accurate allocation sizes.

Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/irda/irlan_common.h | 10 ++++++-
 include/net/irda/irlap_frame.h  | 31 ++++++++++++++++++++--
 include/net/irda/irlmp.h        |  2 +-
 net/irda/af_irda.c              |  3 ++-
 net/irda/ircomm/ircomm_lmp.c    |  4 +--
 net/irda/iriap.c                |  9 ++++---
 net/irda/iriap_event.c          |  2 +-
 net/irda/irlan/irlan_common.c   | 46 +++++++++++++++++++++++++-------
 net/irda/irlan/irlan_provider.c | 12 ++++++---
 net/irda/irlap_frame.c          | 59 ++++++++++++++++++++++-------------------
 net/irda/irlmp.c                |  2 +-
 net/irda/irttp.c                | 14 +++++-----
 12 files changed, 136 insertions(+), 58 deletions(-)

(limited to 'include')

diff --git a/include/net/irda/irlan_common.h b/include/net/irda/irlan_common.h
index 1c73bdbc3eb3..9592c374b41d 100644
--- a/include/net/irda/irlan_common.h
+++ b/include/net/irda/irlan_common.h
@@ -98,7 +98,15 @@
 #define IRLAN_SHORT  1
 #define IRLAN_ARRAY  2
 
-#define IRLAN_MAX_HEADER (TTP_HEADER+LMP_HEADER+LAP_MAX_HEADER)
+/* IrLAN sits on top if IrTTP */
+#define IRLAN_MAX_HEADER (TTP_HEADER+LMP_HEADER)
+/* 1 byte for the command code and 1 byte for the parameter count */
+#define IRLAN_CMD_HEADER 2
+
+#define IRLAN_STRING_PARAMETER_LEN(name, value) (1 + strlen((name)) + 2 \
+						+ strlen ((value)))
+#define IRLAN_BYTE_PARAMETER_LEN(name)          (1 + strlen((name)) + 2 + 1)
+#define IRLAN_SHORT_PARAMETER_LEN(name)         (1 + strlen((name)) + 2 + 2)
 
 /*
  *  IrLAN client
diff --git a/include/net/irda/irlap_frame.h b/include/net/irda/irlap_frame.h
index 3452ae257c84..9dd54a5002b2 100644
--- a/include/net/irda/irlap_frame.h
+++ b/include/net/irda/irlap_frame.h
@@ -74,6 +74,19 @@ struct discovery_t;
 
 #define PF_BIT    0x10 /* Poll/final bit */
 
+/* Some IrLAP field lengths */
+/*
+ * Only baud rate triplet is 4 bytes (PV can be 2 bytes).
+ * All others params (7) are 3 bytes, so that's 7*3 + 1*4 bytes.
+ */
+#define IRLAP_NEGOCIATION_PARAMS_LEN 25
+#define IRLAP_DISCOVERY_INFO_LEN     32
+
+struct disc_frame {
+	__u8 caddr;          /* Connection address */
+	__u8 control;
+} IRDA_PACK;
+
 struct xid_frame {
 	__u8  caddr; /* Connection address */
 	__u8  control;
@@ -95,11 +108,25 @@ struct test_frame {
 struct ua_frame {
 	__u8 caddr;
 	__u8 control;
-
 	__u32 saddr; /* Source device address */
 	__u32 daddr; /* Dest device address */
 } IRDA_PACK;
-	
+
+struct dm_frame {
+	__u8 caddr;          /* Connection address */
+	__u8 control;
+} IRDA_PACK;
+
+struct rd_frame {
+	__u8 caddr;          /* Connection address */
+	__u8 control;
+} IRDA_PACK;
+
+struct rr_frame {
+	__u8 caddr;          /* Connection address */
+	__u8 control;
+} IRDA_PACK;
+
 struct i_frame {
 	__u8 caddr;
 	__u8 control;
diff --git a/include/net/irda/irlmp.h b/include/net/irda/irlmp.h
index 11ecfa58a648..e212b9bc2503 100644
--- a/include/net/irda/irlmp.h
+++ b/include/net/irda/irlmp.h
@@ -48,7 +48,7 @@
 #define DEV_ADDR_ANY  0xffffffff
 
 #define LMP_HEADER          2    /* Dest LSAP + Source LSAP */
-#define LMP_CONTROL_HEADER  4
+#define LMP_CONTROL_HEADER  4    /* LMP_HEADER + opcode + parameter */
 #define LMP_PID_HEADER      1    /* Used by Ultra */
 #define LMP_MAX_HEADER      (LMP_CONTROL_HEADER+LAP_MAX_HEADER)
 
diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c
index 7b7cd5bd2a06..7e1aea89ef05 100644
--- a/net/irda/af_irda.c
+++ b/net/irda/af_irda.c
@@ -309,7 +309,8 @@ static void irda_connect_response(struct irda_sock *self)
 
 	IRDA_ASSERT(self != NULL, return;);
 
-	skb = alloc_skb(64, GFP_ATOMIC);
+	skb = alloc_skb(TTP_MAX_HEADER + TTP_SAR_HEADER,
+			GFP_ATOMIC);
 	if (skb == NULL) {
 		IRDA_DEBUG(0, "%s() Unable to allocate sk_buff!\n",
 			   __FUNCTION__);
diff --git a/net/irda/ircomm/ircomm_lmp.c b/net/irda/ircomm/ircomm_lmp.c
index 959874b6451f..c8e0d89ee11f 100644
--- a/net/irda/ircomm/ircomm_lmp.c
+++ b/net/irda/ircomm/ircomm_lmp.c
@@ -81,7 +81,7 @@ static int ircomm_lmp_connect_response(struct ircomm_cb *self,
 	
 	/* Any userdata supplied? */
 	if (userdata == NULL) {
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC);
 		if (!tx_skb)
 			return -ENOMEM;
 
@@ -115,7 +115,7 @@ static int ircomm_lmp_disconnect_request(struct ircomm_cb *self,
 	IRDA_DEBUG(0, "%s()\n", __FUNCTION__ );
 
         if (!userdata) {
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC);
 		if (!tx_skb)
 			return -ENOMEM;
 		
diff --git a/net/irda/iriap.c b/net/irda/iriap.c
index 61128aa05b40..415cf4eec23b 100644
--- a/net/irda/iriap.c
+++ b/net/irda/iriap.c
@@ -345,10 +345,11 @@ static void iriap_disconnect_request(struct iriap_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IAS_MAGIC, return;);
 
-	tx_skb = alloc_skb(64, GFP_ATOMIC);
+	tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC);
 	if (tx_skb == NULL) {
-		IRDA_DEBUG(0, "%s(), Could not allocate an sk_buff of length %d\n", 
-			__FUNCTION__, 64);
+		IRDA_DEBUG(0,
+			   "%s(), Could not allocate an sk_buff of length %d\n",
+			   __FUNCTION__, LMP_MAX_HEADER);
 		return;
 	}
 
@@ -701,7 +702,7 @@ void iriap_send_ack(struct iriap_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IAS_MAGIC, return;);
 
-	tx_skb = alloc_skb(64, GFP_ATOMIC);
+	tx_skb = alloc_skb(LMP_MAX_HEADER + 1, GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
diff --git a/net/irda/iriap_event.c b/net/irda/iriap_event.c
index da17395df05a..99b18dc7a0b7 100644
--- a/net/irda/iriap_event.c
+++ b/net/irda/iriap_event.c
@@ -365,7 +365,7 @@ static void state_r_disconnect(struct iriap_cb *self, IRIAP_EVENT event,
 
 	switch (event) {
 	case IAP_LM_CONNECT_INDICATION:
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC);
 		if (tx_skb == NULL) {
 			IRDA_WARNING("%s: unable to malloc!\n", __FUNCTION__);
 			return;
diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c
index 7dd0a2fe1d20..9b962f247714 100644
--- a/net/irda/irlan/irlan_common.c
+++ b/net/irda/irlan/irlan_common.c
@@ -636,7 +636,8 @@ void irlan_get_provider_info(struct irlan_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 
-	skb = alloc_skb(64, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER,
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 
@@ -668,7 +669,10 @@ void irlan_open_data_channel(struct irlan_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 	
-	skb = alloc_skb(64, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_STRING_PARAMETER_LEN("MEDIA", "802.3") +
+			IRLAN_STRING_PARAMETER_LEN("ACCESS_TYPE", "DIRECT"),
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 
@@ -704,7 +708,9 @@ void irlan_close_data_channel(struct irlan_cb *self)
 	if (self->client.tsap_ctrl == NULL)
 		return;
 
-	skb = alloc_skb(64, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN"),
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 
@@ -715,7 +721,7 @@ void irlan_close_data_channel(struct irlan_cb *self)
 	
 	/* Build frame */
  	frame[0] = CMD_CLOSE_DATA_CHAN;
-	frame[1] = 0x01; /* Two parameters */
+	frame[1] = 0x01; /* One parameter */
 
 	irlan_insert_byte_param(skb, "DATA_CHAN", self->dtsap_sel_data);
 
@@ -739,7 +745,11 @@ static void irlan_open_unicast_addr(struct irlan_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);	
 	
-	skb = alloc_skb(128, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "DIRECTED") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_MODE", "FILTER"),
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 
@@ -777,7 +787,12 @@ void irlan_set_broadcast_filter(struct irlan_cb *self, int status)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 	
- 	skb = alloc_skb(128, GFP_ATOMIC);
+ 	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "BROADCAST") +
+			/* We may waste one byte here...*/
+			IRLAN_STRING_PARAMETER_LEN("FILTER_MODE", "FILTER"),
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 
@@ -816,7 +831,12 @@ void irlan_set_multicast_filter(struct irlan_cb *self, int status)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 
- 	skb = alloc_skb(128, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "MULTICAST") +
+			/* We may waste one byte here...*/
+			IRLAN_STRING_PARAMETER_LEN("FILTER_MODE", "NONE"),
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 	
@@ -856,7 +876,12 @@ static void irlan_get_unicast_addr(struct irlan_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 	
-	skb = alloc_skb(128, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_BYTE_PARAMETER_LEN("DATA_CHAN") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "DIRECTED") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_OPERATION",
+						   "DYNAMIC"),
+			GFP_ATOMIC);
 	if (!skb)
 		return;
 
@@ -891,7 +916,10 @@ void irlan_get_media_char(struct irlan_cb *self)
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 	
-	skb = alloc_skb(64, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			IRLAN_STRING_PARAMETER_LEN("MEDIA", "802.3"),
+			GFP_ATOMIC);
+
 	if (!skb)
 		return;
 
diff --git a/net/irda/irlan/irlan_provider.c b/net/irda/irlan/irlan_provider.c
index 9c0df86044d7..58efde919667 100644
--- a/net/irda/irlan/irlan_provider.c
+++ b/net/irda/irlan/irlan_provider.c
@@ -296,7 +296,14 @@ void irlan_provider_send_reply(struct irlan_cb *self, int command,
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == IRLAN_MAGIC, return;);
 
-	skb = alloc_skb(128, GFP_ATOMIC);
+	skb = alloc_skb(IRLAN_MAX_HEADER + IRLAN_CMD_HEADER +
+			/* Bigger param length comes from CMD_GET_MEDIA_CHAR */
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "DIRECTED") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "BORADCAST") +
+			IRLAN_STRING_PARAMETER_LEN("FILTER_TYPE", "MULTICAST") +
+			IRLAN_STRING_PARAMETER_LEN("ACCESS_TYPE", "HOSTED"),
+			GFP_ATOMIC);
+
 	if (!skb)
 		return;
 
@@ -354,8 +361,7 @@ void irlan_provider_send_reply(struct irlan_cb *self, int command,
 		} else
 			skb->data[1] = 0x02; /* 2 parameters */
 		irlan_insert_byte_param(skb, "DATA_CHAN", self->stsap_sel_data);
-		irlan_insert_array_param(skb, "RECONNECT_KEY", "LINUX RULES!",
-					 12);
+		irlan_insert_string_param(skb, "RECONNECT_KEY", "LINUX RULES!");
 		break;
 	case CMD_FILTER_OPERATION:
 		irlan_filter_request(self, skb);
diff --git a/net/irda/irlap_frame.c b/net/irda/irlap_frame.c
index ccb983bf0f4a..dba349c832d0 100644
--- a/net/irda/irlap_frame.c
+++ b/net/irda/irlap_frame.c
@@ -117,7 +117,9 @@ void irlap_send_snrm_frame(struct irlap_cb *self, struct qos_info *qos)
 	IRDA_ASSERT(self->magic == LAP_MAGIC, return;);
 
 	/* Allocate frame */
-	tx_skb = alloc_skb(64, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct snrm_frame) +
+			   IRLAP_NEGOCIATION_PARAMS_LEN,
+			   GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
@@ -136,7 +138,7 @@ void irlap_send_snrm_frame(struct irlap_cb *self, struct qos_info *qos)
 	 *  If we are establishing a connection then insert QoS paramerters
 	 */
 	if (qos) {
-		skb_put(tx_skb, 9); /* 21 left */
+		skb_put(tx_skb, 9); /* 25 left */
 		frame->saddr = cpu_to_le32(self->saddr);
 		frame->daddr = cpu_to_le32(self->daddr);
 
@@ -210,7 +212,9 @@ void irlap_send_ua_response_frame(struct irlap_cb *self, struct qos_info *qos)
 	IRDA_ASSERT(self->magic == LAP_MAGIC, return;);
 
 	/* Allocate frame */
-	tx_skb = alloc_skb(64, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct ua_frame) +
+			   IRLAP_NEGOCIATION_PARAMS_LEN,
+			   GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
@@ -245,23 +249,23 @@ void irlap_send_ua_response_frame(struct irlap_cb *self, struct qos_info *qos)
 void irlap_send_dm_frame( struct irlap_cb *self)
 {
 	struct sk_buff *tx_skb = NULL;
-	__u8 *frame;
+	struct dm_frame *frame;
 
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == LAP_MAGIC, return;);
 
-	tx_skb = alloc_skb(32, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct dm_frame), GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
-	frame = skb_put(tx_skb, 2);
+	frame = (struct dm_frame *)skb_put(tx_skb, 2);
 
 	if (self->state == LAP_NDM)
-		frame[0] = CBROADCAST;
+		frame->caddr = CBROADCAST;
 	else
-		frame[0] = self->caddr;
+		frame->caddr = self->caddr;
 
-	frame[1] = DM_RSP | PF_BIT;
+	frame->control = DM_RSP | PF_BIT;
 
 	irlap_queue_xmit(self, tx_skb);
 }
@@ -275,21 +279,21 @@ void irlap_send_dm_frame( struct irlap_cb *self)
 void irlap_send_disc_frame(struct irlap_cb *self)
 {
 	struct sk_buff *tx_skb = NULL;
-	__u8 *frame;
+	struct disc_frame *frame;
 
 	IRDA_DEBUG(3, "%s()\n", __FUNCTION__);
 
 	IRDA_ASSERT(self != NULL, return;);
 	IRDA_ASSERT(self->magic == LAP_MAGIC, return;);
 
-	tx_skb = alloc_skb(16, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct disc_frame), GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
-	frame = skb_put(tx_skb, 2);
+	frame = (struct disc_frame *)skb_put(tx_skb, 2);
 
-	frame[0] = self->caddr | CMD_FRAME;
-	frame[1] = DISC_CMD | PF_BIT;
+	frame->caddr = self->caddr | CMD_FRAME;
+	frame->control = DISC_CMD | PF_BIT;
 
 	irlap_queue_xmit(self, tx_skb);
 }
@@ -315,7 +319,8 @@ void irlap_send_discovery_xid_frame(struct irlap_cb *self, int S, __u8 s,
 	IRDA_ASSERT(self->magic == LAP_MAGIC, return;);
 	IRDA_ASSERT(discovery != NULL, return;);
 
-	tx_skb = alloc_skb(64, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct xid_frame) + IRLAP_DISCOVERY_INFO_LEN,
+			   GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
@@ -573,18 +578,18 @@ static void irlap_recv_discovery_xid_cmd(struct irlap_cb *self,
 void irlap_send_rr_frame(struct irlap_cb *self, int command)
 {
 	struct sk_buff *tx_skb;
-	__u8 *frame;
+	struct rr_frame *frame;
 
-	tx_skb = alloc_skb(16, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct rr_frame), GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
-	frame = skb_put(tx_skb, 2);
+	frame = (struct rr_frame *)skb_put(tx_skb, 2);
 
-	frame[0] = self->caddr;
-	frame[0] |= (command) ? CMD_FRAME : 0;
+	frame->caddr = self->caddr;
+	frame->caddr |= (command) ? CMD_FRAME : 0;
 
-	frame[1] = RR | PF_BIT | (self->vr << 5);
+	frame->control = RR | PF_BIT | (self->vr << 5);
 
 	irlap_queue_xmit(self, tx_skb);
 }
@@ -598,16 +603,16 @@ void irlap_send_rr_frame(struct irlap_cb *self, int command)
 void irlap_send_rd_frame(struct irlap_cb *self)
 {
 	struct sk_buff *tx_skb;
-	__u8 *frame;
+	struct rd_frame *frame;
 
-	tx_skb = alloc_skb(16, GFP_ATOMIC);
+	tx_skb = alloc_skb(sizeof(struct rd_frame), GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
-	frame = skb_put(tx_skb, 2);
+	frame = (struct rd_frame *)skb_put(tx_skb, 2);
 
-	frame[0] = self->caddr;
-	frame[1] = RD_RSP | PF_BIT;
+	frame->caddr = self->caddr;
+	frame->caddr = RD_RSP | PF_BIT;
 
 	irlap_queue_xmit(self, tx_skb);
 }
@@ -1214,7 +1219,7 @@ void irlap_send_test_frame(struct irlap_cb *self, __u8 caddr, __u32 daddr,
 	struct test_frame *frame;
 	__u8 *info;
 
-	tx_skb = alloc_skb(cmd->len+sizeof(struct test_frame), GFP_ATOMIC);
+	tx_skb = alloc_skb(cmd->len + sizeof(struct test_frame), GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
diff --git a/net/irda/irlmp.c b/net/irda/irlmp.c
index c440913dee14..5073261b9d0c 100644
--- a/net/irda/irlmp.c
+++ b/net/irda/irlmp.c
@@ -392,7 +392,7 @@ int irlmp_connect_request(struct lsap_cb *self, __u8 dlsap_sel,
 
 	/* Any userdata? */
 	if (tx_skb == NULL) {
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC);
 		if (!tx_skb)
 			return -ENOMEM;
 
diff --git a/net/irda/irttp.c b/net/irda/irttp.c
index 42acf1cde737..3c2e70b77df1 100644
--- a/net/irda/irttp.c
+++ b/net/irda/irttp.c
@@ -804,12 +804,12 @@ static inline void irttp_give_credit(struct tsap_cb *self)
 		   self->send_credit, self->avail_credit, self->remote_credit);
 
 	/* Give credit to peer */
-	tx_skb = alloc_skb(64, GFP_ATOMIC);
+	tx_skb = alloc_skb(TTP_MAX_HEADER, GFP_ATOMIC);
 	if (!tx_skb)
 		return;
 
 	/* Reserve space for LMP, and LAP header */
-	skb_reserve(tx_skb, self->max_header_size);
+	skb_reserve(tx_skb, LMP_MAX_HEADER);
 
 	/*
 	 *  Since we can transmit and receive frames concurrently,
@@ -1093,7 +1093,8 @@ int irttp_connect_request(struct tsap_cb *self, __u8 dtsap_sel,
 
 	/* Any userdata supplied? */
 	if (userdata == NULL) {
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(TTP_MAX_HEADER + TTP_SAR_HEADER,
+				   GFP_ATOMIC);
 		if (!tx_skb)
 			return -ENOMEM;
 
@@ -1341,7 +1342,8 @@ int irttp_connect_response(struct tsap_cb *self, __u32 max_sdu_size,
 
 	/* Any userdata supplied? */
 	if (userdata == NULL) {
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(TTP_MAX_HEADER + TTP_SAR_HEADER,
+				   GFP_ATOMIC);
 		if (!tx_skb)
 			return -ENOMEM;
 
@@ -1540,14 +1542,14 @@ int irttp_disconnect_request(struct tsap_cb *self, struct sk_buff *userdata,
 
 	if (!userdata) {
 		struct sk_buff *tx_skb;
-		tx_skb = alloc_skb(64, GFP_ATOMIC);
+		tx_skb = alloc_skb(LMP_MAX_HEADER, GFP_ATOMIC);
 		if (!tx_skb)
 			return -ENOMEM;
 
 		/*
 		 *  Reserve space for MUX and LAP header
 		 */
-		skb_reserve(tx_skb, TTP_MAX_HEADER);
+		skb_reserve(tx_skb, LMP_MAX_HEADER);
 
 		userdata = tx_skb;
 	}
-- 
cgit v1.2.3


From cbde1668e4f08e0a150207646010bc65e1e2a42b Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 27 Sep 2006 22:40:19 -0700
Subject: [NET]: Move netlink interface bits to linux/if_link.h.

Moving netlink interface bits to linux/if.h is rather troublesome for
applications including both linux/if.h (which was changed to be included
from linux/rtnetlink.h automatically) and net/if.h.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if.h        | 130 --------------------------------------------
 include/linux/if_link.h   | 136 ++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/rtnetlink.h |   2 +-
 3 files changed, 137 insertions(+), 131 deletions(-)
 create mode 100644 include/linux/if_link.h

(limited to 'include')

diff --git a/include/linux/if.h b/include/linux/if.h
index 8018c2e22c0c..32bf419351f1 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -214,134 +214,4 @@ struct ifconf
 #define	ifc_buf	ifc_ifcu.ifcu_buf		/* buffer address	*/
 #define	ifc_req	ifc_ifcu.ifcu_req		/* array of structures	*/
 
-/* The struct should be in sync with struct net_device_stats */
-struct rtnl_link_stats
-{
-	__u32	rx_packets;		/* total packets received	*/
-	__u32	tx_packets;		/* total packets transmitted	*/
-	__u32	rx_bytes;		/* total bytes received 	*/
-	__u32	tx_bytes;		/* total bytes transmitted	*/
-	__u32	rx_errors;		/* bad packets received		*/
-	__u32	tx_errors;		/* packet transmit problems	*/
-	__u32	rx_dropped;		/* no space in linux buffers	*/
-	__u32	tx_dropped;		/* no space available in linux	*/
-	__u32	multicast;		/* multicast packets received	*/
-	__u32	collisions;
-
-	/* detailed rx_errors: */
-	__u32	rx_length_errors;
-	__u32	rx_over_errors;		/* receiver ring buff overflow	*/
-	__u32	rx_crc_errors;		/* recved pkt with crc error	*/
-	__u32	rx_frame_errors;	/* recv'd frame alignment error */
-	__u32	rx_fifo_errors;		/* recv'r fifo overrun		*/
-	__u32	rx_missed_errors;	/* receiver missed packet	*/
-
-	/* detailed tx_errors */
-	__u32	tx_aborted_errors;
-	__u32	tx_carrier_errors;
-	__u32	tx_fifo_errors;
-	__u32	tx_heartbeat_errors;
-	__u32	tx_window_errors;
-
-	/* for cslip etc */
-	__u32	rx_compressed;
-	__u32	tx_compressed;
-};
-
-/* The struct should be in sync with struct ifmap */
-struct rtnl_link_ifmap
-{
-	__u64	mem_start;
-	__u64	mem_end;
-	__u64	base_addr;
-	__u16	irq;
-	__u8	dma;
-	__u8	port;
-};
-
-enum
-{
-	IFLA_UNSPEC,
-	IFLA_ADDRESS,
-	IFLA_BROADCAST,
-	IFLA_IFNAME,
-	IFLA_MTU,
-	IFLA_LINK,
-	IFLA_QDISC,
-	IFLA_STATS,
-	IFLA_COST,
-#define IFLA_COST IFLA_COST
-	IFLA_PRIORITY,
-#define IFLA_PRIORITY IFLA_PRIORITY
-	IFLA_MASTER,
-#define IFLA_MASTER IFLA_MASTER
-	IFLA_WIRELESS,		/* Wireless Extension event - see wireless.h */
-#define IFLA_WIRELESS IFLA_WIRELESS
-	IFLA_PROTINFO,		/* Protocol specific information for a link */
-#define IFLA_PROTINFO IFLA_PROTINFO
-	IFLA_TXQLEN,
-#define IFLA_TXQLEN IFLA_TXQLEN
-	IFLA_MAP,
-#define IFLA_MAP IFLA_MAP
-	IFLA_WEIGHT,
-#define IFLA_WEIGHT IFLA_WEIGHT
-	IFLA_OPERSTATE,
-	IFLA_LINKMODE,
-	__IFLA_MAX
-};
-
-
-#define IFLA_MAX (__IFLA_MAX - 1)
-
-/* ifi_flags.
-
-   IFF_* flags.
-
-   The only change is:
-   IFF_LOOPBACK, IFF_BROADCAST and IFF_POINTOPOINT are
-   more not changeable by user. They describe link media
-   characteristics and set by device driver.
-
-   Comments:
-   - Combination IFF_BROADCAST|IFF_POINTOPOINT is invalid
-   - If neither of these three flags are set;
-     the interface is NBMA.
-
-   - IFF_MULTICAST does not mean anything special:
-   multicasts can be used on all not-NBMA links.
-   IFF_MULTICAST means that this media uses special encapsulation
-   for multicast frames. Apparently, all IFF_POINTOPOINT and
-   IFF_BROADCAST devices are able to use multicasts too.
- */
-
-/* IFLA_LINK.
-   For usual devices it is equal ifi_index.
-   If it is a "virtual interface" (f.e. tunnel), ifi_link
-   can point to real physical interface (f.e. for bandwidth calculations),
-   or maybe 0, what means, that real media is unknown (usual
-   for IPIP tunnels, when route to endpoint is allowed to change)
- */
-
-/* Subtype attributes for IFLA_PROTINFO */
-enum
-{
-	IFLA_INET6_UNSPEC,
-	IFLA_INET6_FLAGS,	/* link flags			*/
-	IFLA_INET6_CONF,	/* sysctl parameters		*/
-	IFLA_INET6_STATS,	/* statistics			*/
-	IFLA_INET6_MCAST,	/* MC things. What of them?	*/
-	IFLA_INET6_CACHEINFO,	/* time values and max reasm size */
-	__IFLA_INET6_MAX
-};
-
-#define IFLA_INET6_MAX	(__IFLA_INET6_MAX - 1)
-
-struct ifla_cacheinfo
-{
-	__u32	max_reasm_len;
-	__u32	tstamp;		/* ipv6InterfaceTable updated timestamp */
-	__u32	reachable_time;
-	__u32	retrans_time;
-};
-
 #endif /* _LINUX_IF_H */
diff --git a/include/linux/if_link.h b/include/linux/if_link.h
new file mode 100644
index 000000000000..e963a077e6f5
--- /dev/null
+++ b/include/linux/if_link.h
@@ -0,0 +1,136 @@
+#ifndef _LINUX_IF_LINK_H
+#define _LINUX_IF_LINK_H
+
+#include <linux/netlink.h>
+
+/* The struct should be in sync with struct net_device_stats */
+struct rtnl_link_stats
+{
+	__u32	rx_packets;		/* total packets received	*/
+	__u32	tx_packets;		/* total packets transmitted	*/
+	__u32	rx_bytes;		/* total bytes received 	*/
+	__u32	tx_bytes;		/* total bytes transmitted	*/
+	__u32	rx_errors;		/* bad packets received		*/
+	__u32	tx_errors;		/* packet transmit problems	*/
+	__u32	rx_dropped;		/* no space in linux buffers	*/
+	__u32	tx_dropped;		/* no space available in linux	*/
+	__u32	multicast;		/* multicast packets received	*/
+	__u32	collisions;
+
+	/* detailed rx_errors: */
+	__u32	rx_length_errors;
+	__u32	rx_over_errors;		/* receiver ring buff overflow	*/
+	__u32	rx_crc_errors;		/* recved pkt with crc error	*/
+	__u32	rx_frame_errors;	/* recv'd frame alignment error */
+	__u32	rx_fifo_errors;		/* recv'r fifo overrun		*/
+	__u32	rx_missed_errors;	/* receiver missed packet	*/
+
+	/* detailed tx_errors */
+	__u32	tx_aborted_errors;
+	__u32	tx_carrier_errors;
+	__u32	tx_fifo_errors;
+	__u32	tx_heartbeat_errors;
+	__u32	tx_window_errors;
+
+	/* for cslip etc */
+	__u32	rx_compressed;
+	__u32	tx_compressed;
+};
+
+/* The struct should be in sync with struct ifmap */
+struct rtnl_link_ifmap
+{
+	__u64	mem_start;
+	__u64	mem_end;
+	__u64	base_addr;
+	__u16	irq;
+	__u8	dma;
+	__u8	port;
+};
+
+enum
+{
+	IFLA_UNSPEC,
+	IFLA_ADDRESS,
+	IFLA_BROADCAST,
+	IFLA_IFNAME,
+	IFLA_MTU,
+	IFLA_LINK,
+	IFLA_QDISC,
+	IFLA_STATS,
+	IFLA_COST,
+#define IFLA_COST IFLA_COST
+	IFLA_PRIORITY,
+#define IFLA_PRIORITY IFLA_PRIORITY
+	IFLA_MASTER,
+#define IFLA_MASTER IFLA_MASTER
+	IFLA_WIRELESS,		/* Wireless Extension event - see wireless.h */
+#define IFLA_WIRELESS IFLA_WIRELESS
+	IFLA_PROTINFO,		/* Protocol specific information for a link */
+#define IFLA_PROTINFO IFLA_PROTINFO
+	IFLA_TXQLEN,
+#define IFLA_TXQLEN IFLA_TXQLEN
+	IFLA_MAP,
+#define IFLA_MAP IFLA_MAP
+	IFLA_WEIGHT,
+#define IFLA_WEIGHT IFLA_WEIGHT
+	IFLA_OPERSTATE,
+	IFLA_LINKMODE,
+	__IFLA_MAX
+};
+
+
+#define IFLA_MAX (__IFLA_MAX - 1)
+
+/* ifi_flags.
+
+   IFF_* flags.
+
+   The only change is:
+   IFF_LOOPBACK, IFF_BROADCAST and IFF_POINTOPOINT are
+   more not changeable by user. They describe link media
+   characteristics and set by device driver.
+
+   Comments:
+   - Combination IFF_BROADCAST|IFF_POINTOPOINT is invalid
+   - If neither of these three flags are set;
+     the interface is NBMA.
+
+   - IFF_MULTICAST does not mean anything special:
+   multicasts can be used on all not-NBMA links.
+   IFF_MULTICAST means that this media uses special encapsulation
+   for multicast frames. Apparently, all IFF_POINTOPOINT and
+   IFF_BROADCAST devices are able to use multicasts too.
+ */
+
+/* IFLA_LINK.
+   For usual devices it is equal ifi_index.
+   If it is a "virtual interface" (f.e. tunnel), ifi_link
+   can point to real physical interface (f.e. for bandwidth calculations),
+   or maybe 0, what means, that real media is unknown (usual
+   for IPIP tunnels, when route to endpoint is allowed to change)
+ */
+
+/* Subtype attributes for IFLA_PROTINFO */
+enum
+{
+	IFLA_INET6_UNSPEC,
+	IFLA_INET6_FLAGS,	/* link flags			*/
+	IFLA_INET6_CONF,	/* sysctl parameters		*/
+	IFLA_INET6_STATS,	/* statistics			*/
+	IFLA_INET6_MCAST,	/* MC things. What of them?	*/
+	IFLA_INET6_CACHEINFO,	/* time values and max reasm size */
+	__IFLA_INET6_MAX
+};
+
+#define IFLA_INET6_MAX	(__IFLA_INET6_MAX - 1)
+
+struct ifla_cacheinfo
+{
+	__u32	max_reasm_len;
+	__u32	tstamp;		/* ipv6InterfaceTable updated timestamp */
+	__u32	reachable_time;
+	__u32	retrans_time;
+};
+
+#endif /* _LINUX_IF_LINK_H */
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 9c92dc8b9a08..3a18addaed4c 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -2,7 +2,7 @@
 #define __LINUX_RTNETLINK_H
 
 #include <linux/netlink.h>
-#include <linux/if.h>
+#include <linux/if_link.h>
 
 /****
  *		Routing/neighbour discovery messages.
-- 
cgit v1.2.3


From b854d0d218688b30ccea70521d6ea5f3f56d4e12 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Wed, 27 Sep 2006 22:43:05 -0700
Subject: [NET] KBUILD: Add missing entries for new net headers.

Add the following for userspace export by the 'headers_include'
make target: linux/fib_rules.h, linux/if_addr.h, linux/if_link.h,
linux/neighbour.h.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/Kbuild | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 1df2ac30a4d2..f7a52e19b4be 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -58,6 +58,7 @@ header-y += elf-em.h
 header-y += fadvise.h
 header-y += fd.h
 header-y += fdreg.h
+header-y += fib_rules.h
 header-y += ftape-header-segment.h
 header-y += ftape-vendors.h
 header-y += fuse.h
@@ -70,6 +71,7 @@ header-y += hysdn_if.h
 header-y += i2c-dev.h
 header-y += i8k.h
 header-y += icmp.h
+header-y += if_addr.h
 header-y += if_arcnet.h
 header-y += if_arp.h
 header-y += if_bonding.h
@@ -79,6 +81,7 @@ header-y += if_fddi.h
 header-y += if.h
 header-y += if_hippi.h
 header-y += if_infiniband.h
+header-y += if_link.h
 header-y += if_packet.h
 header-y += if_plip.h
 header-y += if_ppp.h
@@ -110,6 +113,7 @@ header-y += mmtimer.h
 header-y += mqueue.h
 header-y += mtio.h
 header-y += ncp_no.h
+header-y += neighbour.h
 header-y += netfilter_arp.h
 header-y += netrom.h
 header-y += nfs2.h
-- 
cgit v1.2.3


From d77072ecfb6d28287d5e2a61d60d87a3a444ac97 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:20:34 -0700
Subject: [NET]: Annotate dst_ops protocol

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 +-
 include/net/dst.h         | 2 +-
 net/core/neighbour.c      | 2 +-
 net/ethernet/eth.c        | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 13d6d4eb8b3a..9264139bd8df 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -187,7 +187,7 @@ struct hh_cache
 {
 	struct hh_cache *hh_next;	/* Next entry			     */
 	atomic_t	hh_refcnt;	/* number of users                   */
-	unsigned short  hh_type;	/* protocol identifier, f.e ETH_P_IP
+	__be16		hh_type;	/* protocol identifier, f.e ETH_P_IP
                                          *  NOTE:  For VLANs, this will be the
                                          *  encapuslated type. --BLG
                                          */
diff --git a/include/net/dst.h b/include/net/dst.h
index a8d825f90305..e156e38e4ac3 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -84,7 +84,7 @@ struct dst_entry
 struct dst_ops
 {
 	unsigned short		family;
-	unsigned short		protocol;
+	__be16			protocol;
 	unsigned		gc_thresh;
 
 	int			(*gc)(void);
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index b6c69e1463e8..8ce8c471d868 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -1079,7 +1079,7 @@ struct neighbour *neigh_event_ns(struct neigh_table *tbl,
 }
 
 static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst,
-			  u16 protocol)
+			  __be16 protocol)
 {
 	struct hh_cache	*hh;
 	struct net_device *dev = dst->dev;
diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 43863933f27f..4bd78c8cfb26 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -223,7 +223,7 @@ static int eth_header_parse(struct sk_buff *skb, unsigned char *haddr)
  */
 int eth_header_cache(struct neighbour *neigh, struct hh_cache *hh)
 {
-	unsigned short type = hh->hh_type;
+	__be16 type = hh->hh_type;
 	struct ethhdr *eth;
 	struct net_device *dev = neigh->dev;
 
-- 
cgit v1.2.3


From 59b8bfd8fd608821e5addc9f4682c7f2424afd8c Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:21:07 -0700
Subject: [NETFILTER]: netfilter misc annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_arp/arp_tables.h | 6 +++---
 include/linux/netfilter_ipv4/ip_queue.h  | 2 +-
 net/ipv4/netfilter.c                     | 4 ++--
 net/ipv4/netfilter/arp_tables.c          | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 149e87c9ab13..44e39b61d9e7 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -46,11 +46,11 @@ struct arpt_arp {
 	struct arpt_devaddr_info tgt_devaddr;
 
 	/* ARP operation code. */
-	u_int16_t arpop, arpop_mask;
+	__be16 arpop, arpop_mask;
 
 	/* ARP hardware address and protocol address format. */
-	u_int16_t arhrd, arhrd_mask;
-	u_int16_t arpro, arpro_mask;
+	__be16 arhrd, arhrd_mask;
+	__be16 arpro, arpro_mask;
 
 	/* The protocol address length is only accepted if it is 4
 	 * so there is no use in offering a way to do filtering on it.
diff --git a/include/linux/netfilter_ipv4/ip_queue.h b/include/linux/netfilter_ipv4/ip_queue.h
index aa08d68c4841..a03507f465f8 100644
--- a/include/linux/netfilter_ipv4/ip_queue.h
+++ b/include/linux/netfilter_ipv4/ip_queue.h
@@ -26,7 +26,7 @@ typedef struct ipq_packet_msg {
 	unsigned int hook;		/* Netfilter hook we rode in on */
 	char indev_name[IFNAMSIZ];	/* Name of incoming interface */
 	char outdev_name[IFNAMSIZ];	/* Name of outgoing interface */
-	unsigned short hw_protocol;	/* Hardware protocol (network order) */
+	__be16 hw_protocol;		/* Hardware protocol (network order) */
 	unsigned short hw_type;		/* Hardware type */
 	unsigned char hw_addrlen;	/* Hardware address length */
 	unsigned char hw_addr[8];	/* Hardware address */
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index f88347de21a9..5ac15379a0cf 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -128,8 +128,8 @@ EXPORT_SYMBOL(ip_nat_decode_session);
  */
 
 struct ip_rt_info {
-	u_int32_t daddr;
-	u_int32_t saddr;
+	__be32 daddr;
+	__be32 saddr;
 	u_int8_t tos;
 };
 
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 85f0d73ebfb4..17e1a687ab45 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -80,7 +80,7 @@ static inline int arp_packet_match(const struct arphdr *arphdr,
 {
 	char *arpptr = (char *)(arphdr + 1);
 	char *src_devaddr, *tgt_devaddr;
-	u32 src_ipaddr, tgt_ipaddr;
+	__be32 src_ipaddr, tgt_ipaddr;
 	int i, ret;
 
 #define FWINV(bool,invflg) ((bool) ^ !!(arpinfo->invflags & invflg))
-- 
cgit v1.2.3


From cdcb71bf964e02e0a22007f5d90ead7bede3b85b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:21:37 -0700
Subject: [NETFILTER]: conntrack annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack.h       |  2 +-
 include/linux/netfilter_ipv4/ip_conntrack_tuple.h | 16 ++---
 net/ipv4/netfilter/ip_conntrack_amanda.c          |  6 +-
 net/ipv4/netfilter/ip_conntrack_core.c            | 12 ++--
 net/ipv4/netfilter/ip_conntrack_ftp.c             |  6 +-
 net/ipv4/netfilter/ip_conntrack_helper_pptp.c     |  4 +-
 net/ipv4/netfilter/ip_conntrack_irc.c             |  5 +-
 net/ipv4/netfilter/ip_conntrack_netbios_ns.c      | 10 +--
 net/ipv4/netfilter/ip_conntrack_netlink.c         | 82 +++++++++++------------
 net/ipv4/netfilter/ip_conntrack_proto_icmp.c      |  4 +-
 net/ipv4/netfilter/ip_conntrack_proto_sctp.c      |  2 +-
 net/ipv4/netfilter/ip_conntrack_proto_tcp.c       |  6 +-
 net/ipv4/netfilter/ip_conntrack_sip.c             | 16 ++---
 net/ipv4/netfilter/ip_conntrack_tftp.c            |  8 +--
 14 files changed, 90 insertions(+), 89 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h
index 51dbec1892c8..64e868034c4a 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack.h
@@ -157,7 +157,7 @@ struct ip_conntrack_expect
 	unsigned int flags;
 
 #ifdef CONFIG_IP_NF_NAT_NEEDED
-	u_int32_t saved_ip;
+	__be32 saved_ip;
 	/* This is the original per-proto part, used to map the
 	 * expected connection the way the recipient expects. */
 	union ip_conntrack_manip_proto saved_proto;
diff --git a/include/linux/netfilter_ipv4/ip_conntrack_tuple.h b/include/linux/netfilter_ipv4/ip_conntrack_tuple.h
index 2fdabdb4c0ef..c228bde74c33 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_tuple.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_tuple.h
@@ -23,13 +23,13 @@ union ip_conntrack_manip_proto
 		__be16 port;
 	} tcp;
 	struct {
-		u_int16_t port;
+		__be16 port;
 	} udp;
 	struct {
-		u_int16_t id;
+		__be16 id;
 	} icmp;
 	struct {
-		u_int16_t port;
+		__be16 port;
 	} sctp;
 	struct {
 		__be16 key;	/* key is 32bit, pptp only uses 16 */
@@ -39,7 +39,7 @@ union ip_conntrack_manip_proto
 /* The manipulable part of the tuple. */
 struct ip_conntrack_manip
 {
-	u_int32_t ip;
+	__be32 ip;
 	union ip_conntrack_manip_proto u;
 };
 
@@ -50,22 +50,22 @@ struct ip_conntrack_tuple
 
 	/* These are the parts of the tuple which are fixed. */
 	struct {
-		u_int32_t ip;
+		__be32 ip;
 		union {
 			/* Add other protocols here. */
 			u_int16_t all;
 
 			struct {
-				u_int16_t port;
+				__be16 port;
 			} tcp;
 			struct {
-				u_int16_t port;
+				__be16 port;
 			} udp;
 			struct {
 				u_int8_t type, code;
 			} icmp;
 			struct {
-				u_int16_t port;
+				__be16 port;
 			} sctp;
 			struct {
 				__be16 key;	/* key is 32bit, 
diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c
index 0a7bd7f04061..6c7383a8e42b 100644
--- a/net/ipv4/netfilter/ip_conntrack_amanda.c
+++ b/net/ipv4/netfilter/ip_conntrack_amanda.c
@@ -155,11 +155,11 @@ static int help(struct sk_buff **pskb,
 		exp->tuple.dst.protonum = IPPROTO_TCP;
 		exp->tuple.dst.u.tcp.port = htons(port);
 
-		exp->mask.src.ip = 0xFFFFFFFF;
+		exp->mask.src.ip = htonl(0xFFFFFFFF);
 		exp->mask.src.u.tcp.port = 0;
-		exp->mask.dst.ip = 0xFFFFFFFF;
+		exp->mask.dst.ip = htonl(0xFFFFFFFF);
 		exp->mask.dst.protonum = 0xFF;
-		exp->mask.dst.u.tcp.port = 0xFFFF;
+		exp->mask.dst.u.tcp.port = htons(0xFFFF);
 
 		if (ip_nat_amanda_hook)
 			ret = ip_nat_amanda_hook(pskb, ctinfo, off - dataoff,
diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c
index c432b3163609..143c4668538b 100644
--- a/net/ipv4/netfilter/ip_conntrack_core.c
+++ b/net/ipv4/netfilter/ip_conntrack_core.c
@@ -149,8 +149,8 @@ static unsigned int ip_conntrack_hash_rnd;
 static u_int32_t __hash_conntrack(const struct ip_conntrack_tuple *tuple,
 			    unsigned int size, unsigned int rnd)
 {
-	return (jhash_3words(tuple->src.ip,
-	                     (tuple->dst.ip ^ tuple->dst.protonum),
+	return (jhash_3words((__force u32)tuple->src.ip,
+	                     ((__force u32)tuple->dst.ip ^ tuple->dst.protonum),
 	                     (tuple->src.u.all | (tuple->dst.u.all << 16)),
 	                     rnd) % size);
 }
@@ -1169,9 +1169,9 @@ void __ip_ct_refresh_acct(struct ip_conntrack *ct,
 int ip_ct_port_tuple_to_nfattr(struct sk_buff *skb,
 			       const struct ip_conntrack_tuple *tuple)
 {
-	NFA_PUT(skb, CTA_PROTO_SRC_PORT, sizeof(u_int16_t),
+	NFA_PUT(skb, CTA_PROTO_SRC_PORT, sizeof(__be16),
 		&tuple->src.u.tcp.port);
-	NFA_PUT(skb, CTA_PROTO_DST_PORT, sizeof(u_int16_t),
+	NFA_PUT(skb, CTA_PROTO_DST_PORT, sizeof(__be16),
 		&tuple->dst.u.tcp.port);
 	return 0;
 
@@ -1186,9 +1186,9 @@ int ip_ct_port_nfattr_to_tuple(struct nfattr *tb[],
 		return -EINVAL;
 
 	t->src.u.tcp.port =
-		*(u_int16_t *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]);
+		*(__be16 *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]);
 	t->dst.u.tcp.port =
-		*(u_int16_t *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]);
+		*(__be16 *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]);
 
 	return 0;
 }
diff --git a/net/ipv4/netfilter/ip_conntrack_ftp.c b/net/ipv4/netfilter/ip_conntrack_ftp.c
index 1d18c863f064..93dcf960662f 100644
--- a/net/ipv4/netfilter/ip_conntrack_ftp.c
+++ b/net/ipv4/netfilter/ip_conntrack_ftp.c
@@ -425,8 +425,8 @@ static int help(struct sk_buff **pskb,
 	exp->tuple.src.u.tcp.port = 0; /* Don't care. */
 	exp->tuple.dst.protonum = IPPROTO_TCP;
 	exp->mask = ((struct ip_conntrack_tuple)
-		{ { 0xFFFFFFFF, { 0 } },
-		  { 0xFFFFFFFF, { .tcp = { 0xFFFF } }, 0xFF }});
+		{ { htonl(0xFFFFFFFF), { 0 } },
+		  { htonl(0xFFFFFFFF), { .tcp = { htons(0xFFFF) } }, 0xFF }});
 
 	exp->expectfn = NULL;
 	exp->flags = 0;
@@ -488,7 +488,7 @@ static int __init ip_conntrack_ftp_init(void)
 	for (i = 0; i < ports_c; i++) {
 		ftp[i].tuple.src.u.tcp.port = htons(ports[i]);
 		ftp[i].tuple.dst.protonum = IPPROTO_TCP;
-		ftp[i].mask.src.u.tcp.port = 0xFFFF;
+		ftp[i].mask.src.u.tcp.port = htons(0xFFFF);
 		ftp[i].mask.dst.protonum = 0xFF;
 		ftp[i].max_expected = 1;
 		ftp[i].timeout = 5 * 60; /* 5 minutes */
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
index fb0aee691721..a2af5e0c7f99 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c
@@ -242,10 +242,10 @@ exp_gre(struct ip_conntrack *ct,
 	exp_orig->tuple.dst.u.gre.key = callid;
 	exp_orig->tuple.dst.protonum = IPPROTO_GRE;
 
-	exp_orig->mask.src.ip = 0xffffffff;
+	exp_orig->mask.src.ip = htonl(0xffffffff);
 	exp_orig->mask.src.u.all = 0;
 	exp_orig->mask.dst.u.gre.key = htons(0xffff);
-	exp_orig->mask.dst.ip = 0xffffffff;
+	exp_orig->mask.dst.ip = htonl(0xffffffff);
 	exp_orig->mask.dst.protonum = 0xff;
 
 	exp_orig->master = ct;
diff --git a/net/ipv4/netfilter/ip_conntrack_irc.c b/net/ipv4/netfilter/ip_conntrack_irc.c
index 44889075f3b2..75f7c3db1619 100644
--- a/net/ipv4/netfilter/ip_conntrack_irc.c
+++ b/net/ipv4/netfilter/ip_conntrack_irc.c
@@ -218,7 +218,8 @@ static int help(struct sk_buff **pskb,
 				    IPPROTO_TCP }});
 			exp->mask = ((struct ip_conntrack_tuple)
 				{ { 0, { 0 } },
-				  { 0xFFFFFFFF, { .tcp = { 0xFFFF } }, 0xFF }});
+				  { htonl(0xFFFFFFFF),
+					{ .tcp = { htons(0xFFFF) } }, 0xFF }});
 			exp->expectfn = NULL;
 			exp->flags = 0;
 			if (ip_nat_irc_hook)
@@ -266,7 +267,7 @@ static int __init ip_conntrack_irc_init(void)
 		hlpr = &irc_helpers[i];
 		hlpr->tuple.src.u.tcp.port = htons(ports[i]);
 		hlpr->tuple.dst.protonum = IPPROTO_TCP;
-		hlpr->mask.src.u.tcp.port = 0xFFFF;
+		hlpr->mask.src.u.tcp.port = htons(0xFFFF);
 		hlpr->mask.dst.protonum = 0xFF;
 		hlpr->max_expected = max_dcc_channels;
 		hlpr->timeout = dcc_timeout;
diff --git a/net/ipv4/netfilter/ip_conntrack_netbios_ns.c b/net/ipv4/netfilter/ip_conntrack_netbios_ns.c
index 4adec47aae32..a1d6a89f64aa 100644
--- a/net/ipv4/netfilter/ip_conntrack_netbios_ns.c
+++ b/net/ipv4/netfilter/ip_conntrack_netbios_ns.c
@@ -78,12 +78,12 @@ static int help(struct sk_buff **pskb,
 		goto out;
 
 	exp->tuple                = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-	exp->tuple.src.u.udp.port = ntohs(NMBD_PORT);
+	exp->tuple.src.u.udp.port = htons(NMBD_PORT);
 
 	exp->mask.src.ip          = mask;
-	exp->mask.src.u.udp.port  = 0xFFFF;
-	exp->mask.dst.ip          = 0xFFFFFFFF;
-	exp->mask.dst.u.udp.port  = 0xFFFF;
+	exp->mask.src.u.udp.port  = htons(0xFFFF);
+	exp->mask.dst.ip          = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.udp.port  = htons(0xFFFF);
 	exp->mask.dst.protonum    = 0xFF;
 
 	exp->expectfn             = NULL;
@@ -115,7 +115,7 @@ static struct ip_conntrack_helper helper = {
 		.src = {
 			.u = {
 				.udp = {
-					.port	= 0xFFFF,
+					.port	= __constant_htons(0xFFFF),
 				}
 			}
 		},
diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c
index 52eddea27e93..53b6dffea6c2 100644
--- a/net/ipv4/netfilter/ip_conntrack_netlink.c
+++ b/net/ipv4/netfilter/ip_conntrack_netlink.c
@@ -78,8 +78,8 @@ ctnetlink_dump_tuples_ip(struct sk_buff *skb,
 {
 	struct nfattr *nest_parms = NFA_NEST(skb, CTA_TUPLE_IP);
 	
-	NFA_PUT(skb, CTA_IP_V4_SRC, sizeof(u_int32_t), &tuple->src.ip);
-	NFA_PUT(skb, CTA_IP_V4_DST, sizeof(u_int32_t), &tuple->dst.ip);
+	NFA_PUT(skb, CTA_IP_V4_SRC, sizeof(__be32), &tuple->src.ip);
+	NFA_PUT(skb, CTA_IP_V4_DST, sizeof(__be32), &tuple->dst.ip);
 
 	NFA_NEST_END(skb, nest_parms);
 
@@ -110,7 +110,7 @@ ctnetlink_dump_tuples(struct sk_buff *skb,
 static inline int
 ctnetlink_dump_status(struct sk_buff *skb, const struct ip_conntrack *ct)
 {
-	u_int32_t status = htonl((u_int32_t) ct->status);
+	__be32 status = htonl((u_int32_t) ct->status);
 	NFA_PUT(skb, CTA_STATUS, sizeof(status), &status);
 	return 0;
 
@@ -122,7 +122,7 @@ static inline int
 ctnetlink_dump_timeout(struct sk_buff *skb, const struct ip_conntrack *ct)
 {
 	long timeout_l = ct->timeout.expires - jiffies;
-	u_int32_t timeout;
+	__be32 timeout;
 
 	if (timeout_l < 0)
 		timeout = 0;
@@ -192,13 +192,13 @@ ctnetlink_dump_counters(struct sk_buff *skb, const struct ip_conntrack *ct,
 {
 	enum ctattr_type type = dir ? CTA_COUNTERS_REPLY: CTA_COUNTERS_ORIG;
 	struct nfattr *nest_count = NFA_NEST(skb, type);
-	u_int32_t tmp;
+	__be32 tmp;
 
 	tmp = htonl(ct->counters[dir].packets);
-	NFA_PUT(skb, CTA_COUNTERS32_PACKETS, sizeof(u_int32_t), &tmp);
+	NFA_PUT(skb, CTA_COUNTERS32_PACKETS, sizeof(__be32), &tmp);
 
 	tmp = htonl(ct->counters[dir].bytes);
-	NFA_PUT(skb, CTA_COUNTERS32_BYTES, sizeof(u_int32_t), &tmp);
+	NFA_PUT(skb, CTA_COUNTERS32_BYTES, sizeof(__be32), &tmp);
 
 	NFA_NEST_END(skb, nest_count);
 
@@ -215,9 +215,9 @@ nfattr_failure:
 static inline int
 ctnetlink_dump_mark(struct sk_buff *skb, const struct ip_conntrack *ct)
 {
-	u_int32_t mark = htonl(ct->mark);
+	__be32 mark = htonl(ct->mark);
 
-	NFA_PUT(skb, CTA_MARK, sizeof(u_int32_t), &mark);
+	NFA_PUT(skb, CTA_MARK, sizeof(__be32), &mark);
 	return 0;
 
 nfattr_failure:
@@ -230,8 +230,8 @@ nfattr_failure:
 static inline int
 ctnetlink_dump_id(struct sk_buff *skb, const struct ip_conntrack *ct)
 {
-	u_int32_t id = htonl(ct->id);
-	NFA_PUT(skb, CTA_ID, sizeof(u_int32_t), &id);
+	__be32 id = htonl(ct->id);
+	NFA_PUT(skb, CTA_ID, sizeof(__be32), &id);
 	return 0;
 
 nfattr_failure:
@@ -241,9 +241,9 @@ nfattr_failure:
 static inline int
 ctnetlink_dump_use(struct sk_buff *skb, const struct ip_conntrack *ct)
 {
-	u_int32_t use = htonl(atomic_read(&ct->ct_general.use));
+	__be32 use = htonl(atomic_read(&ct->ct_general.use));
 	
-	NFA_PUT(skb, CTA_USE, sizeof(u_int32_t), &use);
+	NFA_PUT(skb, CTA_USE, sizeof(__be32), &use);
 	return 0;
 
 nfattr_failure:
@@ -457,8 +457,8 @@ out:
 }
 
 static const size_t cta_min_ip[CTA_IP_MAX] = {
-	[CTA_IP_V4_SRC-1]	= sizeof(u_int32_t),
-	[CTA_IP_V4_DST-1]	= sizeof(u_int32_t),
+	[CTA_IP_V4_SRC-1]	= sizeof(__be32),
+	[CTA_IP_V4_DST-1]	= sizeof(__be32),
 };
 
 static inline int
@@ -475,11 +475,11 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct ip_conntrack_tuple *tuple)
 
 	if (!tb[CTA_IP_V4_SRC-1])
 		return -EINVAL;
-	tuple->src.ip = *(u_int32_t *)NFA_DATA(tb[CTA_IP_V4_SRC-1]);
+	tuple->src.ip = *(__be32 *)NFA_DATA(tb[CTA_IP_V4_SRC-1]);
 
 	if (!tb[CTA_IP_V4_DST-1])
 		return -EINVAL;
-	tuple->dst.ip = *(u_int32_t *)NFA_DATA(tb[CTA_IP_V4_DST-1]);
+	tuple->dst.ip = *(__be32 *)NFA_DATA(tb[CTA_IP_V4_DST-1]);
 
 	DEBUGP("leaving\n");
 
@@ -602,8 +602,8 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr,
 }
 
 static const size_t cta_min_nat[CTA_NAT_MAX] = {
-	[CTA_NAT_MINIP-1]       = sizeof(u_int32_t),
-	[CTA_NAT_MAXIP-1]       = sizeof(u_int32_t),
+	[CTA_NAT_MINIP-1]       = sizeof(__be32),
+	[CTA_NAT_MAXIP-1]       = sizeof(__be32),
 };
 
 static inline int
@@ -623,12 +623,12 @@ ctnetlink_parse_nat(struct nfattr *nat,
 		return -EINVAL;
 
 	if (tb[CTA_NAT_MINIP-1])
-		range->min_ip = *(u_int32_t *)NFA_DATA(tb[CTA_NAT_MINIP-1]);
+		range->min_ip = *(__be32 *)NFA_DATA(tb[CTA_NAT_MINIP-1]);
 
 	if (!tb[CTA_NAT_MAXIP-1])
 		range->max_ip = range->min_ip;
 	else
-		range->max_ip = *(u_int32_t *)NFA_DATA(tb[CTA_NAT_MAXIP-1]);
+		range->max_ip = *(__be32 *)NFA_DATA(tb[CTA_NAT_MAXIP-1]);
 
 	if (range->min_ip)
 		range->flags |= IP_NAT_RANGE_MAP_IPS;
@@ -663,11 +663,11 @@ ctnetlink_parse_help(struct nfattr *attr, char **helper_name)
 }
 
 static const size_t cta_min[CTA_MAX] = {
-	[CTA_STATUS-1] 		= sizeof(u_int32_t),
-	[CTA_TIMEOUT-1] 	= sizeof(u_int32_t),
-	[CTA_MARK-1]		= sizeof(u_int32_t),
-	[CTA_USE-1]		= sizeof(u_int32_t),
-	[CTA_ID-1]		= sizeof(u_int32_t)
+	[CTA_STATUS-1] 		= sizeof(__be32),
+	[CTA_TIMEOUT-1] 	= sizeof(__be32),
+	[CTA_MARK-1]		= sizeof(__be32),
+	[CTA_USE-1]		= sizeof(__be32),
+	[CTA_ID-1]		= sizeof(__be32)
 };
 
 static int
@@ -706,7 +706,7 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb,
 	ct = tuplehash_to_ctrack(h);
 	
 	if (cda[CTA_ID-1]) {
-		u_int32_t id = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_ID-1]));
+		u_int32_t id = ntohl(*(__be32 *)NFA_DATA(cda[CTA_ID-1]));
 		if (ct->id != id) {
 			ip_conntrack_put(ct);
 			return -ENOENT;
@@ -808,7 +808,7 @@ static inline int
 ctnetlink_change_status(struct ip_conntrack *ct, struct nfattr *cda[])
 {
 	unsigned long d;
-	unsigned status = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_STATUS-1]));
+	unsigned status = ntohl(*(__be32 *)NFA_DATA(cda[CTA_STATUS-1]));
 	d = ct->status ^ status;
 
 	if (d & (IPS_EXPECTED|IPS_CONFIRMED|IPS_DYING))
@@ -903,7 +903,7 @@ ctnetlink_change_helper(struct ip_conntrack *ct, struct nfattr *cda[])
 static inline int
 ctnetlink_change_timeout(struct ip_conntrack *ct, struct nfattr *cda[])
 {
-	u_int32_t timeout = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_TIMEOUT-1]));
+	u_int32_t timeout = ntohl(*(__be32 *)NFA_DATA(cda[CTA_TIMEOUT-1]));
 	
 	if (!del_timer(&ct->timeout))
 		return -ETIME;
@@ -966,7 +966,7 @@ ctnetlink_change_conntrack(struct ip_conntrack *ct, struct nfattr *cda[])
 
 #if defined(CONFIG_IP_NF_CONNTRACK_MARK)
 	if (cda[CTA_MARK-1])
-		ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1]));
+		ct->mark = ntohl(*(__be32 *)NFA_DATA(cda[CTA_MARK-1]));
 #endif
 
 	DEBUGP("all done\n");
@@ -989,7 +989,7 @@ ctnetlink_create_conntrack(struct nfattr *cda[],
 
 	if (!cda[CTA_TIMEOUT-1])
 		goto err;
-	ct->timeout.expires = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_TIMEOUT-1]));
+	ct->timeout.expires = ntohl(*(__be32 *)NFA_DATA(cda[CTA_TIMEOUT-1]));
 
 	ct->timeout.expires = jiffies + ct->timeout.expires * HZ;
 	ct->status |= IPS_CONFIRMED;
@@ -1006,7 +1006,7 @@ ctnetlink_create_conntrack(struct nfattr *cda[],
 
 #if defined(CONFIG_IP_NF_CONNTRACK_MARK)
 	if (cda[CTA_MARK-1])
-		ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1]));
+		ct->mark = ntohl(*(__be32 *)NFA_DATA(cda[CTA_MARK-1]));
 #endif
 
 	ct->helper = ip_conntrack_helper_find_get(rtuple);
@@ -1138,8 +1138,8 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb,
                           const struct ip_conntrack_expect *exp)
 {
 	struct ip_conntrack *master = exp->master;
-	u_int32_t timeout = htonl((exp->timeout.expires - jiffies) / HZ);
-	u_int32_t id = htonl(exp->id);
+	__be32 timeout = htonl((exp->timeout.expires - jiffies) / HZ);
+	__be32 id = htonl(exp->id);
 
 	if (ctnetlink_exp_dump_tuple(skb, &exp->tuple, CTA_EXPECT_TUPLE) < 0)
 		goto nfattr_failure;
@@ -1150,8 +1150,8 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb,
 				 CTA_EXPECT_MASTER) < 0)
 		goto nfattr_failure;
 	
-	NFA_PUT(skb, CTA_EXPECT_TIMEOUT, sizeof(timeout), &timeout);
-	NFA_PUT(skb, CTA_EXPECT_ID, sizeof(u_int32_t), &id);
+	NFA_PUT(skb, CTA_EXPECT_TIMEOUT, sizeof(__be32), &timeout);
+	NFA_PUT(skb, CTA_EXPECT_ID, sizeof(__be32), &id);
 
 	return 0;
 	
@@ -1272,8 +1272,8 @@ out:
 }
 
 static const size_t cta_min_exp[CTA_EXPECT_MAX] = {
-	[CTA_EXPECT_TIMEOUT-1]          = sizeof(u_int32_t),
-	[CTA_EXPECT_ID-1]               = sizeof(u_int32_t)
+	[CTA_EXPECT_TIMEOUT-1]          = sizeof(__be32),
+	[CTA_EXPECT_ID-1]               = sizeof(__be32)
 };
 
 static int
@@ -1321,7 +1321,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb,
 		return -ENOENT;
 
 	if (cda[CTA_EXPECT_ID-1]) {
-		u_int32_t id = *(u_int32_t *)NFA_DATA(cda[CTA_EXPECT_ID-1]);
+		__be32 id = *(__be32 *)NFA_DATA(cda[CTA_EXPECT_ID-1]);
 		if (exp->id != ntohl(id)) {
 			ip_conntrack_expect_put(exp);
 			return -ENOENT;
@@ -1375,8 +1375,8 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb,
 			return -ENOENT;
 
 		if (cda[CTA_EXPECT_ID-1]) {
-			u_int32_t id = 
-				*(u_int32_t *)NFA_DATA(cda[CTA_EXPECT_ID-1]);
+			__be32 id =
+				*(__be32 *)NFA_DATA(cda[CTA_EXPECT_ID-1]);
 			if (exp->id != ntohl(id)) {
 				ip_conntrack_expect_put(exp);
 				return -ENOENT;
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_icmp.c b/net/ipv4/netfilter/ip_conntrack_proto_icmp.c
index 09c40ebe3345..295b6fa340db 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_icmp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_icmp.c
@@ -261,7 +261,7 @@ icmp_error(struct sk_buff *skb, enum ip_conntrack_info *ctinfo,
 static int icmp_tuple_to_nfattr(struct sk_buff *skb,
 				const struct ip_conntrack_tuple *t)
 {
-	NFA_PUT(skb, CTA_PROTO_ICMP_ID, sizeof(u_int16_t),
+	NFA_PUT(skb, CTA_PROTO_ICMP_ID, sizeof(__be16),
 		&t->src.u.icmp.id);
 	NFA_PUT(skb, CTA_PROTO_ICMP_TYPE, sizeof(u_int8_t),
 		&t->dst.u.icmp.type);
@@ -287,7 +287,7 @@ static int icmp_nfattr_to_tuple(struct nfattr *tb[],
 	tuple->dst.u.icmp.code =
 			*(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMP_CODE-1]);
 	tuple->src.u.icmp.id =
-			*(u_int16_t *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]);
+			*(__be16 *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]);
 
 	if (tuple->dst.u.icmp.type >= sizeof(invmap)
 	    || !invmap[tuple->dst.u.icmp.type])
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
index b908a4842e18..2443322e4128 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c
@@ -210,7 +210,7 @@ static int sctp_print_conntrack(struct seq_file *s,
 for (offset = skb->nh.iph->ihl * 4 + sizeof(sctp_sctphdr_t), count = 0;	\
 	offset < skb->len &&						\
 	(sch = skb_header_pointer(skb, offset, sizeof(_sch), &_sch));	\
-	offset += (htons(sch->length) + 3) & ~3, count++)
+	offset += (ntohs(sch->length) + 3) & ~3, count++)
 
 /* Some validity checks to make sure the chunks are fine */
 static int do_basic_checks(struct ip_conntrack *conntrack,
diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
index 03ae9a04cb37..06e4e8a6dd9f 100644
--- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c
@@ -519,8 +519,8 @@ static void tcp_sack(const struct sk_buff *skb,
 
 	/* Fast path for timestamp-only option */
 	if (length == TCPOLEN_TSTAMP_ALIGNED*4
-	    && *(__u32 *)ptr ==
-	        __constant_ntohl((TCPOPT_NOP << 24) 
+	    && *(__be32 *)ptr ==
+	        __constant_htonl((TCPOPT_NOP << 24)
 	        		 | (TCPOPT_NOP << 16)
 	        		 | (TCPOPT_TIMESTAMP << 8)
 	        		 | TCPOLEN_TIMESTAMP))
@@ -551,7 +551,7 @@ static void tcp_sack(const struct sk_buff *skb,
 			    	for (i = 0;
 			    	     i < (opsize - TCPOLEN_SACK_BASE);
 			    	     i += TCPOLEN_SACK_PERBLOCK) {
-					tmp = ntohl(*((u_int32_t *)(ptr+i)+1));
+					tmp = ntohl(*((__be32 *)(ptr+i)+1));
 					
 					if (after(tmp, *sack))
 						*sack = tmp;
diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c
index 2893e9c74850..f4f75995a9e4 100644
--- a/net/ipv4/netfilter/ip_conntrack_sip.c
+++ b/net/ipv4/netfilter/ip_conntrack_sip.c
@@ -193,7 +193,7 @@ static int skp_digits_len(const char *dptr, const char *limit, int *shift)
 
 /* Simple ipaddr parser.. */
 static int parse_ipaddr(const char *cp,	const char **endp,
-			u_int32_t *ipaddr, const char *limit)
+			__be32 *ipaddr, const char *limit)
 {
 	unsigned long int val;
 	int i, digit = 0;
@@ -227,7 +227,7 @@ static int parse_ipaddr(const char *cp,	const char **endp,
 static int epaddr_len(const char *dptr, const char *limit, int *shift)
 {
 	const char *aux = dptr;
-	u_int32_t ip;
+	__be32 ip;
 
 	if (parse_ipaddr(dptr, &dptr, &ip, limit) < 0) {
 		DEBUGP("ip: %s parse failed.!\n", dptr);
@@ -302,7 +302,7 @@ int ct_sip_get_info(const char *dptr, size_t dlen,
 static int set_expected_rtp(struct sk_buff **pskb,
 			    struct ip_conntrack *ct,
 			    enum ip_conntrack_info ctinfo,
-			    u_int32_t ipaddr, u_int16_t port,
+			    __be32 ipaddr, u_int16_t port,
 			    const char *dptr)
 {
 	struct ip_conntrack_expect *exp;
@@ -319,10 +319,10 @@ static int set_expected_rtp(struct sk_buff **pskb,
 	exp->tuple.dst.u.udp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_UDP;
 
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.udp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.udp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.udp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 
 	exp->expectfn = NULL;
@@ -349,7 +349,7 @@ static int sip_help(struct sk_buff **pskb,
 	const char *dptr;
 	int ret = NF_ACCEPT;
 	int matchoff, matchlen;
-	u_int32_t ipaddr;
+	__be32 ipaddr;
 	u_int16_t port;
 
 	/* No Data ? */
@@ -439,7 +439,7 @@ static int __init init(void)
 
 		sip[i].tuple.dst.protonum = IPPROTO_UDP;
 		sip[i].tuple.src.u.udp.port = htons(ports[i]);
-		sip[i].mask.src.u.udp.port = 0xFFFF;
+		sip[i].mask.src.u.udp.port = htons(0xFFFF);
 		sip[i].mask.dst.protonum = 0xFF;
 		sip[i].max_expected = 2;
 		sip[i].timeout = 3 * 60; /* 3 minutes */
diff --git a/net/ipv4/netfilter/ip_conntrack_tftp.c b/net/ipv4/netfilter/ip_conntrack_tftp.c
index 7e33d3bed5e3..fe0b634dd377 100644
--- a/net/ipv4/netfilter/ip_conntrack_tftp.c
+++ b/net/ipv4/netfilter/ip_conntrack_tftp.c
@@ -70,10 +70,10 @@ static int tftp_help(struct sk_buff **pskb,
 			return NF_DROP;
 
 		exp->tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		exp->mask.src.ip = 0xffffffff;
+		exp->mask.src.ip = htonl(0xffffffff);
 		exp->mask.src.u.udp.port = 0;
-		exp->mask.dst.ip = 0xffffffff;
-		exp->mask.dst.u.udp.port = 0xffff;
+		exp->mask.dst.ip = htonl(0xffffffff);
+		exp->mask.dst.u.udp.port = htons(0xffff);
 		exp->mask.dst.protonum = 0xff;
 		exp->expectfn = NULL;
 		exp->flags = 0;
@@ -129,7 +129,7 @@ static int __init ip_conntrack_tftp_init(void)
 		tftp[i].tuple.dst.protonum = IPPROTO_UDP;
 		tftp[i].tuple.src.u.udp.port = htons(ports[i]);
 		tftp[i].mask.dst.protonum = 0xFF;
-		tftp[i].mask.src.u.udp.port = 0xFFFF;
+		tftp[i].mask.src.u.udp.port = htons(0xFFFF);
 		tftp[i].max_expected = 1;
 		tftp[i].timeout = 5 * 60; /* 5 minutes */
 		tftp[i].me = THIS_MODULE;
-- 
cgit v1.2.3


From a76b11dd25957287af12ce6855be6d7fd415b3a9 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:22:02 -0700
Subject: [NETFILTER]: NAT annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_nat.h   |  2 +-
 net/ipv4/netfilter/ip_nat_core.c        | 14 ++++++-------
 net/ipv4/netfilter/ip_nat_ftp.c         | 10 ++++-----
 net/ipv4/netfilter/ip_nat_helper.c      | 37 +++++++++++++++------------------
 net/ipv4/netfilter/ip_nat_helper_pptp.c |  2 +-
 net/ipv4/netfilter/ip_nat_proto_icmp.c  |  2 +-
 net/ipv4/netfilter/ip_nat_proto_tcp.c   | 10 ++++-----
 net/ipv4/netfilter/ip_nat_proto_udp.c   | 10 ++++-----
 net/ipv4/netfilter/ip_nat_rule.c        |  6 +++---
 net/ipv4/netfilter/ip_nat_sip.c         |  8 +++----
 net/ipv4/netfilter/ip_nat_snmp_basic.c  |  2 +-
 net/ipv4/netfilter/ip_nat_standalone.c  |  2 +-
 12 files changed, 51 insertions(+), 54 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_nat.h b/include/linux/netfilter_ipv4/ip_nat.h
index 98f8407e4cb5..bdf553620ca1 100644
--- a/include/linux/netfilter_ipv4/ip_nat.h
+++ b/include/linux/netfilter_ipv4/ip_nat.h
@@ -33,7 +33,7 @@ struct ip_nat_range
 	unsigned int flags;
 
 	/* Inclusive: network order. */
-	u_int32_t min_ip, max_ip;
+	__be32 min_ip, max_ip;
 
 	/* Inclusive: network order */
 	union ip_conntrack_manip_proto min, max;
diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c
index 71f3e09cbc84..4b6260a97408 100644
--- a/net/ipv4/netfilter/ip_nat_core.c
+++ b/net/ipv4/netfilter/ip_nat_core.c
@@ -82,7 +82,7 @@ static inline unsigned int
 hash_by_src(const struct ip_conntrack_tuple *tuple)
 {
 	/* Original src, to ensure we map it consistently if poss. */
-	return jhash_3words(tuple->src.ip, tuple->src.u.all,
+	return jhash_3words((__force u32)tuple->src.ip, tuple->src.u.all,
 			    tuple->dst.protonum, 0) % ip_nat_htable_size;
 }
 
@@ -190,7 +190,7 @@ find_best_ips_proto(struct ip_conntrack_tuple *tuple,
 		    const struct ip_conntrack *conntrack,
 		    enum ip_nat_manip_type maniptype)
 {
-	u_int32_t *var_ipp;
+	__be32 *var_ipp;
 	/* Host order */
 	u_int32_t minip, maxip, j;
 
@@ -217,7 +217,7 @@ find_best_ips_proto(struct ip_conntrack_tuple *tuple,
 	 * like this), even across reboots. */
 	minip = ntohl(range->min_ip);
 	maxip = ntohl(range->max_ip);
-	j = jhash_2words(tuple->src.ip, tuple->dst.ip, 0);
+	j = jhash_2words((__force u32)tuple->src.ip, (__force u32)tuple->dst.ip, 0);
 	*var_ipp = htonl(minip + j % (maxip - minip + 1));
 }
 
@@ -534,9 +534,9 @@ int
 ip_nat_port_range_to_nfattr(struct sk_buff *skb, 
 			    const struct ip_nat_range *range)
 {
-	NFA_PUT(skb, CTA_PROTONAT_PORT_MIN, sizeof(u_int16_t),
+	NFA_PUT(skb, CTA_PROTONAT_PORT_MIN, sizeof(__be16),
 		&range->min.tcp.port);
-	NFA_PUT(skb, CTA_PROTONAT_PORT_MAX, sizeof(u_int16_t),
+	NFA_PUT(skb, CTA_PROTONAT_PORT_MAX, sizeof(__be16),
 		&range->max.tcp.port);
 
 	return 0;
@@ -555,7 +555,7 @@ ip_nat_port_nfattr_to_range(struct nfattr *tb[], struct ip_nat_range *range)
 	if (tb[CTA_PROTONAT_PORT_MIN-1]) {
 		ret = 1;
 		range->min.tcp.port = 
-			*(u_int16_t *)NFA_DATA(tb[CTA_PROTONAT_PORT_MIN-1]);
+			*(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MIN-1]);
 	}
 	
 	if (!tb[CTA_PROTONAT_PORT_MAX-1]) {
@@ -564,7 +564,7 @@ ip_nat_port_nfattr_to_range(struct nfattr *tb[], struct ip_nat_range *range)
 	} else {
 		ret = 1;
 		range->max.tcp.port = 
-			*(u_int16_t *)NFA_DATA(tb[CTA_PROTONAT_PORT_MAX-1]);
+			*(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MAX-1]);
 	}
 
 	return ret;
diff --git a/net/ipv4/netfilter/ip_nat_ftp.c b/net/ipv4/netfilter/ip_nat_ftp.c
index 3328fc5c5f50..a71c233d8112 100644
--- a/net/ipv4/netfilter/ip_nat_ftp.c
+++ b/net/ipv4/netfilter/ip_nat_ftp.c
@@ -34,7 +34,7 @@ MODULE_DESCRIPTION("ftp NAT helper");
 
 static int
 mangle_rfc959_packet(struct sk_buff **pskb,
-		     u_int32_t newip,
+		     __be32 newip,
 		     u_int16_t port,
 		     unsigned int matchoff,
 		     unsigned int matchlen,
@@ -57,7 +57,7 @@ mangle_rfc959_packet(struct sk_buff **pskb,
 /* |1|132.235.1.2|6275| */
 static int
 mangle_eprt_packet(struct sk_buff **pskb,
-		   u_int32_t newip,
+		   __be32 newip,
 		   u_int16_t port,
 		   unsigned int matchoff,
 		   unsigned int matchlen,
@@ -79,7 +79,7 @@ mangle_eprt_packet(struct sk_buff **pskb,
 /* |1|132.235.1.2|6275| */
 static int
 mangle_epsv_packet(struct sk_buff **pskb,
-		   u_int32_t newip,
+		   __be32 newip,
 		   u_int16_t port,
 		   unsigned int matchoff,
 		   unsigned int matchlen,
@@ -98,7 +98,7 @@ mangle_epsv_packet(struct sk_buff **pskb,
 					matchlen, buffer, strlen(buffer));
 }
 
-static int (*mangle[])(struct sk_buff **, u_int32_t, u_int16_t,
+static int (*mangle[])(struct sk_buff **, __be32, u_int16_t,
 		     unsigned int,
 		     unsigned int,
 		     struct ip_conntrack *,
@@ -120,7 +120,7 @@ static unsigned int ip_nat_ftp(struct sk_buff **pskb,
 			       struct ip_conntrack_expect *exp,
 			       u32 *seq)
 {
-	u_int32_t newip;
+	__be32 newip;
 	u_int16_t port;
 	int dir = CTINFO2DIR(ctinfo);
 	struct ip_conntrack *ct = exp->master;
diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c
index e9c5187ea5b2..3bf858480558 100644
--- a/net/ipv4/netfilter/ip_nat_helper.c
+++ b/net/ipv4/netfilter/ip_nat_helper.c
@@ -189,7 +189,7 @@ ip_nat_mangle_tcp_packet(struct sk_buff **pskb,
 					   		datalen, 0));
 	} else
 		tcph->check = nf_proto_csum_update(*pskb,
-						   htons(oldlen) ^ 0xFFFF,
+						   htons(oldlen) ^ htons(0xFFFF),
 						   htons(datalen),
 						   tcph->check, 1);
 
@@ -267,7 +267,7 @@ ip_nat_mangle_udp_packet(struct sk_buff **pskb,
 			udph->check = -1;
 	} else
 		udph->check = nf_proto_csum_update(*pskb,
-						   htons(oldlen) ^ 0xFFFF,
+						   htons(oldlen) ^ htons(0xFFFF),
 						   htons(datalen),
 						   udph->check, 1);
 	return 1;
@@ -284,26 +284,24 @@ sack_adjust(struct sk_buff *skb,
 {
 	while (sackoff < sackend) {
 		struct tcp_sack_block_wire *sack;
-		u_int32_t new_start_seq, new_end_seq;
+		__be32 new_start_seq, new_end_seq;
 
 		sack = (void *)skb->data + sackoff;
 		if (after(ntohl(sack->start_seq) - natseq->offset_before,
 			  natseq->correction_pos))
-			new_start_seq = ntohl(sack->start_seq) 
-					- natseq->offset_after;
+			new_start_seq = htonl(ntohl(sack->start_seq)
+					- natseq->offset_after);
 		else
-			new_start_seq = ntohl(sack->start_seq) 
-					- natseq->offset_before;
-		new_start_seq = htonl(new_start_seq);
+			new_start_seq = htonl(ntohl(sack->start_seq)
+					- natseq->offset_before);
 
 		if (after(ntohl(sack->end_seq) - natseq->offset_before,
 			  natseq->correction_pos))
-			new_end_seq = ntohl(sack->end_seq)
-				      - natseq->offset_after;
+			new_end_seq = htonl(ntohl(sack->end_seq)
+				      - natseq->offset_after);
 		else
-			new_end_seq = ntohl(sack->end_seq)
-				      - natseq->offset_before;
-		new_end_seq = htonl(new_end_seq);
+			new_end_seq = htonl(ntohl(sack->end_seq)
+				      - natseq->offset_before);
 
 		DEBUGP("sack_adjust: start_seq: %d->%d, end_seq: %d->%d\n",
 			ntohl(sack->start_seq), new_start_seq,
@@ -375,7 +373,8 @@ ip_nat_seq_adjust(struct sk_buff **pskb,
 		  enum ip_conntrack_info ctinfo)
 {
 	struct tcphdr *tcph;
-	int dir, newseq, newack;
+	int dir;
+	__be32 newseq, newack;
 	struct ip_nat_seq *this_way, *other_way;	
 
 	dir = CTINFO2DIR(ctinfo);
@@ -388,17 +387,15 @@ ip_nat_seq_adjust(struct sk_buff **pskb,
 
 	tcph = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4;
 	if (after(ntohl(tcph->seq), this_way->correction_pos))
-		newseq = ntohl(tcph->seq) + this_way->offset_after;
+		newseq = htonl(ntohl(tcph->seq) + this_way->offset_after);
 	else
-		newseq = ntohl(tcph->seq) + this_way->offset_before;
-	newseq = htonl(newseq);
+		newseq = htonl(ntohl(tcph->seq) + this_way->offset_before);
 
 	if (after(ntohl(tcph->ack_seq) - other_way->offset_before,
 		  other_way->correction_pos))
-		newack = ntohl(tcph->ack_seq) - other_way->offset_after;
+		newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_after);
 	else
-		newack = ntohl(tcph->ack_seq) - other_way->offset_before;
-	newack = htonl(newack);
+		newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before);
 
 	tcph->check = nf_proto_csum_update(*pskb, ~tcph->seq, newseq,
 					   tcph->check, 0);
diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c
index 2ff578807123..329fdcd7d702 100644
--- a/net/ipv4/netfilter/ip_nat_helper_pptp.c
+++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c
@@ -51,7 +51,7 @@
 
 #define IP_NAT_PPTP_VERSION "3.0"
 
-#define REQ_CID(req, off)		(*(u_int16_t *)((char *)(req) + (off)))
+#define REQ_CID(req, off)		(*(__be16 *)((char *)(req) + (off)))
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Harald Welte <laforge@gnumonks.org>");
diff --git a/net/ipv4/netfilter/ip_nat_proto_icmp.c b/net/ipv4/netfilter/ip_nat_proto_icmp.c
index ec50cc295317..3f6efc13ac74 100644
--- a/net/ipv4/netfilter/ip_nat_proto_icmp.c
+++ b/net/ipv4/netfilter/ip_nat_proto_icmp.c
@@ -67,7 +67,7 @@ icmp_manip_pkt(struct sk_buff **pskb,
 
 	hdr = (struct icmphdr *)((*pskb)->data + hdroff);
 	hdr->checksum = nf_proto_csum_update(*pskb,
-					     hdr->un.echo.id ^ 0xFFFF,
+					     hdr->un.echo.id ^ htons(0xFFFF),
 					     tuple->src.u.icmp.id,
 					     hdr->checksum, 0);
 	hdr->un.echo.id = tuple->src.u.icmp.id;
diff --git a/net/ipv4/netfilter/ip_nat_proto_tcp.c b/net/ipv4/netfilter/ip_nat_proto_tcp.c
index 72a6307bd2db..12deb13b93b1 100644
--- a/net/ipv4/netfilter/ip_nat_proto_tcp.c
+++ b/net/ipv4/netfilter/ip_nat_proto_tcp.c
@@ -24,7 +24,7 @@ tcp_in_range(const struct ip_conntrack_tuple *tuple,
 	     const union ip_conntrack_manip_proto *min,
 	     const union ip_conntrack_manip_proto *max)
 {
-	u_int16_t port;
+	__be16 port;
 
 	if (maniptype == IP_NAT_MANIP_SRC)
 		port = tuple->src.u.tcp.port;
@@ -42,7 +42,7 @@ tcp_unique_tuple(struct ip_conntrack_tuple *tuple,
 		 const struct ip_conntrack *conntrack)
 {
 	static u_int16_t port;
-	u_int16_t *portptr;
+	__be16 *portptr;
 	unsigned int range_size, min, i;
 
 	if (maniptype == IP_NAT_MANIP_SRC)
@@ -93,8 +93,8 @@ tcp_manip_pkt(struct sk_buff **pskb,
 	struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff);
 	struct tcphdr *hdr;
 	unsigned int hdroff = iphdroff + iph->ihl*4;
-	u32 oldip, newip;
-	u16 *portptr, newport, oldport;
+	__be32 oldip, newip;
+	__be16 *portptr, newport, oldport;
 	int hdrsize = 8; /* TCP connection tracking guarantees this much */
 
 	/* this could be a inner header returned in icmp packet; in such
@@ -130,7 +130,7 @@ tcp_manip_pkt(struct sk_buff **pskb,
 		return 1;
 
 	hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip, hdr->check, 1);
-	hdr->check = nf_proto_csum_update(*pskb, oldport ^ 0xFFFF, newport,
+	hdr->check = nf_proto_csum_update(*pskb, oldport ^ htons(0xFFFF), newport,
 					  hdr->check, 0);
 	return 1;
 }
diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c
index 5da196ae758c..4bbec7730d18 100644
--- a/net/ipv4/netfilter/ip_nat_proto_udp.c
+++ b/net/ipv4/netfilter/ip_nat_proto_udp.c
@@ -24,7 +24,7 @@ udp_in_range(const struct ip_conntrack_tuple *tuple,
 	     const union ip_conntrack_manip_proto *min,
 	     const union ip_conntrack_manip_proto *max)
 {
-	u_int16_t port;
+	__be16 port;
 
 	if (maniptype == IP_NAT_MANIP_SRC)
 		port = tuple->src.u.udp.port;
@@ -42,7 +42,7 @@ udp_unique_tuple(struct ip_conntrack_tuple *tuple,
 		 const struct ip_conntrack *conntrack)
 {
 	static u_int16_t port;
-	u_int16_t *portptr;
+	__be16 *portptr;
 	unsigned int range_size, min, i;
 
 	if (maniptype == IP_NAT_MANIP_SRC)
@@ -91,8 +91,8 @@ udp_manip_pkt(struct sk_buff **pskb,
 	struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff);
 	struct udphdr *hdr;
 	unsigned int hdroff = iphdroff + iph->ihl*4;
-	u32 oldip, newip;
-	u16 *portptr, newport;
+	__be32 oldip, newip;
+	__be16 *portptr, newport;
 
 	if (!skb_make_writable(pskb, hdroff + sizeof(*hdr)))
 		return 0;
@@ -118,7 +118,7 @@ udp_manip_pkt(struct sk_buff **pskb,
 		hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip,
 						  hdr->check, 1);
 		hdr->check = nf_proto_csum_update(*pskb,
-						  *portptr ^ 0xFFFF, newport,
+						  *portptr ^ htons(0xFFFF), newport,
 						  hdr->check, 0);
 		if (!hdr->check)
 			hdr->check = -1;
diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c
index 7b703839aa58..a176aa3031e0 100644
--- a/net/ipv4/netfilter/ip_nat_rule.c
+++ b/net/ipv4/netfilter/ip_nat_rule.c
@@ -119,7 +119,7 @@ static unsigned int ipt_snat_target(struct sk_buff **pskb,
 }
 
 /* Before 2.6.11 we did implicit source NAT if required. Warn about change. */
-static void warn_if_extra_mangle(u32 dstip, u32 srcip)
+static void warn_if_extra_mangle(__be32 dstip, __be32 srcip)
 {
 	static int warned = 0;
 	struct flowi fl = { .nl_u = { .ip4_u = { .daddr = dstip } } };
@@ -205,7 +205,7 @@ alloc_null_binding(struct ip_conntrack *conntrack,
 	   per-proto parts (hence not IP_NAT_RANGE_PROTO_SPECIFIED).
 	   Use reply in case it's already been mangled (eg local packet).
 	*/
-	u_int32_t ip
+	__be32 ip
 		= (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC
 		   ? conntrack->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip
 		   : conntrack->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip);
@@ -222,7 +222,7 @@ alloc_null_binding_confirmed(struct ip_conntrack *conntrack,
                              struct ip_nat_info *info,
                              unsigned int hooknum)
 {
-	u_int32_t ip
+	__be32 ip
 		= (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC
 		   ? conntrack->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip
 		   : conntrack->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip);
diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c
index 6ffba63adca2..71fc2730a007 100644
--- a/net/ipv4/netfilter/ip_nat_sip.c
+++ b/net/ipv4/netfilter/ip_nat_sip.c
@@ -60,8 +60,8 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb,
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 	char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")];
 	unsigned int bufflen, dataoff;
-	u_int32_t ip;
-	u_int16_t port;
+	__be32 ip;
+	__be16 port;
 
 	dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr);
 
@@ -159,7 +159,7 @@ static int mangle_content_len(struct sk_buff **pskb,
 static unsigned int mangle_sdp(struct sk_buff **pskb,
 			       enum ip_conntrack_info ctinfo,
 			       struct ip_conntrack *ct,
-			       u_int32_t newip, u_int16_t port,
+			       __be32 newip, u_int16_t port,
 			       const char *dptr)
 {
 	char buffer[sizeof("nnn.nnn.nnn.nnn")];
@@ -195,7 +195,7 @@ static unsigned int ip_nat_sdp(struct sk_buff **pskb,
 {
 	struct ip_conntrack *ct = exp->master;
 	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
-	u_int32_t newip;
+	__be32 newip;
 	u_int16_t port;
 
 	DEBUGP("ip_nat_sdp():\n");
diff --git a/net/ipv4/netfilter/ip_nat_snmp_basic.c b/net/ipv4/netfilter/ip_nat_snmp_basic.c
index 18b7fbdccb61..168f45fa1898 100644
--- a/net/ipv4/netfilter/ip_nat_snmp_basic.c
+++ b/net/ipv4/netfilter/ip_nat_snmp_basic.c
@@ -1211,7 +1211,7 @@ static int snmp_translate(struct ip_conntrack *ct,
                           struct sk_buff **pskb)
 {
 	struct iphdr *iph = (*pskb)->nh.iph;
-	struct udphdr *udph = (struct udphdr *)((u_int32_t *)iph + iph->ihl);
+	struct udphdr *udph = (struct udphdr *)((__be32 *)iph + iph->ihl);
 	u_int16_t udplen = ntohs(udph->len);
 	u_int16_t paylen = udplen - sizeof(struct udphdr);
 	int dir = CTINFO2DIR(ctinfo);
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index 9c577db62047..021395b67463 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -191,7 +191,7 @@ ip_nat_in(unsigned int hooknum,
           int (*okfn)(struct sk_buff *))
 {
 	unsigned int ret;
-	u_int32_t daddr = (*pskb)->nh.iph->daddr;
+	__be32 daddr = (*pskb)->nh.iph->daddr;
 
 	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
 	if (ret != NF_DROP && ret != NF_STOLEN
-- 
cgit v1.2.3


From 6a19d61472d0802a24493c0d200e88f99ad39cd8 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:22:24 -0700
Subject: [NETFILTER]: ipt annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ipt_iprange.h |  2 +-
 net/ipv4/netfilter/ipt_CLUSTERIP.c         | 14 +++++++-------
 net/ipv4/netfilter/ipt_ECN.c               | 12 ++++++------
 net/ipv4/netfilter/ipt_NETMAP.c            |  2 +-
 net/ipv4/netfilter/ipt_REJECT.c            |  4 ++--
 net/ipv4/netfilter/ipt_SAME.c              |  3 ++-
 net/ipv4/netfilter/ipt_TCPMSS.c            | 17 +++++++++--------
 net/ipv4/netfilter/ipt_TOS.c               |  4 ++--
 net/ipv4/netfilter/ipt_TTL.c               |  4 ++--
 net/ipv4/netfilter/ipt_hashlimit.c         | 16 +++++++++-------
 net/ipv4/netfilter/ipt_recent.c            | 15 +++++++--------
 net/ipv4/netfilter/iptable_mangle.c        |  2 +-
 12 files changed, 49 insertions(+), 46 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ipt_iprange.h b/include/linux/netfilter_ipv4/ipt_iprange.h
index 3ecb3bd63676..34ab0fb736e2 100644
--- a/include/linux/netfilter_ipv4/ipt_iprange.h
+++ b/include/linux/netfilter_ipv4/ipt_iprange.h
@@ -8,7 +8,7 @@
 
 struct ipt_iprange {
 	/* Inclusive: network order. */
-	u_int32_t min_ip, max_ip;
+	__be32 min_ip, max_ip;
 };
 
 struct ipt_iprange_info
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index 41589665fc5d..7a29d6e7baa7 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -52,7 +52,7 @@ struct clusterip_config {
 	atomic_t entries;			/* number of entries/rules
 						 * referencing us */
 
-	u_int32_t clusterip;			/* the IP address */
+	__be32 clusterip;			/* the IP address */
 	u_int8_t clustermac[ETH_ALEN];		/* the MAC address */
 	struct net_device *dev;			/* device */
 	u_int16_t num_total_nodes;		/* total number of nodes */
@@ -119,7 +119,7 @@ clusterip_config_entry_put(struct clusterip_config *c)
 }
 
 static struct clusterip_config *
-__clusterip_config_find(u_int32_t clusterip)
+__clusterip_config_find(__be32 clusterip)
 {
 	struct list_head *pos;
 
@@ -136,7 +136,7 @@ __clusterip_config_find(u_int32_t clusterip)
 }
 
 static inline struct clusterip_config *
-clusterip_config_find_get(u_int32_t clusterip, int entry)
+clusterip_config_find_get(__be32 clusterip, int entry)
 {
 	struct clusterip_config *c;
 
@@ -166,7 +166,7 @@ clusterip_config_init_nodelist(struct clusterip_config *c,
 }
 
 static struct clusterip_config *
-clusterip_config_init(struct ipt_clusterip_tgt_info *i, u_int32_t ip,
+clusterip_config_init(struct ipt_clusterip_tgt_info *i, __be32 ip,
 			struct net_device *dev)
 {
 	struct clusterip_config *c;
@@ -387,7 +387,7 @@ checkentry(const char *tablename,
 		return 0;
 
 	}
-	if (e->ip.dmsk.s_addr != 0xffffffff
+	if (e->ip.dmsk.s_addr != htonl(0xffffffff)
 	    || e->ip.dst.s_addr == 0) {
 		printk(KERN_ERR "CLUSTERIP: Please specify destination IP\n");
 		return 0;
@@ -476,9 +476,9 @@ static struct ipt_target clusterip_tgt = {
 /* hardcoded for 48bit ethernet and 32bit ipv4 addresses */
 struct arp_payload {
 	u_int8_t src_hw[ETH_ALEN];
-	u_int32_t src_ip;
+	__be32 src_ip;
 	u_int8_t dst_hw[ETH_ALEN];
-	u_int32_t dst_ip;
+	__be32 dst_ip;
 } __attribute__ ((packed));
 
 #ifdef CLUSTERIP_DEBUG
diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c
index 23f9c7ebe7eb..12a818a2462f 100644
--- a/net/ipv4/netfilter/ipt_ECN.c
+++ b/net/ipv4/netfilter/ipt_ECN.c
@@ -28,7 +28,7 @@ static inline int
 set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 {
 	struct iphdr *iph = (*pskb)->nh.iph;
-	u_int16_t oldtos;
+	__be16 oldtos;
 
 	if ((iph->tos & IPT_ECN_IP_MASK) != (einfo->ip_ect & IPT_ECN_IP_MASK)) {
 		if (!skb_make_writable(pskb, sizeof(struct iphdr)))
@@ -37,7 +37,7 @@ set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 		oldtos = iph->tos;
 		iph->tos &= ~IPT_ECN_IP_MASK;
 		iph->tos |= (einfo->ip_ect & IPT_ECN_IP_MASK);
-		iph->check = nf_csum_update(oldtos ^ 0xFFFF, iph->tos,
+		iph->check = nf_csum_update(oldtos ^ htons(0xFFFF), iph->tos,
 					    iph->check);
 	} 
 	return 1;
@@ -48,7 +48,7 @@ static inline int
 set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 {
 	struct tcphdr _tcph, *tcph;
-	u_int16_t oldval;
+	__be16 oldval;
 
 	/* Not enought header? */
 	tcph = skb_header_pointer(*pskb, (*pskb)->nh.iph->ihl*4,
@@ -66,15 +66,15 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo)
 		return 0;
 	tcph = (void *)(*pskb)->nh.iph + (*pskb)->nh.iph->ihl*4;
 
-	oldval = ((u_int16_t *)tcph)[6];
+	oldval = ((__be16 *)tcph)[6];
 	if (einfo->operation & IPT_ECN_OP_SET_ECE)
 		tcph->ece = einfo->proto.tcp.ece;
 	if (einfo->operation & IPT_ECN_OP_SET_CWR)
 		tcph->cwr = einfo->proto.tcp.cwr;
 
 	tcph->check = nf_proto_csum_update((*pskb),
-					   oldval ^ 0xFFFF,
-					   ((u_int16_t *)tcph)[6],
+					   oldval ^ htons(0xFFFF),
+					   ((__be16 *)tcph)[6],
 					   tcph->check, 0);
 	return 1;
 }
diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c
index beb2914225ff..58a88f227108 100644
--- a/net/ipv4/netfilter/ipt_NETMAP.c
+++ b/net/ipv4/netfilter/ipt_NETMAP.c
@@ -58,7 +58,7 @@ target(struct sk_buff **pskb,
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
-	u_int32_t new_ip, netmask;
+	__be32 new_ip, netmask;
 	const struct ip_nat_multi_range_compat *mr = targinfo;
 	struct ip_nat_range newrange;
 
diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c
index b81821edd893..fd0c05efed8a 100644
--- a/net/ipv4/netfilter/ipt_REJECT.c
+++ b/net/ipv4/netfilter/ipt_REJECT.c
@@ -104,8 +104,8 @@ static void send_reset(struct sk_buff *oldskb, int hook)
 	struct iphdr *iph = oldskb->nh.iph;
 	struct tcphdr _otcph, *oth, *tcph;
 	struct rtable *rt;
-	u_int16_t tmp_port;
-	u_int32_t tmp_addr;
+	__be16 tmp_port;
+	__be32 tmp_addr;
 	int needs_ack;
 	int hh_len;
 
diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c
index efbcb1198832..b38b13328d73 100644
--- a/net/ipv4/netfilter/ipt_SAME.c
+++ b/net/ipv4/netfilter/ipt_SAME.c
@@ -135,7 +135,8 @@ same_target(struct sk_buff **pskb,
 {
 	struct ip_conntrack *ct;
 	enum ip_conntrack_info ctinfo;
-	u_int32_t tmpip, aindex, new_ip;
+	u_int32_t tmpip, aindex;
+	__be32 new_ip;
 	const struct ipt_same_info *same = targinfo;
 	struct ip_nat_range newrange;
 	const struct ip_conntrack_tuple *t;
diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c
index 4246c4321e5b..108b6b76311f 100644
--- a/net/ipv4/netfilter/ipt_TCPMSS.c
+++ b/net/ipv4/netfilter/ipt_TCPMSS.c
@@ -42,7 +42,8 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	const struct ipt_tcpmss_info *tcpmssinfo = targinfo;
 	struct tcphdr *tcph;
 	struct iphdr *iph;
-	u_int16_t tcplen, newtotlen, oldval, newmss;
+	u_int16_t tcplen, newmss;
+	__be16 newtotlen, oldval;
 	unsigned int i;
 	u_int8_t *opt;
 
@@ -97,7 +98,7 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 			opt[i+3] = (newmss & 0x00ff);
 
 			tcph->check = nf_proto_csum_update(*pskb,
-							   htons(oldmss)^0xFFFF,
+							   htons(oldmss)^htons(0xFFFF),
 							   htons(newmss),
 							   tcph->check, 0);
 			return IPT_CONTINUE;
@@ -126,7 +127,7 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr));
 
 	tcph->check = nf_proto_csum_update(*pskb,
-					   htons(tcplen) ^ 0xFFFF,
+					   htons(tcplen) ^ htons(0xFFFF),
 				           htons(tcplen + TCPOLEN_MSS),
 					   tcph->check, 1);
 	opt[0] = TCPOPT_MSS;
@@ -134,18 +135,18 @@ ipt_tcpmss_target(struct sk_buff **pskb,
 	opt[2] = (newmss & 0xff00) >> 8;
 	opt[3] = (newmss & 0x00ff);
 
-	tcph->check = nf_proto_csum_update(*pskb, ~0, *((u_int32_t *)opt),
+	tcph->check = nf_proto_csum_update(*pskb, htonl(~0), *((__be32 *)opt),
 					   tcph->check, 0);
 
-	oldval = ((u_int16_t *)tcph)[6];
+	oldval = ((__be16 *)tcph)[6];
 	tcph->doff += TCPOLEN_MSS/4;
 	tcph->check = nf_proto_csum_update(*pskb,
-					   oldval ^ 0xFFFF,
-					   ((u_int16_t *)tcph)[6],
+					   oldval ^ htons(0xFFFF),
+					   ((__be16 *)tcph)[6],
 					   tcph->check, 0);
 
 	newtotlen = htons(ntohs(iph->tot_len) + TCPOLEN_MSS);
-	iph->check = nf_csum_update(iph->tot_len ^ 0xFFFF,
+	iph->check = nf_csum_update(iph->tot_len ^ htons(0xFFFF),
 				    newtotlen, iph->check);
 	iph->tot_len = newtotlen;
 	return IPT_CONTINUE;
diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c
index 471a4c438b0a..6b8b14ccc3d3 100644
--- a/net/ipv4/netfilter/ipt_TOS.c
+++ b/net/ipv4/netfilter/ipt_TOS.c
@@ -30,7 +30,7 @@ target(struct sk_buff **pskb,
 {
 	const struct ipt_tos_target_info *tosinfo = targinfo;
 	struct iphdr *iph = (*pskb)->nh.iph;
-	u_int16_t oldtos;
+	__be16 oldtos;
 
 	if ((iph->tos & IPTOS_TOS_MASK) != tosinfo->tos) {
 		if (!skb_make_writable(pskb, sizeof(struct iphdr)))
@@ -38,7 +38,7 @@ target(struct sk_buff **pskb,
 		iph = (*pskb)->nh.iph;
 		oldtos = iph->tos;
 		iph->tos = (iph->tos & IPTOS_PREC_MASK) | tosinfo->tos;
-		iph->check = nf_csum_update(oldtos ^ 0xFFFF, iph->tos,
+		iph->check = nf_csum_update(oldtos ^ htons(0xFFFF), iph->tos,
 					    iph->check);
 	}
 	return IPT_CONTINUE;
diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c
index 96e79cc6d0f2..ac9517d62af0 100644
--- a/net/ipv4/netfilter/ipt_TTL.c
+++ b/net/ipv4/netfilter/ipt_TTL.c
@@ -54,8 +54,8 @@ ipt_ttl_target(struct sk_buff **pskb,
 	}
 
 	if (new_ttl != iph->ttl) {
-		iph->check = nf_csum_update(ntohs((iph->ttl << 8)) ^ 0xFFFF,
-					    ntohs(new_ttl << 8),
+		iph->check = nf_csum_update(htons((iph->ttl << 8)) ^ htons(0xFFFF),
+					    htons(new_ttl << 8),
 					    iph->check);
 		iph->ttl = new_ttl;
 	}
diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c
index 4f73a61aa3dd..33ccdbf8e794 100644
--- a/net/ipv4/netfilter/ipt_hashlimit.c
+++ b/net/ipv4/netfilter/ipt_hashlimit.c
@@ -50,11 +50,11 @@ static struct file_operations dl_file_ops;
 /* hash table crap */
 
 struct dsthash_dst {
-	u_int32_t src_ip;
-	u_int32_t dst_ip;
+	__be32 src_ip;
+	__be32 dst_ip;
 	/* ports have to be consecutive !!! */
-	u_int16_t src_port;
-	u_int16_t dst_port;
+	__be16 src_port;
+	__be16 dst_port;
 };
 
 struct dsthash_ent {
@@ -106,8 +106,10 @@ static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b)
 static inline u_int32_t
 hash_dst(const struct ipt_hashlimit_htable *ht, const struct dsthash_dst *dst)
 {
-	return (jhash_3words(dst->dst_ip, (dst->dst_port<<16 | dst->src_port), 
-			     dst->src_ip, ht->rnd) % ht->cfg.size);
+	return (jhash_3words((__force u32)dst->dst_ip,
+			    ((__force u32)dst->dst_port<<16 |
+			     (__force u32)dst->src_port),
+			     (__force u32)dst->src_ip, ht->rnd) % ht->cfg.size);
 }
 
 static inline struct dsthash_ent *
@@ -406,7 +408,7 @@ hashlimit_match(const struct sk_buff *skb,
 		dst.src_ip = skb->nh.iph->saddr;
 	if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DPT
 	    ||hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SPT) {
-		u_int16_t _ports[2], *ports;
+		__be16 _ports[2], *ports;
 
 		switch (skb->nh.iph->protocol) {
 		case IPPROTO_TCP:
diff --git a/net/ipv4/netfilter/ipt_recent.c b/net/ipv4/netfilter/ipt_recent.c
index 32ae8d7ac506..126db44e71a8 100644
--- a/net/ipv4/netfilter/ipt_recent.c
+++ b/net/ipv4/netfilter/ipt_recent.c
@@ -50,11 +50,10 @@ MODULE_PARM_DESC(ip_list_perms, "permissions on /proc/net/ipt_recent/* files");
 MODULE_PARM_DESC(ip_list_uid,"owner of /proc/net/ipt_recent/* files");
 MODULE_PARM_DESC(ip_list_gid,"owning group of /proc/net/ipt_recent/* files");
 
-
 struct recent_entry {
 	struct list_head	list;
 	struct list_head	lru_list;
-	u_int32_t		addr;
+	__be32			addr;
 	u_int8_t		ttl;
 	u_int8_t		index;
 	u_int16_t		nstamps;
@@ -85,17 +84,17 @@ static struct file_operations	recent_fops;
 static u_int32_t hash_rnd;
 static int hash_rnd_initted;
 
-static unsigned int recent_entry_hash(u_int32_t addr)
+static unsigned int recent_entry_hash(__be32 addr)
 {
 	if (!hash_rnd_initted) {
 		get_random_bytes(&hash_rnd, 4);
 		hash_rnd_initted = 1;
 	}
-	return jhash_1word(addr, hash_rnd) & (ip_list_hash_size - 1);
+	return jhash_1word((__force u32)addr, hash_rnd) & (ip_list_hash_size - 1);
 }
 
 static struct recent_entry *
-recent_entry_lookup(const struct recent_table *table, u_int32_t addr, u_int8_t ttl)
+recent_entry_lookup(const struct recent_table *table, __be32 addr, u_int8_t ttl)
 {
 	struct recent_entry *e;
 	unsigned int h;
@@ -116,7 +115,7 @@ static void recent_entry_remove(struct recent_table *t, struct recent_entry *e)
 }
 
 static struct recent_entry *
-recent_entry_init(struct recent_table *t, u_int32_t addr, u_int8_t ttl)
+recent_entry_init(struct recent_table *t, __be32 addr, u_int8_t ttl)
 {
 	struct recent_entry *e;
 
@@ -178,7 +177,7 @@ ipt_recent_match(const struct sk_buff *skb,
 	const struct ipt_recent_info *info = matchinfo;
 	struct recent_table *t;
 	struct recent_entry *e;
-	u_int32_t addr;
+	__be32 addr;
 	u_int8_t ttl;
 	int ret = info->invert;
 
@@ -406,7 +405,7 @@ static ssize_t recent_proc_write(struct file *file, const char __user *input,
 	struct recent_table *t = pde->data;
 	struct recent_entry *e;
 	char buf[sizeof("+255.255.255.255")], *c = buf;
-	u_int32_t addr;
+	__be32 addr;
 	int add;
 
 	if (size > sizeof(buf))
diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c
index 79336cb42527..e62ea2bb9c0a 100644
--- a/net/ipv4/netfilter/iptable_mangle.c
+++ b/net/ipv4/netfilter/iptable_mangle.c
@@ -131,7 +131,7 @@ ipt_local_hook(unsigned int hook,
 {
 	unsigned int ret;
 	u_int8_t tos;
-	u_int32_t saddr, daddr;
+	__be32 saddr, daddr;
 	unsigned long nfmark;
 
 	/* root is playing with raw sockets. */
-- 
cgit v1.2.3


From d4263cde88d3fee2af0aac8836bb785cdb6b06c0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:22:51 -0700
Subject: [NETFILTER]: h323 annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4/ip_conntrack_h323.h |  6 +-
 net/ipv4/netfilter/ip_conntrack_helper_h323.c    | 84 ++++++++++++------------
 net/ipv4/netfilter/ip_nat_helper_h323.c          | 16 ++---
 3 files changed, 53 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/include/linux/netfilter_ipv4/ip_conntrack_h323.h b/include/linux/netfilter_ipv4/ip_conntrack_h323.h
index 3cbff7379002..943cc6a4871d 100644
--- a/include/linux/netfilter_ipv4/ip_conntrack_h323.h
+++ b/include/linux/netfilter_ipv4/ip_conntrack_h323.h
@@ -30,7 +30,7 @@ struct ip_ct_h323_master {
 struct ip_conntrack_expect;
 
 extern int get_h225_addr(unsigned char *data, TransportAddress * addr,
-			 u_int32_t * ip, u_int16_t * port);
+			 __be32 * ip, u_int16_t * port);
 extern void ip_conntrack_h245_expect(struct ip_conntrack *new,
 				     struct ip_conntrack_expect *this);
 extern void ip_conntrack_q931_expect(struct ip_conntrack *new,
@@ -38,11 +38,11 @@ extern void ip_conntrack_q931_expect(struct ip_conntrack *new,
 extern int (*set_h245_addr_hook) (struct sk_buff ** pskb,
 				  unsigned char **data, int dataoff,
 				  H245_TransportAddress * addr,
-				  u_int32_t ip, u_int16_t port);
+				  __be32 ip, u_int16_t port);
 extern int (*set_h225_addr_hook) (struct sk_buff ** pskb,
 				  unsigned char **data, int dataoff,
 				  TransportAddress * addr,
-				  u_int32_t ip, u_int16_t port);
+				  __be32 ip, u_int16_t port);
 extern int (*set_sig_addr_hook) (struct sk_buff ** pskb,
 				 struct ip_conntrack * ct,
 				 enum ip_conntrack_info ctinfo,
diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323.c b/net/ipv4/netfilter/ip_conntrack_helper_h323.c
index 9a39e2969712..7b7441202bfd 100644
--- a/net/ipv4/netfilter/ip_conntrack_helper_h323.c
+++ b/net/ipv4/netfilter/ip_conntrack_helper_h323.c
@@ -49,11 +49,11 @@ MODULE_PARM_DESC(callforward_filter, "only create call forwarding expectations "
 int (*set_h245_addr_hook) (struct sk_buff ** pskb,
 			   unsigned char **data, int dataoff,
 			   H245_TransportAddress * addr,
-			   u_int32_t ip, u_int16_t port);
+			   __be32 ip, u_int16_t port);
 int (*set_h225_addr_hook) (struct sk_buff ** pskb,
 			   unsigned char **data, int dataoff,
 			   TransportAddress * addr,
-			   u_int32_t ip, u_int16_t port);
+			   __be32 ip, u_int16_t port);
 int (*set_sig_addr_hook) (struct sk_buff ** pskb,
 			  struct ip_conntrack * ct,
 			  enum ip_conntrack_info ctinfo,
@@ -209,7 +209,7 @@ static int get_tpkt_data(struct sk_buff **pskb, struct ip_conntrack *ct,
 
 /****************************************************************************/
 static int get_h245_addr(unsigned char *data, H245_TransportAddress * addr,
-			 u_int32_t * ip, u_int16_t * port)
+			 __be32 * ip, u_int16_t * port)
 {
 	unsigned char *p;
 
@@ -232,7 +232,7 @@ static int expect_rtp_rtcp(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	u_int16_t rtp_port;
 	struct ip_conntrack_expect *rtp_exp;
@@ -254,10 +254,10 @@ static int expect_rtp_rtcp(struct sk_buff **pskb, struct ip_conntrack *ct,
 	rtp_exp->tuple.dst.ip = ct->tuplehash[!dir].tuple.dst.ip;
 	rtp_exp->tuple.dst.u.udp.port = htons(rtp_port);
 	rtp_exp->tuple.dst.protonum = IPPROTO_UDP;
-	rtp_exp->mask.src.ip = 0xFFFFFFFF;
+	rtp_exp->mask.src.ip = htonl(0xFFFFFFFF);
 	rtp_exp->mask.src.u.udp.port = 0;
-	rtp_exp->mask.dst.ip = 0xFFFFFFFF;
-	rtp_exp->mask.dst.u.udp.port = 0xFFFF;
+	rtp_exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	rtp_exp->mask.dst.u.udp.port = htons(0xFFFF);
 	rtp_exp->mask.dst.protonum = 0xFF;
 	rtp_exp->flags = 0;
 
@@ -271,10 +271,10 @@ static int expect_rtp_rtcp(struct sk_buff **pskb, struct ip_conntrack *ct,
 	rtcp_exp->tuple.dst.ip = ct->tuplehash[!dir].tuple.dst.ip;
 	rtcp_exp->tuple.dst.u.udp.port = htons(rtp_port + 1);
 	rtcp_exp->tuple.dst.protonum = IPPROTO_UDP;
-	rtcp_exp->mask.src.ip = 0xFFFFFFFF;
+	rtcp_exp->mask.src.ip = htonl(0xFFFFFFFF);
 	rtcp_exp->mask.src.u.udp.port = 0;
-	rtcp_exp->mask.dst.ip = 0xFFFFFFFF;
-	rtcp_exp->mask.dst.u.udp.port = 0xFFFF;
+	rtcp_exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	rtcp_exp->mask.dst.u.udp.port = htons(0xFFFF);
 	rtcp_exp->mask.dst.protonum = 0xFF;
 	rtcp_exp->flags = 0;
 
@@ -325,7 +325,7 @@ static int expect_t120(struct sk_buff **pskb,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp = NULL;
 
@@ -342,10 +342,10 @@ static int expect_t120(struct sk_buff **pskb,
 	exp->tuple.dst.ip = ct->tuplehash[!dir].tuple.dst.ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_TCP;
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = IP_CT_EXPECT_PERMANENT;	/* Accept multiple channels */
 
@@ -626,7 +626,7 @@ void ip_conntrack_h245_expect(struct ip_conntrack *new,
 
 /****************************************************************************/
 int get_h225_addr(unsigned char *data, TransportAddress * addr,
-		  u_int32_t * ip, u_int16_t * port)
+		  __be32 * ip, u_int16_t * port)
 {
 	unsigned char *p;
 
@@ -648,7 +648,7 @@ static int expect_h245(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp = NULL;
 
@@ -665,10 +665,10 @@ static int expect_h245(struct sk_buff **pskb, struct ip_conntrack *ct,
 	exp->tuple.dst.ip = ct->tuplehash[!dir].tuple.dst.ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_TCP;
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = 0;
 
@@ -709,7 +709,7 @@ static int expect_callforwarding(struct sk_buff **pskb,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp = NULL;
 
@@ -751,10 +751,10 @@ static int expect_callforwarding(struct sk_buff **pskb,
 	exp->tuple.dst.ip = ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_TCP;
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = 0;
 
@@ -791,7 +791,7 @@ static int process_setup(struct sk_buff **pskb, struct ip_conntrack *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret;
 	int i;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 
 	DEBUGP("ip_ct_q931: Setup\n");
@@ -1188,7 +1188,7 @@ static unsigned char *get_udp_data(struct sk_buff **pskb, int *datalen)
 
 /****************************************************************************/
 static struct ip_conntrack_expect *find_expect(struct ip_conntrack *ct,
-					       u_int32_t ip, u_int16_t port)
+					       __be32 ip, u_int16_t port)
 {
 	struct ip_conntrack_expect *exp;
 	struct ip_conntrack_tuple tuple;
@@ -1228,7 +1228,7 @@ static int expect_q931(struct sk_buff **pskb, struct ip_conntrack *ct,
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
 	int i;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp;
 
@@ -1251,10 +1251,10 @@ static int expect_q931(struct sk_buff **pskb, struct ip_conntrack *ct,
 	exp->tuple.dst.ip = ct->tuplehash[!dir].tuple.dst.ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_TCP;
-	exp->mask.src.ip = gkrouted_only ? 0xFFFFFFFF : 0;
+	exp->mask.src.ip = gkrouted_only ? htonl(0xFFFFFFFF) : 0;
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = IP_CT_EXPECT_PERMANENT;	/* Accept multiple calls */
 
@@ -1307,7 +1307,7 @@ static int process_gcf(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp;
 
@@ -1333,10 +1333,10 @@ static int process_gcf(struct sk_buff **pskb, struct ip_conntrack *ct,
 	exp->tuple.dst.ip = ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_UDP;
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = 0;
 	exp->expectfn = ip_conntrack_ras_expect;
@@ -1477,7 +1477,7 @@ static int process_arq(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	struct ip_ct_h323_master *info = &ct->help.ct_h323_info;
 	int dir = CTINFO2DIR(ctinfo);
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 
 	DEBUGP("ip_ct_ras: ARQ\n");
@@ -1513,7 +1513,7 @@ static int process_acf(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp;
 
@@ -1538,10 +1538,10 @@ static int process_acf(struct sk_buff **pskb, struct ip_conntrack *ct,
 	exp->tuple.dst.ip = ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_TCP;
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = IP_CT_EXPECT_PERMANENT;
 	exp->expectfn = ip_conntrack_q931_expect;
@@ -1581,7 +1581,7 @@ static int process_lcf(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int ret = 0;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 	struct ip_conntrack_expect *exp = NULL;
 
@@ -1598,10 +1598,10 @@ static int process_lcf(struct sk_buff **pskb, struct ip_conntrack *ct,
 	exp->tuple.dst.ip = ip;
 	exp->tuple.dst.u.tcp.port = htons(port);
 	exp->tuple.dst.protonum = IPPROTO_TCP;
-	exp->mask.src.ip = 0xFFFFFFFF;
+	exp->mask.src.ip = htonl(0xFFFFFFFF);
 	exp->mask.src.u.tcp.port = 0;
-	exp->mask.dst.ip = 0xFFFFFFFF;
-	exp->mask.dst.u.tcp.port = 0xFFFF;
+	exp->mask.dst.ip = htonl(0xFFFFFFFF);
+	exp->mask.dst.u.tcp.port = htons(0xFFFF);
 	exp->mask.dst.protonum = 0xFF;
 	exp->flags = IP_CT_EXPECT_PERMANENT;
 	exp->expectfn = ip_conntrack_q931_expect;
diff --git a/net/ipv4/netfilter/ip_nat_helper_h323.c b/net/ipv4/netfilter/ip_nat_helper_h323.c
index 419b878fb467..4a7d34466ee2 100644
--- a/net/ipv4/netfilter/ip_nat_helper_h323.c
+++ b/net/ipv4/netfilter/ip_nat_helper_h323.c
@@ -32,13 +32,13 @@
 /****************************************************************************/
 static int set_addr(struct sk_buff **pskb,
 		    unsigned char **data, int dataoff,
-		    unsigned int addroff, u_int32_t ip, u_int16_t port)
+		    unsigned int addroff, __be32 ip, u_int16_t port)
 {
 	enum ip_conntrack_info ctinfo;
 	struct ip_conntrack *ct = ip_conntrack_get(*pskb, &ctinfo);
 	struct {
-		u_int32_t ip;
-		u_int16_t port;
+		__be32 ip;
+		__be16 port;
 	} __attribute__ ((__packed__)) buf;
 	struct tcphdr _tcph, *th;
 
@@ -86,7 +86,7 @@ static int set_addr(struct sk_buff **pskb,
 static int set_h225_addr(struct sk_buff **pskb,
 			 unsigned char **data, int dataoff,
 			 TransportAddress * addr,
-			 u_int32_t ip, u_int16_t port)
+			 __be32 ip, u_int16_t port)
 {
 	return set_addr(pskb, data, dataoff, addr->ipAddress.ip, ip, port);
 }
@@ -95,7 +95,7 @@ static int set_h225_addr(struct sk_buff **pskb,
 static int set_h245_addr(struct sk_buff **pskb,
 			 unsigned char **data, int dataoff,
 			 H245_TransportAddress * addr,
-			 u_int32_t ip, u_int16_t port)
+			 __be32 ip, u_int16_t port)
 {
 	return set_addr(pskb, data, dataoff,
 			addr->unicastAddress.iPAddress.network, ip, port);
@@ -110,7 +110,7 @@ static int set_sig_addr(struct sk_buff **pskb, struct ip_conntrack *ct,
 	struct ip_ct_h323_master *info = &ct->help.ct_h323_info;
 	int dir = CTINFO2DIR(ctinfo);
 	int i;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 
 	for (i = 0; i < count; i++) {
@@ -164,7 +164,7 @@ static int set_ras_addr(struct sk_buff **pskb, struct ip_conntrack *ct,
 {
 	int dir = CTINFO2DIR(ctinfo);
 	int i;
-	u_int32_t ip;
+	__be32 ip;
 	u_int16_t port;
 
 	for (i = 0; i < count; i++) {
@@ -433,7 +433,7 @@ static int nat_q931(struct sk_buff **pskb, struct ip_conntrack *ct,
 	struct ip_ct_h323_master *info = &ct->help.ct_h323_info;
 	int dir = CTINFO2DIR(ctinfo);
 	u_int16_t nated_port = port;
-	u_int32_t ip;
+	__be32 ip;
 
 	/* Set expectations for NAT */
 	exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port;
-- 
cgit v1.2.3


From 014d730d56b559eacb11e91969a1f41c3feb36f9 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 28 Sep 2006 14:29:52 -0700
Subject: [IPVS]: ipvs annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip_vs.h             | 68 ++++++++++++++++++++---------------------
 net/ipv4/ipvs/ip_vs_conn.c      | 24 +++++++--------
 net/ipv4/ipvs/ip_vs_core.c      | 14 ++++-----
 net/ipv4/ipvs/ip_vs_ctl.c       | 26 ++++++++--------
 net/ipv4/ipvs/ip_vs_dh.c        |  4 +--
 net/ipv4/ipvs/ip_vs_ftp.c       | 12 ++++----
 net/ipv4/ipvs/ip_vs_lblc.c      |  8 ++---
 net/ipv4/ipvs/ip_vs_lblcr.c     |  8 ++---
 net/ipv4/ipvs/ip_vs_proto.c     |  2 +-
 net/ipv4/ipvs/ip_vs_proto_tcp.c | 10 +++---
 net/ipv4/ipvs/ip_vs_proto_udp.c | 14 ++++-----
 net/ipv4/ipvs/ip_vs_sh.c        |  4 +--
 net/ipv4/ipvs/ip_vs_sync.c      | 16 +++++-----
 net/ipv4/ipvs/ip_vs_xmit.c      |  2 +-
 14 files changed, 106 insertions(+), 106 deletions(-)

(limited to 'include')

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 3b57b159b653..49c717e3b040 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -100,22 +100,22 @@
 struct ip_vs_service_user {
 	/* virtual service addresses */
 	u_int16_t		protocol;
-	u_int32_t		addr;		/* virtual ip address */
-	u_int16_t		port;
+	__be32			addr;		/* virtual ip address */
+	__be16			port;
 	u_int32_t		fwmark;		/* firwall mark of service */
 
 	/* virtual service options */
 	char			sched_name[IP_VS_SCHEDNAME_MAXLEN];
 	unsigned		flags;		/* virtual service flags */
 	unsigned		timeout;	/* persistent timeout in sec */
-	u_int32_t		netmask;	/* persistent netmask */
+	__be32			netmask;	/* persistent netmask */
 };
 
 
 struct ip_vs_dest_user {
 	/* destination server address */
-	u_int32_t		addr;
-	u_int16_t		port;
+	__be32			addr;
+	__be16			port;
 
 	/* real server options */
 	unsigned		conn_flags;	/* connection flags */
@@ -163,15 +163,15 @@ struct ip_vs_getinfo {
 struct ip_vs_service_entry {
 	/* which service: user fills in these */
 	u_int16_t		protocol;
-	u_int32_t		addr;		/* virtual address */
-	u_int16_t		port;
+	__be32			addr;		/* virtual address */
+	__be16			port;
 	u_int32_t		fwmark;		/* firwall mark of service */
 
 	/* service options */
 	char			sched_name[IP_VS_SCHEDNAME_MAXLEN];
 	unsigned		flags;          /* virtual service flags */
 	unsigned		timeout;	/* persistent timeout */
-	u_int32_t		netmask;	/* persistent netmask */
+	__be32			netmask;	/* persistent netmask */
 
 	/* number of real servers */
 	unsigned int		num_dests;
@@ -182,8 +182,8 @@ struct ip_vs_service_entry {
 
 
 struct ip_vs_dest_entry {
-	u_int32_t		addr;		/* destination address */
-	u_int16_t		port;
+	__be32			addr;		/* destination address */
+	__be16			port;
 	unsigned		conn_flags;	/* connection flags */
 	int			weight;		/* destination weight */
 
@@ -203,8 +203,8 @@ struct ip_vs_dest_entry {
 struct ip_vs_get_dests {
 	/* which service: user fills in these */
 	u_int16_t		protocol;
-	u_int32_t		addr;		/* virtual address */
-	u_int16_t		port;
+	__be32			addr;		/* virtual address */
+	__be16			port;
 	u_int32_t		fwmark;		/* firwall mark of service */
 
 	/* number of real servers */
@@ -502,12 +502,12 @@ struct ip_vs_conn {
 	struct list_head        c_list;         /* hashed list heads */
 
 	/* Protocol, addresses and port numbers */
-	__u32                   caddr;          /* client address */
-	__u32                   vaddr;          /* virtual address */
-	__u32                   daddr;          /* destination address */
-	__u16                   cport;
-	__u16                   vport;
-	__u16                   dport;
+	__be32                   caddr;          /* client address */
+	__be32                   vaddr;          /* virtual address */
+	__be32                   daddr;          /* destination address */
+	__be16                   cport;
+	__be16                   vport;
+	__be16                   dport;
 	__u16                   protocol;       /* Which protocol (TCP/UDP) */
 
 	/* counter and timer */
@@ -554,12 +554,12 @@ struct ip_vs_service {
 	atomic_t		usecnt;   /* use counter */
 
 	__u16			protocol; /* which protocol (TCP/UDP) */
-	__u32			addr;	  /* IP address for virtual service */
-	__u16			port;	  /* port number for the service */
+	__be32			addr;	  /* IP address for virtual service */
+	__be16			port;	  /* port number for the service */
 	__u32                   fwmark;   /* firewall mark of the service */
 	unsigned		flags;	  /* service status flags */
 	unsigned		timeout;  /* persistent timeout in ticks */
-	__u32			netmask;  /* grouping granularity */
+	__be32			netmask;  /* grouping granularity */
 
 	struct list_head	destinations;  /* real server d-linked list */
 	__u32			num_dests;     /* number of servers */
@@ -581,8 +581,8 @@ struct ip_vs_dest {
 	struct list_head	n_list;   /* for the dests in the service */
 	struct list_head	d_list;   /* for table with all the dests */
 
-	__u32			addr;		/* IP address of the server */
-	__u16			port;		/* port number of the server */
+	__be32			addr;		/* IP address of the server */
+	__be16			port;		/* port number of the server */
 	volatile unsigned	flags;		/* dest status flags */
 	atomic_t		conn_flags;	/* flags to copy to conn */
 	atomic_t		weight;		/* server weight */
@@ -605,8 +605,8 @@ struct ip_vs_dest {
 	/* for virtual service */
 	struct ip_vs_service	*svc;		/* service it belongs to */
 	__u16			protocol;	/* which protocol (TCP/UDP) */
-	__u32			vaddr;		/* virtual IP address */
-	__u16			vport;		/* virtual port number */
+	__be32			vaddr;		/* virtual IP address */
+	__be16			vport;		/* virtual port number */
 	__u32			vfwmark;	/* firewall mark of service */
 };
 
@@ -648,7 +648,7 @@ struct ip_vs_app
 	/* members for application incarnations */
 	struct list_head	p_list;		/* member in proto app list */
 	struct ip_vs_app	*app;		/* its real application */
-	__u16			port;		/* port number in net order */
+	__be16			port;		/* port number in net order */
 	atomic_t		usecnt;		/* usage counter */
 
 	/* output hook: return false if can't linearize. diff set for TCP.  */
@@ -740,11 +740,11 @@ enum {
 };
 
 extern struct ip_vs_conn *ip_vs_conn_in_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port);
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port);
 extern struct ip_vs_conn *ip_vs_ct_in_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port);
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port);
 extern struct ip_vs_conn *ip_vs_conn_out_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port);
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port);
 
 /* put back the conn without restarting its timer */
 static inline void __ip_vs_conn_put(struct ip_vs_conn *cp)
@@ -752,11 +752,11 @@ static inline void __ip_vs_conn_put(struct ip_vs_conn *cp)
 	atomic_dec(&cp->refcnt);
 }
 extern void ip_vs_conn_put(struct ip_vs_conn *cp);
-extern void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __u16 cport);
+extern void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __be16 cport);
 
 extern struct ip_vs_conn *
-ip_vs_conn_new(int proto, __u32 caddr, __u16 cport, __u32 vaddr, __u16 vport,
-	       __u32 daddr, __u16 dport, unsigned flags,
+ip_vs_conn_new(int proto, __be32 caddr, __be16 cport, __be32 vaddr, __be16 vport,
+	       __be32 daddr, __be16 dport, unsigned flags,
 	       struct ip_vs_dest *dest);
 extern void ip_vs_conn_expire_now(struct ip_vs_conn *cp);
 
@@ -887,7 +887,7 @@ extern int sysctl_ip_vs_nat_icmp_send;
 extern struct ip_vs_stats ip_vs_stats;
 
 extern struct ip_vs_service *
-ip_vs_service_get(__u32 fwmark, __u16 protocol, __u32 vaddr, __u16 vport);
+ip_vs_service_get(__u32 fwmark, __u16 protocol, __be32 vaddr, __be16 vport);
 
 static inline void ip_vs_service_put(struct ip_vs_service *svc)
 {
@@ -895,7 +895,7 @@ static inline void ip_vs_service_put(struct ip_vs_service *svc)
 }
 
 extern struct ip_vs_dest *
-ip_vs_lookup_real_service(__u16 protocol, __u32 daddr, __u16 dport);
+ip_vs_lookup_real_service(__u16 protocol, __be32 daddr, __be16 dport);
 extern int ip_vs_use_count_inc(void);
 extern void ip_vs_use_count_dec(void);
 extern int ip_vs_control_init(void);
diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c
index 87b83813cf2c..8832eb517d52 100644
--- a/net/ipv4/ipvs/ip_vs_conn.c
+++ b/net/ipv4/ipvs/ip_vs_conn.c
@@ -115,9 +115,9 @@ static inline void ct_write_unlock_bh(unsigned key)
 /*
  *	Returns hash value for IPVS connection entry
  */
-static unsigned int ip_vs_conn_hashkey(unsigned proto, __u32 addr, __u16 port)
+static unsigned int ip_vs_conn_hashkey(unsigned proto, __be32 addr, __be16 port)
 {
-	return jhash_3words(addr, port, proto, ip_vs_conn_rnd)
+	return jhash_3words((__force u32)addr, (__force u32)port, proto, ip_vs_conn_rnd)
 		& IP_VS_CONN_TAB_MASK;
 }
 
@@ -188,7 +188,7 @@ static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp)
  *	d_addr, d_port: pkt dest address (load balancer)
  */
 static inline struct ip_vs_conn *__ip_vs_conn_in_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port)
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port)
 {
 	unsigned hash;
 	struct ip_vs_conn *cp;
@@ -215,7 +215,7 @@ static inline struct ip_vs_conn *__ip_vs_conn_in_get
 }
 
 struct ip_vs_conn *ip_vs_conn_in_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port)
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port)
 {
 	struct ip_vs_conn *cp;
 
@@ -234,7 +234,7 @@ struct ip_vs_conn *ip_vs_conn_in_get
 
 /* Get reference to connection template */
 struct ip_vs_conn *ip_vs_ct_in_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port)
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port)
 {
 	unsigned hash;
 	struct ip_vs_conn *cp;
@@ -274,7 +274,7 @@ struct ip_vs_conn *ip_vs_ct_in_get
  *	d_addr, d_port: pkt dest address (foreign host)
  */
 struct ip_vs_conn *ip_vs_conn_out_get
-(int protocol, __u32 s_addr, __u16 s_port, __u32 d_addr, __u16 d_port)
+(int protocol, __be32 s_addr, __be16 s_port, __be32 d_addr, __be16 d_port)
 {
 	unsigned hash;
 	struct ip_vs_conn *cp, *ret=NULL;
@@ -324,7 +324,7 @@ void ip_vs_conn_put(struct ip_vs_conn *cp)
 /*
  *	Fill a no_client_port connection with a client port number
  */
-void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __u16 cport)
+void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __be16 cport)
 {
 	if (ip_vs_conn_unhash(cp)) {
 		spin_lock(&cp->lock);
@@ -508,10 +508,10 @@ int ip_vs_check_template(struct ip_vs_conn *ct)
 		/*
 		 * Invalidate the connection template
 		 */
-		if (ct->vport != 65535) {
+		if (ct->vport != htons(0xffff)) {
 			if (ip_vs_conn_unhash(ct)) {
-				ct->dport = 65535;
-				ct->vport = 65535;
+				ct->dport = htons(0xffff);
+				ct->vport = htons(0xffff);
 				ct->cport = 0;
 				ip_vs_conn_hash(ct);
 			}
@@ -596,8 +596,8 @@ void ip_vs_conn_expire_now(struct ip_vs_conn *cp)
  *	Create a new connection entry and hash it into the ip_vs_conn_tab
  */
 struct ip_vs_conn *
-ip_vs_conn_new(int proto, __u32 caddr, __u16 cport, __u32 vaddr, __u16 vport,
-	       __u32 daddr, __u16 dport, unsigned flags,
+ip_vs_conn_new(int proto, __be32 caddr, __be16 cport, __be32 vaddr, __be16 vport,
+	       __be32 daddr, __be16 dport, unsigned flags,
 	       struct ip_vs_dest *dest)
 {
 	struct ip_vs_conn *cp;
diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c
index 3f47ad8e1cad..6dee03935f78 100644
--- a/net/ipv4/ipvs/ip_vs_core.c
+++ b/net/ipv4/ipvs/ip_vs_core.c
@@ -209,14 +209,14 @@ int ip_vs_make_skb_writable(struct sk_buff **pskb, int writable_len)
 static struct ip_vs_conn *
 ip_vs_sched_persist(struct ip_vs_service *svc,
 		    const struct sk_buff *skb,
-		    __u16 ports[2])
+		    __be16 ports[2])
 {
 	struct ip_vs_conn *cp = NULL;
 	struct iphdr *iph = skb->nh.iph;
 	struct ip_vs_dest *dest;
 	struct ip_vs_conn *ct;
-	__u16  dport;	 /* destination port to forward */
-	__u32  snet;	 /* source network of the client, after masking */
+	__be16  dport;	 /* destination port to forward */
+	__be32  snet;	 /* source network of the client, after masking */
 
 	/* Mask saddr with the netmask to adjust template granularity */
 	snet = iph->saddr & svc->netmask;
@@ -383,7 +383,7 @@ ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
 	struct ip_vs_conn *cp = NULL;
 	struct iphdr *iph = skb->nh.iph;
 	struct ip_vs_dest *dest;
-	__u16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr;
 
 	pptr = skb_header_pointer(skb, iph->ihl*4,
 				  sizeof(_ports), _ports);
@@ -446,7 +446,7 @@ ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
 int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 		struct ip_vs_protocol *pp)
 {
-	__u16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr;
 	struct iphdr *iph = skb->nh.iph;
 
 	pptr = skb_header_pointer(skb, iph->ihl*4,
@@ -576,7 +576,7 @@ void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp,
 
 	/* the TCP/UDP port */
 	if (IPPROTO_TCP == ciph->protocol || IPPROTO_UDP == ciph->protocol) {
-		__u16 *ports = (void *)ciph + ciph->ihl*4;
+		__be16 *ports = (void *)ciph + ciph->ihl*4;
 
 		if (inout)
 			ports[1] = cp->vport;
@@ -775,7 +775,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff **pskb,
 		if (sysctl_ip_vs_nat_icmp_send &&
 		    (pp->protocol == IPPROTO_TCP ||
 		     pp->protocol == IPPROTO_UDP)) {
-			__u16 _ports[2], *pptr;
+			__be16 _ports[2], *pptr;
 
 			pptr = skb_header_pointer(skb, ihl,
 						  sizeof(_ports), _ports);
diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c
index 6a28fafe910c..f261616e4602 100644
--- a/net/ipv4/ipvs/ip_vs_ctl.c
+++ b/net/ipv4/ipvs/ip_vs_ctl.c
@@ -283,7 +283,7 @@ static atomic_t ip_vs_nullsvc_counter = ATOMIC_INIT(0);
  *	Returns hash value for virtual service
  */
 static __inline__ unsigned
-ip_vs_svc_hashkey(unsigned proto, __u32 addr, __u16 port)
+ip_vs_svc_hashkey(unsigned proto, __be32 addr, __be16 port)
 {
 	register unsigned porth = ntohs(port);
 
@@ -365,7 +365,7 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc)
  *	Get service by {proto,addr,port} in the service table.
  */
 static __inline__ struct ip_vs_service *
-__ip_vs_service_get(__u16 protocol, __u32 vaddr, __u16 vport)
+__ip_vs_service_get(__u16 protocol, __be32 vaddr, __be16 vport)
 {
 	unsigned hash;
 	struct ip_vs_service *svc;
@@ -410,7 +410,7 @@ static __inline__ struct ip_vs_service *__ip_vs_svc_fwm_get(__u32 fwmark)
 }
 
 struct ip_vs_service *
-ip_vs_service_get(__u32 fwmark, __u16 protocol, __u32 vaddr, __u16 vport)
+ip_vs_service_get(__u32 fwmark, __u16 protocol, __be32 vaddr, __be16 vport)
 {
 	struct ip_vs_service *svc;
 
@@ -480,7 +480,7 @@ __ip_vs_unbind_svc(struct ip_vs_dest *dest)
 /*
  *	Returns hash value for real service
  */
-static __inline__ unsigned ip_vs_rs_hashkey(__u32 addr, __u16 port)
+static __inline__ unsigned ip_vs_rs_hashkey(__be32 addr, __be16 port)
 {
 	register unsigned porth = ntohs(port);
 
@@ -531,7 +531,7 @@ static int ip_vs_rs_unhash(struct ip_vs_dest *dest)
  *	Lookup real service by <proto,addr,port> in the real service table.
  */
 struct ip_vs_dest *
-ip_vs_lookup_real_service(__u16 protocol, __u32 daddr, __u16 dport)
+ip_vs_lookup_real_service(__u16 protocol, __be32 daddr, __be16 dport)
 {
 	unsigned hash;
 	struct ip_vs_dest *dest;
@@ -562,7 +562,7 @@ ip_vs_lookup_real_service(__u16 protocol, __u32 daddr, __u16 dport)
  *	Lookup destination by {addr,port} in the given service
  */
 static struct ip_vs_dest *
-ip_vs_lookup_dest(struct ip_vs_service *svc, __u32 daddr, __u16 dport)
+ip_vs_lookup_dest(struct ip_vs_service *svc, __be32 daddr, __be16 dport)
 {
 	struct ip_vs_dest *dest;
 
@@ -591,7 +591,7 @@ ip_vs_lookup_dest(struct ip_vs_service *svc, __u32 daddr, __u16 dport)
  *  scheduling.
  */
 static struct ip_vs_dest *
-ip_vs_trash_get_dest(struct ip_vs_service *svc, __u32 daddr, __u16 dport)
+ip_vs_trash_get_dest(struct ip_vs_service *svc, __be32 daddr, __be16 dport)
 {
 	struct ip_vs_dest *dest, *nxt;
 
@@ -773,8 +773,8 @@ static int
 ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user *udest)
 {
 	struct ip_vs_dest *dest;
-	__u32 daddr = udest->addr;
-	__u16 dport = udest->port;
+	__be32 daddr = udest->addr;
+	__be16 dport = udest->port;
 	int ret;
 
 	EnterFunction(2);
@@ -879,8 +879,8 @@ static int
 ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user *udest)
 {
 	struct ip_vs_dest *dest;
-	__u32 daddr = udest->addr;
-	__u16 dport = udest->port;
+	__be32 daddr = udest->addr;
+	__be16 dport = udest->port;
 
 	EnterFunction(2);
 
@@ -991,8 +991,8 @@ static int
 ip_vs_del_dest(struct ip_vs_service *svc,struct ip_vs_dest_user *udest)
 {
 	struct ip_vs_dest *dest;
-	__u32 daddr = udest->addr;
-	__u16 dport = udest->port;
+	__be32 daddr = udest->addr;
+	__be16 dport = udest->port;
 
 	EnterFunction(2);
 
diff --git a/net/ipv4/ipvs/ip_vs_dh.c b/net/ipv4/ipvs/ip_vs_dh.c
index 9fee19c4c617..502111fba872 100644
--- a/net/ipv4/ipvs/ip_vs_dh.c
+++ b/net/ipv4/ipvs/ip_vs_dh.c
@@ -66,7 +66,7 @@ struct ip_vs_dh_bucket {
 /*
  *	Returns hash value for IPVS DH entry
  */
-static inline unsigned ip_vs_dh_hashkey(__u32 addr)
+static inline unsigned ip_vs_dh_hashkey(__be32 addr)
 {
 	return (ntohl(addr)*2654435761UL) & IP_VS_DH_TAB_MASK;
 }
@@ -76,7 +76,7 @@ static inline unsigned ip_vs_dh_hashkey(__u32 addr)
  *      Get ip_vs_dest associated with supplied parameters.
  */
 static inline struct ip_vs_dest *
-ip_vs_dh_get(struct ip_vs_dh_bucket *tbl, __u32 addr)
+ip_vs_dh_get(struct ip_vs_dh_bucket *tbl, __be32 addr)
 {
 	return (tbl[ip_vs_dh_hashkey(addr)]).dest;
 }
diff --git a/net/ipv4/ipvs/ip_vs_ftp.c b/net/ipv4/ipvs/ip_vs_ftp.c
index 62505d8f9ff6..b3bbb4a07ae0 100644
--- a/net/ipv4/ipvs/ip_vs_ftp.c
+++ b/net/ipv4/ipvs/ip_vs_ftp.c
@@ -74,7 +74,7 @@ ip_vs_ftp_done_conn(struct ip_vs_app *app, struct ip_vs_conn *cp)
  */
 static int ip_vs_ftp_get_addrport(char *data, char *data_limit,
 				  const char *pattern, size_t plen, char term,
-				  __u32 *addr, __u16 *port,
+				  __be32 *addr, __be16 *port,
 				  char **start, char **end)
 {
 	unsigned char p[6];
@@ -140,8 +140,8 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp,
 	struct tcphdr *th;
 	char *data, *data_limit;
 	char *start, *end;
-	__u32 from;
-	__u16 port;
+	__be32 from;
+	__be16 port;
 	struct ip_vs_conn *n_cp;
 	char buf[24];		/* xxx.xxx.xxx.xxx,ppp,ppp\000 */
 	unsigned buf_len;
@@ -199,7 +199,7 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp,
 		from = n_cp->vaddr;
 		port = n_cp->vport;
 		sprintf(buf,"%d,%d,%d,%d,%d,%d", NIPQUAD(from),
-			port&255, (port>>8)&255);
+			ntohs(port)&255, (ntohs(port)>>8)&255);
 		buf_len = strlen(buf);
 
 		/*
@@ -243,8 +243,8 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp,
 	struct tcphdr *th;
 	char *data, *data_start, *data_limit;
 	char *start, *end;
-	__u32 to;
-	__u16 port;
+	__be32 to;
+	__be16 port;
 	struct ip_vs_conn *n_cp;
 
 	/* no diff required for incoming packets */
diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c
index 6e5cb92a5c83..524751e031de 100644
--- a/net/ipv4/ipvs/ip_vs_lblc.c
+++ b/net/ipv4/ipvs/ip_vs_lblc.c
@@ -87,7 +87,7 @@ static int sysctl_ip_vs_lblc_expiration = 24*60*60*HZ;
  */
 struct ip_vs_lblc_entry {
 	struct list_head        list;
-	__u32                   addr;           /* destination IP address */
+	__be32                  addr;           /* destination IP address */
 	struct ip_vs_dest       *dest;          /* real server (cache) */
 	unsigned long           lastuse;        /* last used time */
 };
@@ -160,7 +160,7 @@ static struct ctl_table_header * sysctl_header;
  *      IP address to a server.
  */
 static inline struct ip_vs_lblc_entry *
-ip_vs_lblc_new(__u32 daddr, struct ip_vs_dest *dest)
+ip_vs_lblc_new(__be32 daddr, struct ip_vs_dest *dest)
 {
 	struct ip_vs_lblc_entry *en;
 
@@ -195,7 +195,7 @@ static inline void ip_vs_lblc_free(struct ip_vs_lblc_entry *en)
 /*
  *	Returns hash value for IPVS LBLC entry
  */
-static inline unsigned ip_vs_lblc_hashkey(__u32 addr)
+static inline unsigned ip_vs_lblc_hashkey(__be32 addr)
 {
 	return (ntohl(addr)*2654435761UL) & IP_VS_LBLC_TAB_MASK;
 }
@@ -234,7 +234,7 @@ ip_vs_lblc_hash(struct ip_vs_lblc_table *tbl, struct ip_vs_lblc_entry *en)
  *  Get ip_vs_lblc_entry associated with supplied parameters.
  */
 static inline struct ip_vs_lblc_entry *
-ip_vs_lblc_get(struct ip_vs_lblc_table *tbl, __u32 addr)
+ip_vs_lblc_get(struct ip_vs_lblc_table *tbl, __be32 addr)
 {
 	unsigned hash;
 	struct ip_vs_lblc_entry *en;
diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c
index 32ba37ba72d8..08990192b6ec 100644
--- a/net/ipv4/ipvs/ip_vs_lblcr.c
+++ b/net/ipv4/ipvs/ip_vs_lblcr.c
@@ -276,7 +276,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_max(struct ip_vs_dest_set *set)
  */
 struct ip_vs_lblcr_entry {
 	struct list_head        list;
-	__u32                   addr;           /* destination IP address */
+	__be32                   addr;           /* destination IP address */
 	struct ip_vs_dest_set   set;            /* destination server set */
 	unsigned long           lastuse;        /* last used time */
 };
@@ -348,7 +348,7 @@ static struct ctl_table_header * sysctl_header;
  *      new/free a ip_vs_lblcr_entry, which is a mapping of a destination
  *      IP address to a server.
  */
-static inline struct ip_vs_lblcr_entry *ip_vs_lblcr_new(__u32 daddr)
+static inline struct ip_vs_lblcr_entry *ip_vs_lblcr_new(__be32 daddr)
 {
 	struct ip_vs_lblcr_entry *en;
 
@@ -381,7 +381,7 @@ static inline void ip_vs_lblcr_free(struct ip_vs_lblcr_entry *en)
 /*
  *	Returns hash value for IPVS LBLCR entry
  */
-static inline unsigned ip_vs_lblcr_hashkey(__u32 addr)
+static inline unsigned ip_vs_lblcr_hashkey(__be32 addr)
 {
 	return (ntohl(addr)*2654435761UL) & IP_VS_LBLCR_TAB_MASK;
 }
@@ -420,7 +420,7 @@ ip_vs_lblcr_hash(struct ip_vs_lblcr_table *tbl, struct ip_vs_lblcr_entry *en)
  *  Get ip_vs_lblcr_entry associated with supplied parameters.
  */
 static inline struct ip_vs_lblcr_entry *
-ip_vs_lblcr_get(struct ip_vs_lblcr_table *tbl, __u32 addr)
+ip_vs_lblcr_get(struct ip_vs_lblcr_table *tbl, __be32 addr)
 {
 	unsigned hash;
 	struct ip_vs_lblcr_entry *en;
diff --git a/net/ipv4/ipvs/ip_vs_proto.c b/net/ipv4/ipvs/ip_vs_proto.c
index 867d4e9c6594..c4528b5c800d 100644
--- a/net/ipv4/ipvs/ip_vs_proto.c
+++ b/net/ipv4/ipvs/ip_vs_proto.c
@@ -176,7 +176,7 @@ ip_vs_tcpudp_debug_packet(struct ip_vs_protocol *pp,
 			pp->name, NIPQUAD(ih->saddr),
 			NIPQUAD(ih->daddr));
 	else {
-		__u16 _ports[2], *pptr
+		__be16 _ports[2], *pptr
 ;
 		pptr = skb_header_pointer(skb, offset + ih->ihl*4,
 					  sizeof(_ports), _ports);
diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c
index 820e8318d10d..bfe779e74590 100644
--- a/net/ipv4/ipvs/ip_vs_proto_tcp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c
@@ -29,7 +29,7 @@ static struct ip_vs_conn *
 tcp_conn_in_get(const struct sk_buff *skb, struct ip_vs_protocol *pp,
 		const struct iphdr *iph, unsigned int proto_off, int inverse)
 {
-	__u16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr;
 
 	pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports);
 	if (pptr == NULL)
@@ -50,7 +50,7 @@ static struct ip_vs_conn *
 tcp_conn_out_get(const struct sk_buff *skb, struct ip_vs_protocol *pp,
 		 const struct iphdr *iph, unsigned int proto_off, int inverse)
 {
-	__u16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr;
 
 	pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports);
 	if (pptr == NULL)
@@ -112,12 +112,12 @@ tcp_conn_schedule(struct sk_buff *skb,
 
 
 static inline void
-tcp_fast_csum_update(struct tcphdr *tcph, u32 oldip, u32 newip,
-		     u16 oldport, u16 newport)
+tcp_fast_csum_update(struct tcphdr *tcph, __be32 oldip, __be32 newip,
+		     __be16 oldport, __be16 newport)
 {
 	tcph->check =
 		ip_vs_check_diff(~oldip, newip,
-				 ip_vs_check_diff(oldport ^ 0xFFFF,
+				 ip_vs_check_diff(oldport ^ htonl(0xFFFF),
 						  newport, tcph->check));
 }
 
diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c
index 90c8166c0ec1..54aa7603591f 100644
--- a/net/ipv4/ipvs/ip_vs_proto_udp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_udp.c
@@ -29,7 +29,7 @@ udp_conn_in_get(const struct sk_buff *skb, struct ip_vs_protocol *pp,
 		const struct iphdr *iph, unsigned int proto_off, int inverse)
 {
 	struct ip_vs_conn *cp;
-	__u16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr;
 
 	pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports);
 	if (pptr == NULL)
@@ -54,7 +54,7 @@ udp_conn_out_get(const struct sk_buff *skb, struct ip_vs_protocol *pp,
 		 const struct iphdr *iph, unsigned int proto_off, int inverse)
 {
 	struct ip_vs_conn *cp;
-	__u16 _ports[2], *pptr;
+	__be16 _ports[2], *pptr;
 
 	pptr = skb_header_pointer(skb, skb->nh.iph->ihl*4,
 				  sizeof(_ports), _ports);
@@ -117,15 +117,15 @@ udp_conn_schedule(struct sk_buff *skb, struct ip_vs_protocol *pp,
 
 
 static inline void
-udp_fast_csum_update(struct udphdr *uhdr, u32 oldip, u32 newip,
-		     u16 oldport, u16 newport)
+udp_fast_csum_update(struct udphdr *uhdr, __be32 oldip, __be32 newip,
+		     __be16 oldport, __be16 newport)
 {
 	uhdr->check =
 		ip_vs_check_diff(~oldip, newip,
-				 ip_vs_check_diff(oldport ^ 0xFFFF,
+				 ip_vs_check_diff(oldport ^ htonl(0xFFFF),
 						  newport, uhdr->check));
 	if (!uhdr->check)
-		uhdr->check = 0xFFFF;
+		uhdr->check = htonl(0xFFFF);
 }
 
 static int
@@ -173,7 +173,7 @@ udp_snat_handler(struct sk_buff **pskb,
 						cp->protocol,
 						(*pskb)->csum);
 		if (udph->check == 0)
-			udph->check = 0xFFFF;
+			udph->check = htonl(0xFFFF);
 		IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n",
 			  pp->name, udph->check,
 			  (char*)&(udph->check) - (char*)udph);
diff --git a/net/ipv4/ipvs/ip_vs_sh.c b/net/ipv4/ipvs/ip_vs_sh.c
index 7775e6cc68be..338668f88fe2 100644
--- a/net/ipv4/ipvs/ip_vs_sh.c
+++ b/net/ipv4/ipvs/ip_vs_sh.c
@@ -63,7 +63,7 @@ struct ip_vs_sh_bucket {
 /*
  *	Returns hash value for IPVS SH entry
  */
-static inline unsigned ip_vs_sh_hashkey(__u32 addr)
+static inline unsigned ip_vs_sh_hashkey(__be32 addr)
 {
 	return (ntohl(addr)*2654435761UL) & IP_VS_SH_TAB_MASK;
 }
@@ -73,7 +73,7 @@ static inline unsigned ip_vs_sh_hashkey(__u32 addr)
  *      Get ip_vs_dest associated with supplied parameters.
  */
 static inline struct ip_vs_dest *
-ip_vs_sh_get(struct ip_vs_sh_bucket *tbl, __u32 addr)
+ip_vs_sh_get(struct ip_vs_sh_bucket *tbl, __be32 addr)
 {
 	return (tbl[ip_vs_sh_hashkey(addr)]).dest;
 }
diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c
index 0c8d20da6139..6ab57d72b615 100644
--- a/net/ipv4/ipvs/ip_vs_sync.c
+++ b/net/ipv4/ipvs/ip_vs_sync.c
@@ -48,16 +48,16 @@ struct ip_vs_sync_conn {
 
 	/* Protocol, addresses and port numbers */
 	__u8			protocol;       /* Which protocol (TCP/UDP) */
-	__u16			cport;
-	__u16                   vport;
-	__u16                   dport;
-	__u32                   caddr;          /* client address */
-	__u32                   vaddr;          /* virtual address */
-	__u32                   daddr;          /* destination address */
+	__be16			cport;
+	__be16                  vport;
+	__be16                  dport;
+	__be32                  caddr;          /* client address */
+	__be32                  vaddr;          /* virtual address */
+	__be32                  daddr;          /* destination address */
 
 	/* Flags and state transition */
-	__u16                   flags;          /* status flags */
-	__u16                   state;          /* state info */
+	__be16                  flags;          /* status flags */
+	__be16                  state;          /* state info */
 
 	/* The sequence options start here */
 };
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index 52c12e9edbbc..e1f77bd7c9a5 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -232,7 +232,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	/* check if it is a connection of no-client-port */
 	if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT)) {
-		__u16 _pt, *p;
+		__be16 _pt, *p;
 		p = skb_header_pointer(skb, iph->ihl*4, sizeof(_pt), &_pt);
 		if (p == NULL)
 			goto tx_error;
-- 
cgit v1.2.3


From 32f50cdee666333168b5203c7864bede159f789e Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Thu, 28 Sep 2006 14:51:47 -0700
Subject: [NetLabel]: add audit support for configuration changes

This patch adds audit support to NetLabel, including six new audit message
types shown below.

 #define AUDIT_MAC_UNLBL_ACCEPT 1406
 #define AUDIT_MAC_UNLBL_DENY   1407
 #define AUDIT_MAC_CIPSOV4_ADD  1408
 #define AUDIT_MAC_CIPSOV4_DEL  1409
 #define AUDIT_MAC_MAP_ADD      1410
 #define AUDIT_MAC_MAP_DEL      1411

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/audit.h              |  6 +++
 include/net/cipso_ipv4.h           |  5 ++-
 include/net/netlabel.h             |  2 +-
 net/ipv4/cipso_ipv4.c              |  8 +++-
 net/netlabel/netlabel_cipso_v4.c   | 43 +++++++++++++-----
 net/netlabel/netlabel_domainhash.c | 54 +++++++++++++++++++---
 net/netlabel/netlabel_domainhash.h |  6 +--
 net/netlabel/netlabel_mgmt.c       | 14 +++---
 net/netlabel/netlabel_unlabeled.c  | 36 ++++++++++++---
 net/netlabel/netlabel_user.c       | 91 ++++++++++++++++++++++++++++++++++++++
 net/netlabel/netlabel_user.h       |  6 +++
 11 files changed, 235 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 40a6c26294ae..42719d07612a 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -95,6 +95,12 @@
 #define AUDIT_MAC_POLICY_LOAD	1403	/* Policy file load */
 #define AUDIT_MAC_STATUS	1404	/* Changed enforcing,permissive,off */
 #define AUDIT_MAC_CONFIG_CHANGE	1405	/* Changes to booleans */
+#define AUDIT_MAC_UNLBL_ACCEPT	1406	/* NetLabel: allow unlabeled traffic */
+#define AUDIT_MAC_UNLBL_DENY	1407	/* NetLabel: deny unlabeled traffic */
+#define AUDIT_MAC_CIPSOV4_ADD	1408	/* NetLabel: add CIPSOv4 DOI entry */
+#define AUDIT_MAC_CIPSOV4_DEL	1409	/* NetLabel: del CIPSOv4 DOI entry */
+#define AUDIT_MAC_MAP_ADD	1410	/* NetLabel: add LSM domain mapping */
+#define AUDIT_MAC_MAP_DEL	1411	/* NetLabel: del LSM domain mapping */
 
 #define AUDIT_FIRST_KERN_ANOM_MSG   1700
 #define AUDIT_LAST_KERN_ANOM_MSG    1799
diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 2d72496c2029..5d6ae1b2b196 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -128,7 +128,9 @@ extern int cipso_v4_rbm_strictvalid;
 
 #ifdef CONFIG_NETLABEL
 int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
-int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head));
+int cipso_v4_doi_remove(u32 doi,
+			u32 audit_secid,
+			void (*callback) (struct rcu_head * head));
 struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi);
 int cipso_v4_doi_walk(u32 *skip_cnt,
 		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
@@ -143,6 +145,7 @@ static inline int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
 }
 
 static inline int cipso_v4_doi_remove(u32 doi,
+				    u32 audit_secid,
 				    void (*callback) (struct rcu_head * head))
 {
 	return 0;
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index 6692430063fd..190bfdbbdba6 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -96,7 +96,7 @@
 struct netlbl_dom_map;
 
 /* Domain mapping operations */
-int netlbl_domhsh_remove(const char *domain);
+int netlbl_domhsh_remove(const char *domain, u32 audit_secid);
 
 /* LSM security attributes */
 struct netlbl_lsm_cache {
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index e6ce0b3ba62a..c4e469ff842d 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -474,6 +474,7 @@ doi_add_failure_rlock:
 /**
  * cipso_v4_doi_remove - Remove an existing DOI from the CIPSO protocol engine
  * @doi: the DOI value
+ * @audit_secid: the LSM secid to use in the audit message
  * @callback: the DOI cleanup/free callback
  *
  * Description:
@@ -483,7 +484,9 @@ doi_add_failure_rlock:
  * success and negative values on failure.
  *
  */
-int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head))
+int cipso_v4_doi_remove(u32 doi,
+			u32 audit_secid,
+			void (*callback) (struct rcu_head * head))
 {
 	struct cipso_v4_doi *doi_def;
 	struct cipso_v4_domhsh_entry *dom_iter;
@@ -502,7 +505,8 @@ int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head))
 		spin_unlock(&cipso_v4_doi_list_lock);
 		list_for_each_entry_rcu(dom_iter, &doi_def->dom_list, list)
 			if (dom_iter->valid)
-				netlbl_domhsh_remove(dom_iter->domain);
+				netlbl_domhsh_remove(dom_iter->domain,
+						     audit_secid);
 		cipso_v4_cache_invalidate();
 		rcu_read_unlock();
 
diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c
index 4125a55f469f..09986ca962a6 100644
--- a/net/netlabel/netlabel_cipso_v4.c
+++ b/net/netlabel/netlabel_cipso_v4.c
@@ -32,6 +32,7 @@
 #include <linux/socket.h>
 #include <linux/string.h>
 #include <linux/skbuff.h>
+#include <linux/audit.h>
 #include <net/sock.h>
 #include <net/netlink.h>
 #include <net/genetlink.h>
@@ -162,8 +163,7 @@ static int netlbl_cipsov4_add_std(struct genl_info *info)
 	int nla_a_rem;
 	int nla_b_rem;
 
-	if (!info->attrs[NLBL_CIPSOV4_A_DOI] ||
-	    !info->attrs[NLBL_CIPSOV4_A_TAGLST] ||
+	if (!info->attrs[NLBL_CIPSOV4_A_TAGLST] ||
 	    !info->attrs[NLBL_CIPSOV4_A_MLSLVLLST])
 		return -EINVAL;
 
@@ -344,8 +344,7 @@ static int netlbl_cipsov4_add_pass(struct genl_info *info)
 	int ret_val;
 	struct cipso_v4_doi *doi_def = NULL;
 
-	if (!info->attrs[NLBL_CIPSOV4_A_DOI] ||
-	    !info->attrs[NLBL_CIPSOV4_A_TAGLST])
+	if (!info->attrs[NLBL_CIPSOV4_A_TAGLST])
 		return -EINVAL;
 
 	doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL);
@@ -381,21 +380,35 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info)
 
 {
 	int ret_val = -EINVAL;
-	u32 map_type;
+	u32 type;
+	u32 doi;
+	const char *type_str = "(unknown)";
+	struct audit_buffer *audit_buf;
 
-	if (!info->attrs[NLBL_CIPSOV4_A_MTYPE])
+	if (!info->attrs[NLBL_CIPSOV4_A_DOI] ||
+	    !info->attrs[NLBL_CIPSOV4_A_MTYPE])
 		return -EINVAL;
 
-	map_type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]);
-	switch (map_type) {
+	type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]);
+	switch (type) {
 	case CIPSO_V4_MAP_STD:
+		type_str = "std";
 		ret_val = netlbl_cipsov4_add_std(info);
 		break;
 	case CIPSO_V4_MAP_PASS:
+		type_str = "pass";
 		ret_val = netlbl_cipsov4_add_pass(info);
 		break;
 	}
 
+	if (ret_val == 0) {
+		doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
+		audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD,
+						      NETLINK_CB(skb).sid);
+		audit_log_format(audit_buf, " doi=%u type=%s", doi, type_str);
+		audit_log_end(audit_buf);
+	}
+
 	return ret_val;
 }
 
@@ -653,11 +666,21 @@ static int netlbl_cipsov4_listall(struct sk_buff *skb,
 static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
 {
 	int ret_val = -EINVAL;
-	u32 doi;
+	u32 doi = 0;
+	struct audit_buffer *audit_buf;
 
 	if (info->attrs[NLBL_CIPSOV4_A_DOI]) {
 		doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
-		ret_val = cipso_v4_doi_remove(doi, netlbl_cipsov4_doi_free);
+		ret_val = cipso_v4_doi_remove(doi,
+					      NETLINK_CB(skb).sid,
+					      netlbl_cipsov4_doi_free);
+	}
+
+	if (ret_val == 0) {
+		audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_DEL,
+						      NETLINK_CB(skb).sid);
+		audit_log_format(audit_buf, " doi=%u", doi);
+		audit_log_end(audit_buf);
 	}
 
 	return ret_val;
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index f56d7a8ac7b7..d64e2ae3b129 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -35,12 +35,14 @@
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
 #include <linux/string.h>
+#include <linux/audit.h>
 #include <net/netlabel.h>
 #include <net/cipso_ipv4.h>
 #include <asm/bug.h>
 
 #include "netlabel_mgmt.h"
 #include "netlabel_domainhash.h"
+#include "netlabel_user.h"
 
 struct netlbl_domhsh_tbl {
 	struct list_head *tbl;
@@ -186,6 +188,7 @@ int netlbl_domhsh_init(u32 size)
 /**
  * netlbl_domhsh_add - Adds a entry to the domain hash table
  * @entry: the entry to add
+ * @audit_secid: the LSM secid to use in the audit message
  *
  * Description:
  * Adds a new entry to the domain hash table and handles any updates to the
@@ -193,10 +196,12 @@ int netlbl_domhsh_init(u32 size)
  * negative on failure.
  *
  */
-int netlbl_domhsh_add(struct netlbl_dom_map *entry)
+int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid)
 {
 	int ret_val;
 	u32 bkt;
+	struct audit_buffer *audit_buf;
+	char *audit_domain;
 
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
@@ -236,6 +241,26 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry)
 		spin_unlock(&netlbl_domhsh_def_lock);
 	} else
 		ret_val = -EINVAL;
+	if (ret_val == 0) {
+		if (entry->domain != NULL)
+			audit_domain = entry->domain;
+		else
+			audit_domain = "(default)";
+		audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD,
+						      audit_secid);
+		audit_log_format(audit_buf, " domain=%s", audit_domain);
+		switch (entry->type) {
+		case NETLBL_NLTYPE_UNLABELED:
+			audit_log_format(audit_buf, " protocol=unlbl");
+			break;
+		case NETLBL_NLTYPE_CIPSOV4:
+			audit_log_format(audit_buf,
+					 " protocol=cipsov4 doi=%u",
+					 entry->type_def.cipsov4->doi);
+			break;
+		}
+		audit_log_end(audit_buf);
+	}
 	rcu_read_unlock();
 
 	if (ret_val != 0) {
@@ -254,6 +279,7 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry)
 /**
  * netlbl_domhsh_add_default - Adds the default entry to the domain hash table
  * @entry: the entry to add
+ * @audit_secid: the LSM secid to use in the audit message
  *
  * Description:
  * Adds a new default entry to the domain hash table and handles any updates
@@ -261,14 +287,15 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry)
  * negative on failure.
  *
  */
-int netlbl_domhsh_add_default(struct netlbl_dom_map *entry)
+int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, u32 audit_secid)
 {
-	return netlbl_domhsh_add(entry);
+	return netlbl_domhsh_add(entry, audit_secid);
 }
 
 /**
  * netlbl_domhsh_remove - Removes an entry from the domain hash table
  * @domain: the domain to remove
+ * @audit_secid: the LSM secid to use in the audit message
  *
  * Description:
  * Removes an entry from the domain hash table and handles any updates to the
@@ -276,10 +303,12 @@ int netlbl_domhsh_add_default(struct netlbl_dom_map *entry)
  * negative on failure.
  *
  */
-int netlbl_domhsh_remove(const char *domain)
+int netlbl_domhsh_remove(const char *domain, u32 audit_secid)
 {
 	int ret_val = -ENOENT;
 	struct netlbl_dom_map *entry;
+	struct audit_buffer *audit_buf;
+	char *audit_domain;
 
 	rcu_read_lock();
 	if (domain != NULL)
@@ -316,8 +345,18 @@ int netlbl_domhsh_remove(const char *domain)
 			ret_val = -ENOENT;
 		spin_unlock(&netlbl_domhsh_def_lock);
 	}
-	if (ret_val == 0)
+	if (ret_val == 0) {
+		if (entry->domain != NULL)
+			audit_domain = entry->domain;
+		else
+			audit_domain = "(default)";
+		audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_DEL,
+						      audit_secid);
+		audit_log_format(audit_buf, " domain=%s", audit_domain);
+		audit_log_end(audit_buf);
+
 		call_rcu(&entry->rcu, netlbl_domhsh_free_entry);
+	}
 
 remove_return:
 	rcu_read_unlock();
@@ -326,6 +365,7 @@ remove_return:
 
 /**
  * netlbl_domhsh_remove_default - Removes the default entry from the table
+ * @audit_secid: the LSM secid to use in the audit message
  *
  * Description:
  * Removes/resets the default entry for the domain hash table and handles any
@@ -333,9 +373,9 @@ remove_return:
  * success, non-zero on failure.
  *
  */
-int netlbl_domhsh_remove_default(void)
+int netlbl_domhsh_remove_default(u32 audit_secid)
 {
-	return netlbl_domhsh_remove(NULL);
+	return netlbl_domhsh_remove(NULL, audit_secid);
 }
 
 /**
diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h
index 02af72a7877c..d50f13cacdca 100644
--- a/net/netlabel/netlabel_domainhash.h
+++ b/net/netlabel/netlabel_domainhash.h
@@ -57,9 +57,9 @@ struct netlbl_dom_map {
 int netlbl_domhsh_init(u32 size);
 
 /* Manipulate the domain hash table */
-int netlbl_domhsh_add(struct netlbl_dom_map *entry);
-int netlbl_domhsh_add_default(struct netlbl_dom_map *entry);
-int netlbl_domhsh_remove_default(void);
+int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid);
+int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, u32 audit_secid);
+int netlbl_domhsh_remove_default(u32 audit_secid);
 struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain);
 int netlbl_domhsh_walk(u32 *skip_bkt,
 		     u32 *skip_chain,
diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c
index 8626c9f678eb..0ac314f18ad1 100644
--- a/net/netlabel/netlabel_mgmt.c
+++ b/net/netlabel/netlabel_mgmt.c
@@ -108,7 +108,7 @@ static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
 
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = netlbl_domhsh_add(entry);
+		ret_val = netlbl_domhsh_add(entry, NETLINK_CB(skb).sid);
 		break;
 	case NETLBL_NLTYPE_CIPSOV4:
 		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
@@ -125,7 +125,7 @@ static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
 			rcu_read_unlock();
 			goto add_failure;
 		}
-		ret_val = netlbl_domhsh_add(entry);
+		ret_val = netlbl_domhsh_add(entry, NETLINK_CB(skb).sid);
 		rcu_read_unlock();
 		break;
 	default:
@@ -161,7 +161,7 @@ static int netlbl_mgmt_remove(struct sk_buff *skb, struct genl_info *info)
 		return -EINVAL;
 
 	domain = nla_data(info->attrs[NLBL_MGMT_A_DOMAIN]);
-	return netlbl_domhsh_remove(domain);
+	return netlbl_domhsh_remove(domain, NETLINK_CB(skb).sid);
 }
 
 /**
@@ -277,7 +277,8 @@ static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = netlbl_domhsh_add_default(entry);
+		ret_val = netlbl_domhsh_add_default(entry,
+						    NETLINK_CB(skb).sid);
 		break;
 	case NETLBL_NLTYPE_CIPSOV4:
 		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
@@ -294,7 +295,8 @@ static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 			rcu_read_unlock();
 			goto adddef_failure;
 		}
-		ret_val = netlbl_domhsh_add_default(entry);
+		ret_val = netlbl_domhsh_add_default(entry,
+						    NETLINK_CB(skb).sid);
 		rcu_read_unlock();
 		break;
 	default:
@@ -322,7 +324,7 @@ adddef_failure:
  */
 static int netlbl_mgmt_removedef(struct sk_buff *skb, struct genl_info *info)
 {
-	return netlbl_domhsh_remove_default();
+	return netlbl_domhsh_remove_default(NETLINK_CB(skb).sid);
 }
 
 /**
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index 440f5c4e1e2d..ab36675fee8c 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -63,6 +63,27 @@ static struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1] = {
 	[NLBL_UNLABEL_A_ACPTFLG] = { .type = NLA_U8 },
 };
 
+/*
+ * Helper Functions
+ */
+
+/**
+ * netlbl_unlabel_acceptflg_set - Set the unlabeled accept flag
+ * @value: desired value
+ * @audit_secid: the LSM secid to use in the audit message
+ *
+ * Description:
+ * Set the value of the unlabeled accept flag to @value.
+ *
+ */
+static void netlbl_unlabel_acceptflg_set(u8 value, u32 audit_secid)
+{
+	atomic_set(&netlabel_unlabel_accept_flg, value);
+	netlbl_audit_nomsg((value ?
+			    AUDIT_MAC_UNLBL_ACCEPT : AUDIT_MAC_UNLBL_DENY),
+			   audit_secid);
+}
+
 /*
  * NetLabel Command Handlers
  */
@@ -79,18 +100,18 @@ static struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1] = {
  */
 static int netlbl_unlabel_accept(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val = -EINVAL;
 	u8 value;
 
 	if (info->attrs[NLBL_UNLABEL_A_ACPTFLG]) {
 		value = nla_get_u8(info->attrs[NLBL_UNLABEL_A_ACPTFLG]);
 		if (value == 1 || value == 0) {
-			atomic_set(&netlabel_unlabel_accept_flg, value);
-			ret_val = 0;
+			netlbl_unlabel_acceptflg_set(value,
+						     NETLINK_CB(skb).sid);
+			return 0;
 		}
 	}
 
-	return ret_val;
+	return -EINVAL;
 }
 
 /**
@@ -229,16 +250,19 @@ int netlbl_unlabel_defconf(void)
 {
 	int ret_val;
 	struct netlbl_dom_map *entry;
+	u32 secid;
+
+	security_task_getsecid(current, &secid);
 
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL)
 		return -ENOMEM;
 	entry->type = NETLBL_NLTYPE_UNLABELED;
-	ret_val = netlbl_domhsh_add_default(entry);
+	ret_val = netlbl_domhsh_add_default(entry, secid);
 	if (ret_val != 0)
 		return ret_val;
 
-	atomic_set(&netlabel_unlabel_accept_flg, 1);
+	netlbl_unlabel_acceptflg_set(1, secid);
 
 	return 0;
 }
diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c
index eeb7d768d2bb..c2343af584cb 100644
--- a/net/netlabel/netlabel_user.c
+++ b/net/netlabel/netlabel_user.c
@@ -32,6 +32,9 @@
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/socket.h>
+#include <linux/audit.h>
+#include <linux/tty.h>
+#include <linux/security.h>
 #include <net/sock.h>
 #include <net/netlink.h>
 #include <net/genetlink.h>
@@ -74,3 +77,91 @@ int netlbl_netlink_init(void)
 
 	return 0;
 }
+
+/*
+ * NetLabel Audit Functions
+ */
+
+/**
+ * netlbl_audit_start_common - Start an audit message
+ * @type: audit message type
+ * @secid: LSM context ID
+ *
+ * Description:
+ * Start an audit message using the type specified in @type and fill the audit
+ * message with some fields common to all NetLabel audit messages.  Returns
+ * a pointer to the audit buffer on success, NULL on failure.
+ *
+ */
+struct audit_buffer *netlbl_audit_start_common(int type, u32 secid)
+{
+	struct audit_context *audit_ctx = current->audit_context;
+	struct audit_buffer *audit_buf;
+	uid_t audit_loginuid;
+	const char *audit_tty;
+	char audit_comm[sizeof(current->comm)];
+	struct vm_area_struct *vma;
+	char *secctx;
+	u32 secctx_len;
+
+	audit_buf = audit_log_start(audit_ctx, GFP_ATOMIC, type);
+	if (audit_buf == NULL)
+		return NULL;
+
+	audit_loginuid = audit_get_loginuid(audit_ctx);
+	if (current->signal &&
+	    current->signal->tty &&
+	    current->signal->tty->name)
+		audit_tty = current->signal->tty->name;
+	else
+		audit_tty = "(none)";
+	get_task_comm(audit_comm, current);
+
+	audit_log_format(audit_buf,
+			 "netlabel: auid=%u uid=%u tty=%s pid=%d",
+			 audit_loginuid,
+			 current->uid,
+			 audit_tty,
+			 current->pid);
+	audit_log_format(audit_buf, " comm=");
+	audit_log_untrustedstring(audit_buf, audit_comm);
+	if (current->mm) {
+		down_read(&current->mm->mmap_sem);
+		vma = current->mm->mmap;
+		while (vma) {
+			if ((vma->vm_flags & VM_EXECUTABLE) &&
+			    vma->vm_file) {
+				audit_log_d_path(audit_buf,
+						 " exe=",
+						 vma->vm_file->f_dentry,
+						 vma->vm_file->f_vfsmnt);
+				break;
+			}
+			vma = vma->vm_next;
+		}
+		up_read(&current->mm->mmap_sem);
+	}
+
+	if (secid != 0 &&
+	    security_secid_to_secctx(secid, &secctx, &secctx_len) == 0)
+		audit_log_format(audit_buf, " subj=%s", secctx);
+
+	return audit_buf;
+}
+
+/**
+ * netlbl_audit_nomsg - Send an audit message without additional text
+ * @type: audit message type
+ * @secid: LSM context ID
+ *
+ * Description:
+ * Send an audit message with only the common NetLabel audit fields.
+ *
+ */
+void netlbl_audit_nomsg(int type, u32 secid)
+{
+	struct audit_buffer *audit_buf;
+
+	audit_buf = netlbl_audit_start_common(type, secid);
+	audit_log_end(audit_buf);
+}
diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h
index 3f9386b917df..ab840acfc964 100644
--- a/net/netlabel/netlabel_user.h
+++ b/net/netlabel/netlabel_user.h
@@ -34,6 +34,7 @@
 #include <linux/types.h>
 #include <linux/skbuff.h>
 #include <linux/capability.h>
+#include <linux/audit.h>
 #include <net/netlink.h>
 #include <net/genetlink.h>
 #include <net/netlabel.h>
@@ -75,4 +76,9 @@ static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb,
 
 int netlbl_netlink_init(void);
 
+/* NetLabel Audit Functions */
+
+struct audit_buffer *netlbl_audit_start_common(int type, u32 secid);
+void netlbl_audit_nomsg(int type, u32 secid);
+
 #endif
-- 
cgit v1.2.3


From 0891a8d706d6e6838a926b6dec42f95581747d0e Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Fri, 29 Sep 2006 01:58:34 -0700
Subject: [PATCH] __percpu_alloc_mask() has to be __always_inline in UP case

...  or we'll end up with cpu_online_map being evaluated on UP.  In
modules.  cpumask.h is very careful to avoid that, and for a very good
reason.  So should we...

PS: yes, it really triggers (on alpha).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/percpu.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 3835a9642f13..46ec72fa2c84 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -74,7 +74,7 @@ static inline int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
 	return 0;
 }
 
-static inline void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+static __always_inline void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
 {
 	return kzalloc(size, gfp);
 }
-- 
cgit v1.2.3


From e04da1dfd9041e306cb33d1b40b6005c23c5b325 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Fri, 29 Sep 2006 01:58:35 -0700
Subject: [PATCH] sys_getcpu() prototype annotated

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/syscalls.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 3f0f716225ec..2d1c3d5c83ac 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -597,6 +597,6 @@ asmlinkage long sys_get_robust_list(int pid,
 				    size_t __user *len_ptr);
 asmlinkage long sys_set_robust_list(struct robust_list_head __user *head,
 				    size_t len);
-asmlinkage long sys_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *cache);
+asmlinkage long sys_getcpu(unsigned __user *cpu, unsigned __user *node, struct getcpu_cache __user *cache);
 
 #endif
-- 
cgit v1.2.3


From f71b2f10f56802075d67c5710cd9f1816382d720 Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@austin.ibm.com>
Date: Fri, 29 Sep 2006 01:58:39 -0700
Subject: [PATCH] JBD: Make journal_brelse_array() static

It's always good to make symbols static when we can, and this also eliminates
the need to rename the function in jbd2

Suggested by Eric Sandeen.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Eric Sandeen <sandeen@sandeen.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/recovery.c   | 2 +-
 include/linux/jbd.h | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
index 445eed6ce5dc..11563fe2a52b 100644
--- a/fs/jbd/recovery.c
+++ b/fs/jbd/recovery.c
@@ -46,7 +46,7 @@ static int scan_revoke_records(journal_t *, struct buffer_head *,
 #ifdef __KERNEL__
 
 /* Release readahead buffers after use */
-void journal_brelse_array(struct buffer_head *b[], int n)
+static void journal_brelse_array(struct buffer_head *b[], int n)
 {
 	while (--n >= 0)
 		brelse (b[n]);
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index a6d9daa38c6d..fe89444b1c6f 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -977,7 +977,6 @@ extern void	   journal_write_revoke_records(journal_t *, transaction_t *);
 extern int	journal_set_revoke(journal_t *, unsigned long, tid_t);
 extern int	journal_test_revoke(journal_t *, unsigned long, tid_t);
 extern void	journal_clear_revoke(journal_t *);
-extern void	journal_brelse_array(struct buffer_head *b[], int n);
 extern void	journal_switch_revoke_table(journal_t *journal);
 
 /*
-- 
cgit v1.2.3


From 2dcea57ae19275451a756a2d5bf96b329487b0e0 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 29 Sep 2006 01:58:41 -0700
Subject: [PATCH] convert s390 page handling macros to functions

Convert s390 page handling macros to functions.  In particular this fixes a
problem with s390's SetPageUptodate macro which uses its input parameter
twice which again can cause subtle bugs.

[akpm@osdl.org: build fix]
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/appldata/appldata_base.c |  2 +-
 include/asm-s390/pgtable.h         | 84 ++++++++++++++++++--------------------
 include/linux/page-flags.h         | 11 +++--
 3 files changed, 46 insertions(+), 51 deletions(-)

(limited to 'include')

diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c
index 9a8f6ff21652..2b1e6c9a6e0e 100644
--- a/arch/s390/appldata/appldata_base.c
+++ b/arch/s390/appldata/appldata_base.c
@@ -16,7 +16,7 @@
 #include <linux/errno.h>
 #include <linux/interrupt.h>
 #include <linux/proc_fs.h>
-#include <linux/page-flags.h>
+#include <linux/mm.h>
 #include <linux/swap.h>
 #include <linux/pagemap.h>
 #include <linux/sysctl.h>
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index 83425cdefc91..ecdff13b2505 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -31,9 +31,9 @@
  * the S390 page table tree.
  */
 #ifndef __ASSEMBLY__
+#include <linux/mm_types.h>
 #include <asm/bug.h>
 #include <asm/processor.h>
-#include <linux/threads.h>
 
 struct vm_area_struct; /* forward declaration (include/linux/mm.h) */
 struct mm_struct;
@@ -597,31 +597,31 @@ ptep_establish(struct vm_area_struct *vma,
  * should therefore only be called if it is not mapped in any
  * address space.
  */
-#define page_test_and_clear_dirty(_page)				  \
-({									  \
-	struct page *__page = (_page);					  \
-	unsigned long __physpage = __pa((__page-mem_map) << PAGE_SHIFT);  \
-	int __skey = page_get_storage_key(__physpage);			  \
-	if (__skey & _PAGE_CHANGED)					  \
-		page_set_storage_key(__physpage, __skey & ~_PAGE_CHANGED);\
-	(__skey & _PAGE_CHANGED);					  \
-})
+static inline int page_test_and_clear_dirty(struct page *page)
+{
+	unsigned long physpage = __pa((page - mem_map) << PAGE_SHIFT);
+	int skey = page_get_storage_key(physpage);
+
+	if (skey & _PAGE_CHANGED)
+		page_set_storage_key(physpage, skey & ~_PAGE_CHANGED);
+	return skey & _PAGE_CHANGED;
+}
 
 /*
  * Test and clear referenced bit in storage key.
  */
-#define page_test_and_clear_young(page)					\
-({									\
-	struct page *__page = (page);					\
-	unsigned long __physpage = __pa((__page-mem_map) << PAGE_SHIFT);\
-	int __ccode;							\
-	asm volatile(							\
-		"	rrbe	0,%1\n"					\
-		"	ipm	%0\n"					\
-		"	srl	%0,28\n"				\
-		: "=d" (__ccode) : "a" (__physpage) : "cc");		\
-	(__ccode & 2);							\
-})
+static inline int page_test_and_clear_young(struct page *page)
+{
+	unsigned long physpage = __pa((page - mem_map) << PAGE_SHIFT);
+	int ccode;
+
+	asm volatile (
+		"rrbe 0,%1\n"
+		"ipm  %0\n"
+		"srl  %0,28\n"
+		: "=d" (ccode) : "a" (physpage) : "cc" );
+	return ccode & 2;
+}
 
 /*
  * Conversion functions: convert a page and protection to a page entry,
@@ -634,32 +634,28 @@ static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
 	return __pte;
 }
 
-#define mk_pte(pg, pgprot)                                                \
-({                                                                        \
-	struct page *__page = (pg);                                       \
-	pgprot_t __pgprot = (pgprot);					  \
-	unsigned long __physpage = __pa((__page-mem_map) << PAGE_SHIFT);  \
-	pte_t __pte = mk_pte_phys(__physpage, __pgprot);                  \
-	__pte;                                                            \
-})
+static inline pte_t mk_pte(struct page *page, pgprot_t pgprot)
+{
+	unsigned long physpage = __pa((page - mem_map) << PAGE_SHIFT);
 
-#define pfn_pte(pfn, pgprot)                                              \
-({                                                                        \
-	pgprot_t __pgprot = (pgprot);					  \
-	unsigned long __physpage = __pa((pfn) << PAGE_SHIFT);             \
-	pte_t __pte = mk_pte_phys(__physpage, __pgprot);                  \
-	__pte;                                                            \
-})
+	return mk_pte_phys(physpage, pgprot);
+}
+
+static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
+{
+	unsigned long physpage = __pa((pfn) << PAGE_SHIFT);
+
+	return mk_pte_phys(physpage, pgprot);
+}
 
 #ifdef __s390x__
 
-#define pfn_pmd(pfn, pgprot)                                              \
-({                                                                        \
-	pgprot_t __pgprot = (pgprot);                                     \
-	unsigned long __physpage = __pa((pfn) << PAGE_SHIFT);             \
-	pmd_t __pmd = __pmd(__physpage + pgprot_val(__pgprot));           \
-	__pmd;                                                            \
-})
+static inline pmd_t pfn_pmd(unsigned long pfn, pgprot_t pgprot)
+{
+	unsigned long physpage = __pa((pfn) << PAGE_SHIFT);
+
+	return __pmd(physpage + pgprot_val(pgprot));
+}
 
 #endif /* __s390x__ */
 
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 9d7921dd50f0..4830a3bedfb2 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -128,12 +128,11 @@
 
 #define PageUptodate(page)	test_bit(PG_uptodate, &(page)->flags)
 #ifdef CONFIG_S390
-#define SetPageUptodate(_page) \
-	do {								      \
-		struct page *__page = (_page);				      \
-		if (!test_and_set_bit(PG_uptodate, &__page->flags))	      \
-			page_test_and_clear_dirty(_page);		      \
-	} while (0)
+static inline void SetPageUptodate(struct page *page)
+{
+	if (!test_and_set_bit(PG_uptodate, &page->flags))
+		page_test_and_clear_dirty(page);
+}
 #else
 #define SetPageUptodate(page)	set_bit(PG_uptodate, &(page)->flags)
 #endif
-- 
cgit v1.2.3


From 3b08606dc2991bcdab14139efd9ed9d492f5f901 Mon Sep 17 00:00:00 2001
From: keith mannthey <kmannth@us.ibm.com>
Date: Fri, 29 Sep 2006 01:58:46 -0700
Subject: [PATCH] convert i386 Summit subarch to use SRAT info for
 apicid_to_node calls

Convert the i386 summit subarch apicid_to_node to use node information
provided by the SRAT.  It was discussed a little on LKML a few weeks ago
and was seen as an acceptable fix.  The current way of obtaining the nodeid

 static inline int apicid_to_node(int logical_apicid)
 {
   return logical_apicid >> 5;
 }

is just not correct for all summit systems/bios.  Assuming the apicid
matches the Linux node number require a leap of faith that the bios mapped
out the apicids a set way.  Modern summit HW (IBM x460) does not layout its
bios in the manner for various reasons and is unable to boot i386 numa.

The best way to get the correct apicid to node information is from the SRAT
table during boot.  It lays out what apicid belongs to what node.  I use
this information to create a table for use at run time.

Signed-off-by: Keith Mannthey <kmannth@us.ibm.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smpboot.c               | 5 ++++-
 arch/i386/kernel/srat.c                  | 7 +++++++
 include/asm-i386/mach-summit/mach_apic.h | 2 +-
 include/asm-i386/smp.h                   | 2 ++
 4 files changed, 14 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 020d873b7d21..82b26d5ce476 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -102,6 +102,8 @@ u8 x86_cpu_to_apicid[NR_CPUS] __read_mostly =
 			{ [0 ... NR_CPUS-1] = 0xff };
 EXPORT_SYMBOL(x86_cpu_to_apicid);
 
+u8 apicid_2_node[MAX_APICID];
+
 /*
  * Trampoline 80x86 program as an array.
  */
@@ -645,7 +647,7 @@ static void map_cpu_to_logical_apicid(void)
 {
 	int cpu = smp_processor_id();
 	int apicid = logical_smp_processor_id();
-	int node = apicid_to_node(apicid);
+	int node = apicid_to_node(hard_smp_processor_id());
 
 	if (!node_online(node))
 		node = first_online_node;
@@ -954,6 +956,7 @@ static int __devinit do_boot_cpu(int apicid, int cpu)
 
 	irq_ctx_init(cpu);
 
+	x86_cpu_to_apicid[cpu] = apicid;
 	/*
 	 * This grunge runs the startup process for
 	 * the targeted processor.
diff --git a/arch/i386/kernel/srat.c b/arch/i386/kernel/srat.c
index 32413122c4c2..f7e735c077c3 100644
--- a/arch/i386/kernel/srat.c
+++ b/arch/i386/kernel/srat.c
@@ -30,6 +30,7 @@
 #include <linux/nodemask.h>
 #include <asm/srat.h>
 #include <asm/topology.h>
+#include <asm/smp.h>
 
 /*
  * proximity macros and definitions
@@ -54,6 +55,7 @@ struct node_memory_chunk_s {
 static struct node_memory_chunk_s node_memory_chunk[MAXCHUNKS];
 
 static int num_memory_chunks;		/* total number of memory chunks */
+static u8 __initdata apicid_to_pxm[MAX_APICID];
 
 extern void * boot_ioremap(unsigned long, unsigned long);
 
@@ -69,6 +71,8 @@ static void __init parse_cpu_affinity_structure(char *p)
 	/* mark this node as "seen" in node bitmap */
 	BMAP_SET(pxm_bitmap, cpu_affinity->proximity_domain);
 
+	apicid_to_pxm[cpu_affinity->apic_id] = cpu_affinity->proximity_domain;
+
 	printk("CPU 0x%02X in proximity domain 0x%02X\n",
 		cpu_affinity->apic_id, cpu_affinity->proximity_domain);
 }
@@ -235,6 +239,9 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp)
 	printk("Number of logical nodes in system = %d\n", num_online_nodes());
 	printk("Number of memory chunks in system = %d\n", num_memory_chunks);
 
+	for (i = 0; i < MAX_APICID; i++)
+		apicid_2_node[i] = pxm_to_node(apicid_to_pxm[i]);
+
 	for (j = 0; j < num_memory_chunks; j++){
 		struct node_memory_chunk_s * chunk = &node_memory_chunk[j];
 		printk("chunk %d nid %d start_pfn %08lx end_pfn %08lx\n",
diff --git a/include/asm-i386/mach-summit/mach_apic.h b/include/asm-i386/mach-summit/mach_apic.h
index a81b05961595..254a0fe01c6a 100644
--- a/include/asm-i386/mach-summit/mach_apic.h
+++ b/include/asm-i386/mach-summit/mach_apic.h
@@ -88,7 +88,7 @@ static inline void clustered_apic_check(void)
 
 static inline int apicid_to_node(int logical_apicid)
 {
-	return logical_apicid >> 5;          /* 2 clusterids per CEC */
+	return apicid_2_node[logical_apicid];
 }
 
 /* Mapping from cpu number to logical apicid */
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 32ac8c91d5c5..915c26a31b79 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -46,6 +46,8 @@ extern u8 x86_cpu_to_apicid[];
 
 #define cpu_physical_id(cpu)	x86_cpu_to_apicid[cpu]
 
+extern u8 apicid_2_node[];
+
 #ifdef CONFIG_HOTPLUG_CPU
 extern void cpu_exit_clear(void);
 extern void cpu_uninit(void);
-- 
cgit v1.2.3


From 5906e4171ad61ce68de95e51b773146707671f80 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Fri, 29 Sep 2006 01:58:54 -0700
Subject: [PATCH] uml: remove pte_mkexec

Andi is making pte_mkexec go away, and UML had one of the last uses.

This removes the use and the definition.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/skas/mmu.c | 6 ++++--
 include/asm-um/pgtable.h  | 6 ------
 2 files changed, 4 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c
index 79c22707a637..4cd2ff546ef6 100644
--- a/arch/um/kernel/skas/mmu.c
+++ b/arch/um/kernel/skas/mmu.c
@@ -61,8 +61,10 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc,
 #endif
 
 	*pte = mk_pte(virt_to_page(kernel), __pgprot(_PAGE_PRESENT));
-	*pte = pte_mkexec(*pte);
-	*pte = pte_wrprotect(*pte);
+	/* This is wrong for the code page, but it doesn't matter since the
+	 * stub is mapped by hand with the correct permissions.
+	 */
+	*pte = pte_mkwrite(*pte);
 	return(0);
 
  out_pmd:
diff --git a/include/asm-um/pgtable.h b/include/asm-um/pgtable.h
index 4862daf8b906..188f72621776 100644
--- a/include/asm-um/pgtable.h
+++ b/include/asm-um/pgtable.h
@@ -274,12 +274,6 @@ static inline pte_t pte_mkread(pte_t pte)
 	return(pte_mknewprot(pte)); 
 }
 
-static inline pte_t pte_mkexec(pte_t pte)
-{ 
-	pte_set_bits(pte, _PAGE_USER);
-	return(pte_mknewprot(pte)); 
-}
-
 static inline pte_t pte_mkdirty(pte_t pte)
 { 
 	pte_set_bits(pte, _PAGE_DIRTY);
-- 
cgit v1.2.3


From 199a9afc3dbe98c35326f1d3907ab94dae953a6e Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Fri, 29 Sep 2006 01:59:00 -0700
Subject: [PATCH] Debug variants of linked list macros

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/list.h | 15 ++++++++++
 lib/Kconfig.debug    |  9 ++++++
 lib/Makefile         |  1 +
 lib/list_debug.c     | 77 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 102 insertions(+)
 create mode 100644 lib/list_debug.c

(limited to 'include')

diff --git a/include/linux/list.h b/include/linux/list.h
index 65a5b5ceda49..a9c90287c0ff 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -39,6 +39,7 @@ static inline void INIT_LIST_HEAD(struct list_head *list)
  * This is only for internal list manipulation where we know
  * the prev/next entries already!
  */
+#ifndef CONFIG_DEBUG_LIST
 static inline void __list_add(struct list_head *new,
 			      struct list_head *prev,
 			      struct list_head *next)
@@ -48,6 +49,11 @@ static inline void __list_add(struct list_head *new,
 	new->prev = prev;
 	prev->next = new;
 }
+#else
+extern void __list_add(struct list_head *new,
+			      struct list_head *prev,
+			      struct list_head *next);
+#endif
 
 /**
  * list_add - add a new entry
@@ -57,10 +63,15 @@ static inline void __list_add(struct list_head *new,
  * Insert a new entry after the specified head.
  * This is good for implementing stacks.
  */
+#ifndef CONFIG_DEBUG_LIST
 static inline void list_add(struct list_head *new, struct list_head *head)
 {
 	__list_add(new, head, head->next);
 }
+#else
+extern void list_add(struct list_head *new, struct list_head *head);
+#endif
+
 
 /**
  * list_add_tail - add a new entry
@@ -153,12 +164,16 @@ static inline void __list_del(struct list_head * prev, struct list_head * next)
  * Note: list_empty on entry does not return true after this, the entry is
  * in an undefined state.
  */
+#ifndef CONFIG_DEBUG_LIST
 static inline void list_del(struct list_head *entry)
 {
 	__list_del(entry->prev, entry->next);
 	entry->next = LIST_POISON1;
 	entry->prev = LIST_POISON2;
 }
+#else
+extern void list_del(struct list_head *entry);
+#endif
 
 /**
  * list_del_rcu - deletes entry from list without re-initialization
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index b0f5ca72599f..f9ae75cc0145 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -320,6 +320,15 @@ config DEBUG_VM
 
 	  If unsure, say N.
 
+config DEBUG_LIST
+	bool "Debug linked list manipulation"
+	depends on DEBUG_KERNEL
+	help
+	  Enable this to turn on extended checks in the linked-list
+	  walking routines.
+
+	  If unsure, say N.
+
 config FRAME_POINTER
 	bool "Compile the kernel with frame pointers"
 	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390 || AVR32 || SUPERH)
diff --git a/lib/Makefile b/lib/Makefile
index ef1d37afbbb6..402762fead70 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -28,6 +28,7 @@ lib-$(CONFIG_GENERIC_HWEIGHT) += hweight.o
 obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o
 obj-$(CONFIG_PLIST) += plist.o
 obj-$(CONFIG_DEBUG_PREEMPT) += smp_processor_id.o
+obj-$(CONFIG_DEBUG_LIST) += list_debug.o
 
 ifneq ($(CONFIG_HAVE_DEC_LOCK),y)
   lib-y += dec_and_lock.o
diff --git a/lib/list_debug.c b/lib/list_debug.c
new file mode 100644
index 000000000000..1aae85cef92c
--- /dev/null
+++ b/lib/list_debug.c
@@ -0,0 +1,77 @@
+/*
+ * Copyright 2006, Red Hat, Inc., Dave Jones
+ * Released under the General Public License (GPL).
+ *
+ * This file contains the linked list implementations for
+ * DEBUG_LIST.
+ */
+
+#include <linux/module.h>
+#include <linux/list.h>
+
+/*
+ * Insert a new entry between two known consecutive entries.
+ *
+ * This is only for internal list manipulation where we know
+ * the prev/next entries already!
+ */
+
+void __list_add(struct list_head *new,
+			      struct list_head *prev,
+			      struct list_head *next)
+{
+	if (unlikely(next->prev != prev)) {
+		printk(KERN_ERR "list_add corruption. next->prev should be %p, but was %p\n",
+			prev, next->prev);
+		BUG();
+	}
+	if (unlikely(prev->next != next)) {
+		printk(KERN_ERR "list_add corruption. prev->next should be %p, but was %p\n",
+			next, prev->next);
+		BUG();
+	}
+	next->prev = new;
+	new->next = next;
+	new->prev = prev;
+	prev->next = new;
+}
+EXPORT_SYMBOL(__list_add);
+
+/**
+ * list_add - add a new entry
+ * @new: new entry to be added
+ * @head: list head to add it after
+ *
+ * Insert a new entry after the specified head.
+ * This is good for implementing stacks.
+ */
+void list_add(struct list_head *new, struct list_head *head)
+{
+	__list_add(new, head, head->next);
+}
+EXPORT_SYMBOL(list_add);
+
+/**
+ * list_del - deletes entry from list.
+ * @entry: the element to delete from the list.
+ * Note: list_empty on entry does not return true after this, the entry is
+ * in an undefined state.
+ */
+void list_del(struct list_head *entry)
+{
+	if (unlikely(entry->prev->next != entry)) {
+		printk(KERN_ERR "list_del corruption. prev->next should be %p, but was %p\n",
+			entry, entry->prev->next);
+		BUG();
+	}
+	if (unlikely(entry->next->prev != entry)) {
+		printk(KERN_ERR "list_del corruption. next->prev should be %p, but was %p\n",
+			entry, entry->next->prev);
+		BUG();
+	}
+	__list_del(entry->prev, entry->next);
+	entry->next = LIST_POISON1;
+	entry->prev = LIST_POISON2;
+}
+EXPORT_SYMBOL(list_del);
+
-- 
cgit v1.2.3


From 9938406ab6b2558d60c0c7200cc8e12f1ea7104a Mon Sep 17 00:00:00 2001
From: Michal Schmidt <xschmi00@stud.feec.vutbr.cz>
Date: Fri, 29 Sep 2006 01:59:03 -0700
Subject: [PATCH] Make touch_nmi_watchdog imply touch_softlockup_watchdog on
 all archs

touch_nmi_watchdog() calls touch_softlockup_watchdog() on both
architectures that implement it (i386 and x86_64).  On other architectures
it does nothing at all.  touch_nmi_watchdog() should imply
touch_softlockup_watchdog() on all architectures.  Suggested by Andi Kleen.

[heiko.carstens@de.ibm.com: s390 fix]
Signed-off-by: Michal Schmidt <xschmi00@stud.feec.vutbr.cz>
Cc: Andi Kleen <ak@muc.de>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michal Schmidt <xschmi00@stud.feec.vutbr.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-s390/irq.h | 3 ---
 include/linux/nmi.h    | 3 ++-
 2 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-s390/irq.h b/include/asm-s390/irq.h
index bd1a721f7aa2..7da991a858f8 100644
--- a/include/asm-s390/irq.h
+++ b/include/asm-s390/irq.h
@@ -19,8 +19,5 @@ enum interruption_class {
 	NR_IRQS,
 };
 
-#define touch_nmi_watchdog() do { } while(0)
-
 #endif /* __KERNEL__ */
 #endif
-
diff --git a/include/linux/nmi.h b/include/linux/nmi.h
index c8f4d2f627d7..e16904e28c3a 100644
--- a/include/linux/nmi.h
+++ b/include/linux/nmi.h
@@ -4,6 +4,7 @@
 #ifndef LINUX_NMI_H
 #define LINUX_NMI_H
 
+#include <linux/sched.h>
 #include <asm/irq.h>
 
 /**
@@ -16,7 +17,7 @@
 #ifdef ARCH_HAS_NMI_WATCHDOG
 extern void touch_nmi_watchdog(void);
 #else
-# define touch_nmi_watchdog() do { } while(0)
+# define touch_nmi_watchdog() touch_softlockup_watchdog()
 #endif
 
 #endif
-- 
cgit v1.2.3


From 684f978347deb42d180373ac4c427f82ef963171 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 29 Sep 2006 01:59:06 -0700
Subject: [PATCH] Let WARN_ON/WARN_ON_ONCE return the condition

Letting WARN_ON/WARN_ON_ONCE return the condition means that you could do

if (WARN_ON(blah)) {
	handle_impossible_case
}

Rather than

if (unlikely(blah)) {
	WARN_ON(1)
	handle_impossible_case
}

I checked all the newly added WARN_ON_ONCE users and none of them test the
return status so we can still change it.

[akpm@osdl.org: warning fix]
[akpm@osdl.org: build fix]
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/bug.h | 32 ++++++++++++++++----------------
 include/asm-powerpc/bug.h | 12 +++++++-----
 2 files changed, 23 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index 8ceab7bcd8b4..a5250895155e 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -16,12 +16,15 @@
 #endif
 
 #ifndef HAVE_ARCH_WARN_ON
-#define WARN_ON(condition) do { \
-	if (unlikely((condition)!=0)) { \
-		printk("BUG: warning at %s:%d/%s()\n", __FILE__, __LINE__, __FUNCTION__); \
-		dump_stack(); \
-	} \
-} while (0)
+#define WARN_ON(condition) ({						\
+	typeof(condition) __ret_warn_on = (condition);			\
+	if (unlikely(__ret_warn_on)) {					\
+		printk("BUG: warning at %s:%d/%s()\n", __FILE__,	\
+			__LINE__, __FUNCTION__);			\
+		dump_stack();						\
+	}								\
+	unlikely(__ret_warn_on);					\
+})
 #endif
 
 #else /* !CONFIG_BUG */
@@ -34,21 +37,18 @@
 #endif
 
 #ifndef HAVE_ARCH_WARN_ON
-#define WARN_ON(condition) do { if (condition) ; } while(0)
+#define WARN_ON(condition) unlikely((condition))
 #endif
 #endif
 
-#define WARN_ON_ONCE(condition)				\
-({							\
+#define WARN_ON_ONCE(condition)	({			\
 	static int __warn_once = 1;			\
-	int __ret = 0;					\
+	typeof(condition) __ret_warn_once = (condition);\
 							\
-	if (unlikely((condition) && __warn_once)) {	\
-		__warn_once = 0;			\
-		WARN_ON(1);				\
-		__ret = 1;				\
-	}						\
-	__ret;						\
+	if (likely(__warn_once))			\
+		if (WARN_ON(__ret_warn_once)) 		\
+			__warn_once = 0;		\
+	unlikely(__ret_warn_once);			\
 })
 
 #ifdef CONFIG_SMP
diff --git a/include/asm-powerpc/bug.h b/include/asm-powerpc/bug.h
index f44b529e3298..978b2c7e84ea 100644
--- a/include/asm-powerpc/bug.h
+++ b/include/asm-powerpc/bug.h
@@ -70,9 +70,10 @@ struct bug_entry *find_bug(unsigned long bugaddr);
 		    "i" (__FILE__), "i" (__FUNCTION__));	\
 } while (0)
 
-#define WARN_ON(x) do {						\
-	if (__builtin_constant_p(x)) {				\
-		if (x)						\
+#define WARN_ON(x) ({						\
+	typeof(x) __ret_warn_on = (x);				\
+	if (__builtin_constant_p(__ret_warn_on)) {		\
+		if (__ret_warn_on)				\
 			__WARN();				\
 	} else {						\
 		__asm__ __volatile__(				\
@@ -80,11 +81,12 @@ struct bug_entry *find_bug(unsigned long bugaddr);
 		".section __bug_table,\"a\"\n"			\
 		"\t"PPC_LONG"	1b,%1,%2,%3\n"			\
 		".previous"					\
-		: : "r" ((long)(x)),				\
+		: : "r" (__ret_warn_on),			\
 		    "i" (__LINE__ + BUG_WARNING_TRAP),		\
 		    "i" (__FILE__), "i" (__FUNCTION__));	\
 	}							\
-} while (0)
+	unlikely(__ret_warn_on);				\
+})
 
 #define HAVE_ARCH_BUG
 #define HAVE_ARCH_BUG_ON
-- 
cgit v1.2.3


From 58012cd788443b9d144bd7c72260a84b6b30f45d Mon Sep 17 00:00:00 2001
From: Chris Boot <bootc@bootc.net>
Date: Fri, 29 Sep 2006 01:59:07 -0700
Subject: [PATCH] scx200_gpio export cleanups

Use EXPORT_SYMBOL_GPL for new symbols, and declare the struct in the header
file for access by other modules.

Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/scx200_gpio.c  | 2 +-
 include/linux/scx200_gpio.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/char/scx200_gpio.c b/drivers/char/scx200_gpio.c
index b1f88c66b2b5..99e5272e3c53 100644
--- a/drivers/char/scx200_gpio.c
+++ b/drivers/char/scx200_gpio.c
@@ -44,7 +44,7 @@ struct nsc_gpio_ops scx200_gpio_ops = {
 	.gpio_change	= scx200_gpio_change,
 	.gpio_current	= scx200_gpio_current
 };
-EXPORT_SYMBOL(scx200_gpio_ops);
+EXPORT_SYMBOL_GPL(scx200_gpio_ops);
 
 static int scx200_gpio_open(struct inode *inode, struct file *file)
 {
diff --git a/include/linux/scx200_gpio.h b/include/linux/scx200_gpio.h
index 90dd069cc145..1a82d30c4b17 100644
--- a/include/linux/scx200_gpio.h
+++ b/include/linux/scx200_gpio.h
@@ -4,6 +4,7 @@ u32 scx200_gpio_configure(unsigned index, u32 set, u32 clear);
 
 extern unsigned scx200_gpio_base;
 extern long scx200_gpio_shadow[2];
+extern struct nsc_gpio_ops scx200_gpio_ops;
 
 #define scx200_gpio_present() (scx200_gpio_base!=0)
 
-- 
cgit v1.2.3


From 6c9979185c7ef4feeb7f8d29be032b8f032a1838 Mon Sep 17 00:00:00 2001
From: "Serge E. Hallyn" <serue@us.ibm.com>
Date: Fri, 29 Sep 2006 01:59:11 -0700
Subject: [PATCH] kthread: convert loop.c to kthread

Convert loop.c from the deprecated kernel_thread to kthread.  This patch
simplifies the code quite a bit and passes similar testing to the previous
submission on both emulated x86 and s390.

Changes since last submission:
	switched to using a rather simple loop based on
	wait_event_interruptible.

Signed-off-by: Serge E. Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/loop.c | 69 ++++++++++++++++++----------------------------------
 include/linux/loop.h |  5 ++--
 2 files changed, 26 insertions(+), 48 deletions(-)

(limited to 'include')

diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index c774121684d7..e87b88731adc 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -72,6 +72,7 @@
 #include <linux/completion.h>
 #include <linux/highmem.h>
 #include <linux/gfp.h>
+#include <linux/kthread.h>
 
 #include <asm/uaccess.h>
 
@@ -522,15 +523,12 @@ static int loop_make_request(request_queue_t *q, struct bio *old_bio)
 		goto out;
 	if (unlikely(rw == WRITE && (lo->lo_flags & LO_FLAGS_READ_ONLY)))
 		goto out;
-	lo->lo_pending++;
 	loop_add_bio(lo, old_bio);
+	wake_up(&lo->lo_event);
 	spin_unlock_irq(&lo->lo_lock);
-	complete(&lo->lo_bh_done);
 	return 0;
 
 out:
-	if (lo->lo_pending == 0)
-		complete(&lo->lo_bh_done);
 	spin_unlock_irq(&lo->lo_lock);
 	bio_io_error(old_bio, old_bio->bi_size);
 	return 0;
@@ -570,14 +568,18 @@ static inline void loop_handle_bio(struct loop_device *lo, struct bio *bio)
  * to avoid blocking in our make_request_fn. it also does loop decrypting
  * on reads for block backed loop, as that is too heavy to do from
  * b_end_io context where irqs may be disabled.
+ *
+ * Loop explanation:  loop_clr_fd() sets lo_state to Lo_rundown before
+ * calling kthread_stop().  Therefore once kthread_should_stop() is
+ * true, make_request will not place any more requests.  Therefore
+ * once kthread_should_stop() is true and lo_bio is NULL, we are
+ * done with the loop.
  */
 static int loop_thread(void *data)
 {
 	struct loop_device *lo = data;
 	struct bio *bio;
 
-	daemonize("loop%d", lo->lo_number);
-
 	/*
 	 * loop can be used in an encrypted device,
 	 * hence, it mustn't be stopped at all
@@ -587,47 +589,21 @@ static int loop_thread(void *data)
 
 	set_user_nice(current, -20);
 
-	lo->lo_state = Lo_bound;
-	lo->lo_pending = 1;
-
-	/*
-	 * complete it, we are running
-	 */
-	complete(&lo->lo_done);
+	while (!kthread_should_stop() || lo->lo_bio) {
 
-	for (;;) {
-		int pending;
+		wait_event_interruptible(lo->lo_event,
+				lo->lo_bio || kthread_should_stop());
 
-		if (wait_for_completion_interruptible(&lo->lo_bh_done))
+		if (!lo->lo_bio)
 			continue;
-
 		spin_lock_irq(&lo->lo_lock);
-
-		/*
-		 * could be completed because of tear-down, not pending work
-		 */
-		if (unlikely(!lo->lo_pending)) {
-			spin_unlock_irq(&lo->lo_lock);
-			break;
-		}
-
 		bio = loop_get_bio(lo);
-		lo->lo_pending--;
-		pending = lo->lo_pending;
 		spin_unlock_irq(&lo->lo_lock);
 
 		BUG_ON(!bio);
 		loop_handle_bio(lo, bio);
-
-		/*
-		 * upped both for pending work and tear-down, lo_pending
-		 * will hit zero then
-		 */
-		if (unlikely(!pending))
-			break;
 	}
 
-	complete(&lo->lo_done);
 	return 0;
 }
 
@@ -840,10 +816,15 @@ static int loop_set_fd(struct loop_device *lo, struct file *lo_file,
 
 	set_blocksize(bdev, lo_blocksize);
 
-	error = kernel_thread(loop_thread, lo, CLONE_KERNEL);
-	if (error < 0)
+	lo->lo_thread = kthread_create(loop_thread, lo, "loop%d",
+						lo->lo_number);
+	if (IS_ERR(lo->lo_thread)) {
+		error = PTR_ERR(lo->lo_thread);
+		lo->lo_thread = NULL;
 		goto out_putf;
-	wait_for_completion(&lo->lo_done);
+	}
+	lo->lo_state = Lo_bound;
+	wake_up_process(lo->lo_thread);
 	return 0;
 
  out_putf:
@@ -907,12 +888,9 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev)
 
 	spin_lock_irq(&lo->lo_lock);
 	lo->lo_state = Lo_rundown;
-	lo->lo_pending--;
-	if (!lo->lo_pending)
-		complete(&lo->lo_bh_done);
 	spin_unlock_irq(&lo->lo_lock);
 
-	wait_for_completion(&lo->lo_done);
+	kthread_stop(lo->lo_thread);
 
 	lo->lo_backing_file = NULL;
 
@@ -925,6 +903,7 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev)
 	lo->lo_sizelimit = 0;
 	lo->lo_encrypt_key_size = 0;
 	lo->lo_flags = 0;
+	lo->lo_thread = NULL;
 	memset(lo->lo_encrypt_key, 0, LO_KEY_SIZE);
 	memset(lo->lo_crypt_name, 0, LO_NAME_SIZE);
 	memset(lo->lo_file_name, 0, LO_NAME_SIZE);
@@ -1287,9 +1266,9 @@ static int __init loop_init(void)
 		if (!lo->lo_queue)
 			goto out_mem4;
 		mutex_init(&lo->lo_ctl_mutex);
-		init_completion(&lo->lo_done);
-		init_completion(&lo->lo_bh_done);
 		lo->lo_number = i;
+		lo->lo_thread = NULL;
+		init_waitqueue_head(&lo->lo_event);
 		spin_lock_init(&lo->lo_lock);
 		disk->major = LOOP_MAJOR;
 		disk->first_minor = i;
diff --git a/include/linux/loop.h b/include/linux/loop.h
index e76c7611d6cc..191a595055f0 100644
--- a/include/linux/loop.h
+++ b/include/linux/loop.h
@@ -59,10 +59,9 @@ struct loop_device {
 	struct bio 		*lo_bio;
 	struct bio		*lo_biotail;
 	int			lo_state;
-	struct completion	lo_done;
-	struct completion	lo_bh_done;
 	struct mutex		lo_ctl_mutex;
-	int			lo_pending;
+	struct task_struct	*lo_thread;
+	wait_queue_head_t	lo_event;
 
 	request_queue_t		*lo_queue;
 };
-- 
cgit v1.2.3


From ad4e09b16ad361c15bd7186dcd118cb901089b97 Mon Sep 17 00:00:00 2001
From: Komal Shah <komal_shah802003@yahoo.com>
Date: Fri, 29 Sep 2006 01:59:19 -0700
Subject: [PATCH] OMAP: Add keypad driver

This patch adds support for keypad driver running on different TI
OMAP(http://www.ti.com/omap) processor based boards like OSK, H2, H3, H4,
Persuas and Nokia 770.

Signed-off-by: Komal Shah <komal_shah802003@yahoo.com>
Acked-by: Dmitry Torokhov <dtor@mail.ru>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/input/keyboard/Kconfig       |   9 +
 drivers/input/keyboard/Makefile      |   1 +
 drivers/input/keyboard/omap-keypad.c | 492 +++++++++++++++++++++++++++++++++++
 include/asm-arm/arch-omap/keypad.h   |   3 +
 4 files changed, 505 insertions(+)
 create mode 100644 drivers/input/keyboard/omap-keypad.c

(limited to 'include')

diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
index a9dda56f62c4..83eac3a66bc8 100644
--- a/drivers/input/keyboard/Kconfig
+++ b/drivers/input/keyboard/Kconfig
@@ -183,4 +183,13 @@ config KEYBOARD_HIL
 	  This driver implements support for HIL-keyboards attached
 	  to your machine, so normally you should say Y here.
 
+config KEYBOARD_OMAP
+	tristate "TI OMAP keypad support"
+	depends on (ARCH_OMAP1 || ARCH_OMAP2)
+	help
+	  Say Y here if you want to use the OMAP keypad.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called omap-keypad.
+
 endif
diff --git a/drivers/input/keyboard/Makefile b/drivers/input/keyboard/Makefile
index 2708167ba175..b265391f1f10 100644
--- a/drivers/input/keyboard/Makefile
+++ b/drivers/input/keyboard/Makefile
@@ -15,4 +15,5 @@ obj-$(CONFIG_KEYBOARD_CORGI)		+= corgikbd.o
 obj-$(CONFIG_KEYBOARD_SPITZ)		+= spitzkbd.o
 obj-$(CONFIG_KEYBOARD_HIL)		+= hil_kbd.o
 obj-$(CONFIG_KEYBOARD_HIL_OLD)		+= hilkbd.o
+obj-$(CONFIG_KEYBOARD_OMAP)             += omap-keypad.o
 
diff --git a/drivers/input/keyboard/omap-keypad.c b/drivers/input/keyboard/omap-keypad.c
new file mode 100644
index 000000000000..d436287d1d2e
--- /dev/null
+++ b/drivers/input/keyboard/omap-keypad.c
@@ -0,0 +1,492 @@
+/*
+ * linux/drivers/input/keyboard/omap-keypad.c
+ *
+ * OMAP Keypad Driver
+ *
+ * Copyright (C) 2003 Nokia Corporation
+ * Written by Timo Teräs <ext-timo.teras@nokia.com>
+ *
+ * Added support for H2 & H3 Keypad
+ * Copyright (C) 2004 Texas Instruments
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/types.h>
+#include <linux/input.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/platform_device.h>
+#include <linux/mutex.h>
+#include <linux/errno.h>
+#include <asm/arch/gpio.h>
+#include <asm/arch/keypad.h>
+#include <asm/arch/menelaus.h>
+#include <asm/irq.h>
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/mach-types.h>
+#include <asm/arch/mux.h>
+
+#undef NEW_BOARD_LEARNING_MODE
+
+static void omap_kp_tasklet(unsigned long);
+static void omap_kp_timer(unsigned long);
+
+static unsigned char keypad_state[8];
+static DEFINE_MUTEX(kp_enable_mutex);
+static int kp_enable = 1;
+static int kp_cur_group = -1;
+
+struct omap_kp {
+	struct input_dev *input;
+	struct timer_list timer;
+	int irq;
+	unsigned int rows;
+	unsigned int cols;
+	unsigned long delay;
+	unsigned int debounce;
+};
+
+DECLARE_TASKLET_DISABLED(kp_tasklet, omap_kp_tasklet, 0);
+
+static int *keymap;
+static unsigned int *row_gpios;
+static unsigned int *col_gpios;
+
+#ifdef CONFIG_ARCH_OMAP2
+static void set_col_gpio_val(struct omap_kp *omap_kp, u8 value)
+{
+	int col;
+	for (col = 0; col < omap_kp->cols; col++) {
+		if (value & (1 << col))
+			omap_set_gpio_dataout(col_gpios[col], 1);
+		else
+			omap_set_gpio_dataout(col_gpios[col], 0);
+	}
+}
+
+static u8 get_row_gpio_val(struct omap_kp *omap_kp)
+{
+	int row;
+	u8 value = 0;
+
+	for (row = 0; row < omap_kp->rows; row++) {
+		if (omap_get_gpio_datain(row_gpios[row]))
+			value |= (1 << row);
+	}
+	return value;
+}
+#else
+#define		set_col_gpio_val(x, y)	do {} while (0)
+#define		get_row_gpio_val(x)	0
+#endif
+
+static irqreturn_t omap_kp_interrupt(int irq, void *dev_id,
+				     struct pt_regs *regs)
+{
+	struct omap_kp *omap_kp = dev_id;
+
+	/* disable keyboard interrupt and schedule for handling */
+	if (cpu_is_omap24xx()) {
+		int i;
+		for (i = 0; i < omap_kp->rows; i++)
+			disable_irq(OMAP_GPIO_IRQ(row_gpios[i]));
+	} else
+		/* disable keyboard interrupt and schedule for handling */
+		omap_writew(1, OMAP_MPUIO_BASE + OMAP_MPUIO_KBD_MASKIT);
+
+	tasklet_schedule(&kp_tasklet);
+
+	return IRQ_HANDLED;
+}
+
+static void omap_kp_timer(unsigned long data)
+{
+	tasklet_schedule(&kp_tasklet);
+}
+
+static void omap_kp_scan_keypad(struct omap_kp *omap_kp, unsigned char *state)
+{
+	int col = 0;
+
+	/* read the keypad status */
+	if (cpu_is_omap24xx()) {
+		int i;
+		for (i = 0; i < omap_kp->rows; i++)
+			disable_irq(OMAP_GPIO_IRQ(row_gpios[i]));
+
+		/* read the keypad status */
+		for (col = 0; col < omap_kp->cols; col++) {
+			set_col_gpio_val(omap_kp, ~(1 << col));
+			state[col] = ~(get_row_gpio_val(omap_kp)) & 0x3f;
+		}
+		set_col_gpio_val(omap_kp, 0);
+
+	} else {
+		/* disable keyboard interrupt and schedule for handling */
+		omap_writew(1, OMAP_MPUIO_BASE + OMAP_MPUIO_KBD_MASKIT);
+
+		/* read the keypad status */
+		omap_writew(0xff, OMAP_MPUIO_BASE + OMAP_MPUIO_KBC);
+		for (col = 0; col < omap_kp->cols; col++) {
+			omap_writew(~(1 << col) & 0xff,
+				    OMAP_MPUIO_BASE + OMAP_MPUIO_KBC);
+
+			udelay(omap_kp->delay);
+
+			state[col] = ~omap_readw(OMAP_MPUIO_BASE +
+						 OMAP_MPUIO_KBR_LATCH) & 0xff;
+		}
+		omap_writew(0x00, OMAP_MPUIO_BASE + OMAP_MPUIO_KBC);
+		udelay(2);
+	}
+}
+
+static inline int omap_kp_find_key(int col, int row)
+{
+	int i, key;
+
+	key = KEY(col, row, 0);
+	for (i = 0; keymap[i] != 0; i++)
+		if ((keymap[i] & 0xff000000) == key)
+			return keymap[i] & 0x00ffffff;
+	return -1;
+}
+
+static void omap_kp_tasklet(unsigned long data)
+{
+	struct omap_kp *omap_kp_data = (struct omap_kp *) data;
+	unsigned char new_state[8], changed, key_down = 0;
+	int col, row;
+	int spurious = 0;
+
+	/* check for any changes */
+	omap_kp_scan_keypad(omap_kp_data, new_state);
+
+	/* check for changes and print those */
+	for (col = 0; col < omap_kp_data->cols; col++) {
+		changed = new_state[col] ^ keypad_state[col];
+		key_down |= new_state[col];
+		if (changed == 0)
+			continue;
+
+		for (row = 0; row < omap_kp_data->rows; row++) {
+			int key;
+			if (!(changed & (1 << row)))
+				continue;
+#ifdef NEW_BOARD_LEARNING_MODE
+			printk(KERN_INFO "omap-keypad: key %d-%d %s\n", col,
+			       row, (new_state[col] & (1 << row)) ?
+			       "pressed" : "released");
+#else
+			key = omap_kp_find_key(col, row);
+			if (key < 0) {
+				printk(KERN_WARNING
+				      "omap-keypad: Spurious key event %d-%d\n",
+				       col, row);
+				/* We scan again after a couple of seconds */
+				spurious = 1;
+				continue;
+			}
+
+			if (!(kp_cur_group == (key & GROUP_MASK) ||
+			      kp_cur_group == -1))
+				continue;
+
+			kp_cur_group = key & GROUP_MASK;
+			input_report_key(omap_kp_data->input, key & ~GROUP_MASK,
+					 new_state[col] & (1 << row));
+#endif
+		}
+	}
+	memcpy(keypad_state, new_state, sizeof(keypad_state));
+
+	if (key_down) {
+                int delay = HZ / 20;
+		/* some key is pressed - keep irq disabled and use timer
+		 * to poll the keypad */
+		if (spurious)
+			delay = 2 * HZ;
+		mod_timer(&omap_kp_data->timer, jiffies + delay);
+	} else {
+		/* enable interrupts */
+		if (cpu_is_omap24xx()) {
+			int i;
+			for (i = 0; i < omap_kp_data->rows; i++)
+				enable_irq(OMAP_GPIO_IRQ(row_gpios[i]));
+		} else {
+			omap_writew(0, OMAP_MPUIO_BASE + OMAP_MPUIO_KBD_MASKIT);
+			kp_cur_group = -1;
+		}
+ 	}
+}
+
+static ssize_t omap_kp_enable_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%u\n", kp_enable);
+}
+
+static ssize_t omap_kp_enable_store(struct device *dev, struct device_attribute *attr,
+				    const char *buf, size_t count)
+{
+	int state;
+
+	if (sscanf(buf, "%u", &state) != 1)
+		return -EINVAL;
+
+	if ((state != 1) && (state != 0))
+		return -EINVAL;
+
+	mutex_lock(&kp_enable_mutex);
+	if (state != kp_enable) {
+		if (state)
+			enable_irq(INT_KEYBOARD);
+		else
+			disable_irq(INT_KEYBOARD);
+		kp_enable = state;
+	}
+	mutex_unlock(&kp_enable_mutex);
+
+	return strnlen(buf, count);
+}
+
+static DEVICE_ATTR(enable, S_IRUGO | S_IWUSR, omap_kp_enable_show, omap_kp_enable_store);
+
+#ifdef CONFIG_PM
+static int omap_kp_suspend(struct platform_device *dev, pm_message_t state)
+{
+	/* Nothing yet */
+
+	return 0;
+}
+
+static int omap_kp_resume(struct platform_device *dev)
+{
+	/* Nothing yet */
+
+	return 0;
+}
+#else
+#define omap_kp_suspend	NULL
+#define omap_kp_resume	NULL
+#endif
+
+static int __init omap_kp_probe(struct platform_device *pdev)
+{
+	struct omap_kp *omap_kp;
+	struct input_dev *input_dev;
+	struct omap_kp_platform_data *pdata =  pdev->dev.platform_data;
+	int i, col_idx, row_idx, irq_idx, ret;
+
+	if (!pdata->rows || !pdata->cols || !pdata->keymap) {
+		printk(KERN_ERR "No rows, cols or keymap from pdata\n");
+		return -EINVAL;
+	}
+
+	omap_kp = kzalloc(sizeof(struct omap_kp), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!omap_kp || !input_dev) {
+		kfree(omap_kp);
+		input_free_device(input_dev);
+		return -ENOMEM;
+	}
+
+	platform_set_drvdata(pdev, omap_kp);
+
+	omap_kp->input = input_dev;
+
+	/* Disable the interrupt for the MPUIO keyboard */
+	if (!cpu_is_omap24xx())
+		omap_writew(1, OMAP_MPUIO_BASE + OMAP_MPUIO_KBD_MASKIT);
+
+	keymap = pdata->keymap;
+
+	if (pdata->rep)
+		set_bit(EV_REP, input_dev->evbit);
+
+	if (pdata->delay)
+		omap_kp->delay = pdata->delay;
+
+	if (pdata->row_gpios && pdata->col_gpios) {
+		row_gpios = pdata->row_gpios;
+		col_gpios = pdata->col_gpios;
+	}
+
+	omap_kp->rows = pdata->rows;
+	omap_kp->cols = pdata->cols;
+
+	if (cpu_is_omap24xx()) {
+		/* Cols: outputs */
+		for (col_idx = 0; col_idx < omap_kp->cols; col_idx++) {
+			if (omap_request_gpio(col_gpios[col_idx]) < 0) {
+				printk(KERN_ERR "Failed to request"
+				       "GPIO%d for keypad\n",
+				       col_gpios[col_idx]);
+				goto err1;
+			}
+			omap_set_gpio_direction(col_gpios[col_idx], 0);
+		}
+		/* Rows: inputs */
+		for (row_idx = 0; row_idx < omap_kp->rows; row_idx++) {
+			if (omap_request_gpio(row_gpios[row_idx]) < 0) {
+				printk(KERN_ERR "Failed to request"
+				       "GPIO%d for keypad\n",
+				       row_gpios[row_idx]);
+				goto err2;
+			}
+			omap_set_gpio_direction(row_gpios[row_idx], 1);
+		}
+	}
+
+	setup_timer(&omap_kp->timer, omap_kp_timer, (unsigned long)omap_kp);
+
+	/* get the irq and init timer*/
+	tasklet_enable(&kp_tasklet);
+	kp_tasklet.data = (unsigned long) omap_kp;
+
+	ret = device_create_file(&pdev->dev, &dev_attr_enable);
+	if (ret < 0)
+		goto err2;
+
+	/* setup input device */
+	set_bit(EV_KEY, input_dev->evbit);
+	for (i = 0; keymap[i] != 0; i++)
+		set_bit(keymap[i] & KEY_MAX, input_dev->keybit);
+	input_dev->name = "omap-keypad";
+	input_dev->phys = "omap-keypad/input0";
+	input_dev->cdev.dev = &pdev->dev;
+	input_dev->private = omap_kp;
+
+	input_dev->id.bustype = BUS_HOST;
+	input_dev->id.vendor = 0x0001;
+	input_dev->id.product = 0x0001;
+	input_dev->id.version = 0x0100;
+
+	input_dev->keycode = keymap;
+	input_dev->keycodesize = sizeof(unsigned int);
+	input_dev->keycodemax = pdata->keymapsize;
+
+	ret = input_register_device(omap_kp->input);
+	if (ret < 0) {
+		printk(KERN_ERR "Unable to register omap-keypad input device\n");
+		goto err3;
+	}
+
+	if (pdata->dbounce)
+		omap_writew(0xff, OMAP_MPUIO_BASE + OMAP_MPUIO_GPIO_DEBOUNCING);
+
+	/* scan current status and enable interrupt */
+	omap_kp_scan_keypad(omap_kp, keypad_state);
+	if (!cpu_is_omap24xx()) {
+		omap_kp->irq = platform_get_irq(pdev, 0);
+		if (omap_kp->irq >= 0) {
+			if (request_irq(omap_kp->irq, omap_kp_interrupt, 0,
+					"omap-keypad", omap_kp) < 0)
+				goto err4;
+		}
+		omap_writew(0, OMAP_MPUIO_BASE + OMAP_MPUIO_KBD_MASKIT);
+	} else {
+		for (irq_idx = 0; irq_idx < omap_kp->rows; irq_idx++) {
+			if (request_irq(OMAP_GPIO_IRQ(row_gpios[irq_idx]),
+				       	omap_kp_interrupt,
+					IRQF_TRIGGER_FALLING,
+				       	"omap-keypad", omap_kp) < 0)
+				goto err5;
+		}
+	}
+	return 0;
+err5:
+	for (i = irq_idx-1; i >=0; i--)
+		free_irq(row_gpios[i], 0);
+err4:
+	input_unregister_device(omap_kp->input);
+	input_dev = NULL;
+err3:
+	device_remove_file(&pdev->dev, &dev_attr_enable);
+err2:
+	for (i = row_idx-1; i >=0; i--)
+		omap_free_gpio(row_gpios[i]);
+err1:
+	for (i = col_idx-1; i >=0; i--)
+		omap_free_gpio(col_gpios[i]);
+
+	kfree(omap_kp);
+	input_free_device(input_dev);
+
+	return -EINVAL;
+}
+
+static int omap_kp_remove(struct platform_device *pdev)
+{
+	struct omap_kp *omap_kp = platform_get_drvdata(pdev);
+
+	/* disable keypad interrupt handling */
+	tasklet_disable(&kp_tasklet);
+	if (cpu_is_omap24xx()) {
+		int i;
+		for (i = 0; i < omap_kp->cols; i++)
+	    		omap_free_gpio(col_gpios[i]);
+		for (i = 0; i < omap_kp->rows; i++) {
+	    		omap_free_gpio(row_gpios[i]);
+			free_irq(OMAP_GPIO_IRQ(row_gpios[i]), 0);
+		}
+	} else {
+		omap_writew(1, OMAP_MPUIO_BASE + OMAP_MPUIO_KBD_MASKIT);
+		free_irq(omap_kp->irq, 0);
+	}
+
+	del_timer_sync(&omap_kp->timer);
+	tasklet_kill(&kp_tasklet);
+
+	/* unregister everything */
+	input_unregister_device(omap_kp->input);
+
+	kfree(omap_kp);
+
+	return 0;
+}
+
+static struct platform_driver omap_kp_driver = {
+	.probe		= omap_kp_probe,
+	.remove		= omap_kp_remove,
+	.suspend	= omap_kp_suspend,
+	.resume		= omap_kp_resume,
+	.driver		= {
+		.name	= "omap-keypad",
+	},
+};
+
+static int __devinit omap_kp_init(void)
+{
+	printk(KERN_INFO "OMAP Keypad Driver\n");
+	return platform_driver_register(&omap_kp_driver);
+}
+
+static void __exit omap_kp_exit(void)
+{
+	platform_driver_unregister(&omap_kp_driver);
+}
+
+module_init(omap_kp_init);
+module_exit(omap_kp_exit);
+
+MODULE_AUTHOR("Timo Teräs");
+MODULE_DESCRIPTION("OMAP Keypad Driver");
+MODULE_LICENSE("GPL");
diff --git a/include/asm-arm/arch-omap/keypad.h b/include/asm-arm/arch-omap/keypad.h
index 8a023a984acb..b7f83075436e 100644
--- a/include/asm-arm/arch-omap/keypad.h
+++ b/include/asm-arm/arch-omap/keypad.h
@@ -14,7 +14,10 @@ struct omap_kp_platform_data {
 	int rows;
 	int cols;
 	int *keymap;
+	unsigned int keymapsize;
 	unsigned int rep:1;
+	unsigned long delay;
+	unsigned int dbounce:1;
 	/* specific to OMAP242x*/
 	unsigned int *row_gpios;
 	unsigned int *col_gpios;
-- 
cgit v1.2.3


From db0b0ead60815155c791e8f479ee4777e7946369 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Fri, 29 Sep 2006 01:59:28 -0700
Subject: [PATCH] lockdep: don't pull in includes when lockdep disabled

Do not pull in various includes through lockdep.h if lockdep is disabled.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/lockdep.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index c040a8c969aa..1314ca0f29be 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -8,13 +8,13 @@
 #ifndef __LINUX_LOCKDEP_H
 #define __LINUX_LOCKDEP_H
 
+#ifdef CONFIG_LOCKDEP
+
 #include <linux/linkage.h>
 #include <linux/list.h>
 #include <linux/debug_locks.h>
 #include <linux/stacktrace.h>
 
-#ifdef CONFIG_LOCKDEP
-
 /*
  * Lock-class usage-state bits:
  */
-- 
cgit v1.2.3


From 650a898342b3fa21c392c06a2b7010fa19823efa Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <miklos@szeredi.hu>
Date: Fri, 29 Sep 2006 01:59:35 -0700
Subject: [PATCH] vfs: define new lookup flag for chdir

In the "operation does permission checking" model used by fuse, chdir
permission is not checked, since there's no chdir method.

For this case set a lookup flag, which will be passed to ->permission(), so
fuse can distinguish it from permission checks for other operations.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/fuse/dir.c         | 2 +-
 fs/open.c             | 3 ++-
 include/linux/namei.h | 1 +
 3 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 409ce6a7cca4..f85b2a282f13 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -776,7 +776,7 @@ static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd)
 		if ((mask & MAY_EXEC) && !S_ISDIR(mode) && !(mode & S_IXUGO))
 			return -EACCES;
 
-		if (nd && (nd->flags & LOOKUP_ACCESS))
+		if (nd && (nd->flags & (LOOKUP_ACCESS | LOOKUP_CHDIR)))
 			return fuse_access(inode, mask);
 		return 0;
 	}
diff --git a/fs/open.c b/fs/open.c
index 303f06d2a7b9..1574d8fe4909 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -546,7 +546,8 @@ asmlinkage long sys_chdir(const char __user * filename)
 	struct nameidata nd;
 	int error;
 
-	error = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &nd);
+	error = __user_walk(filename,
+			    LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_CHDIR, &nd);
 	if (error)
 		goto out;
 
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 45511a5918d3..c6470ba00668 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -54,6 +54,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
 #define LOOKUP_OPEN		(0x0100)
 #define LOOKUP_CREATE		(0x0200)
 #define LOOKUP_ACCESS		(0x0400)
+#define LOOKUP_CHDIR		(0x0800)
 
 extern int FASTCALL(__user_walk(const char __user *, unsigned, struct nameidata *));
 extern int FASTCALL(__user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *));
-- 
cgit v1.2.3


From 2e0c1f6ce7b816f63fea2af3e5e2cb20c66430e9 Mon Sep 17 00:00:00 2001
From: Shem Multinymous <multinymous@gmail.com>
Date: Fri, 29 Sep 2006 01:59:37 -0700
Subject: [PATCH] DMI: Decode and save OEM String information

This teaches dmi_decode() how to decode and save OEM Strings (type 11) DMI
information, which is currently discarded silently.  Existing code using
DMI is not affected.  Follows the "System Management BIOS (SMBIOS)
Specification" (http://www.dmtf.org/standards/smbios), and also the
userspace dmidecode.c code.

OEM Strings are the only safe way to identify some hardware, e.g., the
ThinkPad embedded controller used by the soon-to-be-submitted tp_smapi
driver.  This will also let us eliminate the long whitelist in the mainline
hdaps driver (in a future patch).

Signed-off-by: Shem Multinymous <multinymous@gmail.com>
Cc: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/firmware/dmi_scan.c | 23 +++++++++++++++++++++++
 include/linux/dmi.h         |  3 ++-
 2 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
index b9e3886d9e16..b8b596d5778d 100644
--- a/drivers/firmware/dmi_scan.c
+++ b/drivers/firmware/dmi_scan.c
@@ -123,6 +123,26 @@ static void __init dmi_save_devices(struct dmi_header *dm)
 		dev->type = *d++ & 0x7f;
 		dev->name = dmi_string(dm, *d);
 		dev->device_data = NULL;
+		list_add(&dev->list, &dmi_devices);
+	}
+}
+
+static void __init dmi_save_oem_strings_devices(struct dmi_header *dm)
+{
+	int i, count = *(u8 *)(dm + 1);
+	struct dmi_device *dev;
+
+	for (i = 1; i <= count; i++) {
+		dev = dmi_alloc(sizeof(*dev));
+		if (!dev) {
+			printk(KERN_ERR
+			   "dmi_save_oem_strings_devices: out of memory.\n");
+			break;
+		}
+
+		dev->type = DMI_DEV_TYPE_OEM_STRING;
+		dev->name = dmi_string(dm, i);
+		dev->device_data = NULL;
 
 		list_add(&dev->list, &dmi_devices);
 	}
@@ -181,6 +201,9 @@ static void __init dmi_decode(struct dmi_header *dm)
 	case 10:	/* Onboard Devices Information */
 		dmi_save_devices(dm);
 		break;
+	case 11:	/* OEM Strings */
+		dmi_save_oem_strings_devices(dm);
+		break;
 	case 38:	/* IPMI Device Information */
 		dmi_save_ipmi_device(dm);
 	}
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index b2cd2071d432..38dc403be70b 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -27,7 +27,8 @@ enum dmi_device_type {
 	DMI_DEV_TYPE_ETHERNET,
 	DMI_DEV_TYPE_TOKENRING,
 	DMI_DEV_TYPE_SOUND,
-	DMI_DEV_TYPE_IPMI = -1
+	DMI_DEV_TYPE_IPMI = -1,
+	DMI_DEV_TYPE_OEM_STRING = -2
 };
 
 struct dmi_header {
-- 
cgit v1.2.3


From 402749ea2538be9ddad981a990739b93a0178bc6 Mon Sep 17 00:00:00 2001
From: Matthias Urlichs <smurf@smurf.noris.de>
Date: Fri, 29 Sep 2006 01:59:37 -0700
Subject: [PATCH] Remove unused tty_struct field

Unused: tty_struct.max_flip_cnt

$ git grep max_flip_cnt
include/linux/tty.h:    int max_flip_cnt;
$

Cc: Paul Fulghum <paulkf@microgate.com>
Acked-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/tty.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/tty.h b/include/linux/tty.h
index 04827ca65781..d1dec3d0c814 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -190,7 +190,6 @@ struct tty_struct {
 	struct tty_struct *link;
 	struct fasync_struct *fasync;
 	struct tty_bufhead buf;
-	int max_flip_cnt;
 	int alt_speed;		/* For magic substitution of 38400 bps */
 	wait_queue_head_t write_wait;
 	wait_queue_head_t read_wait;
-- 
cgit v1.2.3


From 3d5b6fccc4b900cc4267692f015ea500bad4c6bf Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 01:59:40 -0700
Subject: [PATCH] task_struct: ifdef Missed'em V IPC

ipc/sem.c only.

$ agrep sysvsem -w -n
ipc/sem.c:912:  undo_list = current->sysvsem.undo_list;
ipc/sem.c:932:  undo_list = current->sysvsem.undo_list;
ipc/sem.c:954:  undo_list = current->sysvsem.undo_list;
ipc/sem.c:963:          current->sysvsem.undo_list = undo_list;
ipc/sem.c:1247:         tsk->sysvsem.undo_list = undo_list;
ipc/sem.c:1249:         tsk->sysvsem.undo_list = NULL;
ipc/sem.c:1271: undo_list = tsk->sysvsem.undo_list;
include/linux/sched.h:876:      struct sysv_sem sysvsem;

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 9d4aa7f95bc8..27122575d902 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -886,8 +886,10 @@ struct task_struct {
 				     - initialized normally by flush_old_exec */
 /* file system info */
 	int link_count, total_link_count;
+#ifdef CONFIG_SYSVIPC
 /* ipc stuff */
 	struct sysv_sem sysvsem;
+#endif
 /* CPU-specific state of this task */
 	struct thread_struct thread;
 /* filesystem information */
-- 
cgit v1.2.3


From 6c5c934153513dc72e2d6464f39e8ef1f27c0a3e Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 01:59:40 -0700
Subject: [PATCH] ifdef blktrace debugging fields

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 block/blktrace.c       | 6 ++++--
 block/ll_rw_blk.c      | 3 +--
 include/linux/blkdev.h | 4 ++--
 include/linux/sched.h  | 3 ++-
 kernel/fork.c          | 2 ++
 5 files changed, 11 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/block/blktrace.c b/block/blktrace.c
index 2b4ef2b89b8d..8ff33441d8a2 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -450,8 +450,10 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
  **/
 void blk_trace_shutdown(request_queue_t *q)
 {
-	blk_trace_startstop(q, 0);
-	blk_trace_remove(q);
+	if (q->blk_trace) {
+		blk_trace_startstop(q, 0);
+		blk_trace_remove(q);
+	}
 }
 
 /*
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 9c3a06bcb7ba..51dc0edf76e0 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -1847,8 +1847,7 @@ static void blk_release_queue(struct kobject *kobj)
 	if (q->queue_tags)
 		__blk_queue_free_tags(q);
 
-	if (q->blk_trace)
-		blk_trace_shutdown(q);
+	blk_trace_shutdown(q);
 
 	kmem_cache_free(requestq_cachep, q);
 }
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index c773ee545ebd..cfde8b3ee919 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -417,9 +417,9 @@ struct request_queue
 	unsigned int		sg_timeout;
 	unsigned int		sg_reserved_size;
 	int			node;
-
+#ifdef CONFIG_BLK_DEV_IO_TRACE
 	struct blk_trace	*blk_trace;
-
+#endif
 	/*
 	 * reserved for flush operations
 	 */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 27122575d902..3696f2f7126d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -784,8 +784,9 @@ struct task_struct {
 	struct prio_array *array;
 
 	unsigned short ioprio;
+#ifdef CONFIG_BLK_DEV_IO_TRACE
 	unsigned int btrace_seq;
-
+#endif
 	unsigned long sleep_avg;
 	unsigned long long timestamp, last_ran;
 	unsigned long long sched_time; /* sched_clock time spent running */
diff --git a/kernel/fork.c b/kernel/fork.c
index 802b1cf0e63f..bca6ce6d3ded 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -183,7 +183,9 @@ static struct task_struct *dup_task_struct(struct task_struct *orig)
 	/* One for us, one for whoever does the "release_task()" (usually parent) */
 	atomic_set(&tsk->usage,2);
 	atomic_set(&tsk->fs_excl, 0);
+#ifdef CONFIG_BLK_DEV_IO_TRACE
 	tsk->btrace_seq = 0;
+#endif
 	tsk->splice_pipe = NULL;
 	return tsk;
 }
-- 
cgit v1.2.3


From d6bd3a39f7c6ebad49c261c3d458df974c880758 Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Fri, 29 Sep 2006 01:59:48 -0700
Subject: [PATCH] Move valid_dma_direction() from x86_64 to generic code

As suggested by Muli Ben-Yehuda this function is moved to generic code as
may be useful for all archs.

[akpm@osdl.org: fix]
Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Cc: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/pci-swiotlb.c | 3 ++-
 drivers/net/fec_8xx/fec_main.c   | 2 +-
 drivers/net/fs_enet/fs_enet.h    | 3 +--
 include/asm-x86_64/dma-mapping.h | 7 -------
 include/linux/dma-mapping.h      | 7 +++++++
 5 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/pci-swiotlb.c b/arch/x86_64/kernel/pci-swiotlb.c
index 6a55f87ba97f..697f0aa794b9 100644
--- a/arch/x86_64/kernel/pci-swiotlb.c
+++ b/arch/x86_64/kernel/pci-swiotlb.c
@@ -3,7 +3,8 @@
 #include <linux/pci.h>
 #include <linux/cache.h>
 #include <linux/module.h>
-#include <asm/dma-mapping.h>
+#include <linux/dma-mapping.h>
+
 #include <asm/proto.h>
 #include <asm/swiotlb.h>
 #include <asm/dma.h>
diff --git a/drivers/net/fec_8xx/fec_main.c b/drivers/net/fec_8xx/fec_main.c
index 22ac2df1aeb0..e17a1449ee10 100644
--- a/drivers/net/fec_8xx/fec_main.c
+++ b/drivers/net/fec_8xx/fec_main.c
@@ -30,6 +30,7 @@
 #include <linux/mii.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
+#include <linux/dma-mapping.h>
 
 #include <asm/8xx_immap.h>
 #include <asm/pgtable.h>
@@ -37,7 +38,6 @@
 #include <asm/irq.h>
 #include <asm/uaccess.h>
 #include <asm/commproc.h>
-#include <asm/dma-mapping.h>
 
 #include "fec_8xx.h"
 
diff --git a/drivers/net/fs_enet/fs_enet.h b/drivers/net/fs_enet/fs_enet.h
index 95022c005f75..92590d8fc24b 100644
--- a/drivers/net/fs_enet/fs_enet.h
+++ b/drivers/net/fs_enet/fs_enet.h
@@ -6,11 +6,10 @@
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/phy.h>
+#include <linux/dma-mapping.h>
 
 #include <linux/fs_enet_pd.h>
 
-#include <asm/dma-mapping.h>
-
 #ifdef CONFIG_CPM1
 #include <asm/commproc.h>
 
diff --git a/include/asm-x86_64/dma-mapping.h b/include/asm-x86_64/dma-mapping.h
index b6da83dcc7a6..10174b110a5c 100644
--- a/include/asm-x86_64/dma-mapping.h
+++ b/include/asm-x86_64/dma-mapping.h
@@ -55,13 +55,6 @@ extern dma_addr_t bad_dma_address;
 extern struct dma_mapping_ops* dma_ops;
 extern int iommu_merge;
 
-static inline int valid_dma_direction(int dma_direction)
-{
-	return ((dma_direction == DMA_BIDIRECTIONAL) ||
-		(dma_direction == DMA_TO_DEVICE) ||
-		(dma_direction == DMA_FROM_DEVICE));
-}
-
 static inline int dma_mapping_error(dma_addr_t dma_addr)
 {
 	if (dma_ops->mapping_error)
diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
index 635690cf3e3d..ff203c465fed 100644
--- a/include/linux/dma-mapping.h
+++ b/include/linux/dma-mapping.h
@@ -24,6 +24,13 @@ enum dma_data_direction {
 #define DMA_28BIT_MASK	0x000000000fffffffULL
 #define DMA_24BIT_MASK	0x0000000000ffffffULL
 
+static inline int valid_dma_direction(int dma_direction)
+{
+	return ((dma_direction == DMA_BIDIRECTIONAL) ||
+		(dma_direction == DMA_TO_DEVICE) ||
+		(dma_direction == DMA_FROM_DEVICE));
+}
+
 #include <asm/dma-mapping.h>
 
 /* Backwards compat, remove in 2.7.x */
-- 
cgit v1.2.3


From cd1c6a48ac16b360746f9f111895931d332c35dd Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Fri, 29 Sep 2006 01:59:49 -0700
Subject: [PATCH] Use valid_dma_direction() in include/asm-i386/dma-mapping.h

Now that the generic DMA code has a function to decide if a given DMA
mapping is valid use it.  This will catch cases where direction is not any
of the defined enum values but some random number outside the valid range.
The current implementation will only catch the defined but invalid case
DMA_NONE.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Acked-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/dma-mapping.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/dma-mapping.h b/include/asm-i386/dma-mapping.h
index 576ae01d71c8..81999a3ebe7c 100644
--- a/include/asm-i386/dma-mapping.h
+++ b/include/asm-i386/dma-mapping.h
@@ -21,7 +21,7 @@ static inline dma_addr_t
 dma_map_single(struct device *dev, void *ptr, size_t size,
 	       enum dma_data_direction direction)
 {
-	BUG_ON(direction == DMA_NONE);
+	BUG_ON(!valid_dma_direction(direction));
 	WARN_ON(size == 0);
 	flush_write_buffers();
 	return virt_to_phys(ptr);
@@ -31,7 +31,7 @@ static inline void
 dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
 		 enum dma_data_direction direction)
 {
-	BUG_ON(direction == DMA_NONE);
+	BUG_ON(!valid_dma_direction(direction));
 }
 
 static inline int
@@ -40,7 +40,7 @@ dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
 {
 	int i;
 
-	BUG_ON(direction == DMA_NONE);
+	BUG_ON(!valid_dma_direction(direction));
 	WARN_ON(nents == 0 || sg[0].length == 0);
 
 	for (i = 0; i < nents; i++ ) {
@@ -57,7 +57,7 @@ static inline dma_addr_t
 dma_map_page(struct device *dev, struct page *page, unsigned long offset,
 	     size_t size, enum dma_data_direction direction)
 {
-	BUG_ON(direction == DMA_NONE);
+	BUG_ON(!valid_dma_direction(direction));
 	return page_to_phys(page) + offset;
 }
 
@@ -65,7 +65,7 @@ static inline void
 dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
 	       enum dma_data_direction direction)
 {
-	BUG_ON(direction == DMA_NONE);
+	BUG_ON(!valid_dma_direction(direction));
 }
 
 
@@ -73,7 +73,7 @@ static inline void
 dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
 	     enum dma_data_direction direction)
 {
-	BUG_ON(direction == DMA_NONE);
+	BUG_ON(!valid_dma_direction(direction));
 }
 
 static inline void
-- 
cgit v1.2.3


From 0e51a720b9d9ea5ebf0fda39108919c6626bffa3 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 01:59:56 -0700
Subject: [PATCH] ifdef ->quota_read, ->quota_write

All suppliers of ->quota_read, ->quota_write (I've found ext2, ext3, UFS,
reiserfs) already have them properly ifdeffed.  All callers of
->quota_read, ->quota_write are under CONFIG_QUOTA umbrella, so...

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/fs.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 8f74dfbb2edd..4caec6cebc42 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1143,9 +1143,10 @@ struct super_operations {
 
 	int (*show_options)(struct seq_file *, struct vfsmount *);
 	int (*show_stats)(struct seq_file *, struct vfsmount *);
-
+#ifdef CONFIG_QUOTA
 	ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
 	ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
+#endif
 };
 
 /* Inode state bits.  Protected by inode_lock. */
-- 
cgit v1.2.3


From 068fbb315dd1e9dd3418aac39a9cfeabe39c16a6 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 01:59:58 -0700
Subject: [PATCH] reiserfs: ifdef xattr_sem

Shrink reiserfs inode by 12 bytes for xattr non-users (me).

	-reiser_inode_cache     356     11
	+reiser_inode_cache     344     11

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: <reiserfs-dev@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/inode.c            | 4 ++--
 include/linux/reiserfs_fs_i.h  | 2 ++
 include/linux/reiserfs_xattr.h | 8 ++++++++
 3 files changed, 12 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 8810fda0da46..78f23f406932 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1129,7 +1129,7 @@ static void init_inode(struct inode *inode, struct path *path)
 	REISERFS_I(inode)->i_jl = NULL;
 	REISERFS_I(inode)->i_acl_access = NULL;
 	REISERFS_I(inode)->i_acl_default = NULL;
-	init_rwsem(&REISERFS_I(inode)->xattr_sem);
+	reiserfs_init_xattr_rwsem(inode);
 
 	if (stat_data_v1(ih)) {
 		struct stat_data_v1 *sd =
@@ -1836,7 +1836,7 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	sd_attrs_to_i_attrs(REISERFS_I(inode)->i_attrs, inode);
 	REISERFS_I(inode)->i_acl_access = NULL;
 	REISERFS_I(inode)->i_acl_default = NULL;
-	init_rwsem(&REISERFS_I(inode)->xattr_sem);
+	reiserfs_init_xattr_rwsem(inode);
 
 	if (old_format_only(sb))
 		make_le_item_head(&ih, NULL, KEY_FORMAT_3_5, SD_OFFSET,
diff --git a/include/linux/reiserfs_fs_i.h b/include/linux/reiserfs_fs_i.h
index 149be8d9a0c9..711e7e7cafa5 100644
--- a/include/linux/reiserfs_fs_i.h
+++ b/include/linux/reiserfs_fs_i.h
@@ -55,7 +55,9 @@ struct reiserfs_inode_info {
 
 	struct posix_acl *i_acl_access;
 	struct posix_acl *i_acl_default;
+#ifdef CONFIG_REISERFS_FS_XATTR
 	struct rw_semaphore xattr_sem;
+#endif
 	struct inode vfs_inode;
 };
 
diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h
index 5e961035c725..966c35851b2e 100644
--- a/include/linux/reiserfs_xattr.h
+++ b/include/linux/reiserfs_xattr.h
@@ -97,6 +97,11 @@ static inline void reiserfs_mark_inode_private(struct inode *inode)
 	inode->i_flags |= S_PRIVATE;
 }
 
+static inline void reiserfs_init_xattr_rwsem(struct inode *inode)
+{
+	init_rwsem(&REISERFS_I(inode)->xattr_sem);
+}
+
 #else
 
 #define is_reiserfs_priv_object(inode) 0
@@ -129,6 +134,9 @@ static inline int reiserfs_xattr_init(struct super_block *sb, int mount_flags)
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL);	/* to be sure */
 	return 0;
 };
+static inline void reiserfs_init_xattr_rwsem(struct inode *inode)
+{
+}
 #endif
 
 #endif				/* __KERNEL__ */
-- 
cgit v1.2.3


From cfe14677f286c9be5d683b88214def8f4b8a6f24 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 02:00:00 -0700
Subject: [PATCH] reiserfs: ifdef ACL stuff from inode

Shrink reiserfs inode more (by 8 bytes) for ACL non-users:

	-reiser_inode_cache     344     11
	+reiser_inode_cache     336     11

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: <reiserfs-dev@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/inode.c           | 10 ++++++----
 fs/reiserfs/super.c           |  6 ++++++
 include/linux/reiserfs_acl.h  | 17 +++++++++++++++++
 include/linux/reiserfs_fs_i.h |  3 ++-
 4 files changed, 31 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 78f23f406932..7e5a2f5ebeb0 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1127,8 +1127,8 @@ static void init_inode(struct inode *inode, struct path *path)
 	REISERFS_I(inode)->i_prealloc_count = 0;
 	REISERFS_I(inode)->i_trans_id = 0;
 	REISERFS_I(inode)->i_jl = NULL;
-	REISERFS_I(inode)->i_acl_access = NULL;
-	REISERFS_I(inode)->i_acl_default = NULL;
+	reiserfs_init_acl_access(inode);
+	reiserfs_init_acl_default(inode);
 	reiserfs_init_xattr_rwsem(inode);
 
 	if (stat_data_v1(ih)) {
@@ -1834,8 +1834,8 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	REISERFS_I(inode)->i_attrs =
 	    REISERFS_I(dir)->i_attrs & REISERFS_INHERIT_MASK;
 	sd_attrs_to_i_attrs(REISERFS_I(inode)->i_attrs, inode);
-	REISERFS_I(inode)->i_acl_access = NULL;
-	REISERFS_I(inode)->i_acl_default = NULL;
+	reiserfs_init_acl_access(inode);
+	reiserfs_init_acl_default(inode);
 	reiserfs_init_xattr_rwsem(inode);
 
 	if (old_format_only(sb))
@@ -1974,11 +1974,13 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	 * iput doesn't deadlock in reiserfs_delete_xattrs. The locking
 	 * code really needs to be reworked, but this will take care of it
 	 * for now. -jeffm */
+#ifdef CONFIG_REISERFS_FS_POSIX_ACL
 	if (REISERFS_I(dir)->i_acl_default && !IS_ERR(REISERFS_I(dir)->i_acl_default)) {
 		reiserfs_write_unlock_xattrs(dir->i_sb);
 		iput(inode);
 		reiserfs_write_lock_xattrs(dir->i_sb);
 	} else
+#endif
 		iput(inode);
 	return err;
 }
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index b40d4d64d598..80fc3b32802f 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -510,8 +510,10 @@ static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags)
 	    SLAB_CTOR_CONSTRUCTOR) {
 		INIT_LIST_HEAD(&ei->i_prealloc_list);
 		inode_init_once(&ei->vfs_inode);
+#ifdef CONFIG_REISERFS_FS_POSIX_ACL
 		ei->i_acl_access = NULL;
 		ei->i_acl_default = NULL;
+#endif
 	}
 }
 
@@ -560,6 +562,7 @@ static void reiserfs_dirty_inode(struct inode *inode)
 	reiserfs_write_unlock(inode->i_sb);
 }
 
+#ifdef CONFIG_REISERFS_FS_POSIX_ACL
 static void reiserfs_clear_inode(struct inode *inode)
 {
 	struct posix_acl *acl;
@@ -574,6 +577,9 @@ static void reiserfs_clear_inode(struct inode *inode)
 		posix_acl_release(acl);
 	REISERFS_I(inode)->i_acl_default = NULL;
 }
+#else
+#define reiserfs_clear_inode NULL
+#endif
 
 #ifdef CONFIG_QUOTA
 static ssize_t reiserfs_quota_write(struct super_block *, int, const char *,
diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h
index 806ec5b06707..fe00f781a622 100644
--- a/include/linux/reiserfs_acl.h
+++ b/include/linux/reiserfs_acl.h
@@ -56,6 +56,16 @@ extern int reiserfs_xattr_posix_acl_init(void) __init;
 extern int reiserfs_xattr_posix_acl_exit(void);
 extern struct reiserfs_xattr_handler posix_acl_default_handler;
 extern struct reiserfs_xattr_handler posix_acl_access_handler;
+
+static inline void reiserfs_init_acl_access(struct inode *inode)
+{
+	REISERFS_I(inode)->i_acl_access = NULL;
+}
+
+static inline void reiserfs_init_acl_default(struct inode *inode)
+{
+	REISERFS_I(inode)->i_acl_default = NULL;
+}
 #else
 
 #define reiserfs_cache_default_acl(inode) 0
@@ -87,4 +97,11 @@ reiserfs_inherit_default_acl(const struct inode *dir, struct dentry *dentry,
 	return 0;
 }
 
+static inline void reiserfs_init_acl_access(struct inode *inode)
+{
+}
+
+static inline void reiserfs_init_acl_default(struct inode *inode)
+{
+}
 #endif
diff --git a/include/linux/reiserfs_fs_i.h b/include/linux/reiserfs_fs_i.h
index 711e7e7cafa5..5b3b297aa2c5 100644
--- a/include/linux/reiserfs_fs_i.h
+++ b/include/linux/reiserfs_fs_i.h
@@ -52,9 +52,10 @@ struct reiserfs_inode_info {
 	 ** flushed */
 	unsigned long i_trans_id;
 	struct reiserfs_journal_list *i_jl;
-
+#ifdef CONFIG_REISERFS_FS_POSIX_ACL
 	struct posix_acl *i_acl_access;
 	struct posix_acl *i_acl_default;
+#endif
 #ifdef CONFIG_REISERFS_FS_XATTR
 	struct rw_semaphore xattr_sem;
 #endif
-- 
cgit v1.2.3


From 50462062a02226a698a211d5bd535376c89b8603 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 02:00:01 -0700
Subject: [PATCH] fs.h: ifdef security fields

[assuming BSD security levels are deleted]
The only user of i_security, f_security, s_security fields is SELinux,
however, quite a few security modules are trying to get into kernel.
So, wrap them under CONFIG_SECURITY. Adding config option for each
security field is likely an overkill.

Following Stephen Smalley's suggestion, i_security initialization is
moved to security_inode_alloc() to not clutter core code with ifdefs
and make alloc_inode() codepath tiny little bit smaller and faster.

The user of (highly greppable) struct fown_struct::security field is
still to be found. I've checked every "fown_struct" and every "f_owner"
occurence. Additionally it's removal doesn't break i386 allmodconfig
build.

struct inode, struct file, struct super_block, struct fown_struct
become smaller.

P.S. Combined with two reiserfs inode shrinking patches sent to
linux-fsdevel, I can finally suck 12 reiserfs inodes into one page.

		/proc/slabinfo

	-ext2_inode_cache	388	10
	+ext2_inode_cache	384	10
	-inode_cache		280	14
	+inode_cache		276	14
	-proc_inode_cache	296	13
	+proc_inode_cache	292	13
	-reiser_inode_cache	336	11
	+reiser_inode_cache	332	12 <=
	-shmem_inode_cache	372	10
	+shmem_inode_cache	368	10

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/inode.c               | 1 -
 include/linux/fs.h       | 8 ++++++--
 include/linux/security.h | 1 +
 3 files changed, 7 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/inode.c b/fs/inode.c
index f5c04dd9ae8a..abf77471e6c4 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -133,7 +133,6 @@ static struct inode *alloc_inode(struct super_block *sb)
 		inode->i_bdev = NULL;
 		inode->i_cdev = NULL;
 		inode->i_rdev = 0;
-		inode->i_security = NULL;
 		inode->dirtied_when = 0;
 		if (security_inode_alloc(inode)) {
 			if (inode->i_sb->s_op->destroy_inode)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4caec6cebc42..6eafbe309483 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -553,7 +553,9 @@ struct inode {
 	unsigned int		i_flags;
 
 	atomic_t		i_writecount;
+#ifdef CONFIG_SECURITY
 	void			*i_security;
+#endif
 	void			*i_private; /* fs or device private pointer */
 #ifdef __NEED_I_SIZE_ORDERED
 	seqcount_t		i_size_seqcount;
@@ -645,7 +647,6 @@ struct fown_struct {
 	rwlock_t lock;          /* protects pid, uid, euid fields */
 	int pid;		/* pid or -pgrp where SIGIO should be sent */
 	uid_t uid, euid;	/* uid/euid of process setting the owner */
-	void *security;
 	int signum;		/* posix.1b rt signal to be delivered on IO */
 };
 
@@ -688,8 +689,9 @@ struct file {
 	struct file_ra_state	f_ra;
 
 	unsigned long		f_version;
+#ifdef CONFIG_SECURITY
 	void			*f_security;
-
+#endif
 	/* needed for tty driver, and maybe others */
 	void			*private_data;
 
@@ -877,7 +879,9 @@ struct super_block {
 	int			s_syncing;
 	int			s_need_sync_fs;
 	atomic_t		s_active;
+#ifdef CONFIG_SECURITY
 	void                    *s_security;
+#endif
 	struct xattr_handler	**s_xattr;
 
 	struct list_head	s_inodes;	/* all inodes */
diff --git a/include/linux/security.h b/include/linux/security.h
index 9f56fb8a4a6c..9b5fea81f55e 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1595,6 +1595,7 @@ static inline void security_sb_post_pivotroot (struct nameidata *old_nd,
 
 static inline int security_inode_alloc (struct inode *inode)
 {
+	inode->i_security = NULL;
 	return security_ops->inode_alloc_security (inode);
 }
 
-- 
cgit v1.2.3


From ae78bf9c4f5fde3c67e2829505f195d7347ce3e4 Mon Sep 17 00:00:00 2001
From: Chris Mason <mason@suse.com>
Date: Fri, 29 Sep 2006 02:00:03 -0700
Subject: [PATCH] add -o flush for fat

Fat is commonly used on removable media.  Mounting with -o flush tells the
FS to write things to disk as quickly as possible.  It is like -o sync, but
much faster (and not as safe).

Signed-off-by: Chris Mason <mason@suse.com>
Cc: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/fat/file.c            | 13 +++++++++++
 fs/fat/inode.c           | 59 ++++++++++++++++++++++++++++++++++++++++++++++--
 fs/msdos/namei.c         | 11 ++++++++-
 include/linux/msdos_fs.h |  3 +++
 4 files changed, 83 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/fat/file.c b/fs/fat/file.c
index 1ee25232e6af..d50fc47169c1 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -13,6 +13,7 @@
 #include <linux/smp_lock.h>
 #include <linux/buffer_head.h>
 #include <linux/writeback.h>
+#include <linux/blkdev.h>
 
 int fat_generic_ioctl(struct inode *inode, struct file *filp,
 		      unsigned int cmd, unsigned long arg)
@@ -112,6 +113,16 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp,
 	}
 }
 
+static int fat_file_release(struct inode *inode, struct file *filp)
+{
+	if ((filp->f_mode & FMODE_WRITE) &&
+	     MSDOS_SB(inode->i_sb)->options.flush) {
+		fat_flush_inodes(inode->i_sb, inode, NULL);
+		blk_congestion_wait(WRITE, HZ/10);
+	}
+	return 0;
+}
+
 const struct file_operations fat_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= do_sync_read,
@@ -121,6 +132,7 @@ const struct file_operations fat_file_operations = {
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
+	.release	= fat_file_release,
 	.ioctl		= fat_generic_ioctl,
 	.fsync		= file_fsync,
 	.sendfile	= generic_file_sendfile,
@@ -289,6 +301,7 @@ void fat_truncate(struct inode *inode)
 	lock_kernel();
 	fat_free(inode, nr_clusters);
 	unlock_kernel();
+	fat_flush_inodes(inode->i_sb, inode, NULL);
 }
 
 struct inode_operations fat_file_inode_operations = {
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index ab96ae823753..045738032a83 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -24,6 +24,7 @@
 #include <linux/vfs.h>
 #include <linux/parser.h>
 #include <linux/uio.h>
+#include <linux/writeback.h>
 #include <asm/unaligned.h>
 
 #ifndef CONFIG_FAT_DEFAULT_IOCHARSET
@@ -853,7 +854,7 @@ enum {
 	Opt_charset, Opt_shortname_lower, Opt_shortname_win95,
 	Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes,
 	Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes,
-	Opt_obsolate, Opt_err,
+	Opt_obsolate, Opt_flush, Opt_err,
 };
 
 static match_table_t fat_tokens = {
@@ -885,7 +886,8 @@ static match_table_t fat_tokens = {
 	{Opt_obsolate, "cvf_format=%20s"},
 	{Opt_obsolate, "cvf_options=%100s"},
 	{Opt_obsolate, "posix"},
-	{Opt_err, NULL}
+	{Opt_flush, "flush"},
+	{Opt_err, NULL},
 };
 static match_table_t msdos_tokens = {
 	{Opt_nodots, "nodots"},
@@ -1026,6 +1028,9 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 				return 0;
 			opts->codepage = option;
 			break;
+		case Opt_flush:
+			opts->flush = 1;
+			break;
 
 		/* msdos specific */
 		case Opt_dots:
@@ -1425,6 +1430,56 @@ out_fail:
 
 EXPORT_SYMBOL_GPL(fat_fill_super);
 
+/*
+ * helper function for fat_flush_inodes.  This writes both the inode
+ * and the file data blocks, waiting for in flight data blocks before
+ * the start of the call.  It does not wait for any io started
+ * during the call
+ */
+static int writeback_inode(struct inode *inode)
+{
+
+	int ret;
+	struct address_space *mapping = inode->i_mapping;
+	struct writeback_control wbc = {
+	       .sync_mode = WB_SYNC_NONE,
+	      .nr_to_write = 0,
+	};
+	/* if we used WB_SYNC_ALL, sync_inode waits for the io for the
+	* inode to finish.  So WB_SYNC_NONE is sent down to sync_inode
+	* and filemap_fdatawrite is used for the data blocks
+	*/
+	ret = sync_inode(inode, &wbc);
+	if (!ret)
+	       ret = filemap_fdatawrite(mapping);
+	return ret;
+}
+
+/*
+ * write data and metadata corresponding to i1 and i2.  The io is
+ * started but we do not wait for any of it to finish.
+ *
+ * filemap_flush is used for the block device, so if there is a dirty
+ * page for a block already in flight, we will not wait and start the
+ * io over again
+ */
+int fat_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2)
+{
+	int ret = 0;
+	if (!MSDOS_SB(sb)->options.flush)
+		return 0;
+	if (i1)
+		ret = writeback_inode(i1);
+	if (!ret && i2)
+		ret = writeback_inode(i2);
+	if (!ret && sb) {
+		struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping;
+		ret = filemap_flush(mapping);
+	}
+	return ret;
+}
+EXPORT_SYMBOL_GPL(fat_flush_inodes);
+
 static int __init init_fat_fs(void)
 {
 	int err;
diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index 9e44158a7540..d220165d4918 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -280,7 +280,7 @@ static int msdos_create(struct inode *dir, struct dentry *dentry, int mode,
 			struct nameidata *nd)
 {
 	struct super_block *sb = dir->i_sb;
-	struct inode *inode;
+	struct inode *inode = NULL;
 	struct fat_slot_info sinfo;
 	struct timespec ts;
 	unsigned char msdos_name[MSDOS_NAME];
@@ -316,6 +316,8 @@ static int msdos_create(struct inode *dir, struct dentry *dentry, int mode,
 	d_instantiate(dentry, inode);
 out:
 	unlock_kernel();
+	if (!err)
+		err = fat_flush_inodes(sb, dir, inode);
 	return err;
 }
 
@@ -348,6 +350,8 @@ static int msdos_rmdir(struct inode *dir, struct dentry *dentry)
 	fat_detach(inode);
 out:
 	unlock_kernel();
+	if (!err)
+		err = fat_flush_inodes(inode->i_sb, dir, inode);
 
 	return err;
 }
@@ -401,6 +405,7 @@ static int msdos_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	d_instantiate(dentry, inode);
 
 	unlock_kernel();
+	fat_flush_inodes(sb, dir, inode);
 	return 0;
 
 out_free:
@@ -430,6 +435,8 @@ static int msdos_unlink(struct inode *dir, struct dentry *dentry)
 	fat_detach(inode);
 out:
 	unlock_kernel();
+	if (!err)
+		err = fat_flush_inodes(inode->i_sb, dir, inode);
 
 	return err;
 }
@@ -635,6 +642,8 @@ static int msdos_rename(struct inode *old_dir, struct dentry *old_dentry,
 			      new_dir, new_msdos_name, new_dentry, is_hid);
 out:
 	unlock_kernel();
+	if (!err)
+		err = fat_flush_inodes(old_dir->i_sb, old_dir, new_dir);
 	return err;
 }
 
diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index bae62d62dc3e..ce6c85815cbd 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -204,6 +204,7 @@ struct fat_mount_options {
 		 unicode_xlate:1, /* create escape sequences for unhandled Unicode */
 		 numtail:1,       /* Does first alias have a numeric '~1' type tail? */
 		 atari:1,         /* Use Atari GEMDOS variation of MS-DOS fs */
+		 flush:1,	  /* write things quickly */
 		 nocase:1;	  /* Does this need case conversion? 0=need case conversion*/
 };
 
@@ -412,6 +413,8 @@ extern int fat_sync_inode(struct inode *inode);
 extern int fat_fill_super(struct super_block *sb, void *data, int silent,
 			struct inode_operations *fs_dir_inode_ops, int isvfat);
 
+extern int fat_flush_inodes(struct super_block *sb, struct inode *i1,
+		            struct inode *i2);
 /* fat/misc.c */
 extern void fat_fs_panic(struct super_block *s, const char *fmt, ...);
 extern void fat_clusters_flush(struct super_block *sb);
-- 
cgit v1.2.3


From ca9bda00b4aafc42cd3d1b9d32934463e2993b4c Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 29 Sep 2006 02:00:03 -0700
Subject: [PATCH] tty locking on resize

The current kernel serializes console resizes but does not serialize the
resize against the tty structure updates.  This means that while two
parallel resizes cannot mess up the console you can get incorrect results
reported.

Secondly while doing this I added vc_lock_resize() to lock and resize the
console.  This leaves all knowledge of the console_sem in the vt/console
driver and kicks it out of the tty layer, which is good

Thirdly while doing this I decided I couldn't stand "disallocate" any
longer so I switched it to "deallocate".

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/selection.c |  2 +-
 drivers/char/tty_io.c    | 29 +++++++++++++++--------------
 drivers/char/vt.c        | 12 +++++++++++-
 drivers/char/vt_ioctl.c  | 17 +++++++----------
 include/linux/vt_kern.h  |  3 ++-
 5 files changed, 36 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/drivers/char/selection.c b/drivers/char/selection.c
index 71093a9fc462..74cff839c857 100644
--- a/drivers/char/selection.c
+++ b/drivers/char/selection.c
@@ -33,7 +33,7 @@ extern void poke_blanked_console(void);
 
 /* Variables for selection control. */
 /* Use a dynamic buffer, instead of static (Dec 1994) */
-struct vc_data *sel_cons;		/* must not be disallocated */
+struct vc_data *sel_cons;		/* must not be deallocated */
 static volatile int sel_start = -1; 	/* cleared by clear_selection */
 static int sel_end;
 static int sel_buffer_lth;
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index eb881cfa53e0..2a1e95b0f282 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -2770,12 +2770,11 @@ static int tiocgwinsz(struct tty_struct *tty, struct winsize __user * arg)
  *	actually has driver level meaning and triggers a VC resize.
  *
  *	Locking:
- *		The console_sem is used to ensure we do not try and resize
- *	the console twice at once.
- *	FIXME: Two racing size sets may leave the console and kernel
- *		parameters disagreeing. Is this exploitable ?
- *	FIXME: Random values racing a window size get is wrong
- *	should lock here against that
+ *		Called function use the console_sem is used to ensure we do
+ *	not try and resize the console twice at once.
+ *		The tty->termios_sem is used to ensure we don't double
+ *	resize and get confused. Lock order - tty->termios.sem before
+ *	console sem
  */
 
 static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
@@ -2785,17 +2784,17 @@ static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
 
 	if (copy_from_user(&tmp_ws, arg, sizeof(*arg)))
 		return -EFAULT;
+
+	down(&tty->termios_sem);
 	if (!memcmp(&tmp_ws, &tty->winsize, sizeof(*arg)))
-		return 0;
+		goto done;
+
 #ifdef CONFIG_VT
 	if (tty->driver->type == TTY_DRIVER_TYPE_CONSOLE) {
-		int rc;
-
-		acquire_console_sem();
-		rc = vc_resize(tty->driver_data, tmp_ws.ws_col, tmp_ws.ws_row);
-		release_console_sem();
-		if (rc)
-			return -ENXIO;
+		if (vc_lock_resize(tty->driver_data, tmp_ws.ws_col, tmp_ws.ws_row)) {
+			up(&tty->termios_sem);
+ 			return -ENXIO;
+		}
 	}
 #endif
 	if (tty->pgrp > 0)
@@ -2804,6 +2803,8 @@ static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
 		kill_pg(real_tty->pgrp, SIGWINCH, 1);
 	tty->winsize = tmp_ws;
 	real_tty->winsize = tmp_ws;
+done:
+	up(&tty->termios_sem);
 	return 0;
 }
 
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 0fca83ededff..b49f03375439 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -885,8 +885,17 @@ int vc_resize(struct vc_data *vc, unsigned int cols, unsigned int lines)
 	return err;
 }
 
+int vc_lock_resize(struct vc_data *vc, unsigned int cols, unsigned int lines)
+{
+	int rc;
+
+	acquire_console_sem();
+	rc = vc_resize(vc, cols, lines);
+	release_console_sem();
+	return rc;
+}
 
-void vc_disallocate(unsigned int currcons)
+void vc_deallocate(unsigned int currcons)
 {
 	WARN_CONSOLE_UNLOCKED();
 
@@ -3790,6 +3799,7 @@ EXPORT_SYMBOL(default_blu);
 EXPORT_SYMBOL(update_region);
 EXPORT_SYMBOL(redraw_screen);
 EXPORT_SYMBOL(vc_resize);
+EXPORT_SYMBOL(vc_lock_resize);
 EXPORT_SYMBOL(fg_console);
 EXPORT_SYMBOL(console_blank_hook);
 EXPORT_SYMBOL(console_blanked);
diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c
index a5628a8b6620..a53e382cc107 100644
--- a/drivers/char/vt_ioctl.c
+++ b/drivers/char/vt_ioctl.c
@@ -96,7 +96,7 @@ do_kdsk_ioctl(int cmd, struct kbentry __user *user_kbe, int perm, struct kbd_str
 		if (!perm)
 			return -EPERM;
 		if (!i && v == K_NOSUCHMAP) {
-			/* disallocate map */
+			/* deallocate map */
 			key_map = key_maps[s];
 			if (s && key_map) {
 			    key_maps[s] = NULL;
@@ -819,20 +819,20 @@ int vt_ioctl(struct tty_struct *tty, struct file * file,
 		if (arg > MAX_NR_CONSOLES)
 			return -ENXIO;
 		if (arg == 0) {
-		    /* disallocate all unused consoles, but leave 0 */
+		    /* deallocate all unused consoles, but leave 0 */
 			acquire_console_sem();
 			for (i=1; i<MAX_NR_CONSOLES; i++)
 				if (! VT_BUSY(i))
-					vc_disallocate(i);
+					vc_deallocate(i);
 			release_console_sem();
 		} else {
-			/* disallocate a single console, if possible */
+			/* deallocate a single console, if possible */
 			arg--;
 			if (VT_BUSY(arg))
 				return -EBUSY;
 			if (arg) {			      /* leave 0 */
 				acquire_console_sem();
-				vc_disallocate(arg);
+				vc_deallocate(arg);
 				release_console_sem();
 			}
 		}
@@ -847,11 +847,8 @@ int vt_ioctl(struct tty_struct *tty, struct file * file,
 		if (get_user(ll, &vtsizes->v_rows) ||
 		    get_user(cc, &vtsizes->v_cols))
 			return -EFAULT;
-		for (i = 0; i < MAX_NR_CONSOLES; i++) {
-			acquire_console_sem();
-			vc_resize(vc_cons[i].d, cc, ll);
-			release_console_sem();
-		}
+		for (i = 0; i < MAX_NR_CONSOLES; i++)
+			vc_lock_resize(vc_cons[i].d, cc, ll);
 		return 0;
 	}
 
diff --git a/include/linux/vt_kern.h b/include/linux/vt_kern.h
index 918a29763aea..1009d3fe1fc2 100644
--- a/include/linux/vt_kern.h
+++ b/include/linux/vt_kern.h
@@ -33,7 +33,8 @@ extern int fg_console, last_console, want_console;
 int vc_allocate(unsigned int console);
 int vc_cons_allocated(unsigned int console);
 int vc_resize(struct vc_data *vc, unsigned int cols, unsigned int lines);
-void vc_disallocate(unsigned int console);
+int vc_lock_resize(struct vc_data *vc, unsigned int cols, unsigned int lines);
+void vc_deallocate(unsigned int console);
 void reset_palette(struct vc_data *vc);
 void do_blank_screen(int entering_gfx);
 void do_unblank_screen(int leaving_gfx);
-- 
cgit v1.2.3


From 3b9b8ab65d8eed784b9164d03807cb2bda7b5cd6 Mon Sep 17 00:00:00 2001
From: Kirill Korotaev <dev@sw.ru>
Date: Fri, 29 Sep 2006 02:00:05 -0700
Subject: [PATCH] Fix unserialized task->files changing

Fixed race on put_files_struct on exec with proc.  Restoring files on
current on error path may lead to proc having a pointer to already kfree-d
files_struct.

->files changing at exit.c and khtread.c are safe as exit_files() makes all
things under lock.

Found during OpenVZ stress testing.

[akpm@osdl.org: add export]
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/binfmt_elf.c      |  6 ++----
 fs/binfmt_misc.c     |  6 ++----
 fs/exec.c            |  3 +--
 include/linux/file.h |  1 +
 kernel/exit.c        | 12 ++++++++++++
 5 files changed, 18 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index dfd8cfb7fb5d..bb43da5cde5c 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1038,10 +1038,8 @@ out_free_interp:
 out_free_file:
 	sys_close(elf_exec_fileno);
 out_free_fh:
-	if (files) {
-		put_files_struct(current->files);
-		current->files = files;
-	}
+	if (files)
+		reset_files_struct(current, files);
 out_free_ph:
 	kfree(elf_phdata);
 	goto out;
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 66ba137f8661..1713c48fef54 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -215,10 +215,8 @@ _error:
 	bprm->interp_flags = 0;
 	bprm->interp_data = 0;
 _unshare:
-	if (files) {
-		put_files_struct(current->files);
-		current->files = files;
-	}
+	if (files)
+		reset_files_struct(current, files);
 	goto _ret;
 }
 
diff --git a/fs/exec.c b/fs/exec.c
index 97df6e0aeaee..a8efe35176b0 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -898,8 +898,7 @@ int flush_old_exec(struct linux_binprm * bprm)
 	return 0;
 
 mmap_failed:
-	put_files_struct(current->files);
-	current->files = files;
+	reset_files_struct(current, files);
 out:
 	return retval;
 }
diff --git a/include/linux/file.h b/include/linux/file.h
index 9f7c2513866f..74183e6f7f45 100644
--- a/include/linux/file.h
+++ b/include/linux/file.h
@@ -112,5 +112,6 @@ struct task_struct;
 
 struct files_struct *get_files_struct(struct task_struct *);
 void FASTCALL(put_files_struct(struct files_struct *fs));
+void reset_files_struct(struct task_struct *, struct files_struct *);
 
 #endif /* __LINUX_FILE_H */
diff --git a/kernel/exit.c b/kernel/exit.c
index d891883420f7..4b6fb054b25d 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -487,6 +487,18 @@ void fastcall put_files_struct(struct files_struct *files)
 
 EXPORT_SYMBOL(put_files_struct);
 
+void reset_files_struct(struct task_struct *tsk, struct files_struct *files)
+{
+	struct files_struct *old;
+
+	old = tsk->files;
+	task_lock(tsk);
+	tsk->files = files;
+	task_unlock(tsk);
+	put_files_struct(old);
+}
+EXPORT_SYMBOL(reset_files_struct);
+
 static inline void __exit_files(struct task_struct *tsk)
 {
 	struct files_struct * files = tsk->files;
-- 
cgit v1.2.3


From f400e198b2ed26ce55b22a1412ded0896e7516ac Mon Sep 17 00:00:00 2001
From: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Date: Fri, 29 Sep 2006 02:00:07 -0700
Subject: [PATCH] pidspace: is_init()

This is an updated version of Eric Biederman's is_init() patch.
(http://lkml.org/lkml/2006/2/6/280).  It applies cleanly to 2.6.18-rc3 and
replaces a few more instances of ->pid == 1 with is_init().

Further, is_init() checks pid and thus removes dependency on Eric's other
patches for now.

Eric's original description:

	There are a lot of places in the kernel where we test for init
	because we give it special properties.  Most  significantly init
	must not die.  This results in code all over the kernel test
	->pid == 1.

	Introduce is_init to capture this case.

	With multiple pid spaces for all of the cases affected we are
	looking for only the first process on the system, not some other
	process that has pid == 1.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Cedric Le Goater <clg@fr.ibm.com>
Cc: <lxc-devel@lists.sourceforge.net>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/mm/fault.c                |  2 +-
 arch/arm/mm/fault.c                  |  2 +-
 arch/arm26/mm/fault.c                |  2 +-
 arch/i386/lib/usercopy.c             |  2 +-
 arch/i386/mm/fault.c                 |  2 +-
 arch/ia64/mm/fault.c                 |  2 +-
 arch/m32r/mm/fault.c                 |  2 +-
 arch/m68k/mm/fault.c                 |  2 +-
 arch/mips/mm/fault.c                 |  2 +-
 arch/powerpc/mm/fault.c              |  2 +-
 arch/powerpc/platforms/pseries/ras.c |  2 +-
 arch/ppc/kernel/traps.c              |  2 +-
 arch/ppc/mm/fault.c                  |  2 +-
 arch/s390/mm/fault.c                 |  2 +-
 arch/sh/mm/fault.c                   |  2 +-
 arch/sh64/mm/fault.c                 |  6 +++---
 arch/um/kernel/trap.c                |  2 +-
 arch/x86_64/mm/fault.c               |  4 ++--
 arch/xtensa/mm/fault.c               |  2 +-
 drivers/char/sysrq.c                 |  2 +-
 include/linux/sched.h                | 10 ++++++++++
 kernel/capability.c                  |  2 +-
 kernel/cpuset.c                      |  2 +-
 kernel/exit.c                        |  2 +-
 kernel/kexec.c                       |  2 +-
 kernel/ptrace.c                      |  1 +
 kernel/sysctl.c                      |  2 +-
 mm/oom_kill.c                        |  2 +-
 security/commoncap.c                 |  2 +-
 29 files changed, 41 insertions(+), 30 deletions(-)

(limited to 'include')

diff --git a/arch/alpha/mm/fault.c b/arch/alpha/mm/fault.c
index 622dabd84680..8871529a34e2 100644
--- a/arch/alpha/mm/fault.c
+++ b/arch/alpha/mm/fault.c
@@ -193,7 +193,7 @@ do_page_fault(unsigned long address, unsigned long mmcsr,
 	/* We ran out of memory, or some other thing happened to us that
 	   made us unable to handle the page fault gracefully.  */
  out_of_memory:
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index a5b33ff3924e..5e658a874498 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -198,7 +198,7 @@ survive:
 		return fault;
 	}
 
-	if (tsk->pid != 1)
+	if (!is_init(tsk))
 		goto out;
 
 	/*
diff --git a/arch/arm26/mm/fault.c b/arch/arm26/mm/fault.c
index a7c4cc922095..a1f6d8a9cc32 100644
--- a/arch/arm26/mm/fault.c
+++ b/arch/arm26/mm/fault.c
@@ -185,7 +185,7 @@ survive:
 	}
 
 	fault = -3; /* out of memory */
-	if (tsk->pid != 1)
+	if (!is_init(tsk))
 		goto out;
 
 	/*
diff --git a/arch/i386/lib/usercopy.c b/arch/i386/lib/usercopy.c
index efc7e7d5f4d0..08502fc6d0cb 100644
--- a/arch/i386/lib/usercopy.c
+++ b/arch/i386/lib/usercopy.c
@@ -739,7 +739,7 @@ survive:
 			retval = get_user_pages(current, current->mm,
 					(unsigned long )to, 1, 1, 0, &pg, NULL);
 
-			if (retval == -ENOMEM && current->pid == 1) {
+			if (retval == -ENOMEM && is_init(current)) {
 				up_read(&current->mm->mmap_sem);
 				blk_congestion_wait(WRITE, HZ/50);
 				goto survive;
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index 50d8617391dd..2581575786c1 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -589,7 +589,7 @@ no_context:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (tsk->pid == 1) {
+	if (is_init(tsk)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c
index d8b1b4ac7f26..59f3ab937615 100644
--- a/arch/ia64/mm/fault.c
+++ b/arch/ia64/mm/fault.c
@@ -280,7 +280,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re
 
   out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/m32r/mm/fault.c b/arch/m32r/mm/fault.c
index dc18a33eefef..8d5f551b5754 100644
--- a/arch/m32r/mm/fault.c
+++ b/arch/m32r/mm/fault.c
@@ -299,7 +299,7 @@ no_context:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (tsk->pid == 1) {
+	if (is_init(tsk)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/m68k/mm/fault.c b/arch/m68k/mm/fault.c
index 5e2d87c10c87..911f2ce3f53e 100644
--- a/arch/m68k/mm/fault.c
+++ b/arch/m68k/mm/fault.c
@@ -181,7 +181,7 @@ good_area:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c
index a4f8c45c4e8e..8423d8590779 100644
--- a/arch/mips/mm/fault.c
+++ b/arch/mips/mm/fault.c
@@ -171,7 +171,7 @@ no_context:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (tsk->pid == 1) {
+	if (is_init(tsk)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index 77953f41d754..e8fa50624b70 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -386,7 +386,7 @@ bad_area_nosemaphore:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c
index 903115d67fdc..311ed1993fc0 100644
--- a/arch/powerpc/platforms/pseries/ras.c
+++ b/arch/powerpc/platforms/pseries/ras.c
@@ -337,7 +337,7 @@ static int recover_mce(struct pt_regs *regs, struct rtas_error_log * err)
 		   err->disposition == RTAS_DISP_NOT_RECOVERED &&
 		   err->target == RTAS_TARGET_MEMORY &&
 		   err->type == RTAS_TYPE_ECC_UNCORR &&
-		   !(current->pid == 0 || current->pid == 1)) {
+		   !(current->pid == 0 || is_init(current))) {
 		/* Kill off a user process with an ECC error */
 		printk(KERN_ERR "MCE: uncorrectable ecc error for pid %d\n",
 		       current->pid);
diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c
index d7a433049b48..aafc8e8893d1 100644
--- a/arch/ppc/kernel/traps.c
+++ b/arch/ppc/kernel/traps.c
@@ -119,7 +119,7 @@ void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr)
 	 * generate the same exception over and over again and we get
 	 * nowhere.  Better to kill it and let the kernel panic.
 	 */
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		__sighandler_t handler;
 
 		spin_lock_irq(&current->sighand->siglock);
diff --git a/arch/ppc/mm/fault.c b/arch/ppc/mm/fault.c
index bc776beb3136..465f451f3bc3 100644
--- a/arch/ppc/mm/fault.c
+++ b/arch/ppc/mm/fault.c
@@ -291,7 +291,7 @@ bad_area:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index f2b9a84dc2bf..9c3c19fe62fc 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -353,7 +353,7 @@ no_context:
 */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (tsk->pid == 1) {
+	if (is_init(tsk)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 507f28914706..68663b8f99ae 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -149,7 +149,7 @@ no_context:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/sh64/mm/fault.c b/arch/sh64/mm/fault.c
index f08d0eaf6497..8e2f6c28b739 100644
--- a/arch/sh64/mm/fault.c
+++ b/arch/sh64/mm/fault.c
@@ -277,7 +277,7 @@ bad_area:
 			show_regs(regs);
 #endif
 		}
-		if (tsk->pid == 1) {
+		if (is_init(tsk)) {
 			panic("INIT had user mode bad_area\n");
 		}
 		tsk->thread.address = address;
@@ -319,14 +319,14 @@ no_context:
  * us unable to handle the page fault gracefully.
  */
 out_of_memory:
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		panic("INIT out of memory\n");
 		yield();
 		goto survive;
 	}
 	printk("fault:Out of memory\n");
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index 61a23fff4395..c7b195c7e51f 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -120,7 +120,7 @@ out_nosemaphore:
  * us unable to handle the page fault gracefully.
  */
 out_of_memory:
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		up_read(&mm->mmap_sem);
 		yield();
 		down_read(&mm->mmap_sem);
diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index 9ba54cc2b5f6..3751b4788e28 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -244,7 +244,7 @@ static int is_errata93(struct pt_regs *regs, unsigned long address)
 
 int unhandled_signal(struct task_struct *tsk, int sig)
 {
-	if (tsk->pid == 1)
+	if (is_init(tsk))
 		return 1;
 	if (tsk->ptrace & PT_PTRACED)
 		return 0;
@@ -580,7 +580,7 @@ no_context:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) { 
+	if (is_init(current)) {
 		yield();
 		goto again;
 	}
diff --git a/arch/xtensa/mm/fault.c b/arch/xtensa/mm/fault.c
index a945a33e85a1..dd0dbec2e57e 100644
--- a/arch/xtensa/mm/fault.c
+++ b/arch/xtensa/mm/fault.c
@@ -144,7 +144,7 @@ bad_area:
 	 */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (current->pid == 1) {
+	if (is_init(current)) {
 		yield();
 		down_read(&mm->mmap_sem);
 		goto survive;
diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index ee3ca8f1768e..0ad6cb081db4 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -208,7 +208,7 @@ static void send_sig_all(int sig)
 	struct task_struct *p;
 
 	for_each_process(p) {
-		if (p->mm && p->pid != 1)
+		if (p->mm && !is_init(p))
 			/* Not swapper, init nor kernel thread */
 			force_sig(sig, p);
 	}
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 3696f2f7126d..ed2af8671589 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1033,6 +1033,16 @@ static inline int pid_alive(struct task_struct *p)
 	return p->pids[PIDTYPE_PID].pid != NULL;
 }
 
+/**
+ * is_init - check if a task structure is the first user space
+ *	     task the kernel created.
+ * @p: Task structure to be checked.
+ */
+static inline int is_init(struct task_struct *tsk)
+{
+	return tsk->pid == 1;
+}
+
 extern void free_task(struct task_struct *tsk);
 #define get_task_struct(tsk) do { atomic_inc(&(tsk)->usage); } while(0)
 
diff --git a/kernel/capability.c b/kernel/capability.c
index c7685ad00a97..edb845a6e84a 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -133,7 +133,7 @@ static inline int cap_set_all(kernel_cap_t *effective,
      int found = 0;
 
      do_each_thread(g, target) {
-             if (target == current || target->pid == 1)
+             if (target == current || is_init(target))
                      continue;
              found = 1;
 	     if (security_capset_check(target, effective, inheritable,
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 1b32c2c04c15..584bb4e6c042 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -240,7 +240,7 @@ static struct super_block *cpuset_sb;
  * A cpuset can only be deleted if both its 'count' of using tasks
  * is zero, and its list of 'children' cpusets is empty.  Since all
  * tasks in the system use _some_ cpuset, and since there is always at
- * least one task in the system (init, pid == 1), therefore, top_cpuset
+ * least one task in the system (init), therefore, top_cpuset
  * always has either children cpusets and/or using tasks.  So we don't
  * need a special hack to ensure that top_cpuset cannot be deleted.
  *
diff --git a/kernel/exit.c b/kernel/exit.c
index 4b6fb054b25d..9961192d6055 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -219,7 +219,7 @@ static int will_become_orphaned_pgrp(int pgrp, struct task_struct *ignored_task)
 	do_each_task_pid(pgrp, PIDTYPE_PGID, p) {
 		if (p == ignored_task
 				|| p->exit_state
-				|| p->real_parent->pid == 1)
+				|| is_init(p->real_parent))
 			continue;
 		if (process_group(p->real_parent) != pgrp
 			    && p->real_parent->signal->session == p->signal->session) {
diff --git a/kernel/kexec.c b/kernel/kexec.c
index 50087ecf337e..37cad75cf494 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -40,7 +40,7 @@ struct resource crashk_res = {
 
 int kexec_should_crash(struct task_struct *p)
 {
-	if (in_interrupt() || !p->pid || p->pid == 1 || panic_on_oops)
+	if (in_interrupt() || !p->pid || is_init(p) || panic_on_oops)
 		return 1;
 	return 0;
 }
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 8aad0331d82e..4d50e06fd745 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -440,6 +440,7 @@ struct task_struct *ptrace_get_task_struct(pid_t pid)
 	child = find_task_by_pid(pid);
 	if (child)
 		get_task_struct(child);
+
 	read_unlock(&tasklist_lock);
 	if (!child)
 		return ERR_PTR(-ESRCH);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 8bfa7d117c54..9535a3839930 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1915,7 +1915,7 @@ int proc_dointvec_bset(ctl_table *table, int write, struct file *filp,
 		return -EPERM;
 	}
 
-	op = (current->pid == 1) ? OP_SET : OP_AND;
+	op = is_init(current) ? OP_SET : OP_AND;
 	return do_proc_dointvec(table,write,filp,buffer,lenp,ppos,
 				do_proc_dointvec_bset_conv,&op);
 }
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index bada3d03119f..f3dd79c1c367 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -255,7 +255,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
  */
 static void __oom_kill_task(struct task_struct *p, const char *message)
 {
-	if (p->pid == 1) {
+	if (is_init(p)) {
 		WARN_ON(1);
 		printk(KERN_WARNING "tried to kill init!\n");
 		return;
diff --git a/security/commoncap.c b/security/commoncap.c
index f50fc298cf80..5a5ef5ca7ea9 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -169,7 +169,7 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 	/* For init, we want to retain the capabilities set
 	 * in the init_task struct. Thus we skip the usual
 	 * capability rules */
-	if (current->pid != 1) {
+	if (!is_init(current)) {
 		current->cap_permitted = new_permitted;
 		current->cap_effective =
 		    cap_intersect (new_permitted, bprm->cap_effective);
-- 
cgit v1.2.3


From 3ca212b813299899d2968aa0a24a797c3746f5ec Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Fri, 29 Sep 2006 02:00:14 -0700
Subject: [PATCH] Remove another config.h

After the asm/ uses of #include <linux/config.h> this one is the next
biggest source of noise.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/vmstat.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 176c7f797339..c89df55f6e03 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -3,7 +3,6 @@
 
 #include <linux/types.h>
 #include <linux/percpu.h>
-#include <linux/config.h>
 #include <linux/mmzone.h>
 #include <asm/atomic.h>
 
-- 
cgit v1.2.3


From af410fc13d95f079910fc3dca7496590c3275967 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Fri, 29 Sep 2006 02:00:14 -0700
Subject: [PATCH] make leds.h include relevant headers

Make it possible to include linux/leds.h without first including list.h and
spinlock.h.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/leds.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include')

diff --git a/include/linux/leds.h b/include/linux/leds.h
index dc23c7c639f3..88afceffb7cb 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -12,6 +12,9 @@
 #ifndef __LINUX_LEDS_H_INCLUDED
 #define __LINUX_LEDS_H_INCLUDED
 
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
 struct device;
 struct class_device;
 /*
-- 
cgit v1.2.3


From 1711ef3866b0360e102327389fe4b76c849bbe83 Mon Sep 17 00:00:00 2001
From: Toyo Abe <toyoa@mvista.com>
Date: Fri, 29 Sep 2006 02:00:28 -0700
Subject: [PATCH] posix-timers: Fix clock_nanosleep() doesn't return the
 remaining time in compatibility mode

The clock_nanosleep() function does not return the time remaining when the
sleep is interrupted by a signal.

This patch creates a new call out, compat_clock_nanosleep_restart(), which
handles returning the remaining time after a sleep is interrupted.  This
patch revives clock_nanosleep_restart().  It is now accessed via the new
call out.  The compat_clock_nanosleep_restart() is used for compatibility
access.

Since this is implemented in compatibility mode the normal path is
virtually unaffected - no real performance impact.

Signed-off-by: Toyo Abe <toyoa@mvista.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/hrtimer.h      |  1 +
 include/linux/posix-timers.h |  4 ++++
 kernel/compat.c              | 33 +++++++++++++++++++++++++++++++++
 kernel/hrtimer.c             | 20 ++++++++++----------
 kernel/posix-cpu-timers.c    | 17 ++++++++++++-----
 kernel/posix-timers.c        | 21 +++++++++++++++++++++
 6 files changed, 81 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 4fc379de6c2f..fca93025ab51 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -138,6 +138,7 @@ extern long hrtimer_nanosleep(struct timespec *rqtp,
 			      struct timespec __user *rmtp,
 			      const enum hrtimer_mode mode,
 			      const clockid_t clockid);
+extern long hrtimer_nanosleep_restart(struct restart_block *restart_block);
 
 extern void hrtimer_init_sleeper(struct hrtimer_sleeper *sl,
 				 struct task_struct *tsk);
diff --git a/include/linux/posix-timers.h b/include/linux/posix-timers.h
index 95572c434bc9..a7dd38f30ade 100644
--- a/include/linux/posix-timers.h
+++ b/include/linux/posix-timers.h
@@ -72,6 +72,7 @@ struct k_clock {
 	int (*timer_create) (struct k_itimer *timer);
 	int (*nsleep) (const clockid_t which_clock, int flags,
 		       struct timespec *, struct timespec __user *);
+	long (*nsleep_restart) (struct restart_block *restart_block);
 	int (*timer_set) (struct k_itimer * timr, int flags,
 			  struct itimerspec * new_setting,
 			  struct itimerspec * old_setting);
@@ -97,6 +98,7 @@ int posix_cpu_clock_set(const clockid_t which_clock, const struct timespec *ts);
 int posix_cpu_timer_create(struct k_itimer *timer);
 int posix_cpu_nsleep(const clockid_t which_clock, int flags,
 		     struct timespec *rqtp, struct timespec __user *rmtp);
+long posix_cpu_nsleep_restart(struct restart_block *restart_block);
 int posix_cpu_timer_set(struct k_itimer *timer, int flags,
 			struct itimerspec *new, struct itimerspec *old);
 int posix_cpu_timer_del(struct k_itimer *timer);
@@ -111,4 +113,6 @@ void posix_cpu_timers_exit_group(struct task_struct *task);
 void set_process_cpu_timer(struct task_struct *task, unsigned int clock_idx,
 			   cputime_t *newval, cputime_t *oldval);
 
+long clock_nanosleep_restart(struct restart_block *restart_block);
+
 #endif
diff --git a/kernel/compat.c b/kernel/compat.c
index 126dee9530aa..75573e5d27b0 100644
--- a/kernel/compat.c
+++ b/kernel/compat.c
@@ -22,6 +22,7 @@
 #include <linux/security.h>
 #include <linux/timex.h>
 #include <linux/migrate.h>
+#include <linux/posix-timers.h>
 
 #include <asm/uaccess.h>
 
@@ -601,6 +602,30 @@ long compat_sys_clock_getres(clockid_t which_clock,
 	return err;
 } 
 
+static long compat_clock_nanosleep_restart(struct restart_block *restart)
+{
+	long err;
+	mm_segment_t oldfs;
+	struct timespec tu;
+	struct compat_timespec *rmtp = (struct compat_timespec *)(restart->arg1);
+
+	restart->arg1 = (unsigned long) &tu;
+	oldfs = get_fs();
+	set_fs(KERNEL_DS);
+	err = clock_nanosleep_restart(restart);
+	set_fs(oldfs);
+
+	if ((err == -ERESTART_RESTARTBLOCK) && rmtp &&
+	    put_compat_timespec(&tu, rmtp))
+		return -EFAULT;
+
+	if (err == -ERESTART_RESTARTBLOCK) {
+		restart->fn = compat_clock_nanosleep_restart;
+		restart->arg1 = (unsigned long) rmtp;
+	}
+	return err;
+}
+
 long compat_sys_clock_nanosleep(clockid_t which_clock, int flags,
 			    struct compat_timespec __user *rqtp,
 			    struct compat_timespec __user *rmtp)
@@ -608,6 +633,7 @@ long compat_sys_clock_nanosleep(clockid_t which_clock, int flags,
 	long err;
 	mm_segment_t oldfs;
 	struct timespec in, out; 
+	struct restart_block *restart;
 
 	if (get_compat_timespec(&in, rqtp)) 
 		return -EFAULT;
@@ -618,9 +644,16 @@ long compat_sys_clock_nanosleep(clockid_t which_clock, int flags,
 				  (struct timespec __user *) &in,
 				  (struct timespec __user *) &out);
 	set_fs(oldfs);
+
 	if ((err == -ERESTART_RESTARTBLOCK) && rmtp &&
 	    put_compat_timespec(&out, rmtp))
 		return -EFAULT;
+
+	if (err == -ERESTART_RESTARTBLOCK) {
+		restart = &current_thread_info()->restart_block;
+		restart->fn = compat_clock_nanosleep_restart;
+		restart->arg1 = (unsigned long) rmtp;
+	}
 	return err;	
 } 
 
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 21c38a7e666b..d0ba190dfeb6 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -693,7 +693,7 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod
 	return t->task == NULL;
 }
 
-static long __sched nanosleep_restart(struct restart_block *restart)
+long __sched hrtimer_nanosleep_restart(struct restart_block *restart)
 {
 	struct hrtimer_sleeper t;
 	struct timespec __user *rmtp;
@@ -702,13 +702,13 @@ static long __sched nanosleep_restart(struct restart_block *restart)
 
 	restart->fn = do_no_restart_syscall;
 
-	hrtimer_init(&t.timer, restart->arg3, HRTIMER_ABS);
-	t.timer.expires.tv64 = ((u64)restart->arg1 << 32) | (u64) restart->arg0;
+	hrtimer_init(&t.timer, restart->arg0, HRTIMER_ABS);
+	t.timer.expires.tv64 = ((u64)restart->arg3 << 32) | (u64) restart->arg2;
 
 	if (do_nanosleep(&t, HRTIMER_ABS))
 		return 0;
 
-	rmtp = (struct timespec __user *) restart->arg2;
+	rmtp = (struct timespec __user *) restart->arg1;
 	if (rmtp) {
 		time = ktime_sub(t.timer.expires, t.timer.base->get_time());
 		if (time.tv64 <= 0)
@@ -718,7 +718,7 @@ static long __sched nanosleep_restart(struct restart_block *restart)
 			return -EFAULT;
 	}
 
-	restart->fn = nanosleep_restart;
+	restart->fn = hrtimer_nanosleep_restart;
 
 	/* The other values in restart are already filled in */
 	return -ERESTART_RESTARTBLOCK;
@@ -751,11 +751,11 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
 	}
 
 	restart = &current_thread_info()->restart_block;
-	restart->fn = nanosleep_restart;
-	restart->arg0 = t.timer.expires.tv64 & 0xFFFFFFFF;
-	restart->arg1 = t.timer.expires.tv64 >> 32;
-	restart->arg2 = (unsigned long) rmtp;
-	restart->arg3 = (unsigned long) t.timer.base->index;
+	restart->fn = hrtimer_nanosleep_restart;
+	restart->arg0 = (unsigned long) t.timer.base->index;
+	restart->arg1 = (unsigned long) rmtp;
+	restart->arg2 = t.timer.expires.tv64 & 0xFFFFFFFF;
+	restart->arg3 = t.timer.expires.tv64 >> 32;
 
 	return -ERESTART_RESTARTBLOCK;
 }
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index d38d9ec3276c..5667fef89dd1 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -1393,8 +1393,6 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx,
 	}
 }
 
-static long posix_cpu_clock_nanosleep_restart(struct restart_block *);
-
 int posix_cpu_nsleep(const clockid_t which_clock, int flags,
 		     struct timespec *rqtp, struct timespec __user *rmtp)
 {
@@ -1471,7 +1469,7 @@ int posix_cpu_nsleep(const clockid_t which_clock, int flags,
 		    copy_to_user(rmtp, &it.it_value, sizeof *rmtp))
 			return -EFAULT;
 
-		restart_block->fn = posix_cpu_clock_nanosleep_restart;
+		restart_block->fn = posix_cpu_nsleep_restart;
 		/* Caller already set restart_block->arg1 */
 		restart_block->arg0 = which_clock;
 		restart_block->arg1 = (unsigned long) rmtp;
@@ -1484,8 +1482,7 @@ int posix_cpu_nsleep(const clockid_t which_clock, int flags,
 	return error;
 }
 
-static long
-posix_cpu_clock_nanosleep_restart(struct restart_block *restart_block)
+long posix_cpu_nsleep_restart(struct restart_block *restart_block)
 {
 	clockid_t which_clock = restart_block->arg0;
 	struct timespec __user *rmtp;
@@ -1524,6 +1521,10 @@ static int process_cpu_nsleep(const clockid_t which_clock, int flags,
 {
 	return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp, rmtp);
 }
+static long process_cpu_nsleep_restart(struct restart_block *restart_block)
+{
+	return -EINVAL;
+}
 static int thread_cpu_clock_getres(const clockid_t which_clock,
 				   struct timespec *tp)
 {
@@ -1544,6 +1545,10 @@ static int thread_cpu_nsleep(const clockid_t which_clock, int flags,
 {
 	return -EINVAL;
 }
+static long thread_cpu_nsleep_restart(struct restart_block *restart_block)
+{
+	return -EINVAL;
+}
 
 static __init int init_posix_cpu_timers(void)
 {
@@ -1553,6 +1558,7 @@ static __init int init_posix_cpu_timers(void)
 		.clock_set = do_posix_clock_nosettime,
 		.timer_create = process_cpu_timer_create,
 		.nsleep = process_cpu_nsleep,
+		.nsleep_restart = process_cpu_nsleep_restart,
 	};
 	struct k_clock thread = {
 		.clock_getres = thread_cpu_clock_getres,
@@ -1560,6 +1566,7 @@ static __init int init_posix_cpu_timers(void)
 		.clock_set = do_posix_clock_nosettime,
 		.timer_create = thread_cpu_timer_create,
 		.nsleep = thread_cpu_nsleep,
+		.nsleep_restart = thread_cpu_nsleep_restart,
 	};
 
 	register_posix_clock(CLOCK_PROCESS_CPUTIME_ID, &process);
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index ac6dc8744429..e5ebcc1ec3a0 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -973,3 +973,24 @@ sys_clock_nanosleep(const clockid_t which_clock, int flags,
 	return CLOCK_DISPATCH(which_clock, nsleep,
 			      (which_clock, flags, &t, rmtp));
 }
+
+/*
+ * nanosleep_restart for monotonic and realtime clocks
+ */
+static int common_nsleep_restart(struct restart_block *restart_block)
+{
+	return hrtimer_nanosleep_restart(restart_block);
+}
+
+/*
+ * This will restart clock_nanosleep. This is required only by
+ * compat_clock_nanosleep_restart for now.
+ */
+long
+clock_nanosleep_restart(struct restart_block *restart_block)
+{
+	clockid_t which_clock = restart_block->arg0;
+
+	return CLOCK_DISPATCH(which_clock, nsleep_restart,
+			      (restart_block));
+}
-- 
cgit v1.2.3


From 3171a0305d62e6627a24bff35af4f997e4988a80 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Fri, 29 Sep 2006 02:00:32 -0700
Subject: [PATCH] simplify update_times (avoid jiffies/jiffies_64 aliasing
 problem)

Pass ticks to do_timer() and update_times(), and adjust x86_64 and s390
timer interrupt handler with this change.

Currently update_times() calculates ticks by "jiffies - wall_jiffies", but
callers of do_timer() should know how many ticks to update.  Passing ticks
get rid of this redundant calculation.  Also there are another redundancy
pointed out by Martin Schwidefsky.

This cleanup make a barrier added by
5aee405c662ca644980c184774277fc6d0769a84 needless.  So this patch removes
it.

As a bonus, this cleanup make wall_jiffies can be removed easily, since now
wall_jiffies is always synced with jiffies.  (This patch does not really
remove wall_jiffies.  It would be another cleanup patch)

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Acked-by: Russell King <rmk@arm.linux.org.uk>
Cc: Ian Molton <spyro@f2s.com>
Cc: Mikael Starvik <starvik@axis.com>
Acked-by: David Howells <dhowells@redhat.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Hirokazu Takata <takata.hirokazu@renesas.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Cc: Richard Curnow <rc@rc0.org.uk>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp>
Cc: Chris Zankel <chris@zankel.net>
Acked-by: "Luck, Tony" <tony.luck@intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/kernel/time.c                 |  2 +-
 arch/arm/kernel/time.c                   |  2 +-
 arch/arm26/kernel/time.c                 |  2 +-
 arch/avr32/kernel/time.c                 |  2 +-
 arch/cris/arch-v10/kernel/time.c         |  2 +-
 arch/cris/arch-v32/kernel/time.c         |  2 +-
 arch/frv/kernel/time.c                   |  2 +-
 arch/h8300/kernel/time.c                 |  2 +-
 arch/ia64/kernel/time.c                  |  2 +-
 arch/m32r/kernel/time.c                  |  2 +-
 arch/m68k/kernel/time.c                  |  2 +-
 arch/m68k/sun3/sun3ints.c                |  2 +-
 arch/m68knommu/kernel/time.c             |  2 +-
 arch/mips/au1000/common/time.c           |  6 +++---
 arch/mips/gt64120/common/time.c          |  2 +-
 arch/mips/kernel/time.c                  |  2 +-
 arch/mips/momentum/ocelot_g/gt-irq.c     |  2 +-
 arch/mips/sgi-ip27/ip27-timer.c          |  2 +-
 arch/parisc/kernel/time.c                |  2 +-
 arch/powerpc/kernel/time.c               |  2 +-
 arch/ppc/kernel/time.c                   |  2 +-
 arch/s390/kernel/time.c                  |  9 ++++-----
 arch/sh/kernel/time.c                    |  2 +-
 arch/sh64/kernel/time.c                  |  2 +-
 arch/sparc/kernel/pcic.c                 |  2 +-
 arch/sparc/kernel/time.c                 |  2 +-
 arch/sparc64/kernel/time.c               |  4 ++--
 arch/um/kernel/time.c                    |  2 +-
 arch/v850/kernel/time.c                  |  2 +-
 arch/x86_64/kernel/time.c                |  8 ++++----
 arch/xtensa/kernel/time.c                |  2 +-
 include/asm-arm/arch-clps711x/time.h     |  2 +-
 include/asm-arm/arch-l7200/time.h        |  2 +-
 include/asm-i386/mach-default/do_timer.h |  2 +-
 include/asm-i386/mach-visws/do_timer.h   |  2 +-
 include/asm-i386/mach-voyager/do_timer.h |  2 +-
 include/linux/sched.h                    |  2 +-
 kernel/timer.c                           | 19 ++++++-------------
 38 files changed, 52 insertions(+), 60 deletions(-)

(limited to 'include')

diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c
index b191cc759737..7c1e44420a78 100644
--- a/arch/alpha/kernel/time.c
+++ b/arch/alpha/kernel/time.c
@@ -132,7 +132,7 @@ irqreturn_t timer_interrupt(int irq, void *dev, struct pt_regs * regs)
 	nticks = delta >> FIX_SHIFT;
 
 	while (nticks > 0) {
-		do_timer(regs);
+		do_timer(1);
 #ifndef CONFIG_SMP
 		update_process_times(user_mode(regs));
 #endif
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index d4dceb5f06e9..f7d5165796ef 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -337,7 +337,7 @@ void timer_tick(struct pt_regs *regs)
 	profile_tick(CPU_PROFILING, regs);
 	do_leds();
 	do_set_rtc();
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/arm26/kernel/time.c b/arch/arm26/kernel/time.c
index db63d75d0715..80adbd005fc5 100644
--- a/arch/arm26/kernel/time.c
+++ b/arch/arm26/kernel/time.c
@@ -194,7 +194,7 @@ EXPORT_SYMBOL(do_settimeofday);
 
 static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-        do_timer(regs);
+        do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
index b0e6b5855a38..3e56b9f4358a 100644
--- a/arch/avr32/kernel/time.c
+++ b/arch/avr32/kernel/time.c
@@ -148,7 +148,7 @@ timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	 * Call the generic timer interrupt handler
 	 */
 	write_seqlock(&xtime_lock);
-	do_timer(regs);
+	do_timer(1);
 	write_sequnlock(&xtime_lock);
 
 	/*
diff --git a/arch/cris/arch-v10/kernel/time.c b/arch/cris/arch-v10/kernel/time.c
index 9c22b76e129a..ebacf1457d91 100644
--- a/arch/cris/arch-v10/kernel/time.c
+++ b/arch/cris/arch-v10/kernel/time.c
@@ -227,7 +227,7 @@ timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	
 	/* call the real timer interrupt handler */
 
-	do_timer(regs);
+	do_timer(1);
 	
         cris_do_profile(regs); /* Save profiling information */
 
diff --git a/arch/cris/arch-v32/kernel/time.c b/arch/cris/arch-v32/kernel/time.c
index 50f3f93293d6..be0a01657d4f 100644
--- a/arch/cris/arch-v32/kernel/time.c
+++ b/arch/cris/arch-v32/kernel/time.c
@@ -219,7 +219,7 @@ timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 		return IRQ_HANDLED;
 
 	/* call the real timer interrupt handler */
-	do_timer(regs);
+	do_timer(1);
 
 	/*
 	 * If we have an externally synchronized Linux clock, then update
diff --git a/arch/frv/kernel/time.c b/arch/frv/kernel/time.c
index 3d0284bccb94..7e55884135ed 100644
--- a/arch/frv/kernel/time.c
+++ b/arch/frv/kernel/time.c
@@ -70,7 +70,7 @@ static irqreturn_t timer_interrupt(int irq, void *dummy, struct pt_regs * regs)
 	 */
 	write_seqlock(&xtime_lock);
 
-	do_timer(regs);
+	do_timer(1);
 	update_process_times(user_mode(regs));
 	profile_tick(CPU_PROFILING, regs);
 
diff --git a/arch/h8300/kernel/time.c b/arch/h8300/kernel/time.c
index 688a5100604c..e569d17b4ae6 100644
--- a/arch/h8300/kernel/time.c
+++ b/arch/h8300/kernel/time.c
@@ -41,7 +41,7 @@ static void timer_interrupt(int irq, void *dummy, struct pt_regs * regs)
 	/* may need to kick the hardware timer */
 	platform_timer_eoi();
 
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 6928ef0d64d8..16262687a103 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -78,7 +78,7 @@ timer_interrupt (int irq, void *dev_id, struct pt_regs *regs)
 			 * xtime_lock.
 			 */
 			write_seqlock(&xtime_lock);
-			do_timer(regs);
+			do_timer(1);
 			local_cpu_data->itm_next = new_itm;
 			write_sequnlock(&xtime_lock);
 		} else
diff --git a/arch/m32r/kernel/time.c b/arch/m32r/kernel/time.c
index ded0be07a476..7a896893cd28 100644
--- a/arch/m32r/kernel/time.c
+++ b/arch/m32r/kernel/time.c
@@ -202,7 +202,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 #ifndef CONFIG_SMP
 	profile_tick(CPU_PROFILING, regs);
 #endif
-	do_timer(regs);
+	do_timer(1);
 
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
diff --git a/arch/m68k/kernel/time.c b/arch/m68k/kernel/time.c
index 98e4b1adfa29..1072e4946a4a 100644
--- a/arch/m68k/kernel/time.c
+++ b/arch/m68k/kernel/time.c
@@ -40,7 +40,7 @@ static inline int set_rtc_mmss(unsigned long nowtime)
  */
 static irqreturn_t timer_interrupt(int irq, void *dummy, struct pt_regs * regs)
 {
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/m68k/sun3/sun3ints.c b/arch/m68k/sun3/sun3ints.c
index f18b9d3ef16d..dc4ea7e074a6 100644
--- a/arch/m68k/sun3/sun3ints.c
+++ b/arch/m68k/sun3/sun3ints.c
@@ -65,7 +65,7 @@ static irqreturn_t sun3_int5(int irq, void *dev_id, struct pt_regs *fp)
 #ifdef CONFIG_SUN3
 	intersil_clear();
 #endif
-        do_timer(fp);
+        do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(fp));
 #endif
diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c
index 1db987272220..db1e1ce0a349 100644
--- a/arch/m68knommu/kernel/time.c
+++ b/arch/m68knommu/kernel/time.c
@@ -51,7 +51,7 @@ static irqreturn_t timer_interrupt(int irq, void *dummy, struct pt_regs * regs)
 
 	write_seqlock(&xtime_lock);
 
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/mips/au1000/common/time.c b/arch/mips/au1000/common/time.c
index 7fbea1bf7b48..0a067f3113a5 100644
--- a/arch/mips/au1000/common/time.c
+++ b/arch/mips/au1000/common/time.c
@@ -96,7 +96,7 @@ void mips_timer_interrupt(struct pt_regs *regs)
 		timerlo = count;
 
 		kstat_this_cpu.irqs[irq]++;
-		do_timer(regs);
+		do_timer(1);
 #ifndef CONFIG_SMP
 		update_process_times(user_mode(regs));
 #endif
@@ -137,7 +137,7 @@ irqreturn_t counter0_irq(int irq, void *dev_id, struct pt_regs *regs)
 	}
 
 	while (time_elapsed > 0) {
-		do_timer(regs);
+		do_timer(1);
 #ifndef CONFIG_SMP
 		update_process_times(user_mode(regs));
 #endif
@@ -156,7 +156,7 @@ irqreturn_t counter0_irq(int irq, void *dev_id, struct pt_regs *regs)
 
 	if (jiffie_drift >= 999) {
 		jiffie_drift -= 999;
-		do_timer(regs); /* increment jiffies by one */
+		do_timer(1); /* increment jiffies by one */
 #ifndef CONFIG_SMP
 		update_process_times(user_mode(regs));
 #endif
diff --git a/arch/mips/gt64120/common/time.c b/arch/mips/gt64120/common/time.c
index d837b26fbe51..7feca49350d1 100644
--- a/arch/mips/gt64120/common/time.c
+++ b/arch/mips/gt64120/common/time.c
@@ -34,7 +34,7 @@ static void gt64120_irq(int irq, void *dev_id, struct pt_regs *regs)
 	if (irq_src & 0x00000800) {	/* Check for timer interrupt */
 		handled = 1;
 		irq_src &= ~0x00000800;
-		do_timer(regs);
+		do_timer(1);
 #ifndef CONFIG_SMP
 		update_process_times(user_mode(regs));
 #endif
diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c
index 170cb67f4ede..6ab8d975a974 100644
--- a/arch/mips/kernel/time.c
+++ b/arch/mips/kernel/time.c
@@ -434,7 +434,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	/*
 	 * call the generic timer interrupt handling
 	 */
-	do_timer(regs);
+	do_timer(1);
 
 	/*
 	 * If we have an externally synchronized Linux clock, then update
diff --git a/arch/mips/momentum/ocelot_g/gt-irq.c b/arch/mips/momentum/ocelot_g/gt-irq.c
index 9fb2493fff02..6cd87cf0195a 100644
--- a/arch/mips/momentum/ocelot_g/gt-irq.c
+++ b/arch/mips/momentum/ocelot_g/gt-irq.c
@@ -133,7 +133,7 @@ static irqreturn_t gt64240_p0int_irq(int irq, void *dev, struct pt_regs *regs)
 		MV_WRITE(TIMER_COUNTER_0_3_INTERRUPT_CAUSE, 0x0);
 
 		/* handle the timer call */
-		do_timer(regs);
+		do_timer(1);
 #ifndef CONFIG_SMP
 		update_process_times(user_mode(regs));
 #endif
diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c
index b029ba79c27a..c62a3a9ef867 100644
--- a/arch/mips/sgi-ip27/ip27-timer.c
+++ b/arch/mips/sgi-ip27/ip27-timer.c
@@ -111,7 +111,7 @@ again:
 	kstat_this_cpu.irqs[irq]++;		/* kstat only for bootcpu? */
 
 	if (cpu == 0)
-		do_timer(regs);
+		do_timer(1);
 
 	update_process_times(user_mode(regs));
 
diff --git a/arch/parisc/kernel/time.c b/arch/parisc/kernel/time.c
index 5facc9bff4ef..700df10924dd 100644
--- a/arch/parisc/kernel/time.c
+++ b/arch/parisc/kernel/time.c
@@ -79,7 +79,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 #endif
 		if (cpu == 0) {
 			write_seqlock(&xtime_lock);
-			do_timer(regs);
+			do_timer(1);
 			write_sequnlock(&xtime_lock);
 		}
 	}
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 7a3c3f791ade..71f71da98e7d 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -693,7 +693,7 @@ void timer_interrupt(struct pt_regs * regs)
 		tb_next_jiffy = tb_last_jiffy + tb_ticks_per_jiffy;
 		if (per_cpu(last_jiffy, cpu) >= tb_next_jiffy) {
 			tb_last_jiffy = tb_next_jiffy;
-			do_timer(regs);
+			do_timer(1);
 			timer_recalc_offset(tb_last_jiffy);
 			timer_check_rtc();
 		}
diff --git a/arch/ppc/kernel/time.c b/arch/ppc/kernel/time.c
index 6ab8cc7226ab..1e1f31554767 100644
--- a/arch/ppc/kernel/time.c
+++ b/arch/ppc/kernel/time.c
@@ -153,7 +153,7 @@ void timer_interrupt(struct pt_regs * regs)
 		/* We are in an interrupt, no need to save/restore flags */
 		write_seqlock(&xtime_lock);
 		tb_last_stamp = jiffy_stamp;
-		do_timer(regs);
+		do_timer(1);
 
 		/*
 		 * update the rtc when needed, this should be performed on the
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 1981c6199fa2..abab42e9f5f8 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -166,7 +166,7 @@ EXPORT_SYMBOL(do_settimeofday);
 void account_ticks(struct pt_regs *regs)
 {
 	__u64 tmp;
-	__u32 ticks, xticks;
+	__u32 ticks;
 
 	/* Calculate how many ticks have passed. */
 	if (S390_lowcore.int_clock < S390_lowcore.jiffy_timer) {
@@ -204,6 +204,7 @@ void account_ticks(struct pt_regs *regs)
 	 */
 	write_seqlock(&xtime_lock);
 	if (S390_lowcore.jiffy_timer > xtime_cc) {
+		__u32 xticks;
 		tmp = S390_lowcore.jiffy_timer - xtime_cc;
 		if (tmp >= 2*CLK_TICKS_PER_JIFFY) {
 			xticks = __div(tmp, CLK_TICKS_PER_JIFFY);
@@ -212,13 +213,11 @@ void account_ticks(struct pt_regs *regs)
 			xticks = 1;
 			xtime_cc += CLK_TICKS_PER_JIFFY;
 		}
-		while (xticks--)
-			do_timer(regs);
+		do_timer(xticks);
 	}
 	write_sequnlock(&xtime_lock);
 #else
-	for (xticks = ticks; xticks > 0; xticks--)
-		do_timer(regs);
+	do_timer(ticks);
 #endif
 
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index 149d9713eddf..f664a196c4f5 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -117,7 +117,7 @@ static long last_rtc_update;
  */
 void handle_timer_tick(struct pt_regs *regs)
 {
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/sh64/kernel/time.c b/arch/sh64/kernel/time.c
index b8162e59030e..3b61e06f9d72 100644
--- a/arch/sh64/kernel/time.c
+++ b/arch/sh64/kernel/time.c
@@ -298,7 +298,7 @@ static inline void do_timer_interrupt(int irq, struct pt_regs *regs)
 	asm ("getcon cr62, %0" : "=r" (current_ctc));
 	ctc_last_interrupt = (unsigned long) current_ctc;
 
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c
index bfd31aac2df3..e19b1bad9bc5 100644
--- a/arch/sparc/kernel/pcic.c
+++ b/arch/sparc/kernel/pcic.c
@@ -712,7 +712,7 @@ static irqreturn_t pcic_timer_handler (int irq, void *h, struct pt_regs *regs)
 {
 	write_seqlock(&xtime_lock);	/* Dummy, to show that we remember */
 	pcic_clear_clock_irq();
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c
index 845081b01267..6f84fa1b58e5 100644
--- a/arch/sparc/kernel/time.c
+++ b/arch/sparc/kernel/time.c
@@ -128,7 +128,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs * regs)
 #endif
 	clear_clock_irq();
 
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c
index b0b4feeec098..ca1193482f07 100644
--- a/arch/sparc64/kernel/time.c
+++ b/arch/sparc64/kernel/time.c
@@ -465,7 +465,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs * regs)
 		profile_tick(CPU_PROFILING, regs);
 		update_process_times(user_mode(regs));
 #endif
-		do_timer(regs);
+		do_timer(1);
 
 		/* Guarantee that the following sequences execute
 		 * uninterrupted.
@@ -496,7 +496,7 @@ void timer_tick_interrupt(struct pt_regs *regs)
 {
 	write_seqlock(&xtime_lock);
 
-	do_timer(regs);
+	do_timer(1);
 
 	timer_check_rtc();
 
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 820affbf3e16..a92965f8f9cd 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -93,7 +93,7 @@ irqreturn_t um_timer(int irq, void *dev, struct pt_regs *regs)
 
 	write_seqlock_irqsave(&xtime_lock, flags);
 
-	do_timer(regs);
+	do_timer(1);
 
 	nsecs = get_time();
 	xtime.tv_sec = nsecs / NSEC_PER_SEC;
diff --git a/arch/v850/kernel/time.c b/arch/v850/kernel/time.c
index a0b46695f186..f4d1a4d3cdc2 100644
--- a/arch/v850/kernel/time.c
+++ b/arch/v850/kernel/time.c
@@ -51,7 +51,7 @@ static irqreturn_t timer_interrupt (int irq, void *dummy, struct pt_regs *regs)
 	if (mach_tick)
 	  mach_tick ();
 
-	do_timer (regs);
+	do_timer (1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 1c255ee76e7c..7ea3bf2a858c 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -415,16 +415,16 @@ void main_timer_handler(struct pt_regs *regs)
 				(((long) offset << US_SCALE) / vxtime.tsc_quot) - 1;
 	}
 
-	if (lost > 0) {
+	if (lost > 0)
 		handle_lost_ticks(lost, regs);
-		jiffies += lost;
-	}
+	else
+		lost = 0;
 
 /*
  * Do the timer stuff.
  */
 
-	do_timer(regs);
+	do_timer(lost + 1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/arch/xtensa/kernel/time.c b/arch/xtensa/kernel/time.c
index 412ab32de391..241db201f40e 100644
--- a/arch/xtensa/kernel/time.c
+++ b/arch/xtensa/kernel/time.c
@@ -175,7 +175,7 @@ again:
 
 		last_ccount_stamp = next;
 		next += CCOUNT_PER_JIFFY;
-		do_timer (regs); /* Linux handler in kernel/timer.c */
+		do_timer (1); /* Linux handler in kernel/timer.c */
 
 		if (ntp_synced() &&
 		    xtime.tv_sec - last_rtc_update >= 659 &&
diff --git a/include/asm-arm/arch-clps711x/time.h b/include/asm-arm/arch-clps711x/time.h
index 9cb27cd4e6ae..0e4a3901d3b3 100644
--- a/include/asm-arm/arch-clps711x/time.h
+++ b/include/asm-arm/arch-clps711x/time.h
@@ -29,7 +29,7 @@ static irqreturn_t
 p720t_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
 	do_leds();
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/include/asm-arm/arch-l7200/time.h b/include/asm-arm/arch-l7200/time.h
index 7b98b533e63a..c69cb508735f 100644
--- a/include/asm-arm/arch-l7200/time.h
+++ b/include/asm-arm/arch-l7200/time.h
@@ -45,7 +45,7 @@
 static irqreturn_t
 timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
diff --git a/include/asm-i386/mach-default/do_timer.h b/include/asm-i386/mach-default/do_timer.h
index 6312c3e79814..4182c347ef85 100644
--- a/include/asm-i386/mach-default/do_timer.h
+++ b/include/asm-i386/mach-default/do_timer.h
@@ -16,7 +16,7 @@
 
 static inline void do_timer_interrupt_hook(struct pt_regs *regs)
 {
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode_vm(regs));
 #endif
diff --git a/include/asm-i386/mach-visws/do_timer.h b/include/asm-i386/mach-visws/do_timer.h
index 95568e6ca91c..8db618c5a72b 100644
--- a/include/asm-i386/mach-visws/do_timer.h
+++ b/include/asm-i386/mach-visws/do_timer.h
@@ -9,7 +9,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
 	/* Clear the interrupt */
 	co_cpu_write(CO_CPU_STAT,co_cpu_read(CO_CPU_STAT) & ~CO_STAT_TIMEINTR);
 
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode_vm(regs));
 #endif
diff --git a/include/asm-i386/mach-voyager/do_timer.h b/include/asm-i386/mach-voyager/do_timer.h
index eaf518098981..099fe9f5c1b2 100644
--- a/include/asm-i386/mach-voyager/do_timer.h
+++ b/include/asm-i386/mach-voyager/do_timer.h
@@ -3,7 +3,7 @@
 
 static inline void do_timer_interrupt_hook(struct pt_regs *regs)
 {
-	do_timer(regs);
+	do_timer(1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode_vm(regs));
 #endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index ed2af8671589..503dea61ff99 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1206,7 +1206,7 @@ extern void switch_uid(struct user_struct *);
 
 #include <asm/current.h>
 
-extern void do_timer(struct pt_regs *);
+extern void do_timer(unsigned long ticks);
 
 extern int FASTCALL(wake_up_state(struct task_struct * tsk, unsigned int state));
 extern int FASTCALL(wake_up_process(struct task_struct * tsk));
diff --git a/kernel/timer.c b/kernel/timer.c
index a2cb1ecb1b28..4f55622b0d38 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1222,10 +1222,8 @@ static inline void calc_load(unsigned long ticks)
 	unsigned long active_tasks; /* fixed-point */
 	static int count = LOAD_FREQ;
 
-	count -= ticks;
-	if (count < 0) {
-		count += LOAD_FREQ;
-		active_tasks = count_active_tasks();
+	active_tasks = count_active_tasks();
+	for (count -= ticks; count < 0; count += LOAD_FREQ) {
 		CALC_LOAD(avenrun[0], EXP_1, active_tasks);
 		CALC_LOAD(avenrun[1], EXP_5, active_tasks);
 		CALC_LOAD(avenrun[2], EXP_15, active_tasks);
@@ -1270,11 +1268,8 @@ void run_local_timers(void)
  * Called by the timer interrupt. xtime_lock must already be taken
  * by the timer IRQ!
  */
-static inline void update_times(void)
+static inline void update_times(unsigned long ticks)
 {
-	unsigned long ticks;
-
-	ticks = jiffies - wall_jiffies;
 	wall_jiffies += ticks;
 	update_wall_time();
 	calc_load(ticks);
@@ -1286,12 +1281,10 @@ static inline void update_times(void)
  * jiffies is defined in the linker script...
  */
 
-void do_timer(struct pt_regs *regs)
+void do_timer(unsigned long ticks)
 {
-	jiffies_64++;
-	/* prevent loading jiffies before storing new jiffies_64 value. */
-	barrier();
-	update_times();
+	jiffies_64 += ticks;
+	update_times(ticks);
 }
 
 #ifdef __ARCH_WANT_SYS_ALARM
-- 
cgit v1.2.3


From 5785c95baede8459d70c4aa0f7becb6e8b5fde4b Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@infradead.org>
Date: Fri, 29 Sep 2006 02:00:43 -0700
Subject: [PATCH] tty: make termios_sem a mutex

[akpm@osdl.org: fix]
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/tty_io.c    | 23 ++++++++++++-----------
 drivers/char/tty_ioctl.c | 17 +++++++++--------
 include/linux/tty.h      |  2 +-
 3 files changed, 22 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index b4f37c65b95c..142427c6e8f3 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -618,9 +618,9 @@ EXPORT_SYMBOL_GPL(tty_prepare_flip_string_flags);
  
 static void tty_set_termios_ldisc(struct tty_struct *tty, int num)
 {
-	down(&tty->termios_sem);
+	mutex_lock(&tty->termios_mutex);
 	tty->termios->c_line = num;
-	up(&tty->termios_sem);
+	mutex_unlock(&tty->termios_mutex);
 }
 
 /*
@@ -1338,9 +1338,9 @@ static void do_tty_hangup(void *data)
 	 */
 	if (tty->driver->flags & TTY_DRIVER_RESET_TERMIOS)
 	{
-		down(&tty->termios_sem);
+		mutex_lock(&tty->termios_mutex);
 		*tty->termios = tty->driver->init_termios;
-		up(&tty->termios_sem);
+		mutex_unlock(&tty->termios_mutex);
 	}
 	
 	/* Defer ldisc switch */
@@ -2750,9 +2750,9 @@ static int tiocgwinsz(struct tty_struct *tty, struct winsize __user * arg)
 {
 	int err;
 
-	down(&tty->termios_sem);
+	mutex_lock(&tty->termios_mutex);
 	err = copy_to_user(arg, &tty->winsize, sizeof(*arg));
-	up(&tty->termios_sem);
+	mutex_unlock(&tty->termios_mutex);
 
 	return err ? -EFAULT: 0;
 }
@@ -2782,14 +2782,15 @@ static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
 	if (copy_from_user(&tmp_ws, arg, sizeof(*arg)))
 		return -EFAULT;
 
-	down(&tty->termios_sem);
+	mutex_lock(&tty->termios_mutex);
 	if (!memcmp(&tmp_ws, &tty->winsize, sizeof(*arg)))
 		goto done;
 
 #ifdef CONFIG_VT
 	if (tty->driver->type == TTY_DRIVER_TYPE_CONSOLE) {
-		if (vc_lock_resize(tty->driver_data, tmp_ws.ws_col, tmp_ws.ws_row)) {
-			up(&tty->termios_sem);
+		if (vc_lock_resize(tty->driver_data, tmp_ws.ws_col,
+					tmp_ws.ws_row)) {
+			mutex_unlock(&tty->termios_mutex);
  			return -ENXIO;
 		}
 	}
@@ -2801,7 +2802,7 @@ static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
 	tty->winsize = tmp_ws;
 	real_tty->winsize = tmp_ws;
 done:
-	up(&tty->termios_sem);
+	mutex_unlock(&tty->termios_mutex);
 	return 0;
 }
 
@@ -3576,7 +3577,7 @@ static void initialize_tty_struct(struct tty_struct *tty)
 	tty_buffer_init(tty);
 	INIT_WORK(&tty->buf.work, flush_to_ldisc, tty);
 	init_MUTEX(&tty->buf.pty_sem);
-	init_MUTEX(&tty->termios_sem);
+	mutex_init(&tty->termios_mutex);
 	init_waitqueue_head(&tty->write_wait);
 	init_waitqueue_head(&tty->read_wait);
 	INIT_WORK(&tty->hangup_work, do_tty_hangup, tty);
diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c
index 4ad47d321bd4..4d540619ac84 100644
--- a/drivers/char/tty_ioctl.c
+++ b/drivers/char/tty_ioctl.c
@@ -20,6 +20,7 @@
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/bitops.h>
+#include <linux/mutex.h>
 
 #include <asm/io.h>
 #include <asm/uaccess.h>
@@ -131,7 +132,7 @@ static void change_termios(struct tty_struct * tty, struct termios * new_termios
 
 	/* FIXME: we need to decide on some locking/ordering semantics
 	   for the set_termios notification eventually */
-	down(&tty->termios_sem);
+	mutex_lock(&tty->termios_mutex);
 
 	*tty->termios = *new_termios;
 	unset_locked_termios(tty->termios, &old_termios, tty->termios_locked);
@@ -176,7 +177,7 @@ static void change_termios(struct tty_struct * tty, struct termios * new_termios
 			(ld->set_termios)(tty, &old_termios);
 		tty_ldisc_deref(ld);
 	}
-	up(&tty->termios_sem);
+	mutex_unlock(&tty->termios_mutex);
 }
 
 /**
@@ -284,13 +285,13 @@ static int get_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb)
 {
 	struct sgttyb tmp;
 
-	down(&tty->termios_sem);
+	mutex_lock(&tty->termios_mutex);
 	tmp.sg_ispeed = 0;
 	tmp.sg_ospeed = 0;
 	tmp.sg_erase = tty->termios->c_cc[VERASE];
 	tmp.sg_kill = tty->termios->c_cc[VKILL];
 	tmp.sg_flags = get_sgflags(tty);
-	up(&tty->termios_sem);
+	mutex_unlock(&tty->termios_mutex);
 	
 	return copy_to_user(sgttyb, &tmp, sizeof(tmp)) ? -EFAULT : 0;
 }
@@ -345,12 +346,12 @@ static int set_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb)
 	if (copy_from_user(&tmp, sgttyb, sizeof(tmp)))
 		return -EFAULT;
 
-	down(&tty->termios_sem);		
+	mutex_lock(&tty->termios_mutex);
 	termios =  *tty->termios;
 	termios.c_cc[VERASE] = tmp.sg_erase;
 	termios.c_cc[VKILL] = tmp.sg_kill;
 	set_sgflags(&termios, tmp.sg_flags);
-	up(&tty->termios_sem);
+	mutex_unlock(&tty->termios_mutex);
 	change_termios(tty, &termios);
 	return 0;
 }
@@ -592,11 +593,11 @@ int n_tty_ioctl(struct tty_struct * tty, struct file * file,
 		case TIOCSSOFTCAR:
 			if (get_user(arg, (unsigned int __user *) arg))
 				return -EFAULT;
-			down(&tty->termios_sem);
+			mutex_lock(&tty->termios_mutex);
 			tty->termios->c_cflag =
 				((tty->termios->c_cflag & ~CLOCAL) |
 				 (arg ? CLOCAL : 0));
-			up(&tty->termios_sem);
+			mutex_unlock(&tty->termios_mutex);
 			return 0;
 		default:
 			return -ENOIOCTLCMD;
diff --git a/include/linux/tty.h b/include/linux/tty.h
index d1dec3d0c814..ea4c2605f8da 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -174,7 +174,7 @@ struct tty_struct {
 	struct tty_driver *driver;
 	int index;
 	struct tty_ldisc ldisc;
-	struct semaphore termios_sem;
+	struct mutex termios_mutex;
 	struct termios *termios, *termios_locked;
 	char name[64];
 	int pgrp;
-- 
cgit v1.2.3


From 416bc51292f977b43b010c6dd937522b90062390 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 29 Sep 2006 02:00:45 -0700
Subject: [PATCH] Use decimal for PTRACE_ATTACH and PTRACE_DETACH.

It is sure confusing that linux/ptrace.h has:
	#define PTRACE_SINGLESTEP	   9
	#define PTRACE_ATTACH		0x10
	#define PTRACE_DETACH		0x11
	#define PTRACE_SYSCALL		  24
All the low-numbered constants are in decimal, but the last two in hex.
It sure makes it likely that someone will look at this and think that
9, 10, 11 are used, and that 16 and 17 are not used.

How about we use the same notation for all the numbers [0,24] in the
same short list?

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/ptrace.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index 8b2749a259dc..eeb1976ef7bf 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -16,8 +16,8 @@
 #define PTRACE_KILL		   8
 #define PTRACE_SINGLESTEP	   9
 
-#define PTRACE_ATTACH		0x10
-#define PTRACE_DETACH		0x11
+#define PTRACE_ATTACH		  16
+#define PTRACE_DETACH		  17
 
 #define PTRACE_SYSCALL		  24
 
-- 
cgit v1.2.3


From 57a6f51c4281aa3119975473c70dce0480d322bd Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 29 Sep 2006 02:00:49 -0700
Subject: [PATCH] introduce is_rt_policy() helper

Imho, makes the code a bit easier to read.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h | 4 ++--
 kernel/sched.c        | 5 ++---
 2 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 503dea61ff99..fbc69cc3923d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -504,8 +504,8 @@ struct signal_struct {
 #define rt_prio(prio)		unlikely((prio) < MAX_RT_PRIO)
 #define rt_task(p)		rt_prio((p)->prio)
 #define batch_task(p)		(unlikely((p)->policy == SCHED_BATCH))
-#define has_rt_policy(p) \
-	unlikely((p)->policy != SCHED_NORMAL && (p)->policy != SCHED_BATCH)
+#define is_rt_policy(p)		((p) != SCHED_NORMAL && (p) != SCHED_BATCH)
+#define has_rt_policy(p)	unlikely(is_rt_policy((p)->policy))
 
 /*
  * Some day this will be a full-fledged user tracking system..
diff --git a/kernel/sched.c b/kernel/sched.c
index f9b3c6a414f1..c3c718aea618 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4109,8 +4109,7 @@ recheck:
 	    (p->mm && param->sched_priority > MAX_USER_RT_PRIO-1) ||
 	    (!p->mm && param->sched_priority > MAX_RT_PRIO-1))
 		return -EINVAL;
-	if ((policy == SCHED_NORMAL || policy == SCHED_BATCH)
-					!= (param->sched_priority == 0))
+	if (is_rt_policy(policy) != (param->sched_priority != 0))
 		return -EINVAL;
 
 	/*
@@ -4134,7 +4133,7 @@ recheck:
 				!rlim_rtprio)
 			return -EPERM;
 		/* can't increase priority */
-		if ((policy != SCHED_NORMAL && policy != SCHED_BATCH) &&
+		if (is_rt_policy(policy) &&
 		    param->sched_priority > p->rt_priority &&
 		    param->sched_priority > rlim_rtprio)
 			return -EPERM;
-- 
cgit v1.2.3


From 9f50b93f066f8dc339de9b0eb78a22a75e6c8f8f Mon Sep 17 00:00:00 2001
From: Josh Triplett <josht@us.ibm.com>
Date: Fri, 29 Sep 2006 02:00:59 -0700
Subject: [PATCH] Make spinlock/rwlock annotations more accurate by using
 parameters, not types

The lock annotations used on spinlocks and rwlocks currently use
__{acquires,releases}(spinlock_t) and __{acquires,releases}(rwlock_t),
respectively.  This loses the information of which lock actually got
acquired or released, and assumes a different type for the parameter of
__acquires and __releases than the rest of the kernel.  While the current
implementations of __acquires and __releases throw away their argument,
this will not always remain the case.  Change this to use the lock
parameter instead, to preserve this information and increase consistency in
usage of __acquires and __releases.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/spinlock_api_smp.h | 50 ++++++++++++++++++++--------------------
 1 file changed, 25 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h
index b2c4f8299464..8828b8155e9c 100644
--- a/include/linux/spinlock_api_smp.h
+++ b/include/linux/spinlock_api_smp.h
@@ -19,41 +19,41 @@ int in_lock_functions(unsigned long addr);
 
 #define assert_spin_locked(x)	BUG_ON(!spin_is_locked(x))
 
-void __lockfunc _spin_lock(spinlock_t *lock)		__acquires(spinlock_t);
+void __lockfunc _spin_lock(spinlock_t *lock)		__acquires(lock);
 void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass)
-							__acquires(spinlock_t);
-void __lockfunc _read_lock(rwlock_t *lock)		__acquires(rwlock_t);
-void __lockfunc _write_lock(rwlock_t *lock)		__acquires(rwlock_t);
-void __lockfunc _spin_lock_bh(spinlock_t *lock)		__acquires(spinlock_t);
-void __lockfunc _read_lock_bh(rwlock_t *lock)		__acquires(rwlock_t);
-void __lockfunc _write_lock_bh(rwlock_t *lock)		__acquires(rwlock_t);
-void __lockfunc _spin_lock_irq(spinlock_t *lock)	__acquires(spinlock_t);
-void __lockfunc _read_lock_irq(rwlock_t *lock)		__acquires(rwlock_t);
-void __lockfunc _write_lock_irq(rwlock_t *lock)		__acquires(rwlock_t);
+							__acquires(lock);
+void __lockfunc _read_lock(rwlock_t *lock)		__acquires(lock);
+void __lockfunc _write_lock(rwlock_t *lock)		__acquires(lock);
+void __lockfunc _spin_lock_bh(spinlock_t *lock)		__acquires(lock);
+void __lockfunc _read_lock_bh(rwlock_t *lock)		__acquires(lock);
+void __lockfunc _write_lock_bh(rwlock_t *lock)		__acquires(lock);
+void __lockfunc _spin_lock_irq(spinlock_t *lock)	__acquires(lock);
+void __lockfunc _read_lock_irq(rwlock_t *lock)		__acquires(lock);
+void __lockfunc _write_lock_irq(rwlock_t *lock)		__acquires(lock);
 unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock)
-							__acquires(spinlock_t);
+							__acquires(lock);
 unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock)
-							__acquires(rwlock_t);
+							__acquires(lock);
 unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock)
-							__acquires(rwlock_t);
+							__acquires(lock);
 int __lockfunc _spin_trylock(spinlock_t *lock);
 int __lockfunc _read_trylock(rwlock_t *lock);
 int __lockfunc _write_trylock(rwlock_t *lock);
 int __lockfunc _spin_trylock_bh(spinlock_t *lock);
-void __lockfunc _spin_unlock(spinlock_t *lock)		__releases(spinlock_t);
-void __lockfunc _read_unlock(rwlock_t *lock)		__releases(rwlock_t);
-void __lockfunc _write_unlock(rwlock_t *lock)		__releases(rwlock_t);
-void __lockfunc _spin_unlock_bh(spinlock_t *lock)	__releases(spinlock_t);
-void __lockfunc _read_unlock_bh(rwlock_t *lock)		__releases(rwlock_t);
-void __lockfunc _write_unlock_bh(rwlock_t *lock)	__releases(rwlock_t);
-void __lockfunc _spin_unlock_irq(spinlock_t *lock)	__releases(spinlock_t);
-void __lockfunc _read_unlock_irq(rwlock_t *lock)	__releases(rwlock_t);
-void __lockfunc _write_unlock_irq(rwlock_t *lock)	__releases(rwlock_t);
+void __lockfunc _spin_unlock(spinlock_t *lock)		__releases(lock);
+void __lockfunc _read_unlock(rwlock_t *lock)		__releases(lock);
+void __lockfunc _write_unlock(rwlock_t *lock)		__releases(lock);
+void __lockfunc _spin_unlock_bh(spinlock_t *lock)	__releases(lock);
+void __lockfunc _read_unlock_bh(rwlock_t *lock)		__releases(lock);
+void __lockfunc _write_unlock_bh(rwlock_t *lock)	__releases(lock);
+void __lockfunc _spin_unlock_irq(spinlock_t *lock)	__releases(lock);
+void __lockfunc _read_unlock_irq(rwlock_t *lock)	__releases(lock);
+void __lockfunc _write_unlock_irq(rwlock_t *lock)	__releases(lock);
 void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags)
-							__releases(spinlock_t);
+							__releases(lock);
 void __lockfunc _read_unlock_irqrestore(rwlock_t *lock, unsigned long flags)
-							__releases(rwlock_t);
+							__releases(lock);
 void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags)
-							__releases(rwlock_t);
+							__releases(lock);
 
 #endif /* __LINUX_SPINLOCK_API_SMP_H */
-- 
cgit v1.2.3


From 303912e2a32aa73785b4c4dee15466d944a38a46 Mon Sep 17 00:00:00 2001
From: Josh Triplett <josht@us.ibm.com>
Date: Fri, 29 Sep 2006 02:01:00 -0700
Subject: [PATCH] Replace _spin_trylock with spin_trylock in the IRQ variants
 to use __cond_lock

spin_trylock_irq and spin_trylock_irqsave use _spin_trylock, which does not
use the __cond_lock wrapper annotation and thus does not affect the lock
context; change them to use spin_trylock instead, which does use
__cond_lock.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/spinlock.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index 31473db92d3b..456e74f0e129 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -241,14 +241,14 @@ do {								\
 #define spin_trylock_irq(lock) \
 ({ \
 	local_irq_disable(); \
-	_spin_trylock(lock) ? \
+	spin_trylock(lock) ? \
 	1 : ({ local_irq_enable(); 0;  }); \
 })
 
 #define spin_trylock_irqsave(lock, flags) \
 ({ \
 	local_irq_save(flags); \
-	_spin_trylock(lock) ? \
+	spin_trylock(lock) ? \
 	1 : ({ local_irq_restore(flags); 0; }); \
 })
 
-- 
cgit v1.2.3


From dcc8e559ee5ae03fa6bdb8611d76d86d0083e793 Mon Sep 17 00:00:00 2001
From: Josh Triplett <josht@us.ibm.com>
Date: Fri, 29 Sep 2006 02:01:03 -0700
Subject: [PATCH] Pass a lock expression to __cond_lock, like __acquire and
 __release

Currently, __acquire and __release take a lock expression, but __cond_lock
takes only a condition, not the lock acquired if the expression evaluates
to true.  Change __cond_lock to accept a lock expression, and change all
the callers to pass in a lock expression.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/compiler.h |  4 ++--
 include/linux/spinlock.h | 10 +++++-----
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 060b96112ec6..0780de440220 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -14,7 +14,7 @@
 # define __releases(x)	__attribute__((context(1,0)))
 # define __acquire(x)	__context__(1)
 # define __release(x)	__context__(-1)
-# define __cond_lock(x)	((x) ? ({ __context__(1); 1; }) : 0)
+# define __cond_lock(x,c)	((c) ? ({ __acquire(x); 1; }) : 0)
 extern void __chk_user_ptr(void __user *);
 extern void __chk_io_ptr(void __iomem *);
 #else
@@ -31,7 +31,7 @@ extern void __chk_io_ptr(void __iomem *);
 # define __releases(x)
 # define __acquire(x) (void)0
 # define __release(x) (void)0
-# define __cond_lock(x) (x)
+# define __cond_lock(x,c) (c)
 #endif
 
 #ifdef __KERNEL__
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index 456e74f0e129..b800d2d68b32 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -167,9 +167,9 @@ do {								\
  * regardless of whether CONFIG_SMP or CONFIG_PREEMPT are set. The various
  * methods are defined as nops in the case they are not required.
  */
-#define spin_trylock(lock)		__cond_lock(_spin_trylock(lock))
-#define read_trylock(lock)		__cond_lock(_read_trylock(lock))
-#define write_trylock(lock)		__cond_lock(_write_trylock(lock))
+#define spin_trylock(lock)		__cond_lock(lock, _spin_trylock(lock))
+#define read_trylock(lock)		__cond_lock(lock, _read_trylock(lock))
+#define write_trylock(lock)		__cond_lock(lock, _write_trylock(lock))
 
 #define spin_lock(lock)			_spin_lock(lock)
 
@@ -236,7 +236,7 @@ do {								\
 					_write_unlock_irqrestore(lock, flags)
 #define write_unlock_bh(lock)		_write_unlock_bh(lock)
 
-#define spin_trylock_bh(lock)		__cond_lock(_spin_trylock_bh(lock))
+#define spin_trylock_bh(lock)	__cond_lock(lock, _spin_trylock_bh(lock))
 
 #define spin_trylock_irq(lock) \
 ({ \
@@ -264,7 +264,7 @@ do {								\
  */
 extern int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock);
 #define atomic_dec_and_lock(atomic, lock) \
-		__cond_lock(_atomic_dec_and_lock(atomic, lock))
+		__cond_lock(lock, _atomic_dec_and_lock(atomic, lock))
 
 /**
  * spin_can_lock - would spin_trylock() succeed?
-- 
cgit v1.2.3


From 368bdb3d616fa352971f45b423ae6344715e620b Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 29 Sep 2006 02:01:05 -0700
Subject: [PATCH] cramfs: make cramfs_uncompress_exit() return void

It always returns 0, so relying on it is useless.  The only caller isn't
checking return value.  In general, un-, de-, -free functions should return
void.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/cramfs/uncompress.c    | 3 +--
 include/linux/cramfs_fs.h | 2 +-
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/cramfs/uncompress.c b/fs/cramfs/uncompress.c
index 8def89f2c438..fc3ccb74626f 100644
--- a/fs/cramfs/uncompress.c
+++ b/fs/cramfs/uncompress.c
@@ -68,11 +68,10 @@ int cramfs_uncompress_init(void)
 	return 0;
 }
 
-int cramfs_uncompress_exit(void)
+void cramfs_uncompress_exit(void)
 {
 	if (!--initialized) {
 		zlib_inflateEnd(&stream);
 		vfree(stream.workspace);
 	}
-	return 0;
 }
diff --git a/include/linux/cramfs_fs.h b/include/linux/cramfs_fs.h
index a41f38428c37..1dba681e428d 100644
--- a/include/linux/cramfs_fs.h
+++ b/include/linux/cramfs_fs.h
@@ -87,6 +87,6 @@ struct cramfs_super {
 /* Uncompression interfaces to the underlying zlib */
 int cramfs_uncompress_block(void *dst, int dstlen, void *src, int srclen);
 int cramfs_uncompress_init(void);
-int cramfs_uncompress_exit(void);
+void cramfs_uncompress_exit(void);
 
 #endif
-- 
cgit v1.2.3


From e8106b941ceab68cc5ff713df7b1276484554584 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 29 Sep 2006 02:01:08 -0700
Subject: [PATCH] lockdep: core, add enable/disable_irq_irqsave/irqrestore()
 APIs

Introduce the disable_irq_nosync_lockdep_irqsave() and
enable_irq_lockdep_irqrestore() APIs.  These are needed for NE2000; basically
NE2000 calls disable_irq and enable_irq as locking against the IRQ handler,
but both in cases where interrupts are on and off.  This means that lockdep
needs to track the old state of the virtual irq flags on disable_irq, and
restore these at enable_irq time.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/net/8390.c        |  6 +++---
 include/linux/interrupt.h | 16 ++++++++++++++++
 2 files changed, 19 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/net/8390.c b/drivers/net/8390.c
index 5b6b05ed8f3c..9d34056147ad 100644
--- a/drivers/net/8390.c
+++ b/drivers/net/8390.c
@@ -299,7 +299,7 @@ static int ei_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 *	Slow phase with lock held.
 	 */
 
-	disable_irq_nosync_lockdep(dev->irq);
+	disable_irq_nosync_lockdep_irqsave(dev->irq, &flags);
 
 	spin_lock(&ei_local->page_lock);
 
@@ -338,7 +338,7 @@ static int ei_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		netif_stop_queue(dev);
 		outb_p(ENISR_ALL, e8390_base + EN0_IMR);
 		spin_unlock(&ei_local->page_lock);
-		enable_irq_lockdep(dev->irq);
+		enable_irq_lockdep_irqrestore(dev->irq, &flags);
 		ei_local->stat.tx_errors++;
 		return 1;
 	}
@@ -379,7 +379,7 @@ static int ei_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	outb_p(ENISR_ALL, e8390_base + EN0_IMR);
 
 	spin_unlock(&ei_local->page_lock);
-	enable_irq_lockdep(dev->irq);
+	enable_irq_lockdep_irqrestore(dev->irq, &flags);
 
 	dev_kfree_skb (skb);
 	ei_local->stat.tx_bytes += send_length;
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index d5afee95fd43..1f97e3d92639 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -123,6 +123,14 @@ static inline void disable_irq_nosync_lockdep(unsigned int irq)
 #endif
 }
 
+static inline void disable_irq_nosync_lockdep_irqsave(unsigned int irq, unsigned long *flags)
+{
+	disable_irq_nosync(irq);
+#ifdef CONFIG_LOCKDEP
+	local_irq_save(*flags);
+#endif
+}
+
 static inline void disable_irq_lockdep(unsigned int irq)
 {
 	disable_irq(irq);
@@ -139,6 +147,14 @@ static inline void enable_irq_lockdep(unsigned int irq)
 	enable_irq(irq);
 }
 
+static inline void enable_irq_lockdep_irqrestore(unsigned int irq, unsigned long *flags)
+{
+#ifdef CONFIG_LOCKDEP
+	local_irq_restore(*flags);
+#endif
+	enable_irq(irq);
+}
+
 /* IRQ wakeup (PM) control: */
 extern int set_irq_wake(unsigned int irq, unsigned int on);
 
-- 
cgit v1.2.3


From 55a101f8f71a3d3dbda7b5c77083ffe47552f831 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 29 Sep 2006 02:01:10 -0700
Subject: [PATCH] kill PF_DEAD flag

After the previous change (->flags & PF_DEAD) <=> (->state == EXIT_DEAD), we
don't need PF_DEAD any longer.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h |  1 -
 kernel/exit.c         |  6 ++----
 kernel/sched.c        | 16 ++++++++--------
 mm/oom_kill.c         |  4 ++--
 4 files changed, 12 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index fbc69cc3923d..9763de334f09 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1061,7 +1061,6 @@ static inline void put_task_struct(struct task_struct *t)
 					/* Not implemented yet, only for 486*/
 #define PF_STARTING	0x00000002	/* being created */
 #define PF_EXITING	0x00000004	/* getting shut down */
-#define PF_DEAD		0x00000008	/* Dead */
 #define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
 #define PF_SUPERPRIV	0x00000100	/* used super-user privileges */
 #define PF_DUMPCORE	0x00000200	/* dumped core */
diff --git a/kernel/exit.c b/kernel/exit.c
index 3d759c98fb11..9dd5f1336da2 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -953,10 +953,8 @@ fastcall NORET_TYPE void do_exit(long code)
 	if (tsk->splice_pipe)
 		__free_pipe_info(tsk->splice_pipe);
 
-	/* PF_DEAD causes final put_task_struct after we schedule. */
 	preempt_disable();
-	BUG_ON(tsk->flags & PF_DEAD);
-	tsk->flags |= PF_DEAD;
+	/* causes final put_task_struct in finish_task_switch(). */
 	tsk->state = EXIT_DEAD;
 
 	schedule();
@@ -972,7 +970,7 @@ NORET_TYPE void complete_and_exit(struct completion *comp, long code)
 {
 	if (comp)
 		complete(comp);
-	
+
 	do_exit(code);
 }
 
diff --git a/kernel/sched.c b/kernel/sched.c
index e1646b044b69..a9405d7cc6ab 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1755,27 +1755,27 @@ static inline void finish_task_switch(struct rq *rq, struct task_struct *prev)
 	__releases(rq->lock)
 {
 	struct mm_struct *mm = rq->prev_mm;
-	unsigned long prev_task_flags;
+	long prev_state;
 
 	rq->prev_mm = NULL;
 
 	/*
 	 * A task struct has one reference for the use as "current".
-	 * If a task dies, then it sets EXIT_ZOMBIE in tsk->exit_state and
-	 * calls schedule one last time. The schedule call will never return,
-	 * and the scheduled task must drop that reference.
-	 * The test for EXIT_ZOMBIE must occur while the runqueue locks are
+	 * If a task dies, then it sets EXIT_DEAD in tsk->state and calls
+	 * schedule one last time. The schedule call will never return, and
+	 * the scheduled task must drop that reference.
+	 * The test for EXIT_DEAD must occur while the runqueue locks are
 	 * still held, otherwise prev could be scheduled on another cpu, die
 	 * there before we look at prev->state, and then the reference would
 	 * be dropped twice.
 	 *		Manfred Spraul <manfred@colorfullife.com>
 	 */
-	prev_task_flags = prev->flags;
+	prev_state = prev->state;
 	finish_arch_switch(prev);
 	finish_lock_switch(rq, prev);
 	if (mm)
 		mmdrop(mm);
-	if (unlikely(prev_task_flags & PF_DEAD)) {
+	if (unlikely(prev_state == EXIT_DEAD)) {
 		/*
 		 * Remove function-return probe instances associated with this
 		 * task and put them back on the free list.
@@ -5153,7 +5153,7 @@ static void migrate_dead(unsigned int dead_cpu, struct task_struct *p)
 	BUG_ON(p->exit_state != EXIT_ZOMBIE && p->exit_state != EXIT_DEAD);
 
 	/* Cannot have done final schedule yet: would have vanished. */
-	BUG_ON(p->flags & PF_DEAD);
+	BUG_ON(p->state == EXIT_DEAD);
 
 	get_task_struct(p);
 
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index f3dd79c1c367..202f186a753a 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -226,8 +226,8 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 		releasing = test_tsk_thread_flag(p, TIF_MEMDIE) ||
 						p->flags & PF_EXITING;
 		if (releasing) {
-			/* PF_DEAD tasks have already released their mm */
-			if (p->flags & PF_DEAD)
+			/* TASK_DEAD tasks have already released their mm */
+			if (p->state == EXIT_DEAD)
 				continue;
 			if (p->flags & PF_EXITING && p == current) {
 				chosen = p;
-- 
cgit v1.2.3


From c394cc9fbb367f87faa2228ec2eabacd2d4701c6 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 29 Sep 2006 02:01:11 -0700
Subject: [PATCH] introduce TASK_DEAD state

I am not sure about this patch, I am asking Ingo to take a decision.

task_struct->state == EXIT_DEAD is a very special case, to avoid a confusion
it makes sense to introduce a new state, TASK_DEAD, while EXIT_DEAD should
live only in ->exit_state as documented in sched.h.

Note that this state is not visible to user-space, get_task_state() masks off
unsuitable states.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h | 1 +
 kernel/exit.c         | 2 +-
 kernel/sched.c        | 8 ++++----
 mm/oom_kill.c         | 2 +-
 4 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 9763de334f09..a06fc89cf6e5 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -148,6 +148,7 @@ extern unsigned long weighted_cpuload(const int cpu);
 #define EXIT_DEAD		32
 /* in tsk->state again */
 #define TASK_NONINTERACTIVE	64
+#define TASK_DEAD		128
 
 #define __set_task_state(tsk, state_value)		\
 	do { (tsk)->state = (state_value); } while (0)
diff --git a/kernel/exit.c b/kernel/exit.c
index 9dd5f1336da2..2e4c13cba95a 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -955,7 +955,7 @@ fastcall NORET_TYPE void do_exit(long code)
 
 	preempt_disable();
 	/* causes final put_task_struct in finish_task_switch(). */
-	tsk->state = EXIT_DEAD;
+	tsk->state = TASK_DEAD;
 
 	schedule();
 	BUG();
diff --git a/kernel/sched.c b/kernel/sched.c
index a9405d7cc6ab..74f169ac0773 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1761,10 +1761,10 @@ static inline void finish_task_switch(struct rq *rq, struct task_struct *prev)
 
 	/*
 	 * A task struct has one reference for the use as "current".
-	 * If a task dies, then it sets EXIT_DEAD in tsk->state and calls
+	 * If a task dies, then it sets TASK_DEAD in tsk->state and calls
 	 * schedule one last time. The schedule call will never return, and
 	 * the scheduled task must drop that reference.
-	 * The test for EXIT_DEAD must occur while the runqueue locks are
+	 * The test for TASK_DEAD must occur while the runqueue locks are
 	 * still held, otherwise prev could be scheduled on another cpu, die
 	 * there before we look at prev->state, and then the reference would
 	 * be dropped twice.
@@ -1775,7 +1775,7 @@ static inline void finish_task_switch(struct rq *rq, struct task_struct *prev)
 	finish_lock_switch(rq, prev);
 	if (mm)
 		mmdrop(mm);
-	if (unlikely(prev_state == EXIT_DEAD)) {
+	if (unlikely(prev_state == TASK_DEAD)) {
 		/*
 		 * Remove function-return probe instances associated with this
 		 * task and put them back on the free list.
@@ -5153,7 +5153,7 @@ static void migrate_dead(unsigned int dead_cpu, struct task_struct *p)
 	BUG_ON(p->exit_state != EXIT_ZOMBIE && p->exit_state != EXIT_DEAD);
 
 	/* Cannot have done final schedule yet: would have vanished. */
-	BUG_ON(p->state == EXIT_DEAD);
+	BUG_ON(p->state == TASK_DEAD);
 
 	get_task_struct(p);
 
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 202f186a753a..21f0a7e8e514 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -227,7 +227,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 						p->flags & PF_EXITING;
 		if (releasing) {
 			/* TASK_DEAD tasks have already released their mm */
-			if (p->state == EXIT_DEAD)
+			if (p->state == TASK_DEAD)
 				continue;
 			if (p->flags & PF_EXITING && p == current) {
 				chosen = p;
-- 
cgit v1.2.3


From 38837fc75acb7fa9b0e111b0241fe4fe76c5d4b3 Mon Sep 17 00:00:00 2001
From: Paul Jackson <pj@sgi.com>
Date: Fri, 29 Sep 2006 02:01:16 -0700
Subject: [PATCH] cpuset: top_cpuset tracks hotplug changes to node_online_map

Change the list of memory nodes allowed to tasks in the top (root) nodeset
to dynamically track what cpus are online, using a call to a cpuset hook
from the memory hotplug code.  Make this top cpus file read-only.

On systems that have cpusets configured in their kernel, but that aren't
actively using cpusets (for some distros, this covers the majority of
systems) all tasks end up in the top cpuset.

If that system does support memory hotplug, then these tasks cannot make
use of memory nodes that are added after system boot, because the memory
nodes are not allowed in the top cpuset.  This is a surprising regression
over earlier kernels that didn't have cpusets enabled.

One key motivation for this change is to remain consistent with the
behaviour for the top_cpuset's 'cpus', which is also read-only, and which
automatically tracks the cpu_online_map.

This change also has the minor benefit that it fixes a long standing,
little noticed, minor bug in cpusets.  The cpuset performance tweak to
short circuit the cpuset_zone_allowed() check on systems with just a single
cpuset (see 'number_of_cpusets', in linux/cpuset.h) meant that simply
changing the 'mems' of the top_cpuset had no affect, even though the change
(the write system call) appeared to succeed.  With the following change,
that write to the 'mems' file fails -EACCES, and the 'mems' file stubbornly
refuses to be changed via user space writes.  Thus no one should be mislead
into thinking they've changed the top_cpusets's 'mems' when in affect they
haven't.

In order to keep the behaviour of cpusets consistent between systems
actively making use of them and systems not using them, this patch changes
the behaviour of the 'mems' file in the top (root) cpuset, making it read
only, and making it automatically track the value of node_online_map.  Thus
tasks in the top cpuset will have automatic use of hot plugged memory nodes
allowed by their cpuset.

[akpm@osdl.org: build fix]
[bunk@stusta.de: build fix]
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/cpusets.txt | 10 +++++-----
 include/linux/cpuset.h    |  4 ++++
 kernel/cpuset.c           | 28 +++++++++++++++++++++++++---
 mm/memory_hotplug.c       |  3 +++
 4 files changed, 37 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/Documentation/cpusets.txt b/Documentation/cpusets.txt
index 76b44290c154..842f0d1ab216 100644
--- a/Documentation/cpusets.txt
+++ b/Documentation/cpusets.txt
@@ -217,11 +217,11 @@ exclusive cpuset.  Also, the use of a Linux virtual file system (vfs)
 to represent the cpuset hierarchy provides for a familiar permission
 and name space for cpusets, with a minimum of additional kernel code.
 
-The cpus file in the root (top_cpuset) cpuset is read-only.
-It automatically tracks the value of cpu_online_map, using a CPU
-hotplug notifier.  If and when memory nodes can be hotplugged,
-we expect to make the mems file in the root cpuset read-only
-as well, and have it track the value of node_online_map.
+The cpus and mems files in the root (top_cpuset) cpuset are
+read-only.  The cpus file automatically tracks the value of
+cpu_online_map using a CPU hotplug notifier, and the mems file
+automatically tracks the value of node_online_map using the
+cpuset_track_online_nodes() hook.
 
 
 1.4 What are exclusive cpusets ?
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 9354722a9217..4d8adf663681 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -63,6 +63,8 @@ static inline int cpuset_do_slab_mem_spread(void)
 	return current->flags & PF_SPREAD_SLAB;
 }
 
+extern void cpuset_track_online_nodes(void);
+
 #else /* !CONFIG_CPUSETS */
 
 static inline int cpuset_init_early(void) { return 0; }
@@ -126,6 +128,8 @@ static inline int cpuset_do_slab_mem_spread(void)
 	return 0;
 }
 
+static inline void cpuset_track_online_nodes(void) {}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 584bb4e6c042..794af5024c2f 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -912,6 +912,10 @@ static int update_nodemask(struct cpuset *cs, char *buf)
 	int fudge;
 	int retval;
 
+	/* top_cpuset.mems_allowed tracks node_online_map; it's read-only */
+	if (cs == &top_cpuset)
+		return -EACCES;
+
 	trialcs = *cs;
 	retval = nodelist_parse(buf, trialcs.mems_allowed);
 	if (retval < 0)
@@ -2042,9 +2046,8 @@ out:
  * (of no affect) on systems that are actively using CPU hotplug
  * but making no active use of cpusets.
  *
- * This handles CPU hotplug (cpuhp) events.  If someday Memory
- * Nodes can be hotplugged (dynamically changing node_online_map)
- * then we should handle that too, perhaps in a similar way.
+ * This routine ensures that top_cpuset.cpus_allowed tracks
+ * cpu_online_map on each CPU hotplug (cpuhp) event.
  */
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -2063,6 +2066,25 @@ static int cpuset_handle_cpuhp(struct notifier_block *nb,
 }
 #endif
 
+/*
+ * Keep top_cpuset.mems_allowed tracking node_online_map.
+ * Call this routine anytime after you change node_online_map.
+ * See also the previous routine cpuset_handle_cpuhp().
+ */
+
+#ifdef CONFIG_MEMORY_HOTPLUG
+void cpuset_track_online_nodes()
+{
+	mutex_lock(&manage_mutex);
+	mutex_lock(&callback_mutex);
+
+	top_cpuset.mems_allowed = node_online_map;
+
+	mutex_unlock(&callback_mutex);
+	mutex_unlock(&manage_mutex);
+}
+#endif
+
 /**
  * cpuset_init_smp - initialize cpus_allowed
  *
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index c37319542b70..9576ed920c0a 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -21,6 +21,7 @@
 #include <linux/highmem.h>
 #include <linux/vmalloc.h>
 #include <linux/ioport.h>
+#include <linux/cpuset.h>
 
 #include <asm/tlbflush.h>
 
@@ -283,6 +284,8 @@ int add_memory(int nid, u64 start, u64 size)
 	/* we online node here. we can't roll back from here. */
 	node_set_online(nid);
 
+	cpuset_track_online_nodes();
+
 	if (new_pgdat) {
 		ret = register_one_node(nid);
 		/*
-- 
cgit v1.2.3


From 2d1d43f6a43b703587e759145f69467e7c6553a7 Mon Sep 17 00:00:00 2001
From: Chandra Seetharaman <sekharan@us.ibm.com>
Date: Fri, 29 Sep 2006 02:01:25 -0700
Subject: [PATCH] call mm/page-writeback.c:set_ratelimit() when new pages are
 hot-added

ratelimit_pages in page-writeback.c is recalculated (in set_ratelimit())
every time a CPU is hot-added/removed.  But this value is not recalculated
when new pages are hot-added.

This patch fixes that problem by calling set_ratelimit() when new pages
are hot-added.

[akpm@osdl.org: cleanups]
Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/writeback.h | 1 +
 mm/memory_hotplug.c       | 2 ++
 mm/page-writeback.c       | 6 +++---
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 56a23a0e7f2e..9d4074ecd0cd 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -117,6 +117,7 @@ int sync_page_range(struct inode *inode, struct address_space *mapping,
 int sync_page_range_nolock(struct inode *inode, struct address_space *mapping,
 			   loff_t pos, loff_t count);
 void set_page_dirty_balance(struct page *page);
+void writeback_set_ratelimit(void);
 
 /* pdflush.c */
 extern int nr_pdflush_threads;	/* Global so it can be exported to sysctl
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 9576ed920c0a..2053bb165a21 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -13,6 +13,7 @@
 #include <linux/compiler.h>
 #include <linux/module.h>
 #include <linux/pagevec.h>
+#include <linux/writeback.h>
 #include <linux/slab.h>
 #include <linux/sysctl.h>
 #include <linux/cpu.h>
@@ -192,6 +193,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages)
 	if (need_zonelists_rebuild)
 		build_all_zonelists();
 	vm_total_pages = nr_free_pagecache_pages();
+	writeback_set_ratelimit();
 	return 0;
 }
 
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index efd2705e4986..488b7088557c 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -501,7 +501,7 @@ void laptop_sync_completion(void)
  * will write six megabyte chunks, max.
  */
 
-static void set_ratelimit(void)
+void writeback_set_ratelimit(void)
 {
 	ratelimit_pages = vm_total_pages / (num_online_cpus() * 32);
 	if (ratelimit_pages < 16)
@@ -513,7 +513,7 @@ static void set_ratelimit(void)
 static int __cpuinit
 ratelimit_handler(struct notifier_block *self, unsigned long u, void *v)
 {
-	set_ratelimit();
+	writeback_set_ratelimit();
 	return 0;
 }
 
@@ -546,7 +546,7 @@ void __init page_writeback_init(void)
 			vm_dirty_ratio = 1;
 	}
 	mod_timer(&wb_timer, jiffies + dirty_writeback_interval);
-	set_ratelimit();
+	writeback_set_ratelimit();
 	register_cpu_notifier(&ratelimit_nb);
 }
 
-- 
cgit v1.2.3


From 04b1db9fd7eea63c9663072feece616ea41b0a79 Mon Sep 17 00:00:00 2001
From: "Ian S. Nelson" <nelsonian@comcast.net>
Date: Fri, 29 Sep 2006 02:01:31 -0700
Subject: [PATCH] /sys/modules: allow full length section names

I've been using systemtap for some debugging and I noticed that it can't
probe a lot of modules.  Turns out it's kind of silly, the sections section
of /sys/module is limited to 32byte filenames and many of the actual
sections are a a bit longer than that.

[akpm@osdl.org: rewrite to use dymanic allocation]
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 CREDITS                |  3 ++-
 include/linux/module.h |  4 ++--
 kernel/module.c        | 26 ++++++++++++++++++++------
 3 files changed, 24 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/CREDITS b/CREDITS
index 8feb2bb49e35..66e82466dde8 100644
--- a/CREDITS
+++ b/CREDITS
@@ -2478,7 +2478,8 @@ S: Derbyshire DE4 3RL
 S: United Kingdom
 
 N: Ian S. Nelson
-E: ian.nelson@echostar.com
+E: nelsonis@earthlink.net
+P: 1024D/00D3D983 3EFD 7B86 B888 D7E2 29B6  9E97 576F 1B97 00D3 D983
 D: Minor mmap and ide hacks
 S: 1370 Atlantis Ave.
 S: Lafayette CO, 80026
diff --git a/include/linux/module.h b/include/linux/module.h
index d4486cc2e7fe..2c599175c583 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -232,17 +232,17 @@ enum module_state
 };
 
 /* Similar stuff for section attributes. */
-#define MODULE_SECT_NAME_LEN 32
 struct module_sect_attr
 {
 	struct module_attribute mattr;
-	char name[MODULE_SECT_NAME_LEN];
+	char *name;
 	unsigned long address;
 };
 
 struct module_sect_attrs
 {
 	struct attribute_group grp;
+	int nsections;
 	struct module_sect_attr attrs[0];
 };
 
diff --git a/kernel/module.c b/kernel/module.c
index b7fe6e840963..05625d5dc758 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -933,6 +933,15 @@ static ssize_t module_sect_show(struct module_attribute *mattr,
 	return sprintf(buf, "0x%lx\n", sattr->address);
 }
 
+static void free_sect_attrs(struct module_sect_attrs *sect_attrs)
+{
+	int section;
+
+	for (section = 0; section < sect_attrs->nsections; section++)
+		kfree(sect_attrs->attrs[section].name);
+	kfree(sect_attrs);
+}
+
 static void add_sect_attrs(struct module *mod, unsigned int nsect,
 		char *secstrings, Elf_Shdr *sechdrs)
 {
@@ -949,21 +958,26 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect,
 			+ nloaded * sizeof(sect_attrs->attrs[0]),
 			sizeof(sect_attrs->grp.attrs[0]));
 	size[1] = (nloaded + 1) * sizeof(sect_attrs->grp.attrs[0]);
-	if (! (sect_attrs = kmalloc(size[0] + size[1], GFP_KERNEL)))
+	sect_attrs = kzalloc(size[0] + size[1], GFP_KERNEL);
+	if (sect_attrs == NULL)
 		return;
 
 	/* Setup section attributes. */
 	sect_attrs->grp.name = "sections";
 	sect_attrs->grp.attrs = (void *)sect_attrs + size[0];
 
+	sect_attrs->nsections = 0;
 	sattr = &sect_attrs->attrs[0];
 	gattr = &sect_attrs->grp.attrs[0];
 	for (i = 0; i < nsect; i++) {
 		if (! (sechdrs[i].sh_flags & SHF_ALLOC))
 			continue;
 		sattr->address = sechdrs[i].sh_addr;
-		strlcpy(sattr->name, secstrings + sechdrs[i].sh_name,
-			MODULE_SECT_NAME_LEN);
+		sattr->name = kstrdup(secstrings + sechdrs[i].sh_name,
+					GFP_KERNEL);
+		if (sattr->name == NULL)
+			goto out;
+		sect_attrs->nsections++;
 		sattr->mattr.show = module_sect_show;
 		sattr->mattr.store = NULL;
 		sattr->mattr.attr.name = sattr->name;
@@ -979,7 +993,7 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect,
 	mod->sect_attrs = sect_attrs;
 	return;
   out:
-	kfree(sect_attrs);
+	free_sect_attrs(sect_attrs);
 }
 
 static void remove_sect_attrs(struct module *mod)
@@ -989,13 +1003,13 @@ static void remove_sect_attrs(struct module *mod)
 				   &mod->sect_attrs->grp);
 		/* We are positive that no one is using any sect attrs
 		 * at this point.  Deallocate immediately. */
-		kfree(mod->sect_attrs);
+		free_sect_attrs(mod->sect_attrs);
 		mod->sect_attrs = NULL;
 	}
 }
 
-
 #else
+
 static inline void add_sect_attrs(struct module *mod, unsigned int nsect,
 		char *sectstrings, Elf_Shdr *sechdrs)
 {
-- 
cgit v1.2.3


From f0c8bd164e1a0585d7e46896553136b4f488bd19 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Fri, 29 Sep 2006 02:01:34 -0700
Subject: [PATCH] Generic infrastructure for acls

The patches solve the following problem: We want to grant access to devices
based on who is logged in from where, etc.  This includes switching back and
forth between multiple user sessions, etc.

Using ACLs to define device access for logged-in users gives us all the
flexibility we need in order to fully solve the problem.

Device special files nowadays usually live on tmpfs, hence tmpfs ACLs.

Different distros have come up with solutions that solve the problem to
different degrees: SUSE uses a resource manager which tracks login sessions
and sets ACLs on device inodes as appropriate.  RedHat uses pam_console, which
changes the primary file ownership to the logged-in user.  Others use a set of
groups that users must be in in order to be granted the appropriate accesses.

The freedesktop.org project plans to implement a combination of a
console-tracker and a HAL-device-list based solution to grant access to
devices to users, and more distros will likely follow this approach.

These patches have first been posted here on 2 February 2005, and again
on 8 January 2006. We have been shipping them in SLES9 and SLES10 with
no problems reported.  The previous submission is archived here:

   http://lkml.org/lkml/2006/1/8/229
   http://lkml.org/lkml/2006/1/8/230
   http://lkml.org/lkml/2006/1/8/231

This patch:

Add some infrastructure for access control lists on in-memory
filesystems such as tmpfs.

Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/Kconfig                  |   4 +
 fs/Makefile                 |   1 +
 fs/generic_acl.c            | 197 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/generic_acl.h |  36 ++++++++
 4 files changed, 238 insertions(+)
 create mode 100644 fs/generic_acl.c
 create mode 100644 include/linux/generic_acl.h

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index d311198bba43..deb9eec9f6ee 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -1940,6 +1940,10 @@ config 9P_FS
 
 	  If unsure, say N.
 
+config GENERIC_ACL
+	bool
+	select FS_POSIX_ACL
+
 endmenu
 
 menu "Partition Types"
diff --git a/fs/Makefile b/fs/Makefile
index 89135428a539..46b8cfe497b2 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -35,6 +35,7 @@ obj-$(CONFIG_BINFMT_FLAT)	+= binfmt_flat.o
 obj-$(CONFIG_FS_MBCACHE)	+= mbcache.o
 obj-$(CONFIG_FS_POSIX_ACL)	+= posix_acl.o xattr_acl.o
 obj-$(CONFIG_NFS_COMMON)	+= nfs_common/
+obj-$(CONFIG_GENERIC_ACL)	+= generic_acl.o
 
 obj-$(CONFIG_QUOTA)		+= dquot.o
 obj-$(CONFIG_QFMT_V1)		+= quota_v1.o
diff --git a/fs/generic_acl.c b/fs/generic_acl.c
new file mode 100644
index 000000000000..9ccb78947171
--- /dev/null
+++ b/fs/generic_acl.c
@@ -0,0 +1,197 @@
+/*
+ * fs/generic_acl.c
+ *
+ * (C) 2005 Andreas Gruenbacher <agruen@suse.de>
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/sched.h>
+#include <linux/fs.h>
+#include <linux/generic_acl.h>
+
+/**
+ * generic_acl_list  -  Generic xattr_handler->list() operation
+ * @ops:	Filesystem specific getacl and setacl callbacks
+ */
+size_t
+generic_acl_list(struct inode *inode, struct generic_acl_operations *ops,
+		 int type, char *list, size_t list_size)
+{
+	struct posix_acl *acl;
+	const char *name;
+	size_t size;
+
+	acl = ops->getacl(inode, type);
+	if (!acl)
+		return 0;
+	posix_acl_release(acl);
+
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			name = POSIX_ACL_XATTR_ACCESS;
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			name = POSIX_ACL_XATTR_DEFAULT;
+			break;
+
+		default:
+			return 0;
+	}
+	size = strlen(name) + 1;
+	if (list && size <= list_size)
+		memcpy(list, name, size);
+	return size;
+}
+
+/**
+ * generic_acl_get  -  Generic xattr_handler->get() operation
+ * @ops:	Filesystem specific getacl and setacl callbacks
+ */
+int
+generic_acl_get(struct inode *inode, struct generic_acl_operations *ops,
+		int type, void *buffer, size_t size)
+{
+	struct posix_acl *acl;
+	int error;
+
+	acl = ops->getacl(inode, type);
+	if (!acl)
+		return -ENODATA;
+	error = posix_acl_to_xattr(acl, buffer, size);
+	posix_acl_release(acl);
+
+	return error;
+}
+
+/**
+ * generic_acl_set  -  Generic xattr_handler->set() operation
+ * @ops:	Filesystem specific getacl and setacl callbacks
+ */
+int
+generic_acl_set(struct inode *inode, struct generic_acl_operations *ops,
+		int type, const void *value, size_t size)
+{
+	struct posix_acl *acl = NULL;
+	int error;
+
+	if (S_ISLNK(inode->i_mode))
+		return -EOPNOTSUPP;
+	if (current->fsuid != inode->i_uid && !capable(CAP_FOWNER))
+		return -EPERM;
+	if (value) {
+		acl = posix_acl_from_xattr(value, size);
+		if (IS_ERR(acl))
+			return PTR_ERR(acl);
+	}
+	if (acl) {
+		mode_t mode;
+
+		error = posix_acl_valid(acl);
+		if (error)
+			goto failed;
+		switch(type) {
+			case ACL_TYPE_ACCESS:
+				mode = inode->i_mode;
+				error = posix_acl_equiv_mode(acl, &mode);
+				if (error < 0)
+					goto failed;
+				inode->i_mode = mode;
+				if (error == 0) {
+					posix_acl_release(acl);
+					acl = NULL;
+				}
+				break;
+
+			case ACL_TYPE_DEFAULT:
+				if (!S_ISDIR(inode->i_mode)) {
+					error = -EINVAL;
+					goto failed;
+				}
+				break;
+		}
+	}
+	ops->setacl(inode, type, acl);
+	error = 0;
+failed:
+	posix_acl_release(acl);
+	return error;
+}
+
+/**
+ * generic_acl_init  -  Take care of acl inheritance at @inode create time
+ * @ops:	Filesystem specific getacl and setacl callbacks
+ *
+ * Files created inside a directory with a default ACL inherit the
+ * directory's default ACL.
+ */
+int
+generic_acl_init(struct inode *inode, struct inode *dir,
+		 struct generic_acl_operations *ops)
+{
+	struct posix_acl *acl = NULL;
+	mode_t mode = inode->i_mode;
+	int error;
+
+	inode->i_mode = mode & ~current->fs->umask;
+	if (!S_ISLNK(inode->i_mode))
+		acl = ops->getacl(dir, ACL_TYPE_DEFAULT);
+	if (acl) {
+		struct posix_acl *clone;
+
+		if (S_ISDIR(inode->i_mode)) {
+			clone = posix_acl_clone(acl, GFP_KERNEL);
+			error = -ENOMEM;
+			if (!clone)
+				goto cleanup;
+			ops->setacl(inode, ACL_TYPE_DEFAULT, clone);
+			posix_acl_release(clone);
+		}
+		clone = posix_acl_clone(acl, GFP_KERNEL);
+		error = -ENOMEM;
+		if (!clone)
+			goto cleanup;
+		error = posix_acl_create_masq(clone, &mode);
+		if (error >= 0) {
+			inode->i_mode = mode;
+			if (error > 0)
+				ops->setacl(inode, ACL_TYPE_ACCESS, clone);
+		}
+		posix_acl_release(clone);
+	}
+	error = 0;
+
+cleanup:
+	posix_acl_release(acl);
+	return error;
+}
+
+/**
+ * generic_acl_chmod  -  change the access acl of @inode upon chmod()
+ * @ops:	FIlesystem specific getacl and setacl callbacks
+ *
+ * A chmod also changes the permissions of the owner, group/mask, and
+ * other ACL entries.
+ */
+int
+generic_acl_chmod(struct inode *inode, struct generic_acl_operations *ops)
+{
+	struct posix_acl *acl, *clone;
+	int error = 0;
+
+	if (S_ISLNK(inode->i_mode))
+		return -EOPNOTSUPP;
+	acl = ops->getacl(inode, ACL_TYPE_ACCESS);
+	if (acl) {
+		clone = posix_acl_clone(acl, GFP_KERNEL);
+		posix_acl_release(acl);
+		if (!clone)
+			return -ENOMEM;
+		error = posix_acl_chmod_masq(clone, inode->i_mode);
+		if (!error)
+			ops->setacl(inode, ACL_TYPE_ACCESS, clone);
+		posix_acl_release(clone);
+	}
+	return error;
+}
diff --git a/include/linux/generic_acl.h b/include/linux/generic_acl.h
new file mode 100644
index 000000000000..80764f40be75
--- /dev/null
+++ b/include/linux/generic_acl.h
@@ -0,0 +1,36 @@
+/*
+ * fs/generic_acl.c
+ *
+ * (C) 2005 Andreas Gruenbacher <agruen@suse.de>
+ *
+ * This file is released under the GPL.
+ */
+
+#ifndef GENERIC_ACL_H
+#define GENERIC_ACL_H
+
+#include <linux/posix_acl.h>
+#include <linux/posix_acl_xattr.h>
+
+/**
+ * struct generic_acl_operations  -  filesystem operations
+ *
+ * Filesystems must make these operations available to the generic
+ * operations.
+ */
+struct generic_acl_operations {
+	struct posix_acl *(*getacl)(struct inode *, int);
+	void (*setacl)(struct inode *, int, struct posix_acl *);
+};
+
+size_t generic_acl_list(struct inode *, struct generic_acl_operations *, int,
+			char *, size_t);
+int generic_acl_get(struct inode *, struct generic_acl_operations *, int,
+		    void *, size_t);
+int generic_acl_set(struct inode *, struct generic_acl_operations *, int,
+		    const void *, size_t);
+int generic_acl_init(struct inode *, struct inode *,
+		     struct generic_acl_operations *);
+int generic_acl_chmod(struct inode *, struct generic_acl_operations *);
+
+#endif
-- 
cgit v1.2.3


From 39f0247d3823e4e0bf8f6838a10362864b1e1053 Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Fri, 29 Sep 2006 02:01:35 -0700
Subject: [PATCH] Access Control Lists for tmpfs

Add access control lists for tmpfs.

Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/Kconfig               |  13 ++++
 include/linux/shmem_fs.h |  24 ++++++
 mm/Makefile              |   1 +
 mm/shmem.c               |  99 +++++++++++++++++++++++-
 mm/shmem_acl.c           | 197 +++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 332 insertions(+), 2 deletions(-)
 create mode 100644 mm/shmem_acl.c

(limited to 'include')

diff --git a/fs/Kconfig b/fs/Kconfig
index deb9eec9f6ee..4fd9efac29ab 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -881,6 +881,19 @@ config TMPFS
 
 	  See <file:Documentation/filesystems/tmpfs.txt> for details.
 
+config TMPFS_POSIX_ACL
+	bool "Tmpfs POSIX Access Control Lists"
+	depends on TMPFS
+	select GENERIC_ACL
+	help
+	  POSIX Access Control Lists (ACLs) support permissions for users and
+	  groups beyond the owner/group/world scheme.
+
+	  To learn more about Access Control Lists, visit the POSIX ACLs for
+	  Linux website <http://acl.bestbits.at/>.
+
+	  If you don't know what Access Control Lists are, say N.
+
 config HUGETLBFS
 	bool "HugeTLB file system support"
 	depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || BROKEN
diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h
index c057f0b32318..f3c51899117f 100644
--- a/include/linux/shmem_fs.h
+++ b/include/linux/shmem_fs.h
@@ -19,6 +19,10 @@ struct shmem_inode_info {
 	swp_entry_t		i_direct[SHMEM_NR_DIRECT]; /* first blocks */
 	struct list_head	swaplist;	/* chain of maybes on swap */
 	struct inode		vfs_inode;
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	struct posix_acl	*i_acl;
+	struct posix_acl	*i_default_acl;
+#endif
 };
 
 struct shmem_sb_info {
@@ -36,4 +40,24 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode)
 	return container_of(inode, struct shmem_inode_info, vfs_inode);
 }
 
+#ifdef CONFIG_TMPFS_POSIX_ACL
+int shmem_permission(struct inode *, int, struct nameidata *);
+int shmem_acl_init(struct inode *, struct inode *);
+void shmem_acl_destroy_inode(struct inode *);
+
+extern struct xattr_handler shmem_xattr_acl_access_handler;
+extern struct xattr_handler shmem_xattr_acl_default_handler;
+
+extern struct generic_acl_operations shmem_acl_ops;
+
+#else
+static inline int shmem_acl_init(struct inode *inode, struct inode *dir)
+{
+	return 0;
+}
+static inline void shmem_acl_destroy_inode(struct inode *inode)
+{
+}
+#endif  /* CONFIG_TMPFS_POSIX_ACL */
+
 #endif
diff --git a/mm/Makefile b/mm/Makefile
index 60c56c0b5e10..6200c6d6afd2 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -17,6 +17,7 @@ obj-$(CONFIG_HUGETLBFS)	+= hugetlb.o
 obj-$(CONFIG_NUMA) 	+= mempolicy.o
 obj-$(CONFIG_SPARSEMEM)	+= sparse.o
 obj-$(CONFIG_SHMEM) += shmem.o
+obj-$(CONFIG_TMPFS_POSIX_ACL) += shmem_acl.o
 obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o
 obj-$(CONFIG_SLOB) += slob.o
 obj-$(CONFIG_SLAB) += slab.o
diff --git a/mm/shmem.c b/mm/shmem.c
index eda907c3a86a..b96de69f236b 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -26,6 +26,8 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/fs.h>
+#include <linux/xattr.h>
+#include <linux/generic_acl.h>
 #include <linux/mm.h>
 #include <linux/mman.h>
 #include <linux/file.h>
@@ -177,6 +179,7 @@ static const struct address_space_operations shmem_aops;
 static struct file_operations shmem_file_operations;
 static struct inode_operations shmem_inode_operations;
 static struct inode_operations shmem_dir_inode_operations;
+static struct inode_operations shmem_special_inode_operations;
 static struct vm_operations_struct shmem_vm_ops;
 
 static struct backing_dev_info shmem_backing_dev_info  __read_mostly = {
@@ -637,7 +640,7 @@ static int shmem_notify_change(struct dentry *dentry, struct iattr *attr)
 	struct page *page = NULL;
 	int error;
 
-	if (attr->ia_valid & ATTR_SIZE) {
+	if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) {
 		if (attr->ia_size < inode->i_size) {
 			/*
 			 * If truncating down to a partial page, then
@@ -670,6 +673,10 @@ static int shmem_notify_change(struct dentry *dentry, struct iattr *attr)
 	error = inode_change_ok(inode, attr);
 	if (!error)
 		error = inode_setattr(inode, attr);
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	if (!error && (attr->ia_valid & ATTR_MODE))
+		error = generic_acl_chmod(inode, &shmem_acl_ops);
+#endif
 	if (page)
 		page_cache_release(page);
 	return error;
@@ -1362,6 +1369,7 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev)
 
 		switch (mode & S_IFMT) {
 		default:
+			inode->i_op = &shmem_special_inode_operations;
 			init_special_inode(inode, mode, dev);
 			break;
 		case S_IFREG:
@@ -1682,7 +1690,11 @@ shmem_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
 				iput(inode);
 				return error;
 			}
-			error = 0;
+		}
+		error = shmem_acl_init(inode, dir);
+		if (error) {
+			iput(inode);
+			return error;
 		}
 		if (dir->i_mode & S_ISGID) {
 			inode->i_gid = dir->i_gid;
@@ -1897,6 +1909,53 @@ static struct inode_operations shmem_symlink_inode_operations = {
 	.put_link	= shmem_put_link,
 };
 
+#ifdef CONFIG_TMPFS_POSIX_ACL
+/**
+ * Superblocks without xattr inode operations will get security.* xattr
+ * support from the VFS "for free". As soon as we have any other xattrs
+ * like ACLs, we also need to implement the security.* handlers at
+ * filesystem level, though.
+ */
+
+static size_t shmem_xattr_security_list(struct inode *inode, char *list,
+					size_t list_len, const char *name,
+					size_t name_len)
+{
+	return security_inode_listsecurity(inode, list, list_len);
+}
+
+static int shmem_xattr_security_get(struct inode *inode, const char *name,
+				    void *buffer, size_t size)
+{
+	if (strcmp(name, "") == 0)
+		return -EINVAL;
+	return security_inode_getsecurity(inode, name, buffer, size,
+					  -EOPNOTSUPP);
+}
+
+static int shmem_xattr_security_set(struct inode *inode, const char *name,
+				    const void *value, size_t size, int flags)
+{
+	if (strcmp(name, "") == 0)
+		return -EINVAL;
+	return security_inode_setsecurity(inode, name, value, size, flags);
+}
+
+struct xattr_handler shmem_xattr_security_handler = {
+	.prefix = XATTR_SECURITY_PREFIX,
+	.list   = shmem_xattr_security_list,
+	.get    = shmem_xattr_security_get,
+	.set    = shmem_xattr_security_set,
+};
+
+static struct xattr_handler *shmem_xattr_handlers[] = {
+	&shmem_xattr_acl_access_handler,
+	&shmem_xattr_acl_default_handler,
+	&shmem_xattr_security_handler,
+	NULL
+};
+#endif
+
 static int shmem_parse_options(char *options, int *mode, uid_t *uid,
 	gid_t *gid, unsigned long *blocks, unsigned long *inodes,
 	int *policy, nodemask_t *policy_nodes)
@@ -2094,6 +2153,10 @@ static int shmem_fill_super(struct super_block *sb,
 	sb->s_magic = TMPFS_MAGIC;
 	sb->s_op = &shmem_ops;
 	sb->s_time_gran = 1;
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	sb->s_xattr = shmem_xattr_handlers;
+	sb->s_flags |= MS_POSIXACL;
+#endif
 
 	inode = shmem_get_inode(sb, S_IFDIR | mode, 0);
 	if (!inode)
@@ -2130,6 +2193,7 @@ static void shmem_destroy_inode(struct inode *inode)
 		/* only struct inode is valid if it's an inline symlink */
 		mpol_free_shared_policy(&SHMEM_I(inode)->policy);
 	}
+	shmem_acl_destroy_inode(inode);
 	kmem_cache_free(shmem_inode_cachep, SHMEM_I(inode));
 }
 
@@ -2141,6 +2205,10 @@ static void init_once(void *foo, struct kmem_cache *cachep,
 	if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
 	    SLAB_CTOR_CONSTRUCTOR) {
 		inode_init_once(&p->vfs_inode);
+#ifdef CONFIG_TMPFS_POSIX_ACL
+		p->i_acl = NULL;
+		p->i_default_acl = NULL;
+#endif
 	}
 }
 
@@ -2184,6 +2252,14 @@ static struct inode_operations shmem_inode_operations = {
 	.truncate	= shmem_truncate,
 	.setattr	= shmem_notify_change,
 	.truncate_range	= shmem_truncate_range,
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= generic_listxattr,
+	.removexattr	= generic_removexattr,
+	.permission	= shmem_permission,
+#endif
+
 };
 
 static struct inode_operations shmem_dir_inode_operations = {
@@ -2198,6 +2274,25 @@ static struct inode_operations shmem_dir_inode_operations = {
 	.mknod		= shmem_mknod,
 	.rename		= shmem_rename,
 #endif
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	.setattr	= shmem_notify_change,
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= generic_listxattr,
+	.removexattr	= generic_removexattr,
+	.permission	= shmem_permission,
+#endif
+};
+
+static struct inode_operations shmem_special_inode_operations = {
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	.setattr	= shmem_notify_change,
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= generic_listxattr,
+	.removexattr	= generic_removexattr,
+	.permission	= shmem_permission,
+#endif
 };
 
 static struct super_operations shmem_ops = {
diff --git a/mm/shmem_acl.c b/mm/shmem_acl.c
new file mode 100644
index 000000000000..c946bf468718
--- /dev/null
+++ b/mm/shmem_acl.c
@@ -0,0 +1,197 @@
+/*
+ * mm/shmem_acl.c
+ *
+ * (C) 2005 Andreas Gruenbacher <agruen@suse.de>
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/fs.h>
+#include <linux/shmem_fs.h>
+#include <linux/xattr.h>
+#include <linux/generic_acl.h>
+
+/**
+ * shmem_get_acl  -   generic_acl_operations->getacl() operation
+ */
+static struct posix_acl *
+shmem_get_acl(struct inode *inode, int type)
+{
+	struct posix_acl *acl = NULL;
+
+	spin_lock(&inode->i_lock);
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			acl = posix_acl_dup(SHMEM_I(inode)->i_acl);
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			acl = posix_acl_dup(SHMEM_I(inode)->i_default_acl);
+			break;
+	}
+	spin_unlock(&inode->i_lock);
+
+	return acl;
+}
+
+/**
+ * shmem_get_acl  -   generic_acl_operations->setacl() operation
+ */
+static void
+shmem_set_acl(struct inode *inode, int type, struct posix_acl *acl)
+{
+	struct posix_acl *free = NULL;
+
+	spin_lock(&inode->i_lock);
+	switch(type) {
+		case ACL_TYPE_ACCESS:
+			free = SHMEM_I(inode)->i_acl;
+			SHMEM_I(inode)->i_acl = posix_acl_dup(acl);
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			free = SHMEM_I(inode)->i_default_acl;
+			SHMEM_I(inode)->i_default_acl = posix_acl_dup(acl);
+			break;
+	}
+	spin_unlock(&inode->i_lock);
+	posix_acl_release(free);
+}
+
+struct generic_acl_operations shmem_acl_ops = {
+	.getacl = shmem_get_acl,
+	.setacl = shmem_set_acl,
+};
+
+/**
+ * shmem_list_acl_access, shmem_get_acl_access, shmem_set_acl_access,
+ * shmem_xattr_acl_access_handler  -  plumbing code to implement the
+ * system.posix_acl_access xattr using the generic acl functions.
+ */
+
+static size_t
+shmem_list_acl_access(struct inode *inode, char *list, size_t list_size,
+		      const char *name, size_t name_len)
+{
+	return generic_acl_list(inode, &shmem_acl_ops, ACL_TYPE_ACCESS,
+				list, list_size);
+}
+
+static int
+shmem_get_acl_access(struct inode *inode, const char *name, void *buffer,
+		     size_t size)
+{
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	return generic_acl_get(inode, &shmem_acl_ops, ACL_TYPE_ACCESS, buffer,
+			       size);
+}
+
+static int
+shmem_set_acl_access(struct inode *inode, const char *name, const void *value,
+		     size_t size, int flags)
+{
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	return generic_acl_set(inode, &shmem_acl_ops, ACL_TYPE_ACCESS, value,
+			       size);
+}
+
+struct xattr_handler shmem_xattr_acl_access_handler = {
+	.prefix = POSIX_ACL_XATTR_ACCESS,
+	.list	= shmem_list_acl_access,
+	.get	= shmem_get_acl_access,
+	.set	= shmem_set_acl_access,
+};
+
+/**
+ * shmem_list_acl_default, shmem_get_acl_default, shmem_set_acl_default,
+ * shmem_xattr_acl_default_handler  -  plumbing code to implement the
+ * system.posix_acl_default xattr using the generic acl functions.
+ */
+
+static size_t
+shmem_list_acl_default(struct inode *inode, char *list, size_t list_size,
+		       const char *name, size_t name_len)
+{
+	return generic_acl_list(inode, &shmem_acl_ops, ACL_TYPE_DEFAULT,
+				list, list_size);
+}
+
+static int
+shmem_get_acl_default(struct inode *inode, const char *name, void *buffer,
+		      size_t size)
+{
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	return generic_acl_get(inode, &shmem_acl_ops, ACL_TYPE_DEFAULT, buffer,
+			       size);
+}
+
+static int
+shmem_set_acl_default(struct inode *inode, const char *name, const void *value,
+		      size_t size, int flags)
+{
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	return generic_acl_set(inode, &shmem_acl_ops, ACL_TYPE_DEFAULT, value,
+			       size);
+}
+
+struct xattr_handler shmem_xattr_acl_default_handler = {
+	.prefix = POSIX_ACL_XATTR_DEFAULT,
+	.list	= shmem_list_acl_default,
+	.get	= shmem_get_acl_default,
+	.set	= shmem_set_acl_default,
+};
+
+/**
+ * shmem_acl_init  -  Inizialize the acl(s) of a new inode
+ */
+int
+shmem_acl_init(struct inode *inode, struct inode *dir)
+{
+	return generic_acl_init(inode, dir, &shmem_acl_ops);
+}
+
+/**
+ * shmem_acl_destroy_inode  -  destroy acls hanging off the in-memory inode
+ *
+ * This is done before destroying the actual inode.
+ */
+
+void
+shmem_acl_destroy_inode(struct inode *inode)
+{
+	if (SHMEM_I(inode)->i_acl)
+		posix_acl_release(SHMEM_I(inode)->i_acl);
+	SHMEM_I(inode)->i_acl = NULL;
+	if (SHMEM_I(inode)->i_default_acl)
+		posix_acl_release(SHMEM_I(inode)->i_default_acl);
+	SHMEM_I(inode)->i_default_acl = NULL;
+}
+
+/**
+ * shmem_check_acl  -  check_acl() callback for generic_permission()
+ */
+static int
+shmem_check_acl(struct inode *inode, int mask)
+{
+	struct posix_acl *acl = shmem_get_acl(inode, ACL_TYPE_ACCESS);
+
+	if (acl) {
+		int error = posix_acl_permission(inode, acl, mask);
+		posix_acl_release(acl);
+		return error;
+	}
+	return -EAGAIN;
+}
+
+/**
+ * shmem_permission  -  permission() inode operation
+ */
+int
+shmem_permission(struct inode *inode, int mask, struct nameidata *nd)
+{
+	return generic_permission(inode, mask, shmem_check_acl);
+}
-- 
cgit v1.2.3


From b885808e185a4ec2dfe16c84434f79e95f0245b0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sat, 30 Sep 2006 01:47:55 +0200
Subject: [PATCH] Add proper sparse __user casts to __copy_to_user_inatomic

Noticed by Al Viro

Cc: viro@ftp.linux.org.uk

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/uaccess.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index e856570c0689..19f99178fe83 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -361,6 +361,11 @@ __must_check unsigned long clear_user(void __user *mem, unsigned long len);
 __must_check unsigned long __clear_user(void __user *mem, unsigned long len);
 
 __must_check long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
-#define __copy_to_user_inatomic copy_user_generic
+
+static __must_check __always_inline int
+__copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
+{
+	return copy_user_generic((__force void *)dst, src, size);
+}
 
 #endif /* __X86_64_UACCESS_H */
-- 
cgit v1.2.3


From 29cbc78b90a73ad80f2f58ba2927956cf663abed Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sat, 30 Sep 2006 01:47:55 +0200
Subject: [PATCH] x86: Clean up x86 NMI sysctls

Use prototypes in headers
Don't define panic_on_unrecovered_nmi for all architectures

Cc: dzickus@redhat.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c   |  3 +++
 arch/i386/kernel/traps.c |  2 ++
 arch/x86_64/kernel/nmi.c |  4 ++++
 include/asm-i386/nmi.h   |  6 ++++++
 include/asm-x86_64/nmi.h |  7 +++++++
 kernel/panic.c           |  1 -
 kernel/sysctl.c          | 11 ++++-------
 7 files changed, 26 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index dbda706fdd14..0fc4997fb143 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -31,6 +31,9 @@
 
 #include "mach_traps.h"
 
+int unknown_nmi_panic;
+int nmi_watchdog_enabled;
+
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
  * - different performance counters/ event selection may be reserved for
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index a13037fe0ee3..6820b8d643c7 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -57,6 +57,8 @@
 
 #include "mach_traps.h"
 
+int panic_on_unrecovered_nmi;
+
 asmlinkage int system_call(void);
 
 struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 },
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 4d6fb047952e..7af9cb3e2d99 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -28,6 +28,10 @@
 #include <asm/mce.h>
 #include <asm/intel_arch_perfmon.h>
 
+int unknown_nmi_panic;
+int nmi_watchdog_enabled;
+int panic_on_unrecovered_nmi;
+
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
  * - different performance counters/ event selection may be reserved for
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 303bcd4592bb..269d315719ca 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -36,4 +36,10 @@ extern unsigned int nmi_watchdog;
 #define NMI_LOCAL_APIC	2
 #define NMI_INVALID	3
 
+struct ctl_table;
+struct file;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
+extern int unknown_nmi_panic;
+
 #endif /* ASM_NMI_H */
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index cbf2669bca71..f367d4014b42 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -70,4 +70,11 @@ extern unsigned int nmi_watchdog;
 #define NMI_LOCAL_APIC	2
 #define NMI_INVALID	3
 
+struct ctl_table;
+struct file;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
+
+extern int unknown_nmi_panic;
+
 #endif /* ASM_NMI_H */
diff --git a/kernel/panic.c b/kernel/panic.c
index 6ceb664fb52a..525e365f7239 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -21,7 +21,6 @@
 #include <linux/debug_locks.h>
 
 int panic_on_oops;
-int panic_on_unrecovered_nmi;
 int tainted;
 static int pause_on_oops;
 static int pause_on_oops_flag;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 9535a3839930..c57c4532e296 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -52,6 +52,10 @@
 extern int proc_nr_files(ctl_table *table, int write, struct file *filp,
                      void __user *buffer, size_t *lenp, loff_t *ppos);
 
+#ifdef CONFIG_X86
+#include <asm/nmi.h>
+#endif
+
 #if defined(CONFIG_SYSCTL)
 
 /* External variables not in a header file. */
@@ -74,13 +78,6 @@ extern int sysctl_drop_caches;
 extern int percpu_pagelist_fraction;
 extern int compat_log;
 
-#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86)
-int unknown_nmi_panic;
-int nmi_watchdog_enabled;
-extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
-			void __user *, size_t *, loff_t *);
-#endif
-
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
 static int maxolduid = 65535;
 static int minolduid;
-- 
cgit v1.2.3


From 34596dc9e59d7bece16fe5aba08116b49465da26 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sat, 30 Sep 2006 01:47:55 +0200
Subject: [PATCH] Define vsyscall cache as blob to make clearer that user space
 shouldn't use it

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/vsyscall.c |  8 ++++----
 include/linux/getcpu.h        | 12 +++++++-----
 kernel/sys.c                  |  8 ++++----
 3 files changed, 15 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index ac48c3857ddb..07c086382059 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -155,8 +155,8 @@ vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 	   We do this here because otherwise user space would do it on
 	   its own in a likely inferior way (no access to jiffies).
 	   If you don't like it pass NULL. */
-	if (tcache && tcache->t0 == (j = __jiffies)) {
-		p = tcache->t1;
+	if (tcache && tcache->blob[0] == (j = __jiffies)) {
+		p = tcache->blob[1];
 	} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
 		/* Load per CPU data from RDTSCP */
 		rdtscp(dummy, dummy, p);
@@ -165,8 +165,8 @@ vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 		asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
 	}
 	if (tcache) {
-		tcache->t0 = j;
-		tcache->t1 = p;
+		tcache->blob[0] = j;
+		tcache->blob[1] = p;
 	}
 	if (cpu)
 		*cpu = p & 0xfff;
diff --git a/include/linux/getcpu.h b/include/linux/getcpu.h
index 031ed3780e45..c7372d7a97be 100644
--- a/include/linux/getcpu.h
+++ b/include/linux/getcpu.h
@@ -1,16 +1,18 @@
 #ifndef _LINUX_GETCPU_H
 #define _LINUX_GETCPU_H 1
 
-/* Cache for getcpu() to speed it up. Results might be upto a jiffie
+/* Cache for getcpu() to speed it up. Results might be a short time
    out of date, but will be faster.
+
    User programs should not refer to the contents of this structure.
-   It is only a cache for vgetcpu(). It might change in future kernels.
+   I repeat they should not refer to it. If they do they will break
+   in future kernels.
+
+   It is only a private cache for vgetcpu(). It will change in future kernels.
    The user program must store this information per thread (__thread)
    If you want 100% accurate information pass NULL instead. */
 struct getcpu_cache {
-	unsigned long t0;
-	unsigned long t1;
-	unsigned long res[4];
+	unsigned long blob[128 / sizeof(long)];
 };
 
 #endif
diff --git a/kernel/sys.c b/kernel/sys.c
index 8647061c084a..b88806c66244 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -2083,12 +2083,12 @@ asmlinkage long sys_getcpu(unsigned __user *cpup, unsigned __user *nodep,
 		 * padding
 		 */
 		unsigned long t0, t1;
-		get_user(t0, &cache->t0);
-		get_user(t1, &cache->t1);
+		get_user(t0, &cache->blob[0]);
+		get_user(t1, &cache->blob[1]);
 		t0++;
 		t1++;
-		put_user(t0, &cache->t0);
-		put_user(t1, &cache->t1);
+		put_user(t0, &cache->blob[0]);
+		put_user(t1, &cache->blob[1]);
 	}
 	return err ? -EFAULT : 0;
 }
-- 
cgit v1.2.3


From c84ef5305930d19bdd6cd576b3a3a73786a82e57 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sat, 30 Sep 2006 01:47:55 +0200
Subject: [PATCH] Use early clobber in semaphores

New code clobbers the result always early, so tell gcc about it
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/semaphore.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-x86_64/semaphore.h b/include/asm-x86_64/semaphore.h
index 107bd90429e8..1194888536b9 100644
--- a/include/asm-x86_64/semaphore.h
+++ b/include/asm-x86_64/semaphore.h
@@ -132,7 +132,7 @@ static inline int down_interruptible(struct semaphore * sem)
 		"jns 2f\n\t"
 		"call __down_failed_interruptible\n"
 		"2:\n"
-		:"=a" (result), "=m" (sem->count)
+		:"=&a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory");
 	return result;
@@ -153,7 +153,7 @@ static inline int down_trylock(struct semaphore * sem)
 		"jns 2f\n\t"
 		"call __down_failed_trylock\n\t"
 		"2:\n"
-		:"=a" (result), "=m" (sem->count)
+		:"=&a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory","cc");
 	return result;
-- 
cgit v1.2.3


From 95d4e6be25a68cd9fbe8c0d356b585504d8db1c7 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Fri, 29 Sep 2006 17:05:05 -0700
Subject: [NetLabel]: audit fixups due to delayed feedback

Fix some issues Steve Grubb had with the way NetLabel was using the audit
subsystem.  This should make NetLabel more consistent with other kernel
generated audit messages specifying configuration changes.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Acked-by: Steve Grubb <sgrubb@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/audit.h              | 11 +++--
 include/net/cipso_ipv4.h           |  4 +-
 include/net/netlabel.h             |  8 +++-
 net/ipv4/cipso_ipv4.c              |  4 +-
 net/netlabel/netlabel_cipso_v4.c   | 48 +++++++++++++---------
 net/netlabel/netlabel_domainhash.c | 82 ++++++++++++++++++++------------------
 net/netlabel/netlabel_domainhash.h |  8 ++--
 net/netlabel/netlabel_mgmt.c       | 27 +++++++++----
 net/netlabel/netlabel_unlabeled.c  | 34 +++++++++++-----
 net/netlabel/netlabel_user.c       | 66 ++++--------------------------
 net/netlabel/netlabel_user.h       | 16 +++++++-
 11 files changed, 157 insertions(+), 151 deletions(-)

(limited to 'include')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 42719d07612a..c3aa09751814 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -95,12 +95,11 @@
 #define AUDIT_MAC_POLICY_LOAD	1403	/* Policy file load */
 #define AUDIT_MAC_STATUS	1404	/* Changed enforcing,permissive,off */
 #define AUDIT_MAC_CONFIG_CHANGE	1405	/* Changes to booleans */
-#define AUDIT_MAC_UNLBL_ACCEPT	1406	/* NetLabel: allow unlabeled traffic */
-#define AUDIT_MAC_UNLBL_DENY	1407	/* NetLabel: deny unlabeled traffic */
-#define AUDIT_MAC_CIPSOV4_ADD	1408	/* NetLabel: add CIPSOv4 DOI entry */
-#define AUDIT_MAC_CIPSOV4_DEL	1409	/* NetLabel: del CIPSOv4 DOI entry */
-#define AUDIT_MAC_MAP_ADD	1410	/* NetLabel: add LSM domain mapping */
-#define AUDIT_MAC_MAP_DEL	1411	/* NetLabel: del LSM domain mapping */
+#define AUDIT_MAC_UNLBL_ALLOW	1406	/* NetLabel: allow unlabeled traffic */
+#define AUDIT_MAC_CIPSOV4_ADD	1407	/* NetLabel: add CIPSOv4 DOI entry */
+#define AUDIT_MAC_CIPSOV4_DEL	1408	/* NetLabel: del CIPSOv4 DOI entry */
+#define AUDIT_MAC_MAP_ADD	1409	/* NetLabel: add LSM domain mapping */
+#define AUDIT_MAC_MAP_DEL	1410	/* NetLabel: del LSM domain mapping */
 
 #define AUDIT_FIRST_KERN_ANOM_MSG   1700
 #define AUDIT_LAST_KERN_ANOM_MSG    1799
diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 5d6ae1b2b196..718b4d9c891f 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -129,7 +129,7 @@ extern int cipso_v4_rbm_strictvalid;
 #ifdef CONFIG_NETLABEL
 int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
 int cipso_v4_doi_remove(u32 doi,
-			u32 audit_secid,
+			struct netlbl_audit *audit_info,
 			void (*callback) (struct rcu_head * head));
 struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi);
 int cipso_v4_doi_walk(u32 *skip_cnt,
@@ -145,7 +145,7 @@ static inline int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
 }
 
 static inline int cipso_v4_doi_remove(u32 doi,
-				    u32 audit_secid,
+				    struct netlbl_audit *audit_info,
 				    void (*callback) (struct rcu_head * head))
 {
 	return 0;
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index 190bfdbbdba6..c63a58058e21 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -92,11 +92,17 @@
  *
  */
 
+/* NetLabel audit information */
+struct netlbl_audit {
+	u32 secid;
+	uid_t loginuid;
+};
+
 /* Domain mapping definition struct */
 struct netlbl_dom_map;
 
 /* Domain mapping operations */
-int netlbl_domhsh_remove(const char *domain, u32 audit_secid);
+int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info);
 
 /* LSM security attributes */
 struct netlbl_lsm_cache {
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index c4e469ff842d..a8e2e879a647 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -485,7 +485,7 @@ doi_add_failure_rlock:
  *
  */
 int cipso_v4_doi_remove(u32 doi,
-			u32 audit_secid,
+			struct netlbl_audit *audit_info,
 			void (*callback) (struct rcu_head * head))
 {
 	struct cipso_v4_doi *doi_def;
@@ -506,7 +506,7 @@ int cipso_v4_doi_remove(u32 doi,
 		list_for_each_entry_rcu(dom_iter, &doi_def->dom_list, list)
 			if (dom_iter->valid)
 				netlbl_domhsh_remove(dom_iter->domain,
-						     audit_secid);
+						     audit_info);
 		cipso_v4_cache_invalidate();
 		rcu_read_unlock();
 
diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c
index 09986ca962a6..a6ce1d6d5c59 100644
--- a/net/netlabel/netlabel_cipso_v4.c
+++ b/net/netlabel/netlabel_cipso_v4.c
@@ -384,11 +384,15 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info)
 	u32 doi;
 	const char *type_str = "(unknown)";
 	struct audit_buffer *audit_buf;
+	struct netlbl_audit audit_info;
 
 	if (!info->attrs[NLBL_CIPSOV4_A_DOI] ||
 	    !info->attrs[NLBL_CIPSOV4_A_MTYPE])
 		return -EINVAL;
 
+	doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]);
 	switch (type) {
 	case CIPSO_V4_MAP_STD:
@@ -401,13 +405,14 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info)
 		break;
 	}
 
-	if (ret_val == 0) {
-		doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
-		audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD,
-						      NETLINK_CB(skb).sid);
-		audit_log_format(audit_buf, " doi=%u type=%s", doi, type_str);
-		audit_log_end(audit_buf);
-	}
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD,
+					      &audit_info);
+	audit_log_format(audit_buf,
+			 " cipso_doi=%u cipso_type=%s res=%u",
+			 doi,
+			 type_str,
+			 ret_val == 0 ? 1 : 0);
+	audit_log_end(audit_buf);
 
 	return ret_val;
 }
@@ -668,20 +673,25 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
 	int ret_val = -EINVAL;
 	u32 doi = 0;
 	struct audit_buffer *audit_buf;
+	struct netlbl_audit audit_info;
 
-	if (info->attrs[NLBL_CIPSOV4_A_DOI]) {
-		doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
-		ret_val = cipso_v4_doi_remove(doi,
-					      NETLINK_CB(skb).sid,
-					      netlbl_cipsov4_doi_free);
-	}
+	if (!info->attrs[NLBL_CIPSOV4_A_DOI])
+		return -EINVAL;
 
-	if (ret_val == 0) {
-		audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_DEL,
-						      NETLINK_CB(skb).sid);
-		audit_log_format(audit_buf, " doi=%u", doi);
-		audit_log_end(audit_buf);
-	}
+	doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
+	ret_val = cipso_v4_doi_remove(doi,
+				      &audit_info,
+				      netlbl_cipsov4_doi_free);
+
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_DEL,
+					      &audit_info);
+	audit_log_format(audit_buf,
+			 " cipso_doi=%u res=%u",
+			 doi,
+			 ret_val == 0 ? 1 : 0);
+	audit_log_end(audit_buf);
 
 	return ret_val;
 }
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index d64e2ae3b129..af4371d3b459 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -188,7 +188,7 @@ int netlbl_domhsh_init(u32 size)
 /**
  * netlbl_domhsh_add - Adds a entry to the domain hash table
  * @entry: the entry to add
- * @audit_secid: the LSM secid to use in the audit message
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Adds a new entry to the domain hash table and handles any updates to the
@@ -196,7 +196,8 @@ int netlbl_domhsh_init(u32 size)
  * negative on failure.
  *
  */
-int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid)
+int netlbl_domhsh_add(struct netlbl_dom_map *entry,
+		      struct netlbl_audit *audit_info)
 {
 	int ret_val;
 	u32 bkt;
@@ -241,26 +242,26 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid)
 		spin_unlock(&netlbl_domhsh_def_lock);
 	} else
 		ret_val = -EINVAL;
-	if (ret_val == 0) {
-		if (entry->domain != NULL)
-			audit_domain = entry->domain;
-		else
-			audit_domain = "(default)";
-		audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD,
-						      audit_secid);
-		audit_log_format(audit_buf, " domain=%s", audit_domain);
-		switch (entry->type) {
-		case NETLBL_NLTYPE_UNLABELED:
-			audit_log_format(audit_buf, " protocol=unlbl");
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			audit_log_format(audit_buf,
-					 " protocol=cipsov4 doi=%u",
-					 entry->type_def.cipsov4->doi);
-			break;
-		}
-		audit_log_end(audit_buf);
+
+	if (entry->domain != NULL)
+		audit_domain = entry->domain;
+	else
+		audit_domain = "(default)";
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD, audit_info);
+	audit_log_format(audit_buf, " nlbl_domain=%s", audit_domain);
+	switch (entry->type) {
+	case NETLBL_NLTYPE_UNLABELED:
+		audit_log_format(audit_buf, " nlbl_protocol=unlbl");
+		break;
+	case NETLBL_NLTYPE_CIPSOV4:
+		audit_log_format(audit_buf,
+				 " nlbl_protocol=cipsov4 cipso_doi=%u",
+				 entry->type_def.cipsov4->doi);
+		break;
 	}
+	audit_log_format(audit_buf, " res=%u", ret_val == 0 ? 1 : 0);
+	audit_log_end(audit_buf);
+
 	rcu_read_unlock();
 
 	if (ret_val != 0) {
@@ -279,7 +280,7 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid)
 /**
  * netlbl_domhsh_add_default - Adds the default entry to the domain hash table
  * @entry: the entry to add
- * @audit_secid: the LSM secid to use in the audit message
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Adds a new default entry to the domain hash table and handles any updates
@@ -287,15 +288,16 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid)
  * negative on failure.
  *
  */
-int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, u32 audit_secid)
+int netlbl_domhsh_add_default(struct netlbl_dom_map *entry,
+			      struct netlbl_audit *audit_info)
 {
-	return netlbl_domhsh_add(entry, audit_secid);
+	return netlbl_domhsh_add(entry, audit_info);
 }
 
 /**
  * netlbl_domhsh_remove - Removes an entry from the domain hash table
  * @domain: the domain to remove
- * @audit_secid: the LSM secid to use in the audit message
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Removes an entry from the domain hash table and handles any updates to the
@@ -303,7 +305,7 @@ int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, u32 audit_secid)
  * negative on failure.
  *
  */
-int netlbl_domhsh_remove(const char *domain, u32 audit_secid)
+int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info)
 {
 	int ret_val = -ENOENT;
 	struct netlbl_dom_map *entry;
@@ -345,18 +347,20 @@ int netlbl_domhsh_remove(const char *domain, u32 audit_secid)
 			ret_val = -ENOENT;
 		spin_unlock(&netlbl_domhsh_def_lock);
 	}
-	if (ret_val == 0) {
-		if (entry->domain != NULL)
-			audit_domain = entry->domain;
-		else
-			audit_domain = "(default)";
-		audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_DEL,
-						      audit_secid);
-		audit_log_format(audit_buf, " domain=%s", audit_domain);
-		audit_log_end(audit_buf);
 
+	if (entry->domain != NULL)
+		audit_domain = entry->domain;
+	else
+		audit_domain = "(default)";
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_DEL, audit_info);
+	audit_log_format(audit_buf,
+			 " nlbl_domain=%s res=%u",
+			 audit_domain,
+			 ret_val == 0 ? 1 : 0);
+	audit_log_end(audit_buf);
+
+	if (ret_val == 0)
 		call_rcu(&entry->rcu, netlbl_domhsh_free_entry);
-	}
 
 remove_return:
 	rcu_read_unlock();
@@ -365,7 +369,7 @@ remove_return:
 
 /**
  * netlbl_domhsh_remove_default - Removes the default entry from the table
- * @audit_secid: the LSM secid to use in the audit message
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Removes/resets the default entry for the domain hash table and handles any
@@ -373,9 +377,9 @@ remove_return:
  * success, non-zero on failure.
  *
  */
-int netlbl_domhsh_remove_default(u32 audit_secid)
+int netlbl_domhsh_remove_default(struct netlbl_audit *audit_info)
 {
-	return netlbl_domhsh_remove(NULL, audit_secid);
+	return netlbl_domhsh_remove(NULL, audit_info);
 }
 
 /**
diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h
index d50f13cacdca..3689956c3436 100644
--- a/net/netlabel/netlabel_domainhash.h
+++ b/net/netlabel/netlabel_domainhash.h
@@ -57,9 +57,11 @@ struct netlbl_dom_map {
 int netlbl_domhsh_init(u32 size);
 
 /* Manipulate the domain hash table */
-int netlbl_domhsh_add(struct netlbl_dom_map *entry, u32 audit_secid);
-int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, u32 audit_secid);
-int netlbl_domhsh_remove_default(u32 audit_secid);
+int netlbl_domhsh_add(struct netlbl_dom_map *entry,
+		      struct netlbl_audit *audit_info);
+int netlbl_domhsh_add_default(struct netlbl_dom_map *entry,
+			      struct netlbl_audit *audit_info);
+int netlbl_domhsh_remove_default(struct netlbl_audit *audit_info);
 struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain);
 int netlbl_domhsh_walk(u32 *skip_bkt,
 		     u32 *skip_chain,
diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c
index 0ac314f18ad1..53c9079ad2c3 100644
--- a/net/netlabel/netlabel_mgmt.c
+++ b/net/netlabel/netlabel_mgmt.c
@@ -87,11 +87,14 @@ static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
 	struct netlbl_dom_map *entry = NULL;
 	size_t tmp_size;
 	u32 tmp_val;
+	struct netlbl_audit audit_info;
 
 	if (!info->attrs[NLBL_MGMT_A_DOMAIN] ||
 	    !info->attrs[NLBL_MGMT_A_PROTOCOL])
 		goto add_failure;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL) {
 		ret_val = -ENOMEM;
@@ -108,7 +111,7 @@ static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
 
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = netlbl_domhsh_add(entry, NETLINK_CB(skb).sid);
+		ret_val = netlbl_domhsh_add(entry, &audit_info);
 		break;
 	case NETLBL_NLTYPE_CIPSOV4:
 		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
@@ -125,7 +128,7 @@ static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
 			rcu_read_unlock();
 			goto add_failure;
 		}
-		ret_val = netlbl_domhsh_add(entry, NETLINK_CB(skb).sid);
+		ret_val = netlbl_domhsh_add(entry, &audit_info);
 		rcu_read_unlock();
 		break;
 	default:
@@ -156,12 +159,15 @@ add_failure:
 static int netlbl_mgmt_remove(struct sk_buff *skb, struct genl_info *info)
 {
 	char *domain;
+	struct netlbl_audit audit_info;
 
 	if (!info->attrs[NLBL_MGMT_A_DOMAIN])
 		return -EINVAL;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	domain = nla_data(info->attrs[NLBL_MGMT_A_DOMAIN]);
-	return netlbl_domhsh_remove(domain, NETLINK_CB(skb).sid);
+	return netlbl_domhsh_remove(domain, &audit_info);
 }
 
 /**
@@ -264,10 +270,13 @@ static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 	int ret_val = -EINVAL;
 	struct netlbl_dom_map *entry = NULL;
 	u32 tmp_val;
+	struct netlbl_audit audit_info;
 
 	if (!info->attrs[NLBL_MGMT_A_PROTOCOL])
 		goto adddef_failure;
 
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL) {
 		ret_val = -ENOMEM;
@@ -277,8 +286,7 @@ static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = netlbl_domhsh_add_default(entry,
-						    NETLINK_CB(skb).sid);
+		ret_val = netlbl_domhsh_add_default(entry, &audit_info);
 		break;
 	case NETLBL_NLTYPE_CIPSOV4:
 		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
@@ -295,8 +303,7 @@ static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 			rcu_read_unlock();
 			goto adddef_failure;
 		}
-		ret_val = netlbl_domhsh_add_default(entry,
-						    NETLINK_CB(skb).sid);
+		ret_val = netlbl_domhsh_add_default(entry, &audit_info);
 		rcu_read_unlock();
 		break;
 	default:
@@ -324,7 +331,11 @@ adddef_failure:
  */
 static int netlbl_mgmt_removedef(struct sk_buff *skb, struct genl_info *info)
 {
-	return netlbl_domhsh_remove_default(NETLINK_CB(skb).sid);
+	struct netlbl_audit audit_info;
+
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
+	return netlbl_domhsh_remove_default(&audit_info);
 }
 
 /**
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index ab36675fee8c..1833ad233b39 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -70,18 +70,25 @@ static struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1] = {
 /**
  * netlbl_unlabel_acceptflg_set - Set the unlabeled accept flag
  * @value: desired value
- * @audit_secid: the LSM secid to use in the audit message
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Set the value of the unlabeled accept flag to @value.
  *
  */
-static void netlbl_unlabel_acceptflg_set(u8 value, u32 audit_secid)
+static void netlbl_unlabel_acceptflg_set(u8 value,
+					 struct netlbl_audit *audit_info)
 {
+	struct audit_buffer *audit_buf;
+	u8 old_val;
+
+	old_val = atomic_read(&netlabel_unlabel_accept_flg);
 	atomic_set(&netlabel_unlabel_accept_flg, value);
-	netlbl_audit_nomsg((value ?
-			    AUDIT_MAC_UNLBL_ACCEPT : AUDIT_MAC_UNLBL_DENY),
-			   audit_secid);
+
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_ALLOW,
+					      audit_info);
+	audit_log_format(audit_buf, " unlbl_accept=%u old=%u", value, old_val);
+	audit_log_end(audit_buf);
 }
 
 /*
@@ -101,12 +108,13 @@ static void netlbl_unlabel_acceptflg_set(u8 value, u32 audit_secid)
 static int netlbl_unlabel_accept(struct sk_buff *skb, struct genl_info *info)
 {
 	u8 value;
+	struct netlbl_audit audit_info;
 
 	if (info->attrs[NLBL_UNLABEL_A_ACPTFLG]) {
 		value = nla_get_u8(info->attrs[NLBL_UNLABEL_A_ACPTFLG]);
 		if (value == 1 || value == 0) {
-			netlbl_unlabel_acceptflg_set(value,
-						     NETLINK_CB(skb).sid);
+			netlbl_netlink_auditinfo(skb, &audit_info);
+			netlbl_unlabel_acceptflg_set(value, &audit_info);
 			return 0;
 		}
 	}
@@ -250,19 +258,23 @@ int netlbl_unlabel_defconf(void)
 {
 	int ret_val;
 	struct netlbl_dom_map *entry;
-	u32 secid;
+	struct netlbl_audit audit_info;
 
-	security_task_getsecid(current, &secid);
+	/* Only the kernel is allowed to call this function and the only time
+	 * it is called is at bootup before the audit subsystem is reporting
+	 * messages so don't worry to much about these values. */
+	security_task_getsecid(current, &audit_info.secid);
+	audit_info.loginuid = 0;
 
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL)
 		return -ENOMEM;
 	entry->type = NETLBL_NLTYPE_UNLABELED;
-	ret_val = netlbl_domhsh_add_default(entry, secid);
+	ret_val = netlbl_domhsh_add_default(entry, &audit_info);
 	if (ret_val != 0)
 		return ret_val;
 
-	netlbl_unlabel_acceptflg_set(1, secid);
+	netlbl_unlabel_acceptflg_set(1, &audit_info);
 
 	return 0;
 }
diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c
index c2343af584cb..98a416381e61 100644
--- a/net/netlabel/netlabel_user.c
+++ b/net/netlabel/netlabel_user.c
@@ -85,7 +85,7 @@ int netlbl_netlink_init(void)
 /**
  * netlbl_audit_start_common - Start an audit message
  * @type: audit message type
- * @secid: LSM context ID
+ * @audit_info: NetLabel audit information
  *
  * Description:
  * Start an audit message using the type specified in @type and fill the audit
@@ -93,14 +93,11 @@ int netlbl_netlink_init(void)
  * a pointer to the audit buffer on success, NULL on failure.
  *
  */
-struct audit_buffer *netlbl_audit_start_common(int type, u32 secid)
+struct audit_buffer *netlbl_audit_start_common(int type,
+					       struct netlbl_audit *audit_info)
 {
 	struct audit_context *audit_ctx = current->audit_context;
 	struct audit_buffer *audit_buf;
-	uid_t audit_loginuid;
-	const char *audit_tty;
-	char audit_comm[sizeof(current->comm)];
-	struct vm_area_struct *vma;
 	char *secctx;
 	u32 secctx_len;
 
@@ -108,60 +105,13 @@ struct audit_buffer *netlbl_audit_start_common(int type, u32 secid)
 	if (audit_buf == NULL)
 		return NULL;
 
-	audit_loginuid = audit_get_loginuid(audit_ctx);
-	if (current->signal &&
-	    current->signal->tty &&
-	    current->signal->tty->name)
-		audit_tty = current->signal->tty->name;
-	else
-		audit_tty = "(none)";
-	get_task_comm(audit_comm, current);
+	audit_log_format(audit_buf, "netlabel: auid=%u", audit_info->loginuid);
 
-	audit_log_format(audit_buf,
-			 "netlabel: auid=%u uid=%u tty=%s pid=%d",
-			 audit_loginuid,
-			 current->uid,
-			 audit_tty,
-			 current->pid);
-	audit_log_format(audit_buf, " comm=");
-	audit_log_untrustedstring(audit_buf, audit_comm);
-	if (current->mm) {
-		down_read(&current->mm->mmap_sem);
-		vma = current->mm->mmap;
-		while (vma) {
-			if ((vma->vm_flags & VM_EXECUTABLE) &&
-			    vma->vm_file) {
-				audit_log_d_path(audit_buf,
-						 " exe=",
-						 vma->vm_file->f_dentry,
-						 vma->vm_file->f_vfsmnt);
-				break;
-			}
-			vma = vma->vm_next;
-		}
-		up_read(&current->mm->mmap_sem);
-	}
-
-	if (secid != 0 &&
-	    security_secid_to_secctx(secid, &secctx, &secctx_len) == 0)
+	if (audit_info->secid != 0 &&
+	    security_secid_to_secctx(audit_info->secid,
+				     &secctx,
+				     &secctx_len) == 0)
 		audit_log_format(audit_buf, " subj=%s", secctx);
 
 	return audit_buf;
 }
-
-/**
- * netlbl_audit_nomsg - Send an audit message without additional text
- * @type: audit message type
- * @secid: LSM context ID
- *
- * Description:
- * Send an audit message with only the common NetLabel audit fields.
- *
- */
-void netlbl_audit_nomsg(int type, u32 secid)
-{
-	struct audit_buffer *audit_buf;
-
-	audit_buf = netlbl_audit_start_common(type, secid);
-	audit_log_end(audit_buf);
-}
diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h
index ab840acfc964..47967ef32964 100644
--- a/net/netlabel/netlabel_user.h
+++ b/net/netlabel/netlabel_user.h
@@ -72,13 +72,25 @@ static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb,
 			   NETLBL_PROTO_VERSION);
 }
 
+/**
+ * netlbl_netlink_auditinfo - Fetch the audit information from a NETLINK msg
+ * @skb: the packet
+ * @audit_info: NetLabel audit information
+ */
+static inline void netlbl_netlink_auditinfo(struct sk_buff *skb,
+					    struct netlbl_audit *audit_info)
+{
+	audit_info->secid = NETLINK_CB(skb).sid;
+	audit_info->loginuid = NETLINK_CB(skb).loginuid;
+}
+
 /* NetLabel NETLINK I/O functions */
 
 int netlbl_netlink_init(void);
 
 /* NetLabel Audit Functions */
 
-struct audit_buffer *netlbl_audit_start_common(int type, u32 secid);
-void netlbl_audit_nomsg(int type, u32 secid);
+struct audit_buffer *netlbl_audit_start_common(int type,
+					      struct netlbl_audit *audit_info);
 
 #endif
-- 
cgit v1.2.3


From f9317a40c4e09e20ef01601fc9f5de9e6acb5b96 Mon Sep 17 00:00:00 2001
From: Michael Chan <mchan@broadcom.com>
Date: Fri, 29 Sep 2006 17:06:23 -0700
Subject: [BNX2]: Disable MSI on 5706 if AMD 8132 bridge is present.

MSI is defined to be 32-bit write.  The 5706 does 64-bit MSI writes
with byte enables disabled on the unused 32-bit word.  This is legal
but causes problems on the AMD 8132 which will eventually stop
responding after a while.

Without this patch, the MSI test done by the driver during open will
pass, but MSI will eventually stop working after a few MSIs are
written by the device.

AMD believes this incompatibility is unique to the 5706, and
prefers to locally disable MSI rather than globally disabling it
using pci_msi_quirk.

Update version to 1.4.45.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bnx2.c      | 32 ++++++++++++++++++++++++++++++--
 include/linux/pci_ids.h |  1 +
 2 files changed, 31 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 7fcf015021ec..6b4edb63c4c4 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -56,8 +56,8 @@
 
 #define DRV_MODULE_NAME		"bnx2"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_MODULE_VERSION	"1.4.44"
-#define DRV_MODULE_RELDATE	"August 10, 2006"
+#define DRV_MODULE_VERSION	"1.4.45"
+#define DRV_MODULE_RELDATE	"September 29, 2006"
 
 #define RUN_AT(x) (jiffies + (x))
 
@@ -5805,6 +5805,34 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		bp->cmd_ticks_int = bp->cmd_ticks;
 	}
 
+	/* Disable MSI on 5706 if AMD 8132 bridge is found.
+	 *
+	 * MSI is defined to be 32-bit write.  The 5706 does 64-bit MSI writes
+	 * with byte enables disabled on the unused 32-bit word.  This is legal
+	 * but causes problems on the AMD 8132 which will eventually stop
+	 * responding after a while.
+	 *
+	 * AMD believes this incompatibility is unique to the 5706, and
+	 * prefers to locally disable MSI rather than globally disabling it
+	 * using pci_msi_quirk.
+	 */
+	if (CHIP_NUM(bp) == CHIP_NUM_5706 && disable_msi == 0) {
+		struct pci_dev *amd_8132 = NULL;
+
+		while ((amd_8132 = pci_get_device(PCI_VENDOR_ID_AMD,
+						  PCI_DEVICE_ID_AMD_8132_BRIDGE,
+						  amd_8132))) {
+			u8 rev;
+
+			pci_read_config_byte(amd_8132, PCI_REVISION_ID, &rev);
+			if (rev >= 0x10 && rev <= 0x13) {
+				disable_msi = 1;
+				pci_dev_put(amd_8132);
+				break;
+			}
+		}
+	}
+
 	bp->autoneg = AUTONEG_SPEED | AUTONEG_FLOW_CTRL;
 	bp->req_line_speed = 0;
 	if (bp->phy_flags & PHY_SERDES_FLAG) {
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index b7e85ff045ea..c9ffbc3843d5 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -507,6 +507,7 @@
 #define PCI_DEVICE_ID_AMD_8151_0	0x7454
 #define PCI_DEVICE_ID_AMD_8131_BRIDGE	0x7450
 #define PCI_DEVICE_ID_AMD_8131_APIC	0x7451
+#define PCI_DEVICE_ID_AMD_8132_BRIDGE	0x7458
 #define PCI_DEVICE_ID_AMD_CS5536_ISA    0x2090
 #define PCI_DEVICE_ID_AMD_CS5536_FLASH  0x2091
 #define PCI_DEVICE_ID_AMD_CS5536_AUDIO  0x2093
-- 
cgit v1.2.3


From 1c9d3e72a7164c590437f2ab6c2c4f6da91f1703 Mon Sep 17 00:00:00 2001
From: Chas Williams <chas@cmf.nrl.navy.mil>
Date: Fri, 29 Sep 2006 17:13:24 -0700
Subject: [ATM]: [lec] header indent, comment and whitespace cleanup

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atmlec.h | 119 ++++++++++++++++++----------------
 net/atm/lec.h          | 172 +++++++++++++++++++++++++++----------------------
 net/atm/lec_arpc.h     | 148 ++++++++++++++++++++++--------------------
 3 files changed, 235 insertions(+), 204 deletions(-)

(limited to 'include')

diff --git a/include/linux/atmlec.h b/include/linux/atmlec.h
index f267f2442766..6f5a1bab8f50 100644
--- a/include/linux/atmlec.h
+++ b/include/linux/atmlec.h
@@ -1,9 +1,7 @@
 /*
- * 
- * ATM Lan Emulation Daemon vs. driver interface
- *
- * mkiiskila@yahoo.com
+ * ATM Lan Emulation Daemon driver interface
  *
+ * Marko Kiiskila <mkiiskila@yahoo.com>
  */
 
 #ifndef _ATMLEC_H_
@@ -13,76 +11,87 @@
 #include <linux/atmioc.h>
 #include <linux/atm.h>
 #include <linux/if_ether.h>
+
 /* ATM lec daemon control socket */
-#define ATMLEC_CTRL _IO('a',ATMIOC_LANE)
-#define ATMLEC_DATA _IO('a',ATMIOC_LANE+1)
-#define ATMLEC_MCAST _IO('a',ATMIOC_LANE+2)
+#define ATMLEC_CTRL	_IO('a', ATMIOC_LANE)
+#define ATMLEC_DATA	_IO('a', ATMIOC_LANE+1)
+#define ATMLEC_MCAST	_IO('a', ATMIOC_LANE+2)
 
 /* Maximum number of LEC interfaces (tweakable) */
 #define MAX_LEC_ITF 48
 
-/* From the total of MAX_LEC_ITF, last NUM_TR_DEVS are reserved for Token Ring.
+/*
+ * From the total of MAX_LEC_ITF, last NUM_TR_DEVS are reserved for Token Ring.
  * E.g. if MAX_LEC_ITF = 48 and NUM_TR_DEVS = 8, then lec0-lec39 are for
  * Ethernet ELANs and lec40-lec47 are for Token Ring ELANS.
  */
 #define NUM_TR_DEVS 8
 
-typedef enum { 
-        l_set_mac_addr,   l_del_mac_addr, 
-        l_svc_setup, 
-        l_addr_delete,    l_topology_change, 
-        l_flush_complete, l_arp_update,
-        l_narp_req, /* LANE2 mandates the use of this */
-        l_config,         l_flush_tran_id, 
-        l_set_lecid,      l_arp_xmt,
-        l_rdesc_arp_xmt,
-        l_associate_req,
-        l_should_bridge   /* should we bridge this MAC? */
+typedef enum {
+	l_set_mac_addr,
+	l_del_mac_addr,
+	l_svc_setup,
+	l_addr_delete,
+	l_topology_change,
+	l_flush_complete,
+	l_arp_update,
+	l_narp_req,		/* LANE2 mandates the use of this */
+	l_config,
+	l_flush_tran_id,
+	l_set_lecid,
+	l_arp_xmt,
+	l_rdesc_arp_xmt,
+	l_associate_req,
+	l_should_bridge		/* should we bridge this MAC? */
 } atmlec_msg_type;
 
 #define ATMLEC_MSG_TYPE_MAX l_should_bridge
 
 struct atmlec_config_msg {
-        unsigned int maximum_unknown_frame_count;
-        unsigned int max_unknown_frame_time;
-        unsigned short max_retry_count;
-        unsigned int aging_time;
-        unsigned int forward_delay_time;
-        unsigned int arp_response_time;
-        unsigned int flush_timeout;
-        unsigned int path_switching_delay;
-        unsigned int  lane_version; /* LANE2: 1 for LANEv1, 2 for LANEv2 */
-        int mtu;
-        int is_proxy;
+	unsigned int maximum_unknown_frame_count;
+	unsigned int max_unknown_frame_time;
+	unsigned short max_retry_count;
+	unsigned int aging_time;
+	unsigned int forward_delay_time;
+	unsigned int arp_response_time;
+	unsigned int flush_timeout;
+	unsigned int path_switching_delay;
+	unsigned int lane_version;	/* LANE2: 1 for LANEv1, 2 for LANEv2 */
+	int mtu;
+	int is_proxy;
 };
- 
+
 struct atmlec_msg {
-        atmlec_msg_type type;
-        int             sizeoftlvs;        /* LANE2: if != 0, tlvs follow */ 
-        union {
-                struct {
-                        unsigned char mac_addr[ETH_ALEN];
-                        unsigned char atm_addr[ATM_ESA_LEN];
-                        unsigned int flag;/* Topology_change flag, 
-                                              remoteflag, permanent flag,
-                                              lecid, transaction id */
-                        unsigned int targetless_le_arp; /* LANE2 */
-                        unsigned int no_source_le_narp; /* LANE2 */
-                } normal;
-                struct atmlec_config_msg config;
-                struct {
-                        uint16_t lec_id;                     /* requestor lec_id  */
-                        uint32_t tran_id;                    /* transaction id    */
-                        unsigned char mac_addr[ETH_ALEN];    /* dst mac addr      */
-                        unsigned char atm_addr[ATM_ESA_LEN]; /* reqestor ATM addr */
-                } proxy;
-		     /* For mapping LE_ARP requests to responses. Filled by */
-        } content;       /* zeppelin, returned by kernel. Used only when proxying */ 
+	atmlec_msg_type type;
+	int sizeoftlvs;		/* LANE2: if != 0, tlvs follow */
+	union {
+		struct {
+			unsigned char mac_addr[ETH_ALEN];
+			unsigned char atm_addr[ATM_ESA_LEN];
+			unsigned int flag;	/*
+						 * Topology_change flag,
+						 * remoteflag, permanent flag,
+						 * lecid, transaction id
+						 */
+			unsigned int targetless_le_arp;	/* LANE2 */
+			unsigned int no_source_le_narp;	/* LANE2 */
+		} normal;
+		struct atmlec_config_msg config;
+		struct {
+			uint16_t lec_id;			/* requestor lec_id  */
+			uint32_t tran_id;			/* transaction id    */
+			unsigned char mac_addr[ETH_ALEN];	/* dst mac addr      */
+			unsigned char atm_addr[ATM_ESA_LEN];	/* reqestor ATM addr */
+		} proxy;	/*
+				 * For mapping LE_ARP requests to responses. Filled by
+				 * zeppelin, returned by kernel. Used only when proxying
+				 */
+	} content;
 } __ATM_API_ALIGN;
 
 struct atmlec_ioc {
-        int dev_num;
-        unsigned char atm_addr[ATM_ESA_LEN];
-        unsigned char receive;    /* 1= receive vcc, 0 = send vcc */
+	int dev_num;
+	unsigned char atm_addr[ATM_ESA_LEN];
+	unsigned char receive;	/* 1= receive vcc, 0 = send vcc */
 };
 #endif /* _ATMLEC_H_ */
diff --git a/net/atm/lec.h b/net/atm/lec.h
index c22a8bfa1f81..c700472f64df 100644
--- a/net/atm/lec.h
+++ b/net/atm/lec.h
@@ -1,14 +1,13 @@
 /*
- *
  * Lan Emulation client header file
  *
- * Marko Kiiskila mkiiskila@yahoo.com
- *
+ * Marko Kiiskila <mkiiskila@yahoo.com>
  */
 
 #ifndef _LEC_H_
 #define _LEC_H_
 
+#include <linux/config.h>
 #include <linux/atmdev.h>
 #include <linux/netdevice.h>
 #include <linux/atmlec.h>
@@ -16,18 +15,18 @@
 #define LEC_HEADER_LEN 16
 
 struct lecdatahdr_8023 {
-  unsigned short le_header;
-  unsigned char h_dest[ETH_ALEN];
-  unsigned char h_source[ETH_ALEN];
-  unsigned short h_type;
+	unsigned short le_header;
+	unsigned char h_dest[ETH_ALEN];
+	unsigned char h_source[ETH_ALEN];
+	unsigned short h_type;
 };
 
 struct lecdatahdr_8025 {
-  unsigned short le_header;
-  unsigned char ac_pad;
-  unsigned char fc;
-  unsigned char h_dest[ETH_ALEN];
-  unsigned char h_source[ETH_ALEN];
+	unsigned short le_header;
+	unsigned char ac_pad;
+	unsigned char fc;
+	unsigned char h_dest[ETH_ALEN];
+	unsigned char h_source[ETH_ALEN];
 };
 
 #define LEC_MINIMUM_8023_SIZE   62
@@ -44,17 +43,18 @@ struct lecdatahdr_8025 {
  *
  */
 struct lane2_ops {
-	int  (*resolve)(struct net_device *dev, u8 *dst_mac, int force,
-                        u8 **tlvs, u32 *sizeoftlvs);
-        int  (*associate_req)(struct net_device *dev, u8 *lan_dst,
-                              u8 *tlvs, u32 sizeoftlvs);
-	void (*associate_indicator)(struct net_device *dev, u8 *mac_addr,
-                                    u8 *tlvs, u32 sizeoftlvs);
+	int (*resolve) (struct net_device *dev, u8 *dst_mac, int force,
+			u8 **tlvs, u32 *sizeoftlvs);
+	int (*associate_req) (struct net_device *dev, u8 *lan_dst,
+			      u8 *tlvs, u32 sizeoftlvs);
+	void (*associate_indicator) (struct net_device *dev, u8 *mac_addr,
+				     u8 *tlvs, u32 sizeoftlvs);
 };
 
 /*
  * ATM LAN Emulation supports both LLC & Dix Ethernet EtherType
  * frames. 
+ *
  * 1. Dix Ethernet EtherType frames encoded by placing EtherType
  *    field in h_type field. Data follows immediatelly after header.
  * 2. LLC Data frames whose total length, including LLC field and data,
@@ -70,72 +70,88 @@ struct lane2_ops {
 #define LEC_ARP_TABLE_SIZE 16
 
 struct lec_priv {
-        struct net_device_stats stats;
-        unsigned short lecid;      /* Lecid of this client */
-        struct lec_arp_table *lec_arp_empty_ones;
-        /* Used for storing VCC's that don't have a MAC address attached yet */
-        struct lec_arp_table *lec_arp_tables[LEC_ARP_TABLE_SIZE];
-        /* Actual LE ARP table */
-        struct lec_arp_table *lec_no_forward;
-        /* Used for storing VCC's (and forward packets from) which are to
-           age out by not using them to forward packets. 
-           This is because to some LE clients there will be 2 VCCs. Only
-           one of them gets used. */
-        struct lec_arp_table *mcast_fwds;
-        /* With LANEv2 it is possible that BUS (or a special multicast server)
-           establishes multiple Multicast Forward VCCs to us. This list
-           collects all those VCCs. LANEv1 client has only one item in this
-           list. These entries are not aged out. */
-        spinlock_t lec_arp_lock;
-        struct atm_vcc *mcast_vcc; /* Default Multicast Send VCC */
-        struct atm_vcc *lecd;
-        struct timer_list lec_arp_timer;
-        /* C10 */
-        unsigned int maximum_unknown_frame_count;
-/* Within the period of time defined by this variable, the client will send 
-   no more than C10 frames to BUS for a given unicast destination. (C11) */
-        unsigned long max_unknown_frame_time;
-/* If no traffic has been sent in this vcc for this period of time,
-   vcc will be torn down (C12)*/
-        unsigned long vcc_timeout_period;
-/* An LE Client MUST not retry an LE_ARP_REQUEST for a 
-   given frame's LAN Destination more than maximum retry count times,
-   after the first LEC_ARP_REQUEST (C13)*/
-        unsigned short max_retry_count;
-/* Max time the client will maintain an entry in its arp cache in
-   absence of a verification of that relationship (C17)*/
-        unsigned long aging_time;
-/* Max time the client will maintain an entry in cache when
-   topology change flag is true (C18) */
-        unsigned long forward_delay_time;
-/* Topology change flag  (C19)*/
-        int topology_change;
-/* Max time the client expects an LE_ARP_REQUEST/LE_ARP_RESPONSE
-   cycle to take (C20)*/
-        unsigned long arp_response_time;
-/* Time limit ot wait to receive an LE_FLUSH_RESPONSE after the
-   LE_FLUSH_REQUEST has been sent before taking recover action. (C21)*/
-        unsigned long flush_timeout;
-/* The time since sending a frame to the bus after which the
-   LE Client may assume that the frame has been either discarded or
-   delivered to the recipient (C22) */
-        unsigned long path_switching_delay;
+	struct net_device_stats stats;
+	unsigned short lecid;			/* Lecid of this client */
+	struct lec_arp_table *lec_arp_empty_ones;
+						/* Used for storing VCC's that don't have a MAC address attached yet */
+	struct lec_arp_table *lec_arp_tables[LEC_ARP_TABLE_SIZE];
+						/* Actual LE ARP table */
+	struct lec_arp_table *lec_no_forward;
+						/*
+						 * Used for storing VCC's (and forward packets from) which are to
+						 * age out by not using them to forward packets.
+						 * This is because to some LE clients there will be 2 VCCs. Only
+						 * one of them gets used.
+						 */
+	struct lec_arp_table *mcast_fwds;
+						/*
+						 * With LANEv2 it is possible that BUS (or a special multicast server)
+						 * establishes multiple Multicast Forward VCCs to us. This list
+						 * collects all those VCCs. LANEv1 client has only one item in this
+						 * list. These entries are not aged out.
+						 */
+	spinlock_t lec_arp_lock;
+	struct atm_vcc *mcast_vcc;		/* Default Multicast Send VCC */
+	struct atm_vcc *lecd;
+	struct timer_list lec_arp_timer;	/* C10 */
+	unsigned int maximum_unknown_frame_count;
+						/*
+						 * Within the period of time defined by this variable, the client will send
+						 * no more than C10 frames to BUS for a given unicast destination. (C11)
+						 */
+	unsigned long max_unknown_frame_time;
+						/*
+						 * If no traffic has been sent in this vcc for this period of time,
+						 * vcc will be torn down (C12)
+						 */
+	unsigned long vcc_timeout_period;
+						/*
+						 * An LE Client MUST not retry an LE_ARP_REQUEST for a
+						 * given frame's LAN Destination more than maximum retry count times,
+						 * after the first LEC_ARP_REQUEST (C13)
+						 */
+	unsigned short max_retry_count;
+						/*
+						 * Max time the client will maintain an entry in its arp cache in
+						 * absence of a verification of that relationship (C17)
+						 */
+	unsigned long aging_time;
+						/*
+						 * Max time the client will maintain an entry in cache when
+						 * topology change flag is true (C18)
+						 */
+	unsigned long forward_delay_time;	/* Topology change flag (C19) */
+	int topology_change;
+						/*
+						 * Max time the client expects an LE_ARP_REQUEST/LE_ARP_RESPONSE
+						 * cycle to take (C20)
+						 */
+	unsigned long arp_response_time;
+						/*
+						 * Time limit ot wait to receive an LE_FLUSH_RESPONSE after the
+						 * LE_FLUSH_REQUEST has been sent before taking recover action. (C21)
+						 */
+	unsigned long flush_timeout;
+						/* The time since sending a frame to the bus after which the
+						 * LE Client may assume that the frame has been either discarded or
+						 * delivered to the recipient (C22)
+						 */
+	unsigned long path_switching_delay;
 
-        u8 *tlvs;          /* LANE2: TLVs are new                */
-        u32 sizeoftlvs;    /* The size of the tlv array in bytes */
-        int lane_version;  /* LANE2                              */
-	int itfnum;        /* e.g. 2 for lec2, 5 for lec5        */
-        struct lane2_ops *lane2_ops; /* can be NULL for LANE v1  */
-        int is_proxy;      /* bridge between ATM and Ethernet    */
-        int is_trdev;      /* Device type, 0 = Ethernet, 1 = TokenRing */
+	u8 *tlvs;				/* LANE2: TLVs are new */
+	u32 sizeoftlvs;				/* The size of the tlv array in bytes */
+	int lane_version;			/* LANE2 */
+	int itfnum;				/* e.g. 2 for lec2, 5 for lec5 */
+	struct lane2_ops *lane2_ops;		/* can be NULL for LANE v1 */
+	int is_proxy;				/* bridge between ATM and Ethernet */
+	int is_trdev;				/* Device type, 0 = Ethernet, 1 = TokenRing */
 };
 
 struct lec_vcc_priv {
-	void (*old_pop)(struct atm_vcc *vcc, struct sk_buff *skb);
+	void (*old_pop) (struct atm_vcc *vcc, struct sk_buff *skb);
 	int xoff;
 };
 
 #define LEC_VCC_PRIV(vcc)	((struct lec_vcc_priv *)((vcc)->user_back))
 
-#endif /* _LEC_H_ */
-
+#endif				/* _LEC_H_ */
diff --git a/net/atm/lec_arpc.h b/net/atm/lec_arpc.h
index 397448094648..0230ca148c77 100644
--- a/net/atm/lec_arpc.h
+++ b/net/atm/lec_arpc.h
@@ -1,92 +1,98 @@
 /*
  * Lec arp cache
- * Marko Kiiskila mkiiskila@yahoo.com
  *
+ * Marko Kiiskila <mkiiskila@yahoo.com>
  */
-#ifndef _LEC_ARP_H
-#define _LEC_ARP_H
+#ifndef _LEC_ARP_H_
+#define _LEC_ARP_H_
 #include <linux/atm.h>
 #include <linux/atmdev.h>
 #include <linux/if_ether.h>
 #include <linux/atmlec.h>
 
 struct lec_arp_table {
-        struct lec_arp_table *next;          /* Linked entry list */
-        unsigned char atm_addr[ATM_ESA_LEN]; /* Atm address */
-        unsigned char mac_addr[ETH_ALEN];    /* Mac address */
-        int is_rdesc;                        /* Mac address is a route descriptor */
-        struct atm_vcc *vcc;                 /* Vcc this entry is attached */
-        struct atm_vcc *recv_vcc;            /* Vcc we receive data from */
-        void (*old_push)(struct atm_vcc *vcc,struct sk_buff *skb); 
-                                             /* Push that leads to daemon */
-        void (*old_recv_push)(struct atm_vcc *vcc, struct sk_buff *skb);
-                                             /* Push that leads to daemon */
-        void (*old_close)(struct atm_vcc *vcc);
-                                             /* We want to see when this
-                                              * vcc gets closed */
-        unsigned long last_used;             /* For expiry */
-        unsigned long timestamp;             /* Used for various timestamping
-                                              * things:
-                                              * 1. FLUSH started 
-                                              *    (status=ESI_FLUSH_PENDING)
-                                              * 2. Counting to 
-                                              *    max_unknown_frame_time
-                                              *    (status=ESI_ARP_PENDING||
-                                              *     status=ESI_VC_PENDING)
-                                              */
-        unsigned char no_tries;              /* No of times arp retry has been 
-                                                tried */
-        unsigned char status;                /* Status of this entry */
-        unsigned short flags;                /* Flags for this entry */
-        unsigned short packets_flooded;      /* Data packets flooded */
-        unsigned long flush_tran_id;         /* Transaction id in flush protocol */
-        struct timer_list timer;             /* Arping timer */
-        struct lec_priv *priv;               /* Pointer back */
+	struct lec_arp_table *next;	/* Linked entry list */
+	unsigned char atm_addr[ATM_ESA_LEN];	/* Atm address */
+	unsigned char mac_addr[ETH_ALEN];	/* Mac address */
+	int is_rdesc;			/* Mac address is a route descriptor */
+	struct atm_vcc *vcc;		/* Vcc this entry is attached */
+	struct atm_vcc *recv_vcc;	/* Vcc we receive data from */
 
-        u8  *tlvs;             /* LANE2: Each MAC address can have TLVs    */
-        u32 sizeoftlvs;        /* associated with it. sizeoftlvs tells the */
-                               /* the length of the tlvs array             */
-        struct sk_buff_head tx_wait; /* wait queue for outgoing packets    */
+	void (*old_push) (struct atm_vcc *vcc, struct sk_buff *skb);
+					/* Push that leads to daemon */
+
+	void (*old_recv_push) (struct atm_vcc *vcc, struct sk_buff *skb);
+					/* Push that leads to daemon */
+
+	void (*old_close) (struct atm_vcc *vcc);
+					/* We want to see when this vcc gets closed */
+
+	unsigned long last_used;	/* For expiry */
+	unsigned long timestamp;	/* Used for various timestamping things:
+					 * 1. FLUSH started
+					 *    (status=ESI_FLUSH_PENDING)
+					 * 2. Counting to
+					 *    max_unknown_frame_time
+					 *    (status=ESI_ARP_PENDING||
+					 *     status=ESI_VC_PENDING)
+					 */
+	unsigned char no_tries;		/* No of times arp retry has been tried */
+	unsigned char status;		/* Status of this entry */
+	unsigned short flags;		/* Flags for this entry */
+	unsigned short packets_flooded;	/* Data packets flooded */
+	unsigned long flush_tran_id;	/* Transaction id in flush protocol */
+	struct timer_list timer;	/* Arping timer */
+	struct lec_priv *priv;		/* Pointer back */
+	u8 *tlvs;
+	u32 sizeoftlvs;			/*
+					 * LANE2: Each MAC address can have TLVs
+					 * associated with it.  sizeoftlvs tells the
+					 * the length of the tlvs array
+					 */
+	struct sk_buff_head tx_wait;	/* wait queue for outgoing packets */
 };
 
-struct tlv {                   /* LANE2: Template tlv struct for accessing */
-                               /* the tlvs in the lec_arp_table->tlvs array*/
-        u32 type;
-        u8  length;
-        u8  value[255];
+/*
+ * LANE2: Template tlv struct for accessing
+ * the tlvs in the lec_arp_table->tlvs array
+ */
+struct tlv {
+	u32 type;
+	u8 length;
+	u8 value[255];
 };
 
 /* Status fields */
-#define ESI_UNKNOWN 0       /*
-                             * Next packet sent to this mac address
-                             * causes ARP-request to be sent 
-                             */
-#define ESI_ARP_PENDING 1   /*
-                             * There is no ATM address associated with this
-                             * 48-bit address.  The LE-ARP protocol is in
-                             * progress.
-                             */
-#define ESI_VC_PENDING 2    /*
-                             * There is a valid ATM address associated with 
-                             * this 48-bit address but there is no VC set 
-                             * up to that ATM address.  The signaling 
-                             * protocol is in process.
-                             */
-#define ESI_FLUSH_PENDING 4 /*
-                             * The LEC has been notified of the FLUSH_START
-                             * status and it is assumed that the flush 
-                             * protocol is in process.
-                             */
-#define ESI_FORWARD_DIRECT 5 /*
-                              * Either the Path Switching Delay (C22) has 
-                              * elapsed or the LEC has notified the Mapping 
-                              * that the flush protocol has completed.  In 
-                              * either case, it is safe to forward packets 
-                              * to this address via the data direct VC.
-                              */
+#define ESI_UNKNOWN 0		/*
+				 * Next packet sent to this mac address
+				 * causes ARP-request to be sent
+				 */
+#define ESI_ARP_PENDING 1	/*
+				 * There is no ATM address associated with this
+				 * 48-bit address.  The LE-ARP protocol is in
+				 * progress.
+				 */
+#define ESI_VC_PENDING 2	/*
+				 * There is a valid ATM address associated with
+				 * this 48-bit address but there is no VC set
+				 * up to that ATM address.  The signaling
+				 * protocol is in process.
+				 */
+#define ESI_FLUSH_PENDING 4	/*
+				 * The LEC has been notified of the FLUSH_START
+				 * status and it is assumed that the flush
+				 * protocol is in process.
+				 */
+#define ESI_FORWARD_DIRECT 5	/*
+				 * Either the Path Switching Delay (C22) has
+				 * elapsed or the LEC has notified the Mapping
+				 * that the flush protocol has completed.  In
+				 * either case, it is safe to forward packets
+				 * to this address via the data direct VC.
+				 */
 
 /* Flag values */
 #define LEC_REMOTE_FLAG      0x0001
 #define LEC_PERMANENT_FLAG   0x0002
 
-#endif
+#endif /* _LEC_ARP_H_ */
-- 
cgit v1.2.3


From 4aff5e2333c9a1609662f2091f55c3f6fffdad36 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Thu, 10 Aug 2006 08:44:47 +0200
Subject: [PATCH] Split struct request ->flags into two parts

Right now ->flags is a bit of a mess: some are request types, and
others are just modifiers. Clean this up by splitting it into
->cmd_type and ->cmd_flags. This allows introduction of generic
Linux block message types, useful for sending generic Linux commands
to block devices.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/as-iosched.c              |   2 +-
 block/elevator.c                |  26 +++---
 block/ll_rw_blk.c               | 101 ++++++++--------------
 block/scsi_ioctl.c              |   6 +-
 drivers/block/floppy.c          |   4 +-
 drivers/block/nbd.c             |   8 +-
 drivers/block/paride/pd.c       |   2 +-
 drivers/block/pktcdvd.c         |   6 +-
 drivers/block/xd.c              |   2 +-
 drivers/cdrom/cdrom.c           |   2 +-
 drivers/cdrom/cdu31a.c          |   4 +-
 drivers/ide/ide-cd.c            |  69 +++++++--------
 drivers/ide/ide-disk.c          |   5 +-
 drivers/ide/ide-dma.c           |   2 +-
 drivers/ide/ide-floppy.c        |  14 ++--
 drivers/ide/ide-io.c            |  36 ++++----
 drivers/ide/ide-lib.c           |   5 +-
 drivers/ide/ide-tape.c          |   8 +-
 drivers/ide/ide-taskfile.c      |   8 +-
 drivers/ide/ide.c               |   4 +-
 drivers/ide/legacy/hd.c         |   2 +-
 drivers/md/dm-emc.c             |   3 +-
 drivers/message/i2o/i2o_block.c |   7 +-
 drivers/mmc/mmc_queue.c         |   6 +-
 drivers/mtd/mtd_blkdevs.c       |   2 +-
 drivers/s390/block/dasd_diag.c  |   2 +-
 drivers/s390/block/dasd_eckd.c  |   2 +-
 drivers/s390/block/dasd_fba.c   |   2 +-
 drivers/scsi/aic7xxx_old.c      |   4 +-
 drivers/scsi/ide-scsi.c         |  14 ++--
 drivers/scsi/pluto.c            |   6 +-
 drivers/scsi/scsi_lib.c         |  37 +++++----
 drivers/scsi/sd.c               |   5 +-
 drivers/scsi/sun3_NCR5380.c     |   2 +-
 drivers/scsi/sun3_scsi.c        |   2 +-
 drivers/scsi/sun3_scsi_vme.c    |   2 +-
 include/linux/blkdev.h          | 180 ++++++++++++++++++++++------------------
 include/linux/blktrace_api.h    |   2 +-
 include/scsi/scsi_tcq.h         |   2 +-
 39 files changed, 295 insertions(+), 301 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index 5da56d48fbd3..ad1cc4077819 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -1335,7 +1335,7 @@ static void as_add_request(request_queue_t *q, struct request *rq)
 	arq->state = AS_RQ_NEW;
 
 	if (rq_data_dir(arq->request) == READ
-			|| (arq->request->flags & REQ_RW_SYNC))
+			|| (arq->request->cmd_flags & REQ_RW_SYNC))
 		arq->is_sync = 1;
 	else
 		arq->is_sync = 0;
diff --git a/block/elevator.c b/block/elevator.c
index 9b72dc7c8a5c..4ac97b642042 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -242,7 +242,7 @@ void elv_dispatch_sort(request_queue_t *q, struct request *rq)
 	list_for_each_prev(entry, &q->queue_head) {
 		struct request *pos = list_entry_rq(entry);
 
-		if (pos->flags & (REQ_SOFTBARRIER|REQ_HARDBARRIER|REQ_STARTED))
+		if (pos->cmd_flags & (REQ_SOFTBARRIER|REQ_HARDBARRIER|REQ_STARTED))
 			break;
 		if (rq->sector >= boundary) {
 			if (pos->sector < boundary)
@@ -313,7 +313,7 @@ void elv_requeue_request(request_queue_t *q, struct request *rq)
 			e->ops->elevator_deactivate_req_fn(q, rq);
 	}
 
-	rq->flags &= ~REQ_STARTED;
+	rq->cmd_flags &= ~REQ_STARTED;
 
 	elv_insert(q, rq, ELEVATOR_INSERT_REQUEUE);
 }
@@ -344,13 +344,13 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
 
 	switch (where) {
 	case ELEVATOR_INSERT_FRONT:
-		rq->flags |= REQ_SOFTBARRIER;
+		rq->cmd_flags |= REQ_SOFTBARRIER;
 
 		list_add(&rq->queuelist, &q->queue_head);
 		break;
 
 	case ELEVATOR_INSERT_BACK:
-		rq->flags |= REQ_SOFTBARRIER;
+		rq->cmd_flags |= REQ_SOFTBARRIER;
 		elv_drain_elevator(q);
 		list_add_tail(&rq->queuelist, &q->queue_head);
 		/*
@@ -369,7 +369,7 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
 
 	case ELEVATOR_INSERT_SORT:
 		BUG_ON(!blk_fs_request(rq));
-		rq->flags |= REQ_SORTED;
+		rq->cmd_flags |= REQ_SORTED;
 		q->nr_sorted++;
 		if (q->last_merge == NULL && rq_mergeable(rq))
 			q->last_merge = rq;
@@ -387,7 +387,7 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
 		 * insertion; otherwise, requests should be requeued
 		 * in ordseq order.
 		 */
-		rq->flags |= REQ_SOFTBARRIER;
+		rq->cmd_flags |= REQ_SOFTBARRIER;
 
 		if (q->ordseq == 0) {
 			list_add(&rq->queuelist, &q->queue_head);
@@ -429,9 +429,9 @@ void __elv_add_request(request_queue_t *q, struct request *rq, int where,
 		       int plug)
 {
 	if (q->ordcolor)
-		rq->flags |= REQ_ORDERED_COLOR;
+		rq->cmd_flags |= REQ_ORDERED_COLOR;
 
-	if (rq->flags & (REQ_SOFTBARRIER | REQ_HARDBARRIER)) {
+	if (rq->cmd_flags & (REQ_SOFTBARRIER | REQ_HARDBARRIER)) {
 		/*
 		 * toggle ordered color
 		 */
@@ -452,7 +452,7 @@ void __elv_add_request(request_queue_t *q, struct request *rq, int where,
 			q->end_sector = rq_end_sector(rq);
 			q->boundary_rq = rq;
 		}
-	} else if (!(rq->flags & REQ_ELVPRIV) && where == ELEVATOR_INSERT_SORT)
+	} else if (!(rq->cmd_flags & REQ_ELVPRIV) && where == ELEVATOR_INSERT_SORT)
 		where = ELEVATOR_INSERT_BACK;
 
 	if (plug)
@@ -493,7 +493,7 @@ struct request *elv_next_request(request_queue_t *q)
 	int ret;
 
 	while ((rq = __elv_next_request(q)) != NULL) {
-		if (!(rq->flags & REQ_STARTED)) {
+		if (!(rq->cmd_flags & REQ_STARTED)) {
 			elevator_t *e = q->elevator;
 
 			/*
@@ -510,7 +510,7 @@ struct request *elv_next_request(request_queue_t *q)
 			 * it, a request that has been delayed should
 			 * not be passed by new incoming requests
 			 */
-			rq->flags |= REQ_STARTED;
+			rq->cmd_flags |= REQ_STARTED;
 			blk_add_trace_rq(q, rq, BLK_TA_ISSUE);
 		}
 
@@ -519,7 +519,7 @@ struct request *elv_next_request(request_queue_t *q)
 			q->boundary_rq = NULL;
 		}
 
-		if ((rq->flags & REQ_DONTPREP) || !q->prep_rq_fn)
+		if ((rq->cmd_flags & REQ_DONTPREP) || !q->prep_rq_fn)
 			break;
 
 		ret = q->prep_rq_fn(q, rq);
@@ -541,7 +541,7 @@ struct request *elv_next_request(request_queue_t *q)
 				nr_bytes = rq->data_len;
 
 			blkdev_dequeue_request(rq);
-			rq->flags |= REQ_QUIET;
+			rq->cmd_flags |= REQ_QUIET;
 			end_that_request_chunk(rq, 0, nr_bytes);
 			end_that_request_last(rq, 0);
 		} else {
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 51dc0edf76e0..9b91bb70c5ed 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -382,8 +382,8 @@ unsigned blk_ordered_req_seq(struct request *rq)
 	if (rq == &q->post_flush_rq)
 		return QUEUE_ORDSEQ_POSTFLUSH;
 
-	if ((rq->flags & REQ_ORDERED_COLOR) ==
-	    (q->orig_bar_rq->flags & REQ_ORDERED_COLOR))
+	if ((rq->cmd_flags & REQ_ORDERED_COLOR) ==
+	    (q->orig_bar_rq->cmd_flags & REQ_ORDERED_COLOR))
 		return QUEUE_ORDSEQ_DRAIN;
 	else
 		return QUEUE_ORDSEQ_DONE;
@@ -446,8 +446,8 @@ static void queue_flush(request_queue_t *q, unsigned which)
 		end_io = post_flush_end_io;
 	}
 
+	rq->cmd_flags = REQ_HARDBARRIER;
 	rq_init(q, rq);
-	rq->flags = REQ_HARDBARRIER;
 	rq->elevator_private = NULL;
 	rq->rq_disk = q->bar_rq.rq_disk;
 	rq->rl = NULL;
@@ -471,9 +471,11 @@ static inline struct request *start_ordered(request_queue_t *q,
 	blkdev_dequeue_request(rq);
 	q->orig_bar_rq = rq;
 	rq = &q->bar_rq;
+	rq->cmd_flags = 0;
 	rq_init(q, rq);
-	rq->flags = bio_data_dir(q->orig_bar_rq->bio);
-	rq->flags |= q->ordered & QUEUE_ORDERED_FUA ? REQ_FUA : 0;
+	if (bio_data_dir(q->orig_bar_rq->bio) == WRITE)
+		rq->cmd_flags |= REQ_RW;
+	rq->cmd_flags |= q->ordered & QUEUE_ORDERED_FUA ? REQ_FUA : 0;
 	rq->elevator_private = NULL;
 	rq->rl = NULL;
 	init_request_from_bio(rq, q->orig_bar_rq->bio);
@@ -1124,7 +1126,7 @@ void blk_queue_end_tag(request_queue_t *q, struct request *rq)
 	}
 
 	list_del_init(&rq->queuelist);
-	rq->flags &= ~REQ_QUEUED;
+	rq->cmd_flags &= ~REQ_QUEUED;
 	rq->tag = -1;
 
 	if (unlikely(bqt->tag_index[tag] == NULL))
@@ -1160,7 +1162,7 @@ int blk_queue_start_tag(request_queue_t *q, struct request *rq)
 	struct blk_queue_tag *bqt = q->queue_tags;
 	int tag;
 
-	if (unlikely((rq->flags & REQ_QUEUED))) {
+	if (unlikely((rq->cmd_flags & REQ_QUEUED))) {
 		printk(KERN_ERR 
 		       "%s: request %p for device [%s] already tagged %d",
 		       __FUNCTION__, rq,
@@ -1174,7 +1176,7 @@ int blk_queue_start_tag(request_queue_t *q, struct request *rq)
 
 	__set_bit(tag, bqt->tag_map);
 
-	rq->flags |= REQ_QUEUED;
+	rq->cmd_flags |= REQ_QUEUED;
 	rq->tag = tag;
 	bqt->tag_index[tag] = rq;
 	blkdev_dequeue_request(rq);
@@ -1210,65 +1212,31 @@ void blk_queue_invalidate_tags(request_queue_t *q)
 			printk(KERN_ERR
 			       "%s: bad tag found on list\n", __FUNCTION__);
 			list_del_init(&rq->queuelist);
-			rq->flags &= ~REQ_QUEUED;
+			rq->cmd_flags &= ~REQ_QUEUED;
 		} else
 			blk_queue_end_tag(q, rq);
 
-		rq->flags &= ~REQ_STARTED;
+		rq->cmd_flags &= ~REQ_STARTED;
 		__elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 0);
 	}
 }
 
 EXPORT_SYMBOL(blk_queue_invalidate_tags);
 
-static const char * const rq_flags[] = {
-	"REQ_RW",
-	"REQ_FAILFAST",
-	"REQ_SORTED",
-	"REQ_SOFTBARRIER",
-	"REQ_HARDBARRIER",
-	"REQ_FUA",
-	"REQ_CMD",
-	"REQ_NOMERGE",
-	"REQ_STARTED",
-	"REQ_DONTPREP",
-	"REQ_QUEUED",
-	"REQ_ELVPRIV",
-	"REQ_PC",
-	"REQ_BLOCK_PC",
-	"REQ_SENSE",
-	"REQ_FAILED",
-	"REQ_QUIET",
-	"REQ_SPECIAL",
-	"REQ_DRIVE_CMD",
-	"REQ_DRIVE_TASK",
-	"REQ_DRIVE_TASKFILE",
-	"REQ_PREEMPT",
-	"REQ_PM_SUSPEND",
-	"REQ_PM_RESUME",
-	"REQ_PM_SHUTDOWN",
-	"REQ_ORDERED_COLOR",
-};
-
 void blk_dump_rq_flags(struct request *rq, char *msg)
 {
 	int bit;
 
-	printk("%s: dev %s: flags = ", msg,
-		rq->rq_disk ? rq->rq_disk->disk_name : "?");
-	bit = 0;
-	do {
-		if (rq->flags & (1 << bit))
-			printk("%s ", rq_flags[bit]);
-		bit++;
-	} while (bit < __REQ_NR_BITS);
+	printk("%s: dev %s: type=%x, flags=%x\n", msg,
+		rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->cmd_type,
+		rq->cmd_flags);
 
 	printk("\nsector %llu, nr/cnr %lu/%u\n", (unsigned long long)rq->sector,
 						       rq->nr_sectors,
 						       rq->current_nr_sectors);
 	printk("bio %p, biotail %p, buffer %p, data %p, len %u\n", rq->bio, rq->biotail, rq->buffer, rq->data, rq->data_len);
 
-	if (rq->flags & (REQ_BLOCK_PC | REQ_PC)) {
+	if (blk_pc_request(rq)) {
 		printk("cdb: ");
 		for (bit = 0; bit < sizeof(rq->cmd); bit++)
 			printk("%02x ", rq->cmd[bit]);
@@ -1441,7 +1409,7 @@ static inline int ll_new_mergeable(request_queue_t *q,
 	int nr_phys_segs = bio_phys_segments(q, bio);
 
 	if (req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
-		req->flags |= REQ_NOMERGE;
+		req->cmd_flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
@@ -1464,7 +1432,7 @@ static inline int ll_new_hw_segment(request_queue_t *q,
 
 	if (req->nr_hw_segments + nr_hw_segs > q->max_hw_segments
 	    || req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
-		req->flags |= REQ_NOMERGE;
+		req->cmd_flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
@@ -1491,7 +1459,7 @@ static int ll_back_merge_fn(request_queue_t *q, struct request *req,
 		max_sectors = q->max_sectors;
 
 	if (req->nr_sectors + bio_sectors(bio) > max_sectors) {
-		req->flags |= REQ_NOMERGE;
+		req->cmd_flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
@@ -1530,7 +1498,7 @@ static int ll_front_merge_fn(request_queue_t *q, struct request *req,
 
 
 	if (req->nr_sectors + bio_sectors(bio) > max_sectors) {
-		req->flags |= REQ_NOMERGE;
+		req->cmd_flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
@@ -2029,7 +1997,7 @@ EXPORT_SYMBOL(blk_get_queue);
 
 static inline void blk_free_request(request_queue_t *q, struct request *rq)
 {
-	if (rq->flags & REQ_ELVPRIV)
+	if (rq->cmd_flags & REQ_ELVPRIV)
 		elv_put_request(q, rq);
 	mempool_free(rq, q->rq.rq_pool);
 }
@@ -2044,17 +2012,17 @@ blk_alloc_request(request_queue_t *q, int rw, struct bio *bio,
 		return NULL;
 
 	/*
-	 * first three bits are identical in rq->flags and bio->bi_rw,
+	 * first three bits are identical in rq->cmd_flags and bio->bi_rw,
 	 * see bio.h and blkdev.h
 	 */
-	rq->flags = rw;
+	rq->cmd_flags = rw;
 
 	if (priv) {
 		if (unlikely(elv_set_request(q, rq, bio, gfp_mask))) {
 			mempool_free(rq, q->rq.rq_pool);
 			return NULL;
 		}
-		rq->flags |= REQ_ELVPRIV;
+		rq->cmd_flags |= REQ_ELVPRIV;
 	}
 
 	return rq;
@@ -2351,7 +2319,8 @@ void blk_insert_request(request_queue_t *q, struct request *rq,
 	 * must not attempt merges on this) and that it acts as a soft
 	 * barrier
 	 */
-	rq->flags |= REQ_SPECIAL | REQ_SOFTBARRIER;
+	rq->cmd_type = REQ_TYPE_SPECIAL;
+	rq->cmd_flags |= REQ_SOFTBARRIER;
 
 	rq->special = data;
 
@@ -2558,7 +2527,7 @@ void blk_execute_rq_nowait(request_queue_t *q, struct gendisk *bd_disk,
 	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;
 
 	rq->rq_disk = bd_disk;
-	rq->flags |= REQ_NOMERGE;
+	rq->cmd_flags |= REQ_NOMERGE;
 	rq->end_io = done;
 	WARN_ON(irqs_disabled());
 	spin_lock_irq(q->queue_lock);
@@ -2728,7 +2697,7 @@ void __blk_put_request(request_queue_t *q, struct request *req)
 	 */
 	if (rl) {
 		int rw = rq_data_dir(req);
-		int priv = req->flags & REQ_ELVPRIV;
+		int priv = req->cmd_flags & REQ_ELVPRIV;
 
 		BUG_ON(!list_empty(&req->queuelist));
 
@@ -2890,22 +2859,22 @@ static inline int attempt_front_merge(request_queue_t *q, struct request *rq)
 
 static void init_request_from_bio(struct request *req, struct bio *bio)
 {
-	req->flags |= REQ_CMD;
+	req->cmd_type = REQ_TYPE_FS;
 
 	/*
 	 * inherit FAILFAST from bio (for read-ahead, and explicit FAILFAST)
 	 */
 	if (bio_rw_ahead(bio) || bio_failfast(bio))
-		req->flags |= REQ_FAILFAST;
+		req->cmd_flags |= REQ_FAILFAST;
 
 	/*
 	 * REQ_BARRIER implies no merging, but lets make it explicit
 	 */
 	if (unlikely(bio_barrier(bio)))
-		req->flags |= (REQ_HARDBARRIER | REQ_NOMERGE);
+		req->cmd_flags |= (REQ_HARDBARRIER | REQ_NOMERGE);
 
 	if (bio_sync(bio))
-		req->flags |= REQ_RW_SYNC;
+		req->cmd_flags |= REQ_RW_SYNC;
 
 	req->errors = 0;
 	req->hard_sector = req->sector = bio->bi_sector;
@@ -3306,7 +3275,7 @@ static int __end_that_request_first(struct request *req, int uptodate,
 		req->errors = 0;
 
 	if (!uptodate) {
-		if (blk_fs_request(req) && !(req->flags & REQ_QUIET))
+		if (blk_fs_request(req) && !(req->cmd_flags & REQ_QUIET))
 			printk("end_request: I/O error, dev %s, sector %llu\n",
 				req->rq_disk ? req->rq_disk->disk_name : "?",
 				(unsigned long long)req->sector);
@@ -3569,8 +3538,8 @@ EXPORT_SYMBOL(end_request);
 
 void blk_rq_bio_prep(request_queue_t *q, struct request *rq, struct bio *bio)
 {
-	/* first two bits are identical in rq->flags and bio->bi_rw */
-	rq->flags |= (bio->bi_rw & 3);
+	/* first two bits are identical in rq->cmd_flags and bio->bi_rw */
+	rq->cmd_flags |= (bio->bi_rw & 3);
 
 	rq->nr_phys_segments = bio_phys_segments(q, bio);
 	rq->nr_hw_segments = bio_hw_segments(q, bio);
diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c
index b33eda26e205..2dc326421a24 100644
--- a/block/scsi_ioctl.c
+++ b/block/scsi_ioctl.c
@@ -294,7 +294,7 @@ static int sg_io(struct file *file, request_queue_t *q,
 	rq->sense = sense;
 	rq->sense_len = 0;
 
-	rq->flags |= REQ_BLOCK_PC;
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	bio = rq->bio;
 
 	/*
@@ -470,7 +470,7 @@ int sg_scsi_ioctl(struct file *file, struct request_queue *q,
 	memset(sense, 0, sizeof(sense));
 	rq->sense = sense;
 	rq->sense_len = 0;
-	rq->flags |= REQ_BLOCK_PC;
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 
 	blk_execute_rq(q, disk, rq, 0);
 
@@ -502,7 +502,7 @@ static int __blk_send_generic(request_queue_t *q, struct gendisk *bd_disk, int c
 	int err;
 
 	rq = blk_get_request(q, WRITE, __GFP_WAIT);
-	rq->flags |= REQ_BLOCK_PC;
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->data = NULL;
 	rq->data_len = 0;
 	rq->timeout = BLK_DEFAULT_TIMEOUT;
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index ad1d7065a1b2..629c5769d994 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -2991,8 +2991,8 @@ static void do_fd_request(request_queue_t * q)
 	if (usage_count == 0) {
 		printk("warning: usage count=0, current_req=%p exiting\n",
 		       current_req);
-		printk("sect=%ld flags=%lx\n", (long)current_req->sector,
-		       current_req->flags);
+		printk("sect=%ld type=%x flags=%x\n", (long)current_req->sector,
+		       current_req->cmd_type, current_req->cmd_flags);
 		return;
 	}
 	if (test_bit(0, &fdc_busy)) {
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index bdbade9a5cf5..9d1035e8d9d8 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -407,10 +407,10 @@ static void do_nbd_request(request_queue_t * q)
 		struct nbd_device *lo;
 
 		blkdev_dequeue_request(req);
-		dprintk(DBG_BLKDEV, "%s: request %p: dequeued (flags=%lx)\n",
-				req->rq_disk->disk_name, req, req->flags);
+		dprintk(DBG_BLKDEV, "%s: request %p: dequeued (flags=%x)\n",
+				req->rq_disk->disk_name, req, req->cmd_type);
 
-		if (!(req->flags & REQ_CMD))
+		if (!blk_fs_request(req))
 			goto error_out;
 
 		lo = req->rq_disk->private_data;
@@ -489,7 +489,7 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
 	switch (cmd) {
 	case NBD_DISCONNECT:
 	        printk(KERN_INFO "%s: NBD_DISCONNECT\n", lo->disk->disk_name);
-		sreq.flags = REQ_SPECIAL;
+		sreq.cmd_type = REQ_TYPE_SPECIAL;
 		nbd_cmd(&sreq) = NBD_CMD_DISC;
 		/*
 		 * Set these to sane values in case server implementation
diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c
index 2403721f9db1..12ff1a274d91 100644
--- a/drivers/block/paride/pd.c
+++ b/drivers/block/paride/pd.c
@@ -437,7 +437,7 @@ static char *pd_buf;		/* buffer for request in progress */
 
 static enum action do_pd_io_start(void)
 {
-	if (pd_req->flags & REQ_SPECIAL) {
+	if (blk_special_request(pd_req)) {
 		phase = pd_special;
 		return pd_special();
 	}
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 451b996bba91..42891d2b054e 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -365,16 +365,16 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	rq->sense = sense;
 	memset(sense, 0, sizeof(sense));
 	rq->sense_len = 0;
-	rq->flags |= REQ_BLOCK_PC | REQ_HARDBARRIER;
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
+	rq->cmd_flags |= REQ_HARDBARRIER;
 	if (cgc->quiet)
-		rq->flags |= REQ_QUIET;
+		rq->cmd_flags |= REQ_QUIET;
 	memcpy(rq->cmd, cgc->cmd, CDROM_PACKET_SIZE);
 	if (sizeof(rq->cmd) > CDROM_PACKET_SIZE)
 		memset(rq->cmd + CDROM_PACKET_SIZE, 0, sizeof(rq->cmd) - CDROM_PACKET_SIZE);
 	rq->cmd_len = COMMAND_SIZE(rq->cmd[0]);
 
 	rq->ref_count++;
-	rq->flags |= REQ_NOMERGE;
 	rq->waiting = &wait;
 	rq->end_io = blk_end_sync_rq;
 	elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 1);
diff --git a/drivers/block/xd.c b/drivers/block/xd.c
index e828e4cbd3e1..ebf3025721d1 100644
--- a/drivers/block/xd.c
+++ b/drivers/block/xd.c
@@ -313,7 +313,7 @@ static void do_xd_request (request_queue_t * q)
 		int res = 0;
 		int retry;
 
-		if (!(req->flags & REQ_CMD)) {
+		if (!blk_fs_request(req)) {
 			end_request(req, 0);
 			continue;
 		}
diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
index d239cf8b20bd..b38c84a7a8e3 100644
--- a/drivers/cdrom/cdrom.c
+++ b/drivers/cdrom/cdrom.c
@@ -2129,7 +2129,7 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 		rq->cmd[9] = 0xf8;
 
 		rq->cmd_len = 12;
-		rq->flags |= REQ_BLOCK_PC;
+		rq->cmd_type = REQ_TYPE_BLOCK_PC;
 		rq->timeout = 60 * HZ;
 		bio = rq->bio;
 
diff --git a/drivers/cdrom/cdu31a.c b/drivers/cdrom/cdu31a.c
index 37bdb0163f0d..ccd91c1a84bd 100644
--- a/drivers/cdrom/cdu31a.c
+++ b/drivers/cdrom/cdu31a.c
@@ -1338,8 +1338,10 @@ static void do_cdu31a_request(request_queue_t * q)
 		}
 
 		/* WTF??? */
-		if (!(req->flags & REQ_CMD))
+		if (!blk_fs_request(req)) {
+			end_request(req, 0);
 			continue;
+		}
 		if (rq_data_dir(req) == WRITE) {
 			end_request(req, 0);
 			continue;
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 654d4cd09847..69bbb6206a00 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -372,7 +372,7 @@ static int cdrom_log_sense(ide_drive_t *drive, struct request *rq,
 {
 	int log = 0;
 
-	if (!sense || !rq || (rq->flags & REQ_QUIET))
+	if (!sense || !rq || (rq->cmd_flags & REQ_QUIET))
 		return 0;
 
 	switch (sense->sense_key) {
@@ -597,7 +597,7 @@ static void cdrom_prepare_request(ide_drive_t *drive, struct request *rq)
 	struct cdrom_info *cd = drive->driver_data;
 
 	ide_init_drive_cmd(rq);
-	rq->flags = REQ_PC;
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->rq_disk = cd->disk;
 }
 
@@ -617,7 +617,7 @@ static void cdrom_queue_request_sense(ide_drive_t *drive, void *sense,
 	rq->cmd[0] = GPCMD_REQUEST_SENSE;
 	rq->cmd[4] = rq->data_len = 18;
 
-	rq->flags = REQ_SENSE;
+	rq->cmd_type = REQ_TYPE_SENSE;
 
 	/* NOTE! Save the failed command in "rq->buffer" */
 	rq->buffer = (void *) failed_command;
@@ -630,10 +630,10 @@ static void cdrom_end_request (ide_drive_t *drive, int uptodate)
 	struct request *rq = HWGROUP(drive)->rq;
 	int nsectors = rq->hard_cur_sectors;
 
-	if ((rq->flags & REQ_SENSE) && uptodate) {
+	if (blk_sense_request(rq) && uptodate) {
 		/*
-		 * For REQ_SENSE, "rq->buffer" points to the original failed
-		 * request
+		 * For REQ_TYPE_SENSE, "rq->buffer" points to the original
+		 * failed request
 		 */
 		struct request *failed = (struct request *) rq->buffer;
 		struct cdrom_info *info = drive->driver_data;
@@ -706,17 +706,17 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 		return 1;
 	}
 
-	if (rq->flags & REQ_SENSE) {
+	if (blk_sense_request(rq)) {
 		/* We got an error trying to get sense info
 		   from the drive (probably while trying
 		   to recover from a former error).  Just give up. */
 
-		rq->flags |= REQ_FAILED;
+		rq->cmd_flags |= REQ_FAILED;
 		cdrom_end_request(drive, 0);
 		ide_error(drive, "request sense failure", stat);
 		return 1;
 
-	} else if (rq->flags & (REQ_PC | REQ_BLOCK_PC)) {
+	} else if (blk_pc_request(rq)) {
 		/* All other functions, except for READ. */
 		unsigned long flags;
 
@@ -724,7 +724,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 		 * if we have an error, pass back CHECK_CONDITION as the
 		 * scsi status byte
 		 */
-		if ((rq->flags & REQ_BLOCK_PC) && !rq->errors)
+		if (!rq->errors)
 			rq->errors = SAM_STAT_CHECK_CONDITION;
 
 		/* Check for tray open. */
@@ -735,12 +735,12 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 			cdrom_saw_media_change (drive);
 			/*printk("%s: media changed\n",drive->name);*/
 			return 0;
-		} else if (!(rq->flags & REQ_QUIET)) {
+		} else if (!(rq->cmd_flags & REQ_QUIET)) {
 			/* Otherwise, print an error. */
 			ide_dump_status(drive, "packet command error", stat);
 		}
 		
-		rq->flags |= REQ_FAILED;
+		rq->cmd_flags |= REQ_FAILED;
 
 		/*
 		 * instead of playing games with moving completions around,
@@ -881,7 +881,7 @@ static int cdrom_timer_expiry(ide_drive_t *drive)
 			wait = ATAPI_WAIT_PC;
 			break;
 		default:
-			if (!(rq->flags & REQ_QUIET))
+			if (!(rq->cmd_flags & REQ_QUIET))
 				printk(KERN_INFO "ide-cd: cmd 0x%x timed out\n", rq->cmd[0]);
 			wait = 0;
 			break;
@@ -1124,7 +1124,7 @@ static ide_startstop_t cdrom_read_intr (ide_drive_t *drive)
 		if (rq->current_nr_sectors > 0) {
 			printk (KERN_ERR "%s: cdrom_read_intr: data underrun (%d blocks)\n",
 				drive->name, rq->current_nr_sectors);
-			rq->flags |= REQ_FAILED;
+			rq->cmd_flags |= REQ_FAILED;
 			cdrom_end_request(drive, 0);
 		} else
 			cdrom_end_request(drive, 1);
@@ -1456,7 +1456,7 @@ static ide_startstop_t cdrom_pc_intr (ide_drive_t *drive)
 			printk ("%s: cdrom_pc_intr: data underrun %d\n",
 				drive->name, pc->buflen);
 			*/
-			rq->flags |= REQ_FAILED;
+			rq->cmd_flags |= REQ_FAILED;
 			cdrom_end_request(drive, 0);
 		}
 		return ide_stopped;
@@ -1509,7 +1509,7 @@ static ide_startstop_t cdrom_pc_intr (ide_drive_t *drive)
 		rq->data += thislen;
 		rq->data_len -= thislen;
 
-		if (rq->flags & REQ_SENSE)
+		if (blk_sense_request(rq))
 			rq->sense_len += thislen;
 	} else {
 confused:
@@ -1517,7 +1517,7 @@ confused:
 			"appears confused (ireason = 0x%02x). "
 			"Trying to recover by ending request.\n",
 			drive->name, ireason);
-		rq->flags |= REQ_FAILED;
+		rq->cmd_flags |= REQ_FAILED;
 		cdrom_end_request(drive, 0);
 		return ide_stopped;
 	}
@@ -1546,7 +1546,7 @@ static ide_startstop_t cdrom_do_packet_command (ide_drive_t *drive)
 	struct cdrom_info *info = drive->driver_data;
 
 	info->dma = 0;
-	rq->flags &= ~REQ_FAILED;
+	rq->cmd_flags &= ~REQ_FAILED;
 	len = rq->data_len;
 
 	/* Start sending the command to the drive. */
@@ -1558,7 +1558,7 @@ static int cdrom_queue_packet_command(ide_drive_t *drive, struct request *rq)
 {
 	struct request_sense sense;
 	int retries = 10;
-	unsigned int flags = rq->flags;
+	unsigned int flags = rq->cmd_flags;
 
 	if (rq->sense == NULL)
 		rq->sense = &sense;
@@ -1567,14 +1567,14 @@ static int cdrom_queue_packet_command(ide_drive_t *drive, struct request *rq)
 	do {
 		int error;
 		unsigned long time = jiffies;
-		rq->flags = flags;
+		rq->cmd_flags = flags;
 
 		error = ide_do_drive_cmd(drive, rq, ide_wait);
 		time = jiffies - time;
 
 		/* FIXME: we should probably abort/retry or something 
 		 * in case of failure */
-		if (rq->flags & REQ_FAILED) {
+		if (rq->cmd_flags & REQ_FAILED) {
 			/* The request failed.  Retry if it was due to a unit
 			   attention status
 			   (usually means media was changed). */
@@ -1596,10 +1596,10 @@ static int cdrom_queue_packet_command(ide_drive_t *drive, struct request *rq)
 		}
 
 		/* End of retry loop. */
-	} while ((rq->flags & REQ_FAILED) && retries >= 0);
+	} while ((rq->cmd_flags & REQ_FAILED) && retries >= 0);
 
 	/* Return an error if the command failed. */
-	return (rq->flags & REQ_FAILED) ? -EIO : 0;
+	return (rq->cmd_flags & REQ_FAILED) ? -EIO : 0;
 }
 
 /*
@@ -1963,7 +1963,7 @@ static ide_startstop_t cdrom_do_block_pc(ide_drive_t *drive, struct request *rq)
 {
 	struct cdrom_info *info = drive->driver_data;
 
-	rq->flags |= REQ_QUIET;
+	rq->cmd_flags |= REQ_QUIET;
 
 	info->dma = 0;
 
@@ -2023,11 +2023,11 @@ ide_do_rw_cdrom (ide_drive_t *drive, struct request *rq, sector_t block)
 		}
 		info->last_block = block;
 		return action;
-	} else if (rq->flags & (REQ_PC | REQ_SENSE)) {
+	} else if (rq->cmd_type == REQ_TYPE_SENSE) {
 		return cdrom_do_packet_command(drive);
-	} else if (rq->flags & REQ_BLOCK_PC) {
+	} else if (blk_pc_request(rq)) {
 		return cdrom_do_block_pc(drive, rq);
-	} else if (rq->flags & REQ_SPECIAL) {
+	} else if (blk_special_request(rq)) {
 		/*
 		 * right now this can only be a reset...
 		 */
@@ -2105,7 +2105,7 @@ static int cdrom_check_status(ide_drive_t *drive, struct request_sense *sense)
 
 	req.sense = sense;
 	req.cmd[0] = GPCMD_TEST_UNIT_READY;
-	req.flags |= REQ_QUIET;
+	req.cmd_flags |= REQ_QUIET;
 
 #if ! STANDARD_ATAPI
         /* the Sanyo 3 CD changer uses byte 7 of TEST_UNIT_READY to 
@@ -2207,7 +2207,7 @@ static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity,
 	req.cmd[0] = GPCMD_READ_CDVD_CAPACITY;
 	req.data = (char *)&capbuf;
 	req.data_len = sizeof(capbuf);
-	req.flags |= REQ_QUIET;
+	req.cmd_flags |= REQ_QUIET;
 
 	stat = cdrom_queue_packet_command(drive, &req);
 	if (stat == 0) {
@@ -2230,7 +2230,7 @@ static int cdrom_read_tocentry(ide_drive_t *drive, int trackno, int msf_flag,
 	req.sense = sense;
 	req.data =  buf;
 	req.data_len = buflen;
-	req.flags |= REQ_QUIET;
+	req.cmd_flags |= REQ_QUIET;
 	req.cmd[0] = GPCMD_READ_TOC_PMA_ATIP;
 	req.cmd[6] = trackno;
 	req.cmd[7] = (buflen >> 8);
@@ -2531,7 +2531,7 @@ static int ide_cdrom_packet(struct cdrom_device_info *cdi,
 	req.timeout = cgc->timeout;
 
 	if (cgc->quiet)
-		req.flags |= REQ_QUIET;
+		req.cmd_flags |= REQ_QUIET;
 
 	req.sense = cgc->sense;
 	cgc->stat = cdrom_queue_packet_command(drive, &req);
@@ -2629,7 +2629,8 @@ int ide_cdrom_reset (struct cdrom_device_info *cdi)
 	int ret;
 
 	cdrom_prepare_request(drive, &req);
-	req.flags = REQ_SPECIAL | REQ_QUIET;
+	req.cmd_type = REQ_TYPE_SPECIAL;
+	req.cmd_flags = REQ_QUIET;
 	ret = ide_do_drive_cmd(drive, &req, ide_wait);
 
 	/*
@@ -3116,9 +3117,9 @@ static int ide_cdrom_prep_pc(struct request *rq)
 
 static int ide_cdrom_prep_fn(request_queue_t *q, struct request *rq)
 {
-	if (rq->flags & REQ_CMD)
+	if (blk_fs_request(rq))
 		return ide_cdrom_prep_fs(q, rq);
-	else if (rq->flags & REQ_BLOCK_PC)
+	else if (blk_pc_request(rq))
 		return ide_cdrom_prep_pc(rq);
 
 	return 0;
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 7cf3eb023521..0a05a377d66a 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -699,7 +699,8 @@ static void idedisk_prepare_flush(request_queue_t *q, struct request *rq)
 		rq->cmd[0] = WIN_FLUSH_CACHE;
 
 
-	rq->flags |= REQ_DRIVE_TASK;
+	rq->cmd_type = REQ_TYPE_ATA_TASK;
+	rq->cmd_flags |= REQ_SOFTBARRIER;
 	rq->buffer = rq->cmd;
 }
 
@@ -740,7 +741,7 @@ static int set_multcount(ide_drive_t *drive, int arg)
 	if (drive->special.b.set_multmode)
 		return -EBUSY;
 	ide_init_drive_cmd (&rq);
-	rq.flags = REQ_DRIVE_CMD;
+	rq.cmd_type = REQ_TYPE_ATA_CMD;
 	drive->mult_req = arg;
 	drive->special.b.set_multmode = 1;
 	(void) ide_do_drive_cmd (drive, &rq, ide_wait);
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 7c3a13e1cf64..c3546fe9af63 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -205,7 +205,7 @@ int ide_build_sglist(ide_drive_t *drive, struct request *rq)
 	ide_hwif_t *hwif = HWIF(drive);
 	struct scatterlist *sg = hwif->sg_table;
 
-	BUG_ON((rq->flags & REQ_DRIVE_TASKFILE) && rq->nr_sectors > 256);
+	BUG_ON((rq->cmd_type == REQ_TYPE_ATA_TASKFILE) && rq->nr_sectors > 256);
 
 	ide_map_sg(drive, rq);
 
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index adbe9f76a505..0edc32204915 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -588,7 +588,7 @@ static int idefloppy_do_end_request(ide_drive_t *drive, int uptodate, int nsecs)
 	/* Why does this happen? */
 	if (!rq)
 		return 0;
-	if (!(rq->flags & REQ_SPECIAL)) { //if (!IDEFLOPPY_RQ_CMD (rq->cmd)) {
+	if (!blk_special_request(rq)) {
 		/* our real local end request function */
 		ide_end_request(drive, uptodate, nsecs);
 		return 0;
@@ -689,7 +689,7 @@ static void idefloppy_queue_pc_head (ide_drive_t *drive,idefloppy_pc_t *pc,struc
 
 	ide_init_drive_cmd(rq);
 	rq->buffer = (char *) pc;
-	rq->flags = REQ_SPECIAL;	//rq->cmd = IDEFLOPPY_PC_RQ;
+	rq->cmd_type = REQ_TYPE_SPECIAL;
 	rq->rq_disk = floppy->disk;
 	(void) ide_do_drive_cmd(drive, rq, ide_preempt);
 }
@@ -1250,7 +1250,7 @@ static void idefloppy_create_rw_cmd (idefloppy_floppy_t *floppy, idefloppy_pc_t
 	pc->callback = &idefloppy_rw_callback;
 	pc->rq = rq;
 	pc->b_count = cmd == READ ? 0 : rq->bio->bi_size;
-	if (rq->flags & REQ_RW)
+	if (rq->cmd_flags & REQ_RW)
 		set_bit(PC_WRITING, &pc->flags);
 	pc->buffer = NULL;
 	pc->request_transfer = pc->buffer_size = blocks * floppy->block_size;
@@ -1303,7 +1303,7 @@ static ide_startstop_t idefloppy_do_request (ide_drive_t *drive, struct request
 		idefloppy_do_end_request(drive, 0, 0);
 		return ide_stopped;
 	}
-	if (rq->flags & REQ_CMD) {
+	if (blk_fs_request(rq)) {
 		if (((long)rq->sector % floppy->bs_factor) ||
 		    (rq->nr_sectors % floppy->bs_factor)) {
 			printk("%s: unsupported r/w request size\n",
@@ -1313,9 +1313,9 @@ static ide_startstop_t idefloppy_do_request (ide_drive_t *drive, struct request
 		}
 		pc = idefloppy_next_pc_storage(drive);
 		idefloppy_create_rw_cmd(floppy, pc, rq, block);
-	} else if (rq->flags & REQ_SPECIAL) {
+	} else if (blk_special_request(rq)) {
 		pc = (idefloppy_pc_t *) rq->buffer;
-	} else if (rq->flags & REQ_BLOCK_PC) {
+	} else if (blk_pc_request(rq)) {
 		pc = idefloppy_next_pc_storage(drive);
 		if (idefloppy_blockpc_cmd(floppy, pc, rq)) {
 			idefloppy_do_end_request(drive, 0, 0);
@@ -1343,7 +1343,7 @@ static int idefloppy_queue_pc_tail (ide_drive_t *drive,idefloppy_pc_t *pc)
 
 	ide_init_drive_cmd (&rq);
 	rq.buffer = (char *) pc;
-	rq.flags = REQ_SPECIAL;		//	rq.cmd = IDEFLOPPY_PC_RQ;
+	rq.cmd_type = REQ_TYPE_SPECIAL;
 	rq.rq_disk = floppy->disk;
 
 	return ide_do_drive_cmd(drive, &rq, ide_wait);
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index fb6795236e76..3436b1f104eb 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -59,7 +59,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq,
 {
 	int ret = 1;
 
-	BUG_ON(!(rq->flags & REQ_STARTED));
+	BUG_ON(!blk_rq_started(rq));
 
 	/*
 	 * if failfast is set on a request, override number of sectors and
@@ -244,7 +244,7 @@ int ide_end_dequeued_request(ide_drive_t *drive, struct request *rq,
 
 	spin_lock_irqsave(&ide_lock, flags);
 
-	BUG_ON(!(rq->flags & REQ_STARTED));
+	BUG_ON(!blk_rq_started(rq));
 
 	/*
 	 * if failfast is set on a request, override number of sectors and
@@ -366,7 +366,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 	rq = HWGROUP(drive)->rq;
 	spin_unlock_irqrestore(&ide_lock, flags);
 
-	if (rq->flags & REQ_DRIVE_CMD) {
+	if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
 		u8 *args = (u8 *) rq->buffer;
 		if (rq->errors == 0)
 			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
@@ -376,7 +376,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			args[1] = err;
 			args[2] = hwif->INB(IDE_NSECTOR_REG);
 		}
-	} else if (rq->flags & REQ_DRIVE_TASK) {
+	} else if (rq->cmd_type == REQ_TYPE_ATA_TASK) {
 		u8 *args = (u8 *) rq->buffer;
 		if (rq->errors == 0)
 			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
@@ -390,7 +390,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			args[5] = hwif->INB(IDE_HCYL_REG);
 			args[6] = hwif->INB(IDE_SELECT_REG);
 		}
-	} else if (rq->flags & REQ_DRIVE_TASKFILE) {
+	} else if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *args = (ide_task_t *) rq->special;
 		if (rq->errors == 0)
 			rq->errors = !OK_STAT(stat,READY_STAT,BAD_STAT);
@@ -587,7 +587,7 @@ ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, u8 stat)
 		return ide_stopped;
 
 	/* retry only "normal" I/O: */
-	if (rq->flags & (REQ_DRIVE_CMD | REQ_DRIVE_TASK | REQ_DRIVE_TASKFILE)) {
+	if (!blk_fs_request(rq)) {
 		rq->errors = 1;
 		ide_end_drive_cmd(drive, stat, err);
 		return ide_stopped;
@@ -638,7 +638,7 @@ ide_startstop_t ide_abort(ide_drive_t *drive, const char *msg)
 		return ide_stopped;
 
 	/* retry only "normal" I/O: */
-	if (rq->flags & (REQ_DRIVE_CMD | REQ_DRIVE_TASK | REQ_DRIVE_TASKFILE)) {
+	if (!blk_fs_request(rq)) {
 		rq->errors = 1;
 		ide_end_drive_cmd(drive, BUSY_STAT, 0);
 		return ide_stopped;
@@ -808,7 +808,7 @@ void ide_map_sg(ide_drive_t *drive, struct request *rq)
 	if (hwif->sg_mapped)	/* needed by ide-scsi */
 		return;
 
-	if ((rq->flags & REQ_DRIVE_TASKFILE) == 0) {
+	if (rq->cmd_type != REQ_TYPE_ATA_TASKFILE) {
 		hwif->sg_nents = blk_rq_map_sg(drive->queue, rq, sg);
 	} else {
 		sg_init_one(sg, rq->buffer, rq->nr_sectors * SECTOR_SIZE);
@@ -844,7 +844,7 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 		struct request *rq)
 {
 	ide_hwif_t *hwif = HWIF(drive);
-	if (rq->flags & REQ_DRIVE_TASKFILE) {
+	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
  		ide_task_t *args = rq->special;
  
 		if (!args)
@@ -866,7 +866,7 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 		if (args->tf_out_flags.all != 0) 
 			return flagged_taskfile(drive, args);
 		return do_rw_taskfile(drive, args);
-	} else if (rq->flags & REQ_DRIVE_TASK) {
+	} else if (rq->cmd_type == REQ_TYPE_ATA_TASK) {
 		u8 *args = rq->buffer;
 		u8 sel;
  
@@ -892,7 +892,7 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
  		hwif->OUTB(sel, IDE_SELECT_REG);
  		ide_cmd(drive, args[0], args[2], &drive_cmd_intr);
  		return ide_started;
- 	} else if (rq->flags & REQ_DRIVE_CMD) {
+ 	} else if (rq->cmd_type == REQ_TYPE_ATA_CMD) {
  		u8 *args = rq->buffer;
 
 		if (!args)
@@ -980,7 +980,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 	ide_startstop_t startstop;
 	sector_t block;
 
-	BUG_ON(!(rq->flags & REQ_STARTED));
+	BUG_ON(!blk_rq_started(rq));
 
 #ifdef DEBUG
 	printk("%s: start_request: current=0x%08lx\n",
@@ -1013,9 +1013,9 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 	if (!drive->special.all) {
 		ide_driver_t *drv;
 
-		if (rq->flags & (REQ_DRIVE_CMD | REQ_DRIVE_TASK))
-			return execute_drive_cmd(drive, rq);
-		else if (rq->flags & REQ_DRIVE_TASKFILE)
+		if (rq->cmd_type == REQ_TYPE_ATA_CMD ||
+		    rq->cmd_type == REQ_TYPE_ATA_TASK ||
+		    rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
 			return execute_drive_cmd(drive, rq);
 		else if (blk_pm_request(rq)) {
 			struct request_pm_state *pm = rq->end_io_data;
@@ -1264,7 +1264,7 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 		 * We count how many times we loop here to make sure we service
 		 * all drives in the hwgroup without looping for ever
 		 */
-		if (drive->blocked && !blk_pm_request(rq) && !(rq->flags & REQ_PREEMPT)) {
+		if (drive->blocked && !blk_pm_request(rq) && !(rq->cmd_flags & REQ_PREEMPT)) {
 			drive = drive->next ? drive->next : hwgroup->drive;
 			if (loops++ < 4 && !blk_queue_plugged(drive->queue))
 				goto again;
@@ -1670,7 +1670,7 @@ irqreturn_t ide_intr (int irq, void *dev_id, struct pt_regs *regs)
 void ide_init_drive_cmd (struct request *rq)
 {
 	memset(rq, 0, sizeof(*rq));
-	rq->flags = REQ_DRIVE_CMD;
+	rq->cmd_type = REQ_TYPE_ATA_CMD;
 	rq->ref_count = 1;
 }
 
@@ -1727,7 +1727,7 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 		hwgroup->rq = NULL;
 	if (action == ide_preempt || action == ide_head_wait) {
 		where = ELEVATOR_INSERT_FRONT;
-		rq->flags |= REQ_PREEMPT;
+		rq->cmd_flags |= REQ_PREEMPT;
 	}
 	__elv_add_request(drive->queue, rq, where, 0);
 	ide_do_request(hwgroup, IDE_NO_IRQ);
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 1feff23487d4..850ef63cc986 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -456,13 +456,14 @@ static void ide_dump_opcode(ide_drive_t *drive)
 	spin_unlock(&ide_lock);
 	if (!rq)
 		return;
-	if (rq->flags & (REQ_DRIVE_CMD | REQ_DRIVE_TASK)) {
+	if (rq->cmd_type == REQ_TYPE_ATA_CMD ||
+	    rq->cmd_type == REQ_TYPE_ATA_TASK) {
 		char *args = rq->buffer;
 		if (args) {
 			opcode = args[0];
 			found = 1;
 		}
-	} else if (rq->flags & REQ_DRIVE_TASKFILE) {
+	} else if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *args = rq->special;
 		if (args) {
 			task_struct_t *tf = (task_struct_t *) args->tfRegister;
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 7067ab997927..643e4b9ac651 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1776,7 +1776,7 @@ static void idetape_create_request_sense_cmd (idetape_pc_t *pc)
 static void idetape_init_rq(struct request *rq, u8 cmd)
 {
 	memset(rq, 0, sizeof(*rq));
-	rq->flags = REQ_SPECIAL;
+	rq->cmd_type = REQ_TYPE_SPECIAL;
 	rq->cmd[0] = cmd;
 }
 
@@ -2433,12 +2433,12 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 			rq->sector, rq->nr_sectors, rq->current_nr_sectors);
 #endif /* IDETAPE_DEBUG_LOG */
 
-	if ((rq->flags & REQ_SPECIAL) == 0) {
+	if (!blk_special_request(rq)) {
 		/*
 		 * We do not support buffer cache originated requests.
 		 */
 		printk(KERN_NOTICE "ide-tape: %s: Unsupported request in "
-			"request queue (%ld)\n", drive->name, rq->flags);
+			"request queue (%d)\n", drive->name, rq->cmd_type);
 		ide_end_request(drive, 0, 0);
 		return ide_stopped;
 	}
@@ -2768,7 +2768,7 @@ static void idetape_wait_for_request (ide_drive_t *drive, struct request *rq)
 	idetape_tape_t *tape = drive->driver_data;
 
 #if IDETAPE_DEBUG_BUGS
-	if (rq == NULL || (rq->flags & REQ_SPECIAL) == 0) {
+	if (rq == NULL || !blk_special_request(rq)) {
 		printk (KERN_ERR "ide-tape: bug: Trying to sleep on non-valid request\n");
 		return;
 	}
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 97a9244312fc..1d0470c1f957 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -363,7 +363,7 @@ static ide_startstop_t task_error(ide_drive_t *drive, struct request *rq,
 
 static void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
 {
-	if (rq->flags & REQ_DRIVE_TASKFILE) {
+	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *task = rq->special;
 
 		if (task->tf_out_flags.all) {
@@ -474,7 +474,7 @@ static int ide_diag_taskfile(ide_drive_t *drive, ide_task_t *args, unsigned long
 	struct request rq;
 
 	memset(&rq, 0, sizeof(rq));
-	rq.flags = REQ_DRIVE_TASKFILE;
+	rq.cmd_type = REQ_TYPE_ATA_TASKFILE;
 	rq.buffer = buf;
 
 	/*
@@ -499,7 +499,7 @@ static int ide_diag_taskfile(ide_drive_t *drive, ide_task_t *args, unsigned long
 		rq.hard_cur_sectors = rq.current_nr_sectors = rq.nr_sectors;
 
 		if (args->command_type == IDE_DRIVE_TASK_RAW_WRITE)
-			rq.flags |= REQ_RW;
+			rq.cmd_flags |= REQ_RW;
 	}
 
 	rq.special = args;
@@ -737,7 +737,7 @@ static int ide_wait_cmd_task(ide_drive_t *drive, u8 *buf)
 	struct request rq;
 
 	ide_init_drive_cmd(&rq);
-	rq.flags = REQ_DRIVE_TASK;
+	rq.cmd_type = REQ_TYPE_ATA_TASK;
 	rq.buffer = buf;
 	return ide_do_drive_cmd(drive, &rq, ide_wait);
 }
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 9c8468de1a75..9384a3fdde6c 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -1217,7 +1217,7 @@ static int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 	memset(&rq, 0, sizeof(rq));
 	memset(&rqpm, 0, sizeof(rqpm));
 	memset(&args, 0, sizeof(args));
-	rq.flags = REQ_PM_SUSPEND;
+	rq.cmd_type = REQ_TYPE_PM_SUSPEND;
 	rq.special = &args;
 	rq.end_io_data = &rqpm;
 	rqpm.pm_step = ide_pm_state_start_suspend;
@@ -1238,7 +1238,7 @@ static int generic_ide_resume(struct device *dev)
 	memset(&rq, 0, sizeof(rq));
 	memset(&rqpm, 0, sizeof(rqpm));
 	memset(&args, 0, sizeof(args));
-	rq.flags = REQ_PM_RESUME;
+	rq.cmd_type = REQ_TYPE_PM_RESUME;
 	rq.special = &args;
 	rq.end_io_data = &rqpm;
 	rqpm.pm_step = ide_pm_state_start_resume;
diff --git a/drivers/ide/legacy/hd.c b/drivers/ide/legacy/hd.c
index aebecd8f51cc..4ab931145673 100644
--- a/drivers/ide/legacy/hd.c
+++ b/drivers/ide/legacy/hd.c
@@ -626,7 +626,7 @@ repeat:
 		req->rq_disk->disk_name, (req->cmd == READ)?"read":"writ",
 		cyl, head, sec, nsect, req->buffer);
 #endif
-	if (req->flags & REQ_CMD) {
+	if (blk_fs_request(req)) {
 		switch (rq_data_dir(req)) {
 		case READ:
 			hd_out(disk,nsect,sec,head,cyl,WIN_READ,&read_intr);
diff --git a/drivers/md/dm-emc.c b/drivers/md/dm-emc.c
index 2a374ccb30dd..2b2d45d7baaa 100644
--- a/drivers/md/dm-emc.c
+++ b/drivers/md/dm-emc.c
@@ -126,7 +126,8 @@ static struct request *get_failover_req(struct emc_handler *h,
 	memset(&rq->cmd, 0, BLK_MAX_CDB);
 
 	rq->timeout = EMC_FAILOVER_TIMEOUT;
-	rq->flags |= (REQ_BLOCK_PC | REQ_FAILFAST | REQ_NOMERGE);
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
+	rq->cmd_flags |= REQ_FAILFAST | REQ_NOMERGE;
 
 	return rq;
 }
diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c
index 1ddc2fb429d5..eaba81bf2eca 100644
--- a/drivers/message/i2o/i2o_block.c
+++ b/drivers/message/i2o/i2o_block.c
@@ -390,9 +390,9 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req)
 	}
 
 	/* request is already processed by us, so return */
-	if (req->flags & REQ_SPECIAL) {
+	if (blk_special_request(req)) {
 		osm_debug("REQ_SPECIAL already set!\n");
-		req->flags |= REQ_DONTPREP;
+		req->cmd_flags |= REQ_DONTPREP;
 		return BLKPREP_OK;
 	}
 
@@ -411,7 +411,8 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req)
 		ireq = req->special;
 
 	/* do not come back here */
-	req->flags |= REQ_DONTPREP | REQ_SPECIAL;
+	req->cmd_type = REQ_TYPE_SPECIAL;
+	req->cmd_flags |= REQ_DONTPREP;
 
 	return BLKPREP_OK;
 };
diff --git a/drivers/mmc/mmc_queue.c b/drivers/mmc/mmc_queue.c
index 74f8cdeeff0f..4ccdd82b680f 100644
--- a/drivers/mmc/mmc_queue.c
+++ b/drivers/mmc/mmc_queue.c
@@ -28,7 +28,7 @@ static int mmc_prep_request(struct request_queue *q, struct request *req)
 	struct mmc_queue *mq = q->queuedata;
 	int ret = BLKPREP_KILL;
 
-	if (req->flags & REQ_SPECIAL) {
+	if (blk_special_request(req)) {
 		/*
 		 * Special commands already have the command
 		 * blocks already setup in req->special.
@@ -36,7 +36,7 @@ static int mmc_prep_request(struct request_queue *q, struct request *req)
 		BUG_ON(!req->special);
 
 		ret = BLKPREP_OK;
-	} else if (req->flags & (REQ_CMD | REQ_BLOCK_PC)) {
+	} else if (blk_fs_request(req) || blk_pc_request(req)) {
 		/*
 		 * Block I/O requests need translating according
 		 * to the protocol.
@@ -50,7 +50,7 @@ static int mmc_prep_request(struct request_queue *q, struct request *req)
 	}
 
 	if (ret == BLKPREP_OK)
-		req->flags |= REQ_DONTPREP;
+		req->cmd_flags |= REQ_DONTPREP;
 
 	return ret;
 }
diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
index 458d3c8ae1ee..6baf5fe14230 100644
--- a/drivers/mtd/mtd_blkdevs.c
+++ b/drivers/mtd/mtd_blkdevs.c
@@ -46,7 +46,7 @@ static int do_blktrans_request(struct mtd_blktrans_ops *tr,
 	nsect = req->current_nr_sectors;
 	buf = req->buffer;
 
-	if (!(req->flags & REQ_CMD))
+	if (!blk_fs_request(req))
 		return 0;
 
 	if (block + nsect > get_capacity(req->rq_disk))
diff --git a/drivers/s390/block/dasd_diag.c b/drivers/s390/block/dasd_diag.c
index 9d051e5687ea..222a8a71a5e8 100644
--- a/drivers/s390/block/dasd_diag.c
+++ b/drivers/s390/block/dasd_diag.c
@@ -529,7 +529,7 @@ dasd_diag_build_cp(struct dasd_device * device, struct request *req)
 	}
 	cqr->retries = DIAG_MAX_RETRIES;
 	cqr->buildclk = get_clock();
-	if (req->flags & REQ_FAILFAST)
+	if (req->cmd_flags & REQ_FAILFAST)
 		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
 	cqr->device = device;
 	cqr->expires = DIAG_TIMEOUT;
diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c
index b7a7fac3f7c3..5ecea3e4fdef 100644
--- a/drivers/s390/block/dasd_eckd.c
+++ b/drivers/s390/block/dasd_eckd.c
@@ -1266,7 +1266,7 @@ dasd_eckd_build_cp(struct dasd_device * device, struct request *req)
 			recid++;
 		}
 	}
-	if (req->flags & REQ_FAILFAST)
+	if (req->cmd_flags & REQ_FAILFAST)
 		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
 	cqr->device = device;
 	cqr->expires = 5 * 60 * HZ;	/* 5 minutes */
diff --git a/drivers/s390/block/dasd_fba.c b/drivers/s390/block/dasd_fba.c
index e85015be109b..80926c548228 100644
--- a/drivers/s390/block/dasd_fba.c
+++ b/drivers/s390/block/dasd_fba.c
@@ -344,7 +344,7 @@ dasd_fba_build_cp(struct dasd_device * device, struct request *req)
 			recid++;
 		}
 	}
-	if (req->flags & REQ_FAILFAST)
+	if (req->cmd_flags & REQ_FAILFAST)
 		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
 	cqr->device = device;
 	cqr->expires = 5 * 60 * HZ;	/* 5 minutes */
diff --git a/drivers/scsi/aic7xxx_old.c b/drivers/scsi/aic7xxx_old.c
index 5dcef48d414f..10353379a074 100644
--- a/drivers/scsi/aic7xxx_old.c
+++ b/drivers/scsi/aic7xxx_old.c
@@ -2862,7 +2862,7 @@ aic7xxx_done(struct aic7xxx_host *p, struct aic7xxx_scb *scb)
       aic_dev->r_total++;
       ptr = aic_dev->r_bins;
     }
-    if(cmd->device->simple_tags && cmd->request->flags & REQ_HARDBARRIER)
+    if(cmd->device->simple_tags && cmd->request->cmd_flags & REQ_HARDBARRIER)
     {
       aic_dev->barrier_total++;
       if(scb->tag_action == MSG_ORDERED_Q_TAG)
@@ -10158,7 +10158,7 @@ aic7xxx_buildscb(struct aic7xxx_host *p, Scsi_Cmnd *cmd,
     /* We always force TEST_UNIT_READY to untagged */
     if (cmd->cmnd[0] != TEST_UNIT_READY && sdptr->simple_tags)
     {
-      if (req->flags & REQ_HARDBARRIER)
+      if (req->cmd_flags & REQ_HARDBARRIER)
       {
 	if(sdptr->ordered_tags)
 	{
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 94d1de55607f..65b19695ebe2 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -344,7 +344,7 @@ static int idescsi_check_condition(ide_drive_t *drive, struct request *failed_co
 	pc->buffer = buf;
 	pc->c[0] = REQUEST_SENSE;
 	pc->c[4] = pc->request_transfer = pc->buffer_size = SCSI_SENSE_BUFFERSIZE;
-	rq->flags = REQ_SENSE;
+	rq->cmd_type = REQ_TYPE_SENSE;
 	pc->timeout = jiffies + WAIT_READY;
 	/* NOTE! Save the failed packet command in "rq->buffer" */
 	rq->buffer = (void *) failed_command->special;
@@ -398,12 +398,12 @@ static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs)
 	int errors = rq->errors;
 	unsigned long flags;
 
-	if (!(rq->flags & (REQ_SPECIAL|REQ_SENSE))) {
+	if (!blk_special_request(rq) && !blk_sense_request(rq)) {
 		ide_end_request(drive, uptodate, nrsecs);
 		return 0;
 	}
 	ide_end_drive_cmd (drive, 0, 0);
-	if (rq->flags & REQ_SENSE) {
+	if (blk_sense_request(rq)) {
 		idescsi_pc_t *opc = (idescsi_pc_t *) rq->buffer;
 		if (log) {
 			printk ("ide-scsi: %s: wrap up check %lu, rst = ", drive->name, opc->scsi_cmd->serial_number);
@@ -712,7 +712,7 @@ static ide_startstop_t idescsi_do_request (ide_drive_t *drive, struct request *r
 	printk (KERN_INFO "sector: %ld, nr_sectors: %ld, current_nr_sectors: %d\n",rq->sector,rq->nr_sectors,rq->current_nr_sectors);
 #endif /* IDESCSI_DEBUG_LOG */
 
-	if (rq->flags & (REQ_SPECIAL|REQ_SENSE)) {
+	if (blk_sense_request(rq) || blk_special_request(rq)) {
 		return idescsi_issue_pc (drive, (idescsi_pc_t *) rq->special);
 	}
 	blk_dump_rq_flags(rq, "ide-scsi: unsup command");
@@ -938,7 +938,7 @@ static int idescsi_queue (struct scsi_cmnd *cmd,
 
 	ide_init_drive_cmd (rq);
 	rq->special = (char *) pc;
-	rq->flags = REQ_SPECIAL;
+	rq->cmd_type = REQ_TYPE_SPECIAL;
 	spin_unlock_irq(host->host_lock);
 	rq->rq_disk = scsi->disk;
 	(void) ide_do_drive_cmd (drive, rq, ide_end);
@@ -992,7 +992,7 @@ static int idescsi_eh_abort (struct scsi_cmnd *cmd)
 		 */
 		printk (KERN_ERR "ide-scsi: cmd aborted!\n");
 
-		if (scsi->pc->rq->flags & REQ_SENSE)
+		if (blk_sense_request(scsi->pc->rq))
 			kfree(scsi->pc->buffer);
 		kfree(scsi->pc->rq);
 		kfree(scsi->pc);
@@ -1042,7 +1042,7 @@ static int idescsi_eh_reset (struct scsi_cmnd *cmd)
 	/* kill current request */
 	blkdev_dequeue_request(req);
 	end_that_request_last(req, 0);
-	if (req->flags & REQ_SENSE)
+	if (blk_sense_request(req))
 		kfree(scsi->pc->buffer);
 	kfree(scsi->pc);
 	scsi->pc = NULL;
diff --git a/drivers/scsi/pluto.c b/drivers/scsi/pluto.c
index 0bd9c60e6455..aa60a5f1fbc3 100644
--- a/drivers/scsi/pluto.c
+++ b/drivers/scsi/pluto.c
@@ -67,7 +67,6 @@ static void __init pluto_detect_done(Scsi_Cmnd *SCpnt)
 
 static void __init pluto_detect_scsi_done(Scsi_Cmnd *SCpnt)
 {
-	SCpnt->request->rq_status = RQ_SCSI_DONE;
 	PLND(("Detect done %08lx\n", (long)SCpnt))
 	if (atomic_dec_and_test (&fcss))
 		up(&fc_sem);
@@ -166,7 +165,7 @@ int __init pluto_detect(struct scsi_host_template *tpnt)
 		
 		SCpnt->cmd_len = COMMAND_SIZE(INQUIRY);
 	
-		SCpnt->request->rq_status = RQ_SCSI_BUSY;
+		SCpnt->request->cmd_flags &= ~REQ_STARTED;
 		
 		SCpnt->done = pluto_detect_done;
 		SCpnt->request_bufflen = 256;
@@ -178,7 +177,8 @@ int __init pluto_detect(struct scsi_host_template *tpnt)
 	for (retry = 0; retry < 5; retry++) {
 		for (i = 0; i < fcscount; i++) {
 			if (!fcs[i].fc) break;
-			if (fcs[i].cmd.request->rq_status != RQ_SCSI_DONE) {
+			if (!(fcs[i].cmd.request->cmd_flags & REQ_STARTED)) {
+				fcs[i].cmd.request->cmd_flags |= REQ_STARTED;
 				disable_irq(fcs[i].fc->irq);
 				PLND(("queuecommand %d %d\n", retry, i))
 				fcp_scsi_queuecommand (&(fcs[i].cmd), 
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index d6743b959a72..71084728eb42 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -82,7 +82,7 @@ static void scsi_unprep_request(struct request *req)
 {
 	struct scsi_cmnd *cmd = req->special;
 
-	req->flags &= ~REQ_DONTPREP;
+	req->cmd_flags &= ~REQ_DONTPREP;
 	req->special = NULL;
 
 	scsi_put_command(cmd);
@@ -196,7 +196,8 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	req->sense_len = 0;
 	req->retries = retries;
 	req->timeout = timeout;
-	req->flags |= flags | REQ_BLOCK_PC | REQ_SPECIAL | REQ_QUIET;
+	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	req->cmd_flags |= flags | REQ_QUIET | REQ_PREEMPT;
 
 	/*
 	 * head injection *required* here otherwise quiesce won't work
@@ -397,7 +398,8 @@ int scsi_execute_async(struct scsi_device *sdev, const unsigned char *cmd,
 	req = blk_get_request(sdev->request_queue, write, gfp);
 	if (!req)
 		goto free_sense;
-	req->flags |= REQ_BLOCK_PC | REQ_QUIET;
+	req->cmd_type = REQ_TYPE_BLOCK_PC;
+	req->cmd_flags |= REQ_QUIET;
 
 	if (use_sg)
 		err = scsi_req_map_sg(req, buffer, use_sg, bufflen, gfp);
@@ -933,7 +935,7 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 					break;
 				}
 			}
-			if (!(req->flags & REQ_QUIET)) {
+			if (!(req->cmd_flags & REQ_QUIET)) {
 				scmd_printk(KERN_INFO, cmd,
 					    "Device not ready: ");
 				scsi_print_sense_hdr("", &sshdr);
@@ -941,7 +943,7 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 			scsi_end_request(cmd, 0, this_count, 1);
 			return;
 		case VOLUME_OVERFLOW:
-			if (!(req->flags & REQ_QUIET)) {
+			if (!(req->cmd_flags & REQ_QUIET)) {
 				scmd_printk(KERN_INFO, cmd,
 					    "Volume overflow, CDB: ");
 				__scsi_print_command(cmd->cmnd);
@@ -963,7 +965,7 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
 		return;
 	}
 	if (result) {
-		if (!(req->flags & REQ_QUIET)) {
+		if (!(req->cmd_flags & REQ_QUIET)) {
 			scmd_printk(KERN_INFO, cmd,
 				    "SCSI error: return code = 0x%08x\n",
 				    result);
@@ -995,7 +997,7 @@ static int scsi_init_io(struct scsi_cmnd *cmd)
 	/*
 	 * if this is a rq->data based REQ_BLOCK_PC, setup for a non-sg xfer
 	 */
-	if ((req->flags & REQ_BLOCK_PC) && !req->bio) {
+	if (blk_pc_request(req) && !req->bio) {
 		cmd->request_bufflen = req->data_len;
 		cmd->request_buffer = req->data;
 		req->buffer = req->data;
@@ -1139,13 +1141,12 @@ static int scsi_prep_fn(struct request_queue *q, struct request *req)
 	 * these two cases differently.  We differentiate by looking
 	 * at request->cmd, as this tells us the real story.
 	 */
-	if (req->flags & REQ_SPECIAL && req->special) {
+	if (blk_special_request(req) && req->special)
 		cmd = req->special;
-	} else if (req->flags & (REQ_CMD | REQ_BLOCK_PC)) {
-
-		if(unlikely(specials_only) && !(req->flags & REQ_SPECIAL)) {
-			if(specials_only == SDEV_QUIESCE ||
-					specials_only == SDEV_BLOCK)
+	else if (blk_pc_request(req) || blk_fs_request(req)) {
+		if (unlikely(specials_only) && !(req->cmd_flags & REQ_PREEMPT)){
+			if (specials_only == SDEV_QUIESCE ||
+			    specials_only == SDEV_BLOCK)
 				goto defer;
 			
 			sdev_printk(KERN_ERR, sdev,
@@ -1153,7 +1154,6 @@ static int scsi_prep_fn(struct request_queue *q, struct request *req)
 			goto kill;
 		}
 			
-			
 		/*
 		 * Now try and find a command block that we can use.
 		 */
@@ -1184,7 +1184,7 @@ static int scsi_prep_fn(struct request_queue *q, struct request *req)
 	 * lock.  We hope REQ_STARTED prevents anything untoward from
 	 * happening now.
 	 */
-	if (req->flags & (REQ_CMD | REQ_BLOCK_PC)) {
+	if (blk_fs_request(req) || blk_pc_request(req)) {
 		int ret;
 
 		/*
@@ -1216,7 +1216,7 @@ static int scsi_prep_fn(struct request_queue *q, struct request *req)
 		/*
 		 * Initialize the actual SCSI command for this request.
 		 */
-		if (req->flags & REQ_BLOCK_PC) {
+		if (blk_pc_request(req)) {
 			scsi_setup_blk_pc_cmnd(cmd);
 		} else if (req->rq_disk) {
 			struct scsi_driver *drv;
@@ -1233,7 +1233,7 @@ static int scsi_prep_fn(struct request_queue *q, struct request *req)
 	/*
 	 * The request is now prepped, no need to come back here
 	 */
-	req->flags |= REQ_DONTPREP;
+	req->cmd_flags |= REQ_DONTPREP;
 	return BLKPREP_OK;
 
  defer:
@@ -1454,8 +1454,9 @@ static void scsi_request_fn(struct request_queue *q)
 		if (unlikely(cmd == NULL)) {
 			printk(KERN_CRIT "impossible request in %s.\n"
 					 "please mail a stack trace to "
-					 "linux-scsi@vger.kernel.org",
+					 "linux-scsi@vger.kernel.org\n",
 					 __FUNCTION__);
+			blk_dump_rq_flags(req, "foo");
 			BUG();
 		}
 		spin_lock(shost->host_lock);
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 638cff41d436..10bc99c911fa 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -443,8 +443,7 @@ static int sd_init_command(struct scsi_cmnd * SCpnt)
 		SCpnt->cmnd[0] = READ_6;
 		SCpnt->sc_data_direction = DMA_FROM_DEVICE;
 	} else {
-		printk(KERN_ERR "sd: Unknown command %lx\n", rq->flags);
-/* overkill 	panic("Unknown sd command %lx\n", rq->flags); */
+		printk(KERN_ERR "sd: Unknown command %x\n", rq->cmd_flags);
 		return 0;
 	}
 
@@ -840,7 +839,7 @@ static int sd_issue_flush(struct device *dev, sector_t *error_sector)
 static void sd_prepare_flush(request_queue_t *q, struct request *rq)
 {
 	memset(rq->cmd, 0, sizeof(rq->cmd));
-	rq->flags |= REQ_BLOCK_PC;
+	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->timeout = SD_TIMEOUT;
 	rq->cmd[0] = SYNCHRONIZE_CACHE;
 	rq->cmd_len = 10;
diff --git a/drivers/scsi/sun3_NCR5380.c b/drivers/scsi/sun3_NCR5380.c
index 2f8073b73bf3..7f9bcef6adfa 100644
--- a/drivers/scsi/sun3_NCR5380.c
+++ b/drivers/scsi/sun3_NCR5380.c
@@ -2017,7 +2017,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance)
 		if((count > SUN3_DMA_MINSIZE) && (sun3_dma_setup_done
 						  != cmd))
 		{
-			if(cmd->request->flags & REQ_CMD) {
+			if(blk_fs_request(cmd->request)) {
 				sun3scsi_dma_setup(d, count,
 						   rq_data_dir(cmd->request));
 				sun3_dma_setup_done = cmd;
diff --git a/drivers/scsi/sun3_scsi.c b/drivers/scsi/sun3_scsi.c
index 837173415d4c..44a99aeb8180 100644
--- a/drivers/scsi/sun3_scsi.c
+++ b/drivers/scsi/sun3_scsi.c
@@ -524,7 +524,7 @@ static inline unsigned long sun3scsi_dma_residual(struct Scsi_Host *instance)
 static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted, Scsi_Cmnd *cmd,
 				    int write_flag)
 {
-	if(cmd->request->flags & REQ_CMD)
+	if(blk_fs_request(cmd->request))
  		return wanted;
 	else
 		return 0;
diff --git a/drivers/scsi/sun3_scsi_vme.c b/drivers/scsi/sun3_scsi_vme.c
index 008a82ab8521..f5742b84b27a 100644
--- a/drivers/scsi/sun3_scsi_vme.c
+++ b/drivers/scsi/sun3_scsi_vme.c
@@ -458,7 +458,7 @@ static inline unsigned long sun3scsi_dma_residual(struct Scsi_Host *instance)
 static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted, Scsi_Cmnd *cmd,
 				    int write_flag)
 {
-	if(cmd->request->flags & REQ_CMD)
+	if(blk_fs_request(cmd->request))
  		return wanted;
 	else
 		return 0;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index cfde8b3ee919..b2a412cf468f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -120,6 +120,86 @@ struct request_list {
 	wait_queue_head_t wait[2];
 };
 
+/*
+ * request command types
+ */
+enum rq_cmd_type_bits {
+	REQ_TYPE_FS		= 1,	/* fs request */
+	REQ_TYPE_BLOCK_PC,		/* scsi command */
+	REQ_TYPE_SENSE,			/* sense request */
+	REQ_TYPE_PM_SUSPEND,		/* suspend request */
+	REQ_TYPE_PM_RESUME,		/* resume request */
+	REQ_TYPE_PM_SHUTDOWN,		/* shutdown request */
+	REQ_TYPE_FLUSH,			/* flush request */
+	REQ_TYPE_SPECIAL,		/* driver defined type */
+	REQ_TYPE_LINUX_BLOCK,		/* generic block layer message */
+	/*
+	 * for ATA/ATAPI devices. this really doesn't belong here, ide should
+	 * use REQ_TYPE_SPECIAL and use rq->cmd[0] with the range of driver
+	 * private REQ_LB opcodes to differentiate what type of request this is
+	 */
+	REQ_TYPE_ATA_CMD,
+	REQ_TYPE_ATA_TASK,
+	REQ_TYPE_ATA_TASKFILE,
+};
+
+/*
+ * For request of type REQ_TYPE_LINUX_BLOCK, rq->cmd[0] is the opcode being
+ * sent down (similar to how REQ_TYPE_BLOCK_PC means that ->cmd[] holds a
+ * SCSI cdb.
+ *
+ * 0x00 -> 0x3f are driver private, to be used for whatever purpose they need,
+ * typically to differentiate REQ_TYPE_SPECIAL requests.
+ *
+ */
+enum {
+	/*
+	 * just examples for now
+	 */
+	REQ_LB_OP_EJECT	= 0x40,		/* eject request */
+	REQ_LB_OP_FLUSH = 0x41,		/* flush device */
+};
+
+/*
+ * request type modified bits. first three bits match BIO_RW* bits, important
+ */
+enum rq_flag_bits {
+	__REQ_RW,		/* not set, read. set, write */
+	__REQ_FAILFAST,		/* no low level driver retries */
+	__REQ_SORTED,		/* elevator knows about this request */
+	__REQ_SOFTBARRIER,	/* may not be passed by ioscheduler */
+	__REQ_HARDBARRIER,	/* may not be passed by drive either */
+	__REQ_FUA,		/* forced unit access */
+	__REQ_NOMERGE,		/* don't touch this for merging */
+	__REQ_STARTED,		/* drive already may have started this one */
+	__REQ_DONTPREP,		/* don't call prep for this one */
+	__REQ_QUEUED,		/* uses queueing */
+	__REQ_ELVPRIV,		/* elevator private data attached */
+	__REQ_FAILED,		/* set if the request failed */
+	__REQ_QUIET,		/* don't worry about errors */
+	__REQ_PREEMPT,		/* set for "ide_preempt" requests */
+	__REQ_ORDERED_COLOR,	/* is before or after barrier */
+	__REQ_RW_SYNC,		/* request is sync (O_DIRECT) */
+	__REQ_NR_BITS,		/* stops here */
+};
+
+#define REQ_RW		(1 << __REQ_RW)
+#define REQ_FAILFAST	(1 << __REQ_FAILFAST)
+#define REQ_SORTED	(1 << __REQ_SORTED)
+#define REQ_SOFTBARRIER	(1 << __REQ_SOFTBARRIER)
+#define REQ_HARDBARRIER	(1 << __REQ_HARDBARRIER)
+#define REQ_FUA		(1 << __REQ_FUA)
+#define REQ_NOMERGE	(1 << __REQ_NOMERGE)
+#define REQ_STARTED	(1 << __REQ_STARTED)
+#define REQ_DONTPREP	(1 << __REQ_DONTPREP)
+#define REQ_QUEUED	(1 << __REQ_QUEUED)
+#define REQ_ELVPRIV	(1 << __REQ_ELVPRIV)
+#define REQ_FAILED	(1 << __REQ_FAILED)
+#define REQ_QUIET	(1 << __REQ_QUIET)
+#define REQ_PREEMPT	(1 << __REQ_PREEMPT)
+#define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
+#define REQ_RW_SYNC	(1 << __REQ_RW_SYNC)
+
 #define BLK_MAX_CDB	16
 
 /*
@@ -129,7 +209,8 @@ struct request {
 	struct list_head queuelist;
 	struct list_head donelist;
 
-	unsigned long flags;		/* see REQ_ bits below */
+	unsigned int cmd_flags;
+	enum rq_cmd_type_bits cmd_type;
 
 	/* Maintain bio traversal state for part by part I/O submission.
 	 * hard_* are block layer internals, no driver should touch them!
@@ -202,73 +283,7 @@ struct request {
 };
 
 /*
- * first three bits match BIO_RW* bits, important
- */
-enum rq_flag_bits {
-	__REQ_RW,		/* not set, read. set, write */
-	__REQ_FAILFAST,		/* no low level driver retries */
-	__REQ_SORTED,		/* elevator knows about this request */
-	__REQ_SOFTBARRIER,	/* may not be passed by ioscheduler */
-	__REQ_HARDBARRIER,	/* may not be passed by drive either */
-	__REQ_FUA,		/* forced unit access */
-	__REQ_CMD,		/* is a regular fs rw request */
-	__REQ_NOMERGE,		/* don't touch this for merging */
-	__REQ_STARTED,		/* drive already may have started this one */
-	__REQ_DONTPREP,		/* don't call prep for this one */
-	__REQ_QUEUED,		/* uses queueing */
-	__REQ_ELVPRIV,		/* elevator private data attached */
-	/*
-	 * for ATA/ATAPI devices
-	 */
-	__REQ_PC,		/* packet command (special) */
-	__REQ_BLOCK_PC,		/* queued down pc from block layer */
-	__REQ_SENSE,		/* sense retrival */
-
-	__REQ_FAILED,		/* set if the request failed */
-	__REQ_QUIET,		/* don't worry about errors */
-	__REQ_SPECIAL,		/* driver suplied command */
-	__REQ_DRIVE_CMD,
-	__REQ_DRIVE_TASK,
-	__REQ_DRIVE_TASKFILE,
-	__REQ_PREEMPT,		/* set for "ide_preempt" requests */
-	__REQ_PM_SUSPEND,	/* suspend request */
-	__REQ_PM_RESUME,	/* resume request */
-	__REQ_PM_SHUTDOWN,	/* shutdown request */
-	__REQ_ORDERED_COLOR,	/* is before or after barrier */
-	__REQ_RW_SYNC,		/* request is sync (O_DIRECT) */
-	__REQ_NR_BITS,		/* stops here */
-};
-
-#define REQ_RW		(1 << __REQ_RW)
-#define REQ_FAILFAST	(1 << __REQ_FAILFAST)
-#define REQ_SORTED	(1 << __REQ_SORTED)
-#define REQ_SOFTBARRIER	(1 << __REQ_SOFTBARRIER)
-#define REQ_HARDBARRIER	(1 << __REQ_HARDBARRIER)
-#define REQ_FUA		(1 << __REQ_FUA)
-#define REQ_CMD		(1 << __REQ_CMD)
-#define REQ_NOMERGE	(1 << __REQ_NOMERGE)
-#define REQ_STARTED	(1 << __REQ_STARTED)
-#define REQ_DONTPREP	(1 << __REQ_DONTPREP)
-#define REQ_QUEUED	(1 << __REQ_QUEUED)
-#define REQ_ELVPRIV	(1 << __REQ_ELVPRIV)
-#define REQ_PC		(1 << __REQ_PC)
-#define REQ_BLOCK_PC	(1 << __REQ_BLOCK_PC)
-#define REQ_SENSE	(1 << __REQ_SENSE)
-#define REQ_FAILED	(1 << __REQ_FAILED)
-#define REQ_QUIET	(1 << __REQ_QUIET)
-#define REQ_SPECIAL	(1 << __REQ_SPECIAL)
-#define REQ_DRIVE_CMD	(1 << __REQ_DRIVE_CMD)
-#define REQ_DRIVE_TASK	(1 << __REQ_DRIVE_TASK)
-#define REQ_DRIVE_TASKFILE	(1 << __REQ_DRIVE_TASKFILE)
-#define REQ_PREEMPT	(1 << __REQ_PREEMPT)
-#define REQ_PM_SUSPEND	(1 << __REQ_PM_SUSPEND)
-#define REQ_PM_RESUME	(1 << __REQ_PM_RESUME)
-#define REQ_PM_SHUTDOWN	(1 << __REQ_PM_SHUTDOWN)
-#define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
-#define REQ_RW_SYNC	(1 << __REQ_RW_SYNC)
-
-/*
- * State information carried for REQ_PM_SUSPEND and REQ_PM_RESUME
+ * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
  * requests. Some step values could eventually be made generic.
  */
 struct request_pm_state
@@ -490,25 +505,28 @@ enum {
 #define blk_queue_stopped(q)	test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags)
 #define blk_queue_flushing(q)	((q)->ordseq)
 
-#define blk_fs_request(rq)	((rq)->flags & REQ_CMD)
-#define blk_pc_request(rq)	((rq)->flags & REQ_BLOCK_PC)
-#define blk_noretry_request(rq)	((rq)->flags & REQ_FAILFAST)
-#define blk_rq_started(rq)	((rq)->flags & REQ_STARTED)
+#define blk_fs_request(rq)	((rq)->cmd_type == REQ_TYPE_FS)
+#define blk_pc_request(rq)	((rq)->cmd_type == REQ_TYPE_BLOCK_PC)
+#define blk_special_request(rq)	((rq)->cmd_type == REQ_TYPE_SPECIAL)
+#define blk_sense_request(rq)	((rq)->cmd_type == REQ_TYPE_SENSE)
+
+#define blk_noretry_request(rq)	((rq)->cmd_flags & REQ_FAILFAST)
+#define blk_rq_started(rq)	((rq)->cmd_flags & REQ_STARTED)
 
 #define blk_account_rq(rq)	(blk_rq_started(rq) && blk_fs_request(rq))
 
-#define blk_pm_suspend_request(rq)	((rq)->flags & REQ_PM_SUSPEND)
-#define blk_pm_resume_request(rq)	((rq)->flags & REQ_PM_RESUME)
+#define blk_pm_suspend_request(rq)	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND)
+#define blk_pm_resume_request(rq)	((rq)->cmd_type == REQ_TYPE_PM_RESUME)
 #define blk_pm_request(rq)	\
-	((rq)->flags & (REQ_PM_SUSPEND | REQ_PM_RESUME))
+	(blk_pm_suspend_request(rq) || blk_pm_resume_request(rq))
 
-#define blk_sorted_rq(rq)	((rq)->flags & REQ_SORTED)
-#define blk_barrier_rq(rq)	((rq)->flags & REQ_HARDBARRIER)
-#define blk_fua_rq(rq)		((rq)->flags & REQ_FUA)
+#define blk_sorted_rq(rq)	((rq)->cmd_flags & REQ_SORTED)
+#define blk_barrier_rq(rq)	((rq)->cmd_flags & REQ_HARDBARRIER)
+#define blk_fua_rq(rq)		((rq)->cmd_flags & REQ_FUA)
 
 #define list_entry_rq(ptr)	list_entry((ptr), struct request, queuelist)
 
-#define rq_data_dir(rq)		((rq)->flags & 1)
+#define rq_data_dir(rq)		((rq)->cmd_flags & 1)
 
 static inline int blk_queue_full(struct request_queue *q, int rw)
 {
@@ -541,7 +559,7 @@ static inline void blk_clear_queue_full(struct request_queue *q, int rw)
 #define RQ_NOMERGE_FLAGS	\
 	(REQ_NOMERGE | REQ_STARTED | REQ_HARDBARRIER | REQ_SOFTBARRIER)
 #define rq_mergeable(rq)	\
-	(!((rq)->flags & RQ_NOMERGE_FLAGS) && blk_fs_request((rq)))
+	(!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && blk_fs_request((rq)))
 
 /*
  * noop, requests are automagically marked as active/inactive by I/O
@@ -737,7 +755,7 @@ extern void blk_put_queue(request_queue_t *);
  */
 #define blk_queue_tag_depth(q)		((q)->queue_tags->busy)
 #define blk_queue_tag_queue(q)		((q)->queue_tags->busy < (q)->queue_tags->max_depth)
-#define blk_rq_tagged(rq)		((rq)->flags & REQ_QUEUED)
+#define blk_rq_tagged(rq)		((rq)->cmd_flags & REQ_QUEUED)
 extern int blk_queue_start_tag(request_queue_t *, struct request *);
 extern struct request *blk_queue_find_tag(request_queue_t *, int);
 extern void blk_queue_end_tag(request_queue_t *, struct request *);
diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index 7520cc1ff9e2..ea48eb1b3fd3 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -148,7 +148,7 @@ static inline void blk_add_trace_rq(struct request_queue *q, struct request *rq,
 				    u32 what)
 {
 	struct blk_trace *bt = q->blk_trace;
-	int rw = rq->flags & 0x03;
+	int rw = rq->cmd_flags & 0x03;
 
 	if (likely(!bt))
 		return;
diff --git a/include/scsi/scsi_tcq.h b/include/scsi/scsi_tcq.h
index d04d05adfa9b..bbf66219b769 100644
--- a/include/scsi/scsi_tcq.h
+++ b/include/scsi/scsi_tcq.h
@@ -100,7 +100,7 @@ static inline int scsi_populate_tag_msg(struct scsi_cmnd *cmd, char *msg)
 	struct scsi_device *sdev = cmd->device;
 
         if (blk_rq_tagged(req)) {
-		if (sdev->ordered_tags && req->flags & REQ_HARDBARRIER)
+		if (sdev->ordered_tags && req->cmd_flags & REQ_HARDBARRIER)
         	        *msg++ = MSG_ORDERED_TAG;
         	else
         	        *msg++ = MSG_SIMPLE_TAG;
-- 
cgit v1.2.3


From 9817064b68fef7e4580c6df1ea597e106b9ff88b Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Fri, 28 Jul 2006 09:23:08 +0200
Subject: [PATCH] elevator: move the backmerging logic into the elevator core

Right now, every IO scheduler implements its own backmerging (except for
noop, which does no merging). That results in duplicated code for
essentially the same operation, which is never a good thing. This patch
moves the backmerging out of the io schedulers and into the elevator
core. We save 1.6kb of text and as a bonus get backmerging for noop as
well. Win-win!

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/as-iosched.c       | 139 +-------------------------------------------
 block/cfq-iosched.c      |  86 +--------------------------
 block/deadline-iosched.c | 128 +----------------------------------------
 block/elevator.c         | 147 ++++++++++++++++++++++++++++++++++++++++++-----
 block/ll_rw_blk.c        |   2 +
 include/linux/blkdev.h   |  17 +-----
 include/linux/elevator.h |   2 +
 7 files changed, 146 insertions(+), 375 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index ad1cc4077819..6db494333c3a 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -14,7 +14,6 @@
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/compiler.h>
-#include <linux/hash.h>
 #include <linux/rbtree.h>
 #include <linux/interrupt.h>
 
@@ -95,7 +94,6 @@ struct as_data {
 
 	struct as_rq *next_arq[2];	/* next in sort order */
 	sector_t last_sector[2];	/* last REQ_SYNC & REQ_ASYNC sectors */
-	struct hlist_head *hash;	/* request hash */
 
 	unsigned long exit_prob;	/* probability a task will exit while
 					   being waited on */
@@ -161,11 +159,6 @@ struct as_rq {
 
 	struct io_context *io_context;	/* The submitting task */
 
-	/*
-	 * request hash, key is the ending offset (for back merge lookup)
-	 */
-	struct hlist_node hash;
-
 	/*
 	 * expire fifo
 	 */
@@ -272,77 +265,6 @@ static void as_put_io_context(struct as_rq *arq)
 	put_io_context(arq->io_context);
 }
 
-/*
- * the back merge hash support functions
- */
-static const int as_hash_shift = 6;
-#define AS_HASH_BLOCK(sec)	((sec) >> 3)
-#define AS_HASH_FN(sec)		(hash_long(AS_HASH_BLOCK((sec)), as_hash_shift))
-#define AS_HASH_ENTRIES		(1 << as_hash_shift)
-#define rq_hash_key(rq)		((rq)->sector + (rq)->nr_sectors)
-
-static inline void __as_del_arq_hash(struct as_rq *arq)
-{
-	hlist_del_init(&arq->hash);
-}
-
-static inline void as_del_arq_hash(struct as_rq *arq)
-{
-	if (!hlist_unhashed(&arq->hash))
-		__as_del_arq_hash(arq);
-}
-
-static void as_add_arq_hash(struct as_data *ad, struct as_rq *arq)
-{
-	struct request *rq = arq->request;
-
-	BUG_ON(!hlist_unhashed(&arq->hash));
-
-	hlist_add_head(&arq->hash, &ad->hash[AS_HASH_FN(rq_hash_key(rq))]);
-}
-
-/*
- * move hot entry to front of chain
- */
-static inline void as_hot_arq_hash(struct as_data *ad, struct as_rq *arq)
-{
-	struct request *rq = arq->request;
-	struct hlist_head *head = &ad->hash[AS_HASH_FN(rq_hash_key(rq))];
-
-	if (hlist_unhashed(&arq->hash)) {
-		WARN_ON(1);
-		return;
-	}
-
-	if (&arq->hash != head->first) {
-		hlist_del(&arq->hash);
-		hlist_add_head(&arq->hash, head);
-	}
-}
-
-static struct request *as_find_arq_hash(struct as_data *ad, sector_t offset)
-{
-	struct hlist_head *hash_list = &ad->hash[AS_HASH_FN(offset)];
-	struct hlist_node *entry, *next;
-	struct as_rq *arq;
-
-	hlist_for_each_entry_safe(arq, entry, next, hash_list, hash) {
-		struct request *__rq = arq->request;
-
-		BUG_ON(hlist_unhashed(&arq->hash));
-
-		if (!rq_mergeable(__rq)) {
-			as_del_arq_hash(arq);
-			continue;
-		}
-
-		if (rq_hash_key(__rq) == offset)
-			return __rq;
-	}
-
-	return NULL;
-}
-
 /*
  * rb tree support functions
  */
@@ -1060,7 +982,6 @@ static void as_remove_queued_request(request_queue_t *q, struct request *rq)
 		ad->next_arq[data_dir] = as_find_next_arq(ad, arq);
 
 	list_del_init(&arq->fifo);
-	as_del_arq_hash(arq);
 	as_del_arq_rb(ad, arq);
 }
 
@@ -1349,8 +1270,6 @@ static void as_add_request(request_queue_t *q, struct request *rq)
 	}
 
 	as_add_arq_rb(ad, arq);
-	if (rq_mergeable(arq->request))
-		as_add_arq_hash(ad, arq);
 
 	/*
 	 * set expire time (only used for reads) and add to fifo list
@@ -1428,42 +1347,17 @@ as_merge(request_queue_t *q, struct request **req, struct bio *bio)
 	struct as_data *ad = q->elevator->elevator_data;
 	sector_t rb_key = bio->bi_sector + bio_sectors(bio);
 	struct request *__rq;
-	int ret;
-
-	/*
-	 * see if the merge hash can satisfy a back merge
-	 */
-	__rq = as_find_arq_hash(ad, bio->bi_sector);
-	if (__rq) {
-		BUG_ON(__rq->sector + __rq->nr_sectors != bio->bi_sector);
-
-		if (elv_rq_merge_ok(__rq, bio)) {
-			ret = ELEVATOR_BACK_MERGE;
-			goto out;
-		}
-	}
 
 	/*
 	 * check for front merge
 	 */
 	__rq = as_find_arq_rb(ad, rb_key, bio_data_dir(bio));
-	if (__rq) {
-		BUG_ON(rb_key != rq_rb_key(__rq));
-
-		if (elv_rq_merge_ok(__rq, bio)) {
-			ret = ELEVATOR_FRONT_MERGE;
-			goto out;
-		}
+	if (__rq && elv_rq_merge_ok(__rq, bio)) {
+		*req = __rq;
+		return ELEVATOR_FRONT_MERGE;
 	}
 
 	return ELEVATOR_NO_MERGE;
-out:
-	if (ret) {
-		if (rq_mergeable(__rq))
-			as_hot_arq_hash(ad, RQ_DATA(__rq));
-	}
-	*req = __rq;
-	return ret;
 }
 
 static void as_merged_request(request_queue_t *q, struct request *req)
@@ -1471,12 +1365,6 @@ static void as_merged_request(request_queue_t *q, struct request *req)
 	struct as_data *ad = q->elevator->elevator_data;
 	struct as_rq *arq = RQ_DATA(req);
 
-	/*
-	 * hash always needs to be repositioned, key is end sector
-	 */
-	as_del_arq_hash(arq);
-	as_add_arq_hash(ad, arq);
-
 	/*
 	 * if the merge was a front merge, we need to reposition request
 	 */
@@ -1501,13 +1389,6 @@ static void as_merged_requests(request_queue_t *q, struct request *req,
 	BUG_ON(!arq);
 	BUG_ON(!anext);
 
-	/*
-	 * reposition arq (this is the merged request) in hash, and in rbtree
-	 * in case of a front merge
-	 */
-	as_del_arq_hash(arq);
-	as_add_arq_hash(ad, arq);
-
 	if (rq_rb_key(req) != arq->rb_key) {
 		as_del_arq_rb(ad, arq);
 		as_add_arq_rb(ad, arq);
@@ -1591,7 +1472,6 @@ static int as_set_request(request_queue_t *q, struct request *rq,
 		arq->request = rq;
 		arq->state = AS_RQ_PRESCHED;
 		arq->io_context = NULL;
-		INIT_HLIST_NODE(&arq->hash);
 		INIT_LIST_HEAD(&arq->fifo);
 		rq->elevator_private = arq;
 		return 0;
@@ -1628,7 +1508,6 @@ static void as_exit_queue(elevator_t *e)
 
 	mempool_destroy(ad->arq_pool);
 	put_io_context(ad->io_context);
-	kfree(ad->hash);
 	kfree(ad);
 }
 
@@ -1639,7 +1518,6 @@ static void as_exit_queue(elevator_t *e)
 static void *as_init_queue(request_queue_t *q, elevator_t *e)
 {
 	struct as_data *ad;
-	int i;
 
 	if (!arq_pool)
 		return NULL;
@@ -1651,17 +1529,9 @@ static void *as_init_queue(request_queue_t *q, elevator_t *e)
 
 	ad->q = q; /* Identify what queue the data belongs to */
 
-	ad->hash = kmalloc_node(sizeof(struct hlist_head)*AS_HASH_ENTRIES,
-				GFP_KERNEL, q->node);
-	if (!ad->hash) {
-		kfree(ad);
-		return NULL;
-	}
-
 	ad->arq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
 				mempool_free_slab, arq_pool, q->node);
 	if (!ad->arq_pool) {
-		kfree(ad->hash);
 		kfree(ad);
 		return NULL;
 	}
@@ -1672,9 +1542,6 @@ static void *as_init_queue(request_queue_t *q, elevator_t *e)
 	init_timer(&ad->antic_timer);
 	INIT_WORK(&ad->antic_work, as_work_handler, q);
 
-	for (i = 0; i < AS_HASH_ENTRIES; i++)
-		INIT_HLIST_HEAD(&ad->hash[i]);
-
 	INIT_LIST_HEAD(&ad->fifo_list[REQ_SYNC]);
 	INIT_LIST_HEAD(&ad->fifo_list[REQ_ASYNC]);
 	ad->sort_list[REQ_SYNC] = RB_ROOT;
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 3a3aee08ec5f..1b803c0c90f1 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -41,16 +41,6 @@ static DEFINE_SPINLOCK(cfq_exit_lock);
 #define CFQ_QHASH_ENTRIES	(1 << CFQ_QHASH_SHIFT)
 #define list_entry_qhash(entry)	hlist_entry((entry), struct cfq_queue, cfq_hash)
 
-/*
- * for the hash of crq inside the cfqq
- */
-#define CFQ_MHASH_SHIFT		6
-#define CFQ_MHASH_BLOCK(sec)	((sec) >> 3)
-#define CFQ_MHASH_ENTRIES	(1 << CFQ_MHASH_SHIFT)
-#define CFQ_MHASH_FN(sec)	hash_long(CFQ_MHASH_BLOCK(sec), CFQ_MHASH_SHIFT)
-#define rq_hash_key(rq)		((rq)->sector + (rq)->nr_sectors)
-#define list_entry_hash(ptr)	hlist_entry((ptr), struct cfq_rq, hash)
-
 #define list_entry_cfqq(ptr)	list_entry((ptr), struct cfq_queue, cfq_list)
 #define list_entry_fifo(ptr)	list_entry((ptr), struct request, queuelist)
 
@@ -112,11 +102,6 @@ struct cfq_data {
 	 */
 	struct hlist_head *cfq_hash;
 
-	/*
-	 * global crq hash for all queues
-	 */
-	struct hlist_head *crq_hash;
-
 	mempool_t *crq_pool;
 
 	int rq_in_driver;
@@ -203,7 +188,6 @@ struct cfq_rq {
 	struct rb_node rb_node;
 	sector_t rb_key;
 	struct request *request;
-	struct hlist_node hash;
 
 	struct cfq_queue *cfq_queue;
 	struct cfq_io_context *io_context;
@@ -271,42 +255,6 @@ static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *, unsigned int, unsi
 static void cfq_dispatch_insert(request_queue_t *, struct cfq_rq *);
 static struct cfq_queue *cfq_get_queue(struct cfq_data *cfqd, unsigned int key, struct task_struct *tsk, gfp_t gfp_mask);
 
-/*
- * lots of deadline iosched dupes, can be abstracted later...
- */
-static inline void cfq_del_crq_hash(struct cfq_rq *crq)
-{
-	hlist_del_init(&crq->hash);
-}
-
-static inline void cfq_add_crq_hash(struct cfq_data *cfqd, struct cfq_rq *crq)
-{
-	const int hash_idx = CFQ_MHASH_FN(rq_hash_key(crq->request));
-
-	hlist_add_head(&crq->hash, &cfqd->crq_hash[hash_idx]);
-}
-
-static struct request *cfq_find_rq_hash(struct cfq_data *cfqd, sector_t offset)
-{
-	struct hlist_head *hash_list = &cfqd->crq_hash[CFQ_MHASH_FN(offset)];
-	struct hlist_node *entry, *next;
-
-	hlist_for_each_safe(entry, next, hash_list) {
-		struct cfq_rq *crq = list_entry_hash(entry);
-		struct request *__rq = crq->request;
-
-		if (!rq_mergeable(__rq)) {
-			cfq_del_crq_hash(crq);
-			continue;
-		}
-
-		if (rq_hash_key(__rq) == offset)
-			return __rq;
-	}
-
-	return NULL;
-}
-
 /*
  * scheduler run of queue, if there are requests pending and no one in the
  * driver that will restart queueing
@@ -677,7 +625,6 @@ static void cfq_remove_request(struct request *rq)
 
 	list_del_init(&rq->queuelist);
 	cfq_del_crq_rb(crq);
-	cfq_del_crq_hash(crq);
 }
 
 static int
@@ -685,34 +632,20 @@ cfq_merge(request_queue_t *q, struct request **req, struct bio *bio)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct request *__rq;
-	int ret;
-
-	__rq = cfq_find_rq_hash(cfqd, bio->bi_sector);
-	if (__rq && elv_rq_merge_ok(__rq, bio)) {
-		ret = ELEVATOR_BACK_MERGE;
-		goto out;
-	}
 
 	__rq = cfq_find_rq_fmerge(cfqd, bio);
 	if (__rq && elv_rq_merge_ok(__rq, bio)) {
-		ret = ELEVATOR_FRONT_MERGE;
-		goto out;
+		*req = __rq;
+		return ELEVATOR_FRONT_MERGE;
 	}
 
 	return ELEVATOR_NO_MERGE;
-out:
-	*req = __rq;
-	return ret;
 }
 
 static void cfq_merged_request(request_queue_t *q, struct request *req)
 {
-	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct cfq_rq *crq = RQ_DATA(req);
 
-	cfq_del_crq_hash(crq);
-	cfq_add_crq_hash(cfqd, crq);
-
 	if (rq_rb_key(req) != crq->rb_key) {
 		struct cfq_queue *cfqq = crq->cfq_queue;
 
@@ -1825,9 +1758,6 @@ static void cfq_insert_request(request_queue_t *q, struct request *rq)
 
 	list_add_tail(&rq->queuelist, &cfqq->fifo);
 
-	if (rq_mergeable(rq))
-		cfq_add_crq_hash(cfqd, crq);
-
 	cfq_crq_enqueued(cfqd, cfqq, crq);
 }
 
@@ -2055,7 +1985,6 @@ cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 		RB_CLEAR_NODE(&crq->rb_node);
 		crq->rb_key = 0;
 		crq->request = rq;
-		INIT_HLIST_NODE(&crq->hash);
 		crq->cfq_queue = cfqq;
 		crq->io_context = cic;
 
@@ -2221,7 +2150,6 @@ static void cfq_exit_queue(elevator_t *e)
 	cfq_shutdown_timer_wq(cfqd);
 
 	mempool_destroy(cfqd->crq_pool);
-	kfree(cfqd->crq_hash);
 	kfree(cfqd->cfq_hash);
 	kfree(cfqd);
 }
@@ -2246,20 +2174,14 @@ static void *cfq_init_queue(request_queue_t *q, elevator_t *e)
 	INIT_LIST_HEAD(&cfqd->empty_list);
 	INIT_LIST_HEAD(&cfqd->cic_list);
 
-	cfqd->crq_hash = kmalloc(sizeof(struct hlist_head) * CFQ_MHASH_ENTRIES, GFP_KERNEL);
-	if (!cfqd->crq_hash)
-		goto out_crqhash;
-
 	cfqd->cfq_hash = kmalloc(sizeof(struct hlist_head) * CFQ_QHASH_ENTRIES, GFP_KERNEL);
 	if (!cfqd->cfq_hash)
-		goto out_cfqhash;
+		goto out_crqhash;
 
 	cfqd->crq_pool = mempool_create_slab_pool(BLKDEV_MIN_RQ, crq_pool);
 	if (!cfqd->crq_pool)
 		goto out_crqpool;
 
-	for (i = 0; i < CFQ_MHASH_ENTRIES; i++)
-		INIT_HLIST_HEAD(&cfqd->crq_hash[i]);
 	for (i = 0; i < CFQ_QHASH_ENTRIES; i++)
 		INIT_HLIST_HEAD(&cfqd->cfq_hash[i]);
 
@@ -2289,8 +2211,6 @@ static void *cfq_init_queue(request_queue_t *q, elevator_t *e)
 	return cfqd;
 out_crqpool:
 	kfree(cfqd->cfq_hash);
-out_cfqhash:
-	kfree(cfqd->crq_hash);
 out_crqhash:
 	kfree(cfqd);
 	return NULL;
diff --git a/block/deadline-iosched.c b/block/deadline-iosched.c
index c7ca9f0b6498..b66e820f544d 100644
--- a/block/deadline-iosched.c
+++ b/block/deadline-iosched.c
@@ -12,7 +12,6 @@
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/compiler.h>
-#include <linux/hash.h>
 #include <linux/rbtree.h>
 
 /*
@@ -24,13 +23,6 @@ static const int writes_starved = 2;    /* max times reads can starve a write */
 static const int fifo_batch = 16;       /* # of sequential requests treated as one
 				     by the above parameters. For throughput. */
 
-static const int deadline_hash_shift = 5;
-#define DL_HASH_BLOCK(sec)	((sec) >> 3)
-#define DL_HASH_FN(sec)		(hash_long(DL_HASH_BLOCK((sec)), deadline_hash_shift))
-#define DL_HASH_ENTRIES		(1 << deadline_hash_shift)
-#define rq_hash_key(rq)		((rq)->sector + (rq)->nr_sectors)
-#define ON_HASH(drq)		(!hlist_unhashed(&(drq)->hash))
-
 struct deadline_data {
 	/*
 	 * run time data
@@ -46,7 +38,6 @@ struct deadline_data {
 	 * next in sort order. read, write or both are NULL
 	 */
 	struct deadline_rq *next_drq[2];
-	struct hlist_head *hash;	/* request hash */
 	unsigned int batching;		/* number of sequential requests made */
 	sector_t last_sector;		/* head position */
 	unsigned int starved;		/* times reads have starved writes */
@@ -74,11 +65,6 @@ struct deadline_rq {
 
 	struct request *request;
 
-	/*
-	 * request hash, key is the ending offset (for back merge lookup)
-	 */
-	struct hlist_node hash;
-
 	/*
 	 * expire fifo
 	 */
@@ -92,69 +78,6 @@ static kmem_cache_t *drq_pool;
 
 #define RQ_DATA(rq)	((struct deadline_rq *) (rq)->elevator_private)
 
-/*
- * the back merge hash support functions
- */
-static inline void __deadline_del_drq_hash(struct deadline_rq *drq)
-{
-	hlist_del_init(&drq->hash);
-}
-
-static inline void deadline_del_drq_hash(struct deadline_rq *drq)
-{
-	if (ON_HASH(drq))
-		__deadline_del_drq_hash(drq);
-}
-
-static inline void
-deadline_add_drq_hash(struct deadline_data *dd, struct deadline_rq *drq)
-{
-	struct request *rq = drq->request;
-
-	BUG_ON(ON_HASH(drq));
-
-	hlist_add_head(&drq->hash, &dd->hash[DL_HASH_FN(rq_hash_key(rq))]);
-}
-
-/*
- * move hot entry to front of chain
- */
-static inline void
-deadline_hot_drq_hash(struct deadline_data *dd, struct deadline_rq *drq)
-{
-	struct request *rq = drq->request;
-	struct hlist_head *head = &dd->hash[DL_HASH_FN(rq_hash_key(rq))];
-
-	if (ON_HASH(drq) && &drq->hash != head->first) {
-		hlist_del(&drq->hash);
-		hlist_add_head(&drq->hash, head);
-	}
-}
-
-static struct request *
-deadline_find_drq_hash(struct deadline_data *dd, sector_t offset)
-{
-	struct hlist_head *hash_list = &dd->hash[DL_HASH_FN(offset)];
-	struct hlist_node *entry, *next;
-	struct deadline_rq *drq;
-
-	hlist_for_each_entry_safe(drq, entry, next, hash_list, hash) {
-		struct request *__rq = drq->request;
-
-		BUG_ON(!ON_HASH(drq));
-
-		if (!rq_mergeable(__rq)) {
-			__deadline_del_drq_hash(drq);
-			continue;
-		}
-
-		if (rq_hash_key(__rq) == offset)
-			return __rq;
-	}
-
-	return NULL;
-}
-
 /*
  * rb tree support functions
  */
@@ -267,22 +190,19 @@ deadline_add_request(struct request_queue *q, struct request *rq)
 {
 	struct deadline_data *dd = q->elevator->elevator_data;
 	struct deadline_rq *drq = RQ_DATA(rq);
-
 	const int data_dir = rq_data_dir(drq->request);
 
 	deadline_add_drq_rb(dd, drq);
+
 	/*
 	 * set expire time (only used for reads) and add to fifo list
 	 */
 	drq->expires = jiffies + dd->fifo_expire[data_dir];
 	list_add_tail(&drq->fifo, &dd->fifo_list[data_dir]);
-
-	if (rq_mergeable(rq))
-		deadline_add_drq_hash(dd, drq);
 }
 
 /*
- * remove rq from rbtree, fifo, and hash
+ * remove rq from rbtree and fifo.
  */
 static void deadline_remove_request(request_queue_t *q, struct request *rq)
 {
@@ -291,7 +211,6 @@ static void deadline_remove_request(request_queue_t *q, struct request *rq)
 
 	list_del_init(&drq->fifo);
 	deadline_del_drq_rb(dd, drq);
-	deadline_del_drq_hash(drq);
 }
 
 static int
@@ -301,19 +220,6 @@ deadline_merge(request_queue_t *q, struct request **req, struct bio *bio)
 	struct request *__rq;
 	int ret;
 
-	/*
-	 * see if the merge hash can satisfy a back merge
-	 */
-	__rq = deadline_find_drq_hash(dd, bio->bi_sector);
-	if (__rq) {
-		BUG_ON(__rq->sector + __rq->nr_sectors != bio->bi_sector);
-
-		if (elv_rq_merge_ok(__rq, bio)) {
-			ret = ELEVATOR_BACK_MERGE;
-			goto out;
-		}
-	}
-
 	/*
 	 * check for front merge
 	 */
@@ -333,8 +239,6 @@ deadline_merge(request_queue_t *q, struct request **req, struct bio *bio)
 
 	return ELEVATOR_NO_MERGE;
 out:
-	if (ret)
-		deadline_hot_drq_hash(dd, RQ_DATA(__rq));
 	*req = __rq;
 	return ret;
 }
@@ -344,12 +248,6 @@ static void deadline_merged_request(request_queue_t *q, struct request *req)
 	struct deadline_data *dd = q->elevator->elevator_data;
 	struct deadline_rq *drq = RQ_DATA(req);
 
-	/*
-	 * hash always needs to be repositioned, key is end sector
-	 */
-	deadline_del_drq_hash(drq);
-	deadline_add_drq_hash(dd, drq);
-
 	/*
 	 * if the merge was a front merge, we need to reposition request
 	 */
@@ -370,13 +268,6 @@ deadline_merged_requests(request_queue_t *q, struct request *req,
 	BUG_ON(!drq);
 	BUG_ON(!dnext);
 
-	/*
-	 * reposition drq (this is the merged request) in hash, and in rbtree
-	 * in case of a front merge
-	 */
-	deadline_del_drq_hash(drq);
-	deadline_add_drq_hash(dd, drq);
-
 	if (rq_rb_key(req) != drq->rb_key) {
 		deadline_del_drq_rb(dd, drq);
 		deadline_add_drq_rb(dd, drq);
@@ -594,7 +485,6 @@ static void deadline_exit_queue(elevator_t *e)
 	BUG_ON(!list_empty(&dd->fifo_list[WRITE]));
 
 	mempool_destroy(dd->drq_pool);
-	kfree(dd->hash);
 	kfree(dd);
 }
 
@@ -605,7 +495,6 @@ static void deadline_exit_queue(elevator_t *e)
 static void *deadline_init_queue(request_queue_t *q, elevator_t *e)
 {
 	struct deadline_data *dd;
-	int i;
 
 	if (!drq_pool)
 		return NULL;
@@ -615,24 +504,13 @@ static void *deadline_init_queue(request_queue_t *q, elevator_t *e)
 		return NULL;
 	memset(dd, 0, sizeof(*dd));
 
-	dd->hash = kmalloc_node(sizeof(struct hlist_head)*DL_HASH_ENTRIES,
-				GFP_KERNEL, q->node);
-	if (!dd->hash) {
-		kfree(dd);
-		return NULL;
-	}
-
 	dd->drq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
 					mempool_free_slab, drq_pool, q->node);
 	if (!dd->drq_pool) {
-		kfree(dd->hash);
 		kfree(dd);
 		return NULL;
 	}
 
-	for (i = 0; i < DL_HASH_ENTRIES; i++)
-		INIT_HLIST_HEAD(&dd->hash[i]);
-
 	INIT_LIST_HEAD(&dd->fifo_list[READ]);
 	INIT_LIST_HEAD(&dd->fifo_list[WRITE]);
 	dd->sort_list[READ] = RB_ROOT;
@@ -667,8 +545,6 @@ deadline_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
 		RB_CLEAR_NODE(&drq->rb_node);
 		drq->request = rq;
 
-		INIT_HLIST_NODE(&drq->hash);
-
 		INIT_LIST_HEAD(&drq->fifo);
 
 		rq->elevator_private = drq;
diff --git a/block/elevator.c b/block/elevator.c
index 4ac97b642042..cff1102dac9d 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -33,12 +33,23 @@
 #include <linux/compiler.h>
 #include <linux/delay.h>
 #include <linux/blktrace_api.h>
+#include <linux/hash.h>
 
 #include <asm/uaccess.h>
 
 static DEFINE_SPINLOCK(elv_list_lock);
 static LIST_HEAD(elv_list);
 
+/*
+ * Merge hash stuff.
+ */
+static const int elv_hash_shift = 6;
+#define ELV_HASH_BLOCK(sec)	((sec) >> 3)
+#define ELV_HASH_FN(sec)	(hash_long(ELV_HASH_BLOCK((sec)), elv_hash_shift))
+#define ELV_HASH_ENTRIES	(1 << elv_hash_shift)
+#define rq_hash_key(rq)		((rq)->sector + (rq)->nr_sectors)
+#define ELV_ON_HASH(rq)		(!hlist_unhashed(&(rq)->hash))
+
 /*
  * can we safely merge with this request?
  */
@@ -153,25 +164,41 @@ static struct kobj_type elv_ktype;
 
 static elevator_t *elevator_alloc(struct elevator_type *e)
 {
-	elevator_t *eq = kmalloc(sizeof(elevator_t), GFP_KERNEL);
-	if (eq) {
-		memset(eq, 0, sizeof(*eq));
-		eq->ops = &e->ops;
-		eq->elevator_type = e;
-		kobject_init(&eq->kobj);
-		snprintf(eq->kobj.name, KOBJ_NAME_LEN, "%s", "iosched");
-		eq->kobj.ktype = &elv_ktype;
-		mutex_init(&eq->sysfs_lock);
-	} else {
-		elevator_put(e);
-	}
+	elevator_t *eq;
+	int i;
+
+	eq = kmalloc(sizeof(elevator_t), GFP_KERNEL);
+	if (unlikely(!eq))
+		goto err;
+
+	memset(eq, 0, sizeof(*eq));
+	eq->ops = &e->ops;
+	eq->elevator_type = e;
+	kobject_init(&eq->kobj);
+	snprintf(eq->kobj.name, KOBJ_NAME_LEN, "%s", "iosched");
+	eq->kobj.ktype = &elv_ktype;
+	mutex_init(&eq->sysfs_lock);
+
+	eq->hash = kmalloc(sizeof(struct hlist_head) * ELV_HASH_ENTRIES, GFP_KERNEL);
+	if (!eq->hash)
+		goto err;
+
+	for (i = 0; i < ELV_HASH_ENTRIES; i++)
+		INIT_HLIST_HEAD(&eq->hash[i]);
+
 	return eq;
+err:
+	kfree(eq);
+	elevator_put(e);
+	return NULL;
 }
 
 static void elevator_release(struct kobject *kobj)
 {
 	elevator_t *e = container_of(kobj, elevator_t, kobj);
+
 	elevator_put(e->elevator_type);
+	kfree(e->hash);
 	kfree(e);
 }
 
@@ -223,6 +250,53 @@ void elevator_exit(elevator_t *e)
 	kobject_put(&e->kobj);
 }
 
+static inline void __elv_rqhash_del(struct request *rq)
+{
+	hlist_del_init(&rq->hash);
+}
+
+static void elv_rqhash_del(request_queue_t *q, struct request *rq)
+{
+	if (ELV_ON_HASH(rq))
+		__elv_rqhash_del(rq);
+}
+
+static void elv_rqhash_add(request_queue_t *q, struct request *rq)
+{
+	elevator_t *e = q->elevator;
+
+	BUG_ON(ELV_ON_HASH(rq));
+	hlist_add_head(&rq->hash, &e->hash[ELV_HASH_FN(rq_hash_key(rq))]);
+}
+
+static void elv_rqhash_reposition(request_queue_t *q, struct request *rq)
+{
+	__elv_rqhash_del(rq);
+	elv_rqhash_add(q, rq);
+}
+
+static struct request *elv_rqhash_find(request_queue_t *q, sector_t offset)
+{
+	elevator_t *e = q->elevator;
+	struct hlist_head *hash_list = &e->hash[ELV_HASH_FN(offset)];
+	struct hlist_node *entry, *next;
+	struct request *rq;
+
+	hlist_for_each_entry_safe(rq, entry, next, hash_list, hash) {
+		BUG_ON(!ELV_ON_HASH(rq));
+
+		if (unlikely(!rq_mergeable(rq))) {
+			__elv_rqhash_del(rq);
+			continue;
+		}
+
+		if (rq_hash_key(rq) == offset)
+			return rq;
+	}
+
+	return NULL;
+}
+
 /*
  * Insert rq into dispatch queue of q.  Queue lock must be held on
  * entry.  If sort != 0, rq is sort-inserted; otherwise, rq will be
@@ -235,6 +309,9 @@ void elv_dispatch_sort(request_queue_t *q, struct request *rq)
 
 	if (q->last_merge == rq)
 		q->last_merge = NULL;
+
+	elv_rqhash_del(q, rq);
+
 	q->nr_sorted--;
 
 	boundary = q->end_sector;
@@ -258,11 +335,32 @@ void elv_dispatch_sort(request_queue_t *q, struct request *rq)
 	list_add(&rq->queuelist, entry);
 }
 
+/*
+ * This should be in elevator.h, but that requires pulling in rq and q
+ */
+void elv_dispatch_add_tail(struct request_queue *q, struct request *rq)
+{
+	if (q->last_merge == rq)
+		q->last_merge = NULL;
+
+	elv_rqhash_del(q, rq);
+
+	q->nr_sorted--;
+
+	q->end_sector = rq_end_sector(rq);
+	q->boundary_rq = rq;
+	list_add_tail(&rq->queuelist, &q->queue_head);
+}
+
 int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
 {
 	elevator_t *e = q->elevator;
+	struct request *__rq;
 	int ret;
 
+	/*
+	 * First try one-hit cache.
+	 */
 	if (q->last_merge) {
 		ret = elv_try_merge(q->last_merge, bio);
 		if (ret != ELEVATOR_NO_MERGE) {
@@ -271,6 +369,15 @@ int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
 		}
 	}
 
+	/*
+	 * See if our hash lookup can find a potential backmerge.
+	 */
+	__rq = elv_rqhash_find(q, bio->bi_sector);
+	if (__rq && elv_rq_merge_ok(__rq, bio)) {
+		*req = __rq;
+		return ELEVATOR_BACK_MERGE;
+	}
+
 	if (e->ops->elevator_merge_fn)
 		return e->ops->elevator_merge_fn(q, req, bio);
 
@@ -284,6 +391,8 @@ void elv_merged_request(request_queue_t *q, struct request *rq)
 	if (e->ops->elevator_merged_fn)
 		e->ops->elevator_merged_fn(q, rq);
 
+	elv_rqhash_reposition(q, rq);
+
 	q->last_merge = rq;
 }
 
@@ -294,8 +403,11 @@ void elv_merge_requests(request_queue_t *q, struct request *rq,
 
 	if (e->ops->elevator_merge_req_fn)
 		e->ops->elevator_merge_req_fn(q, rq, next);
-	q->nr_sorted--;
 
+	elv_rqhash_reposition(q, rq);
+	elv_rqhash_del(q, next);
+
+	q->nr_sorted--;
 	q->last_merge = rq;
 }
 
@@ -371,8 +483,12 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
 		BUG_ON(!blk_fs_request(rq));
 		rq->cmd_flags |= REQ_SORTED;
 		q->nr_sorted++;
-		if (q->last_merge == NULL && rq_mergeable(rq))
-			q->last_merge = rq;
+		if (rq_mergeable(rq)) {
+			elv_rqhash_add(q, rq);
+			if (!q->last_merge)
+				q->last_merge = rq;
+		}
+
 		/*
 		 * Some ioscheds (cfq) run q->request_fn directly, so
 		 * rq cannot be accessed after calling
@@ -557,6 +673,7 @@ struct request *elv_next_request(request_queue_t *q)
 void elv_dequeue_request(request_queue_t *q, struct request *rq)
 {
 	BUG_ON(list_empty(&rq->queuelist));
+	BUG_ON(ELV_ON_HASH(rq));
 
 	list_del_init(&rq->queuelist);
 
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 9b91bb70c5ed..9cbf7b550c78 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -281,6 +281,7 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 {
 	INIT_LIST_HEAD(&rq->queuelist);
 	INIT_LIST_HEAD(&rq->donelist);
+	INIT_HLIST_NODE(&rq->hash);
 
 	rq->errors = 0;
 	rq->rq_status = RQ_ACTIVE;
@@ -2700,6 +2701,7 @@ void __blk_put_request(request_queue_t *q, struct request *req)
 		int priv = req->cmd_flags & REQ_ELVPRIV;
 
 		BUG_ON(!list_empty(&req->queuelist));
+		BUG_ON(!hlist_unhashed(&req->hash));
 
 		blk_free_request(q, req);
 		freed_request(q, rw, priv);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index b2a412cf468f..8f5486964671 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -229,6 +229,8 @@ struct request {
 	struct bio *bio;
 	struct bio *biotail;
 
+	struct hlist_node hash;	/* merge hash */
+
 	void *elevator_private;
 	void *completion_data;
 
@@ -696,21 +698,6 @@ static inline void blkdev_dequeue_request(struct request *req)
 	elv_dequeue_request(req->q, req);
 }
 
-/*
- * This should be in elevator.h, but that requires pulling in rq and q
- */
-static inline void elv_dispatch_add_tail(struct request_queue *q,
-					 struct request *rq)
-{
-	if (q->last_merge == rq)
-		q->last_merge = NULL;
-	q->nr_sorted--;
-
-	q->end_sector = rq_end_sector(rq);
-	q->boundary_rq = rq;
-	list_add_tail(&rq->queuelist, &q->queue_head);
-}
-
 /*
  * Access functions for manipulating queue properties
  */
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 1713ace808bf..2c270e90b33e 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -82,12 +82,14 @@ struct elevator_queue
 	struct kobject kobj;
 	struct elevator_type *elevator_type;
 	struct mutex sysfs_lock;
+	struct hlist_head *hash;
 };
 
 /*
  * block elevator interface
  */
 extern void elv_dispatch_sort(request_queue_t *, struct request *);
+extern void elv_dispatch_add_tail(request_queue_t *, struct request *);
 extern void elv_add_request(request_queue_t *, struct request *, int, int);
 extern void __elv_add_request(request_queue_t *, struct request *, int, int);
 extern void elv_insert(request_queue_t *, struct request *, int);
-- 
cgit v1.2.3


From 10fd48f2376db52f08bf0420d2c4f580e39269e1 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Tue, 11 Jul 2006 21:15:52 +0200
Subject: [PATCH] rbtree: fixed reversed RB_EMPTY_NODE and rb_next/prev

The conditions got reserved. Also make rb_next() and rb_prev() check
for the empty condition.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/as-iosched.c     | 4 ++--
 include/linux/rbtree.h | 2 +-
 lib/rbtree.c           | 6 ++++++
 3 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index 6db494333c3a..9d0f15a54c64 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -336,7 +336,7 @@ static void as_add_arq_rb(struct as_data *ad, struct as_rq *arq)
 
 static inline void as_del_arq_rb(struct as_data *ad, struct as_rq *arq)
 {
-	if (!RB_EMPTY_NODE(&arq->rb_node)) {
+	if (RB_EMPTY_NODE(&arq->rb_node)) {
 		WARN_ON(1);
 		return;
 	}
@@ -1039,7 +1039,7 @@ static void as_move_to_dispatch(struct as_data *ad, struct as_rq *arq)
 	struct request *rq = arq->request;
 	const int data_dir = arq->is_sync;
 
-	BUG_ON(!RB_EMPTY_NODE(&arq->rb_node));
+	BUG_ON(RB_EMPTY_NODE(&arq->rb_node));
 
 	as_antic_stop(ad);
 	ad->antic_status = ANTIC_OFF;
diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h
index 8d5382e62c08..344bc3495ddb 100644
--- a/include/linux/rbtree.h
+++ b/include/linux/rbtree.h
@@ -133,7 +133,7 @@ static inline void rb_set_color(struct rb_node *rb, int color)
 #define	rb_entry(ptr, type, member) container_of(ptr, type, member)
 
 #define RB_EMPTY_ROOT(root)	((root)->rb_node == NULL)
-#define RB_EMPTY_NODE(node)	(rb_parent(node) != node)
+#define RB_EMPTY_NODE(node)	(rb_parent(node) == node)
 #define RB_CLEAR_NODE(node)	(rb_set_parent(node, node))
 
 extern void rb_insert_color(struct rb_node *, struct rb_root *);
diff --git a/lib/rbtree.c b/lib/rbtree.c
index 1e55ba1c2edf..48499c2d88cc 100644
--- a/lib/rbtree.c
+++ b/lib/rbtree.c
@@ -322,6 +322,9 @@ struct rb_node *rb_next(struct rb_node *node)
 {
 	struct rb_node *parent;
 
+	if (rb_parent(node) == node)
+		return NULL;
+
 	/* If we have a right-hand child, go down and then left as far
 	   as we can. */
 	if (node->rb_right) {
@@ -348,6 +351,9 @@ struct rb_node *rb_prev(struct rb_node *node)
 {
 	struct rb_node *parent;
 
+	if (rb_parent(node) == node)
+		return NULL;
+
 	/* If we have a left-hand child, go down and then right as far
 	   as we can. */
 	if (node->rb_left) {
-- 
cgit v1.2.3


From 2e662b65f05d550b6799ed6bfa9963b82279e6b7 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Thu, 13 Jul 2006 11:55:04 +0200
Subject: [PATCH] elevator: abstract out the rbtree sort handling

The rbtree sort/lookup/reposition logic is mostly duplicated in
cfq/deadline/as, so move it to the elevator core. The io schedulers
still provide the actual rb root, as we don't want to impose any sort
of specific handling on the schedulers.

Introduce the helpers and rb_node in struct request to help migrate the
IO schedulers.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/elevator.c         | 123 +++++++++++++++++++++++++++++++++++++++++------
 block/ll_rw_blk.c        |   7 +--
 include/linux/blkdev.h   |   1 +
 include/linux/elevator.h |  18 ++++++-
 4 files changed, 130 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/block/elevator.c b/block/elevator.c
index cff1102dac9d..cbbc36ba016a 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -239,6 +239,8 @@ int elevator_init(request_queue_t *q, char *name)
 	return ret;
 }
 
+EXPORT_SYMBOL(elevator_init);
+
 void elevator_exit(elevator_t *e)
 {
 	mutex_lock(&e->sysfs_lock);
@@ -250,6 +252,8 @@ void elevator_exit(elevator_t *e)
 	kobject_put(&e->kobj);
 }
 
+EXPORT_SYMBOL(elevator_exit);
+
 static inline void __elv_rqhash_del(struct request *rq)
 {
 	hlist_del_init(&rq->hash);
@@ -297,10 +301,69 @@ static struct request *elv_rqhash_find(request_queue_t *q, sector_t offset)
 	return NULL;
 }
 
+/*
+ * RB-tree support functions for inserting/lookup/removal of requests
+ * in a sorted RB tree.
+ */
+struct request *elv_rb_add(struct rb_root *root, struct request *rq)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct request *__rq;
+
+	while (*p) {
+		parent = *p;
+		__rq = rb_entry(parent, struct request, rb_node);
+
+		if (rq->sector < __rq->sector)
+			p = &(*p)->rb_left;
+		else if (rq->sector > __rq->sector)
+			p = &(*p)->rb_right;
+		else
+			return __rq;
+	}
+
+	rb_link_node(&rq->rb_node, parent, p);
+	rb_insert_color(&rq->rb_node, root);
+	return NULL;
+}
+
+EXPORT_SYMBOL(elv_rb_add);
+
+void elv_rb_del(struct rb_root *root, struct request *rq)
+{
+	BUG_ON(RB_EMPTY_NODE(&rq->rb_node));
+	rb_erase(&rq->rb_node, root);
+	RB_CLEAR_NODE(&rq->rb_node);
+}
+
+EXPORT_SYMBOL(elv_rb_del);
+
+struct request *elv_rb_find(struct rb_root *root, sector_t sector)
+{
+	struct rb_node *n = root->rb_node;
+	struct request *rq;
+
+	while (n) {
+		rq = rb_entry(n, struct request, rb_node);
+
+		if (sector < rq->sector)
+			n = n->rb_left;
+		else if (sector > rq->sector)
+			n = n->rb_right;
+		else
+			return rq;
+	}
+
+	return NULL;
+}
+
+EXPORT_SYMBOL(elv_rb_find);
+
 /*
  * Insert rq into dispatch queue of q.  Queue lock must be held on
- * entry.  If sort != 0, rq is sort-inserted; otherwise, rq will be
- * appended to the dispatch queue.  To be used by specific elevators.
+ * entry.  rq is sort insted into the dispatch queue. To be used by
+ * specific elevators.
  */
 void elv_dispatch_sort(request_queue_t *q, struct request *rq)
 {
@@ -335,8 +398,12 @@ void elv_dispatch_sort(request_queue_t *q, struct request *rq)
 	list_add(&rq->queuelist, entry);
 }
 
+EXPORT_SYMBOL(elv_dispatch_sort);
+
 /*
- * This should be in elevator.h, but that requires pulling in rq and q
+ * Insert rq into dispatch queue of q.  Queue lock must be held on
+ * entry.  rq is added to the back of the dispatch queue. To be used by
+ * specific elevators.
  */
 void elv_dispatch_add_tail(struct request_queue *q, struct request *rq)
 {
@@ -352,6 +419,8 @@ void elv_dispatch_add_tail(struct request_queue *q, struct request *rq)
 	list_add_tail(&rq->queuelist, &q->queue_head);
 }
 
+EXPORT_SYMBOL(elv_dispatch_add_tail);
+
 int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
 {
 	elevator_t *e = q->elevator;
@@ -384,14 +453,15 @@ int elv_merge(request_queue_t *q, struct request **req, struct bio *bio)
 	return ELEVATOR_NO_MERGE;
 }
 
-void elv_merged_request(request_queue_t *q, struct request *rq)
+void elv_merged_request(request_queue_t *q, struct request *rq, int type)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_merged_fn)
-		e->ops->elevator_merged_fn(q, rq);
+		e->ops->elevator_merged_fn(q, rq, type);
 
-	elv_rqhash_reposition(q, rq);
+	if (type == ELEVATOR_BACK_MERGE)
+		elv_rqhash_reposition(q, rq);
 
 	q->last_merge = rq;
 }
@@ -577,6 +647,8 @@ void __elv_add_request(request_queue_t *q, struct request *rq, int where,
 	elv_insert(q, rq, where);
 }
 
+EXPORT_SYMBOL(__elv_add_request);
+
 void elv_add_request(request_queue_t *q, struct request *rq, int where,
 		     int plug)
 {
@@ -587,6 +659,8 @@ void elv_add_request(request_queue_t *q, struct request *rq, int where,
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
+EXPORT_SYMBOL(elv_add_request);
+
 static inline struct request *__elv_next_request(request_queue_t *q)
 {
 	struct request *rq;
@@ -670,6 +744,8 @@ struct request *elv_next_request(request_queue_t *q)
 	return rq;
 }
 
+EXPORT_SYMBOL(elv_next_request);
+
 void elv_dequeue_request(request_queue_t *q, struct request *rq)
 {
 	BUG_ON(list_empty(&rq->queuelist));
@@ -686,6 +762,8 @@ void elv_dequeue_request(request_queue_t *q, struct request *rq)
 		q->in_flight++;
 }
 
+EXPORT_SYMBOL(elv_dequeue_request);
+
 int elv_queue_empty(request_queue_t *q)
 {
 	elevator_t *e = q->elevator;
@@ -699,6 +777,8 @@ int elv_queue_empty(request_queue_t *q)
 	return 1;
 }
 
+EXPORT_SYMBOL(elv_queue_empty);
+
 struct request *elv_latter_request(request_queue_t *q, struct request *rq)
 {
 	elevator_t *e = q->elevator;
@@ -1025,11 +1105,26 @@ ssize_t elv_iosched_show(request_queue_t *q, char *name)
 	return len;
 }
 
-EXPORT_SYMBOL(elv_dispatch_sort);
-EXPORT_SYMBOL(elv_add_request);
-EXPORT_SYMBOL(__elv_add_request);
-EXPORT_SYMBOL(elv_next_request);
-EXPORT_SYMBOL(elv_dequeue_request);
-EXPORT_SYMBOL(elv_queue_empty);
-EXPORT_SYMBOL(elevator_exit);
-EXPORT_SYMBOL(elevator_init);
+struct request *elv_rb_former_request(request_queue_t *q, struct request *rq)
+{
+	struct rb_node *rbprev = rb_prev(&rq->rb_node);
+
+	if (rbprev)
+		return rb_entry_rq(rbprev);
+
+	return NULL;
+}
+
+EXPORT_SYMBOL(elv_rb_former_request);
+
+struct request *elv_rb_latter_request(request_queue_t *q, struct request *rq)
+{
+	struct rb_node *rbnext = rb_next(&rq->rb_node);
+
+	if (rbnext)
+		return rb_entry_rq(rbnext);
+
+	return NULL;
+}
+
+EXPORT_SYMBOL(elv_rb_latter_request);
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 9cbf7b550c78..d388486e98bb 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -281,11 +281,12 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 {
 	INIT_LIST_HEAD(&rq->queuelist);
 	INIT_LIST_HEAD(&rq->donelist);
-	INIT_HLIST_NODE(&rq->hash);
 
 	rq->errors = 0;
 	rq->rq_status = RQ_ACTIVE;
 	rq->bio = rq->biotail = NULL;
+	INIT_HLIST_NODE(&rq->hash);
+	RB_CLEAR_NODE(&rq->rb_node);
 	rq->ioprio = 0;
 	rq->buffer = NULL;
 	rq->ref_count = 1;
@@ -2943,7 +2944,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_back_merge(q, req))
-				elv_merged_request(q, req);
+				elv_merged_request(q, req, el_ret);
 			goto out;
 
 		case ELEVATOR_FRONT_MERGE:
@@ -2970,7 +2971,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_front_merge(q, req))
-				elv_merged_request(q, req);
+				elv_merged_request(q, req, el_ret);
 			goto out;
 
 		/* ELV_NO_MERGE: elevator says don't/can't merge. */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8f5486964671..a905c4934a55 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -230,6 +230,7 @@ struct request {
 	struct bio *biotail;
 
 	struct hlist_node hash;	/* merge hash */
+	struct rb_node rb_node;	/* sort/lookup */
 
 	void *elevator_private;
 	void *completion_data;
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 2c270e90b33e..95b2a04b969c 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -6,7 +6,7 @@ typedef int (elevator_merge_fn) (request_queue_t *, struct request **,
 
 typedef void (elevator_merge_req_fn) (request_queue_t *, struct request *, struct request *);
 
-typedef void (elevator_merged_fn) (request_queue_t *, struct request *);
+typedef void (elevator_merged_fn) (request_queue_t *, struct request *, int);
 
 typedef int (elevator_dispatch_fn) (request_queue_t *, int);
 
@@ -96,7 +96,7 @@ extern void elv_insert(request_queue_t *, struct request *, int);
 extern int elv_merge(request_queue_t *, struct request **, struct bio *);
 extern void elv_merge_requests(request_queue_t *, struct request *,
 			       struct request *);
-extern void elv_merged_request(request_queue_t *, struct request *);
+extern void elv_merged_request(request_queue_t *, struct request *, int);
 extern void elv_dequeue_request(request_queue_t *, struct request *);
 extern void elv_requeue_request(request_queue_t *, struct request *);
 extern int elv_queue_empty(request_queue_t *);
@@ -126,6 +126,19 @@ extern int elevator_init(request_queue_t *, char *);
 extern void elevator_exit(elevator_t *);
 extern int elv_rq_merge_ok(struct request *, struct bio *);
 
+/*
+ * Helper functions.
+ */
+extern struct request *elv_rb_former_request(request_queue_t *, struct request *);
+extern struct request *elv_rb_latter_request(request_queue_t *, struct request *);
+
+/*
+ * rb support functions.
+ */
+extern struct request *elv_rb_add(struct rb_root *, struct request *);
+extern void elv_rb_del(struct rb_root *, struct request *);
+extern struct request *elv_rb_find(struct rb_root *, sector_t);
+
 /*
  * Return values from elevator merger
  */
@@ -151,5 +164,6 @@ enum {
 };
 
 #define rq_end_sector(rq)	((rq)->sector + (rq)->nr_sectors)
+#define rb_entry_rq(node)	rb_entry((node), struct request, rb_node)
 
 #endif
-- 
cgit v1.2.3


From 1fbfdfcddff4df188b24d9d05271a76a85064583 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Tue, 11 Jul 2006 21:49:15 +0200
Subject: [PATCH] elevator: introduce a way to reuse rq for internal FIFO
 handling

The io schedulers can use this instead of having to allocate space for
it themselves.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 include/linux/elevator.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include')

diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 95b2a04b969c..0e7b1a733919 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -166,4 +166,16 @@ enum {
 #define rq_end_sector(rq)	((rq)->sector + (rq)->nr_sectors)
 #define rb_entry_rq(node)	rb_entry((node), struct request, rb_node)
 
+/*
+ * Hack to reuse the donelist list_head as the fifo time holder while
+ * the request is in the io scheduler. Saves an unsigned long in rq.
+ */
+#define rq_fifo_time(rq)	((unsigned long) (rq)->donelist.next)
+#define rq_set_fifo_time(rq,exp)	((rq)->donelist.next = (void *) (exp))
+#define rq_entry_fifo(ptr)	list_entry((ptr), struct request, queuelist)
+#define rq_fifo_clear(rq)	do {		\
+	list_del_init(&(rq)->queuelist);	\
+	INIT_LIST_HEAD(&(rq)->donelist);	\
+	} while (0)
+
 #endif
-- 
cgit v1.2.3


From 9e2585a8a23f3a42f815b2a638725d85a921cd65 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Fri, 28 Jul 2006 09:26:13 +0200
Subject: [PATCH] as-iosched: remove arq->is_sync member

We can track this in struct request.

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
---
 block/as-iosched.c     | 36 ++++++++++++++----------------------
 include/linux/blkdev.h |  5 +++++
 2 files changed, 19 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index c2665467950e..dca0b0563ca0 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -151,7 +151,6 @@ struct as_rq {
 
 	struct io_context *io_context;	/* The submitting task */
 
-	unsigned int is_sync;
 	enum arq_state state;
 };
 
@@ -241,7 +240,7 @@ static void as_put_io_context(struct as_rq *arq)
 
 	aic = arq->io_context->aic;
 
-	if (arq->is_sync == REQ_SYNC && aic) {
+	if (rq_is_sync(arq->request) && aic) {
 		spin_lock(&aic->lock);
 		set_bit(AS_TASK_IORUNNING, &aic->state);
 		aic->last_end_request = jiffies;
@@ -254,14 +253,13 @@ static void as_put_io_context(struct as_rq *arq)
 /*
  * rb tree support functions
  */
-#define ARQ_RB_ROOT(ad, arq)	(&(ad)->sort_list[(arq)->is_sync])
+#define RQ_RB_ROOT(ad, rq)	(&(ad)->sort_list[rq_is_sync((rq))])
 
 static void as_add_arq_rb(struct as_data *ad, struct request *rq)
 {
-	struct as_rq *arq = RQ_DATA(rq);
 	struct request *alias;
 
-	while ((unlikely(alias = elv_rb_add(ARQ_RB_ROOT(ad, arq), rq)))) {
+	while ((unlikely(alias = elv_rb_add(RQ_RB_ROOT(ad, rq), rq)))) {
 		as_move_to_dispatch(ad, RQ_DATA(alias));
 		as_antic_stop(ad);
 	}
@@ -269,7 +267,7 @@ static void as_add_arq_rb(struct as_data *ad, struct request *rq)
 
 static inline void as_del_arq_rb(struct as_data *ad, struct request *rq)
 {
-	elv_rb_del(ARQ_RB_ROOT(ad, RQ_DATA(rq)), rq);
+	elv_rb_del(RQ_RB_ROOT(ad, rq), rq);
 }
 
 /*
@@ -300,13 +298,13 @@ as_choose_req(struct as_data *ad, struct as_rq *arq1, struct as_rq *arq2)
 	if (arq2 == NULL)
 		return arq1;
 
-	data_dir = arq1->is_sync;
+	data_dir = rq_is_sync(arq1->request);
 
 	last = ad->last_sector[data_dir];
 	s1 = arq1->request->sector;
 	s2 = arq2->request->sector;
 
-	BUG_ON(data_dir != arq2->is_sync);
+	BUG_ON(data_dir != rq_is_sync(arq2->request));
 
 	/*
 	 * Strict one way elevator _except_ in the case where we allow
@@ -377,7 +375,7 @@ static struct as_rq *as_find_next_arq(struct as_data *ad, struct as_rq *arq)
 	if (rbnext)
 		next = RQ_DATA(rb_entry_rq(rbnext));
 	else {
-		const int data_dir = arq->is_sync;
+		const int data_dir = rq_is_sync(last);
 
 		rbnext = rb_first(&ad->sort_list[data_dir]);
 		if (rbnext && rbnext != &last->rb_node)
@@ -538,8 +536,7 @@ static void as_update_seekdist(struct as_data *ad, struct as_io_context *aic,
 static void as_update_iohist(struct as_data *ad, struct as_io_context *aic,
 				struct request *rq)
 {
-	struct as_rq *arq = RQ_DATA(rq);
-	int data_dir = arq->is_sync;
+	int data_dir = rq_is_sync(rq);
 	unsigned long thinktime = 0;
 	sector_t seek_dist;
 
@@ -674,7 +671,7 @@ static int as_can_break_anticipation(struct as_data *ad, struct as_rq *arq)
 		return 1;
 	}
 
-	if (arq && arq->is_sync == REQ_SYNC && as_close_req(ad, aic, arq)) {
+	if (arq && rq_is_sync(arq->request) && as_close_req(ad, aic, arq)) {
 		/*
 		 * Found a close request that is not one of ours.
 		 *
@@ -758,7 +755,7 @@ static int as_can_anticipate(struct as_data *ad, struct as_rq *arq)
  */
 static void as_update_arq(struct as_data *ad, struct as_rq *arq)
 {
-	const int data_dir = arq->is_sync;
+	const int data_dir = rq_is_sync(arq->request);
 
 	/* keep the next_arq cache up to date */
 	ad->next_arq[data_dir] = as_choose_req(ad, arq, ad->next_arq[data_dir]);
@@ -835,7 +832,7 @@ static void as_completed_request(request_queue_t *q, struct request *rq)
 	 * actually serviced. This should help devices with big TCQ windows
 	 * and writeback caches
 	 */
-	if (ad->new_batch && ad->batch_data_dir == arq->is_sync) {
+	if (ad->new_batch && ad->batch_data_dir == rq_is_sync(rq)) {
 		update_write_batch(ad);
 		ad->current_batch_expires = jiffies +
 				ad->batch_expire[REQ_SYNC];
@@ -868,7 +865,7 @@ out:
 static void as_remove_queued_request(request_queue_t *q, struct request *rq)
 {
 	struct as_rq *arq = RQ_DATA(rq);
-	const int data_dir = arq->is_sync;
+	const int data_dir = rq_is_sync(rq);
 	struct as_data *ad = q->elevator->elevator_data;
 
 	WARN_ON(arq->state != AS_RQ_QUEUED);
@@ -941,7 +938,7 @@ static inline int as_batch_expired(struct as_data *ad)
 static void as_move_to_dispatch(struct as_data *ad, struct as_rq *arq)
 {
 	struct request *rq = arq->request;
-	const int data_dir = arq->is_sync;
+	const int data_dir = rq_is_sync(rq);
 
 	BUG_ON(RB_EMPTY_NODE(&rq->rb_node));
 
@@ -1158,12 +1155,7 @@ static void as_add_request(request_queue_t *q, struct request *rq)
 
 	arq->state = AS_RQ_NEW;
 
-	if (rq_data_dir(arq->request) == READ
-			|| (arq->request->cmd_flags & REQ_RW_SYNC))
-		arq->is_sync = 1;
-	else
-		arq->is_sync = 0;
-	data_dir = arq->is_sync;
+	data_dir = rq_is_sync(rq);
 
 	arq->io_context = as_get_io_context();
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index a905c4934a55..55ef6efe3eb5 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -531,6 +531,11 @@ enum {
 
 #define rq_data_dir(rq)		((rq)->cmd_flags & 1)
 
+/*
+ * We regard a request as sync, if it's a READ or a SYNC write.
+ */
+#define rq_is_sync(rq)		(rq_data_dir((rq)) == READ || (rq)->cmd_flags & REQ_RW_SYNC)
+
 static inline int blk_queue_full(struct request_queue *q, int rw)
 {
 	if (rw == READ)
-- 
cgit v1.2.3


From ff7d145fd911266ae42af7552edc32681c01addb Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Wed, 12 Jul 2006 14:04:37 +0200
Subject: [PATCH] Add one more pointer to struct request for IO scheduler usage

Then we have enough room in the request to get rid of the dynamic
allocations in CFQ/AS.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 include/linux/blkdev.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 55ef6efe3eb5..d2dc17151f6c 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -232,7 +232,13 @@ struct request {
 	struct hlist_node hash;	/* merge hash */
 	struct rb_node rb_node;	/* sort/lookup */
 
+	/*
+	 * two pointers are available for the IO schedulers, if they need
+	 * more they have to dynamically allocate it.
+	 */
 	void *elevator_private;
+	void *elevator_private2;
+
 	void *completion_data;
 
 	int rq_status;	/* should split this into a few status bits */
-- 
cgit v1.2.3


From c00895ab2f08df7044e58ee01c38bf0a661ea0eb Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Sat, 30 Sep 2006 20:29:12 +0200
Subject: [PATCH] Remove ->waiting member from struct request

As the comments indicates in blkdev.h, we can fold it into ->end_io_data
usage as that is really what ->waiting is. Fixup the users of
blk_end_sync_rq().

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/elevator.c          |  3 +--
 block/ll_rw_blk.c         | 13 ++++++-------
 drivers/block/DAC960.c    |  2 +-
 drivers/block/paride/pd.c |  3 +--
 drivers/block/pktcdvd.c   |  2 +-
 drivers/ide/ide-io.c      | 13 ++++++-------
 drivers/ide/ide-tape.c    |  2 +-
 drivers/ide/ide.c         |  4 ++--
 include/linux/blkdev.h    |  3 +--
 9 files changed, 20 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/block/elevator.c b/block/elevator.c
index cbbc36ba016a..924b81b08f86 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -67,8 +67,7 @@ inline int elv_rq_merge_ok(struct request *rq, struct bio *bio)
 	/*
 	 * same device and no special stuff set, merge is ok
 	 */
-	if (rq->rq_disk == bio->bi_bdev->bd_disk &&
-	    !rq->waiting && !rq->special)
+	if (rq->rq_disk == bio->bi_bdev->bd_disk && !rq->special)
 		return 1;
 
 	return 0;
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index d388486e98bb..3b6aad2affd8 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -291,7 +291,6 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 	rq->buffer = NULL;
 	rq->ref_count = 1;
 	rq->q = q;
-	rq->waiting = NULL;
 	rq->special = NULL;
 	rq->data_len = 0;
 	rq->data = NULL;
@@ -451,6 +450,7 @@ static void queue_flush(request_queue_t *q, unsigned which)
 	rq->cmd_flags = REQ_HARDBARRIER;
 	rq_init(q, rq);
 	rq->elevator_private = NULL;
+	rq->elevator_private2 = NULL;
 	rq->rq_disk = q->bar_rq.rq_disk;
 	rq->rl = NULL;
 	rq->end_io = end_io;
@@ -479,6 +479,7 @@ static inline struct request *start_ordered(request_queue_t *q,
 		rq->cmd_flags |= REQ_RW;
 	rq->cmd_flags |= q->ordered & QUEUE_ORDERED_FUA ? REQ_FUA : 0;
 	rq->elevator_private = NULL;
+	rq->elevator_private2 = NULL;
 	rq->rl = NULL;
 	init_request_from_bio(rq, q->orig_bar_rq->bio);
 	rq->end_io = bar_end_io;
@@ -2569,10 +2570,9 @@ int blk_execute_rq(request_queue_t *q, struct gendisk *bd_disk,
 		rq->sense_len = 0;
 	}
 
-	rq->waiting = &wait;
+	rq->end_io_data = &wait;
 	blk_execute_rq_nowait(q, bd_disk, rq, at_head, blk_end_sync_rq);
 	wait_for_completion(&wait);
-	rq->waiting = NULL;
 
 	if (rq->errors)
 		err = -EIO;
@@ -2736,9 +2736,9 @@ EXPORT_SYMBOL(blk_put_request);
  */
 void blk_end_sync_rq(struct request *rq, int error)
 {
-	struct completion *waiting = rq->waiting;
+	struct completion *waiting = rq->end_io_data;
 
-	rq->waiting = NULL;
+	rq->end_io_data = NULL;
 	__blk_put_request(rq->q, rq);
 
 	/*
@@ -2801,7 +2801,7 @@ static int attempt_merge(request_queue_t *q, struct request *req,
 
 	if (rq_data_dir(req) != rq_data_dir(next)
 	    || req->rq_disk != next->rq_disk
-	    || next->waiting || next->special)
+	    || next->special)
 		return 0;
 
 	/*
@@ -2886,7 +2886,6 @@ static void init_request_from_bio(struct request *req, struct bio *bio)
 	req->nr_phys_segments = bio_phys_segments(req->q, bio);
 	req->nr_hw_segments = bio_hw_segments(req->q, bio);
 	req->buffer = bio_data(bio);	/* see ->buffer comment above */
-	req->waiting = NULL;
 	req->bio = req->biotail = bio;
 	req->ioprio = bio_prio(bio);
 	req->rq_disk = bio->bi_bdev->bd_disk;
diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c
index a360215dbce7..2568640430fb 100644
--- a/drivers/block/DAC960.c
+++ b/drivers/block/DAC960.c
@@ -3331,7 +3331,7 @@ static int DAC960_process_queue(DAC960_Controller_T *Controller, struct request_
 		Command->DmaDirection = PCI_DMA_TODEVICE;
 		Command->CommandType = DAC960_WriteCommand;
 	}
-	Command->Completion = Request->waiting;
+	Command->Completion = Request->end_io_data;
 	Command->LogicalDriveNumber = (long)Request->rq_disk->private_data;
 	Command->BlockNumber = Request->sector;
 	Command->BlockCount = Request->nr_sectors;
diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c
index 12ff1a274d91..500d2ebb41e4 100644
--- a/drivers/block/paride/pd.c
+++ b/drivers/block/paride/pd.c
@@ -722,11 +722,10 @@ static int pd_special_command(struct pd_unit *disk,
 	rq.rq_status = RQ_ACTIVE;
 	rq.rq_disk = disk->gd;
 	rq.ref_count = 1;
-	rq.waiting = &wait;
+	rq.end_io_data = &wait;
 	rq.end_io = blk_end_sync_rq;
 	blk_insert_request(disk->gd->queue, &rq, 0, func);
 	wait_for_completion(&wait);
-	rq.waiting = NULL;
 	if (rq.errors)
 		err = -EIO;
 	blk_put_request(&rq);
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 42891d2b054e..888d1aceeeff 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -375,7 +375,7 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	rq->cmd_len = COMMAND_SIZE(rq->cmd[0]);
 
 	rq->ref_count++;
-	rq->waiting = &wait;
+	rq->end_io_data = &wait;
 	rq->end_io = blk_end_sync_rq;
 	elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 1);
 	generic_unplug_device(q);
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 3436b1f104eb..a3ffb04436bd 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -141,7 +141,7 @@ enum {
 
 static void ide_complete_power_step(ide_drive_t *drive, struct request *rq, u8 stat, u8 error)
 {
-	struct request_pm_state *pm = rq->end_io_data;
+	struct request_pm_state *pm = rq->data;
 
 	if (drive->media != ide_disk)
 		return;
@@ -164,7 +164,7 @@ static void ide_complete_power_step(ide_drive_t *drive, struct request *rq, u8 s
 
 static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *rq)
 {
-	struct request_pm_state *pm = rq->end_io_data;
+	struct request_pm_state *pm = rq->data;
 	ide_task_t *args = rq->special;
 
 	memset(args, 0, sizeof(*args));
@@ -421,7 +421,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			}
 		}
 	} else if (blk_pm_request(rq)) {
-		struct request_pm_state *pm = rq->end_io_data;
+		struct request_pm_state *pm = rq->data;
 #ifdef DEBUG_PM
 		printk("%s: complete_power_step(step: %d, stat: %x, err: %x)\n",
 			drive->name, rq->pm->pm_step, stat, err);
@@ -933,7 +933,7 @@ done:
 
 static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 {
-	struct request_pm_state *pm = rq->end_io_data;
+	struct request_pm_state *pm = rq->data;
 
 	if (blk_pm_suspend_request(rq) &&
 	    pm->pm_step == ide_pm_state_start_suspend)
@@ -1018,7 +1018,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 		    rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
 			return execute_drive_cmd(drive, rq);
 		else if (blk_pm_request(rq)) {
-			struct request_pm_state *pm = rq->end_io_data;
+			struct request_pm_state *pm = rq->data;
 #ifdef DEBUG_PM
 			printk("%s: start_power_step(step: %d)\n",
 				drive->name, rq->pm->pm_step);
@@ -1718,7 +1718,7 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 	 */
 	if (must_wait) {
 		rq->ref_count++;
-		rq->waiting = &wait;
+		rq->end_io_data = &wait;
 		rq->end_io = blk_end_sync_rq;
 	}
 
@@ -1736,7 +1736,6 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 	err = 0;
 	if (must_wait) {
 		wait_for_completion(&wait);
-		rq->waiting = NULL;
 		if (rq->errors)
 			err = -EIO;
 
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 643e4b9ac651..66f9678d2f10 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -2773,7 +2773,7 @@ static void idetape_wait_for_request (ide_drive_t *drive, struct request *rq)
 		return;
 	}
 #endif /* IDETAPE_DEBUG_BUGS */
-	rq->waiting = &wait;
+	rq->end_io_data = &wait;
 	rq->end_io = blk_end_sync_rq;
 	spin_unlock_irq(&tape->spinlock);
 	wait_for_completion(&wait);
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 9384a3fdde6c..2b1a1389c318 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -1219,7 +1219,7 @@ static int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 	memset(&args, 0, sizeof(args));
 	rq.cmd_type = REQ_TYPE_PM_SUSPEND;
 	rq.special = &args;
-	rq.end_io_data = &rqpm;
+	rq.data = &rqpm;
 	rqpm.pm_step = ide_pm_state_start_suspend;
 	if (mesg.event == PM_EVENT_PRETHAW)
 		mesg.event = PM_EVENT_FREEZE;
@@ -1240,7 +1240,7 @@ static int generic_ide_resume(struct device *dev)
 	memset(&args, 0, sizeof(args));
 	rq.cmd_type = REQ_TYPE_PM_RESUME;
 	rq.special = &args;
-	rq.end_io_data = &rqpm;
+	rq.data = &rqpm;
 	rqpm.pm_step = ide_pm_state_start_resume;
 	rqpm.pm_state = PM_EVENT_ON;
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d2dc17151f6c..604f23189097 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -266,7 +266,6 @@ struct request {
 	request_queue_t *q;
 	struct request_list *rl;
 
-	struct completion *waiting;
 	void *special;
 	char *buffer;
 
@@ -285,7 +284,7 @@ struct request {
 	int retries;
 
 	/*
-	 * completion callback. end_io_data should be folded in with waiting
+	 * completion callback.
 	 */
 	rq_end_io_fn *end_io;
 	void *end_io_data;
-- 
cgit v1.2.3


From 49171e5c6f414d49a061b5c1c84967c2eb569822 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Thu, 10 Aug 2006 08:59:11 +0200
Subject: [PATCH] Remove struct request_list from struct request

It is always identical to &q->rq, and we only use it for detecting
whether this request came out of our mempool or not. So replace it
with an additional ->flags bit flag.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/ll_rw_blk.c      | 10 ++--------
 include/linux/blkdev.h |  3 ++-
 2 files changed, 4 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 3b6aad2affd8..f7462502bfdf 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -452,7 +452,6 @@ static void queue_flush(request_queue_t *q, unsigned which)
 	rq->elevator_private = NULL;
 	rq->elevator_private2 = NULL;
 	rq->rq_disk = q->bar_rq.rq_disk;
-	rq->rl = NULL;
 	rq->end_io = end_io;
 	q->prepare_flush_fn(q, rq);
 
@@ -480,7 +479,6 @@ static inline struct request *start_ordered(request_queue_t *q,
 	rq->cmd_flags |= q->ordered & QUEUE_ORDERED_FUA ? REQ_FUA : 0;
 	rq->elevator_private = NULL;
 	rq->elevator_private2 = NULL;
-	rq->rl = NULL;
 	init_request_from_bio(rq, q->orig_bar_rq->bio);
 	rq->end_io = bar_end_io;
 
@@ -2018,7 +2016,7 @@ blk_alloc_request(request_queue_t *q, int rw, struct bio *bio,
 	 * first three bits are identical in rq->cmd_flags and bio->bi_rw,
 	 * see bio.h and blkdev.h
 	 */
-	rq->cmd_flags = rw;
+	rq->cmd_flags = rw | REQ_ALLOCED;
 
 	if (priv) {
 		if (unlikely(elv_set_request(q, rq, bio, gfp_mask))) {
@@ -2196,7 +2194,6 @@ rq_starved:
 		ioc->nr_batch_requests--;
 	
 	rq_init(q, rq);
-	rq->rl = rl;
 
 	blk_add_trace_generic(q, bio, rw, BLK_TA_GETRQ);
 out:
@@ -2681,8 +2678,6 @@ EXPORT_SYMBOL_GPL(disk_round_stats);
  */
 void __blk_put_request(request_queue_t *q, struct request *req)
 {
-	struct request_list *rl = req->rl;
-
 	if (unlikely(!q))
 		return;
 	if (unlikely(--req->ref_count))
@@ -2691,13 +2686,12 @@ void __blk_put_request(request_queue_t *q, struct request *req)
 	elv_completed_request(q, req);
 
 	req->rq_status = RQ_INACTIVE;
-	req->rl = NULL;
 
 	/*
 	 * Request may not have originated from ll_rw_blk. if not,
 	 * it didn't come out of our reserved rq pools
 	 */
-	if (rl) {
+	if (req->cmd_flags & REQ_ALLOCED) {
 		int rw = rq_data_dir(req);
 		int priv = req->cmd_flags & REQ_ELVPRIV;
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 604f23189097..d4c1dd046e27 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -180,6 +180,7 @@ enum rq_flag_bits {
 	__REQ_PREEMPT,		/* set for "ide_preempt" requests */
 	__REQ_ORDERED_COLOR,	/* is before or after barrier */
 	__REQ_RW_SYNC,		/* request is sync (O_DIRECT) */
+	__REQ_ALLOCED,		/* request came from our alloc pool */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -199,6 +200,7 @@ enum rq_flag_bits {
 #define REQ_PREEMPT	(1 << __REQ_PREEMPT)
 #define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
 #define REQ_RW_SYNC	(1 << __REQ_RW_SYNC)
+#define REQ_ALLOCED	(1 << __REQ_ALLOCED)
 
 #define BLK_MAX_CDB	16
 
@@ -264,7 +266,6 @@ struct request {
 
 	int ref_count;
 	request_queue_t *q;
-	struct request_list *rl;
 
 	void *special;
 	char *buffer;
-- 
cgit v1.2.3


From cdd6026217c0e4cda2efce1bdc318661bef1f66f Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Fri, 28 Jul 2006 09:32:07 +0200
Subject: [PATCH] Remove ->rq_status from struct request

After Christophs SCSI change, the only usage left is RQ_ACTIVE
and RQ_INACTIVE. The block layer sets RQ_INACTIVE right before freeing
the request, so any check for RQ_INACTIVE in a driver is a bug and
indicates use-after-free.

So kill/clean the remaining users, straight forward.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 arch/um/drivers/ubd_kern.c |  2 --
 block/ll_rw_blk.c          |  3 ---
 drivers/block/paride/pd.c  |  1 -
 drivers/block/swim3.c      |  4 ++--
 drivers/block/swim_iop.c   |  4 ++--
 drivers/fc4/fc.c           |  1 -
 drivers/ide/ide-floppy.c   |  3 +--
 drivers/ide/ide-io.c       |  1 -
 drivers/ide/ide-tape.c     |  4 ++--
 drivers/scsi/ide-scsi.c    |  2 +-
 drivers/scsi/scsi.c        |  2 +-
 include/linux/blkdev.h     | 13 +++++--------
 12 files changed, 14 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 5fa4c8e258a4..fda4a3940698 100644
--- a/arch/um/drivers/ubd_kern.c
+++ b/arch/um/drivers/ubd_kern.c
@@ -981,8 +981,6 @@ static int prepare_request(struct request *req, struct io_thread_req *io_req)
 	__u64 offset;
 	int len;
 
-	if(req->rq_status == RQ_INACTIVE) return(1);
-
 	/* This should be impossible now */
 	if((rq_data_dir(req) == WRITE) && !dev->openflags.w){
 		printk("Write attempted on readonly ubd device %s\n",
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index f7462502bfdf..b94a396aa624 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -283,7 +283,6 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 	INIT_LIST_HEAD(&rq->donelist);
 
 	rq->errors = 0;
-	rq->rq_status = RQ_ACTIVE;
 	rq->bio = rq->biotail = NULL;
 	INIT_HLIST_NODE(&rq->hash);
 	RB_CLEAR_NODE(&rq->rb_node);
@@ -2685,8 +2684,6 @@ void __blk_put_request(request_queue_t *q, struct request *req)
 
 	elv_completed_request(q, req);
 
-	req->rq_status = RQ_INACTIVE;
-
 	/*
 	 * Request may not have originated from ll_rw_blk. if not,
 	 * it didn't come out of our reserved rq pools
diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c
index 500d2ebb41e4..38578b9dbfd1 100644
--- a/drivers/block/paride/pd.c
+++ b/drivers/block/paride/pd.c
@@ -719,7 +719,6 @@ static int pd_special_command(struct pd_unit *disk,
 
 	memset(&rq, 0, sizeof(rq));
 	rq.errors = 0;
-	rq.rq_status = RQ_ACTIVE;
 	rq.rq_disk = disk->gd;
 	rq.ref_count = 1;
 	rq.end_io_data = &wait;
diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c
index cc42e762396f..f2305ee792a1 100644
--- a/drivers/block/swim3.c
+++ b/drivers/block/swim3.c
@@ -319,8 +319,8 @@ static void start_request(struct floppy_state *fs)
 		printk("do_fd_req: dev=%s cmd=%d sec=%ld nr_sec=%ld buf=%p\n",
 		       req->rq_disk->disk_name, req->cmd,
 		       (long)req->sector, req->nr_sectors, req->buffer);
-		printk("           rq_status=%d errors=%d current_nr_sectors=%ld\n",
-		       req->rq_status, req->errors, req->current_nr_sectors);
+		printk("           errors=%d current_nr_sectors=%ld\n",
+		       req->errors, req->current_nr_sectors);
 #endif
 
 		if (req->sector < 0 || req->sector >= fs->total_secs) {
diff --git a/drivers/block/swim_iop.c b/drivers/block/swim_iop.c
index 89e3c2f8b776..dfda796eba56 100644
--- a/drivers/block/swim_iop.c
+++ b/drivers/block/swim_iop.c
@@ -529,8 +529,8 @@ static void start_request(struct floppy_state *fs)
 		printk("do_fd_req: dev=%s cmd=%d sec=%ld nr_sec=%ld buf=%p\n",
 		       CURRENT->rq_disk->disk_name, CURRENT->cmd,
 		       CURRENT->sector, CURRENT->nr_sectors, CURRENT->buffer);
-		printk("           rq_status=%d errors=%d current_nr_sectors=%ld\n",
-		       CURRENT->rq_status, CURRENT->errors, CURRENT->current_nr_sectors);
+		printk("           errors=%d current_nr_sectors=%ld\n",
+		      CURRENT->errors, CURRENT->current_nr_sectors);
 #endif
 
 		if (CURRENT->sector < 0 || CURRENT->sector >= fs->total_secs) {
diff --git a/drivers/fc4/fc.c b/drivers/fc4/fc.c
index 1a159e8843ca..22d17474755f 100644
--- a/drivers/fc4/fc.c
+++ b/drivers/fc4/fc.c
@@ -974,7 +974,6 @@ int fcp_scsi_dev_reset(Scsi_Cmnd *SCpnt)
 	 */
 
 	fc->rst_pkt->device->host->eh_action = &sem;
-	fc->rst_pkt->request->rq_status = RQ_SCSI_BUSY;
 
 	fc->rst_pkt->done = fcp_scsi_reset_done;
 
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 0edc32204915..8ccee9c769f8 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -1281,8 +1281,7 @@ static ide_startstop_t idefloppy_do_request (ide_drive_t *drive, struct request
 	idefloppy_pc_t *pc;
 	unsigned long block = (unsigned long)block_s;
 
-	debug_log(KERN_INFO "rq_status: %d, dev: %s, flags: %lx, errors: %d\n",
-			rq->rq_status,
+	debug_log(KERN_INFO "dev: %s, flags: %lx, errors: %d\n",
 			rq->rq_disk ? rq->rq_disk->disk_name : "?",
 			rq->flags, rq->errors);
 	debug_log(KERN_INFO "sector: %ld, nr_sectors: %ld, "
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index a3ffb04436bd..38479a29d3e1 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1710,7 +1710,6 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 	int must_wait = (action == ide_wait || action == ide_head_wait);
 
 	rq->errors = 0;
-	rq->rq_status = RQ_ACTIVE;
 
 	/*
 	 * we need to hold an extra reference to request for safe inspection
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 66f9678d2f10..2ebc3760f261 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -2423,8 +2423,8 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 #if IDETAPE_DEBUG_LOG
 #if 0
 	if (tape->debug_level >= 5)
-		printk(KERN_INFO "ide-tape: rq_status: %d, "
-			"dev: %s, cmd: %ld, errors: %d\n", rq->rq_status,
+		printk(KERN_INFO "ide-tape:  %d, "
+			"dev: %s, cmd: %ld, errors: %d\n",
 			 rq->rq_disk->disk_name, rq->cmd[0], rq->errors);
 #endif
 	if (tape->debug_level >= 2)
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 65b19695ebe2..1427a41e8441 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -708,7 +708,7 @@ static ide_startstop_t idescsi_issue_pc (ide_drive_t *drive, idescsi_pc_t *pc)
 static ide_startstop_t idescsi_do_request (ide_drive_t *drive, struct request *rq, sector_t block)
 {
 #if IDESCSI_DEBUG_LOG
-	printk (KERN_INFO "rq_status: %d, dev: %s, cmd: %x, errors: %d\n",rq->rq_status, rq->rq_disk->disk_name,rq->cmd[0],rq->errors);
+	printk (KERN_INFO "dev: %s, cmd: %x, errors: %d\n", rq->rq_disk->disk_name,rq->cmd[0],rq->errors);
 	printk (KERN_INFO "sector: %ld, nr_sectors: %ld, current_nr_sectors: %d\n",rq->sector,rq->nr_sectors,rq->current_nr_sectors);
 #endif /* IDESCSI_DEBUG_LOG */
 
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 7a054f9d1ee3..12f6639dda2d 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -1065,7 +1065,7 @@ int scsi_device_cancel(struct scsi_device *sdev, int recovery)
 
 	spin_lock_irqsave(&sdev->list_lock, flags);
 	list_for_each_entry(scmd, &sdev->cmd_list, list) {
-		if (scmd->request && scmd->request->rq_status != RQ_INACTIVE) {
+		if (scmd->request) {
 			/*
 			 * If we are unable to remove the timer, it means
 			 * that the command has already timed out or
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d4c1dd046e27..8a3e309e0842 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -243,8 +243,6 @@ struct request {
 
 	void *completion_data;
 
-	int rq_status;	/* should split this into a few status bits */
-	int errors;
 	struct gendisk *rq_disk;
 	unsigned long start_time;
 
@@ -262,14 +260,16 @@ struct request {
 
 	unsigned short ioprio;
 
-	int tag;
-
-	int ref_count;
 	request_queue_t *q;
 
 	void *special;
 	char *buffer;
 
+	int tag;
+	int errors;
+
+	int ref_count;
+
 	/*
 	 * when request is used as a packet command carrier
 	 */
@@ -456,9 +456,6 @@ struct request_queue
 	struct mutex		sysfs_lock;
 };
 
-#define RQ_INACTIVE		(-1)
-#define RQ_ACTIVE		1
-
 #define QUEUE_FLAG_CLUSTER	0	/* cluster several segments into 1 */
 #define QUEUE_FLAG_QUEUED	1	/* uses generic tag queueing */
 #define QUEUE_FLAG_STOPPED	2	/* queue is stopped */
-- 
cgit v1.2.3


From cb78b285c8f9d59b0d4e4f6a54c2977ce1d9b880 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Fri, 28 Jul 2006 09:32:57 +0200
Subject: [PATCH] Drop useless bio passing in may_queue/set_request API

It's not needed for anything, so kill the bio passing.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/as-iosched.c       | 2 +-
 block/cfq-iosched.c      | 5 ++---
 block/elevator.c         | 9 ++++-----
 block/ll_rw_blk.c        | 9 ++++-----
 include/linux/elevator.h | 8 ++++----
 5 files changed, 15 insertions(+), 18 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index 02eb9333898f..66015bc79e6f 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -1284,7 +1284,7 @@ static void as_work_handler(void *data)
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
-static int as_may_queue(request_queue_t *q, int rw, struct bio *bio)
+static int as_may_queue(request_queue_t *q, int rw)
 {
 	int ret = ELV_MQUEUE_MAY;
 	struct as_data *ad = q->elevator->elevator_data;
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 3c5fd9c2c205..2ac35aacbbf9 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1752,7 +1752,7 @@ __cfq_may_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 	return ELV_MQUEUE_MAY;
 }
 
-static int cfq_may_queue(request_queue_t *q, int rw, struct bio *bio)
+static int cfq_may_queue(request_queue_t *q, int rw)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct task_struct *tsk = current;
@@ -1817,8 +1817,7 @@ static void cfq_put_request(request_queue_t *q, struct request *rq)
  * Allocate cfq data structures associated with this request.
  */
 static int
-cfq_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
-		gfp_t gfp_mask)
+cfq_set_request(request_queue_t *q, struct request *rq, gfp_t gfp_mask)
 {
 	struct cfq_data *cfqd = q->elevator->elevator_data;
 	struct task_struct *tsk = current;
diff --git a/block/elevator.c b/block/elevator.c
index 924b81b08f86..788d2d81994c 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -796,13 +796,12 @@ struct request *elv_former_request(request_queue_t *q, struct request *rq)
 	return NULL;
 }
 
-int elv_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
-		    gfp_t gfp_mask)
+int elv_set_request(request_queue_t *q, struct request *rq, gfp_t gfp_mask)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_set_req_fn)
-		return e->ops->elevator_set_req_fn(q, rq, bio, gfp_mask);
+		return e->ops->elevator_set_req_fn(q, rq, gfp_mask);
 
 	rq->elevator_private = NULL;
 	return 0;
@@ -816,12 +815,12 @@ void elv_put_request(request_queue_t *q, struct request *rq)
 		e->ops->elevator_put_req_fn(q, rq);
 }
 
-int elv_may_queue(request_queue_t *q, int rw, struct bio *bio)
+int elv_may_queue(request_queue_t *q, int rw)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_may_queue_fn)
-		return e->ops->elevator_may_queue_fn(q, rw, bio);
+		return e->ops->elevator_may_queue_fn(q, rw);
 
 	return ELV_MQUEUE_MAY;
 }
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index b94a396aa624..b1ea941f6dc3 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -2003,8 +2003,7 @@ static inline void blk_free_request(request_queue_t *q, struct request *rq)
 }
 
 static inline struct request *
-blk_alloc_request(request_queue_t *q, int rw, struct bio *bio,
-		  int priv, gfp_t gfp_mask)
+blk_alloc_request(request_queue_t *q, int rw, int priv, gfp_t gfp_mask)
 {
 	struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask);
 
@@ -2018,7 +2017,7 @@ blk_alloc_request(request_queue_t *q, int rw, struct bio *bio,
 	rq->cmd_flags = rw | REQ_ALLOCED;
 
 	if (priv) {
-		if (unlikely(elv_set_request(q, rq, bio, gfp_mask))) {
+		if (unlikely(elv_set_request(q, rq, gfp_mask))) {
 			mempool_free(rq, q->rq.rq_pool);
 			return NULL;
 		}
@@ -2109,7 +2108,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 	struct io_context *ioc = NULL;
 	int may_queue, priv;
 
-	may_queue = elv_may_queue(q, rw, bio);
+	may_queue = elv_may_queue(q, rw);
 	if (may_queue == ELV_MQUEUE_NO)
 		goto rq_starved;
 
@@ -2157,7 +2156,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 
 	spin_unlock_irq(q->queue_lock);
 
-	rq = blk_alloc_request(q, rw, bio, priv, gfp_mask);
+	rq = blk_alloc_request(q, rw, priv, gfp_mask);
 	if (unlikely(!rq)) {
 		/*
 		 * Allocation failed presumably due to memory. Undo anything
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 0e7b1a733919..cc81645a3e18 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -14,9 +14,9 @@ typedef void (elevator_add_req_fn) (request_queue_t *, struct request *);
 typedef int (elevator_queue_empty_fn) (request_queue_t *);
 typedef struct request *(elevator_request_list_fn) (request_queue_t *, struct request *);
 typedef void (elevator_completed_req_fn) (request_queue_t *, struct request *);
-typedef int (elevator_may_queue_fn) (request_queue_t *, int, struct bio *);
+typedef int (elevator_may_queue_fn) (request_queue_t *, int);
 
-typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, struct bio *, gfp_t);
+typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, gfp_t);
 typedef void (elevator_put_req_fn) (request_queue_t *, struct request *);
 typedef void (elevator_activate_req_fn) (request_queue_t *, struct request *);
 typedef void (elevator_deactivate_req_fn) (request_queue_t *, struct request *);
@@ -105,9 +105,9 @@ extern struct request *elv_former_request(request_queue_t *, struct request *);
 extern struct request *elv_latter_request(request_queue_t *, struct request *);
 extern int elv_register_queue(request_queue_t *q);
 extern void elv_unregister_queue(request_queue_t *q);
-extern int elv_may_queue(request_queue_t *, int, struct bio *);
+extern int elv_may_queue(request_queue_t *, int);
 extern void elv_completed_request(request_queue_t *, struct request *);
-extern int elv_set_request(request_queue_t *, struct request *, struct bio *, gfp_t);
+extern int elv_set_request(request_queue_t *, struct request *, gfp_t);
 extern void elv_put_request(request_queue_t *, struct request *);
 
 /*
-- 
cgit v1.2.3


From e6a1c874a064e7d07f24986aba7cd537b7f4a25d Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Thu, 10 Aug 2006 09:00:21 +0200
Subject: [PATCH] struct request: shrink and optimize some more

Move some members around and unionize completion_data and rb_node since
they cannot ever be used at the same time.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 include/linux/blkdev.h | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

(limited to 'include')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8a3e309e0842..a1e288069e2e 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -211,6 +211,8 @@ struct request {
 	struct list_head queuelist;
 	struct list_head donelist;
 
+	request_queue_t *q;
+
 	unsigned int cmd_flags;
 	enum rq_cmd_type_bits cmd_type;
 
@@ -219,12 +221,12 @@ struct request {
 	 */
 
 	sector_t sector;		/* next sector to submit */
+	sector_t hard_sector;		/* next sector to complete */
 	unsigned long nr_sectors;	/* no. of sectors left to submit */
+	unsigned long hard_nr_sectors;	/* no. of sectors left to complete */
 	/* no. of sectors left to submit in the current segment */
 	unsigned int current_nr_sectors;
 
-	sector_t hard_sector;		/* next sector to complete */
-	unsigned long hard_nr_sectors;	/* no. of sectors left to complete */
 	/* no. of sectors left to complete in the current segment */
 	unsigned int hard_cur_sectors;
 
@@ -232,7 +234,15 @@ struct request {
 	struct bio *biotail;
 
 	struct hlist_node hash;	/* merge hash */
-	struct rb_node rb_node;	/* sort/lookup */
+	/*
+	 * The rb_node is only used inside the io scheduler, requests
+	 * are pruned when moved to the dispatch queue. So let the
+	 * completion_data share space with the rb_node.
+	 */
+	union {
+		struct rb_node rb_node;	/* sort/lookup */
+		void *completion_data;
+	};
 
 	/*
 	 * two pointers are available for the IO schedulers, if they need
@@ -241,8 +251,6 @@ struct request {
 	void *elevator_private;
 	void *elevator_private2;
 
-	void *completion_data;
-
 	struct gendisk *rq_disk;
 	unsigned long start_time;
 
@@ -260,8 +268,6 @@ struct request {
 
 	unsigned short ioprio;
 
-	request_queue_t *q;
-
 	void *special;
 	char *buffer;
 
-- 
cgit v1.2.3


From fc46379daf90dce57bf765c81d3b39f55150aac2 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Tue, 29 Aug 2006 09:05:44 +0200
Subject: [PATCH] cfq-iosched: kill cfq_exit_lock

cfq_exit_lock is protecting two things now:

- The per-ioc rbtree of cfq_io_contexts

- The per-cfqd linked list of cfq_io_contexts

The per-cfqd linked list can be protected by the queue lock, as it is (by
definition) per cfqd as the queue lock is.

The per-ioc rbtree is mainly used and updated by the process itself only.
The only outside use is the io priority changing. If we move the
priority changing to not browsing the rbtree, we can remove any locking
from the rbtree updates and lookup completely. Let the sys_ioprio syscall
just mark processes as having the iopriority changed and lazily update
the private cfq io contexts the next time io is queued, and we can
remove this locking as well.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/cfq-iosched.c    | 54 ++++++++++++++++----------------------------------
 block/ll_rw_blk.c      |  2 +-
 fs/ioprio.c            |  4 ++--
 include/linux/blkdev.h |  2 +-
 4 files changed, 21 insertions(+), 41 deletions(-)

(limited to 'include')

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index ec24284e9d39..33e0b0c5e31d 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -31,8 +31,6 @@ static int cfq_slice_idle = HZ / 125;
 
 #define CFQ_KEY_ASYNC		(0)
 
-static DEFINE_SPINLOCK(cfq_exit_lock);
-
 /*
  * for the hash of cfqq inside the cfqd
  */
@@ -1084,12 +1082,6 @@ static void cfq_free_io_context(struct io_context *ioc)
 		complete(ioc_gone);
 }
 
-static void cfq_trim(struct io_context *ioc)
-{
-	ioc->set_ioprio = NULL;
-	cfq_free_io_context(ioc);
-}
-
 static void cfq_exit_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 {
 	if (unlikely(cfqq == cfqd->active_queue))
@@ -1101,6 +1093,10 @@ static void cfq_exit_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 static void __cfq_exit_single_io_context(struct cfq_data *cfqd,
 					 struct cfq_io_context *cic)
 {
+	list_del_init(&cic->queue_list);
+	smp_wmb();
+	cic->key = NULL;
+
 	if (cic->cfqq[ASYNC]) {
 		cfq_exit_cfqq(cfqd, cic->cfqq[ASYNC]);
 		cic->cfqq[ASYNC] = NULL;
@@ -1110,9 +1106,6 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd,
 		cfq_exit_cfqq(cfqd, cic->cfqq[SYNC]);
 		cic->cfqq[SYNC] = NULL;
 	}
-
-	cic->key = NULL;
-	list_del_init(&cic->queue_list);
 }
 
 
@@ -1123,27 +1116,23 @@ static void cfq_exit_single_io_context(struct cfq_io_context *cic)
 {
 	struct cfq_data *cfqd = cic->key;
 
-	WARN_ON(!irqs_disabled());
-
 	if (cfqd) {
 		request_queue_t *q = cfqd->queue;
 
-		spin_lock(q->queue_lock);
+		spin_lock_irq(q->queue_lock);
 		__cfq_exit_single_io_context(cfqd, cic);
-		spin_unlock(q->queue_lock);
+		spin_unlock_irq(q->queue_lock);
 	}
 }
 
 static void cfq_exit_io_context(struct io_context *ioc)
 {
 	struct cfq_io_context *__cic;
-	unsigned long flags;
 	struct rb_node *n;
 
 	/*
 	 * put the reference this task is holding to the various queues
 	 */
-	spin_lock_irqsave(&cfq_exit_lock, flags);
 
 	n = rb_first(&ioc->cic_root);
 	while (n != NULL) {
@@ -1152,8 +1141,6 @@ static void cfq_exit_io_context(struct io_context *ioc)
 		cfq_exit_single_io_context(__cic);
 		n = rb_next(n);
 	}
-
-	spin_unlock_irqrestore(&cfq_exit_lock, flags);
 }
 
 static struct cfq_io_context *
@@ -1248,15 +1235,12 @@ static inline void changed_ioprio(struct cfq_io_context *cic)
 	spin_unlock(cfqd->queue->queue_lock);
 }
 
-/*
- * callback from sys_ioprio_set, irqs are disabled
- */
-static int cfq_ioc_set_ioprio(struct io_context *ioc, unsigned int ioprio)
+static void cfq_ioc_set_ioprio(struct io_context *ioc)
 {
 	struct cfq_io_context *cic;
 	struct rb_node *n;
 
-	spin_lock(&cfq_exit_lock);
+	ioc->ioprio_changed = 0;
 
 	n = rb_first(&ioc->cic_root);
 	while (n != NULL) {
@@ -1265,10 +1249,6 @@ static int cfq_ioc_set_ioprio(struct io_context *ioc, unsigned int ioprio)
 		changed_ioprio(cic);
 		n = rb_next(n);
 	}
-
-	spin_unlock(&cfq_exit_lock);
-
-	return 0;
 }
 
 static struct cfq_queue *
@@ -1336,10 +1316,8 @@ out:
 static void
 cfq_drop_dead_cic(struct io_context *ioc, struct cfq_io_context *cic)
 {
-	spin_lock(&cfq_exit_lock);
+	WARN_ON(!list_empty(&cic->queue_list));
 	rb_erase(&cic->rb_node, &ioc->cic_root);
-	list_del_init(&cic->queue_list);
-	spin_unlock(&cfq_exit_lock);
 	kmem_cache_free(cfq_ioc_pool, cic);
 	atomic_dec(&ioc_count);
 }
@@ -1385,7 +1363,6 @@ cfq_cic_link(struct cfq_data *cfqd, struct io_context *ioc,
 	cic->ioc = ioc;
 	cic->key = cfqd;
 
-	ioc->set_ioprio = cfq_ioc_set_ioprio;
 restart:
 	parent = NULL;
 	p = &ioc->cic_root.rb_node;
@@ -1407,11 +1384,12 @@ restart:
 			BUG();
 	}
 
-	spin_lock(&cfq_exit_lock);
 	rb_link_node(&cic->rb_node, parent, p);
 	rb_insert_color(&cic->rb_node, &ioc->cic_root);
+
+	spin_lock_irq(cfqd->queue->queue_lock);
 	list_add(&cic->queue_list, &cfqd->cic_list);
-	spin_unlock(&cfq_exit_lock);
+	spin_unlock_irq(cfqd->queue->queue_lock);
 }
 
 /*
@@ -1441,6 +1419,10 @@ cfq_get_io_context(struct cfq_data *cfqd, gfp_t gfp_mask)
 
 	cfq_cic_link(cfqd, ioc, cic);
 out:
+	smp_read_barrier_depends();
+	if (unlikely(ioc->ioprio_changed))
+		cfq_ioc_set_ioprio(ioc);
+
 	return cic;
 err:
 	put_io_context(ioc);
@@ -1945,7 +1927,6 @@ static void cfq_exit_queue(elevator_t *e)
 
 	cfq_shutdown_timer_wq(cfqd);
 
-	spin_lock(&cfq_exit_lock);
 	spin_lock_irq(q->queue_lock);
 
 	if (cfqd->active_queue)
@@ -1960,7 +1941,6 @@ static void cfq_exit_queue(elevator_t *e)
 	}
 
 	spin_unlock_irq(q->queue_lock);
-	spin_unlock(&cfq_exit_lock);
 
 	cfq_shutdown_timer_wq(cfqd);
 
@@ -2149,7 +2129,7 @@ static struct elevator_type iosched_cfq = {
 		.elevator_may_queue_fn =	cfq_may_queue,
 		.elevator_init_fn =		cfq_init_queue,
 		.elevator_exit_fn =		cfq_exit_queue,
-		.trim =				cfq_trim,
+		.trim =				cfq_free_io_context,
 	},
 	.elevator_attrs =	cfq_attrs,
 	.elevator_name =	"cfq",
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index e25b4cd2dcd1..508548b834f1 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -3654,7 +3654,7 @@ struct io_context *current_io_context(gfp_t gfp_flags)
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
 		ret->task = current;
-		ret->set_ioprio = NULL;
+		ret->ioprio_changed = 0;
 		ret->last_waited = jiffies; /* doesn't matter... */
 		ret->nr_batch_requests = 0; /* because this is 0 */
 		ret->aic = NULL;
diff --git a/fs/ioprio.c b/fs/ioprio.c
index 78b1deae3fa2..0fd1089d7bf6 100644
--- a/fs/ioprio.c
+++ b/fs/ioprio.c
@@ -47,8 +47,8 @@ static int set_task_ioprio(struct task_struct *task, int ioprio)
 	/* see wmb() in current_io_context() */
 	smp_read_barrier_depends();
 
-	if (ioc && ioc->set_ioprio)
-		ioc->set_ioprio(ioc, ioprio);
+	if (ioc)
+		ioc->ioprio_changed = 1;
 
 	task_unlock(task);
 	return 0;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index a1e288069e2e..79cb9fa8034a 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -90,7 +90,7 @@ struct io_context {
 	atomic_t refcount;
 	struct task_struct *task;
 
-	int (*set_ioprio)(struct io_context *, unsigned int);
+	unsigned int ioprio_changed;
 
 	/*
 	 * For request batching
-- 
cgit v1.2.3


From 4a893e837bb470867d74c05d6c6b97bba5a96185 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Sat, 22 Jul 2006 15:37:43 +0200
Subject: [PATCH] elevator: define ioc counting mechanism

None of the in-kernel primitives for handling "atomic" counting seem
to be a good fit. We need something that is essentially free for
incrementing/decrementing, while the read side may be more expensive
as we only ever need to do that when a device is removed from the
kernel.

Use a per-cpu variable for maintaining a per-cpu ioc count and define
a reading mechanism that just sums up the values.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 include/linux/elevator.h | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

(limited to 'include')

diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index cc81645a3e18..9c5a04f6114c 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_ELEVATOR_H
 #define _LINUX_ELEVATOR_H
 
+#include <linux/percpu.h>
+
 typedef int (elevator_merge_fn) (request_queue_t *, struct request **,
 				 struct bio *);
 
@@ -178,4 +180,27 @@ enum {
 	INIT_LIST_HEAD(&(rq)->donelist);	\
 	} while (0)
 
+/*
+ * io context count accounting
+ */
+#define elv_ioc_count_mod(name, __val)				\
+	do {							\
+		preempt_disable();				\
+		__get_cpu_var(name) += (__val);			\
+		preempt_enable();				\
+	} while (0)
+
+#define elv_ioc_count_inc(name)	elv_ioc_count_mod(name, 1)
+#define elv_ioc_count_dec(name)	elv_ioc_count_mod(name, -1)
+
+#define elv_ioc_count_read(name)				\
+({								\
+	unsigned long __val = 0;				\
+	int __cpu;						\
+	smp_wmb();						\
+	for_each_possible_cpu(__cpu)				\
+		__val += per_cpu(name, __cpu);			\
+	__val;							\
+})
+
 #endif
-- 
cgit v1.2.3


From a3b05e8f58c95dfccbf2c824d0c68e5990571f24 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Fri, 28 Jul 2006 09:36:46 +0200
Subject: [PATCH] Kill various deprecated/unused block layer defines/functions

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 drivers/block/cciss.c    |  1 -
 drivers/block/cpqarray.c |  1 -
 include/linux/blkdev.h   | 29 -----------------------------
 include/linux/fs.h       |  1 -
 4 files changed, 32 deletions(-)

(limited to 'include')

diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index 2cd3391ff878..c211065ad829 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -1229,7 +1229,6 @@ static inline void complete_buffers(struct bio *bio, int status)
 		int nr_sectors = bio_sectors(bio);
 
 		bio->bi_next = NULL;
-		blk_finished_io(len);
 		bio_endio(bio, nr_sectors << 9, status ? 0 : -EIO);
 		bio = xbh;
 	}
diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c
index 78082edc14b4..4abc193314ee 100644
--- a/drivers/block/cpqarray.c
+++ b/drivers/block/cpqarray.c
@@ -989,7 +989,6 @@ static inline void complete_buffers(struct bio *bio, int ok)
 		xbh = bio->bi_next;
 		bio->bi_next = NULL;
 		
-		blk_finished_io(nr_sectors);
 		bio_endio(bio, nr_sectors << 9, ok ? 0 : -EIO);
 
 		bio = xbh;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 79cb9fa8034a..593386162f47 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -578,12 +578,6 @@ static inline void blk_clear_queue_full(struct request_queue *q, int rw)
 #define rq_mergeable(rq)	\
 	(!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && blk_fs_request((rq)))
 
-/*
- * noop, requests are automagically marked as active/inactive by I/O
- * scheduler -- see elv_next_request
- */
-#define blk_queue_headactive(q, head_active)
-
 /*
  * q->prep_rq_fn return values
  */
@@ -621,11 +615,6 @@ static inline void blk_queue_bounce(request_queue_t *q, struct bio **bio)
 	if ((rq->bio))			\
 		for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next)
 
-struct sec_size {
-	unsigned block_size;
-	unsigned block_size_bits;
-};
-
 extern int blk_register_queue(struct gendisk *disk);
 extern void blk_unregister_queue(struct gendisk *disk);
 extern void register_disk(struct gendisk *dev);
@@ -690,16 +679,6 @@ extern void end_that_request_last(struct request *, int);
 extern void end_request(struct request *req, int uptodate);
 extern void blk_complete_request(struct request *);
 
-static inline int rq_all_done(struct request *rq, unsigned int nr_bytes)
-{
-	if (blk_fs_request(rq))
-		return (nr_bytes >= (rq->hard_nr_sectors << 9));
-	else if (blk_pc_request(rq))
-		return nr_bytes >= rq->data_len;
-
-	return 0;
-}
-
 /*
  * end_that_request_first/chunk() takes an uptodate argument. we account
  * any value <= as an io error. 0 means -EIO for compatability reasons,
@@ -807,14 +786,6 @@ static inline int queue_dma_alignment(request_queue_t *q)
 	return retval;
 }
 
-static inline int bdev_dma_aligment(struct block_device *bdev)
-{
-	return queue_dma_alignment(bdev_get_queue(bdev));
-}
-
-#define blk_finished_io(nsects)	do { } while (0)
-#define blk_started_io(nsects)	do { } while (0)
-
 /* assumes size > 256 */
 static inline unsigned int blksize_bits(unsigned int size)
 {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6eafbe309483..9306f63bf77e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -79,7 +79,6 @@ extern int dir_notify_enable;
 #define WRITE 1
 #define READA 2		/* read-ahead  - don't block if no resources */
 #define SWRITE 3	/* for ll_rw_block() - wait for buffer lock */
-#define SPECIAL 4	/* For non-blockdevice requests in request queue */
 #define READ_SYNC	(READ | (1 << BIO_RW_SYNC))
 #define WRITE_SYNC	(WRITE | (1 << BIO_RW_SYNC))
 #define WRITE_BARRIER	((1 << BIO_RW) | (1 << BIO_RW_BARRIER))
-- 
cgit v1.2.3


From b5deef901282628d88c784f4c9d2f0583ec3b355 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Wed, 19 Jul 2006 23:39:40 +0200
Subject: [PATCH] Make sure all block/io scheduler setups are node aware

Some were kmalloc_node(), some were still kmalloc(). Change them all to
kmalloc_node().

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/as-iosched.c     |  8 ++++----
 block/cfq-iosched.c    | 13 +++++++------
 block/elevator.c       | 11 ++++++-----
 block/ll_rw_blk.c      | 13 +++++++------
 block/noop-iosched.c   |  2 +-
 include/linux/blkdev.h |  3 +--
 6 files changed, 26 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index 8e1fef1eafc9..f6dc95489316 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -210,9 +210,9 @@ static struct as_io_context *alloc_as_io_context(void)
  * If the current task has no AS IO context then create one and initialise it.
  * Then take a ref on the task's io context and return it.
  */
-static struct io_context *as_get_io_context(void)
+static struct io_context *as_get_io_context(int node)
 {
-	struct io_context *ioc = get_io_context(GFP_ATOMIC);
+	struct io_context *ioc = get_io_context(GFP_ATOMIC, node);
 	if (ioc && !ioc->aic) {
 		ioc->aic = alloc_as_io_context();
 		if (!ioc->aic) {
@@ -1148,7 +1148,7 @@ static void as_add_request(request_queue_t *q, struct request *rq)
 
 	data_dir = rq_is_sync(rq);
 
-	rq->elevator_private = as_get_io_context();
+	rq->elevator_private = as_get_io_context(q->node);
 
 	if (RQ_IOC(rq)) {
 		as_update_iohist(ad, RQ_IOC(rq)->aic, rq);
@@ -1292,7 +1292,7 @@ static int as_may_queue(request_queue_t *q, int rw)
 	struct io_context *ioc;
 	if (ad->antic_status == ANTIC_WAIT_REQ ||
 			ad->antic_status == ANTIC_WAIT_NEXT) {
-		ioc = as_get_io_context();
+		ioc = as_get_io_context(q->node);
 		if (ad->io_context == ioc)
 			ret = ELV_MQUEUE_MUST;
 		put_io_context(ioc);
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 85f1d87e86d4..0452108a932e 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1148,8 +1148,9 @@ static void cfq_exit_io_context(struct io_context *ioc)
 static struct cfq_io_context *
 cfq_alloc_io_context(struct cfq_data *cfqd, gfp_t gfp_mask)
 {
-	struct cfq_io_context *cic = kmem_cache_alloc(cfq_ioc_pool, gfp_mask);
+	struct cfq_io_context *cic;
 
+	cic = kmem_cache_alloc_node(cfq_ioc_pool, gfp_mask, cfqd->queue->node);
 	if (cic) {
 		memset(cic, 0, sizeof(*cic));
 		cic->last_end_request = jiffies;
@@ -1277,11 +1278,11 @@ retry:
 			 * free memory.
 			 */
 			spin_unlock_irq(cfqd->queue->queue_lock);
-			new_cfqq = kmem_cache_alloc(cfq_pool, gfp_mask|__GFP_NOFAIL);
+			new_cfqq = kmem_cache_alloc_node(cfq_pool, gfp_mask|__GFP_NOFAIL, cfqd->queue->node);
 			spin_lock_irq(cfqd->queue->queue_lock);
 			goto retry;
 		} else {
-			cfqq = kmem_cache_alloc(cfq_pool, gfp_mask);
+			cfqq = kmem_cache_alloc_node(cfq_pool, gfp_mask, cfqd->queue->node);
 			if (!cfqq)
 				goto out;
 		}
@@ -1407,7 +1408,7 @@ cfq_get_io_context(struct cfq_data *cfqd, gfp_t gfp_mask)
 
 	might_sleep_if(gfp_mask & __GFP_WAIT);
 
-	ioc = get_io_context(gfp_mask);
+	ioc = get_io_context(gfp_mask, cfqd->queue->node);
 	if (!ioc)
 		return NULL;
 
@@ -1955,7 +1956,7 @@ static void *cfq_init_queue(request_queue_t *q, elevator_t *e)
 	struct cfq_data *cfqd;
 	int i;
 
-	cfqd = kmalloc(sizeof(*cfqd), GFP_KERNEL);
+	cfqd = kmalloc_node(sizeof(*cfqd), GFP_KERNEL, q->node);
 	if (!cfqd)
 		return NULL;
 
@@ -1970,7 +1971,7 @@ static void *cfq_init_queue(request_queue_t *q, elevator_t *e)
 	INIT_LIST_HEAD(&cfqd->empty_list);
 	INIT_LIST_HEAD(&cfqd->cic_list);
 
-	cfqd->cfq_hash = kmalloc(sizeof(struct hlist_head) * CFQ_QHASH_ENTRIES, GFP_KERNEL);
+	cfqd->cfq_hash = kmalloc_node(sizeof(struct hlist_head) * CFQ_QHASH_ENTRIES, GFP_KERNEL, q->node);
 	if (!cfqd->cfq_hash)
 		goto out_free;
 
diff --git a/block/elevator.c b/block/elevator.c
index 788d2d81994c..e643291793a4 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -161,12 +161,12 @@ __setup("elevator=", elevator_setup);
 
 static struct kobj_type elv_ktype;
 
-static elevator_t *elevator_alloc(struct elevator_type *e)
+static elevator_t *elevator_alloc(request_queue_t *q, struct elevator_type *e)
 {
 	elevator_t *eq;
 	int i;
 
-	eq = kmalloc(sizeof(elevator_t), GFP_KERNEL);
+	eq = kmalloc_node(sizeof(elevator_t), GFP_KERNEL, q->node);
 	if (unlikely(!eq))
 		goto err;
 
@@ -178,7 +178,8 @@ static elevator_t *elevator_alloc(struct elevator_type *e)
 	eq->kobj.ktype = &elv_ktype;
 	mutex_init(&eq->sysfs_lock);
 
-	eq->hash = kmalloc(sizeof(struct hlist_head) * ELV_HASH_ENTRIES, GFP_KERNEL);
+	eq->hash = kmalloc_node(sizeof(struct hlist_head) * ELV_HASH_ENTRIES,
+					GFP_KERNEL, q->node);
 	if (!eq->hash)
 		goto err;
 
@@ -224,7 +225,7 @@ int elevator_init(request_queue_t *q, char *name)
 		e = elevator_get("noop");
 	}
 
-	eq = elevator_alloc(e);
+	eq = elevator_alloc(q, e);
 	if (!eq)
 		return -ENOMEM;
 
@@ -987,7 +988,7 @@ static int elevator_switch(request_queue_t *q, struct elevator_type *new_e)
 	/*
 	 * Allocate new elevator
 	 */
-	e = elevator_alloc(new_e);
+	e = elevator_alloc(q, new_e);
 	if (!e)
 		return 0;
 
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 4b7b4461e8d6..c6dfa889206c 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -39,6 +39,7 @@ static void blk_unplug_timeout(unsigned long data);
 static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io);
 static void init_request_from_bio(struct request *req, struct bio *bio);
 static int __make_request(request_queue_t *q, struct bio *bio);
+static struct io_context *current_io_context(gfp_t gfp_flags, int node);
 
 /*
  * For the allocated request tables
@@ -2114,7 +2115,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 
 	if (rl->count[rw]+1 >= queue_congestion_on_threshold(q)) {
 		if (rl->count[rw]+1 >= q->nr_requests) {
-			ioc = current_io_context(GFP_ATOMIC);
+			ioc = current_io_context(GFP_ATOMIC, q->node);
 			/*
 			 * The queue will fill after this allocation, so set
 			 * it as full, and mark this process as "batching".
@@ -2234,7 +2235,7 @@ static struct request *get_request_wait(request_queue_t *q, int rw,
 			 * up to a big batch of them for a small period time.
 			 * See ioc_batching, ioc_set_batching
 			 */
-			ioc = current_io_context(GFP_NOIO);
+			ioc = current_io_context(GFP_NOIO, q->node);
 			ioc_set_batching(q, ioc);
 
 			spin_lock_irq(q->queue_lock);
@@ -3641,7 +3642,7 @@ void exit_io_context(void)
  * but since the current task itself holds a reference, the context can be
  * used in general code, so long as it stays within `current` context.
  */
-struct io_context *current_io_context(gfp_t gfp_flags)
+static struct io_context *current_io_context(gfp_t gfp_flags, int node)
 {
 	struct task_struct *tsk = current;
 	struct io_context *ret;
@@ -3650,7 +3651,7 @@ struct io_context *current_io_context(gfp_t gfp_flags)
 	if (likely(ret))
 		return ret;
 
-	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
+	ret = kmem_cache_alloc_node(iocontext_cachep, gfp_flags, node);
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
 		ret->task = current;
@@ -3674,10 +3675,10 @@ EXPORT_SYMBOL(current_io_context);
  *
  * This is always called in the context of the task which submitted the I/O.
  */
-struct io_context *get_io_context(gfp_t gfp_flags)
+struct io_context *get_io_context(gfp_t gfp_flags, int node)
 {
 	struct io_context *ret;
-	ret = current_io_context(gfp_flags);
+	ret = current_io_context(gfp_flags, node);
 	if (likely(ret))
 		atomic_inc(&ret->refcount);
 	return ret;
diff --git a/block/noop-iosched.c b/block/noop-iosched.c
index 56a7c620574f..79af43179421 100644
--- a/block/noop-iosched.c
+++ b/block/noop-iosched.c
@@ -69,7 +69,7 @@ static void *noop_init_queue(request_queue_t *q, elevator_t *e)
 {
 	struct noop_data *nd;
 
-	nd = kmalloc(sizeof(*nd), GFP_KERNEL);
+	nd = kmalloc_node(sizeof(*nd), GFP_KERNEL, q->node);
 	if (!nd)
 		return NULL;
 	INIT_LIST_HEAD(&nd->queue);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 593386162f47..6609371c303e 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -104,8 +104,7 @@ struct io_context {
 
 void put_io_context(struct io_context *ioc);
 void exit_io_context(void);
-struct io_context *current_io_context(gfp_t gfp_flags);
-struct io_context *get_io_context(gfp_t gfp_flags);
+struct io_context *get_io_context(gfp_t gfp_flags, int node);
 void copy_io_context(struct io_context **pdst, struct io_context **psrc);
 void swap_io_context(struct io_context **ioc1, struct io_context **ioc2);
 
-- 
cgit v1.2.3


From dc72ef4ae35c2016fb594bcc85ce871376682174 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Thu, 20 Jul 2006 14:54:05 +0200
Subject: [PATCH] Add blk_start_queueing() helper

CFQ implements this on its own now, but it's really block layer
knowledge. Tells a device queue to start dispatching requests to
the driver, taking care to unplug if needed. Also fixes the issue
where as/cfq will invoke a stopped queue, which we really don't
want.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/as-iosched.c     |  3 +--
 block/cfq-iosched.c    | 22 ++++------------------
 block/ll_rw_blk.c      | 25 ++++++++++++++++++++-----
 include/linux/blkdev.h |  1 +
 4 files changed, 26 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/block/as-iosched.c b/block/as-iosched.c
index f6dc95489316..bc13dc0b29be 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -1280,8 +1280,7 @@ static void as_work_handler(void *data)
 	unsigned long flags;
 
 	spin_lock_irqsave(q->queue_lock, flags);
-	if (!as_queue_empty(q))
-		q->request_fn(q);
+	blk_start_queueing(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 6fb1613d44d7..9f684cc66bd1 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1552,19 +1552,6 @@ static void cfq_preempt_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 	__cfq_set_active_queue(cfqd, cfqq);
 }
 
-/*
- * should really be a ll_rw_blk.c helper
- */
-static void cfq_start_queueing(struct cfq_data *cfqd, struct cfq_queue *cfqq)
-{
-	request_queue_t *q = cfqd->queue;
-
-	if (!blk_queue_plugged(q))
-		q->request_fn(q);
-	else
-		__generic_unplug_device(q);
-}
-
 /*
  * Called when a new fs request (rq) is added (to cfqq). Check if there's
  * something we should do about it
@@ -1593,7 +1580,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		if (cic == cfqd->active_cic &&
 		    del_timer(&cfqd->idle_slice_timer)) {
 			cfq_slice_expired(cfqd, 0);
-			cfq_start_queueing(cfqd, cfqq);
+			blk_start_queueing(cfqd->queue);
 		}
 		return;
 	}
@@ -1614,7 +1601,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		if (cfq_cfqq_wait_request(cfqq)) {
 			cfq_mark_cfqq_must_dispatch(cfqq);
 			del_timer(&cfqd->idle_slice_timer);
-			cfq_start_queueing(cfqd, cfqq);
+			blk_start_queueing(cfqd->queue);
 		}
 	} else if (cfq_should_preempt(cfqd, cfqq, rq)) {
 		/*
@@ -1624,7 +1611,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		 */
 		cfq_preempt_queue(cfqd, cfqq);
 		cfq_mark_cfqq_must_dispatch(cfqq);
-		cfq_start_queueing(cfqd, cfqq);
+		blk_start_queueing(cfqd->queue);
 	}
 }
 
@@ -1832,8 +1819,7 @@ static void cfq_kick_queue(void *data)
 	unsigned long flags;
 
 	spin_lock_irqsave(q->queue_lock, flags);
-	blk_remove_plug(q);
-	q->request_fn(q);
+	blk_start_queueing(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index c6dfa889206c..346be9ae31f6 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -2266,6 +2266,25 @@ struct request *blk_get_request(request_queue_t *q, int rw, gfp_t gfp_mask)
 }
 EXPORT_SYMBOL(blk_get_request);
 
+/**
+ * blk_start_queueing - initiate dispatch of requests to device
+ * @q:		request queue to kick into gear
+ *
+ * This is basically a helper to remove the need to know whether a queue
+ * is plugged or not if someone just wants to initiate dispatch of requests
+ * for this queue.
+ *
+ * The queue lock must be held with interrupts disabled.
+ */
+void blk_start_queueing(request_queue_t *q)
+{
+	if (!blk_queue_plugged(q))
+		q->request_fn(q);
+	else
+		__generic_unplug_device(q);
+}
+EXPORT_SYMBOL(blk_start_queueing);
+
 /**
  * blk_requeue_request - put a request back on queue
  * @q:		request queue where request should be inserted
@@ -2333,11 +2352,7 @@ void blk_insert_request(request_queue_t *q, struct request *rq,
 
 	drive_stat_acct(rq, rq->nr_sectors, 1);
 	__elv_add_request(q, rq, where, 0);
-
-	if (blk_queue_plugged(q))
-		__generic_unplug_device(q);
-	else
-		q->request_fn(q);
+	blk_start_queueing(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 6609371c303e..5d8e288db163 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -635,6 +635,7 @@ extern void blk_stop_queue(request_queue_t *q);
 extern void blk_sync_queue(struct request_queue *q);
 extern void __blk_stop_queue(request_queue_t *q);
 extern void blk_run_queue(request_queue_t *);
+extern void blk_start_queueing(request_queue_t *);
 extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *);
 extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned int);
 extern int blk_rq_unmap_user(struct bio *, unsigned int);
-- 
cgit v1.2.3


From 5404bc7a87b9949cf61e0174b21f80e73239ab25 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Thu, 10 Aug 2006 09:01:02 +0200
Subject: [PATCH] Allow file systems to differentiate between data and meta
 reads

We can use this information for making more intelligent priority
decisions, and it will also be useful for blktrace.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/ll_rw_blk.c      | 2 ++
 include/linux/bio.h    | 2 ++
 include/linux/blkdev.h | 3 +++
 include/linux/fs.h     | 1 +
 4 files changed, 8 insertions(+)

(limited to 'include')

diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index e3980ec747c1..57992ae511c2 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -2884,6 +2884,8 @@ static void init_request_from_bio(struct request *req, struct bio *bio)
 
 	if (bio_sync(bio))
 		req->cmd_flags |= REQ_RW_SYNC;
+	if (bio_rw_meta(bio))
+		req->cmd_flags |= REQ_RW_META;
 
 	req->errors = 0;
 	req->hard_sector = req->sector = bio->bi_sector;
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 76bdaeab6f62..711c321a7011 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -148,6 +148,7 @@ struct bio {
 #define BIO_RW_BARRIER	2
 #define BIO_RW_FAILFAST	3
 #define BIO_RW_SYNC	4
+#define BIO_RW_META	5
 
 /*
  * upper 16 bits of bi_rw define the io priority of this bio
@@ -178,6 +179,7 @@ struct bio {
 #define bio_sync(bio)		((bio)->bi_rw & (1 << BIO_RW_SYNC))
 #define bio_failfast(bio)	((bio)->bi_rw & (1 << BIO_RW_FAILFAST))
 #define bio_rw_ahead(bio)	((bio)->bi_rw & (1 << BIO_RW_AHEAD))
+#define bio_rw_meta(bio)	((bio)->bi_rw & (1 << BIO_RW_META))
 
 /*
  * will die
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 5d8e288db163..2c01a90998a7 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -180,6 +180,7 @@ enum rq_flag_bits {
 	__REQ_ORDERED_COLOR,	/* is before or after barrier */
 	__REQ_RW_SYNC,		/* request is sync (O_DIRECT) */
 	__REQ_ALLOCED,		/* request came from our alloc pool */
+	__REQ_RW_META,		/* metadata io request */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -200,6 +201,7 @@ enum rq_flag_bits {
 #define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
 #define REQ_RW_SYNC	(1 << __REQ_RW_SYNC)
 #define REQ_ALLOCED	(1 << __REQ_ALLOCED)
+#define REQ_RW_META	(1 << __REQ_RW_META)
 
 #define BLK_MAX_CDB	16
 
@@ -543,6 +545,7 @@ enum {
  * We regard a request as sync, if it's a READ or a SYNC write.
  */
 #define rq_is_sync(rq)		(rq_data_dir((rq)) == READ || (rq)->cmd_flags & REQ_RW_SYNC)
+#define rq_is_meta(rq)		((rq)->cmd_flags & REQ_RW_META)
 
 static inline int blk_queue_full(struct request_queue *q, int rw)
 {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9306f63bf77e..d68c37af4dfb 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -80,6 +80,7 @@ extern int dir_notify_enable;
 #define READA 2		/* read-ahead  - don't block if no resources */
 #define SWRITE 3	/* for ll_rw_block() - wait for buffer lock */
 #define READ_SYNC	(READ | (1 << BIO_RW_SYNC))
+#define READ_META	(READ | (1 << BIO_RW_META))
 #define WRITE_SYNC	(WRITE | (1 << BIO_RW_SYNC))
 #define WRITE_BARRIER	((1 << BIO_RW) | (1 << BIO_RW_BARRIER))
 
-- 
cgit v1.2.3


From 7457e6e2d7406c7009e9ad03db1335fe93b5fb71 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Sun, 23 Jul 2006 02:12:01 +0200
Subject: [PATCH] blktrace: support for logging metadata reads

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 block/blktrace.c             | 5 ++++-
 include/linux/blktrace_api.h | 1 +
 2 files changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/block/blktrace.c b/block/blktrace.c
index 8ff33441d8a2..2592f215a905 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -69,7 +69,7 @@ static u32 ddir_act[2] __read_mostly = { BLK_TC_ACT(BLK_TC_READ), BLK_TC_ACT(BLK
 /*
  * Bio action bits of interest
  */
-static u32 bio_act[5] __read_mostly = { 0, BLK_TC_ACT(BLK_TC_BARRIER), BLK_TC_ACT(BLK_TC_SYNC), 0, BLK_TC_ACT(BLK_TC_AHEAD) };
+static u32 bio_act[9] __read_mostly = { 0, BLK_TC_ACT(BLK_TC_BARRIER), BLK_TC_ACT(BLK_TC_SYNC), 0, BLK_TC_ACT(BLK_TC_AHEAD), 0, 0, 0, BLK_TC_ACT(BLK_TC_META) };
 
 /*
  * More could be added as needed, taking care to increment the decrementer
@@ -81,6 +81,8 @@ static u32 bio_act[5] __read_mostly = { 0, BLK_TC_ACT(BLK_TC_BARRIER), BLK_TC_AC
 	(((rw) & (1 << BIO_RW_SYNC)) >> (BIO_RW_SYNC - 1))
 #define trace_ahead_bit(rw)	\
 	(((rw) & (1 << BIO_RW_AHEAD)) << (2 - BIO_RW_AHEAD))
+#define trace_meta_bit(rw)	\
+	(((rw) & (1 << BIO_RW_META)) >> (BIO_RW_META - 3))
 
 /*
  * The worker for the various blk_add_trace*() types. Fills out a
@@ -103,6 +105,7 @@ void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes,
 	what |= bio_act[trace_barrier_bit(rw)];
 	what |= bio_act[trace_sync_bit(rw)];
 	what |= bio_act[trace_ahead_bit(rw)];
+	what |= bio_act[trace_meta_bit(rw)];
 
 	pid = tsk->pid;
 	if (unlikely(act_log_check(bt, what, sector, pid)))
diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index ea48eb1b3fd3..b99a714fcac6 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -20,6 +20,7 @@ enum blktrace_cat {
 	BLK_TC_PC	= 1 << 9,	/* pc requests */
 	BLK_TC_NOTIFY	= 1 << 10,	/* special message */
 	BLK_TC_AHEAD	= 1 << 11,	/* readahead */
+	BLK_TC_META	= 1 << 12,	/* metadata */
 
 	BLK_TC_END	= 1 << 15,	/* only 16-bits, reminder */
 };
-- 
cgit v1.2.3


From cf9a2ae8d49948f861b56e5333530e491a9da190 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:05:54 +0100
Subject: [PATCH] BLOCK: Move functions out of buffer code [try #6]

Move some functions out of the buffering code that aren't strictly buffering
specific.  This is a precursor to being able to disable the block layer.

 (*) Moved some stuff out of fs/buffer.c:

     (*) The file sync and general sync stuff moved to fs/sync.c.

     (*) The superblock sync stuff moved to fs/super.c.

     (*) do_invalidatepage() moved to mm/truncate.c.

     (*) try_to_release_page() moved to mm/filemap.c.

 (*) Moved some related declarations between header files:

     (*) declarations for do_invalidatepage() and try_to_release_page() moved
     	 to linux/mm.h.

     (*) __set_page_dirty_buffers() moved to linux/buffer_head.h.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/buffer.c                 | 174 --------------------------------------------
 fs/super.c                  |  31 ++++++++
 fs/sync.c                   | 113 ++++++++++++++++++++++++++++
 include/linux/buffer_head.h |   3 +-
 include/linux/fs.h          |   1 +
 include/linux/mm.h          |   4 +-
 mm/filemap.c                |  30 ++++++++
 mm/page-writeback.c         |   1 +
 mm/truncate.c               |  24 ++++++
 9 files changed, 204 insertions(+), 177 deletions(-)

(limited to 'include')

diff --git a/fs/buffer.c b/fs/buffer.c
index 3b6d701073e7..16cfbcd254f1 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -159,31 +159,6 @@ int sync_blockdev(struct block_device *bdev)
 }
 EXPORT_SYMBOL(sync_blockdev);
 
-static void __fsync_super(struct super_block *sb)
-{
-	sync_inodes_sb(sb, 0);
-	DQUOT_SYNC(sb);
-	lock_super(sb);
-	if (sb->s_dirt && sb->s_op->write_super)
-		sb->s_op->write_super(sb);
-	unlock_super(sb);
-	if (sb->s_op->sync_fs)
-		sb->s_op->sync_fs(sb, 1);
-	sync_blockdev(sb->s_bdev);
-	sync_inodes_sb(sb, 1);
-}
-
-/*
- * Write out and wait upon all dirty data associated with this
- * superblock.  Filesystem data as well as the underlying block
- * device.  Takes the superblock lock.
- */
-int fsync_super(struct super_block *sb)
-{
-	__fsync_super(sb);
-	return sync_blockdev(sb->s_bdev);
-}
-
 /*
  * Write out and wait upon all dirty data associated with this
  * device.   Filesystem data as well as the underlying block
@@ -259,118 +234,6 @@ void thaw_bdev(struct block_device *bdev, struct super_block *sb)
 }
 EXPORT_SYMBOL(thaw_bdev);
 
-/*
- * sync everything.  Start out by waking pdflush, because that writes back
- * all queues in parallel.
- */
-static void do_sync(unsigned long wait)
-{
-	wakeup_pdflush(0);
-	sync_inodes(0);		/* All mappings, inodes and their blockdevs */
-	DQUOT_SYNC(NULL);
-	sync_supers();		/* Write the superblocks */
-	sync_filesystems(0);	/* Start syncing the filesystems */
-	sync_filesystems(wait);	/* Waitingly sync the filesystems */
-	sync_inodes(wait);	/* Mappings, inodes and blockdevs, again. */
-	if (!wait)
-		printk("Emergency Sync complete\n");
-	if (unlikely(laptop_mode))
-		laptop_sync_completion();
-}
-
-asmlinkage long sys_sync(void)
-{
-	do_sync(1);
-	return 0;
-}
-
-void emergency_sync(void)
-{
-	pdflush_operation(do_sync, 0);
-}
-
-/*
- * Generic function to fsync a file.
- *
- * filp may be NULL if called via the msync of a vma.
- */
- 
-int file_fsync(struct file *filp, struct dentry *dentry, int datasync)
-{
-	struct inode * inode = dentry->d_inode;
-	struct super_block * sb;
-	int ret, err;
-
-	/* sync the inode to buffers */
-	ret = write_inode_now(inode, 0);
-
-	/* sync the superblock to buffers */
-	sb = inode->i_sb;
-	lock_super(sb);
-	if (sb->s_op->write_super)
-		sb->s_op->write_super(sb);
-	unlock_super(sb);
-
-	/* .. finally sync the buffers to disk */
-	err = sync_blockdev(sb->s_bdev);
-	if (!ret)
-		ret = err;
-	return ret;
-}
-
-long do_fsync(struct file *file, int datasync)
-{
-	int ret;
-	int err;
-	struct address_space *mapping = file->f_mapping;
-
-	if (!file->f_op || !file->f_op->fsync) {
-		/* Why?  We can still call filemap_fdatawrite */
-		ret = -EINVAL;
-		goto out;
-	}
-
-	ret = filemap_fdatawrite(mapping);
-
-	/*
-	 * We need to protect against concurrent writers, which could cause
-	 * livelocks in fsync_buffers_list().
-	 */
-	mutex_lock(&mapping->host->i_mutex);
-	err = file->f_op->fsync(file, file->f_dentry, datasync);
-	if (!ret)
-		ret = err;
-	mutex_unlock(&mapping->host->i_mutex);
-	err = filemap_fdatawait(mapping);
-	if (!ret)
-		ret = err;
-out:
-	return ret;
-}
-
-static long __do_fsync(unsigned int fd, int datasync)
-{
-	struct file *file;
-	int ret = -EBADF;
-
-	file = fget(fd);
-	if (file) {
-		ret = do_fsync(file, datasync);
-		fput(file);
-	}
-	return ret;
-}
-
-asmlinkage long sys_fsync(unsigned int fd)
-{
-	return __do_fsync(fd, 0);
-}
-
-asmlinkage long sys_fdatasync(unsigned int fd)
-{
-	return __do_fsync(fd, 1);
-}
-
 /*
  * Various filesystems appear to want __find_get_block to be non-blocking.
  * But it's the page lock which protects the buffers.  To get around this,
@@ -1550,35 +1413,6 @@ static void discard_buffer(struct buffer_head * bh)
 	unlock_buffer(bh);
 }
 
-/**
- * try_to_release_page() - release old fs-specific metadata on a page
- *
- * @page: the page which the kernel is trying to free
- * @gfp_mask: memory allocation flags (and I/O mode)
- *
- * The address_space is to try to release any data against the page
- * (presumably at page->private).  If the release was successful, return `1'.
- * Otherwise return zero.
- *
- * The @gfp_mask argument specifies whether I/O may be performed to release
- * this page (__GFP_IO), and whether the call may block (__GFP_WAIT).
- *
- * NOTE: @gfp_mask may go away, and this function may become non-blocking.
- */
-int try_to_release_page(struct page *page, gfp_t gfp_mask)
-{
-	struct address_space * const mapping = page->mapping;
-
-	BUG_ON(!PageLocked(page));
-	if (PageWriteback(page))
-		return 0;
-	
-	if (mapping && mapping->a_ops->releasepage)
-		return mapping->a_ops->releasepage(page, gfp_mask);
-	return try_to_free_buffers(page);
-}
-EXPORT_SYMBOL(try_to_release_page);
-
 /**
  * block_invalidatepage - invalidate part of all of a buffer-backed page
  *
@@ -1630,14 +1464,6 @@ out:
 }
 EXPORT_SYMBOL(block_invalidatepage);
 
-void do_invalidatepage(struct page *page, unsigned long offset)
-{
-	void (*invalidatepage)(struct page *, unsigned long);
-	invalidatepage = page->mapping->a_ops->invalidatepage ? :
-		block_invalidatepage;
-	(*invalidatepage)(page, offset);
-}
-
 /*
  * We attach and possibly dirty the buffers atomically wrt
  * __set_page_dirty_buffers() via private_lock.  try_to_free_buffers
diff --git a/fs/super.c b/fs/super.c
index 6987824d0dce..15671cd048b1 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -220,6 +220,37 @@ static int grab_super(struct super_block *s) __releases(sb_lock)
 	return 0;
 }
 
+/*
+ * Write out and wait upon all dirty data associated with this
+ * superblock.  Filesystem data as well as the underlying block
+ * device.  Takes the superblock lock.  Requires a second blkdev
+ * flush by the caller to complete the operation.
+ */
+void __fsync_super(struct super_block *sb)
+{
+	sync_inodes_sb(sb, 0);
+	DQUOT_SYNC(sb);
+	lock_super(sb);
+	if (sb->s_dirt && sb->s_op->write_super)
+		sb->s_op->write_super(sb);
+	unlock_super(sb);
+	if (sb->s_op->sync_fs)
+		sb->s_op->sync_fs(sb, 1);
+	sync_blockdev(sb->s_bdev);
+	sync_inodes_sb(sb, 1);
+}
+
+/*
+ * Write out and wait upon all dirty data associated with this
+ * superblock.  Filesystem data as well as the underlying block
+ * device.  Takes the superblock lock.
+ */
+int fsync_super(struct super_block *sb)
+{
+	__fsync_super(sb);
+	return sync_blockdev(sb->s_bdev);
+}
+
 /**
  *	generic_shutdown_super	-	common helper for ->kill_sb()
  *	@sb: superblock to kill
diff --git a/fs/sync.c b/fs/sync.c
index 955aef04da28..1de747b5ddb9 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -10,10 +10,123 @@
 #include <linux/syscalls.h>
 #include <linux/linkage.h>
 #include <linux/pagemap.h>
+#include <linux/quotaops.h>
+#include <linux/buffer_head.h>
 
 #define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \
 			SYNC_FILE_RANGE_WAIT_AFTER)
 
+/*
+ * sync everything.  Start out by waking pdflush, because that writes back
+ * all queues in parallel.
+ */
+static void do_sync(unsigned long wait)
+{
+	wakeup_pdflush(0);
+	sync_inodes(0);		/* All mappings, inodes and their blockdevs */
+	DQUOT_SYNC(NULL);
+	sync_supers();		/* Write the superblocks */
+	sync_filesystems(0);	/* Start syncing the filesystems */
+	sync_filesystems(wait);	/* Waitingly sync the filesystems */
+	sync_inodes(wait);	/* Mappings, inodes and blockdevs, again. */
+	if (!wait)
+		printk("Emergency Sync complete\n");
+	if (unlikely(laptop_mode))
+		laptop_sync_completion();
+}
+
+asmlinkage long sys_sync(void)
+{
+	do_sync(1);
+	return 0;
+}
+
+void emergency_sync(void)
+{
+	pdflush_operation(do_sync, 0);
+}
+
+/*
+ * Generic function to fsync a file.
+ *
+ * filp may be NULL if called via the msync of a vma.
+ */
+int file_fsync(struct file *filp, struct dentry *dentry, int datasync)
+{
+	struct inode * inode = dentry->d_inode;
+	struct super_block * sb;
+	int ret, err;
+
+	/* sync the inode to buffers */
+	ret = write_inode_now(inode, 0);
+
+	/* sync the superblock to buffers */
+	sb = inode->i_sb;
+	lock_super(sb);
+	if (sb->s_op->write_super)
+		sb->s_op->write_super(sb);
+	unlock_super(sb);
+
+	/* .. finally sync the buffers to disk */
+	err = sync_blockdev(sb->s_bdev);
+	if (!ret)
+		ret = err;
+	return ret;
+}
+
+long do_fsync(struct file *file, int datasync)
+{
+	int ret;
+	int err;
+	struct address_space *mapping = file->f_mapping;
+
+	if (!file->f_op || !file->f_op->fsync) {
+		/* Why?  We can still call filemap_fdatawrite */
+		ret = -EINVAL;
+		goto out;
+	}
+
+	ret = filemap_fdatawrite(mapping);
+
+	/*
+	 * We need to protect against concurrent writers, which could cause
+	 * livelocks in fsync_buffers_list().
+	 */
+	mutex_lock(&mapping->host->i_mutex);
+	err = file->f_op->fsync(file, file->f_dentry, datasync);
+	if (!ret)
+		ret = err;
+	mutex_unlock(&mapping->host->i_mutex);
+	err = filemap_fdatawait(mapping);
+	if (!ret)
+		ret = err;
+out:
+	return ret;
+}
+
+static long __do_fsync(unsigned int fd, int datasync)
+{
+	struct file *file;
+	int ret = -EBADF;
+
+	file = fget(fd);
+	if (file) {
+		ret = do_fsync(file, datasync);
+		fput(file);
+	}
+	return ret;
+}
+
+asmlinkage long sys_fsync(unsigned int fd)
+{
+	return __do_fsync(fd, 0);
+}
+
+asmlinkage long sys_fdatasync(unsigned int fd)
+{
+	return __do_fsync(fd, 1);
+}
+
 /*
  * sys_sync_file_range() permits finely controlled syncing over a segment of
  * a file in the range offset .. (offset+nbytes-1) inclusive.  If nbytes is
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index 737e407d0cd1..64b508e35d2a 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -190,9 +190,7 @@ extern int buffer_heads_over_limit;
  * Generic address_space_operations implementations for buffer_head-backed
  * address_spaces.
  */
-int try_to_release_page(struct page * page, gfp_t gfp_mask);
 void block_invalidatepage(struct page *page, unsigned long offset);
-void do_invalidatepage(struct page *page, unsigned long offset);
 int block_write_full_page(struct page *page, get_block_t *get_block,
 				struct writeback_control *wbc);
 int block_read_full_page(struct page*, get_block_t*);
@@ -302,4 +300,5 @@ static inline void lock_buffer(struct buffer_head *bh)
 		__lock_buffer(bh);
 }
 
+extern int __set_page_dirty_buffers(struct page *page);
 #endif /* _LINUX_BUFFER_HEAD_H */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d68c37af4dfb..1728142ec4b6 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1546,6 +1546,7 @@ extern int __filemap_fdatawrite_range(struct address_space *mapping,
 extern long do_fsync(struct file *file, int datasync);
 extern void sync_supers(void);
 extern void sync_filesystems(int wait);
+extern void __fsync_super(struct super_block *sb);
 extern void emergency_sync(void);
 extern void emergency_remount(void);
 extern int do_remount_sb(struct super_block *sb, int flags,
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7b703b6d4358..4edf1934e5ca 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -743,7 +743,9 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, unsigned long
 		int len, int write, int force, struct page **pages, struct vm_area_struct **vmas);
 void print_bad_pte(struct vm_area_struct *, pte_t, unsigned long);
 
-int __set_page_dirty_buffers(struct page *page);
+extern int try_to_release_page(struct page * page, gfp_t gfp_mask);
+extern void do_invalidatepage(struct page *page, unsigned long offset);
+
 int __set_page_dirty_nobuffers(struct page *page);
 int redirty_page_for_writepage(struct writeback_control *wbc,
 				struct page *page);
diff --git a/mm/filemap.c b/mm/filemap.c
index 3277f3b23524..d6846de08887 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2491,3 +2491,33 @@ generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
 	}
 	return retval;
 }
+
+/**
+ * try_to_release_page() - release old fs-specific metadata on a page
+ *
+ * @page: the page which the kernel is trying to free
+ * @gfp_mask: memory allocation flags (and I/O mode)
+ *
+ * The address_space is to try to release any data against the page
+ * (presumably at page->private).  If the release was successful, return `1'.
+ * Otherwise return zero.
+ *
+ * The @gfp_mask argument specifies whether I/O may be performed to release
+ * this page (__GFP_IO), and whether the call may block (__GFP_WAIT).
+ *
+ * NOTE: @gfp_mask may go away, and this function may become non-blocking.
+ */
+int try_to_release_page(struct page *page, gfp_t gfp_mask)
+{
+	struct address_space * const mapping = page->mapping;
+
+	BUG_ON(!PageLocked(page));
+	if (PageWriteback(page))
+		return 0;
+
+	if (mapping && mapping->a_ops->releasepage)
+		return mapping->a_ops->releasepage(page, gfp_mask);
+	return try_to_free_buffers(page);
+}
+
+EXPORT_SYMBOL(try_to_release_page);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 488b7088557c..9fdcc7903956 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -30,6 +30,7 @@
 #include <linux/sysctl.h>
 #include <linux/cpu.h>
 #include <linux/syscalls.h>
+#include <linux/buffer_head.h>
 
 /*
  * The maximum number of pages to writeout in a single bdflush/kupdate
diff --git a/mm/truncate.c b/mm/truncate.c
index a654928323dc..cd3e34b816db 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -17,6 +17,30 @@
 				   do_invalidatepage */
 
 
+/**
+ * do_invalidatepage - invalidate part of all of a page
+ * @page: the page which is affected
+ * @offset: the index of the truncation point
+ *
+ * do_invalidatepage() is called when all or part of the page has become
+ * invalidated by a truncate operation.
+ *
+ * do_invalidatepage() does not have to release all buffers, but it must
+ * ensure that no dirty buffer is left outside @offset and that no I/O
+ * is underway against any of the blocks which are outside the truncation
+ * point.  Because the caller is about to free (and possibly reuse) those
+ * blocks on-disk.
+ */
+void do_invalidatepage(struct page *page, unsigned long offset)
+{
+	void (*invalidatepage)(struct page *, unsigned long);
+	invalidatepage = page->mapping->a_ops->invalidatepage;
+	if (!invalidatepage)
+		invalidatepage = block_invalidatepage;
+	if (invalidatepage)
+		(*invalidatepage)(page, offset);
+}
+
 static inline void truncate_partial_page(struct page *page, unsigned partial)
 {
 	memclear_highpage_flush(page, partial, PAGE_CACHE_SIZE-partial);
-- 
cgit v1.2.3


From 0d67a46df0125e20d14f12dbd3646f1f1bf23e8c Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:05:56 +0100
Subject: [PATCH] BLOCK: Remove duplicate declaration of exit_io_context() [try
 #6]

Remove the duplicate declaration of exit_io_context() from linux/sched.h.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/linux/sched.h | 1 -
 kernel/exit.c         | 1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index a06fc89cf6e5..fc4a9873ec10 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -710,7 +710,6 @@ extern unsigned int max_cache_size;
 
 
 struct io_context;			/* See blkdev.h */
-void exit_io_context(void);
 struct cpuset;
 
 #define NGROUPS_SMALL		32
diff --git a/kernel/exit.c b/kernel/exit.c
index 2e4c13cba95a..c189de2927ab 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -38,6 +38,7 @@
 #include <linux/pipe_fs_i.h>
 #include <linux/audit.h> /* for audit_free() */
 #include <linux/resource.h>
+#include <linux/blkdev.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
-- 
cgit v1.2.3


From 07f3f05c1e3052b8656129b2a5aca9f888241a34 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Sat, 30 Sep 2006 20:52:18 +0200
Subject: [PATCH] BLOCK: Move extern declarations out of fs/*.c into header
 files [try #6]

Create a new header file, fs/internal.h, for common definitions local to the
sources in the fs/ directory.

Move extern definitions that should be in header files from fs/*.c to
fs/internal.h or other main header files where they span directories.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 arch/mips/kernel/signal_n32.c |  4 ++--
 fs/binfmt_elf.c               |  1 -
 fs/block_dev.c                |  1 +
 fs/char_dev.c                 |  1 +
 fs/compat.c                   | 10 +++-------
 fs/compat_ioctl.c             |  2 --
 fs/dcache.c                   |  4 +---
 fs/fs-writeback.c             |  3 +--
 fs/internal.h                 | 36 ++++++++++++++++++++++++++++++++++++
 fs/namespace.c                |  3 +--
 include/linux/ramfs.h         |  1 +
 include/linux/tty.h           |  3 +++
 kernel/compat.c               |  2 ++
 13 files changed, 52 insertions(+), 19 deletions(-)
 create mode 100644 fs/internal.h

(limited to 'include')

diff --git a/arch/mips/kernel/signal_n32.c b/arch/mips/kernel/signal_n32.c
index 477c5334ec1b..50c17eaa7f25 100644
--- a/arch/mips/kernel/signal_n32.c
+++ b/arch/mips/kernel/signal_n32.c
@@ -42,6 +42,8 @@
 
 #include "signal-common.h"
 
+extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat);
+
 /*
  * Including <asm/unistd.h> would give use the 64-bit syscall numbers ...
  */
@@ -81,8 +83,6 @@ struct rt_sigframe_n32 {
 #endif
 };
 
-extern void sigset_from_compat (sigset_t *set, compat_sigset_t *compat);
-
 save_static_function(sysn32_rt_sigsuspend);
 __attribute_used__ noinline static int
 _sysn32_rt_sigsuspend(nabi_no_regargs struct pt_regs regs)
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 6eb48e1446ec..bad52433de69 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -46,7 +46,6 @@
 static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs);
 static int load_elf_library(struct file *);
 static unsigned long elf_map (struct file *, unsigned long, struct elf_phdr *, int, int);
-extern int dump_fpu (struct pt_regs *, elf_fpregset_t *);
 
 #ifndef elf_addr_t
 #define elf_addr_t unsigned long
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 4346468139e8..20f7333ba372 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -22,6 +22,7 @@
 #include <linux/uio.h>
 #include <linux/namei.h>
 #include <asm/uaccess.h>
+#include "internal.h"
 
 struct bdev_inode {
 	struct block_device bdev;
diff --git a/fs/char_dev.c b/fs/char_dev.c
index 1f3285affa39..a885f46ca001 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -24,6 +24,7 @@
 #ifdef CONFIG_KMOD
 #include <linux/kmod.h>
 #endif
+#include "internal.h"
 
 /*
  * capabilities for /dev/mem, /dev/kmem and similar directly mappable character
diff --git a/fs/compat.c b/fs/compat.c
index ce982f6e8c80..122b4e3992b5 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -52,11 +52,12 @@
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
 #include <asm/ioctls.h>
-
-extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat);
+#include "internal.h"
 
 int compat_log = 1;
 
+extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat);
+
 int compat_printk(const char *fmt, ...)
 {
 	va_list ap;
@@ -313,9 +314,6 @@ out:
 #define IOCTL_HASHSIZE 256
 static struct ioctl_trans *ioctl32_hash_table[IOCTL_HASHSIZE];
 
-extern struct ioctl_trans ioctl_start[];
-extern int ioctl_table_size;
-
 static inline unsigned long ioctl32_hash(unsigned long cmd)
 {
 	return (((cmd >> 6) ^ (cmd >> 4) ^ cmd)) % IOCTL_HASHSIZE;
@@ -838,8 +836,6 @@ static int do_nfs4_super_data_conv(void *raw_data)
 	return 0;
 }
 
-extern int copy_mount_options (const void __user *, unsigned long *);
-
 #define SMBFS_NAME      "smbfs"
 #define NCPFS_NAME      "ncpfs"
 #define NFS4_NAME	"nfs4"
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 4063a9396977..ab74c9bd55fe 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1279,8 +1279,6 @@ static int loop_status(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return err;
 }
 
-extern int tty_ioctl(struct inode * inode, struct file * file, unsigned int cmd, unsigned long arg);
-
 #ifdef CONFIG_VT
 
 static int vt_check(struct file *file)
diff --git a/fs/dcache.c b/fs/dcache.c
index 17b392a2049e..fc2faa44f8d1 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -32,6 +32,7 @@
 #include <linux/seqlock.h>
 #include <linux/swap.h>
 #include <linux/bootmem.h>
+#include "internal.h"
 
 
 int sysctl_vfs_cache_pressure __read_mostly = 100;
@@ -1877,9 +1878,6 @@ kmem_cache_t *filp_cachep __read_mostly;
 
 EXPORT_SYMBOL(d_genocide);
 
-extern void bdev_cache_init(void);
-extern void chrdev_init(void);
-
 void __init vfs_caches_init_early(void)
 {
 	dcache_init_early();
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 892643dc9af1..0639024d83a9 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -22,8 +22,7 @@
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <linux/buffer_head.h>
-
-extern struct super_block *blockdev_superblock;
+#include "internal.h"
 
 /**
  *	__mark_inode_dirty -	internal function
diff --git a/fs/internal.h b/fs/internal.h
new file mode 100644
index 000000000000..c21ecd37b1e7
--- /dev/null
+++ b/fs/internal.h
@@ -0,0 +1,36 @@
+/* fs/ internal definitions
+ *
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/ioctl32.h>
+
+/*
+ * block_dev.c
+ */
+extern struct super_block *blockdev_superblock;
+extern void __init bdev_cache_init(void);
+
+/*
+ * char_dev.c
+ */
+extern void __init chrdev_init(void);
+
+/*
+ * compat_ioctl.c
+ */
+#ifdef CONFIG_COMPAT
+extern struct ioctl_trans ioctl_start[];
+extern int ioctl_table_size;
+#endif
+
+/*
+ * namespace.c
+ */
+extern int copy_mount_options(const void __user *, unsigned long *);
diff --git a/fs/namespace.c b/fs/namespace.c
index 6ede3a539ed8..66d921e14fee 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -24,12 +24,11 @@
 #include <linux/namei.h>
 #include <linux/security.h>
 #include <linux/mount.h>
+#include <linux/ramfs.h>
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
 #include "pnode.h"
 
-extern int __init init_rootfs(void);
-
 /* spinlock for vfsmount related operations, inplace of dcache_lock */
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock);
 
diff --git a/include/linux/ramfs.h b/include/linux/ramfs.h
index 00b340ba6612..b160fb18e8d6 100644
--- a/include/linux/ramfs.h
+++ b/include/linux/ramfs.h
@@ -17,5 +17,6 @@ extern int ramfs_nommu_mmap(struct file *file, struct vm_area_struct *vma);
 
 extern const struct file_operations ramfs_file_operations;
 extern struct vm_operations_struct generic_file_vm_ops;
+extern int __init init_rootfs(void);
 
 #endif
diff --git a/include/linux/tty.h b/include/linux/tty.h
index ea4c2605f8da..44091c0db0b4 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -307,6 +307,9 @@ extern void tty_ldisc_put(int);
 extern void tty_wakeup(struct tty_struct *tty);
 extern void tty_ldisc_flush(struct tty_struct *tty);
 
+extern int tty_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
+		     unsigned long arg);
+
 extern struct mutex tty_mutex;
 
 /* n_tty.c */
diff --git a/kernel/compat.c b/kernel/compat.c
index 75573e5d27b0..b4fbd838cd77 100644
--- a/kernel/compat.c
+++ b/kernel/compat.c
@@ -26,6 +26,8 @@
 
 #include <asm/uaccess.h>
 
+extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat);
+
 int get_compat_timespec(struct timespec *ts, const struct compat_timespec __user *cts)
 {
 	return (!access_ok(VERIFY_READ, cts, sizeof(*cts)) ||
-- 
cgit v1.2.3


From 811d736f9e8013966e1a5a930c0db09508bdbb15 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:06:09 +0100
Subject: [PATCH] BLOCK: Dissociate generic_writepages() from mpage stuff [try
 #6]

Dissociate the generic_writepages() function from the mpage stuff, moving its
declaration to linux/mm.h and actually emitting a full implementation into
mm/page-writeback.c.

The implementation is a partial duplicate of mpage_writepages() with all BIO
references removed.

It is used by NFS to do writeback.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/block_dev.c            |   1 +
 fs/mpage.c                |   2 +
 include/linux/mpage.h     |   6 ---
 include/linux/writeback.h |   2 +
 mm/page-writeback.c       | 134 ++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 139 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/fs/block_dev.c b/fs/block_dev.c
index 20f7333ba372..335c38bb86eb 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -17,6 +17,7 @@
 #include <linux/module.h>
 #include <linux/blkpg.h>
 #include <linux/buffer_head.h>
+#include <linux/writeback.h>
 #include <linux/mpage.h>
 #include <linux/mount.h>
 #include <linux/uio.h>
diff --git a/fs/mpage.c b/fs/mpage.c
index 1e4598247d0b..692a3e578fc8 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -693,6 +693,8 @@ out:
  * the call was made get new I/O started against them.  If wbc->sync_mode is
  * WB_SYNC_ALL then we were called for data integrity and we must wait for
  * existing IO to complete.
+ *
+ * If you fix this you should check generic_writepages() also!
  */
 int
 mpage_writepages(struct address_space *mapping,
diff --git a/include/linux/mpage.h b/include/linux/mpage.h
index 3ca880463c47..517c098fde20 100644
--- a/include/linux/mpage.h
+++ b/include/linux/mpage.h
@@ -20,9 +20,3 @@ int mpage_writepages(struct address_space *mapping,
 		struct writeback_control *wbc, get_block_t get_block);
 int mpage_writepage(struct page *page, get_block_t *get_block,
 		struct writeback_control *wbc);
-
-static inline int
-generic_writepages(struct address_space *mapping, struct writeback_control *wbc)
-{
-	return mpage_writepages(mapping, wbc, NULL);
-}
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 9d4074ecd0cd..4f4d98addb44 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -111,6 +111,8 @@ balance_dirty_pages_ratelimited(struct address_space *mapping)
 }
 
 int pdflush_operation(void (*fn)(unsigned long), unsigned long arg0);
+extern int generic_writepages(struct address_space *mapping,
+			      struct writeback_control *wbc);
 int do_writepages(struct address_space *mapping, struct writeback_control *wbc);
 int sync_page_range(struct inode *inode, struct address_space *mapping,
 			loff_t pos, loff_t count);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 9fdcc7903956..ecf27839c203 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -31,6 +31,7 @@
 #include <linux/cpu.h>
 #include <linux/syscalls.h>
 #include <linux/buffer_head.h>
+#include <linux/pagevec.h>
 
 /*
  * The maximum number of pages to writeout in a single bdflush/kupdate
@@ -551,6 +552,139 @@ void __init page_writeback_init(void)
 	register_cpu_notifier(&ratelimit_nb);
 }
 
+/**
+ * generic_writepages - walk the list of dirty pages of the given
+ *                      address space and writepage() all of them.
+ *
+ * @mapping: address space structure to write
+ * @wbc: subtract the number of written pages from *@wbc->nr_to_write
+ *
+ * This is a library function, which implements the writepages()
+ * address_space_operation.
+ *
+ * If a page is already under I/O, generic_writepages() skips it, even
+ * if it's dirty.  This is desirable behaviour for memory-cleaning writeback,
+ * but it is INCORRECT for data-integrity system calls such as fsync().  fsync()
+ * and msync() need to guarantee that all the data which was dirty at the time
+ * the call was made get new I/O started against them.  If wbc->sync_mode is
+ * WB_SYNC_ALL then we were called for data integrity and we must wait for
+ * existing IO to complete.
+ *
+ * Derived from mpage_writepages() - if you fix this you should check that
+ * also!
+ */
+int generic_writepages(struct address_space *mapping,
+		       struct writeback_control *wbc)
+{
+	struct backing_dev_info *bdi = mapping->backing_dev_info;
+	int ret = 0;
+	int done = 0;
+	int (*writepage)(struct page *page, struct writeback_control *wbc);
+	struct pagevec pvec;
+	int nr_pages;
+	pgoff_t index;
+	pgoff_t end;		/* Inclusive */
+	int scanned = 0;
+	int range_whole = 0;
+
+	if (wbc->nonblocking && bdi_write_congested(bdi)) {
+		wbc->encountered_congestion = 1;
+		return 0;
+	}
+
+	writepage = mapping->a_ops->writepage;
+
+	/* deal with chardevs and other special file */
+	if (!writepage)
+		return 0;
+
+	pagevec_init(&pvec, 0);
+	if (wbc->range_cyclic) {
+		index = mapping->writeback_index; /* Start from prev offset */
+		end = -1;
+	} else {
+		index = wbc->range_start >> PAGE_CACHE_SHIFT;
+		end = wbc->range_end >> PAGE_CACHE_SHIFT;
+		if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
+			range_whole = 1;
+		scanned = 1;
+	}
+retry:
+	while (!done && (index <= end) &&
+	       (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
+					      PAGECACHE_TAG_DIRTY,
+					      min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1))) {
+		unsigned i;
+
+		scanned = 1;
+		for (i = 0; i < nr_pages; i++) {
+			struct page *page = pvec.pages[i];
+
+			/*
+			 * At this point we hold neither mapping->tree_lock nor
+			 * lock on the page itself: the page may be truncated or
+			 * invalidated (changing page->mapping to NULL), or even
+			 * swizzled back from swapper_space to tmpfs file
+			 * mapping
+			 */
+			lock_page(page);
+
+			if (unlikely(page->mapping != mapping)) {
+				unlock_page(page);
+				continue;
+			}
+
+			if (!wbc->range_cyclic && page->index > end) {
+				done = 1;
+				unlock_page(page);
+				continue;
+			}
+
+			if (wbc->sync_mode != WB_SYNC_NONE)
+				wait_on_page_writeback(page);
+
+			if (PageWriteback(page) ||
+			    !clear_page_dirty_for_io(page)) {
+				unlock_page(page);
+				continue;
+			}
+
+			ret = (*writepage)(page, wbc);
+			if (ret) {
+				if (ret == -ENOSPC)
+					set_bit(AS_ENOSPC, &mapping->flags);
+				else
+					set_bit(AS_EIO, &mapping->flags);
+			}
+
+			if (unlikely(ret == AOP_WRITEPAGE_ACTIVATE))
+				unlock_page(page);
+			if (ret || (--(wbc->nr_to_write) <= 0))
+				done = 1;
+			if (wbc->nonblocking && bdi_write_congested(bdi)) {
+				wbc->encountered_congestion = 1;
+				done = 1;
+			}
+		}
+		pagevec_release(&pvec);
+		cond_resched();
+	}
+	if (!scanned && !done) {
+		/*
+		 * We hit the last page and there is more work to be done: wrap
+		 * back to the start of the file
+		 */
+		scanned = 1;
+		index = 0;
+		goto retry;
+	}
+	if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
+		mapping->writeback_index = index;
+	return ret;
+}
+
+EXPORT_SYMBOL(generic_writepages);
+
 int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
 	int ret;
-- 
cgit v1.2.3


From 863d5b822c02d0e7215fb84ca79e9f8c3e35f04e Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:06:14 +0100
Subject: [PATCH] BLOCK: Move the loop device ioctl compat stuff to the loop
 driver [try #6]

Move the loop device ioctl compat stuff from fs/compat_ioctl.c to the loop
driver so that the loop header file doesn't need to be included.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/block/loop.c         | 160 +++++++++++++++++++++++++++++++++++++++++++
 fs/compat_ioctl.c            |  68 ------------------
 include/linux/compat_ioctl.h |   6 --
 3 files changed, 160 insertions(+), 74 deletions(-)

(limited to 'include')

diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 68b0471ad5a6..d6bb8da955a2 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -66,6 +66,7 @@
 #include <linux/swap.h>
 #include <linux/slab.h>
 #include <linux/loop.h>
+#include <linux/compat.h>
 #include <linux/suspend.h>
 #include <linux/writeback.h>
 #include <linux/buffer_head.h>		/* for invalidate_bdev() */
@@ -1165,6 +1166,162 @@ static int lo_ioctl(struct inode * inode, struct file * file,
 	return err;
 }
 
+#ifdef CONFIG_COMPAT
+struct compat_loop_info {
+	compat_int_t	lo_number;      /* ioctl r/o */
+	compat_dev_t	lo_device;      /* ioctl r/o */
+	compat_ulong_t	lo_inode;       /* ioctl r/o */
+	compat_dev_t	lo_rdevice;     /* ioctl r/o */
+	compat_int_t	lo_offset;
+	compat_int_t	lo_encrypt_type;
+	compat_int_t	lo_encrypt_key_size;    /* ioctl w/o */
+	compat_int_t	lo_flags;       /* ioctl r/o */
+	char		lo_name[LO_NAME_SIZE];
+	unsigned char	lo_encrypt_key[LO_KEY_SIZE]; /* ioctl w/o */
+	compat_ulong_t	lo_init[2];
+	char		reserved[4];
+};
+
+/*
+ * Transfer 32-bit compatibility structure in userspace to 64-bit loop info
+ * - noinlined to reduce stack space usage in main part of driver
+ */
+static noinline int
+loop_info64_from_compat(const struct compat_loop_info *arg,
+			struct loop_info64 *info64)
+{
+	struct compat_loop_info info;
+
+	if (copy_from_user(&info, arg, sizeof(info)))
+		return -EFAULT;
+
+	memset(info64, 0, sizeof(*info64));
+	info64->lo_number = info.lo_number;
+	info64->lo_device = info.lo_device;
+	info64->lo_inode = info.lo_inode;
+	info64->lo_rdevice = info.lo_rdevice;
+	info64->lo_offset = info.lo_offset;
+	info64->lo_sizelimit = 0;
+	info64->lo_encrypt_type = info.lo_encrypt_type;
+	info64->lo_encrypt_key_size = info.lo_encrypt_key_size;
+	info64->lo_flags = info.lo_flags;
+	info64->lo_init[0] = info.lo_init[0];
+	info64->lo_init[1] = info.lo_init[1];
+	if (info.lo_encrypt_type == LO_CRYPT_CRYPTOAPI)
+		memcpy(info64->lo_crypt_name, info.lo_name, LO_NAME_SIZE);
+	else
+		memcpy(info64->lo_file_name, info.lo_name, LO_NAME_SIZE);
+	memcpy(info64->lo_encrypt_key, info.lo_encrypt_key, LO_KEY_SIZE);
+	return 0;
+}
+
+/*
+ * Transfer 64-bit loop info to 32-bit compatibility structure in userspace
+ * - noinlined to reduce stack space usage in main part of driver
+ */
+static noinline int
+loop_info64_to_compat(const struct loop_info64 *info64,
+		      struct compat_loop_info __user *arg)
+{
+	struct compat_loop_info info;
+
+	memset(&info, 0, sizeof(info));
+	info.lo_number = info64->lo_number;
+	info.lo_device = info64->lo_device;
+	info.lo_inode = info64->lo_inode;
+	info.lo_rdevice = info64->lo_rdevice;
+	info.lo_offset = info64->lo_offset;
+	info.lo_encrypt_type = info64->lo_encrypt_type;
+	info.lo_encrypt_key_size = info64->lo_encrypt_key_size;
+	info.lo_flags = info64->lo_flags;
+	info.lo_init[0] = info64->lo_init[0];
+	info.lo_init[1] = info64->lo_init[1];
+	if (info.lo_encrypt_type == LO_CRYPT_CRYPTOAPI)
+		memcpy(info.lo_name, info64->lo_crypt_name, LO_NAME_SIZE);
+	else
+		memcpy(info.lo_name, info64->lo_file_name, LO_NAME_SIZE);
+	memcpy(info.lo_encrypt_key, info64->lo_encrypt_key, LO_KEY_SIZE);
+
+	/* error in case values were truncated */
+	if (info.lo_device != info64->lo_device ||
+	    info.lo_rdevice != info64->lo_rdevice ||
+	    info.lo_inode != info64->lo_inode ||
+	    info.lo_offset != info64->lo_offset ||
+	    info.lo_init[0] != info64->lo_init[0] ||
+	    info.lo_init[1] != info64->lo_init[1])
+		return -EOVERFLOW;
+
+	if (copy_to_user(arg, &info, sizeof(info)))
+		return -EFAULT;
+	return 0;
+}
+
+static int
+loop_set_status_compat(struct loop_device *lo,
+		       const struct compat_loop_info __user *arg)
+{
+	struct loop_info64 info64;
+	int ret;
+
+	ret = loop_info64_from_compat(arg, &info64);
+	if (ret < 0)
+		return ret;
+	return loop_set_status(lo, &info64);
+}
+
+static int
+loop_get_status_compat(struct loop_device *lo,
+		       struct compat_loop_info __user *arg)
+{
+	struct loop_info64 info64;
+	int err = 0;
+
+	if (!arg)
+		err = -EINVAL;
+	if (!err)
+		err = loop_get_status(lo, &info64);
+	if (!err)
+		err = loop_info64_to_compat(&info64, arg);
+	return err;
+}
+
+static long lo_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
+	int err;
+
+	lock_kernel();
+	switch(cmd) {
+	case LOOP_SET_STATUS:
+		mutex_lock(&lo->lo_ctl_mutex);
+		err = loop_set_status_compat(
+			lo, (const struct compat_loop_info __user *) arg);
+		mutex_unlock(&lo->lo_ctl_mutex);
+		break;
+	case LOOP_GET_STATUS:
+		mutex_lock(&lo->lo_ctl_mutex);
+		err = loop_get_status_compat(
+			lo, (struct compat_loop_info __user *) arg);
+		mutex_unlock(&lo->lo_ctl_mutex);
+		break;
+	case LOOP_CLR_FD:
+	case LOOP_GET_STATUS64:
+	case LOOP_SET_STATUS64:
+		arg = (unsigned long) compat_ptr(arg);
+	case LOOP_SET_FD:
+	case LOOP_CHANGE_FD:
+		err = lo_ioctl(inode, file, cmd, arg);
+		break;
+	default:
+		err = -ENOIOCTLCMD;
+		break;
+	}
+	unlock_kernel();
+	return err;
+}
+#endif
+
 static int lo_open(struct inode *inode, struct file *file)
 {
 	struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
@@ -1192,6 +1349,9 @@ static struct block_device_operations lo_fops = {
 	.open =		lo_open,
 	.release =	lo_release,
 	.ioctl =	lo_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl =	lo_compat_ioctl,
+#endif
 };
 
 /*
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index ab74c9bd55fe..3b0cf7fbd95a 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -40,7 +40,6 @@
 #include <linux/if_pppox.h>
 #include <linux/mtio.h>
 #include <linux/cdrom.h>
-#include <linux/loop.h>
 #include <linux/auto_fs.h>
 #include <linux/auto_fs4.h>
 #include <linux/tty.h>
@@ -1214,71 +1213,6 @@ static int cdrom_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long ar
 	return err;
 }
 
-struct loop_info32 {
-	compat_int_t	lo_number;      /* ioctl r/o */
-	compat_dev_t	lo_device;      /* ioctl r/o */
-	compat_ulong_t	lo_inode;       /* ioctl r/o */
-	compat_dev_t	lo_rdevice;     /* ioctl r/o */
-	compat_int_t	lo_offset;
-	compat_int_t	lo_encrypt_type;
-	compat_int_t	lo_encrypt_key_size;    /* ioctl w/o */
-	compat_int_t	lo_flags;       /* ioctl r/o */
-	char		lo_name[LO_NAME_SIZE];
-	unsigned char	lo_encrypt_key[LO_KEY_SIZE]; /* ioctl w/o */
-	compat_ulong_t	lo_init[2];
-	char		reserved[4];
-};
-
-static int loop_status(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	mm_segment_t old_fs = get_fs();
-	struct loop_info l;
-	struct loop_info32 __user *ul;
-	int err = -EINVAL;
-
-	ul = compat_ptr(arg);
-	switch(cmd) {
-	case LOOP_SET_STATUS:
-		err = get_user(l.lo_number, &ul->lo_number);
-		err |= __get_user(l.lo_device, &ul->lo_device);
-		err |= __get_user(l.lo_inode, &ul->lo_inode);
-		err |= __get_user(l.lo_rdevice, &ul->lo_rdevice);
-		err |= __copy_from_user(&l.lo_offset, &ul->lo_offset,
-		        8 + (unsigned long)l.lo_init - (unsigned long)&l.lo_offset);
-		if (err) {
-			err = -EFAULT;
-		} else {
-			set_fs (KERNEL_DS);
-			err = sys_ioctl (fd, cmd, (unsigned long)&l);
-			set_fs (old_fs);
-		}
-		break;
-	case LOOP_GET_STATUS:
-		set_fs (KERNEL_DS);
-		err = sys_ioctl (fd, cmd, (unsigned long)&l);
-		set_fs (old_fs);
-		if (!err) {
-			err = put_user(l.lo_number, &ul->lo_number);
-			err |= __put_user(l.lo_device, &ul->lo_device);
-			err |= __put_user(l.lo_inode, &ul->lo_inode);
-			err |= __put_user(l.lo_rdevice, &ul->lo_rdevice);
-			err |= __copy_to_user(&ul->lo_offset, &l.lo_offset,
-				(unsigned long)l.lo_init - (unsigned long)&l.lo_offset);
-			if (err)
-				err = -EFAULT;
-		}
-		break;
-	default: {
-		static int count;
-		if (++count <= 20)
-			printk("%s: Unknown loop ioctl cmd, fd(%d) "
-			       "cmd(%08x) arg(%08lx)\n",
-			       __FUNCTION__, fd, cmd, arg);
-	}
-	}
-	return err;
-}
-
 #ifdef CONFIG_VT
 
 static int vt_check(struct file *file)
@@ -2808,8 +2742,6 @@ HANDLE_IOCTL(MTIOCGET32, mt_ioctl_trans)
 HANDLE_IOCTL(MTIOCPOS32, mt_ioctl_trans)
 HANDLE_IOCTL(CDROMREADAUDIO, cdrom_ioctl_trans)
 HANDLE_IOCTL(CDROM_SEND_PACKET, cdrom_ioctl_trans)
-HANDLE_IOCTL(LOOP_SET_STATUS, loop_status)
-HANDLE_IOCTL(LOOP_GET_STATUS, loop_status)
 #define AUTOFS_IOC_SETTIMEOUT32 _IOWR(0x93,0x64,unsigned int)
 HANDLE_IOCTL(AUTOFS_IOC_SETTIMEOUT32, ioc_settimeout)
 #ifdef CONFIG_VT
diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h
index bea0255196c4..98d40e08ba6e 100644
--- a/include/linux/compat_ioctl.h
+++ b/include/linux/compat_ioctl.h
@@ -395,12 +395,6 @@ COMPATIBLE_IOCTL(DVD_WRITE_STRUCT)
 COMPATIBLE_IOCTL(DVD_AUTH)
 /* pktcdvd */
 COMPATIBLE_IOCTL(PACKET_CTRL_CMD)
-/* Big L */
-ULONG_IOCTL(LOOP_SET_FD)
-ULONG_IOCTL(LOOP_CHANGE_FD)
-COMPATIBLE_IOCTL(LOOP_CLR_FD)
-COMPATIBLE_IOCTL(LOOP_GET_STATUS64)
-COMPATIBLE_IOCTL(LOOP_SET_STATUS64)
 /* Big A */
 /* sparc only */
 /* Big Q for sound/OSS */
-- 
cgit v1.2.3


From 36695673b012096228ebdc1b39a6a5850daa474e Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:06:16 +0100
Subject: [PATCH] BLOCK: Move common FS-specific ioctls to linux/fs.h [try #6]

Move common FS-specific ioctls from linux/ext2_fs.h to linux/fs.h as FS_IOC_*
and FS_IOC32_* and have the users of them use those as a base.

Also move the GETFLAGS/SETFLAGS flags to linux/fs.h as FS_*_FL macros, and then
have the other users use them as a base.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/cifs/ioctl.c             |  7 +++--
 fs/compat_ioctl.c           | 15 -----------
 fs/hfsplus/hfsplus_fs.h     |  8 ++----
 fs/hfsplus/ioctl.c          | 17 ++++++------
 fs/jfs/ioctl.c              | 15 +++++------
 include/linux/ext2_fs.h     | 64 +++++++++++++++++++++++++--------------------
 include/linux/ext3_fs.h     | 20 +++++++++++---
 include/linux/fs.h          | 39 +++++++++++++++++++++++++++
 include/linux/reiserfs_fs.h | 28 +++++++++-----------
 9 files changed, 124 insertions(+), 89 deletions(-)

(limited to 'include')

diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c
index b0ea6687ab55..e34c7db00f6f 100644
--- a/fs/cifs/ioctl.c
+++ b/fs/cifs/ioctl.c
@@ -22,7 +22,6 @@
  */
 
 #include <linux/fs.h>
-#include <linux/ext2_fs.h>
 #include "cifspdu.h"
 #include "cifsglob.h"
 #include "cifsproto.h"
@@ -74,7 +73,7 @@ int cifs_ioctl (struct inode * inode, struct file * filep,
 			}
 			break;
 #ifdef CONFIG_CIFS_POSIX
-		case EXT2_IOC_GETFLAGS:
+		case FS_IOC_GETFLAGS:
 			if(CIFS_UNIX_EXTATTR_CAP & caps) {
 				if (pSMBFile == NULL)
 					break;
@@ -82,12 +81,12 @@ int cifs_ioctl (struct inode * inode, struct file * filep,
 					&ExtAttrBits, &ExtAttrMask);
 				if(rc == 0)
 					rc = put_user(ExtAttrBits &
-						EXT2_FL_USER_VISIBLE,
+						FS_FL_USER_VISIBLE,
 						(int __user *)arg);
 			}
 			break;
 
-		case EXT2_IOC_SETFLAGS:
+		case FS_IOC_SETFLAGS:
 			if(CIFS_UNIX_EXTATTR_CAP & caps) {
 				if(get_user(ExtAttrBits,(int __user *)arg)) {
 					rc = -EFAULT;
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 3b0cf7fbd95a..bd9c4f49d4e5 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -123,21 +123,6 @@
 #include <linux/dvb/video.h>
 #include <linux/lp.h>
 
-/* Aiee. Someone does not find a difference between int and long */
-#define EXT2_IOC32_GETFLAGS               _IOR('f', 1, int)
-#define EXT2_IOC32_SETFLAGS               _IOW('f', 2, int)
-#define EXT3_IOC32_GETVERSION             _IOR('f', 3, int)
-#define EXT3_IOC32_SETVERSION             _IOW('f', 4, int)
-#define EXT3_IOC32_GETRSVSZ               _IOR('f', 5, int)
-#define EXT3_IOC32_SETRSVSZ               _IOW('f', 6, int)
-#define EXT3_IOC32_GROUP_EXTEND           _IOW('f', 7, unsigned int)
-#ifdef CONFIG_JBD_DEBUG
-#define EXT3_IOC32_WAIT_FOR_READONLY      _IOR('f', 99, int)
-#endif
-
-#define EXT2_IOC32_GETVERSION             _IOR('v', 1, int)
-#define EXT2_IOC32_SETVERSION             _IOW('v', 2, int)
-
 static int do_ioctl32_pointer(unsigned int fd, unsigned int cmd,
 			      unsigned long arg, struct file *f)
 {
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index 8a1ca5ef7ada..3915635b4470 100644
--- a/fs/hfsplus/hfsplus_fs.h
+++ b/fs/hfsplus/hfsplus_fs.h
@@ -246,12 +246,8 @@ struct hfsplus_readdir_data {
 
 /* ext2 ioctls (EXT2_IOC_GETFLAGS and EXT2_IOC_SETFLAGS) to support
  * chattr/lsattr */
-#define HFSPLUS_IOC_EXT2_GETFLAGS	_IOR('f', 1, long)
-#define HFSPLUS_IOC_EXT2_SETFLAGS	_IOW('f', 2, long)
-
-#define EXT2_FLAG_IMMUTABLE		0x00000010 /* Immutable file */
-#define EXT2_FLAG_APPEND		0x00000020 /* writes to file may only append */
-#define EXT2_FLAG_NODUMP		0x00000040 /* do not dump file */
+#define HFSPLUS_IOC_EXT2_GETFLAGS	FS_IOC_GETFLAGS
+#define HFSPLUS_IOC_EXT2_SETFLAGS	FS_IOC_SETFLAGS
 
 
 /*
diff --git a/fs/hfsplus/ioctl.c b/fs/hfsplus/ioctl.c
index 13cf848ac833..79fd10402ea3 100644
--- a/fs/hfsplus/ioctl.c
+++ b/fs/hfsplus/ioctl.c
@@ -28,11 +28,11 @@ int hfsplus_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
 	case HFSPLUS_IOC_EXT2_GETFLAGS:
 		flags = 0;
 		if (HFSPLUS_I(inode).rootflags & HFSPLUS_FLG_IMMUTABLE)
-			flags |= EXT2_FLAG_IMMUTABLE; /* EXT2_IMMUTABLE_FL */
+			flags |= FS_IMMUTABLE_FL; /* EXT2_IMMUTABLE_FL */
 		if (HFSPLUS_I(inode).rootflags & HFSPLUS_FLG_APPEND)
-			flags |= EXT2_FLAG_APPEND; /* EXT2_APPEND_FL */
+			flags |= FS_APPEND_FL; /* EXT2_APPEND_FL */
 		if (HFSPLUS_I(inode).userflags & HFSPLUS_FLG_NODUMP)
-			flags |= EXT2_FLAG_NODUMP; /* EXT2_NODUMP_FL */
+			flags |= FS_NODUMP_FL; /* EXT2_NODUMP_FL */
 		return put_user(flags, (int __user *)arg);
 	case HFSPLUS_IOC_EXT2_SETFLAGS: {
 		if (IS_RDONLY(inode))
@@ -44,32 +44,31 @@ int hfsplus_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
 		if (get_user(flags, (int __user *)arg))
 			return -EFAULT;
 
-		if (flags & (EXT2_FLAG_IMMUTABLE|EXT2_FLAG_APPEND) ||
+		if (flags & (FS_IMMUTABLE_FL|FS_APPEND_FL) ||
 		    HFSPLUS_I(inode).rootflags & (HFSPLUS_FLG_IMMUTABLE|HFSPLUS_FLG_APPEND)) {
 			if (!capable(CAP_LINUX_IMMUTABLE))
 				return -EPERM;
 		}
 
 		/* don't silently ignore unsupported ext2 flags */
-		if (flags & ~(EXT2_FLAG_IMMUTABLE|EXT2_FLAG_APPEND|
-			      EXT2_FLAG_NODUMP))
+		if (flags & ~(FS_IMMUTABLE_FL|FS_APPEND_FL|FS_NODUMP_FL))
 			return -EOPNOTSUPP;
 
-		if (flags & EXT2_FLAG_IMMUTABLE) { /* EXT2_IMMUTABLE_FL */
+		if (flags & FS_IMMUTABLE_FL) { /* EXT2_IMMUTABLE_FL */
 			inode->i_flags |= S_IMMUTABLE;
 			HFSPLUS_I(inode).rootflags |= HFSPLUS_FLG_IMMUTABLE;
 		} else {
 			inode->i_flags &= ~S_IMMUTABLE;
 			HFSPLUS_I(inode).rootflags &= ~HFSPLUS_FLG_IMMUTABLE;
 		}
-		if (flags & EXT2_FLAG_APPEND) { /* EXT2_APPEND_FL */
+		if (flags & FS_APPEND_FL) { /* EXT2_APPEND_FL */
 			inode->i_flags |= S_APPEND;
 			HFSPLUS_I(inode).rootflags |= HFSPLUS_FLG_APPEND;
 		} else {
 			inode->i_flags &= ~S_APPEND;
 			HFSPLUS_I(inode).rootflags &= ~HFSPLUS_FLG_APPEND;
 		}
-		if (flags & EXT2_FLAG_NODUMP) /* EXT2_NODUMP_FL */
+		if (flags & FS_NODUMP_FL) /* EXT2_NODUMP_FL */
 			HFSPLUS_I(inode).userflags |= HFSPLUS_FLG_NODUMP;
 		else
 			HFSPLUS_I(inode).userflags &= ~HFSPLUS_FLG_NODUMP;
diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c
index 67b3774820eb..37db52488262 100644
--- a/fs/jfs/ioctl.c
+++ b/fs/jfs/ioctl.c
@@ -6,7 +6,6 @@
  */
 
 #include <linux/fs.h>
-#include <linux/ext2_fs.h>
 #include <linux/ctype.h>
 #include <linux/capability.h>
 #include <linux/time.h>
@@ -22,13 +21,13 @@ static struct {
 	long jfs_flag;
 	long ext2_flag;
 } jfs_map[] = {
-	{JFS_NOATIME_FL, EXT2_NOATIME_FL},
-	{JFS_DIRSYNC_FL, EXT2_DIRSYNC_FL},
-	{JFS_SYNC_FL, EXT2_SYNC_FL},
-	{JFS_SECRM_FL, EXT2_SECRM_FL},
-	{JFS_UNRM_FL, EXT2_UNRM_FL},
-	{JFS_APPEND_FL, EXT2_APPEND_FL},
-	{JFS_IMMUTABLE_FL, EXT2_IMMUTABLE_FL},
+	{JFS_NOATIME_FL,	FS_NOATIME_FL},
+	{JFS_DIRSYNC_FL,	FS_DIRSYNC_FL},
+	{JFS_SYNC_FL,		FS_SYNC_FL},
+	{JFS_SECRM_FL,		FS_SECRM_FL},
+	{JFS_UNRM_FL,		FS_UNRM_FL},
+	{JFS_APPEND_FL,		FS_APPEND_FL},
+	{JFS_IMMUTABLE_FL,	FS_IMMUTABLE_FL},
 	{0, 0},
 };
 
diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index 33a1aa107329..153d755376a4 100644
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -165,41 +165,49 @@ struct ext2_group_desc
 #define	EXT2_N_BLOCKS			(EXT2_TIND_BLOCK + 1)
 
 /*
- * Inode flags
+ * Inode flags (GETFLAGS/SETFLAGS)
  */
-#define	EXT2_SECRM_FL			0x00000001 /* Secure deletion */
-#define	EXT2_UNRM_FL			0x00000002 /* Undelete */
-#define	EXT2_COMPR_FL			0x00000004 /* Compress file */
-#define EXT2_SYNC_FL			0x00000008 /* Synchronous updates */
-#define EXT2_IMMUTABLE_FL		0x00000010 /* Immutable file */
-#define EXT2_APPEND_FL			0x00000020 /* writes to file may only append */
-#define EXT2_NODUMP_FL			0x00000040 /* do not dump file */
-#define EXT2_NOATIME_FL			0x00000080 /* do not update atime */
+#define	EXT2_SECRM_FL			FS_SECRM_FL	/* Secure deletion */
+#define	EXT2_UNRM_FL			FS_UNRM_FL	/* Undelete */
+#define	EXT2_COMPR_FL			FS_COMPR_FL	/* Compress file */
+#define EXT2_SYNC_FL			FS_SYNC_FL	/* Synchronous updates */
+#define EXT2_IMMUTABLE_FL		FS_IMMUTABLE_FL	/* Immutable file */
+#define EXT2_APPEND_FL			FS_APPEND_FL	/* writes to file may only append */
+#define EXT2_NODUMP_FL			FS_NODUMP_FL	/* do not dump file */
+#define EXT2_NOATIME_FL			FS_NOATIME_FL	/* do not update atime */
 /* Reserved for compression usage... */
-#define EXT2_DIRTY_FL			0x00000100
-#define EXT2_COMPRBLK_FL		0x00000200 /* One or more compressed clusters */
-#define EXT2_NOCOMP_FL			0x00000400 /* Don't compress */
-#define EXT2_ECOMPR_FL			0x00000800 /* Compression error */
+#define EXT2_DIRTY_FL			FS_DIRTY_FL
+#define EXT2_COMPRBLK_FL		FS_COMPRBLK_FL	/* One or more compressed clusters */
+#define EXT2_NOCOMP_FL			FS_NOCOMP_FL	/* Don't compress */
+#define EXT2_ECOMPR_FL			FS_ECOMPR_FL	/* Compression error */
 /* End compression flags --- maybe not all used */	
-#define EXT2_BTREE_FL			0x00001000 /* btree format dir */
-#define EXT2_INDEX_FL			0x00001000 /* hash-indexed directory */
-#define EXT2_IMAGIC_FL			0x00002000 /* AFS directory */
-#define EXT2_JOURNAL_DATA_FL		0x00004000 /* Reserved for ext3 */
-#define EXT2_NOTAIL_FL			0x00008000 /* file tail should not be merged */
-#define EXT2_DIRSYNC_FL			0x00010000 /* dirsync behaviour (directories only) */
-#define EXT2_TOPDIR_FL			0x00020000 /* Top of directory hierarchies*/
-#define EXT2_RESERVED_FL		0x80000000 /* reserved for ext2 lib */
-
-#define EXT2_FL_USER_VISIBLE		0x0003DFFF /* User visible flags */
-#define EXT2_FL_USER_MODIFIABLE		0x000380FF /* User modifiable flags */
+#define EXT2_BTREE_FL			FS_BTREE_FL	/* btree format dir */
+#define EXT2_INDEX_FL			FS_INDEX_FL	/* hash-indexed directory */
+#define EXT2_IMAGIC_FL			FS_IMAGIC_FL	/* AFS directory */
+#define EXT2_JOURNAL_DATA_FL		FS_JOURNAL_DATA_FL /* Reserved for ext3 */
+#define EXT2_NOTAIL_FL			FS_NOTAIL_FL	/* file tail should not be merged */
+#define EXT2_DIRSYNC_FL			FS_DIRSYNC_FL	/* dirsync behaviour (directories only) */
+#define EXT2_TOPDIR_FL			FS_TOPDIR_FL	/* Top of directory hierarchies*/
+#define EXT2_RESERVED_FL		FS_RESERVED_FL	/* reserved for ext2 lib */
+
+#define EXT2_FL_USER_VISIBLE		FS_FL_USER_VISIBLE	/* User visible flags */
+#define EXT2_FL_USER_MODIFIABLE		FS_FL_USER_MODIFIABLE	/* User modifiable flags */
 
 /*
  * ioctl commands
  */
-#define	EXT2_IOC_GETFLAGS		_IOR('f', 1, long)
-#define	EXT2_IOC_SETFLAGS		_IOW('f', 2, long)
-#define	EXT2_IOC_GETVERSION		_IOR('v', 1, long)
-#define	EXT2_IOC_SETVERSION		_IOW('v', 2, long)
+#define	EXT2_IOC_GETFLAGS		FS_IOC_GETFLAGS
+#define	EXT2_IOC_SETFLAGS		FS_IOC_SETFLAGS
+#define	EXT2_IOC_GETVERSION		FS_IOC_GETVERSION
+#define	EXT2_IOC_SETVERSION		FS_IOC_SETVERSION
+
+/*
+ * ioctl commands in 32 bit emulation
+ */
+#define EXT2_IOC32_GETFLAGS		FS_IOC32_GETFLAGS
+#define EXT2_IOC32_SETFLAGS		FS_IOC32_SETFLAGS
+#define EXT2_IOC32_GETVERSION		FS_IOC32_GETVERSION
+#define EXT2_IOC32_SETVERSION		FS_IOC32_SETVERSION
 
 /*
  * Structure of an inode on the disk
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index cc08f56750da..a7a01ff44669 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -216,20 +216,32 @@ struct ext3_new_group_data {
 /*
  * ioctl commands
  */
-#define	EXT3_IOC_GETFLAGS		_IOR('f', 1, long)
-#define	EXT3_IOC_SETFLAGS		_IOW('f', 2, long)
+#define	EXT3_IOC_GETFLAGS		FS_IOC_GETFLAGS
+#define	EXT3_IOC_SETFLAGS		FS_IOC_SETFLAGS
 #define	EXT3_IOC_GETVERSION		_IOR('f', 3, long)
 #define	EXT3_IOC_SETVERSION		_IOW('f', 4, long)
 #define EXT3_IOC_GROUP_EXTEND		_IOW('f', 7, unsigned long)
 #define EXT3_IOC_GROUP_ADD		_IOW('f', 8,struct ext3_new_group_input)
-#define	EXT3_IOC_GETVERSION_OLD		_IOR('v', 1, long)
-#define	EXT3_IOC_SETVERSION_OLD		_IOW('v', 2, long)
+#define	EXT3_IOC_GETVERSION_OLD		FS_IOC_GETVERSION
+#define	EXT3_IOC_SETVERSION_OLD		FS_IOC_SETVERSION
 #ifdef CONFIG_JBD_DEBUG
 #define EXT3_IOC_WAIT_FOR_READONLY	_IOR('f', 99, long)
 #endif
 #define EXT3_IOC_GETRSVSZ		_IOR('f', 5, long)
 #define EXT3_IOC_SETRSVSZ		_IOW('f', 6, long)
 
+/*
+ * ioctl commands in 32 bit emulation
+ */
+#define EXT3_IOC32_GETVERSION		_IOR('f', 3, int)
+#define EXT3_IOC32_SETVERSION		_IOW('f', 4, int)
+#define EXT3_IOC32_GETRSVSZ		_IOR('f', 5, int)
+#define EXT3_IOC32_SETRSVSZ		_IOW('f', 6, int)
+#define EXT3_IOC32_GROUP_EXTEND		_IOW('f', 7, unsigned int)
+#ifdef CONFIG_JBD_DEBUG
+#define EXT3_IOC32_WAIT_FOR_READONLY	_IOR('f', 99, int)
+#endif
+
 /*
  *  Mount options
  */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1728142ec4b6..b73a47582dbe 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -217,6 +217,45 @@ extern int dir_notify_enable;
 #define FIBMAP	   _IO(0x00,1)	/* bmap access */
 #define FIGETBSZ   _IO(0x00,2)	/* get the block size used for bmap */
 
+#define	FS_IOC_GETFLAGS			_IOR('f', 1, long)
+#define	FS_IOC_SETFLAGS			_IOW('f', 2, long)
+#define	FS_IOC_GETVERSION		_IOR('v', 1, long)
+#define	FS_IOC_SETVERSION		_IOW('v', 2, long)
+#define FS_IOC32_GETFLAGS		_IOR('f', 1, int)
+#define FS_IOC32_SETFLAGS		_IOW('f', 2, int)
+#define FS_IOC32_GETVERSION		_IOR('v', 1, int)
+#define FS_IOC32_SETVERSION		_IOW('v', 2, int)
+
+/*
+ * Inode flags (FS_IOC_GETFLAGS / FS_IOC_SETFLAGS)
+ */
+#define	FS_SECRM_FL			0x00000001 /* Secure deletion */
+#define	FS_UNRM_FL			0x00000002 /* Undelete */
+#define	FS_COMPR_FL			0x00000004 /* Compress file */
+#define FS_SYNC_FL			0x00000008 /* Synchronous updates */
+#define FS_IMMUTABLE_FL			0x00000010 /* Immutable file */
+#define FS_APPEND_FL			0x00000020 /* writes to file may only append */
+#define FS_NODUMP_FL			0x00000040 /* do not dump file */
+#define FS_NOATIME_FL			0x00000080 /* do not update atime */
+/* Reserved for compression usage... */
+#define FS_DIRTY_FL			0x00000100
+#define FS_COMPRBLK_FL			0x00000200 /* One or more compressed clusters */
+#define FS_NOCOMP_FL			0x00000400 /* Don't compress */
+#define FS_ECOMPR_FL			0x00000800 /* Compression error */
+/* End compression flags --- maybe not all used */
+#define FS_BTREE_FL			0x00001000 /* btree format dir */
+#define FS_INDEX_FL			0x00001000 /* hash-indexed directory */
+#define FS_IMAGIC_FL			0x00002000 /* AFS directory */
+#define FS_JOURNAL_DATA_FL		0x00004000 /* Reserved for ext3 */
+#define FS_NOTAIL_FL			0x00008000 /* file tail should not be merged */
+#define FS_DIRSYNC_FL			0x00010000 /* dirsync behaviour (directories only) */
+#define FS_TOPDIR_FL			0x00020000 /* Top of directory hierarchies*/
+#define FS_RESERVED_FL			0x80000000 /* reserved for ext2 lib */
+
+#define FS_FL_USER_VISIBLE		0x0003DFFF /* User visible flags */
+#define FS_FL_USER_MODIFIABLE		0x000380FF /* User modifiable flags */
+
+
 #define SYNC_FILE_RANGE_WAIT_BEFORE	1
 #define SYNC_FILE_RANGE_WRITE		2
 #define SYNC_FILE_RANGE_WAIT_AFTER	4
diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h
index 28493ffaafe7..0100d6d1d84c 100644
--- a/include/linux/reiserfs_fs.h
+++ b/include/linux/reiserfs_fs.h
@@ -807,21 +807,19 @@ struct stat_data_v1 {
 #define set_sd_v1_first_direct_byte(sdp,v) \
                                 ((sdp)->sd_first_direct_byte = cpu_to_le32(v))
 
-#include <linux/ext2_fs.h>
-
 /* inode flags stored in sd_attrs (nee sd_reserved) */
 
 /* we want common flags to have the same values as in ext2,
    so chattr(1) will work without problems */
-#define REISERFS_IMMUTABLE_FL EXT2_IMMUTABLE_FL
-#define REISERFS_APPEND_FL    EXT2_APPEND_FL
-#define REISERFS_SYNC_FL      EXT2_SYNC_FL
-#define REISERFS_NOATIME_FL   EXT2_NOATIME_FL
-#define REISERFS_NODUMP_FL    EXT2_NODUMP_FL
-#define REISERFS_SECRM_FL     EXT2_SECRM_FL
-#define REISERFS_UNRM_FL      EXT2_UNRM_FL
-#define REISERFS_COMPR_FL     EXT2_COMPR_FL
-#define REISERFS_NOTAIL_FL    EXT2_NOTAIL_FL
+#define REISERFS_IMMUTABLE_FL FS_IMMUTABLE_FL
+#define REISERFS_APPEND_FL    FS_APPEND_FL
+#define REISERFS_SYNC_FL      FS_SYNC_FL
+#define REISERFS_NOATIME_FL   FS_NOATIME_FL
+#define REISERFS_NODUMP_FL    FS_NODUMP_FL
+#define REISERFS_SECRM_FL     FS_SECRM_FL
+#define REISERFS_UNRM_FL      FS_UNRM_FL
+#define REISERFS_COMPR_FL     FS_COMPR_FL
+#define REISERFS_NOTAIL_FL    FS_NOTAIL_FL
 
 /* persistent flags that file inherits from the parent directory */
 #define REISERFS_INHERIT_MASK ( REISERFS_IMMUTABLE_FL |	\
@@ -2168,10 +2166,10 @@ int reiserfs_ioctl(struct inode *inode, struct file *filp,
 #define REISERFS_IOC_UNPACK		_IOW(0xCD,1,long)
 /* define following flags to be the same as in ext2, so that chattr(1),
    lsattr(1) will work with us. */
-#define REISERFS_IOC_GETFLAGS		EXT2_IOC_GETFLAGS
-#define REISERFS_IOC_SETFLAGS		EXT2_IOC_SETFLAGS
-#define REISERFS_IOC_GETVERSION		EXT2_IOC_GETVERSION
-#define REISERFS_IOC_SETVERSION		EXT2_IOC_SETVERSION
+#define REISERFS_IOC_GETFLAGS		FS_IOC_GETFLAGS
+#define REISERFS_IOC_SETFLAGS		FS_IOC_SETFLAGS
+#define REISERFS_IOC_GETVERSION		FS_IOC_GETVERSION
+#define REISERFS_IOC_SETVERSION		FS_IOC_SETVERSION
 
 /* Locking primitives */
 /* Right now we are still falling back to (un)lock_kernel, but eventually that
-- 
cgit v1.2.3


From 52b499c438ff60991eb3855ca090782569b3e8cf Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:06:18 +0100
Subject: [PATCH] BLOCK: Move the ReiserFS device ioctl compat stuff to the
 ReiserFS driver [try #6]

Move the ReiserFS device ioctl compat stuff from fs/compat_ioctl.c to the
ReiserFS driver so that the ReiserFS header file doesn't need to be included.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/compat_ioctl.c           | 12 ------------
 fs/reiserfs/dir.c           |  3 +++
 fs/reiserfs/file.c          |  4 ++++
 fs/reiserfs/ioctl.c         | 35 +++++++++++++++++++++++++++++++++++
 include/linux/reiserfs_fs.h |  9 +++++++++
 5 files changed, 51 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index bd9c4f49d4e5..0346f2ab57c4 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -59,7 +59,6 @@
 #include <linux/pci.h>
 #include <linux/module.h>
 #include <linux/serial.h>
-#include <linux/reiserfs_fs.h>
 #include <linux/if_tun.h>
 #include <linux/ctype.h>
 #include <linux/ioctl32.h>
@@ -2014,16 +2013,6 @@ static int vfat_ioctl32(unsigned fd, unsigned cmd, unsigned long arg)
 	return ret;
 }
 
-#define REISERFS_IOC_UNPACK32               _IOW(0xCD,1,int)
-
-static int reiserfs_ioctl32(unsigned fd, unsigned cmd, unsigned long ptr)
-{
-        if (cmd == REISERFS_IOC_UNPACK32)
-                cmd = REISERFS_IOC_UNPACK;
-
-        return sys_ioctl(fd,cmd,ptr);
-}
-
 struct raw32_config_request
 {
         compat_int_t    raw_minor;
@@ -2784,7 +2773,6 @@ HANDLE_IOCTL(BLKGETSIZE64_32, do_blkgetsize64)
 /* vfat */
 HANDLE_IOCTL(VFAT_IOCTL_READDIR_BOTH32, vfat_ioctl32)
 HANDLE_IOCTL(VFAT_IOCTL_READDIR_SHORT32, vfat_ioctl32)
-HANDLE_IOCTL(REISERFS_IOC_UNPACK32, reiserfs_ioctl32)
 /* Raw devices */
 HANDLE_IOCTL(RAW_SETBIND, raw_ioctl)
 HANDLE_IOCTL(RAW_GETBIND, raw_ioctl)
diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c
index 9aabcc0ccd2d..657050ad7430 100644
--- a/fs/reiserfs/dir.c
+++ b/fs/reiserfs/dir.c
@@ -22,6 +22,9 @@ const struct file_operations reiserfs_dir_operations = {
 	.readdir = reiserfs_readdir,
 	.fsync = reiserfs_dir_fsync,
 	.ioctl = reiserfs_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = reiserfs_compat_ioctl,
+#endif
 };
 
 static int reiserfs_dir_fsync(struct file *filp, struct dentry *dentry,
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index 1cfbe857ba27..3e08f7161a3d 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -2,6 +2,7 @@
  * Copyright 2000 by Hans Reiser, licensing governed by reiserfs/README
  */
 
+#include <linux/config.h>
 #include <linux/time.h>
 #include <linux/reiserfs_fs.h>
 #include <linux/reiserfs_acl.h>
@@ -1568,6 +1569,9 @@ const struct file_operations reiserfs_file_operations = {
 	.read = generic_file_read,
 	.write = reiserfs_file_write,
 	.ioctl = reiserfs_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = reiserfs_compat_ioctl,
+#endif
 	.mmap = generic_file_mmap,
 	.release = reiserfs_file_release,
 	.fsync = reiserfs_sync_file,
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index a986b5e1e288..9c57578cb831 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -9,6 +9,7 @@
 #include <asm/uaccess.h>
 #include <linux/pagemap.h>
 #include <linux/smp_lock.h>
+#include <linux/compat.h>
 
 static int reiserfs_unpack(struct inode *inode, struct file *filp);
 
@@ -94,6 +95,40 @@ int reiserfs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
 	}
 }
 
+#ifdef CONFIG_COMPAT
+long reiserfs_compat_ioctl(struct file *file, unsigned int cmd,
+				unsigned long arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	int ret;
+
+	/* These are just misnamed, they actually get/put from/to user an int */
+	switch (cmd) {
+	case REISERFS_IOC32_UNPACK:
+		cmd = REISERFS_IOC_UNPACK;
+		break;
+	case REISERFS_IOC32_GETFLAGS:
+		cmd = REISERFS_IOC_GETFLAGS;
+		break;
+	case REISERFS_IOC32_SETFLAGS:
+		cmd = REISERFS_IOC_SETFLAGS;
+		break;
+	case REISERFS_IOC32_GETVERSION:
+		cmd = REISERFS_IOC_GETVERSION;
+		break;
+	case REISERFS_IOC32_SETVERSION:
+		cmd = REISERFS_IOC_SETVERSION;
+		break;
+	default:
+		return -ENOIOCTLCMD;
+	}
+	lock_kernel();
+	ret = reiserfs_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg));
+	unlock_kernel();
+	return ret;
+}
+#endif
+
 /*
 ** reiserfs_unpack
 ** Function try to convert tail from direct item into indirect.
diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h
index 0100d6d1d84c..9c63abffd7b2 100644
--- a/include/linux/reiserfs_fs.h
+++ b/include/linux/reiserfs_fs.h
@@ -2161,6 +2161,8 @@ __u32 r5_hash(const signed char *msg, int len);
 /* prototypes from ioctl.c */
 int reiserfs_ioctl(struct inode *inode, struct file *filp,
 		   unsigned int cmd, unsigned long arg);
+long reiserfs_compat_ioctl(struct file *filp,
+		   unsigned int cmd, unsigned long arg);
 
 /* ioctl's command */
 #define REISERFS_IOC_UNPACK		_IOW(0xCD,1,long)
@@ -2171,6 +2173,13 @@ int reiserfs_ioctl(struct inode *inode, struct file *filp,
 #define REISERFS_IOC_GETVERSION		FS_IOC_GETVERSION
 #define REISERFS_IOC_SETVERSION		FS_IOC_SETVERSION
 
+/* the 32 bit compat definitions with int argument */
+#define REISERFS_IOC32_UNPACK		_IOW(0xCD, 1, int)
+#define REISERFS_IOC32_GETFLAGS		FS_IOC32_GETFLAGS
+#define REISERFS_IOC32_SETFLAGS		FS_IOC32_SETFLAGS
+#define REISERFS_IOC32_GETVERSION	FS_IOC32_GETVERSION
+#define REISERFS_IOC32_SETVERSION	FS_IOC32_SETVERSION
+
 /* Locking primitives */
 /* Right now we are still falling back to (un)lock_kernel, but eventually that
    would evolve into real per-fs locks */
-- 
cgit v1.2.3


From 52a700c5675f399c07e6e57328291e57f13ef3bb Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 29 Aug 2006 19:06:23 +0100
Subject: [PATCH] BLOCK: Move the Ext3 device ioctl compat stuff to the Ext3
 driver [try #6]

Move the Ext3 device ioctl compat stuff from fs/compat_ioctl.c to the Ext3
driver so that the Ext3 header file doesn't need to be included.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/compat_ioctl.c       | 27 ------------------------
 fs/ext3/dir.c           |  3 +++
 fs/ext3/file.c          |  3 +++
 fs/ext3/ioctl.c         | 55 ++++++++++++++++++++++++++++++++++++++++++++++++-
 include/linux/ext3_fs.h |  6 ++++++
 5 files changed, 66 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 3594668559af..e5eb0f10f05a 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -45,8 +45,6 @@
 #include <linux/tty.h>
 #include <linux/vt_kern.h>
 #include <linux/fb.h>
-#include <linux/ext3_jbd.h>
-#include <linux/ext3_fs.h>
 #include <linux/videodev.h>
 #include <linux/netdevice.h>
 #include <linux/raw.h>
@@ -158,22 +156,6 @@ static int rw_long(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return err;
 }
 
-static int do_ext3_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	/* These are just misnamed, they actually get/put from/to user an int */
-	switch (cmd) {
-	case EXT3_IOC32_GETVERSION: cmd = EXT3_IOC_GETVERSION; break;
-	case EXT3_IOC32_SETVERSION: cmd = EXT3_IOC_SETVERSION; break;
-	case EXT3_IOC32_GETRSVSZ: cmd = EXT3_IOC_GETRSVSZ; break;
-	case EXT3_IOC32_SETRSVSZ: cmd = EXT3_IOC_SETRSVSZ; break;
-	case EXT3_IOC32_GROUP_EXTEND: cmd = EXT3_IOC_GROUP_EXTEND; break;
-#ifdef CONFIG_JBD_DEBUG
-	case EXT3_IOC32_WAIT_FOR_READONLY: cmd = EXT3_IOC_WAIT_FOR_READONLY; break;
-#endif
-	}
-	return sys_ioctl(fd, cmd, (unsigned long)compat_ptr(arg));
-}
-
 struct compat_video_event {
 	int32_t		type;
 	compat_time_t	timestamp;
@@ -2712,15 +2694,6 @@ HANDLE_IOCTL(PIO_UNIMAP, do_unimap_ioctl)
 HANDLE_IOCTL(GIO_UNIMAP, do_unimap_ioctl)
 HANDLE_IOCTL(KDFONTOP, do_kdfontop_ioctl)
 #endif
-HANDLE_IOCTL(EXT3_IOC32_GETVERSION, do_ext3_ioctl)
-HANDLE_IOCTL(EXT3_IOC32_SETVERSION, do_ext3_ioctl)
-HANDLE_IOCTL(EXT3_IOC32_GETRSVSZ, do_ext3_ioctl)
-HANDLE_IOCTL(EXT3_IOC32_SETRSVSZ, do_ext3_ioctl)
-HANDLE_IOCTL(EXT3_IOC32_GROUP_EXTEND, do_ext3_ioctl)
-COMPATIBLE_IOCTL(EXT3_IOC_GROUP_ADD)
-#ifdef CONFIG_JBD_DEBUG
-HANDLE_IOCTL(EXT3_IOC32_WAIT_FOR_READONLY, do_ext3_ioctl)
-#endif
 /* One SMB ioctl needs translations. */
 #define SMB_IOC_GETMOUNTUID_32 _IOR('u', 1, compat_uid_t)
 HANDLE_IOCTL(SMB_IOC_GETMOUNTUID_32, do_smb_getmountuid)
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 429acbb4e064..d0b54f30b914 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -44,6 +44,9 @@ const struct file_operations ext3_dir_operations = {
 	.read		= generic_read_dir,
 	.readdir	= ext3_readdir,		/* we take BKL. needed?*/
 	.ioctl		= ext3_ioctl,		/* BKL held */
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= ext3_compat_ioctl,
+#endif
 	.fsync		= ext3_sync_file,	/* BKL held */
 #ifdef CONFIG_EXT3_INDEX
 	.release	= ext3_release_dir,
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 994efd189f4e..74ff20f9d09b 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -114,6 +114,9 @@ const struct file_operations ext3_file_operations = {
 	.readv		= generic_file_readv,
 	.writev		= generic_file_writev,
 	.ioctl		= ext3_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= ext3_compat_ioctl,
+#endif
 	.mmap		= generic_file_mmap,
 	.open		= generic_file_open,
 	.release	= ext3_release_file,
diff --git a/fs/ext3/ioctl.c b/fs/ext3/ioctl.c
index 3a6b012d120c..12daa6869572 100644
--- a/fs/ext3/ioctl.c
+++ b/fs/ext3/ioctl.c
@@ -13,9 +13,10 @@
 #include <linux/ext3_fs.h>
 #include <linux/ext3_jbd.h>
 #include <linux/time.h>
+#include <linux/compat.h>
+#include <linux/smp_lock.h>
 #include <asm/uaccess.h>
 
-
 int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
 		unsigned long arg)
 {
@@ -252,3 +253,55 @@ flags_err:
 		return -ENOTTY;
 	}
 }
+
+#ifdef CONFIG_COMPAT
+long ext3_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	int ret;
+
+	/* These are just misnamed, they actually get/put from/to user an int */
+	switch (cmd) {
+	case EXT3_IOC32_GETFLAGS:
+		cmd = EXT3_IOC_GETFLAGS;
+		break;
+	case EXT3_IOC32_SETFLAGS:
+		cmd = EXT3_IOC_SETFLAGS;
+		break;
+	case EXT3_IOC32_GETVERSION:
+		cmd = EXT3_IOC_GETVERSION;
+		break;
+	case EXT3_IOC32_SETVERSION:
+		cmd = EXT3_IOC_SETVERSION;
+		break;
+	case EXT3_IOC32_GROUP_EXTEND:
+		cmd = EXT3_IOC_GROUP_EXTEND;
+		break;
+	case EXT3_IOC32_GETVERSION_OLD:
+		cmd = EXT3_IOC_GETVERSION_OLD;
+		break;
+	case EXT3_IOC32_SETVERSION_OLD:
+		cmd = EXT3_IOC_SETVERSION_OLD;
+		break;
+#ifdef CONFIG_JBD_DEBUG
+	case EXT3_IOC32_WAIT_FOR_READONLY:
+		cmd = EXT3_IOC_WAIT_FOR_READONLY;
+		break;
+#endif
+	case EXT3_IOC32_GETRSVSZ:
+		cmd = EXT3_IOC_GETRSVSZ;
+		break;
+	case EXT3_IOC32_SETRSVSZ:
+		cmd = EXT3_IOC_SETRSVSZ;
+		break;
+	case EXT3_IOC_GROUP_ADD:
+		break;
+	default:
+		return -ENOIOCTLCMD;
+	}
+	lock_kernel();
+	ret = ext3_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg));
+	unlock_kernel();
+	return ret;
+}
+#endif
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index a7a01ff44669..11cca1bdc0c7 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -233,6 +233,8 @@ struct ext3_new_group_data {
 /*
  * ioctl commands in 32 bit emulation
  */
+#define EXT3_IOC32_GETFLAGS		FS_IOC32_GETFLAGS
+#define EXT3_IOC32_SETFLAGS		FS_IOC32_SETFLAGS
 #define EXT3_IOC32_GETVERSION		_IOR('f', 3, int)
 #define EXT3_IOC32_SETVERSION		_IOW('f', 4, int)
 #define EXT3_IOC32_GETRSVSZ		_IOR('f', 5, int)
@@ -241,6 +243,9 @@ struct ext3_new_group_data {
 #ifdef CONFIG_JBD_DEBUG
 #define EXT3_IOC32_WAIT_FOR_READONLY	_IOR('f', 99, int)
 #endif
+#define EXT3_IOC32_GETVERSION_OLD	FS_IOC32_GETVERSION
+#define EXT3_IOC32_SETVERSION_OLD	FS_IOC32_SETVERSION
+
 
 /*
  *  Mount options
@@ -824,6 +829,7 @@ extern void ext3_set_aops(struct inode *inode);
 /* ioctl.c */
 extern int ext3_ioctl (struct inode *, struct file *, unsigned int,
 		       unsigned long);
+extern long ext3_compat_ioctl (struct file *, unsigned int, unsigned long);
 
 /* namei.c */
 extern int ext3_orphan_add(handle_t *, struct inode *);
-- 
cgit v1.2.3


From 9361401eb7619c033e2394e4f9f6d410d6719ac7 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Sat, 30 Sep 2006 20:45:40 +0200
Subject: [PATCH] BLOCK: Make it possible to disable the block layer [try #6]

Make it possible to disable the block layer.  Not all embedded devices require
it, some can make do with just JFFS2, NFS, ramfs, etc - none of which require
the block layer to be present.

This patch does the following:

 (*) Introduces CONFIG_BLOCK to disable the block layer, buffering and blockdev
     support.

 (*) Adds dependencies on CONFIG_BLOCK to any configuration item that controls
     an item that uses the block layer.  This includes:

     (*) Block I/O tracing.

     (*) Disk partition code.

     (*) All filesystems that are block based, eg: Ext3, ReiserFS, ISOFS.

     (*) The SCSI layer.  As far as I can tell, even SCSI chardevs use the
     	 block layer to do scheduling.  Some drivers that use SCSI facilities -
     	 such as USB storage - end up disabled indirectly from this.

     (*) Various block-based device drivers, such as IDE and the old CDROM
     	 drivers.

     (*) MTD blockdev handling and FTL.

     (*) JFFS - which uses set_bdev_super(), something it could avoid doing by
     	 taking a leaf out of JFFS2's book.

 (*) Makes most of the contents of linux/blkdev.h, linux/buffer_head.h and
     linux/elevator.h contingent on CONFIG_BLOCK being set.  sector_div() is,
     however, still used in places, and so is still available.

 (*) Also made contingent are the contents of linux/mpage.h, linux/genhd.h and
     parts of linux/fs.h.

 (*) Makes a number of files in fs/ contingent on CONFIG_BLOCK.

 (*) Makes mm/bounce.c (bounce buffering) contingent on CONFIG_BLOCK.

 (*) set_page_dirty() doesn't call __set_page_dirty_buffers() if CONFIG_BLOCK
     is not enabled.

 (*) fs/no-block.c is created to hold out-of-line stubs and things that are
     required when CONFIG_BLOCK is not set:

     (*) Default blockdev file operations (to give error ENODEV on opening).

 (*) Makes some /proc changes:

     (*) /proc/devices does not list any blockdevs.

     (*) /proc/diskstats and /proc/partitions are contingent on CONFIG_BLOCK.

 (*) Makes some compat ioctl handling contingent on CONFIG_BLOCK.

 (*) If CONFIG_BLOCK is not defined, makes sys_quotactl() return -ENODEV if
     given command other than Q_SYNC or if a special device is specified.

 (*) In init/do_mounts.c, no reference is made to the blockdev routines if
     CONFIG_BLOCK is not defined.  This does not prohibit NFS roots or JFFS2.

 (*) The bdflush, ioprio_set and ioprio_get syscalls can now be absent (return
     error ENOSYS by way of cond_syscall if so).

 (*) The seclvl_bd_claim() and seclvl_bd_release() security calls do nothing if
     CONFIG_BLOCK is not set, since they can't then happen.

Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/Kconfig                | 20 ++++++++++++++++++
 block/Kconfig.iosched        |  3 +++
 block/Makefile               |  2 +-
 drivers/block/Kconfig        |  4 ++++
 drivers/cdrom/Kconfig        |  2 +-
 drivers/char/Kconfig         |  1 +
 drivers/char/random.c        |  4 ++++
 drivers/ide/Kconfig          |  4 ++++
 drivers/md/Kconfig           |  3 +++
 drivers/message/i2o/Kconfig  |  2 +-
 drivers/mmc/Kconfig          |  2 +-
 drivers/mmc/Makefile         |  3 ++-
 drivers/mtd/Kconfig          | 12 +++++------
 drivers/mtd/devices/Kconfig  |  2 +-
 drivers/s390/block/Kconfig   |  2 +-
 drivers/scsi/Kconfig         |  2 ++
 fs/Kconfig                   | 31 ++++++++++++++++++++-------
 fs/Makefile                  | 14 +++++++++----
 fs/compat_ioctl.c            | 18 ++++++++++++++++
 fs/internal.h                |  6 ++++++
 fs/no-block.c                | 22 +++++++++++++++++++
 fs/partitions/Makefile       |  2 +-
 fs/proc/proc_misc.c          | 11 +++++++++-
 fs/quota.c                   | 44 ++++++++++++++++++++++++++------------
 fs/super.c                   |  4 ++++
 fs/xfs/Kconfig               |  1 +
 include/linux/blkdev.h       | 50 +++++++++++++++++++++++++++++++-------------
 include/linux/buffer_head.h  | 16 ++++++++++++++
 include/linux/compat_ioctl.h |  2 ++
 include/linux/elevator.h     |  3 +++
 include/linux/fs.h           | 25 +++++++++++++++++++---
 include/linux/genhd.h        |  4 ++++
 include/linux/mpage.h        |  3 +++
 include/linux/raid/md.h      |  3 +++
 include/linux/raid/md_k.h    |  3 +++
 include/scsi/scsi_tcq.h      |  3 ++-
 init/Kconfig                 |  2 +-
 init/do_mounts.c             | 13 +++++++++++-
 kernel/sys_ni.c              |  5 +++++
 mm/Makefile                  |  2 +-
 mm/filemap.c                 |  4 ++++
 mm/migrate.c                 |  2 ++
 mm/page-writeback.c          |  8 ++++---
 mm/truncate.c                |  2 ++
 44 files changed, 308 insertions(+), 63 deletions(-)
 create mode 100644 fs/no-block.c

(limited to 'include')

diff --git a/block/Kconfig b/block/Kconfig
index b6f5f0a79655..9af6c614dfde 100644
--- a/block/Kconfig
+++ b/block/Kconfig
@@ -1,6 +1,24 @@
 #
 # Block layer core configuration
 #
+config BLOCK
+       bool "Enable the block layer"
+       default y
+       help
+	 This permits the block layer to be removed from the kernel if it's not
+	 needed (on some embedded devices for example).  If this option is
+	 disabled, then blockdev files will become unusable and some
+	 filesystems (such as ext3) will become unavailable.
+
+	 This option will also disable SCSI character devices and USB storage
+	 since they make use of various block layer definitions and
+	 facilities.
+
+	 Say Y here unless you know you really don't want to mount disks and
+	 suchlike.
+
+if BLOCK
+
 #XXX - it makes sense to enable this only for 32-bit subarch's, not for x86_64
 #for instance.
 config LBD
@@ -33,4 +51,6 @@ config LSF
 
 	  If unsure, say Y.
 
+endif
+
 source block/Kconfig.iosched
diff --git a/block/Kconfig.iosched b/block/Kconfig.iosched
index 48d090e266fc..903f0d3b6852 100644
--- a/block/Kconfig.iosched
+++ b/block/Kconfig.iosched
@@ -1,3 +1,4 @@
+if BLOCK
 
 menu "IO Schedulers"
 
@@ -67,3 +68,5 @@ config DEFAULT_IOSCHED
 	default "noop" if DEFAULT_NOOP
 
 endmenu
+
+endif
diff --git a/block/Makefile b/block/Makefile
index c05de0e0037f..4b84d0d5947b 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the kernel block layer
 #
 
-obj-y	:= elevator.o ll_rw_blk.o ioctl.o genhd.o scsi_ioctl.o
+obj-$(CONFIG_BLOCK) := elevator.o ll_rw_blk.o ioctl.o genhd.o scsi_ioctl.o
 
 obj-$(CONFIG_IOSCHED_NOOP)	+= noop-iosched.o
 obj-$(CONFIG_IOSCHED_AS)	+= as-iosched.o
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index b5382cedf0c0..422e31d5f8e5 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -2,6 +2,8 @@
 # Block device driver configuration
 #
 
+if BLOCK
+
 menu "Block devices"
 
 config BLK_DEV_FD
@@ -468,3 +470,5 @@ config ATA_OVER_ETH
 	devices like the Coraid EtherDrive (R) Storage Blade.
 
 endmenu
+
+endif
diff --git a/drivers/cdrom/Kconfig b/drivers/cdrom/Kconfig
index ff5652d40619..4b12e9031fb3 100644
--- a/drivers/cdrom/Kconfig
+++ b/drivers/cdrom/Kconfig
@@ -3,7 +3,7 @@
 #
 
 menu "Old CD-ROM drivers (not SCSI, not IDE)"
-	depends on ISA
+	depends on ISA && BLOCK
 
 config CD_NO_IDESCSI
 	bool "Support non-SCSI/IDE/ATAPI CDROM drives"
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 4cc619edf424..bde1c665d9f4 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -1006,6 +1006,7 @@ config GPIO_VR41XX
 
 config RAW_DRIVER
 	tristate "RAW driver (/dev/raw/rawN) (OBSOLETE)"
+	depends on BLOCK
 	help
 	  The raw driver permits block devices to be bound to /dev/raw/rawN. 
 	  Once bound, I/O against /dev/raw/rawN uses efficient zero-copy I/O. 
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 4c3a5ca9d8f7..b430a12eb819 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -655,6 +655,7 @@ void add_interrupt_randomness(int irq)
 	add_timer_randomness(irq_timer_state[irq], 0x100 + irq);
 }
 
+#ifdef CONFIG_BLOCK
 void add_disk_randomness(struct gendisk *disk)
 {
 	if (!disk || !disk->random)
@@ -667,6 +668,7 @@ void add_disk_randomness(struct gendisk *disk)
 }
 
 EXPORT_SYMBOL(add_disk_randomness);
+#endif
 
 #define EXTRACT_SIZE 10
 
@@ -918,6 +920,7 @@ void rand_initialize_irq(int irq)
 	}
 }
 
+#ifdef CONFIG_BLOCK
 void rand_initialize_disk(struct gendisk *disk)
 {
 	struct timer_rand_state *state;
@@ -932,6 +935,7 @@ void rand_initialize_disk(struct gendisk *disk)
 		disk->random = state;
 	}
 }
+#endif
 
 static ssize_t
 random_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos)
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index b6fb167e20f6..69d627bd537a 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -4,6 +4,8 @@
 # Andre Hedrick <andre@linux-ide.org>
 #
 
+if BLOCK
+
 menu "ATA/ATAPI/MFM/RLL support"
 
 config IDE
@@ -1082,3 +1084,5 @@ config BLK_DEV_HD
 endif
 
 endmenu
+
+endif
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index bf869ed03eed..6dd31a291d84 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -2,6 +2,8 @@
 # Block device driver configuration
 #
 
+if BLOCK
+
 menu "Multi-device support (RAID and LVM)"
 
 config MD
@@ -251,3 +253,4 @@ config DM_MULTIPATH_EMC
 
 endmenu
 
+endif
diff --git a/drivers/message/i2o/Kconfig b/drivers/message/i2o/Kconfig
index fef677103880..6443392bffff 100644
--- a/drivers/message/i2o/Kconfig
+++ b/drivers/message/i2o/Kconfig
@@ -88,7 +88,7 @@ config I2O_BUS
 
 config I2O_BLOCK
 	tristate "I2O Block OSM"
-	depends on I2O
+	depends on I2O && BLOCK
 	---help---
 	  Include support for the I2O Block OSM. The Block OSM presents disk
 	  and other structured block devices to the operating system. If you
diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
index 45bcf098e762..f540bd88dc5a 100644
--- a/drivers/mmc/Kconfig
+++ b/drivers/mmc/Kconfig
@@ -21,7 +21,7 @@ config MMC_DEBUG
 
 config MMC_BLOCK
 	tristate "MMC block device driver"
-	depends on MMC
+	depends on MMC && BLOCK
 	default y
 	help
 	  Say Y here to enable the MMC block device driver support.
diff --git a/drivers/mmc/Makefile b/drivers/mmc/Makefile
index d2957e35cc6f..b1f6e03e7aa9 100644
--- a/drivers/mmc/Makefile
+++ b/drivers/mmc/Makefile
@@ -24,7 +24,8 @@ obj-$(CONFIG_MMC_AU1X)		+= au1xmmc.o
 obj-$(CONFIG_MMC_OMAP)		+= omap.o
 obj-$(CONFIG_MMC_AT91RM9200)	+= at91_mci.o
 
-mmc_core-y := mmc.o mmc_queue.o mmc_sysfs.o
+mmc_core-y := mmc.o mmc_sysfs.o
+mmc_core-$(CONFIG_BLOCK) += mmc_queue.o
 
 ifeq ($(CONFIG_MMC_DEBUG),y)
 EXTRA_CFLAGS += -DDEBUG
diff --git a/drivers/mtd/Kconfig b/drivers/mtd/Kconfig
index a03e862851db..a304b34c2632 100644
--- a/drivers/mtd/Kconfig
+++ b/drivers/mtd/Kconfig
@@ -166,7 +166,7 @@ config MTD_CHAR
 
 config MTD_BLOCK
 	tristate "Caching block device access to MTD devices"
-	depends on MTD
+	depends on MTD && BLOCK
 	---help---
 	  Although most flash chips have an erase size too large to be useful
 	  as block devices, it is possible to use MTD devices which are based
@@ -188,7 +188,7 @@ config MTD_BLOCK
 
 config MTD_BLOCK_RO
 	tristate "Readonly block device access to MTD devices"
-	depends on MTD_BLOCK!=y && MTD
+	depends on MTD_BLOCK!=y && MTD && BLOCK
 	help
 	  This allows you to mount read-only file systems (such as cramfs)
 	  from an MTD device, without the overhead (and danger) of the caching
@@ -199,7 +199,7 @@ config MTD_BLOCK_RO
 
 config FTL
 	tristate "FTL (Flash Translation Layer) support"
-	depends on MTD
+	depends on MTD && BLOCK
 	---help---
 	  This provides support for the original Flash Translation Layer which
 	  is part of the PCMCIA specification. It uses a kind of pseudo-
@@ -215,7 +215,7 @@ config FTL
 
 config NFTL
 	tristate "NFTL (NAND Flash Translation Layer) support"
-	depends on MTD
+	depends on MTD && BLOCK
 	---help---
 	  This provides support for the NAND Flash Translation Layer which is
 	  used on M-Systems' DiskOnChip devices. It uses a kind of pseudo-
@@ -238,7 +238,7 @@ config NFTL_RW
 
 config INFTL
 	tristate "INFTL (Inverse NAND Flash Translation Layer) support"
-	depends on MTD
+	depends on MTD && BLOCK
 	---help---
 	  This provides support for the Inverse NAND Flash Translation
 	  Layer which is used on M-Systems' newer DiskOnChip devices. It
@@ -255,7 +255,7 @@ config INFTL
 
 config RFD_FTL
         tristate "Resident Flash Disk (Flash Translation Layer) support"
-	depends on MTD
+	depends on MTD && BLOCK
 	---help---
 	  This provides support for the flash translation layer known
 	  as the Resident Flash Disk (RFD), as used by the Embedded BIOS
diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig
index 16c02b5ccf7e..440f6851da69 100644
--- a/drivers/mtd/devices/Kconfig
+++ b/drivers/mtd/devices/Kconfig
@@ -136,7 +136,7 @@ config MTDRAM_ABS_POS
 
 config MTD_BLOCK2MTD
 	tristate "MTD using block device"
-	depends on MTD
+	depends on MTD && BLOCK
 	help
 	  This driver allows a block device to appear as an MTD. It would
 	  generally be used in the following cases:
diff --git a/drivers/s390/block/Kconfig b/drivers/s390/block/Kconfig
index 929d6fff6152..b250c5354503 100644
--- a/drivers/s390/block/Kconfig
+++ b/drivers/s390/block/Kconfig
@@ -1,4 +1,4 @@
-if S390
+if S390 && BLOCK
 
 comment "S/390 block device drivers"
 	depends on S390
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index c4dfcc91ddda..dab082002e6f 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -3,11 +3,13 @@ menu "SCSI device support"
 config RAID_ATTRS
 	tristate "RAID Transport Class"
 	default n
+	depends on BLOCK
 	---help---
 	  Provides RAID
 
 config SCSI
 	tristate "SCSI device support"
+	depends on BLOCK
 	---help---
 	  If you want to use a SCSI hard disk, SCSI tape drive, SCSI CD-ROM or
 	  any other SCSI device under Linux, say Y and make sure that you know
diff --git a/fs/Kconfig b/fs/Kconfig
index 4fd9efac29ab..1453d2d164f7 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -4,6 +4,8 @@
 
 menu "File systems"
 
+if BLOCK
+
 config EXT2_FS
 	tristate "Second extended fs support"
 	help
@@ -399,6 +401,8 @@ config ROMFS_FS
 	  If you don't know whether you need it, then you don't need it:
 	  answer N.
 
+endif
+
 config INOTIFY
 	bool "Inotify file change notification support"
 	default y
@@ -530,6 +534,7 @@ config FUSE_FS
 	  If you want to develop a userspace FS, or if you want to use
 	  a filesystem based on FUSE, answer Y or M.
 
+if BLOCK
 menu "CD-ROM/DVD Filesystems"
 
 config ISO9660_FS
@@ -597,7 +602,9 @@ config UDF_NLS
 	depends on (UDF_FS=m && NLS) || (UDF_FS=y && NLS=y)
 
 endmenu
+endif
 
+if BLOCK
 menu "DOS/FAT/NT Filesystems"
 
 config FAT_FS
@@ -782,6 +789,7 @@ config NTFS_RW
 	  It is perfectly safe to say N here.
 
 endmenu
+endif
 
 menu "Pseudo filesystems"
 
@@ -939,7 +947,7 @@ menu "Miscellaneous filesystems"
 
 config ADFS_FS
 	tristate "ADFS file system support (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	depends on BLOCK && EXPERIMENTAL
 	help
 	  The Acorn Disc Filing System is the standard file system of the
 	  RiscOS operating system which runs on Acorn's ARM-based Risc PC
@@ -967,7 +975,7 @@ config ADFS_FS_RW
 
 config AFFS_FS
 	tristate "Amiga FFS file system support (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	depends on BLOCK && EXPERIMENTAL
 	help
 	  The Fast File System (FFS) is the common file system used on hard
 	  disks by Amiga(tm) systems since AmigaOS Version 1.3 (34.20).  Say Y
@@ -989,7 +997,7 @@ config AFFS_FS
 
 config HFS_FS
 	tristate "Apple Macintosh file system support (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	depends on BLOCK && EXPERIMENTAL
 	select NLS
 	help
 	  If you say Y here, you will be able to mount Macintosh-formatted
@@ -1002,6 +1010,7 @@ config HFS_FS
 
 config HFSPLUS_FS
 	tristate "Apple Extended HFS file system support"
+	depends on BLOCK
 	select NLS
 	select NLS_UTF8
 	help
@@ -1015,7 +1024,7 @@ config HFSPLUS_FS
 
 config BEFS_FS
 	tristate "BeOS file system (BeFS) support (read only) (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	depends on BLOCK && EXPERIMENTAL
 	select NLS
 	help
 	  The BeOS File System (BeFS) is the native file system of Be, Inc's
@@ -1042,7 +1051,7 @@ config BEFS_DEBUG
 
 config BFS_FS
 	tristate "BFS file system support (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	depends on BLOCK && EXPERIMENTAL
 	help
 	  Boot File System (BFS) is a file system used under SCO UnixWare to
 	  allow the bootloader access to the kernel image and other important
@@ -1064,7 +1073,7 @@ config BFS_FS
 
 config EFS_FS
 	tristate "EFS file system support (read only) (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	depends on BLOCK && EXPERIMENTAL
 	help
 	  EFS is an older file system used for non-ISO9660 CD-ROMs and hard
 	  disk partitions by SGI's IRIX operating system (IRIX 6.0 and newer
@@ -1079,7 +1088,7 @@ config EFS_FS
 
 config JFFS_FS
 	tristate "Journalling Flash File System (JFFS) support"
-	depends on MTD
+	depends on MTD && BLOCK
 	help
 	  JFFS is the Journaling Flash File System developed by Axis
 	  Communications in Sweden, aimed at providing a crash/powerdown-safe
@@ -1264,6 +1273,7 @@ endchoice
 
 config CRAMFS
 	tristate "Compressed ROM file system support (cramfs)"
+	depends on BLOCK
 	select ZLIB_INFLATE
 	help
 	  Saying Y here includes support for CramFs (Compressed ROM File
@@ -1283,6 +1293,7 @@ config CRAMFS
 
 config VXFS_FS
 	tristate "FreeVxFS file system support (VERITAS VxFS(TM) compatible)"
+	depends on BLOCK
 	help
 	  FreeVxFS is a file system driver that support the VERITAS VxFS(TM)
 	  file system format.  VERITAS VxFS(TM) is the standard file system
@@ -1300,6 +1311,7 @@ config VXFS_FS
 
 config HPFS_FS
 	tristate "OS/2 HPFS file system support"
+	depends on BLOCK
 	help
 	  OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS
 	  is the file system used for organizing files on OS/2 hard disk
@@ -1316,6 +1328,7 @@ config HPFS_FS
 
 config QNX4FS_FS
 	tristate "QNX4 file system support (read only)"
+	depends on BLOCK
 	help
 	  This is the file system used by the real-time operating systems
 	  QNX 4 and QNX 6 (the latter is also called QNX RTP).
@@ -1343,6 +1356,7 @@ config QNX4FS_RW
 
 config SYSV_FS
 	tristate "System V/Xenix/V7/Coherent file system support"
+	depends on BLOCK
 	help
 	  SCO, Xenix and Coherent are commercial Unix systems for Intel
 	  machines, and Version 7 was used on the DEC PDP-11. Saying Y
@@ -1381,6 +1395,7 @@ config SYSV_FS
 
 config UFS_FS
 	tristate "UFS file system support (read only)"
+	depends on BLOCK
 	help
 	  BSD and derivate versions of Unix (such as SunOS, FreeBSD, NetBSD,
 	  OpenBSD and NeXTstep) use a file system called UFS. Some System V
@@ -1959,11 +1974,13 @@ config GENERIC_ACL
 
 endmenu
 
+if BLOCK
 menu "Partition Types"
 
 source "fs/partitions/Kconfig"
 
 endmenu
+endif
 
 source "fs/nls/Kconfig"
 
diff --git a/fs/Makefile b/fs/Makefile
index 46b8cfe497b2..a503e6ce0f32 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -5,12 +5,18 @@
 # Rewritten to use lists instead of if-statements.
 # 
 
-obj-y :=	open.o read_write.o file_table.o buffer.o  bio.o super.o \
-		block_dev.o char_dev.o stat.o exec.o pipe.o namei.o fcntl.o \
+obj-y :=	open.o read_write.o file_table.o super.o \
+		char_dev.o stat.o exec.o pipe.o namei.o fcntl.o \
 		ioctl.o readdir.o select.o fifo.o locks.o dcache.o inode.o \
 		attr.o bad_inode.o file.o filesystems.o namespace.o aio.o \
-		seq_file.o xattr.o libfs.o fs-writeback.o mpage.o direct-io.o \
-		ioprio.o pnode.o drop_caches.o splice.o sync.o
+		seq_file.o xattr.o libfs.o fs-writeback.o \
+		pnode.o drop_caches.o splice.o sync.o
+
+ifeq ($(CONFIG_BLOCK),y)
+obj-y +=	buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o
+else
+obj-y +=	no-block.o
+endif
 
 obj-$(CONFIG_INOTIFY)		+= inotify.o
 obj-$(CONFIG_INOTIFY_USER)	+= inotify_user.o
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index e1a56437040a..64b34533edea 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -645,6 +645,7 @@ out:
 }
 #endif
 
+#ifdef CONFIG_BLOCK
 struct hd_geometry32 {
 	unsigned char heads;
 	unsigned char sectors;
@@ -869,6 +870,7 @@ static int sg_grt_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 	}
 	return err;
 }
+#endif /* CONFIG_BLOCK */
 
 struct sock_fprog32 {
 	unsigned short	len;
@@ -992,6 +994,7 @@ static int ppp_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 }
 
 
+#ifdef CONFIG_BLOCK
 struct mtget32 {
 	compat_long_t	mt_type;
 	compat_long_t	mt_resid;
@@ -1164,6 +1167,7 @@ static int cdrom_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long ar
 
 	return err;
 }
+#endif /* CONFIG_BLOCK */
 
 #ifdef CONFIG_VT
 
@@ -1491,6 +1495,7 @@ ret_einval(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return -EINVAL;
 }
 
+#ifdef CONFIG_BLOCK
 static int broken_blkgetsize(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
 	/* The mkswap binary hard codes it to Intel value :-((( */
@@ -1525,12 +1530,14 @@ static int blkpg_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long ar
 
 	return sys_ioctl(fd, cmd, (unsigned long)a);
 }
+#endif
 
 static int ioc_settimeout(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
 	return rw_long(fd, AUTOFS_IOC_SETTIMEOUT, arg);
 }
 
+#ifdef CONFIG_BLOCK
 /* Fix sizeof(sizeof()) breakage */
 #define BLKBSZGET_32   _IOR(0x12,112,int)
 #define BLKBSZSET_32   _IOW(0x12,113,int)
@@ -1551,6 +1558,7 @@ static int do_blkgetsize64(unsigned int fd, unsigned int cmd,
 {
        return sys_ioctl(fd, BLKGETSIZE64, (unsigned long)compat_ptr(arg));
 }
+#endif
 
 /* Bluetooth ioctls */
 #define HCIUARTSETPROTO	_IOW('U', 200, int)
@@ -1571,6 +1579,7 @@ static int do_blkgetsize64(unsigned int fd, unsigned int cmd,
 #define HIDPGETCONNLIST	_IOR('H', 210, int)
 #define HIDPGETCONNINFO	_IOR('H', 211, int)
 
+#ifdef CONFIG_BLOCK
 struct floppy_struct32 {
 	compat_uint_t	size;
 	compat_uint_t	sect;
@@ -1895,6 +1904,7 @@ out:
 	kfree(karg);
 	return err;
 }
+#endif
 
 struct mtd_oob_buf32 {
 	u_int32_t start;
@@ -1936,6 +1946,7 @@ static int mtd_rw_oob(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return err;
 }	
 
+#ifdef CONFIG_BLOCK
 struct raw32_config_request
 {
         compat_int_t    raw_minor;
@@ -2000,6 +2011,7 @@ static int raw_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
         }
         return ret;
 }
+#endif /* CONFIG_BLOCK */
 
 struct serial_struct32 {
         compat_int_t    type;
@@ -2606,6 +2618,7 @@ HANDLE_IOCTL(SIOCBRDELIF, dev_ifsioc)
 HANDLE_IOCTL(SIOCRTMSG, ret_einval)
 HANDLE_IOCTL(SIOCGSTAMP, do_siocgstamp)
 #endif
+#ifdef CONFIG_BLOCK
 HANDLE_IOCTL(HDIO_GETGEO, hdio_getgeo)
 HANDLE_IOCTL(BLKRAGET, w_long)
 HANDLE_IOCTL(BLKGETSIZE, w_long)
@@ -2631,14 +2644,17 @@ HANDLE_IOCTL(FDGETFDCSTAT32, fd_ioctl_trans)
 HANDLE_IOCTL(FDWERRORGET32, fd_ioctl_trans)
 HANDLE_IOCTL(SG_IO,sg_ioctl_trans)
 HANDLE_IOCTL(SG_GET_REQUEST_TABLE, sg_grt_trans)
+#endif
 HANDLE_IOCTL(PPPIOCGIDLE32, ppp_ioctl_trans)
 HANDLE_IOCTL(PPPIOCSCOMPRESS32, ppp_ioctl_trans)
 HANDLE_IOCTL(PPPIOCSPASS32, ppp_sock_fprog_ioctl_trans)
 HANDLE_IOCTL(PPPIOCSACTIVE32, ppp_sock_fprog_ioctl_trans)
+#ifdef CONFIG_BLOCK
 HANDLE_IOCTL(MTIOCGET32, mt_ioctl_trans)
 HANDLE_IOCTL(MTIOCPOS32, mt_ioctl_trans)
 HANDLE_IOCTL(CDROMREADAUDIO, cdrom_ioctl_trans)
 HANDLE_IOCTL(CDROM_SEND_PACKET, cdrom_ioctl_trans)
+#endif
 #define AUTOFS_IOC_SETTIMEOUT32 _IOWR(0x93,0x64,unsigned int)
 HANDLE_IOCTL(AUTOFS_IOC_SETTIMEOUT32, ioc_settimeout)
 #ifdef CONFIG_VT
@@ -2677,12 +2693,14 @@ HANDLE_IOCTL(SONET_SETFRAMING, do_atm_ioctl)
 HANDLE_IOCTL(SONET_GETFRAMING, do_atm_ioctl)
 HANDLE_IOCTL(SONET_GETFRSENSE, do_atm_ioctl)
 /* block stuff */
+#ifdef CONFIG_BLOCK
 HANDLE_IOCTL(BLKBSZGET_32, do_blkbszget)
 HANDLE_IOCTL(BLKBSZSET_32, do_blkbszset)
 HANDLE_IOCTL(BLKGETSIZE64_32, do_blkgetsize64)
 /* Raw devices */
 HANDLE_IOCTL(RAW_SETBIND, raw_ioctl)
 HANDLE_IOCTL(RAW_GETBIND, raw_ioctl)
+#endif
 /* Serial */
 HANDLE_IOCTL(TIOCGSERIAL, serial_struct_ioctl)
 HANDLE_IOCTL(TIOCSSERIAL, serial_struct_ioctl)
diff --git a/fs/internal.h b/fs/internal.h
index f662b703bb97..f07147d63255 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -14,10 +14,16 @@
 /*
  * block_dev.c
  */
+#ifdef CONFIG_BLOCK
 extern struct super_block *blockdev_superblock;
 extern void __init bdev_cache_init(void);
 
 #define sb_is_blkdev_sb(sb) ((sb) == blockdev_superblock)
+#else
+static inline void bdev_cache_init(void) {}
+
+#define sb_is_blkdev_sb(sb) 0
+#endif
 
 /*
  * char_dev.c
diff --git a/fs/no-block.c b/fs/no-block.c
new file mode 100644
index 000000000000..d269a93d3467
--- /dev/null
+++ b/fs/no-block.c
@@ -0,0 +1,22 @@
+/* no-block.c: implementation of routines required for non-BLOCK configuration
+ *
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/fs.h>
+
+static int no_blkdev_open(struct inode * inode, struct file * filp)
+{
+	return -ENODEV;
+}
+
+const struct file_operations def_blk_fops = {
+	.open		= no_blkdev_open,
+};
diff --git a/fs/partitions/Makefile b/fs/partitions/Makefile
index d713ce6b3e12..67e665fdb7fc 100644
--- a/fs/partitions/Makefile
+++ b/fs/partitions/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y := check.o
+obj-$(CONFIG_BLOCK) := check.o
 
 obj-$(CONFIG_ACORN_PARTITION) += acorn.o
 obj-$(CONFIG_AMIGA_PARTITION) += amiga.o
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 5bbd60896050..66bc425f2f3d 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -277,12 +277,15 @@ static int devinfo_show(struct seq_file *f, void *v)
 		if (i == 0)
 			seq_printf(f, "Character devices:\n");
 		chrdev_show(f, i);
-	} else {
+	}
+#ifdef CONFIG_BLOCK
+	else {
 		i -= CHRDEV_MAJOR_HASH_SIZE;
 		if (i == 0)
 			seq_printf(f, "\nBlock devices:\n");
 		blkdev_show(f, i);
 	}
+#endif
 	return 0;
 }
 
@@ -355,6 +358,7 @@ static int stram_read_proc(char *page, char **start, off_t off,
 }
 #endif
 
+#ifdef CONFIG_BLOCK
 extern struct seq_operations partitions_op;
 static int partitions_open(struct inode *inode, struct file *file)
 {
@@ -378,6 +382,7 @@ static struct file_operations proc_diskstats_operations = {
 	.llseek		= seq_lseek,
 	.release	= seq_release,
 };
+#endif
 
 #ifdef CONFIG_MODULES
 extern struct seq_operations modules_op;
@@ -695,7 +700,9 @@ void __init proc_misc_init(void)
 		entry->proc_fops = &proc_kmsg_operations;
 	create_seq_entry("devices", 0, &proc_devinfo_operations);
 	create_seq_entry("cpuinfo", 0, &proc_cpuinfo_operations);
+#ifdef CONFIG_BLOCK
 	create_seq_entry("partitions", 0, &proc_partitions_operations);
+#endif
 	create_seq_entry("stat", 0, &proc_stat_operations);
 	create_seq_entry("interrupts", 0, &proc_interrupts_operations);
 #ifdef CONFIG_SLAB
@@ -707,7 +714,9 @@ void __init proc_misc_init(void)
 	create_seq_entry("buddyinfo",S_IRUGO, &fragmentation_file_operations);
 	create_seq_entry("vmstat",S_IRUGO, &proc_vmstat_file_operations);
 	create_seq_entry("zoneinfo",S_IRUGO, &proc_zoneinfo_file_operations);
+#ifdef CONFIG_BLOCK
 	create_seq_entry("diskstats", 0, &proc_diskstats_operations);
+#endif
 #ifdef CONFIG_MODULES
 	create_seq_entry("modules", 0, &proc_modules_operations);
 #endif
diff --git a/fs/quota.c b/fs/quota.c
index d6a2be826e29..b9dae76a0b6e 100644
--- a/fs/quota.c
+++ b/fs/quota.c
@@ -337,6 +337,34 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, void
 	return 0;
 }
 
+/*
+ * look up a superblock on which quota ops will be performed
+ * - use the name of a block device to find the superblock thereon
+ */
+static inline struct super_block *quotactl_block(const char __user *special)
+{
+#ifdef CONFIG_BLOCK
+	struct block_device *bdev;
+	struct super_block *sb;
+	char *tmp = getname(special);
+
+	if (IS_ERR(tmp))
+		return ERR_PTR(PTR_ERR(tmp));
+	bdev = lookup_bdev(tmp);
+	putname(tmp);
+	if (IS_ERR(bdev))
+		return ERR_PTR(PTR_ERR(bdev));
+	sb = get_super(bdev);
+	bdput(bdev);
+	if (!sb)
+		return ERR_PTR(-ENODEV);
+
+	return sb;
+#else
+	return ERR_PTR(-ENODEV);
+#endif
+}
+
 /*
  * This is the system call interface. This communicates with
  * the user-level programs. Currently this only supports diskquota
@@ -347,25 +375,15 @@ asmlinkage long sys_quotactl(unsigned int cmd, const char __user *special, qid_t
 {
 	uint cmds, type;
 	struct super_block *sb = NULL;
-	struct block_device *bdev;
-	char *tmp;
 	int ret;
 
 	cmds = cmd >> SUBCMDSHIFT;
 	type = cmd & SUBCMDMASK;
 
 	if (cmds != Q_SYNC || special) {
-		tmp = getname(special);
-		if (IS_ERR(tmp))
-			return PTR_ERR(tmp);
-		bdev = lookup_bdev(tmp);
-		putname(tmp);
-		if (IS_ERR(bdev))
-			return PTR_ERR(bdev);
-		sb = get_super(bdev);
-		bdput(bdev);
-		if (!sb)
-			return -ENODEV;
+		sb = quotactl_block(special);
+		if (IS_ERR(sb))
+			return PTR_ERR(sb);
 	}
 
 	ret = check_quotactl_valid(sb, type, cmds, id);
diff --git a/fs/super.c b/fs/super.c
index 15671cd048b1..aec99ddbe53f 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -571,8 +571,10 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 {
 	int retval;
 	
+#ifdef CONFIG_BLOCK
 	if (!(flags & MS_RDONLY) && bdev_read_only(sb->s_bdev))
 		return -EACCES;
+#endif
 	if (flags & MS_RDONLY)
 		acct_auto_close(sb);
 	shrink_dcache_sb(sb);
@@ -692,6 +694,7 @@ void kill_litter_super(struct super_block *sb)
 
 EXPORT_SYMBOL(kill_litter_super);
 
+#ifdef CONFIG_BLOCK
 static int set_bdev_super(struct super_block *s, void *data)
 {
 	s->s_bdev = data;
@@ -787,6 +790,7 @@ void kill_block_super(struct super_block *sb)
 }
 
 EXPORT_SYMBOL(kill_block_super);
+#endif
 
 int get_sb_nodev(struct file_system_type *fs_type,
 	int flags, void *data,
diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig
index 26b364c9d62c..35115bca036e 100644
--- a/fs/xfs/Kconfig
+++ b/fs/xfs/Kconfig
@@ -1,5 +1,6 @@
 config XFS_FS
 	tristate "XFS filesystem support"
+	depends on BLOCK
 	help
 	  XFS is a high performance journaling filesystem which originated
 	  on the SGI IRIX platform.  It is completely multi-threaded, can
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2c01a90998a7..3e36107d342a 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -16,6 +16,22 @@
 
 #include <asm/scatterlist.h>
 
+#ifdef CONFIG_LBD
+# include <asm/div64.h>
+# define sector_div(a, b) do_div(a, b)
+#else
+# define sector_div(n, b)( \
+{ \
+	int _res; \
+	_res = (n) % (b); \
+	(n) /= (b); \
+	_res; \
+} \
+)
+#endif
+
+#ifdef CONFIG_BLOCK
+
 struct scsi_ioctl_command;
 
 struct request_queue;
@@ -818,24 +834,30 @@ struct work_struct;
 int kblockd_schedule_work(struct work_struct *work);
 void kblockd_flush(void);
 
-#ifdef CONFIG_LBD
-# include <asm/div64.h>
-# define sector_div(a, b) do_div(a, b)
-#else
-# define sector_div(n, b)( \
-{ \
-	int _res; \
-	_res = (n) % (b); \
-	(n) /= (b); \
-	_res; \
-} \
-)
-#endif 
-
 #define MODULE_ALIAS_BLOCKDEV(major,minor) \
 	MODULE_ALIAS("block-major-" __stringify(major) "-" __stringify(minor))
 #define MODULE_ALIAS_BLOCKDEV_MAJOR(major) \
 	MODULE_ALIAS("block-major-" __stringify(major) "-*")
 
 
+#else /* CONFIG_BLOCK */
+/*
+ * stubs for when the block layer is configured out
+ */
+#define buffer_heads_over_limit 0
+
+static inline long blk_congestion_wait(int rw, long timeout)
+{
+	return timeout;
+}
+
+static inline long nr_blockdev_pages(void)
+{
+	return 0;
+}
+
+static inline void exit_io_context(void) {}
+
+#endif /* CONFIG_BLOCK */
+
 #endif
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index 64b508e35d2a..131ffd37e716 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -14,6 +14,8 @@
 #include <linux/wait.h>
 #include <asm/atomic.h>
 
+#ifdef CONFIG_BLOCK
+
 enum bh_state_bits {
 	BH_Uptodate,	/* Contains valid data */
 	BH_Dirty,	/* Is dirty */
@@ -301,4 +303,18 @@ static inline void lock_buffer(struct buffer_head *bh)
 }
 
 extern int __set_page_dirty_buffers(struct page *page);
+
+#else /* CONFIG_BLOCK */
+
+static inline void buffer_init(void) {}
+static inline int try_to_free_buffers(struct page *page) { return 1; }
+static inline int sync_blockdev(struct block_device *bdev) { return 0; }
+static inline int inode_has_buffers(struct inode *inode) { return 0; }
+static inline void invalidate_inode_buffers(struct inode *inode) {}
+static inline int remove_inode_buffers(struct inode *inode) { return 1; }
+static inline int sync_mapping_buffers(struct address_space *mapping) { return 0; }
+static inline void invalidate_bdev(struct block_device *bdev, int destroy_dirty_buffers) {}
+
+
+#endif /* CONFIG_BLOCK */
 #endif /* _LINUX_BUFFER_HEAD_H */
diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h
index 98d40e08ba6e..d61ef5951538 100644
--- a/include/linux/compat_ioctl.h
+++ b/include/linux/compat_ioctl.h
@@ -90,6 +90,7 @@ COMPATIBLE_IOCTL(FDTWADDLE)
 COMPATIBLE_IOCTL(FDFMTTRK)
 COMPATIBLE_IOCTL(FDRAWCMD)
 /* 0x12 */
+#ifdef CONFIG_BLOCK
 COMPATIBLE_IOCTL(BLKRASET)
 COMPATIBLE_IOCTL(BLKROSET)
 COMPATIBLE_IOCTL(BLKROGET)
@@ -103,6 +104,7 @@ COMPATIBLE_IOCTL(BLKTRACESETUP)
 COMPATIBLE_IOCTL(BLKTRACETEARDOWN)
 ULONG_IOCTL(BLKRASET)
 ULONG_IOCTL(BLKFRASET)
+#endif
 /* RAID */
 COMPATIBLE_IOCTL(RAID_VERSION)
 COMPATIBLE_IOCTL(GET_ARRAY_INFO)
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 9c5a04f6114c..b3370ef5164d 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -3,6 +3,8 @@
 
 #include <linux/percpu.h>
 
+#ifdef CONFIG_BLOCK
+
 typedef int (elevator_merge_fn) (request_queue_t *, struct request **,
 				 struct bio *);
 
@@ -203,4 +205,5 @@ enum {
 	__val;							\
 })
 
+#endif /* CONFIG_BLOCK */
 #endif
diff --git a/include/linux/fs.h b/include/linux/fs.h
index b73a47582dbe..5baf3a153403 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1482,6 +1482,7 @@ extern void __init vfs_caches_init(unsigned long);
 extern void putname(const char *name);
 #endif
 
+#ifdef CONFIG_BLOCK
 extern int register_blkdev(unsigned int, const char *);
 extern int unregister_blkdev(unsigned int, const char *);
 extern struct block_device *bdget(dev_t);
@@ -1490,11 +1491,15 @@ extern void bd_forget(struct inode *inode);
 extern void bdput(struct block_device *);
 extern struct block_device *open_by_devnum(dev_t, unsigned);
 extern struct block_device *open_partition_by_devnum(dev_t, unsigned);
-extern const struct file_operations def_blk_fops;
 extern const struct address_space_operations def_blk_aops;
+#else
+static inline void bd_forget(struct inode *inode) {}
+#endif
+extern const struct file_operations def_blk_fops;
 extern const struct file_operations def_chr_fops;
 extern const struct file_operations bad_sock_fops;
 extern const struct file_operations def_fifo_fops;
+#ifdef CONFIG_BLOCK
 extern int ioctl_by_bdev(struct block_device *, unsigned, unsigned long);
 extern int blkdev_ioctl(struct inode *, struct file *, unsigned, unsigned long);
 extern long compat_blkdev_ioctl(struct file *, unsigned, unsigned long);
@@ -1510,6 +1515,7 @@ extern void bd_release_from_disk(struct block_device *, struct gendisk *);
 #define bd_claim_by_disk(bdev, holder, disk)	bd_claim(bdev, holder)
 #define bd_release_from_disk(bdev, disk)	bd_release(bdev)
 #endif
+#endif
 
 /* fs/char_dev.c */
 #define CHRDEV_MAJOR_HASH_SIZE	255
@@ -1523,14 +1529,19 @@ extern int chrdev_open(struct inode *, struct file *);
 extern void chrdev_show(struct seq_file *,off_t);
 
 /* fs/block_dev.c */
-#define BLKDEV_MAJOR_HASH_SIZE	255
 #define BDEVNAME_SIZE	32	/* Largest string for a blockdev identifier */
+
+#ifdef CONFIG_BLOCK
+#define BLKDEV_MAJOR_HASH_SIZE	255
 extern const char *__bdevname(dev_t, char *buffer);
 extern const char *bdevname(struct block_device *bdev, char *buffer);
 extern struct block_device *lookup_bdev(const char *);
 extern struct block_device *open_bdev_excl(const char *, int, void *);
 extern void close_bdev_excl(struct block_device *);
 extern void blkdev_show(struct seq_file *,off_t);
+#else
+#define BLKDEV_MAJOR_HASH_SIZE	0
+#endif
 
 extern void init_special_inode(struct inode *, umode_t, dev_t);
 
@@ -1544,6 +1555,7 @@ extern const struct file_operations rdwr_fifo_fops;
 
 extern int fs_may_remount_ro(struct super_block *);
 
+#ifdef CONFIG_BLOCK
 /*
  * return READ, READA, or WRITE
  */
@@ -1555,9 +1567,10 @@ extern int fs_may_remount_ro(struct super_block *);
 #define bio_data_dir(bio)	((bio)->bi_rw & 1)
 
 extern int check_disk_change(struct block_device *);
-extern int invalidate_inodes(struct super_block *);
 extern int __invalidate_device(struct block_device *);
 extern int invalidate_partition(struct gendisk *, int);
+#endif
+extern int invalidate_inodes(struct super_block *);
 unsigned long invalidate_mapping_pages(struct address_space *mapping,
 					pgoff_t start, pgoff_t end);
 unsigned long invalidate_inode_pages(struct address_space *mapping);
@@ -1590,7 +1603,9 @@ extern void emergency_sync(void);
 extern void emergency_remount(void);
 extern int do_remount_sb(struct super_block *sb, int flags,
 			 void *data, int force);
+#ifdef CONFIG_BLOCK
 extern sector_t bmap(struct inode *, sector_t);
+#endif
 extern int notify_change(struct dentry *, struct iattr *);
 extern int permission(struct inode *, int, struct nameidata *);
 extern int generic_permission(struct inode *, int,
@@ -1673,9 +1688,11 @@ static inline void insert_inode_hash(struct inode *inode) {
 extern struct file * get_empty_filp(void);
 extern void file_move(struct file *f, struct list_head *list);
 extern void file_kill(struct file *f);
+#ifdef CONFIG_BLOCK
 struct bio;
 extern void submit_bio(int, struct bio *);
 extern int bdev_read_only(struct block_device *);
+#endif
 extern int set_blocksize(struct block_device *, int);
 extern int sb_set_blocksize(struct super_block *, int);
 extern int sb_min_blocksize(struct super_block *, int);
@@ -1756,6 +1773,7 @@ static inline void do_generic_file_read(struct file * filp, loff_t *ppos,
 				actor);
 }
 
+#ifdef CONFIG_BLOCK
 ssize_t __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	struct block_device *bdev, const struct iovec *iov, loff_t offset,
 	unsigned long nr_segs, get_block_t get_block, dio_iodone_t end_io,
@@ -1793,6 +1811,7 @@ static inline ssize_t blockdev_direct_IO_own_locking(int rw, struct kiocb *iocb,
 	return __blockdev_direct_IO(rw, iocb, inode, bdev, iov, offset,
 				nr_segs, get_block, end_io, DIO_OWN_LOCKING);
 }
+#endif
 
 extern const struct file_operations generic_ro_fops;
 
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index e4af57e87c17..41f276fdd185 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -11,6 +11,8 @@
 
 #include <linux/types.h>
 
+#ifdef CONFIG_BLOCK
+
 enum {
 /* These three have identical behaviour; use the second one if DOS FDISK gets
    confused about extended/logical partitions starting past cylinder 1023. */
@@ -420,3 +422,5 @@ static inline struct block_device *bdget_disk(struct gendisk *disk, int index)
 #endif
 
 #endif
+
+#endif
diff --git a/include/linux/mpage.h b/include/linux/mpage.h
index 517c098fde20..cc5fb75af78a 100644
--- a/include/linux/mpage.h
+++ b/include/linux/mpage.h
@@ -9,6 +9,7 @@
  * (And no, it doesn't do the #ifdef __MPAGE_H thing, and it doesn't do
  * nested includes.  Get it right in the .c file).
  */
+#ifdef CONFIG_BLOCK
 
 struct writeback_control;
 typedef int (writepage_t)(struct page *page, struct writeback_control *wbc);
@@ -20,3 +21,5 @@ int mpage_writepages(struct address_space *mapping,
 		struct writeback_control *wbc, get_block_t get_block);
 int mpage_writepage(struct page *page, get_block_t *get_block,
 		struct writeback_control *wbc);
+
+#endif
diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h
index eb3e547c8fee..c588709acbbc 100644
--- a/include/linux/raid/md.h
+++ b/include/linux/raid/md.h
@@ -53,6 +53,8 @@
 #include <linux/raid/md_u.h>
 #include <linux/raid/md_k.h>
 
+#ifdef CONFIG_MD
+
 /*
  * Different major versions are not compatible.
  * Different minor versions are only downward compatible.
@@ -95,5 +97,6 @@ extern void md_new_event(mddev_t *mddev);
 
 extern void md_update_sb(mddev_t * mddev);
 
+#endif /* CONFIG_MD */
 #endif 
 
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index d28890295852..920b94fe31fa 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -18,6 +18,8 @@
 /* and dm-bio-list.h is not under include/linux because.... ??? */
 #include "../../../drivers/md/dm-bio-list.h"
 
+#ifdef CONFIG_BLOCK
+
 #define	LEVEL_MULTIPATH		(-4)
 #define	LEVEL_LINEAR		(-1)
 #define	LEVEL_FAULTY		(-5)
@@ -362,5 +364,6 @@ static inline void safe_put_page(struct page *p)
 	if (p) put_page(p);
 }
 
+#endif /* CONFIG_BLOCK */
 #endif
 
diff --git a/include/scsi/scsi_tcq.h b/include/scsi/scsi_tcq.h
index bbf66219b769..c247a28259bc 100644
--- a/include/scsi/scsi_tcq.h
+++ b/include/scsi/scsi_tcq.h
@@ -6,7 +6,6 @@
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_host.h>
 
-
 #define MSG_SIMPLE_TAG	0x20
 #define MSG_HEAD_TAG	0x21
 #define MSG_ORDERED_TAG	0x22
@@ -14,6 +13,7 @@
 #define SCSI_NO_TAG	(-1)    /* identify no tag in use */
 
 
+#ifdef CONFIG_BLOCK
 
 /**
  * scsi_get_tag_type - get the type of tag the device supports
@@ -144,4 +144,5 @@ static inline int scsi_init_shared_tag_map(struct Scsi_Host *shost, int depth)
 	return shost->bqt ? 0 : -ENOMEM;
 }
 
+#endif /* CONFIG_BLOCK */
 #endif /* _SCSI_SCSI_TCQ_H */
diff --git a/init/Kconfig b/init/Kconfig
index 4381006dd666..d2eb7a84a264 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -92,7 +92,7 @@ config LOCALVERSION_AUTO
 
 config SWAP
 	bool "Support for paging of anonymous memory (swap)"
-	depends on MMU
+	depends on MMU && BLOCK
 	default y
 	help
 	  This option allows you to choose whether you want to have support
diff --git a/init/do_mounts.c b/init/do_mounts.c
index b290aadb1d3f..dc1ec0803ef9 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -285,7 +285,11 @@ void __init mount_block_root(char *name, int flags)
 {
 	char *fs_names = __getname();
 	char *p;
+#ifdef CONFIG_BLOCK
 	char b[BDEVNAME_SIZE];
+#else
+	const char *b = name;
+#endif
 
 	get_fs_names(fs_names);
 retry:
@@ -304,7 +308,9 @@ retry:
 		 * Allow the user to distinguish between failed sys_open
 		 * and bad superblock on root device.
 		 */
+#ifdef CONFIG_BLOCK
 		__bdevname(ROOT_DEV, b);
+#endif
 		printk("VFS: Cannot open root device \"%s\" or %s\n",
 				root_device_name, b);
 		printk("Please append a correct \"root=\" boot option\n");
@@ -316,7 +322,10 @@ retry:
 	for (p = fs_names; *p; p += strlen(p)+1)
 		printk(" %s", p);
 	printk("\n");
-	panic("VFS: Unable to mount root fs on %s", __bdevname(ROOT_DEV, b));
+#ifdef CONFIG_BLOCK
+	__bdevname(ROOT_DEV, b);
+#endif
+	panic("VFS: Unable to mount root fs on %s", b);
 out:
 	putname(fs_names);
 }
@@ -387,8 +396,10 @@ void __init mount_root(void)
 			change_floppy("root floppy");
 	}
 #endif
+#ifdef CONFIG_BLOCK
 	create_dev("/dev/root", ROOT_DEV);
 	mount_block_root("/dev/root", root_mountflags);
+#endif
 }
 
 /*
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 6991bece67e8..7a3b2e75f040 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -134,3 +134,8 @@ cond_syscall(sys_madvise);
 cond_syscall(sys_mremap);
 cond_syscall(sys_remap_file_pages);
 cond_syscall(compat_sys_move_pages);
+
+/* block-layer dependent */
+cond_syscall(sys_bdflush);
+cond_syscall(sys_ioprio_set);
+cond_syscall(sys_ioprio_get);
diff --git a/mm/Makefile b/mm/Makefile
index 4f2166a833b9..12b3a4eee88d 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -12,7 +12,7 @@ obj-y			:= bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
 			   readahead.o swap.o truncate.o vmscan.o \
 			   prio_tree.o util.o mmzone.o vmstat.o $(mmu-y)
 
-ifeq ($(CONFIG_MMU),y)
+ifeq ($(CONFIG_MMU)$(CONFIG_BLOCK),yy)
 obj-y			+= bounce.o
 endif
 obj-$(CONFIG_SWAP)	+= page_io.o swap_state.o swapfile.o thrash.o
diff --git a/mm/filemap.c b/mm/filemap.c
index d6846de08887..c4fe97f5ace0 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2020,6 +2020,7 @@ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, i
 		if (unlikely(*pos + *count > inode->i_sb->s_maxbytes))
 			*count = inode->i_sb->s_maxbytes - *pos;
 	} else {
+#ifdef CONFIG_BLOCK
 		loff_t isize;
 		if (bdev_read_only(I_BDEV(inode)))
 			return -EPERM;
@@ -2031,6 +2032,9 @@ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, i
 
 		if (*pos + *count > isize)
 			*count = isize - *pos;
+#else
+		return -EPERM;
+#endif
 	}
 	return 0;
 }
diff --git a/mm/migrate.c b/mm/migrate.c
index 7f50e3ff54cd..ba2453f9483d 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -409,6 +409,7 @@ int migrate_page(struct address_space *mapping,
 }
 EXPORT_SYMBOL(migrate_page);
 
+#ifdef CONFIG_BLOCK
 /*
  * Migration function for pages with buffers. This function can only be used
  * if the underlying filesystem guarantees that no other references to "page"
@@ -466,6 +467,7 @@ int buffer_migrate_page(struct address_space *mapping,
 	return 0;
 }
 EXPORT_SYMBOL(buffer_migrate_page);
+#endif
 
 /*
  * Writeback a page to clean the dirty state
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index ecf27839c203..c0d4ce144dec 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -807,9 +807,11 @@ int fastcall set_page_dirty(struct page *page)
 
 	if (likely(mapping)) {
 		int (*spd)(struct page *) = mapping->a_ops->set_page_dirty;
-		if (spd)
-			return (*spd)(page);
-		return __set_page_dirty_buffers(page);
+#ifdef CONFIG_BLOCK
+		if (!spd)
+			spd = __set_page_dirty_buffers;
+#endif
+		return (*spd)(page);
 	}
 	if (!PageDirty(page)) {
 		if (!TestSetPageDirty(page))
diff --git a/mm/truncate.c b/mm/truncate.c
index cd3e34b816db..8fde6580657e 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -35,8 +35,10 @@ void do_invalidatepage(struct page *page, unsigned long offset)
 {
 	void (*invalidatepage)(struct page *, unsigned long);
 	invalidatepage = page->mapping->a_ops->invalidatepage;
+#ifdef CONFIG_BLOCK
 	if (!invalidatepage)
 		invalidatepage = block_invalidatepage;
+#endif
 	if (invalidatepage)
 		(*invalidatepage)(page, offset);
 }
-- 
cgit v1.2.3


From bcfd8d36151e531e1c6c731f1fbf792509a1c494 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Thu, 31 Aug 2006 12:56:06 +0200
Subject: [PATCH] CONFIG_BLOCK: blk_congestion_wait() fix

Don't just do nothing: it'll cause busywaits all over writeback and page
reclaim.

For now, take a fixed-length nap.  Will improve when NFS starts waking up
throttled processes.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/linux/blkdev.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 3e36107d342a..1d79b8d4ca6d 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_BLKDEV_H
 #define _LINUX_BLKDEV_H
 
+#include <linux/sched.h>
 #include <linux/major.h>
 #include <linux/genhd.h>
 #include <linux/list.h>
@@ -848,7 +849,7 @@ void kblockd_flush(void);
 
 static inline long blk_congestion_wait(int rw, long timeout)
 {
-	return timeout;
+	return io_schedule_timeout(timeout);
 }
 
 static inline long nr_blockdev_pages(void)
@@ -856,7 +857,9 @@ static inline long nr_blockdev_pages(void)
 	return 0;
 }
 
-static inline void exit_io_context(void) {}
+static inline void exit_io_context(void)
+{
+}
 
 #endif /* CONFIG_BLOCK */
 
-- 
cgit v1.2.3


From cb5d9e0948122dd8b808550574d95bd60674ba3b Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Sat, 30 Sep 2006 21:23:44 -0400
Subject: [PATCH] scsi: device_reprobe() can fail

device_reprobe() should return an error code.  When it does so,
scsi_device_reprobe() should propagate it back.

Acked-by: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/scsi/scsi_device.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index 895d212864cd..b401c82036be 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -298,9 +298,9 @@ extern int scsi_execute_async(struct scsi_device *sdev,
 			      void (*done)(void *, char *, int, int),
 			      gfp_t gfp);
 
-static inline void scsi_device_reprobe(struct scsi_device *sdev)
+static inline int __must_check scsi_device_reprobe(struct scsi_device *sdev)
 {
-	device_reprobe(&sdev->sdev_gendev);
+	return device_reprobe(&sdev->sdev_gendev);
 }
 
 static inline unsigned int sdev_channel(struct scsi_device *sdev)
-- 
cgit v1.2.3


From 236561e5df009f79f1939e3ca269b9b6f18092f5 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Sat, 30 Sep 2006 23:27:03 -0700
Subject: [PATCH] PCI quirks update

This fixes two things

Firstly someone mistakenly used "errata" for the singular.  This causes
Dave Woodhouse to emit diagnostics whenever the string is read, and so
should be fixed.

Secondly the AMD AGP tunnel has an erratum which causes hangs if you try
and do direct PCI to AGP transfers in some cases.  We have a flag for
PCI/PCI failures but we need a different flag for this really as in this
case we don't want to stop PCI/PCI transfers using things like IOAT and the
new RAID offload work.

I'll post some updates to make proper use of the PCIAGP flag in the
media/video drivers to Mauro.

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pci/quirks.c | 17 +++++++++++++++--
 include/linux/pci.h  |  5 +++--
 2 files changed, 18 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 08cd86a6dd66..23b599d6a9d5 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -93,8 +93,21 @@ static void __devinit quirk_nopcipci(struct pci_dev *dev)
 		pci_pci_problems |= PCIPCI_FAIL;
 	}
 }
+
+static void __devinit quirk_nopciamd(struct pci_dev *dev)
+{
+	u8 rev;
+	pci_read_config_byte(dev, 0x08, &rev);
+	if (rev == 0x13) {
+		/* Erratum 24 */
+		printk(KERN_INFO "Chipset erratum: Disabling direct PCI/AGP transfers.\n");
+		pci_pci_problems |= PCIAGP_FAIL;
+	}
+}
+
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_5597,		quirk_nopcipci );
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_496,		quirk_nopcipci );
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_8151_0,	quirk_nopciamd );
 
 /*
  *	Triton requires workarounds to be used by the drivers
@@ -555,7 +568,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237,		quirk_via_vt
  * is currently marked NoFix
  *
  * We have multiple reports of hangs with this chipset that went away with
- * noapic specified. For the moment we assume its the errata. We may be wrong
+ * noapic specified. For the moment we assume it's the erratum. We may be wrong
  * of course. However the advice is demonstrably good even if so..
  */
 static void __devinit quirk_amd_ioapic(struct pci_dev *dev)
@@ -564,7 +577,7 @@ static void __devinit quirk_amd_ioapic(struct pci_dev *dev)
 
 	pci_read_config_byte(dev, PCI_REVISION_ID, &rev);
 	if (rev >= 0x02) {
-		printk(KERN_WARNING "I/O APIC: AMD Errata #22 may be present. In the event of instability try\n");
+		printk(KERN_WARNING "I/O APIC: AMD Erratum #22 may be present. In the event of instability try\n");
 		printk(KERN_WARNING "        : booting with the \"noapic\" option.\n");
 	}
 }
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 5c3a4176eb64..4431ce4e1e6f 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -787,12 +787,13 @@ enum pci_fixup_pass {
 void pci_fixup_device(enum pci_fixup_pass pass, struct pci_dev *dev);
 
 extern int pci_pci_problems;
-#define PCIPCI_FAIL		1
+#define PCIPCI_FAIL		1	/* No PCI PCI DMA */
 #define PCIPCI_TRITON		2
 #define PCIPCI_NATOMA		4
 #define PCIPCI_VIAETBF		8
 #define PCIPCI_VSFX		16
-#define PCIPCI_ALIMAGIK		32
+#define PCIPCI_ALIMAGIK		32	/* Need low latency setting */
+#define PCIAGP_FAIL		64	/* No PCI to AGP DMA */
 
 #endif /* __KERNEL__ */
 #endif /* LINUX_PCI_H */
-- 
cgit v1.2.3


From f28c5edc06ecd8068b38b7662ad19f4d20d741af Mon Sep 17 00:00:00 2001
From: Keith Mannthey <kmannth@us.ibm.com>
Date: Sat, 30 Sep 2006 23:27:04 -0700
Subject: [PATCH] hot-add-mem x86_64: fixup externs

Fix up externs in memory_hotplug.c.  Cleanup.

Signed-off-by: Keith Mannthey <kmannth@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/memory_hotplug.h | 2 ++
 include/linux/mm.h             | 2 ++
 mm/memory_hotplug.c            | 4 ----
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index 218501cfaeb9..7b54666cea8e 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -172,5 +172,7 @@ static inline int __remove_pages(struct zone *zone, unsigned long start_pfn,
 extern int add_memory(int nid, u64 start, u64 size);
 extern int arch_add_memory(int nid, u64 start, u64 size);
 extern int remove_memory(u64 start, u64 size);
+extern int sparse_add_one_section(struct zone *zone, unsigned long start_pfn,
+								int nr_pages);
 
 #endif /* __LINUX_MEMORY_HOTPLUG_H */
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 4edf1934e5ca..b7966ab8cb6a 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -946,6 +946,8 @@ extern void mem_init(void);
 extern void show_mem(void);
 extern void si_meminfo(struct sysinfo * val);
 extern void si_meminfo_node(struct sysinfo *val, int nid);
+extern void zonetable_add(struct zone *zone, int nid, enum zone_type zid,
+					unsigned long pfn, unsigned long size);
 
 #ifdef CONFIG_NUMA
 extern void setup_per_cpu_pageset(void);
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 2053bb165a21..63b14d4f68fb 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -26,8 +26,6 @@
 
 #include <asm/tlbflush.h>
 
-extern void zonetable_add(struct zone *zone, int nid, int zid, unsigned long pfn,
-			  unsigned long size);
 static int __add_zone(struct zone *zone, unsigned long phys_start_pfn)
 {
 	struct pglist_data *pgdat = zone->zone_pgdat;
@@ -47,8 +45,6 @@ static int __add_zone(struct zone *zone, unsigned long phys_start_pfn)
 	return 0;
 }
 
-extern int sparse_add_one_section(struct zone *zone, unsigned long start_pfn,
-				  int nr_pages);
 static int __add_section(struct zone *zone, unsigned long phys_start_pfn)
 {
 	int nr_pages = PAGES_PER_SECTION;
-- 
cgit v1.2.3


From 53947027ad90542ddb2bb746e3175827c270610a Mon Sep 17 00:00:00 2001
From: Keith Mannthey <kmannth@us.ibm.com>
Date: Sat, 30 Sep 2006 23:27:08 -0700
Subject: [PATCH] hot-add-mem x86_64: use CONFIG_MEMORY_HOTPLUG_SPARSE

Migate CONFIG_MEMORY_HOTPLUG to CONFIG_MEMORY_HOTPLUG_SPARSE where needed.

Signed-off-by: Keith Mannthey <kmannth@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/base/Makefile  | 2 +-
 include/linux/memory.h | 4 ++--
 mm/memory_hotplug.c    | 2 ++
 3 files changed, 5 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/drivers/base/Makefile b/drivers/base/Makefile
index b539e5e75b56..7bbb9eeda235 100644
--- a/drivers/base/Makefile
+++ b/drivers/base/Makefile
@@ -8,7 +8,7 @@ obj-y			+= power/
 obj-$(CONFIG_ISA)	+= isa.o
 obj-$(CONFIG_FW_LOADER)	+= firmware_class.o
 obj-$(CONFIG_NUMA)	+= node.o
-obj-$(CONFIG_MEMORY_HOTPLUG) += memory.o
+obj-$(CONFIG_MEMORY_HOTPLUG_SPARSE) += memory.o
 obj-$(CONFIG_SMP)	+= topology.o
 obj-$(CONFIG_SYS_HYPERVISOR) += hypervisor.o
 
diff --git a/include/linux/memory.h b/include/linux/memory.h
index 8f04143ca363..654ef5544878 100644
--- a/include/linux/memory.h
+++ b/include/linux/memory.h
@@ -57,7 +57,7 @@ struct memory_block {
 struct notifier_block;
 struct mem_section;
 
-#ifndef CONFIG_MEMORY_HOTPLUG
+#ifndef CONFIG_MEMORY_HOTPLUG_SPARSE
 static inline int memory_dev_init(void)
 {
 	return 0;
@@ -78,7 +78,7 @@ extern int remove_memory_block(unsigned long, struct mem_section *, int);
 #define CONFIG_MEM_BLOCK_SIZE	(PAGES_PER_SECTION<<PAGE_SHIFT)
 
 
-#endif /* CONFIG_MEMORY_HOTPLUG */
+#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */
 
 #define hotplug_memory_notifier(fn, pri) {			\
 	static struct notifier_block fn##_mem_nb =		\
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 63b14d4f68fb..7666dbd328df 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -26,6 +26,7 @@
 
 #include <asm/tlbflush.h>
 
+#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
 static int __add_zone(struct zone *zone, unsigned long phys_start_pfn)
 {
 	struct pglist_data *pgdat = zone->zone_pgdat;
@@ -192,6 +193,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages)
 	writeback_set_ratelimit();
 	return 0;
 }
+#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */
 
 static pg_data_t *hotadd_new_pgdat(int nid, u64 start)
 {
-- 
cgit v1.2.3


From 6e21828743247270d09a86756a0c11702500dbfb Mon Sep 17 00:00:00 2001
From: Richard Knutsson <ricknu-0@student.ltu.se>
Date: Sat, 30 Sep 2006 23:27:11 -0700
Subject: [PATCH] Generic boolean

This patch defines:
* a generic boolean-type, named 'bool'
* aliases to 0 and 1, named 'false' and 'true'

Removing colliding definitions of 'bool', 'false' and 'true'.

Signed-off-by: Richard Knutsson <ricknu-0@student.ltu.se>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/DAC960.h            | 2 +-
 drivers/media/video/cpia2/cpia2.h | 4 ----
 drivers/net/dgrs.c                | 1 -
 drivers/scsi/BusLogic.h           | 5 +----
 include/linux/stddef.h            | 6 ++++++
 include/linux/types.h             | 2 ++
 6 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include')

diff --git a/drivers/block/DAC960.h b/drivers/block/DAC960.h
index a82f37f749a5..f9217c34bc2b 100644
--- a/drivers/block/DAC960.h
+++ b/drivers/block/DAC960.h
@@ -71,7 +71,7 @@
   Define a Boolean data type.
 */
 
-typedef enum { false, true } __attribute__ ((packed)) boolean;
+typedef bool boolean;
 
 
 /*
diff --git a/drivers/media/video/cpia2/cpia2.h b/drivers/media/video/cpia2/cpia2.h
index c5ecb2be5f93..8d2dfc128821 100644
--- a/drivers/media/video/cpia2/cpia2.h
+++ b/drivers/media/video/cpia2/cpia2.h
@@ -50,10 +50,6 @@
 /***
  * Image defines
  ***/
-#ifndef true
-#define true 1
-#define false 0
-#endif
 
 /*  Misc constants */
 #define ALLOW_CORRUPT 0		/* Causes collater to discard checksum */
diff --git a/drivers/net/dgrs.c b/drivers/net/dgrs.c
index e9361cb5f4cd..d0842527b369 100644
--- a/drivers/net/dgrs.c
+++ b/drivers/net/dgrs.c
@@ -110,7 +110,6 @@ static char version[] __initdata =
  *	DGRS include files
  */
 typedef unsigned char uchar;
-typedef unsigned int bool;
 #define vol volatile
 
 #include "dgrs.h"
diff --git a/drivers/scsi/BusLogic.h b/drivers/scsi/BusLogic.h
index 9792e5af5252..d6d1d5613c8a 100644
--- a/drivers/scsi/BusLogic.h
+++ b/drivers/scsi/BusLogic.h
@@ -237,10 +237,7 @@ enum BusLogic_BIOS_DiskGeometryTranslation {
   Define a Boolean data type.
 */
 
-typedef enum {
-	false,
-	true
-} PACKED boolean;
+typedef bool boolean;
 
 /*
   Define a 10^18 Statistics Byte Counter data type.
diff --git a/include/linux/stddef.h b/include/linux/stddef.h
index ea65dfb60cd8..6a40c76bdcf1 100644
--- a/include/linux/stddef.h
+++ b/include/linux/stddef.h
@@ -11,6 +11,12 @@
 #endif
 
 #ifdef __KERNEL__
+
+enum {
+	false	= 0,
+	true	= 1
+};
+
 #undef offsetof
 #ifdef __compiler_offsetof
 #define offsetof(TYPE,MEMBER) __compiler_offsetof(TYPE,MEMBER)
diff --git a/include/linux/types.h b/include/linux/types.h
index 3f235660a3cd..406d4ae57631 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -33,6 +33,8 @@ typedef __kernel_clockid_t	clockid_t;
 typedef __kernel_mqd_t		mqd_t;
 
 #ifdef __KERNEL__
+typedef _Bool			bool;
+
 typedef __kernel_uid32_t	uid_t;
 typedef __kernel_gid32_t	gid_t;
 typedef __kernel_uid16_t        uid16_t;
-- 
cgit v1.2.3


From 5c87579e65ee4f419b2369407f82326d38b5d2d8 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Sat, 30 Sep 2006 23:27:17 -0700
Subject: [PATCH] maximum latency tracking infrastructure

Add infrastructure to track "maximum allowable latency" for power saving
policies.

The reason for adding this infrastructure is that power management in the
idle loop needs to make a tradeoff between latency and power savings
(deeper power save modes have a longer latency to running code again).  The
code that today makes this tradeoff just does a rather simple algorithm;
however this is not good enough: There are devices and use cases where a
lower latency is required than that the higher power saving states provide.
 An example would be audio playback, but another example is the ipw2100
wireless driver that right now has a very direct and ugly acpi hook to
disable some higher power states randomly when it gets certain types of
error.

The proposed solution is to have an interface where drivers can

* announce the maximum latency (in microseconds) that they can deal with
* modify this latency
* give up their constraint

and a function where the code that decides on power saving strategy can
query the current global desired maximum.

This patch has a user of each side: on the consumer side, ACPI is patched
to use this, on the producer side the ipw2100 driver is patched.

A generic maximum latency is also registered of 2 timer ticks (more and you
lose accurate time tracking after all).

While the existing users of the patch are x86 specific, the infrastructure
is not.  I'd like to ask the arch maintainers of other architectures if the
infrastructure is generic enough for their use (assuming the architecture
has such a tradeoff as concept at all), and the sound/multimedia driver
owners to look at the driver facing API to see if this is something they
can use.

[akpm@osdl.org: cleanups]
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jesse Barnes <jesse.barnes@intel.com>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/acpi/processor_idle.c  |  38 +++++-
 drivers/net/wireless/ipw2100.c |  10 ++
 include/linux/latency.h        |  25 ++++
 kernel/Makefile                |   2 +-
 kernel/latency.c               | 279 +++++++++++++++++++++++++++++++++++++++++
 5 files changed, 349 insertions(+), 5 deletions(-)
 create mode 100644 include/linux/latency.h
 create mode 100644 kernel/latency.c

(limited to 'include')

diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 71066066d626..0a395fca843b 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -38,6 +38,7 @@
 #include <linux/dmi.h>
 #include <linux/moduleparam.h>
 #include <linux/sched.h>	/* need_resched() */
+#include <linux/latency.h>
 
 #include <asm/io.h>
 #include <asm/uaccess.h>
@@ -453,7 +454,8 @@ static void acpi_processor_idle(void)
 	 */
 	if (cx->promotion.state &&
 	    ((cx->promotion.state - pr->power.states) <= max_cstate)) {
-		if (sleep_ticks > cx->promotion.threshold.ticks) {
+		if (sleep_ticks > cx->promotion.threshold.ticks &&
+		  cx->promotion.state->latency <= system_latency_constraint()) {
 			cx->promotion.count++;
 			cx->demotion.count = 0;
 			if (cx->promotion.count >=
@@ -494,8 +496,10 @@ static void acpi_processor_idle(void)
       end:
 	/*
 	 * Demote if current state exceeds max_cstate
+	 * or if the latency of the current state is unacceptable
 	 */
-	if ((pr->power.state - pr->power.states) > max_cstate) {
+	if ((pr->power.state - pr->power.states) > max_cstate ||
+		pr->power.state->latency > system_latency_constraint()) {
 		if (cx->demotion.state)
 			next_state = cx->demotion.state;
 	}
@@ -1009,9 +1013,11 @@ static int acpi_processor_power_seq_show(struct seq_file *seq, void *offset)
 
 	seq_printf(seq, "active state:            C%zd\n"
 		   "max_cstate:              C%d\n"
-		   "bus master activity:     %08x\n",
+		   "bus master activity:     %08x\n"
+		   "maximum allowed latency: %d usec\n",
 		   pr->power.state ? pr->power.state - pr->power.states : 0,
-		   max_cstate, (unsigned)pr->power.bm_activity);
+		   max_cstate, (unsigned)pr->power.bm_activity,
+		   system_latency_constraint());
 
 	seq_puts(seq, "states:\n");
 
@@ -1077,6 +1083,28 @@ static const struct file_operations acpi_processor_power_fops = {
 	.release = single_release,
 };
 
+static void smp_callback(void *v)
+{
+	/* we already woke the CPU up, nothing more to do */
+}
+
+/*
+ * This function gets called when a part of the kernel has a new latency
+ * requirement.  This means we need to get all processors out of their C-state,
+ * and then recalculate a new suitable C-state. Just do a cross-cpu IPI; that
+ * wakes them all right up.
+ */
+static int acpi_processor_latency_notify(struct notifier_block *b,
+		unsigned long l, void *v)
+{
+	smp_call_function(smp_callback, NULL, 0, 1);
+	return NOTIFY_OK;
+}
+
+static struct notifier_block acpi_processor_latency_notifier = {
+	.notifier_call = acpi_processor_latency_notify,
+};
+
 int acpi_processor_power_init(struct acpi_processor *pr,
 			      struct acpi_device *device)
 {
@@ -1093,6 +1121,7 @@ int acpi_processor_power_init(struct acpi_processor *pr,
 			       "ACPI: processor limited to max C-state %d\n",
 			       max_cstate);
 		first_run++;
+		register_latency_notifier(&acpi_processor_latency_notifier);
 	}
 
 	if (!pr)
@@ -1164,6 +1193,7 @@ int acpi_processor_power_exit(struct acpi_processor *pr,
 		 * copies of pm_idle before proceeding.
 		 */
 		cpu_idle_wait();
+		unregister_latency_notifier(&acpi_processor_latency_notifier);
 	}
 
 	return 0;
diff --git a/drivers/net/wireless/ipw2100.c b/drivers/net/wireless/ipw2100.c
index 6c5add701a6f..97937809de09 100644
--- a/drivers/net/wireless/ipw2100.c
+++ b/drivers/net/wireless/ipw2100.c
@@ -163,6 +163,7 @@ that only one external action is invoked at a time.
 #include <linux/firmware.h>
 #include <linux/acpi.h>
 #include <linux/ctype.h>
+#include <linux/latency.h>
 
 #include "ipw2100.h"
 
@@ -1697,6 +1698,11 @@ static int ipw2100_up(struct ipw2100_priv *priv, int deferred)
 		return 0;
 	}
 
+	/* the ipw2100 hardware really doesn't want power management delays
+	 * longer than 175usec
+	 */
+	modify_acceptable_latency("ipw2100", 175);
+
 	/* If the interrupt is enabled, turn it off... */
 	spin_lock_irqsave(&priv->low_lock, flags);
 	ipw2100_disable_interrupts(priv);
@@ -1849,6 +1855,8 @@ static void ipw2100_down(struct ipw2100_priv *priv)
 	ipw2100_disable_interrupts(priv);
 	spin_unlock_irqrestore(&priv->low_lock, flags);
 
+	modify_acceptable_latency("ipw2100", INFINITE_LATENCY);
+
 #ifdef ACPI_CSTATE_LIMIT_DEFINED
 	if (priv->config & CFG_C3_DISABLED) {
 		IPW_DEBUG_INFO(": Resetting C3 transitions.\n");
@@ -6534,6 +6542,7 @@ static int __init ipw2100_init(void)
 
 	ret = pci_register_driver(&ipw2100_pci_driver);
 
+	set_acceptable_latency("ipw2100", INFINITE_LATENCY);
 #ifdef CONFIG_IPW2100_DEBUG
 	ipw2100_debug_level = debug;
 	driver_create_file(&ipw2100_pci_driver.driver,
@@ -6554,6 +6563,7 @@ static void __exit ipw2100_exit(void)
 			   &driver_attr_debug_level);
 #endif
 	pci_unregister_driver(&ipw2100_pci_driver);
+	remove_acceptable_latency("ipw2100");
 }
 
 module_init(ipw2100_init);
diff --git a/include/linux/latency.h b/include/linux/latency.h
new file mode 100644
index 000000000000..c08b52bb55b0
--- /dev/null
+++ b/include/linux/latency.h
@@ -0,0 +1,25 @@
+/*
+ * latency.h: Explicit system-wide latency-expectation infrastructure
+ *
+ * (C) Copyright 2006 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ */
+
+#ifndef _INCLUDE_GUARD_LATENCY_H_
+#define _INCLUDE_GUARD_LATENCY_H_
+
+#include <linux/notifier.h>
+
+void set_acceptable_latency(char *identifier, int usecs);
+void modify_acceptable_latency(char *identifier, int usecs);
+void remove_acceptable_latency(char *identifier);
+void synchronize_acceptable_latency(void);
+int system_latency_constraint(void);
+
+int register_latency_notifier(struct notifier_block * nb);
+int unregister_latency_notifier(struct notifier_block * nb);
+
+#define INFINITE_LATENCY 1000000
+
+#endif
diff --git a/kernel/Makefile b/kernel/Makefile
index d62ec66c1af2..e210e8cf7237 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -8,7 +8,7 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
 	    signal.o sys.o kmod.o workqueue.o pid.o \
 	    rcupdate.o extable.o params.o posix-timers.o \
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
-	    hrtimer.o rwsem.o
+	    hrtimer.o rwsem.o latency.o
 
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
 obj-y += time/
diff --git a/kernel/latency.c b/kernel/latency.c
new file mode 100644
index 000000000000..258f2555abbc
--- /dev/null
+++ b/kernel/latency.c
@@ -0,0 +1,279 @@
+/*
+ * latency.c: Explicit system-wide latency-expectation infrastructure
+ *
+ * The purpose of this infrastructure is to allow device drivers to set
+ * latency constraint they have and to collect and summarize these
+ * expectations globally. The cummulated result can then be used by
+ * power management and similar users to make decisions that have
+ * tradoffs with a latency component.
+ *
+ * An example user of this are the x86 C-states; each higher C state saves
+ * more power, but has a higher exit latency. For the idle loop power
+ * code to make a good decision which C-state to use, information about
+ * acceptable latencies is required.
+ *
+ * An example announcer of latency is an audio driver that knowns it
+ * will get an interrupt when the hardware has 200 usec of samples
+ * left in the DMA buffer; in that case the driver can set a latency
+ * constraint of, say, 150 usec.
+ *
+ * Multiple drivers can each announce their maximum accepted latency,
+ * to keep these appart, a string based identifier is used.
+ *
+ *
+ * (C) Copyright 2006 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/latency.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
+#include <asm/atomic.h>
+
+struct latency_info {
+	struct list_head list;
+	int usecs;
+	char *identifier;
+};
+
+/*
+ * locking rule: all modifications to current_max_latency and
+ * latency_list need to be done while holding the latency_lock.
+ * latency_lock needs to be taken _irqsave.
+ */
+static atomic_t current_max_latency;
+static DEFINE_SPINLOCK(latency_lock);
+
+static LIST_HEAD(latency_list);
+static BLOCKING_NOTIFIER_HEAD(latency_notifier);
+
+/*
+ * This function returns the maximum latency allowed, which
+ * happens to be the minimum of all maximum latencies on the
+ * list.
+ */
+static int __find_max_latency(void)
+{
+	int min = INFINITE_LATENCY;
+	struct latency_info *info;
+
+	list_for_each_entry(info, &latency_list, list) {
+		if (info->usecs < min)
+			min = info->usecs;
+	}
+	return min;
+}
+
+/**
+ * set_acceptable_latency - sets the maximum latency acceptable
+ * @identifier: string that identifies this driver
+ * @usecs: maximum acceptable latency for this driver
+ *
+ * This function informs the kernel that this device(driver)
+ * can accept at most usecs latency. This setting is used for
+ * power management and similar tradeoffs.
+ *
+ * This function sleeps and can only be called from process
+ * context.
+ * Calling this function with an existing identifier is valid
+ * and will cause the existing latency setting to be changed.
+ */
+void set_acceptable_latency(char *identifier, int usecs)
+{
+	struct latency_info *info, *iter;
+	unsigned long flags;
+	int found_old = 0;
+
+	info = kzalloc(sizeof(struct latency_info), GFP_KERNEL);
+	if (!info)
+		return;
+	info->usecs = usecs;
+	info->identifier = kstrdup(identifier, GFP_KERNEL);
+	if (!info->identifier)
+		goto free_info;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	list_for_each_entry(iter, &latency_list, list) {
+		if (strcmp(iter->identifier, identifier)==0) {
+			found_old = 1;
+			iter->usecs = usecs;
+			break;
+		}
+	}
+	if (!found_old)
+		list_add(&info->list, &latency_list);
+
+	if (usecs < atomic_read(&current_max_latency))
+		atomic_set(&current_max_latency, usecs);
+
+	spin_unlock_irqrestore(&latency_lock, flags);
+
+	blocking_notifier_call_chain(&latency_notifier,
+		atomic_read(&current_max_latency), NULL);
+
+	/*
+	 * if we inserted the new one, we're done; otherwise there was
+	 * an existing one so we need to free the redundant data
+	 */
+	if (!found_old)
+		return;
+
+	kfree(info->identifier);
+free_info:
+	kfree(info);
+}
+EXPORT_SYMBOL_GPL(set_acceptable_latency);
+
+/**
+ * modify_acceptable_latency - changes the maximum latency acceptable
+ * @identifier: string that identifies this driver
+ * @usecs: maximum acceptable latency for this driver
+ *
+ * This function informs the kernel that this device(driver)
+ * can accept at most usecs latency. This setting is used for
+ * power management and similar tradeoffs.
+ *
+ * This function does not sleep and can be called in any context.
+ * Trying to use a non-existing identifier silently gets ignored.
+ *
+ * Due to the atomic nature of this function, the modified latency
+ * value will only be used for future decisions; past decisions
+ * can still lead to longer latencies in the near future.
+ */
+void modify_acceptable_latency(char *identifier, int usecs)
+{
+	struct latency_info *iter;
+	unsigned long flags;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	list_for_each_entry(iter, &latency_list, list) {
+		if (strcmp(iter->identifier, identifier) == 0) {
+			iter->usecs = usecs;
+			break;
+		}
+	}
+	if (usecs < atomic_read(&current_max_latency))
+		atomic_set(&current_max_latency, usecs);
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+EXPORT_SYMBOL_GPL(modify_acceptable_latency);
+
+/**
+ * remove_acceptable_latency - removes the maximum latency acceptable
+ * @identifier: string that identifies this driver
+ *
+ * This function removes a previously set maximum latency setting
+ * for the driver and frees up any resources associated with the
+ * bookkeeping needed for this.
+ *
+ * This function does not sleep and can be called in any context.
+ * Trying to use a non-existing identifier silently gets ignored.
+ */
+void remove_acceptable_latency(char *identifier)
+{
+	unsigned long flags;
+	int newmax = 0;
+	struct latency_info *iter, *temp;
+
+	spin_lock_irqsave(&latency_lock, flags);
+
+	list_for_each_entry_safe(iter,  temp, &latency_list, list) {
+		if (strcmp(iter->identifier, identifier) == 0) {
+			list_del(&iter->list);
+			newmax = iter->usecs;
+			kfree(iter->identifier);
+			kfree(iter);
+			break;
+		}
+	}
+
+	/* If we just deleted the system wide value, we need to
+	 * recalculate with a full search
+	 */
+	if (newmax == atomic_read(&current_max_latency)) {
+		newmax = __find_max_latency();
+		atomic_set(&current_max_latency, newmax);
+	}
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+EXPORT_SYMBOL_GPL(remove_acceptable_latency);
+
+/**
+ * system_latency_constraint - queries the system wide latency maximum
+ *
+ * This function returns the system wide maximum latency in
+ * microseconds.
+ *
+ * This function does not sleep and can be called in any context.
+ */
+int system_latency_constraint(void)
+{
+	return atomic_read(&current_max_latency);
+}
+EXPORT_SYMBOL_GPL(system_latency_constraint);
+
+/**
+ * synchronize_acceptable_latency - recalculates all latency decisions
+ *
+ * This function will cause a callback to various kernel pieces that
+ * will make those pieces rethink their latency decisions. This implies
+ * that if there are overlong latencies in hardware state already, those
+ * latencies get taken right now. When this call completes no overlong
+ * latency decisions should be active anymore.
+ *
+ * Typical usecase of this is after a modify_acceptable_latency() call,
+ * which in itself is non-blocking and non-synchronizing.
+ *
+ * This function blocks and should not be called with locks held.
+ */
+
+void synchronize_acceptable_latency(void)
+{
+	blocking_notifier_call_chain(&latency_notifier,
+		atomic_read(&current_max_latency), NULL);
+}
+EXPORT_SYMBOL_GPL(synchronize_acceptable_latency);
+
+/*
+ * Latency notifier: this notifier gets called when a non-atomic new
+ * latency value gets set. The expectation nof the caller of the
+ * non-atomic set is that when the call returns, future latencies
+ * are within bounds, so the functions on the notifier list are
+ * expected to take the overlong latencies immediately, inside the
+ * callback, and not make a overlong latency decision anymore.
+ *
+ * The callback gets called when the new latency value is made
+ * active so system_latency_constraint() returns the new latency.
+ */
+int register_latency_notifier(struct notifier_block * nb)
+{
+	return blocking_notifier_chain_register(&latency_notifier, nb);
+}
+EXPORT_SYMBOL_GPL(register_latency_notifier);
+
+int unregister_latency_notifier(struct notifier_block * nb)
+{
+	return blocking_notifier_chain_unregister(&latency_notifier, nb);
+}
+EXPORT_SYMBOL_GPL(unregister_latency_notifier);
+
+static __init int latency_init(void)
+{
+	atomic_set(&current_max_latency, INFINITE_LATENCY);
+	/*
+	 * we don't want by default to have longer latencies than 2 ticks,
+	 * since that would cause lost ticks
+	 */
+	set_acceptable_latency("kernel", 2*1000000/HZ);
+	return 0;
+}
+
+module_init(latency_init);
-- 
cgit v1.2.3


From 9442e691e4aec85eba43ac60a3e77c77fd2e73a4 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Sat, 30 Sep 2006 23:27:19 -0700
Subject: [PATCH] maximum latency tracking: ALSA support

Add maximum latency tracking to the ALSA subsystem for PCM playback.  In
ALSA, the playback application controls the buffer size and thus indirectly
the period of latency that it can deal with.  This patch uses 75% of the
total available latency as threshold to announce to the latency subsystem;
While 75% is a crude heuristic it's a quite reasonable one; the remaining
25% can be used for all driver processing for the next samples which is
also proportional to the size of the buffer.

With ogg123 a latency setting of about 4msec was seen (at 44Khz), while
with the "play" command a much longer maximum tolerable latency was seen.
Other, more multimedia oriented players as well as games, will have a lot
smaller buffers to allow better synchronization and those will actually get
into the latency domains where there is impact on the power management
rules.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/sound/pcm.h     |  1 +
 sound/core/pcm.c        |  3 +++
 sound/core/pcm_native.c | 23 ++++++++++++++++++++++-
 3 files changed, 26 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/sound/pcm.h b/include/sound/pcm.h
index 60d40b34efc0..afaf3e88e086 100644
--- a/include/sound/pcm.h
+++ b/include/sound/pcm.h
@@ -347,6 +347,7 @@ struct snd_pcm_substream {
 	int number;
 	char name[32];			/* substream name */
 	int stream;			/* stream (direction) */
+	char latency_id[20];		/* latency identifier */
 	size_t buffer_bytes_max;	/* limit ring buffer size */
 	struct snd_dma_buffer dma_buffer;
 	unsigned int dma_buf_id;
diff --git a/sound/core/pcm.c b/sound/core/pcm.c
index bf8f412988b8..fbbbcd20c4cc 100644
--- a/sound/core/pcm.c
+++ b/sound/core/pcm.c
@@ -629,6 +629,9 @@ int snd_pcm_new_stream(struct snd_pcm *pcm, int stream, int substream_count)
 		substream->number = idx;
 		substream->stream = stream;
 		sprintf(substream->name, "subdevice #%i", idx);
+		snprintf(substream->latency_id, sizeof(substream->latency_id),
+			 "ALSA-PCM%d-%d%c%d", pcm->card->number, pcm->device,
+			 (stream ? 'c' : 'p'), idx);
 		substream->buffer_bytes_max = UINT_MAX;
 		if (prev == NULL)
 			pstr->substream = substream;
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index 0224c70414f5..891d7140553c 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -25,6 +25,7 @@
 #include <linux/file.h>
 #include <linux/slab.h>
 #include <linux/time.h>
+#include <linux/latency.h>
 #include <linux/uio.h>
 #include <sound/core.h>
 #include <sound/control.h>
@@ -347,11 +348,26 @@ out:
 	return err;
 }
 
+static int period_to_usecs(struct snd_pcm_runtime *runtime)
+{
+	int usecs;
+
+	if (! runtime->rate)
+		return -1; /* invalid */
+
+	/* take 75% of period time as the deadline */
+	usecs = (750000 / runtime->rate) * runtime->period_size;
+	usecs += ((750000 % runtime->rate) * runtime->period_size) /
+		runtime->rate;
+
+	return usecs;
+}
+
 static int snd_pcm_hw_params(struct snd_pcm_substream *substream,
 			     struct snd_pcm_hw_params *params)
 {
 	struct snd_pcm_runtime *runtime;
-	int err;
+	int err, usecs;
 	unsigned int bits;
 	snd_pcm_uframes_t frames;
 
@@ -431,6 +447,10 @@ static int snd_pcm_hw_params(struct snd_pcm_substream *substream,
 
 	snd_pcm_timer_resolution_change(substream);
 	runtime->status->state = SNDRV_PCM_STATE_SETUP;
+
+	remove_acceptable_latency(substream->latency_id);
+	if ((usecs = period_to_usecs(runtime)) >= 0)
+		set_acceptable_latency(substream->latency_id, usecs);
 	return 0;
  _error:
 	/* hardware might be unuseable from this time,
@@ -490,6 +510,7 @@ static int snd_pcm_hw_free(struct snd_pcm_substream *substream)
 	if (substream->ops->hw_free)
 		result = substream->ops->hw_free(substream);
 	runtime->status->state = SNDRV_PCM_STATE_OPEN;
+	remove_acceptable_latency(substream->latency_id);
 	return result;
 }
 
-- 
cgit v1.2.3


From 1a2f67b459bb7846d4a15924face63eb2683acc2 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sat, 30 Sep 2006 23:27:20 -0700
Subject: [PATCH] kmemdup: introduce

One of idiomatic ways to duplicate a region of memory is

	dst = kmalloc(len, GFP_KERNEL);
	if (!dst)
		return -ENOMEM;
	memcpy(dst, src, len);

which is neat code except a programmer needs to write size twice.  Which
sometimes leads to mistakes.  If len passed to kmalloc is smaller that len
passed to memcpy, it's straight overwrite-beyond-end.  If len passed to
memcpy is smaller than len passed to kmalloc, it's either a) legit
behaviour ;-), or b) cloned buffer will contain garbage in second half.

Slight trolling of commit lists shows several duplications bugs
done exactly because of diverged lenghts:

	Linux:
		[CRYPTO]: Fix memcpy/memset args.
		[PATCH] memcpy/memset fixes
	OpenBSD:
		kerberosV/src/lib/asn1: der_copy.c:1.4

If programmer is given only one place to play with lengths, I believe, such
mistakes could be avoided.

With kmemdup, the snippet above will be rewritten as:

	dst = kmemdup(src, len, GFP_KERNEL);
	if (!dst)
		return -ENOMEM;

This also leads to smaller code (kzalloc effect). Quick grep shows
200+ places where kmemdup() can be used.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/string.h |  1 +
 mm/util.c              | 18 ++++++++++++++++++
 2 files changed, 19 insertions(+)

(limited to 'include')

diff --git a/include/linux/string.h b/include/linux/string.h
index e4c755860316..4f69ef9e6eb5 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -99,6 +99,7 @@ extern void * memchr(const void *,int,__kernel_size_t);
 #endif
 
 extern char *kstrdup(const char *s, gfp_t gfp);
+extern void *kmemdup(const void *src, size_t len, gfp_t gfp);
 
 #ifdef __cplusplus
 }
diff --git a/mm/util.c b/mm/util.c
index 7368479220b3..e14fa84ef39a 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -40,6 +40,24 @@ char *kstrdup(const char *s, gfp_t gfp)
 }
 EXPORT_SYMBOL(kstrdup);
 
+/**
+ * kmemdup - duplicate region of memory
+ *
+ * @src: memory region to duplicate
+ * @len: memory region length
+ * @gfp: GFP mask to use
+ */
+void *kmemdup(const void *src, size_t len, gfp_t gfp)
+{
+	void *p;
+
+	p = ____kmalloc(len, gfp);
+	if (p)
+		memcpy(p, src, len);
+	return p;
+}
+EXPORT_SYMBOL(kmemdup);
+
 /*
  * strndup_user - duplicate an existing string from user space
  *
-- 
cgit v1.2.3


From 82b0547cfae1fb2ee26cad588f6d49a347d24740 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sat, 30 Sep 2006 23:27:22 -0700
Subject: [PATCH] Create fs/utimes.c

* fs/open.c is getting bit crowdy
* preparation to lutimes(2)

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/Makefile           |   2 +-
 fs/open.c             | 134 ------------------------------------------------
 fs/utimes.c           | 137 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/namei.h |   1 +
 include/linux/utime.h |   2 +
 5 files changed, 141 insertions(+), 135 deletions(-)
 create mode 100644 fs/utimes.c

(limited to 'include')

diff --git a/fs/Makefile b/fs/Makefile
index a503e6ce0f32..819b2a93bebe 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -10,7 +10,7 @@ obj-y :=	open.o read_write.o file_table.o super.o \
 		ioctl.o readdir.o select.o fifo.o locks.o dcache.o inode.o \
 		attr.o bad_inode.o file.o filesystems.o namespace.o aio.o \
 		seq_file.o xattr.o libfs.o fs-writeback.o \
-		pnode.o drop_caches.o splice.o sync.o
+		pnode.o drop_caches.o splice.o sync.o utimes.o
 
 ifeq ($(CONFIG_BLOCK),y)
 obj-y +=	buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o
diff --git a/fs/open.c b/fs/open.c
index 304c1c7814cb..35c3e454458e 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -6,7 +6,6 @@
 
 #include <linux/string.h>
 #include <linux/mm.h>
-#include <linux/utime.h>
 #include <linux/file.h>
 #include <linux/smp_lock.h>
 #include <linux/quotaops.h>
@@ -29,8 +28,6 @@
 #include <linux/rcupdate.h>
 #include <linux/audit.h>
 
-#include <asm/unistd.h>
-
 int vfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	int retval = -ENODEV;
@@ -353,137 +350,6 @@ asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length)
 }
 #endif
 
-#ifdef __ARCH_WANT_SYS_UTIME
-
-/*
- * sys_utime() can be implemented in user-level using sys_utimes().
- * Is this for backwards compatibility?  If so, why not move it
- * into the appropriate arch directory (for those architectures that
- * need it).
- */
-
-/* If times==NULL, set access and modification to current time,
- * must be owner or have write permission.
- * Else, update from *times, must be owner or super user.
- */
-asmlinkage long sys_utime(char __user * filename, struct utimbuf __user * times)
-{
-	int error;
-	struct nameidata nd;
-	struct inode * inode;
-	struct iattr newattrs;
-
-	error = user_path_walk(filename, &nd);
-	if (error)
-		goto out;
-	inode = nd.dentry->d_inode;
-
-	error = -EROFS;
-	if (IS_RDONLY(inode))
-		goto dput_and_out;
-
-	/* Don't worry, the checks are done in inode_change_ok() */
-	newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
-	if (times) {
-		error = -EPERM;
-		if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
-			goto dput_and_out;
-
-		error = get_user(newattrs.ia_atime.tv_sec, &times->actime);
-		newattrs.ia_atime.tv_nsec = 0;
-		if (!error)
-			error = get_user(newattrs.ia_mtime.tv_sec, &times->modtime);
-		newattrs.ia_mtime.tv_nsec = 0;
-		if (error)
-			goto dput_and_out;
-
-		newattrs.ia_valid |= ATTR_ATIME_SET | ATTR_MTIME_SET;
-	} else {
-                error = -EACCES;
-                if (IS_IMMUTABLE(inode))
-                        goto dput_and_out;
-
-		if (current->fsuid != inode->i_uid &&
-		    (error = vfs_permission(&nd, MAY_WRITE)) != 0)
-			goto dput_and_out;
-	}
-	mutex_lock(&inode->i_mutex);
-	error = notify_change(nd.dentry, &newattrs);
-	mutex_unlock(&inode->i_mutex);
-dput_and_out:
-	path_release(&nd);
-out:
-	return error;
-}
-
-#endif
-
-/* If times==NULL, set access and modification to current time,
- * must be owner or have write permission.
- * Else, update from *times, must be owner or super user.
- */
-long do_utimes(int dfd, char __user *filename, struct timeval *times)
-{
-	int error;
-	struct nameidata nd;
-	struct inode * inode;
-	struct iattr newattrs;
-
-	error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
-
-	if (error)
-		goto out;
-	inode = nd.dentry->d_inode;
-
-	error = -EROFS;
-	if (IS_RDONLY(inode))
-		goto dput_and_out;
-
-	/* Don't worry, the checks are done in inode_change_ok() */
-	newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
-	if (times) {
-		error = -EPERM;
-                if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
-                        goto dput_and_out;
-
-		newattrs.ia_atime.tv_sec = times[0].tv_sec;
-		newattrs.ia_atime.tv_nsec = times[0].tv_usec * 1000;
-		newattrs.ia_mtime.tv_sec = times[1].tv_sec;
-		newattrs.ia_mtime.tv_nsec = times[1].tv_usec * 1000;
-		newattrs.ia_valid |= ATTR_ATIME_SET | ATTR_MTIME_SET;
-	} else {
-		error = -EACCES;
-                if (IS_IMMUTABLE(inode))
-                        goto dput_and_out;
-
-		if (current->fsuid != inode->i_uid &&
-		    (error = vfs_permission(&nd, MAY_WRITE)) != 0)
-			goto dput_and_out;
-	}
-	mutex_lock(&inode->i_mutex);
-	error = notify_change(nd.dentry, &newattrs);
-	mutex_unlock(&inode->i_mutex);
-dput_and_out:
-	path_release(&nd);
-out:
-	return error;
-}
-
-asmlinkage long sys_futimesat(int dfd, char __user *filename, struct timeval __user *utimes)
-{
-	struct timeval times[2];
-
-	if (utimes && copy_from_user(&times, utimes, sizeof(times)))
-		return -EFAULT;
-	return do_utimes(dfd, filename, utimes ? times : NULL);
-}
-
-asmlinkage long sys_utimes(char __user *filename, struct timeval __user *utimes)
-{
-	return sys_futimesat(AT_FDCWD, filename, utimes);
-}
-
-
 /*
  * access() needs to use the real uid/gid, not the effective uid/gid.
  * We do this by temporarily clearing all FS-related capabilities and
diff --git a/fs/utimes.c b/fs/utimes.c
new file mode 100644
index 000000000000..1bcd852fc4a9
--- /dev/null
+++ b/fs/utimes.c
@@ -0,0 +1,137 @@
+#include <linux/compiler.h>
+#include <linux/fs.h>
+#include <linux/linkage.h>
+#include <linux/namei.h>
+#include <linux/utime.h>
+#include <asm/uaccess.h>
+#include <asm/unistd.h>
+
+#ifdef __ARCH_WANT_SYS_UTIME
+
+/*
+ * sys_utime() can be implemented in user-level using sys_utimes().
+ * Is this for backwards compatibility?  If so, why not move it
+ * into the appropriate arch directory (for those architectures that
+ * need it).
+ */
+
+/* If times==NULL, set access and modification to current time,
+ * must be owner or have write permission.
+ * Else, update from *times, must be owner or super user.
+ */
+asmlinkage long sys_utime(char __user * filename, struct utimbuf __user * times)
+{
+	int error;
+	struct nameidata nd;
+	struct inode * inode;
+	struct iattr newattrs;
+
+	error = user_path_walk(filename, &nd);
+	if (error)
+		goto out;
+	inode = nd.dentry->d_inode;
+
+	error = -EROFS;
+	if (IS_RDONLY(inode))
+		goto dput_and_out;
+
+	/* Don't worry, the checks are done in inode_change_ok() */
+	newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
+	if (times) {
+		error = -EPERM;
+		if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+			goto dput_and_out;
+
+		error = get_user(newattrs.ia_atime.tv_sec, &times->actime);
+		newattrs.ia_atime.tv_nsec = 0;
+		if (!error)
+			error = get_user(newattrs.ia_mtime.tv_sec, &times->modtime);
+		newattrs.ia_mtime.tv_nsec = 0;
+		if (error)
+			goto dput_and_out;
+
+		newattrs.ia_valid |= ATTR_ATIME_SET | ATTR_MTIME_SET;
+	} else {
+                error = -EACCES;
+                if (IS_IMMUTABLE(inode))
+                        goto dput_and_out;
+
+		if (current->fsuid != inode->i_uid &&
+		    (error = vfs_permission(&nd, MAY_WRITE)) != 0)
+			goto dput_and_out;
+	}
+	mutex_lock(&inode->i_mutex);
+	error = notify_change(nd.dentry, &newattrs);
+	mutex_unlock(&inode->i_mutex);
+dput_and_out:
+	path_release(&nd);
+out:
+	return error;
+}
+
+#endif
+
+/* If times==NULL, set access and modification to current time,
+ * must be owner or have write permission.
+ * Else, update from *times, must be owner or super user.
+ */
+long do_utimes(int dfd, char __user *filename, struct timeval *times)
+{
+	int error;
+	struct nameidata nd;
+	struct inode * inode;
+	struct iattr newattrs;
+
+	error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
+
+	if (error)
+		goto out;
+	inode = nd.dentry->d_inode;
+
+	error = -EROFS;
+	if (IS_RDONLY(inode))
+		goto dput_and_out;
+
+	/* Don't worry, the checks are done in inode_change_ok() */
+	newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
+	if (times) {
+		error = -EPERM;
+                if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+                        goto dput_and_out;
+
+		newattrs.ia_atime.tv_sec = times[0].tv_sec;
+		newattrs.ia_atime.tv_nsec = times[0].tv_usec * 1000;
+		newattrs.ia_mtime.tv_sec = times[1].tv_sec;
+		newattrs.ia_mtime.tv_nsec = times[1].tv_usec * 1000;
+		newattrs.ia_valid |= ATTR_ATIME_SET | ATTR_MTIME_SET;
+	} else {
+		error = -EACCES;
+                if (IS_IMMUTABLE(inode))
+                        goto dput_and_out;
+
+		if (current->fsuid != inode->i_uid &&
+		    (error = vfs_permission(&nd, MAY_WRITE)) != 0)
+			goto dput_and_out;
+	}
+	mutex_lock(&inode->i_mutex);
+	error = notify_change(nd.dentry, &newattrs);
+	mutex_unlock(&inode->i_mutex);
+dput_and_out:
+	path_release(&nd);
+out:
+	return error;
+}
+
+asmlinkage long sys_futimesat(int dfd, char __user *filename, struct timeval __user *utimes)
+{
+	struct timeval times[2];
+
+	if (utimes && copy_from_user(&times, utimes, sizeof(times)))
+		return -EFAULT;
+	return do_utimes(dfd, filename, utimes ? times : NULL);
+}
+
+asmlinkage long sys_utimes(char __user *filename, struct timeval __user *utimes)
+{
+	return sys_futimesat(AT_FDCWD, filename, utimes);
+}
diff --git a/include/linux/namei.h b/include/linux/namei.h
index c6470ba00668..f5f19606effb 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_NAMEI_H
 #define _LINUX_NAMEI_H
 
+#include <linux/dcache.h>
 #include <linux/linkage.h>
 
 struct vfsmount;
diff --git a/include/linux/utime.h b/include/linux/utime.h
index c6bf27b7897e..640be6a1959e 100644
--- a/include/linux/utime.h
+++ b/include/linux/utime.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_UTIME_H
 #define _LINUX_UTIME_H
 
+#include <linux/types.h>
+
 struct utimbuf {
 	time_t actime;
 	time_t modtime;
-- 
cgit v1.2.3


From ef6edc9746dc2bfdacf44eefd5f881179971c478 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Sat, 30 Sep 2006 23:27:43 -0700
Subject: [PATCH] Directed yield: cpu_relax variants for spinlocks and rw-locks

On systems running with virtual cpus there is optimization potential in
regard to spinlocks and rw-locks.  If the virtual cpu that has taken a lock
is known to a cpu that wants to acquire the same lock it is beneficial to
yield the timeslice of the virtual cpu in favour of the cpu that has the
lock (directed yield).

With CONFIG_PREEMPT="n" this can be implemented by the architecture without
common code changes.  Powerpc already does this.

With CONFIG_PREEMPT="y" the lock loops are coded with _raw_spin_trylock,
_raw_read_trylock and _raw_write_trylock in kernel/spinlock.c.  If the lock
could not be taken cpu_relax is called.  A directed yield is not possible
because cpu_relax doesn't know anything about the lock.  To be able to
yield the lock in favour of the current lock holder variants of cpu_relax
for spinlocks and rw-locks are needed.  The new _raw_spin_relax,
_raw_read_relax and _raw_write_relax primitives differ from cpu_relax
insofar that they have an argument: a pointer to the lock structure.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-alpha/spinlock.h         | 4 ++++
 include/asm-arm/spinlock.h           | 4 ++++
 include/asm-cris/arch-v32/spinlock.h | 4 ++++
 include/asm-i386/spinlock.h          | 4 ++++
 include/asm-ia64/spinlock.h          | 4 ++++
 include/asm-m32r/spinlock.h          | 4 ++++
 include/asm-mips/spinlock.h          | 4 ++++
 include/asm-parisc/spinlock.h        | 4 ++++
 include/asm-powerpc/spinlock.h       | 4 ++++
 include/asm-ppc/spinlock.h           | 4 ++++
 include/asm-s390/spinlock.h          | 4 ++++
 include/asm-sh/spinlock.h            | 4 ++++
 include/asm-sparc/spinlock.h         | 4 ++++
 include/asm-sparc64/spinlock.h       | 4 ++++
 include/asm-x86_64/spinlock.h        | 4 ++++
 kernel/spinlock.c                    | 4 ++--
 16 files changed, 62 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-alpha/spinlock.h b/include/asm-alpha/spinlock.h
index 0c294c9b0c55..aeeb125f6851 100644
--- a/include/asm-alpha/spinlock.h
+++ b/include/asm-alpha/spinlock.h
@@ -166,4 +166,8 @@ static inline void __raw_write_unlock(raw_rwlock_t * lock)
 	lock->lock = 0;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* _ALPHA_SPINLOCK_H */
diff --git a/include/asm-arm/spinlock.h b/include/asm-arm/spinlock.h
index 01b7c26a3038..861092fbaa53 100644
--- a/include/asm-arm/spinlock.h
+++ b/include/asm-arm/spinlock.h
@@ -218,4 +218,8 @@ static inline int __raw_read_trylock(raw_rwlock_t *rw)
 /* read_can_lock - would read_trylock() succeed? */
 #define __raw_read_can_lock(x)		((x)->lock < 0x80000000)
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-cris/arch-v32/spinlock.h b/include/asm-cris/arch-v32/spinlock.h
index 52df72a62232..5f43df0a5fb4 100644
--- a/include/asm-cris/arch-v32/spinlock.h
+++ b/include/asm-cris/arch-v32/spinlock.h
@@ -160,4 +160,8 @@ static __inline__ int is_write_locked(rwlock_t *rw)
 	return rw->counter < 0;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_ARCH_SPINLOCK_H */
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index b0b3043f05e1..c18b71fae6b3 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -205,4 +205,8 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw)
 				 : "+m" (rw->lock) : : "memory");
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-ia64/spinlock.h b/include/asm-ia64/spinlock.h
index 9e83210dc312..ff857e31738a 100644
--- a/include/asm-ia64/spinlock.h
+++ b/include/asm-ia64/spinlock.h
@@ -213,4 +213,8 @@ static inline int __raw_read_trylock(raw_rwlock_t *x)
 	return (u32)ia64_cmpxchg4_acq((__u32 *)(x), new.word, old.word) == old.word;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /*  _ASM_IA64_SPINLOCK_H */
diff --git a/include/asm-m32r/spinlock.h b/include/asm-m32r/spinlock.h
index f9f90727a4a1..f5cfba81ee10 100644
--- a/include/asm-m32r/spinlock.h
+++ b/include/asm-m32r/spinlock.h
@@ -316,4 +316,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock)
 	return 0;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif	/* _ASM_M32R_SPINLOCK_H */
diff --git a/include/asm-mips/spinlock.h b/include/asm-mips/spinlock.h
index 4c1a1b53aeaf..c8d5587467bb 100644
--- a/include/asm-mips/spinlock.h
+++ b/include/asm-mips/spinlock.h
@@ -328,4 +328,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 }
 
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* _ASM_SPINLOCK_H */
diff --git a/include/asm-parisc/spinlock.h b/include/asm-parisc/spinlock.h
index a93960e232cf..e1825530365d 100644
--- a/include/asm-parisc/spinlock.h
+++ b/include/asm-parisc/spinlock.h
@@ -152,4 +152,8 @@ static __inline__ int __raw_write_can_lock(raw_rwlock_t *rw)
 	return !rw->counter;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-powerpc/spinlock.h b/include/asm-powerpc/spinlock.h
index c31e4382a775..eaccac8327f2 100644
--- a/include/asm-powerpc/spinlock.h
+++ b/include/asm-powerpc/spinlock.h
@@ -285,5 +285,9 @@ static __inline__ void __raw_write_unlock(raw_rwlock_t *rw)
 	rw->lock = 0;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-ppc/spinlock.h b/include/asm-ppc/spinlock.h
index 5c64b75f0295..fccaf5531e57 100644
--- a/include/asm-ppc/spinlock.h
+++ b/include/asm-ppc/spinlock.h
@@ -161,4 +161,8 @@ static __inline__ void __raw_write_unlock(raw_rwlock_t *rw)
 	rw->lock = 0;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h
index ce3edf6d63b3..5f00feaf1be6 100644
--- a/include/asm-s390/spinlock.h
+++ b/include/asm-s390/spinlock.h
@@ -154,4 +154,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 	return _raw_write_trylock_retry(rw);
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SPINLOCK_H */
diff --git a/include/asm-sh/spinlock.h b/include/asm-sh/spinlock.h
index 846322d4c35d..54458fd24981 100644
--- a/include/asm-sh/spinlock.h
+++ b/include/asm-sh/spinlock.h
@@ -100,4 +100,8 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 	return 0;
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SH_SPINLOCK_H */
diff --git a/include/asm-sparc/spinlock.h b/include/asm-sparc/spinlock.h
index 1c75474ba1df..557d08959d2f 100644
--- a/include/asm-sparc/spinlock.h
+++ b/include/asm-sparc/spinlock.h
@@ -154,6 +154,10 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 #define __raw_read_trylock(lock) generic__raw_read_trylock(lock)
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #define __raw_read_can_lock(rw) (!((rw)->lock & 0xff))
 #define __raw_write_can_lock(rw) (!(rw)->lock)
 
diff --git a/include/asm-sparc64/spinlock.h b/include/asm-sparc64/spinlock.h
index bd5ffc76bc7e..0006fe9f8c7a 100644
--- a/include/asm-sparc64/spinlock.h
+++ b/include/asm-sparc64/spinlock.h
@@ -241,6 +241,10 @@ static int inline __write_trylock(raw_rwlock_t *lock)
 #define __raw_read_can_lock(rw)		(!((rw)->lock & 0x80000000UL))
 #define __raw_write_can_lock(rw)	(!(rw)->lock)
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* !(__ASSEMBLY__) */
 
 #endif /* !(__SPARC64_SPINLOCK_H) */
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index 3daf5b005905..05ef097ba55b 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -133,4 +133,8 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw)
 				: "=m" (rw->lock) : : "memory");
 }
 
+#define _raw_spin_relax(lock)	cpu_relax()
+#define _raw_read_relax(lock)	cpu_relax()
+#define _raw_write_relax(lock)	cpu_relax()
+
 #endif /* __ASM_SPINLOCK_H */
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index d48143eafbfd..476c3741511b 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -215,7 +215,7 @@ void __lockfunc _##op##_lock(locktype##_t *lock)			\
 		if (!(lock)->break_lock)				\
 			(lock)->break_lock = 1;				\
 		while (!op##_can_lock(lock) && (lock)->break_lock)	\
-			cpu_relax();					\
+			_raw_##op##_relax(&lock->raw_lock);		\
 	}								\
 	(lock)->break_lock = 0;						\
 }									\
@@ -237,7 +237,7 @@ unsigned long __lockfunc _##op##_lock_irqsave(locktype##_t *lock)	\
 		if (!(lock)->break_lock)				\
 			(lock)->break_lock = 1;				\
 		while (!op##_can_lock(lock) && (lock)->break_lock)	\
-			cpu_relax();					\
+			_raw_##op##_relax(&lock->raw_lock);		\
 	}								\
 	(lock)->break_lock = 0;						\
 	return flags;							\
-- 
cgit v1.2.3


From cdc39363d33506b0e067d41fc91f89d186bdf7f7 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Sat, 30 Sep 2006 23:27:44 -0700
Subject: [PATCH] Directed yield: direct yield of spinlocks for powerpc

Powerpc already has a directed yield for CONFIG_PREEMPT="n".  To make it
work with CONFIG_PREEMPT="y" as well the _raw_{spin,read,write}_relax
primitives need to be defined to call __spin_yield() for spinlocks and
__rw_yield() for rw-locks.

Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-powerpc/spinlock.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/asm-powerpc/spinlock.h b/include/asm-powerpc/spinlock.h
index eaccac8327f2..cc4cfceac67c 100644
--- a/include/asm-powerpc/spinlock.h
+++ b/include/asm-powerpc/spinlock.h
@@ -285,9 +285,9 @@ static __inline__ void __raw_write_unlock(raw_rwlock_t *rw)
 	rw->lock = 0;
 }
 
-#define _raw_spin_relax(lock)	cpu_relax()
-#define _raw_read_relax(lock)	cpu_relax()
-#define _raw_write_relax(lock)	cpu_relax()
+#define _raw_spin_relax(lock)	__spin_yield(lock)
+#define _raw_read_relax(lock)	__rw_yield(lock)
+#define _raw_write_relax(lock)	__rw_yield(lock)
 
 #endif /* __KERNEL__ */
 #endif /* __ASM_SPINLOCK_H */
-- 
cgit v1.2.3


From 3c1fcfe229e99752c74efb945a4a3f560be04204 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Sat, 30 Sep 2006 23:27:45 -0700
Subject: [PATCH] Directed yield: direct yield of spinlocks for s390.

Use the new diagnose 0x9c in the spinlock implementation for s390.  It
yields the remaining timeslice of the virtual cpu that tries to acquire a
lock to the virtual cpu that is the current holder of the lock.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/head31.S         | 11 +++++++
 arch/s390/kernel/head64.S         | 11 +++++++
 arch/s390/lib/spinlock.c          | 62 ++++++++++++++++++++++++---------------
 include/asm-s390/setup.h          |  1 +
 include/asm-s390/spinlock.h       | 33 ++++++++++++++-------
 include/asm-s390/spinlock_types.h |  6 ++--
 6 files changed, 88 insertions(+), 36 deletions(-)

(limited to 'include')

diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S
index 1fa9fa1ca740..1b952a3664e2 100644
--- a/arch/s390/kernel/head31.S
+++ b/arch/s390/kernel/head31.S
@@ -254,6 +254,16 @@ startup_continue:
 	oi	3(%r12),0x80		# set IDTE flag
 .Lchkidte:
 
+#
+# find out if the diag 0x9c is available
+#
+	mvc	__LC_PGM_NEW_PSW(8),.Lpcdiag9c-.LPG1(%r13)
+	stap   __LC_CPUID+4		# store cpu address
+	lh     %r1,__LC_CPUID+4
+	diag   %r1,0,0x9c		# test diag 0x9c
+	oi     2(%r12),1		# set diag9c flag
+.Lchkdiag9c:
+
 	lpsw  .Lentry-.LPG1(13)		# jump to _stext in primary-space,
 					# virtual and never return ...
 	.align	8
@@ -281,6 +291,7 @@ startup_continue:
 .Lpccsp:.long	0x00080000,0x80000000 + .Lchkcsp
 .Lpcmvpg:.long	0x00080000,0x80000000 + .Lchkmvpg
 .Lpcidte:.long	0x00080000,0x80000000 + .Lchkidte
+.Lpcdiag9c:.long 0x00080000,0x80000000 + .Lchkdiag9c
 .Lmemsize:.long memory_size
 .Lmchunk:.long	memory_chunk
 .Lmflags:.long	machine_flags
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 48998d50b00a..b30e5897cdf7 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -250,6 +250,17 @@ startup_continue:
 	oi	7(%r12),0x80		# set IDTE flag
 0:
 
+#
+# find out if the diag 0x9c is available
+#
+	la     %r1,0f-.LPG1(%r13)	# set program check address
+	stg    %r1,__LC_PGM_NEW_PSW+8
+	stap   __LC_CPUID+4		# store cpu address
+	lh     %r1,__LC_CPUID+4
+	diag   %r1,0,0x9c		# test diag 0x9c
+	oi     6(%r12),1		# set diag9c flag
+0:
+
 #
 # find out if we have the MVCOS instruction
 #
diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c
index b9b7958a226a..8d76403fcf89 100644
--- a/arch/s390/lib/spinlock.c
+++ b/arch/s390/lib/spinlock.c
@@ -24,57 +24,76 @@ static int __init spin_retry_setup(char *str)
 }
 __setup("spin_retry=", spin_retry_setup);
 
-static inline void
-_diag44(void)
+static inline void _raw_yield(void)
 {
-#ifdef CONFIG_64BIT
 	if (MACHINE_HAS_DIAG44)
-#endif
 		asm volatile("diag 0,0,0x44");
 }
 
-void
-_raw_spin_lock_wait(raw_spinlock_t *lp, unsigned int pc)
+static inline void _raw_yield_cpu(int cpu)
+{
+	if (MACHINE_HAS_DIAG9C)
+		asm volatile("diag %0,0,0x9c"
+			     : : "d" (__cpu_logical_map[cpu]));
+	else
+		_raw_yield();
+}
+
+void _raw_spin_lock_wait(raw_spinlock_t *lp, unsigned int pc)
 {
 	int count = spin_retry;
+	unsigned int cpu = ~smp_processor_id();
 
 	while (1) {
 		if (count-- <= 0) {
-			_diag44();
+			unsigned int owner = lp->owner_cpu;
+			if (owner != 0)
+				_raw_yield_cpu(~owner);
 			count = spin_retry;
 		}
 		if (__raw_spin_is_locked(lp))
 			continue;
-		if (_raw_compare_and_swap(&lp->lock, 0, pc) == 0)
+		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) {
+			lp->owner_pc = pc;
 			return;
+		}
 	}
 }
 EXPORT_SYMBOL(_raw_spin_lock_wait);
 
-int
-_raw_spin_trylock_retry(raw_spinlock_t *lp, unsigned int pc)
+int _raw_spin_trylock_retry(raw_spinlock_t *lp, unsigned int pc)
 {
-	int count = spin_retry;
+	unsigned int cpu = ~smp_processor_id();
+	int count;
 
-	while (count-- > 0) {
+	for (count = spin_retry; count > 0; count--) {
 		if (__raw_spin_is_locked(lp))
 			continue;
-		if (_raw_compare_and_swap(&lp->lock, 0, pc) == 0)
+		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0) {
+			lp->owner_pc = pc;
 			return 1;
+		}
 	}
 	return 0;
 }
 EXPORT_SYMBOL(_raw_spin_trylock_retry);
 
-void
-_raw_read_lock_wait(raw_rwlock_t *rw)
+void _raw_spin_relax(raw_spinlock_t *lock)
+{
+	unsigned int cpu = lock->owner_cpu;
+	if (cpu != 0)
+		_raw_yield_cpu(~cpu);
+}
+EXPORT_SYMBOL(_raw_spin_relax);
+
+void _raw_read_lock_wait(raw_rwlock_t *rw)
 {
 	unsigned int old;
 	int count = spin_retry;
 
 	while (1) {
 		if (count-- <= 0) {
-			_diag44();
+			_raw_yield();
 			count = spin_retry;
 		}
 		if (!__raw_read_can_lock(rw))
@@ -86,8 +105,7 @@ _raw_read_lock_wait(raw_rwlock_t *rw)
 }
 EXPORT_SYMBOL(_raw_read_lock_wait);
 
-int
-_raw_read_trylock_retry(raw_rwlock_t *rw)
+int _raw_read_trylock_retry(raw_rwlock_t *rw)
 {
 	unsigned int old;
 	int count = spin_retry;
@@ -103,14 +121,13 @@ _raw_read_trylock_retry(raw_rwlock_t *rw)
 }
 EXPORT_SYMBOL(_raw_read_trylock_retry);
 
-void
-_raw_write_lock_wait(raw_rwlock_t *rw)
+void _raw_write_lock_wait(raw_rwlock_t *rw)
 {
 	int count = spin_retry;
 
 	while (1) {
 		if (count-- <= 0) {
-			_diag44();
+			_raw_yield();
 			count = spin_retry;
 		}
 		if (!__raw_write_can_lock(rw))
@@ -121,8 +138,7 @@ _raw_write_lock_wait(raw_rwlock_t *rw)
 }
 EXPORT_SYMBOL(_raw_write_lock_wait);
 
-int
-_raw_write_trylock_retry(raw_rwlock_t *rw)
+int _raw_write_trylock_retry(raw_rwlock_t *rw)
 {
 	int count = spin_retry;
 
diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h
index f1959732b6fd..5d72eda8a11b 100644
--- a/include/asm-s390/setup.h
+++ b/include/asm-s390/setup.h
@@ -39,6 +39,7 @@ extern unsigned long machine_flags;
 #define MACHINE_IS_P390		(machine_flags & 4)
 #define MACHINE_HAS_MVPG	(machine_flags & 16)
 #define MACHINE_HAS_IDTE	(machine_flags & 128)
+#define MACHINE_HAS_DIAG9C	(machine_flags & 256)
 
 #ifndef __s390x__
 #define MACHINE_HAS_IEEE	(machine_flags & 2)
diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h
index 5f00feaf1be6..6b78af16999b 100644
--- a/include/asm-s390/spinlock.h
+++ b/include/asm-s390/spinlock.h
@@ -13,6 +13,8 @@
 
 #if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
 
+#include <linux/smp.h>
+
 static inline int
 _raw_compare_and_swap(volatile unsigned int *lock,
 		      unsigned int old, unsigned int new)
@@ -50,34 +52,46 @@ _raw_compare_and_swap(volatile unsigned int *lock,
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-#define __raw_spin_is_locked(x) ((x)->lock != 0)
+#define __raw_spin_is_locked(x) ((x)->owner_cpu != 0)
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 #define __raw_spin_unlock_wait(lock) \
-	do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
+	do { while (__raw_spin_is_locked(lock)) \
+		 _raw_spin_relax(lock); } while (0)
 
-extern void _raw_spin_lock_wait(raw_spinlock_t *lp, unsigned int pc);
-extern int _raw_spin_trylock_retry(raw_spinlock_t *lp, unsigned int pc);
+extern void _raw_spin_lock_wait(raw_spinlock_t *, unsigned int pc);
+extern int _raw_spin_trylock_retry(raw_spinlock_t *, unsigned int pc);
+extern void _raw_spin_relax(raw_spinlock_t *lock);
 
 static inline void __raw_spin_lock(raw_spinlock_t *lp)
 {
 	unsigned long pc = 1 | (unsigned long) __builtin_return_address(0);
-
-	if (unlikely(_raw_compare_and_swap(&lp->lock, 0, pc) != 0))
-		_raw_spin_lock_wait(lp, pc);
+	int old;
+
+	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
+	if (likely(old == 0)) {
+		lp->owner_pc = pc;
+		return;
+	}
+	_raw_spin_lock_wait(lp, pc);
 }
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lp)
 {
 	unsigned long pc = 1 | (unsigned long) __builtin_return_address(0);
+	int old;
 
-	if (likely(_raw_compare_and_swap(&lp->lock, 0, pc) == 0))
+	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
+	if (likely(old == 0)) {
+		lp->owner_pc = pc;
 		return 1;
+	}
 	return _raw_spin_trylock_retry(lp, pc);
 }
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lp)
 {
-	_raw_compare_and_swap(&lp->lock, lp->lock, 0);
+	lp->owner_pc = 0;
+	_raw_compare_and_swap(&lp->owner_cpu, lp->owner_cpu, 0);
 }
 		
 /*
@@ -154,7 +168,6 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 	return _raw_write_trylock_retry(rw);
 }
 
-#define _raw_spin_relax(lock)	cpu_relax()
 #define _raw_read_relax(lock)	cpu_relax()
 #define _raw_write_relax(lock)	cpu_relax()
 
diff --git a/include/asm-s390/spinlock_types.h b/include/asm-s390/spinlock_types.h
index f79a2216204f..b7ac13f7aa37 100644
--- a/include/asm-s390/spinlock_types.h
+++ b/include/asm-s390/spinlock_types.h
@@ -6,16 +6,16 @@
 #endif
 
 typedef struct {
-	volatile unsigned int lock;
+	volatile unsigned int owner_cpu;
+	volatile unsigned int owner_pc;
 } __attribute__ ((aligned (4))) raw_spinlock_t;
 
 #define __RAW_SPIN_LOCK_UNLOCKED	{ 0 }
 
 typedef struct {
 	volatile unsigned int lock;
-	volatile unsigned int owner_pc;
 } raw_rwlock_t;
 
-#define __RAW_RW_LOCK_UNLOCKED		{ 0, 0 }
+#define __RAW_RW_LOCK_UNLOCKED		{ 0 }
 
 #endif
-- 
cgit v1.2.3


From cb10dc9ac7eea2c891df6b79b9ef1fbe59cb5429 Mon Sep 17 00:00:00 2001
From: Paul Fulghum <paulkf@microgate.com>
Date: Sat, 30 Sep 2006 23:27:45 -0700
Subject: [PATCH] synclink_gt: add bisync and monosync modes

Add bisync and monosync serial protocol support to the synclink_gt driver.

Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/synclink_gt.c | 88 +++++++++++++++++++++++++++++++++-------------
 include/linux/synclink.h   |  4 ++-
 2 files changed, 66 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 2f07b085536b..bd3821679ac8 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -461,7 +461,7 @@ static int  adapter_test(struct slgt_info *info);
 static void reset_adapter(struct slgt_info *info);
 static void reset_port(struct slgt_info *info);
 static void async_mode(struct slgt_info *info);
-static void hdlc_mode(struct slgt_info *info);
+static void sync_mode(struct slgt_info *info);
 
 static void rx_stop(struct slgt_info *info);
 static void rx_start(struct slgt_info *info);
@@ -881,7 +881,9 @@ static int write(struct tty_struct *tty,
 	if (!count)
 		goto cleanup;
 
-	if (info->params.mode == MGSL_MODE_RAW) {
+	if (info->params.mode == MGSL_MODE_RAW ||
+	    info->params.mode == MGSL_MODE_MONOSYNC ||
+	    info->params.mode == MGSL_MODE_BISYNC) {
 		unsigned int bufs_needed = (count/DMABUFSIZE);
 		unsigned int bufs_free = free_tbuf_count(info);
 		if (count % DMABUFSIZE)
@@ -1897,6 +1899,8 @@ static void bh_handler(void* context)
 				while(rx_get_frame(info));
 				break;
 			case MGSL_MODE_RAW:
+			case MGSL_MODE_MONOSYNC:
+			case MGSL_MODE_BISYNC:
 				while(rx_get_buf(info));
 				break;
 			}
@@ -2362,10 +2366,9 @@ static void program_hw(struct slgt_info *info)
 	rx_stop(info);
 	tx_stop(info);
 
-	if (info->params.mode == MGSL_MODE_HDLC ||
-	    info->params.mode == MGSL_MODE_RAW ||
+	if (info->params.mode != MGSL_MODE_ASYNC ||
 	    info->netcount)
-		hdlc_mode(info);
+		sync_mode(info);
 	else
 		async_mode(info);
 
@@ -2564,6 +2567,10 @@ static int rx_enable(struct slgt_info *info, int enable)
 	if (enable) {
 		if (!info->rx_enabled)
 			rx_start(info);
+		else if (enable == 2) {
+			/* force hunt mode (write 1 to RCR[3]) */
+			wr_reg16(info, RCR, rd_reg16(info, RCR) | BIT3);
+		}
 	} else {
 		if (info->rx_enabled)
 			rx_stop(info);
@@ -3748,7 +3755,7 @@ static void tx_start(struct slgt_info *info)
 {
 	if (!info->tx_enabled) {
 		wr_reg16(info, TCR,
-			(unsigned short)(rd_reg16(info, TCR) | BIT1));
+			 (unsigned short)((rd_reg16(info, TCR) | BIT1) & ~BIT2));
 		info->tx_enabled = TRUE;
 	}
 
@@ -3775,13 +3782,18 @@ static void tx_start(struct slgt_info *info)
 				tdma_reset(info);
 				/* set 1st descriptor address */
 				wr_reg32(info, TDDAR, info->tbufs[info->tbuf_start].pdesc);
-				if (info->params.mode == MGSL_MODE_RAW)
+				switch(info->params.mode) {
+				case MGSL_MODE_RAW:
+				case MGSL_MODE_MONOSYNC:
+				case MGSL_MODE_BISYNC:
 					wr_reg32(info, TDCSR, BIT2 + BIT0); /* IRQ + DMA enable */
-				else
+					break;
+				default:
 					wr_reg32(info, TDCSR, BIT0); /* DMA enable */
+				}
 			}
 
-			if (info->params.mode != MGSL_MODE_RAW) {
+			if (info->params.mode == MGSL_MODE_HDLC) {
 				info->tx_timer.expires = jiffies + msecs_to_jiffies(5000);
 				add_timer(&info->tx_timer);
 			}
@@ -3814,7 +3826,6 @@ static void tx_stop(struct slgt_info *info)
 	/* reset and disable transmitter */
 	val = rd_reg16(info, TCR) & ~BIT1;          /* clear enable bit */
 	wr_reg16(info, TCR, (unsigned short)(val | BIT2)); /* set reset bit */
-	wr_reg16(info, TCR, val);                  /* clear reset */
 
 	slgt_irq_off(info, IRQ_TXDATA + IRQ_TXIDLE + IRQ_TXUNDER);
 
@@ -3982,7 +3993,7 @@ static void async_mode(struct slgt_info *info)
 		enable_loopback(info);
 }
 
-static void hdlc_mode(struct slgt_info *info)
+static void sync_mode(struct slgt_info *info)
 {
 	unsigned short val;
 
@@ -3992,7 +4003,7 @@ static void hdlc_mode(struct slgt_info *info)
 
 	/* TCR (tx control)
 	 *
-	 * 15..13  mode, 000=HDLC 001=raw sync
+	 * 15..13  mode, 000=HDLC 001=raw 010=async 011=monosync 100=bisync
 	 * 12..10  encoding
 	 * 09      CRC enable
 	 * 08      CRC32
@@ -4006,8 +4017,11 @@ static void hdlc_mode(struct slgt_info *info)
 	 */
 	val = 0;
 
-	if (info->params.mode == MGSL_MODE_RAW)
-		val |= BIT13;
+	switch(info->params.mode) {
+	case MGSL_MODE_MONOSYNC: val |= BIT14 + BIT13; break;
+	case MGSL_MODE_BISYNC:   val |= BIT15; break;
+	case MGSL_MODE_RAW:      val |= BIT13; break;
+	}
 	if (info->if_mode & MGSL_INTERFACE_RTS_EN)
 		val |= BIT7;
 
@@ -4058,7 +4072,7 @@ static void hdlc_mode(struct slgt_info *info)
 
 	/* RCR (rx control)
 	 *
-	 * 15..13  mode, 000=HDLC 001=raw sync
+	 * 15..13  mode, 000=HDLC 001=raw 010=async 011=monosync 100=bisync
 	 * 12..10  encoding
 	 * 09      CRC enable
 	 * 08      CRC32
@@ -4069,8 +4083,11 @@ static void hdlc_mode(struct slgt_info *info)
 	 */
 	val = 0;
 
-	if (info->params.mode == MGSL_MODE_RAW)
-		val |= BIT13;
+	switch(info->params.mode) {
+	case MGSL_MODE_MONOSYNC: val |= BIT14 + BIT13; break;
+	case MGSL_MODE_BISYNC:   val |= BIT15; break;
+	case MGSL_MODE_RAW:      val |= BIT13; break;
+	}
 
 	switch(info->params.encoding)
 	{
@@ -4309,10 +4326,15 @@ static void free_rbufs(struct slgt_info *info, unsigned int i, unsigned int last
 	while(!done) {
 		/* reset current buffer for reuse */
 		info->rbufs[i].status = 0;
-		if (info->params.mode == MGSL_MODE_RAW)
+		switch(info->params.mode) {
+		case MGSL_MODE_RAW:
+		case MGSL_MODE_MONOSYNC:
+		case MGSL_MODE_BISYNC:
 			set_desc_count(info->rbufs[i], info->raw_rx_size);
-		else
+			break;
+		default:
 			set_desc_count(info->rbufs[i], DMABUFSIZE);
+		}
 
 		if (i == last)
 			done = 1;
@@ -4477,13 +4499,24 @@ cleanup:
 static int rx_get_buf(struct slgt_info *info)
 {
 	unsigned int i = info->rbuf_current;
+	unsigned int count;
 
 	if (!desc_complete(info->rbufs[i]))
 		return 0;
-	DBGDATA(info, info->rbufs[i].buf, desc_count(info->rbufs[i]), "rx");
-	DBGINFO(("rx_get_buf size=%d\n", desc_count(info->rbufs[i])));
-	ldisc_receive_buf(info->tty, info->rbufs[i].buf,
-			  info->flag_buf, desc_count(info->rbufs[i]));
+	count = desc_count(info->rbufs[i]);
+	switch(info->params.mode) {
+	case MGSL_MODE_MONOSYNC:
+	case MGSL_MODE_BISYNC:
+		/* ignore residue in byte synchronous modes */
+		if (desc_residue(info->rbufs[i]))
+			count--;
+		break;
+	}
+	DBGDATA(info, info->rbufs[i].buf, count, "rx");
+	DBGINFO(("rx_get_buf size=%d\n", count));
+	if (count)
+		ldisc_receive_buf(info->tty, info->rbufs[i].buf,
+				  info->flag_buf, count);
 	free_rbufs(info, i, i);
 	return 1;
 }
@@ -4549,8 +4582,13 @@ static void tx_load(struct slgt_info *info, const char *buf, unsigned int size)
 		size -= count;
 		buf  += count;
 
-		if (!size && info->params.mode != MGSL_MODE_RAW)
-			set_desc_eof(*d, 1); /* HDLC: set EOF of last desc */
+		/*
+		 * set EOF bit for last buffer of HDLC frame or
+		 * for every buffer in raw mode
+		 */
+		if ((!size && info->params.mode == MGSL_MODE_HDLC) ||
+		    info->params.mode == MGSL_MODE_RAW)
+			set_desc_eof(*d, 1);
 		else
 			set_desc_eof(*d, 0);
 
diff --git a/include/linux/synclink.h b/include/linux/synclink.h
index 0577f5284cbc..c8b042667af1 100644
--- a/include/linux/synclink.h
+++ b/include/linux/synclink.h
@@ -1,7 +1,7 @@
 /*
  * SyncLink Multiprotocol Serial Adapter Driver
  *
- * $Id: synclink.h,v 3.13 2006/05/23 18:25:06 paulkf Exp $
+ * $Id: synclink.h,v 3.14 2006/07/17 20:15:43 paulkf Exp $
  *
  * Copyright (C) 1998-2000 by Microgate Corporation
  *
@@ -124,6 +124,8 @@
 
 #define MGSL_MODE_ASYNC		1
 #define MGSL_MODE_HDLC		2
+#define MGSL_MODE_MONOSYNC	3
+#define MGSL_MODE_BISYNC	4
 #define MGSL_MODE_RAW		6
 
 #define MGSL_BUS_TYPE_ISA	1
-- 
cgit v1.2.3


From fb48388337182013bce811b9c336e8e64b0c858b Mon Sep 17 00:00:00 2001
From: Olaf Hering <olaf@aepfle.de>
Date: Sat, 30 Sep 2006 23:27:51 -0700
Subject: [PATCH] remove SYSRQ_KEY and related defines from ppc/sh/h8300

Remove unused global SYSRQ_KEY from ppc and powerpc
Remove unused define SYSRQ_KEY from sh/sh64 and h8300
Remove unused pckbd_sysrq_xlate and kbd_sysrq_xlate usage

Signed-off-by: Olaf Hering <olaf@aepfle.de>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/kernel/setup_32.c    | 4 ----
 arch/powerpc/kernel/setup_64.c    | 5 -----
 arch/ppc/kernel/setup.c           | 4 ----
 include/asm-h8300/keyboard.h      | 8 --------
 include/asm-sh/ec3104/keyboard.h  | 2 --
 include/asm-sh/mpc1211/keyboard.h | 4 ----
 include/asm-sh64/keyboard.h       | 4 ----
 7 files changed, 31 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index e0df2ba1ab9f..79a17795d17b 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -67,10 +67,6 @@ int have_of = 1;
 dev_t boot_dev;
 #endif /* CONFIG_PPC_MULTIPLATFORM */
 
-#ifdef CONFIG_MAGIC_SYSRQ
-unsigned long SYSRQ_KEY = 0x54;
-#endif /* CONFIG_MAGIC_SYSRQ */
-
 #ifdef CONFIG_VGA_CONSOLE
 unsigned long vgacon_remap_base;
 #endif
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 00d6b8addd78..962ad5ebc767 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -93,11 +93,6 @@ int dcache_bsize;
 int icache_bsize;
 int ucache_bsize;
 
-#ifdef CONFIG_MAGIC_SYSRQ
-unsigned long SYSRQ_KEY;
-#endif /* CONFIG_MAGIC_SYSRQ */
-
-
 #ifdef CONFIG_SMP
 
 static int smt_enabled_cmdline;
diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c
index 5458ac5da7c3..75fe13815be2 100644
--- a/arch/ppc/kernel/setup.c
+++ b/arch/ppc/kernel/setup.c
@@ -86,10 +86,6 @@ int ppc_do_canonicalize_irqs;
 EXPORT_SYMBOL(ppc_do_canonicalize_irqs);
 #endif
 
-#ifdef CONFIG_MAGIC_SYSRQ
-unsigned long SYSRQ_KEY = 0x54;
-#endif /* CONFIG_MAGIC_SYSRQ */
-
 #ifdef CONFIG_VGA_CONSOLE
 unsigned long vgacon_remap_base;
 #endif
diff --git a/include/asm-h8300/keyboard.h b/include/asm-h8300/keyboard.h
index fbad65e8a5c0..90efbd655390 100644
--- a/include/asm-h8300/keyboard.h
+++ b/include/asm-h8300/keyboard.h
@@ -18,14 +18,6 @@
 #define kbd_enable_irq(x...)	do {;} while (0)
 #define kbd_disable_irq(x...)	do {;} while (0)
 
-
-/* needed if MAGIC_SYSRQ is enabled for serial console */
-#ifndef SYSRQ_KEY
-#define SYSRQ_KEY		((unsigned char)(-1))
-#define kbd_sysrq_xlate         ((unsigned char *)NULL)
-#endif
-
-
 #endif  /* _H8300_KEYBOARD_H */
 
 
diff --git a/include/asm-sh/ec3104/keyboard.h b/include/asm-sh/ec3104/keyboard.h
index 0dee7b05b49d..c1253a683197 100644
--- a/include/asm-sh/ec3104/keyboard.h
+++ b/include/asm-sh/ec3104/keyboard.h
@@ -6,8 +6,6 @@ extern char ec3104_kbd_unexpected_up(unsigned char);
 extern void ec3104_kbd_leds(unsigned char);
 extern void ec3104_kbd_init_hw(void);
 
-#define SYSRQ_KEY 0x54
-
 #define kbd_sysrq_xlate ec3104_kbd_sysrq_xlate
 #define kbd_setkeycode ec3104_kbd_setkeycode
 #define kbd_getkeycode ec3104_kbd_getkeycode
diff --git a/include/asm-sh/mpc1211/keyboard.h b/include/asm-sh/mpc1211/keyboard.h
index 71ef4cf4242d..9020feee7b4c 100644
--- a/include/asm-sh/mpc1211/keyboard.h
+++ b/include/asm-sh/mpc1211/keyboard.h
@@ -24,7 +24,6 @@ extern void pckbd_leds(unsigned char leds);
 extern void pckbd_init_hw(void);
 extern int pckbd_pm_resume(struct pm_dev *, pm_request_t, void *);
 extern pm_callback pm_kbd_request_override;
-extern unsigned char pckbd_sysrq_xlate[128];
 
 #define kbd_setkeycode		pckbd_setkeycode
 #define kbd_getkeycode		pckbd_getkeycode
@@ -32,9 +31,6 @@ extern unsigned char pckbd_sysrq_xlate[128];
 #define kbd_unexpected_up	pckbd_unexpected_up
 #define kbd_leds		pckbd_leds
 #define kbd_init_hw		pckbd_init_hw
-#define kbd_sysrq_xlate		pckbd_sysrq_xlate
-
-#define SYSRQ_KEY 0x54
 
 /* resource allocation */
 #define kbd_request_region()
diff --git a/include/asm-sh64/keyboard.h b/include/asm-sh64/keyboard.h
index 1fab96d792bf..0b01c3beb2f8 100644
--- a/include/asm-sh64/keyboard.h
+++ b/include/asm-sh64/keyboard.h
@@ -30,7 +30,6 @@ extern int pckbd_translate(unsigned char scancode, unsigned char *keycode,
 extern char pckbd_unexpected_up(unsigned char keycode);
 extern void pckbd_leds(unsigned char leds);
 extern void pckbd_init_hw(void);
-extern unsigned char pckbd_sysrq_xlate[128];
 
 #define kbd_setkeycode		pckbd_setkeycode
 #define kbd_getkeycode		pckbd_getkeycode
@@ -38,9 +37,6 @@ extern unsigned char pckbd_sysrq_xlate[128];
 #define kbd_unexpected_up	pckbd_unexpected_up
 #define kbd_leds		pckbd_leds
 #define kbd_init_hw		pckbd_init_hw
-#define kbd_sysrq_xlate		pckbd_sysrq_xlate
-
-#define SYSRQ_KEY 0x54
 
 /* resource allocation */
 #define kbd_request_region()
-- 
cgit v1.2.3


From 89bbc03c01f68e627a2b120963f136e2815f0d84 Mon Sep 17 00:00:00 2001
From: Thomas Petazzoni <thomas.petazzoni@enix.org>
Date: Sat, 30 Sep 2006 23:27:54 -0700
Subject: [PATCH] Prevent multiple inclusion of linux/sysrq.h

Prevent multiple inclusions of include/linux/sysrq.h using traditional
#ifndef..#endif.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@enix.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sysrq.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/sysrq.h b/include/linux/sysrq.h
index 4812ff60561c..e657e523b9bf 100644
--- a/include/linux/sysrq.h
+++ b/include/linux/sysrq.h
@@ -11,6 +11,8 @@
  *	based upon discusions in irc://irc.openprojects.net/#kernelnewbies
  */
 
+#ifndef _LINUX_SYSRQ_H
+#define _LINUX_SYSRQ_H
 
 struct pt_regs;
 struct tty_struct;
@@ -57,3 +59,5 @@ static inline int __reterr(void)
 #define unregister_sysrq_key(ig,nore) __reterr()
 
 #endif
+
+#endif /* _LINUX_SYSRQ_H */
-- 
cgit v1.2.3


From 54f67f631dfc25ca7a8b19200e34013abc974337 Mon Sep 17 00:00:00 2001
From: Petr Vandrovec <petr@vandrovec.name>
Date: Sat, 30 Sep 2006 23:27:55 -0700
Subject: [PATCH] Move ncpfs 32bit compat ioctl to ncpfs

The ncp specific compat ioctls are clearly local to one file system, so the
code can better live there.

This version of the patch moves everything into the generic ioctl handler
and uses it for both 32 and 64 bit calls.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Petr Vandrovec <petr@vandrovec.name>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/compat_ioctl.c            | 198 -----------------------------------
 fs/ncpfs/dir.c               |   3 +
 fs/ncpfs/file.c              |   3 +
 fs/ncpfs/ioctl.c             | 239 +++++++++++++++++++++++++++++++++++++------
 include/linux/compat_ioctl.h |  12 ---
 include/linux/ncp_fs.h       |   1 +
 6 files changed, 217 insertions(+), 239 deletions(-)

(limited to 'include')

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 64b34533edea..27ca1aa30562 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -60,7 +60,6 @@
 #include <linux/ctype.h>
 #include <linux/ioctl32.h>
 #include <linux/syscalls.h>
-#include <linux/ncp_fs.h>
 #include <linux/i2c.h>
 #include <linux/i2c-dev.h>
 #include <linux/wireless.h>
@@ -2348,193 +2347,6 @@ static int rtc_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
 	}
 }
 
-#if defined(CONFIG_NCP_FS) || defined(CONFIG_NCP_FS_MODULE)
-struct ncp_ioctl_request_32 {
-	u32 function;
-	u32 size;
-	compat_caddr_t data;
-};
-
-struct ncp_fs_info_v2_32 {
-	s32 version;
-	u32 mounted_uid;
-	u32 connection;
-	u32 buffer_size;
-
-	u32 volume_number;
-	u32 directory_id;
-
-	u32 dummy1;
-	u32 dummy2;
-	u32 dummy3;
-};
-
-struct ncp_objectname_ioctl_32
-{
-	s32		auth_type;
-	u32		object_name_len;
-	compat_caddr_t	object_name;	/* an userspace data, in most cases user name */
-};
-
-struct ncp_privatedata_ioctl_32
-{
-	u32		len;
-	compat_caddr_t	data;		/* ~1000 for NDS */
-};
-
-#define	NCP_IOC_NCPREQUEST_32		_IOR('n', 1, struct ncp_ioctl_request_32)
-#define NCP_IOC_GETMOUNTUID2_32		_IOW('n', 2, u32)
-#define NCP_IOC_GET_FS_INFO_V2_32	_IOWR('n', 4, struct ncp_fs_info_v2_32)
-#define NCP_IOC_GETOBJECTNAME_32	_IOWR('n', 9, struct ncp_objectname_ioctl_32)
-#define NCP_IOC_SETOBJECTNAME_32	_IOR('n', 9, struct ncp_objectname_ioctl_32)
-#define NCP_IOC_GETPRIVATEDATA_32	_IOWR('n', 10, struct ncp_privatedata_ioctl_32)
-#define NCP_IOC_SETPRIVATEDATA_32	_IOR('n', 10, struct ncp_privatedata_ioctl_32)
-
-static int do_ncp_ncprequest(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct ncp_ioctl_request_32 n32;
-	struct ncp_ioctl_request __user *p = compat_alloc_user_space(sizeof(*p));
-
-	if (copy_from_user(&n32, compat_ptr(arg), sizeof(n32)) ||
-	    put_user(n32.function, &p->function) ||
-	    put_user(n32.size, &p->size) ||
-	    put_user(compat_ptr(n32.data), &p->data))
-		return -EFAULT;
-
-	return sys_ioctl(fd, NCP_IOC_NCPREQUEST, (unsigned long)p);
-}
-
-static int do_ncp_getmountuid2(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	mm_segment_t old_fs = get_fs();
-	__kernel_uid_t kuid;
-	int err;
-
-	cmd = NCP_IOC_GETMOUNTUID2;
-
-	set_fs(KERNEL_DS);
-	err = sys_ioctl(fd, cmd, (unsigned long)&kuid);
-	set_fs(old_fs);
-
-	if (!err)
-		err = put_user(kuid,
-			       (unsigned int __user *) compat_ptr(arg));
-
-	return err;
-}
-
-static int do_ncp_getfsinfo2(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	mm_segment_t old_fs = get_fs();
-	struct ncp_fs_info_v2_32 n32;
-	struct ncp_fs_info_v2 n;
-	int err;
-
-	if (copy_from_user(&n32, compat_ptr(arg), sizeof(n32)))
-		return -EFAULT;
-	if (n32.version != NCP_GET_FS_INFO_VERSION_V2)
-		return -EINVAL;
-	n.version = NCP_GET_FS_INFO_VERSION_V2;
-
-	set_fs(KERNEL_DS);
-	err = sys_ioctl(fd, NCP_IOC_GET_FS_INFO_V2, (unsigned long)&n);
-	set_fs(old_fs);
-
-	if (!err) {
-		n32.version = n.version;
-		n32.mounted_uid = n.mounted_uid;
-		n32.connection = n.connection;
-		n32.buffer_size = n.buffer_size;
-		n32.volume_number = n.volume_number;
-		n32.directory_id = n.directory_id;
-		n32.dummy1 = n.dummy1;
-		n32.dummy2 = n.dummy2;
-		n32.dummy3 = n.dummy3;
-		err = copy_to_user(compat_ptr(arg), &n32, sizeof(n32)) ? -EFAULT : 0;
-	}
-	return err;
-}
-
-static int do_ncp_getobjectname(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct ncp_objectname_ioctl_32 n32, __user *p32 = compat_ptr(arg);
-	struct ncp_objectname_ioctl __user *p = compat_alloc_user_space(sizeof(*p));
-	s32 auth_type;
-	u32 name_len;
-	int err;
-
-	if (copy_from_user(&n32, p32, sizeof(n32)) ||
-	    put_user(n32.object_name_len, &p->object_name_len) ||
-	    put_user(compat_ptr(n32.object_name), &p->object_name))
-		return -EFAULT;
-
-	err = sys_ioctl(fd, NCP_IOC_GETOBJECTNAME, (unsigned long)p);
-        if (err)
-		return err;
-
-	if (get_user(auth_type, &p->auth_type) ||
-	    put_user(auth_type, &p32->auth_type) ||
-	    get_user(name_len, &p->object_name_len) ||
-	    put_user(name_len, &p32->object_name_len))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int do_ncp_setobjectname(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct ncp_objectname_ioctl_32 n32, __user *p32 = compat_ptr(arg);
-	struct ncp_objectname_ioctl __user *p = compat_alloc_user_space(sizeof(*p));
-
-	if (copy_from_user(&n32, p32, sizeof(n32)) ||
-	    put_user(n32.auth_type, &p->auth_type) ||
-	    put_user(n32.object_name_len, &p->object_name_len) ||
-	    put_user(compat_ptr(n32.object_name), &p->object_name))
-		return -EFAULT;
-
-	return sys_ioctl(fd, NCP_IOC_SETOBJECTNAME, (unsigned long)p);
-}
-
-static int do_ncp_getprivatedata(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct ncp_privatedata_ioctl_32 n32, __user *p32 = compat_ptr(arg);
-	struct ncp_privatedata_ioctl __user *p =
-		compat_alloc_user_space(sizeof(*p));
-	u32 len;
-	int err;
-
-	if (copy_from_user(&n32, p32, sizeof(n32)) ||
-	    put_user(n32.len, &p->len) ||
-	    put_user(compat_ptr(n32.data), &p->data))
-		return -EFAULT;
-
-	err = sys_ioctl(fd, NCP_IOC_GETPRIVATEDATA, (unsigned long)p);
-        if (err)
-		return err;
-
-	if (get_user(len, &p->len) ||
-	    put_user(len, &p32->len))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int do_ncp_setprivatedata(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct ncp_privatedata_ioctl_32 n32;
-	struct ncp_privatedata_ioctl_32 __user *p32 = compat_ptr(arg);
-	struct ncp_privatedata_ioctl __user *p =
-		compat_alloc_user_space(sizeof(*p));
-
-	if (copy_from_user(&n32, p32, sizeof(n32)) ||
-	    put_user(n32.len, &p->len) ||
-	    put_user(compat_ptr(n32.data), &p->data))
-		return -EFAULT;
-
-	return sys_ioctl(fd, NCP_IOC_SETPRIVATEDATA, (unsigned long)p);
-}
-#endif
-
 static int
 lp_timeout_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
@@ -2748,16 +2560,6 @@ HANDLE_IOCTL(RTC_IRQP_SET32, rtc_ioctl)
 HANDLE_IOCTL(RTC_EPOCH_READ32, rtc_ioctl)
 HANDLE_IOCTL(RTC_EPOCH_SET32, rtc_ioctl)
 
-#if defined(CONFIG_NCP_FS) || defined(CONFIG_NCP_FS_MODULE)
-HANDLE_IOCTL(NCP_IOC_NCPREQUEST_32, do_ncp_ncprequest)
-HANDLE_IOCTL(NCP_IOC_GETMOUNTUID2_32, do_ncp_getmountuid2)
-HANDLE_IOCTL(NCP_IOC_GET_FS_INFO_V2_32, do_ncp_getfsinfo2)
-HANDLE_IOCTL(NCP_IOC_GETOBJECTNAME_32, do_ncp_getobjectname)
-HANDLE_IOCTL(NCP_IOC_SETOBJECTNAME_32, do_ncp_setobjectname)
-HANDLE_IOCTL(NCP_IOC_GETPRIVATEDATA_32, do_ncp_getprivatedata)
-HANDLE_IOCTL(NCP_IOC_SETPRIVATEDATA_32, do_ncp_setprivatedata)
-#endif
-
 /* dvb */
 HANDLE_IOCTL(VIDEO_GET_EVENT, do_video_get_event)
 HANDLE_IOCTL(VIDEO_STILLPICTURE, do_video_stillpicture)
diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c
index b4ee89250e95..458b3b785194 100644
--- a/fs/ncpfs/dir.c
+++ b/fs/ncpfs/dir.c
@@ -53,6 +53,9 @@ const struct file_operations ncp_dir_operations =
 	.read		= generic_read_dir,
 	.readdir	= ncp_readdir,
 	.ioctl		= ncp_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= ncp_compat_ioctl,
+#endif
 };
 
 struct inode_operations ncp_dir_inode_operations =
diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c
index e6b7c67cf057..df37524b85db 100644
--- a/fs/ncpfs/file.c
+++ b/fs/ncpfs/file.c
@@ -289,6 +289,9 @@ const struct file_operations ncp_file_operations =
 	.read		= ncp_file_read,
 	.write		= ncp_file_write,
 	.ioctl		= ncp_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= ncp_compat_ioctl,
+#endif
 	.mmap		= ncp_mmap,
 	.release	= ncp_release,
 	.fsync		= ncp_fsync,
diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c
index 42039fe0653c..a89ac84a8241 100644
--- a/fs/ncpfs/ioctl.c
+++ b/fs/ncpfs/ioctl.c
@@ -7,19 +7,21 @@
  *
  */
 
-
-#include <asm/uaccess.h>
 #include <linux/capability.h>
+#include <linux/compat.h>
 #include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/ioctl.h>
 #include <linux/time.h>
 #include <linux/mm.h>
 #include <linux/highuid.h>
+#include <linux/smp_lock.h>
 #include <linux/vmalloc.h>
 
 #include <linux/ncp_fs.h>
 
+#include <asm/uaccess.h>
+
 #include "ncplib_kernel.h"
 
 /* maximum limit for ncp_objectname_ioctl */
@@ -89,6 +91,82 @@ ncp_get_fs_info_v2(struct ncp_server * server, struct file *file,
 	return 0;
 }
 
+#ifdef CONFIG_COMPAT
+struct compat_ncp_objectname_ioctl
+{
+	s32		auth_type;
+	u32		object_name_len;
+	compat_caddr_t	object_name;	/* an userspace data, in most cases user name */
+};
+
+struct compat_ncp_fs_info_v2 {
+	s32 version;
+	u32 mounted_uid;
+	u32 connection;
+	u32 buffer_size;
+
+	u32 volume_number;
+	u32 directory_id;
+
+	u32 dummy1;
+	u32 dummy2;
+	u32 dummy3;
+};
+
+struct compat_ncp_ioctl_request {
+	u32 function;
+	u32 size;
+	compat_caddr_t data;
+};
+
+struct compat_ncp_privatedata_ioctl
+{
+	u32		len;
+	compat_caddr_t	data;		/* ~1000 for NDS */
+};
+
+#define NCP_IOC_GET_FS_INFO_V2_32	_IOWR('n', 4, struct compat_ncp_fs_info_v2)
+#define NCP_IOC_NCPREQUEST_32		_IOR('n', 1, struct compat_ncp_ioctl_request)
+#define NCP_IOC_GETOBJECTNAME_32	_IOWR('n', 9, struct compat_ncp_objectname_ioctl)
+#define NCP_IOC_SETOBJECTNAME_32	_IOR('n', 9, struct compat_ncp_objectname_ioctl)
+#define NCP_IOC_GETPRIVATEDATA_32	_IOWR('n', 10, struct compat_ncp_privatedata_ioctl)
+#define NCP_IOC_SETPRIVATEDATA_32	_IOR('n', 10, struct compat_ncp_privatedata_ioctl)
+
+static int
+ncp_get_compat_fs_info_v2(struct ncp_server * server, struct file *file,
+		   struct compat_ncp_fs_info_v2 __user * arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct compat_ncp_fs_info_v2 info2;
+
+	if ((file_permission(file, MAY_WRITE) != 0)
+	    && (current->uid != server->m.mounted_uid)) {
+		return -EACCES;
+	}
+	if (copy_from_user(&info2, arg, sizeof(info2)))
+		return -EFAULT;
+
+	if (info2.version != NCP_GET_FS_INFO_VERSION_V2) {
+		DPRINTK("info.version invalid: %d\n", info2.version);
+		return -EINVAL;
+	}
+	info2.mounted_uid   = server->m.mounted_uid;
+	info2.connection    = server->connection;
+	info2.buffer_size   = server->buffer_size;
+	info2.volume_number = NCP_FINFO(inode)->volNumber;
+	info2.directory_id  = NCP_FINFO(inode)->DosDirNum;
+	info2.dummy1 = info2.dummy2 = info2.dummy3 = 0;
+
+	if (copy_to_user(arg, &info2, sizeof(info2)))
+		return -EFAULT;
+	return 0;
+}
+#endif
+
+#define NCP_IOC_GETMOUNTUID16		_IOW('n', 2, u16)
+#define NCP_IOC_GETMOUNTUID32		_IOW('n', 2, u32)
+#define NCP_IOC_GETMOUNTUID64		_IOW('n', 2, u64)
+
 #ifdef CONFIG_NCPFS_NLS
 /* Here we are select the iocharset and the codepage for NLS.
  * Thanks Petr Vandrovec for idea and many hints.
@@ -192,12 +270,24 @@ int ncp_ioctl(struct inode *inode, struct file *filp,
 	void __user *argp = (void __user *)arg;
 
 	switch (cmd) {
+#ifdef CONFIG_COMPAT
+	case NCP_IOC_NCPREQUEST_32:
+#endif
 	case NCP_IOC_NCPREQUEST:
-
 		if ((file_permission(filp, MAY_WRITE) != 0)
 		    && (current->uid != server->m.mounted_uid)) {
 			return -EACCES;
 		}
+#ifdef CONFIG_COMPAT
+		if (cmd == NCP_IOC_NCPREQUEST_32) {
+			struct compat_ncp_ioctl_request request32;
+			if (copy_from_user(&request32, argp, sizeof(request32)))
+				return -EFAULT;
+			request.function = request32.function;
+			request.size = request32.size;
+			request.data = compat_ptr(request32.data);
+		} else
+#endif
 		if (copy_from_user(&request, argp, sizeof(request)))
 			return -EFAULT;
 
@@ -254,19 +344,35 @@ int ncp_ioctl(struct inode *inode, struct file *filp,
 	case NCP_IOC_GET_FS_INFO_V2:
 		return ncp_get_fs_info_v2(server, filp, argp);
 
-	case NCP_IOC_GETMOUNTUID2:
-		{
-			unsigned long tmp = server->m.mounted_uid;
-
-			if ((file_permission(filp, MAY_READ) != 0)
-			    && (current->uid != server->m.mounted_uid))
-			{
-				return -EACCES;
-			}
-			if (put_user(tmp, (unsigned long __user *)argp)) 
+#ifdef CONFIG_COMPAT
+	case NCP_IOC_GET_FS_INFO_V2_32:
+		return ncp_get_compat_fs_info_v2(server, filp, argp);
+#endif
+	/* we have too many combinations of CONFIG_COMPAT,
+	 * CONFIG_64BIT and CONFIG_UID16, so just handle
+	 * any of the possible ioctls */
+	case NCP_IOC_GETMOUNTUID16:
+	case NCP_IOC_GETMOUNTUID32:
+	case NCP_IOC_GETMOUNTUID64:
+		if ((file_permission(filp, MAY_READ) != 0)
+			&& (current->uid != server->m.mounted_uid)) {
+			return -EACCES;
+		}
+		if (cmd == NCP_IOC_GETMOUNTUID16) {
+			u16 uid;
+			SET_UID(uid, server->m.mounted_uid);
+			if (put_user(uid, (u16 __user *)argp))
+				return -EFAULT;
+		} else if (cmd == NCP_IOC_GETMOUNTUID32) {
+			if (put_user(server->m.mounted_uid,
+						(u32 __user *)argp))
+				return -EFAULT;
+		} else {
+			if (put_user(server->m.mounted_uid,
+						(u64 __user *)argp))
 				return -EFAULT;
-			return 0;
 		}
+		return 0;
 
 	case NCP_IOC_GETROOT:
 		{
@@ -476,6 +582,32 @@ outrel:
 		}
 #endif	/* CONFIG_NCPFS_IOCTL_LOCKING */
 
+#ifdef CONFIG_COMPAT
+	case NCP_IOC_GETOBJECTNAME_32:
+		if (current->uid != server->m.mounted_uid) {
+			return -EACCES;
+		}
+		{
+			struct compat_ncp_objectname_ioctl user;
+			size_t outl;
+
+			if (copy_from_user(&user, argp, sizeof(user)))
+				return -EFAULT;
+			user.auth_type = server->auth.auth_type;
+			outl = user.object_name_len;
+			user.object_name_len = server->auth.object_name_len;
+			if (outl > user.object_name_len)
+				outl = user.object_name_len;
+			if (outl) {
+				if (copy_to_user(compat_ptr(user.object_name),
+						 server->auth.object_name,
+						 outl)) return -EFAULT;
+			}
+			if (copy_to_user(argp, &user, sizeof(user)))
+				return -EFAULT;
+			return 0;
+		}
+#endif
 	case NCP_IOC_GETOBJECTNAME:
 		if (current->uid != server->m.mounted_uid) {
 			return -EACCES;
@@ -500,6 +632,9 @@ outrel:
 				return -EFAULT;
 			return 0;
 		}
+#ifdef CONFIG_COMPAT
+	case NCP_IOC_SETOBJECTNAME_32:
+#endif
 	case NCP_IOC_SETOBJECTNAME:
 		if (current->uid != server->m.mounted_uid) {
 			return -EACCES;
@@ -512,8 +647,19 @@ outrel:
 			void* oldprivate;
 			size_t oldprivatelen;
 
+#ifdef CONFIG_COMPAT
+			if (cmd == NCP_IOC_SETOBJECTNAME_32) {
+				struct compat_ncp_objectname_ioctl user32;
+				if (copy_from_user(&user32, argp, sizeof(user32)))
+					return -EFAULT;
+				user.auth_type = user32.auth_type;
+				user.object_name_len = user32.object_name_len;
+				user.object_name = compat_ptr(user32.object_name);
+			} else
+#endif
 			if (copy_from_user(&user, argp, sizeof(user)))
 				return -EFAULT;
+
 			if (user.object_name_len > NCP_OBJECT_NAME_MAX_LEN)
 				return -ENOMEM;
 			if (user.object_name_len) {
@@ -544,6 +690,9 @@ outrel:
 			kfree(oldname);
 			return 0;
 		}
+#ifdef CONFIG_COMPAT
+	case NCP_IOC_GETPRIVATEDATA_32:
+#endif
 	case NCP_IOC_GETPRIVATEDATA:
 		if (current->uid != server->m.mounted_uid) {
 			return -EACCES;
@@ -552,8 +701,18 @@ outrel:
 			struct ncp_privatedata_ioctl user;
 			size_t outl;
 
+#ifdef CONFIG_COMPAT
+			if (cmd == NCP_IOC_GETPRIVATEDATA_32) {
+				struct compat_ncp_privatedata_ioctl user32;
+				if (copy_from_user(&user32, argp, sizeof(user32)))
+					return -EFAULT;
+				user.len = user32.len;
+				user.data = compat_ptr(user32.data);
+			} else
+#endif
 			if (copy_from_user(&user, argp, sizeof(user)))
 				return -EFAULT;
+
 			outl = user.len;
 			user.len = server->priv.len;
 			if (outl > user.len) outl = user.len;
@@ -562,10 +721,23 @@ outrel:
 						 server->priv.data,
 						 outl)) return -EFAULT;
 			}
+#ifdef CONFIG_COMPAT
+			if (cmd == NCP_IOC_GETPRIVATEDATA_32) {
+				struct compat_ncp_privatedata_ioctl user32;
+				user32.len = user.len;
+				user32.data = (unsigned long) user.data;
+				if (copy_to_user(&user32, argp, sizeof(user32)))
+					return -EFAULT;
+			} else
+#endif
 			if (copy_to_user(argp, &user, sizeof(user)))
 				return -EFAULT;
+
 			return 0;
 		}
+#ifdef CONFIG_COMPAT
+	case NCP_IOC_SETPRIVATEDATA_32:
+#endif
 	case NCP_IOC_SETPRIVATEDATA:
 		if (current->uid != server->m.mounted_uid) {
 			return -EACCES;
@@ -576,8 +748,18 @@ outrel:
 			void* old;
 			size_t oldlen;
 
+#ifdef CONFIG_COMPAT
+			if (cmd == NCP_IOC_SETPRIVATEDATA_32) {
+				struct compat_ncp_privatedata_ioctl user32;
+				if (copy_from_user(&user32, argp, sizeof(user32)))
+					return -EFAULT;
+				user.len = user32.len;
+				user.data = compat_ptr(user32.data);
+			} else
+#endif
 			if (copy_from_user(&user, argp, sizeof(user)))
 				return -EFAULT;
+
 			if (user.len > NCP_PRIVATE_DATA_MAX_LEN)
 				return -ENOMEM;
 			if (user.len) {
@@ -636,20 +818,19 @@ outrel:
 		}
 
 	}
-/* #ifdef CONFIG_UID16 */
-	/* NCP_IOC_GETMOUNTUID may be same as NCP_IOC_GETMOUNTUID2,
-           so we have this out of switch */
-	if (cmd == NCP_IOC_GETMOUNTUID) {
-		__kernel_uid_t uid = 0;
-		if ((file_permission(filp, MAY_READ) != 0)
-		    && (current->uid != server->m.mounted_uid)) {
-			return -EACCES;
-		}
-		SET_UID(uid, server->m.mounted_uid);
-		if (put_user(uid, (__kernel_uid_t __user *)argp))
-			return -EFAULT;
-		return 0;
-	}
-/* #endif */
 	return -EINVAL;
 }
+
+#ifdef CONFIG_COMPAT
+long ncp_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	int ret;
+
+	lock_kernel();
+	arg = (unsigned long) compat_ptr(arg);
+	ret = ncp_ioctl(inode, file, cmd, arg);
+	unlock_kernel();
+	return ret;
+}
+#endif
diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h
index d61ef5951538..d5b7abc4f409 100644
--- a/include/linux/compat_ioctl.h
+++ b/include/linux/compat_ioctl.h
@@ -569,18 +569,6 @@ COMPATIBLE_IOCTL(RAW_SETBIND)
 COMPATIBLE_IOCTL(RAW_GETBIND)
 /* SMB ioctls which do not need any translations */
 COMPATIBLE_IOCTL(SMB_IOC_NEWCONN)
-/* NCP ioctls which do not need any translations */
-COMPATIBLE_IOCTL(NCP_IOC_CONN_LOGGED_IN)
-COMPATIBLE_IOCTL(NCP_IOC_SIGN_INIT)
-COMPATIBLE_IOCTL(NCP_IOC_SIGN_WANTED)
-COMPATIBLE_IOCTL(NCP_IOC_SET_SIGN_WANTED)
-COMPATIBLE_IOCTL(NCP_IOC_LOCKUNLOCK)
-COMPATIBLE_IOCTL(NCP_IOC_GETROOT)
-COMPATIBLE_IOCTL(NCP_IOC_SETROOT)
-COMPATIBLE_IOCTL(NCP_IOC_GETCHARSETS)
-COMPATIBLE_IOCTL(NCP_IOC_SETCHARSETS)
-COMPATIBLE_IOCTL(NCP_IOC_GETDENTRYTTL)
-COMPATIBLE_IOCTL(NCP_IOC_SETDENTRYTTL)
 /* Little a */
 COMPATIBLE_IOCTL(ATMSIGD_CTRL)
 COMPATIBLE_IOCTL(ATMARPD_CTRL)
diff --git a/include/linux/ncp_fs.h b/include/linux/ncp_fs.h
index 02e352be717e..0ea7f89e613c 100644
--- a/include/linux/ncp_fs.h
+++ b/include/linux/ncp_fs.h
@@ -212,6 +212,7 @@ void ncp_date_unix2dos(int unix_date, __le16 * time, __le16 * date);
 
 /* linux/fs/ncpfs/ioctl.c */
 int ncp_ioctl(struct inode *, struct file *, unsigned int, unsigned long);
+long ncp_compat_ioctl(struct file *, unsigned int, unsigned long);
 
 /* linux/fs/ncpfs/sock.c */
 int ncp_request2(struct ncp_server *server, int function,
-- 
cgit v1.2.3


From c69c31270c35a6b8421a8e4ba81de1247ac6df95 Mon Sep 17 00:00:00 2001
From: Corey Minyard <minyard@acm.org>
Date: Sat, 30 Sep 2006 23:27:56 -0700
Subject: [PATCH] IPMI: per-channel command registration

This patch adds the ability to register for a command per-channel in the
IPMI driver.

If your BMC supports multiple channels, incoming messages can be useful to
have the ability to register to receive commands on a specific channel
instead the current behaviour of all channels.

Signed-off-by: David Barksdale <amatus@ocgnet.org>
Signed-off-by: Corey Minyard <minyard@acm.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/IPMI.txt              |  9 +++--
 drivers/char/ipmi/ipmi_devintf.c    | 34 ++++++++++++++++-
 drivers/char/ipmi/ipmi_msghandler.c | 75 +++++++++++++++++++++++++++----------
 include/linux/ipmi.h                | 48 ++++++++++++++++++++++--
 4 files changed, 138 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/Documentation/IPMI.txt b/Documentation/IPMI.txt
index 0256805b548f..7756e09ea759 100644
--- a/Documentation/IPMI.txt
+++ b/Documentation/IPMI.txt
@@ -326,9 +326,12 @@ for events, they will all receive all events that come in.
 
 For receiving commands, you have to individually register commands you
 want to receive.  Call ipmi_register_for_cmd() and supply the netfn
-and command name for each command you want to receive.  Only one user
-may be registered for each netfn/cmd, but different users may register
-for different commands.
+and command name for each command you want to receive.  You also
+specify a bitmask of the channels you want to receive the command from
+(or use IPMI_CHAN_ALL for all channels if you don't care).  Only one
+user may be registered for each netfn/cmd/channel, but different users
+may register for different commands, or the same command if the
+channel bitmasks do not overlap.
 
 From userland, equivalent IOCTLs are provided to do these functions.
 
diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c
index 68d7c61a864e..81fcf0ce21d1 100644
--- a/drivers/char/ipmi/ipmi_devintf.c
+++ b/drivers/char/ipmi/ipmi_devintf.c
@@ -377,7 +377,8 @@ static int ipmi_ioctl(struct inode  *inode,
 			break;
 		}
 
-		rv = ipmi_register_for_cmd(priv->user, val.netfn, val.cmd);
+		rv = ipmi_register_for_cmd(priv->user, val.netfn, val.cmd,
+					   IPMI_CHAN_ALL);
 		break;
 	}
 
@@ -390,7 +391,36 @@ static int ipmi_ioctl(struct inode  *inode,
 			break;
 		}
 
-		rv = ipmi_unregister_for_cmd(priv->user, val.netfn, val.cmd);
+		rv = ipmi_unregister_for_cmd(priv->user, val.netfn, val.cmd,
+					     IPMI_CHAN_ALL);
+		break;
+	}
+
+	case IPMICTL_REGISTER_FOR_CMD_CHANS:
+	{
+		struct ipmi_cmdspec_chans val;
+
+		if (copy_from_user(&val, arg, sizeof(val))) {
+			rv = -EFAULT;
+			break;
+		}
+
+		rv = ipmi_register_for_cmd(priv->user, val.netfn, val.cmd,
+					   val.chans);
+		break;
+	}
+
+	case IPMICTL_UNREGISTER_FOR_CMD_CHANS:
+	{
+		struct ipmi_cmdspec_chans val;
+
+		if (copy_from_user(&val, arg, sizeof(val))) {
+			rv = -EFAULT;
+			break;
+		}
+
+		rv = ipmi_unregister_for_cmd(priv->user, val.netfn, val.cmd,
+					     val.chans);
 		break;
 	}
 
diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c
index 843d34c8627c..2455e8d478ac 100644
--- a/drivers/char/ipmi/ipmi_msghandler.c
+++ b/drivers/char/ipmi/ipmi_msghandler.c
@@ -96,6 +96,7 @@ struct cmd_rcvr
 	ipmi_user_t   user;
 	unsigned char netfn;
 	unsigned char cmd;
+	unsigned int  chans;
 
 	/*
 	 * This is used to form a linked lised during mass deletion.
@@ -953,24 +954,41 @@ int ipmi_set_gets_events(ipmi_user_t user, int val)
 
 static struct cmd_rcvr *find_cmd_rcvr(ipmi_smi_t    intf,
 				      unsigned char netfn,
-				      unsigned char cmd)
+				      unsigned char cmd,
+				      unsigned char chan)
 {
 	struct cmd_rcvr *rcvr;
 
 	list_for_each_entry_rcu(rcvr, &intf->cmd_rcvrs, link) {
-		if ((rcvr->netfn == netfn) && (rcvr->cmd == cmd))
+		if ((rcvr->netfn == netfn) && (rcvr->cmd == cmd)
+					&& (rcvr->chans & (1 << chan)))
 			return rcvr;
 	}
 	return NULL;
 }
 
+static int is_cmd_rcvr_exclusive(ipmi_smi_t    intf,
+				 unsigned char netfn,
+				 unsigned char cmd,
+				 unsigned int  chans)
+{
+	struct cmd_rcvr *rcvr;
+
+	list_for_each_entry_rcu(rcvr, &intf->cmd_rcvrs, link) {
+		if ((rcvr->netfn == netfn) && (rcvr->cmd == cmd)
+					&& (rcvr->chans & chans))
+			return 0;
+	}
+	return 1;
+}
+
 int ipmi_register_for_cmd(ipmi_user_t   user,
 			  unsigned char netfn,
-			  unsigned char cmd)
+			  unsigned char cmd,
+			  unsigned int  chans)
 {
 	ipmi_smi_t      intf = user->intf;
 	struct cmd_rcvr *rcvr;
-	struct cmd_rcvr *entry;
 	int             rv = 0;
 
 
@@ -979,12 +997,12 @@ int ipmi_register_for_cmd(ipmi_user_t   user,
 		return -ENOMEM;
 	rcvr->cmd = cmd;
 	rcvr->netfn = netfn;
+	rcvr->chans = chans;
 	rcvr->user = user;
 
 	mutex_lock(&intf->cmd_rcvrs_mutex);
 	/* Make sure the command/netfn is not already registered. */
-	entry = find_cmd_rcvr(intf, netfn, cmd);
-	if (entry) {
+	if (!is_cmd_rcvr_exclusive(intf, netfn, cmd, chans)) {
 		rv = -EBUSY;
 		goto out_unlock;
 	}
@@ -1001,24 +1019,39 @@ int ipmi_register_for_cmd(ipmi_user_t   user,
 
 int ipmi_unregister_for_cmd(ipmi_user_t   user,
 			    unsigned char netfn,
-			    unsigned char cmd)
+			    unsigned char cmd,
+			    unsigned int  chans)
 {
 	ipmi_smi_t      intf = user->intf;
 	struct cmd_rcvr *rcvr;
+	struct cmd_rcvr *rcvrs = NULL;
+	int i, rv = -ENOENT;
 
 	mutex_lock(&intf->cmd_rcvrs_mutex);
-	/* Make sure the command/netfn is not already registered. */
-	rcvr = find_cmd_rcvr(intf, netfn, cmd);
-	if ((rcvr) && (rcvr->user == user)) {
-		list_del_rcu(&rcvr->link);
-		mutex_unlock(&intf->cmd_rcvrs_mutex);
-		synchronize_rcu();
+	for (i = 0; i < IPMI_NUM_CHANNELS; i++) {
+		if (((1 << i) & chans) == 0)
+			continue;
+		rcvr = find_cmd_rcvr(intf, netfn, cmd, i);
+		if (rcvr == NULL)
+			continue;
+		if (rcvr->user == user) {
+			rv = 0;
+			rcvr->chans &= ~chans;
+			if (rcvr->chans == 0) {
+				list_del_rcu(&rcvr->link);
+				rcvr->next = rcvrs;
+				rcvrs = rcvr;
+			}
+		}
+	}
+	mutex_unlock(&intf->cmd_rcvrs_mutex);
+	synchronize_rcu();
+	while (rcvrs) {
+		rcvr = rcvrs;
+		rcvrs = rcvr->next;
 		kfree(rcvr);
-		return 0;
-	} else {
-		mutex_unlock(&intf->cmd_rcvrs_mutex);
-		return -ENOENT;
 	}
+	return rv;
 }
 
 void ipmi_user_set_run_to_completion(ipmi_user_t user, int val)
@@ -2548,6 +2581,7 @@ static int handle_ipmb_get_msg_cmd(ipmi_smi_t          intf,
 	int                      rv = 0;
 	unsigned char            netfn;
 	unsigned char            cmd;
+	unsigned char            chan;
 	ipmi_user_t              user = NULL;
 	struct ipmi_ipmb_addr    *ipmb_addr;
 	struct ipmi_recv_msg     *recv_msg;
@@ -2568,9 +2602,10 @@ static int handle_ipmb_get_msg_cmd(ipmi_smi_t          intf,
 
 	netfn = msg->rsp[4] >> 2;
 	cmd = msg->rsp[8];
+	chan = msg->rsp[3] & 0xf;
 
 	rcu_read_lock();
-	rcvr = find_cmd_rcvr(intf, netfn, cmd);
+	rcvr = find_cmd_rcvr(intf, netfn, cmd, chan);
 	if (rcvr) {
 		user = rcvr->user;
 		kref_get(&user->refcount);
@@ -2728,6 +2763,7 @@ static int handle_lan_get_msg_cmd(ipmi_smi_t          intf,
 	int                      rv = 0;
 	unsigned char            netfn;
 	unsigned char            cmd;
+	unsigned char            chan;
 	ipmi_user_t              user = NULL;
 	struct ipmi_lan_addr     *lan_addr;
 	struct ipmi_recv_msg     *recv_msg;
@@ -2748,9 +2784,10 @@ static int handle_lan_get_msg_cmd(ipmi_smi_t          intf,
 
 	netfn = msg->rsp[6] >> 2;
 	cmd = msg->rsp[10];
+	chan = msg->rsp[3] & 0xf;
 
 	rcu_read_lock();
-	rcvr = find_cmd_rcvr(intf, netfn, cmd);
+	rcvr = find_cmd_rcvr(intf, netfn, cmd, chan);
 	if (rcvr) {
 		user = rcvr->user;
 		kref_get(&user->refcount);
diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h
index d09fbeabf1dc..796ca009fd46 100644
--- a/include/linux/ipmi.h
+++ b/include/linux/ipmi.h
@@ -148,6 +148,13 @@ struct ipmi_lan_addr
 #define IPMI_BMC_CHANNEL  0xf
 #define IPMI_NUM_CHANNELS 0x10
 
+/*
+ * Used to signify an "all channel" bitmask.  This is more than the
+ * actual number of channels because this is used in userland and
+ * will cover us if the number of channels is extended.
+ */
+#define IPMI_CHAN_ALL     (~0)
+
 
 /*
  * A raw IPMI message without any addressing.  This covers both
@@ -350,18 +357,21 @@ int ipmi_request_supply_msgs(ipmi_user_t          user,
 
 /*
  * When commands come in to the SMS, the user can register to receive
- * them.  Only one user can be listening on a specific netfn/cmd pair
+ * them.  Only one user can be listening on a specific netfn/cmd/chan tuple
  * at a time, you will get an EBUSY error if the command is already
  * registered.  If a command is received that does not have a user
  * registered, the driver will automatically return the proper
- * error.
+ * error.  Channels are specified as a bitfield, use IPMI_CHAN_ALL to
+ * mean all channels.
  */
 int ipmi_register_for_cmd(ipmi_user_t   user,
 			  unsigned char netfn,
-			  unsigned char cmd);
+			  unsigned char cmd,
+			  unsigned int  chans);
 int ipmi_unregister_for_cmd(ipmi_user_t   user,
 			    unsigned char netfn,
-			    unsigned char cmd);
+			    unsigned char cmd,
+			    unsigned int  chans);
 
 /*
  * Allow run-to-completion mode to be set for the interface of
@@ -571,6 +581,36 @@ struct ipmi_cmdspec
 #define IPMICTL_UNREGISTER_FOR_CMD	_IOR(IPMI_IOC_MAGIC, 15,	\
 					     struct ipmi_cmdspec)
 
+/*
+ * Register to get commands from other entities on specific channels.
+ * This way, you can only listen on specific channels, or have messages
+ * from some channels go to one place and other channels to someplace
+ * else.  The chans field is a bitmask, (1 << channel) for each channel.
+ * It may be IPMI_CHAN_ALL for all channels.
+ */
+struct ipmi_cmdspec_chans
+{
+	unsigned int netfn;
+	unsigned int cmd;
+	unsigned int chans;
+};
+
+/*
+ * Register to receive a specific command on specific channels.  error values:
+ *   - EFAULT - an address supplied was invalid.
+ *   - EBUSY - One of the netfn/cmd/chans supplied was already in use.
+ *   - ENOMEM - could not allocate memory for the entry.
+ */
+#define IPMICTL_REGISTER_FOR_CMD_CHANS	_IOR(IPMI_IOC_MAGIC, 28,	\
+					     struct ipmi_cmdspec_chans)
+/*
+ * Unregister some netfn/cmd/chans.  error values:
+ *  - EFAULT - an address supplied was invalid.
+ *  - ENOENT - None of the netfn/cmd/chans were found registered for this user.
+ */
+#define IPMICTL_UNREGISTER_FOR_CMD_CHANS _IOR(IPMI_IOC_MAGIC, 29,	\
+					     struct ipmi_cmdspec_chans)
+
 /* 
  * Set whether this interface receives events.  Note that the first
  * user registered for events will get all pending events for the
-- 
cgit v1.2.3


From 30cbc22217bb3d5d4c74c88127fbf595460bdb76 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olaf@aepfle.de>
Date: Sat, 30 Sep 2006 23:27:57 -0700
Subject: [PATCH] update legacy io handling for pmac

ppc can boot one single binary on prep, chrp and pmac boards.  ppc64 can
boot one single binary on pseries and G5 boards.  pmac has no legacy io,
probing for PC style legacy hardware (or accessing the legacy io area
regulary) may lead to a hard crash:

* add check for parport_pc, exit on pmac.  32bit chrp has no
  ->check_legacy_ioport, the probe is always called.  64bit chrp has
  check_legacy_ioport, check for a "parallel" node

* add check for isapnp, only PReP boards may have real ISA slots.  32bit
  PReP will have no ->check_legacy_ioport, the probe is always called.

* update code in i8042_platform_init.  Run ->check_legacy_ioport first,
  always call request_region.  No functional change.  Remove whitespace
  before i8042_reset init.

Signed-off-by: Olaf Hering <olaf@aepfle.de>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Adam Belay <ambx1@neo.rr.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/platforms/pseries/setup.c |  6 ++++++
 drivers/input/serio/i8042-io.h         | 15 ++++++---------
 drivers/parport/parport_pc.c           |  4 ++++
 drivers/pnp/pnpbios/core.c             |  8 ++++++++
 include/asm-powerpc/io.h               |  2 ++
 5 files changed, 26 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index a6398fbe530d..8ed362140452 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -415,6 +415,12 @@ static int pSeries_check_legacy_ioport(unsigned int baseport)
 			return -ENODEV;
 		of_node_put(np);
 		break;
+	case PARALLEL_BASE:
+		np = of_find_node_by_type(NULL, "parallel");
+		if (np == NULL)
+			return -ENODEV;
+		of_node_put(np);
+		break;
 	}
 	return 0;
 }
diff --git a/drivers/input/serio/i8042-io.h b/drivers/input/serio/i8042-io.h
index cc21914fbc72..3b4e13b9ce1b 100644
--- a/drivers/input/serio/i8042-io.h
+++ b/drivers/input/serio/i8042-io.h
@@ -67,25 +67,22 @@ static inline int i8042_platform_init(void)
  * On some platforms touching the i8042 data register region can do really
  * bad things. Because of this the region is always reserved on such boxes.
  */
-#if !defined(__sh__) && !defined(__alpha__) && !defined(__mips__) && !defined(CONFIG_PPC_MERGE)
-	if (!request_region(I8042_DATA_REG, 16, "i8042"))
-		return -EBUSY;
-#endif
-
-        i8042_reset = 1;
-
 #if defined(CONFIG_PPC_MERGE)
 	if (check_legacy_ioport(I8042_DATA_REG))
-		return -EBUSY;
+		return -ENODEV;
+#endif
+#if !defined(__sh__) && !defined(__alpha__) && !defined(__mips__)
 	if (!request_region(I8042_DATA_REG, 16, "i8042"))
 		return -EBUSY;
 #endif
+
+	i8042_reset = 1;
 	return 0;
 }
 
 static inline void i8042_platform_exit(void)
 {
-#if !defined(__sh__) && !defined(__alpha__) && !defined(CONFIG_PPC64)
+#if !defined(__sh__) && !defined(__alpha__)
 	release_region(I8042_DATA_REG, 16);
 #endif
 }
diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c
index fe800dc0be9f..43894ddec7dc 100644
--- a/drivers/parport/parport_pc.c
+++ b/drivers/parport/parport_pc.c
@@ -3374,6 +3374,10 @@ __setup("parport_init_mode=",parport_init_mode_setup);
 
 static int __init parport_pc_init(void)
 {
+#if defined(CONFIG_PPC_MERGE)
+	if (check_legacy_ioport(PARALLEL_BASE))
+		return -ENODEV;
+#endif
 	if (parse_parport_params())
 		return -EINVAL;
 
diff --git a/drivers/pnp/pnpbios/core.c b/drivers/pnp/pnpbios/core.c
index 551f58e29810..81a6c83d89a6 100644
--- a/drivers/pnp/pnpbios/core.c
+++ b/drivers/pnp/pnpbios/core.c
@@ -526,6 +526,10 @@ static int __init pnpbios_init(void)
 {
 	int ret;
 
+#if defined(CONFIG_PPC_MERGE)
+	if (check_legacy_ioport(PNPBIOS_BASE))
+		return -ENODEV;
+#endif
 	if (pnpbios_disabled || dmi_check_system(pnpbios_dmi_table)) {
 		printk(KERN_INFO "PnPBIOS: Disabled\n");
 		return -ENODEV;
@@ -575,6 +579,10 @@ subsys_initcall(pnpbios_init);
 
 static int __init pnpbios_thread_init(void)
 {
+#if defined(CONFIG_PPC_MERGE)
+	if (check_legacy_ioport(PNPBIOS_BASE))
+		return 0;
+#endif
 	if (pnpbios_disabled)
 		return 0;
 #ifdef CONFIG_HOTPLUG
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 46bae1cf385b..19b2ec1ec665 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -11,6 +11,8 @@
 
 /* Check of existence of legacy devices */
 extern int check_legacy_ioport(unsigned long base_port);
+#define PARALLEL_BASE	0x378
+#define PNPBIOS_BASE	0xf000	/* only relevant for PReP */
 
 #ifndef CONFIG_PPC64
 #include <asm-ppc/io.h>
-- 
cgit v1.2.3


From 3a2711116073db258224afd2cc0f478bdf305575 Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Sat, 30 Sep 2006 23:28:09 -0700
Subject: [PATCH] Remove BUG_ON(unlikely) in include/linux/aio.h

BUG_ON() does this unlikely check itself, as bugs in Linux are unlikely
anyway :)

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Acked-by: Zach Brown <zach.brown@oracle.com>
Acked-by: Benjamin LaHaise <benjamin.c.lahaise@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/aio.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/aio.h b/include/linux/aio.h
index 00c8efa95cc3..8a0193385a9b 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -213,11 +213,11 @@ int FASTCALL(io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
 				  struct iocb *iocb));
 
 #define get_ioctx(kioctx) do {						\
-	BUG_ON(unlikely(atomic_read(&(kioctx)->users) <= 0));		\
+	BUG_ON(atomic_read(&(kioctx)->users) <= 0);			\
 	atomic_inc(&(kioctx)->users);					\
 } while (0)
 #define put_ioctx(kioctx) do {						\
-	BUG_ON(unlikely(atomic_read(&(kioctx)->users) <= 0));		\
+	BUG_ON(atomic_read(&(kioctx)->users) <= 0);			\
 	if (unlikely(atomic_dec_and_test(&(kioctx)->users))) 		\
 		__put_ioctx(kioctx);					\
 } while (0)
-- 
cgit v1.2.3


From ff8371ac9a5a55c956991fed8e5f58640c7a32f3 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sat, 30 Sep 2006 23:28:17 -0700
Subject: [PATCH] constify rtc_class_ops: update drivers

Update RTC framework so that drivers can constify their method tables, moving
them from ".data" to ".rodata".  Then update the drivers.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/rtc/class.c       | 2 +-
 drivers/rtc/rtc-at91.c    | 2 +-
 drivers/rtc/rtc-dev.c     | 4 ++--
 drivers/rtc/rtc-ds1307.c  | 2 +-
 drivers/rtc/rtc-ds1553.c  | 2 +-
 drivers/rtc/rtc-ds1672.c  | 2 +-
 drivers/rtc/rtc-ds1742.c  | 2 +-
 drivers/rtc/rtc-ep93xx.c  | 2 +-
 drivers/rtc/rtc-isl1208.c | 2 +-
 drivers/rtc/rtc-m48t86.c  | 2 +-
 drivers/rtc/rtc-max6902.c | 2 +-
 drivers/rtc/rtc-pcf8563.c | 2 +-
 drivers/rtc/rtc-pcf8583.c | 2 +-
 drivers/rtc/rtc-pl031.c   | 2 +-
 drivers/rtc/rtc-proc.c    | 2 +-
 drivers/rtc/rtc-rs5c348.c | 2 +-
 drivers/rtc/rtc-rs5c372.c | 2 +-
 drivers/rtc/rtc-s3c.c     | 2 +-
 drivers/rtc/rtc-sa1100.c  | 2 +-
 drivers/rtc/rtc-test.c    | 2 +-
 drivers/rtc/rtc-v3020.c   | 2 +-
 drivers/rtc/rtc-vr41xx.c  | 2 +-
 drivers/rtc/rtc-x1205.c   | 2 +-
 include/linux/rtc.h       | 4 ++--
 24 files changed, 26 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index 306d600a764a..7a0d8ee2de9c 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -39,7 +39,7 @@ static void rtc_device_release(struct class_device *class_dev)
  * Returns the pointer to the new struct class device.
  */
 struct rtc_device *rtc_device_register(const char *name, struct device *dev,
-					struct rtc_class_ops *ops,
+					const struct rtc_class_ops *ops,
 					struct module *owner)
 {
 	struct rtc_device *rtc;
diff --git a/drivers/rtc/rtc-at91.c b/drivers/rtc/rtc-at91.c
index dfd0ce86f6a0..3cf3529888c7 100644
--- a/drivers/rtc/rtc-at91.c
+++ b/drivers/rtc/rtc-at91.c
@@ -267,7 +267,7 @@ static irqreturn_t at91_rtc_interrupt(int irq, void *dev_id,
 	return IRQ_NONE;		/* not handled */
 }
 
-static struct rtc_class_ops at91_rtc_ops = {
+static const struct rtc_class_ops at91_rtc_ops = {
 	.ioctl		= at91_rtc_ioctl,
 	.read_time	= at91_rtc_readtime,
 	.set_time	= at91_rtc_settime,
diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c
index 629d47cc7e88..583789c66cdb 100644
--- a/drivers/rtc/rtc-dev.c
+++ b/drivers/rtc/rtc-dev.c
@@ -24,7 +24,7 @@ static int rtc_dev_open(struct inode *inode, struct file *file)
 	int err;
 	struct rtc_device *rtc = container_of(inode->i_cdev,
 					struct rtc_device, char_dev);
-	struct rtc_class_ops *ops = rtc->ops;
+	const struct rtc_class_ops *ops = rtc->ops;
 
 	/* We keep the lock as long as the device is in use
 	 * and return immediately if busy
@@ -209,7 +209,7 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file,
 	int err = 0;
 	struct class_device *class_dev = file->private_data;
 	struct rtc_device *rtc = to_rtc_device(class_dev);
-	struct rtc_class_ops *ops = rtc->ops;
+	const struct rtc_class_ops *ops = rtc->ops;
 	struct rtc_time tm;
 	struct rtc_wkalrm alarm;
 	void __user *uarg = (void __user *) arg;
diff --git a/drivers/rtc/rtc-ds1307.c b/drivers/rtc/rtc-ds1307.c
index e8afb9384786..cc5032b6f42a 100644
--- a/drivers/rtc/rtc-ds1307.c
+++ b/drivers/rtc/rtc-ds1307.c
@@ -178,7 +178,7 @@ static int ds1307_set_time(struct device *dev, struct rtc_time *t)
 	return 0;
 }
 
-static struct rtc_class_ops ds13xx_rtc_ops = {
+static const struct rtc_class_ops ds13xx_rtc_ops = {
 	.read_time	= ds1307_get_time,
 	.set_time	= ds1307_set_time,
 };
diff --git a/drivers/rtc/rtc-ds1553.c b/drivers/rtc/rtc-ds1553.c
index 209001495474..4fc9422ed86d 100644
--- a/drivers/rtc/rtc-ds1553.c
+++ b/drivers/rtc/rtc-ds1553.c
@@ -250,7 +250,7 @@ static int ds1553_rtc_ioctl(struct device *dev, unsigned int cmd,
 	return 0;
 }
 
-static struct rtc_class_ops ds1553_rtc_ops = {
+static const struct rtc_class_ops ds1553_rtc_ops = {
 	.read_time	= ds1553_rtc_read_time,
 	.set_time	= ds1553_rtc_set_time,
 	.read_alarm	= ds1553_rtc_read_alarm,
diff --git a/drivers/rtc/rtc-ds1672.c b/drivers/rtc/rtc-ds1672.c
index 9be81fd4737c..9c68ec99afa5 100644
--- a/drivers/rtc/rtc-ds1672.c
+++ b/drivers/rtc/rtc-ds1672.c
@@ -156,7 +156,7 @@ static ssize_t show_control(struct device *dev, struct device_attribute *attr, c
 }
 static DEVICE_ATTR(control, S_IRUGO, show_control, NULL);
 
-static struct rtc_class_ops ds1672_rtc_ops = {
+static const struct rtc_class_ops ds1672_rtc_ops = {
 	.read_time	= ds1672_rtc_read_time,
 	.set_time	= ds1672_rtc_set_time,
 	.set_mmss	= ds1672_rtc_set_mmss,
diff --git a/drivers/rtc/rtc-ds1742.c b/drivers/rtc/rtc-ds1742.c
index 8e47e5a06d2a..01da5ab15fbc 100644
--- a/drivers/rtc/rtc-ds1742.c
+++ b/drivers/rtc/rtc-ds1742.c
@@ -116,7 +116,7 @@ static int ds1742_rtc_read_time(struct device *dev, struct rtc_time *tm)
 	return 0;
 }
 
-static struct rtc_class_ops ds1742_rtc_ops = {
+static const struct rtc_class_ops ds1742_rtc_ops = {
 	.read_time	= ds1742_rtc_read_time,
 	.set_time	= ds1742_rtc_set_time,
 };
diff --git a/drivers/rtc/rtc-ep93xx.c b/drivers/rtc/rtc-ep93xx.c
index e1a1169e4664..ef4f147f3c0c 100644
--- a/drivers/rtc/rtc-ep93xx.c
+++ b/drivers/rtc/rtc-ep93xx.c
@@ -73,7 +73,7 @@ static int ep93xx_rtc_proc(struct device *dev, struct seq_file *seq)
 	return 0;
 }
 
-static struct rtc_class_ops ep93xx_rtc_ops = {
+static const struct rtc_class_ops ep93xx_rtc_ops = {
 	.read_time	= ep93xx_rtc_read_time,
 	.set_time	= ep93xx_rtc_set_time,
 	.set_mmss	= ep93xx_rtc_set_mmss,
diff --git a/drivers/rtc/rtc-isl1208.c b/drivers/rtc/rtc-isl1208.c
index f324d0a635d4..1c743641b73b 100644
--- a/drivers/rtc/rtc-isl1208.c
+++ b/drivers/rtc/rtc-isl1208.c
@@ -390,7 +390,7 @@ static int isl1208_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm)
 	return isl1208_i2c_read_alarm(to_i2c_client(dev), alarm);
 }
 
-static struct rtc_class_ops isl1208_rtc_ops = {
+static const struct rtc_class_ops isl1208_rtc_ops = {
 	.proc		= isl1208_rtc_proc,
 	.read_time	= isl1208_rtc_read_time,
 	.set_time	= isl1208_rtc_set_time,
diff --git a/drivers/rtc/rtc-m48t86.c b/drivers/rtc/rtc-m48t86.c
index 8c0d1a6739ad..8ff4a1221f59 100644
--- a/drivers/rtc/rtc-m48t86.c
+++ b/drivers/rtc/rtc-m48t86.c
@@ -138,7 +138,7 @@ static int m48t86_rtc_proc(struct device *dev, struct seq_file *seq)
 	return 0;
 }
 
-static struct rtc_class_ops m48t86_rtc_ops = {
+static const struct rtc_class_ops m48t86_rtc_ops = {
 	.read_time	= m48t86_rtc_read_time,
 	.set_time	= m48t86_rtc_set_time,
 	.proc		= m48t86_rtc_proc,
diff --git a/drivers/rtc/rtc-max6902.c b/drivers/rtc/rtc-max6902.c
index 2c9739562b5c..9eeef964663a 100644
--- a/drivers/rtc/rtc-max6902.c
+++ b/drivers/rtc/rtc-max6902.c
@@ -207,7 +207,7 @@ static int max6902_set_time(struct device *dev, struct rtc_time *tm)
 	return max6902_set_datetime(dev, tm);
 }
 
-static struct rtc_class_ops max6902_rtc_ops = {
+static const struct rtc_class_ops max6902_rtc_ops = {
 	.read_time	= max6902_read_time,
 	.set_time	= max6902_set_time,
 };
diff --git a/drivers/rtc/rtc-pcf8563.c b/drivers/rtc/rtc-pcf8563.c
index bd4310643038..a760cf69af90 100644
--- a/drivers/rtc/rtc-pcf8563.c
+++ b/drivers/rtc/rtc-pcf8563.c
@@ -227,7 +227,7 @@ static int pcf8563_rtc_set_time(struct device *dev, struct rtc_time *tm)
 	return pcf8563_set_datetime(to_i2c_client(dev), tm);
 }
 
-static struct rtc_class_ops pcf8563_rtc_ops = {
+static const struct rtc_class_ops pcf8563_rtc_ops = {
 	.read_time	= pcf8563_rtc_read_time,
 	.set_time	= pcf8563_rtc_set_time,
 };
diff --git a/drivers/rtc/rtc-pcf8583.c b/drivers/rtc/rtc-pcf8583.c
index b235a30cb661..5875ebb8c79d 100644
--- a/drivers/rtc/rtc-pcf8583.c
+++ b/drivers/rtc/rtc-pcf8583.c
@@ -273,7 +273,7 @@ static int pcf8583_rtc_set_time(struct device *dev, struct rtc_time *tm)
 	return ret;
 }
 
-static struct rtc_class_ops pcf8583_rtc_ops = {
+static const struct rtc_class_ops pcf8583_rtc_ops = {
 	.read_time	= pcf8583_rtc_read_time,
 	.set_time	= pcf8583_rtc_set_time,
 };
diff --git a/drivers/rtc/rtc-pl031.c b/drivers/rtc/rtc-pl031.c
index d6d1c5726b0e..739d1a6e14eb 100644
--- a/drivers/rtc/rtc-pl031.c
+++ b/drivers/rtc/rtc-pl031.c
@@ -128,7 +128,7 @@ static int pl031_set_alarm(struct device *dev, struct rtc_wkalrm *alarm)
 	return 0;
 }
 
-static struct rtc_class_ops pl031_ops = {
+static const struct rtc_class_ops pl031_ops = {
 	.open = pl031_open,
 	.release = pl031_release,
 	.ioctl = pl031_ioctl,
diff --git a/drivers/rtc/rtc-proc.c b/drivers/rtc/rtc-proc.c
index 2943d83edfd1..d51d8f20e634 100644
--- a/drivers/rtc/rtc-proc.c
+++ b/drivers/rtc/rtc-proc.c
@@ -23,7 +23,7 @@ static int rtc_proc_show(struct seq_file *seq, void *offset)
 {
 	int err;
 	struct class_device *class_dev = seq->private;
-	struct rtc_class_ops *ops = to_rtc_device(class_dev)->ops;
+	const struct rtc_class_ops *ops = to_rtc_device(class_dev)->ops;
 	struct rtc_wkalrm alrm;
 	struct rtc_time tm;
 
diff --git a/drivers/rtc/rtc-rs5c348.c b/drivers/rtc/rtc-rs5c348.c
index 25589061f931..f50f3fc353cd 100644
--- a/drivers/rtc/rtc-rs5c348.c
+++ b/drivers/rtc/rtc-rs5c348.c
@@ -140,7 +140,7 @@ rs5c348_rtc_read_time(struct device *dev, struct rtc_time *tm)
 	return 0;
 }
 
-static struct rtc_class_ops rs5c348_rtc_ops = {
+static const struct rtc_class_ops rs5c348_rtc_ops = {
 	.read_time	= rs5c348_rtc_read_time,
 	.set_time	= rs5c348_rtc_set_time,
 };
diff --git a/drivers/rtc/rtc-rs5c372.c b/drivers/rtc/rtc-rs5c372.c
index 7553d797603f..bbdad099471d 100644
--- a/drivers/rtc/rtc-rs5c372.c
+++ b/drivers/rtc/rtc-rs5c372.c
@@ -160,7 +160,7 @@ static int rs5c372_rtc_proc(struct device *dev, struct seq_file *seq)
 	return 0;
 }
 
-static struct rtc_class_ops rs5c372_rtc_ops = {
+static const struct rtc_class_ops rs5c372_rtc_ops = {
 	.proc		= rs5c372_rtc_proc,
 	.read_time	= rs5c372_rtc_read_time,
 	.set_time	= rs5c372_rtc_set_time,
diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c
index 2c7de79c83b9..625dad2eeb4f 100644
--- a/drivers/rtc/rtc-s3c.c
+++ b/drivers/rtc/rtc-s3c.c
@@ -386,7 +386,7 @@ static void s3c_rtc_release(struct device *dev)
 	free_irq(s3c_rtc_tickno, rtc_dev);
 }
 
-static struct rtc_class_ops s3c_rtcops = {
+static const struct rtc_class_ops s3c_rtcops = {
 	.open		= s3c_rtc_open,
 	.release	= s3c_rtc_release,
 	.ioctl		= s3c_rtc_ioctl,
diff --git a/drivers/rtc/rtc-sa1100.c b/drivers/rtc/rtc-sa1100.c
index ee4b61ee67b0..439c41aea31c 100644
--- a/drivers/rtc/rtc-sa1100.c
+++ b/drivers/rtc/rtc-sa1100.c
@@ -303,7 +303,7 @@ static int sa1100_rtc_proc(struct device *dev, struct seq_file *seq)
 	return 0;
 }
 
-static struct rtc_class_ops sa1100_rtc_ops = {
+static const struct rtc_class_ops sa1100_rtc_ops = {
 	.open = sa1100_rtc_open,
 	.read_callback = sa1100_rtc_read_callback,
 	.release = sa1100_rtc_release,
diff --git a/drivers/rtc/rtc-test.c b/drivers/rtc/rtc-test.c
index e1fa5fe7901f..bc4bd24508a2 100644
--- a/drivers/rtc/rtc-test.c
+++ b/drivers/rtc/rtc-test.c
@@ -75,7 +75,7 @@ static int test_rtc_ioctl(struct device *dev, unsigned int cmd,
 	}
 }
 
-static struct rtc_class_ops test_rtc_ops = {
+static const struct rtc_class_ops test_rtc_ops = {
 	.proc = test_rtc_proc,
 	.read_time = test_rtc_read_time,
 	.set_time = test_rtc_set_time,
diff --git a/drivers/rtc/rtc-v3020.c b/drivers/rtc/rtc-v3020.c
index e28cc4b0901a..09b714f1cdc3 100644
--- a/drivers/rtc/rtc-v3020.c
+++ b/drivers/rtc/rtc-v3020.c
@@ -149,7 +149,7 @@ static int v3020_set_time(struct device *dev, struct rtc_time *dt)
 	return 0;
 }
 
-static struct rtc_class_ops v3020_rtc_ops = {
+static const struct rtc_class_ops v3020_rtc_ops = {
 	.read_time	= v3020_read_time,
 	.set_time	= v3020_set_time,
 };
diff --git a/drivers/rtc/rtc-vr41xx.c b/drivers/rtc/rtc-vr41xx.c
index 596764fd29f5..58e5ed0aa127 100644
--- a/drivers/rtc/rtc-vr41xx.c
+++ b/drivers/rtc/rtc-vr41xx.c
@@ -296,7 +296,7 @@ static irqreturn_t rtclong1_interrupt(int irq, void *dev_id, struct pt_regs *reg
 	return IRQ_HANDLED;
 }
 
-static struct rtc_class_ops vr41xx_rtc_ops = {
+static const struct rtc_class_ops vr41xx_rtc_ops = {
 	.release	= vr41xx_rtc_release,
 	.ioctl		= vr41xx_rtc_ioctl,
 	.read_time	= vr41xx_rtc_read_time,
diff --git a/drivers/rtc/rtc-x1205.c b/drivers/rtc/rtc-x1205.c
index 788b6d1f8f2f..522c69753bbf 100644
--- a/drivers/rtc/rtc-x1205.c
+++ b/drivers/rtc/rtc-x1205.c
@@ -460,7 +460,7 @@ static int x1205_rtc_proc(struct device *dev, struct seq_file *seq)
 	return 0;
 }
 
-static struct rtc_class_ops x1205_rtc_ops = {
+static const struct rtc_class_ops x1205_rtc_ops = {
 	.proc		= x1205_rtc_proc,
 	.read_time	= x1205_rtc_read_time,
 	.set_time	= x1205_rtc_set_time,
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index 5371e4e74595..b89f09357054 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -141,7 +141,7 @@ struct rtc_device
 	int id;
 	char name[RTC_DEVICE_NAME_SIZE];
 
-	struct rtc_class_ops *ops;
+	const struct rtc_class_ops *ops;
 	struct mutex ops_lock;
 
 	struct class_device *rtc_dev;
@@ -172,7 +172,7 @@ struct rtc_device
 
 extern struct rtc_device *rtc_device_register(const char *name,
 					struct device *dev,
-					struct rtc_class_ops *ops,
+					const struct rtc_class_ops *ops,
 					struct module *owner);
 extern void rtc_device_unregister(struct rtc_device *rdev);
 extern int rtc_interface_register(struct class_interface *intf);
-- 
cgit v1.2.3


From c902e0a0102f1095eec4b3511c13c84ca2bc4577 Mon Sep 17 00:00:00 2001
From: Josh Triplett <josht@us.ibm.com>
Date: Sat, 30 Sep 2006 23:28:21 -0700
Subject: [PATCH] Pass sparse the lock expression given to lock annotations

The lock annotation macros __acquires, __releases, __acquire, and __release
all currently throw away the lock expression passed as an argument.  Now
that sparse can parse __context__ and __attribute__((context)) with a
context expression, pass the lock expression down to sparse as the context
expression.  This requires a version of sparse from GIT commit
37475a6c1c3e66219e68d912d5eb833f4098fd72 or later.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/compiler.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 0780de440220..538423d4a865 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -10,10 +10,10 @@
 # define __force	__attribute__((force))
 # define __nocast	__attribute__((nocast))
 # define __iomem	__attribute__((noderef, address_space(2)))
-# define __acquires(x)	__attribute__((context(0,1)))
-# define __releases(x)	__attribute__((context(1,0)))
-# define __acquire(x)	__context__(1)
-# define __release(x)	__context__(-1)
+# define __acquires(x)	__attribute__((context(x,0,1)))
+# define __releases(x)	__attribute__((context(x,1,0)))
+# define __acquire(x)	__context__(x,1)
+# define __release(x)	__context__(x,-1)
 # define __cond_lock(x,c)	((c) ? ({ __acquire(x); 1; }) : 0)
 extern void __chk_user_ptr(void __user *);
 extern void __chk_io_ptr(void __iomem *);
-- 
cgit v1.2.3


From 4c7ee8de956fc250fe31e2fa91f6da980fabe317 Mon Sep 17 00:00:00 2001
From: john stultz <johnstul@us.ibm.com>
Date: Sat, 30 Sep 2006 23:28:22 -0700
Subject: [PATCH] NTP: Move all the NTP related code to ntp.c

Move all the NTP related code to ntp.c

[akpm@osdl.org: cleanups, build fix]
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h |  10 +-
 kernel/time.c         | 173 ----------------------
 kernel/time/Makefile  |   2 +-
 kernel/time/ntp.c     | 389 ++++++++++++++++++++++++++++++++++++++++++++++++++
 kernel/timer.c        | 211 +--------------------------
 5 files changed, 399 insertions(+), 386 deletions(-)
 create mode 100644 kernel/time/ntp.c

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index d543d3871e38..2a21485bf183 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -294,11 +294,15 @@ extern void register_time_interpolator(struct time_interpolator *);
 extern void unregister_time_interpolator(struct time_interpolator *);
 extern void time_interpolator_reset(void);
 extern unsigned long time_interpolator_get_offset(void);
+extern void time_interpolator_update(long delta_nsec);
 
 #else /* !CONFIG_TIME_INTERPOLATION */
 
-static inline void
-time_interpolator_reset(void)
+static inline void time_interpolator_reset(void)
+{
+}
+
+static inline void time_interpolator_update(long delta_nsec)
 {
 }
 
@@ -309,6 +313,8 @@ time_interpolator_reset(void)
 /* Returns how long ticks are at present, in ns / 2^(SHIFT_SCALE-10). */
 extern u64 current_tick_length(void);
 
+extern void second_overflow(void);
+extern void update_ntp_one_tick(void);
 extern int do_adjtimex(struct timex *);
 
 #endif /* KERNEL */
diff --git a/kernel/time.c b/kernel/time.c
index 5bd489747643..0e017bff4c19 100644
--- a/kernel/time.c
+++ b/kernel/time.c
@@ -202,179 +202,6 @@ asmlinkage long sys_settimeofday(struct timeval __user *tv,
 	return do_sys_settimeofday(tv ? &new_ts : NULL, tz ? &new_tz : NULL);
 }
 
-/* we call this to notify the arch when the clock is being
- * controlled.  If no such arch routine, do nothing.
- */
-void __attribute__ ((weak)) notify_arch_cmos_timer(void)
-{
-	return;
-}
-
-/* adjtimex mainly allows reading (and writing, if superuser) of
- * kernel time-keeping variables. used by xntpd.
- */
-int do_adjtimex(struct timex *txc)
-{
-        long ltemp, mtemp, save_adjust;
-	int result;
-
-	/* In order to modify anything, you gotta be super-user! */
-	if (txc->modes && !capable(CAP_SYS_TIME))
-		return -EPERM;
-		
-	/* Now we validate the data before disabling interrupts */
-
-	if ((txc->modes & ADJ_OFFSET_SINGLESHOT) == ADJ_OFFSET_SINGLESHOT)
-	  /* singleshot must not be used with any other mode bits */
-		if (txc->modes != ADJ_OFFSET_SINGLESHOT)
-			return -EINVAL;
-
-	if (txc->modes != ADJ_OFFSET_SINGLESHOT && (txc->modes & ADJ_OFFSET))
-	  /* adjustment Offset limited to +- .512 seconds */
-		if (txc->offset <= - MAXPHASE || txc->offset >= MAXPHASE )
-			return -EINVAL;	
-
-	/* if the quartz is off by more than 10% something is VERY wrong ! */
-	if (txc->modes & ADJ_TICK)
-		if (txc->tick <  900000/USER_HZ ||
-		    txc->tick > 1100000/USER_HZ)
-			return -EINVAL;
-
-	write_seqlock_irq(&xtime_lock);
-	result = time_state;	/* mostly `TIME_OK' */
-
-	/* Save for later - semantics of adjtime is to return old value */
-	save_adjust = time_next_adjust ? time_next_adjust : time_adjust;
-
-#if 0	/* STA_CLOCKERR is never set yet */
-	time_status &= ~STA_CLOCKERR;		/* reset STA_CLOCKERR */
-#endif
-	/* If there are input parameters, then process them */
-	if (txc->modes)
-	{
-	    if (txc->modes & ADJ_STATUS)	/* only set allowed bits */
-		time_status =  (txc->status & ~STA_RONLY) |
-			      (time_status & STA_RONLY);
-
-	    if (txc->modes & ADJ_FREQUENCY) {	/* p. 22 */
-		if (txc->freq > MAXFREQ || txc->freq < -MAXFREQ) {
-		    result = -EINVAL;
-		    goto leave;
-		}
-		time_freq = txc->freq;
-	    }
-
-	    if (txc->modes & ADJ_MAXERROR) {
-		if (txc->maxerror < 0 || txc->maxerror >= NTP_PHASE_LIMIT) {
-		    result = -EINVAL;
-		    goto leave;
-		}
-		time_maxerror = txc->maxerror;
-	    }
-
-	    if (txc->modes & ADJ_ESTERROR) {
-		if (txc->esterror < 0 || txc->esterror >= NTP_PHASE_LIMIT) {
-		    result = -EINVAL;
-		    goto leave;
-		}
-		time_esterror = txc->esterror;
-	    }
-
-	    if (txc->modes & ADJ_TIMECONST) {	/* p. 24 */
-		if (txc->constant < 0) {	/* NTP v4 uses values > 6 */
-		    result = -EINVAL;
-		    goto leave;
-		}
-		time_constant = txc->constant;
-	    }
-
-	    if (txc->modes & ADJ_OFFSET) {	/* values checked earlier */
-		if (txc->modes == ADJ_OFFSET_SINGLESHOT) {
-		    /* adjtime() is independent from ntp_adjtime() */
-		    if ((time_next_adjust = txc->offset) == 0)
-			 time_adjust = 0;
-		}
-		else if (time_status & STA_PLL) {
-		    ltemp = txc->offset;
-
-		    /*
-		     * Scale the phase adjustment and
-		     * clamp to the operating range.
-		     */
-		    if (ltemp > MAXPHASE)
-		        time_offset = MAXPHASE << SHIFT_UPDATE;
-		    else if (ltemp < -MAXPHASE)
-			time_offset = -(MAXPHASE << SHIFT_UPDATE);
-		    else
-		        time_offset = ltemp << SHIFT_UPDATE;
-
-		    /*
-		     * Select whether the frequency is to be controlled
-		     * and in which mode (PLL or FLL). Clamp to the operating
-		     * range. Ugly multiply/divide should be replaced someday.
-		     */
-
-		    if (time_status & STA_FREQHOLD || time_reftime == 0)
-		        time_reftime = xtime.tv_sec;
-		    mtemp = xtime.tv_sec - time_reftime;
-		    time_reftime = xtime.tv_sec;
-		    if (time_status & STA_FLL) {
-		        if (mtemp >= MINSEC) {
-			    ltemp = (time_offset / mtemp) << (SHIFT_USEC -
-							      SHIFT_UPDATE);
-			    time_freq += shift_right(ltemp, SHIFT_KH);
-			} else /* calibration interval too short (p. 12) */
-				result = TIME_ERROR;
-		    } else {	/* PLL mode */
-		        if (mtemp < MAXSEC) {
-			    ltemp *= mtemp;
-			    time_freq += shift_right(ltemp,(time_constant +
-						       time_constant +
-						       SHIFT_KF - SHIFT_USEC));
-			} else /* calibration interval too long (p. 12) */
-				result = TIME_ERROR;
-		    }
-		    time_freq = min(time_freq, time_tolerance);
-		    time_freq = max(time_freq, -time_tolerance);
-		} /* STA_PLL */
-	    } /* txc->modes & ADJ_OFFSET */
-	    if (txc->modes & ADJ_TICK) {
-		tick_usec = txc->tick;
-		tick_nsec = TICK_USEC_TO_NSEC(tick_usec);
-	    }
-	} /* txc->modes */
-leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
-		result = TIME_ERROR;
-	
-	if ((txc->modes & ADJ_OFFSET_SINGLESHOT) == ADJ_OFFSET_SINGLESHOT)
-	    txc->offset	   = save_adjust;
-	else {
-	    txc->offset = shift_right(time_offset, SHIFT_UPDATE);
-	}
-	txc->freq	   = time_freq;
-	txc->maxerror	   = time_maxerror;
-	txc->esterror	   = time_esterror;
-	txc->status	   = time_status;
-	txc->constant	   = time_constant;
-	txc->precision	   = time_precision;
-	txc->tolerance	   = time_tolerance;
-	txc->tick	   = tick_usec;
-
-	/* PPS is not implemented, so these are zero */
-	txc->ppsfreq	   = 0;
-	txc->jitter	   = 0;
-	txc->shift	   = 0;
-	txc->stabil	   = 0;
-	txc->jitcnt	   = 0;
-	txc->calcnt	   = 0;
-	txc->errcnt	   = 0;
-	txc->stbcnt	   = 0;
-	write_sequnlock_irq(&xtime_lock);
-	do_gettimeofday(&txc->time);
-	notify_arch_cmos_timer();
-	return(result);
-}
-
 asmlinkage long sys_adjtimex(struct timex __user *txc_p)
 {
 	struct timex txc;		/* Local copy of parameter */
diff --git a/kernel/time/Makefile b/kernel/time/Makefile
index e1dfd8e86cce..61a3907d16fb 100644
--- a/kernel/time/Makefile
+++ b/kernel/time/Makefile
@@ -1 +1 @@
-obj-y += clocksource.o jiffies.o
+obj-y += ntp.o clocksource.o jiffies.o
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
new file mode 100644
index 000000000000..8ccce15b4b23
--- /dev/null
+++ b/kernel/time/ntp.c
@@ -0,0 +1,389 @@
+/*
+ * linux/kernel/time/ntp.c
+ *
+ * NTP state machine interfaces and logic.
+ *
+ * This code was mainly moved from kernel/timer.c and kernel/time.c
+ * Please see those files for relevant copyright info and historical
+ * changelogs.
+ */
+
+#include <linux/mm.h>
+#include <linux/time.h>
+#include <linux/timex.h>
+
+#include <asm/div64.h>
+#include <asm/timex.h>
+
+/* Don't completely fail for HZ > 500.  */
+int tickadj = 500/HZ ? : 1;		/* microsecs */
+
+/*
+ * phase-lock loop variables
+ */
+/* TIME_ERROR prevents overwriting the CMOS clock */
+int time_state = TIME_OK;		/* clock synchronization status	*/
+int time_status = STA_UNSYNC;		/* clock status bits		*/
+long time_offset;			/* time adjustment (us)		*/
+long time_constant = 2;			/* pll time constant		*/
+long time_tolerance = MAXFREQ;		/* frequency tolerance (ppm)	*/
+long time_precision = 1;		/* clock precision (us)		*/
+long time_maxerror = NTP_PHASE_LIMIT;	/* maximum error (us)		*/
+long time_esterror = NTP_PHASE_LIMIT;	/* estimated error (us)		*/
+long time_freq = (((NSEC_PER_SEC + HZ/2) % HZ - HZ/2) << SHIFT_USEC) / NSEC_PER_USEC;
+					/* frequency offset (scaled ppm)*/
+static long time_adj;			/* tick adjust (scaled 1 / HZ)	*/
+long time_reftime;			/* time at last adjustment (s)	*/
+long time_adjust;
+long time_next_adjust;
+
+/*
+ * this routine handles the overflow of the microsecond field
+ *
+ * The tricky bits of code to handle the accurate clock support
+ * were provided by Dave Mills (Mills@UDEL.EDU) of NTP fame.
+ * They were originally developed for SUN and DEC kernels.
+ * All the kudos should go to Dave for this stuff.
+ */
+void second_overflow(void)
+{
+	long ltemp;
+
+	/* Bump the maxerror field */
+	time_maxerror += time_tolerance >> SHIFT_USEC;
+	if (time_maxerror > NTP_PHASE_LIMIT) {
+		time_maxerror = NTP_PHASE_LIMIT;
+		time_status |= STA_UNSYNC;
+	}
+
+	/*
+	 * Leap second processing. If in leap-insert state at the end of the
+	 * day, the system clock is set back one second; if in leap-delete
+	 * state, the system clock is set ahead one second. The microtime()
+	 * routine or external clock driver will insure that reported time is
+	 * always monotonic. The ugly divides should be replaced.
+	 */
+	switch (time_state) {
+	case TIME_OK:
+		if (time_status & STA_INS)
+			time_state = TIME_INS;
+		else if (time_status & STA_DEL)
+			time_state = TIME_DEL;
+		break;
+	case TIME_INS:
+		if (xtime.tv_sec % 86400 == 0) {
+			xtime.tv_sec--;
+			wall_to_monotonic.tv_sec++;
+			/*
+			 * The timer interpolator will make time change
+			 * gradually instead of an immediate jump by one second
+			 */
+			time_interpolator_update(-NSEC_PER_SEC);
+			time_state = TIME_OOP;
+			clock_was_set();
+			printk(KERN_NOTICE "Clock: inserting leap second "
+					"23:59:60 UTC\n");
+		}
+		break;
+	case TIME_DEL:
+		if ((xtime.tv_sec + 1) % 86400 == 0) {
+			xtime.tv_sec++;
+			wall_to_monotonic.tv_sec--;
+			/*
+			 * Use of time interpolator for a gradual change of
+			 * time
+			 */
+			time_interpolator_update(NSEC_PER_SEC);
+			time_state = TIME_WAIT;
+			clock_was_set();
+			printk(KERN_NOTICE "Clock: deleting leap second "
+					"23:59:59 UTC\n");
+		}
+		break;
+	case TIME_OOP:
+		time_state = TIME_WAIT;
+		break;
+	case TIME_WAIT:
+		if (!(time_status & (STA_INS | STA_DEL)))
+		time_state = TIME_OK;
+	}
+
+	/*
+	 * Compute the phase adjustment for the next second. In PLL mode, the
+	 * offset is reduced by a fixed factor times the time constant. In FLL
+	 * mode the offset is used directly. In either mode, the maximum phase
+	 * adjustment for each second is clamped so as to spread the adjustment
+	 * over not more than the number of seconds between updates.
+	 */
+	ltemp = time_offset;
+	if (!(time_status & STA_FLL))
+		ltemp = shift_right(ltemp, SHIFT_KG + time_constant);
+	ltemp = min(ltemp, (MAXPHASE / MINSEC) << SHIFT_UPDATE);
+	ltemp = max(ltemp, -(MAXPHASE / MINSEC) << SHIFT_UPDATE);
+	time_offset -= ltemp;
+	time_adj = ltemp << (SHIFT_SCALE - SHIFT_HZ - SHIFT_UPDATE);
+
+	/*
+	 * Compute the frequency estimate and additional phase adjustment due
+	 * to frequency error for the next second.
+	 */
+	ltemp = time_freq;
+	time_adj += shift_right(ltemp,(SHIFT_USEC + SHIFT_HZ - SHIFT_SCALE));
+
+#if HZ == 100
+	/*
+	 * Compensate for (HZ==100) != (1 << SHIFT_HZ).  Add 25% and 3.125% to
+	 * get 128.125; => only 0.125% error (p. 14)
+	 */
+	time_adj += shift_right(time_adj, 2) + shift_right(time_adj, 5);
+#endif
+#if HZ == 250
+	/*
+	 * Compensate for (HZ==250) != (1 << SHIFT_HZ).  Add 1.5625% and
+	 * 0.78125% to get 255.85938; => only 0.05% error (p. 14)
+	 */
+	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
+#endif
+#if HZ == 1000
+	/*
+	 * Compensate for (HZ==1000) != (1 << SHIFT_HZ).  Add 1.5625% and
+	 * 0.78125% to get 1023.4375; => only 0.05% error (p. 14)
+	 */
+	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
+#endif
+}
+
+/*
+ * Returns how many microseconds we need to add to xtime this tick
+ * in doing an adjustment requested with adjtime.
+ */
+static long adjtime_adjustment(void)
+{
+	long time_adjust_step;
+
+	time_adjust_step = time_adjust;
+	if (time_adjust_step) {
+		/*
+		 * We are doing an adjtime thing.  Prepare time_adjust_step to
+		 * be within bounds.  Note that a positive time_adjust means we
+		 * want the clock to run faster.
+		 *
+		 * Limit the amount of the step to be in the range
+		 * -tickadj .. +tickadj
+		 */
+		time_adjust_step = min(time_adjust_step, (long)tickadj);
+		time_adjust_step = max(time_adjust_step, (long)-tickadj);
+	}
+	return time_adjust_step;
+}
+
+/* in the NTP reference this is called "hardclock()" */
+void update_ntp_one_tick(void)
+{
+	long time_adjust_step;
+
+	time_adjust_step = adjtime_adjustment();
+	if (time_adjust_step)
+		/* Reduce by this step the amount of time left  */
+		time_adjust -= time_adjust_step;
+
+	/* Changes by adjtime() do not take effect till next tick. */
+	if (time_next_adjust != 0) {
+		time_adjust = time_next_adjust;
+		time_next_adjust = 0;
+	}
+}
+
+/*
+ * Return how long ticks are at the moment, that is, how much time
+ * update_wall_time_one_tick will add to xtime next time we call it
+ * (assuming no calls to do_adjtimex in the meantime).
+ * The return value is in fixed-point nanoseconds shifted by the
+ * specified number of bits to the right of the binary point.
+ * This function has no side-effects.
+ */
+u64 current_tick_length(void)
+{
+	long delta_nsec;
+	u64 ret;
+
+	/* calculate the finest interval NTP will allow.
+	 *    ie: nanosecond value shifted by (SHIFT_SCALE - 10)
+	 */
+	delta_nsec = tick_nsec + adjtime_adjustment() * 1000;
+	ret = (u64)delta_nsec << TICK_LENGTH_SHIFT;
+	ret += (s64)time_adj << (TICK_LENGTH_SHIFT - (SHIFT_SCALE - 10));
+
+	return ret;
+}
+
+
+void __attribute__ ((weak)) notify_arch_cmos_timer(void)
+{
+	return;
+}
+
+/* adjtimex mainly allows reading (and writing, if superuser) of
+ * kernel time-keeping variables. used by xntpd.
+ */
+int do_adjtimex(struct timex *txc)
+{
+	long ltemp, mtemp, save_adjust;
+	int result;
+
+	/* In order to modify anything, you gotta be super-user! */
+	if (txc->modes && !capable(CAP_SYS_TIME))
+		return -EPERM;
+
+	/* Now we validate the data before disabling interrupts */
+
+	if ((txc->modes & ADJ_OFFSET_SINGLESHOT) == ADJ_OFFSET_SINGLESHOT)
+	  /* singleshot must not be used with any other mode bits */
+		if (txc->modes != ADJ_OFFSET_SINGLESHOT)
+			return -EINVAL;
+
+	if (txc->modes != ADJ_OFFSET_SINGLESHOT && (txc->modes & ADJ_OFFSET))
+	  /* adjustment Offset limited to +- .512 seconds */
+		if (txc->offset <= - MAXPHASE || txc->offset >= MAXPHASE )
+			return -EINVAL;
+
+	/* if the quartz is off by more than 10% something is VERY wrong ! */
+	if (txc->modes & ADJ_TICK)
+		if (txc->tick <  900000/USER_HZ ||
+		    txc->tick > 1100000/USER_HZ)
+			return -EINVAL;
+
+	write_seqlock_irq(&xtime_lock);
+	result = time_state;	/* mostly `TIME_OK' */
+
+	/* Save for later - semantics of adjtime is to return old value */
+	save_adjust = time_next_adjust ? time_next_adjust : time_adjust;
+
+#if 0	/* STA_CLOCKERR is never set yet */
+	time_status &= ~STA_CLOCKERR;		/* reset STA_CLOCKERR */
+#endif
+	/* If there are input parameters, then process them */
+	if (txc->modes)
+	{
+	    if (txc->modes & ADJ_STATUS)	/* only set allowed bits */
+		time_status =  (txc->status & ~STA_RONLY) |
+			      (time_status & STA_RONLY);
+
+	    if (txc->modes & ADJ_FREQUENCY) {	/* p. 22 */
+		if (txc->freq > MAXFREQ || txc->freq < -MAXFREQ) {
+		    result = -EINVAL;
+		    goto leave;
+		}
+		time_freq = txc->freq;
+	    }
+
+	    if (txc->modes & ADJ_MAXERROR) {
+		if (txc->maxerror < 0 || txc->maxerror >= NTP_PHASE_LIMIT) {
+		    result = -EINVAL;
+		    goto leave;
+		}
+		time_maxerror = txc->maxerror;
+	    }
+
+	    if (txc->modes & ADJ_ESTERROR) {
+		if (txc->esterror < 0 || txc->esterror >= NTP_PHASE_LIMIT) {
+		    result = -EINVAL;
+		    goto leave;
+		}
+		time_esterror = txc->esterror;
+	    }
+
+	    if (txc->modes & ADJ_TIMECONST) {	/* p. 24 */
+		if (txc->constant < 0) {	/* NTP v4 uses values > 6 */
+		    result = -EINVAL;
+		    goto leave;
+		}
+		time_constant = txc->constant;
+	    }
+
+	    if (txc->modes & ADJ_OFFSET) {	/* values checked earlier */
+		if (txc->modes == ADJ_OFFSET_SINGLESHOT) {
+		    /* adjtime() is independent from ntp_adjtime() */
+		    if ((time_next_adjust = txc->offset) == 0)
+			 time_adjust = 0;
+		}
+		else if (time_status & STA_PLL) {
+		    ltemp = txc->offset;
+
+		    /*
+		     * Scale the phase adjustment and
+		     * clamp to the operating range.
+		     */
+		    if (ltemp > MAXPHASE)
+		        time_offset = MAXPHASE << SHIFT_UPDATE;
+		    else if (ltemp < -MAXPHASE)
+			time_offset = -(MAXPHASE << SHIFT_UPDATE);
+		    else
+		        time_offset = ltemp << SHIFT_UPDATE;
+
+		    /*
+		     * Select whether the frequency is to be controlled
+		     * and in which mode (PLL or FLL). Clamp to the operating
+		     * range. Ugly multiply/divide should be replaced someday.
+		     */
+
+		    if (time_status & STA_FREQHOLD || time_reftime == 0)
+		        time_reftime = xtime.tv_sec;
+		    mtemp = xtime.tv_sec - time_reftime;
+		    time_reftime = xtime.tv_sec;
+		    if (time_status & STA_FLL) {
+		        if (mtemp >= MINSEC) {
+			    ltemp = (time_offset / mtemp) << (SHIFT_USEC -
+							      SHIFT_UPDATE);
+			    time_freq += shift_right(ltemp, SHIFT_KH);
+			} else /* calibration interval too short (p. 12) */
+				result = TIME_ERROR;
+		    } else {	/* PLL mode */
+		        if (mtemp < MAXSEC) {
+			    ltemp *= mtemp;
+			    time_freq += shift_right(ltemp,(time_constant +
+						       time_constant +
+						       SHIFT_KF - SHIFT_USEC));
+			} else /* calibration interval too long (p. 12) */
+				result = TIME_ERROR;
+		    }
+		    time_freq = min(time_freq, time_tolerance);
+		    time_freq = max(time_freq, -time_tolerance);
+		} /* STA_PLL */
+	    } /* txc->modes & ADJ_OFFSET */
+	    if (txc->modes & ADJ_TICK) {
+		tick_usec = txc->tick;
+		tick_nsec = TICK_USEC_TO_NSEC(tick_usec);
+	    }
+	} /* txc->modes */
+leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
+		result = TIME_ERROR;
+
+	if ((txc->modes & ADJ_OFFSET_SINGLESHOT) == ADJ_OFFSET_SINGLESHOT)
+	    txc->offset	   = save_adjust;
+	else {
+	    txc->offset = shift_right(time_offset, SHIFT_UPDATE);
+	}
+	txc->freq	   = time_freq;
+	txc->maxerror	   = time_maxerror;
+	txc->esterror	   = time_esterror;
+	txc->status	   = time_status;
+	txc->constant	   = time_constant;
+	txc->precision	   = time_precision;
+	txc->tolerance	   = time_tolerance;
+	txc->tick	   = tick_usec;
+
+	/* PPS is not implemented, so these are zero */
+	txc->ppsfreq	   = 0;
+	txc->jitter	   = 0;
+	txc->shift	   = 0;
+	txc->stabil	   = 0;
+	txc->jitcnt	   = 0;
+	txc->calcnt	   = 0;
+	txc->errcnt	   = 0;
+	txc->stbcnt	   = 0;
+	write_sequnlock_irq(&xtime_lock);
+	do_gettimeofday(&txc->time);
+	notify_arch_cmos_timer();
+	return(result);
+}
diff --git a/kernel/timer.c b/kernel/timer.c
index 4f55622b0d38..5fccc7cbf3b4 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -41,12 +41,6 @@
 #include <asm/timex.h>
 #include <asm/io.h>
 
-#ifdef CONFIG_TIME_INTERPOLATION
-static void time_interpolator_update(long delta_nsec);
-#else
-#define time_interpolator_update(x)
-#endif
-
 u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES;
 
 EXPORT_SYMBOL(jiffies_64);
@@ -587,209 +581,6 @@ struct timespec wall_to_monotonic __attribute__ ((aligned (16)));
 
 EXPORT_SYMBOL(xtime);
 
-/* Don't completely fail for HZ > 500.  */
-int tickadj = 500/HZ ? : 1;		/* microsecs */
-
-
-/*
- * phase-lock loop variables
- */
-/* TIME_ERROR prevents overwriting the CMOS clock */
-int time_state = TIME_OK;		/* clock synchronization status	*/
-int time_status = STA_UNSYNC;		/* clock status bits		*/
-long time_offset;			/* time adjustment (us)		*/
-long time_constant = 2;			/* pll time constant		*/
-long time_tolerance = MAXFREQ;		/* frequency tolerance (ppm)	*/
-long time_precision = 1;		/* clock precision (us)		*/
-long time_maxerror = NTP_PHASE_LIMIT;	/* maximum error (us)		*/
-long time_esterror = NTP_PHASE_LIMIT;	/* estimated error (us)		*/
-long time_freq = (((NSEC_PER_SEC + HZ/2) % HZ - HZ/2) << SHIFT_USEC) / NSEC_PER_USEC;
-					/* frequency offset (scaled ppm)*/
-static long time_adj;			/* tick adjust (scaled 1 / HZ)	*/
-long time_reftime;			/* time at last adjustment (s)	*/
-long time_adjust;
-long time_next_adjust;
-
-/*
- * this routine handles the overflow of the microsecond field
- *
- * The tricky bits of code to handle the accurate clock support
- * were provided by Dave Mills (Mills@UDEL.EDU) of NTP fame.
- * They were originally developed for SUN and DEC kernels.
- * All the kudos should go to Dave for this stuff.
- *
- */
-static void second_overflow(void)
-{
-	long ltemp;
-
-	/* Bump the maxerror field */
-	time_maxerror += time_tolerance >> SHIFT_USEC;
-	if (time_maxerror > NTP_PHASE_LIMIT) {
-		time_maxerror = NTP_PHASE_LIMIT;
-		time_status |= STA_UNSYNC;
-	}
-
-	/*
-	 * Leap second processing. If in leap-insert state at the end of the
-	 * day, the system clock is set back one second; if in leap-delete
-	 * state, the system clock is set ahead one second. The microtime()
-	 * routine or external clock driver will insure that reported time is
-	 * always monotonic. The ugly divides should be replaced.
-	 */
-	switch (time_state) {
-	case TIME_OK:
-		if (time_status & STA_INS)
-			time_state = TIME_INS;
-		else if (time_status & STA_DEL)
-			time_state = TIME_DEL;
-		break;
-	case TIME_INS:
-		if (xtime.tv_sec % 86400 == 0) {
-			xtime.tv_sec--;
-			wall_to_monotonic.tv_sec++;
-			/*
-			 * The timer interpolator will make time change
-			 * gradually instead of an immediate jump by one second
-			 */
-			time_interpolator_update(-NSEC_PER_SEC);
-			time_state = TIME_OOP;
-			clock_was_set();
-			printk(KERN_NOTICE "Clock: inserting leap second "
-					"23:59:60 UTC\n");
-		}
-		break;
-	case TIME_DEL:
-		if ((xtime.tv_sec + 1) % 86400 == 0) {
-			xtime.tv_sec++;
-			wall_to_monotonic.tv_sec--;
-			/*
-			 * Use of time interpolator for a gradual change of
-			 * time
-			 */
-			time_interpolator_update(NSEC_PER_SEC);
-			time_state = TIME_WAIT;
-			clock_was_set();
-			printk(KERN_NOTICE "Clock: deleting leap second "
-					"23:59:59 UTC\n");
-		}
-		break;
-	case TIME_OOP:
-		time_state = TIME_WAIT;
-		break;
-	case TIME_WAIT:
-		if (!(time_status & (STA_INS | STA_DEL)))
-		time_state = TIME_OK;
-	}
-
-	/*
-	 * Compute the phase adjustment for the next second. In PLL mode, the
-	 * offset is reduced by a fixed factor times the time constant. In FLL
-	 * mode the offset is used directly. In either mode, the maximum phase
-	 * adjustment for each second is clamped so as to spread the adjustment
-	 * over not more than the number of seconds between updates.
-	 */
-	ltemp = time_offset;
-	if (!(time_status & STA_FLL))
-		ltemp = shift_right(ltemp, SHIFT_KG + time_constant);
-	ltemp = min(ltemp, (MAXPHASE / MINSEC) << SHIFT_UPDATE);
-	ltemp = max(ltemp, -(MAXPHASE / MINSEC) << SHIFT_UPDATE);
-	time_offset -= ltemp;
-	time_adj = ltemp << (SHIFT_SCALE - SHIFT_HZ - SHIFT_UPDATE);
-
-	/*
-	 * Compute the frequency estimate and additional phase adjustment due
-	 * to frequency error for the next second.
-	 */
-	ltemp = time_freq;
-	time_adj += shift_right(ltemp,(SHIFT_USEC + SHIFT_HZ - SHIFT_SCALE));
-
-#if HZ == 100
-	/*
-	 * Compensate for (HZ==100) != (1 << SHIFT_HZ).  Add 25% and 3.125% to
-	 * get 128.125; => only 0.125% error (p. 14)
-	 */
-	time_adj += shift_right(time_adj, 2) + shift_right(time_adj, 5);
-#endif
-#if HZ == 250
-	/*
-	 * Compensate for (HZ==250) != (1 << SHIFT_HZ).  Add 1.5625% and
-	 * 0.78125% to get 255.85938; => only 0.05% error (p. 14)
-	 */
-	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
-#endif
-#if HZ == 1000
-	/*
-	 * Compensate for (HZ==1000) != (1 << SHIFT_HZ).  Add 1.5625% and
-	 * 0.78125% to get 1023.4375; => only 0.05% error (p. 14)
-	 */
-	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
-#endif
-}
-
-/*
- * Returns how many microseconds we need to add to xtime this tick
- * in doing an adjustment requested with adjtime.
- */
-static long adjtime_adjustment(void)
-{
-	long time_adjust_step;
-
-	time_adjust_step = time_adjust;
-	if (time_adjust_step) {
-		/*
-		 * We are doing an adjtime thing.  Prepare time_adjust_step to
-		 * be within bounds.  Note that a positive time_adjust means we
-		 * want the clock to run faster.
-		 *
-		 * Limit the amount of the step to be in the range
-		 * -tickadj .. +tickadj
-		 */
-		time_adjust_step = min(time_adjust_step, (long)tickadj);
-		time_adjust_step = max(time_adjust_step, (long)-tickadj);
-	}
-	return time_adjust_step;
-}
-
-/* in the NTP reference this is called "hardclock()" */
-static void update_ntp_one_tick(void)
-{
-	long time_adjust_step;
-
-	time_adjust_step = adjtime_adjustment();
-	if (time_adjust_step)
-		/* Reduce by this step the amount of time left  */
-		time_adjust -= time_adjust_step;
-
-	/* Changes by adjtime() do not take effect till next tick. */
-	if (time_next_adjust != 0) {
-		time_adjust = time_next_adjust;
-		time_next_adjust = 0;
-	}
-}
-
-/*
- * Return how long ticks are at the moment, that is, how much time
- * update_wall_time_one_tick will add to xtime next time we call it
- * (assuming no calls to do_adjtimex in the meantime).
- * The return value is in fixed-point nanoseconds shifted by the
- * specified number of bits to the right of the binary point.
- * This function has no side-effects.
- */
-u64 current_tick_length(void)
-{
-	long delta_nsec;
-	u64 ret;
-
-	/* calculate the finest interval NTP will allow.
-	 *    ie: nanosecond value shifted by (SHIFT_SCALE - 10)
-	 */
-	delta_nsec = tick_nsec + adjtime_adjustment() * 1000;
-	ret = (u64)delta_nsec << TICK_LENGTH_SHIFT;
-	ret += (s64)time_adj << (TICK_LENGTH_SHIFT - (SHIFT_SCALE - 10));
-
-	return ret;
-}
 
 /* XXX - all of this timekeeping code should be later moved to time.c */
 #include <linux/clocksource.h>
@@ -1775,7 +1566,7 @@ unsigned long time_interpolator_get_offset(void)
 #define INTERPOLATOR_ADJUST 65536
 #define INTERPOLATOR_MAX_SKIP 10*INTERPOLATOR_ADJUST
 
-static void time_interpolator_update(long delta_nsec)
+void time_interpolator_update(long delta_nsec)
 {
 	u64 counter;
 	unsigned long offset;
-- 
cgit v1.2.3


From b0ee75561beadc4db4d9a899c8ef4a7db50aa0ab Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:22 -0700
Subject: [PATCH] ntp: add ntp_update_frequency

This introduces ntp_update_frequency() and deinlines ntp_clear() (as it's not
performance critical).  ntp_update_frequency() calculates the base tick length
using tick_usec and adds a base adjustment, in case the frequency doesn't
divide evenly by HZ.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h | 14 ++------------
 kernel/time/ntp.c     | 50 ++++++++++++++++++++++++++++++++++++++++++++------
 kernel/timer.c        | 11 ++++-------
 3 files changed, 50 insertions(+), 25 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index 2a21485bf183..b589c8218bb9 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -219,18 +219,8 @@ extern long time_reftime;	/* time at last adjustment (s) */
 extern long time_adjust;	/* The amount of adjtime left */
 extern long time_next_adjust;	/* Value for time_adjust at next tick */
 
-/**
- * ntp_clear - Clears the NTP state variables
- *
- * Must be called while holding a write on the xtime_lock
- */
-static inline void ntp_clear(void)
-{
-	time_adjust = 0;		/* stop active adjtime() */
-	time_status |= STA_UNSYNC;
-	time_maxerror = NTP_PHASE_LIMIT;
-	time_esterror = NTP_PHASE_LIMIT;
-}
+extern void ntp_clear(void);
+extern void ntp_update_frequency(void);
 
 /**
  * ntp_synced - Returns 1 if the NTP status is not UNSYNC
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 8ccce15b4b23..77137bec2aea 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -15,6 +15,13 @@
 #include <asm/div64.h>
 #include <asm/timex.h>
 
+/*
+ * Timekeeping variables
+ */
+unsigned long tick_usec = TICK_USEC; 		/* USER_HZ period (usec) */
+unsigned long tick_nsec;			/* ACTHZ period (nsec) */
+static u64 tick_length, tick_length_base;
+
 /* Don't completely fail for HZ > 500.  */
 int tickadj = 500/HZ ? : 1;		/* microsecs */
 
@@ -37,6 +44,36 @@ long time_reftime;			/* time at last adjustment (s)	*/
 long time_adjust;
 long time_next_adjust;
 
+/**
+ * ntp_clear - Clears the NTP state variables
+ *
+ * Must be called while holding a write on the xtime_lock
+ */
+void ntp_clear(void)
+{
+	time_adjust = 0;		/* stop active adjtime() */
+	time_status |= STA_UNSYNC;
+	time_maxerror = NTP_PHASE_LIMIT;
+	time_esterror = NTP_PHASE_LIMIT;
+
+	ntp_update_frequency();
+
+	tick_length = tick_length_base;
+}
+
+#define CLOCK_TICK_OVERFLOW	(LATCH * HZ - CLOCK_TICK_RATE)
+#define CLOCK_TICK_ADJUST	(((s64)CLOCK_TICK_OVERFLOW * NSEC_PER_SEC) / (s64)CLOCK_TICK_RATE)
+
+void ntp_update_frequency(void)
+{
+	tick_length_base = (u64)(tick_usec * NSEC_PER_USEC * USER_HZ) << TICK_LENGTH_SHIFT;
+	tick_length_base += (s64)CLOCK_TICK_ADJUST << TICK_LENGTH_SHIFT;
+
+	do_div(tick_length_base, HZ);
+
+	tick_nsec = tick_length_base >> TICK_LENGTH_SHIFT;
+}
+
 /*
  * this routine handles the overflow of the microsecond field
  *
@@ -151,6 +188,7 @@ void second_overflow(void)
 	 */
 	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
 #endif
+	tick_length = tick_length_base;
 }
 
 /*
@@ -204,14 +242,13 @@ void update_ntp_one_tick(void)
  */
 u64 current_tick_length(void)
 {
-	long delta_nsec;
 	u64 ret;
 
 	/* calculate the finest interval NTP will allow.
 	 *    ie: nanosecond value shifted by (SHIFT_SCALE - 10)
 	 */
-	delta_nsec = tick_nsec + adjtime_adjustment() * 1000;
-	ret = (u64)delta_nsec << TICK_LENGTH_SHIFT;
+	ret = tick_length;
+	ret += (u64)(adjtime_adjustment() * 1000) << TICK_LENGTH_SHIFT;
 	ret += (s64)time_adj << (TICK_LENGTH_SHIFT - (SHIFT_SCALE - 10));
 
 	return ret;
@@ -351,10 +388,11 @@ int do_adjtimex(struct timex *txc)
 		    time_freq = max(time_freq, -time_tolerance);
 		} /* STA_PLL */
 	    } /* txc->modes & ADJ_OFFSET */
-	    if (txc->modes & ADJ_TICK) {
+	    if (txc->modes & ADJ_TICK)
 		tick_usec = txc->tick;
-		tick_nsec = TICK_USEC_TO_NSEC(tick_usec);
-	    }
+
+	    if (txc->modes & ADJ_TICK)
+		    ntp_update_frequency();
 	} /* txc->modes */
 leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
 		result = TIME_ERROR;
diff --git a/kernel/timer.c b/kernel/timer.c
index 5fccc7cbf3b4..78d3fa10fcd6 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -562,12 +562,6 @@ found:
 
 /******************************************************************/
 
-/*
- * Timekeeping variables
- */
-unsigned long tick_usec = TICK_USEC; 		/* USER_HZ period (usec) */
-unsigned long tick_nsec = TICK_NSEC;		/* ACTHZ period (nsec) */
-
 /* 
  * The current time 
  * wall_to_monotonic is what we need to add to xtime (or xtime corrected 
@@ -757,10 +751,13 @@ void __init timekeeping_init(void)
 	unsigned long flags;
 
 	write_seqlock_irqsave(&xtime_lock, flags);
+
+	ntp_clear();
+
 	clock = clocksource_get_next();
 	clocksource_calculate_interval(clock, tick_nsec);
 	clock->cycle_last = clocksource_read(clock);
-	ntp_clear();
+
 	write_sequnlock_irqrestore(&xtime_lock, flags);
 }
 
-- 
cgit v1.2.3


From 3d3675cc3d04d7fd4bb11e8c1ea79e5ade4f5e44 Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:25 -0700
Subject: [PATCH] ntp: prescale time_offset

This converts time_offset into a scaled per tick value.  This avoids now
completely the crude compensation in second_overflow().

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h |  2 +-
 kernel/time/ntp.c     | 64 +++++++++++++--------------------------------------
 2 files changed, 17 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index b589c8218bb9..1cde6f6a2712 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -89,7 +89,7 @@
  * FINENSEC is 1 ns in SHIFT_UPDATE units of the time_phase variable.
  */
 #define SHIFT_SCALE 22		/* phase scale (shift) */
-#define SHIFT_UPDATE (SHIFT_KG + MAXTC) /* time offset scale (shift) */
+#define SHIFT_UPDATE (SHIFT_HZ + 1) /* time offset scale (shift) */
 #define SHIFT_USEC 16		/* frequency offset scale (shift) */
 #define FINENSEC (1L << (SHIFT_SCALE - 10)) /* ~1 ns in phase units */
 
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index ab21eb06e09b..238ce47ef09d 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -31,7 +31,7 @@ int tickadj = 500/HZ ? : 1;		/* microsecs */
 /* TIME_ERROR prevents overwriting the CMOS clock */
 int time_state = TIME_OK;		/* clock synchronization status	*/
 int time_status = STA_UNSYNC;		/* clock status bits		*/
-long time_offset;			/* time adjustment (us)		*/
+long time_offset;			/* time adjustment (ns)		*/
 long time_constant = 2;			/* pll time constant		*/
 long time_tolerance = MAXFREQ;		/* frequency tolerance (ppm)	*/
 long time_precision = 1;		/* clock precision (us)		*/
@@ -57,6 +57,7 @@ void ntp_clear(void)
 	ntp_update_frequency();
 
 	tick_length = tick_length_base;
+	time_offset = 0;
 }
 
 #define CLOCK_TICK_OVERFLOW	(LATCH * HZ - CLOCK_TICK_RATE)
@@ -83,7 +84,7 @@ void ntp_update_frequency(void)
  */
 void second_overflow(void)
 {
-	long ltemp, time_adj;
+	long time_adj;
 
 	/* Bump the maxerror field */
 	time_maxerror += time_tolerance >> SHIFT_USEC;
@@ -151,42 +152,14 @@ void second_overflow(void)
 	 * adjustment for each second is clamped so as to spread the adjustment
 	 * over not more than the number of seconds between updates.
 	 */
-	ltemp = time_offset;
-	if (!(time_status & STA_FLL))
-		ltemp = shift_right(ltemp, SHIFT_KG + time_constant);
-	ltemp = min(ltemp, (MAXPHASE / MINSEC) << SHIFT_UPDATE);
-	ltemp = max(ltemp, -(MAXPHASE / MINSEC) << SHIFT_UPDATE);
-	time_offset -= ltemp;
-	time_adj = ltemp << (SHIFT_SCALE - SHIFT_HZ - SHIFT_UPDATE);
-
-	/*
-	 * Compute the frequency estimate and additional phase adjustment due
-	 * to frequency error for the next second.
-	 */
-
-#if HZ == 100
-	/*
-	 * Compensate for (HZ==100) != (1 << SHIFT_HZ).  Add 25% and 3.125% to
-	 * get 128.125; => only 0.125% error (p. 14)
-	 */
-	time_adj += shift_right(time_adj, 2) + shift_right(time_adj, 5);
-#endif
-#if HZ == 250
-	/*
-	 * Compensate for (HZ==250) != (1 << SHIFT_HZ).  Add 1.5625% and
-	 * 0.78125% to get 255.85938; => only 0.05% error (p. 14)
-	 */
-	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
-#endif
-#if HZ == 1000
-	/*
-	 * Compensate for (HZ==1000) != (1 << SHIFT_HZ).  Add 1.5625% and
-	 * 0.78125% to get 1023.4375; => only 0.05% error (p. 14)
-	 */
-	time_adj += shift_right(time_adj, 6) + shift_right(time_adj, 7);
-#endif
 	tick_length = tick_length_base;
-	tick_length += (s64)time_adj << (TICK_LENGTH_SHIFT - (SHIFT_SCALE - 10));
+	time_adj = time_offset;
+	if (!(time_status & STA_FLL))
+		time_adj = shift_right(time_adj, SHIFT_KG + time_constant);
+	time_adj = min(time_adj, -((MAXPHASE / HZ) << SHIFT_UPDATE) / MINSEC);
+	time_adj = max(time_adj, ((MAXPHASE / HZ) << SHIFT_UPDATE) / MINSEC);
+	time_offset -= time_adj;
+	tick_length += (s64)time_adj << (TICK_LENGTH_SHIFT - SHIFT_UPDATE);
 }
 
 /*
@@ -347,12 +320,8 @@ int do_adjtimex(struct timex *txc)
 		     * Scale the phase adjustment and
 		     * clamp to the operating range.
 		     */
-		    if (ltemp > MAXPHASE)
-		        time_offset = MAXPHASE << SHIFT_UPDATE;
-		    else if (ltemp < -MAXPHASE)
-			time_offset = -(MAXPHASE << SHIFT_UPDATE);
-		    else
-		        time_offset = ltemp << SHIFT_UPDATE;
+		    time_offset = min(ltemp, MAXPHASE);
+		    time_offset = max(time_offset, -MAXPHASE);
 
 		    /*
 		     * Select whether the frequency is to be controlled
@@ -366,8 +335,7 @@ int do_adjtimex(struct timex *txc)
 		    time_reftime = xtime.tv_sec;
 		    if (time_status & STA_FLL) {
 		        if (mtemp >= MINSEC) {
-			    ltemp = (time_offset / mtemp) << (SHIFT_USEC -
-							      SHIFT_UPDATE);
+			    ltemp = ((time_offset << 12) / mtemp) << (SHIFT_USEC - 12);
 			    time_freq += shift_right(ltemp, SHIFT_KH);
 			} else /* calibration interval too short (p. 12) */
 				result = TIME_ERROR;
@@ -382,6 +350,7 @@ int do_adjtimex(struct timex *txc)
 		    }
 		    time_freq = min(time_freq, time_tolerance);
 		    time_freq = max(time_freq, -time_tolerance);
+		    time_offset = (time_offset * NSEC_PER_USEC / HZ) << SHIFT_UPDATE;
 		} /* STA_PLL */
 	    } /* txc->modes & ADJ_OFFSET */
 	    if (txc->modes & ADJ_TICK)
@@ -395,9 +364,8 @@ leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
 
 	if ((txc->modes & ADJ_OFFSET_SINGLESHOT) == ADJ_OFFSET_SINGLESHOT)
 	    txc->offset	   = save_adjust;
-	else {
-	    txc->offset = shift_right(time_offset, SHIFT_UPDATE);
-	}
+	else
+	    txc->offset    = shift_right(time_offset, SHIFT_UPDATE) * HZ / 1000;
 	txc->freq	   = time_freq;
 	txc->maxerror	   = time_maxerror;
 	txc->esterror	   = time_esterror;
-- 
cgit v1.2.3


From 8f807f8d2137ba728d22820103131038639b68a9 Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:25 -0700
Subject: [PATCH] ntp: add time_adjust to tick length

This folds update_ntp_one_tick() into second_overflow() and adds time_adjust
to the tick length, this makes time_next_adjust unnecessary.  This slightly
changes the adjtime() behaviour, instead of applying it to the next tick, it's
applied to the next second.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h |  1 -
 kernel/time/ntp.c     | 71 +++++++++++++--------------------------------------
 kernel/timer.c        |  2 --
 3 files changed, 18 insertions(+), 56 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index 1cde6f6a2712..b5f297e17668 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -217,7 +217,6 @@ extern long time_freq;		/* frequency offset (scaled ppm) */
 extern long time_reftime;	/* time at last adjustment (s) */
 
 extern long time_adjust;	/* The amount of adjtime left */
-extern long time_next_adjust;	/* Value for time_adjust at next tick */
 
 extern void ntp_clear(void);
 extern void ntp_update_frequency(void);
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 238ce47ef09d..65223b7ed810 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -22,8 +22,9 @@ unsigned long tick_usec = TICK_USEC; 		/* USER_HZ period (usec) */
 unsigned long tick_nsec;			/* ACTHZ period (nsec) */
 static u64 tick_length, tick_length_base;
 
-/* Don't completely fail for HZ > 500.  */
-int tickadj = 500/HZ ? : 1;		/* microsecs */
+#define MAX_TICKADJ		500		/* microsecs */
+#define MAX_TICKADJ_SCALED	(((u64)(MAX_TICKADJ * NSEC_PER_USEC) << \
+				  TICK_LENGTH_SHIFT) / HZ)
 
 /*
  * phase-lock loop variables
@@ -40,7 +41,6 @@ long time_esterror = NTP_PHASE_LIMIT;	/* estimated error (us)		*/
 long time_freq;				/* frequency offset (scaled ppm)*/
 long time_reftime;			/* time at last adjustment (s)	*/
 long time_adjust;
-long time_next_adjust;
 
 /**
  * ntp_clear - Clears the NTP state variables
@@ -160,46 +160,19 @@ void second_overflow(void)
 	time_adj = max(time_adj, ((MAXPHASE / HZ) << SHIFT_UPDATE) / MINSEC);
 	time_offset -= time_adj;
 	tick_length += (s64)time_adj << (TICK_LENGTH_SHIFT - SHIFT_UPDATE);
-}
-
-/*
- * Returns how many microseconds we need to add to xtime this tick
- * in doing an adjustment requested with adjtime.
- */
-static long adjtime_adjustment(void)
-{
-	long time_adjust_step;
-
-	time_adjust_step = time_adjust;
-	if (time_adjust_step) {
-		/*
-		 * We are doing an adjtime thing.  Prepare time_adjust_step to
-		 * be within bounds.  Note that a positive time_adjust means we
-		 * want the clock to run faster.
-		 *
-		 * Limit the amount of the step to be in the range
-		 * -tickadj .. +tickadj
-		 */
-		time_adjust_step = min(time_adjust_step, (long)tickadj);
-		time_adjust_step = max(time_adjust_step, (long)-tickadj);
-	}
-	return time_adjust_step;
-}
 
-/* in the NTP reference this is called "hardclock()" */
-void update_ntp_one_tick(void)
-{
-	long time_adjust_step;
-
-	time_adjust_step = adjtime_adjustment();
-	if (time_adjust_step)
-		/* Reduce by this step the amount of time left  */
-		time_adjust -= time_adjust_step;
-
-	/* Changes by adjtime() do not take effect till next tick. */
-	if (time_next_adjust != 0) {
-		time_adjust = time_next_adjust;
-		time_next_adjust = 0;
+	if (unlikely(time_adjust)) {
+		if (time_adjust > MAX_TICKADJ) {
+			time_adjust -= MAX_TICKADJ;
+			tick_length += MAX_TICKADJ_SCALED;
+		} else if (time_adjust < -MAX_TICKADJ) {
+			time_adjust += MAX_TICKADJ;
+			tick_length -= MAX_TICKADJ_SCALED;
+		} else {
+			time_adjust = 0;
+			tick_length += (s64)(time_adjust * NSEC_PER_USEC /
+					     HZ) << TICK_LENGTH_SHIFT;
+		}
 	}
 }
 
@@ -213,14 +186,7 @@ void update_ntp_one_tick(void)
  */
 u64 current_tick_length(void)
 {
-	u64 ret;
-
-	/* calculate the finest interval NTP will allow.
-	 */
-	ret = tick_length;
-	ret += (u64)(adjtime_adjustment() * 1000) << TICK_LENGTH_SHIFT;
-
-	return ret;
+	return tick_length;
 }
 
 
@@ -263,7 +229,7 @@ int do_adjtimex(struct timex *txc)
 	result = time_state;	/* mostly `TIME_OK' */
 
 	/* Save for later - semantics of adjtime is to return old value */
-	save_adjust = time_next_adjust ? time_next_adjust : time_adjust;
+	save_adjust = time_adjust;
 
 #if 0	/* STA_CLOCKERR is never set yet */
 	time_status &= ~STA_CLOCKERR;		/* reset STA_CLOCKERR */
@@ -310,8 +276,7 @@ int do_adjtimex(struct timex *txc)
 	    if (txc->modes & ADJ_OFFSET) {	/* values checked earlier */
 		if (txc->modes == ADJ_OFFSET_SINGLESHOT) {
 		    /* adjtime() is independent from ntp_adjtime() */
-		    if ((time_next_adjust = txc->offset) == 0)
-			 time_adjust = 0;
+		    time_adjust = txc->offset;
 		}
 		else if (time_status & STA_PLL) {
 		    ltemp = txc->offset;
diff --git a/kernel/timer.c b/kernel/timer.c
index 78d3fa10fcd6..654dd5df2417 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -937,8 +937,6 @@ static void update_wall_time(void)
 		/* interpolator bits */
 		time_interpolator_update(clock->xtime_interval
 						>> clock->shift);
-		/* increment the NTP state machine */
-		update_ntp_one_tick();
 
 		/* accumulate error between NTP and clock interval */
 		clock->error += current_tick_length();
-- 
cgit v1.2.3


From 97eebe138caaf78354b1fad233e63bafdcc4fd54 Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:26 -0700
Subject: [PATCH] ntp: remove time_tolerance

time_tolerance isn't changed at all in the kernel, so simply remove it, this
simplifies the next patch, as it avoids a number of conversions.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h | 1 -
 kernel/time/ntp.c     | 9 ++++-----
 2 files changed, 4 insertions(+), 6 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index b5f297e17668..7715b4c0caf9 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -208,7 +208,6 @@ extern int time_state;		/* clock status */
 extern int time_status;		/* clock synchronization status bits */
 extern long time_offset;	/* time adjustment (us) */
 extern long time_constant;	/* pll time constant */
-extern long time_tolerance;	/* frequency tolerance (ppm) */
 extern long time_precision;	/* clock precision (us) */
 extern long time_maxerror;	/* maximum error */
 extern long time_esterror;	/* estimated error */
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 65223b7ed810..af7563f5d4e2 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -34,7 +34,6 @@ int time_state = TIME_OK;		/* clock synchronization status	*/
 int time_status = STA_UNSYNC;		/* clock status bits		*/
 long time_offset;			/* time adjustment (ns)		*/
 long time_constant = 2;			/* pll time constant		*/
-long time_tolerance = MAXFREQ;		/* frequency tolerance (ppm)	*/
 long time_precision = 1;		/* clock precision (us)		*/
 long time_maxerror = NTP_PHASE_LIMIT;	/* maximum error (us)		*/
 long time_esterror = NTP_PHASE_LIMIT;	/* estimated error (us)		*/
@@ -87,7 +86,7 @@ void second_overflow(void)
 	long time_adj;
 
 	/* Bump the maxerror field */
-	time_maxerror += time_tolerance >> SHIFT_USEC;
+	time_maxerror += MAXFREQ >> SHIFT_USEC;
 	if (time_maxerror > NTP_PHASE_LIMIT) {
 		time_maxerror = NTP_PHASE_LIMIT;
 		time_status |= STA_UNSYNC;
@@ -313,8 +312,8 @@ int do_adjtimex(struct timex *txc)
 			} else /* calibration interval too long (p. 12) */
 				result = TIME_ERROR;
 		    }
-		    time_freq = min(time_freq, time_tolerance);
-		    time_freq = max(time_freq, -time_tolerance);
+		    time_freq = min(time_freq, MAXFREQ);
+		    time_freq = max(time_freq, -MAXFREQ);
 		    time_offset = (time_offset * NSEC_PER_USEC / HZ) << SHIFT_UPDATE;
 		} /* STA_PLL */
 	    } /* txc->modes & ADJ_OFFSET */
@@ -337,7 +336,7 @@ leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
 	txc->status	   = time_status;
 	txc->constant	   = time_constant;
 	txc->precision	   = time_precision;
-	txc->tolerance	   = time_tolerance;
+	txc->tolerance	   = MAXFREQ;
 	txc->tick	   = tick_usec;
 
 	/* PPS is not implemented, so these are zero */
-- 
cgit v1.2.3


From 04b617e71e363e640e88be1e43f53fa6a3afef9f Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:27 -0700
Subject: [PATCH] ntp: convert time_freq to nsec value

This converts time_freq to a scaled nsec value and adds around 6bit of extra
resolution.  This pushes the time_freq to its 32bit limits so the calculatons
have to be done with 64bit.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h |  2 ++
 kernel/time/ntp.c     | 36 ++++++++++++++++++++++--------------
 2 files changed, 24 insertions(+), 14 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index 7715b4c0caf9..671609ee1a3d 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -91,10 +91,12 @@
 #define SHIFT_SCALE 22		/* phase scale (shift) */
 #define SHIFT_UPDATE (SHIFT_HZ + 1) /* time offset scale (shift) */
 #define SHIFT_USEC 16		/* frequency offset scale (shift) */
+#define SHIFT_NSEC 12		/* kernel frequency offset scale */
 #define FINENSEC (1L << (SHIFT_SCALE - 10)) /* ~1 ns in phase units */
 
 #define MAXPHASE 512000L        /* max phase error (us) */
 #define MAXFREQ (512L << SHIFT_USEC)  /* max frequency error (ppm) */
+#define MAXFREQ_NSEC (512000L << SHIFT_NSEC) /* max frequency error (ppb) */
 #define MINSEC 16L              /* min interval between updates (s) */
 #define MAXSEC 1200L            /* max interval between updates (s) */
 #define	NTP_PHASE_LIMIT	(MAXPHASE << 5)	/* beyond max. dispersion */
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index af7563f5d4e2..9137b54613e0 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -66,7 +66,7 @@ void ntp_update_frequency(void)
 {
 	tick_length_base = (u64)(tick_usec * NSEC_PER_USEC * USER_HZ) << TICK_LENGTH_SHIFT;
 	tick_length_base += (s64)CLOCK_TICK_ADJUST << TICK_LENGTH_SHIFT;
-	tick_length_base += ((s64)time_freq * NSEC_PER_USEC) << (TICK_LENGTH_SHIFT - SHIFT_USEC);
+	tick_length_base += (s64)time_freq << (TICK_LENGTH_SHIFT - SHIFT_NSEC);
 
 	do_div(tick_length_base, HZ);
 
@@ -200,6 +200,7 @@ void __attribute__ ((weak)) notify_arch_cmos_timer(void)
 int do_adjtimex(struct timex *txc)
 {
 	long ltemp, mtemp, save_adjust;
+	s64 freq_adj;
 	int result;
 
 	/* In order to modify anything, you gotta be super-user! */
@@ -245,7 +246,7 @@ int do_adjtimex(struct timex *txc)
 		    result = -EINVAL;
 		    goto leave;
 		}
-		time_freq = txc->freq;
+		time_freq = ((s64)txc->freq * NSEC_PER_USEC) >> (SHIFT_USEC - SHIFT_NSEC);
 	    }
 
 	    if (txc->modes & ADJ_MAXERROR) {
@@ -278,14 +279,14 @@ int do_adjtimex(struct timex *txc)
 		    time_adjust = txc->offset;
 		}
 		else if (time_status & STA_PLL) {
-		    ltemp = txc->offset;
+		    ltemp = txc->offset * NSEC_PER_USEC;
 
 		    /*
 		     * Scale the phase adjustment and
 		     * clamp to the operating range.
 		     */
-		    time_offset = min(ltemp, MAXPHASE);
-		    time_offset = max(time_offset, -MAXPHASE);
+		    time_offset = min(ltemp, MAXPHASE * NSEC_PER_USEC);
+		    time_offset = max(time_offset, -MAXPHASE * NSEC_PER_USEC);
 
 		    /*
 		     * Select whether the frequency is to be controlled
@@ -297,24 +298,31 @@ int do_adjtimex(struct timex *txc)
 		        time_reftime = xtime.tv_sec;
 		    mtemp = xtime.tv_sec - time_reftime;
 		    time_reftime = xtime.tv_sec;
+		    freq_adj = 0;
 		    if (time_status & STA_FLL) {
 		        if (mtemp >= MINSEC) {
-			    ltemp = ((time_offset << 12) / mtemp) << (SHIFT_USEC - 12);
-			    time_freq += shift_right(ltemp, SHIFT_KH);
+			    freq_adj = (s64)time_offset << (SHIFT_NSEC - SHIFT_KH);
+			    if (time_offset < 0) {
+				freq_adj = -freq_adj;
+				do_div(freq_adj, mtemp);
+				freq_adj = -freq_adj;
+			    } else
+				do_div(freq_adj, mtemp);
 			} else /* calibration interval too short (p. 12) */
 				result = TIME_ERROR;
 		    } else {	/* PLL mode */
 		        if (mtemp < MAXSEC) {
-			    ltemp *= mtemp;
-			    time_freq += shift_right(ltemp,(time_constant +
+			    freq_adj = (s64)ltemp * mtemp;
+			    freq_adj = shift_right(freq_adj,(time_constant +
 						       time_constant +
-						       SHIFT_KF - SHIFT_USEC));
+						       SHIFT_KF - SHIFT_NSEC));
 			} else /* calibration interval too long (p. 12) */
 				result = TIME_ERROR;
 		    }
-		    time_freq = min(time_freq, MAXFREQ);
-		    time_freq = max(time_freq, -MAXFREQ);
-		    time_offset = (time_offset * NSEC_PER_USEC / HZ) << SHIFT_UPDATE;
+		    freq_adj += time_freq;
+		    freq_adj = min(freq_adj, (s64)MAXFREQ_NSEC);
+		    time_freq = max(freq_adj, (s64)-MAXFREQ_NSEC);
+		    time_offset = (time_offset / HZ) << SHIFT_UPDATE;
 		} /* STA_PLL */
 	    } /* txc->modes & ADJ_OFFSET */
 	    if (txc->modes & ADJ_TICK)
@@ -330,7 +338,7 @@ leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
 	    txc->offset	   = save_adjust;
 	else
 	    txc->offset    = shift_right(time_offset, SHIFT_UPDATE) * HZ / 1000;
-	txc->freq	   = time_freq;
+	txc->freq	   = (time_freq / NSEC_PER_USEC) << (SHIFT_USEC - SHIFT_NSEC);
 	txc->maxerror	   = time_maxerror;
 	txc->esterror	   = time_esterror;
 	txc->status	   = time_status;
-- 
cgit v1.2.3


From f19923937321244e7dc334767eb4b67e0e3d5c74 Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:28 -0700
Subject: [PATCH] ntp: convert to the NTP4 reference model

This converts the kernel ntp model into a model which matches the nanokernel
reference implementations.  The previous patches already increased the
resolution and precision of the computations, so that this conversion becomes
quite simple.

<linux@horizon.com> explains:

The original NTP kernel interface was defined in units of microseconds.
That's what Linux implements.  As computers have gotten faster and can now
split microseconds easily, a new kernel interface using nanosecond units was
defined ("the nanokernel", confusing as that name is to OS hackers), and
there's an STA_NANO bit in the adjtimex() status field to tell the application
which units it's using.

The current ntpd supports both, but Linux loses some possible timing
resolution because of quantization effects, and the ntpd hackers would really
like to be able to drop the backwards compatibility code.

Ulrich Windl has been maintaining a patch set to do the conversion for years,
but it's hard to keep in sync.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h | 11 +++++------
 kernel/time/ntp.c     | 51 +++++++++++++++++++--------------------------------
 2 files changed, 24 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index 671609ee1a3d..ac808f13fa0e 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -69,10 +69,9 @@
  * zero to MAXTC, the PLL will converge in 15 minutes to 16 hours,
  * respectively.
  */
-#define SHIFT_KG 6		/* phase factor (shift) */
-#define SHIFT_KF 16		/* PLL frequency factor (shift) */
-#define SHIFT_KH 2		/* FLL frequency factor (shift) */
-#define MAXTC 6			/* maximum time constant (shift) */
+#define SHIFT_PLL	4	/* PLL frequency factor (shift) */
+#define SHIFT_FLL	2	/* FLL frequency factor (shift) */
+#define MAXTC		10	/* maximum time constant (shift) */
 
 /*
  * The SHIFT_SCALE define establishes the decimal point of the time_phase
@@ -97,8 +96,8 @@
 #define MAXPHASE 512000L        /* max phase error (us) */
 #define MAXFREQ (512L << SHIFT_USEC)  /* max frequency error (ppm) */
 #define MAXFREQ_NSEC (512000L << SHIFT_NSEC) /* max frequency error (ppb) */
-#define MINSEC 16L              /* min interval between updates (s) */
-#define MAXSEC 1200L            /* max interval between updates (s) */
+#define MINSEC 256		/* min interval between updates (s) */
+#define MAXSEC 2048		/* max interval between updates (s) */
 #define	NTP_PHASE_LIMIT	(MAXPHASE << 5)	/* beyond max. dispersion */
 
 /*
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 9137b54613e0..1ab5e9d7fa50 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -145,18 +145,11 @@ void second_overflow(void)
 	}
 
 	/*
-	 * Compute the phase adjustment for the next second. In PLL mode, the
-	 * offset is reduced by a fixed factor times the time constant. In FLL
-	 * mode the offset is used directly. In either mode, the maximum phase
-	 * adjustment for each second is clamped so as to spread the adjustment
-	 * over not more than the number of seconds between updates.
+	 * Compute the phase adjustment for the next second. The offset is
+	 * reduced by a fixed factor times the time constant.
 	 */
 	tick_length = tick_length_base;
-	time_adj = time_offset;
-	if (!(time_status & STA_FLL))
-		time_adj = shift_right(time_adj, SHIFT_KG + time_constant);
-	time_adj = min(time_adj, -((MAXPHASE / HZ) << SHIFT_UPDATE) / MINSEC);
-	time_adj = max(time_adj, ((MAXPHASE / HZ) << SHIFT_UPDATE) / MINSEC);
+	time_adj = shift_right(time_offset, SHIFT_PLL + time_constant);
 	time_offset -= time_adj;
 	tick_length += (s64)time_adj << (TICK_LENGTH_SHIFT - SHIFT_UPDATE);
 
@@ -200,7 +193,7 @@ void __attribute__ ((weak)) notify_arch_cmos_timer(void)
 int do_adjtimex(struct timex *txc)
 {
 	long ltemp, mtemp, save_adjust;
-	s64 freq_adj;
+	s64 freq_adj, temp64;
 	int result;
 
 	/* In order to modify anything, you gotta be super-user! */
@@ -270,7 +263,7 @@ int do_adjtimex(struct timex *txc)
 		    result = -EINVAL;
 		    goto leave;
 		}
-		time_constant = txc->constant;
+		time_constant = min(txc->constant + 4, (long)MAXTC);
 	    }
 
 	    if (txc->modes & ADJ_OFFSET) {	/* values checked earlier */
@@ -298,26 +291,20 @@ int do_adjtimex(struct timex *txc)
 		        time_reftime = xtime.tv_sec;
 		    mtemp = xtime.tv_sec - time_reftime;
 		    time_reftime = xtime.tv_sec;
-		    freq_adj = 0;
-		    if (time_status & STA_FLL) {
-		        if (mtemp >= MINSEC) {
-			    freq_adj = (s64)time_offset << (SHIFT_NSEC - SHIFT_KH);
-			    if (time_offset < 0) {
-				freq_adj = -freq_adj;
-				do_div(freq_adj, mtemp);
-				freq_adj = -freq_adj;
-			    } else
-				do_div(freq_adj, mtemp);
-			} else /* calibration interval too short (p. 12) */
-				result = TIME_ERROR;
-		    } else {	/* PLL mode */
-		        if (mtemp < MAXSEC) {
-			    freq_adj = (s64)ltemp * mtemp;
-			    freq_adj = shift_right(freq_adj,(time_constant +
-						       time_constant +
-						       SHIFT_KF - SHIFT_NSEC));
-			} else /* calibration interval too long (p. 12) */
-				result = TIME_ERROR;
+
+		    freq_adj = (s64)time_offset * mtemp;
+		    freq_adj = shift_right(freq_adj, time_constant * 2 +
+					   (SHIFT_PLL + 2) * 2 - SHIFT_NSEC);
+		    if (mtemp >= MINSEC && (time_status & STA_FLL || mtemp > MAXSEC)) {
+			temp64 = (s64)time_offset << (SHIFT_NSEC - SHIFT_FLL);
+			if (time_offset < 0) {
+			    temp64 = -temp64;
+			    do_div(temp64, mtemp);
+			    freq_adj -= temp64;
+			} else {
+			    do_div(temp64, mtemp);
+			    freq_adj += temp64;
+			}
 		    }
 		    freq_adj += time_freq;
 		    freq_adj = min(freq_adj, (s64)MAXFREQ_NSEC);
-- 
cgit v1.2.3


From 0883d899ef862c1b0f8b2c2d38098470c193a3dd Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Sat, 30 Sep 2006 23:28:29 -0700
Subject: [PATCH] ntp: cleanup defines and comments

Remove a few unused defines and remove obsolete information from comments.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/timex.h    |  5 -----
 include/asm-m32r/timex.h   |  3 ---
 include/asm-sh64/timex.h   |  3 ---
 include/asm-xtensa/timex.h |  3 ---
 include/linux/timex.h      | 13 +++----------
 5 files changed, 3 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/include/asm-frv/timex.h b/include/asm-frv/timex.h
index 2aa562fa067b..a89bddefdacf 100644
--- a/include/asm-frv/timex.h
+++ b/include/asm-frv/timex.h
@@ -6,11 +6,6 @@
 #define CLOCK_TICK_RATE		1193180 /* Underlying HZ */
 #define CLOCK_TICK_FACTOR	20	/* Factor of both 1000000 and CLOCK_TICK_RATE */
 
-#define FINETUNE							\
-((((((long)LATCH * HZ - CLOCK_TICK_RATE) << SHIFT_HZ) *			\
-   (1000000/CLOCK_TICK_FACTOR) / (CLOCK_TICK_RATE/CLOCK_TICK_FACTOR))	\
-  << (SHIFT_SCALE-SHIFT_HZ)) / HZ)
-
 typedef unsigned long cycles_t;
 
 static inline cycles_t get_cycles(void)
diff --git a/include/asm-m32r/timex.h b/include/asm-m32r/timex.h
index e89bfd17db51..019441c1d7a0 100644
--- a/include/asm-m32r/timex.h
+++ b/include/asm-m32r/timex.h
@@ -12,9 +12,6 @@
 
 #define CLOCK_TICK_RATE	(CONFIG_BUS_CLOCK / CONFIG_TIMER_DIVIDE)
 #define CLOCK_TICK_FACTOR	20	/* Factor of both 1000000 and CLOCK_TICK_RATE */
-#define FINETUNE ((((((long)LATCH * HZ - CLOCK_TICK_RATE) << SHIFT_HZ) * \
-	(1000000/CLOCK_TICK_FACTOR) / (CLOCK_TICK_RATE/CLOCK_TICK_FACTOR)) \
-		<< (SHIFT_SCALE-SHIFT_HZ)) / HZ)
 
 #ifdef __KERNEL__
 /*
diff --git a/include/asm-sh64/timex.h b/include/asm-sh64/timex.h
index af0b79269661..163e2b62fe27 100644
--- a/include/asm-sh64/timex.h
+++ b/include/asm-sh64/timex.h
@@ -17,9 +17,6 @@
 
 #define CLOCK_TICK_RATE	1193180 /* Underlying HZ */
 #define CLOCK_TICK_FACTOR	20	/* Factor of both 1000000 and CLOCK_TICK_RATE */
-#define FINETUNE ((((((long)LATCH * HZ - CLOCK_TICK_RATE) << SHIFT_HZ) * \
-	(1000000/CLOCK_TICK_FACTOR) / (CLOCK_TICK_RATE/CLOCK_TICK_FACTOR)) \
-		<< (SHIFT_SCALE-SHIFT_HZ)) / HZ)
 
 typedef unsigned long cycles_t;
 
diff --git a/include/asm-xtensa/timex.h b/include/asm-xtensa/timex.h
index d14a3755a12b..c7b705e66655 100644
--- a/include/asm-xtensa/timex.h
+++ b/include/asm-xtensa/timex.h
@@ -31,9 +31,6 @@
 
 #define CLOCK_TICK_RATE 	1193180	/* (everyone is using this value) */
 #define CLOCK_TICK_FACTOR       20 /* Factor of both 10^6 and CLOCK_TICK_RATE */
-#define FINETUNE ((((((long)LATCH * HZ - CLOCK_TICK_RATE) << SHIFT_HZ) * \
-	(1000000/CLOCK_TICK_FACTOR) / (CLOCK_TICK_RATE/CLOCK_TICK_FACTOR)) \
-		<< (SHIFT_SCALE-SHIFT_HZ)) / HZ)
 
 #ifdef CONFIG_XTENSA_CALIBRATE_CCOUNT
 extern unsigned long ccount_per_jiffy;
diff --git a/include/linux/timex.h b/include/linux/timex.h
index ac808f13fa0e..261381b5da82 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -74,24 +74,17 @@
 #define MAXTC		10	/* maximum time constant (shift) */
 
 /*
- * The SHIFT_SCALE define establishes the decimal point of the time_phase
- * variable which serves as an extension to the low-order bits of the
- * system clock variable. The SHIFT_UPDATE define establishes the decimal
- * point of the time_offset variable which represents the current offset
- * with respect to standard time. The FINENSEC define represents 1 nsec in
- * scaled units.
+ * The SHIFT_UPDATE define establishes the decimal point of the
+ * time_offset variable which represents the current offset with
+ * respect to standard time.
  *
  * SHIFT_USEC defines the scaling (shift) of the time_freq and
  * time_tolerance variables, which represent the current frequency
  * offset and maximum frequency tolerance.
- *
- * FINENSEC is 1 ns in SHIFT_UPDATE units of the time_phase variable.
  */
-#define SHIFT_SCALE 22		/* phase scale (shift) */
 #define SHIFT_UPDATE (SHIFT_HZ + 1) /* time offset scale (shift) */
 #define SHIFT_USEC 16		/* frequency offset scale (shift) */
 #define SHIFT_NSEC 12		/* kernel frequency offset scale */
-#define FINENSEC (1L << (SHIFT_SCALE - 10)) /* ~1 ns in phase units */
 
 #define MAXPHASE 512000L        /* max phase error (us) */
 #define MAXFREQ (512L << SHIFT_USEC)  /* max frequency error (ppm) */
-- 
cgit v1.2.3


From 70bc42f90a3f4721c89dbe865e6c95da8565b41c Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sat, 30 Sep 2006 23:28:29 -0700
Subject: [PATCH] kernel/time/ntp.c: possible cleanups

This patch contains the following possible cleanups:
- make the following needlessly global function static:
  - ntp_update_frequency()
- make the following needlessly global variables static:
  - time_state
  - time_offset
  - time_constant
  - time_reftime
- remove the following read-only global variable:
  - time_precision

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h |  6 ------
 kernel/time/ntp.c     | 40 ++++++++++++++++++++--------------------
 2 files changed, 20 insertions(+), 26 deletions(-)

(limited to 'include')

diff --git a/include/linux/timex.h b/include/linux/timex.h
index 261381b5da82..049dfe4a11f2 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -198,21 +198,15 @@ extern int tickadj;			/* amount of adjustment per tick */
 /*
  * phase-lock loop variables
  */
-extern int time_state;		/* clock status */
 extern int time_status;		/* clock synchronization status bits */
-extern long time_offset;	/* time adjustment (us) */
-extern long time_constant;	/* pll time constant */
-extern long time_precision;	/* clock precision (us) */
 extern long time_maxerror;	/* maximum error */
 extern long time_esterror;	/* estimated error */
 
 extern long time_freq;		/* frequency offset (scaled ppm) */
-extern long time_reftime;	/* time at last adjustment (s) */
 
 extern long time_adjust;	/* The amount of adjtime left */
 
 extern void ntp_clear(void);
-extern void ntp_update_frequency(void);
 
 /**
  * ntp_synced - Returns 1 if the NTP status is not UNSYNC
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 1ab5e9d7fa50..47195fa0ec4f 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -30,17 +30,31 @@ static u64 tick_length, tick_length_base;
  * phase-lock loop variables
  */
 /* TIME_ERROR prevents overwriting the CMOS clock */
-int time_state = TIME_OK;		/* clock synchronization status	*/
+static int time_state = TIME_OK;	/* clock synchronization status	*/
 int time_status = STA_UNSYNC;		/* clock status bits		*/
-long time_offset;			/* time adjustment (ns)		*/
-long time_constant = 2;			/* pll time constant		*/
-long time_precision = 1;		/* clock precision (us)		*/
+static long time_offset;		/* time adjustment (ns)		*/
+static long time_constant = 2;		/* pll time constant		*/
 long time_maxerror = NTP_PHASE_LIMIT;	/* maximum error (us)		*/
 long time_esterror = NTP_PHASE_LIMIT;	/* estimated error (us)		*/
 long time_freq;				/* frequency offset (scaled ppm)*/
-long time_reftime;			/* time at last adjustment (s)	*/
+static long time_reftime;		/* time at last adjustment (s)	*/
 long time_adjust;
 
+#define CLOCK_TICK_OVERFLOW	(LATCH * HZ - CLOCK_TICK_RATE)
+#define CLOCK_TICK_ADJUST	(((s64)CLOCK_TICK_OVERFLOW * NSEC_PER_SEC) / \
+					(s64)CLOCK_TICK_RATE)
+
+static void ntp_update_frequency(void)
+{
+	tick_length_base = (u64)(tick_usec * NSEC_PER_USEC * USER_HZ) << TICK_LENGTH_SHIFT;
+	tick_length_base += (s64)CLOCK_TICK_ADJUST << TICK_LENGTH_SHIFT;
+	tick_length_base += (s64)time_freq << (TICK_LENGTH_SHIFT - SHIFT_NSEC);
+
+	do_div(tick_length_base, HZ);
+
+	tick_nsec = tick_length_base >> TICK_LENGTH_SHIFT;
+}
+
 /**
  * ntp_clear - Clears the NTP state variables
  *
@@ -59,20 +73,6 @@ void ntp_clear(void)
 	time_offset = 0;
 }
 
-#define CLOCK_TICK_OVERFLOW	(LATCH * HZ - CLOCK_TICK_RATE)
-#define CLOCK_TICK_ADJUST	(((s64)CLOCK_TICK_OVERFLOW * NSEC_PER_SEC) / (s64)CLOCK_TICK_RATE)
-
-void ntp_update_frequency(void)
-{
-	tick_length_base = (u64)(tick_usec * NSEC_PER_USEC * USER_HZ) << TICK_LENGTH_SHIFT;
-	tick_length_base += (s64)CLOCK_TICK_ADJUST << TICK_LENGTH_SHIFT;
-	tick_length_base += (s64)time_freq << (TICK_LENGTH_SHIFT - SHIFT_NSEC);
-
-	do_div(tick_length_base, HZ);
-
-	tick_nsec = tick_length_base >> TICK_LENGTH_SHIFT;
-}
-
 /*
  * this routine handles the overflow of the microsecond field
  *
@@ -330,7 +330,7 @@ leave:	if ((time_status & (STA_UNSYNC|STA_CLOCKERR)) != 0)
 	txc->esterror	   = time_esterror;
 	txc->status	   = time_status;
 	txc->constant	   = time_constant;
-	txc->precision	   = time_precision;
+	txc->precision	   = 1;
 	txc->tolerance	   = MAXFREQ;
 	txc->tick	   = tick_usec;
 
-- 
cgit v1.2.3


From 8ef386092d7c2891bd7acefb2a87f878f7e9a0d6 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 30 Sep 2006 23:28:31 -0700
Subject: [PATCH] kill wall_jiffies

With 2.6.18-rc4-mm2, now wall_jiffies will always be the same as jiffies.
So we can kill wall_jiffies completely.

This is just a cleanup and logically should not change any real behavior
except for one thing: RTC updating code in (old) ppc and xtensa use a
condition "jiffies - wall_jiffies == 1".  This condition is never met so I
suppose it is just a bug.  I just remove that condition only instead of
kill the whole "if" block.

[heiko.carstens@de.ibm.com: s390 build fix and cleanup]
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: Andi Kleen <ak@muc.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ian Molton <spyro@f2s.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Hirokazu Takata <takata.hirokazu@renesas.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp>
Cc: Richard Curnow <rc@rc0.org.uk>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp>
Cc: Chris Zankel <chris@zankel.net>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/kernel/time.c         | 15 +++++----------
 arch/arm/kernel/time.c           | 10 +---------
 arch/arm26/kernel/time.c         | 12 ++----------
 arch/cris/kernel/time.c          |  7 -------
 arch/i386/kernel/time.c          |  3 ---
 arch/ia64/kernel/time.c          |  2 --
 arch/m32r/kernel/time.c          | 11 +----------
 arch/m68k/kernel/time.c          | 16 +++-------------
 arch/m68knommu/kernel/time.c     |  7 +------
 arch/mips/kernel/time.c          | 12 +-----------
 arch/mips/sgi-ip27/ip27-timer.c  |  2 --
 arch/parisc/kernel/time.c        |  5 +----
 arch/powerpc/kernel/time.c       |  7 -------
 arch/ppc/kernel/time.c           | 11 +++--------
 arch/s390/kernel/time.c          |  7 ++-----
 arch/sh/kernel/time.c            | 10 +---------
 arch/sh64/kernel/time.c          | 12 +-----------
 arch/sparc/kernel/pcic.c         | 16 ++--------------
 arch/sparc/kernel/time.c         | 18 +++---------------
 arch/sparc64/kernel/time.c       |  2 --
 arch/x86_64/kernel/time.c        | 12 +++---------
 arch/x86_64/kernel/vmlinux.lds.S |  3 ---
 arch/x86_64/kernel/vsyscall.c    |  3 +--
 arch/xtensa/kernel/time.c        | 13 +++----------
 include/asm-x86_64/vsyscall.h    |  3 ---
 kernel/timer.c                   |  6 +-----
 26 files changed, 35 insertions(+), 190 deletions(-)

(limited to 'include')

diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c
index 7c1e44420a78..581ddcc22fc5 100644
--- a/arch/alpha/kernel/time.c
+++ b/arch/alpha/kernel/time.c
@@ -54,8 +54,6 @@
 #include "proto.h"
 #include "irq_impl.h"
 
-extern unsigned long wall_jiffies;	/* kernel/timer.c */
-
 static int set_rtc_mmss(unsigned long);
 
 DEFINE_SPINLOCK(rtc_lock);
@@ -413,7 +411,7 @@ void
 do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
-	unsigned long sec, usec, lost, seq;
+	unsigned long sec, usec, seq;
 	unsigned long delta_cycles, delta_usec, partial_tick;
 
 	do {
@@ -423,14 +421,13 @@ do_gettimeofday(struct timeval *tv)
 		sec = xtime.tv_sec;
 		usec = (xtime.tv_nsec / 1000);
 		partial_tick = state.partial_tick;
-		lost = jiffies - wall_jiffies;
 
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
 
 #ifdef CONFIG_SMP
 	/* Until and unless we figure out how to get cpu cycle counters
 	   in sync and keep them there, we can't use the rpcc tricks.  */
-	delta_usec = lost * (1000000 / HZ);
+	delta_usec = 0;
 #else
 	/*
 	 * usec = cycles * ticks_per_cycle * 2**48 * 1e6 / (2**48 * ticks)
@@ -446,8 +443,7 @@ do_gettimeofday(struct timeval *tv)
 	 */
 
 	delta_usec = (delta_cycles * state.scaled_ticks_per_cycle 
-		      + partial_tick
-		      + (lost << FIX_SHIFT)) * 15625;
+		      + partial_tick) * 15625;
 	delta_usec = ((delta_usec / ((1UL << (FIX_SHIFT-6-1)) * HZ)) + 1) / 2;
 #endif
 
@@ -480,12 +476,11 @@ do_settimeofday(struct timespec *tv)
 	   time.  Without this, a full-tick error is possible.  */
 
 #ifdef CONFIG_SMP
-	delta_nsec = (jiffies - wall_jiffies) * (NSEC_PER_SEC / HZ);
+	delta_nsec = 0;
 #else
 	delta_nsec = rpcc() - state.last_time;
 	delta_nsec = (delta_nsec * state.scaled_ticks_per_cycle 
-		      + state.partial_tick
-		      + ((jiffies - wall_jiffies) << FIX_SHIFT)) * 15625;
+		      + state.partial_tick) * 15625;
 	delta_nsec = ((delta_nsec / ((1UL << (FIX_SHIFT-6-1)) * HZ)) + 1) / 2;
 	delta_nsec *= 1000;
 #endif
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index f7d5165796ef..b030320b17c7 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -37,8 +37,6 @@
  */
 struct sys_timer *system_timer;
 
-extern unsigned long wall_jiffies;
-
 /* this needs a better home */
 DEFINE_SPINLOCK(rtc_lock);
 
@@ -237,16 +235,11 @@ void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
 	unsigned long seq;
-	unsigned long usec, sec, lost;
+	unsigned long usec, sec;
 
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 		usec = system_timer->offset();
-
-		lost = jiffies - wall_jiffies;
-		if (lost)
-			usec += lost * USECS_PER_JIFFY;
-
 		sec = xtime.tv_sec;
 		usec += xtime.tv_nsec / 1000;
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
@@ -279,7 +272,6 @@ int do_settimeofday(struct timespec *tv)
 	 * done, and then undo it!
 	 */
 	nsec -= system_timer->offset() * NSEC_PER_USEC;
-	nsec -= (jiffies - wall_jiffies) * TICK_NSEC;
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/arm26/kernel/time.c b/arch/arm26/kernel/time.c
index 80adbd005fc5..1206469b2b86 100644
--- a/arch/arm26/kernel/time.c
+++ b/arch/arm26/kernel/time.c
@@ -33,8 +33,6 @@
 #include <asm/irq.h>
 #include <asm/ioc.h>
 
-extern unsigned long wall_jiffies;
-
 /* this needs a better home */
 DEFINE_SPINLOCK(rtc_lock);
 
@@ -136,16 +134,11 @@ void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
 	unsigned long seq;
-	unsigned long usec, sec, lost;
+	unsigned long usec, sec;
 
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 		usec = gettimeoffset();
-
-		lost = jiffies - wall_jiffies;
-		if (lost)
-			usec += lost * USECS_PER_JIFFY;
-
 		sec = xtime.tv_sec;
 		usec += xtime.tv_nsec / 1000;
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
@@ -174,8 +167,7 @@ int do_settimeofday(struct timespec *tv)
 	 * wall time.  Discover what correction gettimeofday() would have
 	 * done, and then undo it!
 	 */
-	tv->tv_nsec -= 1000 * (gettimeoffset() +
-			(jiffies - wall_jiffies) * USECS_PER_JIFFY);
+	tv->tv_nsec -= 1000 * gettimeoffset();
 
 	while (tv->tv_nsec < 0) {
 		tv->tv_nsec += NSEC_PER_SEC;
diff --git a/arch/cris/kernel/time.c b/arch/cris/kernel/time.c
index 66ba8898db07..0f9213cbd48e 100644
--- a/arch/cris/kernel/time.c
+++ b/arch/cris/kernel/time.c
@@ -37,7 +37,6 @@ int have_rtc;  /* used to remember if we have an RTC or not */;
 
 #define TICK_SIZE tick
 
-extern unsigned long wall_jiffies;
 extern unsigned long loops_per_jiffy; /* init/main.c */
 unsigned long loops_per_usec;
 
@@ -58,11 +57,6 @@ void do_gettimeofday(struct timeval *tv)
 	local_irq_save(flags);
 	local_irq_disable();
 	usec = do_gettimeoffset();
-	{
-		unsigned long lost = jiffies - wall_jiffies;
-		if (lost)
-			usec += lost * (1000000 / HZ);
-	}
 
         /*
 	 * If time_adjust is negative then NTP is slowing the clock
@@ -103,7 +97,6 @@ int do_settimeofday(struct timespec *tv)
 	 * made, and then undo it!
 	 */
 	nsec -= do_gettimeoffset() * NSEC_PER_USEC;
-	nsec -= (jiffies - wall_jiffies) * TICK_NSEC;
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index 86944acfb647..58a2d5582419 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -76,8 +76,6 @@ int pit_latch_buggy;              /* extern */
 unsigned int cpu_khz;	/* Detected as we calibrate the TSC */
 EXPORT_SYMBOL(cpu_khz);
 
-extern unsigned long wall_jiffies;
-
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
 
@@ -329,7 +327,6 @@ static int timer_resume(struct sys_device *dev)
 	do_settimeofday(&ts);
 	write_seqlock_irqsave(&xtime_lock, flags);
 	jiffies_64 += sleep_length;
-	wall_jiffies += sleep_length;
 	write_sequnlock_irqrestore(&xtime_lock, flags);
 	touch_softlockup_watchdog();
 	return 0;
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 16262687a103..62e07f906e05 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -29,8 +29,6 @@
 #include <asm/sections.h>
 #include <asm/system.h>
 
-extern unsigned long wall_jiffies;
-
 volatile int time_keeper_id = 0; /* smp_processor_id() of time-keeper */
 
 #ifdef CONFIG_IA64_DEBUG_IRQ
diff --git a/arch/m32r/kernel/time.c b/arch/m32r/kernel/time.c
index 7a896893cd28..d8af155db984 100644
--- a/arch/m32r/kernel/time.c
+++ b/arch/m32r/kernel/time.c
@@ -38,7 +38,6 @@ extern void send_IPI_allbutself(int, int);
 extern void smp_local_timer_interrupt(struct pt_regs *);
 #endif
 
-extern unsigned long wall_jiffies;
 #define TICK_SIZE	(tick_nsec / 1000)
 
 /*
@@ -108,24 +107,17 @@ void do_gettimeofday(struct timeval *tv)
 	unsigned long max_ntp_tick = tick_usec - tickadj;
 
 	do {
-		unsigned long lost;
-
 		seq = read_seqbegin(&xtime_lock);
 
 		usec = do_gettimeoffset();
-		lost = jiffies - wall_jiffies;
 
 		/*
 		 * If time_adjust is negative then NTP is slowing the clock
 		 * so make sure not to go into next possible interval.
 		 * Better to lose some accuracy than have time go backwards..
 		 */
-		if (unlikely(time_adjust < 0)) {
+		if (unlikely(time_adjust < 0))
 			usec = min(usec, max_ntp_tick);
-			if (lost)
-				usec += lost * max_ntp_tick;
-		} else if (unlikely(lost))
-			usec += lost * tick_usec;
 
 		sec = xtime.tv_sec;
 		usec += (xtime.tv_nsec / 1000);
@@ -158,7 +150,6 @@ int do_settimeofday(struct timespec *tv)
 	 * made, and then undo it!
 	 */
 	nsec -= do_gettimeoffset() * NSEC_PER_USEC;
-	nsec -= (jiffies - wall_jiffies) * TICK_NSEC;
 
 	wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/m68k/kernel/time.c b/arch/m68k/kernel/time.c
index 1072e4946a4a..6cfc984380d9 100644
--- a/arch/m68k/kernel/time.c
+++ b/arch/m68k/kernel/time.c
@@ -96,31 +96,23 @@ void time_init(void)
 void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
-	extern unsigned long wall_jiffies;
 	unsigned long seq;
-	unsigned long usec, sec, lost;
+	unsigned long usec, sec;
 	unsigned long max_ntp_tick = tick_usec - tickadj;
 
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 
 		usec = mach_gettimeoffset();
-		lost = jiffies - wall_jiffies;
 
 		/*
 		 * If time_adjust is negative then NTP is slowing the clock
 		 * so make sure not to go into next possible interval.
 		 * Better to lose some accuracy than have time go backwards..
 		 */
-		if (unlikely(time_adjust < 0)) {
+		if (unlikely(time_adjust < 0))
 			usec = min(usec, max_ntp_tick);
 
-			if (lost)
-				usec += lost * max_ntp_tick;
-		}
-		else if (unlikely(lost))
-			usec += lost * tick_usec;
-
 		sec = xtime.tv_sec;
 		usec += xtime.tv_nsec/1000;
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
@@ -141,7 +133,6 @@ int do_settimeofday(struct timespec *tv)
 {
 	time_t wtm_sec, sec = tv->tv_sec;
 	long wtm_nsec, nsec = tv->tv_nsec;
-	extern unsigned long wall_jiffies;
 
 	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
 		return -EINVAL;
@@ -153,8 +144,7 @@ int do_settimeofday(struct timespec *tv)
 	 * Discover what correction gettimeofday
 	 * would have done, and then undo it!
 	 */
-	nsec -= 1000 * (mach_gettimeoffset() +
-			(jiffies - wall_jiffies) * (1000000 / HZ));
+	nsec -= 1000 * mach_gettimeoffset();
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c
index db1e1ce0a349..c5667bdddd5e 100644
--- a/arch/m68knommu/kernel/time.c
+++ b/arch/m68knommu/kernel/time.c
@@ -26,8 +26,6 @@
 
 #define	TICK_SIZE (tick_nsec / 1000)
 
-extern unsigned long wall_jiffies;
-
 
 static inline int set_rtc_mmss(unsigned long nowtime)
 {
@@ -124,15 +122,12 @@ void time_init(void)
 void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
-	unsigned long lost, seq;
+	unsigned long seq;
 	unsigned long usec, sec;
 
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 		usec = mach_gettimeoffset ? mach_gettimeoffset() : 0;
-		lost = jiffies - wall_jiffies;
-		if (lost)
-			usec += lost * (1000000 / HZ);
 		sec = xtime.tv_sec;
 		usec += (xtime.tv_nsec / 1000);
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c
index 6ab8d975a974..845c7e55505d 100644
--- a/arch/mips/kernel/time.c
+++ b/arch/mips/kernel/time.c
@@ -47,8 +47,6 @@
 /*
  * forward reference
  */
-extern volatile unsigned long wall_jiffies;
-
 DEFINE_SPINLOCK(rtc_lock);
 
 /*
@@ -159,7 +157,6 @@ void (*mips_hpt_init)(unsigned int);
 void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long seq;
-	unsigned long lost;
 	unsigned long usec, sec;
 	unsigned long max_ntp_tick;
 
@@ -168,8 +165,6 @@ void do_gettimeofday(struct timeval *tv)
 
 		usec = do_gettimeoffset();
 
-		lost = jiffies - wall_jiffies;
-
 		/*
 		 * If time_adjust is negative then NTP is slowing the clock
 		 * so make sure not to go into next possible interval.
@@ -178,11 +173,7 @@ void do_gettimeofday(struct timeval *tv)
 		if (unlikely(time_adjust < 0)) {
 			max_ntp_tick = (USEC_PER_SEC / HZ) - tickadj;
 			usec = min(usec, max_ntp_tick);
-
-			if (lost)
-				usec += lost * max_ntp_tick;
-		} else if (unlikely(lost))
-			usec += lost * (USEC_PER_SEC / HZ);
+		}
 
 		sec = xtime.tv_sec;
 		usec += (xtime.tv_nsec / 1000);
@@ -217,7 +208,6 @@ int do_settimeofday(struct timespec *tv)
 	 * made, and then undo it!
 	 */
 	nsec -= do_gettimeoffset() * NSEC_PER_USEC;
-	nsec -= (jiffies - wall_jiffies) * tick_nsec;
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c
index c62a3a9ef867..257ce118e380 100644
--- a/arch/mips/sgi-ip27/ip27-timer.c
+++ b/arch/mips/sgi-ip27/ip27-timer.c
@@ -42,8 +42,6 @@
 static unsigned long ct_cur[NR_CPUS];	/* What counter should be at next timer irq */
 static long last_rtc_update;		/* Last time the rtc clock got updated */
 
-extern volatile unsigned long wall_jiffies;
-
 #if 0
 static int set_rtc_mmss(unsigned long nowtime)
 {
diff --git a/arch/parisc/kernel/time.c b/arch/parisc/kernel/time.c
index 700df10924dd..ab641d67f551 100644
--- a/arch/parisc/kernel/time.c
+++ b/arch/parisc/kernel/time.c
@@ -32,9 +32,6 @@
 
 #include <linux/timex.h>
 
-/* xtime and wall_jiffies keep wall-clock time */
-extern unsigned long wall_jiffies;
-
 static long clocktick __read_mostly;	/* timer cycles per tick */
 static long halftick __read_mostly;
 
@@ -112,7 +109,7 @@ EXPORT_SYMBOL(profile_pc);
 /*** converted from ia64 ***/
 /*
  * Return the number of micro-seconds that elapsed since the last
- * update to wall time (aka xtime aka wall_jiffies).  The xtime_lock
+ * update to wall time (aka xtime).  The xtime_lock
  * must be at least read-locked when calling this routine.
  */
 static inline unsigned long
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 71f71da98e7d..8b278d85ca4e 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -117,8 +117,6 @@ unsigned tb_to_ns_shift;
 
 struct gettimeofday_struct do_gtod;
 
-extern unsigned long wall_jiffies;
-
 extern struct timezone sys_tz;
 static long timezone_offset;
 
@@ -816,11 +814,6 @@ int do_settimeofday(struct timespec *tv)
 	/*
 	 * Subtract off the number of nanoseconds since the
 	 * beginning of the last tick.
-	 * Note that since we don't increment jiffies_64 anywhere other
-	 * than in do_timer (since we don't have a lost tick problem),
-	 * wall_jiffies will always be the same as jiffies,
-	 * and therefore the (jiffies - wall_jiffies) computation
-	 * has been removed.
 	 */
 	tb_delta = tb_ticks_since(tb_last_jiffy);
 	tb_delta = mulhdu(tb_delta, do_gtod.varp->tb_to_xs); /* in xsec */
diff --git a/arch/ppc/kernel/time.c b/arch/ppc/kernel/time.c
index 1e1f31554767..187388625a76 100644
--- a/arch/ppc/kernel/time.c
+++ b/arch/ppc/kernel/time.c
@@ -80,8 +80,6 @@ unsigned tb_to_us;
 unsigned tb_last_stamp;
 unsigned long tb_to_ns_scale;
 
-extern unsigned long wall_jiffies;
-
 /* used for timezone offset */
 static long timezone_offset;
 
@@ -173,8 +171,7 @@ void timer_interrupt(struct pt_regs * regs)
 		 */
 		if ( ppc_md.set_rtc_time && ntp_synced() &&
 		     xtime.tv_sec - last_rtc_update >= 659 &&
-		     abs((xtime.tv_nsec / 1000) - (1000000-1000000/HZ)) < 500000/HZ &&
-		     jiffies - wall_jiffies == 1) {
+		     abs((xtime.tv_nsec / 1000) - (1000000-1000000/HZ)) < 500000/HZ) {
 		  	if (ppc_md.set_rtc_time(xtime.tv_sec+1 + timezone_offset) == 0)
 				last_rtc_update = xtime.tv_sec+1;
 			else
@@ -200,7 +197,7 @@ void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
 	unsigned long seq;
-	unsigned delta, lost_ticks, usec, sec;
+	unsigned delta, usec, sec;
 
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
@@ -214,10 +211,9 @@ void do_gettimeofday(struct timeval *tv)
 		if (!smp_tb_synchronized)
 			delta = 0;
 #endif /* CONFIG_SMP */
-		lost_ticks = jiffies - wall_jiffies;
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
 
-	usec += mulhwu(tb_to_us, tb_ticks_per_jiffy * lost_ticks + delta);
+	usec += mulhwu(tb_to_us, delta);
 	while (usec >= 1000000) {
 	  	sec++;
 		usec -= 1000000;
@@ -258,7 +254,6 @@ int do_settimeofday(struct timespec *tv)
 	 * still reasonable when gettimeofday resolution is 1 jiffy.
 	 */
 	tb_delta = tb_ticks_since(last_jiffy_stamp(smp_processor_id()));
-	tb_delta += (jiffies - wall_jiffies) * tb_ticks_per_jiffy;
 
 	new_nsec -= 1000 * mulhwu(tb_to_us, tb_delta);
 
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index abab42e9f5f8..4bf66cc4a267 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -53,8 +53,6 @@ static u64 init_timer_cc;
 static u64 jiffies_timer_cc;
 static u64 xtime_cc;
 
-extern unsigned long wall_jiffies;
-
 /*
  * Scheduler clock - returns current time in nanosec units.
  */
@@ -87,9 +85,8 @@ static inline unsigned long do_gettimeoffset(void)
 {
 	__u64 now;
 
-        now = (get_clock() - jiffies_timer_cc) >> 12;
-	/* We require the offset from the latest update of xtime */
-	now -= (__u64) wall_jiffies*USECS_PER_JIFFY;
+	now = (get_clock() - jiffies_timer_cc) >> 12;
+	now -= (__u64) jiffies * USECS_PER_JIFFY;
 	return (unsigned long) now;
 }
 
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index f664a196c4f5..450c68f1df05 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -18,7 +18,6 @@
 #include <asm/timer.h>
 #include <asm/kgdb.h>
 
-extern unsigned long wall_jiffies;
 struct sys_timer *sys_timer;
 
 /* Move this somewhere more sensible.. */
@@ -52,16 +51,10 @@ void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long seq;
 	unsigned long usec, sec;
-	unsigned long lost;
 
 	do {
 		seq = read_seqbegin(&xtime_lock);
 		usec = get_timer_offset();
-
-		lost = jiffies - wall_jiffies;
-		if (lost)
-			usec += lost * (1000000 / HZ);
-
 		sec = xtime.tv_sec;
 		usec += xtime.tv_nsec / 1000;
 	} while (read_seqretry(&xtime_lock, seq));
@@ -91,8 +84,7 @@ int do_settimeofday(struct timespec *tv)
 	 * wall time.  Discover what correction gettimeofday() would have
 	 * made, and then undo it!
 	 */
-	nsec -= 1000 * (get_timer_offset() +
-				(jiffies - wall_jiffies) * (1000000 / HZ));
+	nsec -= 1000 * get_timer_offset();
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/sh64/kernel/time.c b/arch/sh64/kernel/time.c
index 3b61e06f9d72..9c4a38a8698c 100644
--- a/arch/sh64/kernel/time.c
+++ b/arch/sh64/kernel/time.c
@@ -107,8 +107,6 @@
 
 #define TICK_SIZE (tick_nsec / 1000)
 
-extern unsigned long wall_jiffies;
-
 static unsigned long tmu_base, rtc_base;
 unsigned long cprc_base;
 
@@ -194,13 +192,6 @@ void do_gettimeofday(struct timeval *tv)
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 		usec = usecs_since_tick();
-		{
-			unsigned long lost = jiffies - wall_jiffies;
-
-			if (lost)
-				usec += lost * (1000000 / HZ);
-		}
-
 		sec = xtime.tv_sec;
 		usec += xtime.tv_nsec / 1000;
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
@@ -229,8 +220,7 @@ int do_settimeofday(struct timespec *tv)
 	 * wall time.  Discover what correction gettimeofday() would have
 	 * made, and then undo it!
 	 */
-	nsec -= 1000 * (usecs_since_tick() +
-				(jiffies - wall_jiffies) * (1000000 / HZ));
+	nsec -= 1000 * usecs_since_tick();
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c
index e19b1bad9bc5..edb6cc665f56 100644
--- a/arch/sparc/kernel/pcic.c
+++ b/arch/sparc/kernel/pcic.c
@@ -765,8 +765,6 @@ static __inline__ unsigned long do_gettimeoffset(void)
 	return count;
 }
 
-extern unsigned long wall_jiffies;
-
 static void pci_do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
@@ -775,26 +773,17 @@ static void pci_do_gettimeofday(struct timeval *tv)
 	unsigned long max_ntp_tick = tick_usec - tickadj;
 
 	do {
-		unsigned long lost;
-
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 		usec = do_gettimeoffset();
-		lost = jiffies - wall_jiffies;
 
 		/*
 		 * If time_adjust is negative then NTP is slowing the clock
 		 * so make sure not to go into next possible interval.
 		 * Better to lose some accuracy than have time go backwards..
 		 */
-		if (unlikely(time_adjust < 0)) {
+		if (unlikely(time_adjust < 0))
 			usec = min(usec, max_ntp_tick);
 
-			if (lost)
-				usec += lost * max_ntp_tick;
-		}
-		else if (unlikely(lost))
-			usec += lost * tick_usec;
-
 		sec = xtime.tv_sec;
 		usec += (xtime.tv_nsec / 1000);
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
@@ -819,8 +808,7 @@ static int pci_do_settimeofday(struct timespec *tv)
 	 * wall time.  Discover what correction gettimeofday() would have
 	 * made, and then undo it!
 	 */
-	tv->tv_nsec -= 1000 * (do_gettimeoffset() + 
-				(jiffies - wall_jiffies) * (USEC_PER_SEC / HZ));
+	tv->tv_nsec -= 1000 * do_gettimeoffset();
 	while (tv->tv_nsec < 0) {
 		tv->tv_nsec += NSEC_PER_SEC;
 		tv->tv_sec--;
diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c
index 6f84fa1b58e5..e10dc831944d 100644
--- a/arch/sparc/kernel/time.c
+++ b/arch/sparc/kernel/time.c
@@ -43,8 +43,6 @@
 #include <asm/pcic.h>
 #include <asm/of_device.h>
 
-extern unsigned long wall_jiffies;
-
 DEFINE_SPINLOCK(rtc_lock);
 enum sparc_clock_type sp_clock_typ;
 DEFINE_SPINLOCK(mostek_lock);
@@ -449,7 +447,7 @@ unsigned long long sched_clock(void)
 
 /* Ok, my cute asm atomicity trick doesn't work anymore.
  * There are just too many variables that need to be protected
- * now (both members of xtime, wall_jiffies, et al.)
+ * now (both members of xtime, et al.)
  */
 void do_gettimeofday(struct timeval *tv)
 {
@@ -459,26 +457,17 @@ void do_gettimeofday(struct timeval *tv)
 	unsigned long max_ntp_tick = tick_usec - tickadj;
 
 	do {
-		unsigned long lost;
-
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
 		usec = do_gettimeoffset();
-		lost = jiffies - wall_jiffies;
 
 		/*
 		 * If time_adjust is negative then NTP is slowing the clock
 		 * so make sure not to go into next possible interval.
 		 * Better to lose some accuracy than have time go backwards..
 		 */
-		if (unlikely(time_adjust < 0)) {
+		if (unlikely(time_adjust < 0))
 			usec = min(usec, max_ntp_tick);
 
-			if (lost)
-				usec += lost * max_ntp_tick;
-		}
-		else if (unlikely(lost))
-			usec += lost * tick_usec;
-
 		sec = xtime.tv_sec;
 		usec += (xtime.tv_nsec / 1000);
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
@@ -521,8 +510,7 @@ static int sbus_do_settimeofday(struct timespec *tv)
 	 * wall time.  Discover what correction gettimeofday() would have
 	 * made, and then undo it!
 	 */
-	nsec -= 1000 * (do_gettimeoffset() +
-			(jiffies - wall_jiffies) * (USEC_PER_SEC / HZ));
+	nsec -= 1000 * do_gettimeoffset();
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c
index ca1193482f07..00f6fc4aaaff 100644
--- a/arch/sparc64/kernel/time.c
+++ b/arch/sparc64/kernel/time.c
@@ -53,8 +53,6 @@ void __iomem *mstk48t02_regs = NULL;
 unsigned long ds1287_regs = 0UL;
 #endif
 
-extern unsigned long wall_jiffies;
-
 static void __iomem *mstk48t08_regs;
 static void __iomem *mstk48t59_regs;
 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 7ea3bf2a858c..557e92af7bea 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -77,7 +77,6 @@ unsigned long long monotonic_base;
 struct vxtime_data __vxtime __section_vxtime;	/* for vsyscalls */
 
 volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES;
-unsigned long __wall_jiffies __section_wall_jiffies = INITIAL_JIFFIES;
 struct timespec __xtime __section_xtime;
 struct timezone __sys_tz __section_sys_tz;
 
@@ -119,7 +118,7 @@ unsigned int (*do_gettimeoffset)(void) = do_gettimeoffset_tsc;
 
 void do_gettimeofday(struct timeval *tv)
 {
-	unsigned long seq, t;
+	unsigned long seq;
  	unsigned int sec, usec;
 
 	do {
@@ -136,10 +135,7 @@ void do_gettimeofday(struct timeval *tv)
 		   be found. Note when you fix it here you need to do the same
 		   in arch/x86_64/kernel/vsyscall.c and export all needed
 		   variables in vmlinux.lds. -AK */ 
-
-		t = (jiffies - wall_jiffies) * USEC_PER_TICK +
-			do_gettimeoffset();
-		usec += t;
+		usec += do_gettimeoffset();
 
 	} while (read_seqretry(&xtime_lock, seq));
 
@@ -165,8 +161,7 @@ int do_settimeofday(struct timespec *tv)
 
 	write_seqlock_irq(&xtime_lock);
 
-	nsec -= do_gettimeoffset() * NSEC_PER_USEC +
-		(jiffies - wall_jiffies) * NSEC_PER_TICK;
+	nsec -= do_gettimeoffset() * NSEC_PER_USEC;
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
@@ -1071,7 +1066,6 @@ static int timer_resume(struct sys_device *dev)
 		vxtime.last_tsc = get_cycles_sync();
 	write_sequnlock_irqrestore(&xtime_lock,flags);
 	jiffies += sleep_length;
-	wall_jiffies += sleep_length;
 	monotonic_base += sleep_length * (NSEC_PER_SEC/HZ);
 	touch_softlockup_watchdog();
 	return 0;
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index f8aeccf105fa..b9df2ab6529f 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -101,9 +101,6 @@ SECTIONS
   .vgetcpu_mode : AT(VLOAD(.vgetcpu_mode)) { *(.vgetcpu_mode) }
   vgetcpu_mode = VVIRT(.vgetcpu_mode);
 
-  .wall_jiffies : AT(VLOAD(.wall_jiffies)) { *(.wall_jiffies) }
-  wall_jiffies = VVIRT(.wall_jiffies);
-
   .sys_tz : AT(VLOAD(.sys_tz)) { *(.sys_tz) }
   sys_tz = VVIRT(.sys_tz);
 
diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index 07c086382059..a98b460af6a1 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -66,8 +66,7 @@ static __always_inline void do_vgettimeofday(struct timeval * tv)
 		sequence = read_seqbegin(&__xtime_lock);
 		
 		sec = __xtime.tv_sec;
-		usec = (__xtime.tv_nsec / 1000) +
-			(__jiffies - __wall_jiffies) * (1000000 / HZ);
+		usec = __xtime.tv_nsec / 1000;
 
 		if (__vxtime.mode != VXTIME_HPET) {
 			t = get_cycles_sync();
diff --git a/arch/xtensa/kernel/time.c b/arch/xtensa/kernel/time.c
index 241db201f40e..37347e369987 100644
--- a/arch/xtensa/kernel/time.c
+++ b/arch/xtensa/kernel/time.c
@@ -26,8 +26,6 @@
 #include <asm/platform.h>
 
 
-extern volatile unsigned long wall_jiffies;
-
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
 
@@ -110,7 +108,6 @@ int do_settimeofday(struct timespec *tv)
 	 */
 	ccount = get_ccount();
 	nsec -= (ccount - last_ccount_stamp) * CCOUNT_NSEC;
-	nsec -= (jiffies - wall_jiffies) * CCOUNT_PER_JIFFY * CCOUNT_NSEC;
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
@@ -129,7 +126,7 @@ EXPORT_SYMBOL(do_settimeofday);
 void do_gettimeofday(struct timeval *tv)
 {
 	unsigned long flags;
-	unsigned long sec, usec, delta, lost, seq;
+	unsigned long sec, usec, delta, seq;
 
 	do {
 		seq = read_seqbegin_irqsave(&xtime_lock, flags);
@@ -137,12 +134,9 @@ void do_gettimeofday(struct timeval *tv)
 		delta = get_ccount() - last_ccount_stamp;
 		sec = xtime.tv_sec;
 		usec = (xtime.tv_nsec / NSEC_PER_USEC);
-
-		lost = jiffies - wall_jiffies;
-
 	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
 
-	usec += lost * (1000000UL/HZ) + (delta * CCOUNT_NSEC) / NSEC_PER_USEC;
+	usec += (delta * CCOUNT_NSEC) / NSEC_PER_USEC;
 	for (; usec >= 1000000; sec++, usec -= 1000000)
 		;
 
@@ -179,8 +173,7 @@ again:
 
 		if (ntp_synced() &&
 		    xtime.tv_sec - last_rtc_update >= 659 &&
-		    abs((xtime.tv_nsec/1000)-(1000000-1000000/HZ))<5000000/HZ &&
-		    jiffies - wall_jiffies == 1) {
+		    abs((xtime.tv_nsec/1000)-(1000000-1000000/HZ))<5000000/HZ) {
 
 			if (platform_set_rtc_time(xtime.tv_sec+1) == 0)
 				last_rtc_update = xtime.tv_sec+1;
diff --git a/include/asm-x86_64/vsyscall.h b/include/asm-x86_64/vsyscall.h
index 2281e9399b96..fd452fc2c037 100644
--- a/include/asm-x86_64/vsyscall.h
+++ b/include/asm-x86_64/vsyscall.h
@@ -17,7 +17,6 @@ enum vsyscall_num {
 
 #define __section_vxtime __attribute__ ((unused, __section__ (".vxtime"), aligned(16)))
 #define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16)))
-#define __section_wall_jiffies __attribute__ ((unused, __section__ (".wall_jiffies"), aligned(16)))
 #define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16)))
 #define __section_sys_tz __attribute__ ((unused, __section__ (".sys_tz"), aligned(16)))
 #define __section_sysctl_vsyscall __attribute__ ((unused, __section__ (".sysctl_vsyscall"), aligned(16)))
@@ -48,14 +47,12 @@ extern struct vxtime_data __vxtime;
 extern int __vgetcpu_mode;
 extern struct timespec __xtime;
 extern volatile unsigned long __jiffies;
-extern unsigned long __wall_jiffies;
 extern struct timezone __sys_tz;
 extern seqlock_t __xtime_lock;
 
 /* kernel space (writeable) */
 extern struct vxtime_data vxtime;
 extern int vgetcpu_mode;
-extern unsigned long wall_jiffies;
 extern struct timezone sys_tz;
 extern int sysctl_vsyscall;
 extern seqlock_t xtime_lock;
diff --git a/kernel/timer.c b/kernel/timer.c
index 654dd5df2417..c1c7fbcffec1 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -768,7 +768,7 @@ static int timekeeping_suspended;
  * @dev:	unused
  *
  * This is for the generic clocksource timekeeping.
- * xtime/wall_to_monotonic/jiffies/wall_jiffies/etc are
+ * xtime/wall_to_monotonic/jiffies/etc are
  * still managed by arch specific suspend/resume code.
  */
 static int timekeeping_resume(struct sys_device *dev)
@@ -1016,9 +1016,6 @@ static inline void calc_load(unsigned long ticks)
 	}
 }
 
-/* jiffies at the most recent update of wall time */
-unsigned long wall_jiffies = INITIAL_JIFFIES;
-
 /*
  * This read-write spinlock protects us from races in SMP while
  * playing with xtime and avenrun.
@@ -1056,7 +1053,6 @@ void run_local_timers(void)
  */
 static inline void update_times(unsigned long ticks)
 {
-	wall_jiffies += ticks;
 	update_wall_time();
 	calc_load(ticks);
 }
-- 
cgit v1.2.3


From e1fabd3ccf02901374bffa434e0af472749a5bd9 Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Sat, 30 Sep 2006 23:28:40 -0700
Subject: [PATCH] reiserfs: fix is_reusable bitmap check to not traverse the
 bitmap info array

There is a check in is_reusable to determine if a particular block is a bitmap
block.  It verifies this by going through the array of bitmap block buffer
heads and comparing the block number to each one.

Bitmap blocks are at defined locations on the disk in both old and current
formats.  Simply checking against the known good values is enough.

This is a trivial optimization for a non-production codepath, but this is the
first in a series of patches that will ultimately remove the buffer heads from
that array.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: <reiserfs-dev@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/bitmap.c           | 40 +++++++++++++++++++++++++---------------
 fs/reiserfs/super.c            |  2 ++
 include/linux/reiserfs_fs_sb.h |  1 +
 3 files changed, 28 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c
index 4a7dbdee1b6d..1022347a211f 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -50,16 +50,15 @@ static inline void get_bit_address(struct super_block *s,
 {
 	/* It is in the bitmap block number equal to the block
 	 * number divided by the number of bits in a block. */
-	*bmap_nr = block / (s->s_blocksize << 3);
+	*bmap_nr = block >> (s->s_blocksize_bits + 3);
 	/* Within that bitmap block it is located at bit offset *offset. */
 	*offset = block & ((s->s_blocksize << 3) - 1);
-	return;
 }
 
 #ifdef CONFIG_REISERFS_CHECK
 int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value)
 {
-	int i, j;
+	int bmap, offset;
 
 	if (block == 0 || block >= SB_BLOCK_COUNT(s)) {
 		reiserfs_warning(s,
@@ -68,34 +67,45 @@ int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value)
 		return 0;
 	}
 
-	/* it can't be one of the bitmap blocks */
-	for (i = 0; i < SB_BMAP_NR(s); i++)
-		if (block == SB_AP_BITMAP(s)[i].bh->b_blocknr) {
+	get_bit_address(s, block, &bmap, &offset);
+
+	/* Old format filesystem? Unlikely, but the bitmaps are all up front so
+	 * we need to account for it. */
+	if (unlikely(test_bit(REISERFS_OLD_FORMAT,
+			      &(REISERFS_SB(s)->s_properties)))) {
+		b_blocknr_t bmap1 = REISERFS_SB(s)->s_sbh->b_blocknr + 1;
+		if (block >= bmap1 && block <= bmap1 + SB_BMAP_NR(s)) {
+			reiserfs_warning(s, "vs: 4019: is_reusable: "
+					 "bitmap block %lu(%u) can't be freed or reused",
+					 block, SB_BMAP_NR(s));
+			return 0;
+		}
+	} else {
+		if (offset == 0) {
 			reiserfs_warning(s, "vs: 4020: is_reusable: "
 					 "bitmap block %lu(%u) can't be freed or reused",
 					 block, SB_BMAP_NR(s));
 			return 0;
 		}
+	}
 
-	get_bit_address(s, block, &i, &j);
-
-	if (i >= SB_BMAP_NR(s)) {
+	if (bmap >= SB_BMAP_NR(s)) {
 		reiserfs_warning(s,
 				 "vs-4030: is_reusable: there is no so many bitmap blocks: "
-				 "block=%lu, bitmap_nr=%d", block, i);
+				 "block=%lu, bitmap_nr=%d", block, bmap);
 		return 0;
 	}
 
 	if ((bit_value == 0 &&
-	     reiserfs_test_le_bit(j, SB_AP_BITMAP(s)[i].bh->b_data)) ||
+	     reiserfs_test_le_bit(offset, SB_AP_BITMAP(s)[bmap].bh->b_data)) ||
 	    (bit_value == 1 &&
-	     reiserfs_test_le_bit(j, SB_AP_BITMAP(s)[i].bh->b_data) == 0)) {
+	     reiserfs_test_le_bit(offset, SB_AP_BITMAP(s)[bmap].bh->b_data) == 0)) {
 		reiserfs_warning(s,
 				 "vs-4040: is_reusable: corresponding bit of block %lu does not "
-				 "match required value (i==%d, j==%d) test_bit==%d",
-				 block, i, j, reiserfs_test_le_bit(j,
+				 "match required value (bmap==%d, offset==%d) test_bit==%d",
+				 block, bmap, offset, reiserfs_test_le_bit(offset,
 								   SB_AP_BITMAP
-								   (s)[i].bh->
+								   (s)[bmap].bh->
 								   b_data));
 
 		return 0;
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 80fc3b32802f..db2c581df766 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -1818,6 +1818,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	if (is_reiserfs_3_5(rs)
 	    || (is_reiserfs_jr(rs) && SB_VERSION(s) == REISERFS_VERSION_1))
 		set_bit(REISERFS_3_5, &(sbi->s_properties));
+	else if (old_format)
+		set_bit(REISERFS_OLD_FORMAT, &(sbi->s_properties));
 	else
 		set_bit(REISERFS_3_6, &(sbi->s_properties));
 
diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index 31b4c0bd4fa0..4f21ad388c79 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -414,6 +414,7 @@ struct reiserfs_sb_info {
 /* Definitions of reiserfs on-disk properties: */
 #define REISERFS_3_5 0
 #define REISERFS_3_6 1
+#define REISERFS_OLD_FORMAT 2
 
 enum reiserfs_mount_options {
 /* Mount options */
-- 
cgit v1.2.3


From 6f01046b35d940079822827498a7dd6d3eec8c6b Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Sat, 30 Sep 2006 23:28:43 -0700
Subject: [PATCH] reiserfs: reorganize bitmap loading functions

This patch moves the bitmap loading code from super.c to bitmap.c

The code is also restructured somewhat.  The only difference between new
format bitmaps and old format bitmaps is where they are.  That's a two liner
before loading the block to use the correct one.  There's no need for an
entirely separate code path.

The load path is generally the same, with the pattern being to throw out a
bunch of requests and then wait for them, then cache the metadata from the
contents.

Again, like the previous patches, the purpose is to set up for later ones.

Update: There was a bug in the previously posted version of this that resulted
in corruption.  The problem was that bitmap 0 on new format file systems must
be treated specially, and wasn't.  A stupid bug with an easy fix.

This is hopefully the last fix for the disaster that is the reiserfs bitmap
patch set.

If a bitmap block was full, first_zero_hint would end up at zero since it
would never be changed from it's zeroed out value.  This just sets it
beyond the end of the bitmap block.  If any bits are freed, it will be
reset to a valid bit.  When info->free_count = 0, then we already know it's
full.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: <reiserfs-dev@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/bitmap.c        |  88 ++++++++++++++++++++++++++++++++++
 fs/reiserfs/resize.c        |   1 +
 fs/reiserfs/super.c         | 114 +-------------------------------------------
 include/linux/reiserfs_fs.h |   4 ++
 4 files changed, 94 insertions(+), 113 deletions(-)

(limited to 'include')

diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c
index 44d9410e9d6a..abdd6d9c4558 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -9,6 +9,7 @@
 #include <linux/buffer_head.h>
 #include <linux/kernel.h>
 #include <linux/pagemap.h>
+#include <linux/vmalloc.h>
 #include <linux/reiserfs_fs_sb.h>
 #include <linux/reiserfs_fs_i.h>
 #include <linux/quotaops.h>
@@ -1285,3 +1286,90 @@ int reiserfs_can_fit_pages(struct super_block *sb	/* superblock of filesystem
 
 	return space > 0 ? space : 0;
 }
+
+void reiserfs_cache_bitmap_metadata(struct super_block *sb,
+                                    struct buffer_head *bh,
+                                    struct reiserfs_bitmap_info *info)
+{
+	unsigned long *cur = (unsigned long *)(bh->b_data + bh->b_size);
+
+	info->first_zero_hint = 1 << (sb->s_blocksize_bits + 3);
+
+	while (--cur >= (unsigned long *)bh->b_data) {
+		int base = ((char *)cur - bh->b_data) << 3;
+
+		/* 0 and ~0 are special, we can optimize for them */
+		if (*cur == 0) {
+			info->first_zero_hint = base;
+			info->free_count += BITS_PER_LONG;
+		} else if (*cur != ~0L) {       /* A mix, investigate */
+			int b;
+			for (b = BITS_PER_LONG - 1; b >= 0; b--) {
+				if (!reiserfs_test_le_bit(b, cur)) {
+					info->first_zero_hint = base + b;
+					info->free_count++;
+				}
+			}
+		}
+	}
+	/* The first bit must ALWAYS be 1 */
+	BUG_ON(info->first_zero_hint == 0);
+}
+
+struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
+                                               unsigned int bitmap)
+{
+	b_blocknr_t block = (sb->s_blocksize << 3) * bitmap;
+	struct buffer_head *bh;
+
+	/* Way old format filesystems had the bitmaps packed up front.
+	 * I doubt there are any of these left, but just in case... */
+	if (unlikely(test_bit(REISERFS_OLD_FORMAT,
+	                      &(REISERFS_SB(sb)->s_properties))))
+		block = REISERFS_SB(sb)->s_sbh->b_blocknr + 1 + bitmap;
+	else if (bitmap == 0)
+		block = (REISERFS_DISK_OFFSET_IN_BYTES >> sb->s_blocksize_bits) + 1;
+
+	bh = sb_getblk(sb, block);
+	if (!buffer_uptodate(bh))
+		ll_rw_block(READ, 1, &bh);
+
+	return bh;
+}
+
+int reiserfs_init_bitmap_cache(struct super_block *sb)
+{
+	struct reiserfs_bitmap_info *bitmap;
+	int i;
+
+	bitmap = vmalloc(sizeof (*bitmap) * SB_BMAP_NR(sb));
+	if (bitmap == NULL)
+		return -ENOMEM;
+
+	memset(bitmap, 0, sizeof (*bitmap) * SB_BMAP_NR(sb));
+
+	for (i = 0; i < SB_BMAP_NR(sb); i++)
+		bitmap[i].bh = reiserfs_read_bitmap_block(sb, i);
+
+	/* make sure we have them all */
+	for (i = 0; i < SB_BMAP_NR(sb); i++) {
+		wait_on_buffer(bitmap[i].bh);
+		if (!buffer_uptodate(bitmap[i].bh)) {
+			reiserfs_warning(sb, "sh-2029: %s: "
+					 "bitmap block (#%lu) reading failed",
+			                 __FUNCTION__, bitmap[i].bh->b_blocknr);
+			for (i = 0; i < SB_BMAP_NR(sb); i++)
+				brelse(bitmap[i].bh);
+			vfree(bitmap);
+			return -EIO;
+		}
+	}
+
+	/* Cache the info on the bitmaps before we get rolling */
+	for (i = 0; i < SB_BMAP_NR(sb); i++)
+		reiserfs_cache_bitmap_metadata(sb, bitmap[i].bh, &bitmap[i]);
+
+	SB_AP_BITMAP(sb) = bitmap;
+
+	return 0;
+}
diff --git a/fs/reiserfs/resize.c b/fs/reiserfs/resize.c
index 958b75978994..90d39fd3096f 100644
--- a/fs/reiserfs/resize.c
+++ b/fs/reiserfs/resize.c
@@ -132,6 +132,7 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 			get_bh(bh);
 			memset(bh->b_data, 0, sb_blocksize(sb));
 			reiserfs_test_and_set_le_bit(0, bh->b_data);
+			reiserfs_cache_bitmap_metadata(s, bh, bitmap + i);
 
 			set_buffer_uptodate(bh);
 			mark_buffer_dirty(bh);
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index db2c581df766..c78e99e196fa 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -1243,118 +1243,6 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
 	return 0;
 }
 
-/* load_bitmap_info_data - Sets up the reiserfs_bitmap_info structure from disk.
- * @sb - superblock for this filesystem
- * @bi - the bitmap info to be loaded. Requires that bi->bh is valid.
- *
- * This routine counts how many free bits there are, finding the first zero
- * as a side effect. Could also be implemented as a loop of test_bit() calls, or
- * a loop of find_first_zero_bit() calls. This implementation is similar to
- * find_first_zero_bit(), but doesn't return after it finds the first bit.
- * Should only be called on fs mount, but should be fairly efficient anyways.
- *
- * bi->first_zero_hint is considered unset if it == 0, since the bitmap itself
- * will * invariably occupt block 0 represented in the bitmap. The only
- * exception to this is when free_count also == 0, since there will be no
- * free blocks at all.
- */
-
-static void load_bitmap_info_data(struct super_block *sb,
-				  struct reiserfs_bitmap_info *bi)
-{
-	unsigned long *cur = (unsigned long *)bi->bh->b_data;
-
-	while ((char *)cur < (bi->bh->b_data + sb->s_blocksize)) {
-
-		/* No need to scan if all 0's or all 1's.
-		 * Since we're only counting 0's, we can simply ignore all 1's */
-		if (*cur == 0) {
-			if (bi->first_zero_hint == 0) {
-				bi->first_zero_hint =
-				    ((char *)cur - bi->bh->b_data) << 3;
-			}
-			bi->free_count += sizeof(unsigned long) * 8;
-		} else if (*cur != ~0L) {
-			int b;
-			for (b = 0; b < sizeof(unsigned long) * 8; b++) {
-				if (!reiserfs_test_le_bit(b, cur)) {
-					bi->free_count++;
-					if (bi->first_zero_hint == 0)
-						bi->first_zero_hint =
-						    (((char *)cur -
-						      bi->bh->b_data) << 3) + b;
-				}
-			}
-		}
-		cur++;
-	}
-
-#ifdef CONFIG_REISERFS_CHECK
-// This outputs a lot of unneded info on big FSes
-//    reiserfs_warning ("bitmap loaded from block %d: %d free blocks",
-//                    bi->bh->b_blocknr, bi->free_count);
-#endif
-}
-
-static int read_bitmaps(struct super_block *s)
-{
-	int i, bmap_nr;
-
-	SB_AP_BITMAP(s) =
-	    vmalloc(sizeof(struct reiserfs_bitmap_info) * SB_BMAP_NR(s));
-	if (SB_AP_BITMAP(s) == 0)
-		return 1;
-	memset(SB_AP_BITMAP(s), 0,
-	       sizeof(struct reiserfs_bitmap_info) * SB_BMAP_NR(s));
-	for (i = 0, bmap_nr =
-	     REISERFS_DISK_OFFSET_IN_BYTES / s->s_blocksize + 1;
-	     i < SB_BMAP_NR(s); i++, bmap_nr = s->s_blocksize * 8 * i) {
-		SB_AP_BITMAP(s)[i].bh = sb_getblk(s, bmap_nr);
-		if (!buffer_uptodate(SB_AP_BITMAP(s)[i].bh))
-			ll_rw_block(READ, 1, &SB_AP_BITMAP(s)[i].bh);
-	}
-	for (i = 0; i < SB_BMAP_NR(s); i++) {
-		wait_on_buffer(SB_AP_BITMAP(s)[i].bh);
-		if (!buffer_uptodate(SB_AP_BITMAP(s)[i].bh)) {
-			reiserfs_warning(s, "sh-2029: reiserfs read_bitmaps: "
-					 "bitmap block (#%lu) reading failed",
-					 SB_AP_BITMAP(s)[i].bh->b_blocknr);
-			for (i = 0; i < SB_BMAP_NR(s); i++)
-				brelse(SB_AP_BITMAP(s)[i].bh);
-			vfree(SB_AP_BITMAP(s));
-			SB_AP_BITMAP(s) = NULL;
-			return 1;
-		}
-		load_bitmap_info_data(s, SB_AP_BITMAP(s) + i);
-	}
-	return 0;
-}
-
-static int read_old_bitmaps(struct super_block *s)
-{
-	int i;
-	struct reiserfs_super_block *rs = SB_DISK_SUPER_BLOCK(s);
-	int bmp1 = (REISERFS_OLD_DISK_OFFSET_IN_BYTES / s->s_blocksize) + 1;	/* first of bitmap blocks */
-
-	/* read true bitmap */
-	SB_AP_BITMAP(s) =
-	    vmalloc(sizeof(struct reiserfs_buffer_info *) * sb_bmap_nr(rs));
-	if (SB_AP_BITMAP(s) == 0)
-		return 1;
-
-	memset(SB_AP_BITMAP(s), 0,
-	       sizeof(struct reiserfs_buffer_info *) * sb_bmap_nr(rs));
-
-	for (i = 0; i < sb_bmap_nr(rs); i++) {
-		SB_AP_BITMAP(s)[i].bh = sb_bread(s, bmp1 + i);
-		if (!SB_AP_BITMAP(s)[i].bh)
-			return 1;
-		load_bitmap_info_data(s, SB_AP_BITMAP(s) + i);
-	}
-
-	return 0;
-}
-
 static int read_super_block(struct super_block *s, int offset)
 {
 	struct buffer_head *bh;
@@ -1736,7 +1624,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	sbi->s_mount_state = SB_REISERFS_STATE(s);
 	sbi->s_mount_state = REISERFS_VALID_FS;
 
-	if (old_format ? read_old_bitmaps(s) : read_bitmaps(s)) {
+	if ((errval = reiserfs_init_bitmap_cache(s))) {
 		SWARN(silent, s,
 		      "jmacd-8: reiserfs_fill_super: unable to read bitmap");
 		goto error;
diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h
index 9c63abffd7b2..7bc6bfb86253 100644
--- a/include/linux/reiserfs_fs.h
+++ b/include/linux/reiserfs_fs.h
@@ -2073,6 +2073,10 @@ void reiserfs_init_alloc_options(struct super_block *s);
  */
 __le32 reiserfs_choose_packing(struct inode *dir);
 
+int reiserfs_init_bitmap_cache(struct super_block *sb);
+void reiserfs_free_bitmap_cache(struct super_block *sb);
+void reiserfs_cache_bitmap_metadata(struct super_block *sb, struct buffer_head *bh, struct reiserfs_bitmap_info *info);
+struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb, unsigned int bitmap);
 int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value);
 void reiserfs_free_block(struct reiserfs_transaction_handle *th, struct inode *,
 			 b_blocknr_t, int for_unformatted);
-- 
cgit v1.2.3


From 5065227b46235ec0131b383cc2f537069b55c6b6 Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Sat, 30 Sep 2006 23:28:44 -0700
Subject: [PATCH] reiserfs: on-demand bitmap loading

This is the patch the three previous ones have been leading up to.

It changes the behavior of ReiserFS from loading and caching all the bitmaps
as special, to treating the bitmaps like any other bit of metadata and just
letting the system-wide caches figure out what to hang on to.

Buffer heads are allocated on the fly, so there is no need to retain pointers
to all of them.  The caching of the metadata occurs when the data is read and
updated, and is considered invalid and uncached until then.

I needed to remove the vs-4040 check for performing a duplicate operation on a
particular bit.  The reason is that while the other sites for working with
bitmaps are allowed to schedule, is_reusable() is called from do_balance(),
which will panic if a schedule occurs in certain places.

The benefit of on-demand bitmaps clearly outweighs a sanity check that depends
on a compile-time option that is discouraged.

[akpm@osdl.org: warning fix]
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: <reiserfs-dev@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/bitmap.c           | 97 +++++++++++++++++++-----------------------
 fs/reiserfs/resize.c           | 24 ++++++++---
 fs/reiserfs/super.c            | 39 ++++-------------
 include/linux/reiserfs_fs_sb.h |  1 -
 4 files changed, 70 insertions(+), 91 deletions(-)

(limited to 'include')

diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c
index abdd6d9c4558..cca1dbf5458f 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -60,7 +60,6 @@ static inline void get_bit_address(struct super_block *s,
 int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value)
 {
 	int bmap, offset;
-	struct buffer_head *bh;
 
 	if (block == 0 || block >= SB_BLOCK_COUNT(s)) {
 		reiserfs_warning(s,
@@ -98,22 +97,6 @@ int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value)
 		return 0;
 	}
 
-	bh = SB_AP_BITMAP(s)[bmap].bh;
-	get_bh(bh);
-
-	if ((bit_value == 0 && reiserfs_test_le_bit(offset, bh->b_data)) ||
-	    (bit_value == 1 && reiserfs_test_le_bit(offset, bh->b_data) == 0)) {
-		reiserfs_warning(s,
-				 "vs-4040: is_reusable: corresponding bit of block %lu does not "
-				 "match required value (bmap==%d, offset==%d) test_bit==%d",
-				 block, bmap, offset,
-				 reiserfs_test_le_bit(offset, bh->b_data));
-
-		brelse(bh);
-		return 0;
-	}
-	brelse(bh);
-
 	if (bit_value == 0 && block == SB_ROOT_BLOCK(s)) {
 		reiserfs_warning(s,
 				 "vs-4050: is_reusable: this is root block (%u), "
@@ -173,13 +156,10 @@ static int scan_bitmap_block(struct reiserfs_transaction_handle *th,
 				 bmap_n);
 		return 0;
 	}
-	bh = bi->bh;
-	get_bh(bh);
 
-	if (buffer_locked(bh)) {
-		PROC_INFO_INC(s, scan_bitmap.wait);
-		__wait_on_buffer(bh);
-	}
+	bh = reiserfs_read_bitmap_block(s, bmap_n);
+	if (bh == NULL)
+		return 0;
 
 	while (1) {
 	      cont:
@@ -285,9 +265,20 @@ static int bmap_hash_id(struct super_block *s, u32 id)
  */
 static inline int block_group_used(struct super_block *s, u32 id)
 {
-	int bm;
-	bm = bmap_hash_id(s, id);
-	if (SB_AP_BITMAP(s)[bm].free_count > ((s->s_blocksize << 3) * 60 / 100)) {
+	int bm = bmap_hash_id(s, id);
+	struct reiserfs_bitmap_info *info = &SB_AP_BITMAP(s)[bm];
+
+	/* If we don't have cached information on this bitmap block, we're
+	 * going to have to load it later anyway. Loading it here allows us
+	 * to make a better decision. This favors long-term performace gain
+	 * with a better on-disk layout vs. a short term gain of skipping the
+	 * read and potentially having a bad placement. */
+	if (info->first_zero_hint == 0) {
+		struct buffer_head *bh = reiserfs_read_bitmap_block(s, bm);
+		brelse(bh);
+	}
+
+	if (info->free_count > ((s->s_blocksize << 3) * 60 / 100)) {
 		return 0;
 	}
 	return 1;
@@ -413,8 +404,9 @@ static void _reiserfs_free_block(struct reiserfs_transaction_handle *th,
 		return;
 	}
 
-	bmbh = apbi[nr].bh;
-	get_bh(bmbh);
+	bmbh = reiserfs_read_bitmap_block(s, nr);
+	if (!bmbh)
+		return;
 
 	reiserfs_prepare_for_journal(s, bmbh, 1);
 
@@ -1320,6 +1312,7 @@ struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
                                                unsigned int bitmap)
 {
 	b_blocknr_t block = (sb->s_blocksize << 3) * bitmap;
+	struct reiserfs_bitmap_info *info = SB_AP_BITMAP(sb) + bitmap;
 	struct buffer_head *bh;
 
 	/* Way old format filesystems had the bitmaps packed up front.
@@ -1330,9 +1323,21 @@ struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
 	else if (bitmap == 0)
 		block = (REISERFS_DISK_OFFSET_IN_BYTES >> sb->s_blocksize_bits) + 1;
 
-	bh = sb_getblk(sb, block);
-	if (!buffer_uptodate(bh))
-		ll_rw_block(READ, 1, &bh);
+	bh = sb_bread(sb, block);
+	if (bh == NULL)
+		reiserfs_warning(sb, "sh-2029: %s: bitmap block (#%lu) "
+		                 "reading failed", __FUNCTION__, bh->b_blocknr);
+	else {
+		if (buffer_locked(bh)) {
+			PROC_INFO_INC(sb, scan_bitmap.wait);
+			__wait_on_buffer(bh);
+		}
+		BUG_ON(!buffer_uptodate(bh));
+		BUG_ON(atomic_read(&bh->b_count) == 0);
+
+		if (info->first_zero_hint == 0)
+			reiserfs_cache_bitmap_metadata(sb, bh, info);
+	}
 
 	return bh;
 }
@@ -1340,7 +1345,6 @@ struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
 int reiserfs_init_bitmap_cache(struct super_block *sb)
 {
 	struct reiserfs_bitmap_info *bitmap;
-	int i;
 
 	bitmap = vmalloc(sizeof (*bitmap) * SB_BMAP_NR(sb));
 	if (bitmap == NULL)
@@ -1348,28 +1352,15 @@ int reiserfs_init_bitmap_cache(struct super_block *sb)
 
 	memset(bitmap, 0, sizeof (*bitmap) * SB_BMAP_NR(sb));
 
-	for (i = 0; i < SB_BMAP_NR(sb); i++)
-		bitmap[i].bh = reiserfs_read_bitmap_block(sb, i);
-
-	/* make sure we have them all */
-	for (i = 0; i < SB_BMAP_NR(sb); i++) {
-		wait_on_buffer(bitmap[i].bh);
-		if (!buffer_uptodate(bitmap[i].bh)) {
-			reiserfs_warning(sb, "sh-2029: %s: "
-					 "bitmap block (#%lu) reading failed",
-			                 __FUNCTION__, bitmap[i].bh->b_blocknr);
-			for (i = 0; i < SB_BMAP_NR(sb); i++)
-				brelse(bitmap[i].bh);
-			vfree(bitmap);
-			return -EIO;
-		}
-	}
-
-	/* Cache the info on the bitmaps before we get rolling */
-	for (i = 0; i < SB_BMAP_NR(sb); i++)
-		reiserfs_cache_bitmap_metadata(sb, bitmap[i].bh, &bitmap[i]);
-
 	SB_AP_BITMAP(sb) = bitmap;
 
 	return 0;
 }
+
+void reiserfs_free_bitmap_cache(struct super_block *sb)
+{
+	if (SB_AP_BITMAP(sb)) {
+		vfree(SB_AP_BITMAP(sb));
+		SB_AP_BITMAP(sb) = NULL;
+	}
+}
diff --git a/fs/reiserfs/resize.c b/fs/reiserfs/resize.c
index 90d39fd3096f..315684793d1d 100644
--- a/fs/reiserfs/resize.c
+++ b/fs/reiserfs/resize.c
@@ -128,8 +128,9 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 		 * transaction begins, and the new bitmaps don't matter if the
 		 * transaction fails. */
 		for (i = bmap_nr; i < bmap_nr_new; i++) {
-			bh = sb_getblk(s, i * s->s_blocksize * 8);
-			get_bh(bh);
+			/* don't use read_bitmap_block since it will cache
+			 * the uninitialized bitmap */
+			bh = sb_bread(s, i * s->s_blocksize * 8);
 			memset(bh->b_data, 0, sb_blocksize(sb));
 			reiserfs_test_and_set_le_bit(0, bh->b_data);
 			reiserfs_cache_bitmap_metadata(s, bh, bitmap + i);
@@ -140,7 +141,6 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 			// update bitmap_info stuff
 			bitmap[i].first_zero_hint = 1;
 			bitmap[i].free_count = sb_blocksize(sb) * 8 - 1;
-			bitmap[i].bh = bh;
 			brelse(bh);
 		}
 		/* free old bitmap blocks array */
@@ -157,8 +157,13 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 
 	/* Extend old last bitmap block - new blocks have been made available */
 	info = SB_AP_BITMAP(s) + bmap_nr - 1;
-	bh = info->bh;
-	get_bh(bh);
+	bh = reiserfs_read_bitmap_block(s, bmap_nr - 1);
+	if (!bh) {
+		int jerr = journal_end(&th, s, 10);
+		if (jerr)
+			return jerr;
+		return -EIO;
+	}
 
 	reiserfs_prepare_for_journal(s, bh, 1);
 	for (i = block_r; i < s->s_blocksize * 8; i++)
@@ -172,8 +177,13 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 
 	/* Correct new last bitmap block - It may not be full */
 	info = SB_AP_BITMAP(s) + bmap_nr_new - 1;
-	bh = info->bh;
-	get_bh(bh);
+	bh = reiserfs_read_bitmap_block(s, bmap_nr_new - 1);
+	if (!bh) {
+		int jerr = journal_end(&th, s, 10);
+		if (jerr)
+			return jerr;
+		return -EIO;
+	}
 
 	reiserfs_prepare_for_journal(s, bh, 1);
 	for (i = block_r_new; i < s->s_blocksize * 8; i++)
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index c78e99e196fa..c89aa2338191 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -432,7 +432,6 @@ int remove_save_link(struct inode *inode, int truncate)
 
 static void reiserfs_put_super(struct super_block *s)
 {
-	int i;
 	struct reiserfs_transaction_handle th;
 	th.t_trans_id = 0;
 
@@ -462,10 +461,7 @@ static void reiserfs_put_super(struct super_block *s)
 	 */
 	journal_release(&th, s);
 
-	for (i = 0; i < SB_BMAP_NR(s); i++)
-		brelse(SB_AP_BITMAP(s)[i].bh);
-
-	vfree(SB_AP_BITMAP(s));
+	reiserfs_free_bitmap_cache(s);
 
 	brelse(SB_BUFFER_WITH_SB(s));
 
@@ -1344,7 +1340,6 @@ static int read_super_block(struct super_block *s, int offset)
 /* after journal replay, reread all bitmap and super blocks */
 static int reread_meta_blocks(struct super_block *s)
 {
-	int i;
 	ll_rw_block(READ, 1, &(SB_BUFFER_WITH_SB(s)));
 	wait_on_buffer(SB_BUFFER_WITH_SB(s));
 	if (!buffer_uptodate(SB_BUFFER_WITH_SB(s))) {
@@ -1353,20 +1348,7 @@ static int reread_meta_blocks(struct super_block *s)
 		return 1;
 	}
 
-	for (i = 0; i < SB_BMAP_NR(s); i++) {
-		ll_rw_block(READ, 1, &(SB_AP_BITMAP(s)[i].bh));
-		wait_on_buffer(SB_AP_BITMAP(s)[i].bh);
-		if (!buffer_uptodate(SB_AP_BITMAP(s)[i].bh)) {
-			reiserfs_warning(s,
-					 "reread_meta_blocks, error reading bitmap block number %d at %llu",
-					 i,
-					 (unsigned long long)SB_AP_BITMAP(s)[i].
-					 bh->b_blocknr);
-			return 1;
-		}
-	}
 	return 0;
-
 }
 
 /////////////////////////////////////////////////////
@@ -1547,7 +1529,6 @@ static int function2code(hashf_t func)
 static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 {
 	struct inode *root_inode;
-	int j;
 	struct reiserfs_transaction_handle th;
 	int old_format = 0;
 	unsigned long blocks;
@@ -1793,19 +1774,17 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	if (jinit_done) {	/* kill the commit thread, free journal ram */
 		journal_release_error(NULL, s);
 	}
-	if (SB_DISK_SUPER_BLOCK(s)) {
-		for (j = 0; j < SB_BMAP_NR(s); j++) {
-			if (SB_AP_BITMAP(s))
-				brelse(SB_AP_BITMAP(s)[j].bh);
-		}
-		vfree(SB_AP_BITMAP(s));
-	}
+
+	reiserfs_free_bitmap_cache(s);
 	if (SB_BUFFER_WITH_SB(s))
 		brelse(SB_BUFFER_WITH_SB(s));
 #ifdef CONFIG_QUOTA
-	for (j = 0; j < MAXQUOTAS; j++) {
-		kfree(sbi->s_qf_names[j]);
-		sbi->s_qf_names[j] = NULL;
+	{
+		int j;
+		for (j = 0; j < MAXQUOTAS; j++) {
+			kfree(sbi->s_qf_names[j]);
+			sbi->s_qf_names[j] = NULL;
+		}
 	}
 #endif
 	kfree(sbi);
diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index 4f21ad388c79..73e0becec086 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -267,7 +267,6 @@ struct reiserfs_bitmap_info {
 	// FIXME: Won't work with block sizes > 8K
 	__u16 first_zero_hint;
 	__u16 free_count;
-	struct buffer_head *bh;	/* the actual bitmap */
 };
 
 struct proc_dir_entry;
-- 
cgit v1.2.3


From 027445c37282bc1ed26add45e573ad2d3e4860a5 Mon Sep 17 00:00:00 2001
From: Badari Pulavarty <pbadari@us.ibm.com>
Date: Sat, 30 Sep 2006 23:28:46 -0700
Subject: [PATCH] Vectorize aio_read/aio_write fileop methods

This patch vectorizes aio_read() and aio_write() methods to prepare for
collapsing all aio & vectored operations into one interface - which is
aio_read()/aio_write().

Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Michael Holzheu <HOLZHEU@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/filesystems/Locking |  5 +--
 Documentation/filesystems/vfs.txt |  4 +-
 arch/s390/hypfs/inode.c           | 17 +++++++--
 drivers/char/raw.c                | 14 +------
 drivers/usb/gadget/inode.c        | 80 +++++++++++++++++++++++++++------------
 fs/aio.c                          | 15 ++++++--
 fs/block_dev.c                    | 10 +----
 fs/cifs/cifsfs.c                  |  6 +--
 fs/ext3/file.c                    |  5 ++-
 fs/nfs/direct.c                   | 26 ++++++++++---
 fs/nfs/file.c                     | 34 +++++++++--------
 fs/ntfs/file.c                    |  8 ++--
 fs/ocfs2/file.c                   | 28 +++++++-------
 fs/read_write.c                   | 20 ++++++++--
 fs/reiserfs/file.c                |  4 +-
 fs/xfs/linux-2.6/xfs_file.c       | 46 +++++++++++-----------
 include/linux/aio.h               |  2 +
 include/linux/fs.h                | 10 ++---
 include/linux/nfs_fs.h            | 10 +++--
 include/net/sock.h                |  1 -
 mm/filemap.c                      | 39 +++++++++----------
 net/socket.c                      | 49 +++++++++++-------------
 22 files changed, 242 insertions(+), 191 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 247d7f619aa2..eb1a6cad21e6 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -356,10 +356,9 @@ The last two are called only from check_disk_change().
 prototypes:
 	loff_t (*llseek) (struct file *, loff_t, int);
 	ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
-	ssize_t (*aio_read) (struct kiocb *, char __user *, size_t, loff_t);
 	ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
-	ssize_t (*aio_write) (struct kiocb *, const char __user *, size_t,
-			loff_t);
+	ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
+	ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
 	int (*readdir) (struct file *, void *, filldir_t);
 	unsigned int (*poll) (struct file *, struct poll_table_struct *);
 	int (*ioctl) (struct inode *, struct file *, unsigned int,
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 1cb7e8be927a..cd07c21b8400 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -699,9 +699,9 @@ This describes how the VFS can manipulate an open file. As of kernel
 struct file_operations {
 	loff_t (*llseek) (struct file *, loff_t, int);
 	ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
-	ssize_t (*aio_read) (struct kiocb *, char __user *, size_t, loff_t);
 	ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
-	ssize_t (*aio_write) (struct kiocb *, const char __user *, size_t, loff_t);
+	ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
+	ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
 	int (*readdir) (struct file *, void *, filldir_t);
 	unsigned int (*poll) (struct file *, struct poll_table_struct *);
 	int (*ioctl) (struct inode *, struct file *, unsigned int, unsigned long);
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index 813fc21358f9..cd702ae45d6d 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -134,12 +134,20 @@ static int hypfs_open(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-static ssize_t hypfs_aio_read(struct kiocb *iocb, __user char *buf,
-			      size_t count, loff_t offset)
+static ssize_t hypfs_aio_read(struct kiocb *iocb, const struct iovec *iov,
+			      unsigned long nr_segs, loff_t offset)
 {
 	char *data;
 	size_t len;
 	struct file *filp = iocb->ki_filp;
+	/* XXX: temporary */
+	char __user *buf = iov[0].iov_base;
+	size_t count = iov[0].iov_len;
+
+	if (nr_segs != 1) {
+		count = -EINVAL;
+		goto out;
+	}
 
 	data = filp->private_data;
 	len = strlen(data);
@@ -158,12 +166,13 @@ static ssize_t hypfs_aio_read(struct kiocb *iocb, __user char *buf,
 out:
 	return count;
 }
-static ssize_t hypfs_aio_write(struct kiocb *iocb, const char __user *buf,
-			       size_t count, loff_t pos)
+static ssize_t hypfs_aio_write(struct kiocb *iocb, const struct iovec *iov,
+			      unsigned long nr_segs, loff_t offset)
 {
 	int rc;
 	struct super_block *sb;
 	struct hypfs_sb_info *fs_info;
+	size_t count = iov_length(iov, nr_segs);
 
 	sb = iocb->ki_filp->f_dentry->d_inode->i_sb;
 	fs_info = sb->s_fs_info;
diff --git a/drivers/char/raw.c b/drivers/char/raw.c
index c596a08c07b3..173fb08555d5 100644
--- a/drivers/char/raw.c
+++ b/drivers/char/raw.c
@@ -249,23 +249,11 @@ static ssize_t raw_file_write(struct file *file, const char __user *buf,
 	return generic_file_write_nolock(file, &local_iov, 1, ppos);
 }
 
-static ssize_t raw_file_aio_write(struct kiocb *iocb, const char __user *buf,
-					size_t count, loff_t pos)
-{
-	struct iovec local_iov = {
-		.iov_base = (char __user *)buf,
-		.iov_len = count
-	};
-
-	return generic_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
-}
-
-
 static const struct file_operations raw_fops = {
 	.read	=	generic_file_read,
 	.aio_read = 	generic_file_aio_read,
 	.write	=	raw_file_write,
-	.aio_write = 	raw_file_aio_write,
+	.aio_write = 	generic_file_aio_write_nolock,
 	.open	=	raw_open,
 	.release=	raw_release,
 	.ioctl	=	raw_ioctl,
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 4a000d846a93..86924f9cdd7e 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -533,7 +533,8 @@ struct kiocb_priv {
 	struct usb_request	*req;
 	struct ep_data		*epdata;
 	void			*buf;
-	char __user		*ubuf;		/* NULL for writes */
+	const struct iovec	*iv;
+	unsigned long		nr_segs;
 	unsigned		actual;
 };
 
@@ -561,17 +562,32 @@ static int ep_aio_cancel(struct kiocb *iocb, struct io_event *e)
 static ssize_t ep_aio_read_retry(struct kiocb *iocb)
 {
 	struct kiocb_priv	*priv = iocb->private;
-	ssize_t			status = priv->actual;
-
-	/* we "retry" to get the right mm context for this: */
-	status = copy_to_user(priv->ubuf, priv->buf, priv->actual);
-	if (unlikely(0 != status))
-		status = -EFAULT;
-	else
-		status = priv->actual;
-	kfree(priv->buf);
-	kfree(priv);
-	return status;
+	ssize_t			len, total;
+	int			i;
+
+  	/* we "retry" to get the right mm context for this: */
+
+ 	/* copy stuff into user buffers */
+ 	total = priv->actual;
+ 	len = 0;
+ 	for (i=0; i < priv->nr_segs; i++) {
+ 		ssize_t this = min((ssize_t)(priv->iv[i].iov_len), total);
+
+ 		if (copy_to_user(priv->iv[i].iov_base, priv->buf, this)) {
+ 			if (len == 0)
+ 				len = -EFAULT;
+ 			break;
+ 		}
+
+ 		total -= this;
+ 		len += this;
+ 		if (total == 0)
+ 			break;
+ 	}
+  	kfree(priv->buf);
+  	kfree(priv);
+  	aio_put_req(iocb);
+ 	return len;
 }
 
 static void ep_aio_complete(struct usb_ep *ep, struct usb_request *req)
@@ -584,7 +600,7 @@ static void ep_aio_complete(struct usb_ep *ep, struct usb_request *req)
 	spin_lock(&epdata->dev->lock);
 	priv->req = NULL;
 	priv->epdata = NULL;
-	if (priv->ubuf == NULL
+	if (priv->iv == NULL
 			|| unlikely(req->actual == 0)
 			|| unlikely(kiocbIsCancelled(iocb))) {
 		kfree(req->buf);
@@ -619,7 +635,8 @@ ep_aio_rwtail(
 	char		*buf,
 	size_t		len,
 	struct ep_data	*epdata,
-	char __user	*ubuf
+	const struct iovec *iv,
+	unsigned long 	nr_segs
 )
 {
 	struct kiocb_priv	*priv;
@@ -634,7 +651,8 @@ fail:
 		return value;
 	}
 	iocb->private = priv;
-	priv->ubuf = ubuf;
+	priv->iv = iv;
+	priv->nr_segs = nr_segs;
 
 	value = get_ready_ep(iocb->ki_filp->f_flags, epdata);
 	if (unlikely(value < 0)) {
@@ -674,41 +692,53 @@ fail:
 		kfree(priv);
 		put_ep(epdata);
 	} else
-		value = (ubuf ? -EIOCBRETRY : -EIOCBQUEUED);
+		value = (iv ? -EIOCBRETRY : -EIOCBQUEUED);
 	return value;
 }
 
 static ssize_t
-ep_aio_read(struct kiocb *iocb, char __user *ubuf, size_t len, loff_t o)
+ep_aio_read(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t o)
 {
 	struct ep_data		*epdata = iocb->ki_filp->private_data;
 	char			*buf;
 
 	if (unlikely(epdata->desc.bEndpointAddress & USB_DIR_IN))
 		return -EINVAL;
-	buf = kmalloc(len, GFP_KERNEL);
+
+	buf = kmalloc(iocb->ki_left, GFP_KERNEL);
 	if (unlikely(!buf))
 		return -ENOMEM;
+
 	iocb->ki_retry = ep_aio_read_retry;
-	return ep_aio_rwtail(iocb, buf, len, epdata, ubuf);
+	return ep_aio_rwtail(iocb, buf, iocb->ki_left, epdata, iov, nr_segs);
 }
 
 static ssize_t
-ep_aio_write(struct kiocb *iocb, const char __user *ubuf, size_t len, loff_t o)
+ep_aio_write(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t o)
 {
 	struct ep_data		*epdata = iocb->ki_filp->private_data;
 	char			*buf;
+	size_t			len = 0;
+	int			i = 0;
 
 	if (unlikely(!(epdata->desc.bEndpointAddress & USB_DIR_IN)))
 		return -EINVAL;
-	buf = kmalloc(len, GFP_KERNEL);
+
+	buf = kmalloc(iocb->ki_left, GFP_KERNEL);
 	if (unlikely(!buf))
 		return -ENOMEM;
-	if (unlikely(copy_from_user(buf, ubuf, len) != 0)) {
-		kfree(buf);
-		return -EFAULT;
+
+	for (i=0; i < nr_segs; i++) {
+		if (unlikely(copy_from_user(&buf[len], iov[i].iov_base,
+				iov[i].iov_len) != 0)) {
+			kfree(buf);
+			return -EFAULT;
+		}
+		len += iov[i].iov_len;
 	}
-	return ep_aio_rwtail(iocb, buf, len, epdata, NULL);
+	return ep_aio_rwtail(iocb, buf, len, epdata, NULL, 0);
 }
 
 /*----------------------------------------------------------------------*/
diff --git a/fs/aio.c b/fs/aio.c
index 950630187acc..27ff56540c73 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -15,6 +15,7 @@
 #include <linux/aio_abi.h>
 #include <linux/module.h>
 #include <linux/syscalls.h>
+#include <linux/uio.h>
 
 #define DEBUG 0
 
@@ -1315,8 +1316,11 @@ static ssize_t aio_pread(struct kiocb *iocb)
 	ssize_t ret = 0;
 
 	do {
-		ret = file->f_op->aio_read(iocb, iocb->ki_buf,
-			iocb->ki_left, iocb->ki_pos);
+		iocb->ki_inline_vec.iov_base = iocb->ki_buf;
+		iocb->ki_inline_vec.iov_len = iocb->ki_left;
+
+		ret = file->f_op->aio_read(iocb, &iocb->ki_inline_vec,
+						1, iocb->ki_pos);
 		/*
 		 * Can't just depend on iocb->ki_left to determine
 		 * whether we are done. This may have been a short read.
@@ -1349,8 +1353,11 @@ static ssize_t aio_pwrite(struct kiocb *iocb)
 	ssize_t ret = 0;
 
 	do {
-		ret = file->f_op->aio_write(iocb, iocb->ki_buf,
-			iocb->ki_left, iocb->ki_pos);
+		iocb->ki_inline_vec.iov_base = iocb->ki_buf;
+		iocb->ki_inline_vec.iov_len = iocb->ki_left;
+
+		ret = file->f_op->aio_write(iocb, &iocb->ki_inline_vec,
+						1, iocb->ki_pos);
 		if (ret > 0) {
 			iocb->ki_buf += ret;
 			iocb->ki_left -= ret;
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 0c361ea7e5a6..8c819117310b 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1162,14 +1162,6 @@ static ssize_t blkdev_file_write(struct file *file, const char __user *buf,
 	return generic_file_write_nolock(file, &local_iov, 1, ppos);
 }
 
-static ssize_t blkdev_file_aio_write(struct kiocb *iocb, const char __user *buf,
-				   size_t count, loff_t pos)
-{
-	struct iovec local_iov = { .iov_base = (void __user *)buf, .iov_len = count };
-
-	return generic_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
-}
-
 static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
 	return blkdev_ioctl(file->f_mapping->host, file, cmd, arg);
@@ -1192,7 +1184,7 @@ const struct file_operations def_blk_fops = {
 	.read		= generic_file_read,
 	.write		= blkdev_file_write,
   	.aio_read	= generic_file_aio_read,
-  	.aio_write	= blkdev_file_aio_write, 
+  	.aio_write	= generic_file_aio_write_nolock,
 	.mmap		= generic_file_mmap,
 	.fsync		= block_fsync,
 	.unlocked_ioctl	= block_ioctl,
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 22bcf4d7e7ae..5abb42a7c53e 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -492,13 +492,13 @@ static ssize_t cifs_file_writev(struct file *file, const struct iovec *iov,
 	return written;
 }
 
-static ssize_t cifs_file_aio_write(struct kiocb *iocb, const char __user *buf,
-				   size_t count, loff_t pos)
+static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
+				   unsigned long nr_segs, loff_t pos)
 {
 	struct inode *inode = iocb->ki_filp->f_dentry->d_inode;
 	ssize_t written;
 
-	written = generic_file_aio_write(iocb, buf, count, pos);
+	written = generic_file_aio_write(iocb, iov, nr_segs, pos);
 	if (!CIFS_I(inode)->clientCanCacheAll)
 		filemap_fdatawrite(inode->i_mapping);
 	return written;
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 74ff20f9d09b..5c762457bc89 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -48,14 +48,15 @@ static int ext3_release_file (struct inode * inode, struct file * filp)
 }
 
 static ssize_t
-ext3_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t pos)
+ext3_file_write(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_dentry->d_inode;
 	ssize_t ret;
 	int err;
 
-	ret = generic_file_aio_write(iocb, buf, count, pos);
+	ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
 
 	/*
 	 * Skip flushing if there was an error, or if nothing was written.
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 377839bed172..9f7f8b9ea1e2 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -707,8 +707,8 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz
 /**
  * nfs_file_direct_read - file direct read operation for NFS files
  * @iocb: target I/O control block
- * @buf: user's buffer into which to read data
- * @count: number of bytes to read
+ * @iov: vector of user buffers into which to read data
+ * @nr_segs: size of iov vector
  * @pos: byte offset in file where reading starts
  *
  * We use this function for direct reads instead of calling
@@ -725,17 +725,24 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz
  * client must read the updated atime from the server back into its
  * cache.
  */
-ssize_t nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, loff_t pos)
+ssize_t nfs_file_direct_read(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos)
 {
 	ssize_t retval = -EINVAL;
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
+	/* XXX: temporary */
+	const char __user *buf = iov[0].iov_base;
+	size_t count = iov[0].iov_len;
 
 	dprintk("nfs: direct read(%s/%s, %lu@%Ld)\n",
 		file->f_dentry->d_parent->d_name.name,
 		file->f_dentry->d_name.name,
 		(unsigned long) count, (long long) pos);
 
+	if (nr_segs != 1)
+		return -EINVAL;
+
 	if (count < 0)
 		goto out;
 	retval = -EFAULT;
@@ -760,8 +767,8 @@ out:
 /**
  * nfs_file_direct_write - file direct write operation for NFS files
  * @iocb: target I/O control block
- * @buf: user's buffer from which to write data
- * @count: number of bytes to write
+ * @iov: vector of user buffers from which to write data
+ * @nr_segs: size of iov vector
  * @pos: byte offset in file where writing starts
  *
  * We use this function for direct writes instead of calling
@@ -782,17 +789,24 @@ out:
  * Note that O_APPEND is not supported for NFS direct writes, as there
  * is no atomic O_APPEND write facility in the NFS protocol.
  */
-ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t pos)
+ssize_t nfs_file_direct_write(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos)
 {
 	ssize_t retval;
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
+	/* XXX: temporary */
+	const char __user *buf = iov[0].iov_base;
+	size_t count = iov[0].iov_len;
 
 	dfprintk(VFS, "nfs: direct write(%s/%s, %lu@%Ld)\n",
 		file->f_dentry->d_parent->d_name.name,
 		file->f_dentry->d_name.name,
 		(unsigned long) count, (long long) pos);
 
+	if (nr_segs != 1)
+		return -EINVAL;
+
 	retval = generic_write_checks(file, &pos, &count, 0);
 	if (retval)
 		goto out;
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index be997d649127..cc93865cea93 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -41,8 +41,10 @@ static int nfs_file_release(struct inode *, struct file *);
 static loff_t nfs_file_llseek(struct file *file, loff_t offset, int origin);
 static int  nfs_file_mmap(struct file *, struct vm_area_struct *);
 static ssize_t nfs_file_sendfile(struct file *, loff_t *, size_t, read_actor_t, void *);
-static ssize_t nfs_file_read(struct kiocb *, char __user *, size_t, loff_t);
-static ssize_t nfs_file_write(struct kiocb *, const char __user *, size_t, loff_t);
+static ssize_t nfs_file_read(struct kiocb *, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos);
+static ssize_t nfs_file_write(struct kiocb *, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos);
 static int  nfs_file_flush(struct file *, fl_owner_t id);
 static int  nfs_fsync(struct file *, struct dentry *dentry, int datasync);
 static int nfs_check_flags(int flags);
@@ -53,8 +55,8 @@ const struct file_operations nfs_file_operations = {
 	.llseek		= nfs_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
-	.aio_read		= nfs_file_read,
-	.aio_write		= nfs_file_write,
+	.aio_read	= nfs_file_read,
+	.aio_write	= nfs_file_write,
 	.mmap		= nfs_file_mmap,
 	.open		= nfs_file_open,
 	.flush		= nfs_file_flush,
@@ -196,15 +198,17 @@ nfs_file_flush(struct file *file, fl_owner_t id)
 }
 
 static ssize_t
-nfs_file_read(struct kiocb *iocb, char __user * buf, size_t count, loff_t pos)
+nfs_file_read(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t pos)
 {
 	struct dentry * dentry = iocb->ki_filp->f_dentry;
 	struct inode * inode = dentry->d_inode;
 	ssize_t result;
+	size_t count = iov_length(iov, nr_segs);
 
 #ifdef CONFIG_NFS_DIRECTIO
 	if (iocb->ki_filp->f_flags & O_DIRECT)
-		return nfs_file_direct_read(iocb, buf, count, pos);
+		return nfs_file_direct_read(iocb, iov, nr_segs, pos);
 #endif
 
 	dfprintk(VFS, "nfs: read(%s/%s, %lu@%lu)\n",
@@ -214,7 +218,7 @@ nfs_file_read(struct kiocb *iocb, char __user * buf, size_t count, loff_t pos)
 	result = nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping);
 	nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, count);
 	if (!result)
-		result = generic_file_aio_read(iocb, buf, count, pos);
+		result = generic_file_aio_read(iocb, iov, nr_segs, pos);
 	return result;
 }
 
@@ -336,24 +340,22 @@ const struct address_space_operations nfs_file_aops = {
 #endif
 };
 
-/* 
- * Write to a file (through the page cache).
- */
-static ssize_t
-nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t pos)
+static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos)
 {
 	struct dentry * dentry = iocb->ki_filp->f_dentry;
 	struct inode * inode = dentry->d_inode;
 	ssize_t result;
+	size_t count = iov_length(iov, nr_segs);
 
 #ifdef CONFIG_NFS_DIRECTIO
 	if (iocb->ki_filp->f_flags & O_DIRECT)
-		return nfs_file_direct_write(iocb, buf, count, pos);
+		return nfs_file_direct_write(iocb, iov, nr_segs, pos);
 #endif
 
-	dfprintk(VFS, "nfs: write(%s/%s(%ld), %lu@%lu)\n",
+	dfprintk(VFS, "nfs: write(%s/%s(%ld), %lu@%Ld)\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name,
-		inode->i_ino, (unsigned long) count, (unsigned long) pos);
+		inode->i_ino, (unsigned long) count, (long long) pos);
 
 	result = -EBUSY;
 	if (IS_SWAPFILE(inode))
@@ -372,7 +374,7 @@ nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t
 		goto out;
 
 	nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count);
-	result = generic_file_aio_write(iocb, buf, count, pos);
+	result = generic_file_aio_write(iocb, iov, nr_segs, pos);
 out:
 	return result;
 
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 585a79d39c9d..0c46f5c86b71 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -2176,20 +2176,18 @@ out:
 /**
  * ntfs_file_aio_write -
  */
-static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const char __user *buf,
-		size_t count, loff_t pos)
+static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
 	struct inode *inode = mapping->host;
 	ssize_t ret;
-	struct iovec local_iov = { .iov_base = (void __user *)buf,
-				   .iov_len = count };
 
 	BUG_ON(iocb->ki_pos != pos);
 
 	mutex_lock(&inode->i_mutex);
-	ret = ntfs_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
+	ret = ntfs_file_aio_write_nolock(iocb, iov, nr_segs, &iocb->ki_pos);
 	mutex_unlock(&inode->i_mutex);
 	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
 		int err = sync_page_range(inode, mapping, pos, ret);
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 2bbfa17090cf..d9ba0a931a03 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -961,25 +961,23 @@ static inline int ocfs2_write_should_remove_suid(struct inode *inode)
 }
 
 static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
-				    const char __user *buf,
-				    size_t count,
+				    const struct iovec *iov,
+				    unsigned long nr_segs,
 				    loff_t pos)
 {
-	struct iovec local_iov = { .iov_base = (void __user *)buf,
-				   .iov_len = count };
 	int ret, rw_level = -1, meta_level = -1, have_alloc_sem = 0;
 	u32 clusters;
 	struct file *filp = iocb->ki_filp;
 	struct inode *inode = filp->f_dentry->d_inode;
 	loff_t newsize, saved_pos;
 
-	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", filp, buf,
-		   (unsigned int)count,
+	mlog_entry("(0x%p, %u, '%.*s')\n", filp,
+		   (unsigned int)nr_segs,
 		   filp->f_dentry->d_name.len,
 		   filp->f_dentry->d_name.name);
 
 	/* happy write of zero bytes */
-	if (count == 0)
+	if (iocb->ki_left == 0)
 		return 0;
 
 	if (!inode) {
@@ -1048,7 +1046,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
 		} else {
 			saved_pos = iocb->ki_pos;
 		}
-		newsize = count + saved_pos;
+		newsize = iocb->ki_left + saved_pos;
 
 		mlog(0, "pos=%lld newsize=%lld cursize=%lld\n",
 		     (long long) saved_pos, (long long) newsize,
@@ -1081,7 +1079,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
 		if (!clusters)
 			break;
 
-		ret = ocfs2_extend_file(inode, NULL, newsize, count);
+		ret = ocfs2_extend_file(inode, NULL, newsize, iocb->ki_left);
 		if (ret < 0) {
 			if (ret != -ENOSPC)
 				mlog_errno(ret);
@@ -1098,7 +1096,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
 	/* communicate with ocfs2_dio_end_io */
 	ocfs2_iocb_set_rw_locked(iocb);
 
-	ret = generic_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos);
+	ret = generic_file_aio_write_nolock(iocb, iov, nr_segs, iocb->ki_pos);
 
 	/* buffered aio wouldn't have proper lock coverage today */
 	BUG_ON(ret == -EIOCBQUEUED && !(filp->f_flags & O_DIRECT));
@@ -1132,16 +1130,16 @@ out:
 }
 
 static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
-				   char __user *buf,
-				   size_t count,
+				   const struct iovec *iov,
+				   unsigned long nr_segs,
 				   loff_t pos)
 {
 	int ret = 0, rw_level = -1, have_alloc_sem = 0;
 	struct file *filp = iocb->ki_filp;
 	struct inode *inode = filp->f_dentry->d_inode;
 
-	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", filp, buf,
-		   (unsigned int)count,
+	mlog_entry("(0x%p, %u, '%.*s')\n", filp,
+		   (unsigned int)nr_segs,
 		   filp->f_dentry->d_name.len,
 		   filp->f_dentry->d_name.name);
 
@@ -1185,7 +1183,7 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
 	}
 	ocfs2_meta_unlock(inode, 0);
 
-	ret = generic_file_aio_read(iocb, buf, count, iocb->ki_pos);
+	ret = generic_file_aio_read(iocb, iov, nr_segs, iocb->ki_pos);
 	if (ret == -EINVAL)
 		mlog(ML_ERROR, "generic_file_aio_read returned -EINVAL\n");
 
diff --git a/fs/read_write.c b/fs/read_write.c
index d4cb3183c99c..679dd535380f 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -227,14 +227,20 @@ static void wait_on_retry_sync_kiocb(struct kiocb *iocb)
 
 ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
 {
+	struct iovec iov = { .iov_base = buf, .iov_len = len };
 	struct kiocb kiocb;
 	ssize_t ret;
 
 	init_sync_kiocb(&kiocb, filp);
 	kiocb.ki_pos = *ppos;
-	while (-EIOCBRETRY ==
-		(ret = filp->f_op->aio_read(&kiocb, buf, len, kiocb.ki_pos)))
+	kiocb.ki_left = len;
+
+	for (;;) {
+		ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
+		if (ret != -EIOCBRETRY)
+			break;
 		wait_on_retry_sync_kiocb(&kiocb);
+	}
 
 	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&kiocb);
@@ -279,14 +285,20 @@ EXPORT_SYMBOL(vfs_read);
 
 ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
 {
+	struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = len };
 	struct kiocb kiocb;
 	ssize_t ret;
 
 	init_sync_kiocb(&kiocb, filp);
 	kiocb.ki_pos = *ppos;
-	while (-EIOCBRETRY ==
-	       (ret = filp->f_op->aio_write(&kiocb, buf, len, kiocb.ki_pos)))
+	kiocb.ki_left = len;
+
+	for (;;) {
+		ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
+		if (ret != -EIOCBRETRY)
+			break;
 		wait_on_retry_sync_kiocb(&kiocb);
+	}
 
 	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&kiocb);
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index c11f6118c9ca..41f24369e47a 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -1334,7 +1334,7 @@ static ssize_t reiserfs_file_write(struct file *file,	/* the file we are going t
 			if (err)
 				return err;
 		}
-		result = generic_file_write(file, buf, count, ppos);
+		result = do_sync_write(file, buf, count, ppos);
 
 		if (after_file_end) {	/* Now update i_size and remove the savelink */
 			struct reiserfs_transaction_handle th;
@@ -1566,7 +1566,7 @@ static ssize_t reiserfs_file_write(struct file *file,	/* the file we are going t
 }
 
 const struct file_operations reiserfs_file_operations = {
-	.read = generic_file_read,
+	.read = do_sync_read,
 	.write = reiserfs_file_write,
 	.ioctl = reiserfs_ioctl,
 #ifdef CONFIG_COMPAT
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 41cfcba7ce49..4737971c6a39 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -49,50 +49,49 @@ static struct vm_operations_struct xfs_dmapi_file_vm_ops;
 STATIC inline ssize_t
 __xfs_file_read(
 	struct kiocb		*iocb,
-	char			__user *buf,
+	const struct iovec	*iov,
+	unsigned long		nr_segs,
 	int			ioflags,
-	size_t			count,
 	loff_t			pos)
 {
-	struct iovec		iov = {buf, count};
 	struct file		*file = iocb->ki_filp;
 	bhv_vnode_t		*vp = vn_from_inode(file->f_dentry->d_inode);
 
 	BUG_ON(iocb->ki_pos != pos);
 	if (unlikely(file->f_flags & O_DIRECT))
 		ioflags |= IO_ISDIRECT;
-	return bhv_vop_read(vp, iocb, &iov, 1, &iocb->ki_pos, ioflags, NULL);
+	return bhv_vop_read(vp, iocb, iov, nr_segs, &iocb->ki_pos,
+				ioflags, NULL);
 }
 
 STATIC ssize_t
 xfs_file_aio_read(
 	struct kiocb		*iocb,
-	char			__user *buf,
-	size_t			count,
+	const struct iovec	*iov,
+	unsigned long		nr_segs,
 	loff_t			pos)
 {
-	return __xfs_file_read(iocb, buf, IO_ISAIO, count, pos);
+	return __xfs_file_read(iocb, iov, nr_segs, IO_ISAIO, pos);
 }
 
 STATIC ssize_t
 xfs_file_aio_read_invis(
 	struct kiocb		*iocb,
-	char			__user *buf,
-	size_t			count,
+	const struct iovec	*iov,
+	unsigned long		nr_segs,
 	loff_t			pos)
 {
-	return __xfs_file_read(iocb, buf, IO_ISAIO|IO_INVIS, count, pos);
+	return __xfs_file_read(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
 }
 
 STATIC inline ssize_t
 __xfs_file_write(
-	struct kiocb	*iocb,
-	const char	__user *buf,
-	int		ioflags,
-	size_t		count,
-	loff_t		pos)
+	struct kiocb		*iocb,
+	const struct iovec	*iov,
+	unsigned long		nr_segs,
+	int			ioflags,
+	loff_t			pos)
 {
-	struct iovec	iov = {(void __user *)buf, count};
 	struct file	*file = iocb->ki_filp;
 	struct inode	*inode = file->f_mapping->host;
 	bhv_vnode_t	*vp = vn_from_inode(inode);
@@ -100,27 +99,28 @@ __xfs_file_write(
 	BUG_ON(iocb->ki_pos != pos);
 	if (unlikely(file->f_flags & O_DIRECT))
 		ioflags |= IO_ISDIRECT;
-	return bhv_vop_write(vp, iocb, &iov, 1, &iocb->ki_pos, ioflags, NULL);
+	return bhv_vop_write(vp, iocb, iov, nr_segs, &iocb->ki_pos,
+				ioflags, NULL);
 }
 
 STATIC ssize_t
 xfs_file_aio_write(
 	struct kiocb		*iocb,
-	const char		__user *buf,
-	size_t			count,
+	const struct iovec	*iov,
+	unsigned long		nr_segs,
 	loff_t			pos)
 {
-	return __xfs_file_write(iocb, buf, IO_ISAIO, count, pos);
+	return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO, pos);
 }
 
 STATIC ssize_t
 xfs_file_aio_write_invis(
 	struct kiocb		*iocb,
-	const char		__user *buf,
-	size_t			count,
+	const struct iovec	*iov,
+	unsigned long		nr_segs,
 	loff_t			pos)
 {
-	return __xfs_file_write(iocb, buf, IO_ISAIO|IO_INVIS, count, pos);
+	return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
 }
 
 STATIC inline ssize_t
diff --git a/include/linux/aio.h b/include/linux/aio.h
index 8a0193385a9b..58349e58b749 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -4,6 +4,7 @@
 #include <linux/list.h>
 #include <linux/workqueue.h>
 #include <linux/aio_abi.h>
+#include <linux/uio.h>
 
 #include <asm/atomic.h>
 
@@ -112,6 +113,7 @@ struct kiocb {
 	long			ki_retried; 	/* just for testing */
 	long			ki_kicked; 	/* just for testing */
 	long			ki_queued; 	/* just for testing */
+	struct iovec		ki_inline_vec;	/* inline vector */
 
 	struct list_head	ki_list;	/* the aio core uses this
 						 * for cancellation */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 5baf3a153403..257bae16f545 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1097,9 +1097,9 @@ struct file_operations {
 	struct module *owner;
 	loff_t (*llseek) (struct file *, loff_t, int);
 	ssize_t (*read) (struct file *, char __user *, size_t, loff_t *);
-	ssize_t (*aio_read) (struct kiocb *, char __user *, size_t, loff_t);
 	ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *);
-	ssize_t (*aio_write) (struct kiocb *, const char __user *, size_t, loff_t);
+	ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
+	ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t);
 	int (*readdir) (struct file *, void *, filldir_t);
 	unsigned int (*poll) (struct file *, struct poll_table_struct *);
 	int (*ioctl) (struct inode *, struct file *, unsigned int, unsigned long);
@@ -1704,11 +1704,11 @@ extern int file_send_actor(read_descriptor_t * desc, struct page *page, unsigned
 extern ssize_t generic_file_read(struct file *, char __user *, size_t, loff_t *);
 int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk);
 extern ssize_t generic_file_write(struct file *, const char __user *, size_t, loff_t *);
-extern ssize_t generic_file_aio_read(struct kiocb *, char __user *, size_t, loff_t);
+extern ssize_t generic_file_aio_read(struct kiocb *, const struct iovec *, unsigned long, loff_t);
 extern ssize_t __generic_file_aio_read(struct kiocb *, const struct iovec *, unsigned long, loff_t *);
-extern ssize_t generic_file_aio_write(struct kiocb *, const char __user *, size_t, loff_t);
+extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, loff_t);
 extern ssize_t generic_file_aio_write_nolock(struct kiocb *, const struct iovec *,
-		unsigned long, loff_t *);
+		unsigned long, loff_t);
 extern ssize_t generic_file_direct_write(struct kiocb *, const struct iovec *,
 		unsigned long *, loff_t, loff_t *, size_t, size_t);
 extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *,
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 98c9b9f667a5..76ff54846ada 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -367,10 +367,12 @@ extern int nfs3_removexattr (struct dentry *, const char *name);
  */
 extern ssize_t nfs_direct_IO(int, struct kiocb *, const struct iovec *, loff_t,
 			unsigned long);
-extern ssize_t nfs_file_direct_read(struct kiocb *iocb, char __user *buf,
-			size_t count, loff_t pos);
-extern ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf,
-			size_t count, loff_t pos);
+extern ssize_t nfs_file_direct_read(struct kiocb *iocb,
+			const struct iovec *iov, unsigned long nr_segs,
+			loff_t pos);
+extern ssize_t nfs_file_direct_write(struct kiocb *iocb,
+			const struct iovec *iov, unsigned long nr_segs,
+			loff_t pos);
 
 /*
  * linux/fs/nfs/dir.c
diff --git a/include/net/sock.h b/include/net/sock.h
index edd4d73ce7f5..40bb90ebb2d1 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -665,7 +665,6 @@ struct sock_iocb {
 	struct sock		*sk;
 	struct scm_cookie	*scm;
 	struct msghdr		*msg, async_msg;
-	struct iovec		async_iov;
 	struct kiocb		*kiocb;
 };
 
diff --git a/mm/filemap.c b/mm/filemap.c
index c4fe97f5ace0..f6c1d22b504f 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1226,12 +1226,11 @@ out:
 EXPORT_SYMBOL(__generic_file_aio_read);
 
 ssize_t
-generic_file_aio_read(struct kiocb *iocb, char __user *buf, size_t count, loff_t pos)
+generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t pos)
 {
-	struct iovec local_iov = { .iov_base = buf, .iov_len = count };
-
 	BUG_ON(iocb->ki_pos != pos);
-	return __generic_file_aio_read(iocb, &local_iov, 1, &iocb->ki_pos);
+	return __generic_file_aio_read(iocb, iov, nr_segs, &iocb->ki_pos);
 }
 EXPORT_SYMBOL(generic_file_aio_read);
 
@@ -2315,22 +2314,22 @@ out:
 	current->backing_dev_info = NULL;
 	return written ? written : err;
 }
-EXPORT_SYMBOL(generic_file_aio_write_nolock);
 
-ssize_t
-generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
-				unsigned long nr_segs, loff_t *ppos)
+ssize_t generic_file_aio_write_nolock(struct kiocb *iocb,
+		const struct iovec *iov, unsigned long nr_segs, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
 	struct inode *inode = mapping->host;
 	ssize_t ret;
-	loff_t pos = *ppos;
 
-	ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs, ppos);
+	BUG_ON(iocb->ki_pos != pos);
+
+	ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs,
+			&iocb->ki_pos);
 
 	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
-		int err;
+		ssize_t err;
 
 		err = sync_page_range_nolock(inode, mapping, pos, ret);
 		if (err < 0)
@@ -2338,6 +2337,7 @@ generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
 	}
 	return ret;
 }
+EXPORT_SYMBOL(generic_file_aio_write_nolock);
 
 static ssize_t
 __generic_file_write_nolock(struct file *file, const struct iovec *iov,
@@ -2347,8 +2347,9 @@ __generic_file_write_nolock(struct file *file, const struct iovec *iov,
 	ssize_t ret;
 
 	init_sync_kiocb(&kiocb, file);
+	kiocb.ki_pos = *ppos;
 	ret = __generic_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
-	if (ret == -EIOCBQUEUED)
+	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&kiocb);
 	return ret;
 }
@@ -2361,28 +2362,28 @@ generic_file_write_nolock(struct file *file, const struct iovec *iov,
 	ssize_t ret;
 
 	init_sync_kiocb(&kiocb, file);
-	ret = generic_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
+	kiocb.ki_pos = *ppos;
+	ret = generic_file_aio_write_nolock(&kiocb, iov, nr_segs, *ppos);
 	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&kiocb);
+	*ppos = kiocb.ki_pos;
 	return ret;
 }
 EXPORT_SYMBOL(generic_file_write_nolock);
 
-ssize_t generic_file_aio_write(struct kiocb *iocb, const char __user *buf,
-			       size_t count, loff_t pos)
+ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t pos)
 {
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
 	struct inode *inode = mapping->host;
 	ssize_t ret;
-	struct iovec local_iov = { .iov_base = (void __user *)buf,
-					.iov_len = count };
 
 	BUG_ON(iocb->ki_pos != pos);
 
 	mutex_lock(&inode->i_mutex);
-	ret = __generic_file_aio_write_nolock(iocb, &local_iov, 1,
-						&iocb->ki_pos);
+	ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs,
+			&iocb->ki_pos);
 	mutex_unlock(&inode->i_mutex);
 
 	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
diff --git a/net/socket.c b/net/socket.c
index 1bc4167e0da8..df92e4252749 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -95,10 +95,10 @@
 #include <linux/netfilter.h>
 
 static int sock_no_open(struct inode *irrelevant, struct file *dontcare);
-static ssize_t sock_aio_read(struct kiocb *iocb, char __user *buf,
-			     size_t size, loff_t pos);
-static ssize_t sock_aio_write(struct kiocb *iocb, const char __user *buf,
-			      size_t size, loff_t pos);
+static ssize_t sock_aio_read(struct kiocb *iocb, const struct iovec *iov,
+			 unsigned long nr_segs, loff_t pos);
+static ssize_t sock_aio_write(struct kiocb *iocb, const struct iovec *iov,
+			  unsigned long nr_segs, loff_t pos);
 static int sock_mmap(struct file *file, struct vm_area_struct *vma);
 
 static int sock_close(struct inode *inode, struct file *file);
@@ -664,7 +664,6 @@ static ssize_t sock_sendpage(struct file *file, struct page *page,
 }
 
 static struct sock_iocb *alloc_sock_iocb(struct kiocb *iocb,
-					 char __user *ubuf, size_t size,
 					 struct sock_iocb *siocb)
 {
 	if (!is_sync_kiocb(iocb)) {
@@ -675,16 +674,13 @@ static struct sock_iocb *alloc_sock_iocb(struct kiocb *iocb,
 	}
 
 	siocb->kiocb = iocb;
-	siocb->async_iov.iov_base = ubuf;
-	siocb->async_iov.iov_len = size;
-
 	iocb->private = siocb;
 	return siocb;
 }
 
 static ssize_t do_sock_read(struct msghdr *msg, struct kiocb *iocb,
-			    struct file *file, struct iovec *iov,
-			    unsigned long nr_segs)
+		struct file *file, const struct iovec *iov,
+		unsigned long nr_segs)
 {
 	struct socket *sock = file->private_data;
 	size_t size = 0;
@@ -715,32 +711,33 @@ static ssize_t sock_readv(struct file *file, const struct iovec *iov,
 	init_sync_kiocb(&iocb, NULL);
 	iocb.private = &siocb;
 
-	ret = do_sock_read(&msg, &iocb, file, (struct iovec *)iov, nr_segs);
+	ret = do_sock_read(&msg, &iocb, file, iov, nr_segs);
 	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&iocb);
 	return ret;
 }
 
-static ssize_t sock_aio_read(struct kiocb *iocb, char __user *ubuf,
-			     size_t count, loff_t pos)
+static ssize_t sock_aio_read(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos)
 {
 	struct sock_iocb siocb, *x;
 
 	if (pos != 0)
 		return -ESPIPE;
-	if (count == 0)		/* Match SYS5 behaviour */
+
+	if (iocb->ki_left == 0)	/* Match SYS5 behaviour */
 		return 0;
 
-	x = alloc_sock_iocb(iocb, ubuf, count, &siocb);
+
+	x = alloc_sock_iocb(iocb, &siocb);
 	if (!x)
 		return -ENOMEM;
-	return do_sock_read(&x->async_msg, iocb, iocb->ki_filp,
-			    &x->async_iov, 1);
+	return do_sock_read(&x->async_msg, iocb, iocb->ki_filp, iov, nr_segs);
 }
 
 static ssize_t do_sock_write(struct msghdr *msg, struct kiocb *iocb,
-			     struct file *file, struct iovec *iov,
-			     unsigned long nr_segs)
+			struct file *file, const struct iovec *iov,
+			unsigned long nr_segs)
 {
 	struct socket *sock = file->private_data;
 	size_t size = 0;
@@ -773,28 +770,28 @@ static ssize_t sock_writev(struct file *file, const struct iovec *iov,
 	init_sync_kiocb(&iocb, NULL);
 	iocb.private = &siocb;
 
-	ret = do_sock_write(&msg, &iocb, file, (struct iovec *)iov, nr_segs);
+	ret = do_sock_write(&msg, &iocb, file, iov, nr_segs);
 	if (-EIOCBQUEUED == ret)
 		ret = wait_on_sync_kiocb(&iocb);
 	return ret;
 }
 
-static ssize_t sock_aio_write(struct kiocb *iocb, const char __user *ubuf,
-			      size_t count, loff_t pos)
+static ssize_t sock_aio_write(struct kiocb *iocb, const struct iovec *iov,
+			  unsigned long nr_segs, loff_t pos)
 {
 	struct sock_iocb siocb, *x;
 
 	if (pos != 0)
 		return -ESPIPE;
-	if (count == 0)		/* Match SYS5 behaviour */
+
+	if (iocb->ki_left == 0)	/* Match SYS5 behaviour */
 		return 0;
 
-	x = alloc_sock_iocb(iocb, (void __user *)ubuf, count, &siocb);
+	x = alloc_sock_iocb(iocb, &siocb);
 	if (!x)
 		return -ENOMEM;
 
-	return do_sock_write(&x->async_msg, iocb, iocb->ki_filp,
-			     &x->async_iov, 1);
+	return do_sock_write(&x->async_msg, iocb, iocb->ki_filp, iov, nr_segs);
 }
 
 /*
-- 
cgit v1.2.3


From ee0b3e671baff681d69fbf0db33b47603c0a8280 Mon Sep 17 00:00:00 2001
From: Badari Pulavarty <pbadari@us.ibm.com>
Date: Sat, 30 Sep 2006 23:28:47 -0700
Subject: [PATCH] Remove readv/writev methods and use aio_read/aio_write
 instead

This patch removes readv() and writev() methods and replaces them with
aio_read()/aio_write() methods.

Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/raw.c          |   2 -
 drivers/net/tun.c           |  37 +++++-----------
 fs/bad_inode.c              |   2 -
 fs/block_dev.c              |   2 -
 fs/cifs/cifsfs.c            |  16 -------
 fs/compat.c                 |  44 +++++--------------
 fs/ext2/file.c              |   2 -
 fs/ext3/file.c              |   2 -
 fs/fat/file.c               |   2 -
 fs/fuse/dev.c               |  37 +++++-----------
 fs/hostfs/hostfs_kern.c     |   2 -
 fs/jfs/file.c               |   2 -
 fs/ntfs/file.c              |   2 -
 fs/pipe.c                   |  59 ++++++++++----------------
 fs/read_write.c             | 101 +++++++++++++++++++++++++++++---------------
 fs/read_write.h             |  14 ++++++
 fs/xfs/linux-2.6/xfs_file.c |  94 -----------------------------------------
 include/linux/fs.h          |   6 ---
 mm/filemap.c                |  36 ----------------
 net/socket.c                |  40 ------------------
 sound/core/pcm_native.c     |  40 +++++++++---------
 21 files changed, 154 insertions(+), 388 deletions(-)
 create mode 100644 fs/read_write.h

(limited to 'include')

diff --git a/drivers/char/raw.c b/drivers/char/raw.c
index 173fb08555d5..490db531e2d8 100644
--- a/drivers/char/raw.c
+++ b/drivers/char/raw.c
@@ -257,8 +257,6 @@ static const struct file_operations raw_fops = {
 	.open	=	raw_open,
 	.release=	raw_release,
 	.ioctl	=	raw_ioctl,
-	.readv	= 	generic_file_readv,
-	.writev	= 	generic_file_writev,
 	.owner	=	THIS_MODULE,
 };
 
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index de8da6dee1b0..bc0face01d25 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -288,11 +288,10 @@ static inline size_t iov_total(const struct iovec *iv, unsigned long count)
 	return len;
 }
 
-/* Writev */
-static ssize_t tun_chr_writev(struct file * file, const struct iovec *iv,
-			      unsigned long count, loff_t *pos)
+static ssize_t tun_chr_aio_write(struct kiocb *iocb, const struct iovec *iv,
+			      unsigned long count, loff_t pos)
 {
-	struct tun_struct *tun = file->private_data;
+	struct tun_struct *tun = iocb->ki_filp->private_data;
 
 	if (!tun)
 		return -EBADFD;
@@ -302,14 +301,6 @@ static ssize_t tun_chr_writev(struct file * file, const struct iovec *iv,
 	return tun_get_user(tun, (struct iovec *) iv, iov_total(iv, count));
 }
 
-/* Write */
-static ssize_t tun_chr_write(struct file * file, const char __user * buf,
-			     size_t count, loff_t *pos)
-{
-	struct iovec iv = { (void __user *) buf, count };
-	return tun_chr_writev(file, &iv, 1, pos);
-}
-
 /* Put packet to the user space buffer */
 static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
 				       struct sk_buff *skb,
@@ -343,10 +334,10 @@ static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
 	return total;
 }
 
-/* Readv */
-static ssize_t tun_chr_readv(struct file *file, const struct iovec *iv,
-			    unsigned long count, loff_t *pos)
+static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
+			    unsigned long count, loff_t pos)
 {
+	struct file *file = iocb->ki_filp;
 	struct tun_struct *tun = file->private_data;
 	DECLARE_WAITQUEUE(wait, current);
 	struct sk_buff *skb;
@@ -426,14 +417,6 @@ static ssize_t tun_chr_readv(struct file *file, const struct iovec *iv,
 	return ret;
 }
 
-/* Read */
-static ssize_t tun_chr_read(struct file * file, char __user * buf,
-			    size_t count, loff_t *pos)
-{
-	struct iovec iv = { buf, count };
-	return tun_chr_readv(file, &iv, 1, pos);
-}
-
 static void tun_setup(struct net_device *dev)
 {
 	struct tun_struct *tun = netdev_priv(dev);
@@ -764,10 +747,10 @@ static int tun_chr_close(struct inode *inode, struct file *file)
 static struct file_operations tun_fops = {
 	.owner	= THIS_MODULE,
 	.llseek = no_llseek,
-	.read	= tun_chr_read,
-	.readv	= tun_chr_readv,
-	.write	= tun_chr_write,
-	.writev = tun_chr_writev,
+	.read  = do_sync_read,
+	.aio_read  = tun_chr_aio_read,
+	.write = do_sync_write,
+	.aio_write = tun_chr_aio_write,
 	.poll	= tun_chr_poll,
 	.ioctl	= tun_chr_ioctl,
 	.open	= tun_chr_open,
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index 80599ae33966..34e6d7b220c3 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -40,8 +40,6 @@ static const struct file_operations bad_file_ops =
 	.aio_fsync	= EIO_ERROR,
 	.fasync		= EIO_ERROR,
 	.lock		= EIO_ERROR,
-	.readv		= EIO_ERROR,
-	.writev		= EIO_ERROR,
 	.sendfile	= EIO_ERROR,
 	.sendpage	= EIO_ERROR,
 	.get_unmapped_area = EIO_ERROR,
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 8c819117310b..0f143094ef1d 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1191,8 +1191,6 @@ const struct file_operations def_blk_fops = {
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= compat_blkdev_ioctl,
 #endif
-	.readv		= generic_file_readv,
-	.writev		= generic_file_write_nolock,
 	.sendfile	= generic_file_sendfile,
 	.splice_read	= generic_file_splice_read,
 	.splice_write	= generic_file_splice_write,
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 5abb42a7c53e..c00c654f2e11 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -480,18 +480,6 @@ cifs_get_sb(struct file_system_type *fs_type,
 	return simple_set_mnt(mnt, sb);
 }
 
-static ssize_t cifs_file_writev(struct file *file, const struct iovec *iov,
-				unsigned long nr_segs, loff_t *ppos)
-{
-	struct inode *inode = file->f_dentry->d_inode;
-	ssize_t written;
-
-	written = generic_file_writev(file, iov, nr_segs, ppos);
-	if (!CIFS_I(inode)->clientCanCacheAll)
-		filemap_fdatawrite(inode->i_mapping);
-	return written;
-}
-
 static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 				   unsigned long nr_segs, loff_t pos)
 {
@@ -577,8 +565,6 @@ struct inode_operations cifs_symlink_inode_ops = {
 const struct file_operations cifs_file_ops = {
 	.read = do_sync_read,
 	.write = do_sync_write,
-	.readv = generic_file_readv,
-	.writev = cifs_file_writev,
 	.aio_read = generic_file_aio_read,
 	.aio_write = cifs_file_aio_write,
 	.open = cifs_open,
@@ -620,8 +606,6 @@ const struct file_operations cifs_file_direct_ops = {
 const struct file_operations cifs_file_nobrl_ops = {
 	.read = do_sync_read,
 	.write = do_sync_write,
-	.readv = generic_file_readv,
-	.writev = cifs_file_writev,
 	.aio_read = generic_file_aio_read,
 	.aio_write = cifs_file_aio_write,
 	.open = cifs_open,
diff --git a/fs/compat.c b/fs/compat.c
index 122b4e3992b5..6b90bf35f61d 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -70,6 +70,8 @@ int compat_printk(const char *fmt, ...)
 	return ret;
 }
 
+#include "read_write.h"
+
 /*
  * Not all architectures have sys_utime, so implement this in terms
  * of sys_utimes.
@@ -1149,9 +1151,6 @@ static ssize_t compat_do_readv_writev(int type, struct file *file,
 			       const struct compat_iovec __user *uvector,
 			       unsigned long nr_segs, loff_t *pos)
 {
-	typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
-	typedef ssize_t (*iov_fn_t)(struct file *, const struct iovec *, unsigned long, loff_t *);
-
 	compat_ssize_t tot_len;
 	struct iovec iovstack[UIO_FASTIOV];
 	struct iovec *iov=iovstack, *vector;
@@ -1234,39 +1233,18 @@ static ssize_t compat_do_readv_writev(int type, struct file *file,
 	fnv = NULL;
 	if (type == READ) {
 		fn = file->f_op->read;
-		fnv = file->f_op->readv;
+		fnv = file->f_op->aio_read;
 	} else {
 		fn = (io_fn_t)file->f_op->write;
-		fnv = file->f_op->writev;
-	}
-	if (fnv) {
-		ret = fnv(file, iov, nr_segs, pos);
-		goto out;
+		fnv = file->f_op->aio_write;
 	}
 
-	/* Do it by hand, with file-ops */
-	ret = 0;
-	vector = iov;
-	while (nr_segs > 0) {
-		void __user * base;
-		size_t len;
-		ssize_t nr;
-
-		base = vector->iov_base;
-		len = vector->iov_len;
-		vector++;
-		nr_segs--;
-
-		nr = fn(file, base, len, pos);
+	if (fnv)
+		ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
+						pos, fnv);
+	else
+		ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn);
 
-		if (nr < 0) {
-			if (!ret) ret = nr;
-			break;
-		}
-		ret += nr;
-		if (nr != len)
-			break;
-	}
 out:
 	if (iov != iovstack)
 		kfree(iov);
@@ -1294,7 +1272,7 @@ compat_sys_readv(unsigned long fd, const struct compat_iovec __user *vec, unsign
 		goto out;
 
 	ret = -EINVAL;
-	if (!file->f_op || (!file->f_op->readv && !file->f_op->read))
+	if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read))
 		goto out;
 
 	ret = compat_do_readv_writev(READ, file, vec, vlen, &file->f_pos);
@@ -1317,7 +1295,7 @@ compat_sys_writev(unsigned long fd, const struct compat_iovec __user *vec, unsig
 		goto out;
 
 	ret = -EINVAL;
-	if (!file->f_op || (!file->f_op->writev && !file->f_op->write))
+	if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write))
 		goto out;
 
 	ret = compat_do_readv_writev(WRITE, file, vec, vlen, &file->f_pos);
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index e8bbed9dd268..e893e2be9ed4 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -53,8 +53,6 @@ const struct file_operations ext2_file_operations = {
 	.open		= generic_file_open,
 	.release	= ext2_release_file,
 	.fsync		= ext2_sync_file,
-	.readv		= generic_file_readv,
-	.writev		= generic_file_writev,
 	.sendfile	= generic_file_sendfile,
 	.splice_read	= generic_file_splice_read,
 	.splice_write	= generic_file_splice_write,
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 5c762457bc89..e96c388047e0 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -112,8 +112,6 @@ const struct file_operations ext3_file_operations = {
 	.write		= do_sync_write,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= ext3_file_write,
-	.readv		= generic_file_readv,
-	.writev		= generic_file_writev,
 	.ioctl		= ext3_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= ext3_compat_ioctl,
diff --git a/fs/fat/file.c b/fs/fat/file.c
index d50fc47169c1..f4b8f8b3fbdd 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -127,8 +127,6 @@ const struct file_operations fat_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
-	.readv		= generic_file_readv,
-	.writev		= generic_file_writev,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 4fc557c40cc0..66571eafbb1e 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -680,14 +680,15 @@ static int fuse_read_interrupt(struct fuse_conn *fc, struct fuse_req *req,
  * request_end().  Otherwise add it to the processing list, and set
  * the 'sent' flag.
  */
-static ssize_t fuse_dev_readv(struct file *file, const struct iovec *iov,
-			      unsigned long nr_segs, loff_t *off)
+static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov,
+			      unsigned long nr_segs, loff_t pos)
 {
 	int err;
 	struct fuse_req *req;
 	struct fuse_in *in;
 	struct fuse_copy_state cs;
 	unsigned reqsize;
+	struct file *file = iocb->ki_filp;
 	struct fuse_conn *fc = fuse_get_conn(file);
 	if (!fc)
 		return -EPERM;
@@ -761,15 +762,6 @@ static ssize_t fuse_dev_readv(struct file *file, const struct iovec *iov,
 	return err;
 }
 
-static ssize_t fuse_dev_read(struct file *file, char __user *buf,
-			     size_t nbytes, loff_t *off)
-{
-	struct iovec iov;
-	iov.iov_len = nbytes;
-	iov.iov_base = buf;
-	return fuse_dev_readv(file, &iov, 1, off);
-}
-
 /* Look up request on processing list by unique ID */
 static struct fuse_req *request_find(struct fuse_conn *fc, u64 unique)
 {
@@ -814,15 +806,15 @@ static int copy_out_args(struct fuse_copy_state *cs, struct fuse_out *out,
  * it from the list and copy the rest of the buffer to the request.
  * The request is finished by calling request_end()
  */
-static ssize_t fuse_dev_writev(struct file *file, const struct iovec *iov,
-			       unsigned long nr_segs, loff_t *off)
+static ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov,
+			       unsigned long nr_segs, loff_t pos)
 {
 	int err;
 	unsigned nbytes = iov_length(iov, nr_segs);
 	struct fuse_req *req;
 	struct fuse_out_header oh;
 	struct fuse_copy_state cs;
-	struct fuse_conn *fc = fuse_get_conn(file);
+	struct fuse_conn *fc = fuse_get_conn(iocb->ki_filp);
 	if (!fc)
 		return -EPERM;
 
@@ -898,15 +890,6 @@ static ssize_t fuse_dev_writev(struct file *file, const struct iovec *iov,
 	return err;
 }
 
-static ssize_t fuse_dev_write(struct file *file, const char __user *buf,
-			      size_t nbytes, loff_t *off)
-{
-	struct iovec iov;
-	iov.iov_len = nbytes;
-	iov.iov_base = (char __user *) buf;
-	return fuse_dev_writev(file, &iov, 1, off);
-}
-
 static unsigned fuse_dev_poll(struct file *file, poll_table *wait)
 {
 	unsigned mask = POLLOUT | POLLWRNORM;
@@ -1041,10 +1024,10 @@ static int fuse_dev_fasync(int fd, struct file *file, int on)
 const struct file_operations fuse_dev_operations = {
 	.owner		= THIS_MODULE,
 	.llseek		= no_llseek,
-	.read		= fuse_dev_read,
-	.readv		= fuse_dev_readv,
-	.write		= fuse_dev_write,
-	.writev		= fuse_dev_writev,
+	.read		= do_sync_read,
+	.aio_read	= fuse_dev_read,
+	.write		= do_sync_write,
+	.aio_write	= fuse_dev_write,
 	.poll		= fuse_dev_poll,
 	.release	= fuse_dev_release,
 	.fasync		= fuse_dev_fasync,
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 322e876c35ed..4908c38a5885 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -389,8 +389,6 @@ static const struct file_operations hostfs_file_fops = {
 	.sendfile	= generic_file_sendfile,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
-	.readv		= generic_file_readv,
-	.writev		= generic_file_writev,
 	.write		= generic_file_write,
 	.mmap		= generic_file_mmap,
 	.open		= hostfs_file_open,
diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index 1c9745be5ada..f535f2911c12 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -108,8 +108,6 @@ const struct file_operations jfs_file_operations = {
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
-	.readv		= generic_file_readv,
-	.writev		= generic_file_writev,
  	.sendfile	= generic_file_sendfile,
 	.fsync		= jfs_fsync,
 	.release	= jfs_release,
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 0c46f5c86b71..2f9b5a0953ff 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -2298,11 +2298,9 @@ const struct file_operations ntfs_file_ops = {
 	.llseek		= generic_file_llseek,	 /* Seek inside file. */
 	.read		= generic_file_read,	 /* Read from file. */
 	.aio_read	= generic_file_aio_read, /* Async read from file. */
-	.readv		= generic_file_readv,	 /* Read from file. */
 #ifdef NTFS_RW
 	.write		= ntfs_file_write,	 /* Write to file. */
 	.aio_write	= ntfs_file_aio_write,	 /* Async write to file. */
-	.writev		= ntfs_file_writev,	 /* Write to file. */
 	/*.release	= ,*/			 /* Last file is closed.  See
 						    fs/ext2/file.c::
 						    ext2_release_file() for
diff --git a/fs/pipe.c b/fs/pipe.c
index f3b6f71e9d0b..2e60e1c8815e 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -218,9 +218,10 @@ static struct pipe_buf_operations anon_pipe_buf_ops = {
 };
 
 static ssize_t
-pipe_readv(struct file *filp, const struct iovec *_iov,
-	   unsigned long nr_segs, loff_t *ppos)
+pipe_read(struct kiocb *iocb, const struct iovec *_iov,
+	   unsigned long nr_segs, loff_t pos)
 {
+	struct file *filp = iocb->ki_filp;
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct pipe_inode_info *pipe;
 	int do_wakeup;
@@ -330,17 +331,10 @@ redo:
 }
 
 static ssize_t
-pipe_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
-{
-	struct iovec iov = { .iov_base = buf, .iov_len = count };
-
-	return pipe_readv(filp, &iov, 1, ppos);
-}
-
-static ssize_t
-pipe_writev(struct file *filp, const struct iovec *_iov,
-	    unsigned long nr_segs, loff_t *ppos)
+pipe_write(struct kiocb *iocb, const struct iovec *_iov,
+	    unsigned long nr_segs, loff_t ppos)
 {
+	struct file *filp = iocb->ki_filp;
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct pipe_inode_info *pipe;
 	ssize_t ret;
@@ -509,15 +503,6 @@ out:
 	return ret;
 }
 
-static ssize_t
-pipe_write(struct file *filp, const char __user *buf,
-	   size_t count, loff_t *ppos)
-{
-	struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = count };
-
-	return pipe_writev(filp, &iov, 1, ppos);
-}
-
 static ssize_t
 bad_pipe_r(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
 {
@@ -736,8 +721,8 @@ pipe_rdwr_open(struct inode *inode, struct file *filp)
  */
 const struct file_operations read_fifo_fops = {
 	.llseek		= no_llseek,
-	.read		= pipe_read,
-	.readv		= pipe_readv,
+	.read		= do_sync_read,
+	.aio_read	= pipe_read,
 	.write		= bad_pipe_w,
 	.poll		= pipe_poll,
 	.ioctl		= pipe_ioctl,
@@ -749,8 +734,8 @@ const struct file_operations read_fifo_fops = {
 const struct file_operations write_fifo_fops = {
 	.llseek		= no_llseek,
 	.read		= bad_pipe_r,
-	.write		= pipe_write,
-	.writev		= pipe_writev,
+	.write		= do_sync_write,
+	.aio_write	= pipe_write,
 	.poll		= pipe_poll,
 	.ioctl		= pipe_ioctl,
 	.open		= pipe_write_open,
@@ -760,10 +745,10 @@ const struct file_operations write_fifo_fops = {
 
 const struct file_operations rdwr_fifo_fops = {
 	.llseek		= no_llseek,
-	.read		= pipe_read,
-	.readv		= pipe_readv,
-	.write		= pipe_write,
-	.writev		= pipe_writev,
+	.read		= do_sync_read,
+	.aio_read	= pipe_read,
+	.write		= do_sync_write,
+	.aio_write	= pipe_write,
 	.poll		= pipe_poll,
 	.ioctl		= pipe_ioctl,
 	.open		= pipe_rdwr_open,
@@ -773,8 +758,8 @@ const struct file_operations rdwr_fifo_fops = {
 
 static struct file_operations read_pipe_fops = {
 	.llseek		= no_llseek,
-	.read		= pipe_read,
-	.readv		= pipe_readv,
+	.read		= do_sync_read,
+	.aio_read	= pipe_read,
 	.write		= bad_pipe_w,
 	.poll		= pipe_poll,
 	.ioctl		= pipe_ioctl,
@@ -786,8 +771,8 @@ static struct file_operations read_pipe_fops = {
 static struct file_operations write_pipe_fops = {
 	.llseek		= no_llseek,
 	.read		= bad_pipe_r,
-	.write		= pipe_write,
-	.writev		= pipe_writev,
+	.write		= do_sync_write,
+	.aio_write	= pipe_write,
 	.poll		= pipe_poll,
 	.ioctl		= pipe_ioctl,
 	.open		= pipe_write_open,
@@ -797,10 +782,10 @@ static struct file_operations write_pipe_fops = {
 
 static struct file_operations rdwr_pipe_fops = {
 	.llseek		= no_llseek,
-	.read		= pipe_read,
-	.readv		= pipe_readv,
-	.write		= pipe_write,
-	.writev		= pipe_writev,
+	.read		= do_sync_read,
+	.aio_read	= pipe_read,
+	.write		= do_sync_write,
+	.aio_write	= pipe_write,
 	.poll		= pipe_poll,
 	.ioctl		= pipe_ioctl,
 	.open		= pipe_rdwr_open,
diff --git a/fs/read_write.c b/fs/read_write.c
index 679dd535380f..32d54cca9bd9 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/syscalls.h>
 #include <linux/pagemap.h>
+#include "read_write.h"
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -450,6 +451,62 @@ unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to)
 
 EXPORT_UNUSED_SYMBOL(iov_shorten);  /*  June 2006  */
 
+ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
+		unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn)
+{
+	struct kiocb kiocb;
+	ssize_t ret;
+
+	init_sync_kiocb(&kiocb, filp);
+	kiocb.ki_pos = *ppos;
+	kiocb.ki_left = len;
+	kiocb.ki_nbytes = len;
+
+	for (;;) {
+		ret = fn(&kiocb, iov, nr_segs, kiocb.ki_pos);
+		if (ret != -EIOCBRETRY)
+			break;
+		wait_on_retry_sync_kiocb(&kiocb);
+	}
+
+	if (ret == -EIOCBQUEUED)
+		ret = wait_on_sync_kiocb(&kiocb);
+	*ppos = kiocb.ki_pos;
+	return ret;
+}
+
+/* Do it by hand, with file-ops */
+ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov,
+		unsigned long nr_segs, loff_t *ppos, io_fn_t fn)
+{
+	struct iovec *vector = iov;
+	ssize_t ret = 0;
+
+	while (nr_segs > 0) {
+		void __user *base;
+		size_t len;
+		ssize_t nr;
+
+		base = vector->iov_base;
+		len = vector->iov_len;
+		vector++;
+		nr_segs--;
+
+		nr = fn(filp, base, len, ppos);
+
+		if (nr < 0) {
+			if (!ret)
+				ret = nr;
+			break;
+		}
+		ret += nr;
+		if (nr != len)
+			break;
+	}
+
+	return ret;
+}
+
 /* A write operation does a read from user space and vice versa */
 #define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ)
 
@@ -457,12 +514,9 @@ static ssize_t do_readv_writev(int type, struct file *file,
 			       const struct iovec __user * uvector,
 			       unsigned long nr_segs, loff_t *pos)
 {
-	typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
-	typedef ssize_t (*iov_fn_t)(struct file *, const struct iovec *, unsigned long, loff_t *);
-
 	size_t tot_len;
 	struct iovec iovstack[UIO_FASTIOV];
-	struct iovec *iov=iovstack, *vector;
+	struct iovec *iov = iovstack;
 	ssize_t ret;
 	int seg;
 	io_fn_t fn;
@@ -532,39 +586,18 @@ static ssize_t do_readv_writev(int type, struct file *file,
 	fnv = NULL;
 	if (type == READ) {
 		fn = file->f_op->read;
-		fnv = file->f_op->readv;
+		fnv = file->f_op->aio_read;
 	} else {
 		fn = (io_fn_t)file->f_op->write;
-		fnv = file->f_op->writev;
-	}
-	if (fnv) {
-		ret = fnv(file, iov, nr_segs, pos);
-		goto out;
+		fnv = file->f_op->aio_write;
 	}
 
-	/* Do it by hand, with file-ops */
-	ret = 0;
-	vector = iov;
-	while (nr_segs > 0) {
-		void __user * base;
-		size_t len;
-		ssize_t nr;
-
-		base = vector->iov_base;
-		len = vector->iov_len;
-		vector++;
-		nr_segs--;
-
-		nr = fn(file, base, len, pos);
+	if (fnv)
+		ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
+						pos, fnv);
+	else
+		ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn);
 
-		if (nr < 0) {
-			if (!ret) ret = nr;
-			break;
-		}
-		ret += nr;
-		if (nr != len)
-			break;
-	}
 out:
 	if (iov != iovstack)
 		kfree(iov);
@@ -585,7 +618,7 @@ ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
 {
 	if (!(file->f_mode & FMODE_READ))
 		return -EBADF;
-	if (!file->f_op || (!file->f_op->readv && !file->f_op->read))
+	if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read))
 		return -EINVAL;
 
 	return do_readv_writev(READ, file, vec, vlen, pos);
@@ -598,7 +631,7 @@ ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
 {
 	if (!(file->f_mode & FMODE_WRITE))
 		return -EBADF;
-	if (!file->f_op || (!file->f_op->writev && !file->f_op->write))
+	if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write))
 		return -EINVAL;
 
 	return do_readv_writev(WRITE, file, vec, vlen, pos);
diff --git a/fs/read_write.h b/fs/read_write.h
new file mode 100644
index 000000000000..d07b954c6e0c
--- /dev/null
+++ b/fs/read_write.h
@@ -0,0 +1,14 @@
+/*
+ * This file is only for sharing some helpers from read_write.c with compat.c.
+ * Don't use anywhere else.
+ */
+
+
+typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
+typedef ssize_t (*iov_fn_t)(struct kiocb *, const struct iovec *,
+		unsigned long, loff_t);
+
+ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
+		unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn);
+ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov,
+		unsigned long nr_segs, loff_t *ppos, io_fn_t fn);
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 4737971c6a39..d93d8dd1958d 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -123,96 +123,6 @@ xfs_file_aio_write_invis(
 	return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
 }
 
-STATIC inline ssize_t
-__xfs_file_readv(
-	struct file		*file,
-	const struct iovec 	*iov,
-	int			ioflags,
-	unsigned long		nr_segs,
-	loff_t			*ppos)
-{
-	struct inode	*inode = file->f_mapping->host;
-	bhv_vnode_t	*vp = vn_from_inode(inode);
-	struct kiocb	kiocb;
-	ssize_t		rval;
-
-	init_sync_kiocb(&kiocb, file);
-	kiocb.ki_pos = *ppos;
-
-	if (unlikely(file->f_flags & O_DIRECT))
-		ioflags |= IO_ISDIRECT;
-	rval = bhv_vop_read(vp, &kiocb, iov, nr_segs,
-				&kiocb.ki_pos, ioflags, NULL);
-
-	*ppos = kiocb.ki_pos;
-	return rval;
-}
-
-STATIC ssize_t
-xfs_file_readv(
-	struct file		*file,
-	const struct iovec 	*iov,
-	unsigned long		nr_segs,
-	loff_t			*ppos)
-{
-	return __xfs_file_readv(file, iov, 0, nr_segs, ppos);
-}
-
-STATIC ssize_t
-xfs_file_readv_invis(
-	struct file		*file,
-	const struct iovec 	*iov,
-	unsigned long		nr_segs,
-	loff_t			*ppos)
-{
-	return __xfs_file_readv(file, iov, IO_INVIS, nr_segs, ppos);
-}
-
-STATIC inline ssize_t
-__xfs_file_writev(
-	struct file		*file,
-	const struct iovec 	*iov,
-	int			ioflags,
-	unsigned long		nr_segs,
-	loff_t			*ppos)
-{
-	struct inode	*inode = file->f_mapping->host;
-	bhv_vnode_t	*vp = vn_from_inode(inode);
-	struct kiocb	kiocb;
-	ssize_t		rval;
-
-	init_sync_kiocb(&kiocb, file);
-	kiocb.ki_pos = *ppos;
-	if (unlikely(file->f_flags & O_DIRECT))
-		ioflags |= IO_ISDIRECT;
-
-	rval = bhv_vop_write(vp, &kiocb, iov, nr_segs,
-				 &kiocb.ki_pos, ioflags, NULL);
-
-	*ppos = kiocb.ki_pos;
-	return rval;
-}
-
-STATIC ssize_t
-xfs_file_writev(
-	struct file		*file,
-	const struct iovec 	*iov,
-	unsigned long		nr_segs,
-	loff_t			*ppos)
-{
-	return __xfs_file_writev(file, iov, 0, nr_segs, ppos);
-}
-
-STATIC ssize_t
-xfs_file_writev_invis(
-	struct file		*file,
-	const struct iovec 	*iov,
-	unsigned long		nr_segs,
-	loff_t			*ppos)
-{
-	return __xfs_file_writev(file, iov, IO_INVIS, nr_segs, ppos);
-}
-
 STATIC ssize_t
 xfs_file_sendfile(
 	struct file		*filp,
@@ -540,8 +450,6 @@ const struct file_operations xfs_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
-	.readv		= xfs_file_readv,
-	.writev		= xfs_file_writev,
 	.aio_read	= xfs_file_aio_read,
 	.aio_write	= xfs_file_aio_write,
 	.sendfile	= xfs_file_sendfile,
@@ -565,8 +473,6 @@ const struct file_operations xfs_invis_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
-	.readv		= xfs_file_readv_invis,
-	.writev		= xfs_file_writev_invis,
 	.aio_read	= xfs_file_aio_read_invis,
 	.aio_write	= xfs_file_aio_write_invis,
 	.sendfile	= xfs_file_sendfile_invis,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 257bae16f545..afb6e6f89a01 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1113,8 +1113,6 @@ struct file_operations {
 	int (*aio_fsync) (struct kiocb *, int datasync);
 	int (*fasync) (int, struct file *, int);
 	int (*lock) (struct file *, int, struct file_lock *);
-	ssize_t (*readv) (struct file *, const struct iovec *, unsigned long, loff_t *);
-	ssize_t (*writev) (struct file *, const struct iovec *, unsigned long, loff_t *);
 	ssize_t (*sendfile) (struct file *, loff_t *, size_t, read_actor_t, void *);
 	ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
@@ -1734,10 +1732,6 @@ extern long do_splice_direct(struct file *in, loff_t *ppos, struct file *out,
 
 extern void
 file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping);
-extern ssize_t generic_file_readv(struct file *filp, const struct iovec *iov, 
-	unsigned long nr_segs, loff_t *ppos);
-ssize_t generic_file_writev(struct file *filp, const struct iovec *iov, 
-			unsigned long nr_segs, loff_t *ppos);
 extern loff_t no_llseek(struct file *file, loff_t offset, int origin);
 extern loff_t generic_file_llseek(struct file *file, loff_t offset, int origin);
 extern loff_t remote_llseek(struct file *file, loff_t offset, int origin);
diff --git a/mm/filemap.c b/mm/filemap.c
index f6c1d22b504f..48497094ae83 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2421,42 +2421,6 @@ ssize_t generic_file_write(struct file *file, const char __user *buf,
 }
 EXPORT_SYMBOL(generic_file_write);
 
-ssize_t generic_file_readv(struct file *filp, const struct iovec *iov,
-			unsigned long nr_segs, loff_t *ppos)
-{
-	struct kiocb kiocb;
-	ssize_t ret;
-
-	init_sync_kiocb(&kiocb, filp);
-	ret = __generic_file_aio_read(&kiocb, iov, nr_segs, ppos);
-	if (-EIOCBQUEUED == ret)
-		ret = wait_on_sync_kiocb(&kiocb);
-	return ret;
-}
-EXPORT_SYMBOL(generic_file_readv);
-
-ssize_t generic_file_writev(struct file *file, const struct iovec *iov,
-			unsigned long nr_segs, loff_t *ppos)
-{
-	struct address_space *mapping = file->f_mapping;
-	struct inode *inode = mapping->host;
-	ssize_t ret;
-
-	mutex_lock(&inode->i_mutex);
-	ret = __generic_file_write_nolock(file, iov, nr_segs, ppos);
-	mutex_unlock(&inode->i_mutex);
-
-	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
-		int err;
-
-		err = sync_page_range(inode, mapping, *ppos - ret, ret);
-		if (err < 0)
-			ret = err;
-	}
-	return ret;
-}
-EXPORT_SYMBOL(generic_file_writev);
-
 /*
  * Called under i_mutex for writes to S_ISREG files.   Returns -EIO if something
  * went wrong during pagecache shootdown.
diff --git a/net/socket.c b/net/socket.c
index df92e4252749..01918f7a301a 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -110,10 +110,6 @@ static long compat_sock_ioctl(struct file *file,
 			      unsigned int cmd, unsigned long arg);
 #endif
 static int sock_fasync(int fd, struct file *filp, int on);
-static ssize_t sock_readv(struct file *file, const struct iovec *vector,
-			  unsigned long count, loff_t *ppos);
-static ssize_t sock_writev(struct file *file, const struct iovec *vector,
-			   unsigned long count, loff_t *ppos);
 static ssize_t sock_sendpage(struct file *file, struct page *page,
 			     int offset, size_t size, loff_t *ppos, int more);
 
@@ -136,8 +132,6 @@ static struct file_operations socket_file_ops = {
 	.open =		sock_no_open,	/* special open code to disallow open via /proc */
 	.release =	sock_close,
 	.fasync =	sock_fasync,
-	.readv =	sock_readv,
-	.writev =	sock_writev,
 	.sendpage =	sock_sendpage,
 	.splice_write = generic_splice_sendpage,
 };
@@ -700,23 +694,6 @@ static ssize_t do_sock_read(struct msghdr *msg, struct kiocb *iocb,
 	return __sock_recvmsg(iocb, sock, msg, size, msg->msg_flags);
 }
 
-static ssize_t sock_readv(struct file *file, const struct iovec *iov,
-			  unsigned long nr_segs, loff_t *ppos)
-{
-	struct kiocb iocb;
-	struct sock_iocb siocb;
-	struct msghdr msg;
-	int ret;
-
-	init_sync_kiocb(&iocb, NULL);
-	iocb.private = &siocb;
-
-	ret = do_sock_read(&msg, &iocb, file, iov, nr_segs);
-	if (-EIOCBQUEUED == ret)
-		ret = wait_on_sync_kiocb(&iocb);
-	return ret;
-}
-
 static ssize_t sock_aio_read(struct kiocb *iocb, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos)
 {
@@ -759,23 +736,6 @@ static ssize_t do_sock_write(struct msghdr *msg, struct kiocb *iocb,
 	return __sock_sendmsg(iocb, sock, msg, size);
 }
 
-static ssize_t sock_writev(struct file *file, const struct iovec *iov,
-			   unsigned long nr_segs, loff_t *ppos)
-{
-	struct msghdr msg;
-	struct kiocb iocb;
-	struct sock_iocb siocb;
-	int ret;
-
-	init_sync_kiocb(&iocb, NULL);
-	iocb.private = &siocb;
-
-	ret = do_sock_write(&msg, &iocb, file, iov, nr_segs);
-	if (-EIOCBQUEUED == ret)
-		ret = wait_on_sync_kiocb(&iocb);
-	return ret;
-}
-
 static ssize_t sock_aio_write(struct kiocb *iocb, const struct iovec *iov,
 			  unsigned long nr_segs, loff_t pos)
 {
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index 891d7140553c..37b4b10850ae 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -2852,8 +2852,8 @@ static ssize_t snd_pcm_write(struct file *file, const char __user *buf,
 	return result;
 }
 
-static ssize_t snd_pcm_readv(struct file *file, const struct iovec *_vector,
-			     unsigned long count, loff_t * offset)
+static ssize_t snd_pcm_aio_read(struct kiocb *iocb, const struct iovec *iov,
+			     unsigned long nr_segs, loff_t pos)
 
 {
 	struct snd_pcm_file *pcm_file;
@@ -2864,22 +2864,22 @@ static ssize_t snd_pcm_readv(struct file *file, const struct iovec *_vector,
 	void __user **bufs;
 	snd_pcm_uframes_t frames;
 
-	pcm_file = file->private_data;
+	pcm_file = iocb->ki_filp->private_data;
 	substream = pcm_file->substream;
 	snd_assert(substream != NULL, return -ENXIO);
 	runtime = substream->runtime;
 	if (runtime->status->state == SNDRV_PCM_STATE_OPEN)
 		return -EBADFD;
-	if (count > 1024 || count != runtime->channels)
+	if (nr_segs > 1024 || nr_segs != runtime->channels)
 		return -EINVAL;
-	if (!frame_aligned(runtime, _vector->iov_len))
+	if (!frame_aligned(runtime, iov->iov_len))
 		return -EINVAL;
-	frames = bytes_to_samples(runtime, _vector->iov_len);
-	bufs = kmalloc(sizeof(void *) * count, GFP_KERNEL);
+	frames = bytes_to_samples(runtime, iov->iov_len);
+	bufs = kmalloc(sizeof(void *) * nr_segs, GFP_KERNEL);
 	if (bufs == NULL)
 		return -ENOMEM;
-	for (i = 0; i < count; ++i)
-		bufs[i] = _vector[i].iov_base;
+	for (i = 0; i < nr_segs; ++i)
+		bufs[i] = iov[i].iov_base;
 	result = snd_pcm_lib_readv(substream, bufs, frames);
 	if (result > 0)
 		result = frames_to_bytes(runtime, result);
@@ -2887,8 +2887,8 @@ static ssize_t snd_pcm_readv(struct file *file, const struct iovec *_vector,
 	return result;
 }
 
-static ssize_t snd_pcm_writev(struct file *file, const struct iovec *_vector,
-			      unsigned long count, loff_t * offset)
+static ssize_t snd_pcm_aio_write(struct kiocb *iocb, const struct iovec *iov,
+			      unsigned long nr_segs, loff_t pos)
 {
 	struct snd_pcm_file *pcm_file;
 	struct snd_pcm_substream *substream;
@@ -2898,7 +2898,7 @@ static ssize_t snd_pcm_writev(struct file *file, const struct iovec *_vector,
 	void __user **bufs;
 	snd_pcm_uframes_t frames;
 
-	pcm_file = file->private_data;
+	pcm_file = iocb->ki_filp->private_data;
 	substream = pcm_file->substream;
 	snd_assert(substream != NULL, result = -ENXIO; goto end);
 	runtime = substream->runtime;
@@ -2906,17 +2906,17 @@ static ssize_t snd_pcm_writev(struct file *file, const struct iovec *_vector,
 		result = -EBADFD;
 		goto end;
 	}
-	if (count > 128 || count != runtime->channels ||
-	    !frame_aligned(runtime, _vector->iov_len)) {
+	if (nr_segs > 128 || nr_segs != runtime->channels ||
+	    !frame_aligned(runtime, iov->iov_len)) {
 		result = -EINVAL;
 		goto end;
 	}
-	frames = bytes_to_samples(runtime, _vector->iov_len);
-	bufs = kmalloc(sizeof(void *) * count, GFP_KERNEL);
+	frames = bytes_to_samples(runtime, iov->iov_len);
+	bufs = kmalloc(sizeof(void *) * nr_segs, GFP_KERNEL);
 	if (bufs == NULL)
 		return -ENOMEM;
-	for (i = 0; i < count; ++i)
-		bufs[i] = _vector[i].iov_base;
+	for (i = 0; i < nr_segs; ++i)
+		bufs[i] = iov[i].iov_base;
 	result = snd_pcm_lib_writev(substream, bufs, frames);
 	if (result > 0)
 		result = frames_to_bytes(runtime, result);
@@ -3426,7 +3426,7 @@ struct file_operations snd_pcm_f_ops[2] = {
 	{
 		.owner =		THIS_MODULE,
 		.write =		snd_pcm_write,
-		.writev =		snd_pcm_writev,
+		.aio_write =		snd_pcm_aio_write,
 		.open =			snd_pcm_playback_open,
 		.release =		snd_pcm_release,
 		.poll =			snd_pcm_playback_poll,
@@ -3438,7 +3438,7 @@ struct file_operations snd_pcm_f_ops[2] = {
 	{
 		.owner =		THIS_MODULE,
 		.read =			snd_pcm_read,
-		.readv =		snd_pcm_readv,
+		.aio_read =		snd_pcm_aio_read,
 		.open =			snd_pcm_capture_open,
 		.release =		snd_pcm_release,
 		.poll =			snd_pcm_capture_poll,
-- 
cgit v1.2.3


From 543ade1fc901db4c3dbe9fb27241fb977f1f3eea Mon Sep 17 00:00:00 2001
From: Badari Pulavarty <pbadari@us.ibm.com>
Date: Sat, 30 Sep 2006 23:28:48 -0700
Subject: [PATCH] Streamline generic_file_* interfaces and filemap cleanups

This patch cleans up generic_file_*_read/write() interfaces.  Christoph
Hellwig gave me the idea for this clean ups.

In a nutshell, all filesystems should set .aio_read/.aio_write methods and use
do_sync_read/ do_sync_write() as their .read/.write methods.  This allows us
to cleanup all variants of generic_file_* routines.

Final available interfaces:

generic_file_aio_read() - read handler
generic_file_aio_write() - write handler
generic_file_aio_write_nolock() - no lock write handler

__generic_file_aio_write_nolock() - internal worker routine

Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/raw.c         | 15 ++------
 fs/adfs/file.c             |  6 ++--
 fs/affs/file.c             |  6 ++--
 fs/bfs/file.c              |  6 ++--
 fs/block_dev.c             | 12 ++-----
 fs/ext2/file.c             |  4 +--
 fs/fuse/file.c             |  6 ++--
 fs/hfs/inode.c             |  6 ++--
 fs/hfsplus/inode.c         |  6 ++--
 fs/hostfs/hostfs_kern.c    |  4 +--
 fs/hpfs/file.c             |  6 ++--
 fs/jffs/inode-v23.c        |  6 ++--
 fs/jffs2/file.c            |  6 ++--
 fs/jfs/file.c              |  4 +--
 fs/minix/file.c            |  6 ++--
 fs/ntfs/file.c             |  2 +-
 fs/qnx4/file.c             |  6 ++--
 fs/ramfs/file-mmu.c        |  6 ++--
 fs/ramfs/file-nommu.c      |  6 ++--
 fs/read_write.c            |  3 +-
 fs/smbfs/file.c            | 24 ++++++++-----
 fs/sysv/file.c             |  6 ++--
 fs/udf/file.c              | 16 +++++----
 fs/ufs/file.c              |  6 ++--
 fs/xfs/linux-2.6/xfs_lrw.c |  4 ++-
 include/linux/fs.h         |  5 ---
 mm/filemap.c               | 87 +++-------------------------------------------
 27 files changed, 105 insertions(+), 165 deletions(-)

(limited to 'include')

diff --git a/drivers/char/raw.c b/drivers/char/raw.c
index 490db531e2d8..89b718e326e5 100644
--- a/drivers/char/raw.c
+++ b/drivers/char/raw.c
@@ -238,21 +238,10 @@ out:
 	return err;
 }
 
-static ssize_t raw_file_write(struct file *file, const char __user *buf,
-				   size_t count, loff_t *ppos)
-{
-	struct iovec local_iov = {
-		.iov_base = (char __user *)buf,
-		.iov_len = count
-	};
-
-	return generic_file_write_nolock(file, &local_iov, 1, ppos);
-}
-
 static const struct file_operations raw_fops = {
-	.read	=	generic_file_read,
+	.read	=	do_sync_read,
 	.aio_read = 	generic_file_aio_read,
-	.write	=	raw_file_write,
+	.write	=	do_sync_write,
 	.aio_write = 	generic_file_aio_write_nolock,
 	.open	=	raw_open,
 	.release=	raw_release,
diff --git a/fs/adfs/file.c b/fs/adfs/file.c
index 1014b9f2117b..6101ea679cb1 100644
--- a/fs/adfs/file.c
+++ b/fs/adfs/file.c
@@ -27,10 +27,12 @@
 
 const struct file_operations adfs_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
 	.mmap		= generic_file_mmap,
 	.fsync		= file_fsync,
-	.write		= generic_file_write,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.sendfile	= generic_file_sendfile,
 };
 
diff --git a/fs/affs/file.c b/fs/affs/file.c
index 3de8590e4f6a..05b5e22de759 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -27,8 +27,10 @@ static int affs_file_release(struct inode *inode, struct file *filp);
 
 const struct file_operations affs_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.open		= affs_file_open,
 	.release	= affs_file_release,
diff --git a/fs/bfs/file.c b/fs/bfs/file.c
index 3d5aca28a0a0..a9164a87f8de 100644
--- a/fs/bfs/file.c
+++ b/fs/bfs/file.c
@@ -19,8 +19,10 @@
 
 const struct file_operations bfs_file_operations = {
 	.llseek 	= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.sendfile	= generic_file_sendfile,
 };
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 0f143094ef1d..bc8f27cc4483 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1154,14 +1154,6 @@ static int blkdev_close(struct inode * inode, struct file * filp)
 	return blkdev_put(bdev);
 }
 
-static ssize_t blkdev_file_write(struct file *file, const char __user *buf,
-				   size_t count, loff_t *ppos)
-{
-	struct iovec local_iov = { .iov_base = (void __user *)buf, .iov_len = count };
-
-	return generic_file_write_nolock(file, &local_iov, 1, ppos);
-}
-
 static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
 	return blkdev_ioctl(file->f_mapping->host, file, cmd, arg);
@@ -1181,8 +1173,8 @@ const struct file_operations def_blk_fops = {
 	.open		= blkdev_open,
 	.release	= blkdev_close,
 	.llseek		= block_llseek,
-	.read		= generic_file_read,
-	.write		= blkdev_file_write,
+	.read		= do_sync_read,
+	.write		= do_sync_write,
   	.aio_read	= generic_file_aio_read,
   	.aio_write	= generic_file_aio_write_nolock,
 	.mmap		= generic_file_mmap,
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index e893e2be9ed4..2dba473c524a 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -41,8 +41,8 @@ static int ext2_release_file (struct inode * inode, struct file * filp)
  */
 const struct file_operations ext2_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.write		= do_sync_write,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
 	.ioctl		= ext2_ioctl,
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 5c4fcd1dbf59..183626868eea 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -753,8 +753,10 @@ static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl)
 
 static const struct file_operations fuse_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= fuse_file_mmap,
 	.open		= fuse_open,
 	.flush		= fuse_flush,
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index d05641c35fc9..02f5573e0349 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -601,8 +601,10 @@ int hfs_inode_setattr(struct dentry *dentry, struct iattr * attr)
 
 static const struct file_operations hfs_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.sendfile	= generic_file_sendfile,
 	.fsync		= file_fsync,
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 0eb1a6092668..9e3675249633 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -282,8 +282,10 @@ static struct inode_operations hfsplus_file_inode_operations = {
 
 static const struct file_operations hfsplus_file_operations = {
 	.llseek 	= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.sendfile	= generic_file_sendfile,
 	.fsync		= file_fsync,
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 4908c38a5885..b6bd33ca3731 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -385,11 +385,11 @@ int hostfs_fsync(struct file *file, struct dentry *dentry, int datasync)
 
 static const struct file_operations hostfs_file_fops = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
+	.read		= do_sync_read,
 	.sendfile	= generic_file_sendfile,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
-	.write		= generic_file_write,
+	.write		= do_sync_write,
 	.mmap		= generic_file_mmap,
 	.open		= hostfs_file_open,
 	.release	= NULL,
diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c
index d9eb19b7b8ae..8b94d24855f0 100644
--- a/fs/hpfs/file.c
+++ b/fs/hpfs/file.c
@@ -113,7 +113,7 @@ static ssize_t hpfs_file_write(struct file *file, const char __user *buf,
 {
 	ssize_t retval;
 
-	retval = generic_file_write(file, buf, count, ppos);
+	retval = do_sync_write(file, buf, count, ppos);
 	if (retval > 0)
 		hpfs_i(file->f_dentry->d_inode)->i_dirty = 1;
 	return retval;
@@ -122,8 +122,10 @@ static ssize_t hpfs_file_write(struct file *file, const char __user *buf,
 const struct file_operations hpfs_file_ops =
 {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
 	.write		= hpfs_file_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.release	= hpfs_file_release,
 	.fsync		= hpfs_file_fsync,
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index f5cf9c93e243..068ef0de8de2 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -1632,8 +1632,10 @@ static const struct file_operations jffs_file_operations =
 {
 	.open		= generic_file_open,
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.ioctl		= jffs_ioctl,
 	.mmap		= generic_file_readonly_mmap,
 	.fsync		= jffs_fsync,
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 3ed6e3e120b6..242875f77cb3 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -42,8 +42,10 @@ const struct file_operations jffs2_file_operations =
 {
 	.llseek =	generic_file_llseek,
 	.open =		generic_file_open,
-	.read =		generic_file_read,
-	.write =	generic_file_write,
+ 	.read =		do_sync_read,
+ 	.aio_read =	generic_file_aio_read,
+ 	.write =	do_sync_write,
+ 	.aio_write =	generic_file_aio_write,
 	.ioctl =	jffs2_ioctl,
 	.mmap =		generic_file_readonly_mmap,
 	.fsync =	jffs2_fsync,
diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index f535f2911c12..976e90dc2d1b 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -103,8 +103,8 @@ struct inode_operations jfs_file_inode_operations = {
 const struct file_operations jfs_file_operations = {
 	.open		= jfs_open,
 	.llseek		= generic_file_llseek,
-	.write		= generic_file_write,
-	.read		= generic_file_read,
+	.write		= do_sync_write,
+	.read		= do_sync_read,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
diff --git a/fs/minix/file.c b/fs/minix/file.c
index 420b32882a10..40eac2e60d25 100644
--- a/fs/minix/file.c
+++ b/fs/minix/file.c
@@ -17,8 +17,10 @@ int minix_sync_file(struct file *, struct dentry *, int);
 
 const struct file_operations minix_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.fsync		= minix_sync_file,
 	.sendfile	= generic_file_sendfile,
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 2f9b5a0953ff..ae2fe0016d2c 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -2296,7 +2296,7 @@ static int ntfs_file_fsync(struct file *filp, struct dentry *dentry,
 
 const struct file_operations ntfs_file_ops = {
 	.llseek		= generic_file_llseek,	 /* Seek inside file. */
-	.read		= generic_file_read,	 /* Read from file. */
+	.read		= do_sync_read,		 /* Read from file. */
 	.aio_read	= generic_file_aio_read, /* Async read from file. */
 #ifdef NTFS_RW
 	.write		= ntfs_file_write,	 /* Write to file. */
diff --git a/fs/qnx4/file.c b/fs/qnx4/file.c
index 62af4b1348bd..467e5ac7280e 100644
--- a/fs/qnx4/file.c
+++ b/fs/qnx4/file.c
@@ -22,11 +22,13 @@
 const struct file_operations qnx4_file_operations =
 {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
 	.mmap		= generic_file_mmap,
 	.sendfile	= generic_file_sendfile,
 #ifdef CONFIG_QNX4FS_RW
-	.write		= generic_file_write,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.fsync		= qnx4_sync_file,
 #endif
 };
diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c
index 86f14cacf641..0947fb57dcf3 100644
--- a/fs/ramfs/file-mmu.c
+++ b/fs/ramfs/file-mmu.c
@@ -33,8 +33,10 @@ const struct address_space_operations ramfs_aops = {
 };
 
 const struct file_operations ramfs_file_operations = {
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.fsync		= simple_sync_file,
 	.sendfile	= generic_file_sendfile,
diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c
index 677139b48e00..bfe5dbf1002e 100644
--- a/fs/ramfs/file-nommu.c
+++ b/fs/ramfs/file-nommu.c
@@ -36,8 +36,10 @@ const struct address_space_operations ramfs_aops = {
 const struct file_operations ramfs_file_operations = {
 	.mmap			= ramfs_nommu_mmap,
 	.get_unmapped_area	= ramfs_nommu_get_unmapped_area,
-	.read			= generic_file_read,
-	.write			= generic_file_write,
+	.read			= do_sync_read,
+	.aio_read		= generic_file_aio_read,
+	.write			= do_sync_write,
+	.aio_write		= generic_file_aio_write,
 	.fsync			= simple_sync_file,
 	.sendfile		= generic_file_sendfile,
 	.llseek			= generic_file_llseek,
diff --git a/fs/read_write.c b/fs/read_write.c
index 32d54cca9bd9..4ed839bcb91c 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -22,7 +22,8 @@
 
 const struct file_operations generic_ro_fops = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
 	.mmap		= generic_file_readonly_mmap,
 	.sendfile	= generic_file_sendfile,
 };
diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c
index dae67048baba..50784d13c87b 100644
--- a/fs/smbfs/file.c
+++ b/fs/smbfs/file.c
@@ -214,13 +214,15 @@ smb_updatepage(struct file *file, struct page *page, unsigned long offset,
 }
 
 static ssize_t
-smb_file_read(struct file * file, char __user * buf, size_t count, loff_t *ppos)
+smb_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
+			unsigned long nr_segs, loff_t pos)
 {
+	struct file * file = iocb->ki_filp;
 	struct dentry * dentry = file->f_dentry;
 	ssize_t	status;
 
 	VERBOSE("file %s/%s, count=%lu@%lu\n", DENTRY_PATH(dentry),
-		(unsigned long) count, (unsigned long) *ppos);
+		(unsigned long) iocb->ki_left, (unsigned long) pos);
 
 	status = smb_revalidate_inode(dentry);
 	if (status) {
@@ -233,7 +235,7 @@ smb_file_read(struct file * file, char __user * buf, size_t count, loff_t *ppos)
 		(long)dentry->d_inode->i_size,
 		dentry->d_inode->i_flags, dentry->d_inode->i_atime);
 
-	status = generic_file_read(file, buf, count, ppos);
+	status = generic_file_aio_read(iocb, iov, nr_segs, pos);
 out:
 	return status;
 }
@@ -317,14 +319,16 @@ const struct address_space_operations smb_file_aops = {
  * Write to a file (through the page cache).
  */
 static ssize_t
-smb_file_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
+smb_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
+			       unsigned long nr_segs, loff_t pos)
 {
+	struct file * file = iocb->ki_filp;
 	struct dentry * dentry = file->f_dentry;
 	ssize_t	result;
 
 	VERBOSE("file %s/%s, count=%lu@%lu\n",
 		DENTRY_PATH(dentry),
-		(unsigned long) count, (unsigned long) *ppos);
+		(unsigned long) iocb->ki_left, (unsigned long) pos);
 
 	result = smb_revalidate_inode(dentry);
 	if (result) {
@@ -337,8 +341,8 @@ smb_file_write(struct file *file, const char __user *buf, size_t count, loff_t *
 	if (result)
 		goto out;
 
-	if (count > 0) {
-		result = generic_file_write(file, buf, count, ppos);
+	if (iocb->ki_left > 0) {
+		result = generic_file_aio_write(iocb, iov, nr_segs, pos);
 		VERBOSE("pos=%ld, size=%ld, mtime=%ld, atime=%ld\n",
 			(long) file->f_pos, (long) dentry->d_inode->i_size,
 			dentry->d_inode->i_mtime, dentry->d_inode->i_atime);
@@ -402,8 +406,10 @@ smb_file_permission(struct inode *inode, int mask, struct nameidata *nd)
 const struct file_operations smb_file_operations =
 {
 	.llseek		= remote_llseek,
-	.read		= smb_file_read,
-	.write		= smb_file_write,
+	.read		= do_sync_read,
+	.aio_read	= smb_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= smb_file_aio_write,
 	.ioctl		= smb_ioctl,
 	.mmap		= smb_file_mmap,
 	.open		= smb_file_open,
diff --git a/fs/sysv/file.c b/fs/sysv/file.c
index a59e303135fa..47a4b728f15b 100644
--- a/fs/sysv/file.c
+++ b/fs/sysv/file.c
@@ -21,8 +21,10 @@
  */
 const struct file_operations sysv_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.fsync		= sysv_sync_file,
 	.sendfile	= generic_file_sendfile,
diff --git a/fs/udf/file.c b/fs/udf/file.c
index a59e5f33daf6..7aedd552cba1 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -103,19 +103,21 @@ const struct address_space_operations udf_adinicb_aops = {
 	.commit_write		= udf_adinicb_commit_write,
 };
 
-static ssize_t udf_file_write(struct file * file, const char __user * buf,
-	size_t count, loff_t *ppos)
+static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
+			      unsigned long nr_segs, loff_t ppos)
 {
 	ssize_t retval;
+	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_dentry->d_inode;
 	int err, pos;
+	size_t count = iocb->ki_left;
 
 	if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB)
 	{
 		if (file->f_flags & O_APPEND)
 			pos = inode->i_size;
 		else
-			pos = *ppos;
+			pos = ppos;
 
 		if (inode->i_sb->s_blocksize < (udf_file_entry_alloc_offset(inode) +
 			pos + count))
@@ -136,7 +138,7 @@ static ssize_t udf_file_write(struct file * file, const char __user * buf,
 		}
 	}
 
-	retval = generic_file_write(file, buf, count, ppos);
+	retval = generic_file_aio_write(iocb, iov, nr_segs, ppos);
 
 	if (retval > 0)
 		mark_inode_dirty(inode);
@@ -249,11 +251,13 @@ static int udf_release_file(struct inode * inode, struct file * filp)
 }
 
 const struct file_operations udf_file_operations = {
-	.read			= generic_file_read,
+	.read			= do_sync_read,
+	.aio_read		= generic_file_aio_read,
 	.ioctl			= udf_ioctl,
 	.open			= generic_file_open,
 	.mmap			= generic_file_mmap,
-	.write			= udf_file_write,
+	.write			= do_sync_write,
+	.aio_write		= udf_file_aio_write,
 	.release		= udf_release_file,
 	.fsync			= udf_fsync_file,
 	.sendfile		= generic_file_sendfile,
diff --git a/fs/ufs/file.c b/fs/ufs/file.c
index a9c6e5f04fae..1e096323bad4 100644
--- a/fs/ufs/file.c
+++ b/fs/ufs/file.c
@@ -53,8 +53,10 @@ static int ufs_sync_file(struct file *file, struct dentry *dentry, int datasync)
  
 const struct file_operations ufs_file_operations = {
 	.llseek		= generic_file_llseek,
-	.read		= generic_file_read,
-	.write		= generic_file_write,
+	.read		= do_sync_read,
+	.aio_read	= generic_file_aio_read,
+	.write		= do_sync_write,
+	.aio_write	= generic_file_aio_write,
 	.mmap		= generic_file_mmap,
 	.open           = generic_file_open,
 	.fsync		= ufs_sync_file,
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 55992b40353c..fa842f1c9fa2 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -279,7 +279,9 @@ xfs_read(
 
 	xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore,
 				(void *)iovp, segs, *offset, ioflags);
-	ret = __generic_file_aio_read(iocb, iovp, segs, offset);
+
+	iocb->ki_pos = *offset;
+	ret = generic_file_aio_read(iocb, iovp, segs, *offset);
 	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
 		ret = wait_on_sync_kiocb(iocb);
 	if (ret > 0)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index afb6e6f89a01..011129f8803e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1699,11 +1699,8 @@ extern int generic_file_mmap(struct file *, struct vm_area_struct *);
 extern int generic_file_readonly_mmap(struct file *, struct vm_area_struct *);
 extern int file_read_actor(read_descriptor_t * desc, struct page *page, unsigned long offset, unsigned long size);
 extern int file_send_actor(read_descriptor_t * desc, struct page *page, unsigned long offset, unsigned long size);
-extern ssize_t generic_file_read(struct file *, char __user *, size_t, loff_t *);
 int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk);
-extern ssize_t generic_file_write(struct file *, const char __user *, size_t, loff_t *);
 extern ssize_t generic_file_aio_read(struct kiocb *, const struct iovec *, unsigned long, loff_t);
-extern ssize_t __generic_file_aio_read(struct kiocb *, const struct iovec *, unsigned long, loff_t *);
 extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, loff_t);
 extern ssize_t generic_file_aio_write_nolock(struct kiocb *, const struct iovec *,
 		unsigned long, loff_t);
@@ -1713,8 +1710,6 @@ extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *,
 		unsigned long, loff_t, loff_t *, size_t, ssize_t);
 extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos);
 extern ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos);
-ssize_t generic_file_write_nolock(struct file *file, const struct iovec *iov,
-				unsigned long nr_segs, loff_t *ppos);
 extern ssize_t generic_file_sendfile(struct file *, loff_t *, size_t, read_actor_t, void *);
 extern void do_generic_mapping_read(struct address_space *mapping,
 				    struct file_ra_state *, struct file *,
diff --git a/mm/filemap.c b/mm/filemap.c
index 48497094ae83..ec469235985d 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1149,13 +1149,14 @@ success:
  * that can use the page cache directly.
  */
 ssize_t
-__generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
-		unsigned long nr_segs, loff_t *ppos)
+generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
+		unsigned long nr_segs, loff_t pos)
 {
 	struct file *filp = iocb->ki_filp;
 	ssize_t retval;
 	unsigned long seg;
 	size_t count;
+	loff_t *ppos = &iocb->ki_pos;
 
 	count = 0;
 	for (seg = 0; seg < nr_segs; seg++) {
@@ -1179,7 +1180,7 @@ __generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 
 	/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */
 	if (filp->f_flags & O_DIRECT) {
-		loff_t pos = *ppos, size;
+		loff_t size;
 		struct address_space *mapping;
 		struct inode *inode;
 
@@ -1223,32 +1224,8 @@ __generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 out:
 	return retval;
 }
-EXPORT_SYMBOL(__generic_file_aio_read);
-
-ssize_t
-generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
-		unsigned long nr_segs, loff_t pos)
-{
-	BUG_ON(iocb->ki_pos != pos);
-	return __generic_file_aio_read(iocb, iov, nr_segs, &iocb->ki_pos);
-}
 EXPORT_SYMBOL(generic_file_aio_read);
 
-ssize_t
-generic_file_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
-{
-	struct iovec local_iov = { .iov_base = buf, .iov_len = count };
-	struct kiocb kiocb;
-	ssize_t ret;
-
-	init_sync_kiocb(&kiocb, filp);
-	ret = __generic_file_aio_read(&kiocb, &local_iov, 1, ppos);
-	if (-EIOCBQUEUED == ret)
-		ret = wait_on_sync_kiocb(&kiocb);
-	return ret;
-}
-EXPORT_SYMBOL(generic_file_read);
-
 int file_send_actor(read_descriptor_t * desc, struct page *page, unsigned long offset, unsigned long size)
 {
 	ssize_t written;
@@ -2339,38 +2316,6 @@ ssize_t generic_file_aio_write_nolock(struct kiocb *iocb,
 }
 EXPORT_SYMBOL(generic_file_aio_write_nolock);
 
-static ssize_t
-__generic_file_write_nolock(struct file *file, const struct iovec *iov,
-				unsigned long nr_segs, loff_t *ppos)
-{
-	struct kiocb kiocb;
-	ssize_t ret;
-
-	init_sync_kiocb(&kiocb, file);
-	kiocb.ki_pos = *ppos;
-	ret = __generic_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
-	if (-EIOCBQUEUED == ret)
-		ret = wait_on_sync_kiocb(&kiocb);
-	return ret;
-}
-
-ssize_t
-generic_file_write_nolock(struct file *file, const struct iovec *iov,
-				unsigned long nr_segs, loff_t *ppos)
-{
-	struct kiocb kiocb;
-	ssize_t ret;
-
-	init_sync_kiocb(&kiocb, file);
-	kiocb.ki_pos = *ppos;
-	ret = generic_file_aio_write_nolock(&kiocb, iov, nr_segs, *ppos);
-	if (-EIOCBQUEUED == ret)
-		ret = wait_on_sync_kiocb(&kiocb);
-	*ppos = kiocb.ki_pos;
-	return ret;
-}
-EXPORT_SYMBOL(generic_file_write_nolock);
-
 ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 		unsigned long nr_segs, loff_t pos)
 {
@@ -2397,30 +2342,6 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 }
 EXPORT_SYMBOL(generic_file_aio_write);
 
-ssize_t generic_file_write(struct file *file, const char __user *buf,
-			   size_t count, loff_t *ppos)
-{
-	struct address_space *mapping = file->f_mapping;
-	struct inode *inode = mapping->host;
-	ssize_t	ret;
-	struct iovec local_iov = { .iov_base = (void __user *)buf,
-					.iov_len = count };
-
-	mutex_lock(&inode->i_mutex);
-	ret = __generic_file_write_nolock(file, &local_iov, 1, ppos);
-	mutex_unlock(&inode->i_mutex);
-
-	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
-		ssize_t err;
-
-		err = sync_page_range(inode, mapping, *ppos - ret, ret);
-		if (err < 0)
-			ret = err;
-	}
-	return ret;
-}
-EXPORT_SYMBOL(generic_file_write);
-
 /*
  * Called under i_mutex for writes to S_ISREG files.   Returns -EIO if something
  * went wrong during pagecache shootdown.
-- 
cgit v1.2.3


From eed4e51fb60c3863c134a5e9f6006b29805ead97 Mon Sep 17 00:00:00 2001
From: Badari Pulavarty <pbadari@us.ibm.com>
Date: Sat, 30 Sep 2006 23:28:49 -0700
Subject: [PATCH] Add vector AIO support

This work is initially done by Zach Brown to add support for vectored aio.
These are the core changes for AIO to support
IOCB_CMD_PREADV/IOCB_CMD_PWRITEV.

[akpm@osdl.org: huge build fix]
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Acked-by: Benjamin LaHaise <bcrl@kvack.org>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/aio.c                | 171 +++++++++++++++++++++++++++++++++---------------
 fs/read_write.c         | 129 +++++++++++++++++++++---------------
 include/linux/aio.h     |   4 ++
 include/linux/aio_abi.h |   2 +
 include/linux/fs.h      |   5 ++
 5 files changed, 203 insertions(+), 108 deletions(-)

(limited to 'include')

diff --git a/fs/aio.c b/fs/aio.c
index 27ff56540c73..2e0d1505ee36 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -415,6 +415,7 @@ static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx)
 	req->ki_retry = NULL;
 	req->ki_dtor = NULL;
 	req->private = NULL;
+	req->ki_iovec = NULL;
 	INIT_LIST_HEAD(&req->ki_run_list);
 
 	/* Check if the completion queue has enough free space to
@@ -460,6 +461,8 @@ static inline void really_put_req(struct kioctx *ctx, struct kiocb *req)
 
 	if (req->ki_dtor)
 		req->ki_dtor(req);
+	if (req->ki_iovec != &req->ki_inline_vec)
+		kfree(req->ki_iovec);
 	kmem_cache_free(kiocb_cachep, req);
 	ctx->reqs_active--;
 
@@ -1301,69 +1304,63 @@ asmlinkage long sys_io_destroy(aio_context_t ctx)
 	return -EINVAL;
 }
 
-/*
- * aio_p{read,write} are the default  ki_retry methods for
- * IO_CMD_P{READ,WRITE}.  They maintains kiocb retry state around potentially
- * multiple calls to f_op->aio_read().  They loop around partial progress
- * instead of returning -EIOCBRETRY because they don't have the means to call
- * kick_iocb().
- */
-static ssize_t aio_pread(struct kiocb *iocb)
+static void aio_advance_iovec(struct kiocb *iocb, ssize_t ret)
 {
-	struct file *file = iocb->ki_filp;
-	struct address_space *mapping = file->f_mapping;
-	struct inode *inode = mapping->host;
-	ssize_t ret = 0;
-
-	do {
-		iocb->ki_inline_vec.iov_base = iocb->ki_buf;
-		iocb->ki_inline_vec.iov_len = iocb->ki_left;
-
-		ret = file->f_op->aio_read(iocb, &iocb->ki_inline_vec,
-						1, iocb->ki_pos);
-		/*
-		 * Can't just depend on iocb->ki_left to determine
-		 * whether we are done. This may have been a short read.
-		 */
-		if (ret > 0) {
-			iocb->ki_buf += ret;
-			iocb->ki_left -= ret;
+	struct iovec *iov = &iocb->ki_iovec[iocb->ki_cur_seg];
+
+	BUG_ON(ret <= 0);
+
+	while (iocb->ki_cur_seg < iocb->ki_nr_segs && ret > 0) {
+		ssize_t this = min((ssize_t)iov->iov_len, ret);
+		iov->iov_base += this;
+		iov->iov_len -= this;
+		iocb->ki_left -= this;
+		ret -= this;
+		if (iov->iov_len == 0) {
+			iocb->ki_cur_seg++;
+			iov++;
 		}
+	}
 
-		/*
-		 * For pipes and sockets we return once we have some data; for
-		 * regular files we retry till we complete the entire read or
-		 * find that we can't read any more data (e.g short reads).
-		 */
-	} while (ret > 0 && iocb->ki_left > 0 &&
-		 !S_ISFIFO(inode->i_mode) && !S_ISSOCK(inode->i_mode));
-
-	/* This means we must have transferred all that we could */
-	/* No need to retry anymore */
-	if ((ret == 0) || (iocb->ki_left == 0))
-		ret = iocb->ki_nbytes - iocb->ki_left;
-
-	return ret;
+	/* the caller should not have done more io than what fit in
+	 * the remaining iovecs */
+	BUG_ON(ret > 0 && iocb->ki_left == 0);
 }
 
-/* see aio_pread() */
-static ssize_t aio_pwrite(struct kiocb *iocb)
+static ssize_t aio_rw_vect_retry(struct kiocb *iocb)
 {
 	struct file *file = iocb->ki_filp;
+	struct address_space *mapping = file->f_mapping;
+	struct inode *inode = mapping->host;
+	ssize_t (*rw_op)(struct kiocb *, const struct iovec *,
+			 unsigned long, loff_t);
 	ssize_t ret = 0;
+	unsigned short opcode;
+
+	if ((iocb->ki_opcode == IOCB_CMD_PREADV) ||
+		(iocb->ki_opcode == IOCB_CMD_PREAD)) {
+		rw_op = file->f_op->aio_read;
+		opcode = IOCB_CMD_PREADV;
+	} else {
+		rw_op = file->f_op->aio_write;
+		opcode = IOCB_CMD_PWRITEV;
+	}
 
 	do {
-		iocb->ki_inline_vec.iov_base = iocb->ki_buf;
-		iocb->ki_inline_vec.iov_len = iocb->ki_left;
-
-		ret = file->f_op->aio_write(iocb, &iocb->ki_inline_vec,
-						1, iocb->ki_pos);
-		if (ret > 0) {
-			iocb->ki_buf += ret;
-			iocb->ki_left -= ret;
-		}
-	} while (ret > 0 && iocb->ki_left > 0);
+		ret = rw_op(iocb, &iocb->ki_iovec[iocb->ki_cur_seg],
+			    iocb->ki_nr_segs - iocb->ki_cur_seg,
+			    iocb->ki_pos);
+		if (ret > 0)
+			aio_advance_iovec(iocb, ret);
+
+	/* retry all partial writes.  retry partial reads as long as its a
+	 * regular file. */
+	} while (ret > 0 && iocb->ki_left > 0 &&
+		 (opcode == IOCB_CMD_PWRITEV ||
+		  (!S_ISFIFO(inode->i_mode) && !S_ISSOCK(inode->i_mode))));
 
+	/* This means we must have transferred all that we could */
+	/* No need to retry anymore */
 	if ((ret == 0) || (iocb->ki_left == 0))
 		ret = iocb->ki_nbytes - iocb->ki_left;
 
@@ -1390,6 +1387,38 @@ static ssize_t aio_fsync(struct kiocb *iocb)
 	return ret;
 }
 
+static ssize_t aio_setup_vectored_rw(int type, struct kiocb *kiocb)
+{
+	ssize_t ret;
+
+	ret = rw_copy_check_uvector(type, (struct iovec __user *)kiocb->ki_buf,
+				    kiocb->ki_nbytes, 1,
+				    &kiocb->ki_inline_vec, &kiocb->ki_iovec);
+	if (ret < 0)
+		goto out;
+
+	kiocb->ki_nr_segs = kiocb->ki_nbytes;
+	kiocb->ki_cur_seg = 0;
+	/* ki_nbytes/left now reflect bytes instead of segs */
+	kiocb->ki_nbytes = ret;
+	kiocb->ki_left = ret;
+
+	ret = 0;
+out:
+	return ret;
+}
+
+static ssize_t aio_setup_single_vector(struct kiocb *kiocb)
+{
+	kiocb->ki_iovec = &kiocb->ki_inline_vec;
+	kiocb->ki_iovec->iov_base = kiocb->ki_buf;
+	kiocb->ki_iovec->iov_len = kiocb->ki_left;
+	kiocb->ki_nr_segs = 1;
+	kiocb->ki_cur_seg = 0;
+	kiocb->ki_nbytes = kiocb->ki_left;
+	return 0;
+}
+
 /*
  * aio_setup_iocb:
  *	Performs the initial checks and aio retry method
@@ -1412,9 +1441,12 @@ static ssize_t aio_setup_iocb(struct kiocb *kiocb)
 		ret = security_file_permission(file, MAY_READ);
 		if (unlikely(ret))
 			break;
+		ret = aio_setup_single_vector(kiocb);
+		if (ret)
+			break;
 		ret = -EINVAL;
 		if (file->f_op->aio_read)
-			kiocb->ki_retry = aio_pread;
+			kiocb->ki_retry = aio_rw_vect_retry;
 		break;
 	case IOCB_CMD_PWRITE:
 		ret = -EBADF;
@@ -1427,9 +1459,40 @@ static ssize_t aio_setup_iocb(struct kiocb *kiocb)
 		ret = security_file_permission(file, MAY_WRITE);
 		if (unlikely(ret))
 			break;
+		ret = aio_setup_single_vector(kiocb);
+		if (ret)
+			break;
+		ret = -EINVAL;
+		if (file->f_op->aio_write)
+			kiocb->ki_retry = aio_rw_vect_retry;
+		break;
+	case IOCB_CMD_PREADV:
+		ret = -EBADF;
+		if (unlikely(!(file->f_mode & FMODE_READ)))
+			break;
+		ret = security_file_permission(file, MAY_READ);
+		if (unlikely(ret))
+			break;
+		ret = aio_setup_vectored_rw(READ, kiocb);
+		if (ret)
+			break;
+		ret = -EINVAL;
+		if (file->f_op->aio_read)
+			kiocb->ki_retry = aio_rw_vect_retry;
+		break;
+	case IOCB_CMD_PWRITEV:
+		ret = -EBADF;
+		if (unlikely(!(file->f_mode & FMODE_WRITE)))
+			break;
+		ret = security_file_permission(file, MAY_WRITE);
+		if (unlikely(ret))
+			break;
+		ret = aio_setup_vectored_rw(WRITE, kiocb);
+		if (ret)
+			break;
 		ret = -EINVAL;
 		if (file->f_op->aio_write)
-			kiocb->ki_retry = aio_pwrite;
+			kiocb->ki_retry = aio_rw_vect_retry;
 		break;
 	case IOCB_CMD_FDSYNC:
 		ret = -EINVAL;
diff --git a/fs/read_write.c b/fs/read_write.c
index 4ed839bcb91c..f792000a28e6 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -511,6 +511,74 @@ ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov,
 /* A write operation does a read from user space and vice versa */
 #define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ)
 
+ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
+			      unsigned long nr_segs, unsigned long fast_segs,
+			      struct iovec *fast_pointer,
+			      struct iovec **ret_pointer)
+  {
+	unsigned long seg;
+  	ssize_t ret;
+	struct iovec *iov = fast_pointer;
+
+  	/*
+  	 * SuS says "The readv() function *may* fail if the iovcnt argument
+  	 * was less than or equal to 0, or greater than {IOV_MAX}.  Linux has
+  	 * traditionally returned zero for zero segments, so...
+  	 */
+	if (nr_segs == 0) {
+		ret = 0;
+  		goto out;
+	}
+
+  	/*
+  	 * First get the "struct iovec" from user memory and
+  	 * verify all the pointers
+  	 */
+	if (nr_segs > UIO_MAXIOV) {
+		ret = -EINVAL;
+  		goto out;
+	}
+	if (nr_segs > fast_segs) {
+  		iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL);
+		if (iov == NULL) {
+			ret = -ENOMEM;
+  			goto out;
+		}
+  	}
+	if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector))) {
+		ret = -EFAULT;
+  		goto out;
+	}
+
+  	/*
+	 * According to the Single Unix Specification we should return EINVAL
+	 * if an element length is < 0 when cast to ssize_t or if the
+	 * total length would overflow the ssize_t return value of the
+	 * system call.
+  	 */
+	ret = 0;
+  	for (seg = 0; seg < nr_segs; seg++) {
+  		void __user *buf = iov[seg].iov_base;
+  		ssize_t len = (ssize_t)iov[seg].iov_len;
+
+		/* see if we we're about to use an invalid len or if
+		 * it's about to overflow ssize_t */
+		if (len < 0 || (ret + len < ret)) {
+			ret = -EINVAL;
+  			goto out;
+		}
+		if (unlikely(!access_ok(vrfy_dir(type), buf, len))) {
+			ret = -EFAULT;
+  			goto out;
+		}
+
+		ret += len;
+  	}
+out:
+	*ret_pointer = iov;
+	return ret;
+}
+
 static ssize_t do_readv_writev(int type, struct file *file,
 			       const struct iovec __user * uvector,
 			       unsigned long nr_segs, loff_t *pos)
@@ -519,64 +587,20 @@ static ssize_t do_readv_writev(int type, struct file *file,
 	struct iovec iovstack[UIO_FASTIOV];
 	struct iovec *iov = iovstack;
 	ssize_t ret;
-	int seg;
 	io_fn_t fn;
 	iov_fn_t fnv;
 
-	/*
-	 * SuS says "The readv() function *may* fail if the iovcnt argument
-	 * was less than or equal to 0, or greater than {IOV_MAX}.  Linux has
-	 * traditionally returned zero for zero segments, so...
-	 */
-	ret = 0;
-	if (nr_segs == 0)
-		goto out;
-
-	/*
-	 * First get the "struct iovec" from user memory and
-	 * verify all the pointers
-	 */
-	ret = -EINVAL;
-	if (nr_segs > UIO_MAXIOV)
-		goto out;
-	if (!file->f_op)
-		goto out;
-	if (nr_segs > UIO_FASTIOV) {
-		ret = -ENOMEM;
-		iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL);
-		if (!iov)
-			goto out;
-	}
-	ret = -EFAULT;
-	if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector)))
+	if (!file->f_op) {
+		ret = -EINVAL;
 		goto out;
+	}
 
-	/*
-	 * Single unix specification:
-	 * We should -EINVAL if an element length is not >= 0 and fitting an
-	 * ssize_t.  The total length is fitting an ssize_t
-	 *
-	 * Be careful here because iov_len is a size_t not an ssize_t
-	 */
-	tot_len = 0;
-	ret = -EINVAL;
-	for (seg = 0; seg < nr_segs; seg++) {
-		void __user *buf = iov[seg].iov_base;
-		ssize_t len = (ssize_t)iov[seg].iov_len;
-
-		if (len < 0)	/* size_t not fitting an ssize_t .. */
-			goto out;
-		if (unlikely(!access_ok(vrfy_dir(type), buf, len)))
-			goto Efault;
-		tot_len += len;
-		if ((ssize_t)tot_len < 0) /* maths overflow on the ssize_t */
-			goto out;
-	}
-	if (tot_len == 0) {
-		ret = 0;
+	ret = rw_copy_check_uvector(type, uvector, nr_segs,
+			ARRAY_SIZE(iovstack), iovstack, &iov);
+	if (ret <= 0)
 		goto out;
-	}
 
+	tot_len = ret;
 	ret = rw_verify_area(type, file, pos, tot_len);
 	if (ret < 0)
 		goto out;
@@ -609,9 +633,6 @@ out:
 			fsnotify_modify(file->f_dentry);
 	}
 	return ret;
-Efault:
-	ret = -EFAULT;
-	goto out;
 }
 
 ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
diff --git a/include/linux/aio.h b/include/linux/aio.h
index 58349e58b749..5722568fc71e 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -7,6 +7,7 @@
 #include <linux/uio.h>
 
 #include <asm/atomic.h>
+#include <linux/uio.h>
 
 #define AIO_MAXSEGS		4
 #define AIO_KIOGRP_NR_ATOMIC	8
@@ -114,6 +115,9 @@ struct kiocb {
 	long			ki_kicked; 	/* just for testing */
 	long			ki_queued; 	/* just for testing */
 	struct iovec		ki_inline_vec;	/* inline vector */
+ 	struct iovec		*ki_iovec;
+ 	unsigned long		ki_nr_segs;
+ 	unsigned long		ki_cur_seg;
 
 	struct list_head	ki_list;	/* the aio core uses this
 						 * for cancellation */
diff --git a/include/linux/aio_abi.h b/include/linux/aio_abi.h
index 30fdcc89d142..3466b1d0ffd2 100644
--- a/include/linux/aio_abi.h
+++ b/include/linux/aio_abi.h
@@ -41,6 +41,8 @@ enum {
 	 * IOCB_CMD_POLL = 5,
 	 */
 	IOCB_CMD_NOOP = 6,
+	IOCB_CMD_PREADV = 7,
+	IOCB_CMD_PWRITEV = 8,
 };
 
 /* read() from /dev/aio returns these structures. */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 011129f8803e..4bb70871873f 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1150,6 +1150,11 @@ struct inode_operations {
 
 struct seq_file;
 
+ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
+				unsigned long nr_segs, unsigned long fast_segs,
+				struct iovec *fast_pointer,
+				struct iovec **ret_pointer);
+
 extern ssize_t vfs_read(struct file *, char __user *, size_t, loff_t *);
 extern ssize_t vfs_write(struct file *, const char __user *, size_t, loff_t *);
 extern ssize_t vfs_readv(struct file *, const struct iovec __user *,
-- 
cgit v1.2.3


From 31608214fe21dc31d8046679054ab033b1fe5cf1 Mon Sep 17 00:00:00 2001
From: "Chen, Kenneth W" <kenneth.w.chen@intel.com>
Date: Sat, 30 Sep 2006 23:28:50 -0700
Subject: [PATCH] clean up unused kiocb variables

Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Acked-by: Zach Brown <zach.brown@oracle.com>
Cc: Suparna Bhattacharya <suparna@in.ibm.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/aio.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/aio.h b/include/linux/aio.h
index 5722568fc71e..0d71c0041f13 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -112,8 +112,6 @@ struct kiocb {
 	char 			__user *ki_buf;	/* remaining iocb->aio_buf */
 	size_t			ki_left; 	/* remaining bytes */
 	long			ki_retried; 	/* just for testing */
-	long			ki_kicked; 	/* just for testing */
-	long			ki_queued; 	/* just for testing */
 	struct iovec		ki_inline_vec;	/* inline vector */
  	struct iovec		*ki_iovec;
  	unsigned long		ki_nr_segs;
-- 
cgit v1.2.3


From 17db952cd16cecc76937b138c685ae3d198ab17c Mon Sep 17 00:00:00 2001
From: Balbir Singh <balbir@in.ibm.com>
Date: Sat, 30 Sep 2006 23:28:51 -0700
Subject: [PATCH] Add genetlink utilities for payload length calculation

Add two utility helper functions genlmsg_msg_size() and genlmsg_total_size().
These functions are derived from their netlink counterparts.

Signed-off-by: Balbir Singh <balbir@in.ibm.com>
Cc: Jamal Hadi <hadi@cyberus.ca>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jay Lan <jlan@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/net/genetlink.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

(limited to 'include')

diff --git a/include/net/genetlink.h b/include/net/genetlink.h
index 4a38d85e4e25..b619314218a6 100644
--- a/include/net/genetlink.h
+++ b/include/net/genetlink.h
@@ -169,4 +169,22 @@ static inline int genlmsg_len(const struct genlmsghdr *gnlh)
 	return (nlh->nlmsg_len - GENL_HDRLEN - NLMSG_HDRLEN);
 }
 
+/**
+ * genlmsg_msg_size - length of genetlink message not including padding
+ * @payload: length of message payload
+ */
+static inline int genlmsg_msg_size(int payload)
+{
+	return GENL_HDRLEN + payload;
+}
+
+/**
+ * genlmsg_total_size - length of genetlink message including padding
+ * @payload: length of message payload
+ */
+static inline int genlmsg_total_size(int payload)
+{
+	return NLMSG_ALIGN(genlmsg_msg_size(payload));
+}
+
 #endif	/* __NET_GENERIC_NETLINK_H */
-- 
cgit v1.2.3


From f3cef7a99469afc159fec3a61b42dc7ca5b6824f Mon Sep 17 00:00:00 2001
From: Jay Lan <jlan@engr.sgi.com>
Date: Sat, 30 Sep 2006 23:28:55 -0700
Subject: [PATCH] csa: basic accounting over taskstats

Add some basic accounting fields to the taskstats struct, add a new
kernel/tsacct.c to handle basic accounting data handling upon exit.  A handle
is added to taskstats.c to invoke the basic accounting data handling.

Signed-off-by: Jay Lan <jlan@sgi.com>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Chris Sturtivant <csturtiv@sgi.com>
Cc: Tony Ernst <tee@sgi.com>
Cc: Guillaume Thouvenin <guillaume.thouvenin@bull.net>
Cc: "Michal Piotrowski" <michal.k.k.piotrowski@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/taskstats.h   | 29 ++++++++++++++----
 include/linux/tsacct_kern.h | 19 ++++++++++++
 kernel/Makefile             |  2 +-
 kernel/taskstats.c          |  4 +++
 kernel/tsacct.c             | 72 +++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 120 insertions(+), 6 deletions(-)
 create mode 100644 include/linux/tsacct_kern.h
 create mode 100644 kernel/tsacct.c

(limited to 'include')

diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h
index f1cb6cddd19d..af93a63a5092 100644
--- a/include/linux/taskstats.h
+++ b/include/linux/taskstats.h
@@ -2,6 +2,7 @@
  *
  * Copyright (C) Shailabh Nagar, IBM Corp. 2006
  *           (C) Balbir Singh,   IBM Corp. 2006
+ *           (C) Jay Lan,        SGI, 2006
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2.1 of the GNU Lesser General Public License
@@ -29,16 +30,18 @@
  *	c) add new fields after version comment; maintain 64-bit alignment
  */
 
-#define TASKSTATS_VERSION	1
+
+#define TASKSTATS_VERSION	2
+#define TS_COMM_LEN		16	/* should sync up with TASK_COMM_LEN
+					 * in linux/sched.h */
 
 struct taskstats {
 
 	/* Version 1 */
 	__u16	version;
-	__u16	padding[3];	/* Userspace should not interpret the padding
-				 * field which can be replaced by useful
-				 * fields if struct taskstats is extended.
-				 */
+	__u32	ac_exitcode;	/* Exit status */
+	__u8	ac_flag;		/* Record flags */
+	__u8	ac_nice;		/* task_nice */
 
 	/* Delay accounting fields start
 	 *
@@ -88,6 +91,22 @@ struct taskstats {
 	__u64	cpu_run_virtual_total;
 	/* Delay accounting fields end */
 	/* version 1 ends here */
+
+	/* Basic Accounting Fields start */
+	char	ac_comm[TS_COMM_LEN];	/* Command name */
+	__u8	ac_sched;		/* Scheduling discipline */
+	__u8	ac_pad[3];
+	__u32	ac_uid;			/* User ID */
+	__u32	ac_gid;			/* Group ID */
+	__u32	ac_pid;			/* Process ID */
+	__u32	ac_ppid;		/* Parent process ID */
+	__u32	ac_btime;		/* Begin time [sec since 1970] */
+	__u64	ac_etime;		/* Elapsed time [usec] */
+	__u64	ac_utime;		/* User CPU time [usec] */
+	__u64	ac_stime;		/* SYstem CPU time [usec] */
+	__u64	ac_minflt;		/* Minor Page Fault */
+	__u64	ac_majflt;		/* Major Page Fault */
+	/* Basic Accounting Fields end */
 };
 
 
diff --git a/include/linux/tsacct_kern.h b/include/linux/tsacct_kern.h
new file mode 100644
index 000000000000..7e8196a02118
--- /dev/null
+++ b/include/linux/tsacct_kern.h
@@ -0,0 +1,19 @@
+/*
+ * tsacct_kern.h - kernel header for system accounting over taskstats interface
+ *
+ * Copyright (C) Jay Lan	SGI
+ */
+
+#ifndef _LINUX_TSACCT_KERN_H
+#define _LINUX_TSACCT_KERN_H
+
+#include <linux/taskstats.h>
+
+#ifdef CONFIG_TASKSTATS
+extern void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk);
+#else
+static inline void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
+{}
+#endif /* CONFIG_TASKSTATS */
+
+#endif
diff --git a/kernel/Makefile b/kernel/Makefile
index e210e8cf7237..aacaafb28b9d 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -49,7 +49,7 @@ obj-$(CONFIG_SECCOMP) += seccomp.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
-obj-$(CONFIG_TASKSTATS) += taskstats.o
+obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
 
 ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
 # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index c451af2ddb50..6c38dce88e8c 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -18,6 +18,7 @@
 
 #include <linux/kernel.h>
 #include <linux/taskstats_kern.h>
+#include <linux/tsacct_kern.h>
 #include <linux/delayacct.h>
 #include <linux/cpumask.h>
 #include <linux/percpu.h>
@@ -198,7 +199,10 @@ static int fill_pid(pid_t pid, struct task_struct *pidtsk,
 	 */
 
 	delayacct_add_tsk(stats, tsk);
+
+	/* fill in basic acct fields */
 	stats->version = TASKSTATS_VERSION;
+	bacct_add_tsk(stats, tsk);
 
 	/* Define err: label here if needed */
 	put_task_struct(tsk);
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
new file mode 100644
index 000000000000..899067950a88
--- /dev/null
+++ b/kernel/tsacct.c
@@ -0,0 +1,72 @@
+/*
+ * tsacct.c - System accounting over taskstats interface
+ *
+ * Copyright (C) Jay Lan,	<jlan@sgi.com>
+ *
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/tsacct_kern.h>
+#include <linux/acct.h>
+
+
+#define USEC_PER_TICK	(USEC_PER_SEC/HZ)
+/*
+ * fill in basic accounting fields
+ */
+void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
+{
+	struct timespec uptime, ts;
+	s64 ac_etime;
+
+	BUILD_BUG_ON(TS_COMM_LEN < TASK_COMM_LEN);
+
+	/* calculate task elapsed time in timespec */
+	do_posix_clock_monotonic_gettime(&uptime);
+	ts = timespec_sub(uptime, current->group_leader->start_time);
+	/* rebase elapsed time to usec */
+	ac_etime = timespec_to_ns(&ts);
+	do_div(ac_etime, NSEC_PER_USEC);
+	stats->ac_etime = ac_etime;
+	stats->ac_btime = xtime.tv_sec - ts.tv_sec;
+	if (thread_group_leader(tsk)) {
+		stats->ac_exitcode = tsk->exit_code;
+		if (tsk->flags & PF_FORKNOEXEC)
+			stats->ac_flag |= AFORK;
+	}
+	if (tsk->flags & PF_SUPERPRIV)
+		stats->ac_flag |= ASU;
+	if (tsk->flags & PF_DUMPCORE)
+		stats->ac_flag |= ACORE;
+	if (tsk->flags & PF_SIGNALED)
+		stats->ac_flag |= AXSIG;
+	stats->ac_nice	 = task_nice(tsk);
+	stats->ac_sched	 = tsk->policy;
+	stats->ac_uid	 = tsk->uid;
+	stats->ac_gid	 = tsk->gid;
+	stats->ac_pid	 = tsk->pid;
+	stats->ac_ppid	 = (tsk->parent) ? tsk->parent->pid : 0;
+	stats->ac_utime	 = cputime_to_msecs(tsk->utime) * USEC_PER_MSEC;
+	stats->ac_stime	 = cputime_to_msecs(tsk->stime) * USEC_PER_MSEC;
+	stats->ac_minflt = tsk->min_flt;
+	stats->ac_majflt = tsk->maj_flt;
+	/* Each process gets a minimum of one usec cpu time */
+	if ((stats->ac_utime == 0) && (stats->ac_stime == 0)) {
+		stats->ac_stime = 1;
+	}
+
+	strncpy(stats->ac_comm, tsk->comm, sizeof(stats->ac_comm));
+}
+
-- 
cgit v1.2.3


From 9acc1853519a0473620d424105f9d49ea5b4e62e Mon Sep 17 00:00:00 2001
From: Jay Lan <jlan@engr.sgi.com>
Date: Sat, 30 Sep 2006 23:28:58 -0700
Subject: [PATCH] csa: Extended system accounting over taskstats

Add extended system accounting handling over taskstats interface.  A
CONFIG_TASK_XACCT flag is created to enable the extended accounting code.

Signed-off-by: Jay Lan <jlan@sgi.com>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Chris Sturtivant <csturtiv@sgi.com>
Cc: Tony Ernst <tee@sgi.com>
Cc: Guillaume Thouvenin <guillaume.thouvenin@bull.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/taskstats.h   | 11 +++++++++++
 include/linux/tsacct_kern.h |  9 +++++++++
 init/Kconfig                |  9 +++++++++
 kernel/taskstats.c          |  4 ++++
 kernel/tsacct.c             | 19 +++++++++++++++++++
 5 files changed, 52 insertions(+)

(limited to 'include')

diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h
index af93a63a5092..3d2c304886b0 100644
--- a/include/linux/taskstats.h
+++ b/include/linux/taskstats.h
@@ -107,6 +107,17 @@ struct taskstats {
 	__u64	ac_minflt;		/* Minor Page Fault */
 	__u64	ac_majflt;		/* Major Page Fault */
 	/* Basic Accounting Fields end */
+
+ 	/* Extended accounting fields start */
+ 	__u64	acct_rss_mem1;		/* accumulated rss usage */
+ 	__u64	acct_vm_mem1;		/* accumulated virtual memory usage */
+ 	__u64	hiwater_rss;		/* High-watermark of RSS usage */
+ 	__u64	hiwater_vm;		/* High-water virtual memory usage */
+ 	__u64	read_char;		/* bytes read */
+ 	__u64	write_char;		/* bytes written */
+ 	__u64	read_syscalls;		/* read syscalls */
+ 	__u64	write_syscalls;		/* write syscalls */
+ 	/* Extended accounting fields end */
 };
 
 
diff --git a/include/linux/tsacct_kern.h b/include/linux/tsacct_kern.h
index 7e8196a02118..74102dcae67a 100644
--- a/include/linux/tsacct_kern.h
+++ b/include/linux/tsacct_kern.h
@@ -16,4 +16,13 @@ static inline void bacct_add_tsk(struct taskstats *stats, struct task_struct *ts
 {}
 #endif /* CONFIG_TASKSTATS */
 
+#ifdef CONFIG_TASK_XACCT
+extern void xacct_add_tsk(struct taskstats *stats, struct task_struct *p);
+#else
+static inline void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
+{}
+#endif /* CONFIG_TASK_XACCT */
+
 #endif
+
+
diff --git a/init/Kconfig b/init/Kconfig
index d2d72704f875..f7a04d0daf07 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -257,6 +257,15 @@ config CC_OPTIMIZE_FOR_SIZE
 
 	  If unsure, say N.
 
+config TASK_XACCT
+	bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
+	depends on TASKSTATS
+	help
+	  Collect extended task accounting data and send the data
+	  to userland for processing over the taskstats interface.
+
+	  Say N if unsure.
+
 config SYSCTL
 	bool
 
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index 6c38dce88e8c..5d6a8c54ee85 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -20,6 +20,7 @@
 #include <linux/taskstats_kern.h>
 #include <linux/tsacct_kern.h>
 #include <linux/delayacct.h>
+#include <linux/tsacct_kern.h>
 #include <linux/cpumask.h>
 #include <linux/percpu.h>
 #include <net/genetlink.h>
@@ -204,6 +205,9 @@ static int fill_pid(pid_t pid, struct task_struct *pidtsk,
 	stats->version = TASKSTATS_VERSION;
 	bacct_add_tsk(stats, tsk);
 
+	/* fill in extended acct fields */
+	xacct_add_tsk(stats, tsk);
+
 	/* Define err: label here if needed */
 	put_task_struct(tsk);
 	return rc;
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 899067950a88..410483490cf6 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -70,3 +70,22 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
 	strncpy(stats->ac_comm, tsk->comm, sizeof(stats->ac_comm));
 }
 
+
+#ifdef CONFIG_TASK_XACCT
+/*
+ * fill in extended accounting fields
+ */
+void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
+{
+	stats->acct_rss_mem1 = p->acct_rss_mem1;
+	stats->acct_vm_mem1  = p->acct_vm_mem1;
+	if (p->mm) {
+		stats->hiwater_rss   = p->mm->hiwater_rss;
+		stats->hiwater_vm    = p->mm->hiwater_vm;
+	}
+	stats->read_char	= p->rchar;
+	stats->write_char	= p->wchar;
+	stats->read_syscalls	= p->syscr;
+	stats->write_syscalls	= p->syscw;
+}
+#endif
-- 
cgit v1.2.3


From 8f0ab5147951267134612570604cf8341901a80c Mon Sep 17 00:00:00 2001
From: Jay Lan <jlan@engr.sgi.com>
Date: Sat, 30 Sep 2006 23:28:59 -0700
Subject: [PATCH] csa: convert CONFIG tag for extended accounting routines

There were a few accounting data/macros that are used in CSA but are #ifdef'ed
inside CONFIG_BSD_PROCESS_ACCT.  This patch is to change those ifdef's from
CONFIG_BSD_PROCESS_ACCT to CONFIG_TASK_XACCT.  A few defines are moved from
kernel/acct.c and include/linux/acct.h to kernel/tsacct.c and
include/linux/tsacct_kern.h.

Signed-off-by: Jay Lan <jlan@sgi.com>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Chris Sturtivant <csturtiv@sgi.com>
Cc: Tony Ernst <tee@sgi.com>
Cc: Guillaume Thouvenin <guillaume.thouvenin@bull.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/compat.c                 |  2 +-
 fs/exec.c                   |  2 +-
 include/linux/acct.h        |  4 ----
 include/linux/sched.h       |  2 +-
 include/linux/tsacct_kern.h |  6 ++++++
 kernel/acct.c               | 30 ------------------------------
 kernel/exit.c               |  1 +
 kernel/fork.c               |  1 +
 kernel/sched.c              |  2 +-
 kernel/tsacct.c             | 30 ++++++++++++++++++++++++++++++
 10 files changed, 42 insertions(+), 38 deletions(-)

(limited to 'include')

diff --git a/fs/compat.c b/fs/compat.c
index 6b90bf35f61d..13fb08d096c4 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -44,7 +44,7 @@
 #include <linux/nfsd/syscall.h>
 #include <linux/personality.h>
 #include <linux/rwsem.h>
-#include <linux/acct.h>
+#include <linux/tsacct_kern.h>
 #include <linux/mm.h>
 
 #include <net/sock.h>		/* siocdevprivate_ioctl */
diff --git a/fs/exec.c b/fs/exec.c
index a8efe35176b0..0db3fc3c5f0f 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -46,7 +46,7 @@
 #include <linux/security.h>
 #include <linux/syscalls.h>
 #include <linux/rmap.h>
-#include <linux/acct.h>
+#include <linux/tsacct_kern.h>
 #include <linux/cn_proc.h>
 #include <linux/audit.h>
 
diff --git a/include/linux/acct.h b/include/linux/acct.h
index e86bae7324d2..0496d1f09952 100644
--- a/include/linux/acct.h
+++ b/include/linux/acct.h
@@ -124,16 +124,12 @@ extern void acct_auto_close(struct super_block *sb);
 extern void acct_init_pacct(struct pacct_struct *pacct);
 extern void acct_collect(long exitcode, int group_dead);
 extern void acct_process(void);
-extern void acct_update_integrals(struct task_struct *tsk);
-extern void acct_clear_integrals(struct task_struct *tsk);
 #else
 #define acct_auto_close_mnt(x)	do { } while (0)
 #define acct_auto_close(x)	do { } while (0)
 #define acct_init_pacct(x)	do { } while (0)
 #define acct_collect(x,y)	do { } while (0)
 #define acct_process()		do { } while (0)
-#define acct_update_integrals(x)		do { } while (0)
-#define acct_clear_integrals(task)	do { } while (0)
 #endif
 
 /*
diff --git a/include/linux/sched.h b/include/linux/sched.h
index fc4a9873ec10..4ddeb0f982fb 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -981,7 +981,7 @@ struct task_struct {
 	wait_queue_t *io_wait;
 /* i/o counters(bytes read/written, #syscalls */
 	u64 rchar, wchar, syscr, syscw;
-#if defined(CONFIG_BSD_PROCESS_ACCT)
+#if defined(CONFIG_TASK_XACCT)
 	u64 acct_rss_mem1;	/* accumulated rss usage */
 	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
 	clock_t acct_stimexpd;	/* clock_t-converted stime since last update */
diff --git a/include/linux/tsacct_kern.h b/include/linux/tsacct_kern.h
index 74102dcae67a..7e50ac795b0b 100644
--- a/include/linux/tsacct_kern.h
+++ b/include/linux/tsacct_kern.h
@@ -18,9 +18,15 @@ static inline void bacct_add_tsk(struct taskstats *stats, struct task_struct *ts
 
 #ifdef CONFIG_TASK_XACCT
 extern void xacct_add_tsk(struct taskstats *stats, struct task_struct *p);
+extern void acct_update_integrals(struct task_struct *tsk);
+extern void acct_clear_integrals(struct task_struct *tsk);
 #else
 static inline void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 {}
+static inline void acct_update_integrals(struct task_struct *tsk)
+{}
+static inline void acct_clear_integrals(struct task_struct *tsk)
+{}
 #endif /* CONFIG_TASK_XACCT */
 
 #endif
diff --git a/kernel/acct.c b/kernel/acct.c
index f4330acead46..0aad5ca36a81 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -602,33 +602,3 @@ void acct_process(void)
 	do_acct_process(file);
 	fput(file);
 }
-
-
-/**
- * acct_update_integrals - update mm integral fields in task_struct
- * @tsk: task_struct for accounting
- */
-void acct_update_integrals(struct task_struct *tsk)
-{
-	if (likely(tsk->mm)) {
-		long delta =
-			cputime_to_jiffies(tsk->stime) - tsk->acct_stimexpd;
-
-		if (delta == 0)
-			return;
-		tsk->acct_stimexpd = tsk->stime;
-		tsk->acct_rss_mem1 += delta * get_mm_rss(tsk->mm);
-		tsk->acct_vm_mem1 += delta * tsk->mm->total_vm;
-	}
-}
-
-/**
- * acct_clear_integrals - clear the mm integral fields in task_struct
- * @tsk: task_struct whose accounting fields are cleared
- */
-void acct_clear_integrals(struct task_struct *tsk)
-{
-	tsk->acct_stimexpd = 0;
-	tsk->acct_rss_mem1 = 0;
-	tsk->acct_vm_mem1 = 0;
-}
diff --git a/kernel/exit.c b/kernel/exit.c
index c189de2927ab..3b47f26985f2 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -18,6 +18,7 @@
 #include <linux/security.h>
 #include <linux/cpu.h>
 #include <linux/acct.h>
+#include <linux/tsacct_kern.h>
 #include <linux/file.h>
 #include <linux/binfmts.h>
 #include <linux/ptrace.h>
diff --git a/kernel/fork.c b/kernel/fork.c
index 1c999f3e0b47..89f666491d1f 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -42,6 +42,7 @@
 #include <linux/profile.h>
 #include <linux/rmap.h>
 #include <linux/acct.h>
+#include <linux/tsacct_kern.h>
 #include <linux/cn_proc.h>
 #include <linux/delayacct.h>
 #include <linux/taskstats_kern.h>
diff --git a/kernel/sched.c b/kernel/sched.c
index 74f169ac0773..2bbd948f0169 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -49,7 +49,7 @@
 #include <linux/seq_file.h>
 #include <linux/syscalls.h>
 #include <linux/times.h>
-#include <linux/acct.h>
+#include <linux/tsacct_kern.h>
 #include <linux/kprobes.h>
 #include <linux/delayacct.h>
 #include <asm/tlb.h>
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 410483490cf6..47c71daa416f 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -88,4 +88,34 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 	stats->read_syscalls	= p->syscr;
 	stats->write_syscalls	= p->syscw;
 }
+
+
+/**
+ * acct_update_integrals - update mm integral fields in task_struct
+ * @tsk: task_struct for accounting
+ */
+void acct_update_integrals(struct task_struct *tsk)
+{
+	if (likely(tsk->mm)) {
+		long delta =
+			cputime_to_jiffies(tsk->stime) - tsk->acct_stimexpd;
+
+		if (delta == 0)
+			return;
+		tsk->acct_stimexpd = tsk->stime;
+		tsk->acct_rss_mem1 += delta * get_mm_rss(tsk->mm);
+		tsk->acct_vm_mem1 += delta * tsk->mm->total_vm;
+	}
+}
+
+/**
+ * acct_clear_integrals - clear the mm integral fields in task_struct
+ * @tsk: task_struct whose accounting fields are cleared
+ */
+void acct_clear_integrals(struct task_struct *tsk)
+{
+	tsk->acct_stimexpd = 0;
+	tsk->acct_rss_mem1 = 0;
+	tsk->acct_vm_mem1 = 0;
+}
 #endif
-- 
cgit v1.2.3


From db5fed26b2e0beed939b773dd5896077a1794d65 Mon Sep 17 00:00:00 2001
From: Jay Lan <jlan@sgi.com>
Date: Sat, 30 Sep 2006 23:29:00 -0700
Subject: [PATCH] csa accounting taskstats update

ChangeLog:
   Feedbacks from Andrew Morton:
   - define TS_COMM_LEN to 32
   - change acct_stimexpd field of task_struct to be of
     cputime_t, which is to be used to save the tsk->stime
     of last timer interrupt update.
   - a new Documentation/accounting/taskstats-struct.txt
     to describe fields of taskstats struct.

   Feedback from Balbir Singh:
   - keep the stime of a task to be zero when both stime
     and utime are zero as recoreded in task_struct.

   Misc:
   - convert accumulated RSS/VM from platform dependent
     pages-ticks to MBytes-usecs in the kernel

Cc: Shailabh Nagar <nagar@watson.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Chris Sturtivant <csturtiv@sgi.com>
Cc: Tony Ernst <tee@sgi.com>
Cc: Guillaume Thouvenin <guillaume.thouvenin@bull.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/accounting/taskstats-struct.txt | 161 ++++++++++++++++++++++++++
 include/linux/sched.h                         |   2 +-
 include/linux/taskstats.h                     |  40 +++++--
 kernel/tsacct.c                               |  25 ++--
 4 files changed, 207 insertions(+), 21 deletions(-)
 create mode 100644 Documentation/accounting/taskstats-struct.txt

(limited to 'include')

diff --git a/Documentation/accounting/taskstats-struct.txt b/Documentation/accounting/taskstats-struct.txt
new file mode 100644
index 000000000000..661c797eaf79
--- /dev/null
+++ b/Documentation/accounting/taskstats-struct.txt
@@ -0,0 +1,161 @@
+The struct taskstats
+--------------------
+
+This document contains an explanation of the struct taskstats fields.
+
+There are three different groups of fields in the struct taskstats:
+
+1) Common and basic accounting fields
+    If CONFIG_TASKSTATS is set, the taskstats inteface is enabled and
+    the common fields and basic accounting fields are collected for
+    delivery at do_exit() of a task.
+2) Delay accounting fields
+    These fields are placed between
+    /* Delay accounting fields start */
+    and
+    /* Delay accounting fields end */
+    Their values are collected if CONFIG_TASK_DELAY_ACCT is set.
+3) Extended accounting fields
+    These fields are placed between
+    /* Extended accounting fields start */
+    and
+    /* Extended accounting fields end */
+    Their values are collected if CONFIG_TASK_XACCT is set.
+
+Future extension should add fields to the end of the taskstats struct, and
+should not change the relative position of each field within the struct.
+
+
+struct taskstats {
+
+1) Common and basic accounting fields:
+	/* The version number of this struct. This field is always set to
+	 * TAKSTATS_VERSION, which is defined in <linux/taskstats.h>.
+	 * Each time the struct is changed, the value should be incremented.
+	 */
+	__u16	version;
+
+  	/* The exit code of a task. */
+	__u32	ac_exitcode;		/* Exit status */
+
+  	/* The accounting flags of a task as defined in <linux/acct.h>
+	 * Defined values are AFORK, ASU, ACOMPAT, ACORE, and AXSIG.
+	 */
+	__u8	ac_flag;		/* Record flags */
+
+  	/* The value of task_nice() of a task. */
+	__u8	ac_nice;		/* task_nice */
+
+  	/* The name of the command that started this task. */
+	char	ac_comm[TS_COMM_LEN];	/* Command name */
+
+  	/* The scheduling discipline as set in task->policy field. */
+	__u8	ac_sched;		/* Scheduling discipline */
+
+	__u8	ac_pad[3];
+	__u32	ac_uid;			/* User ID */
+	__u32	ac_gid;			/* Group ID */
+	__u32	ac_pid;			/* Process ID */
+	__u32	ac_ppid;		/* Parent process ID */
+
+  	/* The time when a task begins, in [secs] since 1970. */
+	__u32	ac_btime;		/* Begin time [sec since 1970] */
+
+  	/* The elapsed time of a task, in [usec]. */
+	__u64	ac_etime;		/* Elapsed time [usec] */
+
+  	/* The user CPU time of a task, in [usec]. */
+	__u64	ac_utime;		/* User CPU time [usec] */
+
+  	/* The system CPU time of a task, in [usec]. */
+	__u64	ac_stime;		/* System CPU time [usec] */
+
+  	/* The minor page fault count of a task, as set in task->min_flt. */
+	__u64	ac_minflt;		/* Minor Page Fault Count */
+
+	/* The major page fault count of a task, as set in task->maj_flt. */
+	__u64	ac_majflt;		/* Major Page Fault Count */
+
+
+2) Delay accounting fields:
+	/* Delay accounting fields start
+	 *
+	 * All values, until the comment "Delay accounting fields end" are
+	 * available only if delay accounting is enabled, even though the last
+	 * few fields are not delays
+	 *
+	 * xxx_count is the number of delay values recorded
+	 * xxx_delay_total is the corresponding cumulative delay in nanoseconds
+	 *
+	 * xxx_delay_total wraps around to zero on overflow
+	 * xxx_count incremented regardless of overflow
+	 */
+
+	/* Delay waiting for cpu, while runnable
+	 * count, delay_total NOT updated atomically
+	 */
+	__u64	cpu_count;
+	__u64	cpu_delay_total;
+
+	/* Following four fields atomically updated using task->delays->lock */
+
+	/* Delay waiting for synchronous block I/O to complete
+	 * does not account for delays in I/O submission
+	 */
+	__u64	blkio_count;
+	__u64	blkio_delay_total;
+
+	/* Delay waiting for page fault I/O (swap in only) */
+	__u64	swapin_count;
+	__u64	swapin_delay_total;
+
+	/* cpu "wall-clock" running time
+	 * On some architectures, value will adjust for cpu time stolen
+	 * from the kernel in involuntary waits due to virtualization.
+	 * Value is cumulative, in nanoseconds, without a corresponding count
+	 * and wraps around to zero silently on overflow
+	 */
+	__u64	cpu_run_real_total;
+
+	/* cpu "virtual" running time
+	 * Uses time intervals seen by the kernel i.e. no adjustment
+	 * for kernel's involuntary waits due to virtualization.
+	 * Value is cumulative, in nanoseconds, without a corresponding count
+	 * and wraps around to zero silently on overflow
+	 */
+	__u64	cpu_run_virtual_total;
+	/* Delay accounting fields end */
+	/* version 1 ends here */
+
+
+3) Extended accounting fields
+	/* Extended accounting fields start */
+
+	/* Accumulated RSS usage in duration of a task, in MBytes-usecs.
+	 * The current rss usage is added to this counter every time
+	 * a tick is charged to a task's system time. So, at the end we
+	 * will have memory usage multiplied by system time. Thus an
+	 * average usage per system time unit can be calculated.
+	 */
+	__u64	coremem;		/* accumulated RSS usage in MB-usec */
+
+  	/* Accumulated virtual memory usage in duration of a task.
+	 * Same as acct_rss_mem1 above except that we keep track of VM usage.
+	 */
+	__u64	virtmem;		/* accumulated VM usage in MB-usec */
+
+  	/* High watermark of RSS usage in duration of a task, in KBytes. */
+	__u64	hiwater_rss;		/* High-watermark of RSS usage */
+
+  	/* High watermark of VM  usage in duration of a task, in KBytes. */
+	__u64	hiwater_vm;		/* High-water virtual memory usage */
+
+	/* The following four fields are I/O statistics of a task. */
+	__u64	read_char;		/* bytes read */
+	__u64	write_char;		/* bytes written */
+	__u64	read_syscalls;		/* read syscalls */
+	__u64	write_syscalls;		/* write syscalls */
+
+	/* Extended accounting fields end */
+
+}
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4ddeb0f982fb..7ef899c47c29 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -984,7 +984,7 @@ struct task_struct {
 #if defined(CONFIG_TASK_XACCT)
 	u64 acct_rss_mem1;	/* accumulated rss usage */
 	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
-	clock_t acct_stimexpd;	/* clock_t-converted stime since last update */
+	cputime_t acct_stimexpd;/* stime since last update */
 #endif
 #ifdef CONFIG_NUMA
   	struct mempolicy *mempolicy;
diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h
index 3d2c304886b0..45248806ae9c 100644
--- a/include/linux/taskstats.h
+++ b/include/linux/taskstats.h
@@ -32,14 +32,21 @@
 
 
 #define TASKSTATS_VERSION	2
-#define TS_COMM_LEN		16	/* should sync up with TASK_COMM_LEN
+#define TS_COMM_LEN		32	/* should be >= TASK_COMM_LEN
 					 * in linux/sched.h */
 
 struct taskstats {
 
-	/* Version 1 */
+	/* The version number of this struct. This field is always set to
+	 * TAKSTATS_VERSION, which is defined in <linux/taskstats.h>.
+	 * Each time the struct is changed, the value should be incremented.
+	 */
 	__u16	version;
-	__u32	ac_exitcode;	/* Exit status */
+	__u32	ac_exitcode;		/* Exit status */
+
+	/* The accounting flags of a task as defined in <linux/acct.h>
+	 * Defined values are AFORK, ASU, ACOMPAT, ACORE, and AXSIG.
+	 */
 	__u8	ac_flag;		/* Record flags */
 	__u8	ac_nice;		/* task_nice */
 
@@ -104,15 +111,30 @@ struct taskstats {
 	__u64	ac_etime;		/* Elapsed time [usec] */
 	__u64	ac_utime;		/* User CPU time [usec] */
 	__u64	ac_stime;		/* SYstem CPU time [usec] */
-	__u64	ac_minflt;		/* Minor Page Fault */
-	__u64	ac_majflt;		/* Major Page Fault */
+	__u64	ac_minflt;		/* Minor Page Fault Count */
+	__u64	ac_majflt;		/* Major Page Fault Count */
 	/* Basic Accounting Fields end */
 
  	/* Extended accounting fields start */
- 	__u64	acct_rss_mem1;		/* accumulated rss usage */
- 	__u64	acct_vm_mem1;		/* accumulated virtual memory usage */
- 	__u64	hiwater_rss;		/* High-watermark of RSS usage */
- 	__u64	hiwater_vm;		/* High-water virtual memory usage */
+	/* Accumulated RSS usage in duration of a task, in MBytes-usecs.
+	 * The current rss usage is added to this counter every time
+	 * a tick is charged to a task's system time. So, at the end we
+	 * will have memory usage multiplied by system time. Thus an
+	 * average usage per system time unit can be calculated.
+	 */
+ 	__u64	coremem;		/* accumulated RSS usage in MB-usec */
+	/* Accumulated virtual memory usage in duration of a task.
+	 * Same as acct_rss_mem1 above except that we keep track of VM usage.
+	 */
+ 	__u64	virtmem;		/* accumulated VM  usage in MB-usec */
+
+	/* High watermark of RSS and virtual memory usage in duration of
+	 * a task, in KBytes.
+	 */
+ 	__u64	hiwater_rss;		/* High-watermark of RSS usage, in KB */
+ 	__u64	hiwater_vm;		/* High-water VM usage, in KB */
+
+	/* The following four fields are I/O statistics of a task. */
  	__u64	read_char;		/* bytes read */
  	__u64	write_char;		/* bytes written */
  	__u64	read_syscalls;		/* read syscalls */
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 47c71daa416f..db443221ba5b 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -20,6 +20,7 @@
 #include <linux/sched.h>
 #include <linux/tsacct_kern.h>
 #include <linux/acct.h>
+#include <linux/jiffies.h>
 
 
 #define USEC_PER_TICK	(USEC_PER_SEC/HZ)
@@ -62,33 +63,35 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
 	stats->ac_stime	 = cputime_to_msecs(tsk->stime) * USEC_PER_MSEC;
 	stats->ac_minflt = tsk->min_flt;
 	stats->ac_majflt = tsk->maj_flt;
-	/* Each process gets a minimum of one usec cpu time */
-	if ((stats->ac_utime == 0) && (stats->ac_stime == 0)) {
-		stats->ac_stime = 1;
-	}
 
 	strncpy(stats->ac_comm, tsk->comm, sizeof(stats->ac_comm));
 }
 
 
 #ifdef CONFIG_TASK_XACCT
+
+#define KB 1024
+#define MB (1024*KB)
 /*
  * fill in extended accounting fields
  */
 void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 {
-	stats->acct_rss_mem1 = p->acct_rss_mem1;
-	stats->acct_vm_mem1  = p->acct_vm_mem1;
+	/* convert pages-jiffies to Mbyte-usec */
+	stats->coremem = jiffies_to_usecs(p->acct_rss_mem1) * PAGE_SIZE / MB;
+	stats->virtmem = jiffies_to_usecs(p->acct_vm_mem1) * PAGE_SIZE / MB;
 	if (p->mm) {
-		stats->hiwater_rss   = p->mm->hiwater_rss;
-		stats->hiwater_vm    = p->mm->hiwater_vm;
+		/* adjust to KB unit */
+		stats->hiwater_rss   = p->mm->hiwater_rss * PAGE_SIZE / KB;
+		stats->hiwater_vm    = p->mm->hiwater_vm * PAGE_SIZE / KB;
 	}
 	stats->read_char	= p->rchar;
 	stats->write_char	= p->wchar;
 	stats->read_syscalls	= p->syscr;
 	stats->write_syscalls	= p->syscw;
 }
-
+#undef KB
+#undef MB
 
 /**
  * acct_update_integrals - update mm integral fields in task_struct
@@ -97,8 +100,8 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 void acct_update_integrals(struct task_struct *tsk)
 {
 	if (likely(tsk->mm)) {
-		long delta =
-			cputime_to_jiffies(tsk->stime) - tsk->acct_stimexpd;
+		long delta = cputime_to_jiffies(
+			cputime_sub(tsk->stime, tsk->acct_stimexpd));
 
 		if (delta == 0)
 			return;
-- 
cgit v1.2.3


From 9a53c3a783c2fa9b969628e65695c11c3e51e673 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Sat, 30 Sep 2006 23:29:03 -0700
Subject: [PATCH] r/o bind mounts: unlink: monitor i_nlink

When a filesystem decrements i_nlink to zero, it means that a write must be
performed in order to drop the inode from the filesystem.

We're shortly going to have keep filesystems from being remounted r/o between
the time that this i_nlink decrement and that write occurs.

So, add a little helper function to do the decrements.  We'll tie into it in a
bit to note when i_nlink hits zero.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/usb/core/inode.c |  7 ++++---
 fs/autofs/root.c         |  2 +-
 fs/autofs4/root.c        |  2 +-
 fs/bfs/dir.c             |  9 +++------
 fs/cifs/inode.c          | 10 +++++-----
 fs/coda/dir.c            |  4 ++--
 fs/ext2/namei.c          |  2 +-
 fs/ext3/namei.c          | 14 +++++++-------
 fs/hfs/dir.c             |  2 +-
 fs/hfsplus/dir.c         |  2 +-
 fs/hpfs/namei.c          |  6 +++---
 fs/jffs/inode-v23.c      |  3 +--
 fs/jffs2/dir.c           |  6 +++---
 fs/jfs/namei.c           | 14 ++++++--------
 fs/libfs.c               | 10 +++++-----
 fs/minix/namei.c         |  2 +-
 fs/msdos/namei.c         |  9 ++++-----
 fs/nfs/dir.c             |  6 +++---
 fs/ocfs2/namei.c         |  4 ++--
 fs/qnx4/namei.c          |  6 ++----
 fs/reiserfs/namei.c      |  6 +++---
 fs/sysv/namei.c          |  2 +-
 fs/udf/namei.c           | 18 ++++++------------
 fs/ufs/namei.c           |  2 +-
 fs/vfat/namei.c          |  9 ++++-----
 include/linux/fs.h       |  7 ++++++-
 ipc/mqueue.c             |  2 +-
 mm/shmem.c               | 10 +++++-----
 28 files changed, 83 insertions(+), 93 deletions(-)

(limited to 'include')

diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index df3d152f0493..88002e45a6b4 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -332,7 +332,7 @@ static int usbfs_unlink (struct inode *dir, struct dentry *dentry)
 {
 	struct inode *inode = dentry->d_inode;
 	mutex_lock(&inode->i_mutex);
-	dentry->d_inode->i_nlink--;
+	drop_nlink(dentry->d_inode);
 	dput(dentry);
 	mutex_unlock(&inode->i_mutex);
 	d_delete(dentry);
@@ -347,10 +347,11 @@ static int usbfs_rmdir(struct inode *dir, struct dentry *dentry)
 	mutex_lock(&inode->i_mutex);
 	dentry_unhash(dentry);
 	if (usbfs_empty(dentry)) {
-		dentry->d_inode->i_nlink -= 2;
+		drop_nlink(dentry->d_inode);
+		drop_nlink(dentry->d_inode);
 		dput(dentry);
 		inode->i_flags |= S_DEAD;
-		dir->i_nlink--;
+		drop_nlink(dir);
 		error = 0;
 	}
 	mutex_unlock(&inode->i_mutex);
diff --git a/fs/autofs/root.c b/fs/autofs/root.c
index 9cac08d6a873..54ad70731927 100644
--- a/fs/autofs/root.c
+++ b/fs/autofs/root.c
@@ -414,7 +414,7 @@ static int autofs_root_rmdir(struct inode *dir, struct dentry *dentry)
 
 	dentry->d_time = (unsigned long)(struct autofs_dir_ent *)NULL;
 	autofs_hash_delete(ent);
-	dir->i_nlink--;
+	drop_nlink(dir);
 	d_drop(dentry);
 	unlock_kernel();
 
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 563ef9d7da9f..348bec0982b0 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -676,7 +676,7 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
 	dentry->d_inode->i_nlink = 0;
 
 	if (dir->i_nlink)
-		dir->i_nlink--;
+		drop_nlink(dir);
 
 	return 0;
 }
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index dcf04cb13283..ce05d1643dd1 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -117,8 +117,7 @@ static int bfs_create(struct inode * dir, struct dentry * dentry, int mode,
 
 	err = bfs_add_entry(dir, dentry->d_name.name, dentry->d_name.len, inode->i_ino);
 	if (err) {
-		inode->i_nlink--;
-		mark_inode_dirty(inode);
+		inode_dec_link_count(inode);
 		iput(inode);
 		unlock_kernel();
 		return err;
@@ -196,9 +195,8 @@ static int bfs_unlink(struct inode * dir, struct dentry * dentry)
 	mark_buffer_dirty(bh);
 	dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	mark_inode_dirty(dir);
-	inode->i_nlink--;
 	inode->i_ctime = dir->i_ctime;
-	mark_inode_dirty(inode);
+	inode_dec_link_count(inode);
 	error = 0;
 
 out_brelse:
@@ -249,9 +247,8 @@ static int bfs_rename(struct inode * old_dir, struct dentry * old_dentry,
 	old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC;
 	mark_inode_dirty(old_dir);
 	if (new_inode) {
-		new_inode->i_nlink--;
 		new_inode->i_ctime = CURRENT_TIME_SEC;
-		mark_inode_dirty(new_inode);
+		inode_dec_link_count(new_inode);
 	}
 	mark_buffer_dirty(old_bh);
 	error = 0;
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 05f874c7441b..74441a17e186 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -590,7 +590,7 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry)
 
 	if (!rc) {
 		if (direntry->d_inode)
-			direntry->d_inode->i_nlink--;
+			drop_nlink(direntry->d_inode);
 	} else if (rc == -ENOENT) {
 		d_drop(direntry);
 	} else if (rc == -ETXTBSY) {
@@ -609,7 +609,7 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry)
 						CIFS_MOUNT_MAP_SPECIAL_CHR);
 			CIFSSMBClose(xid, pTcon, netfid);
 			if (direntry->d_inode)
-				direntry->d_inode->i_nlink--;
+				drop_nlink(direntry->d_inode);
 		}
 	} else if (rc == -EACCES) {
 		/* try only if r/o attribute set in local lookup data? */
@@ -663,7 +663,7 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry)
 						CIFS_MOUNT_MAP_SPECIAL_CHR);
 			if (!rc) {
 				if (direntry->d_inode)
-					direntry->d_inode->i_nlink--;
+					drop_nlink(direntry->d_inode);
 			} else if (rc == -ETXTBSY) {
 				int oplock = FALSE;
 				__u16 netfid;
@@ -684,7 +684,7 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry)
 						    CIFS_MOUNT_MAP_SPECIAL_CHR);
 					CIFSSMBClose(xid, pTcon, netfid);
 					if (direntry->d_inode)
-			                        direntry->d_inode->i_nlink--;
+						drop_nlink(direntry->d_inode);
 				}
 			/* BB if rc = -ETXTBUSY goto the rename logic BB */
 			}
@@ -816,7 +816,7 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
 			  cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
 
 	if (!rc) {
-		inode->i_nlink--;
+		drop_nlink(inode);
 		i_size_write(direntry->d_inode,0);
 		direntry->d_inode->i_nlink = 0;
 	}
diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 8651ea6a23b7..0a2fd8bb7579 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -367,7 +367,7 @@ int coda_unlink(struct inode *dir, struct dentry *de)
         }
 
 	coda_dir_changed(dir, 0);
-	de->d_inode->i_nlink--;
+	drop_nlink(de->d_inode);
 	unlock_kernel();
 
         return 0;
@@ -394,7 +394,7 @@ int coda_rmdir(struct inode *dir, struct dentry *de)
         }
 
 	coda_dir_changed(dir, -1);
-	de->d_inode->i_nlink--;
+	drop_nlink(de->d_inode);
 	d_delete(de);
 	unlock_kernel();
 
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index 4ca824985321..e1af5b4cf80c 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -326,7 +326,7 @@ static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry,
 		ext2_set_link(new_dir, new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 235e77b52ea5..14c55adfae83 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -1621,7 +1621,7 @@ static inline void ext3_inc_count(handle_t *handle, struct inode *inode)
 
 static inline void ext3_dec_count(handle_t *handle, struct inode *inode)
 {
-	inode->i_nlink--;
+	drop_nlink(inode);
 }
 
 static int ext3_add_nondir(handle_t *handle,
@@ -1743,7 +1743,7 @@ retry:
 	inode->i_size = EXT3_I(inode)->i_disksize = inode->i_sb->s_blocksize;
 	dir_block = ext3_bread (handle, inode, 0, 1, &err);
 	if (!dir_block) {
-		inode->i_nlink--; /* is this nlink == 0? */
+		drop_nlink(inode); /* is this nlink == 0? */
 		ext3_mark_inode_dirty(handle, inode);
 		iput (inode);
 		goto out_stop;
@@ -2053,7 +2053,7 @@ static int ext3_rmdir (struct inode * dir, struct dentry *dentry)
 	ext3_orphan_add(handle, inode);
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	ext3_mark_inode_dirty(handle, inode);
-	dir->i_nlink--;
+	drop_nlink(dir);
 	ext3_update_dx_flag(dir);
 	ext3_mark_inode_dirty(handle, dir);
 
@@ -2104,7 +2104,7 @@ static int ext3_unlink(struct inode * dir, struct dentry *dentry)
 	dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	ext3_update_dx_flag(dir);
 	ext3_mark_inode_dirty(handle, dir);
-	inode->i_nlink--;
+	drop_nlink(inode);
 	if (!inode->i_nlink)
 		ext3_orphan_add(handle, inode);
 	inode->i_ctime = dir->i_ctime;
@@ -2326,7 +2326,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry,
 	}
 
 	if (new_inode) {
-		new_inode->i_nlink--;
+		drop_nlink(new_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 	}
 	old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC;
@@ -2337,9 +2337,9 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry,
 		PARENT_INO(dir_bh->b_data) = cpu_to_le32(new_dir->i_ino);
 		BUFFER_TRACE(dir_bh, "call ext3_journal_dirty_metadata");
 		ext3_journal_dirty_metadata(handle, dir_bh);
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 		if (new_inode) {
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		} else {
 			new_dir->i_nlink++;
 			ext3_update_dx_flag(new_dir);
diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c
index 7cd8cc03aea7..cfef6ad529a7 100644
--- a/fs/hfs/dir.c
+++ b/fs/hfs/dir.c
@@ -246,7 +246,7 @@ static int hfs_unlink(struct inode *dir, struct dentry *dentry)
 	if (res)
 		return res;
 
-	inode->i_nlink--;
+	drop_nlink(inode);
 	hfs_delete_inode(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	mark_inode_dirty(inode);
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 1f9ece0de326..9ceb0dfaa1cc 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -338,7 +338,7 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry)
 		return res;
 
 	if (inode->i_nlink > 0)
-		inode->i_nlink--;
+		drop_nlink(inode);
 	hfsplus_delete_inode(inode);
 	if (inode->i_ino != cnid && !inode->i_nlink) {
 		if (!atomic_read(&HFSPLUS_I(inode).opencnt)) {
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 59e7dc182a0c..4078b0becc5e 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -434,7 +434,7 @@ again:
 		unlock_kernel();
 		return -ENOSPC;
 	default:
-		inode->i_nlink--;
+		drop_nlink(inode);
 		err = 0;
 	}
 	goto out;
@@ -494,7 +494,7 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 		err = -ENOSPC;
 		break;
 	default:
-		dir->i_nlink--;
+		drop_nlink(dir);
 		inode->i_nlink = 0;
 		err = 0;
 	}
@@ -636,7 +636,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	hpfs_i(i)->i_parent_dir = new_dir->i_ino;
 	if (S_ISDIR(i->i_mode)) {
 		new_dir->i_nlink++;
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 	}
 	if ((fnode = hpfs_map_fnode(i->i_sb, i->i_ino, &bh))) {
 		fnode->up = new_dir->i_ino;
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index 068ef0de8de2..3f7899ea4cba 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -1052,9 +1052,8 @@ jffs_remove(struct inode *dir, struct dentry *dentry, int type)
 
 	dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	mark_inode_dirty(dir);
-	inode->i_nlink--;
 	inode->i_ctime = dir->i_ctime;
-	mark_inode_dirty(inode);
+	inode_dec_link_count(inode);
 
 	d_delete(dentry);	/* This also frees the inode */
 
diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index edd8371fc6a5..a5e9f2205b33 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -615,7 +615,7 @@ static int jffs2_rmdir (struct inode *dir_i, struct dentry *dentry)
 	}
 	ret = jffs2_unlink(dir_i, dentry);
 	if (!ret)
-		dir_i->i_nlink--;
+		drop_nlink(dir_i);
 	return ret;
 }
 
@@ -823,7 +823,7 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry,
 
 	if (victim_f) {
 		/* There was a victim. Kill it off nicely */
-		new_dentry->d_inode->i_nlink--;
+		drop_nlink(new_dentry->d_inode);
 		/* Don't oops if the victim was a dirent pointing to an
 		   inode which didn't exist. */
 		if (victim_f->inocache) {
@@ -862,7 +862,7 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry,
 	}
 
 	if (S_ISDIR(old_dentry->d_inode->i_mode))
-		old_dir_i->i_nlink--;
+		drop_nlink(old_dir_i);
 
 	new_dir_i->i_mtime = new_dir_i->i_ctime = old_dir_i->i_mtime = old_dir_i->i_ctime = ITIME(now);
 
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 295268ad231b..088b85976ac0 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -393,9 +393,8 @@ static int jfs_rmdir(struct inode *dip, struct dentry *dentry)
 	/* update parent directory's link count corresponding
 	 * to ".." entry of the target directory deleted
 	 */
-	dip->i_nlink--;
 	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
-	mark_inode_dirty(dip);
+	inode_dec_link_count(dip);
 
 	/*
 	 * OS/2 could have created EA and/or ACL
@@ -515,8 +514,7 @@ static int jfs_unlink(struct inode *dip, struct dentry *dentry)
 	mark_inode_dirty(dip);
 
 	/* update target's inode */
-	ip->i_nlink--;
-	mark_inode_dirty(ip);
+	inode_dec_link_count(ip);
 
 	/*
 	 *      commit zero link count object
@@ -835,7 +833,7 @@ static int jfs_link(struct dentry *old_dentry,
 	rc = txCommit(tid, 2, &iplist[0], 0);
 
 	if (rc) {
-		ip->i_nlink--;
+		ip->i_nlink--; /* never instantiated */
 		iput(ip);
 	} else
 		d_instantiate(dentry, ip);
@@ -1155,9 +1153,9 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			      old_ip->i_ino, JFS_RENAME);
 		if (rc)
 			goto out4;
-		new_ip->i_nlink--;
+		drop_nlink(new_ip);
 		if (S_ISDIR(new_ip->i_mode)) {
-			new_ip->i_nlink--;
+			drop_nlink(new_ip);
 			if (new_ip->i_nlink) {
 				mutex_unlock(&JFS_IP(new_ip)->commit_mutex);
 				if (old_dir != new_dir)
@@ -1223,7 +1221,7 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		goto out4;
 	}
 	if (S_ISDIR(old_ip->i_mode)) {
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 		if (old_dir != new_dir) {
 			/*
 			 * Change inode number of parent for moved directory
diff --git a/fs/libfs.c b/fs/libfs.c
index 3793aaa14577..9204feba75ac 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -275,7 +275,7 @@ int simple_unlink(struct inode *dir, struct dentry *dentry)
 	struct inode *inode = dentry->d_inode;
 
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
-	inode->i_nlink--;
+	drop_nlink(inode);
 	dput(dentry);
 	return 0;
 }
@@ -285,9 +285,9 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 	if (!simple_empty(dentry))
 		return -ENOTEMPTY;
 
-	dentry->d_inode->i_nlink--;
+	drop_nlink(dentry->d_inode);
 	simple_unlink(dir, dentry);
-	dir->i_nlink--;
+	drop_nlink(dir);
 	return 0;
 }
 
@@ -303,9 +303,9 @@ int simple_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (new_dentry->d_inode) {
 		simple_unlink(new_dir, new_dentry);
 		if (they_are_dirs)
-			old_dir->i_nlink--;
+			drop_nlink(old_dir);
 	} else if (they_are_dirs) {
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 		new_dir->i_nlink++;
 	}
 
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index 5b6a4540a05b..299bb66e3bde 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -249,7 +249,7 @@ static int minix_rename(struct inode * old_dir, struct dentry *old_dentry,
 		minix_set_link(new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index d220165d4918..635613f2f65a 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -343,7 +343,7 @@ static int msdos_rmdir(struct inode *dir, struct dentry *dentry)
 	err = fat_remove_entries(dir, &sinfo);	/* and releases bh */
 	if (err)
 		goto out;
-	dir->i_nlink--;
+	drop_nlink(dir);
 
 	inode->i_nlink = 0;
 	inode->i_ctime = CURRENT_TIME_SEC;
@@ -549,7 +549,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 			if (err)
 				goto error_dotdot;
 		}
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 		if (!new_inode)
 			new_dir->i_nlink++;
 	}
@@ -566,10 +566,9 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 		mark_inode_dirty(old_dir);
 
 	if (new_inode) {
+		drop_nlink(new_inode);
 		if (is_dir)
-			new_inode->i_nlink -= 2;
-		else
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		new_inode->i_ctime = ts;
 	}
 out:
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 7432f1a43f3d..26eecb86f9b0 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -843,7 +843,7 @@ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode)
 	nfs_inode_return_delegation(inode);
 	if (dentry->d_flags & DCACHE_NFSFS_RENAMED) {
 		lock_kernel();
-		inode->i_nlink--;
+		drop_nlink(inode);
 		nfs_complete_unlink(dentry);
 		unlock_kernel();
 	}
@@ -1401,7 +1401,7 @@ static int nfs_safe_remove(struct dentry *dentry)
 		error = NFS_PROTO(dir)->remove(dir, &dentry->d_name);
 		/* The VFS may want to delete this inode */
 		if (error == 0)
-			inode->i_nlink--;
+			drop_nlink(inode);
 		nfs_mark_for_revalidate(inode);
 		nfs_end_data_update(inode);
 	} else
@@ -1639,7 +1639,7 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto out;
 		}
 	} else
-		new_inode->i_nlink--;
+		drop_nlink(new_inode);
 
 go_ahead:
 	/*
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 849c3b4bb94a..40f83f53053a 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -739,7 +739,7 @@ static int ocfs2_link(struct dentry *old_dentry,
 	err = ocfs2_journal_dirty(handle, fe_bh);
 	if (err < 0) {
 		le16_add_cpu(&fe->i_links_count, -1);
-		inode->i_nlink--;
+		drop_nlink(inode);
 		mlog_errno(err);
 		goto bail;
 	}
@@ -749,7 +749,7 @@ static int ocfs2_link(struct dentry *old_dentry,
 			      parent_fe_bh, de_bh);
 	if (err) {
 		le16_add_cpu(&fe->i_links_count, -1);
-		inode->i_nlink--;
+		drop_nlink(inode);
 		mlog_errno(err);
 		goto bail;
 	}
diff --git a/fs/qnx4/namei.c b/fs/qnx4/namei.c
index c3d83f67154a..ad5afa4ea8f3 100644
--- a/fs/qnx4/namei.c
+++ b/fs/qnx4/namei.c
@@ -189,8 +189,7 @@ int qnx4_rmdir(struct inode *dir, struct dentry *dentry)
 	inode->i_nlink = 0;
 	mark_inode_dirty(inode);
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
-	dir->i_nlink--;
-	mark_inode_dirty(dir);
+	inode_dec_link_count(dir);
 	retval = 0;
 
       end_rmdir:
@@ -234,9 +233,8 @@ int qnx4_unlink(struct inode *dir, struct dentry *dentry)
 	mark_buffer_dirty(bh);
 	dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	mark_inode_dirty(dir);
-	inode->i_nlink--;
 	inode->i_ctime = dir->i_ctime;
-	mark_inode_dirty(inode);
+	inode_dec_link_count(inode);
 	retval = 0;
 
 end_unlink:
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index c61710e49c62..c76d427e027b 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -20,7 +20,7 @@
 #include <linux/quotaops.h>
 
 #define INC_DIR_INODE_NLINK(i) if (i->i_nlink != 1) { i->i_nlink++; if (i->i_nlink >= REISERFS_LINK_MAX) i->i_nlink=1; }
-#define DEC_DIR_INODE_NLINK(i) if (i->i_nlink != 1) i->i_nlink--;
+#define DEC_DIR_INODE_NLINK(i) if (i->i_nlink != 1) drop_nlink(i);
 
 // directory item contains array of entry headers. This performs
 // binary search through that array
@@ -994,7 +994,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
 		inode->i_nlink = 1;
 	}
 
-	inode->i_nlink--;
+	drop_nlink(inode);
 
 	/*
 	 * we schedule before doing the add_save_link call, save the link
@@ -1475,7 +1475,7 @@ static int reiserfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (S_ISDIR(new_dentry_inode->i_mode)) {
 			new_dentry_inode->i_nlink = 0;
 		} else {
-			new_dentry_inode->i_nlink--;
+			drop_nlink(new_dentry_inode);
 		}
 		new_dentry_inode->i_ctime = ctime;
 		savelink = new_dentry_inode->i_nlink;
diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c
index b8a73f716fbe..f7c08db8e34c 100644
--- a/fs/sysv/namei.c
+++ b/fs/sysv/namei.c
@@ -250,7 +250,7 @@ static int sysv_rename(struct inode * old_dir, struct dentry * old_dentry,
 		sysv_set_link(new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index ab9a7629d23e..d14d25534aa8 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -878,8 +878,7 @@ static int udf_rmdir(struct inode * dir, struct dentry * dentry)
 			inode->i_nlink);
 	inode->i_nlink = 0;
 	inode->i_size = 0;
-	mark_inode_dirty(inode);
-	dir->i_nlink --;
+	inode_dec_link_count(inode);
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = current_fs_time(dir->i_sb);
 	mark_inode_dirty(dir);
 
@@ -923,8 +922,7 @@ static int udf_unlink(struct inode * dir, struct dentry * dentry)
 		goto end_unlink;
 	dir->i_ctime = dir->i_mtime = current_fs_time(dir->i_sb);
 	mark_inode_dirty(dir);
-	inode->i_nlink--;
-	mark_inode_dirty(inode);
+	inode_dec_link_count(inode);
 	inode->i_ctime = dir->i_ctime;
 	retval = 0;
 
@@ -1101,8 +1099,7 @@ out:
 	return err;
 
 out_no_entry:
-	inode->i_nlink--;
-	mark_inode_dirty(inode);
+	inode_dec_link_count(inode);
 	iput(inode);
 	goto out;
 }
@@ -1261,9 +1258,8 @@ static int udf_rename (struct inode * old_dir, struct dentry * old_dentry,
 
 	if (new_inode)
 	{
-		new_inode->i_nlink--;
 		new_inode->i_ctime = current_fs_time(new_inode->i_sb);
-		mark_inode_dirty(new_inode);
+		inode_dec_link_count(new_inode);
 	}
 	old_dir->i_ctime = old_dir->i_mtime = current_fs_time(old_dir->i_sb);
 	mark_inode_dirty(old_dir);
@@ -1279,12 +1275,10 @@ static int udf_rename (struct inode * old_dir, struct dentry * old_dentry,
 		}
 		else
 			mark_buffer_dirty_inode(dir_bh, old_inode);
-		old_dir->i_nlink --;
-		mark_inode_dirty(old_dir);
+		inode_dec_link_count(old_dir);
 		if (new_inode)
 		{
-			new_inode->i_nlink --;
-			mark_inode_dirty(new_inode);
+			inode_dec_link_count(new_inode);
 		}
 		else
 		{
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index d344b411e261..e84c0ecf0730 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -308,7 +308,7 @@ static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		ufs_set_link(new_dir, new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		inode_dec_link_count(new_inode);
 	} else {
 		if (dir_de) {
diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c
index 9a8f48bae956..090d74ffa061 100644
--- a/fs/vfat/namei.c
+++ b/fs/vfat/namei.c
@@ -782,7 +782,7 @@ static int vfat_rmdir(struct inode *dir, struct dentry *dentry)
 	err = fat_remove_entries(dir, &sinfo);	/* and releases bh */
 	if (err)
 		goto out;
-	dir->i_nlink--;
+	drop_nlink(dir);
 
 	inode->i_nlink = 0;
 	inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC;
@@ -930,7 +930,7 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 			if (err)
 				goto error_dotdot;
 		}
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 		if (!new_inode)
  			new_dir->i_nlink++;
 	}
@@ -947,10 +947,9 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 		mark_inode_dirty(old_dir);
 
 	if (new_inode) {
+		drop_nlink(new_inode);
 		if (is_dir)
-			new_inode->i_nlink -= 2;
-		else
-			new_inode->i_nlink--;
+			drop_nlink(new_inode);
 		new_inode->i_ctime = ts;
 	}
 out:
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4bb70871873f..26d3c61116c0 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1225,9 +1225,14 @@ static inline void inode_inc_link_count(struct inode *inode)
 	mark_inode_dirty(inode);
 }
 
-static inline void inode_dec_link_count(struct inode *inode)
+static inline void drop_nlink(struct inode *inode)
 {
 	inode->i_nlink--;
+}
+
+static inline void inode_dec_link_count(struct inode *inode)
+{
+	drop_nlink(inode);
 	mark_inode_dirty(inode);
 }
 
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 840f8a6fb85f..10aa8eeeb112 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -307,7 +307,7 @@ static int mqueue_unlink(struct inode *dir, struct dentry *dentry)
 
 	dir->i_ctime = dir->i_mtime = dir->i_atime = CURRENT_TIME;
 	dir->i_size -= DIRENT_SIZE;
-  	inode->i_nlink--;
+  	drop_nlink(inode);
   	dput(dentry);
   	return 0;
 }
diff --git a/mm/shmem.c b/mm/shmem.c
index b96de69f236b..908dd947b1ea 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1772,7 +1772,7 @@ static int shmem_unlink(struct inode *dir, struct dentry *dentry)
 
 	dir->i_size -= BOGO_DIRENT_SIZE;
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
-	inode->i_nlink--;
+	drop_nlink(inode);
 	dput(dentry);	/* Undo the count from "create" - this does all the work */
 	return 0;
 }
@@ -1782,8 +1782,8 @@ static int shmem_rmdir(struct inode *dir, struct dentry *dentry)
 	if (!simple_empty(dentry))
 		return -ENOTEMPTY;
 
-	dentry->d_inode->i_nlink--;
-	dir->i_nlink--;
+	drop_nlink(dentry->d_inode);
+	drop_nlink(dir);
 	return shmem_unlink(dir, dentry);
 }
 
@@ -1804,9 +1804,9 @@ static int shmem_rename(struct inode *old_dir, struct dentry *old_dentry, struct
 	if (new_dentry->d_inode) {
 		(void) shmem_unlink(new_dir, new_dentry);
 		if (they_are_dirs)
-			old_dir->i_nlink--;
+			drop_nlink(old_dir);
 	} else if (they_are_dirs) {
-		old_dir->i_nlink--;
+		drop_nlink(old_dir);
 		new_dir->i_nlink++;
 	}
 
-- 
cgit v1.2.3


From d8c76e6f45c111c32a4b3e50a2adc9210737b0d8 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Sat, 30 Sep 2006 23:29:04 -0700
Subject: [PATCH] r/o bind mount prepwork: inc_nlink() helper

This is mostly included for parity with dec_nlink(), where we will have some
more hooks.  This one should stay pretty darn straightforward for now.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/infiniband/hw/ipath/ipath_fs.c | 4 ++--
 drivers/usb/core/inode.c               | 4 ++--
 fs/9p/vfs_inode.c                      | 2 +-
 fs/autofs/root.c                       | 2 +-
 fs/autofs4/root.c                      | 2 +-
 fs/bfs/dir.c                           | 2 +-
 fs/cifs/inode.c                        | 2 +-
 fs/coda/dir.c                          | 2 +-
 fs/configfs/dir.c                      | 4 ++--
 fs/configfs/mount.c                    | 2 +-
 fs/debugfs/inode.c                     | 4 ++--
 fs/ext3/namei.c                        | 6 +++---
 fs/fuse/control.c                      | 2 +-
 fs/hfsplus/dir.c                       | 2 +-
 fs/hpfs/namei.c                        | 4 ++--
 fs/hugetlbfs/inode.c                   | 4 ++--
 fs/jffs2/dir.c                         | 6 +++---
 fs/jffs2/fs.c                          | 6 +++---
 fs/jfs/namei.c                         | 6 +++---
 fs/libfs.c                             | 4 ++--
 fs/msdos/namei.c                       | 4 ++--
 fs/ocfs2/dlm/dlmfs.c                   | 6 +++---
 fs/ocfs2/namei.c                       | 8 ++++----
 fs/ramfs/inode.c                       | 4 ++--
 fs/reiserfs/namei.c                    | 6 +++---
 fs/sysfs/dir.c                         | 4 ++--
 fs/sysfs/mount.c                       | 2 +-
 fs/udf/inode.c                         | 2 +-
 fs/udf/namei.c                         | 6 +++---
 fs/vfat/namei.c                        | 4 ++--
 include/linux/fs.h                     | 7 ++++++-
 ipc/mqueue.c                           | 2 +-
 kernel/cpuset.c                        | 8 ++++----
 mm/shmem.c                             | 8 ++++----
 net/sunrpc/rpc_pipe.c                  | 6 +++---
 security/inode.c                       | 4 ++--
 security/selinux/selinuxfs.c           | 4 ++--
 37 files changed, 80 insertions(+), 75 deletions(-)

(limited to 'include')

diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c
index a507d0b5be6c..d9ff283f725e 100644
--- a/drivers/infiniband/hw/ipath/ipath_fs.c
+++ b/drivers/infiniband/hw/ipath/ipath_fs.c
@@ -66,8 +66,8 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry,
 	inode->i_private = data;
 	if ((mode & S_IFMT) == S_IFDIR) {
 		inode->i_op = &simple_dir_inode_operations;
-		inode->i_nlink++;
-		dir->i_nlink++;
+		inc_nlink(inode);
+		inc_nlink(dir);
 	}
 
 	inode->i_fop = fops;
diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index 88002e45a6b4..7c77c2d8d300 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -263,7 +263,7 @@ static struct inode *usbfs_get_inode (struct super_block *sb, int mode, dev_t de
 			inode->i_fop = &simple_dir_operations;
 
 			/* directory inodes start off with i_nlink == 2 (for "." entry) */
-			inode->i_nlink++;
+			inc_nlink(inode);
 			break;
 		}
 	}
@@ -295,7 +295,7 @@ static int usbfs_mkdir (struct inode *dir, struct dentry *dentry, int mode)
 	mode = (mode & (S_IRWXUGO | S_ISVTX)) | S_IFDIR;
 	res = usbfs_mknod (dir, dentry, mode, 0);
 	if (!res)
-		dir->i_nlink++;
+		inc_nlink(dir);
 	return res;
 }
 
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 7a7ec2d1d2f4..5241c600ce28 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -233,7 +233,7 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode)
 			inode->i_op = &v9fs_symlink_inode_operations;
 			break;
 		case S_IFDIR:
-			inode->i_nlink++;
+			inc_nlink(inode);
 			if(v9ses->extended)
 				inode->i_op = &v9fs_dir_inode_operations_ext;
 			else
diff --git a/fs/autofs/root.c b/fs/autofs/root.c
index 54ad70731927..368a1c33a3c8 100644
--- a/fs/autofs/root.c
+++ b/fs/autofs/root.c
@@ -466,7 +466,7 @@ static int autofs_root_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	ent->dentry = dentry;
 	autofs_hash_insert(dh,ent);
 
-	dir->i_nlink++;
+	inc_nlink(dir);
 	d_instantiate(dentry, iget(dir->i_sb,ino));
 	unlock_kernel();
 
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 348bec0982b0..e21bb4668201 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -713,7 +713,7 @@ static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (p_ino && dentry->d_parent != dentry)
 		atomic_inc(&p_ino->count);
 	ino->inode = inode;
-	dir->i_nlink++;
+	inc_nlink(dir);
 	dir->i_mtime = CURRENT_TIME;
 
 	return 0;
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index ce05d1643dd1..a650f1d0b85e 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -163,7 +163,7 @@ static int bfs_link(struct dentry * old, struct inode * dir, struct dentry * new
 		unlock_kernel();
 		return err;
 	}
-	inode->i_nlink++;
+	inc_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	mark_inode_dirty(inode);
 	atomic_inc(&inode->i_count);
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 74441a17e186..76b7fb34101a 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -735,7 +735,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode)
 		cFYI(1, ("cifs_mkdir returned 0x%x", rc));
 		d_drop(direntry);
 	} else {
-		inode->i_nlink++;
+		inc_nlink(inode);
 		if (pTcon->ses->capabilities & CAP_UNIX)
 			rc = cifs_get_inode_info_unix(&newinode, full_path,
 						      inode->i_sb,xid);
diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 0a2fd8bb7579..0102b28a15fb 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -304,7 +304,7 @@ static int coda_link(struct dentry *source_de, struct inode *dir_inode,
 	coda_dir_changed(dir_inode, 0);
 	atomic_inc(&inode->i_count);
 	d_instantiate(de, inode);
-	inode->i_nlink++;
+	inc_nlink(inode);
         
 out:
 	unlock_kernel();
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 816e8ef64560..8a3b6a1a6ad1 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -139,7 +139,7 @@ static int init_dir(struct inode * inode)
 	inode->i_fop = &configfs_dir_operations;
 
 	/* directory inodes start off with i_nlink == 2 (for "." entry) */
-	inode->i_nlink++;
+	inc_nlink(inode);
 	return 0;
 }
 
@@ -169,7 +169,7 @@ static int create_dir(struct config_item * k, struct dentry * p,
 	if (!error) {
 		error = configfs_create(d, mode, init_dir);
 		if (!error) {
-			p->d_inode->i_nlink++;
+			inc_nlink(p->d_inode);
 			(d)->d_op = &configfs_dentry_ops;
 		} else {
 			struct configfs_dirent *sd = d->d_fsdata;
diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index 3e5fe843e1df..68bd5c93ca52 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -84,7 +84,7 @@ static int configfs_fill_super(struct super_block *sb, void *data, int silent)
 		inode->i_op = &configfs_dir_inode_operations;
 		inode->i_fop = &configfs_dir_operations;
 		/* directory inodes start off with i_nlink == 2 (for "." entry) */
-		inode->i_nlink++;
+		inc_nlink(inode);
 	} else {
 		pr_debug("configfs: could not get root inode\n");
 		return -ENOMEM;
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 269e649e6dc6..ecf3da9edf21 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -54,7 +54,7 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d
 			inode->i_fop = &simple_dir_operations;
 
 			/* directory inodes start off with i_nlink == 2 (for "." entry) */
-			inode->i_nlink++;
+			inc_nlink(inode);
 			break;
 		}
 	}
@@ -87,7 +87,7 @@ static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	mode = (mode & (S_IRWXUGO | S_ISVTX)) | S_IFDIR;
 	res = debugfs_mknod(dir, dentry, mode, 0);
 	if (!res)
-		dir->i_nlink++;
+		inc_nlink(dir);
 	return res;
 }
 
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 14c55adfae83..b45c88bd5f73 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -1616,7 +1616,7 @@ static int ext3_delete_entry (handle_t *handle,
  */
 static inline void ext3_inc_count(handle_t *handle, struct inode *inode)
 {
-	inode->i_nlink++;
+	inc_nlink(inode);
 }
 
 static inline void ext3_dec_count(handle_t *handle, struct inode *inode)
@@ -1775,7 +1775,7 @@ retry:
 		iput (inode);
 		goto out_stop;
 	}
-	dir->i_nlink++;
+	inc_nlink(dir);
 	ext3_update_dx_flag(dir);
 	ext3_mark_inode_dirty(handle, dir);
 	d_instantiate(dentry, inode);
@@ -2341,7 +2341,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry,
 		if (new_inode) {
 			drop_nlink(new_inode);
 		} else {
-			new_dir->i_nlink++;
+			inc_nlink(new_dir);
 			ext3_update_dx_flag(new_dir);
 			ext3_mark_inode_dirty(handle, new_dir);
 		}
diff --git a/fs/fuse/control.c b/fs/fuse/control.c
index 79ec1f23d4d2..16b39c053d47 100644
--- a/fs/fuse/control.c
+++ b/fs/fuse/control.c
@@ -116,7 +116,7 @@ int fuse_ctl_add_conn(struct fuse_conn *fc)
 		return 0;
 
 	parent = fuse_control_sb->s_root;
-	parent->d_inode->i_nlink++;
+	inc_nlink(parent->d_inode);
 	sprintf(name, "%llu", (unsigned long long) fc->id);
 	parent = fuse_ctl_add_dentry(parent, fc, name, S_IFDIR | 0500, 2,
 				     &simple_dir_inode_operations,
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 9ceb0dfaa1cc..99b4ed1b87d2 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -298,7 +298,7 @@ static int hfsplus_link(struct dentry *src_dentry, struct inode *dst_dir,
 	if (res)
 		return res;
 
-	inode->i_nlink++;
+	inc_nlink(inode);
 	hfsplus_instantiate(dst_dentry, inode, cnid);
 	atomic_inc(&inode->i_count);
 	inode->i_ctime = CURRENT_TIME_SEC;
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 4078b0becc5e..25dd6f81eca7 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -89,7 +89,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	brelse(bh);
 	hpfs_mark_4buffers_dirty(&qbh0);
 	hpfs_brelse4(&qbh0);
-	dir->i_nlink++;
+	inc_nlink(dir);
 	insert_inode_hash(result);
 
 	if (result->i_uid != current->fsuid ||
@@ -635,7 +635,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	end:
 	hpfs_i(i)->i_parent_dir = new_dir->i_ino;
 	if (S_ISDIR(i->i_mode)) {
-		new_dir->i_nlink++;
+		inc_nlink(new_dir);
 		drop_nlink(old_dir);
 	}
 	if ((fnode = hpfs_map_fnode(i->i_sb, i->i_ino, &bh))) {
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index f5b8f329aca6..5e03b2f67b93 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -377,7 +377,7 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid,
 			inode->i_fop = &simple_dir_operations;
 
 			/* directory inodes start off with i_nlink == 2 (for "." entry) */
-			inode->i_nlink++;
+			inc_nlink(inode);
 			break;
 		case S_IFLNK:
 			inode->i_op = &page_symlink_inode_operations;
@@ -418,7 +418,7 @@ static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
 	int retval = hugetlbfs_mknod(dir, dentry, mode | S_IFDIR, 0);
 	if (!retval)
-		dir->i_nlink++;
+		inc_nlink(dir);
 	return retval;
 }
 
diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index a5e9f2205b33..9def6adf4a5d 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -588,7 +588,7 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode)
 	}
 
 	dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
-	dir_i->i_nlink++;
+	inc_nlink(dir_i);
 
 	jffs2_free_raw_dirent(rd);
 
@@ -836,7 +836,7 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry,
 	/* If it was a directory we moved, and there was no victim,
 	   increase i_nlink on its new parent */
 	if (S_ISDIR(old_dentry->d_inode->i_mode) && !victim_f)
-		new_dir_i->i_nlink++;
+		inc_nlink(new_dir_i);
 
 	/* Unlink the original */
 	ret = jffs2_do_unlink(c, JFFS2_INODE_INFO(old_dir_i),
@@ -848,7 +848,7 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry,
 		/* Oh shit. We really ought to make a single node which can do both atomically */
 		struct jffs2_inode_info *f = JFFS2_INODE_INFO(old_dentry->d_inode);
 		down(&f->sem);
-		old_dentry->d_inode->i_nlink++;
+		inc_nlink(old_dentry->d_inode);
 		if (f->inocache)
 			f->inocache->nlink++;
 		up(&f->sem);
diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index 72d9909d95ff..7bc1a4201c0c 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -277,13 +277,13 @@ void jffs2_read_inode (struct inode *inode)
 
 		for (fd=f->dents; fd; fd = fd->next) {
 			if (fd->type == DT_DIR && fd->ino)
-				inode->i_nlink++;
+				inc_nlink(inode);
 		}
 		/* and '..' */
-		inode->i_nlink++;
+		inc_nlink(inode);
 		/* Root dir gets i_nlink 3 for some reason */
 		if (inode->i_ino == 1)
-			inode->i_nlink++;
+			inc_nlink(inode);
 
 		inode->i_op = &jffs2_dir_inode_operations;
 		inode->i_fop = &jffs2_dir_operations;
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 088b85976ac0..8cef88170aa4 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -292,7 +292,7 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, int mode)
 	mark_inode_dirty(ip);
 
 	/* update parent directory inode */
-	dip->i_nlink++;		/* for '..' from child directory */
+	inc_nlink(dip);		/* for '..' from child directory */
 	dip->i_ctime = dip->i_mtime = CURRENT_TIME;
 	mark_inode_dirty(dip);
 
@@ -822,7 +822,7 @@ static int jfs_link(struct dentry *old_dentry,
 		goto free_dname;
 
 	/* update object inode */
-	ip->i_nlink++;		/* for new link */
+	inc_nlink(ip);		/* for new link */
 	ip->i_ctime = CURRENT_TIME;
 	dir->i_ctime = dir->i_mtime = CURRENT_TIME;
 	mark_inode_dirty(dir);
@@ -1206,7 +1206,7 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto out4;
 		}
 		if (S_ISDIR(old_ip->i_mode))
-			new_dir->i_nlink++;
+			inc_nlink(new_dir);
 	}
 	/*
 	 * Remove old directory entry
diff --git a/fs/libfs.c b/fs/libfs.c
index 9204feba75ac..bd08e0e64a8c 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -243,7 +243,7 @@ int simple_link(struct dentry *old_dentry, struct inode *dir, struct dentry *den
 	struct inode *inode = old_dentry->d_inode;
 
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
-	inode->i_nlink++;
+	inc_nlink(inode);
 	atomic_inc(&inode->i_count);
 	dget(dentry);
 	d_instantiate(dentry, inode);
@@ -306,7 +306,7 @@ int simple_rename(struct inode *old_dir, struct dentry *old_dentry,
 			drop_nlink(old_dir);
 	} else if (they_are_dirs) {
 		drop_nlink(old_dir);
-		new_dir->i_nlink++;
+		inc_nlink(new_dir);
 	}
 
 	old_dir->i_ctime = old_dir->i_mtime = new_dir->i_ctime =
diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index 635613f2f65a..fa868c755907 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -389,7 +389,7 @@ static int msdos_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	err = msdos_add_entry(dir, msdos_name, 1, is_hid, cluster, &ts, &sinfo);
 	if (err)
 		goto out_free;
-	dir->i_nlink++;
+	inc_nlink(dir);
 
 	inode = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
 	brelse(sinfo.bh);
@@ -551,7 +551,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name,
 		}
 		drop_nlink(old_dir);
 		if (!new_inode)
-			new_dir->i_nlink++;
+			inc_nlink(new_dir);
 	}
 
 	err = fat_remove_entries(old_dir, &old_sinfo);	/* and releases bh */
diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c
index 0368c6402182..16b8d1ba7066 100644
--- a/fs/ocfs2/dlm/dlmfs.c
+++ b/fs/ocfs2/dlm/dlmfs.c
@@ -338,7 +338,7 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb)
 		inode->i_blocks = 0;
 		inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-		inode->i_nlink++;
+		inc_nlink(inode);
 
 		inode->i_fop = &simple_dir_operations;
 		inode->i_op = &dlmfs_root_inode_operations;
@@ -395,7 +395,7 @@ static struct inode *dlmfs_get_inode(struct inode *parent,
 
 		/* directory inodes start off with i_nlink ==
 		 * 2 (for "." entry) */
-		inode->i_nlink++;
+		inc_nlink(inode);
 		break;
 	}
 
@@ -449,7 +449,7 @@ static int dlmfs_mkdir(struct inode * dir,
 	}
 	ip->ip_dlm = dlm;
 
-	dir->i_nlink++;
+	inc_nlink(dir);
 	d_instantiate(dentry, inode);
 	dget(dentry);	/* Extra count - pin the dentry in core */
 
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 40f83f53053a..8c370a39e0c1 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -429,7 +429,7 @@ static int ocfs2_mknod(struct inode *dir,
 			mlog_errno(status);
 			goto leave;
 		}
-		dir->i_nlink++;
+		inc_nlink(dir);
 	}
 
 	status = ocfs2_add_entry(handle, dentry, inode,
@@ -730,7 +730,7 @@ static int ocfs2_link(struct dentry *old_dentry,
 		goto bail;
 	}
 
-	inode->i_nlink++;
+	inc_nlink(inode);
 	inode->i_ctime = CURRENT_TIME;
 	fe->i_links_count = cpu_to_le16(inode->i_nlink);
 	fe->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
@@ -952,7 +952,7 @@ static int ocfs2_unlink(struct inode *dir,
 						parent_node_bh);
 		if (status < 0) {
 			mlog_errno(status);
-			dir->i_nlink++;
+			inc_nlink(dir);
 		}
 	}
 
@@ -1382,7 +1382,7 @@ static int ocfs2_rename(struct inode *old_dir,
 		if (new_inode) {
 			new_inode->i_nlink--;
 		} else {
-			new_dir->i_nlink++;
+			inc_nlink(new_dir);
 			mark_inode_dirty(new_dir);
 		}
 	}
diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c
index bc0e51662424..2faf4cdf61b0 100644
--- a/fs/ramfs/inode.c
+++ b/fs/ramfs/inode.c
@@ -75,7 +75,7 @@ struct inode *ramfs_get_inode(struct super_block *sb, int mode, dev_t dev)
 			inode->i_fop = &simple_dir_operations;
 
 			/* directory inodes start off with i_nlink == 2 (for "." entry) */
-			inode->i_nlink++;
+			inc_nlink(inode);
 			break;
 		case S_IFLNK:
 			inode->i_op = &page_symlink_inode_operations;
@@ -113,7 +113,7 @@ static int ramfs_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 {
 	int retval = ramfs_mknod(dir, dentry, mode | S_IFDIR, 0);
 	if (!retval)
-		dir->i_nlink++;
+		inc_nlink(dir);
 	return retval;
 }
 
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index c76d427e027b..cf92e89515f2 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -19,7 +19,7 @@
 #include <linux/smp_lock.h>
 #include <linux/quotaops.h>
 
-#define INC_DIR_INODE_NLINK(i) if (i->i_nlink != 1) { i->i_nlink++; if (i->i_nlink >= REISERFS_LINK_MAX) i->i_nlink=1; }
+#define INC_DIR_INODE_NLINK(i) if (i->i_nlink != 1) { inc_nlink(i); if (i->i_nlink >= REISERFS_LINK_MAX) i->i_nlink=1; }
 #define DEC_DIR_INODE_NLINK(i) if (i->i_nlink != 1) drop_nlink(i);
 
 // directory item contains array of entry headers. This performs
@@ -1006,7 +1006,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
 	    reiserfs_cut_from_item(&th, &path, &(de.de_entry_key), dir, NULL,
 				   0);
 	if (retval < 0) {
-		inode->i_nlink++;
+		inc_nlink(inode);
 		goto end_unlink;
 	}
 	inode->i_ctime = CURRENT_TIME_SEC;
@@ -1143,7 +1143,7 @@ static int reiserfs_link(struct dentry *old_dentry, struct inode *dir,
 	}
 
 	/* inc before scheduling so reiserfs_unlink knows we are here */
-	inode->i_nlink++;
+	inc_nlink(inode);
 
 	retval = journal_begin(&th, dir->i_sb, jbegin_count);
 	if (retval) {
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 5f3d725d1125..3aa3434621ca 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -103,7 +103,7 @@ static int init_dir(struct inode * inode)
 	inode->i_fop = &sysfs_dir_operations;
 
 	/* directory inodes start off with i_nlink == 2 (for "." entry) */
-	inode->i_nlink++;
+	inc_nlink(inode);
 	return 0;
 }
 
@@ -137,7 +137,7 @@ static int create_dir(struct kobject * k, struct dentry * p,
 		if (!error) {
 			error = sysfs_create(*d, mode, init_dir);
 			if (!error) {
-				p->d_inode->i_nlink++;
+				inc_nlink(p->d_inode);
 				(*d)->d_op = &sysfs_dentry_ops;
 				d_rehash(*d);
 			}
diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c
index 40190c489271..20551a1b8a56 100644
--- a/fs/sysfs/mount.c
+++ b/fs/sysfs/mount.c
@@ -49,7 +49,7 @@ static int sysfs_fill_super(struct super_block *sb, void *data, int silent)
 		inode->i_op = &sysfs_dir_inode_operations;
 		inode->i_fop = &sysfs_dir_operations;
 		/* directory inodes start off with i_nlink == 2 (for "." entry) */
-		inode->i_nlink++;	
+		inc_nlink(inode);
 	} else {
 		pr_debug("sysfs: could not get root inode\n");
 		return -ENOMEM;
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index b223b32db991..ae21a0e59e95 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -1165,7 +1165,7 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh)
 			inode->i_op = &udf_dir_inode_operations;
 			inode->i_fop = &udf_dir_operations;
 			inode->i_mode |= S_IFDIR;
-			inode->i_nlink ++;
+			inc_nlink(inode);
 			break;
 		}
 		case ICBTAG_FILE_TYPE_REALTIME:
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index d14d25534aa8..e40c95e65117 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -762,7 +762,7 @@ static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 		cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL);
 	cfi.fileCharacteristics |= FID_FILE_CHAR_DIRECTORY;
 	udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL);
-	dir->i_nlink++;
+	inc_nlink(dir);
 	mark_inode_dirty(dir);
 	d_instantiate(dentry, inode);
 	if (fibh.sbh != fibh.ebh)
@@ -1147,7 +1147,7 @@ static int udf_link(struct dentry * old_dentry, struct inode * dir,
 	if (fibh.sbh != fibh.ebh)
 		udf_release_data(fibh.ebh);
 	udf_release_data(fibh.sbh);
-	inode->i_nlink ++;
+	inc_nlink(inode);
 	inode->i_ctime = current_fs_time(inode->i_sb);
 	mark_inode_dirty(inode);
 	atomic_inc(&inode->i_count);
@@ -1282,7 +1282,7 @@ static int udf_rename (struct inode * old_dir, struct dentry * old_dentry,
 		}
 		else
 		{
-			new_dir->i_nlink ++;
+			inc_nlink(new_dir);
 			mark_inode_dirty(new_dir);
 		}
 	}
diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c
index 090d74ffa061..5846ba2d5d9f 100644
--- a/fs/vfat/namei.c
+++ b/fs/vfat/namei.c
@@ -837,7 +837,7 @@ static int vfat_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (err)
 		goto out_free;
 	dir->i_version++;
-	dir->i_nlink++;
+	inc_nlink(dir);
 
 	inode = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
 	brelse(sinfo.bh);
@@ -932,7 +932,7 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry,
 		}
 		drop_nlink(old_dir);
 		if (!new_inode)
- 			new_dir->i_nlink++;
+ 			inc_nlink(new_dir);
 	}
 
 	err = fat_remove_entries(old_dir, &old_sinfo);	/* and releases bh */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 26d3c61116c0..6a5267da565f 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1219,9 +1219,14 @@ static inline void mark_inode_dirty_sync(struct inode *inode)
 	__mark_inode_dirty(inode, I_DIRTY_SYNC);
 }
 
-static inline void inode_inc_link_count(struct inode *inode)
+static inline void inc_nlink(struct inode *inode)
 {
 	inode->i_nlink++;
+}
+
+static inline void inode_inc_link_count(struct inode *inode)
+{
+	inc_nlink(inode);
 	mark_inode_dirty(inode);
 }
 
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 10aa8eeeb112..d75d0ba83360 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -168,7 +168,7 @@ static struct inode *mqueue_get_inode(struct super_block *sb, int mode,
 			/* all is ok */
 			info->user = get_uid(u);
 		} else if (S_ISDIR(mode)) {
-			inode->i_nlink++;
+			inc_nlink(inode);
 			/* Some things misbehave if size == 0 on a directory */
 			inode->i_size = 2 * DIRENT_SIZE;
 			inode->i_op = &mqueue_dir_inode_operations;
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 8c3c400cce91..9d850ae13b1b 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -377,7 +377,7 @@ static int cpuset_fill_super(struct super_block *sb, void *unused_data,
 		inode->i_op = &simple_dir_inode_operations;
 		inode->i_fop = &simple_dir_operations;
 		/* directories start off with i_nlink == 2 (for "." entry) */
-		inode->i_nlink++;
+		inc_nlink(inode);
 	} else {
 		return -ENOMEM;
 	}
@@ -1565,7 +1565,7 @@ static int cpuset_create_file(struct dentry *dentry, int mode)
 		inode->i_fop = &simple_dir_operations;
 
 		/* start off with i_nlink == 2 (for "." entry) */
-		inode->i_nlink++;
+		inc_nlink(inode);
 	} else if (S_ISREG(mode)) {
 		inode->i_size = 0;
 		inode->i_fop = &cpuset_file_operations;
@@ -1598,7 +1598,7 @@ static int cpuset_create_dir(struct cpuset *cs, const char *name, int mode)
 	error = cpuset_create_file(dentry, S_IFDIR | mode);
 	if (!error) {
 		dentry->d_fsdata = cs;
-		parent->d_inode->i_nlink++;
+		inc_nlink(parent->d_inode);
 		cs->dentry = dentry;
 	}
 	dput(dentry);
@@ -2033,7 +2033,7 @@ int __init cpuset_init(void)
 	}
 	root = cpuset_mount->mnt_sb->s_root;
 	root->d_fsdata = &top_cpuset;
-	root->d_inode->i_nlink++;
+	inc_nlink(root->d_inode);
 	top_cpuset.dentry = root;
 	root->d_inode->i_op = &cpuset_dir_inode_operations;
 	number_of_cpusets = 1;
diff --git a/mm/shmem.c b/mm/shmem.c
index 908dd947b1ea..bb8ca7ef7094 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1379,7 +1379,7 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev)
 							&sbinfo->policy_nodes);
 			break;
 		case S_IFDIR:
-			inode->i_nlink++;
+			inc_nlink(inode);
 			/* Some things misbehave if size == 0 on a directory */
 			inode->i_size = 2 * BOGO_DIRENT_SIZE;
 			inode->i_op = &shmem_dir_inode_operations;
@@ -1715,7 +1715,7 @@ static int shmem_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 
 	if ((error = shmem_mknod(dir, dentry, mode | S_IFDIR, 0)))
 		return error;
-	dir->i_nlink++;
+	inc_nlink(dir);
 	return 0;
 }
 
@@ -1750,7 +1750,7 @@ static int shmem_link(struct dentry *old_dentry, struct inode *dir, struct dentr
 
 	dir->i_size += BOGO_DIRENT_SIZE;
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
-	inode->i_nlink++;
+	inc_nlink(inode);
 	atomic_inc(&inode->i_count);	/* New dentry reference */
 	dget(dentry);		/* Extra pinning count for the created dentry */
 	d_instantiate(dentry, inode);
@@ -1807,7 +1807,7 @@ static int shmem_rename(struct inode *old_dir, struct dentry *old_dentry, struct
 			drop_nlink(old_dir);
 	} else if (they_are_dirs) {
 		drop_nlink(old_dir);
-		new_dir->i_nlink++;
+		inc_nlink(new_dir);
 	}
 
 	old_dir->i_size -= BOGO_DIRENT_SIZE;
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index 700c6e061a04..9a0b41a97f90 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -494,7 +494,7 @@ rpc_get_inode(struct super_block *sb, int mode)
 		case S_IFDIR:
 			inode->i_fop = &simple_dir_operations;
 			inode->i_op = &simple_dir_inode_operations;
-			inode->i_nlink++;
+			inc_nlink(inode);
 		default:
 			break;
 	}
@@ -571,7 +571,7 @@ rpc_populate(struct dentry *parent,
 		if (private)
 			rpc_inode_setowner(inode, private);
 		if (S_ISDIR(mode))
-			dir->i_nlink++;
+			inc_nlink(dir);
 		d_add(dentry, inode);
 	}
 	mutex_unlock(&dir->i_mutex);
@@ -593,7 +593,7 @@ __rpc_mkdir(struct inode *dir, struct dentry *dentry)
 		goto out_err;
 	inode->i_ino = iunique(dir->i_sb, 100);
 	d_instantiate(dentry, inode);
-	dir->i_nlink++;
+	inc_nlink(dir);
 	inode_dir_notify(dir, DN_CREATE);
 	return 0;
 out_err:
diff --git a/security/inode.c b/security/inode.c
index 49ee51529396..9b16e14f3a80 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -78,7 +78,7 @@ static struct inode *get_inode(struct super_block *sb, int mode, dev_t dev)
 			inode->i_fop = &simple_dir_operations;
 
 			/* directory inodes start off with i_nlink == 2 (for "." entry) */
-			inode->i_nlink++;
+			inc_nlink(inode);
 			break;
 		}
 	}
@@ -111,7 +111,7 @@ static int mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	mode = (mode & (S_IRWXUGO | S_ISVTX)) | S_IFDIR;
 	res = mknod(dir, dentry, mode, 0);
 	if (!res)
-		dir->i_nlink++;
+		inc_nlink(dir);
 	return res;
 }
 
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index bab7b386cb8d..cd244419c980 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -1253,10 +1253,10 @@ static int sel_make_dir(struct inode *dir, struct dentry *dentry)
 	inode->i_op = &simple_dir_inode_operations;
 	inode->i_fop = &simple_dir_operations;
 	/* directory inodes start off with i_nlink == 2 (for "." entry) */
-	inode->i_nlink++;
+	inc_nlink(inode);
 	d_add(dentry, inode);
 	/* bump link count on parent directory, too */
-	dir->i_nlink++;
+	inc_nlink(dir);
 out:
 	return ret;
 }
-- 
cgit v1.2.3


From ce71ec36840368b877fb63bd14c8e67ab62d08b1 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Sat, 30 Sep 2006 23:29:06 -0700
Subject: [PATCH] r/o bind mounts: monitor zeroing of i_nlink

Some filesystems, instead of simply decrementing i_nlink, simply zero it
during an unlink operation.  We need to catch these in addition to the
decrement operations.

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/autofs4/root.c   | 4 ++--
 fs/cifs/inode.c     | 2 +-
 fs/ext3/namei.c     | 2 +-
 fs/fuse/dir.c       | 4 ++--
 fs/hfs/dir.c        | 2 +-
 fs/hfsplus/dir.c    | 4 ++--
 fs/hpfs/namei.c     | 4 ++--
 fs/jfs/namei.c      | 2 +-
 fs/msdos/namei.c    | 4 ++--
 fs/nfs/dir.c        | 2 +-
 fs/qnx4/namei.c     | 2 +-
 fs/reiserfs/namei.c | 4 ++--
 fs/udf/namei.c      | 2 +-
 fs/vfat/namei.c     | 4 ++--
 include/linux/fs.h  | 5 +++++
 15 files changed, 26 insertions(+), 21 deletions(-)

(limited to 'include')

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index e21bb4668201..c1493524da4d 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -638,7 +638,7 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry)
 	dput(ino->dentry);
 
 	dentry->d_inode->i_size = 0;
-	dentry->d_inode->i_nlink = 0;
+	clear_nlink(dentry->d_inode);
 
 	dir->i_mtime = CURRENT_TIME;
 
@@ -673,7 +673,7 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
 	}
 	dput(ino->dentry);
 	dentry->d_inode->i_size = 0;
-	dentry->d_inode->i_nlink = 0;
+	clear_nlink(dentry->d_inode);
 
 	if (dir->i_nlink)
 		drop_nlink(dir);
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 76b7fb34101a..6b90ef98e4cf 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -818,7 +818,7 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
 	if (!rc) {
 		drop_nlink(inode);
 		i_size_write(direntry->d_inode,0);
-		direntry->d_inode->i_nlink = 0;
+		clear_nlink(direntry->d_inode);
 	}
 
 	cifsInode = CIFS_I(direntry->d_inode);
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index b45c88bd5f73..906731a20f1a 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -2045,7 +2045,7 @@ static int ext3_rmdir (struct inode * dir, struct dentry *dentry)
 			      "empty directory has nlink!=2 (%d)",
 			      inode->i_nlink);
 	inode->i_version++;
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	/* There's no need to set i_disksize: the fact that i_nlink is
 	 * zero will ensure that the right thing happens during any
 	 * recovery. */
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index f85b2a282f13..8605155db171 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -508,7 +508,7 @@ static int fuse_unlink(struct inode *dir, struct dentry *entry)
 		/* Set nlink to zero so the inode can be cleared, if
                    the inode does have more links this will be
                    discovered at the next lookup/getattr */
-		inode->i_nlink = 0;
+		clear_nlink(inode);
 		fuse_invalidate_attr(inode);
 		fuse_invalidate_attr(dir);
 		fuse_invalidate_entry_cache(entry);
@@ -534,7 +534,7 @@ static int fuse_rmdir(struct inode *dir, struct dentry *entry)
 	err = req->out.h.error;
 	fuse_put_request(fc, req);
 	if (!err) {
-		entry->d_inode->i_nlink = 0;
+		clear_nlink(entry->d_inode);
 		fuse_invalidate_attr(dir);
 		fuse_invalidate_entry_cache(entry);
 	} else if (err == -EINTR)
diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c
index cfef6ad529a7..37d681b4f216 100644
--- a/fs/hfs/dir.c
+++ b/fs/hfs/dir.c
@@ -273,7 +273,7 @@ static int hfs_rmdir(struct inode *dir, struct dentry *dentry)
 	res = hfs_cat_delete(inode->i_ino, dir, &dentry->d_name);
 	if (res)
 		return res;
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	hfs_delete_inode(inode);
 	mark_inode_dirty(inode);
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 99b4ed1b87d2..7e309751645f 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -348,7 +348,7 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry)
 		} else
 			inode->i_flags |= S_DEAD;
 	} else
-		inode->i_nlink = 0;
+		clear_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	mark_inode_dirty(inode);
 
@@ -387,7 +387,7 @@ static int hfsplus_rmdir(struct inode *dir, struct dentry *dentry)
 	res = hfsplus_delete_cat(inode->i_ino, dir, &dentry->d_name);
 	if (res)
 		return res;
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	hfsplus_delete_inode(inode);
 	mark_inode_dirty(inode);
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 25dd6f81eca7..2507e7393f3c 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -495,7 +495,7 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 		break;
 	default:
 		drop_nlink(dir);
-		inode->i_nlink = 0;
+		clear_nlink(inode);
 		err = 0;
 	}
 	goto out;
@@ -590,7 +590,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		int r;
 		if ((r = hpfs_remove_dirent(old_dir, dno, dep, &qbh, 1)) != 2) {
 			if ((nde = map_dirent(new_dir, hpfs_i(new_dir)->i_dno, (char *)new_name, new_len, NULL, &qbh1))) {
-				new_inode->i_nlink = 0;
+				clear_nlink(new_inode);
 				copy_de(nde, &de);
 				memcpy(nde->name, new_name, new_len);
 				hpfs_mark_4buffers_dirty(&qbh1);
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 8cef88170aa4..b8d16a6aa88f 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -414,7 +414,7 @@ static int jfs_rmdir(struct inode *dip, struct dentry *dentry)
 	JFS_IP(ip)->acl.flag = 0;
 
 	/* mark the target directory as deleted */
-	ip->i_nlink = 0;
+	clear_nlink(ip);
 	mark_inode_dirty(ip);
 
 	rc = txCommit(tid, 2, &iplist[0], 0);
diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index fa868c755907..b0f01b3b0536 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -345,7 +345,7 @@ static int msdos_rmdir(struct inode *dir, struct dentry *dentry)
 		goto out;
 	drop_nlink(dir);
 
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	fat_detach(inode);
 out:
@@ -430,7 +430,7 @@ static int msdos_unlink(struct inode *dir, struct dentry *dentry)
 	err = fat_remove_entries(dir, &sinfo);	/* and releases bh */
 	if (err)
 		goto out;
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	fat_detach(inode);
 out:
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 26eecb86f9b0..481f8892a919 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1286,7 +1286,7 @@ static int nfs_rmdir(struct inode *dir, struct dentry *dentry)
 	error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name);
 	/* Ensure the VFS deletes this inode */
 	if (error == 0 && dentry->d_inode != NULL)
-		dentry->d_inode->i_nlink = 0;
+		clear_nlink(dentry->d_inode);
 	nfs_end_data_update(dir);
 	unlock_kernel();
 
diff --git a/fs/qnx4/namei.c b/fs/qnx4/namei.c
index ad5afa4ea8f3..733cdf01d645 100644
--- a/fs/qnx4/namei.c
+++ b/fs/qnx4/namei.c
@@ -186,7 +186,7 @@ int qnx4_rmdir(struct inode *dir, struct dentry *dentry)
 	memset(de->di_fname, 0, sizeof de->di_fname);
 	de->di_mode = 0;
 	mark_buffer_dirty(bh);
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	mark_inode_dirty(inode);
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	inode_dec_link_count(dir);
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index cf92e89515f2..16e9cff8f15d 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -913,7 +913,7 @@ static int reiserfs_rmdir(struct inode *dir, struct dentry *dentry)
 		reiserfs_warning(inode->i_sb, "%s: empty directory has nlink "
 				 "!= 2 (%d)", __FUNCTION__, inode->i_nlink);
 
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC;
 	reiserfs_update_sd(&th, inode);
 
@@ -1473,7 +1473,7 @@ static int reiserfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (new_dentry_inode) {
 		// adjust link number of the victim
 		if (S_ISDIR(new_dentry_inode->i_mode)) {
-			new_dentry_inode->i_nlink = 0;
+			clear_nlink(new_dentry_inode);
 		} else {
 			drop_nlink(new_dentry_inode);
 		}
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index e40c95e65117..73163325e5ec 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -876,7 +876,7 @@ static int udf_rmdir(struct inode * dir, struct dentry * dentry)
 		udf_warning(inode->i_sb, "udf_rmdir",
 			"empty directory has nlink != 2 (%d)",
 			inode->i_nlink);
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_size = 0;
 	inode_dec_link_count(inode);
 	inode->i_ctime = dir->i_ctime = dir->i_mtime = current_fs_time(dir->i_sb);
diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c
index 5846ba2d5d9f..edb711ff7b05 100644
--- a/fs/vfat/namei.c
+++ b/fs/vfat/namei.c
@@ -784,7 +784,7 @@ static int vfat_rmdir(struct inode *dir, struct dentry *dentry)
 		goto out;
 	drop_nlink(dir);
 
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC;
 	fat_detach(inode);
 out:
@@ -808,7 +808,7 @@ static int vfat_unlink(struct inode *dir, struct dentry *dentry)
 	err = fat_remove_entries(dir, &sinfo);	/* and releases bh */
 	if (err)
 		goto out;
-	inode->i_nlink = 0;
+	clear_nlink(inode);
 	inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC;
 	fat_detach(inode);
 out:
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6a5267da565f..3493d2828f7d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1235,6 +1235,11 @@ static inline void drop_nlink(struct inode *inode)
 	inode->i_nlink--;
 }
 
+static inline void clear_nlink(struct inode *inode)
+{
+	inode->i_nlink = 0;
+}
+
 static inline void inode_dec_link_count(struct inode *inode)
 {
 	drop_nlink(inode);
-- 
cgit v1.2.3


From dc2bc768a009b9ad8711894c544dc6b0d8c0ce57 Mon Sep 17 00:00:00 2001
From: Fernando Vazquez <fernando@intellilink.co.jp>
Date: Sat, 30 Sep 2006 23:29:07 -0700
Subject: [PATCH] stack overflow safe kdump: safe_smp_processor_id()

This is a the first of a series of patch-sets aiming at making kdump more
robust against stack overflows.

This patch set does the following:

* Add safe_smp_processor_id function to i386 architecture (this function was
  inspired by the x86_64 function of the same name).

* Substitute "smp_processor_id" with the stack overflow-safe
  "safe_smp_processor_id" in the reboot path to the second kernel.

This patch:

On the event of a stack overflow critical data that usually resides at the
bottom of the stack is likely to be stomped and, consequently, its use should
be avoided.

In particular, in the i386 and IA64 architectures the macro smp_processor_id
ultimately makes use of the "cpu" member of struct thread_info which resides
at the bottom of the stack.  x86_64, on the other hand, is not affected by
this problem because it benefits from the use of the PDA infrastructure.

To circumvent this problem I suggest implementing "safe_smp_processor_id()"
(it already exists in x86_64) for i386 and IA64 and use it as a replacement
for smp_processor_id in the reboot path to the dump capture kernel.  This is a
possible implementation for i386.

Signed-off-by: Fernando Vazquez <fernando@intellilink.co.jp>
Looks-reasonable-to: Andi Kleen <ak@muc.de>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smp.c | 27 +++++++++++++++++++++++++++
 include/asm-i386/smp.h |  2 ++
 2 files changed, 29 insertions(+)

(limited to 'include')

diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index 465188e2d701..1b080ab8a49f 100644
--- a/arch/i386/kernel/smp.c
+++ b/arch/i386/kernel/smp.c
@@ -700,3 +700,30 @@ int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
 	return 0;
 }
 EXPORT_SYMBOL(smp_call_function_single);
+
+static int convert_apicid_to_cpu(int apic_id)
+{
+	int i;
+
+	for (i = 0; i < NR_CPUS; i++) {
+		if (x86_cpu_to_apicid[i] == apic_id)
+			return i;
+	}
+	return -1;
+}
+
+int safe_smp_processor_id(void)
+{
+	int apicid, cpuid;
+
+	if (!boot_cpu_has(X86_FEATURE_APIC))
+		return 0;
+
+	apicid = hard_smp_processor_id();
+	if (apicid == BAD_APICID)
+		return 0;
+
+	cpuid = convert_apicid_to_cpu(apicid);
+
+	return cpuid >= 0 ? cpuid : 0;
+}
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 915c26a31b79..6aa1206f6e2a 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -84,6 +84,7 @@ static inline int hard_smp_processor_id(void)
 #endif
 #endif
 
+extern int safe_smp_processor_id(void);
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 extern unsigned int num_processors;
@@ -92,6 +93,7 @@ extern unsigned int num_processors;
 
 #else /* CONFIG_SMP */
 
+#define safe_smp_processor_id()		0
 #define cpu_physical_id(cpu)		boot_cpu_physical_apicid
 
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
-- 
cgit v1.2.3


From 74588d8ba34ff1bda027cfa737972af01ab00c8b Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Sat, 30 Sep 2006 23:29:12 -0700
Subject: [PATCH] Generic ioremap_page_range: implementation

This patch adds a generic implementation of ioremap_page_range() in
lib/ioremap.c based on the i386 implementation. It differs from the
i386 version in the following ways:

  * The PTE flags are passed as a pgprot_t argument and must be
    determined up front by the arch-specific code. No additional
    PTE flags are added.
  * Uses set_pte_at() instead of set_pte()

[bunk@stusta.de: warning fix]
]dhowells@redhat.com: nommu build fix]
Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Andi Kleen <ak@suse.de>
Cc: <linux-m32r@ml.linux-m32r.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Kyle McMartin <kyle@parisc-linux.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/io.h |  4 +++
 lib/Makefile       |  1 +
 lib/ioremap.c      | 94 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 99 insertions(+)
 create mode 100644 lib/ioremap.c

(limited to 'include')

diff --git a/include/linux/io.h b/include/linux/io.h
index 420e2fdf26f6..aa3f5af670b5 100644
--- a/include/linux/io.h
+++ b/include/linux/io.h
@@ -19,8 +19,12 @@
 #define _LINUX_IO_H
 
 #include <asm/io.h>
+#include <asm/page.h>
 
 void __iowrite32_copy(void __iomem *to, const void *from, size_t count);
 void __iowrite64_copy(void __iomem *to, const void *from, size_t count);
 
+int ioremap_page_range(unsigned long addr, unsigned long end,
+		       unsigned long phys_addr, pgprot_t prot);
+
 #endif /* _LINUX_IO_H */
diff --git a/lib/Makefile b/lib/Makefile
index 402762fead70..ddf3e676e1f4 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -7,6 +7,7 @@ lib-y := errno.o ctype.o string.o vsprintf.o cmdline.o \
 	 idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \
 	 sha1.o
 
+lib-$(CONFIG_MMU) += ioremap.o
 lib-$(CONFIG_SMP) += cpumask.o
 
 lib-y	+= kobject.o kref.o kobject_uevent.o klist.o
diff --git a/lib/ioremap.c b/lib/ioremap.c
new file mode 100644
index 000000000000..29c810ec9813
--- /dev/null
+++ b/lib/ioremap.c
@@ -0,0 +1,94 @@
+/*
+ * Re-map IO memory to kernel address space so that we can access it.
+ * This is needed for high PCI addresses that aren't mapped in the
+ * 640k-1MB IO memory area on PC's
+ *
+ * (C) Copyright 1995 1996 Linus Torvalds
+ */
+#include <linux/io.h>
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+
+#include <asm/cacheflush.h>
+#include <asm/pgtable.h>
+
+static int ioremap_pte_range(pmd_t *pmd, unsigned long addr,
+		unsigned long end, unsigned long phys_addr, pgprot_t prot)
+{
+	pte_t *pte;
+	unsigned long pfn;
+
+	pfn = phys_addr >> PAGE_SHIFT;
+	pte = pte_alloc_kernel(pmd, addr);
+	if (!pte)
+		return -ENOMEM;
+	do {
+		BUG_ON(!pte_none(*pte));
+		set_pte_at(&init_mm, addr, pte, pfn_pte(pfn, prot));
+		pfn++;
+	} while (pte++, addr += PAGE_SIZE, addr != end);
+	return 0;
+}
+
+static inline int ioremap_pmd_range(pud_t *pud, unsigned long addr,
+		unsigned long end, unsigned long phys_addr, pgprot_t prot)
+{
+	pmd_t *pmd;
+	unsigned long next;
+
+	phys_addr -= addr;
+	pmd = pmd_alloc(&init_mm, pud, addr);
+	if (!pmd)
+		return -ENOMEM;
+	do {
+		next = pmd_addr_end(addr, end);
+		if (ioremap_pte_range(pmd, addr, next, phys_addr + addr, prot))
+			return -ENOMEM;
+	} while (pmd++, addr = next, addr != end);
+	return 0;
+}
+
+static inline int ioremap_pud_range(pgd_t *pgd, unsigned long addr,
+		unsigned long end, unsigned long phys_addr, pgprot_t prot)
+{
+	pud_t *pud;
+	unsigned long next;
+
+	phys_addr -= addr;
+	pud = pud_alloc(&init_mm, pgd, addr);
+	if (!pud)
+		return -ENOMEM;
+	do {
+		next = pud_addr_end(addr, end);
+		if (ioremap_pmd_range(pud, addr, next, phys_addr + addr, prot))
+			return -ENOMEM;
+	} while (pud++, addr = next, addr != end);
+	return 0;
+}
+
+int ioremap_page_range(unsigned long addr,
+		       unsigned long end, unsigned long phys_addr, pgprot_t prot)
+{
+	pgd_t *pgd;
+	unsigned long start;
+	unsigned long next;
+	int err;
+
+	BUG_ON(addr >= end);
+
+	flush_cache_all();
+
+	start = addr;
+	phys_addr -= addr;
+	pgd = pgd_offset_k(addr);
+	do {
+		next = pgd_addr_end(addr, end);
+		err = ioremap_pud_range(pgd, addr, next, phys_addr+addr, prot);
+		if (err)
+			break;
+	} while (pgd++, addr = next, addr != end);
+
+	flush_tlb_all();
+
+	return err;
+}
-- 
cgit v1.2.3


From d6cbd281d189977b38eac7eb2a4678de19b6b483 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sat, 30 Sep 2006 23:29:26 -0700
Subject: [PATCH] Some cleanup in the pipe code

Split the big and hard to read do_pipe function into smaller pieces.

This creates new create_write_pipe/free_write_pipe/create_read_pipe
functions.  These functions are made global so that they can be used by
other parts of the kernel.

The resulting code is more generic and easier to read and has cleaner error
handling and less gotos.

[akpm@osdl.org: cleanup]
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/pipe.c          | 155 ++++++++++++++++++++++++++++++++---------------------
 include/linux/fs.h |   3 ++
 2 files changed, 96 insertions(+), 62 deletions(-)

(limited to 'include')

diff --git a/fs/pipe.c b/fs/pipe.c
index 2e60e1c8815e..b1626f269a34 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -874,87 +874,118 @@ fail_inode:
 	return NULL;
 }
 
-int do_pipe(int *fd)
+struct file *create_write_pipe(void)
 {
-	struct qstr this;
-	char name[32];
+	int err;
+	struct inode *inode;
+	struct file *f;
 	struct dentry *dentry;
-	struct inode * inode;
-	struct file *f1, *f2;
-	int error;
-	int i, j;
-
-	error = -ENFILE;
-	f1 = get_empty_filp();
-	if (!f1)
-		goto no_files;
-
-	f2 = get_empty_filp();
-	if (!f2)
-		goto close_f1;
+	char name[32];
+	struct qstr this;
 
+	f = get_empty_filp();
+	if (!f)
+		return ERR_PTR(-ENFILE);
+	err = -ENFILE;
 	inode = get_pipe_inode();
 	if (!inode)
-		goto close_f12;
+		goto err_file;
 
-	error = get_unused_fd();
-	if (error < 0)
-		goto close_f12_inode;
-	i = error;
-
-	error = get_unused_fd();
-	if (error < 0)
-		goto close_f12_inode_i;
-	j = error;
-
-	error = -ENOMEM;
 	sprintf(name, "[%lu]", inode->i_ino);
 	this.name = name;
 	this.len = strlen(name);
 	this.hash = inode->i_ino; /* will go */
+	err = -ENOMEM;
 	dentry = d_alloc(pipe_mnt->mnt_sb->s_root, &this);
 	if (!dentry)
-		goto close_f12_inode_i_j;
+		goto err_inode;
 
 	dentry->d_op = &pipefs_dentry_operations;
 	d_add(dentry, inode);
-	f1->f_vfsmnt = f2->f_vfsmnt = mntget(mntget(pipe_mnt));
-	f1->f_dentry = f2->f_dentry = dget(dentry);
-	f1->f_mapping = f2->f_mapping = inode->i_mapping;
-
-	/* read file */
-	f1->f_pos = f2->f_pos = 0;
-	f1->f_flags = O_RDONLY;
-	f1->f_op = &read_pipe_fops;
-	f1->f_mode = FMODE_READ;
-	f1->f_version = 0;
-
-	/* write file */
-	f2->f_flags = O_WRONLY;
-	f2->f_op = &write_pipe_fops;
-	f2->f_mode = FMODE_WRITE;
-	f2->f_version = 0;
-
-	fd_install(i, f1);
-	fd_install(j, f2);
-	fd[0] = i;
-	fd[1] = j;
+	f->f_vfsmnt = mntget(pipe_mnt);
+	f->f_dentry = dentry;
+	f->f_mapping = inode->i_mapping;
 
-	return 0;
+	f->f_flags = O_WRONLY;
+	f->f_op = &write_pipe_fops;
+	f->f_mode = FMODE_WRITE;
+	f->f_version = 0;
 
-close_f12_inode_i_j:
-	put_unused_fd(j);
-close_f12_inode_i:
-	put_unused_fd(i);
-close_f12_inode:
+	return f;
+
+ err_inode:
 	free_pipe_info(inode);
 	iput(inode);
-close_f12:
-	put_filp(f2);
-close_f1:
-	put_filp(f1);
-no_files:
-	return error;	
+ err_file:
+	put_filp(f);
+	return ERR_PTR(err);
+}
+
+void free_write_pipe(struct file *f)
+{
+	mntput(f->f_vfsmnt);
+	dput(f->f_dentry);
+	put_filp(f);
+}
+
+struct file *create_read_pipe(struct file *wrf)
+{
+	struct file *f = get_empty_filp();
+	if (!f)
+		return ERR_PTR(-ENFILE);
+
+	/* Grab pipe from the writer */
+	f->f_vfsmnt = mntget(wrf->f_vfsmnt);
+	f->f_dentry = dget(wrf->f_dentry);
+	f->f_mapping = wrf->f_dentry->d_inode->i_mapping;
+
+	f->f_pos = 0;
+	f->f_flags = O_RDONLY;
+	f->f_op = &read_pipe_fops;
+	f->f_mode = FMODE_READ;
+	f->f_version = 0;
+
+	return f;
+}
+
+int do_pipe(int *fd)
+{
+	struct file *fw, *fr;
+	int error;
+	int fdw, fdr;
+
+	fw = create_write_pipe();
+	if (IS_ERR(fw))
+		return PTR_ERR(fw);
+	fr = create_read_pipe(fw);
+	error = PTR_ERR(fr);
+	if (IS_ERR(fr))
+		goto err_write_pipe;
+
+	error = get_unused_fd();
+	if (error < 0)
+		goto err_read_pipe;
+	fdr = error;
+
+	error = get_unused_fd();
+	if (error < 0)
+		goto err_fdr;
+	fdw = error;
+
+	fd_install(fdr, fr);
+	fd_install(fdw, fw);
+	fd[0] = fdr;
+	fd[1] = fdw;
+
+	return 0;
+
+ err_fdr:
+	put_unused_fd(fdr);
+ err_read_pipe:
+	put_filp(fr);
+ err_write_pipe:
+	free_write_pipe(fw);
+	return error;
 }
 
 /*
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 3493d2828f7d..2e29a2edaeec 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1641,6 +1641,9 @@ static inline void allow_write_access(struct file *file)
 		atomic_inc(&file->f_dentry->d_inode->i_writecount);
 }
 extern int do_pipe(int *);
+extern struct file *create_read_pipe(struct file *f);
+extern struct file *create_write_pipe(void);
+extern void free_write_pipe(struct file *);
 
 extern int open_namei(int dfd, const char *, int, int, struct nameidata *);
 extern int may_open(struct nameidata *, int, int);
-- 
cgit v1.2.3


From e239ca540594cff00adcce163dc332b27015d8e5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sat, 30 Sep 2006 23:29:27 -0700
Subject: [PATCH] Create call_usermodehelper_pipe()

A new member in the ever growing family of call_usermode* functions is
born.  The new call_usermodehelper_pipe() function allows to pipe data to
the stdin of the called user mode progam and behaves otherwise like the
normal call_usermodehelp() (except that it always waits for the child to
finish)

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kmod.h |  4 ++++
 kernel/kmod.c        | 55 +++++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 58 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/linux/kmod.h b/include/linux/kmod.h
index 0db22a1ab474..10f505c8431d 100644
--- a/include/linux/kmod.h
+++ b/include/linux/kmod.h
@@ -47,4 +47,8 @@ call_usermodehelper(char *path, char **argv, char **envp, int wait)
 
 extern void usermodehelper_init(void);
 
+struct file;
+extern int call_usermodehelper_pipe(char *path, char *argv[], char *envp[],
+				    struct file **filp);
+
 #endif /* __LINUX_KMOD_H__ */
diff --git a/kernel/kmod.c b/kernel/kmod.c
index 842f8015d7fd..5c63c53014a9 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -122,6 +122,7 @@ struct subprocess_info {
 	struct key *ring;
 	int wait;
 	int retval;
+	struct file *stdin;
 };
 
 /*
@@ -145,12 +146,26 @@ static int ____call_usermodehelper(void *data)
 
 	key_put(old_session);
 
+	/* Install input pipe when needed */
+	if (sub_info->stdin) {
+		struct files_struct *f = current->files;
+		struct fdtable *fdt;
+		/* no races because files should be private here */
+		sys_close(0);
+		fd_install(0, sub_info->stdin);
+		spin_lock(&f->file_lock);
+		fdt = files_fdtable(f);
+		FD_SET(0, fdt->open_fds);
+		FD_CLR(0, fdt->close_on_exec);
+		spin_unlock(&f->file_lock);
+	}
+
 	/* We can run anywhere, unlike our parent keventd(). */
 	set_cpus_allowed(current, CPU_MASK_ALL);
 
 	retval = -EPERM;
 	if (current->fs->root)
-		retval = execve(sub_info->path, sub_info->argv,sub_info->envp);
+		retval = execve(sub_info->path, sub_info->argv, sub_info->envp);
 
 	/* Exec failed? */
 	sub_info->retval = retval;
@@ -268,6 +283,44 @@ int call_usermodehelper_keys(char *path, char **argv, char **envp,
 }
 EXPORT_SYMBOL(call_usermodehelper_keys);
 
+int call_usermodehelper_pipe(char *path, char **argv, char **envp,
+			     struct file **filp)
+{
+	DECLARE_COMPLETION(done);
+	struct subprocess_info sub_info = {
+		.complete	= &done,
+		.path		= path,
+		.argv		= argv,
+		.envp		= envp,
+		.retval		= 0,
+	};
+	struct file *f;
+	DECLARE_WORK(work, __call_usermodehelper, &sub_info);
+
+	if (!khelper_wq)
+		return -EBUSY;
+
+	if (path[0] == '\0')
+		return 0;
+
+	f = create_write_pipe();
+	if (!f)
+		return -ENOMEM;
+	*filp = f;
+
+	f = create_read_pipe(f);
+	if (!f) {
+		free_write_pipe(*filp);
+		return -ENOMEM;
+	}
+	sub_info.stdin = f;
+
+	queue_work(khelper_wq, &work);
+	wait_for_completion(&done);
+	return sub_info.retval;
+}
+EXPORT_SYMBOL(call_usermodehelper_pipe);
+
 void __init usermodehelper_init(void)
 {
 	khelper_wq = create_singlethread_workqueue("khelper");
-- 
cgit v1.2.3


From 9888a1cae3f859db38b9604e3df1c02177161bb0 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:31 -0700
Subject: [PATCH] paravirt: pte clear not present

Change pte_clear_full to a more appropriately named pte_clear_not_present,
allowing optimizations when not-present mapping changes need not be reflected
in the hardware TLB for protected page table modes.  There is also another
case that can use it in the fremap code.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/pgtable.h | 9 +++++++--
 mm/fremap.c                   | 2 +-
 mm/memory.c                   | 2 +-
 3 files changed, 9 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 349260cd86ed..78740716c9e7 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -112,8 +112,13 @@ do {				  					  \
 })
 #endif
 
-#ifndef __HAVE_ARCH_PTE_CLEAR_FULL
-#define pte_clear_full(__mm, __address, __ptep, __full)			\
+/*
+ * Some architectures may be able to avoid expensive synchronization
+ * primitives when modifications are made to PTE's which are already
+ * not present, or in the process of an address space destruction.
+ */
+#ifndef __HAVE_ARCH_PTE_CLEAR_NOT_PRESENT_FULL
+#define pte_clear_not_present_full(__mm, __address, __ptep, __full)	\
 do {									\
 	pte_clear((__mm), (__address), (__ptep));			\
 } while (0)
diff --git a/mm/fremap.c b/mm/fremap.c
index aa30618ec6b2..7a9d0f5d246d 100644
--- a/mm/fremap.c
+++ b/mm/fremap.c
@@ -39,7 +39,7 @@ static int zap_pte(struct mm_struct *mm, struct vm_area_struct *vma,
 	} else {
 		if (!pte_file(pte))
 			free_swap_and_cache(pte_to_swp_entry(pte));
-		pte_clear(mm, addr, ptep);
+		pte_clear_not_present_full(mm, addr, ptep, 0);
 	}
 	return !!page;
 }
diff --git a/mm/memory.c b/mm/memory.c
index 7707187484cc..2e754621d333 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -690,7 +690,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 			continue;
 		if (!pte_file(ptent))
 			free_swap_and_cache(pte_to_swp_entry(ptent));
-		pte_clear_full(mm, addr, pte, tlb->fullmm);
+		pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
 	} while (pte++, addr += PAGE_SIZE, (addr != end && *zap_work > 0));
 
 	add_mm_rss(mm, file_rss, anon_rss);
-- 
cgit v1.2.3


From 6606c3e0da5360799e07ae24b05080cc85c68e72 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:33 -0700
Subject: [PATCH] paravirt: lazy mmu mode hooks.patch

Implement lazy MMU update hooks which are SMP safe for both direct and shadow
page tables.  The idea is that PTE updates and page invalidations while in
lazy mode can be batched into a single hypercall.  We use this in VMI for
shadow page table synchronization, and it is a win.  It also can be used by
PPC and for direct page tables on Xen.

For SMP, the enter / leave must happen under protection of the page table
locks for page tables which are being modified.  This is because otherwise,
you end up with stale state in the batched hypercall, which other CPUs can
race ahead of.  Doing this under the protection of the locks guarantees the
synchronization is correct, and also means that spurious faults which are
generated during this window by remote CPUs are properly handled, as the page
fault handler must re-check the PTE under protection of the same lock.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/pgtable.h | 20 ++++++++++++++++++++
 mm/memory.c                   |  8 ++++++++
 mm/mprotect.c                 |  2 ++
 mm/mremap.c                   |  2 ++
 4 files changed, 32 insertions(+)

(limited to 'include')

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 78740716c9e7..56627fa453a6 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -170,6 +170,26 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addres
 #define move_pte(pte, prot, old_addr, new_addr)	(pte)
 #endif
 
+/*
+ * A facility to provide lazy MMU batching.  This allows PTE updates and
+ * page invalidations to be delayed until a call to leave lazy MMU mode
+ * is issued.  Some architectures may benefit from doing this, and it is
+ * beneficial for both shadow and direct mode hypervisors, which may batch
+ * the PTE updates which happen during this window.  Note that using this
+ * interface requires that read hazards be removed from the code.  A read
+ * hazard could result in the direct mode hypervisor case, since the actual
+ * write to the page tables may not yet have taken place, so reads though
+ * a raw PTE pointer after it has been modified are not guaranteed to be
+ * up to date.  This mode can only be entered and left under the protection of
+ * the page table locks for all page tables which may be modified.  In the UP
+ * case, this is required so that preemption is disabled, and in the SMP case,
+ * it must synchronize the delayed page table writes properly on other CPUs.
+ */
+#ifndef __HAVE_ARCH_ENTER_LAZY_MMU_MODE
+#define arch_enter_lazy_mmu_mode()	do {} while (0)
+#define arch_leave_lazy_mmu_mode()	do {} while (0)
+#endif
+
 /*
  * When walking page tables, get the address of the next boundary,
  * or the end address of the range if that comes earlier.  Although no
diff --git a/mm/memory.c b/mm/memory.c
index 2e754621d333..9cf3f341a28a 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -506,6 +506,7 @@ again:
 	src_pte = pte_offset_map_nested(src_pmd, addr);
 	src_ptl = pte_lockptr(src_mm, src_pmd);
 	spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
+	arch_enter_lazy_mmu_mode();
 
 	do {
 		/*
@@ -527,6 +528,7 @@ again:
 		progress += 8;
 	} while (dst_pte++, src_pte++, addr += PAGE_SIZE, addr != end);
 
+	arch_leave_lazy_mmu_mode();
 	spin_unlock(src_ptl);
 	pte_unmap_nested(src_pte - 1);
 	add_mm_rss(dst_mm, rss[0], rss[1]);
@@ -628,6 +630,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 	int anon_rss = 0;
 
 	pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+	arch_enter_lazy_mmu_mode();
 	do {
 		pte_t ptent = *pte;
 		if (pte_none(ptent)) {
@@ -694,6 +697,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 	} while (pte++, addr += PAGE_SIZE, (addr != end && *zap_work > 0));
 
 	add_mm_rss(mm, file_rss, anon_rss);
+	arch_leave_lazy_mmu_mode();
 	pte_unmap_unlock(pte - 1, ptl);
 
 	return addr;
@@ -1109,6 +1113,7 @@ static int zeromap_pte_range(struct mm_struct *mm, pmd_t *pmd,
 	pte = pte_alloc_map_lock(mm, pmd, addr, &ptl);
 	if (!pte)
 		return -ENOMEM;
+	arch_enter_lazy_mmu_mode();
 	do {
 		struct page *page = ZERO_PAGE(addr);
 		pte_t zero_pte = pte_wrprotect(mk_pte(page, prot));
@@ -1118,6 +1123,7 @@ static int zeromap_pte_range(struct mm_struct *mm, pmd_t *pmd,
 		BUG_ON(!pte_none(*pte));
 		set_pte_at(mm, addr, pte, zero_pte);
 	} while (pte++, addr += PAGE_SIZE, addr != end);
+	arch_leave_lazy_mmu_mode();
 	pte_unmap_unlock(pte - 1, ptl);
 	return 0;
 }
@@ -1275,11 +1281,13 @@ static int remap_pte_range(struct mm_struct *mm, pmd_t *pmd,
 	pte = pte_alloc_map_lock(mm, pmd, addr, &ptl);
 	if (!pte)
 		return -ENOMEM;
+	arch_enter_lazy_mmu_mode();
 	do {
 		BUG_ON(!pte_none(*pte));
 		set_pte_at(mm, addr, pte, pfn_pte(pfn, prot));
 		pfn++;
 	} while (pte++, addr += PAGE_SIZE, addr != end);
+	arch_leave_lazy_mmu_mode();
 	pte_unmap_unlock(pte - 1, ptl);
 	return 0;
 }
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 955f9d0e38aa..3b8f3c0c63f3 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -34,6 +34,7 @@ static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
 	spinlock_t *ptl;
 
 	pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+	arch_enter_lazy_mmu_mode();
 	do {
 		oldpte = *pte;
 		if (pte_present(oldpte)) {
@@ -70,6 +71,7 @@ static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
 		}
 
 	} while (pte++, addr += PAGE_SIZE, addr != end);
+	arch_leave_lazy_mmu_mode();
 	pte_unmap_unlock(pte - 1, ptl);
 }
 
diff --git a/mm/mremap.c b/mm/mremap.c
index 7c15cf3373ad..9c769fa29f32 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -98,6 +98,7 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd,
 	new_ptl = pte_lockptr(mm, new_pmd);
 	if (new_ptl != old_ptl)
 		spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
+	arch_enter_lazy_mmu_mode();
 
 	for (; old_addr < old_end; old_pte++, old_addr += PAGE_SIZE,
 				   new_pte++, new_addr += PAGE_SIZE) {
@@ -109,6 +110,7 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd,
 		set_pte_at(mm, new_addr, new_pte, pte);
 	}
 
+	arch_leave_lazy_mmu_mode();
 	if (new_ptl != old_ptl)
 		spin_unlock(new_ptl);
 	pte_unmap_nested(new_pte - 1);
-- 
cgit v1.2.3


From 25e4df5bae333a06cd2c9b88baf14432652dc9f7 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:34 -0700
Subject: [PATCH] paravirt: combine flush accessed dirty.patch

Remove ptep_test_and_clear_{dirty|young} from i386, and instead use the
dominating functions, ptep_clear_flush_{dirty|young}.  This allows the TLB
page flush to be contained in the same macro, and allows for an eager
optimization - if reading the PTE initially returned dirty/accessed, we can
assume the fact that no subsequent update to the PTE which cleared accessed /
dirty has occurred, as the only way A/D bits can change without holding the
page table lock is if a remote processor clears them.  This eliminates an
extra branch which came from the generic version of the code, as we know that
no other CPU could have cleared the A/D bit, so the flush will always be
needed.

We still export these two defines, even though we do not actually define
the macros in the i386 code:

 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY

The reason for this is that the only use of these functions is within the
generic clear_flush functions, and we want a strong guarantee that there
are no other users of these functions, so we want to prevent the generic
code from defining them for us.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/pgtable.h | 41 ++++++++++++++++++++++++++++-------------
 1 file changed, 28 insertions(+), 13 deletions(-)

(limited to 'include')

diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 541b3e234335..94c87ff4b5ac 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -262,21 +262,36 @@ do {									\
 	}								\
 } while (0)
 
+/*
+ * We don't actually have these, but we want to advertise them so that
+ * we can encompass the flush here.
+ */
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
-static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
-{
-	if (!pte_dirty(*ptep))
-		return 0;
-	return test_and_clear_bit(_PAGE_BIT_DIRTY, &ptep->pte_low);
-}
-
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
-{
-	if (!pte_young(*ptep))
-		return 0;
-	return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte_low);
-}
+
+#define __HAVE_ARCH_PTEP_CLEAR_DIRTY_FLUSH
+#define ptep_clear_flush_dirty(vma, address, ptep)			\
+({									\
+	int __dirty;							\
+	__dirty = pte_dirty(*(ptep));					\
+	if (__dirty) {							\
+		clear_bit(_PAGE_BIT_DIRTY, &(ptep)->pte_low);		\
+		flush_tlb_page(vma, address);				\
+	}								\
+	__dirty;							\
+})
+
+#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
+#define ptep_clear_flush_young(vma, address, ptep)			\
+({									\
+	int __young;							\
+	__young = pte_young(*(ptep));					\
+	if (__young) {							\
+		clear_bit(_PAGE_BIT_ACCESSED, &(ptep)->pte_low);	\
+		flush_tlb_page(vma, address);				\
+	}								\
+	__young;							\
+})
 
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
 static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
-- 
cgit v1.2.3


From 23002d88be309a7c78db69363c9d933a29a3b0bb Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:35 -0700
Subject: [PATCH] paravirt: kpte flush

Create a new PTE function which combines clearing a kernel PTE with the
subsequent flush.  This allows the two to be easily combined into a single
hypercall or paravirt-op.  More subtly, reverse the order of the flush for
kmap_atomic.  Instead of flushing on establishing a mapping, flush on clearing
a mapping.  This eliminates the possibility of leaving stale kmap entries
which may still have valid TLB mappings.  This is required for direct mode
hypervisors, which need to reprotect all mappings of a given page when
changing the page type from a normal page to a protected page (such as a page
table or descriptor table page).  But it also provides some nicer semantics
for real hardware, by providing extra debug-proofing against using stale
mappings, as well as ensuring that no stale mappings exist when changing the
cacheability attributes of a page, which could lead to cache conflicts when
two different types of mappings exist for the same page.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/highmem.c     | 18 +++++++-----------
 include/asm-i386/pgtable.h |  7 +++++++
 2 files changed, 14 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c
index ba44000b9069..f9f647cdbc7b 100644
--- a/arch/i386/mm/highmem.c
+++ b/arch/i386/mm/highmem.c
@@ -38,22 +38,19 @@ void *kmap_atomic(struct page *page, enum km_type type)
 
 	idx = type + KM_TYPE_NR*smp_processor_id();
 	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
-#ifdef CONFIG_DEBUG_HIGHMEM
 	if (!pte_none(*(kmap_pte-idx)))
 		BUG();
-#endif
 	set_pte(kmap_pte-idx, mk_pte(page, kmap_prot));
-	__flush_tlb_one(vaddr);
 
 	return (void*) vaddr;
 }
 
 void kunmap_atomic(void *kvaddr, enum km_type type)
 {
-#ifdef CONFIG_DEBUG_HIGHMEM
 	unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
 	enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
 
+#ifdef CONFIG_DEBUG_HIGHMEM
 	if (vaddr >= PAGE_OFFSET && vaddr < (unsigned long)high_memory) {
 		dec_preempt_count();
 		preempt_check_resched();
@@ -62,14 +59,14 @@ void kunmap_atomic(void *kvaddr, enum km_type type)
 
 	if (vaddr != __fix_to_virt(FIX_KMAP_BEGIN+idx))
 		BUG();
-
+#endif
 	/*
-	 * force other mappings to Oops if they'll try to access
-	 * this pte without first remap it
+	 * Force other mappings to Oops if they'll try to access this pte
+	 * without first remap it.  Keeping stale mappings around is a bad idea
+	 * also, in case the page changes cacheability attributes or becomes
+	 * a protected page in a hypervisor.
 	 */
-	pte_clear(&init_mm, vaddr, kmap_pte-idx);
-	__flush_tlb_one(vaddr);
-#endif
+	kpte_clear_flush(kmap_pte-idx, vaddr);
 
 	dec_preempt_count();
 	preempt_check_resched();
@@ -88,7 +85,6 @@ void *kmap_atomic_pfn(unsigned long pfn, enum km_type type)
 	idx = type + KM_TYPE_NR*smp_processor_id();
 	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
 	set_pte(kmap_pte-idx, pfn_pte(pfn, kmap_prot));
-	__flush_tlb_one(vaddr);
 
 	return (void*) vaddr;
 }
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 94c87ff4b5ac..ee9696d2f67f 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -441,6 +441,13 @@ extern pte_t *lookup_address(unsigned long address);
 #define pte_unmap_nested(pte) do { } while (0)
 #endif
 
+/* Clear a kernel PTE and flush it from the TLB */
+#define kpte_clear_flush(ptep, vaddr)					\
+do {									\
+	pte_clear(&init_mm, vaddr, ptep);				\
+	__flush_tlb_one(vaddr);						\
+} while (0)
+
 /*
  * The i386 doesn't have any external MMU info: the kernel page
  * tables contain all the necessary information.
-- 
cgit v1.2.3


From d6d861e3c963b4077c83e078e3e300c4b81f93e7 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:36 -0700
Subject: [PATCH] paravirt: optimize ptep establish for pae

The ptep_establish macro is only used on user-level PTEs, for P->P mapping
changes.  Since these always happen under protection of the pagetable lock,
the strong synchronization of a 64-bit cmpxchg is not needed, in fact, not
even a lock prefix needs to be used.  We can simply instead clear the P-bit,
followed by a normal set.  The write ordering is still important to avoid the
possibility of the TLB snooping a partially written PTE and getting a bad
mapping installed.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/pgtable-2level.h |  1 +
 include/asm-i386/pgtable-3level.h | 15 +++++++++++++++
 include/asm-i386/pgtable.h        | 11 +++++++++++
 3 files changed, 27 insertions(+)

(limited to 'include')

diff --git a/include/asm-i386/pgtable-2level.h b/include/asm-i386/pgtable-2level.h
index 201c86a6711e..8d8d3b9ecdb0 100644
--- a/include/asm-i386/pgtable-2level.h
+++ b/include/asm-i386/pgtable-2level.h
@@ -16,6 +16,7 @@
 #define set_pte(pteptr, pteval) (*(pteptr) = pteval)
 #define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
 #define set_pte_atomic(pteptr, pteval) set_pte(pteptr,pteval)
+#define set_pte_present(mm,addr,ptep,pteval) set_pte_at(mm,addr,ptep,pteval)
 #define set_pmd(pmdptr, pmdval) (*(pmdptr) = (pmdval))
 
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index 0d899173232e..7c58debdb39e 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -58,6 +58,21 @@ static inline void set_pte(pte_t *ptep, pte_t pte)
 }
 #define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
 
+/*
+ * Since this is only called on user PTEs, and the page fault handler
+ * must handle the already racy situation of simultaneous page faults,
+ * we are justified in merely clearing the PTE present bit, followed
+ * by a set.  The ordering here is important.
+ */
+static inline void set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
+{
+	ptep->pte_low = 0;
+	smp_wmb();
+	ptep->pte_high = pte.pte_high;
+	smp_wmb();
+	ptep->pte_low = pte.pte_low;
+}
+
 #define __HAVE_ARCH_SET_PTE_ATOMIC
 #define set_pte_atomic(pteptr,pteval) \
 		set_64bit((unsigned long long *)(pteptr),pte_val(pteval))
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index ee9696d2f67f..8cb708a6bed0 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -269,6 +269,17 @@ do {									\
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
 
+/*
+ * Rules for using ptep_establish: the pte MUST be a user pte, and
+ * must be a present->present transition.
+ */
+#define __HAVE_ARCH_PTEP_ESTABLISH
+#define ptep_establish(vma, address, ptep, pteval)			\
+do {									\
+	set_pte_present((vma)->vm_mm, address, ptep, pteval);		\
+	flush_tlb_page(vma, address);					\
+} while (0)
+
 #define __HAVE_ARCH_PTEP_CLEAR_DIRTY_FLUSH
 #define ptep_clear_flush_dirty(vma, address, ptep)			\
 ({									\
-- 
cgit v1.2.3


From a93cb055a23f3172c1e6a22ac1dc4f1c07929b08 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:37 -0700
Subject: [PATCH] paravirt: remove set pte atomic

Now that ptep_establish has a definition in PAE i386 3-level paging code, the
only paging model which is insane enough to have multi-word hardware PTEs
which are not efficient to set atomically, we can remove the ghost of
set_pte_atomic from other architectures which falesly duplicated it, and
remove all knowledge of it from the generic pgtable code.

set_pte_atomic is now a private pte operator which is specific to i386

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/pgtable.h         | 2 --
 include/asm-generic/pgtable.h     | 8 --------
 include/asm-i386/pgtable-3level.h | 1 -
 include/asm-m32r/pgtable-2level.h | 2 +-
 4 files changed, 1 insertion(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h
index 2fb3c6f05e03..ba1b37df69d5 100644
--- a/include/asm-frv/pgtable.h
+++ b/include/asm-frv/pgtable.h
@@ -176,8 +176,6 @@ do {							\
 } while(0)
 #define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
 
-#define set_pte_atomic(pteptr, pteval)		set_pte((pteptr), (pteval))
-
 /*
  * pgd_offset() returns a (pgd_t *)
  * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 56627fa453a6..9d774d07d95b 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -15,19 +15,11 @@
  * Note: the old pte is known to not be writable, so we don't need to
  * worry about dirty bits etc getting lost.
  */
-#ifndef __HAVE_ARCH_SET_PTE_ATOMIC
 #define ptep_establish(__vma, __address, __ptep, __entry)		\
 do {				  					\
 	set_pte_at((__vma)->vm_mm, (__address), __ptep, __entry);	\
 	flush_tlb_page(__vma, __address);				\
 } while (0)
-#else /* __HAVE_ARCH_SET_PTE_ATOMIC */
-#define ptep_establish(__vma, __address, __ptep, __entry)		\
-do {				  					\
-	set_pte_atomic(__ptep, __entry);				\
-	flush_tlb_page(__vma, __address);				\
-} while (0)
-#endif /* __HAVE_ARCH_SET_PTE_ATOMIC */
 #endif
 
 #ifndef __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index 7c58debdb39e..c2d701ea35be 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -73,7 +73,6 @@ static inline void set_pte_present(struct mm_struct *mm, unsigned long addr, pte
 	ptep->pte_low = pte.pte_low;
 }
 
-#define __HAVE_ARCH_SET_PTE_ATOMIC
 #define set_pte_atomic(pteptr,pteval) \
 		set_64bit((unsigned long long *)(pteptr),pte_val(pteval))
 #define set_pmd(pmdptr,pmdval) \
diff --git a/include/asm-m32r/pgtable-2level.h b/include/asm-m32r/pgtable-2level.h
index 6a674e3d37a2..84152760e0b5 100644
--- a/include/asm-m32r/pgtable-2level.h
+++ b/include/asm-m32r/pgtable-2level.h
@@ -44,7 +44,7 @@ static inline int pgd_present(pgd_t pgd)	{ return 1; }
  */
 #define set_pte(pteptr, pteval) (*(pteptr) = pteval)
 #define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
-#define set_pte_atomic(pteptr, pteval)	set_pte(pteptr, pteval)
+
 /*
  * (pmds are folded into pgds so this doesnt get actually called,
  * but the define is needed for a generic inline function.)
-- 
cgit v1.2.3


From 789e6ac0a7cbbb38402293256a295302fd8a1100 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:38 -0700
Subject: [PATCH] paravirt: update pte hook

Add a pte_update_hook which notifies about pte changes that have been made
without using the set_pte / clear_pte interfaces.  This allows shadow mode
hypervisors which do not trap on page table access to maintain synchronized
shadows.

It also turns out, there was one pte update in PAE mode that wasn't using any
accessor interface at all for setting NX protection.  Considering it is PAE
specific, and the accessor is i386 specific, I didn't want to add a generic
encapsulation of this behavior yet.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/init.c        |  1 +
 include/asm-i386/pgtable.h | 21 +++++++++++++++++++++
 2 files changed, 22 insertions(+)

(limited to 'include')

diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 4a5a914b3432..90089c14c23d 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -493,6 +493,7 @@ int __init set_kernel_exec(unsigned long vaddr, int enable)
 		pte->pte_high &= ~(1 << (_PAGE_BIT_NX - 32));
 	else
 		pte->pte_high |= 1 << (_PAGE_BIT_NX - 32);
+	pte_update_defer(&init_mm, vaddr, pte);
 	__flush_tlb_all();
 out:
 	return ret;
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 8cb708a6bed0..7d398f493dde 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -246,6 +246,23 @@ static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return p
 # include <asm/pgtable-2level.h>
 #endif
 
+/*
+ * Rules for using pte_update - it must be called after any PTE update which
+ * has not been done using the set_pte / clear_pte interfaces.  It is used by
+ * shadow mode hypervisors to resynchronize the shadow page tables.  Kernel PTE
+ * updates should either be sets, clears, or set_pte_atomic for P->P
+ * transitions, which means this hook should only be called for user PTEs.
+ * This hook implies a P->P protection or access change has taken place, which
+ * requires a subsequent TLB flush.  The notification can optionally be delayed
+ * until the TLB flush event by using the pte_update_defer form of the
+ * interface, but care must be taken to assure that the flush happens while
+ * still holding the same page table lock so that the shadow and primary pages
+ * do not become out of sync on SMP.
+ */
+#define pte_update(mm, addr, ptep)		do { } while (0)
+#define pte_update_defer(mm, addr, ptep)	do { } while (0)
+
+
 /*
  * We only update the dirty/accessed state if we set
  * the dirty bit by hand in the kernel, since the hardware
@@ -258,6 +275,7 @@ static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return p
 do {									\
 	if (dirty) {							\
 		(ptep)->pte_low = (entry).pte_low;			\
+		pte_update_defer((vma)->vm_mm, (addr), (ptep));		\
 		flush_tlb_page(vma, address);				\
 	}								\
 } while (0)
@@ -287,6 +305,7 @@ do {									\
 	__dirty = pte_dirty(*(ptep));					\
 	if (__dirty) {							\
 		clear_bit(_PAGE_BIT_DIRTY, &(ptep)->pte_low);		\
+		pte_update_defer((vma)->vm_mm, (addr), (ptep));		\
 		flush_tlb_page(vma, address);				\
 	}								\
 	__dirty;							\
@@ -299,6 +318,7 @@ do {									\
 	__young = pte_young(*(ptep));					\
 	if (__young) {							\
 		clear_bit(_PAGE_BIT_ACCESSED, &(ptep)->pte_low);	\
+		pte_update_defer((vma)->vm_mm, (addr), (ptep));		\
 		flush_tlb_page(vma, address);				\
 	}								\
 	__young;							\
@@ -321,6 +341,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	clear_bit(_PAGE_BIT_RW, &ptep->pte_low);
+	pte_update(mm, addr, ptep);
 }
 
 /*
-- 
cgit v1.2.3


From 5a73fdc5ea836d2edc5e02890b51185fe304d7d3 Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Sat, 30 Sep 2006 23:29:38 -0700
Subject: [PATCH] Some config.h removals

During tracking down a PAE compile failure, I found that config.h was being
included in a bunch of places in i386 code.  It is no longer necessary, so
drop it.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/boot/video.S | 2 --
 arch/i386/kernel/nmi.c | 1 -
 arch/i386/lib/delay.c  | 1 -
 include/linux/unwind.h | 2 --
 4 files changed, 6 deletions(-)

(limited to 'include')

diff --git a/arch/i386/boot/video.S b/arch/i386/boot/video.S
index 8c2a6faeeae5..2c5b5cc55f79 100644
--- a/arch/i386/boot/video.S
+++ b/arch/i386/boot/video.S
@@ -11,8 +11,6 @@
  *
  */
 
-#include <linux/config.h> /* for CONFIG_VIDEO_* */
-
 /* Enable autodetection of SVGA adapters and modes. */
 #undef CONFIG_VIDEO_SVGA
 
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 0fc4997fb143..3e8e3adb0489 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -13,7 +13,6 @@
  *  Mikael Pettersson	: PM converted to driver model. Disable/enable API.
  */
 
-#include <linux/config.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
diff --git a/arch/i386/lib/delay.c b/arch/i386/lib/delay.c
index 3c0714c4b669..f6edb11364df 100644
--- a/arch/i386/lib/delay.c
+++ b/arch/i386/lib/delay.c
@@ -11,7 +11,6 @@
  */
 
 #include <linux/module.h>
-#include <linux/config.h>
 #include <linux/sched.h>
 #include <linux/delay.h>
 
diff --git a/include/linux/unwind.h b/include/linux/unwind.h
index ce48e2cd37a2..73e1751d03dd 100644
--- a/include/linux/unwind.h
+++ b/include/linux/unwind.h
@@ -12,8 +12,6 @@
  * is not much point in implementing the full Dwarf2 unwind API.
  */
 
-#include <linux/config.h>
-
 struct module;
 
 #ifdef CONFIG_STACK_UNWIND
-- 
cgit v1.2.3